Merge tag 'drm-next-2019-05-09' of git://anongit.freedesktop.org/drm/drm

Pull drm updates from Dave Airlie:
 "This has two exciting community drivers for ARM Mali accelerators.
  Since ARM has never been open source friendly on the GPU side of the
  house, the community has had to create open source drivers for the
  Mali GPUs. Lima covers the older t4xx and panfrost the newer 6xx/7xx
  series. Well done to all involved and hopefully this will help ARM
  head in the right direction.

  There is also now the ability if you don't have any of the legacy
  drivers enabled (pre-KMS) to remove all the pre-KMS support code from
  the core drm, this saves 10% or so in codesize on my machine.

  i915 also enable Icelake/Elkhart Lake Gen11 GPUs by default, vboxvideo
  moves out of staging.

  There are also some rcar-du patches which crossover with media tree
  but all should be acked by Mauro.

  Summary:

  uapi changes:
   - Colorspace connector property
   - fourcc - new YUV formts
   - timeline sync objects initially merged
   - expose FB_DAMAGE_CLIPS to atomic userspace

  new drivers:
   - vboxvideo: moved out of staging
   - aspeed: ASPEED SoC BMC chip display support
   - lima: ARM Mali4xx GPU acceleration driver support
   - panfrost: ARM Mali6xx/7xx Midgard/Bitfrost acceleration driver support

  core:
   - component helper docs
   - unplugging fixes
   - devm device init
   - MIPI/DSI rate control
   - shmem backed gem objects
   - connector, display_info, edid_quirks cleanups
   - dma_buf fence chain support
   - 64-bit dma-fence seqno comparison fixes
   - move initial fb config code to core
   - gem fence array helpers for Lima
   - ability to remove legacy support code if no drivers requires it (removes 10% of drm.ko size)
   - lease fixes

  ttm:
   - unified DRM_FILE_PAGE_OFFSET handling
   - Account for kernel allocations in kernel zone only

  panel:
   - OSD070T1718-19TS panel support
   - panel-tpo-td028ttec1 backlight support
   - Ronbo RB070D30 MIPI/DSI
   - Feiyang FY07024DI26A30-D MIPI-DSI panel
   - Rocktech jh057n00900 MIPI-DSI panel

  i915:
   - Comet Lake (Gen9) PCI IDs
   - Updated Icelake PCI IDs
   - Elkhartlake (Gen11) support
   - DP MST property addtions
   - plane and watermark fixes
   - Icelake port sync and VEBOX disable fixes
   - struct_mutex usage reduction
   - Icelake gamma fix
   - GuC reset fixes
   - make mmap more asynchronous
   - sound display power well race fixes
   - DDI/MIPI-DSI clocks for Icelake
   - Icelake RPS frequency changing support
   - Icelake workarounds

  amdgpu:
   - Use HMM for userptr
   - vega20 experimental smu11 support
   - RAS support for vega20
   - BACO support for vega12 + fixes for vega20
   - reworked IH interrupt handling
   - amdkfd RAS support
   - Freesync improvements
   - initial timeline sync object support
   - DC Z ordering fixes
   - NV12 planes support
   - colorspace properties for planes=
   - eDP opts if eDP already initialized

  nouveau:
   - misc fixes

  etnaviv:
   - misc fixes

  msm:
   - GPU zap shader support expansion
   - robustness ABI addition

  exynos:
   - Logging cleanups

  tegra:
   - Shared reset fix
   - CPU cache maintenance fix

  cirrus:
   - driver rewritten using simple helpers

  meson:
   - G12A support

  vmwgfx:
   - Resource dirtying management improvements
   - Userspace logging improvements

  virtio:
   - PRIME fixes

  rockchip:
   - rk3066 hdmi support

  sun4i:
   - DSI burst mode support

  vc4:
   - load tracker to detect underflow

  v3d:
   - v3d v4.2 support

  malidp:
   - initial Mali D71 support in komeda driver

  tfp410:
   - omap related improvement

  omapdrm:
   - drm bridge/panel support
   - drop some omap specific panels

  rcar-du:
   - Display writeback support"

* tag 'drm-next-2019-05-09' of git://anongit.freedesktop.org/drm/drm: (1507 commits)
  drm/msm/a6xx: No zap shader is not an error
  drm/cma-helper: Fix drm_gem_cma_free_object()
  drm: Fix timestamp docs for variable refresh properties.
  drm/komeda: Mark the local functions as static
  drm/komeda: Fixed warning: Function parameter or member not described
  drm/komeda: Expose bus_width to Komeda-CORE
  drm/komeda: Add sysfs attribute: core_id and config_id
  drm: add non-desktop quirk for Valve HMDs
  drm/panfrost: Show stored feature registers
  drm/panfrost: Don't scream about deferred probe
  drm/panfrost: Disable PM on probe failure
  drm/panfrost: Set DMA masks earlier
  drm/panfrost: Add sanity checks to submit IOCTL
  drm/etnaviv: initialize idle mask before querying the HW db
  drm: introduce a capability flag for syncobj timeline support
  drm: report consistent errors when checking syncobj capibility
  drm/nouveau/nouveau: forward error generated while resuming objects tree
  drm/nouveau/fb/ramgk104: fix spelling mistake "sucessfully" -> "successfully"
  drm/nouveau/i2c: Disable i2c bus access after ->fini()
  drm/nouveau: Remove duplicate ACPI_VIDEO_NOTIFY_PROBE definition
  ...

932 files changed, 61573 insertions, 23510 deletions

diff --git a/drivers/dma-buf/Makefile b/drivers/dma-buf/Makefile
index 0913a6ccab5a..1f006e083eb9 100644
--- a/drivers/dma-buf/Makefile
+++ b/drivers/dma-buf/Makefile
@@ -1,4 +1,5 @@
-obj-y := dma-buf.o dma-fence.o dma-fence-array.o reservation.o seqno-fence.o
+obj-y := dma-buf.o dma-fence.o dma-fence-array.o dma-fence-chain.o \
+	 reservation.o seqno-fence.o
 obj-$(CONFIG_SYNC_FILE)		+= sync_file.o
 obj-$(CONFIG_SW_SYNC)		+= sw_sync.o sync_debug.o
 obj-$(CONFIG_UDMABUF)		+= udmabuf.o
diff --git a/drivers/dma-buf/dma-fence-chain.c b/drivers/dma-buf/dma-fence-chain.c
new file mode 100644
index 000000000000..93c42078cb57
--- /dev/null
+++ b/drivers/dma-buf/dma-fence-chain.c
@@ -0,0 +1,242 @@
+/*
+ * fence-chain: chain fences together in a timeline
+ *
+ * Copyright (C) 2018 Advanced Micro Devices, Inc.
+ * Authors:
+ *	Christian König <christian.koenig@amd.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ */
+
+#include <linux/dma-fence-chain.h>
+
+static bool dma_fence_chain_enable_signaling(struct dma_fence *fence);
+
+/**
+ * dma_fence_chain_get_prev - use RCU to get a reference to the previous fence
+ * @chain: chain node to get the previous node from
+ *
+ * Use dma_fence_get_rcu_safe to get a reference to the previous fence of the
+ * chain node.
+ */
+static struct dma_fence *dma_fence_chain_get_prev(struct dma_fence_chain *chain)
+{
+	struct dma_fence *prev;
+
+	rcu_read_lock();
+	prev = dma_fence_get_rcu_safe(&chain->prev);
+	rcu_read_unlock();
+	return prev;
+}
+
+/**
+ * dma_fence_chain_walk - chain walking function
+ * @fence: current chain node
+ *
+ * Walk the chain to the next node. Returns the next fence or NULL if we are at
+ * the end of the chain. Garbage collects chain nodes which are already
+ * signaled.
+ */
+struct dma_fence *dma_fence_chain_walk(struct dma_fence *fence)
+{
+	struct dma_fence_chain *chain, *prev_chain;
+	struct dma_fence *prev, *replacement, *tmp;
+
+	chain = to_dma_fence_chain(fence);
+	if (!chain) {
+		dma_fence_put(fence);
+		return NULL;
+	}
+
+	while ((prev = dma_fence_chain_get_prev(chain))) {
+
+		prev_chain = to_dma_fence_chain(prev);
+		if (prev_chain) {
+			if (!dma_fence_is_signaled(prev_chain->fence))
+				break;
+
+			replacement = dma_fence_chain_get_prev(prev_chain);
+		} else {
+			if (!dma_fence_is_signaled(prev))
+				break;
+
+			replacement = NULL;
+		}
+
+		tmp = cmpxchg((void **)&chain->prev, (void *)prev, (void *)replacement);
+		if (tmp == prev)
+			dma_fence_put(tmp);
+		else
+			dma_fence_put(replacement);
+		dma_fence_put(prev);
+	}
+
+	dma_fence_put(fence);
+	return prev;
+}
+EXPORT_SYMBOL(dma_fence_chain_walk);
+
+/**
+ * dma_fence_chain_find_seqno - find fence chain node by seqno
+ * @pfence: pointer to the chain node where to start
+ * @seqno: the sequence number to search for
+ *
+ * Advance the fence pointer to the chain node which will signal this sequence
+ * number. If no sequence number is provided then this is a no-op.
+ *
+ * Returns EINVAL if the fence is not a chain node or the sequence number has
+ * not yet advanced far enough.
+ */
+int dma_fence_chain_find_seqno(struct dma_fence **pfence, uint64_t seqno)
+{
+	struct dma_fence_chain *chain;
+
+	if (!seqno)
+		return 0;
+
+	chain = to_dma_fence_chain(*pfence);
+	if (!chain || chain->base.seqno < seqno)
+		return -EINVAL;
+
+	dma_fence_chain_for_each(*pfence, &chain->base) {
+		if ((*pfence)->context != chain->base.context ||
+		    to_dma_fence_chain(*pfence)->prev_seqno < seqno)
+			break;
+	}
+	dma_fence_put(&chain->base);
+
+	return 0;
+}
+EXPORT_SYMBOL(dma_fence_chain_find_seqno);
+
+static const char *dma_fence_chain_get_driver_name(struct dma_fence *fence)
+{
+        return "dma_fence_chain";
+}
+
+static const char *dma_fence_chain_get_timeline_name(struct dma_fence *fence)
+{
+        return "unbound";
+}
+
+static void dma_fence_chain_irq_work(struct irq_work *work)
+{
+	struct dma_fence_chain *chain;
+
+	chain = container_of(work, typeof(*chain), work);
+
+	/* Try to rearm the callback */
+	if (!dma_fence_chain_enable_signaling(&chain->base))
+		/* Ok, we are done. No more unsignaled fences left */
+		dma_fence_signal(&chain->base);
+	dma_fence_put(&chain->base);
+}
+
+static void dma_fence_chain_cb(struct dma_fence *f, struct dma_fence_cb *cb)
+{
+	struct dma_fence_chain *chain;
+
+	chain = container_of(cb, typeof(*chain), cb);
+	irq_work_queue(&chain->work);
+	dma_fence_put(f);
+}
+
+static bool dma_fence_chain_enable_signaling(struct dma_fence *fence)
+{
+	struct dma_fence_chain *head = to_dma_fence_chain(fence);
+
+	dma_fence_get(&head->base);
+	dma_fence_chain_for_each(fence, &head->base) {
+		struct dma_fence_chain *chain = to_dma_fence_chain(fence);
+		struct dma_fence *f = chain ? chain->fence : fence;
+
+		dma_fence_get(f);
+		if (!dma_fence_add_callback(f, &head->cb, dma_fence_chain_cb)) {
+			dma_fence_put(fence);
+			return true;
+		}
+		dma_fence_put(f);
+	}
+	dma_fence_put(&head->base);
+	return false;
+}
+
+static bool dma_fence_chain_signaled(struct dma_fence *fence)
+{
+	dma_fence_chain_for_each(fence, fence) {
+		struct dma_fence_chain *chain = to_dma_fence_chain(fence);
+		struct dma_fence *f = chain ? chain->fence : fence;
+
+		if (!dma_fence_is_signaled(f)) {
+			dma_fence_put(fence);
+			return false;
+		}
+	}
+
+	return true;
+}
+
+static void dma_fence_chain_release(struct dma_fence *fence)
+{
+	struct dma_fence_chain *chain = to_dma_fence_chain(fence);
+
+	dma_fence_put(rcu_dereference_protected(chain->prev, true));
+	dma_fence_put(chain->fence);
+	dma_fence_free(fence);
+}
+
+const struct dma_fence_ops dma_fence_chain_ops = {
+	.use_64bit_seqno = true,
+	.get_driver_name = dma_fence_chain_get_driver_name,
+	.get_timeline_name = dma_fence_chain_get_timeline_name,
+	.enable_signaling = dma_fence_chain_enable_signaling,
+	.signaled = dma_fence_chain_signaled,
+	.release = dma_fence_chain_release,
+};
+EXPORT_SYMBOL(dma_fence_chain_ops);
+
+/**
+ * dma_fence_chain_init - initialize a fence chain
+ * @chain: the chain node to initialize
+ * @prev: the previous fence
+ * @fence: the current fence
+ *
+ * Initialize a new chain node and either start a new chain or add the node to
+ * the existing chain of the previous fence.
+ */
+void dma_fence_chain_init(struct dma_fence_chain *chain,
+			  struct dma_fence *prev,
+			  struct dma_fence *fence,
+			  uint64_t seqno)
+{
+	struct dma_fence_chain *prev_chain = to_dma_fence_chain(prev);
+	uint64_t context;
+
+	spin_lock_init(&chain->lock);
+	rcu_assign_pointer(chain->prev, prev);
+	chain->fence = fence;
+	chain->prev_seqno = 0;
+	init_irq_work(&chain->work, dma_fence_chain_irq_work);
+
+	/* Try to reuse the context of the previous chain node. */
+	if (prev_chain && __dma_fence_is_later(seqno, prev->seqno, prev->ops)) {
+		context = prev->context;
+		chain->prev_seqno = prev->seqno;
+	} else {
+		context = dma_fence_context_alloc(1);
+		/* Make sure that we always have a valid sequence number. */
+		if (prev_chain)
+			seqno = max(prev->seqno, seqno);
+	}
+
+	dma_fence_init(&chain->base, &dma_fence_chain_ops,
+		       &chain->lock, context, seqno);
+}
+EXPORT_SYMBOL(dma_fence_chain_init);
diff --git a/drivers/dma-buf/reservation.c b/drivers/dma-buf/reservation.c
index c1618335ca99..4d32e2c67862 100644
--- a/drivers/dma-buf/reservation.c
+++ b/drivers/dma-buf/reservation.c
@@ -73,6 +73,8 @@ int reservation_object_reserve_shared(struct reservation_object *obj,
 	struct reservation_object_list *old, *new;
 	unsigned int i, j, k, max;
 
+	reservation_object_assert_held(obj);
+
 	old = reservation_object_get_list(obj);
 
 	if (old && old->shared_max) {
@@ -151,6 +153,8 @@ void reservation_object_add_shared_fence(struct reservation_object *obj,
 
 	dma_fence_get(fence);
 
+	reservation_object_assert_held(obj);
+
 	fobj = reservation_object_get_list(obj);
 	count = fobj->shared_count;
 
@@ -196,6 +200,8 @@ void reservation_object_add_excl_fence(struct reservation_object *obj,
 	struct reservation_object_list *old;
 	u32 i = 0;
 
+	reservation_object_assert_held(obj);
+
 	old = reservation_object_get_list(obj);
 	if (old)
 		i = old->shared_count;
@@ -236,6 +242,8 @@ int reservation_object_copy_fences(struct reservation_object *dst,
 	size_t size;
 	unsigned i;
 
+	reservation_object_assert_held(dst);
+
 	rcu_read_lock();
 	src_list = rcu_dereference(src->fence);
 
diff --git a/drivers/dma-buf/sw_sync.c b/drivers/dma-buf/sw_sync.c
index 32dcf7b4c935..119b2ffbc2c9 100644
--- a/drivers/dma-buf/sw_sync.c
+++ b/drivers/dma-buf/sw_sync.c
@@ -161,7 +161,7 @@ static bool timeline_fence_signaled(struct dma_fence *fence)
 {
 	struct sync_timeline *parent = dma_fence_parent(fence);
 
-	return !__dma_fence_is_later(fence->seqno, parent->value);
+	return !__dma_fence_is_later(fence->seqno, parent->value, fence->ops);
 }
 
 static bool timeline_fence_enable_signaling(struct dma_fence *fence)
diff --git a/drivers/dma-buf/sync_file.c b/drivers/dma-buf/sync_file.c
index 4f6305ca52c8..ed3fb6e5224c 100644
--- a/drivers/dma-buf/sync_file.c
+++ b/drivers/dma-buf/sync_file.c
@@ -258,7 +258,8 @@ static struct sync_file *sync_file_merge(const char *name, struct sync_file *a,
 
 			i_b++;
 		} else {
-			if (__dma_fence_is_later(pt_a->seqno, pt_b->seqno))
+			if (__dma_fence_is_later(pt_a->seqno, pt_b->seqno,
+						 pt_a->ops))
 				add_fence(fences, &i, pt_a);
 			else
 				add_fence(fences, &i, pt_b);
diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index bd943a71756c..2267e84d5cb4 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -173,6 +173,12 @@ config DRM_KMS_CMA_HELPER
 	help
 	  Choose this if you need the KMS CMA helper functions
 
+config DRM_GEM_SHMEM_HELPER
+	bool
+	depends on DRM
+	help
+	  Choose this if you need the GEM shmem helper functions
+
 config DRM_VM
 	bool
 	depends on DRM && MMU
@@ -225,8 +231,6 @@ config DRM_AMDGPU
 
 source "drivers/gpu/drm/amd/amdgpu/Kconfig"
 
-source "drivers/gpu/drm/amd/lib/Kconfig"
-
 source "drivers/gpu/drm/nouveau/Kconfig"
 
 source "drivers/gpu/drm/i915/Kconfig"
@@ -251,6 +255,9 @@ config DRM_VKMS
 
 	  If M is selected the module will be called vkms.
 
+config DRM_ATI_PCIGART
+        bool
+
 source "drivers/gpu/drm/exynos/Kconfig"
 
 source "drivers/gpu/drm/rockchip/Kconfig"
@@ -329,12 +336,21 @@ source "drivers/gpu/drm/tve200/Kconfig"
 
 source "drivers/gpu/drm/xen/Kconfig"
 
+source "drivers/gpu/drm/vboxvideo/Kconfig"
+
+source "drivers/gpu/drm/lima/Kconfig"
+
+source "drivers/gpu/drm/panfrost/Kconfig"
+
+source "drivers/gpu/drm/aspeed/Kconfig"
+
 # Keep legacy drivers last
 
 menuconfig DRM_LEGACY
 	bool "Enable legacy drivers (DANGEROUS)"
 	depends on DRM && MMU
 	select DRM_VM
+	select DRM_ATI_PCIGART if PCI
 	help
 	  Enable legacy DRI1 drivers. Those drivers expose unsafe and dangerous
 	  APIs to user-space, which can be used to circumvent access
diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile
index 1ac55c65eac0..72f5036d9bfa 100644
--- a/drivers/gpu/drm/Makefile
+++ b/drivers/gpu/drm/Makefile
@@ -3,11 +3,9 @@
 # Makefile for the drm device driver.  This driver provides support for the
 # Direct Rendering Infrastructure (DRI) in XFree86 4.1.0 and higher.
 
-drm-y       :=	drm_auth.o drm_bufs.o drm_cache.o \
-		drm_context.o drm_dma.o \
+drm-y       :=	drm_auth.o drm_cache.o \
 		drm_file.o drm_gem.o drm_ioctl.o drm_irq.o \
-		drm_lock.o drm_memory.o drm_drv.o \
-		drm_scatter.o drm_pci.o \
+		drm_memory.o drm_drv.o drm_pci.o \
 		drm_sysfs.o drm_hashtab.o drm_mm.o \
 		drm_crtc.o drm_fourcc.o drm_modes.o drm_edid.o \
 		drm_encoder_slave.o \
@@ -21,11 +19,13 @@ drm-y       :=	drm_auth.o drm_bufs.o drm_cache.o \
 		drm_syncobj.o drm_lease.o drm_writeback.o drm_client.o \
 		drm_atomic_uapi.o
 
+drm-$(CONFIG_DRM_LEGACY) += drm_legacy_misc.o drm_bufs.o drm_context.o drm_dma.o drm_scatter.o drm_lock.o
 drm-$(CONFIG_DRM_LIB_RANDOM) += lib/drm_random.o
 drm-$(CONFIG_DRM_VM) += drm_vm.o
 drm-$(CONFIG_COMPAT) += drm_ioc32.o
 drm-$(CONFIG_DRM_GEM_CMA_HELPER) += drm_gem_cma_helper.o
-drm-$(CONFIG_PCI) += ati_pcigart.o
+drm-$(CONFIG_DRM_GEM_SHMEM_HELPER) += drm_gem_shmem_helper.o
+drm-$(CONFIG_DRM_ATI_PCIGART) += ati_pcigart.o
 drm-$(CONFIG_DRM_PANEL) += drm_panel.o
 drm-$(CONFIG_OF) += drm_of.o
 drm-$(CONFIG_AGP) += drm_agpsupport.o
@@ -37,7 +37,8 @@ drm_kms_helper-y := drm_crtc_helper.o drm_dp_helper.o drm_dsc.o drm_probe_helper
 		drm_kms_helper_common.o drm_dp_dual_mode_helper.o \
 		drm_simple_kms_helper.o drm_modeset_helper.o \
 		drm_scdc_helper.o drm_gem_framebuffer_helper.o \
-		drm_atomic_state_helper.o drm_damage_helper.o
+		drm_atomic_state_helper.o drm_damage_helper.o \
+		drm_format_helper.o
 
 drm_kms_helper-$(CONFIG_DRM_PANEL_BRIDGE) += bridge/panel.o
 drm_kms_helper-$(CONFIG_DRM_FBDEV_EMULATION) += drm_fb_helper.o
@@ -56,7 +57,6 @@ obj-$(CONFIG_DRM_TTM)	+= ttm/
 obj-$(CONFIG_DRM_SCHED)	+= scheduler/
 obj-$(CONFIG_DRM_TDFX)	+= tdfx/
 obj-$(CONFIG_DRM_R128)	+= r128/
-obj-y			+= amd/lib/
 obj-$(CONFIG_HSA_AMD) += amd/amdkfd/
 obj-$(CONFIG_DRM_RADEON)+= radeon/
 obj-$(CONFIG_DRM_AMDGPU)+= amd/amdgpu/
@@ -109,3 +109,7 @@ obj-$(CONFIG_DRM_TINYDRM) += tinydrm/
 obj-$(CONFIG_DRM_PL111) += pl111/
 obj-$(CONFIG_DRM_TVE200) += tve200/
 obj-$(CONFIG_DRM_XEN) += xen/
+obj-$(CONFIG_DRM_VBOXVIDEO) += vboxvideo/
+obj-$(CONFIG_DRM_LIMA)  += lima/
+obj-$(CONFIG_DRM_PANFROST) += panfrost/
+obj-$(CONFIG_DRM_ASPEED_GFX) += aspeed/
diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index 466da5954a68..fdd0ca4b0f0b 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -23,7 +23,7 @@
 # Makefile for the drm device driver.  This driver provides support for the
 # Direct Rendering Infrastructure (DRI) in XFree86 4.1.0 and higher.
 
-FULL_AMD_PATH=$(src)/..
+FULL_AMD_PATH=$(srctree)/$(src)/..
 DISPLAY_FOLDER_NAME=display
 FULL_AMD_DISPLAY_PATH = $(FULL_AMD_PATH)/$(DISPLAY_FOLDER_NAME)
 
@@ -53,7 +53,8 @@ amdgpu-y += amdgpu_device.o amdgpu_kms.o \
 	amdgpu_ucode.o amdgpu_bo_list.o amdgpu_ctx.o amdgpu_sync.o \
 	amdgpu_gtt_mgr.o amdgpu_vram_mgr.o amdgpu_virt.o amdgpu_atomfirmware.o \
 	amdgpu_vf_error.o amdgpu_sched.o amdgpu_debugfs.o amdgpu_ids.o \
-	amdgpu_gmc.o amdgpu_xgmi.o amdgpu_csa.o
+	amdgpu_gmc.o amdgpu_xgmi.o amdgpu_csa.o amdgpu_ras.o amdgpu_vm_cpu.o \
+	amdgpu_vm_sdma.o
 
 # add asic specific block
 amdgpu-$(CONFIG_DRM_AMDGPU_CIK)+= cik.o cik_ih.o kv_smc.o kv_dpm.o \
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 8d0d7f3dd5fb..14398f55f602 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -83,6 +83,7 @@
 #include "amdgpu_gem.h"
 #include "amdgpu_doorbell.h"
 #include "amdgpu_amdkfd.h"
+#include "amdgpu_smu.h"
 
 #define MAX_GPU_INSTANCE		16
 
@@ -156,6 +157,8 @@ extern int amdgpu_emu_mode;
 extern uint amdgpu_smu_memory_pool_size;
 extern uint amdgpu_dc_feature_mask;
 extern struct amdgpu_mgpu_info mgpu_info;
+extern int amdgpu_ras_enable;
+extern uint amdgpu_ras_mask;
 
 #ifdef CONFIG_DRM_AMDGPU_SI
 extern int amdgpu_si_support;
@@ -433,6 +436,12 @@ struct amdgpu_cs_chunk {
 	void			*kdata;
 };
 
+struct amdgpu_cs_post_dep {
+	struct drm_syncobj *syncobj;
+	struct dma_fence_chain *chain;
+	u64 point;
+};
+
 struct amdgpu_cs_parser {
 	struct amdgpu_device	*adev;
 	struct drm_file		*filp;
@@ -462,8 +471,8 @@ struct amdgpu_cs_parser {
 	/* user fence */
 	struct amdgpu_bo_list_entry	uf_entry;
 
-	unsigned num_post_dep_syncobjs;
-	struct drm_syncobj **post_dep_syncobjs;
+	unsigned			num_post_deps;
+	struct amdgpu_cs_post_dep	*post_deps;
 };
 
 static inline u32 amdgpu_get_ib_value(struct amdgpu_cs_parser *p,
@@ -702,7 +711,6 @@ enum amd_hw_ip_block_type {
 struct amd_powerplay {
 	void *pp_handle;
 	const struct amd_pm_funcs *pp_funcs;
-	uint32_t pp_feature;
 };
 
 #define AMDGPU_RESET_MAGIC_NUM 64
@@ -825,6 +833,7 @@ struct amdgpu_device {
 	/* For pre-DCE11. DCE11 and later are in "struct amdgpu_device->dm" */
 	struct work_struct		hotplug_work;
 	struct amdgpu_irq_src		crtc_irq;
+	struct amdgpu_irq_src		vupdate_irq;
 	struct amdgpu_irq_src		pageflip_irq;
 	struct amdgpu_irq_src		hpd_irq;
 
@@ -842,6 +851,9 @@ struct amdgpu_device {
 	struct amd_powerplay		powerplay;
 	bool				pp_force_state_enabled;
 
+	/* smu */
+	struct smu_context		smu;
+
 	/* dpm */
 	struct amdgpu_pm		pm;
 	u32				cg_flags;
@@ -922,6 +934,8 @@ struct amdgpu_device {
 
 	int asic_reset_res;
 	struct work_struct		xgmi_reset_work;
+
+	bool                            in_baco_reset;
 };
 
 static inline struct amdgpu_device *amdgpu_ttm_adev(struct ttm_bo_device *bdev)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index fe1d7368c1e6..aeead072fa79 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -335,6 +335,43 @@ void amdgpu_amdkfd_free_gtt_mem(struct kgd_dev *kgd, void *mem_obj)
 	amdgpu_bo_unref(&(bo));
 }
 
+uint32_t amdgpu_amdkfd_get_fw_version(struct kgd_dev *kgd,
+				      enum kgd_engine_type type)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
+
+	switch (type) {
+	case KGD_ENGINE_PFP:
+		return adev->gfx.pfp_fw_version;
+
+	case KGD_ENGINE_ME:
+		return adev->gfx.me_fw_version;
+
+	case KGD_ENGINE_CE:
+		return adev->gfx.ce_fw_version;
+
+	case KGD_ENGINE_MEC1:
+		return adev->gfx.mec_fw_version;
+
+	case KGD_ENGINE_MEC2:
+		return adev->gfx.mec2_fw_version;
+
+	case KGD_ENGINE_RLC:
+		return adev->gfx.rlc_fw_version;
+
+	case KGD_ENGINE_SDMA1:
+		return adev->sdma.instance[0].fw_version;
+
+	case KGD_ENGINE_SDMA2:
+		return adev->sdma.instance[1].fw_version;
+
+	default:
+		return 0;
+	}
+
+	return 0;
+}
+
 void amdgpu_amdkfd_get_local_mem_info(struct kgd_dev *kgd,
 				      struct kfd_local_mem_info *mem_info)
 {
@@ -640,4 +677,8 @@ int kgd2kfd_post_reset(struct kfd_dev *kfd)
 void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry)
 {
 }
+
+void kgd2kfd_set_sram_ecc_flag(struct kfd_dev *kfd)
+{
+}
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 0b31a1859023..4e37fa7e85b1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -81,6 +81,18 @@ struct amdgpu_kfd_dev {
 	uint64_t vram_used;
 };
 
+enum kgd_engine_type {
+	KGD_ENGINE_PFP = 1,
+	KGD_ENGINE_ME,
+	KGD_ENGINE_CE,
+	KGD_ENGINE_MEC1,
+	KGD_ENGINE_MEC2,
+	KGD_ENGINE_RLC,
+	KGD_ENGINE_SDMA1,
+	KGD_ENGINE_SDMA2,
+	KGD_ENGINE_MAX
+};
+
 struct amdgpu_amdkfd_fence *amdgpu_amdkfd_fence_create(u64 context,
 						       struct mm_struct *mm);
 bool amdkfd_fence_check_mm(struct dma_fence *f, struct mm_struct *mm);
@@ -142,6 +154,8 @@ int amdgpu_amdkfd_alloc_gtt_mem(struct kgd_dev *kgd, size_t size,
 				void **mem_obj, uint64_t *gpu_addr,
 				void **cpu_ptr, bool mqd_gfx9);
 void amdgpu_amdkfd_free_gtt_mem(struct kgd_dev *kgd, void *mem_obj);
+uint32_t amdgpu_amdkfd_get_fw_version(struct kgd_dev *kgd,
+				      enum kgd_engine_type type);
 void amdgpu_amdkfd_get_local_mem_info(struct kgd_dev *kgd,
 				      struct kfd_local_mem_info *mem_info);
 uint64_t amdgpu_amdkfd_get_gpu_clock_counter(struct kgd_dev *kgd);
@@ -230,5 +244,6 @@ int kgd2kfd_quiesce_mm(struct mm_struct *mm);
 int kgd2kfd_resume_mm(struct mm_struct *mm);
 int kgd2kfd_schedule_evict_and_restore_process(struct mm_struct *mm,
 					       struct dma_fence *fence);
+void kgd2kfd_set_sram_ecc_flag(struct kfd_dev *kfd);
 
 #endif /* AMDGPU_AMDKFD_H_INCLUDED */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
index ff7fac7df34b..fa09e11a600c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
@@ -22,14 +22,12 @@
 
 #include <linux/fdtable.h>
 #include <linux/uaccess.h>
-#include <linux/firmware.h>
 #include <linux/mmu_context.h>
 #include <drm/drmP.h>
 #include "amdgpu.h"
 #include "amdgpu_amdkfd.h"
 #include "cikd.h"
 #include "cik_sdma.h"
-#include "amdgpu_ucode.h"
 #include "gfx_v7_0.h"
 #include "gca/gfx_7_2_d.h"
 #include "gca/gfx_7_2_enum.h"
@@ -139,7 +137,6 @@ static bool get_atc_vmid_pasid_mapping_valid(struct kgd_dev *kgd, uint8_t vmid);
 static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd,
 							uint8_t vmid);
 
-static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type);
 static void set_scratch_backing_va(struct kgd_dev *kgd,
 					uint64_t va, uint32_t vmid);
 static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
@@ -191,7 +188,6 @@ static const struct kfd2kgd_calls kfd2kgd = {
 	.address_watch_get_offset = kgd_address_watch_get_offset,
 	.get_atc_vmid_pasid_mapping_pasid = get_atc_vmid_pasid_mapping_pasid,
 	.get_atc_vmid_pasid_mapping_valid = get_atc_vmid_pasid_mapping_valid,
-	.get_fw_version = get_fw_version,
 	.set_scratch_backing_va = set_scratch_backing_va,
 	.get_tile_config = get_tile_config,
 	.set_vm_context_page_table_base = set_vm_context_page_table_base,
@@ -792,63 +788,6 @@ static void set_scratch_backing_va(struct kgd_dev *kgd,
 	unlock_srbm(kgd);
 }
 
-static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *) kgd;
-	const union amdgpu_firmware_header *hdr;
-
-	switch (type) {
-	case KGD_ENGINE_PFP:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.pfp_fw->data;
-		break;
-
-	case KGD_ENGINE_ME:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.me_fw->data;
-		break;
-
-	case KGD_ENGINE_CE:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.ce_fw->data;
-		break;
-
-	case KGD_ENGINE_MEC1:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.mec_fw->data;
-		break;
-
-	case KGD_ENGINE_MEC2:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.mec2_fw->data;
-		break;
-
-	case KGD_ENGINE_RLC:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.rlc_fw->data;
-		break;
-
-	case KGD_ENGINE_SDMA1:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->sdma.instance[0].fw->data;
-		break;
-
-	case KGD_ENGINE_SDMA2:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->sdma.instance[1].fw->data;
-		break;
-
-	default:
-		return 0;
-	}
-
-	if (hdr == NULL)
-		return 0;
-
-	/* Only 12 bit in use*/
-	return hdr->common.ucode_version;
-}
-
 static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
 			uint64_t page_table_base)
 {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
index 56ea929f524b..fec3a6aa1de6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
@@ -23,12 +23,10 @@
 #include <linux/module.h>
 #include <linux/fdtable.h>
 #include <linux/uaccess.h>
-#include <linux/firmware.h>
 #include <linux/mmu_context.h>
 #include <drm/drmP.h>
 #include "amdgpu.h"
 #include "amdgpu_amdkfd.h"
-#include "amdgpu_ucode.h"
 #include "gfx_v8_0.h"
 #include "gca/gfx_8_0_sh_mask.h"
 #include "gca/gfx_8_0_d.h"
@@ -95,7 +93,6 @@ static bool get_atc_vmid_pasid_mapping_valid(struct kgd_dev *kgd,
 		uint8_t vmid);
 static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd,
 		uint8_t vmid);
-static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type);
 static void set_scratch_backing_va(struct kgd_dev *kgd,
 					uint64_t va, uint32_t vmid);
 static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
@@ -148,7 +145,6 @@ static const struct kfd2kgd_calls kfd2kgd = {
 			get_atc_vmid_pasid_mapping_pasid,
 	.get_atc_vmid_pasid_mapping_valid =
 			get_atc_vmid_pasid_mapping_valid,
-	.get_fw_version = get_fw_version,
 	.set_scratch_backing_va = set_scratch_backing_va,
 	.get_tile_config = get_tile_config,
 	.set_vm_context_page_table_base = set_vm_context_page_table_base,
@@ -751,63 +747,6 @@ static void set_scratch_backing_va(struct kgd_dev *kgd,
 	unlock_srbm(kgd);
 }
 
-static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *) kgd;
-	const union amdgpu_firmware_header *hdr;
-
-	switch (type) {
-	case KGD_ENGINE_PFP:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.pfp_fw->data;
-		break;
-
-	case KGD_ENGINE_ME:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.me_fw->data;
-		break;
-
-	case KGD_ENGINE_CE:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.ce_fw->data;
-		break;
-
-	case KGD_ENGINE_MEC1:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.mec_fw->data;
-		break;
-
-	case KGD_ENGINE_MEC2:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.mec2_fw->data;
-		break;
-
-	case KGD_ENGINE_RLC:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.rlc_fw->data;
-		break;
-
-	case KGD_ENGINE_SDMA1:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->sdma.instance[0].fw->data;
-		break;
-
-	case KGD_ENGINE_SDMA2:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->sdma.instance[1].fw->data;
-		break;
-
-	default:
-		return 0;
-	}
-
-	if (hdr == NULL)
-		return 0;
-
-	/* Only 12 bit in use*/
-	return hdr->common.ucode_version;
-}
-
 static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
 		uint64_t page_table_base)
 {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
index 5c51d4910650..ef3d93b995b2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
@@ -25,12 +25,10 @@
 #include <linux/module.h>
 #include <linux/fdtable.h>
 #include <linux/uaccess.h>
-#include <linux/firmware.h>
 #include <linux/mmu_context.h>
 #include <drm/drmP.h>
 #include "amdgpu.h"
 #include "amdgpu_amdkfd.h"
-#include "amdgpu_ucode.h"
 #include "soc15_hw_ip.h"
 #include "gc/gc_9_0_offset.h"
 #include "gc/gc_9_0_sh_mask.h"
@@ -111,7 +109,6 @@ static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd,
 		uint8_t vmid);
 static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
 		uint64_t page_table_base);
-static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type);
 static void set_scratch_backing_va(struct kgd_dev *kgd,
 					uint64_t va, uint32_t vmid);
 static int invalidate_tlbs(struct kgd_dev *kgd, uint16_t pasid);
@@ -158,7 +155,6 @@ static const struct kfd2kgd_calls kfd2kgd = {
 			get_atc_vmid_pasid_mapping_pasid,
 	.get_atc_vmid_pasid_mapping_valid =
 			get_atc_vmid_pasid_mapping_valid,
-	.get_fw_version = get_fw_version,
 	.set_scratch_backing_va = set_scratch_backing_va,
 	.get_tile_config = amdgpu_amdkfd_get_tile_config,
 	.set_vm_context_page_table_base = set_vm_context_page_table_base,
@@ -874,56 +870,6 @@ static void set_scratch_backing_va(struct kgd_dev *kgd,
 	 */
 }
 
-/* FIXME: Does this need to be ASIC-specific code? */
-static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *) kgd;
-	const union amdgpu_firmware_header *hdr;
-
-	switch (type) {
-	case KGD_ENGINE_PFP:
-		hdr = (const union amdgpu_firmware_header *)adev->gfx.pfp_fw->data;
-		break;
-
-	case KGD_ENGINE_ME:
-		hdr = (const union amdgpu_firmware_header *)adev->gfx.me_fw->data;
-		break;
-
-	case KGD_ENGINE_CE:
-		hdr = (const union amdgpu_firmware_header *)adev->gfx.ce_fw->data;
-		break;
-
-	case KGD_ENGINE_MEC1:
-		hdr = (const union amdgpu_firmware_header *)adev->gfx.mec_fw->data;
-		break;
-
-	case KGD_ENGINE_MEC2:
-		hdr = (const union amdgpu_firmware_header *)adev->gfx.mec2_fw->data;
-		break;
-
-	case KGD_ENGINE_RLC:
-		hdr = (const union amdgpu_firmware_header *)adev->gfx.rlc_fw->data;
-		break;
-
-	case KGD_ENGINE_SDMA1:
-		hdr = (const union amdgpu_firmware_header *)adev->sdma.instance[0].fw->data;
-		break;
-
-	case KGD_ENGINE_SDMA2:
-		hdr = (const union amdgpu_firmware_header *)adev->sdma.instance[1].fw->data;
-		break;
-
-	default:
-		return 0;
-	}
-
-	if (hdr == NULL)
-		return 0;
-
-	/* Only 12 bit in use*/
-	return hdr->common.ucode_version;
-}
-
 static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
 		uint64_t page_table_base)
 {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index 1921dec3df7a..a6e5184d436c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -410,15 +410,7 @@ static int add_bo_to_vm(struct amdgpu_device *adev, struct kgd_mem *mem,
 	if (p_bo_va_entry)
 		*p_bo_va_entry = bo_va_entry;
 
-	/* Allocate new page tables if needed and validate
-	 * them.
-	 */
-	ret = amdgpu_vm_alloc_pts(adev, vm, va, amdgpu_bo_size(bo));
-	if (ret) {
-		pr_err("Failed to allocate pts, err=%d\n", ret);
-		goto err_alloc_pts;
-	}
-
+	/* Allocate validate page tables if needed */
 	ret = vm_validate_pt_pd_bos(vm);
 	if (ret) {
 		pr_err("validate_pt_pd_bos() failed\n");
@@ -741,13 +733,7 @@ static int update_gpuvm_pte(struct amdgpu_device *adev,
 		struct amdgpu_sync *sync)
 {
 	int ret;
-	struct amdgpu_vm *vm;
-	struct amdgpu_bo_va *bo_va;
-	struct amdgpu_bo *bo;
-
-	bo_va = entry->bo_va;
-	vm = bo_va->base.vm;
-	bo = bo_va->base.bo;
+	struct amdgpu_bo_va *bo_va = entry->bo_va;
 
 	/* Update the page tables  */
 	ret = amdgpu_vm_bo_update(adev, bo_va, false);
@@ -906,7 +892,8 @@ static int init_kfd_vm(struct amdgpu_vm *vm, void **process_info,
 		pr_err("validate_pt_pd_bos() failed\n");
 		goto validate_pd_fail;
 	}
-	amdgpu_bo_sync_wait(vm->root.base.bo, AMDGPU_FENCE_OWNER_KFD, false);
+	ret = amdgpu_bo_sync_wait(vm->root.base.bo,
+				  AMDGPU_FENCE_OWNER_KFD, false);
 	if (ret)
 		goto wait_pd_fail;
 	amdgpu_bo_fence(vm->root.base.bo,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c
index b61e1dc61b4c..f96d75c6e099 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c
@@ -28,8 +28,6 @@
 #include "atom.h"
 #include "atombios.h"
 
-#define get_index_into_master_table(master_table, table_name) (offsetof(struct master_table, table_name) / sizeof(uint16_t))
-
 bool amdgpu_atomfirmware_gpu_supports_virtualization(struct amdgpu_device *adev)
 {
 	int index = get_index_into_master_table(atom_master_list_of_data_tables_v2_1,
@@ -238,10 +236,71 @@ int amdgpu_atomfirmware_get_vram_type(struct amdgpu_device *adev)
 	return 0;
 }
 
+/*
+ * Return true if vbios enabled ecc by default, if umc info table is available
+ * or false if ecc is not enabled or umc info table is not available
+ */
+bool amdgpu_atomfirmware_mem_ecc_supported(struct amdgpu_device *adev)
+{
+	struct amdgpu_mode_info *mode_info = &adev->mode_info;
+	int index;
+	u16 data_offset, size;
+	union umc_info *umc_info;
+	u8 frev, crev;
+	bool ecc_default_enabled = false;
+
+	index = get_index_into_master_table(atom_master_list_of_data_tables_v2_1,
+			umc_info);
+
+	if (amdgpu_atom_parse_data_header(mode_info->atom_context,
+				index, &size, &frev, &crev, &data_offset)) {
+		/* support umc_info 3.1+ */
+		if ((frev == 3 && crev >= 1) || (frev > 3)) {
+			umc_info = (union umc_info *)
+				(mode_info->atom_context->bios + data_offset);
+			ecc_default_enabled =
+				(le32_to_cpu(umc_info->v31.umc_config) &
+				 UMC_CONFIG__DEFAULT_MEM_ECC_ENABLE) ? true : false;
+		}
+	}
+
+	return ecc_default_enabled;
+}
+
 union firmware_info {
 	struct atom_firmware_info_v3_1 v31;
 };
 
+/*
+ * Return true if vbios supports sram ecc or false if not
+ */
+bool amdgpu_atomfirmware_sram_ecc_supported(struct amdgpu_device *adev)
+{
+	struct amdgpu_mode_info *mode_info = &adev->mode_info;
+	int index;
+	u16 data_offset, size;
+	union firmware_info *firmware_info;
+	u8 frev, crev;
+	bool sram_ecc_supported = false;
+
+	index = get_index_into_master_table(atom_master_list_of_data_tables_v2_1,
+			firmwareinfo);
+
+	if (amdgpu_atom_parse_data_header(adev->mode_info.atom_context,
+				index, &size, &frev, &crev, &data_offset)) {
+		/* support firmware_info 3.1 + */
+		if ((frev == 3 && crev >=1) || (frev > 3)) {
+			firmware_info = (union firmware_info *)
+				(mode_info->atom_context->bios + data_offset);
+			sram_ecc_supported =
+				(le32_to_cpu(firmware_info->v31.firmware_capability) &
+				 ATOM_FIRMWARE_CAP_SRAM_ECC) ? true : false;
+		}
+	}
+
+	return sram_ecc_supported;
+}
+
 union smu_info {
 	struct atom_smu_info_v3_1 v31;
 };
@@ -346,11 +405,11 @@ int amdgpu_atomfirmware_get_gfx_info(struct amdgpu_device *adev)
 			(mode_info->atom_context->bios + data_offset);
 		switch (crev) {
 		case 4:
-			adev->gfx.config.max_shader_engines = gfx_info->v24.gc_num_se;
-			adev->gfx.config.max_cu_per_sh = gfx_info->v24.gc_num_cu_per_sh;
-			adev->gfx.config.max_sh_per_se = gfx_info->v24.gc_num_sh_per_se;
-			adev->gfx.config.max_backends_per_se = gfx_info->v24.gc_num_rb_per_se;
-			adev->gfx.config.max_texture_channel_caches = gfx_info->v24.gc_num_tccs;
+			adev->gfx.config.max_shader_engines = gfx_info->v24.max_shader_engines;
+			adev->gfx.config.max_cu_per_sh = gfx_info->v24.max_cu_per_sh;
+			adev->gfx.config.max_sh_per_se = gfx_info->v24.max_sh_per_se;
+			adev->gfx.config.max_backends_per_se = gfx_info->v24.max_backends_per_se;
+			adev->gfx.config.max_texture_channel_caches = gfx_info->v24.max_texture_channel_caches;
 			adev->gfx.config.max_gprs = le16_to_cpu(gfx_info->v24.gc_num_gprs);
 			adev->gfx.config.max_gs_threads = gfx_info->v24.gc_num_max_gs_thds;
 			adev->gfx.config.gs_vgt_table_depth = gfx_info->v24.gc_gs_table_depth;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.h
index 20f158fd3b76..5ec6f92f353c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.h
@@ -24,6 +24,8 @@
 #ifndef __AMDGPU_ATOMFIRMWARE_H__
 #define __AMDGPU_ATOMFIRMWARE_H__
 
+#define get_index_into_master_table(master_table, table_name) (offsetof(struct master_table, table_name) / sizeof(uint16_t))
+
 bool amdgpu_atomfirmware_gpu_supports_virtualization(struct amdgpu_device *adev);
 void amdgpu_atomfirmware_scratch_regs_init(struct amdgpu_device *adev);
 int amdgpu_atomfirmware_allocate_fb_scratch(struct amdgpu_device *adev);
@@ -31,5 +33,7 @@ int amdgpu_atomfirmware_get_vram_width(struct amdgpu_device *adev);
 int amdgpu_atomfirmware_get_vram_type(struct amdgpu_device *adev);
 int amdgpu_atomfirmware_get_clock_info(struct amdgpu_device *adev);
 int amdgpu_atomfirmware_get_gfx_info(struct amdgpu_device *adev);
+bool amdgpu_atomfirmware_mem_ecc_supported(struct amdgpu_device *adev);
+bool amdgpu_atomfirmware_sram_ecc_supported(struct amdgpu_device *adev);
 
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 52a5e4fdc95b..2f6239b6be6f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -215,6 +215,8 @@ static int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, union drm_amdgpu_cs
 		case AMDGPU_CHUNK_ID_SYNCOBJ_IN:
 		case AMDGPU_CHUNK_ID_SYNCOBJ_OUT:
 		case AMDGPU_CHUNK_ID_SCHEDULED_DEPENDENCIES:
+		case AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_WAIT:
+		case AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_SIGNAL:
 			break;
 
 		default:
@@ -804,9 +806,11 @@ static void amdgpu_cs_parser_fini(struct amdgpu_cs_parser *parser, int error,
 		ttm_eu_backoff_reservation(&parser->ticket,
 					   &parser->validated);
 
-	for (i = 0; i < parser->num_post_dep_syncobjs; i++)
-		drm_syncobj_put(parser->post_dep_syncobjs[i]);
-	kfree(parser->post_dep_syncobjs);
+	for (i = 0; i < parser->num_post_deps; i++) {
+		drm_syncobj_put(parser->post_deps[i].syncobj);
+		kfree(parser->post_deps[i].chain);
+	}
+	kfree(parser->post_deps);
 
 	dma_fence_put(parser->fence);
 
@@ -1117,13 +1121,18 @@ static int amdgpu_cs_process_fence_dep(struct amdgpu_cs_parser *p,
 }
 
 static int amdgpu_syncobj_lookup_and_add_to_sync(struct amdgpu_cs_parser *p,
-						 uint32_t handle)
+						 uint32_t handle, u64 point,
+						 u64 flags)
 {
-	int r;
 	struct dma_fence *fence;
-	r = drm_syncobj_find_fence(p->filp, handle, 0, 0, &fence);
-	if (r)
+	int r;
+
+	r = drm_syncobj_find_fence(p->filp, handle, point, flags, &fence);
+	if (r) {
+		DRM_ERROR("syncobj %u failed to find fence @ %llu (%d)!\n",
+			  handle, point, r);
 		return r;
+	}
 
 	r = amdgpu_sync_fence(p->adev, &p->job->sync, fence, true);
 	dma_fence_put(fence);
@@ -1134,46 +1143,118 @@ static int amdgpu_syncobj_lookup_and_add_to_sync(struct amdgpu_cs_parser *p,
 static int amdgpu_cs_process_syncobj_in_dep(struct amdgpu_cs_parser *p,
 					    struct amdgpu_cs_chunk *chunk)
 {
+	struct drm_amdgpu_cs_chunk_sem *deps;
 	unsigned num_deps;
 	int i, r;
-	struct drm_amdgpu_cs_chunk_sem *deps;
 
 	deps = (struct drm_amdgpu_cs_chunk_sem *)chunk->kdata;
 	num_deps = chunk->length_dw * 4 /
 		sizeof(struct drm_amdgpu_cs_chunk_sem);
+	for (i = 0; i < num_deps; ++i) {
+		r = amdgpu_syncobj_lookup_and_add_to_sync(p, deps[i].handle,
+							  0, 0);
+		if (r)
+			return r;
+	}
+
+	return 0;
+}
+
 
+static int amdgpu_cs_process_syncobj_timeline_in_dep(struct amdgpu_cs_parser *p,
+						     struct amdgpu_cs_chunk *chunk)
+{
+	struct drm_amdgpu_cs_chunk_syncobj *syncobj_deps;
+	unsigned num_deps;
+	int i, r;
+
+	syncobj_deps = (struct drm_amdgpu_cs_chunk_syncobj *)chunk->kdata;
+	num_deps = chunk->length_dw * 4 /
+		sizeof(struct drm_amdgpu_cs_chunk_syncobj);
 	for (i = 0; i < num_deps; ++i) {
-		r = amdgpu_syncobj_lookup_and_add_to_sync(p, deps[i].handle);
+		r = amdgpu_syncobj_lookup_and_add_to_sync(p,
+							  syncobj_deps[i].handle,
+							  syncobj_deps[i].point,
+							  syncobj_deps[i].flags);
 		if (r)
 			return r;
 	}
+
 	return 0;
 }
 
 static int amdgpu_cs_process_syncobj_out_dep(struct amdgpu_cs_parser *p,
 					     struct amdgpu_cs_chunk *chunk)
 {
+	struct drm_amdgpu_cs_chunk_sem *deps;
 	unsigned num_deps;
 	int i;
-	struct drm_amdgpu_cs_chunk_sem *deps;
+
 	deps = (struct drm_amdgpu_cs_chunk_sem *)chunk->kdata;
 	num_deps = chunk->length_dw * 4 /
 		sizeof(struct drm_amdgpu_cs_chunk_sem);
 
-	p->post_dep_syncobjs = kmalloc_array(num_deps,
-					     sizeof(struct drm_syncobj *),
-					     GFP_KERNEL);
-	p->num_post_dep_syncobjs = 0;
+	p->post_deps = kmalloc_array(num_deps, sizeof(*p->post_deps),
+				     GFP_KERNEL);
+	p->num_post_deps = 0;
 
-	if (!p->post_dep_syncobjs)
+	if (!p->post_deps)
 		return -ENOMEM;
 
+
 	for (i = 0; i < num_deps; ++i) {
-		p->post_dep_syncobjs[i] = drm_syncobj_find(p->filp, deps[i].handle);
-		if (!p->post_dep_syncobjs[i])
+		p->post_deps[i].syncobj =
+			drm_syncobj_find(p->filp, deps[i].handle);
+		if (!p->post_deps[i].syncobj)
 			return -EINVAL;
-		p->num_post_dep_syncobjs++;
+		p->post_deps[i].chain = NULL;
+		p->post_deps[i].point = 0;
+		p->num_post_deps++;
 	}
+
+	return 0;
+}
+
+
+static int amdgpu_cs_process_syncobj_timeline_out_dep(struct amdgpu_cs_parser *p,
+						      struct amdgpu_cs_chunk
+						      *chunk)
+{
+	struct drm_amdgpu_cs_chunk_syncobj *syncobj_deps;
+	unsigned num_deps;
+	int i;
+
+	syncobj_deps = (struct drm_amdgpu_cs_chunk_syncobj *)chunk->kdata;
+	num_deps = chunk->length_dw * 4 /
+		sizeof(struct drm_amdgpu_cs_chunk_syncobj);
+
+	p->post_deps = kmalloc_array(num_deps, sizeof(*p->post_deps),
+				     GFP_KERNEL);
+	p->num_post_deps = 0;
+
+	if (!p->post_deps)
+		return -ENOMEM;
+
+	for (i = 0; i < num_deps; ++i) {
+		struct amdgpu_cs_post_dep *dep = &p->post_deps[i];
+
+		dep->chain = NULL;
+		if (syncobj_deps[i].point) {
+			dep->chain = kmalloc(sizeof(*dep->chain), GFP_KERNEL);
+			if (!dep->chain)
+				return -ENOMEM;
+		}
+
+		dep->syncobj = drm_syncobj_find(p->filp,
+						syncobj_deps[i].handle);
+		if (!dep->syncobj) {
+			kfree(dep->chain);
+			return -EINVAL;
+		}
+		dep->point = syncobj_deps[i].point;
+		p->num_post_deps++;
+	}
+
 	return 0;
 }
 
@@ -1187,19 +1268,33 @@ static int amdgpu_cs_dependencies(struct amdgpu_device *adev,
 
 		chunk = &p->chunks[i];
 
-		if (chunk->chunk_id == AMDGPU_CHUNK_ID_DEPENDENCIES ||
-		    chunk->chunk_id == AMDGPU_CHUNK_ID_SCHEDULED_DEPENDENCIES) {
+		switch (chunk->chunk_id) {
+		case AMDGPU_CHUNK_ID_DEPENDENCIES:
+		case AMDGPU_CHUNK_ID_SCHEDULED_DEPENDENCIES:
 			r = amdgpu_cs_process_fence_dep(p, chunk);
 			if (r)
 				return r;
-		} else if (chunk->chunk_id == AMDGPU_CHUNK_ID_SYNCOBJ_IN) {
+			break;
+		case AMDGPU_CHUNK_ID_SYNCOBJ_IN:
 			r = amdgpu_cs_process_syncobj_in_dep(p, chunk);
 			if (r)
 				return r;
-		} else if (chunk->chunk_id == AMDGPU_CHUNK_ID_SYNCOBJ_OUT) {
+			break;
+		case AMDGPU_CHUNK_ID_SYNCOBJ_OUT:
 			r = amdgpu_cs_process_syncobj_out_dep(p, chunk);
 			if (r)
 				return r;
+			break;
+		case AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_WAIT:
+			r = amdgpu_cs_process_syncobj_timeline_in_dep(p, chunk);
+			if (r)
+				return r;
+			break;
+		case AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_SIGNAL:
+			r = amdgpu_cs_process_syncobj_timeline_out_dep(p, chunk);
+			if (r)
+				return r;
+			break;
 		}
 	}
 
@@ -1210,8 +1305,17 @@ static void amdgpu_cs_post_dependencies(struct amdgpu_cs_parser *p)
 {
 	int i;
 
-	for (i = 0; i < p->num_post_dep_syncobjs; ++i)
-		drm_syncobj_replace_fence(p->post_dep_syncobjs[i], p->fence);
+	for (i = 0; i < p->num_post_deps; ++i) {
+		if (p->post_deps[i].chain && p->post_deps[i].point) {
+			drm_syncobj_add_point(p->post_deps[i].syncobj,
+					      p->post_deps[i].chain,
+					      p->fence, p->post_deps[i].point);
+			p->post_deps[i].chain = NULL;
+		} else {
+			drm_syncobj_replace_fence(p->post_deps[i].syncobj,
+						  p->fence);
+		}
+	}
 }
 
 static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c
index 7e22be7ca68a..54dd02a898b9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c
@@ -92,15 +92,6 @@ int amdgpu_map_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 		return -ENOMEM;
 	}
 
-	r = amdgpu_vm_alloc_pts(adev, (*bo_va)->base.vm, csa_addr,
-				size);
-	if (r) {
-		DRM_ERROR("failed to allocate pts for static CSA, err=%d\n", r);
-		amdgpu_vm_bo_rmv(adev, *bo_va);
-		ttm_eu_backoff_reservation(&ticket, &list);
-		return r;
-	}
-
 	r = amdgpu_vm_bo_map(adev, *bo_va, csa_addr, 0, size,
 			     AMDGPU_PTE_READABLE | AMDGPU_PTE_WRITEABLE |
 			     AMDGPU_PTE_EXECUTABLE);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
index 7b526593eb77..a28a3d722ba2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
@@ -26,6 +26,7 @@
 #include <drm/drm_auth.h>
 #include "amdgpu.h"
 #include "amdgpu_sched.h"
+#include "amdgpu_ras.h"
 
 #define to_amdgpu_ctx_entity(e)	\
 	container_of((e), struct amdgpu_ctx_entity, entity)
@@ -344,6 +345,7 @@ static int amdgpu_ctx_query2(struct amdgpu_device *adev,
 {
 	struct amdgpu_ctx *ctx;
 	struct amdgpu_ctx_mgr *mgr;
+	uint32_t ras_counter;
 
 	if (!fpriv)
 		return -EINVAL;
@@ -368,6 +370,21 @@ static int amdgpu_ctx_query2(struct amdgpu_device *adev,
 	if (atomic_read(&ctx->guilty))
 		out->state.flags |= AMDGPU_CTX_QUERY2_FLAGS_GUILTY;
 
+	/*query ue count*/
+	ras_counter = amdgpu_ras_query_error_count(adev, false);
+	/*ras counter is monotonic increasing*/
+	if (ras_counter != ctx->ras_counter_ue) {
+		out->state.flags |= AMDGPU_CTX_QUERY2_FLAGS_RAS_UE;
+		ctx->ras_counter_ue = ras_counter;
+	}
+
+	/*query ce count*/
+	ras_counter = amdgpu_ras_query_error_count(adev, true);
+	if (ras_counter != ctx->ras_counter_ce) {
+		out->state.flags |= AMDGPU_CTX_QUERY2_FLAGS_RAS_CE;
+		ctx->ras_counter_ce = ras_counter;
+	}
+
 	mutex_unlock(&mgr->lock);
 	return 0;
 }
@@ -541,32 +558,26 @@ void amdgpu_ctx_mgr_init(struct amdgpu_ctx_mgr *mgr)
 	idr_init(&mgr->ctx_handles);
 }
 
-void amdgpu_ctx_mgr_entity_flush(struct amdgpu_ctx_mgr *mgr)
+long amdgpu_ctx_mgr_entity_flush(struct amdgpu_ctx_mgr *mgr, long timeout)
 {
 	unsigned num_entities = amdgput_ctx_total_num_entities();
 	struct amdgpu_ctx *ctx;
 	struct idr *idp;
 	uint32_t id, i;
-	long max_wait = MAX_WAIT_SCHED_ENTITY_Q_EMPTY;
 
 	idp = &mgr->ctx_handles;
 
 	mutex_lock(&mgr->lock);
 	idr_for_each_entry(idp, ctx, id) {
-
-		if (!ctx->adev) {
-			mutex_unlock(&mgr->lock);
-			return;
-		}
-
 		for (i = 0; i < num_entities; i++) {
 			struct drm_sched_entity *entity;
 
 			entity = &ctx->entities[0][i].entity;
-			max_wait = drm_sched_entity_flush(entity, max_wait);
+			timeout = drm_sched_entity_flush(entity, timeout);
 		}
 	}
 	mutex_unlock(&mgr->lock);
+	return timeout;
 }
 
 void amdgpu_ctx_mgr_entity_fini(struct amdgpu_ctx_mgr *mgr)
@@ -579,10 +590,6 @@ void amdgpu_ctx_mgr_entity_fini(struct amdgpu_ctx_mgr *mgr)
 	idp = &mgr->ctx_handles;
 
 	idr_for_each_entry(idp, ctx, id) {
-
-		if (!ctx->adev)
-			return;
-
 		if (kref_read(&ctx->refcount) != 1) {
 			DRM_ERROR("ctx %p is still alive\n", ctx);
 			continue;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h
index b3b012c0a7da..5f1b54c9bcdb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h
@@ -49,6 +49,8 @@ struct amdgpu_ctx {
 	enum drm_sched_priority		override_priority;
 	struct mutex			lock;
 	atomic_t			guilty;
+	uint32_t			ras_counter_ce;
+	uint32_t			ras_counter_ue;
 };
 
 struct amdgpu_ctx_mgr {
@@ -82,7 +84,7 @@ int amdgpu_ctx_wait_prev_fence(struct amdgpu_ctx *ctx,
 
 void amdgpu_ctx_mgr_init(struct amdgpu_ctx_mgr *mgr);
 void amdgpu_ctx_mgr_entity_fini(struct amdgpu_ctx_mgr *mgr);
-void amdgpu_ctx_mgr_entity_flush(struct amdgpu_ctx_mgr *mgr);
+long amdgpu_ctx_mgr_entity_flush(struct amdgpu_ctx_mgr *mgr, long timeout);
 void amdgpu_ctx_mgr_fini(struct amdgpu_ctx_mgr *mgr);
 
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
index 4ae3ff9a1d4c..8930d66f2204 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
@@ -568,10 +568,9 @@ static ssize_t amdgpu_debugfs_sensor_read(struct file *f, char __user *buf,
 	idx = *pos >> 2;
 
 	valuesize = sizeof(values);
-	if (adev->powerplay.pp_funcs && adev->powerplay.pp_funcs->read_sensor)
-		r = amdgpu_dpm_read_sensor(adev, idx, &values[0], &valuesize);
-	else
-		return -EINVAL;
+	r = amdgpu_dpm_read_sensor(adev, idx, &values[0], &valuesize);
+	if (r)
+		return r;
 
 	if (size > valuesize)
 		return -EINVAL;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 79fb302fb954..cc8ad3831982 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -60,6 +60,7 @@
 #include "amdgpu_pm.h"
 
 #include "amdgpu_xgmi.h"
+#include "amdgpu_ras.h"
 
 MODULE_FIRMWARE("amdgpu/vega10_gpu_info.bin");
 MODULE_FIRMWARE("amdgpu/vega12_gpu_info.bin");
@@ -1506,7 +1507,9 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev)
 			return -EAGAIN;
 	}
 
-	adev->powerplay.pp_feature = amdgpu_pp_feature_mask;
+	adev->pm.pp_feature = amdgpu_pp_feature_mask;
+	if (amdgpu_sriov_vf(adev))
+		adev->pm.pp_feature &= ~PP_GFXOFF_MASK;
 
 	for (i = 0; i < adev->num_ip_blocks; i++) {
 		if ((amdgpu_ip_block_mask & (1 << i)) == 0) {
@@ -1638,6 +1641,10 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
 {
 	int i, r;
 
+	r = amdgpu_ras_init(adev);
+	if (r)
+		return r;
+
 	for (i = 0; i < adev->num_ip_blocks; i++) {
 		if (!adev->ip_blocks[i].status.valid)
 			continue;
@@ -1681,6 +1688,13 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
 		}
 	}
 
+	r = amdgpu_ib_pool_init(adev);
+	if (r) {
+		dev_err(adev->dev, "IB initialization failed (%d).\n", r);
+		amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_IB_INIT_FAIL, 0, r);
+		goto init_failed;
+	}
+
 	r = amdgpu_ucode_create_bo(adev); /* create ucode bo when sw_init complete*/
 	if (r)
 		goto init_failed;
@@ -1869,6 +1883,8 @@ static int amdgpu_device_ip_fini(struct amdgpu_device *adev)
 {
 	int i, r;
 
+	amdgpu_ras_pre_fini(adev);
+
 	if (adev->gmc.xgmi.num_physical_nodes > 1)
 		amdgpu_xgmi_remove_device(adev);
 
@@ -1917,6 +1933,7 @@ static int amdgpu_device_ip_fini(struct amdgpu_device *adev)
 			amdgpu_free_static_csa(&adev->virt.csa_obj);
 			amdgpu_device_wb_fini(adev);
 			amdgpu_device_vram_scratch_fini(adev);
+			amdgpu_ib_pool_fini(adev);
 		}
 
 		r = adev->ip_blocks[i].version->funcs->sw_fini((void *)adev);
@@ -1937,6 +1954,8 @@ static int amdgpu_device_ip_fini(struct amdgpu_device *adev)
 		adev->ip_blocks[i].status.late_initialized = false;
 	}
 
+	amdgpu_ras_fini(adev);
+
 	if (amdgpu_sriov_vf(adev))
 		if (amdgpu_virt_release_full_gpu(adev, false))
 			DRM_ERROR("failed to release exclusive mode on fini\n");
@@ -1999,6 +2018,10 @@ static void amdgpu_device_ip_late_init_func_handler(struct work_struct *work)
 	r = amdgpu_device_enable_mgpu_fan_boost();
 	if (r)
 		DRM_ERROR("enable mgpu fan boost failed (%d).\n", r);
+
+	/*set to low pstate by default */
+	amdgpu_xgmi_set_pstate(adev, 0);
+
 }
 
 static void amdgpu_device_delay_enable_gfx_off(struct work_struct *work)
@@ -2369,7 +2392,7 @@ static void amdgpu_device_xgmi_reset_func(struct work_struct *__work)
 
 	adev->asic_reset_res =  amdgpu_asic_reset(adev);
 	if (adev->asic_reset_res)
-		DRM_WARN("ASIC reset failed with err r, %d for drm dev, %s",
+		DRM_WARN("ASIC reset failed with error, %d for drm dev, %s",
 			 adev->asic_reset_res, adev->ddev->unique);
 }
 
@@ -2448,6 +2471,7 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 	mutex_init(&adev->virt.vf_errors.lock);
 	hash_init(adev->mn_hash);
 	mutex_init(&adev->lock_reset);
+	mutex_init(&adev->virt.dpm_mutex);
 
 	amdgpu_device_check_arguments(adev);
 
@@ -2642,13 +2666,6 @@ fence_driver_init:
 	/* Get a log2 for easy divisions. */
 	adev->mm_stats.log2_max_MBps = ilog2(max(1u, max_MBps));
 
-	r = amdgpu_ib_pool_init(adev);
-	if (r) {
-		dev_err(adev->dev, "IB initialization failed (%d).\n", r);
-		amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_IB_INIT_FAIL, 0, r);
-		goto failed;
-	}
-
 	amdgpu_fbdev_init(adev);
 
 	r = amdgpu_pm_sysfs_init(adev);
@@ -2694,6 +2711,9 @@ fence_driver_init:
 		goto failed;
 	}
 
+	/* must succeed. */
+	amdgpu_ras_post_init(adev);
+
 	return 0;
 
 failed:
@@ -2726,7 +2746,6 @@ void amdgpu_device_fini(struct amdgpu_device *adev)
 		else
 			drm_atomic_helper_shutdown(adev->ddev);
 	}
-	amdgpu_ib_pool_fini(adev);
 	amdgpu_fence_driver_fini(adev);
 	amdgpu_pm_sysfs_fini(adev);
 	amdgpu_fbdev_fini(adev);
@@ -3225,6 +3244,8 @@ static int amdgpu_device_reset_sriov(struct amdgpu_device *adev,
 	if (r)
 		return r;
 
+	amdgpu_amdkfd_pre_reset(adev);
+
 	/* Resume IP prior to SMC */
 	r = amdgpu_device_ip_reinit_early_sriov(adev);
 	if (r)
@@ -3244,6 +3265,7 @@ static int amdgpu_device_reset_sriov(struct amdgpu_device *adev,
 
 	amdgpu_irq_gpu_reset_resume_helper(adev);
 	r = amdgpu_ib_ring_tests(adev);
+	amdgpu_amdkfd_post_reset(adev);
 
 error:
 	amdgpu_virt_init_data_exchange(adev);
@@ -3376,7 +3398,7 @@ static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive,
 				r = amdgpu_asic_reset(tmp_adev);
 
 			if (r) {
-				DRM_ERROR("ASIC reset failed with err r, %d for drm dev, %s",
+				DRM_ERROR("ASIC reset failed with error, %d for drm dev, %s",
 					 r, tmp_adev->ddev->unique);
 				break;
 			}
@@ -3393,6 +3415,11 @@ static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive,
 						break;
 				}
 			}
+
+			list_for_each_entry(tmp_adev, device_list_handle,
+					gmc.xgmi.head) {
+				amdgpu_ras_reserve_bad_pages(tmp_adev);
+			}
 		}
 	}
 
@@ -3411,7 +3438,7 @@ static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive,
 
 				vram_lost = amdgpu_device_check_vram_lost(tmp_adev);
 				if (vram_lost) {
-					DRM_ERROR("VRAM is lost!\n");
+					DRM_INFO("VRAM is lost due to GPU reset!\n");
 					atomic_inc(&tmp_adev->vram_lost_counter);
 				}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c
index 344967df3137..523b8ab6b04e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c
@@ -904,3 +904,19 @@ amdgpu_get_vce_clock_state(void *handle, u32 idx)
 
 	return NULL;
 }
+
+int amdgpu_dpm_get_sclk(struct amdgpu_device *adev, bool low)
+{
+	if (is_support_sw_smu(adev))
+		return smu_get_sclk(&adev->smu, low);
+	else
+		return (adev)->powerplay.pp_funcs->get_sclk((adev)->powerplay.pp_handle, (low));
+}
+
+int amdgpu_dpm_get_mclk(struct amdgpu_device *adev, bool low)
+{
+	if (is_support_sw_smu(adev))
+		return smu_get_mclk(&adev->smu, low);
+	else
+		return (adev)->powerplay.pp_funcs->get_mclk((adev)->powerplay.pp_handle, (low));
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h
index e871e022c129..dca35407879d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h
@@ -260,9 +260,6 @@ enum amdgpu_pcie_gen {
 #define amdgpu_dpm_enable_bapm(adev, e) \
 		((adev)->powerplay.pp_funcs->enable_bapm((adev)->powerplay.pp_handle, (e)))
 
-#define amdgpu_dpm_read_sensor(adev, idx, value, size) \
-		((adev)->powerplay.pp_funcs->read_sensor((adev)->powerplay.pp_handle, (idx), (value), (size)))
-
 #define amdgpu_dpm_set_fan_control_mode(adev, m) \
 		((adev)->powerplay.pp_funcs->set_fan_control_mode((adev)->powerplay.pp_handle, (m)))
 
@@ -281,18 +278,18 @@ enum amdgpu_pcie_gen {
 #define amdgpu_dpm_set_fan_speed_rpm(adev, s) \
 		((adev)->powerplay.pp_funcs->set_fan_speed_rpm)((adev)->powerplay.pp_handle, (s))
 
-#define amdgpu_dpm_get_sclk(adev, l) \
-		((adev)->powerplay.pp_funcs->get_sclk((adev)->powerplay.pp_handle, (l)))
-
-#define amdgpu_dpm_get_mclk(adev, l)  \
-		((adev)->powerplay.pp_funcs->get_mclk((adev)->powerplay.pp_handle, (l)))
-
 #define amdgpu_dpm_force_performance_level(adev, l) \
 		((adev)->powerplay.pp_funcs->force_performance_level((adev)->powerplay.pp_handle, (l)))
 
 #define amdgpu_dpm_get_current_power_state(adev) \
 		((adev)->powerplay.pp_funcs->get_current_power_state((adev)->powerplay.pp_handle))
 
+#define amdgpu_smu_get_current_power_state(adev) \
+		((adev)->smu.ppt_funcs->get_current_power_state(&((adev)->smu)))
+
+#define amdgpu_smu_set_power_state(adev) \
+		((adev)->smu.ppt_funcs->set_power_state(&((adev)->smu)))
+
 #define amdgpu_dpm_get_pp_num_states(adev, data) \
 		((adev)->powerplay.pp_funcs->get_pp_num_states((adev)->powerplay.pp_handle, data))
 
@@ -448,6 +445,9 @@ struct amdgpu_pm {
 	uint32_t                smu_prv_buffer_size;
 	struct amdgpu_bo        *smu_prv_buffer;
 	bool ac_power;
+	/* powerplay feature */
+	uint32_t pp_feature;
+
 };
 
 #define R600_SSTU_DFLT                               0
@@ -486,6 +486,8 @@ void amdgpu_dpm_print_ps_status(struct amdgpu_device *adev,
 u32 amdgpu_dpm_get_vblank_time(struct amdgpu_device *adev);
 u32 amdgpu_dpm_get_vrefresh(struct amdgpu_device *adev);
 void amdgpu_dpm_get_active_displays(struct amdgpu_device *adev);
+int amdgpu_dpm_read_sensor(struct amdgpu_device *adev, enum amd_pp_sensors sensor,
+			   void *data, uint32_t *size);
 
 bool amdgpu_is_internal_thermal_sensor(enum amdgpu_int_thermal_type sensor);
 
@@ -504,4 +506,8 @@ enum amdgpu_pcie_gen amdgpu_get_pcie_gen_support(struct amdgpu_device *adev,
 struct amd_vce_state*
 amdgpu_get_vce_clock_state(void *handle, u32 idx);
 
+extern int amdgpu_dpm_get_sclk(struct amdgpu_device *adev, bool low);
+
+extern int amdgpu_dpm_get_mclk(struct amdgpu_device *adev, bool low);
+
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 7419ea8a388b..1e2cc9d68a05 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -74,9 +74,11 @@
  * - 3.28.0 - Add AMDGPU_CHUNK_ID_SCHEDULED_DEPENDENCIES
  * - 3.29.0 - Add AMDGPU_IB_FLAG_RESET_GDS_MAX_WAVE_ID
  * - 3.30.0 - Add AMDGPU_SCHED_OP_CONTEXT_PRIORITY_OVERRIDE.
+ * - 3.31.0 - Add support for per-flip tiling attribute changes with DC
+ * - 3.32.0 - Add syncobj timeline support to AMDGPU_CS.
  */
 #define KMS_DRIVER_MAJOR	3
-#define KMS_DRIVER_MINOR	30
+#define KMS_DRIVER_MINOR	32
 #define KMS_DRIVER_PATCHLEVEL	0
 
 int amdgpu_vram_limit = 0;
@@ -117,8 +119,8 @@ uint amdgpu_pg_mask = 0xffffffff;
 uint amdgpu_sdma_phase_quantum = 32;
 char *amdgpu_disable_cu = NULL;
 char *amdgpu_virtual_display = NULL;
-/* OverDrive(bit 14),gfxoff(bit 15),stutter mode(bit 17) disabled by default*/
-uint amdgpu_pp_feature_mask = 0xfffd3fff;
+/* OverDrive(bit 14) disabled by default*/
+uint amdgpu_pp_feature_mask = 0xffffbfff;
 int amdgpu_ngg = 0;
 int amdgpu_prim_buf_per_se = 0;
 int amdgpu_pos_buf_per_se = 0;
@@ -136,6 +138,8 @@ uint amdgpu_dc_feature_mask = 0;
 struct amdgpu_mgpu_info mgpu_info = {
 	.mutex = __MUTEX_INITIALIZER(mgpu_info.mutex),
 };
+int amdgpu_ras_enable = -1;
+uint amdgpu_ras_mask = 0xffffffff;
 
 /**
  * DOC: vramlimit (int)
@@ -495,6 +499,21 @@ MODULE_PARM_DESC(emu_mode, "Emulation mode, (1 = enable, 0 = disable)");
 module_param_named(emu_mode, amdgpu_emu_mode, int, 0444);
 
 /**
+ * DOC: ras_enable (int)
+ * Enable RAS features on the GPU (0 = disable, 1 = enable, -1 = auto (default))
+ */
+MODULE_PARM_DESC(ras_enable, "Enable RAS features on the GPU (0 = disable, 1 = enable, -1 = auto (default))");
+module_param_named(ras_enable, amdgpu_ras_enable, int, 0444);
+
+/**
+ * DOC: ras_mask (uint)
+ * Mask of RAS features to enable (default 0xffffffff), only valid when ras_enable == 1
+ * See the flags in drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
+ */
+MODULE_PARM_DESC(ras_mask, "Mask of RAS features to enable (default 0xffffffff), only valid when ras_enable == 1");
+module_param_named(ras_mask, amdgpu_ras_mask, uint, 0444);
+
+/**
  * DOC: si_support (int)
  * Set SI support driver. This parameter works after set config CONFIG_DRM_AMDGPU_SI. For SI asic, when radeon driver is enabled,
  * set value 0 to use radeon driver, while set value 1 to use amdgpu driver. The default is using radeon driver when it available,
@@ -974,6 +993,7 @@ amdgpu_pci_remove(struct pci_dev *pdev)
 
 	DRM_ERROR("Device removal is currently not supported outside of fbcon\n");
 	drm_dev_unplug(dev);
+	drm_dev_put(dev);
 	pci_disable_device(pdev);
 	pci_set_drvdata(pdev, NULL);
 }
@@ -1158,13 +1178,14 @@ static int amdgpu_flush(struct file *f, fl_owner_t id)
 {
 	struct drm_file *file_priv = f->private_data;
 	struct amdgpu_fpriv *fpriv = file_priv->driver_priv;
+	long timeout = MAX_WAIT_SCHED_ENTITY_Q_EMPTY;
 
-	amdgpu_ctx_mgr_entity_flush(&fpriv->ctx_mgr);
+	timeout = amdgpu_ctx_mgr_entity_flush(&fpriv->ctx_mgr, timeout);
+	timeout = amdgpu_vm_wait_idle(&fpriv->vm, timeout);
 
-	return 0;
+	return timeout >= 0 ? 0 : timeout;
 }
 
-
 static const struct file_operations amdgpu_driver_kms_fops = {
 	.owner = THIS_MODULE,
 	.open = drm_open,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c
index 5cbde74b97dd..e47609218839 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c
@@ -49,12 +49,11 @@
 static int
 amdgpufb_open(struct fb_info *info, int user)
 {
-	struct amdgpu_fbdev *rfbdev = info->par;
-	struct amdgpu_device *adev = rfbdev->adev;
-	int ret = pm_runtime_get_sync(adev->ddev->dev);
+	struct drm_fb_helper *fb_helper = info->par;
+	int ret = pm_runtime_get_sync(fb_helper->dev->dev);
 	if (ret < 0 && ret != -EACCES) {
-		pm_runtime_mark_last_busy(adev->ddev->dev);
-		pm_runtime_put_autosuspend(adev->ddev->dev);
+		pm_runtime_mark_last_busy(fb_helper->dev->dev);
+		pm_runtime_put_autosuspend(fb_helper->dev->dev);
 		return ret;
 	}
 	return 0;
@@ -63,11 +62,10 @@ amdgpufb_open(struct fb_info *info, int user)
 static int
 amdgpufb_release(struct fb_info *info, int user)
 {
-	struct amdgpu_fbdev *rfbdev = info->par;
-	struct amdgpu_device *adev = rfbdev->adev;
+	struct drm_fb_helper *fb_helper = info->par;
 
-	pm_runtime_mark_last_busy(adev->ddev->dev);
-	pm_runtime_put_autosuspend(adev->ddev->dev);
+	pm_runtime_mark_last_busy(fb_helper->dev->dev);
+	pm_runtime_put_autosuspend(fb_helper->dev->dev);
 	return 0;
 }
 
@@ -233,9 +231,6 @@ static int amdgpufb_create(struct drm_fb_helper *helper,
 		goto out;
 	}
 
-	info->par = rfbdev;
-	info->skip_vt_switch = true;
-
 	ret = amdgpu_display_framebuffer_init(adev->ddev, &rfbdev->rfb,
 					      &mode_cmd, gobj);
 	if (ret) {
@@ -248,10 +243,6 @@ static int amdgpufb_create(struct drm_fb_helper *helper,
 	/* setup helper */
 	rfbdev->helper.fb = fb;
 
-	strcpy(info->fix.id, "amdgpudrmfb");
-
-	drm_fb_helper_fill_fix(info, fb->pitches[0], fb->format->depth);
-
 	info->fbops = &amdgpufb_ops;
 
 	tmp = amdgpu_bo_gpu_offset(abo) - adev->gmc.vram_start;
@@ -260,7 +251,7 @@ static int amdgpufb_create(struct drm_fb_helper *helper,
 	info->screen_base = amdgpu_bo_kptr(abo);
 	info->screen_size = amdgpu_bo_size(abo);
 
-	drm_fb_helper_fill_var(info, &rfbdev->helper, sizes->fb_width, sizes->fb_height);
+	drm_fb_helper_fill_info(info, &rfbdev->helper, sizes);
 
 	/* setup aperture base/size for vesafb takeover */
 	info->apertures->ranges[0].base = adev->ddev->mode_config.fb_base;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index ee47c11e92ce..4dee2326b29c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -136,8 +136,9 @@ int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f,
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct amdgpu_fence *fence;
-	struct dma_fence *old, **ptr;
+	struct dma_fence __rcu **ptr;
 	uint32_t seq;
+	int r;
 
 	fence = kmem_cache_alloc(amdgpu_fence_slab, GFP_KERNEL);
 	if (fence == NULL)
@@ -153,15 +154,24 @@ int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f,
 			       seq, flags | AMDGPU_FENCE_FLAG_INT);
 
 	ptr = &ring->fence_drv.fences[seq & ring->fence_drv.num_fences_mask];
+	if (unlikely(rcu_dereference_protected(*ptr, 1))) {
+		struct dma_fence *old;
+
+		rcu_read_lock();
+		old = dma_fence_get_rcu_safe(ptr);
+		rcu_read_unlock();
+
+		if (old) {
+			r = dma_fence_wait(old, false);
+			dma_fence_put(old);
+			if (r)
+				return r;
+		}
+	}
+
 	/* This function can't be called concurrently anyway, otherwise
 	 * emitting the fence would mess up the hardware ring buffer.
 	 */
-	old = rcu_dereference_protected(*ptr, 1);
-	if (old && !dma_fence_is_signaled(old)) {
-		DRM_INFO("rcu slot is busy\n");
-		dma_fence_wait(old, false);
-	}
-
 	rcu_assign_pointer(*ptr, dma_fence_get(&fence->base));
 
 	*f = &fence->base;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index d21dd2f369da..d4fcf5475464 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -31,6 +31,7 @@
 #include <drm/amdgpu_drm.h>
 #include "amdgpu.h"
 #include "amdgpu_display.h"
+#include "amdgpu_xgmi.h"
 
 void amdgpu_gem_object_free(struct drm_gem_object *gobj)
 {
@@ -627,11 +628,6 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 
 	switch (args->operation) {
 	case AMDGPU_VA_OP_MAP:
-		r = amdgpu_vm_alloc_pts(adev, bo_va->base.vm, args->va_address,
-					args->map_size);
-		if (r)
-			goto error_backoff;
-
 		va_flags = amdgpu_gmc_get_pte_flags(adev, args->flags);
 		r = amdgpu_vm_bo_map(adev, bo_va, args->va_address,
 				     args->offset_in_bo, args->map_size,
@@ -647,11 +643,6 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 						args->map_size);
 		break;
 	case AMDGPU_VA_OP_REPLACE:
-		r = amdgpu_vm_alloc_pts(adev, bo_va->base.vm, args->va_address,
-					args->map_size);
-		if (r)
-			goto error_backoff;
-
 		va_flags = amdgpu_gmc_get_pte_flags(adev, args->flags);
 		r = amdgpu_vm_bo_replace_map(adev, bo_va, args->va_address,
 					     args->offset_in_bo, args->map_size,
@@ -678,6 +669,7 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,
 	struct amdgpu_device *adev = dev->dev_private;
 	struct drm_amdgpu_gem_op *args = data;
 	struct drm_gem_object *gobj;
+	struct amdgpu_vm_bo_base *base;
 	struct amdgpu_bo *robj;
 	int r;
 
@@ -716,6 +708,15 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,
 			amdgpu_bo_unreserve(robj);
 			break;
 		}
+		for (base = robj->vm_bo; base; base = base->next)
+			if (amdgpu_xgmi_same_hive(amdgpu_ttm_adev(robj->tbo.bdev),
+				amdgpu_ttm_adev(base->vm->root.base.bo->tbo.bdev))) {
+				r = -EINVAL;
+				amdgpu_bo_unreserve(robj);
+				goto out;
+			}
+
+
 		robj->preferred_domains = args->value & (AMDGPU_GEM_DOMAIN_VRAM |
 							AMDGPU_GEM_DOMAIN_GTT |
 							AMDGPU_GEM_DOMAIN_CPU);
@@ -745,17 +746,25 @@ int amdgpu_mode_dumb_create(struct drm_file *file_priv,
 	struct amdgpu_device *adev = dev->dev_private;
 	struct drm_gem_object *gobj;
 	uint32_t handle;
+	u64 flags = AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED;
 	u32 domain;
 	int r;
 
+	/*
+	 * The buffer returned from this function should be cleared, but
+	 * it can only be done if the ring is enabled or we'll fail to
+	 * create the buffer.
+	 */
+	if (adev->mman.buffer_funcs_enabled)
+		flags |= AMDGPU_GEM_CREATE_VRAM_CLEARED;
+
 	args->pitch = amdgpu_align_pitch(adev, args->width,
 					 DIV_ROUND_UP(args->bpp, 8), 0);
 	args->size = (u64)args->pitch * args->height;
 	args->size = ALIGN(args->size, PAGE_SIZE);
 	domain = amdgpu_bo_get_preferred_pin_domain(adev,
 				amdgpu_display_supported_domains(adev));
-	r = amdgpu_gem_object_create(adev, args->size, 0, domain,
-				     AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED,
+	r = amdgpu_gem_object_create(adev, args->size, 0, domain, flags,
 				     ttm_bo_type_device, NULL, &gobj);
 	if (r)
 		return -ENOMEM;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 97a60da62004..997932ebbb83 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -390,7 +390,7 @@ void amdgpu_gfx_compute_mqd_sw_fini(struct amdgpu_device *adev)
 
 void amdgpu_gfx_off_ctrl(struct amdgpu_device *adev, bool enable)
 {
-	if (!(adev->powerplay.pp_feature & PP_GFXOFF_MASK))
+	if (!(adev->pm.pp_feature & PP_GFXOFF_MASK))
 		return;
 
 	if (!adev->powerplay.pp_funcs || !adev->powerplay.pp_funcs->set_powergating_by_smu)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index f790e15bcd08..09fc53af3d35 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -258,6 +258,9 @@ struct amdgpu_gfx {
 	/* pipe reservation */
 	struct mutex			pipe_reserve_mutex;
 	DECLARE_BITMAP			(pipe_reserve_bitmap, AMDGPU_MAX_COMPUTE_QUEUES);
+
+	/*ras */
+	struct ras_common_if		*ras_if;
 };
 
 #define amdgpu_gfx_get_gpu_clock_counter(adev) (adev)->gfx.funcs->get_gpu_clock_counter((adev))
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
index d73367cab4f3..250d9212cc38 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
@@ -80,6 +80,33 @@ uint64_t amdgpu_gmc_pd_addr(struct amdgpu_bo *bo)
 }
 
 /**
+ * amdgpu_gmc_set_pte_pde - update the page tables using CPU
+ *
+ * @adev: amdgpu_device pointer
+ * @cpu_pt_addr: cpu address of the page table
+ * @gpu_page_idx: entry in the page table to update
+ * @addr: dst addr to write into pte/pde
+ * @flags: access flags
+ *
+ * Update the page tables using CPU.
+ */
+int amdgpu_gmc_set_pte_pde(struct amdgpu_device *adev, void *cpu_pt_addr,
+				uint32_t gpu_page_idx, uint64_t addr,
+				uint64_t flags)
+{
+	void __iomem *ptr = (void *)cpu_pt_addr;
+	uint64_t value;
+
+	/*
+	 * The following is for PTE only. GART does not have PDEs.
+	*/
+	value = addr & 0x0000FFFFFFFFF000ULL;
+	value |= flags;
+	writeq(value, ptr + (gpu_page_idx * 8));
+	return 0;
+}
+
+/**
  * amdgpu_gmc_agp_addr - return the address in the AGP address space
  *
  * @tbo: TTM BO which needs the address, must be in GTT domain
@@ -213,3 +240,58 @@ void amdgpu_gmc_agp_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc)
 	dev_info(adev->dev, "AGP: %lluM 0x%016llX - 0x%016llX\n",
 			mc->agp_size >> 20, mc->agp_start, mc->agp_end);
 }
+
+/**
+ * amdgpu_gmc_filter_faults - filter VM faults
+ *
+ * @adev: amdgpu device structure
+ * @addr: address of the VM fault
+ * @pasid: PASID of the process causing the fault
+ * @timestamp: timestamp of the fault
+ *
+ * Returns:
+ * True if the fault was filtered and should not be processed further.
+ * False if the fault is a new one and needs to be handled.
+ */
+bool amdgpu_gmc_filter_faults(struct amdgpu_device *adev, uint64_t addr,
+			      uint16_t pasid, uint64_t timestamp)
+{
+	struct amdgpu_gmc *gmc = &adev->gmc;
+
+	uint64_t stamp, key = addr << 4 | pasid;
+	struct amdgpu_gmc_fault *fault;
+	uint32_t hash;
+
+	/* If we don't have space left in the ring buffer return immediately */
+	stamp = max(timestamp, AMDGPU_GMC_FAULT_TIMEOUT + 1) -
+		AMDGPU_GMC_FAULT_TIMEOUT;
+	if (gmc->fault_ring[gmc->last_fault].timestamp >= stamp)
+		return true;
+
+	/* Try to find the fault in the hash */
+	hash = hash_64(key, AMDGPU_GMC_FAULT_HASH_ORDER);
+	fault = &gmc->fault_ring[gmc->fault_hash[hash].idx];
+	while (fault->timestamp >= stamp) {
+		uint64_t tmp;
+
+		if (fault->key == key)
+			return true;
+
+		tmp = fault->timestamp;
+		fault = &gmc->fault_ring[fault->next];
+
+		/* Check if the entry was reused */
+		if (fault->timestamp >= tmp)
+			break;
+	}
+
+	/* Add the fault to the ring */
+	fault = &gmc->fault_ring[gmc->last_fault];
+	fault->key = key;
+	fault->timestamp = timestamp;
+
+	/* And update the hash */
+	fault->next = gmc->fault_hash[hash].idx;
+	gmc->fault_hash[hash].idx = gmc->last_fault++;
+	return false;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
index 81e6070d255b..071145ac67b5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
@@ -43,9 +43,35 @@
  */
 #define AMDGPU_GMC_HOLE_MASK	0x0000ffffffffffffULL
 
+/*
+ * Ring size as power of two for the log of recent faults.
+ */
+#define AMDGPU_GMC_FAULT_RING_ORDER	8
+#define AMDGPU_GMC_FAULT_RING_SIZE	(1 << AMDGPU_GMC_FAULT_RING_ORDER)
+
+/*
+ * Hash size as power of two for the log of recent faults
+ */
+#define AMDGPU_GMC_FAULT_HASH_ORDER	8
+#define AMDGPU_GMC_FAULT_HASH_SIZE	(1 << AMDGPU_GMC_FAULT_HASH_ORDER)
+
+/*
+ * Number of IH timestamp ticks until a fault is considered handled
+ */
+#define AMDGPU_GMC_FAULT_TIMEOUT	5000ULL
+
 struct firmware;
 
 /*
+ * GMC page fault information
+ */
+struct amdgpu_gmc_fault {
+	uint64_t	timestamp;
+	uint64_t	next:AMDGPU_GMC_FAULT_RING_ORDER;
+	uint64_t	key:52;
+};
+
+/*
  * VMHUB structures, functions & helpers
  */
 struct amdgpu_vmhub {
@@ -71,12 +97,6 @@ struct amdgpu_gmc_funcs {
 	/* Change the VMID -> PASID mapping */
 	void (*emit_pasid_mapping)(struct amdgpu_ring *ring, unsigned vmid,
 				   unsigned pasid);
-	/* write pte/pde updates using the cpu */
-	int (*set_pte_pde)(struct amdgpu_device *adev,
-			   void *cpu_pt_addr, /* cpu addr of page table */
-			   uint32_t gpu_page_idx, /* pte/pde to update */
-			   uint64_t addr, /* addr to write into pte/pde */
-			   uint64_t flags); /* access flags */
 	/* enable/disable PRT support */
 	void (*set_prt)(struct amdgpu_device *adev, bool enable);
 	/* set pte flags based per asic */
@@ -147,15 +167,22 @@ struct amdgpu_gmc {
 	struct kfd_vm_fault_info *vm_fault_info;
 	atomic_t		vm_fault_info_updated;
 
+	struct amdgpu_gmc_fault	fault_ring[AMDGPU_GMC_FAULT_RING_SIZE];
+	struct {
+		uint64_t	idx:AMDGPU_GMC_FAULT_RING_ORDER;
+	} fault_hash[AMDGPU_GMC_FAULT_HASH_SIZE];
+	uint64_t		last_fault:AMDGPU_GMC_FAULT_RING_ORDER;
+
 	const struct amdgpu_gmc_funcs	*gmc_funcs;
 
 	struct amdgpu_xgmi xgmi;
+	struct amdgpu_irq_src	ecc_irq;
+	struct ras_common_if    *ras_if;
 };
 
 #define amdgpu_gmc_flush_gpu_tlb(adev, vmid, type) (adev)->gmc.gmc_funcs->flush_gpu_tlb((adev), (vmid), (type))
 #define amdgpu_gmc_emit_flush_gpu_tlb(r, vmid, addr) (r)->adev->gmc.gmc_funcs->emit_flush_gpu_tlb((r), (vmid), (addr))
 #define amdgpu_gmc_emit_pasid_mapping(r, vmid, pasid) (r)->adev->gmc.gmc_funcs->emit_pasid_mapping((r), (vmid), (pasid))
-#define amdgpu_gmc_set_pte_pde(adev, pt, idx, addr, flags) (adev)->gmc.gmc_funcs->set_pte_pde((adev), (pt), (idx), (addr), (flags))
 #define amdgpu_gmc_get_vm_pde(adev, level, dst, flags) (adev)->gmc.gmc_funcs->get_vm_pde((adev), (level), (dst), (flags))
 #define amdgpu_gmc_get_pte_flags(adev, flags) (adev)->gmc.gmc_funcs->get_vm_pte_flags((adev),(flags))
 
@@ -189,6 +216,9 @@ static inline uint64_t amdgpu_gmc_sign_extend(uint64_t addr)
 
 void amdgpu_gmc_get_pde_for_bo(struct amdgpu_bo *bo, int level,
 			       uint64_t *addr, uint64_t *flags);
+int amdgpu_gmc_set_pte_pde(struct amdgpu_device *adev, void *cpu_pt_addr,
+				uint32_t gpu_page_idx, uint64_t addr,
+				uint64_t flags);
 uint64_t amdgpu_gmc_pd_addr(struct amdgpu_bo *bo);
 uint64_t amdgpu_gmc_agp_addr(struct ttm_buffer_object *bo);
 void amdgpu_gmc_vram_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc,
@@ -197,5 +227,7 @@ void amdgpu_gmc_gart_location(struct amdgpu_device *adev,
 			      struct amdgpu_gmc *mc);
 void amdgpu_gmc_agp_location(struct amdgpu_device *adev,
 			     struct amdgpu_gmc *mc);
+bool amdgpu_gmc_filter_faults(struct amdgpu_device *adev, uint64_t addr,
+			      uint16_t pasid, uint64_t timestamp);
 
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c
index da7b1b92d9cf..62591d081856 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c
@@ -37,6 +37,47 @@ struct amdgpu_gtt_node {
 };
 
 /**
+ * DOC: mem_info_gtt_total
+ *
+ * The amdgpu driver provides a sysfs API for reporting current total size of
+ * the GTT.
+ * The file mem_info_gtt_total is used for this, and returns the total size of
+ * the GTT block, in bytes
+ */
+static ssize_t amdgpu_mem_info_gtt_total_show(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = ddev->dev_private;
+
+	return snprintf(buf, PAGE_SIZE, "%llu\n",
+			(adev->mman.bdev.man[TTM_PL_TT].size) * PAGE_SIZE);
+}
+
+/**
+ * DOC: mem_info_gtt_used
+ *
+ * The amdgpu driver provides a sysfs API for reporting current total amount of
+ * used GTT.
+ * The file mem_info_gtt_used is used for this, and returns the current used
+ * size of the GTT block, in bytes
+ */
+static ssize_t amdgpu_mem_info_gtt_used_show(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = ddev->dev_private;
+
+	return snprintf(buf, PAGE_SIZE, "%llu\n",
+			amdgpu_gtt_mgr_usage(&adev->mman.bdev.man[TTM_PL_TT]));
+}
+
+static DEVICE_ATTR(mem_info_gtt_total, S_IRUGO,
+	           amdgpu_mem_info_gtt_total_show, NULL);
+static DEVICE_ATTR(mem_info_gtt_used, S_IRUGO,
+	           amdgpu_mem_info_gtt_used_show, NULL);
+
+/**
  * amdgpu_gtt_mgr_init - init GTT manager and DRM MM
  *
  * @man: TTM memory type manager
@@ -50,6 +91,7 @@ static int amdgpu_gtt_mgr_init(struct ttm_mem_type_manager *man,
 	struct amdgpu_device *adev = amdgpu_ttm_adev(man->bdev);
 	struct amdgpu_gtt_mgr *mgr;
 	uint64_t start, size;
+	int ret;
 
 	mgr = kzalloc(sizeof(*mgr), GFP_KERNEL);
 	if (!mgr)
@@ -61,6 +103,18 @@ static int amdgpu_gtt_mgr_init(struct ttm_mem_type_manager *man,
 	spin_lock_init(&mgr->lock);
 	atomic64_set(&mgr->available, p_size);
 	man->priv = mgr;
+
+	ret = device_create_file(adev->dev, &dev_attr_mem_info_gtt_total);
+	if (ret) {
+		DRM_ERROR("Failed to create device file mem_info_gtt_total\n");
+		return ret;
+	}
+	ret = device_create_file(adev->dev, &dev_attr_mem_info_gtt_used);
+	if (ret) {
+		DRM_ERROR("Failed to create device file mem_info_gtt_used\n");
+		return ret;
+	}
+
 	return 0;
 }
 
@@ -74,12 +128,17 @@ static int amdgpu_gtt_mgr_init(struct ttm_mem_type_manager *man,
  */
 static int amdgpu_gtt_mgr_fini(struct ttm_mem_type_manager *man)
 {
+	struct amdgpu_device *adev = amdgpu_ttm_adev(man->bdev);
 	struct amdgpu_gtt_mgr *mgr = man->priv;
 	spin_lock(&mgr->lock);
 	drm_mm_takedown(&mgr->mm);
 	spin_unlock(&mgr->lock);
 	kfree(mgr);
 	man->priv = NULL;
+
+	device_remove_file(adev->dev, &dev_attr_mem_info_gtt_total);
+	device_remove_file(adev->dev, &dev_attr_mem_info_gtt_used);
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
index 1c50be3ab8a9..934dfdcb4e73 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
@@ -142,6 +142,7 @@ void amdgpu_ih_ring_fini(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih)
  */
 int amdgpu_ih_process(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih)
 {
+	unsigned int count = AMDGPU_IH_MAX_NUM_IVS;
 	u32 wptr;
 
 	if (!ih->enabled || adev->shutdown)
@@ -159,7 +160,7 @@ restart_ih:
 	/* Order reading of wptr vs. reading of IH ring data */
 	rmb();
 
-	while (ih->rptr != wptr) {
+	while (ih->rptr != wptr && --count) {
 		amdgpu_irq_dispatch(adev, ih);
 		ih->rptr &= ih->ptr_mask;
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h
index 113a1ba13d4a..4e0bb645176d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h
@@ -24,6 +24,9 @@
 #ifndef __AMDGPU_IH_H__
 #define __AMDGPU_IH_H__
 
+/* Maximum number of IVs processed at once */
+#define AMDGPU_IH_MAX_NUM_IVS	32
+
 struct amdgpu_device;
 struct amdgpu_iv_entry;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index e860412043bb..b17d0545728e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -39,6 +39,7 @@
 #include "amdgpu_amdkfd.h"
 #include "amdgpu_gem.h"
 #include "amdgpu_display.h"
+#include "amdgpu_ras.h"
 
 static void amdgpu_unregister_gpu_instance(struct amdgpu_device *adev)
 {
@@ -296,6 +297,17 @@ static int amdgpu_firmware_info(struct drm_amdgpu_info_firmware *fw_info,
 		fw_info->ver = adev->pm.fw_version;
 		fw_info->feature = 0;
 		break;
+	case AMDGPU_INFO_FW_TA:
+		if (query_fw->index > 1)
+			return -EINVAL;
+		if (query_fw->index == 0) {
+			fw_info->ver = adev->psp.ta_fw_version;
+			fw_info->feature = adev->psp.ta_xgmi_ucode_version;
+		} else {
+			fw_info->ver = adev->psp.ta_fw_version;
+			fw_info->feature = adev->psp.ta_ras_ucode_version;
+		}
+		break;
 	case AMDGPU_INFO_FW_SDMA:
 		if (query_fw->index >= adev->sdma.num_instances)
 			return -EINVAL;
@@ -684,6 +696,10 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 		if (adev->pm.dpm_enabled) {
 			dev_info.max_engine_clock = amdgpu_dpm_get_sclk(adev, false) * 10;
 			dev_info.max_memory_clock = amdgpu_dpm_get_mclk(adev, false) * 10;
+		} else if (amdgpu_sriov_vf(adev) && amdgim_is_hwperf(adev) &&
+			   adev->virt.ops->get_pp_clk) {
+			dev_info.max_engine_clock = amdgpu_virt_get_sclk(adev, false) * 10;
+			dev_info.max_memory_clock = amdgpu_virt_get_mclk(adev, false) * 10;
 		} else {
 			dev_info.max_engine_clock = adev->clock.default_sclk * 10;
 			dev_info.max_memory_clock = adev->clock.default_mclk * 10;
@@ -909,6 +925,18 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 	case AMDGPU_INFO_VRAM_LOST_COUNTER:
 		ui32 = atomic_read(&adev->vram_lost_counter);
 		return copy_to_user(out, &ui32, min(size, 4u)) ? -EFAULT : 0;
+	case AMDGPU_INFO_RAS_ENABLED_FEATURES: {
+		struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
+		uint64_t ras_mask;
+
+		if (!ras)
+			return -EINVAL;
+		ras_mask = (uint64_t)ras->supported << 32 | ras->features;
+
+		return copy_to_user(out, &ras_mask,
+				min_t(u64, size, sizeof(ras_mask))) ?
+			-EFAULT : 0;
+	}
 	default:
 		DRM_DEBUG_KMS("Invalid request %d\n", info->query);
 		return -EINVAL;
@@ -1328,6 +1356,16 @@ static int amdgpu_debugfs_firmware_info(struct seq_file *m, void *data)
 	seq_printf(m, "ASD feature version: %u, firmware version: 0x%08x\n",
 		   fw_info.feature, fw_info.ver);
 
+	query_fw.fw_type = AMDGPU_INFO_FW_TA;
+	for (i = 0; i < 2; i++) {
+		query_fw.index = i;
+		ret = amdgpu_firmware_info(&fw_info, &query_fw, adev);
+		if (ret)
+			continue;
+		seq_printf(m, "TA %s feature version: %u, firmware version: 0x%08x\n",
+				i ? "RAS" : "XGMI", fw_info.feature, fw_info.ver);
+	}
+
 	/* SMC */
 	query_fw.fw_type = AMDGPU_INFO_FW_SMC;
 	ret = amdgpu_firmware_info(&fw_info, &query_fw, adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
index 889e443eeee7..2e9e3db778c6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
@@ -58,7 +58,7 @@ struct amdgpu_hpd;
 #define to_amdgpu_encoder(x) container_of(x, struct amdgpu_encoder, base)
 #define to_amdgpu_framebuffer(x) container_of(x, struct amdgpu_framebuffer, base)
 
-#define to_dm_plane_state(x)	container_of(x, struct dm_plane_state, base);
+#define to_dm_plane_state(x)	container_of(x, struct dm_plane_state, base)
 
 #define AMDGPU_MAX_HPD_PINS 6
 #define AMDGPU_MAX_CRTCS 6
@@ -406,7 +406,7 @@ struct amdgpu_crtc {
 	struct amdgpu_flip_work *pflip_works;
 	enum amdgpu_flip_status pflip_status;
 	int deferred_flip_completion;
-	u64 last_flip_vblank;
+	u32 last_flip_vblank;
 	/* pll sharing */
 	struct amdgpu_atom_ss ss;
 	bool ss_enabled;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index ec9e45004bff..93b2c5a48a71 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -88,12 +88,14 @@ static void amdgpu_bo_destroy(struct ttm_buffer_object *tbo)
 	if (bo->gem_base.import_attach)
 		drm_prime_gem_destroy(&bo->gem_base, bo->tbo.sg);
 	drm_gem_object_release(&bo->gem_base);
-	amdgpu_bo_unref(&bo->parent);
+	/* in case amdgpu_device_recover_vram got NULL of bo->parent */
 	if (!list_empty(&bo->shadow_list)) {
 		mutex_lock(&adev->shadow_list_lock);
 		list_del_init(&bo->shadow_list);
 		mutex_unlock(&adev->shadow_list_lock);
 	}
+	amdgpu_bo_unref(&bo->parent);
+
 	kfree(bo->metadata);
 	kfree(bo);
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
index 220a6a7b1bc1..c430e8259038 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
@@ -72,6 +72,8 @@ struct amdgpu_bo_va {
 
 	/* If the mappings are cleared or filled */
 	bool				cleared;
+
+	bool				is_xgmi;
 };
 
 struct amdgpu_bo {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c
index a7adb7b6bd98..95144e49c7f9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c
@@ -28,6 +28,7 @@
 #include "amdgpu_pm.h"
 #include "amdgpu_dpm.h"
 #include "amdgpu_display.h"
+#include "amdgpu_smu.h"
 #include "atom.h"
 #include <linux/power_supply.h>
 #include <linux/hwmon.h>
@@ -80,6 +81,27 @@ void amdgpu_pm_acpi_event_handler(struct amdgpu_device *adev)
 	}
 }
 
+int amdgpu_dpm_read_sensor(struct amdgpu_device *adev, enum amd_pp_sensors sensor,
+			   void *data, uint32_t *size)
+{
+	int ret = 0;
+
+	if (!data || !size)
+		return -EINVAL;
+
+	if (is_support_sw_smu(adev))
+		ret = smu_read_sensor(&adev->smu, sensor, data, size);
+	else {
+		if (adev->powerplay.pp_funcs && adev->powerplay.pp_funcs->read_sensor)
+			ret = adev->powerplay.pp_funcs->read_sensor((adev)->powerplay.pp_handle,
+								    sensor, data, size);
+		else
+			ret = -EINVAL;
+	}
+
+	return ret;
+}
+
 /**
  * DOC: power_dpm_state
  *
@@ -122,7 +144,9 @@ static ssize_t amdgpu_get_dpm_state(struct device *dev,
 	struct amdgpu_device *adev = ddev->dev_private;
 	enum amd_pm_state_type pm;
 
-	if (adev->powerplay.pp_funcs->get_current_power_state)
+	if (is_support_sw_smu(adev) && adev->smu.ppt_funcs->get_current_power_state)
+		pm = amdgpu_smu_get_current_power_state(adev);
+	else if (adev->powerplay.pp_funcs->get_current_power_state)
 		pm = amdgpu_dpm_get_current_power_state(adev);
 	else
 		pm = adev->pm.dpm.user_state;
@@ -240,7 +264,9 @@ static ssize_t amdgpu_get_dpm_forced_performance_level(struct device *dev,
 	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON))
 		return snprintf(buf, PAGE_SIZE, "off\n");
 
-	if (adev->powerplay.pp_funcs->get_performance_level)
+	if (is_support_sw_smu(adev))
+		level = smu_get_performance_level(&adev->smu);
+	else if (adev->powerplay.pp_funcs->get_performance_level)
 		level = amdgpu_dpm_get_performance_level(adev);
 	else
 		level = adev->pm.dpm.forced_level;
@@ -273,7 +299,9 @@ static ssize_t amdgpu_set_dpm_forced_performance_level(struct device *dev,
 	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON))
 		return -EINVAL;
 
-	if (adev->powerplay.pp_funcs->get_performance_level)
+	if (is_support_sw_smu(adev))
+		current_level = smu_get_performance_level(&adev->smu);
+	else if (adev->powerplay.pp_funcs->get_performance_level)
 		current_level = amdgpu_dpm_get_performance_level(adev);
 
 	if (strncmp("low", buf, strlen("low")) == 0) {
@@ -299,10 +327,35 @@ static ssize_t amdgpu_set_dpm_forced_performance_level(struct device *dev,
 		goto fail;
 	}
 
+        if (amdgpu_sriov_vf(adev)) {
+                if (amdgim_is_hwperf(adev) &&
+                    adev->virt.ops->force_dpm_level) {
+                        mutex_lock(&adev->pm.mutex);
+                        adev->virt.ops->force_dpm_level(adev, level);
+                        mutex_unlock(&adev->pm.mutex);
+                        return count;
+                } else {
+                        return -EINVAL;
+		}
+        }
+
 	if (current_level == level)
 		return count;
 
-	if (adev->powerplay.pp_funcs->force_performance_level) {
+	if (is_support_sw_smu(adev)) {
+		mutex_lock(&adev->pm.mutex);
+		if (adev->pm.dpm.thermal_active) {
+			count = -EINVAL;
+			mutex_unlock(&adev->pm.mutex);
+			goto fail;
+		}
+		ret = smu_force_performance_level(&adev->smu, level);
+		if (ret)
+			count = -EINVAL;
+		else
+			adev->pm.dpm.forced_level = level;
+		mutex_unlock(&adev->pm.mutex);
+	} else if (adev->powerplay.pp_funcs->force_performance_level) {
 		mutex_lock(&adev->pm.mutex);
 		if (adev->pm.dpm.thermal_active) {
 			count = -EINVAL;
@@ -328,9 +381,13 @@ static ssize_t amdgpu_get_pp_num_states(struct device *dev,
 	struct drm_device *ddev = dev_get_drvdata(dev);
 	struct amdgpu_device *adev = ddev->dev_private;
 	struct pp_states_info data;
-	int i, buf_len;
+	int i, buf_len, ret;
 
-	if (adev->powerplay.pp_funcs->get_pp_num_states)
+	if (is_support_sw_smu(adev)) {
+		ret = smu_get_power_num_states(&adev->smu, &data);
+		if (ret)
+			return ret;
+	} else if (adev->powerplay.pp_funcs->get_pp_num_states)
 		amdgpu_dpm_get_pp_num_states(adev, &data);
 
 	buf_len = snprintf(buf, PAGE_SIZE, "states: %d\n", data.nums);
@@ -351,23 +408,29 @@ static ssize_t amdgpu_get_pp_cur_state(struct device *dev,
 	struct drm_device *ddev = dev_get_drvdata(dev);
 	struct amdgpu_device *adev = ddev->dev_private;
 	struct pp_states_info data;
+	struct smu_context *smu = &adev->smu;
 	enum amd_pm_state_type pm = 0;
-	int i = 0;
+	int i = 0, ret = 0;
 
-	if (adev->powerplay.pp_funcs->get_current_power_state
+	if (is_support_sw_smu(adev)) {
+		pm = smu_get_current_power_state(smu);
+		ret = smu_get_power_num_states(smu, &data);
+		if (ret)
+			return ret;
+	} else if (adev->powerplay.pp_funcs->get_current_power_state
 		 && adev->powerplay.pp_funcs->get_pp_num_states) {
 		pm = amdgpu_dpm_get_current_power_state(adev);
 		amdgpu_dpm_get_pp_num_states(adev, &data);
+	}
 
-		for (i = 0; i < data.nums; i++) {
-			if (pm == data.states[i])
-				break;
-		}
-
-		if (i == data.nums)
-			i = -EINVAL;
+	for (i = 0; i < data.nums; i++) {
+		if (pm == data.states[i])
+			break;
 	}
 
+	if (i == data.nums)
+		i = -EINVAL;
+
 	return snprintf(buf, PAGE_SIZE, "%d\n", i);
 }
 
@@ -397,6 +460,8 @@ static ssize_t amdgpu_set_pp_force_state(struct device *dev,
 
 	if (strlen(buf) == 1)
 		adev->pp_force_state_enabled = false;
+	else if (is_support_sw_smu(adev))
+		adev->pp_force_state_enabled = false;
 	else if (adev->powerplay.pp_funcs->dispatch_tasks &&
 			adev->powerplay.pp_funcs->get_pp_num_states) {
 		struct pp_states_info data;
@@ -442,7 +507,12 @@ static ssize_t amdgpu_get_pp_table(struct device *dev,
 	char *table = NULL;
 	int size;
 
-	if (adev->powerplay.pp_funcs->get_pp_table)
+	if (is_support_sw_smu(adev)) {
+		size = smu_sys_get_pp_table(&adev->smu, (void **)&table);
+		if (size < 0)
+			return size;
+	}
+	else if (adev->powerplay.pp_funcs->get_pp_table)
 		size = amdgpu_dpm_get_pp_table(adev, &table);
 	else
 		return 0;
@@ -462,8 +532,13 @@ static ssize_t amdgpu_set_pp_table(struct device *dev,
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
 	struct amdgpu_device *adev = ddev->dev_private;
+	int ret = 0;
 
-	if (adev->powerplay.pp_funcs->set_pp_table)
+	if (is_support_sw_smu(adev)) {
+		ret = smu_sys_set_pp_table(&adev->smu, (void *)buf, count);
+		if (ret)
+			return ret;
+	} else if (adev->powerplay.pp_funcs->set_pp_table)
 		amdgpu_dpm_set_pp_table(adev, buf, count);
 
 	return count;
@@ -586,19 +661,29 @@ static ssize_t amdgpu_set_pp_od_clk_voltage(struct device *dev,
 			tmp_str++;
 	}
 
-	if (adev->powerplay.pp_funcs->odn_edit_dpm_table)
-		ret = amdgpu_dpm_odn_edit_dpm_table(adev, type,
-						parameter, parameter_size);
+	if (is_support_sw_smu(adev)) {
+		ret = smu_od_edit_dpm_table(&adev->smu, type,
+					    parameter, parameter_size);
 
-	if (ret)
-		return -EINVAL;
+		if (ret)
+			return -EINVAL;
+	} else {
+		if (adev->powerplay.pp_funcs->odn_edit_dpm_table)
+			ret = amdgpu_dpm_odn_edit_dpm_table(adev, type,
+						parameter, parameter_size);
 
-	if (type == PP_OD_COMMIT_DPM_TABLE) {
-		if (adev->powerplay.pp_funcs->dispatch_tasks) {
-			amdgpu_dpm_dispatch_task(adev, AMD_PP_TASK_READJUST_POWER_STATE, NULL);
-			return count;
-		} else {
+		if (ret)
 			return -EINVAL;
+
+		if (type == PP_OD_COMMIT_DPM_TABLE) {
+			if (adev->powerplay.pp_funcs->dispatch_tasks) {
+				amdgpu_dpm_dispatch_task(adev,
+						AMD_PP_TASK_READJUST_POWER_STATE,
+						NULL);
+				return count;
+			} else {
+				return -EINVAL;
+			}
 		}
 	}
 
@@ -613,7 +698,13 @@ static ssize_t amdgpu_get_pp_od_clk_voltage(struct device *dev,
 	struct amdgpu_device *adev = ddev->dev_private;
 	uint32_t size = 0;
 
-	if (adev->powerplay.pp_funcs->print_clock_levels) {
+	if (is_support_sw_smu(adev)) {
+		size = smu_print_clk_levels(&adev->smu, OD_SCLK, buf);
+		size += smu_print_clk_levels(&adev->smu, OD_MCLK, buf+size);
+		size += smu_print_clk_levels(&adev->smu, OD_VDDC_CURVE, buf+size);
+		size += smu_print_clk_levels(&adev->smu, OD_RANGE, buf+size);
+		return size;
+	} else if (adev->powerplay.pp_funcs->print_clock_levels) {
 		size = amdgpu_dpm_print_clock_levels(adev, OD_SCLK, buf);
 		size += amdgpu_dpm_print_clock_levels(adev, OD_MCLK, buf+size);
 		size += amdgpu_dpm_print_clock_levels(adev, OD_VDDC_CURVE, buf+size);
@@ -711,7 +802,13 @@ static ssize_t amdgpu_get_pp_dpm_sclk(struct device *dev,
 	struct drm_device *ddev = dev_get_drvdata(dev);
 	struct amdgpu_device *adev = ddev->dev_private;
 
-	if (adev->powerplay.pp_funcs->print_clock_levels)
+	if (amdgpu_sriov_vf(adev) && amdgim_is_hwperf(adev) &&
+	    adev->virt.ops->get_pp_clk)
+		return adev->virt.ops->get_pp_clk(adev, PP_SCLK, buf);
+
+	if (is_support_sw_smu(adev))
+		return smu_print_clk_levels(&adev->smu, PP_SCLK, buf);
+	else if (adev->powerplay.pp_funcs->print_clock_levels)
 		return amdgpu_dpm_print_clock_levels(adev, PP_SCLK, buf);
 	else
 		return snprintf(buf, PAGE_SIZE, "\n");
@@ -767,7 +864,9 @@ static ssize_t amdgpu_set_pp_dpm_sclk(struct device *dev,
 	if (ret)
 		return ret;
 
-	if (adev->powerplay.pp_funcs->force_clock_level)
+	if (is_support_sw_smu(adev))
+		ret = smu_force_clk_levels(&adev->smu, PP_SCLK, mask);
+	else if (adev->powerplay.pp_funcs->force_clock_level)
 		ret = amdgpu_dpm_force_clock_level(adev, PP_SCLK, mask);
 
 	if (ret)
@@ -783,7 +882,9 @@ static ssize_t amdgpu_get_pp_dpm_mclk(struct device *dev,
 	struct drm_device *ddev = dev_get_drvdata(dev);
 	struct amdgpu_device *adev = ddev->dev_private;
 
-	if (adev->powerplay.pp_funcs->print_clock_levels)
+	if (is_support_sw_smu(adev))
+		return smu_print_clk_levels(&adev->smu, PP_MCLK, buf);
+	else if (adev->powerplay.pp_funcs->print_clock_levels)
 		return amdgpu_dpm_print_clock_levels(adev, PP_MCLK, buf);
 	else
 		return snprintf(buf, PAGE_SIZE, "\n");
@@ -803,7 +904,9 @@ static ssize_t amdgpu_set_pp_dpm_mclk(struct device *dev,
 	if (ret)
 		return ret;
 
-	if (adev->powerplay.pp_funcs->force_clock_level)
+	if (is_support_sw_smu(adev))
+		ret = smu_force_clk_levels(&adev->smu, PP_MCLK, mask);
+	else if (adev->powerplay.pp_funcs->force_clock_level)
 		ret = amdgpu_dpm_force_clock_level(adev, PP_MCLK, mask);
 
 	if (ret)
@@ -819,7 +922,9 @@ static ssize_t amdgpu_get_pp_dpm_socclk(struct device *dev,
 	struct drm_device *ddev = dev_get_drvdata(dev);
 	struct amdgpu_device *adev = ddev->dev_private;
 
-	if (adev->powerplay.pp_funcs->print_clock_levels)
+	if (is_support_sw_smu(adev))
+		return smu_print_clk_levels(&adev->smu, PP_SOCCLK, buf);
+	else if (adev->powerplay.pp_funcs->print_clock_levels)
 		return amdgpu_dpm_print_clock_levels(adev, PP_SOCCLK, buf);
 	else
 		return snprintf(buf, PAGE_SIZE, "\n");
@@ -839,7 +944,9 @@ static ssize_t amdgpu_set_pp_dpm_socclk(struct device *dev,
 	if (ret)
 		return ret;
 
-	if (adev->powerplay.pp_funcs->force_clock_level)
+	if (is_support_sw_smu(adev))
+		ret = smu_force_clk_levels(&adev->smu, PP_SOCCLK, mask);
+	else if (adev->powerplay.pp_funcs->force_clock_level)
 		ret = amdgpu_dpm_force_clock_level(adev, PP_SOCCLK, mask);
 
 	if (ret)
@@ -855,7 +962,9 @@ static ssize_t amdgpu_get_pp_dpm_fclk(struct device *dev,
 	struct drm_device *ddev = dev_get_drvdata(dev);
 	struct amdgpu_device *adev = ddev->dev_private;
 
-	if (adev->powerplay.pp_funcs->print_clock_levels)
+	if (is_support_sw_smu(adev))
+		return smu_print_clk_levels(&adev->smu, PP_FCLK, buf);
+	else if (adev->powerplay.pp_funcs->print_clock_levels)
 		return amdgpu_dpm_print_clock_levels(adev, PP_FCLK, buf);
 	else
 		return snprintf(buf, PAGE_SIZE, "\n");
@@ -875,7 +984,9 @@ static ssize_t amdgpu_set_pp_dpm_fclk(struct device *dev,
 	if (ret)
 		return ret;
 
-	if (adev->powerplay.pp_funcs->force_clock_level)
+	if (is_support_sw_smu(adev))
+		ret = smu_force_clk_levels(&adev->smu, PP_FCLK, mask);
+	else if (adev->powerplay.pp_funcs->force_clock_level)
 		ret = amdgpu_dpm_force_clock_level(adev, PP_FCLK, mask);
 
 	if (ret)
@@ -891,7 +1002,9 @@ static ssize_t amdgpu_get_pp_dpm_dcefclk(struct device *dev,
 	struct drm_device *ddev = dev_get_drvdata(dev);
 	struct amdgpu_device *adev = ddev->dev_private;
 
-	if (adev->powerplay.pp_funcs->print_clock_levels)
+	if (is_support_sw_smu(adev))
+		return smu_print_clk_levels(&adev->smu, PP_DCEFCLK, buf);
+	else if (adev->powerplay.pp_funcs->print_clock_levels)
 		return amdgpu_dpm_print_clock_levels(adev, PP_DCEFCLK, buf);
 	else
 		return snprintf(buf, PAGE_SIZE, "\n");
@@ -911,7 +1024,9 @@ static ssize_t amdgpu_set_pp_dpm_dcefclk(struct device *dev,
 	if (ret)
 		return ret;
 
-	if (adev->powerplay.pp_funcs->force_clock_level)
+	if (is_support_sw_smu(adev))
+		ret = smu_force_clk_levels(&adev->smu, PP_DCEFCLK, mask);
+	else if (adev->powerplay.pp_funcs->force_clock_level)
 		ret = amdgpu_dpm_force_clock_level(adev, PP_DCEFCLK, mask);
 
 	if (ret)
@@ -927,7 +1042,9 @@ static ssize_t amdgpu_get_pp_dpm_pcie(struct device *dev,
 	struct drm_device *ddev = dev_get_drvdata(dev);
 	struct amdgpu_device *adev = ddev->dev_private;
 
-	if (adev->powerplay.pp_funcs->print_clock_levels)
+	if (is_support_sw_smu(adev))
+		return smu_print_clk_levels(&adev->smu, PP_PCIE, buf);
+	else if (adev->powerplay.pp_funcs->print_clock_levels)
 		return amdgpu_dpm_print_clock_levels(adev, PP_PCIE, buf);
 	else
 		return snprintf(buf, PAGE_SIZE, "\n");
@@ -947,7 +1064,9 @@ static ssize_t amdgpu_set_pp_dpm_pcie(struct device *dev,
 	if (ret)
 		return ret;
 
-	if (adev->powerplay.pp_funcs->force_clock_level)
+	if (is_support_sw_smu(adev))
+		ret = smu_force_clk_levels(&adev->smu, PP_PCIE, mask);
+	else if (adev->powerplay.pp_funcs->force_clock_level)
 		ret = amdgpu_dpm_force_clock_level(adev, PP_PCIE, mask);
 
 	if (ret)
@@ -964,7 +1083,9 @@ static ssize_t amdgpu_get_pp_sclk_od(struct device *dev,
 	struct amdgpu_device *adev = ddev->dev_private;
 	uint32_t value = 0;
 
-	if (adev->powerplay.pp_funcs->get_sclk_od)
+	if (is_support_sw_smu(adev))
+		value = smu_get_od_percentage(&(adev->smu), OD_SCLK);
+	else if (adev->powerplay.pp_funcs->get_sclk_od)
 		value = amdgpu_dpm_get_sclk_od(adev);
 
 	return snprintf(buf, PAGE_SIZE, "%d\n", value);
@@ -986,14 +1107,19 @@ static ssize_t amdgpu_set_pp_sclk_od(struct device *dev,
 		count = -EINVAL;
 		goto fail;
 	}
-	if (adev->powerplay.pp_funcs->set_sclk_od)
-		amdgpu_dpm_set_sclk_od(adev, (uint32_t)value);
 
-	if (adev->powerplay.pp_funcs->dispatch_tasks) {
-		amdgpu_dpm_dispatch_task(adev, AMD_PP_TASK_READJUST_POWER_STATE, NULL);
+	if (is_support_sw_smu(adev)) {
+		value = smu_set_od_percentage(&(adev->smu), OD_SCLK, (uint32_t)value);
 	} else {
-		adev->pm.dpm.current_ps = adev->pm.dpm.boot_ps;
-		amdgpu_pm_compute_clocks(adev);
+		if (adev->powerplay.pp_funcs->set_sclk_od)
+			amdgpu_dpm_set_sclk_od(adev, (uint32_t)value);
+
+		if (adev->powerplay.pp_funcs->dispatch_tasks) {
+			amdgpu_dpm_dispatch_task(adev, AMD_PP_TASK_READJUST_POWER_STATE, NULL);
+		} else {
+			adev->pm.dpm.current_ps = adev->pm.dpm.boot_ps;
+			amdgpu_pm_compute_clocks(adev);
+		}
 	}
 
 fail:
@@ -1008,7 +1134,9 @@ static ssize_t amdgpu_get_pp_mclk_od(struct device *dev,
 	struct amdgpu_device *adev = ddev->dev_private;
 	uint32_t value = 0;
 
-	if (adev->powerplay.pp_funcs->get_mclk_od)
+	if (is_support_sw_smu(adev))
+		value = smu_get_od_percentage(&(adev->smu), OD_MCLK);
+	else if (adev->powerplay.pp_funcs->get_mclk_od)
 		value = amdgpu_dpm_get_mclk_od(adev);
 
 	return snprintf(buf, PAGE_SIZE, "%d\n", value);
@@ -1030,14 +1158,19 @@ static ssize_t amdgpu_set_pp_mclk_od(struct device *dev,
 		count = -EINVAL;
 		goto fail;
 	}
-	if (adev->powerplay.pp_funcs->set_mclk_od)
-		amdgpu_dpm_set_mclk_od(adev, (uint32_t)value);
 
-	if (adev->powerplay.pp_funcs->dispatch_tasks) {
-		amdgpu_dpm_dispatch_task(adev, AMD_PP_TASK_READJUST_POWER_STATE, NULL);
+	if (is_support_sw_smu(adev)) {
+		value = smu_set_od_percentage(&(adev->smu), OD_MCLK, (uint32_t)value);
 	} else {
-		adev->pm.dpm.current_ps = adev->pm.dpm.boot_ps;
-		amdgpu_pm_compute_clocks(adev);
+		if (adev->powerplay.pp_funcs->set_mclk_od)
+			amdgpu_dpm_set_mclk_od(adev, (uint32_t)value);
+
+		if (adev->powerplay.pp_funcs->dispatch_tasks) {
+			amdgpu_dpm_dispatch_task(adev, AMD_PP_TASK_READJUST_POWER_STATE, NULL);
+		} else {
+			adev->pm.dpm.current_ps = adev->pm.dpm.boot_ps;
+			amdgpu_pm_compute_clocks(adev);
+		}
 	}
 
 fail:
@@ -1071,7 +1204,9 @@ static ssize_t amdgpu_get_pp_power_profile_mode(struct device *dev,
 	struct drm_device *ddev = dev_get_drvdata(dev);
 	struct amdgpu_device *adev = ddev->dev_private;
 
-	if (adev->powerplay.pp_funcs->get_power_profile_mode)
+	if (is_support_sw_smu(adev))
+		return smu_get_power_profile_mode(&adev->smu, buf);
+	else if (adev->powerplay.pp_funcs->get_power_profile_mode)
 		return amdgpu_dpm_get_power_profile_mode(adev, buf);
 
 	return snprintf(buf, PAGE_SIZE, "\n");
@@ -1121,9 +1256,10 @@ static ssize_t amdgpu_set_pp_power_profile_mode(struct device *dev,
 		}
 	}
 	parameter[parameter_size] = profile_mode;
-	if (adev->powerplay.pp_funcs->set_power_profile_mode)
+	if (is_support_sw_smu(adev))
+		ret = smu_set_power_profile_mode(&adev->smu, parameter, parameter_size);
+	else if (adev->powerplay.pp_funcs->set_power_profile_mode)
 		ret = amdgpu_dpm_set_power_profile_mode(adev, parameter, parameter_size);
-
 	if (!ret)
 		return count;
 fail:
@@ -1146,14 +1282,10 @@ static ssize_t amdgpu_get_busy_percent(struct device *dev,
 	struct amdgpu_device *adev = ddev->dev_private;
 	int r, value, size = sizeof(value);
 
-	/* sanity check PP is enabled */
-	if (!(adev->powerplay.pp_funcs &&
-	      adev->powerplay.pp_funcs->read_sensor))
-		return -EINVAL;
-
 	/* read the IP busy sensor */
 	r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GPU_LOAD,
 				   (void *)&value, &size);
+
 	if (r)
 		return r;
 
@@ -1247,11 +1379,6 @@ static ssize_t amdgpu_hwmon_show_temp(struct device *dev,
 	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON))
 		return -EINVAL;
 
-	/* sanity check PP is enabled */
-	if (!(adev->powerplay.pp_funcs &&
-	      adev->powerplay.pp_funcs->read_sensor))
-		return -EINVAL;
-
 	/* get the temperature */
 	r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GPU_TEMP,
 				   (void *)&temp, &size);
@@ -1283,11 +1410,14 @@ static ssize_t amdgpu_hwmon_get_pwm1_enable(struct device *dev,
 {
 	struct amdgpu_device *adev = dev_get_drvdata(dev);
 	u32 pwm_mode = 0;
+	if (is_support_sw_smu(adev)) {
+		pwm_mode = smu_get_fan_control_mode(&adev->smu);
+	} else {
+		if (!adev->powerplay.pp_funcs->get_fan_control_mode)
+			return -EINVAL;
 
-	if (!adev->powerplay.pp_funcs->get_fan_control_mode)
-		return -EINVAL;
-
-	pwm_mode = amdgpu_dpm_get_fan_control_mode(adev);
+		pwm_mode = amdgpu_dpm_get_fan_control_mode(adev);
+	}
 
 	return sprintf(buf, "%i\n", pwm_mode);
 }
@@ -1306,14 +1436,22 @@ static ssize_t amdgpu_hwmon_set_pwm1_enable(struct device *dev,
 	     (adev->ddev->switch_power_state != DRM_SWITCH_POWER_ON))
 		return -EINVAL;
 
-	if (!adev->powerplay.pp_funcs->set_fan_control_mode)
-		return -EINVAL;
+	if (is_support_sw_smu(adev)) {
+		err = kstrtoint(buf, 10, &value);
+		if (err)
+			return err;
 
-	err = kstrtoint(buf, 10, &value);
-	if (err)
-		return err;
+		smu_set_fan_control_mode(&adev->smu, value);
+	} else {
+		if (!adev->powerplay.pp_funcs->set_fan_control_mode)
+			return -EINVAL;
+
+		err = kstrtoint(buf, 10, &value);
+		if (err)
+			return err;
 
-	amdgpu_dpm_set_fan_control_mode(adev, value);
+		amdgpu_dpm_set_fan_control_mode(adev, value);
+	}
 
 	return count;
 }
@@ -1345,8 +1483,10 @@ static ssize_t amdgpu_hwmon_set_pwm1(struct device *dev,
 	if  ((adev->flags & AMD_IS_PX) &&
 	     (adev->ddev->switch_power_state != DRM_SWITCH_POWER_ON))
 		return -EINVAL;
-
-	pwm_mode = amdgpu_dpm_get_fan_control_mode(adev);
+	if (is_support_sw_smu(adev))
+		pwm_mode = smu_get_fan_control_mode(&adev->smu);
+	else
+		pwm_mode = amdgpu_dpm_get_fan_control_mode(adev);
 	if (pwm_mode != AMD_FAN_CTRL_MANUAL) {
 		pr_info("manual fan speed control should be enabled first\n");
 		return -EINVAL;
@@ -1358,7 +1498,11 @@ static ssize_t amdgpu_hwmon_set_pwm1(struct device *dev,
 
 	value = (value * 100) / 255;
 
-	if (adev->powerplay.pp_funcs->set_fan_speed_percent) {
+	if (is_support_sw_smu(adev)) {
+		err = smu_set_fan_speed_percent(&adev->smu, value);
+		if (err)
+			return err;
+	} else if (adev->powerplay.pp_funcs->set_fan_speed_percent) {
 		err = amdgpu_dpm_set_fan_speed_percent(adev, value);
 		if (err)
 			return err;
@@ -1380,7 +1524,11 @@ static ssize_t amdgpu_hwmon_get_pwm1(struct device *dev,
 	     (adev->ddev->switch_power_state != DRM_SWITCH_POWER_ON))
 		return -EINVAL;
 
-	if (adev->powerplay.pp_funcs->get_fan_speed_percent) {
+	if (is_support_sw_smu(adev)) {
+		err = smu_get_fan_speed_percent(&adev->smu, &speed);
+		if (err)
+			return err;
+	} else if (adev->powerplay.pp_funcs->get_fan_speed_percent) {
 		err = amdgpu_dpm_get_fan_speed_percent(adev, &speed);
 		if (err)
 			return err;
@@ -1404,7 +1552,11 @@ static ssize_t amdgpu_hwmon_get_fan1_input(struct device *dev,
 	     (adev->ddev->switch_power_state != DRM_SWITCH_POWER_ON))
 		return -EINVAL;
 
-	if (adev->powerplay.pp_funcs->get_fan_speed_rpm) {
+	if (is_support_sw_smu(adev)) {
+		err = smu_get_current_rpm(&adev->smu, &speed);
+		if (err)
+			return err;
+	} else if (adev->powerplay.pp_funcs->get_fan_speed_rpm) {
 		err = amdgpu_dpm_get_fan_speed_rpm(adev, &speed);
 		if (err)
 			return err;
@@ -1422,9 +1574,6 @@ static ssize_t amdgpu_hwmon_get_fan1_min(struct device *dev,
 	u32 size = sizeof(min_rpm);
 	int r;
 
-	if (!adev->powerplay.pp_funcs->read_sensor)
-		return -EINVAL;
-
 	r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_MIN_FAN_RPM,
 				   (void *)&min_rpm, &size);
 	if (r)
@@ -1442,9 +1591,6 @@ static ssize_t amdgpu_hwmon_get_fan1_max(struct device *dev,
 	u32 size = sizeof(max_rpm);
 	int r;
 
-	if (!adev->powerplay.pp_funcs->read_sensor)
-		return -EINVAL;
-
 	r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_MAX_FAN_RPM,
 				   (void *)&max_rpm, &size);
 	if (r)
@@ -1466,7 +1612,11 @@ static ssize_t amdgpu_hwmon_get_fan1_target(struct device *dev,
 	     (adev->ddev->switch_power_state != DRM_SWITCH_POWER_ON))
 		return -EINVAL;
 
-	if (adev->powerplay.pp_funcs->get_fan_speed_rpm) {
+	if (is_support_sw_smu(adev)) {
+		err = smu_get_current_rpm(&adev->smu, &rpm);
+		if (err)
+			return err;
+	} else if (adev->powerplay.pp_funcs->get_fan_speed_rpm) {
 		err = amdgpu_dpm_get_fan_speed_rpm(adev, &rpm);
 		if (err)
 			return err;
@@ -1484,7 +1634,11 @@ static ssize_t amdgpu_hwmon_set_fan1_target(struct device *dev,
 	u32 value;
 	u32 pwm_mode;
 
-	pwm_mode = amdgpu_dpm_get_fan_control_mode(adev);
+	if (is_support_sw_smu(adev))
+		pwm_mode = smu_get_fan_control_mode(&adev->smu);
+	else
+		pwm_mode = amdgpu_dpm_get_fan_control_mode(adev);
+
 	if (pwm_mode != AMD_FAN_CTRL_MANUAL)
 		return -ENODATA;
 
@@ -1497,7 +1651,11 @@ static ssize_t amdgpu_hwmon_set_fan1_target(struct device *dev,
 	if (err)
 		return err;
 
-	if (adev->powerplay.pp_funcs->set_fan_speed_rpm) {
+	if (is_support_sw_smu(adev)) {
+		err = smu_set_fan_speed_rpm(&adev->smu, value);
+		if (err)
+			return err;
+	} else if (adev->powerplay.pp_funcs->set_fan_speed_rpm) {
 		err = amdgpu_dpm_set_fan_speed_rpm(adev, value);
 		if (err)
 			return err;
@@ -1513,11 +1671,14 @@ static ssize_t amdgpu_hwmon_get_fan1_enable(struct device *dev,
 	struct amdgpu_device *adev = dev_get_drvdata(dev);
 	u32 pwm_mode = 0;
 
-	if (!adev->powerplay.pp_funcs->get_fan_control_mode)
-		return -EINVAL;
-
-	pwm_mode = amdgpu_dpm_get_fan_control_mode(adev);
+	if (is_support_sw_smu(adev)) {
+		pwm_mode = smu_get_fan_control_mode(&adev->smu);
+	} else {
+		if (!adev->powerplay.pp_funcs->get_fan_control_mode)
+			return -EINVAL;
 
+		pwm_mode = amdgpu_dpm_get_fan_control_mode(adev);
+	}
 	return sprintf(buf, "%i\n", pwm_mode == AMD_FAN_CTRL_AUTO ? 0 : 1);
 }
 
@@ -1536,8 +1697,6 @@ static ssize_t amdgpu_hwmon_set_fan1_enable(struct device *dev,
 	     (adev->ddev->switch_power_state != DRM_SWITCH_POWER_ON))
 		return -EINVAL;
 
-	if (!adev->powerplay.pp_funcs->set_fan_control_mode)
-		return -EINVAL;
 
 	err = kstrtoint(buf, 10, &value);
 	if (err)
@@ -1550,7 +1709,13 @@ static ssize_t amdgpu_hwmon_set_fan1_enable(struct device *dev,
 	else
 		return -EINVAL;
 
-	amdgpu_dpm_set_fan_control_mode(adev, pwm_mode);
+	if (is_support_sw_smu(adev)) {
+		smu_set_fan_control_mode(&adev->smu, pwm_mode);
+	} else {
+		if (!adev->powerplay.pp_funcs->set_fan_control_mode)
+			return -EINVAL;
+		amdgpu_dpm_set_fan_control_mode(adev, pwm_mode);
+	}
 
 	return count;
 }
@@ -1569,11 +1734,6 @@ static ssize_t amdgpu_hwmon_show_vddgfx(struct device *dev,
 	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON))
 		return -EINVAL;
 
-	/* sanity check PP is enabled */
-	if (!(adev->powerplay.pp_funcs &&
-	      adev->powerplay.pp_funcs->read_sensor))
-	      return -EINVAL;
-
 	/* get the voltage */
 	r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_VDDGFX,
 				   (void *)&vddgfx, &size);
@@ -1608,11 +1768,6 @@ static ssize_t amdgpu_hwmon_show_vddnb(struct device *dev,
 	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON))
 		return -EINVAL;
 
-	/* sanity check PP is enabled */
-	if (!(adev->powerplay.pp_funcs &&
-	      adev->powerplay.pp_funcs->read_sensor))
-	      return -EINVAL;
-
 	/* get the voltage */
 	r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_VDDNB,
 				   (void *)&vddnb, &size);
@@ -1644,11 +1799,6 @@ static ssize_t amdgpu_hwmon_show_power_avg(struct device *dev,
 	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON))
 		return -EINVAL;
 
-	/* sanity check PP is enabled */
-	if (!(adev->powerplay.pp_funcs &&
-	      adev->powerplay.pp_funcs->read_sensor))
-	      return -EINVAL;
-
 	/* get the voltage */
 	r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GPU_POWER,
 				   (void *)&query, &size);
@@ -1675,7 +1825,10 @@ static ssize_t amdgpu_hwmon_show_power_cap_max(struct device *dev,
 	struct amdgpu_device *adev = dev_get_drvdata(dev);
 	uint32_t limit = 0;
 
-	if (adev->powerplay.pp_funcs && adev->powerplay.pp_funcs->get_power_limit) {
+	if (is_support_sw_smu(adev)) {
+		smu_get_power_limit(&adev->smu, &limit, true);
+		return snprintf(buf, PAGE_SIZE, "%u\n", limit * 1000000);
+	} else if (adev->powerplay.pp_funcs && adev->powerplay.pp_funcs->get_power_limit) {
 		adev->powerplay.pp_funcs->get_power_limit(adev->powerplay.pp_handle, &limit, true);
 		return snprintf(buf, PAGE_SIZE, "%u\n", limit * 1000000);
 	} else {
@@ -1690,7 +1843,10 @@ static ssize_t amdgpu_hwmon_show_power_cap(struct device *dev,
 	struct amdgpu_device *adev = dev_get_drvdata(dev);
 	uint32_t limit = 0;
 
-	if (adev->powerplay.pp_funcs && adev->powerplay.pp_funcs->get_power_limit) {
+	if (is_support_sw_smu(adev)) {
+		smu_get_power_limit(&adev->smu, &limit, false);
+		return snprintf(buf, PAGE_SIZE, "%u\n", limit * 1000000);
+	} else if (adev->powerplay.pp_funcs && adev->powerplay.pp_funcs->get_power_limit) {
 		adev->powerplay.pp_funcs->get_power_limit(adev->powerplay.pp_handle, &limit, false);
 		return snprintf(buf, PAGE_SIZE, "%u\n", limit * 1000000);
 	} else {
@@ -1713,7 +1869,9 @@ static ssize_t amdgpu_hwmon_set_power_cap(struct device *dev,
 		return err;
 
 	value = value / 1000000; /* convert to Watt */
-	if (adev->powerplay.pp_funcs && adev->powerplay.pp_funcs->set_power_limit) {
+	if (is_support_sw_smu(adev)) {
+		adev->smu.funcs->set_power_limit(&adev->smu, value);
+	} else if (adev->powerplay.pp_funcs && adev->powerplay.pp_funcs->set_power_limit) {
 		err = adev->powerplay.pp_funcs->set_power_limit(adev->powerplay.pp_handle, value);
 		if (err)
 			return err;
@@ -1967,18 +2125,20 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj,
 	     attr == &sensor_dev_attr_fan1_enable.dev_attr.attr))
 		return 0;
 
-	/* mask fan attributes if we have no bindings for this asic to expose */
-	if ((!adev->powerplay.pp_funcs->get_fan_speed_percent &&
-	     attr == &sensor_dev_attr_pwm1.dev_attr.attr) || /* can't query fan */
-	    (!adev->powerplay.pp_funcs->get_fan_control_mode &&
-	     attr == &sensor_dev_attr_pwm1_enable.dev_attr.attr)) /* can't query state */
-		effective_mode &= ~S_IRUGO;
+	if (!is_support_sw_smu(adev)) {
+		/* mask fan attributes if we have no bindings for this asic to expose */
+		if ((!adev->powerplay.pp_funcs->get_fan_speed_percent &&
+		     attr == &sensor_dev_attr_pwm1.dev_attr.attr) || /* can't query fan */
+		    (!adev->powerplay.pp_funcs->get_fan_control_mode &&
+		     attr == &sensor_dev_attr_pwm1_enable.dev_attr.attr)) /* can't query state */
+			effective_mode &= ~S_IRUGO;
 
-	if ((!adev->powerplay.pp_funcs->set_fan_speed_percent &&
-	     attr == &sensor_dev_attr_pwm1.dev_attr.attr) || /* can't manage fan */
-	    (!adev->powerplay.pp_funcs->set_fan_control_mode &&
-	     attr == &sensor_dev_attr_pwm1_enable.dev_attr.attr)) /* can't manage state */
-		effective_mode &= ~S_IWUSR;
+		if ((!adev->powerplay.pp_funcs->set_fan_speed_percent &&
+		     attr == &sensor_dev_attr_pwm1.dev_attr.attr) || /* can't manage fan */
+		    (!adev->powerplay.pp_funcs->set_fan_control_mode &&
+		     attr == &sensor_dev_attr_pwm1_enable.dev_attr.attr)) /* can't manage state */
+			effective_mode &= ~S_IWUSR;
+	}
 
 	if ((adev->flags & AMD_IS_APU) &&
 	    (attr == &sensor_dev_attr_power1_average.dev_attr.attr ||
@@ -1987,20 +2147,22 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj,
 	     attr == &sensor_dev_attr_power1_cap.dev_attr.attr))
 		return 0;
 
-	/* hide max/min values if we can't both query and manage the fan */
-	if ((!adev->powerplay.pp_funcs->set_fan_speed_percent &&
-	     !adev->powerplay.pp_funcs->get_fan_speed_percent) &&
-	     (!adev->powerplay.pp_funcs->set_fan_speed_rpm &&
-	     !adev->powerplay.pp_funcs->get_fan_speed_rpm) &&
-	    (attr == &sensor_dev_attr_pwm1_max.dev_attr.attr ||
-	     attr == &sensor_dev_attr_pwm1_min.dev_attr.attr))
-		return 0;
+	if (!is_support_sw_smu(adev)) {
+		/* hide max/min values if we can't both query and manage the fan */
+		if ((!adev->powerplay.pp_funcs->set_fan_speed_percent &&
+		     !adev->powerplay.pp_funcs->get_fan_speed_percent) &&
+		     (!adev->powerplay.pp_funcs->set_fan_speed_rpm &&
+		     !adev->powerplay.pp_funcs->get_fan_speed_rpm) &&
+		    (attr == &sensor_dev_attr_pwm1_max.dev_attr.attr ||
+		     attr == &sensor_dev_attr_pwm1_min.dev_attr.attr))
+			return 0;
 
-	if ((!adev->powerplay.pp_funcs->set_fan_speed_rpm &&
-	     !adev->powerplay.pp_funcs->get_fan_speed_rpm) &&
-	    (attr == &sensor_dev_attr_fan1_max.dev_attr.attr ||
-	     attr == &sensor_dev_attr_fan1_min.dev_attr.attr))
-		return 0;
+		if ((!adev->powerplay.pp_funcs->set_fan_speed_rpm &&
+		     !adev->powerplay.pp_funcs->get_fan_speed_rpm) &&
+		    (attr == &sensor_dev_attr_fan1_max.dev_attr.attr ||
+		     attr == &sensor_dev_attr_fan1_min.dev_attr.attr))
+			return 0;
+	}
 
 	/* only APUs have vddnb */
 	if (!(adev->flags & AMD_IS_APU) &&
@@ -2039,9 +2201,7 @@ void amdgpu_dpm_thermal_work_handler(struct work_struct *work)
 	if (!adev->pm.dpm_enabled)
 		return;
 
-	if (adev->powerplay.pp_funcs &&
-	    adev->powerplay.pp_funcs->read_sensor &&
-	    !amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GPU_TEMP,
+	if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GPU_TEMP,
 				    (void *)&temp, &size)) {
 		if (temp < adev->pm.dpm.thermal.min_temp)
 			/* switch back the user state */
@@ -2267,7 +2427,13 @@ static void amdgpu_dpm_change_power_state_locked(struct amdgpu_device *adev)
 
 void amdgpu_dpm_enable_uvd(struct amdgpu_device *adev, bool enable)
 {
-	if (adev->powerplay.pp_funcs->set_powergating_by_smu) {
+	int ret = 0;
+	if (is_support_sw_smu(adev)) {
+	    ret = smu_dpm_set_power_gate(&adev->smu, AMD_IP_BLOCK_TYPE_UVD, enable);
+	    if (ret)
+		DRM_ERROR("[SW SMU]: dpm enable uvd failed, state = %s, ret = %d. \n",
+			  enable ? "true" : "false", ret);
+	} else if (adev->powerplay.pp_funcs->set_powergating_by_smu) {
 		/* enable/disable UVD */
 		mutex_lock(&adev->pm.mutex);
 		amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_UVD, !enable);
@@ -2288,7 +2454,13 @@ void amdgpu_dpm_enable_uvd(struct amdgpu_device *adev, bool enable)
 
 void amdgpu_dpm_enable_vce(struct amdgpu_device *adev, bool enable)
 {
-	if (adev->powerplay.pp_funcs->set_powergating_by_smu) {
+	int ret = 0;
+	if (is_support_sw_smu(adev)) {
+	    ret = smu_dpm_set_power_gate(&adev->smu, AMD_IP_BLOCK_TYPE_VCE, enable);
+	    if (ret)
+		DRM_ERROR("[SW SMU]: dpm enable vce failed, state = %s, ret = %d. \n",
+			  enable ? "true" : "false", ret);
+	} else if (adev->powerplay.pp_funcs->set_powergating_by_smu) {
 		/* enable/disable VCE */
 		mutex_lock(&adev->pm.mutex);
 		amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_VCE, !enable);
@@ -2413,7 +2585,8 @@ int amdgpu_pm_sysfs_init(struct amdgpu_device *adev)
 				"pp_power_profile_mode\n");
 		return ret;
 	}
-	if (hwmgr->od_enabled) {
+	if ((is_support_sw_smu(adev) && adev->smu.od_enabled) ||
+	    (!is_support_sw_smu(adev) && hwmgr->od_enabled)) {
 		ret = device_create_file(adev->dev,
 				&dev_attr_pp_od_clk_voltage);
 		if (ret) {
@@ -2489,7 +2662,8 @@ void amdgpu_pm_sysfs_fini(struct amdgpu_device *adev)
 	device_remove_file(adev->dev, &dev_attr_pp_mclk_od);
 	device_remove_file(adev->dev,
 			&dev_attr_pp_power_profile_mode);
-	if (hwmgr->od_enabled)
+	if ((is_support_sw_smu(adev) && adev->smu.od_enabled) ||
+	    (!is_support_sw_smu(adev) && hwmgr->od_enabled))
 		device_remove_file(adev->dev,
 				&dev_attr_pp_od_clk_voltage);
 	device_remove_file(adev->dev, &dev_attr_gpu_busy_percent);
@@ -2516,28 +2690,38 @@ void amdgpu_pm_compute_clocks(struct amdgpu_device *adev)
 			amdgpu_fence_wait_empty(ring);
 	}
 
-	if (adev->powerplay.pp_funcs->dispatch_tasks) {
-		if (!amdgpu_device_has_dc_support(adev)) {
+	if (is_support_sw_smu(adev)) {
+		struct smu_context *smu = &adev->smu;
+		struct smu_dpm_context *smu_dpm = &adev->smu.smu_dpm;
+		mutex_lock(&(smu->mutex));
+		smu_handle_task(&adev->smu,
+				smu_dpm->dpm_level,
+				AMD_PP_TASK_DISPLAY_CONFIG_CHANGE);
+		mutex_unlock(&(smu->mutex));
+	} else {
+		if (adev->powerplay.pp_funcs->dispatch_tasks) {
+			if (!amdgpu_device_has_dc_support(adev)) {
+				mutex_lock(&adev->pm.mutex);
+				amdgpu_dpm_get_active_displays(adev);
+				adev->pm.pm_display_cfg.num_display = adev->pm.dpm.new_active_crtc_count;
+				adev->pm.pm_display_cfg.vrefresh = amdgpu_dpm_get_vrefresh(adev);
+				adev->pm.pm_display_cfg.min_vblank_time = amdgpu_dpm_get_vblank_time(adev);
+				/* we have issues with mclk switching with refresh rates over 120 hz on the non-DC code. */
+				if (adev->pm.pm_display_cfg.vrefresh > 120)
+					adev->pm.pm_display_cfg.min_vblank_time = 0;
+				if (adev->powerplay.pp_funcs->display_configuration_change)
+					adev->powerplay.pp_funcs->display_configuration_change(
+									adev->powerplay.pp_handle,
+									&adev->pm.pm_display_cfg);
+				mutex_unlock(&adev->pm.mutex);
+			}
+			amdgpu_dpm_dispatch_task(adev, AMD_PP_TASK_DISPLAY_CONFIG_CHANGE, NULL);
+		} else {
 			mutex_lock(&adev->pm.mutex);
 			amdgpu_dpm_get_active_displays(adev);
-			adev->pm.pm_display_cfg.num_display = adev->pm.dpm.new_active_crtc_count;
-			adev->pm.pm_display_cfg.vrefresh = amdgpu_dpm_get_vrefresh(adev);
-			adev->pm.pm_display_cfg.min_vblank_time = amdgpu_dpm_get_vblank_time(adev);
-			/* we have issues with mclk switching with refresh rates over 120 hz on the non-DC code. */
-			if (adev->pm.pm_display_cfg.vrefresh > 120)
-				adev->pm.pm_display_cfg.min_vblank_time = 0;
-			if (adev->powerplay.pp_funcs->display_configuration_change)
-				adev->powerplay.pp_funcs->display_configuration_change(
-								adev->powerplay.pp_handle,
-								&adev->pm.pm_display_cfg);
+			amdgpu_dpm_change_power_state_locked(adev);
 			mutex_unlock(&adev->pm.mutex);
 		}
-		amdgpu_dpm_dispatch_task(adev, AMD_PP_TASK_DISPLAY_CONFIG_CHANGE, NULL);
-	} else {
-		mutex_lock(&adev->pm.mutex);
-		amdgpu_dpm_get_active_displays(adev);
-		amdgpu_dpm_change_power_state_locked(adev);
-		mutex_unlock(&adev->pm.mutex);
 	}
 }
 
@@ -2553,11 +2737,6 @@ static int amdgpu_debugfs_pm_info_pp(struct seq_file *m, struct amdgpu_device *a
 	uint32_t query = 0;
 	int size;
 
-	/* sanity check PP is enabled */
-	if (!(adev->powerplay.pp_funcs &&
-	      adev->powerplay.pp_funcs->read_sensor))
-	      return -EINVAL;
-
 	/* GPU Clocks */
 	size = sizeof(value);
 	seq_printf(m, "GFX Clocks and Power:\n");
@@ -2649,7 +2828,7 @@ static int amdgpu_debugfs_pm_info(struct seq_file *m, void *data)
 	if  ((adev->flags & AMD_IS_PX) &&
 	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON)) {
 		seq_printf(m, "PX asic powered off\n");
-	} else if (adev->powerplay.pp_funcs->debugfs_print_current_performance_level) {
+	} else if (!is_support_sw_smu(adev) && adev->powerplay.pp_funcs->debugfs_print_current_performance_level) {
 		mutex_lock(&adev->pm.mutex);
 		if (adev->powerplay.pp_funcs->debugfs_print_current_performance_level)
 			adev->powerplay.pp_funcs->debugfs_print_current_performance_level(adev, m);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 3091488cd8cc..905cce1814f3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -120,6 +120,7 @@ psp_cmd_submit_buf(struct psp_context *psp,
 {
 	int ret;
 	int index;
+	int timeout = 2000;
 
 	memset(psp->cmd_buf_mem, 0, PSP_CMD_BUFFER_SIZE);
 
@@ -133,8 +134,11 @@ psp_cmd_submit_buf(struct psp_context *psp,
 		return ret;
 	}
 
-	while (*((unsigned int *)psp->fence_buf) != index)
+	while (*((unsigned int *)psp->fence_buf) != index) {
+		if (--timeout == 0)
+			break;
 		msleep(1);
+	}
 
 	/* In some cases, psp response status is not 0 even there is no
 	 * problem while the command is submitted. Some version of PSP FW
@@ -143,12 +147,14 @@ psp_cmd_submit_buf(struct psp_context *psp,
 	 * during psp initialization to avoid breaking hw_init and it doesn't
 	 * return -EINVAL.
 	 */
-	if (psp->cmd_buf_mem->resp.status) {
+	if (psp->cmd_buf_mem->resp.status || !timeout) {
 		if (ucode)
 			DRM_WARN("failed to load ucode id (%d) ",
 				  ucode->ucode_id);
 		DRM_WARN("psp command failed and response status is (%d)\n",
 			  psp->cmd_buf_mem->resp.status);
+		if (!timeout)
+			return -EINVAL;
 	}
 
 	/* get xGMI session id from response buffer */
@@ -181,13 +187,13 @@ static int psp_tmr_init(struct psp_context *psp)
 	int ret;
 
 	/*
-	 * Allocate 3M memory aligned to 1M from Frame Buffer (local
-	 * physical).
+	 * According to HW engineer, they prefer the TMR address be "naturally
+	 * aligned" , e.g. the start address be an integer divide of TMR size.
 	 *
 	 * Note: this memory need be reserved till the driver
 	 * uninitializes.
 	 */
-	ret = amdgpu_bo_create_kernel(psp->adev, PSP_TMR_SIZE, 0x100000,
+	ret = amdgpu_bo_create_kernel(psp->adev, PSP_TMR_SIZE, PSP_TMR_SIZE,
 				      AMDGPU_GEM_DOMAIN_VRAM,
 				      &psp->tmr_bo, &psp->tmr_mc_addr, &psp->tmr_buf);
 
@@ -466,6 +472,206 @@ static int psp_xgmi_initialize(struct psp_context *psp)
 	return ret;
 }
 
+// ras begin
+static void psp_prep_ras_ta_load_cmd_buf(struct psp_gfx_cmd_resp *cmd,
+		uint64_t ras_ta_mc, uint64_t ras_mc_shared,
+		uint32_t ras_ta_size, uint32_t shared_size)
+{
+	cmd->cmd_id = GFX_CMD_ID_LOAD_TA;
+	cmd->cmd.cmd_load_ta.app_phy_addr_lo = lower_32_bits(ras_ta_mc);
+	cmd->cmd.cmd_load_ta.app_phy_addr_hi = upper_32_bits(ras_ta_mc);
+	cmd->cmd.cmd_load_ta.app_len = ras_ta_size;
+
+	cmd->cmd.cmd_load_ta.cmd_buf_phy_addr_lo = lower_32_bits(ras_mc_shared);
+	cmd->cmd.cmd_load_ta.cmd_buf_phy_addr_hi = upper_32_bits(ras_mc_shared);
+	cmd->cmd.cmd_load_ta.cmd_buf_len = shared_size;
+}
+
+static int psp_ras_init_shared_buf(struct psp_context *psp)
+{
+	int ret;
+
+	/*
+	 * Allocate 16k memory aligned to 4k from Frame Buffer (local
+	 * physical) for ras ta <-> Driver
+	 */
+	ret = amdgpu_bo_create_kernel(psp->adev, PSP_RAS_SHARED_MEM_SIZE,
+			PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM,
+			&psp->ras.ras_shared_bo,
+			&psp->ras.ras_shared_mc_addr,
+			&psp->ras.ras_shared_buf);
+
+	return ret;
+}
+
+static int psp_ras_load(struct psp_context *psp)
+{
+	int ret;
+	struct psp_gfx_cmd_resp *cmd;
+
+	/*
+	 * TODO: bypass the loading in sriov for now
+	 */
+	if (amdgpu_sriov_vf(psp->adev))
+		return 0;
+
+	cmd = kzalloc(sizeof(struct psp_gfx_cmd_resp), GFP_KERNEL);
+	if (!cmd)
+		return -ENOMEM;
+
+	memset(psp->fw_pri_buf, 0, PSP_1_MEG);
+	memcpy(psp->fw_pri_buf, psp->ta_ras_start_addr, psp->ta_ras_ucode_size);
+
+	psp_prep_ras_ta_load_cmd_buf(cmd, psp->fw_pri_mc_addr,
+			psp->ras.ras_shared_mc_addr,
+			psp->ta_ras_ucode_size, PSP_RAS_SHARED_MEM_SIZE);
+
+	ret = psp_cmd_submit_buf(psp, NULL, cmd,
+			psp->fence_buf_mc_addr);
+
+	if (!ret) {
+		psp->ras.ras_initialized = 1;
+		psp->ras.session_id = cmd->resp.session_id;
+	}
+
+	kfree(cmd);
+
+	return ret;
+}
+
+static void psp_prep_ras_ta_unload_cmd_buf(struct psp_gfx_cmd_resp *cmd,
+						uint32_t ras_session_id)
+{
+	cmd->cmd_id = GFX_CMD_ID_UNLOAD_TA;
+	cmd->cmd.cmd_unload_ta.session_id = ras_session_id;
+}
+
+static int psp_ras_unload(struct psp_context *psp)
+{
+	int ret;
+	struct psp_gfx_cmd_resp *cmd;
+
+	/*
+	 * TODO: bypass the unloading in sriov for now
+	 */
+	if (amdgpu_sriov_vf(psp->adev))
+		return 0;
+
+	cmd = kzalloc(sizeof(struct psp_gfx_cmd_resp), GFP_KERNEL);
+	if (!cmd)
+		return -ENOMEM;
+
+	psp_prep_ras_ta_unload_cmd_buf(cmd, psp->ras.session_id);
+
+	ret = psp_cmd_submit_buf(psp, NULL, cmd,
+			psp->fence_buf_mc_addr);
+
+	kfree(cmd);
+
+	return ret;
+}
+
+static void psp_prep_ras_ta_invoke_cmd_buf(struct psp_gfx_cmd_resp *cmd,
+		uint32_t ta_cmd_id,
+		uint32_t ras_session_id)
+{
+	cmd->cmd_id = GFX_CMD_ID_INVOKE_CMD;
+	cmd->cmd.cmd_invoke_cmd.session_id = ras_session_id;
+	cmd->cmd.cmd_invoke_cmd.ta_cmd_id = ta_cmd_id;
+	/* Note: cmd_invoke_cmd.buf is not used for now */
+}
+
+int psp_ras_invoke(struct psp_context *psp, uint32_t ta_cmd_id)
+{
+	int ret;
+	struct psp_gfx_cmd_resp *cmd;
+
+	/*
+	 * TODO: bypass the loading in sriov for now
+	 */
+	if (amdgpu_sriov_vf(psp->adev))
+		return 0;
+
+	cmd = kzalloc(sizeof(struct psp_gfx_cmd_resp), GFP_KERNEL);
+	if (!cmd)
+		return -ENOMEM;
+
+	psp_prep_ras_ta_invoke_cmd_buf(cmd, ta_cmd_id,
+			psp->ras.session_id);
+
+	ret = psp_cmd_submit_buf(psp, NULL, cmd,
+			psp->fence_buf_mc_addr);
+
+	kfree(cmd);
+
+	return ret;
+}
+
+int psp_ras_enable_features(struct psp_context *psp,
+		union ta_ras_cmd_input *info, bool enable)
+{
+	struct ta_ras_shared_memory *ras_cmd;
+	int ret;
+
+	if (!psp->ras.ras_initialized)
+		return -EINVAL;
+
+	ras_cmd = (struct ta_ras_shared_memory *)psp->ras.ras_shared_buf;
+	memset(ras_cmd, 0, sizeof(struct ta_ras_shared_memory));
+
+	if (enable)
+		ras_cmd->cmd_id = TA_RAS_COMMAND__ENABLE_FEATURES;
+	else
+		ras_cmd->cmd_id = TA_RAS_COMMAND__DISABLE_FEATURES;
+
+	ras_cmd->ras_in_message = *info;
+
+	ret = psp_ras_invoke(psp, ras_cmd->cmd_id);
+	if (ret)
+		return -EINVAL;
+
+	return ras_cmd->ras_status;
+}
+
+static int psp_ras_terminate(struct psp_context *psp)
+{
+	int ret;
+
+	if (!psp->ras.ras_initialized)
+		return 0;
+
+	ret = psp_ras_unload(psp);
+	if (ret)
+		return ret;
+
+	psp->ras.ras_initialized = 0;
+
+	/* free ras shared memory */
+	amdgpu_bo_free_kernel(&psp->ras.ras_shared_bo,
+			&psp->ras.ras_shared_mc_addr,
+			&psp->ras.ras_shared_buf);
+
+	return 0;
+}
+
+static int psp_ras_initialize(struct psp_context *psp)
+{
+	int ret;
+
+	if (!psp->ras.ras_initialized) {
+		ret = psp_ras_init_shared_buf(psp);
+		if (ret)
+			return ret;
+	}
+
+	ret = psp_ras_load(psp);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+// ras end
+
 static int psp_hw_start(struct psp_context *psp)
 {
 	struct amdgpu_device *adev = psp->adev;
@@ -473,25 +679,35 @@ static int psp_hw_start(struct psp_context *psp)
 
 	if (!amdgpu_sriov_vf(adev) || !adev->in_gpu_reset) {
 		ret = psp_bootloader_load_sysdrv(psp);
-		if (ret)
+		if (ret) {
+			DRM_ERROR("PSP load sysdrv failed!\n");
 			return ret;
+		}
 
 		ret = psp_bootloader_load_sos(psp);
-		if (ret)
+		if (ret) {
+			DRM_ERROR("PSP load sos failed!\n");
 			return ret;
+		}
 	}
 
 	ret = psp_ring_create(psp, PSP_RING_TYPE__KM);
-	if (ret)
+	if (ret) {
+		DRM_ERROR("PSP create ring failed!\n");
 		return ret;
+	}
 
 	ret = psp_tmr_load(psp);
-	if (ret)
+	if (ret) {
+		DRM_ERROR("PSP load tmr failed!\n");
 		return ret;
+	}
 
 	ret = psp_asd_load(psp);
-	if (ret)
+	if (ret) {
+		DRM_ERROR("PSP load asd failed!\n");
 		return ret;
+	}
 
 	if (adev->gmc.xgmi.num_physical_nodes > 1) {
 		ret = psp_xgmi_initialize(psp);
@@ -502,6 +718,15 @@ static int psp_hw_start(struct psp_context *psp)
 			dev_err(psp->adev->dev,
 				"XGMI: Failed to initialize XGMI session\n");
 	}
+
+
+	if (psp->adev->psp.ta_fw) {
+		ret = psp_ras_initialize(psp);
+		if (ret)
+			dev_err(psp->adev->dev,
+					"RAS: Failed to initialize RAS\n");
+	}
+
 	return 0;
 }
 
@@ -665,53 +890,52 @@ static int psp_load_fw(struct amdgpu_device *adev)
 					&psp->fence_buf_mc_addr,
 					&psp->fence_buf);
 	if (ret)
-		goto failed_mem2;
+		goto failed;
 
 	ret = amdgpu_bo_create_kernel(adev, PSP_CMD_BUFFER_SIZE, PAGE_SIZE,
 				      AMDGPU_GEM_DOMAIN_VRAM,
 				      &psp->cmd_buf_bo, &psp->cmd_buf_mc_addr,
 				      (void **)&psp->cmd_buf_mem);
 	if (ret)
-		goto failed_mem1;
+		goto failed;
 
 	memset(psp->fence_buf, 0, PSP_FENCE_BUFFER_SIZE);
 
 	ret = psp_ring_init(psp, PSP_RING_TYPE__KM);
-	if (ret)
-		goto failed_mem;
+	if (ret) {
+		DRM_ERROR("PSP ring init failed!\n");
+		goto failed;
+	}
 
 	ret = psp_tmr_init(psp);
-	if (ret)
-		goto failed_mem;
+	if (ret) {
+		DRM_ERROR("PSP tmr init failed!\n");
+		goto failed;
+	}
 
 	ret = psp_asd_init(psp);
-	if (ret)
-		goto failed_mem;
+	if (ret) {
+		DRM_ERROR("PSP asd init failed!\n");
+		goto failed;
+	}
 
 skip_memalloc:
 	ret = psp_hw_start(psp);
 	if (ret)
-		goto failed_mem;
+		goto failed;
 
 	ret = psp_np_fw_load(psp);
 	if (ret)
-		goto failed_mem;
+		goto failed;
 
 	return 0;
 
-failed_mem:
-	amdgpu_bo_free_kernel(&psp->cmd_buf_bo,
-			      &psp->cmd_buf_mc_addr,
-			      (void **)&psp->cmd_buf_mem);
-failed_mem1:
-	amdgpu_bo_free_kernel(&psp->fence_buf_bo,
-			      &psp->fence_buf_mc_addr, &psp->fence_buf);
-failed_mem2:
-	amdgpu_bo_free_kernel(&psp->fw_pri_bo,
-			      &psp->fw_pri_mc_addr, &psp->fw_pri_buf);
 failed:
-	kfree(psp->cmd);
-	psp->cmd = NULL;
+	/*
+	 * all cleanup jobs (xgmi terminate, ras terminate,
+	 * ring destroy, cmd/fence/fw buffers destory,
+	 * psp->cmd destory) are delayed to psp_hw_fini
+	 */
 	return ret;
 }
 
@@ -753,6 +977,9 @@ static int psp_hw_fini(void *handle)
 	    psp->xgmi_context.initialized == 1)
                 psp_xgmi_terminate(psp);
 
+	if (psp->adev->psp.ta_fw)
+		psp_ras_terminate(psp);
+
 	psp_ring_destroy(psp, PSP_RING_TYPE__KM);
 
 	amdgpu_bo_free_kernel(&psp->tmr_bo, &psp->tmr_mc_addr, &psp->tmr_buf);
@@ -786,6 +1013,14 @@ static int psp_suspend(void *handle)
 		}
 	}
 
+	if (psp->adev->psp.ta_fw) {
+		ret = psp_ras_terminate(psp);
+		if (ret) {
+			DRM_ERROR("Failed to terminate ras ta\n");
+			return ret;
+		}
+	}
+
 	ret = psp_ring_stop(psp, PSP_RING_TYPE__KM);
 	if (ret) {
 		DRM_ERROR("PSP ring stop failed\n");
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
index 2ef98cc755d6..cde113f07c96 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
@@ -28,11 +28,13 @@
 #include "amdgpu.h"
 #include "psp_gfx_if.h"
 #include "ta_xgmi_if.h"
+#include "ta_ras_if.h"
 
 #define PSP_FENCE_BUFFER_SIZE	0x1000
 #define PSP_CMD_BUFFER_SIZE	0x1000
 #define PSP_ASD_SHARED_MEM_SIZE 0x4000
 #define PSP_XGMI_SHARED_MEM_SIZE 0x4000
+#define PSP_RAS_SHARED_MEM_SIZE 0x4000
 #define PSP_1_MEG		0x100000
 #define PSP_TMR_SIZE	0x400000
 
@@ -88,6 +90,9 @@ struct psp_funcs
 	int (*xgmi_set_topology_info)(struct psp_context *psp, int number_devices,
 				      struct psp_xgmi_topology_info *topology);
 	bool (*support_vmr_ring)(struct psp_context *psp);
+	int (*ras_trigger_error)(struct psp_context *psp,
+			struct ta_ras_trigger_error_input *info);
+	int (*ras_cure_posion)(struct psp_context *psp, uint64_t *mode_ptr);
 };
 
 struct psp_xgmi_context {
@@ -98,6 +103,16 @@ struct psp_xgmi_context {
 	void                            *xgmi_shared_buf;
 };
 
+struct psp_ras_context {
+	/*ras fw*/
+	bool			ras_initialized;
+	uint32_t		session_id;
+	struct amdgpu_bo	*ras_shared_bo;
+	uint64_t		ras_shared_mc_addr;
+	void			*ras_shared_buf;
+	struct amdgpu_ras	*ras;
+};
+
 struct psp_context
 {
 	struct amdgpu_device            *adev;
@@ -150,10 +165,15 @@ struct psp_context
 
 	/* xgmi ta firmware and buffer */
 	const struct firmware		*ta_fw;
+	uint32_t			ta_fw_version;
 	uint32_t			ta_xgmi_ucode_version;
 	uint32_t			ta_xgmi_ucode_size;
 	uint8_t				*ta_xgmi_start_addr;
+	uint32_t			ta_ras_ucode_version;
+	uint32_t			ta_ras_ucode_size;
+	uint8_t				*ta_ras_start_addr;
 	struct psp_xgmi_context		xgmi_context;
+	struct psp_ras_context		ras;
 };
 
 struct amdgpu_psp_funcs {
@@ -207,6 +227,13 @@ struct psp_xgmi_topology_info {
 
 #define amdgpu_psp_check_fw_loading_status(adev, i) (adev)->firmware.funcs->check_fw_loading_status((adev), (i))
 
+#define psp_ras_trigger_error(psp, info) \
+	((psp)->funcs->ras_trigger_error ? \
+	(psp)->funcs->ras_trigger_error((psp), (info)) : -EINVAL)
+#define psp_ras_cure_posion(psp, addr) \
+	((psp)->funcs->ras_cure_posion ? \
+	(psp)->funcs->ras_cure_posion(psp, (addr)) : -EINVAL)
+
 extern const struct amd_ip_funcs psp_ip_funcs;
 
 extern const struct amdgpu_ip_block_version psp_v3_1_ip_block;
@@ -217,6 +244,11 @@ extern const struct amdgpu_ip_block_version psp_v10_0_ip_block;
 
 int psp_gpu_reset(struct amdgpu_device *adev);
 int psp_xgmi_invoke(struct psp_context *psp, uint32_t ta_cmd_id);
+
+int psp_ras_invoke(struct psp_context *psp, uint32_t ta_cmd_id);
+int psp_ras_enable_features(struct psp_context *psp,
+		union ta_ras_cmd_input *info, bool enable);
+
 extern const struct amdgpu_ip_block_version psp_v11_0_ip_block;
 
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
new file mode 100644
index 000000000000..22bd21efe6b1
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -0,0 +1,1482 @@
+/*
+ * Copyright 2018 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ *
+ */
+#include <linux/debugfs.h>
+#include <linux/list.h>
+#include <linux/module.h>
+#include "amdgpu.h"
+#include "amdgpu_ras.h"
+#include "amdgpu_atomfirmware.h"
+
+struct ras_ih_data {
+	/* interrupt bottom half */
+	struct work_struct ih_work;
+	int inuse;
+	/* IP callback */
+	ras_ih_cb cb;
+	/* full of entries */
+	unsigned char *ring;
+	unsigned int ring_size;
+	unsigned int element_size;
+	unsigned int aligned_element_size;
+	unsigned int rptr;
+	unsigned int wptr;
+};
+
+struct ras_fs_data {
+	char sysfs_name[32];
+	char debugfs_name[32];
+};
+
+struct ras_err_data {
+	unsigned long ue_count;
+	unsigned long ce_count;
+};
+
+struct ras_err_handler_data {
+	/* point to bad pages array */
+	struct {
+		unsigned long bp;
+		struct amdgpu_bo *bo;
+	} *bps;
+	/* the count of entries */
+	int count;
+	/* the space can place new entries */
+	int space_left;
+	/* last reserved entry's index + 1 */
+	int last_reserved;
+};
+
+struct ras_manager {
+	struct ras_common_if head;
+	/* reference count */
+	int use;
+	/* ras block link */
+	struct list_head node;
+	/* the device */
+	struct amdgpu_device *adev;
+	/* debugfs */
+	struct dentry *ent;
+	/* sysfs */
+	struct device_attribute sysfs_attr;
+	int attr_inuse;
+
+	/* fs node name */
+	struct ras_fs_data fs_data;
+
+	/* IH data */
+	struct ras_ih_data ih_data;
+
+	struct ras_err_data err_data;
+};
+
+const char *ras_error_string[] = {
+	"none",
+	"parity",
+	"single_correctable",
+	"multi_uncorrectable",
+	"poison",
+};
+
+const char *ras_block_string[] = {
+	"umc",
+	"sdma",
+	"gfx",
+	"mmhub",
+	"athub",
+	"pcie_bif",
+	"hdp",
+	"xgmi_wafl",
+	"df",
+	"smn",
+	"sem",
+	"mp0",
+	"mp1",
+	"fuse",
+};
+
+#define ras_err_str(i) (ras_error_string[ffs(i)])
+#define ras_block_str(i) (ras_block_string[i])
+
+#define AMDGPU_RAS_FLAG_INIT_BY_VBIOS 1
+#define RAS_DEFAULT_FLAGS (AMDGPU_RAS_FLAG_INIT_BY_VBIOS)
+
+static void amdgpu_ras_self_test(struct amdgpu_device *adev)
+{
+	/* TODO */
+}
+
+static ssize_t amdgpu_ras_debugfs_read(struct file *f, char __user *buf,
+					size_t size, loff_t *pos)
+{
+	struct ras_manager *obj = (struct ras_manager *)file_inode(f)->i_private;
+	struct ras_query_if info = {
+		.head = obj->head,
+	};
+	ssize_t s;
+	char val[128];
+
+	if (amdgpu_ras_error_query(obj->adev, &info))
+		return -EINVAL;
+
+	s = snprintf(val, sizeof(val), "%s: %lu\n%s: %lu\n",
+			"ue", info.ue_count,
+			"ce", info.ce_count);
+	if (*pos >= s)
+		return 0;
+
+	s -= *pos;
+	s = min_t(u64, s, size);
+
+
+	if (copy_to_user(buf, &val[*pos], s))
+		return -EINVAL;
+
+	*pos += s;
+
+	return s;
+}
+
+static const struct file_operations amdgpu_ras_debugfs_ops = {
+	.owner = THIS_MODULE,
+	.read = amdgpu_ras_debugfs_read,
+	.write = NULL,
+	.llseek = default_llseek
+};
+
+static int amdgpu_ras_find_block_id_by_name(const char *name, int *block_id)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(ras_block_string); i++) {
+		*block_id = i;
+		if (strcmp(name, ras_block_str(i)) == 0)
+			return 0;
+	}
+	return -EINVAL;
+}
+
+static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f,
+		const char __user *buf, size_t size,
+		loff_t *pos, struct ras_debug_if *data)
+{
+	ssize_t s = min_t(u64, 64, size);
+	char str[65];
+	char block_name[33];
+	char err[9] = "ue";
+	int op = -1;
+	int block_id;
+	u64 address, value;
+
+	if (*pos)
+		return -EINVAL;
+	*pos = size;
+
+	memset(str, 0, sizeof(str));
+	memset(data, 0, sizeof(*data));
+
+	if (copy_from_user(str, buf, s))
+		return -EINVAL;
+
+	if (sscanf(str, "disable %32s", block_name) == 1)
+		op = 0;
+	else if (sscanf(str, "enable %32s %8s", block_name, err) == 2)
+		op = 1;
+	else if (sscanf(str, "inject %32s %8s", block_name, err) == 2)
+		op = 2;
+	else if (str[0] && str[1] && str[2] && str[3])
+		/* ascii string, but commands are not matched. */
+		return -EINVAL;
+
+	if (op != -1) {
+		if (amdgpu_ras_find_block_id_by_name(block_name, &block_id))
+			return -EINVAL;
+
+		data->head.block = block_id;
+		data->head.type = memcmp("ue", err, 2) == 0 ?
+			AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE :
+			AMDGPU_RAS_ERROR__SINGLE_CORRECTABLE;
+		data->op = op;
+
+		if (op == 2) {
+			if (sscanf(str, "%*s %*s %*s %llu %llu",
+						&address, &value) != 2)
+				if (sscanf(str, "%*s %*s %*s 0x%llx 0x%llx",
+							&address, &value) != 2)
+					return -EINVAL;
+			data->inject.address = address;
+			data->inject.value = value;
+		}
+	} else {
+		if (size < sizeof(*data))
+			return -EINVAL;
+
+		if (copy_from_user(data, buf, sizeof(*data)))
+			return -EINVAL;
+	}
+
+	return 0;
+}
+/*
+ * DOC: ras debugfs control interface
+ *
+ * It accepts struct ras_debug_if who has two members.
+ *
+ * First member: ras_debug_if::head or ras_debug_if::inject.
+ *
+ * head is used to indicate which IP block will be under control.
+ *
+ * head has four members, they are block, type, sub_block_index, name.
+ * block: which IP will be under control.
+ * type: what kind of error will be enabled/disabled/injected.
+ * sub_block_index: some IPs have subcomponets. say, GFX, sDMA.
+ * name: the name of IP.
+ *
+ * inject has two more members than head, they are address, value.
+ * As their names indicate, inject operation will write the
+ * value to the address.
+ *
+ * Second member: struct ras_debug_if::op.
+ * It has three kinds of operations.
+ *  0: disable RAS on the block. Take ::head as its data.
+ *  1: enable RAS on the block. Take ::head as its data.
+ *  2: inject errors on the block. Take ::inject as its data.
+ *
+ * How to use the interface?
+ * programs:
+ * copy the struct ras_debug_if in your codes and initialize it.
+ * write the struct to the control node.
+ *
+ * bash:
+ * echo op block [error [address value]] > .../ras/ras_ctrl
+ *	op: disable, enable, inject
+ *		disable: only block is needed
+ *		enable: block and error are needed
+ *		inject: error, address, value are needed
+ *	block: umc, smda, gfx, .........
+ *		see ras_block_string[] for details
+ *	error: ue, ce
+ *		ue: multi_uncorrectable
+ *		ce: single_correctable
+ *
+ * here are some examples for bash commands,
+ *	echo inject umc ue 0x0 0x0 > /sys/kernel/debug/dri/0/ras/ras_ctrl
+ *	echo inject umc ce 0 0 > /sys/kernel/debug/dri/0/ras/ras_ctrl
+ *	echo disable umc > /sys/kernel/debug/dri/0/ras/ras_ctrl
+ *
+ * How to check the result?
+ *
+ * For disable/enable, please check ras features at
+ * /sys/class/drm/card[0/1/2...]/device/ras/features
+ *
+ * For inject, please check corresponding err count at
+ * /sys/class/drm/card[0/1/2...]/device/ras/[gfx/sdma/...]_err_count
+ *
+ * NOTE: operation is only allowed on blocks which are supported.
+ * Please check ras mask at /sys/module/amdgpu/parameters/ras_mask
+ */
+static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f, const char __user *buf,
+		size_t size, loff_t *pos)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)file_inode(f)->i_private;
+	struct ras_debug_if data;
+	int ret = 0;
+
+	ret = amdgpu_ras_debugfs_ctrl_parse_data(f, buf, size, pos, &data);
+	if (ret)
+		return -EINVAL;
+
+	if (!amdgpu_ras_is_supported(adev, data.head.block))
+		return -EINVAL;
+
+	switch (data.op) {
+	case 0:
+		ret = amdgpu_ras_feature_enable(adev, &data.head, 0);
+		break;
+	case 1:
+		ret = amdgpu_ras_feature_enable(adev, &data.head, 1);
+		break;
+	case 2:
+		ret = amdgpu_ras_error_inject(adev, &data.inject);
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	};
+
+	if (ret)
+		return -EINVAL;
+
+	return size;
+}
+
+static const struct file_operations amdgpu_ras_debugfs_ctrl_ops = {
+	.owner = THIS_MODULE,
+	.read = NULL,
+	.write = amdgpu_ras_debugfs_ctrl_write,
+	.llseek = default_llseek
+};
+
+static ssize_t amdgpu_ras_sysfs_read(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	struct ras_manager *obj = container_of(attr, struct ras_manager, sysfs_attr);
+	struct ras_query_if info = {
+		.head = obj->head,
+	};
+
+	if (amdgpu_ras_error_query(obj->adev, &info))
+		return -EINVAL;
+
+	return snprintf(buf, PAGE_SIZE, "%s: %lu\n%s: %lu\n",
+			"ue", info.ue_count,
+			"ce", info.ce_count);
+}
+
+/* obj begin */
+
+#define get_obj(obj) do { (obj)->use++; } while (0)
+#define alive_obj(obj) ((obj)->use)
+
+static inline void put_obj(struct ras_manager *obj)
+{
+	if (obj && --obj->use == 0)
+		list_del(&obj->node);
+	if (obj && obj->use < 0) {
+		 DRM_ERROR("RAS ERROR: Unbalance obj(%s) use\n", obj->head.name);
+	}
+}
+
+/* make one obj and return it. */
+static struct ras_manager *amdgpu_ras_create_obj(struct amdgpu_device *adev,
+		struct ras_common_if *head)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct ras_manager *obj;
+
+	if (!con)
+		return NULL;
+
+	if (head->block >= AMDGPU_RAS_BLOCK_COUNT)
+		return NULL;
+
+	obj = &con->objs[head->block];
+	/* already exist. return obj? */
+	if (alive_obj(obj))
+		return NULL;
+
+	obj->head = *head;
+	obj->adev = adev;
+	list_add(&obj->node, &con->head);
+	get_obj(obj);
+
+	return obj;
+}
+
+/* return an obj equal to head, or the first when head is NULL */
+static struct ras_manager *amdgpu_ras_find_obj(struct amdgpu_device *adev,
+		struct ras_common_if *head)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct ras_manager *obj;
+	int i;
+
+	if (!con)
+		return NULL;
+
+	if (head) {
+		if (head->block >= AMDGPU_RAS_BLOCK_COUNT)
+			return NULL;
+
+		obj = &con->objs[head->block];
+
+		if (alive_obj(obj)) {
+			WARN_ON(head->block != obj->head.block);
+			return obj;
+		}
+	} else {
+		for (i = 0; i < AMDGPU_RAS_BLOCK_COUNT; i++) {
+			obj = &con->objs[i];
+			if (alive_obj(obj)) {
+				WARN_ON(i != obj->head.block);
+				return obj;
+			}
+		}
+	}
+
+	return NULL;
+}
+/* obj end */
+
+/* feature ctl begin */
+static int amdgpu_ras_is_feature_allowed(struct amdgpu_device *adev,
+		struct ras_common_if *head)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+
+	return con->hw_supported & BIT(head->block);
+}
+
+static int amdgpu_ras_is_feature_enabled(struct amdgpu_device *adev,
+		struct ras_common_if *head)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+
+	return con->features & BIT(head->block);
+}
+
+/*
+ * if obj is not created, then create one.
+ * set feature enable flag.
+ */
+static int __amdgpu_ras_feature_enable(struct amdgpu_device *adev,
+		struct ras_common_if *head, int enable)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct ras_manager *obj = amdgpu_ras_find_obj(adev, head);
+
+	/* If hardware does not support ras, then do not create obj.
+	 * But if hardware support ras, we can create the obj.
+	 * Ras framework checks con->hw_supported to see if it need do
+	 * corresponding initialization.
+	 * IP checks con->support to see if it need disable ras.
+	 */
+	if (!amdgpu_ras_is_feature_allowed(adev, head))
+		return 0;
+	if (!(!!enable ^ !!amdgpu_ras_is_feature_enabled(adev, head)))
+		return 0;
+
+	if (enable) {
+		if (!obj) {
+			obj = amdgpu_ras_create_obj(adev, head);
+			if (!obj)
+				return -EINVAL;
+		} else {
+			/* In case we create obj somewhere else */
+			get_obj(obj);
+		}
+		con->features |= BIT(head->block);
+	} else {
+		if (obj && amdgpu_ras_is_feature_enabled(adev, head)) {
+			con->features &= ~BIT(head->block);
+			put_obj(obj);
+		}
+	}
+
+	return 0;
+}
+
+/* wrapper of psp_ras_enable_features */
+int amdgpu_ras_feature_enable(struct amdgpu_device *adev,
+		struct ras_common_if *head, bool enable)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	union ta_ras_cmd_input info;
+	int ret;
+
+	if (!con)
+		return -EINVAL;
+
+	if (!enable) {
+		info.disable_features = (struct ta_ras_disable_features_input) {
+			.block_id =  amdgpu_ras_block_to_ta(head->block),
+			.error_type = amdgpu_ras_error_to_ta(head->type),
+		};
+	} else {
+		info.enable_features = (struct ta_ras_enable_features_input) {
+			.block_id =  amdgpu_ras_block_to_ta(head->block),
+			.error_type = amdgpu_ras_error_to_ta(head->type),
+		};
+	}
+
+	/* Do not enable if it is not allowed. */
+	WARN_ON(enable && !amdgpu_ras_is_feature_allowed(adev, head));
+	/* Are we alerady in that state we are going to set? */
+	if (!(!!enable ^ !!amdgpu_ras_is_feature_enabled(adev, head)))
+		return 0;
+
+	ret = psp_ras_enable_features(&adev->psp, &info, enable);
+	if (ret) {
+		DRM_ERROR("RAS ERROR: %s %s feature failed ret %d\n",
+				enable ? "enable":"disable",
+				ras_block_str(head->block),
+				ret);
+		return -EINVAL;
+	}
+
+	/* setup the obj */
+	__amdgpu_ras_feature_enable(adev, head, enable);
+
+	return 0;
+}
+
+/* Only used in device probe stage and called only once. */
+int amdgpu_ras_feature_enable_on_boot(struct amdgpu_device *adev,
+		struct ras_common_if *head, bool enable)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	int ret;
+
+	if (!con)
+		return -EINVAL;
+
+	if (con->flags & AMDGPU_RAS_FLAG_INIT_BY_VBIOS) {
+		/* If ras is enabled by vbios, we set up ras object first in
+		 * both case. For enable, that is all what we need do. For
+		 * disable, we need perform a ras TA disable cmd after that.
+		 */
+		ret = __amdgpu_ras_feature_enable(adev, head, 1);
+		if (ret)
+			return ret;
+
+		if (!enable)
+			ret = amdgpu_ras_feature_enable(adev, head, 0);
+	} else
+		ret = amdgpu_ras_feature_enable(adev, head, enable);
+
+	return ret;
+}
+
+static int amdgpu_ras_disable_all_features(struct amdgpu_device *adev,
+		bool bypass)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct ras_manager *obj, *tmp;
+
+	list_for_each_entry_safe(obj, tmp, &con->head, node) {
+		/* bypass psp.
+		 * aka just release the obj and corresponding flags
+		 */
+		if (bypass) {
+			if (__amdgpu_ras_feature_enable(adev, &obj->head, 0))
+				break;
+		} else {
+			if (amdgpu_ras_feature_enable(adev, &obj->head, 0))
+				break;
+		}
+	}
+
+	return con->features;
+}
+
+static int amdgpu_ras_enable_all_features(struct amdgpu_device *adev,
+		bool bypass)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	int ras_block_count = AMDGPU_RAS_BLOCK_COUNT;
+	int i;
+	const enum amdgpu_ras_error_type default_ras_type =
+		AMDGPU_RAS_ERROR__NONE;
+
+	for (i = 0; i < ras_block_count; i++) {
+		struct ras_common_if head = {
+			.block = i,
+			.type = default_ras_type,
+			.sub_block_index = 0,
+		};
+		strcpy(head.name, ras_block_str(i));
+		if (bypass) {
+			/*
+			 * bypass psp. vbios enable ras for us.
+			 * so just create the obj
+			 */
+			if (__amdgpu_ras_feature_enable(adev, &head, 1))
+				break;
+		} else {
+			if (amdgpu_ras_feature_enable(adev, &head, 1))
+				break;
+		}
+	}
+
+	return con->features;
+}
+/* feature ctl end */
+
+/* query/inject/cure begin */
+int amdgpu_ras_error_query(struct amdgpu_device *adev,
+		struct ras_query_if *info)
+{
+	struct ras_manager *obj = amdgpu_ras_find_obj(adev, &info->head);
+
+	if (!obj)
+		return -EINVAL;
+	/* TODO might read the register to read the count */
+
+	info->ue_count = obj->err_data.ue_count;
+	info->ce_count = obj->err_data.ce_count;
+
+	return 0;
+}
+
+/* wrapper of psp_ras_trigger_error */
+int amdgpu_ras_error_inject(struct amdgpu_device *adev,
+		struct ras_inject_if *info)
+{
+	struct ras_manager *obj = amdgpu_ras_find_obj(adev, &info->head);
+	struct ta_ras_trigger_error_input block_info = {
+		.block_id =  amdgpu_ras_block_to_ta(info->head.block),
+		.inject_error_type = amdgpu_ras_error_to_ta(info->head.type),
+		.sub_block_index = info->head.sub_block_index,
+		.address = info->address,
+		.value = info->value,
+	};
+	int ret = 0;
+
+	if (!obj)
+		return -EINVAL;
+
+	ret = psp_ras_trigger_error(&adev->psp, &block_info);
+	if (ret)
+		DRM_ERROR("RAS ERROR: inject %s error failed ret %d\n",
+				ras_block_str(info->head.block),
+				ret);
+
+	return ret;
+}
+
+int amdgpu_ras_error_cure(struct amdgpu_device *adev,
+		struct ras_cure_if *info)
+{
+	/* psp fw has no cure interface for now. */
+	return 0;
+}
+
+/* get the total error counts on all IPs */
+int amdgpu_ras_query_error_count(struct amdgpu_device *adev,
+		bool is_ce)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct ras_manager *obj;
+	struct ras_err_data data = {0, 0};
+
+	if (!con)
+		return -EINVAL;
+
+	list_for_each_entry(obj, &con->head, node) {
+		struct ras_query_if info = {
+			.head = obj->head,
+		};
+
+		if (amdgpu_ras_error_query(adev, &info))
+			return -EINVAL;
+
+		data.ce_count += info.ce_count;
+		data.ue_count += info.ue_count;
+	}
+
+	return is_ce ? data.ce_count : data.ue_count;
+}
+/* query/inject/cure end */
+
+
+/* sysfs begin */
+
+static ssize_t amdgpu_ras_sysfs_features_read(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	struct amdgpu_ras *con =
+		container_of(attr, struct amdgpu_ras, features_attr);
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = ddev->dev_private;
+	struct ras_common_if head;
+	int ras_block_count = AMDGPU_RAS_BLOCK_COUNT;
+	int i;
+	ssize_t s;
+	struct ras_manager *obj;
+
+	s = scnprintf(buf, PAGE_SIZE, "feature mask: 0x%x\n", con->features);
+
+	for (i = 0; i < ras_block_count; i++) {
+		head.block = i;
+
+		if (amdgpu_ras_is_feature_enabled(adev, &head)) {
+			obj = amdgpu_ras_find_obj(adev, &head);
+			s += scnprintf(&buf[s], PAGE_SIZE - s,
+					"%s: %s\n",
+					ras_block_str(i),
+					ras_err_str(obj->head.type));
+		} else
+			s += scnprintf(&buf[s], PAGE_SIZE - s,
+					"%s: disabled\n",
+					ras_block_str(i));
+	}
+
+	return s;
+}
+
+static int amdgpu_ras_sysfs_create_feature_node(struct amdgpu_device *adev)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct attribute *attrs[] = {
+		&con->features_attr.attr,
+		NULL
+	};
+	struct attribute_group group = {
+		.name = "ras",
+		.attrs = attrs,
+	};
+
+	con->features_attr = (struct device_attribute) {
+		.attr = {
+			.name = "features",
+			.mode = S_IRUGO,
+		},
+			.show = amdgpu_ras_sysfs_features_read,
+	};
+	sysfs_attr_init(attrs[0]);
+
+	return sysfs_create_group(&adev->dev->kobj, &group);
+}
+
+static int amdgpu_ras_sysfs_remove_feature_node(struct amdgpu_device *adev)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct attribute *attrs[] = {
+		&con->features_attr.attr,
+		NULL
+	};
+	struct attribute_group group = {
+		.name = "ras",
+		.attrs = attrs,
+	};
+
+	sysfs_remove_group(&adev->dev->kobj, &group);
+
+	return 0;
+}
+
+int amdgpu_ras_sysfs_create(struct amdgpu_device *adev,
+		struct ras_fs_if *head)
+{
+	struct ras_manager *obj = amdgpu_ras_find_obj(adev, &head->head);
+
+	if (!obj || obj->attr_inuse)
+		return -EINVAL;
+
+	get_obj(obj);
+
+	memcpy(obj->fs_data.sysfs_name,
+			head->sysfs_name,
+			sizeof(obj->fs_data.sysfs_name));
+
+	obj->sysfs_attr = (struct device_attribute){
+		.attr = {
+			.name = obj->fs_data.sysfs_name,
+			.mode = S_IRUGO,
+		},
+			.show = amdgpu_ras_sysfs_read,
+	};
+	sysfs_attr_init(&obj->sysfs_attr.attr);
+
+	if (sysfs_add_file_to_group(&adev->dev->kobj,
+				&obj->sysfs_attr.attr,
+				"ras")) {
+		put_obj(obj);
+		return -EINVAL;
+	}
+
+	obj->attr_inuse = 1;
+
+	return 0;
+}
+
+int amdgpu_ras_sysfs_remove(struct amdgpu_device *adev,
+		struct ras_common_if *head)
+{
+	struct ras_manager *obj = amdgpu_ras_find_obj(adev, head);
+
+	if (!obj || !obj->attr_inuse)
+		return -EINVAL;
+
+	sysfs_remove_file_from_group(&adev->dev->kobj,
+				&obj->sysfs_attr.attr,
+				"ras");
+	obj->attr_inuse = 0;
+	put_obj(obj);
+
+	return 0;
+}
+
+static int amdgpu_ras_sysfs_remove_all(struct amdgpu_device *adev)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct ras_manager *obj, *tmp;
+
+	list_for_each_entry_safe(obj, tmp, &con->head, node) {
+		amdgpu_ras_sysfs_remove(adev, &obj->head);
+	}
+
+	amdgpu_ras_sysfs_remove_feature_node(adev);
+
+	return 0;
+}
+/* sysfs end */
+
+/* debugfs begin */
+static int amdgpu_ras_debugfs_create_ctrl_node(struct amdgpu_device *adev)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct drm_minor *minor = adev->ddev->primary;
+	struct dentry *root = minor->debugfs_root, *dir;
+	struct dentry *ent;
+
+	dir = debugfs_create_dir("ras", root);
+	if (IS_ERR(dir))
+		return -EINVAL;
+
+	con->dir = dir;
+
+	ent = debugfs_create_file("ras_ctrl",
+			S_IWUGO | S_IRUGO, con->dir,
+			adev, &amdgpu_ras_debugfs_ctrl_ops);
+	if (IS_ERR(ent)) {
+		debugfs_remove(con->dir);
+		return -EINVAL;
+	}
+
+	con->ent = ent;
+	return 0;
+}
+
+int amdgpu_ras_debugfs_create(struct amdgpu_device *adev,
+		struct ras_fs_if *head)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct ras_manager *obj = amdgpu_ras_find_obj(adev, &head->head);
+	struct dentry *ent;
+
+	if (!obj || obj->ent)
+		return -EINVAL;
+
+	get_obj(obj);
+
+	memcpy(obj->fs_data.debugfs_name,
+			head->debugfs_name,
+			sizeof(obj->fs_data.debugfs_name));
+
+	ent = debugfs_create_file(obj->fs_data.debugfs_name,
+			S_IWUGO | S_IRUGO, con->dir,
+			obj, &amdgpu_ras_debugfs_ops);
+
+	if (IS_ERR(ent))
+		return -EINVAL;
+
+	obj->ent = ent;
+
+	return 0;
+}
+
+int amdgpu_ras_debugfs_remove(struct amdgpu_device *adev,
+		struct ras_common_if *head)
+{
+	struct ras_manager *obj = amdgpu_ras_find_obj(adev, head);
+
+	if (!obj || !obj->ent)
+		return 0;
+
+	debugfs_remove(obj->ent);
+	obj->ent = NULL;
+	put_obj(obj);
+
+	return 0;
+}
+
+static int amdgpu_ras_debugfs_remove_all(struct amdgpu_device *adev)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct ras_manager *obj, *tmp;
+
+	list_for_each_entry_safe(obj, tmp, &con->head, node) {
+		amdgpu_ras_debugfs_remove(adev, &obj->head);
+	}
+
+	debugfs_remove(con->ent);
+	debugfs_remove(con->dir);
+	con->dir = NULL;
+	con->ent = NULL;
+
+	return 0;
+}
+/* debugfs end */
+
+/* ras fs */
+
+static int amdgpu_ras_fs_init(struct amdgpu_device *adev)
+{
+	amdgpu_ras_sysfs_create_feature_node(adev);
+	amdgpu_ras_debugfs_create_ctrl_node(adev);
+
+	return 0;
+}
+
+static int amdgpu_ras_fs_fini(struct amdgpu_device *adev)
+{
+	amdgpu_ras_debugfs_remove_all(adev);
+	amdgpu_ras_sysfs_remove_all(adev);
+	return 0;
+}
+/* ras fs end */
+
+/* ih begin */
+static void amdgpu_ras_interrupt_handler(struct ras_manager *obj)
+{
+	struct ras_ih_data *data = &obj->ih_data;
+	struct amdgpu_iv_entry entry;
+	int ret;
+
+	while (data->rptr != data->wptr) {
+		rmb();
+		memcpy(&entry, &data->ring[data->rptr],
+				data->element_size);
+
+		wmb();
+		data->rptr = (data->aligned_element_size +
+				data->rptr) % data->ring_size;
+
+		/* Let IP handle its data, maybe we need get the output
+		 * from the callback to udpate the error type/count, etc
+		 */
+		if (data->cb) {
+			ret = data->cb(obj->adev, &entry);
+			/* ue will trigger an interrupt, and in that case
+			 * we need do a reset to recovery the whole system.
+			 * But leave IP do that recovery, here we just dispatch
+			 * the error.
+			 */
+			if (ret == AMDGPU_RAS_UE) {
+				obj->err_data.ue_count++;
+			}
+			/* Might need get ce count by register, but not all IP
+			 * saves ce count, some IP just use one bit or two bits
+			 * to indicate ce happened.
+			 */
+		}
+	}
+}
+
+static void amdgpu_ras_interrupt_process_handler(struct work_struct *work)
+{
+	struct ras_ih_data *data =
+		container_of(work, struct ras_ih_data, ih_work);
+	struct ras_manager *obj =
+		container_of(data, struct ras_manager, ih_data);
+
+	amdgpu_ras_interrupt_handler(obj);
+}
+
+int amdgpu_ras_interrupt_dispatch(struct amdgpu_device *adev,
+		struct ras_dispatch_if *info)
+{
+	struct ras_manager *obj = amdgpu_ras_find_obj(adev, &info->head);
+	struct ras_ih_data *data = &obj->ih_data;
+
+	if (!obj)
+		return -EINVAL;
+
+	if (data->inuse == 0)
+		return 0;
+
+	/* Might be overflow... */
+	memcpy(&data->ring[data->wptr], info->entry,
+			data->element_size);
+
+	wmb();
+	data->wptr = (data->aligned_element_size +
+			data->wptr) % data->ring_size;
+
+	schedule_work(&data->ih_work);
+
+	return 0;
+}
+
+int amdgpu_ras_interrupt_remove_handler(struct amdgpu_device *adev,
+		struct ras_ih_if *info)
+{
+	struct ras_manager *obj = amdgpu_ras_find_obj(adev, &info->head);
+	struct ras_ih_data *data;
+
+	if (!obj)
+		return -EINVAL;
+
+	data = &obj->ih_data;
+	if (data->inuse == 0)
+		return 0;
+
+	cancel_work_sync(&data->ih_work);
+
+	kfree(data->ring);
+	memset(data, 0, sizeof(*data));
+	put_obj(obj);
+
+	return 0;
+}
+
+int amdgpu_ras_interrupt_add_handler(struct amdgpu_device *adev,
+		struct ras_ih_if *info)
+{
+	struct ras_manager *obj = amdgpu_ras_find_obj(adev, &info->head);
+	struct ras_ih_data *data;
+
+	if (!obj) {
+		/* in case we registe the IH before enable ras feature */
+		obj = amdgpu_ras_create_obj(adev, &info->head);
+		if (!obj)
+			return -EINVAL;
+	} else
+		get_obj(obj);
+
+	data = &obj->ih_data;
+	/* add the callback.etc */
+	*data = (struct ras_ih_data) {
+		.inuse = 0,
+		.cb = info->cb,
+		.element_size = sizeof(struct amdgpu_iv_entry),
+		.rptr = 0,
+		.wptr = 0,
+	};
+
+	INIT_WORK(&data->ih_work, amdgpu_ras_interrupt_process_handler);
+
+	data->aligned_element_size = ALIGN(data->element_size, 8);
+	/* the ring can store 64 iv entries. */
+	data->ring_size = 64 * data->aligned_element_size;
+	data->ring = kmalloc(data->ring_size, GFP_KERNEL);
+	if (!data->ring) {
+		put_obj(obj);
+		return -ENOMEM;
+	}
+
+	/* IH is ready */
+	data->inuse = 1;
+
+	return 0;
+}
+
+static int amdgpu_ras_interrupt_remove_all(struct amdgpu_device *adev)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct ras_manager *obj, *tmp;
+
+	list_for_each_entry_safe(obj, tmp, &con->head, node) {
+		struct ras_ih_if info = {
+			.head = obj->head,
+		};
+		amdgpu_ras_interrupt_remove_handler(adev, &info);
+	}
+
+	return 0;
+}
+/* ih end */
+
+/* recovery begin */
+static void amdgpu_ras_do_recovery(struct work_struct *work)
+{
+	struct amdgpu_ras *ras =
+		container_of(work, struct amdgpu_ras, recovery_work);
+
+	amdgpu_device_gpu_recover(ras->adev, 0);
+	atomic_set(&ras->in_recovery, 0);
+}
+
+static int amdgpu_ras_release_vram(struct amdgpu_device *adev,
+		struct amdgpu_bo **bo_ptr)
+{
+	/* no need to free it actually. */
+	amdgpu_bo_free_kernel(bo_ptr, NULL, NULL);
+	return 0;
+}
+
+/* reserve vram with size@offset */
+static int amdgpu_ras_reserve_vram(struct amdgpu_device *adev,
+		uint64_t offset, uint64_t size,
+		struct amdgpu_bo **bo_ptr)
+{
+	struct ttm_operation_ctx ctx = { false, false };
+	struct amdgpu_bo_param bp;
+	int r = 0;
+	int i;
+	struct amdgpu_bo *bo;
+
+	if (bo_ptr)
+		*bo_ptr = NULL;
+	memset(&bp, 0, sizeof(bp));
+	bp.size = size;
+	bp.byte_align = PAGE_SIZE;
+	bp.domain = AMDGPU_GEM_DOMAIN_VRAM;
+	bp.flags = AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS |
+		AMDGPU_GEM_CREATE_NO_CPU_ACCESS;
+	bp.type = ttm_bo_type_kernel;
+	bp.resv = NULL;
+
+	r = amdgpu_bo_create(adev, &bp, &bo);
+	if (r)
+		return -EINVAL;
+
+	r = amdgpu_bo_reserve(bo, false);
+	if (r)
+		goto error_reserve;
+
+	offset = ALIGN(offset, PAGE_SIZE);
+	for (i = 0; i < bo->placement.num_placement; ++i) {
+		bo->placements[i].fpfn = offset >> PAGE_SHIFT;
+		bo->placements[i].lpfn = (offset + size) >> PAGE_SHIFT;
+	}
+
+	ttm_bo_mem_put(&bo->tbo, &bo->tbo.mem);
+	r = ttm_bo_mem_space(&bo->tbo, &bo->placement, &bo->tbo.mem, &ctx);
+	if (r)
+		goto error_pin;
+
+	r = amdgpu_bo_pin_restricted(bo,
+			AMDGPU_GEM_DOMAIN_VRAM,
+			offset,
+			offset + size);
+	if (r)
+		goto error_pin;
+
+	if (bo_ptr)
+		*bo_ptr = bo;
+
+	amdgpu_bo_unreserve(bo);
+	return r;
+
+error_pin:
+	amdgpu_bo_unreserve(bo);
+error_reserve:
+	amdgpu_bo_unref(&bo);
+	return r;
+}
+
+/* alloc/realloc bps array */
+static int amdgpu_ras_realloc_eh_data_space(struct amdgpu_device *adev,
+		struct ras_err_handler_data *data, int pages)
+{
+	unsigned int old_space = data->count + data->space_left;
+	unsigned int new_space = old_space + pages;
+	unsigned int align_space = ALIGN(new_space, 1024);
+	void *tmp = kmalloc(align_space * sizeof(*data->bps), GFP_KERNEL);
+
+	if (!tmp)
+		return -ENOMEM;
+
+	if (data->bps) {
+		memcpy(tmp, data->bps,
+				data->count * sizeof(*data->bps));
+		kfree(data->bps);
+	}
+
+	data->bps = tmp;
+	data->space_left += align_space - old_space;
+	return 0;
+}
+
+/* it deal with vram only. */
+int amdgpu_ras_add_bad_pages(struct amdgpu_device *adev,
+		unsigned long *bps, int pages)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct ras_err_handler_data *data;
+	int i = pages;
+	int ret = 0;
+
+	if (!con || !con->eh_data || !bps || pages <= 0)
+		return 0;
+
+	mutex_lock(&con->recovery_lock);
+	data = con->eh_data;
+	if (!data)
+		goto out;
+
+	if (data->space_left <= pages)
+		if (amdgpu_ras_realloc_eh_data_space(adev, data, pages)) {
+			ret = -ENOMEM;
+			goto out;
+		}
+
+	while (i--)
+		data->bps[data->count++].bp = bps[i];
+
+	data->space_left -= pages;
+out:
+	mutex_unlock(&con->recovery_lock);
+
+	return ret;
+}
+
+/* called in gpu recovery/init */
+int amdgpu_ras_reserve_bad_pages(struct amdgpu_device *adev)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct ras_err_handler_data *data;
+	uint64_t bp;
+	struct amdgpu_bo *bo;
+	int i;
+
+	if (!con || !con->eh_data)
+		return 0;
+
+	mutex_lock(&con->recovery_lock);
+	data = con->eh_data;
+	if (!data)
+		goto out;
+	/* reserve vram at driver post stage. */
+	for (i = data->last_reserved; i < data->count; i++) {
+		bp = data->bps[i].bp;
+
+		if (amdgpu_ras_reserve_vram(adev, bp << PAGE_SHIFT,
+					PAGE_SIZE, &bo))
+			DRM_ERROR("RAS ERROR: reserve vram %llx fail\n", bp);
+
+		data->bps[i].bo = bo;
+		data->last_reserved = i + 1;
+	}
+out:
+	mutex_unlock(&con->recovery_lock);
+	return 0;
+}
+
+/* called when driver unload */
+static int amdgpu_ras_release_bad_pages(struct amdgpu_device *adev)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct ras_err_handler_data *data;
+	struct amdgpu_bo *bo;
+	int i;
+
+	if (!con || !con->eh_data)
+		return 0;
+
+	mutex_lock(&con->recovery_lock);
+	data = con->eh_data;
+	if (!data)
+		goto out;
+
+	for (i = data->last_reserved - 1; i >= 0; i--) {
+		bo = data->bps[i].bo;
+
+		amdgpu_ras_release_vram(adev, &bo);
+
+		data->bps[i].bo = bo;
+		data->last_reserved = i;
+	}
+out:
+	mutex_unlock(&con->recovery_lock);
+	return 0;
+}
+
+static int amdgpu_ras_save_bad_pages(struct amdgpu_device *adev)
+{
+	/* TODO
+	 * write the array to eeprom when SMU disabled.
+	 */
+	return 0;
+}
+
+static int amdgpu_ras_load_bad_pages(struct amdgpu_device *adev)
+{
+	/* TODO
+	 * read the array to eeprom when SMU disabled.
+	 */
+	return 0;
+}
+
+static int amdgpu_ras_recovery_init(struct amdgpu_device *adev)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct ras_err_handler_data **data = &con->eh_data;
+
+	*data = kmalloc(sizeof(**data),
+			GFP_KERNEL|__GFP_ZERO);
+	if (!*data)
+		return -ENOMEM;
+
+	mutex_init(&con->recovery_lock);
+	INIT_WORK(&con->recovery_work, amdgpu_ras_do_recovery);
+	atomic_set(&con->in_recovery, 0);
+	con->adev = adev;
+
+	amdgpu_ras_load_bad_pages(adev);
+	amdgpu_ras_reserve_bad_pages(adev);
+
+	return 0;
+}
+
+static int amdgpu_ras_recovery_fini(struct amdgpu_device *adev)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct ras_err_handler_data *data = con->eh_data;
+
+	cancel_work_sync(&con->recovery_work);
+	amdgpu_ras_save_bad_pages(adev);
+	amdgpu_ras_release_bad_pages(adev);
+
+	mutex_lock(&con->recovery_lock);
+	con->eh_data = NULL;
+	kfree(data->bps);
+	kfree(data);
+	mutex_unlock(&con->recovery_lock);
+
+	return 0;
+}
+/* recovery end */
+
+/*
+ * check hardware's ras ability which will be saved in hw_supported.
+ * if hardware does not support ras, we can skip some ras initializtion and
+ * forbid some ras operations from IP.
+ * if software itself, say boot parameter, limit the ras ability. We still
+ * need allow IP do some limited operations, like disable. In such case,
+ * we have to initialize ras as normal. but need check if operation is
+ * allowed or not in each function.
+ */
+static void amdgpu_ras_check_supported(struct amdgpu_device *adev,
+		uint32_t *hw_supported, uint32_t *supported)
+{
+	*hw_supported = 0;
+	*supported = 0;
+
+	if (amdgpu_sriov_vf(adev) ||
+			adev->asic_type != CHIP_VEGA20)
+		return;
+
+	if (adev->is_atom_fw &&
+			(amdgpu_atomfirmware_mem_ecc_supported(adev) ||
+			 amdgpu_atomfirmware_sram_ecc_supported(adev)))
+		*hw_supported = AMDGPU_RAS_BLOCK_MASK;
+
+	*supported = amdgpu_ras_enable == 0 ?
+				0 : *hw_supported & amdgpu_ras_mask;
+}
+
+int amdgpu_ras_init(struct amdgpu_device *adev)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+
+	if (con)
+		return 0;
+
+	con = kmalloc(sizeof(struct amdgpu_ras) +
+			sizeof(struct ras_manager) * AMDGPU_RAS_BLOCK_COUNT,
+			GFP_KERNEL|__GFP_ZERO);
+	if (!con)
+		return -ENOMEM;
+
+	con->objs = (struct ras_manager *)(con + 1);
+
+	amdgpu_ras_set_context(adev, con);
+
+	amdgpu_ras_check_supported(adev, &con->hw_supported,
+			&con->supported);
+	con->features = 0;
+	INIT_LIST_HEAD(&con->head);
+	/* Might need get this flag from vbios. */
+	con->flags = RAS_DEFAULT_FLAGS;
+
+	if (amdgpu_ras_recovery_init(adev))
+		goto recovery_out;
+
+	amdgpu_ras_mask &= AMDGPU_RAS_BLOCK_MASK;
+
+	if (amdgpu_ras_fs_init(adev))
+		goto fs_out;
+
+	amdgpu_ras_self_test(adev);
+
+	DRM_INFO("RAS INFO: ras initialized successfully, "
+			"hardware ability[%x] ras_mask[%x]\n",
+			con->hw_supported, con->supported);
+	return 0;
+fs_out:
+	amdgpu_ras_recovery_fini(adev);
+recovery_out:
+	amdgpu_ras_set_context(adev, NULL);
+	kfree(con);
+
+	return -EINVAL;
+}
+
+/* do some init work after IP late init as dependence */
+void amdgpu_ras_post_init(struct amdgpu_device *adev)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct ras_manager *obj, *tmp;
+
+	if (!con)
+		return;
+
+	if (con->flags & AMDGPU_RAS_FLAG_INIT_BY_VBIOS) {
+		/* Set up all other IPs which are not implemented. There is a
+		 * tricky thing that IP's actual ras error type should be
+		 * MULTI_UNCORRECTABLE, but as driver does not handle it, so
+		 * ERROR_NONE make sense anyway.
+		 */
+		amdgpu_ras_enable_all_features(adev, 1);
+
+		/* We enable ras on all hw_supported block, but as boot
+		 * parameter might disable some of them and one or more IP has
+		 * not implemented yet. So we disable them on behalf.
+		 */
+		list_for_each_entry_safe(obj, tmp, &con->head, node) {
+			if (!amdgpu_ras_is_supported(adev, obj->head.block)) {
+				amdgpu_ras_feature_enable(adev, &obj->head, 0);
+				/* there should be no any reference. */
+				WARN_ON(alive_obj(obj));
+			}
+		}
+	}
+}
+
+/* do some fini work before IP fini as dependence */
+int amdgpu_ras_pre_fini(struct amdgpu_device *adev)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+
+	if (!con)
+		return 0;
+
+	/* Need disable ras on all IPs here before ip [hw/sw]fini */
+	amdgpu_ras_disable_all_features(adev, 0);
+	amdgpu_ras_recovery_fini(adev);
+	return 0;
+}
+
+int amdgpu_ras_fini(struct amdgpu_device *adev)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+
+	if (!con)
+		return 0;
+
+	amdgpu_ras_fs_fini(adev);
+	amdgpu_ras_interrupt_remove_all(adev);
+
+	WARN(con->features, "Feature mask is not cleared");
+
+	if (con->features)
+		amdgpu_ras_disable_all_features(adev, 1);
+
+	amdgpu_ras_set_context(adev, NULL);
+	kfree(con);
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
new file mode 100644
index 000000000000..eaef5edefc34
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
@@ -0,0 +1,294 @@
+/*
+ * Copyright 2018 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ *
+ */
+#ifndef _AMDGPU_RAS_H
+#define _AMDGPU_RAS_H
+
+#include <linux/debugfs.h>
+#include <linux/list.h>
+#include "amdgpu.h"
+#include "amdgpu_psp.h"
+#include "ta_ras_if.h"
+
+enum amdgpu_ras_block {
+	AMDGPU_RAS_BLOCK__UMC = 0,
+	AMDGPU_RAS_BLOCK__SDMA,
+	AMDGPU_RAS_BLOCK__GFX,
+	AMDGPU_RAS_BLOCK__MMHUB,
+	AMDGPU_RAS_BLOCK__ATHUB,
+	AMDGPU_RAS_BLOCK__PCIE_BIF,
+	AMDGPU_RAS_BLOCK__HDP,
+	AMDGPU_RAS_BLOCK__XGMI_WAFL,
+	AMDGPU_RAS_BLOCK__DF,
+	AMDGPU_RAS_BLOCK__SMN,
+	AMDGPU_RAS_BLOCK__SEM,
+	AMDGPU_RAS_BLOCK__MP0,
+	AMDGPU_RAS_BLOCK__MP1,
+	AMDGPU_RAS_BLOCK__FUSE,
+
+	AMDGPU_RAS_BLOCK__LAST
+};
+
+#define AMDGPU_RAS_BLOCK_COUNT	AMDGPU_RAS_BLOCK__LAST
+#define AMDGPU_RAS_BLOCK_MASK	((1ULL << AMDGPU_RAS_BLOCK_COUNT) - 1)
+
+enum amdgpu_ras_error_type {
+	AMDGPU_RAS_ERROR__NONE							= 0,
+	AMDGPU_RAS_ERROR__PARITY						= 1,
+	AMDGPU_RAS_ERROR__SINGLE_CORRECTABLE					= 2,
+	AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE					= 4,
+	AMDGPU_RAS_ERROR__POISON						= 8,
+};
+
+enum amdgpu_ras_ret {
+	AMDGPU_RAS_SUCCESS = 0,
+	AMDGPU_RAS_FAIL,
+	AMDGPU_RAS_UE,
+	AMDGPU_RAS_CE,
+	AMDGPU_RAS_PT,
+};
+
+struct ras_common_if {
+	enum amdgpu_ras_block block;
+	enum amdgpu_ras_error_type type;
+	uint32_t sub_block_index;
+	/* block name */
+	char name[32];
+};
+
+typedef int (*ras_ih_cb)(struct amdgpu_device *adev,
+		struct amdgpu_iv_entry *entry);
+
+struct amdgpu_ras {
+	/* ras infrastructure */
+	/* for ras itself. */
+	uint32_t hw_supported;
+	/* for IP to check its ras ability. */
+	uint32_t supported;
+	uint32_t features;
+	struct list_head head;
+	/* debugfs */
+	struct dentry *dir;
+	/* debugfs ctrl */
+	struct dentry *ent;
+	/* sysfs */
+	struct device_attribute features_attr;
+	/* block array */
+	struct ras_manager *objs;
+
+	/* gpu recovery */
+	struct work_struct recovery_work;
+	atomic_t in_recovery;
+	struct amdgpu_device *adev;
+	/* error handler data */
+	struct ras_err_handler_data *eh_data;
+	struct mutex recovery_lock;
+
+	uint32_t flags;
+};
+
+/* interfaces for IP */
+
+struct ras_fs_if {
+	struct ras_common_if head;
+	char sysfs_name[32];
+	char debugfs_name[32];
+};
+
+struct ras_query_if {
+	struct ras_common_if head;
+	unsigned long ue_count;
+	unsigned long ce_count;
+};
+
+struct ras_inject_if {
+	struct ras_common_if head;
+	uint64_t address;
+	uint64_t value;
+};
+
+struct ras_cure_if {
+	struct ras_common_if head;
+	uint64_t address;
+};
+
+struct ras_ih_if {
+	struct ras_common_if head;
+	ras_ih_cb cb;
+};
+
+struct ras_dispatch_if {
+	struct ras_common_if head;
+	struct amdgpu_iv_entry *entry;
+};
+
+struct ras_debug_if {
+	union {
+		struct ras_common_if head;
+		struct ras_inject_if inject;
+	};
+	int op;
+};
+/* work flow
+ * vbios
+ * 1: ras feature enable (enabled by default)
+ * psp
+ * 2: ras framework init (in ip_init)
+ * IP
+ * 3: IH add
+ * 4: debugfs/sysfs create
+ * 5: query/inject
+ * 6: debugfs/sysfs remove
+ * 7: IH remove
+ * 8: feature disable
+ */
+
+#define amdgpu_ras_get_context(adev)		((adev)->psp.ras.ras)
+#define amdgpu_ras_set_context(adev, ras_con)	((adev)->psp.ras.ras = (ras_con))
+
+/* check if ras is supported on block, say, sdma, gfx */
+static inline int amdgpu_ras_is_supported(struct amdgpu_device *adev,
+		unsigned int block)
+{
+	struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
+
+	return ras && (ras->supported & (1 << block));
+}
+
+int amdgpu_ras_query_error_count(struct amdgpu_device *adev,
+		bool is_ce);
+
+/* error handling functions */
+int amdgpu_ras_add_bad_pages(struct amdgpu_device *adev,
+		unsigned long *bps, int pages);
+
+int amdgpu_ras_reserve_bad_pages(struct amdgpu_device *adev);
+
+static inline int amdgpu_ras_reset_gpu(struct amdgpu_device *adev,
+		bool is_baco)
+{
+	/* remove me when gpu reset works on vega20 A1. */
+#if 0
+	struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
+
+	if (atomic_cmpxchg(&ras->in_recovery, 0, 1) == 0)
+		schedule_work(&ras->recovery_work);
+#endif
+	return 0;
+}
+
+static inline enum ta_ras_block
+amdgpu_ras_block_to_ta(enum amdgpu_ras_block block) {
+	switch (block) {
+	case AMDGPU_RAS_BLOCK__UMC:
+		return TA_RAS_BLOCK__UMC;
+	case AMDGPU_RAS_BLOCK__SDMA:
+		return TA_RAS_BLOCK__SDMA;
+	case AMDGPU_RAS_BLOCK__GFX:
+		return TA_RAS_BLOCK__GFX;
+	case AMDGPU_RAS_BLOCK__MMHUB:
+		return TA_RAS_BLOCK__MMHUB;
+	case AMDGPU_RAS_BLOCK__ATHUB:
+		return TA_RAS_BLOCK__ATHUB;
+	case AMDGPU_RAS_BLOCK__PCIE_BIF:
+		return TA_RAS_BLOCK__PCIE_BIF;
+	case AMDGPU_RAS_BLOCK__HDP:
+		return TA_RAS_BLOCK__HDP;
+	case AMDGPU_RAS_BLOCK__XGMI_WAFL:
+		return TA_RAS_BLOCK__XGMI_WAFL;
+	case AMDGPU_RAS_BLOCK__DF:
+		return TA_RAS_BLOCK__DF;
+	case AMDGPU_RAS_BLOCK__SMN:
+		return TA_RAS_BLOCK__SMN;
+	case AMDGPU_RAS_BLOCK__SEM:
+		return TA_RAS_BLOCK__SEM;
+	case AMDGPU_RAS_BLOCK__MP0:
+		return TA_RAS_BLOCK__MP0;
+	case AMDGPU_RAS_BLOCK__MP1:
+		return TA_RAS_BLOCK__MP1;
+	case AMDGPU_RAS_BLOCK__FUSE:
+		return TA_RAS_BLOCK__FUSE;
+	default:
+		WARN_ONCE(1, "RAS ERROR: unexpected block id %d\n", block);
+		return TA_RAS_BLOCK__UMC;
+	}
+}
+
+static inline enum ta_ras_error_type
+amdgpu_ras_error_to_ta(enum amdgpu_ras_error_type error) {
+	switch (error) {
+	case AMDGPU_RAS_ERROR__NONE:
+		return TA_RAS_ERROR__NONE;
+	case AMDGPU_RAS_ERROR__PARITY:
+		return TA_RAS_ERROR__PARITY;
+	case AMDGPU_RAS_ERROR__SINGLE_CORRECTABLE:
+		return TA_RAS_ERROR__SINGLE_CORRECTABLE;
+	case AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE:
+		return TA_RAS_ERROR__MULTI_UNCORRECTABLE;
+	case AMDGPU_RAS_ERROR__POISON:
+		return TA_RAS_ERROR__POISON;
+	default:
+		WARN_ONCE(1, "RAS ERROR: unexpected error type %d\n", error);
+		return TA_RAS_ERROR__NONE;
+	}
+}
+
+/* called in ip_init and ip_fini */
+int amdgpu_ras_init(struct amdgpu_device *adev);
+void amdgpu_ras_post_init(struct amdgpu_device *adev);
+int amdgpu_ras_fini(struct amdgpu_device *adev);
+int amdgpu_ras_pre_fini(struct amdgpu_device *adev);
+
+int amdgpu_ras_feature_enable(struct amdgpu_device *adev,
+		struct ras_common_if *head, bool enable);
+
+int amdgpu_ras_feature_enable_on_boot(struct amdgpu_device *adev,
+		struct ras_common_if *head, bool enable);
+
+int amdgpu_ras_sysfs_create(struct amdgpu_device *adev,
+		struct ras_fs_if *head);
+
+int amdgpu_ras_sysfs_remove(struct amdgpu_device *adev,
+		struct ras_common_if *head);
+
+int amdgpu_ras_debugfs_create(struct amdgpu_device *adev,
+		struct ras_fs_if *head);
+
+int amdgpu_ras_debugfs_remove(struct amdgpu_device *adev,
+		struct ras_common_if *head);
+
+int amdgpu_ras_error_query(struct amdgpu_device *adev,
+		struct ras_query_if *info);
+
+int amdgpu_ras_error_inject(struct amdgpu_device *adev,
+		struct ras_inject_if *info);
+
+int amdgpu_ras_interrupt_add_handler(struct amdgpu_device *adev,
+		struct ras_ih_if *info);
+
+int amdgpu_ras_interrupt_remove_handler(struct amdgpu_device *adev,
+		struct ras_ih_if *info);
+
+int amdgpu_ras_interrupt_dispatch(struct amdgpu_device *adev,
+		struct ras_dispatch_if *info);
+#endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index 335a0edf114b..8f5026c123ef 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -248,6 +248,8 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring,
 	 */
 	if (ring->funcs->type == AMDGPU_RING_TYPE_KIQ)
 		sched_hw_submission = max(sched_hw_submission, 256);
+	else if (ring == &adev->sdma.instance[0].page)
+		sched_hw_submission = 256;
 
 	if (ring->adev == NULL) {
 		if (adev->num_rings >= AMDGPU_MAX_RINGS)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
index 16b1a6ae5ba6..1ba9ba3b54f7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
@@ -28,9 +28,8 @@
 #define AMDGPU_MAX_SDMA_INSTANCES		2
 
 enum amdgpu_sdma_irq {
-	AMDGPU_SDMA_IRQ_TRAP0 = 0,
-	AMDGPU_SDMA_IRQ_TRAP1,
-
+	AMDGPU_SDMA_IRQ_INSTANCE0  = 0,
+	AMDGPU_SDMA_IRQ_INSTANCE1,
 	AMDGPU_SDMA_IRQ_LAST
 };
 
@@ -49,9 +48,11 @@ struct amdgpu_sdma {
 	struct amdgpu_sdma_instance instance[AMDGPU_MAX_SDMA_INSTANCES];
 	struct amdgpu_irq_src	trap_irq;
 	struct amdgpu_irq_src	illegal_inst_irq;
+	struct amdgpu_irq_src	ecc_irq;
 	int			num_instances;
 	uint32_t                    srbm_soft_reset;
 	bool			has_page_queue;
+	struct ras_common_if	*ras_if;
 };
 
 /*
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 73e71e61dc99..0c52d1f9fe0f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -50,8 +50,6 @@
 #include "amdgpu_sdma.h"
 #include "bif/bif_4_1_d.h"
 
-#define DRM_FILE_PAGE_OFFSET (0x100000000ULL >> PAGE_SHIFT)
-
 static int amdgpu_map_buffer(struct ttm_buffer_object *bo,
 			     struct ttm_mem_reg *mem, unsigned num_pages,
 			     uint64_t offset, unsigned window,
@@ -1424,6 +1422,13 @@ static bool amdgpu_ttm_bo_eviction_valuable(struct ttm_buffer_object *bo,
 	struct dma_fence *f;
 	int i;
 
+	/* Don't evict VM page tables while they are busy, otherwise we can't
+	 * cleanly handle page faults.
+	 */
+	if (bo->type == ttm_bo_type_kernel &&
+	    !reservation_object_test_signaled_rcu(bo->resv, true))
+		return false;
+
 	/* If bo is a KFD BO, check if the bo belongs to the current process.
 	 * If true, then return false as any KFD process needs all its BOs to
 	 * be resident to run successfully
@@ -1671,7 +1676,6 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 	r = ttm_bo_device_init(&adev->mman.bdev,
 			       &amdgpu_bo_driver,
 			       adev->ddev->anon_inode->i_mapping,
-			       DRM_FILE_PAGE_OFFSET,
 			       adev->need_dma32);
 	if (r) {
 		DRM_ERROR("failed initializing buffer object driver(%d).\n", r);
@@ -1877,14 +1881,9 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)
 
 int amdgpu_mmap(struct file *filp, struct vm_area_struct *vma)
 {
-	struct drm_file *file_priv;
-	struct amdgpu_device *adev;
-
-	if (unlikely(vma->vm_pgoff < DRM_FILE_PAGE_OFFSET))
-		return -EINVAL;
+	struct drm_file *file_priv = filp->private_data;
+	struct amdgpu_device *adev = file_priv->minor->dev->dev_private;
 
-	file_priv = filp->private_data;
-	adev = file_priv->minor->dev->dev_private;
 	if (adev == NULL)
 		return -EINVAL;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
index 462a04e0f5e6..7d484fad3909 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
@@ -36,6 +36,7 @@ void amdgpu_virt_init_setting(struct amdgpu_device *adev)
 	/* enable virtual display */
 	adev->mode_info.num_crtc = 1;
 	adev->enable_virtual_display = true;
+	adev->ddev->driver->driver_features &= ~DRIVER_ATOMIC;
 	adev->cg_flags = 0;
 	adev->pg_flags = 0;
 }
@@ -375,4 +376,53 @@ void amdgpu_virt_init_data_exchange(struct amdgpu_device *adev)
 	}
 }
 
+static uint32_t parse_clk(char *buf, bool min)
+{
+        char *ptr = buf;
+        uint32_t clk = 0;
+
+        do {
+                ptr = strchr(ptr, ':');
+                if (!ptr)
+                        break;
+                ptr+=2;
+                clk = simple_strtoul(ptr, NULL, 10);
+        } while (!min);
+
+        return clk * 100;
+}
+
+uint32_t amdgpu_virt_get_sclk(struct amdgpu_device *adev, bool lowest)
+{
+	char *buf = NULL;
+	uint32_t clk = 0;
+
+	buf = kzalloc(PAGE_SIZE, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	adev->virt.ops->get_pp_clk(adev, PP_SCLK, buf);
+	clk = parse_clk(buf, lowest);
+
+	kfree(buf);
+
+	return clk;
+}
+
+uint32_t amdgpu_virt_get_mclk(struct amdgpu_device *adev, bool lowest)
+{
+	char *buf = NULL;
+	uint32_t clk = 0;
+
+	buf = kzalloc(PAGE_SIZE, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	adev->virt.ops->get_pp_clk(adev, PP_MCLK, buf);
+	clk = parse_clk(buf, lowest);
+
+	kfree(buf);
+
+	return clk;
+}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
index 722deefc0a7e..584947b7ccf3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
@@ -57,6 +57,8 @@ struct amdgpu_virt_ops {
 	int (*reset_gpu)(struct amdgpu_device *adev);
 	int (*wait_reset)(struct amdgpu_device *adev);
 	void (*trans_msg)(struct amdgpu_device *adev, u32 req, u32 data1, u32 data2, u32 data3);
+	int (*get_pp_clk)(struct amdgpu_device *adev, u32 type, char *buf);
+	int (*force_dpm_level)(struct amdgpu_device *adev, u32 level);
 };
 
 /*
@@ -83,6 +85,8 @@ enum AMDGIM_FEATURE_FLAG {
 	AMDGIM_FEATURE_GIM_LOAD_UCODES   = 0x2,
 	/* VRAM LOST by GIM */
 	AMDGIM_FEATURE_GIM_FLR_VRAMLOST = 0x4,
+	/* HW PERF SIM in GIM */
+	AMDGIM_FEATURE_HW_PERF_SIMULATION = (1 << 3),
 };
 
 struct amd_sriov_msg_pf2vf_info_header {
@@ -252,6 +256,8 @@ struct amdgpu_virt {
 	struct amdgpu_vf_error_buffer   vf_errors;
 	struct amdgpu_virt_fw_reserve	fw_reserve;
 	uint32_t gim_feature;
+	/* protect DPM events to GIM */
+	struct mutex                    dpm_mutex;
 };
 
 #define amdgpu_sriov_enabled(adev) \
@@ -278,6 +284,9 @@ static inline bool is_virtual_machine(void)
 #endif
 }
 
+#define amdgim_is_hwperf(adev) \
+	((adev)->virt.gim_feature & AMDGIM_FEATURE_HW_PERF_SIMULATION)
+
 bool amdgpu_virt_mmio_blocked(struct amdgpu_device *adev);
 void amdgpu_virt_init_setting(struct amdgpu_device *adev);
 uint32_t amdgpu_virt_kiq_rreg(struct amdgpu_device *adev, uint32_t reg);
@@ -295,5 +304,7 @@ int amdgpu_virt_fw_reserve_get_checksum(void *obj, unsigned long obj_size,
 					unsigned int key,
 					unsigned int chksum);
 void amdgpu_virt_init_data_exchange(struct amdgpu_device *adev);
+uint32_t amdgpu_virt_get_sclk(struct amdgpu_device *adev, bool lowest);
+uint32_t amdgpu_virt_get_mclk(struct amdgpu_device *adev, bool lowest);
 
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 16fcb56c232b..a07c85815b7a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -34,6 +34,7 @@
 #include "amdgpu_trace.h"
 #include "amdgpu_amdkfd.h"
 #include "amdgpu_gmc.h"
+#include "amdgpu_xgmi.h"
 
 /**
  * DOC: GPUVM
@@ -66,50 +67,6 @@ INTERVAL_TREE_DEFINE(struct amdgpu_bo_va_mapping, rb, uint64_t, __subtree_last,
 #undef LAST
 
 /**
- * struct amdgpu_pte_update_params - Local structure
- *
- * Encapsulate some VM table update parameters to reduce
- * the number of function parameters
- *
- */
-struct amdgpu_pte_update_params {
-
-	/**
-	 * @adev: amdgpu device we do this update for
-	 */
-	struct amdgpu_device *adev;
-
-	/**
-	 * @vm: optional amdgpu_vm we do this update for
-	 */
-	struct amdgpu_vm *vm;
-
-	/**
-	 * @src: address where to copy page table entries from
-	 */
-	uint64_t src;
-
-	/**
-	 * @ib: indirect buffer to fill with commands
-	 */
-	struct amdgpu_ib *ib;
-
-	/**
-	 * @func: Function which actually does the update
-	 */
-	void (*func)(struct amdgpu_pte_update_params *params,
-		     struct amdgpu_bo *bo, uint64_t pe,
-		     uint64_t addr, unsigned count, uint32_t incr,
-		     uint64_t flags);
-	/**
-	 * @pages_addr:
-	 *
-	 * DMA addresses to use for mapping, used during VM update by CPU
-	 */
-	dma_addr_t *pages_addr;
-};
-
-/**
  * struct amdgpu_prt_cb - Helper to disable partial resident texture feature from a fence callback
  */
 struct amdgpu_prt_cb {
@@ -183,6 +140,22 @@ static unsigned amdgpu_vm_num_entries(struct amdgpu_device *adev,
 }
 
 /**
+ * amdgpu_vm_num_ats_entries - return the number of ATS entries in the root PD
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Returns:
+ * The number of entries in the root page directory which needs the ATS setting.
+ */
+static unsigned amdgpu_vm_num_ats_entries(struct amdgpu_device *adev)
+{
+	unsigned shift;
+
+	shift = amdgpu_vm_level_shift(adev, adev->vm_manager.root_level);
+	return AMDGPU_GMC_HOLE_START >> (shift + AMDGPU_GPU_PAGE_SHIFT);
+}
+
+/**
  * amdgpu_vm_entries_mask - the mask to get the entry number of a PD/PT
  *
  * @adev: amdgpu_device pointer
@@ -333,7 +306,7 @@ static void amdgpu_vm_bo_base_init(struct amdgpu_vm_bo_base *base,
 		return;
 
 	vm->bulk_moveable = false;
-	if (bo->tbo.type == ttm_bo_type_kernel)
+	if (bo->tbo.type == ttm_bo_type_kernel && bo->parent)
 		amdgpu_vm_bo_relocated(base);
 	else
 		amdgpu_vm_bo_idle(base);
@@ -505,61 +478,39 @@ static void amdgpu_vm_pt_next(struct amdgpu_device *adev,
 }
 
 /**
- * amdgpu_vm_pt_first_leaf - get first leaf PD/PT
+ * amdgpu_vm_pt_first_dfs - start a deep first search
  *
- * @adev: amdgpu_device pointer
+ * @adev: amdgpu_device structure
  * @vm: amdgpu_vm structure
- * @start: start addr of the walk
  * @cursor: state to initialize
  *
- * Start a walk and go directly to the leaf node.
- */
-static void amdgpu_vm_pt_first_leaf(struct amdgpu_device *adev,
-				    struct amdgpu_vm *vm, uint64_t start,
-				    struct amdgpu_vm_pt_cursor *cursor)
-{
-	amdgpu_vm_pt_start(adev, vm, start, cursor);
-	while (amdgpu_vm_pt_descendant(adev, cursor));
-}
-
-/**
- * amdgpu_vm_pt_next_leaf - get next leaf PD/PT
- *
- * @adev: amdgpu_device pointer
- * @cursor: current state
- *
- * Walk the PD/PT tree to the next leaf node.
+ * Starts a deep first traversal of the PD/PT tree.
  */
-static void amdgpu_vm_pt_next_leaf(struct amdgpu_device *adev,
+static void amdgpu_vm_pt_first_dfs(struct amdgpu_device *adev,
+				   struct amdgpu_vm *vm,
+				   struct amdgpu_vm_pt_cursor *start,
 				   struct amdgpu_vm_pt_cursor *cursor)
 {
-	amdgpu_vm_pt_next(adev, cursor);
-	if (cursor->pfn != ~0ll)
-		while (amdgpu_vm_pt_descendant(adev, cursor));
+	if (start)
+		*cursor = *start;
+	else
+		amdgpu_vm_pt_start(adev, vm, 0, cursor);
+	while (amdgpu_vm_pt_descendant(adev, cursor));
 }
 
 /**
- * for_each_amdgpu_vm_pt_leaf - walk over all leaf PDs/PTs in the hierarchy
- */
-#define for_each_amdgpu_vm_pt_leaf(adev, vm, start, end, cursor)		\
-	for (amdgpu_vm_pt_first_leaf((adev), (vm), (start), &(cursor));		\
-	     (cursor).pfn <= end; amdgpu_vm_pt_next_leaf((adev), &(cursor)))
-
-/**
- * amdgpu_vm_pt_first_dfs - start a deep first search
+ * amdgpu_vm_pt_continue_dfs - check if the deep first search should continue
  *
- * @adev: amdgpu_device structure
- * @vm: amdgpu_vm structure
- * @cursor: state to initialize
+ * @start: starting point for the search
+ * @entry: current entry
  *
- * Starts a deep first traversal of the PD/PT tree.
+ * Returns:
+ * True when the search should continue, false otherwise.
  */
-static void amdgpu_vm_pt_first_dfs(struct amdgpu_device *adev,
-				   struct amdgpu_vm *vm,
-				   struct amdgpu_vm_pt_cursor *cursor)
+static bool amdgpu_vm_pt_continue_dfs(struct amdgpu_vm_pt_cursor *start,
+				      struct amdgpu_vm_pt *entry)
 {
-	amdgpu_vm_pt_start(adev, vm, 0, cursor);
-	while (amdgpu_vm_pt_descendant(adev, cursor));
+	return entry && (!start || entry != start->entry);
 }
 
 /**
@@ -587,11 +538,11 @@ static void amdgpu_vm_pt_next_dfs(struct amdgpu_device *adev,
 /**
  * for_each_amdgpu_vm_pt_dfs_safe - safe deep first search of all PDs/PTs
  */
-#define for_each_amdgpu_vm_pt_dfs_safe(adev, vm, cursor, entry)			\
-	for (amdgpu_vm_pt_first_dfs((adev), (vm), &(cursor)),			\
+#define for_each_amdgpu_vm_pt_dfs_safe(adev, vm, start, cursor, entry)		\
+	for (amdgpu_vm_pt_first_dfs((adev), (vm), (start), &(cursor)),		\
 	     (entry) = (cursor).entry, amdgpu_vm_pt_next_dfs((adev), &(cursor));\
-	     (entry); (entry) = (cursor).entry,					\
-	     amdgpu_vm_pt_next_dfs((adev), &(cursor)))
+	     amdgpu_vm_pt_continue_dfs((start), (entry));			\
+	     (entry) = (cursor).entry, amdgpu_vm_pt_next_dfs((adev), &(cursor)))
 
 /**
  * amdgpu_vm_get_pd_bo - add the VM PD to a validation list
@@ -712,18 +663,11 @@ int amdgpu_vm_validate_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 		if (bo->tbo.type != ttm_bo_type_kernel) {
 			amdgpu_vm_bo_moved(bo_base);
 		} else {
-			if (vm->use_cpu_for_update)
-				r = amdgpu_bo_kmap(bo, NULL);
+			vm->update_funcs->map_table(bo);
+			if (bo->parent)
+				amdgpu_vm_bo_relocated(bo_base);
 			else
-				r = amdgpu_ttm_alloc_gart(&bo->tbo);
-			if (r)
-				break;
-			if (bo->shadow) {
-				r = amdgpu_ttm_alloc_gart(&bo->shadow->tbo);
-				if (r)
-					break;
-			}
-			amdgpu_vm_bo_relocated(bo_base);
+				amdgpu_vm_bo_idle(bo_base);
 		}
 	}
 
@@ -751,8 +695,6 @@ bool amdgpu_vm_ready(struct amdgpu_vm *vm)
  * @adev: amdgpu_device pointer
  * @vm: VM to clear BO from
  * @bo: BO to clear
- * @level: level this BO is at
- * @pte_support_ats: indicate ATS support from PTE
  *
  * Root PD needs to be reserved when calling this.
  *
@@ -760,99 +702,112 @@ bool amdgpu_vm_ready(struct amdgpu_vm *vm)
  * 0 on success, errno otherwise.
  */
 static int amdgpu_vm_clear_bo(struct amdgpu_device *adev,
-			      struct amdgpu_vm *vm, struct amdgpu_bo *bo,
-			      unsigned level, bool pte_support_ats)
+			      struct amdgpu_vm *vm,
+			      struct amdgpu_bo *bo)
 {
 	struct ttm_operation_ctx ctx = { true, false };
-	struct dma_fence *fence = NULL;
+	unsigned level = adev->vm_manager.root_level;
+	struct amdgpu_vm_update_params params;
+	struct amdgpu_bo *ancestor = bo;
 	unsigned entries, ats_entries;
-	struct amdgpu_ring *ring;
-	struct amdgpu_job *job;
 	uint64_t addr;
 	int r;
 
+	/* Figure out our place in the hierarchy */
+	if (ancestor->parent) {
+		++level;
+		while (ancestor->parent->parent) {
+			++level;
+			ancestor = ancestor->parent;
+		}
+	}
+
 	entries = amdgpu_bo_size(bo) / 8;
+	if (!vm->pte_support_ats) {
+		ats_entries = 0;
+
+	} else if (!bo->parent) {
+		ats_entries = amdgpu_vm_num_ats_entries(adev);
+		ats_entries = min(ats_entries, entries);
+		entries -= ats_entries;
+
+	} else {
+		struct amdgpu_vm_pt *pt;
 
-	if (pte_support_ats) {
-		if (level == adev->vm_manager.root_level) {
-			ats_entries = amdgpu_vm_level_shift(adev, level);
-			ats_entries += AMDGPU_GPU_PAGE_SHIFT;
-			ats_entries = AMDGPU_GMC_HOLE_START >> ats_entries;
-			ats_entries = min(ats_entries, entries);
-			entries -= ats_entries;
+		pt = container_of(ancestor->vm_bo, struct amdgpu_vm_pt, base);
+		ats_entries = amdgpu_vm_num_ats_entries(adev);
+		if ((pt - vm->root.entries) >= ats_entries) {
+			ats_entries = 0;
 		} else {
 			ats_entries = entries;
 			entries = 0;
 		}
-	} else {
-		ats_entries = 0;
 	}
 
-	ring = container_of(vm->entity.rq->sched, struct amdgpu_ring, sched);
-
 	r = ttm_bo_validate(&bo->tbo, &bo->placement, &ctx);
 	if (r)
-		goto error;
+		return r;
+
+	if (bo->shadow) {
+		r = ttm_bo_validate(&bo->shadow->tbo, &bo->shadow->placement,
+				    &ctx);
+		if (r)
+			return r;
+	}
 
-	r = amdgpu_ttm_alloc_gart(&bo->tbo);
+	r = vm->update_funcs->map_table(bo);
 	if (r)
 		return r;
 
-	r = amdgpu_job_alloc_with_ib(adev, 64, &job);
+	memset(&params, 0, sizeof(params));
+	params.adev = adev;
+	params.vm = vm;
+
+	r = vm->update_funcs->prepare(&params, AMDGPU_FENCE_OWNER_KFD, NULL);
 	if (r)
-		goto error;
+		return r;
 
-	addr = amdgpu_bo_gpu_offset(bo);
+	addr = 0;
 	if (ats_entries) {
-		uint64_t ats_value;
+		uint64_t value = 0, flags;
 
-		ats_value = AMDGPU_PTE_DEFAULT_ATC;
-		if (level != AMDGPU_VM_PTB)
-			ats_value |= AMDGPU_PDE_PTE;
+		flags = AMDGPU_PTE_DEFAULT_ATC;
+		if (level != AMDGPU_VM_PTB) {
+			/* Handle leaf PDEs as PTEs */
+			flags |= AMDGPU_PDE_PTE;
+			amdgpu_gmc_get_vm_pde(adev, level, &value, &flags);
+		}
+
+		r = vm->update_funcs->update(&params, bo, addr, 0, ats_entries,
+					     value, flags);
+		if (r)
+			return r;
 
-		amdgpu_vm_set_pte_pde(adev, &job->ibs[0], addr, 0,
-				      ats_entries, 0, ats_value);
 		addr += ats_entries * 8;
 	}
 
 	if (entries) {
-		uint64_t value = 0;
-
-		/* Workaround for fault priority problem on GMC9 */
-		if (level == AMDGPU_VM_PTB && adev->asic_type >= CHIP_VEGA10)
-			value = AMDGPU_PTE_EXECUTABLE;
+		uint64_t value = 0, flags = 0;
+
+		if (adev->asic_type >= CHIP_VEGA10) {
+			if (level != AMDGPU_VM_PTB) {
+				/* Handle leaf PDEs as PTEs */
+				flags |= AMDGPU_PDE_PTE;
+				amdgpu_gmc_get_vm_pde(adev, level,
+						      &value, &flags);
+			} else {
+				/* Workaround for fault priority problem on GMC9 */
+				flags = AMDGPU_PTE_EXECUTABLE;
+			}
+		}
 
-		amdgpu_vm_set_pte_pde(adev, &job->ibs[0], addr, 0,
-				      entries, 0, value);
+		r = vm->update_funcs->update(&params, bo, addr, 0, entries,
+					     value, flags);
+		if (r)
+			return r;
 	}
 
-	amdgpu_ring_pad_ib(ring, &job->ibs[0]);
-
-	WARN_ON(job->ibs[0].length_dw > 64);
-	r = amdgpu_sync_resv(adev, &job->sync, bo->tbo.resv,
-			     AMDGPU_FENCE_OWNER_KFD, false);
-	if (r)
-		goto error_free;
-
-	r = amdgpu_job_submit(job, &vm->entity, AMDGPU_FENCE_OWNER_UNDEFINED,
-			      &fence);
-	if (r)
-		goto error_free;
-
-	amdgpu_bo_fence(bo, fence, true);
-	dma_fence_put(fence);
-
-	if (bo->shadow)
-		return amdgpu_vm_clear_bo(adev, vm, bo->shadow,
-					  level, pte_support_ats);
-
-	return 0;
-
-error_free:
-	amdgpu_job_free(job);
-
-error:
-	return r;
+	return vm->update_funcs->commit(&params, NULL);
 }
 
 /**
@@ -883,89 +838,56 @@ static void amdgpu_vm_bo_param(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 }
 
 /**
- * amdgpu_vm_alloc_pts - Allocate page tables.
+ * amdgpu_vm_alloc_pts - Allocate a specific page table
  *
  * @adev: amdgpu_device pointer
  * @vm: VM to allocate page tables for
- * @saddr: Start address which needs to be allocated
- * @size: Size from start address we need.
+ * @cursor: Which page table to allocate
  *
- * Make sure the page directories and page tables are allocated
+ * Make sure a specific page table or directory is allocated.
  *
  * Returns:
- * 0 on success, errno otherwise.
+ * 1 if page table needed to be allocated, 0 if page table was already
+ * allocated, negative errno if an error occurred.
  */
-int amdgpu_vm_alloc_pts(struct amdgpu_device *adev,
-			struct amdgpu_vm *vm,
-			uint64_t saddr, uint64_t size)
+static int amdgpu_vm_alloc_pts(struct amdgpu_device *adev,
+			       struct amdgpu_vm *vm,
+			       struct amdgpu_vm_pt_cursor *cursor)
 {
-	struct amdgpu_vm_pt_cursor cursor;
+	struct amdgpu_vm_pt *entry = cursor->entry;
+	struct amdgpu_bo_param bp;
 	struct amdgpu_bo *pt;
-	bool ats = false;
-	uint64_t eaddr;
 	int r;
 
-	/* validate the parameters */
-	if (saddr & AMDGPU_GPU_PAGE_MASK || size & AMDGPU_GPU_PAGE_MASK)
-		return -EINVAL;
-
-	eaddr = saddr + size - 1;
-
-	if (vm->pte_support_ats)
-		ats = saddr < AMDGPU_GMC_HOLE_START;
-
-	saddr /= AMDGPU_GPU_PAGE_SIZE;
-	eaddr /= AMDGPU_GPU_PAGE_SIZE;
+	if (cursor->level < AMDGPU_VM_PTB && !entry->entries) {
+		unsigned num_entries;
 
-	if (eaddr >= adev->vm_manager.max_pfn) {
-		dev_err(adev->dev, "va above limit (0x%08llX >= 0x%08llX)\n",
-			eaddr, adev->vm_manager.max_pfn);
-		return -EINVAL;
+		num_entries = amdgpu_vm_num_entries(adev, cursor->level);
+		entry->entries = kvmalloc_array(num_entries,
+						sizeof(*entry->entries),
+						GFP_KERNEL | __GFP_ZERO);
+		if (!entry->entries)
+			return -ENOMEM;
 	}
 
-	for_each_amdgpu_vm_pt_leaf(adev, vm, saddr, eaddr, cursor) {
-		struct amdgpu_vm_pt *entry = cursor.entry;
-		struct amdgpu_bo_param bp;
-
-		if (cursor.level < AMDGPU_VM_PTB) {
-			unsigned num_entries;
-
-			num_entries = amdgpu_vm_num_entries(adev, cursor.level);
-			entry->entries = kvmalloc_array(num_entries,
-							sizeof(*entry->entries),
-							GFP_KERNEL |
-							__GFP_ZERO);
-			if (!entry->entries)
-				return -ENOMEM;
-		}
-
-
-		if (entry->base.bo)
-			continue;
-
-		amdgpu_vm_bo_param(adev, vm, cursor.level, &bp);
+	if (entry->base.bo)
+		return 0;
 
-		r = amdgpu_bo_create(adev, &bp, &pt);
-		if (r)
-			return r;
+	amdgpu_vm_bo_param(adev, vm, cursor->level, &bp);
 
-		if (vm->use_cpu_for_update) {
-			r = amdgpu_bo_kmap(pt, NULL);
-			if (r)
-				goto error_free_pt;
-		}
-
-		/* Keep a reference to the root directory to avoid
-		* freeing them up in the wrong order.
-		*/
-		pt->parent = amdgpu_bo_ref(cursor.parent->base.bo);
+	r = amdgpu_bo_create(adev, &bp, &pt);
+	if (r)
+		return r;
 
-		amdgpu_vm_bo_base_init(&entry->base, vm, pt);
+	/* Keep a reference to the root directory to avoid
+	 * freeing them up in the wrong order.
+	 */
+	pt->parent = amdgpu_bo_ref(cursor->parent->base.bo);
+	amdgpu_vm_bo_base_init(&entry->base, vm, pt);
 
-		r = amdgpu_vm_clear_bo(adev, vm, pt, cursor.level, ats);
-		if (r)
-			goto error_free_pt;
-	}
+	r = amdgpu_vm_clear_bo(adev, vm, pt);
+	if (r)
+		goto error_free_pt;
 
 	return 0;
 
@@ -976,31 +898,45 @@ error_free_pt:
 }
 
 /**
+ * amdgpu_vm_free_table - fre one PD/PT
+ *
+ * @entry: PDE to free
+ */
+static void amdgpu_vm_free_table(struct amdgpu_vm_pt *entry)
+{
+	if (entry->base.bo) {
+		entry->base.bo->vm_bo = NULL;
+		list_del(&entry->base.vm_status);
+		amdgpu_bo_unref(&entry->base.bo->shadow);
+		amdgpu_bo_unref(&entry->base.bo);
+	}
+	kvfree(entry->entries);
+	entry->entries = NULL;
+}
+
+/**
  * amdgpu_vm_free_pts - free PD/PT levels
  *
  * @adev: amdgpu device structure
  * @vm: amdgpu vm structure
+ * @start: optional cursor where to start freeing PDs/PTs
  *
  * Free the page directory or page table level and all sub levels.
  */
 static void amdgpu_vm_free_pts(struct amdgpu_device *adev,
-			       struct amdgpu_vm *vm)
+			       struct amdgpu_vm *vm,
+			       struct amdgpu_vm_pt_cursor *start)
 {
 	struct amdgpu_vm_pt_cursor cursor;
 	struct amdgpu_vm_pt *entry;
 
-	for_each_amdgpu_vm_pt_dfs_safe(adev, vm, cursor, entry) {
+	vm->bulk_moveable = false;
 
-		if (entry->base.bo) {
-			entry->base.bo->vm_bo = NULL;
-			list_del(&entry->base.vm_status);
-			amdgpu_bo_unref(&entry->base.bo->shadow);
-			amdgpu_bo_unref(&entry->base.bo);
-		}
-		kvfree(entry->entries);
-	}
+	for_each_amdgpu_vm_pt_dfs_safe(adev, vm, start, cursor, entry)
+		amdgpu_vm_free_table(entry);
 
-	BUG_ON(vm->root.base.bo);
+	if (start)
+		amdgpu_vm_free_table(start->entry);
 }
 
 /**
@@ -1212,66 +1148,6 @@ struct amdgpu_bo_va *amdgpu_vm_bo_find(struct amdgpu_vm *vm,
 }
 
 /**
- * amdgpu_vm_do_set_ptes - helper to call the right asic function
- *
- * @params: see amdgpu_pte_update_params definition
- * @bo: PD/PT to update
- * @pe: addr of the page entry
- * @addr: dst addr to write into pe
- * @count: number of page entries to update
- * @incr: increase next addr by incr bytes
- * @flags: hw access flags
- *
- * Traces the parameters and calls the right asic functions
- * to setup the page table using the DMA.
- */
-static void amdgpu_vm_do_set_ptes(struct amdgpu_pte_update_params *params,
-				  struct amdgpu_bo *bo,
-				  uint64_t pe, uint64_t addr,
-				  unsigned count, uint32_t incr,
-				  uint64_t flags)
-{
-	pe += amdgpu_bo_gpu_offset(bo);
-	trace_amdgpu_vm_set_ptes(pe, addr, count, incr, flags);
-
-	if (count < 3) {
-		amdgpu_vm_write_pte(params->adev, params->ib, pe,
-				    addr | flags, count, incr);
-
-	} else {
-		amdgpu_vm_set_pte_pde(params->adev, params->ib, pe, addr,
-				      count, incr, flags);
-	}
-}
-
-/**
- * amdgpu_vm_do_copy_ptes - copy the PTEs from the GART
- *
- * @params: see amdgpu_pte_update_params definition
- * @bo: PD/PT to update
- * @pe: addr of the page entry
- * @addr: dst addr to write into pe
- * @count: number of page entries to update
- * @incr: increase next addr by incr bytes
- * @flags: hw access flags
- *
- * Traces the parameters and calls the DMA function to copy the PTEs.
- */
-static void amdgpu_vm_do_copy_ptes(struct amdgpu_pte_update_params *params,
-				   struct amdgpu_bo *bo,
-				   uint64_t pe, uint64_t addr,
-				   unsigned count, uint32_t incr,
-				   uint64_t flags)
-{
-	uint64_t src = (params->src + (addr >> 12) * 8);
-
-	pe += amdgpu_bo_gpu_offset(bo);
-	trace_amdgpu_vm_copy_ptes(pe, src, count);
-
-	amdgpu_vm_copy_pte(params->adev, params->ib, pe, src, count);
-}
-
-/**
  * amdgpu_vm_map_gart - Resolve gart mapping of addr
  *
  * @pages_addr: optional DMA address to use for lookup
@@ -1283,7 +1159,7 @@ static void amdgpu_vm_do_copy_ptes(struct amdgpu_pte_update_params *params,
  * Returns:
  * The pointer for the page table entry.
  */
-static uint64_t amdgpu_vm_map_gart(const dma_addr_t *pages_addr, uint64_t addr)
+uint64_t amdgpu_vm_map_gart(const dma_addr_t *pages_addr, uint64_t addr)
 {
 	uint64_t result;
 
@@ -1298,88 +1174,31 @@ static uint64_t amdgpu_vm_map_gart(const dma_addr_t *pages_addr, uint64_t addr)
 	return result;
 }
 
-/**
- * amdgpu_vm_cpu_set_ptes - helper to update page tables via CPU
- *
- * @params: see amdgpu_pte_update_params definition
- * @bo: PD/PT to update
- * @pe: kmap addr of the page entry
- * @addr: dst addr to write into pe
- * @count: number of page entries to update
- * @incr: increase next addr by incr bytes
- * @flags: hw access flags
- *
- * Write count number of PT/PD entries directly.
- */
-static void amdgpu_vm_cpu_set_ptes(struct amdgpu_pte_update_params *params,
-				   struct amdgpu_bo *bo,
-				   uint64_t pe, uint64_t addr,
-				   unsigned count, uint32_t incr,
-				   uint64_t flags)
-{
-	unsigned int i;
-	uint64_t value;
-
-	pe += (unsigned long)amdgpu_bo_kptr(bo);
-
-	trace_amdgpu_vm_set_ptes(pe, addr, count, incr, flags);
-
-	for (i = 0; i < count; i++) {
-		value = params->pages_addr ?
-			amdgpu_vm_map_gart(params->pages_addr, addr) :
-			addr;
-		amdgpu_gmc_set_pte_pde(params->adev, (void *)(uintptr_t)pe,
-				       i, value, flags);
-		addr += incr;
-	}
-}
-
-/**
- * amdgpu_vm_update_func - helper to call update function
- *
- * Calls the update function for both the given BO as well as its shadow.
- */
-static void amdgpu_vm_update_func(struct amdgpu_pte_update_params *params,
-				  struct amdgpu_bo *bo,
-				  uint64_t pe, uint64_t addr,
-				  unsigned count, uint32_t incr,
-				  uint64_t flags)
-{
-	if (bo->shadow)
-		params->func(params, bo->shadow, pe, addr, count, incr, flags);
-	params->func(params, bo, pe, addr, count, incr, flags);
-}
-
 /*
  * amdgpu_vm_update_pde - update a single level in the hierarchy
  *
  * @param: parameters for the update
  * @vm: requested vm
- * @parent: parent directory
  * @entry: entry to update
  *
  * Makes sure the requested entry in parent is up to date.
  */
-static void amdgpu_vm_update_pde(struct amdgpu_pte_update_params *params,
-				 struct amdgpu_vm *vm,
-				 struct amdgpu_vm_pt *parent,
-				 struct amdgpu_vm_pt *entry)
+static int amdgpu_vm_update_pde(struct amdgpu_vm_update_params *params,
+				struct amdgpu_vm *vm,
+				struct amdgpu_vm_pt *entry)
 {
+	struct amdgpu_vm_pt *parent = amdgpu_vm_pt_parent(entry);
 	struct amdgpu_bo *bo = parent->base.bo, *pbo;
 	uint64_t pde, pt, flags;
 	unsigned level;
 
-	/* Don't update huge pages here */
-	if (entry->huge)
-		return;
-
 	for (level = 0, pbo = bo->parent; pbo; ++level)
 		pbo = pbo->parent;
 
 	level += params->adev->vm_manager.root_level;
 	amdgpu_gmc_get_pde_for_bo(entry->base.bo, level, &pt, &flags);
 	pde = (entry - parent->entries) * 8;
-	amdgpu_vm_update_func(params, bo, pde, pt, 1, 0, flags);
+	return vm->update_funcs->update(params, bo, pde, pt, 1, 0, flags);
 }
 
 /*
@@ -1396,7 +1215,7 @@ static void amdgpu_vm_invalidate_pds(struct amdgpu_device *adev,
 	struct amdgpu_vm_pt_cursor cursor;
 	struct amdgpu_vm_pt *entry;
 
-	for_each_amdgpu_vm_pt_dfs_safe(adev, vm, cursor, entry)
+	for_each_amdgpu_vm_pt_dfs_safe(adev, vm, NULL, cursor, entry)
 		if (entry->base.bo && !entry->base.moved)
 			amdgpu_vm_bo_relocated(&entry->base);
 }
@@ -1415,89 +1234,39 @@ static void amdgpu_vm_invalidate_pds(struct amdgpu_device *adev,
 int amdgpu_vm_update_directories(struct amdgpu_device *adev,
 				 struct amdgpu_vm *vm)
 {
-	struct amdgpu_pte_update_params params;
-	struct amdgpu_job *job;
-	unsigned ndw = 0;
-	int r = 0;
+	struct amdgpu_vm_update_params params;
+	int r;
 
 	if (list_empty(&vm->relocated))
 		return 0;
 
-restart:
 	memset(&params, 0, sizeof(params));
 	params.adev = adev;
+	params.vm = vm;
 
-	if (vm->use_cpu_for_update) {
-		r = amdgpu_bo_sync_wait(vm->root.base.bo,
-					AMDGPU_FENCE_OWNER_VM, true);
-		if (unlikely(r))
-			return r;
-
-		params.func = amdgpu_vm_cpu_set_ptes;
-	} else {
-		ndw = 512 * 8;
-		r = amdgpu_job_alloc_with_ib(adev, ndw * 4, &job);
-		if (r)
-			return r;
-
-		params.ib = &job->ibs[0];
-		params.func = amdgpu_vm_do_set_ptes;
-	}
+	r = vm->update_funcs->prepare(&params, AMDGPU_FENCE_OWNER_VM, NULL);
+	if (r)
+		return r;
 
 	while (!list_empty(&vm->relocated)) {
-		struct amdgpu_vm_pt *pt, *entry;
+		struct amdgpu_vm_pt *entry;
 
 		entry = list_first_entry(&vm->relocated, struct amdgpu_vm_pt,
 					 base.vm_status);
 		amdgpu_vm_bo_idle(&entry->base);
 
-		pt = amdgpu_vm_pt_parent(entry);
-		if (!pt)
-			continue;
-
-		amdgpu_vm_update_pde(&params, vm, pt, entry);
-
-		if (!vm->use_cpu_for_update &&
-		    (ndw - params.ib->length_dw) < 32)
-			break;
-	}
-
-	if (vm->use_cpu_for_update) {
-		/* Flush HDP */
-		mb();
-		amdgpu_asic_flush_hdp(adev, NULL);
-	} else if (params.ib->length_dw == 0) {
-		amdgpu_job_free(job);
-	} else {
-		struct amdgpu_bo *root = vm->root.base.bo;
-		struct amdgpu_ring *ring;
-		struct dma_fence *fence;
-
-		ring = container_of(vm->entity.rq->sched, struct amdgpu_ring,
-				    sched);
-
-		amdgpu_ring_pad_ib(ring, params.ib);
-		amdgpu_sync_resv(adev, &job->sync, root->tbo.resv,
-				 AMDGPU_FENCE_OWNER_VM, false);
-		WARN_ON(params.ib->length_dw > ndw);
-		r = amdgpu_job_submit(job, &vm->entity, AMDGPU_FENCE_OWNER_VM,
-				      &fence);
+		r = amdgpu_vm_update_pde(&params, vm, entry);
 		if (r)
 			goto error;
-
-		amdgpu_bo_fence(root, fence, true);
-		dma_fence_put(vm->last_update);
-		vm->last_update = fence;
 	}
 
-	if (!list_empty(&vm->relocated))
-		goto restart;
-
+	r = vm->update_funcs->commit(&params, &vm->last_update);
+	if (r)
+		goto error;
 	return 0;
 
 error:
 	amdgpu_vm_invalidate_pds(adev, vm);
-	amdgpu_job_free(job);
 	return r;
 }
 
@@ -1506,7 +1275,7 @@ error:
  *
  * Make sure to set the right flags for the PTEs at the desired level.
  */
-static void amdgpu_vm_update_flags(struct amdgpu_pte_update_params *params,
+static void amdgpu_vm_update_flags(struct amdgpu_vm_update_params *params,
 				   struct amdgpu_bo *bo, unsigned level,
 				   uint64_t pe, uint64_t addr,
 				   unsigned count, uint32_t incr,
@@ -1525,13 +1294,14 @@ static void amdgpu_vm_update_flags(struct amdgpu_pte_update_params *params,
 		flags |= AMDGPU_PTE_EXECUTABLE;
 	}
 
-	amdgpu_vm_update_func(params, bo, pe, addr, count, incr, flags);
+	params->vm->update_funcs->update(params, bo, pe, addr, count, incr,
+					 flags);
 }
 
 /**
  * amdgpu_vm_fragment - get fragment for PTEs
  *
- * @params: see amdgpu_pte_update_params definition
+ * @params: see amdgpu_vm_update_params definition
  * @start: first PTE to handle
  * @end: last PTE to handle
  * @flags: hw mapping flags
@@ -1540,7 +1310,7 @@ static void amdgpu_vm_update_flags(struct amdgpu_pte_update_params *params,
  *
  * Returns the first possible fragment for the start and end address.
  */
-static void amdgpu_vm_fragment(struct amdgpu_pte_update_params *params,
+static void amdgpu_vm_fragment(struct amdgpu_vm_update_params *params,
 			       uint64_t start, uint64_t end, uint64_t flags,
 			       unsigned int *frag, uint64_t *frag_end)
 {
@@ -1573,7 +1343,7 @@ static void amdgpu_vm_fragment(struct amdgpu_pte_update_params *params,
 		max_frag = 31;
 
 	/* system pages are non continuously */
-	if (params->src) {
+	if (params->pages_addr) {
 		*frag = 0;
 		*frag_end = end;
 		return;
@@ -1592,7 +1362,7 @@ static void amdgpu_vm_fragment(struct amdgpu_pte_update_params *params,
 /**
  * amdgpu_vm_update_ptes - make sure that page tables are valid
  *
- * @params: see amdgpu_pte_update_params definition
+ * @params: see amdgpu_vm_update_params definition
  * @start: start of GPU address range
  * @end: end of GPU address range
  * @dst: destination address to map to, the next dst inside the function
@@ -1603,7 +1373,7 @@ static void amdgpu_vm_fragment(struct amdgpu_pte_update_params *params,
  * Returns:
  * 0 for success, -EINVAL for failure.
  */
-static int amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params,
+static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params,
 				 uint64_t start, uint64_t end,
 				 uint64_t dst, uint64_t flags)
 {
@@ -1611,6 +1381,7 @@ static int amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params,
 	struct amdgpu_vm_pt_cursor cursor;
 	uint64_t frag_start = start, frag_end;
 	unsigned int frag;
+	int r;
 
 	/* figure out the initial fragment */
 	amdgpu_vm_fragment(params, frag_start, end, flags, &frag, &frag_end);
@@ -1618,12 +1389,15 @@ static int amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params,
 	/* walk over the address space and update the PTs */
 	amdgpu_vm_pt_start(adev, params->vm, start, &cursor);
 	while (cursor.pfn < end) {
-		struct amdgpu_bo *pt = cursor.entry->base.bo;
 		unsigned shift, parent_shift, mask;
 		uint64_t incr, entry_end, pe_start;
+		struct amdgpu_bo *pt;
 
-		if (!pt)
-			return -ENOENT;
+		r = amdgpu_vm_alloc_pts(params->adev, params->vm, &cursor);
+		if (r)
+			return r;
+
+		pt = cursor.entry->base.bo;
 
 		/* The root level can't be a huge page */
 		if (cursor.level == adev->vm_manager.root_level) {
@@ -1632,16 +1406,10 @@ static int amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params,
 			continue;
 		}
 
-		/* If it isn't already handled it can't be a huge page */
-		if (cursor.entry->huge) {
-			/* Add the entry to the relocated list to update it. */
-			cursor.entry->huge = false;
-			amdgpu_vm_bo_relocated(&cursor.entry->base);
-		}
-
 		shift = amdgpu_vm_level_shift(adev, cursor.level);
 		parent_shift = amdgpu_vm_level_shift(adev, cursor.level - 1);
-		if (adev->asic_type < CHIP_VEGA10) {
+		if (adev->asic_type < CHIP_VEGA10 &&
+		    (flags & AMDGPU_PTE_VALID)) {
 			/* No huge page support before GMC v9 */
 			if (cursor.level != AMDGPU_VM_PTB) {
 				if (!amdgpu_vm_pt_descendant(adev, &cursor))
@@ -1697,9 +1465,9 @@ static int amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params,
 		} while (frag_start < entry_end);
 
 		if (amdgpu_vm_pt_descendant(adev, &cursor)) {
-			/* Mark all child entries as huge */
+			/* Free all child entries */
 			while (cursor.pfn < frag_start) {
-				cursor.entry->huge = true;
+				amdgpu_vm_free_pts(adev, params->vm, &cursor);
 				amdgpu_vm_pt_next(adev, &cursor);
 			}
 
@@ -1738,137 +1506,28 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev,
 				       uint64_t flags, uint64_t addr,
 				       struct dma_fence **fence)
 {
-	struct amdgpu_ring *ring;
+	struct amdgpu_vm_update_params params;
 	void *owner = AMDGPU_FENCE_OWNER_VM;
-	unsigned nptes, ncmds, ndw;
-	struct amdgpu_job *job;
-	struct amdgpu_pte_update_params params;
-	struct dma_fence *f = NULL;
 	int r;
 
 	memset(&params, 0, sizeof(params));
 	params.adev = adev;
 	params.vm = vm;
+	params.pages_addr = pages_addr;
 
 	/* sync to everything except eviction fences on unmapping */
 	if (!(flags & AMDGPU_PTE_VALID))
 		owner = AMDGPU_FENCE_OWNER_KFD;
 
-	if (vm->use_cpu_for_update) {
-		/* params.src is used as flag to indicate system Memory */
-		if (pages_addr)
-			params.src = ~0;
-
-		/* Wait for PT BOs to be idle. PTs share the same resv. object
-		 * as the root PD BO
-		 */
-		r = amdgpu_bo_sync_wait(vm->root.base.bo, owner, true);
-		if (unlikely(r))
-			return r;
-
-		/* Wait for any BO move to be completed */
-		if (exclusive) {
-			r = dma_fence_wait(exclusive, true);
-			if (unlikely(r))
-				return r;
-		}
-
-		params.func = amdgpu_vm_cpu_set_ptes;
-		params.pages_addr = pages_addr;
-		return amdgpu_vm_update_ptes(&params, start, last + 1,
-					     addr, flags);
-	}
-
-	ring = container_of(vm->entity.rq->sched, struct amdgpu_ring, sched);
-
-	nptes = last - start + 1;
-
-	/*
-	 * reserve space for two commands every (1 << BLOCK_SIZE)
-	 *  entries or 2k dwords (whatever is smaller)
-	 */
-	ncmds = ((nptes >> min(adev->vm_manager.block_size, 11u)) + 1);
-
-	/* The second command is for the shadow pagetables. */
-	if (vm->root.base.bo->shadow)
-		ncmds *= 2;
-
-	/* padding, etc. */
-	ndw = 64;
-
-	if (pages_addr) {
-		/* copy commands needed */
-		ndw += ncmds * adev->vm_manager.vm_pte_funcs->copy_pte_num_dw;
-
-		/* and also PTEs */
-		ndw += nptes * 2;
-
-		params.func = amdgpu_vm_do_copy_ptes;
-
-	} else {
-		/* set page commands needed */
-		ndw += ncmds * 10;
-
-		/* extra commands for begin/end fragments */
-		ncmds = 2 * adev->vm_manager.fragment_size;
-		if (vm->root.base.bo->shadow)
-			ncmds *= 2;
-
-		ndw += 10 * ncmds;
-
-		params.func = amdgpu_vm_do_set_ptes;
-	}
-
-	r = amdgpu_job_alloc_with_ib(adev, ndw * 4, &job);
+	r = vm->update_funcs->prepare(&params, owner, exclusive);
 	if (r)
 		return r;
 
-	params.ib = &job->ibs[0];
-
-	if (pages_addr) {
-		uint64_t *pte;
-		unsigned i;
-
-		/* Put the PTEs at the end of the IB. */
-		i = ndw - nptes * 2;
-		pte= (uint64_t *)&(job->ibs->ptr[i]);
-		params.src = job->ibs->gpu_addr + i * 4;
-
-		for (i = 0; i < nptes; ++i) {
-			pte[i] = amdgpu_vm_map_gart(pages_addr, addr + i *
-						    AMDGPU_GPU_PAGE_SIZE);
-			pte[i] |= flags;
-		}
-		addr = 0;
-	}
-
-	r = amdgpu_sync_fence(adev, &job->sync, exclusive, false);
-	if (r)
-		goto error_free;
-
-	r = amdgpu_sync_resv(adev, &job->sync, vm->root.base.bo->tbo.resv,
-			     owner, false);
-	if (r)
-		goto error_free;
-
 	r = amdgpu_vm_update_ptes(&params, start, last + 1, addr, flags);
 	if (r)
-		goto error_free;
-
-	amdgpu_ring_pad_ib(ring, params.ib);
-	WARN_ON(params.ib->length_dw > ndw);
-	r = amdgpu_job_submit(job, &vm->entity, AMDGPU_FENCE_OWNER_VM, &f);
-	if (r)
-		goto error_free;
-
-	amdgpu_bo_fence(vm->root.base.bo, f, true);
-	dma_fence_put(*fence);
-	*fence = f;
-	return 0;
+		return r;
 
-error_free:
-	amdgpu_job_free(job);
-	return r;
+	return vm->update_funcs->commit(&params, fence);
 }
 
 /**
@@ -1880,6 +1539,7 @@ error_free:
  * @vm: requested vm
  * @mapping: mapped range and flags to use for the update
  * @flags: HW flags for the mapping
+ * @bo_adev: amdgpu_device pointer that bo actually been allocated
  * @nodes: array of drm_mm_nodes with the MC addresses
  * @fence: optional resulting fence
  *
@@ -1895,6 +1555,7 @@ static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev,
 				      struct amdgpu_vm *vm,
 				      struct amdgpu_bo_va_mapping *mapping,
 				      uint64_t flags,
+				      struct amdgpu_device *bo_adev,
 				      struct drm_mm_node *nodes,
 				      struct dma_fence **fence)
 {
@@ -1949,7 +1610,6 @@ static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev,
 		if (pages_addr) {
 			uint64_t count;
 
-			max_entries = min(max_entries, 16ull * 1024ull);
 			for (count = 1;
 			     count < max_entries / AMDGPU_GPU_PAGES_IN_CPU_PAGE;
 			     ++count) {
@@ -1969,7 +1629,7 @@ static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev,
 			}
 
 		} else if (flags & AMDGPU_PTE_VALID) {
-			addr += adev->vm_manager.vram_base_offset;
+			addr += bo_adev->vm_manager.vram_base_offset;
 			addr += pfn << PAGE_SHIFT;
 		}
 
@@ -2016,6 +1676,7 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev,
 	struct drm_mm_node *nodes;
 	struct dma_fence *exclusive, **last_update;
 	uint64_t flags;
+	struct amdgpu_device *bo_adev = adev;
 	int r;
 
 	if (clear || !bo) {
@@ -2034,10 +1695,12 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev,
 		exclusive = reservation_object_get_excl(bo->tbo.resv);
 	}
 
-	if (bo)
+	if (bo) {
 		flags = amdgpu_ttm_tt_pte_flags(adev, bo->tbo.ttm, mem);
-	else
+		bo_adev = amdgpu_ttm_adev(bo->tbo.bdev);
+	} else {
 		flags = 0x0;
+	}
 
 	if (clear || (bo && bo->tbo.resv == vm->root.base.bo->tbo.resv))
 		last_update = &vm->last_update;
@@ -2054,7 +1717,7 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev,
 
 	list_for_each_entry(mapping, &bo_va->invalids, list) {
 		r = amdgpu_vm_bo_split_mapping(adev, exclusive, pages_addr, vm,
-					       mapping, flags, nodes,
+					       mapping, flags, bo_adev, nodes,
 					       last_update);
 		if (r)
 			return r;
@@ -2374,6 +2037,16 @@ struct amdgpu_bo_va *amdgpu_vm_bo_add(struct amdgpu_device *adev,
 	INIT_LIST_HEAD(&bo_va->valids);
 	INIT_LIST_HEAD(&bo_va->invalids);
 
+	if (bo && amdgpu_xgmi_same_hive(adev, amdgpu_ttm_adev(bo->tbo.bdev)) &&
+	    (bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM)) {
+		bo_va->is_xgmi = true;
+		mutex_lock(&adev->vm_manager.lock_pstate);
+		/* Power up XGMI if it can be potentially used */
+		if (++adev->vm_manager.xgmi_map_counter == 1)
+			amdgpu_xgmi_set_pstate(adev, 1);
+		mutex_unlock(&adev->vm_manager.lock_pstate);
+	}
+
 	return bo_va;
 }
 
@@ -2792,6 +2465,14 @@ void amdgpu_vm_bo_rmv(struct amdgpu_device *adev,
 	}
 
 	dma_fence_put(bo_va->last_pt_update);
+
+	if (bo && bo_va->is_xgmi) {
+		mutex_lock(&adev->vm_manager.lock_pstate);
+		if (--adev->vm_manager.xgmi_map_counter == 0)
+			amdgpu_xgmi_set_pstate(adev, 0);
+		mutex_unlock(&adev->vm_manager.lock_pstate);
+	}
+
 	kfree(bo_va);
 }
 
@@ -2949,20 +2630,16 @@ void amdgpu_vm_adjust_size(struct amdgpu_device *adev, uint32_t min_vm_size,
 		 adev->vm_manager.fragment_size);
 }
 
-static struct amdgpu_retryfault_hashtable *init_fault_hash(void)
+/**
+ * amdgpu_vm_wait_idle - wait for the VM to become idle
+ *
+ * @vm: VM object to wait for
+ * @timeout: timeout to wait for VM to become idle
+ */
+long amdgpu_vm_wait_idle(struct amdgpu_vm *vm, long timeout)
 {
-	struct amdgpu_retryfault_hashtable *fault_hash;
-
-	fault_hash = kmalloc(sizeof(*fault_hash), GFP_KERNEL);
-	if (!fault_hash)
-		return fault_hash;
-
-	INIT_CHASH_TABLE(fault_hash->hash,
-			AMDGPU_PAGEFAULT_HASH_BITS, 8, 0);
-	spin_lock_init(&fault_hash->lock);
-	fault_hash->count = 0;
-
-	return fault_hash;
+	return reservation_object_wait_timeout_rcu(vm->root.base.bo->tbo.resv,
+						   true, true, timeout);
 }
 
 /**
@@ -3018,6 +2695,11 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 			 vm->use_cpu_for_update ? "CPU" : "SDMA");
 	WARN_ONCE((vm->use_cpu_for_update && !amdgpu_gmc_vram_full_visible(&adev->gmc)),
 		  "CPU update of VM recommended only for large BAR system\n");
+
+	if (vm->use_cpu_for_update)
+		vm->update_funcs = &amdgpu_vm_cpu_funcs;
+	else
+		vm->update_funcs = &amdgpu_vm_sdma_funcs;
 	vm->last_update = NULL;
 
 	amdgpu_vm_bo_param(adev, vm, adev->vm_manager.root_level, &bp);
@@ -3037,9 +2719,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 
 	amdgpu_vm_bo_base_init(&vm->root.base, vm, root);
 
-	r = amdgpu_vm_clear_bo(adev, vm, root,
-			       adev->vm_manager.root_level,
-			       vm->pte_support_ats);
+	r = amdgpu_vm_clear_bo(adev, vm, root);
 	if (r)
 		goto error_unreserve;
 
@@ -3058,12 +2738,6 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 		vm->pasid = pasid;
 	}
 
-	vm->fault_hash = init_fault_hash();
-	if (!vm->fault_hash) {
-		r = -ENOMEM;
-		goto error_free_root;
-	}
-
 	INIT_KFIFO(vm->faults);
 
 	return 0;
@@ -3134,9 +2808,8 @@ int amdgpu_vm_make_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm, uns
 	 * changing any other state, in case it fails.
 	 */
 	if (pte_support_ats != vm->pte_support_ats) {
-		r = amdgpu_vm_clear_bo(adev, vm, vm->root.base.bo,
-			       adev->vm_manager.root_level,
-			       pte_support_ats);
+		vm->pte_support_ats = pte_support_ats;
+		r = amdgpu_vm_clear_bo(adev, vm, vm->root.base.bo);
 		if (r)
 			goto free_idr;
 	}
@@ -3144,7 +2817,6 @@ int amdgpu_vm_make_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm, uns
 	/* Update VM state */
 	vm->use_cpu_for_update = !!(adev->vm_manager.vm_update_mode &
 				    AMDGPU_VM_USE_CPU_FOR_COMPUTE);
-	vm->pte_support_ats = pte_support_ats;
 	DRM_DEBUG_DRIVER("VM update mode is %s\n",
 			 vm->use_cpu_for_update ? "CPU" : "SDMA");
 	WARN_ONCE((vm->use_cpu_for_update && !amdgpu_gmc_vram_full_visible(&adev->gmc)),
@@ -3219,15 +2891,10 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 	struct amdgpu_bo_va_mapping *mapping, *tmp;
 	bool prt_fini_needed = !!adev->gmc.gmc_funcs->set_prt;
 	struct amdgpu_bo *root;
-	u64 fault;
 	int i, r;
 
 	amdgpu_amdkfd_gpuvm_destroy_cb(adev, vm);
 
-	/* Clear pending page faults from IH when the VM is destroyed */
-	while (kfifo_get(&vm->faults, &fault))
-		amdgpu_vm_clear_fault(vm->fault_hash, fault);
-
 	if (vm->pasid) {
 		unsigned long flags;
 
@@ -3236,9 +2903,6 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 		spin_unlock_irqrestore(&adev->vm_manager.pasid_lock, flags);
 	}
 
-	kfree(vm->fault_hash);
-	vm->fault_hash = NULL;
-
 	drm_sched_entity_destroy(&vm->entity);
 
 	if (!RB_EMPTY_ROOT(&vm->va.rb_root)) {
@@ -3267,10 +2931,11 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 	if (r) {
 		dev_err(adev->dev, "Leaking page tables because BO reservation failed\n");
 	} else {
-		amdgpu_vm_free_pts(adev, vm);
+		amdgpu_vm_free_pts(adev, vm, NULL);
 		amdgpu_bo_unreserve(root);
 	}
 	amdgpu_bo_unref(&root);
+	WARN_ON(vm->root.base.bo);
 	dma_fence_put(vm->last_update);
 	for (i = 0; i < AMDGPU_MAX_VMHUBS; i++)
 		amdgpu_vmid_free_reserved(adev, vm, i);
@@ -3315,6 +2980,9 @@ void amdgpu_vm_manager_init(struct amdgpu_device *adev)
 
 	idr_init(&adev->vm_manager.pasid_idr);
 	spin_lock_init(&adev->vm_manager.pasid_lock);
+
+	adev->vm_manager.xgmi_map_counter = 0;
+	mutex_init(&adev->vm_manager.lock_pstate);
 }
 
 /**
@@ -3405,78 +3073,3 @@ void amdgpu_vm_set_task_info(struct amdgpu_vm *vm)
 		}
 	}
 }
-
-/**
- * amdgpu_vm_add_fault - Add a page fault record to fault hash table
- *
- * @fault_hash: fault hash table
- * @key: 64-bit encoding of PASID and address
- *
- * This should be called when a retry page fault interrupt is
- * received. If this is a new page fault, it will be added to a hash
- * table. The return value indicates whether this is a new fault, or
- * a fault that was already known and is already being handled.
- *
- * If there are too many pending page faults, this will fail. Retry
- * interrupts should be ignored in this case until there is enough
- * free space.
- *
- * Returns 0 if the fault was added, 1 if the fault was already known,
- * -ENOSPC if there are too many pending faults.
- */
-int amdgpu_vm_add_fault(struct amdgpu_retryfault_hashtable *fault_hash, u64 key)
-{
-	unsigned long flags;
-	int r = -ENOSPC;
-
-	if (WARN_ON_ONCE(!fault_hash))
-		/* Should be allocated in amdgpu_vm_init
-		 */
-		return r;
-
-	spin_lock_irqsave(&fault_hash->lock, flags);
-
-	/* Only let the hash table fill up to 50% for best performance */
-	if (fault_hash->count >= (1 << (AMDGPU_PAGEFAULT_HASH_BITS-1)))
-		goto unlock_out;
-
-	r = chash_table_copy_in(&fault_hash->hash, key, NULL);
-	if (!r)
-		fault_hash->count++;
-
-	/* chash_table_copy_in should never fail unless we're losing count */
-	WARN_ON_ONCE(r < 0);
-
-unlock_out:
-	spin_unlock_irqrestore(&fault_hash->lock, flags);
-	return r;
-}
-
-/**
- * amdgpu_vm_clear_fault - Remove a page fault record
- *
- * @fault_hash: fault hash table
- * @key: 64-bit encoding of PASID and address
- *
- * This should be called when a page fault has been handled. Any
- * future interrupt with this key will be processed as a new
- * page fault.
- */
-void amdgpu_vm_clear_fault(struct amdgpu_retryfault_hashtable *fault_hash, u64 key)
-{
-	unsigned long flags;
-	int r;
-
-	if (!fault_hash)
-		return;
-
-	spin_lock_irqsave(&fault_hash->lock, flags);
-
-	r = chash_table_remove(&fault_hash->hash, key, NULL);
-	if (!WARN_ON_ONCE(r < 0)) {
-		fault_hash->count--;
-		WARN_ON_ONCE(fault_hash->count < 0);
-	}
-
-	spin_unlock_irqrestore(&fault_hash->lock, flags);
-}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index 81ff8177f092..91baf95212a6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -30,7 +30,6 @@
 #include <drm/gpu_scheduler.h>
 #include <drm/drm_file.h>
 #include <drm/ttm/ttm_bo_driver.h>
-#include <linux/chash.h>
 
 #include "amdgpu_sync.h"
 #include "amdgpu_ring.h"
@@ -140,7 +139,6 @@ struct amdgpu_vm_bo_base {
 
 struct amdgpu_vm_pt {
 	struct amdgpu_vm_bo_base	base;
-	bool				huge;
 
 	/* array of page tables, one for each directory entry */
 	struct amdgpu_vm_pt		*entries;
@@ -167,11 +165,6 @@ struct amdgpu_vm_pte_funcs {
 			    uint32_t incr, uint64_t flags);
 };
 
-#define AMDGPU_VM_FAULT(pasid, addr) (((u64)(pasid) << 48) | (addr))
-#define AMDGPU_VM_FAULT_PASID(fault) ((u64)(fault) >> 48)
-#define AMDGPU_VM_FAULT_ADDR(fault)  ((u64)(fault) & 0xfffffffff000ULL)
-
-
 struct amdgpu_task_info {
 	char	process_name[TASK_COMM_LEN];
 	char	task_name[TASK_COMM_LEN];
@@ -179,11 +172,52 @@ struct amdgpu_task_info {
 	pid_t	tgid;
 };
 
-#define AMDGPU_PAGEFAULT_HASH_BITS 8
-struct amdgpu_retryfault_hashtable {
-	DECLARE_CHASH_TABLE(hash, AMDGPU_PAGEFAULT_HASH_BITS, 8, 0);
-	spinlock_t	lock;
-	int		count;
+/**
+ * struct amdgpu_vm_update_params
+ *
+ * Encapsulate some VM table update parameters to reduce
+ * the number of function parameters
+ *
+ */
+struct amdgpu_vm_update_params {
+
+	/**
+	 * @adev: amdgpu device we do this update for
+	 */
+	struct amdgpu_device *adev;
+
+	/**
+	 * @vm: optional amdgpu_vm we do this update for
+	 */
+	struct amdgpu_vm *vm;
+
+	/**
+	 * @pages_addr:
+	 *
+	 * DMA addresses to use for mapping
+	 */
+	dma_addr_t *pages_addr;
+
+	/**
+	 * @job: job to used for hw submission
+	 */
+	struct amdgpu_job *job;
+
+	/**
+	 * @num_dw_left: number of dw left for the IB
+	 */
+	unsigned int num_dw_left;
+};
+
+struct amdgpu_vm_update_funcs {
+	int (*map_table)(struct amdgpu_bo *bo);
+	int (*prepare)(struct amdgpu_vm_update_params *p, void * owner,
+		       struct dma_fence *exclusive);
+	int (*update)(struct amdgpu_vm_update_params *p,
+		      struct amdgpu_bo *bo, uint64_t pe, uint64_t addr,
+		      unsigned count, uint32_t incr, uint64_t flags);
+	int (*commit)(struct amdgpu_vm_update_params *p,
+		      struct dma_fence **fence);
 };
 
 struct amdgpu_vm {
@@ -221,7 +255,10 @@ struct amdgpu_vm {
 	struct amdgpu_vmid	*reserved_vmid[AMDGPU_MAX_VMHUBS];
 
 	/* Flag to indicate if VM tables are updated by CPU or GPU (SDMA) */
-	bool                    use_cpu_for_update;
+	bool					use_cpu_for_update;
+
+	/* Functions to use for VM table updates */
+	const struct amdgpu_vm_update_funcs	*update_funcs;
 
 	/* Flag to indicate ATS support from PTE for GFX9 */
 	bool			pte_support_ats;
@@ -245,7 +282,6 @@ struct amdgpu_vm {
 	struct ttm_lru_bulk_move lru_bulk_move;
 	/* mark whether can do the bulk move */
 	bool			bulk_moveable;
-	struct amdgpu_retryfault_hashtable *fault_hash;
 };
 
 struct amdgpu_vm_manager {
@@ -267,6 +303,7 @@ struct amdgpu_vm_manager {
 	const struct amdgpu_vm_pte_funcs	*vm_pte_funcs;
 	struct drm_sched_rq			*vm_pte_rqs[AMDGPU_MAX_RINGS];
 	unsigned				vm_pte_num_rqs;
+	struct amdgpu_ring			*page_fault;
 
 	/* partial resident texture handling */
 	spinlock_t				prt_lock;
@@ -283,14 +320,23 @@ struct amdgpu_vm_manager {
 	 */
 	struct idr				pasid_idr;
 	spinlock_t				pasid_lock;
+
+	/* counter of mapped memory through xgmi */
+	uint32_t				xgmi_map_counter;
+	struct mutex				lock_pstate;
 };
 
 #define amdgpu_vm_copy_pte(adev, ib, pe, src, count) ((adev)->vm_manager.vm_pte_funcs->copy_pte((ib), (pe), (src), (count)))
 #define amdgpu_vm_write_pte(adev, ib, pe, value, count, incr) ((adev)->vm_manager.vm_pte_funcs->write_pte((ib), (pe), (value), (count), (incr)))
 #define amdgpu_vm_set_pte_pde(adev, ib, pe, addr, count, incr, flags) ((adev)->vm_manager.vm_pte_funcs->set_pte_pde((ib), (pe), (addr), (count), (incr), (flags)))
 
+extern const struct amdgpu_vm_update_funcs amdgpu_vm_cpu_funcs;
+extern const struct amdgpu_vm_update_funcs amdgpu_vm_sdma_funcs;
+
 void amdgpu_vm_manager_init(struct amdgpu_device *adev);
 void amdgpu_vm_manager_fini(struct amdgpu_device *adev);
+
+long amdgpu_vm_wait_idle(struct amdgpu_vm *vm, long timeout);
 int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 		   int vm_context, unsigned int pasid);
 int amdgpu_vm_make_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm, unsigned int pasid);
@@ -303,9 +349,6 @@ bool amdgpu_vm_ready(struct amdgpu_vm *vm);
 int amdgpu_vm_validate_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 			      int (*callback)(void *p, struct amdgpu_bo *bo),
 			      void *param);
-int amdgpu_vm_alloc_pts(struct amdgpu_device *adev,
-			struct amdgpu_vm *vm,
-			uint64_t saddr, uint64_t size);
 int amdgpu_vm_flush(struct amdgpu_ring *ring, struct amdgpu_job *job, bool need_pipe_sync);
 int amdgpu_vm_update_directories(struct amdgpu_device *adev,
 				 struct amdgpu_vm *vm);
@@ -319,6 +362,7 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev,
 			bool clear);
 void amdgpu_vm_bo_invalidate(struct amdgpu_device *adev,
 			     struct amdgpu_bo *bo, bool evicted);
+uint64_t amdgpu_vm_map_gart(const dma_addr_t *pages_addr, uint64_t addr);
 struct amdgpu_bo_va *amdgpu_vm_bo_find(struct amdgpu_vm *vm,
 				       struct amdgpu_bo *bo);
 struct amdgpu_bo_va *amdgpu_vm_bo_add(struct amdgpu_device *adev,
@@ -358,11 +402,6 @@ void amdgpu_vm_set_task_info(struct amdgpu_vm *vm);
 
 void amdgpu_vm_move_to_lru_tail(struct amdgpu_device *adev,
 				struct amdgpu_vm *vm);
-
-int amdgpu_vm_add_fault(struct amdgpu_retryfault_hashtable *fault_hash, u64 key);
-
-void amdgpu_vm_clear_fault(struct amdgpu_retryfault_hashtable *fault_hash, u64 key);
-
 void amdgpu_vm_del_from_lru_notify(struct ttm_buffer_object *bo);
 
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c
new file mode 100644
index 000000000000..5222d165abfc
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c
@@ -0,0 +1,127 @@
+/*
+ * Copyright 2019 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#include "amdgpu_vm.h"
+#include "amdgpu_object.h"
+#include "amdgpu_trace.h"
+
+/**
+ * amdgpu_vm_cpu_map_table - make sure new PDs/PTs are kmapped
+ *
+ * @table: newly allocated or validated PD/PT
+ */
+static int amdgpu_vm_cpu_map_table(struct amdgpu_bo *table)
+{
+	return amdgpu_bo_kmap(table, NULL);
+}
+
+/**
+ * amdgpu_vm_cpu_prepare - prepare page table update with the CPU
+ *
+ * @p: see amdgpu_vm_update_params definition
+ * @owner: owner we need to sync to
+ * @exclusive: exclusive move fence we need to sync to
+ *
+ * Returns:
+ * Negativ errno, 0 for success.
+ */
+static int amdgpu_vm_cpu_prepare(struct amdgpu_vm_update_params *p, void *owner,
+				 struct dma_fence *exclusive)
+{
+	int r;
+
+	/* Wait for PT BOs to be idle. PTs share the same resv. object
+	 * as the root PD BO
+	 */
+	r = amdgpu_bo_sync_wait(p->vm->root.base.bo, owner, true);
+	if (unlikely(r))
+		return r;
+
+	/* Wait for any BO move to be completed */
+	if (exclusive) {
+		r = dma_fence_wait(exclusive, true);
+		if (unlikely(r))
+			return r;
+	}
+
+	return 0;
+}
+
+/**
+ * amdgpu_vm_cpu_update - helper to update page tables via CPU
+ *
+ * @p: see amdgpu_vm_update_params definition
+ * @bo: PD/PT to update
+ * @pe: kmap addr of the page entry
+ * @addr: dst addr to write into pe
+ * @count: number of page entries to update
+ * @incr: increase next addr by incr bytes
+ * @flags: hw access flags
+ *
+ * Write count number of PT/PD entries directly.
+ */
+static int amdgpu_vm_cpu_update(struct amdgpu_vm_update_params *p,
+				struct amdgpu_bo *bo, uint64_t pe,
+				uint64_t addr, unsigned count, uint32_t incr,
+				uint64_t flags)
+{
+	unsigned int i;
+	uint64_t value;
+
+	pe += (unsigned long)amdgpu_bo_kptr(bo);
+
+	trace_amdgpu_vm_set_ptes(pe, addr, count, incr, flags);
+
+	for (i = 0; i < count; i++) {
+		value = p->pages_addr ?
+			amdgpu_vm_map_gart(p->pages_addr, addr) :
+			addr;
+		amdgpu_gmc_set_pte_pde(p->adev, (void *)(uintptr_t)pe,
+				       i, value, flags);
+		addr += incr;
+	}
+	return 0;
+}
+
+/**
+ * amdgpu_vm_cpu_commit - commit page table update to the HW
+ *
+ * @p: see amdgpu_vm_update_params definition
+ * @fence: unused
+ *
+ * Make sure that the hardware sees the page table updates.
+ */
+static int amdgpu_vm_cpu_commit(struct amdgpu_vm_update_params *p,
+				struct dma_fence **fence)
+{
+	/* Flush HDP */
+	mb();
+	amdgpu_asic_flush_hdp(p->adev, NULL);
+	return 0;
+}
+
+const struct amdgpu_vm_update_funcs amdgpu_vm_cpu_funcs = {
+	.map_table = amdgpu_vm_cpu_map_table,
+	.prepare = amdgpu_vm_cpu_prepare,
+	.update = amdgpu_vm_cpu_update,
+	.commit = amdgpu_vm_cpu_commit
+};
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c
new file mode 100644
index 000000000000..ddd181f5ed37
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c
@@ -0,0 +1,270 @@
+/*
+ * Copyright 2019 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#include "amdgpu_vm.h"
+#include "amdgpu_job.h"
+#include "amdgpu_object.h"
+#include "amdgpu_trace.h"
+
+#define AMDGPU_VM_SDMA_MIN_NUM_DW	256u
+#define AMDGPU_VM_SDMA_MAX_NUM_DW	(16u * 1024u)
+
+/**
+ * amdgpu_vm_sdma_map_table - make sure new PDs/PTs are GTT mapped
+ *
+ * @table: newly allocated or validated PD/PT
+ */
+static int amdgpu_vm_sdma_map_table(struct amdgpu_bo *table)
+{
+	int r;
+
+	r = amdgpu_ttm_alloc_gart(&table->tbo);
+	if (r)
+		return r;
+
+	if (table->shadow)
+		r = amdgpu_ttm_alloc_gart(&table->shadow->tbo);
+
+	return r;
+}
+
+/**
+ * amdgpu_vm_sdma_prepare - prepare SDMA command submission
+ *
+ * @p: see amdgpu_vm_update_params definition
+ * @owner: owner we need to sync to
+ * @exclusive: exclusive move fence we need to sync to
+ *
+ * Returns:
+ * Negativ errno, 0 for success.
+ */
+static int amdgpu_vm_sdma_prepare(struct amdgpu_vm_update_params *p,
+				  void *owner, struct dma_fence *exclusive)
+{
+	struct amdgpu_bo *root = p->vm->root.base.bo;
+	unsigned int ndw = AMDGPU_VM_SDMA_MIN_NUM_DW;
+	int r;
+
+	r = amdgpu_job_alloc_with_ib(p->adev, ndw * 4, &p->job);
+	if (r)
+		return r;
+
+	r = amdgpu_sync_fence(p->adev, &p->job->sync, exclusive, false);
+	if (r)
+		return r;
+
+	r = amdgpu_sync_resv(p->adev, &p->job->sync, root->tbo.resv,
+			     owner, false);
+	if (r)
+		return r;
+
+	p->num_dw_left = ndw;
+	return 0;
+}
+
+/**
+ * amdgpu_vm_sdma_commit - commit SDMA command submission
+ *
+ * @p: see amdgpu_vm_update_params definition
+ * @fence: resulting fence
+ *
+ * Returns:
+ * Negativ errno, 0 for success.
+ */
+static int amdgpu_vm_sdma_commit(struct amdgpu_vm_update_params *p,
+				 struct dma_fence **fence)
+{
+	struct amdgpu_bo *root = p->vm->root.base.bo;
+	struct amdgpu_ib *ib = p->job->ibs;
+	struct amdgpu_ring *ring;
+	struct dma_fence *f;
+	int r;
+
+	ring = container_of(p->vm->entity.rq->sched, struct amdgpu_ring, sched);
+
+	WARN_ON(ib->length_dw == 0);
+	amdgpu_ring_pad_ib(ring, ib);
+	WARN_ON(ib->length_dw > p->num_dw_left);
+	r = amdgpu_job_submit(p->job, &p->vm->entity,
+			      AMDGPU_FENCE_OWNER_VM, &f);
+	if (r)
+		goto error;
+
+	amdgpu_bo_fence(root, f, true);
+	if (fence)
+		swap(*fence, f);
+	dma_fence_put(f);
+	return 0;
+
+error:
+	amdgpu_job_free(p->job);
+	return r;
+}
+
+
+/**
+ * amdgpu_vm_sdma_copy_ptes - copy the PTEs from mapping
+ *
+ * @p: see amdgpu_vm_update_params definition
+ * @bo: PD/PT to update
+ * @pe: addr of the page entry
+ * @count: number of page entries to copy
+ *
+ * Traces the parameters and calls the DMA function to copy the PTEs.
+ */
+static void amdgpu_vm_sdma_copy_ptes(struct amdgpu_vm_update_params *p,
+				     struct amdgpu_bo *bo, uint64_t pe,
+				     unsigned count)
+{
+	struct amdgpu_ib *ib = p->job->ibs;
+	uint64_t src = ib->gpu_addr;
+
+	src += p->num_dw_left * 4;
+
+	pe += amdgpu_bo_gpu_offset(bo);
+	trace_amdgpu_vm_copy_ptes(pe, src, count);
+
+	amdgpu_vm_copy_pte(p->adev, ib, pe, src, count);
+}
+
+/**
+ * amdgpu_vm_sdma_set_ptes - helper to call the right asic function
+ *
+ * @p: see amdgpu_vm_update_params definition
+ * @bo: PD/PT to update
+ * @pe: addr of the page entry
+ * @addr: dst addr to write into pe
+ * @count: number of page entries to update
+ * @incr: increase next addr by incr bytes
+ * @flags: hw access flags
+ *
+ * Traces the parameters and calls the right asic functions
+ * to setup the page table using the DMA.
+ */
+static void amdgpu_vm_sdma_set_ptes(struct amdgpu_vm_update_params *p,
+				    struct amdgpu_bo *bo, uint64_t pe,
+				    uint64_t addr, unsigned count,
+				    uint32_t incr, uint64_t flags)
+{
+	struct amdgpu_ib *ib = p->job->ibs;
+
+	pe += amdgpu_bo_gpu_offset(bo);
+	trace_amdgpu_vm_set_ptes(pe, addr, count, incr, flags);
+	if (count < 3) {
+		amdgpu_vm_write_pte(p->adev, ib, pe, addr | flags,
+				    count, incr);
+	} else {
+		amdgpu_vm_set_pte_pde(p->adev, ib, pe, addr,
+				      count, incr, flags);
+	}
+}
+
+/**
+ * amdgpu_vm_sdma_update - execute VM update
+ *
+ * @p: see amdgpu_vm_update_params definition
+ * @bo: PD/PT to update
+ * @pe: addr of the page entry
+ * @addr: dst addr to write into pe
+ * @count: number of page entries to update
+ * @incr: increase next addr by incr bytes
+ * @flags: hw access flags
+ *
+ * Reserve space in the IB, setup mapping buffer on demand and write commands to
+ * the IB.
+ */
+static int amdgpu_vm_sdma_update(struct amdgpu_vm_update_params *p,
+				 struct amdgpu_bo *bo, uint64_t pe,
+				 uint64_t addr, unsigned count, uint32_t incr,
+				 uint64_t flags)
+{
+	unsigned int i, ndw, nptes;
+	uint64_t *pte;
+	int r;
+
+	do {
+		ndw = p->num_dw_left;
+		ndw -= p->job->ibs->length_dw;
+
+		if (ndw < 32) {
+			r = amdgpu_vm_sdma_commit(p, NULL);
+			if (r)
+				return r;
+
+			/* estimate how many dw we need */
+			ndw = 32;
+			if (p->pages_addr)
+				ndw += count * 2;
+			ndw = max(ndw, AMDGPU_VM_SDMA_MIN_NUM_DW);
+			ndw = min(ndw, AMDGPU_VM_SDMA_MAX_NUM_DW);
+
+			r = amdgpu_job_alloc_with_ib(p->adev, ndw * 4, &p->job);
+			if (r)
+				return r;
+
+			p->num_dw_left = ndw;
+		}
+
+		if (!p->pages_addr) {
+			/* set page commands needed */
+			if (bo->shadow)
+				amdgpu_vm_sdma_set_ptes(p, bo->shadow, pe, addr,
+							count, incr, flags);
+			amdgpu_vm_sdma_set_ptes(p, bo, pe, addr, count,
+						incr, flags);
+			return 0;
+		}
+
+		/* copy commands needed */
+		ndw -= p->adev->vm_manager.vm_pte_funcs->copy_pte_num_dw *
+			(bo->shadow ? 2 : 1);
+
+		/* for padding */
+		ndw -= 7;
+
+		nptes = min(count, ndw / 2);
+
+		/* Put the PTEs at the end of the IB. */
+		p->num_dw_left -= nptes * 2;
+		pte = (uint64_t *)&(p->job->ibs->ptr[p->num_dw_left]);
+		for (i = 0; i < nptes; ++i, addr += incr) {
+			pte[i] = amdgpu_vm_map_gart(p->pages_addr, addr);
+			pte[i] |= flags;
+		}
+
+		if (bo->shadow)
+			amdgpu_vm_sdma_copy_ptes(p, bo->shadow, pe, nptes);
+		amdgpu_vm_sdma_copy_ptes(p, bo, pe, nptes);
+
+		pe += nptes * 8;
+		count -= nptes;
+	} while (count);
+
+	return 0;
+}
+
+const struct amdgpu_vm_update_funcs amdgpu_vm_sdma_funcs = {
+	.map_table = amdgpu_vm_sdma_map_table,
+	.prepare = amdgpu_vm_sdma_prepare,
+	.update = amdgpu_vm_sdma_update,
+	.commit = amdgpu_vm_sdma_commit
+};
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
index 3f9d5d00c9b3..ec9ea3fdbb4a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
@@ -33,6 +33,85 @@ struct amdgpu_vram_mgr {
 };
 
 /**
+ * DOC: mem_info_vram_total
+ *
+ * The amdgpu driver provides a sysfs API for reporting current total VRAM
+ * available on the device
+ * The file mem_info_vram_total is used for this and returns the total
+ * amount of VRAM in bytes
+ */
+static ssize_t amdgpu_mem_info_vram_total_show(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = ddev->dev_private;
+
+	return snprintf(buf, PAGE_SIZE, "%llu\n", adev->gmc.real_vram_size);
+}
+
+/**
+ * DOC: mem_info_vis_vram_total
+ *
+ * The amdgpu driver provides a sysfs API for reporting current total
+ * visible VRAM available on the device
+ * The file mem_info_vis_vram_total is used for this and returns the total
+ * amount of visible VRAM in bytes
+ */
+static ssize_t amdgpu_mem_info_vis_vram_total_show(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = ddev->dev_private;
+
+	return snprintf(buf, PAGE_SIZE, "%llu\n", adev->gmc.visible_vram_size);
+}
+
+/**
+ * DOC: mem_info_vram_used
+ *
+ * The amdgpu driver provides a sysfs API for reporting current total VRAM
+ * available on the device
+ * The file mem_info_vram_used is used for this and returns the total
+ * amount of currently used VRAM in bytes
+ */
+static ssize_t amdgpu_mem_info_vram_used_show(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = ddev->dev_private;
+
+	return snprintf(buf, PAGE_SIZE, "%llu\n",
+		amdgpu_vram_mgr_usage(&adev->mman.bdev.man[TTM_PL_VRAM]));
+}
+
+/**
+ * DOC: mem_info_vis_vram_used
+ *
+ * The amdgpu driver provides a sysfs API for reporting current total of
+ * used visible VRAM
+ * The file mem_info_vis_vram_used is used for this and returns the total
+ * amount of currently used visible VRAM in bytes
+ */
+static ssize_t amdgpu_mem_info_vis_vram_used_show(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = ddev->dev_private;
+
+	return snprintf(buf, PAGE_SIZE, "%llu\n",
+		amdgpu_vram_mgr_vis_usage(&adev->mman.bdev.man[TTM_PL_VRAM]));
+}
+
+static DEVICE_ATTR(mem_info_vram_total, S_IRUGO,
+		   amdgpu_mem_info_vram_total_show, NULL);
+static DEVICE_ATTR(mem_info_vis_vram_total, S_IRUGO,
+		   amdgpu_mem_info_vis_vram_total_show,NULL);
+static DEVICE_ATTR(mem_info_vram_used, S_IRUGO,
+		   amdgpu_mem_info_vram_used_show, NULL);
+static DEVICE_ATTR(mem_info_vis_vram_used, S_IRUGO,
+		   amdgpu_mem_info_vis_vram_used_show, NULL);
+
+/**
  * amdgpu_vram_mgr_init - init VRAM manager and DRM MM
  *
  * @man: TTM memory type manager
@@ -43,7 +122,9 @@ struct amdgpu_vram_mgr {
 static int amdgpu_vram_mgr_init(struct ttm_mem_type_manager *man,
 				unsigned long p_size)
 {
+	struct amdgpu_device *adev = amdgpu_ttm_adev(man->bdev);
 	struct amdgpu_vram_mgr *mgr;
+	int ret;
 
 	mgr = kzalloc(sizeof(*mgr), GFP_KERNEL);
 	if (!mgr)
@@ -52,6 +133,29 @@ static int amdgpu_vram_mgr_init(struct ttm_mem_type_manager *man,
 	drm_mm_init(&mgr->mm, 0, p_size);
 	spin_lock_init(&mgr->lock);
 	man->priv = mgr;
+
+	/* Add the two VRAM-related sysfs files */
+	ret = device_create_file(adev->dev, &dev_attr_mem_info_vram_total);
+	if (ret) {
+		DRM_ERROR("Failed to create device file mem_info_vram_total\n");
+		return ret;
+	}
+	ret = device_create_file(adev->dev, &dev_attr_mem_info_vis_vram_total);
+	if (ret) {
+		DRM_ERROR("Failed to create device file mem_info_vis_vram_total\n");
+		return ret;
+	}
+	ret = device_create_file(adev->dev, &dev_attr_mem_info_vram_used);
+	if (ret) {
+		DRM_ERROR("Failed to create device file mem_info_vram_used\n");
+		return ret;
+	}
+	ret = device_create_file(adev->dev, &dev_attr_mem_info_vis_vram_used);
+	if (ret) {
+		DRM_ERROR("Failed to create device file mem_info_vis_vram_used\n");
+		return ret;
+	}
+
 	return 0;
 }
 
@@ -65,6 +169,7 @@ static int amdgpu_vram_mgr_init(struct ttm_mem_type_manager *man,
  */
 static int amdgpu_vram_mgr_fini(struct ttm_mem_type_manager *man)
 {
+	struct amdgpu_device *adev = amdgpu_ttm_adev(man->bdev);
 	struct amdgpu_vram_mgr *mgr = man->priv;
 
 	spin_lock(&mgr->lock);
@@ -72,6 +177,10 @@ static int amdgpu_vram_mgr_fini(struct ttm_mem_type_manager *man)
 	spin_unlock(&mgr->lock);
 	kfree(mgr);
 	man->priv = NULL;
+	device_remove_file(adev->dev, &dev_attr_mem_info_vram_total);
+	device_remove_file(adev->dev, &dev_attr_mem_info_vis_vram_total);
+	device_remove_file(adev->dev, &dev_attr_mem_info_vram_used);
+	device_remove_file(adev->dev, &dev_attr_mem_info_vis_vram_used);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
index 407dd16cc35c..a48c84c51775 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
@@ -24,6 +24,7 @@
 #include <linux/list.h>
 #include "amdgpu.h"
 #include "amdgpu_xgmi.h"
+#include "amdgpu_smu.h"
 
 
 static DEFINE_MUTEX(xgmi_mutex);
@@ -34,12 +35,132 @@ static DEFINE_MUTEX(xgmi_mutex);
 static struct amdgpu_hive_info xgmi_hives[AMDGPU_MAX_XGMI_HIVE];
 static unsigned hive_count = 0;
 
-
 void *amdgpu_xgmi_hive_try_lock(struct amdgpu_hive_info *hive)
 {
 	return &hive->device_list;
 }
 
+static ssize_t amdgpu_xgmi_show_hive_id(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	struct amdgpu_hive_info *hive =
+			container_of(attr, struct amdgpu_hive_info, dev_attr);
+
+	return snprintf(buf, PAGE_SIZE, "%llu\n", hive->hive_id);
+}
+
+static int amdgpu_xgmi_sysfs_create(struct amdgpu_device *adev,
+				    struct amdgpu_hive_info *hive)
+{
+	int ret = 0;
+
+	if (WARN_ON(hive->kobj))
+		return -EINVAL;
+
+	hive->kobj = kobject_create_and_add("xgmi_hive_info", &adev->dev->kobj);
+	if (!hive->kobj) {
+		dev_err(adev->dev, "XGMI: Failed to allocate sysfs entry!\n");
+		return -EINVAL;
+	}
+
+	hive->dev_attr = (struct device_attribute) {
+		.attr = {
+			.name = "xgmi_hive_id",
+			.mode = S_IRUGO,
+
+		},
+		.show = amdgpu_xgmi_show_hive_id,
+	};
+
+	ret = sysfs_create_file(hive->kobj, &hive->dev_attr.attr);
+	if (ret) {
+		dev_err(adev->dev, "XGMI: Failed to create device file xgmi_hive_id\n");
+		kobject_del(hive->kobj);
+		kobject_put(hive->kobj);
+		hive->kobj = NULL;
+	}
+
+	return ret;
+}
+
+static void amdgpu_xgmi_sysfs_destroy(struct amdgpu_device *adev,
+				    struct amdgpu_hive_info *hive)
+{
+	sysfs_remove_file(hive->kobj, &hive->dev_attr.attr);
+	kobject_del(hive->kobj);
+	kobject_put(hive->kobj);
+	hive->kobj = NULL;
+}
+
+static ssize_t amdgpu_xgmi_show_device_id(struct device *dev,
+				     struct device_attribute *attr,
+				     char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = ddev->dev_private;
+
+	return snprintf(buf, PAGE_SIZE, "%llu\n", adev->gmc.xgmi.node_id);
+
+}
+
+
+static DEVICE_ATTR(xgmi_device_id, S_IRUGO, amdgpu_xgmi_show_device_id, NULL);
+
+
+static int amdgpu_xgmi_sysfs_add_dev_info(struct amdgpu_device *adev,
+					 struct amdgpu_hive_info *hive)
+{
+	int ret = 0;
+	char node[10] = { 0 };
+
+	/* Create xgmi device id file */
+	ret = device_create_file(adev->dev, &dev_attr_xgmi_device_id);
+	if (ret) {
+		dev_err(adev->dev, "XGMI: Failed to create device file xgmi_device_id\n");
+		return ret;
+	}
+
+	/* Create sysfs link to hive info folder on the first device */
+	if (adev != hive->adev) {
+		ret = sysfs_create_link(&adev->dev->kobj, hive->kobj,
+					"xgmi_hive_info");
+		if (ret) {
+			dev_err(adev->dev, "XGMI: Failed to create link to hive info");
+			goto remove_file;
+		}
+	}
+
+	sprintf(node, "node%d", hive->number_devices);
+	/* Create sysfs link form the hive folder to yourself */
+	ret = sysfs_create_link(hive->kobj, &adev->dev->kobj, node);
+	if (ret) {
+		dev_err(adev->dev, "XGMI: Failed to create link from hive info");
+		goto remove_link;
+	}
+
+	goto success;
+
+
+remove_link:
+	sysfs_remove_link(&adev->dev->kobj, adev->ddev->unique);
+
+remove_file:
+	device_remove_file(adev->dev, &dev_attr_xgmi_device_id);
+
+success:
+	return ret;
+}
+
+static void amdgpu_xgmi_sysfs_rem_dev_info(struct amdgpu_device *adev,
+					  struct amdgpu_hive_info *hive)
+{
+	device_remove_file(adev->dev, &dev_attr_xgmi_device_id);
+	sysfs_remove_link(&adev->dev->kobj, adev->ddev->unique);
+	sysfs_remove_link(hive->kobj, adev->ddev->unique);
+}
+
+
+
 struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev, int lock)
 {
 	int i;
@@ -66,18 +187,50 @@ struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev, int lo
 
 	/* initialize new hive if not exist */
 	tmp = &xgmi_hives[hive_count++];
+
+	if (amdgpu_xgmi_sysfs_create(adev, tmp)) {
+		mutex_unlock(&xgmi_mutex);
+		return NULL;
+	}
+
+	tmp->adev = adev;
 	tmp->hive_id = adev->gmc.xgmi.hive_id;
 	INIT_LIST_HEAD(&tmp->device_list);
 	mutex_init(&tmp->hive_lock);
 	mutex_init(&tmp->reset_lock);
+
 	if (lock)
 		mutex_lock(&tmp->hive_lock);
-
+	tmp->pstate = -1;
 	mutex_unlock(&xgmi_mutex);
 
 	return tmp;
 }
 
+int amdgpu_xgmi_set_pstate(struct amdgpu_device *adev, int pstate)
+{
+	int ret = 0;
+	struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev, 0);
+
+	if (!hive)
+		return 0;
+
+	if (hive->pstate == pstate)
+		return 0;
+
+	dev_dbg(adev->dev, "Set xgmi pstate %d.\n", pstate);
+
+	if (is_support_sw_smu(adev))
+		ret = smu_set_xgmi_pstate(&adev->smu, pstate);
+	if (ret)
+		dev_err(adev->dev,
+			"XGMI: Set pstate failure on device %llx, hive %llx, ret %d",
+			adev->gmc.xgmi.node_id,
+			adev->gmc.xgmi.hive_id, ret);
+
+	return ret;
+}
+
 int amdgpu_xgmi_update_topology(struct amdgpu_hive_info *hive, struct amdgpu_device *adev)
 {
 	int ret = -EINVAL;
@@ -156,8 +309,17 @@ int amdgpu_xgmi_add_device(struct amdgpu_device *adev)
 			break;
 	}
 
-	dev_info(adev->dev, "XGMI: Add node %d, hive 0x%llx.\n",
-		 adev->gmc.xgmi.physical_node_id, adev->gmc.xgmi.hive_id);
+	if (!ret)
+		ret = amdgpu_xgmi_sysfs_add_dev_info(adev, hive);
+
+	if (!ret)
+		dev_info(adev->dev, "XGMI: Add node %d, hive 0x%llx.\n",
+			 adev->gmc.xgmi.physical_node_id, adev->gmc.xgmi.hive_id);
+	else
+		dev_err(adev->dev, "XGMI: Failed to add node %d, hive 0x%llx ret: %d\n",
+			adev->gmc.xgmi.physical_node_id, adev->gmc.xgmi.hive_id,
+			ret);
+
 
 	mutex_unlock(&hive->hive_lock);
 exit:
@@ -176,9 +338,11 @@ void amdgpu_xgmi_remove_device(struct amdgpu_device *adev)
 		return;
 
 	if (!(hive->number_devices--)) {
+		amdgpu_xgmi_sysfs_destroy(adev, hive);
 		mutex_destroy(&hive->hive_lock);
 		mutex_destroy(&hive->reset_lock);
 	} else {
+		amdgpu_xgmi_sysfs_rem_dev_info(adev, hive);
 		mutex_unlock(&hive->hive_lock);
 	}
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h
index 14bc60664159..3e9c91e9a4bf 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h
@@ -29,13 +29,25 @@ struct amdgpu_hive_info {
 	struct list_head	device_list;
 	struct psp_xgmi_topology_info	topology_info;
 	int number_devices;
-	struct mutex hive_lock,
-		     reset_lock;
+	struct mutex hive_lock, reset_lock;
+	struct kobject *kobj;
+	struct device_attribute dev_attr;
+	struct amdgpu_device *adev;
+	int pstate; /*0 -- low , 1 -- high , -1 unknown*/
 };
 
 struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev, int lock);
 int amdgpu_xgmi_update_topology(struct amdgpu_hive_info *hive, struct amdgpu_device *adev);
 int amdgpu_xgmi_add_device(struct amdgpu_device *adev);
 void amdgpu_xgmi_remove_device(struct amdgpu_device *adev);
+int amdgpu_xgmi_set_pstate(struct amdgpu_device *adev, int pstate);
+
+static inline bool amdgpu_xgmi_same_hive(struct amdgpu_device *adev,
+		struct amdgpu_device *bo_adev)
+{
+	return (adev != bo_adev &&
+		adev->gmc.xgmi.hive_id &&
+		adev->gmc.xgmi.hive_id == bo_adev->gmc.xgmi.hive_id);
+}
 
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
index 189599b694e8..d42808b05971 100644
--- a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
@@ -977,8 +977,8 @@ static int cik_sdma_sw_init(void *handle)
 		r = amdgpu_ring_init(adev, ring, 1024,
 				     &adev->sdma.trap_irq,
 				     (i == 0) ?
-				     AMDGPU_SDMA_IRQ_TRAP0 :
-				     AMDGPU_SDMA_IRQ_TRAP1);
+				     AMDGPU_SDMA_IRQ_INSTANCE0 :
+				     AMDGPU_SDMA_IRQ_INSTANCE1);
 		if (r)
 			return r;
 	}
@@ -1114,7 +1114,7 @@ static int cik_sdma_set_trap_irq_state(struct amdgpu_device *adev,
 	u32 sdma_cntl;
 
 	switch (type) {
-	case AMDGPU_SDMA_IRQ_TRAP0:
+	case AMDGPU_SDMA_IRQ_INSTANCE0:
 		switch (state) {
 		case AMDGPU_IRQ_STATE_DISABLE:
 			sdma_cntl = RREG32(mmSDMA0_CNTL + SDMA0_REGISTER_OFFSET);
@@ -1130,7 +1130,7 @@ static int cik_sdma_set_trap_irq_state(struct amdgpu_device *adev,
 			break;
 		}
 		break;
-	case AMDGPU_SDMA_IRQ_TRAP1:
+	case AMDGPU_SDMA_IRQ_INSTANCE1:
 		switch (state) {
 		case AMDGPU_IRQ_STATE_DISABLE:
 			sdma_cntl = RREG32(mmSDMA0_CNTL + SDMA1_REGISTER_OFFSET);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
index 305276c7e4bf..c0cb244f58cd 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
@@ -782,6 +782,25 @@ static void gfx_v6_0_tiling_mode_table_init(struct amdgpu_device *adev)
 				BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
 				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
 				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2);
+		tilemode[18] =  MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+				ARRAY_MODE(ARRAY_1D_TILED_THICK) |
+				PIPE_CONFIG(ADDR_SURF_P4_8x16);
+		tilemode[19] =  MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+				ARRAY_MODE(ARRAY_2D_TILED_XTHICK) |
+				PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+				BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+				NUM_BANKS(ADDR_SURF_16_BANK) |
+				TILE_SPLIT(split_equal_to_row_size);
+		tilemode[20] =  MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+				ARRAY_MODE(ARRAY_2D_TILED_THICK) |
+				PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+				BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+				NUM_BANKS(ADDR_SURF_16_BANK) |
+				TILE_SPLIT(split_equal_to_row_size);
 		tilemode[21] =  MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
 				ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
 				PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index b8e50a34bdb3..02955e6e9dd9 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -3236,6 +3236,7 @@ static void gfx_v8_0_tiling_mode_table_init(struct amdgpu_device *adev)
 		dev_warn(adev->dev,
 			 "Unknown chip type (%d) in function gfx_v8_0_tiling_mode_table_init() falling through to CHIP_CARRIZO\n",
 			 adev->asic_type);
+		/* fall through */
 
 	case CHIP_CARRIZO:
 		modearray[0] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index a11db2b1a63f..ba67d1023264 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -40,6 +40,8 @@
 
 #include "ivsrcid/gfx/irqsrcs_gfx_9_0.h"
 
+#include "amdgpu_ras.h"
+
 #define GFX9_NUM_GFX_RINGS     1
 #define GFX9_MEC_HPD_SIZE 4096
 #define RLCG_UCODE_LOADING_START_ADDRESS 0x00002000L
@@ -576,6 +578,27 @@ static void gfx_v9_0_check_fw_write_wait(struct amdgpu_device *adev)
 	}
 }
 
+static void gfx_v9_0_check_if_need_gfxoff(struct amdgpu_device *adev)
+{
+	switch (adev->asic_type) {
+	case CHIP_VEGA10:
+	case CHIP_VEGA12:
+	case CHIP_VEGA20:
+		break;
+	case CHIP_RAVEN:
+		if (adev->rev_id >= 0x8 || adev->pdev->device == 0x15d8)
+			break;
+		if ((adev->gfx.rlc_fw_version < 531) ||
+		    (adev->gfx.rlc_fw_version == 53815) ||
+		    (adev->gfx.rlc_feature_version < 1) ||
+		    !adev->gfx.rlc.is_rlc_v2_1)
+			adev->pm.pp_feature &= ~PP_GFXOFF_MASK;
+		break;
+	default:
+		break;
+	}
+}
+
 static int gfx_v9_0_init_microcode(struct amdgpu_device *adev)
 {
 	const char *chip_name;
@@ -828,6 +851,7 @@ static int gfx_v9_0_init_microcode(struct amdgpu_device *adev)
 	}
 
 out:
+	gfx_v9_0_check_if_need_gfxoff(adev);
 	gfx_v9_0_check_fw_write_wait(adev);
 	if (err) {
 		dev_err(adev->dev,
@@ -1639,6 +1663,18 @@ static int gfx_v9_0_sw_init(void *handle)
 	if (r)
 		return r;
 
+	/* ECC error */
+	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_GRBM_CP, GFX_9_0__SRCID__CP_ECC_ERROR,
+			      &adev->gfx.cp_ecc_error_irq);
+	if (r)
+		return r;
+
+	/* FUE error */
+	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_GRBM_CP, GFX_9_0__SRCID__CP_FUE_ERROR,
+			      &adev->gfx.cp_ecc_error_irq);
+	if (r)
+		return r;
+
 	adev->gfx.gfx_current_status = AMDGPU_GFX_NORMAL_MODE;
 
 	gfx_v9_0_scratch_init(adev);
@@ -1731,6 +1767,20 @@ static int gfx_v9_0_sw_fini(void *handle)
 	int i;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX) &&
+			adev->gfx.ras_if) {
+		struct ras_common_if *ras_if = adev->gfx.ras_if;
+		struct ras_ih_if ih_info = {
+			.head = *ras_if,
+		};
+
+		amdgpu_ras_debugfs_remove(adev, ras_if);
+		amdgpu_ras_sysfs_remove(adev, ras_if);
+		amdgpu_ras_interrupt_remove_handler(adev,  &ih_info);
+		amdgpu_ras_feature_enable(adev, ras_if, 0);
+		kfree(ras_if);
+	}
+
 	amdgpu_bo_free_kernel(&adev->gds.oa_gfx_bo, NULL, NULL);
 	amdgpu_bo_free_kernel(&adev->gds.gws_gfx_bo, NULL, NULL);
 	amdgpu_bo_free_kernel(&adev->gds.gds_gfx_bo, NULL, NULL);
@@ -3303,6 +3353,7 @@ static int gfx_v9_0_hw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	amdgpu_irq_put(adev, &adev->gfx.cp_ecc_error_irq, 0);
 	amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0);
 	amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0);
 
@@ -3492,6 +3543,80 @@ static int gfx_v9_0_early_init(void *handle)
 	return 0;
 }
 
+static int gfx_v9_0_process_ras_data_cb(struct amdgpu_device *adev,
+		struct amdgpu_iv_entry *entry);
+
+static int gfx_v9_0_ecc_late_init(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	struct ras_common_if **ras_if = &adev->gfx.ras_if;
+	struct ras_ih_if ih_info = {
+		.cb = gfx_v9_0_process_ras_data_cb,
+	};
+	struct ras_fs_if fs_info = {
+		.sysfs_name = "gfx_err_count",
+		.debugfs_name = "gfx_err_inject",
+	};
+	struct ras_common_if ras_block = {
+		.block = AMDGPU_RAS_BLOCK__GFX,
+		.type = AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE,
+		.sub_block_index = 0,
+		.name = "gfx",
+	};
+	int r;
+
+	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX)) {
+		amdgpu_ras_feature_enable_on_boot(adev, &ras_block, 0);
+		return 0;
+	}
+
+	if (*ras_if)
+		goto resume;
+
+	*ras_if = kmalloc(sizeof(**ras_if), GFP_KERNEL);
+	if (!*ras_if)
+		return -ENOMEM;
+
+	**ras_if = ras_block;
+
+	r = amdgpu_ras_feature_enable_on_boot(adev, *ras_if, 1);
+	if (r)
+		goto feature;
+
+	ih_info.head = **ras_if;
+	fs_info.head = **ras_if;
+
+	r = amdgpu_ras_interrupt_add_handler(adev, &ih_info);
+	if (r)
+		goto interrupt;
+
+	r = amdgpu_ras_debugfs_create(adev, &fs_info);
+	if (r)
+		goto debugfs;
+
+	r = amdgpu_ras_sysfs_create(adev, &fs_info);
+	if (r)
+		goto sysfs;
+resume:
+	r = amdgpu_irq_get(adev, &adev->gfx.cp_ecc_error_irq, 0);
+	if (r)
+		goto irq;
+
+	return 0;
+irq:
+	amdgpu_ras_sysfs_remove(adev, *ras_if);
+sysfs:
+	amdgpu_ras_debugfs_remove(adev, *ras_if);
+debugfs:
+	amdgpu_ras_interrupt_remove_handler(adev, &ih_info);
+interrupt:
+	amdgpu_ras_feature_enable(adev, *ras_if, 0);
+feature:
+	kfree(*ras_if);
+	*ras_if = NULL;
+	return -EINVAL;
+}
+
 static int gfx_v9_0_late_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
@@ -3505,6 +3630,10 @@ static int gfx_v9_0_late_init(void *handle)
 	if (r)
 		return r;
 
+	r = gfx_v9_0_ecc_late_init(handle);
+	if (r)
+		return r;
+
 	return 0;
 }
 
@@ -4541,6 +4670,45 @@ static int gfx_v9_0_set_priv_inst_fault_state(struct amdgpu_device *adev,
 	return 0;
 }
 
+#define ENABLE_ECC_ON_ME_PIPE(me, pipe)				\
+	WREG32_FIELD15(GC, 0, CP_ME##me##_PIPE##pipe##_INT_CNTL,\
+			CP_ECC_ERROR_INT_ENABLE, 1)
+
+#define DISABLE_ECC_ON_ME_PIPE(me, pipe)			\
+	WREG32_FIELD15(GC, 0, CP_ME##me##_PIPE##pipe##_INT_CNTL,\
+			CP_ECC_ERROR_INT_ENABLE, 0)
+
+static int gfx_v9_0_set_cp_ecc_error_state(struct amdgpu_device *adev,
+					      struct amdgpu_irq_src *source,
+					      unsigned type,
+					      enum amdgpu_interrupt_state state)
+{
+	switch (state) {
+	case AMDGPU_IRQ_STATE_DISABLE:
+		WREG32_FIELD15(GC, 0, CP_INT_CNTL_RING0,
+				CP_ECC_ERROR_INT_ENABLE, 0);
+		DISABLE_ECC_ON_ME_PIPE(1, 0);
+		DISABLE_ECC_ON_ME_PIPE(1, 1);
+		DISABLE_ECC_ON_ME_PIPE(1, 2);
+		DISABLE_ECC_ON_ME_PIPE(1, 3);
+		break;
+
+	case AMDGPU_IRQ_STATE_ENABLE:
+		WREG32_FIELD15(GC, 0, CP_INT_CNTL_RING0,
+				CP_ECC_ERROR_INT_ENABLE, 1);
+		ENABLE_ECC_ON_ME_PIPE(1, 0);
+		ENABLE_ECC_ON_ME_PIPE(1, 1);
+		ENABLE_ECC_ON_ME_PIPE(1, 2);
+		ENABLE_ECC_ON_ME_PIPE(1, 3);
+		break;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+
 static int gfx_v9_0_set_eop_interrupt_state(struct amdgpu_device *adev,
 					    struct amdgpu_irq_src *src,
 					    unsigned type,
@@ -4657,6 +4825,34 @@ static int gfx_v9_0_priv_inst_irq(struct amdgpu_device *adev,
 	return 0;
 }
 
+static int gfx_v9_0_process_ras_data_cb(struct amdgpu_device *adev,
+		struct amdgpu_iv_entry *entry)
+{
+	/* TODO ue will trigger an interrupt. */
+	kgd2kfd_set_sram_ecc_flag(adev->kfd.dev);
+	amdgpu_ras_reset_gpu(adev, 0);
+	return AMDGPU_RAS_UE;
+}
+
+static int gfx_v9_0_cp_ecc_error_irq(struct amdgpu_device *adev,
+				  struct amdgpu_irq_src *source,
+				  struct amdgpu_iv_entry *entry)
+{
+	struct ras_common_if *ras_if = adev->gfx.ras_if;
+	struct ras_dispatch_if ih_data = {
+		.entry = entry,
+	};
+
+	if (!ras_if)
+		return 0;
+
+	ih_data.head = *ras_if;
+
+	DRM_ERROR("CP ECC ERROR IRQ\n");
+	amdgpu_ras_interrupt_dispatch(adev, &ih_data);
+	return 0;
+}
+
 static const struct amd_ip_funcs gfx_v9_0_ip_funcs = {
 	.name = "gfx_v9_0",
 	.early_init = gfx_v9_0_early_init,
@@ -4818,6 +5014,12 @@ static const struct amdgpu_irq_src_funcs gfx_v9_0_priv_inst_irq_funcs = {
 	.process = gfx_v9_0_priv_inst_irq,
 };
 
+static const struct amdgpu_irq_src_funcs gfx_v9_0_cp_ecc_error_irq_funcs = {
+	.set = gfx_v9_0_set_cp_ecc_error_state,
+	.process = gfx_v9_0_cp_ecc_error_irq,
+};
+
+
 static void gfx_v9_0_set_irq_funcs(struct amdgpu_device *adev)
 {
 	adev->gfx.eop_irq.num_types = AMDGPU_CP_IRQ_LAST;
@@ -4828,6 +5030,9 @@ static void gfx_v9_0_set_irq_funcs(struct amdgpu_device *adev)
 
 	adev->gfx.priv_inst_irq.num_types = 1;
 	adev->gfx.priv_inst_irq.funcs = &gfx_v9_0_priv_inst_irq_funcs;
+
+	adev->gfx.cp_ecc_error_irq.num_types = 2; /*C5 ECC error and C9 FUE error*/
+	adev->gfx.cp_ecc_error_irq.funcs = &gfx_v9_0_cp_ecc_error_irq_funcs;
 }
 
 static void gfx_v9_0_set_rlc_funcs(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c
index f5edddf3b29d..7bb5359d0bbd 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c
@@ -143,7 +143,7 @@ static void gfxhub_v1_0_init_cache_regs(struct amdgpu_device *adev)
 	/* XXX for emulation, Refer to closed source code.*/
 	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, L2_PDE0_CACHE_TAG_GENERATION_MODE,
 			    0);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, PDE_FAULT_CLASSIFICATION, 1);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, PDE_FAULT_CLASSIFICATION, 0);
 	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, CONTEXT1_IDENTITY_ACCESS_MODE, 1);
 	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, IDENTITY_MODE_FRAGMENT_SIZE, 0);
 	WREG32_SOC15(GC, 0, mmVM_L2_CNTL, tmp);
@@ -236,7 +236,7 @@ static void gfxhub_v1_0_setup_vmid_config(struct amdgpu_device *adev)
 				    block_size);
 		/* Send no-retry XNACK on fault to suppress VM fault storm. */
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 0);
+				    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 1);
 		WREG32_SOC15_OFFSET(GC, 0, mmVM_CONTEXT1_CNTL, i, tmp);
 		WREG32_SOC15_OFFSET(GC, 0, mmVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32, i*2, 0);
 		WREG32_SOC15_OFFSET(GC, 0, mmVM_CONTEXT1_PAGE_TABLE_START_ADDR_HI32, i*2, 0);
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
index 9fc3296592fe..b06d876da2d9 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
@@ -225,7 +225,7 @@ static void gmc_v6_0_vram_gtt_location(struct amdgpu_device *adev,
 	u64 base = RREG32(mmMC_VM_FB_LOCATION) & 0xFFFF;
 	base <<= 24;
 
-	amdgpu_gmc_vram_location(adev, &adev->gmc, base);
+	amdgpu_gmc_vram_location(adev, mc, base);
 	amdgpu_gmc_gart_location(adev, mc);
 }
 
@@ -383,20 +383,6 @@ static uint64_t gmc_v6_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring,
 	return pd_addr;
 }
 
-static int gmc_v6_0_set_pte_pde(struct amdgpu_device *adev, void *cpu_pt_addr,
-				uint32_t gpu_page_idx, uint64_t addr,
-				uint64_t flags)
-{
-	void __iomem *ptr = (void *)cpu_pt_addr;
-	uint64_t value;
-
-	value = addr & 0xFFFFFFFFFFFFF000ULL;
-	value |= flags;
-	writeq(value, ptr + (gpu_page_idx * 8));
-
-	return 0;
-}
-
 static uint64_t gmc_v6_0_get_vm_pte_flags(struct amdgpu_device *adev,
 					  uint32_t flags)
 {
@@ -886,7 +872,7 @@ static int gmc_v6_0_sw_init(void *handle)
 		pci_set_consistent_dma_mask(adev->pdev, DMA_BIT_MASK(32));
 		dev_warn(adev->dev, "amdgpu: No coherent DMA available.\n");
 	}
-	adev->need_swiotlb = drm_get_max_iomem() > ((u64)1 << dma_bits);
+	adev->need_swiotlb = drm_need_swiotlb(dma_bits);
 
 	r = gmc_v6_0_init_microcode(adev);
 	if (r) {
@@ -1169,7 +1155,6 @@ static const struct amd_ip_funcs gmc_v6_0_ip_funcs = {
 static const struct amdgpu_gmc_funcs gmc_v6_0_gmc_funcs = {
 	.flush_gpu_tlb = gmc_v6_0_flush_gpu_tlb,
 	.emit_flush_gpu_tlb = gmc_v6_0_emit_flush_gpu_tlb,
-	.set_pte_pde = gmc_v6_0_set_pte_pde,
 	.set_prt = gmc_v6_0_set_prt,
 	.get_vm_pde = gmc_v6_0_get_vm_pde,
 	.get_vm_pte_flags = gmc_v6_0_get_vm_pte_flags
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
index 761dcfb2fec0..75aa3332aee2 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
@@ -242,7 +242,7 @@ static void gmc_v7_0_vram_gtt_location(struct amdgpu_device *adev,
 	u64 base = RREG32(mmMC_VM_FB_LOCATION) & 0xFFFF;
 	base <<= 24;
 
-	amdgpu_gmc_vram_location(adev, &adev->gmc, base);
+	amdgpu_gmc_vram_location(adev, mc, base);
 	amdgpu_gmc_gart_location(adev, mc);
 }
 
@@ -460,31 +460,6 @@ static void gmc_v7_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned vmid,
 	amdgpu_ring_emit_wreg(ring, mmIH_VMID_0_LUT + vmid, pasid);
 }
 
-/**
- * gmc_v7_0_set_pte_pde - update the page tables using MMIO
- *
- * @adev: amdgpu_device pointer
- * @cpu_pt_addr: cpu address of the page table
- * @gpu_page_idx: entry in the page table to update
- * @addr: dst addr to write into pte/pde
- * @flags: access flags
- *
- * Update the page tables using the CPU.
- */
-static int gmc_v7_0_set_pte_pde(struct amdgpu_device *adev, void *cpu_pt_addr,
-				 uint32_t gpu_page_idx, uint64_t addr,
-				 uint64_t flags)
-{
-	void __iomem *ptr = (void *)cpu_pt_addr;
-	uint64_t value;
-
-	value = addr & 0xFFFFFFFFFFFFF000ULL;
-	value |= flags;
-	writeq(value, ptr + (gpu_page_idx * 8));
-
-	return 0;
-}
-
 static uint64_t gmc_v7_0_get_vm_pte_flags(struct amdgpu_device *adev,
 					  uint32_t flags)
 {
@@ -1030,7 +1005,7 @@ static int gmc_v7_0_sw_init(void *handle)
 		pci_set_consistent_dma_mask(adev->pdev, DMA_BIT_MASK(32));
 		pr_warn("amdgpu: No coherent DMA available\n");
 	}
-	adev->need_swiotlb = drm_get_max_iomem() > ((u64)1 << dma_bits);
+	adev->need_swiotlb = drm_need_swiotlb(dma_bits);
 
 	r = gmc_v7_0_init_microcode(adev);
 	if (r) {
@@ -1376,7 +1351,6 @@ static const struct amdgpu_gmc_funcs gmc_v7_0_gmc_funcs = {
 	.flush_gpu_tlb = gmc_v7_0_flush_gpu_tlb,
 	.emit_flush_gpu_tlb = gmc_v7_0_emit_flush_gpu_tlb,
 	.emit_pasid_mapping = gmc_v7_0_emit_pasid_mapping,
-	.set_pte_pde = gmc_v7_0_set_pte_pde,
 	.set_prt = gmc_v7_0_set_prt,
 	.get_vm_pte_flags = gmc_v7_0_get_vm_pte_flags,
 	.get_vm_pde = gmc_v7_0_get_vm_pde
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
index 34440672f938..8a3b5e6fc6c9 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
@@ -433,7 +433,7 @@ static void gmc_v8_0_vram_gtt_location(struct amdgpu_device *adev,
 		base = RREG32(mmMC_VM_FB_LOCATION) & 0xFFFF;
 	base <<= 24;
 
-	amdgpu_gmc_vram_location(adev, &adev->gmc, base);
+	amdgpu_gmc_vram_location(adev, mc, base);
 	amdgpu_gmc_gart_location(adev, mc);
 }
 
@@ -662,50 +662,26 @@ static void gmc_v8_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned vmid,
 	amdgpu_ring_emit_wreg(ring, mmIH_VMID_0_LUT + vmid, pasid);
 }
 
-/**
- * gmc_v8_0_set_pte_pde - update the page tables using MMIO
- *
- * @adev: amdgpu_device pointer
- * @cpu_pt_addr: cpu address of the page table
- * @gpu_page_idx: entry in the page table to update
- * @addr: dst addr to write into pte/pde
- * @flags: access flags
+/*
+ * PTE format on VI:
+ * 63:40 reserved
+ * 39:12 4k physical page base address
+ * 11:7 fragment
+ * 6 write
+ * 5 read
+ * 4 exe
+ * 3 reserved
+ * 2 snooped
+ * 1 system
+ * 0 valid
  *
- * Update the page tables using the CPU.
+ * PDE format on VI:
+ * 63:59 block fragment size
+ * 58:40 reserved
+ * 39:1 physical base address of PTE
+ * bits 5:1 must be 0.
+ * 0 valid
  */
-static int gmc_v8_0_set_pte_pde(struct amdgpu_device *adev, void *cpu_pt_addr,
-				uint32_t gpu_page_idx, uint64_t addr,
-				uint64_t flags)
-{
-	void __iomem *ptr = (void *)cpu_pt_addr;
-	uint64_t value;
-
-	/*
-	 * PTE format on VI:
-	 * 63:40 reserved
-	 * 39:12 4k physical page base address
-	 * 11:7 fragment
-	 * 6 write
-	 * 5 read
-	 * 4 exe
-	 * 3 reserved
-	 * 2 snooped
-	 * 1 system
-	 * 0 valid
-	 *
-	 * PDE format on VI:
-	 * 63:59 block fragment size
-	 * 58:40 reserved
-	 * 39:1 physical base address of PTE
-	 * bits 5:1 must be 0.
-	 * 0 valid
-	 */
-	value = addr & 0x000000FFFFFFF000ULL;
-	value |= flags;
-	writeq(value, ptr + (gpu_page_idx * 8));
-
-	return 0;
-}
 
 static uint64_t gmc_v8_0_get_vm_pte_flags(struct amdgpu_device *adev,
 					  uint32_t flags)
@@ -1155,7 +1131,7 @@ static int gmc_v8_0_sw_init(void *handle)
 		pci_set_consistent_dma_mask(adev->pdev, DMA_BIT_MASK(32));
 		pr_warn("amdgpu: No coherent DMA available\n");
 	}
-	adev->need_swiotlb = drm_get_max_iomem() > ((u64)1 << dma_bits);
+	adev->need_swiotlb = drm_need_swiotlb(dma_bits);
 
 	r = gmc_v8_0_init_microcode(adev);
 	if (r) {
@@ -1743,7 +1719,6 @@ static const struct amdgpu_gmc_funcs gmc_v8_0_gmc_funcs = {
 	.flush_gpu_tlb = gmc_v8_0_flush_gpu_tlb,
 	.emit_flush_gpu_tlb = gmc_v8_0_emit_flush_gpu_tlb,
 	.emit_pasid_mapping = gmc_v8_0_emit_pasid_mapping,
-	.set_pte_pde = gmc_v8_0_set_pte_pde,
 	.set_prt = gmc_v8_0_set_prt,
 	.get_vm_pte_flags = gmc_v8_0_get_vm_pte_flags,
 	.get_vm_pde = gmc_v8_0_get_vm_pde
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 2fe8397241ea..3fd79e07944d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -47,6 +47,8 @@
 
 #include "ivsrcid/vmc/irqsrcs_vmc_1_0.h"
 
+#include "amdgpu_ras.h"
+
 /* add these here since we already include dce12 headers and these are for DCN */
 #define mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION                                                          0x055d
 #define mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION_BASE_IDX                                                 2
@@ -84,121 +86,182 @@ static const struct soc15_reg_golden golden_settings_athub_1_0_0[] =
 	SOC15_REG_GOLDEN_VALUE(ATHUB, 0, mmRPB_ARB_CNTL2, 0x00ff00ff, 0x00080008)
 };
 
-/* Ecc related register addresses, (BASE + reg offset) */
-/* Universal Memory Controller caps (may be fused). */
-/* UMCCH:UmcLocalCap */
-#define UMCLOCALCAPS_ADDR0	(0x00014306 + 0x00000000)
-#define UMCLOCALCAPS_ADDR1	(0x00014306 + 0x00000800)
-#define UMCLOCALCAPS_ADDR2	(0x00014306 + 0x00001000)
-#define UMCLOCALCAPS_ADDR3	(0x00014306 + 0x00001800)
-#define UMCLOCALCAPS_ADDR4	(0x00054306 + 0x00000000)
-#define UMCLOCALCAPS_ADDR5	(0x00054306 + 0x00000800)
-#define UMCLOCALCAPS_ADDR6	(0x00054306 + 0x00001000)
-#define UMCLOCALCAPS_ADDR7	(0x00054306 + 0x00001800)
-#define UMCLOCALCAPS_ADDR8	(0x00094306 + 0x00000000)
-#define UMCLOCALCAPS_ADDR9	(0x00094306 + 0x00000800)
-#define UMCLOCALCAPS_ADDR10	(0x00094306 + 0x00001000)
-#define UMCLOCALCAPS_ADDR11	(0x00094306 + 0x00001800)
-#define UMCLOCALCAPS_ADDR12	(0x000d4306 + 0x00000000)
-#define UMCLOCALCAPS_ADDR13	(0x000d4306 + 0x00000800)
-#define UMCLOCALCAPS_ADDR14	(0x000d4306 + 0x00001000)
-#define UMCLOCALCAPS_ADDR15	(0x000d4306 + 0x00001800)
-
-/* Universal Memory Controller Channel config. */
-/* UMCCH:UMC_CONFIG */
-#define UMCCH_UMC_CONFIG_ADDR0	(0x00014040 + 0x00000000)
-#define UMCCH_UMC_CONFIG_ADDR1	(0x00014040 + 0x00000800)
-#define UMCCH_UMC_CONFIG_ADDR2	(0x00014040 + 0x00001000)
-#define UMCCH_UMC_CONFIG_ADDR3	(0x00014040 + 0x00001800)
-#define UMCCH_UMC_CONFIG_ADDR4	(0x00054040 + 0x00000000)
-#define UMCCH_UMC_CONFIG_ADDR5	(0x00054040 + 0x00000800)
-#define UMCCH_UMC_CONFIG_ADDR6	(0x00054040 + 0x00001000)
-#define UMCCH_UMC_CONFIG_ADDR7	(0x00054040 + 0x00001800)
-#define UMCCH_UMC_CONFIG_ADDR8	(0x00094040 + 0x00000000)
-#define UMCCH_UMC_CONFIG_ADDR9	(0x00094040 + 0x00000800)
-#define UMCCH_UMC_CONFIG_ADDR10	(0x00094040 + 0x00001000)
-#define UMCCH_UMC_CONFIG_ADDR11	(0x00094040 + 0x00001800)
-#define UMCCH_UMC_CONFIG_ADDR12	(0x000d4040 + 0x00000000)
-#define UMCCH_UMC_CONFIG_ADDR13	(0x000d4040 + 0x00000800)
-#define UMCCH_UMC_CONFIG_ADDR14	(0x000d4040 + 0x00001000)
-#define UMCCH_UMC_CONFIG_ADDR15	(0x000d4040 + 0x00001800)
-
-/* Universal Memory Controller Channel Ecc config. */
-/* UMCCH:EccCtrl */
-#define UMCCH_ECCCTRL_ADDR0	(0x00014053 + 0x00000000)
-#define UMCCH_ECCCTRL_ADDR1	(0x00014053 + 0x00000800)
-#define UMCCH_ECCCTRL_ADDR2	(0x00014053 + 0x00001000)
-#define UMCCH_ECCCTRL_ADDR3	(0x00014053 + 0x00001800)
-#define UMCCH_ECCCTRL_ADDR4	(0x00054053 + 0x00000000)
-#define UMCCH_ECCCTRL_ADDR5	(0x00054053 + 0x00000800)
-#define UMCCH_ECCCTRL_ADDR6	(0x00054053 + 0x00001000)
-#define UMCCH_ECCCTRL_ADDR7	(0x00054053 + 0x00001800)
-#define UMCCH_ECCCTRL_ADDR8	(0x00094053 + 0x00000000)
-#define UMCCH_ECCCTRL_ADDR9	(0x00094053 + 0x00000800)
-#define UMCCH_ECCCTRL_ADDR10	(0x00094053 + 0x00001000)
-#define UMCCH_ECCCTRL_ADDR11	(0x00094053 + 0x00001800)
-#define UMCCH_ECCCTRL_ADDR12	(0x000d4053 + 0x00000000)
-#define UMCCH_ECCCTRL_ADDR13	(0x000d4053 + 0x00000800)
-#define UMCCH_ECCCTRL_ADDR14	(0x000d4053 + 0x00001000)
-#define UMCCH_ECCCTRL_ADDR15	(0x000d4053 + 0x00001800)
-
-static const uint32_t ecc_umclocalcap_addrs[] = {
-	UMCLOCALCAPS_ADDR0,
-	UMCLOCALCAPS_ADDR1,
-	UMCLOCALCAPS_ADDR2,
-	UMCLOCALCAPS_ADDR3,
-	UMCLOCALCAPS_ADDR4,
-	UMCLOCALCAPS_ADDR5,
-	UMCLOCALCAPS_ADDR6,
-	UMCLOCALCAPS_ADDR7,
-	UMCLOCALCAPS_ADDR8,
-	UMCLOCALCAPS_ADDR9,
-	UMCLOCALCAPS_ADDR10,
-	UMCLOCALCAPS_ADDR11,
-	UMCLOCALCAPS_ADDR12,
-	UMCLOCALCAPS_ADDR13,
-	UMCLOCALCAPS_ADDR14,
-	UMCLOCALCAPS_ADDR15,
+static const uint32_t ecc_umc_mcumc_ctrl_addrs[] = {
+	(0x000143c0 + 0x00000000),
+	(0x000143c0 + 0x00000800),
+	(0x000143c0 + 0x00001000),
+	(0x000143c0 + 0x00001800),
+	(0x000543c0 + 0x00000000),
+	(0x000543c0 + 0x00000800),
+	(0x000543c0 + 0x00001000),
+	(0x000543c0 + 0x00001800),
+	(0x000943c0 + 0x00000000),
+	(0x000943c0 + 0x00000800),
+	(0x000943c0 + 0x00001000),
+	(0x000943c0 + 0x00001800),
+	(0x000d43c0 + 0x00000000),
+	(0x000d43c0 + 0x00000800),
+	(0x000d43c0 + 0x00001000),
+	(0x000d43c0 + 0x00001800),
+	(0x001143c0 + 0x00000000),
+	(0x001143c0 + 0x00000800),
+	(0x001143c0 + 0x00001000),
+	(0x001143c0 + 0x00001800),
+	(0x001543c0 + 0x00000000),
+	(0x001543c0 + 0x00000800),
+	(0x001543c0 + 0x00001000),
+	(0x001543c0 + 0x00001800),
+	(0x001943c0 + 0x00000000),
+	(0x001943c0 + 0x00000800),
+	(0x001943c0 + 0x00001000),
+	(0x001943c0 + 0x00001800),
+	(0x001d43c0 + 0x00000000),
+	(0x001d43c0 + 0x00000800),
+	(0x001d43c0 + 0x00001000),
+	(0x001d43c0 + 0x00001800),
 };
 
-static const uint32_t ecc_umcch_umc_config_addrs[] = {
-	UMCCH_UMC_CONFIG_ADDR0,
-	UMCCH_UMC_CONFIG_ADDR1,
-	UMCCH_UMC_CONFIG_ADDR2,
-	UMCCH_UMC_CONFIG_ADDR3,
-	UMCCH_UMC_CONFIG_ADDR4,
-	UMCCH_UMC_CONFIG_ADDR5,
-	UMCCH_UMC_CONFIG_ADDR6,
-	UMCCH_UMC_CONFIG_ADDR7,
-	UMCCH_UMC_CONFIG_ADDR8,
-	UMCCH_UMC_CONFIG_ADDR9,
-	UMCCH_UMC_CONFIG_ADDR10,
-	UMCCH_UMC_CONFIG_ADDR11,
-	UMCCH_UMC_CONFIG_ADDR12,
-	UMCCH_UMC_CONFIG_ADDR13,
-	UMCCH_UMC_CONFIG_ADDR14,
-	UMCCH_UMC_CONFIG_ADDR15,
+static const uint32_t ecc_umc_mcumc_ctrl_mask_addrs[] = {
+	(0x000143e0 + 0x00000000),
+	(0x000143e0 + 0x00000800),
+	(0x000143e0 + 0x00001000),
+	(0x000143e0 + 0x00001800),
+	(0x000543e0 + 0x00000000),
+	(0x000543e0 + 0x00000800),
+	(0x000543e0 + 0x00001000),
+	(0x000543e0 + 0x00001800),
+	(0x000943e0 + 0x00000000),
+	(0x000943e0 + 0x00000800),
+	(0x000943e0 + 0x00001000),
+	(0x000943e0 + 0x00001800),
+	(0x000d43e0 + 0x00000000),
+	(0x000d43e0 + 0x00000800),
+	(0x000d43e0 + 0x00001000),
+	(0x000d43e0 + 0x00001800),
+	(0x001143e0 + 0x00000000),
+	(0x001143e0 + 0x00000800),
+	(0x001143e0 + 0x00001000),
+	(0x001143e0 + 0x00001800),
+	(0x001543e0 + 0x00000000),
+	(0x001543e0 + 0x00000800),
+	(0x001543e0 + 0x00001000),
+	(0x001543e0 + 0x00001800),
+	(0x001943e0 + 0x00000000),
+	(0x001943e0 + 0x00000800),
+	(0x001943e0 + 0x00001000),
+	(0x001943e0 + 0x00001800),
+	(0x001d43e0 + 0x00000000),
+	(0x001d43e0 + 0x00000800),
+	(0x001d43e0 + 0x00001000),
+	(0x001d43e0 + 0x00001800),
 };
 
-static const uint32_t ecc_umcch_eccctrl_addrs[] = {
-	UMCCH_ECCCTRL_ADDR0,
-	UMCCH_ECCCTRL_ADDR1,
-	UMCCH_ECCCTRL_ADDR2,
-	UMCCH_ECCCTRL_ADDR3,
-	UMCCH_ECCCTRL_ADDR4,
-	UMCCH_ECCCTRL_ADDR5,
-	UMCCH_ECCCTRL_ADDR6,
-	UMCCH_ECCCTRL_ADDR7,
-	UMCCH_ECCCTRL_ADDR8,
-	UMCCH_ECCCTRL_ADDR9,
-	UMCCH_ECCCTRL_ADDR10,
-	UMCCH_ECCCTRL_ADDR11,
-	UMCCH_ECCCTRL_ADDR12,
-	UMCCH_ECCCTRL_ADDR13,
-	UMCCH_ECCCTRL_ADDR14,
-	UMCCH_ECCCTRL_ADDR15,
+static const uint32_t ecc_umc_mcumc_status_addrs[] = {
+	(0x000143c2 + 0x00000000),
+	(0x000143c2 + 0x00000800),
+	(0x000143c2 + 0x00001000),
+	(0x000143c2 + 0x00001800),
+	(0x000543c2 + 0x00000000),
+	(0x000543c2 + 0x00000800),
+	(0x000543c2 + 0x00001000),
+	(0x000543c2 + 0x00001800),
+	(0x000943c2 + 0x00000000),
+	(0x000943c2 + 0x00000800),
+	(0x000943c2 + 0x00001000),
+	(0x000943c2 + 0x00001800),
+	(0x000d43c2 + 0x00000000),
+	(0x000d43c2 + 0x00000800),
+	(0x000d43c2 + 0x00001000),
+	(0x000d43c2 + 0x00001800),
+	(0x001143c2 + 0x00000000),
+	(0x001143c2 + 0x00000800),
+	(0x001143c2 + 0x00001000),
+	(0x001143c2 + 0x00001800),
+	(0x001543c2 + 0x00000000),
+	(0x001543c2 + 0x00000800),
+	(0x001543c2 + 0x00001000),
+	(0x001543c2 + 0x00001800),
+	(0x001943c2 + 0x00000000),
+	(0x001943c2 + 0x00000800),
+	(0x001943c2 + 0x00001000),
+	(0x001943c2 + 0x00001800),
+	(0x001d43c2 + 0x00000000),
+	(0x001d43c2 + 0x00000800),
+	(0x001d43c2 + 0x00001000),
+	(0x001d43c2 + 0x00001800),
 };
 
+static int gmc_v9_0_ecc_interrupt_state(struct amdgpu_device *adev,
+		struct amdgpu_irq_src *src,
+		unsigned type,
+		enum amdgpu_interrupt_state state)
+{
+	u32 bits, i, tmp, reg;
+
+	bits = 0x7f;
+
+	switch (state) {
+	case AMDGPU_IRQ_STATE_DISABLE:
+		for (i = 0; i < ARRAY_SIZE(ecc_umc_mcumc_ctrl_addrs); i++) {
+			reg = ecc_umc_mcumc_ctrl_addrs[i];
+			tmp = RREG32(reg);
+			tmp &= ~bits;
+			WREG32(reg, tmp);
+		}
+		for (i = 0; i < ARRAY_SIZE(ecc_umc_mcumc_ctrl_mask_addrs); i++) {
+			reg = ecc_umc_mcumc_ctrl_mask_addrs[i];
+			tmp = RREG32(reg);
+			tmp &= ~bits;
+			WREG32(reg, tmp);
+		}
+		break;
+	case AMDGPU_IRQ_STATE_ENABLE:
+		for (i = 0; i < ARRAY_SIZE(ecc_umc_mcumc_ctrl_addrs); i++) {
+			reg = ecc_umc_mcumc_ctrl_addrs[i];
+			tmp = RREG32(reg);
+			tmp |= bits;
+			WREG32(reg, tmp);
+		}
+		for (i = 0; i < ARRAY_SIZE(ecc_umc_mcumc_ctrl_mask_addrs); i++) {
+			reg = ecc_umc_mcumc_ctrl_mask_addrs[i];
+			tmp = RREG32(reg);
+			tmp |= bits;
+			WREG32(reg, tmp);
+		}
+		break;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+static int gmc_v9_0_process_ras_data_cb(struct amdgpu_device *adev,
+		struct amdgpu_iv_entry *entry)
+{
+	kgd2kfd_set_sram_ecc_flag(adev->kfd.dev);
+	amdgpu_ras_reset_gpu(adev, 0);
+	return AMDGPU_RAS_UE;
+}
+
+static int gmc_v9_0_process_ecc_irq(struct amdgpu_device *adev,
+		struct amdgpu_irq_src *source,
+		struct amdgpu_iv_entry *entry)
+{
+	struct ras_common_if *ras_if = adev->gmc.ras_if;
+	struct ras_dispatch_if ih_data = {
+		.entry = entry,
+	};
+
+	if (!ras_if)
+		return 0;
+
+	ih_data.head = *ras_if;
+
+	amdgpu_ras_interrupt_dispatch(adev, &ih_data);
+	return 0;
+}
+
 static int gmc_v9_0_vm_fault_interrupt_state(struct amdgpu_device *adev,
 					struct amdgpu_irq_src *src,
 					unsigned type,
@@ -244,62 +307,6 @@ static int gmc_v9_0_vm_fault_interrupt_state(struct amdgpu_device *adev,
 	return 0;
 }
 
-/**
- * vega10_ih_prescreen_iv - prescreen an interrupt vector
- *
- * @adev: amdgpu_device pointer
- *
- * Returns true if the interrupt vector should be further processed.
- */
-static bool gmc_v9_0_prescreen_iv(struct amdgpu_device *adev,
-				  struct amdgpu_iv_entry *entry,
-				  uint64_t addr)
-{
-	struct amdgpu_vm *vm;
-	u64 key;
-	int r;
-
-	/* No PASID, can't identify faulting process */
-	if (!entry->pasid)
-		return true;
-
-	/* Not a retry fault */
-	if (!(entry->src_data[1] & 0x80))
-		return true;
-
-	/* Track retry faults in per-VM fault FIFO. */
-	spin_lock(&adev->vm_manager.pasid_lock);
-	vm = idr_find(&adev->vm_manager.pasid_idr, entry->pasid);
-	if (!vm) {
-		/* VM not found, process it normally */
-		spin_unlock(&adev->vm_manager.pasid_lock);
-		return true;
-	}
-
-	key = AMDGPU_VM_FAULT(entry->pasid, addr);
-	r = amdgpu_vm_add_fault(vm->fault_hash, key);
-
-	/* Hash table is full or the fault is already being processed,
-	 * ignore further page faults
-	 */
-	if (r != 0) {
-		spin_unlock(&adev->vm_manager.pasid_lock);
-		return false;
-	}
-	/* No locking required with single writer and single reader */
-	r = kfifo_put(&vm->faults, key);
-	if (!r) {
-		/* FIFO is full. Ignore it until there is space */
-		amdgpu_vm_clear_fault(vm->fault_hash, key);
-		spin_unlock(&adev->vm_manager.pasid_lock);
-		return false;
-	}
-
-	spin_unlock(&adev->vm_manager.pasid_lock);
-	/* It's the first fault for this address, process it normally */
-	return true;
-}
-
 static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 				struct amdgpu_irq_src *source,
 				struct amdgpu_iv_entry *entry)
@@ -312,9 +319,11 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 	addr = (u64)entry->src_data[0] << 12;
 	addr |= ((u64)entry->src_data[1] & 0xf) << 44;
 
-	if (!gmc_v9_0_prescreen_iv(adev, entry, addr))
+	if (retry_fault && amdgpu_gmc_filter_faults(adev, addr, entry->pasid,
+						    entry->timestamp))
 		return 1; /* This also prevents sending it to KFD */
 
+	/* If it's the first fault for this address, process it normally */
 	if (!amdgpu_sriov_vf(adev)) {
 		status = RREG32(hub->vm_l2_pro_fault_status);
 		WREG32_P(hub->vm_l2_pro_fault_cntl, 1, ~1);
@@ -350,10 +359,19 @@ static const struct amdgpu_irq_src_funcs gmc_v9_0_irq_funcs = {
 	.process = gmc_v9_0_process_interrupt,
 };
 
+
+static const struct amdgpu_irq_src_funcs gmc_v9_0_ecc_funcs = {
+	.set = gmc_v9_0_ecc_interrupt_state,
+	.process = gmc_v9_0_process_ecc_irq,
+};
+
 static void gmc_v9_0_set_irq_funcs(struct amdgpu_device *adev)
 {
 	adev->gmc.vm_fault.num_types = 1;
 	adev->gmc.vm_fault.funcs = &gmc_v9_0_irq_funcs;
+
+	adev->gmc.ecc_irq.num_types = 1;
+	adev->gmc.ecc_irq.funcs = &gmc_v9_0_ecc_funcs;
 }
 
 static uint32_t gmc_v9_0_get_invalidate_req(unsigned int vmid,
@@ -466,64 +484,37 @@ static void gmc_v9_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned vmid,
 	amdgpu_ring_emit_wreg(ring, reg, pasid);
 }
 
-/**
- * gmc_v9_0_set_pte_pde - update the page tables using MMIO
- *
- * @adev: amdgpu_device pointer
- * @cpu_pt_addr: cpu address of the page table
- * @gpu_page_idx: entry in the page table to update
- * @addr: dst addr to write into pte/pde
- * @flags: access flags
+/*
+ * PTE format on VEGA 10:
+ * 63:59 reserved
+ * 58:57 mtype
+ * 56 F
+ * 55 L
+ * 54 P
+ * 53 SW
+ * 52 T
+ * 50:48 reserved
+ * 47:12 4k physical page base address
+ * 11:7 fragment
+ * 6 write
+ * 5 read
+ * 4 exe
+ * 3 Z
+ * 2 snooped
+ * 1 system
+ * 0 valid
  *
- * Update the page tables using the CPU.
+ * PDE format on VEGA 10:
+ * 63:59 block fragment size
+ * 58:55 reserved
+ * 54 P
+ * 53:48 reserved
+ * 47:6 physical base address of PD or PTE
+ * 5:3 reserved
+ * 2 C
+ * 1 system
+ * 0 valid
  */
-static int gmc_v9_0_set_pte_pde(struct amdgpu_device *adev, void *cpu_pt_addr,
-				uint32_t gpu_page_idx, uint64_t addr,
-				uint64_t flags)
-{
-	void __iomem *ptr = (void *)cpu_pt_addr;
-	uint64_t value;
-
-	/*
-	 * PTE format on VEGA 10:
-	 * 63:59 reserved
-	 * 58:57 mtype
-	 * 56 F
-	 * 55 L
-	 * 54 P
-	 * 53 SW
-	 * 52 T
-	 * 50:48 reserved
-	 * 47:12 4k physical page base address
-	 * 11:7 fragment
-	 * 6 write
-	 * 5 read
-	 * 4 exe
-	 * 3 Z
-	 * 2 snooped
-	 * 1 system
-	 * 0 valid
-	 *
-	 * PDE format on VEGA 10:
-	 * 63:59 block fragment size
-	 * 58:55 reserved
-	 * 54 P
-	 * 53:48 reserved
-	 * 47:6 physical base address of PD or PTE
-	 * 5:3 reserved
-	 * 2 C
-	 * 1 system
-	 * 0 valid
-	 */
-
-	/*
-	 * The following is for PTE only. GART does not have PDEs.
-	*/
-	value = addr & 0x0000FFFFFFFFF000ULL;
-	value |= flags;
-	writeq(value, ptr + (gpu_page_idx * 8));
-	return 0;
-}
 
 static uint64_t gmc_v9_0_get_vm_pte_flags(struct amdgpu_device *adev,
 						uint32_t flags)
@@ -593,7 +584,6 @@ static const struct amdgpu_gmc_funcs gmc_v9_0_gmc_funcs = {
 	.flush_gpu_tlb = gmc_v9_0_flush_gpu_tlb,
 	.emit_flush_gpu_tlb = gmc_v9_0_emit_flush_gpu_tlb,
 	.emit_pasid_mapping = gmc_v9_0_emit_pasid_mapping,
-	.set_pte_pde = gmc_v9_0_set_pte_pde,
 	.get_vm_pte_flags = gmc_v9_0_get_vm_pte_flags,
 	.get_vm_pde = gmc_v9_0_get_vm_pde
 };
@@ -620,85 +610,6 @@ static int gmc_v9_0_early_init(void *handle)
 	return 0;
 }
 
-static int gmc_v9_0_ecc_available(struct amdgpu_device *adev)
-{
-	uint32_t reg_val;
-	uint32_t reg_addr;
-	uint32_t field_val;
-	size_t i;
-	uint32_t fv2;
-	size_t lost_sheep;
-
-	DRM_DEBUG("ecc: gmc_v9_0_ecc_available()\n");
-
-	lost_sheep = 0;
-	for (i = 0; i < ARRAY_SIZE(ecc_umclocalcap_addrs); ++i) {
-		reg_addr = ecc_umclocalcap_addrs[i];
-		DRM_DEBUG("ecc: "
-			  "UMCCH_UmcLocalCap[%zu]: reg_addr: 0x%08x\n",
-			  i, reg_addr);
-		reg_val = RREG32(reg_addr);
-		field_val = REG_GET_FIELD(reg_val, UMCCH0_0_UmcLocalCap,
-					  EccDis);
-		DRM_DEBUG("ecc: "
-			  "reg_val: 0x%08x, "
-			  "EccDis: 0x%08x, ",
-			  reg_val, field_val);
-		if (field_val) {
-			DRM_ERROR("ecc: UmcLocalCap:EccDis is set.\n");
-			++lost_sheep;
-		}
-	}
-
-	for (i = 0; i < ARRAY_SIZE(ecc_umcch_umc_config_addrs); ++i) {
-		reg_addr = ecc_umcch_umc_config_addrs[i];
-		DRM_DEBUG("ecc: "
-			  "UMCCH0_0_UMC_CONFIG[%zu]: reg_addr: 0x%08x",
-			  i, reg_addr);
-		reg_val = RREG32(reg_addr);
-		field_val = REG_GET_FIELD(reg_val, UMCCH0_0_UMC_CONFIG,
-					  DramReady);
-		DRM_DEBUG("ecc: "
-			  "reg_val: 0x%08x, "
-			  "DramReady: 0x%08x\n",
-			  reg_val, field_val);
-
-		if (!field_val) {
-			DRM_ERROR("ecc: UMC_CONFIG:DramReady is not set.\n");
-			++lost_sheep;
-		}
-	}
-
-	for (i = 0; i < ARRAY_SIZE(ecc_umcch_eccctrl_addrs); ++i) {
-		reg_addr = ecc_umcch_eccctrl_addrs[i];
-		DRM_DEBUG("ecc: "
-			  "UMCCH_EccCtrl[%zu]: reg_addr: 0x%08x, ",
-			  i, reg_addr);
-		reg_val = RREG32(reg_addr);
-		field_val = REG_GET_FIELD(reg_val, UMCCH0_0_EccCtrl,
-					  WrEccEn);
-		fv2 = REG_GET_FIELD(reg_val, UMCCH0_0_EccCtrl,
-				    RdEccEn);
-		DRM_DEBUG("ecc: "
-			  "reg_val: 0x%08x, "
-			  "WrEccEn: 0x%08x, "
-			  "RdEccEn: 0x%08x\n",
-			  reg_val, field_val, fv2);
-
-		if (!field_val) {
-			DRM_DEBUG("ecc: WrEccEn is not set\n");
-			++lost_sheep;
-		}
-		if (!fv2) {
-			DRM_DEBUG("ecc: RdEccEn is not set\n");
-			++lost_sheep;
-		}
-	}
-
-	DRM_DEBUG("ecc: lost_sheep: %zu\n", lost_sheep);
-	return lost_sheep == 0;
-}
-
 static bool gmc_v9_0_keep_stolen_memory(struct amdgpu_device *adev)
 {
 
@@ -751,31 +662,119 @@ static int gmc_v9_0_allocate_vm_inv_eng(struct amdgpu_device *adev)
 	return 0;
 }
 
-static int gmc_v9_0_late_init(void *handle)
+static int gmc_v9_0_ecc_late_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	struct ras_common_if **ras_if = &adev->gmc.ras_if;
+	struct ras_ih_if ih_info = {
+		.cb = gmc_v9_0_process_ras_data_cb,
+	};
+	struct ras_fs_if fs_info = {
+		.sysfs_name = "umc_err_count",
+		.debugfs_name = "umc_err_inject",
+	};
+	struct ras_common_if ras_block = {
+		.block = AMDGPU_RAS_BLOCK__UMC,
+		.type = AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE,
+		.sub_block_index = 0,
+		.name = "umc",
+	};
 	int r;
 
+	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__UMC)) {
+		amdgpu_ras_feature_enable_on_boot(adev, &ras_block, 0);
+		return 0;
+	}
+	/* handle resume path. */
+	if (*ras_if)
+		goto resume;
+
+	*ras_if = kmalloc(sizeof(**ras_if), GFP_KERNEL);
+	if (!*ras_if)
+		return -ENOMEM;
+
+	**ras_if = ras_block;
+
+	r = amdgpu_ras_feature_enable_on_boot(adev, *ras_if, 1);
+	if (r)
+		goto feature;
+
+	ih_info.head = **ras_if;
+	fs_info.head = **ras_if;
+
+	r = amdgpu_ras_interrupt_add_handler(adev, &ih_info);
+	if (r)
+		goto interrupt;
+
+	r = amdgpu_ras_debugfs_create(adev, &fs_info);
+	if (r)
+		goto debugfs;
+
+	r = amdgpu_ras_sysfs_create(adev, &fs_info);
+	if (r)
+		goto sysfs;
+resume:
+	r = amdgpu_irq_get(adev, &adev->gmc.ecc_irq, 0);
+	if (r)
+		goto irq;
+
+	return 0;
+irq:
+	amdgpu_ras_sysfs_remove(adev, *ras_if);
+sysfs:
+	amdgpu_ras_debugfs_remove(adev, *ras_if);
+debugfs:
+	amdgpu_ras_interrupt_remove_handler(adev, &ih_info);
+interrupt:
+	amdgpu_ras_feature_enable(adev, *ras_if, 0);
+feature:
+	kfree(*ras_if);
+	*ras_if = NULL;
+	return -EINVAL;
+}
+
+
+static int gmc_v9_0_late_init(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	bool r;
+
 	if (!gmc_v9_0_keep_stolen_memory(adev))
 		amdgpu_bo_late_init(adev);
 
 	r = gmc_v9_0_allocate_vm_inv_eng(adev);
 	if (r)
 		return r;
+	/* Check if ecc is available */
+	if (!amdgpu_sriov_vf(adev)) {
+		switch (adev->asic_type) {
+		case CHIP_VEGA10:
+		case CHIP_VEGA20:
+			r = amdgpu_atomfirmware_mem_ecc_supported(adev);
+			if (!r) {
+				DRM_INFO("ECC is not present.\n");
+				if (adev->df_funcs->enable_ecc_force_par_wr_rmw)
+					adev->df_funcs->enable_ecc_force_par_wr_rmw(adev, false);
+			} else {
+				DRM_INFO("ECC is active.\n");
+			}
 
-	if (adev->asic_type == CHIP_VEGA10 && !amdgpu_sriov_vf(adev)) {
-		r = gmc_v9_0_ecc_available(adev);
-		if (r == 1) {
-			DRM_INFO("ECC is active.\n");
-		} else if (r == 0) {
-			DRM_INFO("ECC is not present.\n");
-			adev->df_funcs->enable_ecc_force_par_wr_rmw(adev, false);
-		} else {
-			DRM_ERROR("gmc_v9_0_ecc_available() failed. r: %d\n", r);
-			return r;
+			r = amdgpu_atomfirmware_sram_ecc_supported(adev);
+			if (!r) {
+				DRM_INFO("SRAM ECC is not present.\n");
+			} else {
+				DRM_INFO("SRAM ECC is active.\n");
+			}
+			break;
+		default:
+			break;
 		}
 	}
 
+	r = gmc_v9_0_ecc_late_init(handle);
+	if (r)
+		return r;
+
 	return amdgpu_irq_get(adev, &adev->gmc.vm_fault, 0);
 }
 
@@ -787,7 +786,7 @@ static void gmc_v9_0_vram_gtt_location(struct amdgpu_device *adev,
 		base = mmhub_v1_0_get_fb_location(adev);
 	/* add the xgmi offset of the physical node */
 	base += adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size;
-	amdgpu_gmc_vram_location(adev, &adev->gmc, base);
+	amdgpu_gmc_vram_location(adev, mc, base);
 	amdgpu_gmc_gart_location(adev, mc);
 	if (!amdgpu_sriov_vf(adev))
 		amdgpu_gmc_agp_location(adev, mc);
@@ -987,6 +986,12 @@ static int gmc_v9_0_sw_init(void *handle)
 	if (r)
 		return r;
 
+	/* interrupt sent to DF. */
+	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_DF, 0,
+			&adev->gmc.ecc_irq);
+	if (r)
+		return r;
+
 	/* Set the internal MC address mask
 	 * This is the max address of the GPU's
 	 * internal address space.
@@ -1011,7 +1016,7 @@ static int gmc_v9_0_sw_init(void *handle)
 		pci_set_consistent_dma_mask(adev->pdev, DMA_BIT_MASK(32));
 		printk(KERN_WARNING "amdgpu: No coherent DMA available.\n");
 	}
-	adev->need_swiotlb = drm_get_max_iomem() > ((u64)1 << dma_bits);
+	adev->need_swiotlb = drm_need_swiotlb(dma_bits);
 
 	if (adev->gmc.xgmi.supported) {
 		r = gfxhub_v1_1_get_xgmi_info(adev);
@@ -1052,6 +1057,22 @@ static int gmc_v9_0_sw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__UMC) &&
+			adev->gmc.ras_if) {
+		struct ras_common_if *ras_if = adev->gmc.ras_if;
+		struct ras_ih_if ih_info = {
+			.head = *ras_if,
+		};
+
+		/*remove fs first*/
+		amdgpu_ras_debugfs_remove(adev, ras_if);
+		amdgpu_ras_sysfs_remove(adev, ras_if);
+		/*remove the IH*/
+		amdgpu_ras_interrupt_remove_handler(adev, &ih_info);
+		amdgpu_ras_feature_enable(adev, ras_if, 0);
+		kfree(ras_if);
+	}
+
 	amdgpu_gem_force_release(adev);
 	amdgpu_vm_manager_fini(adev);
 
@@ -1198,6 +1219,7 @@ static int gmc_v9_0_hw_fini(void *handle)
 		return 0;
 	}
 
+	amdgpu_irq_put(adev, &adev->gmc.ecc_irq, 0);
 	amdgpu_irq_put(adev, &adev->gmc.vm_fault, 0);
 	gmc_v9_0_gart_disable(adev);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/kv_dpm.c b/drivers/gpu/drm/amd/amdgpu/kv_dpm.c
index 0c9a2c03504e..f2e6b148ccad 100644
--- a/drivers/gpu/drm/amd/amdgpu/kv_dpm.c
+++ b/drivers/gpu/drm/amd/amdgpu/kv_dpm.c
@@ -2824,7 +2824,7 @@ static int kv_dpm_init(struct amdgpu_device *adev)
 		pi->caps_tcp_ramping = true;
 	}
 
-	if (adev->powerplay.pp_feature & PP_SCLK_DEEP_SLEEP_MASK)
+	if (adev->pm.pp_feature & PP_SCLK_DEEP_SLEEP_MASK)
 		pi->caps_sclk_ds = true;
 	else
 		pi->caps_sclk_ds = false;
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
index 1696644ec022..41a9a5779623 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
@@ -163,7 +163,7 @@ static void mmhub_v1_0_init_cache_regs(struct amdgpu_device *adev)
 	/* XXX for emulation, Refer to closed source code.*/
 	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, L2_PDE0_CACHE_TAG_GENERATION_MODE,
 			    0);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, PDE_FAULT_CLASSIFICATION, 1);
+	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, PDE_FAULT_CLASSIFICATION, 0);
 	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, CONTEXT1_IDENTITY_ACCESS_MODE, 1);
 	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, IDENTITY_MODE_FRAGMENT_SIZE, 0);
 	WREG32_SOC15(MMHUB, 0, mmVM_L2_CNTL, tmp);
@@ -256,7 +256,7 @@ static void mmhub_v1_0_setup_vmid_config(struct amdgpu_device *adev)
 				    block_size);
 		/* Send no-retry XNACK on fault to suppress VM fault storm. */
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 0);
+				    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 1);
 		WREG32_SOC15_OFFSET(MMHUB, 0, mmVM_CONTEXT1_CNTL, i, tmp);
 		WREG32_SOC15_OFFSET(MMHUB, 0, mmVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32, i*2, 0);
 		WREG32_SOC15_OFFSET(MMHUB, 0, mmVM_CONTEXT1_PAGE_TABLE_START_ADDR_HI32, i*2, 0);
diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c
index 73851ebb3833..8dbad496b29f 100644
--- a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c
+++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c
@@ -157,6 +157,82 @@ static void xgpu_ai_mailbox_trans_msg (struct amdgpu_device *adev,
 	xgpu_ai_mailbox_set_valid(adev, false);
 }
 
+static int xgpu_ai_get_pp_clk(struct amdgpu_device *adev, u32 type, char *buf)
+{
+        int r = 0;
+        u32 req, val, size;
+
+        if (!amdgim_is_hwperf(adev) || buf == NULL)
+                return -EBADRQC;
+
+        switch(type) {
+        case PP_SCLK:
+                req = IDH_IRQ_GET_PP_SCLK;
+                break;
+        case PP_MCLK:
+                req = IDH_IRQ_GET_PP_MCLK;
+                break;
+        default:
+                return -EBADRQC;
+        }
+
+        mutex_lock(&adev->virt.dpm_mutex);
+
+        xgpu_ai_mailbox_trans_msg(adev, req, 0, 0, 0);
+
+        r = xgpu_ai_poll_msg(adev, IDH_SUCCESS);
+        if (!r && adev->fw_vram_usage.va != NULL) {
+                val = RREG32_NO_KIQ(
+                        SOC15_REG_OFFSET(NBIO, 0,
+                                         mmBIF_BX_PF0_MAILBOX_MSGBUF_RCV_DW1));
+                size = strnlen((((char *)adev->virt.fw_reserve.p_pf2vf) +
+                                val), PAGE_SIZE);
+
+                if (size < PAGE_SIZE)
+                        strcpy(buf,((char *)adev->virt.fw_reserve.p_pf2vf + val));
+                else
+                        size = 0;
+
+                r = size;
+                goto out;
+        }
+
+        r = xgpu_ai_poll_msg(adev, IDH_FAIL);
+        if(r)
+                pr_info("%s DPM request failed",
+                        (type == PP_SCLK)? "SCLK" : "MCLK");
+
+out:
+        mutex_unlock(&adev->virt.dpm_mutex);
+        return r;
+}
+
+static int xgpu_ai_force_dpm_level(struct amdgpu_device *adev, u32 level)
+{
+        int r = 0;
+        u32 req = IDH_IRQ_FORCE_DPM_LEVEL;
+
+        if (!amdgim_is_hwperf(adev))
+                return -EBADRQC;
+
+        mutex_lock(&adev->virt.dpm_mutex);
+        xgpu_ai_mailbox_trans_msg(adev, req, level, 0, 0);
+
+        r = xgpu_ai_poll_msg(adev, IDH_SUCCESS);
+        if (!r)
+                goto out;
+
+        r = xgpu_ai_poll_msg(adev, IDH_FAIL);
+        if (!r)
+                pr_info("DPM request failed");
+        else
+                pr_info("Mailbox is broken");
+
+out:
+        mutex_unlock(&adev->virt.dpm_mutex);
+        return r;
+}
+
 static int xgpu_ai_send_access_requests(struct amdgpu_device *adev,
 					enum idh_request req)
 {
@@ -375,4 +451,6 @@ const struct amdgpu_virt_ops xgpu_ai_virt_ops = {
 	.reset_gpu = xgpu_ai_request_reset,
 	.wait_reset = NULL,
 	.trans_msg = xgpu_ai_mailbox_trans_msg,
+	.get_pp_clk = xgpu_ai_get_pp_clk,
+	.force_dpm_level = xgpu_ai_force_dpm_level,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h
index b4a9ceea334b..39d151b79153 100644
--- a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h
+++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h
@@ -35,6 +35,10 @@ enum idh_request {
 	IDH_REL_GPU_FINI_ACCESS,
 	IDH_REQ_GPU_RESET_ACCESS,
 
+	IDH_IRQ_FORCE_DPM_LEVEL = 10,
+	IDH_IRQ_GET_PP_SCLK,
+	IDH_IRQ_GET_PP_MCLK,
+
 	IDH_LOG_VF_ERROR       = 200,
 };
 
@@ -43,6 +47,8 @@ enum idh_event {
 	IDH_READY_TO_ACCESS_GPU,
 	IDH_FLR_NOTIFICATION,
 	IDH_FLR_NOTIFICATION_CMPL,
+	IDH_SUCCESS,
+	IDH_FAIL,
 	IDH_EVENT_MAX
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c
index 6a0fcd67662a..aef9d059ae52 100644
--- a/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c
+++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c
@@ -515,7 +515,7 @@ static void xgpu_vi_mailbox_flr_work(struct work_struct *work)
 
 	/* wait until RCV_MSG become 3 */
 	if (xgpu_vi_poll_msg(adev, IDH_FLR_NOTIFICATION_CMPL)) {
-		pr_err("failed to recieve FLR_CMPL\n");
+		pr_err("failed to receive FLR_CMPL\n");
 		return;
 	}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v6_1.c b/drivers/gpu/drm/amd/amdgpu/nbio_v6_1.c
index cc967dbfd631..6590143c3f75 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v6_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v6_1.c
@@ -118,7 +118,8 @@ static void nbio_v6_1_ih_doorbell_range(struct amdgpu_device *adev,
 
 	if (use_doorbell) {
 		ih_doorbell_range = REG_SET_FIELD(ih_doorbell_range, BIF_IH_DOORBELL_RANGE, OFFSET, doorbell_index);
-		ih_doorbell_range = REG_SET_FIELD(ih_doorbell_range, BIF_IH_DOORBELL_RANGE, SIZE, 2);
+		ih_doorbell_range = REG_SET_FIELD(ih_doorbell_range,
+						  BIF_IH_DOORBELL_RANGE, SIZE, 6);
 	} else
 		ih_doorbell_range = REG_SET_FIELD(ih_doorbell_range, BIF_IH_DOORBELL_RANGE, SIZE, 0);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h
index f3a7d207af07..2f79765b4bdb 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h
+++ b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h
@@ -43,6 +43,7 @@ enum psp_gfx_crtl_cmd_id
     GFX_CTRL_CMD_ID_ENABLE_INT      = 0x00050000,   /* enable PSP-to-Gfx interrupt */
     GFX_CTRL_CMD_ID_DISABLE_INT     = 0x00060000,   /* disable PSP-to-Gfx interrupt */
     GFX_CTRL_CMD_ID_MODE1_RST       = 0x00070000,   /* trigger the Mode 1 reset */
+    GFX_CTRL_CMD_ID_GBR_IH_SET      = 0x00080000,   /* set Gbr IH_RB_CNTL registers */
     GFX_CTRL_CMD_ID_CONSUME_CMD     = 0x000A0000,   /* send interrupt to psp for updating write pointer of vf */
     GFX_CTRL_CMD_ID_DESTROY_GPCOM_RING = 0x000C0000, /* destroy GPCOM ring */
 
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c
index 860b70d80d3c..b91df7bd1d98 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c
@@ -33,6 +33,9 @@
 #include "sdma0/sdma0_4_0_offset.h"
 #include "nbio/nbio_7_4_offset.h"
 
+#include "oss/osssys_4_0_offset.h"
+#include "oss/osssys_4_0_sh_mask.h"
+
 MODULE_FIRMWARE("amdgpu/vega20_sos.bin");
 MODULE_FIRMWARE("amdgpu/vega20_asd.bin");
 MODULE_FIRMWARE("amdgpu/vega20_ta.bin");
@@ -113,6 +116,13 @@ static int psp_v11_0_init_microcode(struct psp_context *psp)
 		adev->psp.ta_xgmi_ucode_size = le32_to_cpu(ta_hdr->ta_xgmi_size_bytes);
 		adev->psp.ta_xgmi_start_addr = (uint8_t *)ta_hdr +
 			le32_to_cpu(ta_hdr->header.ucode_array_offset_bytes);
+
+		adev->psp.ta_fw_version = le32_to_cpu(ta_hdr->header.ucode_version);
+
+		adev->psp.ta_ras_ucode_version = le32_to_cpu(ta_hdr->ta_ras_ucode_version);
+		adev->psp.ta_ras_ucode_size = le32_to_cpu(ta_hdr->ta_ras_size_bytes);
+		adev->psp.ta_ras_start_addr = (uint8_t *)adev->psp.ta_xgmi_start_addr +
+			le32_to_cpu(ta_hdr->ta_ras_offset_bytes);
 	}
 
 	return 0;
@@ -217,6 +227,37 @@ static int psp_v11_0_bootloader_load_sos(struct psp_context *psp)
 	return ret;
 }
 
+static void psp_v11_0_reroute_ih(struct psp_context *psp)
+{
+	struct amdgpu_device *adev = psp->adev;
+	uint32_t tmp;
+
+	/* Change IH ring for VMC */
+	tmp = REG_SET_FIELD(0, IH_CLIENT_CFG_DATA, CREDIT_RETURN_ADDR, 0x1244b);
+	tmp = REG_SET_FIELD(tmp, IH_CLIENT_CFG_DATA, CLIENT_TYPE, 1);
+	tmp = REG_SET_FIELD(tmp, IH_CLIENT_CFG_DATA, RING_ID, 1);
+
+	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_69, 3);
+	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_70, tmp);
+	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, GFX_CTRL_CMD_ID_GBR_IH_SET);
+
+	mdelay(20);
+	psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64),
+		     0x80000000, 0x8000FFFF, false);
+
+	/* Change IH ring for UMC */
+	tmp = REG_SET_FIELD(0, IH_CLIENT_CFG_DATA, CREDIT_RETURN_ADDR, 0x1216b);
+	tmp = REG_SET_FIELD(tmp, IH_CLIENT_CFG_DATA, RING_ID, 1);
+
+	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_69, 4);
+	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_70, tmp);
+	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, GFX_CTRL_CMD_ID_GBR_IH_SET);
+
+	mdelay(20);
+	psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64),
+		     0x80000000, 0x8000FFFF, false);
+}
+
 static int psp_v11_0_ring_init(struct psp_context *psp,
 			      enum psp_ring_type ring_type)
 {
@@ -224,6 +265,8 @@ static int psp_v11_0_ring_init(struct psp_context *psp,
 	struct psp_ring *ring;
 	struct amdgpu_device *adev = psp->adev;
 
+	psp_v11_0_reroute_ih(psp);
+
 	ring = &psp->km_ring;
 
 	ring->ring_type = ring_type;
@@ -631,7 +674,7 @@ static int psp_v11_0_xgmi_set_topology_info(struct psp_context *psp,
 	for (i = 0; i < topology_info_input->num_nodes; i++) {
 		topology_info_input->nodes[i].node_id = topology->nodes[i].node_id;
 		topology_info_input->nodes[i].num_hops = topology->nodes[i].num_hops;
-		topology_info_input->nodes[i].is_sharing_enabled = topology->nodes[i].is_sharing_enabled;
+		topology_info_input->nodes[i].is_sharing_enabled = 1;
 		topology_info_input->nodes[i].sdma_engine = topology->nodes[i].sdma_engine;
 	}
 
@@ -679,6 +722,54 @@ static int psp_v11_0_xgmi_get_node_id(struct psp_context *psp, uint64_t *node_id
 	return 0;
 }
 
+static int psp_v11_0_ras_trigger_error(struct psp_context *psp,
+		struct ta_ras_trigger_error_input *info)
+{
+	struct ta_ras_shared_memory *ras_cmd;
+	int ret;
+
+	if (!psp->ras.ras_initialized)
+		return -EINVAL;
+
+	ras_cmd = (struct ta_ras_shared_memory *)psp->ras.ras_shared_buf;
+	memset(ras_cmd, 0, sizeof(struct ta_ras_shared_memory));
+
+	ras_cmd->cmd_id = TA_RAS_COMMAND__TRIGGER_ERROR;
+	ras_cmd->ras_in_message.trigger_error = *info;
+
+	ret = psp_ras_invoke(psp, ras_cmd->cmd_id);
+	if (ret)
+		return -EINVAL;
+
+	return ras_cmd->ras_status;
+}
+
+static int psp_v11_0_ras_cure_posion(struct psp_context *psp, uint64_t *mode_ptr)
+{
+#if 0
+	// not support yet.
+	struct ta_ras_shared_memory *ras_cmd;
+	int ret;
+
+	if (!psp->ras.ras_initialized)
+		return -EINVAL;
+
+	ras_cmd = (struct ta_ras_shared_memory *)psp->ras.ras_shared_buf;
+	memset(ras_cmd, 0, sizeof(struct ta_ras_shared_memory));
+
+	ras_cmd->cmd_id = TA_RAS_COMMAND__CURE_POISON;
+	ras_cmd->ras_in_message.cure_poison.mode_ptr = mode_ptr;
+
+	ret = psp_ras_invoke(psp, ras_cmd->cmd_id);
+	if (ret)
+		return -EINVAL;
+
+	return ras_cmd->ras_status;
+#else
+	return -EINVAL;
+#endif
+}
+
 static const struct psp_funcs psp_v11_0_funcs = {
 	.init_microcode = psp_v11_0_init_microcode,
 	.bootloader_load_sysdrv = psp_v11_0_bootloader_load_sysdrv,
@@ -695,6 +786,8 @@ static const struct psp_funcs psp_v11_0_funcs = {
 	.xgmi_get_hive_id = psp_v11_0_xgmi_get_hive_id,
 	.xgmi_get_node_id = psp_v11_0_xgmi_get_node_id,
 	.support_vmr_ring = psp_v11_0_support_vmr_ring,
+	.ras_trigger_error = psp_v11_0_ras_trigger_error,
+	.ras_cure_posion = psp_v11_0_ras_cure_posion,
 };
 
 void psp_v11_0_set_psp_funcs(struct psp_context *psp)
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c
index 0487e3a4e9e7..143f0fae69d5 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c
@@ -37,6 +37,9 @@
 #include "sdma0/sdma0_4_0_offset.h"
 #include "nbio/nbio_6_1_offset.h"
 
+#include "oss/osssys_4_0_offset.h"
+#include "oss/osssys_4_0_sh_mask.h"
+
 MODULE_FIRMWARE("amdgpu/vega10_sos.bin");
 MODULE_FIRMWARE("amdgpu/vega10_asd.bin");
 MODULE_FIRMWARE("amdgpu/vega12_sos.bin");
@@ -252,6 +255,37 @@ static int psp_v3_1_ring_init(struct psp_context *psp,
 	return 0;
 }
 
+static void psp_v3_1_reroute_ih(struct psp_context *psp)
+{
+	struct amdgpu_device *adev = psp->adev;
+	uint32_t tmp;
+
+	/* Change IH ring for VMC */
+	tmp = REG_SET_FIELD(0, IH_CLIENT_CFG_DATA, CREDIT_RETURN_ADDR, 0x1244b);
+	tmp = REG_SET_FIELD(tmp, IH_CLIENT_CFG_DATA, CLIENT_TYPE, 1);
+	tmp = REG_SET_FIELD(tmp, IH_CLIENT_CFG_DATA, RING_ID, 1);
+
+	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_69, 3);
+	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_70, tmp);
+	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, GFX_CTRL_CMD_ID_GBR_IH_SET);
+
+	mdelay(20);
+	psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64),
+		     0x80000000, 0x8000FFFF, false);
+
+	/* Change IH ring for UMC */
+	tmp = REG_SET_FIELD(0, IH_CLIENT_CFG_DATA, CREDIT_RETURN_ADDR, 0x1216b);
+	tmp = REG_SET_FIELD(tmp, IH_CLIENT_CFG_DATA, RING_ID, 1);
+
+	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_69, 4);
+	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_70, tmp);
+	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, GFX_CTRL_CMD_ID_GBR_IH_SET);
+
+	mdelay(20);
+	psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64),
+		     0x80000000, 0x8000FFFF, false);
+}
+
 static int psp_v3_1_ring_create(struct psp_context *psp,
 				enum psp_ring_type ring_type)
 {
@@ -260,6 +294,8 @@ static int psp_v3_1_ring_create(struct psp_context *psp,
 	struct psp_ring *ring = &psp->km_ring;
 	struct amdgpu_device *adev = psp->adev;
 
+	psp_v3_1_reroute_ih(psp);
+
 	/* Write low address of the ring to C2PMSG_69 */
 	psp_ring_reg = lower_32_bits(ring->ring_mem_mc_addr);
 	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_69, psp_ring_reg);
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
index cca3552b36ed..36196372e8db 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
@@ -870,8 +870,8 @@ static int sdma_v2_4_sw_init(void *handle)
 		r = amdgpu_ring_init(adev, ring, 1024,
 				     &adev->sdma.trap_irq,
 				     (i == 0) ?
-				     AMDGPU_SDMA_IRQ_TRAP0 :
-				     AMDGPU_SDMA_IRQ_TRAP1);
+				     AMDGPU_SDMA_IRQ_INSTANCE0 :
+				     AMDGPU_SDMA_IRQ_INSTANCE1);
 		if (r)
 			return r;
 	}
@@ -1006,7 +1006,7 @@ static int sdma_v2_4_set_trap_irq_state(struct amdgpu_device *adev,
 	u32 sdma_cntl;
 
 	switch (type) {
-	case AMDGPU_SDMA_IRQ_TRAP0:
+	case AMDGPU_SDMA_IRQ_INSTANCE0:
 		switch (state) {
 		case AMDGPU_IRQ_STATE_DISABLE:
 			sdma_cntl = RREG32(mmSDMA0_CNTL + SDMA0_REGISTER_OFFSET);
@@ -1022,7 +1022,7 @@ static int sdma_v2_4_set_trap_irq_state(struct amdgpu_device *adev,
 			break;
 		}
 		break;
-	case AMDGPU_SDMA_IRQ_TRAP1:
+	case AMDGPU_SDMA_IRQ_INSTANCE1:
 		switch (state) {
 		case AMDGPU_IRQ_STATE_DISABLE:
 			sdma_cntl = RREG32(mmSDMA0_CNTL + SDMA1_REGISTER_OFFSET);
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
index 0ce8331baeb2..6d39544e7829 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
@@ -1154,8 +1154,8 @@ static int sdma_v3_0_sw_init(void *handle)
 		r = amdgpu_ring_init(adev, ring, 1024,
 				     &adev->sdma.trap_irq,
 				     (i == 0) ?
-				     AMDGPU_SDMA_IRQ_TRAP0 :
-				     AMDGPU_SDMA_IRQ_TRAP1);
+				     AMDGPU_SDMA_IRQ_INSTANCE0 :
+				     AMDGPU_SDMA_IRQ_INSTANCE1);
 		if (r)
 			return r;
 	}
@@ -1340,7 +1340,7 @@ static int sdma_v3_0_set_trap_irq_state(struct amdgpu_device *adev,
 	u32 sdma_cntl;
 
 	switch (type) {
-	case AMDGPU_SDMA_IRQ_TRAP0:
+	case AMDGPU_SDMA_IRQ_INSTANCE0:
 		switch (state) {
 		case AMDGPU_IRQ_STATE_DISABLE:
 			sdma_cntl = RREG32(mmSDMA0_CNTL + SDMA0_REGISTER_OFFSET);
@@ -1356,7 +1356,7 @@ static int sdma_v3_0_set_trap_irq_state(struct amdgpu_device *adev,
 			break;
 		}
 		break;
-	case AMDGPU_SDMA_IRQ_TRAP1:
+	case AMDGPU_SDMA_IRQ_INSTANCE1:
 		switch (state) {
 		case AMDGPU_IRQ_STATE_DISABLE:
 			sdma_cntl = RREG32(mmSDMA0_CNTL + SDMA1_REGISTER_OFFSET);
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
index c816e55d43a9..9c88ce513d78 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
@@ -41,6 +41,8 @@
 #include "ivsrcid/sdma0/irqsrcs_sdma0_4_0.h"
 #include "ivsrcid/sdma1/irqsrcs_sdma1_4_0.h"
 
+#include "amdgpu_ras.h"
+
 MODULE_FIRMWARE("amdgpu/vega10_sdma.bin");
 MODULE_FIRMWARE("amdgpu/vega10_sdma1.bin");
 MODULE_FIRMWARE("amdgpu/vega12_sdma.bin");
@@ -154,7 +156,6 @@ static const struct soc15_reg_golden golden_settings_sdma0_4_2[] =
 	SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_RLC7_RB_RPTR_ADDR_LO, 0xfffffffd, 0x00000001),
 	SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_RLC7_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
 	SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_UTCL1_PAGE, 0x000003ff, 0x000003c0),
-	SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_UTCL1_WATERMK, 0xFE000000, 0x00000000),
 };
 
 static const struct soc15_reg_golden golden_settings_sdma1_4_2[] = {
@@ -184,7 +185,6 @@ static const struct soc15_reg_golden golden_settings_sdma1_4_2[] = {
 	SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_RLC7_RB_RPTR_ADDR_LO, 0xfffffffd, 0x00000001),
 	SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_RLC7_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
 	SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_UTCL1_PAGE, 0x000003ff, 0x000003c0),
-	SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_UTCL1_WATERMK, 0xFE000000, 0x00000000),
 };
 
 static const struct soc15_reg_golden golden_settings_sdma_rv1[] =
@@ -849,7 +849,7 @@ static void sdma_v4_0_gfx_resume(struct amdgpu_device *adev, unsigned int i)
 	wptr_poll_cntl = RREG32_SDMA(i, mmSDMA0_GFX_RB_WPTR_POLL_CNTL);
 	wptr_poll_cntl = REG_SET_FIELD(wptr_poll_cntl,
 				       SDMA0_GFX_RB_WPTR_POLL_CNTL,
-				       F32_POLL_ENABLE, amdgpu_sriov_vf(adev));
+				       F32_POLL_ENABLE, amdgpu_sriov_vf(adev)? 1 : 0);
 	WREG32_SDMA(i, mmSDMA0_GFX_RB_WPTR_POLL_CNTL, wptr_poll_cntl);
 
 	/* enable DMA RB */
@@ -940,7 +940,7 @@ static void sdma_v4_0_page_resume(struct amdgpu_device *adev, unsigned int i)
 	wptr_poll_cntl = RREG32_SDMA(i, mmSDMA0_PAGE_RB_WPTR_POLL_CNTL);
 	wptr_poll_cntl = REG_SET_FIELD(wptr_poll_cntl,
 				       SDMA0_PAGE_RB_WPTR_POLL_CNTL,
-				       F32_POLL_ENABLE, amdgpu_sriov_vf(adev));
+				       F32_POLL_ENABLE, amdgpu_sriov_vf(adev)? 1 : 0);
 	WREG32_SDMA(i, mmSDMA0_PAGE_RB_WPTR_POLL_CNTL, wptr_poll_cntl);
 
 	/* enable DMA RB */
@@ -1493,6 +1493,87 @@ static int sdma_v4_0_early_init(void *handle)
 	return 0;
 }
 
+static int sdma_v4_0_process_ras_data_cb(struct amdgpu_device *adev,
+		struct amdgpu_iv_entry *entry);
+
+static int sdma_v4_0_late_init(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	struct ras_common_if **ras_if = &adev->sdma.ras_if;
+	struct ras_ih_if ih_info = {
+		.cb = sdma_v4_0_process_ras_data_cb,
+	};
+	struct ras_fs_if fs_info = {
+		.sysfs_name = "sdma_err_count",
+		.debugfs_name = "sdma_err_inject",
+	};
+	struct ras_common_if ras_block = {
+		.block = AMDGPU_RAS_BLOCK__SDMA,
+		.type = AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE,
+		.sub_block_index = 0,
+		.name = "sdma",
+	};
+	int r;
+
+	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__SDMA)) {
+		amdgpu_ras_feature_enable_on_boot(adev, &ras_block, 0);
+		return 0;
+	}
+
+	/* handle resume path. */
+	if (*ras_if)
+		goto resume;
+
+	*ras_if = kmalloc(sizeof(**ras_if), GFP_KERNEL);
+	if (!*ras_if)
+		return -ENOMEM;
+
+	**ras_if = ras_block;
+
+	r = amdgpu_ras_feature_enable_on_boot(adev, *ras_if, 1);
+	if (r)
+		goto feature;
+
+	ih_info.head = **ras_if;
+	fs_info.head = **ras_if;
+
+	r = amdgpu_ras_interrupt_add_handler(adev, &ih_info);
+	if (r)
+		goto interrupt;
+
+	r = amdgpu_ras_debugfs_create(adev, &fs_info);
+	if (r)
+		goto debugfs;
+
+	r = amdgpu_ras_sysfs_create(adev, &fs_info);
+	if (r)
+		goto sysfs;
+resume:
+	r = amdgpu_irq_get(adev, &adev->sdma.ecc_irq, AMDGPU_SDMA_IRQ_INSTANCE0);
+	if (r)
+		goto irq;
+
+	r = amdgpu_irq_get(adev, &adev->sdma.ecc_irq, AMDGPU_SDMA_IRQ_INSTANCE1);
+	if (r) {
+		amdgpu_irq_put(adev, &adev->sdma.ecc_irq, AMDGPU_SDMA_IRQ_INSTANCE0);
+		goto irq;
+	}
+
+	return 0;
+irq:
+	amdgpu_ras_sysfs_remove(adev, *ras_if);
+sysfs:
+	amdgpu_ras_debugfs_remove(adev, *ras_if);
+debugfs:
+	amdgpu_ras_interrupt_remove_handler(adev, &ih_info);
+interrupt:
+	amdgpu_ras_feature_enable(adev, *ras_if, 0);
+feature:
+	kfree(*ras_if);
+	*ras_if = NULL;
+	return -EINVAL;
+}
+
 static int sdma_v4_0_sw_init(void *handle)
 {
 	struct amdgpu_ring *ring;
@@ -1511,6 +1592,18 @@ static int sdma_v4_0_sw_init(void *handle)
 	if (r)
 		return r;
 
+	/* SDMA SRAM ECC event */
+	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_SDMA0, SDMA0_4_0__SRCID__SDMA_SRAM_ECC,
+			&adev->sdma.ecc_irq);
+	if (r)
+		return r;
+
+	/* SDMA SRAM ECC event */
+	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_SDMA1, SDMA1_4_0__SRCID__SDMA_SRAM_ECC,
+			&adev->sdma.ecc_irq);
+	if (r)
+		return r;
+
 	for (i = 0; i < adev->sdma.num_instances; i++) {
 		ring = &adev->sdma.instance[i].ring;
 		ring->ring_obj = NULL;
@@ -1526,8 +1619,8 @@ static int sdma_v4_0_sw_init(void *handle)
 		r = amdgpu_ring_init(adev, ring, 1024,
 				     &adev->sdma.trap_irq,
 				     (i == 0) ?
-				     AMDGPU_SDMA_IRQ_TRAP0 :
-				     AMDGPU_SDMA_IRQ_TRAP1);
+				     AMDGPU_SDMA_IRQ_INSTANCE0 :
+				     AMDGPU_SDMA_IRQ_INSTANCE1);
 		if (r)
 			return r;
 
@@ -1546,8 +1639,8 @@ static int sdma_v4_0_sw_init(void *handle)
 			r = amdgpu_ring_init(adev, ring, 1024,
 					     &adev->sdma.trap_irq,
 					     (i == 0) ?
-					     AMDGPU_SDMA_IRQ_TRAP0 :
-					     AMDGPU_SDMA_IRQ_TRAP1);
+					     AMDGPU_SDMA_IRQ_INSTANCE0 :
+					     AMDGPU_SDMA_IRQ_INSTANCE1);
 			if (r)
 				return r;
 		}
@@ -1561,6 +1654,22 @@ static int sdma_v4_0_sw_fini(void *handle)
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	int i;
 
+	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__SDMA) &&
+			adev->sdma.ras_if) {
+		struct ras_common_if *ras_if = adev->sdma.ras_if;
+		struct ras_ih_if ih_info = {
+			.head = *ras_if,
+		};
+
+		/*remove fs first*/
+		amdgpu_ras_debugfs_remove(adev, ras_if);
+		amdgpu_ras_sysfs_remove(adev, ras_if);
+		/*remove the IH*/
+		amdgpu_ras_interrupt_remove_handler(adev, &ih_info);
+		amdgpu_ras_feature_enable(adev, ras_if, 0);
+		kfree(ras_if);
+	}
+
 	for (i = 0; i < adev->sdma.num_instances; i++) {
 		amdgpu_ring_fini(&adev->sdma.instance[i].ring);
 		if (adev->sdma.has_page_queue)
@@ -1598,6 +1707,9 @@ static int sdma_v4_0_hw_fini(void *handle)
 	if (amdgpu_sriov_vf(adev))
 		return 0;
 
+	amdgpu_irq_put(adev, &adev->sdma.ecc_irq, AMDGPU_SDMA_IRQ_INSTANCE0);
+	amdgpu_irq_put(adev, &adev->sdma.ecc_irq, AMDGPU_SDMA_IRQ_INSTANCE1);
+
 	sdma_v4_0_ctx_switch_enable(adev, false);
 	sdma_v4_0_enable(adev, false);
 
@@ -1666,13 +1778,12 @@ static int sdma_v4_0_set_trap_irq_state(struct amdgpu_device *adev,
 					unsigned type,
 					enum amdgpu_interrupt_state state)
 {
-	unsigned int instance = (type == AMDGPU_SDMA_IRQ_TRAP0) ? 0 : 1;
 	u32 sdma_cntl;
 
-	sdma_cntl = RREG32_SDMA(instance, mmSDMA0_CNTL);
+	sdma_cntl = RREG32_SDMA(type, mmSDMA0_CNTL);
 	sdma_cntl = REG_SET_FIELD(sdma_cntl, SDMA0_CNTL, TRAP_ENABLE,
 		       state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0);
-	WREG32_SDMA(instance, mmSDMA0_CNTL, sdma_cntl);
+	WREG32_SDMA(type, mmSDMA0_CNTL, sdma_cntl);
 
 	return 0;
 }
@@ -1714,6 +1825,58 @@ static int sdma_v4_0_process_trap_irq(struct amdgpu_device *adev,
 	return 0;
 }
 
+static int sdma_v4_0_process_ras_data_cb(struct amdgpu_device *adev,
+		struct amdgpu_iv_entry *entry)
+{
+	uint32_t instance, err_source;
+
+	switch (entry->client_id) {
+	case SOC15_IH_CLIENTID_SDMA0:
+		instance = 0;
+		break;
+	case SOC15_IH_CLIENTID_SDMA1:
+		instance = 1;
+		break;
+	default:
+		return 0;
+	}
+
+	switch (entry->src_id) {
+	case SDMA0_4_0__SRCID__SDMA_SRAM_ECC:
+		err_source = 0;
+		break;
+	case SDMA0_4_0__SRCID__SDMA_ECC:
+		err_source = 1;
+		break;
+	default:
+		return 0;
+	}
+
+	kgd2kfd_set_sram_ecc_flag(adev->kfd.dev);
+
+	amdgpu_ras_reset_gpu(adev, 0);
+
+	return AMDGPU_RAS_UE;
+}
+
+static int sdma_v4_0_process_ecc_irq(struct amdgpu_device *adev,
+				      struct amdgpu_irq_src *source,
+				      struct amdgpu_iv_entry *entry)
+{
+	struct ras_common_if *ras_if = adev->sdma.ras_if;
+	struct ras_dispatch_if ih_data = {
+		.entry = entry,
+	};
+
+	if (!ras_if)
+		return 0;
+
+	ih_data.head = *ras_if;
+
+	amdgpu_ras_interrupt_dispatch(adev, &ih_data);
+	return 0;
+}
+
 static int sdma_v4_0_process_illegal_inst_irq(struct amdgpu_device *adev,
 					      struct amdgpu_irq_src *source,
 					      struct amdgpu_iv_entry *entry)
@@ -1741,6 +1904,25 @@ static int sdma_v4_0_process_illegal_inst_irq(struct amdgpu_device *adev,
 	return 0;
 }
 
+static int sdma_v4_0_set_ecc_irq_state(struct amdgpu_device *adev,
+					struct amdgpu_irq_src *source,
+					unsigned type,
+					enum amdgpu_interrupt_state state)
+{
+	u32 sdma_edc_config;
+
+	u32 reg_offset = (type == AMDGPU_SDMA_IRQ_INSTANCE0) ?
+		sdma_v4_0_get_reg_offset(adev, 0, mmSDMA0_EDC_CONFIG) :
+		sdma_v4_0_get_reg_offset(adev, 1, mmSDMA0_EDC_CONFIG);
+
+	sdma_edc_config = RREG32(reg_offset);
+	sdma_edc_config = REG_SET_FIELD(sdma_edc_config, SDMA0_EDC_CONFIG, ECC_INT_ENABLE,
+		       state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0);
+	WREG32(reg_offset, sdma_edc_config);
+
+	return 0;
+}
+
 static void sdma_v4_0_update_medium_grain_clock_gating(
 		struct amdgpu_device *adev,
 		bool enable)
@@ -1906,7 +2088,7 @@ static void sdma_v4_0_get_clockgating_state(void *handle, u32 *flags)
 const struct amd_ip_funcs sdma_v4_0_ip_funcs = {
 	.name = "sdma_v4_0",
 	.early_init = sdma_v4_0_early_init,
-	.late_init = NULL,
+	.late_init = sdma_v4_0_late_init,
 	.sw_init = sdma_v4_0_sw_init,
 	.sw_fini = sdma_v4_0_sw_fini,
 	.hw_init = sdma_v4_0_hw_init,
@@ -2008,11 +2190,20 @@ static const struct amdgpu_irq_src_funcs sdma_v4_0_illegal_inst_irq_funcs = {
 	.process = sdma_v4_0_process_illegal_inst_irq,
 };
 
+static const struct amdgpu_irq_src_funcs sdma_v4_0_ecc_irq_funcs = {
+	.set = sdma_v4_0_set_ecc_irq_state,
+	.process = sdma_v4_0_process_ecc_irq,
+};
+
+
+
 static void sdma_v4_0_set_irq_funcs(struct amdgpu_device *adev)
 {
 	adev->sdma.trap_irq.num_types = AMDGPU_SDMA_IRQ_LAST;
 	adev->sdma.trap_irq.funcs = &sdma_v4_0_trap_irq_funcs;
 	adev->sdma.illegal_inst_irq.funcs = &sdma_v4_0_illegal_inst_irq_funcs;
+	adev->sdma.ecc_irq.num_types = AMDGPU_SDMA_IRQ_LAST;
+	adev->sdma.ecc_irq.funcs = &sdma_v4_0_ecc_irq_funcs;
 }
 
 /**
@@ -2077,8 +2268,8 @@ static const struct amdgpu_buffer_funcs sdma_v4_0_buffer_funcs = {
 static void sdma_v4_0_set_buffer_funcs(struct amdgpu_device *adev)
 {
 	adev->mman.buffer_funcs = &sdma_v4_0_buffer_funcs;
-	if (adev->sdma.has_page_queue)
-		adev->mman.buffer_funcs_ring = &adev->sdma.instance[0].page;
+	if (adev->sdma.has_page_queue && adev->sdma.num_instances > 1)
+		adev->mman.buffer_funcs_ring = &adev->sdma.instance[1].page;
 	else
 		adev->mman.buffer_funcs_ring = &adev->sdma.instance[0].ring;
 }
@@ -2097,15 +2288,22 @@ static void sdma_v4_0_set_vm_pte_funcs(struct amdgpu_device *adev)
 	unsigned i;
 
 	adev->vm_manager.vm_pte_funcs = &sdma_v4_0_vm_pte_funcs;
-	for (i = 0; i < adev->sdma.num_instances; i++) {
-		if (adev->sdma.has_page_queue)
+	if (adev->sdma.has_page_queue && adev->sdma.num_instances > 1) {
+		for (i = 1; i < adev->sdma.num_instances; i++) {
 			sched = &adev->sdma.instance[i].page.sched;
-		else
+			adev->vm_manager.vm_pte_rqs[i - 1] =
+				&sched->sched_rq[DRM_SCHED_PRIORITY_KERNEL];
+		}
+		adev->vm_manager.vm_pte_num_rqs = adev->sdma.num_instances - 1;
+		adev->vm_manager.page_fault = &adev->sdma.instance[0].page;
+	} else {
+		for (i = 0; i < adev->sdma.num_instances; i++) {
 			sched = &adev->sdma.instance[i].ring.sched;
-		adev->vm_manager.vm_pte_rqs[i] =
-			&sched->sched_rq[DRM_SCHED_PRIORITY_KERNEL];
+			adev->vm_manager.vm_pte_rqs[i] =
+				&sched->sched_rq[DRM_SCHED_PRIORITY_KERNEL];
+		}
+		adev->vm_manager.vm_pte_num_rqs = adev->sdma.num_instances;
 	}
-	adev->vm_manager.vm_pte_num_rqs = adev->sdma.num_instances;
 }
 
 const struct amdgpu_ip_block_version sdma_v4_0_ip_block = {
diff --git a/drivers/gpu/drm/amd/amdgpu/si_dma.c b/drivers/gpu/drm/amd/amdgpu/si_dma.c
index f15f196684ba..3eeefd40dae0 100644
--- a/drivers/gpu/drm/amd/amdgpu/si_dma.c
+++ b/drivers/gpu/drm/amd/amdgpu/si_dma.c
@@ -503,8 +503,8 @@ static int si_dma_sw_init(void *handle)
 		r = amdgpu_ring_init(adev, ring, 1024,
 				     &adev->sdma.trap_irq,
 				     (i == 0) ?
-				     AMDGPU_SDMA_IRQ_TRAP0 :
-				     AMDGPU_SDMA_IRQ_TRAP1);
+				     AMDGPU_SDMA_IRQ_INSTANCE0 :
+				     AMDGPU_SDMA_IRQ_INSTANCE1);
 		if (r)
 			return r;
 	}
@@ -591,7 +591,7 @@ static int si_dma_set_trap_irq_state(struct amdgpu_device *adev,
 	u32 sdma_cntl;
 
 	switch (type) {
-	case AMDGPU_SDMA_IRQ_TRAP0:
+	case AMDGPU_SDMA_IRQ_INSTANCE0:
 		switch (state) {
 		case AMDGPU_IRQ_STATE_DISABLE:
 			sdma_cntl = RREG32(DMA_CNTL + DMA0_REGISTER_OFFSET);
@@ -607,7 +607,7 @@ static int si_dma_set_trap_irq_state(struct amdgpu_device *adev,
 			break;
 		}
 		break;
-	case AMDGPU_SDMA_IRQ_TRAP1:
+	case AMDGPU_SDMA_IRQ_INSTANCE1:
 		switch (state) {
 		case AMDGPU_IRQ_STATE_DISABLE:
 			sdma_cntl = RREG32(DMA_CNTL + DMA1_REGISTER_OFFSET);
diff --git a/drivers/gpu/drm/amd/amdgpu/si_dpm.c b/drivers/gpu/drm/amd/amdgpu/si_dpm.c
index 41e01a7f57a4..d57e75e5c71f 100644
--- a/drivers/gpu/drm/amd/amdgpu/si_dpm.c
+++ b/drivers/gpu/drm/amd/amdgpu/si_dpm.c
@@ -4098,14 +4098,13 @@ static int si_notify_smc_display_change(struct amdgpu_device *adev,
 
 static void si_program_response_times(struct amdgpu_device *adev)
 {
-	u32 voltage_response_time, backbias_response_time, acpi_delay_time, vbi_time_out;
+	u32 voltage_response_time, acpi_delay_time, vbi_time_out;
 	u32 vddc_dly, acpi_dly, vbi_dly;
 	u32 reference_clock;
 
 	si_write_smc_soft_register(adev, SI_SMC_SOFT_REGISTER_mvdd_chg_time, 1);
 
 	voltage_response_time = (u32)adev->pm.dpm.voltage_response_time;
-	backbias_response_time = (u32)adev->pm.dpm.backbias_response_time;
 
 	if (voltage_response_time == 0)
 		voltage_response_time = 1000;
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index ed89a101f73f..4900e4958dec 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -63,6 +63,7 @@
 #include "vcn_v1_0.h"
 #include "dce_virtual.h"
 #include "mxgpu_ai.h"
+#include "amdgpu_smu.h"
 
 #define mmMP0_MISC_CGTT_CTRL0                                                                   0x01b9
 #define mmMP0_MISC_CGTT_CTRL0_BASE_IDX                                                          0
@@ -392,6 +393,7 @@ void soc15_program_register_sequence(struct amdgpu_device *adev,
 static int soc15_asic_mode1_reset(struct amdgpu_device *adev)
 {
 	u32 i;
+	int ret = 0;
 
 	amdgpu_atombios_scratch_regs_engine_hung(adev, true);
 
@@ -402,7 +404,9 @@ static int soc15_asic_mode1_reset(struct amdgpu_device *adev)
 
 	pci_save_state(adev->pdev);
 
-	psp_gpu_reset(adev);
+	ret = psp_gpu_reset(adev);
+	if (ret)
+		dev_err(adev->dev, "GPU mode1 reset failed\n");
 
 	pci_restore_state(adev->pdev);
 
@@ -417,7 +421,7 @@ static int soc15_asic_mode1_reset(struct amdgpu_device *adev)
 
 	amdgpu_atombios_scratch_regs_engine_hung(adev, false);
 
-	return 0;
+	return ret;
 }
 
 static int soc15_asic_get_baco_capability(struct amdgpu_device *adev, bool *cap)
@@ -451,6 +455,8 @@ static int soc15_asic_baco_reset(struct amdgpu_device *adev)
 
 	dev_info(adev->dev, "GPU BACO reset\n");
 
+	adev->in_baco_reset = 1;
+
 	return 0;
 }
 
@@ -461,8 +467,15 @@ static int soc15_asic_reset(struct amdgpu_device *adev)
 
 	switch (adev->asic_type) {
 	case CHIP_VEGA10:
+	case CHIP_VEGA12:
 		soc15_asic_get_baco_capability(adev, &baco_reset);
 		break;
+	case CHIP_VEGA20:
+		if (adev->psp.sos_fw_version >= 0x80067)
+			soc15_asic_get_baco_capability(adev, &baco_reset);
+		else
+			baco_reset = false;
+		break;
 	default:
 		baco_reset = false;
 		break;
@@ -602,8 +615,12 @@ int soc15_set_ip_blocks(struct amdgpu_device *adev)
 		}
 		amdgpu_device_ip_block_add(adev, &gfx_v9_0_ip_block);
 		amdgpu_device_ip_block_add(adev, &sdma_v4_0_ip_block);
-		if (!amdgpu_sriov_vf(adev))
-			amdgpu_device_ip_block_add(adev, &pp_smu_ip_block);
+		if (!amdgpu_sriov_vf(adev)) {
+			if (is_support_sw_smu(adev))
+				amdgpu_device_ip_block_add(adev, &smu_v11_0_ip_block);
+			else
+				amdgpu_device_ip_block_add(adev, &pp_smu_ip_block);
+		}
 		if (adev->enable_virtual_display || amdgpu_sriov_vf(adev))
 			amdgpu_device_ip_block_add(adev, &dce_virtual_ip_block);
 #if defined(CONFIG_DRM_AMD_DC)
@@ -884,7 +901,8 @@ static int soc15_common_early_init(void *handle)
 
 			adev->pg_flags = AMD_PG_SUPPORT_SDMA | AMD_PG_SUPPORT_VCN;
 		} else if (adev->pdev->device == 0x15d8) {
-			adev->cg_flags = AMD_CG_SUPPORT_GFX_MGLS |
+			adev->cg_flags = AMD_CG_SUPPORT_GFX_MGCG |
+				AMD_CG_SUPPORT_GFX_MGLS |
 				AMD_CG_SUPPORT_GFX_CP_LS |
 				AMD_CG_SUPPORT_GFX_3D_CGCG |
 				AMD_CG_SUPPORT_GFX_3D_CGLS |
@@ -927,7 +945,7 @@ static int soc15_common_early_init(void *handle)
 			adev->pg_flags = AMD_PG_SUPPORT_SDMA | AMD_PG_SUPPORT_VCN;
 		}
 
-		if (adev->powerplay.pp_feature & PP_GFXOFF_MASK)
+		if (adev->pm.pp_feature & PP_GFXOFF_MASK)
 			adev->pg_flags |= AMD_PG_SUPPORT_GFX_PG |
 				AMD_PG_SUPPORT_CP |
 				AMD_PG_SUPPORT_RLC_SMU_HS;
diff --git a/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h b/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h
new file mode 100644
index 000000000000..0b4e7b55595a
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h
@@ -0,0 +1,108 @@
+/****************************************************************************\
+* 
+*  File Name      ta_ras_if.h
+*  Project        AMD PSP SW IP Module
+*
+*  Description    Interface to the RAS Trusted Application
+*
+*  Copyright 2019 Advanced Micro Devices, Inc.
+*
+* Permission is hereby granted, free of charge, to any person obtaining a copy of this software 
+* and associated documentation files (the "Software"), to deal in the Software without restriction,
+* including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense,
+* and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so,
+* subject to the following conditions:
+*
+* The above copyright notice and this permission notice shall be included in all copies or substantial
+* portions of the Software.
+*
+* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+* THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+* OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+* ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+* OTHER DEALINGS IN THE SOFTWARE.
+*/
+#ifndef _TA_RAS_IF_H
+#define _TA_RAS_IF_H
+
+/* Responses have bit 31 set */
+#define RSP_ID_MASK (1U << 31)
+#define RSP_ID(cmdId) (((uint32_t)(cmdId)) | RSP_ID_MASK)
+
+#define TA_NUM_BLOCK_MAX		14
+
+enum ras_command {
+	TA_RAS_COMMAND__ENABLE_FEATURES = 0,
+	TA_RAS_COMMAND__DISABLE_FEATURES,
+	TA_RAS_COMMAND__TRIGGER_ERROR,
+};
+
+enum ta_ras_status {
+	TA_RAS_STATUS__SUCCESS				= 0x00,
+	TA_RAS_STATUS__RESET_NEEDED			= 0x01,
+	TA_RAS_STATUS__ERROR_INVALID_PARAMETER		= 0x02,
+	TA_RAS_STATUS__ERROR_RAS_NOT_AVAILABLE		= 0x03,
+	TA_RAS_STATUS__ERROR_RAS_DUPLICATE_CMD		= 0x04,
+	TA_RAS_STATUS__ERROR_INJECTION_FAILED		= 0x05
+};
+
+enum ta_ras_block {
+	TA_RAS_BLOCK__UMC = 0,
+	TA_RAS_BLOCK__SDMA,
+	TA_RAS_BLOCK__GFX,
+	TA_RAS_BLOCK__MMHUB,
+	TA_RAS_BLOCK__ATHUB,
+	TA_RAS_BLOCK__PCIE_BIF,
+	TA_RAS_BLOCK__HDP,
+	TA_RAS_BLOCK__XGMI_WAFL,
+	TA_RAS_BLOCK__DF,
+	TA_RAS_BLOCK__SMN,
+	TA_RAS_BLOCK__SEM,
+	TA_RAS_BLOCK__MP0,
+	TA_RAS_BLOCK__MP1,
+	TA_RAS_BLOCK__FUSE = (TA_NUM_BLOCK_MAX - 1),
+};
+
+enum ta_ras_error_type {
+	TA_RAS_ERROR__NONE				= 0,
+	TA_RAS_ERROR__PARITY				= 1,
+	TA_RAS_ERROR__SINGLE_CORRECTABLE		= 2,
+	TA_RAS_ERROR__MULTI_UNCORRECTABLE		= 4,
+	TA_RAS_ERROR__POISON				= 8
+};
+
+struct ta_ras_enable_features_input {
+	enum ta_ras_block       block_id;
+	enum ta_ras_error_type  error_type;
+};
+
+struct ta_ras_disable_features_input {
+	enum ta_ras_block       block_id;
+	enum ta_ras_error_type  error_type;
+};
+
+struct ta_ras_trigger_error_input {
+	enum ta_ras_block		block_id;
+	enum ta_ras_error_type		inject_error_type;
+	uint32_t			sub_block_index;
+	uint64_t			address;
+	uint64_t			value;
+};
+
+union ta_ras_cmd_input {
+	struct ta_ras_enable_features_input	enable_features;
+	struct ta_ras_disable_features_input	disable_features;
+	struct ta_ras_trigger_error_input	trigger_error;
+};
+
+struct ta_ras_shared_memory {
+	uint32_t			cmd_id;
+	uint32_t			resp_id;
+	enum ta_ras_status		ras_status;
+	uint32_t			reserved;
+	union ta_ras_cmd_input		ras_in_message;
+};
+
+#endif // TL_RAS_IF_H_
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c
index bed78a778e3f..40363ca6c5f1 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c
@@ -283,7 +283,7 @@ static int vce_v2_0_stop(struct amdgpu_device *adev)
 	}
 
 	if (vce_v2_0_wait_for_idle(adev)) {
-		DRM_INFO("VCE is busy, Can't set clock gateing");
+		DRM_INFO("VCE is busy, Can't set clock gating");
 		return 0;
 	}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
index aadc3e66ebd7..f3f5938430d4 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
@@ -382,6 +382,7 @@ static int vce_v4_0_start(struct amdgpu_device *adev)
 static int vce_v4_0_stop(struct amdgpu_device *adev)
 {
 
+	/* Disable VCPU */
 	WREG32_P(SOC15_REG_OFFSET(VCE, 0, mmVCE_VCPU_CNTL), 0, ~0x200001);
 
 	/* hold on ECPU */
@@ -389,8 +390,8 @@ static int vce_v4_0_stop(struct amdgpu_device *adev)
 			VCE_SOFT_RESET__ECPU_SOFT_RESET_MASK,
 			~VCE_SOFT_RESET__ECPU_SOFT_RESET_MASK);
 
-	/* clear BUSY flag */
-	WREG32_P(SOC15_REG_OFFSET(VCE, 0, mmVCE_STATUS), 0, ~VCE_STATUS__JOB_BUSY_MASK);
+	/* clear VCE_STATUS */
+	WREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_STATUS), 0);
 
 	/* Set Clock-Gating off */
 	/* if (adev->cg_flags & AMD_CG_SUPPORT_VCE_MGCG)
@@ -922,6 +923,7 @@ static int vce_v4_0_set_clockgating_state(void *handle,
 
 	return 0;
 }
+#endif
 
 static int vce_v4_0_set_powergating_state(void *handle,
 					  enum amd_powergating_state state)
@@ -935,16 +937,11 @@ static int vce_v4_0_set_powergating_state(void *handle,
 	 */
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	if (!(adev->pg_flags & AMD_PG_SUPPORT_VCE))
-		return 0;
-
 	if (state == AMD_PG_STATE_GATE)
-		/* XXX do we need a vce_v4_0_stop()? */
-		return 0;
+		return vce_v4_0_stop(adev);
 	else
 		return vce_v4_0_start(adev);
 }
-#endif
 
 static void vce_v4_0_ring_emit_ib(struct amdgpu_ring *ring, struct amdgpu_job *job,
 					struct amdgpu_ib *ib, uint32_t flags)
@@ -1059,7 +1056,7 @@ const struct amd_ip_funcs vce_v4_0_ip_funcs = {
 	.soft_reset = NULL /* vce_v4_0_soft_reset */,
 	.post_soft_reset = NULL /* vce_v4_0_post_soft_reset */,
 	.set_clockgating_state = vce_v4_0_set_clockgating_state,
-	.set_powergating_state = NULL /* vce_v4_0_set_powergating_state */,
+	.set_powergating_state = vce_v4_0_set_powergating_state,
 };
 
 static const struct amdgpu_ring_funcs vce_v4_0_ring_vm_funcs = {
diff --git a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c
index 6d1f804277f8..1b2f69a9a24e 100644
--- a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c
@@ -136,6 +136,25 @@ static uint32_t vega10_ih_rb_cntl(struct amdgpu_ih_ring *ih, uint32_t ih_rb_cntl
 	return ih_rb_cntl;
 }
 
+static uint32_t vega10_ih_doorbell_rptr(struct amdgpu_ih_ring *ih)
+{
+	u32 ih_doorbell_rtpr = 0;
+
+	if (ih->use_doorbell) {
+		ih_doorbell_rtpr = REG_SET_FIELD(ih_doorbell_rtpr,
+						 IH_DOORBELL_RPTR, OFFSET,
+						 ih->doorbell_index);
+		ih_doorbell_rtpr = REG_SET_FIELD(ih_doorbell_rtpr,
+						 IH_DOORBELL_RPTR,
+						 ENABLE, 1);
+	} else {
+		ih_doorbell_rtpr = REG_SET_FIELD(ih_doorbell_rtpr,
+						 IH_DOORBELL_RPTR,
+						 ENABLE, 0);
+	}
+	return ih_doorbell_rtpr;
+}
+
 /**
  * vega10_ih_irq_init - init and enable the interrupt ring
  *
@@ -150,8 +169,8 @@ static uint32_t vega10_ih_rb_cntl(struct amdgpu_ih_ring *ih, uint32_t ih_rb_cntl
 static int vega10_ih_irq_init(struct amdgpu_device *adev)
 {
 	struct amdgpu_ih_ring *ih;
+	u32 ih_rb_cntl;
 	int ret = 0;
-	u32 ih_rb_cntl, ih_doorbell_rtpr;
 	u32 tmp;
 
 	/* disable irqs */
@@ -177,23 +196,11 @@ static int vega10_ih_irq_init(struct amdgpu_device *adev)
 		     upper_32_bits(ih->wptr_addr) & 0xFFFF);
 
 	/* set rptr, wptr to 0 */
-	WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR, 0);
 	WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR, 0);
+	WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR, 0);
 
-	ih_doorbell_rtpr = RREG32_SOC15(OSSSYS, 0, mmIH_DOORBELL_RPTR);
-	if (adev->irq.ih.use_doorbell) {
-		ih_doorbell_rtpr = REG_SET_FIELD(ih_doorbell_rtpr,
-						 IH_DOORBELL_RPTR, OFFSET,
-						 adev->irq.ih.doorbell_index);
-		ih_doorbell_rtpr = REG_SET_FIELD(ih_doorbell_rtpr,
-						 IH_DOORBELL_RPTR,
-						 ENABLE, 1);
-	} else {
-		ih_doorbell_rtpr = REG_SET_FIELD(ih_doorbell_rtpr,
-						 IH_DOORBELL_RPTR,
-						 ENABLE, 0);
-	}
-	WREG32_SOC15(OSSSYS, 0, mmIH_DOORBELL_RPTR, ih_doorbell_rtpr);
+	WREG32_SOC15(OSSSYS, 0, mmIH_DOORBELL_RPTR,
+		     vega10_ih_doorbell_rptr(ih));
 
 	ih = &adev->irq.ih1;
 	if (ih->ring_size) {
@@ -203,11 +210,18 @@ static int vega10_ih_irq_init(struct amdgpu_device *adev)
 
 		ih_rb_cntl = RREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING1);
 		ih_rb_cntl = vega10_ih_rb_cntl(ih, ih_rb_cntl);
+		ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL,
+					   WPTR_OVERFLOW_ENABLE, 0);
+		ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL,
+					   RB_FULL_DRAIN_ENABLE, 1);
 		WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING1, ih_rb_cntl);
 
 		/* set rptr, wptr to 0 */
-		WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR_RING1, 0);
 		WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR_RING1, 0);
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR_RING1, 0);
+
+		WREG32_SOC15(OSSSYS, 0, mmIH_DOORBELL_RPTR_RING1,
+			     vega10_ih_doorbell_rptr(ih));
 	}
 
 	ih = &adev->irq.ih2;
@@ -216,13 +230,16 @@ static int vega10_ih_irq_init(struct amdgpu_device *adev)
 		WREG32_SOC15(OSSSYS, 0, mmIH_RB_BASE_HI_RING2,
 			     (ih->gpu_addr >> 40) & 0xff);
 
-		ih_rb_cntl = RREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING1);
+		ih_rb_cntl = RREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING2);
 		ih_rb_cntl = vega10_ih_rb_cntl(ih, ih_rb_cntl);
 		WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING2, ih_rb_cntl);
 
 		/* set rptr, wptr to 0 */
-		WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR_RING2, 0);
 		WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR_RING2, 0);
+		WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR_RING2, 0);
+
+		WREG32_SOC15(OSSSYS, 0, mmIH_DOORBELL_RPTR_RING2,
+			     vega10_ih_doorbell_rptr(ih));
 	}
 
 	tmp = RREG32_SOC15(OSSSYS, 0, mmIH_STORM_CLIENT_LIST_CNTL);
@@ -449,20 +466,23 @@ static int vega10_ih_sw_init(void *handle)
 	if (r)
 		return r;
 
-	if (adev->asic_type == CHIP_VEGA10) {
-		r = amdgpu_ih_ring_init(adev, &adev->irq.ih1, PAGE_SIZE, true);
-		if (r)
-			return r;
-
-		r = amdgpu_ih_ring_init(adev, &adev->irq.ih2, PAGE_SIZE, true);
-		if (r)
-			return r;
-	}
-
-	/* TODO add doorbell for IH1 & IH2 as well */
 	adev->irq.ih.use_doorbell = true;
 	adev->irq.ih.doorbell_index = adev->doorbell_index.ih << 1;
 
+	r = amdgpu_ih_ring_init(adev, &adev->irq.ih1, PAGE_SIZE, true);
+	if (r)
+		return r;
+
+	adev->irq.ih1.use_doorbell = true;
+	adev->irq.ih1.doorbell_index = (adev->doorbell_index.ih + 1) << 1;
+
+	r = amdgpu_ih_ring_init(adev, &adev->irq.ih2, PAGE_SIZE, true);
+	if (r)
+		return r;
+
+	adev->irq.ih2.use_doorbell = true;
+	adev->irq.ih2.doorbell_index = (adev->doorbell_index.ih + 2) << 1;
+
 	r = amdgpu_irq_init(adev);
 
 	return r;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index cf9a49f49d3a..c1e4d44d6137 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -467,6 +467,8 @@ struct kfd_dev *kgd2kfd_probe(struct kgd_dev *kgd,
 	memset(&kfd->doorbell_available_index, 0,
 		sizeof(kfd->doorbell_available_index));
 
+	atomic_set(&kfd->sram_ecc_flag, 0);
+
 	return kfd;
 }
 
@@ -492,9 +494,9 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 {
 	unsigned int size;
 
-	kfd->mec_fw_version = kfd->kfd2kgd->get_fw_version(kfd->kgd,
+	kfd->mec_fw_version = amdgpu_amdkfd_get_fw_version(kfd->kgd,
 			KGD_ENGINE_MEC1);
-	kfd->sdma_fw_version = kfd->kfd2kgd->get_fw_version(kfd->kgd,
+	kfd->sdma_fw_version = amdgpu_amdkfd_get_fw_version(kfd->kgd,
 			KGD_ENGINE_SDMA1);
 	kfd->shared_resources = *gpu_resources;
 
@@ -662,6 +664,9 @@ int kgd2kfd_post_reset(struct kfd_dev *kfd)
 		return ret;
 	count = atomic_dec_return(&kfd_locked);
 	WARN_ONCE(count != 0, "KFD reset ref. error");
+
+	atomic_set(&kfd->sram_ecc_flag, 0);
+
 	return 0;
 }
 
@@ -1025,6 +1030,12 @@ int kfd_gtt_sa_free(struct kfd_dev *kfd, struct kfd_mem_obj *mem_obj)
 	return 0;
 }
 
+void kgd2kfd_set_sram_ecc_flag(struct kfd_dev *kfd)
+{
+	if (kfd)
+		atomic_inc(&kfd->sram_ecc_flag);
+}
+
 #if defined(CONFIG_DEBUG_FS)
 
 /* This function will send a package to HIQ to hang the HWS
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
index e9f0e0a1b41c..6e1d41c5bf86 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
@@ -1011,25 +1011,41 @@ void kfd_signal_vm_fault_event(struct kfd_dev *dev, unsigned int pasid,
 void kfd_signal_reset_event(struct kfd_dev *dev)
 {
 	struct kfd_hsa_hw_exception_data hw_exception_data;
+	struct kfd_hsa_memory_exception_data memory_exception_data;
 	struct kfd_process *p;
 	struct kfd_event *ev;
 	unsigned int temp;
 	uint32_t id, idx;
+	int reset_cause = atomic_read(&dev->sram_ecc_flag) ?
+			KFD_HW_EXCEPTION_ECC :
+			KFD_HW_EXCEPTION_GPU_HANG;
 
 	/* Whole gpu reset caused by GPU hang and memory is lost */
 	memset(&hw_exception_data, 0, sizeof(hw_exception_data));
 	hw_exception_data.gpu_id = dev->id;
 	hw_exception_data.memory_lost = 1;
+	hw_exception_data.reset_cause = reset_cause;
+
+	memset(&memory_exception_data, 0, sizeof(memory_exception_data));
+	memory_exception_data.ErrorType = KFD_MEM_ERR_SRAM_ECC;
+	memory_exception_data.gpu_id = dev->id;
+	memory_exception_data.failure.imprecise = true;
 
 	idx = srcu_read_lock(&kfd_processes_srcu);
 	hash_for_each_rcu(kfd_processes_table, temp, p, kfd_processes) {
 		mutex_lock(&p->event_mutex);
 		id = KFD_FIRST_NONSIGNAL_EVENT_ID;
-		idr_for_each_entry_continue(&p->event_idr, ev, id)
+		idr_for_each_entry_continue(&p->event_idr, ev, id) {
 			if (ev->type == KFD_EVENT_TYPE_HW_EXCEPTION) {
 				ev->hw_exception_data = hw_exception_data;
 				set_event(ev);
 			}
+			if (ev->type == KFD_EVENT_TYPE_MEMORY &&
+			    reset_cause == KFD_HW_EXCEPTION_ECC) {
+				ev->memory_exception_data = memory_exception_data;
+				set_event(ev);
+			}
+		}
 		mutex_unlock(&p->event_mutex);
 	}
 	srcu_read_unlock(&kfd_processes_srcu, idx);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 0eeee3c6d6dc..9e0230965675 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -276,6 +276,9 @@ struct kfd_dev {
 	uint64_t hive_id;
 
 	bool pci_atomic_requested;
+
+	/* SRAM ECC flag */
+	atomic_t sram_ecc_flag;
 };
 
 enum kfd_mempool {
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index 09da91644f9f..2cb09e088dce 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -37,6 +37,7 @@
 #include "kfd_device_queue_manager.h"
 #include "kfd_iommu.h"
 #include "amdgpu_amdkfd.h"
+#include "amdgpu_ras.h"
 
 /* topology_device_list - Master list of all topology devices */
 static struct list_head topology_device_list;
@@ -1197,6 +1198,7 @@ int kfd_topology_add_device(struct kfd_dev *gpu)
 	void *crat_image = NULL;
 	size_t image_size = 0;
 	int proximity_domain;
+	struct amdgpu_ras *ctx;
 
 	INIT_LIST_HEAD(&temp_topology_device_list);
 
@@ -1328,6 +1330,20 @@ int kfd_topology_add_device(struct kfd_dev *gpu)
 		dev->node_props.capability |= HSA_CAP_ATS_PRESENT;
 	}
 
+	ctx = amdgpu_ras_get_context((struct amdgpu_device *)(dev->gpu->kgd));
+	if (ctx) {
+		/* kfd only concerns sram ecc on GFX/SDMA and HBM ecc on UMC */
+		dev->node_props.capability |=
+			(((ctx->features & BIT(AMDGPU_RAS_BLOCK__SDMA)) != 0) ||
+			 ((ctx->features & BIT(AMDGPU_RAS_BLOCK__GFX)) != 0)) ?
+			HSA_CAP_SRAM_EDCSUPPORTED : 0;
+		dev->node_props.capability |= ((ctx->features & BIT(AMDGPU_RAS_BLOCK__UMC)) != 0) ?
+			HSA_CAP_MEM_EDCSUPPORTED : 0;
+
+		dev->node_props.capability |= (ctx->features != 0) ?
+			HSA_CAP_RASEVENTNOTIFY : 0;
+	}
+
 	kfd_debug_print_topology();
 
 	if (!res)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.h b/drivers/gpu/drm/amd/amdkfd/kfd_topology.h
index 92a19be07344..84710cfd23c2 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.h
@@ -48,6 +48,10 @@
 #define HSA_CAP_DOORBELL_TYPE_2_0		0x2
 #define HSA_CAP_AQL_QUEUE_DOUBLE_MAP		0x00004000
 
+#define HSA_CAP_SRAM_EDCSUPPORTED		0x00080000
+#define HSA_CAP_MEM_EDCSUPPORTED		0x00100000
+#define HSA_CAP_RASEVENTNOTIFY			0x00200000
+
 struct kfd_node_properties {
 	uint64_t hive_id;
 	uint32_t cpu_cores_count;
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 3082b55b1e77..1854506e3e8f 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -111,7 +111,8 @@ amdgpu_dm_update_connector_after_detect(struct amdgpu_dm_connector *aconnector);
 
 static int amdgpu_dm_plane_init(struct amdgpu_display_manager *dm,
 				struct drm_plane *plane,
-				unsigned long possible_crtcs);
+				unsigned long possible_crtcs,
+				const struct dc_plane_cap *plane_cap);
 static int amdgpu_dm_crtc_init(struct amdgpu_display_manager *dm,
 			       struct drm_plane *plane,
 			       uint32_t link_index);
@@ -137,30 +138,6 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev,
 static void handle_cursor_update(struct drm_plane *plane,
 				 struct drm_plane_state *old_plane_state);
 
-
-
-static const enum drm_plane_type dm_plane_type_default[AMDGPU_MAX_PLANES] = {
-	DRM_PLANE_TYPE_PRIMARY,
-	DRM_PLANE_TYPE_PRIMARY,
-	DRM_PLANE_TYPE_PRIMARY,
-	DRM_PLANE_TYPE_PRIMARY,
-	DRM_PLANE_TYPE_PRIMARY,
-	DRM_PLANE_TYPE_PRIMARY,
-};
-
-static const enum drm_plane_type dm_plane_type_carizzo[AMDGPU_MAX_PLANES] = {
-	DRM_PLANE_TYPE_PRIMARY,
-	DRM_PLANE_TYPE_PRIMARY,
-	DRM_PLANE_TYPE_PRIMARY,
-	DRM_PLANE_TYPE_OVERLAY,/* YUV Capable Underlay */
-};
-
-static const enum drm_plane_type dm_plane_type_stoney[AMDGPU_MAX_PLANES] = {
-	DRM_PLANE_TYPE_PRIMARY,
-	DRM_PLANE_TYPE_PRIMARY,
-	DRM_PLANE_TYPE_OVERLAY, /* YUV Capable Underlay */
-};
-
 /*
  * dm_vblank_get_counter
  *
@@ -275,12 +252,22 @@ get_crtc_by_otg_inst(struct amdgpu_device *adev,
 	return NULL;
 }
 
+static inline bool amdgpu_dm_vrr_active(struct dm_crtc_state *dm_state)
+{
+	return dm_state->freesync_config.state == VRR_STATE_ACTIVE_VARIABLE ||
+	       dm_state->freesync_config.state == VRR_STATE_ACTIVE_FIXED;
+}
+
 static void dm_pflip_high_irq(void *interrupt_params)
 {
 	struct amdgpu_crtc *amdgpu_crtc;
 	struct common_irq_params *irq_params = interrupt_params;
 	struct amdgpu_device *adev = irq_params->adev;
 	unsigned long flags;
+	struct drm_pending_vblank_event *e;
+	struct dm_crtc_state *acrtc_state;
+	uint32_t vpos, hpos, v_blank_start, v_blank_end;
+	bool vrr_active;
 
 	amdgpu_crtc = get_crtc_by_otg_inst(adev, irq_params->irq_src - IRQ_TYPE_PFLIP);
 
@@ -303,26 +290,116 @@ static void dm_pflip_high_irq(void *interrupt_params)
 		return;
 	}
 
-	/* Update to correct count(s) if racing with vblank irq */
-	amdgpu_crtc->last_flip_vblank = drm_crtc_accurate_vblank_count(&amdgpu_crtc->base);
+	/* page flip completed. */
+	e = amdgpu_crtc->event;
+	amdgpu_crtc->event = NULL;
 
-	/* wake up userspace */
-	if (amdgpu_crtc->event) {
-		drm_crtc_send_vblank_event(&amdgpu_crtc->base, amdgpu_crtc->event);
+	if (!e)
+		WARN_ON(1);
 
-		/* page flip completed. clean up */
-		amdgpu_crtc->event = NULL;
+	acrtc_state = to_dm_crtc_state(amdgpu_crtc->base.state);
+	vrr_active = amdgpu_dm_vrr_active(acrtc_state);
+
+	/* Fixed refresh rate, or VRR scanout position outside front-porch? */
+	if (!vrr_active ||
+	    !dc_stream_get_scanoutpos(acrtc_state->stream, &v_blank_start,
+				      &v_blank_end, &hpos, &vpos) ||
+	    (vpos < v_blank_start)) {
+		/* Update to correct count and vblank timestamp if racing with
+		 * vblank irq. This also updates to the correct vblank timestamp
+		 * even in VRR mode, as scanout is past the front-porch atm.
+		 */
+		drm_crtc_accurate_vblank_count(&amdgpu_crtc->base);
 
-	} else
-		WARN_ON(1);
+		/* Wake up userspace by sending the pageflip event with proper
+		 * count and timestamp of vblank of flip completion.
+		 */
+		if (e) {
+			drm_crtc_send_vblank_event(&amdgpu_crtc->base, e);
+
+			/* Event sent, so done with vblank for this flip */
+			drm_crtc_vblank_put(&amdgpu_crtc->base);
+		}
+	} else if (e) {
+		/* VRR active and inside front-porch: vblank count and
+		 * timestamp for pageflip event will only be up to date after
+		 * drm_crtc_handle_vblank() has been executed from late vblank
+		 * irq handler after start of back-porch (vline 0). We queue the
+		 * pageflip event for send-out by drm_crtc_handle_vblank() with
+		 * updated timestamp and count, once it runs after us.
+		 *
+		 * We need to open-code this instead of using the helper
+		 * drm_crtc_arm_vblank_event(), as that helper would
+		 * call drm_crtc_accurate_vblank_count(), which we must
+		 * not call in VRR mode while we are in front-porch!
+		 */
+
+		/* sequence will be replaced by real count during send-out. */
+		e->sequence = drm_crtc_vblank_count(&amdgpu_crtc->base);
+		e->pipe = amdgpu_crtc->crtc_id;
+
+		list_add_tail(&e->base.link, &adev->ddev->vblank_event_list);
+		e = NULL;
+	}
+
+	/* Keep track of vblank of this flip for flip throttling. We use the
+	 * cooked hw counter, as that one incremented at start of this vblank
+	 * of pageflip completion, so last_flip_vblank is the forbidden count
+	 * for queueing new pageflips if vsync + VRR is enabled.
+	 */
+	amdgpu_crtc->last_flip_vblank = amdgpu_get_vblank_counter_kms(adev->ddev,
+							amdgpu_crtc->crtc_id);
 
 	amdgpu_crtc->pflip_status = AMDGPU_FLIP_NONE;
 	spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
 
-	DRM_DEBUG_DRIVER("%s - crtc :%d[%p], pflip_stat:AMDGPU_FLIP_NONE\n",
-					__func__, amdgpu_crtc->crtc_id, amdgpu_crtc);
+	DRM_DEBUG_DRIVER("crtc:%d[%p], pflip_stat:AMDGPU_FLIP_NONE, vrr[%d]-fp %d\n",
+			 amdgpu_crtc->crtc_id, amdgpu_crtc,
+			 vrr_active, (int) !e);
+}
+
+static void dm_vupdate_high_irq(void *interrupt_params)
+{
+	struct common_irq_params *irq_params = interrupt_params;
+	struct amdgpu_device *adev = irq_params->adev;
+	struct amdgpu_crtc *acrtc;
+	struct dm_crtc_state *acrtc_state;
+	unsigned long flags;
+
+	acrtc = get_crtc_by_otg_inst(adev, irq_params->irq_src - IRQ_TYPE_VUPDATE);
+
+	if (acrtc) {
+		acrtc_state = to_dm_crtc_state(acrtc->base.state);
+
+		DRM_DEBUG_DRIVER("crtc:%d, vupdate-vrr:%d\n", acrtc->crtc_id,
+				 amdgpu_dm_vrr_active(acrtc_state));
 
-	drm_crtc_vblank_put(&amdgpu_crtc->base);
+		/* Core vblank handling is done here after end of front-porch in
+		 * vrr mode, as vblank timestamping will give valid results
+		 * while now done after front-porch. This will also deliver
+		 * page-flip completion events that have been queued to us
+		 * if a pageflip happened inside front-porch.
+		 */
+		if (amdgpu_dm_vrr_active(acrtc_state)) {
+			drm_crtc_handle_vblank(&acrtc->base);
+
+			/* BTR processing for pre-DCE12 ASICs */
+			if (acrtc_state->stream &&
+			    adev->family < AMDGPU_FAMILY_AI) {
+				spin_lock_irqsave(&adev->ddev->event_lock, flags);
+				mod_freesync_handle_v_update(
+				    adev->dm.freesync_module,
+				    acrtc_state->stream,
+				    &acrtc_state->vrr_params);
+
+				dc_stream_adjust_vmin_vmax(
+				    adev->dm.dc,
+				    acrtc_state->stream,
+				    &acrtc_state->vrr_params.adjust);
+				spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
+			}
+		}
+	}
 }
 
 static void dm_crtc_high_irq(void *interrupt_params)
@@ -331,18 +408,33 @@ static void dm_crtc_high_irq(void *interrupt_params)
 	struct amdgpu_device *adev = irq_params->adev;
 	struct amdgpu_crtc *acrtc;
 	struct dm_crtc_state *acrtc_state;
+	unsigned long flags;
 
 	acrtc = get_crtc_by_otg_inst(adev, irq_params->irq_src - IRQ_TYPE_VBLANK);
 
 	if (acrtc) {
-		drm_crtc_handle_vblank(&acrtc->base);
-		amdgpu_dm_crtc_handle_crc_irq(&acrtc->base);
-
 		acrtc_state = to_dm_crtc_state(acrtc->base.state);
 
-		if (acrtc_state->stream &&
+		DRM_DEBUG_DRIVER("crtc:%d, vupdate-vrr:%d\n", acrtc->crtc_id,
+				 amdgpu_dm_vrr_active(acrtc_state));
+
+		/* Core vblank handling at start of front-porch is only possible
+		 * in non-vrr mode, as only there vblank timestamping will give
+		 * valid results while done in front-porch. Otherwise defer it
+		 * to dm_vupdate_high_irq after end of front-porch.
+		 */
+		if (!amdgpu_dm_vrr_active(acrtc_state))
+			drm_crtc_handle_vblank(&acrtc->base);
+
+		/* Following stuff must happen at start of vblank, for crc
+		 * computation and below-the-range btr support in vrr mode.
+		 */
+		amdgpu_dm_crtc_handle_crc_irq(&acrtc->base);
+
+		if (acrtc_state->stream && adev->family >= AMDGPU_FAMILY_AI &&
 		    acrtc_state->vrr_params.supported &&
 		    acrtc_state->freesync_config.state == VRR_STATE_ACTIVE_VARIABLE) {
+			spin_lock_irqsave(&adev->ddev->event_lock, flags);
 			mod_freesync_handle_v_update(
 				adev->dm.freesync_module,
 				acrtc_state->stream,
@@ -352,6 +444,7 @@ static void dm_crtc_high_irq(void *interrupt_params)
 				adev->dm.dc,
 				acrtc_state->stream,
 				&acrtc_state->vrr_params.adjust);
+			spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
 		}
 	}
 }
@@ -462,6 +555,8 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 	if (amdgpu_dc_feature_mask & DC_FBC_MASK)
 		init_data.flags.fbc_support = true;
 
+	init_data.flags.power_down_display_on_boot = true;
+
 	/* Display Core create. */
 	adev->dm.dc = dc_create(&init_data);
 
@@ -912,9 +1007,16 @@ static int dm_resume(void *handle)
 	struct drm_plane *plane;
 	struct drm_plane_state *new_plane_state;
 	struct dm_plane_state *dm_new_plane_state;
+	struct dm_atomic_state *dm_state = to_dm_atomic_state(dm->atomic_obj.state);
 	enum dc_connection_type new_connection_type = dc_connection_none;
 	int i;
 
+	/* Recreate dc_state - DC invalidates it when setting power state to S3. */
+	dc_release_state(dm_state->context);
+	dm_state->context = dc_create_state(dm->dc);
+	/* TODO: Remove dc_state->dccg, use dc->dccg directly. */
+	dc_resource_state_construct(dm->dc, dm_state->context);
+
 	/* power on hardware */
 	dc_set_power_state(dm->dc, DC_ACPI_CM_POWER_STATE_D0);
 
@@ -1457,6 +1559,27 @@ static int dce110_register_irq_handlers(struct amdgpu_device *adev)
 				dm_crtc_high_irq, c_irq_params);
 	}
 
+	/* Use VUPDATE interrupt */
+	for (i = VISLANDS30_IV_SRCID_D1_V_UPDATE_INT; i <= VISLANDS30_IV_SRCID_D6_V_UPDATE_INT; i += 2) {
+		r = amdgpu_irq_add_id(adev, client_id, i, &adev->vupdate_irq);
+		if (r) {
+			DRM_ERROR("Failed to add vupdate irq id!\n");
+			return r;
+		}
+
+		int_params.int_context = INTERRUPT_HIGH_IRQ_CONTEXT;
+		int_params.irq_source =
+			dc_interrupt_to_irq_source(dc, i, 0);
+
+		c_irq_params = &adev->dm.vupdate_params[int_params.irq_source - DC_IRQ_SOURCE_VUPDATE1];
+
+		c_irq_params->adev = adev;
+		c_irq_params->irq_src = int_params.irq_source;
+
+		amdgpu_dm_irq_register_interrupt(adev, &int_params,
+				dm_vupdate_high_irq, c_irq_params);
+	}
+
 	/* Use GRPH_PFLIP interrupt */
 	for (i = VISLANDS30_IV_SRCID_D1_GRPH_PFLIP;
 			i <= VISLANDS30_IV_SRCID_D6_GRPH_PFLIP; i += 2) {
@@ -1542,6 +1665,34 @@ static int dcn10_register_irq_handlers(struct amdgpu_device *adev)
 				dm_crtc_high_irq, c_irq_params);
 	}
 
+	/* Use VUPDATE_NO_LOCK interrupt on DCN, which seems to correspond to
+	 * the regular VUPDATE interrupt on DCE. We want DC_IRQ_SOURCE_VUPDATEx
+	 * to trigger at end of each vblank, regardless of state of the lock,
+	 * matching DCE behaviour.
+	 */
+	for (i = DCN_1_0__SRCID__OTG0_IHC_V_UPDATE_NO_LOCK_INTERRUPT;
+	     i <= DCN_1_0__SRCID__OTG0_IHC_V_UPDATE_NO_LOCK_INTERRUPT + adev->mode_info.num_crtc - 1;
+	     i++) {
+		r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_DCE, i, &adev->vupdate_irq);
+
+		if (r) {
+			DRM_ERROR("Failed to add vupdate irq id!\n");
+			return r;
+		}
+
+		int_params.int_context = INTERRUPT_HIGH_IRQ_CONTEXT;
+		int_params.irq_source =
+			dc_interrupt_to_irq_source(dc, i, 0);
+
+		c_irq_params = &adev->dm.vupdate_params[int_params.irq_source - DC_IRQ_SOURCE_VUPDATE1];
+
+		c_irq_params->adev = adev;
+		c_irq_params->irq_src = int_params.irq_source;
+
+		amdgpu_dm_irq_register_interrupt(adev, &int_params,
+				dm_vupdate_high_irq, c_irq_params);
+	}
+
 	/* Use GRPH_PFLIP interrupt */
 	for (i = DCN_1_0__SRCID__HUBP0_FLIP_INTERRUPT;
 			i <= DCN_1_0__SRCID__HUBP0_FLIP_INTERRUPT + adev->mode_info.num_crtc - 1;
@@ -1593,15 +1744,10 @@ static int dm_atomic_get_state(struct drm_atomic_state *state,
 	struct amdgpu_device *adev = dev->dev_private;
 	struct amdgpu_display_manager *dm = &adev->dm;
 	struct drm_private_state *priv_state;
-	int ret;
 
 	if (*dm_state)
 		return 0;
 
-	ret = drm_modeset_lock(&dm->atomic_obj_lock, state->acquire_ctx);
-	if (ret)
-		return ret;
-
 	priv_state = drm_atomic_get_private_obj_state(state, &dm->atomic_obj);
 	if (IS_ERR(priv_state))
 		return PTR_ERR(priv_state);
@@ -1658,17 +1804,16 @@ dm_atomic_duplicate_state(struct drm_private_obj *obj)
 
 	__drm_atomic_helper_private_obj_duplicate_state(obj, &new_state->base);
 
-	new_state->context = dc_create_state();
+	old_state = to_dm_atomic_state(obj->state);
+
+	if (old_state && old_state->context)
+		new_state->context = dc_copy_state(old_state->context);
+
 	if (!new_state->context) {
 		kfree(new_state);
 		return NULL;
 	}
 
-	old_state = to_dm_atomic_state(obj->state);
-	if (old_state && old_state->context)
-		dc_resource_state_copy_construct(old_state->context,
-						 new_state->context);
-
 	return &new_state->base;
 }
 
@@ -1708,13 +1853,11 @@ static int amdgpu_dm_mode_config_init(struct amdgpu_device *adev)
 
 	adev->ddev->mode_config.fb_base = adev->gmc.aper_base;
 
-	drm_modeset_lock_init(&adev->dm.atomic_obj_lock);
-
 	state = kzalloc(sizeof(*state), GFP_KERNEL);
 	if (!state)
 		return -ENOMEM;
 
-	state->context = dc_create_state();
+	state->context = dc_create_state(adev->dm.dc);
 	if (!state->context) {
 		kfree(state);
 		return -ENOMEM;
@@ -1841,39 +1984,42 @@ amdgpu_dm_register_backlight_device(struct amdgpu_display_manager *dm)
 #endif
 
 static int initialize_plane(struct amdgpu_display_manager *dm,
-			     struct amdgpu_mode_info *mode_info,
-			     int plane_id)
+			    struct amdgpu_mode_info *mode_info, int plane_id,
+			    enum drm_plane_type plane_type,
+			    const struct dc_plane_cap *plane_cap)
 {
 	struct drm_plane *plane;
 	unsigned long possible_crtcs;
 	int ret = 0;
 
 	plane = kzalloc(sizeof(struct drm_plane), GFP_KERNEL);
-	mode_info->planes[plane_id] = plane;
-
 	if (!plane) {
 		DRM_ERROR("KMS: Failed to allocate plane\n");
 		return -ENOMEM;
 	}
-	plane->type = mode_info->plane_type[plane_id];
+	plane->type = plane_type;
 
 	/*
-	 * HACK: IGT tests expect that each plane can only have
-	 * one possible CRTC. For now, set one CRTC for each
-	 * plane that is not an underlay, but still allow multiple
-	 * CRTCs for underlay planes.
+	 * HACK: IGT tests expect that the primary plane for a CRTC
+	 * can only have one possible CRTC. Only expose support for
+	 * any CRTC if they're not going to be used as a primary plane
+	 * for a CRTC - like overlay or underlay planes.
 	 */
 	possible_crtcs = 1 << plane_id;
 	if (plane_id >= dm->dc->caps.max_streams)
 		possible_crtcs = 0xff;
 
-	ret = amdgpu_dm_plane_init(dm, mode_info->planes[plane_id], possible_crtcs);
+	ret = amdgpu_dm_plane_init(dm, plane, possible_crtcs, plane_cap);
 
 	if (ret) {
 		DRM_ERROR("KMS: Failed to initialize plane\n");
+		kfree(plane);
 		return ret;
 	}
 
+	if (mode_info)
+		mode_info->planes[plane_id] = plane;
+
 	return ret;
 }
 
@@ -1916,8 +2062,9 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
 	struct amdgpu_encoder *aencoder = NULL;
 	struct amdgpu_mode_info *mode_info = &adev->mode_info;
 	uint32_t link_cnt;
-	int32_t total_overlay_planes, total_primary_planes;
+	int32_t primary_planes;
 	enum dc_connection_type new_connection_type = dc_connection_none;
+	const struct dc_plane_cap *plane;
 
 	link_cnt = dm->dc->caps.max_links;
 	if (amdgpu_dm_mode_config_init(dm->adev)) {
@@ -1925,24 +2072,53 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
 		return -EINVAL;
 	}
 
-	/* Identify the number of planes to be initialized */
-	total_overlay_planes = dm->dc->caps.max_slave_planes;
-	total_primary_planes = dm->dc->caps.max_planes - dm->dc->caps.max_slave_planes;
+	/* There is one primary plane per CRTC */
+	primary_planes = dm->dc->caps.max_streams;
+	ASSERT(primary_planes <= AMDGPU_MAX_PLANES);
 
-	/* First initialize overlay planes, index starting after primary planes */
-	for (i = (total_overlay_planes - 1); i >= 0; i--) {
-		if (initialize_plane(dm, mode_info, (total_primary_planes + i))) {
-			DRM_ERROR("KMS: Failed to initialize overlay plane\n");
+	/*
+	 * Initialize primary planes, implicit planes for legacy IOCTLS.
+	 * Order is reversed to match iteration order in atomic check.
+	 */
+	for (i = (primary_planes - 1); i >= 0; i--) {
+		plane = &dm->dc->caps.planes[i];
+
+		if (initialize_plane(dm, mode_info, i,
+				     DRM_PLANE_TYPE_PRIMARY, plane)) {
+			DRM_ERROR("KMS: Failed to initialize primary plane\n");
 			goto fail;
 		}
 	}
 
-	/* Initialize primary planes */
-	for (i = (total_primary_planes - 1); i >= 0; i--) {
-		if (initialize_plane(dm, mode_info, i)) {
-			DRM_ERROR("KMS: Failed to initialize primary plane\n");
+	/*
+	 * Initialize overlay planes, index starting after primary planes.
+	 * These planes have a higher DRM index than the primary planes since
+	 * they should be considered as having a higher z-order.
+	 * Order is reversed to match iteration order in atomic check.
+	 *
+	 * Only support DCN for now, and only expose one so we don't encourage
+	 * userspace to use up all the pipes.
+	 */
+	for (i = 0; i < dm->dc->caps.max_planes; ++i) {
+		struct dc_plane_cap *plane = &dm->dc->caps.planes[i];
+
+		if (plane->type != DC_PLANE_TYPE_DCN_UNIVERSAL)
+			continue;
+
+		if (!plane->blends_with_above || !plane->blends_with_below)
+			continue;
+
+		if (!plane->pixel_format_support.argb8888)
+			continue;
+
+		if (initialize_plane(dm, NULL, primary_planes + i,
+				     DRM_PLANE_TYPE_OVERLAY, plane)) {
+			DRM_ERROR("KMS: Failed to initialize overlay plane\n");
 			goto fail;
 		}
+
+		/* Only create one overlay plane. */
+		break;
 	}
 
 	for (i = 0; i < dm->dc->caps.max_streams; i++)
@@ -2042,8 +2218,7 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
 fail:
 	kfree(aencoder);
 	kfree(aconnector);
-	for (i = 0; i < dm->dc->caps.max_planes; i++)
-		kfree(mode_info->planes[i]);
+
 	return -EINVAL;
 }
 
@@ -2124,53 +2299,45 @@ static int dm_early_init(void *handle)
 		adev->mode_info.num_crtc = 6;
 		adev->mode_info.num_hpd = 6;
 		adev->mode_info.num_dig = 6;
-		adev->mode_info.plane_type = dm_plane_type_default;
 		break;
 	case CHIP_KAVERI:
 		adev->mode_info.num_crtc = 4;
 		adev->mode_info.num_hpd = 6;
 		adev->mode_info.num_dig = 7;
-		adev->mode_info.plane_type = dm_plane_type_default;
 		break;
 	case CHIP_KABINI:
 	case CHIP_MULLINS:
 		adev->mode_info.num_crtc = 2;
 		adev->mode_info.num_hpd = 6;
 		adev->mode_info.num_dig = 6;
-		adev->mode_info.plane_type = dm_plane_type_default;
 		break;
 	case CHIP_FIJI:
 	case CHIP_TONGA:
 		adev->mode_info.num_crtc = 6;
 		adev->mode_info.num_hpd = 6;
 		adev->mode_info.num_dig = 7;
-		adev->mode_info.plane_type = dm_plane_type_default;
 		break;
 	case CHIP_CARRIZO:
 		adev->mode_info.num_crtc = 3;
 		adev->mode_info.num_hpd = 6;
 		adev->mode_info.num_dig = 9;
-		adev->mode_info.plane_type = dm_plane_type_carizzo;
 		break;
 	case CHIP_STONEY:
 		adev->mode_info.num_crtc = 2;
 		adev->mode_info.num_hpd = 6;
 		adev->mode_info.num_dig = 9;
-		adev->mode_info.plane_type = dm_plane_type_stoney;
 		break;
 	case CHIP_POLARIS11:
 	case CHIP_POLARIS12:
 		adev->mode_info.num_crtc = 5;
 		adev->mode_info.num_hpd = 5;
 		adev->mode_info.num_dig = 5;
-		adev->mode_info.plane_type = dm_plane_type_default;
 		break;
 	case CHIP_POLARIS10:
 	case CHIP_VEGAM:
 		adev->mode_info.num_crtc = 6;
 		adev->mode_info.num_hpd = 6;
 		adev->mode_info.num_dig = 6;
-		adev->mode_info.plane_type = dm_plane_type_default;
 		break;
 	case CHIP_VEGA10:
 	case CHIP_VEGA12:
@@ -2178,14 +2345,12 @@ static int dm_early_init(void *handle)
 		adev->mode_info.num_crtc = 6;
 		adev->mode_info.num_hpd = 6;
 		adev->mode_info.num_dig = 6;
-		adev->mode_info.plane_type = dm_plane_type_default;
 		break;
 #if defined(CONFIG_DRM_AMD_DC_DCN1_0)
 	case CHIP_RAVEN:
 		adev->mode_info.num_crtc = 4;
 		adev->mode_info.num_hpd = 4;
 		adev->mode_info.num_dig = 4;
-		adev->mode_info.plane_type = dm_plane_type_default;
 		break;
 #endif
 	default:
@@ -2243,56 +2408,63 @@ static const struct drm_encoder_funcs amdgpu_dm_encoder_funcs = {
 	.destroy = amdgpu_dm_encoder_destroy,
 };
 
-static bool fill_rects_from_plane_state(const struct drm_plane_state *state,
-					struct dc_plane_state *plane_state)
+
+static int fill_dc_scaling_info(const struct drm_plane_state *state,
+				struct dc_scaling_info *scaling_info)
 {
-	plane_state->src_rect.x = state->src_x >> 16;
-	plane_state->src_rect.y = state->src_y >> 16;
-	/* we ignore the mantissa for now and do not deal with floating pixels :( */
-	plane_state->src_rect.width = state->src_w >> 16;
+	int scale_w, scale_h;
 
-	if (plane_state->src_rect.width == 0)
-		return false;
+	memset(scaling_info, 0, sizeof(*scaling_info));
 
-	plane_state->src_rect.height = state->src_h >> 16;
-	if (plane_state->src_rect.height == 0)
-		return false;
+	/* Source is fixed 16.16 but we ignore mantissa for now... */
+	scaling_info->src_rect.x = state->src_x >> 16;
+	scaling_info->src_rect.y = state->src_y >> 16;
 
-	plane_state->dst_rect.x = state->crtc_x;
-	plane_state->dst_rect.y = state->crtc_y;
+	scaling_info->src_rect.width = state->src_w >> 16;
+	if (scaling_info->src_rect.width == 0)
+		return -EINVAL;
+
+	scaling_info->src_rect.height = state->src_h >> 16;
+	if (scaling_info->src_rect.height == 0)
+		return -EINVAL;
+
+	scaling_info->dst_rect.x = state->crtc_x;
+	scaling_info->dst_rect.y = state->crtc_y;
 
 	if (state->crtc_w == 0)
-		return false;
+		return -EINVAL;
 
-	plane_state->dst_rect.width = state->crtc_w;
+	scaling_info->dst_rect.width = state->crtc_w;
 
 	if (state->crtc_h == 0)
-		return false;
+		return -EINVAL;
 
-	plane_state->dst_rect.height = state->crtc_h;
+	scaling_info->dst_rect.height = state->crtc_h;
 
-	plane_state->clip_rect = plane_state->dst_rect;
+	/* DRM doesn't specify clipping on destination output. */
+	scaling_info->clip_rect = scaling_info->dst_rect;
 
-	switch (state->rotation & DRM_MODE_ROTATE_MASK) {
-	case DRM_MODE_ROTATE_0:
-		plane_state->rotation = ROTATION_ANGLE_0;
-		break;
-	case DRM_MODE_ROTATE_90:
-		plane_state->rotation = ROTATION_ANGLE_90;
-		break;
-	case DRM_MODE_ROTATE_180:
-		plane_state->rotation = ROTATION_ANGLE_180;
-		break;
-	case DRM_MODE_ROTATE_270:
-		plane_state->rotation = ROTATION_ANGLE_270;
-		break;
-	default:
-		plane_state->rotation = ROTATION_ANGLE_0;
-		break;
-	}
+	/* TODO: Validate scaling per-format with DC plane caps */
+	scale_w = scaling_info->dst_rect.width * 1000 /
+		  scaling_info->src_rect.width;
 
-	return true;
+	if (scale_w < 250 || scale_w > 16000)
+		return -EINVAL;
+
+	scale_h = scaling_info->dst_rect.height * 1000 /
+		  scaling_info->src_rect.height;
+
+	if (scale_h < 250 || scale_h > 16000)
+		return -EINVAL;
+
+	/*
+	 * The "scaling_quality" can be ignored for now, quality = 0 has DC
+	 * assume reasonable defaults based on the format.
+	 */
+
+	return 0;
 }
+
 static int get_fb_info(const struct amdgpu_framebuffer *amdgpu_fb,
 		       uint64_t *tiling_flags)
 {
@@ -2321,10 +2493,16 @@ static inline uint64_t get_dcc_address(uint64_t address, uint64_t tiling_flags)
 	return offset ? (address + offset * 256) : 0;
 }
 
-static bool fill_plane_dcc_attributes(struct amdgpu_device *adev,
-				      const struct amdgpu_framebuffer *afb,
-				      struct dc_plane_state *plane_state,
-				      uint64_t info)
+static int
+fill_plane_dcc_attributes(struct amdgpu_device *adev,
+			  const struct amdgpu_framebuffer *afb,
+			  const enum surface_pixel_format format,
+			  const enum dc_rotation_angle rotation,
+			  const union plane_size *plane_size,
+			  const union dc_tiling_info *tiling_info,
+			  const uint64_t info,
+			  struct dc_plane_dcc_param *dcc,
+			  struct dc_plane_address *address)
 {
 	struct dc *dc = adev->dm.dc;
 	struct dc_dcc_surface_param input;
@@ -2337,133 +2515,103 @@ static bool fill_plane_dcc_attributes(struct amdgpu_device *adev,
 	memset(&output, 0, sizeof(output));
 
 	if (!offset)
-		return false;
+		return 0;
+
+	if (format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN)
+		return 0;
 
 	if (!dc->cap_funcs.get_dcc_compression_cap)
-		return false;
+		return -EINVAL;
 
-	input.format = plane_state->format;
-	input.surface_size.width =
-		plane_state->plane_size.grph.surface_size.width;
-	input.surface_size.height =
-		plane_state->plane_size.grph.surface_size.height;
-	input.swizzle_mode = plane_state->tiling_info.gfx9.swizzle;
+	input.format = format;
+	input.surface_size.width = plane_size->grph.surface_size.width;
+	input.surface_size.height = plane_size->grph.surface_size.height;
+	input.swizzle_mode = tiling_info->gfx9.swizzle;
 
-	if (plane_state->rotation == ROTATION_ANGLE_0 ||
-	    plane_state->rotation == ROTATION_ANGLE_180)
+	if (rotation == ROTATION_ANGLE_0 || rotation == ROTATION_ANGLE_180)
 		input.scan = SCAN_DIRECTION_HORIZONTAL;
-	else if (plane_state->rotation == ROTATION_ANGLE_90 ||
-		 plane_state->rotation == ROTATION_ANGLE_270)
+	else if (rotation == ROTATION_ANGLE_90 || rotation == ROTATION_ANGLE_270)
 		input.scan = SCAN_DIRECTION_VERTICAL;
 
 	if (!dc->cap_funcs.get_dcc_compression_cap(dc, &input, &output))
-		return false;
+		return -EINVAL;
 
 	if (!output.capable)
-		return false;
+		return -EINVAL;
 
 	if (i64b == 0 && output.grph.rgb.independent_64b_blks != 0)
-		return false;
+		return -EINVAL;
 
-	plane_state->dcc.enable = 1;
-	plane_state->dcc.grph.meta_pitch =
+	dcc->enable = 1;
+	dcc->grph.meta_pitch =
 		AMDGPU_TILING_GET(info, DCC_PITCH_MAX) + 1;
-	plane_state->dcc.grph.independent_64b_blks = i64b;
+	dcc->grph.independent_64b_blks = i64b;
 
 	dcc_address = get_dcc_address(afb->address, info);
-	plane_state->address.grph.meta_addr.low_part =
-		lower_32_bits(dcc_address);
-	plane_state->address.grph.meta_addr.high_part =
-		upper_32_bits(dcc_address);
+	address->grph.meta_addr.low_part = lower_32_bits(dcc_address);
+	address->grph.meta_addr.high_part = upper_32_bits(dcc_address);
 
-	return true;
+	return 0;
 }
 
-static int fill_plane_attributes_from_fb(struct amdgpu_device *adev,
-					 struct dc_plane_state *plane_state,
-					 const struct amdgpu_framebuffer *amdgpu_fb)
-{
-	uint64_t tiling_flags;
-	unsigned int awidth;
-	const struct drm_framebuffer *fb = &amdgpu_fb->base;
-	int ret = 0;
-	struct drm_format_name_buf format_name;
-
-	ret = get_fb_info(
-		amdgpu_fb,
-		&tiling_flags);
-
-	if (ret)
-		return ret;
-
-	switch (fb->format->format) {
-	case DRM_FORMAT_C8:
-		plane_state->format = SURFACE_PIXEL_FORMAT_GRPH_PALETA_256_COLORS;
-		break;
-	case DRM_FORMAT_RGB565:
-		plane_state->format = SURFACE_PIXEL_FORMAT_GRPH_RGB565;
-		break;
-	case DRM_FORMAT_XRGB8888:
-	case DRM_FORMAT_ARGB8888:
-		plane_state->format = SURFACE_PIXEL_FORMAT_GRPH_ARGB8888;
-		break;
-	case DRM_FORMAT_XRGB2101010:
-	case DRM_FORMAT_ARGB2101010:
-		plane_state->format = SURFACE_PIXEL_FORMAT_GRPH_ARGB2101010;
-		break;
-	case DRM_FORMAT_XBGR2101010:
-	case DRM_FORMAT_ABGR2101010:
-		plane_state->format = SURFACE_PIXEL_FORMAT_GRPH_ABGR2101010;
-		break;
-	case DRM_FORMAT_XBGR8888:
-	case DRM_FORMAT_ABGR8888:
-		plane_state->format = SURFACE_PIXEL_FORMAT_GRPH_ABGR8888;
-		break;
-	case DRM_FORMAT_NV21:
-		plane_state->format = SURFACE_PIXEL_FORMAT_VIDEO_420_YCbCr;
-		break;
-	case DRM_FORMAT_NV12:
-		plane_state->format = SURFACE_PIXEL_FORMAT_VIDEO_420_YCrCb;
-		break;
-	default:
-		DRM_ERROR("Unsupported screen format %s\n",
-			  drm_get_format_name(fb->format->format, &format_name));
-		return -EINVAL;
-	}
-
-	memset(&plane_state->address, 0, sizeof(plane_state->address));
-	memset(&plane_state->tiling_info, 0, sizeof(plane_state->tiling_info));
-	memset(&plane_state->dcc, 0, sizeof(plane_state->dcc));
-
-	if (plane_state->format < SURFACE_PIXEL_FORMAT_VIDEO_BEGIN) {
-		plane_state->address.type = PLN_ADDR_TYPE_GRAPHICS;
-		plane_state->plane_size.grph.surface_size.x = 0;
-		plane_state->plane_size.grph.surface_size.y = 0;
-		plane_state->plane_size.grph.surface_size.width = fb->width;
-		plane_state->plane_size.grph.surface_size.height = fb->height;
-		plane_state->plane_size.grph.surface_pitch =
-				fb->pitches[0] / fb->format->cpp[0];
-		/* TODO: unhardcode */
-		plane_state->color_space = COLOR_SPACE_SRGB;
+static int
+fill_plane_buffer_attributes(struct amdgpu_device *adev,
+			     const struct amdgpu_framebuffer *afb,
+			     const enum surface_pixel_format format,
+			     const enum dc_rotation_angle rotation,
+			     const uint64_t tiling_flags,
+			     union dc_tiling_info *tiling_info,
+			     union plane_size *plane_size,
+			     struct dc_plane_dcc_param *dcc,
+			     struct dc_plane_address *address)
+{
+	const struct drm_framebuffer *fb = &afb->base;
+	int ret;
 
+	memset(tiling_info, 0, sizeof(*tiling_info));
+	memset(plane_size, 0, sizeof(*plane_size));
+	memset(dcc, 0, sizeof(*dcc));
+	memset(address, 0, sizeof(*address));
+
+	if (format < SURFACE_PIXEL_FORMAT_VIDEO_BEGIN) {
+		plane_size->grph.surface_size.x = 0;
+		plane_size->grph.surface_size.y = 0;
+		plane_size->grph.surface_size.width = fb->width;
+		plane_size->grph.surface_size.height = fb->height;
+		plane_size->grph.surface_pitch =
+			fb->pitches[0] / fb->format->cpp[0];
+
+		address->type = PLN_ADDR_TYPE_GRAPHICS;
+		address->grph.addr.low_part = lower_32_bits(afb->address);
+		address->grph.addr.high_part = upper_32_bits(afb->address);
 	} else {
-		awidth = ALIGN(fb->width, 64);
-		plane_state->address.type = PLN_ADDR_TYPE_VIDEO_PROGRESSIVE;
-		plane_state->plane_size.video.luma_size.x = 0;
-		plane_state->plane_size.video.luma_size.y = 0;
-		plane_state->plane_size.video.luma_size.width = awidth;
-		plane_state->plane_size.video.luma_size.height = fb->height;
-		/* TODO: unhardcode */
-		plane_state->plane_size.video.luma_pitch = awidth;
-
-		plane_state->plane_size.video.chroma_size.x = 0;
-		plane_state->plane_size.video.chroma_size.y = 0;
-		plane_state->plane_size.video.chroma_size.width = awidth;
-		plane_state->plane_size.video.chroma_size.height = fb->height;
-		plane_state->plane_size.video.chroma_pitch = awidth / 2;
-
-		/* TODO: unhardcode */
-		plane_state->color_space = COLOR_SPACE_YCBCR709;
+		uint64_t chroma_addr = afb->address + fb->offsets[1];
+
+		plane_size->video.luma_size.x = 0;
+		plane_size->video.luma_size.y = 0;
+		plane_size->video.luma_size.width = fb->width;
+		plane_size->video.luma_size.height = fb->height;
+		plane_size->video.luma_pitch =
+			fb->pitches[0] / fb->format->cpp[0];
+
+		plane_size->video.chroma_size.x = 0;
+		plane_size->video.chroma_size.y = 0;
+		/* TODO: set these based on surface format */
+		plane_size->video.chroma_size.width = fb->width / 2;
+		plane_size->video.chroma_size.height = fb->height / 2;
+
+		plane_size->video.chroma_pitch =
+			fb->pitches[1] / fb->format->cpp[1];
+
+		address->type = PLN_ADDR_TYPE_VIDEO_PROGRESSIVE;
+		address->video_progressive.luma_addr.low_part =
+			lower_32_bits(afb->address);
+		address->video_progressive.luma_addr.high_part =
+			upper_32_bits(afb->address);
+		address->video_progressive.chroma_addr.low_part =
+			lower_32_bits(chroma_addr);
+		address->video_progressive.chroma_addr.high_part =
+			upper_32_bits(chroma_addr);
 	}
 
 	/* Fill GFX8 params */
@@ -2477,21 +2625,21 @@ static int fill_plane_attributes_from_fb(struct amdgpu_device *adev,
 		num_banks = AMDGPU_TILING_GET(tiling_flags, NUM_BANKS);
 
 		/* XXX fix me for VI */
-		plane_state->tiling_info.gfx8.num_banks = num_banks;
-		plane_state->tiling_info.gfx8.array_mode =
+		tiling_info->gfx8.num_banks = num_banks;
+		tiling_info->gfx8.array_mode =
 				DC_ARRAY_2D_TILED_THIN1;
-		plane_state->tiling_info.gfx8.tile_split = tile_split;
-		plane_state->tiling_info.gfx8.bank_width = bankw;
-		plane_state->tiling_info.gfx8.bank_height = bankh;
-		plane_state->tiling_info.gfx8.tile_aspect = mtaspect;
-		plane_state->tiling_info.gfx8.tile_mode =
+		tiling_info->gfx8.tile_split = tile_split;
+		tiling_info->gfx8.bank_width = bankw;
+		tiling_info->gfx8.bank_height = bankh;
+		tiling_info->gfx8.tile_aspect = mtaspect;
+		tiling_info->gfx8.tile_mode =
 				DC_ADDR_SURF_MICRO_TILING_DISPLAY;
 	} else if (AMDGPU_TILING_GET(tiling_flags, ARRAY_MODE)
 			== DC_ARRAY_1D_TILED_THIN1) {
-		plane_state->tiling_info.gfx8.array_mode = DC_ARRAY_1D_TILED_THIN1;
+		tiling_info->gfx8.array_mode = DC_ARRAY_1D_TILED_THIN1;
 	}
 
-	plane_state->tiling_info.gfx8.pipe_config =
+	tiling_info->gfx8.pipe_config =
 			AMDGPU_TILING_GET(tiling_flags, PIPE_CONFIG);
 
 	if (adev->asic_type == CHIP_VEGA10 ||
@@ -2499,60 +2647,249 @@ static int fill_plane_attributes_from_fb(struct amdgpu_device *adev,
 	    adev->asic_type == CHIP_VEGA20 ||
 	    adev->asic_type == CHIP_RAVEN) {
 		/* Fill GFX9 params */
-		plane_state->tiling_info.gfx9.num_pipes =
+		tiling_info->gfx9.num_pipes =
 			adev->gfx.config.gb_addr_config_fields.num_pipes;
-		plane_state->tiling_info.gfx9.num_banks =
+		tiling_info->gfx9.num_banks =
 			adev->gfx.config.gb_addr_config_fields.num_banks;
-		plane_state->tiling_info.gfx9.pipe_interleave =
+		tiling_info->gfx9.pipe_interleave =
 			adev->gfx.config.gb_addr_config_fields.pipe_interleave_size;
-		plane_state->tiling_info.gfx9.num_shader_engines =
+		tiling_info->gfx9.num_shader_engines =
 			adev->gfx.config.gb_addr_config_fields.num_se;
-		plane_state->tiling_info.gfx9.max_compressed_frags =
+		tiling_info->gfx9.max_compressed_frags =
 			adev->gfx.config.gb_addr_config_fields.max_compress_frags;
-		plane_state->tiling_info.gfx9.num_rb_per_se =
+		tiling_info->gfx9.num_rb_per_se =
 			adev->gfx.config.gb_addr_config_fields.num_rb_per_se;
-		plane_state->tiling_info.gfx9.swizzle =
+		tiling_info->gfx9.swizzle =
 			AMDGPU_TILING_GET(tiling_flags, SWIZZLE_MODE);
-		plane_state->tiling_info.gfx9.shaderEnable = 1;
+		tiling_info->gfx9.shaderEnable = 1;
 
-		fill_plane_dcc_attributes(adev, amdgpu_fb, plane_state,
-					  tiling_flags);
+		ret = fill_plane_dcc_attributes(adev, afb, format, rotation,
+						plane_size, tiling_info,
+						tiling_flags, dcc, address);
+		if (ret)
+			return ret;
 	}
 
-	plane_state->visible = true;
-	plane_state->scaling_quality.h_taps_c = 0;
-	plane_state->scaling_quality.v_taps_c = 0;
+	return 0;
+}
 
-	/* is this needed? is plane_state zeroed at allocation? */
-	plane_state->scaling_quality.h_taps = 0;
-	plane_state->scaling_quality.v_taps = 0;
-	plane_state->stereo_format = PLANE_STEREO_FORMAT_NONE;
+static void
+fill_blending_from_plane_state(const struct drm_plane_state *plane_state,
+			       bool *per_pixel_alpha, bool *global_alpha,
+			       int *global_alpha_value)
+{
+	*per_pixel_alpha = false;
+	*global_alpha = false;
+	*global_alpha_value = 0xff;
 
-	return ret;
+	if (plane_state->plane->type != DRM_PLANE_TYPE_OVERLAY)
+		return;
 
+	if (plane_state->pixel_blend_mode == DRM_MODE_BLEND_PREMULTI) {
+		static const uint32_t alpha_formats[] = {
+			DRM_FORMAT_ARGB8888,
+			DRM_FORMAT_RGBA8888,
+			DRM_FORMAT_ABGR8888,
+		};
+		uint32_t format = plane_state->fb->format->format;
+		unsigned int i;
+
+		for (i = 0; i < ARRAY_SIZE(alpha_formats); ++i) {
+			if (format == alpha_formats[i]) {
+				*per_pixel_alpha = true;
+				break;
+			}
+		}
+	}
+
+	if (plane_state->alpha < 0xffff) {
+		*global_alpha = true;
+		*global_alpha_value = plane_state->alpha >> 8;
+	}
 }
 
-static int fill_plane_attributes(struct amdgpu_device *adev,
-				 struct dc_plane_state *dc_plane_state,
-				 struct drm_plane_state *plane_state,
-				 struct drm_crtc_state *crtc_state)
+static int
+fill_plane_color_attributes(const struct drm_plane_state *plane_state,
+			    const enum surface_pixel_format format,
+			    enum dc_color_space *color_space)
 {
-	const struct amdgpu_framebuffer *amdgpu_fb =
+	bool full_range;
+
+	*color_space = COLOR_SPACE_SRGB;
+
+	/* DRM color properties only affect non-RGB formats. */
+	if (format < SURFACE_PIXEL_FORMAT_VIDEO_BEGIN)
+		return 0;
+
+	full_range = (plane_state->color_range == DRM_COLOR_YCBCR_FULL_RANGE);
+
+	switch (plane_state->color_encoding) {
+	case DRM_COLOR_YCBCR_BT601:
+		if (full_range)
+			*color_space = COLOR_SPACE_YCBCR601;
+		else
+			*color_space = COLOR_SPACE_YCBCR601_LIMITED;
+		break;
+
+	case DRM_COLOR_YCBCR_BT709:
+		if (full_range)
+			*color_space = COLOR_SPACE_YCBCR709;
+		else
+			*color_space = COLOR_SPACE_YCBCR709_LIMITED;
+		break;
+
+	case DRM_COLOR_YCBCR_BT2020:
+		if (full_range)
+			*color_space = COLOR_SPACE_2020_YCBCR;
+		else
+			return -EINVAL;
+		break;
+
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int
+fill_dc_plane_info_and_addr(struct amdgpu_device *adev,
+			    const struct drm_plane_state *plane_state,
+			    const uint64_t tiling_flags,
+			    struct dc_plane_info *plane_info,
+			    struct dc_plane_address *address)
+{
+	const struct drm_framebuffer *fb = plane_state->fb;
+	const struct amdgpu_framebuffer *afb =
 		to_amdgpu_framebuffer(plane_state->fb);
-	const struct drm_crtc *crtc = plane_state->crtc;
-	int ret = 0;
+	struct drm_format_name_buf format_name;
+	int ret;
 
-	if (!fill_rects_from_plane_state(plane_state, dc_plane_state))
+	memset(plane_info, 0, sizeof(*plane_info));
+
+	switch (fb->format->format) {
+	case DRM_FORMAT_C8:
+		plane_info->format =
+			SURFACE_PIXEL_FORMAT_GRPH_PALETA_256_COLORS;
+		break;
+	case DRM_FORMAT_RGB565:
+		plane_info->format = SURFACE_PIXEL_FORMAT_GRPH_RGB565;
+		break;
+	case DRM_FORMAT_XRGB8888:
+	case DRM_FORMAT_ARGB8888:
+		plane_info->format = SURFACE_PIXEL_FORMAT_GRPH_ARGB8888;
+		break;
+	case DRM_FORMAT_XRGB2101010:
+	case DRM_FORMAT_ARGB2101010:
+		plane_info->format = SURFACE_PIXEL_FORMAT_GRPH_ARGB2101010;
+		break;
+	case DRM_FORMAT_XBGR2101010:
+	case DRM_FORMAT_ABGR2101010:
+		plane_info->format = SURFACE_PIXEL_FORMAT_GRPH_ABGR2101010;
+		break;
+	case DRM_FORMAT_XBGR8888:
+	case DRM_FORMAT_ABGR8888:
+		plane_info->format = SURFACE_PIXEL_FORMAT_GRPH_ABGR8888;
+		break;
+	case DRM_FORMAT_NV21:
+		plane_info->format = SURFACE_PIXEL_FORMAT_VIDEO_420_YCbCr;
+		break;
+	case DRM_FORMAT_NV12:
+		plane_info->format = SURFACE_PIXEL_FORMAT_VIDEO_420_YCrCb;
+		break;
+	default:
+		DRM_ERROR(
+			"Unsupported screen format %s\n",
+			drm_get_format_name(fb->format->format, &format_name));
 		return -EINVAL;
+	}
+
+	switch (plane_state->rotation & DRM_MODE_ROTATE_MASK) {
+	case DRM_MODE_ROTATE_0:
+		plane_info->rotation = ROTATION_ANGLE_0;
+		break;
+	case DRM_MODE_ROTATE_90:
+		plane_info->rotation = ROTATION_ANGLE_90;
+		break;
+	case DRM_MODE_ROTATE_180:
+		plane_info->rotation = ROTATION_ANGLE_180;
+		break;
+	case DRM_MODE_ROTATE_270:
+		plane_info->rotation = ROTATION_ANGLE_270;
+		break;
+	default:
+		plane_info->rotation = ROTATION_ANGLE_0;
+		break;
+	}
 
-	ret = fill_plane_attributes_from_fb(
-		crtc->dev->dev_private,
-		dc_plane_state,
-		amdgpu_fb);
+	plane_info->visible = true;
+	plane_info->stereo_format = PLANE_STEREO_FORMAT_NONE;
 
+	ret = fill_plane_color_attributes(plane_state, plane_info->format,
+					  &plane_info->color_space);
 	if (ret)
 		return ret;
 
+	ret = fill_plane_buffer_attributes(adev, afb, plane_info->format,
+					   plane_info->rotation, tiling_flags,
+					   &plane_info->tiling_info,
+					   &plane_info->plane_size,
+					   &plane_info->dcc, address);
+	if (ret)
+		return ret;
+
+	fill_blending_from_plane_state(
+		plane_state, &plane_info->per_pixel_alpha,
+		&plane_info->global_alpha, &plane_info->global_alpha_value);
+
+	return 0;
+}
+
+static int fill_dc_plane_attributes(struct amdgpu_device *adev,
+				    struct dc_plane_state *dc_plane_state,
+				    struct drm_plane_state *plane_state,
+				    struct drm_crtc_state *crtc_state)
+{
+	const struct amdgpu_framebuffer *amdgpu_fb =
+		to_amdgpu_framebuffer(plane_state->fb);
+	struct dc_scaling_info scaling_info;
+	struct dc_plane_info plane_info;
+	uint64_t tiling_flags;
+	int ret;
+
+	ret = fill_dc_scaling_info(plane_state, &scaling_info);
+	if (ret)
+		return ret;
+
+	dc_plane_state->src_rect = scaling_info.src_rect;
+	dc_plane_state->dst_rect = scaling_info.dst_rect;
+	dc_plane_state->clip_rect = scaling_info.clip_rect;
+	dc_plane_state->scaling_quality = scaling_info.scaling_quality;
+
+	ret = get_fb_info(amdgpu_fb, &tiling_flags);
+	if (ret)
+		return ret;
+
+	ret = fill_dc_plane_info_and_addr(adev, plane_state, tiling_flags,
+					  &plane_info,
+					  &dc_plane_state->address);
+	if (ret)
+		return ret;
+
+	dc_plane_state->format = plane_info.format;
+	dc_plane_state->color_space = plane_info.color_space;
+	dc_plane_state->format = plane_info.format;
+	dc_plane_state->plane_size = plane_info.plane_size;
+	dc_plane_state->rotation = plane_info.rotation;
+	dc_plane_state->horizontal_mirror = plane_info.horizontal_mirror;
+	dc_plane_state->stereo_format = plane_info.stereo_format;
+	dc_plane_state->tiling_info = plane_info.tiling_info;
+	dc_plane_state->visible = plane_info.visible;
+	dc_plane_state->per_pixel_alpha = plane_info.per_pixel_alpha;
+	dc_plane_state->global_alpha = plane_info.global_alpha;
+	dc_plane_state->global_alpha_value = plane_info.global_alpha_value;
+	dc_plane_state->dcc = plane_info.dcc;
+
 	/*
 	 * Always set input transfer function, since plane state is refreshed
 	 * every time.
@@ -3124,6 +3461,8 @@ dm_crtc_duplicate_state(struct drm_crtc *crtc)
 		dc_stream_retain(state->stream);
 	}
 
+	state->active_planes = cur->active_planes;
+	state->interrupts_enabled = cur->interrupts_enabled;
 	state->vrr_params = cur->vrr_params;
 	state->vrr_infopacket = cur->vrr_infopacket;
 	state->abm_level = cur->abm_level;
@@ -3136,12 +3475,41 @@ dm_crtc_duplicate_state(struct drm_crtc *crtc)
 	return &state->base;
 }
 
+static inline int dm_set_vupdate_irq(struct drm_crtc *crtc, bool enable)
+{
+	enum dc_irq_source irq_source;
+	struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc);
+	struct amdgpu_device *adev = crtc->dev->dev_private;
+	int rc;
+
+	irq_source = IRQ_TYPE_VUPDATE + acrtc->otg_inst;
+
+	rc = dc_interrupt_set(adev->dm.dc, irq_source, enable) ? 0 : -EBUSY;
+
+	DRM_DEBUG_DRIVER("crtc %d - vupdate irq %sabling: r=%d\n",
+			 acrtc->crtc_id, enable ? "en" : "dis", rc);
+	return rc;
+}
 
 static inline int dm_set_vblank(struct drm_crtc *crtc, bool enable)
 {
 	enum dc_irq_source irq_source;
 	struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc);
 	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct dm_crtc_state *acrtc_state = to_dm_crtc_state(crtc->state);
+	int rc = 0;
+
+	if (enable) {
+		/* vblank irq on -> Only need vupdate irq in vrr mode */
+		if (amdgpu_dm_vrr_active(acrtc_state))
+			rc = dm_set_vupdate_irq(crtc, true);
+	} else {
+		/* vblank irq off -> vupdate irq off */
+		rc = dm_set_vupdate_irq(crtc, false);
+	}
+
+	if (rc)
+		return rc;
 
 	irq_source = IRQ_TYPE_VBLANK + acrtc->otg_inst;
 	return dc_interrupt_set(adev->dm.dc, irq_source, enable) ? 0 : -EBUSY;
@@ -3519,6 +3887,76 @@ static void dm_crtc_helper_disable(struct drm_crtc *crtc)
 {
 }
 
+static bool does_crtc_have_active_cursor(struct drm_crtc_state *new_crtc_state)
+{
+	struct drm_device *dev = new_crtc_state->crtc->dev;
+	struct drm_plane *plane;
+
+	drm_for_each_plane_mask(plane, dev, new_crtc_state->plane_mask) {
+		if (plane->type == DRM_PLANE_TYPE_CURSOR)
+			return true;
+	}
+
+	return false;
+}
+
+static int count_crtc_active_planes(struct drm_crtc_state *new_crtc_state)
+{
+	struct drm_atomic_state *state = new_crtc_state->state;
+	struct drm_plane *plane;
+	int num_active = 0;
+
+	drm_for_each_plane_mask(plane, state->dev, new_crtc_state->plane_mask) {
+		struct drm_plane_state *new_plane_state;
+
+		/* Cursor planes are "fake". */
+		if (plane->type == DRM_PLANE_TYPE_CURSOR)
+			continue;
+
+		new_plane_state = drm_atomic_get_new_plane_state(state, plane);
+
+		if (!new_plane_state) {
+			/*
+			 * The plane is enable on the CRTC and hasn't changed
+			 * state. This means that it previously passed
+			 * validation and is therefore enabled.
+			 */
+			num_active += 1;
+			continue;
+		}
+
+		/* We need a framebuffer to be considered enabled. */
+		num_active += (new_plane_state->fb != NULL);
+	}
+
+	return num_active;
+}
+
+/*
+ * Sets whether interrupts should be enabled on a specific CRTC.
+ * We require that the stream be enabled and that there exist active
+ * DC planes on the stream.
+ */
+static void
+dm_update_crtc_interrupt_state(struct drm_crtc *crtc,
+			       struct drm_crtc_state *new_crtc_state)
+{
+	struct dm_crtc_state *dm_new_crtc_state =
+		to_dm_crtc_state(new_crtc_state);
+
+	dm_new_crtc_state->active_planes = 0;
+	dm_new_crtc_state->interrupts_enabled = false;
+
+	if (!dm_new_crtc_state->stream)
+		return;
+
+	dm_new_crtc_state->active_planes =
+		count_crtc_active_planes(new_crtc_state);
+
+	dm_new_crtc_state->interrupts_enabled =
+		dm_new_crtc_state->active_planes > 0;
+}
+
 static int dm_crtc_helper_atomic_check(struct drm_crtc *crtc,
 				       struct drm_crtc_state *state)
 {
@@ -3527,6 +3965,14 @@ static int dm_crtc_helper_atomic_check(struct drm_crtc *crtc,
 	struct dm_crtc_state *dm_crtc_state = to_dm_crtc_state(state);
 	int ret = -EINVAL;
 
+	/*
+	 * Update interrupt state for the CRTC. This needs to happen whenever
+	 * the CRTC has changed or whenever any of its planes have changed.
+	 * Atomic check satisfies both of these requirements since the CRTC
+	 * is added to the state by DRM during drm_atomic_helper_check_planes.
+	 */
+	dm_update_crtc_interrupt_state(crtc, state);
+
 	if (unlikely(!dm_crtc_state->stream &&
 		     modeset_required(state, NULL, dm_crtc_state->stream))) {
 		WARN_ON(1);
@@ -3537,6 +3983,15 @@ static int dm_crtc_helper_atomic_check(struct drm_crtc *crtc,
 	if (!dm_crtc_state->stream)
 		return 0;
 
+	/*
+	 * We want at least one hardware plane enabled to use
+	 * the stream with a cursor enabled.
+	 */
+	if (state->enable && state->active &&
+	    does_crtc_have_active_cursor(state) &&
+	    dm_crtc_state->active_planes == 0)
+		return -EINVAL;
+
 	if (dc_validate_stream(dc, dm_crtc_state->stream) == DC_OK)
 		return 0;
 
@@ -3583,11 +4038,8 @@ static void dm_drm_plane_reset(struct drm_plane *plane)
 	amdgpu_state = kzalloc(sizeof(*amdgpu_state), GFP_KERNEL);
 	WARN_ON(amdgpu_state == NULL);
 
-	if (amdgpu_state) {
-		plane->state = &amdgpu_state->base;
-		plane->state->plane = plane;
-		plane->state->rotation = DRM_MODE_ROTATE_0;
-	}
+	if (amdgpu_state)
+		__drm_atomic_helper_plane_reset(plane, &amdgpu_state->base);
 }
 
 static struct drm_plane_state *
@@ -3637,10 +4089,8 @@ static int dm_plane_helper_prepare_fb(struct drm_plane *plane,
 	struct drm_gem_object *obj;
 	struct amdgpu_device *adev;
 	struct amdgpu_bo *rbo;
-	uint64_t chroma_addr = 0;
 	struct dm_plane_state *dm_plane_state_new, *dm_plane_state_old;
-	uint64_t tiling_flags, dcc_address;
-	unsigned int awidth;
+	uint64_t tiling_flags;
 	uint32_t domain;
 	int r;
 
@@ -3693,29 +4143,11 @@ static int dm_plane_helper_prepare_fb(struct drm_plane *plane,
 			dm_plane_state_old->dc_state != dm_plane_state_new->dc_state) {
 		struct dc_plane_state *plane_state = dm_plane_state_new->dc_state;
 
-		if (plane_state->format < SURFACE_PIXEL_FORMAT_VIDEO_BEGIN) {
-			plane_state->address.grph.addr.low_part = lower_32_bits(afb->address);
-			plane_state->address.grph.addr.high_part = upper_32_bits(afb->address);
-
-			dcc_address =
-				get_dcc_address(afb->address, tiling_flags);
-			plane_state->address.grph.meta_addr.low_part =
-				lower_32_bits(dcc_address);
-			plane_state->address.grph.meta_addr.high_part =
-				upper_32_bits(dcc_address);
-		} else {
-			awidth = ALIGN(new_state->fb->width, 64);
-			plane_state->address.type = PLN_ADDR_TYPE_VIDEO_PROGRESSIVE;
-			plane_state->address.video_progressive.luma_addr.low_part
-							= lower_32_bits(afb->address);
-			plane_state->address.video_progressive.luma_addr.high_part
-							= upper_32_bits(afb->address);
-			chroma_addr = afb->address + (u64)awidth * new_state->fb->height;
-			plane_state->address.video_progressive.chroma_addr.low_part
-							= lower_32_bits(chroma_addr);
-			plane_state->address.video_progressive.chroma_addr.high_part
-							= upper_32_bits(chroma_addr);
-		}
+		fill_plane_buffer_attributes(
+			adev, afb, plane_state->format, plane_state->rotation,
+			tiling_flags, &plane_state->tiling_info,
+			&plane_state->plane_size, &plane_state->dcc,
+			&plane_state->address);
 	}
 
 	return 0;
@@ -3747,13 +4179,18 @@ static int dm_plane_atomic_check(struct drm_plane *plane,
 {
 	struct amdgpu_device *adev = plane->dev->dev_private;
 	struct dc *dc = adev->dm.dc;
-	struct dm_plane_state *dm_plane_state = to_dm_plane_state(state);
+	struct dm_plane_state *dm_plane_state;
+	struct dc_scaling_info scaling_info;
+	int ret;
+
+	dm_plane_state = to_dm_plane_state(state);
 
 	if (!dm_plane_state->dc_state)
 		return 0;
 
-	if (!fill_rects_from_plane_state(state, dm_plane_state->dc_state))
-		return -EINVAL;
+	ret = fill_dc_scaling_info(state, &scaling_info);
+	if (ret)
+		return ret;
 
 	if (dc_validate_plane(dc, dm_plane_state->dc_state) == DC_OK)
 		return 0;
@@ -3826,64 +4263,115 @@ static const uint32_t rgb_formats[] = {
 	DRM_FORMAT_ABGR2101010,
 	DRM_FORMAT_XBGR8888,
 	DRM_FORMAT_ABGR8888,
+	DRM_FORMAT_RGB565,
 };
 
-static const uint32_t yuv_formats[] = {
-	DRM_FORMAT_NV12,
-	DRM_FORMAT_NV21,
+static const uint32_t overlay_formats[] = {
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_RGBA8888,
+	DRM_FORMAT_XBGR8888,
+	DRM_FORMAT_ABGR8888,
+	DRM_FORMAT_RGB565
 };
 
 static const u32 cursor_formats[] = {
 	DRM_FORMAT_ARGB8888
 };
 
-static int amdgpu_dm_plane_init(struct amdgpu_display_manager *dm,
-				struct drm_plane *plane,
-				unsigned long possible_crtcs)
+static int get_plane_formats(const struct drm_plane *plane,
+			     const struct dc_plane_cap *plane_cap,
+			     uint32_t *formats, int max_formats)
 {
-	int res = -EPERM;
+	int i, num_formats = 0;
+
+	/*
+	 * TODO: Query support for each group of formats directly from
+	 * DC plane caps. This will require adding more formats to the
+	 * caps list.
+	 */
 
 	switch (plane->type) {
 	case DRM_PLANE_TYPE_PRIMARY:
-		res = drm_universal_plane_init(
-				dm->adev->ddev,
-				plane,
-				possible_crtcs,
-				&dm_plane_funcs,
-				rgb_formats,
-				ARRAY_SIZE(rgb_formats),
-				NULL, plane->type, NULL);
+		for (i = 0; i < ARRAY_SIZE(rgb_formats); ++i) {
+			if (num_formats >= max_formats)
+				break;
+
+			formats[num_formats++] = rgb_formats[i];
+		}
+
+		if (plane_cap && plane_cap->pixel_format_support.nv12)
+			formats[num_formats++] = DRM_FORMAT_NV12;
 		break;
+
 	case DRM_PLANE_TYPE_OVERLAY:
-		res = drm_universal_plane_init(
-				dm->adev->ddev,
-				plane,
-				possible_crtcs,
-				&dm_plane_funcs,
-				yuv_formats,
-				ARRAY_SIZE(yuv_formats),
-				NULL, plane->type, NULL);
+		for (i = 0; i < ARRAY_SIZE(overlay_formats); ++i) {
+			if (num_formats >= max_formats)
+				break;
+
+			formats[num_formats++] = overlay_formats[i];
+		}
 		break;
+
 	case DRM_PLANE_TYPE_CURSOR:
-		res = drm_universal_plane_init(
-				dm->adev->ddev,
-				plane,
-				possible_crtcs,
-				&dm_plane_funcs,
-				cursor_formats,
-				ARRAY_SIZE(cursor_formats),
-				NULL, plane->type, NULL);
+		for (i = 0; i < ARRAY_SIZE(cursor_formats); ++i) {
+			if (num_formats >= max_formats)
+				break;
+
+			formats[num_formats++] = cursor_formats[i];
+		}
 		break;
 	}
 
+	return num_formats;
+}
+
+static int amdgpu_dm_plane_init(struct amdgpu_display_manager *dm,
+				struct drm_plane *plane,
+				unsigned long possible_crtcs,
+				const struct dc_plane_cap *plane_cap)
+{
+	uint32_t formats[32];
+	int num_formats;
+	int res = -EPERM;
+
+	num_formats = get_plane_formats(plane, plane_cap, formats,
+					ARRAY_SIZE(formats));
+
+	res = drm_universal_plane_init(dm->adev->ddev, plane, possible_crtcs,
+				       &dm_plane_funcs, formats, num_formats,
+				       NULL, plane->type, NULL);
+	if (res)
+		return res;
+
+	if (plane->type == DRM_PLANE_TYPE_OVERLAY &&
+	    plane_cap && plane_cap->per_pixel_alpha) {
+		unsigned int blend_caps = BIT(DRM_MODE_BLEND_PIXEL_NONE) |
+					  BIT(DRM_MODE_BLEND_PREMULTI);
+
+		drm_plane_create_alpha_property(plane);
+		drm_plane_create_blend_mode_property(plane, blend_caps);
+	}
+
+	if (plane->type == DRM_PLANE_TYPE_PRIMARY &&
+	    plane_cap && plane_cap->pixel_format_support.nv12) {
+		/* This only affects YUV formats. */
+		drm_plane_create_color_properties(
+			plane,
+			BIT(DRM_COLOR_YCBCR_BT601) |
+			BIT(DRM_COLOR_YCBCR_BT709),
+			BIT(DRM_COLOR_YCBCR_LIMITED_RANGE) |
+			BIT(DRM_COLOR_YCBCR_FULL_RANGE),
+			DRM_COLOR_YCBCR_BT709, DRM_COLOR_YCBCR_LIMITED_RANGE);
+	}
+
 	drm_plane_helper_add(plane, &dm_plane_helper_funcs);
 
 	/* Create (reset) the plane state */
 	if (plane->funcs->reset)
 		plane->funcs->reset(plane);
 
-
-	return res;
+	return 0;
 }
 
 static int amdgpu_dm_crtc_init(struct amdgpu_display_manager *dm,
@@ -3900,7 +4388,7 @@ static int amdgpu_dm_crtc_init(struct amdgpu_display_manager *dm,
 		goto fail;
 
 	cursor_plane->type = DRM_PLANE_TYPE_CURSOR;
-	res = amdgpu_dm_plane_init(dm, cursor_plane, 0);
+	res = amdgpu_dm_plane_init(dm, cursor_plane, 0, NULL);
 
 	acrtc = kzalloc(sizeof(struct amdgpu_crtc), GFP_KERNEL);
 	if (!acrtc)
@@ -4334,6 +4822,8 @@ static int amdgpu_dm_connector_init(struct amdgpu_display_manager *dm,
 		DRM_ERROR("Failed to create debugfs for connector");
 		goto out_free;
 	}
+	aconnector->debugfs_dpcd_address = 0;
+	aconnector->debugfs_dpcd_size = 0;
 #endif
 
 	if (connector_type == DRM_MODE_CONNECTOR_DisplayPort
@@ -4473,9 +4963,13 @@ static int get_cursor_position(struct drm_plane *plane, struct drm_crtc *crtc,
 
 	x = plane->state->crtc_x;
 	y = plane->state->crtc_y;
-	/* avivo cursor are offset into the total surface */
-	x += crtc->primary->state->src_x >> 16;
-	y += crtc->primary->state->src_y >> 16;
+
+	if (crtc->primary->state) {
+		/* avivo cursor are offset into the total surface */
+		x += crtc->primary->state->src_x >> 16;
+		y += crtc->primary->state->src_y >> 16;
+	}
+
 	if (x < 0) {
 		xorigin = min(-x, amdgpu_crtc->max_cursor_width - 1);
 		x = 0;
@@ -4582,9 +5076,10 @@ static void update_freesync_state_on_stream(
 	struct dc_plane_state *surface,
 	u32 flip_timestamp_in_us)
 {
-	struct mod_vrr_params vrr_params = new_crtc_state->vrr_params;
+	struct mod_vrr_params vrr_params;
 	struct dc_info_packet vrr_infopacket = {0};
-	struct mod_freesync_config config = new_crtc_state->freesync_config;
+	struct amdgpu_device *adev = dm->adev;
+	unsigned long flags;
 
 	if (!new_stream)
 		return;
@@ -4597,19 +5092,8 @@ static void update_freesync_state_on_stream(
 	if (!new_stream->timing.h_total || !new_stream->timing.v_total)
 		return;
 
-	if (new_crtc_state->vrr_supported &&
-	    config.min_refresh_in_uhz &&
-	    config.max_refresh_in_uhz) {
-		config.state = new_crtc_state->base.vrr_enabled ?
-			VRR_STATE_ACTIVE_VARIABLE :
-			VRR_STATE_INACTIVE;
-	} else {
-		config.state = VRR_STATE_UNSUPPORTED;
-	}
-
-	mod_freesync_build_vrr_params(dm->freesync_module,
-				      new_stream,
-				      &config, &vrr_params);
+	spin_lock_irqsave(&adev->ddev->event_lock, flags);
+	vrr_params = new_crtc_state->vrr_params;
 
 	if (surface) {
 		mod_freesync_handle_preflip(
@@ -4618,6 +5102,12 @@ static void update_freesync_state_on_stream(
 			new_stream,
 			flip_timestamp_in_us,
 			&vrr_params);
+
+		if (adev->family < AMDGPU_FAMILY_AI &&
+		    amdgpu_dm_vrr_active(new_crtc_state)) {
+			mod_freesync_handle_v_update(dm->freesync_module,
+						     new_stream, &vrr_params);
+		}
 	}
 
 	mod_freesync_build_vrr_infopacket(
@@ -4649,6 +5139,100 @@ static void update_freesync_state_on_stream(
 			      new_crtc_state->base.crtc->base.id,
 			      (int)new_crtc_state->base.vrr_enabled,
 			      (int)vrr_params.state);
+
+	spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
+}
+
+static void pre_update_freesync_state_on_stream(
+	struct amdgpu_display_manager *dm,
+	struct dm_crtc_state *new_crtc_state)
+{
+	struct dc_stream_state *new_stream = new_crtc_state->stream;
+	struct mod_vrr_params vrr_params;
+	struct mod_freesync_config config = new_crtc_state->freesync_config;
+	struct amdgpu_device *adev = dm->adev;
+	unsigned long flags;
+
+	if (!new_stream)
+		return;
+
+	/*
+	 * TODO: Determine why min/max totals and vrefresh can be 0 here.
+	 * For now it's sufficient to just guard against these conditions.
+	 */
+	if (!new_stream->timing.h_total || !new_stream->timing.v_total)
+		return;
+
+	spin_lock_irqsave(&adev->ddev->event_lock, flags);
+	vrr_params = new_crtc_state->vrr_params;
+
+	if (new_crtc_state->vrr_supported &&
+	    config.min_refresh_in_uhz &&
+	    config.max_refresh_in_uhz) {
+		config.state = new_crtc_state->base.vrr_enabled ?
+			VRR_STATE_ACTIVE_VARIABLE :
+			VRR_STATE_INACTIVE;
+	} else {
+		config.state = VRR_STATE_UNSUPPORTED;
+	}
+
+	mod_freesync_build_vrr_params(dm->freesync_module,
+				      new_stream,
+				      &config, &vrr_params);
+
+	new_crtc_state->freesync_timing_changed |=
+		(memcmp(&new_crtc_state->vrr_params.adjust,
+			&vrr_params.adjust,
+			sizeof(vrr_params.adjust)) != 0);
+
+	new_crtc_state->vrr_params = vrr_params;
+	spin_unlock_irqrestore(&adev->ddev->event_lock, flags);
+}
+
+static void amdgpu_dm_handle_vrr_transition(struct dm_crtc_state *old_state,
+					    struct dm_crtc_state *new_state)
+{
+	bool old_vrr_active = amdgpu_dm_vrr_active(old_state);
+	bool new_vrr_active = amdgpu_dm_vrr_active(new_state);
+
+	if (!old_vrr_active && new_vrr_active) {
+		/* Transition VRR inactive -> active:
+		 * While VRR is active, we must not disable vblank irq, as a
+		 * reenable after disable would compute bogus vblank/pflip
+		 * timestamps if it likely happened inside display front-porch.
+		 *
+		 * We also need vupdate irq for the actual core vblank handling
+		 * at end of vblank.
+		 */
+		dm_set_vupdate_irq(new_state->base.crtc, true);
+		drm_crtc_vblank_get(new_state->base.crtc);
+		DRM_DEBUG_DRIVER("%s: crtc=%u VRR off->on: Get vblank ref\n",
+				 __func__, new_state->base.crtc->base.id);
+	} else if (old_vrr_active && !new_vrr_active) {
+		/* Transition VRR active -> inactive:
+		 * Allow vblank irq disable again for fixed refresh rate.
+		 */
+		dm_set_vupdate_irq(new_state->base.crtc, false);
+		drm_crtc_vblank_put(new_state->base.crtc);
+		DRM_DEBUG_DRIVER("%s: crtc=%u VRR on->off: Drop vblank ref\n",
+				 __func__, new_state->base.crtc->base.id);
+	}
+}
+
+static void amdgpu_dm_commit_cursors(struct drm_atomic_state *state)
+{
+	struct drm_plane *plane;
+	struct drm_plane_state *old_plane_state, *new_plane_state;
+	int i;
+
+	/*
+	 * TODO: Make this per-stream so we don't issue redundant updates for
+	 * commits with multiple streams.
+	 */
+	for_each_oldnew_plane_in_state(state, plane, old_plane_state,
+				       new_plane_state, i)
+		if (plane->type == DRM_PLANE_TYPE_CURSOR)
+			handle_cursor_update(plane, old_plane_state);
 }
 
 static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
@@ -4656,7 +5240,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 				    struct drm_device *dev,
 				    struct amdgpu_display_manager *dm,
 				    struct drm_crtc *pcrtc,
-				    bool *wait_for_vblank)
+				    bool wait_for_vblank)
 {
 	uint32_t i, r;
 	uint64_t timestamp_ns;
@@ -4668,42 +5252,42 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 	struct dm_crtc_state *acrtc_state = to_dm_crtc_state(new_pcrtc_state);
 	struct dm_crtc_state *dm_old_crtc_state =
 			to_dm_crtc_state(drm_atomic_get_old_crtc_state(state, pcrtc));
-	int flip_count = 0, planes_count = 0, vpos, hpos;
+	int planes_count = 0, vpos, hpos;
 	unsigned long flags;
 	struct amdgpu_bo *abo;
-	uint64_t tiling_flags, dcc_address;
-	uint32_t target, target_vblank;
-	uint64_t last_flip_vblank;
-	bool vrr_active = acrtc_state->freesync_config.state == VRR_STATE_ACTIVE_VARIABLE;
-
-	struct {
-		struct dc_surface_update surface_updates[MAX_SURFACES];
-		struct dc_flip_addrs flip_addrs[MAX_SURFACES];
-		struct dc_stream_update stream_update;
-	} *flip;
-
+	uint64_t tiling_flags;
+	uint32_t target_vblank, last_flip_vblank;
+	bool vrr_active = amdgpu_dm_vrr_active(acrtc_state);
+	bool pflip_present = false;
 	struct {
 		struct dc_surface_update surface_updates[MAX_SURFACES];
 		struct dc_plane_info plane_infos[MAX_SURFACES];
 		struct dc_scaling_info scaling_infos[MAX_SURFACES];
+		struct dc_flip_addrs flip_addrs[MAX_SURFACES];
 		struct dc_stream_update stream_update;
-	} *full;
+	} *bundle;
 
-	flip = kzalloc(sizeof(*flip), GFP_KERNEL);
-	full = kzalloc(sizeof(*full), GFP_KERNEL);
+	bundle = kzalloc(sizeof(*bundle), GFP_KERNEL);
 
-	if (!flip || !full) {
-		dm_error("Failed to allocate update bundles\n");
+	if (!bundle) {
+		dm_error("Failed to allocate update bundle\n");
 		goto cleanup;
 	}
 
+	/*
+	 * Disable the cursor first if we're disabling all the planes.
+	 * It'll remain on the screen after the planes are re-enabled
+	 * if we don't.
+	 */
+	if (acrtc_state->active_planes == 0)
+		amdgpu_dm_commit_cursors(state);
+
 	/* update planes when needed */
 	for_each_oldnew_plane_in_state(state, plane, old_plane_state, new_plane_state, i) {
 		struct drm_crtc *crtc = new_plane_state->crtc;
 		struct drm_crtc_state *new_crtc_state;
 		struct drm_framebuffer *fb = new_plane_state->fb;
-		struct amdgpu_framebuffer *afb = to_amdgpu_framebuffer(fb);
-		bool pflip_needed;
+		bool plane_needs_flip;
 		struct dc_plane_state *dc_plane;
 		struct dm_plane_state *dm_new_plane_state = to_dm_plane_state(new_plane_state);
 
@@ -4718,122 +5302,96 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 		if (!new_crtc_state->active)
 			continue;
 
-		pflip_needed = old_plane_state->fb &&
-			old_plane_state->fb != new_plane_state->fb;
-
 		dc_plane = dm_new_plane_state->dc_state;
 
-		if (pflip_needed) {
-			/*
-			 * Assume even ONE crtc with immediate flip means
-			 * entire can't wait for VBLANK
-			 * TODO Check if it's correct
-			 */
-			if (new_pcrtc_state->pageflip_flags & DRM_MODE_PAGE_FLIP_ASYNC)
-				*wait_for_vblank = false;
+		bundle->surface_updates[planes_count].surface = dc_plane;
+		if (new_pcrtc_state->color_mgmt_changed) {
+			bundle->surface_updates[planes_count].gamma = dc_plane->gamma_correction;
+			bundle->surface_updates[planes_count].in_transfer_func = dc_plane->in_transfer_func;
+		}
 
-			/*
-			 * TODO This might fail and hence better not used, wait
-			 * explicitly on fences instead
-			 * and in general should be called for
-			 * blocking commit to as per framework helpers
-			 */
-			abo = gem_to_amdgpu_bo(fb->obj[0]);
-			r = amdgpu_bo_reserve(abo, true);
-			if (unlikely(r != 0))
-				DRM_ERROR("failed to reserve buffer before flip\n");
+		fill_dc_scaling_info(new_plane_state,
+				     &bundle->scaling_infos[planes_count]);
 
-			/*
-			 * Wait for all fences on this FB. Do limited wait to avoid
-			 * deadlock during GPU reset when this fence will not signal
-			 * but we hold reservation lock for the BO.
-			 */
-			r = reservation_object_wait_timeout_rcu(abo->tbo.resv,
-								true, false,
-								msecs_to_jiffies(5000));
-			if (unlikely(r == 0))
-				DRM_ERROR("Waiting for fences timed out.");
+		bundle->surface_updates[planes_count].scaling_info =
+			&bundle->scaling_infos[planes_count];
 
+		plane_needs_flip = old_plane_state->fb && new_plane_state->fb;
 
+		pflip_present = pflip_present || plane_needs_flip;
 
-			amdgpu_bo_get_tiling_flags(abo, &tiling_flags);
+		if (!plane_needs_flip) {
+			planes_count += 1;
+			continue;
+		}
 
-			amdgpu_bo_unreserve(abo);
+		abo = gem_to_amdgpu_bo(fb->obj[0]);
 
-			flip->flip_addrs[flip_count].address.grph.addr.low_part = lower_32_bits(afb->address);
-			flip->flip_addrs[flip_count].address.grph.addr.high_part = upper_32_bits(afb->address);
+		/*
+		 * Wait for all fences on this FB. Do limited wait to avoid
+		 * deadlock during GPU reset when this fence will not signal
+		 * but we hold reservation lock for the BO.
+		 */
+		r = reservation_object_wait_timeout_rcu(abo->tbo.resv, true,
+							false,
+							msecs_to_jiffies(5000));
+		if (unlikely(r <= 0))
+			DRM_ERROR("Waiting for fences timed out or interrupted!");
 
-			dcc_address = get_dcc_address(afb->address, tiling_flags);
-			flip->flip_addrs[flip_count].address.grph.meta_addr.low_part = lower_32_bits(dcc_address);
-			flip->flip_addrs[flip_count].address.grph.meta_addr.high_part = upper_32_bits(dcc_address);
+		/*
+		 * TODO This might fail and hence better not used, wait
+		 * explicitly on fences instead
+		 * and in general should be called for
+		 * blocking commit to as per framework helpers
+		 */
+		r = amdgpu_bo_reserve(abo, true);
+		if (unlikely(r != 0))
+			DRM_ERROR("failed to reserve buffer before flip\n");
 
-			flip->flip_addrs[flip_count].flip_immediate =
-					(crtc->state->pageflip_flags & DRM_MODE_PAGE_FLIP_ASYNC) != 0;
+		amdgpu_bo_get_tiling_flags(abo, &tiling_flags);
 
-			timestamp_ns = ktime_get_ns();
-			flip->flip_addrs[flip_count].flip_timestamp_in_us = div_u64(timestamp_ns, 1000);
-			flip->surface_updates[flip_count].flip_addr = &flip->flip_addrs[flip_count];
-			flip->surface_updates[flip_count].surface = dc_plane;
+		amdgpu_bo_unreserve(abo);
 
-			if (!flip->surface_updates[flip_count].surface) {
-				DRM_ERROR("No surface for CRTC: id=%d\n",
-						acrtc_attach->crtc_id);
-				continue;
-			}
+		fill_dc_plane_info_and_addr(
+			dm->adev, new_plane_state, tiling_flags,
+			&bundle->plane_infos[planes_count],
+			&bundle->flip_addrs[planes_count].address);
 
-			if (plane == pcrtc->primary)
-				update_freesync_state_on_stream(
-					dm,
-					acrtc_state,
-					acrtc_state->stream,
-					dc_plane,
-					flip->flip_addrs[flip_count].flip_timestamp_in_us);
+		bundle->surface_updates[planes_count].plane_info =
+			&bundle->plane_infos[planes_count];
 
-			DRM_DEBUG_DRIVER("%s Flipping to hi: 0x%x, low: 0x%x\n",
-					 __func__,
-					 flip->flip_addrs[flip_count].address.grph.addr.high_part,
-					 flip->flip_addrs[flip_count].address.grph.addr.low_part);
+		bundle->flip_addrs[planes_count].flip_immediate =
+				(crtc->state->pageflip_flags & DRM_MODE_PAGE_FLIP_ASYNC) != 0;
 
-			flip_count += 1;
-		}
+		timestamp_ns = ktime_get_ns();
+		bundle->flip_addrs[planes_count].flip_timestamp_in_us = div_u64(timestamp_ns, 1000);
+		bundle->surface_updates[planes_count].flip_addr = &bundle->flip_addrs[planes_count];
+		bundle->surface_updates[planes_count].surface = dc_plane;
 
-		full->surface_updates[planes_count].surface = dc_plane;
-		if (new_pcrtc_state->color_mgmt_changed) {
-			full->surface_updates[planes_count].gamma = dc_plane->gamma_correction;
-			full->surface_updates[planes_count].in_transfer_func = dc_plane->in_transfer_func;
+		if (!bundle->surface_updates[planes_count].surface) {
+			DRM_ERROR("No surface for CRTC: id=%d\n",
+					acrtc_attach->crtc_id);
+			continue;
 		}
 
+		if (plane == pcrtc->primary)
+			update_freesync_state_on_stream(
+				dm,
+				acrtc_state,
+				acrtc_state->stream,
+				dc_plane,
+				bundle->flip_addrs[planes_count].flip_timestamp_in_us);
 
-		full->scaling_infos[planes_count].scaling_quality = dc_plane->scaling_quality;
-		full->scaling_infos[planes_count].src_rect = dc_plane->src_rect;
-		full->scaling_infos[planes_count].dst_rect = dc_plane->dst_rect;
-		full->scaling_infos[planes_count].clip_rect = dc_plane->clip_rect;
-		full->surface_updates[planes_count].scaling_info = &full->scaling_infos[planes_count];
-
-
-		full->plane_infos[planes_count].color_space = dc_plane->color_space;
-		full->plane_infos[planes_count].format = dc_plane->format;
-		full->plane_infos[planes_count].plane_size = dc_plane->plane_size;
-		full->plane_infos[planes_count].rotation = dc_plane->rotation;
-		full->plane_infos[planes_count].horizontal_mirror = dc_plane->horizontal_mirror;
-		full->plane_infos[planes_count].stereo_format = dc_plane->stereo_format;
-		full->plane_infos[planes_count].tiling_info = dc_plane->tiling_info;
-		full->plane_infos[planes_count].visible = dc_plane->visible;
-		full->plane_infos[planes_count].per_pixel_alpha = dc_plane->per_pixel_alpha;
-		full->plane_infos[planes_count].dcc = dc_plane->dcc;
-		full->surface_updates[planes_count].plane_info = &full->plane_infos[planes_count];
+		DRM_DEBUG_DRIVER("%s Flipping to hi: 0x%x, low: 0x%x\n",
+				 __func__,
+				 bundle->flip_addrs[planes_count].address.grph.addr.high_part,
+				 bundle->flip_addrs[planes_count].address.grph.addr.low_part);
 
 		planes_count += 1;
 
 	}
 
-	/*
-	 * TODO: For proper atomic behaviour, we should be calling into DC once with
-	 * all the changes.  However, DC refuses to do pageflips and non-pageflip
-	 * changes in the same call.  Change DC to respect atomic behaviour,
-	 * hopefully eliminating dc_*_update structs in their entirety.
-	 */
-	if (flip_count) {
+	if (pflip_present) {
 		if (!vrr_active) {
 			/* Use old throttling in non-vrr fixed refresh rate mode
 			 * to keep flip scheduling based on target vblank counts
@@ -4841,7 +5399,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 			 * clients using the GLX_OML_sync_control extension or
 			 * DRI3/Present extension with defined target_msc.
 			 */
-			last_flip_vblank = drm_crtc_vblank_count(pcrtc);
+			last_flip_vblank = amdgpu_get_vblank_counter_kms(dm->ddev, acrtc_attach->crtc_id);
 		}
 		else {
 			/* For variable refresh rate mode only:
@@ -4857,11 +5415,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 			spin_unlock_irqrestore(&pcrtc->dev->event_lock, flags);
 		}
 
-		target = (uint32_t)last_flip_vblank + *wait_for_vblank;
-
-		/* Prepare wait for target vblank early - before the fence-waits */
-		target_vblank = target - (uint32_t)drm_crtc_vblank_count(pcrtc) +
-				amdgpu_get_vblank_counter_kms(pcrtc->dev, acrtc_attach->crtc_id);
+		target_vblank = last_flip_vblank + wait_for_vblank;
 
 		/*
 		 * Wait until we're out of the vertical blank period before the one
@@ -4892,54 +5446,102 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 		if (acrtc_state->stream) {
 
 			if (acrtc_state->freesync_timing_changed)
-				flip->stream_update.adjust =
+				bundle->stream_update.adjust =
 					&acrtc_state->stream->adjust;
 
 			if (acrtc_state->freesync_vrr_info_changed)
-				flip->stream_update.vrr_infopacket =
+				bundle->stream_update.vrr_infopacket =
 					&acrtc_state->stream->vrr_infopacket;
 		}
-
-		mutex_lock(&dm->dc_lock);
-		dc_commit_updates_for_stream(dm->dc,
-						     flip->surface_updates,
-						     flip_count,
-						     acrtc_state->stream,
-						     &flip->stream_update,
-						     dc_state);
-		mutex_unlock(&dm->dc_lock);
 	}
 
-	if (planes_count) {
+	/* Update the planes if changed or disable if we don't have any. */
+	if (planes_count || acrtc_state->active_planes == 0) {
 		if (new_pcrtc_state->mode_changed) {
-			full->stream_update.src = acrtc_state->stream->src;
-			full->stream_update.dst = acrtc_state->stream->dst;
+			bundle->stream_update.src = acrtc_state->stream->src;
+			bundle->stream_update.dst = acrtc_state->stream->dst;
 		}
 
 		if (new_pcrtc_state->color_mgmt_changed)
-			full->stream_update.out_transfer_func = acrtc_state->stream->out_transfer_func;
+			bundle->stream_update.out_transfer_func = acrtc_state->stream->out_transfer_func;
 
 		acrtc_state->stream->abm_level = acrtc_state->abm_level;
 		if (acrtc_state->abm_level != dm_old_crtc_state->abm_level)
-			full->stream_update.abm_level = &acrtc_state->abm_level;
+			bundle->stream_update.abm_level = &acrtc_state->abm_level;
 
 		mutex_lock(&dm->dc_lock);
 		dc_commit_updates_for_stream(dm->dc,
-						     full->surface_updates,
+						     bundle->surface_updates,
 						     planes_count,
 						     acrtc_state->stream,
-						     &full->stream_update,
+						     &bundle->stream_update,
 						     dc_state);
 		mutex_unlock(&dm->dc_lock);
 	}
 
-	for_each_oldnew_plane_in_state(state, plane, old_plane_state, new_plane_state, i)
-		if (plane->type == DRM_PLANE_TYPE_CURSOR)
-			handle_cursor_update(plane, old_plane_state);
+	/*
+	 * Update cursor state *after* programming all the planes.
+	 * This avoids redundant programming in the case where we're going
+	 * to be disabling a single plane - those pipes are being disabled.
+	 */
+	if (acrtc_state->active_planes)
+		amdgpu_dm_commit_cursors(state);
 
 cleanup:
-	kfree(flip);
-	kfree(full);
+	kfree(bundle);
+}
+
+/*
+ * Enable interrupts on CRTCs that are newly active, undergone
+ * a modeset, or have active planes again.
+ *
+ * Done in two passes, based on the for_modeset flag:
+ * Pass 1: For CRTCs going through modeset
+ * Pass 2: For CRTCs going from 0 to n active planes
+ *
+ * Interrupts can only be enabled after the planes are programmed,
+ * so this requires a two-pass approach since we don't want to
+ * just defer the interrupts until after commit planes every time.
+ */
+static void amdgpu_dm_enable_crtc_interrupts(struct drm_device *dev,
+					     struct drm_atomic_state *state,
+					     bool for_modeset)
+{
+	struct amdgpu_device *adev = dev->dev_private;
+	struct drm_crtc *crtc;
+	struct drm_crtc_state *old_crtc_state, *new_crtc_state;
+	int i;
+
+	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state,
+				      new_crtc_state, i) {
+		struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc);
+		struct dm_crtc_state *dm_new_crtc_state =
+			to_dm_crtc_state(new_crtc_state);
+		struct dm_crtc_state *dm_old_crtc_state =
+			to_dm_crtc_state(old_crtc_state);
+		bool modeset = drm_atomic_crtc_needs_modeset(new_crtc_state);
+		bool run_pass;
+
+		run_pass = (for_modeset && modeset) ||
+			   (!for_modeset && !modeset &&
+			    !dm_old_crtc_state->interrupts_enabled);
+
+		if (!run_pass)
+			continue;
+
+		if (!dm_new_crtc_state->interrupts_enabled)
+			continue;
+
+		manage_dm_interrupts(adev, acrtc, true);
+
+#ifdef CONFIG_DEBUG_FS
+		/* The stream has changed so CRC capture needs to re-enabled. */
+		if (dm_new_crtc_state->crc_enabled) {
+			dm_new_crtc_state->crc_enabled = false;
+			amdgpu_dm_crtc_set_crc_source(crtc, "auto");
+		}
+#endif
+	}
 }
 
 /*
@@ -4953,8 +5555,7 @@ cleanup:
 static void amdgpu_dm_crtc_copy_transient_flags(struct drm_crtc_state *crtc_state,
 						struct dc_stream_state *stream_state)
 {
-	stream_state->mode_changed =
-		crtc_state->mode_changed || crtc_state->active_changed;
+	stream_state->mode_changed = drm_atomic_crtc_needs_modeset(crtc_state);
 }
 
 static int amdgpu_dm_atomic_commit(struct drm_device *dev,
@@ -4967,30 +5568,41 @@ static int amdgpu_dm_atomic_commit(struct drm_device *dev,
 	int i;
 
 	/*
-	 * We evade vblanks and pflips on crtc that
-	 * should be changed. We do it here to flush & disable
-	 * interrupts before drm_swap_state is called in drm_atomic_helper_commit
-	 * it will update crtc->dm_crtc_state->stream pointer which is used in
-	 * the ISRs.
+	 * We evade vblank and pflip interrupts on CRTCs that are undergoing
+	 * a modeset, being disabled, or have no active planes.
+	 *
+	 * It's done in atomic commit rather than commit tail for now since
+	 * some of these interrupt handlers access the current CRTC state and
+	 * potentially the stream pointer itself.
+	 *
+	 * Since the atomic state is swapped within atomic commit and not within
+	 * commit tail this would leave to new state (that hasn't been committed yet)
+	 * being accesssed from within the handlers.
+	 *
+	 * TODO: Fix this so we can do this in commit tail and not have to block
+	 * in atomic check.
 	 */
 	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
 		struct dm_crtc_state *dm_old_crtc_state = to_dm_crtc_state(old_crtc_state);
 		struct dm_crtc_state *dm_new_crtc_state = to_dm_crtc_state(new_crtc_state);
 		struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc);
 
-		if (drm_atomic_crtc_needs_modeset(new_crtc_state)
-		    && dm_old_crtc_state->stream) {
+		if (dm_old_crtc_state->interrupts_enabled &&
+		    (!dm_new_crtc_state->interrupts_enabled ||
+		     drm_atomic_crtc_needs_modeset(new_crtc_state))) {
 			/*
-			 * If the stream is removed and CRC capture was
-			 * enabled on the CRTC the extra vblank reference
-			 * needs to be dropped since CRC capture will be
-			 * disabled.
+			 * Drop the extra vblank reference added by CRC
+			 * capture if applicable.
 			 */
-			if (!dm_new_crtc_state->stream
-			    && dm_new_crtc_state->crc_enabled) {
+			if (dm_new_crtc_state->crc_enabled)
 				drm_crtc_vblank_put(crtc);
+
+			/*
+			 * Only keep CRC capture enabled if there's
+			 * still a stream for the CRTC.
+			 */
+			if (!dm_new_crtc_state->stream)
 				dm_new_crtc_state->crc_enabled = false;
-			}
 
 			manage_dm_interrupts(adev, acrtc, false);
 		}
@@ -5037,7 +5649,7 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 		dc_state = dm_state->context;
 	} else {
 		/* No state changes, retain current state. */
-		dc_state_temp = dc_create_state();
+		dc_state_temp = dc_create_state(dm->dc);
 		ASSERT(dc_state_temp);
 		dc_state = dc_state_temp;
 		dc_resource_state_copy_construct_current(dm->dc, dc_state);
@@ -5206,45 +5818,41 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 		mutex_unlock(&dm->dc_lock);
 	}
 
+	/* Count number of newly disabled CRTCs for dropping PM refs later. */
 	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state,
-			new_crtc_state, i) {
-		/*
-		 * loop to enable interrupts on newly arrived crtc
-		 */
-		struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc);
-		bool modeset_needed;
-
+				      new_crtc_state, i) {
 		if (old_crtc_state->active && !new_crtc_state->active)
 			crtc_disable_count++;
 
 		dm_new_crtc_state = to_dm_crtc_state(new_crtc_state);
 		dm_old_crtc_state = to_dm_crtc_state(old_crtc_state);
-		modeset_needed = modeset_required(
-				new_crtc_state,
-				dm_new_crtc_state->stream,
-				dm_old_crtc_state->stream);
 
-		if (dm_new_crtc_state->stream == NULL || !modeset_needed)
-			continue;
-
-		manage_dm_interrupts(adev, acrtc, true);
+		/* Update freesync active state. */
+		pre_update_freesync_state_on_stream(dm, dm_new_crtc_state);
 
-#ifdef CONFIG_DEBUG_FS
-		/* The stream has changed so CRC capture needs to re-enabled. */
-		if (dm_new_crtc_state->crc_enabled)
-			amdgpu_dm_crtc_set_crc_source(crtc, "auto");
-#endif
+		/* Handle vrr on->off / off->on transitions */
+		amdgpu_dm_handle_vrr_transition(dm_old_crtc_state,
+						dm_new_crtc_state);
 	}
 
+	/* Enable interrupts for CRTCs going through a modeset. */
+	amdgpu_dm_enable_crtc_interrupts(dev, state, true);
+
+	for_each_new_crtc_in_state(state, crtc, new_crtc_state, j)
+		if (new_crtc_state->pageflip_flags & DRM_MODE_PAGE_FLIP_ASYNC)
+			wait_for_vblank = false;
+
 	/* update planes when needed per crtc*/
 	for_each_new_crtc_in_state(state, crtc, new_crtc_state, j) {
 		dm_new_crtc_state = to_dm_crtc_state(new_crtc_state);
 
 		if (dm_new_crtc_state->stream)
 			amdgpu_dm_commit_planes(state, dc_state, dev,
-						dm, crtc, &wait_for_vblank);
+						dm, crtc, wait_for_vblank);
 	}
 
+	/* Enable interrupts for CRTCs going from 0 to n active planes. */
+	amdgpu_dm_enable_crtc_interrupts(dev, state, false);
 
 	/*
 	 * send vblank event on all events not handled in flip and
@@ -5484,21 +6092,12 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm,
 	struct amdgpu_dm_connector *aconnector = NULL;
 	struct drm_connector_state *drm_new_conn_state = NULL, *drm_old_conn_state = NULL;
 	struct dm_connector_state *dm_new_conn_state = NULL, *dm_old_conn_state = NULL;
-	struct drm_plane_state *new_plane_state = NULL;
 
 	new_stream = NULL;
 
 	dm_old_crtc_state = to_dm_crtc_state(old_crtc_state);
 	dm_new_crtc_state = to_dm_crtc_state(new_crtc_state);
 	acrtc = to_amdgpu_crtc(crtc);
-
-	new_plane_state = drm_atomic_get_new_plane_state(state, new_crtc_state->crtc->primary);
-
-	if (new_crtc_state->enable && new_plane_state && !new_plane_state->fb) {
-		ret = -EINVAL;
-		goto fail;
-	}
-
 	aconnector = amdgpu_dm_find_first_crtc_matching_connector(state, crtc);
 
 	/* TODO This hack should go away */
@@ -5661,6 +6260,9 @@ skip_modeset:
 		update_stream_scaling_settings(
 			&new_crtc_state->mode, dm_new_conn_state, dm_new_crtc_state->stream);
 
+	/* ABM settings */
+	dm_new_crtc_state->abm_level = dm_new_conn_state->abm_level;
+
 	/*
 	 * Color management settings. We also update color properties
 	 * when a modeset is needed, to ensure it gets reprogrammed.
@@ -5685,6 +6287,69 @@ fail:
 	return ret;
 }
 
+static bool should_reset_plane(struct drm_atomic_state *state,
+			       struct drm_plane *plane,
+			       struct drm_plane_state *old_plane_state,
+			       struct drm_plane_state *new_plane_state)
+{
+	struct drm_plane *other;
+	struct drm_plane_state *old_other_state, *new_other_state;
+	struct drm_crtc_state *new_crtc_state;
+	int i;
+
+	/*
+	 * TODO: Remove this hack once the checks below are sufficient
+	 * enough to determine when we need to reset all the planes on
+	 * the stream.
+	 */
+	if (state->allow_modeset)
+		return true;
+
+	/* Exit early if we know that we're adding or removing the plane. */
+	if (old_plane_state->crtc != new_plane_state->crtc)
+		return true;
+
+	/* old crtc == new_crtc == NULL, plane not in context. */
+	if (!new_plane_state->crtc)
+		return false;
+
+	new_crtc_state =
+		drm_atomic_get_new_crtc_state(state, new_plane_state->crtc);
+
+	if (!new_crtc_state)
+		return true;
+
+	if (drm_atomic_crtc_needs_modeset(new_crtc_state))
+		return true;
+
+	/*
+	 * If there are any new primary or overlay planes being added or
+	 * removed then the z-order can potentially change. To ensure
+	 * correct z-order and pipe acquisition the current DC architecture
+	 * requires us to remove and recreate all existing planes.
+	 *
+	 * TODO: Come up with a more elegant solution for this.
+	 */
+	for_each_oldnew_plane_in_state(state, other, old_other_state, new_other_state, i) {
+		if (other->type == DRM_PLANE_TYPE_CURSOR)
+			continue;
+
+		if (old_other_state->crtc != new_plane_state->crtc &&
+		    new_other_state->crtc != new_plane_state->crtc)
+			continue;
+
+		if (old_other_state->crtc != new_other_state->crtc)
+			return true;
+
+		/* TODO: Remove this once we can handle fast format changes. */
+		if (old_other_state->fb && new_other_state->fb &&
+		    old_other_state->fb->format != new_other_state->fb->format)
+			return true;
+	}
+
+	return false;
+}
+
 static int dm_update_plane_state(struct dc *dc,
 				 struct drm_atomic_state *state,
 				 struct drm_plane *plane,
@@ -5699,8 +6364,7 @@ static int dm_update_plane_state(struct dc *dc,
 	struct drm_crtc_state *old_crtc_state, *new_crtc_state;
 	struct dm_crtc_state *dm_new_crtc_state, *dm_old_crtc_state;
 	struct dm_plane_state *dm_new_plane_state, *dm_old_plane_state;
-	/* TODO return page_flip_needed() function */
-	bool pflip_needed  = !state->allow_modeset;
+	bool needs_reset;
 	int ret = 0;
 
 
@@ -5713,10 +6377,12 @@ static int dm_update_plane_state(struct dc *dc,
 	if (plane->type == DRM_PLANE_TYPE_CURSOR)
 		return 0;
 
+	needs_reset = should_reset_plane(state, plane, old_plane_state,
+					 new_plane_state);
+
 	/* Remove any changed/removed planes */
 	if (!enable) {
-		if (pflip_needed &&
-		    plane->type != DRM_PLANE_TYPE_OVERLAY)
+		if (!needs_reset)
 			return 0;
 
 		if (!old_plane_crtc)
@@ -5767,7 +6433,7 @@ static int dm_update_plane_state(struct dc *dc,
 		if (!dm_new_crtc_state->stream)
 			return 0;
 
-		if (pflip_needed && plane->type != DRM_PLANE_TYPE_OVERLAY)
+		if (!needs_reset)
 			return 0;
 
 		WARN_ON(dm_new_plane_state->dc_state);
@@ -5779,7 +6445,7 @@ static int dm_update_plane_state(struct dc *dc,
 		DRM_DEBUG_DRIVER("Enabling DRM plane: %d on DRM crtc %d\n",
 				plane->base.id, new_plane_crtc->base.id);
 
-		ret = fill_plane_attributes(
+		ret = fill_dc_plane_attributes(
 			new_plane_crtc->dev->dev_private,
 			dc_new_plane_state,
 			new_plane_state,
@@ -5827,10 +6493,11 @@ static int dm_update_plane_state(struct dc *dc,
 }
 
 static int
-dm_determine_update_type_for_commit(struct dc *dc,
+dm_determine_update_type_for_commit(struct amdgpu_display_manager *dm,
 				    struct drm_atomic_state *state,
 				    enum surface_update_type *out_type)
 {
+	struct dc *dc = dm->dc;
 	struct dm_atomic_state *dm_state = NULL, *old_dm_state = NULL;
 	int i, j, num_plane, ret = 0;
 	struct drm_plane_state *old_plane_state, *new_plane_state;
@@ -5844,21 +6511,22 @@ dm_determine_update_type_for_commit(struct dc *dc,
 	struct dc_stream_status *status = NULL;
 
 	struct dc_surface_update *updates;
-	struct dc_plane_state *surface;
 	enum surface_update_type update_type = UPDATE_TYPE_FAST;
 
 	updates = kcalloc(MAX_SURFACES, sizeof(*updates), GFP_KERNEL);
-	surface = kcalloc(MAX_SURFACES, sizeof(*surface), GFP_KERNEL);
 
-	if (!updates || !surface) {
-		DRM_ERROR("Plane or surface update failed to allocate");
+	if (!updates) {
+		DRM_ERROR("Failed to allocate plane updates\n");
 		/* Set type to FULL to avoid crashing in DC*/
 		update_type = UPDATE_TYPE_FULL;
 		goto cleanup;
 	}
 
 	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
-		struct dc_stream_update stream_update = { 0 };
+		struct dc_scaling_info scaling_info;
+		struct dc_stream_update stream_update;
+
+		memset(&stream_update, 0, sizeof(stream_update));
 
 		new_dm_crtc_state = to_dm_crtc_state(new_crtc_state);
 		old_dm_crtc_state = to_dm_crtc_state(old_crtc_state);
@@ -5886,23 +6554,12 @@ dm_determine_update_type_for_commit(struct dc *dc,
 				goto cleanup;
 			}
 
-			if (!state->allow_modeset)
-				continue;
-
 			if (crtc != new_plane_crtc)
 				continue;
 
-			updates[num_plane].surface = &surface[num_plane];
+			updates[num_plane].surface = new_dm_plane_state->dc_state;
 
 			if (new_crtc_state->mode_changed) {
-				updates[num_plane].surface->src_rect =
-						new_dm_plane_state->dc_state->src_rect;
-				updates[num_plane].surface->dst_rect =
-						new_dm_plane_state->dc_state->dst_rect;
-				updates[num_plane].surface->rotation =
-						new_dm_plane_state->dc_state->rotation;
-				updates[num_plane].surface->in_transfer_func =
-						new_dm_plane_state->dc_state->in_transfer_func;
 				stream_update.dst = new_dm_crtc_state->stream->dst;
 				stream_update.src = new_dm_crtc_state->stream->src;
 			}
@@ -5918,6 +6575,13 @@ dm_determine_update_type_for_commit(struct dc *dc,
 						new_dm_crtc_state->stream->out_transfer_func;
 			}
 
+			ret = fill_dc_scaling_info(new_plane_state,
+						   &scaling_info);
+			if (ret)
+				goto cleanup;
+
+			updates[num_plane].scaling_info = &scaling_info;
+
 			num_plane++;
 		}
 
@@ -5937,8 +6601,14 @@ dm_determine_update_type_for_commit(struct dc *dc,
 		status = dc_stream_get_status_from_state(old_dm_state->context,
 							 new_dm_crtc_state->stream);
 
+		/*
+		 * TODO: DC modifies the surface during this call so we need
+		 * to lock here - find a way to do this without locking.
+		 */
+		mutex_lock(&dm->dc_lock);
 		update_type = dc_check_update_surfaces_for_stream(dc, updates, num_plane,
 								  &stream_update, status);
+		mutex_unlock(&dm->dc_lock);
 
 		if (update_type > UPDATE_TYPE_MED) {
 			update_type = UPDATE_TYPE_FULL;
@@ -5948,7 +6618,6 @@ dm_determine_update_type_for_commit(struct dc *dc,
 
 cleanup:
 	kfree(updates);
-	kfree(surface);
 
 	*out_type = update_type;
 	return ret;
@@ -6132,7 +6801,7 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev,
 		lock_and_validation_needed = true;
 	}
 
-	ret = dm_determine_update_type_for_commit(dc, state, &update_type);
+	ret = dm_determine_update_type_for_commit(&adev->dm, state, &update_type);
 	if (ret)
 		goto fail;
 
@@ -6147,9 +6816,6 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev,
 	 */
 	if (lock_and_validation_needed && overall_update_type <= UPDATE_TYPE_FAST)
 		WARN(1, "Global lock should be Set, overall_update_type should be UPDATE_TYPE_MED or UPDATE_TYPE_FULL");
-	else if (!lock_and_validation_needed && overall_update_type > UPDATE_TYPE_FAST)
-		WARN(1, "Global lock should NOT be set, overall_update_type should be UPDATE_TYPE_FAST");
-
 
 	if (overall_update_type > UPDATE_TYPE_FAST) {
 		ret = dm_atomic_get_state(state, &dm_state);
@@ -6160,7 +6826,7 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev,
 		if (ret)
 			goto fail;
 
-		if (dc_validate_global_state(dc, dm_state->context) != DC_OK) {
+		if (dc_validate_global_state(dc, dm_state->context, false) != DC_OK) {
 			ret = -EINVAL;
 			goto fail;
 		}
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
index fbd161ddc3f4..978ff14a7d45 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
@@ -132,8 +132,6 @@ struct amdgpu_display_manager {
 	 */
 	struct drm_private_obj atomic_obj;
 
-	struct drm_modeset_lock atomic_obj_lock;
-
 	/**
 	 * @dc_lock:
 	 *
@@ -184,6 +182,15 @@ struct amdgpu_display_manager {
 	struct common_irq_params
 	vblank_params[DC_IRQ_SOURCE_VBLANK6 - DC_IRQ_SOURCE_VBLANK1 + 1];
 
+	/**
+	 * @vupdate_params:
+	 *
+	 * Vertical update IRQ parameters, passed to registered handlers when
+	 * triggered.
+	 */
+	struct common_irq_params
+	vupdate_params[DC_IRQ_SOURCE_VUPDATE6 - DC_IRQ_SOURCE_VUPDATE1 + 1];
+
 	spinlock_t irq_handler_list_table_lock;
 
 	struct backlight_device *backlight_dev;
@@ -240,6 +247,10 @@ struct amdgpu_dm_connector {
 	struct mutex hpd_lock;
 
 	bool fake_enable;
+#ifdef CONFIG_DEBUG_FS
+	uint32_t debugfs_dpcd_address;
+	uint32_t debugfs_dpcd_size;
+#endif
 };
 
 #define to_amdgpu_dm_connector(x) container_of(x, struct amdgpu_dm_connector, base)
@@ -260,6 +271,9 @@ struct dm_crtc_state {
 	struct drm_crtc_state base;
 	struct dc_stream_state *stream;
 
+	int active_planes;
+	bool interrupts_enabled;
+
 	int crc_skip_count;
 	bool crc_enabled;
 
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
index 216e48cec716..7258c992a2bf 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
@@ -126,46 +126,51 @@ int amdgpu_dm_set_regamma_lut(struct dm_crtc_state *crtc)
 		crtc->base.state->dev->dev_private;
 	struct drm_color_lut *lut;
 	uint32_t lut_size;
-	struct dc_gamma *gamma;
+	struct dc_gamma *gamma = NULL;
 	enum dc_transfer_func_type old_type = stream->out_transfer_func->type;
 
 	bool ret;
 
-	if (!blob) {
+	if (!blob && adev->asic_type <= CHIP_RAVEN) {
 		/* By default, use the SRGB predefined curve.*/
 		stream->out_transfer_func->type = TF_TYPE_PREDEFINED;
 		stream->out_transfer_func->tf = TRANSFER_FUNCTION_SRGB;
 		return 0;
 	}
 
-	lut = (struct drm_color_lut *)blob->data;
-	lut_size = blob->length / sizeof(struct drm_color_lut);
-
-	gamma = dc_create_gamma();
-	if (!gamma)
-		return -ENOMEM;
+	if (blob) {
+		lut = (struct drm_color_lut *)blob->data;
+		lut_size = blob->length / sizeof(struct drm_color_lut);
+
+		gamma = dc_create_gamma();
+		if (!gamma)
+			return -ENOMEM;
+
+		gamma->num_entries = lut_size;
+		if (gamma->num_entries == MAX_COLOR_LEGACY_LUT_ENTRIES)
+			gamma->type = GAMMA_RGB_256;
+		else if (gamma->num_entries == MAX_COLOR_LUT_ENTRIES)
+			gamma->type = GAMMA_CS_TFM_1D;
+		else {
+			/* Invalid lut size */
+			dc_gamma_release(&gamma);
+			return -EINVAL;
+		}
 
-	gamma->num_entries = lut_size;
-	if (gamma->num_entries == MAX_COLOR_LEGACY_LUT_ENTRIES)
-		gamma->type = GAMMA_RGB_256;
-	else if (gamma->num_entries == MAX_COLOR_LUT_ENTRIES)
-		gamma->type = GAMMA_CS_TFM_1D;
-	else {
-		/* Invalid lut size */
-		dc_gamma_release(&gamma);
-		return -EINVAL;
+		/* Convert drm_lut into dc_gamma */
+		__drm_lut_to_dc_gamma(lut, gamma, gamma->type == GAMMA_RGB_256);
 	}
 
-	/* Convert drm_lut into dc_gamma */
-	__drm_lut_to_dc_gamma(lut, gamma, gamma->type == GAMMA_RGB_256);
-
-	/* Call color module to translate into something DC understands. Namely
-	 * a transfer function.
+	/* predefined gamma ROM only exist for RAVEN and pre-RAVEN ASIC,
+	 * set canRomBeUsed accordingly
 	 */
 	stream->out_transfer_func->type = TF_TYPE_DISTRIBUTED_POINTS;
 	ret = mod_color_calculate_regamma_params(stream->out_transfer_func,
-						 gamma, true, adev->asic_type <= CHIP_RAVEN, NULL);
-	dc_gamma_release(&gamma);
+			gamma, true, adev->asic_type <= CHIP_RAVEN, NULL);
+
+	if (gamma)
+		dc_gamma_release(&gamma);
+
 	if (!ret) {
 		stream->out_transfer_func->type = old_type;
 		DRM_ERROR("Out of memory when calculating regamma params\n");
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
index 4a55cde027cf..1d5fc5ad3bee 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
@@ -29,6 +29,7 @@
 #include "amdgpu.h"
 #include "amdgpu_dm.h"
 #include "amdgpu_dm_debugfs.h"
+#include "dm_helpers.h"
 
 /* function description
  * get/ set DP configuration: lane_count, link_rate, spread_spectrum
@@ -688,8 +689,131 @@ static int vrr_range_show(struct seq_file *m, void *data)
 
 	return 0;
 }
+
+/* function description
+ *
+ * generic SDP message access for testing
+ *
+ * debugfs sdp_message is located at /syskernel/debug/dri/0/DP-x
+ *
+ * SDP header
+ * Hb0 : Secondary-Data Packet ID
+ * Hb1 : Secondary-Data Packet type
+ * Hb2 : Secondary-Data-packet-specific header, Byte 0
+ * Hb3 : Secondary-Data-packet-specific header, Byte 1
+ *
+ * for using custom sdp message: input 4 bytes SDP header and 32 bytes raw data
+ */
+static ssize_t dp_sdp_message_debugfs_write(struct file *f, const char __user *buf,
+				 size_t size, loff_t *pos)
+{
+	int r;
+	uint8_t data[36];
+	struct amdgpu_dm_connector *connector = file_inode(f)->i_private;
+	struct dm_crtc_state *acrtc_state;
+	uint32_t write_size = 36;
+
+	if (connector->base.status != connector_status_connected)
+		return -ENODEV;
+
+	if (size == 0)
+		return 0;
+
+	acrtc_state = to_dm_crtc_state(connector->base.state->crtc->state);
+
+	r = copy_from_user(data, buf, write_size);
+
+	write_size -= r;
+
+	dc_stream_send_dp_sdp(acrtc_state->stream, data, write_size);
+
+	return write_size;
+}
+
 DEFINE_SHOW_ATTRIBUTE(vrr_range);
 
+static ssize_t dp_dpcd_address_write(struct file *f, const char __user *buf,
+				 size_t size, loff_t *pos)
+{
+	int r;
+	struct amdgpu_dm_connector *connector = file_inode(f)->i_private;
+
+	if (size < sizeof(connector->debugfs_dpcd_address))
+		return 0;
+
+	r = copy_from_user(&connector->debugfs_dpcd_address,
+			buf, sizeof(connector->debugfs_dpcd_address));
+
+	return size - r;
+}
+
+static ssize_t dp_dpcd_size_write(struct file *f, const char __user *buf,
+				 size_t size, loff_t *pos)
+{
+	int r;
+	struct amdgpu_dm_connector *connector = file_inode(f)->i_private;
+
+	if (size < sizeof(connector->debugfs_dpcd_size))
+		return 0;
+
+	r = copy_from_user(&connector->debugfs_dpcd_size,
+			buf, sizeof(connector->debugfs_dpcd_size));
+
+	if (connector->debugfs_dpcd_size > 256)
+		connector->debugfs_dpcd_size = 0;
+
+	return size - r;
+}
+
+static ssize_t dp_dpcd_data_write(struct file *f, const char __user *buf,
+				 size_t size, loff_t *pos)
+{
+	int r;
+	char *data;
+	struct amdgpu_dm_connector *connector = file_inode(f)->i_private;
+	struct dc_link *link = connector->dc_link;
+	uint32_t write_size = connector->debugfs_dpcd_size;
+
+	if (size < write_size)
+		return 0;
+
+	data = kzalloc(write_size, GFP_KERNEL);
+	if (!data)
+		return 0;
+
+	r = copy_from_user(data, buf, write_size);
+
+	dm_helpers_dp_write_dpcd(link->ctx, link,
+			connector->debugfs_dpcd_address, data, write_size - r);
+	kfree(data);
+	return write_size - r;
+}
+
+static ssize_t dp_dpcd_data_read(struct file *f, char __user *buf,
+				 size_t size, loff_t *pos)
+{
+	int r;
+	char *data;
+	struct amdgpu_dm_connector *connector = file_inode(f)->i_private;
+	struct dc_link *link = connector->dc_link;
+	uint32_t read_size = connector->debugfs_dpcd_size;
+
+	if (size < read_size)
+		return 0;
+
+	data = kzalloc(read_size, GFP_KERNEL);
+	if (!data)
+		return 0;
+
+	dm_helpers_dp_read_dpcd(link->ctx, link,
+			connector->debugfs_dpcd_address, data, read_size);
+
+	r = copy_to_user(buf, data, read_size);
+
+	kfree(data);
+	return read_size - r;
+}
+
 static const struct file_operations dp_link_settings_debugfs_fops = {
 	.owner = THIS_MODULE,
 	.read = dp_link_settings_read,
@@ -710,6 +834,31 @@ static const struct file_operations dp_phy_test_pattern_fops = {
 	.llseek = default_llseek
 };
 
+static const struct file_operations sdp_message_fops = {
+	.owner = THIS_MODULE,
+	.write = dp_sdp_message_debugfs_write,
+	.llseek = default_llseek
+};
+
+static const struct file_operations dp_dpcd_address_debugfs_fops = {
+	.owner = THIS_MODULE,
+	.write = dp_dpcd_address_write,
+	.llseek = default_llseek
+};
+
+static const struct file_operations dp_dpcd_size_debugfs_fops = {
+	.owner = THIS_MODULE,
+	.write = dp_dpcd_size_write,
+	.llseek = default_llseek
+};
+
+static const struct file_operations dp_dpcd_data_debugfs_fops = {
+	.owner = THIS_MODULE,
+	.read = dp_dpcd_data_read,
+	.write = dp_dpcd_data_write,
+	.llseek = default_llseek
+};
+
 static const struct {
 	char *name;
 	const struct file_operations *fops;
@@ -717,7 +866,11 @@ static const struct {
 		{"link_settings", &dp_link_settings_debugfs_fops},
 		{"phy_settings", &dp_phy_settings_debugfs_fop},
 		{"test_pattern", &dp_phy_test_pattern_fops},
-		{"vrr_range", &vrr_range_fops}
+		{"vrr_range", &vrr_range_fops},
+		{"sdp_message", &sdp_message_fops},
+		{"aux_dpcd_address", &dp_dpcd_address_debugfs_fops},
+		{"aux_dpcd_size", &dp_dpcd_size_debugfs_fops},
+		{"aux_dpcd_data", &dp_dpcd_data_debugfs_fops}
 };
 
 int connector_debugfs_init(struct amdgpu_dm_connector *connector)
@@ -842,6 +995,35 @@ static const struct drm_info_list amdgpu_dm_debugfs_list[] = {
 	{"amdgpu_target_backlight_pwm", &target_backlight_read},
 };
 
+/*
+ * Sets the DC visual confirm debug option from the given string.
+ * Example usage: echo 1 > /sys/kernel/debug/dri/0/amdgpu_visual_confirm
+ */
+static int visual_confirm_set(void *data, u64 val)
+{
+	struct amdgpu_device *adev = data;
+
+	adev->dm.dc->debug.visual_confirm = (enum visual_confirm)val;
+
+	return 0;
+}
+
+/*
+ * Reads the DC visual confirm debug option value into the given buffer.
+ * Example usage: cat /sys/kernel/debug/dri/0/amdgpu_dm_visual_confirm
+ */
+static int visual_confirm_get(void *data, u64 *val)
+{
+	struct amdgpu_device *adev = data;
+
+	*val = adev->dm.dc->debug.visual_confirm;
+
+	return 0;
+}
+
+DEFINE_DEBUGFS_ATTRIBUTE(visual_confirm_fops, visual_confirm_get,
+			 visual_confirm_set, "%llu\n");
+
 int dtn_debugfs_init(struct amdgpu_device *adev)
 {
 	static const struct file_operations dtn_log_fops = {
@@ -867,5 +1049,13 @@ int dtn_debugfs_init(struct amdgpu_device *adev)
 		adev,
 		&dtn_log_fops);
 
-	return PTR_ERR_OR_ZERO(ent);
+	if (IS_ERR(ent))
+		return PTR_ERR(ent);
+
+	ent = debugfs_create_file_unsafe("amdgpu_dm_visual_confirm", 0644, root,
+					 adev, &visual_confirm_fops);
+	if (IS_ERR(ent))
+		return PTR_ERR(ent);
+
+	return 0;
 }
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
index b39766bd2840..e6cd67342df8 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
@@ -264,7 +264,7 @@ bool dm_helpers_dp_mst_write_payload_allocation_table(
 }
 
 /*
- * poll pending down reply before clear payload allocation table
+ * poll pending down reply
  */
 void dm_helpers_dp_mst_poll_pending_down_reply(
 	struct dc_context *ctx,
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c
index cd10f77cdeb0..fd22b4474dbf 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c
@@ -674,11 +674,30 @@ static int amdgpu_dm_set_crtc_irq_state(struct amdgpu_device *adev,
 		__func__);
 }
 
+static int amdgpu_dm_set_vupdate_irq_state(struct amdgpu_device *adev,
+					   struct amdgpu_irq_src *source,
+					   unsigned int crtc_id,
+					   enum amdgpu_interrupt_state state)
+{
+	return dm_irq_state(
+		adev,
+		source,
+		crtc_id,
+		state,
+		IRQ_TYPE_VUPDATE,
+		__func__);
+}
+
 static const struct amdgpu_irq_src_funcs dm_crtc_irq_funcs = {
 	.set = amdgpu_dm_set_crtc_irq_state,
 	.process = amdgpu_dm_irq_handler,
 };
 
+static const struct amdgpu_irq_src_funcs dm_vupdate_irq_funcs = {
+	.set = amdgpu_dm_set_vupdate_irq_state,
+	.process = amdgpu_dm_irq_handler,
+};
+
 static const struct amdgpu_irq_src_funcs dm_pageflip_irq_funcs = {
 	.set = amdgpu_dm_set_pflip_irq_state,
 	.process = amdgpu_dm_irq_handler,
@@ -695,6 +714,9 @@ void amdgpu_dm_set_irq_funcs(struct amdgpu_device *adev)
 	adev->crtc_irq.num_types = adev->mode_info.num_crtc;
 	adev->crtc_irq.funcs = &dm_crtc_irq_funcs;
 
+	adev->vupdate_irq.num_types = adev->mode_info.num_crtc;
+	adev->vupdate_irq.funcs = &dm_vupdate_irq_funcs;
+
 	adev->pageflip_irq.num_types = adev->mode_info.num_crtc;
 	adev->pageflip_irq.funcs = &dm_pageflip_irq_funcs;
 
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
index c4ea3a91f17a..6e205ee36ac3 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
@@ -84,6 +84,7 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux,
 {
 	ssize_t result = 0;
 	struct aux_payload payload;
+	enum aux_channel_operation_result operation_result;
 
 	if (WARN_ON(msg->size > 16))
 		return -E2BIG;
@@ -97,13 +98,27 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux,
 	payload.mot = (msg->request & DP_AUX_I2C_MOT) != 0;
 	payload.defer_delay = 0;
 
-	result = dc_link_aux_transfer(TO_DM_AUX(aux)->ddc_service, &payload);
+	result = dc_link_aux_transfer_raw(TO_DM_AUX(aux)->ddc_service, &payload,
+				      &operation_result);
 
 	if (payload.write)
 		result = msg->size;
 
-	if (result < 0) /* DC doesn't know about kernel error codes */
-		result = -EIO;
+	if (result < 0)
+		switch (operation_result) {
+		case AUX_CHANNEL_OPERATION_SUCCEEDED:
+			break;
+		case AUX_CHANNEL_OPERATION_FAILED_HPD_DISCON:
+		case AUX_CHANNEL_OPERATION_FAILED_REASON_UNKNOWN:
+			result = -EIO;
+			break;
+		case AUX_CHANNEL_OPERATION_FAILED_INVALID_REPLY:
+			result = -EBUSY;
+			break;
+		case AUX_CHANNEL_OPERATION_FAILED_TIMEOUT:
+			result = -ETIMEDOUT;
+			break;
+		}
 
 	return result;
 }
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c
index a114954d6a5b..350e7a620d45 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c
@@ -33,6 +33,7 @@
 #include "amdgpu_dm_irq.h"
 #include "amdgpu_pm.h"
 #include "dm_pp_smu.h"
+#include "amdgpu_smu.h"
 
 
 bool dm_pp_apply_display_requirements(
@@ -40,6 +41,7 @@ bool dm_pp_apply_display_requirements(
 		const struct dm_pp_display_configuration *pp_display_cfg)
 {
 	struct amdgpu_device *adev = ctx->driver_context;
+	struct smu_context *smu = &adev->smu;
 	int i;
 
 	if (adev->pm.dpm_enabled) {
@@ -105,6 +107,9 @@ bool dm_pp_apply_display_requirements(
 			adev->powerplay.pp_funcs->display_configuration_change(
 				adev->powerplay.pp_handle,
 				&adev->pm.pm_display_cfg);
+		else
+			smu_display_configuration_change(smu,
+							 &adev->pm.pm_display_cfg);
 
 		amdgpu_pm_compute_clocks(adev);
 	}
@@ -308,6 +313,12 @@ bool dm_pp_get_clock_levels_by_type(
 		if (adev->powerplay.pp_funcs->get_clock_by_type(pp_handle,
 			dc_to_pp_clock_type(clk_type), &pp_clks)) {
 		/* Error in pplib. Provide default values. */
+			return true;
+		}
+	} else if (adev->smu.funcs && adev->smu.funcs->get_clock_by_type) {
+		if (smu_get_clock_by_type(&adev->smu,
+					  dc_to_pp_clock_type(clk_type),
+					  &pp_clks)) {
 			get_default_clock_levels(clk_type, dc_clks);
 			return true;
 		}
@@ -324,6 +335,13 @@ bool dm_pp_get_clock_levels_by_type(
 			validation_clks.memory_max_clock = 80000;
 			validation_clks.level = 0;
 		}
+	} else if (adev->smu.funcs && adev->smu.funcs->get_max_high_clocks) {
+		if (smu_get_max_high_clocks(&adev->smu, &validation_clks)) {
+			DRM_INFO("DM_PPLIB: Warning: using default validation clocks!\n");
+			validation_clks.engine_max_clock = 72000;
+			validation_clks.memory_max_clock = 80000;
+			validation_clks.level = 0;
+		}
 	}
 
 	DRM_INFO("DM_PPLIB: Validation clocks:\n");
@@ -374,14 +392,21 @@ bool dm_pp_get_clock_levels_by_type_with_latency(
 	void *pp_handle = adev->powerplay.pp_handle;
 	struct pp_clock_levels_with_latency pp_clks = { 0 };
 	const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
+	int ret;
+
+	if (pp_funcs && pp_funcs->get_clock_by_type_with_latency) {
+		ret = pp_funcs->get_clock_by_type_with_latency(pp_handle,
+						dc_to_pp_clock_type(clk_type),
+						&pp_clks);
+		if (ret)
+			return false;
+	} else if (adev->smu.ppt_funcs && adev->smu.ppt_funcs->get_clock_by_type_with_latency) {
+		if (smu_get_clock_by_type_with_latency(&adev->smu,
+						       dc_to_pp_clock_type(clk_type),
+						       &pp_clks))
+			return false;
+	}
 
-	if (!pp_funcs || !pp_funcs->get_clock_by_type_with_latency)
-		return false;
-
-	if (pp_funcs->get_clock_by_type_with_latency(pp_handle,
-						     dc_to_pp_clock_type(clk_type),
-						     &pp_clks))
-		return false;
 
 	pp_to_dc_clock_levels_with_latency(&pp_clks, clk_level_info, clk_type);
 
@@ -397,14 +422,20 @@ bool dm_pp_get_clock_levels_by_type_with_voltage(
 	void *pp_handle = adev->powerplay.pp_handle;
 	struct pp_clock_levels_with_voltage pp_clk_info = {0};
 	const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
-
-	if (!pp_funcs || !pp_funcs->get_clock_by_type_with_voltage)
-		return false;
-
-	if (pp_funcs->get_clock_by_type_with_voltage(pp_handle,
-						     dc_to_pp_clock_type(clk_type),
-						     &pp_clk_info))
-		return false;
+	int ret;
+
+	if (pp_funcs && pp_funcs->get_clock_by_type_with_voltage) {
+		ret = pp_funcs->get_clock_by_type_with_voltage(pp_handle,
+						dc_to_pp_clock_type(clk_type),
+						&pp_clk_info);
+		if (ret)
+			return false;
+	} else if (adev->smu.ppt_funcs && adev->smu.ppt_funcs->get_clock_by_type_with_voltage) {
+		if (smu_get_clock_by_type_with_voltage(&adev->smu,
+						       dc_to_pp_clock_type(clk_type),
+						       &pp_clk_info))
+			return false;
+	}
 
 	pp_to_dc_clock_levels_with_voltage(&pp_clk_info, clk_level_info, clk_type);
 
@@ -445,6 +476,10 @@ bool dm_pp_apply_clock_for_voltage_request(
 		ret = adev->powerplay.pp_funcs->display_clock_voltage_request(
 			adev->powerplay.pp_handle,
 			&pp_clock_request);
+	else if (adev->smu.funcs &&
+		 adev->smu.funcs->display_clock_voltage_request)
+		ret = smu_display_clock_voltage_request(&adev->smu,
+							&pp_clock_request);
 	if (ret)
 		return false;
 	return true;
@@ -462,6 +497,8 @@ bool dm_pp_get_static_clocks(
 		ret = adev->powerplay.pp_funcs->get_current_clocks(
 			adev->powerplay.pp_handle,
 			&pp_clk_info);
+	else if (adev->smu.funcs)
+		ret = smu_get_current_clocks(&adev->smu, &pp_clk_info);
 	if (ret)
 		return false;
 
@@ -472,27 +509,6 @@ bool dm_pp_get_static_clocks(
 	return true;
 }
 
-void pp_rv_set_display_requirement(struct pp_smu *pp,
-		struct pp_smu_display_requirement_rv *req)
-{
-	const struct dc_context *ctx = pp->dm;
-	struct amdgpu_device *adev = ctx->driver_context;
-	void *pp_handle = adev->powerplay.pp_handle;
-	const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
-	struct pp_display_clock_request clock = {0};
-
-	if (!pp_funcs || !pp_funcs->display_clock_voltage_request)
-		return;
-
-	clock.clock_type = amd_pp_dcf_clock;
-	clock.clock_freq_in_khz = req->hard_min_dcefclk_mhz * 1000;
-	pp_funcs->display_clock_voltage_request(pp_handle, &clock);
-
-	clock.clock_type = amd_pp_f_clock;
-	clock.clock_freq_in_khz = req->hard_min_fclk_mhz * 1000;
-	pp_funcs->display_clock_voltage_request(pp_handle, &clock);
-}
-
 void pp_rv_set_wm_ranges(struct pp_smu *pp,
 		struct pp_smu_wm_range_sets *ranges)
 {
@@ -508,9 +524,6 @@ void pp_rv_set_wm_ranges(struct pp_smu *pp,
 	wm_with_clock_ranges.num_wm_dmif_sets = ranges->num_reader_wm_sets;
 	wm_with_clock_ranges.num_wm_mcif_sets = ranges->num_writer_wm_sets;
 
-	if (!pp_funcs || !pp_funcs->set_watermarks_for_clocks_ranges)
-		return;
-
 	for (i = 0; i < wm_with_clock_ranges.num_wm_dmif_sets; i++) {
 		if (ranges->reader_wm_sets[i].wm_inst > 3)
 			wm_dce_clocks[i].wm_set_id = WM_SET_A;
@@ -543,7 +556,13 @@ void pp_rv_set_wm_ranges(struct pp_smu *pp,
 				ranges->writer_wm_sets[i].min_drain_clk_mhz * 1000;
 	}
 
-	pp_funcs->set_watermarks_for_clocks_ranges(pp_handle, &wm_with_clock_ranges);
+	if (pp_funcs && pp_funcs->set_watermarks_for_clocks_ranges)
+		pp_funcs->set_watermarks_for_clocks_ranges(pp_handle,
+							   &wm_with_clock_ranges);
+	else if (adev->smu.funcs &&
+		 adev->smu.funcs->set_watermarks_for_clock_ranges)
+		smu_set_watermarks_for_clock_ranges(&adev->smu,
+						    &wm_with_clock_ranges);
 }
 
 void pp_rv_set_pme_wa_enable(struct pp_smu *pp)
@@ -553,10 +572,10 @@ void pp_rv_set_pme_wa_enable(struct pp_smu *pp)
 	void *pp_handle = adev->powerplay.pp_handle;
 	const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
 
-	if (!pp_funcs || !pp_funcs->notify_smu_enable_pwe)
-		return;
-
-	pp_funcs->notify_smu_enable_pwe(pp_handle);
+	if (pp_funcs && pp_funcs->notify_smu_enable_pwe)
+		pp_funcs->notify_smu_enable_pwe(pp_handle);
+	else if (adev->smu.funcs)
+		smu_notify_smu_enable_pwe(&adev->smu);
 }
 
 void pp_rv_set_active_display_count(struct pp_smu *pp, int count)
@@ -611,17 +630,16 @@ void pp_rv_set_hard_min_fclk_by_freq(struct pp_smu *pp, int mhz)
 	pp_funcs->set_hard_min_fclk_by_freq(pp_handle, mhz);
 }
 
-void dm_pp_get_funcs_rv(
+void dm_pp_get_funcs(
 		struct dc_context *ctx,
-		struct pp_smu_funcs_rv *funcs)
+		struct pp_smu_funcs *funcs)
 {
-	funcs->pp_smu.dm = ctx;
-	funcs->set_display_requirement = pp_rv_set_display_requirement;
-	funcs->set_wm_ranges = pp_rv_set_wm_ranges;
-	funcs->set_pme_wa_enable = pp_rv_set_pme_wa_enable;
-	funcs->set_display_count = pp_rv_set_active_display_count;
-	funcs->set_min_deep_sleep_dcfclk = pp_rv_set_min_deep_sleep_dcfclk;
-	funcs->set_hard_min_dcfclk_by_freq = pp_rv_set_hard_min_dcefclk_by_freq;
-	funcs->set_hard_min_fclk_by_freq = pp_rv_set_hard_min_fclk_by_freq;
+	funcs->rv_funcs.pp_smu.dm = ctx;
+	funcs->rv_funcs.set_wm_ranges = pp_rv_set_wm_ranges;
+	funcs->rv_funcs.set_pme_wa_enable = pp_rv_set_pme_wa_enable;
+	funcs->rv_funcs.set_display_count = pp_rv_set_active_display_count;
+	funcs->rv_funcs.set_min_deep_sleep_dcfclk = pp_rv_set_min_deep_sleep_dcfclk;
+	funcs->rv_funcs.set_hard_min_dcfclk_by_freq = pp_rv_set_hard_min_dcefclk_by_freq;
+	funcs->rv_funcs.set_hard_min_fclk_by_freq = pp_rv_set_hard_min_fclk_by_freq;
 }
 
diff --git a/drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c b/drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c
index f28989860fd8..1e9a2d352068 100644
--- a/drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c
+++ b/drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c
@@ -449,6 +449,11 @@ static inline unsigned int clamp_ux_dy(
 		return min_clamp;
 }
 
+unsigned int dc_fixpt_u4d19(struct fixed31_32 arg)
+{
+	return ux_dy(arg.value, 4, 19);
+}
+
 unsigned int dc_fixpt_u3d19(struct fixed31_32 arg)
 {
 	return ux_dy(arg.value, 3, 19);
diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c
index eb62d10bb65c..1b4b51657f5e 100644
--- a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c
+++ b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c
@@ -247,6 +247,53 @@ static enum dcn_bw_defs tl_pixel_format_to_bw_defs(enum surface_pixel_format for
 	}
 }
 
+enum source_macro_tile_size swizzle_mode_to_macro_tile_size(enum swizzle_mode_values sw_mode)
+{
+	switch (sw_mode) {
+	/* for 4/8/16 high tiles */
+	case DC_SW_LINEAR:
+		return dm_4k_tile;
+	case DC_SW_4KB_S:
+	case DC_SW_4KB_S_X:
+		return dm_4k_tile;
+	case DC_SW_64KB_S:
+	case DC_SW_64KB_S_X:
+	case DC_SW_64KB_S_T:
+		return dm_64k_tile;
+	case DC_SW_VAR_S:
+	case DC_SW_VAR_S_X:
+		return dm_256k_tile;
+
+	/* For 64bpp 2 high tiles */
+	case DC_SW_4KB_D:
+	case DC_SW_4KB_D_X:
+		return dm_4k_tile;
+	case DC_SW_64KB_D:
+	case DC_SW_64KB_D_X:
+	case DC_SW_64KB_D_T:
+		return dm_64k_tile;
+	case DC_SW_VAR_D:
+	case DC_SW_VAR_D_X:
+		return dm_256k_tile;
+
+	case DC_SW_4KB_R:
+	case DC_SW_4KB_R_X:
+		return dm_4k_tile;
+	case DC_SW_64KB_R:
+	case DC_SW_64KB_R_X:
+		return dm_64k_tile;
+	case DC_SW_VAR_R:
+	case DC_SW_VAR_R_X:
+		return dm_256k_tile;
+
+	/* Unsupported swizzle modes for dcn */
+	case DC_SW_256B_S:
+	default:
+		ASSERT(0); /* Not supported */
+		return 0;
+	}
+}
+
 static void pipe_ctx_to_e2e_pipe_params (
 		const struct pipe_ctx *pipe,
 		struct _vcs_dpi_display_pipe_params_st *input)
@@ -287,46 +334,7 @@ static void pipe_ctx_to_e2e_pipe_params (
 	input->src.cur0_src_width      = 128; /* TODO: Cursor calcs, not curently stored */
 	input->src.cur0_bpp            = 32;
 
-	switch (pipe->plane_state->tiling_info.gfx9.swizzle) {
-	/* for 4/8/16 high tiles */
-	case DC_SW_LINEAR:
-		input->src.macro_tile_size = dm_4k_tile;
-		break;
-	case DC_SW_4KB_S:
-	case DC_SW_4KB_S_X:
-		input->src.macro_tile_size = dm_4k_tile;
-		break;
-	case DC_SW_64KB_S:
-	case DC_SW_64KB_S_X:
-	case DC_SW_64KB_S_T:
-		input->src.macro_tile_size = dm_64k_tile;
-		break;
-	case DC_SW_VAR_S:
-	case DC_SW_VAR_S_X:
-		input->src.macro_tile_size = dm_256k_tile;
-		break;
-
-	/* For 64bpp 2 high tiles */
-	case DC_SW_4KB_D:
-	case DC_SW_4KB_D_X:
-		input->src.macro_tile_size = dm_4k_tile;
-		break;
-	case DC_SW_64KB_D:
-	case DC_SW_64KB_D_X:
-	case DC_SW_64KB_D_T:
-		input->src.macro_tile_size = dm_64k_tile;
-		break;
-	case DC_SW_VAR_D:
-	case DC_SW_VAR_D_X:
-		input->src.macro_tile_size = dm_256k_tile;
-		break;
-
-	/* Unsupported swizzle modes for dcn */
-	case DC_SW_256B_S:
-	default:
-		ASSERT(0); /* Not supported */
-		break;
-	}
+	input->src.macro_tile_size = swizzle_mode_to_macro_tile_size(pipe->plane_state->tiling_info.gfx9.swizzle);
 
 	switch (pipe->plane_state->rotation) {
 	case ROTATION_ANGLE_0:
@@ -466,7 +474,7 @@ static void dcn_bw_calc_rq_dlg_ttu(
 	input.clks_cfg.dcfclk_mhz = v->dcfclk;
 	input.clks_cfg.dispclk_mhz = v->dispclk;
 	input.clks_cfg.dppclk_mhz = v->dppclk;
-	input.clks_cfg.refclk_mhz = dc->res_pool->ref_clock_inKhz / 1000.0;
+	input.clks_cfg.refclk_mhz = dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000.0;
 	input.clks_cfg.socclk_mhz = v->socclk;
 	input.clks_cfg.voltage = v->voltage_level;
 //	dc->dml.logger = pool->base.logger;
@@ -536,28 +544,28 @@ static void calc_wm_sets_and_perf_params(
 		v->fabric_and_dram_bandwidth = v->fabric_and_dram_bandwidth_vnom0p8;
 		dispclkdppclkdcfclk_deep_sleep_prefetch_parameters_watermarks_and_performance_calculation(v);
 
-		context->bw.dcn.watermarks.b.cstate_pstate.cstate_exit_ns =
+		context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.cstate_exit_ns =
 			v->stutter_exit_watermark * 1000;
-		context->bw.dcn.watermarks.b.cstate_pstate.cstate_enter_plus_exit_ns =
+		context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.cstate_enter_plus_exit_ns =
 				v->stutter_enter_plus_exit_watermark * 1000;
-		context->bw.dcn.watermarks.b.cstate_pstate.pstate_change_ns =
+		context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.pstate_change_ns =
 				v->dram_clock_change_watermark * 1000;
-		context->bw.dcn.watermarks.b.pte_meta_urgent_ns = v->ptemeta_urgent_watermark * 1000;
-		context->bw.dcn.watermarks.b.urgent_ns = v->urgent_watermark * 1000;
+		context->bw_ctx.bw.dcn.watermarks.b.pte_meta_urgent_ns = v->ptemeta_urgent_watermark * 1000;
+		context->bw_ctx.bw.dcn.watermarks.b.urgent_ns = v->urgent_watermark * 1000;
 
 		v->dcfclk_per_state[1] = v->dcfclkv_nom0p8;
 		v->dcfclk_per_state[0] = v->dcfclkv_nom0p8;
 		v->dcfclk = v->dcfclkv_nom0p8;
 		dispclkdppclkdcfclk_deep_sleep_prefetch_parameters_watermarks_and_performance_calculation(v);
 
-		context->bw.dcn.watermarks.c.cstate_pstate.cstate_exit_ns =
+		context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_exit_ns =
 			v->stutter_exit_watermark * 1000;
-		context->bw.dcn.watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns =
+		context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns =
 				v->stutter_enter_plus_exit_watermark * 1000;
-		context->bw.dcn.watermarks.c.cstate_pstate.pstate_change_ns =
+		context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.pstate_change_ns =
 				v->dram_clock_change_watermark * 1000;
-		context->bw.dcn.watermarks.c.pte_meta_urgent_ns = v->ptemeta_urgent_watermark * 1000;
-		context->bw.dcn.watermarks.c.urgent_ns = v->urgent_watermark * 1000;
+		context->bw_ctx.bw.dcn.watermarks.c.pte_meta_urgent_ns = v->ptemeta_urgent_watermark * 1000;
+		context->bw_ctx.bw.dcn.watermarks.c.urgent_ns = v->urgent_watermark * 1000;
 	}
 
 	if (v->voltage_level < 3) {
@@ -571,14 +579,14 @@ static void calc_wm_sets_and_perf_params(
 		v->dcfclk = v->dcfclkv_max0p9;
 		dispclkdppclkdcfclk_deep_sleep_prefetch_parameters_watermarks_and_performance_calculation(v);
 
-		context->bw.dcn.watermarks.d.cstate_pstate.cstate_exit_ns =
+		context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.cstate_exit_ns =
 			v->stutter_exit_watermark * 1000;
-		context->bw.dcn.watermarks.d.cstate_pstate.cstate_enter_plus_exit_ns =
+		context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.cstate_enter_plus_exit_ns =
 				v->stutter_enter_plus_exit_watermark * 1000;
-		context->bw.dcn.watermarks.d.cstate_pstate.pstate_change_ns =
+		context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.pstate_change_ns =
 				v->dram_clock_change_watermark * 1000;
-		context->bw.dcn.watermarks.d.pte_meta_urgent_ns = v->ptemeta_urgent_watermark * 1000;
-		context->bw.dcn.watermarks.d.urgent_ns = v->urgent_watermark * 1000;
+		context->bw_ctx.bw.dcn.watermarks.d.pte_meta_urgent_ns = v->ptemeta_urgent_watermark * 1000;
+		context->bw_ctx.bw.dcn.watermarks.d.urgent_ns = v->urgent_watermark * 1000;
 	}
 
 	v->fabric_and_dram_bandwidth_per_state[2] = v->fabric_and_dram_bandwidth_vnom0p8;
@@ -591,20 +599,20 @@ static void calc_wm_sets_and_perf_params(
 	v->dcfclk = v->dcfclk_per_state[v->voltage_level];
 	dispclkdppclkdcfclk_deep_sleep_prefetch_parameters_watermarks_and_performance_calculation(v);
 
-	context->bw.dcn.watermarks.a.cstate_pstate.cstate_exit_ns =
+	context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_exit_ns =
 		v->stutter_exit_watermark * 1000;
-	context->bw.dcn.watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns =
+	context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns =
 			v->stutter_enter_plus_exit_watermark * 1000;
-	context->bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns =
+	context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns =
 			v->dram_clock_change_watermark * 1000;
-	context->bw.dcn.watermarks.a.pte_meta_urgent_ns = v->ptemeta_urgent_watermark * 1000;
-	context->bw.dcn.watermarks.a.urgent_ns = v->urgent_watermark * 1000;
+	context->bw_ctx.bw.dcn.watermarks.a.pte_meta_urgent_ns = v->ptemeta_urgent_watermark * 1000;
+	context->bw_ctx.bw.dcn.watermarks.a.urgent_ns = v->urgent_watermark * 1000;
 	if (v->voltage_level >= 2) {
-		context->bw.dcn.watermarks.b = context->bw.dcn.watermarks.a;
-		context->bw.dcn.watermarks.c = context->bw.dcn.watermarks.a;
+		context->bw_ctx.bw.dcn.watermarks.b = context->bw_ctx.bw.dcn.watermarks.a;
+		context->bw_ctx.bw.dcn.watermarks.c = context->bw_ctx.bw.dcn.watermarks.a;
 	}
 	if (v->voltage_level >= 3)
-		context->bw.dcn.watermarks.d = context->bw.dcn.watermarks.a;
+		context->bw_ctx.bw.dcn.watermarks.d = context->bw_ctx.bw.dcn.watermarks.a;
 }
 #endif
 
@@ -693,8 +701,15 @@ static void hack_bounding_box(struct dcn_bw_internal_vars *v,
 
 bool dcn_validate_bandwidth(
 		struct dc *dc,
-		struct dc_state *context)
+		struct dc_state *context,
+		bool fast_validate)
 {
+	/*
+	 * we want a breakdown of the various stages of validation, which the
+	 * perf_trace macro doesn't support
+	 */
+	BW_VAL_TRACE_SETUP();
+
 	const struct resource_pool *pool = dc->res_pool;
 	struct dcn_bw_internal_vars *v = &context->dcn_bw_vars;
 	int i, input_idx;
@@ -703,6 +718,9 @@ bool dcn_validate_bandwidth(
 	float bw_limit;
 
 	PERFORMANCE_TRACE_START();
+
+	BW_VAL_TRACE_COUNT();
+
 	if (dcn_bw_apply_registry_override(dc))
 		dcn_bw_sync_calcs_and_dml(dc);
 
@@ -1000,13 +1018,16 @@ bool dcn_validate_bandwidth(
 				dc->debug.sr_enter_plus_exit_time_dpm0_ns / 1000.0f;
 		if (dc->debug.sr_exit_time_dpm0_ns)
 			v->sr_exit_time =  dc->debug.sr_exit_time_dpm0_ns / 1000.0f;
-		dc->dml.soc.sr_enter_plus_exit_time_us = v->sr_enter_plus_exit_time;
-		dc->dml.soc.sr_exit_time_us = v->sr_exit_time;
+		context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = v->sr_enter_plus_exit_time;
+		context->bw_ctx.dml.soc.sr_exit_time_us = v->sr_exit_time;
 		mode_support_and_system_configuration(v);
 	}
 
-	if (v->voltage_level != 5) {
+	BW_VAL_TRACE_END_VOLTAGE_LEVEL();
+
+	if (v->voltage_level != number_of_states_plus_one && !fast_validate) {
 		float bw_consumed = v->total_bandwidth_consumed_gbyte_per_second;
+
 		if (bw_consumed < v->fabric_and_dram_bandwidth_vmin0p65)
 			bw_consumed = v->fabric_and_dram_bandwidth_vmin0p65;
 		else if (bw_consumed < v->fabric_and_dram_bandwidth_vmid0p72)
@@ -1027,58 +1048,60 @@ bool dcn_validate_bandwidth(
 		 */
 		dispclkdppclkdcfclk_deep_sleep_prefetch_parameters_watermarks_and_performance_calculation(v);
 
-		context->bw.dcn.watermarks.a.cstate_pstate.cstate_exit_ns =
+		context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_exit_ns =
 			v->stutter_exit_watermark * 1000;
-		context->bw.dcn.watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns =
+		context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns =
 				v->stutter_enter_plus_exit_watermark * 1000;
-		context->bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns =
+		context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns =
 				v->dram_clock_change_watermark * 1000;
-		context->bw.dcn.watermarks.a.pte_meta_urgent_ns = v->ptemeta_urgent_watermark * 1000;
-		context->bw.dcn.watermarks.a.urgent_ns = v->urgent_watermark * 1000;
-		context->bw.dcn.watermarks.b = context->bw.dcn.watermarks.a;
-		context->bw.dcn.watermarks.c = context->bw.dcn.watermarks.a;
-		context->bw.dcn.watermarks.d = context->bw.dcn.watermarks.a;
+		context->bw_ctx.bw.dcn.watermarks.a.pte_meta_urgent_ns = v->ptemeta_urgent_watermark * 1000;
+		context->bw_ctx.bw.dcn.watermarks.a.urgent_ns = v->urgent_watermark * 1000;
+		context->bw_ctx.bw.dcn.watermarks.b = context->bw_ctx.bw.dcn.watermarks.a;
+		context->bw_ctx.bw.dcn.watermarks.c = context->bw_ctx.bw.dcn.watermarks.a;
+		context->bw_ctx.bw.dcn.watermarks.d = context->bw_ctx.bw.dcn.watermarks.a;
 
-		context->bw.dcn.clk.fclk_khz = (int)(bw_consumed * 1000000 /
+		context->bw_ctx.bw.dcn.clk.fclk_khz = (int)(bw_consumed * 1000000 /
 				(ddr4_dram_factor_single_Channel * v->number_of_channels));
 		if (bw_consumed == v->fabric_and_dram_bandwidth_vmin0p65) {
-			context->bw.dcn.clk.fclk_khz = (int)(bw_consumed * 1000000 / 32);
+			context->bw_ctx.bw.dcn.clk.fclk_khz = (int)(bw_consumed * 1000000 / 32);
 		}
 
-		context->bw.dcn.clk.dcfclk_deep_sleep_khz = (int)(v->dcf_clk_deep_sleep * 1000);
-		context->bw.dcn.clk.dcfclk_khz = (int)(v->dcfclk * 1000);
+		context->bw_ctx.bw.dcn.clk.dcfclk_deep_sleep_khz = (int)(v->dcf_clk_deep_sleep * 1000);
+		context->bw_ctx.bw.dcn.clk.dcfclk_khz = (int)(v->dcfclk * 1000);
 
-		context->bw.dcn.clk.dispclk_khz = (int)(v->dispclk * 1000);
+		context->bw_ctx.bw.dcn.clk.dispclk_khz = (int)(v->dispclk * 1000);
 		if (dc->debug.max_disp_clk == true)
-			context->bw.dcn.clk.dispclk_khz = (int)(dc->dcn_soc->max_dispclk_vmax0p9 * 1000);
+			context->bw_ctx.bw.dcn.clk.dispclk_khz = (int)(dc->dcn_soc->max_dispclk_vmax0p9 * 1000);
 
-		if (context->bw.dcn.clk.dispclk_khz <
+		if (context->bw_ctx.bw.dcn.clk.dispclk_khz <
 				dc->debug.min_disp_clk_khz) {
-			context->bw.dcn.clk.dispclk_khz =
+			context->bw_ctx.bw.dcn.clk.dispclk_khz =
 					dc->debug.min_disp_clk_khz;
 		}
 
-		context->bw.dcn.clk.dppclk_khz = context->bw.dcn.clk.dispclk_khz / v->dispclk_dppclk_ratio;
-		context->bw.dcn.clk.phyclk_khz = v->phyclk_per_state[v->voltage_level];
+		context->bw_ctx.bw.dcn.clk.dppclk_khz = context->bw_ctx.bw.dcn.clk.dispclk_khz / v->dispclk_dppclk_ratio;
+		context->bw_ctx.bw.dcn.clk.phyclk_khz = v->phyclk_per_state[v->voltage_level];
 		switch (v->voltage_level) {
 		case 0:
-			context->bw.dcn.clk.max_supported_dppclk_khz =
+			context->bw_ctx.bw.dcn.clk.max_supported_dppclk_khz =
 					(int)(dc->dcn_soc->max_dppclk_vmin0p65 * 1000);
 			break;
 		case 1:
-			context->bw.dcn.clk.max_supported_dppclk_khz =
+			context->bw_ctx.bw.dcn.clk.max_supported_dppclk_khz =
 					(int)(dc->dcn_soc->max_dppclk_vmid0p72 * 1000);
 			break;
 		case 2:
-			context->bw.dcn.clk.max_supported_dppclk_khz =
+			context->bw_ctx.bw.dcn.clk.max_supported_dppclk_khz =
 					(int)(dc->dcn_soc->max_dppclk_vnom0p8 * 1000);
 			break;
 		default:
-			context->bw.dcn.clk.max_supported_dppclk_khz =
+			context->bw_ctx.bw.dcn.clk.max_supported_dppclk_khz =
 					(int)(dc->dcn_soc->max_dppclk_vmax0p9 * 1000);
 			break;
 		}
 
+		BW_VAL_TRACE_END_WATERMARKS();
+
 		for (i = 0, input_idx = 0; i < pool->pipe_count; i++) {
 			struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
 
@@ -1141,7 +1164,7 @@ bool dcn_validate_bandwidth(
 						hsplit_pipe->pipe_dlg_param.vblank_end = pipe->pipe_dlg_param.vblank_end;
 					} else {
 						/* pipe not split previously needs split */
-						hsplit_pipe = find_idle_secondary_pipe(&context->res_ctx, pool);
+						hsplit_pipe = find_idle_secondary_pipe(&context->res_ctx, pool, pipe);
 						ASSERT(hsplit_pipe);
 						split_stream_across_pipes(
 							&context->res_ctx, pool,
@@ -1169,13 +1192,17 @@ bool dcn_validate_bandwidth(
 
 			input_idx++;
 		}
+	} else if (v->voltage_level == number_of_states_plus_one) {
+		BW_VAL_TRACE_SKIP(fail);
+	} else if (fast_validate) {
+		BW_VAL_TRACE_SKIP(fast);
 	}
 
 	if (v->voltage_level == 0) {
 
-		dc->dml.soc.sr_enter_plus_exit_time_us =
+		context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us =
 				dc->dcn_soc->sr_enter_plus_exit_time;
-		dc->dml.soc.sr_exit_time_us = dc->dcn_soc->sr_exit_time;
+		context->bw_ctx.dml.soc.sr_exit_time_us = dc->dcn_soc->sr_exit_time;
 	}
 
 	/*
@@ -1188,6 +1215,7 @@ bool dcn_validate_bandwidth(
 	kernel_fpu_end();
 
 	PERFORMANCE_TRACE_END();
+	BW_VAL_TRACE_FINISH();
 
 	if (bw_limit_pass && v->voltage_level != 5)
 		return true;
@@ -1395,12 +1423,14 @@ void dcn_bw_update_from_pplib(struct dc *dc)
 
 void dcn_bw_notify_pplib_of_wm_ranges(struct dc *dc)
 {
-	struct pp_smu_funcs_rv *pp = dc->res_pool->pp_smu;
+	struct pp_smu_funcs_rv *pp = NULL;
 	struct pp_smu_wm_range_sets ranges = {0};
 	int min_fclk_khz, min_dcfclk_khz, socclk_khz;
 	const int overdrive = 5000000; /* 5 GHz to cover Overdrive */
 
-	if (!pp->set_wm_ranges)
+	if (dc->res_pool->pp_smu)
+		pp = &dc->res_pool->pp_smu->rv_funcs;
+	if (!pp || !pp->set_wm_ranges)
 		return;
 
 	kernel_fpu_begin();
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index a6cda201c964..18c775a950cc 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -524,6 +524,14 @@ void dc_link_set_preferred_link_settings(struct dc *dc,
 	struct dc_stream_state *link_stream;
 	struct dc_link_settings store_settings = *link_setting;
 
+	link->preferred_link_setting = store_settings;
+
+	/* Retrain with preferred link settings only relevant for
+	 * DP signal type
+	 */
+	if (!dc_is_dp_signal(link->connector_signal))
+		return;
+
 	for (i = 0; i < MAX_PIPES; i++) {
 		pipe = &dc->current_state->res_ctx.pipe_ctx[i];
 		if (pipe->stream && pipe->stream->link) {
@@ -538,7 +546,10 @@ void dc_link_set_preferred_link_settings(struct dc *dc,
 
 	link_stream = link->dc->current_state->res_ctx.pipe_ctx[i].stream;
 
-	link->preferred_link_setting = store_settings;
+	/* Cannot retrain link if backend is off */
+	if (link_stream->dpms_off)
+		return;
+
 	if (link_stream)
 		decide_link_settings(link_stream, &store_settings);
 
@@ -573,6 +584,28 @@ void dc_link_set_test_pattern(struct dc_link *link,
 			cust_pattern_size);
 }
 
+uint32_t dc_link_bandwidth_kbps(
+	const struct dc_link *link,
+	const struct dc_link_settings *link_setting)
+{
+	uint32_t link_bw_kbps = link_setting->link_rate * LINK_RATE_REF_FREQ_IN_KHZ; /* bytes per sec */
+
+	link_bw_kbps *= 8;   /* 8 bits per byte*/
+	link_bw_kbps *= link_setting->lane_count;
+
+	return link_bw_kbps;
+
+}
+
+const struct dc_link_settings *dc_link_get_link_cap(
+		const struct dc_link *link)
+{
+	if (link->preferred_link_setting.lane_count != LANE_COUNT_UNKNOWN &&
+			link->preferred_link_setting.link_rate != LINK_RATE_UNKNOWN)
+		return &link->preferred_link_setting;
+	return &link->verified_link_cap;
+}
+
 static void destruct(struct dc *dc)
 {
 	dc_release_state(dc->current_state);
@@ -621,6 +654,10 @@ static bool construct(struct dc *dc,
 #endif
 
 	enum dce_version dc_version = DCE_VERSION_UNKNOWN;
+	dc->config = init_params->flags;
+
+	memcpy(&dc->bb_overrides, &init_params->bb_overrides, sizeof(dc->bb_overrides));
+
 	dc_dceip = kzalloc(sizeof(*dc_dceip), GFP_KERNEL);
 	if (!dc_dceip) {
 		dm_error("%s: failed to create dceip\n", __func__);
@@ -668,13 +705,6 @@ static bool construct(struct dc *dc,
 	dc_ctx->dc_stream_id_count = 0;
 	dc->ctx = dc_ctx;
 
-	dc->current_state = dc_create_state();
-
-	if (!dc->current_state) {
-		dm_error("%s: failed to create validate ctx\n", __func__);
-		goto fail;
-	}
-
 	/* Create logger */
 
 	dc_ctx->dce_environment = init_params->dce_environment;
@@ -722,14 +752,22 @@ static bool construct(struct dc *dc,
 		goto fail;
 	}
 
-	dc->res_pool = dc_create_resource_pool(
-			dc,
-			init_params->num_virtual_links,
-			dc_version,
-			init_params->asic_id);
+	dc->res_pool = dc_create_resource_pool(dc, init_params, dc_version);
 	if (!dc->res_pool)
 		goto fail;
 
+	/* Creation of current_state must occur after dc->dml
+	 * is initialized in dc_create_resource_pool because
+	 * on creation it copies the contents of dc->dml
+	 */
+
+	dc->current_state = dc_create_state(dc);
+
+	if (!dc->current_state) {
+		dm_error("%s: failed to create validate ctx\n", __func__);
+		goto fail;
+	}
+
 	dc_resource_state_construct(dc, dc->current_state);
 
 	if (!create_links(dc, init_params->num_virtual_links))
@@ -746,7 +784,7 @@ fail:
 static void disable_dangling_plane(struct dc *dc, struct dc_state *context)
 {
 	int i, j;
-	struct dc_state *dangling_context = dc_create_state();
+	struct dc_state *dangling_context = dc_create_state(dc);
 	struct dc_state *current_ctx;
 
 	if (dangling_context == NULL)
@@ -811,8 +849,6 @@ struct dc *dc_create(const struct dc_init_data *init_params)
 	if (dc->res_pool->dmcu != NULL)
 		dc->versions.dmcu_version = dc->res_pool->dmcu->dmcu_version;
 
-	dc->config = init_params->flags;
-
 	dc->build_id = DC_BUILD_ID;
 
 	DC_LOG_DC("Display Core initialized\n");
@@ -969,7 +1005,7 @@ static bool context_changed(
 	return false;
 }
 
-bool dc_validate_seamless_boot_timing(struct dc *dc,
+bool dc_validate_seamless_boot_timing(const struct dc *dc,
 				const struct dc_sink *sink,
 				struct dc_crtc_timing *crtc_timing)
 {
@@ -1060,7 +1096,13 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c
 	if (!dcb->funcs->is_accelerated_mode(dcb))
 		dc->hwss.enable_accelerated_mode(dc, context);
 
-	dc->hwss.prepare_bandwidth(dc, context);
+	for (i = 0; i < context->stream_count; i++) {
+		if (context->streams[i]->apply_seamless_boot_optimization)
+			dc->optimize_seamless_boot = true;
+	}
+
+	if (!dc->optimize_seamless_boot)
+		dc->hwss.prepare_bandwidth(dc, context);
 
 	/* re-program planes for existing stream, in case we need to
 	 * free up plane resource for later use
@@ -1135,12 +1177,15 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c
 
 	dc_enable_stereo(dc, context, dc_streams, context->stream_count);
 
-	/* pplib is notified if disp_num changed */
-	dc->hwss.optimize_bandwidth(dc, context);
+	if (!dc->optimize_seamless_boot)
+		/* pplib is notified if disp_num changed */
+		dc->hwss.optimize_bandwidth(dc, context);
 
 	for (i = 0; i < context->stream_count; i++)
 		context->streams[i]->mode_changed = false;
 
+	memset(&context->commit_hints, 0, sizeof(context->commit_hints));
+
 	dc_release_state(dc->current_state);
 
 	dc->current_state = context;
@@ -1177,7 +1222,7 @@ bool dc_post_update_surfaces_to_stream(struct dc *dc)
 	int i;
 	struct dc_state *context = dc->current_state;
 
-	if (dc->optimized_required == false)
+	if (!dc->optimized_required || dc->optimize_seamless_boot)
 		return true;
 
 	post_surface_trace(dc);
@@ -1195,18 +1240,60 @@ bool dc_post_update_surfaces_to_stream(struct dc *dc)
 	return true;
 }
 
-struct dc_state *dc_create_state(void)
+struct dc_state *dc_create_state(struct dc *dc)
 {
 	struct dc_state *context = kzalloc(sizeof(struct dc_state),
 					   GFP_KERNEL);
 
 	if (!context)
 		return NULL;
+	/* Each context must have their own instance of VBA and in order to
+	 * initialize and obtain IP and SOC the base DML instance from DC is
+	 * initially copied into every context
+	 */
+#ifdef CONFIG_DRM_AMD_DC_DCN1_0
+	memcpy(&context->bw_ctx.dml, &dc->dml, sizeof(struct display_mode_lib));
+#endif
 
 	kref_init(&context->refcount);
+
 	return context;
 }
 
+struct dc_state *dc_copy_state(struct dc_state *src_ctx)
+{
+	int i, j;
+	struct dc_state *new_ctx = kzalloc(sizeof(struct dc_state),
+					   GFP_KERNEL);
+
+	if (!new_ctx)
+		return NULL;
+
+	memcpy(new_ctx, src_ctx, sizeof(struct dc_state));
+
+	for (i = 0; i < MAX_PIPES; i++) {
+			struct pipe_ctx *cur_pipe = &new_ctx->res_ctx.pipe_ctx[i];
+
+			if (cur_pipe->top_pipe)
+				cur_pipe->top_pipe =  &new_ctx->res_ctx.pipe_ctx[cur_pipe->top_pipe->pipe_idx];
+
+			if (cur_pipe->bottom_pipe)
+				cur_pipe->bottom_pipe = &new_ctx->res_ctx.pipe_ctx[cur_pipe->bottom_pipe->pipe_idx];
+
+	}
+
+	for (i = 0; i < new_ctx->stream_count; i++) {
+			dc_stream_retain(new_ctx->streams[i]);
+			for (j = 0; j < new_ctx->stream_status[i].plane_count; j++)
+				dc_plane_state_retain(
+					new_ctx->stream_status[i].plane_states[j]);
+	}
+
+	kref_init(&new_ctx->refcount);
+
+	return new_ctx;
+}
+
 void dc_retain_state(struct dc_state *context)
 {
 	kref_get(&context->refcount);
@@ -1666,6 +1753,7 @@ static void commit_planes_do_stream_update(struct dc *dc,
 				continue;
 
 			if (stream_update->dpms_off) {
+				dc->hwss.pipe_control_lock(dc, pipe_ctx, true);
 				if (*stream_update->dpms_off) {
 					core_link_disable_stream(pipe_ctx, KEEP_ACQUIRED_RESOURCE);
 					dc->hwss.optimize_bandwidth(dc, dc->current_state);
@@ -1673,6 +1761,7 @@ static void commit_planes_do_stream_update(struct dc *dc,
 					dc->hwss.prepare_bandwidth(dc, dc->current_state);
 					core_link_enable_stream(dc->current_state, pipe_ctx);
 				}
+				dc->hwss.pipe_control_lock(dc, pipe_ctx, false);
 			}
 
 			if (stream_update->abm_level && pipe_ctx->stream_res.abm) {
@@ -1700,7 +1789,16 @@ static void commit_planes_for_stream(struct dc *dc,
 	int i, j;
 	struct pipe_ctx *top_pipe_to_program = NULL;
 
-	if (update_type == UPDATE_TYPE_FULL) {
+	if (dc->optimize_seamless_boot && surface_count > 0) {
+		/* Optimize seamless boot flag keeps clocks and watermarks high until
+		 * first flip. After first flip, optimization is required to lower
+		 * bandwidth.
+		 */
+		dc->optimize_seamless_boot = false;
+		dc->optimized_required = true;
+	}
+
+	if (update_type == UPDATE_TYPE_FULL && !dc->optimize_seamless_boot) {
 		dc->hwss.prepare_bandwidth(dc, context);
 		context_clock_trace(dc, context);
 	}
@@ -1800,7 +1898,7 @@ void dc_commit_updates_for_stream(struct dc *dc,
 	if (update_type >= UPDATE_TYPE_FULL) {
 
 		/* initialize scratch memory for building context */
-		context = dc_create_state();
+		context = dc_create_state(dc);
 		if (context == NULL) {
 			DC_ERROR("Failed to allocate new validate context!\n");
 			return;
@@ -2099,13 +2197,13 @@ void dc_link_remove_remote_sink(struct dc_link *link, struct dc_sink *sink)
 
 void get_clock_requirements_for_state(struct dc_state *state, struct AsicStateEx *info)
 {
-	info->displayClock				= (unsigned int)state->bw.dcn.clk.dispclk_khz;
-	info->engineClock				= (unsigned int)state->bw.dcn.clk.dcfclk_khz;
-	info->memoryClock				= (unsigned int)state->bw.dcn.clk.dramclk_khz;
-	info->maxSupportedDppClock		= (unsigned int)state->bw.dcn.clk.max_supported_dppclk_khz;
-	info->dppClock					= (unsigned int)state->bw.dcn.clk.dppclk_khz;
-	info->socClock					= (unsigned int)state->bw.dcn.clk.socclk_khz;
-	info->dcfClockDeepSleep			= (unsigned int)state->bw.dcn.clk.dcfclk_deep_sleep_khz;
-	info->fClock					= (unsigned int)state->bw.dcn.clk.fclk_khz;
-	info->phyClock					= (unsigned int)state->bw.dcn.clk.phyclk_khz;
+	info->displayClock				= (unsigned int)state->bw_ctx.bw.dcn.clk.dispclk_khz;
+	info->engineClock				= (unsigned int)state->bw_ctx.bw.dcn.clk.dcfclk_khz;
+	info->memoryClock				= (unsigned int)state->bw_ctx.bw.dcn.clk.dramclk_khz;
+	info->maxSupportedDppClock		= (unsigned int)state->bw_ctx.bw.dcn.clk.max_supported_dppclk_khz;
+	info->dppClock					= (unsigned int)state->bw_ctx.bw.dcn.clk.dppclk_khz;
+	info->socClock					= (unsigned int)state->bw_ctx.bw.dcn.clk.socclk_khz;
+	info->dcfClockDeepSleep			= (unsigned int)state->bw_ctx.bw.dcn.clk.dcfclk_deep_sleep_khz;
+	info->fClock					= (unsigned int)state->bw_ctx.bw.dcn.clk.fclk_khz;
+	info->phyClock					= (unsigned int)state->bw_ctx.bw.dcn.clk.phyclk_khz;
 }
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_debug.c b/drivers/gpu/drm/amd/display/dc/core/dc_debug.c
index 73d049506618..5903e7822f98 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_debug.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_debug.c
@@ -351,19 +351,19 @@ void context_clock_trace(
 	DC_LOGGER_INIT(dc->ctx->logger);
 	CLOCK_TRACE("Current: dispclk_khz:%d  max_dppclk_khz:%d  dcfclk_khz:%d\n"
 			"dcfclk_deep_sleep_khz:%d  fclk_khz:%d  socclk_khz:%d\n",
-			context->bw.dcn.clk.dispclk_khz,
-			context->bw.dcn.clk.dppclk_khz,
-			context->bw.dcn.clk.dcfclk_khz,
-			context->bw.dcn.clk.dcfclk_deep_sleep_khz,
-			context->bw.dcn.clk.fclk_khz,
-			context->bw.dcn.clk.socclk_khz);
+			context->bw_ctx.bw.dcn.clk.dispclk_khz,
+			context->bw_ctx.bw.dcn.clk.dppclk_khz,
+			context->bw_ctx.bw.dcn.clk.dcfclk_khz,
+			context->bw_ctx.bw.dcn.clk.dcfclk_deep_sleep_khz,
+			context->bw_ctx.bw.dcn.clk.fclk_khz,
+			context->bw_ctx.bw.dcn.clk.socclk_khz);
 	CLOCK_TRACE("Calculated: dispclk_khz:%d  max_dppclk_khz:%d  dcfclk_khz:%d\n"
 			"dcfclk_deep_sleep_khz:%d  fclk_khz:%d  socclk_khz:%d\n",
-			context->bw.dcn.clk.dispclk_khz,
-			context->bw.dcn.clk.dppclk_khz,
-			context->bw.dcn.clk.dcfclk_khz,
-			context->bw.dcn.clk.dcfclk_deep_sleep_khz,
-			context->bw.dcn.clk.fclk_khz,
-			context->bw.dcn.clk.socclk_khz);
+			context->bw_ctx.bw.dcn.clk.dispclk_khz,
+			context->bw_ctx.bw.dcn.clk.dppclk_khz,
+			context->bw_ctx.bw.dcn.clk.dcfclk_khz,
+			context->bw_ctx.bw.dcn.clk.dcfclk_deep_sleep_khz,
+			context->bw_ctx.bw.dcn.clk.fclk_khz,
+			context->bw_ctx.bw.dcn.clk.socclk_khz);
 #endif
 }
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link.c b/drivers/gpu/drm/amd/display/dc/core/dc_link.c
index ea18e9c2d8ce..b37ecc3ede61 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c
@@ -58,7 +58,6 @@
  ******************************************************************************/
 
 enum {
-	LINK_RATE_REF_FREQ_IN_MHZ = 27,
 	PEAK_FACTOR_X1000 = 1006,
 	/*
 	* Some receivers fail to train on first try and are good
@@ -515,6 +514,40 @@ static void link_disconnect_remap(struct dc_sink *prev_sink, struct dc_link *lin
 }
 
 
+static void read_edp_current_link_settings_on_detect(struct dc_link *link)
+{
+	union lane_count_set lane_count_set = { {0} };
+	uint8_t link_bw_set;
+	uint8_t link_rate_set;
+
+	// Read DPCD 00101h to find out the number of lanes currently set
+	core_link_read_dpcd(link, DP_LANE_COUNT_SET,
+			&lane_count_set.raw, sizeof(lane_count_set));
+	link->cur_link_settings.lane_count = lane_count_set.bits.LANE_COUNT_SET;
+
+	// Read DPCD 00100h to find if standard link rates are set
+	core_link_read_dpcd(link, DP_LINK_BW_SET,
+			&link_bw_set, sizeof(link_bw_set));
+
+	if (link_bw_set == 0) {
+		/* If standard link rates are not being used,
+		 * Read DPCD 00115h to find the link rate set used
+		 */
+		core_link_read_dpcd(link, DP_LINK_RATE_SET,
+				&link_rate_set, sizeof(link_rate_set));
+
+		if (link_rate_set < link->dpcd_caps.edp_supported_link_rates_count) {
+			link->cur_link_settings.link_rate =
+				link->dpcd_caps.edp_supported_link_rates[link_rate_set];
+			link->cur_link_settings.link_rate_set = link_rate_set;
+			link->cur_link_settings.use_link_rate_set = true;
+		}
+	} else {
+		link->cur_link_settings.link_rate = link_bw_set;
+		link->cur_link_settings.use_link_rate_set = false;
+	}
+}
+
 static bool detect_dp(
 	struct dc_link *link,
 	struct display_sink_capability *sink_caps,
@@ -640,7 +673,8 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason)
 	bool same_dpcd = true;
 	enum dc_connection_type new_connection_type = dc_connection_none;
 	DC_LOGGER_INIT(link->ctx->logger);
-	if (link->connector_signal == SIGNAL_TYPE_VIRTUAL)
+
+	if (dc_is_virtual_signal(link->connector_signal))
 		return false;
 
 	if (false == dc_link_detect_sink(link, &new_connection_type)) {
@@ -648,9 +682,14 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason)
 		return false;
 	}
 
-	if (link->connector_signal == SIGNAL_TYPE_EDP &&
-			link->local_sink)
-		return true;
+	if (link->connector_signal == SIGNAL_TYPE_EDP) {
+		/* On detect, we want to make sure current link settings are
+		 * up to date, especially if link was powered on by GOP.
+		 */
+		read_edp_current_link_settings_on_detect(link);
+		if (link->local_sink)
+			return true;
+	}
 
 	if (link->connector_signal == SIGNAL_TYPE_LVDS &&
 			link->local_sink)
@@ -720,9 +759,8 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason)
 					same_dpcd = false;
 			}
 			/* Active dongle plug in without display or downstream unplug*/
-			if (link->type == dc_connection_active_dongle
-					&& link->dpcd_caps.sink_count.
-					bits.SINK_COUNT == 0) {
+			if (link->type == dc_connection_active_dongle &&
+				link->dpcd_caps.sink_count.bits.SINK_COUNT == 0) {
 				if (prev_sink != NULL) {
 					/* Downstream unplug */
 					dc_sink_release(prev_sink);
@@ -1172,8 +1210,6 @@ static bool construct(
 		goto create_fail;
 	}
 
-
-
 	/* TODO: #DAL3 Implement id to str function.*/
 	LINK_INFO("Connector[%d] description:"
 			"signal %d\n",
@@ -1207,7 +1243,7 @@ static bool construct(
 	link->link_enc = link->dc->res_pool->funcs->link_enc_create(
 								&enc_init_data);
 
-	if( link->link_enc == NULL) {
+	if (link->link_enc == NULL) {
 		DC_ERROR("Failed to create link encoder!\n");
 		goto link_enc_create_fail;
 	}
@@ -1399,9 +1435,24 @@ static enum dc_status enable_link_dp(
 	/* get link settings for video mode timing */
 	decide_link_settings(stream, &link_settings);
 
+	if (pipe_ctx->stream->signal == SIGNAL_TYPE_EDP) {
+		/* If link settings are different than current and link already enabled
+		 * then need to disable before programming to new rate.
+		 */
+		if (link->link_status.link_active &&
+			(link->cur_link_settings.lane_count != link_settings.lane_count ||
+			 link->cur_link_settings.link_rate != link_settings.link_rate)) {
+			dp_disable_link_phy(link, pipe_ctx->stream->signal);
+		}
+
+		/*in case it is not on*/
+		link->dc->hwss.edp_power_control(link, true);
+		link->dc->hwss.edp_wait_for_hpd_ready(link, true);
+	}
+
 	pipe_ctx->stream_res.pix_clk_params.requested_sym_clk =
 			link_settings.link_rate * LINK_RATE_REF_FREQ_IN_KHZ;
-	state->dccg->funcs->update_clocks(state->dccg, state, false);
+	state->clk_mgr->funcs->update_clocks(state->clk_mgr, state, false);
 
 	dp_enable_link_phy(
 		link,
@@ -1442,15 +1493,9 @@ static enum dc_status enable_link_edp(
 		struct pipe_ctx *pipe_ctx)
 {
 	enum dc_status status;
-	struct dc_stream_state *stream = pipe_ctx->stream;
-	struct dc_link *link = stream->link;
-	/*in case it is not on*/
-	link->dc->hwss.edp_power_control(link, true);
-	link->dc->hwss.edp_wait_for_hpd_ready(link, true);
 
 	status = enable_link_dp(state, pipe_ctx);
 
-
 	return status;
 }
 
@@ -1466,14 +1511,14 @@ static enum dc_status enable_link_dp_mst(
 	if (link->cur_link_settings.lane_count != LANE_COUNT_UNKNOWN)
 		return DC_OK;
 
+	/* clear payload table */
+	dm_helpers_dp_mst_clear_payload_allocation_table(link->ctx, link);
+
 	/* to make sure the pending down rep can be processed
-	 * before clear payload table
+	 * before enabling the link
 	 */
 	dm_helpers_dp_mst_poll_pending_down_reply(link->ctx, link);
 
-	/* clear payload table */
-	dm_helpers_dp_mst_clear_payload_allocation_table(link->ctx, link);
-
 	/* set the sink to MST mode before enabling the link */
 	dp_enable_mst_on_sink(link, true);
 
@@ -1982,7 +2027,7 @@ static void enable_link_hdmi(struct pipe_ctx *pipe_ctx)
 			pipe_ctx->stream->signal,
 			stream->phy_pix_clk);
 
-	if (pipe_ctx->stream->signal == SIGNAL_TYPE_HDMI_TYPE_A)
+	if (dc_is_hdmi_signal(pipe_ctx->stream->signal))
 		dal_ddc_service_read_scdc_data(link->ddc);
 }
 
@@ -2074,11 +2119,28 @@ static void disable_link(struct dc_link *link, enum signal_type signal)
 	}
 }
 
+static uint32_t get_timing_pixel_clock_100hz(const struct dc_crtc_timing *timing)
+{
+
+	uint32_t pxl_clk = timing->pix_clk_100hz;
+
+	if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR420)
+		pxl_clk /= 2;
+	else if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR422)
+		pxl_clk = pxl_clk * 2 / 3;
+
+	if (timing->display_color_depth == COLOR_DEPTH_101010)
+		pxl_clk = pxl_clk * 10 / 8;
+	else if (timing->display_color_depth == COLOR_DEPTH_121212)
+		pxl_clk = pxl_clk * 12 / 8;
+
+	return pxl_clk;
+}
+
 static bool dp_active_dongle_validate_timing(
 		const struct dc_crtc_timing *timing,
 		const struct dpcd_caps *dpcd_caps)
 {
-	unsigned int required_pix_clk_100hz = timing->pix_clk_100hz;
 	const struct dc_dongle_caps *dongle_caps = &dpcd_caps->dongle_caps;
 
 	switch (dpcd_caps->dongle_type) {
@@ -2115,13 +2177,6 @@ static bool dp_active_dongle_validate_timing(
 		return false;
 	}
 
-
-	/* Check Color Depth and Pixel Clock */
-	if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR420)
-		required_pix_clk_100hz /= 2;
-	else if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR422)
-		required_pix_clk_100hz = required_pix_clk_100hz * 2 / 3;
-
 	switch (timing->display_color_depth) {
 	case COLOR_DEPTH_666:
 	case COLOR_DEPTH_888:
@@ -2130,14 +2185,11 @@ static bool dp_active_dongle_validate_timing(
 	case COLOR_DEPTH_101010:
 		if (dongle_caps->dp_hdmi_max_bpc < 10)
 			return false;
-		required_pix_clk_100hz = required_pix_clk_100hz * 10 / 8;
 		break;
 	case COLOR_DEPTH_121212:
 		if (dongle_caps->dp_hdmi_max_bpc < 12)
 			return false;
-		required_pix_clk_100hz = required_pix_clk_100hz * 12 / 8;
 		break;
-
 	case COLOR_DEPTH_141414:
 	case COLOR_DEPTH_161616:
 	default:
@@ -2145,7 +2197,7 @@ static bool dp_active_dongle_validate_timing(
 		return false;
 	}
 
-	if (required_pix_clk_100hz > (dongle_caps->dp_hdmi_max_pixel_clk * 10))
+	if (get_timing_pixel_clock_100hz(timing) > (dongle_caps->dp_hdmi_max_pixel_clk_in_khz * 10))
 		return false;
 
 	return true;
@@ -2166,7 +2218,7 @@ enum dc_status dc_link_validate_mode_timing(
 		return DC_OK;
 
 	/* Passive Dongle */
-	if (0 != max_pix_clk && timing->pix_clk_100hz > max_pix_clk)
+	if (max_pix_clk != 0 && get_timing_pixel_clock_100hz(timing) > max_pix_clk)
 		return DC_EXCEED_DONGLE_CAP;
 
 	/* Active Dongle*/
@@ -2284,14 +2336,13 @@ void core_link_resume(struct dc_link *link)
 
 static struct fixed31_32 get_pbn_per_slot(struct dc_stream_state *stream)
 {
-	struct dc_link_settings *link_settings =
-			&stream->link->cur_link_settings;
-	uint32_t link_rate_in_mbps =
-			link_settings->link_rate * LINK_RATE_REF_FREQ_IN_MHZ;
-	struct fixed31_32 mbps = dc_fixpt_from_int(
-			link_rate_in_mbps * link_settings->lane_count);
-
-	return dc_fixpt_div_int(mbps, 54);
+	struct fixed31_32 mbytes_per_sec;
+	uint32_t link_rate_in_mbytes_per_sec = dc_link_bandwidth_kbps(stream->link, &stream->link->cur_link_settings);
+	link_rate_in_mbytes_per_sec /= 8000; /* Kbits to MBytes */
+
+	mbytes_per_sec = dc_fixpt_from_int(link_rate_in_mbytes_per_sec);
+
+	return dc_fixpt_div_int(mbytes_per_sec, 54);
 }
 
 static int get_color_depth(enum dc_color_depth color_depth)
@@ -2316,7 +2367,7 @@ static struct fixed31_32 get_pbn_from_timing(struct pipe_ctx *pipe_ctx)
 	uint32_t denominator;
 
 	bpc = get_color_depth(pipe_ctx->stream_res.pix_clk_params.color_depth);
-	kbps = pipe_ctx->stream_res.pix_clk_params.requested_pix_clk_100hz / 10 * bpc * 3;
+	kbps = dc_bandwidth_in_kbps_from_timing(&pipe_ctx->stream->timing);
 
 	/*
 	 * margin 5300ppm + 300ppm ~ 0.6% as per spec, factor is 1.006
@@ -2551,12 +2602,12 @@ void core_link_enable_stream(
 		struct dc_state *state,
 		struct pipe_ctx *pipe_ctx)
 {
-	struct dc  *core_dc = pipe_ctx->stream->ctx->dc;
+	struct dc *core_dc = pipe_ctx->stream->ctx->dc;
 	struct dc_stream_state *stream = pipe_ctx->stream;
 	enum dc_status status;
 	DC_LOGGER_INIT(pipe_ctx->stream->ctx->logger);
 
-	if (pipe_ctx->stream->signal != SIGNAL_TYPE_VIRTUAL) {
+	if (!dc_is_virtual_signal(pipe_ctx->stream->signal)) {
 		stream->link->link_enc->funcs->setup(
 			stream->link->link_enc,
 			pipe_ctx->stream->signal);
@@ -2570,9 +2621,10 @@ void core_link_enable_stream(
 		pipe_ctx->stream_res.stream_enc->funcs->dp_set_stream_attribute(
 			pipe_ctx->stream_res.stream_enc,
 			&stream->timing,
-			stream->output_color_space);
+			stream->output_color_space,
+			stream->link->dpcd_caps.dprx_feature.bits.SST_SPLIT_SDP_CAP);
 
-	if (dc_is_hdmi_signal(pipe_ctx->stream->signal))
+	if (dc_is_hdmi_tmds_signal(pipe_ctx->stream->signal))
 		pipe_ctx->stream_res.stream_enc->funcs->hdmi_set_stream_attribute(
 			pipe_ctx->stream_res.stream_enc,
 			&stream->timing,
@@ -2736,3 +2788,49 @@ void dc_link_enable_hpd_filter(struct dc_link *link, bool enable)
 	}
 }
 
+uint32_t dc_bandwidth_in_kbps_from_timing(
+	const struct dc_crtc_timing *timing)
+{
+	uint32_t bits_per_channel = 0;
+	uint32_t kbps;
+
+	switch (timing->display_color_depth) {
+	case COLOR_DEPTH_666:
+		bits_per_channel = 6;
+		break;
+	case COLOR_DEPTH_888:
+		bits_per_channel = 8;
+		break;
+	case COLOR_DEPTH_101010:
+		bits_per_channel = 10;
+		break;
+	case COLOR_DEPTH_121212:
+		bits_per_channel = 12;
+		break;
+	case COLOR_DEPTH_141414:
+		bits_per_channel = 14;
+		break;
+	case COLOR_DEPTH_161616:
+		bits_per_channel = 16;
+		break;
+	default:
+		break;
+	}
+
+	ASSERT(bits_per_channel != 0);
+
+	kbps = timing->pix_clk_100hz / 10;
+	kbps *= bits_per_channel;
+
+	if (timing->flags.Y_ONLY != 1) {
+		/*Only YOnly make reduce bandwidth by 1/3 compares to RGB*/
+		kbps *= 3;
+		if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR420)
+			kbps /= 2;
+		else if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR422)
+			kbps = kbps * 2 / 3;
+	}
+
+	return kbps;
+
+}
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c
index b7ee63cd8dc7..f02092a0dc76 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c
@@ -573,12 +573,28 @@ bool dal_ddc_service_query_ddc_data(
 	return ret;
 }
 
-int dc_link_aux_transfer(struct ddc_service *ddc,
-		struct aux_payload *payload)
+/* dc_link_aux_transfer_raw() - Attempt to transfer
+ * the given aux payload.  This function does not perform
+ * retries or handle error states.  The reply is returned
+ * in the payload->reply and the result through
+ * *operation_result.  Returns the number of bytes transferred,
+ * or -1 on a failure.
+ */
+int dc_link_aux_transfer_raw(struct ddc_service *ddc,
+		struct aux_payload *payload,
+		enum aux_channel_operation_result *operation_result)
 {
-	return dce_aux_transfer(ddc, payload);
+	return dce_aux_transfer_raw(ddc, payload, operation_result);
 }
 
+/* dc_link_aux_transfer_with_retries() - Attempt to submit an
+ * aux payload, retrying on timeouts, defers, and busy states
+ * as outlined in the DP spec.  Returns true if the request
+ * was successful.
+ *
+ * Unless you want to implement your own retry semantics, this
+ * is probably the one you want.
+ */
 bool dc_link_aux_transfer_with_retries(struct ddc_service *ddc,
 		struct aux_payload *payload)
 {
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c
index 09d301216076..1ee544a32ebb 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c
@@ -93,12 +93,10 @@ static void dpcd_set_link_settings(
 	struct dc_link *link,
 	const struct link_training_settings *lt_settings)
 {
-	uint8_t rate = (uint8_t)
-	(lt_settings->link_settings.link_rate);
+	uint8_t rate;
 
 	union down_spread_ctrl downspread = { {0} };
 	union lane_count_set lane_count_set = { {0} };
-	uint8_t link_set_buffer[2];
 
 	downspread.raw = (uint8_t)
 	(lt_settings->link_settings.link_spread);
@@ -111,29 +109,42 @@ static void dpcd_set_link_settings(
 	lane_count_set.bits.POST_LT_ADJ_REQ_GRANTED =
 		link->dpcd_caps.max_ln_count.bits.POST_LT_ADJ_REQ_SUPPORTED;
 
-	link_set_buffer[0] = rate;
-	link_set_buffer[1] = lane_count_set.raw;
-
-	core_link_write_dpcd(link, DP_LINK_BW_SET,
-	link_set_buffer, 2);
 	core_link_write_dpcd(link, DP_DOWNSPREAD_CTRL,
 	&downspread.raw, sizeof(downspread));
 
+	core_link_write_dpcd(link, DP_LANE_COUNT_SET,
+	&lane_count_set.raw, 1);
+
 	if (link->dpcd_caps.dpcd_rev.raw >= DPCD_REV_14 &&
-		(link->dpcd_caps.link_rate_set >= 1 &&
-		link->dpcd_caps.link_rate_set <= 8)) {
+			lt_settings->link_settings.use_link_rate_set == true) {
+		rate = 0;
+		core_link_write_dpcd(link, DP_LINK_BW_SET, &rate, 1);
 		core_link_write_dpcd(link, DP_LINK_RATE_SET,
-		&link->dpcd_caps.link_rate_set, 1);
+				&lt_settings->link_settings.link_rate_set, 1);
+	} else {
+		rate = (uint8_t) (lt_settings->link_settings.link_rate);
+		core_link_write_dpcd(link, DP_LINK_BW_SET, &rate, 1);
 	}
 
-	DC_LOG_HW_LINK_TRAINING("%s\n %x rate = %x\n %x lane = %x\n %x spread = %x\n",
-		__func__,
-		DP_LINK_BW_SET,
-		lt_settings->link_settings.link_rate,
-		DP_LANE_COUNT_SET,
-		lt_settings->link_settings.lane_count,
-		DP_DOWNSPREAD_CTRL,
-		lt_settings->link_settings.link_spread);
+	if (rate) {
+		DC_LOG_HW_LINK_TRAINING("%s\n %x rate = %x\n %x lane = %x\n %x spread = %x\n",
+			__func__,
+			DP_LINK_BW_SET,
+			lt_settings->link_settings.link_rate,
+			DP_LANE_COUNT_SET,
+			lt_settings->link_settings.lane_count,
+			DP_DOWNSPREAD_CTRL,
+			lt_settings->link_settings.link_spread);
+	} else {
+		DC_LOG_HW_LINK_TRAINING("%s\n %x rate set = %x\n %x lane = %x\n %x spread = %x\n",
+			__func__,
+			DP_LINK_RATE_SET,
+			lt_settings->link_settings.link_rate_set,
+			DP_LANE_COUNT_SET,
+			lt_settings->link_settings.lane_count,
+			DP_DOWNSPREAD_CTRL,
+			lt_settings->link_settings.link_spread);
+	}
 
 }
 
@@ -952,6 +963,8 @@ enum link_training_result dc_link_dp_perform_link_training(
 
 	lt_settings.link_settings.link_rate = link_setting->link_rate;
 	lt_settings.link_settings.lane_count = link_setting->lane_count;
+	lt_settings.link_settings.use_link_rate_set = link_setting->use_link_rate_set;
+	lt_settings.link_settings.link_rate_set = link_setting->link_rate_set;
 
 	/*@todo[vdevulap] move SS to LS, should not be handled by displaypath*/
 
@@ -1075,7 +1088,7 @@ static struct dc_link_settings get_max_link_cap(struct dc_link *link)
 {
 	/* Set Default link settings */
 	struct dc_link_settings max_link_cap = {LANE_COUNT_FOUR, LINK_RATE_HIGH,
-			LINK_SPREAD_05_DOWNSPREAD_30KHZ};
+			LINK_SPREAD_05_DOWNSPREAD_30KHZ, false, 0};
 
 	/* Higher link settings based on feature supported */
 	if (link->link_enc->features.flags.bits.IS_HBR2_CAPABLE)
@@ -1520,69 +1533,6 @@ static bool decide_fallback_link_setting(
 	return true;
 }
 
-static uint32_t bandwidth_in_kbps_from_timing(
-	const struct dc_crtc_timing *timing)
-{
-	uint32_t bits_per_channel = 0;
-	uint32_t kbps;
-
-	switch (timing->display_color_depth) {
-	case COLOR_DEPTH_666:
-		bits_per_channel = 6;
-		break;
-	case COLOR_DEPTH_888:
-		bits_per_channel = 8;
-		break;
-	case COLOR_DEPTH_101010:
-		bits_per_channel = 10;
-		break;
-	case COLOR_DEPTH_121212:
-		bits_per_channel = 12;
-		break;
-	case COLOR_DEPTH_141414:
-		bits_per_channel = 14;
-		break;
-	case COLOR_DEPTH_161616:
-		bits_per_channel = 16;
-		break;
-	default:
-		break;
-	}
-
-	ASSERT(bits_per_channel != 0);
-
-	kbps = timing->pix_clk_100hz / 10;
-	kbps *= bits_per_channel;
-
-	if (timing->flags.Y_ONLY != 1) {
-		/*Only YOnly make reduce bandwidth by 1/3 compares to RGB*/
-		kbps *= 3;
-		if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR420)
-			kbps /= 2;
-		else if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR422)
-			kbps = kbps * 2 / 3;
-	}
-
-	return kbps;
-
-}
-
-static uint32_t bandwidth_in_kbps_from_link_settings(
-	const struct dc_link_settings *link_setting)
-{
-	uint32_t link_rate_in_kbps = link_setting->link_rate *
-		LINK_RATE_REF_FREQ_IN_KHZ;
-
-	uint32_t lane_count  = link_setting->lane_count;
-	uint32_t kbps = link_rate_in_kbps;
-
-	kbps *= lane_count;
-	kbps *= 8;   /* 8 bits per byte*/
-
-	return kbps;
-
-}
-
 bool dp_validate_mode_timing(
 	struct dc_link *link,
 	const struct dc_crtc_timing *timing)
@@ -1598,8 +1548,7 @@ bool dp_validate_mode_timing(
 		timing->v_addressable == (uint32_t) 480)
 		return true;
 
-	/* We always use verified link settings */
-	link_setting = &link->verified_link_cap;
+	link_setting = dc_link_get_link_cap(link);
 
 	/* TODO: DYNAMIC_VALIDATION needs to be implemented */
 	/*if (flags.DYNAMIC_VALIDATION == 1 &&
@@ -1607,8 +1556,8 @@ bool dp_validate_mode_timing(
 		link_setting = &link->verified_link_cap;
 	*/
 
-	req_bw = bandwidth_in_kbps_from_timing(timing);
-	max_bw = bandwidth_in_kbps_from_link_settings(link_setting);
+	req_bw = dc_bandwidth_in_kbps_from_timing(timing);
+	max_bw = dc_link_bandwidth_kbps(link, link_setting);
 
 	if (req_bw <= max_bw) {
 		/* remember the biggest mode here, during
@@ -1629,58 +1578,78 @@ bool dp_validate_mode_timing(
 		return false;
 }
 
-void decide_link_settings(struct dc_stream_state *stream,
-	struct dc_link_settings *link_setting)
+static bool decide_dp_link_settings(struct dc_link *link, struct dc_link_settings *link_setting, uint32_t req_bw)
 {
-
 	struct dc_link_settings initial_link_setting = {
-		LANE_COUNT_ONE, LINK_RATE_LOW, LINK_SPREAD_DISABLED};
+		LANE_COUNT_ONE, LINK_RATE_LOW, LINK_SPREAD_DISABLED, false, 0};
 	struct dc_link_settings current_link_setting =
 			initial_link_setting;
-	struct dc_link *link;
-	uint32_t req_bw;
 	uint32_t link_bw;
 
-	req_bw = bandwidth_in_kbps_from_timing(&stream->timing);
-
-	link = stream->link;
-
-	/* if preferred is specified through AMDDP, use it, if it's enough
-	 * to drive the mode
+	/* search for the minimum link setting that:
+	 * 1. is supported according to the link training result
+	 * 2. could support the b/w requested by the timing
 	 */
-	if (link->preferred_link_setting.lane_count !=
-			LANE_COUNT_UNKNOWN &&
-			link->preferred_link_setting.link_rate !=
-					LINK_RATE_UNKNOWN) {
-		*link_setting =  link->preferred_link_setting;
-		return;
-	}
+	while (current_link_setting.link_rate <=
+			link->verified_link_cap.link_rate) {
+		link_bw = dc_link_bandwidth_kbps(
+				link,
+				&current_link_setting);
+		if (req_bw <= link_bw) {
+			*link_setting = current_link_setting;
+			return true;
+		}
 
-	/* MST doesn't perform link training for now
-	 * TODO: add MST specific link training routine
-	 */
-	if (stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) {
-		*link_setting = link->verified_link_cap;
-		return;
+		if (current_link_setting.lane_count <
+				link->verified_link_cap.lane_count) {
+			current_link_setting.lane_count =
+					increase_lane_count(
+							current_link_setting.lane_count);
+		} else {
+			current_link_setting.link_rate =
+					increase_link_rate(
+							current_link_setting.link_rate);
+			current_link_setting.lane_count =
+					initial_link_setting.lane_count;
+		}
 	}
 
-	/* EDP use the link cap setting */
-	if (link->connector_signal == SIGNAL_TYPE_EDP) {
+	return false;
+}
+
+static bool decide_edp_link_settings(struct dc_link *link, struct dc_link_settings *link_setting, uint32_t req_bw)
+{
+	struct dc_link_settings initial_link_setting;
+	struct dc_link_settings current_link_setting;
+	uint32_t link_bw;
+
+	if (link->dpcd_caps.dpcd_rev.raw < DPCD_REV_14 ||
+			link->dpcd_caps.edp_supported_link_rates_count == 0 ||
+			link->dc->config.optimize_edp_link_rate == false) {
 		*link_setting = link->verified_link_cap;
-		return;
+		return true;
 	}
 
+	memset(&initial_link_setting, 0, sizeof(initial_link_setting));
+	initial_link_setting.lane_count = LANE_COUNT_ONE;
+	initial_link_setting.link_rate = link->dpcd_caps.edp_supported_link_rates[0];
+	initial_link_setting.link_spread = LINK_SPREAD_DISABLED;
+	initial_link_setting.use_link_rate_set = true;
+	initial_link_setting.link_rate_set = 0;
+	current_link_setting = initial_link_setting;
+
 	/* search for the minimum link setting that:
 	 * 1. is supported according to the link training result
 	 * 2. could support the b/w requested by the timing
 	 */
 	while (current_link_setting.link_rate <=
 			link->verified_link_cap.link_rate) {
-		link_bw = bandwidth_in_kbps_from_link_settings(
+		link_bw = dc_link_bandwidth_kbps(
+				link,
 				&current_link_setting);
 		if (req_bw <= link_bw) {
 			*link_setting = current_link_setting;
-			return;
+			return true;
 		}
 
 		if (current_link_setting.lane_count <
@@ -1689,13 +1658,53 @@ void decide_link_settings(struct dc_stream_state *stream,
 					increase_lane_count(
 							current_link_setting.lane_count);
 		} else {
-			current_link_setting.link_rate =
-					increase_link_rate(
-							current_link_setting.link_rate);
-			current_link_setting.lane_count =
-					initial_link_setting.lane_count;
+			if (current_link_setting.link_rate_set < link->dpcd_caps.edp_supported_link_rates_count) {
+				current_link_setting.link_rate_set++;
+				current_link_setting.link_rate =
+					link->dpcd_caps.edp_supported_link_rates[current_link_setting.link_rate_set];
+				current_link_setting.lane_count =
+									initial_link_setting.lane_count;
+			} else
+				break;
 		}
 	}
+	return false;
+}
+
+void decide_link_settings(struct dc_stream_state *stream,
+	struct dc_link_settings *link_setting)
+{
+	struct dc_link *link;
+	uint32_t req_bw;
+
+	req_bw = dc_bandwidth_in_kbps_from_timing(&stream->timing);
+
+	link = stream->link;
+
+	/* if preferred is specified through AMDDP, use it, if it's enough
+	 * to drive the mode
+	 */
+	if (link->preferred_link_setting.lane_count !=
+			LANE_COUNT_UNKNOWN &&
+			link->preferred_link_setting.link_rate !=
+					LINK_RATE_UNKNOWN) {
+		*link_setting =  link->preferred_link_setting;
+		return;
+	}
+
+	/* MST doesn't perform link training for now
+	 * TODO: add MST specific link training routine
+	 */
+	if (stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) {
+		*link_setting = link->verified_link_cap;
+		return;
+	}
+
+	if (link->connector_signal == SIGNAL_TYPE_EDP) {
+		if (decide_edp_link_settings(link, link_setting, req_bw))
+			return;
+	} else if (decide_dp_link_settings(link, link_setting, req_bw))
+		return;
 
 	BREAK_TO_DEBUGGER();
 	ASSERT(link->verified_link_cap.lane_count != LANE_COUNT_UNKNOWN);
@@ -2155,11 +2164,7 @@ bool is_mst_supported(struct dc_link *link)
 
 bool is_dp_active_dongle(const struct dc_link *link)
 {
-	enum display_dongle_type dongle_type = link->dpcd_caps.dongle_type;
-
-	return (dongle_type == DISPLAY_DONGLE_DP_VGA_CONVERTER) ||
-			(dongle_type == DISPLAY_DONGLE_DP_DVI_CONVERTER) ||
-			(dongle_type == DISPLAY_DONGLE_DP_HDMI_CONVERTER);
+	return link->dpcd_caps.is_branch_dev;
 }
 
 static int translate_dpcd_max_bpc(enum dpcd_downstream_port_max_bpc bpc)
@@ -2180,6 +2185,30 @@ static int translate_dpcd_max_bpc(enum dpcd_downstream_port_max_bpc bpc)
 	return -1;
 }
 
+static void read_dp_device_vendor_id(struct dc_link *link)
+{
+	struct dp_device_vendor_id dp_id;
+
+	/* read IEEE branch device id */
+	core_link_read_dpcd(
+		link,
+		DP_BRANCH_OUI,
+		(uint8_t *)&dp_id,
+		sizeof(dp_id));
+
+	link->dpcd_caps.branch_dev_id =
+		(dp_id.ieee_oui[0] << 16) +
+		(dp_id.ieee_oui[1] << 8) +
+		dp_id.ieee_oui[2];
+
+	memmove(
+		link->dpcd_caps.branch_dev_name,
+		dp_id.ieee_device_id,
+		sizeof(dp_id.ieee_device_id));
+}
+
+
+
 static void get_active_converter_info(
 	uint8_t data, struct dc_link *link)
 {
@@ -2193,6 +2222,9 @@ static void get_active_converter_info(
 		return;
 	}
 
+	/* DPCD 0x5 bit 0 = 1, it indicate it's branch device */
+	link->dpcd_caps.is_branch_dev = ds_port.fields.PORT_PRESENT;
+
 	switch (ds_port.fields.PORT_TYPE) {
 	case DOWNSTREAM_VGA:
 		link->dpcd_caps.dongle_type = DISPLAY_DONGLE_DP_VGA_CONVERTER;
@@ -2234,8 +2266,8 @@ static void get_active_converter_info(
 					hdmi_caps = {.raw = det_caps[3] };
 				union dwnstream_port_caps_byte2
 					hdmi_color_caps = {.raw = det_caps[2] };
-				link->dpcd_caps.dongle_caps.dp_hdmi_max_pixel_clk =
-					det_caps[1] * 25000;
+				link->dpcd_caps.dongle_caps.dp_hdmi_max_pixel_clk_in_khz =
+					det_caps[1] * 2500;
 
 				link->dpcd_caps.dongle_caps.is_dp_hdmi_s3d_converter =
 					hdmi_caps.bits.FRAME_SEQ_TO_FRAME_PACK;
@@ -2252,7 +2284,7 @@ static void get_active_converter_info(
 					translate_dpcd_max_bpc(
 						hdmi_color_caps.bits.MAX_BITS_PER_COLOR_COMPONENT);
 
-				if (link->dpcd_caps.dongle_caps.dp_hdmi_max_pixel_clk != 0)
+				if (link->dpcd_caps.dongle_caps.dp_hdmi_max_pixel_clk_in_khz != 0)
 					link->dpcd_caps.dongle_caps.extendedCapValid = true;
 			}
 
@@ -2263,27 +2295,6 @@ static void get_active_converter_info(
 	ddc_service_set_dongle_type(link->ddc, link->dpcd_caps.dongle_type);
 
 	{
-		struct dp_device_vendor_id dp_id;
-
-		/* read IEEE branch device id */
-		core_link_read_dpcd(
-			link,
-			DP_BRANCH_OUI,
-			(uint8_t *)&dp_id,
-			sizeof(dp_id));
-
-		link->dpcd_caps.branch_dev_id =
-			(dp_id.ieee_oui[0] << 16) +
-			(dp_id.ieee_oui[1] << 8) +
-			dp_id.ieee_oui[2];
-
-		memmove(
-			link->dpcd_caps.branch_dev_name,
-			dp_id.ieee_device_id,
-			sizeof(dp_id.ieee_device_id));
-	}
-
-	{
 		struct dp_sink_hw_fw_revision dp_hw_fw_revision;
 
 		core_link_read_dpcd(
@@ -2347,6 +2358,10 @@ static bool retrieve_link_cap(struct dc_link *link)
 {
 	uint8_t dpcd_data[DP_ADAPTER_CAP - DP_DPCD_REV + 1];
 
+	/*Only need to read 1 byte starting from DP_DPRX_FEATURE_ENUMERATION_LIST.
+	 */
+	uint8_t dpcd_dprx_data = '\0';
+
 	struct dp_device_vendor_id sink_id;
 	union down_stream_port_count down_strm_port_count;
 	union edp_configuration_cap edp_config_cap;
@@ -2383,7 +2398,10 @@ static bool retrieve_link_cap(struct dc_link *link)
 		aux_rd_interval.raw =
 			dpcd_data[DP_TRAINING_AUX_RD_INTERVAL];
 
-		if (aux_rd_interval.bits.EXT_RECIEVER_CAP_FIELD_PRESENT == 1) {
+		link->dpcd_caps.ext_receiver_cap_field_present =
+				aux_rd_interval.bits.EXT_RECEIVER_CAP_FIELD_PRESENT == 1 ? true:false;
+
+		if (aux_rd_interval.bits.EXT_RECEIVER_CAP_FIELD_PRESENT == 1) {
 			uint8_t ext_cap_data[16];
 
 			memset(ext_cap_data, '\0', sizeof(ext_cap_data));
@@ -2404,11 +2422,44 @@ static bool retrieve_link_cap(struct dc_link *link)
 	}
 
 	link->dpcd_caps.dpcd_rev.raw =
-		dpcd_data[DP_DPCD_REV - DP_DPCD_REV];
+			dpcd_data[DP_DPCD_REV - DP_DPCD_REV];
+
+	if (link->dpcd_caps.dpcd_rev.raw >= 0x14) {
+		for (i = 0; i < read_dpcd_retry_cnt; i++) {
+			status = core_link_read_dpcd(
+					link,
+					DP_DPRX_FEATURE_ENUMERATION_LIST,
+					&dpcd_dprx_data,
+					sizeof(dpcd_dprx_data));
+			if (status == DC_OK)
+				break;
+		}
+
+		link->dpcd_caps.dprx_feature.raw = dpcd_dprx_data;
+
+		if (status != DC_OK)
+			dm_error("%s: Read DPRX caps data failed.\n", __func__);
+	}
+
+	else {
+		link->dpcd_caps.dprx_feature.raw = 0;
+	}
+
+
+	/* Error condition checking...
+	 * It is impossible for Sink to report Max Lane Count = 0.
+	 * It is possible for Sink to report Max Link Rate = 0, if it is
+	 * an eDP device that is reporting specialized link rates in the
+	 * SUPPORTED_LINK_RATE table.
+	 */
+	if (dpcd_data[DP_MAX_LANE_COUNT - DP_DPCD_REV] == 0)
+		return false;
 
 	ds_port.byte = dpcd_data[DP_DOWNSTREAMPORT_PRESENT -
 				 DP_DPCD_REV];
 
+	read_dp_device_vendor_id(link);
+
 	get_active_converter_info(ds_port.byte, link);
 
 	dp_wa_power_up_0010FA(link, dpcd_data, sizeof(dpcd_data));
@@ -2536,31 +2587,31 @@ enum dc_link_rate linkRateInKHzToLinkRateMultiplier(uint32_t link_rate_in_khz)
 
 void detect_edp_sink_caps(struct dc_link *link)
 {
-	uint8_t supported_link_rates[16] = {0};
+	uint8_t supported_link_rates[16];
 	uint32_t entry;
 	uint32_t link_rate_in_khz;
 	enum dc_link_rate link_rate = LINK_RATE_UNKNOWN;
 
 	retrieve_link_cap(link);
+	link->dpcd_caps.edp_supported_link_rates_count = 0;
+	memset(supported_link_rates, 0, sizeof(supported_link_rates));
 
-	if (link->dpcd_caps.dpcd_rev.raw >= DPCD_REV_14) {
+	if (link->dpcd_caps.dpcd_rev.raw >= DPCD_REV_14 &&
+			link->dc->config.optimize_edp_link_rate) {
 		// Read DPCD 00010h - 0001Fh 16 bytes at one shot
 		core_link_read_dpcd(link, DP_SUPPORTED_LINK_RATES,
 							supported_link_rates, sizeof(supported_link_rates));
 
-		link->dpcd_caps.link_rate_set = 0;
 		for (entry = 0; entry < 16; entry += 2) {
 			// DPCD register reports per-lane link rate = 16-bit link rate capability
-			// value X 200 kHz. Need multipler to find link rate in kHz.
+			// value X 200 kHz. Need multiplier to find link rate in kHz.
 			link_rate_in_khz = (supported_link_rates[entry+1] * 0x100 +
 										supported_link_rates[entry]) * 200;
 
 			if (link_rate_in_khz != 0) {
 				link_rate = linkRateInKHzToLinkRateMultiplier(link_rate_in_khz);
-				if (link->reported_link_cap.link_rate < link_rate) {
-					link->reported_link_cap.link_rate = link_rate;
-					link->dpcd_caps.link_rate_set = entry;
-				}
+				link->dpcd_caps.edp_supported_link_rates[link->dpcd_caps.edp_supported_link_rates_count] = link_rate;
+				link->dpcd_caps.edp_supported_link_rates_count++;
 			}
 		}
 	}
@@ -2601,6 +2652,7 @@ static void set_crtc_test_pattern(struct dc_link *link,
 	enum dc_color_depth color_depth = pipe_ctx->
 		stream->timing.display_color_depth;
 	struct bit_depth_reduction_params params;
+	struct output_pixel_processor *opp = pipe_ctx->stream_res.opp;
 
 	memset(&params, 0, sizeof(params));
 
@@ -2640,8 +2692,7 @@ static void set_crtc_test_pattern(struct dc_link *link,
 	{
 		/* disable bit depth reduction */
 		pipe_ctx->stream->bit_depth_params = params;
-		pipe_ctx->stream_res.opp->funcs->
-			opp_program_bit_depth_reduction(pipe_ctx->stream_res.opp, &params);
+		opp->funcs->opp_program_bit_depth_reduction(opp, &params);
 		if (pipe_ctx->stream_res.tg->funcs->set_test_pattern)
 			pipe_ctx->stream_res.tg->funcs->set_test_pattern(pipe_ctx->stream_res.tg,
 				controller_test_pattern, color_depth);
@@ -2650,11 +2701,9 @@ static void set_crtc_test_pattern(struct dc_link *link,
 	case DP_TEST_PATTERN_VIDEO_MODE:
 	{
 		/* restore bitdepth reduction */
-		resource_build_bit_depth_reduction_params(pipe_ctx->stream,
-					&params);
+		resource_build_bit_depth_reduction_params(pipe_ctx->stream, &params);
 		pipe_ctx->stream->bit_depth_params = params;
-		pipe_ctx->stream_res.opp->funcs->
-			opp_program_bit_depth_reduction(pipe_ctx->stream_res.opp, &params);
+		opp->funcs->opp_program_bit_depth_reduction(opp, &params);
 		if (pipe_ctx->stream_res.tg->funcs->set_test_pattern)
 			pipe_ctx->stream_res.tg->funcs->set_test_pattern(pipe_ctx->stream_res.tg,
 				CONTROLLER_DP_TEST_PATTERN_VIDEOMODE,
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c
index f7f7515f65f4..b0dea759cd86 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c
@@ -58,6 +58,8 @@ void dp_enable_link_phy(
 	const struct dc_link_settings *link_settings)
 {
 	struct link_encoder *link_enc = link->link_enc;
+	struct dc  *core_dc = link->ctx->dc;
+	struct dmcu *dmcu = core_dc->res_pool->dmcu;
 
 	struct pipe_ctx *pipes =
 			link->dc->current_state->res_ctx.pipe_ctx;
@@ -84,6 +86,9 @@ void dp_enable_link_phy(
 		}
 	}
 
+	if (dmcu != NULL && dmcu->funcs->lock_phy)
+		dmcu->funcs->lock_phy(dmcu);
+
 	if (dc_is_dp_sst_signal(signal)) {
 		link_enc->funcs->enable_dp_output(
 						link_enc,
@@ -95,6 +100,10 @@ void dp_enable_link_phy(
 						link_settings,
 						clock_source);
 	}
+
+	if (dmcu != NULL && dmcu->funcs->unlock_phy)
+		dmcu->funcs->unlock_phy(dmcu);
+
 	link->cur_link_settings = *link_settings;
 
 	dp_receiver_power_ctrl(link, true);
@@ -150,15 +159,25 @@ bool edp_receiver_ready_T7(struct dc_link *link)
 
 void dp_disable_link_phy(struct dc_link *link, enum signal_type signal)
 {
+	struct dc  *core_dc = link->ctx->dc;
+	struct dmcu *dmcu = core_dc->res_pool->dmcu;
+
 	if (!link->wa_flags.dp_keep_receiver_powered)
 		dp_receiver_power_ctrl(link, false);
 
 	if (signal == SIGNAL_TYPE_EDP) {
 		link->link_enc->funcs->disable_output(link->link_enc, signal);
 		link->dc->hwss.edp_power_control(link, false);
-	} else
+	} else {
+		if (dmcu != NULL && dmcu->funcs->lock_phy)
+			dmcu->funcs->lock_phy(dmcu);
+
 		link->link_enc->funcs->disable_output(link->link_enc, signal);
 
+		if (dmcu != NULL && dmcu->funcs->unlock_phy)
+			dmcu->funcs->unlock_phy(dmcu);
+	}
+
 	/* Clear current link setting.*/
 	memset(&link->cur_link_settings, 0,
 			sizeof(link->cur_link_settings));
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
index 349ab8017776..eac7186e4f08 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
@@ -31,6 +31,8 @@
 #include "opp.h"
 #include "timing_generator.h"
 #include "transform.h"
+#include "dccg.h"
+#include "dchubbub.h"
 #include "dpp.h"
 #include "core_types.h"
 #include "set_mode_types.h"
@@ -104,44 +106,43 @@ enum dce_version resource_parse_asic_id(struct hw_asic_id asic_id)
 	return dc_version;
 }
 
-struct resource_pool *dc_create_resource_pool(
-				struct dc  *dc,
-				int num_virtual_links,
-				enum dce_version dc_version,
-				struct hw_asic_id asic_id)
+struct resource_pool *dc_create_resource_pool(struct dc  *dc,
+					      const struct dc_init_data *init_data,
+					      enum dce_version dc_version)
 {
 	struct resource_pool *res_pool = NULL;
 
 	switch (dc_version) {
 	case DCE_VERSION_8_0:
 		res_pool = dce80_create_resource_pool(
-			num_virtual_links, dc);
+				init_data->num_virtual_links, dc);
 		break;
 	case DCE_VERSION_8_1:
 		res_pool = dce81_create_resource_pool(
-			num_virtual_links, dc);
+				init_data->num_virtual_links, dc);
 		break;
 	case DCE_VERSION_8_3:
 		res_pool = dce83_create_resource_pool(
-			num_virtual_links, dc);
+				init_data->num_virtual_links, dc);
 		break;
 	case DCE_VERSION_10_0:
 		res_pool = dce100_create_resource_pool(
-				num_virtual_links, dc);
+				init_data->num_virtual_links, dc);
 		break;
 	case DCE_VERSION_11_0:
 		res_pool = dce110_create_resource_pool(
-			num_virtual_links, dc, asic_id);
+				init_data->num_virtual_links, dc,
+				init_data->asic_id);
 		break;
 	case DCE_VERSION_11_2:
 	case DCE_VERSION_11_22:
 		res_pool = dce112_create_resource_pool(
-			num_virtual_links, dc);
+				init_data->num_virtual_links, dc);
 		break;
 	case DCE_VERSION_12_0:
 	case DCE_VERSION_12_1:
 		res_pool = dce120_create_resource_pool(
-			num_virtual_links, dc);
+				init_data->num_virtual_links, dc);
 		break;
 
 #if defined(CONFIG_DRM_AMD_DC_DCN1_0)
@@ -149,8 +150,7 @@ struct resource_pool *dc_create_resource_pool(
 #if defined(CONFIG_DRM_AMD_DC_DCN1_01)
 	case DCN_VERSION_1_01:
 #endif
-		res_pool = dcn10_create_resource_pool(
-				num_virtual_links, dc);
+		res_pool = dcn10_create_resource_pool(init_data, dc);
 		break;
 #endif
 
@@ -163,7 +163,28 @@ struct resource_pool *dc_create_resource_pool(
 
 		if (dc->ctx->dc_bios->funcs->get_firmware_info(
 				dc->ctx->dc_bios, &fw_info) == BP_RESULT_OK) {
-				res_pool->ref_clock_inKhz = fw_info.pll_info.crystal_frequency;
+				res_pool->ref_clocks.xtalin_clock_inKhz = fw_info.pll_info.crystal_frequency;
+
+				if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
+					// On FPGA these dividers are currently not configured by GDB
+					res_pool->ref_clocks.dccg_ref_clock_inKhz = res_pool->ref_clocks.xtalin_clock_inKhz;
+					res_pool->ref_clocks.dchub_ref_clock_inKhz = res_pool->ref_clocks.xtalin_clock_inKhz;
+				} else if (res_pool->dccg && res_pool->hubbub) {
+					// If DCCG reference frequency cannot be determined (usually means not set to xtalin) then this is a critical error
+					// as this value must be known for DCHUB programming
+					(res_pool->dccg->funcs->get_dccg_ref_freq)(res_pool->dccg,
+							fw_info.pll_info.crystal_frequency,
+							&res_pool->ref_clocks.dccg_ref_clock_inKhz);
+
+					// Similarly, if DCHUB reference frequency cannot be determined, then it is also a critical error
+					(res_pool->hubbub->funcs->get_dchub_ref_freq)(res_pool->hubbub,
+							res_pool->ref_clocks.dccg_ref_clock_inKhz,
+							&res_pool->ref_clocks.dchub_ref_clock_inKhz);
+				} else {
+					// Not all ASICs have DCCG sw component
+					res_pool->ref_clocks.dccg_ref_clock_inKhz = res_pool->ref_clocks.xtalin_clock_inKhz;
+					res_pool->ref_clocks.dchub_ref_clock_inKhz = res_pool->ref_clocks.xtalin_clock_inKhz;
+				}
 			} else
 				ASSERT_CRITICAL(false);
 	}
@@ -260,6 +281,7 @@ bool resource_construct(
 			pool->stream_enc_count++;
 		}
 	}
+
 	dc->caps.dynamic_audio = false;
 	if (pool->audio_count < pool->stream_enc_count) {
 		dc->caps.dynamic_audio = true;
@@ -1014,24 +1036,60 @@ enum dc_status resource_build_scaling_params_for_context(
 
 struct pipe_ctx *find_idle_secondary_pipe(
 		struct resource_context *res_ctx,
-		const struct resource_pool *pool)
+		const struct resource_pool *pool,
+		const struct pipe_ctx *primary_pipe)
 {
 	int i;
 	struct pipe_ctx *secondary_pipe = NULL;
 
 	/*
-	 * search backwards for the second pipe to keep pipe
-	 * assignment more consistent
+	 * We add a preferred pipe mapping to avoid the chance that
+	 * MPCCs already in use will need to be reassigned to other trees.
+	 * For example, if we went with the strict, assign backwards logic:
+	 *
+	 * (State 1)
+	 * Display A on, no surface, top pipe = 0
+	 * Display B on, no surface, top pipe = 1
+	 *
+	 * (State 2)
+	 * Display A on, no surface, top pipe = 0
+	 * Display B on, surface enable, top pipe = 1, bottom pipe = 5
+	 *
+	 * (State 3)
+	 * Display A on, surface enable, top pipe = 0, bottom pipe = 5
+	 * Display B on, surface enable, top pipe = 1, bottom pipe = 4
+	 *
+	 * The state 2->3 transition requires remapping MPCC 5 from display B
+	 * to display A.
+	 *
+	 * However, with the preferred pipe logic, state 2 would look like:
+	 *
+	 * (State 2)
+	 * Display A on, no surface, top pipe = 0
+	 * Display B on, surface enable, top pipe = 1, bottom pipe = 4
+	 *
+	 * This would then cause 2->3 to not require remapping any MPCCs.
 	 */
-
-	for (i = pool->pipe_count - 1; i >= 0; i--) {
-		if (res_ctx->pipe_ctx[i].stream == NULL) {
-			secondary_pipe = &res_ctx->pipe_ctx[i];
-			secondary_pipe->pipe_idx = i;
-			break;
+	if (primary_pipe) {
+		int preferred_pipe_idx = (pool->pipe_count - 1) - primary_pipe->pipe_idx;
+		if (res_ctx->pipe_ctx[preferred_pipe_idx].stream == NULL) {
+			secondary_pipe = &res_ctx->pipe_ctx[preferred_pipe_idx];
+			secondary_pipe->pipe_idx = preferred_pipe_idx;
 		}
 	}
 
+	/*
+	 * search backwards for the second pipe to keep pipe
+	 * assignment more consistent
+	 */
+	if (!secondary_pipe)
+		for (i = pool->pipe_count - 1; i >= 0; i--) {
+			if (res_ctx->pipe_ctx[i].stream == NULL) {
+				secondary_pipe = &res_ctx->pipe_ctx[i];
+				secondary_pipe->pipe_idx = i;
+				break;
+			}
+		}
 
 	return secondary_pipe;
 }
@@ -1214,6 +1272,9 @@ bool dc_add_plane_to_context(
 		free_pipe->clock_source = tail_pipe->clock_source;
 		free_pipe->top_pipe = tail_pipe;
 		tail_pipe->bottom_pipe = free_pipe;
+	} else if (free_pipe->bottom_pipe && free_pipe->bottom_pipe->plane_state == NULL) {
+		ASSERT(free_pipe->bottom_pipe->stream_res.opp != free_pipe->stream_res.opp);
+		free_pipe->bottom_pipe->plane_state = plane_state;
 	}
 
 	/* assign new surfaces*/
@@ -1224,6 +1285,35 @@ bool dc_add_plane_to_context(
 	return true;
 }
 
+struct pipe_ctx *dc_res_get_odm_bottom_pipe(struct pipe_ctx *pipe_ctx)
+{
+	struct pipe_ctx *bottom_pipe = pipe_ctx->bottom_pipe;
+
+	/* ODM should only be updated once per otg */
+	if (pipe_ctx->top_pipe)
+		return NULL;
+
+	while (bottom_pipe) {
+		if (bottom_pipe->stream_res.opp != pipe_ctx->stream_res.opp)
+			break;
+		bottom_pipe = bottom_pipe->bottom_pipe;
+	}
+
+	return bottom_pipe;
+}
+
+bool dc_res_is_odm_head_pipe(struct pipe_ctx *pipe_ctx)
+{
+	struct pipe_ctx *top_pipe = pipe_ctx->top_pipe;
+
+	if (!top_pipe)
+		return false;
+	if (top_pipe && top_pipe->stream_res.opp == pipe_ctx->stream_res.opp)
+		return false;
+
+	return true;
+}
+
 bool dc_remove_plane_from_context(
 		const struct dc *dc,
 		struct dc_stream_state *stream,
@@ -1247,10 +1337,14 @@ bool dc_remove_plane_from_context(
 
 	/* release pipe for plane*/
 	for (i = pool->pipe_count - 1; i >= 0; i--) {
-		struct pipe_ctx *pipe_ctx;
+		struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
 
-		if (context->res_ctx.pipe_ctx[i].plane_state == plane_state) {
-			pipe_ctx = &context->res_ctx.pipe_ctx[i];
+		if (pipe_ctx->plane_state == plane_state) {
+			if (dc_res_is_odm_head_pipe(pipe_ctx)) {
+				pipe_ctx->plane_state = NULL;
+				pipe_ctx->bottom_pipe = NULL;
+				continue;
+			}
 
 			if (pipe_ctx->top_pipe)
 				pipe_ctx->top_pipe->bottom_pipe = pipe_ctx->bottom_pipe;
@@ -1268,8 +1362,9 @@ bool dc_remove_plane_from_context(
 			 */
 			if (!pipe_ctx->top_pipe) {
 				pipe_ctx->plane_state = NULL;
-				pipe_ctx->bottom_pipe = NULL;
-			} else  {
+				if (!dc_res_get_odm_bottom_pipe(pipe_ctx))
+					pipe_ctx->bottom_pipe = NULL;
+			} else {
 				memset(pipe_ctx, 0, sizeof(*pipe_ctx));
 			}
 		}
@@ -1674,6 +1769,9 @@ enum dc_status dc_remove_stream_from_ctx(
 	for (i = 0; i < MAX_PIPES; i++) {
 		if (new_ctx->res_ctx.pipe_ctx[i].stream == stream &&
 				!new_ctx->res_ctx.pipe_ctx[i].top_pipe) {
+			struct pipe_ctx *odm_pipe =
+					dc_res_get_odm_bottom_pipe(&new_ctx->res_ctx.pipe_ctx[i]);
+
 			del_pipe = &new_ctx->res_ctx.pipe_ctx[i];
 
 			ASSERT(del_pipe->stream_res.stream_enc);
@@ -1698,6 +1796,8 @@ enum dc_status dc_remove_stream_from_ctx(
 				dc->res_pool->funcs->remove_stream_from_ctx(dc, new_ctx, stream);
 
 			memset(del_pipe, 0, sizeof(*del_pipe));
+			if (odm_pipe)
+				memset(odm_pipe, 0, sizeof(*odm_pipe));
 
 			break;
 		}
@@ -1855,6 +1955,7 @@ enum dc_status resource_map_pool_resources(
 	struct dc_context *dc_ctx = dc->ctx;
 	struct pipe_ctx *pipe_ctx = NULL;
 	int pipe_idx = -1;
+	struct dc_bios *dcb = dc->ctx->dc_bios;
 
 	/* TODO Check if this is needed */
 	/*if (!resource_is_stream_unchanged(old_context, stream)) {
@@ -1869,6 +1970,13 @@ enum dc_status resource_map_pool_resources(
 
 	calculate_phy_pix_clks(stream);
 
+	/* TODO: Check Linux */
+	if (dc->config.allow_seamless_boot_optimization &&
+			!dcb->funcs->is_accelerated_mode(dcb)) {
+		if (dc_validate_seamless_boot_timing(dc, stream->sink, &stream->timing))
+			stream->apply_seamless_boot_optimization = true;
+	}
+
 	if (stream->apply_seamless_boot_optimization)
 		pipe_idx = acquire_resource_from_hw_enabled_state(
 				&context->res_ctx,
@@ -1951,7 +2059,7 @@ void dc_resource_state_construct(
 		const struct dc *dc,
 		struct dc_state *dst_ctx)
 {
-	dst_ctx->dccg = dc->res_pool->clk_mgr;
+	dst_ctx->clk_mgr = dc->res_pool->clk_mgr;
 }
 
 /**
@@ -1959,12 +2067,14 @@ void dc_resource_state_construct(
  * Checks HW resource availability and bandwidth requirement.
  * @dc: dc struct for this driver
  * @new_ctx: state to be validated
+ * @fast_validate: set to true if only yes/no to support matters
  *
  * Return: DC_OK if the result can be programmed.  Otherwise, an error code.
  */
 enum dc_status dc_validate_global_state(
 		struct dc *dc,
-		struct dc_state *new_ctx)
+		struct dc_state *new_ctx,
+		bool fast_validate)
 {
 	enum dc_status result = DC_ERROR_UNEXPECTED;
 	int i, j;
@@ -2019,7 +2129,7 @@ enum dc_status dc_validate_global_state(
 	result = resource_build_scaling_params_for_context(dc, new_ctx);
 
 	if (result == DC_OK)
-		if (!dc->res_pool->funcs->validate_bandwidth(dc, new_ctx))
+		if (!dc->res_pool->funcs->validate_bandwidth(dc, new_ctx, fast_validate))
 			result = DC_FAIL_BANDWIDTH_VALIDATE;
 
 	return result;
@@ -2315,6 +2425,21 @@ static void set_spd_info_packet(
 	*info_packet = stream->vrr_infopacket;
 }
 
+static void set_dp_sdp_info_packet(
+		struct dc_info_packet *info_packet,
+		struct dc_stream_state *stream)
+{
+	/* SPD info packet for custom sdp message */
+
+	/* Return if false. If true,
+	 * set the corresponding bit in the info packet
+	 */
+	if (!stream->dpsdp_infopacket.valid)
+		return;
+
+	*info_packet = stream->dpsdp_infopacket;
+}
+
 static void set_hdr_static_info_packet(
 		struct dc_info_packet *info_packet,
 		struct dc_stream_state *stream)
@@ -2411,6 +2536,7 @@ void resource_build_info_frame(struct pipe_ctx *pipe_ctx)
 	info->spd.valid = false;
 	info->hdrsmd.valid = false;
 	info->vsc.valid = false;
+	info->dpsdp.valid = false;
 
 	signal = pipe_ctx->stream->signal;
 
@@ -2430,6 +2556,8 @@ void resource_build_info_frame(struct pipe_ctx *pipe_ctx)
 		set_spd_info_packet(&info->spd, pipe_ctx->stream);
 
 		set_hdr_static_info_packet(&info->hdrsmd, pipe_ctx->stream);
+
+		set_dp_sdp_info_packet(&info->dpsdp, pipe_ctx->stream);
 	}
 
 	patch_gamut_packet_checksum(&info->gamut);
@@ -2657,10 +2785,11 @@ enum dc_status dc_validate_stream(struct dc *dc, struct dc_stream_state *stream)
 	if (!tg->funcs->validate_timing(tg, &stream->timing))
 		res = DC_FAIL_CONTROLLER_VALIDATE;
 
-	if (res == DC_OK)
+	if (res == DC_OK) {
 		if (!link->link_enc->funcs->validate_output_with_stream(
 						link->link_enc, stream))
 			res = DC_FAIL_ENC_VALIDATE;
+	}
 
 	/* TODO: validate audio ASIC caps, encoder */
 
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
index 996298c35f42..96e97d25d639 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
@@ -29,6 +29,9 @@
 #include "resource.h"
 #include "ipp.h"
 #include "timing_generator.h"
+#if defined(CONFIG_DRM_AMD_DC_DCN1_0)
+#include "dcn10/dcn10_hw_sequencer.h"
+#endif
 
 #define DC_LOGGER dc->ctx->logger
 
@@ -160,6 +163,27 @@ struct dc_stream_state *dc_create_stream_for_sink(
 	return stream;
 }
 
+struct dc_stream_state *dc_copy_stream(const struct dc_stream_state *stream)
+{
+	struct dc_stream_state *new_stream;
+
+	new_stream = kzalloc(sizeof(struct dc_stream_state), GFP_KERNEL);
+	if (!new_stream)
+		return NULL;
+
+	memcpy(new_stream, stream, sizeof(struct dc_stream_state));
+
+	if (new_stream->sink)
+		dc_sink_retain(new_stream->sink);
+
+	if (new_stream->out_transfer_func)
+		dc_transfer_func_retain(new_stream->out_transfer_func);
+
+	kref_init(&new_stream->refcount);
+
+	return new_stream;
+}
+
 /**
  * dc_stream_get_status_from_state - Get stream status from given dc state
  * @state: DC state to find the stream status in
@@ -196,6 +220,35 @@ struct dc_stream_status *dc_stream_get_status(
 	return dc_stream_get_status_from_state(dc->current_state, stream);
 }
 
+static void delay_cursor_until_vupdate(struct pipe_ctx *pipe_ctx, struct dc *dc)
+{
+#if defined(CONFIG_DRM_AMD_DC_DCN1_0)
+	unsigned int vupdate_line;
+	unsigned int lines_to_vupdate, us_to_vupdate, vpos, nvpos;
+	struct dc_stream_state *stream = pipe_ctx->stream;
+	unsigned int us_per_line;
+
+	if (stream->ctx->asic_id.chip_family == FAMILY_RV &&
+			ASIC_REV_IS_RAVEN(stream->ctx->asic_id.hw_internal_rev)) {
+
+		vupdate_line = get_vupdate_offset_from_vsync(pipe_ctx);
+		if (!dc_stream_get_crtc_position(dc, &stream, 1, &vpos, &nvpos))
+			return;
+
+		if (vpos >= vupdate_line)
+			return;
+
+		us_per_line = stream->timing.h_total * 10000 / stream->timing.pix_clk_100hz;
+		lines_to_vupdate = vupdate_line - vpos;
+		us_to_vupdate = lines_to_vupdate * us_per_line;
+
+		/* 70 us is a conservative estimate of cursor update time*/
+		if (us_to_vupdate < 70)
+			udelay(us_to_vupdate);
+	}
+#endif
+}
+
 /**
  * dc_stream_set_cursor_attributes() - Update cursor attributes and set cursor surface address
  */
@@ -234,6 +287,8 @@ bool dc_stream_set_cursor_attributes(
 
 		if (!pipe_to_program) {
 			pipe_to_program = pipe_ctx;
+
+			delay_cursor_until_vupdate(pipe_ctx, core_dc);
 			core_dc->hwss.pipe_control_lock(core_dc, pipe_to_program, true);
 		}
 
@@ -278,11 +333,13 @@ bool dc_stream_set_cursor_position(
 				(!pipe_ctx->plane_res.mi  && !pipe_ctx->plane_res.hubp) ||
 				!pipe_ctx->plane_state ||
 				(!pipe_ctx->plane_res.xfm && !pipe_ctx->plane_res.dpp) ||
-				!pipe_ctx->plane_res.ipp)
+				(!pipe_ctx->plane_res.ipp && !pipe_ctx->plane_res.dpp))
 			continue;
 
 		if (!pipe_to_program) {
 			pipe_to_program = pipe_ctx;
+
+			delay_cursor_until_vupdate(pipe_ctx, core_dc);
 			core_dc->hwss.pipe_control_lock(core_dc, pipe_to_program, true);
 		}
 
@@ -314,6 +371,68 @@ uint32_t dc_stream_get_vblank_counter(const struct dc_stream_state *stream)
 	return 0;
 }
 
+static void build_dp_sdp_info_frame(struct pipe_ctx *pipe_ctx,
+		const uint8_t  *custom_sdp_message,
+		unsigned int sdp_message_size)
+{
+	uint8_t i;
+	struct encoder_info_frame *info = &pipe_ctx->stream_res.encoder_info_frame;
+
+	/* set valid info */
+	info->dpsdp.valid = true;
+
+	/* set sdp message header */
+	info->dpsdp.hb0 = custom_sdp_message[0]; /* package id */
+	info->dpsdp.hb1 = custom_sdp_message[1]; /* package type */
+	info->dpsdp.hb2 = custom_sdp_message[2]; /* package specific byte 0 any data */
+	info->dpsdp.hb3 = custom_sdp_message[3]; /* package specific byte 0 any data */
+
+	/* set sdp message data */
+	for (i = 0; i < 32; i++)
+		info->dpsdp.sb[i] = (custom_sdp_message[i+4]);
+
+}
+
+static void invalid_dp_sdp_info_frame(struct pipe_ctx *pipe_ctx)
+{
+	struct encoder_info_frame *info = &pipe_ctx->stream_res.encoder_info_frame;
+
+	/* in-valid info */
+	info->dpsdp.valid = false;
+}
+
+bool dc_stream_send_dp_sdp(const struct dc_stream_state *stream,
+		const uint8_t *custom_sdp_message,
+		unsigned int sdp_message_size)
+{
+	int i;
+	struct dc  *core_dc;
+	struct resource_context *res_ctx;
+
+	if (stream == NULL) {
+		dm_error("DC: dc_stream is NULL!\n");
+		return false;
+	}
+
+	core_dc = stream->ctx->dc;
+	res_ctx = &core_dc->current_state->res_ctx;
+
+	for (i = 0; i < MAX_PIPES; i++) {
+		struct pipe_ctx *pipe_ctx = &res_ctx->pipe_ctx[i];
+
+		if (pipe_ctx->stream != stream)
+			continue;
+
+		build_dp_sdp_info_frame(pipe_ctx, custom_sdp_message, sdp_message_size);
+
+		core_dc->hwss.update_info_frame(pipe_ctx);
+
+		invalid_dp_sdp_info_frame(pipe_ctx);
+	}
+
+	return true;
+}
+
 bool dc_stream_get_scanoutpos(const struct dc_stream_state *stream,
 				  uint32_t *v_blank_start,
 				  uint32_t *v_blank_end,
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_surface.c b/drivers/gpu/drm/amd/display/dc/core/dc_surface.c
index ee6bd50f60b8..a5e86f9b148f 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_surface.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_surface.c
@@ -119,6 +119,19 @@ const struct dc_plane_status *dc_plane_get_status(
 	if (core_dc->current_state == NULL)
 		return NULL;
 
+	/* Find the current plane state and set its pending bit to false */
+	for (i = 0; i < core_dc->res_pool->pipe_count; i++) {
+		struct pipe_ctx *pipe_ctx =
+				&core_dc->current_state->res_ctx.pipe_ctx[i];
+
+		if (pipe_ctx->plane_state != plane_state)
+			continue;
+
+		pipe_ctx->plane_state->status.is_flip_pending = false;
+
+		break;
+	}
+
 	for (i = 0; i < core_dc->res_pool->pipe_count; i++) {
 		struct pipe_ctx *pipe_ctx =
 				&core_dc->current_state->res_ctx.pipe_ctx[i];
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 0515095574e7..44e4b0465587 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -39,9 +39,10 @@
 #include "inc/hw/dmcu.h"
 #include "dml/display_mode_lib.h"
 
-#define DC_VER "3.2.17"
+#define DC_VER "3.2.27"
 
 #define MAX_SURFACES 3
+#define MAX_PLANES 6
 #define MAX_STREAMS 6
 #define MAX_SINKS_PER_LINK 4
 
@@ -53,6 +54,41 @@ struct dc_versions {
 	struct dmcu_version dmcu_version;
 };
 
+enum dc_plane_type {
+	DC_PLANE_TYPE_INVALID,
+	DC_PLANE_TYPE_DCE_RGB,
+	DC_PLANE_TYPE_DCE_UNDERLAY,
+	DC_PLANE_TYPE_DCN_UNIVERSAL,
+};
+
+struct dc_plane_cap {
+	enum dc_plane_type type;
+	uint32_t blends_with_above : 1;
+	uint32_t blends_with_below : 1;
+	uint32_t per_pixel_alpha : 1;
+	struct {
+		uint32_t argb8888 : 1;
+		uint32_t nv12 : 1;
+		uint32_t fp16 : 1;
+	} pixel_format_support;
+	// max upscaling factor x1000
+	// upscaling factors are always >= 1
+	// for example, 1080p -> 8K is 4.0, or 4000 raw value
+	struct {
+		uint32_t argb8888;
+		uint32_t nv12;
+		uint32_t fp16;
+	} max_upscale_factor;
+	// max downscale factor x1000
+	// downscale factors are always <= 1
+	// for example, 8K -> 1080p is 0.25, or 250 raw value
+	struct {
+		uint32_t argb8888;
+		uint32_t nv12;
+		uint32_t fp16;
+	} max_downscale_factor;
+};
+
 struct dc_caps {
 	uint32_t max_streams;
 	uint32_t max_links;
@@ -73,6 +109,7 @@ struct dc_caps {
 	bool force_dp_tps4_for_cp2520;
 	bool disable_dp_clk_share;
 	bool psp_setup_panel_mode;
+	struct dc_plane_cap planes[MAX_PLANES];
 };
 
 struct dc_dcc_surface_param {
@@ -164,6 +201,10 @@ struct dc_config {
 	bool gpu_vm_support;
 	bool disable_disp_pll_sharing;
 	bool fbc_support;
+	bool optimize_edp_link_rate;
+	bool disable_fractional_pwm;
+	bool allow_seamless_boot_optimization;
+	bool power_down_display_on_boot;
 };
 
 enum visual_confirm {
@@ -203,7 +244,59 @@ struct dc_clocks {
 	int fclk_khz;
 	int phyclk_khz;
 	int dramclk_khz;
-};
+	bool p_state_change_support;
+};
+
+struct dc_bw_validation_profile {
+	bool enable;
+
+	unsigned long long total_ticks;
+	unsigned long long voltage_level_ticks;
+	unsigned long long watermark_ticks;
+	unsigned long long rq_dlg_ticks;
+
+	unsigned long long total_count;
+	unsigned long long skip_fast_count;
+	unsigned long long skip_pass_count;
+	unsigned long long skip_fail_count;
+};
+
+#define BW_VAL_TRACE_SETUP() \
+		unsigned long long end_tick = 0; \
+		unsigned long long voltage_level_tick = 0; \
+		unsigned long long watermark_tick = 0; \
+		unsigned long long start_tick = dc->debug.bw_val_profile.enable ? \
+				dm_get_timestamp(dc->ctx) : 0
+
+#define BW_VAL_TRACE_COUNT() \
+		if (dc->debug.bw_val_profile.enable) \
+			dc->debug.bw_val_profile.total_count++
+
+#define BW_VAL_TRACE_SKIP(status) \
+		if (dc->debug.bw_val_profile.enable) { \
+			if (!voltage_level_tick) \
+				voltage_level_tick = dm_get_timestamp(dc->ctx); \
+			dc->debug.bw_val_profile.skip_ ## status ## _count++; \
+		}
+
+#define BW_VAL_TRACE_END_VOLTAGE_LEVEL() \
+		if (dc->debug.bw_val_profile.enable) \
+			voltage_level_tick = dm_get_timestamp(dc->ctx)
+
+#define BW_VAL_TRACE_END_WATERMARKS() \
+		if (dc->debug.bw_val_profile.enable) \
+			watermark_tick = dm_get_timestamp(dc->ctx)
+
+#define BW_VAL_TRACE_FINISH() \
+		if (dc->debug.bw_val_profile.enable) { \
+			end_tick = dm_get_timestamp(dc->ctx); \
+			dc->debug.bw_val_profile.total_ticks += end_tick - start_tick; \
+			dc->debug.bw_val_profile.voltage_level_ticks += voltage_level_tick - start_tick; \
+			if (watermark_tick) { \
+				dc->debug.bw_val_profile.watermark_ticks += watermark_tick - voltage_level_tick; \
+				dc->debug.bw_val_profile.rq_dlg_ticks += end_tick - watermark_tick; \
+			} \
+		}
 
 struct dc_debug_options {
 	enum visual_confirm visual_confirm;
@@ -257,6 +350,8 @@ struct dc_debug_options {
 	bool skip_detection_link_training;
 	unsigned int force_odm_combine; //bit vector based on otg inst
 	unsigned int force_fclk_khz;
+	bool disable_tri_buf;
+	struct dc_bw_validation_profile bw_val_profile;
 };
 
 struct dc_debug_data {
@@ -265,6 +360,14 @@ struct dc_debug_data {
 	uint32_t auxErrorCount;
 };
 
+struct dc_bounding_box_overrides {
+	int sr_exit_time_ns;
+	int sr_enter_plus_exit_time_ns;
+	int urgent_latency_ns;
+	int percent_of_ideal_drambw;
+	int dram_clock_change_latency_ns;
+};
+
 struct dc_state;
 struct resource_pool;
 struct dce_hwseq;
@@ -274,6 +377,7 @@ struct dc {
 	struct dc_cap_funcs cap_funcs;
 	struct dc_config config;
 	struct dc_debug_options debug;
+	struct dc_bounding_box_overrides bb_overrides;
 	struct dc_context *ctx;
 
 	uint8_t link_count;
@@ -298,8 +402,12 @@ struct dc {
 	struct hw_sequencer_funcs hwss;
 	struct dce_hwseq *hwseq;
 
+	/* Require to optimize clocks and bandwidth for added/removed planes */
 	bool optimized_required;
 
+	/* Require to maintain clocks and bandwidth for UEFI enabled HW */
+	bool optimize_seamless_boot;
+
 	/* FBC compressor */
 	struct compressor *fbc_compressor;
 
@@ -327,6 +435,7 @@ struct dc_init_data {
 	struct hw_asic_id asic_id;
 	void *driver; /* ctx */
 	struct cgs_device *cgs_device;
+	struct dc_bounding_box_overrides bb_overrides;
 
 	int num_virtual_links;
 	/*
@@ -597,7 +706,7 @@ struct dc_validation_set {
 	uint8_t plane_count;
 };
 
-bool dc_validate_seamless_boot_timing(struct dc *dc,
+bool dc_validate_seamless_boot_timing(const struct dc *dc,
 				const struct dc_sink *sink,
 				struct dc_crtc_timing *crtc_timing);
 
@@ -605,9 +714,14 @@ enum dc_status dc_validate_plane(struct dc *dc, const struct dc_plane_state *pla
 
 void get_clock_requirements_for_state(struct dc_state *state, struct AsicStateEx *info);
 
+/*
+ * fast_validate: we return after determining if we can support the new state,
+ * but before we populate the programming info
+ */
 enum dc_status dc_validate_global_state(
 		struct dc *dc,
-		struct dc_state *new_ctx);
+		struct dc_state *new_ctx,
+		bool fast_validate);
 
 
 void dc_resource_state_construct(
@@ -636,7 +750,8 @@ void dc_resource_state_destruct(struct dc_state *context);
 bool dc_commit_state(struct dc *dc, struct dc_state *context);
 
 
-struct dc_state *dc_create_state(void);
+struct dc_state *dc_create_state(struct dc *dc);
+struct dc_state *dc_copy_state(struct dc_state *src_ctx);
 void dc_retain_state(struct dc_state *context);
 void dc_release_state(struct dc_state *context);
 
@@ -648,9 +763,16 @@ struct dpcd_caps {
 	union dpcd_rev dpcd_rev;
 	union max_lane_count max_ln_count;
 	union max_down_spread max_down_spread;
+	union dprx_feature dprx_feature;
+
+	/* valid only for eDP v1.4 or higher*/
+	uint8_t edp_supported_link_rates_count;
+	enum dc_link_rate edp_supported_link_rates[8];
 
 	/* dongle type (DP converter, CV smart dongle) */
 	enum display_dongle_type dongle_type;
+	/* branch device or sink device */
+	bool is_branch_dev;
 	/* Dongle's downstream count. */
 	union sink_count sink_count;
 	/* If dongle_type == DISPLAY_DONGLE_DP_HDMI_CONVERTER,
@@ -666,11 +788,11 @@ struct dpcd_caps {
 	int8_t branch_dev_name[6];
 	int8_t branch_hw_revision;
 	int8_t branch_fw_revision[2];
-	uint8_t link_rate_set;
 
 	bool allow_invalid_MSA_timing_param;
 	bool panel_mode_edp;
 	bool dpcd_display_control_capable;
+	bool ext_receiver_cap_field_present;
 };
 
 #include "dc_link.h"
diff --git a/drivers/gpu/drm/amd/display/dc/dc_ddc_types.h b/drivers/gpu/drm/amd/display/dc/dc_ddc_types.h
index 05c8c31d8b31..4ef97f65e55d 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_ddc_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_ddc_types.h
@@ -68,6 +68,8 @@ enum aux_transaction_reply {
 	AUX_TRANSACTION_REPLY_AUX_ACK = 0x00,
 	AUX_TRANSACTION_REPLY_AUX_NACK = 0x01,
 	AUX_TRANSACTION_REPLY_AUX_DEFER = 0x02,
+	AUX_TRANSACTION_REPLY_I2C_OVER_AUX_NACK = 0x04,
+	AUX_TRANSACTION_REPLY_I2C_OVER_AUX_DEFER = 0x08,
 
 	AUX_TRANSACTION_REPLY_I2C_ACK = 0x00,
 	AUX_TRANSACTION_REPLY_I2C_NACK = 0x10,
diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
index d4eab33c453b..11c68a399267 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
@@ -94,6 +94,8 @@ struct dc_link_settings {
 	enum dc_lane_count lane_count;
 	enum dc_link_rate link_rate;
 	enum dc_link_spread link_spread;
+	bool use_link_rate_set;
+	uint8_t link_rate_set;
 };
 
 struct dc_lane_settings {
@@ -420,10 +422,24 @@ union edp_configuration_cap {
 	uint8_t raw;
 };
 
+union dprx_feature {
+	struct {
+		uint8_t GTC_CAP:1;                             // bit 0: DP 1.3+
+		uint8_t SST_SPLIT_SDP_CAP:1;                   // bit 1: DP 1.4
+		uint8_t AV_SYNC_CAP:1;                         // bit 2: DP 1.3+
+		uint8_t VSC_SDP_COLORIMETRY_SUPPORTED:1;       // bit 3: DP 1.3+
+		uint8_t VSC_EXT_VESA_SDP_SUPPORTED:1;          // bit 4: DP 1.4
+		uint8_t VSC_EXT_VESA_SDP_CHAINING_SUPPORTED:1; // bit 5: DP 1.4
+		uint8_t VSC_EXT_CEA_SDP_SUPPORTED:1;           // bit 6: DP 1.4
+		uint8_t VSC_EXT_CEA_SDP_CHAINING_SUPPORTED:1;  // bit 7: DP 1.4
+	} bits;
+	uint8_t raw;
+};
+
 union training_aux_rd_interval {
 	struct {
 		uint8_t TRAINIG_AUX_RD_INTERVAL:7;
-		uint8_t EXT_RECIEVER_CAP_FIELD_PRESENT:1;
+		uint8_t EXT_RECEIVER_CAP_FIELD_PRESENT:1;
 	} bits;
 	uint8_t raw;
 };
diff --git a/drivers/gpu/drm/amd/display/dc/dc_helper.c b/drivers/gpu/drm/amd/display/dc/dc_helper.c
index 597d38393379..5e6c5eff49cf 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_helper.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_helper.c
@@ -51,20 +51,16 @@ static inline void set_reg_field_value_masks(
 	field_value_mask->mask = field_value_mask->mask | mask;
 }
 
-uint32_t generic_reg_update_ex(const struct dc_context *ctx,
-		uint32_t addr, uint32_t reg_val, int n,
+static void set_reg_field_values(struct dc_reg_value_masks *field_value_mask,
+		uint32_t addr, int n,
 		uint8_t shift1, uint32_t mask1, uint32_t field_value1,
-		...)
+		va_list ap)
 {
-	struct dc_reg_value_masks field_value_mask = {0};
 	uint32_t shift, mask, field_value;
 	int i = 1;
 
-	va_list ap;
-	va_start(ap, field_value1);
-
 	/* gather all bits value/mask getting updated in this register */
-	set_reg_field_value_masks(&field_value_mask,
+	set_reg_field_value_masks(field_value_mask,
 			field_value1, mask1, shift1);
 
 	while (i < n) {
@@ -72,10 +68,48 @@ uint32_t generic_reg_update_ex(const struct dc_context *ctx,
 		mask = va_arg(ap, uint32_t);
 		field_value = va_arg(ap, uint32_t);
 
-		set_reg_field_value_masks(&field_value_mask,
+		set_reg_field_value_masks(field_value_mask,
 				field_value, mask, shift);
 		i++;
 	}
+}
+
+uint32_t generic_reg_update_ex(const struct dc_context *ctx,
+		uint32_t addr, int n,
+		uint8_t shift1, uint32_t mask1, uint32_t field_value1,
+		...)
+{
+	struct dc_reg_value_masks field_value_mask = {0};
+	uint32_t reg_val;
+	va_list ap;
+
+	va_start(ap, field_value1);
+
+	set_reg_field_values(&field_value_mask, addr, n, shift1, mask1,
+			field_value1, ap);
+
+	va_end(ap);
+
+	/* mmio write directly */
+	reg_val = dm_read_reg(ctx, addr);
+	reg_val = (reg_val & ~field_value_mask.mask) | field_value_mask.value;
+	dm_write_reg(ctx, addr, reg_val);
+	return reg_val;
+}
+
+uint32_t generic_reg_set_ex(const struct dc_context *ctx,
+		uint32_t addr, uint32_t reg_val, int n,
+		uint8_t shift1, uint32_t mask1, uint32_t field_value1,
+		...)
+{
+	struct dc_reg_value_masks field_value_mask = {0};
+	va_list ap;
+
+	va_start(ap, field_value1);
+
+	set_reg_field_values(&field_value_mask, addr, n, shift1, mask1,
+			field_value1, ap);
+
 	va_end(ap);
 
 
@@ -85,6 +119,24 @@ uint32_t generic_reg_update_ex(const struct dc_context *ctx,
 	return reg_val;
 }
 
+uint32_t dm_read_reg_func(
+	const struct dc_context *ctx,
+	uint32_t address,
+	const char *func_name)
+{
+	uint32_t value;
+#ifdef DM_CHECK_ADDR_0
+	if (address == 0) {
+		DC_ERR("invalid register read; address = 0\n");
+		return 0;
+	}
+#endif
+	value = cgs_read_register(ctx->cgs_device, address);
+	trace_amdgpu_dc_rreg(&ctx->perf_trace->read_count, address, value);
+
+	return value;
+}
+
 uint32_t generic_reg_get(const struct dc_context *ctx, uint32_t addr,
 		uint8_t shift, uint32_t mask, uint32_t *field_value)
 {
@@ -235,7 +287,7 @@ uint32_t generic_reg_get(const struct dc_context *ctx,
 }
 */
 
-uint32_t generic_reg_wait(const struct dc_context *ctx,
+void generic_reg_wait(const struct dc_context *ctx,
 	uint32_t addr, uint32_t shift, uint32_t mask, uint32_t condition_value,
 	unsigned int delay_between_poll_us, unsigned int time_out_num_tries,
 	const char *func_name, int line)
@@ -265,7 +317,7 @@ uint32_t generic_reg_wait(const struct dc_context *ctx,
 				DC_LOG_DC("REG_WAIT taking a while: %dms in %s line:%d\n",
 						delay_between_poll_us * i / 1000,
 						func_name, line);
-			return reg_val;
+			return;
 		}
 	}
 
@@ -275,8 +327,6 @@ uint32_t generic_reg_wait(const struct dc_context *ctx,
 
 	if (!IS_FPGA_MAXIMUS_DC(ctx->dce_environment))
 		BREAK_TO_DEBUGGER();
-
-	return reg_val;
 }
 
 void generic_write_indirect_reg(const struct dc_context *ctx,
diff --git a/drivers/gpu/drm/amd/display/dc/dc_link.h b/drivers/gpu/drm/amd/display/dc/dc_link.h
index 8fc223defed4..7b9429e30d82 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_link.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_link.h
@@ -120,6 +120,7 @@ struct dc_link {
 	/* MST record stream using this link */
 	struct link_flags {
 		bool dp_keep_receiver_powered;
+		bool dp_skip_DID2;
 	} wa_flags;
 	struct link_mst_stream_allocation_table mst_stream_alloc_table;
 
@@ -246,10 +247,18 @@ void dc_link_set_test_pattern(struct dc_link *link,
 			const struct link_training_settings *p_link_settings,
 			const unsigned char *p_custom_pattern,
 			unsigned int cust_pattern_size);
+uint32_t dc_link_bandwidth_kbps(
+	const struct dc_link *link,
+	const struct dc_link_settings *link_setting);
+
+const struct dc_link_settings *dc_link_get_link_cap(
+		const struct dc_link *link);
 
 bool dc_submit_i2c(
 		struct dc *dc,
 		uint32_t link_index,
 		struct i2c_command *cmd);
 
+uint32_t dc_bandwidth_in_kbps_from_timing(
+	const struct dc_crtc_timing *timing);
 #endif /* DC_LINK_H_ */
diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h
index 5657cb3a2ad3..189bdab929a5 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_stream.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h
@@ -80,6 +80,7 @@ struct dc_stream_state {
 	struct dc_info_packet vrr_infopacket;
 	struct dc_info_packet vsc_infopacket;
 	struct dc_info_packet vsp_infopacket;
+	struct dc_info_packet dpsdp_infopacket;
 
 	struct rect src; /* composition area */
 	struct rect dst; /* stream addressable area */
@@ -221,6 +222,13 @@ struct dc_stream_state *dc_get_stream_at_index(struct dc *dc, uint8_t i);
  */
 uint32_t dc_stream_get_vblank_counter(const struct dc_stream_state *stream);
 
+/*
+ * Send dp sdp message.
+ */
+bool dc_stream_send_dp_sdp(const struct dc_stream_state *stream,
+		const uint8_t *custom_sdp_message,
+		unsigned int sdp_message_size);
+
 /* TODO: Return parsed values rather than direct register read
  * This has a dependency on the caller (amdgpu_display_get_crtc_scanoutpos)
  * being refactored properly to be dce-specific
@@ -299,6 +307,8 @@ enum surface_update_type dc_check_update_surfaces_for_stream(
  */
 struct dc_stream_state *dc_create_stream_for_sink(struct dc_sink *dc_sink);
 
+struct dc_stream_state *dc_copy_stream(const struct dc_stream_state *stream);
+
 void update_stream_signal(struct dc_stream_state *stream, struct dc_sink *sink);
 
 void dc_stream_retain(struct dc_stream_state *dc_stream);
diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h
index da2009a108cf..6c2a3d9a4c2e 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_types.h
@@ -103,7 +103,7 @@ struct dc_context {
 };
 
 
-#define DC_MAX_EDID_BUFFER_SIZE 512
+#define DC_MAX_EDID_BUFFER_SIZE 1024
 #define EDID_BLOCK_SIZE 128
 #define MAX_SURFACE_NUM 4
 #define NUM_PIXEL_FORMATS 10
@@ -395,7 +395,7 @@ struct dc_dongle_caps {
 	bool is_dp_hdmi_ycbcr422_converter;
 	bool is_dp_hdmi_ycbcr420_converter;
 	uint32_t dp_hdmi_max_bpc;
-	uint32_t dp_hdmi_max_pixel_clk;
+	uint32_t dp_hdmi_max_pixel_clk_in_khz;
 };
 /* Scaling format */
 enum scaling_transformation {
@@ -550,9 +550,9 @@ struct psr_config {
 	unsigned char psr_version;
 	unsigned int psr_rfb_setup_time;
 	bool psr_exit_link_training_required;
-
 	bool psr_frame_capture_indication_req;
 	unsigned int psr_sdp_transmit_line_num_deadline;
+	bool allow_smu_optimizations;
 };
 
 union dmcu_psr_level {
@@ -654,6 +654,7 @@ struct psr_context {
 	 * continuing powerd own
 	 */
 	unsigned int frame_delay;
+	bool allow_smu_optimizations;
 };
 
 struct colorspace_transform {
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c
index 4fe3664fb495..bd33c47183fc 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c
@@ -171,24 +171,24 @@ static void submit_channel_request(
 		 (request->action == I2CAUX_TRANSACTION_ACTION_I2C_WRITE_MOT)));
 	if (REG(AUXN_IMPCAL)) {
 		/* clear_aux_error */
-		REG_UPDATE_SEQ(AUXN_IMPCAL, AUXN_CALOUT_ERROR_AK,
-				1,
-				0);
+		REG_UPDATE_SEQ_2(AUXN_IMPCAL,
+				AUXN_CALOUT_ERROR_AK, 1,
+				AUXN_CALOUT_ERROR_AK, 0);
 
-		REG_UPDATE_SEQ(AUXP_IMPCAL, AUXP_CALOUT_ERROR_AK,
-				1,
-				0);
+		REG_UPDATE_SEQ_2(AUXP_IMPCAL,
+				AUXP_CALOUT_ERROR_AK, 1,
+				AUXP_CALOUT_ERROR_AK, 0);
 
 		/* force_default_calibrate */
-		REG_UPDATE_1BY1_2(AUXN_IMPCAL,
+		REG_UPDATE_SEQ_2(AUXN_IMPCAL,
 				AUXN_IMPCAL_ENABLE, 1,
 				AUXN_IMPCAL_OVERRIDE_ENABLE, 0);
 
 		/* bug? why AUXN update EN and OVERRIDE_EN 1 by 1 while AUX P toggles OVERRIDE? */
 
-		REG_UPDATE_SEQ(AUXP_IMPCAL, AUXP_IMPCAL_OVERRIDE_ENABLE,
-				1,
-				0);
+		REG_UPDATE_SEQ_2(AUXP_IMPCAL,
+				AUXP_IMPCAL_OVERRIDE_ENABLE, 1,
+				AUXP_IMPCAL_OVERRIDE_ENABLE, 0);
 	}
 
 	REG_UPDATE(AUX_INTERRUPT_CONTROL, AUX_SW_DONE_ACK, 1);
@@ -270,7 +270,7 @@ static int read_channel_reply(struct dce_aux *engine, uint32_t size,
 	if (!bytes_replied)
 		return -1;
 
-	REG_UPDATE_1BY1_3(AUX_SW_DATA,
+	REG_UPDATE_SEQ_3(AUX_SW_DATA,
 			  AUX_SW_INDEX, 0,
 			  AUX_SW_AUTOINCREMENT_DISABLE, 1,
 			  AUX_SW_DATA_RW, 1);
@@ -320,9 +320,10 @@ static enum aux_channel_operation_result get_channel_status(
 	*returned_bytes = 0;
 
 	/* poll to make sure that SW_DONE is asserted */
-	value = REG_WAIT(AUX_SW_STATUS, AUX_SW_DONE, 1,
+	REG_WAIT(AUX_SW_STATUS, AUX_SW_DONE, 1,
 				10, aux110->timeout_period/10);
 
+	value = REG_READ(AUX_SW_STATUS);
 	/* in case HPD is LOW, exit AUX transaction */
 	if ((value & AUX_SW_STATUS__AUX_SW_HPD_DISCON_MASK))
 		return AUX_CHANNEL_OPERATION_FAILED_HPD_DISCON;
@@ -377,7 +378,6 @@ static bool acquire(
 	struct dce_aux *engine,
 	struct ddc *ddc)
 {
-
 	enum gpio_result result;
 
 	if (!is_engine_available(engine))
@@ -442,12 +442,12 @@ static enum i2caux_transaction_action i2caux_action_from_payload(struct aux_payl
 	return I2CAUX_TRANSACTION_ACTION_DP_READ;
 }
 
-int dce_aux_transfer(struct ddc_service *ddc,
-		struct aux_payload *payload)
+int dce_aux_transfer_raw(struct ddc_service *ddc,
+		struct aux_payload *payload,
+		enum aux_channel_operation_result *operation_result)
 {
 	struct ddc *ddc_pin = ddc->ddc_pin;
 	struct dce_aux *aux_engine;
-	enum aux_channel_operation_result operation_result;
 	struct aux_request_transaction_data aux_req;
 	struct aux_reply_transaction_data aux_rep;
 	uint8_t returned_bytes = 0;
@@ -458,7 +458,8 @@ int dce_aux_transfer(struct ddc_service *ddc,
 	memset(&aux_rep, 0, sizeof(aux_rep));
 
 	aux_engine = ddc->ctx->dc->res_pool->engines[ddc_pin->pin_data->en];
-	acquire(aux_engine, ddc_pin);
+	if (!acquire(aux_engine, ddc_pin))
+		return -1;
 
 	if (payload->i2c_over_aux)
 		aux_req.type = AUX_TRANSACTION_TYPE_I2C;
@@ -473,28 +474,26 @@ int dce_aux_transfer(struct ddc_service *ddc,
 	aux_req.data = payload->data;
 
 	submit_channel_request(aux_engine, &aux_req);
-	operation_result = get_channel_status(aux_engine, &returned_bytes);
-
-	switch (operation_result) {
-	case AUX_CHANNEL_OPERATION_SUCCEEDED:
-		res = read_channel_reply(aux_engine, payload->length,
-							payload->data, payload->reply,
-							&status);
-		break;
-	case AUX_CHANNEL_OPERATION_FAILED_HPD_DISCON:
-		res = 0;
-		break;
-	case AUX_CHANNEL_OPERATION_FAILED_REASON_UNKNOWN:
-	case AUX_CHANNEL_OPERATION_FAILED_INVALID_REPLY:
-	case AUX_CHANNEL_OPERATION_FAILED_TIMEOUT:
+	*operation_result = get_channel_status(aux_engine, &returned_bytes);
+
+	if (*operation_result == AUX_CHANNEL_OPERATION_SUCCEEDED) {
+		read_channel_reply(aux_engine, payload->length,
+					 payload->data, payload->reply,
+					 &status);
+		res = returned_bytes;
+	} else {
 		res = -1;
-		break;
 	}
+
 	release_engine(aux_engine);
 	return res;
 }
 
-#define AUX_RETRY_MAX 7
+#define AUX_MAX_RETRIES 7
+#define AUX_MAX_DEFER_RETRIES 7
+#define AUX_MAX_I2C_DEFER_RETRIES 7
+#define AUX_MAX_INVALID_REPLY_RETRIES 2
+#define AUX_MAX_TIMEOUT_RETRIES 3
 
 bool dce_aux_transfer_with_retries(struct ddc_service *ddc,
 		struct aux_payload *payload)
@@ -502,24 +501,85 @@ bool dce_aux_transfer_with_retries(struct ddc_service *ddc,
 	int i, ret = 0;
 	uint8_t reply;
 	bool payload_reply = true;
+	enum aux_channel_operation_result operation_result;
+	int aux_ack_retries = 0,
+		aux_defer_retries = 0,
+		aux_i2c_defer_retries = 0,
+		aux_timeout_retries = 0,
+		aux_invalid_reply_retries = 0;
 
 	if (!payload->reply) {
 		payload_reply = false;
 		payload->reply = &reply;
 	}
 
-	for (i = 0; i < AUX_RETRY_MAX; i++) {
-		ret = dce_aux_transfer(ddc, payload);
-
-		if (ret >= 0) {
-			if (*payload->reply == 0) {
-				if (!payload_reply)
-					payload->reply = NULL;
-				return true;
+	for (i = 0; i < AUX_MAX_RETRIES; i++) {
+		ret = dce_aux_transfer_raw(ddc, payload, &operation_result);
+		switch (operation_result) {
+		case AUX_CHANNEL_OPERATION_SUCCEEDED:
+			aux_timeout_retries = 0;
+			aux_invalid_reply_retries = 0;
+
+			switch (*payload->reply) {
+			case AUX_TRANSACTION_REPLY_AUX_ACK:
+				if (!payload->write && payload->length != ret) {
+					if (++aux_ack_retries >= AUX_MAX_RETRIES)
+						goto fail;
+					else
+						udelay(300);
+				} else
+					return true;
+			break;
+
+			case AUX_TRANSACTION_REPLY_AUX_DEFER:
+			case AUX_TRANSACTION_REPLY_I2C_OVER_AUX_NACK:
+			case AUX_TRANSACTION_REPLY_I2C_OVER_AUX_DEFER:
+				if (++aux_defer_retries >= AUX_MAX_DEFER_RETRIES)
+					goto fail;
+				break;
+
+			case AUX_TRANSACTION_REPLY_I2C_DEFER:
+				aux_defer_retries = 0;
+				if (++aux_i2c_defer_retries >= AUX_MAX_I2C_DEFER_RETRIES)
+					goto fail;
+				break;
+
+			case AUX_TRANSACTION_REPLY_AUX_NACK:
+			case AUX_TRANSACTION_REPLY_HPD_DISCON:
+			default:
+				goto fail;
 			}
-		}
+			break;
+
+		case AUX_CHANNEL_OPERATION_FAILED_INVALID_REPLY:
+			if (++aux_invalid_reply_retries >= AUX_MAX_INVALID_REPLY_RETRIES)
+				goto fail;
+			else
+				udelay(400);
+			break;
+
+		case AUX_CHANNEL_OPERATION_FAILED_TIMEOUT:
+			if (++aux_timeout_retries >= AUX_MAX_TIMEOUT_RETRIES)
+				goto fail;
+			else {
+				/*
+				 * DP 1.4, 2.8.2:  AUX Transaction Response/Reply Timeouts
+				 * According to the DP spec there should be 3 retries total
+				 * with a 400us wait inbetween each. Hardware already waits
+				 * for 550us therefore no wait is required here.
+				 */
+			}
+			break;
 
-		udelay(1000);
+		case AUX_CHANNEL_OPERATION_FAILED_HPD_DISCON:
+		case AUX_CHANNEL_OPERATION_FAILED_REASON_UNKNOWN:
+		default:
+			goto fail;
+		}
 	}
+
+fail:
+	if (!payload_reply)
+		payload->reply = NULL;
 	return false;
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.h b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.h
index e28ed6a00ff4..ce6a26d189b0 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.h
@@ -123,8 +123,9 @@ bool dce110_aux_engine_acquire(
 	struct dce_aux *aux_engine,
 	struct ddc *ddc);
 
-int dce_aux_transfer(struct ddc_service *ddc,
-		struct aux_payload *cmd);
+int dce_aux_transfer_raw(struct ddc_service *ddc,
+		struct aux_payload *cmd,
+		enum aux_channel_operation_result *operation_result);
 
 bool dce_aux_transfer_with_retries(struct ddc_service *ddc,
 		struct aux_payload *cmd);
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c
index 6e142c2db986..963686380738 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c
@@ -222,7 +222,7 @@ static enum dm_pp_clocks_state dce_get_required_clocks_state(
 	 * all required clocks
 	 */
 	for (i = clk_mgr_dce->max_clks_state; i >= DM_PP_CLOCKS_STATE_ULTRA_LOW; i--)
-		if (context->bw.dce.dispclk_khz >
+		if (context->bw_ctx.bw.dce.dispclk_khz >
 				clk_mgr_dce->max_clks_by_state[i].display_clk_khz
 			|| max_pix_clk >
 				clk_mgr_dce->max_clks_by_state[i].pixel_clk_khz)
@@ -232,7 +232,7 @@ static enum dm_pp_clocks_state dce_get_required_clocks_state(
 	if (low_req_clk > clk_mgr_dce->max_clks_state) {
 		/* set max clock state for high phyclock, invalid on exceeding display clock */
 		if (clk_mgr_dce->max_clks_by_state[clk_mgr_dce->max_clks_state].display_clk_khz
-				< context->bw.dce.dispclk_khz)
+				< context->bw_ctx.bw.dce.dispclk_khz)
 			low_req_clk = DM_PP_CLOCKS_STATE_INVALID;
 		else
 			low_req_clk = clk_mgr_dce->max_clks_state;
@@ -610,22 +610,22 @@ static void dce11_pplib_apply_display_requirements(
 	struct dm_pp_display_configuration *pp_display_cfg = &context->pp_display_cfg;
 
 	pp_display_cfg->all_displays_in_sync =
-		context->bw.dce.all_displays_in_sync;
+		context->bw_ctx.bw.dce.all_displays_in_sync;
 	pp_display_cfg->nb_pstate_switch_disable =
-			context->bw.dce.nbp_state_change_enable == false;
+			context->bw_ctx.bw.dce.nbp_state_change_enable == false;
 	pp_display_cfg->cpu_cc6_disable =
-			context->bw.dce.cpuc_state_change_enable == false;
+			context->bw_ctx.bw.dce.cpuc_state_change_enable == false;
 	pp_display_cfg->cpu_pstate_disable =
-			context->bw.dce.cpup_state_change_enable == false;
+			context->bw_ctx.bw.dce.cpup_state_change_enable == false;
 	pp_display_cfg->cpu_pstate_separation_time =
-			context->bw.dce.blackout_recovery_time_us;
+			context->bw_ctx.bw.dce.blackout_recovery_time_us;
 
-	pp_display_cfg->min_memory_clock_khz = context->bw.dce.yclk_khz
+	pp_display_cfg->min_memory_clock_khz = context->bw_ctx.bw.dce.yclk_khz
 		/ MEMORY_TYPE_MULTIPLIER_CZ;
 
 	pp_display_cfg->min_engine_clock_khz = determine_sclk_from_bounding_box(
 			dc,
-			context->bw.dce.sclk_khz);
+			context->bw_ctx.bw.dce.sclk_khz);
 
 	/*
 	 * As workaround for >4x4K lightup set dcfclock to min_engine_clock value.
@@ -638,7 +638,7 @@ static void dce11_pplib_apply_display_requirements(
 			pp_display_cfg->min_engine_clock_khz : 0;
 
 	pp_display_cfg->min_engine_clock_deep_sleep_khz
-			= context->bw.dce.sclk_deep_sleep_khz;
+			= context->bw_ctx.bw.dce.sclk_deep_sleep_khz;
 
 	pp_display_cfg->avail_mclk_switch_time_us =
 						dce110_get_min_vblank_time_us(context);
@@ -669,7 +669,7 @@ static void dce_update_clocks(struct clk_mgr *clk_mgr,
 {
 	struct dce_clk_mgr *clk_mgr_dce = TO_DCE_CLK_MGR(clk_mgr);
 	struct dm_pp_power_level_change_request level_change_req;
-	int patched_disp_clk = context->bw.dce.dispclk_khz;
+	int patched_disp_clk = context->bw_ctx.bw.dce.dispclk_khz;
 
 	/*TODO: W/A for dal3 linux, investigate why this works */
 	if (!clk_mgr_dce->dfs_bypass_active)
@@ -696,7 +696,7 @@ static void dce11_update_clocks(struct clk_mgr *clk_mgr,
 {
 	struct dce_clk_mgr *clk_mgr_dce = TO_DCE_CLK_MGR(clk_mgr);
 	struct dm_pp_power_level_change_request level_change_req;
-	int patched_disp_clk = context->bw.dce.dispclk_khz;
+	int patched_disp_clk = context->bw_ctx.bw.dce.dispclk_khz;
 
 	/*TODO: W/A for dal3 linux, investigate why this works */
 	if (!clk_mgr_dce->dfs_bypass_active)
@@ -711,7 +711,7 @@ static void dce11_update_clocks(struct clk_mgr *clk_mgr,
 	}
 
 	if (should_set_clock(safe_to_lower, patched_disp_clk, clk_mgr->clks.dispclk_khz)) {
-		context->bw.dce.dispclk_khz = dce_set_clock(clk_mgr, patched_disp_clk);
+		context->bw_ctx.bw.dce.dispclk_khz = dce_set_clock(clk_mgr, patched_disp_clk);
 		clk_mgr->clks.dispclk_khz = patched_disp_clk;
 	}
 	dce11_pplib_apply_display_requirements(clk_mgr->ctx->dc, context);
@@ -723,7 +723,7 @@ static void dce112_update_clocks(struct clk_mgr *clk_mgr,
 {
 	struct dce_clk_mgr *clk_mgr_dce = TO_DCE_CLK_MGR(clk_mgr);
 	struct dm_pp_power_level_change_request level_change_req;
-	int patched_disp_clk = context->bw.dce.dispclk_khz;
+	int patched_disp_clk = context->bw_ctx.bw.dce.dispclk_khz;
 
 	/*TODO: W/A for dal3 linux, investigate why this works */
 	if (!clk_mgr_dce->dfs_bypass_active)
@@ -751,7 +751,7 @@ static void dce12_update_clocks(struct clk_mgr *clk_mgr,
 	struct dce_clk_mgr *clk_mgr_dce = TO_DCE_CLK_MGR(clk_mgr);
 	struct dm_pp_clock_for_voltage_req clock_voltage_req = {0};
 	int max_pix_clk = get_max_pixel_clock_for_all_paths(context);
-	int patched_disp_clk = context->bw.dce.dispclk_khz;
+	int patched_disp_clk = context->bw_ctx.bw.dce.dispclk_khz;
 
 	/*TODO: W/A for dal3 linux, investigate why this works */
 	if (!clk_mgr_dce->dfs_bypass_active)
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c
index 71d5777de961..f70437aae8e0 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c
@@ -978,7 +978,7 @@ static bool dce110_clock_source_power_down(
 }
 
 static bool get_pixel_clk_frequency_100hz(
-		struct clock_source *clock_source,
+		const struct clock_source *clock_source,
 		unsigned int inst,
 		unsigned int *pixel_clk_khz)
 {
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c
index c2926cf19dee..818536eea00a 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c
@@ -51,6 +51,9 @@
 #define PSR_SET_WAITLOOP 0x31
 #define MCP_INIT_DMCU 0x88
 #define MCP_INIT_IRAM 0x89
+#define MCP_SYNC_PHY_LOCK 0x90
+#define MCP_SYNC_PHY_UNLOCK 0x91
+#define MCP_BL_SET_PWM_FRAC 0x6A  /* Enable or disable Fractional PWM */
 #define MASTER_COMM_CNTL_REG__MASTER_COMM_INTERRUPT_MASK   0x00000001L
 
 static bool dce_dmcu_init(struct dmcu *dmcu)
@@ -213,9 +216,6 @@ static bool dce_dmcu_setup_psr(struct dmcu *dmcu,
 	link->link_enc->funcs->psr_program_secondary_packet(link->link_enc,
 			psr_context->sdpTransmitLineNumDeadline);
 
-	if (psr_context->psr_level.bits.SKIP_SMU_NOTIFICATION)
-		REG_UPDATE(SMU_INTERRUPT_CONTROL, DC_SMU_INT_ENABLE, 1);
-
 	/* waitDMCUReadyForCmd */
 	REG_WAIT(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 0,
 					dmcu_wait_reg_ready_interval,
@@ -342,9 +342,32 @@ static void dcn10_get_dmcu_version(struct dmcu *dmcu)
 			IRAM_RD_ADDR_AUTO_INC, 0);
 }
 
+static void dcn10_dmcu_enable_fractional_pwm(struct dmcu *dmcu,
+		uint32_t fractional_pwm)
+{
+	struct dce_dmcu *dmcu_dce = TO_DCE_DMCU(dmcu);
+
+	/* Wait until microcontroller is ready to process interrupt */
+	REG_WAIT(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 0, 100, 800);
+
+	/* Set PWM fractional enable/disable */
+	REG_WRITE(MASTER_COMM_DATA_REG1, fractional_pwm);
+
+	/* Set command to enable or disable fractional PWM microcontroller */
+	REG_UPDATE(MASTER_COMM_CMD_REG, MASTER_COMM_CMD_REG_BYTE0,
+			MCP_BL_SET_PWM_FRAC);
+
+	/* Notify microcontroller of new command */
+	REG_UPDATE(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 1);
+
+	/* Ensure command has been executed before continuing */
+	REG_WAIT(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 0, 100, 800);
+}
+
 static bool dcn10_dmcu_init(struct dmcu *dmcu)
 {
 	struct dce_dmcu *dmcu_dce = TO_DCE_DMCU(dmcu);
+	const struct dc_config *config = &dmcu->ctx->dc->config;
 	bool status = false;
 
 	/*  Definition of DC_DMCU_SCRATCH
@@ -382,9 +405,14 @@ static bool dcn10_dmcu_init(struct dmcu *dmcu)
 		if (dmcu->dmcu_state == DMCU_RUNNING) {
 			/* Retrieve and cache the DMCU firmware version. */
 			dcn10_get_dmcu_version(dmcu);
+
+			/* Initialize DMCU to use fractional PWM or not */
+			dcn10_dmcu_enable_fractional_pwm(dmcu,
+				(config->disable_fractional_pwm == false) ? 1 : 0);
 			status = true;
-		} else
+		} else {
 			status = false;
+		}
 
 		break;
 	case DMCU_RUNNING:
@@ -594,7 +622,7 @@ static bool dcn10_dmcu_setup_psr(struct dmcu *dmcu,
 	link->link_enc->funcs->psr_program_secondary_packet(link->link_enc,
 			psr_context->sdpTransmitLineNumDeadline);
 
-	if (psr_context->psr_level.bits.SKIP_SMU_NOTIFICATION)
+	if (psr_context->allow_smu_optimizations)
 		REG_UPDATE(SMU_INTERRUPT_CONTROL, DC_SMU_INT_ENABLE, 1);
 
 	/* waitDMCUReadyForCmd */
@@ -615,6 +643,7 @@ static bool dcn10_dmcu_setup_psr(struct dmcu *dmcu,
 			psr_context->psrFrameCaptureIndicationReq;
 	masterCmdData1.bits.aux_chan = psr_context->channel;
 	masterCmdData1.bits.aux_repeat = psr_context->aux_repeats;
+	masterCmdData1.bits.allow_smu_optimizations = psr_context->allow_smu_optimizations;
 	dm_write_reg(dmcu->ctx, REG(MASTER_COMM_DATA_REG1),
 					masterCmdData1.u32All);
 
@@ -635,6 +664,7 @@ static bool dcn10_dmcu_setup_psr(struct dmcu *dmcu,
 	dm_write_reg(dmcu->ctx, REG(MASTER_COMM_DATA_REG3),
 			masterCmdData3.u32All);
 
+
 	/* setDMCUParam_Cmd */
 	REG_UPDATE(MASTER_COMM_CMD_REG,
 			MASTER_COMM_CMD_REG_BYTE0, PSR_SET);
@@ -691,7 +721,7 @@ static bool dcn10_is_dmcu_initialized(struct dmcu *dmcu)
 	return true;
 }
 
-#endif
+#endif //(CONFIG_DRM_AMD_DC_DCN1_0)
 
 static const struct dmcu_funcs dce_funcs = {
 	.dmcu_init = dce_dmcu_init,
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.h b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.h
index c24c0e5ea44e..60ce56f60ae3 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.h
@@ -199,16 +199,16 @@ struct dce_dmcu {
  ******************************************************************/
 union dce_dmcu_psr_config_data_reg1 {
 	struct {
-		unsigned int timehyst_frames:8;    /*[7:0]*/
-		unsigned int hyst_lines:7;         /*[14:8]*/
-		unsigned int rfb_update_auto_en:1; /*[15:15]*/
-		unsigned int dp_port_num:3;        /*[18:16]*/
-		unsigned int dcp_sel:3;            /*[21:19]*/
-		unsigned int phy_type:1;           /*[22:22]*/
-		unsigned int frame_cap_ind:1;      /*[23:23]*/
-		unsigned int aux_chan:3;           /*[26:24]*/
-		unsigned int aux_repeat:4;         /*[30:27]*/
-		unsigned int reserved:1;           /*[31:31]*/
+		unsigned int timehyst_frames:8;                  /*[7:0]*/
+		unsigned int hyst_lines:7;                       /*[14:8]*/
+		unsigned int rfb_update_auto_en:1;               /*[15:15]*/
+		unsigned int dp_port_num:3;                      /*[18:16]*/
+		unsigned int dcp_sel:3;                          /*[21:19]*/
+		unsigned int phy_type:1;                         /*[22:22]*/
+		unsigned int frame_cap_ind:1;                    /*[23:23]*/
+		unsigned int aux_chan:3;                         /*[26:24]*/
+		unsigned int aux_repeat:4;                       /*[30:27]*/
+		unsigned int allow_smu_optimizations:1;         /*[31:31]*/
 	} bits;
 	unsigned int u32All;
 };
@@ -236,7 +236,7 @@ union dce_dmcu_psr_config_data_reg3 {
 	struct {
 		unsigned int psr_level:16;      /*[15:0]*/
 		unsigned int link_rate:4;       /*[19:16]*/
-		unsigned int reserved:12;       /*[31:20]*/
+		unsigned int reserved:12;        /*[31:20]*/
 	} bits;
 	unsigned int u32All;
 };
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c
index 40f2d6e0b122..cd26161bcc4d 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c
@@ -346,6 +346,16 @@ static void release_engine(
 
 }
 
+static bool is_engine_available(struct dce_i2c_hw *dce_i2c_hw)
+{
+	unsigned int arbitrate;
+
+	REG_GET(DC_I2C_ARBITRATION, DC_I2C_REG_RW_CNTL_STATUS, &arbitrate);
+	if (arbitrate == DC_I2C_REG_RW_CNTL_STATUS_DMCU_ONLY)
+		return false;
+	return true;
+}
+
 struct dce_i2c_hw *acquire_i2c_hw_engine(
 	struct resource_pool *pool,
 	struct ddc *ddc)
@@ -368,7 +378,7 @@ struct dce_i2c_hw *acquire_i2c_hw_engine(
 	if (!dce_i2c_hw)
 		return NULL;
 
-	if (pool->i2c_hw_buffer_in_use)
+	if (pool->i2c_hw_buffer_in_use || !is_engine_available(dce_i2c_hw))
 		return NULL;
 
 	do {
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.h b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.h
index 7f19bb439665..575500755b2e 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.h
@@ -29,7 +29,8 @@
 enum dc_i2c_status {
 	DC_I2C_STATUS__DC_I2C_STATUS_IDLE,
 	DC_I2C_STATUS__DC_I2C_STATUS_USED_BY_SW,
-	DC_I2C_STATUS__DC_I2C_STATUS_USED_BY_HW
+	DC_I2C_STATUS__DC_I2C_STATUS_USED_BY_HW,
+	DC_I2C_REG_RW_CNTL_STATUS_DMCU_ONLY = 2,
 };
 
 enum dc_i2c_arbitration {
@@ -129,7 +130,8 @@ enum {
 	I2C_SF(DC_I2C_DATA, DC_I2C_DATA, mask_sh),\
 	I2C_SF(DC_I2C_DATA, DC_I2C_INDEX, mask_sh),\
 	I2C_SF(DC_I2C_DATA, DC_I2C_INDEX_WRITE, mask_sh),\
-	I2C_SF(MICROSECOND_TIME_BASE_DIV, XTAL_REF_DIV, mask_sh)
+	I2C_SF(MICROSECOND_TIME_BASE_DIV, XTAL_REF_DIV, mask_sh),\
+	I2C_SF(DC_I2C_ARBITRATION, DC_I2C_REG_RW_CNTL_STATUS, mask_sh)
 
 #define I2C_COMMON_MASK_SH_LIST_DCE110(mask_sh)\
 	I2C_COMMON_MASK_SH_LIST_DCE_COMMON_BASE(mask_sh),\
@@ -170,6 +172,7 @@ struct dce_i2c_shift {
 	uint8_t DC_I2C_INDEX;
 	uint8_t DC_I2C_INDEX_WRITE;
 	uint8_t XTAL_REF_DIV;
+	uint8_t DC_I2C_REG_RW_CNTL_STATUS;
 };
 
 struct dce_i2c_mask {
@@ -207,6 +210,7 @@ struct dce_i2c_mask {
 	uint32_t DC_I2C_INDEX;
 	uint32_t DC_I2C_INDEX_WRITE;
 	uint32_t XTAL_REF_DIV;
+	uint32_t DC_I2C_REG_RW_CNTL_STATUS;
 };
 
 struct dce_i2c_registers {
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c
index 1fa2d4fd7a35..14309fe6f2e6 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c
@@ -272,7 +272,8 @@ static void dce110_update_hdmi_info_packet(
 static void dce110_stream_encoder_dp_set_stream_attribute(
 	struct stream_encoder *enc,
 	struct dc_crtc_timing *crtc_timing,
-	enum dc_color_space output_color_space)
+	enum dc_color_space output_color_space,
+	uint32_t enable_sdp_splitting)
 {
 #if defined(CONFIG_DRM_AMD_DC_DCN1_0)
 	uint32_t h_active_start;
@@ -977,7 +978,7 @@ static void dce110_stream_encoder_dp_unblank(
 
 		uint64_t m_vid_l = n_vid;
 
-		m_vid_l *= param->pixel_clk_khz;
+		m_vid_l *= param->timing.pix_clk_100hz / 10;
 		m_vid_l = div_u64(m_vid_l,
 			param->link_settings.link_rate
 				* LINK_RATE_REF_FREQ_IN_KHZ);
diff --git a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c
index 23044e6723e8..e938bf9986d3 100644
--- a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c
@@ -378,6 +378,28 @@ static const struct resource_caps res_cap = {
 	.num_ddc = 6,
 };
 
+static const struct dc_plane_cap plane_cap = {
+	.type = DC_PLANE_TYPE_DCE_RGB,
+
+	.pixel_format_support = {
+			.argb8888 = true,
+			.nv12 = false,
+			.fp16 = false
+	},
+
+	.max_upscale_factor = {
+			.argb8888 = 16000,
+			.nv12 = 1,
+			.fp16 = 1
+	},
+
+	.max_downscale_factor = {
+			.argb8888 = 250,
+			.nv12 = 1,
+			.fp16 = 1
+	}
+};
+
 #define CTX  ctx
 #define REG(reg) mm ## reg
 
@@ -756,7 +778,8 @@ static enum dc_status build_mapped_resource(
 
 bool dce100_validate_bandwidth(
 	struct dc  *dc,
-	struct dc_state *context)
+	struct dc_state *context,
+	bool fast_validate)
 {
 	int i;
 	bool at_least_one_pipe = false;
@@ -768,11 +791,11 @@ bool dce100_validate_bandwidth(
 
 	if (at_least_one_pipe) {
 		/* TODO implement when needed but for now hardcode max value*/
-		context->bw.dce.dispclk_khz = 681000;
-		context->bw.dce.yclk_khz = 250000 * MEMORY_TYPE_MULTIPLIER_CZ;
+		context->bw_ctx.bw.dce.dispclk_khz = 681000;
+		context->bw_ctx.bw.dce.yclk_khz = 250000 * MEMORY_TYPE_MULTIPLIER_CZ;
 	} else {
-		context->bw.dce.dispclk_khz = 0;
-		context->bw.dce.yclk_khz = 0;
+		context->bw_ctx.bw.dce.dispclk_khz = 0;
+		context->bw_ctx.bw.dce.yclk_khz = 0;
 	}
 
 	return true;
@@ -1023,6 +1046,9 @@ static bool construct(
 
 	dc->caps.max_planes =  pool->base.pipe_count;
 
+	for (i = 0; i < dc->caps.max_planes; ++i)
+		dc->caps.planes[i] = plane_cap;
+
 	if (!resource_construct(num_virtual_links, dc, &pool->base,
 			&res_create_funcs))
 		goto res_create_fail;
diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
index 5e4db3712eef..7ac50ab1b762 100644
--- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
@@ -616,7 +616,7 @@ dce110_set_output_transfer_func(struct pipe_ctx *pipe_ctx,
 
 void dce110_update_info_frame(struct pipe_ctx *pipe_ctx)
 {
-	bool is_hdmi;
+	bool is_hdmi_tmds;
 	bool is_dp;
 
 	ASSERT(pipe_ctx->stream);
@@ -624,13 +624,13 @@ void dce110_update_info_frame(struct pipe_ctx *pipe_ctx)
 	if (pipe_ctx->stream_res.stream_enc == NULL)
 		return;  /* this is not root pipe */
 
-	is_hdmi = dc_is_hdmi_signal(pipe_ctx->stream->signal);
+	is_hdmi_tmds = dc_is_hdmi_tmds_signal(pipe_ctx->stream->signal);
 	is_dp = dc_is_dp_signal(pipe_ctx->stream->signal);
 
-	if (!is_hdmi && !is_dp)
+	if (!is_hdmi_tmds && !is_dp)
 		return;
 
-	if (is_hdmi)
+	if (is_hdmi_tmds)
 		pipe_ctx->stream_res.stream_enc->funcs->update_hdmi_info_packets(
 			pipe_ctx->stream_res.stream_enc,
 			&pipe_ctx->stream_res.encoder_info_frame);
@@ -935,13 +935,31 @@ void hwss_edp_backlight_control(
 		edp_receiver_ready_T9(link);
 }
 
+// Static helper function which calls the correct function
+// based on pp_smu version
+static void set_pme_wa_enable_by_version(struct dc *dc)
+{
+	struct pp_smu_funcs *pp_smu = NULL;
+
+	if (dc->res_pool->pp_smu)
+		pp_smu = dc->res_pool->pp_smu;
+
+	if (pp_smu) {
+		if (pp_smu->ctx.ver == PP_SMU_VER_RV && pp_smu->rv_funcs.set_pme_wa_enable)
+			pp_smu->rv_funcs.set_pme_wa_enable(&(pp_smu->ctx));
+	}
+}
+
 void dce110_enable_audio_stream(struct pipe_ctx *pipe_ctx)
 {
-	struct dc *core_dc = pipe_ctx->stream->ctx->dc;
 	/* notify audio driver for audio modes of monitor */
-	struct pp_smu_funcs_rv *pp_smu = core_dc->res_pool->pp_smu;
+	struct dc *core_dc = pipe_ctx->stream->ctx->dc;
+	struct pp_smu_funcs *pp_smu = NULL;
 	unsigned int i, num_audio = 1;
 
+	if (core_dc->res_pool->pp_smu)
+		pp_smu = core_dc->res_pool->pp_smu;
+
 	if (pipe_ctx->stream_res.audio) {
 		for (i = 0; i < MAX_PIPES; i++) {
 			/*current_state not updated yet*/
@@ -951,30 +969,31 @@ void dce110_enable_audio_stream(struct pipe_ctx *pipe_ctx)
 
 		pipe_ctx->stream_res.audio->funcs->az_enable(pipe_ctx->stream_res.audio);
 
-		if (num_audio >= 1 && pp_smu != NULL && pp_smu->set_pme_wa_enable != NULL)
+		if (num_audio >= 1 && pp_smu != NULL)
 			/*this is the first audio. apply the PME w/a in order to wake AZ from D3*/
-			pp_smu->set_pme_wa_enable(&pp_smu->pp_smu);
+			set_pme_wa_enable_by_version(core_dc);
 		/* un-mute audio */
 		/* TODO: audio should be per stream rather than per link */
 		pipe_ctx->stream_res.stream_enc->funcs->audio_mute_control(
-			pipe_ctx->stream_res.stream_enc, false);
+					pipe_ctx->stream_res.stream_enc, false);
 	}
 }
 
 void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx, int option)
 {
 	struct dc *dc = pipe_ctx->stream->ctx->dc;
+	struct pp_smu_funcs *pp_smu = NULL;
 
 	pipe_ctx->stream_res.stream_enc->funcs->audio_mute_control(
 			pipe_ctx->stream_res.stream_enc, true);
 	if (pipe_ctx->stream_res.audio) {
-		struct pp_smu_funcs_rv *pp_smu = dc->res_pool->pp_smu;
+		if (dc->res_pool->pp_smu)
+			pp_smu = dc->res_pool->pp_smu;
 
 		if (option != KEEP_ACQUIRED_RESOURCE ||
-				!dc->debug.az_endpoint_mute_only) {
+				!dc->debug.az_endpoint_mute_only)
 			/*only disalbe az_endpoint if power down or free*/
 			pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio);
-		}
 
 		if (dc_is_dp_signal(pipe_ctx->stream->signal))
 			pipe_ctx->stream_res.stream_enc->funcs->dp_audio_disable(
@@ -989,9 +1008,9 @@ void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx, int option)
 			update_audio_usage(&dc->current_state->res_ctx, dc->res_pool, pipe_ctx->stream_res.audio, false);
 			pipe_ctx->stream_res.audio = NULL;
 		}
-		if (pp_smu != NULL && pp_smu->set_pme_wa_enable != NULL)
+		if (pp_smu != NULL)
 			/*this is the first audio. apply the PME w/a in order to wake AZ from D3*/
-			pp_smu->set_pme_wa_enable(&pp_smu->pp_smu);
+			set_pme_wa_enable_by_version(dc);
 
 		/* TODO: notify audio driver for if audio modes list changed
 		 * add audio mode list change flag */
@@ -1007,7 +1026,7 @@ void dce110_disable_stream(struct pipe_ctx *pipe_ctx, int option)
 	struct dc_link *link = stream->link;
 	struct dc *dc = pipe_ctx->stream->ctx->dc;
 
-	if (dc_is_hdmi_signal(pipe_ctx->stream->signal))
+	if (dc_is_hdmi_tmds_signal(pipe_ctx->stream->signal))
 		pipe_ctx->stream_res.stream_enc->funcs->stop_hdmi_info_packets(
 			pipe_ctx->stream_res.stream_enc);
 
@@ -1032,7 +1051,7 @@ void dce110_unblank_stream(struct pipe_ctx *pipe_ctx,
 	struct dc_link *link = stream->link;
 
 	/* only 3 items below are used by unblank */
-	params.pixel_clk_khz = pipe_ctx->stream->timing.pix_clk_100hz / 10;
+	params.timing = pipe_ctx->stream->timing;
 	params.link_settings.link_rate = link_settings->link_rate;
 
 	if (dc_is_dp_signal(pipe_ctx->stream->signal))
@@ -1147,8 +1166,8 @@ static void build_audio_output(
 	if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT ||
 			pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) {
 		audio_output->pll_info.dp_dto_source_clock_in_khz =
-				state->dccg->funcs->get_dp_ref_clk_frequency(
-						state->dccg);
+				state->clk_mgr->funcs->get_dp_ref_clk_frequency(
+						state->clk_mgr);
 	}
 
 	audio_output->pll_info.feed_back_divider =
@@ -1349,7 +1368,7 @@ static enum dc_status apply_single_controller_ctx_to_hw(
 		pipe_ctx->stream_res.tg->funcs->set_static_screen_control(
 				pipe_ctx->stream_res.tg, event_triggers);
 
-	if (pipe_ctx->stream->signal != SIGNAL_TYPE_VIRTUAL)
+	if (!dc_is_virtual_signal(pipe_ctx->stream->signal))
 		pipe_ctx->stream_res.stream_enc->funcs->dig_connect_to_otg(
 			pipe_ctx->stream_res.stream_enc,
 			pipe_ctx->stream_res.tg->inst);
@@ -1358,7 +1377,7 @@ static enum dc_status apply_single_controller_ctx_to_hw(
 			pipe_ctx->stream_res.opp,
 			COLOR_SPACE_YCBCR601,
 			stream->timing.display_color_depth,
-			pipe_ctx->stream->signal);
+			stream->signal);
 
 	pipe_ctx->stream_res.opp->funcs->opp_program_fmt(
 		pipe_ctx->stream_res.opp,
@@ -1532,6 +1551,9 @@ void dce110_enable_accelerated_mode(struct dc *dc, struct dc_state *context)
 		}
 	}
 
+	if (dc->hwss.init_pipes)
+		dc->hwss.init_pipes(dc, context);
+
 	if (edp_link) {
 		/* this seems to cause blank screens on DCE8 */
 		if ((dc->ctx->dce_version == DCE_VERSION_8_0) ||
@@ -1608,18 +1630,18 @@ static void dce110_set_displaymarks(
 			dc->bw_vbios->blackout_duration, pipe_ctx->stream);
 		pipe_ctx->plane_res.mi->funcs->mem_input_program_display_marks(
 			pipe_ctx->plane_res.mi,
-			context->bw.dce.nbp_state_change_wm_ns[num_pipes],
-			context->bw.dce.stutter_exit_wm_ns[num_pipes],
-			context->bw.dce.stutter_entry_wm_ns[num_pipes],
-			context->bw.dce.urgent_wm_ns[num_pipes],
+			context->bw_ctx.bw.dce.nbp_state_change_wm_ns[num_pipes],
+			context->bw_ctx.bw.dce.stutter_exit_wm_ns[num_pipes],
+			context->bw_ctx.bw.dce.stutter_entry_wm_ns[num_pipes],
+			context->bw_ctx.bw.dce.urgent_wm_ns[num_pipes],
 			total_dest_line_time_ns);
 		if (i == underlay_idx) {
 			num_pipes++;
 			pipe_ctx->plane_res.mi->funcs->mem_input_program_chroma_display_marks(
 				pipe_ctx->plane_res.mi,
-				context->bw.dce.nbp_state_change_wm_ns[num_pipes],
-				context->bw.dce.stutter_exit_wm_ns[num_pipes],
-				context->bw.dce.urgent_wm_ns[num_pipes],
+				context->bw_ctx.bw.dce.nbp_state_change_wm_ns[num_pipes],
+				context->bw_ctx.bw.dce.stutter_exit_wm_ns[num_pipes],
+				context->bw_ctx.bw.dce.urgent_wm_ns[num_pipes],
 				total_dest_line_time_ns);
 		}
 		num_pipes++;
@@ -2612,7 +2634,7 @@ void dce110_set_cursor_position(struct pipe_ctx *pipe_ctx)
 	struct mem_input *mi = pipe_ctx->plane_res.mi;
 	struct dc_cursor_mi_param param = {
 		.pixel_clk_khz = pipe_ctx->stream->timing.pix_clk_100hz / 10,
-		.ref_clk_khz = pipe_ctx->stream->ctx->dc->res_pool->ref_clock_inKhz,
+		.ref_clk_khz = pipe_ctx->stream->ctx->dc->res_pool->ref_clocks.xtalin_clock_inKhz,
 		.viewport = pipe_ctx->plane_res.scl_data.viewport,
 		.h_scale_ratio = pipe_ctx->plane_res.scl_data.ratios.horz,
 		.v_scale_ratio = pipe_ctx->plane_res.scl_data.ratios.vert,
diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c
index 7549adaa1542..dcd04e9ea76b 100644
--- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c
@@ -392,6 +392,55 @@ static const struct resource_caps stoney_resource_cap = {
 		.num_ddc = 3,
 };
 
+static const struct dc_plane_cap plane_cap = {
+		.type = DC_PLANE_TYPE_DCE_RGB,
+		.blends_with_below = true,
+		.blends_with_above = true,
+		.per_pixel_alpha = 1,
+
+		.pixel_format_support = {
+				.argb8888 = true,
+				.nv12 = false,
+				.fp16 = false
+		},
+
+		.max_upscale_factor = {
+				.argb8888 = 16000,
+				.nv12 = 1,
+				.fp16 = 1
+		},
+
+		.max_downscale_factor = {
+				.argb8888 = 250,
+				.nv12 = 1,
+				.fp16 = 1
+		}
+};
+
+static const struct dc_plane_cap underlay_plane_cap = {
+		.type = DC_PLANE_TYPE_DCE_UNDERLAY,
+		.blends_with_above = true,
+		.per_pixel_alpha = 1,
+
+		.pixel_format_support = {
+				.argb8888 = false,
+				.nv12 = true,
+				.fp16 = false
+		},
+
+		.max_upscale_factor = {
+				.argb8888 = 1,
+				.nv12 = 16000,
+				.fp16 = 1
+		},
+
+		.max_downscale_factor = {
+				.argb8888 = 1,
+				.nv12 = 250,
+				.fp16 = 1
+		}
+};
+
 #define CTX  ctx
 #define REG(reg) mm ## reg
 
@@ -854,7 +903,8 @@ static enum dc_status build_mapped_resource(
 
 static bool dce110_validate_bandwidth(
 	struct dc *dc,
-	struct dc_state *context)
+	struct dc_state *context,
+	bool fast_validate)
 {
 	bool result = false;
 
@@ -868,7 +918,7 @@ static bool dce110_validate_bandwidth(
 			dc->bw_vbios,
 			context->res_ctx.pipe_ctx,
 			dc->res_pool->pipe_count,
-			&context->bw.dce))
+			&context->bw_ctx.bw.dce))
 		result =  true;
 
 	if (!result)
@@ -878,8 +928,8 @@ static bool dce110_validate_bandwidth(
 			context->streams[0]->timing.v_addressable,
 			context->streams[0]->timing.pix_clk_100hz / 10);
 
-	if (memcmp(&dc->current_state->bw.dce,
-			&context->bw.dce, sizeof(context->bw.dce))) {
+	if (memcmp(&dc->current_state->bw_ctx.bw.dce,
+			&context->bw_ctx.bw.dce, sizeof(context->bw_ctx.bw.dce))) {
 
 		DC_LOG_BANDWIDTH_CALCS(
 			"%s: finish,\n"
@@ -893,34 +943,34 @@ static bool dce110_validate_bandwidth(
 			"sclk: %d sclk_sleep: %d yclk: %d blackout_recovery_time_us: %d\n"
 			,
 			__func__,
-			context->bw.dce.nbp_state_change_wm_ns[0].b_mark,
-			context->bw.dce.nbp_state_change_wm_ns[0].a_mark,
-			context->bw.dce.urgent_wm_ns[0].b_mark,
-			context->bw.dce.urgent_wm_ns[0].a_mark,
-			context->bw.dce.stutter_exit_wm_ns[0].b_mark,
-			context->bw.dce.stutter_exit_wm_ns[0].a_mark,
-			context->bw.dce.nbp_state_change_wm_ns[1].b_mark,
-			context->bw.dce.nbp_state_change_wm_ns[1].a_mark,
-			context->bw.dce.urgent_wm_ns[1].b_mark,
-			context->bw.dce.urgent_wm_ns[1].a_mark,
-			context->bw.dce.stutter_exit_wm_ns[1].b_mark,
-			context->bw.dce.stutter_exit_wm_ns[1].a_mark,
-			context->bw.dce.nbp_state_change_wm_ns[2].b_mark,
-			context->bw.dce.nbp_state_change_wm_ns[2].a_mark,
-			context->bw.dce.urgent_wm_ns[2].b_mark,
-			context->bw.dce.urgent_wm_ns[2].a_mark,
-			context->bw.dce.stutter_exit_wm_ns[2].b_mark,
-			context->bw.dce.stutter_exit_wm_ns[2].a_mark,
-			context->bw.dce.stutter_mode_enable,
-			context->bw.dce.cpuc_state_change_enable,
-			context->bw.dce.cpup_state_change_enable,
-			context->bw.dce.nbp_state_change_enable,
-			context->bw.dce.all_displays_in_sync,
-			context->bw.dce.dispclk_khz,
-			context->bw.dce.sclk_khz,
-			context->bw.dce.sclk_deep_sleep_khz,
-			context->bw.dce.yclk_khz,
-			context->bw.dce.blackout_recovery_time_us);
+			context->bw_ctx.bw.dce.nbp_state_change_wm_ns[0].b_mark,
+			context->bw_ctx.bw.dce.nbp_state_change_wm_ns[0].a_mark,
+			context->bw_ctx.bw.dce.urgent_wm_ns[0].b_mark,
+			context->bw_ctx.bw.dce.urgent_wm_ns[0].a_mark,
+			context->bw_ctx.bw.dce.stutter_exit_wm_ns[0].b_mark,
+			context->bw_ctx.bw.dce.stutter_exit_wm_ns[0].a_mark,
+			context->bw_ctx.bw.dce.nbp_state_change_wm_ns[1].b_mark,
+			context->bw_ctx.bw.dce.nbp_state_change_wm_ns[1].a_mark,
+			context->bw_ctx.bw.dce.urgent_wm_ns[1].b_mark,
+			context->bw_ctx.bw.dce.urgent_wm_ns[1].a_mark,
+			context->bw_ctx.bw.dce.stutter_exit_wm_ns[1].b_mark,
+			context->bw_ctx.bw.dce.stutter_exit_wm_ns[1].a_mark,
+			context->bw_ctx.bw.dce.nbp_state_change_wm_ns[2].b_mark,
+			context->bw_ctx.bw.dce.nbp_state_change_wm_ns[2].a_mark,
+			context->bw_ctx.bw.dce.urgent_wm_ns[2].b_mark,
+			context->bw_ctx.bw.dce.urgent_wm_ns[2].a_mark,
+			context->bw_ctx.bw.dce.stutter_exit_wm_ns[2].b_mark,
+			context->bw_ctx.bw.dce.stutter_exit_wm_ns[2].a_mark,
+			context->bw_ctx.bw.dce.stutter_mode_enable,
+			context->bw_ctx.bw.dce.cpuc_state_change_enable,
+			context->bw_ctx.bw.dce.cpup_state_change_enable,
+			context->bw_ctx.bw.dce.nbp_state_change_enable,
+			context->bw_ctx.bw.dce.all_displays_in_sync,
+			context->bw_ctx.bw.dce.dispclk_khz,
+			context->bw_ctx.bw.dce.sclk_khz,
+			context->bw_ctx.bw.dce.sclk_deep_sleep_khz,
+			context->bw_ctx.bw.dce.yclk_khz,
+			context->bw_ctx.bw.dce.blackout_recovery_time_us);
 	}
 	return result;
 }
@@ -1371,6 +1421,11 @@ static bool construct(
 
 	dc->caps.max_planes =  pool->base.pipe_count;
 
+	for (i = 0; i < pool->base.underlay_pipe_index; ++i)
+		dc->caps.planes[i] = plane_cap;
+
+	dc->caps.planes[pool->base.underlay_pipe_index] = underlay_plane_cap;
+
 	bw_calcs_init(dc->bw_dceip, dc->bw_vbios, dc->ctx->asic_id);
 
 	bw_calcs_data_update_from_pplib(dc);
diff --git a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c
index ea3065d63372..a480b15f6885 100644
--- a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c
@@ -397,6 +397,28 @@ static const struct resource_caps polaris_11_resource_cap = {
 		.num_ddc = 5,
 };
 
+static const struct dc_plane_cap plane_cap = {
+	.type = DC_PLANE_TYPE_DCE_RGB,
+
+	.pixel_format_support = {
+			.argb8888 = true,
+			.nv12 = false,
+			.fp16 = false
+	},
+
+	.max_upscale_factor = {
+			.argb8888 = 16000,
+			.nv12 = 1,
+			.fp16 = 1
+	},
+
+	.max_downscale_factor = {
+			.argb8888 = 250,
+			.nv12 = 1,
+			.fp16 = 1
+	}
+};
+
 #define CTX  ctx
 #define REG(reg) mm ## reg
 
@@ -804,7 +826,8 @@ static enum dc_status build_mapped_resource(
 
 bool dce112_validate_bandwidth(
 	struct dc *dc,
-	struct dc_state *context)
+	struct dc_state *context,
+	bool fast_validate)
 {
 	bool result = false;
 
@@ -818,7 +841,7 @@ bool dce112_validate_bandwidth(
 			dc->bw_vbios,
 			context->res_ctx.pipe_ctx,
 			dc->res_pool->pipe_count,
-			&context->bw.dce))
+			&context->bw_ctx.bw.dce))
 		result = true;
 
 	if (!result)
@@ -826,8 +849,8 @@ bool dce112_validate_bandwidth(
 			"%s: Bandwidth validation failed!",
 			__func__);
 
-	if (memcmp(&dc->current_state->bw.dce,
-			&context->bw.dce, sizeof(context->bw.dce))) {
+	if (memcmp(&dc->current_state->bw_ctx.bw.dce,
+			&context->bw_ctx.bw.dce, sizeof(context->bw_ctx.bw.dce))) {
 
 		DC_LOG_BANDWIDTH_CALCS(
 			"%s: finish,\n"
@@ -841,34 +864,34 @@ bool dce112_validate_bandwidth(
 			"sclk: %d sclk_sleep: %d yclk: %d blackout_recovery_time_us: %d\n"
 			,
 			__func__,
-			context->bw.dce.nbp_state_change_wm_ns[0].b_mark,
-			context->bw.dce.nbp_state_change_wm_ns[0].a_mark,
-			context->bw.dce.urgent_wm_ns[0].b_mark,
-			context->bw.dce.urgent_wm_ns[0].a_mark,
-			context->bw.dce.stutter_exit_wm_ns[0].b_mark,
-			context->bw.dce.stutter_exit_wm_ns[0].a_mark,
-			context->bw.dce.nbp_state_change_wm_ns[1].b_mark,
-			context->bw.dce.nbp_state_change_wm_ns[1].a_mark,
-			context->bw.dce.urgent_wm_ns[1].b_mark,
-			context->bw.dce.urgent_wm_ns[1].a_mark,
-			context->bw.dce.stutter_exit_wm_ns[1].b_mark,
-			context->bw.dce.stutter_exit_wm_ns[1].a_mark,
-			context->bw.dce.nbp_state_change_wm_ns[2].b_mark,
-			context->bw.dce.nbp_state_change_wm_ns[2].a_mark,
-			context->bw.dce.urgent_wm_ns[2].b_mark,
-			context->bw.dce.urgent_wm_ns[2].a_mark,
-			context->bw.dce.stutter_exit_wm_ns[2].b_mark,
-			context->bw.dce.stutter_exit_wm_ns[2].a_mark,
-			context->bw.dce.stutter_mode_enable,
-			context->bw.dce.cpuc_state_change_enable,
-			context->bw.dce.cpup_state_change_enable,
-			context->bw.dce.nbp_state_change_enable,
-			context->bw.dce.all_displays_in_sync,
-			context->bw.dce.dispclk_khz,
-			context->bw.dce.sclk_khz,
-			context->bw.dce.sclk_deep_sleep_khz,
-			context->bw.dce.yclk_khz,
-			context->bw.dce.blackout_recovery_time_us);
+			context->bw_ctx.bw.dce.nbp_state_change_wm_ns[0].b_mark,
+			context->bw_ctx.bw.dce.nbp_state_change_wm_ns[0].a_mark,
+			context->bw_ctx.bw.dce.urgent_wm_ns[0].b_mark,
+			context->bw_ctx.bw.dce.urgent_wm_ns[0].a_mark,
+			context->bw_ctx.bw.dce.stutter_exit_wm_ns[0].b_mark,
+			context->bw_ctx.bw.dce.stutter_exit_wm_ns[0].a_mark,
+			context->bw_ctx.bw.dce.nbp_state_change_wm_ns[1].b_mark,
+			context->bw_ctx.bw.dce.nbp_state_change_wm_ns[1].a_mark,
+			context->bw_ctx.bw.dce.urgent_wm_ns[1].b_mark,
+			context->bw_ctx.bw.dce.urgent_wm_ns[1].a_mark,
+			context->bw_ctx.bw.dce.stutter_exit_wm_ns[1].b_mark,
+			context->bw_ctx.bw.dce.stutter_exit_wm_ns[1].a_mark,
+			context->bw_ctx.bw.dce.nbp_state_change_wm_ns[2].b_mark,
+			context->bw_ctx.bw.dce.nbp_state_change_wm_ns[2].a_mark,
+			context->bw_ctx.bw.dce.urgent_wm_ns[2].b_mark,
+			context->bw_ctx.bw.dce.urgent_wm_ns[2].a_mark,
+			context->bw_ctx.bw.dce.stutter_exit_wm_ns[2].b_mark,
+			context->bw_ctx.bw.dce.stutter_exit_wm_ns[2].a_mark,
+			context->bw_ctx.bw.dce.stutter_mode_enable,
+			context->bw_ctx.bw.dce.cpuc_state_change_enable,
+			context->bw_ctx.bw.dce.cpup_state_change_enable,
+			context->bw_ctx.bw.dce.nbp_state_change_enable,
+			context->bw_ctx.bw.dce.all_displays_in_sync,
+			context->bw_ctx.bw.dce.dispclk_khz,
+			context->bw_ctx.bw.dce.sclk_khz,
+			context->bw_ctx.bw.dce.sclk_deep_sleep_khz,
+			context->bw_ctx.bw.dce.yclk_khz,
+			context->bw_ctx.bw.dce.blackout_recovery_time_us);
 	}
 	return result;
 }
@@ -887,7 +910,7 @@ enum dc_status resource_map_phy_clock_resources(
 		return DC_ERROR_UNEXPECTED;
 
 	if (dc_is_dp_signal(pipe_ctx->stream->signal)
-		|| pipe_ctx->stream->signal == SIGNAL_TYPE_VIRTUAL)
+		|| dc_is_virtual_signal(pipe_ctx->stream->signal))
 		pipe_ctx->clock_source =
 				dc->res_pool->dp_clock_source;
 	else
@@ -1310,6 +1333,9 @@ static bool construct(
 
 	dc->caps.max_planes =  pool->base.pipe_count;
 
+	for (i = 0; i < dc->caps.max_planes; ++i)
+		dc->caps.planes[i] = plane_cap;
+
 	/* Create hardware sequencer */
 	dce112_hw_sequencer_construct(dc);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.h b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.h
index 95a403396219..1f57ebc6f9b4 100644
--- a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.h
@@ -44,7 +44,8 @@ enum dc_status dce112_validate_with_context(
 
 bool dce112_validate_bandwidth(
 	struct dc *dc,
-	struct dc_state *context);
+	struct dc_state *context,
+	bool fast_validate);
 
 enum dc_status dce112_add_stream_to_ctx(
 		struct dc *dc,
diff --git a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c
index 312a0aebf91f..6d49c7143c67 100644
--- a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c
@@ -454,6 +454,28 @@ static const struct resource_caps res_cap = {
 		.num_ddc = 6,
 };
 
+static const struct dc_plane_cap plane_cap = {
+	.type = DC_PLANE_TYPE_DCE_RGB,
+
+	.pixel_format_support = {
+			.argb8888 = true,
+			.nv12 = false,
+			.fp16 = false
+	},
+
+	.max_upscale_factor = {
+			.argb8888 = 16000,
+			.nv12 = 1,
+			.fp16 = 1
+	},
+
+	.max_downscale_factor = {
+			.argb8888 = 250,
+			.nv12 = 1,
+			.fp16 = 1
+	}
+};
+
 static const struct dc_debug_options debug_defaults = {
 		.disable_clock_gate = true,
 };
@@ -1171,6 +1193,9 @@ static bool construct(
 
 	dc->caps.max_planes =  pool->base.pipe_count;
 
+	for (i = 0; i < dc->caps.max_planes; ++i)
+		dc->caps.planes[i] = plane_cap;
+
 	bw_calcs_init(dc->bw_dceip, dc->bw_vbios, dc->ctx->asic_id);
 
 	bw_calcs_data_update_from_pplib(dc);
diff --git a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c
index c109ace96be9..27d0cc394963 100644
--- a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c
@@ -387,6 +387,28 @@ static const struct resource_caps res_cap_83 = {
 		.num_ddc = 2,
 };
 
+static const struct dc_plane_cap plane_cap = {
+	.type = DC_PLANE_TYPE_DCE_RGB,
+
+	.pixel_format_support = {
+			.argb8888 = true,
+			.nv12 = false,
+			.fp16 = false
+	},
+
+	.max_upscale_factor = {
+			.argb8888 = 16000,
+			.nv12 = 1,
+			.fp16 = 1
+	},
+
+	.max_downscale_factor = {
+			.argb8888 = 250,
+			.nv12 = 1,
+			.fp16 = 1
+	}
+};
+
 static const struct dce_dmcu_registers dmcu_regs = {
 		DMCU_DCE80_REG_LIST()
 };
@@ -790,7 +812,8 @@ static void destruct(struct dce110_resource_pool *pool)
 
 bool dce80_validate_bandwidth(
 	struct dc *dc,
-	struct dc_state *context)
+	struct dc_state *context,
+	bool fast_validate)
 {
 	int i;
 	bool at_least_one_pipe = false;
@@ -802,11 +825,11 @@ bool dce80_validate_bandwidth(
 
 	if (at_least_one_pipe) {
 		/* TODO implement when needed but for now hardcode max value*/
-		context->bw.dce.dispclk_khz = 681000;
-		context->bw.dce.yclk_khz = 250000 * MEMORY_TYPE_MULTIPLIER_CZ;
+		context->bw_ctx.bw.dce.dispclk_khz = 681000;
+		context->bw_ctx.bw.dce.yclk_khz = 250000 * MEMORY_TYPE_MULTIPLIER_CZ;
 	} else {
-		context->bw.dce.dispclk_khz = 0;
-		context->bw.dce.yclk_khz = 0;
+		context->bw_ctx.bw.dce.dispclk_khz = 0;
+		context->bw_ctx.bw.dce.yclk_khz = 0;
 	}
 
 	return true;
@@ -1032,6 +1055,10 @@ static bool dce80_construct(
 	}
 
 	dc->caps.max_planes =  pool->base.pipe_count;
+
+	for (i = 0; i < dc->caps.max_planes; ++i)
+		dc->caps.planes[i] = plane_cap;
+
 	dc->caps.disable_dp_clk_share = true;
 
 	if (!resource_construct(num_virtual_links, dc, &pool->base,
@@ -1237,6 +1264,10 @@ static bool dce81_construct(
 	}
 
 	dc->caps.max_planes =  pool->base.pipe_count;
+
+	for (i = 0; i < dc->caps.max_planes; ++i)
+		dc->caps.planes[i] = plane_cap;
+
 	dc->caps.disable_dp_clk_share = true;
 
 	if (!resource_construct(num_virtual_links, dc, &pool->base,
@@ -1438,6 +1469,10 @@ static bool dce83_construct(
 	}
 
 	dc->caps.max_planes =  pool->base.pipe_count;
+
+	for (i = 0; i < dc->caps.max_planes; ++i)
+		dc->caps.planes[i] = plane_cap;
+
 	dc->caps.disable_dp_clk_share = true;
 
 	if (!resource_construct(num_virtual_links, dc, &pool->base,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.c
index afe8c42211cd..2b2de1d913c9 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.c
@@ -43,23 +43,6 @@
 #define DC_LOGGER \
 	clk_mgr->ctx->logger
 
-void dcn1_pplib_apply_display_requirements(
-	struct dc *dc,
-	struct dc_state *context)
-{
-	struct dm_pp_display_configuration *pp_display_cfg = &context->pp_display_cfg;
-
-	pp_display_cfg->min_engine_clock_khz = dc->res_pool->clk_mgr->clks.dcfclk_khz;
-	pp_display_cfg->min_memory_clock_khz = dc->res_pool->clk_mgr->clks.fclk_khz;
-	pp_display_cfg->min_engine_clock_deep_sleep_khz = dc->res_pool->clk_mgr->clks.dcfclk_deep_sleep_khz;
-	pp_display_cfg->min_dcfc_deep_sleep_clock_khz = dc->res_pool->clk_mgr->clks.dcfclk_deep_sleep_khz;
-	pp_display_cfg->min_dcfclock_khz = dc->res_pool->clk_mgr->clks.dcfclk_khz;
-	pp_display_cfg->disp_clk_khz = dc->res_pool->clk_mgr->clks.dispclk_khz;
-	dce110_fill_display_configs(context, pp_display_cfg);
-
-	dm_pp_apply_display_requirements(dc->ctx, pp_display_cfg);
-}
-
 static int dcn1_determine_dppclk_threshold(struct clk_mgr *clk_mgr, struct dc_clocks *new_clocks)
 {
 	bool request_dpp_div = new_clocks->dispclk_khz > new_clocks->dppclk_khz;
@@ -167,11 +150,8 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr,
 {
 	struct dc *dc = clk_mgr->ctx->dc;
 	struct dc_debug_options *debug = &dc->debug;
-	struct dc_clocks *new_clocks = &context->bw.dcn.clk;
-	struct pp_smu_display_requirement_rv *smu_req_cur =
-			&dc->res_pool->pp_smu_req;
-	struct pp_smu_display_requirement_rv smu_req = *smu_req_cur;
-	struct pp_smu_funcs_rv *pp_smu = dc->res_pool->pp_smu;
+	struct dc_clocks *new_clocks = &context->bw_ctx.bw.dcn.clk;
+	struct pp_smu_funcs_rv *pp_smu = NULL;
 	bool send_request_to_increase = false;
 	bool send_request_to_lower = false;
 	int display_count;
@@ -179,7 +159,8 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr,
 	bool enter_display_off = false;
 
 	display_count = get_active_display_cnt(dc, context);
-
+	if (dc->res_pool->pp_smu)
+		pp_smu = &dc->res_pool->pp_smu->rv_funcs;
 	if (display_count == 0)
 		enter_display_off = true;
 
@@ -189,10 +170,8 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr,
 		 * if function pointer not set up, this message is
 		 * sent as part of pplib_apply_display_requirements.
 		 */
-		if (pp_smu->set_display_count)
+		if (pp_smu && pp_smu->set_display_count)
 			pp_smu->set_display_count(&pp_smu->pp_smu, display_count);
-
-		smu_req.display_count = display_count;
 	}
 
 	if (new_clocks->dispclk_khz > clk_mgr->clks.dispclk_khz
@@ -203,7 +182,6 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr,
 
 	if (should_set_clock(safe_to_lower, new_clocks->phyclk_khz, clk_mgr->clks.phyclk_khz)) {
 		clk_mgr->clks.phyclk_khz = new_clocks->phyclk_khz;
-
 		send_request_to_lower = true;
 	}
 
@@ -213,24 +191,18 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr,
 
 	if (should_set_clock(safe_to_lower, new_clocks->fclk_khz, clk_mgr->clks.fclk_khz)) {
 		clk_mgr->clks.fclk_khz = new_clocks->fclk_khz;
-		smu_req.hard_min_fclk_mhz = new_clocks->fclk_khz / 1000;
-
 		send_request_to_lower = true;
 	}
 
 	//DCF Clock
 	if (should_set_clock(safe_to_lower, new_clocks->dcfclk_khz, clk_mgr->clks.dcfclk_khz)) {
 		clk_mgr->clks.dcfclk_khz = new_clocks->dcfclk_khz;
-		smu_req.hard_min_dcefclk_mhz = new_clocks->dcfclk_khz / 1000;
-
 		send_request_to_lower = true;
 	}
 
 	if (should_set_clock(safe_to_lower,
 			new_clocks->dcfclk_deep_sleep_khz, clk_mgr->clks.dcfclk_deep_sleep_khz)) {
 		clk_mgr->clks.dcfclk_deep_sleep_khz = new_clocks->dcfclk_deep_sleep_khz;
-		smu_req.min_deep_sleep_dcefclk_mhz = (new_clocks->dcfclk_deep_sleep_khz + 999) / 1000;
-
 		send_request_to_lower = true;
 	}
 
@@ -239,17 +211,13 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr,
 	 */
 	if (send_request_to_increase) {
 		/*use dcfclk to request voltage*/
-		if (pp_smu->set_hard_min_fclk_by_freq &&
+		if (pp_smu && pp_smu->set_hard_min_fclk_by_freq &&
 				pp_smu->set_hard_min_dcfclk_by_freq &&
 				pp_smu->set_min_deep_sleep_dcfclk) {
 
-			pp_smu->set_hard_min_fclk_by_freq(&pp_smu->pp_smu, smu_req.hard_min_fclk_mhz);
-			pp_smu->set_hard_min_dcfclk_by_freq(&pp_smu->pp_smu, smu_req.hard_min_dcefclk_mhz);
-			pp_smu->set_min_deep_sleep_dcfclk(&pp_smu->pp_smu, smu_req.min_deep_sleep_dcefclk_mhz);
-		} else {
-			if (pp_smu->set_display_requirement)
-				pp_smu->set_display_requirement(&pp_smu->pp_smu, &smu_req);
-			dcn1_pplib_apply_display_requirements(dc, context);
+			pp_smu->set_hard_min_fclk_by_freq(&pp_smu->pp_smu, new_clocks->fclk_khz / 1000);
+			pp_smu->set_hard_min_dcfclk_by_freq(&pp_smu->pp_smu, new_clocks->dcfclk_khz / 1000);
+			pp_smu->set_min_deep_sleep_dcfclk(&pp_smu->pp_smu, (new_clocks->dcfclk_deep_sleep_khz + 999) / 1000);
 		}
 	}
 
@@ -259,27 +227,20 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr,
 			|| new_clocks->dispclk_khz == clk_mgr->clks.dispclk_khz) {
 		dcn1_ramp_up_dispclk_with_dpp(clk_mgr, new_clocks);
 		clk_mgr->clks.dispclk_khz = new_clocks->dispclk_khz;
-
 		send_request_to_lower = true;
 	}
 
 	if (!send_request_to_increase && send_request_to_lower) {
 		/*use dcfclk to request voltage*/
-		if (pp_smu->set_hard_min_fclk_by_freq &&
+		if (pp_smu && pp_smu->set_hard_min_fclk_by_freq &&
 				pp_smu->set_hard_min_dcfclk_by_freq &&
 				pp_smu->set_min_deep_sleep_dcfclk) {
 
-			pp_smu->set_hard_min_fclk_by_freq(&pp_smu->pp_smu, smu_req.hard_min_fclk_mhz);
-			pp_smu->set_hard_min_dcfclk_by_freq(&pp_smu->pp_smu, smu_req.hard_min_dcefclk_mhz);
-			pp_smu->set_min_deep_sleep_dcfclk(&pp_smu->pp_smu, smu_req.min_deep_sleep_dcefclk_mhz);
-		} else {
-			if (pp_smu->set_display_requirement)
-				pp_smu->set_display_requirement(&pp_smu->pp_smu, &smu_req);
-			dcn1_pplib_apply_display_requirements(dc, context);
+			pp_smu->set_hard_min_fclk_by_freq(&pp_smu->pp_smu, new_clocks->fclk_khz / 1000);
+			pp_smu->set_hard_min_dcfclk_by_freq(&pp_smu->pp_smu, new_clocks->dcfclk_khz / 1000);
+			pp_smu->set_min_deep_sleep_dcfclk(&pp_smu->pp_smu, (new_clocks->dcfclk_deep_sleep_khz + 999) / 1000);
 		}
 	}
-
-	*smu_req_cur = smu_req;
 }
 static const struct clk_mgr_funcs dcn1_funcs = {
 	.get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.h
index a995eda443a3..97007cf33665 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.h
@@ -34,10 +34,6 @@ struct clk_bypass {
 	uint32_t dprefclk_bypass;
 };
 
-void dcn1_pplib_apply_display_requirements(
-	struct dc *dc,
-	struct dc_state *context);
-
 struct clk_mgr *dcn1_clk_mgr_create(struct dc_context *ctx);
 
 #endif //__DCN10_CLK_MGR_H__
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c
index cd1ebe57ed59..6f4b24756323 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c
@@ -91,13 +91,6 @@ enum dscl_mode_sel {
 	DSCL_MODE_DSCL_BYPASS = 6
 };
 
-enum gamut_remap_select {
-	GAMUT_REMAP_BYPASS = 0,
-	GAMUT_REMAP_COEFF,
-	GAMUT_REMAP_COMA_COEFF,
-	GAMUT_REMAP_COMB_COEFF
-};
-
 void dpp_read_state(struct dpp *dpp_base,
 		struct dcn_dpp_state *s)
 {
@@ -392,6 +385,10 @@ void dpp1_cnv_setup (
 	default:
 		break;
 	}
+
+	/* Set default color space based on format if none is given. */
+	color_space = input_color_space ? input_color_space : color_space;
+
 	REG_SET(CNVC_SURFACE_PIXEL_FORMAT, 0,
 			CNVC_SURFACE_PIXEL_FORMAT, pixel_format);
 	REG_UPDATE(FORMAT_CONTROL, FORMAT_CONTROL__ALPHA_EN, alpha_en);
@@ -403,7 +400,7 @@ void dpp1_cnv_setup (
 		for (i = 0; i < 12; i++)
 			tbl_entry.regval[i] = input_csc_color_matrix.matrix[i];
 
-		tbl_entry.color_space = input_color_space;
+		tbl_entry.color_space = color_space;
 
 		if (color_space >= COLOR_SPACE_YCBCR601)
 			select = INPUT_CSC_SELECT_ICSC;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_cm.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_cm.c
index 41f0f4c912e7..882bcc5a40f6 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_cm.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_cm.c
@@ -88,13 +88,6 @@ enum dscl_mode_sel {
 	DSCL_MODE_DSCL_BYPASS = 6
 };
 
-enum gamut_remap_select {
-	GAMUT_REMAP_BYPASS = 0,
-	GAMUT_REMAP_COEFF,
-	GAMUT_REMAP_COMA_COEFF,
-	GAMUT_REMAP_COMB_COEFF
-};
-
 static const struct dpp_input_csc_matrix dpp_input_csc_matrix[] = {
 	{COLOR_SPACE_SRGB,
 		{0x2000, 0, 0, 0, 0, 0x2000, 0, 0, 0, 0, 0x2000, 0} },
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c
index c7642e748297..ce21a290bf3e 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c
@@ -406,15 +406,25 @@ void dpp1_dscl_calc_lb_num_partitions(
 		int *num_part_y,
 		int *num_part_c)
 {
+	int lb_memory_size, lb_memory_size_c, lb_memory_size_a, num_partitions_a,
+	lb_bpc, memory_line_size_y, memory_line_size_c, memory_line_size_a;
+
 	int line_size = scl_data->viewport.width < scl_data->recout.width ?
 			scl_data->viewport.width : scl_data->recout.width;
 	int line_size_c = scl_data->viewport_c.width < scl_data->recout.width ?
 			scl_data->viewport_c.width : scl_data->recout.width;
-	int lb_bpc = dpp1_dscl_get_lb_depth_bpc(scl_data->lb_params.depth);
-	int memory_line_size_y = (line_size * lb_bpc + 71) / 72; /* +71 to ceil */
-	int memory_line_size_c = (line_size_c * lb_bpc + 71) / 72; /* +71 to ceil */
-	int memory_line_size_a = (line_size + 5) / 6; /* +5 to ceil */
-	int lb_memory_size, lb_memory_size_c, lb_memory_size_a, num_partitions_a;
+
+	if (line_size == 0)
+		line_size = 1;
+
+	if (line_size_c == 0)
+		line_size_c = 1;
+
+
+	lb_bpc = dpp1_dscl_get_lb_depth_bpc(scl_data->lb_params.depth);
+	memory_line_size_y = (line_size * lb_bpc + 71) / 72; /* +71 to ceil */
+	memory_line_size_c = (line_size_c * lb_bpc + 71) / 72; /* +71 to ceil */
+	memory_line_size_a = (line_size + 5) / 6; /* +5 to ceil */
 
 	if (lb_config == LB_MEMORY_CONFIG_1) {
 		lb_memory_size = 816;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c
index e161ad836812..0db2a6e96fc0 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c
@@ -258,8 +258,9 @@ void hubbub1_wm_change_req_wa(struct hubbub *hubbub)
 {
 	struct dcn10_hubbub *hubbub1 = TO_DCN10_HUBBUB(hubbub);
 
-	REG_UPDATE_SEQ(DCHUBBUB_ARB_WATERMARK_CHANGE_CNTL,
-			DCHUBBUB_ARB_WATERMARK_CHANGE_REQUEST, 0, 1);
+	REG_UPDATE_SEQ_2(DCHUBBUB_ARB_WATERMARK_CHANGE_CNTL,
+			DCHUBBUB_ARB_WATERMARK_CHANGE_REQUEST, 0,
+			DCHUBBUB_ARB_WATERMARK_CHANGE_REQUEST, 1);
 }
 
 void hubbub1_program_watermarks(
@@ -282,7 +283,8 @@ void hubbub1_program_watermarks(
 		hubbub1->watermarks.a.urgent_ns = watermarks->a.urgent_ns;
 		prog_wm_value = convert_and_clamp(watermarks->a.urgent_ns,
 				refclk_mhz, 0x1fffff);
-		REG_WRITE(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, prog_wm_value);
+		REG_SET(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, 0,
+				DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, prog_wm_value);
 
 		DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_A calculated =%d\n"
 			"HW register value = 0x%x\n",
@@ -309,7 +311,8 @@ void hubbub1_program_watermarks(
 			prog_wm_value = convert_and_clamp(
 					watermarks->a.cstate_pstate.cstate_enter_plus_exit_ns,
 					refclk_mhz, 0x1fffff);
-			REG_WRITE(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A, prog_wm_value);
+			REG_SET(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A, 0,
+					DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A, prog_wm_value);
 			DC_LOG_BANDWIDTH_CALCS("SR_ENTER_EXIT_WATERMARK_A calculated =%d\n"
 				"HW register value = 0x%x\n",
 				watermarks->a.cstate_pstate.cstate_enter_plus_exit_ns, prog_wm_value);
@@ -322,7 +325,8 @@ void hubbub1_program_watermarks(
 			prog_wm_value = convert_and_clamp(
 					watermarks->a.cstate_pstate.cstate_exit_ns,
 					refclk_mhz, 0x1fffff);
-			REG_WRITE(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A, prog_wm_value);
+			REG_SET(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A, 0,
+					DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A, prog_wm_value);
 			DC_LOG_BANDWIDTH_CALCS("SR_EXIT_WATERMARK_A calculated =%d\n"
 				"HW register value = 0x%x\n",
 				watermarks->a.cstate_pstate.cstate_exit_ns, prog_wm_value);
@@ -336,7 +340,8 @@ void hubbub1_program_watermarks(
 		prog_wm_value = convert_and_clamp(
 				watermarks->a.cstate_pstate.pstate_change_ns,
 				refclk_mhz, 0x1fffff);
-		REG_WRITE(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, prog_wm_value);
+		REG_SET(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, 0,
+				DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, prog_wm_value);
 		DC_LOG_BANDWIDTH_CALCS("DRAM_CLK_CHANGE_WATERMARK_A calculated =%d\n"
 			"HW register value = 0x%x\n\n",
 			watermarks->a.cstate_pstate.pstate_change_ns, prog_wm_value);
@@ -347,7 +352,8 @@ void hubbub1_program_watermarks(
 		hubbub1->watermarks.b.urgent_ns = watermarks->b.urgent_ns;
 		prog_wm_value = convert_and_clamp(watermarks->b.urgent_ns,
 				refclk_mhz, 0x1fffff);
-		REG_WRITE(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B, prog_wm_value);
+		REG_SET(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B, 0,
+				DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B, prog_wm_value);
 
 		DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_B calculated =%d\n"
 			"HW register value = 0x%x\n",
@@ -374,7 +380,8 @@ void hubbub1_program_watermarks(
 			prog_wm_value = convert_and_clamp(
 					watermarks->b.cstate_pstate.cstate_enter_plus_exit_ns,
 					refclk_mhz, 0x1fffff);
-			REG_WRITE(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B, prog_wm_value);
+			REG_SET(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B, 0,
+					DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B, prog_wm_value);
 			DC_LOG_BANDWIDTH_CALCS("SR_ENTER_EXIT_WATERMARK_B calculated =%d\n"
 				"HW register value = 0x%x\n",
 				watermarks->b.cstate_pstate.cstate_enter_plus_exit_ns, prog_wm_value);
@@ -387,7 +394,8 @@ void hubbub1_program_watermarks(
 			prog_wm_value = convert_and_clamp(
 					watermarks->b.cstate_pstate.cstate_exit_ns,
 					refclk_mhz, 0x1fffff);
-			REG_WRITE(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B, prog_wm_value);
+			REG_SET(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B, 0,
+					DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B, prog_wm_value);
 			DC_LOG_BANDWIDTH_CALCS("SR_EXIT_WATERMARK_B calculated =%d\n"
 				"HW register value = 0x%x\n",
 				watermarks->b.cstate_pstate.cstate_exit_ns, prog_wm_value);
@@ -401,7 +409,8 @@ void hubbub1_program_watermarks(
 		prog_wm_value = convert_and_clamp(
 				watermarks->b.cstate_pstate.pstate_change_ns,
 				refclk_mhz, 0x1fffff);
-		REG_WRITE(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, prog_wm_value);
+		REG_SET(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, 0,
+				DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, prog_wm_value);
 		DC_LOG_BANDWIDTH_CALCS("DRAM_CLK_CHANGE_WATERMARK_B calculated =%d\n"
 			"HW register value = 0x%x\n\n",
 			watermarks->b.cstate_pstate.pstate_change_ns, prog_wm_value);
@@ -412,7 +421,8 @@ void hubbub1_program_watermarks(
 		hubbub1->watermarks.c.urgent_ns = watermarks->c.urgent_ns;
 		prog_wm_value = convert_and_clamp(watermarks->c.urgent_ns,
 				refclk_mhz, 0x1fffff);
-		REG_WRITE(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C, prog_wm_value);
+		REG_SET(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C, 0,
+				DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C, prog_wm_value);
 
 		DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_C calculated =%d\n"
 			"HW register value = 0x%x\n",
@@ -439,7 +449,8 @@ void hubbub1_program_watermarks(
 			prog_wm_value = convert_and_clamp(
 					watermarks->c.cstate_pstate.cstate_enter_plus_exit_ns,
 					refclk_mhz, 0x1fffff);
-			REG_WRITE(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C, prog_wm_value);
+			REG_SET(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C, 0,
+					DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C, prog_wm_value);
 			DC_LOG_BANDWIDTH_CALCS("SR_ENTER_EXIT_WATERMARK_C calculated =%d\n"
 				"HW register value = 0x%x\n",
 				watermarks->c.cstate_pstate.cstate_enter_plus_exit_ns, prog_wm_value);
@@ -452,7 +463,8 @@ void hubbub1_program_watermarks(
 			prog_wm_value = convert_and_clamp(
 					watermarks->c.cstate_pstate.cstate_exit_ns,
 					refclk_mhz, 0x1fffff);
-			REG_WRITE(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C, prog_wm_value);
+			REG_SET(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C, 0,
+					DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C, prog_wm_value);
 			DC_LOG_BANDWIDTH_CALCS("SR_EXIT_WATERMARK_C calculated =%d\n"
 				"HW register value = 0x%x\n",
 				watermarks->c.cstate_pstate.cstate_exit_ns, prog_wm_value);
@@ -466,7 +478,8 @@ void hubbub1_program_watermarks(
 		prog_wm_value = convert_and_clamp(
 				watermarks->c.cstate_pstate.pstate_change_ns,
 				refclk_mhz, 0x1fffff);
-		REG_WRITE(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, prog_wm_value);
+		REG_SET(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, 0,
+				DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, prog_wm_value);
 		DC_LOG_BANDWIDTH_CALCS("DRAM_CLK_CHANGE_WATERMARK_C calculated =%d\n"
 			"HW register value = 0x%x\n\n",
 			watermarks->c.cstate_pstate.pstate_change_ns, prog_wm_value);
@@ -477,7 +490,8 @@ void hubbub1_program_watermarks(
 		hubbub1->watermarks.d.urgent_ns = watermarks->d.urgent_ns;
 		prog_wm_value = convert_and_clamp(watermarks->d.urgent_ns,
 				refclk_mhz, 0x1fffff);
-		REG_WRITE(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D, prog_wm_value);
+		REG_SET(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D, 0,
+				DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D, prog_wm_value);
 
 		DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_D calculated =%d\n"
 			"HW register value = 0x%x\n",
@@ -504,7 +518,8 @@ void hubbub1_program_watermarks(
 			prog_wm_value = convert_and_clamp(
 					watermarks->d.cstate_pstate.cstate_enter_plus_exit_ns,
 					refclk_mhz, 0x1fffff);
-			REG_WRITE(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D, prog_wm_value);
+			REG_SET(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D, 0,
+					DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D, prog_wm_value);
 			DC_LOG_BANDWIDTH_CALCS("SR_ENTER_EXIT_WATERMARK_D calculated =%d\n"
 				"HW register value = 0x%x\n",
 				watermarks->d.cstate_pstate.cstate_enter_plus_exit_ns, prog_wm_value);
@@ -517,7 +532,8 @@ void hubbub1_program_watermarks(
 			prog_wm_value = convert_and_clamp(
 					watermarks->d.cstate_pstate.cstate_exit_ns,
 					refclk_mhz, 0x1fffff);
-			REG_WRITE(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D, prog_wm_value);
+			REG_SET(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D, 0,
+					DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D, prog_wm_value);
 			DC_LOG_BANDWIDTH_CALCS("SR_EXIT_WATERMARK_D calculated =%d\n"
 				"HW register value = 0x%x\n",
 				watermarks->d.cstate_pstate.cstate_exit_ns, prog_wm_value);
@@ -531,7 +547,8 @@ void hubbub1_program_watermarks(
 		prog_wm_value = convert_and_clamp(
 				watermarks->d.cstate_pstate.pstate_change_ns,
 				refclk_mhz, 0x1fffff);
-		REG_WRITE(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, prog_wm_value);
+		REG_SET(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, 0,
+				DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, prog_wm_value);
 		DC_LOG_BANDWIDTH_CALCS("DRAM_CLK_CHANGE_WATERMARK_D calculated =%d\n"
 			"HW register value = 0x%x\n\n",
 			watermarks->d.cstate_pstate.pstate_change_ns, prog_wm_value);
@@ -866,6 +883,7 @@ static const struct hubbub_funcs hubbub1_funcs = {
 	.dcc_support_pixel_format = hubbub1_dcc_support_pixel_format,
 	.get_dcc_compression_cap = hubbub1_get_dcc_compression_cap,
 	.wm_read_state = hubbub1_wm_read_state,
+	.program_watermarks = hubbub1_program_watermarks,
 };
 
 void hubbub1_construct(struct hubbub *hubbub,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h
index 9cd4a5194154..85811b24a497 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h
@@ -32,18 +32,14 @@
 #define TO_DCN10_HUBBUB(hubbub)\
 	container_of(hubbub, struct dcn10_hubbub, base)
 
-#define HUBHUB_REG_LIST_DCN()\
+#define HUBBUB_REG_LIST_DCN_COMMON()\
 	SR(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A),\
-	SR(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_A),\
 	SR(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A),\
 	SR(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B),\
-	SR(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_B),\
 	SR(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B),\
 	SR(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C),\
-	SR(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_C),\
 	SR(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C),\
 	SR(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D),\
-	SR(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_D),\
 	SR(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D),\
 	SR(DCHUBBUB_ARB_WATERMARK_CHANGE_CNTL),\
 	SR(DCHUBBUB_ARB_DRAM_STATE_CNTL),\
@@ -54,6 +50,12 @@
 	SR(DCHUBBUB_TEST_DEBUG_DATA),\
 	SR(DCHUBBUB_SOFT_RESET)
 
+#define HUBBUB_VM_REG_LIST() \
+	SR(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_A),\
+	SR(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_B),\
+	SR(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_C),\
+	SR(DCHUBBUB_ARB_PTE_META_URGENCY_WATERMARK_D)
+
 #define HUBBUB_SR_WATERMARK_REG_LIST()\
 	SR(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A),\
 	SR(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A),\
@@ -65,7 +67,8 @@
 	SR(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D)
 
 #define HUBBUB_REG_LIST_DCN10(id)\
-	HUBHUB_REG_LIST_DCN(), \
+	HUBBUB_REG_LIST_DCN_COMMON(), \
+	HUBBUB_VM_REG_LIST(), \
 	HUBBUB_SR_WATERMARK_REG_LIST(), \
 	SR(DCHUBBUB_SDPIF_FB_TOP),\
 	SR(DCHUBBUB_SDPIF_FB_BASE),\
@@ -122,8 +125,7 @@ struct dcn_hubbub_registers {
 #define HUBBUB_SF(reg_name, field_name, post_fix)\
 	.field_name = reg_name ## __ ## field_name ## post_fix
 
-
-#define HUBBUB_MASK_SH_LIST_DCN(mask_sh)\
+#define HUBBUB_MASK_SH_LIST_DCN_COMMON(mask_sh)\
 		HUBBUB_SF(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, mask_sh), \
 		HUBBUB_SF(DCHUBBUB_SOFT_RESET, DCHUBBUB_GLOBAL_SOFT_RESET, mask_sh), \
 		HUBBUB_SF(DCHUBBUB_ARB_WATERMARK_CHANGE_CNTL, DCHUBBUB_ARB_WATERMARK_CHANGE_REQUEST, mask_sh), \
@@ -133,10 +135,29 @@ struct dcn_hubbub_registers {
 		HUBBUB_SF(DCHUBBUB_ARB_DRAM_STATE_CNTL, DCHUBBUB_ARB_ALLOW_PSTATE_CHANGE_FORCE_VALUE, mask_sh), \
 		HUBBUB_SF(DCHUBBUB_ARB_DRAM_STATE_CNTL, DCHUBBUB_ARB_ALLOW_PSTATE_CHANGE_FORCE_ENABLE, mask_sh), \
 		HUBBUB_SF(DCHUBBUB_ARB_SAT_LEVEL, DCHUBBUB_ARB_SAT_LEVEL, mask_sh), \
-		HUBBUB_SF(DCHUBBUB_ARB_DF_REQ_OUTSTAND, DCHUBBUB_ARB_MIN_REQ_OUTSTAND, mask_sh)
+		HUBBUB_SF(DCHUBBUB_ARB_DF_REQ_OUTSTAND, DCHUBBUB_ARB_MIN_REQ_OUTSTAND, mask_sh), \
+		HUBBUB_SF(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, mask_sh), \
+		HUBBUB_SF(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B, DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B, mask_sh), \
+		HUBBUB_SF(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C, DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C, mask_sh), \
+		HUBBUB_SF(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D, DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D, mask_sh), \
+		HUBBUB_SF(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, mask_sh), \
+		HUBBUB_SF(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, mask_sh), \
+		HUBBUB_SF(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, mask_sh), \
+		HUBBUB_SF(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, mask_sh)
+
+#define HUBBUB_MASK_SH_LIST_STUTTER(mask_sh) \
+		HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A, DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A, mask_sh), \
+		HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B, DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B, mask_sh), \
+		HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C, DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C, mask_sh), \
+		HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D, DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D, mask_sh), \
+		HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A, DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A, mask_sh), \
+		HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B, DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B, mask_sh), \
+		HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C, DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C, mask_sh), \
+		HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D, DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D, mask_sh)
 
 #define HUBBUB_MASK_SH_LIST_DCN10(mask_sh)\
-		HUBBUB_MASK_SH_LIST_DCN(mask_sh), \
+		HUBBUB_MASK_SH_LIST_DCN_COMMON(mask_sh), \
+		HUBBUB_MASK_SH_LIST_STUTTER(mask_sh), \
 		HUBBUB_SF(DCHUBBUB_SDPIF_FB_TOP, SDPIF_FB_TOP, mask_sh), \
 		HUBBUB_SF(DCHUBBUB_SDPIF_FB_BASE, SDPIF_FB_BASE, mask_sh), \
 		HUBBUB_SF(DCHUBBUB_SDPIF_FB_OFFSET, SDPIF_FB_OFFSET, mask_sh), \
@@ -167,15 +188,35 @@ struct dcn_hubbub_registers {
 		type FB_OFFSET;\
 		type AGP_BOT;\
 		type AGP_TOP;\
-		type AGP_BASE
+		type AGP_BASE;\
+		type DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A;\
+		type DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B;\
+		type DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C;\
+		type DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D;\
+		type DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A;\
+		type DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B;\
+		type DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C;\
+		type DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D
+
+#define HUBBUB_STUTTER_REG_FIELD_LIST(type) \
+		type DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A;\
+		type DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B;\
+		type DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C;\
+		type DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D;\
+		type DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A;\
+		type DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B;\
+		type DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C;\
+		type DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D
 
 
 struct dcn_hubbub_shift {
 	DCN_HUBBUB_REG_FIELD_LIST(uint8_t);
+	HUBBUB_STUTTER_REG_FIELD_LIST(uint8_t);
 };
 
 struct dcn_hubbub_mask {
 	DCN_HUBBUB_REG_FIELD_LIST(uint32_t);
+	HUBBUB_STUTTER_REG_FIELD_LIST(uint32_t);
 };
 
 struct dc;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c
index 0ba68d41b9c3..54b219a710d8 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c
@@ -1178,6 +1178,10 @@ void hubp1_vtg_sel(struct hubp *hubp, uint32_t otg_inst)
 	REG_UPDATE(DCHUBP_CNTL, HUBP_VTG_SEL, otg_inst);
 }
 
+void hubp1_init(struct hubp *hubp)
+{
+	//do nothing
+}
 static const struct hubp_funcs dcn10_hubp_funcs = {
 	.hubp_program_surface_flip_and_addr =
 			hubp1_program_surface_flip_and_addr,
@@ -1201,7 +1205,7 @@ static const struct hubp_funcs dcn10_hubp_funcs = {
 	.hubp_clear_underflow = hubp1_clear_underflow,
 	.hubp_disable_control =  hubp1_disable_control,
 	.hubp_get_underflow_status = hubp1_get_underflow_status,
-
+	.hubp_init = hubp1_init,
 };
 
 /*****************************************/
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h
index a6d6dfe00617..99d2b7e2a578 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h
@@ -34,6 +34,7 @@
 #define HUBP_REG_LIST_DCN(id)\
 	SRI(DCHUBP_CNTL, HUBP, id),\
 	SRI(HUBPREQ_DEBUG_DB, HUBP, id),\
+	SRI(HUBPREQ_DEBUG, HUBP, id),\
 	SRI(DCSURF_ADDR_CONFIG, HUBP, id),\
 	SRI(DCSURF_TILING_CONFIG, HUBP, id),\
 	SRI(DCSURF_SURFACE_PITCH, HUBPREQ, id),\
@@ -138,6 +139,7 @@
 #define HUBP_COMMON_REG_VARIABLE_LIST \
 	uint32_t DCHUBP_CNTL; \
 	uint32_t HUBPREQ_DEBUG_DB; \
+	uint32_t HUBPREQ_DEBUG; \
 	uint32_t DCSURF_ADDR_CONFIG; \
 	uint32_t DCSURF_TILING_CONFIG; \
 	uint32_t DCSURF_SURFACE_PITCH; \
@@ -247,7 +249,7 @@
 	.field_name = reg_name ## __ ## field_name ## post_fix
 
 /* Mask/shift struct generation macro for all ASICs (including those with reduced functionality) */
-#define HUBP_MASK_SH_LIST_DCN(mask_sh)\
+#define HUBP_MASK_SH_LIST_DCN_COMMON(mask_sh)\
 	HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_BLANK_EN, mask_sh),\
 	HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_TTU_DISABLE, mask_sh),\
 	HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_UNDERFLOW_STATUS, mask_sh),\
@@ -331,7 +333,6 @@
 	HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG, META_CHUNK_SIZE, mask_sh),\
 	HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG, MIN_META_CHUNK_SIZE, mask_sh),\
 	HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG, DPTE_GROUP_SIZE, mask_sh),\
-	HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG, MPTE_GROUP_SIZE, mask_sh),\
 	HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG, SWATH_HEIGHT, mask_sh),\
 	HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG, PTE_ROW_HEIGHT_LINEAR, mask_sh),\
 	HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG_C, CHUNK_SIZE_C, mask_sh),\
@@ -339,7 +340,6 @@
 	HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG_C, META_CHUNK_SIZE_C, mask_sh),\
 	HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG_C, MIN_META_CHUNK_SIZE_C, mask_sh),\
 	HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG_C, DPTE_GROUP_SIZE_C, mask_sh),\
-	HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG_C, MPTE_GROUP_SIZE_C, mask_sh),\
 	HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG_C, SWATH_HEIGHT_C, mask_sh),\
 	HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG_C, PTE_ROW_HEIGHT_LINEAR_C, mask_sh),\
 	HUBP_SF(HUBPREQ0_BLANK_OFFSET_0, REFCYC_H_BLANK_END, mask_sh),\
@@ -373,6 +373,11 @@
 	HUBP_SF(HUBPREQ0_DCN_SURF0_TTU_CNTL1, REFCYC_PER_REQ_DELIVERY_PRE, mask_sh),\
 	HUBP_SF(HUBP0_HUBP_CLK_CNTL, HUBP_CLOCK_ENABLE, mask_sh)
 
+#define HUBP_MASK_SH_LIST_DCN(mask_sh)\
+	HUBP_MASK_SH_LIST_DCN_COMMON(mask_sh),\
+	HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG, MPTE_GROUP_SIZE, mask_sh),\
+	HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG_C, MPTE_GROUP_SIZE_C, mask_sh)
+
 /* Mask/shift struct generation macro for ASICs with VM */
 #define HUBP_MASK_SH_LIST_DCN_VM(mask_sh)\
 	HUBP_SF(HUBPREQ0_NOM_PARAMETERS_0, DST_Y_PER_PTE_ROW_NOM_L, mask_sh),\
@@ -595,6 +600,9 @@
 	type AGP_BASE;\
 	type AGP_BOT;\
 	type AGP_TOP;\
+	type DCN_VM_SYSTEM_APERTURE_DEFAULT_SYSTEM;\
+	type DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB;\
+	type DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB;\
 	/* todo:  get these from GVM instead of reading registers ourselves */\
 	type PAGE_DIRECTORY_ENTRY_HI32;\
 	type PAGE_DIRECTORY_ENTRY_LO32;\
@@ -743,4 +751,6 @@ enum cursor_pitch hubp1_get_cursor_pitch(unsigned int pitch);
 void hubp1_vready_workaround(struct hubp *hubp,
 		struct _vcs_dpi_display_pipe_dest_params_st *pipe_dest);
 
+void hubp1_init(struct hubp *hubp);
+
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
index d1a8f1c302a9..33d311cea28c 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
@@ -65,7 +65,7 @@ void print_microsec(struct dc_context *dc_ctx,
 	struct dc_log_buffer_ctx *log_ctx,
 	uint32_t ref_cycle)
 {
-	const uint32_t ref_clk_mhz = dc_ctx->dc->res_pool->ref_clock_inKhz / 1000;
+	const uint32_t ref_clk_mhz = dc_ctx->dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000;
 	static const unsigned int frac = 1000;
 	uint32_t us_x10 = (ref_cycle * frac) / ref_clk_mhz;
 
@@ -345,13 +345,13 @@ void dcn10_log_hw_state(struct dc *dc,
 
 	DTN_INFO("\nCALCULATED Clocks: dcfclk_khz:%d  dcfclk_deep_sleep_khz:%d  dispclk_khz:%d\n"
 		"dppclk_khz:%d  max_supported_dppclk_khz:%d  fclk_khz:%d  socclk_khz:%d\n\n",
-			dc->current_state->bw.dcn.clk.dcfclk_khz,
-			dc->current_state->bw.dcn.clk.dcfclk_deep_sleep_khz,
-			dc->current_state->bw.dcn.clk.dispclk_khz,
-			dc->current_state->bw.dcn.clk.dppclk_khz,
-			dc->current_state->bw.dcn.clk.max_supported_dppclk_khz,
-			dc->current_state->bw.dcn.clk.fclk_khz,
-			dc->current_state->bw.dcn.clk.socclk_khz);
+			dc->current_state->bw_ctx.bw.dcn.clk.dcfclk_khz,
+			dc->current_state->bw_ctx.bw.dcn.clk.dcfclk_deep_sleep_khz,
+			dc->current_state->bw_ctx.bw.dcn.clk.dispclk_khz,
+			dc->current_state->bw_ctx.bw.dcn.clk.dppclk_khz,
+			dc->current_state->bw_ctx.bw.dcn.clk.max_supported_dppclk_khz,
+			dc->current_state->bw_ctx.bw.dcn.clk.fclk_khz,
+			dc->current_state->bw_ctx.bw.dcn.clk.socclk_khz);
 
 	log_mpc_crc(dc, log_ctx);
 
@@ -714,7 +714,7 @@ static enum dc_status dcn10_enable_stream_timing(
 	return DC_OK;
 }
 
-static void reset_back_end_for_pipe(
+static void dcn10_reset_back_end_for_pipe(
 		struct dc *dc,
 		struct pipe_ctx *pipe_ctx,
 		struct dc_state *context)
@@ -889,22 +889,23 @@ void hwss1_plane_atomic_disconnect(struct dc *dc, struct pipe_ctx *pipe_ctx)
 		dcn10_verify_allow_pstate_change_high(dc);
 }
 
-static void plane_atomic_power_down(struct dc *dc, struct pipe_ctx *pipe_ctx)
+static void plane_atomic_power_down(struct dc *dc,
+		struct dpp *dpp,
+		struct hubp *hubp)
 {
 	struct dce_hwseq *hws = dc->hwseq;
-	struct dpp *dpp = pipe_ctx->plane_res.dpp;
 	DC_LOGGER_INIT(dc->ctx->logger);
 
 	if (REG(DC_IP_REQUEST_CNTL)) {
 		REG_SET(DC_IP_REQUEST_CNTL, 0,
 				IP_REQUEST_EN, 1);
 		dpp_pg_control(hws, dpp->inst, false);
-		hubp_pg_control(hws, pipe_ctx->plane_res.hubp->inst, false);
+		hubp_pg_control(hws, hubp->inst, false);
 		dpp->funcs->dpp_reset(dpp);
 		REG_SET(DC_IP_REQUEST_CNTL, 0,
 				IP_REQUEST_EN, 0);
 		DC_LOG_DEBUG(
-				"Power gated front end %d\n", pipe_ctx->pipe_idx);
+				"Power gated front end %d\n", hubp->inst);
 	}
 }
 
@@ -931,7 +932,9 @@ static void plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx)
 	hubp->power_gated = true;
 	dc->optimized_required = false; /* We're powering off, no need to optimize */
 
-	plane_atomic_power_down(dc, pipe_ctx);
+	plane_atomic_power_down(dc,
+			pipe_ctx->plane_res.dpp,
+			pipe_ctx->plane_res.hubp);
 
 	pipe_ctx->stream = NULL;
 	memset(&pipe_ctx->stream_res, 0, sizeof(pipe_ctx->stream_res));
@@ -976,16 +979,14 @@ static void dcn10_init_pipes(struct dc *dc, struct dc_state *context)
 		 * to non-preferred front end. If pipe_ctx->stream is not NULL,
 		 * we will use the pipe, so don't disable
 		 */
-		if (pipe_ctx->stream != NULL)
+		if (pipe_ctx->stream != NULL && can_apply_seamless_boot)
 			continue;
 
-		if (tg->funcs->is_tg_enabled(tg))
-			tg->funcs->lock(tg);
-
 		/* Blank controller using driver code instead of
 		 * command table.
 		 */
 		if (tg->funcs->is_tg_enabled(tg)) {
+			tg->funcs->lock(tg);
 			tg->funcs->set_blank(tg, true);
 			hwss_wait_for_blank_complete(tg);
 		}
@@ -1001,16 +1002,19 @@ static void dcn10_init_pipes(struct dc *dc, struct dc_state *context)
 		struct dpp *dpp = dc->res_pool->dpps[i];
 		struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
 
-		// W/A for issue with dc_post_update_surfaces_to_stream
-		hubp->power_gated = true;
-
 		/* There is assumption that pipe_ctx is not mapping irregularly
 		 * to non-preferred front end. If pipe_ctx->stream is not NULL,
 		 * we will use the pipe, so don't disable
 		 */
-		if (pipe_ctx->stream != NULL)
+		if (can_apply_seamless_boot &&
+			pipe_ctx->stream != NULL &&
+			pipe_ctx->stream_res.tg->funcs->is_tg_enabled(
+				pipe_ctx->stream_res.tg))
 			continue;
 
+		/* Disable on the current state so the new one isn't cleared. */
+		pipe_ctx = &dc->current_state->res_ctx.pipe_ctx[i];
+
 		dpp->funcs->dpp_reset(dpp);
 
 		pipe_ctx->stream_res.tg = tg;
@@ -1108,6 +1112,25 @@ static void dcn10_init_hw(struct dc *dc)
 			link->link_status.link_active = true;
 	}
 
+	/* If taking control over from VBIOS, we may want to optimize our first
+	 * mode set, so we need to skip powering down pipes until we know which
+	 * pipes we want to use.
+	 * Otherwise, if taking control is not possible, we need to power
+	 * everything down.
+	 */
+	if (dcb->funcs->is_accelerated_mode(dcb) || dc->config.power_down_display_on_boot) {
+		for (i = 0; i < dc->res_pool->pipe_count; i++) {
+			struct hubp *hubp = dc->res_pool->hubps[i];
+			struct dpp *dpp = dc->res_pool->dpps[i];
+
+			hubp->funcs->hubp_init(hubp);
+			dc->res_pool->opps[i]->mpc_tree_params.opp_id = dc->res_pool->opps[i]->inst;
+			plane_atomic_power_down(dc, dpp, hubp);
+		}
+
+		apply_DEGVIDCN10_253_wa(dc);
+	}
+
 	for (i = 0; i < dc->res_pool->audio_count; i++) {
 		struct audio *audio = dc->res_pool->audios[i];
 
@@ -1137,12 +1160,9 @@ static void dcn10_init_hw(struct dc *dc)
 	enable_power_gating_plane(dc->hwseq, true);
 
 	memset(&dc->res_pool->clk_mgr->clks, 0, sizeof(dc->res_pool->clk_mgr->clks));
-
-	if (dc->hwss.init_pipes)
-		dc->hwss.init_pipes(dc, dc->current_state);
 }
 
-static void reset_hw_ctx_wrap(
+static void dcn10_reset_hw_ctx_wrap(
 		struct dc *dc,
 		struct dc_state *context)
 {
@@ -1164,10 +1184,9 @@ static void reset_hw_ctx_wrap(
 				pipe_need_reprogram(pipe_ctx_old, pipe_ctx)) {
 			struct clock_source *old_clk = pipe_ctx_old->clock_source;
 
-			reset_back_end_for_pipe(dc, pipe_ctx_old, dc->current_state);
-			if (dc->hwss.enable_stream_gating) {
+			dcn10_reset_back_end_for_pipe(dc, pipe_ctx_old, dc->current_state);
+			if (dc->hwss.enable_stream_gating)
 				dc->hwss.enable_stream_gating(dc, pipe_ctx);
-			}
 			if (old_clk)
 				old_clk->funcs->cs_power_down(old_clk);
 		}
@@ -1836,7 +1855,7 @@ void dcn10_get_hdr_visual_confirm_color(
 
 	switch (top_pipe_ctx->plane_res.scl_data.format) {
 	case PIXEL_FORMAT_ARGB2101010:
-		if (top_pipe_ctx->stream->out_transfer_func->tf == TRANSFER_FUNCTION_UNITY) {
+		if (top_pipe_ctx->stream->out_transfer_func->tf == TRANSFER_FUNCTION_PQ) {
 			/* HDR10, ARGB2101010 - set boarder color to red */
 			color->color_r_cr = color_value;
 		}
@@ -1931,7 +1950,7 @@ static void update_dpp(struct dpp *dpp, struct dc_plane_state *plane_state)
 			plane_state->format,
 			EXPANSION_MODE_ZERO,
 			plane_state->input_csc_color_matrix,
-			COLOR_SPACE_YCBCR601_LIMITED);
+			plane_state->color_space);
 
 	//set scale and bias registers
 	build_prescale_params(&bns_params, plane_state);
@@ -2051,7 +2070,7 @@ void update_dchubp_dpp(
 	 * divided by 2
 	 */
 	if (plane_state->update_flags.bits.full_update) {
-		bool should_divided_by_2 = context->bw.dcn.clk.dppclk_khz <=
+		bool should_divided_by_2 = context->bw_ctx.bw.dcn.clk.dppclk_khz <=
 				dc->res_pool->clk_mgr->clks.dispclk_khz / 2;
 
 		dpp->funcs->dpp_dppclk_control(
@@ -2120,6 +2139,9 @@ void update_dchubp_dpp(
 	if (pipe_ctx->stream->cursor_attributes.address.quad_part != 0) {
 		dc->hwss.set_cursor_position(pipe_ctx);
 		dc->hwss.set_cursor_attribute(pipe_ctx);
+
+		if (dc->hwss.set_cursor_sdr_white_level)
+			dc->hwss.set_cursor_sdr_white_level(pipe_ctx);
 	}
 
 	if (plane_state->update_flags.bits.full_update) {
@@ -2310,6 +2332,7 @@ static void dcn10_apply_ctx_for_surface(
 	int i;
 	struct timing_generator *tg;
 	bool removed_pipe[4] = { false };
+	bool interdependent_update = false;
 	struct pipe_ctx *top_pipe_to_program =
 			find_top_pipe_for_stream(dc, context, stream);
 	DC_LOGGER_INIT(dc->ctx->logger);
@@ -2319,7 +2342,13 @@ static void dcn10_apply_ctx_for_surface(
 
 	tg = top_pipe_to_program->stream_res.tg;
 
-	dcn10_pipe_control_lock(dc, top_pipe_to_program, true);
+	interdependent_update = top_pipe_to_program->plane_state &&
+		top_pipe_to_program->plane_state->update_flags.bits.full_update;
+
+	if (interdependent_update)
+		lock_all_pipes(dc, context, true);
+	else
+		dcn10_pipe_control_lock(dc, top_pipe_to_program, true);
 
 	if (num_planes == 0) {
 		/* OTG blank before remove all front end */
@@ -2339,15 +2368,9 @@ static void dcn10_apply_ctx_for_surface(
 		 */
 		if (pipe_ctx->plane_state && !old_pipe_ctx->plane_state) {
 			if (old_pipe_ctx->stream_res.tg == tg &&
-				old_pipe_ctx->plane_res.hubp &&
-				old_pipe_ctx->plane_res.hubp->opp_id != 0xf) {
+			    old_pipe_ctx->plane_res.hubp &&
+			    old_pipe_ctx->plane_res.hubp->opp_id != 0xf)
 				dcn10_disable_plane(dc, old_pipe_ctx);
-				/*
-				 * power down fe will unlock when calling reset, need
-				 * to lock it back here. Messy, need rework.
-				 */
-				pipe_ctx->stream_res.tg->funcs->lock(pipe_ctx->stream_res.tg);
-			}
 		}
 
 		if ((!pipe_ctx->plane_state ||
@@ -2366,29 +2389,25 @@ static void dcn10_apply_ctx_for_surface(
 	if (num_planes > 0)
 		program_all_pipe_in_tree(dc, top_pipe_to_program, context);
 
-	dcn10_pipe_control_lock(dc, top_pipe_to_program, false);
-
-	if (top_pipe_to_program->plane_state &&
-			top_pipe_to_program->plane_state->update_flags.bits.full_update)
+	if (interdependent_update)
 		for (i = 0; i < dc->res_pool->pipe_count; i++) {
 			struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
-			tg = pipe_ctx->stream_res.tg;
 			/* Skip inactive pipes and ones already updated */
-			if (!pipe_ctx->stream || pipe_ctx->stream == stream
-					|| !pipe_ctx->plane_state
-					|| !tg->funcs->is_tg_enabled(tg))
+			if (!pipe_ctx->stream || pipe_ctx->stream == stream ||
+			    !pipe_ctx->plane_state || !tg->funcs->is_tg_enabled(tg))
 				continue;
 
-			tg->funcs->lock(tg);
-
 			pipe_ctx->plane_res.hubp->funcs->hubp_setup_interdependent(
 				pipe_ctx->plane_res.hubp,
 				&pipe_ctx->dlg_regs,
 				&pipe_ctx->ttu_regs);
-
-			tg->funcs->unlock(tg);
 		}
 
+	if (interdependent_update)
+		lock_all_pipes(dc, context, false);
+	else
+		dcn10_pipe_control_lock(dc, top_pipe_to_program, false);
+
 	if (num_planes == 0)
 		false_optc_underflow_wa(dc, stream, tg);
 
@@ -2420,12 +2439,14 @@ static void dcn10_prepare_bandwidth(
 		struct dc *dc,
 		struct dc_state *context)
 {
+	struct hubbub *hubbub = dc->res_pool->hubbub;
+
 	if (dc->debug.sanity_checks)
 		dcn10_verify_allow_pstate_change_high(dc);
 
 	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
 		if (context->stream_count == 0)
-			context->bw.dcn.clk.phyclk_khz = 0;
+			context->bw_ctx.bw.dcn.clk.phyclk_khz = 0;
 
 		dc->res_pool->clk_mgr->funcs->update_clocks(
 				dc->res_pool->clk_mgr,
@@ -2433,9 +2454,9 @@ static void dcn10_prepare_bandwidth(
 				false);
 	}
 
-	hubbub1_program_watermarks(dc->res_pool->hubbub,
-			&context->bw.dcn.watermarks,
-			dc->res_pool->ref_clock_inKhz / 1000,
+	hubbub->funcs->program_watermarks(hubbub,
+			&context->bw_ctx.bw.dcn.watermarks,
+			dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000,
 			true);
 	dcn10_stereo_hw_frame_pack_wa(dc, context);
 
@@ -2450,12 +2471,14 @@ static void dcn10_optimize_bandwidth(
 		struct dc *dc,
 		struct dc_state *context)
 {
+	struct hubbub *hubbub = dc->res_pool->hubbub;
+
 	if (dc->debug.sanity_checks)
 		dcn10_verify_allow_pstate_change_high(dc);
 
 	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
 		if (context->stream_count == 0)
-			context->bw.dcn.clk.phyclk_khz = 0;
+			context->bw_ctx.bw.dcn.clk.phyclk_khz = 0;
 
 		dc->res_pool->clk_mgr->funcs->update_clocks(
 				dc->res_pool->clk_mgr,
@@ -2463,9 +2486,9 @@ static void dcn10_optimize_bandwidth(
 				true);
 	}
 
-	hubbub1_program_watermarks(dc->res_pool->hubbub,
-			&context->bw.dcn.watermarks,
-			dc->res_pool->ref_clock_inKhz / 1000,
+	hubbub->funcs->program_watermarks(hubbub,
+			&context->bw_ctx.bw.dcn.watermarks,
+			dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000,
 			true);
 	dcn10_stereo_hw_frame_pack_wa(dc, context);
 
@@ -2654,7 +2677,7 @@ static void dcn10_update_pending_status(struct pipe_ctx *pipe_ctx)
 	flip_pending = pipe_ctx->plane_res.hubp->funcs->hubp_is_flip_pending(
 					pipe_ctx->plane_res.hubp);
 
-	plane_state->status.is_flip_pending = flip_pending;
+	plane_state->status.is_flip_pending = plane_state->status.is_flip_pending || flip_pending;
 
 	if (!flip_pending)
 		plane_state->status.current_address = plane_state->status.requested_address;
@@ -2685,16 +2708,22 @@ static void dcn10_set_cursor_position(struct pipe_ctx *pipe_ctx)
 	struct dpp *dpp = pipe_ctx->plane_res.dpp;
 	struct dc_cursor_mi_param param = {
 		.pixel_clk_khz = pipe_ctx->stream->timing.pix_clk_100hz / 10,
-		.ref_clk_khz = pipe_ctx->stream->ctx->dc->res_pool->ref_clock_inKhz,
+		.ref_clk_khz = pipe_ctx->stream->ctx->dc->res_pool->ref_clocks.dchub_ref_clock_inKhz,
 		.viewport = pipe_ctx->plane_res.scl_data.viewport,
 		.h_scale_ratio = pipe_ctx->plane_res.scl_data.ratios.horz,
 		.v_scale_ratio = pipe_ctx->plane_res.scl_data.ratios.vert,
 		.rotation = pipe_ctx->plane_state->rotation,
 		.mirror = pipe_ctx->plane_state->horizontal_mirror
 	};
+	uint32_t x_plane = pipe_ctx->plane_state->dst_rect.x;
+	uint32_t y_plane = pipe_ctx->plane_state->dst_rect.y;
+	uint32_t x_offset = min(x_plane, pos_cpy.x);
+	uint32_t y_offset = min(y_plane, pos_cpy.y);
 
-	pos_cpy.x_hotspot += pipe_ctx->plane_state->dst_rect.x;
-	pos_cpy.y_hotspot += pipe_ctx->plane_state->dst_rect.y;
+	pos_cpy.x -= x_offset;
+	pos_cpy.y -= y_offset;
+	pos_cpy.x_hotspot += (x_plane - x_offset);
+	pos_cpy.y_hotspot += (y_plane - y_offset);
 
 	if (pipe_ctx->plane_state->address.type
 			== PLN_ADDR_TYPE_VIDEO_PROGRESSIVE)
@@ -2789,6 +2818,33 @@ int get_vupdate_offset_from_vsync(struct pipe_ctx *pipe_ctx)
 	return vertical_line_start;
 }
 
+void lock_all_pipes(struct dc *dc,
+	struct dc_state *context,
+	bool lock)
+{
+	struct pipe_ctx *pipe_ctx;
+	struct timing_generator *tg;
+	int i;
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		pipe_ctx = &context->res_ctx.pipe_ctx[i];
+		tg = pipe_ctx->stream_res.tg;
+		/*
+		 * Only lock the top pipe's tg to prevent redundant
+		 * (un)locking. Also skip if pipe is disabled.
+		 */
+		if (pipe_ctx->top_pipe ||
+		    !pipe_ctx->stream || !pipe_ctx->plane_state ||
+		    !tg->funcs->is_tg_enabled(tg))
+			continue;
+
+		if (lock)
+			tg->funcs->lock(tg);
+		else
+			tg->funcs->unlock(tg);
+	}
+}
+
 static void calc_vupdate_position(
 		struct pipe_ctx *pipe_ctx,
 		uint32_t *start_line,
@@ -2882,6 +2938,29 @@ static void dcn10_setup_vupdate_interrupt(struct pipe_ctx *pipe_ctx)
 		tg->funcs->setup_vertical_interrupt2(tg, start_line);
 }
 
+static void dcn10_unblank_stream(struct pipe_ctx *pipe_ctx,
+		struct dc_link_settings *link_settings)
+{
+	struct encoder_unblank_param params = { { 0 } };
+	struct dc_stream_state *stream = pipe_ctx->stream;
+	struct dc_link *link = stream->link;
+
+	/* only 3 items below are used by unblank */
+	params.timing = pipe_ctx->stream->timing;
+
+	params.link_settings.link_rate = link_settings->link_rate;
+
+	if (dc_is_dp_signal(pipe_ctx->stream->signal)) {
+		if (params.timing.pixel_encoding == PIXEL_ENCODING_YCBCR420)
+			params.timing.pix_clk_100hz /= 2;
+		pipe_ctx->stream_res.stream_enc->funcs->dp_unblank(pipe_ctx->stream_res.stream_enc, &params);
+	}
+
+	if (link->local_sink && link->local_sink->sink_signal == SIGNAL_TYPE_EDP) {
+		link->dc->hwss.edp_backlight_control(link, true);
+	}
+}
+
 static const struct hw_sequencer_funcs dcn10_funcs = {
 	.program_gamut_remap = program_gamut_remap,
 	.init_hw = dcn10_init_hw,
@@ -2903,7 +2982,7 @@ static const struct hw_sequencer_funcs dcn10_funcs = {
 	.update_info_frame = dce110_update_info_frame,
 	.enable_stream = dce110_enable_stream,
 	.disable_stream = dce110_disable_stream,
-	.unblank_stream = dce110_unblank_stream,
+	.unblank_stream = dcn10_unblank_stream,
 	.blank_stream = dce110_blank_stream,
 	.enable_audio_stream = dce110_enable_audio_stream,
 	.disable_audio_stream = dce110_disable_audio_stream,
@@ -2913,7 +2992,7 @@ static const struct hw_sequencer_funcs dcn10_funcs = {
 	.pipe_control_lock = dcn10_pipe_control_lock,
 	.prepare_bandwidth = dcn10_prepare_bandwidth,
 	.optimize_bandwidth = dcn10_optimize_bandwidth,
-	.reset_hw_ctx_wrap = reset_hw_ctx_wrap,
+	.reset_hw_ctx_wrap = dcn10_reset_hw_ctx_wrap,
 	.enable_stream_timing = dcn10_enable_stream_timing,
 	.set_drr = set_drr,
 	.get_position = get_position,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h
index 6d66084df55f..4b3b27a5d23b 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h
@@ -83,4 +83,8 @@ struct pipe_ctx *find_top_pipe_for_stream(
 
 int get_vupdate_offset_from_vsync(struct pipe_ctx *pipe_ctx);
 
+void lock_all_pipes(struct dc *dc,
+	struct dc_state *context,
+	bool lock);
+
 #endif /* __DC_HWSS_DCN10_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c
index 98f41d250978..991622da9ed5 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c
@@ -77,7 +77,7 @@ static unsigned int dcn10_get_hubbub_state(struct dc *dc, char *pBuf, unsigned i
 	unsigned int chars_printed = 0;
 	unsigned int remaining_buffer = bufSize;
 
-	const uint32_t ref_clk_mhz = dc_ctx->dc->res_pool->ref_clock_inKhz / 1000;
+	const uint32_t ref_clk_mhz = dc_ctx->dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000;
 	static const unsigned int frac = 1000;
 
 	memset(&wm, 0, sizeof(struct dcn_hubbub_wm));
@@ -115,7 +115,7 @@ static unsigned int dcn10_get_hubp_states(struct dc *dc, char *pBuf, unsigned in
 	unsigned int chars_printed = 0;
 	unsigned int remaining_buffer = bufSize;
 
-	const uint32_t ref_clk_mhz = dc_ctx->dc->res_pool->ref_clock_inKhz / 1000;
+	const uint32_t ref_clk_mhz = dc_ctx->dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000;
 	static const unsigned int frac = 1000;
 
 	if (invarOnly)
@@ -472,12 +472,12 @@ static unsigned int dcn10_get_clock_states(struct dc *dc, char *pBuf, unsigned i
 	chars_printed = snprintf_count(pBuf, bufSize, "dcfclk,dcfclk_deep_sleep,dispclk,"
 		"dppclk,fclk,socclk\n"
 		"%d,%d,%d,%d,%d,%d\n",
-		dc->current_state->bw.dcn.clk.dcfclk_khz,
-		dc->current_state->bw.dcn.clk.dcfclk_deep_sleep_khz,
-		dc->current_state->bw.dcn.clk.dispclk_khz,
-		dc->current_state->bw.dcn.clk.dppclk_khz,
-		dc->current_state->bw.dcn.clk.fclk_khz,
-		dc->current_state->bw.dcn.clk.socclk_khz);
+		dc->current_state->bw_ctx.bw.dcn.clk.dcfclk_khz,
+		dc->current_state->bw_ctx.bw.dcn.clk.dcfclk_deep_sleep_khz,
+		dc->current_state->bw_ctx.bw.dcn.clk.dispclk_khz,
+		dc->current_state->bw_ctx.bw.dcn.clk.dppclk_khz,
+		dc->current_state->bw_ctx.bw.dcn.clk.fclk_khz,
+		dc->current_state->bw_ctx.bw.dcn.clk.socclk_khz);
 
 	remaining_buffer -= chars_printed;
 	pBuf += chars_printed;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c
index a9db372688ff..0126a44ba012 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c
@@ -1304,7 +1304,6 @@ void dcn10_link_encoder_connect_dig_be_to_fe(
 #define HPD_REG_UPDATE_N(reg_name, n, ...)	\
 		generic_reg_update_ex(CTX, \
 				HPD_REG(reg_name), \
-				HPD_REG_READ(reg_name), \
 				n, __VA_ARGS__)
 
 #define HPD_REG_UPDATE(reg_name, field, val)	\
@@ -1337,7 +1336,6 @@ void dcn10_link_encoder_disable_hpd(struct link_encoder *enc)
 #define AUX_REG_UPDATE_N(reg_name, n, ...)	\
 		generic_reg_update_ex(CTX, \
 				AUX_REG(reg_name), \
-				AUX_REG_READ(reg_name), \
 				n, __VA_ARGS__)
 
 #define AUX_REG_UPDATE(reg_name, field, val)	\
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
index 09d74070a49b..7eccb54c421d 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
@@ -516,6 +516,31 @@ static const struct resource_caps rv2_res_cap = {
 };
 #endif
 
+static const struct dc_plane_cap plane_cap = {
+	.type = DC_PLANE_TYPE_DCN_UNIVERSAL,
+	.blends_with_above = true,
+	.blends_with_below = true,
+	.per_pixel_alpha = true,
+
+	.pixel_format_support = {
+			.argb8888 = true,
+			.nv12 = true,
+			.fp16 = true
+	},
+
+	.max_upscale_factor = {
+			.argb8888 = 16000,
+			.nv12 = 16000,
+			.fp16 = 1
+	},
+
+	.max_downscale_factor = {
+			.argb8888 = 250,
+			.nv12 = 250,
+			.fp16 = 1
+	}
+};
+
 static const struct dc_debug_options debug_defaults_drv = {
 		.sanity_checks = true,
 		.disable_dmcu = true,
@@ -848,14 +873,14 @@ void dcn10_clock_source_destroy(struct clock_source **clk_src)
 	*clk_src = NULL;
 }
 
-static struct pp_smu_funcs_rv *dcn10_pp_smu_create(struct dc_context *ctx)
+static struct pp_smu_funcs *dcn10_pp_smu_create(struct dc_context *ctx)
 {
-	struct pp_smu_funcs_rv *pp_smu = kzalloc(sizeof(*pp_smu), GFP_KERNEL);
+	struct pp_smu_funcs *pp_smu = kzalloc(sizeof(*pp_smu), GFP_KERNEL);
 
 	if (!pp_smu)
 		return pp_smu;
 
-	dm_pp_get_funcs_rv(ctx, pp_smu);
+	dm_pp_get_funcs(ctx, pp_smu);
 	return pp_smu;
 }
 
@@ -865,10 +890,7 @@ static void destruct(struct dcn10_resource_pool *pool)
 
 	for (i = 0; i < pool->base.stream_enc_count; i++) {
 		if (pool->base.stream_enc[i] != NULL) {
-			/* TODO: free dcn version of stream encoder once implemented
-			 * rather than using virtual stream encoder
-			 */
-			kfree(pool->base.stream_enc[i]);
+			kfree(DCN10STRENC_FROM_STRENC(pool->base.stream_enc[i]));
 			pool->base.stream_enc[i] = NULL;
 		}
 	}
@@ -921,9 +943,6 @@ static void destruct(struct dcn10_resource_pool *pool)
 		}
 	}
 
-	for (i = 0; i < pool->base.stream_enc_count; i++)
-		kfree(pool->base.stream_enc[i]);
-
 	for (i = 0; i < pool->base.audio_count; i++) {
 		if (pool->base.audios[i])
 			dce_aud_destroy(&pool->base.audios[i]);
@@ -1078,7 +1097,7 @@ static struct pipe_ctx *dcn10_acquire_idle_pipe_for_layer(
 {
 	struct resource_context *res_ctx = &context->res_ctx;
 	struct pipe_ctx *head_pipe = resource_get_head_pipe_for_stream(res_ctx, stream);
-	struct pipe_ctx *idle_pipe = find_idle_secondary_pipe(res_ctx, pool);
+	struct pipe_ctx *idle_pipe = find_idle_secondary_pipe(res_ctx, pool, head_pipe);
 
 	if (!head_pipe) {
 		ASSERT(0);
@@ -1143,7 +1162,7 @@ static enum dc_status dcn10_validate_global(struct dc *dc, struct dc_state *cont
 			continue;
 
 		if (context->stream_status[i].plane_count > 2)
-			return false;
+			return DC_FAIL_UNSUPPORTED_1;
 
 		for (j = 0; j < context->stream_status[i].plane_count; j++) {
 			struct dc_plane_state *plane =
@@ -1351,7 +1370,7 @@ static bool construct(
 		goto fail;
 	}
 
-	dml_init_instance(&dc->dml, DML_PROJECT_RAVEN1);
+	dml_init_instance(&dc->dml, &dcn1_0_soc, &dcn1_0_ip, DML_PROJECT_RAVEN1);
 	memcpy(dc->dcn_ip, &dcn10_ip_defaults, sizeof(dcn10_ip_defaults));
 	memcpy(dc->dcn_soc, &dcn10_soc_defaults, sizeof(dcn10_soc_defaults));
 
@@ -1510,6 +1529,9 @@ static bool construct(
 	dcn10_hw_sequencer_construct(dc);
 	dc->caps.max_planes =  pool->base.pipe_count;
 
+	for (i = 0; i < dc->caps.max_planes; ++i)
+		dc->caps.planes[i] = plane_cap;
+
 	dc->cap_funcs = cap_funcs;
 
 	return true;
@@ -1522,7 +1544,7 @@ fail:
 }
 
 struct resource_pool *dcn10_create_resource_pool(
-		uint8_t num_virtual_links,
+		const struct dc_init_data *init_data,
 		struct dc *dc)
 {
 	struct dcn10_resource_pool *pool =
@@ -1531,7 +1553,7 @@ struct resource_pool *dcn10_create_resource_pool(
 	if (!pool)
 		return NULL;
 
-	if (construct(num_virtual_links, dc, pool))
+	if (construct(init_data->num_virtual_links, dc, pool))
 		return &pool->base;
 
 	BREAK_TO_DEBUGGER();
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.h
index 8f71225bc61b..999c684a0b36 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.h
@@ -39,7 +39,7 @@ struct dcn10_resource_pool {
 	struct resource_pool base;
 };
 struct resource_pool *dcn10_create_resource_pool(
-		uint8_t num_virtual_links,
+		const struct dc_init_data *init_data,
 		struct dc *dc);
 
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c
index b08254121251..8ee9f6dc1d62 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c
@@ -245,7 +245,8 @@ static void enc1_update_hdmi_info_packet(
 void enc1_stream_encoder_dp_set_stream_attribute(
 	struct stream_encoder *enc,
 	struct dc_crtc_timing *crtc_timing,
-	enum dc_color_space output_color_space)
+	enum dc_color_space output_color_space,
+	uint32_t enable_sdp_splitting)
 {
 	uint32_t h_active_start;
 	uint32_t v_active_start;
@@ -298,7 +299,6 @@ void enc1_stream_encoder_dp_set_stream_attribute(
 		break;
 	case PIXEL_ENCODING_YCBCR420:
 		dp_pixel_encoding = DP_PIXEL_ENCODING_TYPE_YCBCR420;
-		REG_UPDATE(DP_VID_TIMING, DP_VID_N_MUL, 1);
 		break;
 	default:
 		dp_pixel_encoding = DP_PIXEL_ENCODING_TYPE_RGB444;
@@ -726,13 +726,19 @@ void enc1_stream_encoder_update_dp_info_packets(
 				3,  /* packetIndex */
 				&info_frame->hdrsmd);
 
+	if (info_frame->dpsdp.valid)
+		enc1_update_generic_info_packet(
+				enc1,
+				4,/* packetIndex */
+				&info_frame->dpsdp);
+
 	/* enable/disable transmission of packet(s).
 	 * If enabled, packet transmission begins on the next frame
 	 */
 	REG_UPDATE(DP_SEC_CNTL, DP_SEC_GSP0_ENABLE, info_frame->vsc.valid);
 	REG_UPDATE(DP_SEC_CNTL, DP_SEC_GSP2_ENABLE, info_frame->spd.valid);
 	REG_UPDATE(DP_SEC_CNTL, DP_SEC_GSP3_ENABLE, info_frame->hdrsmd.valid);
-
+	REG_UPDATE(DP_SEC_CNTL, DP_SEC_GSP4_ENABLE, info_frame->dpsdp.valid);
 
 	/* This bit is the master enable bit.
 	 * When enabling secondary stream engine,
@@ -797,10 +803,10 @@ void enc1_stream_encoder_dp_blank(
 	 */
 	REG_UPDATE(DP_VID_STREAM_CNTL, DP_VID_STREAM_DIS_DEFER, 2);
 	/* Larger delay to wait until VBLANK - use max retry of
-	 * 10us*3000=30ms. This covers 16.6ms of typical 60 Hz mode +
+	 * 10us*5000=50ms. This covers 41.7ms of minimum 24 Hz mode +
 	 * a little more because we may not trust delay accuracy.
 	 */
-	max_retries = DP_BLANK_MAX_RETRY * 150;
+	max_retries = DP_BLANK_MAX_RETRY * 250;
 
 	/* disable DP stream */
 	REG_UPDATE(DP_VID_STREAM_CNTL, DP_VID_STREAM_ENABLE, 0);
@@ -833,14 +839,19 @@ void enc1_stream_encoder_dp_unblank(
 	if (param->link_settings.link_rate != LINK_RATE_UNKNOWN) {
 		uint32_t n_vid = 0x8000;
 		uint32_t m_vid;
+		uint32_t n_multiply = 0;
+		uint64_t m_vid_l = n_vid;
 
+		/* YCbCr 4:2:0 : Computed VID_M will be 2X the input rate */
+		if (param->timing.pixel_encoding == PIXEL_ENCODING_YCBCR420) {
+			/*this param->pixel_clk_khz is half of 444 rate for 420 already*/
+			n_multiply = 1;
+		}
 		/* M / N = Fstream / Flink
 		 * m_vid / n_vid = pixel rate / link rate
 		 */
 
-		uint64_t m_vid_l = n_vid;
-
-		m_vid_l *= param->pixel_clk_khz;
+		m_vid_l *= param->timing.pix_clk_100hz / 10;
 		m_vid_l = div_u64(m_vid_l,
 			param->link_settings.link_rate
 				* LINK_RATE_REF_FREQ_IN_KHZ);
@@ -859,7 +870,9 @@ void enc1_stream_encoder_dp_unblank(
 
 		REG_UPDATE(DP_VID_M, DP_VID_M, m_vid);
 
-		REG_UPDATE(DP_VID_TIMING, DP_VID_M_N_GEN_EN, 1);
+		REG_UPDATE_2(DP_VID_TIMING,
+				DP_VID_M_N_GEN_EN, 1,
+				DP_VID_N_MUL, n_multiply);
 	}
 
 	/* set DIG_START to 0x1 to resync FIFO */
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h
index b7c800e10a32..e654c2f55971 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h
@@ -462,7 +462,8 @@ void enc1_update_generic_info_packet(
 void enc1_stream_encoder_dp_set_stream_attribute(
 	struct stream_encoder *enc,
 	struct dc_crtc_timing *crtc_timing,
-	enum dc_color_space output_color_space);
+	enum dc_color_space output_color_space,
+	uint32_t enable_sdp_splitting);
 
 void enc1_stream_encoder_hdmi_set_stream_attribute(
 	struct stream_encoder *enc,
diff --git a/drivers/gpu/drm/amd/display/dc/dm_helpers.h b/drivers/gpu/drm/amd/display/dc/dm_helpers.h
index e81b24374bcb..ccbfe9680d27 100644
--- a/drivers/gpu/drm/amd/display/dc/dm_helpers.h
+++ b/drivers/gpu/drm/amd/display/dc/dm_helpers.h
@@ -58,7 +58,7 @@ bool dm_helpers_dp_mst_write_payload_allocation_table(
 		bool enable);
 
 /*
- * poll pending down reply before clear payload allocation table
+ * poll pending down reply
  */
 void dm_helpers_dp_mst_poll_pending_down_reply(
 	struct dc_context *ctx,
diff --git a/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h b/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h
index 14bed5b1fa97..4fc4208d1472 100644
--- a/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h
+++ b/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h
@@ -30,6 +30,8 @@
  * interface to PPLIB/SMU to setup clocks and pstate requirements on SoC
  */
 
+typedef bool BOOLEAN;
+
 enum pp_smu_ver {
 	/*
 	 * PP_SMU_INTERFACE_X should be interpreted as the interface defined
@@ -72,29 +74,6 @@ struct pp_smu_wm_range_sets {
 	struct pp_smu_wm_set_range writer_wm_sets[MAX_WATERMARK_SETS];
 };
 
-struct pp_smu_display_requirement_rv {
-	/* PPSMC_MSG_SetDisplayCount: count
-	 *  0 triggers S0i2 optimization
-	 */
-	unsigned int display_count;
-
-	/* PPSMC_MSG_SetHardMinFclkByFreq: mhz
-	 *  FCLK will vary with DPM, but never below requested hard min
-	 */
-	unsigned int hard_min_fclk_mhz;
-
-	/* PPSMC_MSG_SetHardMinDcefclkByFreq: mhz
-	 *  fixed clock at requested freq, either from FCH bypass or DFS
-	 */
-	unsigned int hard_min_dcefclk_mhz;
-
-	/* PPSMC_MSG_SetMinDeepSleepDcefclk: mhz
-	 *  when DF is in cstate, dcf clock is further divided down
-	 *  to just above given frequency
-	 */
-	unsigned int min_deep_sleep_dcefclk_mhz;
-};
-
 struct pp_smu_funcs_rv {
 	struct pp_smu pp_smu;
 
@@ -137,12 +116,6 @@ struct pp_smu_funcs_rv {
 	/* PME w/a */
 	void (*set_pme_wa_enable)(struct pp_smu *pp);
 
-	/*
-	 * Legacy functions.  Used for backwards comp. with existing
-	 * PPlib code.
-	 */
-	void (*set_display_requirement)(struct pp_smu *pp,
-			struct pp_smu_display_requirement_rv *req);
 };
 
 struct pp_smu_funcs {
diff --git a/drivers/gpu/drm/amd/display/dc/dm_services.h b/drivers/gpu/drm/amd/display/dc/dm_services.h
index 1961cc6d9143..b426ba02b793 100644
--- a/drivers/gpu/drm/amd/display/dc/dm_services.h
+++ b/drivers/gpu/drm/amd/display/dc/dm_services.h
@@ -52,30 +52,17 @@ irq_handler_idx dm_register_interrupt(
  * GPU registers access
  *
  */
-
+uint32_t dm_read_reg_func(
+	const struct dc_context *ctx,
+	uint32_t address,
+	const char *func_name);
 /* enable for debugging new code, this adds 50k to the driver size. */
 /* #define DM_CHECK_ADDR_0 */
 
 #define dm_read_reg(ctx, address)	\
 		dm_read_reg_func(ctx, address, __func__)
 
-static inline uint32_t dm_read_reg_func(
-	const struct dc_context *ctx,
-	uint32_t address,
-	const char *func_name)
-{
-	uint32_t value;
-#ifdef DM_CHECK_ADDR_0
-	if (address == 0) {
-		DC_ERR("invalid register read; address = 0\n");
-		return 0;
-	}
-#endif
-	value = cgs_read_register(ctx->cgs_device, address);
-	trace_amdgpu_dc_rreg(&ctx->perf_trace->read_count, address, value);
 
-	return value;
-}
 
 #define dm_write_reg(ctx, address, value)	\
 	dm_write_reg_func(ctx, address, value, __func__)
@@ -144,10 +131,14 @@ static inline uint32_t set_reg_field_value_ex(
 		reg_name ## __ ## reg_field ## _MASK,\
 		reg_name ## __ ## reg_field ## __SHIFT)
 
-uint32_t generic_reg_update_ex(const struct dc_context *ctx,
+uint32_t generic_reg_set_ex(const struct dc_context *ctx,
 		uint32_t addr, uint32_t reg_val, int n,
 		uint8_t shift1, uint32_t mask1, uint32_t field_value1, ...);
 
+uint32_t generic_reg_update_ex(const struct dc_context *ctx,
+		uint32_t addr, int n,
+		uint8_t shift1, uint32_t mask1, uint32_t field_value1, ...);
+
 #define FD(reg_field)	reg_field ## __SHIFT, \
 						reg_field ## _MASK
 
@@ -155,7 +146,7 @@ uint32_t generic_reg_update_ex(const struct dc_context *ctx,
  * return number of poll before condition is met
  * return 0 if condition is not meet after specified time out tries
  */
-unsigned int generic_reg_wait(const struct dc_context *ctx,
+void generic_reg_wait(const struct dc_context *ctx,
 	uint32_t addr, uint32_t mask, uint32_t shift, uint32_t condition_value,
 	unsigned int delay_between_poll_us, unsigned int time_out_num_tries,
 	const char *func_name, int line);
@@ -172,11 +163,10 @@ unsigned int generic_reg_wait(const struct dc_context *ctx,
 
 #define generic_reg_update_soc15(ctx, inst_offset, reg_name, n, ...)\
 		generic_reg_update_ex(ctx, DCE_BASE.instance[0].segment[mm##reg_name##_BASE_IDX] +  mm##reg_name + inst_offset, \
-		dm_read_reg_func(ctx, mm##reg_name + DCE_BASE.instance[0].segment[mm##reg_name##_BASE_IDX] + inst_offset, __func__), \
 		n, __VA_ARGS__)
 
 #define generic_reg_set_soc15(ctx, inst_offset, reg_name, n, ...)\
-		generic_reg_update_ex(ctx, DCE_BASE.instance[0].segment[mm##reg_name##_BASE_IDX] + mm##reg_name + inst_offset, 0, \
+		generic_reg_set_ex(ctx, DCE_BASE.instance[0].segment[mm##reg_name##_BASE_IDX] + mm##reg_name + inst_offset, 0, \
 		n, __VA_ARGS__)
 
 #define get_reg_field_value_soc15(reg_value, block, reg_num, reg_name, reg_field)\
@@ -223,8 +213,8 @@ bool dm_pp_notify_wm_clock_changes(
 	const struct dc_context *ctx,
 	struct dm_pp_wm_sets_with_clock_ranges *wm_with_clock_ranges);
 
-void dm_pp_get_funcs_rv(struct dc_context *ctx,
-		struct pp_smu_funcs_rv *funcs);
+void dm_pp_get_funcs(struct dc_context *ctx,
+		struct pp_smu_funcs *funcs);
 
 /* DAL calls this function to notify PP about completion of Mode Set.
  * For PP it means that current DCE clocks are those which were returned
diff --git a/drivers/gpu/drm/amd/display/dc/dm_services_types.h b/drivers/gpu/drm/amd/display/dc/dm_services_types.h
index 77200711abbe..a3d1be20dd9d 100644
--- a/drivers/gpu/drm/amd/display/dc/dm_services_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dm_services_types.h
@@ -29,7 +29,7 @@
 #include "os_types.h"
 #include "dc_types.h"
 
-struct pp_smu_funcs_rv;
+struct pp_smu_funcs;
 
 struct dm_pp_clock_range {
 	int min_khz;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c
index d303b789adfe..80ffd7d958b2 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c
@@ -26,40 +26,14 @@
 #include "display_mode_lib.h"
 #include "dc_features.h"
 
-extern const struct _vcs_dpi_ip_params_st dcn1_0_ip;
-extern const struct _vcs_dpi_soc_bounding_box_st dcn1_0_soc;
-
-static void set_soc_bounding_box(struct _vcs_dpi_soc_bounding_box_st *soc, enum dml_project project)
-{
-	switch (project) {
-	case DML_PROJECT_RAVEN1:
-		*soc = dcn1_0_soc;
-		break;
-	default:
-		ASSERT(0);
-		break;
-	}
-}
-
-static void set_ip_params(struct _vcs_dpi_ip_params_st *ip, enum dml_project project)
+void dml_init_instance(struct display_mode_lib *lib,
+		const struct _vcs_dpi_soc_bounding_box_st *soc_bb,
+		const struct _vcs_dpi_ip_params_st *ip_params,
+		enum dml_project project)
 {
-	switch (project) {
-	case DML_PROJECT_RAVEN1:
-		*ip = dcn1_0_ip;
-		break;
-	default:
-		ASSERT(0);
-		break;
-	}
-}
-
-void dml_init_instance(struct display_mode_lib *lib, enum dml_project project)
-{
-	if (lib->project != project) {
-		set_soc_bounding_box(&lib->soc, project);
-		set_ip_params(&lib->ip, project);
-		lib->project = project;
-	}
+	lib->soc = *soc_bb;
+	lib->ip = *ip_params;
+	lib->project = project;
 }
 
 const char *dml_get_status_message(enum dm_validation_status status)
diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h
index a730e0209c05..1b546dba34bd 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h
@@ -41,7 +41,10 @@ struct display_mode_lib {
 	struct dal_logger *logger;
 };
 
-void dml_init_instance(struct display_mode_lib *lib, enum dml_project project);
+void dml_init_instance(struct display_mode_lib *lib,
+		const struct _vcs_dpi_soc_bounding_box_st *soc_bb,
+		const struct _vcs_dpi_ip_params_st *ip_params,
+		enum dml_project project);
 
 const char *dml_get_status_message(enum dm_validation_status status);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h
index 391183e3428f..c5b791d158a7 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h
@@ -25,6 +25,8 @@
 #ifndef __DISPLAY_MODE_STRUCTS_H__
 #define __DISPLAY_MODE_STRUCTS_H__
 
+#define MAX_CLOCK_LIMIT_STATES 8
+
 typedef struct _vcs_dpi_voltage_scaling_st voltage_scaling_st;
 typedef struct _vcs_dpi_soc_bounding_box_st soc_bounding_box_st;
 typedef struct _vcs_dpi_ip_params_st ip_params_st;
@@ -103,7 +105,7 @@ struct _vcs_dpi_soc_bounding_box_st {
 	double xfc_xbuf_latency_tolerance_us;
 	int use_urgent_burst_bw;
 	unsigned int num_states;
-	struct _vcs_dpi_voltage_scaling_st clock_limits[8];
+	struct _vcs_dpi_voltage_scaling_st clock_limits[MAX_CLOCK_LIMIT_STATES];
 };
 
 struct _vcs_dpi_ip_params_st {
@@ -416,6 +418,7 @@ struct _vcs_dpi_display_dlg_regs_st {
 	unsigned int refcyc_per_vm_group_flip;
 	unsigned int refcyc_per_vm_req_vblank;
 	unsigned int refcyc_per_vm_req_flip;
+	unsigned int refcyc_per_vm_dmdata;
 };
 
 struct _vcs_dpi_display_ttu_regs_st {
diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_rq_dlg_helpers.c b/drivers/gpu/drm/amd/display/dc/dml/display_rq_dlg_helpers.c
index 48400d642610..e2d82aacd3bc 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/display_rq_dlg_helpers.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/display_rq_dlg_helpers.c
@@ -321,6 +321,9 @@ void print__dlg_regs_st(struct display_mode_lib *mode_lib, display_dlg_regs_st d
 	dml_print(
 			"DML_RQ_DLG_CALC:    xfc_reg_remote_surface_flip_latency = 0x%0x\n",
 			dlg_regs.xfc_reg_remote_surface_flip_latency);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_vm_dmdata            = 0x%0x\n",
+			dlg_regs.refcyc_per_vm_dmdata);
 
 	dml_print("DML_RQ_DLG_CALC: =====================================\n");
 }
diff --git a/drivers/gpu/drm/amd/display/dc/inc/clock_source.h b/drivers/gpu/drm/amd/display/dc/inc/clock_source.h
index fe6301cb8681..1b01a9a58d14 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/clock_source.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/clock_source.h
@@ -167,7 +167,7 @@ struct clock_source_funcs {
 			struct pixel_clk_params *,
 			struct pll_settings *);
 	bool (*get_pixel_clk_frequency_100hz)(
-			struct clock_source *clock_source,
+			const struct clock_source *clock_source,
 			unsigned int inst,
 			unsigned int *pixel_clk_khz);
 };
diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_status.h b/drivers/gpu/drm/amd/display/dc/inc/core_status.h
index 2e61a22ef4b2..8dca3b7700e5 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/core_status.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/core_status.h
@@ -43,7 +43,7 @@ enum dc_status {
 	DC_FAIL_BANDWIDTH_VALIDATE = 13, /* BW and Watermark validation */
 	DC_FAIL_SCALING = 14,
 	DC_FAIL_DP_LINK_TRAINING = 15,
-
+	DC_FAIL_UNSUPPORTED_1 = 18,
 	DC_ERROR_UNEXPECTED = -1
 };
 
diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
index 986ed1728644..6f5ab05d6467 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
@@ -95,10 +95,10 @@ struct resource_funcs {
 	void (*link_init)(struct dc_link *link);
 	struct link_encoder *(*link_enc_create)(
 			const struct encoder_init_data *init);
-
 	bool (*validate_bandwidth)(
 					struct dc *dc,
-					struct dc_state *context);
+					struct dc_state *context,
+					bool fast_validate);
 
 	enum dc_status (*validate_global)(
 		struct dc *dc,
@@ -144,8 +144,7 @@ struct resource_pool {
 	struct stream_encoder *stream_enc[MAX_PIPES * 2];
 	struct hubbub *hubbub;
 	struct mpc *mpc;
-	struct pp_smu_funcs_rv *pp_smu;
-	struct pp_smu_display_requirement_rv pp_smu_req;
+	struct pp_smu_funcs *pp_smu;
 	struct dce_aux *engines[MAX_PIPES];
 	struct dce_i2c_hw *hw_i2cs[MAX_PIPES];
 	struct dce_i2c_sw *sw_i2cs[MAX_PIPES];
@@ -154,7 +153,12 @@ struct resource_pool {
 	unsigned int pipe_count;
 	unsigned int underlay_pipe_index;
 	unsigned int stream_enc_count;
-	unsigned int ref_clock_inKhz;
+
+	struct {
+		unsigned int xtalin_clock_inKhz;
+		unsigned int dccg_ref_clock_inKhz;
+		unsigned int dchub_ref_clock_inKhz;
+	} ref_clocks;
 	unsigned int timing_generator_count;
 
 	/*
@@ -262,18 +266,22 @@ struct dcn_bw_output {
 	struct dcn_watermark_set watermarks;
 };
 
-union bw_context {
+union bw_output {
 	struct dcn_bw_output dcn;
 	struct dce_bw_output dce;
 };
 
+struct bw_context {
+	union bw_output bw;
+	struct display_mode_lib dml;
+};
 /**
  * struct dc_state - The full description of a state requested by a user
  *
  * @streams: Stream properties
  * @stream_status: The planes on a given stream
  * @res_ctx: Persistent state of resources
- * @bw: The output from bandwidth and watermark calculations
+ * @bw_ctx: The output from bandwidth and watermark calculations and the DML
  * @pp_display_cfg: PowerPlay clocks and settings
  * @dcn_bw_vars: non-stack memory to support bandwidth calculations
  *
@@ -285,7 +293,7 @@ struct dc_state {
 
 	struct resource_context res_ctx;
 
-	union bw_context bw;
+	struct bw_context bw_ctx;
 
 	/* Note: these are big structures, do *not* put on stack! */
 	struct dm_pp_display_configuration pp_display_cfg;
@@ -293,7 +301,11 @@ struct dc_state {
 	struct dcn_bw_internal_vars dcn_bw_vars;
 #endif
 
-	struct clk_mgr *dccg;
+	struct clk_mgr *clk_mgr;
+
+	struct {
+		bool full_update_needed : 1;
+	} commit_hints;
 
 	struct kref refcount;
 };
diff --git a/drivers/gpu/drm/amd/display/dc/inc/dc_link_ddc.h b/drivers/gpu/drm/amd/display/dc/inc/dc_link_ddc.h
index 16fd4dc6c4dd..b1fab251c09b 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/dc_link_ddc.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/dc_link_ddc.h
@@ -95,8 +95,9 @@ bool dal_ddc_service_query_ddc_data(
 		uint8_t *read_buf,
 		uint32_t read_size);
 
-int dc_link_aux_transfer(struct ddc_service *ddc,
-		struct aux_payload *payload);
+int dc_link_aux_transfer_raw(struct ddc_service *ddc,
+		struct aux_payload *payload,
+		enum aux_channel_operation_result *operation_result);
 
 bool dc_link_aux_transfer_with_retries(struct ddc_service *ddc,
 		struct aux_payload *payload);
diff --git a/drivers/gpu/drm/amd/display/dc/inc/dcn_calcs.h b/drivers/gpu/drm/amd/display/dc/inc/dcn_calcs.h
index ece954a40a8e..263c09630c06 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/dcn_calcs.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/dcn_calcs.h
@@ -621,7 +621,8 @@ extern const struct dcn_ip_params dcn10_ip_defaults;
 
 bool dcn_validate_bandwidth(
 		struct dc *dc,
-		struct dc_state *context);
+		struct dc_state *context,
+		bool fast_validate);
 
 unsigned int dcn_find_dcfclk_suits_all(
 	const struct dc *dc,
@@ -631,5 +632,7 @@ void dcn_bw_update_from_pplib(struct dc *dc);
 void dcn_bw_notify_pplib_of_wm_ranges(struct dc *dc);
 void dcn_bw_sync_calcs_and_dml(struct dc *dc);
 
+enum source_macro_tile_size swizzle_mode_to_macro_tile_size(enum swizzle_mode_values sw_mode);
+
 #endif /* __DCN_CALCS_H__ */
 
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h
index 23a4b18e5fee..31bd6d5183ab 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h
@@ -42,6 +42,8 @@ struct clk_mgr_funcs {
 			bool safe_to_lower);
 
 	int (*get_dp_ref_clk_frequency)(struct clk_mgr *clk_mgr);
+
+	void (*init_clocks)(struct clk_mgr *clk_mgr);
 };
 
 #endif /* __DAL_CLK_MGR_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h
index 95a56d012626..05ee5295d2c1 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h
@@ -39,6 +39,10 @@ struct dccg_funcs {
 	void (*update_dpp_dto)(struct dccg *dccg,
 			int dpp_inst,
 			int req_dppclk);
+	void (*get_dccg_ref_freq)(struct dccg *dccg,
+			unsigned int xtalin_freq_inKhz,
+			unsigned int *dccg_ref_freq_inKhz);
+	void (*dccg_init)(struct dccg *dccg);
 };
 
 #endif //__DAL_DCCG_H__
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h
index 9d2d8e51306c..93667e8b23b3 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h
@@ -73,6 +73,16 @@ struct hubbub_funcs {
 
 	void (*wm_read_state)(struct hubbub *hubbub,
 			struct dcn_hubbub_wm *wm);
+
+	void (*get_dchub_ref_freq)(struct hubbub *hubbub,
+			unsigned int dccg_ref_freq_inKhz,
+			unsigned int *dchub_ref_freq_inKhz);
+
+	void (*program_watermarks)(
+			struct hubbub *hubbub,
+			struct dcn_watermark_set *watermarks,
+			unsigned int refclk_mhz,
+			bool safe_to_lower);
 };
 
 struct hubbub {
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dmcu.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dmcu.h
index cbaa43853611..c68f0ce346c7 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/dmcu.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dmcu.h
@@ -70,6 +70,8 @@ struct dmcu_funcs {
 	void (*get_psr_wait_loop)(struct dmcu *dmcu,
 			unsigned int *psr_wait_loop_number);
 	bool (*is_dmcu_initialized)(struct dmcu *dmcu);
+	bool (*lock_phy)(struct dmcu *dmcu);
+	bool (*unlock_phy)(struct dmcu *dmcu);
 };
 
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h
index 1cd07e94ee63..455df4999797 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h
@@ -130,6 +130,7 @@ struct hubp_funcs {
 	void (*hubp_clear_underflow)(struct hubp *hubp);
 	void (*hubp_disable_control)(struct hubp *hubp, bool disable_hubp);
 	unsigned int (*hubp_get_underflow_status)(struct hubp *hubp);
+	void (*hubp_init)(struct hubp *hubp);
 
 };
 
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/hw_shared.h b/drivers/gpu/drm/amd/display/dc/inc/hw/hw_shared.h
index da85537a4488..4c8e2c6fb6db 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/hw_shared.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/hw_shared.h
@@ -146,6 +146,12 @@ struct out_csc_color_matrix {
 	uint16_t regval[12];
 };
 
+enum gamut_remap_select {
+	GAMUT_REMAP_BYPASS = 0,
+	GAMUT_REMAP_COEFF,
+	GAMUT_REMAP_COMA_COEFF,
+	GAMUT_REMAP_COMB_COEFF
+};
 
 enum opp_regamma {
 	OPP_REGAMMA_BYPASS = 0,
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h b/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h
index 4051493557bc..49854eb73d1d 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h
@@ -63,11 +63,13 @@ struct encoder_info_frame {
 	struct dc_info_packet vsc;
 	/* HDR Static MetaData */
 	struct dc_info_packet hdrsmd;
+	/* custom sdp message */
+	struct dc_info_packet dpsdp;
 };
 
 struct encoder_unblank_param {
 	struct dc_link_settings link_settings;
-	unsigned int pixel_clk_khz;
+	struct dc_crtc_timing timing;
 };
 
 struct encoder_set_dp_phy_pattern_param {
@@ -88,7 +90,8 @@ struct stream_encoder_funcs {
 	void (*dp_set_stream_attribute)(
 		struct stream_encoder *enc,
 		struct dc_crtc_timing *crtc_timing,
-		enum dc_color_space output_color_space);
+		enum dc_color_space output_color_space,
+		uint32_t enable_sdp_splitting);
 
 	void (*hdmi_set_stream_attribute)(
 		struct stream_encoder *enc,
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h
index c25f7df7b5e3..067d53caf28a 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h
@@ -187,8 +187,10 @@ struct timing_generator_funcs {
 	bool (*did_triggered_reset_occur)(struct timing_generator *tg);
 	void (*setup_global_swap_lock)(struct timing_generator *tg,
 							const struct dcp_gsl_params *gsl_params);
+	void (*setup_global_lock)(struct timing_generator *tg);
 	void (*unlock)(struct timing_generator *tg);
 	void (*lock)(struct timing_generator *tg);
+	void (*lock_global)(struct timing_generator *tg);
 	void (*enable_reset_trigger)(struct timing_generator *tg,
 				     int source_tg_inst);
 	void (*enable_crtc_reset)(struct timing_generator *tg,
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
index 7676f25216b1..33905468e2b9 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
@@ -176,6 +176,10 @@ struct hw_sequencer_funcs {
 				struct dc *dc,
 				struct pipe_ctx *pipe,
 				bool lock);
+	void (*pipe_control_lock_global)(
+				struct dc *dc,
+				struct pipe_ctx *pipe,
+				bool lock);
 	void (*blank_pixel_data)(
 			struct dc *dc,
 			struct pipe_ctx *pipe_ctx,
diff --git a/drivers/gpu/drm/amd/display/dc/inc/reg_helper.h b/drivers/gpu/drm/amd/display/dc/inc/reg_helper.h
index cf5a84b9e27c..8503d9cc4763 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/reg_helper.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/reg_helper.h
@@ -52,7 +52,7 @@
 
 /* macro to set register fields. */
 #define REG_SET_N(reg_name, n, initial_val, ...)	\
-		generic_reg_update_ex(CTX, \
+		generic_reg_set_ex(CTX, \
 				REG(reg_name), \
 				initial_val, \
 				n, __VA_ARGS__)
@@ -225,7 +225,6 @@
 #define REG_UPDATE_N(reg_name, n, ...)	\
 		generic_reg_update_ex(CTX, \
 				REG(reg_name), \
-				REG_READ(reg_name), \
 				n, __VA_ARGS__)
 
 #define REG_UPDATE(reg_name, field, val)	\
@@ -380,16 +379,11 @@
 /* macro to update a register field to specified values in given sequences.
  * useful when toggling bits
  */
-#define REG_UPDATE_SEQ(reg, field, value1, value2) \
-{	uint32_t val = REG_UPDATE(reg, field, value1); \
-	REG_SET(reg, val, field, value2); }
-
-/* macro to update fields in register 1 field at a time in given order */
-#define REG_UPDATE_1BY1_2(reg, f1, v1, f2, v2) \
+#define REG_UPDATE_SEQ_2(reg, f1, v1, f2, v2) \
 {	uint32_t val = REG_UPDATE(reg, f1, v1); \
 	REG_SET(reg, val, f2, v2); }
 
-#define REG_UPDATE_1BY1_3(reg, f1, v1, f2, v2, f3, v3) \
+#define REG_UPDATE_SEQ_3(reg, f1, v1, f2, v2, f3, v3) \
 {	uint32_t val = REG_UPDATE(reg, f1, v1); \
 	val = REG_SET(reg, val, f2, v2); \
 	REG_SET(reg, val, f3, v3); }
diff --git a/drivers/gpu/drm/amd/display/dc/inc/resource.h b/drivers/gpu/drm/amd/display/dc/inc/resource.h
index 0086a2f1d21a..3ce0a4fc5822 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/resource.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/resource.h
@@ -70,11 +70,9 @@ bool resource_construct(
 	struct resource_pool *pool,
 	const struct resource_create_funcs *create_funcs);
 
-struct resource_pool *dc_create_resource_pool(
-				struct dc *dc,
-				int num_virtual_links,
-				enum dce_version dc_version,
-				struct hw_asic_id asic_id);
+struct resource_pool *dc_create_resource_pool(struct dc  *dc,
+					      const struct dc_init_data *init_data,
+					      enum dce_version dc_version);
 
 void dc_destroy_resource_pool(struct dc *dc);
 
@@ -131,7 +129,8 @@ bool resource_attach_surfaces_to_context(
 
 struct pipe_ctx *find_idle_secondary_pipe(
 		struct resource_context *res_ctx,
-		const struct resource_pool *pool);
+		const struct resource_pool *pool,
+		const struct pipe_ctx *primary_pipe);
 
 bool resource_is_stream_unchanged(
 	struct dc_state *old_context, struct dc_stream_state *stream);
@@ -172,4 +171,7 @@ void update_audio_usage(
 
 unsigned int resource_pixel_format_to_bpp(enum surface_pixel_format format);
 
+struct pipe_ctx *dc_res_get_odm_bottom_pipe(struct pipe_ctx *pipe_ctx);
+bool dc_res_is_odm_head_pipe(struct pipe_ctx *pipe_ctx);
+
 #endif /* DRIVERS_GPU_DRM_AMD_DC_DEV_DC_INC_RESOURCE_H_ */
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dce110/irq_service_dce110.c b/drivers/gpu/drm/amd/display/dc/irq/dce110/irq_service_dce110.c
index afe0876fe6f8..86987f5e8bd5 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dce110/irq_service_dce110.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dce110/irq_service_dce110.c
@@ -81,6 +81,11 @@ static const struct irq_source_info_funcs vblank_irq_info_funcs = {
 	.ack = NULL
 };
 
+static const struct irq_source_info_funcs vupdate_irq_info_funcs = {
+	.set = NULL,
+	.ack = NULL
+};
+
 #define hpd_int_entry(reg_num)\
 	[DC_IRQ_SOURCE_HPD1 + reg_num] = {\
 		.enable_reg = mmHPD ## reg_num ## _DC_HPD_INT_CONTROL,\
@@ -137,7 +142,7 @@ static const struct irq_source_info_funcs vblank_irq_info_funcs = {
 		CRTC_V_UPDATE_INT_STATUS__CRTC_V_UPDATE_INT_CLEAR_MASK,\
 		.ack_value =\
 		CRTC_V_UPDATE_INT_STATUS__CRTC_V_UPDATE_INT_CLEAR_MASK,\
-		.funcs = &vblank_irq_info_funcs\
+		.funcs = &vupdate_irq_info_funcs\
 	}
 
 #define vblank_int_entry(reg_num)\
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dce120/irq_service_dce120.c b/drivers/gpu/drm/amd/display/dc/irq/dce120/irq_service_dce120.c
index 1ea7256ec89b..750ba0ab4106 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dce120/irq_service_dce120.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dce120/irq_service_dce120.c
@@ -84,6 +84,11 @@ static const struct irq_source_info_funcs vblank_irq_info_funcs = {
 	.ack = NULL
 };
 
+static const struct irq_source_info_funcs vupdate_irq_info_funcs = {
+	.set = NULL,
+	.ack = NULL
+};
+
 #define BASE_INNER(seg) \
 	DCE_BASE__INST0_SEG ## seg
 
@@ -140,7 +145,7 @@ static const struct irq_source_info_funcs vblank_irq_info_funcs = {
 		IRQ_REG_ENTRY(CRTC, reg_num,\
 			CRTC_INTERRUPT_CONTROL, CRTC_V_UPDATE_INT_MSK,\
 			CRTC_V_UPDATE_INT_STATUS, CRTC_V_UPDATE_INT_CLEAR),\
-		.funcs = &vblank_irq_info_funcs\
+		.funcs = &vupdate_irq_info_funcs\
 	}
 
 #define vblank_int_entry(reg_num)\
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dce80/irq_service_dce80.c b/drivers/gpu/drm/amd/display/dc/irq/dce80/irq_service_dce80.c
index 8a2066c313fe..de218fe84a43 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dce80/irq_service_dce80.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dce80/irq_service_dce80.c
@@ -84,6 +84,10 @@ static const struct irq_source_info_funcs vblank_irq_info_funcs = {
 	.ack = NULL
 };
 
+static const struct irq_source_info_funcs vupdate_irq_info_funcs = {
+	.set = NULL,
+	.ack = NULL
+};
 
 #define hpd_int_entry(reg_num)\
 	[DC_IRQ_SOURCE_INVALID + reg_num] = {\
@@ -142,7 +146,7 @@ static const struct irq_source_info_funcs vblank_irq_info_funcs = {
 		CRTC_V_UPDATE_INT_STATUS__CRTC_V_UPDATE_INT_CLEAR_MASK,\
 		.ack_value =\
 		CRTC_V_UPDATE_INT_STATUS__CRTC_V_UPDATE_INT_CLEAR_MASK,\
-		.funcs = &vblank_irq_info_funcs\
+		.funcs = &vupdate_irq_info_funcs\
 	}
 
 #define vblank_int_entry(reg_num)\
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn10/irq_service_dcn10.c b/drivers/gpu/drm/amd/display/dc/irq/dcn10/irq_service_dcn10.c
index e04ae49243f6..10ac6deff5ff 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dcn10/irq_service_dcn10.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dcn10/irq_service_dcn10.c
@@ -56,6 +56,18 @@ enum dc_irq_source to_dal_irq_source_dcn10(
 		return DC_IRQ_SOURCE_VBLANK5;
 	case DCN_1_0__SRCID__DC_D6_OTG_VSTARTUP:
 		return DC_IRQ_SOURCE_VBLANK6;
+	case DCN_1_0__SRCID__OTG0_IHC_V_UPDATE_NO_LOCK_INTERRUPT:
+		return DC_IRQ_SOURCE_VUPDATE1;
+	case DCN_1_0__SRCID__OTG1_IHC_V_UPDATE_NO_LOCK_INTERRUPT:
+		return DC_IRQ_SOURCE_VUPDATE2;
+	case DCN_1_0__SRCID__OTG2_IHC_V_UPDATE_NO_LOCK_INTERRUPT:
+		return DC_IRQ_SOURCE_VUPDATE3;
+	case DCN_1_0__SRCID__OTG3_IHC_V_UPDATE_NO_LOCK_INTERRUPT:
+		return DC_IRQ_SOURCE_VUPDATE4;
+	case DCN_1_0__SRCID__OTG4_IHC_V_UPDATE_NO_LOCK_INTERRUPT:
+		return DC_IRQ_SOURCE_VUPDATE5;
+	case DCN_1_0__SRCID__OTG5_IHC_V_UPDATE_NO_LOCK_INTERRUPT:
+		return DC_IRQ_SOURCE_VUPDATE6;
 	case DCN_1_0__SRCID__HUBP0_FLIP_INTERRUPT:
 		return DC_IRQ_SOURCE_PFLIP1;
 	case DCN_1_0__SRCID__HUBP1_FLIP_INTERRUPT:
@@ -153,6 +165,11 @@ static const struct irq_source_info_funcs vblank_irq_info_funcs = {
 	.ack = NULL
 };
 
+static const struct irq_source_info_funcs vupdate_no_lock_irq_info_funcs = {
+	.set = NULL,
+	.ack = NULL
+};
+
 #define BASE_INNER(seg) \
 	DCE_BASE__INST0_SEG ## seg
 
@@ -203,12 +220,15 @@ static const struct irq_source_info_funcs vblank_irq_info_funcs = {
 		.funcs = &pflip_irq_info_funcs\
 	}
 
-#define vupdate_int_entry(reg_num)\
+/* vupdate_no_lock_int_entry maps to DC_IRQ_SOURCE_VUPDATEx, to match semantic
+ * of DCE's DC_IRQ_SOURCE_VUPDATEx.
+ */
+#define vupdate_no_lock_int_entry(reg_num)\
 	[DC_IRQ_SOURCE_VUPDATE1 + reg_num] = {\
 		IRQ_REG_ENTRY(OTG, reg_num,\
-			OTG_GLOBAL_SYNC_STATUS, VUPDATE_INT_EN,\
-			OTG_GLOBAL_SYNC_STATUS, VUPDATE_EVENT_CLEAR),\
-		.funcs = &vblank_irq_info_funcs\
+			OTG_GLOBAL_SYNC_STATUS, VUPDATE_NO_LOCK_INT_EN,\
+			OTG_GLOBAL_SYNC_STATUS, VUPDATE_NO_LOCK_EVENT_CLEAR),\
+		.funcs = &vupdate_no_lock_irq_info_funcs\
 	}
 
 #define vblank_int_entry(reg_num)\
@@ -315,12 +335,12 @@ irq_source_info_dcn10[DAL_IRQ_SOURCES_NUMBER] = {
 	dc_underflow_int_entry(6),
 	[DC_IRQ_SOURCE_DMCU_SCP] = dummy_irq_entry(),
 	[DC_IRQ_SOURCE_VBIOS_SW] = dummy_irq_entry(),
-	vupdate_int_entry(0),
-	vupdate_int_entry(1),
-	vupdate_int_entry(2),
-	vupdate_int_entry(3),
-	vupdate_int_entry(4),
-	vupdate_int_entry(5),
+	vupdate_no_lock_int_entry(0),
+	vupdate_no_lock_int_entry(1),
+	vupdate_no_lock_int_entry(2),
+	vupdate_no_lock_int_entry(3),
+	vupdate_no_lock_int_entry(4),
+	vupdate_no_lock_int_entry(5),
 	vblank_int_entry(0),
 	vblank_int_entry(1),
 	vblank_int_entry(2),
diff --git a/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c
index 3dc1733eea20..fdcf9e66d852 100644
--- a/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c
@@ -29,7 +29,8 @@
 static void virtual_stream_encoder_dp_set_stream_attribute(
 	struct stream_encoder *enc,
 	struct dc_crtc_timing *crtc_timing,
-	enum dc_color_space output_color_space) {}
+	enum dc_color_space output_color_space,
+	uint32_t enable_sdp_splitting) {}
 
 static void virtual_stream_encoder_hdmi_set_stream_attribute(
 	struct stream_encoder *enc,
diff --git a/drivers/gpu/drm/amd/display/include/fixed31_32.h b/drivers/gpu/drm/amd/display/include/fixed31_32.h
index 52a73332befb..89ef9f6860e5 100644
--- a/drivers/gpu/drm/amd/display/include/fixed31_32.h
+++ b/drivers/gpu/drm/amd/display/include/fixed31_32.h
@@ -503,6 +503,8 @@ static inline int dc_fixpt_ceil(struct fixed31_32 arg)
  * fractional
  */
 
+unsigned int dc_fixpt_u4d19(struct fixed31_32 arg);
+
 unsigned int dc_fixpt_u3d19(struct fixed31_32 arg);
 
 unsigned int dc_fixpt_u2d19(struct fixed31_32 arg);
diff --git a/drivers/gpu/drm/amd/display/include/signal_types.h b/drivers/gpu/drm/amd/display/include/signal_types.h
index f56d2891475f..beed70179bb5 100644
--- a/drivers/gpu/drm/amd/display/include/signal_types.h
+++ b/drivers/gpu/drm/amd/display/include/signal_types.h
@@ -45,6 +45,11 @@ enum signal_type {
 };
 
 /* help functions for signal types manipulation */
+static inline bool dc_is_hdmi_tmds_signal(enum signal_type signal)
+{
+	return (signal == SIGNAL_TYPE_HDMI_TYPE_A);
+}
+
 static inline bool dc_is_hdmi_signal(enum signal_type signal)
 {
 	return (signal == SIGNAL_TYPE_HDMI_TYPE_A);
diff --git a/drivers/gpu/drm/amd/display/modules/color/color_gamma.c b/drivers/gpu/drm/amd/display/modules/color/color_gamma.c
index 0fbc8fbc3541..a1055413bade 100644
--- a/drivers/gpu/drm/amd/display/modules/color/color_gamma.c
+++ b/drivers/gpu/drm/amd/display/modules/color/color_gamma.c
@@ -1854,6 +1854,8 @@ bool mod_color_calculate_degamma_params(struct dc_transfer_func *input_tf,
 			coordinates_x, axis_x, curve,
 			MAX_HW_POINTS, tf_pts,
 			mapUserRamp && ramp && ramp->type == GAMMA_RGB_256);
+	if (ramp->type == GAMMA_CUSTOM)
+		apply_lut_1d(ramp, MAX_HW_POINTS, tf_pts);
 
 	ret = true;
 
diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
index bfd27f10879e..19b1eaebe484 100644
--- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
+++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
@@ -37,6 +37,8 @@
 #define RENDER_TIMES_MAX_COUNT 10
 /* Threshold to exit BTR (to avoid frequent enter-exits at the lower limit) */
 #define BTR_EXIT_MARGIN 2000
+/* Threshold to change BTR multiplier (to avoid frequent changes) */
+#define BTR_DRIFT_MARGIN 2000
 /*Threshold to exit fixed refresh rate*/
 #define FIXED_REFRESH_EXIT_MARGIN_IN_HZ 4
 /* Number of consecutive frames to check before entering/exiting fixed refresh*/
@@ -48,6 +50,93 @@ struct core_freesync {
 	struct dc *dc;
 };
 
+void setFieldWithMask(unsigned char *dest, unsigned int mask, unsigned int value)
+{
+	unsigned int shift = 0;
+
+	if (!mask || !dest)
+		return;
+
+	while (!((mask >> shift) & 1))
+		shift++;
+
+	//reset
+	*dest = *dest & ~mask;
+	//set
+	//dont let value span past mask
+	value = value & (mask >> shift);
+	//insert value
+	*dest = *dest | (value << shift);
+}
+
+// VTEM Byte Offset
+#define VRR_VTEM_PB0		0
+#define VRR_VTEM_PB1		1
+#define VRR_VTEM_PB2		2
+#define VRR_VTEM_PB3		3
+#define VRR_VTEM_PB4		4
+#define VRR_VTEM_PB5		5
+#define VRR_VTEM_PB6		6
+
+#define VRR_VTEM_MD0		7
+#define VRR_VTEM_MD1		8
+#define VRR_VTEM_MD2		9
+#define VRR_VTEM_MD3		10
+
+
+// VTEM Byte Masks
+//PB0
+#define MASK__VRR_VTEM_PB0__RESERVED0  0x01
+#define MASK__VRR_VTEM_PB0__SYNC       0x02
+#define MASK__VRR_VTEM_PB0__VFR        0x04
+#define MASK__VRR_VTEM_PB0__AFR        0x08
+#define MASK__VRR_VTEM_PB0__DS_TYPE    0x30
+	//0: Periodic pseudo-static EM Data Set
+	//1: Periodic dynamic EM Data Set
+	//2: Unique EM Data Set
+	//3: Reserved
+#define MASK__VRR_VTEM_PB0__END        0x40
+#define MASK__VRR_VTEM_PB0__NEW        0x80
+
+//PB1
+#define MASK__VRR_VTEM_PB1__RESERVED1 0xFF
+
+//PB2
+#define MASK__VRR_VTEM_PB2__ORGANIZATION_ID 0xFF
+	//0: This is a Vendor Specific EM Data Set
+	//1: This EM Data Set is defined by This Specification (HDMI 2.1 r102.clean)
+	//2: This EM Data Set is defined by CTA-861-G
+	//3: This EM Data Set is defined by VESA
+//PB3
+#define MASK__VRR_VTEM_PB3__DATA_SET_TAG_MSB    0xFF
+//PB4
+#define MASK__VRR_VTEM_PB4__DATA_SET_TAG_LSB    0xFF
+//PB5
+#define MASK__VRR_VTEM_PB5__DATA_SET_LENGTH_MSB 0xFF
+//PB6
+#define MASK__VRR_VTEM_PB6__DATA_SET_LENGTH_LSB 0xFF
+
+
+
+//PB7-27 (20 bytes):
+//PB7 = MD0
+#define MASK__VRR_VTEM_MD0__VRR_EN         0x01
+#define MASK__VRR_VTEM_MD0__M_CONST        0x02
+#define MASK__VRR_VTEM_MD0__RESERVED2      0x0C
+#define MASK__VRR_VTEM_MD0__FVA_FACTOR_M1  0xF0
+
+//MD1
+#define MASK__VRR_VTEM_MD1__BASE_VFRONT    0xFF
+
+//MD2
+#define MASK__VRR_VTEM_MD2__BASE_REFRESH_RATE_98  0x03
+#define MASK__VRR_VTEM_MD2__RB                    0x04
+#define MASK__VRR_VTEM_MD2__RESERVED3             0xF8
+
+//MD3
+#define MASK__VRR_VTEM_MD3__BASE_REFRESH_RATE_07  0xFF
+
+
 #define MOD_FREESYNC_TO_CORE(mod_freesync)\
 		container_of(mod_freesync, struct core_freesync, public)
 
@@ -248,6 +337,7 @@ static void apply_below_the_range(struct core_freesync *core_freesync,
 	unsigned int frames_to_insert = 0;
 	unsigned int min_frame_duration_in_ns = 0;
 	unsigned int max_render_time_in_us = in_out_vrr->max_duration_in_us;
+	unsigned int delta_from_mid_point_delta_in_us;
 
 	min_frame_duration_in_ns = ((unsigned int) (div64_u64(
 		(1000000000ULL * 1000000),
@@ -318,10 +408,27 @@ static void apply_below_the_range(struct core_freesync *core_freesync,
 		/* Choose number of frames to insert based on how close it
 		 * can get to the mid point of the variable range.
 		 */
-		if (delta_from_mid_point_in_us_1 < delta_from_mid_point_in_us_2)
+		if (delta_from_mid_point_in_us_1 < delta_from_mid_point_in_us_2) {
 			frames_to_insert = mid_point_frames_ceil;
-		else
+			delta_from_mid_point_delta_in_us = delta_from_mid_point_in_us_2 -
+					delta_from_mid_point_in_us_1;
+		} else {
 			frames_to_insert = mid_point_frames_floor;
+			delta_from_mid_point_delta_in_us = delta_from_mid_point_in_us_1 -
+					delta_from_mid_point_in_us_2;
+		}
+
+		/* Prefer current frame multiplier when BTR is enabled unless it drifts
+		 * too far from the midpoint
+		 */
+		if (in_out_vrr->btr.frames_to_insert != 0 &&
+				delta_from_mid_point_delta_in_us < BTR_DRIFT_MARGIN) {
+			if (((last_render_time_in_us / in_out_vrr->btr.frames_to_insert) <
+					in_out_vrr->max_duration_in_us) &&
+				((last_render_time_in_us / in_out_vrr->btr.frames_to_insert) >
+					in_out_vrr->min_duration_in_us))
+				frames_to_insert = in_out_vrr->btr.frames_to_insert;
+		}
 
 		/* Either we've calculated the number of frames to insert,
 		 * or we need to insert min duration frames
@@ -330,10 +437,8 @@ static void apply_below_the_range(struct core_freesync *core_freesync,
 			inserted_frame_duration_in_us = last_render_time_in_us /
 							frames_to_insert;
 
-		if (inserted_frame_duration_in_us <
-			(1000000 / in_out_vrr->max_refresh_in_uhz))
-			inserted_frame_duration_in_us =
-				(1000000 / in_out_vrr->max_refresh_in_uhz);
+		if (inserted_frame_duration_in_us < in_out_vrr->min_duration_in_us)
+			inserted_frame_duration_in_us = in_out_vrr->min_duration_in_us;
 
 		/* Cache the calculated variables */
 		in_out_vrr->btr.inserted_duration_in_us =
@@ -469,16 +574,14 @@ static void build_vrr_infopacket_header_vtem(enum signal_type signal,
 	// HB0, HB1, HB2 indicates PacketType VTEMPacket
 	infopacket->hb0 = 0x7F;
 	infopacket->hb1 = 0xC0;
-	infopacket->hb2 = 0x00;
-	/* HB3 Bit Fields
-	 * Reserved :1 = 0
-	 * Sync     :1 = 0
-	 * VFR      :1 = 1
-	 * Ds_Type  :2 = 0
-	 * End      :1 = 0
-	 * New      :1 = 0
-	 */
-	infopacket->hb3 = 0x20;
+	infopacket->hb2 = 0x00; //sequence_index
+
+	setFieldWithMask(&infopacket->sb[VRR_VTEM_PB0], MASK__VRR_VTEM_PB0__VFR, 1);
+	setFieldWithMask(&infopacket->sb[VRR_VTEM_PB2], MASK__VRR_VTEM_PB2__ORGANIZATION_ID, 1);
+	setFieldWithMask(&infopacket->sb[VRR_VTEM_PB3], MASK__VRR_VTEM_PB3__DATA_SET_TAG_MSB, 0);
+	setFieldWithMask(&infopacket->sb[VRR_VTEM_PB4], MASK__VRR_VTEM_PB4__DATA_SET_TAG_LSB, 1);
+	setFieldWithMask(&infopacket->sb[VRR_VTEM_PB5], MASK__VRR_VTEM_PB5__DATA_SET_LENGTH_MSB, 0);
+	setFieldWithMask(&infopacket->sb[VRR_VTEM_PB6], MASK__VRR_VTEM_PB6__DATA_SET_LENGTH_LSB, 4);
 }
 
 static void build_vrr_infopacket_header_v1(enum signal_type signal,
@@ -583,45 +686,36 @@ static void build_vrr_vtem_infopacket_data(const struct dc_stream_state *stream,
 		const struct mod_vrr_params *vrr,
 		struct dc_info_packet *infopacket)
 {
-	/* dc_info_packet to VtemPacket Translation of Bit-fields,
-	 * SB[6]
-	 * unsigned char VRR_EN        :1
-	 * unsigned char M_CONST       :1
-	 * unsigned char Reserved2     :2
-	 * unsigned char FVA_Factor_M1 :4
-	 * SB[7]
-	 * unsigned char Base_Vfront   :8
-	 * SB[8]
-	 * unsigned char Base_Refresh_Rate_98 :2
-	 * unsigned char RB                   :1
-	 * unsigned char Reserved3            :5
-	 * SB[9]
-	 * unsigned char Base_RefreshRate_07  :8
-	 */
 	unsigned int fieldRateInHz;
 
 	if (vrr->state == VRR_STATE_ACTIVE_VARIABLE ||
-				vrr->state == VRR_STATE_ACTIVE_FIXED){
-		infopacket->sb[6] |= 0x80; //VRR_EN Bit = 1
+				vrr->state == VRR_STATE_ACTIVE_FIXED) {
+		setFieldWithMask(&infopacket->sb[VRR_VTEM_MD0], MASK__VRR_VTEM_MD0__VRR_EN, 1);
 	} else {
-		infopacket->sb[6] &= 0x7F; //VRR_EN Bit = 0
+		setFieldWithMask(&infopacket->sb[VRR_VTEM_MD0], MASK__VRR_VTEM_MD0__VRR_EN, 0);
 	}
 
 	if (!stream->timing.vic) {
-		infopacket->sb[7] = stream->timing.v_front_porch;
+		setFieldWithMask(&infopacket->sb[VRR_VTEM_MD1], MASK__VRR_VTEM_MD1__BASE_VFRONT,
+				stream->timing.v_front_porch);
+
 
 		/* TODO: In dal2, we check mode flags for a reduced blanking timing.
 		 * Need a way to relay that information to this function.
 		 * if("ReducedBlanking")
 		 * {
-		 *   infopacket->sb[8] |= 0x20; //Set 3rd bit to 1
+		 *   setFieldWithMask(&infopacket->sb[VRR_VTEM_MD2], MASK__VRR_VTEM_MD2__RB, 1;
 		 * }
 		 */
+
+		//TODO: DAL2 does FixPoint and rounding. Here we might need to account for that
 		fieldRateInHz = (stream->timing.pix_clk_100hz * 100)/
-				(stream->timing.h_total * stream->timing.v_total);
+			(stream->timing.h_total * stream->timing.v_total);
 
-		infopacket->sb[8] |= ((fieldRateInHz & 0x300) >> 2);
-		infopacket->sb[9] |= fieldRateInHz & 0xFF;
+		setFieldWithMask(&infopacket->sb[VRR_VTEM_MD2],  MASK__VRR_VTEM_MD2__BASE_REFRESH_RATE_98,
+				fieldRateInHz >> 8);
+		setFieldWithMask(&infopacket->sb[VRR_VTEM_MD3], MASK__VRR_VTEM_MD3__BASE_REFRESH_RATE_07,
+				fieldRateInHz);
 
 	}
 	infopacket->valid = true;
@@ -745,6 +839,8 @@ static void build_vrr_infopacket_vtem(const struct dc_stream_state *stream,
 {
 	//VTEM info packet for HdmiVrr
 
+	memset(infopacket, 0, sizeof(struct dc_info_packet));
+
 	//VTEM Packet is structured differently
 	build_vrr_infopacket_header_vtem(stream->signal, infopacket);
 	build_vrr_vtem_infopacket_data(stream, vrr, infopacket);
diff --git a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c
index 038b88221c5f..b3810b864676 100644
--- a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c
+++ b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c
@@ -41,9 +41,12 @@ static const unsigned char min_reduction_table[13] = {
 static const unsigned char max_reduction_table[13] = {
 0xf5, 0xe5, 0xd9, 0xcd, 0xb1, 0xa5, 0xa5, 0x80, 0x65, 0x4d, 0x4d, 0x4d, 0x32};
 
-/* ABM 2.2 Min Reduction effectively disabled (100% for all configs)*/
+/* Possible ABM 2.2 Min Reduction configs from least aggressive to most aggressive
+ *  0    1     2     3     4     5     6     7     8     9     10    11   12
+ * 100  100   100   100   100   100   100   100  100  92.2  83.1  75.3  75.3 %
+ */
 static const unsigned char min_reduction_table_v_2_2[13] = {
-0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff};
+0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xeb, 0xd4, 0xc0, 0xc0};
 
 /* Possible ABM 2.2 Max Reduction configs from least aggressive to most aggressive
  *  0    1     2     3     4     5     6     7     8     9     10    11   12
@@ -408,9 +411,9 @@ void fill_iram_v_2_2(struct iram_table_v_2_2 *ram_table, struct dmcu_iram_parame
 	ram_table->flags = 0x0;
 
 	ram_table->deviation_gain[0] = 0xb3;
-	ram_table->deviation_gain[1] = 0xb3;
-	ram_table->deviation_gain[2] = 0xb3;
-	ram_table->deviation_gain[3] = 0xb3;
+	ram_table->deviation_gain[1] = 0xa8;
+	ram_table->deviation_gain[2] = 0x98;
+	ram_table->deviation_gain[3] = 0x68;
 
 	ram_table->min_reduction[0][0] = min_reduction_table_v_2_2[abm_config[set][0]];
 	ram_table->min_reduction[1][0] = min_reduction_table_v_2_2[abm_config[set][0]];
@@ -505,7 +508,7 @@ void fill_iram_v_2_2(struct iram_table_v_2_2 *ram_table, struct dmcu_iram_parame
 
 	ram_table->contrastFactor[0] = 0x99;
 	ram_table->contrastFactor[1] = 0x99;
-	ram_table->contrastFactor[2] = 0x99;
+	ram_table->contrastFactor[2] = 0x90;
 	ram_table->contrastFactor[3] = 0x80;
 
 	ram_table->iir_curve[0] = 0x65;
diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h
index 470d7b89071a..574bf6e70763 100644
--- a/drivers/gpu/drm/amd/include/amd_shared.h
+++ b/drivers/gpu/drm/amd/include/amd_shared.h
@@ -137,6 +137,7 @@ enum DC_FEATURE_MASK {
 	DC_FBC_MASK = 0x1,
 };
 
+enum amd_dpm_forced_level;
 /**
  * struct amd_ip_funcs - general hooks for managing amdgpu IP Blocks
  */
@@ -186,6 +187,8 @@ struct amd_ip_funcs {
 				     enum amd_powergating_state state);
 	/** @get_clockgating_state: get current clockgating status */
 	void (*get_clockgating_state)(void *handle, u32 *flags);
+	/** @enable_umd_pstate: enable UMD powerstate */
+	int (*enable_umd_pstate)(void *handle, enum amd_dpm_forced_level *level);
 };
 
 
diff --git a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_1_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_1_0_offset.h
index 721c61171045..5a44e614ab7e 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_1_0_offset.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_1_0_offset.h
@@ -2347,6 +2347,8 @@
 #define mmHUBP0_DCHUBP_VMPG_CONFIG_BASE_IDX                                                            2
 #define mmHUBP0_HUBPREQ_DEBUG_DB                                                                       0x0569
 #define mmHUBP0_HUBPREQ_DEBUG_DB_BASE_IDX                                                              2
+#define mmHUBP0_HUBPREQ_DEBUG                                                                          0x056a
+#define mmHUBP0_HUBPREQ_DEBUG_BASE_IDX                                                                 2
 #define mmHUBP0_HUBP_MEASURE_WIN_CTRL_DCFCLK                                                           0x056e
 #define mmHUBP0_HUBP_MEASURE_WIN_CTRL_DCFCLK_BASE_IDX                                                  2
 #define mmHUBP0_HUBP_MEASURE_WIN_CTRL_DPPCLK                                                           0x056f
@@ -2631,6 +2633,8 @@
 #define mmHUBP1_DCHUBP_VMPG_CONFIG_BASE_IDX                                                            2
 #define mmHUBP1_HUBPREQ_DEBUG_DB                                                                       0x062d
 #define mmHUBP1_HUBPREQ_DEBUG_DB_BASE_IDX                                                              2
+#define mmHUBP1_HUBPREQ_DEBUG                                                                          0x062e
+#define mmHUBP1_HUBPREQ_DEBUG_BASE_IDX                                                                 2
 #define mmHUBP1_HUBP_MEASURE_WIN_CTRL_DCFCLK                                                           0x0632
 #define mmHUBP1_HUBP_MEASURE_WIN_CTRL_DCFCLK_BASE_IDX                                                  2
 #define mmHUBP1_HUBP_MEASURE_WIN_CTRL_DPPCLK                                                           0x0633
@@ -2915,6 +2919,8 @@
 #define mmHUBP2_DCHUBP_VMPG_CONFIG_BASE_IDX                                                            2
 #define mmHUBP2_HUBPREQ_DEBUG_DB                                                                       0x06f1
 #define mmHUBP2_HUBPREQ_DEBUG_DB_BASE_IDX                                                              2
+#define mmHUBP2_HUBPREQ_DEBUG                                                                          0x06f2
+#define mmHUBP2_HUBPREQ_DEBUG_BASE_IDX                                                                 2
 #define mmHUBP2_HUBP_MEASURE_WIN_CTRL_DCFCLK                                                           0x06f6
 #define mmHUBP2_HUBP_MEASURE_WIN_CTRL_DCFCLK_BASE_IDX                                                  2
 #define mmHUBP2_HUBP_MEASURE_WIN_CTRL_DPPCLK                                                           0x06f7
@@ -3199,6 +3205,8 @@
 #define mmHUBP3_DCHUBP_VMPG_CONFIG_BASE_IDX                                                            2
 #define mmHUBP3_HUBPREQ_DEBUG_DB                                                                       0x07b5
 #define mmHUBP3_HUBPREQ_DEBUG_DB_BASE_IDX                                                              2
+#define mmHUBP3_HUBPREQ_DEBUG                                                                          0x07b6
+#define mmHUBP3_HUBPREQ_DEBUG_BASE_IDX                                                                 2
 #define mmHUBP3_HUBP_MEASURE_WIN_CTRL_DCFCLK                                                           0x07ba
 #define mmHUBP3_HUBP_MEASURE_WIN_CTRL_DCFCLK_BASE_IDX                                                  2
 #define mmHUBP3_HUBP_MEASURE_WIN_CTRL_DPPCLK                                                           0x07bb
diff --git a/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_1_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_1_0_offset.h
index 442ca7c471a5..6109f5ad25ad 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_1_0_offset.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_1_0_offset.h
@@ -141,6 +141,8 @@
 #define mmUVD_GPCOM_VCPU_DATA0_BASE_IDX                                                                1
 #define mmUVD_GPCOM_VCPU_DATA1                                                                         0x03c5
 #define mmUVD_GPCOM_VCPU_DATA1_BASE_IDX                                                                1
+#define mmUVD_ENGINE_CNTL                                                                              0x03c6
+#define mmUVD_ENGINE_CNTL_BASE_IDX                                                                     1
 #define mmUVD_UDEC_DBW_UV_ADDR_CONFIG                                                                  0x03d2
 #define mmUVD_UDEC_DBW_UV_ADDR_CONFIG_BASE_IDX                                                         1
 #define mmUVD_UDEC_ADDR_CONFIG                                                                         0x03d3
diff --git a/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_1_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_1_0_sh_mask.h
index 63457f9df4c5..f84bed6eecb9 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_1_0_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_1_0_sh_mask.h
@@ -312,6 +312,11 @@
 //UVD_GPCOM_VCPU_DATA1
 #define UVD_GPCOM_VCPU_DATA1__DATA1__SHIFT                                                                    0x0
 #define UVD_GPCOM_VCPU_DATA1__DATA1_MASK                                                                      0xFFFFFFFFL
+//UVD_ENGINE_CNTL
+#define UVD_ENGINE_CNTL__ENGINE_START_MASK 0x1
+#define UVD_ENGINE_CNTL__ENGINE_START__SHIFT 0x0
+#define UVD_ENGINE_CNTL__ENGINE_START_MODE_MASK 0x2
+#define UVD_ENGINE_CNTL__ENGINE_START_MODE__SHIFT 0x1
 //UVD_UDEC_DBW_UV_ADDR_CONFIG
 #define UVD_UDEC_DBW_UV_ADDR_CONFIG__NUM_PIPES__SHIFT                                                         0x0
 #define UVD_UDEC_DBW_UV_ADDR_CONFIG__PIPE_INTERLEAVE_SIZE__SHIFT                                              0x3
diff --git a/drivers/gpu/drm/amd/include/atomfirmware.h b/drivers/gpu/drm/amd/include/atomfirmware.h
index 8eb0bb241210..d3075adb3297 100644
--- a/drivers/gpu/drm/amd/include/atomfirmware.h
+++ b/drivers/gpu/drm/amd/include/atomfirmware.h
@@ -494,6 +494,9 @@ enum atombios_firmware_capability
   ATOM_FIRMWARE_CAP_FIRMWARE_POSTED = 0x00000001,
   ATOM_FIRMWARE_CAP_GPU_VIRTUALIZATION  = 0x00000002,
   ATOM_FIRMWARE_CAP_WMI_SUPPORT  = 0x00000040,
+  ATOM_FIRMWARE_CAP_HWEMU_ENABLE  = 0x00000080,
+  ATOM_FIRMWARE_CAP_HWEMU_UMC_CFG = 0x00000100,
+  ATOM_FIRMWARE_CAP_SRAM_ECC      = 0x00000200,
 };
 
 enum atom_cooling_solution_id{
@@ -528,6 +531,35 @@ struct atom_firmware_info_v3_2 {
   uint32_t reserved2[3];
 };
 
+struct atom_firmware_info_v3_3
+{
+  struct atom_common_table_header table_header;
+  uint32_t firmware_revision;
+  uint32_t bootup_sclk_in10khz;
+  uint32_t bootup_mclk_in10khz;
+  uint32_t firmware_capability;             // enum atombios_firmware_capability
+  uint32_t main_call_parser_entry;          /* direct address of main parser call in VBIOS binary. */
+  uint32_t bios_scratch_reg_startaddr;      // 1st bios scratch register dword address
+  uint16_t bootup_vddc_mv;
+  uint16_t bootup_vddci_mv;
+  uint16_t bootup_mvddc_mv;
+  uint16_t bootup_vddgfx_mv;
+  uint8_t  mem_module_id;
+  uint8_t  coolingsolution_id;              /*0: Air cooling; 1: Liquid cooling ... */
+  uint8_t  reserved1[2];
+  uint32_t mc_baseaddr_high;
+  uint32_t mc_baseaddr_low;
+  uint8_t  board_i2c_feature_id;            // enum of atom_board_i2c_feature_id_def
+  uint8_t  board_i2c_feature_gpio_id;       // i2c id find in gpio_lut data table gpio_id
+  uint8_t  board_i2c_feature_slave_addr;
+  uint8_t  reserved3;
+  uint16_t bootup_mvddq_mv;
+  uint16_t bootup_mvpp_mv;
+  uint32_t zfbstartaddrin16mb;
+  uint32_t pplib_pptable_id;                // if pplib_pptable_id!=0, pplib get powerplay table inside driver instead of from VBIOS
+  uint32_t reserved2[2];
+};
+
 /* 
   ***************************************************************************
     Data Table lcd_info  structure
@@ -686,6 +718,7 @@ enum atom_encoder_caps_def
   ATOM_ENCODER_CAP_RECORD_HBR2_EN               =0x02,         // DP1.2 HBR2 setting is qualified and HBR2 can be enabled 
   ATOM_ENCODER_CAP_RECORD_HDMI6Gbps_EN          =0x04,         // HDMI2.0 6Gbps enable or not. 
   ATOM_ENCODER_CAP_RECORD_HBR3_EN               =0x08,         // DP1.3 HBR3 is supported by board. 
+  ATOM_ENCODER_CAP_RECORD_USB_C_TYPE            =0x100,        // the DP connector is a USB-C type.
 };
 
 struct  atom_encoder_caps_record
@@ -1226,16 +1259,17 @@ struct  atom_gfx_info_v2_3 {
   uint32_t rm21_sram_vmin_value;
 };
 
-struct  atom_gfx_info_v2_4 {
+struct  atom_gfx_info_v2_4
+{
   struct  atom_common_table_header  table_header;
   uint8_t gfxip_min_ver;
   uint8_t gfxip_max_ver;
-  uint8_t gc_num_se;
-  uint8_t max_tile_pipes;
-  uint8_t gc_num_cu_per_sh;
-  uint8_t gc_num_sh_per_se;
-  uint8_t gc_num_rb_per_se;
-  uint8_t gc_num_tccs;
+  uint8_t max_shader_engines;
+  uint8_t reserved;
+  uint8_t max_cu_per_sh;
+  uint8_t max_sh_per_se;
+  uint8_t max_backends_per_se;
+  uint8_t max_texture_channel_caches;
   uint32_t regaddr_cp_dma_src_addr;
   uint32_t regaddr_cp_dma_src_addr_hi;
   uint32_t regaddr_cp_dma_dst_addr;
@@ -1780,6 +1814,56 @@ struct atom_umc_info_v3_1
   uint32_t mem_refclk_10khz;
 };
 
+// umc_info.umc_config
+enum atom_umc_config_def {
+  UMC_CONFIG__ENABLE_1KB_INTERLEAVE_MODE  =   0x00000001,
+  UMC_CONFIG__DEFAULT_MEM_ECC_ENABLE      =   0x00000002,
+  UMC_CONFIG__ENABLE_HBM_LANE_REPAIR      =   0x00000004,
+  UMC_CONFIG__ENABLE_BANK_HARVESTING      =   0x00000008,
+  UMC_CONFIG__ENABLE_PHY_REINIT           =   0x00000010,
+  UMC_CONFIG__DISABLE_UCODE_CHKSTATUS     =   0x00000020,
+};
+
+struct atom_umc_info_v3_2
+{
+  struct  atom_common_table_header  table_header;
+  uint32_t ucode_version;
+  uint32_t ucode_rom_startaddr;
+  uint32_t ucode_length;
+  uint16_t umc_reg_init_offset;
+  uint16_t customer_ucode_name_offset;
+  uint16_t mclk_ss_percentage;
+  uint16_t mclk_ss_rate_10hz;
+  uint8_t umcip_min_ver;
+  uint8_t umcip_max_ver;
+  uint8_t vram_type;              //enum of atom_dgpu_vram_type
+  uint8_t umc_config;
+  uint32_t mem_refclk_10khz;
+  uint32_t pstate_uclk_10khz[4];
+  uint16_t umcgoldenoffset;
+  uint16_t densitygoldenoffset;
+};
+
+struct atom_umc_info_v3_3
+{
+  struct  atom_common_table_header  table_header;
+  uint32_t ucode_reserved;
+  uint32_t ucode_rom_startaddr;
+  uint32_t ucode_length;
+  uint16_t umc_reg_init_offset;
+  uint16_t customer_ucode_name_offset;
+  uint16_t mclk_ss_percentage;
+  uint16_t mclk_ss_rate_10hz;
+  uint8_t umcip_min_ver;
+  uint8_t umcip_max_ver;
+  uint8_t vram_type;              //enum of atom_dgpu_vram_type
+  uint8_t umc_config;
+  uint32_t mem_refclk_10khz;
+  uint32_t pstate_uclk_10khz[4];
+  uint16_t umcgoldenoffset;
+  uint16_t densitygoldenoffset;
+  uint32_t reserved[4];
+};
 
 /* 
   ***************************************************************************
diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
index 5f3c10ebff08..b897aca9b4c9 100644
--- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
+++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
@@ -85,18 +85,6 @@ enum kgd_memory_pool {
 	KGD_POOL_FRAMEBUFFER = 3,
 };
 
-enum kgd_engine_type {
-	KGD_ENGINE_PFP = 1,
-	KGD_ENGINE_ME,
-	KGD_ENGINE_CE,
-	KGD_ENGINE_MEC1,
-	KGD_ENGINE_MEC2,
-	KGD_ENGINE_RLC,
-	KGD_ENGINE_SDMA1,
-	KGD_ENGINE_SDMA2,
-	KGD_ENGINE_MAX
-};
-
 /**
  * enum kfd_sched_policy
  *
@@ -230,8 +218,6 @@ struct tile_config {
  * @hqd_sdma_destroy: Destructs and preempts the SDMA queue assigned to that
  * SDMA hqd slot.
  *
- * @get_fw_version: Returns FW versions from the header
- *
  * @set_scratch_backing_va: Sets VA for scratch backing memory of a VMID.
  * Only used for no cp scheduling mode
  *
@@ -311,8 +297,6 @@ struct kfd2kgd_calls {
 					struct kgd_dev *kgd,
 					uint8_t vmid);
 
-	uint16_t (*get_fw_version)(struct kgd_dev *kgd,
-				enum kgd_engine_type type);
 	void (*set_scratch_backing_va)(struct kgd_dev *kgd,
 				uint64_t va, uint32_t vmid);
 	int (*get_tile_config)(struct kgd_dev *kgd, struct tile_config *config);
diff --git a/drivers/gpu/drm/amd/include/linux/chash.h b/drivers/gpu/drm/amd/include/linux/chash.h
deleted file mode 100644
index 6dc159924ed1..000000000000
--- a/drivers/gpu/drm/amd/include/linux/chash.h
+++ /dev/null
@@ -1,366 +0,0 @@
-/*
- * Copyright 2017 Advanced Micro Devices, Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- *
- */
-
-#ifndef _LINUX_CHASH_H
-#define _LINUX_CHASH_H
-
-#include <linux/types.h>
-#include <linux/hash.h>
-#include <linux/bug.h>
-#include <asm/bitsperlong.h>
-
-#if BITS_PER_LONG == 32
-# define _CHASH_LONG_SHIFT 5
-#elif BITS_PER_LONG == 64
-# define _CHASH_LONG_SHIFT 6
-#else
-# error "Unexpected BITS_PER_LONG"
-#endif
-
-struct __chash_table {
-	u8 bits;
-	u8 key_size;
-	unsigned int value_size;
-	u32 size_mask;
-	unsigned long *occup_bitmap, *valid_bitmap;
-	union {
-		u32 *keys32;
-		u64 *keys64;
-	};
-	u8 *values;
-
-#ifdef CONFIG_CHASH_STATS
-	u64 hits, hits_steps, hits_time_ns;
-	u64 miss, miss_steps, miss_time_ns;
-	u64 relocs, reloc_dist;
-#endif
-};
-
-#define __CHASH_BITMAP_SIZE(bits)				\
-	(((1 << (bits)) + BITS_PER_LONG - 1) / BITS_PER_LONG)
-#define __CHASH_ARRAY_SIZE(bits, size)				\
-	((((size) << (bits)) + sizeof(long) - 1) / sizeof(long))
-
-#define __CHASH_DATA_SIZE(bits, key_size, value_size)	\
-	(__CHASH_BITMAP_SIZE(bits) * 2 +		\
-	 __CHASH_ARRAY_SIZE(bits, key_size) +		\
-	 __CHASH_ARRAY_SIZE(bits, value_size))
-
-#define STRUCT_CHASH_TABLE(bits, key_size, value_size)			\
-	struct {							\
-		struct __chash_table table;				\
-		unsigned long data					\
-			[__CHASH_DATA_SIZE(bits, key_size, value_size)];\
-	}
-
-/**
- * struct chash_table - Dynamically allocated closed hash table
- *
- * Use this struct for dynamically allocated hash tables (using
- * chash_table_alloc and chash_table_free), where the size is
- * determined at runtime.
- */
-struct chash_table {
-	struct __chash_table table;
-	unsigned long *data;
-};
-
-/**
- * DECLARE_CHASH_TABLE - macro to declare a closed hash table
- * @table: name of the declared hash table
- * @bts: Table size will be 2^bits entries
- * @key_sz: Size of hash keys in bytes, 4 or 8
- * @val_sz: Size of data values in bytes, can be 0
- *
- * This declares the hash table variable with a static size.
- *
- * The closed hash table stores key-value pairs with low memory and
- * lookup overhead. In operation it performs no dynamic memory
- * management. The data being stored does not require any
- * list_heads. The hash table performs best with small @val_sz and as
- * long as some space (about 50%) is left free in the table. But the
- * table can still work reasonably efficiently even when filled up to
- * about 90%. If bigger data items need to be stored and looked up,
- * store the pointer to it as value in the hash table.
- *
- * @val_sz may be 0. This can be useful when all the stored
- * information is contained in the key itself and the fact that it is
- * in the hash table (or not).
- */
-#define DECLARE_CHASH_TABLE(table, bts, key_sz, val_sz)		\
-	STRUCT_CHASH_TABLE(bts, key_sz, val_sz) table
-
-#ifdef CONFIG_CHASH_STATS
-#define __CHASH_STATS_INIT(prefix),		\
-		prefix.hits = 0,		\
-		prefix.hits_steps = 0,		\
-		prefix.hits_time_ns = 0,	\
-		prefix.miss = 0,		\
-		prefix.miss_steps = 0,		\
-		prefix.miss_time_ns = 0,	\
-		prefix.relocs = 0,		\
-		prefix.reloc_dist = 0
-#else
-#define __CHASH_STATS_INIT(prefix)
-#endif
-
-#define __CHASH_TABLE_INIT(prefix, data, bts, key_sz, val_sz)	\
-	prefix.bits = (bts),					\
-		prefix.key_size = (key_sz),			\
-		prefix.value_size = (val_sz),			\
-		prefix.size_mask = ((1 << bts) - 1),		\
-		prefix.occup_bitmap = &data[0],			\
-		prefix.valid_bitmap = &data			\
-			[__CHASH_BITMAP_SIZE(bts)],		\
-		prefix.keys64 = (u64 *)&data			\
-			[__CHASH_BITMAP_SIZE(bts) * 2],		\
-		prefix.values = (u8 *)&data			\
-			[__CHASH_BITMAP_SIZE(bts) * 2 +		\
-			 __CHASH_ARRAY_SIZE(bts, key_sz)]	\
-		__CHASH_STATS_INIT(prefix)
-
-/**
- * DEFINE_CHASH_TABLE - macro to define and initialize a closed hash table
- * @tbl: name of the declared hash table
- * @bts: Table size will be 2^bits entries
- * @key_sz: Size of hash keys in bytes, 4 or 8
- * @val_sz: Size of data values in bytes, can be 0
- *
- * Note: the macro can be used for global and local hash table variables.
- */
-#define DEFINE_CHASH_TABLE(tbl, bts, key_sz, val_sz)			\
-	DECLARE_CHASH_TABLE(tbl, bts, key_sz, val_sz) = {		\
-		.table = {						\
-			__CHASH_TABLE_INIT(, (tbl).data, bts, key_sz, val_sz) \
-		},							\
-		.data = {0}						\
-	}
-
-/**
- * INIT_CHASH_TABLE - Initialize a hash table declared by DECLARE_CHASH_TABLE
- * @tbl: name of the declared hash table
- * @bts: Table size will be 2^bits entries
- * @key_sz: Size of hash keys in bytes, 4 or 8
- * @val_sz: Size of data values in bytes, can be 0
- */
-#define INIT_CHASH_TABLE(tbl, bts, key_sz, val_sz)			\
-	__CHASH_TABLE_INIT(((tbl).table), (tbl).data, bts, key_sz, val_sz)
-
-int chash_table_alloc(struct chash_table *table, u8 bits, u8 key_size,
-		      unsigned int value_size, gfp_t gfp_mask);
-void chash_table_free(struct chash_table *table);
-
-/**
- * chash_table_dump_stats - Dump statistics of a closed hash table
- * @tbl: Pointer to the table structure
- *
- * Dumps some performance statistics of the table gathered in operation
- * in the kernel log using pr_debug. If CONFIG_DYNAMIC_DEBUG is enabled,
- * user must turn on messages for chash.c (file chash.c +p).
- */
-#ifdef CONFIG_CHASH_STATS
-#define chash_table_dump_stats(tbl) __chash_table_dump_stats(&(*tbl).table)
-
-void __chash_table_dump_stats(struct __chash_table *table);
-#else
-#define chash_table_dump_stats(tbl)
-#endif
-
-/**
- * chash_table_reset_stats - Reset statistics of a closed hash table
- * @tbl: Pointer to the table structure
- */
-#ifdef CONFIG_CHASH_STATS
-#define chash_table_reset_stats(tbl) __chash_table_reset_stats(&(*tbl).table)
-
-static inline void __chash_table_reset_stats(struct __chash_table *table)
-{
-	(void)table __CHASH_STATS_INIT((*table));
-}
-#else
-#define chash_table_reset_stats(tbl)
-#endif
-
-/**
- * chash_table_copy_in - Copy a new value into the hash table
- * @tbl: Pointer to the table structure
- * @key: Key of the entry to add or update
- * @value: Pointer to value to copy, may be NULL
- *
- * If @key already has an entry, its value is replaced. Otherwise a
- * new entry is added. If @value is NULL, the value is left unchanged
- * or uninitialized. Returns 1 if an entry already existed, 0 if a new
- * entry was added or %-ENOMEM if there was no free space in the
- * table.
- */
-#define chash_table_copy_in(tbl, key, value)			\
-	__chash_table_copy_in(&(*tbl).table, key, value)
-
-int __chash_table_copy_in(struct __chash_table *table, u64 key,
-			  const void *value);
-
-/**
- * chash_table_copy_out - Copy a value out of the hash table
- * @tbl: Pointer to the table structure
- * @key: Key of the entry to find
- * @value: Pointer to value to copy, may be NULL
- *
- * If @value is not NULL and the table has a non-0 value_size, the
- * value at @key is copied to @value. Returns the slot index of the
- * entry or %-EINVAL if @key was not found.
- */
-#define chash_table_copy_out(tbl, key, value)			\
-	__chash_table_copy_out(&(*tbl).table, key, value, false)
-
-int __chash_table_copy_out(struct __chash_table *table, u64 key,
-			   void *value, bool remove);
-
-/**
- * chash_table_remove - Remove an entry from the hash table
- * @tbl: Pointer to the table structure
- * @key: Key of the entry to find
- * @value: Pointer to value to copy, may be NULL
- *
- * If @value is not NULL and the table has a non-0 value_size, the
- * value at @key is copied to @value. The entry is removed from the
- * table. Returns the slot index of the removed entry or %-EINVAL if
- * @key was not found.
- */
-#define chash_table_remove(tbl, key, value)			\
-	__chash_table_copy_out(&(*tbl).table, key, value, true)
-
-/*
- * Low level iterator API used internally by the above functions.
- */
-struct chash_iter {
-	struct __chash_table *table;
-	unsigned long mask;
-	int slot;
-};
-
-/**
- * CHASH_ITER_INIT - Initialize a hash table iterator
- * @tbl: Pointer to hash table to iterate over
- * @s: Initial slot number
- */
-#define CHASH_ITER_INIT(table, s) {			\
-		table,					\
-		1UL << ((s) & (BITS_PER_LONG - 1)),	\
-		s					\
-	}
-/**
- * CHASH_ITER_SET - Set hash table iterator to new slot
- * @iter: Iterator
- * @s: Slot number
- */
-#define CHASH_ITER_SET(iter, s)					\
-	(iter).mask = 1UL << ((s) & (BITS_PER_LONG - 1)),	\
-	(iter).slot = (s)
-/**
- * CHASH_ITER_INC - Increment hash table iterator
- * @table: Hash table to iterate over
- *
- * Wraps around at the end.
- */
-#define CHASH_ITER_INC(iter) do {					\
-		(iter).mask = (iter).mask << 1 |			\
-			(iter).mask >> (BITS_PER_LONG - 1);		\
-		(iter).slot = ((iter).slot + 1) & (iter).table->size_mask; \
-	} while (0)
-
-static inline bool chash_iter_is_valid(const struct chash_iter iter)
-{
-	BUG_ON((unsigned)iter.slot >= (1 << iter.table->bits));
-	return !!(iter.table->valid_bitmap[iter.slot >> _CHASH_LONG_SHIFT] &
-		  iter.mask);
-}
-static inline bool chash_iter_is_empty(const struct chash_iter iter)
-{
-	BUG_ON((unsigned)iter.slot >= (1 << iter.table->bits));
-	return !(iter.table->occup_bitmap[iter.slot >> _CHASH_LONG_SHIFT] &
-		 iter.mask);
-}
-
-static inline void chash_iter_set_valid(const struct chash_iter iter)
-{
-	BUG_ON((unsigned)iter.slot >= (1 << iter.table->bits));
-	iter.table->valid_bitmap[iter.slot >> _CHASH_LONG_SHIFT] |= iter.mask;
-	iter.table->occup_bitmap[iter.slot >> _CHASH_LONG_SHIFT] |= iter.mask;
-}
-static inline void chash_iter_set_invalid(const struct chash_iter iter)
-{
-	BUG_ON((unsigned)iter.slot >= (1 << iter.table->bits));
-	iter.table->valid_bitmap[iter.slot >> _CHASH_LONG_SHIFT] &= ~iter.mask;
-}
-static inline void chash_iter_set_empty(const struct chash_iter iter)
-{
-	BUG_ON((unsigned)iter.slot >= (1 << iter.table->bits));
-	iter.table->occup_bitmap[iter.slot >> _CHASH_LONG_SHIFT] &= ~iter.mask;
-}
-
-static inline u32 chash_iter_key32(const struct chash_iter iter)
-{
-	BUG_ON(iter.table->key_size != 4);
-	BUG_ON((unsigned)iter.slot >= (1 << iter.table->bits));
-	return iter.table->keys32[iter.slot];
-}
-static inline u64 chash_iter_key64(const struct chash_iter iter)
-{
-	BUG_ON(iter.table->key_size != 8);
-	BUG_ON((unsigned)iter.slot >= (1 << iter.table->bits));
-	return iter.table->keys64[iter.slot];
-}
-static inline u64 chash_iter_key(const struct chash_iter iter)
-{
-	BUG_ON((unsigned)iter.slot >= (1 << iter.table->bits));
-	return (iter.table->key_size == 4) ?
-		iter.table->keys32[iter.slot] : iter.table->keys64[iter.slot];
-}
-
-static inline u32 chash_iter_hash32(const struct chash_iter iter)
-{
-	BUG_ON(iter.table->key_size != 4);
-	return hash_32(chash_iter_key32(iter), iter.table->bits);
-}
-
-static inline u32 chash_iter_hash64(const struct chash_iter iter)
-{
-	BUG_ON(iter.table->key_size != 8);
-	return hash_64(chash_iter_key64(iter), iter.table->bits);
-}
-
-static inline u32 chash_iter_hash(const struct chash_iter iter)
-{
-	return (iter.table->key_size == 4) ?
-		hash_32(chash_iter_key32(iter), iter.table->bits) :
-		hash_64(chash_iter_key64(iter), iter.table->bits);
-}
-
-static inline void *chash_iter_value(const struct chash_iter iter)
-{
-	BUG_ON((unsigned)iter.slot >= (1 << iter.table->bits));
-	return iter.table->values +
-		((unsigned long)iter.slot * iter.table->value_size);
-}
-
-#endif /* _LINUX_CHASH_H */
diff --git a/drivers/gpu/drm/amd/lib/Kconfig b/drivers/gpu/drm/amd/lib/Kconfig
deleted file mode 100644
index 776ef3434c10..000000000000
--- a/drivers/gpu/drm/amd/lib/Kconfig
+++ /dev/null
@@ -1,28 +0,0 @@
-menu "AMD Library routines"
-
-#
-# Closed hash table
-#
-config CHASH
-	tristate
-	default DRM_AMDGPU
-	help
-	 Statically sized closed hash table implementation with low
-	 memory and CPU overhead.
-
-config CHASH_STATS
-	bool "Closed hash table performance statistics"
-	depends on CHASH
-	default n
-	help
-	 Enable collection of performance statistics for closed hash tables.
-
-config CHASH_SELFTEST
-	bool "Closed hash table self test"
-	depends on CHASH
-	default n
-	help
-	 Runs a selftest during module load. Several module parameters
-	 are available to modify the behaviour of the test.
-
-endmenu
diff --git a/drivers/gpu/drm/amd/lib/Makefile b/drivers/gpu/drm/amd/lib/Makefile
deleted file mode 100644
index 690243001e1a..000000000000
--- a/drivers/gpu/drm/amd/lib/Makefile
+++ /dev/null
@@ -1,32 +0,0 @@
-#
-# Copyright 2017 Advanced Micro Devices, Inc.
-#
-# Permission is hereby granted, free of charge, to any person obtaining a
-# copy of this software and associated documentation files (the "Software"),
-# to deal in the Software without restriction, including without limitation
-# the rights to use, copy, modify, merge, publish, distribute, sublicense,
-# and/or sell copies of the Software, and to permit persons to whom the
-# Software is furnished to do so, subject to the following conditions:
-#
-# The above copyright notice and this permission notice shall be included in
-# all copies or substantial portions of the Software.
-#
-# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
-# THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
-# OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
-# ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
-# OTHER DEALINGS IN THE SOFTWARE.
-#
-#
-# Makefile for AMD library routines, which are used by AMD driver
-# components.
-#
-# This is for common library routines that can be shared between AMD
-# driver components or later moved to kernel/lib for sharing with
-# other drivers.
-
-ccflags-y := -I$(src)/../include
-
-obj-$(CONFIG_CHASH) += chash.o
diff --git a/drivers/gpu/drm/amd/lib/chash.c b/drivers/gpu/drm/amd/lib/chash.c
deleted file mode 100644
index b8e45f356a1c..000000000000
--- a/drivers/gpu/drm/amd/lib/chash.c
+++ /dev/null
@@ -1,638 +0,0 @@
-/*
- * Copyright 2017 Advanced Micro Devices, Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- *
- */
-
-#include <linux/types.h>
-#include <linux/hash.h>
-#include <linux/bug.h>
-#include <linux/slab.h>
-#include <linux/module.h>
-#include <linux/sched/clock.h>
-#include <asm/div64.h>
-#include <linux/chash.h>
-
-/**
- * chash_table_alloc - Allocate closed hash table
- * @table: Pointer to the table structure
- * @bits: Table size will be 2^bits entries
- * @key_size: Size of hash keys in bytes, 4 or 8
- * @value_size: Size of data values in bytes, can be 0
- */
-int chash_table_alloc(struct chash_table *table, u8 bits, u8 key_size,
-		      unsigned int value_size, gfp_t gfp_mask)
-{
-	if (bits > 31)
-		return -EINVAL;
-
-	if (key_size != 4 && key_size != 8)
-		return -EINVAL;
-
-	table->data = kcalloc(__CHASH_DATA_SIZE(bits, key_size, value_size),
-		       sizeof(long), gfp_mask);
-	if (!table->data)
-		return -ENOMEM;
-
-	__CHASH_TABLE_INIT(table->table, table->data,
-			   bits, key_size, value_size);
-
-	return 0;
-}
-EXPORT_SYMBOL(chash_table_alloc);
-
-/**
- * chash_table_free - Free closed hash table
- * @table: Pointer to the table structure
- */
-void chash_table_free(struct chash_table *table)
-{
-	kfree(table->data);
-}
-EXPORT_SYMBOL(chash_table_free);
-
-#ifdef CONFIG_CHASH_STATS
-
-#define DIV_FRAC(nom, denom, quot, frac, frac_digits) do {		\
-		u64 __nom = (nom);					\
-		u64 __denom = (denom);					\
-		u64 __quot, __frac;					\
-		u32 __rem;						\
-									\
-		while (__denom >> 32) {					\
-			__nom   >>= 1;					\
-			__denom >>= 1;					\
-		}							\
-		__quot = __nom;						\
-		__rem  = do_div(__quot, __denom);			\
-		__frac = __rem * (frac_digits) + (__denom >> 1);	\
-		do_div(__frac, __denom);				\
-		(quot) = __quot;					\
-		(frac) = __frac;					\
-	} while (0)
-
-void __chash_table_dump_stats(struct __chash_table *table)
-{
-	struct chash_iter iter = CHASH_ITER_INIT(table, 0);
-	u32 filled = 0, empty = 0, tombstones = 0;
-	u64 quot1, quot2;
-	u32 frac1, frac2;
-
-	do {
-		if (chash_iter_is_valid(iter))
-			filled++;
-		else if (chash_iter_is_empty(iter))
-			empty++;
-		else
-			tombstones++;
-		CHASH_ITER_INC(iter);
-	} while (iter.slot);
-
-	pr_debug("chash: key size %u, value size %u\n",
-		 table->key_size, table->value_size);
-	pr_debug("  Slots total/filled/empty/tombstones: %u / %u / %u / %u\n",
-		 1 << table->bits, filled, empty, tombstones);
-	if (table->hits > 0) {
-		DIV_FRAC(table->hits_steps, table->hits, quot1, frac1, 1000);
-		DIV_FRAC(table->hits * 1000, table->hits_time_ns,
-			 quot2, frac2, 1000);
-	} else {
-		quot1 = quot2 = 0;
-		frac1 = frac2 = 0;
-	}
-	pr_debug("  Hits   (avg.cost, rate): %llu (%llu.%03u, %llu.%03u M/s)\n",
-		 table->hits, quot1, frac1, quot2, frac2);
-	if (table->miss > 0) {
-		DIV_FRAC(table->miss_steps, table->miss, quot1, frac1, 1000);
-		DIV_FRAC(table->miss * 1000, table->miss_time_ns,
-			 quot2, frac2, 1000);
-	} else {
-		quot1 = quot2 = 0;
-		frac1 = frac2 = 0;
-	}
-	pr_debug("  Misses (avg.cost, rate): %llu (%llu.%03u, %llu.%03u M/s)\n",
-		 table->miss, quot1, frac1, quot2, frac2);
-	if (table->hits + table->miss > 0) {
-		DIV_FRAC(table->hits_steps + table->miss_steps,
-			 table->hits + table->miss, quot1, frac1, 1000);
-		DIV_FRAC((table->hits + table->miss) * 1000,
-			 (table->hits_time_ns + table->miss_time_ns),
-			 quot2, frac2, 1000);
-	} else {
-		quot1 = quot2 = 0;
-		frac1 = frac2 = 0;
-	}
-	pr_debug("  Total  (avg.cost, rate): %llu (%llu.%03u, %llu.%03u M/s)\n",
-		 table->hits + table->miss, quot1, frac1, quot2, frac2);
-	if (table->relocs > 0) {
-		DIV_FRAC(table->hits + table->miss, table->relocs,
-			 quot1, frac1, 1000);
-		DIV_FRAC(table->reloc_dist, table->relocs, quot2, frac2, 1000);
-		pr_debug("  Relocations (freq, avg.dist): %llu (1:%llu.%03u, %llu.%03u)\n",
-			 table->relocs, quot1, frac1, quot2, frac2);
-	} else {
-		pr_debug("  No relocations\n");
-	}
-}
-EXPORT_SYMBOL(__chash_table_dump_stats);
-
-#undef DIV_FRAC
-#endif
-
-#define CHASH_INC(table, a) ((a) = ((a) + 1) & (table)->size_mask)
-#define CHASH_ADD(table, a, b) (((a) + (b)) & (table)->size_mask)
-#define CHASH_SUB(table, a, b) (((a) - (b)) & (table)->size_mask)
-#define CHASH_IN_RANGE(table, slot, first, last) \
-	(CHASH_SUB(table, slot, first) <= CHASH_SUB(table, last, first))
-
-/*#define CHASH_DEBUG Uncomment this to enable verbose debug output*/
-#ifdef CHASH_DEBUG
-static void chash_table_dump(struct __chash_table *table)
-{
-	struct chash_iter iter = CHASH_ITER_INIT(table, 0);
-
-	do {
-		if ((iter.slot & 3) == 0)
-			pr_debug("%04x: ", iter.slot);
-
-		if (chash_iter_is_valid(iter))
-			pr_debug("[%016llx] ", chash_iter_key(iter));
-		else if (chash_iter_is_empty(iter))
-			pr_debug("[    <empty>     ] ");
-		else
-			pr_debug("[  <tombstone>   ] ");
-
-		if ((iter.slot & 3) == 3)
-			pr_debug("\n");
-
-		CHASH_ITER_INC(iter);
-	} while (iter.slot);
-
-	if ((iter.slot & 3) != 0)
-		pr_debug("\n");
-}
-
-static int chash_table_check(struct __chash_table *table)
-{
-	u32 hash;
-	struct chash_iter iter = CHASH_ITER_INIT(table, 0);
-	struct chash_iter cur = CHASH_ITER_INIT(table, 0);
-
-	do {
-		if (!chash_iter_is_valid(iter)) {
-			CHASH_ITER_INC(iter);
-			continue;
-		}
-
-		hash = chash_iter_hash(iter);
-		CHASH_ITER_SET(cur, hash);
-		while (cur.slot != iter.slot) {
-			if (chash_iter_is_empty(cur)) {
-				pr_err("Path to element at %x with hash %x broken at slot %x\n",
-				       iter.slot, hash, cur.slot);
-				chash_table_dump(table);
-				return -EINVAL;
-			}
-			CHASH_ITER_INC(cur);
-		}
-
-		CHASH_ITER_INC(iter);
-	} while (iter.slot);
-
-	return 0;
-}
-#endif
-
-static void chash_iter_relocate(struct chash_iter dst, struct chash_iter src)
-{
-	BUG_ON(src.table == dst.table && src.slot == dst.slot);
-	BUG_ON(src.table->key_size != dst.table->key_size);
-	BUG_ON(src.table->value_size != dst.table->value_size);
-
-	if (dst.table->key_size == 4)
-		dst.table->keys32[dst.slot] = src.table->keys32[src.slot];
-	else
-		dst.table->keys64[dst.slot] = src.table->keys64[src.slot];
-
-	if (dst.table->value_size)
-		memcpy(chash_iter_value(dst), chash_iter_value(src),
-		       dst.table->value_size);
-
-	chash_iter_set_valid(dst);
-	chash_iter_set_invalid(src);
-
-#ifdef CONFIG_CHASH_STATS
-	if (src.table == dst.table) {
-		dst.table->relocs++;
-		dst.table->reloc_dist +=
-			CHASH_SUB(dst.table, src.slot, dst.slot);
-	}
-#endif
-}
-
-/**
- * __chash_table_find - Helper for looking up a hash table entry
- * @iter: Pointer to hash table iterator
- * @key: Key of the entry to find
- * @for_removal: set to true if the element will be removed soon
- *
- * Searches for an entry in the hash table with a given key. iter must
- * be initialized by the caller to point to the home position of the
- * hypothetical entry, i.e. it must be initialized with the hash table
- * and the key's hash as the initial slot for the search.
- *
- * This function also does some local clean-up to speed up future
- * look-ups by relocating entries to better slots and removing
- * tombstones that are no longer needed.
- *
- * If @for_removal is true, the function avoids relocating the entry
- * that is being returned.
- *
- * Returns 0 if the search is successful. In this case iter is updated
- * to point to the found entry. Otherwise %-EINVAL is returned and the
- * iter is updated to point to the first available slot for the given
- * key. If the table is full, the slot is set to -1.
- */
-static int chash_table_find(struct chash_iter *iter, u64 key,
-			    bool for_removal)
-{
-#ifdef CONFIG_CHASH_STATS
-	u64 ts1 = local_clock();
-#endif
-	u32 hash = iter->slot;
-	struct chash_iter first_redundant = CHASH_ITER_INIT(iter->table, -1);
-	int first_avail = (for_removal ? -2 : -1);
-
-	while (!chash_iter_is_valid(*iter) || chash_iter_key(*iter) != key) {
-		if (chash_iter_is_empty(*iter)) {
-			/* Found an empty slot, which ends the
-			 * search. Clean up any preceding tombstones
-			 * that are no longer needed because they lead
-			 * to no-where
-			 */
-			if ((int)first_redundant.slot < 0)
-				goto not_found;
-			while (first_redundant.slot != iter->slot) {
-				if (!chash_iter_is_valid(first_redundant))
-					chash_iter_set_empty(first_redundant);
-				CHASH_ITER_INC(first_redundant);
-			}
-#ifdef CHASH_DEBUG
-			chash_table_check(iter->table);
-#endif
-			goto not_found;
-		} else if (!chash_iter_is_valid(*iter)) {
-			/* Found a tombstone. Remember it as candidate
-			 * for relocating the entry we're looking for
-			 * or for adding a new entry with the given key
-			 */
-			if (first_avail == -1)
-				first_avail = iter->slot;
-			/* Or mark it as the start of a series of
-			 * potentially redundant tombstones
-			 */
-			else if (first_redundant.slot == -1)
-				CHASH_ITER_SET(first_redundant, iter->slot);
-		} else if (first_redundant.slot >= 0) {
-			/* Found a valid, occupied slot with a
-			 * preceding series of tombstones. Relocate it
-			 * to a better position that no longer depends
-			 * on those tombstones
-			 */
-			u32 cur_hash = chash_iter_hash(*iter);
-
-			if (!CHASH_IN_RANGE(iter->table, cur_hash,
-					    first_redundant.slot + 1,
-					    iter->slot)) {
-				/* This entry has a hash at or before
-				 * the first tombstone we found. We
-				 * can relocate it to that tombstone
-				 * and advance to the next tombstone
-				 */
-				chash_iter_relocate(first_redundant, *iter);
-				do {
-					CHASH_ITER_INC(first_redundant);
-				} while (chash_iter_is_valid(first_redundant));
-			} else if (cur_hash != iter->slot) {
-				/* Relocate entry to its home position
-				 * or as close as possible so it no
-				 * longer depends on any preceding
-				 * tombstones
-				 */
-				struct chash_iter new_iter =
-					CHASH_ITER_INIT(iter->table, cur_hash);
-
-				while (new_iter.slot != iter->slot &&
-				       chash_iter_is_valid(new_iter))
-					CHASH_ITER_INC(new_iter);
-
-				if (new_iter.slot != iter->slot)
-					chash_iter_relocate(new_iter, *iter);
-			}
-		}
-
-		CHASH_ITER_INC(*iter);
-		if (iter->slot == hash) {
-			iter->slot = -1;
-			goto not_found;
-		}
-	}
-
-#ifdef CONFIG_CHASH_STATS
-	iter->table->hits++;
-	iter->table->hits_steps += CHASH_SUB(iter->table, iter->slot, hash) + 1;
-#endif
-
-	if (first_avail >= 0) {
-		CHASH_ITER_SET(first_redundant, first_avail);
-		chash_iter_relocate(first_redundant, *iter);
-		iter->slot = first_redundant.slot;
-		iter->mask = first_redundant.mask;
-	}
-
-#ifdef CONFIG_CHASH_STATS
-	iter->table->hits_time_ns += local_clock() - ts1;
-#endif
-
-	return 0;
-
-not_found:
-#ifdef CONFIG_CHASH_STATS
-	iter->table->miss++;
-	iter->table->miss_steps += (iter->slot < 0) ?
-		(1 << iter->table->bits) :
-		CHASH_SUB(iter->table, iter->slot, hash) + 1;
-#endif
-
-	if (first_avail >= 0)
-		CHASH_ITER_SET(*iter, first_avail);
-
-#ifdef CONFIG_CHASH_STATS
-	iter->table->miss_time_ns += local_clock() - ts1;
-#endif
-
-	return -EINVAL;
-}
-
-int __chash_table_copy_in(struct __chash_table *table, u64 key,
-			  const void *value)
-{
-	u32 hash = (table->key_size == 4) ?
-		hash_32(key, table->bits) : hash_64(key, table->bits);
-	struct chash_iter iter = CHASH_ITER_INIT(table, hash);
-	int r = chash_table_find(&iter, key, false);
-
-	/* Found an existing entry */
-	if (!r) {
-		if (value && table->value_size)
-			memcpy(chash_iter_value(iter), value,
-			       table->value_size);
-		return 1;
-	}
-
-	/* Is there a place to add a new entry? */
-	if (iter.slot < 0) {
-		pr_err("Hash table overflow\n");
-		return -ENOMEM;
-	}
-
-	chash_iter_set_valid(iter);
-
-	if (table->key_size == 4)
-		table->keys32[iter.slot] = key;
-	else
-		table->keys64[iter.slot] = key;
-	if (value && table->value_size)
-		memcpy(chash_iter_value(iter), value, table->value_size);
-
-	return 0;
-}
-EXPORT_SYMBOL(__chash_table_copy_in);
-
-int __chash_table_copy_out(struct __chash_table *table, u64 key,
-			   void *value, bool remove)
-{
-	u32 hash = (table->key_size == 4) ?
-		hash_32(key, table->bits) : hash_64(key, table->bits);
-	struct chash_iter iter = CHASH_ITER_INIT(table, hash);
-	int r = chash_table_find(&iter, key, remove);
-
-	if (r < 0)
-		return r;
-
-	if (value && table->value_size)
-		memcpy(value, chash_iter_value(iter), table->value_size);
-
-	if (remove)
-		chash_iter_set_invalid(iter);
-
-	return iter.slot;
-}
-EXPORT_SYMBOL(__chash_table_copy_out);
-
-#ifdef CONFIG_CHASH_SELFTEST
-/**
- * chash_self_test - Run a self-test of the hash table implementation
- * @bits: Table size will be 2^bits entries
- * @key_size: Size of hash keys in bytes, 4 or 8
- * @min_fill: Minimum fill level during the test
- * @max_fill: Maximum fill level during the test
- * @iterations: Number of test iterations
- *
- * The test adds and removes entries from a hash table, cycling the
- * fill level between min_fill and max_fill entries. Also tests lookup
- * and value retrieval.
- */
-static int __init chash_self_test(u8 bits, u8 key_size,
-				  int min_fill, int max_fill,
-				  u64 iterations)
-{
-	struct chash_table table;
-	int ret;
-	u64 add_count, rmv_count;
-	u64 value;
-
-	if (key_size == 4 && iterations > 0xffffffff)
-		return -EINVAL;
-	if (min_fill >= max_fill)
-		return -EINVAL;
-
-	ret = chash_table_alloc(&table, bits, key_size, sizeof(u64),
-				GFP_KERNEL);
-	if (ret) {
-		pr_err("chash_table_alloc failed: %d\n", ret);
-		return ret;
-	}
-
-	for (add_count = 0, rmv_count = 0; add_count < iterations;
-	     add_count++) {
-		/* When we hit the max_fill level, remove entries down
-		 * to min_fill
-		 */
-		if (add_count - rmv_count == max_fill) {
-			u64 find_count = rmv_count;
-
-			/* First try to find all entries that we're
-			 * about to remove, confirm their value, test
-			 * writing them back a second time.
-			 */
-			for (; add_count - find_count > min_fill;
-			     find_count++) {
-				ret = chash_table_copy_out(&table, find_count,
-							   &value);
-				if (ret < 0) {
-					pr_err("chash_table_copy_out failed: %d\n",
-					       ret);
-					goto out;
-				}
-				if (value != ~find_count) {
-					pr_err("Wrong value retrieved for key 0x%llx, expected 0x%llx got 0x%llx\n",
-					       find_count, ~find_count, value);
-#ifdef CHASH_DEBUG
-					chash_table_dump(&table.table);
-#endif
-					ret = -EFAULT;
-					goto out;
-				}
-				ret = chash_table_copy_in(&table, find_count,
-							  &value);
-				if (ret != 1) {
-					pr_err("copy_in second time returned %d, expected 1\n",
-					       ret);
-					ret = -EFAULT;
-					goto out;
-				}
-			}
-			/* Remove them until we hit min_fill level */
-			for (; add_count - rmv_count > min_fill; rmv_count++) {
-				ret = chash_table_remove(&table, rmv_count,
-							 NULL);
-				if (ret < 0) {
-					pr_err("chash_table_remove failed: %d\n",
-					       ret);
-					goto out;
-				}
-			}
-		}
-
-		/* Add a new value */
-		value = ~add_count;
-		ret = chash_table_copy_in(&table, add_count, &value);
-		if (ret != 0) {
-			pr_err("copy_in first time returned %d, expected 0\n",
-			       ret);
-			ret = -EFAULT;
-			goto out;
-		}
-	}
-
-	chash_table_dump_stats(&table);
-	chash_table_reset_stats(&table);
-
-out:
-	chash_table_free(&table);
-	return ret;
-}
-
-static unsigned int chash_test_bits = 10;
-MODULE_PARM_DESC(test_bits,
-		 "Selftest number of hash bits ([4..20], default=10)");
-module_param_named(test_bits, chash_test_bits, uint, 0444);
-
-static unsigned int chash_test_keysize = 8;
-MODULE_PARM_DESC(test_keysize, "Selftest keysize (4 or 8, default=8)");
-module_param_named(test_keysize, chash_test_keysize, uint, 0444);
-
-static unsigned int chash_test_minfill;
-MODULE_PARM_DESC(test_minfill, "Selftest minimum #entries (default=50%)");
-module_param_named(test_minfill, chash_test_minfill, uint, 0444);
-
-static unsigned int chash_test_maxfill;
-MODULE_PARM_DESC(test_maxfill, "Selftest maximum #entries (default=80%)");
-module_param_named(test_maxfill, chash_test_maxfill, uint, 0444);
-
-static unsigned long chash_test_iters;
-MODULE_PARM_DESC(test_iters, "Selftest iterations (default=1000 x #entries)");
-module_param_named(test_iters, chash_test_iters, ulong, 0444);
-
-static int __init chash_init(void)
-{
-	int ret;
-	u64 ts1_ns;
-
-	/* Skip self test on user errors */
-	if (chash_test_bits < 4 || chash_test_bits > 20) {
-		pr_err("chash: test_bits out of range [4..20].\n");
-		return 0;
-	}
-	if (chash_test_keysize != 4 && chash_test_keysize != 8) {
-		pr_err("chash: test_keysize invalid. Must be 4 or 8.\n");
-		return 0;
-	}
-
-	if (!chash_test_minfill)
-		chash_test_minfill = (1 << chash_test_bits) / 2;
-	if (!chash_test_maxfill)
-		chash_test_maxfill = (1 << chash_test_bits) * 4 / 5;
-	if (!chash_test_iters)
-		chash_test_iters = (1 << chash_test_bits) * 1000;
-
-	if (chash_test_minfill >= (1 << chash_test_bits)) {
-		pr_err("chash: test_minfill too big. Must be < table size.\n");
-		return 0;
-	}
-	if (chash_test_maxfill >= (1 << chash_test_bits)) {
-		pr_err("chash: test_maxfill too big. Must be < table size.\n");
-		return 0;
-	}
-	if (chash_test_minfill >= chash_test_maxfill) {
-		pr_err("chash: test_minfill must be < test_maxfill.\n");
-		return 0;
-	}
-	if (chash_test_keysize == 4 && chash_test_iters > 0xffffffff) {
-		pr_err("chash: test_iters must be < 4G for 4 byte keys.\n");
-		return 0;
-	}
-
-	ts1_ns = local_clock();
-	ret = chash_self_test(chash_test_bits, chash_test_keysize,
-			      chash_test_minfill, chash_test_maxfill,
-			      chash_test_iters);
-	if (!ret) {
-		u64 ts_delta_us = local_clock() - ts1_ns;
-		u64 iters_per_second = (u64)chash_test_iters * 1000000;
-
-		do_div(ts_delta_us, 1000);
-		do_div(iters_per_second, ts_delta_us);
-		pr_info("chash: self test took %llu us, %llu iterations/s\n",
-			ts_delta_us, iters_per_second);
-	} else {
-		pr_err("chash: self test failed: %d\n", ret);
-	}
-
-	return ret;
-}
-
-module_init(chash_init);
-
-#endif /* CONFIG_CHASH_SELFTEST */
-
-MODULE_DESCRIPTION("Closed hash table");
-MODULE_LICENSE("GPL and additional rights");
diff --git a/drivers/gpu/drm/amd/powerplay/Makefile b/drivers/gpu/drm/amd/powerplay/Makefile
index 231785a9e24c..ec87b3430d12 100644
--- a/drivers/gpu/drm/amd/powerplay/Makefile
+++ b/drivers/gpu/drm/amd/powerplay/Makefile
@@ -35,7 +35,7 @@ AMD_POWERPLAY = $(addsuffix /Makefile,$(addprefix $(FULL_AMD_PATH)/powerplay/,$(
 
 include $(AMD_POWERPLAY)
 
-POWER_MGR = amd_powerplay.o
+POWER_MGR = amd_powerplay.o amdgpu_smu.o smu_v11_0.o vega20_ppt.o
 
 AMD_PP_POWER = $(addprefix $(AMD_PP_PATH)/,$(POWER_MGR))
 
diff --git a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
index 3f73f7cd18b9..bea1587d352d 100644
--- a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
+++ b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
@@ -53,7 +53,7 @@ static int amd_powerplay_create(struct amdgpu_device *adev)
 	mutex_init(&hwmgr->smu_lock);
 	hwmgr->chip_family = adev->family;
 	hwmgr->chip_id = adev->asic_type;
-	hwmgr->feature_mask = adev->powerplay.pp_feature;
+	hwmgr->feature_mask = adev->pm.pp_feature;
 	hwmgr->display_config = &adev->pm.pm_display_cfg;
 	adev->powerplay.pp_handle = hwmgr;
 	adev->powerplay.pp_funcs = &pp_dpm_funcs;
@@ -1304,7 +1304,7 @@ static int pp_notify_smu_enable_pwe(void *handle)
 
 	if (hwmgr->hwmgr_func->smus_notify_pwe == NULL) {
 		pr_info_ratelimited("%s was not implemented.\n", __func__);
-		return -EINVAL;;
+		return -EINVAL;
 	}
 
 	mutex_lock(&hwmgr->smu_lock);
@@ -1341,7 +1341,7 @@ static int pp_set_min_deep_sleep_dcefclk(void *handle, uint32_t clock)
 
 	if (hwmgr->hwmgr_func->set_min_deep_sleep_dcefclk == NULL) {
 		pr_debug("%s was not implemented.\n", __func__);
-		return -EINVAL;;
+		return -EINVAL;
 	}
 
 	mutex_lock(&hwmgr->smu_lock);
@@ -1360,7 +1360,7 @@ static int pp_set_hard_min_dcefclk_by_freq(void *handle, uint32_t clock)
 
 	if (hwmgr->hwmgr_func->set_hard_min_dcefclk_by_freq == NULL) {
 		pr_debug("%s was not implemented.\n", __func__);
-		return -EINVAL;;
+		return -EINVAL;
 	}
 
 	mutex_lock(&hwmgr->smu_lock);
@@ -1379,7 +1379,7 @@ static int pp_set_hard_min_fclk_by_freq(void *handle, uint32_t clock)
 
 	if (hwmgr->hwmgr_func->set_hard_min_fclk_by_freq == NULL) {
 		pr_debug("%s was not implemented.\n", __func__);
-		return -EINVAL;;
+		return -EINVAL;
 	}
 
 	mutex_lock(&hwmgr->smu_lock);
diff --git a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
new file mode 100644
index 000000000000..c058c784180e
--- /dev/null
+++ b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
@@ -0,0 +1,1253 @@
+/*
+ * Copyright 2019 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#include "pp_debug.h"
+#include <linux/firmware.h>
+#include <drm/drmP.h>
+#include "amdgpu.h"
+#include "amdgpu_smu.h"
+#include "soc15_common.h"
+#include "smu_v11_0.h"
+#include "atom.h"
+#include "amd_pcie.h"
+
+int smu_dpm_set_power_gate(struct smu_context *smu, uint32_t block_type,
+			   bool gate)
+{
+	int ret = 0;
+
+	switch (block_type) {
+	case AMD_IP_BLOCK_TYPE_UVD:
+		ret = smu_dpm_set_uvd_enable(smu, gate);
+		break;
+	case AMD_IP_BLOCK_TYPE_VCE:
+		ret = smu_dpm_set_vce_enable(smu, gate);
+		break;
+	default:
+		break;
+	}
+
+	return ret;
+}
+
+enum amd_pm_state_type smu_get_current_power_state(struct smu_context *smu)
+{
+	/* not support power state */
+	return POWER_STATE_TYPE_DEFAULT;
+}
+
+int smu_get_power_num_states(struct smu_context *smu,
+			     struct pp_states_info *state_info)
+{
+	if (!state_info)
+		return -EINVAL;
+
+	/* not support power state */
+	memset(state_info, 0, sizeof(struct pp_states_info));
+	state_info->nums = 0;
+
+	return 0;
+}
+
+int smu_common_read_sensor(struct smu_context *smu, enum amd_pp_sensors sensor,
+			   void *data, uint32_t *size)
+{
+	int ret = 0;
+
+	switch (sensor) {
+	case AMDGPU_PP_SENSOR_STABLE_PSTATE_SCLK:
+		*((uint32_t *)data) = smu->pstate_sclk;
+		*size = 4;
+		break;
+	case AMDGPU_PP_SENSOR_STABLE_PSTATE_MCLK:
+		*((uint32_t *)data) = smu->pstate_mclk;
+		*size = 4;
+		break;
+	case AMDGPU_PP_SENSOR_ENABLED_SMC_FEATURES_MASK:
+		ret = smu_feature_get_enabled_mask(smu, (uint32_t *)data, 2);
+		*size = 8;
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+
+	if (ret)
+		*size = 0;
+
+	return ret;
+}
+
+int smu_update_table_with_arg(struct smu_context *smu, uint16_t table_id, uint16_t exarg,
+		     void *table_data, bool drv2smu)
+{
+	struct smu_table_context *smu_table = &smu->smu_table;
+	struct smu_table *table = NULL;
+	int ret = 0;
+	uint32_t table_index;
+
+	if (!table_data || table_id >= smu_table->table_count)
+		return -EINVAL;
+
+	table_index = (exarg << 16) | table_id;
+
+	table = &smu_table->tables[table_id];
+
+	if (drv2smu)
+		memcpy(table->cpu_addr, table_data, table->size);
+
+	ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetDriverDramAddrHigh,
+					  upper_32_bits(table->mc_address));
+	if (ret)
+		return ret;
+	ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetDriverDramAddrLow,
+					  lower_32_bits(table->mc_address));
+	if (ret)
+		return ret;
+	ret = smu_send_smc_msg_with_param(smu, drv2smu ?
+					  SMU_MSG_TransferTableDram2Smu :
+					  SMU_MSG_TransferTableSmu2Dram,
+					  table_index);
+	if (ret)
+		return ret;
+
+	if (!drv2smu)
+		memcpy(table_data, table->cpu_addr, table->size);
+
+	return ret;
+}
+
+bool is_support_sw_smu(struct amdgpu_device *adev)
+{
+	if (amdgpu_dpm != 1)
+		return false;
+
+	if (adev->asic_type >= CHIP_VEGA20 && adev->asic_type != CHIP_RAVEN)
+		return true;
+
+	return false;
+}
+
+int smu_sys_get_pp_table(struct smu_context *smu, void **table)
+{
+	struct smu_table_context *smu_table = &smu->smu_table;
+
+	if (!smu_table->power_play_table && !smu_table->hardcode_pptable)
+		return -EINVAL;
+
+	if (smu_table->hardcode_pptable)
+		*table = smu_table->hardcode_pptable;
+	else
+		*table = smu_table->power_play_table;
+
+	return smu_table->power_play_table_size;
+}
+
+int smu_sys_set_pp_table(struct smu_context *smu,  void *buf, size_t size)
+{
+	struct smu_table_context *smu_table = &smu->smu_table;
+	ATOM_COMMON_TABLE_HEADER *header = (ATOM_COMMON_TABLE_HEADER *)buf;
+	int ret = 0;
+
+	if (header->usStructureSize != size) {
+		pr_err("pp table size not matched !\n");
+		return -EIO;
+	}
+
+	mutex_lock(&smu->mutex);
+	if (!smu_table->hardcode_pptable)
+		smu_table->hardcode_pptable = kzalloc(size, GFP_KERNEL);
+	if (!smu_table->hardcode_pptable) {
+		ret = -ENOMEM;
+		goto failed;
+	}
+
+	memcpy(smu_table->hardcode_pptable, buf, size);
+	smu_table->power_play_table = smu_table->hardcode_pptable;
+	smu_table->power_play_table_size = size;
+	mutex_unlock(&smu->mutex);
+
+	ret = smu_reset(smu);
+	if (ret)
+		pr_info("smu reset failed, ret = %d\n", ret);
+
+	return ret;
+
+failed:
+	mutex_unlock(&smu->mutex);
+	return ret;
+}
+
+int smu_feature_init_dpm(struct smu_context *smu)
+{
+	struct smu_feature *feature = &smu->smu_feature;
+	int ret = 0;
+	uint32_t unallowed_feature_mask[SMU_FEATURE_MAX/32];
+
+	mutex_lock(&feature->mutex);
+	bitmap_fill(feature->allowed, SMU_FEATURE_MAX);
+	mutex_unlock(&feature->mutex);
+
+	ret = smu_get_unallowed_feature_mask(smu, unallowed_feature_mask,
+					     SMU_FEATURE_MAX/32);
+	if (ret)
+		return ret;
+
+	mutex_lock(&feature->mutex);
+	bitmap_andnot(feature->allowed, feature->allowed,
+		      (unsigned long *)unallowed_feature_mask,
+		      feature->feature_num);
+	mutex_unlock(&feature->mutex);
+
+	return ret;
+}
+
+int smu_feature_is_enabled(struct smu_context *smu, int feature_id)
+{
+	struct smu_feature *feature = &smu->smu_feature;
+	int ret = 0;
+
+	WARN_ON(feature_id > feature->feature_num);
+
+	mutex_lock(&feature->mutex);
+	ret = test_bit(feature_id, feature->enabled);
+	mutex_unlock(&feature->mutex);
+
+	return ret;
+}
+
+int smu_feature_set_enabled(struct smu_context *smu, int feature_id, bool enable)
+{
+	struct smu_feature *feature = &smu->smu_feature;
+	int ret = 0;
+
+	WARN_ON(feature_id > feature->feature_num);
+
+	mutex_lock(&feature->mutex);
+	ret = smu_feature_update_enable_state(smu, feature_id, enable);
+	if (ret)
+		goto failed;
+
+	if (enable)
+		test_and_set_bit(feature_id, feature->enabled);
+	else
+		test_and_clear_bit(feature_id, feature->enabled);
+
+failed:
+	mutex_unlock(&feature->mutex);
+
+	return ret;
+}
+
+int smu_feature_is_supported(struct smu_context *smu, int feature_id)
+{
+	struct smu_feature *feature = &smu->smu_feature;
+	int ret = 0;
+
+	WARN_ON(feature_id > feature->feature_num);
+
+	mutex_lock(&feature->mutex);
+	ret = test_bit(feature_id, feature->supported);
+	mutex_unlock(&feature->mutex);
+
+	return ret;
+}
+
+int smu_feature_set_supported(struct smu_context *smu, int feature_id,
+			      bool enable)
+{
+	struct smu_feature *feature = &smu->smu_feature;
+	int ret = 0;
+
+	WARN_ON(feature_id > feature->feature_num);
+
+	mutex_unlock(&feature->mutex);
+	if (enable)
+		test_and_set_bit(feature_id, feature->supported);
+	else
+		test_and_clear_bit(feature_id, feature->supported);
+	mutex_unlock(&feature->mutex);
+
+	return ret;
+}
+
+static int smu_set_funcs(struct amdgpu_device *adev)
+{
+	struct smu_context *smu = &adev->smu;
+
+	switch (adev->asic_type) {
+	case CHIP_VEGA20:
+		adev->pm.pp_feature &= ~PP_GFXOFF_MASK;
+		if (adev->pm.pp_feature & PP_OVERDRIVE_MASK)
+			smu->od_enabled = true;
+		smu_v11_0_set_smu_funcs(smu);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int smu_early_init(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	struct smu_context *smu = &adev->smu;
+
+	smu->adev = adev;
+	mutex_init(&smu->mutex);
+
+	return smu_set_funcs(adev);
+}
+
+static int smu_late_init(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	struct smu_context *smu = &adev->smu;
+	mutex_lock(&smu->mutex);
+	smu_handle_task(&adev->smu,
+			smu->smu_dpm.dpm_level,
+			AMD_PP_TASK_COMPLETE_INIT);
+	mutex_unlock(&smu->mutex);
+
+	return 0;
+}
+
+int smu_get_atom_data_table(struct smu_context *smu, uint32_t table,
+			    uint16_t *size, uint8_t *frev, uint8_t *crev,
+			    uint8_t **addr)
+{
+	struct amdgpu_device *adev = smu->adev;
+	uint16_t data_start;
+
+	if (!amdgpu_atom_parse_data_header(adev->mode_info.atom_context, table,
+					   size, frev, crev, &data_start))
+		return -EINVAL;
+
+	*addr = (uint8_t *)adev->mode_info.atom_context->bios + data_start;
+
+	return 0;
+}
+
+static int smu_initialize_pptable(struct smu_context *smu)
+{
+	/* TODO */
+	return 0;
+}
+
+static int smu_smc_table_sw_init(struct smu_context *smu)
+{
+	int ret;
+
+	ret = smu_initialize_pptable(smu);
+	if (ret) {
+		pr_err("Failed to init smu_initialize_pptable!\n");
+		return ret;
+	}
+
+	/**
+	 * Create smu_table structure, and init smc tables such as
+	 * TABLE_PPTABLE, TABLE_WATERMARKS, TABLE_SMU_METRICS, and etc.
+	 */
+	ret = smu_init_smc_tables(smu);
+	if (ret) {
+		pr_err("Failed to init smc tables!\n");
+		return ret;
+	}
+
+	/**
+	 * Create smu_power_context structure, and allocate smu_dpm_context and
+	 * context size to fill the smu_power_context data.
+	 */
+	ret = smu_init_power(smu);
+	if (ret) {
+		pr_err("Failed to init smu_init_power!\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static int smu_smc_table_sw_fini(struct smu_context *smu)
+{
+	int ret;
+
+	ret = smu_fini_smc_tables(smu);
+	if (ret) {
+		pr_err("Failed to smu_fini_smc_tables!\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static int smu_sw_init(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	struct smu_context *smu = &adev->smu;
+	int ret;
+
+	if (!is_support_sw_smu(adev))
+		return -EINVAL;
+
+	smu->pool_size = adev->pm.smu_prv_buffer_size;
+	smu->smu_feature.feature_num = SMU_FEATURE_MAX;
+	mutex_init(&smu->smu_feature.mutex);
+	bitmap_zero(smu->smu_feature.supported, SMU_FEATURE_MAX);
+	bitmap_zero(smu->smu_feature.enabled, SMU_FEATURE_MAX);
+	bitmap_zero(smu->smu_feature.allowed, SMU_FEATURE_MAX);
+	smu->watermarks_bitmap = 0;
+	smu->power_profile_mode = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;
+	smu->default_power_profile_mode = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;
+
+	smu->workload_mask = 1 << smu->workload_prority[PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT];
+	smu->workload_prority[PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT] = 0;
+	smu->workload_prority[PP_SMC_POWER_PROFILE_FULLSCREEN3D] = 1;
+	smu->workload_prority[PP_SMC_POWER_PROFILE_POWERSAVING] = 2;
+	smu->workload_prority[PP_SMC_POWER_PROFILE_VIDEO] = 3;
+	smu->workload_prority[PP_SMC_POWER_PROFILE_VR] = 4;
+	smu->workload_prority[PP_SMC_POWER_PROFILE_COMPUTE] = 5;
+	smu->workload_prority[PP_SMC_POWER_PROFILE_CUSTOM] = 6;
+
+	smu->workload_setting[0] = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;
+	smu->workload_setting[1] = PP_SMC_POWER_PROFILE_FULLSCREEN3D;
+	smu->workload_setting[2] = PP_SMC_POWER_PROFILE_POWERSAVING;
+	smu->workload_setting[3] = PP_SMC_POWER_PROFILE_VIDEO;
+	smu->workload_setting[4] = PP_SMC_POWER_PROFILE_VR;
+	smu->workload_setting[5] = PP_SMC_POWER_PROFILE_COMPUTE;
+	smu->workload_setting[6] = PP_SMC_POWER_PROFILE_CUSTOM;
+	smu->display_config = &adev->pm.pm_display_cfg;
+
+	smu->smu_dpm.dpm_level = AMD_DPM_FORCED_LEVEL_AUTO;
+	smu->smu_dpm.requested_dpm_level = AMD_DPM_FORCED_LEVEL_AUTO;
+	ret = smu_init_microcode(smu);
+	if (ret) {
+		pr_err("Failed to load smu firmware!\n");
+		return ret;
+	}
+
+	ret = smu_smc_table_sw_init(smu);
+	if (ret) {
+		pr_err("Failed to sw init smc table!\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static int smu_sw_fini(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	struct smu_context *smu = &adev->smu;
+	int ret;
+
+	if (!is_support_sw_smu(adev))
+		return -EINVAL;
+
+	ret = smu_smc_table_sw_fini(smu);
+	if (ret) {
+		pr_err("Failed to sw fini smc table!\n");
+		return ret;
+	}
+
+	ret = smu_fini_power(smu);
+	if (ret) {
+		pr_err("Failed to init smu_fini_power!\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static int smu_init_fb_allocations(struct smu_context *smu)
+{
+	struct amdgpu_device *adev = smu->adev;
+	struct smu_table_context *smu_table = &smu->smu_table;
+	struct smu_table *tables = smu_table->tables;
+	uint32_t table_count = smu_table->table_count;
+	uint32_t i = 0;
+	int32_t ret = 0;
+
+	if (table_count <= 0)
+		return -EINVAL;
+
+	for (i = 0 ; i < table_count; i++) {
+		if (tables[i].size == 0)
+			continue;
+		ret = amdgpu_bo_create_kernel(adev,
+					      tables[i].size,
+					      tables[i].align,
+					      tables[i].domain,
+					      &tables[i].bo,
+					      &tables[i].mc_address,
+					      &tables[i].cpu_addr);
+		if (ret)
+			goto failed;
+	}
+
+	return 0;
+failed:
+	for (; i > 0; i--) {
+		if (tables[i].size == 0)
+			continue;
+		amdgpu_bo_free_kernel(&tables[i].bo,
+				      &tables[i].mc_address,
+				      &tables[i].cpu_addr);
+
+	}
+	return ret;
+}
+
+static int smu_fini_fb_allocations(struct smu_context *smu)
+{
+	struct smu_table_context *smu_table = &smu->smu_table;
+	struct smu_table *tables = smu_table->tables;
+	uint32_t table_count = smu_table->table_count;
+	uint32_t i = 0;
+
+	if (table_count == 0 || tables == NULL)
+		return 0;
+
+	for (i = 0 ; i < table_count; i++) {
+		if (tables[i].size == 0)
+			continue;
+		amdgpu_bo_free_kernel(&tables[i].bo,
+				      &tables[i].mc_address,
+				      &tables[i].cpu_addr);
+	}
+
+	return 0;
+}
+
+static int smu_override_pcie_parameters(struct smu_context *smu)
+{
+	struct amdgpu_device *adev = smu->adev;
+	uint32_t pcie_gen = 0, pcie_width = 0, smu_pcie_arg;
+	int ret;
+
+	if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN4)
+		pcie_gen = 3;
+	else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3)
+		pcie_gen = 2;
+	else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2)
+		pcie_gen = 1;
+	else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN1)
+		pcie_gen = 0;
+
+	/* Bit 31:16: LCLK DPM level. 0 is DPM0, and 1 is DPM1
+	 * Bit 15:8:  PCIE GEN, 0 to 3 corresponds to GEN1 to GEN4
+	 * Bit 7:0:   PCIE lane width, 1 to 7 corresponds is x1 to x32
+	 */
+	if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X16)
+		pcie_width = 6;
+	else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X12)
+		pcie_width = 5;
+	else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X8)
+		pcie_width = 4;
+	else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X4)
+		pcie_width = 3;
+	else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X2)
+		pcie_width = 2;
+	else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X1)
+		pcie_width = 1;
+
+	smu_pcie_arg = (1 << 16) | (pcie_gen << 8) | pcie_width;
+	ret = smu_send_smc_msg_with_param(smu,
+					  SMU_MSG_OverridePcieParameters,
+					  smu_pcie_arg);
+	if (ret)
+		pr_err("[%s] Attempt to override pcie params failed!\n", __func__);
+	return ret;
+}
+
+static int smu_smc_table_hw_init(struct smu_context *smu,
+				 bool initialize)
+{
+	struct amdgpu_device *adev = smu->adev;
+	int ret;
+
+	if (smu_is_dpm_running(smu) && adev->in_suspend) {
+		pr_info("dpm has been enabled\n");
+		return 0;
+	}
+
+	ret = smu_init_display(smu);
+	if (ret)
+		return ret;
+
+	if (initialize) {
+		ret = smu_read_pptable_from_vbios(smu);
+		if (ret)
+			return ret;
+
+		/* get boot_values from vbios to set revision, gfxclk, and etc. */
+		ret = smu_get_vbios_bootup_values(smu);
+		if (ret)
+			return ret;
+
+		ret = smu_get_clk_info_from_vbios(smu);
+		if (ret)
+			return ret;
+
+		/*
+		 * check if the format_revision in vbios is up to pptable header
+		 * version, and the structure size is not 0.
+		 */
+		ret = smu_get_clk_info_from_vbios(smu);
+		if (ret)
+			return ret;
+
+		ret = smu_check_pptable(smu);
+		if (ret)
+			return ret;
+
+		/*
+		 * allocate vram bos to store smc table contents.
+		 */
+		ret = smu_init_fb_allocations(smu);
+		if (ret)
+			return ret;
+
+		/*
+		 * Parse pptable format and fill PPTable_t smc_pptable to
+		 * smu_table_context structure. And read the smc_dpm_table from vbios,
+		 * then fill it into smc_pptable.
+		 */
+		ret = smu_parse_pptable(smu);
+		if (ret)
+			return ret;
+
+		/*
+		 * Send msg GetDriverIfVersion to check if the return value is equal
+		 * with DRIVER_IF_VERSION of smc header.
+		 */
+		ret = smu_check_fw_version(smu);
+		if (ret)
+			return ret;
+	}
+
+	/*
+	 * Copy pptable bo in the vram to smc with SMU MSGs such as
+	 * SetDriverDramAddr and TransferTableDram2Smu.
+	 */
+	ret = smu_write_pptable(smu);
+	if (ret)
+		return ret;
+
+	/* issue RunAfllBtc msg */
+	ret = smu_run_afll_btc(smu);
+	if (ret)
+		return ret;
+
+	ret = smu_feature_set_allowed_mask(smu);
+	if (ret)
+		return ret;
+
+	ret = smu_system_features_control(smu, true);
+	if (ret)
+		return ret;
+
+	ret = smu_override_pcie_parameters(smu);
+	if (ret)
+		return ret;
+
+	ret = smu_notify_display_change(smu);
+	if (ret)
+		return ret;
+
+	/*
+	 * Set min deep sleep dce fclk with bootup value from vbios via
+	 * SetMinDeepSleepDcefclk MSG.
+	 */
+	ret = smu_set_min_dcef_deep_sleep(smu);
+	if (ret)
+		return ret;
+
+	/*
+	 * Set initialized values (get from vbios) to dpm tables context such as
+	 * gfxclk, memclk, dcefclk, and etc. And enable the DPM feature for each
+	 * type of clks.
+	 */
+	if (initialize) {
+		ret = smu_populate_smc_pptable(smu);
+		if (ret)
+			return ret;
+
+		ret = smu_init_max_sustainable_clocks(smu);
+		if (ret)
+			return ret;
+	}
+
+	ret = smu_set_od8_default_settings(smu, initialize);
+	if (ret)
+		return ret;
+
+	if (initialize) {
+		ret = smu_populate_umd_state_clk(smu);
+		if (ret)
+			return ret;
+
+		ret = smu_get_power_limit(smu, &smu->default_power_limit, false);
+		if (ret)
+			return ret;
+	}
+
+	/*
+	 * Set PMSTATUSLOG table bo address with SetToolsDramAddr MSG for tools.
+	 */
+	ret = smu_set_tool_table_location(smu);
+
+	return ret;
+}
+
+/**
+ * smu_alloc_memory_pool - allocate memory pool in the system memory
+ *
+ * @smu: amdgpu_device pointer
+ *
+ * This memory pool will be used for SMC use and msg SetSystemVirtualDramAddr
+ * and DramLogSetDramAddr can notify it changed.
+ *
+ * Returns 0 on success, error on failure.
+ */
+static int smu_alloc_memory_pool(struct smu_context *smu)
+{
+	struct amdgpu_device *adev = smu->adev;
+	struct smu_table_context *smu_table = &smu->smu_table;
+	struct smu_table *memory_pool = &smu_table->memory_pool;
+	uint64_t pool_size = smu->pool_size;
+	int ret = 0;
+
+	if (pool_size == SMU_MEMORY_POOL_SIZE_ZERO)
+		return ret;
+
+	memory_pool->size = pool_size;
+	memory_pool->align = PAGE_SIZE;
+	memory_pool->domain = AMDGPU_GEM_DOMAIN_GTT;
+
+	switch (pool_size) {
+	case SMU_MEMORY_POOL_SIZE_256_MB:
+	case SMU_MEMORY_POOL_SIZE_512_MB:
+	case SMU_MEMORY_POOL_SIZE_1_GB:
+	case SMU_MEMORY_POOL_SIZE_2_GB:
+		ret = amdgpu_bo_create_kernel(adev,
+					      memory_pool->size,
+					      memory_pool->align,
+					      memory_pool->domain,
+					      &memory_pool->bo,
+					      &memory_pool->mc_address,
+					      &memory_pool->cpu_addr);
+		break;
+	default:
+		break;
+	}
+
+	return ret;
+}
+
+static int smu_free_memory_pool(struct smu_context *smu)
+{
+	struct smu_table_context *smu_table = &smu->smu_table;
+	struct smu_table *memory_pool = &smu_table->memory_pool;
+	int ret = 0;
+
+	if (memory_pool->size == SMU_MEMORY_POOL_SIZE_ZERO)
+		return ret;
+
+	amdgpu_bo_free_kernel(&memory_pool->bo,
+			      &memory_pool->mc_address,
+			      &memory_pool->cpu_addr);
+
+	memset(memory_pool, 0, sizeof(struct smu_table));
+
+	return ret;
+}
+
+static int smu_hw_init(void *handle)
+{
+	int ret;
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	struct smu_context *smu = &adev->smu;
+
+	if (!is_support_sw_smu(adev))
+		return -EINVAL;
+
+	if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
+		ret = smu_load_microcode(smu);
+		if (ret)
+			return ret;
+	}
+
+	ret = smu_check_fw_status(smu);
+	if (ret) {
+		pr_err("SMC firmware status is not correct\n");
+		return ret;
+	}
+
+	mutex_lock(&smu->mutex);
+
+	ret = smu_feature_init_dpm(smu);
+	if (ret)
+		goto failed;
+
+	ret = smu_smc_table_hw_init(smu, true);
+	if (ret)
+		goto failed;
+
+	ret = smu_alloc_memory_pool(smu);
+	if (ret)
+		goto failed;
+
+	/*
+	 * Use msg SetSystemVirtualDramAddr and DramLogSetDramAddr can notify
+	 * pool location.
+	 */
+	ret = smu_notify_memory_pool_location(smu);
+	if (ret)
+		goto failed;
+
+	ret = smu_start_thermal_control(smu);
+	if (ret)
+		goto failed;
+
+	mutex_unlock(&smu->mutex);
+
+	adev->pm.dpm_enabled = true;
+
+	pr_info("SMU is initialized successfully!\n");
+
+	return 0;
+
+failed:
+	mutex_unlock(&smu->mutex);
+	return ret;
+}
+
+static int smu_hw_fini(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	struct smu_context *smu = &adev->smu;
+	struct smu_table_context *table_context = &smu->smu_table;
+	int ret = 0;
+
+	if (!is_support_sw_smu(adev))
+		return -EINVAL;
+
+	kfree(table_context->driver_pptable);
+	table_context->driver_pptable = NULL;
+
+	kfree(table_context->max_sustainable_clocks);
+	table_context->max_sustainable_clocks = NULL;
+
+	kfree(table_context->od_feature_capabilities);
+	table_context->od_feature_capabilities = NULL;
+
+	kfree(table_context->od_settings_max);
+	table_context->od_settings_max = NULL;
+
+	kfree(table_context->od_settings_min);
+	table_context->od_settings_min = NULL;
+
+	kfree(table_context->overdrive_table);
+	table_context->overdrive_table = NULL;
+
+	kfree(table_context->od8_settings);
+	table_context->od8_settings = NULL;
+
+	ret = smu_fini_fb_allocations(smu);
+	if (ret)
+		return ret;
+
+	ret = smu_free_memory_pool(smu);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+int smu_reset(struct smu_context *smu)
+{
+	struct amdgpu_device *adev = smu->adev;
+	int ret = 0;
+
+	ret = smu_hw_fini(adev);
+	if (ret)
+		return ret;
+
+	ret = smu_hw_init(adev);
+	if (ret)
+		return ret;
+
+	return ret;
+}
+
+static int smu_suspend(void *handle)
+{
+	int ret;
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	struct smu_context *smu = &adev->smu;
+
+	if (!is_support_sw_smu(adev))
+		return -EINVAL;
+
+	ret = smu_system_features_control(smu, false);
+	if (ret)
+		return ret;
+
+	smu->watermarks_bitmap &= ~(WATERMARKS_LOADED);
+
+	return 0;
+}
+
+static int smu_resume(void *handle)
+{
+	int ret;
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	struct smu_context *smu = &adev->smu;
+
+	if (!is_support_sw_smu(adev))
+		return -EINVAL;
+
+	pr_info("SMU is resuming...\n");
+
+	mutex_lock(&smu->mutex);
+
+	ret = smu_smc_table_hw_init(smu, false);
+	if (ret)
+		goto failed;
+
+	ret = smu_start_thermal_control(smu);
+	if (ret)
+		goto failed;
+
+	mutex_unlock(&smu->mutex);
+
+	pr_info("SMU is resumed successfully!\n");
+
+	return 0;
+failed:
+	mutex_unlock(&smu->mutex);
+	return ret;
+}
+
+int smu_display_configuration_change(struct smu_context *smu,
+				     const struct amd_pp_display_configuration *display_config)
+{
+	int index = 0;
+	int num_of_active_display = 0;
+
+	if (!is_support_sw_smu(smu->adev))
+		return -EINVAL;
+
+	if (!display_config)
+		return -EINVAL;
+
+	mutex_lock(&smu->mutex);
+
+	smu_set_deep_sleep_dcefclk(smu,
+				   display_config->min_dcef_deep_sleep_set_clk / 100);
+
+	for (index = 0; index < display_config->num_path_including_non_display; index++) {
+		if (display_config->displays[index].controller_id != 0)
+			num_of_active_display++;
+	}
+
+	smu_set_active_display_count(smu, num_of_active_display);
+
+	smu_store_cc6_data(smu, display_config->cpu_pstate_separation_time,
+			   display_config->cpu_cc6_disable,
+			   display_config->cpu_pstate_disable,
+			   display_config->nb_pstate_switch_disable);
+
+	mutex_unlock(&smu->mutex);
+
+	return 0;
+}
+
+static int smu_get_clock_info(struct smu_context *smu,
+			      struct smu_clock_info *clk_info,
+			      enum smu_perf_level_designation designation)
+{
+	int ret;
+	struct smu_performance_level level = {0};
+
+	if (!clk_info)
+		return -EINVAL;
+
+	ret = smu_get_perf_level(smu, PERF_LEVEL_ACTIVITY, &level);
+	if (ret)
+		return -EINVAL;
+
+	clk_info->min_mem_clk = level.memory_clock;
+	clk_info->min_eng_clk = level.core_clock;
+	clk_info->min_bus_bandwidth = level.non_local_mem_freq * level.non_local_mem_width;
+
+	ret = smu_get_perf_level(smu, designation, &level);
+	if (ret)
+		return -EINVAL;
+
+	clk_info->min_mem_clk = level.memory_clock;
+	clk_info->min_eng_clk = level.core_clock;
+	clk_info->min_bus_bandwidth = level.non_local_mem_freq * level.non_local_mem_width;
+
+	return 0;
+}
+
+int smu_get_current_clocks(struct smu_context *smu,
+			   struct amd_pp_clock_info *clocks)
+{
+	struct amd_pp_simple_clock_info simple_clocks = {0};
+	struct smu_clock_info hw_clocks;
+	int ret = 0;
+
+	if (!is_support_sw_smu(smu->adev))
+		return -EINVAL;
+
+	mutex_lock(&smu->mutex);
+
+	smu_get_dal_power_level(smu, &simple_clocks);
+
+	if (smu->support_power_containment)
+		ret = smu_get_clock_info(smu, &hw_clocks,
+					 PERF_LEVEL_POWER_CONTAINMENT);
+	else
+		ret = smu_get_clock_info(smu, &hw_clocks, PERF_LEVEL_ACTIVITY);
+
+	if (ret) {
+		pr_err("Error in smu_get_clock_info\n");
+		goto failed;
+	}
+
+	clocks->min_engine_clock = hw_clocks.min_eng_clk;
+	clocks->max_engine_clock = hw_clocks.max_eng_clk;
+	clocks->min_memory_clock = hw_clocks.min_mem_clk;
+	clocks->max_memory_clock = hw_clocks.max_mem_clk;
+	clocks->min_bus_bandwidth = hw_clocks.min_bus_bandwidth;
+	clocks->max_bus_bandwidth = hw_clocks.max_bus_bandwidth;
+	clocks->max_engine_clock_in_sr = hw_clocks.max_eng_clk;
+	clocks->min_engine_clock_in_sr = hw_clocks.min_eng_clk;
+
+        if (simple_clocks.level == 0)
+                clocks->max_clocks_state = PP_DAL_POWERLEVEL_7;
+        else
+                clocks->max_clocks_state = simple_clocks.level;
+
+        if (!smu_get_current_shallow_sleep_clocks(smu, &hw_clocks)) {
+                clocks->max_engine_clock_in_sr = hw_clocks.max_eng_clk;
+                clocks->min_engine_clock_in_sr = hw_clocks.min_eng_clk;
+        }
+
+failed:
+	mutex_unlock(&smu->mutex);
+	return ret;
+}
+
+static int smu_set_clockgating_state(void *handle,
+				     enum amd_clockgating_state state)
+{
+	return 0;
+}
+
+static int smu_set_powergating_state(void *handle,
+				     enum amd_powergating_state state)
+{
+	return 0;
+}
+
+static int smu_enable_umd_pstate(void *handle,
+		      enum amd_dpm_forced_level *level)
+{
+	uint32_t profile_mode_mask = AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD |
+					AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK |
+					AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK |
+					AMD_DPM_FORCED_LEVEL_PROFILE_PEAK;
+
+	struct smu_context *smu = (struct smu_context*)(handle);
+	struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm);
+	if (!smu_dpm_ctx->dpm_context)
+		return -EINVAL;
+
+	if (!(smu_dpm_ctx->dpm_level & profile_mode_mask)) {
+		/* enter umd pstate, save current level, disable gfx cg*/
+		if (*level & profile_mode_mask) {
+			smu_dpm_ctx->saved_dpm_level = smu_dpm_ctx->dpm_level;
+			smu_dpm_ctx->enable_umd_pstate = true;
+			amdgpu_device_ip_set_clockgating_state(smu->adev,
+							       AMD_IP_BLOCK_TYPE_GFX,
+							       AMD_CG_STATE_UNGATE);
+			amdgpu_device_ip_set_powergating_state(smu->adev,
+							       AMD_IP_BLOCK_TYPE_GFX,
+							       AMD_PG_STATE_UNGATE);
+		}
+	} else {
+		/* exit umd pstate, restore level, enable gfx cg*/
+		if (!(*level & profile_mode_mask)) {
+			if (*level == AMD_DPM_FORCED_LEVEL_PROFILE_EXIT)
+				*level = smu_dpm_ctx->saved_dpm_level;
+			smu_dpm_ctx->enable_umd_pstate = false;
+			amdgpu_device_ip_set_clockgating_state(smu->adev,
+							       AMD_IP_BLOCK_TYPE_GFX,
+							       AMD_CG_STATE_GATE);
+			amdgpu_device_ip_set_powergating_state(smu->adev,
+							       AMD_IP_BLOCK_TYPE_GFX,
+							       AMD_PG_STATE_GATE);
+		}
+	}
+
+	return 0;
+}
+
+int smu_adjust_power_state_dynamic(struct smu_context *smu,
+				   enum amd_dpm_forced_level level,
+				   bool skip_display_settings)
+{
+	int ret = 0;
+	int index = 0;
+	uint32_t sclk_mask, mclk_mask, soc_mask;
+	long workload;
+	struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm);
+
+	if (!skip_display_settings) {
+		ret = smu_display_config_changed(smu);
+		if (ret) {
+			pr_err("Failed to change display config!");
+			return ret;
+		}
+	}
+
+	ret = smu_apply_clocks_adjust_rules(smu);
+	if (ret) {
+		pr_err("Failed to apply clocks adjust rules!");
+		return ret;
+	}
+
+	if (!skip_display_settings) {
+		ret = smu_notify_smc_dispaly_config(smu);
+		if (ret) {
+			pr_err("Failed to notify smc display config!");
+			return ret;
+		}
+	}
+
+	if (smu_dpm_ctx->dpm_level != level) {
+		switch (level) {
+		case AMD_DPM_FORCED_LEVEL_HIGH:
+			ret = smu_force_dpm_limit_value(smu, true);
+			break;
+		case AMD_DPM_FORCED_LEVEL_LOW:
+			ret = smu_force_dpm_limit_value(smu, false);
+			break;
+
+		case AMD_DPM_FORCED_LEVEL_AUTO:
+			ret = smu_unforce_dpm_levels(smu);
+			break;
+
+		case AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD:
+		case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK:
+		case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK:
+		case AMD_DPM_FORCED_LEVEL_PROFILE_PEAK:
+			ret = smu_get_profiling_clk_mask(smu, level,
+							 &sclk_mask,
+							 &mclk_mask,
+							 &soc_mask);
+			if (ret)
+				return ret;
+			smu_force_clk_levels(smu, PP_SCLK, 1 << sclk_mask);
+			smu_force_clk_levels(smu, PP_MCLK, 1 << mclk_mask);
+			break;
+
+		case AMD_DPM_FORCED_LEVEL_MANUAL:
+		case AMD_DPM_FORCED_LEVEL_PROFILE_EXIT:
+		default:
+			break;
+		}
+
+		if (!ret)
+			smu_dpm_ctx->dpm_level = level;
+	}
+
+	if (smu_dpm_ctx->dpm_level != AMD_DPM_FORCED_LEVEL_MANUAL) {
+		index = fls(smu->workload_mask);
+		index = index > 0 && index <= WORKLOAD_POLICY_MAX ? index - 1 : 0;
+		workload = smu->workload_setting[index];
+
+		if (smu->power_profile_mode != workload)
+			smu_set_power_profile_mode(smu, &workload, 0);
+	}
+
+	return ret;
+}
+
+int smu_handle_task(struct smu_context *smu,
+		    enum amd_dpm_forced_level level,
+		    enum amd_pp_task task_id)
+{
+	int ret = 0;
+
+	switch (task_id) {
+	case AMD_PP_TASK_DISPLAY_CONFIG_CHANGE:
+		ret = smu_pre_display_config_changed(smu);
+		if (ret)
+			return ret;
+		ret = smu_set_cpu_power_state(smu);
+		if (ret)
+			return ret;
+		ret = smu_adjust_power_state_dynamic(smu, level, false);
+		break;
+	case AMD_PP_TASK_COMPLETE_INIT:
+	case AMD_PP_TASK_READJUST_POWER_STATE:
+		ret = smu_adjust_power_state_dynamic(smu, level, true);
+		break;
+	default:
+		break;
+	}
+
+	return ret;
+}
+
+const struct amd_ip_funcs smu_ip_funcs = {
+	.name = "smu",
+	.early_init = smu_early_init,
+	.late_init = smu_late_init,
+	.sw_init = smu_sw_init,
+	.sw_fini = smu_sw_fini,
+	.hw_init = smu_hw_init,
+	.hw_fini = smu_hw_fini,
+	.suspend = smu_suspend,
+	.resume = smu_resume,
+	.is_idle = NULL,
+	.check_soft_reset = NULL,
+	.wait_for_idle = NULL,
+	.soft_reset = NULL,
+	.set_clockgating_state = smu_set_clockgating_state,
+	.set_powergating_state = smu_set_powergating_state,
+	.enable_umd_pstate = smu_enable_umd_pstate,
+};
+
+const struct amdgpu_ip_block_version smu_v11_0_ip_block =
+{
+	.type = AMD_IP_BLOCK_TYPE_SMC,
+	.major = 11,
+	.minor = 0,
+	.rev = 0,
+	.funcs = &smu_ip_funcs,
+};
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/Makefile b/drivers/gpu/drm/amd/powerplay/hwmgr/Makefile
index 0b3c6d1d52e4..cc63705920dc 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/Makefile
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/Makefile
@@ -35,7 +35,8 @@ HARDWARE_MGR = hwmgr.o processpptables.o \
 		vega12_thermal.o \
 		pp_overdriver.o smu_helper.o \
 		vega20_processpptables.o vega20_hwmgr.o vega20_powertune.o \
-		vega20_thermal.o common_baco.o vega10_baco.o  vega20_baco.o
+		vega20_thermal.o common_baco.o vega10_baco.o  vega20_baco.o \
+		vega12_baco.o smu9_baco.o
 
 AMD_PP_HWMGR = $(addprefix $(AMD_PP_PATH)/hwmgr/,$(HARDWARE_MGR))
 
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c b/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c
index c1c51c115e57..70f7f47a2fcf 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c
@@ -76,7 +76,7 @@ int phm_set_power_state(struct pp_hwmgr *hwmgr,
 int phm_enable_dynamic_state_management(struct pp_hwmgr *hwmgr)
 {
 	struct amdgpu_device *adev = NULL;
-	int ret = -EINVAL;;
+	int ret = -EINVAL;
 	PHM_FUNC_CHECK(hwmgr);
 	adev = hwmgr->adev;
 
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c
index 0ad8fe4a6277..9a595f7525e6 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c
@@ -35,6 +35,7 @@
 #include "smu10_hwmgr.h"
 #include "power_state.h"
 #include "soc15_common.h"
+#include "smu10.h"
 
 #define SMU10_MAX_DEEPSLEEP_DIVIDER_ID     5
 #define SMU10_MINIMUM_ENGINE_CLOCK         800   /* 8Mhz, the low boundary of engine clock allowed on this chip */
@@ -114,11 +115,6 @@ static int smu10_initialize_dpm_defaults(struct pp_hwmgr *hwmgr)
 	smu10_data->num_active_display = 0;
 	smu10_data->deep_sleep_dcefclk = 0;
 
-	if (hwmgr->feature_mask & PP_GFXOFF_MASK)
-		smu10_data->gfx_off_controled_by_driver = true;
-	else
-		smu10_data->gfx_off_controled_by_driver = false;
-
 	phm_cap_unset(hwmgr->platform_descriptor.platformCaps,
 					PHM_PlatformCaps_SclkDeepSleep);
 
@@ -209,18 +205,13 @@ static int smu10_set_clock_limit(struct pp_hwmgr *hwmgr, const void *input)
 	return 0;
 }
 
-static inline uint32_t convert_10k_to_mhz(uint32_t clock)
-{
-	return (clock + 99) / 100;
-}
-
 static int smu10_set_min_deep_sleep_dcefclk(struct pp_hwmgr *hwmgr, uint32_t clock)
 {
 	struct smu10_hwmgr *smu10_data = (struct smu10_hwmgr *)(hwmgr->backend);
 
 	if (smu10_data->need_min_deep_sleep_dcefclk &&
-		smu10_data->deep_sleep_dcefclk != convert_10k_to_mhz(clock)) {
-		smu10_data->deep_sleep_dcefclk = convert_10k_to_mhz(clock);
+		smu10_data->deep_sleep_dcefclk != clock) {
+		smu10_data->deep_sleep_dcefclk = clock;
 		smum_send_msg_to_smc_with_parameter(hwmgr,
 					PPSMC_MSG_SetMinDeepSleepDcefclk,
 					smu10_data->deep_sleep_dcefclk);
@@ -233,8 +224,8 @@ static int smu10_set_hard_min_dcefclk_by_freq(struct pp_hwmgr *hwmgr, uint32_t c
 	struct smu10_hwmgr *smu10_data = (struct smu10_hwmgr *)(hwmgr->backend);
 
 	if (smu10_data->dcf_actual_hard_min_freq &&
-		smu10_data->dcf_actual_hard_min_freq != convert_10k_to_mhz(clock)) {
-		smu10_data->dcf_actual_hard_min_freq = convert_10k_to_mhz(clock);
+		smu10_data->dcf_actual_hard_min_freq != clock) {
+		smu10_data->dcf_actual_hard_min_freq = clock;
 		smum_send_msg_to_smc_with_parameter(hwmgr,
 					PPSMC_MSG_SetHardMinDcefclkByFreq,
 					smu10_data->dcf_actual_hard_min_freq);
@@ -247,8 +238,8 @@ static int smu10_set_hard_min_fclk_by_freq(struct pp_hwmgr *hwmgr, uint32_t cloc
 	struct smu10_hwmgr *smu10_data = (struct smu10_hwmgr *)(hwmgr->backend);
 
 	if (smu10_data->f_actual_hard_min_freq &&
-		smu10_data->f_actual_hard_min_freq != convert_10k_to_mhz(clock)) {
-		smu10_data->f_actual_hard_min_freq = convert_10k_to_mhz(clock);
+		smu10_data->f_actual_hard_min_freq != clock) {
+		smu10_data->f_actual_hard_min_freq = clock;
 		smum_send_msg_to_smc_with_parameter(hwmgr,
 					PPSMC_MSG_SetHardMinFclkByFreq,
 					smu10_data->f_actual_hard_min_freq);
@@ -330,9 +321,9 @@ static bool smu10_is_gfx_on(struct pp_hwmgr *hwmgr)
 
 static int smu10_disable_gfx_off(struct pp_hwmgr *hwmgr)
 {
-	struct smu10_hwmgr *smu10_data = (struct smu10_hwmgr *)(hwmgr->backend);
+	struct amdgpu_device *adev = hwmgr->adev;
 
-	if (smu10_data->gfx_off_controled_by_driver) {
+	if (adev->pm.pp_feature & PP_GFXOFF_MASK) {
 		smum_send_msg_to_smc(hwmgr, PPSMC_MSG_DisableGfxOff);
 
 		/* confirm gfx is back to "on" state */
@@ -350,9 +341,9 @@ static int smu10_disable_dpm_tasks(struct pp_hwmgr *hwmgr)
 
 static int smu10_enable_gfx_off(struct pp_hwmgr *hwmgr)
 {
-	struct smu10_hwmgr *smu10_data = (struct smu10_hwmgr *)(hwmgr->backend);
+	struct amdgpu_device *adev = hwmgr->adev;
 
-	if (smu10_data->gfx_off_controled_by_driver)
+	if (adev->pm.pp_feature & PP_GFXOFF_MASK)
 		smum_send_msg_to_smc(hwmgr, PPSMC_MSG_EnableGfxOff);
 
 	return 0;
@@ -577,7 +568,6 @@ static int smu10_dpm_force_dpm_level(struct pp_hwmgr *hwmgr,
 				enum amd_dpm_forced_level level)
 {
 	struct smu10_hwmgr *data = hwmgr->backend;
-	struct amdgpu_device *adev = hwmgr->adev;
 	uint32_t min_sclk = hwmgr->display_config->min_core_set_clock;
 	uint32_t min_mclk = hwmgr->display_config->min_mem_set_clock/100;
 
@@ -586,11 +576,6 @@ static int smu10_dpm_force_dpm_level(struct pp_hwmgr *hwmgr,
 		return 0;
 	}
 
-	/* Disable UMDPSTATE support on rv2 temporarily */
-	if ((adev->asic_type == CHIP_RAVEN) &&
-	    (adev->rev_id >= 8))
-		return 0;
-
 	if (min_sclk < data->gfx_min_freq_limit)
 		min_sclk = data->gfx_min_freq_limit;
 
@@ -1205,6 +1190,94 @@ static void smu10_powergate_vcn(struct pp_hwmgr *hwmgr, bool bgate)
 	}
 }
 
+static int conv_power_profile_to_pplib_workload(int power_profile)
+{
+	int pplib_workload = 0;
+
+	switch (power_profile) {
+	case PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT:
+		pplib_workload = WORKLOAD_DEFAULT_BIT;
+		break;
+	case PP_SMC_POWER_PROFILE_FULLSCREEN3D:
+		pplib_workload = WORKLOAD_PPLIB_FULL_SCREEN_3D_BIT;
+		break;
+	case PP_SMC_POWER_PROFILE_POWERSAVING:
+		pplib_workload = WORKLOAD_PPLIB_POWER_SAVING_BIT;
+		break;
+	case PP_SMC_POWER_PROFILE_VIDEO:
+		pplib_workload = WORKLOAD_PPLIB_VIDEO_BIT;
+		break;
+	case PP_SMC_POWER_PROFILE_VR:
+		pplib_workload = WORKLOAD_PPLIB_VR_BIT;
+		break;
+	case PP_SMC_POWER_PROFILE_COMPUTE:
+		pplib_workload = WORKLOAD_PPLIB_COMPUTE_BIT;
+		break;
+	}
+
+	return pplib_workload;
+}
+
+static int smu10_get_power_profile_mode(struct pp_hwmgr *hwmgr, char *buf)
+{
+	uint32_t i, size = 0;
+	static const uint8_t
+		profile_mode_setting[6][4] = {{70, 60, 0, 0,},
+						{70, 60, 1, 3,},
+						{90, 60, 0, 0,},
+						{70, 60, 0, 0,},
+						{70, 90, 0, 0,},
+						{30, 60, 0, 6,},
+						};
+	static const char *profile_name[6] = {
+					"BOOTUP_DEFAULT",
+					"3D_FULL_SCREEN",
+					"POWER_SAVING",
+					"VIDEO",
+					"VR",
+					"COMPUTE"};
+	static const char *title[6] = {"NUM",
+			"MODE_NAME",
+			"BUSY_SET_POINT",
+			"FPS",
+			"USE_RLC_BUSY",
+			"MIN_ACTIVE_LEVEL"};
+
+	if (!buf)
+		return -EINVAL;
+
+	size += sprintf(buf + size, "%s %16s %s %s %s %s\n",title[0],
+			title[1], title[2], title[3], title[4], title[5]);
+
+	for (i = 0; i <= PP_SMC_POWER_PROFILE_COMPUTE; i++)
+		size += sprintf(buf + size, "%3d %14s%s: %14d %3d %10d %14d\n",
+			i, profile_name[i], (i == hwmgr->power_profile_mode) ? "*" : " ",
+			profile_mode_setting[i][0], profile_mode_setting[i][1],
+			profile_mode_setting[i][2], profile_mode_setting[i][3]);
+
+	return size;
+}
+
+static int smu10_set_power_profile_mode(struct pp_hwmgr *hwmgr, long *input, uint32_t size)
+{
+	int workload_type = 0;
+
+	if (input[size] > PP_SMC_POWER_PROFILE_COMPUTE) {
+		pr_err("Invalid power profile mode %ld\n", input[size]);
+		return -EINVAL;
+	}
+	hwmgr->power_profile_mode = input[size];
+
+	/* conv PP_SMC_POWER_PROFILE* to WORKLOAD_PPLIB_*_BIT */
+	workload_type =
+		conv_power_profile_to_pplib_workload(hwmgr->power_profile_mode);
+	smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_ActiveProcessNotify,
+						1 << workload_type);
+
+	return 0;
+}
+
+
 static const struct pp_hwmgr_func smu10_hwmgr_funcs = {
 	.backend_init = smu10_hwmgr_backend_init,
 	.backend_fini = smu10_hwmgr_backend_fini,
@@ -1246,6 +1319,8 @@ static const struct pp_hwmgr_func smu10_hwmgr_funcs = {
 	.powergate_sdma = smu10_powergate_sdma,
 	.set_hard_min_dcefclk_by_freq = smu10_set_hard_min_dcefclk_by_freq,
 	.set_hard_min_fclk_by_freq = smu10_set_hard_min_fclk_by_freq,
+	.get_power_profile_mode = smu10_get_power_profile_mode,
+	.set_power_profile_mode = smu10_set_power_profile_mode,
 };
 
 int smu10_init_function_pointers(struct pp_hwmgr *hwmgr)
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c
index 83d3d935f3ac..048757e8f494 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c
@@ -77,7 +77,7 @@
 #define PCIE_BUS_CLK                10000
 #define TCLK                        (PCIE_BUS_CLK / 10)
 
-static const struct profile_mode_setting smu7_profiling[7] =
+static struct profile_mode_setting smu7_profiling[7] =
 					{{0, 0, 0, 0, 0, 0, 0, 0},
 					 {1, 0, 100, 30, 1, 0, 100, 10},
 					 {1, 10, 0, 30, 0, 0, 0, 0},
@@ -4984,17 +4984,27 @@ static int smu7_set_power_profile_mode(struct pp_hwmgr *hwmgr, long *input, uint
 	mode = input[size];
 	switch (mode) {
 	case PP_SMC_POWER_PROFILE_CUSTOM:
-		if (size < 8)
+		if (size < 8 && size != 0)
 			return -EINVAL;
-
-		tmp.bupdate_sclk = input[0];
-		tmp.sclk_up_hyst = input[1];
-		tmp.sclk_down_hyst = input[2];
-		tmp.sclk_activity = input[3];
-		tmp.bupdate_mclk = input[4];
-		tmp.mclk_up_hyst = input[5];
-		tmp.mclk_down_hyst = input[6];
-		tmp.mclk_activity = input[7];
+		/* If only CUSTOM is passed in, use the saved values. Check
+		 * that we actually have a CUSTOM profile by ensuring that
+		 * the "use sclk" or the "use mclk" bits are set
+		 */
+		tmp = smu7_profiling[PP_SMC_POWER_PROFILE_CUSTOM];
+		if (size == 0) {
+			if (tmp.bupdate_sclk == 0 && tmp.bupdate_mclk == 0)
+				return -EINVAL;
+		} else {
+			tmp.bupdate_sclk = input[0];
+			tmp.sclk_up_hyst = input[1];
+			tmp.sclk_down_hyst = input[2];
+			tmp.sclk_activity = input[3];
+			tmp.bupdate_mclk = input[4];
+			tmp.mclk_up_hyst = input[5];
+			tmp.mclk_down_hyst = input[6];
+			tmp.mclk_activity = input[7];
+			smu7_profiling[PP_SMC_POWER_PROFILE_CUSTOM] = tmp;
+		}
 		if (!smum_update_dpm_settings(hwmgr, &tmp)) {
 			memcpy(&data->current_profile_setting, &tmp, sizeof(struct profile_mode_setting));
 			hwmgr->power_profile_mode = mode;
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu9_baco.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu9_baco.c
new file mode 100644
index 000000000000..de0a37f7c632
--- /dev/null
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu9_baco.c
@@ -0,0 +1,66 @@
+/*
+ * Copyright 2018 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#include "amdgpu.h"
+#include "soc15.h"
+#include "soc15_hw_ip.h"
+#include "vega10_ip_offset.h"
+#include "soc15_common.h"
+#include "vega10_inc.h"
+#include "smu9_baco.h"
+
+int smu9_baco_get_capability(struct pp_hwmgr *hwmgr, bool *cap)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)(hwmgr->adev);
+	uint32_t reg, data;
+
+	*cap = false;
+	if (!phm_cap_enabled(hwmgr->platform_descriptor.platformCaps, PHM_PlatformCaps_BACO))
+		return 0;
+
+	WREG32(0x12074, 0xFFF0003B);
+	data = RREG32(0x12075);
+
+	if (data == 0x1) {
+		reg = RREG32_SOC15(NBIF, 0, mmRCC_BIF_STRAP0);
+
+		if (reg & RCC_BIF_STRAP0__STRAP_PX_CAPABLE_MASK)
+			*cap = true;
+	}
+
+	return 0;
+}
+
+int smu9_baco_get_state(struct pp_hwmgr *hwmgr, enum BACO_STATE *state)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)(hwmgr->adev);
+	uint32_t reg;
+
+	reg = RREG32_SOC15(NBIF, 0, mmBACO_CNTL);
+
+	if (reg & BACO_CNTL__BACO_MODE_MASK)
+		/* gfx has already entered BACO state */
+		*state = BACO_STATE_IN;
+	else
+		*state = BACO_STATE_OUT;
+	return 0;
+}
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu9_baco.h b/drivers/gpu/drm/amd/powerplay/hwmgr/smu9_baco.h
new file mode 100644
index 000000000000..84e90f801ac3
--- /dev/null
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu9_baco.h
@@ -0,0 +1,31 @@
+/*
+ * Copyright 2018 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#ifndef __SMU9_BACO_H__
+#define __SMU9_BACO_H__
+#include "hwmgr.h"
+#include "common_baco.h"
+
+extern int smu9_baco_get_capability(struct pp_hwmgr *hwmgr, bool *cap);
+extern int smu9_baco_get_state(struct pp_hwmgr *hwmgr, enum BACO_STATE *state);
+
+#endif
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_baco.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_baco.c
index 7337be5602e4..d168af4a4d78 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_baco.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_baco.c
@@ -85,48 +85,11 @@ static const struct soc15_baco_cmd_entry clean_baco_tbl[] =
 	{CMD_WRITE, SOC15_REG_ENTRY(NBIF, 0, mmBIOS_SCRATCH_7), 0, 0, 0, 0},
 };
 
-int vega10_baco_get_capability(struct pp_hwmgr *hwmgr, bool *cap)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)(hwmgr->adev);
-	uint32_t reg, data;
-
-	*cap = false;
-	if (!phm_cap_enabled(hwmgr->platform_descriptor.platformCaps, PHM_PlatformCaps_BACO))
-		return 0;
-
-	WREG32(0x12074, 0xFFF0003B);
-	data = RREG32(0x12075);
-
-	if (data == 0x1) {
-		reg = RREG32_SOC15(NBIF, 0, mmRCC_BIF_STRAP0);
-
-		if (reg & RCC_BIF_STRAP0__STRAP_PX_CAPABLE_MASK)
-			*cap = true;
-	}
-
-	return 0;
-}
-
-int vega10_baco_get_state(struct pp_hwmgr *hwmgr, enum BACO_STATE *state)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)(hwmgr->adev);
-	uint32_t reg;
-
-	reg = RREG32_SOC15(NBIF, 0, mmBACO_CNTL);
-
-	if (reg & BACO_CNTL__BACO_MODE_MASK)
-		/* gfx has already entered BACO state */
-		*state = BACO_STATE_IN;
-	else
-		*state = BACO_STATE_OUT;
-	return 0;
-}
-
 int vega10_baco_set_state(struct pp_hwmgr *hwmgr, enum BACO_STATE state)
 {
 	enum BACO_STATE cur_state;
 
-	vega10_baco_get_state(hwmgr, &cur_state);
+	smu9_baco_get_state(hwmgr, &cur_state);
 
 	if (cur_state == state)
 		/* aisc already in the target state */
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_baco.h b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_baco.h
index f7a3ffa744b3..96d793f026a5 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_baco.h
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_baco.h
@@ -22,11 +22,8 @@
  */
 #ifndef __VEGA10_BACO_H__
 #define __VEGA10_BACO_H__
-#include "hwmgr.h"
-#include "common_baco.h"
+#include "smu9_baco.h"
 
-extern int vega10_baco_get_capability(struct pp_hwmgr *hwmgr, bool *cap);
-extern int vega10_baco_get_state(struct pp_hwmgr *hwmgr, enum BACO_STATE *state);
 extern int vega10_baco_set_state(struct pp_hwmgr *hwmgr, enum BACO_STATE state);
 
 #endif
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c
index 5c4f701939ea..384c37875cd0 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c
@@ -1427,6 +1427,15 @@ static int vega10_setup_default_dpm_tables(struct pp_hwmgr *hwmgr)
 
 	vega10_setup_default_pcie_table(hwmgr);
 
+	/* Zero out the saved copy of the CUSTOM profile
+	 * This will be checked when trying to set the profile
+	 * and will require that new values be passed in
+	 */
+	data->custom_profile_mode[0] = 0;
+	data->custom_profile_mode[1] = 0;
+	data->custom_profile_mode[2] = 0;
+	data->custom_profile_mode[3] = 0;
+
 	/* save a copy of the default DPM table */
 	memcpy(&(data->golden_dpm_table), &(data->dpm_table),
 			sizeof(struct vega10_dpm_table));
@@ -4904,16 +4913,23 @@ static int vega10_set_power_profile_mode(struct pp_hwmgr *hwmgr, long *input, ui
 	uint8_t FPS;
 	uint8_t use_rlc_busy;
 	uint8_t min_active_level;
+	uint32_t power_profile_mode = input[size];
 
-	hwmgr->power_profile_mode = input[size];
-
-	smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetWorkloadMask,
-						1<<hwmgr->power_profile_mode);
-
-	if (hwmgr->power_profile_mode == PP_SMC_POWER_PROFILE_CUSTOM) {
-		if (size == 0 || size > 4)
+	if (power_profile_mode == PP_SMC_POWER_PROFILE_CUSTOM) {
+		if (size != 0 && size != 4)
 			return -EINVAL;
 
+		/* If size = 0 and the CUSTOM profile has been set already
+		 * then just apply the profile. The copy stored in the hwmgr
+		 * is zeroed out on init
+		 */
+		if (size == 0) {
+			if (data->custom_profile_mode[0] != 0)
+				goto out;
+			else
+				return -EINVAL;
+		}
+
 		data->custom_profile_mode[0] = busy_set_point = input[0];
 		data->custom_profile_mode[1] = FPS = input[1];
 		data->custom_profile_mode[2] = use_rlc_busy = input[2];
@@ -4924,6 +4940,11 @@ static int vega10_set_power_profile_mode(struct pp_hwmgr *hwmgr, long *input, ui
 					use_rlc_busy << 16 | min_active_level<<24);
 	}
 
+out:
+	smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetWorkloadMask,
+						1 << power_profile_mode);
+	hwmgr->power_profile_mode = power_profile_mode;
+
 	return 0;
 }
 
@@ -5170,8 +5191,8 @@ static const struct pp_hwmgr_func vega10_hwmgr_funcs = {
 	.set_power_limit = vega10_set_power_limit,
 	.odn_edit_dpm_table = vega10_odn_edit_dpm_table,
 	.get_performance_level = vega10_get_performance_level,
-	.get_asic_baco_capability = vega10_baco_get_capability,
-	.get_asic_baco_state = vega10_baco_get_state,
+	.get_asic_baco_capability = smu9_baco_get_capability,
+	.get_asic_baco_state = smu9_baco_get_state,
 	.set_asic_baco_state = vega10_baco_set_state,
 	.enable_mgpu_fan_boost = vega10_enable_mgpu_fan_boost,
 	.get_ppfeature_status = vega10_get_ppfeature_status,
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_baco.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_baco.c
new file mode 100644
index 000000000000..9d8ca94a8f0c
--- /dev/null
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_baco.c
@@ -0,0 +1,119 @@
+/*
+ * Copyright 2019 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#include "amdgpu.h"
+#include "soc15.h"
+#include "soc15_hw_ip.h"
+#include "vega10_ip_offset.h"
+#include "soc15_common.h"
+#include "vega12_inc.h"
+#include "vega12_ppsmc.h"
+#include "vega12_baco.h"
+
+static const struct soc15_baco_cmd_entry  pre_baco_tbl[] =
+{
+	{ CMD_READMODIFYWRITE, NBIF_HWID, 0, mmBIF_DOORBELL_CNTL_BASE_IDX, mmBIF_DOORBELL_CNTL, BIF_DOORBELL_CNTL__DOORBELL_MONITOR_EN_MASK, BIF_DOORBELL_CNTL__DOORBELL_MONITOR_EN__SHIFT, 0, 0 },
+	{ CMD_WRITE, NBIF_HWID, 0, mmBIF_FB_EN_BASE_IDX, mmBIF_FB_EN, 0, 0, 0, 0 },
+	{ CMD_READMODIFYWRITE, NBIF_HWID, 0, mmRCC_BACO_CNTL_MISC_BASE_IDX, mmBACO_CNTL, BACO_CNTL__BACO_DSTATE_BYPASS_MASK, BACO_CNTL__BACO_DSTATE_BYPASS__SHIFT, 0, 1 },
+	{ CMD_READMODIFYWRITE, NBIF_HWID, 0, mmRCC_BACO_CNTL_MISC_BASE_IDX, mmBACO_CNTL, BACO_CNTL__BACO_RST_INTR_MASK_MASK, BACO_CNTL__BACO_RST_INTR_MASK__SHIFT, 0, 1 }
+};
+
+static const struct soc15_baco_cmd_entry enter_baco_tbl[] =
+{
+	{ CMD_WAITFOR, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__SOC_DOMAIN_IDLE_MASK, THM_BACO_CNTL__SOC_DOMAIN_IDLE__SHIFT, 0xffffffff, 0x80000000 },
+	{ CMD_READMODIFYWRITE, NBIF_HWID, 0, mmRCC_BACO_CNTL_MISC_BASE_IDX, mmBACO_CNTL, BACO_CNTL__BACO_EN_MASK, BACO_CNTL__BACO_EN__SHIFT, 0, 1 },
+	{ CMD_READMODIFYWRITE, NBIF_HWID, 0, mmRCC_BACO_CNTL_MISC_BASE_IDX, mmBACO_CNTL, BACO_CNTL__BACO_BIF_LCLK_SWITCH_MASK, BACO_CNTL__BACO_BIF_LCLK_SWITCH__SHIFT, 0, 1 },
+	{ CMD_READMODIFYWRITE, NBIF_HWID, 0, mmRCC_BACO_CNTL_MISC_BASE_IDX, mmBACO_CNTL, BACO_CNTL__BACO_DUMMY_EN_MASK, BACO_CNTL__BACO_DUMMY_EN__SHIFT, 0, 1 },
+	{ CMD_READMODIFYWRITE, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__BACO_SOC_VDCI_RESET_MASK, THM_BACO_CNTL__BACO_SOC_VDCI_RESET__SHIFT, 0, 1 },
+	{ CMD_READMODIFYWRITE, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__BACO_SMNCLK_MUX_MASK, THM_BACO_CNTL__BACO_SMNCLK_MUX__SHIFT, 0, 1 },
+	{ CMD_READMODIFYWRITE, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__BACO_ISO_EN_MASK, THM_BACO_CNTL__BACO_ISO_EN__SHIFT, 0, 1 },
+	{ CMD_READMODIFYWRITE, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__BACO_AEB_ISO_EN_MASK, THM_BACO_CNTL__BACO_AEB_ISO_EN__SHIFT, 0, 1 },
+	{ CMD_READMODIFYWRITE, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__BACO_ANA_ISO_EN_MASK, THM_BACO_CNTL__BACO_ANA_ISO_EN__SHIFT, 0, 1 },
+	{ CMD_READMODIFYWRITE, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__BACO_SOC_REFCLK_OFF_MASK, THM_BACO_CNTL__BACO_SOC_REFCLK_OFF__SHIFT, 0, 1 },
+	{ CMD_READMODIFYWRITE, NBIF_HWID, 0, mmRCC_BACO_CNTL_MISC_BASE_IDX, mmBACO_CNTL, BACO_CNTL__BACO_POWER_OFF_MASK, BACO_CNTL__BACO_POWER_OFF__SHIFT, 0, 1 },
+	{ CMD_DELAY_MS, 0, 0, 0, 5, 0 },
+	{ CMD_READMODIFYWRITE, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__BACO_RESET_EN_MASK, THM_BACO_CNTL__BACO_RESET_EN__SHIFT, 0, 1 },
+	{ CMD_READMODIFYWRITE, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__BACO_PWROKRAW_CNTL_MASK, THM_BACO_CNTL__BACO_PWROKRAW_CNTL__SHIFT, 0, 0 },
+	{ CMD_WAITFOR, NBIF_HWID, 0, mmRCC_BACO_CNTL_MISC_BASE_IDX, mmBACO_CNTL, BACO_CNTL__BACO_MODE_MASK, BACO_CNTL__BACO_MODE__SHIFT, 0xffffffff, 0x100 }
+};
+
+static const struct soc15_baco_cmd_entry exit_baco_tbl[] =
+{
+	{ CMD_READMODIFYWRITE, NBIF_HWID, 0, mmRCC_BACO_CNTL_MISC_BASE_IDX, mmBACO_CNTL, BACO_CNTL__BACO_POWER_OFF_MASK, BACO_CNTL__BACO_POWER_OFF__SHIFT, 0, 0 },
+	{ CMD_DELAY_MS, 0, 0, 0, 0, 0, 0, 10, 0 },
+	{ CMD_READMODIFYWRITE, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__BACO_SOC_REFCLK_OFF_MASK, THM_BACO_CNTL__BACO_SOC_REFCLK_OFF__SHIFT, 0, 0 },
+	{ CMD_READMODIFYWRITE, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__BACO_ANA_ISO_EN_MASK, THM_BACO_CNTL__BACO_ANA_ISO_EN__SHIFT, 0, 0 },
+	{ CMD_READMODIFYWRITE, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__BACO_AEB_ISO_EN_MASK, THM_BACO_CNTL__BACO_AEB_ISO_EN__SHIFT, 0, 0 },
+	{ CMD_READMODIFYWRITE, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__BACO_ISO_EN_MASK, THM_BACO_CNTL__BACO_ISO_EN__SHIFT, 0, 0 },
+	{ CMD_READMODIFYWRITE, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__BACO_PWROKRAW_CNTL_MASK, THM_BACO_CNTL__BACO_PWROKRAW_CNTL__SHIFT, 0, 1 },
+	{ CMD_READMODIFYWRITE, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__BACO_SMNCLK_MUX_MASK, THM_BACO_CNTL__BACO_SMNCLK_MUX__SHIFT, 0, 0 },
+	{ CMD_READMODIFYWRITE, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__BACO_SOC_VDCI_RESET_MASK, THM_BACO_CNTL__BACO_SOC_VDCI_RESET__SHIFT, 0, 0 },
+	{ CMD_READMODIFYWRITE, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__BACO_EXIT_MASK, THM_BACO_CNTL__BACO_EXIT__SHIFT, 0, 1 },
+	{ CMD_READMODIFYWRITE, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__BACO_RESET_EN_MASK, THM_BACO_CNTL__BACO_RESET_EN__SHIFT, 0, 0 },
+	{ CMD_WAITFOR, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__BACO_EXIT_MASK, 0, 0xffffffff, 0 },
+	{ CMD_READMODIFYWRITE, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__BACO_SB_AXI_FENCE_MASK, THM_BACO_CNTL__BACO_SB_AXI_FENCE__SHIFT, 0, 0 },
+	{ CMD_READMODIFYWRITE, NBIF_HWID, 0, mmRCC_BACO_CNTL_MISC_BASE_IDX, mmBACO_CNTL, BACO_CNTL__BACO_DUMMY_EN_MASK, BACO_CNTL__BACO_DUMMY_EN__SHIFT, 0, 0 },
+	{ CMD_READMODIFYWRITE, NBIF_HWID, 0, mmRCC_BACO_CNTL_MISC_BASE_IDX, mmBACO_CNTL, BACO_CNTL__BACO_BIF_LCLK_SWITCH_MASK, BACO_CNTL__BACO_BIF_LCLK_SWITCH__SHIFT, 0, 0 },
+	{ CMD_READMODIFYWRITE, NBIF_HWID, 0, mmRCC_BACO_CNTL_MISC_BASE_IDX, mmBACO_CNTL, BACO_CNTL__BACO_EN_MASK, BACO_CNTL__BACO_EN__SHIFT, 0, 0 },
+	{ CMD_WAITFOR, NBIF_HWID, 0, mmRCC_BACO_CNTL_MISC_BASE_IDX, mmBACO_CNTL, BACO_CNTL__BACO_MODE_MASK, 0, 0xffffffff, 0 }
+};
+
+static const struct soc15_baco_cmd_entry clean_baco_tbl[] =
+{
+	{ CMD_WRITE, NBIF_HWID, 0, mmBIOS_SCRATCH_6_BASE_IDX, mmBIOS_SCRATCH_6, 0, 0, 0, 0 },
+	{ CMD_WRITE, NBIF_HWID, 0, mmBIOS_SCRATCH_7_BASE_IDX, mmBIOS_SCRATCH_7, 0, 0, 0, 0 }
+};
+
+int vega12_baco_set_state(struct pp_hwmgr *hwmgr, enum BACO_STATE state)
+{
+	enum BACO_STATE cur_state;
+
+	smu9_baco_get_state(hwmgr, &cur_state);
+
+	if (cur_state == state)
+		/* aisc already in the target state */
+		return 0;
+
+	if (state == BACO_STATE_IN) {
+		if (soc15_baco_program_registers(hwmgr, pre_baco_tbl,
+					     ARRAY_SIZE(pre_baco_tbl))) {
+			if (smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_EnterBaco, 0))
+				return -EINVAL;
+
+			if (soc15_baco_program_registers(hwmgr, enter_baco_tbl,
+						   ARRAY_SIZE(enter_baco_tbl)))
+				return 0;
+		}
+	} else if (state == BACO_STATE_OUT) {
+		/* HW requires at least 20ms between regulator off and on */
+		msleep(20);
+		/* Execute Hardware BACO exit sequence */
+		if (soc15_baco_program_registers(hwmgr, exit_baco_tbl,
+					     ARRAY_SIZE(exit_baco_tbl))) {
+			if (soc15_baco_program_registers(hwmgr, clean_baco_tbl,
+						     ARRAY_SIZE(clean_baco_tbl)))
+				return 0;
+		}
+	}
+
+	return -EINVAL;
+}
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_baco.h b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_baco.h
new file mode 100644
index 000000000000..57b72e5a95ae
--- /dev/null
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_baco.h
@@ -0,0 +1,29 @@
+/*
+ * Copyright 2019 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#ifndef __VEGA12_BACO_H__
+#define __VEGA12_BACO_H__
+#include "smu9_baco.h"
+
+extern int vega12_baco_set_state(struct pp_hwmgr *hwmgr, enum BACO_STATE state);
+
+#endif
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c
index bdb48e94eff6..707cd4b0357f 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c
@@ -45,6 +45,7 @@
 #include "ppinterrupt.h"
 #include "pp_overdriver.h"
 #include "pp_thermal.h"
+#include "vega12_baco.h"
 
 
 static int vega12_force_clock_level(struct pp_hwmgr *hwmgr,
@@ -2626,8 +2627,12 @@ static const struct pp_hwmgr_func vega12_hwmgr_funcs = {
 	.start_thermal_controller = vega12_start_thermal_controller,
 	.powergate_gfx = vega12_gfx_off_control,
 	.get_performance_level = vega12_get_performance_level,
+	.get_asic_baco_capability = smu9_baco_get_capability,
+	.get_asic_baco_state = smu9_baco_get_state,
+	.set_asic_baco_state = vega12_baco_set_state,
 	.get_ppfeature_status = vega12_get_ppfeature_status,
 	.set_ppfeature_status = vega12_set_ppfeature_status,
+
 };
 
 int vega12_hwmgr_init(struct pp_hwmgr *hwmgr)
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_inc.h b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_inc.h
index 30b278c50222..e6d9e84059e1 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_inc.h
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_inc.h
@@ -35,5 +35,7 @@
 #include "asic_reg/gc/gc_9_2_1_sh_mask.h"
 
 #include "asic_reg/nbio/nbio_6_1_offset.h"
+#include "asic_reg/nbio/nbio_6_1_offset.h"
+#include "asic_reg/nbio/nbio_6_1_sh_mask.h"
 
 #endif
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.c
index 5e8602a79b1c..df6ff9252401 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.c
@@ -27,6 +27,7 @@
 #include "vega20_inc.h"
 #include "vega20_ppsmc.h"
 #include "vega20_baco.h"
+#include "vega20_smumgr.h"
 
 
 
@@ -101,3 +102,14 @@ int vega20_baco_set_state(struct pp_hwmgr *hwmgr, enum BACO_STATE state)
 
 	return 0;
 }
+
+int vega20_baco_apply_vdci_flush_workaround(struct pp_hwmgr *hwmgr)
+{
+	int ret = 0;
+
+	ret = vega20_set_pptable_driver_address(hwmgr);
+	if (ret)
+		return ret;
+
+	return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_BacoWorkAroundFlushVDCI);
+}
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.h b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.h
index 51c7f8392925..f06471e712dc 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.h
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.h
@@ -28,5 +28,6 @@
 extern int vega20_baco_get_capability(struct pp_hwmgr *hwmgr, bool *cap);
 extern int vega20_baco_get_state(struct pp_hwmgr *hwmgr, enum BACO_STATE *state);
 extern int vega20_baco_set_state(struct pp_hwmgr *hwmgr, enum BACO_STATE state);
+extern int vega20_baco_apply_vdci_flush_workaround(struct pp_hwmgr *hwmgr);
 
 #endif
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c
index 23b5b94a4939..9b9f87b84910 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c
@@ -434,6 +434,7 @@ static int vega20_hwmgr_backend_init(struct pp_hwmgr *hwmgr)
 	hwmgr->platform_descriptor.clockStep.memoryClock = 500;
 
 	data->total_active_cus = adev->gfx.cu_info.number;
+	data->is_custom_profile_set = false;
 
 	return 0;
 }
@@ -450,6 +451,7 @@ static int vega20_init_sclk_threshold(struct pp_hwmgr *hwmgr)
 
 static int vega20_setup_asic_task(struct pp_hwmgr *hwmgr)
 {
+	struct amdgpu_device *adev = (struct amdgpu_device *)(hwmgr->adev);
 	int ret = 0;
 
 	ret = vega20_init_sclk_threshold(hwmgr);
@@ -457,7 +459,15 @@ static int vega20_setup_asic_task(struct pp_hwmgr *hwmgr)
 			"Failed to init sclk threshold!",
 			return ret);
 
-	return 0;
+	if (adev->in_baco_reset) {
+		adev->in_baco_reset = 0;
+
+		ret = vega20_baco_apply_vdci_flush_workaround(hwmgr);
+		if (ret)
+			pr_err("Failed to apply vega20 baco workaround!\n");
+	}
+
+	return ret;
 }
 
 /*
@@ -3450,7 +3460,18 @@ static void vega20_power_gate_vce(struct pp_hwmgr *hwmgr, bool bgate)
 		return ;
 
 	data->vce_power_gated = bgate;
-	vega20_enable_disable_vce_dpm(hwmgr, !bgate);
+	if (bgate) {
+		vega20_enable_disable_vce_dpm(hwmgr, !bgate);
+		amdgpu_device_ip_set_powergating_state(hwmgr->adev,
+						AMD_IP_BLOCK_TYPE_VCE,
+						AMD_PG_STATE_GATE);
+	} else {
+		amdgpu_device_ip_set_powergating_state(hwmgr->adev,
+						AMD_IP_BLOCK_TYPE_VCE,
+						AMD_PG_STATE_UNGATE);
+		vega20_enable_disable_vce_dpm(hwmgr, !bgate);
+	}
+
 }
 
 static void vega20_power_gate_uvd(struct pp_hwmgr *hwmgr, bool bgate)
@@ -3826,16 +3847,19 @@ static int vega20_set_power_profile_mode(struct pp_hwmgr *hwmgr, long *input, ui
 {
 	DpmActivityMonitorCoeffInt_t activity_monitor;
 	int workload_type, result = 0;
+	uint32_t power_profile_mode = input[size];
 
-	hwmgr->power_profile_mode = input[size];
-
-	if (hwmgr->power_profile_mode > PP_SMC_POWER_PROFILE_CUSTOM) {
-		pr_err("Invalid power profile mode %d\n", hwmgr->power_profile_mode);
+	if (power_profile_mode > PP_SMC_POWER_PROFILE_CUSTOM) {
+		pr_err("Invalid power profile mode %d\n", power_profile_mode);
 		return -EINVAL;
 	}
 
-	if (hwmgr->power_profile_mode == PP_SMC_POWER_PROFILE_CUSTOM) {
-		if (size < 10)
+	if (power_profile_mode == PP_SMC_POWER_PROFILE_CUSTOM) {
+		struct vega20_hwmgr *data =
+			(struct vega20_hwmgr *)(hwmgr->backend);
+		if (size == 0 && !data->is_custom_profile_set)
+			return -EINVAL;
+		if (size < 10 && size != 0)
 			return -EINVAL;
 
 		result = vega20_get_activity_monitor_coeff(hwmgr,
@@ -3845,6 +3869,13 @@ static int vega20_set_power_profile_mode(struct pp_hwmgr *hwmgr, long *input, ui
 				"[SetPowerProfile] Failed to get activity monitor!",
 				return result);
 
+		/* If size==0, then we want to apply the already-configured
+		 * CUSTOM profile again. Just apply it, since we checked its
+		 * validity above
+		 */
+		if (size == 0)
+			goto out;
+
 		switch (input[0]) {
 		case 0: /* Gfxclk */
 			activity_monitor.Gfx_FPS = input[1];
@@ -3895,17 +3926,21 @@ static int vega20_set_power_profile_mode(struct pp_hwmgr *hwmgr, long *input, ui
 		result = vega20_set_activity_monitor_coeff(hwmgr,
 				(uint8_t *)(&activity_monitor),
 				WORKLOAD_PPLIB_CUSTOM_BIT);
+		data->is_custom_profile_set = true;
 		PP_ASSERT_WITH_CODE(!result,
 				"[SetPowerProfile] Failed to set activity monitor!",
 				return result);
 	}
 
+out:
 	/* conv PP_SMC_POWER_PROFILE* to WORKLOAD_PPLIB_*_BIT */
 	workload_type =
-		conv_power_profile_to_pplib_workload(hwmgr->power_profile_mode);
+		conv_power_profile_to_pplib_workload(power_profile_mode);
 	smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetWorkloadMask,
 						1 << workload_type);
 
+	hwmgr->power_profile_mode = power_profile_mode;
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.h b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.h
index ac2a3118a0ae..2c3125f82b24 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.h
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.h
@@ -531,6 +531,8 @@ struct vega20_hwmgr {
 	bool                           pcie_parameters_override;
 	uint32_t                       pcie_gen_level1;
 	uint32_t                       pcie_width_level1;
+
+	bool                           is_custom_profile_set;
 };
 
 #define VEGA20_DPM2_NEAR_TDP_DEC                      10
diff --git a/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h
new file mode 100644
index 000000000000..c8b168b3413b
--- /dev/null
+++ b/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h
@@ -0,0 +1,770 @@
+/*
+ * Copyright 2019 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#ifndef __AMDGPU_SMU_H__
+#define __AMDGPU_SMU_H__
+
+#include "amdgpu.h"
+#include "kgd_pp_interface.h"
+#include "dm_pp_interface.h"
+
+struct smu_hw_power_state {
+	unsigned int magic;
+};
+
+struct smu_power_state;
+
+enum smu_state_ui_label {
+	SMU_STATE_UI_LABEL_NONE,
+	SMU_STATE_UI_LABEL_BATTERY,
+	SMU_STATE_UI_TABEL_MIDDLE_LOW,
+	SMU_STATE_UI_LABEL_BALLANCED,
+	SMU_STATE_UI_LABEL_MIDDLE_HIGHT,
+	SMU_STATE_UI_LABEL_PERFORMANCE,
+	SMU_STATE_UI_LABEL_BACO,
+};
+
+enum smu_state_classification_flag {
+	SMU_STATE_CLASSIFICATION_FLAG_BOOT                     = 0x0001,
+	SMU_STATE_CLASSIFICATION_FLAG_THERMAL                  = 0x0002,
+	SMU_STATE_CLASSIFICATIN_FLAG_LIMITED_POWER_SOURCE      = 0x0004,
+	SMU_STATE_CLASSIFICATION_FLAG_RESET                    = 0x0008,
+	SMU_STATE_CLASSIFICATION_FLAG_FORCED                   = 0x0010,
+	SMU_STATE_CLASSIFICATION_FLAG_USER_3D_PERFORMANCE      = 0x0020,
+	SMU_STATE_CLASSIFICATION_FLAG_USER_2D_PERFORMANCE      = 0x0040,
+	SMU_STATE_CLASSIFICATION_FLAG_3D_PERFORMANCE           = 0x0080,
+	SMU_STATE_CLASSIFICATION_FLAG_AC_OVERDIRVER_TEMPLATE   = 0x0100,
+	SMU_STATE_CLASSIFICATION_FLAG_UVD                      = 0x0200,
+	SMU_STATE_CLASSIFICATION_FLAG_3D_PERFORMANCE_LOW       = 0x0400,
+	SMU_STATE_CLASSIFICATION_FLAG_ACPI                     = 0x0800,
+	SMU_STATE_CLASSIFICATION_FLAG_HD2                      = 0x1000,
+	SMU_STATE_CLASSIFICATION_FLAG_UVD_HD                   = 0x2000,
+	SMU_STATE_CLASSIFICATION_FLAG_UVD_SD                   = 0x4000,
+	SMU_STATE_CLASSIFICATION_FLAG_USER_DC_PERFORMANCE      = 0x8000,
+	SMU_STATE_CLASSIFICATION_FLAG_DC_OVERDIRVER_TEMPLATE   = 0x10000,
+	SMU_STATE_CLASSIFICATION_FLAG_BACO                     = 0x20000,
+	SMU_STATE_CLASSIFICATIN_FLAG_LIMITED_POWER_SOURCE2      = 0x40000,
+	SMU_STATE_CLASSIFICATION_FLAG_ULV                      = 0x80000,
+	SMU_STATE_CLASSIFICATION_FLAG_UVD_MVC                  = 0x100000,
+};
+
+struct smu_state_classification_block {
+	enum smu_state_ui_label         ui_label;
+	enum smu_state_classification_flag  flags;
+	int                          bios_index;
+	bool                      temporary_state;
+	bool                      to_be_deleted;
+};
+
+struct smu_state_pcie_block {
+	unsigned int lanes;
+};
+
+enum smu_refreshrate_source {
+	SMU_REFRESHRATE_SOURCE_EDID,
+	SMU_REFRESHRATE_SOURCE_EXPLICIT
+};
+
+struct smu_state_display_block {
+	bool              disable_frame_modulation;
+	bool              limit_refreshrate;
+	enum smu_refreshrate_source refreshrate_source;
+	int                  explicit_refreshrate;
+	int                  edid_refreshrate_index;
+	bool              enable_vari_bright;
+};
+
+struct smu_state_memroy_block {
+	bool              dll_off;
+	uint8_t                 m3arb;
+	uint8_t                 unused[3];
+};
+
+struct smu_state_software_algorithm_block {
+	bool disable_load_balancing;
+	bool enable_sleep_for_timestamps;
+};
+
+struct smu_temperature_range {
+	int min;
+	int max;
+};
+
+struct smu_state_validation_block {
+	bool single_display_only;
+	bool disallow_on_dc;
+	uint8_t supported_power_levels;
+};
+
+struct smu_uvd_clocks {
+	uint32_t vclk;
+	uint32_t dclk;
+};
+
+/**
+* Structure to hold a SMU Power State.
+*/
+struct smu_power_state {
+	uint32_t                                      id;
+	struct list_head                              ordered_list;
+	struct list_head                              all_states_list;
+
+	struct smu_state_classification_block         classification;
+	struct smu_state_validation_block             validation;
+	struct smu_state_pcie_block                   pcie;
+	struct smu_state_display_block                display;
+	struct smu_state_memroy_block                 memory;
+	struct smu_temperature_range                  temperatures;
+	struct smu_state_software_algorithm_block     software;
+	struct smu_uvd_clocks                         uvd_clocks;
+	struct smu_hw_power_state                     hardware;
+};
+
+enum smu_message_type
+{
+	SMU_MSG_TestMessage = 0,
+	SMU_MSG_GetSmuVersion,
+	SMU_MSG_GetDriverIfVersion,
+	SMU_MSG_SetAllowedFeaturesMaskLow,
+	SMU_MSG_SetAllowedFeaturesMaskHigh,
+	SMU_MSG_EnableAllSmuFeatures,
+	SMU_MSG_DisableAllSmuFeatures,
+	SMU_MSG_EnableSmuFeaturesLow,
+	SMU_MSG_EnableSmuFeaturesHigh,
+	SMU_MSG_DisableSmuFeaturesLow,
+	SMU_MSG_DisableSmuFeaturesHigh,
+	SMU_MSG_GetEnabledSmuFeaturesLow,
+	SMU_MSG_GetEnabledSmuFeaturesHigh,
+	SMU_MSG_SetWorkloadMask,
+	SMU_MSG_SetPptLimit,
+	SMU_MSG_SetDriverDramAddrHigh,
+	SMU_MSG_SetDriverDramAddrLow,
+	SMU_MSG_SetToolsDramAddrHigh,
+	SMU_MSG_SetToolsDramAddrLow,
+	SMU_MSG_TransferTableSmu2Dram,
+	SMU_MSG_TransferTableDram2Smu,
+	SMU_MSG_UseDefaultPPTable,
+	SMU_MSG_UseBackupPPTable,
+	SMU_MSG_RunBtc,
+	SMU_MSG_RequestI2CBus,
+	SMU_MSG_ReleaseI2CBus,
+	SMU_MSG_SetFloorSocVoltage,
+	SMU_MSG_SoftReset,
+	SMU_MSG_StartBacoMonitor,
+	SMU_MSG_CancelBacoMonitor,
+	SMU_MSG_EnterBaco,
+	SMU_MSG_SetSoftMinByFreq,
+	SMU_MSG_SetSoftMaxByFreq,
+	SMU_MSG_SetHardMinByFreq,
+	SMU_MSG_SetHardMaxByFreq,
+	SMU_MSG_GetMinDpmFreq,
+	SMU_MSG_GetMaxDpmFreq,
+	SMU_MSG_GetDpmFreqByIndex,
+	SMU_MSG_GetDpmClockFreq,
+	SMU_MSG_GetSsVoltageByDpm,
+	SMU_MSG_SetMemoryChannelConfig,
+	SMU_MSG_SetGeminiMode,
+	SMU_MSG_SetGeminiApertureHigh,
+	SMU_MSG_SetGeminiApertureLow,
+	SMU_MSG_SetMinLinkDpmByIndex,
+	SMU_MSG_OverridePcieParameters,
+	SMU_MSG_OverDriveSetPercentage,
+	SMU_MSG_SetMinDeepSleepDcefclk,
+	SMU_MSG_ReenableAcDcInterrupt,
+	SMU_MSG_NotifyPowerSource,
+	SMU_MSG_SetUclkFastSwitch,
+	SMU_MSG_SetUclkDownHyst,
+	SMU_MSG_GfxDeviceDriverReset,
+	SMU_MSG_GetCurrentRpm,
+	SMU_MSG_SetVideoFps,
+	SMU_MSG_SetTjMax,
+	SMU_MSG_SetFanTemperatureTarget,
+	SMU_MSG_PrepareMp1ForUnload,
+	SMU_MSG_DramLogSetDramAddrHigh,
+	SMU_MSG_DramLogSetDramAddrLow,
+	SMU_MSG_DramLogSetDramSize,
+	SMU_MSG_SetFanMaxRpm,
+	SMU_MSG_SetFanMinPwm,
+	SMU_MSG_ConfigureGfxDidt,
+	SMU_MSG_NumOfDisplays,
+	SMU_MSG_RemoveMargins,
+	SMU_MSG_ReadSerialNumTop32,
+	SMU_MSG_ReadSerialNumBottom32,
+	SMU_MSG_SetSystemVirtualDramAddrHigh,
+	SMU_MSG_SetSystemVirtualDramAddrLow,
+	SMU_MSG_WaflTest,
+	SMU_MSG_SetFclkGfxClkRatio,
+	SMU_MSG_AllowGfxOff,
+	SMU_MSG_DisallowGfxOff,
+	SMU_MSG_GetPptLimit,
+	SMU_MSG_GetDcModeMaxDpmFreq,
+	SMU_MSG_GetDebugData,
+	SMU_MSG_SetXgmiMode,
+	SMU_MSG_RunAfllBtc,
+	SMU_MSG_ExitBaco,
+	SMU_MSG_PrepareMp1ForReset,
+	SMU_MSG_PrepareMp1ForShutdown,
+	SMU_MSG_SetMGpuFanBoostLimitRpm,
+	SMU_MSG_GetAVFSVoltageByDpm,
+	SMU_MSG_MAX_COUNT,
+};
+
+enum smu_memory_pool_size
+{
+    SMU_MEMORY_POOL_SIZE_ZERO   = 0,
+    SMU_MEMORY_POOL_SIZE_256_MB = 0x10000000,
+    SMU_MEMORY_POOL_SIZE_512_MB = 0x20000000,
+    SMU_MEMORY_POOL_SIZE_1_GB   = 0x40000000,
+    SMU_MEMORY_POOL_SIZE_2_GB   = 0x80000000,
+};
+
+#define SMU_TABLE_INIT(tables, table_id, s, a, d)	\
+	do {						\
+		tables[table_id].size = s;		\
+		tables[table_id].align = a;		\
+		tables[table_id].domain = d;		\
+	} while (0)
+
+struct smu_table {
+	uint64_t size;
+	uint32_t align;
+	uint8_t domain;
+	uint64_t mc_address;
+	void *cpu_addr;
+	struct amdgpu_bo *bo;
+};
+
+enum smu_perf_level_designation {
+	PERF_LEVEL_ACTIVITY,
+	PERF_LEVEL_POWER_CONTAINMENT,
+};
+
+struct smu_performance_level {
+	uint32_t core_clock;
+	uint32_t memory_clock;
+	uint32_t vddc;
+	uint32_t vddci;
+	uint32_t non_local_mem_freq;
+	uint32_t non_local_mem_width;
+};
+
+struct smu_clock_info {
+	uint32_t min_mem_clk;
+	uint32_t max_mem_clk;
+	uint32_t min_eng_clk;
+	uint32_t max_eng_clk;
+	uint32_t min_bus_bandwidth;
+	uint32_t max_bus_bandwidth;
+};
+
+struct smu_bios_boot_up_values
+{
+	uint32_t			revision;
+	uint32_t			gfxclk;
+	uint32_t			uclk;
+	uint32_t			socclk;
+	uint32_t			dcefclk;
+	uint32_t			eclk;
+	uint32_t			vclk;
+	uint32_t			dclk;
+	uint16_t			vddc;
+	uint16_t			vddci;
+	uint16_t			mvddc;
+	uint16_t			vdd_gfx;
+	uint8_t				cooling_id;
+	uint32_t			pp_table_id;
+};
+
+struct smu_table_context
+{
+	void				*power_play_table;
+	uint32_t			power_play_table_size;
+	void				*hardcode_pptable;
+
+	void				*max_sustainable_clocks;
+	struct smu_bios_boot_up_values	boot_values;
+	void                            *driver_pptable;
+	struct smu_table		*tables;
+	uint32_t			table_count;
+	struct smu_table		memory_pool;
+	uint16_t                        software_shutdown_temp;
+	uint8_t                         thermal_controller_type;
+	uint16_t			TDPODLimit;
+
+	uint8_t				*od_feature_capabilities;
+	uint32_t			*od_settings_max;
+	uint32_t			*od_settings_min;
+	void				*overdrive_table;
+	void				*od8_settings;
+	bool				od_gfxclk_update;
+	bool				od_memclk_update;
+};
+
+struct smu_dpm_context {
+	uint32_t dpm_context_size;
+	void *dpm_context;
+	void *golden_dpm_context;
+	bool enable_umd_pstate;
+	enum amd_dpm_forced_level dpm_level;
+	enum amd_dpm_forced_level saved_dpm_level;
+	enum amd_dpm_forced_level requested_dpm_level;
+	struct smu_power_state *dpm_request_power_state;
+	struct smu_power_state *dpm_current_power_state;
+	struct mclock_latency_table *mclk_latency_table;
+};
+
+struct smu_power_context {
+	void *power_context;
+	uint32_t power_context_size;
+};
+
+
+#define SMU_FEATURE_MAX	(64)
+struct smu_feature
+{
+	uint32_t feature_num;
+	DECLARE_BITMAP(supported, SMU_FEATURE_MAX);
+	DECLARE_BITMAP(allowed, SMU_FEATURE_MAX);
+	DECLARE_BITMAP(enabled, SMU_FEATURE_MAX);
+	struct mutex mutex;
+};
+
+struct smu_clocks {
+	uint32_t engine_clock;
+	uint32_t memory_clock;
+	uint32_t bus_bandwidth;
+	uint32_t engine_clock_in_sr;
+	uint32_t dcef_clock;
+	uint32_t dcef_clock_in_sr;
+};
+
+#define MAX_REGULAR_DPM_NUM 16
+struct mclk_latency_entries {
+	uint32_t  frequency;
+	uint32_t  latency;
+};
+struct mclock_latency_table {
+	uint32_t  count;
+	struct mclk_latency_entries  entries[MAX_REGULAR_DPM_NUM];
+};
+
+#define WORKLOAD_POLICY_MAX 7
+struct smu_context
+{
+	struct amdgpu_device            *adev;
+
+	const struct smu_funcs		*funcs;
+	const struct pptable_funcs	*ppt_funcs;
+	struct mutex			mutex;
+	uint64_t pool_size;
+
+	struct smu_table_context	smu_table;
+	struct smu_dpm_context		smu_dpm;
+	struct smu_power_context	smu_power;
+	struct smu_feature		smu_feature;
+	struct amd_pp_display_configuration  *display_config;
+
+	uint32_t pstate_sclk;
+	uint32_t pstate_mclk;
+
+	bool od_enabled;
+	uint32_t power_limit;
+	uint32_t default_power_limit;
+
+	bool support_power_containment;
+	bool disable_watermark;
+
+#define WATERMARKS_EXIST	(1 << 0)
+#define WATERMARKS_LOADED	(1 << 1)
+	uint32_t watermarks_bitmap;
+
+	uint32_t workload_mask;
+	uint32_t workload_prority[WORKLOAD_POLICY_MAX];
+	uint32_t workload_setting[WORKLOAD_POLICY_MAX];
+	uint32_t power_profile_mode;
+	uint32_t default_power_profile_mode;
+
+	uint32_t smc_if_version;
+};
+
+struct pptable_funcs {
+	int (*alloc_dpm_context)(struct smu_context *smu);
+	int (*store_powerplay_table)(struct smu_context *smu);
+	int (*check_powerplay_table)(struct smu_context *smu);
+	int (*append_powerplay_table)(struct smu_context *smu);
+	int (*get_smu_msg_index)(struct smu_context *smu, uint32_t index);
+	int (*run_afll_btc)(struct smu_context *smu);
+	int (*get_unallowed_feature_mask)(struct smu_context *smu, uint32_t *feature_mask, uint32_t num);
+	enum amd_pm_state_type (*get_current_power_state)(struct smu_context *smu);
+	int (*set_default_dpm_table)(struct smu_context *smu);
+	int (*set_power_state)(struct smu_context *smu);
+	int (*populate_umd_state_clk)(struct smu_context *smu);
+	int (*print_clk_levels)(struct smu_context *smu, enum pp_clock_type type, char *buf);
+	int (*force_clk_levels)(struct smu_context *smu, enum pp_clock_type type, uint32_t mask);
+	int (*set_default_od8_settings)(struct smu_context *smu);
+	int (*update_specified_od8_value)(struct smu_context *smu,
+					  uint32_t index,
+					  uint32_t value);
+	int (*get_od_percentage)(struct smu_context *smu, enum pp_clock_type type);
+	int (*set_od_percentage)(struct smu_context *smu,
+				 enum pp_clock_type type,
+				 uint32_t value);
+	int (*od_edit_dpm_table)(struct smu_context *smu,
+				 enum PP_OD_DPM_TABLE_COMMAND type,
+				 long *input, uint32_t size);
+	int (*get_clock_by_type_with_latency)(struct smu_context *smu,
+					      enum amd_pp_clock_type type,
+					      struct
+					      pp_clock_levels_with_latency
+					      *clocks);
+	int (*get_clock_by_type_with_voltage)(struct smu_context *smu,
+					      enum amd_pp_clock_type type,
+					      struct
+					      pp_clock_levels_with_voltage
+					      *clocks);
+	int (*get_power_profile_mode)(struct smu_context *smu, char *buf);
+	int (*set_power_profile_mode)(struct smu_context *smu, long *input, uint32_t size);
+	enum amd_dpm_forced_level (*get_performance_level)(struct smu_context *smu);
+	int (*force_performance_level)(struct smu_context *smu, enum amd_dpm_forced_level level);
+	int (*pre_display_config_changed)(struct smu_context *smu);
+	int (*display_config_changed)(struct smu_context *smu);
+	int (*apply_clocks_adjust_rules)(struct smu_context *smu);
+	int (*notify_smc_dispaly_config)(struct smu_context *smu);
+	int (*force_dpm_limit_value)(struct smu_context *smu, bool highest);
+	int (*unforce_dpm_levels)(struct smu_context *smu);
+	int (*upload_dpm_level)(struct smu_context *smu, bool max,
+				uint32_t feature_mask);
+	int (*get_profiling_clk_mask)(struct smu_context *smu,
+				      enum amd_dpm_forced_level level,
+				      uint32_t *sclk_mask,
+				      uint32_t *mclk_mask,
+				      uint32_t *soc_mask);
+	int (*set_cpu_power_state)(struct smu_context *smu);
+};
+
+struct smu_funcs
+{
+	int (*init_microcode)(struct smu_context *smu);
+	int (*init_smc_tables)(struct smu_context *smu);
+	int (*fini_smc_tables)(struct smu_context *smu);
+	int (*init_power)(struct smu_context *smu);
+	int (*fini_power)(struct smu_context *smu);
+	int (*load_microcode)(struct smu_context *smu);
+	int (*check_fw_status)(struct smu_context *smu);
+	int (*read_pptable_from_vbios)(struct smu_context *smu);
+	int (*get_vbios_bootup_values)(struct smu_context *smu);
+	int (*get_clk_info_from_vbios)(struct smu_context *smu);
+	int (*check_pptable)(struct smu_context *smu);
+	int (*parse_pptable)(struct smu_context *smu);
+	int (*populate_smc_pptable)(struct smu_context *smu);
+	int (*check_fw_version)(struct smu_context *smu);
+	int (*write_pptable)(struct smu_context *smu);
+	int (*set_min_dcef_deep_sleep)(struct smu_context *smu);
+	int (*set_tool_table_location)(struct smu_context *smu);
+	int (*notify_memory_pool_location)(struct smu_context *smu);
+	int (*write_watermarks_table)(struct smu_context *smu);
+	int (*set_last_dcef_min_deep_sleep_clk)(struct smu_context *smu);
+	int (*system_features_control)(struct smu_context *smu, bool en);
+	int (*send_smc_msg)(struct smu_context *smu, uint16_t msg);
+	int (*send_smc_msg_with_param)(struct smu_context *smu, uint16_t msg, uint32_t param);
+	int (*read_smc_arg)(struct smu_context *smu, uint32_t *arg);
+	int (*init_display)(struct smu_context *smu);
+	int (*set_allowed_mask)(struct smu_context *smu);
+	int (*get_enabled_mask)(struct smu_context *smu, uint32_t *feature_mask, uint32_t num);
+	bool (*is_dpm_running)(struct smu_context *smu);
+	int (*update_feature_enable_state)(struct smu_context *smu, uint32_t feature_id, bool enabled);
+	int (*notify_display_change)(struct smu_context *smu);
+	int (*get_power_limit)(struct smu_context *smu, uint32_t *limit, bool def);
+	int (*set_power_limit)(struct smu_context *smu, uint32_t n);
+	int (*get_current_clk_freq)(struct smu_context *smu, uint32_t clk_id, uint32_t *value);
+	int (*init_max_sustainable_clocks)(struct smu_context *smu);
+	int (*start_thermal_control)(struct smu_context *smu);
+	int (*read_sensor)(struct smu_context *smu, enum amd_pp_sensors sensor,
+			   void *data, uint32_t *size);
+	int (*set_deep_sleep_dcefclk)(struct smu_context *smu, uint32_t clk);
+	int (*set_active_display_count)(struct smu_context *smu, uint32_t count);
+	int (*store_cc6_data)(struct smu_context *smu, uint32_t separation_time,
+			      bool cc6_disable, bool pstate_disable,
+			      bool pstate_switch_disable);
+	int (*get_clock_by_type)(struct smu_context *smu,
+				 enum amd_pp_clock_type type,
+				 struct amd_pp_clocks *clocks);
+	int (*get_max_high_clocks)(struct smu_context *smu,
+				   struct amd_pp_simple_clock_info *clocks);
+	int (*display_clock_voltage_request)(struct smu_context *smu, struct
+					     pp_display_clock_request
+					     *clock_req);
+	int (*get_dal_power_level)(struct smu_context *smu,
+				   struct amd_pp_simple_clock_info *clocks);
+	int (*get_perf_level)(struct smu_context *smu,
+			      enum smu_perf_level_designation designation,
+			      struct smu_performance_level *level);
+	int (*get_current_shallow_sleep_clocks)(struct smu_context *smu,
+						struct smu_clock_info *clocks);
+	int (*notify_smu_enable_pwe)(struct smu_context *smu);
+	int (*set_watermarks_for_clock_ranges)(struct smu_context *smu,
+					       struct dm_pp_wm_sets_with_clock_ranges_soc15 *clock_ranges);
+	int (*set_od8_default_settings)(struct smu_context *smu,
+					bool initialize);
+	int (*conv_power_profile_to_pplib_workload)(int power_profile);
+	int (*get_power_profile_mode)(struct smu_context *smu, char *buf);
+	int (*set_power_profile_mode)(struct smu_context *smu, long *input, uint32_t size);
+	int (*update_od8_settings)(struct smu_context *smu,
+				   uint32_t index,
+				   uint32_t value);
+	int (*dpm_set_uvd_enable)(struct smu_context *smu, bool enable);
+	int (*dpm_set_vce_enable)(struct smu_context *smu, bool enable);
+	uint32_t (*get_sclk)(struct smu_context *smu, bool low);
+	uint32_t (*get_mclk)(struct smu_context *smu, bool low);
+	int (*get_current_rpm)(struct smu_context *smu, uint32_t *speed);
+	uint32_t (*get_fan_control_mode)(struct smu_context *smu);
+	int (*set_fan_control_mode)(struct smu_context *smu, uint32_t mode);
+	int (*get_fan_speed_percent)(struct smu_context *smu, uint32_t *speed);
+	int (*set_fan_speed_percent)(struct smu_context *smu, uint32_t speed);
+	int (*set_fan_speed_rpm)(struct smu_context *smu, uint32_t speed);
+	int (*set_xgmi_pstate)(struct smu_context *smu, uint32_t pstate);
+
+};
+
+#define smu_init_microcode(smu) \
+	((smu)->funcs->init_microcode ? (smu)->funcs->init_microcode((smu)) : 0)
+#define smu_init_smc_tables(smu) \
+	((smu)->funcs->init_smc_tables ? (smu)->funcs->init_smc_tables((smu)) : 0)
+#define smu_fini_smc_tables(smu) \
+	((smu)->funcs->fini_smc_tables ? (smu)->funcs->fini_smc_tables((smu)) : 0)
+#define smu_init_power(smu) \
+	((smu)->funcs->init_power ? (smu)->funcs->init_power((smu)) : 0)
+#define smu_fini_power(smu) \
+	((smu)->funcs->fini_power ? (smu)->funcs->fini_power((smu)) : 0)
+#define smu_load_microcode(smu) \
+	((smu)->funcs->load_microcode ? (smu)->funcs->load_microcode((smu)) : 0)
+#define smu_check_fw_status(smu) \
+	((smu)->funcs->check_fw_status ? (smu)->funcs->check_fw_status((smu)) : 0)
+#define smu_read_pptable_from_vbios(smu) \
+	((smu)->funcs->read_pptable_from_vbios ? (smu)->funcs->read_pptable_from_vbios((smu)) : 0)
+#define smu_get_vbios_bootup_values(smu) \
+	((smu)->funcs->get_vbios_bootup_values ? (smu)->funcs->get_vbios_bootup_values((smu)) : 0)
+#define smu_get_clk_info_from_vbios(smu) \
+	((smu)->funcs->get_clk_info_from_vbios ? (smu)->funcs->get_clk_info_from_vbios((smu)) : 0)
+#define smu_check_pptable(smu) \
+	((smu)->funcs->check_pptable ? (smu)->funcs->check_pptable((smu)) : 0)
+#define smu_parse_pptable(smu) \
+	((smu)->funcs->parse_pptable ? (smu)->funcs->parse_pptable((smu)) : 0)
+#define smu_populate_smc_pptable(smu) \
+	((smu)->funcs->populate_smc_pptable ? (smu)->funcs->populate_smc_pptable((smu)) : 0)
+#define smu_check_fw_version(smu) \
+	((smu)->funcs->check_fw_version ? (smu)->funcs->check_fw_version((smu)) : 0)
+#define smu_write_pptable(smu) \
+	((smu)->funcs->write_pptable ? (smu)->funcs->write_pptable((smu)) : 0)
+#define smu_set_min_dcef_deep_sleep(smu) \
+	((smu)->funcs->set_min_dcef_deep_sleep ? (smu)->funcs->set_min_dcef_deep_sleep((smu)) : 0)
+#define smu_set_tool_table_location(smu) \
+	((smu)->funcs->set_tool_table_location ? (smu)->funcs->set_tool_table_location((smu)) : 0)
+#define smu_notify_memory_pool_location(smu) \
+	((smu)->funcs->notify_memory_pool_location ? (smu)->funcs->notify_memory_pool_location((smu)) : 0)
+#define smu_write_watermarks_table(smu) \
+	((smu)->funcs->write_watermarks_table ? (smu)->funcs->write_watermarks_table((smu)) : 0)
+#define smu_set_last_dcef_min_deep_sleep_clk(smu) \
+	((smu)->funcs->set_last_dcef_min_deep_sleep_clk ? (smu)->funcs->set_last_dcef_min_deep_sleep_clk((smu)) : 0)
+#define smu_system_features_control(smu, en) \
+	((smu)->funcs->system_features_control ? (smu)->funcs->system_features_control((smu), (en)) : 0)
+#define smu_init_max_sustainable_clocks(smu) \
+	((smu)->funcs->init_max_sustainable_clocks ? (smu)->funcs->init_max_sustainable_clocks((smu)) : 0)
+#define smu_set_od8_default_settings(smu, initialize) \
+	((smu)->funcs->set_od8_default_settings ? (smu)->funcs->set_od8_default_settings((smu), (initialize)) : 0)
+#define smu_update_od8_settings(smu, index, value) \
+	((smu)->funcs->update_od8_settings ? (smu)->funcs->update_od8_settings((smu), (index), (value)) : 0)
+#define smu_get_current_rpm(smu, speed) \
+	((smu)->funcs->get_current_rpm ? (smu)->funcs->get_current_rpm((smu), (speed)) : 0)
+#define smu_set_fan_speed_rpm(smu, speed) \
+	((smu)->funcs->set_fan_speed_rpm ? (smu)->funcs->set_fan_speed_rpm((smu), (speed)) : 0)
+#define smu_send_smc_msg(smu, msg) \
+	((smu)->funcs->send_smc_msg? (smu)->funcs->send_smc_msg((smu), (msg)) : 0)
+#define smu_send_smc_msg_with_param(smu, msg, param) \
+	((smu)->funcs->send_smc_msg_with_param? (smu)->funcs->send_smc_msg_with_param((smu), (msg), (param)) : 0)
+#define smu_read_smc_arg(smu, arg) \
+	((smu)->funcs->read_smc_arg? (smu)->funcs->read_smc_arg((smu), (arg)) : 0)
+#define smu_alloc_dpm_context(smu) \
+	((smu)->ppt_funcs->alloc_dpm_context ? (smu)->ppt_funcs->alloc_dpm_context((smu)) : 0)
+#define smu_init_display(smu) \
+	((smu)->funcs->init_display ? (smu)->funcs->init_display((smu)) : 0)
+#define smu_feature_set_allowed_mask(smu) \
+	((smu)->funcs->set_allowed_mask? (smu)->funcs->set_allowed_mask((smu)) : 0)
+#define smu_feature_get_enabled_mask(smu, mask, num) \
+	((smu)->funcs->get_enabled_mask? (smu)->funcs->get_enabled_mask((smu), (mask), (num)) : 0)
+#define smu_is_dpm_running(smu) \
+	((smu)->funcs->is_dpm_running ? (smu)->funcs->is_dpm_running((smu)) : 0)
+#define smu_feature_update_enable_state(smu, feature_id, enabled) \
+	((smu)->funcs->update_feature_enable_state? (smu)->funcs->update_feature_enable_state((smu), (feature_id), (enabled)) : 0)
+#define smu_notify_display_change(smu) \
+	((smu)->funcs->notify_display_change? (smu)->funcs->notify_display_change((smu)) : 0)
+#define smu_store_powerplay_table(smu) \
+	((smu)->ppt_funcs->store_powerplay_table ? (smu)->ppt_funcs->store_powerplay_table((smu)) : 0)
+#define smu_check_powerplay_table(smu) \
+	((smu)->ppt_funcs->check_powerplay_table ? (smu)->ppt_funcs->check_powerplay_table((smu)) : 0)
+#define smu_append_powerplay_table(smu) \
+	((smu)->ppt_funcs->append_powerplay_table ? (smu)->ppt_funcs->append_powerplay_table((smu)) : 0)
+#define smu_set_default_dpm_table(smu) \
+	((smu)->ppt_funcs->set_default_dpm_table ? (smu)->ppt_funcs->set_default_dpm_table((smu)) : 0)
+#define smu_populate_umd_state_clk(smu) \
+	((smu)->ppt_funcs->populate_umd_state_clk ? (smu)->ppt_funcs->populate_umd_state_clk((smu)) : 0)
+#define smu_set_default_od8_settings(smu) \
+	((smu)->ppt_funcs->set_default_od8_settings ? (smu)->ppt_funcs->set_default_od8_settings((smu)) : 0)
+#define smu_update_specified_od8_value(smu, index, value) \
+	((smu)->ppt_funcs->update_specified_od8_value ? (smu)->ppt_funcs->update_specified_od8_value((smu), (index), (value)) : 0)
+#define smu_get_power_limit(smu, limit, def) \
+	((smu)->funcs->get_power_limit ? (smu)->funcs->get_power_limit((smu), (limit), (def)) : 0)
+#define smu_set_power_limit(smu, limit) \
+	((smu)->funcs->set_power_limit ? (smu)->funcs->set_power_limit((smu), (limit)) : 0)
+#define smu_get_current_clk_freq(smu, clk_id, value) \
+	((smu)->funcs->get_current_clk_freq? (smu)->funcs->get_current_clk_freq((smu), (clk_id), (value)) : 0)
+#define smu_print_clk_levels(smu, type, buf) \
+	((smu)->ppt_funcs->print_clk_levels ? (smu)->ppt_funcs->print_clk_levels((smu), (type), (buf)) : 0)
+#define smu_force_clk_levels(smu, type, level) \
+	((smu)->ppt_funcs->force_clk_levels ? (smu)->ppt_funcs->force_clk_levels((smu), (type), (level)) : 0)
+#define smu_get_od_percentage(smu, type) \
+	((smu)->ppt_funcs->get_od_percentage ? (smu)->ppt_funcs->get_od_percentage((smu), (type)) : 0)
+#define smu_set_od_percentage(smu, type, value) \
+	((smu)->ppt_funcs->set_od_percentage ? (smu)->ppt_funcs->set_od_percentage((smu), (type), (value)) : 0)
+#define smu_od_edit_dpm_table(smu, type, input, size) \
+	((smu)->ppt_funcs->od_edit_dpm_table ? (smu)->ppt_funcs->od_edit_dpm_table((smu), (type), (input), (size)) : 0)
+#define smu_start_thermal_control(smu) \
+	((smu)->funcs->start_thermal_control? (smu)->funcs->start_thermal_control((smu)) : 0)
+#define smu_read_sensor(smu, sensor, data, size) \
+	((smu)->funcs->read_sensor? (smu)->funcs->read_sensor((smu), (sensor), (data), (size)) : 0)
+#define smu_get_power_profile_mode(smu, buf) \
+	((smu)->funcs->get_power_profile_mode ? (smu)->funcs->get_power_profile_mode((smu), buf) : 0)
+#define smu_set_power_profile_mode(smu, param, param_size) \
+	((smu)->funcs->set_power_profile_mode ? (smu)->funcs->set_power_profile_mode((smu), (param), (param_size)) : 0)
+#define smu_get_performance_level(smu) \
+	((smu)->ppt_funcs->get_performance_level ? (smu)->ppt_funcs->get_performance_level((smu)) : 0)
+#define smu_force_performance_level(smu, level) \
+	((smu)->ppt_funcs->force_performance_level ? (smu)->ppt_funcs->force_performance_level((smu), (level)) : 0)
+#define smu_pre_display_config_changed(smu) \
+	((smu)->ppt_funcs->pre_display_config_changed ? (smu)->ppt_funcs->pre_display_config_changed((smu)) : 0)
+#define smu_display_config_changed(smu) \
+	((smu)->ppt_funcs->display_config_changed ? (smu)->ppt_funcs->display_config_changed((smu)) : 0)
+#define smu_apply_clocks_adjust_rules(smu) \
+	((smu)->ppt_funcs->apply_clocks_adjust_rules ? (smu)->ppt_funcs->apply_clocks_adjust_rules((smu)) : 0)
+#define smu_notify_smc_dispaly_config(smu) \
+	((smu)->ppt_funcs->notify_smc_dispaly_config ? (smu)->ppt_funcs->notify_smc_dispaly_config((smu)) : 0)
+#define smu_force_dpm_limit_value(smu, highest) \
+	((smu)->ppt_funcs->force_dpm_limit_value ? (smu)->ppt_funcs->force_dpm_limit_value((smu), (highest)) : 0)
+#define smu_unforce_dpm_levels(smu) \
+	((smu)->ppt_funcs->unforce_dpm_levels ? (smu)->ppt_funcs->unforce_dpm_levels((smu)) : 0)
+#define smu_upload_dpm_level(smu, max, feature_mask) \
+	((smu)->ppt_funcs->upload_dpm_level ? (smu)->ppt_funcs->upload_dpm_level((smu), (max), (feature_mask)) : 0)
+#define smu_get_profiling_clk_mask(smu, level, sclk_mask, mclk_mask, soc_mask) \
+	((smu)->ppt_funcs->get_profiling_clk_mask ? (smu)->ppt_funcs->get_profiling_clk_mask((smu), (level), (sclk_mask), (mclk_mask), (soc_mask)) : 0)
+#define smu_set_cpu_power_state(smu) \
+	((smu)->ppt_funcs->set_cpu_power_state ? (smu)->ppt_funcs->set_cpu_power_state((smu)) : 0)
+#define smu_get_fan_control_mode(smu) \
+	((smu)->funcs->get_fan_control_mode ? (smu)->funcs->get_fan_control_mode((smu)) : 0)
+#define smu_set_fan_control_mode(smu, value) \
+	((smu)->funcs->set_fan_control_mode ? (smu)->funcs->set_fan_control_mode((smu), (value)) : 0)
+#define smu_get_fan_speed_percent(smu, speed) \
+	((smu)->funcs->get_fan_speed_percent ? (smu)->funcs->get_fan_speed_percent((smu), (speed)) : 0)
+#define smu_set_fan_speed_percent(smu, speed) \
+	((smu)->funcs->set_fan_speed_percent ? (smu)->funcs->set_fan_speed_percent((smu), (speed)) : 0)
+
+#define smu_msg_get_index(smu, msg) \
+	((smu)->ppt_funcs? ((smu)->ppt_funcs->get_smu_msg_index? (smu)->ppt_funcs->get_smu_msg_index((smu), (msg)) : -EINVAL) : -EINVAL)
+#define smu_run_afll_btc(smu) \
+	((smu)->ppt_funcs? ((smu)->ppt_funcs->run_afll_btc? (smu)->ppt_funcs->run_afll_btc((smu)) : 0) : 0)
+#define smu_get_unallowed_feature_mask(smu, feature_mask, num) \
+	((smu)->ppt_funcs? ((smu)->ppt_funcs->get_unallowed_feature_mask? (smu)->ppt_funcs->get_unallowed_feature_mask((smu), (feature_mask), (num)) : 0) : 0)
+#define smu_set_deep_sleep_dcefclk(smu, clk) \
+	((smu)->funcs->set_deep_sleep_dcefclk ? (smu)->funcs->set_deep_sleep_dcefclk((smu), (clk)) : 0)
+#define smu_set_active_display_count(smu, count) \
+	((smu)->funcs->set_active_display_count ? (smu)->funcs->set_active_display_count((smu), (count)) : 0)
+#define smu_store_cc6_data(smu, st, cc6_dis, pst_dis, pst_sw_dis) \
+	((smu)->funcs->store_cc6_data ? (smu)->funcs->store_cc6_data((smu), (st), (cc6_dis), (pst_dis), (pst_sw_dis)) : 0)
+#define smu_get_clock_by_type(smu, type, clocks) \
+	((smu)->funcs->get_clock_by_type ? (smu)->funcs->get_clock_by_type((smu), (type), (clocks)) : 0)
+#define smu_get_max_high_clocks(smu, clocks) \
+	((smu)->funcs->get_max_high_clocks ? (smu)->funcs->get_max_high_clocks((smu), (clocks)) : 0)
+#define smu_get_clock_by_type_with_latency(smu, type, clocks) \
+	((smu)->ppt_funcs->get_clock_by_type_with_latency ? (smu)->ppt_funcs->get_clock_by_type_with_latency((smu), (type), (clocks)) : 0)
+#define smu_get_clock_by_type_with_voltage(smu, type, clocks) \
+	((smu)->ppt_funcs->get_clock_by_type_with_voltage ? (smu)->ppt_funcs->get_clock_by_type_with_voltage((smu), (type), (clocks)) : 0)
+#define smu_display_clock_voltage_request(smu, clock_req) \
+	((smu)->funcs->display_clock_voltage_request ? (smu)->funcs->display_clock_voltage_request((smu), (clock_req)) : 0)
+#define smu_get_dal_power_level(smu, clocks) \
+	((smu)->funcs->get_dal_power_level ? (smu)->funcs->get_dal_power_level((smu), (clocks)) : 0)
+#define smu_get_perf_level(smu, designation, level) \
+	((smu)->funcs->get_perf_level ? (smu)->funcs->get_perf_level((smu), (designation), (level)) : 0)
+#define smu_get_current_shallow_sleep_clocks(smu, clocks) \
+	((smu)->funcs->get_current_shallow_sleep_clocks ? (smu)->funcs->get_current_shallow_sleep_clocks((smu), (clocks)) : 0)
+#define smu_notify_smu_enable_pwe(smu) \
+	((smu)->funcs->notify_smu_enable_pwe ? (smu)->funcs->notify_smu_enable_pwe((smu)) : 0)
+#define smu_set_watermarks_for_clock_ranges(smu, clock_ranges) \
+	((smu)->funcs->set_watermarks_for_clock_ranges ? (smu)->funcs->set_watermarks_for_clock_ranges((smu), (clock_ranges)) : 0)
+#define smu_dpm_set_uvd_enable(smu, enable) \
+	((smu)->funcs->dpm_set_uvd_enable ? (smu)->funcs->dpm_set_uvd_enable((smu), (enable)) : 0)
+#define smu_dpm_set_vce_enable(smu, enable) \
+	((smu)->funcs->dpm_set_vce_enable ? (smu)->funcs->dpm_set_vce_enable((smu), (enable)) : 0)
+#define smu_get_sclk(smu, low) \
+	((smu)->funcs->get_sclk ? (smu)->funcs->get_sclk((smu), (low)) : 0)
+#define smu_get_mclk(smu, low) \
+	((smu)->funcs->get_mclk ? (smu)->funcs->get_mclk((smu), (low)) : 0)
+#define smu_set_xgmi_pstate(smu, pstate) \
+		((smu)->funcs->set_xgmi_pstate ? (smu)->funcs->set_xgmi_pstate((smu), (pstate)) : 0)
+
+
+extern int smu_get_atom_data_table(struct smu_context *smu, uint32_t table,
+				   uint16_t *size, uint8_t *frev, uint8_t *crev,
+				   uint8_t **addr);
+
+extern const struct amd_ip_funcs smu_ip_funcs;
+
+extern const struct amdgpu_ip_block_version smu_v11_0_ip_block;
+extern int smu_feature_init_dpm(struct smu_context *smu);
+
+extern int smu_feature_is_enabled(struct smu_context *smu, int feature_id);
+extern int smu_feature_set_enabled(struct smu_context *smu, int feature_id, bool enable);
+extern int smu_feature_is_supported(struct smu_context *smu, int feature_id);
+extern int smu_feature_set_supported(struct smu_context *smu, int feature_id, bool enable);
+
+int smu_update_table_with_arg(struct smu_context *smu, uint16_t table_id, uint16_t exarg,
+		     void *table_data, bool drv2smu);
+#define smu_update_table(smu, table_id, table_data, drv2smu) \
+	smu_update_table_with_arg((smu), (table_id), 0, (table_data), (drv2smu))
+
+bool is_support_sw_smu(struct amdgpu_device *adev);
+int smu_reset(struct smu_context *smu);
+int smu_common_read_sensor(struct smu_context *smu, enum amd_pp_sensors sensor,
+			   void *data, uint32_t *size);
+int smu_sys_get_pp_table(struct smu_context *smu, void **table);
+int smu_sys_set_pp_table(struct smu_context *smu,  void *buf, size_t size);
+int smu_get_power_num_states(struct smu_context *smu, struct pp_states_info *state_info);
+enum amd_pm_state_type smu_get_current_power_state(struct smu_context *smu);
+
+/* smu to display interface */
+extern int smu_display_configuration_change(struct smu_context *smu, const
+					    struct amd_pp_display_configuration
+					    *display_config);
+extern int smu_get_current_clocks(struct smu_context *smu,
+				  struct amd_pp_clock_info *clocks);
+extern int smu_dpm_set_power_gate(struct smu_context *smu,uint32_t block_type, bool gate);
+extern int smu_handle_task(struct smu_context *smu,
+			   enum amd_dpm_forced_level level,
+			   enum amd_pp_task task_id);
+#endif
diff --git a/drivers/gpu/drm/amd/powerplay/inc/rv_ppsmc.h b/drivers/gpu/drm/amd/powerplay/inc/rv_ppsmc.h
index a2991fa2e6f8..90879e4092a3 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/rv_ppsmc.h
+++ b/drivers/gpu/drm/amd/powerplay/inc/rv_ppsmc.h
@@ -85,7 +85,6 @@
 #define PPSMC_MSG_SetRccPfcPmeRestoreRegister   0x36
 #define PPSMC_Message_Count                     0x37
 
-
 typedef uint16_t PPSMC_Result;
 typedef int      PPSMC_Msg;
 
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu10.h b/drivers/gpu/drm/amd/powerplay/inc/smu10.h
index 9e837a5014c5..b96520528240 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu10.h
+++ b/drivers/gpu/drm/amd/powerplay/inc/smu10.h
@@ -136,12 +136,14 @@
 #define FEATURE_CORE_CSTATES_MASK     (1 << FEATURE_CORE_CSTATES_BIT)
 
 /* Workload bits */
-#define WORKLOAD_PPLIB_FULL_SCREEN_3D_BIT 0
-#define WORKLOAD_PPLIB_VIDEO_BIT          2
-#define WORKLOAD_PPLIB_VR_BIT             3
-#define WORKLOAD_PPLIB_COMPUTE_BIT        4
-#define WORKLOAD_PPLIB_CUSTOM_BIT         5
-#define WORKLOAD_PPLIB_COUNT              6
+#define WORKLOAD_DEFAULT_BIT              0
+#define WORKLOAD_PPLIB_FULL_SCREEN_3D_BIT 1
+#define WORKLOAD_PPLIB_POWER_SAVING_BIT   2
+#define WORKLOAD_PPLIB_VIDEO_BIT          3
+#define WORKLOAD_PPLIB_VR_BIT             4
+#define WORKLOAD_PPLIB_COMPUTE_BIT        5
+#define WORKLOAD_PPLIB_CUSTOM_BIT         6
+#define WORKLOAD_PPLIB_COUNT              7
 
 typedef struct {
 	/* MP1_EXT_SCRATCH0 */
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h b/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h
new file mode 100644
index 000000000000..aa8d81f4111e
--- /dev/null
+++ b/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h
@@ -0,0 +1,89 @@
+/*
+ * Copyright 2019 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#ifndef __SMU_V11_0_H__
+#define __SMU_V11_0_H__
+
+#include "amdgpu_smu.h"
+
+/* MP Apertures */
+#define MP0_Public			0x03800000
+#define MP0_SRAM			0x03900000
+#define MP1_Public			0x03b00000
+#define MP1_SRAM			0x03c00004
+
+/* address block */
+#define smnMP1_FIRMWARE_FLAGS		0x3010024
+#define smnMP0_FW_INTF			0x30101c0
+#define smnMP1_PUB_CTRL			0x3010b14
+
+struct smu_11_0_max_sustainable_clocks {
+	uint32_t display_clock;
+	uint32_t phy_clock;
+	uint32_t pixel_clock;
+	uint32_t uclock;
+	uint32_t dcef_clock;
+	uint32_t soc_clock;
+};
+
+struct smu_11_0_dpm_table {
+	uint32_t    min;        /* MHz */
+	uint32_t    max;        /* MHz */
+};
+
+struct smu_11_0_dpm_tables {
+	struct smu_11_0_dpm_table        soc_table;
+	struct smu_11_0_dpm_table        gfx_table;
+	struct smu_11_0_dpm_table        uclk_table;
+	struct smu_11_0_dpm_table        eclk_table;
+	struct smu_11_0_dpm_table        vclk_table;
+	struct smu_11_0_dpm_table        dclk_table;
+	struct smu_11_0_dpm_table        dcef_table;
+	struct smu_11_0_dpm_table        pixel_table;
+	struct smu_11_0_dpm_table        display_table;
+	struct smu_11_0_dpm_table        phy_table;
+	struct smu_11_0_dpm_table        fclk_table;
+};
+
+struct smu_11_0_dpm_context {
+	struct smu_11_0_dpm_tables  dpm_tables;
+	uint32_t                    workload_policy_mask;
+	uint32_t                    dcef_min_ds_clk;
+};
+
+enum smu_11_0_power_state {
+	SMU_11_0_POWER_STATE__D0 = 0,
+	SMU_11_0_POWER_STATE__D1,
+	SMU_11_0_POWER_STATE__D3, /* Sleep*/
+	SMU_11_0_POWER_STATE__D4, /* Hibernate*/
+	SMU_11_0_POWER_STATE__D5, /* Power off*/
+};
+
+struct smu_11_0_power_context {
+	uint32_t	power_source;
+	uint8_t		in_power_limit_boost_mode;
+	enum smu_11_0_power_state power_state;
+};
+
+void smu_v11_0_set_smu_funcs(struct smu_context *smu);
+
+#endif
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0_ppsmc.h b/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0_ppsmc.h
new file mode 100644
index 000000000000..f466f624ad32
--- /dev/null
+++ b/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0_ppsmc.h
@@ -0,0 +1,128 @@
+/*
+ * Copyright 2018 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#ifndef SMU_V11_0_PPSMC_H
+#define SMU_V11_0_PPSMC_H
+
+// SMU Response Codes:
+#define PPSMC_Result_OK                    0x1
+#define PPSMC_Result_Failed                0xFF
+#define PPSMC_Result_UnknownCmd            0xFE
+#define PPSMC_Result_CmdRejectedPrereq     0xFD
+#define PPSMC_Result_CmdRejectedBusy       0xFC
+
+// Message Definitions:
+// BASIC
+#define PPSMC_MSG_TestMessage                    0x1
+#define PPSMC_MSG_GetSmuVersion                  0x2
+#define PPSMC_MSG_GetDriverIfVersion             0x3
+#define PPSMC_MSG_SetAllowedFeaturesMaskLow      0x4
+#define PPSMC_MSG_SetAllowedFeaturesMaskHigh     0x5
+#define PPSMC_MSG_EnableAllSmuFeatures           0x6
+#define PPSMC_MSG_DisableAllSmuFeatures          0x7
+#define PPSMC_MSG_EnableSmuFeaturesLow           0x8
+#define PPSMC_MSG_EnableSmuFeaturesHigh          0x9
+#define PPSMC_MSG_DisableSmuFeaturesLow          0xA
+#define PPSMC_MSG_DisableSmuFeaturesHigh         0xB
+#define PPSMC_MSG_GetEnabledSmuFeaturesLow       0xC
+#define PPSMC_MSG_GetEnabledSmuFeaturesHigh      0xD
+#define PPSMC_MSG_SetDriverDramAddrHigh          0xE
+#define PPSMC_MSG_SetDriverDramAddrLow           0xF
+#define PPSMC_MSG_SetToolsDramAddrHigh           0x10
+#define PPSMC_MSG_SetToolsDramAddrLow            0x11
+#define PPSMC_MSG_TransferTableSmu2Dram          0x12
+#define PPSMC_MSG_TransferTableDram2Smu          0x13
+#define PPSMC_MSG_UseDefaultPPTable              0x14
+#define PPSMC_MSG_UseBackupPPTable               0x15
+#define PPSMC_MSG_SetSystemVirtualDramAddrHigh   0x16
+#define PPSMC_MSG_SetSystemVirtualDramAddrLow    0x17
+
+//BACO/BAMACO/BOMACO
+#define PPSMC_MSG_EnterBaco                      0x18
+#define PPSMC_MSG_ExitBaco                       0x19
+
+//DPM
+#define PPSMC_MSG_SetSoftMinByFreq               0x1A
+#define PPSMC_MSG_SetSoftMaxByFreq               0x1B
+#define PPSMC_MSG_SetHardMinByFreq               0x1C
+#define PPSMC_MSG_SetHardMaxByFreq               0x1D 
+#define PPSMC_MSG_GetMinDpmFreq                  0x1E
+#define PPSMC_MSG_GetMaxDpmFreq                  0x1F
+#define PPSMC_MSG_GetDpmFreqByIndex              0x20
+#define PPSMC_MSG_OverridePcieParameters         0x21
+#define PPSMC_MSG_SetMinDeepSleepDcefclk         0x22
+#define PPSMC_MSG_SetWorkloadMask                0x23 
+#define PPSMC_MSG_SetUclkFastSwitch              0x24
+#define PPSMC_MSG_GetAvfsVoltageByDpm            0x25
+#define PPSMC_MSG_SetVideoFps                    0x26
+#define PPSMC_MSG_GetDcModeMaxDpmFreq            0x27
+
+//Power Gating
+#define PPSMC_MSG_AllowGfxOff                    0x28
+#define PPSMC_MSG_DisallowGfxOff                 0x29
+#define PPSMC_MSG_PowerUpVcn					 0x2A
+#define PPSMC_MSG_PowerDownVcn					 0x2B	
+#define PPSMC_MSG_PowerUpJpeg                    0x2C
+#define PPSMC_MSG_PowerDownJpeg					 0x2D
+//reserve 0x2A to 0x2F for PG harvesting TBD
+
+//I2C Interface
+#define PPSMC_RequestI2cTransaction              0x30
+
+//Resets
+#define PPSMC_MSG_SoftReset                      0x31  //FIXME Need confirmation from driver
+#define PPSMC_MSG_PrepareMp1ForUnload            0x32
+#define PPSMC_MSG_PrepareMp1ForReset             0x33
+#define PPSMC_MSG_PrepareMp1ForShutdown          0x34
+
+//ACDC Power Source
+#define PPSMC_MSG_SetPptLimit                    0x35
+#define PPSMC_MSG_GetPptLimit                    0x36
+#define PPSMC_MSG_ReenableAcDcInterrupt          0x37
+#define PPSMC_MSG_NotifyPowerSource              0x38
+//#define PPSMC_MSG_GfxDeviceDriverReset           0x39 //FIXME mode1 and 2 resets will go directly go PSP
+
+//BTC
+#define PPSMC_MSG_RunBtc                         0x3A
+
+//Debug
+#define PPSMC_MSG_DramLogSetDramAddrHigh         0x3B
+#define PPSMC_MSG_DramLogSetDramAddrLow          0x3C
+#define PPSMC_MSG_DramLogSetDramSize             0x3D
+#define PPSMC_MSG_GetDebugData                   0x3E
+
+//Others
+#define PPSMC_MSG_ConfigureGfxDidt               0x3F
+#define PPSMC_MSG_NumOfDisplays                  0x40
+
+#define PPSMC_MSG_SetMemoryChannelConfig         0x41 
+#define PPSMC_MSG_SetGeminiMode                  0x42
+#define PPSMC_MSG_SetGeminiApertureHigh          0x43
+#define PPSMC_MSG_SetGeminiApertureLow           0x44
+
+#define PPSMC_Message_Count                      0x45
+
+typedef uint32_t PPSMC_Result;
+typedef uint32_t PPSMC_Msg;
+
+#endif
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0_pptable.h b/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0_pptable.h
new file mode 100644
index 000000000000..92c65b80bde2
--- /dev/null
+++ b/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0_pptable.h
@@ -0,0 +1,147 @@
+/*
+ * Copyright 2018 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#ifndef SMU_11_0_PPTABLE_H
+#define SMU_11_0_PPTABLE_H
+
+
+#define SMU_11_0_TABLE_FORMAT_REVISION                  12
+
+//// POWERPLAYTABLE::ulPlatformCaps
+#define SMU_11_0_PP_PLATFORM_CAP_POWERPLAY              0x1
+#define SMU_11_0_PP_PLATFORM_CAP_SBIOSPOWERSOURCE       0x2
+#define SMU_11_0_PP_PLATFORM_CAP_HARDWAREDC             0x4
+#define SMU_11_0_PP_PLATFORM_CAP_BACO                   0x8
+#define SMU_11_0_PP_PLATFORM_CAP_MACO                   0x10
+#define SMU_11_0_PP_PLATFORM_CAP_SHADOWPSTATE           0x20
+
+// SMU_11_0_PP_THERMALCONTROLLER - Thermal Controller Type
+#define SMU_11_0_PP_THERMALCONTROLLER_NONE              0
+
+#define SMU_11_0_PP_OVERDRIVE_VERSION                   0x0800
+#define SMU_11_0_PP_POWERSAVINGCLOCK_VERSION            0x0100
+
+enum SMU_11_0_ODFEATURE_ID {
+    SMU_11_0_ODFEATURE_GFXCLK_LIMITS        = 1 << 0,         //GFXCLK Limit feature
+    SMU_11_0_ODFEATURE_GFXCLK_CURVE         = 1 << 1,         //GFXCLK Curve feature
+    SMU_11_0_ODFEATURE_UCLK_MAX             = 1 << 2,         //UCLK Limit feature
+    SMU_11_0_ODFEATURE_POWER_LIMIT          = 1 << 3,         //Power Limit feature
+    SMU_11_0_ODFEATURE_FAN_ACOUSTIC_LIMIT   = 1 << 4,         //Fan Acoustic RPM feature
+    SMU_11_0_ODFEATURE_FAN_SPEED_MIN        = 1 << 5,         //Minimum Fan Speed feature
+    SMU_11_0_ODFEATURE_TEMPERATURE_FAN      = 1 << 6,         //Fan Target Temperature Limit feature
+    SMU_11_0_ODFEATURE_TEMPERATURE_SYSTEM   = 1 << 7,         //Operating Temperature Limit feature
+    SMU_11_0_ODFEATURE_MEMORY_TIMING_TUNE   = 1 << 8,         //AC Timing Tuning feature
+    SMU_11_0_ODFEATURE_FAN_ZERO_RPM_CONTROL = 1 << 9,         //Zero RPM feature
+    SMU_11_0_ODFEATURE_AUTO_UV_ENGINE       = 1 << 10,        //Auto Under Volt GFXCLK feature
+    SMU_11_0_ODFEATURE_AUTO_OC_ENGINE       = 1 << 11,        //Auto Over Clock GFXCLK feature
+    SMU_11_0_ODFEATURE_AUTO_OC_MEMORY       = 1 << 12,        //Auto Over Clock MCLK feature
+    SMU_11_0_ODFEATURE_FAN_CURVE            = 1 << 13,        //VICTOR TODO
+    SMU_11_0_ODFEATURE_COUNT                = 14,
+};
+#define SMU_11_0_MAX_ODFEATURE    32          //Maximum Number of OD Features
+
+enum SMU_11_0_ODSETTING_ID {
+    SMU_11_0_ODSETTING_GFXCLKFMAX = 0,
+    SMU_11_0_ODSETTING_GFXCLKFMIN,
+    SMU_11_0_ODSETTING_VDDGFXCURVEFREQ_P1,
+    SMU_11_0_ODSETTING_VDDGFXCURVEVOLTAGE_P1,
+    SMU_11_0_ODSETTING_VDDGFXCURVEFREQ_P2,
+    SMU_11_0_ODSETTING_VDDGFXCURVEVOLTAGE_P2,
+    SMU_11_0_ODSETTING_VDDGFXCURVEFREQ_P3,
+    SMU_11_0_ODSETTING_VDDGFXCURVEVOLTAGE_P3,
+    SMU_11_0_ODSETTING_UCLKFMAX,
+    SMU_11_0_ODSETTING_POWERPERCENTAGE,
+    SMU_11_0_ODSETTING_FANRPMMIN,
+    SMU_11_0_ODSETTING_FANRPMACOUSTICLIMIT,
+    SMU_11_0_ODSETTING_FANTARGETTEMPERATURE,
+    SMU_11_0_ODSETTING_OPERATINGTEMPMAX,
+    SMU_11_0_ODSETTING_ACTIMING,
+    SMU_11_0_ODSETTING_FAN_ZERO_RPM_CONTROL,
+    SMU_11_0_ODSETTING_AUTOUVENGINE,
+    SMU_11_0_ODSETTING_AUTOOCENGINE,
+    SMU_11_0_ODSETTING_AUTOOCMEMORY,
+    SMU_11_0_ODSETTING_COUNT,
+};
+#define SMU_11_0_MAX_ODSETTING    32          //Maximum Number of ODSettings
+
+struct smu_11_0_overdrive_table
+{
+    uint8_t  revision;                                        //Revision = SMU_11_0_PP_OVERDRIVE_VERSION
+    uint8_t  reserve[3];                                      //Zero filled field reserved for future use
+    uint32_t feature_count;                                   //Total number of supported features
+    uint32_t setting_count;                                   //Total number of supported settings
+    uint8_t  cap[SMU_11_0_MAX_ODFEATURE];                     //OD feature support flags
+    uint32_t max[SMU_11_0_MAX_ODSETTING];                     //default maximum settings
+    uint32_t min[SMU_11_0_MAX_ODSETTING];                     //default minimum settings
+} __attribute__((packed));
+
+enum SMU_11_0_PPCLOCK_ID {
+    SMU_11_0_PPCLOCK_GFXCLK = 0,
+    SMU_11_0_PPCLOCK_VCLK,
+    SMU_11_0_PPCLOCK_DCLK,
+    SMU_11_0_PPCLOCK_ECLK,
+    SMU_11_0_PPCLOCK_SOCCLK,
+    SMU_11_0_PPCLOCK_UCLK,
+    SMU_11_0_PPCLOCK_DCEFCLK,
+    SMU_11_0_PPCLOCK_DISPCLK,
+    SMU_11_0_PPCLOCK_PIXCLK,
+    SMU_11_0_PPCLOCK_PHYCLK,
+    SMU_11_0_PPCLOCK_COUNT,
+};
+#define SMU_11_0_MAX_PPCLOCK      16          //Maximum Number of PP Clocks
+
+struct smu_11_0_power_saving_clock_table
+{
+    uint8_t  revision;                                        //Revision = SMU_11_0_PP_POWERSAVINGCLOCK_VERSION
+    uint8_t  reserve[3];                                      //Zero filled field reserved for future use
+    uint32_t count;                                           //power_saving_clock_count = SMU_11_0_PPCLOCK_COUNT
+    uint32_t max[SMU_11_0_MAX_PPCLOCK];                       //PowerSavingClock Mode Clock Maximum array In MHz
+    uint32_t min[SMU_11_0_MAX_PPCLOCK];                       //PowerSavingClock Mode Clock Minimum array In MHz
+} __attribute__((packed));
+
+struct smu_11_0_powerplay_table
+{
+      struct atom_common_table_header header;
+      uint8_t  table_revision;
+      uint32_t table_size;                          //Driver portion table size. The offset to smc_pptable including header size
+      uint32_t golden_pp_id;
+      uint32_t golden_revision;
+      uint16_t format_id;
+      uint32_t platform_caps;                       //POWERPLAYABLE::ulPlatformCaps
+                                                    
+      uint8_t  thermal_controller_type;             //one of SMU_11_0_PP_THERMALCONTROLLER
+
+      uint16_t small_power_limit1;
+      uint16_t small_power_limit2;
+      uint16_t boost_power_limit;
+      uint16_t od_turbo_power_limit;                //Power limit setting for Turbo mode in Performance UI Tuning. 
+      uint16_t od_power_save_power_limit;           //Power limit setting for PowerSave/Optimal mode in Performance UI Tuning. 
+      uint16_t software_shutdown_temp;
+
+      uint16_t reserve[6];                          //Zero filled field reserved for future use
+
+      struct smu_11_0_power_saving_clock_table      power_saving_clock;
+      struct smu_11_0_overdrive_table               overdrive_table;
+
+      PPTable_t smc_pptable;                        //PPTable_t in smu11_driver_if.h
+} __attribute__((packed));
+
+#endif
diff --git a/drivers/gpu/drm/amd/powerplay/inc/vega20_ppsmc.h b/drivers/gpu/drm/amd/powerplay/inc/vega20_ppsmc.h
index 4f63a736ea0e..a0883038f3c3 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/vega20_ppsmc.h
+++ b/drivers/gpu/drm/amd/powerplay/inc/vega20_ppsmc.h
@@ -119,7 +119,8 @@
 #define PPSMC_MSG_PrepareMp1ForShutdown          0x5A
 #define PPSMC_MSG_SetMGpuFanBoostLimitRpm        0x5D
 #define PPSMC_MSG_GetAVFSVoltageByDpm            0x5F
-#define PPSMC_Message_Count                      0x60
+#define PPSMC_MSG_BacoWorkAroundFlushVDCI        0x60
+#define PPSMC_Message_Count                      0x61
 
 typedef uint32_t PPSMC_Result;
 typedef uint32_t PPSMC_Msg;
diff --git a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c
new file mode 100644
index 000000000000..92903a4cc4d8
--- /dev/null
+++ b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c
@@ -0,0 +1,1977 @@
+/*
+ * Copyright 2019 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#include "pp_debug.h"
+#include <linux/firmware.h>
+#include "amdgpu.h"
+#include "amdgpu_smu.h"
+#include "atomfirmware.h"
+#include "amdgpu_atomfirmware.h"
+#include "smu_v11_0.h"
+#include "smu11_driver_if.h"
+#include "soc15_common.h"
+#include "atom.h"
+#include "vega20_ppt.h"
+#include "pp_thermal.h"
+
+#include "asic_reg/thm/thm_11_0_2_offset.h"
+#include "asic_reg/thm/thm_11_0_2_sh_mask.h"
+#include "asic_reg/mp/mp_9_0_offset.h"
+#include "asic_reg/mp/mp_9_0_sh_mask.h"
+#include "asic_reg/nbio/nbio_7_4_offset.h"
+#include "asic_reg/smuio/smuio_9_0_offset.h"
+#include "asic_reg/smuio/smuio_9_0_sh_mask.h"
+
+MODULE_FIRMWARE("amdgpu/vega20_smc.bin");
+
+#define SMU11_TOOL_SIZE		0x19000
+#define SMU11_THERMAL_MINIMUM_ALERT_TEMP      0
+#define SMU11_THERMAL_MAXIMUM_ALERT_TEMP      255
+
+#define SMU11_TEMPERATURE_UNITS_PER_CENTIGRADES 1000
+#define SMU11_VOLTAGE_SCALE 4
+
+#define SMC_DPM_FEATURE (FEATURE_DPM_PREFETCHER_MASK | \
+			 FEATURE_DPM_GFXCLK_MASK | \
+			 FEATURE_DPM_UCLK_MASK | \
+			 FEATURE_DPM_SOCCLK_MASK | \
+			 FEATURE_DPM_UVD_MASK | \
+			 FEATURE_DPM_VCE_MASK | \
+			 FEATURE_DPM_MP0CLK_MASK | \
+			 FEATURE_DPM_LINK_MASK | \
+			 FEATURE_DPM_DCEFCLK_MASK)
+
+static int smu_v11_0_send_msg_without_waiting(struct smu_context *smu,
+					      uint16_t msg)
+{
+	struct amdgpu_device *adev = smu->adev;
+	WREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_66, msg);
+	return 0;
+}
+
+static int smu_v11_0_read_arg(struct smu_context *smu, uint32_t *arg)
+{
+	struct amdgpu_device *adev = smu->adev;
+
+	*arg = RREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_82);
+	return 0;
+}
+
+static int smu_v11_0_wait_for_response(struct smu_context *smu)
+{
+	struct amdgpu_device *adev = smu->adev;
+	uint32_t cur_value, i;
+
+	for (i = 0; i < adev->usec_timeout; i++) {
+		cur_value = RREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_90);
+		if ((cur_value & MP1_C2PMSG_90__CONTENT_MASK) != 0)
+			break;
+		udelay(1);
+	}
+
+	/* timeout means wrong logic */
+	if (i == adev->usec_timeout)
+		return -ETIME;
+
+	return RREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_90) == 0x1 ? 0 : -EIO;
+}
+
+static int smu_v11_0_send_msg(struct smu_context *smu, uint16_t msg)
+{
+	struct amdgpu_device *adev = smu->adev;
+	int ret = 0, index = 0;
+
+	index = smu_msg_get_index(smu, msg);
+	if (index < 0)
+		return index;
+
+	smu_v11_0_wait_for_response(smu);
+
+	WREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_90, 0);
+
+	smu_v11_0_send_msg_without_waiting(smu, (uint16_t)index);
+
+	ret = smu_v11_0_wait_for_response(smu);
+
+	if (ret)
+		pr_err("Failed to send message 0x%x, response 0x%x\n", index,
+		       ret);
+
+	return ret;
+
+}
+
+static int
+smu_v11_0_send_msg_with_param(struct smu_context *smu, uint16_t msg,
+			      uint32_t param)
+{
+
+	struct amdgpu_device *adev = smu->adev;
+	int ret = 0, index = 0;
+
+	index = smu_msg_get_index(smu, msg);
+	if (index < 0)
+		return index;
+
+	ret = smu_v11_0_wait_for_response(smu);
+	if (ret)
+		pr_err("Failed to send message 0x%x, response 0x%x, param 0x%x\n",
+		       index, ret, param);
+
+	WREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_90, 0);
+
+	WREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_82, param);
+
+	smu_v11_0_send_msg_without_waiting(smu, (uint16_t)index);
+
+	ret = smu_v11_0_wait_for_response(smu);
+	if (ret)
+		pr_err("Failed to send message 0x%x, response 0x%x param 0x%x\n",
+		       index, ret, param);
+
+	return ret;
+}
+
+static int smu_v11_0_init_microcode(struct smu_context *smu)
+{
+	struct amdgpu_device *adev = smu->adev;
+	const char *chip_name;
+	char fw_name[30];
+	int err = 0;
+	const struct smc_firmware_header_v1_0 *hdr;
+	const struct common_firmware_header *header;
+	struct amdgpu_firmware_info *ucode = NULL;
+
+	switch (adev->asic_type) {
+	case CHIP_VEGA20:
+		chip_name = "vega20";
+		break;
+	default:
+		BUG();
+	}
+
+	snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_smc.bin", chip_name);
+
+	err = request_firmware(&adev->pm.fw, fw_name, adev->dev);
+	if (err)
+		goto out;
+	err = amdgpu_ucode_validate(adev->pm.fw);
+	if (err)
+		goto out;
+
+	hdr = (const struct smc_firmware_header_v1_0 *) adev->pm.fw->data;
+	amdgpu_ucode_print_smc_hdr(&hdr->header);
+	adev->pm.fw_version = le32_to_cpu(hdr->header.ucode_version);
+
+	if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) {
+		ucode = &adev->firmware.ucode[AMDGPU_UCODE_ID_SMC];
+		ucode->ucode_id = AMDGPU_UCODE_ID_SMC;
+		ucode->fw = adev->pm.fw;
+		header = (const struct common_firmware_header *)ucode->fw->data;
+		adev->firmware.fw_size +=
+			ALIGN(le32_to_cpu(header->ucode_size_bytes), PAGE_SIZE);
+	}
+
+out:
+	if (err) {
+		DRM_ERROR("smu_v11_0: Failed to load firmware \"%s\"\n",
+			  fw_name);
+		release_firmware(adev->pm.fw);
+		adev->pm.fw = NULL;
+	}
+	return err;
+}
+
+static int smu_v11_0_load_microcode(struct smu_context *smu)
+{
+	return 0;
+}
+
+static int smu_v11_0_check_fw_status(struct smu_context *smu)
+{
+	struct amdgpu_device *adev = smu->adev;
+	uint32_t mp1_fw_flags;
+
+	mp1_fw_flags = RREG32_PCIE(MP1_Public |
+				   (smnMP1_FIRMWARE_FLAGS & 0xffffffff));
+
+	if ((mp1_fw_flags & MP1_FIRMWARE_FLAGS__INTERRUPTS_ENABLED_MASK) >>
+	    MP1_FIRMWARE_FLAGS__INTERRUPTS_ENABLED__SHIFT)
+		return 0;
+
+	return -EIO;
+}
+
+static int smu_v11_0_check_fw_version(struct smu_context *smu)
+{
+	uint32_t smu_version = 0xff;
+	int ret = 0;
+
+	ret = smu_send_smc_msg(smu, SMU_MSG_GetDriverIfVersion);
+	if (ret)
+		goto err;
+
+	ret = smu_read_smc_arg(smu, &smu_version);
+	if (ret)
+		goto err;
+
+	if (smu_version != smu->smc_if_version)
+		ret = -EINVAL;
+err:
+	return ret;
+}
+
+static int smu_v11_0_read_pptable_from_vbios(struct smu_context *smu)
+{
+	int ret, index;
+	uint16_t size;
+	uint8_t frev, crev;
+	void *table;
+
+	index = get_index_into_master_table(atom_master_list_of_data_tables_v2_1,
+					    powerplayinfo);
+
+	ret = smu_get_atom_data_table(smu, index, &size, &frev, &crev,
+				      (uint8_t **)&table);
+	if (ret)
+		return ret;
+
+	if (!smu->smu_table.power_play_table)
+		smu->smu_table.power_play_table = table;
+	if (!smu->smu_table.power_play_table_size)
+		smu->smu_table.power_play_table_size = size;
+
+	return 0;
+}
+
+static int smu_v11_0_init_dpm_context(struct smu_context *smu)
+{
+	struct smu_dpm_context *smu_dpm = &smu->smu_dpm;
+
+	if (smu_dpm->dpm_context || smu_dpm->dpm_context_size != 0)
+		return -EINVAL;
+
+	return smu_alloc_dpm_context(smu);
+}
+
+static int smu_v11_0_fini_dpm_context(struct smu_context *smu)
+{
+	struct smu_dpm_context *smu_dpm = &smu->smu_dpm;
+
+	if (!smu_dpm->dpm_context || smu_dpm->dpm_context_size == 0)
+		return -EINVAL;
+
+	kfree(smu_dpm->dpm_context);
+	kfree(smu_dpm->golden_dpm_context);
+	kfree(smu_dpm->dpm_current_power_state);
+	kfree(smu_dpm->dpm_request_power_state);
+	smu_dpm->dpm_context = NULL;
+	smu_dpm->golden_dpm_context = NULL;
+	smu_dpm->dpm_context_size = 0;
+	smu_dpm->dpm_current_power_state = NULL;
+	smu_dpm->dpm_request_power_state = NULL;
+
+	return 0;
+}
+
+static int smu_v11_0_init_smc_tables(struct smu_context *smu)
+{
+	struct smu_table_context *smu_table = &smu->smu_table;
+	struct smu_table *tables = NULL;
+	int ret = 0;
+
+	if (smu_table->tables || smu_table->table_count != 0)
+		return -EINVAL;
+
+	tables = kcalloc(TABLE_COUNT, sizeof(struct smu_table), GFP_KERNEL);
+	if (!tables)
+		return -ENOMEM;
+
+	smu_table->tables = tables;
+	smu_table->table_count = TABLE_COUNT;
+
+	SMU_TABLE_INIT(tables, TABLE_PPTABLE, sizeof(PPTable_t),
+		       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
+	SMU_TABLE_INIT(tables, TABLE_WATERMARKS, sizeof(Watermarks_t),
+		       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
+	SMU_TABLE_INIT(tables, TABLE_SMU_METRICS, sizeof(SmuMetrics_t),
+		       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
+	SMU_TABLE_INIT(tables, TABLE_OVERDRIVE, sizeof(OverDriveTable_t),
+		       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
+	SMU_TABLE_INIT(tables, TABLE_PMSTATUSLOG, SMU11_TOOL_SIZE, PAGE_SIZE,
+		       AMDGPU_GEM_DOMAIN_VRAM);
+	SMU_TABLE_INIT(tables, TABLE_ACTIVITY_MONITOR_COEFF,
+		       sizeof(DpmActivityMonitorCoeffInt_t),
+		       PAGE_SIZE,
+		       AMDGPU_GEM_DOMAIN_VRAM);
+
+	ret = smu_v11_0_init_dpm_context(smu);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static int smu_v11_0_fini_smc_tables(struct smu_context *smu)
+{
+	struct smu_table_context *smu_table = &smu->smu_table;
+	int ret = 0;
+
+	if (!smu_table->tables || smu_table->table_count == 0)
+		return -EINVAL;
+
+	kfree(smu_table->tables);
+	smu_table->tables = NULL;
+	smu_table->table_count = 0;
+
+	ret = smu_v11_0_fini_dpm_context(smu);
+	if (ret)
+		return ret;
+	return 0;
+}
+
+static int smu_v11_0_init_power(struct smu_context *smu)
+{
+	struct smu_power_context *smu_power = &smu->smu_power;
+
+	if (smu_power->power_context || smu_power->power_context_size != 0)
+		return -EINVAL;
+
+	smu_power->power_context = kzalloc(sizeof(struct smu_11_0_dpm_context),
+					   GFP_KERNEL);
+	if (!smu_power->power_context)
+		return -ENOMEM;
+	smu_power->power_context_size = sizeof(struct smu_11_0_dpm_context);
+
+	return 0;
+}
+
+static int smu_v11_0_fini_power(struct smu_context *smu)
+{
+	struct smu_power_context *smu_power = &smu->smu_power;
+
+	if (!smu_power->power_context || smu_power->power_context_size == 0)
+		return -EINVAL;
+
+	kfree(smu_power->power_context);
+	smu_power->power_context = NULL;
+	smu_power->power_context_size = 0;
+
+	return 0;
+}
+
+int smu_v11_0_get_vbios_bootup_values(struct smu_context *smu)
+{
+	int ret, index;
+	uint16_t size;
+	uint8_t frev, crev;
+	struct atom_common_table_header *header;
+	struct atom_firmware_info_v3_3 *v_3_3;
+	struct atom_firmware_info_v3_1 *v_3_1;
+
+	index = get_index_into_master_table(atom_master_list_of_data_tables_v2_1,
+					    firmwareinfo);
+
+	ret = smu_get_atom_data_table(smu, index, &size, &frev, &crev,
+				      (uint8_t **)&header);
+	if (ret)
+		return ret;
+
+	if (header->format_revision != 3) {
+		pr_err("unknown atom_firmware_info version! for smu11\n");
+		return -EINVAL;
+	}
+
+	switch (header->content_revision) {
+	case 0:
+	case 1:
+	case 2:
+		v_3_1 = (struct atom_firmware_info_v3_1 *)header;
+		smu->smu_table.boot_values.revision = v_3_1->firmware_revision;
+		smu->smu_table.boot_values.gfxclk = v_3_1->bootup_sclk_in10khz;
+		smu->smu_table.boot_values.uclk = v_3_1->bootup_mclk_in10khz;
+		smu->smu_table.boot_values.socclk = 0;
+		smu->smu_table.boot_values.dcefclk = 0;
+		smu->smu_table.boot_values.vddc = v_3_1->bootup_vddc_mv;
+		smu->smu_table.boot_values.vddci = v_3_1->bootup_vddci_mv;
+		smu->smu_table.boot_values.mvddc = v_3_1->bootup_mvddc_mv;
+		smu->smu_table.boot_values.vdd_gfx = v_3_1->bootup_vddgfx_mv;
+		smu->smu_table.boot_values.cooling_id = v_3_1->coolingsolution_id;
+		smu->smu_table.boot_values.pp_table_id = 0;
+		break;
+	case 3:
+	default:
+		v_3_3 = (struct atom_firmware_info_v3_3 *)header;
+		smu->smu_table.boot_values.revision = v_3_3->firmware_revision;
+		smu->smu_table.boot_values.gfxclk = v_3_3->bootup_sclk_in10khz;
+		smu->smu_table.boot_values.uclk = v_3_3->bootup_mclk_in10khz;
+		smu->smu_table.boot_values.socclk = 0;
+		smu->smu_table.boot_values.dcefclk = 0;
+		smu->smu_table.boot_values.vddc = v_3_3->bootup_vddc_mv;
+		smu->smu_table.boot_values.vddci = v_3_3->bootup_vddci_mv;
+		smu->smu_table.boot_values.mvddc = v_3_3->bootup_mvddc_mv;
+		smu->smu_table.boot_values.vdd_gfx = v_3_3->bootup_vddgfx_mv;
+		smu->smu_table.boot_values.cooling_id = v_3_3->coolingsolution_id;
+		smu->smu_table.boot_values.pp_table_id = v_3_3->pplib_pptable_id;
+	}
+
+	return 0;
+}
+
+static int smu_v11_0_get_clk_info_from_vbios(struct smu_context *smu)
+{
+	int ret, index;
+	struct amdgpu_device *adev = smu->adev;
+	struct atom_get_smu_clock_info_parameters_v3_1 input = {0};
+	struct atom_get_smu_clock_info_output_parameters_v3_1 *output;
+
+	input.clk_id = SMU11_SYSPLL0_SOCCLK_ID;
+	input.command = GET_SMU_CLOCK_INFO_V3_1_GET_CLOCK_FREQ;
+	index = get_index_into_master_table(atom_master_list_of_command_functions_v2_1,
+					    getsmuclockinfo);
+
+	ret = amdgpu_atom_execute_table(adev->mode_info.atom_context, index,
+					(uint32_t *)&input);
+	if (ret)
+		return -EINVAL;
+
+	output = (struct atom_get_smu_clock_info_output_parameters_v3_1 *)&input;
+	smu->smu_table.boot_values.socclk = le32_to_cpu(output->atom_smu_outputclkfreq.smu_clock_freq_hz) / 10000;
+
+	memset(&input, 0, sizeof(input));
+	input.clk_id = SMU11_SYSPLL0_DCEFCLK_ID;
+	input.command = GET_SMU_CLOCK_INFO_V3_1_GET_CLOCK_FREQ;
+	index = get_index_into_master_table(atom_master_list_of_command_functions_v2_1,
+					    getsmuclockinfo);
+
+	ret = amdgpu_atom_execute_table(adev->mode_info.atom_context, index,
+					(uint32_t *)&input);
+	if (ret)
+		return -EINVAL;
+
+	output = (struct atom_get_smu_clock_info_output_parameters_v3_1 *)&input;
+	smu->smu_table.boot_values.dcefclk = le32_to_cpu(output->atom_smu_outputclkfreq.smu_clock_freq_hz) / 10000;
+
+	memset(&input, 0, sizeof(input));
+	input.clk_id = SMU11_SYSPLL0_ECLK_ID;
+	input.command = GET_SMU_CLOCK_INFO_V3_1_GET_CLOCK_FREQ;
+	index = get_index_into_master_table(atom_master_list_of_command_functions_v2_1,
+					    getsmuclockinfo);
+
+	ret = amdgpu_atom_execute_table(adev->mode_info.atom_context, index,
+					(uint32_t *)&input);
+	if (ret)
+		return -EINVAL;
+
+	output = (struct atom_get_smu_clock_info_output_parameters_v3_1 *)&input;
+	smu->smu_table.boot_values.eclk = le32_to_cpu(output->atom_smu_outputclkfreq.smu_clock_freq_hz) / 10000;
+
+	memset(&input, 0, sizeof(input));
+	input.clk_id = SMU11_SYSPLL0_VCLK_ID;
+	input.command = GET_SMU_CLOCK_INFO_V3_1_GET_CLOCK_FREQ;
+	index = get_index_into_master_table(atom_master_list_of_command_functions_v2_1,
+					    getsmuclockinfo);
+
+	ret = amdgpu_atom_execute_table(adev->mode_info.atom_context, index,
+					(uint32_t *)&input);
+	if (ret)
+		return -EINVAL;
+
+	output = (struct atom_get_smu_clock_info_output_parameters_v3_1 *)&input;
+	smu->smu_table.boot_values.vclk = le32_to_cpu(output->atom_smu_outputclkfreq.smu_clock_freq_hz) / 10000;
+
+	memset(&input, 0, sizeof(input));
+	input.clk_id = SMU11_SYSPLL0_DCLK_ID;
+	input.command = GET_SMU_CLOCK_INFO_V3_1_GET_CLOCK_FREQ;
+	index = get_index_into_master_table(atom_master_list_of_command_functions_v2_1,
+					    getsmuclockinfo);
+
+	ret = amdgpu_atom_execute_table(adev->mode_info.atom_context, index,
+					(uint32_t *)&input);
+	if (ret)
+		return -EINVAL;
+
+	output = (struct atom_get_smu_clock_info_output_parameters_v3_1 *)&input;
+	smu->smu_table.boot_values.dclk = le32_to_cpu(output->atom_smu_outputclkfreq.smu_clock_freq_hz) / 10000;
+
+	return 0;
+}
+
+static int smu_v11_0_notify_memory_pool_location(struct smu_context *smu)
+{
+	struct smu_table_context *smu_table = &smu->smu_table;
+	struct smu_table *memory_pool = &smu_table->memory_pool;
+	int ret = 0;
+	uint64_t address;
+	uint32_t address_low, address_high;
+
+	if (memory_pool->size == 0 || memory_pool->cpu_addr == NULL)
+		return ret;
+
+	address = (uintptr_t)memory_pool->cpu_addr;
+	address_high = (uint32_t)upper_32_bits(address);
+	address_low  = (uint32_t)lower_32_bits(address);
+
+	ret = smu_send_smc_msg_with_param(smu,
+					  SMU_MSG_SetSystemVirtualDramAddrHigh,
+					  address_high);
+	if (ret)
+		return ret;
+	ret = smu_send_smc_msg_with_param(smu,
+					  SMU_MSG_SetSystemVirtualDramAddrLow,
+					  address_low);
+	if (ret)
+		return ret;
+
+	address = memory_pool->mc_address;
+	address_high = (uint32_t)upper_32_bits(address);
+	address_low  = (uint32_t)lower_32_bits(address);
+
+	ret = smu_send_smc_msg_with_param(smu, SMU_MSG_DramLogSetDramAddrHigh,
+					  address_high);
+	if (ret)
+		return ret;
+	ret = smu_send_smc_msg_with_param(smu, SMU_MSG_DramLogSetDramAddrLow,
+					  address_low);
+	if (ret)
+		return ret;
+	ret = smu_send_smc_msg_with_param(smu, SMU_MSG_DramLogSetDramSize,
+					  (uint32_t)memory_pool->size);
+	if (ret)
+		return ret;
+
+	return ret;
+}
+
+static int smu_v11_0_check_pptable(struct smu_context *smu)
+{
+	int ret;
+
+	ret = smu_check_powerplay_table(smu);
+	return ret;
+}
+
+static int smu_v11_0_parse_pptable(struct smu_context *smu)
+{
+	int ret;
+
+	struct smu_table_context *table_context = &smu->smu_table;
+
+	if (table_context->driver_pptable)
+		return -EINVAL;
+
+	table_context->driver_pptable = kzalloc(sizeof(PPTable_t), GFP_KERNEL);
+
+	if (!table_context->driver_pptable)
+		return -ENOMEM;
+
+	ret = smu_store_powerplay_table(smu);
+	if (ret)
+		return -EINVAL;
+
+	ret = smu_append_powerplay_table(smu);
+
+	return ret;
+}
+
+static int smu_v11_0_populate_smc_pptable(struct smu_context *smu)
+{
+	int ret;
+
+	ret = smu_set_default_dpm_table(smu);
+
+	return ret;
+}
+
+static int smu_v11_0_write_pptable(struct smu_context *smu)
+{
+	struct smu_table_context *table_context = &smu->smu_table;
+	int ret = 0;
+
+	ret = smu_update_table(smu, TABLE_PPTABLE, table_context->driver_pptable, true);
+
+	return ret;
+}
+
+static int smu_v11_0_write_watermarks_table(struct smu_context *smu)
+{
+	return smu_update_table(smu, TABLE_WATERMARKS,
+				smu->smu_table.tables[TABLE_WATERMARKS].cpu_addr, true);
+}
+
+static int smu_v11_0_set_deep_sleep_dcefclk(struct smu_context *smu, uint32_t clk)
+{
+	int ret;
+
+	ret = smu_send_smc_msg_with_param(smu,
+					  SMU_MSG_SetMinDeepSleepDcefclk, clk);
+	if (ret)
+		pr_err("SMU11 attempt to set divider for DCEFCLK Failed!");
+
+	return ret;
+}
+
+static int smu_v11_0_set_min_dcef_deep_sleep(struct smu_context *smu)
+{
+	struct smu_table_context *table_context = &smu->smu_table;
+
+	if (!table_context)
+		return -EINVAL;
+
+	return smu_set_deep_sleep_dcefclk(smu,
+					  table_context->boot_values.dcefclk / 100);
+}
+
+static int smu_v11_0_set_tool_table_location(struct smu_context *smu)
+{
+	int ret = 0;
+	struct smu_table *tool_table = &smu->smu_table.tables[TABLE_PMSTATUSLOG];
+
+	if (tool_table->mc_address) {
+		ret = smu_send_smc_msg_with_param(smu,
+				SMU_MSG_SetToolsDramAddrHigh,
+				upper_32_bits(tool_table->mc_address));
+		if (!ret)
+			ret = smu_send_smc_msg_with_param(smu,
+				SMU_MSG_SetToolsDramAddrLow,
+				lower_32_bits(tool_table->mc_address));
+	}
+
+	return ret;
+}
+
+static int smu_v11_0_init_display(struct smu_context *smu)
+{
+	int ret = 0;
+	ret = smu_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, 0);
+	return ret;
+}
+
+static int smu_v11_0_update_feature_enable_state(struct smu_context *smu, uint32_t feature_id, bool enabled)
+{
+	uint32_t feature_low = 0, feature_high = 0;
+	int ret = 0;
+
+	if (feature_id >= 0 && feature_id < 31)
+		feature_low = (1 << feature_id);
+	else if (feature_id > 31 && feature_id < 63)
+		feature_high = (1 << feature_id);
+	else
+		return -EINVAL;
+
+	if (enabled) {
+		ret = smu_send_smc_msg_with_param(smu, SMU_MSG_EnableSmuFeaturesLow,
+						  feature_low);
+		if (ret)
+			return ret;
+		ret = smu_send_smc_msg_with_param(smu, SMU_MSG_EnableSmuFeaturesHigh,
+						  feature_high);
+		if (ret)
+			return ret;
+
+	} else {
+		ret = smu_send_smc_msg_with_param(smu, SMU_MSG_DisableSmuFeaturesLow,
+						  feature_low);
+		if (ret)
+			return ret;
+		ret = smu_send_smc_msg_with_param(smu, SMU_MSG_DisableSmuFeaturesHigh,
+						  feature_high);
+		if (ret)
+			return ret;
+
+	}
+
+	return ret;
+}
+
+static int smu_v11_0_set_allowed_mask(struct smu_context *smu)
+{
+	struct smu_feature *feature = &smu->smu_feature;
+	int ret = 0;
+	uint32_t feature_mask[2];
+
+	mutex_lock(&feature->mutex);
+	if (bitmap_empty(feature->allowed, SMU_FEATURE_MAX) || feature->feature_num < 64)
+		goto failed;
+
+	bitmap_copy((unsigned long *)feature_mask, feature->allowed, 64);
+
+	ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetAllowedFeaturesMaskHigh,
+					  feature_mask[1]);
+	if (ret)
+		goto failed;
+
+	ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetAllowedFeaturesMaskLow,
+					  feature_mask[0]);
+	if (ret)
+		goto failed;
+
+failed:
+	mutex_unlock(&feature->mutex);
+	return ret;
+}
+
+static int smu_v11_0_get_enabled_mask(struct smu_context *smu,
+				      uint32_t *feature_mask, uint32_t num)
+{
+	uint32_t feature_mask_high = 0, feature_mask_low = 0;
+	int ret = 0;
+
+	if (!feature_mask || num < 2)
+		return -EINVAL;
+
+	ret = smu_send_smc_msg(smu, SMU_MSG_GetEnabledSmuFeaturesHigh);
+	if (ret)
+		return ret;
+	ret = smu_read_smc_arg(smu, &feature_mask_high);
+	if (ret)
+		return ret;
+
+	ret = smu_send_smc_msg(smu, SMU_MSG_GetEnabledSmuFeaturesLow);
+	if (ret)
+		return ret;
+	ret = smu_read_smc_arg(smu, &feature_mask_low);
+	if (ret)
+		return ret;
+
+	feature_mask[0] = feature_mask_low;
+	feature_mask[1] = feature_mask_high;
+
+	return ret;
+}
+
+static bool smu_v11_0_is_dpm_running(struct smu_context *smu)
+{
+	int ret = 0;
+	uint32_t feature_mask[2];
+	unsigned long feature_enabled;
+	ret = smu_v11_0_get_enabled_mask(smu, feature_mask, 2);
+	feature_enabled = (unsigned long)((uint64_t)feature_mask[0] |
+			   ((uint64_t)feature_mask[1] << 32));
+	return !!(feature_enabled & SMC_DPM_FEATURE);
+}
+
+static int smu_v11_0_system_features_control(struct smu_context *smu,
+					     bool en)
+{
+	struct smu_feature *feature = &smu->smu_feature;
+	uint32_t feature_mask[2];
+	int ret = 0;
+
+	ret = smu_send_smc_msg(smu, (en ? SMU_MSG_EnableAllSmuFeatures :
+				     SMU_MSG_DisableAllSmuFeatures));
+	if (ret)
+		return ret;
+	ret = smu_feature_get_enabled_mask(smu, feature_mask, 2);
+	if (ret)
+		return ret;
+
+	bitmap_copy(feature->enabled, (unsigned long *)&feature_mask,
+		    feature->feature_num);
+	bitmap_copy(feature->supported, (unsigned long *)&feature_mask,
+		    feature->feature_num);
+
+	return ret;
+}
+
+static int smu_v11_0_notify_display_change(struct smu_context *smu)
+{
+	int ret = 0;
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_UCLK_BIT))
+	    ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetUclkFastSwitch, 1);
+
+	return ret;
+}
+
+static int
+smu_v11_0_get_max_sustainable_clock(struct smu_context *smu, uint32_t *clock,
+				    PPCLK_e clock_select)
+{
+	int ret = 0;
+
+	ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetDcModeMaxDpmFreq,
+					  clock_select << 16);
+	if (ret) {
+		pr_err("[GetMaxSustainableClock] Failed to get max DC clock from SMC!");
+		return ret;
+	}
+
+	ret = smu_read_smc_arg(smu, clock);
+	if (ret)
+		return ret;
+
+	if (*clock != 0)
+		return 0;
+
+	/* if DC limit is zero, return AC limit */
+	ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetMaxDpmFreq,
+					  clock_select << 16);
+	if (ret) {
+		pr_err("[GetMaxSustainableClock] failed to get max AC clock from SMC!");
+		return ret;
+	}
+
+	ret = smu_read_smc_arg(smu, clock);
+
+	return ret;
+}
+
+static int smu_v11_0_init_max_sustainable_clocks(struct smu_context *smu)
+{
+	struct smu_11_0_max_sustainable_clocks *max_sustainable_clocks;
+	int ret = 0;
+
+	max_sustainable_clocks = kzalloc(sizeof(struct smu_11_0_max_sustainable_clocks),
+					 GFP_KERNEL);
+	smu->smu_table.max_sustainable_clocks = (void *)max_sustainable_clocks;
+
+	max_sustainable_clocks->uclock = smu->smu_table.boot_values.uclk / 100;
+	max_sustainable_clocks->soc_clock = smu->smu_table.boot_values.socclk / 100;
+	max_sustainable_clocks->dcef_clock = smu->smu_table.boot_values.dcefclk / 100;
+	max_sustainable_clocks->display_clock = 0xFFFFFFFF;
+	max_sustainable_clocks->phy_clock = 0xFFFFFFFF;
+	max_sustainable_clocks->pixel_clock = 0xFFFFFFFF;
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_UCLK_BIT)) {
+		ret = smu_v11_0_get_max_sustainable_clock(smu,
+							  &(max_sustainable_clocks->uclock),
+							  PPCLK_UCLK);
+		if (ret) {
+			pr_err("[%s] failed to get max UCLK from SMC!",
+			       __func__);
+			return ret;
+		}
+	}
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_SOCCLK_BIT)) {
+		ret = smu_v11_0_get_max_sustainable_clock(smu,
+							  &(max_sustainable_clocks->soc_clock),
+							  PPCLK_SOCCLK);
+		if (ret) {
+			pr_err("[%s] failed to get max SOCCLK from SMC!",
+			       __func__);
+			return ret;
+		}
+	}
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_DCEFCLK_BIT)) {
+		ret = smu_v11_0_get_max_sustainable_clock(smu,
+							  &(max_sustainable_clocks->dcef_clock),
+							  PPCLK_DCEFCLK);
+		if (ret) {
+			pr_err("[%s] failed to get max DCEFCLK from SMC!",
+			       __func__);
+			return ret;
+		}
+
+		ret = smu_v11_0_get_max_sustainable_clock(smu,
+							  &(max_sustainable_clocks->display_clock),
+							  PPCLK_DISPCLK);
+		if (ret) {
+			pr_err("[%s] failed to get max DISPCLK from SMC!",
+			       __func__);
+			return ret;
+		}
+		ret = smu_v11_0_get_max_sustainable_clock(smu,
+							  &(max_sustainable_clocks->phy_clock),
+							  PPCLK_PHYCLK);
+		if (ret) {
+			pr_err("[%s] failed to get max PHYCLK from SMC!",
+			       __func__);
+			return ret;
+		}
+		ret = smu_v11_0_get_max_sustainable_clock(smu,
+							  &(max_sustainable_clocks->pixel_clock),
+							  PPCLK_PIXCLK);
+		if (ret) {
+			pr_err("[%s] failed to get max PIXCLK from SMC!",
+			       __func__);
+			return ret;
+		}
+	}
+
+	if (max_sustainable_clocks->soc_clock < max_sustainable_clocks->uclock)
+		max_sustainable_clocks->uclock = max_sustainable_clocks->soc_clock;
+
+	return 0;
+}
+
+static int smu_v11_0_get_power_limit(struct smu_context *smu,
+				     uint32_t *limit,
+				     bool get_default)
+{
+	int ret = 0;
+
+	if (get_default) {
+		mutex_lock(&smu->mutex);
+		*limit = smu->default_power_limit;
+		if (smu->od_enabled) {
+			*limit *= (100 + smu->smu_table.TDPODLimit);
+			*limit /= 100;
+		}
+		mutex_unlock(&smu->mutex);
+	} else {
+		ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetPptLimit,
+						  POWER_SOURCE_AC << 16);
+		if (ret) {
+			pr_err("[%s] get PPT limit failed!", __func__);
+			return ret;
+		}
+		smu_read_smc_arg(smu, limit);
+		smu->power_limit = *limit;
+	}
+
+	return ret;
+}
+
+static int smu_v11_0_set_power_limit(struct smu_context *smu, uint32_t n)
+{
+	uint32_t max_power_limit;
+	int ret = 0;
+
+	if (n == 0)
+		n = smu->default_power_limit;
+
+	max_power_limit = smu->default_power_limit;
+
+	if (smu->od_enabled) {
+		max_power_limit *= (100 + smu->smu_table.TDPODLimit);
+		max_power_limit /= 100;
+	}
+
+	if (smu_feature_is_enabled(smu, FEATURE_PPT_BIT))
+		ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetPptLimit, n);
+	if (ret) {
+		pr_err("[%s] Set power limit Failed!", __func__);
+		return ret;
+	}
+
+	return ret;
+}
+
+static int smu_v11_0_get_current_clk_freq(struct smu_context *smu, uint32_t clk_id, uint32_t *value)
+{
+	int ret = 0;
+	uint32_t freq;
+
+	if (clk_id >= PPCLK_COUNT || !value)
+		return -EINVAL;
+
+	ret = smu_send_smc_msg_with_param(smu,
+			SMU_MSG_GetDpmClockFreq, (clk_id << 16));
+	if (ret)
+		return ret;
+
+	ret = smu_read_smc_arg(smu, &freq);
+	if (ret)
+		return ret;
+
+	freq *= 100;
+	*value = freq;
+
+	return ret;
+}
+
+static int smu_v11_0_get_thermal_range(struct smu_context *smu,
+				struct PP_TemperatureRange *range)
+{
+	memcpy(range, &SMU7ThermalWithDelayPolicy[0], sizeof(struct PP_TemperatureRange));
+
+	range->max = smu->smu_table.software_shutdown_temp *
+		PP_TEMPERATURE_UNITS_PER_CENTIGRADES;
+
+	return 0;
+}
+
+static int smu_v11_0_set_thermal_range(struct smu_context *smu,
+			struct PP_TemperatureRange *range)
+{
+	struct amdgpu_device *adev = smu->adev;
+	int low = SMU11_THERMAL_MINIMUM_ALERT_TEMP *
+		PP_TEMPERATURE_UNITS_PER_CENTIGRADES;
+	int high = SMU11_THERMAL_MAXIMUM_ALERT_TEMP *
+		PP_TEMPERATURE_UNITS_PER_CENTIGRADES;
+	uint32_t val;
+
+	if (low < range->min)
+		low = range->min;
+	if (high > range->max)
+		high = range->max;
+
+	if (low > high)
+		return -EINVAL;
+
+	val = RREG32_SOC15(THM, 0, mmTHM_THERMAL_INT_CTRL);
+	val = REG_SET_FIELD(val, THM_THERMAL_INT_CTRL, MAX_IH_CREDIT, 5);
+	val = REG_SET_FIELD(val, THM_THERMAL_INT_CTRL, THERM_IH_HW_ENA, 1);
+	val = REG_SET_FIELD(val, THM_THERMAL_INT_CTRL, DIG_THERM_INTH, (high / PP_TEMPERATURE_UNITS_PER_CENTIGRADES));
+	val = REG_SET_FIELD(val, THM_THERMAL_INT_CTRL, DIG_THERM_INTL, (low / PP_TEMPERATURE_UNITS_PER_CENTIGRADES));
+	val = val & (~THM_THERMAL_INT_CTRL__THERM_TRIGGER_MASK_MASK);
+
+	WREG32_SOC15(THM, 0, mmTHM_THERMAL_INT_CTRL, val);
+
+	return 0;
+}
+
+static int smu_v11_0_enable_thermal_alert(struct smu_context *smu)
+{
+	struct amdgpu_device *adev = smu->adev;
+	uint32_t val = 0;
+
+	val |= (1 << THM_THERMAL_INT_ENA__THERM_INTH_CLR__SHIFT);
+	val |= (1 << THM_THERMAL_INT_ENA__THERM_INTL_CLR__SHIFT);
+	val |= (1 << THM_THERMAL_INT_ENA__THERM_TRIGGER_CLR__SHIFT);
+
+	WREG32_SOC15(THM, 0, mmTHM_THERMAL_INT_ENA, val);
+
+	return 0;
+}
+
+static int smu_v11_0_set_thermal_fan_table(struct smu_context *smu)
+{
+	int ret;
+	struct smu_table_context *table_context = &smu->smu_table;
+	PPTable_t *pptable = table_context->driver_pptable;
+
+	ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetFanTemperatureTarget,
+			(uint32_t)pptable->FanTargetTemperature);
+
+	return ret;
+}
+
+static int smu_v11_0_start_thermal_control(struct smu_context *smu)
+{
+	int ret = 0;
+	struct PP_TemperatureRange range;
+	struct amdgpu_device *adev = smu->adev;
+
+	smu_v11_0_get_thermal_range(smu, &range);
+
+	if (smu->smu_table.thermal_controller_type) {
+		ret = smu_v11_0_set_thermal_range(smu, &range);
+		if (ret)
+			return ret;
+
+		ret = smu_v11_0_enable_thermal_alert(smu);
+		if (ret)
+			return ret;
+		ret = smu_v11_0_set_thermal_fan_table(smu);
+		if (ret)
+			return ret;
+	}
+
+	adev->pm.dpm.thermal.min_temp = range.min;
+	adev->pm.dpm.thermal.max_temp = range.max;
+
+	return ret;
+}
+
+static int smu_v11_0_get_current_activity_percent(struct smu_context *smu,
+						  uint32_t *value)
+{
+	int ret = 0;
+	SmuMetrics_t metrics;
+
+	if (!value)
+		return -EINVAL;
+
+	ret = smu_update_table(smu, TABLE_SMU_METRICS, (void *)&metrics, false);
+	if (ret)
+		return ret;
+
+	*value = metrics.AverageGfxActivity;
+
+	return 0;
+}
+
+static int smu_v11_0_thermal_get_temperature(struct smu_context *smu, uint32_t *value)
+{
+	struct amdgpu_device *adev = smu->adev;
+	uint32_t temp = 0;
+
+	if (!value)
+		return -EINVAL;
+
+	temp = RREG32_SOC15(THM, 0, mmCG_MULT_THERMAL_STATUS);
+	temp = (temp & CG_MULT_THERMAL_STATUS__CTF_TEMP_MASK) >>
+			CG_MULT_THERMAL_STATUS__CTF_TEMP__SHIFT;
+
+	temp = temp & 0x1ff;
+	temp *= SMU11_TEMPERATURE_UNITS_PER_CENTIGRADES;
+
+	*value = temp;
+
+	return 0;
+}
+
+static int smu_v11_0_get_gpu_power(struct smu_context *smu, uint32_t *value)
+{
+	int ret = 0;
+	SmuMetrics_t metrics;
+
+	if (!value)
+		return -EINVAL;
+
+	ret = smu_update_table(smu, TABLE_SMU_METRICS, (void *)&metrics, false);
+	if (ret)
+		return ret;
+
+	*value = metrics.CurrSocketPower << 8;
+
+	return 0;
+}
+
+static uint16_t convert_to_vddc(uint8_t vid)
+{
+	return (uint16_t) ((6200 - (vid * 25)) / SMU11_VOLTAGE_SCALE);
+}
+
+static int smu_v11_0_get_gfx_vdd(struct smu_context *smu, uint32_t *value)
+{
+	struct amdgpu_device *adev = smu->adev;
+	uint32_t vdd = 0, val_vid = 0;
+
+	if (!value)
+		return -EINVAL;
+	val_vid = (RREG32_SOC15(SMUIO, 0, mmSMUSVI0_TEL_PLANE0) &
+		SMUSVI0_TEL_PLANE0__SVI0_PLANE0_VDDCOR_MASK) >>
+		SMUSVI0_TEL_PLANE0__SVI0_PLANE0_VDDCOR__SHIFT;
+
+	vdd = (uint32_t)convert_to_vddc((uint8_t)val_vid);
+
+	*value = vdd;
+
+	return 0;
+
+}
+
+static int smu_v11_0_read_sensor(struct smu_context *smu,
+				 enum amd_pp_sensors sensor,
+				 void *data, uint32_t *size)
+{
+	struct smu_table_context *table_context = &smu->smu_table;
+	PPTable_t *pptable = table_context->driver_pptable;
+	int ret = 0;
+	switch (sensor) {
+	case AMDGPU_PP_SENSOR_GPU_LOAD:
+		ret = smu_v11_0_get_current_activity_percent(smu,
+							     (uint32_t *)data);
+		*size = 4;
+		break;
+	case AMDGPU_PP_SENSOR_GFX_MCLK:
+		ret = smu_get_current_clk_freq(smu, PPCLK_UCLK, (uint32_t *)data);
+		*size = 4;
+		break;
+	case AMDGPU_PP_SENSOR_GFX_SCLK:
+		ret = smu_get_current_clk_freq(smu, PPCLK_GFXCLK, (uint32_t *)data);
+		*size = 4;
+		break;
+	case AMDGPU_PP_SENSOR_GPU_TEMP:
+		ret = smu_v11_0_thermal_get_temperature(smu, (uint32_t *)data);
+		*size = 4;
+		break;
+	case AMDGPU_PP_SENSOR_GPU_POWER:
+		ret = smu_v11_0_get_gpu_power(smu, (uint32_t *)data);
+		*size = 4;
+		break;
+	case AMDGPU_PP_SENSOR_VDDGFX:
+		ret = smu_v11_0_get_gfx_vdd(smu, (uint32_t *)data);
+		*size = 4;
+		break;
+	case AMDGPU_PP_SENSOR_UVD_POWER:
+		*(uint32_t *)data = smu_feature_is_enabled(smu, FEATURE_DPM_UVD_BIT) ? 1 : 0;
+		*size = 4;
+		break;
+	case AMDGPU_PP_SENSOR_VCE_POWER:
+		*(uint32_t *)data = smu_feature_is_enabled(smu, FEATURE_DPM_VCE_BIT) ? 1 : 0;
+		*size = 4;
+		break;
+	case AMDGPU_PP_SENSOR_MIN_FAN_RPM:
+		*(uint32_t *)data = 0;
+		*size = 4;
+		break;
+	case AMDGPU_PP_SENSOR_MAX_FAN_RPM:
+		*(uint32_t *)data = pptable->FanMaximumRpm;
+		*size = 4;
+		break;
+	default:
+		ret = smu_common_read_sensor(smu, sensor, data, size);
+		break;
+	}
+
+	if (ret)
+		*size = 0;
+
+	return ret;
+}
+
+static int
+smu_v11_0_display_clock_voltage_request(struct smu_context *smu,
+					struct pp_display_clock_request
+					*clock_req)
+{
+	enum amd_pp_clock_type clk_type = clock_req->clock_type;
+	int ret = 0;
+	PPCLK_e clk_select = 0;
+	uint32_t clk_freq = clock_req->clock_freq_in_khz / 1000;
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_DCEFCLK_BIT)) {
+		switch (clk_type) {
+		case amd_pp_dcef_clock:
+			clk_select = PPCLK_DCEFCLK;
+			break;
+		case amd_pp_disp_clock:
+			clk_select = PPCLK_DISPCLK;
+			break;
+		case amd_pp_pixel_clock:
+			clk_select = PPCLK_PIXCLK;
+			break;
+		case amd_pp_phy_clock:
+			clk_select = PPCLK_PHYCLK;
+			break;
+		default:
+			pr_info("[%s] Invalid Clock Type!", __func__);
+			ret = -EINVAL;
+			break;
+		}
+
+		if (ret)
+			goto failed;
+
+		ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinByFreq,
+						  (clk_select << 16) | clk_freq);
+	}
+
+failed:
+	return ret;
+}
+
+static int smu_v11_0_set_watermarks_table(struct smu_context *smu,
+					  Watermarks_t *table, struct
+					  dm_pp_wm_sets_with_clock_ranges_soc15
+					  *clock_ranges)
+{
+	int i;
+
+	if (!table || !clock_ranges)
+		return -EINVAL;
+
+	if (clock_ranges->num_wm_dmif_sets > 4 ||
+	    clock_ranges->num_wm_mcif_sets > 4)
+                return -EINVAL;
+
+        for (i = 0; i < clock_ranges->num_wm_dmif_sets; i++) {
+		table->WatermarkRow[1][i].MinClock =
+			cpu_to_le16((uint16_t)
+			(clock_ranges->wm_dmif_clocks_ranges[i].wm_min_dcfclk_clk_in_khz /
+			1000));
+		table->WatermarkRow[1][i].MaxClock =
+			cpu_to_le16((uint16_t)
+			(clock_ranges->wm_dmif_clocks_ranges[i].wm_max_dcfclk_clk_in_khz /
+			1000));
+		table->WatermarkRow[1][i].MinUclk =
+			cpu_to_le16((uint16_t)
+			(clock_ranges->wm_dmif_clocks_ranges[i].wm_min_mem_clk_in_khz /
+			1000));
+		table->WatermarkRow[1][i].MaxUclk =
+			cpu_to_le16((uint16_t)
+			(clock_ranges->wm_dmif_clocks_ranges[i].wm_max_mem_clk_in_khz /
+			1000));
+		table->WatermarkRow[1][i].WmSetting = (uint8_t)
+				clock_ranges->wm_dmif_clocks_ranges[i].wm_set_id;
+        }
+
+	for (i = 0; i < clock_ranges->num_wm_mcif_sets; i++) {
+		table->WatermarkRow[0][i].MinClock =
+			cpu_to_le16((uint16_t)
+			(clock_ranges->wm_mcif_clocks_ranges[i].wm_min_socclk_clk_in_khz /
+			1000));
+		table->WatermarkRow[0][i].MaxClock =
+			cpu_to_le16((uint16_t)
+			(clock_ranges->wm_mcif_clocks_ranges[i].wm_max_socclk_clk_in_khz /
+			1000));
+		table->WatermarkRow[0][i].MinUclk =
+			cpu_to_le16((uint16_t)
+			(clock_ranges->wm_mcif_clocks_ranges[i].wm_min_mem_clk_in_khz /
+			1000));
+		table->WatermarkRow[0][i].MaxUclk =
+			cpu_to_le16((uint16_t)
+			(clock_ranges->wm_mcif_clocks_ranges[i].wm_max_mem_clk_in_khz /
+			1000));
+		table->WatermarkRow[0][i].WmSetting = (uint8_t)
+				clock_ranges->wm_mcif_clocks_ranges[i].wm_set_id;
+        }
+
+	return 0;
+}
+
+static int
+smu_v11_0_set_watermarks_for_clock_ranges(struct smu_context *smu, struct
+					  dm_pp_wm_sets_with_clock_ranges_soc15
+					  *clock_ranges)
+{
+	int ret = 0;
+	struct smu_table *watermarks = &smu->smu_table.tables[TABLE_WATERMARKS];
+	Watermarks_t *table = watermarks->cpu_addr;
+
+	if (!smu->disable_watermark &&
+	    smu_feature_is_enabled(smu, FEATURE_DPM_DCEFCLK_BIT) &&
+	    smu_feature_is_enabled(smu, FEATURE_DPM_SOCCLK_BIT)) {
+		smu_v11_0_set_watermarks_table(smu, table, clock_ranges);
+		smu->watermarks_bitmap |= WATERMARKS_EXIST;
+		smu->watermarks_bitmap &= ~WATERMARKS_LOADED;
+	}
+
+	return ret;
+}
+
+static int smu_v11_0_get_clock_ranges(struct smu_context *smu,
+				      uint32_t *clock,
+				      PPCLK_e clock_select,
+				      bool max)
+{
+	int ret;
+	*clock = 0;
+	if (max) {
+		ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetMaxDpmFreq,
+					    (clock_select << 16));
+		if (ret) {
+			pr_err("[GetClockRanges] Failed to get max clock from SMC!\n");
+			return ret;
+		}
+		smu_read_smc_arg(smu, clock);
+	} else {
+		ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetMinDpmFreq,
+					    (clock_select << 16));
+		if (ret) {
+			pr_err("[GetClockRanges] Failed to get min clock from SMC!\n");
+			return ret;
+		}
+		smu_read_smc_arg(smu, clock);
+	}
+
+	return 0;
+}
+
+static uint32_t smu_v11_0_dpm_get_sclk(struct smu_context *smu, bool low)
+{
+	uint32_t gfx_clk;
+	int ret;
+
+	if (!smu_feature_is_enabled(smu, FEATURE_DPM_GFXCLK_BIT)) {
+		pr_err("[GetSclks]: gfxclk dpm not enabled!\n");
+		return -EPERM;
+	}
+
+	if (low) {
+		ret = smu_v11_0_get_clock_ranges(smu, &gfx_clk, PPCLK_GFXCLK, false);
+		if (ret) {
+			pr_err("[GetSclks]: fail to get min PPCLK_GFXCLK\n");
+			return ret;
+		}
+	} else {
+		ret = smu_v11_0_get_clock_ranges(smu, &gfx_clk, PPCLK_GFXCLK, true);
+		if (ret) {
+			pr_err("[GetSclks]: fail to get max PPCLK_GFXCLK\n");
+			return ret;
+		}
+	}
+
+	return (gfx_clk * 100);
+}
+
+static uint32_t smu_v11_0_dpm_get_mclk(struct smu_context *smu, bool low)
+{
+	uint32_t mem_clk;
+	int ret;
+
+	if (!smu_feature_is_enabled(smu, FEATURE_DPM_UCLK_BIT)) {
+		pr_err("[GetMclks]: memclk dpm not enabled!\n");
+		return -EPERM;
+	}
+
+	if (low) {
+		ret = smu_v11_0_get_clock_ranges(smu, &mem_clk, PPCLK_UCLK, false);
+		if (ret) {
+			pr_err("[GetMclks]: fail to get min PPCLK_UCLK\n");
+			return ret;
+		}
+	} else {
+		ret = smu_v11_0_get_clock_ranges(smu, &mem_clk, PPCLK_GFXCLK, true);
+		if (ret) {
+			pr_err("[GetMclks]: fail to get max PPCLK_UCLK\n");
+			return ret;
+		}
+	}
+
+	return (mem_clk * 100);
+}
+
+static int smu_v11_0_set_od8_default_settings(struct smu_context *smu,
+					      bool initialize)
+{
+	struct smu_table_context *table_context = &smu->smu_table;
+	int ret;
+
+	if (initialize) {
+		if (table_context->overdrive_table)
+			return -EINVAL;
+
+		table_context->overdrive_table = kzalloc(sizeof(OverDriveTable_t), GFP_KERNEL);
+
+		if (!table_context->overdrive_table)
+			return -ENOMEM;
+
+		ret = smu_update_table(smu, TABLE_OVERDRIVE, table_context->overdrive_table, false);
+		if (ret) {
+			pr_err("Failed to export over drive table!\n");
+			return ret;
+		}
+
+		smu_set_default_od8_settings(smu);
+	}
+
+	ret = smu_update_table(smu, TABLE_OVERDRIVE, table_context->overdrive_table, true);
+	if (ret) {
+		pr_err("Failed to import over drive table!\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static int smu_v11_0_conv_power_profile_to_pplib_workload(int power_profile)
+{
+	int pplib_workload = 0;
+
+	switch (power_profile) {
+	case PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT:
+	     pplib_workload = WORKLOAD_DEFAULT_BIT;
+	     break;
+	case PP_SMC_POWER_PROFILE_FULLSCREEN3D:
+	     pplib_workload = WORKLOAD_PPLIB_FULL_SCREEN_3D_BIT;
+	     break;
+	case PP_SMC_POWER_PROFILE_POWERSAVING:
+	     pplib_workload = WORKLOAD_PPLIB_POWER_SAVING_BIT;
+	     break;
+	case PP_SMC_POWER_PROFILE_VIDEO:
+	     pplib_workload = WORKLOAD_PPLIB_VIDEO_BIT;
+	     break;
+	case PP_SMC_POWER_PROFILE_VR:
+	     pplib_workload = WORKLOAD_PPLIB_VR_BIT;
+	     break;
+	case PP_SMC_POWER_PROFILE_COMPUTE:
+	     pplib_workload = WORKLOAD_PPLIB_COMPUTE_BIT;
+	     break;
+	case PP_SMC_POWER_PROFILE_CUSTOM:
+		pplib_workload = WORKLOAD_PPLIB_CUSTOM_BIT;
+		break;
+	}
+
+	return pplib_workload;
+}
+
+static int smu_v11_0_get_power_profile_mode(struct smu_context *smu, char *buf)
+{
+	DpmActivityMonitorCoeffInt_t activity_monitor;
+	uint32_t i, size = 0;
+	uint16_t workload_type = 0;
+	static const char *profile_name[] = {
+					"BOOTUP_DEFAULT",
+					"3D_FULL_SCREEN",
+					"POWER_SAVING",
+					"VIDEO",
+					"VR",
+					"COMPUTE",
+					"CUSTOM"};
+	static const char *title[] = {
+			"PROFILE_INDEX(NAME)",
+			"CLOCK_TYPE(NAME)",
+			"FPS",
+			"UseRlcBusy",
+			"MinActiveFreqType",
+			"MinActiveFreq",
+			"BoosterFreqType",
+			"BoosterFreq",
+			"PD_Data_limit_c",
+			"PD_Data_error_coeff",
+			"PD_Data_error_rate_coeff"};
+	int result = 0;
+
+	if (!buf)
+		return -EINVAL;
+
+	size += sprintf(buf + size, "%16s %s %s %s %s %s %s %s %s %s %s\n",
+			title[0], title[1], title[2], title[3], title[4], title[5],
+			title[6], title[7], title[8], title[9], title[10]);
+
+	for (i = 0; i <= PP_SMC_POWER_PROFILE_CUSTOM; i++) {
+		/* conv PP_SMC_POWER_PROFILE* to WORKLOAD_PPLIB_*_BIT */
+		workload_type = smu_v11_0_conv_power_profile_to_pplib_workload(i);
+		result = smu_update_table_with_arg(smu, TABLE_ACTIVITY_MONITOR_COEFF,
+						   workload_type, &activity_monitor, false);
+		if (result) {
+			pr_err("[%s] Failed to get activity monitor!", __func__);
+			return result;
+		}
+
+		size += sprintf(buf + size, "%2d %14s%s:\n",
+			i, profile_name[i], (i == smu->power_profile_mode) ? "*" : " ");
+
+		size += sprintf(buf + size, "%19s %d(%13s) %7d %7d %7d %7d %7d %7d %7d %7d %7d\n",
+			" ",
+			0,
+			"GFXCLK",
+			activity_monitor.Gfx_FPS,
+			activity_monitor.Gfx_UseRlcBusy,
+			activity_monitor.Gfx_MinActiveFreqType,
+			activity_monitor.Gfx_MinActiveFreq,
+			activity_monitor.Gfx_BoosterFreqType,
+			activity_monitor.Gfx_BoosterFreq,
+			activity_monitor.Gfx_PD_Data_limit_c,
+			activity_monitor.Gfx_PD_Data_error_coeff,
+			activity_monitor.Gfx_PD_Data_error_rate_coeff);
+
+		size += sprintf(buf + size, "%19s %d(%13s) %7d %7d %7d %7d %7d %7d %7d %7d %7d\n",
+			" ",
+			1,
+			"SOCCLK",
+			activity_monitor.Soc_FPS,
+			activity_monitor.Soc_UseRlcBusy,
+			activity_monitor.Soc_MinActiveFreqType,
+			activity_monitor.Soc_MinActiveFreq,
+			activity_monitor.Soc_BoosterFreqType,
+			activity_monitor.Soc_BoosterFreq,
+			activity_monitor.Soc_PD_Data_limit_c,
+			activity_monitor.Soc_PD_Data_error_coeff,
+			activity_monitor.Soc_PD_Data_error_rate_coeff);
+
+		size += sprintf(buf + size, "%19s %d(%13s) %7d %7d %7d %7d %7d %7d %7d %7d %7d\n",
+			" ",
+			2,
+			"UCLK",
+			activity_monitor.Mem_FPS,
+			activity_monitor.Mem_UseRlcBusy,
+			activity_monitor.Mem_MinActiveFreqType,
+			activity_monitor.Mem_MinActiveFreq,
+			activity_monitor.Mem_BoosterFreqType,
+			activity_monitor.Mem_BoosterFreq,
+			activity_monitor.Mem_PD_Data_limit_c,
+			activity_monitor.Mem_PD_Data_error_coeff,
+			activity_monitor.Mem_PD_Data_error_rate_coeff);
+
+		size += sprintf(buf + size, "%19s %d(%13s) %7d %7d %7d %7d %7d %7d %7d %7d %7d\n",
+			" ",
+			3,
+			"FCLK",
+			activity_monitor.Fclk_FPS,
+			activity_monitor.Fclk_UseRlcBusy,
+			activity_monitor.Fclk_MinActiveFreqType,
+			activity_monitor.Fclk_MinActiveFreq,
+			activity_monitor.Fclk_BoosterFreqType,
+			activity_monitor.Fclk_BoosterFreq,
+			activity_monitor.Fclk_PD_Data_limit_c,
+			activity_monitor.Fclk_PD_Data_error_coeff,
+			activity_monitor.Fclk_PD_Data_error_rate_coeff);
+	}
+
+	return size;
+}
+
+static int smu_v11_0_set_power_profile_mode(struct smu_context *smu, long *input, uint32_t size)
+{
+	DpmActivityMonitorCoeffInt_t activity_monitor;
+	int workload_type = 0, ret = 0;
+
+	smu->power_profile_mode = input[size];
+
+	if (smu->power_profile_mode > PP_SMC_POWER_PROFILE_CUSTOM) {
+		pr_err("Invalid power profile mode %d\n", smu->power_profile_mode);
+		return -EINVAL;
+	}
+
+	if (smu->power_profile_mode == PP_SMC_POWER_PROFILE_CUSTOM) {
+		ret = smu_update_table_with_arg(smu, TABLE_ACTIVITY_MONITOR_COEFF,
+						WORKLOAD_PPLIB_CUSTOM_BIT, &activity_monitor, false);
+		if (ret) {
+			pr_err("[%s] Failed to get activity monitor!", __func__);
+			return ret;
+		}
+
+		switch (input[0]) {
+		case 0: /* Gfxclk */
+			activity_monitor.Gfx_FPS = input[1];
+			activity_monitor.Gfx_UseRlcBusy = input[2];
+			activity_monitor.Gfx_MinActiveFreqType = input[3];
+			activity_monitor.Gfx_MinActiveFreq = input[4];
+			activity_monitor.Gfx_BoosterFreqType = input[5];
+			activity_monitor.Gfx_BoosterFreq = input[6];
+			activity_monitor.Gfx_PD_Data_limit_c = input[7];
+			activity_monitor.Gfx_PD_Data_error_coeff = input[8];
+			activity_monitor.Gfx_PD_Data_error_rate_coeff = input[9];
+			break;
+		case 1: /* Socclk */
+			activity_monitor.Soc_FPS = input[1];
+			activity_monitor.Soc_UseRlcBusy = input[2];
+			activity_monitor.Soc_MinActiveFreqType = input[3];
+			activity_monitor.Soc_MinActiveFreq = input[4];
+			activity_monitor.Soc_BoosterFreqType = input[5];
+			activity_monitor.Soc_BoosterFreq = input[6];
+			activity_monitor.Soc_PD_Data_limit_c = input[7];
+			activity_monitor.Soc_PD_Data_error_coeff = input[8];
+			activity_monitor.Soc_PD_Data_error_rate_coeff = input[9];
+			break;
+		case 2: /* Uclk */
+			activity_monitor.Mem_FPS = input[1];
+			activity_monitor.Mem_UseRlcBusy = input[2];
+			activity_monitor.Mem_MinActiveFreqType = input[3];
+			activity_monitor.Mem_MinActiveFreq = input[4];
+			activity_monitor.Mem_BoosterFreqType = input[5];
+			activity_monitor.Mem_BoosterFreq = input[6];
+			activity_monitor.Mem_PD_Data_limit_c = input[7];
+			activity_monitor.Mem_PD_Data_error_coeff = input[8];
+			activity_monitor.Mem_PD_Data_error_rate_coeff = input[9];
+			break;
+		case 3: /* Fclk */
+			activity_monitor.Fclk_FPS = input[1];
+			activity_monitor.Fclk_UseRlcBusy = input[2];
+			activity_monitor.Fclk_MinActiveFreqType = input[3];
+			activity_monitor.Fclk_MinActiveFreq = input[4];
+			activity_monitor.Fclk_BoosterFreqType = input[5];
+			activity_monitor.Fclk_BoosterFreq = input[6];
+			activity_monitor.Fclk_PD_Data_limit_c = input[7];
+			activity_monitor.Fclk_PD_Data_error_coeff = input[8];
+			activity_monitor.Fclk_PD_Data_error_rate_coeff = input[9];
+			break;
+		}
+
+		ret = smu_update_table_with_arg(smu, TABLE_ACTIVITY_MONITOR_COEFF,
+						WORKLOAD_PPLIB_COMPUTE_BIT, &activity_monitor, true);
+		if (ret) {
+			pr_err("[%s] Failed to set activity monitor!", __func__);
+			return ret;
+		}
+	}
+
+	/* conv PP_SMC_POWER_PROFILE* to WORKLOAD_PPLIB_*_BIT */
+	workload_type =
+		smu_v11_0_conv_power_profile_to_pplib_workload(smu->power_profile_mode);
+	smu_send_smc_msg_with_param(smu, SMU_MSG_SetWorkloadMask,
+				    1 << workload_type);
+
+	return ret;
+}
+
+static int smu_v11_0_update_od8_settings(struct smu_context *smu,
+					uint32_t index,
+					uint32_t value)
+{
+	struct smu_table_context *table_context = &smu->smu_table;
+	int ret;
+
+	ret = smu_update_table(smu, TABLE_OVERDRIVE,
+			       table_context->overdrive_table, false);
+	if (ret) {
+		pr_err("Failed to export over drive table!\n");
+		return ret;
+	}
+
+	smu_update_specified_od8_value(smu, index, value);
+
+	ret = smu_update_table(smu, TABLE_OVERDRIVE,
+			       table_context->overdrive_table, true);
+	if (ret) {
+		pr_err("Failed to import over drive table!\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static int smu_v11_0_dpm_set_uvd_enable(struct smu_context *smu, bool enable)
+{
+	if (!smu_feature_is_supported(smu, FEATURE_DPM_VCE_BIT))
+		return 0;
+
+	if (enable == smu_feature_is_enabled(smu, FEATURE_DPM_VCE_BIT))
+		return 0;
+
+	return smu_feature_set_enabled(smu, FEATURE_DPM_VCE_BIT, enable);
+}
+
+static int smu_v11_0_dpm_set_vce_enable(struct smu_context *smu, bool enable)
+{
+	if (!smu_feature_is_supported(smu, FEATURE_DPM_UVD_BIT))
+		return 0;
+
+	if (enable == smu_feature_is_enabled(smu, FEATURE_DPM_UVD_BIT))
+		return 0;
+
+	return smu_feature_set_enabled(smu, FEATURE_DPM_UVD_BIT, enable);
+}
+
+static int smu_v11_0_get_current_rpm(struct smu_context *smu,
+				     uint32_t *current_rpm)
+{
+	int ret;
+
+	ret = smu_send_smc_msg(smu, SMU_MSG_GetCurrentRpm);
+
+	if (ret) {
+		pr_err("Attempt to get current RPM from SMC Failed!\n");
+		return ret;
+	}
+
+	smu_read_smc_arg(smu, current_rpm);
+
+	return 0;
+}
+
+static uint32_t
+smu_v11_0_get_fan_control_mode(struct smu_context *smu)
+{
+	if (!smu_feature_is_enabled(smu, FEATURE_FAN_CONTROL_BIT))
+		return AMD_FAN_CTRL_MANUAL;
+	else
+		return AMD_FAN_CTRL_AUTO;
+}
+
+static int
+smu_v11_0_get_fan_speed_percent(struct smu_context *smu,
+					   uint32_t *speed)
+{
+	int ret = 0;
+	uint32_t percent = 0;
+	uint32_t current_rpm;
+	PPTable_t *pptable = smu->smu_table.driver_pptable;
+
+	ret = smu_v11_0_get_current_rpm(smu, &current_rpm);
+	percent = current_rpm * 100 / pptable->FanMaximumRpm;
+	*speed = percent > 100 ? 100 : percent;
+
+	return ret;
+}
+
+static int
+smu_v11_0_smc_fan_control(struct smu_context *smu, bool start)
+{
+	int ret = 0;
+
+	if (smu_feature_is_supported(smu, FEATURE_FAN_CONTROL_BIT))
+		return 0;
+
+	ret = smu_feature_set_enabled(smu, FEATURE_FAN_CONTROL_BIT, start);
+	if (ret)
+		pr_err("[%s]%s smc FAN CONTROL feature failed!",
+		       __func__, (start ? "Start" : "Stop"));
+
+	return ret;
+}
+
+static int
+smu_v11_0_set_fan_static_mode(struct smu_context *smu, uint32_t mode)
+{
+	struct amdgpu_device *adev = smu->adev;
+
+	WREG32_SOC15(THM, 0, mmCG_FDO_CTRL2,
+		     REG_SET_FIELD(RREG32_SOC15(THM, 0, mmCG_FDO_CTRL2),
+				   CG_FDO_CTRL2, TMIN, 0));
+	WREG32_SOC15(THM, 0, mmCG_FDO_CTRL2,
+		     REG_SET_FIELD(RREG32_SOC15(THM, 0, mmCG_FDO_CTRL2),
+				   CG_FDO_CTRL2, FDO_PWM_MODE, mode));
+
+	return 0;
+}
+
+static int
+smu_v11_0_set_fan_speed_percent(struct smu_context *smu, uint32_t speed)
+{
+	struct amdgpu_device *adev = smu->adev;
+	uint32_t duty100;
+	uint32_t duty;
+	uint64_t tmp64;
+	bool stop = 0;
+
+	if (speed > 100)
+		speed = 100;
+
+	if (smu_v11_0_smc_fan_control(smu, stop))
+		return -EINVAL;
+	duty100 = REG_GET_FIELD(RREG32_SOC15(THM, 0, mmCG_FDO_CTRL1),
+				CG_FDO_CTRL1, FMAX_DUTY100);
+	if (!duty100)
+		return -EINVAL;
+
+	tmp64 = (uint64_t)speed * duty100;
+	do_div(tmp64, 100);
+	duty = (uint32_t)tmp64;
+
+	WREG32_SOC15(THM, 0, mmCG_FDO_CTRL0,
+		     REG_SET_FIELD(RREG32_SOC15(THM, 0, mmCG_FDO_CTRL0),
+				   CG_FDO_CTRL0, FDO_STATIC_DUTY, duty));
+
+	return smu_v11_0_set_fan_static_mode(smu, FDO_PWM_MODE_STATIC);
+}
+
+static int
+smu_v11_0_set_fan_control_mode(struct smu_context *smu,
+			       uint32_t mode)
+{
+	int ret = 0;
+	bool start = 1;
+	bool stop  = 0;
+
+	switch (mode) {
+	case AMD_FAN_CTRL_NONE:
+		ret = smu_v11_0_set_fan_speed_percent(smu, 100);
+		break;
+	case AMD_FAN_CTRL_MANUAL:
+		ret = smu_v11_0_smc_fan_control(smu, stop);
+		break;
+	case AMD_FAN_CTRL_AUTO:
+		ret = smu_v11_0_smc_fan_control(smu, start);
+		break;
+	default:
+		break;
+	}
+
+	if (ret) {
+		pr_err("[%s]Set fan control mode failed!", __func__);
+		return -EINVAL;
+	}
+
+	return ret;
+}
+
+static int smu_v11_0_set_fan_speed_rpm(struct smu_context *smu,
+				       uint32_t speed)
+{
+	struct amdgpu_device *adev = smu->adev;
+	int ret;
+	uint32_t tach_period, crystal_clock_freq;
+	bool stop = 0;
+
+	if (!speed)
+		return -EINVAL;
+
+	mutex_lock(&(smu->mutex));
+	ret = smu_v11_0_smc_fan_control(smu, stop);
+	if (ret)
+		goto set_fan_speed_rpm_failed;
+
+	crystal_clock_freq = amdgpu_asic_get_xclk(adev);
+	tach_period = 60 * crystal_clock_freq * 10000 / (8 * speed);
+	WREG32_SOC15(THM, 0, mmCG_TACH_CTRL,
+		     REG_SET_FIELD(RREG32_SOC15(THM, 0, mmCG_TACH_CTRL),
+				   CG_TACH_CTRL, TARGET_PERIOD,
+				   tach_period));
+
+	ret = smu_v11_0_set_fan_static_mode(smu, FDO_PWM_MODE_STATIC_RPM);
+
+set_fan_speed_rpm_failed:
+	mutex_unlock(&(smu->mutex));
+	return ret;
+}
+
+static int smu_v11_0_set_xgmi_pstate(struct smu_context *smu,
+				     uint32_t pstate)
+{
+	int ret = 0;
+	mutex_lock(&(smu->mutex));
+	ret = smu_send_smc_msg_with_param(smu,
+					  SMU_MSG_SetXgmiMode,
+					  pstate ? XGMI_STATE_D0 : XGMI_STATE_D3);
+	mutex_unlock(&(smu->mutex));
+	return ret;
+}
+
+static const struct smu_funcs smu_v11_0_funcs = {
+	.init_microcode = smu_v11_0_init_microcode,
+	.load_microcode = smu_v11_0_load_microcode,
+	.check_fw_status = smu_v11_0_check_fw_status,
+	.check_fw_version = smu_v11_0_check_fw_version,
+	.send_smc_msg = smu_v11_0_send_msg,
+	.send_smc_msg_with_param = smu_v11_0_send_msg_with_param,
+	.read_smc_arg = smu_v11_0_read_arg,
+	.read_pptable_from_vbios = smu_v11_0_read_pptable_from_vbios,
+	.init_smc_tables = smu_v11_0_init_smc_tables,
+	.fini_smc_tables = smu_v11_0_fini_smc_tables,
+	.init_power = smu_v11_0_init_power,
+	.fini_power = smu_v11_0_fini_power,
+	.get_vbios_bootup_values = smu_v11_0_get_vbios_bootup_values,
+	.get_clk_info_from_vbios = smu_v11_0_get_clk_info_from_vbios,
+	.notify_memory_pool_location = smu_v11_0_notify_memory_pool_location,
+	.check_pptable = smu_v11_0_check_pptable,
+	.parse_pptable = smu_v11_0_parse_pptable,
+	.populate_smc_pptable = smu_v11_0_populate_smc_pptable,
+	.write_pptable = smu_v11_0_write_pptable,
+	.write_watermarks_table = smu_v11_0_write_watermarks_table,
+	.set_min_dcef_deep_sleep = smu_v11_0_set_min_dcef_deep_sleep,
+	.set_tool_table_location = smu_v11_0_set_tool_table_location,
+	.init_display = smu_v11_0_init_display,
+	.set_allowed_mask = smu_v11_0_set_allowed_mask,
+	.get_enabled_mask = smu_v11_0_get_enabled_mask,
+	.is_dpm_running = smu_v11_0_is_dpm_running,
+	.system_features_control = smu_v11_0_system_features_control,
+	.update_feature_enable_state = smu_v11_0_update_feature_enable_state,
+	.notify_display_change = smu_v11_0_notify_display_change,
+	.get_power_limit = smu_v11_0_get_power_limit,
+	.set_power_limit = smu_v11_0_set_power_limit,
+	.get_current_clk_freq = smu_v11_0_get_current_clk_freq,
+	.init_max_sustainable_clocks = smu_v11_0_init_max_sustainable_clocks,
+	.start_thermal_control = smu_v11_0_start_thermal_control,
+	.read_sensor = smu_v11_0_read_sensor,
+	.set_deep_sleep_dcefclk = smu_v11_0_set_deep_sleep_dcefclk,
+	.display_clock_voltage_request = smu_v11_0_display_clock_voltage_request,
+	.set_watermarks_for_clock_ranges = smu_v11_0_set_watermarks_for_clock_ranges,
+	.get_sclk = smu_v11_0_dpm_get_sclk,
+	.get_mclk = smu_v11_0_dpm_get_mclk,
+	.set_od8_default_settings = smu_v11_0_set_od8_default_settings,
+	.conv_power_profile_to_pplib_workload = smu_v11_0_conv_power_profile_to_pplib_workload,
+	.get_power_profile_mode = smu_v11_0_get_power_profile_mode,
+	.set_power_profile_mode = smu_v11_0_set_power_profile_mode,
+	.update_od8_settings = smu_v11_0_update_od8_settings,
+	.dpm_set_uvd_enable = smu_v11_0_dpm_set_uvd_enable,
+	.dpm_set_vce_enable = smu_v11_0_dpm_set_vce_enable,
+	.get_current_rpm = smu_v11_0_get_current_rpm,
+	.get_fan_control_mode = smu_v11_0_get_fan_control_mode,
+	.set_fan_control_mode = smu_v11_0_set_fan_control_mode,
+	.get_fan_speed_percent = smu_v11_0_get_fan_speed_percent,
+	.set_fan_speed_percent = smu_v11_0_set_fan_speed_percent,
+	.set_fan_speed_rpm = smu_v11_0_set_fan_speed_rpm,
+	.set_xgmi_pstate = smu_v11_0_set_xgmi_pstate,
+};
+
+void smu_v11_0_set_smu_funcs(struct smu_context *smu)
+{
+	struct amdgpu_device *adev = smu->adev;
+
+	smu->funcs = &smu_v11_0_funcs;
+	switch (adev->asic_type) {
+	case CHIP_VEGA20:
+		vega20_set_ppt_funcs(smu);
+		break;
+	default:
+		pr_warn("Unknown asic for smu11\n");
+	}
+}
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smu10_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/smu10_smumgr.c
index d111dd4e03d7..6d11076a79ba 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/smu10_smumgr.c
+++ b/drivers/gpu/drm/amd/powerplay/smumgr/smu10_smumgr.c
@@ -212,6 +212,10 @@ static int smu10_start_smu(struct pp_hwmgr *hwmgr)
 	hwmgr->smu_version = smu10_read_arg_from_smc(hwmgr);
 	adev->pm.fw_version = hwmgr->smu_version >> 8;
 
+	if (adev->rev_id < 0x8 && adev->pdev->device != 0x15d8 &&
+	    adev->pm.fw_version < 0x1e45)
+		adev->pm.pp_feature &= ~PP_GFXOFF_MASK;
+
 	if (smu10_verify_smc_interface(hwmgr))
 		return -EINVAL;
 
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.c
index ba00744c3413..f301a73f6df1 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.c
+++ b/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.c
@@ -367,6 +367,26 @@ static int vega20_set_tools_address(struct pp_hwmgr *hwmgr)
 	return ret;
 }
 
+int vega20_set_pptable_driver_address(struct pp_hwmgr *hwmgr)
+{
+	struct vega20_smumgr *priv =
+			(struct vega20_smumgr *)(hwmgr->smu_backend);
+	int ret = 0;
+
+	PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc_with_parameter(hwmgr,
+			PPSMC_MSG_SetDriverDramAddrHigh,
+			upper_32_bits(priv->smu_tables.entry[TABLE_PPTABLE].mc_addr))) == 0,
+			"[SetPPtabeDriverAddress] Attempt to Set Dram Addr High Failed!",
+			return ret);
+	PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc_with_parameter(hwmgr,
+			PPSMC_MSG_SetDriverDramAddrLow,
+			lower_32_bits(priv->smu_tables.entry[TABLE_PPTABLE].mc_addr))) == 0,
+			"[SetPPtabeDriverAddress] Attempt to Set Dram Addr Low Failed!",
+			return ret);
+
+	return ret;
+}
+
 static int vega20_smu_init(struct pp_hwmgr *hwmgr)
 {
 	struct vega20_smumgr *priv;
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.h b/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.h
index 77349c3f0162..ec953ab13e87 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.h
+++ b/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.h
@@ -55,6 +55,7 @@ int vega20_set_activity_monitor_coeff(struct pp_hwmgr *hwmgr,
 		uint8_t *table, uint16_t workload_type);
 int vega20_get_activity_monitor_coeff(struct pp_hwmgr *hwmgr,
 		uint8_t *table, uint16_t workload_type);
+int vega20_set_pptable_driver_address(struct pp_hwmgr *hwmgr);
 
 #endif
 
diff --git a/drivers/gpu/drm/amd/powerplay/vega20_ppt.c b/drivers/gpu/drm/amd/powerplay/vega20_ppt.c
new file mode 100644
index 000000000000..8fafcbdb1dfd
--- /dev/null
+++ b/drivers/gpu/drm/amd/powerplay/vega20_ppt.c
@@ -0,0 +1,2413 @@
+/*
+ * Copyright 2019 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#include "pp_debug.h"
+#include <linux/firmware.h>
+#include "amdgpu.h"
+#include "amdgpu_smu.h"
+#include "atomfirmware.h"
+#include "amdgpu_atomfirmware.h"
+#include "smu_v11_0.h"
+#include "smu11_driver_if.h"
+#include "soc15_common.h"
+#include "atom.h"
+#include "power_state.h"
+#include "vega20_ppt.h"
+#include "vega20_pptable.h"
+#include "vega20_ppsmc.h"
+#include "nbio/nbio_7_4_sh_mask.h"
+
+#define smnPCIE_LC_SPEED_CNTL			0x11140290
+#define smnPCIE_LC_LINK_WIDTH_CNTL		0x11140288
+
+#define MSG_MAP(msg) \
+	[SMU_MSG_##msg] = PPSMC_MSG_##msg
+
+static int vega20_message_map[SMU_MSG_MAX_COUNT] = {
+	MSG_MAP(TestMessage),
+	MSG_MAP(GetSmuVersion),
+	MSG_MAP(GetDriverIfVersion),
+	MSG_MAP(SetAllowedFeaturesMaskLow),
+	MSG_MAP(SetAllowedFeaturesMaskHigh),
+	MSG_MAP(EnableAllSmuFeatures),
+	MSG_MAP(DisableAllSmuFeatures),
+	MSG_MAP(EnableSmuFeaturesLow),
+	MSG_MAP(EnableSmuFeaturesHigh),
+	MSG_MAP(DisableSmuFeaturesLow),
+	MSG_MAP(DisableSmuFeaturesHigh),
+	MSG_MAP(GetEnabledSmuFeaturesLow),
+	MSG_MAP(GetEnabledSmuFeaturesHigh),
+	MSG_MAP(SetWorkloadMask),
+	MSG_MAP(SetPptLimit),
+	MSG_MAP(SetDriverDramAddrHigh),
+	MSG_MAP(SetDriverDramAddrLow),
+	MSG_MAP(SetToolsDramAddrHigh),
+	MSG_MAP(SetToolsDramAddrLow),
+	MSG_MAP(TransferTableSmu2Dram),
+	MSG_MAP(TransferTableDram2Smu),
+	MSG_MAP(UseDefaultPPTable),
+	MSG_MAP(UseBackupPPTable),
+	MSG_MAP(RunBtc),
+	MSG_MAP(RequestI2CBus),
+	MSG_MAP(ReleaseI2CBus),
+	MSG_MAP(SetFloorSocVoltage),
+	MSG_MAP(SoftReset),
+	MSG_MAP(StartBacoMonitor),
+	MSG_MAP(CancelBacoMonitor),
+	MSG_MAP(EnterBaco),
+	MSG_MAP(SetSoftMinByFreq),
+	MSG_MAP(SetSoftMaxByFreq),
+	MSG_MAP(SetHardMinByFreq),
+	MSG_MAP(SetHardMaxByFreq),
+	MSG_MAP(GetMinDpmFreq),
+	MSG_MAP(GetMaxDpmFreq),
+	MSG_MAP(GetDpmFreqByIndex),
+	MSG_MAP(GetDpmClockFreq),
+	MSG_MAP(GetSsVoltageByDpm),
+	MSG_MAP(SetMemoryChannelConfig),
+	MSG_MAP(SetGeminiMode),
+	MSG_MAP(SetGeminiApertureHigh),
+	MSG_MAP(SetGeminiApertureLow),
+	MSG_MAP(SetMinLinkDpmByIndex),
+	MSG_MAP(OverridePcieParameters),
+	MSG_MAP(OverDriveSetPercentage),
+	MSG_MAP(SetMinDeepSleepDcefclk),
+	MSG_MAP(ReenableAcDcInterrupt),
+	MSG_MAP(NotifyPowerSource),
+	MSG_MAP(SetUclkFastSwitch),
+	MSG_MAP(SetUclkDownHyst),
+	MSG_MAP(GetCurrentRpm),
+	MSG_MAP(SetVideoFps),
+	MSG_MAP(SetTjMax),
+	MSG_MAP(SetFanTemperatureTarget),
+	MSG_MAP(PrepareMp1ForUnload),
+	MSG_MAP(DramLogSetDramAddrHigh),
+	MSG_MAP(DramLogSetDramAddrLow),
+	MSG_MAP(DramLogSetDramSize),
+	MSG_MAP(SetFanMaxRpm),
+	MSG_MAP(SetFanMinPwm),
+	MSG_MAP(ConfigureGfxDidt),
+	MSG_MAP(NumOfDisplays),
+	MSG_MAP(RemoveMargins),
+	MSG_MAP(ReadSerialNumTop32),
+	MSG_MAP(ReadSerialNumBottom32),
+	MSG_MAP(SetSystemVirtualDramAddrHigh),
+	MSG_MAP(SetSystemVirtualDramAddrLow),
+	MSG_MAP(WaflTest),
+	MSG_MAP(SetFclkGfxClkRatio),
+	MSG_MAP(AllowGfxOff),
+	MSG_MAP(DisallowGfxOff),
+	MSG_MAP(GetPptLimit),
+	MSG_MAP(GetDcModeMaxDpmFreq),
+	MSG_MAP(GetDebugData),
+	MSG_MAP(SetXgmiMode),
+	MSG_MAP(RunAfllBtc),
+	MSG_MAP(ExitBaco),
+	MSG_MAP(PrepareMp1ForReset),
+	MSG_MAP(PrepareMp1ForShutdown),
+	MSG_MAP(SetMGpuFanBoostLimitRpm),
+	MSG_MAP(GetAVFSVoltageByDpm),
+};
+
+static int vega20_get_smu_msg_index(struct smu_context *smc, uint32_t index)
+{
+	int val;
+
+	if (index >= SMU_MSG_MAX_COUNT)
+		return -EINVAL;
+
+	val = vega20_message_map[index];
+	if (val > PPSMC_Message_Count)
+		return -EINVAL;
+
+	return val;
+}
+
+static int vega20_allocate_dpm_context(struct smu_context *smu)
+{
+	struct smu_dpm_context *smu_dpm = &smu->smu_dpm;
+
+	if (smu_dpm->dpm_context)
+		return -EINVAL;
+
+	smu_dpm->dpm_context = kzalloc(sizeof(struct vega20_dpm_table),
+				       GFP_KERNEL);
+	if (!smu_dpm->dpm_context)
+		return -ENOMEM;
+
+	if (smu_dpm->golden_dpm_context)
+		return -EINVAL;
+
+	smu_dpm->golden_dpm_context = kzalloc(sizeof(struct vega20_dpm_table),
+					      GFP_KERNEL);
+	if (!smu_dpm->golden_dpm_context)
+		return -ENOMEM;
+
+	smu_dpm->dpm_context_size = sizeof(struct vega20_dpm_table);
+
+	smu_dpm->dpm_current_power_state = kzalloc(sizeof(struct smu_power_state),
+				       GFP_KERNEL);
+	if (!smu_dpm->dpm_current_power_state)
+		return -ENOMEM;
+
+	smu_dpm->dpm_request_power_state = kzalloc(sizeof(struct smu_power_state),
+				       GFP_KERNEL);
+	if (!smu_dpm->dpm_request_power_state)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static int vega20_setup_od8_information(struct smu_context *smu)
+{
+	ATOM_Vega20_POWERPLAYTABLE *powerplay_table = NULL;
+	struct smu_table_context *table_context = &smu->smu_table;
+
+	uint32_t od_feature_count, od_feature_array_size,
+		 od_setting_count, od_setting_array_size;
+
+	if (!table_context->power_play_table)
+		return -EINVAL;
+
+	powerplay_table = table_context->power_play_table;
+
+	if (powerplay_table->OverDrive8Table.ucODTableRevision == 1) {
+		/* Setup correct ODFeatureCount, and store ODFeatureArray from
+		 * powerplay table to od_feature_capabilities */
+		od_feature_count =
+			(le32_to_cpu(powerplay_table->OverDrive8Table.ODFeatureCount) >
+			 ATOM_VEGA20_ODFEATURE_COUNT) ?
+			ATOM_VEGA20_ODFEATURE_COUNT :
+			le32_to_cpu(powerplay_table->OverDrive8Table.ODFeatureCount);
+
+		od_feature_array_size = sizeof(uint8_t) * od_feature_count;
+
+		if (table_context->od_feature_capabilities)
+			return -EINVAL;
+
+		table_context->od_feature_capabilities = kmemdup(&powerplay_table->OverDrive8Table.ODFeatureCapabilities,
+								 od_feature_array_size,
+								 GFP_KERNEL);
+		if (!table_context->od_feature_capabilities)
+			return -ENOMEM;
+
+		/* Setup correct ODSettingCount, and store ODSettingArray from
+		 * powerplay table to od_settings_max and od_setting_min */
+		od_setting_count =
+			(le32_to_cpu(powerplay_table->OverDrive8Table.ODSettingCount) >
+			 ATOM_VEGA20_ODSETTING_COUNT) ?
+			ATOM_VEGA20_ODSETTING_COUNT :
+			le32_to_cpu(powerplay_table->OverDrive8Table.ODSettingCount);
+
+		od_setting_array_size = sizeof(uint32_t) * od_setting_count;
+
+		if (table_context->od_settings_max)
+			return -EINVAL;
+
+		table_context->od_settings_max = kmemdup(&powerplay_table->OverDrive8Table.ODSettingsMax,
+							 od_setting_array_size,
+							 GFP_KERNEL);
+
+		if (!table_context->od_settings_max) {
+			kfree(table_context->od_feature_capabilities);
+			table_context->od_feature_capabilities = NULL;
+			return -ENOMEM;
+		}
+
+		if (table_context->od_settings_min)
+			return -EINVAL;
+
+		table_context->od_settings_min = kmemdup(&powerplay_table->OverDrive8Table.ODSettingsMin,
+							 od_setting_array_size,
+							 GFP_KERNEL);
+
+		if (!table_context->od_settings_min) {
+			kfree(table_context->od_feature_capabilities);
+			table_context->od_feature_capabilities = NULL;
+			kfree(table_context->od_settings_max);
+			table_context->od_settings_max = NULL;
+			return -ENOMEM;
+		}
+	}
+
+	return 0;
+}
+
+static int vega20_store_powerplay_table(struct smu_context *smu)
+{
+	ATOM_Vega20_POWERPLAYTABLE *powerplay_table = NULL;
+	struct smu_table_context *table_context = &smu->smu_table;
+	int ret;
+
+	if (!table_context->power_play_table)
+		return -EINVAL;
+
+	powerplay_table = table_context->power_play_table;
+
+	memcpy(table_context->driver_pptable, &powerplay_table->smcPPTable,
+	       sizeof(PPTable_t));
+
+	table_context->software_shutdown_temp = powerplay_table->usSoftwareShutdownTemp;
+	table_context->thermal_controller_type = powerplay_table->ucThermalControllerType;
+	table_context->TDPODLimit = le32_to_cpu(powerplay_table->OverDrive8Table.ODSettingsMax[ATOM_VEGA20_ODSETTING_POWERPERCENTAGE]);
+
+	ret = vega20_setup_od8_information(smu);
+
+	return ret;
+}
+
+static int vega20_append_powerplay_table(struct smu_context *smu)
+{
+	struct smu_table_context *table_context = &smu->smu_table;
+	PPTable_t *smc_pptable = table_context->driver_pptable;
+	struct atom_smc_dpm_info_v4_4 *smc_dpm_table;
+	int index, i, ret;
+
+	index = get_index_into_master_table(atom_master_list_of_data_tables_v2_1,
+					   smc_dpm_info);
+
+	ret = smu_get_atom_data_table(smu, index, NULL, NULL, NULL,
+				      (uint8_t **)&smc_dpm_table);
+	if (ret)
+		return ret;
+
+	smc_pptable->MaxVoltageStepGfx = smc_dpm_table->maxvoltagestepgfx;
+	smc_pptable->MaxVoltageStepSoc = smc_dpm_table->maxvoltagestepsoc;
+
+	smc_pptable->VddGfxVrMapping = smc_dpm_table->vddgfxvrmapping;
+	smc_pptable->VddSocVrMapping = smc_dpm_table->vddsocvrmapping;
+	smc_pptable->VddMem0VrMapping = smc_dpm_table->vddmem0vrmapping;
+	smc_pptable->VddMem1VrMapping = smc_dpm_table->vddmem1vrmapping;
+
+	smc_pptable->GfxUlvPhaseSheddingMask = smc_dpm_table->gfxulvphasesheddingmask;
+	smc_pptable->SocUlvPhaseSheddingMask = smc_dpm_table->soculvphasesheddingmask;
+	smc_pptable->ExternalSensorPresent = smc_dpm_table->externalsensorpresent;
+
+	smc_pptable->GfxMaxCurrent = smc_dpm_table->gfxmaxcurrent;
+	smc_pptable->GfxOffset = smc_dpm_table->gfxoffset;
+	smc_pptable->Padding_TelemetryGfx = smc_dpm_table->padding_telemetrygfx;
+
+	smc_pptable->SocMaxCurrent = smc_dpm_table->socmaxcurrent;
+	smc_pptable->SocOffset = smc_dpm_table->socoffset;
+	smc_pptable->Padding_TelemetrySoc = smc_dpm_table->padding_telemetrysoc;
+
+	smc_pptable->Mem0MaxCurrent = smc_dpm_table->mem0maxcurrent;
+	smc_pptable->Mem0Offset = smc_dpm_table->mem0offset;
+	smc_pptable->Padding_TelemetryMem0 = smc_dpm_table->padding_telemetrymem0;
+
+	smc_pptable->Mem1MaxCurrent = smc_dpm_table->mem1maxcurrent;
+	smc_pptable->Mem1Offset = smc_dpm_table->mem1offset;
+	smc_pptable->Padding_TelemetryMem1 = smc_dpm_table->padding_telemetrymem1;
+
+	smc_pptable->AcDcGpio = smc_dpm_table->acdcgpio;
+	smc_pptable->AcDcPolarity = smc_dpm_table->acdcpolarity;
+	smc_pptable->VR0HotGpio = smc_dpm_table->vr0hotgpio;
+	smc_pptable->VR0HotPolarity = smc_dpm_table->vr0hotpolarity;
+
+	smc_pptable->VR1HotGpio = smc_dpm_table->vr1hotgpio;
+	smc_pptable->VR1HotPolarity = smc_dpm_table->vr1hotpolarity;
+	smc_pptable->Padding1 = smc_dpm_table->padding1;
+	smc_pptable->Padding2 = smc_dpm_table->padding2;
+
+	smc_pptable->LedPin0 = smc_dpm_table->ledpin0;
+	smc_pptable->LedPin1 = smc_dpm_table->ledpin1;
+	smc_pptable->LedPin2 = smc_dpm_table->ledpin2;
+
+	smc_pptable->PllGfxclkSpreadEnabled = smc_dpm_table->pllgfxclkspreadenabled;
+	smc_pptable->PllGfxclkSpreadPercent = smc_dpm_table->pllgfxclkspreadpercent;
+	smc_pptable->PllGfxclkSpreadFreq = smc_dpm_table->pllgfxclkspreadfreq;
+
+	smc_pptable->UclkSpreadEnabled = 0;
+	smc_pptable->UclkSpreadPercent = smc_dpm_table->uclkspreadpercent;
+	smc_pptable->UclkSpreadFreq = smc_dpm_table->uclkspreadfreq;
+
+	smc_pptable->FclkSpreadEnabled = smc_dpm_table->fclkspreadenabled;
+	smc_pptable->FclkSpreadPercent = smc_dpm_table->fclkspreadpercent;
+	smc_pptable->FclkSpreadFreq = smc_dpm_table->fclkspreadfreq;
+
+	smc_pptable->FllGfxclkSpreadEnabled = smc_dpm_table->fllgfxclkspreadenabled;
+	smc_pptable->FllGfxclkSpreadPercent = smc_dpm_table->fllgfxclkspreadpercent;
+	smc_pptable->FllGfxclkSpreadFreq = smc_dpm_table->fllgfxclkspreadfreq;
+
+	for (i = 0; i < I2C_CONTROLLER_NAME_COUNT; i++) {
+		smc_pptable->I2cControllers[i].Enabled =
+			smc_dpm_table->i2ccontrollers[i].enabled;
+		smc_pptable->I2cControllers[i].SlaveAddress =
+			smc_dpm_table->i2ccontrollers[i].slaveaddress;
+		smc_pptable->I2cControllers[i].ControllerPort =
+			smc_dpm_table->i2ccontrollers[i].controllerport;
+		smc_pptable->I2cControllers[i].ThermalThrottler =
+			smc_dpm_table->i2ccontrollers[i].thermalthrottler;
+		smc_pptable->I2cControllers[i].I2cProtocol =
+			smc_dpm_table->i2ccontrollers[i].i2cprotocol;
+		smc_pptable->I2cControllers[i].I2cSpeed =
+			smc_dpm_table->i2ccontrollers[i].i2cspeed;
+	}
+
+	return 0;
+}
+
+static int vega20_check_powerplay_table(struct smu_context *smu)
+{
+	ATOM_Vega20_POWERPLAYTABLE *powerplay_table = NULL;
+	struct smu_table_context *table_context = &smu->smu_table;
+
+	powerplay_table = table_context->power_play_table;
+
+	if (powerplay_table->sHeader.format_revision < ATOM_VEGA20_TABLE_REVISION_VEGA20) {
+		pr_err("Unsupported PPTable format!");
+		return -EINVAL;
+	}
+
+	if (!powerplay_table->sHeader.structuresize) {
+		pr_err("Invalid PowerPlay Table!");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int vega20_run_btc_afll(struct smu_context *smu)
+{
+	return smu_send_smc_msg(smu, SMU_MSG_RunAfllBtc);
+}
+
+static int
+vega20_get_unallowed_feature_mask(struct smu_context *smu,
+				  uint32_t *feature_mask, uint32_t num)
+{
+	if (num > 2)
+		return -EINVAL;
+
+	feature_mask[0] = 0xE0041C00;
+	feature_mask[1] = 0xFFFFFFFE; /* bit32~bit63 is Unsupported */
+
+	return 0;
+}
+
+static enum
+amd_pm_state_type vega20_get_current_power_state(struct smu_context *smu)
+{
+	enum amd_pm_state_type pm_type;
+	struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm);
+
+	if (!smu_dpm_ctx->dpm_context ||
+	    !smu_dpm_ctx->dpm_current_power_state)
+		return -EINVAL;
+
+	mutex_lock(&(smu->mutex));
+	switch (smu_dpm_ctx->dpm_current_power_state->classification.ui_label) {
+	case SMU_STATE_UI_LABEL_BATTERY:
+		pm_type = POWER_STATE_TYPE_BATTERY;
+		break;
+	case SMU_STATE_UI_LABEL_BALLANCED:
+		pm_type = POWER_STATE_TYPE_BALANCED;
+		break;
+	case SMU_STATE_UI_LABEL_PERFORMANCE:
+		pm_type = POWER_STATE_TYPE_PERFORMANCE;
+		break;
+	default:
+		if (smu_dpm_ctx->dpm_current_power_state->classification.flags & SMU_STATE_CLASSIFICATION_FLAG_BOOT)
+			pm_type = POWER_STATE_TYPE_INTERNAL_BOOT;
+		else
+			pm_type = POWER_STATE_TYPE_DEFAULT;
+		break;
+	}
+	mutex_unlock(&(smu->mutex));
+
+	return pm_type;
+}
+
+static int
+vega20_set_single_dpm_table(struct smu_context *smu,
+			    struct vega20_single_dpm_table *single_dpm_table,
+			    PPCLK_e clk_id)
+{
+	int ret = 0;
+	uint32_t i, num_of_levels = 0, clk;
+
+	ret = smu_send_smc_msg_with_param(smu,
+			SMU_MSG_GetDpmFreqByIndex,
+			(clk_id << 16 | 0xFF));
+	if (ret) {
+		pr_err("[GetNumOfDpmLevel] failed to get dpm levels!");
+		return ret;
+	}
+
+	smu_read_smc_arg(smu, &num_of_levels);
+	if (!num_of_levels) {
+		pr_err("[GetNumOfDpmLevel] number of clk levels is invalid!");
+		return -EINVAL;
+	}
+
+	single_dpm_table->count = num_of_levels;
+
+	for (i = 0; i < num_of_levels; i++) {
+		ret = smu_send_smc_msg_with_param(smu,
+				SMU_MSG_GetDpmFreqByIndex,
+				(clk_id << 16 | i));
+		if (ret) {
+			pr_err("[GetDpmFreqByIndex] failed to get dpm freq by index!");
+			return ret;
+		}
+		smu_read_smc_arg(smu, &clk);
+		if (!clk) {
+			pr_err("[GetDpmFreqByIndex] clk value is invalid!");
+			return -EINVAL;
+		}
+		single_dpm_table->dpm_levels[i].value = clk;
+		single_dpm_table->dpm_levels[i].enabled = true;
+	}
+	return 0;
+}
+
+static void vega20_init_single_dpm_state(struct vega20_dpm_state *dpm_state)
+{
+	dpm_state->soft_min_level = 0x0;
+	dpm_state->soft_max_level = 0xffff;
+        dpm_state->hard_min_level = 0x0;
+        dpm_state->hard_max_level = 0xffff;
+}
+
+static int vega20_set_default_dpm_table(struct smu_context *smu)
+{
+	int ret;
+
+	struct smu_dpm_context *smu_dpm = &smu->smu_dpm;
+	struct vega20_dpm_table *dpm_table = NULL;
+	struct vega20_single_dpm_table *single_dpm_table;
+
+	dpm_table = smu_dpm->dpm_context;
+
+	/* socclk */
+	single_dpm_table = &(dpm_table->soc_table);
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_SOCCLK_BIT)) {
+		ret = vega20_set_single_dpm_table(smu, single_dpm_table,
+						  PPCLK_SOCCLK);
+		if (ret) {
+			pr_err("[SetupDefaultDpmTable] failed to get socclk dpm levels!");
+			return ret;
+		}
+	} else {
+		single_dpm_table->count = 1;
+		single_dpm_table->dpm_levels[0].value = smu->smu_table.boot_values.socclk / 100;
+	}
+	vega20_init_single_dpm_state(&(single_dpm_table->dpm_state));
+
+	/* gfxclk */
+	single_dpm_table = &(dpm_table->gfx_table);
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_GFXCLK_BIT)) {
+		ret = vega20_set_single_dpm_table(smu, single_dpm_table,
+						  PPCLK_GFXCLK);
+		if (ret) {
+			pr_err("[SetupDefaultDpmTable] failed to get gfxclk dpm levels!");
+			return ret;
+		}
+	} else {
+		single_dpm_table->count = 1;
+		single_dpm_table->dpm_levels[0].value = smu->smu_table.boot_values.gfxclk / 100;
+	}
+	vega20_init_single_dpm_state(&(single_dpm_table->dpm_state));
+
+	/* memclk */
+	single_dpm_table = &(dpm_table->mem_table);
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_UCLK_BIT)) {
+		ret = vega20_set_single_dpm_table(smu, single_dpm_table,
+						  PPCLK_UCLK);
+		if (ret) {
+			pr_err("[SetupDefaultDpmTable] failed to get memclk dpm levels!");
+			return ret;
+		}
+	} else {
+		single_dpm_table->count = 1;
+		single_dpm_table->dpm_levels[0].value = smu->smu_table.boot_values.uclk / 100;
+	}
+	vega20_init_single_dpm_state(&(single_dpm_table->dpm_state));
+
+	/* eclk */
+	single_dpm_table = &(dpm_table->eclk_table);
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_VCE_BIT)) {
+		ret = vega20_set_single_dpm_table(smu, single_dpm_table, PPCLK_ECLK);
+		if (ret) {
+			pr_err("[SetupDefaultDpmTable] failed to get eclk dpm levels!");
+			return ret;
+		}
+	} else {
+		single_dpm_table->count = 1;
+		single_dpm_table->dpm_levels[0].value = smu->smu_table.boot_values.eclk / 100;
+	}
+	vega20_init_single_dpm_state(&(single_dpm_table->dpm_state));
+
+	/* vclk */
+	single_dpm_table = &(dpm_table->vclk_table);
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_UVD_BIT)) {
+		ret = vega20_set_single_dpm_table(smu, single_dpm_table, PPCLK_VCLK);
+		if (ret) {
+			pr_err("[SetupDefaultDpmTable] failed to get vclk dpm levels!");
+			return ret;
+		}
+	} else {
+		single_dpm_table->count = 1;
+		single_dpm_table->dpm_levels[0].value = smu->smu_table.boot_values.vclk / 100;
+	}
+	vega20_init_single_dpm_state(&(single_dpm_table->dpm_state));
+
+	/* dclk */
+	single_dpm_table = &(dpm_table->dclk_table);
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_UVD_BIT)) {
+		ret = vega20_set_single_dpm_table(smu, single_dpm_table, PPCLK_DCLK);
+		if (ret) {
+			pr_err("[SetupDefaultDpmTable] failed to get dclk dpm levels!");
+			return ret;
+		}
+	} else {
+		single_dpm_table->count = 1;
+		single_dpm_table->dpm_levels[0].value = smu->smu_table.boot_values.dclk / 100;
+	}
+	vega20_init_single_dpm_state(&(single_dpm_table->dpm_state));
+
+	/* dcefclk */
+	single_dpm_table = &(dpm_table->dcef_table);
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_DCEFCLK_BIT)) {
+		ret = vega20_set_single_dpm_table(smu, single_dpm_table,
+						  PPCLK_DCEFCLK);
+		if (ret) {
+			pr_err("[SetupDefaultDpmTable] failed to get dcefclk dpm levels!");
+			return ret;
+		}
+	} else {
+		single_dpm_table->count = 1;
+		single_dpm_table->dpm_levels[0].value = smu->smu_table.boot_values.dcefclk / 100;
+	}
+	vega20_init_single_dpm_state(&(single_dpm_table->dpm_state));
+
+	/* pixclk */
+	single_dpm_table = &(dpm_table->pixel_table);
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_DCEFCLK_BIT)) {
+		ret = vega20_set_single_dpm_table(smu, single_dpm_table,
+						  PPCLK_PIXCLK);
+		if (ret) {
+			pr_err("[SetupDefaultDpmTable] failed to get pixclk dpm levels!");
+			return ret;
+		}
+	} else {
+		single_dpm_table->count = 0;
+	}
+	vega20_init_single_dpm_state(&(single_dpm_table->dpm_state));
+
+	/* dispclk */
+	single_dpm_table = &(dpm_table->display_table);
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_DCEFCLK_BIT)) {
+		ret = vega20_set_single_dpm_table(smu, single_dpm_table,
+						  PPCLK_DISPCLK);
+		if (ret) {
+			pr_err("[SetupDefaultDpmTable] failed to get dispclk dpm levels!");
+			return ret;
+		}
+	} else {
+		single_dpm_table->count = 0;
+	}
+	vega20_init_single_dpm_state(&(single_dpm_table->dpm_state));
+
+	/* phyclk */
+	single_dpm_table = &(dpm_table->phy_table);
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_DCEFCLK_BIT)) {
+		ret = vega20_set_single_dpm_table(smu, single_dpm_table,
+						  PPCLK_PHYCLK);
+		if (ret) {
+			pr_err("[SetupDefaultDpmTable] failed to get phyclk dpm levels!");
+			return ret;
+		}
+	} else {
+		single_dpm_table->count = 0;
+	}
+	vega20_init_single_dpm_state(&(single_dpm_table->dpm_state));
+
+	/* fclk */
+	single_dpm_table = &(dpm_table->fclk_table);
+
+	if (smu_feature_is_enabled(smu,FEATURE_DPM_FCLK_BIT)) {
+		ret = vega20_set_single_dpm_table(smu, single_dpm_table,
+						  PPCLK_FCLK);
+		if (ret) {
+			pr_err("[SetupDefaultDpmTable] failed to get fclk dpm levels!");
+			return ret;
+		}
+	} else {
+		single_dpm_table->count = 0;
+	}
+	vega20_init_single_dpm_state(&(single_dpm_table->dpm_state));
+
+	memcpy(smu_dpm->golden_dpm_context, dpm_table,
+	       sizeof(struct vega20_dpm_table));
+
+	return 0;
+}
+
+static int vega20_populate_umd_state_clk(struct smu_context *smu)
+{
+	struct smu_dpm_context *smu_dpm = &smu->smu_dpm;
+	struct vega20_dpm_table *dpm_table = NULL;
+	struct vega20_single_dpm_table *gfx_table = NULL;
+	struct vega20_single_dpm_table *mem_table = NULL;
+
+	dpm_table = smu_dpm->dpm_context;
+	gfx_table = &(dpm_table->gfx_table);
+	mem_table = &(dpm_table->mem_table);
+
+	smu->pstate_sclk = gfx_table->dpm_levels[0].value;
+	smu->pstate_mclk = mem_table->dpm_levels[0].value;
+
+	if (gfx_table->count > VEGA20_UMD_PSTATE_GFXCLK_LEVEL &&
+	    mem_table->count > VEGA20_UMD_PSTATE_MCLK_LEVEL) {
+		smu->pstate_sclk = gfx_table->dpm_levels[VEGA20_UMD_PSTATE_GFXCLK_LEVEL].value;
+		smu->pstate_mclk = mem_table->dpm_levels[VEGA20_UMD_PSTATE_MCLK_LEVEL].value;
+	}
+
+	smu->pstate_sclk = smu->pstate_sclk * 100;
+	smu->pstate_mclk = smu->pstate_mclk * 100;
+
+	return 0;
+}
+
+static int vega20_get_clk_table(struct smu_context *smu,
+			struct pp_clock_levels_with_latency *clocks,
+			struct vega20_single_dpm_table *dpm_table)
+{
+	int i, count;
+
+	count = (dpm_table->count > MAX_NUM_CLOCKS) ? MAX_NUM_CLOCKS : dpm_table->count;
+	clocks->num_levels = count;
+
+	for (i = 0; i < count; i++) {
+		clocks->data[i].clocks_in_khz =
+			dpm_table->dpm_levels[i].value * 1000;
+		clocks->data[i].latency_in_us = 0;
+	}
+
+	return 0;
+}
+
+static int vega20_print_clk_levels(struct smu_context *smu,
+			enum pp_clock_type type, char *buf)
+{
+	int i, now, size = 0;
+	int ret = 0;
+	uint32_t gen_speed, lane_width;
+	struct amdgpu_device *adev = smu->adev;
+	struct pp_clock_levels_with_latency clocks;
+	struct vega20_single_dpm_table *single_dpm_table;
+	struct smu_table_context *table_context = &smu->smu_table;
+	struct smu_dpm_context *smu_dpm = &smu->smu_dpm;
+	struct vega20_dpm_table *dpm_table = NULL;
+	struct vega20_od8_settings *od8_settings =
+		(struct vega20_od8_settings *)table_context->od8_settings;
+	OverDriveTable_t *od_table =
+		(OverDriveTable_t *)(table_context->overdrive_table);
+	PPTable_t *pptable = (PPTable_t *)table_context->driver_pptable;
+
+	dpm_table = smu_dpm->dpm_context;
+
+	switch (type) {
+	case PP_SCLK:
+		ret = smu_get_current_clk_freq(smu, PPCLK_GFXCLK, &now);
+		if (ret) {
+			pr_err("Attempt to get current gfx clk Failed!");
+			return ret;
+		}
+
+		single_dpm_table = &(dpm_table->gfx_table);
+		ret = vega20_get_clk_table(smu, &clocks, single_dpm_table);
+		if (ret) {
+			pr_err("Attempt to get gfx clk levels Failed!");
+			return ret;
+		}
+
+		for (i = 0; i < clocks.num_levels; i++)
+			size += sprintf(buf + size, "%d: %uMhz %s\n", i,
+					clocks.data[i].clocks_in_khz / 1000,
+					(clocks.data[i].clocks_in_khz == now * 10)
+					? "*" : "");
+		break;
+
+	case PP_MCLK:
+		ret = smu_get_current_clk_freq(smu, PPCLK_UCLK, &now);
+		if (ret) {
+			pr_err("Attempt to get current mclk Failed!");
+			return ret;
+		}
+
+		single_dpm_table = &(dpm_table->mem_table);
+		ret = vega20_get_clk_table(smu, &clocks, single_dpm_table);
+		if (ret) {
+			pr_err("Attempt to get memory clk levels Failed!");
+			return ret;
+		}
+
+		for (i = 0; i < clocks.num_levels; i++)
+			size += sprintf(buf + size, "%d: %uMhz %s\n",
+				i, clocks.data[i].clocks_in_khz / 1000,
+				(clocks.data[i].clocks_in_khz == now * 10)
+				? "*" : "");
+		break;
+
+	case PP_SOCCLK:
+		ret = smu_get_current_clk_freq(smu, PPCLK_SOCCLK, &now);
+		if (ret) {
+			pr_err("Attempt to get current socclk Failed!");
+			return ret;
+		}
+
+		single_dpm_table = &(dpm_table->soc_table);
+		ret = vega20_get_clk_table(smu, &clocks, single_dpm_table);
+		if (ret) {
+			pr_err("Attempt to get socclk levels Failed!");
+			return ret;
+		}
+
+		for (i = 0; i < clocks.num_levels; i++)
+			size += sprintf(buf + size, "%d: %uMhz %s\n",
+				i, clocks.data[i].clocks_in_khz / 1000,
+				(clocks.data[i].clocks_in_khz == now * 10)
+				? "*" : "");
+		break;
+
+	case PP_FCLK:
+		ret = smu_get_current_clk_freq(smu, PPCLK_FCLK, &now);
+		if (ret) {
+			pr_err("Attempt to get current fclk Failed!");
+			return ret;
+		}
+
+		single_dpm_table = &(dpm_table->fclk_table);
+		for (i = 0; i < single_dpm_table->count; i++)
+			size += sprintf(buf + size, "%d: %uMhz %s\n",
+				i, single_dpm_table->dpm_levels[i].value,
+				(single_dpm_table->dpm_levels[i].value == now / 100)
+				? "*" : "");
+		break;
+
+	case PP_DCEFCLK:
+		ret = smu_get_current_clk_freq(smu, PPCLK_DCEFCLK, &now);
+		if (ret) {
+			pr_err("Attempt to get current dcefclk Failed!");
+			return ret;
+		}
+
+		single_dpm_table = &(dpm_table->dcef_table);
+		ret = vega20_get_clk_table(smu, &clocks, single_dpm_table);
+		if (ret) {
+			pr_err("Attempt to get dcefclk levels Failed!");
+			return ret;
+		}
+
+		for (i = 0; i < clocks.num_levels; i++)
+			size += sprintf(buf + size, "%d: %uMhz %s\n",
+				i, clocks.data[i].clocks_in_khz / 1000,
+				(clocks.data[i].clocks_in_khz == now * 10) ? "*" : "");
+		break;
+
+	case PP_PCIE:
+		gen_speed = (RREG32_PCIE(smnPCIE_LC_SPEED_CNTL) &
+			     PSWUSP0_PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE_MASK)
+			>> PSWUSP0_PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE__SHIFT;
+		lane_width = (RREG32_PCIE(smnPCIE_LC_LINK_WIDTH_CNTL) &
+			      PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD_MASK)
+			>> PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD__SHIFT;
+		for (i = 0; i < NUM_LINK_LEVELS; i++)
+			size += sprintf(buf + size, "%d: %s %s %dMhz %s\n", i,
+					(pptable->PcieGenSpeed[i] == 0) ? "2.5GT/s," :
+					(pptable->PcieGenSpeed[i] == 1) ? "5.0GT/s," :
+					(pptable->PcieGenSpeed[i] == 2) ? "8.0GT/s," :
+					(pptable->PcieGenSpeed[i] == 3) ? "16.0GT/s," : "",
+					(pptable->PcieLaneCount[i] == 1) ? "x1" :
+					(pptable->PcieLaneCount[i] == 2) ? "x2" :
+					(pptable->PcieLaneCount[i] == 3) ? "x4" :
+					(pptable->PcieLaneCount[i] == 4) ? "x8" :
+					(pptable->PcieLaneCount[i] == 5) ? "x12" :
+					(pptable->PcieLaneCount[i] == 6) ? "x16" : "",
+					pptable->LclkFreq[i],
+					(gen_speed == pptable->PcieGenSpeed[i]) &&
+					(lane_width == pptable->PcieLaneCount[i]) ?
+					"*" : "");
+		break;
+
+	case OD_SCLK:
+		if (od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FMIN].feature_id &&
+		    od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FMAX].feature_id) {
+			size = sprintf(buf, "%s:\n", "OD_SCLK");
+			size += sprintf(buf + size, "0: %10uMhz\n",
+					od_table->GfxclkFmin);
+			size += sprintf(buf + size, "1: %10uMhz\n",
+					od_table->GfxclkFmax);
+		}
+
+		break;
+
+	case OD_MCLK:
+		if (od8_settings->od8_settings_array[OD8_SETTING_UCLK_FMAX].feature_id) {
+			size = sprintf(buf, "%s:\n", "OD_MCLK");
+			size += sprintf(buf + size, "1: %10uMhz\n",
+					 od_table->UclkFmax);
+		}
+
+		break;
+
+	case OD_VDDC_CURVE:
+		if (od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FREQ1].feature_id &&
+		    od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FREQ2].feature_id &&
+		    od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FREQ3].feature_id &&
+		    od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE1].feature_id &&
+		    od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE2].feature_id &&
+		    od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE3].feature_id) {
+			size = sprintf(buf, "%s:\n", "OD_VDDC_CURVE");
+			size += sprintf(buf + size, "0: %10uMhz %10dmV\n",
+					od_table->GfxclkFreq1,
+					od_table->GfxclkVolt1 / VOLTAGE_SCALE);
+			size += sprintf(buf + size, "1: %10uMhz %10dmV\n",
+					od_table->GfxclkFreq2,
+					od_table->GfxclkVolt2 / VOLTAGE_SCALE);
+			size += sprintf(buf + size, "2: %10uMhz %10dmV\n",
+					od_table->GfxclkFreq3,
+					od_table->GfxclkVolt3 / VOLTAGE_SCALE);
+		}
+
+		break;
+
+	case OD_RANGE:
+		size = sprintf(buf, "%s:\n", "OD_RANGE");
+
+		if (od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FMIN].feature_id &&
+		    od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FMAX].feature_id) {
+			size += sprintf(buf + size, "SCLK: %7uMhz %10uMhz\n",
+					od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FMIN].min_value,
+					od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FMAX].max_value);
+		}
+
+		if (od8_settings->od8_settings_array[OD8_SETTING_UCLK_FMAX].feature_id) {
+			single_dpm_table = &(dpm_table->mem_table);
+			ret = vega20_get_clk_table(smu, &clocks, single_dpm_table);
+			if (ret) {
+				pr_err("Attempt to get memory clk levels Failed!");
+				return ret;
+			}
+
+			size += sprintf(buf + size, "MCLK: %7uMhz %10uMhz\n",
+					clocks.data[0].clocks_in_khz / 1000,
+					od8_settings->od8_settings_array[OD8_SETTING_UCLK_FMAX].max_value);
+		}
+
+		if (od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FREQ1].feature_id &&
+		    od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FREQ2].feature_id &&
+		    od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FREQ3].feature_id &&
+		    od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE1].feature_id &&
+		    od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE2].feature_id &&
+		    od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE3].feature_id) {
+			size += sprintf(buf + size, "VDDC_CURVE_SCLK[0]: %7uMhz %10uMhz\n",
+					od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FREQ1].min_value,
+					od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FREQ1].max_value);
+			size += sprintf(buf + size, "VDDC_CURVE_VOLT[0]: %7dmV %11dmV\n",
+					od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE1].min_value,
+					od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE1].max_value);
+			size += sprintf(buf + size, "VDDC_CURVE_SCLK[1]: %7uMhz %10uMhz\n",
+					od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FREQ2].min_value,
+					od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FREQ2].max_value);
+			size += sprintf(buf + size, "VDDC_CURVE_VOLT[1]: %7dmV %11dmV\n",
+					od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE2].min_value,
+					od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE2].max_value);
+			size += sprintf(buf + size, "VDDC_CURVE_SCLK[2]: %7uMhz %10uMhz\n",
+					od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FREQ3].min_value,
+					od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FREQ3].max_value);
+			size += sprintf(buf + size, "VDDC_CURVE_VOLT[2]: %7dmV %11dmV\n",
+					od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE3].min_value,
+					od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE3].max_value);
+		}
+
+		break;
+
+	default:
+		break;
+	}
+	return size;
+}
+
+static int vega20_upload_dpm_level(struct smu_context *smu, bool max,
+				   uint32_t feature_mask)
+{
+	struct vega20_dpm_table *dpm_table;
+	struct vega20_single_dpm_table *single_dpm_table;
+	uint32_t freq;
+	int ret = 0;
+
+	dpm_table = smu->smu_dpm.dpm_context;
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_GFXCLK_BIT) &&
+	    (feature_mask & FEATURE_DPM_GFXCLK_MASK)) {
+		single_dpm_table = &(dpm_table->gfx_table);
+		freq = max ? single_dpm_table->dpm_state.soft_max_level :
+			single_dpm_table->dpm_state.soft_min_level;
+		ret = smu_send_smc_msg_with_param(smu,
+			(max ? SMU_MSG_SetSoftMaxByFreq : SMU_MSG_SetSoftMinByFreq),
+			(PPCLK_GFXCLK << 16) | (freq & 0xffff));
+		if (ret) {
+			pr_err("Failed to set soft %s gfxclk !\n",
+						max ? "max" : "min");
+			return ret;
+		}
+	}
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_UCLK_BIT) &&
+	    (feature_mask & FEATURE_DPM_UCLK_MASK)) {
+		single_dpm_table = &(dpm_table->mem_table);
+		freq = max ? single_dpm_table->dpm_state.soft_max_level :
+			single_dpm_table->dpm_state.soft_min_level;
+		ret = smu_send_smc_msg_with_param(smu,
+			(max ? SMU_MSG_SetSoftMaxByFreq : SMU_MSG_SetSoftMinByFreq),
+			(PPCLK_UCLK << 16) | (freq & 0xffff));
+		if (ret) {
+			pr_err("Failed to set soft %s memclk !\n",
+						max ? "max" : "min");
+			return ret;
+		}
+	}
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_SOCCLK_BIT) &&
+	    (feature_mask & FEATURE_DPM_SOCCLK_MASK)) {
+		single_dpm_table = &(dpm_table->soc_table);
+		freq = max ? single_dpm_table->dpm_state.soft_max_level :
+			single_dpm_table->dpm_state.soft_min_level;
+		ret = smu_send_smc_msg_with_param(smu,
+			(max ? SMU_MSG_SetSoftMaxByFreq : SMU_MSG_SetSoftMinByFreq),
+			(PPCLK_SOCCLK << 16) | (freq & 0xffff));
+		if (ret) {
+			pr_err("Failed to set soft %s socclk !\n",
+						max ? "max" : "min");
+			return ret;
+		}
+	}
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_FCLK_BIT) &&
+	    (feature_mask & FEATURE_DPM_FCLK_MASK)) {
+		single_dpm_table = &(dpm_table->fclk_table);
+		freq = max ? single_dpm_table->dpm_state.soft_max_level :
+			single_dpm_table->dpm_state.soft_min_level;
+		ret = smu_send_smc_msg_with_param(smu,
+			(max ? SMU_MSG_SetSoftMaxByFreq : SMU_MSG_SetSoftMinByFreq),
+			(PPCLK_FCLK << 16) | (freq & 0xffff));
+		if (ret) {
+			pr_err("Failed to set soft %s fclk !\n",
+						max ? "max" : "min");
+			return ret;
+		}
+	}
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_DCEFCLK_BIT) &&
+	    (feature_mask & FEATURE_DPM_DCEFCLK_MASK)) {
+		single_dpm_table = &(dpm_table->dcef_table);
+		freq = single_dpm_table->dpm_state.hard_min_level;
+		if (!max) {
+			ret = smu_send_smc_msg_with_param(smu,
+				SMU_MSG_SetHardMinByFreq,
+				(PPCLK_DCEFCLK << 16) | (freq & 0xffff));
+			if (ret) {
+				pr_err("Failed to set hard min dcefclk !\n");
+				return ret;
+			}
+		}
+	}
+
+	return ret;
+}
+
+static int vega20_force_clk_levels(struct smu_context *smu,
+			enum pp_clock_type type, uint32_t mask)
+{
+	struct vega20_dpm_table *dpm_table;
+	struct vega20_single_dpm_table *single_dpm_table;
+	uint32_t soft_min_level, soft_max_level, hard_min_level;
+	struct smu_dpm_context *smu_dpm = &smu->smu_dpm;
+	int ret = 0;
+
+	if (smu_dpm->dpm_level != AMD_DPM_FORCED_LEVEL_MANUAL) {
+		pr_info("force clock level is for dpm manual mode only.\n");
+		return -EINVAL;
+	}
+
+	mutex_lock(&(smu->mutex));
+
+	soft_min_level = mask ? (ffs(mask) - 1) : 0;
+	soft_max_level = mask ? (fls(mask) - 1) : 0;
+
+	dpm_table = smu->smu_dpm.dpm_context;
+
+	switch (type) {
+	case PP_SCLK:
+		single_dpm_table = &(dpm_table->gfx_table);
+
+		if (soft_max_level >= single_dpm_table->count) {
+			pr_err("Clock level specified %d is over max allowed %d\n",
+					soft_max_level, single_dpm_table->count - 1);
+			ret = -EINVAL;
+			break;
+		}
+
+		single_dpm_table->dpm_state.soft_min_level =
+			single_dpm_table->dpm_levels[soft_min_level].value;
+		single_dpm_table->dpm_state.soft_max_level =
+			single_dpm_table->dpm_levels[soft_max_level].value;
+
+		ret = vega20_upload_dpm_level(smu, false, FEATURE_DPM_GFXCLK_MASK);
+		if (ret) {
+			pr_err("Failed to upload boot level to lowest!\n");
+			break;
+		}
+
+		ret = vega20_upload_dpm_level(smu, true, FEATURE_DPM_GFXCLK_MASK);
+		if (ret)
+			pr_err("Failed to upload dpm max level to highest!\n");
+
+		break;
+
+	case PP_MCLK:
+		single_dpm_table = &(dpm_table->mem_table);
+
+		if (soft_max_level >= single_dpm_table->count) {
+			pr_err("Clock level specified %d is over max allowed %d\n",
+					soft_max_level, single_dpm_table->count - 1);
+			ret = -EINVAL;
+			break;
+		}
+
+		single_dpm_table->dpm_state.soft_min_level =
+			single_dpm_table->dpm_levels[soft_min_level].value;
+		single_dpm_table->dpm_state.soft_max_level =
+			single_dpm_table->dpm_levels[soft_max_level].value;
+
+		ret = vega20_upload_dpm_level(smu, false, FEATURE_DPM_UCLK_MASK);
+		if (ret) {
+			pr_err("Failed to upload boot level to lowest!\n");
+			break;
+		}
+
+		ret = vega20_upload_dpm_level(smu, true, FEATURE_DPM_UCLK_MASK);
+		if (ret)
+			pr_err("Failed to upload dpm max level to highest!\n");
+
+		break;
+
+	case PP_SOCCLK:
+		single_dpm_table = &(dpm_table->soc_table);
+
+		if (soft_max_level >= single_dpm_table->count) {
+			pr_err("Clock level specified %d is over max allowed %d\n",
+					soft_max_level, single_dpm_table->count - 1);
+			ret = -EINVAL;
+			break;
+		}
+
+		single_dpm_table->dpm_state.soft_min_level =
+			single_dpm_table->dpm_levels[soft_min_level].value;
+		single_dpm_table->dpm_state.soft_max_level =
+			single_dpm_table->dpm_levels[soft_max_level].value;
+
+		ret = vega20_upload_dpm_level(smu, false, FEATURE_DPM_SOCCLK_MASK);
+		if (ret) {
+			pr_err("Failed to upload boot level to lowest!\n");
+			break;
+		}
+
+		ret = vega20_upload_dpm_level(smu, true, FEATURE_DPM_SOCCLK_MASK);
+		if (ret)
+			pr_err("Failed to upload dpm max level to highest!\n");
+
+		break;
+
+	case PP_FCLK:
+		single_dpm_table = &(dpm_table->fclk_table);
+
+		if (soft_max_level >= single_dpm_table->count) {
+			pr_err("Clock level specified %d is over max allowed %d\n",
+					soft_max_level, single_dpm_table->count - 1);
+			ret = -EINVAL;
+			break;
+		}
+
+		single_dpm_table->dpm_state.soft_min_level =
+			single_dpm_table->dpm_levels[soft_min_level].value;
+		single_dpm_table->dpm_state.soft_max_level =
+			single_dpm_table->dpm_levels[soft_max_level].value;
+
+		ret = vega20_upload_dpm_level(smu, false, FEATURE_DPM_FCLK_MASK);
+		if (ret) {
+			pr_err("Failed to upload boot level to lowest!\n");
+			break;
+		}
+
+		ret = vega20_upload_dpm_level(smu, true, FEATURE_DPM_FCLK_MASK);
+		if (ret)
+			pr_err("Failed to upload dpm max level to highest!\n");
+
+		break;
+
+	case PP_DCEFCLK:
+		hard_min_level = soft_min_level;
+		single_dpm_table = &(dpm_table->dcef_table);
+
+		if (hard_min_level >= single_dpm_table->count) {
+			pr_err("Clock level specified %d is over max allowed %d\n",
+					hard_min_level, single_dpm_table->count - 1);
+			ret = -EINVAL;
+			break;
+		}
+
+		single_dpm_table->dpm_state.hard_min_level =
+			single_dpm_table->dpm_levels[hard_min_level].value;
+
+		ret = vega20_upload_dpm_level(smu, false, FEATURE_DPM_DCEFCLK_MASK);
+		if (ret)
+			pr_err("Failed to upload boot level to lowest!\n");
+
+		break;
+
+	case PP_PCIE:
+		if (soft_min_level >= NUM_LINK_LEVELS ||
+		    soft_max_level >= NUM_LINK_LEVELS) {
+			ret = -EINVAL;
+			break;
+		}
+
+		ret = smu_send_smc_msg_with_param(smu,
+				SMU_MSG_SetMinLinkDpmByIndex, soft_min_level);
+		if (ret)
+			pr_err("Failed to set min link dpm level!\n");
+
+		break;
+
+	default:
+		break;
+	}
+
+	mutex_unlock(&(smu->mutex));
+	return ret;
+}
+
+static int vega20_get_clock_by_type_with_latency(struct smu_context *smu,
+						 enum amd_pp_clock_type type,
+						 struct pp_clock_levels_with_latency *clocks)
+{
+	int ret;
+	struct vega20_single_dpm_table *single_dpm_table;
+	struct smu_dpm_context *smu_dpm = &smu->smu_dpm;
+	struct vega20_dpm_table *dpm_table = NULL;
+
+	dpm_table = smu_dpm->dpm_context;
+
+	mutex_lock(&smu->mutex);
+
+	switch (type) {
+	case amd_pp_sys_clock:
+		single_dpm_table = &(dpm_table->gfx_table);
+		ret = vega20_get_clk_table(smu, clocks, single_dpm_table);
+		break;
+	case amd_pp_mem_clock:
+		single_dpm_table = &(dpm_table->mem_table);
+		ret = vega20_get_clk_table(smu, clocks, single_dpm_table);
+		break;
+	case amd_pp_dcef_clock:
+		single_dpm_table = &(dpm_table->dcef_table);
+		ret = vega20_get_clk_table(smu, clocks, single_dpm_table);
+		break;
+	case amd_pp_soc_clock:
+		single_dpm_table = &(dpm_table->soc_table);
+		ret = vega20_get_clk_table(smu, clocks, single_dpm_table);
+		break;
+	default:
+		ret = -EINVAL;
+	}
+
+	mutex_unlock(&smu->mutex);
+	return ret;
+}
+
+static int vega20_overdrive_get_gfx_clk_base_voltage(struct smu_context *smu,
+						     uint32_t *voltage,
+						     uint32_t freq)
+{
+	int ret;
+
+	ret = smu_send_smc_msg_with_param(smu,
+			SMU_MSG_GetAVFSVoltageByDpm,
+			((AVFS_CURVE << 24) | (OD8_HOTCURVE_TEMPERATURE << 16) | freq));
+	if (ret) {
+		pr_err("[GetBaseVoltage] failed to get GFXCLK AVFS voltage from SMU!");
+		return ret;
+	}
+
+	smu_read_smc_arg(smu, voltage);
+	*voltage = *voltage / VOLTAGE_SCALE;
+
+	return 0;
+}
+
+static int vega20_set_default_od8_setttings(struct smu_context *smu)
+{
+	struct smu_table_context *table_context = &smu->smu_table;
+	OverDriveTable_t *od_table = (OverDriveTable_t *)(table_context->overdrive_table);
+	struct vega20_od8_settings *od8_settings = NULL;
+	PPTable_t *smc_pptable = table_context->driver_pptable;
+	int i, ret;
+
+	if (table_context->od8_settings)
+		return -EINVAL;
+
+	table_context->od8_settings = kzalloc(sizeof(struct vega20_od8_settings), GFP_KERNEL);
+
+	if (!table_context->od8_settings)
+		return -ENOMEM;
+
+	memset(table_context->od8_settings, 0, sizeof(struct vega20_od8_settings));
+	od8_settings = (struct vega20_od8_settings *)table_context->od8_settings;
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_SOCCLK_BIT)) {
+		if (table_context->od_feature_capabilities[ATOM_VEGA20_ODFEATURE_GFXCLK_LIMITS] &&
+		    table_context->od_settings_max[OD8_SETTING_GFXCLK_FMAX] > 0 &&
+		    table_context->od_settings_min[OD8_SETTING_GFXCLK_FMIN] > 0 &&
+		    (table_context->od_settings_max[OD8_SETTING_GFXCLK_FMAX] >=
+		     table_context->od_settings_min[OD8_SETTING_GFXCLK_FMIN])) {
+			od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FMIN].feature_id =
+				OD8_GFXCLK_LIMITS;
+			od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FMAX].feature_id =
+				OD8_GFXCLK_LIMITS;
+			od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FMIN].default_value =
+				od_table->GfxclkFmin;
+			od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FMAX].default_value =
+				od_table->GfxclkFmax;
+		}
+
+		if (table_context->od_feature_capabilities[ATOM_VEGA20_ODFEATURE_GFXCLK_CURVE] &&
+		    (table_context->od_settings_min[OD8_SETTING_GFXCLK_VOLTAGE1] >=
+		     smc_pptable->MinVoltageGfx / VOLTAGE_SCALE) &&
+		    (table_context->od_settings_max[OD8_SETTING_GFXCLK_VOLTAGE3] <=
+		     smc_pptable->MaxVoltageGfx / VOLTAGE_SCALE) &&
+		    (table_context->od_settings_min[OD8_SETTING_GFXCLK_VOLTAGE1] <=
+		     table_context->od_settings_max[OD8_SETTING_GFXCLK_VOLTAGE3])) {
+			od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FREQ1].feature_id =
+				OD8_GFXCLK_CURVE;
+			od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE1].feature_id =
+				OD8_GFXCLK_CURVE;
+			od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FREQ2].feature_id =
+				OD8_GFXCLK_CURVE;
+			od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE2].feature_id =
+				OD8_GFXCLK_CURVE;
+			od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FREQ3].feature_id =
+				OD8_GFXCLK_CURVE;
+			od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE3].feature_id =
+				OD8_GFXCLK_CURVE;
+
+			od_table->GfxclkFreq1 = od_table->GfxclkFmin;
+			od_table->GfxclkFreq2 = (od_table->GfxclkFmin + od_table->GfxclkFmax) / 2;
+			od_table->GfxclkFreq3 = od_table->GfxclkFmax;
+			od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FREQ1].default_value =
+				od_table->GfxclkFreq1;
+			od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FREQ2].default_value =
+				od_table->GfxclkFreq2;
+			od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FREQ3].default_value =
+				od_table->GfxclkFreq3;
+
+			ret = vega20_overdrive_get_gfx_clk_base_voltage(smu,
+				&od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE1].default_value,
+				od_table->GfxclkFreq1);
+			if (ret)
+				od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE1].default_value = 0;
+			od_table->GfxclkVolt1 =
+				od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE1].default_value
+				* VOLTAGE_SCALE;
+			ret = vega20_overdrive_get_gfx_clk_base_voltage(smu,
+				&od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE2].default_value,
+				od_table->GfxclkFreq2);
+			if (ret)
+				od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE2].default_value = 0;
+			od_table->GfxclkVolt2 =
+				od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE2].default_value
+				* VOLTAGE_SCALE;
+			ret = vega20_overdrive_get_gfx_clk_base_voltage(smu,
+				&od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE3].default_value,
+				od_table->GfxclkFreq3);
+			if (ret)
+				od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE3].default_value = 0;
+			od_table->GfxclkVolt3 =
+				od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE3].default_value
+				* VOLTAGE_SCALE;
+		}
+	}
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_UCLK_BIT)) {
+		if (table_context->od_feature_capabilities[ATOM_VEGA20_ODFEATURE_UCLK_MAX] &&
+		    table_context->od_settings_min[OD8_SETTING_UCLK_FMAX] > 0 &&
+		    table_context->od_settings_max[OD8_SETTING_UCLK_FMAX] > 0 &&
+		    (table_context->od_settings_max[OD8_SETTING_UCLK_FMAX] >=
+		     table_context->od_settings_min[OD8_SETTING_UCLK_FMAX])) {
+			od8_settings->od8_settings_array[OD8_SETTING_UCLK_FMAX].feature_id =
+				OD8_UCLK_MAX;
+			od8_settings->od8_settings_array[OD8_SETTING_UCLK_FMAX].default_value =
+				od_table->UclkFmax;
+		}
+	}
+
+	if (table_context->od_feature_capabilities[ATOM_VEGA20_ODFEATURE_POWER_LIMIT] &&
+	    table_context->od_settings_min[OD8_SETTING_POWER_PERCENTAGE] > 0 &&
+	    table_context->od_settings_min[OD8_SETTING_POWER_PERCENTAGE] <= 100 &&
+	    table_context->od_settings_max[OD8_SETTING_POWER_PERCENTAGE] > 0 &&
+	    table_context->od_settings_max[OD8_SETTING_POWER_PERCENTAGE] <= 100) {
+		od8_settings->od8_settings_array[OD8_SETTING_POWER_PERCENTAGE].feature_id =
+			OD8_POWER_LIMIT;
+		od8_settings->od8_settings_array[OD8_SETTING_POWER_PERCENTAGE].default_value =
+			od_table->OverDrivePct;
+	}
+
+	if (smu_feature_is_enabled(smu, FEATURE_FAN_CONTROL_BIT)) {
+		if (table_context->od_feature_capabilities[ATOM_VEGA20_ODFEATURE_FAN_ACOUSTIC_LIMIT] &&
+		    table_context->od_settings_min[OD8_SETTING_FAN_ACOUSTIC_LIMIT] > 0 &&
+		    table_context->od_settings_max[OD8_SETTING_FAN_ACOUSTIC_LIMIT] > 0 &&
+		    (table_context->od_settings_max[OD8_SETTING_FAN_ACOUSTIC_LIMIT] >=
+		     table_context->od_settings_min[OD8_SETTING_FAN_ACOUSTIC_LIMIT])) {
+			od8_settings->od8_settings_array[OD8_SETTING_FAN_ACOUSTIC_LIMIT].feature_id =
+				OD8_ACOUSTIC_LIMIT_SCLK;
+			od8_settings->od8_settings_array[OD8_SETTING_FAN_ACOUSTIC_LIMIT].default_value =
+				od_table->FanMaximumRpm;
+		}
+
+		if (table_context->od_feature_capabilities[ATOM_VEGA20_ODFEATURE_FAN_SPEED_MIN] &&
+		    table_context->od_settings_min[OD8_SETTING_FAN_MIN_SPEED] > 0 &&
+		    table_context->od_settings_max[OD8_SETTING_FAN_MIN_SPEED] > 0 &&
+		    (table_context->od_settings_max[OD8_SETTING_FAN_MIN_SPEED] >=
+		     table_context->od_settings_min[OD8_SETTING_FAN_MIN_SPEED])) {
+			od8_settings->od8_settings_array[OD8_SETTING_FAN_MIN_SPEED].feature_id =
+				OD8_FAN_SPEED_MIN;
+			od8_settings->od8_settings_array[OD8_SETTING_FAN_MIN_SPEED].default_value =
+				od_table->FanMinimumPwm * smc_pptable->FanMaximumRpm / 100;
+		}
+	}
+
+	if (smu_feature_is_enabled(smu, FEATURE_THERMAL_BIT)) {
+		if (table_context->od_feature_capabilities[ATOM_VEGA20_ODFEATURE_TEMPERATURE_FAN] &&
+		    table_context->od_settings_min[OD8_SETTING_FAN_TARGET_TEMP] > 0 &&
+		    table_context->od_settings_max[OD8_SETTING_FAN_TARGET_TEMP] > 0 &&
+		    (table_context->od_settings_max[OD8_SETTING_FAN_TARGET_TEMP] >=
+		     table_context->od_settings_min[OD8_SETTING_FAN_TARGET_TEMP])) {
+			od8_settings->od8_settings_array[OD8_SETTING_FAN_TARGET_TEMP].feature_id =
+				OD8_TEMPERATURE_FAN;
+			od8_settings->od8_settings_array[OD8_SETTING_FAN_TARGET_TEMP].default_value =
+				od_table->FanTargetTemperature;
+		}
+
+		if (table_context->od_feature_capabilities[ATOM_VEGA20_ODFEATURE_TEMPERATURE_SYSTEM] &&
+		    table_context->od_settings_min[OD8_SETTING_OPERATING_TEMP_MAX] > 0 &&
+		    table_context->od_settings_max[OD8_SETTING_OPERATING_TEMP_MAX] > 0 &&
+		    (table_context->od_settings_max[OD8_SETTING_OPERATING_TEMP_MAX] >=
+		     table_context->od_settings_min[OD8_SETTING_OPERATING_TEMP_MAX])) {
+			od8_settings->od8_settings_array[OD8_SETTING_OPERATING_TEMP_MAX].feature_id =
+				OD8_TEMPERATURE_SYSTEM;
+			od8_settings->od8_settings_array[OD8_SETTING_OPERATING_TEMP_MAX].default_value =
+				od_table->MaxOpTemp;
+		}
+	}
+
+	for (i = 0; i < OD8_SETTING_COUNT; i++) {
+		if (od8_settings->od8_settings_array[i].feature_id) {
+			od8_settings->od8_settings_array[i].min_value =
+				table_context->od_settings_min[i];
+			od8_settings->od8_settings_array[i].max_value =
+				table_context->od_settings_max[i];
+			od8_settings->od8_settings_array[i].current_value =
+				od8_settings->od8_settings_array[i].default_value;
+		} else {
+			od8_settings->od8_settings_array[i].min_value = 0;
+			od8_settings->od8_settings_array[i].max_value = 0;
+			od8_settings->od8_settings_array[i].current_value = 0;
+		}
+	}
+
+	return 0;
+}
+
+static int vega20_get_od_percentage(struct smu_context *smu,
+				    enum pp_clock_type type)
+{
+	struct smu_dpm_context *smu_dpm = &smu->smu_dpm;
+	struct vega20_dpm_table *dpm_table = NULL;
+	struct vega20_dpm_table *golden_table = NULL;
+	struct vega20_single_dpm_table *single_dpm_table;
+	struct vega20_single_dpm_table *golden_dpm_table;
+	int value, golden_value;
+
+	dpm_table = smu_dpm->dpm_context;
+	golden_table = smu_dpm->golden_dpm_context;
+
+	switch (type) {
+	case OD_SCLK:
+		single_dpm_table = &(dpm_table->gfx_table);
+		golden_dpm_table = &(golden_table->gfx_table);
+		break;
+	case OD_MCLK:
+		single_dpm_table = &(dpm_table->mem_table);
+		golden_dpm_table = &(golden_table->mem_table);
+		break;
+	default:
+		return -EINVAL;
+		break;
+	}
+
+	value = single_dpm_table->dpm_levels[single_dpm_table->count - 1].value;
+	golden_value = golden_dpm_table->dpm_levels[golden_dpm_table->count - 1].value;
+
+	value -= golden_value;
+	value = DIV_ROUND_UP(value * 100, golden_value);
+
+	return value;
+}
+
+static int
+vega20_get_profiling_clk_mask(struct smu_context *smu,
+			      enum amd_dpm_forced_level level,
+			      uint32_t *sclk_mask,
+			      uint32_t *mclk_mask,
+			      uint32_t *soc_mask)
+{
+	struct vega20_dpm_table *dpm_table = (struct vega20_dpm_table *)smu->smu_dpm.dpm_context;
+	struct vega20_single_dpm_table *gfx_dpm_table;
+	struct vega20_single_dpm_table *mem_dpm_table;
+	struct vega20_single_dpm_table *soc_dpm_table;
+
+	if (!smu->smu_dpm.dpm_context)
+		return -EINVAL;
+
+	gfx_dpm_table = &dpm_table->gfx_table;
+	mem_dpm_table = &dpm_table->mem_table;
+	soc_dpm_table = &dpm_table->soc_table;
+
+	*sclk_mask = 0;
+	*mclk_mask = 0;
+	*soc_mask  = 0;
+
+	if (gfx_dpm_table->count > VEGA20_UMD_PSTATE_GFXCLK_LEVEL &&
+	    mem_dpm_table->count > VEGA20_UMD_PSTATE_MCLK_LEVEL &&
+	    soc_dpm_table->count > VEGA20_UMD_PSTATE_SOCCLK_LEVEL) {
+		*sclk_mask = VEGA20_UMD_PSTATE_GFXCLK_LEVEL;
+		*mclk_mask = VEGA20_UMD_PSTATE_MCLK_LEVEL;
+		*soc_mask  = VEGA20_UMD_PSTATE_SOCCLK_LEVEL;
+	}
+
+	if (level == AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK) {
+		*sclk_mask = 0;
+	} else if (level == AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK) {
+		*mclk_mask = 0;
+	} else if (level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK) {
+		*sclk_mask = gfx_dpm_table->count - 1;
+		*mclk_mask = mem_dpm_table->count - 1;
+		*soc_mask  = soc_dpm_table->count - 1;
+	}
+
+	return 0;
+}
+
+static int
+vega20_set_uclk_to_highest_dpm_level(struct smu_context *smu,
+				     struct vega20_single_dpm_table *dpm_table)
+{
+	int ret = 0;
+	struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm);
+	if (!smu_dpm_ctx->dpm_context)
+		return -EINVAL;
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_UCLK_BIT)) {
+		if (dpm_table->count <= 0) {
+			pr_err("[%s] Dpm table has no entry!", __func__);
+				return -EINVAL;
+		}
+
+		if (dpm_table->count > NUM_UCLK_DPM_LEVELS) {
+			pr_err("[%s] Dpm table has too many entries!", __func__);
+				return -EINVAL;
+		}
+
+		dpm_table->dpm_state.hard_min_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+		ret = smu_send_smc_msg_with_param(smu,
+				SMU_MSG_SetHardMinByFreq,
+				(PPCLK_UCLK << 16) | dpm_table->dpm_state.hard_min_level);
+		if (ret) {
+			pr_err("[%s] Set hard min uclk failed!", __func__);
+				return ret;
+		}
+	}
+
+	return ret;
+}
+
+static int vega20_pre_display_config_changed(struct smu_context *smu)
+{
+	int ret = 0;
+	struct vega20_dpm_table *dpm_table = smu->smu_dpm.dpm_context;
+
+	if (!smu->smu_dpm.dpm_context)
+		return -EINVAL;
+
+	smu_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, 0);
+	ret = vega20_set_uclk_to_highest_dpm_level(smu,
+						   &dpm_table->mem_table);
+	if (ret)
+		pr_err("Failed to set uclk to highest dpm level");
+	return ret;
+}
+
+static int vega20_display_config_changed(struct smu_context *smu)
+{
+	int ret = 0;
+
+	if (!smu->funcs)
+		return -EINVAL;
+
+	if (!smu->smu_dpm.dpm_context ||
+	    !smu->smu_table.tables ||
+	    !smu->smu_table.tables[TABLE_WATERMARKS].cpu_addr)
+		return -EINVAL;
+
+	if ((smu->watermarks_bitmap & WATERMARKS_EXIST) &&
+	    !(smu->watermarks_bitmap & WATERMARKS_LOADED)) {
+		ret = smu->funcs->write_watermarks_table(smu);
+		if (ret) {
+			pr_err("Failed to update WMTABLE!");
+			return ret;
+		}
+		smu->watermarks_bitmap |= WATERMARKS_LOADED;
+	}
+
+	if ((smu->watermarks_bitmap & WATERMARKS_EXIST) &&
+	    smu_feature_is_supported(smu, FEATURE_DPM_DCEFCLK_BIT) &&
+	    smu_feature_is_supported(smu, FEATURE_DPM_SOCCLK_BIT)) {
+		smu_send_smc_msg_with_param(smu,
+					    SMU_MSG_NumOfDisplays,
+					    smu->display_config->num_display);
+	}
+
+	return ret;
+}
+
+static int vega20_apply_clocks_adjust_rules(struct smu_context *smu)
+{
+	struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm);
+	struct vega20_dpm_table *dpm_ctx = (struct vega20_dpm_table *)(smu_dpm_ctx->dpm_context);
+	struct vega20_single_dpm_table *dpm_table;
+	bool vblank_too_short = false;
+	bool disable_mclk_switching;
+	uint32_t i, latency;
+
+	disable_mclk_switching = ((1 < smu->display_config->num_display) &&
+				  !smu->display_config->multi_monitor_in_sync) || vblank_too_short;
+	latency = smu->display_config->dce_tolerable_mclk_in_active_latency;
+
+	/* gfxclk */
+	dpm_table = &(dpm_ctx->gfx_table);
+	dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[0].value;
+	dpm_table->dpm_state.soft_max_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+	dpm_table->dpm_state.hard_min_level = dpm_table->dpm_levels[0].value;
+	dpm_table->dpm_state.hard_max_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+
+		if (VEGA20_UMD_PSTATE_GFXCLK_LEVEL < dpm_table->count) {
+			dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[VEGA20_UMD_PSTATE_GFXCLK_LEVEL].value;
+			dpm_table->dpm_state.soft_max_level = dpm_table->dpm_levels[VEGA20_UMD_PSTATE_GFXCLK_LEVEL].value;
+		}
+
+		if (smu_dpm_ctx->dpm_level == AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK) {
+			dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[0].value;
+			dpm_table->dpm_state.soft_max_level = dpm_table->dpm_levels[0].value;
+		}
+
+		if (smu_dpm_ctx->dpm_level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK) {
+			dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+			dpm_table->dpm_state.soft_max_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+		}
+
+	/* memclk */
+	dpm_table = &(dpm_ctx->mem_table);
+	dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[0].value;
+	dpm_table->dpm_state.soft_max_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+	dpm_table->dpm_state.hard_min_level = dpm_table->dpm_levels[0].value;
+	dpm_table->dpm_state.hard_max_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+
+		if (VEGA20_UMD_PSTATE_MCLK_LEVEL < dpm_table->count) {
+			dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[VEGA20_UMD_PSTATE_MCLK_LEVEL].value;
+			dpm_table->dpm_state.soft_max_level = dpm_table->dpm_levels[VEGA20_UMD_PSTATE_MCLK_LEVEL].value;
+		}
+
+		if (smu_dpm_ctx->dpm_level == AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK) {
+			dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[0].value;
+			dpm_table->dpm_state.soft_max_level = dpm_table->dpm_levels[0].value;
+		}
+
+		if (smu_dpm_ctx->dpm_level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK) {
+			dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+			dpm_table->dpm_state.soft_max_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+		}
+
+	/* honour DAL's UCLK Hardmin */
+	if (dpm_table->dpm_state.hard_min_level < (smu->display_config->min_mem_set_clock / 100))
+		dpm_table->dpm_state.hard_min_level = smu->display_config->min_mem_set_clock / 100;
+
+	/* Hardmin is dependent on displayconfig */
+	if (disable_mclk_switching) {
+		dpm_table->dpm_state.hard_min_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+		for (i = 0; i < smu_dpm_ctx->mclk_latency_table->count - 1; i++) {
+			if (smu_dpm_ctx->mclk_latency_table->entries[i].latency <= latency) {
+				if (dpm_table->dpm_levels[i].value >= (smu->display_config->min_mem_set_clock / 100)) {
+					dpm_table->dpm_state.hard_min_level = dpm_table->dpm_levels[i].value;
+					break;
+				}
+			}
+		}
+	}
+
+	if (smu->display_config->nb_pstate_switch_disable)
+		dpm_table->dpm_state.hard_min_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+
+	/* vclk */
+	dpm_table = &(dpm_ctx->vclk_table);
+	dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[0].value;
+	dpm_table->dpm_state.soft_max_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+	dpm_table->dpm_state.hard_min_level = dpm_table->dpm_levels[0].value;
+	dpm_table->dpm_state.hard_max_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+
+		if (VEGA20_UMD_PSTATE_UVDCLK_LEVEL < dpm_table->count) {
+			dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[VEGA20_UMD_PSTATE_UVDCLK_LEVEL].value;
+			dpm_table->dpm_state.soft_max_level = dpm_table->dpm_levels[VEGA20_UMD_PSTATE_UVDCLK_LEVEL].value;
+		}
+
+		if (smu_dpm_ctx->dpm_level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK) {
+			dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+			dpm_table->dpm_state.soft_max_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+		}
+
+	/* dclk */
+	dpm_table = &(dpm_ctx->dclk_table);
+	dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[0].value;
+	dpm_table->dpm_state.soft_max_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+	dpm_table->dpm_state.hard_min_level = dpm_table->dpm_levels[0].value;
+	dpm_table->dpm_state.hard_max_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+
+		if (VEGA20_UMD_PSTATE_UVDCLK_LEVEL < dpm_table->count) {
+			dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[VEGA20_UMD_PSTATE_UVDCLK_LEVEL].value;
+			dpm_table->dpm_state.soft_max_level = dpm_table->dpm_levels[VEGA20_UMD_PSTATE_UVDCLK_LEVEL].value;
+		}
+
+		if (smu_dpm_ctx->dpm_level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK) {
+			dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+			dpm_table->dpm_state.soft_max_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+		}
+
+	/* socclk */
+	dpm_table = &(dpm_ctx->soc_table);
+	dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[0].value;
+	dpm_table->dpm_state.soft_max_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+	dpm_table->dpm_state.hard_min_level = dpm_table->dpm_levels[0].value;
+	dpm_table->dpm_state.hard_max_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+
+		if (VEGA20_UMD_PSTATE_SOCCLK_LEVEL < dpm_table->count) {
+			dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[VEGA20_UMD_PSTATE_SOCCLK_LEVEL].value;
+			dpm_table->dpm_state.soft_max_level = dpm_table->dpm_levels[VEGA20_UMD_PSTATE_SOCCLK_LEVEL].value;
+		}
+
+		if (smu_dpm_ctx->dpm_level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK) {
+			dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+			dpm_table->dpm_state.soft_max_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+		}
+
+	/* eclk */
+	dpm_table = &(dpm_ctx->eclk_table);
+	dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[0].value;
+	dpm_table->dpm_state.soft_max_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+	dpm_table->dpm_state.hard_min_level = dpm_table->dpm_levels[0].value;
+	dpm_table->dpm_state.hard_max_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+
+		if (VEGA20_UMD_PSTATE_VCEMCLK_LEVEL < dpm_table->count) {
+			dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[VEGA20_UMD_PSTATE_VCEMCLK_LEVEL].value;
+			dpm_table->dpm_state.soft_max_level = dpm_table->dpm_levels[VEGA20_UMD_PSTATE_VCEMCLK_LEVEL].value;
+		}
+
+		if (smu_dpm_ctx->dpm_level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK) {
+			dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+			dpm_table->dpm_state.soft_max_level = dpm_table->dpm_levels[dpm_table->count - 1].value;
+		}
+	return 0;
+}
+
+static int
+vega20_notify_smc_dispaly_config(struct smu_context *smu)
+{
+	struct vega20_dpm_table *dpm_table = smu->smu_dpm.dpm_context;
+	struct vega20_single_dpm_table *memtable = &dpm_table->mem_table;
+	struct smu_clocks min_clocks = {0};
+	struct pp_display_clock_request clock_req;
+	int ret = 0;
+
+	min_clocks.dcef_clock = smu->display_config->min_dcef_set_clk;
+	min_clocks.dcef_clock_in_sr = smu->display_config->min_dcef_deep_sleep_set_clk;
+	min_clocks.memory_clock = smu->display_config->min_mem_set_clock;
+
+	if (smu_feature_is_supported(smu, FEATURE_DPM_DCEFCLK_BIT)) {
+		clock_req.clock_type = amd_pp_dcef_clock;
+		clock_req.clock_freq_in_khz = min_clocks.dcef_clock * 10;
+		if (!smu->funcs->display_clock_voltage_request(smu, &clock_req)) {
+			if (smu_feature_is_supported(smu, FEATURE_DS_DCEFCLK_BIT)) {
+				ret = smu_send_smc_msg_with_param(smu,
+								  SMU_MSG_SetMinDeepSleepDcefclk,
+								  min_clocks.dcef_clock_in_sr/100);
+				if (ret) {
+					pr_err("Attempt to set divider for DCEFCLK Failed!");
+					return ret;
+				}
+			}
+		} else {
+			pr_info("Attempt to set Hard Min for DCEFCLK Failed!");
+		}
+	}
+
+	if (smu_feature_is_enabled(smu, FEATURE_DPM_UCLK_BIT)) {
+		memtable->dpm_state.hard_min_level = min_clocks.memory_clock/100;
+		ret = smu_send_smc_msg_with_param(smu,
+						  SMU_MSG_SetHardMinByFreq,
+						  (PPCLK_UCLK << 16) | memtable->dpm_state.hard_min_level);
+		if (ret) {
+			pr_err("[%s] Set hard min uclk failed!", __func__);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+static uint32_t vega20_find_lowest_dpm_level(struct vega20_single_dpm_table *table)
+{
+	uint32_t i;
+
+	for (i = 0; i < table->count; i++) {
+		if (table->dpm_levels[i].enabled)
+			break;
+	}
+	if (i >= table->count) {
+		i = 0;
+		table->dpm_levels[i].enabled = true;
+	}
+
+	return i;
+}
+
+static uint32_t vega20_find_highest_dpm_level(struct vega20_single_dpm_table *table)
+{
+	int i = 0;
+
+	if (!table) {
+		pr_err("[%s] DPM Table does not exist!", __func__);
+		return 0;
+	}
+	if (table->count <= 0) {
+		pr_err("[%s] DPM Table has no entry!", __func__);
+		return 0;
+	}
+	if (table->count > MAX_REGULAR_DPM_NUMBER) {
+		pr_err("[%s] DPM Table has too many entries!", __func__);
+		return MAX_REGULAR_DPM_NUMBER - 1;
+	}
+
+	for (i = table->count - 1; i >= 0; i--) {
+		if (table->dpm_levels[i].enabled)
+			break;
+	}
+	if (i < 0) {
+		i = 0;
+		table->dpm_levels[i].enabled = true;
+	}
+
+	return i;
+}
+
+static int vega20_force_dpm_limit_value(struct smu_context *smu, bool highest)
+{
+	uint32_t soft_level;
+	int ret = 0;
+	struct vega20_dpm_table *dpm_table =
+		(struct vega20_dpm_table *)smu->smu_dpm.dpm_context;
+
+	if (highest)
+		soft_level = vega20_find_highest_dpm_level(&(dpm_table->gfx_table));
+	else
+		soft_level = vega20_find_lowest_dpm_level(&(dpm_table->gfx_table));
+
+	dpm_table->gfx_table.dpm_state.soft_min_level =
+		dpm_table->gfx_table.dpm_state.soft_max_level =
+		dpm_table->gfx_table.dpm_levels[soft_level].value;
+
+	if (highest)
+		soft_level = vega20_find_highest_dpm_level(&(dpm_table->mem_table));
+	else
+		soft_level = vega20_find_lowest_dpm_level(&(dpm_table->mem_table));
+
+	dpm_table->mem_table.dpm_state.soft_min_level =
+		dpm_table->mem_table.dpm_state.soft_max_level =
+		dpm_table->mem_table.dpm_levels[soft_level].value;
+
+	if (highest)
+		soft_level = vega20_find_highest_dpm_level(&(dpm_table->soc_table));
+	else
+		soft_level = vega20_find_lowest_dpm_level(&(dpm_table->soc_table));
+
+	dpm_table->soc_table.dpm_state.soft_min_level =
+		dpm_table->soc_table.dpm_state.soft_max_level =
+		dpm_table->soc_table.dpm_levels[soft_level].value;
+
+	ret = vega20_upload_dpm_level(smu, false, 0xFFFFFFFF);
+	if (ret) {
+		pr_err("Failed to upload boot level to %s!\n",
+				highest ? "highest" : "lowest");
+		return ret;
+	}
+
+	ret = vega20_upload_dpm_level(smu, true, 0xFFFFFFFF);
+	if (ret) {
+		pr_err("Failed to upload dpm max level to %s!\n!",
+				highest ? "highest" : "lowest");
+		return ret;
+	}
+
+	return ret;
+}
+
+static int vega20_unforce_dpm_levels(struct smu_context *smu)
+{
+	uint32_t soft_min_level, soft_max_level;
+	int ret = 0;
+	struct vega20_dpm_table *dpm_table =
+		(struct vega20_dpm_table *)smu->smu_dpm.dpm_context;
+
+	soft_min_level = vega20_find_lowest_dpm_level(&(dpm_table->gfx_table));
+	soft_max_level = vega20_find_highest_dpm_level(&(dpm_table->gfx_table));
+	dpm_table->gfx_table.dpm_state.soft_min_level =
+		dpm_table->gfx_table.dpm_levels[soft_min_level].value;
+	dpm_table->gfx_table.dpm_state.soft_max_level =
+		dpm_table->gfx_table.dpm_levels[soft_max_level].value;
+
+	soft_min_level = vega20_find_lowest_dpm_level(&(dpm_table->mem_table));
+	soft_max_level = vega20_find_highest_dpm_level(&(dpm_table->mem_table));
+	dpm_table->mem_table.dpm_state.soft_min_level =
+		dpm_table->gfx_table.dpm_levels[soft_min_level].value;
+	dpm_table->mem_table.dpm_state.soft_max_level =
+		dpm_table->gfx_table.dpm_levels[soft_max_level].value;
+
+	soft_min_level = vega20_find_lowest_dpm_level(&(dpm_table->soc_table));
+	soft_max_level = vega20_find_highest_dpm_level(&(dpm_table->soc_table));
+	dpm_table->soc_table.dpm_state.soft_min_level =
+		dpm_table->soc_table.dpm_levels[soft_min_level].value;
+	dpm_table->soc_table.dpm_state.soft_max_level =
+		dpm_table->soc_table.dpm_levels[soft_max_level].value;
+
+	ret = smu_upload_dpm_level(smu, false, 0xFFFFFFFF);
+	if (ret) {
+		pr_err("Failed to upload DPM Bootup Levels!");
+		return ret;
+	}
+
+	ret = smu_upload_dpm_level(smu, true, 0xFFFFFFFF);
+	if (ret) {
+		pr_err("Failed to upload DPM Max Levels!");
+		return ret;
+	}
+
+	return ret;
+}
+
+static enum amd_dpm_forced_level vega20_get_performance_level(struct smu_context *smu)
+{
+	struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm);
+	if (!smu_dpm_ctx->dpm_context)
+		return -EINVAL;
+
+	if (smu_dpm_ctx->dpm_level != smu_dpm_ctx->saved_dpm_level) {
+		mutex_lock(&(smu->mutex));
+		smu_dpm_ctx->saved_dpm_level = smu_dpm_ctx->dpm_level;
+		mutex_unlock(&(smu->mutex));
+	}
+	return smu_dpm_ctx->dpm_level;
+}
+
+static int
+vega20_force_performance_level(struct smu_context *smu, enum amd_dpm_forced_level level)
+{
+	int ret = 0;
+	int i;
+	struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm);
+
+	if (!smu_dpm_ctx->dpm_context)
+		return -EINVAL;
+
+	for (i = 0; i < smu->adev->num_ip_blocks; i++) {
+		if (smu->adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SMC)
+			break;
+	}
+
+	mutex_lock(&smu->mutex);
+
+	smu->adev->ip_blocks[i].version->funcs->enable_umd_pstate(smu, &level);
+	ret = smu_handle_task(smu, level,
+			      AMD_PP_TASK_READJUST_POWER_STATE);
+
+	mutex_unlock(&smu->mutex);
+
+	return ret;
+}
+
+static int vega20_update_specified_od8_value(struct smu_context *smu,
+					     uint32_t index,
+					     uint32_t value)
+{
+	struct smu_table_context *table_context = &smu->smu_table;
+	OverDriveTable_t *od_table =
+		(OverDriveTable_t *)(table_context->overdrive_table);
+	struct vega20_od8_settings *od8_settings =
+		(struct vega20_od8_settings *)table_context->od8_settings;
+
+	switch (index) {
+	case OD8_SETTING_GFXCLK_FMIN:
+		od_table->GfxclkFmin = (uint16_t)value;
+		break;
+
+	case OD8_SETTING_GFXCLK_FMAX:
+		if (value < od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FMAX].min_value ||
+		    value > od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FMAX].max_value)
+			return -EINVAL;
+		od_table->GfxclkFmax = (uint16_t)value;
+		break;
+
+	case OD8_SETTING_GFXCLK_FREQ1:
+		od_table->GfxclkFreq1 = (uint16_t)value;
+		break;
+
+	case OD8_SETTING_GFXCLK_VOLTAGE1:
+		od_table->GfxclkVolt1 = (uint16_t)value;
+		break;
+
+	case OD8_SETTING_GFXCLK_FREQ2:
+		od_table->GfxclkFreq2 = (uint16_t)value;
+		break;
+
+	case OD8_SETTING_GFXCLK_VOLTAGE2:
+		od_table->GfxclkVolt2 = (uint16_t)value;
+		break;
+
+	case OD8_SETTING_GFXCLK_FREQ3:
+		od_table->GfxclkFreq3 = (uint16_t)value;
+		break;
+
+	case OD8_SETTING_GFXCLK_VOLTAGE3:
+		od_table->GfxclkVolt3 = (uint16_t)value;
+		break;
+
+	case OD8_SETTING_UCLK_FMAX:
+		if (value < od8_settings->od8_settings_array[OD8_SETTING_UCLK_FMAX].min_value ||
+		    value > od8_settings->od8_settings_array[OD8_SETTING_UCLK_FMAX].max_value)
+			return -EINVAL;
+		od_table->UclkFmax = (uint16_t)value;
+		break;
+
+	case OD8_SETTING_POWER_PERCENTAGE:
+		od_table->OverDrivePct = (int16_t)value;
+		break;
+
+	case OD8_SETTING_FAN_ACOUSTIC_LIMIT:
+		od_table->FanMaximumRpm = (uint16_t)value;
+		break;
+
+	case OD8_SETTING_FAN_MIN_SPEED:
+		od_table->FanMinimumPwm = (uint16_t)value;
+		break;
+
+	case OD8_SETTING_FAN_TARGET_TEMP:
+		od_table->FanTargetTemperature = (uint16_t)value;
+		break;
+
+	case OD8_SETTING_OPERATING_TEMP_MAX:
+		od_table->MaxOpTemp = (uint16_t)value;
+		break;
+	}
+
+	return 0;
+}
+
+static int vega20_set_od_percentage(struct smu_context *smu,
+				    enum pp_clock_type type,
+				    uint32_t value)
+{
+	struct smu_dpm_context *smu_dpm = &smu->smu_dpm;
+	struct vega20_dpm_table *dpm_table = NULL;
+	struct vega20_dpm_table *golden_table = NULL;
+	struct vega20_single_dpm_table *single_dpm_table;
+	struct vega20_single_dpm_table *golden_dpm_table;
+	uint32_t od_clk, index;
+	int ret = 0;
+	int feature_enabled;
+	PPCLK_e clk_id;
+
+	mutex_lock(&(smu->mutex));
+
+	dpm_table = smu_dpm->dpm_context;
+	golden_table = smu_dpm->golden_dpm_context;
+
+	switch (type) {
+	case OD_SCLK:
+		single_dpm_table = &(dpm_table->gfx_table);
+		golden_dpm_table = &(golden_table->gfx_table);
+		feature_enabled = smu_feature_is_enabled(smu, FEATURE_DPM_GFXCLK_BIT);
+		clk_id = PPCLK_GFXCLK;
+		index = OD8_SETTING_GFXCLK_FMAX;
+		break;
+	case OD_MCLK:
+		single_dpm_table = &(dpm_table->mem_table);
+		golden_dpm_table = &(golden_table->mem_table);
+		feature_enabled = smu_feature_is_enabled(smu, FEATURE_DPM_UCLK_BIT);
+		clk_id = PPCLK_UCLK;
+		index = OD8_SETTING_UCLK_FMAX;
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+
+	if (ret)
+		goto set_od_failed;
+
+	od_clk = golden_dpm_table->dpm_levels[golden_dpm_table->count - 1].value * value;
+	od_clk /= 100;
+	od_clk += golden_dpm_table->dpm_levels[golden_dpm_table->count - 1].value;
+
+	ret = smu_update_od8_settings(smu, index, od_clk);
+	if (ret) {
+		pr_err("[Setoverdrive] failed to set od clk!\n");
+		goto set_od_failed;
+	}
+
+	if (feature_enabled) {
+		ret = vega20_set_single_dpm_table(smu, single_dpm_table,
+						  clk_id);
+		if (ret) {
+			pr_err("[Setoverdrive] failed to refresh dpm table!\n");
+			goto set_od_failed;
+		}
+	} else {
+		single_dpm_table->count = 1;
+		single_dpm_table->dpm_levels[0].value = smu->smu_table.boot_values.gfxclk / 100;
+	}
+
+	ret = smu_handle_task(smu, smu_dpm->dpm_level,
+			      AMD_PP_TASK_READJUST_POWER_STATE);
+
+set_od_failed:
+	mutex_unlock(&(smu->mutex));
+
+	return ret;
+}
+
+static int vega20_odn_edit_dpm_table(struct smu_context *smu,
+				     enum PP_OD_DPM_TABLE_COMMAND type,
+				     long *input, uint32_t size)
+{
+	struct smu_table_context *table_context = &smu->smu_table;
+	OverDriveTable_t *od_table =
+		(OverDriveTable_t *)(table_context->overdrive_table);
+	struct smu_dpm_context *smu_dpm = &smu->smu_dpm;
+	struct vega20_dpm_table *dpm_table = NULL;
+	struct vega20_single_dpm_table *single_dpm_table;
+	struct vega20_od8_settings *od8_settings =
+		(struct vega20_od8_settings *)table_context->od8_settings;
+	struct pp_clock_levels_with_latency clocks;
+	int32_t input_index, input_clk, input_vol, i;
+	int od8_id;
+	int ret = 0;
+
+	dpm_table = smu_dpm->dpm_context;
+
+	if (!input) {
+		pr_warn("NULL user input for clock and voltage\n");
+		return -EINVAL;
+	}
+
+	switch (type) {
+	case PP_OD_EDIT_SCLK_VDDC_TABLE:
+		if (!(od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FMIN].feature_id &&
+		      od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FMAX].feature_id)) {
+			pr_info("Sclk min/max frequency overdrive not supported\n");
+			return -EOPNOTSUPP;
+		}
+
+		for (i = 0; i < size; i += 2) {
+			if (i + 2 > size) {
+				pr_info("invalid number of input parameters %d\n", size);
+				return -EINVAL;
+			}
+
+			input_index = input[i];
+			input_clk = input[i + 1];
+
+			if (input_index != 0 && input_index != 1) {
+				pr_info("Invalid index %d\n", input_index);
+				pr_info("Support min/max sclk frequency settingonly which index by 0/1\n");
+				return -EINVAL;
+			}
+
+			if (input_clk < od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FMIN].min_value ||
+			    input_clk > od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FMAX].max_value) {
+				pr_info("clock freq %d is not within allowed range [%d - %d]\n",
+					input_clk,
+					od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FMIN].min_value,
+					od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FMAX].max_value);
+				return -EINVAL;
+			}
+
+			if (input_index == 0 && od_table->GfxclkFmin != input_clk) {
+				od_table->GfxclkFmin = input_clk;
+				table_context->od_gfxclk_update = true;
+			} else if (input_index == 1 && od_table->GfxclkFmax != input_clk) {
+				od_table->GfxclkFmax = input_clk;
+				table_context->od_gfxclk_update = true;
+			}
+		}
+
+		break;
+
+	case PP_OD_EDIT_MCLK_VDDC_TABLE:
+		if (!od8_settings->od8_settings_array[OD8_SETTING_UCLK_FMAX].feature_id) {
+			pr_info("Mclk max frequency overdrive not supported\n");
+			return -EOPNOTSUPP;
+		}
+
+		single_dpm_table = &(dpm_table->mem_table);
+		ret = vega20_get_clk_table(smu, &clocks, single_dpm_table);
+		if (ret) {
+			pr_err("Attempt to get memory clk levels Failed!");
+			return ret;
+		}
+
+		for (i = 0; i < size; i += 2) {
+			if (i + 2 > size) {
+				pr_info("invalid number of input parameters %d\n",
+					 size);
+				return -EINVAL;
+			}
+
+			input_index = input[i];
+			input_clk = input[i + 1];
+
+			if (input_index != 1) {
+				pr_info("Invalid index %d\n", input_index);
+				pr_info("Support max Mclk frequency setting only which index by 1\n");
+				return -EINVAL;
+			}
+
+			if (input_clk < clocks.data[0].clocks_in_khz / 1000 ||
+			    input_clk > od8_settings->od8_settings_array[OD8_SETTING_UCLK_FMAX].max_value) {
+				pr_info("clock freq %d is not within allowed range [%d - %d]\n",
+					input_clk,
+					clocks.data[0].clocks_in_khz / 1000,
+					od8_settings->od8_settings_array[OD8_SETTING_UCLK_FMAX].max_value);
+				return -EINVAL;
+			}
+
+			if (input_index == 1 && od_table->UclkFmax != input_clk) {
+				table_context->od_gfxclk_update = true;
+				od_table->UclkFmax = input_clk;
+			}
+		}
+
+		break;
+
+	case PP_OD_EDIT_VDDC_CURVE:
+		if (!(od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FREQ1].feature_id &&
+		      od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FREQ2].feature_id &&
+		      od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_FREQ3].feature_id &&
+		      od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE1].feature_id &&
+		      od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE2].feature_id &&
+		      od8_settings->od8_settings_array[OD8_SETTING_GFXCLK_VOLTAGE3].feature_id)) {
+			pr_info("Voltage curve calibrate not supported\n");
+			return -EOPNOTSUPP;
+		}
+
+		for (i = 0; i < size; i += 3) {
+			if (i + 3 > size) {
+				pr_info("invalid number of input parameters %d\n",
+					size);
+				return -EINVAL;
+			}
+
+			input_index = input[i];
+			input_clk = input[i + 1];
+			input_vol = input[i + 2];
+
+			if (input_index > 2) {
+				pr_info("Setting for point %d is not supported\n",
+					input_index + 1);
+				pr_info("Three supported points index by 0, 1, 2\n");
+				return -EINVAL;
+			}
+
+			od8_id = OD8_SETTING_GFXCLK_FREQ1 + 2 * input_index;
+			if (input_clk < od8_settings->od8_settings_array[od8_id].min_value ||
+			    input_clk > od8_settings->od8_settings_array[od8_id].max_value) {
+				pr_info("clock freq %d is not within allowed range [%d - %d]\n",
+					input_clk,
+					od8_settings->od8_settings_array[od8_id].min_value,
+					od8_settings->od8_settings_array[od8_id].max_value);
+				return -EINVAL;
+			}
+
+			od8_id = OD8_SETTING_GFXCLK_VOLTAGE1 + 2 * input_index;
+			if (input_vol < od8_settings->od8_settings_array[od8_id].min_value ||
+			    input_vol > od8_settings->od8_settings_array[od8_id].max_value) {
+				pr_info("clock voltage %d is not within allowed range [%d- %d]\n",
+					input_vol,
+					od8_settings->od8_settings_array[od8_id].min_value,
+					od8_settings->od8_settings_array[od8_id].max_value);
+				return -EINVAL;
+			}
+
+			switch (input_index) {
+			case 0:
+				od_table->GfxclkFreq1 = input_clk;
+				od_table->GfxclkVolt1 = input_vol * VOLTAGE_SCALE;
+				break;
+			case 1:
+				od_table->GfxclkFreq2 = input_clk;
+				od_table->GfxclkVolt2 = input_vol * VOLTAGE_SCALE;
+				break;
+			case 2:
+				od_table->GfxclkFreq3 = input_clk;
+				od_table->GfxclkVolt3 = input_vol * VOLTAGE_SCALE;
+				break;
+			}
+		}
+
+		break;
+
+	case PP_OD_RESTORE_DEFAULT_TABLE:
+		ret = smu_update_table(smu, TABLE_OVERDRIVE, table_context->overdrive_table, false);
+		if (ret) {
+			pr_err("Failed to export over drive table!\n");
+			return ret;
+		}
+
+		break;
+
+	case PP_OD_COMMIT_DPM_TABLE:
+		ret = smu_update_table(smu, TABLE_OVERDRIVE, table_context->overdrive_table, true);
+		if (ret) {
+			pr_err("Failed to import over drive table!\n");
+			return ret;
+		}
+
+		/* retrieve updated gfxclk table */
+		if (table_context->od_gfxclk_update) {
+			table_context->od_gfxclk_update = false;
+			single_dpm_table = &(dpm_table->gfx_table);
+
+			if (smu_feature_is_enabled(smu, FEATURE_DPM_GFXCLK_BIT)) {
+				ret = vega20_set_single_dpm_table(smu, single_dpm_table,
+								  PPCLK_GFXCLK);
+				if (ret) {
+					pr_err("[Setoverdrive] failed to refresh dpm table!\n");
+					return ret;
+				}
+			} else {
+				single_dpm_table->count = 1;
+				single_dpm_table->dpm_levels[0].value = smu->smu_table.boot_values.gfxclk / 100;
+			}
+		}
+
+		break;
+
+	default:
+		return -EINVAL;
+	}
+
+	if (type == PP_OD_COMMIT_DPM_TABLE) {
+		mutex_lock(&(smu->mutex));
+		ret = smu_handle_task(smu, smu_dpm->dpm_level,
+				      AMD_PP_TASK_READJUST_POWER_STATE);
+		mutex_unlock(&(smu->mutex));
+	}
+
+	return ret;
+}
+
+static const struct pptable_funcs vega20_ppt_funcs = {
+	.alloc_dpm_context = vega20_allocate_dpm_context,
+	.store_powerplay_table = vega20_store_powerplay_table,
+	.check_powerplay_table = vega20_check_powerplay_table,
+	.append_powerplay_table = vega20_append_powerplay_table,
+	.get_smu_msg_index = vega20_get_smu_msg_index,
+	.run_afll_btc = vega20_run_btc_afll,
+	.get_unallowed_feature_mask = vega20_get_unallowed_feature_mask,
+	.get_current_power_state = vega20_get_current_power_state,
+	.set_default_dpm_table = vega20_set_default_dpm_table,
+	.set_power_state = NULL,
+	.populate_umd_state_clk = vega20_populate_umd_state_clk,
+	.print_clk_levels = vega20_print_clk_levels,
+	.force_clk_levels = vega20_force_clk_levels,
+	.get_clock_by_type_with_latency = vega20_get_clock_by_type_with_latency,
+	.set_default_od8_settings = vega20_set_default_od8_setttings,
+	.get_od_percentage = vega20_get_od_percentage,
+	.get_performance_level = vega20_get_performance_level,
+	.force_performance_level = vega20_force_performance_level,
+	.update_specified_od8_value = vega20_update_specified_od8_value,
+	.set_od_percentage = vega20_set_od_percentage,
+	.od_edit_dpm_table = vega20_odn_edit_dpm_table,
+	.pre_display_config_changed = vega20_pre_display_config_changed,
+	.display_config_changed = vega20_display_config_changed,
+	.apply_clocks_adjust_rules = vega20_apply_clocks_adjust_rules,
+	.notify_smc_dispaly_config = vega20_notify_smc_dispaly_config,
+	.force_dpm_limit_value = vega20_force_dpm_limit_value,
+	.unforce_dpm_levels = vega20_unforce_dpm_levels,
+	.upload_dpm_level = vega20_upload_dpm_level,
+	.get_profiling_clk_mask = vega20_get_profiling_clk_mask,
+};
+
+void vega20_set_ppt_funcs(struct smu_context *smu)
+{
+	smu->ppt_funcs = &vega20_ppt_funcs;
+	smu->smc_if_version = SMU11_DRIVER_IF_VERSION;
+}
diff --git a/drivers/gpu/drm/amd/powerplay/vega20_ppt.h b/drivers/gpu/drm/amd/powerplay/vega20_ppt.h
new file mode 100644
index 000000000000..5a0d2af63173
--- /dev/null
+++ b/drivers/gpu/drm/amd/powerplay/vega20_ppt.h
@@ -0,0 +1,129 @@
+/*
+ * Copyright 2019 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#ifndef __VEGA20_PPT_H__
+#define __VEGA20_PPT_H__
+
+#define VEGA20_UMD_PSTATE_GFXCLK_LEVEL         0x3
+#define VEGA20_UMD_PSTATE_SOCCLK_LEVEL         0x3
+#define VEGA20_UMD_PSTATE_MCLK_LEVEL           0x2
+#define VEGA20_UMD_PSTATE_UVDCLK_LEVEL         0x3
+#define VEGA20_UMD_PSTATE_VCEMCLK_LEVEL        0x3
+
+#define MAX_REGULAR_DPM_NUMBER 16
+#define MAX_PCIE_CONF 2
+
+#define VOLTAGE_SCALE 4
+#define AVFS_CURVE 0
+#define OD8_HOTCURVE_TEMPERATURE 85
+
+struct vega20_dpm_level {
+        bool            enabled;
+        uint32_t        value;
+        uint32_t        param1;
+};
+
+struct vega20_dpm_state {
+        uint32_t  soft_min_level;
+        uint32_t  soft_max_level;
+        uint32_t  hard_min_level;
+        uint32_t  hard_max_level;
+};
+
+struct vega20_single_dpm_table {
+        uint32_t                count;
+        struct vega20_dpm_state dpm_state;
+        struct vega20_dpm_level dpm_levels[MAX_REGULAR_DPM_NUMBER];
+};
+
+struct vega20_pcie_table {
+        uint16_t count;
+        uint8_t  pcie_gen[MAX_PCIE_CONF];
+        uint8_t  pcie_lane[MAX_PCIE_CONF];
+        uint32_t lclk[MAX_PCIE_CONF];
+};
+
+struct vega20_dpm_table {
+	struct vega20_single_dpm_table  soc_table;
+        struct vega20_single_dpm_table  gfx_table;
+        struct vega20_single_dpm_table  mem_table;
+        struct vega20_single_dpm_table  eclk_table;
+        struct vega20_single_dpm_table  vclk_table;
+        struct vega20_single_dpm_table  dclk_table;
+        struct vega20_single_dpm_table  dcef_table;
+        struct vega20_single_dpm_table  pixel_table;
+        struct vega20_single_dpm_table  display_table;
+        struct vega20_single_dpm_table  phy_table;
+        struct vega20_single_dpm_table  fclk_table;
+        struct vega20_pcie_table        pcie_table;
+};
+
+enum OD8_FEATURE_ID
+{
+	OD8_GFXCLK_LIMITS               = 1 << 0,
+	OD8_GFXCLK_CURVE                = 1 << 1,
+	OD8_UCLK_MAX                    = 1 << 2,
+	OD8_POWER_LIMIT                 = 1 << 3,
+	OD8_ACOUSTIC_LIMIT_SCLK         = 1 << 4,   //FanMaximumRpm
+	OD8_FAN_SPEED_MIN               = 1 << 5,   //FanMinimumPwm
+	OD8_TEMPERATURE_FAN             = 1 << 6,   //FanTargetTemperature
+	OD8_TEMPERATURE_SYSTEM          = 1 << 7,   //MaxOpTemp
+	OD8_MEMORY_TIMING_TUNE          = 1 << 8,
+	OD8_FAN_ZERO_RPM_CONTROL        = 1 << 9
+};
+
+enum OD8_SETTING_ID
+{
+	OD8_SETTING_GFXCLK_FMIN = 0,
+	OD8_SETTING_GFXCLK_FMAX,
+	OD8_SETTING_GFXCLK_FREQ1,
+	OD8_SETTING_GFXCLK_VOLTAGE1,
+	OD8_SETTING_GFXCLK_FREQ2,
+	OD8_SETTING_GFXCLK_VOLTAGE2,
+	OD8_SETTING_GFXCLK_FREQ3,
+	OD8_SETTING_GFXCLK_VOLTAGE3,
+	OD8_SETTING_UCLK_FMAX,
+	OD8_SETTING_POWER_PERCENTAGE,
+	OD8_SETTING_FAN_ACOUSTIC_LIMIT,
+	OD8_SETTING_FAN_MIN_SPEED,
+	OD8_SETTING_FAN_TARGET_TEMP,
+	OD8_SETTING_OPERATING_TEMP_MAX,
+	OD8_SETTING_AC_TIMING,
+	OD8_SETTING_FAN_ZERO_RPM_CONTROL,
+	OD8_SETTING_COUNT
+};
+
+struct vega20_od8_single_setting {
+	uint32_t	feature_id;
+	int32_t		min_value;
+	int32_t		max_value;
+	int32_t		current_value;
+	int32_t		default_value;
+};
+
+struct vega20_od8_settings {
+	struct vega20_od8_single_setting	od8_settings_array[OD8_SETTING_COUNT];
+};
+
+extern void vega20_set_ppt_funcs(struct smu_context *smu);
+
+#endif
diff --git a/drivers/gpu/drm/arm/display/include/malidp_product.h b/drivers/gpu/drm/arm/display/include/malidp_product.h
index b35fc5db866b..1053b11352eb 100644
--- a/drivers/gpu/drm/arm/display/include/malidp_product.h
+++ b/drivers/gpu/drm/arm/display/include/malidp_product.h
@@ -20,4 +20,16 @@
 /* Mali-display product IDs */
 #define MALIDP_D71_PRODUCT_ID   0x0071
 
+union komeda_config_id {
+	struct {
+		__u32	max_line_sz:16,
+			n_pipelines:2,
+			n_scalers:2, /* number of scalers per pipeline */
+			n_layers:3, /* number of layers per pipeline */
+			n_richs:3, /* number of rich layers per pipeline */
+			reserved_bits:6;
+	};
+	__u32 value;
+};
+
 #endif /* _MALIDP_PRODUCT_H_ */
diff --git a/drivers/gpu/drm/arm/display/include/malidp_utils.h b/drivers/gpu/drm/arm/display/include/malidp_utils.h
index 63cc47cefcf8..8cfd91196e15 100644
--- a/drivers/gpu/drm/arm/display/include/malidp_utils.h
+++ b/drivers/gpu/drm/arm/display/include/malidp_utils.h
@@ -7,10 +7,41 @@
 #ifndef _MALIDP_UTILS_
 #define _MALIDP_UTILS_
 
+#include <linux/delay.h>
+
 #define has_bit(nr, mask)	(BIT(nr) & (mask))
 #define has_bits(bits, mask)	(((bits) & (mask)) == (bits))
 
 #define dp_for_each_set_bit(bit, mask) \
 	for_each_set_bit((bit), ((unsigned long *)&(mask)), sizeof(mask) * 8)
 
+#define dp_wait_cond(__cond, __tries, __min_range, __max_range)	\
+({							\
+	int num_tries = __tries;			\
+	while (!__cond && (num_tries > 0)) {		\
+		usleep_range(__min_range, __max_range);	\
+		if (__cond)				\
+			break;				\
+		num_tries--;				\
+	}						\
+	num_tries;					\
+})
+
+/* the restriction of range is [start, end] */
+struct malidp_range {
+	u32 start;
+	u32 end;
+};
+
+static inline void set_range(struct malidp_range *rg, u32 start, u32 end)
+{
+	rg->start = start;
+	rg->end   = end;
+}
+
+static inline bool in_range(struct malidp_range *rg, u32 v)
+{
+	return (v >= rg->start) && (v <= rg->end);
+}
+
 #endif /* _MALIDP_UTILS_ */
diff --git a/drivers/gpu/drm/arm/display/komeda/Makefile b/drivers/gpu/drm/arm/display/komeda/Makefile
index 1b875e5dc0f6..412eeba8c39f 100644
--- a/drivers/gpu/drm/arm/display/komeda/Makefile
+++ b/drivers/gpu/drm/arm/display/komeda/Makefile
@@ -1,14 +1,15 @@
 # SPDX-License-Identifier: GPL-2.0
 
 ccflags-y := \
-	-I$(src)/../include \
-	-I$(src)
+	-I $(srctree)/$(src)/../include \
+	-I $(srctree)/$(src)
 
 komeda-y := \
 	komeda_drv.o \
 	komeda_dev.o \
 	komeda_format_caps.o \
 	komeda_pipeline.o \
+	komeda_pipeline_state.o \
 	komeda_framebuffer.o \
 	komeda_kms.o \
 	komeda_crtc.o \
@@ -16,6 +17,7 @@ komeda-y := \
 	komeda_private_obj.o
 
 komeda-y += \
-	d71/d71_dev.o
+	d71/d71_dev.o \
+	d71/d71_component.o
 
 obj-$(CONFIG_DRM_KOMEDA) += komeda.o
diff --git a/drivers/gpu/drm/arm/display/komeda/d71/d71_component.c b/drivers/gpu/drm/arm/display/komeda/d71/d71_component.c
new file mode 100644
index 000000000000..031e5f305a3c
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/komeda/d71/d71_component.c
@@ -0,0 +1,685 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+
+#include <drm/drm_print.h>
+#include "d71_dev.h"
+#include "komeda_kms.h"
+#include "malidp_io.h"
+#include "komeda_framebuffer.h"
+
+static void get_resources_id(u32 hw_id, u32 *pipe_id, u32 *comp_id)
+{
+	u32 id = BLOCK_INFO_BLK_ID(hw_id);
+	u32 pipe = id;
+
+	switch (BLOCK_INFO_BLK_TYPE(hw_id)) {
+	case D71_BLK_TYPE_LPU_WB_LAYER:
+		id = KOMEDA_COMPONENT_WB_LAYER;
+		break;
+	case D71_BLK_TYPE_CU_SPLITTER:
+		id = KOMEDA_COMPONENT_SPLITTER;
+		break;
+	case D71_BLK_TYPE_CU_SCALER:
+		pipe = id / D71_PIPELINE_MAX_SCALERS;
+		id %= D71_PIPELINE_MAX_SCALERS;
+		id += KOMEDA_COMPONENT_SCALER0;
+		break;
+	case D71_BLK_TYPE_CU:
+		id += KOMEDA_COMPONENT_COMPIZ0;
+		break;
+	case D71_BLK_TYPE_LPU_LAYER:
+		pipe = id / D71_PIPELINE_MAX_LAYERS;
+		id %= D71_PIPELINE_MAX_LAYERS;
+		id += KOMEDA_COMPONENT_LAYER0;
+		break;
+	case D71_BLK_TYPE_DOU_IPS:
+		id += KOMEDA_COMPONENT_IPS0;
+		break;
+	case D71_BLK_TYPE_CU_MERGER:
+		id = KOMEDA_COMPONENT_MERGER;
+		break;
+	case D71_BLK_TYPE_DOU:
+		id = KOMEDA_COMPONENT_TIMING_CTRLR;
+		break;
+	default:
+		id = 0xFFFFFFFF;
+	}
+
+	if (comp_id)
+		*comp_id = id;
+
+	if (pipe_id)
+		*pipe_id = pipe;
+}
+
+static u32 get_valid_inputs(struct block_header *blk)
+{
+	u32 valid_inputs = 0, comp_id;
+	int i;
+
+	for (i = 0; i < PIPELINE_INFO_N_VALID_INPUTS(blk->pipeline_info); i++) {
+		get_resources_id(blk->input_ids[i], NULL, &comp_id);
+		if (comp_id == 0xFFFFFFFF)
+			continue;
+		valid_inputs |= BIT(comp_id);
+	}
+
+	return valid_inputs;
+}
+
+static void get_values_from_reg(void __iomem *reg, u32 offset,
+				u32 count, u32 *val)
+{
+	u32 i, addr;
+
+	for (i = 0; i < count; i++) {
+		addr = offset + (i << 2);
+		/* 0xA4 is WO register */
+		if (addr != 0xA4)
+			val[i] = malidp_read32(reg, addr);
+		else
+			val[i] = 0xDEADDEAD;
+	}
+}
+
+static void dump_block_header(struct seq_file *sf, void __iomem *reg)
+{
+	struct block_header hdr;
+	u32 i, n_input, n_output;
+
+	d71_read_block_header(reg, &hdr);
+	seq_printf(sf, "BLOCK_INFO:\t\t0x%X\n", hdr.block_info);
+	seq_printf(sf, "PIPELINE_INFO:\t\t0x%X\n", hdr.pipeline_info);
+
+	n_output = PIPELINE_INFO_N_OUTPUTS(hdr.pipeline_info);
+	n_input  = PIPELINE_INFO_N_VALID_INPUTS(hdr.pipeline_info);
+
+	for (i = 0; i < n_input; i++)
+		seq_printf(sf, "VALID_INPUT_ID%u:\t0x%X\n",
+			   i, hdr.input_ids[i]);
+
+	for (i = 0; i < n_output; i++)
+		seq_printf(sf, "OUTPUT_ID%u:\t\t0x%X\n",
+			   i, hdr.output_ids[i]);
+}
+
+static u32 to_rot_ctrl(u32 rot)
+{
+	u32 lr_ctrl = 0;
+
+	switch (rot & DRM_MODE_ROTATE_MASK) {
+	case DRM_MODE_ROTATE_0:
+		lr_ctrl |= L_ROT(L_ROT_R0);
+		break;
+	case DRM_MODE_ROTATE_90:
+		lr_ctrl |= L_ROT(L_ROT_R90);
+		break;
+	case DRM_MODE_ROTATE_180:
+		lr_ctrl |= L_ROT(L_ROT_R180);
+		break;
+	case DRM_MODE_ROTATE_270:
+		lr_ctrl |= L_ROT(L_ROT_R270);
+		break;
+	}
+
+	if (rot & DRM_MODE_REFLECT_X)
+		lr_ctrl |= L_HFLIP;
+	if (rot & DRM_MODE_REFLECT_Y)
+		lr_ctrl |= L_VFLIP;
+
+	return lr_ctrl;
+}
+
+static inline u32 to_d71_input_id(struct komeda_component_output *output)
+{
+	struct komeda_component *comp = output->component;
+
+	return comp ? (comp->hw_id + output->output_port) : 0;
+}
+
+static void d71_layer_disable(struct komeda_component *c)
+{
+	malidp_write32_mask(c->reg, BLK_CONTROL, L_EN, 0);
+}
+
+static void d71_layer_update(struct komeda_component *c,
+			     struct komeda_component_state *state)
+{
+	struct komeda_layer_state *st = to_layer_st(state);
+	struct drm_plane_state *plane_st = state->plane->state;
+	struct drm_framebuffer *fb = plane_st->fb;
+	struct komeda_fb *kfb = to_kfb(fb);
+	u32 __iomem *reg = c->reg;
+	u32 ctrl_mask = L_EN | L_ROT(L_ROT_R270) | L_HFLIP | L_VFLIP | L_TBU_EN;
+	u32 ctrl = L_EN | to_rot_ctrl(st->rot);
+	int i;
+
+	for (i = 0; i < fb->format->num_planes; i++) {
+		malidp_write32(reg,
+			       BLK_P0_PTR_LOW + i * LAYER_PER_PLANE_REGS * 4,
+			       lower_32_bits(st->addr[i]));
+		malidp_write32(reg,
+			       BLK_P0_PTR_HIGH + i * LAYER_PER_PLANE_REGS * 4,
+			       upper_32_bits(st->addr[i]));
+		if (i >= 2)
+			break;
+
+		malidp_write32(reg,
+			       BLK_P0_STRIDE + i * LAYER_PER_PLANE_REGS * 4,
+			       fb->pitches[i] & 0xFFFF);
+	}
+
+	malidp_write32(reg, LAYER_FMT, kfb->format_caps->hw_id);
+	malidp_write32(reg, BLK_IN_SIZE, HV_SIZE(st->hsize, st->vsize));
+
+	malidp_write32_mask(reg, BLK_CONTROL, ctrl_mask, ctrl);
+}
+
+static void d71_layer_dump(struct komeda_component *c, struct seq_file *sf)
+{
+	u32 v[15], i;
+	bool rich, rgb2rgb;
+	char *prefix;
+
+	get_values_from_reg(c->reg, LAYER_INFO, 1, &v[14]);
+	if (v[14] & 0x1) {
+		rich = true;
+		prefix = "LR_";
+	} else {
+		rich = false;
+		prefix = "LS_";
+	}
+
+	rgb2rgb = !!(v[14] & L_INFO_CM);
+
+	dump_block_header(sf, c->reg);
+
+	seq_printf(sf, "%sLAYER_INFO:\t\t0x%X\n", prefix, v[14]);
+
+	get_values_from_reg(c->reg, 0xD0, 1, v);
+	seq_printf(sf, "%sCONTROL:\t\t0x%X\n", prefix, v[0]);
+	if (rich) {
+		get_values_from_reg(c->reg, 0xD4, 1, v);
+		seq_printf(sf, "LR_RICH_CONTROL:\t0x%X\n", v[0]);
+	}
+	get_values_from_reg(c->reg, 0xD8, 4, v);
+	seq_printf(sf, "%sFORMAT:\t\t0x%X\n", prefix, v[0]);
+	seq_printf(sf, "%sIT_COEFFTAB:\t\t0x%X\n", prefix, v[1]);
+	seq_printf(sf, "%sIN_SIZE:\t\t0x%X\n", prefix, v[2]);
+	seq_printf(sf, "%sPALPHA:\t\t0x%X\n", prefix, v[3]);
+
+	get_values_from_reg(c->reg, 0x100, 3, v);
+	seq_printf(sf, "%sP0_PTR_LOW:\t\t0x%X\n", prefix, v[0]);
+	seq_printf(sf, "%sP0_PTR_HIGH:\t\t0x%X\n", prefix, v[1]);
+	seq_printf(sf, "%sP0_STRIDE:\t\t0x%X\n", prefix, v[2]);
+
+	get_values_from_reg(c->reg, 0x110, 2, v);
+	seq_printf(sf, "%sP1_PTR_LOW:\t\t0x%X\n", prefix, v[0]);
+	seq_printf(sf, "%sP1_PTR_HIGH:\t\t0x%X\n", prefix, v[1]);
+	if (rich) {
+		get_values_from_reg(c->reg, 0x118, 1, v);
+		seq_printf(sf, "LR_P1_STRIDE:\t\t0x%X\n", v[0]);
+
+		get_values_from_reg(c->reg, 0x120, 2, v);
+		seq_printf(sf, "LR_P2_PTR_LOW:\t\t0x%X\n", v[0]);
+		seq_printf(sf, "LR_P2_PTR_HIGH:\t\t0x%X\n", v[1]);
+
+		get_values_from_reg(c->reg, 0x130, 12, v);
+		for (i = 0; i < 12; i++)
+			seq_printf(sf, "LR_YUV_RGB_COEFF%u:\t0x%X\n", i, v[i]);
+	}
+
+	if (rgb2rgb) {
+		get_values_from_reg(c->reg, LAYER_RGB_RGB_COEFF0, 12, v);
+		for (i = 0; i < 12; i++)
+			seq_printf(sf, "LS_RGB_RGB_COEFF%u:\t0x%X\n", i, v[i]);
+	}
+
+	get_values_from_reg(c->reg, 0x160, 3, v);
+	seq_printf(sf, "%sAD_CONTROL:\t\t0x%X\n", prefix, v[0]);
+	seq_printf(sf, "%sAD_H_CROP:\t\t0x%X\n", prefix, v[1]);
+	seq_printf(sf, "%sAD_V_CROP:\t\t0x%X\n", prefix, v[2]);
+}
+
+static struct komeda_component_funcs d71_layer_funcs = {
+	.update		= d71_layer_update,
+	.disable	= d71_layer_disable,
+	.dump_register	= d71_layer_dump,
+};
+
+static int d71_layer_init(struct d71_dev *d71,
+			  struct block_header *blk, u32 __iomem *reg)
+{
+	struct komeda_component *c;
+	struct komeda_layer *layer;
+	u32 pipe_id, layer_id, layer_info;
+
+	get_resources_id(blk->block_info, &pipe_id, &layer_id);
+	c = komeda_component_add(&d71->pipes[pipe_id]->base, sizeof(*layer),
+				 layer_id,
+				 BLOCK_INFO_INPUT_ID(blk->block_info),
+				 &d71_layer_funcs, 0,
+				 get_valid_inputs(blk),
+				 1, reg, "LPU%d_LAYER%d", pipe_id, layer_id);
+	if (IS_ERR(c)) {
+		DRM_ERROR("Failed to add layer component\n");
+		return PTR_ERR(c);
+	}
+
+	layer = to_layer(c);
+	layer_info = malidp_read32(reg, LAYER_INFO);
+
+	if (layer_info & L_INFO_RF)
+		layer->layer_type = KOMEDA_FMT_RICH_LAYER;
+	else
+		layer->layer_type = KOMEDA_FMT_SIMPLE_LAYER;
+
+	set_range(&layer->hsize_in, 4, d71->max_line_size);
+	set_range(&layer->vsize_in, 4, d71->max_vsize);
+
+	malidp_write32(reg, LAYER_PALPHA, D71_PALPHA_DEF_MAP);
+
+	layer->supported_rots = DRM_MODE_ROTATE_MASK | DRM_MODE_REFLECT_MASK;
+
+	return 0;
+}
+
+static int d71_wb_layer_init(struct d71_dev *d71,
+			     struct block_header *blk, u32 __iomem *reg)
+{
+	DRM_DEBUG("Detect D71_Wb_Layer.\n");
+
+	return 0;
+}
+
+static void d71_component_disable(struct komeda_component *c)
+{
+	u32 __iomem *reg = c->reg;
+	u32 i;
+
+	malidp_write32(reg, BLK_CONTROL, 0);
+
+	for (i = 0; i < c->max_active_inputs; i++)
+		malidp_write32(reg, BLK_INPUT_ID0 + (i << 2), 0);
+}
+
+static void compiz_enable_input(u32 __iomem *id_reg,
+				u32 __iomem *cfg_reg,
+				u32 input_hw_id,
+				struct komeda_compiz_input_cfg *cin)
+{
+	u32 ctrl = CU_INPUT_CTRL_EN;
+	u8 blend = cin->pixel_blend_mode;
+
+	if (blend == DRM_MODE_BLEND_PIXEL_NONE)
+		ctrl |= CU_INPUT_CTRL_PAD;
+	else if (blend == DRM_MODE_BLEND_PREMULTI)
+		ctrl |= CU_INPUT_CTRL_PMUL;
+
+	ctrl |= CU_INPUT_CTRL_ALPHA(cin->layer_alpha);
+
+	malidp_write32(id_reg, BLK_INPUT_ID0, input_hw_id);
+
+	malidp_write32(cfg_reg, CU_INPUT0_SIZE,
+		       HV_SIZE(cin->hsize, cin->vsize));
+	malidp_write32(cfg_reg, CU_INPUT0_OFFSET,
+		       HV_OFFSET(cin->hoffset, cin->voffset));
+	malidp_write32(cfg_reg, CU_INPUT0_CONTROL, ctrl);
+}
+
+static void d71_compiz_update(struct komeda_component *c,
+			      struct komeda_component_state *state)
+{
+	struct komeda_compiz_state *st = to_compiz_st(state);
+	u32 __iomem *reg = c->reg;
+	u32 __iomem *id_reg, *cfg_reg;
+	u32 index, input_hw_id;
+
+	for_each_changed_input(state, index) {
+		id_reg = reg + index;
+		cfg_reg = reg + index * CU_PER_INPUT_REGS;
+		input_hw_id = to_d71_input_id(&state->inputs[index]);
+		if (state->active_inputs & BIT(index)) {
+			compiz_enable_input(id_reg, cfg_reg,
+					    input_hw_id, &st->cins[index]);
+		} else {
+			malidp_write32(id_reg, BLK_INPUT_ID0, 0);
+			malidp_write32(cfg_reg, CU_INPUT0_CONTROL, 0);
+		}
+	}
+
+	malidp_write32(reg, BLK_SIZE, HV_SIZE(st->hsize, st->vsize));
+}
+
+static void d71_compiz_dump(struct komeda_component *c, struct seq_file *sf)
+{
+	u32 v[8], i;
+
+	dump_block_header(sf, c->reg);
+
+	get_values_from_reg(c->reg, 0x80, 5, v);
+	for (i = 0; i < 5; i++)
+		seq_printf(sf, "CU_INPUT_ID%u:\t\t0x%X\n", i, v[i]);
+
+	get_values_from_reg(c->reg, 0xA0, 5, v);
+	seq_printf(sf, "CU_IRQ_RAW_STATUS:\t0x%X\n", v[0]);
+	seq_printf(sf, "CU_IRQ_CLEAR:\t\t0x%X\n", v[1]);
+	seq_printf(sf, "CU_IRQ_MASK:\t\t0x%X\n", v[2]);
+	seq_printf(sf, "CU_IRQ_STATUS:\t\t0x%X\n", v[3]);
+	seq_printf(sf, "CU_STATUS:\t\t0x%X\n", v[4]);
+
+	get_values_from_reg(c->reg, 0xD0, 2, v);
+	seq_printf(sf, "CU_CONTROL:\t\t0x%X\n", v[0]);
+	seq_printf(sf, "CU_SIZE:\t\t0x%X\n", v[1]);
+
+	get_values_from_reg(c->reg, 0xDC, 1, v);
+	seq_printf(sf, "CU_BG_COLOR:\t\t0x%X\n", v[0]);
+
+	for (i = 0, v[4] = 0xE0; i < 5; i++, v[4] += 0x10) {
+		get_values_from_reg(c->reg, v[4], 3, v);
+		seq_printf(sf, "CU_INPUT%u_SIZE:\t\t0x%X\n", i, v[0]);
+		seq_printf(sf, "CU_INPUT%u_OFFSET:\t0x%X\n", i, v[1]);
+		seq_printf(sf, "CU_INPUT%u_CONTROL:\t0x%X\n", i, v[2]);
+	}
+
+	get_values_from_reg(c->reg, 0x130, 2, v);
+	seq_printf(sf, "CU_USER_LOW:\t\t0x%X\n", v[0]);
+	seq_printf(sf, "CU_USER_HIGH:\t\t0x%X\n", v[1]);
+}
+
+static struct komeda_component_funcs d71_compiz_funcs = {
+	.update		= d71_compiz_update,
+	.disable	= d71_component_disable,
+	.dump_register	= d71_compiz_dump,
+};
+
+static int d71_compiz_init(struct d71_dev *d71,
+			   struct block_header *blk, u32 __iomem *reg)
+{
+	struct komeda_component *c;
+	struct komeda_compiz *compiz;
+	u32 pipe_id, comp_id;
+
+	get_resources_id(blk->block_info, &pipe_id, &comp_id);
+
+	c = komeda_component_add(&d71->pipes[pipe_id]->base, sizeof(*compiz),
+				 comp_id,
+				 BLOCK_INFO_INPUT_ID(blk->block_info),
+				 &d71_compiz_funcs,
+				 CU_NUM_INPUT_IDS, get_valid_inputs(blk),
+				 CU_NUM_OUTPUT_IDS, reg,
+				 "CU%d", pipe_id);
+	if (IS_ERR(c))
+		return PTR_ERR(c);
+
+	compiz = to_compiz(c);
+
+	set_range(&compiz->hsize, D71_MIN_LINE_SIZE, d71->max_line_size);
+	set_range(&compiz->vsize, D71_MIN_VERTICAL_SIZE, d71->max_vsize);
+
+	return 0;
+}
+
+static void d71_improc_update(struct komeda_component *c,
+			      struct komeda_component_state *state)
+{
+	struct komeda_improc_state *st = to_improc_st(state);
+	u32 __iomem *reg = c->reg;
+	u32 index, input_hw_id;
+
+	for_each_changed_input(state, index) {
+		input_hw_id = state->active_inputs & BIT(index) ?
+			      to_d71_input_id(&state->inputs[index]) : 0;
+		malidp_write32(reg, BLK_INPUT_ID0 + index * 4, input_hw_id);
+	}
+
+	malidp_write32(reg, BLK_SIZE, HV_SIZE(st->hsize, st->vsize));
+}
+
+static void d71_improc_dump(struct komeda_component *c, struct seq_file *sf)
+{
+	u32 v[12], i;
+
+	dump_block_header(sf, c->reg);
+
+	get_values_from_reg(c->reg, 0x80, 2, v);
+	seq_printf(sf, "IPS_INPUT_ID0:\t\t0x%X\n", v[0]);
+	seq_printf(sf, "IPS_INPUT_ID1:\t\t0x%X\n", v[1]);
+
+	get_values_from_reg(c->reg, 0xC0, 1, v);
+	seq_printf(sf, "IPS_INFO:\t\t0x%X\n", v[0]);
+
+	get_values_from_reg(c->reg, 0xD0, 3, v);
+	seq_printf(sf, "IPS_CONTROL:\t\t0x%X\n", v[0]);
+	seq_printf(sf, "IPS_SIZE:\t\t0x%X\n", v[1]);
+	seq_printf(sf, "IPS_DEPTH:\t\t0x%X\n", v[2]);
+
+	get_values_from_reg(c->reg, 0x130, 12, v);
+	for (i = 0; i < 12; i++)
+		seq_printf(sf, "IPS_RGB_RGB_COEFF%u:\t0x%X\n", i, v[i]);
+
+	get_values_from_reg(c->reg, 0x170, 12, v);
+	for (i = 0; i < 12; i++)
+		seq_printf(sf, "IPS_RGB_YUV_COEFF%u:\t0x%X\n", i, v[i]);
+}
+
+static struct komeda_component_funcs d71_improc_funcs = {
+	.update		= d71_improc_update,
+	.disable	= d71_component_disable,
+	.dump_register	= d71_improc_dump,
+};
+
+static int d71_improc_init(struct d71_dev *d71,
+			   struct block_header *blk, u32 __iomem *reg)
+{
+	struct komeda_component *c;
+	struct komeda_improc *improc;
+	u32 pipe_id, comp_id, value;
+
+	get_resources_id(blk->block_info, &pipe_id, &comp_id);
+
+	c = komeda_component_add(&d71->pipes[pipe_id]->base, sizeof(*improc),
+				 comp_id,
+				 BLOCK_INFO_INPUT_ID(blk->block_info),
+				 &d71_improc_funcs, IPS_NUM_INPUT_IDS,
+				 get_valid_inputs(blk),
+				 IPS_NUM_OUTPUT_IDS, reg, "DOU%d_IPS", pipe_id);
+	if (IS_ERR(c)) {
+		DRM_ERROR("Failed to add improc component\n");
+		return PTR_ERR(c);
+	}
+
+	improc = to_improc(c);
+	improc->supported_color_depths = BIT(8) | BIT(10);
+	improc->supported_color_formats = DRM_COLOR_FORMAT_RGB444 |
+					  DRM_COLOR_FORMAT_YCRCB444 |
+					  DRM_COLOR_FORMAT_YCRCB422;
+	value = malidp_read32(reg, BLK_INFO);
+	if (value & IPS_INFO_CHD420)
+		improc->supported_color_formats |= DRM_COLOR_FORMAT_YCRCB420;
+
+	improc->supports_csc = true;
+	improc->supports_gamma = true;
+
+	return 0;
+}
+
+static void d71_timing_ctrlr_disable(struct komeda_component *c)
+{
+	malidp_write32_mask(c->reg, BLK_CONTROL, BS_CTRL_EN, 0);
+}
+
+static void d71_timing_ctrlr_update(struct komeda_component *c,
+				    struct komeda_component_state *state)
+{
+	struct drm_crtc_state *crtc_st = state->crtc->state;
+	u32 __iomem *reg = c->reg;
+	struct videomode vm;
+	u32 value;
+
+	drm_display_mode_to_videomode(&crtc_st->adjusted_mode, &vm);
+
+	malidp_write32(reg, BS_ACTIVESIZE, HV_SIZE(vm.hactive, vm.vactive));
+	malidp_write32(reg, BS_HINTERVALS, BS_H_INTVALS(vm.hfront_porch,
+							vm.hback_porch));
+	malidp_write32(reg, BS_VINTERVALS, BS_V_INTVALS(vm.vfront_porch,
+							vm.vback_porch));
+
+	value = BS_SYNC_VSW(vm.vsync_len) | BS_SYNC_HSW(vm.hsync_len);
+	value |= vm.flags & DISPLAY_FLAGS_VSYNC_HIGH ? BS_SYNC_VSP : 0;
+	value |= vm.flags & DISPLAY_FLAGS_HSYNC_HIGH ? BS_SYNC_HSP : 0;
+	malidp_write32(reg, BS_SYNC, value);
+
+	malidp_write32(reg, BS_PROG_LINE, D71_DEFAULT_PREPRETCH_LINE - 1);
+	malidp_write32(reg, BS_PREFETCH_LINE, D71_DEFAULT_PREPRETCH_LINE);
+
+	/* configure bs control register */
+	value = BS_CTRL_EN | BS_CTRL_VM;
+
+	malidp_write32(reg, BLK_CONTROL, value);
+}
+
+static void d71_timing_ctrlr_dump(struct komeda_component *c,
+				  struct seq_file *sf)
+{
+	u32 v[8], i;
+
+	dump_block_header(sf, c->reg);
+
+	get_values_from_reg(c->reg, 0xC0, 1, v);
+	seq_printf(sf, "BS_INFO:\t\t0x%X\n", v[0]);
+
+	get_values_from_reg(c->reg, 0xD0, 8, v);
+	seq_printf(sf, "BS_CONTROL:\t\t0x%X\n", v[0]);
+	seq_printf(sf, "BS_PROG_LINE:\t\t0x%X\n", v[1]);
+	seq_printf(sf, "BS_PREFETCH_LINE:\t0x%X\n", v[2]);
+	seq_printf(sf, "BS_BG_COLOR:\t\t0x%X\n", v[3]);
+	seq_printf(sf, "BS_ACTIVESIZE:\t\t0x%X\n", v[4]);
+	seq_printf(sf, "BS_HINTERVALS:\t\t0x%X\n", v[5]);
+	seq_printf(sf, "BS_VINTERVALS:\t\t0x%X\n", v[6]);
+	seq_printf(sf, "BS_SYNC:\t\t0x%X\n", v[7]);
+
+	get_values_from_reg(c->reg, 0x100, 3, v);
+	seq_printf(sf, "BS_DRIFT_TO:\t\t0x%X\n", v[0]);
+	seq_printf(sf, "BS_FRAME_TO:\t\t0x%X\n", v[1]);
+	seq_printf(sf, "BS_TE_TO:\t\t0x%X\n", v[2]);
+
+	get_values_from_reg(c->reg, 0x110, 3, v);
+	for (i = 0; i < 3; i++)
+		seq_printf(sf, "BS_T%u_INTERVAL:\t\t0x%X\n", i, v[i]);
+
+	get_values_from_reg(c->reg, 0x120, 5, v);
+	for (i = 0; i < 2; i++) {
+		seq_printf(sf, "BS_CRC%u_LOW:\t\t0x%X\n", i, v[i << 1]);
+		seq_printf(sf, "BS_CRC%u_HIGH:\t\t0x%X\n", i, v[(i << 1) + 1]);
+	}
+	seq_printf(sf, "BS_USER:\t\t0x%X\n", v[4]);
+}
+
+static struct komeda_component_funcs d71_timing_ctrlr_funcs = {
+	.update		= d71_timing_ctrlr_update,
+	.disable	= d71_timing_ctrlr_disable,
+	.dump_register	= d71_timing_ctrlr_dump,
+};
+
+static int d71_timing_ctrlr_init(struct d71_dev *d71,
+				 struct block_header *blk, u32 __iomem *reg)
+{
+	struct komeda_component *c;
+	struct komeda_timing_ctrlr *ctrlr;
+	u32 pipe_id, comp_id;
+
+	get_resources_id(blk->block_info, &pipe_id, &comp_id);
+
+	c = komeda_component_add(&d71->pipes[pipe_id]->base, sizeof(*ctrlr),
+				 KOMEDA_COMPONENT_TIMING_CTRLR,
+				 BLOCK_INFO_INPUT_ID(blk->block_info),
+				 &d71_timing_ctrlr_funcs,
+				 1, BIT(KOMEDA_COMPONENT_IPS0 + pipe_id),
+				 BS_NUM_OUTPUT_IDS, reg, "DOU%d_BS", pipe_id);
+	if (IS_ERR(c)) {
+		DRM_ERROR("Failed to add display_ctrl component\n");
+		return PTR_ERR(c);
+	}
+
+	ctrlr = to_ctrlr(c);
+
+	ctrlr->supports_dual_link = true;
+
+	return 0;
+}
+
+int d71_probe_block(struct d71_dev *d71,
+		    struct block_header *blk, u32 __iomem *reg)
+{
+	struct d71_pipeline *pipe;
+	int blk_id = BLOCK_INFO_BLK_ID(blk->block_info);
+
+	int err = 0;
+
+	switch (BLOCK_INFO_BLK_TYPE(blk->block_info)) {
+	case D71_BLK_TYPE_GCU:
+		break;
+
+	case D71_BLK_TYPE_LPU:
+		pipe = d71->pipes[blk_id];
+		pipe->lpu_addr = reg;
+		break;
+
+	case D71_BLK_TYPE_LPU_LAYER:
+		err = d71_layer_init(d71, blk, reg);
+		break;
+
+	case D71_BLK_TYPE_LPU_WB_LAYER:
+		err = d71_wb_layer_init(d71, blk, reg);
+		break;
+
+	case D71_BLK_TYPE_CU:
+		pipe = d71->pipes[blk_id];
+		pipe->cu_addr = reg;
+		err = d71_compiz_init(d71, blk, reg);
+		break;
+
+	case D71_BLK_TYPE_CU_SPLITTER:
+	case D71_BLK_TYPE_CU_SCALER:
+	case D71_BLK_TYPE_CU_MERGER:
+		break;
+
+	case D71_BLK_TYPE_DOU:
+		pipe = d71->pipes[blk_id];
+		pipe->dou_addr = reg;
+		break;
+
+	case D71_BLK_TYPE_DOU_IPS:
+		err = d71_improc_init(d71, blk, reg);
+		break;
+
+	case D71_BLK_TYPE_DOU_FT_COEFF:
+		pipe = d71->pipes[blk_id];
+		pipe->dou_ft_coeff_addr = reg;
+		break;
+
+	case D71_BLK_TYPE_DOU_BS:
+		err = d71_timing_ctrlr_init(d71, blk, reg);
+		break;
+
+	case D71_BLK_TYPE_GLB_LT_COEFF:
+		break;
+
+	case D71_BLK_TYPE_GLB_SCL_COEFF:
+		d71->glb_scl_coeff_addr[blk_id] = reg;
+		break;
+
+	default:
+		DRM_ERROR("Unknown block (block_info: 0x%x) is found\n",
+			  blk->block_info);
+		err = -EINVAL;
+		break;
+	}
+
+	return err;
+}
diff --git a/drivers/gpu/drm/arm/display/komeda/d71/d71_dev.c b/drivers/gpu/drm/arm/display/komeda/d71/d71_dev.c
index edbf9daa1545..34506ef7ad40 100644
--- a/drivers/gpu/drm/arm/display/komeda/d71/d71_dev.c
+++ b/drivers/gpu/drm/arm/display/komeda/d71/d71_dev.c
@@ -4,13 +4,425 @@
  * Author: James.Qian.Wang <james.qian.wang@arm.com>
  *
  */
+
+#include <drm/drm_print.h>
+#include "d71_dev.h"
 #include "malidp_io.h"
-#include "komeda_dev.h"
+
+static u64 get_lpu_event(struct d71_pipeline *d71_pipeline)
+{
+	u32 __iomem *reg = d71_pipeline->lpu_addr;
+	u32 status, raw_status;
+	u64 evts = 0ULL;
+
+	raw_status = malidp_read32(reg, BLK_IRQ_RAW_STATUS);
+	if (raw_status & LPU_IRQ_IBSY)
+		evts |= KOMEDA_EVENT_IBSY;
+	if (raw_status & LPU_IRQ_EOW)
+		evts |= KOMEDA_EVENT_EOW;
+
+	if (raw_status & (LPU_IRQ_ERR | LPU_IRQ_IBSY)) {
+		u32 restore = 0, tbu_status;
+		/* Check error of LPU status */
+		status = malidp_read32(reg, BLK_STATUS);
+		if (status & LPU_STATUS_AXIE) {
+			restore |= LPU_STATUS_AXIE;
+			evts |= KOMEDA_ERR_AXIE;
+		}
+		if (status & LPU_STATUS_ACE0) {
+			restore |= LPU_STATUS_ACE0;
+			evts |= KOMEDA_ERR_ACE0;
+		}
+		if (status & LPU_STATUS_ACE1) {
+			restore |= LPU_STATUS_ACE1;
+			evts |= KOMEDA_ERR_ACE1;
+		}
+		if (status & LPU_STATUS_ACE2) {
+			restore |= LPU_STATUS_ACE2;
+			evts |= KOMEDA_ERR_ACE2;
+		}
+		if (status & LPU_STATUS_ACE3) {
+			restore |= LPU_STATUS_ACE3;
+			evts |= KOMEDA_ERR_ACE3;
+		}
+		if (restore != 0)
+			malidp_write32_mask(reg, BLK_STATUS, restore, 0);
+
+		restore = 0;
+		/* Check errors of TBU status */
+		tbu_status = malidp_read32(reg, LPU_TBU_STATUS);
+		if (tbu_status & LPU_TBU_STATUS_TCF) {
+			restore |= LPU_TBU_STATUS_TCF;
+			evts |= KOMEDA_ERR_TCF;
+		}
+		if (tbu_status & LPU_TBU_STATUS_TTNG) {
+			restore |= LPU_TBU_STATUS_TTNG;
+			evts |= KOMEDA_ERR_TTNG;
+		}
+		if (tbu_status & LPU_TBU_STATUS_TITR) {
+			restore |= LPU_TBU_STATUS_TITR;
+			evts |= KOMEDA_ERR_TITR;
+		}
+		if (tbu_status & LPU_TBU_STATUS_TEMR) {
+			restore |= LPU_TBU_STATUS_TEMR;
+			evts |= KOMEDA_ERR_TEMR;
+		}
+		if (tbu_status & LPU_TBU_STATUS_TTF) {
+			restore |= LPU_TBU_STATUS_TTF;
+			evts |= KOMEDA_ERR_TTF;
+		}
+		if (restore != 0)
+			malidp_write32_mask(reg, LPU_TBU_STATUS, restore, 0);
+	}
+
+	malidp_write32(reg, BLK_IRQ_CLEAR, raw_status);
+	return evts;
+}
+
+static u64 get_cu_event(struct d71_pipeline *d71_pipeline)
+{
+	u32 __iomem *reg = d71_pipeline->cu_addr;
+	u32 status, raw_status;
+	u64 evts = 0ULL;
+
+	raw_status = malidp_read32(reg, BLK_IRQ_RAW_STATUS);
+	if (raw_status & CU_IRQ_OVR)
+		evts |= KOMEDA_EVENT_OVR;
+
+	if (raw_status & (CU_IRQ_ERR | CU_IRQ_OVR)) {
+		status = malidp_read32(reg, BLK_STATUS) & 0x7FFFFFFF;
+		if (status & CU_STATUS_CPE)
+			evts |= KOMEDA_ERR_CPE;
+		if (status & CU_STATUS_ZME)
+			evts |= KOMEDA_ERR_ZME;
+		if (status & CU_STATUS_CFGE)
+			evts |= KOMEDA_ERR_CFGE;
+		if (status)
+			malidp_write32_mask(reg, BLK_STATUS, status, 0);
+	}
+
+	malidp_write32(reg, BLK_IRQ_CLEAR, raw_status);
+
+	return evts;
+}
+
+static u64 get_dou_event(struct d71_pipeline *d71_pipeline)
+{
+	u32 __iomem *reg = d71_pipeline->dou_addr;
+	u32 status, raw_status;
+	u64 evts = 0ULL;
+
+	raw_status = malidp_read32(reg, BLK_IRQ_RAW_STATUS);
+	if (raw_status & DOU_IRQ_PL0)
+		evts |= KOMEDA_EVENT_VSYNC;
+	if (raw_status & DOU_IRQ_UND)
+		evts |= KOMEDA_EVENT_URUN;
+
+	if (raw_status & (DOU_IRQ_ERR | DOU_IRQ_UND)) {
+		u32 restore  = 0;
+
+		status = malidp_read32(reg, BLK_STATUS);
+		if (status & DOU_STATUS_DRIFTTO) {
+			restore |= DOU_STATUS_DRIFTTO;
+			evts |= KOMEDA_ERR_DRIFTTO;
+		}
+		if (status & DOU_STATUS_FRAMETO) {
+			restore |= DOU_STATUS_FRAMETO;
+			evts |= KOMEDA_ERR_FRAMETO;
+		}
+		if (status & DOU_STATUS_TETO) {
+			restore |= DOU_STATUS_TETO;
+			evts |= KOMEDA_ERR_TETO;
+		}
+		if (status & DOU_STATUS_CSCE) {
+			restore |= DOU_STATUS_CSCE;
+			evts |= KOMEDA_ERR_CSCE;
+		}
+
+		if (restore != 0)
+			malidp_write32_mask(reg, BLK_STATUS, restore, 0);
+	}
+
+	malidp_write32(reg, BLK_IRQ_CLEAR, raw_status);
+	return evts;
+}
+
+static u64 get_pipeline_event(struct d71_pipeline *d71_pipeline, u32 gcu_status)
+{
+	u32 evts = 0ULL;
+
+	if (gcu_status & (GLB_IRQ_STATUS_LPU0 | GLB_IRQ_STATUS_LPU1))
+		evts |= get_lpu_event(d71_pipeline);
+
+	if (gcu_status & (GLB_IRQ_STATUS_CU0 | GLB_IRQ_STATUS_CU1))
+		evts |= get_cu_event(d71_pipeline);
+
+	if (gcu_status & (GLB_IRQ_STATUS_DOU0 | GLB_IRQ_STATUS_DOU1))
+		evts |= get_dou_event(d71_pipeline);
+
+	return evts;
+}
+
+static irqreturn_t
+d71_irq_handler(struct komeda_dev *mdev, struct komeda_events *evts)
+{
+	struct d71_dev *d71 = mdev->chip_data;
+	u32 status, gcu_status, raw_status;
+
+	gcu_status = malidp_read32(d71->gcu_addr, GLB_IRQ_STATUS);
+
+	if (gcu_status & GLB_IRQ_STATUS_GCU) {
+		raw_status = malidp_read32(d71->gcu_addr, BLK_IRQ_RAW_STATUS);
+		if (raw_status & GCU_IRQ_CVAL0)
+			evts->pipes[0] |= KOMEDA_EVENT_FLIP;
+		if (raw_status & GCU_IRQ_CVAL1)
+			evts->pipes[1] |= KOMEDA_EVENT_FLIP;
+		if (raw_status & GCU_IRQ_ERR) {
+			status = malidp_read32(d71->gcu_addr, BLK_STATUS);
+			if (status & GCU_STATUS_MERR) {
+				evts->global |= KOMEDA_ERR_MERR;
+				malidp_write32_mask(d71->gcu_addr, BLK_STATUS,
+						    GCU_STATUS_MERR, 0);
+			}
+		}
+
+		malidp_write32(d71->gcu_addr, BLK_IRQ_CLEAR, raw_status);
+	}
+
+	if (gcu_status & GLB_IRQ_STATUS_PIPE0)
+		evts->pipes[0] |= get_pipeline_event(d71->pipes[0], gcu_status);
+
+	if (gcu_status & GLB_IRQ_STATUS_PIPE1)
+		evts->pipes[1] |= get_pipeline_event(d71->pipes[1], gcu_status);
+
+	return gcu_status ? IRQ_HANDLED : IRQ_NONE;
+}
+
+#define ENABLED_GCU_IRQS	(GCU_IRQ_CVAL0 | GCU_IRQ_CVAL1 | \
+				 GCU_IRQ_MODE | GCU_IRQ_ERR)
+#define ENABLED_LPU_IRQS	(LPU_IRQ_IBSY | LPU_IRQ_ERR | LPU_IRQ_EOW)
+#define ENABLED_CU_IRQS		(CU_IRQ_OVR | CU_IRQ_ERR)
+#define ENABLED_DOU_IRQS	(DOU_IRQ_UND | DOU_IRQ_ERR)
+
+static int d71_enable_irq(struct komeda_dev *mdev)
+{
+	struct d71_dev *d71 = mdev->chip_data;
+	struct d71_pipeline *pipe;
+	u32 i;
+
+	malidp_write32_mask(d71->gcu_addr, BLK_IRQ_MASK,
+			    ENABLED_GCU_IRQS, ENABLED_GCU_IRQS);
+	for (i = 0; i < d71->num_pipelines; i++) {
+		pipe = d71->pipes[i];
+		malidp_write32_mask(pipe->cu_addr,  BLK_IRQ_MASK,
+				    ENABLED_CU_IRQS, ENABLED_CU_IRQS);
+		malidp_write32_mask(pipe->lpu_addr, BLK_IRQ_MASK,
+				    ENABLED_LPU_IRQS, ENABLED_LPU_IRQS);
+		malidp_write32_mask(pipe->dou_addr, BLK_IRQ_MASK,
+				    ENABLED_DOU_IRQS, ENABLED_DOU_IRQS);
+	}
+	return 0;
+}
+
+static int d71_disable_irq(struct komeda_dev *mdev)
+{
+	struct d71_dev *d71 = mdev->chip_data;
+	struct d71_pipeline *pipe;
+	u32 i;
+
+	malidp_write32_mask(d71->gcu_addr, BLK_IRQ_MASK, ENABLED_GCU_IRQS, 0);
+	for (i = 0; i < d71->num_pipelines; i++) {
+		pipe = d71->pipes[i];
+		malidp_write32_mask(pipe->cu_addr,  BLK_IRQ_MASK,
+				    ENABLED_CU_IRQS, 0);
+		malidp_write32_mask(pipe->lpu_addr, BLK_IRQ_MASK,
+				    ENABLED_LPU_IRQS, 0);
+		malidp_write32_mask(pipe->dou_addr, BLK_IRQ_MASK,
+				    ENABLED_DOU_IRQS, 0);
+	}
+	return 0;
+}
+
+static void d71_on_off_vblank(struct komeda_dev *mdev, int master_pipe, bool on)
+{
+	struct d71_dev *d71 = mdev->chip_data;
+	struct d71_pipeline *pipe = d71->pipes[master_pipe];
+
+	malidp_write32_mask(pipe->dou_addr, BLK_IRQ_MASK,
+			    DOU_IRQ_PL0, on ? DOU_IRQ_PL0 : 0);
+}
+
+static int to_d71_opmode(int core_mode)
+{
+	switch (core_mode) {
+	case KOMEDA_MODE_DISP0:
+		return DO0_ACTIVE_MODE;
+	case KOMEDA_MODE_DISP1:
+		return DO1_ACTIVE_MODE;
+	case KOMEDA_MODE_DUAL_DISP:
+		return DO01_ACTIVE_MODE;
+	case KOMEDA_MODE_INACTIVE:
+		return INACTIVE_MODE;
+	default:
+		WARN(1, "Unknown operation mode");
+		return INACTIVE_MODE;
+	}
+}
+
+static int d71_change_opmode(struct komeda_dev *mdev, int new_mode)
+{
+	struct d71_dev *d71 = mdev->chip_data;
+	u32 opmode = to_d71_opmode(new_mode);
+	int ret;
+
+	malidp_write32_mask(d71->gcu_addr, BLK_CONTROL, 0x7, opmode);
+
+	ret = dp_wait_cond(((malidp_read32(d71->gcu_addr, BLK_CONTROL) & 0x7) == opmode),
+			   100, 1000, 10000);
+
+	return ret > 0 ? 0 : -ETIMEDOUT;
+}
+
+static void d71_flush(struct komeda_dev *mdev,
+		      int master_pipe, u32 active_pipes)
+{
+	struct d71_dev *d71 = mdev->chip_data;
+	u32 reg_offset = (master_pipe == 0) ?
+			 GCU_CONFIG_VALID0 : GCU_CONFIG_VALID1;
+
+	malidp_write32(d71->gcu_addr, reg_offset, GCU_CONFIG_CVAL);
+}
+
+static int d71_reset(struct d71_dev *d71)
+{
+	u32 __iomem *gcu = d71->gcu_addr;
+	int ret;
+
+	malidp_write32_mask(gcu, BLK_CONTROL,
+			    GCU_CONTROL_SRST, GCU_CONTROL_SRST);
+
+	ret = dp_wait_cond(!(malidp_read32(gcu, BLK_CONTROL) & GCU_CONTROL_SRST),
+			   100, 1000, 10000);
+
+	return ret > 0 ? 0 : -ETIMEDOUT;
+}
+
+void d71_read_block_header(u32 __iomem *reg, struct block_header *blk)
+{
+	int i;
+
+	blk->block_info = malidp_read32(reg, BLK_BLOCK_INFO);
+	if (BLOCK_INFO_BLK_TYPE(blk->block_info) == D71_BLK_TYPE_RESERVED)
+		return;
+
+	blk->pipeline_info = malidp_read32(reg, BLK_PIPELINE_INFO);
+
+	/* get valid input and output ids */
+	for (i = 0; i < PIPELINE_INFO_N_VALID_INPUTS(blk->pipeline_info); i++)
+		blk->input_ids[i] = malidp_read32(reg + i, BLK_VALID_INPUT_ID0);
+	for (i = 0; i < PIPELINE_INFO_N_OUTPUTS(blk->pipeline_info); i++)
+		blk->output_ids[i] = malidp_read32(reg + i, BLK_OUTPUT_ID0);
+}
+
+static void d71_cleanup(struct komeda_dev *mdev)
+{
+	struct d71_dev *d71 = mdev->chip_data;
+
+	if (!d71)
+		return;
+
+	devm_kfree(mdev->dev, d71);
+	mdev->chip_data = NULL;
+}
 
 static int d71_enum_resources(struct komeda_dev *mdev)
 {
-	/* TODO add enum resources */
-	return -1;
+	struct d71_dev *d71;
+	struct komeda_pipeline *pipe;
+	struct block_header blk;
+	u32 __iomem *blk_base;
+	u32 i, value, offset;
+	int err;
+
+	d71 = devm_kzalloc(mdev->dev, sizeof(*d71), GFP_KERNEL);
+	if (!d71)
+		return -ENOMEM;
+
+	mdev->chip_data = d71;
+	d71->mdev = mdev;
+	d71->gcu_addr = mdev->reg_base;
+	d71->periph_addr = mdev->reg_base + (D71_BLOCK_OFFSET_PERIPH >> 2);
+
+	err = d71_reset(d71);
+	if (err) {
+		DRM_ERROR("Fail to reset d71 device.\n");
+		goto err_cleanup;
+	}
+
+	/* probe GCU */
+	value = malidp_read32(d71->gcu_addr, GLB_CORE_INFO);
+	d71->num_blocks = value & 0xFF;
+	d71->num_pipelines = (value >> 8) & 0x7;
+
+	if (d71->num_pipelines > D71_MAX_PIPELINE) {
+		DRM_ERROR("d71 supports %d pipelines, but got: %d.\n",
+			  D71_MAX_PIPELINE, d71->num_pipelines);
+		err = -EINVAL;
+		goto err_cleanup;
+	}
+
+	/* probe PERIPH */
+	value = malidp_read32(d71->periph_addr, BLK_BLOCK_INFO);
+	if (BLOCK_INFO_BLK_TYPE(value) != D71_BLK_TYPE_PERIPH) {
+		DRM_ERROR("access blk periph but got blk: %d.\n",
+			  BLOCK_INFO_BLK_TYPE(value));
+		err = -EINVAL;
+		goto err_cleanup;
+	}
+
+	value = malidp_read32(d71->periph_addr, PERIPH_CONFIGURATION_ID);
+
+	d71->max_line_size	= value & PERIPH_MAX_LINE_SIZE ? 4096 : 2048;
+	d71->max_vsize		= 4096;
+	d71->num_rich_layers	= value & PERIPH_NUM_RICH_LAYERS ? 2 : 1;
+	d71->supports_dual_link	= value & PERIPH_SPLIT_EN ? true : false;
+	d71->integrates_tbu	= value & PERIPH_TBU_EN ? true : false;
+
+	for (i = 0; i < d71->num_pipelines; i++) {
+		pipe = komeda_pipeline_add(mdev, sizeof(struct d71_pipeline),
+					   NULL);
+		if (IS_ERR(pipe)) {
+			err = PTR_ERR(pipe);
+			goto err_cleanup;
+		}
+		d71->pipes[i] = to_d71_pipeline(pipe);
+	}
+
+	/* loop the register blks and probe */
+	i = 2; /* exclude GCU and PERIPH */
+	offset = D71_BLOCK_SIZE; /* skip GCU */
+	while (i < d71->num_blocks) {
+		blk_base = mdev->reg_base + (offset >> 2);
+
+		d71_read_block_header(blk_base, &blk);
+		if (BLOCK_INFO_BLK_TYPE(blk.block_info) != D71_BLK_TYPE_RESERVED) {
+			err = d71_probe_block(d71, &blk, blk_base);
+			if (err)
+				goto err_cleanup;
+			i++;
+		}
+
+		offset += D71_BLOCK_SIZE;
+	}
+
+	DRM_DEBUG("total %d (out of %d) blocks are found.\n",
+		  i, d71->num_blocks);
+
+	return 0;
+
+err_cleanup:
+	d71_cleanup(mdev);
+	return err;
 }
 
 #define __HW_ID(__group, __format) \
@@ -93,19 +505,22 @@ static void d71_init_fmt_tbl(struct komeda_dev *mdev)
 static struct komeda_dev_funcs d71_chip_funcs = {
 	.init_format_table = d71_init_fmt_tbl,
 	.enum_resources	= d71_enum_resources,
-	.cleanup	= NULL,
+	.cleanup	= d71_cleanup,
+	.irq_handler	= d71_irq_handler,
+	.enable_irq	= d71_enable_irq,
+	.disable_irq	= d71_disable_irq,
+	.on_off_vblank	= d71_on_off_vblank,
+	.change_opmode	= d71_change_opmode,
+	.flush		= d71_flush,
 };
 
-#define GLB_ARCH_ID		0x000
-#define GLB_CORE_ID		0x004
-#define GLB_CORE_INFO		0x008
-
 struct komeda_dev_funcs *
 d71_identify(u32 __iomem *reg_base, struct komeda_chip_info *chip)
 {
 	chip->arch_id	= malidp_read32(reg_base, GLB_ARCH_ID);
 	chip->core_id	= malidp_read32(reg_base, GLB_CORE_ID);
 	chip->core_info	= malidp_read32(reg_base, GLB_CORE_INFO);
+	chip->bus_width	= D71_BUS_WIDTH_16_BYTES;
 
 	return &d71_chip_funcs;
 }
diff --git a/drivers/gpu/drm/arm/display/komeda/d71/d71_dev.h b/drivers/gpu/drm/arm/display/komeda/d71/d71_dev.h
new file mode 100644
index 000000000000..7465c57d9774
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/komeda/d71/d71_dev.h
@@ -0,0 +1,50 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+#ifndef _D71_DEV_H_
+#define _D71_DEV_H_
+
+#include "komeda_dev.h"
+#include "komeda_pipeline.h"
+#include "d71_regs.h"
+
+struct d71_pipeline {
+	struct komeda_pipeline base;
+
+	/* d71 private pipeline blocks */
+	u32 __iomem	*lpu_addr;
+	u32 __iomem	*cu_addr;
+	u32 __iomem	*dou_addr;
+	u32 __iomem	*dou_ft_coeff_addr; /* forward transform coeffs table */
+};
+
+struct d71_dev {
+	struct komeda_dev *mdev;
+
+	int	num_blocks;
+	int	num_pipelines;
+	int	num_rich_layers;
+	u32	max_line_size;
+	u32	max_vsize;
+	u32	supports_dual_link : 1;
+	u32	integrates_tbu : 1;
+
+	/* global register blocks */
+	u32 __iomem	*gcu_addr;
+	/* scaling coeffs table */
+	u32 __iomem	*glb_scl_coeff_addr[D71_MAX_GLB_SCL_COEFF];
+	u32 __iomem	*periph_addr;
+
+	struct d71_pipeline *pipes[D71_MAX_PIPELINE];
+};
+
+#define to_d71_pipeline(x)	container_of(x, struct d71_pipeline, base)
+
+int d71_probe_block(struct d71_dev *d71,
+		    struct block_header *blk, u32 __iomem *reg);
+void d71_read_block_header(u32 __iomem *reg, struct block_header *blk);
+
+#endif /* !_D71_DEV_H_ */
diff --git a/drivers/gpu/drm/arm/display/komeda/d71/d71_regs.h b/drivers/gpu/drm/arm/display/komeda/d71/d71_regs.h
new file mode 100644
index 000000000000..2d5e6d00b42c
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/komeda/d71/d71_regs.h
@@ -0,0 +1,530 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+#ifndef _D71_REG_H_
+#define _D71_REG_H_
+
+/* Common block registers offset */
+#define BLK_BLOCK_INFO		0x000
+#define BLK_PIPELINE_INFO	0x004
+#define BLK_VALID_INPUT_ID0	0x020
+#define BLK_OUTPUT_ID0		0x060
+#define BLK_INPUT_ID0		0x080
+#define BLK_IRQ_RAW_STATUS	0x0A0
+#define BLK_IRQ_CLEAR		0x0A4
+#define BLK_IRQ_MASK		0x0A8
+#define BLK_IRQ_STATUS		0x0AC
+#define BLK_STATUS		0x0B0
+#define BLK_INFO		0x0C0
+#define BLK_CONTROL		0x0D0
+#define BLK_SIZE		0x0D4
+#define BLK_IN_SIZE		0x0E0
+
+#define BLK_P0_PTR_LOW		0x100
+#define BLK_P0_PTR_HIGH		0x104
+#define BLK_P0_STRIDE		0x108
+#define BLK_P1_PTR_LOW		0x110
+#define BLK_P1_PTR_HIGH		0x114
+#define BLK_P1_STRIDE		0x118
+#define BLK_P2_PTR_LOW		0x120
+#define BLK_P2_PTR_HIGH		0x124
+
+#define BLOCK_INFO_N_SUBBLKS(x)	((x) & 0x000F)
+#define BLOCK_INFO_BLK_ID(x)	(((x) & 0x00F0) >> 4)
+#define BLOCK_INFO_BLK_TYPE(x)	(((x) & 0xFF00) >> 8)
+#define BLOCK_INFO_INPUT_ID(x)	((x) & 0xFFF0)
+#define BLOCK_INFO_TYPE_ID(x)	(((x) & 0x0FF0) >> 4)
+
+#define PIPELINE_INFO_N_OUTPUTS(x)	((x) & 0x000F)
+#define PIPELINE_INFO_N_VALID_INPUTS(x)	(((x) & 0x0F00) >> 8)
+
+/* Common block control register bits */
+#define BLK_CTRL_EN		BIT(0)
+/* Common size macro */
+#define HV_SIZE(h, v)		(((h) & 0x1FFF) + (((v) & 0x1FFF) << 16))
+#define HV_OFFSET(h, v)		(((h) & 0xFFF) + (((v) & 0xFFF) << 16))
+#define HV_CROP(h, v)		(((h) & 0xFFF) + (((v) & 0xFFF) << 16))
+
+/* AD_CONTROL register */
+#define AD_CONTROL		0x160
+
+/* AD_CONTROL register bits */
+#define AD_AEN			BIT(0)
+#define AD_YT			BIT(1)
+#define AD_BS			BIT(2)
+#define AD_WB			BIT(3)
+#define AD_TH			BIT(4)
+
+/* Global Control Unit */
+#define GLB_ARCH_ID		0x000
+#define GLB_CORE_ID		0x004
+#define GLB_CORE_INFO		0x008
+#define GLB_IRQ_STATUS		0x010
+
+#define GCU_CONFIG_VALID0	0x0D4
+#define GCU_CONFIG_VALID1	0x0D8
+
+/* GCU_CONTROL_BITS */
+#define GCU_CONTROL_MODE(x)	((x) & 0x7)
+#define GCU_CONTROL_SRST	BIT(16)
+
+/* GCU opmode */
+#define INACTIVE_MODE		0
+#define TBU_CONNECT_MODE	1
+#define TBU_DISCONNECT_MODE	2
+#define DO0_ACTIVE_MODE		3
+#define DO1_ACTIVE_MODE		4
+#define DO01_ACTIVE_MODE	5
+
+/* GLB_IRQ_STATUS bits */
+#define GLB_IRQ_STATUS_GCU	BIT(0)
+#define GLB_IRQ_STATUS_LPU0	BIT(8)
+#define GLB_IRQ_STATUS_LPU1	BIT(9)
+#define GLB_IRQ_STATUS_ATU0	BIT(10)
+#define GLB_IRQ_STATUS_ATU1	BIT(11)
+#define GLB_IRQ_STATUS_ATU2	BIT(12)
+#define GLB_IRQ_STATUS_ATU3	BIT(13)
+#define GLB_IRQ_STATUS_CU0	BIT(16)
+#define GLB_IRQ_STATUS_CU1	BIT(17)
+#define GLB_IRQ_STATUS_DOU0	BIT(24)
+#define GLB_IRQ_STATUS_DOU1	BIT(25)
+
+#define GLB_IRQ_STATUS_PIPE0	(GLB_IRQ_STATUS_LPU0 |\
+				 GLB_IRQ_STATUS_ATU0 |\
+				 GLB_IRQ_STATUS_ATU1 |\
+				 GLB_IRQ_STATUS_CU0 |\
+				 GLB_IRQ_STATUS_DOU0)
+
+#define GLB_IRQ_STATUS_PIPE1	(GLB_IRQ_STATUS_LPU1 |\
+				 GLB_IRQ_STATUS_ATU2 |\
+				 GLB_IRQ_STATUS_ATU3 |\
+				 GLB_IRQ_STATUS_CU1 |\
+				 GLB_IRQ_STATUS_DOU1)
+
+#define GLB_IRQ_STATUS_ATU	(GLB_IRQ_STATUS_ATU0 |\
+				 GLB_IRQ_STATUS_ATU1 |\
+				 GLB_IRQ_STATUS_ATU2 |\
+				 GLB_IRQ_STATUS_ATU3)
+
+/* GCU_IRQ_BITS */
+#define GCU_IRQ_CVAL0		BIT(0)
+#define GCU_IRQ_CVAL1		BIT(1)
+#define GCU_IRQ_MODE		BIT(4)
+#define GCU_IRQ_ERR		BIT(11)
+
+/* GCU_STATUS_BITS */
+#define GCU_STATUS_MODE(x)	((x) & 0x7)
+#define GCU_STATUS_MERR		BIT(4)
+#define GCU_STATUS_TCS0		BIT(8)
+#define GCU_STATUS_TCS1		BIT(9)
+#define GCU_STATUS_ACTIVE	BIT(31)
+
+/* GCU_CONFIG_VALIDx BITS */
+#define GCU_CONFIG_CVAL		BIT(0)
+
+/* PERIPHERAL registers */
+#define PERIPH_MAX_LINE_SIZE	BIT(0)
+#define PERIPH_NUM_RICH_LAYERS	BIT(4)
+#define PERIPH_SPLIT_EN		BIT(8)
+#define PERIPH_TBU_EN		BIT(12)
+#define PERIPH_AFBC_DMA_EN	BIT(16)
+#define PERIPH_CONFIGURATION_ID	0x1D4
+
+/* LPU register */
+#define LPU_TBU_STATUS		0x0B4
+#define LPU_RAXI_CONTROL	0x0D0
+#define LPU_WAXI_CONTROL	0x0D4
+#define LPU_TBU_CONTROL		0x0D8
+
+/* LPU_xAXI_CONTROL_BITS */
+#define TO_RAXI_AOUTSTDCAPB(x)	(x)
+#define TO_RAXI_BOUTSTDCAPB(x)	((x) << 8)
+#define TO_RAXI_BEN(x)		((x) << 15)
+#define TO_xAXI_BURSTLEN(x)	((x) << 16)
+#define TO_xAXI_AxQOS(x)	((x) << 24)
+#define TO_xAXI_ORD(x)		((x) << 31)
+#define TO_WAXI_OUTSTDCAPB(x)	(x)
+
+#define RAXI_AOUTSTDCAPB_MASK	0x7F
+#define RAXI_BOUTSTDCAPB_MASK	0x7F00
+#define RAXI_BEN_MASK		BIT(15)
+#define xAXI_BURSTLEN_MASK	0x3F0000
+#define xAXI_AxQOS_MASK		0xF000000
+#define xAXI_ORD_MASK		BIT(31)
+#define WAXI_OUTSTDCAPB_MASK	0x3F
+
+/* LPU_TBU_CONTROL BITS */
+#define TO_TBU_DOUTSTDCAPB(x)	(x)
+#define TBU_DOUTSTDCAPB_MASK	0x3F
+
+/* LPU_IRQ_BITS */
+#define LPU_IRQ_IBSY		BIT(10)
+#define LPU_IRQ_ERR		BIT(11)
+#define LPU_IRQ_EOW		BIT(12)
+#define LPU_IRQ_PL0		BIT(13)
+
+/* LPU_STATUS_BITS */
+#define LPU_STATUS_AXIED(x)	((x) & 0xF)
+#define LPU_STATUS_AXIE		BIT(4)
+#define LPU_STATUS_AXIRP	BIT(5)
+#define LPU_STATUS_AXIWP	BIT(6)
+#define LPU_STATUS_ACE0		BIT(16)
+#define LPU_STATUS_ACE1		BIT(17)
+#define LPU_STATUS_ACE2		BIT(18)
+#define LPU_STATUS_ACE3		BIT(19)
+#define LPU_STATUS_ACTIVE	BIT(31)
+
+#define AXIEID_MASK		0xF
+#define AXIE_MASK		LPU_STATUS_AXIE
+#define AXIRP_MASK		LPU_STATUS_AXIRP
+#define AXIWP_MASK		LPU_STATUS_AXIWP
+
+#define FROM_AXIEID(reg)	((reg) & AXIEID_MASK)
+#define TO_AXIE(x)		((x) << 4)
+#define FROM_AXIRP(reg)		(((reg) & AXIRP_MASK) >> 5)
+#define FROM_AXIWP(reg)		(((reg) & AXIWP_MASK) >> 6)
+
+/* LPU_TBU_STATUS_BITS */
+#define LPU_TBU_STATUS_TCF	BIT(1)
+#define LPU_TBU_STATUS_TTNG	BIT(2)
+#define LPU_TBU_STATUS_TITR	BIT(8)
+#define LPU_TBU_STATUS_TEMR	BIT(16)
+#define LPU_TBU_STATUS_TTF	BIT(31)
+
+/* LPU_TBU_CONTROL BITS */
+#define LPU_TBU_CTRL_TLBPEN	BIT(16)
+
+/* CROSSBAR CONTROL BITS */
+#define CBU_INPUT_CTRL_EN	BIT(0)
+#define CBU_NUM_INPUT_IDS	5
+#define CBU_NUM_OUTPUT_IDS	5
+
+/* CU register */
+#define CU_BG_COLOR		0x0DC
+#define CU_INPUT0_SIZE		0x0E0
+#define CU_INPUT0_OFFSET	0x0E4
+#define CU_INPUT0_CONTROL	0x0E8
+#define CU_INPUT1_SIZE		0x0F0
+#define CU_INPUT1_OFFSET	0x0F4
+#define CU_INPUT1_CONTROL	0x0F8
+#define CU_INPUT2_SIZE		0x100
+#define CU_INPUT2_OFFSET	0x104
+#define CU_INPUT2_CONTROL	0x108
+#define CU_INPUT3_SIZE		0x110
+#define CU_INPUT3_OFFSET	0x114
+#define CU_INPUT3_CONTROL	0x118
+#define CU_INPUT4_SIZE		0x120
+#define CU_INPUT4_OFFSET	0x124
+#define CU_INPUT4_CONTROL	0x128
+
+#define CU_PER_INPUT_REGS	4
+
+#define CU_NUM_INPUT_IDS	5
+#define CU_NUM_OUTPUT_IDS	1
+
+/* CU control register bits */
+#define CU_CTRL_COPROC		BIT(0)
+
+/* CU_IRQ_BITS */
+#define CU_IRQ_OVR		BIT(9)
+#define CU_IRQ_ERR		BIT(11)
+
+/* CU_STATUS_BITS */
+#define CU_STATUS_CPE		BIT(0)
+#define CU_STATUS_ZME		BIT(1)
+#define CU_STATUS_CFGE		BIT(2)
+#define CU_STATUS_ACTIVE	BIT(31)
+
+/* CU input control register bits */
+#define CU_INPUT_CTRL_EN	BIT(0)
+#define CU_INPUT_CTRL_PAD	BIT(1)
+#define CU_INPUT_CTRL_PMUL	BIT(2)
+#define CU_INPUT_CTRL_ALPHA(x)	(((x) & 0xFF) << 8)
+
+/* DOU register */
+
+/* DOU_IRQ_BITS */
+#define DOU_IRQ_UND		BIT(8)
+#define DOU_IRQ_ERR		BIT(11)
+#define DOU_IRQ_PL0		BIT(13)
+#define DOU_IRQ_PL1		BIT(14)
+
+/* DOU_STATUS_BITS */
+#define DOU_STATUS_DRIFTTO	BIT(0)
+#define DOU_STATUS_FRAMETO	BIT(1)
+#define DOU_STATUS_TETO		BIT(2)
+#define DOU_STATUS_CSCE		BIT(8)
+#define DOU_STATUS_ACTIVE	BIT(31)
+
+/* Layer registers */
+#define LAYER_INFO		0x0C0
+#define LAYER_R_CONTROL		0x0D4
+#define LAYER_FMT		0x0D8
+#define LAYER_LT_COEFFTAB	0x0DC
+#define LAYER_PALPHA		0x0E4
+
+#define LAYER_YUV_RGB_COEFF0	0x130
+
+#define LAYER_AD_H_CROP		0x164
+#define LAYER_AD_V_CROP		0x168
+
+#define LAYER_RGB_RGB_COEFF0	0x170
+
+/* L_CONTROL_BITS */
+#define L_EN			BIT(0)
+#define L_IT			BIT(4)
+#define L_R2R			BIT(5)
+#define L_FT			BIT(6)
+#define L_ROT(x)		(((x) & 3) << 8)
+#define L_HFLIP			BIT(10)
+#define L_VFLIP			BIT(11)
+#define L_TBU_EN		BIT(16)
+#define L_A_RCACHE(x)		(((x) & 0xF) << 28)
+#define L_ROT_R0		0
+#define L_ROT_R90		1
+#define L_ROT_R180		2
+#define L_ROT_R270		3
+
+/* LAYER_R_CONTROL BITS */
+#define LR_CHI422_BILINEAR	0
+#define LR_CHI422_REPLICATION	1
+#define LR_CHI420_JPEG		(0 << 2)
+#define LR_CHI420_MPEG		(1 << 2)
+
+#define L_ITSEL(x)		((x) & 0xFFF)
+#define L_FTSEL(x)		(((x) & 0xFFF) << 16)
+
+#define LAYER_PER_PLANE_REGS	4
+
+/* Layer_WR registers */
+#define LAYER_WR_PROG_LINE	0x0D4
+#define LAYER_WR_FORMAT		0x0D8
+
+/* Layer_WR control bits */
+#define LW_OFM			BIT(4)
+#define LW_LALPHA(x)		(((x) & 0xFF) << 8)
+#define LW_A_WCACHE(x)		(((x) & 0xF) << 28)
+#define LW_TBU_EN		BIT(16)
+
+#define AxCACHE_MASK		0xF0000000
+
+/* Layer AXI R/W cache setting */
+#define AxCACHE_B		BIT(0)	/* Bufferable */
+#define AxCACHE_M		BIT(1)	/* Modifiable */
+#define AxCACHE_RA		BIT(2)	/* Read-Allocate */
+#define AxCACHE_WA		BIT(3)	/* Write-Allocate */
+
+/* Layer info bits */
+#define L_INFO_RF		BIT(0)
+#define L_INFO_CM		BIT(1)
+#define L_INFO_ABUF_SIZE(x)	(((x) >> 4) & 0x7)
+
+/* Scaler registers */
+#define SC_COEFFTAB		0x0DC
+#define SC_OUT_SIZE		0x0E4
+#define SC_H_CROP		0x0E8
+#define SC_V_CROP		0x0EC
+#define SC_H_INIT_PH		0x0F0
+#define SC_H_DELTA_PH		0x0F4
+#define SC_V_INIT_PH		0x0F8
+#define SC_V_DELTA_PH		0x0FC
+#define SC_ENH_LIMITS		0x130
+#define SC_ENH_COEFF0		0x134
+
+#define SC_MAX_ENH_COEFF	9
+
+/* SC_CTRL_BITS */
+#define SC_CTRL_SCL		BIT(0)
+#define SC_CTRL_LS		BIT(1)
+#define SC_CTRL_AP		BIT(4)
+#define SC_CTRL_IENH		BIT(8)
+#define SC_CTRL_RGBSM		BIT(16)
+#define SC_CTRL_ASM		BIT(17)
+
+#define SC_VTSEL(vtal)		((vtal) << 16)
+
+#define SC_NUM_INPUTS_IDS	1
+#define SC_NUM_OUTPUTS_IDS	1
+
+#define MG_NUM_INPUTS_IDS	2
+#define MG_NUM_OUTPUTS_IDS	1
+
+/* Merger registers */
+#define MG_INPUT_ID0		BLK_INPUT_ID0
+#define MG_INPUT_ID1		(MG_INPUT_ID0 + 4)
+#define MG_SIZE			BLK_SIZE
+
+/* Splitter registers */
+#define SP_OVERLAP_SIZE		0xD8
+
+/* Backend registers */
+#define BS_INFO			0x0C0
+#define BS_PROG_LINE		0x0D4
+#define BS_PREFETCH_LINE	0x0D8
+#define BS_BG_COLOR		0x0DC
+#define BS_ACTIVESIZE		0x0E0
+#define BS_HINTERVALS		0x0E4
+#define BS_VINTERVALS		0x0E8
+#define BS_SYNC			0x0EC
+#define BS_DRIFT_TO		0x100
+#define BS_FRAME_TO		0x104
+#define BS_TE_TO		0x108
+#define BS_T0_INTERVAL		0x110
+#define BS_T1_INTERVAL		0x114
+#define BS_T2_INTERVAL		0x118
+#define BS_CRC0_LOW		0x120
+#define BS_CRC0_HIGH		0x124
+#define BS_CRC1_LOW		0x128
+#define BS_CRC1_HIGH		0x12C
+#define BS_USER			0x130
+
+/* BS control register bits */
+#define BS_CTRL_EN		BIT(0)
+#define BS_CTRL_VM		BIT(1)
+#define BS_CTRL_BM		BIT(2)
+#define BS_CTRL_HMASK		BIT(4)
+#define BS_CTRL_VD		BIT(5)
+#define BS_CTRL_TE		BIT(8)
+#define BS_CTRL_TS		BIT(9)
+#define BS_CTRL_TM		BIT(12)
+#define BS_CTRL_DL		BIT(16)
+#define BS_CTRL_SBS		BIT(17)
+#define BS_CTRL_CRC		BIT(18)
+#define BS_CTRL_PM		BIT(20)
+
+/* BS active size/intervals */
+#define BS_H_INTVALS(hfp, hbp)	(((hfp) & 0xFFF) + (((hbp) & 0x3FF) << 16))
+#define BS_V_INTVALS(vfp, vbp)  (((vfp) & 0x3FFF) + (((vbp) & 0xFF) << 16))
+
+/* BS_SYNC bits */
+#define BS_SYNC_HSW(x)		((x) & 0x3FF)
+#define BS_SYNC_HSP		BIT(12)
+#define BS_SYNC_VSW(x)		(((x) & 0xFF) << 16)
+#define BS_SYNC_VSP		BIT(28)
+
+#define BS_NUM_INPUT_IDS	0
+#define BS_NUM_OUTPUT_IDS	0
+
+/* Image process registers */
+#define IPS_DEPTH		0x0D8
+#define IPS_RGB_RGB_COEFF0	0x130
+#define IPS_RGB_YUV_COEFF0	0x170
+
+#define IPS_DEPTH_MARK		0xF
+
+/* IPS control register bits */
+#define IPS_CTRL_RGB		BIT(0)
+#define IPS_CTRL_FT		BIT(4)
+#define IPS_CTRL_YUV		BIT(8)
+#define IPS_CTRL_CHD422		BIT(9)
+#define IPS_CTRL_CHD420		BIT(10)
+#define IPS_CTRL_LPF		BIT(11)
+#define IPS_CTRL_DITH		BIT(12)
+#define IPS_CTRL_CLAMP		BIT(16)
+#define IPS_CTRL_SBS		BIT(17)
+
+/* IPS info register bits */
+#define IPS_INFO_CHD420		BIT(10)
+
+#define IPS_NUM_INPUT_IDS	2
+#define IPS_NUM_OUTPUT_IDS	1
+
+/* FT_COEFF block registers */
+#define FT_COEFF0		0x80
+#define GLB_IT_COEFF		0x80
+
+/* GLB_SC_COEFF registers */
+#define GLB_SC_COEFF_ADDR	0x0080
+#define GLB_SC_COEFF_DATA	0x0084
+#define GLB_LT_COEFF_DATA	0x0080
+
+#define GLB_SC_COEFF_MAX_NUM	1024
+#define GLB_LT_COEFF_NUM	65
+/* GLB_SC_ADDR */
+#define SC_COEFF_R_ADDR		BIT(18)
+#define SC_COEFF_G_ADDR		BIT(17)
+#define SC_COEFF_B_ADDR		BIT(16)
+
+#define SC_COEFF_DATA(x, y)	(((y) & 0xFFFF) | (((x) & 0xFFFF) << 16))
+
+enum d71_blk_type {
+	D71_BLK_TYPE_GCU		= 0x00,
+	D71_BLK_TYPE_LPU		= 0x01,
+	D71_BLK_TYPE_CU			= 0x02,
+	D71_BLK_TYPE_DOU		= 0x03,
+	D71_BLK_TYPE_AEU		= 0x04,
+	D71_BLK_TYPE_GLB_LT_COEFF	= 0x05,
+	D71_BLK_TYPE_GLB_SCL_COEFF	= 0x06, /* SH/SV scaler coeff */
+	D71_BLK_TYPE_GLB_SC_COEFF	= 0x07,
+	D71_BLK_TYPE_PERIPH		= 0x08,
+	D71_BLK_TYPE_LPU_TRUSTED	= 0x09,
+	D71_BLK_TYPE_AEU_TRUSTED	= 0x0A,
+	D71_BLK_TYPE_LPU_LAYER		= 0x10,
+	D71_BLK_TYPE_LPU_WB_LAYER	= 0x11,
+	D71_BLK_TYPE_CU_SPLITTER	= 0x20,
+	D71_BLK_TYPE_CU_SCALER		= 0x21,
+	D71_BLK_TYPE_CU_MERGER		= 0x22,
+	D71_BLK_TYPE_DOU_IPS		= 0x30,
+	D71_BLK_TYPE_DOU_BS		= 0x31,
+	D71_BLK_TYPE_DOU_FT_COEFF	= 0x32,
+	D71_BLK_TYPE_AEU_DS		= 0x40,
+	D71_BLK_TYPE_AEU_AES		= 0x41,
+	D71_BLK_TYPE_RESERVED		= 0xFF
+};
+
+/* Constant of components */
+#define D71_MAX_PIPELINE		2
+#define D71_PIPELINE_MAX_SCALERS	2
+#define D71_PIPELINE_MAX_LAYERS		4
+
+#define D71_MAX_GLB_IT_COEFF		3
+#define D71_MAX_GLB_SCL_COEFF		4
+
+#define D71_MAX_LAYERS_PER_LPU		4
+#define D71_BLOCK_MAX_INPUT		9
+#define D71_BLOCK_MAX_OUTPUT		5
+#define D71_MAX_SC_PER_CU		2
+
+#define D71_BLOCK_OFFSET_PERIPH		0xFE00
+#define D71_BLOCK_SIZE			0x0200
+
+#define D71_DEFAULT_PREPRETCH_LINE	5
+#define D71_BUS_WIDTH_16_BYTES		16
+
+#define D71_MIN_LINE_SIZE		64
+#define D71_MIN_VERTICAL_SIZE		64
+#define D71_SC_MIN_LIN_SIZE		4
+#define D71_SC_MIN_VERTICAL_SIZE	4
+#define D71_SC_MAX_LIN_SIZE		2048
+#define D71_SC_MAX_VERTICAL_SIZE	4096
+
+#define D71_SC_MAX_UPSCALING		64
+#define D71_SC_MAX_DOWNSCALING		6
+#define D71_SC_SPLIT_OVERLAP		8
+#define D71_SC_ENH_SPLIT_OVERLAP	1
+
+#define D71_MG_MIN_MERGED_SIZE		4
+#define D71_MG_MAX_MERGED_HSIZE		4032
+#define D71_MG_MAX_MERGED_VSIZE		4096
+
+#define D71_PALPHA_DEF_MAP		0xFFAA5500
+#define D71_LAYER_CONTROL_DEFAULT	0x30000000
+#define D71_WB_LAYER_CONTROL_DEFAULT	0x3000FF00
+#define D71_BS_CONTROL_DEFAULT		0x00000002
+
+struct block_header {
+	u32 block_info;
+	u32 pipeline_info;
+	u32 input_ids[D71_BLOCK_MAX_INPUT];
+	u32 output_ids[D71_BLOCK_MAX_OUTPUT];
+};
+
+static inline u32 get_block_type(struct block_header *blk)
+{
+	return BLOCK_INFO_BLK_TYPE(blk->block_info);
+}
+
+#endif /* !_D71_REG_H_ */
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_crtc.c b/drivers/gpu/drm/arm/display/komeda/komeda_crtc.c
index 3ca5718aa0c2..62fad59f5a6a 100644
--- a/drivers/gpu/drm/arm/display/komeda/komeda_crtc.c
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_crtc.c
@@ -18,10 +18,415 @@
 #include "komeda_dev.h"
 #include "komeda_kms.h"
 
-struct drm_crtc_helper_funcs komeda_crtc_helper_funcs = {
+/**
+ * komeda_crtc_atomic_check - build display output data flow
+ * @crtc: DRM crtc
+ * @state: the crtc state object
+ *
+ * crtc_atomic_check is the final check stage, so beside build a display data
+ * pipeline according to the crtc_state, but still needs to release or disable
+ * the unclaimed pipeline resources.
+ *
+ * RETURNS:
+ * Zero for success or -errno
+ */
+static int
+komeda_crtc_atomic_check(struct drm_crtc *crtc,
+			 struct drm_crtc_state *state)
+{
+	struct komeda_crtc *kcrtc = to_kcrtc(crtc);
+	struct komeda_crtc_state *kcrtc_st = to_kcrtc_st(state);
+	int err;
+
+	if (state->active) {
+		err = komeda_build_display_data_flow(kcrtc, kcrtc_st);
+		if (err)
+			return err;
+	}
+
+	/* release unclaimed pipeline resources */
+	err = komeda_release_unclaimed_resources(kcrtc->master, kcrtc_st);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+static u32 komeda_calc_mclk(struct komeda_crtc_state *kcrtc_st)
+{
+	unsigned long mclk = kcrtc_st->base.adjusted_mode.clock * 1000;
+
+	return mclk;
+}
+
+/* For active a crtc, mainly need two parts of preparation
+ * 1. adjust display operation mode.
+ * 2. enable needed clk
+ */
+static int
+komeda_crtc_prepare(struct komeda_crtc *kcrtc)
+{
+	struct komeda_dev *mdev = kcrtc->base.dev->dev_private;
+	struct komeda_pipeline *master = kcrtc->master;
+	struct komeda_crtc_state *kcrtc_st = to_kcrtc_st(kcrtc->base.state);
+	unsigned long pxlclk_rate = kcrtc_st->base.adjusted_mode.clock * 1000;
+	u32 new_mode;
+	int err;
+
+	mutex_lock(&mdev->lock);
+
+	new_mode = mdev->dpmode | BIT(master->id);
+	if (WARN_ON(new_mode == mdev->dpmode)) {
+		err = 0;
+		goto unlock;
+	}
+
+	err = mdev->funcs->change_opmode(mdev, new_mode);
+	if (err) {
+		DRM_ERROR("failed to change opmode: 0x%x -> 0x%x.\n,",
+			  mdev->dpmode, new_mode);
+		goto unlock;
+	}
+
+	mdev->dpmode = new_mode;
+	/* Only need to enable mclk on single display mode, but no need to
+	 * enable mclk it on dual display mode, since the dual mode always
+	 * switch from single display mode, the mclk already enabled, no need
+	 * to enable it again.
+	 */
+	if (new_mode != KOMEDA_MODE_DUAL_DISP) {
+		err = clk_set_rate(mdev->mclk, komeda_calc_mclk(kcrtc_st));
+		if (err)
+			DRM_ERROR("failed to set mclk.\n");
+		err = clk_prepare_enable(mdev->mclk);
+		if (err)
+			DRM_ERROR("failed to enable mclk.\n");
+	}
+
+	err = clk_prepare_enable(master->aclk);
+	if (err)
+		DRM_ERROR("failed to enable axi clk for pipe%d.\n", master->id);
+	err = clk_set_rate(master->pxlclk, pxlclk_rate);
+	if (err)
+		DRM_ERROR("failed to set pxlclk for pipe%d\n", master->id);
+	err = clk_prepare_enable(master->pxlclk);
+	if (err)
+		DRM_ERROR("failed to enable pxl clk for pipe%d.\n", master->id);
+
+unlock:
+	mutex_unlock(&mdev->lock);
+
+	return err;
+}
+
+static int
+komeda_crtc_unprepare(struct komeda_crtc *kcrtc)
+{
+	struct komeda_dev *mdev = kcrtc->base.dev->dev_private;
+	struct komeda_pipeline *master = kcrtc->master;
+	u32 new_mode;
+	int err;
+
+	mutex_lock(&mdev->lock);
+
+	new_mode = mdev->dpmode & (~BIT(master->id));
+
+	if (WARN_ON(new_mode == mdev->dpmode)) {
+		err = 0;
+		goto unlock;
+	}
+
+	err = mdev->funcs->change_opmode(mdev, new_mode);
+	if (err) {
+		DRM_ERROR("failed to change opmode: 0x%x -> 0x%x.\n,",
+			  mdev->dpmode, new_mode);
+		goto unlock;
+	}
+
+	mdev->dpmode = new_mode;
+
+	clk_disable_unprepare(master->pxlclk);
+	clk_disable_unprepare(master->aclk);
+	if (new_mode == KOMEDA_MODE_INACTIVE)
+		clk_disable_unprepare(mdev->mclk);
+
+unlock:
+	mutex_unlock(&mdev->lock);
+
+	return err;
+}
+
+void komeda_crtc_handle_event(struct komeda_crtc   *kcrtc,
+			      struct komeda_events *evts)
+{
+	struct drm_crtc *crtc = &kcrtc->base;
+	u32 events = evts->pipes[kcrtc->master->id];
+
+	if (events & KOMEDA_EVENT_VSYNC)
+		drm_crtc_handle_vblank(crtc);
+
+	/* will handle it together with the write back support */
+	if (events & KOMEDA_EVENT_EOW)
+		DRM_DEBUG("EOW.\n");
+
+	if (events & KOMEDA_EVENT_FLIP) {
+		unsigned long flags;
+		struct drm_pending_vblank_event *event;
+
+		spin_lock_irqsave(&crtc->dev->event_lock, flags);
+		if (kcrtc->disable_done) {
+			complete_all(kcrtc->disable_done);
+			kcrtc->disable_done = NULL;
+		} else if (crtc->state->event) {
+			event = crtc->state->event;
+			/*
+			 * Consume event before notifying drm core that flip
+			 * happened.
+			 */
+			crtc->state->event = NULL;
+			drm_crtc_send_vblank_event(crtc, event);
+		} else {
+			DRM_WARN("CRTC[%d]: FLIP happen but no pending commit.\n",
+				 drm_crtc_index(&kcrtc->base));
+		}
+		spin_unlock_irqrestore(&crtc->dev->event_lock, flags);
+	}
+}
+
+static void
+komeda_crtc_do_flush(struct drm_crtc *crtc,
+		     struct drm_crtc_state *old)
+{
+	struct komeda_crtc *kcrtc = to_kcrtc(crtc);
+	struct komeda_crtc_state *kcrtc_st = to_kcrtc_st(crtc->state);
+	struct komeda_dev *mdev = kcrtc->base.dev->dev_private;
+	struct komeda_pipeline *master = kcrtc->master;
+
+	DRM_DEBUG_ATOMIC("CRTC%d_FLUSH: active_pipes: 0x%x, affected: 0x%x.\n",
+			 drm_crtc_index(crtc),
+			 kcrtc_st->active_pipes, kcrtc_st->affected_pipes);
+
+	/* step 1: update the pipeline/component state to HW */
+	if (has_bit(master->id, kcrtc_st->affected_pipes))
+		komeda_pipeline_update(master, old->state);
+
+	/* step 2: notify the HW to kickoff the update */
+	mdev->funcs->flush(mdev, master->id, kcrtc_st->active_pipes);
+}
+
+static void
+komeda_crtc_atomic_enable(struct drm_crtc *crtc,
+			  struct drm_crtc_state *old)
+{
+	komeda_crtc_prepare(to_kcrtc(crtc));
+	drm_crtc_vblank_on(crtc);
+	komeda_crtc_do_flush(crtc, old);
+}
+
+static void
+komeda_crtc_atomic_disable(struct drm_crtc *crtc,
+			   struct drm_crtc_state *old)
+{
+	struct komeda_crtc *kcrtc = to_kcrtc(crtc);
+	struct komeda_crtc_state *old_st = to_kcrtc_st(old);
+	struct komeda_dev *mdev = crtc->dev->dev_private;
+	struct komeda_pipeline *master = kcrtc->master;
+	struct completion *disable_done = &crtc->state->commit->flip_done;
+	struct completion temp;
+	int timeout;
+
+	DRM_DEBUG_ATOMIC("CRTC%d_DISABLE: active_pipes: 0x%x, affected: 0x%x.\n",
+			 drm_crtc_index(crtc),
+			 old_st->active_pipes, old_st->affected_pipes);
+
+	if (has_bit(master->id, old_st->active_pipes))
+		komeda_pipeline_disable(master, old->state);
+
+	/* crtc_disable has two scenarios according to the state->active switch.
+	 * 1. active -> inactive
+	 *    this commit is a disable commit. and the commit will be finished
+	 *    or done after the disable operation. on this case we can directly
+	 *    use the crtc->state->event to tracking the HW disable operation.
+	 * 2. active -> active
+	 *    the crtc->commit is not for disable, but a modeset operation when
+	 *    crtc is active, such commit actually has been completed by 3
+	 *    DRM operations:
+	 *    crtc_disable, update_planes(crtc_flush), crtc_enable
+	 *    so on this case the crtc->commit is for the whole process.
+	 *    we can not use it for tracing the disable, we need a temporary
+	 *    flip_done for tracing the disable. and crtc->state->event for
+	 *    the crtc_enable operation.
+	 *    That's also the reason why skip modeset commit in
+	 *    komeda_crtc_atomic_flush()
+	 */
+	if (crtc->state->active) {
+		struct komeda_pipeline_state *pipe_st;
+		/* clear the old active_comps to zero */
+		pipe_st = komeda_pipeline_get_old_state(master, old->state);
+		pipe_st->active_comps = 0;
+
+		init_completion(&temp);
+		kcrtc->disable_done = &temp;
+		disable_done = &temp;
+	}
+
+	mdev->funcs->flush(mdev, master->id, 0);
+
+	/* wait the disable take affect.*/
+	timeout = wait_for_completion_timeout(disable_done, HZ);
+	if (timeout == 0) {
+		DRM_ERROR("disable pipeline%d timeout.\n", kcrtc->master->id);
+		if (crtc->state->active) {
+			unsigned long flags;
+
+			spin_lock_irqsave(&crtc->dev->event_lock, flags);
+			kcrtc->disable_done = NULL;
+			spin_unlock_irqrestore(&crtc->dev->event_lock, flags);
+		}
+	}
+
+	drm_crtc_vblank_off(crtc);
+	komeda_crtc_unprepare(kcrtc);
+}
+
+static void
+komeda_crtc_atomic_flush(struct drm_crtc *crtc,
+			 struct drm_crtc_state *old)
+{
+	/* commit with modeset will be handled in enable/disable */
+	if (drm_atomic_crtc_needs_modeset(crtc->state))
+		return;
+
+	komeda_crtc_do_flush(crtc, old);
+}
+
+static enum drm_mode_status
+komeda_crtc_mode_valid(struct drm_crtc *crtc, const struct drm_display_mode *m)
+{
+	struct komeda_dev *mdev = crtc->dev->dev_private;
+	struct komeda_crtc *kcrtc = to_kcrtc(crtc);
+	struct komeda_pipeline *master = kcrtc->master;
+	long mode_clk, pxlclk;
+
+	if (m->flags & DRM_MODE_FLAG_INTERLACE)
+		return MODE_NO_INTERLACE;
+
+	/* main clock/AXI clk must be faster than pxlclk*/
+	mode_clk = m->clock * 1000;
+	pxlclk = clk_round_rate(master->pxlclk, mode_clk);
+	if (pxlclk != mode_clk) {
+		DRM_DEBUG_ATOMIC("pxlclk doesn't support %ld Hz\n", mode_clk);
+
+		return MODE_NOCLOCK;
+	}
+
+	if (clk_round_rate(mdev->mclk, mode_clk) < pxlclk) {
+		DRM_DEBUG_ATOMIC("mclk can't satisfy the requirement of %s-clk: %ld.\n",
+				 m->name, pxlclk);
+
+		return MODE_CLOCK_HIGH;
+	}
+
+	if (clk_round_rate(master->aclk, mode_clk) < pxlclk) {
+		DRM_DEBUG_ATOMIC("aclk can't satisfy the requirement of %s-clk: %ld.\n",
+				 m->name, pxlclk);
+
+		return MODE_CLOCK_HIGH;
+	}
+
+	return MODE_OK;
+}
+
+static bool komeda_crtc_mode_fixup(struct drm_crtc *crtc,
+				   const struct drm_display_mode *m,
+				   struct drm_display_mode *adjusted_mode)
+{
+	struct komeda_crtc *kcrtc = to_kcrtc(crtc);
+	struct komeda_pipeline *master = kcrtc->master;
+	long mode_clk = m->clock * 1000;
+
+	adjusted_mode->clock = clk_round_rate(master->pxlclk, mode_clk) / 1000;
+
+	return true;
+}
+
+static struct drm_crtc_helper_funcs komeda_crtc_helper_funcs = {
+	.atomic_check	= komeda_crtc_atomic_check,
+	.atomic_flush	= komeda_crtc_atomic_flush,
+	.atomic_enable	= komeda_crtc_atomic_enable,
+	.atomic_disable	= komeda_crtc_atomic_disable,
+	.mode_valid	= komeda_crtc_mode_valid,
+	.mode_fixup	= komeda_crtc_mode_fixup,
 };
 
+static void komeda_crtc_reset(struct drm_crtc *crtc)
+{
+	struct komeda_crtc_state *state;
+
+	if (crtc->state)
+		__drm_atomic_helper_crtc_destroy_state(crtc->state);
+
+	kfree(to_kcrtc_st(crtc->state));
+	crtc->state = NULL;
+
+	state = kzalloc(sizeof(*state), GFP_KERNEL);
+	if (state) {
+		crtc->state = &state->base;
+		crtc->state->crtc = crtc;
+	}
+}
+
+static struct drm_crtc_state *
+komeda_crtc_atomic_duplicate_state(struct drm_crtc *crtc)
+{
+	struct komeda_crtc_state *old = to_kcrtc_st(crtc->state);
+	struct komeda_crtc_state *new;
+
+	new = kzalloc(sizeof(*new), GFP_KERNEL);
+	if (!new)
+		return NULL;
+
+	__drm_atomic_helper_crtc_duplicate_state(crtc, &new->base);
+
+	new->affected_pipes = old->active_pipes;
+
+	return &new->base;
+}
+
+static void komeda_crtc_atomic_destroy_state(struct drm_crtc *crtc,
+					     struct drm_crtc_state *state)
+{
+	__drm_atomic_helper_crtc_destroy_state(state);
+	kfree(to_kcrtc_st(state));
+}
+
+static int komeda_crtc_vblank_enable(struct drm_crtc *crtc)
+{
+	struct komeda_dev *mdev = crtc->dev->dev_private;
+	struct komeda_crtc *kcrtc = to_kcrtc(crtc);
+
+	mdev->funcs->on_off_vblank(mdev, kcrtc->master->id, true);
+	return 0;
+}
+
+static void komeda_crtc_vblank_disable(struct drm_crtc *crtc)
+{
+	struct komeda_dev *mdev = crtc->dev->dev_private;
+	struct komeda_crtc *kcrtc = to_kcrtc(crtc);
+
+	mdev->funcs->on_off_vblank(mdev, kcrtc->master->id, false);
+}
+
 static const struct drm_crtc_funcs komeda_crtc_funcs = {
+	.gamma_set		= drm_atomic_helper_legacy_gamma_set,
+	.destroy		= drm_crtc_cleanup,
+	.set_config		= drm_atomic_helper_set_config,
+	.page_flip		= drm_atomic_helper_page_flip,
+	.reset			= komeda_crtc_reset,
+	.atomic_duplicate_state	= komeda_crtc_atomic_duplicate_state,
+	.atomic_destroy_state	= komeda_crtc_atomic_destroy_state,
+	.enable_vblank		= komeda_crtc_vblank_enable,
+	.disable_vblank		= komeda_crtc_vblank_disable,
 };
 
 int komeda_kms_setup_crtcs(struct komeda_kms_dev *kms,
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_dev.c b/drivers/gpu/drm/arm/display/komeda/komeda_dev.c
index 70e9bb7fa30c..ca3599e4a4d3 100644
--- a/drivers/gpu/drm/arm/display/komeda/komeda_dev.c
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_dev.c
@@ -8,11 +8,99 @@
 #include <linux/of_device.h>
 #include <linux/of_graph.h>
 #include <linux/platform_device.h>
+#ifdef CONFIG_DEBUG_FS
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
+#endif
 
 #include <drm/drm_print.h>
 
 #include "komeda_dev.h"
 
+static int komeda_register_show(struct seq_file *sf, void *x)
+{
+	struct komeda_dev *mdev = sf->private;
+	int i;
+
+	if (mdev->funcs->dump_register)
+		mdev->funcs->dump_register(mdev, sf);
+
+	for (i = 0; i < mdev->n_pipelines; i++)
+		komeda_pipeline_dump_register(mdev->pipelines[i], sf);
+
+	return 0;
+}
+
+static int komeda_register_open(struct inode *inode, struct file *filp)
+{
+	return single_open(filp, komeda_register_show, inode->i_private);
+}
+
+static const struct file_operations komeda_register_fops = {
+	.owner		= THIS_MODULE,
+	.open		= komeda_register_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
+#ifdef CONFIG_DEBUG_FS
+static void komeda_debugfs_init(struct komeda_dev *mdev)
+{
+	if (!debugfs_initialized())
+		return;
+
+	mdev->debugfs_root = debugfs_create_dir("komeda", NULL);
+	if (IS_ERR_OR_NULL(mdev->debugfs_root))
+		return;
+
+	debugfs_create_file("register", 0444, mdev->debugfs_root,
+			    mdev, &komeda_register_fops);
+}
+#endif
+
+static ssize_t
+core_id_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct komeda_dev *mdev = dev_to_mdev(dev);
+
+	return snprintf(buf, PAGE_SIZE, "0x%08x\n", mdev->chip.core_id);
+}
+static DEVICE_ATTR_RO(core_id);
+
+static ssize_t
+config_id_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct komeda_dev *mdev = dev_to_mdev(dev);
+	struct komeda_pipeline *pipe = mdev->pipelines[0];
+	union komeda_config_id config_id;
+	int i;
+
+	memset(&config_id, 0, sizeof(config_id));
+
+	config_id.max_line_sz = pipe->layers[0]->hsize_in.end;
+	config_id.n_pipelines = mdev->n_pipelines;
+	config_id.n_scalers = pipe->n_scalers;
+	config_id.n_layers = pipe->n_layers;
+	config_id.n_richs = 0;
+	for (i = 0; i < pipe->n_layers; i++) {
+		if (pipe->layers[i]->layer_type == KOMEDA_FMT_RICH_LAYER)
+			config_id.n_richs++;
+	}
+	return snprintf(buf, PAGE_SIZE, "0x%08x\n", config_id.value);
+}
+static DEVICE_ATTR_RO(config_id);
+
+static struct attribute *komeda_sysfs_entries[] = {
+	&dev_attr_core_id.attr,
+	&dev_attr_config_id.attr,
+	NULL,
+};
+
+static struct attribute_group komeda_sysfs_attr_group = {
+	.attrs = komeda_sysfs_entries,
+};
+
 static int komeda_parse_pipe_dt(struct komeda_dev *mdev, struct device_node *np)
 {
 	struct komeda_pipeline *pipe;
@@ -53,6 +141,7 @@ static int komeda_parse_pipe_dt(struct komeda_dev *mdev, struct device_node *np)
 
 static int komeda_parse_dt(struct device *dev, struct komeda_dev *mdev)
 {
+	struct platform_device *pdev = to_platform_device(dev);
 	struct device_node *child, *np = dev->of_node;
 	struct clk *clk;
 	int ret;
@@ -62,6 +151,11 @@ static int komeda_parse_dt(struct device *dev, struct komeda_dev *mdev)
 		return PTR_ERR(clk);
 
 	mdev->mclk = clk;
+	mdev->irq  = platform_get_irq(pdev, 0);
+	if (mdev->irq < 0) {
+		DRM_ERROR("could not get IRQ number.\n");
+		return mdev->irq;
+	}
 
 	for_each_available_child_of_node(np, child) {
 		if (of_node_cmp(child->name, "pipeline") == 0) {
@@ -99,6 +193,8 @@ struct komeda_dev *komeda_dev_create(struct device *dev)
 	if (!mdev)
 		return ERR_PTR(-ENOMEM);
 
+	mutex_init(&mdev->lock);
+
 	mdev->dev = dev;
 	mdev->reg_base = devm_ioremap_resource(dev, io_res);
 	if (IS_ERR(mdev->reg_base)) {
@@ -147,6 +243,22 @@ struct komeda_dev *komeda_dev_create(struct device *dev)
 		goto err_cleanup;
 	}
 
+	err = komeda_assemble_pipelines(mdev);
+	if (err) {
+		DRM_ERROR("assemble display pipelines failed.\n");
+		goto err_cleanup;
+	}
+
+	err = sysfs_create_group(&dev->kobj, &komeda_sysfs_attr_group);
+	if (err) {
+		DRM_ERROR("create sysfs group failed.\n");
+		goto err_cleanup;
+	}
+
+#ifdef CONFIG_DEBUG_FS
+	komeda_debugfs_init(mdev);
+#endif
+
 	return mdev;
 
 err_cleanup:
@@ -160,6 +272,12 @@ void komeda_dev_destroy(struct komeda_dev *mdev)
 	struct komeda_dev_funcs *funcs = mdev->funcs;
 	int i;
 
+	sysfs_remove_group(&dev->kobj, &komeda_sysfs_attr_group);
+
+#ifdef CONFIG_DEBUG_FS
+	debugfs_remove_recursive(mdev->debugfs_root);
+#endif
+
 	for (i = 0; i < mdev->n_pipelines; i++) {
 		komeda_pipeline_destroy(mdev, mdev->pipelines[i]);
 		mdev->pipelines[i] = NULL;
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_dev.h b/drivers/gpu/drm/arm/display/komeda/komeda_dev.h
index 0f77dead6a23..29e03c4e1ffc 100644
--- a/drivers/gpu/drm/arm/display/komeda/komeda_dev.h
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_dev.h
@@ -13,6 +13,33 @@
 #include "malidp_product.h"
 #include "komeda_format_caps.h"
 
+#define KOMEDA_EVENT_VSYNC		BIT_ULL(0)
+#define KOMEDA_EVENT_FLIP		BIT_ULL(1)
+#define KOMEDA_EVENT_URUN		BIT_ULL(2)
+#define KOMEDA_EVENT_IBSY		BIT_ULL(3)
+#define KOMEDA_EVENT_OVR		BIT_ULL(4)
+#define KOMEDA_EVENT_EOW		BIT_ULL(5)
+#define KOMEDA_EVENT_MODE		BIT_ULL(6)
+
+#define KOMEDA_ERR_TETO			BIT_ULL(14)
+#define KOMEDA_ERR_TEMR			BIT_ULL(15)
+#define KOMEDA_ERR_TITR			BIT_ULL(16)
+#define KOMEDA_ERR_CPE			BIT_ULL(17)
+#define KOMEDA_ERR_CFGE			BIT_ULL(18)
+#define KOMEDA_ERR_AXIE			BIT_ULL(19)
+#define KOMEDA_ERR_ACE0			BIT_ULL(20)
+#define KOMEDA_ERR_ACE1			BIT_ULL(21)
+#define KOMEDA_ERR_ACE2			BIT_ULL(22)
+#define KOMEDA_ERR_ACE3			BIT_ULL(23)
+#define KOMEDA_ERR_DRIFTTO		BIT_ULL(24)
+#define KOMEDA_ERR_FRAMETO		BIT_ULL(25)
+#define KOMEDA_ERR_CSCE			BIT_ULL(26)
+#define KOMEDA_ERR_ZME			BIT_ULL(27)
+#define KOMEDA_ERR_MERR			BIT_ULL(28)
+#define KOMEDA_ERR_TCF			BIT_ULL(29)
+#define KOMEDA_ERR_TTNG			BIT_ULL(30)
+#define KOMEDA_ERR_TTF			BIT_ULL(31)
+
 /* malidp device id */
 enum {
 	MALI_D71 = 0,
@@ -39,6 +66,11 @@ struct komeda_product_data {
 
 struct komeda_dev;
 
+struct komeda_events {
+	u64 global;
+	u64 pipes[KOMEDA_MAX_PIPELINES];
+};
+
 /**
  * struct komeda_dev_funcs
  *
@@ -60,6 +92,49 @@ struct komeda_dev_funcs {
 	int (*enum_resources)(struct komeda_dev *mdev);
 	/** @cleanup: call to chip to cleanup komeda_dev->chip data */
 	void (*cleanup)(struct komeda_dev *mdev);
+	/**
+	 * @irq_handler:
+	 *
+	 * for CORE to get the HW event from the CHIP when interrupt happened.
+	 */
+	irqreturn_t (*irq_handler)(struct komeda_dev *mdev,
+				   struct komeda_events *events);
+	/** @enable_irq: enable irq */
+	int (*enable_irq)(struct komeda_dev *mdev);
+	/** @disable_irq: disable irq */
+	int (*disable_irq)(struct komeda_dev *mdev);
+	/** @on_off_vblank: notify HW to on/off vblank */
+	void (*on_off_vblank)(struct komeda_dev *mdev,
+			      int master_pipe, bool on);
+
+	/** @dump_register: Optional, dump registers to seq_file */
+	void (*dump_register)(struct komeda_dev *mdev, struct seq_file *seq);
+	/**
+	 * @change_opmode:
+	 *
+	 * Notify HW to switch to a new display operation mode.
+	 */
+	int (*change_opmode)(struct komeda_dev *mdev, int new_mode);
+	/** @flush: Notify the HW to flush or kickoff the update */
+	void (*flush)(struct komeda_dev *mdev,
+		      int master_pipe, u32 active_pipes);
+};
+
+/*
+ * DISPLAY_MODE describes how many display been enabled, and which will be
+ * passed to CHIP by &komeda_dev_funcs->change_opmode(), then CHIP can do the
+ * pipeline resources assignment according to this usage hint.
+ * -   KOMEDA_MODE_DISP0: Only one display enabled, pipeline-0 work as master.
+ * -   KOMEDA_MODE_DISP1: Only one display enabled, pipeline-0 work as master.
+ * -   KOMEDA_MODE_DUAL_DISP: Dual display mode, both display has been enabled.
+ * And D71 supports assign two pipelines to one single display on mode
+ * KOMEDA_MODE_DISP0/DISP1
+ */
+enum {
+	KOMEDA_MODE_INACTIVE	= 0,
+	KOMEDA_MODE_DISP0	= BIT(0),
+	KOMEDA_MODE_DISP1	= BIT(1),
+	KOMEDA_MODE_DUAL_DISP	= KOMEDA_MODE_DISP0 | KOMEDA_MODE_DISP1,
 };
 
 /**
@@ -70,18 +145,31 @@ struct komeda_dev_funcs {
  * control-abilites of device.
  */
 struct komeda_dev {
+	/** @dev: the base device structure */
 	struct device *dev;
+	/** @reg_base: the base address of komeda io space */
 	u32 __iomem   *reg_base;
 
+	/** @chip: the basic chip information */
 	struct komeda_chip_info chip;
 	/** @fmt_tbl: initialized by &komeda_dev_funcs->init_format_table */
 	struct komeda_format_caps_table fmt_tbl;
 	/** @pclk: APB clock for register access */
 	struct clk *pclk;
-	/** @mck: HW main engine clk */
+	/** @mclk: HW main engine clk */
 	struct clk *mclk;
 
+	/** @irq: irq number */
+	int irq;
+
+	/** @lock: used to protect dpmode */
+	struct mutex lock;
+	/** @dpmode: current display mode */
+	u32 dpmode;
+
+	/** @n_pipelines: the number of pipe in @pipelines */
 	int n_pipelines;
+	/** @pipelines: the komeda pipelines */
 	struct komeda_pipeline *pipelines[KOMEDA_MAX_PIPELINES];
 
 	/** @funcs: chip funcs to access to HW */
@@ -93,6 +181,9 @@ struct komeda_dev {
 	 * destroyed by &komeda_dev_funcs.cleanup()
 	 */
 	void *chip_data;
+
+	/** @debugfs_root: root directory of komeda debugfs */
+	struct dentry *debugfs_root;
 };
 
 static inline bool
@@ -107,4 +198,6 @@ d71_identify(u32 __iomem *reg, struct komeda_chip_info *chip);
 struct komeda_dev *komeda_dev_create(struct device *dev);
 void komeda_dev_destroy(struct komeda_dev *mdev);
 
+struct komeda_dev *dev_to_mdev(struct device *dev);
+
 #endif /*_KOMEDA_DEV_H_*/
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_drv.c b/drivers/gpu/drm/arm/display/komeda/komeda_drv.c
index 2bdd189b041d..cfa5068d9d1e 100644
--- a/drivers/gpu/drm/arm/display/komeda/komeda_drv.c
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_drv.c
@@ -17,6 +17,13 @@ struct komeda_drv {
 	struct komeda_kms_dev *kms;
 };
 
+struct komeda_dev *dev_to_mdev(struct device *dev)
+{
+	struct komeda_drv *mdrv = dev_get_drvdata(dev);
+
+	return mdrv ? mdrv->mdev : NULL;
+}
+
 static void komeda_unbind(struct device *dev)
 {
 	struct komeda_drv *mdrv = dev_get_drvdata(dev);
@@ -120,7 +127,7 @@ static const struct komeda_product_data komeda_products[] = {
 	},
 };
 
-const struct of_device_id komeda_of_match[] = {
+static const struct of_device_id komeda_of_match[] = {
 	{ .compatible = "arm,mali-d71", .data = &komeda_products[MALI_D71], },
 	{},
 };
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.h b/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.h
index 0de2e4a2afd2..ea2fe190c1e3 100644
--- a/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.h
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.h
@@ -10,11 +10,16 @@
 #include <drm/drm_framebuffer.h>
 #include "komeda_format_caps.h"
 
-/** struct komeda_fb - entend drm_framebuffer with komeda attribute */
+/**
+ * struct komeda_fb - Entending drm_framebuffer with komeda attribute
+ */
 struct komeda_fb {
 	/** @base: &drm_framebuffer */
 	struct drm_framebuffer base;
-	/* @format_caps: &komeda_format_caps */
+	/**
+	 * @format_caps:
+	 * extends drm_format_info for komeda specific information
+	 */
 	const struct komeda_format_caps *format_caps;
 	/** @aligned_w: aligned frame buffer width */
 	u32 aligned_w;
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_kms.c b/drivers/gpu/drm/arm/display/komeda/komeda_kms.c
index 47a58ab20434..86f6542afb40 100644
--- a/drivers/gpu/drm/arm/display/komeda/komeda_kms.c
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_kms.c
@@ -13,6 +13,7 @@
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_irq.h>
 #include <drm/drm_vblank.h>
 
 #include "komeda_dev.h"
@@ -25,18 +26,39 @@ static int komeda_gem_cma_dumb_create(struct drm_file *file,
 				      struct drm_device *dev,
 				      struct drm_mode_create_dumb *args)
 {
-	u32 alignment = 16; /* TODO get alignment from dev */
+	struct komeda_dev *mdev = dev->dev_private;
+	u32 pitch = DIV_ROUND_UP(args->width * args->bpp, 8);
 
-	args->pitch = ALIGN(DIV_ROUND_UP(args->width * args->bpp, 8),
-			    alignment);
+	args->pitch = ALIGN(pitch, mdev->chip.bus_width);
 
 	return drm_gem_cma_dumb_create_internal(file, dev, args);
 }
 
+static irqreturn_t komeda_kms_irq_handler(int irq, void *data)
+{
+	struct drm_device *drm = data;
+	struct komeda_dev *mdev = drm->dev_private;
+	struct komeda_kms_dev *kms = to_kdev(drm);
+	struct komeda_events evts;
+	irqreturn_t status;
+	u32 i;
+
+	/* Call into the CHIP to recognize events */
+	memset(&evts, 0, sizeof(evts));
+	status = mdev->funcs->irq_handler(mdev, &evts);
+
+	/* Notify the crtc to handle the events */
+	for (i = 0; i < kms->n_crtcs; i++)
+		komeda_crtc_handle_event(&kms->crtcs[i], &evts);
+
+	return status;
+}
+
 static struct drm_driver komeda_kms_driver = {
 	.driver_features = DRIVER_GEM | DRIVER_MODESET | DRIVER_ATOMIC |
-			   DRIVER_PRIME,
+			   DRIVER_PRIME | DRIVER_HAVE_IRQ,
 	.lastclose			= drm_fb_helper_lastclose,
+	.irq_handler			= komeda_kms_irq_handler,
 	.gem_free_object_unlocked	= drm_gem_cma_free_object,
 	.gem_vm_ops			= &drm_gem_cma_vm_ops,
 	.dumb_create			= komeda_gem_cma_dumb_create,
@@ -78,9 +100,37 @@ static const struct drm_mode_config_helper_funcs komeda_mode_config_helpers = {
 	.atomic_commit_tail = komeda_kms_commit_tail,
 };
 
+static int komeda_kms_check(struct drm_device *dev,
+			    struct drm_atomic_state *state)
+{
+	struct drm_crtc *crtc;
+	struct drm_crtc_state *old_crtc_st, *new_crtc_st;
+	int i, err;
+
+	err = drm_atomic_helper_check_modeset(dev, state);
+	if (err)
+		return err;
+
+	/* komeda need to re-calculate resource assumption in every commit
+	 * so need to add all affected_planes (even unchanged) to
+	 * drm_atomic_state.
+	 */
+	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_st, new_crtc_st, i) {
+		err = drm_atomic_add_affected_planes(state, crtc);
+		if (err)
+			return err;
+	}
+
+	err = drm_atomic_helper_check_planes(dev, state);
+	if (err)
+		return err;
+
+	return 0;
+}
+
 static const struct drm_mode_config_funcs komeda_mode_config_funcs = {
 	.fb_create		= komeda_fb_create,
-	.atomic_check		= drm_atomic_helper_check,
+	.atomic_check		= komeda_kms_check,
 	.atomic_commit		= drm_atomic_helper_commit,
 };
 
@@ -144,14 +194,25 @@ struct komeda_kms_dev *komeda_kms_attach(struct komeda_dev *mdev)
 
 	drm_mode_config_reset(drm);
 
-	err = drm_dev_register(drm, 0);
+	err = drm_irq_install(drm, mdev->irq);
 	if (err)
 		goto cleanup_mode_config;
 
+	err = mdev->funcs->enable_irq(mdev);
+	if (err)
+		goto uninstall_irq;
+
+	err = drm_dev_register(drm, 0);
+	if (err)
+		goto uninstall_irq;
+
 	return kms;
 
+uninstall_irq:
+	drm_irq_uninstall(drm);
 cleanup_mode_config:
 	drm_mode_config_cleanup(drm);
+	komeda_kms_cleanup_private_objs(kms);
 free_kms:
 	kfree(kms);
 	return ERR_PTR(err);
@@ -162,9 +223,11 @@ void komeda_kms_detach(struct komeda_kms_dev *kms)
 	struct drm_device *drm = &kms->base;
 	struct komeda_dev *mdev = drm->dev_private;
 
+	mdev->funcs->disable_irq(mdev);
 	drm_dev_unregister(drm);
+	drm_irq_uninstall(drm);
 	component_unbind_all(mdev->dev, drm);
-	komeda_kms_cleanup_private_objs(mdev);
+	komeda_kms_cleanup_private_objs(kms);
 	drm_mode_config_cleanup(drm);
 	drm->dev_private = NULL;
 	drm_dev_put(drm);
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_kms.h b/drivers/gpu/drm/arm/display/komeda/komeda_kms.h
index 874e9c9f0749..ac3d9209b4d9 100644
--- a/drivers/gpu/drm/arm/display/komeda/komeda_kms.h
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_kms.h
@@ -12,8 +12,12 @@
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_device.h>
 #include <drm/drm_writeback.h>
+#include <video/videomode.h>
+#include <video/display_timing.h>
 
-/** struct komeda_plane - komeda instance of drm_plane */
+/**
+ * struct komeda_plane - komeda instance of drm_plane
+ */
 struct komeda_plane {
 	/** @base: &drm_plane */
 	struct drm_plane base;
@@ -68,9 +72,14 @@ struct komeda_crtc {
 	 * merge into the master.
 	 */
 	struct komeda_pipeline *slave;
+
+	/** @disable_done: this flip_done is for tracing the disable */
+	struct completion *disable_done;
 };
 
-/** struct komeda_crtc_state */
+/**
+ * struct komeda_crtc_state
+ */
 struct komeda_crtc_state {
 	/** @base: &drm_crtc_state */
 	struct drm_crtc_state base;
@@ -78,7 +87,15 @@ struct komeda_crtc_state {
 	/* private properties */
 
 	/* computed state which are used by validate/check */
+	/**
+	 * @affected_pipes:
+	 * the affected pipelines in once display instance
+	 */
 	u32 affected_pipes;
+	/**
+	 * @active_pipes:
+	 * the active pipelines in once display instance
+	 */
 	u32 active_pipes;
 };
 
@@ -106,7 +123,10 @@ int komeda_kms_add_crtcs(struct komeda_kms_dev *kms, struct komeda_dev *mdev);
 int komeda_kms_add_planes(struct komeda_kms_dev *kms, struct komeda_dev *mdev);
 int komeda_kms_add_private_objs(struct komeda_kms_dev *kms,
 				struct komeda_dev *mdev);
-void komeda_kms_cleanup_private_objs(struct komeda_dev *mdev);
+void komeda_kms_cleanup_private_objs(struct komeda_kms_dev *kms);
+
+void komeda_crtc_handle_event(struct komeda_crtc   *kcrtc,
+			      struct komeda_events *evts);
 
 struct komeda_kms_dev *komeda_kms_attach(struct komeda_dev *mdev);
 void komeda_kms_detach(struct komeda_kms_dev *kms);
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.c b/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.c
index f1908e9ef128..c379439c6194 100644
--- a/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.c
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.c
@@ -19,17 +19,17 @@ komeda_pipeline_add(struct komeda_dev *mdev, size_t size,
 	if (mdev->n_pipelines + 1 > KOMEDA_MAX_PIPELINES) {
 		DRM_ERROR("Exceed max support %d pipelines.\n",
 			  KOMEDA_MAX_PIPELINES);
-		return NULL;
+		return ERR_PTR(-ENOSPC);
 	}
 
 	if (size < sizeof(*pipe)) {
 		DRM_ERROR("Request pipeline size too small.\n");
-		return NULL;
+		return ERR_PTR(-EINVAL);
 	}
 
 	pipe = devm_kzalloc(mdev->dev, size, GFP_KERNEL);
 	if (!pipe)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	pipe->mdev = mdev;
 	pipe->id   = mdev->n_pipelines;
@@ -62,7 +62,7 @@ void komeda_pipeline_destroy(struct komeda_dev *mdev,
 	devm_kfree(mdev->dev, pipe);
 }
 
-struct komeda_component **
+static struct komeda_component **
 komeda_pipeline_get_component_pos(struct komeda_pipeline *pipe, int id)
 {
 	struct komeda_dev *mdev = pipe->mdev;
@@ -142,32 +142,32 @@ komeda_component_add(struct komeda_pipeline *pipe,
 	if (max_active_inputs > KOMEDA_COMPONENT_N_INPUTS) {
 		WARN(1, "please large KOMEDA_COMPONENT_N_INPUTS to %d.\n",
 		     max_active_inputs);
-		return NULL;
+		return ERR_PTR(-ENOSPC);
 	}
 
 	pos = komeda_pipeline_get_component_pos(pipe, id);
 	if (!pos || (*pos))
-		return NULL;
+		return ERR_PTR(-EINVAL);
 
 	if (has_bit(id, KOMEDA_PIPELINE_LAYERS)) {
 		idx = id - KOMEDA_COMPONENT_LAYER0;
 		num = &pipe->n_layers;
 		if (idx != pipe->n_layers) {
 			DRM_ERROR("please add Layer by id sequence.\n");
-			return NULL;
+			return ERR_PTR(-EINVAL);
 		}
 	} else if (has_bit(id,  KOMEDA_PIPELINE_SCALERS)) {
 		idx = id - KOMEDA_COMPONENT_SCALER0;
 		num = &pipe->n_scalers;
 		if (idx != pipe->n_scalers) {
 			DRM_ERROR("please add Scaler by id sequence.\n");
-			return NULL;
+			return ERR_PTR(-EINVAL);
 		}
 	}
 
 	c = devm_kzalloc(pipe->mdev->dev, comp_sz, GFP_KERNEL);
 	if (!c)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	c->id = id;
 	c->hw_id = hw_id;
@@ -200,3 +200,98 @@ void komeda_component_destroy(struct komeda_dev *mdev,
 {
 	devm_kfree(mdev->dev, c);
 }
+
+static void komeda_component_dump(struct komeda_component *c)
+{
+	if (!c)
+		return;
+
+	DRM_DEBUG("	%s: ID %d-0x%08lx.\n",
+		  c->name, c->id, BIT(c->id));
+	DRM_DEBUG("		max_active_inputs:%d, supported_inputs: 0x%08x.\n",
+		  c->max_active_inputs, c->supported_inputs);
+	DRM_DEBUG("		max_active_outputs:%d, supported_outputs: 0x%08x.\n",
+		  c->max_active_outputs, c->supported_outputs);
+}
+
+static void komeda_pipeline_dump(struct komeda_pipeline *pipe)
+{
+	struct komeda_component *c;
+	int id;
+
+	DRM_INFO("Pipeline-%d: n_layers: %d, n_scalers: %d, output: %s\n",
+		 pipe->id, pipe->n_layers, pipe->n_scalers,
+		 pipe->of_output_dev ? pipe->of_output_dev->full_name : "none");
+
+	dp_for_each_set_bit(id, pipe->avail_comps) {
+		c = komeda_pipeline_get_component(pipe, id);
+
+		komeda_component_dump(c);
+	}
+}
+
+static void komeda_component_verify_inputs(struct komeda_component *c)
+{
+	struct komeda_pipeline *pipe = c->pipeline;
+	struct komeda_component *input;
+	int id;
+
+	dp_for_each_set_bit(id, c->supported_inputs) {
+		input = komeda_pipeline_get_component(pipe, id);
+		if (!input) {
+			c->supported_inputs &= ~(BIT(id));
+			DRM_WARN("Can not find input(ID-%d) for component: %s.\n",
+				 id, c->name);
+			continue;
+		}
+
+		input->supported_outputs |= BIT(c->id);
+	}
+}
+
+static void komeda_pipeline_assemble(struct komeda_pipeline *pipe)
+{
+	struct komeda_component *c;
+	int id;
+
+	dp_for_each_set_bit(id, pipe->avail_comps) {
+		c = komeda_pipeline_get_component(pipe, id);
+
+		komeda_component_verify_inputs(c);
+	}
+}
+
+int komeda_assemble_pipelines(struct komeda_dev *mdev)
+{
+	struct komeda_pipeline *pipe;
+	int i;
+
+	for (i = 0; i < mdev->n_pipelines; i++) {
+		pipe = mdev->pipelines[i];
+
+		komeda_pipeline_assemble(pipe);
+		komeda_pipeline_dump(pipe);
+	}
+
+	return 0;
+}
+
+void komeda_pipeline_dump_register(struct komeda_pipeline *pipe,
+				   struct seq_file *sf)
+{
+	struct komeda_component *c;
+	u32 id;
+
+	seq_printf(sf, "\n======== Pipeline-%d ==========\n", pipe->id);
+
+	if (pipe->funcs && pipe->funcs->dump_register)
+		pipe->funcs->dump_register(pipe, sf);
+
+	dp_for_each_set_bit(id, pipe->avail_comps) {
+		c = komeda_pipeline_get_component(pipe, id);
+
+		seq_printf(sf, "\n------%s------\n", c->name);
+		if (c->funcs->dump_register)
+			c->funcs->dump_register(c, sf);
+	}
+}
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.h b/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.h
index 8c950bc8ae96..b1f813a349a4 100644
--- a/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.h
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_pipeline.h
@@ -90,32 +90,35 @@ struct komeda_component {
 	u32 __iomem *reg;
 	/** @id: component id */
 	u32 id;
-	/** @hw_ic: component hw id,
-	 *  which is initialized by chip and used by chip only
+	/**
+	 * @hw_id: component hw id,
+	 * which is initialized by chip and used by chip only
 	 */
 	u32 hw_id;
 
 	/**
 	 * @max_active_inputs:
-	 * @max_active_outpus:
+	 * @max_active_outputs:
 	 *
-	 * maximum number of inputs/outputs that can be active in the same time
+	 * maximum number of inputs/outputs that can be active at the same time
 	 * Note:
 	 * the number isn't the bit number of @supported_inputs or
 	 * @supported_outputs, but may be less than it, since component may not
 	 * support enabling all @supported_inputs/outputs at the same time.
 	 */
 	u8 max_active_inputs;
+	/** @max_active_outputs: maximum number of outputs */
 	u8 max_active_outputs;
 	/**
 	 * @supported_inputs:
 	 * @supported_outputs:
 	 *
-	 * bitmask of BIT(component->id) for the supported inputs/outputs
+	 * bitmask of BIT(component->id) for the supported inputs/outputs,
 	 * describes the possibilities of how a component is linked into a
 	 * pipeline.
 	 */
 	u32 supported_inputs;
+	/** @supported_outputs: bitmask of supported output componenet ids */
 	u32 supported_outputs;
 
 	/**
@@ -134,7 +137,8 @@ struct komeda_component {
 struct komeda_component_output {
 	/** @component: indicate which component the data comes from */
 	struct komeda_component *component;
-	/** @output_port:
+	/**
+	 * @output_port:
 	 * the output port of the &komeda_component_output.component
 	 */
 	u8 output_port;
@@ -150,11 +154,12 @@ struct komeda_component_output {
 struct komeda_component_state {
 	/** @obj: tracking component_state by drm_atomic_state */
 	struct drm_private_state obj;
+	/** @component: backpointer to the component */
 	struct komeda_component *component;
 	/**
 	 * @binding_user:
-	 * currently bound user, the user can be crtc/plane/wb_conn, which is
-	 * valid decided by @component and @inputs
+	 * currently bound user, the user can be @crtc, @plane or @wb_conn,
+	 * which is valid decided by @component and @inputs
 	 *
 	 * -  Layer: its user always is plane.
 	 * -  compiz/improc/timing_ctrlr: the user is crtc.
@@ -162,20 +167,24 @@ struct komeda_component_state {
 	 * -  scaler: plane when input is layer, wb_conn if input is compiz.
 	 */
 	union {
+		/** @crtc: backpointer for user crtc */
 		struct drm_crtc *crtc;
+		/** @plane: backpointer for user plane */
 		struct drm_plane *plane;
+		/** @wb_conn: backpointer for user wb_connector  */
 		struct drm_connector *wb_conn;
 		void *binding_user;
 	};
+
 	/**
 	 * @active_inputs:
 	 *
 	 * active_inputs is bitmask of @inputs index
 	 *
-	 * -  active_inputs = changed_active_inputs + unchanged_active_inputs
-	 * -  affected_inputs = old->active_inputs + new->active_inputs;
+	 * -  active_inputs = changed_active_inputs | unchanged_active_inputs
+	 * -  affected_inputs = old->active_inputs | new->active_inputs;
 	 * -  disabling_inputs = affected_inputs ^ active_inputs;
-	 * -  changed_inputs = disabling_inputs + changed_active_inputs;
+	 * -  changed_inputs = disabling_inputs | changed_active_inputs;
 	 *
 	 * NOTE:
 	 * changed_inputs doesn't include all active_input but only
@@ -183,7 +192,9 @@ struct komeda_component_state {
 	 * level for dirty update.
 	 */
 	u16 active_inputs;
+	/** @changed_active_inputs: bitmask of the changed @active_inputs */
 	u16 changed_active_inputs;
+	/** @affected_inputs: bitmask for affected @inputs */
 	u16 affected_inputs;
 	/**
 	 * @inputs:
@@ -204,57 +215,96 @@ static inline u16 component_changed_inputs(struct komeda_component_state *st)
 	return component_disabling_inputs(st) | st->changed_active_inputs;
 }
 
+#define for_each_changed_input(st, i)	\
+	for ((i) = 0; (i) < (st)->component->max_active_inputs; (i)++)	\
+		if (has_bit((i), component_changed_inputs(st)))
+
 #define to_comp(__c)	(((__c) == NULL) ? NULL : &((__c)->base))
 #define to_cpos(__c)	((struct komeda_component **)&(__c))
 
-/* these structures are going to be filled in in uture patches */
 struct komeda_layer {
 	struct komeda_component base;
-	/* layer specific features and caps */
-	int layer_type; /* RICH, SIMPLE or WB */
+	/* accepted h/v input range before rotation */
+	struct malidp_range hsize_in, vsize_in;
+	u32 layer_type; /* RICH, SIMPLE or WB */
+	u32 supported_rots;
 };
 
 struct komeda_layer_state {
 	struct komeda_component_state base;
 	/* layer specific configuration state */
+	u16 hsize, vsize;
+	u32 rot;
+	dma_addr_t addr[3];
 };
 
-struct komeda_compiz {
+struct komeda_scaler {
 	struct komeda_component base;
-	/* compiz specific features and caps */
+	/* scaler features and caps */
 };
 
-struct komeda_compiz_state {
+struct komeda_scaler_state {
 	struct komeda_component_state base;
-	/* compiz specific configuration state */
 };
 
-struct komeda_scaler {
+struct komeda_compiz {
 	struct komeda_component base;
-	/* scaler features and caps */
+	struct malidp_range hsize, vsize;
 };
 
-struct komeda_scaler_state {
+struct komeda_compiz_input_cfg {
+	u16 hsize, vsize;
+	u16 hoffset, voffset;
+	u8 pixel_blend_mode, layer_alpha;
+};
+
+struct komeda_compiz_state {
 	struct komeda_component_state base;
+	/* composition size */
+	u16 hsize, vsize;
+	struct komeda_compiz_input_cfg cins[KOMEDA_COMPONENT_N_INPUTS];
 };
 
 struct komeda_improc {
 	struct komeda_component base;
+	u32 supported_color_formats;  /* DRM_RGB/YUV444/YUV420*/
+	u32 supported_color_depths; /* BIT(8) | BIT(10)*/
+	u8 supports_degamma : 1;
+	u8 supports_csc : 1;
+	u8 supports_gamma : 1;
 };
 
 struct komeda_improc_state {
 	struct komeda_component_state base;
+	u16 hsize, vsize;
 };
 
 /* display timing controller */
 struct komeda_timing_ctrlr {
 	struct komeda_component base;
+	u8 supports_dual_link : 1;
 };
 
 struct komeda_timing_ctrlr_state {
 	struct komeda_component_state base;
 };
 
+/* Why define A separated structure but not use plane_state directly ?
+ * 1. Komeda supports layer_split which means a plane_state can be split and
+ *    handled by two layers, one layer only handle half of plane image.
+ * 2. Fix up the user properties according to HW's capabilities, like user
+ *    set rotation to R180, but HW only supports REFLECT_X+Y. the rot here is
+ *    after drm_rotation_simplify()
+ */
+struct komeda_data_flow_cfg {
+	struct komeda_component_output input;
+	u16 in_x, in_y, in_w, in_h;
+	u32 out_x, out_y, out_w, out_h;
+	u32 rot;
+	int blending_zorder;
+	u8 pixel_blend_mode, layer_alpha;
+};
+
 /** struct komeda_pipeline_funcs */
 struct komeda_pipeline_funcs {
 	/* dump_register: Optional, dump registers to seq_file */
@@ -280,14 +330,23 @@ struct komeda_pipeline {
 	int id;
 	/** @avail_comps: available components mask of pipeline */
 	u32 avail_comps;
+	/** @n_layers: the number of layer on @layers */
 	int n_layers;
+	/** @layers: the pipeline layers */
 	struct komeda_layer *layers[KOMEDA_PIPELINE_MAX_LAYERS];
+	/** @n_scalers: the number of scaler on @scalers */
 	int n_scalers;
+	/** @scalers: the pipeline scalers */
 	struct komeda_scaler *scalers[KOMEDA_PIPELINE_MAX_SCALERS];
+	/** @compiz: compositor */
 	struct komeda_compiz *compiz;
+	/** @wb_layer: writeback layer */
 	struct komeda_layer  *wb_layer;
+	/** @improc: post image processor */
 	struct komeda_improc *improc;
+	/** @ctrlr: timing controller */
 	struct komeda_timing_ctrlr *ctrlr;
+	/** @funcs: chip pipeline functions */
 	struct komeda_pipeline_funcs *funcs; /* private pipeline functions */
 
 	/** @of_node: pipeline dt node */
@@ -308,6 +367,7 @@ struct komeda_pipeline {
 struct komeda_pipeline_state {
 	/** @obj: tracking pipeline_state by drm_atomic_state */
 	struct drm_private_state obj;
+	/** @pipe: backpointer to the pipeline */
 	struct komeda_pipeline *pipe;
 	/** @crtc: currently bound crtc */
 	struct drm_crtc *crtc;
@@ -340,10 +400,13 @@ komeda_pipeline_add(struct komeda_dev *mdev, size_t size,
 		    struct komeda_pipeline_funcs *funcs);
 void komeda_pipeline_destroy(struct komeda_dev *mdev,
 			     struct komeda_pipeline *pipe);
-
+int komeda_assemble_pipelines(struct komeda_dev *mdev);
 struct komeda_component *
 komeda_pipeline_get_component(struct komeda_pipeline *pipe, int id);
 
+void komeda_pipeline_dump_register(struct komeda_pipeline *pipe,
+				   struct seq_file *sf);
+
 /* component APIs */
 struct komeda_component *
 komeda_component_add(struct komeda_pipeline *pipe,
@@ -356,4 +419,26 @@ komeda_component_add(struct komeda_pipeline *pipe,
 void komeda_component_destroy(struct komeda_dev *mdev,
 			      struct komeda_component *c);
 
+struct komeda_plane_state;
+struct komeda_crtc_state;
+struct komeda_crtc;
+
+int komeda_build_layer_data_flow(struct komeda_layer *layer,
+				 struct komeda_plane_state *kplane_st,
+				 struct komeda_crtc_state *kcrtc_st,
+				 struct komeda_data_flow_cfg *dflow);
+int komeda_build_display_data_flow(struct komeda_crtc *kcrtc,
+				   struct komeda_crtc_state *kcrtc_st);
+
+int komeda_release_unclaimed_resources(struct komeda_pipeline *pipe,
+				       struct komeda_crtc_state *kcrtc_st);
+
+struct komeda_pipeline_state *
+komeda_pipeline_get_old_state(struct komeda_pipeline *pipe,
+			      struct drm_atomic_state *state);
+void komeda_pipeline_disable(struct komeda_pipeline *pipe,
+			     struct drm_atomic_state *old_state);
+void komeda_pipeline_update(struct komeda_pipeline *pipe,
+			    struct drm_atomic_state *old_state);
+
 #endif /* _KOMEDA_PIPELINE_H_*/
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_pipeline_state.c b/drivers/gpu/drm/arm/display/komeda/komeda_pipeline_state.c
new file mode 100644
index 000000000000..36570d7dad61
--- /dev/null
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_pipeline_state.c
@@ -0,0 +1,610 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * (C) COPYRIGHT 2018 ARM Limited. All rights reserved.
+ * Author: James.Qian.Wang <james.qian.wang@arm.com>
+ *
+ */
+
+#include <drm/drm_print.h>
+#include <linux/clk.h>
+#include "komeda_dev.h"
+#include "komeda_kms.h"
+#include "komeda_pipeline.h"
+#include "komeda_framebuffer.h"
+
+static inline bool is_switching_user(void *old, void *new)
+{
+	if (!old || !new)
+		return false;
+
+	return old != new;
+}
+
+static struct komeda_pipeline_state *
+komeda_pipeline_get_state(struct komeda_pipeline *pipe,
+			  struct drm_atomic_state *state)
+{
+	struct drm_private_state *priv_st;
+
+	priv_st = drm_atomic_get_private_obj_state(state, &pipe->obj);
+	if (IS_ERR(priv_st))
+		return ERR_CAST(priv_st);
+
+	return priv_to_pipe_st(priv_st);
+}
+
+struct komeda_pipeline_state *
+komeda_pipeline_get_old_state(struct komeda_pipeline *pipe,
+			      struct drm_atomic_state *state)
+{
+	struct drm_private_state *priv_st;
+
+	priv_st = drm_atomic_get_old_private_obj_state(state, &pipe->obj);
+	if (priv_st)
+		return priv_to_pipe_st(priv_st);
+	return NULL;
+}
+
+static struct komeda_pipeline_state *
+komeda_pipeline_get_new_state(struct komeda_pipeline *pipe,
+			      struct drm_atomic_state *state)
+{
+	struct drm_private_state *priv_st;
+
+	priv_st = drm_atomic_get_new_private_obj_state(state, &pipe->obj);
+	if (priv_st)
+		return priv_to_pipe_st(priv_st);
+	return NULL;
+}
+
+/* Assign pipeline for crtc */
+static struct komeda_pipeline_state *
+komeda_pipeline_get_state_and_set_crtc(struct komeda_pipeline *pipe,
+				       struct drm_atomic_state *state,
+				       struct drm_crtc *crtc)
+{
+	struct komeda_pipeline_state *st;
+
+	st = komeda_pipeline_get_state(pipe, state);
+	if (IS_ERR(st))
+		return st;
+
+	if (is_switching_user(crtc, st->crtc)) {
+		DRM_DEBUG_ATOMIC("CRTC%d required pipeline%d is busy.\n",
+				 drm_crtc_index(crtc), pipe->id);
+		return ERR_PTR(-EBUSY);
+	}
+
+	/* pipeline only can be disabled when the it is free or unused */
+	if (!crtc && st->active_comps) {
+		DRM_DEBUG_ATOMIC("Disabling a busy pipeline:%d.\n", pipe->id);
+		return ERR_PTR(-EBUSY);
+	}
+
+	st->crtc = crtc;
+
+	if (crtc) {
+		struct komeda_crtc_state *kcrtc_st;
+
+		kcrtc_st = to_kcrtc_st(drm_atomic_get_new_crtc_state(state,
+								     crtc));
+
+		kcrtc_st->active_pipes |= BIT(pipe->id);
+		kcrtc_st->affected_pipes |= BIT(pipe->id);
+	}
+	return st;
+}
+
+static struct komeda_component_state *
+komeda_component_get_state(struct komeda_component *c,
+			   struct drm_atomic_state *state)
+{
+	struct drm_private_state *priv_st;
+
+	WARN_ON(!drm_modeset_is_locked(&c->pipeline->obj.lock));
+
+	priv_st = drm_atomic_get_private_obj_state(state, &c->obj);
+	if (IS_ERR(priv_st))
+		return ERR_CAST(priv_st);
+
+	return priv_to_comp_st(priv_st);
+}
+
+static struct komeda_component_state *
+komeda_component_get_old_state(struct komeda_component *c,
+			       struct drm_atomic_state *state)
+{
+	struct drm_private_state *priv_st;
+
+	priv_st = drm_atomic_get_old_private_obj_state(state, &c->obj);
+	if (priv_st)
+		return priv_to_comp_st(priv_st);
+	return NULL;
+}
+
+/**
+ * komeda_component_get_state_and_set_user()
+ *
+ * @c: component to get state and set user
+ * @state: global atomic state
+ * @user: direct user, the binding user
+ * @crtc: the CRTC user, the big boss :)
+ *
+ * This function accepts two users:
+ * -   The direct user: can be plane/crtc/wb_connector depends on component
+ * -   The big boss (CRTC)
+ * CRTC is the big boss (the final user), because all component resources
+ * eventually will be assigned to CRTC, like the layer will be binding to
+ * kms_plane, but kms plane will be binding to a CRTC eventually.
+ *
+ * The big boss (CRTC) is for pipeline assignment, since &komeda_component isn't
+ * independent and can be assigned to CRTC freely, but belongs to a specific
+ * pipeline, only pipeline can be shared between crtc, and pipeline as a whole
+ * (include all the internal components) assigned to a specific CRTC.
+ *
+ * So when set a user to komeda_component, need first to check the status of
+ * component->pipeline to see if the pipeline is available on this specific
+ * CRTC. if the pipeline is busy (assigned to another CRTC), even the required
+ * component is free, the component still cannot be assigned to the direct user.
+ */
+static struct komeda_component_state *
+komeda_component_get_state_and_set_user(struct komeda_component *c,
+					struct drm_atomic_state *state,
+					void *user,
+					struct drm_crtc *crtc)
+{
+	struct komeda_pipeline_state *pipe_st;
+	struct komeda_component_state *st;
+
+	/* First check if the pipeline is available */
+	pipe_st = komeda_pipeline_get_state_and_set_crtc(c->pipeline,
+							 state, crtc);
+	if (IS_ERR(pipe_st))
+		return ERR_CAST(pipe_st);
+
+	st = komeda_component_get_state(c, state);
+	if (IS_ERR(st))
+		return st;
+
+	/* check if the component has been occupied */
+	if (is_switching_user(user, st->binding_user)) {
+		DRM_DEBUG_ATOMIC("required %s is busy.\n", c->name);
+		return ERR_PTR(-EBUSY);
+	}
+
+	st->binding_user = user;
+	/* mark the component as active if user is valid */
+	if (st->binding_user)
+		pipe_st->active_comps |= BIT(c->id);
+
+	return st;
+}
+
+static void
+komeda_component_add_input(struct komeda_component_state *state,
+			   struct komeda_component_output *input,
+			   int idx)
+{
+	struct komeda_component *c = state->component;
+
+	WARN_ON((idx < 0 || idx >= c->max_active_inputs));
+
+	/* since the inputs[i] is only valid when it is active. So if a input[i]
+	 * is a newly enabled input which switches from disable to enable, then
+	 * the old inputs[i] is undefined (NOT zeroed), we can not rely on
+	 * memcmp, but directly mark it changed
+	 */
+	if (!has_bit(idx, state->affected_inputs) ||
+	    memcmp(&state->inputs[idx], input, sizeof(*input))) {
+		memcpy(&state->inputs[idx], input, sizeof(*input));
+		state->changed_active_inputs |= BIT(idx);
+	}
+	state->active_inputs |= BIT(idx);
+	state->affected_inputs |= BIT(idx);
+}
+
+static int
+komeda_component_check_input(struct komeda_component_state *state,
+			     struct komeda_component_output *input,
+			     int idx)
+{
+	struct komeda_component *c = state->component;
+
+	if ((idx < 0) || (idx >= c->max_active_inputs)) {
+		DRM_DEBUG_ATOMIC("%s invalid input id: %d.\n", c->name, idx);
+		return -EINVAL;
+	}
+
+	if (has_bit(idx, state->active_inputs)) {
+		DRM_DEBUG_ATOMIC("%s required input_id: %d has been occupied already.\n",
+				 c->name, idx);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static void
+komeda_component_set_output(struct komeda_component_output *output,
+			    struct komeda_component *comp,
+			    u8 output_port)
+{
+	output->component = comp;
+	output->output_port = output_port;
+}
+
+static int
+komeda_component_validate_private(struct komeda_component *c,
+				  struct komeda_component_state *st)
+{
+	int err;
+
+	if (!c->funcs->validate)
+		return 0;
+
+	err = c->funcs->validate(c, st);
+	if (err)
+		DRM_DEBUG_ATOMIC("%s validate private failed.\n", c->name);
+
+	return err;
+}
+
+static int
+komeda_layer_check_cfg(struct komeda_layer *layer,
+		       struct komeda_plane_state *kplane_st,
+		       struct komeda_data_flow_cfg *dflow)
+{
+	if (!in_range(&layer->hsize_in, dflow->in_w)) {
+		DRM_DEBUG_ATOMIC("src_w: %d is out of range.\n", dflow->in_w);
+		return -EINVAL;
+	}
+
+	if (!in_range(&layer->vsize_in, dflow->in_h)) {
+		DRM_DEBUG_ATOMIC("src_h: %d is out of range.\n", dflow->in_h);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int
+komeda_layer_validate(struct komeda_layer *layer,
+		      struct komeda_plane_state *kplane_st,
+		      struct komeda_data_flow_cfg *dflow)
+{
+	struct drm_plane_state *plane_st = &kplane_st->base;
+	struct drm_framebuffer *fb = plane_st->fb;
+	struct komeda_fb *kfb = to_kfb(fb);
+	struct komeda_component_state *c_st;
+	struct komeda_layer_state *st;
+	int i, err;
+
+	err = komeda_layer_check_cfg(layer, kplane_st, dflow);
+	if (err)
+		return err;
+
+	c_st = komeda_component_get_state_and_set_user(&layer->base,
+			plane_st->state, plane_st->plane, plane_st->crtc);
+	if (IS_ERR(c_st))
+		return PTR_ERR(c_st);
+
+	st = to_layer_st(c_st);
+
+	st->rot = dflow->rot;
+	st->hsize = kfb->aligned_w;
+	st->vsize = kfb->aligned_h;
+
+	for (i = 0; i < fb->format->num_planes; i++)
+		st->addr[i] = komeda_fb_get_pixel_addr(kfb, dflow->in_x,
+						       dflow->in_y, i);
+
+	err = komeda_component_validate_private(&layer->base, c_st);
+	if (err)
+		return err;
+
+	/* update the data flow for the next stage */
+	komeda_component_set_output(&dflow->input, &layer->base, 0);
+
+	return 0;
+}
+
+static void pipeline_composition_size(struct komeda_crtc_state *kcrtc_st,
+				      u16 *hsize, u16 *vsize)
+{
+	struct drm_display_mode *m = &kcrtc_st->base.adjusted_mode;
+
+	if (hsize)
+		*hsize = m->hdisplay;
+	if (vsize)
+		*vsize = m->vdisplay;
+}
+
+static int
+komeda_compiz_set_input(struct komeda_compiz *compiz,
+			struct komeda_crtc_state *kcrtc_st,
+			struct komeda_data_flow_cfg *dflow)
+{
+	struct drm_atomic_state *drm_st = kcrtc_st->base.state;
+	struct komeda_component_state *c_st, *old_st;
+	struct komeda_compiz_input_cfg *cin;
+	u16 compiz_w, compiz_h;
+	int idx = dflow->blending_zorder;
+
+	pipeline_composition_size(kcrtc_st, &compiz_w, &compiz_h);
+	/* check display rect */
+	if ((dflow->out_x + dflow->out_w > compiz_w) ||
+	    (dflow->out_y + dflow->out_h > compiz_h) ||
+	     dflow->out_w == 0 || dflow->out_h == 0) {
+		DRM_DEBUG_ATOMIC("invalid disp rect [x=%d, y=%d, w=%d, h=%d]\n",
+				 dflow->out_x, dflow->out_y,
+				 dflow->out_w, dflow->out_h);
+		return -EINVAL;
+	}
+
+	c_st = komeda_component_get_state_and_set_user(&compiz->base, drm_st,
+			kcrtc_st->base.crtc, kcrtc_st->base.crtc);
+	if (IS_ERR(c_st))
+		return PTR_ERR(c_st);
+
+	if (komeda_component_check_input(c_st, &dflow->input, idx))
+		return -EINVAL;
+
+	cin = &(to_compiz_st(c_st)->cins[idx]);
+
+	cin->hsize   = dflow->out_w;
+	cin->vsize   = dflow->out_h;
+	cin->hoffset = dflow->out_x;
+	cin->voffset = dflow->out_y;
+	cin->pixel_blend_mode = dflow->pixel_blend_mode;
+	cin->layer_alpha = dflow->layer_alpha;
+
+	old_st = komeda_component_get_old_state(&compiz->base, drm_st);
+	WARN_ON(!old_st);
+
+	/* compare with old to check if this input has been changed */
+	if (memcmp(&(to_compiz_st(old_st)->cins[idx]), cin, sizeof(*cin)))
+		c_st->changed_active_inputs |= BIT(idx);
+
+	komeda_component_add_input(c_st, &dflow->input, idx);
+
+	return 0;
+}
+
+static int
+komeda_compiz_validate(struct komeda_compiz *compiz,
+		       struct komeda_crtc_state *state,
+		       struct komeda_data_flow_cfg *dflow)
+{
+	struct komeda_component_state *c_st;
+	struct komeda_compiz_state *st;
+
+	c_st = komeda_component_get_state_and_set_user(&compiz->base,
+			state->base.state, state->base.crtc, state->base.crtc);
+	if (IS_ERR(c_st))
+		return PTR_ERR(c_st);
+
+	st = to_compiz_st(c_st);
+
+	pipeline_composition_size(state, &st->hsize, &st->vsize);
+
+	komeda_component_set_output(&dflow->input, &compiz->base, 0);
+
+	/* compiz output dflow will be fed to the next pipeline stage, prepare
+	 * the data flow configuration for the next stage
+	 */
+	if (dflow) {
+		dflow->in_w = st->hsize;
+		dflow->in_h = st->vsize;
+		dflow->out_w = dflow->in_w;
+		dflow->out_h = dflow->in_h;
+		/* the output data of compiz doesn't have alpha, it only can be
+		 * used as bottom layer when blend it with master layers
+		 */
+		dflow->pixel_blend_mode = DRM_MODE_BLEND_PIXEL_NONE;
+		dflow->layer_alpha = 0xFF;
+		dflow->blending_zorder = 0;
+	}
+
+	return 0;
+}
+
+static int
+komeda_improc_validate(struct komeda_improc *improc,
+		       struct komeda_crtc_state *kcrtc_st,
+		       struct komeda_data_flow_cfg *dflow)
+{
+	struct drm_crtc *crtc = kcrtc_st->base.crtc;
+	struct komeda_component_state *c_st;
+	struct komeda_improc_state *st;
+
+	c_st = komeda_component_get_state_and_set_user(&improc->base,
+			kcrtc_st->base.state, crtc, crtc);
+	if (IS_ERR(c_st))
+		return PTR_ERR(c_st);
+
+	st = to_improc_st(c_st);
+
+	st->hsize = dflow->in_w;
+	st->vsize = dflow->in_h;
+
+	komeda_component_add_input(&st->base, &dflow->input, 0);
+	komeda_component_set_output(&dflow->input, &improc->base, 0);
+
+	return 0;
+}
+
+static int
+komeda_timing_ctrlr_validate(struct komeda_timing_ctrlr *ctrlr,
+			     struct komeda_crtc_state *kcrtc_st,
+			     struct komeda_data_flow_cfg *dflow)
+{
+	struct drm_crtc *crtc = kcrtc_st->base.crtc;
+	struct komeda_timing_ctrlr_state *st;
+	struct komeda_component_state *c_st;
+
+	c_st = komeda_component_get_state_and_set_user(&ctrlr->base,
+			kcrtc_st->base.state, crtc, crtc);
+	if (IS_ERR(c_st))
+		return PTR_ERR(c_st);
+
+	st = to_ctrlr_st(c_st);
+
+	komeda_component_add_input(&st->base, &dflow->input, 0);
+	komeda_component_set_output(&dflow->input, &ctrlr->base, 0);
+
+	return 0;
+}
+
+int komeda_build_layer_data_flow(struct komeda_layer *layer,
+				 struct komeda_plane_state *kplane_st,
+				 struct komeda_crtc_state *kcrtc_st,
+				 struct komeda_data_flow_cfg *dflow)
+{
+	struct drm_plane *plane = kplane_st->base.plane;
+	struct komeda_pipeline *pipe = layer->base.pipeline;
+	int err;
+
+	DRM_DEBUG_ATOMIC("%s handling [PLANE:%d:%s]: src[x/y:%d/%d, w/h:%d/%d] disp[x/y:%d/%d, w/h:%d/%d]",
+			 layer->base.name, plane->base.id, plane->name,
+			 dflow->in_x, dflow->in_y, dflow->in_w, dflow->in_h,
+			 dflow->out_x, dflow->out_y, dflow->out_w, dflow->out_h);
+
+	err = komeda_layer_validate(layer, kplane_st, dflow);
+	if (err)
+		return err;
+
+	err = komeda_compiz_set_input(pipe->compiz, kcrtc_st, dflow);
+
+	return err;
+}
+
+/* build display output data flow, the data path is:
+ * compiz -> improc -> timing_ctrlr
+ */
+int komeda_build_display_data_flow(struct komeda_crtc *kcrtc,
+				   struct komeda_crtc_state *kcrtc_st)
+{
+	struct komeda_pipeline *master = kcrtc->master;
+	struct komeda_data_flow_cfg m_dflow; /* master data flow */
+	int err;
+
+	memset(&m_dflow, 0, sizeof(m_dflow));
+
+	err = komeda_compiz_validate(master->compiz, kcrtc_st, &m_dflow);
+	if (err)
+		return err;
+
+	err = komeda_improc_validate(master->improc, kcrtc_st, &m_dflow);
+	if (err)
+		return err;
+
+	err = komeda_timing_ctrlr_validate(master->ctrlr, kcrtc_st, &m_dflow);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+static void
+komeda_pipeline_unbound_components(struct komeda_pipeline *pipe,
+				   struct komeda_pipeline_state *new)
+{
+	struct drm_atomic_state *drm_st = new->obj.state;
+	struct komeda_pipeline_state *old = priv_to_pipe_st(pipe->obj.state);
+	struct komeda_component_state *c_st;
+	struct komeda_component *c;
+	u32 disabling_comps, id;
+
+	WARN_ON(!old);
+
+	disabling_comps = (~new->active_comps) & old->active_comps;
+
+	/* unbound all disabling component */
+	dp_for_each_set_bit(id, disabling_comps) {
+		c = komeda_pipeline_get_component(pipe, id);
+		c_st = komeda_component_get_state_and_set_user(c,
+				drm_st, NULL, new->crtc);
+		WARN_ON(IS_ERR(c_st));
+	}
+}
+
+/* release unclaimed pipeline resource */
+int komeda_release_unclaimed_resources(struct komeda_pipeline *pipe,
+				       struct komeda_crtc_state *kcrtc_st)
+{
+	struct drm_atomic_state *drm_st = kcrtc_st->base.state;
+	struct komeda_pipeline_state *st;
+
+	/* ignore the pipeline which is not affected */
+	if (!pipe || !has_bit(pipe->id, kcrtc_st->affected_pipes))
+		return 0;
+
+	if (has_bit(pipe->id, kcrtc_st->active_pipes))
+		st = komeda_pipeline_get_new_state(pipe, drm_st);
+	else
+		st = komeda_pipeline_get_state_and_set_crtc(pipe, drm_st, NULL);
+
+	if (WARN_ON(IS_ERR_OR_NULL(st)))
+		return -EINVAL;
+
+	komeda_pipeline_unbound_components(pipe, st);
+
+	return 0;
+}
+
+void komeda_pipeline_disable(struct komeda_pipeline *pipe,
+			     struct drm_atomic_state *old_state)
+{
+	struct komeda_pipeline_state *old;
+	struct komeda_component *c;
+	struct komeda_component_state *c_st;
+	u32 id, disabling_comps = 0;
+
+	old = komeda_pipeline_get_old_state(pipe, old_state);
+
+	disabling_comps = old->active_comps;
+	DRM_DEBUG_ATOMIC("PIPE%d: disabling_comps: 0x%x.\n",
+			 pipe->id, disabling_comps);
+
+	dp_for_each_set_bit(id, disabling_comps) {
+		c = komeda_pipeline_get_component(pipe, id);
+		c_st = priv_to_comp_st(c->obj.state);
+
+		/*
+		 * If we disabled a component then all active_inputs should be
+		 * put in the list of changed_active_inputs, so they get
+		 * re-enabled.
+		 * This usually happens during a modeset when the pipeline is
+		 * first disabled and then the actual state gets committed
+		 * again.
+		 */
+		c_st->changed_active_inputs |= c_st->active_inputs;
+
+		c->funcs->disable(c);
+	}
+}
+
+void komeda_pipeline_update(struct komeda_pipeline *pipe,
+			    struct drm_atomic_state *old_state)
+{
+	struct komeda_pipeline_state *new = priv_to_pipe_st(pipe->obj.state);
+	struct komeda_pipeline_state *old;
+	struct komeda_component *c;
+	u32 id, changed_comps = 0;
+
+	old = komeda_pipeline_get_old_state(pipe, old_state);
+
+	changed_comps = new->active_comps | old->active_comps;
+
+	DRM_DEBUG_ATOMIC("PIPE%d: active_comps: 0x%x, changed: 0x%x.\n",
+			 pipe->id, new->active_comps, changed_comps);
+
+	dp_for_each_set_bit(id, changed_comps) {
+		c = komeda_pipeline_get_component(pipe, id);
+
+		if (new->active_comps & BIT(c->id))
+			c->funcs->update(c, priv_to_comp_st(c->obj.state));
+		else
+			c->funcs->disable(c);
+	}
+}
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_plane.c b/drivers/gpu/drm/arm/display/komeda/komeda_plane.c
index 0a4953a9a909..07ed0cc1bc44 100644
--- a/drivers/gpu/drm/arm/display/komeda/komeda_plane.c
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_plane.c
@@ -7,10 +7,96 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_plane_helper.h>
+#include <drm/drm_print.h>
 #include "komeda_dev.h"
 #include "komeda_kms.h"
 
+static int
+komeda_plane_init_data_flow(struct drm_plane_state *st,
+			    struct komeda_data_flow_cfg *dflow)
+{
+	struct drm_framebuffer *fb = st->fb;
+
+	memset(dflow, 0, sizeof(*dflow));
+
+	dflow->blending_zorder = st->zpos;
+
+	/* if format doesn't have alpha, fix blend mode to PIXEL_NONE */
+	dflow->pixel_blend_mode = fb->format->has_alpha ?
+		st->pixel_blend_mode : DRM_MODE_BLEND_PIXEL_NONE;
+	dflow->layer_alpha = st->alpha >> 8;
+
+	dflow->out_x = st->crtc_x;
+	dflow->out_y = st->crtc_y;
+	dflow->out_w = st->crtc_w;
+	dflow->out_h = st->crtc_h;
+
+	dflow->in_x = st->src_x >> 16;
+	dflow->in_y = st->src_y >> 16;
+	dflow->in_w = st->src_w >> 16;
+	dflow->in_h = st->src_h >> 16;
+
+	return 0;
+}
+
+/**
+ * komeda_plane_atomic_check - build input data flow
+ * @plane: DRM plane
+ * @state: the plane state object
+ *
+ * RETURNS:
+ * Zero for success or -errno
+ */
+static int
+komeda_plane_atomic_check(struct drm_plane *plane,
+			  struct drm_plane_state *state)
+{
+	struct komeda_plane *kplane = to_kplane(plane);
+	struct komeda_plane_state *kplane_st = to_kplane_st(state);
+	struct komeda_layer *layer = kplane->layer;
+	struct drm_crtc_state *crtc_st;
+	struct komeda_crtc *kcrtc;
+	struct komeda_crtc_state *kcrtc_st;
+	struct komeda_data_flow_cfg dflow;
+	int err;
+
+	if (!state->crtc || !state->fb)
+		return 0;
+
+	crtc_st = drm_atomic_get_crtc_state(state->state, state->crtc);
+	if (!crtc_st->enable) {
+		DRM_DEBUG_ATOMIC("Cannot update plane on a disabled CRTC.\n");
+		return -EINVAL;
+	}
+
+	/* crtc is inactive, skip the resource assignment */
+	if (!crtc_st->active)
+		return 0;
+
+	kcrtc = to_kcrtc(state->crtc);
+	kcrtc_st = to_kcrtc_st(crtc_st);
+
+	err = komeda_plane_init_data_flow(state, &dflow);
+	if (err)
+		return err;
+
+	err = komeda_build_layer_data_flow(layer, kplane_st, kcrtc_st, &dflow);
+
+	return err;
+}
+
+/* plane doesn't represent a real HW, so there is no HW update for plane.
+ * komeda handles all the HW update in crtc->atomic_flush
+ */
+static void
+komeda_plane_atomic_update(struct drm_plane *plane,
+			   struct drm_plane_state *old_state)
+{
+}
+
 static const struct drm_plane_helper_funcs komeda_plane_helper_funcs = {
+	.atomic_check	= komeda_plane_atomic_check,
+	.atomic_update	= komeda_plane_atomic_update,
 };
 
 static void komeda_plane_destroy(struct drm_plane *plane)
@@ -20,7 +106,60 @@ static void komeda_plane_destroy(struct drm_plane *plane)
 	kfree(to_kplane(plane));
 }
 
+static void komeda_plane_reset(struct drm_plane *plane)
+{
+	struct komeda_plane_state *state;
+	struct komeda_plane *kplane = to_kplane(plane);
+
+	if (plane->state)
+		__drm_atomic_helper_plane_destroy_state(plane->state);
+
+	kfree(plane->state);
+	plane->state = NULL;
+
+	state = kzalloc(sizeof(*state), GFP_KERNEL);
+	if (state) {
+		state->base.rotation = DRM_MODE_ROTATE_0;
+		state->base.pixel_blend_mode = DRM_MODE_BLEND_PREMULTI;
+		state->base.alpha = DRM_BLEND_ALPHA_OPAQUE;
+		state->base.zpos = kplane->layer->base.id;
+		plane->state = &state->base;
+		plane->state->plane = plane;
+	}
+}
+
+static struct drm_plane_state *
+komeda_plane_atomic_duplicate_state(struct drm_plane *plane)
+{
+	struct komeda_plane_state *new;
+
+	if (WARN_ON(!plane->state))
+		return NULL;
+
+	new = kzalloc(sizeof(*new), GFP_KERNEL);
+	if (!new)
+		return NULL;
+
+	__drm_atomic_helper_plane_duplicate_state(plane, &new->base);
+
+	return &new->base;
+}
+
+static void
+komeda_plane_atomic_destroy_state(struct drm_plane *plane,
+				  struct drm_plane_state *state)
+{
+	__drm_atomic_helper_plane_destroy_state(state);
+	kfree(to_kplane_st(state));
+}
+
 static const struct drm_plane_funcs komeda_plane_funcs = {
+	.update_plane		= drm_atomic_helper_update_plane,
+	.disable_plane		= drm_atomic_helper_disable_plane,
+	.destroy		= komeda_plane_destroy,
+	.reset			= komeda_plane_reset,
+	.atomic_duplicate_state	= komeda_plane_atomic_duplicate_state,
+	.atomic_destroy_state	= komeda_plane_atomic_destroy_state,
 };
 
 /* for komeda, which is pipeline can be share between crtcs */
diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_private_obj.c b/drivers/gpu/drm/arm/display/komeda/komeda_private_obj.c
index f1c9e3fefa86..a54878cbd6e4 100644
--- a/drivers/gpu/drm/arm/display/komeda/komeda_private_obj.c
+++ b/drivers/gpu/drm/arm/display/komeda/komeda_private_obj.c
@@ -7,6 +7,188 @@
 #include "komeda_dev.h"
 #include "komeda_kms.h"
 
+static void
+komeda_component_state_reset(struct komeda_component_state *st)
+{
+	st->binding_user = NULL;
+	st->affected_inputs = st->active_inputs;
+	st->active_inputs = 0;
+	st->changed_active_inputs = 0;
+}
+
+static struct drm_private_state *
+komeda_layer_atomic_duplicate_state(struct drm_private_obj *obj)
+{
+	struct komeda_layer_state *st;
+
+	st = kmemdup(obj->state, sizeof(*st), GFP_KERNEL);
+	if (!st)
+		return NULL;
+
+	komeda_component_state_reset(&st->base);
+	__drm_atomic_helper_private_obj_duplicate_state(obj, &st->base.obj);
+
+	return &st->base.obj;
+}
+
+static void
+komeda_layer_atomic_destroy_state(struct drm_private_obj *obj,
+				  struct drm_private_state *state)
+{
+	struct komeda_layer_state *st = to_layer_st(priv_to_comp_st(state));
+
+	kfree(st);
+}
+
+static const struct drm_private_state_funcs komeda_layer_obj_funcs = {
+	.atomic_duplicate_state	= komeda_layer_atomic_duplicate_state,
+	.atomic_destroy_state	= komeda_layer_atomic_destroy_state,
+};
+
+static int komeda_layer_obj_add(struct komeda_kms_dev *kms,
+				struct komeda_layer *layer)
+{
+	struct komeda_layer_state *st;
+
+	st = kzalloc(sizeof(*st), GFP_KERNEL);
+	if (!st)
+		return -ENOMEM;
+
+	st->base.component = &layer->base;
+	drm_atomic_private_obj_init(&kms->base, &layer->base.obj, &st->base.obj,
+				    &komeda_layer_obj_funcs);
+	return 0;
+}
+
+static struct drm_private_state *
+komeda_compiz_atomic_duplicate_state(struct drm_private_obj *obj)
+{
+	struct komeda_compiz_state *st;
+
+	st = kmemdup(obj->state, sizeof(*st), GFP_KERNEL);
+	if (!st)
+		return NULL;
+
+	komeda_component_state_reset(&st->base);
+	__drm_atomic_helper_private_obj_duplicate_state(obj, &st->base.obj);
+
+	return &st->base.obj;
+}
+
+static void
+komeda_compiz_atomic_destroy_state(struct drm_private_obj *obj,
+				   struct drm_private_state *state)
+{
+	kfree(to_compiz_st(priv_to_comp_st(state)));
+}
+
+static const struct drm_private_state_funcs komeda_compiz_obj_funcs = {
+	.atomic_duplicate_state	= komeda_compiz_atomic_duplicate_state,
+	.atomic_destroy_state	= komeda_compiz_atomic_destroy_state,
+};
+
+static int komeda_compiz_obj_add(struct komeda_kms_dev *kms,
+				 struct komeda_compiz *compiz)
+{
+	struct komeda_compiz_state *st;
+
+	st = kzalloc(sizeof(*st), GFP_KERNEL);
+	if (!st)
+		return -ENOMEM;
+
+	st->base.component = &compiz->base;
+	drm_atomic_private_obj_init(&kms->base, &compiz->base.obj, &st->base.obj,
+				    &komeda_compiz_obj_funcs);
+
+	return 0;
+}
+
+static struct drm_private_state *
+komeda_improc_atomic_duplicate_state(struct drm_private_obj *obj)
+{
+	struct komeda_improc_state *st;
+
+	st = kmemdup(obj->state, sizeof(*st), GFP_KERNEL);
+	if (!st)
+		return NULL;
+
+	komeda_component_state_reset(&st->base);
+	__drm_atomic_helper_private_obj_duplicate_state(obj, &st->base.obj);
+
+	return &st->base.obj;
+}
+
+static void
+komeda_improc_atomic_destroy_state(struct drm_private_obj *obj,
+				   struct drm_private_state *state)
+{
+	kfree(to_improc_st(priv_to_comp_st(state)));
+}
+
+static const struct drm_private_state_funcs komeda_improc_obj_funcs = {
+	.atomic_duplicate_state	= komeda_improc_atomic_duplicate_state,
+	.atomic_destroy_state	= komeda_improc_atomic_destroy_state,
+};
+
+static int komeda_improc_obj_add(struct komeda_kms_dev *kms,
+				 struct komeda_improc *improc)
+{
+	struct komeda_improc_state *st;
+
+	st = kzalloc(sizeof(*st), GFP_KERNEL);
+	if (!st)
+		return -ENOMEM;
+
+	st->base.component = &improc->base;
+	drm_atomic_private_obj_init(&kms->base, &improc->base.obj, &st->base.obj,
+				    &komeda_improc_obj_funcs);
+
+	return 0;
+}
+
+static struct drm_private_state *
+komeda_timing_ctrlr_atomic_duplicate_state(struct drm_private_obj *obj)
+{
+	struct komeda_timing_ctrlr_state *st;
+
+	st = kmemdup(obj->state, sizeof(*st), GFP_KERNEL);
+	if (!st)
+		return NULL;
+
+	komeda_component_state_reset(&st->base);
+	__drm_atomic_helper_private_obj_duplicate_state(obj, &st->base.obj);
+
+	return &st->base.obj;
+}
+
+static void
+komeda_timing_ctrlr_atomic_destroy_state(struct drm_private_obj *obj,
+					 struct drm_private_state *state)
+{
+	kfree(to_ctrlr_st(priv_to_comp_st(state)));
+}
+
+static const struct drm_private_state_funcs komeda_timing_ctrlr_obj_funcs = {
+	.atomic_duplicate_state	= komeda_timing_ctrlr_atomic_duplicate_state,
+	.atomic_destroy_state	= komeda_timing_ctrlr_atomic_destroy_state,
+};
+
+static int komeda_timing_ctrlr_obj_add(struct komeda_kms_dev *kms,
+				       struct komeda_timing_ctrlr *ctrlr)
+{
+	struct komeda_compiz_state *st;
+
+	st = kzalloc(sizeof(*st), GFP_KERNEL);
+	if (!st)
+		return -ENOMEM;
+
+	st->base.component = &ctrlr->base;
+	drm_atomic_private_obj_init(&kms->base, &ctrlr->base.obj, &st->base.obj,
+				    &komeda_timing_ctrlr_obj_funcs);
+
+	return 0;
+}
+
 static struct drm_private_state *
 komeda_pipeline_atomic_duplicate_state(struct drm_private_obj *obj)
 {
@@ -55,7 +237,7 @@ int komeda_kms_add_private_objs(struct komeda_kms_dev *kms,
 				struct komeda_dev *mdev)
 {
 	struct komeda_pipeline *pipe;
-	int i, err;
+	int i, j, err;
 
 	for (i = 0; i < mdev->n_pipelines; i++) {
 		pipe = mdev->pipelines[i];
@@ -64,25 +246,33 @@ int komeda_kms_add_private_objs(struct komeda_kms_dev *kms,
 		if (err)
 			return err;
 
-		/* Add component */
+		for (j = 0; j < pipe->n_layers; j++) {
+			err = komeda_layer_obj_add(kms, pipe->layers[j]);
+			if (err)
+				return err;
+		}
+
+		err = komeda_compiz_obj_add(kms, pipe->compiz);
+		if (err)
+			return err;
+
+		err = komeda_improc_obj_add(kms, pipe->improc);
+		if (err)
+			return err;
+
+		err = komeda_timing_ctrlr_obj_add(kms, pipe->ctrlr);
+		if (err)
+			return err;
 	}
 
 	return 0;
 }
 
-void komeda_kms_cleanup_private_objs(struct komeda_dev *mdev)
+void komeda_kms_cleanup_private_objs(struct komeda_kms_dev *kms)
 {
-	struct komeda_pipeline *pipe;
-	struct komeda_component *c;
-	int i, id;
+	struct drm_mode_config *config = &kms->base.mode_config;
+	struct drm_private_obj *obj, *next;
 
-	for (i = 0; i < mdev->n_pipelines; i++) {
-		pipe = mdev->pipelines[i];
-		dp_for_each_set_bit(id, pipe->avail_comps) {
-			c = komeda_pipeline_get_component(pipe, id);
-
-			drm_atomic_private_obj_fini(&c->obj);
-		}
-		drm_atomic_private_obj_fini(&pipe->obj);
-	}
+	list_for_each_entry_safe(obj, next, &config->privobj_list, head)
+		drm_atomic_private_obj_fini(obj);
 }
diff --git a/drivers/gpu/drm/arm/malidp_drv.c b/drivers/gpu/drm/arm/malidp_drv.c
index ab50ad06e271..21725c9b9f5e 100644
--- a/drivers/gpu/drm/arm/malidp_drv.c
+++ b/drivers/gpu/drm/arm/malidp_drv.c
@@ -264,37 +264,17 @@ static bool
 malidp_verify_afbc_framebuffer_caps(struct drm_device *dev,
 				    const struct drm_mode_fb_cmd2 *mode_cmd)
 {
-	const struct drm_format_info *info;
-
-	if ((mode_cmd->modifier[0] >> 56) != DRM_FORMAT_MOD_VENDOR_ARM) {
-		DRM_DEBUG_KMS("Unknown modifier (not Arm)\n");
-		return false;
-	}
-
-	if (mode_cmd->modifier[0] &
-	    ~DRM_FORMAT_MOD_ARM_AFBC(AFBC_MOD_VALID_BITS)) {
-		DRM_DEBUG_KMS("Unsupported modifiers\n");
-		return false;
-	}
-
-	info = drm_get_format_info(dev, mode_cmd);
-	if (!info) {
-		DRM_DEBUG_KMS("Unable to get the format information\n");
+	if (malidp_format_mod_supported(dev, mode_cmd->pixel_format,
+					mode_cmd->modifier[0]) == false)
 		return false;
-	}
-
-	if (info->num_planes != 1) {
-		DRM_DEBUG_KMS("AFBC buffers expect one plane\n");
-		return false;
-	}
 
 	if (mode_cmd->offsets[0] != 0) {
 		DRM_DEBUG_KMS("AFBC buffers' plane offset should be 0\n");
 		return false;
 	}
 
-	switch (mode_cmd->modifier[0] & AFBC_FORMAT_MOD_BLOCK_SIZE_MASK) {
-	case AFBC_FORMAT_MOD_BLOCK_SIZE_16x16:
+	switch (mode_cmd->modifier[0] & AFBC_SIZE_MASK) {
+	case AFBC_SIZE_16X16:
 		if ((mode_cmd->width % 16) || (mode_cmd->height % 16)) {
 			DRM_DEBUG_KMS("AFBC buffers must be aligned to 16 pixels\n");
 			return false;
@@ -318,9 +298,10 @@ malidp_verify_afbc_framebuffer_size(struct drm_device *dev,
 	struct drm_gem_object *objs = NULL;
 	u32 afbc_superblock_size = 0, afbc_superblock_height = 0;
 	u32 afbc_superblock_width = 0, afbc_size = 0;
+	int bpp = 0;
 
-	switch (mode_cmd->modifier[0] & AFBC_FORMAT_MOD_BLOCK_SIZE_MASK) {
-	case AFBC_FORMAT_MOD_BLOCK_SIZE_16x16:
+	switch (mode_cmd->modifier[0] & AFBC_SIZE_MASK) {
+	case AFBC_SIZE_16X16:
 		afbc_superblock_height = 16;
 		afbc_superblock_width = 16;
 		break;
@@ -334,15 +315,19 @@ malidp_verify_afbc_framebuffer_size(struct drm_device *dev,
 	n_superblocks = (mode_cmd->width / afbc_superblock_width) *
 		(mode_cmd->height / afbc_superblock_height);
 
-	afbc_superblock_size = info->cpp[0] * afbc_superblock_width *
-		afbc_superblock_height;
+	bpp = malidp_format_get_bpp(info->format);
+
+	afbc_superblock_size = (bpp * afbc_superblock_width * afbc_superblock_height)
+				/ BITS_PER_BYTE;
 
 	afbc_size = ALIGN(n_superblocks * AFBC_HEADER_SIZE, AFBC_SUPERBLK_ALIGNMENT);
 	afbc_size += n_superblocks * ALIGN(afbc_superblock_size, AFBC_SUPERBLK_ALIGNMENT);
 
-	if (mode_cmd->width * info->cpp[0] != mode_cmd->pitches[0]) {
-		DRM_DEBUG_KMS("Invalid value of pitch (=%u) should be same as width (=%u) * cpp (=%u)\n",
-			      mode_cmd->pitches[0], mode_cmd->width, info->cpp[0]);
+	if ((mode_cmd->width * bpp) != (mode_cmd->pitches[0] * BITS_PER_BYTE)) {
+		DRM_DEBUG_KMS("Invalid value of (pitch * BITS_PER_BYTE) (=%u) "
+			      "should be same as width (=%u) * bpp (=%u)\n",
+			      (mode_cmd->pitches[0] * BITS_PER_BYTE),
+			      mode_cmd->width, bpp);
 		return false;
 	}
 
@@ -406,6 +391,7 @@ static int malidp_init(struct drm_device *drm)
 	drm->mode_config.max_height = hwdev->max_line_size;
 	drm->mode_config.funcs = &malidp_mode_config_funcs;
 	drm->mode_config.helper_private = &malidp_mode_config_helpers;
+	drm->mode_config.allow_fb_modifiers = true;
 
 	ret = malidp_crtc_init(drm);
 	if (ret)
diff --git a/drivers/gpu/drm/arm/malidp_drv.h b/drivers/gpu/drm/arm/malidp_drv.h
index b76c86f18a56..019a682b2716 100644
--- a/drivers/gpu/drm/arm/malidp_drv.h
+++ b/drivers/gpu/drm/arm/malidp_drv.h
@@ -90,6 +90,12 @@ struct malidp_crtc_state {
 int malidp_de_planes_init(struct drm_device *drm);
 int malidp_crtc_init(struct drm_device *drm);
 
+bool malidp_hw_format_is_linear_only(u32 format);
+bool malidp_hw_format_is_afbc_only(u32 format);
+
+bool malidp_format_mod_supported(struct drm_device *drm,
+				 u32 format, u64 modifier);
+
 #ifdef CONFIG_DEBUG_FS
 void malidp_error(struct malidp_drm *malidp,
 		  struct malidp_error_stats *error_stats, u32 status,
diff --git a/drivers/gpu/drm/arm/malidp_hw.c b/drivers/gpu/drm/arm/malidp_hw.c
index b9bed1138fa3..8df12e9a33bb 100644
--- a/drivers/gpu/drm/arm/malidp_hw.c
+++ b/drivers/gpu/drm/arm/malidp_hw.c
@@ -49,11 +49,19 @@ static const struct malidp_format_id malidp500_de_formats[] = {
 	{ DRM_FORMAT_YUYV, DE_VIDEO1, 13 },
 	{ DRM_FORMAT_NV12, DE_VIDEO1 | SE_MEMWRITE, 14 },
 	{ DRM_FORMAT_YUV420, DE_VIDEO1, 15 },
+	{ DRM_FORMAT_XYUV8888, DE_VIDEO1, 16 },
+	/* These are supported with AFBC only */
+	{ DRM_FORMAT_YUV420_8BIT, DE_VIDEO1, 14 },
+	{ DRM_FORMAT_VUY888, DE_VIDEO1, 16 },
+	{ DRM_FORMAT_VUY101010, DE_VIDEO1, 17 },
+	{ DRM_FORMAT_YUV420_10BIT, DE_VIDEO1, 18 }
 };
 
 #define MALIDP_ID(__group, __format) \
 	((((__group) & 0x7) << 3) | ((__format) & 0x7))
 
+#define AFBC_YUV_422_FORMAT_ID	MALIDP_ID(5, 1)
+
 #define MALIDP_COMMON_FORMATS \
 	/*    fourcc,   layers supporting the format,      internal id   */ \
 	{ DRM_FORMAT_ARGB2101010, DE_VIDEO1 | DE_GRAPHICS1 | DE_VIDEO2 | SE_MEMWRITE, MALIDP_ID(0, 0) }, \
@@ -74,11 +82,25 @@ static const struct malidp_format_id malidp500_de_formats[] = {
 	{ DRM_FORMAT_ABGR1555, DE_VIDEO1 | DE_GRAPHICS1 | DE_VIDEO2, MALIDP_ID(4, 1) }, \
 	{ DRM_FORMAT_RGB565, DE_VIDEO1 | DE_GRAPHICS1 | DE_VIDEO2, MALIDP_ID(4, 2) }, \
 	{ DRM_FORMAT_BGR565, DE_VIDEO1 | DE_GRAPHICS1 | DE_VIDEO2, MALIDP_ID(4, 3) }, \
+	/* This is only supported with linear modifier */	\
+	{ DRM_FORMAT_XYUV8888, DE_VIDEO1 | DE_VIDEO2, MALIDP_ID(5, 0) },\
+	/* This is only supported with AFBC modifier */		\
+	{ DRM_FORMAT_VUY888, DE_VIDEO1 | DE_VIDEO2, MALIDP_ID(5, 0) }, \
 	{ DRM_FORMAT_YUYV, DE_VIDEO1 | DE_VIDEO2, MALIDP_ID(5, 2) },	\
+	/* This is only supported with linear modifier */ \
 	{ DRM_FORMAT_UYVY, DE_VIDEO1 | DE_VIDEO2, MALIDP_ID(5, 3) },	\
 	{ DRM_FORMAT_NV12, DE_VIDEO1 | DE_VIDEO2 | SE_MEMWRITE, MALIDP_ID(5, 6) },	\
+	/* This is only supported with AFBC modifier */ \
+	{ DRM_FORMAT_YUV420_8BIT, DE_VIDEO1 | DE_VIDEO2, MALIDP_ID(5, 6) }, \
 	{ DRM_FORMAT_YUV420, DE_VIDEO1 | DE_VIDEO2, MALIDP_ID(5, 7) }, \
-	{ DRM_FORMAT_X0L2, DE_VIDEO1 | DE_VIDEO2, MALIDP_ID(6, 6)}
+	/* This is only supported with linear modifier */ \
+	{ DRM_FORMAT_XVYU2101010, DE_VIDEO1 | DE_VIDEO2, MALIDP_ID(6, 0)}, \
+	/* This is only supported with AFBC modifier */ \
+	{ DRM_FORMAT_VUY101010, DE_VIDEO1 | DE_VIDEO2, MALIDP_ID(6, 0)}, \
+	{ DRM_FORMAT_X0L2, DE_VIDEO1 | DE_VIDEO2, MALIDP_ID(6, 6)}, \
+	/* This is only supported with AFBC modifier */ \
+	{ DRM_FORMAT_YUV420_10BIT, DE_VIDEO1 | DE_VIDEO2, MALIDP_ID(6, 7)}, \
+	{ DRM_FORMAT_P010, DE_VIDEO1 | DE_VIDEO2, MALIDP_ID(6, 7)}
 
 static const struct malidp_format_id malidp550_de_formats[] = {
 	MALIDP_COMMON_FORMATS,
@@ -94,11 +116,14 @@ static const struct malidp_layer malidp500_layers[] = {
 	 *	yuv2rgb matrix offset, mmu control register offset, rotation_features
 	 */
 	{ DE_VIDEO1, MALIDP500_DE_LV_BASE, MALIDP500_DE_LV_PTR_BASE,
-		MALIDP_DE_LV_STRIDE0, MALIDP500_LV_YUV2RGB, 0, ROTATE_ANY },
+		MALIDP_DE_LV_STRIDE0, MALIDP500_LV_YUV2RGB, 0, ROTATE_ANY,
+		MALIDP500_DE_LV_AD_CTRL },
 	{ DE_GRAPHICS1, MALIDP500_DE_LG1_BASE, MALIDP500_DE_LG1_PTR_BASE,
-		MALIDP_DE_LG_STRIDE, 0, 0, ROTATE_ANY },
+		MALIDP_DE_LG_STRIDE, 0, 0, ROTATE_ANY,
+		MALIDP500_DE_LG1_AD_CTRL },
 	{ DE_GRAPHICS2, MALIDP500_DE_LG2_BASE, MALIDP500_DE_LG2_PTR_BASE,
-		MALIDP_DE_LG_STRIDE, 0, 0, ROTATE_ANY },
+		MALIDP_DE_LG_STRIDE, 0, 0, ROTATE_ANY,
+		MALIDP500_DE_LG2_AD_CTRL },
 };
 
 static const struct malidp_layer malidp550_layers[] = {
@@ -106,13 +131,16 @@ static const struct malidp_layer malidp550_layers[] = {
 	 *	yuv2rgb matrix offset, mmu control register offset, rotation_features
 	 */
 	{ DE_VIDEO1, MALIDP550_DE_LV1_BASE, MALIDP550_DE_LV1_PTR_BASE,
-		MALIDP_DE_LV_STRIDE0, MALIDP550_LV_YUV2RGB, 0, ROTATE_ANY },
+		MALIDP_DE_LV_STRIDE0, MALIDP550_LV_YUV2RGB, 0, ROTATE_ANY,
+		MALIDP550_DE_LV1_AD_CTRL },
 	{ DE_GRAPHICS1, MALIDP550_DE_LG_BASE, MALIDP550_DE_LG_PTR_BASE,
-		MALIDP_DE_LG_STRIDE, 0, 0, ROTATE_ANY },
+		MALIDP_DE_LG_STRIDE, 0, 0, ROTATE_ANY,
+		MALIDP550_DE_LG_AD_CTRL },
 	{ DE_VIDEO2, MALIDP550_DE_LV2_BASE, MALIDP550_DE_LV2_PTR_BASE,
-		MALIDP_DE_LV_STRIDE0, MALIDP550_LV_YUV2RGB, 0, ROTATE_ANY },
+		MALIDP_DE_LV_STRIDE0, MALIDP550_LV_YUV2RGB, 0, ROTATE_ANY,
+		MALIDP550_DE_LV2_AD_CTRL },
 	{ DE_SMART, MALIDP550_DE_LS_BASE, MALIDP550_DE_LS_PTR_BASE,
-		MALIDP550_DE_LS_R1_STRIDE, 0, 0, ROTATE_NONE },
+		MALIDP550_DE_LS_R1_STRIDE, 0, 0, ROTATE_NONE, 0 },
 };
 
 static const struct malidp_layer malidp650_layers[] = {
@@ -122,16 +150,44 @@ static const struct malidp_layer malidp650_layers[] = {
 	 */
 	{ DE_VIDEO1, MALIDP550_DE_LV1_BASE, MALIDP550_DE_LV1_PTR_BASE,
 		MALIDP_DE_LV_STRIDE0, MALIDP550_LV_YUV2RGB,
-		MALIDP650_DE_LV_MMU_CTRL, ROTATE_ANY },
+		MALIDP650_DE_LV_MMU_CTRL, ROTATE_ANY,
+		MALIDP550_DE_LV1_AD_CTRL },
 	{ DE_GRAPHICS1, MALIDP550_DE_LG_BASE, MALIDP550_DE_LG_PTR_BASE,
 		MALIDP_DE_LG_STRIDE, 0, MALIDP650_DE_LG_MMU_CTRL,
-		ROTATE_COMPRESSED },
+		ROTATE_COMPRESSED, MALIDP550_DE_LG_AD_CTRL },
 	{ DE_VIDEO2, MALIDP550_DE_LV2_BASE, MALIDP550_DE_LV2_PTR_BASE,
 		MALIDP_DE_LV_STRIDE0, MALIDP550_LV_YUV2RGB,
-		MALIDP650_DE_LV_MMU_CTRL, ROTATE_ANY },
+		MALIDP650_DE_LV_MMU_CTRL, ROTATE_ANY,
+		MALIDP550_DE_LV2_AD_CTRL },
 	{ DE_SMART, MALIDP550_DE_LS_BASE, MALIDP550_DE_LS_PTR_BASE,
 		MALIDP550_DE_LS_R1_STRIDE, 0, MALIDP650_DE_LS_MMU_CTRL,
-		ROTATE_NONE },
+		ROTATE_NONE, 0 },
+};
+
+const u64 malidp_format_modifiers[] = {
+	/* All RGB formats (except XRGB, RGBX, XBGR, BGRX) */
+	DRM_FORMAT_MOD_ARM_AFBC(AFBC_SIZE_16X16 | AFBC_YTR | AFBC_SPARSE),
+	DRM_FORMAT_MOD_ARM_AFBC(AFBC_SIZE_16X16 | AFBC_YTR),
+
+	/* All RGB formats > 16bpp (except XRGB, RGBX, XBGR, BGRX) */
+	DRM_FORMAT_MOD_ARM_AFBC(AFBC_SIZE_16X16 | AFBC_YTR | AFBC_SPARSE | AFBC_SPLIT),
+
+	/* All 8 or 10 bit YUV 444 formats. */
+	/* In DP550, 10 bit YUV 420 format also supported */
+	DRM_FORMAT_MOD_ARM_AFBC(AFBC_SIZE_16X16 | AFBC_SPARSE | AFBC_SPLIT),
+
+	/* YUV 420, 422 P1 8 bit and YUV 444 8 bit/10 bit formats */
+	DRM_FORMAT_MOD_ARM_AFBC(AFBC_SIZE_16X16 | AFBC_SPARSE),
+	DRM_FORMAT_MOD_ARM_AFBC(AFBC_SIZE_16X16),
+
+	/* YUV 420, 422 P1 8, 10 bit formats */
+	DRM_FORMAT_MOD_ARM_AFBC(AFBC_SIZE_16X16 | AFBC_CBR | AFBC_SPARSE),
+	DRM_FORMAT_MOD_ARM_AFBC(AFBC_SIZE_16X16 | AFBC_CBR),
+
+	/* All formats */
+	DRM_FORMAT_MOD_LINEAR,
+
+	DRM_FORMAT_MOD_INVALID
 };
 
 #define SE_N_SCALING_COEFFS	96
@@ -324,14 +380,39 @@ static void malidp500_modeset(struct malidp_hw_device *hwdev, struct videomode *
 		malidp_hw_clearbits(hwdev, MALIDP_DISP_FUNC_ILACED, MALIDP_DE_DISPLAY_FUNC);
 }
 
-static int malidp500_rotmem_required(struct malidp_hw_device *hwdev, u16 w, u16 h, u32 fmt)
+int malidp_format_get_bpp(u32 fmt)
+{
+	int bpp = drm_format_plane_cpp(fmt, 0) * 8;
+
+	if (bpp == 0) {
+		switch (fmt) {
+		case DRM_FORMAT_VUY101010:
+			bpp = 30;
+		case DRM_FORMAT_YUV420_10BIT:
+			bpp = 15;
+			break;
+		case DRM_FORMAT_YUV420_8BIT:
+			bpp = 12;
+			break;
+		default:
+			bpp = 0;
+		}
+	}
+
+	return bpp;
+}
+
+static int malidp500_rotmem_required(struct malidp_hw_device *hwdev, u16 w,
+				     u16 h, u32 fmt, bool has_modifier)
 {
 	/*
 	 * Each layer needs enough rotation memory to fit 8 lines
 	 * worth of pixel data. Required size is then:
 	 *    size = rotated_width * (bpp / 8) * 8;
 	 */
-	return w * drm_format_plane_cpp(fmt, 0) * 8;
+	int bpp = malidp_format_get_bpp(fmt);
+
+	return w * bpp;
 }
 
 static void malidp500_se_write_pp_coefftab(struct malidp_hw_device *hwdev,
@@ -609,9 +690,9 @@ static void malidp550_modeset(struct malidp_hw_device *hwdev, struct videomode *
 		malidp_hw_clearbits(hwdev, MALIDP_DISP_FUNC_ILACED, MALIDP_DE_DISPLAY_FUNC);
 }
 
-static int malidp550_rotmem_required(struct malidp_hw_device *hwdev, u16 w, u16 h, u32 fmt)
+static int malidpx50_get_bytes_per_column(u32 fmt)
 {
-	u32 bytes_per_col;
+	u32 bytes_per_column;
 
 	switch (fmt) {
 	/* 8 lines at 4 bytes per pixel */
@@ -637,19 +718,77 @@ static int malidp550_rotmem_required(struct malidp_hw_device *hwdev, u16 w, u16
 	case DRM_FORMAT_UYVY:
 	case DRM_FORMAT_YUYV:
 	case DRM_FORMAT_X0L0:
-	case DRM_FORMAT_X0L2:
-		bytes_per_col = 32;
+		bytes_per_column = 32;
 		break;
 	/* 16 lines at 1.5 bytes per pixel */
 	case DRM_FORMAT_NV12:
 	case DRM_FORMAT_YUV420:
-		bytes_per_col = 24;
+	/* 8 lines at 3 bytes per pixel */
+	case DRM_FORMAT_VUY888:
+	/* 16 lines at 12 bits per pixel */
+	case DRM_FORMAT_YUV420_8BIT:
+	/* 8 lines at 3 bytes per pixel */
+	case DRM_FORMAT_P010:
+		bytes_per_column = 24;
+		break;
+	/* 8 lines at 30 bits per pixel */
+	case DRM_FORMAT_VUY101010:
+	/* 16 lines at 15 bits per pixel */
+	case DRM_FORMAT_YUV420_10BIT:
+		bytes_per_column = 30;
 		break;
 	default:
 		return -EINVAL;
 	}
 
-	return w * bytes_per_col;
+	return bytes_per_column;
+}
+
+static int malidp550_rotmem_required(struct malidp_hw_device *hwdev, u16 w,
+				     u16 h, u32 fmt, bool has_modifier)
+{
+	int bytes_per_column = 0;
+
+	switch (fmt) {
+	/* 8 lines at 15 bits per pixel */
+	case DRM_FORMAT_YUV420_10BIT:
+		bytes_per_column = 15;
+		break;
+	/* Uncompressed YUV 420 10 bit single plane cannot be rotated */
+	case DRM_FORMAT_X0L2:
+		if (has_modifier)
+			bytes_per_column = 8;
+		else
+			return -EINVAL;
+		break;
+	default:
+		bytes_per_column = malidpx50_get_bytes_per_column(fmt);
+	}
+
+	if (bytes_per_column == -EINVAL)
+		return bytes_per_column;
+
+	return w * bytes_per_column;
+}
+
+static int malidp650_rotmem_required(struct malidp_hw_device *hwdev, u16 w,
+				     u16 h, u32 fmt, bool has_modifier)
+{
+	int bytes_per_column = 0;
+
+	switch (fmt) {
+	/* 16 lines at 2 bytes per pixel */
+	case DRM_FORMAT_X0L2:
+		bytes_per_column = 32;
+		break;
+	default:
+		bytes_per_column = malidpx50_get_bytes_per_column(fmt);
+	}
+
+	if (bytes_per_column == -EINVAL)
+		return bytes_per_column;
+
+	return w * bytes_per_column;
 }
 
 static int malidp550_se_set_scaling_coeffs(struct malidp_hw_device *hwdev,
@@ -838,7 +977,10 @@ const struct malidp_hw malidp_device[MALIDP_MAX_DEVICES] = {
 			.se_base = MALIDP550_SE_BASE,
 			.dc_base = MALIDP550_DC_BASE,
 			.out_depth_base = MALIDP550_DE_OUTPUT_DEPTH,
-			.features = MALIDP_REGMAP_HAS_CLEARIRQ,
+			.features = MALIDP_REGMAP_HAS_CLEARIRQ |
+				    MALIDP_DEVICE_AFBC_SUPPORT_SPLIT |
+				    MALIDP_DEVICE_AFBC_YUV_420_10_SUPPORT_SPLIT |
+				    MALIDP_DEVICE_AFBC_YUYV_USE_422_P2,
 			.n_layers = ARRAY_SIZE(malidp550_layers),
 			.layers = malidp550_layers,
 			.de_irq_map = {
@@ -884,7 +1026,9 @@ const struct malidp_hw malidp_device[MALIDP_MAX_DEVICES] = {
 			.se_base = MALIDP550_SE_BASE,
 			.dc_base = MALIDP550_DC_BASE,
 			.out_depth_base = MALIDP550_DE_OUTPUT_DEPTH,
-			.features = MALIDP_REGMAP_HAS_CLEARIRQ,
+			.features = MALIDP_REGMAP_HAS_CLEARIRQ |
+				    MALIDP_DEVICE_AFBC_SUPPORT_SPLIT |
+				    MALIDP_DEVICE_AFBC_YUYV_USE_422_P2,
 			.n_layers = ARRAY_SIZE(malidp650_layers),
 			.layers = malidp650_layers,
 			.de_irq_map = {
@@ -923,7 +1067,7 @@ const struct malidp_hw malidp_device[MALIDP_MAX_DEVICES] = {
 		.in_config_mode = malidp550_in_config_mode,
 		.set_config_valid = malidp550_set_config_valid,
 		.modeset = malidp550_modeset,
-		.rotmem_required = malidp550_rotmem_required,
+		.rotmem_required = malidp650_rotmem_required,
 		.se_set_scaling_coeffs = malidp550_se_set_scaling_coeffs,
 		.se_calc_mclk = malidp550_se_calc_mclk,
 		.enable_memwrite = malidp550_enable_memwrite,
@@ -933,19 +1077,72 @@ const struct malidp_hw malidp_device[MALIDP_MAX_DEVICES] = {
 };
 
 u8 malidp_hw_get_format_id(const struct malidp_hw_regmap *map,
-			   u8 layer_id, u32 format)
+			   u8 layer_id, u32 format, bool has_modifier)
 {
 	unsigned int i;
 
 	for (i = 0; i < map->n_pixel_formats; i++) {
 		if (((map->pixel_formats[i].layer & layer_id) == layer_id) &&
-		    (map->pixel_formats[i].format == format))
-			return map->pixel_formats[i].id;
+		    (map->pixel_formats[i].format == format)) {
+			/*
+			 * In some DP550 and DP650, DRM_FORMAT_YUYV + AFBC modifier
+			 * is supported by a different h/w format id than
+			 * DRM_FORMAT_YUYV (only).
+			 */
+			if (format == DRM_FORMAT_YUYV &&
+			    (has_modifier) &&
+			    (map->features & MALIDP_DEVICE_AFBC_YUYV_USE_422_P2))
+				return AFBC_YUV_422_FORMAT_ID;
+			else
+				return map->pixel_formats[i].id;
+		}
 	}
 
 	return MALIDP_INVALID_FORMAT_ID;
 }
 
+bool malidp_hw_format_is_linear_only(u32 format)
+{
+	switch (format) {
+	case DRM_FORMAT_ARGB2101010:
+	case DRM_FORMAT_RGBA1010102:
+	case DRM_FORMAT_BGRA1010102:
+	case DRM_FORMAT_ARGB8888:
+	case DRM_FORMAT_RGBA8888:
+	case DRM_FORMAT_BGRA8888:
+	case DRM_FORMAT_XBGR8888:
+	case DRM_FORMAT_XRGB8888:
+	case DRM_FORMAT_RGBX8888:
+	case DRM_FORMAT_BGRX8888:
+	case DRM_FORMAT_RGB888:
+	case DRM_FORMAT_RGB565:
+	case DRM_FORMAT_ARGB1555:
+	case DRM_FORMAT_RGBA5551:
+	case DRM_FORMAT_BGRA5551:
+	case DRM_FORMAT_UYVY:
+	case DRM_FORMAT_XYUV8888:
+	case DRM_FORMAT_XVYU2101010:
+	case DRM_FORMAT_X0L2:
+	case DRM_FORMAT_X0L0:
+		return true;
+	default:
+		return false;
+	}
+}
+
+bool malidp_hw_format_is_afbc_only(u32 format)
+{
+	switch (format) {
+	case DRM_FORMAT_VUY888:
+	case DRM_FORMAT_VUY101010:
+	case DRM_FORMAT_YUV420_8BIT:
+	case DRM_FORMAT_YUV420_10BIT:
+		return true;
+	default:
+		return false;
+	}
+}
+
 static void malidp_hw_clear_irq(struct malidp_hw_device *hwdev, u8 block, u32 irq)
 {
 	u32 base = malidp_get_block_base(hwdev, block);
diff --git a/drivers/gpu/drm/arm/malidp_hw.h b/drivers/gpu/drm/arm/malidp_hw.h
index 40155e2ea9d9..207c3ce52f1a 100644
--- a/drivers/gpu/drm/arm/malidp_hw.h
+++ b/drivers/gpu/drm/arm/malidp_hw.h
@@ -70,6 +70,8 @@ struct malidp_layer {
 	s16 yuv2rgb_offset;	/* offset to the YUV->RGB matrix entries */
 	u16 mmu_ctrl_offset;    /* offset to the MMU control register */
 	enum rotation_features rot;	/* type of rotation supported */
+	/* address offset for the AFBC decoder registers */
+	u16 afbc_decoder_offset;
 };
 
 enum malidp_scaling_coeff_set {
@@ -93,7 +95,10 @@ struct malidp_se_config {
 };
 
 /* regmap features */
-#define MALIDP_REGMAP_HAS_CLEARIRQ	(1 << 0)
+#define MALIDP_REGMAP_HAS_CLEARIRQ				BIT(0)
+#define MALIDP_DEVICE_AFBC_SUPPORT_SPLIT			BIT(1)
+#define MALIDP_DEVICE_AFBC_YUV_420_10_SUPPORT_SPLIT		BIT(2)
+#define MALIDP_DEVICE_AFBC_YUYV_USE_422_P2			BIT(3)
 
 struct malidp_hw_regmap {
 	/* address offset of the DE register bank */
@@ -179,7 +184,8 @@ struct malidp_hw {
 	 * Calculate the required rotation memory given the active area
 	 * and the buffer format.
 	 */
-	int (*rotmem_required)(struct malidp_hw_device *hwdev, u16 w, u16 h, u32 fmt);
+	int (*rotmem_required)(struct malidp_hw_device *hwdev, u16 w, u16 h,
+			       u32 fmt, bool has_modifier);
 
 	int (*se_set_scaling_coeffs)(struct malidp_hw_device *hwdev,
 				     struct malidp_se_config *se_config,
@@ -319,7 +325,9 @@ int malidp_se_irq_init(struct drm_device *drm, int irq);
 void malidp_se_irq_fini(struct malidp_hw_device *hwdev);
 
 u8 malidp_hw_get_format_id(const struct malidp_hw_regmap *map,
-			   u8 layer_id, u32 format);
+			   u8 layer_id, u32 format, bool has_modifier);
+
+int malidp_format_get_bpp(u32 fmt);
 
 static inline u8 malidp_hw_get_pitch_align(struct malidp_hw_device *hwdev, bool rotated)
 {
@@ -388,9 +396,18 @@ static inline void malidp_se_set_enh_coeffs(struct malidp_hw_device *hwdev)
 
 #define MALIDP_GAMMA_LUT_SIZE		4096
 
-#define AFBC_MOD_VALID_BITS (AFBC_FORMAT_MOD_BLOCK_SIZE_MASK | \
-			AFBC_FORMAT_MOD_YTR | AFBC_FORMAT_MOD_SPLIT | \
-			AFBC_FORMAT_MOD_SPARSE | AFBC_FORMAT_MOD_CBR | \
-			AFBC_FORMAT_MOD_TILED | AFBC_FORMAT_MOD_SC)
+#define AFBC_SIZE_MASK		AFBC_FORMAT_MOD_BLOCK_SIZE_MASK
+#define AFBC_SIZE_16X16		AFBC_FORMAT_MOD_BLOCK_SIZE_16x16
+#define AFBC_YTR		AFBC_FORMAT_MOD_YTR
+#define AFBC_SPARSE		AFBC_FORMAT_MOD_SPARSE
+#define AFBC_CBR		AFBC_FORMAT_MOD_CBR
+#define AFBC_SPLIT		AFBC_FORMAT_MOD_SPLIT
+#define AFBC_TILED		AFBC_FORMAT_MOD_TILED
+#define AFBC_SC			AFBC_FORMAT_MOD_SC
+
+#define AFBC_MOD_VALID_BITS	(AFBC_SIZE_MASK | AFBC_YTR | AFBC_SPLIT | \
+				 AFBC_SPARSE | AFBC_CBR | AFBC_TILED | AFBC_SC)
+
+extern const u64 malidp_format_modifiers[];
 
 #endif  /* __MALIDP_HW_H__ */
diff --git a/drivers/gpu/drm/arm/malidp_mw.c b/drivers/gpu/drm/arm/malidp_mw.c
index 041a64dc7167..5f102bdaf841 100644
--- a/drivers/gpu/drm/arm/malidp_mw.c
+++ b/drivers/gpu/drm/arm/malidp_mw.c
@@ -141,9 +141,14 @@ malidp_mw_encoder_atomic_check(struct drm_encoder *encoder,
 		return -EINVAL;
 	}
 
+	if (fb->modifier) {
+		DRM_DEBUG_KMS("Writeback framebuffer does not support modifiers\n");
+		return -EINVAL;
+	}
+
 	mw_state->format =
 		malidp_hw_get_format_id(&malidp->dev->hw->map, SE_MEMWRITE,
-					fb->format->format);
+					fb->format->format, !!fb->modifier);
 	if (mw_state->format == MALIDP_INVALID_FORMAT_ID) {
 		struct drm_format_name_buf format_name;
 
@@ -252,8 +257,7 @@ void malidp_mw_atomic_commit(struct drm_device *drm,
 				     &mw_state->addrs[0],
 				     mw_state->format);
 
-		drm_writeback_queue_job(mw_conn, conn_state->writeback_job);
-		conn_state->writeback_job = NULL;
+		drm_writeback_queue_job(mw_conn, conn_state);
 		hwdev->hw->enable_memwrite(hwdev, mw_state->addrs,
 					   mw_state->pitches, mw_state->n_planes,
 					   fb->width, fb->height, mw_state->format,
diff --git a/drivers/gpu/drm/arm/malidp_planes.c b/drivers/gpu/drm/arm/malidp_planes.c
index c9a6d3e0cada..d42e0ea9a303 100644
--- a/drivers/gpu/drm/arm/malidp_planes.c
+++ b/drivers/gpu/drm/arm/malidp_planes.c
@@ -52,6 +52,8 @@
 #define MALIDP550_LS_ENABLE		0x01c
 #define MALIDP550_LS_R1_IN_SIZE		0x020
 
+#define MODIFIERS_COUNT_MAX		15
+
 /*
  * This 4-entry look-up-table is used to determine the full 8-bit alpha value
  * for formats with 1- or 2-bit alpha channels.
@@ -145,6 +147,119 @@ static void malidp_plane_atomic_print_state(struct drm_printer *p,
 	drm_printf(p, "\tmmu_prefetch_pgsize=%d\n", ms->mmu_prefetch_pgsize);
 }
 
+bool malidp_format_mod_supported(struct drm_device *drm,
+				 u32 format, u64 modifier)
+{
+	const struct drm_format_info *info;
+	const u64 *modifiers;
+	struct malidp_drm *malidp = drm->dev_private;
+	const struct malidp_hw_regmap *map = &malidp->dev->hw->map;
+
+	if (WARN_ON(modifier == DRM_FORMAT_MOD_INVALID))
+		return false;
+
+	/* Some pixel formats are supported without any modifier */
+	if (modifier == DRM_FORMAT_MOD_LINEAR) {
+		/*
+		 * However these pixel formats need to be supported with
+		 * modifiers only
+		 */
+		return !malidp_hw_format_is_afbc_only(format);
+	}
+
+	if ((modifier >> 56) != DRM_FORMAT_MOD_VENDOR_ARM) {
+		DRM_ERROR("Unknown modifier (not Arm)\n");
+		return false;
+	}
+
+	if (modifier &
+	    ~DRM_FORMAT_MOD_ARM_AFBC(AFBC_MOD_VALID_BITS)) {
+		DRM_DEBUG_KMS("Unsupported modifiers\n");
+		return false;
+	}
+
+	modifiers = malidp_format_modifiers;
+
+	/* SPLIT buffers must use SPARSE layout */
+	if (WARN_ON_ONCE((modifier & AFBC_SPLIT) && !(modifier & AFBC_SPARSE)))
+		return false;
+
+	/* CBR only applies to YUV formats, where YTR should be always 0 */
+	if (WARN_ON_ONCE((modifier & AFBC_CBR) && (modifier & AFBC_YTR)))
+		return false;
+
+	while (*modifiers != DRM_FORMAT_MOD_INVALID) {
+		if (*modifiers == modifier)
+			break;
+
+		modifiers++;
+	}
+
+	/* return false, if the modifier was not found */
+	if (*modifiers == DRM_FORMAT_MOD_INVALID) {
+		DRM_DEBUG_KMS("Unsupported modifier\n");
+		return false;
+	}
+
+	info = drm_format_info(format);
+
+	if (info->num_planes != 1) {
+		DRM_DEBUG_KMS("AFBC buffers expect one plane\n");
+		return false;
+	}
+
+	if (malidp_hw_format_is_linear_only(format) == true) {
+		DRM_DEBUG_KMS("Given format (0x%x) is supported is linear mode only\n",
+			      format);
+		return false;
+	}
+
+	/*
+	 * RGB formats need to provide YTR modifier and YUV formats should not
+	 * provide YTR modifier.
+	 */
+	if (!(info->is_yuv) != !!(modifier & AFBC_FORMAT_MOD_YTR)) {
+		DRM_DEBUG_KMS("AFBC_FORMAT_MOD_YTR is %s for %s formats\n",
+			      info->is_yuv ? "disallowed" : "mandatory",
+			      info->is_yuv ? "YUV" : "RGB");
+		return false;
+	}
+
+	if (modifier & AFBC_SPLIT) {
+		if (!info->is_yuv) {
+			if (drm_format_plane_cpp(format, 0) <= 2) {
+				DRM_DEBUG_KMS("RGB formats <= 16bpp are not supported with SPLIT\n");
+				return false;
+			}
+		}
+
+		if ((drm_format_horz_chroma_subsampling(format) != 1) ||
+		    (drm_format_vert_chroma_subsampling(format) != 1)) {
+			if (!(format == DRM_FORMAT_YUV420_10BIT &&
+			      (map->features & MALIDP_DEVICE_AFBC_YUV_420_10_SUPPORT_SPLIT))) {
+				DRM_DEBUG_KMS("Formats which are sub-sampled should never be split\n");
+				return false;
+			}
+		}
+	}
+
+	if (modifier & AFBC_CBR) {
+		if ((drm_format_horz_chroma_subsampling(format) == 1) ||
+		    (drm_format_vert_chroma_subsampling(format) == 1)) {
+			DRM_DEBUG_KMS("Formats which are not sub-sampled should not have CBR set\n");
+			return false;
+		}
+	}
+
+	return true;
+}
+
+static bool malidp_format_mod_supported_per_plane(struct drm_plane *plane,
+						  u32 format, u64 modifier)
+{
+	return malidp_format_mod_supported(plane->dev, format, modifier);
+}
+
 static const struct drm_plane_funcs malidp_de_plane_funcs = {
 	.update_plane = drm_atomic_helper_update_plane,
 	.disable_plane = drm_atomic_helper_disable_plane,
@@ -153,6 +268,7 @@ static const struct drm_plane_funcs malidp_de_plane_funcs = {
 	.atomic_duplicate_state = malidp_duplicate_plane_state,
 	.atomic_destroy_state = malidp_destroy_plane_state,
 	.atomic_print_state = malidp_plane_atomic_print_state,
+	.format_mod_supported = malidp_format_mod_supported_per_plane,
 };
 
 static int malidp_se_check_scaling(struct malidp_plane *mp,
@@ -406,8 +522,8 @@ static int malidp_de_plane_check(struct drm_plane *plane,
 	fb = state->fb;
 
 	ms->format = malidp_hw_get_format_id(&mp->hwdev->hw->map,
-					     mp->layer->id,
-					     fb->format->format);
+					     mp->layer->id, fb->format->format,
+					     !!fb->modifier);
 	if (ms->format == MALIDP_INVALID_FORMAT_ID)
 		return -EINVAL;
 
@@ -415,8 +531,8 @@ static int malidp_de_plane_check(struct drm_plane *plane,
 	for (i = 0; i < ms->n_planes; i++) {
 		u8 alignment = malidp_hw_get_pitch_align(mp->hwdev, rotated);
 
-		if ((fb->pitches[i] * drm_format_info_block_height(fb->format, i))
-				& (alignment - 1)) {
+		if (((fb->pitches[i] * drm_format_info_block_height(fb->format, i))
+				& (alignment - 1)) && !(fb->modifier)) {
 			DRM_DEBUG_KMS("Invalid pitch %u for plane %d\n",
 				      fb->pitches[i], i);
 			return -EINVAL;
@@ -469,13 +585,20 @@ static int malidp_de_plane_check(struct drm_plane *plane,
 			return -EINVAL;
 	}
 
+	/* SMART layer does not support AFBC */
+	if (mp->layer->id == DE_SMART && fb->modifier) {
+		DRM_ERROR("AFBC framebuffer not supported in SMART layer");
+		return -EINVAL;
+	}
+
 	ms->rotmem_size = 0;
 	if (state->rotation & MALIDP_ROTATED_MASK) {
 		int val;
 
 		val = mp->hwdev->hw->rotmem_required(mp->hwdev, state->crtc_w,
 						     state->crtc_h,
-						     fb->format->format);
+						     fb->format->format,
+						     !!(fb->modifier));
 		if (val < 0)
 			return val;
 
@@ -592,6 +715,83 @@ static void malidp_de_set_mmu_control(struct malidp_plane *mp,
 			mp->layer->base + mp->layer->mmu_ctrl_offset);
 }
 
+static void malidp_set_plane_base_addr(struct drm_framebuffer *fb,
+				       struct malidp_plane *mp,
+				       int plane_index)
+{
+	dma_addr_t paddr;
+	u16 ptr;
+	struct drm_plane *plane = &mp->base;
+	bool afbc = fb->modifier ? true : false;
+
+	ptr = mp->layer->ptr + (plane_index << 4);
+
+	/*
+	 * drm_fb_cma_get_gem_addr() alters the physical base address of the
+	 * framebuffer as per the plane's src_x, src_y co-ordinates (ie to
+	 * take care of source cropping).
+	 * For AFBC, this is not needed as the cropping is handled by _AD_CROP_H
+	 * and _AD_CROP_V registers.
+	 */
+	if (!afbc) {
+		paddr = drm_fb_cma_get_gem_addr(fb, plane->state,
+						plane_index);
+	} else {
+		struct drm_gem_cma_object *obj;
+
+		obj = drm_fb_cma_get_gem_obj(fb, plane_index);
+
+		if (WARN_ON(!obj))
+			return;
+		paddr = obj->paddr;
+	}
+
+	malidp_hw_write(mp->hwdev, lower_32_bits(paddr), ptr);
+	malidp_hw_write(mp->hwdev, upper_32_bits(paddr), ptr + 4);
+}
+
+static void malidp_de_set_plane_afbc(struct drm_plane *plane)
+{
+	struct malidp_plane *mp;
+	u32 src_w, src_h, val = 0, src_x, src_y;
+	struct drm_framebuffer *fb = plane->state->fb;
+
+	mp = to_malidp_plane(plane);
+
+	/* no afbc_decoder_offset means AFBC is not supported on this plane */
+	if (!mp->layer->afbc_decoder_offset)
+		return;
+
+	if (!fb->modifier) {
+		malidp_hw_write(mp->hwdev, 0, mp->layer->afbc_decoder_offset);
+		return;
+	}
+
+	/* convert src values from Q16 fixed point to integer */
+	src_w = plane->state->src_w >> 16;
+	src_h = plane->state->src_h >> 16;
+	src_x = plane->state->src_x >> 16;
+	src_y = plane->state->src_y >> 16;
+
+	val = ((fb->width - (src_x + src_w)) << MALIDP_AD_CROP_RIGHT_OFFSET) |
+		   src_x;
+	malidp_hw_write(mp->hwdev, val,
+			mp->layer->afbc_decoder_offset + MALIDP_AD_CROP_H);
+
+	val = ((fb->height - (src_y + src_h)) << MALIDP_AD_CROP_BOTTOM_OFFSET) |
+		   src_y;
+	malidp_hw_write(mp->hwdev, val,
+			mp->layer->afbc_decoder_offset + MALIDP_AD_CROP_V);
+
+	val = MALIDP_AD_EN;
+	if (fb->modifier & AFBC_FORMAT_MOD_SPLIT)
+		val |= MALIDP_AD_BS;
+	if (fb->modifier & AFBC_FORMAT_MOD_YTR)
+		val |= MALIDP_AD_YTR;
+
+	malidp_hw_write(mp->hwdev, val, mp->layer->afbc_decoder_offset);
+}
+
 static void malidp_de_plane_update(struct drm_plane *plane,
 				   struct drm_plane_state *old_state)
 {
@@ -602,12 +802,23 @@ static void malidp_de_plane_update(struct drm_plane *plane,
 	u8 plane_alpha = state->alpha >> 8;
 	u32 src_w, src_h, dest_w, dest_h, val;
 	int i;
+	struct drm_framebuffer *fb = plane->state->fb;
 
 	mp = to_malidp_plane(plane);
 
-	/* convert src values from Q16 fixed point to integer */
-	src_w = state->src_w >> 16;
-	src_h = state->src_h >> 16;
+	/*
+	 * For AFBC framebuffer, use the framebuffer width and height for
+	 * configuring layer input size register.
+	 */
+	if (fb->modifier) {
+		src_w = fb->width;
+		src_h = fb->height;
+	} else {
+		/* convert src values from Q16 fixed point to integer */
+		src_w = state->src_w >> 16;
+		src_h = state->src_h >> 16;
+	}
+
 	dest_w = state->crtc_w;
 	dest_h = state->crtc_h;
 
@@ -615,15 +826,8 @@ static void malidp_de_plane_update(struct drm_plane *plane,
 	val = (val & ~LAYER_FORMAT_MASK) | ms->format;
 	malidp_hw_write(mp->hwdev, val, mp->layer->base);
 
-	for (i = 0; i < ms->n_planes; i++) {
-		/* calculate the offset for the layer's plane registers */
-		u16 ptr = mp->layer->ptr + (i << 4);
-		dma_addr_t fb_addr = drm_fb_cma_get_gem_addr(state->fb,
-							     state, i);
-
-		malidp_hw_write(mp->hwdev, lower_32_bits(fb_addr), ptr);
-		malidp_hw_write(mp->hwdev, upper_32_bits(fb_addr), ptr + 4);
-	}
+	for (i = 0; i < ms->n_planes; i++)
+		malidp_set_plane_base_addr(fb, mp, i);
 
 	malidp_de_set_mmu_control(mp, ms);
 
@@ -657,6 +861,8 @@ static void malidp_de_plane_update(struct drm_plane *plane,
 				mp->layer->base + MALIDP550_LS_R1_IN_SIZE);
 	}
 
+	malidp_de_set_plane_afbc(plane);
+
 	/* first clear the rotation bits */
 	val = malidp_hw_read(mp->hwdev, mp->layer->base + MALIDP_LAYER_CONTROL);
 	val &= ~LAYER_ROT_MASK;
@@ -733,7 +939,26 @@ int malidp_de_planes_init(struct drm_device *drm)
 				  BIT(DRM_MODE_BLEND_PREMULTI)   |
 				  BIT(DRM_MODE_BLEND_COVERAGE);
 	u32 *formats;
-	int ret, i, j, n;
+	int ret, i = 0, j = 0, n;
+	u64 supported_modifiers[MODIFIERS_COUNT_MAX];
+	const u64 *modifiers;
+
+	modifiers = malidp_format_modifiers;
+
+	if (!(map->features & MALIDP_DEVICE_AFBC_SUPPORT_SPLIT)) {
+		/*
+		 * Since our hardware does not support SPLIT, so build the list
+		 * of supported modifiers excluding SPLIT ones.
+		 */
+		while (*modifiers != DRM_FORMAT_MOD_INVALID) {
+			if (!(*modifiers & AFBC_SPLIT))
+				supported_modifiers[j++] = *modifiers;
+
+			modifiers++;
+		}
+		supported_modifiers[j++] = DRM_FORMAT_MOD_INVALID;
+		modifiers = supported_modifiers;
+	}
 
 	formats = kcalloc(map->n_pixel_formats, sizeof(*formats), GFP_KERNEL);
 	if (!formats) {
@@ -758,9 +983,15 @@ int malidp_de_planes_init(struct drm_device *drm)
 
 		plane_type = (i == 0) ? DRM_PLANE_TYPE_PRIMARY :
 					DRM_PLANE_TYPE_OVERLAY;
+
+		/*
+		 * All the layers except smart layer supports AFBC modifiers.
+		 */
 		ret = drm_universal_plane_init(drm, &plane->base, crtcs,
-					       &malidp_de_plane_funcs, formats,
-					       n, NULL, plane_type, NULL);
+				&malidp_de_plane_funcs, formats, n,
+				(id == DE_SMART) ? NULL : modifiers, plane_type,
+				NULL);
+
 		if (ret < 0)
 			goto cleanup;
 
diff --git a/drivers/gpu/drm/arm/malidp_regs.h b/drivers/gpu/drm/arm/malidp_regs.h
index 7ce3e141464d..a0dd6e1676a8 100644
--- a/drivers/gpu/drm/arm/malidp_regs.h
+++ b/drivers/gpu/drm/arm/malidp_regs.h
@@ -198,10 +198,13 @@
 #define MALIDP500_LV_YUV2RGB		((s16)(-0xB8))
 #define MALIDP500_DE_LV_BASE		0x00100
 #define MALIDP500_DE_LV_PTR_BASE	0x00124
+#define MALIDP500_DE_LV_AD_CTRL		0x00400
 #define MALIDP500_DE_LG1_BASE		0x00200
 #define MALIDP500_DE_LG1_PTR_BASE	0x0021c
+#define MALIDP500_DE_LG1_AD_CTRL	0x0040c
 #define MALIDP500_DE_LG2_BASE		0x00300
 #define MALIDP500_DE_LG2_PTR_BASE	0x0031c
+#define MALIDP500_DE_LG2_AD_CTRL	0x00418
 #define MALIDP500_SE_BASE		0x00c00
 #define MALIDP500_SE_CONTROL		0x00c0c
 #define MALIDP500_SE_MEMWRITE_OUT_SIZE	0x00c2c
@@ -228,10 +231,13 @@
 #define MALIDP550_LV_YUV2RGB		0x00084
 #define MALIDP550_DE_LV1_BASE		0x00100
 #define MALIDP550_DE_LV1_PTR_BASE	0x00124
+#define MALIDP550_DE_LV1_AD_CTRL	0x001B8
 #define MALIDP550_DE_LV2_BASE		0x00200
 #define MALIDP550_DE_LV2_PTR_BASE	0x00224
+#define MALIDP550_DE_LV2_AD_CTRL	0x002B8
 #define MALIDP550_DE_LG_BASE		0x00300
 #define MALIDP550_DE_LG_PTR_BASE	0x0031c
+#define MALIDP550_DE_LG_AD_CTRL		0x00330
 #define MALIDP550_DE_LS_BASE		0x00400
 #define MALIDP550_DE_LS_PTR_BASE	0x0042c
 #define MALIDP550_DE_PERF_BASE		0x00500
@@ -258,6 +264,20 @@
 #define MALIDP_MMU_CTRL_PX_PS(x)	(1 << (8 + (x)))
 #define MALIDP_MMU_CTRL_PP_NUM_REQ(x)	(((x) & 0x7f) << 12)
 
+/* AFBC register offsets relative to MALIDPXXX_DE_LX_AD_CTRL */
+/* The following register offsets are common for DP500, DP550 and DP650 */
+#define MALIDP_AD_CROP_H                0x4
+#define MALIDP_AD_CROP_V                0x8
+#define MALIDP_AD_END_PTR_LOW           0xc
+#define MALIDP_AD_END_PTR_HIGH          0x10
+
+/* AFBC decoder Registers */
+#define MALIDP_AD_EN                    BIT(0)
+#define MALIDP_AD_YTR                   BIT(4)
+#define MALIDP_AD_BS                    BIT(8)
+#define MALIDP_AD_CROP_RIGHT_OFFSET     16
+#define MALIDP_AD_CROP_BOTTOM_OFFSET    16
+
 /*
  * Starting with DP550 the register map blocks has been standardised to the
  * following layout:
diff --git a/drivers/gpu/drm/armada/armada_fbdev.c b/drivers/gpu/drm/armada/armada_fbdev.c
index 8d23700848df..1e7140f005a5 100644
--- a/drivers/gpu/drm/armada/armada_fbdev.c
+++ b/drivers/gpu/drm/armada/armada_fbdev.c
@@ -78,8 +78,6 @@ static int armada_fbdev_create(struct drm_fb_helper *fbh,
 		goto err_fballoc;
 	}
 
-	strlcpy(info->fix.id, "armada-drmfb", sizeof(info->fix.id));
-	info->par = fbh;
 	info->fbops = &armada_fb_ops;
 	info->fix.smem_start = obj->phys_addr;
 	info->fix.smem_len = obj->obj.size;
@@ -87,9 +85,7 @@ static int armada_fbdev_create(struct drm_fb_helper *fbh,
 	info->screen_base = ptr;
 	fbh->fb = &dfb->fb;
 
-	drm_fb_helper_fill_fix(info, dfb->fb.pitches[0],
-			       dfb->fb.format->depth);
-	drm_fb_helper_fill_var(info, fbh, sizes->fb_width, sizes->fb_height);
+	drm_fb_helper_fill_info(info, fbh, sizes);
 
 	DRM_DEBUG_KMS("allocated %dx%d %dbpp fb: 0x%08llx\n",
 		dfb->fb.width, dfb->fb.height, dfb->fb.format->cpp[0] * 8,
diff --git a/drivers/gpu/drm/aspeed/Kconfig b/drivers/gpu/drm/aspeed/Kconfig
new file mode 100644
index 000000000000..cccab520e02f
--- /dev/null
+++ b/drivers/gpu/drm/aspeed/Kconfig
@@ -0,0 +1,14 @@
+config DRM_ASPEED_GFX
+	tristate "ASPEED BMC Display Controller"
+	depends on DRM && OF
+	depends on (COMPILE_TEST || ARCH_ASPEED)
+	select DRM_KMS_HELPER
+	select DRM_KMS_CMA_HELPER
+	select DMA_CMA if HAVE_DMA_CONTIGUOUS
+	select CMA if HAVE_DMA_CONTIGUOUS
+	select MFD_SYSCON
+	help
+	  Chose this option if you have an ASPEED AST2500 SOC Display
+	  Controller (aka GFX).
+
+	  If M is selected this module will be called aspeed_gfx.
diff --git a/drivers/gpu/drm/aspeed/Makefile b/drivers/gpu/drm/aspeed/Makefile
new file mode 100644
index 000000000000..6e194cd790d8
--- /dev/null
+++ b/drivers/gpu/drm/aspeed/Makefile
@@ -0,0 +1,3 @@
+aspeed_gfx-y := aspeed_gfx_drv.o aspeed_gfx_crtc.o aspeed_gfx_out.o
+
+obj-$(CONFIG_DRM_ASPEED_GFX) += aspeed_gfx.o
diff --git a/drivers/gpu/drm/aspeed/aspeed_gfx.h b/drivers/gpu/drm/aspeed/aspeed_gfx.h
new file mode 100644
index 000000000000..a10358bb61ec
--- /dev/null
+++ b/drivers/gpu/drm/aspeed/aspeed_gfx.h
@@ -0,0 +1,104 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright 2018 IBM Corporation */
+
+#include <drm/drm_device.h>
+#include <drm/drm_simple_kms_helper.h>
+
+struct aspeed_gfx {
+	void __iomem			*base;
+	struct clk			*clk;
+	struct reset_control		*rst;
+	struct regmap			*scu;
+
+	struct drm_simple_display_pipe	pipe;
+	struct drm_connector		connector;
+	struct drm_fbdev_cma		*fbdev;
+};
+
+int aspeed_gfx_create_pipe(struct drm_device *drm);
+int aspeed_gfx_create_output(struct drm_device *drm);
+
+#define CRT_CTRL1		0x60 /* CRT Control I */
+#define CRT_CTRL2		0x64 /* CRT Control II */
+#define CRT_STATUS		0x68 /* CRT Status */
+#define CRT_MISC		0x6c /* CRT Misc Setting */
+#define CRT_HORIZ0		0x70 /* CRT Horizontal Total & Display Enable End */
+#define CRT_HORIZ1		0x74 /* CRT Horizontal Retrace Start & End */
+#define CRT_VERT0		0x78 /* CRT Vertical Total & Display Enable End */
+#define CRT_VERT1		0x7C /* CRT Vertical Retrace Start & End */
+#define CRT_ADDR		0x80 /* CRT Display Starting Address */
+#define CRT_OFFSET		0x84 /* CRT Display Offset & Terminal Count */
+#define CRT_THROD		0x88 /* CRT Threshold */
+#define CRT_XSCALE		0x8C /* CRT Scaling-Up Factor */
+#define CRT_CURSOR0		0x90 /* CRT Hardware Cursor X & Y Offset */
+#define CRT_CURSOR1		0x94 /* CRT Hardware Cursor X & Y Position */
+#define CRT_CURSOR2		0x98 /* CRT Hardware Cursor Pattern Address */
+#define CRT_9C			0x9C
+#define CRT_OSD_H		0xA0 /* CRT OSD Horizontal Start/End */
+#define CRT_OSD_V		0xA4 /* CRT OSD Vertical Start/End */
+#define CRT_OSD_ADDR		0xA8 /* CRT OSD Pattern Address */
+#define CRT_OSD_DISP		0xAC /* CRT OSD Offset */
+#define CRT_OSD_THRESH		0xB0 /* CRT OSD Threshold & Alpha */
+#define CRT_B4			0xB4
+#define CRT_STS_V		0xB8 /* CRT Status V */
+#define CRT_SCRATCH		0xBC /* Scratchpad */
+#define CRT_BB0_ADDR		0xD0 /* CRT Display BB0 Starting Address */
+#define CRT_BB1_ADDR		0xD4 /* CRT Display BB1 Starting Address */
+#define CRT_BB_COUNT		0xD8 /* CRT Display BB Terminal Count */
+#define OSD_COLOR1		0xE0 /* OSD Color Palette Index 1 & 0 */
+#define OSD_COLOR2		0xE4 /* OSD Color Palette Index 3 & 2 */
+#define OSD_COLOR3		0xE8 /* OSD Color Palette Index 5 & 4 */
+#define OSD_COLOR4		0xEC /* OSD Color Palette Index 7 & 6 */
+#define OSD_COLOR5		0xF0 /* OSD Color Palette Index 9 & 8 */
+#define OSD_COLOR6		0xF4 /* OSD Color Palette Index 11 & 10 */
+#define OSD_COLOR7		0xF8 /* OSD Color Palette Index 13 & 12 */
+#define OSD_COLOR8		0xFC /* OSD Color Palette Index 15 & 14 */
+
+/* CTRL1 */
+#define CRT_CTRL_EN			BIT(0)
+#define CRT_CTRL_HW_CURSOR_EN		BIT(1)
+#define CRT_CTRL_OSD_EN			BIT(2)
+#define CRT_CTRL_INTERLACED		BIT(3)
+#define CRT_CTRL_COLOR_RGB565		(0 << 7)
+#define CRT_CTRL_COLOR_YUV444		(1 << 7)
+#define CRT_CTRL_COLOR_XRGB8888		(2 << 7)
+#define CRT_CTRL_COLOR_RGB888		(3 << 7)
+#define CRT_CTRL_COLOR_YUV444_2RGB	(5 << 7)
+#define CRT_CTRL_COLOR_YUV422		(7 << 7)
+#define CRT_CTRL_COLOR_MASK		GENMASK(9, 7)
+#define CRT_CTRL_HSYNC_NEGATIVE		BIT(16)
+#define CRT_CTRL_VSYNC_NEGATIVE		BIT(17)
+#define CRT_CTRL_VERTICAL_INTR_EN	BIT(30)
+#define CRT_CTRL_VERTICAL_INTR_STS	BIT(31)
+
+/* CTRL2 */
+#define CRT_CTRL_DAC_EN			BIT(0)
+#define CRT_CTRL_VBLANK_LINE(x)		(((x) << 20) & CRT_CTRL_VBLANK_LINE_MASK)
+#define CRT_CTRL_VBLANK_LINE_MASK	GENMASK(20, 31)
+
+/* CRT_HORIZ0 */
+#define CRT_H_TOTAL(x)			(x)
+#define CRT_H_DE(x)			((x) << 16)
+
+/* CRT_HORIZ1 */
+#define CRT_H_RS_START(x)		(x)
+#define CRT_H_RS_END(x)			((x) << 16)
+
+/* CRT_VIRT0 */
+#define CRT_V_TOTAL(x)			(x)
+#define CRT_V_DE(x)			((x) << 16)
+
+/* CRT_VIRT1 */
+#define CRT_V_RS_START(x)		(x)
+#define CRT_V_RS_END(x)			((x) << 16)
+
+/* CRT_OFFSET */
+#define CRT_DISP_OFFSET(x)		(x)
+#define CRT_TERM_COUNT(x)		((x) << 16)
+
+/* CRT_THROD */
+#define CRT_THROD_LOW(x)		(x)
+#define CRT_THROD_HIGH(x)		((x) << 8)
+
+/* Default Threshold Seting */
+#define G5_CRT_THROD_VAL	(CRT_THROD_LOW(0x24) | CRT_THROD_HIGH(0x3C))
diff --git a/drivers/gpu/drm/aspeed/aspeed_gfx_crtc.c b/drivers/gpu/drm/aspeed/aspeed_gfx_crtc.c
new file mode 100644
index 000000000000..15db9e426ec4
--- /dev/null
+++ b/drivers/gpu/drm/aspeed/aspeed_gfx_crtc.c
@@ -0,0 +1,241 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright 2018 IBM Corporation
+
+#include <linux/clk.h>
+#include <linux/reset.h>
+#include <linux/regmap.h>
+
+#include <drm/drm_crtc_helper.h>
+#include <drm/drm_device.h>
+#include <drm/drm_fb_cma_helper.h>
+#include <drm/drm_fourcc.h>
+#include <drm/drm_gem_cma_helper.h>
+#include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_panel.h>
+#include <drm/drm_simple_kms_helper.h>
+#include <drm/drm_vblank.h>
+
+#include "aspeed_gfx.h"
+
+static struct aspeed_gfx *
+drm_pipe_to_aspeed_gfx(struct drm_simple_display_pipe *pipe)
+{
+	return container_of(pipe, struct aspeed_gfx, pipe);
+}
+
+static int aspeed_gfx_set_pixel_fmt(struct aspeed_gfx *priv, u32 *bpp)
+{
+	struct drm_crtc *crtc = &priv->pipe.crtc;
+	struct drm_device *drm = crtc->dev;
+	const u32 format = crtc->primary->state->fb->format->format;
+	u32 ctrl1;
+
+	ctrl1 = readl(priv->base + CRT_CTRL1);
+	ctrl1 &= ~CRT_CTRL_COLOR_MASK;
+
+	switch (format) {
+	case DRM_FORMAT_RGB565:
+		dev_dbg(drm->dev, "Setting up RGB565 mode\n");
+		ctrl1 |= CRT_CTRL_COLOR_RGB565;
+		*bpp = 16;
+		break;
+	case DRM_FORMAT_XRGB8888:
+		dev_dbg(drm->dev, "Setting up XRGB8888 mode\n");
+		ctrl1 |= CRT_CTRL_COLOR_XRGB8888;
+		*bpp = 32;
+		break;
+	default:
+		dev_err(drm->dev, "Unhandled pixel format %08x\n", format);
+		return -EINVAL;
+	}
+
+	writel(ctrl1, priv->base + CRT_CTRL1);
+
+	return 0;
+}
+
+static void aspeed_gfx_enable_controller(struct aspeed_gfx *priv)
+{
+	u32 ctrl1 = readl(priv->base + CRT_CTRL1);
+	u32 ctrl2 = readl(priv->base + CRT_CTRL2);
+
+	/* SCU2C: set DAC source for display output to Graphics CRT (GFX) */
+	regmap_update_bits(priv->scu, 0x2c, BIT(16), BIT(16));
+
+	writel(ctrl1 | CRT_CTRL_EN, priv->base + CRT_CTRL1);
+	writel(ctrl2 | CRT_CTRL_DAC_EN, priv->base + CRT_CTRL2);
+}
+
+static void aspeed_gfx_disable_controller(struct aspeed_gfx *priv)
+{
+	u32 ctrl1 = readl(priv->base + CRT_CTRL1);
+	u32 ctrl2 = readl(priv->base + CRT_CTRL2);
+
+	writel(ctrl1 & ~CRT_CTRL_EN, priv->base + CRT_CTRL1);
+	writel(ctrl2 & ~CRT_CTRL_DAC_EN, priv->base + CRT_CTRL2);
+
+	regmap_update_bits(priv->scu, 0x2c, BIT(16), 0);
+}
+
+static void aspeed_gfx_crtc_mode_set_nofb(struct aspeed_gfx *priv)
+{
+	struct drm_display_mode *m = &priv->pipe.crtc.state->adjusted_mode;
+	u32 ctrl1, d_offset, t_count, bpp;
+	int err;
+
+	err = aspeed_gfx_set_pixel_fmt(priv, &bpp);
+	if (err)
+		return;
+
+#if 0
+	/* TODO: we have only been able to test with the 40MHz USB clock. The
+	 * clock is fixed, so we cannot adjust it here. */
+	clk_set_rate(priv->pixel_clk, m->crtc_clock * 1000);
+#endif
+
+	ctrl1 = readl(priv->base + CRT_CTRL1);
+	ctrl1 &= ~(CRT_CTRL_INTERLACED |
+			CRT_CTRL_HSYNC_NEGATIVE |
+			CRT_CTRL_VSYNC_NEGATIVE);
+
+	if (m->flags & DRM_MODE_FLAG_INTERLACE)
+		ctrl1 |= CRT_CTRL_INTERLACED;
+
+	if (!(m->flags & DRM_MODE_FLAG_PHSYNC))
+		ctrl1 |= CRT_CTRL_HSYNC_NEGATIVE;
+
+	if (!(m->flags & DRM_MODE_FLAG_PVSYNC))
+		ctrl1 |= CRT_CTRL_VSYNC_NEGATIVE;
+
+	writel(ctrl1, priv->base + CRT_CTRL1);
+
+	/* Horizontal timing */
+	writel(CRT_H_TOTAL(m->htotal - 1) | CRT_H_DE(m->hdisplay - 1),
+			priv->base + CRT_HORIZ0);
+	writel(CRT_H_RS_START(m->hsync_start - 1) | CRT_H_RS_END(m->hsync_end),
+			priv->base + CRT_HORIZ1);
+
+
+	/* Vertical timing */
+	writel(CRT_V_TOTAL(m->vtotal - 1) | CRT_V_DE(m->vdisplay - 1),
+			priv->base + CRT_VERT0);
+	writel(CRT_V_RS_START(m->vsync_start) | CRT_V_RS_END(m->vsync_end),
+			priv->base + CRT_VERT1);
+
+	/*
+	 * Display Offset: address difference between consecutive scan lines
+	 * Terminal Count: memory size of one scan line
+	 */
+	d_offset = m->hdisplay * bpp / 8;
+	t_count = (m->hdisplay * bpp + 127) / 128;
+	writel(CRT_DISP_OFFSET(d_offset) | CRT_TERM_COUNT(t_count),
+			priv->base + CRT_OFFSET);
+
+	/*
+	 * Threshold: FIFO thresholds of refill and stop (16 byte chunks
+	 * per line, rounded up)
+	 */
+	writel(G5_CRT_THROD_VAL, priv->base + CRT_THROD);
+}
+
+static void aspeed_gfx_pipe_enable(struct drm_simple_display_pipe *pipe,
+			      struct drm_crtc_state *crtc_state,
+			      struct drm_plane_state *plane_state)
+{
+	struct aspeed_gfx *priv = drm_pipe_to_aspeed_gfx(pipe);
+	struct drm_crtc *crtc = &pipe->crtc;
+
+	aspeed_gfx_crtc_mode_set_nofb(priv);
+	aspeed_gfx_enable_controller(priv);
+	drm_crtc_vblank_on(crtc);
+}
+
+static void aspeed_gfx_pipe_disable(struct drm_simple_display_pipe *pipe)
+{
+	struct aspeed_gfx *priv = drm_pipe_to_aspeed_gfx(pipe);
+	struct drm_crtc *crtc = &pipe->crtc;
+
+	drm_crtc_vblank_off(crtc);
+	aspeed_gfx_disable_controller(priv);
+}
+
+static void aspeed_gfx_pipe_update(struct drm_simple_display_pipe *pipe,
+				   struct drm_plane_state *plane_state)
+{
+	struct aspeed_gfx *priv = drm_pipe_to_aspeed_gfx(pipe);
+	struct drm_crtc *crtc = &pipe->crtc;
+	struct drm_framebuffer *fb = pipe->plane.state->fb;
+	struct drm_pending_vblank_event *event;
+	struct drm_gem_cma_object *gem;
+
+	spin_lock_irq(&crtc->dev->event_lock);
+	event = crtc->state->event;
+	if (event) {
+		crtc->state->event = NULL;
+
+		if (drm_crtc_vblank_get(crtc) == 0)
+			drm_crtc_arm_vblank_event(crtc, event);
+		else
+			drm_crtc_send_vblank_event(crtc, event);
+	}
+	spin_unlock_irq(&crtc->dev->event_lock);
+
+	if (!fb)
+		return;
+
+	gem = drm_fb_cma_get_gem_obj(fb, 0);
+	if (!gem)
+		return;
+	writel(gem->paddr, priv->base + CRT_ADDR);
+}
+
+static int aspeed_gfx_enable_vblank(struct drm_simple_display_pipe *pipe)
+{
+	struct aspeed_gfx *priv = drm_pipe_to_aspeed_gfx(pipe);
+	u32 reg = readl(priv->base + CRT_CTRL1);
+
+	/* Clear pending VBLANK IRQ */
+	writel(reg | CRT_CTRL_VERTICAL_INTR_STS, priv->base + CRT_CTRL1);
+
+	reg |= CRT_CTRL_VERTICAL_INTR_EN;
+	writel(reg, priv->base + CRT_CTRL1);
+
+	return 0;
+}
+
+static void aspeed_gfx_disable_vblank(struct drm_simple_display_pipe *pipe)
+{
+	struct aspeed_gfx *priv = drm_pipe_to_aspeed_gfx(pipe);
+	u32 reg = readl(priv->base + CRT_CTRL1);
+
+	reg &= ~CRT_CTRL_VERTICAL_INTR_EN;
+	writel(reg, priv->base + CRT_CTRL1);
+
+	/* Clear pending VBLANK IRQ */
+	writel(reg | CRT_CTRL_VERTICAL_INTR_STS, priv->base + CRT_CTRL1);
+}
+
+static struct drm_simple_display_pipe_funcs aspeed_gfx_funcs = {
+	.enable		= aspeed_gfx_pipe_enable,
+	.disable	= aspeed_gfx_pipe_disable,
+	.update		= aspeed_gfx_pipe_update,
+	.prepare_fb	= drm_gem_fb_simple_display_pipe_prepare_fb,
+	.enable_vblank	= aspeed_gfx_enable_vblank,
+	.disable_vblank	= aspeed_gfx_disable_vblank,
+};
+
+static const uint32_t aspeed_gfx_formats[] = {
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_RGB565,
+};
+
+int aspeed_gfx_create_pipe(struct drm_device *drm)
+{
+	struct aspeed_gfx *priv = drm->dev_private;
+
+	return drm_simple_display_pipe_init(drm, &priv->pipe, &aspeed_gfx_funcs,
+					    aspeed_gfx_formats,
+					    ARRAY_SIZE(aspeed_gfx_formats),
+					    NULL,
+					    &priv->connector);
+}
diff --git a/drivers/gpu/drm/aspeed/aspeed_gfx_drv.c b/drivers/gpu/drm/aspeed/aspeed_gfx_drv.c
new file mode 100644
index 000000000000..eeb22eccd1fc
--- /dev/null
+++ b/drivers/gpu/drm/aspeed/aspeed_gfx_drv.c
@@ -0,0 +1,269 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright 2018 IBM Corporation
+
+#include <linux/clk.h>
+#include <linux/dma-mapping.h>
+#include <linux/irq.h>
+#include <linux/mfd/syscon.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_reserved_mem.h>
+#include <linux/platform_device.h>
+#include <linux/regmap.h>
+#include <linux/reset.h>
+
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_crtc_helper.h>
+#include <drm/drm_device.h>
+#include <drm/drm_fb_cma_helper.h>
+#include <drm/drm_fb_helper.h>
+#include <drm/drm_gem_cma_helper.h>
+#include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_probe_helper.h>
+#include <drm/drm_simple_kms_helper.h>
+#include <drm/drm_vblank.h>
+#include <drm/drm_drv.h>
+
+#include "aspeed_gfx.h"
+
+/**
+ * DOC: ASPEED GFX Driver
+ *
+ * This driver is for the ASPEED BMC SoC's 'GFX' display hardware, also called
+ * the 'SOC Display Controller' in the datasheet. This driver runs on the ARM
+ * based BMC systems, unlike the ast driver which runs on a host CPU and is for
+ * a PCIe graphics device.
+ *
+ * The AST2500 supports a total of 3 output paths:
+ *
+ *   1. VGA output, the output target can choose either or both to the DAC
+ *   or DVO interface.
+ *
+ *   2. Graphics CRT output, the output target can choose either or both to
+ *   the DAC or DVO interface.
+ *
+ *   3. Video input from DVO, the video input can be used for video engine
+ *   capture or DAC display output.
+ *
+ * Output options are selected in SCU2C.
+ *
+ * The "VGA mode" device is the PCI attached controller. The "Graphics CRT"
+ * is the ARM's internal display controller.
+ *
+ * The driver only supports a simple configuration consisting of a 40MHz
+ * pixel clock, fixed by hardware limitations, and the VGA output path.
+ *
+ * The driver was written with the 'AST2500 Software Programming Guide' v17,
+ * which is available under NDA from ASPEED.
+ */
+
+static const struct drm_mode_config_funcs aspeed_gfx_mode_config_funcs = {
+	.fb_create		= drm_gem_fb_create,
+	.atomic_check		= drm_atomic_helper_check,
+	.atomic_commit		= drm_atomic_helper_commit,
+};
+
+static void aspeed_gfx_setup_mode_config(struct drm_device *drm)
+{
+	drm_mode_config_init(drm);
+
+	drm->mode_config.min_width = 0;
+	drm->mode_config.min_height = 0;
+	drm->mode_config.max_width = 800;
+	drm->mode_config.max_height = 600;
+	drm->mode_config.funcs = &aspeed_gfx_mode_config_funcs;
+}
+
+static irqreturn_t aspeed_gfx_irq_handler(int irq, void *data)
+{
+	struct drm_device *drm = data;
+	struct aspeed_gfx *priv = drm->dev_private;
+	u32 reg;
+
+	reg = readl(priv->base + CRT_CTRL1);
+
+	if (reg & CRT_CTRL_VERTICAL_INTR_STS) {
+		drm_crtc_handle_vblank(&priv->pipe.crtc);
+		writel(reg, priv->base + CRT_CTRL1);
+		return IRQ_HANDLED;
+	}
+
+	return IRQ_NONE;
+}
+
+
+
+static int aspeed_gfx_load(struct drm_device *drm)
+{
+	struct platform_device *pdev = to_platform_device(drm->dev);
+	struct aspeed_gfx *priv;
+	struct resource *res;
+	int ret;
+
+	priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+	drm->dev_private = priv;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	priv->base = devm_ioremap_resource(drm->dev, res);
+	if (IS_ERR(priv->base))
+		return PTR_ERR(priv->base);
+
+	priv->scu = syscon_regmap_lookup_by_compatible("aspeed,ast2500-scu");
+	if (IS_ERR(priv->scu)) {
+		dev_err(&pdev->dev, "failed to find SCU regmap\n");
+		return PTR_ERR(priv->scu);
+	}
+
+	ret = of_reserved_mem_device_init(drm->dev);
+	if (ret) {
+		dev_err(&pdev->dev,
+			"failed to initialize reserved mem: %d\n", ret);
+		return ret;
+	}
+
+	ret = dma_set_mask_and_coherent(drm->dev, DMA_BIT_MASK(32));
+	if (ret) {
+		dev_err(&pdev->dev, "failed to set DMA mask: %d\n", ret);
+		return ret;
+	}
+
+	priv->rst = devm_reset_control_get_exclusive(&pdev->dev, NULL);
+	if (IS_ERR(priv->rst)) {
+		dev_err(&pdev->dev,
+			"missing or invalid reset controller device tree entry");
+		return PTR_ERR(priv->rst);
+	}
+	reset_control_deassert(priv->rst);
+
+	priv->clk = devm_clk_get(drm->dev, NULL);
+	if (IS_ERR(priv->clk)) {
+		dev_err(&pdev->dev,
+			"missing or invalid clk device tree entry");
+		return PTR_ERR(priv->clk);
+	}
+	clk_prepare_enable(priv->clk);
+
+	/* Sanitize control registers */
+	writel(0, priv->base + CRT_CTRL1);
+	writel(0, priv->base + CRT_CTRL2);
+
+	aspeed_gfx_setup_mode_config(drm);
+
+	ret = drm_vblank_init(drm, 1);
+	if (ret < 0) {
+		dev_err(drm->dev, "Failed to initialise vblank\n");
+		return ret;
+	}
+
+	ret = aspeed_gfx_create_output(drm);
+	if (ret < 0) {
+		dev_err(drm->dev, "Failed to create outputs\n");
+		return ret;
+	}
+
+	ret = aspeed_gfx_create_pipe(drm);
+	if (ret < 0) {
+		dev_err(drm->dev, "Cannot setup simple display pipe\n");
+		return ret;
+	}
+
+	ret = devm_request_irq(drm->dev, platform_get_irq(pdev, 0),
+			       aspeed_gfx_irq_handler, 0, "aspeed gfx", drm);
+	if (ret < 0) {
+		dev_err(drm->dev, "Failed to install IRQ handler\n");
+		return ret;
+	}
+
+	drm_mode_config_reset(drm);
+
+	drm_fbdev_generic_setup(drm, 32);
+
+	return 0;
+}
+
+static void aspeed_gfx_unload(struct drm_device *drm)
+{
+	drm_kms_helper_poll_fini(drm);
+	drm_mode_config_cleanup(drm);
+
+	drm->dev_private = NULL;
+}
+
+DEFINE_DRM_GEM_CMA_FOPS(fops);
+
+static struct drm_driver aspeed_gfx_driver = {
+	.driver_features        = DRIVER_GEM | DRIVER_MODESET |
+				DRIVER_PRIME | DRIVER_ATOMIC,
+	.gem_create_object	= drm_cma_gem_create_object_default_funcs,
+	.dumb_create		= drm_gem_cma_dumb_create,
+	.prime_handle_to_fd	= drm_gem_prime_handle_to_fd,
+	.prime_fd_to_handle	= drm_gem_prime_fd_to_handle,
+	.gem_prime_import_sg_table = drm_gem_cma_prime_import_sg_table,
+	.gem_prime_mmap		= drm_gem_prime_mmap,
+	.fops = &fops,
+	.name = "aspeed-gfx-drm",
+	.desc = "ASPEED GFX DRM",
+	.date = "20180319",
+	.major = 1,
+	.minor = 0,
+};
+
+static const struct of_device_id aspeed_gfx_match[] = {
+	{ .compatible = "aspeed,ast2500-gfx" },
+	{ }
+};
+
+static int aspeed_gfx_probe(struct platform_device *pdev)
+{
+	struct drm_device *drm;
+	int ret;
+
+	drm = drm_dev_alloc(&aspeed_gfx_driver, &pdev->dev);
+	if (IS_ERR(drm))
+		return PTR_ERR(drm);
+
+	ret = aspeed_gfx_load(drm);
+	if (ret)
+		goto err_free;
+
+	ret = drm_dev_register(drm, 0);
+	if (ret)
+		goto err_unload;
+
+	return 0;
+
+err_unload:
+	aspeed_gfx_unload(drm);
+err_free:
+	drm_dev_put(drm);
+
+	return ret;
+}
+
+static int aspeed_gfx_remove(struct platform_device *pdev)
+{
+	struct drm_device *drm = platform_get_drvdata(pdev);
+
+	drm_dev_unregister(drm);
+	aspeed_gfx_unload(drm);
+	drm_dev_put(drm);
+
+	return 0;
+}
+
+static struct platform_driver aspeed_gfx_platform_driver = {
+	.probe		= aspeed_gfx_probe,
+	.remove		= aspeed_gfx_remove,
+	.driver = {
+		.name = "aspeed_gfx",
+		.of_match_table = aspeed_gfx_match,
+	},
+};
+
+module_platform_driver(aspeed_gfx_platform_driver);
+
+MODULE_AUTHOR("Joel Stanley <joel@jms.id.au>");
+MODULE_DESCRIPTION("ASPEED BMC DRM/KMS driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/aspeed/aspeed_gfx_out.c b/drivers/gpu/drm/aspeed/aspeed_gfx_out.c
new file mode 100644
index 000000000000..67ee5fa10055
--- /dev/null
+++ b/drivers/gpu/drm/aspeed/aspeed_gfx_out.c
@@ -0,0 +1,42 @@
+// SPDX-License-Identifier: GPL-2.0+
+// Copyright 2018 IBM Corporation
+
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_connector.h>
+#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
+
+#include "aspeed_gfx.h"
+
+static int aspeed_gfx_get_modes(struct drm_connector *connector)
+{
+	return drm_add_modes_noedid(connector, 800, 600);
+}
+
+static const struct
+drm_connector_helper_funcs aspeed_gfx_connector_helper_funcs = {
+	.get_modes = aspeed_gfx_get_modes,
+};
+
+static const struct drm_connector_funcs aspeed_gfx_connector_funcs = {
+	.fill_modes		= drm_helper_probe_single_connector_modes,
+	.destroy		= drm_connector_cleanup,
+	.reset			= drm_atomic_helper_connector_reset,
+	.atomic_duplicate_state	= drm_atomic_helper_connector_duplicate_state,
+	.atomic_destroy_state	= drm_atomic_helper_connector_destroy_state,
+};
+
+int aspeed_gfx_create_output(struct drm_device *drm)
+{
+	struct aspeed_gfx *priv = drm->dev_private;
+	int ret;
+
+	priv->connector.dpms = DRM_MODE_DPMS_OFF;
+	priv->connector.polled = 0;
+	drm_connector_helper_add(&priv->connector,
+				 &aspeed_gfx_connector_helper_funcs);
+	ret = drm_connector_init(drm, &priv->connector,
+				 &aspeed_gfx_connector_funcs,
+				 DRM_MODE_CONNECTOR_Unknown);
+	return ret;
+}
diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h
index bfc65040dfcb..1cf0c75e411d 100644
--- a/drivers/gpu/drm/ast/ast_drv.h
+++ b/drivers/gpu/drm/ast/ast_drv.h
@@ -259,7 +259,7 @@ struct ast_framebuffer {
 };
 
 struct ast_fbdev {
-	struct drm_fb_helper helper;
+	struct drm_fb_helper helper; /* must be first */
 	struct ast_framebuffer afb;
 	void *sysram;
 	int size;
@@ -353,8 +353,6 @@ extern int ast_dumb_mmap_offset(struct drm_file *file,
 				uint32_t handle,
 				uint64_t *offset);
 
-#define DRM_FILE_PAGE_OFFSET (0x100000000ULL >> PAGE_SHIFT)
-
 int ast_mm_init(struct ast_private *ast);
 void ast_mm_fini(struct ast_private *ast);
 
diff --git a/drivers/gpu/drm/ast/ast_fb.c b/drivers/gpu/drm/ast/ast_fb.c
index 2c9f8dd9733a..e718d0f60d6b 100644
--- a/drivers/gpu/drm/ast/ast_fb.c
+++ b/drivers/gpu/drm/ast/ast_fb.c
@@ -217,8 +217,6 @@ static int astfb_create(struct drm_fb_helper *helper,
 		ret = PTR_ERR(info);
 		goto out;
 	}
-	info->par = afbdev;
-
 	ret = ast_framebuffer_init(dev, &afbdev->afb, &mode_cmd, gobj);
 	if (ret)
 		goto out;
@@ -229,15 +227,12 @@ static int astfb_create(struct drm_fb_helper *helper,
 	fb = &afbdev->afb.base;
 	afbdev->helper.fb = fb;
 
-	strcpy(info->fix.id, "astdrmfb");
-
 	info->fbops = &astfb_ops;
 
 	info->apertures->ranges[0].base = pci_resource_start(dev->pdev, 0);
 	info->apertures->ranges[0].size = pci_resource_len(dev->pdev, 0);
 
-	drm_fb_helper_fill_fix(info, fb->pitches[0], fb->format->depth);
-	drm_fb_helper_fill_var(info, &afbdev->helper, sizes->fb_width, sizes->fb_height);
+	drm_fb_helper_fill_info(info, &afbdev->helper, sizes);
 
 	info->screen_base = sysram;
 	info->screen_size = size;
diff --git a/drivers/gpu/drm/ast/ast_ttm.c b/drivers/gpu/drm/ast/ast_ttm.c
index c168d62fe8f9..75d477b37854 100644
--- a/drivers/gpu/drm/ast/ast_ttm.c
+++ b/drivers/gpu/drm/ast/ast_ttm.c
@@ -178,7 +178,6 @@ int ast_mm_init(struct ast_private *ast)
 	ret = ttm_bo_device_init(&ast->ttm.bdev,
 				 &ast_bo_driver,
 				 dev->anon_inode->i_mapping,
-				 DRM_FILE_PAGE_OFFSET,
 				 true);
 	if (ret) {
 		DRM_ERROR("Error initialising bo driver; %d\n", ret);
@@ -344,13 +343,8 @@ int ast_bo_push_sysram(struct ast_bo *bo)
 
 int ast_mmap(struct file *filp, struct vm_area_struct *vma)
 {
-	struct drm_file *file_priv;
-	struct ast_private *ast;
+	struct drm_file *file_priv = filp->private_data;
+	struct ast_private *ast = file_priv->minor->dev->dev_private;
 
-	if (unlikely(vma->vm_pgoff < DRM_FILE_PAGE_OFFSET))
-		return -EINVAL;
-
-	file_priv = filp->private_data;
-	ast = file_priv->minor->dev->dev_private;
 	return ttm_bo_mmap(filp, vma, &ast->ttm.bdev);
 }
diff --git a/drivers/gpu/drm/bochs/bochs.h b/drivers/gpu/drm/bochs/bochs.h
index 03711394f1ed..341cc9d1bab4 100644
--- a/drivers/gpu/drm/bochs/bochs.h
+++ b/drivers/gpu/drm/bochs/bochs.h
@@ -7,6 +7,7 @@
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_encoder.h>
 #include <drm/drm_fb_helper.h>
+#include <drm/drm_simple_kms_helper.h>
 
 #include <drm/drm_gem.h>
 
@@ -69,11 +70,9 @@ struct bochs_device {
 	struct edid *edid;
 
 	/* drm */
-	struct drm_device  *dev;
-	struct drm_crtc crtc;
-	struct drm_encoder encoder;
+	struct drm_device *dev;
+	struct drm_simple_display_pipe pipe;
 	struct drm_connector connector;
-	bool mode_config_initialized;
 
 	/* ttm */
 	struct {
@@ -101,8 +100,6 @@ static inline struct bochs_bo *gem_to_bochs_bo(struct drm_gem_object *gem)
 	return container_of(gem, struct bochs_bo, gem);
 }
 
-#define DRM_FILE_PAGE_OFFSET (0x100000000ULL >> PAGE_SHIFT)
-
 static inline u64 bochs_bo_mmap_offset(struct bochs_bo *bo)
 {
 	return drm_vma_node_offset_addr(&bo->bo.vma_node);
diff --git a/drivers/gpu/drm/bochs/bochs_kms.c b/drivers/gpu/drm/bochs/bochs_kms.c
index 9cd82e3631fb..5e905f50449d 100644
--- a/drivers/gpu/drm/bochs/bochs_kms.c
+++ b/drivers/gpu/drm/bochs/bochs_kms.c
@@ -22,76 +22,55 @@ MODULE_PARM_DESC(defy, "default y resolution");
 
 /* ---------------------------------------------------------------------- */
 
-static void bochs_crtc_mode_set_nofb(struct drm_crtc *crtc)
+static const uint32_t bochs_formats[] = {
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_BGRX8888,
+};
+
+static void bochs_plane_update(struct bochs_device *bochs,
+			       struct drm_plane_state *state)
 {
-	struct bochs_device *bochs =
-		container_of(crtc, struct bochs_device, crtc);
+	struct bochs_bo *bo;
 
-	bochs_hw_setmode(bochs, &crtc->mode);
+	if (!state->fb || !bochs->stride)
+		return;
+
+	bo = gem_to_bochs_bo(state->fb->obj[0]);
+	bochs_hw_setbase(bochs,
+			 state->crtc_x,
+			 state->crtc_y,
+			 bo->bo.offset);
+	bochs_hw_setformat(bochs, state->fb->format);
 }
 
-static void bochs_crtc_atomic_enable(struct drm_crtc *crtc,
-				     struct drm_crtc_state *old_crtc_state)
+static void bochs_pipe_enable(struct drm_simple_display_pipe *pipe,
+			      struct drm_crtc_state *crtc_state,
+			      struct drm_plane_state *plane_state)
 {
+	struct bochs_device *bochs = pipe->crtc.dev->dev_private;
+
+	bochs_hw_setmode(bochs, &crtc_state->mode);
+	bochs_plane_update(bochs, plane_state);
 }
 
-static void bochs_crtc_atomic_flush(struct drm_crtc *crtc,
-				    struct drm_crtc_state *old_crtc_state)
+static void bochs_pipe_update(struct drm_simple_display_pipe *pipe,
+			      struct drm_plane_state *old_state)
 {
-	struct drm_device *dev = crtc->dev;
-	struct drm_pending_vblank_event *event;
+	struct bochs_device *bochs = pipe->crtc.dev->dev_private;
+	struct drm_crtc *crtc = &pipe->crtc;
 
-	if (crtc->state && crtc->state->event) {
-		unsigned long irqflags;
+	bochs_plane_update(bochs, pipe->plane.state);
 
-		spin_lock_irqsave(&dev->event_lock, irqflags);
-		event = crtc->state->event;
+	if (crtc->state->event) {
+		spin_lock_irq(&crtc->dev->event_lock);
+		drm_crtc_send_vblank_event(crtc, crtc->state->event);
 		crtc->state->event = NULL;
-		drm_crtc_send_vblank_event(crtc, event);
-		spin_unlock_irqrestore(&dev->event_lock, irqflags);
+		spin_unlock_irq(&crtc->dev->event_lock);
 	}
 }
 
-
-/* These provide the minimum set of functions required to handle a CRTC */
-static const struct drm_crtc_funcs bochs_crtc_funcs = {
-	.set_config = drm_atomic_helper_set_config,
-	.destroy = drm_crtc_cleanup,
-	.page_flip = drm_atomic_helper_page_flip,
-	.reset = drm_atomic_helper_crtc_reset,
-	.atomic_duplicate_state = drm_atomic_helper_crtc_duplicate_state,
-	.atomic_destroy_state = drm_atomic_helper_crtc_destroy_state,
-};
-
-static const struct drm_crtc_helper_funcs bochs_helper_funcs = {
-	.mode_set_nofb = bochs_crtc_mode_set_nofb,
-	.atomic_enable = bochs_crtc_atomic_enable,
-	.atomic_flush = bochs_crtc_atomic_flush,
-};
-
-static const uint32_t bochs_formats[] = {
-	DRM_FORMAT_XRGB8888,
-	DRM_FORMAT_BGRX8888,
-};
-
-static void bochs_plane_atomic_update(struct drm_plane *plane,
-				      struct drm_plane_state *old_state)
-{
-	struct bochs_device *bochs = plane->dev->dev_private;
-	struct bochs_bo *bo;
-
-	if (!plane->state->fb)
-		return;
-	bo = gem_to_bochs_bo(plane->state->fb->obj[0]);
-	bochs_hw_setbase(bochs,
-			 plane->state->crtc_x,
-			 plane->state->crtc_y,
-			 bo->bo.offset);
-	bochs_hw_setformat(bochs, plane->state->fb->format);
-}
-
-static int bochs_plane_prepare_fb(struct drm_plane *plane,
-				struct drm_plane_state *new_state)
+static int bochs_pipe_prepare_fb(struct drm_simple_display_pipe *pipe,
+				 struct drm_plane_state *new_state)
 {
 	struct bochs_bo *bo;
 
@@ -101,8 +80,8 @@ static int bochs_plane_prepare_fb(struct drm_plane *plane,
 	return bochs_bo_pin(bo, TTM_PL_FLAG_VRAM);
 }
 
-static void bochs_plane_cleanup_fb(struct drm_plane *plane,
-				   struct drm_plane_state *old_state)
+static void bochs_pipe_cleanup_fb(struct drm_simple_display_pipe *pipe,
+				  struct drm_plane_state *old_state)
 {
 	struct bochs_bo *bo;
 
@@ -112,73 +91,13 @@ static void bochs_plane_cleanup_fb(struct drm_plane *plane,
 	bochs_bo_unpin(bo);
 }
 
-static const struct drm_plane_helper_funcs bochs_plane_helper_funcs = {
-	.atomic_update = bochs_plane_atomic_update,
-	.prepare_fb = bochs_plane_prepare_fb,
-	.cleanup_fb = bochs_plane_cleanup_fb,
-};
-
-static const struct drm_plane_funcs bochs_plane_funcs = {
-       .update_plane   = drm_atomic_helper_update_plane,
-       .disable_plane  = drm_atomic_helper_disable_plane,
-       .destroy        = drm_primary_helper_destroy,
-       .reset          = drm_atomic_helper_plane_reset,
-       .atomic_duplicate_state = drm_atomic_helper_plane_duplicate_state,
-       .atomic_destroy_state = drm_atomic_helper_plane_destroy_state,
-};
-
-static struct drm_plane *bochs_primary_plane(struct drm_device *dev)
-{
-	struct drm_plane *primary;
-	int ret;
-
-	primary = kzalloc(sizeof(*primary), GFP_KERNEL);
-	if (primary == NULL) {
-		DRM_DEBUG_KMS("Failed to allocate primary plane\n");
-		return NULL;
-	}
-
-	ret = drm_universal_plane_init(dev, primary, 0,
-				       &bochs_plane_funcs,
-				       bochs_formats,
-				       ARRAY_SIZE(bochs_formats),
-				       NULL,
-				       DRM_PLANE_TYPE_PRIMARY, NULL);
-	if (ret) {
-		kfree(primary);
-		return NULL;
-	}
-
-	drm_plane_helper_add(primary, &bochs_plane_helper_funcs);
-	return primary;
-}
-
-static void bochs_crtc_init(struct drm_device *dev)
-{
-	struct bochs_device *bochs = dev->dev_private;
-	struct drm_crtc *crtc = &bochs->crtc;
-	struct drm_plane *primary = bochs_primary_plane(dev);
-
-	drm_crtc_init_with_planes(dev, crtc, primary, NULL,
-				  &bochs_crtc_funcs, NULL);
-	drm_crtc_helper_add(crtc, &bochs_helper_funcs);
-}
-
-static const struct drm_encoder_funcs bochs_encoder_encoder_funcs = {
-	.destroy = drm_encoder_cleanup,
+static const struct drm_simple_display_pipe_funcs bochs_pipe_funcs = {
+	.enable	    = bochs_pipe_enable,
+	.update	    = bochs_pipe_update,
+	.prepare_fb = bochs_pipe_prepare_fb,
+	.cleanup_fb = bochs_pipe_cleanup_fb,
 };
 
-static void bochs_encoder_init(struct drm_device *dev)
-{
-	struct bochs_device *bochs = dev->dev_private;
-	struct drm_encoder *encoder = &bochs->encoder;
-
-	encoder->possible_crtcs = 0x1;
-	drm_encoder_init(dev, encoder, &bochs_encoder_encoder_funcs,
-			 DRM_MODE_ENCODER_DAC, NULL);
-}
-
-
 static int bochs_connector_get_modes(struct drm_connector *connector)
 {
 	struct bochs_device *bochs =
@@ -214,20 +133,9 @@ static enum drm_mode_status bochs_connector_mode_valid(struct drm_connector *con
 	return MODE_OK;
 }
 
-static struct drm_encoder *
-bochs_connector_best_encoder(struct drm_connector *connector)
-{
-	int enc_id = connector->encoder_ids[0];
-	/* pick the encoder ids */
-	if (enc_id)
-		return drm_encoder_find(connector->dev, NULL, enc_id);
-	return NULL;
-}
-
 static const struct drm_connector_helper_funcs bochs_connector_connector_helper_funcs = {
 	.get_modes = bochs_connector_get_modes,
 	.mode_valid = bochs_connector_mode_valid,
-	.best_encoder = bochs_connector_best_encoder,
 };
 
 static const struct drm_connector_funcs bochs_connector_connector_funcs = {
@@ -278,7 +186,6 @@ const struct drm_mode_config_funcs bochs_mode_funcs = {
 int bochs_kms_init(struct bochs_device *bochs)
 {
 	drm_mode_config_init(bochs->dev);
-	bochs->mode_config_initialized = true;
 
 	bochs->dev->mode_config.max_width = 8192;
 	bochs->dev->mode_config.max_height = 8192;
@@ -290,11 +197,14 @@ int bochs_kms_init(struct bochs_device *bochs)
 
 	bochs->dev->mode_config.funcs = &bochs_mode_funcs;
 
-	bochs_crtc_init(bochs->dev);
-	bochs_encoder_init(bochs->dev);
 	bochs_connector_init(bochs->dev);
-	drm_connector_attach_encoder(&bochs->connector,
-					  &bochs->encoder);
+	drm_simple_display_pipe_init(bochs->dev,
+				     &bochs->pipe,
+				     &bochs_pipe_funcs,
+				     bochs_formats,
+				     ARRAY_SIZE(bochs_formats),
+				     NULL,
+				     &bochs->connector);
 
 	drm_mode_config_reset(bochs->dev);
 
@@ -303,8 +213,6 @@ int bochs_kms_init(struct bochs_device *bochs)
 
 void bochs_kms_fini(struct bochs_device *bochs)
 {
-	if (bochs->mode_config_initialized) {
-		drm_mode_config_cleanup(bochs->dev);
-		bochs->mode_config_initialized = false;
-	}
+	drm_atomic_helper_shutdown(bochs->dev);
+	drm_mode_config_cleanup(bochs->dev);
 }
diff --git a/drivers/gpu/drm/bochs/bochs_mm.c b/drivers/gpu/drm/bochs/bochs_mm.c
index 49463348a07a..4a40308169c4 100644
--- a/drivers/gpu/drm/bochs/bochs_mm.c
+++ b/drivers/gpu/drm/bochs/bochs_mm.c
@@ -156,7 +156,6 @@ int bochs_mm_init(struct bochs_device *bochs)
 	ret = ttm_bo_device_init(&bochs->ttm.bdev,
 				 &bochs_bo_driver,
 				 bochs->dev->anon_inode->i_mapping,
-				 DRM_FILE_PAGE_OFFSET,
 				 true);
 	if (ret) {
 		DRM_ERROR("Error initialising bo driver; %d\n", ret);
@@ -264,14 +263,9 @@ int bochs_bo_unpin(struct bochs_bo *bo)
 
 int bochs_mmap(struct file *filp, struct vm_area_struct *vma)
 {
-	struct drm_file *file_priv;
-	struct bochs_device *bochs;
+	struct drm_file *file_priv = filp->private_data;
+	struct bochs_device *bochs = file_priv->minor->dev->dev_private;
 
-	if (unlikely(vma->vm_pgoff < DRM_FILE_PAGE_OFFSET))
-		return -EINVAL;
-
-	file_priv = filp->private_data;
-	bochs = file_priv->minor->dev->dev_private;
 	return ttm_bo_mmap(filp, vma, &bochs->ttm.bdev);
 }
 
diff --git a/drivers/gpu/drm/bridge/dumb-vga-dac.c b/drivers/gpu/drm/bridge/dumb-vga-dac.c
index 0805801f4e94..e64736c39a9f 100644
--- a/drivers/gpu/drm/bridge/dumb-vga-dac.c
+++ b/drivers/gpu/drm/bridge/dumb-vga-dac.c
@@ -234,7 +234,7 @@ static int dumb_vga_remove(struct platform_device *pdev)
  */
 static const struct drm_bridge_timings default_dac_timings = {
 	/* Timing specifications, datasheet page 7 */
-	.sampling_edge = DRM_BUS_FLAG_PIXDATA_POSEDGE,
+	.input_bus_flags = DRM_BUS_FLAG_PIXDATA_SAMPLE_POSEDGE,
 	.setup_time_ps = 500,
 	.hold_time_ps = 1500,
 };
@@ -245,7 +245,7 @@ static const struct drm_bridge_timings default_dac_timings = {
  */
 static const struct drm_bridge_timings ti_ths8134_dac_timings = {
 	/* From timing diagram, datasheet page 9 */
-	.sampling_edge = DRM_BUS_FLAG_PIXDATA_POSEDGE,
+	.input_bus_flags = DRM_BUS_FLAG_PIXDATA_SAMPLE_POSEDGE,
 	/* From datasheet, page 12 */
 	.setup_time_ps = 3000,
 	/* I guess this means latched input */
@@ -258,7 +258,7 @@ static const struct drm_bridge_timings ti_ths8134_dac_timings = {
  */
 static const struct drm_bridge_timings ti_ths8135_dac_timings = {
 	/* From timing diagram, datasheet page 14 */
-	.sampling_edge = DRM_BUS_FLAG_PIXDATA_POSEDGE,
+	.input_bus_flags = DRM_BUS_FLAG_PIXDATA_SAMPLE_POSEDGE,
 	/* From datasheet, page 16 */
 	.setup_time_ps = 2000,
 	.hold_time_ps = 500,
diff --git a/drivers/gpu/drm/bridge/tc358767.c b/drivers/gpu/drm/bridge/tc358767.c
index 888980d4bc74..e570c9dee180 100644
--- a/drivers/gpu/drm/bridge/tc358767.c
+++ b/drivers/gpu/drm/bridge/tc358767.c
@@ -1222,8 +1222,8 @@ static int tc_bridge_attach(struct drm_bridge *bridge)
 					 &bus_format, 1);
 	tc->connector.display_info.bus_flags =
 		DRM_BUS_FLAG_DE_HIGH |
-		DRM_BUS_FLAG_PIXDATA_NEGEDGE |
-		DRM_BUS_FLAG_SYNC_NEGEDGE;
+		DRM_BUS_FLAG_PIXDATA_DRIVE_NEGEDGE |
+		DRM_BUS_FLAG_SYNC_DRIVE_NEGEDGE;
 	drm_connector_attach_encoder(&tc->connector, tc->bridge.encoder);
 
 	return 0;
diff --git a/drivers/gpu/drm/bridge/ti-tfp410.c b/drivers/gpu/drm/bridge/ti-tfp410.c
index 7bfb4f338813..8b0e71bd3ca7 100644
--- a/drivers/gpu/drm/bridge/ti-tfp410.c
+++ b/drivers/gpu/drm/bridge/ti-tfp410.c
@@ -27,10 +27,16 @@
 struct tfp410 {
 	struct drm_bridge	bridge;
 	struct drm_connector	connector;
+	unsigned int		connector_type;
 
+	u32			bus_format;
 	struct i2c_adapter	*ddc;
 	struct gpio_desc	*hpd;
+	int			hpd_irq;
 	struct delayed_work	hpd_work;
+	struct gpio_desc	*powerdown;
+
+	struct drm_bridge_timings timings;
 
 	struct device *dev;
 };
@@ -120,26 +126,47 @@ static int tfp410_attach(struct drm_bridge *bridge)
 		return -ENODEV;
 	}
 
-	if (dvi->hpd)
+	if (dvi->hpd_irq >= 0)
 		dvi->connector.polled = DRM_CONNECTOR_POLL_HPD;
+	else
+		dvi->connector.polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT;
 
 	drm_connector_helper_add(&dvi->connector,
 				 &tfp410_con_helper_funcs);
 	ret = drm_connector_init(bridge->dev, &dvi->connector,
-				 &tfp410_con_funcs, DRM_MODE_CONNECTOR_HDMIA);
+				 &tfp410_con_funcs, dvi->connector_type);
 	if (ret) {
 		dev_err(dvi->dev, "drm_connector_init() failed: %d\n", ret);
 		return ret;
 	}
 
+	drm_display_info_set_bus_formats(&dvi->connector.display_info,
+					 &dvi->bus_format, 1);
+
 	drm_connector_attach_encoder(&dvi->connector,
 					  bridge->encoder);
 
 	return 0;
 }
 
+static void tfp410_enable(struct drm_bridge *bridge)
+{
+	struct tfp410 *dvi = drm_bridge_to_tfp410(bridge);
+
+	gpiod_set_value_cansleep(dvi->powerdown, 0);
+}
+
+static void tfp410_disable(struct drm_bridge *bridge)
+{
+	struct tfp410 *dvi = drm_bridge_to_tfp410(bridge);
+
+	gpiod_set_value_cansleep(dvi->powerdown, 1);
+}
+
 static const struct drm_bridge_funcs tfp410_bridge_funcs = {
 	.attach		= tfp410_attach,
+	.enable		= tfp410_enable,
+	.disable	= tfp410_disable,
 };
 
 static void tfp410_hpd_work_func(struct work_struct *work)
@@ -162,6 +189,83 @@ static irqreturn_t tfp410_hpd_irq_thread(int irq, void *arg)
 	return IRQ_HANDLED;
 }
 
+static const struct drm_bridge_timings tfp410_default_timings = {
+	.input_bus_flags = DRM_BUS_FLAG_PIXDATA_SAMPLE_POSEDGE
+			 | DRM_BUS_FLAG_DE_HIGH,
+	.setup_time_ps = 1200,
+	.hold_time_ps = 1300,
+};
+
+static int tfp410_parse_timings(struct tfp410 *dvi, bool i2c)
+{
+	struct drm_bridge_timings *timings = &dvi->timings;
+	struct device_node *ep;
+	u32 pclk_sample = 0;
+	u32 bus_width = 24;
+	s32 deskew = 0;
+
+	/* Start with defaults. */
+	*timings = tfp410_default_timings;
+
+	if (i2c)
+		/*
+		 * In I2C mode timings are configured through the I2C interface.
+		 * As the driver doesn't support I2C configuration yet, we just
+		 * go with the defaults (BSEL=1, DSEL=1, DKEN=0, EDGE=1).
+		 */
+		return 0;
+
+	/*
+	 * In non-I2C mode, timings are configured through the BSEL, DSEL, DKEN
+	 * and EDGE pins. They are specified in DT through endpoint properties
+	 * and vendor-specific properties.
+	 */
+	ep = of_graph_get_endpoint_by_regs(dvi->dev->of_node, 0, 0);
+	if (!ep)
+		return -EINVAL;
+
+	/* Get the sampling edge from the endpoint. */
+	of_property_read_u32(ep, "pclk-sample", &pclk_sample);
+	of_property_read_u32(ep, "bus-width", &bus_width);
+	of_node_put(ep);
+
+	timings->input_bus_flags = DRM_BUS_FLAG_DE_HIGH;
+
+	switch (pclk_sample) {
+	case 0:
+		timings->input_bus_flags |= DRM_BUS_FLAG_PIXDATA_SAMPLE_NEGEDGE
+					 |  DRM_BUS_FLAG_SYNC_SAMPLE_NEGEDGE;
+		break;
+	case 1:
+		timings->input_bus_flags |= DRM_BUS_FLAG_PIXDATA_SAMPLE_POSEDGE
+					 |  DRM_BUS_FLAG_SYNC_SAMPLE_POSEDGE;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	switch (bus_width) {
+	case 12:
+		dvi->bus_format = MEDIA_BUS_FMT_RGB888_2X12_LE;
+		break;
+	case 24:
+		dvi->bus_format = MEDIA_BUS_FMT_RGB888_1X24;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	/* Get the setup and hold time from vendor-specific properties. */
+	of_property_read_u32(dvi->dev->of_node, "ti,deskew", (u32 *)&deskew);
+	if (deskew < -4 || deskew > 3)
+		return -EINVAL;
+
+	timings->setup_time_ps = min(0, 1200 - 350 * deskew);
+	timings->hold_time_ps = min(0, 1300 + 350 * deskew);
+
+	return 0;
+}
+
 static int tfp410_get_connector_properties(struct tfp410 *dvi)
 {
 	struct device_node *connector_node, *ddc_phandle;
@@ -172,6 +276,11 @@ static int tfp410_get_connector_properties(struct tfp410 *dvi)
 	if (!connector_node)
 		return -ENODEV;
 
+	if (of_device_is_compatible(connector_node, "hdmi-connector"))
+		dvi->connector_type = DRM_MODE_CONNECTOR_HDMIA;
+	else
+		dvi->connector_type = DRM_MODE_CONNECTOR_DVID;
+
 	dvi->hpd = fwnode_get_named_gpiod(&connector_node->fwnode,
 					"hpd-gpios", 0, GPIOD_IN, "hpd");
 	if (IS_ERR(dvi->hpd)) {
@@ -200,7 +309,7 @@ fail:
 	return ret;
 }
 
-static int tfp410_init(struct device *dev)
+static int tfp410_init(struct device *dev, bool i2c)
 {
 	struct tfp410 *dvi;
 	int ret;
@@ -217,16 +326,33 @@ static int tfp410_init(struct device *dev)
 
 	dvi->bridge.funcs = &tfp410_bridge_funcs;
 	dvi->bridge.of_node = dev->of_node;
+	dvi->bridge.timings = &dvi->timings;
 	dvi->dev = dev;
 
+	ret = tfp410_parse_timings(dvi, i2c);
+	if (ret)
+		goto fail;
+
 	ret = tfp410_get_connector_properties(dvi);
 	if (ret)
 		goto fail;
 
-	if (dvi->hpd) {
+	dvi->powerdown = devm_gpiod_get_optional(dev, "powerdown",
+						 GPIOD_OUT_HIGH);
+	if (IS_ERR(dvi->powerdown)) {
+		dev_err(dev, "failed to parse powerdown gpio\n");
+		return PTR_ERR(dvi->powerdown);
+	}
+
+	if (dvi->hpd)
+		dvi->hpd_irq = gpiod_to_irq(dvi->hpd);
+	else
+		dvi->hpd_irq = -ENXIO;
+
+	if (dvi->hpd_irq >= 0) {
 		INIT_DELAYED_WORK(&dvi->hpd_work, tfp410_hpd_work_func);
 
-		ret = devm_request_threaded_irq(dev, gpiod_to_irq(dvi->hpd),
+		ret = devm_request_threaded_irq(dev, dvi->hpd_irq,
 			NULL, tfp410_hpd_irq_thread, IRQF_TRIGGER_RISING |
 			IRQF_TRIGGER_FALLING | IRQF_ONESHOT,
 			"hdmi-hpd", dvi);
@@ -264,7 +390,7 @@ static int tfp410_fini(struct device *dev)
 
 static int tfp410_probe(struct platform_device *pdev)
 {
-	return tfp410_init(&pdev->dev);
+	return tfp410_init(&pdev->dev, false);
 }
 
 static int tfp410_remove(struct platform_device *pdev)
@@ -301,7 +427,7 @@ static int tfp410_i2c_probe(struct i2c_client *client,
 		return -ENXIO;
 	}
 
-	return tfp410_init(&client->dev);
+	return tfp410_init(&client->dev, true);
 }
 
 static int tfp410_i2c_remove(struct i2c_client *client)
diff --git a/drivers/gpu/drm/cirrus/Kconfig b/drivers/gpu/drm/cirrus/Kconfig
index fc78c90ee931..dd4f52a0bc1c 100644
--- a/drivers/gpu/drm/cirrus/Kconfig
+++ b/drivers/gpu/drm/cirrus/Kconfig
@@ -2,7 +2,7 @@ config DRM_CIRRUS_QEMU
 	tristate "Cirrus driver for QEMU emulated device"
 	depends on DRM && PCI && MMU
 	select DRM_KMS_HELPER
-	select DRM_TTM
+	select DRM_GEM_SHMEM_HELPER
 	help
 	 This is a KMS driver for emulated cirrus device in qemu.
 	 It is *NOT* intended for real cirrus devices. This requires
diff --git a/drivers/gpu/drm/cirrus/Makefile b/drivers/gpu/drm/cirrus/Makefile
index 919c0a336c97..acf8971d37a1 100644
--- a/drivers/gpu/drm/cirrus/Makefile
+++ b/drivers/gpu/drm/cirrus/Makefile
@@ -1,4 +1 @@
-cirrus-y  := cirrus_main.o cirrus_mode.o \
-	cirrus_drv.o cirrus_fbdev.o cirrus_ttm.o
-
 obj-$(CONFIG_DRM_CIRRUS_QEMU) += cirrus.o
diff --git a/drivers/gpu/drm/cirrus/cirrus.c b/drivers/gpu/drm/cirrus/cirrus.c
new file mode 100644
index 000000000000..be4ea370ba31
--- /dev/null
+++ b/drivers/gpu/drm/cirrus/cirrus.c
@@ -0,0 +1,657 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright 2012-2019 Red Hat
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License version 2. See the file COPYING in the main
+ * directory of this archive for more details.
+ *
+ * Authors: Matthew Garrett
+ *	    Dave Airlie
+ *	    Gerd Hoffmann
+ *
+ * Portions of this code derived from cirrusfb.c:
+ * drivers/video/cirrusfb.c - driver for Cirrus Logic chipsets
+ *
+ * Copyright 1999-2001 Jeff Garzik <jgarzik@pobox.com>
+ */
+
+#include <linux/console.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+
+#include <video/cirrus.h>
+#include <video/vga.h>
+
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_atomic_state_helper.h>
+#include <drm/drm_connector.h>
+#include <drm/drm_damage_helper.h>
+#include <drm/drm_drv.h>
+#include <drm/drm_fb_helper.h>
+#include <drm/drm_file.h>
+#include <drm/drm_format_helper.h>
+#include <drm/drm_fourcc.h>
+#include <drm/drm_gem_shmem_helper.h>
+#include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_ioctl.h>
+#include <drm/drm_modeset_helper_vtables.h>
+#include <drm/drm_probe_helper.h>
+#include <drm/drm_simple_kms_helper.h>
+#include <drm/drm_vblank.h>
+
+#define DRIVER_NAME "cirrus"
+#define DRIVER_DESC "qemu cirrus vga"
+#define DRIVER_DATE "2019"
+#define DRIVER_MAJOR 2
+#define DRIVER_MINOR 0
+
+#define CIRRUS_MAX_PITCH (0x1FF << 3)      /* (4096 - 1) & ~111b bytes */
+#define CIRRUS_VRAM_SIZE (4 * 1024 * 1024) /* 4 MB */
+
+struct cirrus_device {
+	struct drm_device	       dev;
+	struct drm_simple_display_pipe pipe;
+	struct drm_connector	       conn;
+	unsigned int		       cpp;
+	unsigned int		       pitch;
+	void __iomem		       *vram;
+	void __iomem		       *mmio;
+};
+
+/* ------------------------------------------------------------------ */
+/*
+ * The meat of this driver. The core passes us a mode and we have to program
+ * it. The modesetting here is the bare minimum required to satisfy the qemu
+ * emulation of this hardware, and running this against a real device is
+ * likely to result in an inadequately programmed mode. We've already had
+ * the opportunity to modify the mode, so whatever we receive here should
+ * be something that can be correctly programmed and displayed
+ */
+
+#define SEQ_INDEX 4
+#define SEQ_DATA 5
+
+static u8 rreg_seq(struct cirrus_device *cirrus, u8 reg)
+{
+	iowrite8(reg, cirrus->mmio + SEQ_INDEX);
+	return ioread8(cirrus->mmio + SEQ_DATA);
+}
+
+static void wreg_seq(struct cirrus_device *cirrus, u8 reg, u8 val)
+{
+	iowrite8(reg, cirrus->mmio + SEQ_INDEX);
+	iowrite8(val, cirrus->mmio + SEQ_DATA);
+}
+
+#define CRT_INDEX 0x14
+#define CRT_DATA 0x15
+
+static u8 rreg_crt(struct cirrus_device *cirrus, u8 reg)
+{
+	iowrite8(reg, cirrus->mmio + CRT_INDEX);
+	return ioread8(cirrus->mmio + CRT_DATA);
+}
+
+static void wreg_crt(struct cirrus_device *cirrus, u8 reg, u8 val)
+{
+	iowrite8(reg, cirrus->mmio + CRT_INDEX);
+	iowrite8(val, cirrus->mmio + CRT_DATA);
+}
+
+#define GFX_INDEX 0xe
+#define GFX_DATA 0xf
+
+static void wreg_gfx(struct cirrus_device *cirrus, u8 reg, u8 val)
+{
+	iowrite8(reg, cirrus->mmio + GFX_INDEX);
+	iowrite8(val, cirrus->mmio + GFX_DATA);
+}
+
+#define VGA_DAC_MASK  0x06
+
+static void wreg_hdr(struct cirrus_device *cirrus, u8 val)
+{
+	ioread8(cirrus->mmio + VGA_DAC_MASK);
+	ioread8(cirrus->mmio + VGA_DAC_MASK);
+	ioread8(cirrus->mmio + VGA_DAC_MASK);
+	ioread8(cirrus->mmio + VGA_DAC_MASK);
+	iowrite8(val, cirrus->mmio + VGA_DAC_MASK);
+}
+
+static int cirrus_convert_to(struct drm_framebuffer *fb)
+{
+	if (fb->format->cpp[0] == 4 && fb->pitches[0] > CIRRUS_MAX_PITCH) {
+		if (fb->width * 3 <= CIRRUS_MAX_PITCH)
+			/* convert from XR24 to RG24 */
+			return 3;
+		else
+			/* convert from XR24 to RG16 */
+			return 2;
+	}
+	return 0;
+}
+
+static int cirrus_cpp(struct drm_framebuffer *fb)
+{
+	int convert_cpp = cirrus_convert_to(fb);
+
+	if (convert_cpp)
+		return convert_cpp;
+	return fb->format->cpp[0];
+}
+
+static int cirrus_pitch(struct drm_framebuffer *fb)
+{
+	int convert_cpp = cirrus_convert_to(fb);
+
+	if (convert_cpp)
+		return convert_cpp * fb->width;
+	return fb->pitches[0];
+}
+
+static void cirrus_set_start_address(struct cirrus_device *cirrus, u32 offset)
+{
+	u32 addr;
+	u8 tmp;
+
+	addr = offset >> 2;
+	wreg_crt(cirrus, 0x0c, (u8)((addr >> 8) & 0xff));
+	wreg_crt(cirrus, 0x0d, (u8)(addr & 0xff));
+
+	tmp = rreg_crt(cirrus, 0x1b);
+	tmp &= 0xf2;
+	tmp |= (addr >> 16) & 0x01;
+	tmp |= (addr >> 15) & 0x0c;
+	wreg_crt(cirrus, 0x1b, tmp);
+
+	tmp = rreg_crt(cirrus, 0x1d);
+	tmp &= 0x7f;
+	tmp |= (addr >> 12) & 0x80;
+	wreg_crt(cirrus, 0x1d, tmp);
+}
+
+static int cirrus_mode_set(struct cirrus_device *cirrus,
+			   struct drm_display_mode *mode,
+			   struct drm_framebuffer *fb)
+{
+	int hsyncstart, hsyncend, htotal, hdispend;
+	int vtotal, vdispend;
+	int tmp;
+	int sr07 = 0, hdr = 0;
+
+	htotal = mode->htotal / 8;
+	hsyncend = mode->hsync_end / 8;
+	hsyncstart = mode->hsync_start / 8;
+	hdispend = mode->hdisplay / 8;
+
+	vtotal = mode->vtotal;
+	vdispend = mode->vdisplay;
+
+	vdispend -= 1;
+	vtotal -= 2;
+
+	htotal -= 5;
+	hdispend -= 1;
+	hsyncstart += 1;
+	hsyncend += 1;
+
+	wreg_crt(cirrus, VGA_CRTC_V_SYNC_END, 0x20);
+	wreg_crt(cirrus, VGA_CRTC_H_TOTAL, htotal);
+	wreg_crt(cirrus, VGA_CRTC_H_DISP, hdispend);
+	wreg_crt(cirrus, VGA_CRTC_H_SYNC_START, hsyncstart);
+	wreg_crt(cirrus, VGA_CRTC_H_SYNC_END, hsyncend);
+	wreg_crt(cirrus, VGA_CRTC_V_TOTAL, vtotal & 0xff);
+	wreg_crt(cirrus, VGA_CRTC_V_DISP_END, vdispend & 0xff);
+
+	tmp = 0x40;
+	if ((vdispend + 1) & 512)
+		tmp |= 0x20;
+	wreg_crt(cirrus, VGA_CRTC_MAX_SCAN, tmp);
+
+	/*
+	 * Overflow bits for values that don't fit in the standard registers
+	 */
+	tmp = 0x10;
+	if (vtotal & 0x100)
+		tmp |= 0x01;
+	if (vdispend & 0x100)
+		tmp |= 0x02;
+	if ((vdispend + 1) & 0x100)
+		tmp |= 0x08;
+	if (vtotal & 0x200)
+		tmp |= 0x20;
+	if (vdispend & 0x200)
+		tmp |= 0x40;
+	wreg_crt(cirrus, VGA_CRTC_OVERFLOW, tmp);
+
+	tmp = 0;
+
+	/* More overflow bits */
+
+	if ((htotal + 5) & 0x40)
+		tmp |= 0x10;
+	if ((htotal + 5) & 0x80)
+		tmp |= 0x20;
+	if (vtotal & 0x100)
+		tmp |= 0x40;
+	if (vtotal & 0x200)
+		tmp |= 0x80;
+
+	wreg_crt(cirrus, CL_CRT1A, tmp);
+
+	/* Disable Hercules/CGA compatibility */
+	wreg_crt(cirrus, VGA_CRTC_MODE, 0x03);
+
+	sr07 = rreg_seq(cirrus, 0x07);
+	sr07 &= 0xe0;
+	hdr = 0;
+
+	cirrus->cpp = cirrus_cpp(fb);
+	switch (cirrus->cpp * 8) {
+	case 8:
+		sr07 |= 0x11;
+		break;
+	case 16:
+		sr07 |= 0x17;
+		hdr = 0xc1;
+		break;
+	case 24:
+		sr07 |= 0x15;
+		hdr = 0xc5;
+		break;
+	case 32:
+		sr07 |= 0x19;
+		hdr = 0xc5;
+		break;
+	default:
+		return -1;
+	}
+
+	wreg_seq(cirrus, 0x7, sr07);
+
+	/* Program the pitch */
+	cirrus->pitch = cirrus_pitch(fb);
+	tmp = cirrus->pitch / 8;
+	wreg_crt(cirrus, VGA_CRTC_OFFSET, tmp);
+
+	/* Enable extended blanking and pitch bits, and enable full memory */
+	tmp = 0x22;
+	tmp |= (cirrus->pitch >> 7) & 0x10;
+	tmp |= (cirrus->pitch >> 6) & 0x40;
+	wreg_crt(cirrus, 0x1b, tmp);
+
+	/* Enable high-colour modes */
+	wreg_gfx(cirrus, VGA_GFX_MODE, 0x40);
+
+	/* And set graphics mode */
+	wreg_gfx(cirrus, VGA_GFX_MISC, 0x01);
+
+	wreg_hdr(cirrus, hdr);
+
+	cirrus_set_start_address(cirrus, 0);
+
+	/* Unblank (needed on S3 resume, vgabios doesn't do it then) */
+	outb(0x20, 0x3c0);
+	return 0;
+}
+
+static int cirrus_fb_blit_rect(struct drm_framebuffer *fb,
+			       struct drm_rect *rect)
+{
+	struct cirrus_device *cirrus = fb->dev->dev_private;
+	void *vmap;
+
+	vmap = drm_gem_shmem_vmap(fb->obj[0]);
+	if (!vmap)
+		return -ENOMEM;
+
+	if (cirrus->cpp == fb->format->cpp[0])
+		drm_fb_memcpy_dstclip(cirrus->vram,
+				      vmap, fb, rect);
+
+	else if (fb->format->cpp[0] == 4 && cirrus->cpp == 2)
+		drm_fb_xrgb8888_to_rgb565_dstclip(cirrus->vram,
+						  cirrus->pitch,
+						  vmap, fb, rect, false);
+
+	else if (fb->format->cpp[0] == 4 && cirrus->cpp == 3)
+		drm_fb_xrgb8888_to_rgb888_dstclip(cirrus->vram,
+						  cirrus->pitch,
+						  vmap, fb, rect);
+
+	else
+		WARN_ON_ONCE("cpp mismatch");
+
+	drm_gem_shmem_vunmap(fb->obj[0], vmap);
+	return 0;
+}
+
+static int cirrus_fb_blit_fullscreen(struct drm_framebuffer *fb)
+{
+	struct drm_rect fullscreen = {
+		.x1 = 0,
+		.x2 = fb->width,
+		.y1 = 0,
+		.y2 = fb->height,
+	};
+	return cirrus_fb_blit_rect(fb, &fullscreen);
+}
+
+static int cirrus_check_size(int width, int height,
+			     struct drm_framebuffer *fb)
+{
+	int pitch = width * 2;
+
+	if (fb)
+		pitch = cirrus_pitch(fb);
+
+	if (pitch > CIRRUS_MAX_PITCH)
+		return -EINVAL;
+	if (pitch * height > CIRRUS_VRAM_SIZE)
+		return -EINVAL;
+	return 0;
+}
+
+/* ------------------------------------------------------------------ */
+/* cirrus connector						      */
+
+static int cirrus_conn_get_modes(struct drm_connector *conn)
+{
+	int count;
+
+	count = drm_add_modes_noedid(conn,
+				     conn->dev->mode_config.max_width,
+				     conn->dev->mode_config.max_height);
+	drm_set_preferred_mode(conn, 1024, 768);
+	return count;
+}
+
+static const struct drm_connector_helper_funcs cirrus_conn_helper_funcs = {
+	.get_modes = cirrus_conn_get_modes,
+};
+
+static const struct drm_connector_funcs cirrus_conn_funcs = {
+	.fill_modes = drm_helper_probe_single_connector_modes,
+	.destroy = drm_connector_cleanup,
+	.reset = drm_atomic_helper_connector_reset,
+	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
+};
+
+static int cirrus_conn_init(struct cirrus_device *cirrus)
+{
+	drm_connector_helper_add(&cirrus->conn, &cirrus_conn_helper_funcs);
+	return drm_connector_init(&cirrus->dev, &cirrus->conn,
+				  &cirrus_conn_funcs, DRM_MODE_CONNECTOR_VGA);
+
+}
+
+/* ------------------------------------------------------------------ */
+/* cirrus (simple) display pipe					      */
+
+static enum drm_mode_status cirrus_pipe_mode_valid(struct drm_crtc *crtc,
+						   const struct drm_display_mode *mode)
+{
+	if (cirrus_check_size(mode->hdisplay, mode->vdisplay, NULL) < 0)
+		return MODE_BAD;
+	return MODE_OK;
+}
+
+static int cirrus_pipe_check(struct drm_simple_display_pipe *pipe,
+			     struct drm_plane_state *plane_state,
+			     struct drm_crtc_state *crtc_state)
+{
+	struct drm_framebuffer *fb = plane_state->fb;
+
+	if (!fb)
+		return 0;
+	return cirrus_check_size(fb->width, fb->height, fb);
+}
+
+static void cirrus_pipe_enable(struct drm_simple_display_pipe *pipe,
+			       struct drm_crtc_state *crtc_state,
+			       struct drm_plane_state *plane_state)
+{
+	struct cirrus_device *cirrus = pipe->crtc.dev->dev_private;
+
+	cirrus_mode_set(cirrus, &crtc_state->mode, plane_state->fb);
+	cirrus_fb_blit_fullscreen(plane_state->fb);
+}
+
+static void cirrus_pipe_update(struct drm_simple_display_pipe *pipe,
+			       struct drm_plane_state *old_state)
+{
+	struct cirrus_device *cirrus = pipe->crtc.dev->dev_private;
+	struct drm_plane_state *state = pipe->plane.state;
+	struct drm_crtc *crtc = &pipe->crtc;
+	struct drm_rect rect;
+
+	if (pipe->plane.state->fb &&
+	    cirrus->cpp != cirrus_cpp(pipe->plane.state->fb))
+		cirrus_mode_set(cirrus, &crtc->mode,
+				pipe->plane.state->fb);
+
+	if (drm_atomic_helper_damage_merged(old_state, state, &rect))
+		cirrus_fb_blit_rect(pipe->plane.state->fb, &rect);
+
+	if (crtc->state->event) {
+		spin_lock_irq(&crtc->dev->event_lock);
+		drm_crtc_send_vblank_event(crtc, crtc->state->event);
+		crtc->state->event = NULL;
+		spin_unlock_irq(&crtc->dev->event_lock);
+	}
+}
+
+static const struct drm_simple_display_pipe_funcs cirrus_pipe_funcs = {
+	.mode_valid = cirrus_pipe_mode_valid,
+	.check	    = cirrus_pipe_check,
+	.enable	    = cirrus_pipe_enable,
+	.update	    = cirrus_pipe_update,
+};
+
+static const uint32_t cirrus_formats[] = {
+	DRM_FORMAT_RGB565,
+	DRM_FORMAT_RGB888,
+	DRM_FORMAT_XRGB8888,
+};
+
+static const uint64_t cirrus_modifiers[] = {
+	DRM_FORMAT_MOD_LINEAR,
+	DRM_FORMAT_MOD_INVALID
+};
+
+static int cirrus_pipe_init(struct cirrus_device *cirrus)
+{
+	return drm_simple_display_pipe_init(&cirrus->dev,
+					    &cirrus->pipe,
+					    &cirrus_pipe_funcs,
+					    cirrus_formats,
+					    ARRAY_SIZE(cirrus_formats),
+					    cirrus_modifiers,
+					    &cirrus->conn);
+}
+
+/* ------------------------------------------------------------------ */
+/* cirrus framebuffers & mode config				      */
+
+static struct drm_framebuffer*
+cirrus_fb_create(struct drm_device *dev, struct drm_file *file_priv,
+		 const struct drm_mode_fb_cmd2 *mode_cmd)
+{
+	if (mode_cmd->pixel_format != DRM_FORMAT_RGB565 &&
+	    mode_cmd->pixel_format != DRM_FORMAT_RGB888 &&
+	    mode_cmd->pixel_format != DRM_FORMAT_XRGB8888)
+		return ERR_PTR(-EINVAL);
+	if (cirrus_check_size(mode_cmd->width, mode_cmd->height, NULL) < 0)
+		return ERR_PTR(-EINVAL);
+	return drm_gem_fb_create_with_dirty(dev, file_priv, mode_cmd);
+}
+
+static const struct drm_mode_config_funcs cirrus_mode_config_funcs = {
+	.fb_create = cirrus_fb_create,
+	.atomic_check = drm_atomic_helper_check,
+	.atomic_commit = drm_atomic_helper_commit,
+};
+
+static void cirrus_mode_config_init(struct cirrus_device *cirrus)
+{
+	struct drm_device *dev = &cirrus->dev;
+
+	drm_mode_config_init(dev);
+	dev->mode_config.min_width = 0;
+	dev->mode_config.min_height = 0;
+	dev->mode_config.max_width = CIRRUS_MAX_PITCH / 2;
+	dev->mode_config.max_height = 1024;
+	dev->mode_config.preferred_depth = 16;
+	dev->mode_config.prefer_shadow = 0;
+	dev->mode_config.funcs = &cirrus_mode_config_funcs;
+}
+
+/* ------------------------------------------------------------------ */
+
+DEFINE_DRM_GEM_SHMEM_FOPS(cirrus_fops);
+
+static struct drm_driver cirrus_driver = {
+	.driver_features = DRIVER_MODESET | DRIVER_GEM | DRIVER_ATOMIC | DRIVER_PRIME,
+
+	.name		 = DRIVER_NAME,
+	.desc		 = DRIVER_DESC,
+	.date		 = DRIVER_DATE,
+	.major		 = DRIVER_MAJOR,
+	.minor		 = DRIVER_MINOR,
+
+	.fops		 = &cirrus_fops,
+	DRM_GEM_SHMEM_DRIVER_OPS,
+};
+
+static int cirrus_pci_probe(struct pci_dev *pdev,
+			    const struct pci_device_id *ent)
+{
+	struct drm_device *dev;
+	struct cirrus_device *cirrus;
+	int ret;
+
+	ret = drm_fb_helper_remove_conflicting_pci_framebuffers(pdev, 0, "cirrusdrmfb");
+	if (ret)
+		return ret;
+
+	ret = pci_enable_device(pdev);
+	if (ret)
+		return ret;
+
+	ret = pci_request_regions(pdev, DRIVER_NAME);
+	if (ret)
+		return ret;
+
+	ret = -ENOMEM;
+	cirrus = kzalloc(sizeof(*cirrus), GFP_KERNEL);
+	if (cirrus == NULL)
+		goto err_pci_release;
+
+	dev = &cirrus->dev;
+	ret = drm_dev_init(dev, &cirrus_driver, &pdev->dev);
+	if (ret)
+		goto err_free_cirrus;
+	dev->dev_private = cirrus;
+
+	ret = -ENOMEM;
+	cirrus->vram = ioremap(pci_resource_start(pdev, 0),
+			       pci_resource_len(pdev, 0));
+	if (cirrus->vram == NULL)
+		goto err_dev_put;
+
+	cirrus->mmio = ioremap(pci_resource_start(pdev, 1),
+			       pci_resource_len(pdev, 1));
+	if (cirrus->mmio == NULL)
+		goto err_unmap_vram;
+
+	cirrus_mode_config_init(cirrus);
+
+	ret = cirrus_conn_init(cirrus);
+	if (ret < 0)
+		goto err_cleanup;
+
+	ret = cirrus_pipe_init(cirrus);
+	if (ret < 0)
+		goto err_cleanup;
+
+	drm_mode_config_reset(dev);
+
+	dev->pdev = pdev;
+	pci_set_drvdata(pdev, dev);
+	ret = drm_dev_register(dev, 0);
+	if (ret)
+		goto err_cleanup;
+
+	drm_fbdev_generic_setup(dev, dev->mode_config.preferred_depth);
+	return 0;
+
+err_cleanup:
+	drm_mode_config_cleanup(dev);
+	iounmap(cirrus->mmio);
+err_unmap_vram:
+	iounmap(cirrus->vram);
+err_dev_put:
+	drm_dev_put(dev);
+err_free_cirrus:
+	kfree(cirrus);
+err_pci_release:
+	pci_release_regions(pdev);
+	return ret;
+}
+
+static void cirrus_pci_remove(struct pci_dev *pdev)
+{
+	struct drm_device *dev = pci_get_drvdata(pdev);
+	struct cirrus_device *cirrus = dev->dev_private;
+
+	drm_dev_unregister(dev);
+	drm_mode_config_cleanup(dev);
+	iounmap(cirrus->mmio);
+	iounmap(cirrus->vram);
+	drm_dev_put(dev);
+	kfree(cirrus);
+	pci_release_regions(pdev);
+}
+
+static const struct pci_device_id pciidlist[] = {
+	{
+		.vendor    = PCI_VENDOR_ID_CIRRUS,
+		.device    = PCI_DEVICE_ID_CIRRUS_5446,
+		/* only bind to the cirrus chip in qemu */
+		.subvendor = PCI_SUBVENDOR_ID_REDHAT_QUMRANET,
+		.subdevice = PCI_SUBDEVICE_ID_QEMU,
+	}, {
+		.vendor    = PCI_VENDOR_ID_CIRRUS,
+		.device    = PCI_DEVICE_ID_CIRRUS_5446,
+		.subvendor = PCI_VENDOR_ID_XEN,
+		.subdevice = 0x0001,
+	},
+	{ /* end if list */ }
+};
+
+static struct pci_driver cirrus_pci_driver = {
+	.name = DRIVER_NAME,
+	.id_table = pciidlist,
+	.probe = cirrus_pci_probe,
+	.remove = cirrus_pci_remove,
+};
+
+static int __init cirrus_init(void)
+{
+	if (vgacon_text_force())
+		return -EINVAL;
+	return pci_register_driver(&cirrus_pci_driver);
+}
+
+static void __exit cirrus_exit(void)
+{
+	pci_unregister_driver(&cirrus_pci_driver);
+}
+
+module_init(cirrus_init);
+module_exit(cirrus_exit);
+
+MODULE_DEVICE_TABLE(pci, pciidlist);
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/cirrus/cirrus_drv.c b/drivers/gpu/drm/cirrus/cirrus_drv.c
deleted file mode 100644
index 8ec880f3a322..000000000000
--- a/drivers/gpu/drm/cirrus/cirrus_drv.c
+++ /dev/null
@@ -1,161 +0,0 @@
-/*
- * Copyright 2012 Red Hat <mjg@redhat.com>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License version 2. See the file COPYING in the main
- * directory of this archive for more details.
- *
- * Authors: Matthew Garrett
- *          Dave Airlie
- */
-#include <linux/module.h>
-#include <linux/console.h>
-#include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
-#include <drm/drm_probe_helper.h>
-
-#include "cirrus_drv.h"
-
-int cirrus_modeset = -1;
-int cirrus_bpp = 16;
-
-MODULE_PARM_DESC(modeset, "Disable/Enable modesetting");
-module_param_named(modeset, cirrus_modeset, int, 0400);
-MODULE_PARM_DESC(bpp, "Max bits-per-pixel (default:16)");
-module_param_named(bpp, cirrus_bpp, int, 0400);
-
-/*
- * This is the generic driver code. This binds the driver to the drm core,
- * which then performs further device association and calls our graphics init
- * functions
- */
-
-static struct drm_driver driver;
-
-/* only bind to the cirrus chip in qemu */
-static const struct pci_device_id pciidlist[] = {
-	{ PCI_VENDOR_ID_CIRRUS, PCI_DEVICE_ID_CIRRUS_5446,
-	  PCI_SUBVENDOR_ID_REDHAT_QUMRANET, PCI_SUBDEVICE_ID_QEMU,
-	  0, 0, 0 },
-	{ PCI_VENDOR_ID_CIRRUS, PCI_DEVICE_ID_CIRRUS_5446, PCI_VENDOR_ID_XEN,
-	  0x0001, 0, 0, 0 },
-	{0,}
-};
-
-
-static int cirrus_pci_probe(struct pci_dev *pdev,
-			    const struct pci_device_id *ent)
-{
-	int ret;
-
-	ret = drm_fb_helper_remove_conflicting_pci_framebuffers(pdev, 0, "cirrusdrmfb");
-	if (ret)
-		return ret;
-
-	return drm_get_pci_dev(pdev, ent, &driver);
-}
-
-static void cirrus_pci_remove(struct pci_dev *pdev)
-{
-	struct drm_device *dev = pci_get_drvdata(pdev);
-
-	drm_put_dev(dev);
-}
-
-#ifdef CONFIG_PM_SLEEP
-static int cirrus_pm_suspend(struct device *dev)
-{
-	struct pci_dev *pdev = to_pci_dev(dev);
-	struct drm_device *drm_dev = pci_get_drvdata(pdev);
-	struct cirrus_device *cdev = drm_dev->dev_private;
-
-	drm_kms_helper_poll_disable(drm_dev);
-
-	if (cdev->mode_info.gfbdev) {
-		console_lock();
-		drm_fb_helper_set_suspend(&cdev->mode_info.gfbdev->helper, 1);
-		console_unlock();
-	}
-
-	return 0;
-}
-
-static int cirrus_pm_resume(struct device *dev)
-{
-	struct pci_dev *pdev = to_pci_dev(dev);
-	struct drm_device *drm_dev = pci_get_drvdata(pdev);
-	struct cirrus_device *cdev = drm_dev->dev_private;
-
-	drm_helper_resume_force_mode(drm_dev);
-
-	if (cdev->mode_info.gfbdev) {
-		console_lock();
-		drm_fb_helper_set_suspend(&cdev->mode_info.gfbdev->helper, 0);
-		console_unlock();
-	}
-
-	drm_kms_helper_poll_enable(drm_dev);
-	return 0;
-}
-#endif
-
-static const struct file_operations cirrus_driver_fops = {
-	.owner = THIS_MODULE,
-	.open = drm_open,
-	.release = drm_release,
-	.unlocked_ioctl = drm_ioctl,
-	.mmap = cirrus_mmap,
-	.poll = drm_poll,
-	.compat_ioctl = drm_compat_ioctl,
-};
-static struct drm_driver driver = {
-	.driver_features = DRIVER_MODESET | DRIVER_GEM,
-	.load = cirrus_driver_load,
-	.unload = cirrus_driver_unload,
-	.fops = &cirrus_driver_fops,
-	.name = DRIVER_NAME,
-	.desc = DRIVER_DESC,
-	.date = DRIVER_DATE,
-	.major = DRIVER_MAJOR,
-	.minor = DRIVER_MINOR,
-	.patchlevel = DRIVER_PATCHLEVEL,
-	.gem_free_object_unlocked = cirrus_gem_free_object,
-	.dumb_create = cirrus_dumb_create,
-	.dumb_map_offset = cirrus_dumb_mmap_offset,
-};
-
-static const struct dev_pm_ops cirrus_pm_ops = {
-	SET_SYSTEM_SLEEP_PM_OPS(cirrus_pm_suspend,
-				cirrus_pm_resume)
-};
-
-static struct pci_driver cirrus_pci_driver = {
-	.name = DRIVER_NAME,
-	.id_table = pciidlist,
-	.probe = cirrus_pci_probe,
-	.remove = cirrus_pci_remove,
-	.driver.pm = &cirrus_pm_ops,
-};
-
-static int __init cirrus_init(void)
-{
-	if (vgacon_text_force() && cirrus_modeset == -1)
-		return -EINVAL;
-
-	if (cirrus_modeset == 0)
-		return -EINVAL;
-	return pci_register_driver(&cirrus_pci_driver);
-}
-
-static void __exit cirrus_exit(void)
-{
-	pci_unregister_driver(&cirrus_pci_driver);
-}
-
-module_init(cirrus_init);
-module_exit(cirrus_exit);
-
-MODULE_DEVICE_TABLE(pci, pciidlist);
-MODULE_AUTHOR(DRIVER_AUTHOR);
-MODULE_DESCRIPTION(DRIVER_DESC);
-MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/cirrus/cirrus_drv.h b/drivers/gpu/drm/cirrus/cirrus_drv.h
index f2b2e0d169fa..1bd816be3aae 100644
--- a/drivers/gpu/drm/cirrus/cirrus_drv.h
+++ b/drivers/gpu/drm/cirrus/cirrus_drv.h
@@ -101,7 +101,6 @@ struct cirrus_crtc {
 
 struct cirrus_fbdev;
 struct cirrus_mode_info {
-	bool				mode_config_initialized;
 	struct cirrus_crtc		*crtc;
 	/* pointer to fbdev info structure */
 	struct cirrus_fbdev		*gfbdev;
@@ -143,7 +142,7 @@ struct cirrus_device {
 
 
 struct cirrus_fbdev {
-	struct drm_fb_helper helper;
+	struct drm_fb_helper helper; /* must be first */
 	struct drm_framebuffer *gfb;
 	void *sysram;
 	int size;
@@ -169,7 +168,6 @@ cirrus_bo(struct ttm_buffer_object *bo)
 
 
 #define to_cirrus_obj(x) container_of(x, struct cirrus_gem_object, base)
-#define DRM_FILE_PAGE_OFFSET (0x100000000ULL >> PAGE_SHIFT)
 
 				/* cirrus_main.c */
 int cirrus_device_init(struct cirrus_device *cdev,
diff --git a/drivers/gpu/drm/cirrus/cirrus_fbdev.c b/drivers/gpu/drm/cirrus/cirrus_fbdev.c
deleted file mode 100644
index 39df62acac69..000000000000
--- a/drivers/gpu/drm/cirrus/cirrus_fbdev.c
+++ /dev/null
@@ -1,315 +0,0 @@
-/*
- * Copyright 2012 Red Hat
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License version 2. See the file COPYING in the main
- * directory of this archive for more details.
- *
- * Authors: Matthew Garrett
- *          Dave Airlie
- */
-#include <linux/module.h>
-#include <drm/drmP.h>
-#include <drm/drm_util.h>
-#include <drm/drm_fb_helper.h>
-#include <drm/drm_crtc_helper.h>
-
-#include "cirrus_drv.h"
-
-static void cirrus_dirty_update(struct cirrus_fbdev *afbdev,
-			     int x, int y, int width, int height)
-{
-	int i;
-	struct drm_gem_object *obj;
-	struct cirrus_bo *bo;
-	int src_offset, dst_offset;
-	int bpp = afbdev->gfb->format->cpp[0];
-	int ret = -EBUSY;
-	bool unmap = false;
-	bool store_for_later = false;
-	int x2, y2;
-	unsigned long flags;
-
-	obj = afbdev->gfb->obj[0];
-	bo = gem_to_cirrus_bo(obj);
-
-	/*
-	 * try and reserve the BO, if we fail with busy
-	 * then the BO is being moved and we should
-	 * store up the damage until later.
-	 */
-	if (drm_can_sleep())
-		ret = cirrus_bo_reserve(bo, true);
-	if (ret) {
-		if (ret != -EBUSY)
-			return;
-		store_for_later = true;
-	}
-
-	x2 = x + width - 1;
-	y2 = y + height - 1;
-	spin_lock_irqsave(&afbdev->dirty_lock, flags);
-
-	if (afbdev->y1 < y)
-		y = afbdev->y1;
-	if (afbdev->y2 > y2)
-		y2 = afbdev->y2;
-	if (afbdev->x1 < x)
-		x = afbdev->x1;
-	if (afbdev->x2 > x2)
-		x2 = afbdev->x2;
-
-	if (store_for_later) {
-		afbdev->x1 = x;
-		afbdev->x2 = x2;
-		afbdev->y1 = y;
-		afbdev->y2 = y2;
-		spin_unlock_irqrestore(&afbdev->dirty_lock, flags);
-		return;
-	}
-
-	afbdev->x1 = afbdev->y1 = INT_MAX;
-	afbdev->x2 = afbdev->y2 = 0;
-	spin_unlock_irqrestore(&afbdev->dirty_lock, flags);
-
-	if (!bo->kmap.virtual) {
-		ret = ttm_bo_kmap(&bo->bo, 0, bo->bo.num_pages, &bo->kmap);
-		if (ret) {
-			DRM_ERROR("failed to kmap fb updates\n");
-			cirrus_bo_unreserve(bo);
-			return;
-		}
-		unmap = true;
-	}
-	for (i = y; i < y + height; i++) {
-		/* assume equal stride for now */
-		src_offset = dst_offset = i * afbdev->gfb->pitches[0] + (x * bpp);
-		memcpy_toio(bo->kmap.virtual + src_offset, afbdev->sysram + src_offset, width * bpp);
-
-	}
-	if (unmap)
-		ttm_bo_kunmap(&bo->kmap);
-
-	cirrus_bo_unreserve(bo);
-}
-
-static void cirrus_fillrect(struct fb_info *info,
-			 const struct fb_fillrect *rect)
-{
-	struct cirrus_fbdev *afbdev = info->par;
-	drm_fb_helper_sys_fillrect(info, rect);
-	cirrus_dirty_update(afbdev, rect->dx, rect->dy, rect->width,
-			 rect->height);
-}
-
-static void cirrus_copyarea(struct fb_info *info,
-			 const struct fb_copyarea *area)
-{
-	struct cirrus_fbdev *afbdev = info->par;
-	drm_fb_helper_sys_copyarea(info, area);
-	cirrus_dirty_update(afbdev, area->dx, area->dy, area->width,
-			 area->height);
-}
-
-static void cirrus_imageblit(struct fb_info *info,
-			  const struct fb_image *image)
-{
-	struct cirrus_fbdev *afbdev = info->par;
-	drm_fb_helper_sys_imageblit(info, image);
-	cirrus_dirty_update(afbdev, image->dx, image->dy, image->width,
-			 image->height);
-}
-
-
-static struct fb_ops cirrusfb_ops = {
-	.owner = THIS_MODULE,
-	.fb_check_var = drm_fb_helper_check_var,
-	.fb_set_par = drm_fb_helper_set_par,
-	.fb_fillrect = cirrus_fillrect,
-	.fb_copyarea = cirrus_copyarea,
-	.fb_imageblit = cirrus_imageblit,
-	.fb_pan_display = drm_fb_helper_pan_display,
-	.fb_blank = drm_fb_helper_blank,
-	.fb_setcmap = drm_fb_helper_setcmap,
-};
-
-static int cirrusfb_create_object(struct cirrus_fbdev *afbdev,
-			       const struct drm_mode_fb_cmd2 *mode_cmd,
-			       struct drm_gem_object **gobj_p)
-{
-	struct drm_device *dev = afbdev->helper.dev;
-	struct cirrus_device *cdev = dev->dev_private;
-	u32 bpp;
-	u32 size;
-	struct drm_gem_object *gobj;
-	int ret = 0;
-
-	bpp = drm_format_plane_cpp(mode_cmd->pixel_format, 0) * 8;
-
-	if (!cirrus_check_framebuffer(cdev, mode_cmd->width, mode_cmd->height,
-				      bpp, mode_cmd->pitches[0]))
-		return -EINVAL;
-
-	size = mode_cmd->pitches[0] * mode_cmd->height;
-	ret = cirrus_gem_create(dev, size, true, &gobj);
-	if (ret)
-		return ret;
-
-	*gobj_p = gobj;
-	return ret;
-}
-
-static int cirrusfb_create(struct drm_fb_helper *helper,
-			   struct drm_fb_helper_surface_size *sizes)
-{
-	struct cirrus_fbdev *gfbdev =
-		container_of(helper, struct cirrus_fbdev, helper);
-	struct cirrus_device *cdev = gfbdev->helper.dev->dev_private;
-	struct fb_info *info;
-	struct drm_framebuffer *fb;
-	struct drm_mode_fb_cmd2 mode_cmd;
-	void *sysram;
-	struct drm_gem_object *gobj = NULL;
-	int size, ret;
-
-	mode_cmd.width = sizes->surface_width;
-	mode_cmd.height = sizes->surface_height;
-	mode_cmd.pitches[0] = mode_cmd.width * ((sizes->surface_bpp + 7) / 8);
-	mode_cmd.pixel_format = drm_mode_legacy_fb_format(sizes->surface_bpp,
-							  sizes->surface_depth);
-	size = mode_cmd.pitches[0] * mode_cmd.height;
-
-	ret = cirrusfb_create_object(gfbdev, &mode_cmd, &gobj);
-	if (ret) {
-		DRM_ERROR("failed to create fbcon backing object %d\n", ret);
-		return ret;
-	}
-
-	sysram = vmalloc(size);
-	if (!sysram)
-		return -ENOMEM;
-
-	info = drm_fb_helper_alloc_fbi(helper);
-	if (IS_ERR(info)) {
-		ret = PTR_ERR(info);
-		goto err_vfree;
-	}
-
-	info->par = gfbdev;
-
-	fb = kzalloc(sizeof(*fb), GFP_KERNEL);
-	if (!fb) {
-		ret = -ENOMEM;
-		goto err_drm_gem_object_put_unlocked;
-	}
-
-	ret = cirrus_framebuffer_init(cdev->dev, fb, &mode_cmd, gobj);
-	if (ret)
-		goto err_kfree;
-
-	gfbdev->sysram = sysram;
-	gfbdev->size = size;
-	gfbdev->gfb = fb;
-
-	/* setup helper */
-	gfbdev->helper.fb = fb;
-
-	strcpy(info->fix.id, "cirrusdrmfb");
-
-	info->fbops = &cirrusfb_ops;
-
-	drm_fb_helper_fill_fix(info, fb->pitches[0], fb->format->depth);
-	drm_fb_helper_fill_var(info, &gfbdev->helper, sizes->fb_width,
-			       sizes->fb_height);
-
-	/* setup aperture base/size for vesafb takeover */
-	info->apertures->ranges[0].base = cdev->dev->mode_config.fb_base;
-	info->apertures->ranges[0].size = cdev->mc.vram_size;
-
-	info->fix.smem_start = cdev->dev->mode_config.fb_base;
-	info->fix.smem_len = cdev->mc.vram_size;
-
-	info->screen_base = sysram;
-	info->screen_size = size;
-
-	info->fix.mmio_start = 0;
-	info->fix.mmio_len = 0;
-
-	DRM_INFO("fb mappable at 0x%lX\n", info->fix.smem_start);
-	DRM_INFO("vram aper at 0x%lX\n", (unsigned long)info->fix.smem_start);
-	DRM_INFO("size %lu\n", (unsigned long)info->fix.smem_len);
-	DRM_INFO("fb depth is %d\n", fb->format->depth);
-	DRM_INFO("   pitch is %d\n", fb->pitches[0]);
-
-	return 0;
-
-err_kfree:
-	kfree(fb);
-err_drm_gem_object_put_unlocked:
-	drm_gem_object_put_unlocked(gobj);
-err_vfree:
-	vfree(sysram);
-	return ret;
-}
-
-static int cirrus_fbdev_destroy(struct drm_device *dev,
-				struct cirrus_fbdev *gfbdev)
-{
-	struct drm_framebuffer *gfb = gfbdev->gfb;
-
-	drm_helper_force_disable_all(dev);
-
-	drm_fb_helper_unregister_fbi(&gfbdev->helper);
-
-	vfree(gfbdev->sysram);
-	drm_fb_helper_fini(&gfbdev->helper);
-	if (gfb)
-		drm_framebuffer_put(gfb);
-
-	return 0;
-}
-
-static const struct drm_fb_helper_funcs cirrus_fb_helper_funcs = {
-	.fb_probe = cirrusfb_create,
-};
-
-int cirrus_fbdev_init(struct cirrus_device *cdev)
-{
-	struct cirrus_fbdev *gfbdev;
-	int ret;
-
-	/*bpp_sel = 8;*/
-	gfbdev = kzalloc(sizeof(struct cirrus_fbdev), GFP_KERNEL);
-	if (!gfbdev)
-		return -ENOMEM;
-
-	cdev->mode_info.gfbdev = gfbdev;
-	spin_lock_init(&gfbdev->dirty_lock);
-
-	drm_fb_helper_prepare(cdev->dev, &gfbdev->helper,
-			      &cirrus_fb_helper_funcs);
-
-	ret = drm_fb_helper_init(cdev->dev, &gfbdev->helper,
-				 CIRRUSFB_CONN_LIMIT);
-	if (ret)
-		return ret;
-
-	ret = drm_fb_helper_single_add_all_connectors(&gfbdev->helper);
-	if (ret)
-		return ret;
-
-	/* disable all the possible outputs/crtcs before entering KMS mode */
-	drm_helper_disable_unused_functions(cdev->dev);
-
-	return drm_fb_helper_initial_config(&gfbdev->helper, cirrus_bpp);
-}
-
-void cirrus_fbdev_fini(struct cirrus_device *cdev)
-{
-	if (!cdev->mode_info.gfbdev)
-		return;
-
-	cirrus_fbdev_destroy(cdev->dev, cdev->mode_info.gfbdev);
-	kfree(cdev->mode_info.gfbdev);
-	cdev->mode_info.gfbdev = NULL;
-}
diff --git a/drivers/gpu/drm/cirrus/cirrus_main.c b/drivers/gpu/drm/cirrus/cirrus_main.c
deleted file mode 100644
index 57f8fe6d020b..000000000000
--- a/drivers/gpu/drm/cirrus/cirrus_main.c
+++ /dev/null
@@ -1,328 +0,0 @@
-/*
- * Copyright 2012 Red Hat
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License version 2. See the file COPYING in the main
- * directory of this archive for more details.
- *
- * Authors: Matthew Garrett
- *          Dave Airlie
- */
-#include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
-#include <drm/drm_gem_framebuffer_helper.h>
-
-#include "cirrus_drv.h"
-
-static const struct drm_framebuffer_funcs cirrus_fb_funcs = {
-	.create_handle = drm_gem_fb_create_handle,
-	.destroy = drm_gem_fb_destroy,
-};
-
-int cirrus_framebuffer_init(struct drm_device *dev,
-			    struct drm_framebuffer *gfb,
-			    const struct drm_mode_fb_cmd2 *mode_cmd,
-			    struct drm_gem_object *obj)
-{
-	int ret;
-
-	drm_helper_mode_fill_fb_struct(dev, gfb, mode_cmd);
-	gfb->obj[0] = obj;
-	ret = drm_framebuffer_init(dev, gfb, &cirrus_fb_funcs);
-	if (ret) {
-		DRM_ERROR("drm_framebuffer_init failed: %d\n", ret);
-		return ret;
-	}
-	return 0;
-}
-
-static struct drm_framebuffer *
-cirrus_user_framebuffer_create(struct drm_device *dev,
-			       struct drm_file *filp,
-			       const struct drm_mode_fb_cmd2 *mode_cmd)
-{
-	struct cirrus_device *cdev = dev->dev_private;
-	struct drm_gem_object *obj;
-	struct drm_framebuffer *fb;
-	u32 bpp;
-	int ret;
-
-	bpp = drm_format_plane_cpp(mode_cmd->pixel_format, 0) * 8;
-
-	if (!cirrus_check_framebuffer(cdev, mode_cmd->width, mode_cmd->height,
-				      bpp, mode_cmd->pitches[0]))
-		return ERR_PTR(-EINVAL);
-
-	obj = drm_gem_object_lookup(filp, mode_cmd->handles[0]);
-	if (obj == NULL)
-		return ERR_PTR(-ENOENT);
-
-	fb = kzalloc(sizeof(*fb), GFP_KERNEL);
-	if (!fb) {
-		drm_gem_object_put_unlocked(obj);
-		return ERR_PTR(-ENOMEM);
-	}
-
-	ret = cirrus_framebuffer_init(dev, fb, mode_cmd, obj);
-	if (ret) {
-		drm_gem_object_put_unlocked(obj);
-		kfree(fb);
-		return ERR_PTR(ret);
-	}
-	return fb;
-}
-
-static const struct drm_mode_config_funcs cirrus_mode_funcs = {
-	.fb_create = cirrus_user_framebuffer_create,
-};
-
-/* Unmap the framebuffer from the core and release the memory */
-static void cirrus_vram_fini(struct cirrus_device *cdev)
-{
-	iounmap(cdev->rmmio);
-	cdev->rmmio = NULL;
-	if (cdev->mc.vram_base)
-		release_mem_region(cdev->mc.vram_base, cdev->mc.vram_size);
-}
-
-/* Map the framebuffer from the card and configure the core */
-static int cirrus_vram_init(struct cirrus_device *cdev)
-{
-	/* BAR 0 is VRAM */
-	cdev->mc.vram_base = pci_resource_start(cdev->dev->pdev, 0);
-	cdev->mc.vram_size = pci_resource_len(cdev->dev->pdev, 0);
-
-	if (!request_mem_region(cdev->mc.vram_base, cdev->mc.vram_size,
-				"cirrusdrmfb_vram")) {
-		DRM_ERROR("can't reserve VRAM\n");
-		return -ENXIO;
-	}
-
-	return 0;
-}
-
-/*
- * Our emulated hardware has two sets of memory. One is video RAM and can
- * simply be used as a linear framebuffer - the other provides mmio access
- * to the display registers. The latter can also be accessed via IO port
- * access, but we map the range and use mmio to program them instead
- */
-
-int cirrus_device_init(struct cirrus_device *cdev,
-		       struct drm_device *ddev,
-		       struct pci_dev *pdev, uint32_t flags)
-{
-	int ret;
-
-	cdev->dev = ddev;
-	cdev->flags = flags;
-
-	/* Hardcode the number of CRTCs to 1 */
-	cdev->num_crtc = 1;
-
-	/* BAR 0 is the framebuffer, BAR 1 contains registers */
-	cdev->rmmio_base = pci_resource_start(cdev->dev->pdev, 1);
-	cdev->rmmio_size = pci_resource_len(cdev->dev->pdev, 1);
-
-	if (!request_mem_region(cdev->rmmio_base, cdev->rmmio_size,
-				"cirrusdrmfb_mmio")) {
-		DRM_ERROR("can't reserve mmio registers\n");
-		return -ENOMEM;
-	}
-
-	cdev->rmmio = ioremap(cdev->rmmio_base, cdev->rmmio_size);
-
-	if (cdev->rmmio == NULL)
-		return -ENOMEM;
-
-	ret = cirrus_vram_init(cdev);
-	if (ret) {
-		release_mem_region(cdev->rmmio_base, cdev->rmmio_size);
-		return ret;
-	}
-
-	return 0;
-}
-
-void cirrus_device_fini(struct cirrus_device *cdev)
-{
-	release_mem_region(cdev->rmmio_base, cdev->rmmio_size);
-	cirrus_vram_fini(cdev);
-}
-
-/*
- * Functions here will be called by the core once it's bound the driver to
- * a PCI device
- */
-
-int cirrus_driver_load(struct drm_device *dev, unsigned long flags)
-{
-	struct cirrus_device *cdev;
-	int r;
-
-	cdev = kzalloc(sizeof(struct cirrus_device), GFP_KERNEL);
-	if (cdev == NULL)
-		return -ENOMEM;
-	dev->dev_private = (void *)cdev;
-
-	r = cirrus_device_init(cdev, dev, dev->pdev, flags);
-	if (r) {
-		dev_err(&dev->pdev->dev, "Fatal error during GPU init: %d\n", r);
-		goto out;
-	}
-
-	r = cirrus_mm_init(cdev);
-	if (r) {
-		dev_err(&dev->pdev->dev, "fatal err on mm init\n");
-		goto out;
-	}
-
-	/*
-	 * cirrus_modeset_init() is initializing/registering the emulated fbdev
-	 * and DRM internals can access/test some of the fields in
-	 * mode_config->funcs as part of the fbdev registration process.
-	 * Make sure dev->mode_config.funcs is properly set to avoid
-	 * dereferencing a NULL pointer.
-	 * FIXME: mode_config.funcs assignment should probably be done in
-	 * cirrus_modeset_init() (that's a common pattern seen in other DRM
-	 * drivers).
-	 */
-	dev->mode_config.funcs = &cirrus_mode_funcs;
-	r = cirrus_modeset_init(cdev);
-	if (r) {
-		dev_err(&dev->pdev->dev, "Fatal error during modeset init: %d\n", r);
-		goto out;
-	}
-
-	return 0;
-out:
-	cirrus_driver_unload(dev);
-	return r;
-}
-
-void cirrus_driver_unload(struct drm_device *dev)
-{
-	struct cirrus_device *cdev = dev->dev_private;
-
-	if (cdev == NULL)
-		return;
-	cirrus_modeset_fini(cdev);
-	cirrus_mm_fini(cdev);
-	cirrus_device_fini(cdev);
-	kfree(cdev);
-	dev->dev_private = NULL;
-}
-
-int cirrus_gem_create(struct drm_device *dev,
-		   u32 size, bool iskernel,
-		   struct drm_gem_object **obj)
-{
-	struct cirrus_bo *cirrusbo;
-	int ret;
-
-	*obj = NULL;
-
-	size = roundup(size, PAGE_SIZE);
-	if (size == 0)
-		return -EINVAL;
-
-	ret = cirrus_bo_create(dev, size, 0, 0, &cirrusbo);
-	if (ret) {
-		if (ret != -ERESTARTSYS)
-			DRM_ERROR("failed to allocate GEM object\n");
-		return ret;
-	}
-	*obj = &cirrusbo->gem;
-	return 0;
-}
-
-int cirrus_dumb_create(struct drm_file *file,
-		    struct drm_device *dev,
-		    struct drm_mode_create_dumb *args)
-{
-	int ret;
-	struct drm_gem_object *gobj;
-	u32 handle;
-
-	args->pitch = args->width * ((args->bpp + 7) / 8);
-	args->size = args->pitch * args->height;
-
-	ret = cirrus_gem_create(dev, args->size, false,
-			     &gobj);
-	if (ret)
-		return ret;
-
-	ret = drm_gem_handle_create(file, gobj, &handle);
-	drm_gem_object_put_unlocked(gobj);
-	if (ret)
-		return ret;
-
-	args->handle = handle;
-	return 0;
-}
-
-static void cirrus_bo_unref(struct cirrus_bo **bo)
-{
-	struct ttm_buffer_object *tbo;
-
-	if ((*bo) == NULL)
-		return;
-
-	tbo = &((*bo)->bo);
-	ttm_bo_put(tbo);
-	*bo = NULL;
-}
-
-void cirrus_gem_free_object(struct drm_gem_object *obj)
-{
-	struct cirrus_bo *cirrus_bo = gem_to_cirrus_bo(obj);
-
-	cirrus_bo_unref(&cirrus_bo);
-}
-
-
-static inline u64 cirrus_bo_mmap_offset(struct cirrus_bo *bo)
-{
-	return drm_vma_node_offset_addr(&bo->bo.vma_node);
-}
-
-int
-cirrus_dumb_mmap_offset(struct drm_file *file,
-		     struct drm_device *dev,
-		     uint32_t handle,
-		     uint64_t *offset)
-{
-	struct drm_gem_object *obj;
-	struct cirrus_bo *bo;
-
-	obj = drm_gem_object_lookup(file, handle);
-	if (obj == NULL)
-		return -ENOENT;
-
-	bo = gem_to_cirrus_bo(obj);
-	*offset = cirrus_bo_mmap_offset(bo);
-
-	drm_gem_object_put_unlocked(obj);
-
-	return 0;
-}
-
-bool cirrus_check_framebuffer(struct cirrus_device *cdev, int width, int height,
-			      int bpp, int pitch)
-{
-	const int max_pitch = 0x1FF << 3; /* (4096 - 1) & ~111b bytes */
-	const int max_size = cdev->mc.vram_size;
-
-	if (bpp > cirrus_bpp)
-		return false;
-	if (bpp > 32)
-		return false;
-
-	if (pitch > max_pitch)
-		return false;
-
-	if (pitch * height > max_size)
-		return false;
-
-	return true;
-}
diff --git a/drivers/gpu/drm/cirrus/cirrus_mode.c b/drivers/gpu/drm/cirrus/cirrus_mode.c
deleted file mode 100644
index 7f9bc32af685..000000000000
--- a/drivers/gpu/drm/cirrus/cirrus_mode.c
+++ /dev/null
@@ -1,621 +0,0 @@
-
-/*
- * Copyright 2012 Red Hat
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License version 2. See the file COPYING in the main
- * directory of this archive for more details.
- *
- * Authors: Matthew Garrett
- *          Dave Airlie
- *
- * Portions of this code derived from cirrusfb.c:
- * drivers/video/cirrusfb.c - driver for Cirrus Logic chipsets
- *
- * Copyright 1999-2001 Jeff Garzik <jgarzik@pobox.com>
- */
-#include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
-#include <drm/drm_plane_helper.h>
-#include <drm/drm_probe_helper.h>
-
-#include <video/cirrus.h>
-
-#include "cirrus_drv.h"
-
-#define CIRRUS_LUT_SIZE 256
-
-#define PALETTE_INDEX 0x8
-#define PALETTE_DATA 0x9
-
-/*
- * This file contains setup code for the CRTC.
- */
-
-/*
- * The DRM core requires DPMS functions, but they make little sense in our
- * case and so are just stubs
- */
-
-static void cirrus_crtc_dpms(struct drm_crtc *crtc, int mode)
-{
-	struct drm_device *dev = crtc->dev;
-	struct cirrus_device *cdev = dev->dev_private;
-	u8 sr01, gr0e;
-
-	switch (mode) {
-	case DRM_MODE_DPMS_ON:
-		sr01 = 0x00;
-		gr0e = 0x00;
-		break;
-	case DRM_MODE_DPMS_STANDBY:
-		sr01 = 0x20;
-		gr0e = 0x02;
-		break;
-	case DRM_MODE_DPMS_SUSPEND:
-		sr01 = 0x20;
-		gr0e = 0x04;
-		break;
-	case DRM_MODE_DPMS_OFF:
-		sr01 = 0x20;
-		gr0e = 0x06;
-		break;
-	default:
-		return;
-	}
-
-	WREG8(SEQ_INDEX, 0x1);
-	sr01 |= RREG8(SEQ_DATA) & ~0x20;
-	WREG_SEQ(0x1, sr01);
-
-	WREG8(GFX_INDEX, 0xe);
-	gr0e |= RREG8(GFX_DATA) & ~0x06;
-	WREG_GFX(0xe, gr0e);
-}
-
-static void cirrus_set_start_address(struct drm_crtc *crtc, unsigned offset)
-{
-	struct cirrus_device *cdev = crtc->dev->dev_private;
-	u32 addr;
-	u8 tmp;
-
-	addr = offset >> 2;
-	WREG_CRT(0x0c, (u8)((addr >> 8) & 0xff));
-	WREG_CRT(0x0d, (u8)(addr & 0xff));
-
-	WREG8(CRT_INDEX, 0x1b);
-	tmp = RREG8(CRT_DATA);
-	tmp &= 0xf2;
-	tmp |= (addr >> 16) & 0x01;
-	tmp |= (addr >> 15) & 0x0c;
-	WREG_CRT(0x1b, tmp);
-	WREG8(CRT_INDEX, 0x1d);
-	tmp = RREG8(CRT_DATA);
-	tmp &= 0x7f;
-	tmp |= (addr >> 12) & 0x80;
-	WREG_CRT(0x1d, tmp);
-}
-
-/* cirrus is different - we will force move buffers out of VRAM */
-static int cirrus_crtc_do_set_base(struct drm_crtc *crtc,
-				struct drm_framebuffer *fb,
-				int x, int y, int atomic)
-{
-	struct cirrus_device *cdev = crtc->dev->dev_private;
-	struct cirrus_bo *bo;
-	int ret;
-	u64 gpu_addr;
-
-	/* push the previous fb to system ram */
-	if (!atomic && fb) {
-		bo = gem_to_cirrus_bo(fb->obj[0]);
-		ret = cirrus_bo_reserve(bo, false);
-		if (ret)
-			return ret;
-		cirrus_bo_push_sysram(bo);
-		cirrus_bo_unreserve(bo);
-	}
-
-	bo = gem_to_cirrus_bo(crtc->primary->fb->obj[0]);
-
-	ret = cirrus_bo_reserve(bo, false);
-	if (ret)
-		return ret;
-
-	ret = cirrus_bo_pin(bo, TTM_PL_FLAG_VRAM, &gpu_addr);
-	if (ret) {
-		cirrus_bo_unreserve(bo);
-		return ret;
-	}
-
-	if (cdev->mode_info.gfbdev->gfb == crtc->primary->fb) {
-		/* if pushing console in kmap it */
-		ret = ttm_bo_kmap(&bo->bo, 0, bo->bo.num_pages, &bo->kmap);
-		if (ret)
-			DRM_ERROR("failed to kmap fbcon\n");
-	}
-	cirrus_bo_unreserve(bo);
-
-	cirrus_set_start_address(crtc, (u32)gpu_addr);
-	return 0;
-}
-
-static int cirrus_crtc_mode_set_base(struct drm_crtc *crtc, int x, int y,
-			     struct drm_framebuffer *old_fb)
-{
-	return cirrus_crtc_do_set_base(crtc, old_fb, x, y, 0);
-}
-
-/*
- * The meat of this driver. The core passes us a mode and we have to program
- * it. The modesetting here is the bare minimum required to satisfy the qemu
- * emulation of this hardware, and running this against a real device is
- * likely to result in an inadequately programmed mode. We've already had
- * the opportunity to modify the mode, so whatever we receive here should
- * be something that can be correctly programmed and displayed
- */
-static int cirrus_crtc_mode_set(struct drm_crtc *crtc,
-				struct drm_display_mode *mode,
-				struct drm_display_mode *adjusted_mode,
-				int x, int y, struct drm_framebuffer *old_fb)
-{
-	struct drm_device *dev = crtc->dev;
-	struct cirrus_device *cdev = dev->dev_private;
-	const struct drm_framebuffer *fb = crtc->primary->fb;
-	int hsyncstart, hsyncend, htotal, hdispend;
-	int vtotal, vdispend;
-	int tmp;
-	int sr07 = 0, hdr = 0;
-
-	htotal = mode->htotal / 8;
-	hsyncend = mode->hsync_end / 8;
-	hsyncstart = mode->hsync_start / 8;
-	hdispend = mode->hdisplay / 8;
-
-	vtotal = mode->vtotal;
-	vdispend = mode->vdisplay;
-
-	vdispend -= 1;
-	vtotal -= 2;
-
-	htotal -= 5;
-	hdispend -= 1;
-	hsyncstart += 1;
-	hsyncend += 1;
-
-	WREG_CRT(VGA_CRTC_V_SYNC_END, 0x20);
-	WREG_CRT(VGA_CRTC_H_TOTAL, htotal);
-	WREG_CRT(VGA_CRTC_H_DISP, hdispend);
-	WREG_CRT(VGA_CRTC_H_SYNC_START, hsyncstart);
-	WREG_CRT(VGA_CRTC_H_SYNC_END, hsyncend);
-	WREG_CRT(VGA_CRTC_V_TOTAL, vtotal & 0xff);
-	WREG_CRT(VGA_CRTC_V_DISP_END, vdispend & 0xff);
-
-	tmp = 0x40;
-	if ((vdispend + 1) & 512)
-		tmp |= 0x20;
-	WREG_CRT(VGA_CRTC_MAX_SCAN, tmp);
-
-	/*
-	 * Overflow bits for values that don't fit in the standard registers
-	 */
-	tmp = 16;
-	if (vtotal & 256)
-		tmp |= 1;
-	if (vdispend & 256)
-		tmp |= 2;
-	if ((vdispend + 1) & 256)
-		tmp |= 8;
-	if (vtotal & 512)
-		tmp |= 32;
-	if (vdispend & 512)
-		tmp |= 64;
-	WREG_CRT(VGA_CRTC_OVERFLOW, tmp);
-
-	tmp = 0;
-
-	/* More overflow bits */
-
-	if ((htotal + 5) & 64)
-		tmp |= 16;
-	if ((htotal + 5) & 128)
-		tmp |= 32;
-	if (vtotal & 256)
-		tmp |= 64;
-	if (vtotal & 512)
-		tmp |= 128;
-
-	WREG_CRT(CL_CRT1A, tmp);
-
-	/* Disable Hercules/CGA compatibility */
-	WREG_CRT(VGA_CRTC_MODE, 0x03);
-
-	WREG8(SEQ_INDEX, 0x7);
-	sr07 = RREG8(SEQ_DATA);
-	sr07 &= 0xe0;
-	hdr = 0;
-	switch (fb->format->cpp[0] * 8) {
-	case 8:
-		sr07 |= 0x11;
-		break;
-	case 16:
-		sr07 |= 0x17;
-		hdr = 0xc1;
-		break;
-	case 24:
-		sr07 |= 0x15;
-		hdr = 0xc5;
-		break;
-	case 32:
-		sr07 |= 0x19;
-		hdr = 0xc5;
-		break;
-	default:
-		return -1;
-	}
-
-	WREG_SEQ(0x7, sr07);
-
-	/* Program the pitch */
-	tmp = fb->pitches[0] / 8;
-	WREG_CRT(VGA_CRTC_OFFSET, tmp);
-
-	/* Enable extended blanking and pitch bits, and enable full memory */
-	tmp = 0x22;
-	tmp |= (fb->pitches[0] >> 7) & 0x10;
-	tmp |= (fb->pitches[0] >> 6) & 0x40;
-	WREG_CRT(0x1b, tmp);
-
-	/* Enable high-colour modes */
-	WREG_GFX(VGA_GFX_MODE, 0x40);
-
-	/* And set graphics mode */
-	WREG_GFX(VGA_GFX_MISC, 0x01);
-
-	WREG_HDR(hdr);
-	cirrus_crtc_do_set_base(crtc, old_fb, x, y, 0);
-
-	/* Unblank (needed on S3 resume, vgabios doesn't do it then) */
-	outb(0x20, 0x3c0);
-	return 0;
-}
-
-/*
- * This is called before a mode is programmed. A typical use might be to
- * enable DPMS during the programming to avoid seeing intermediate stages,
- * but that's not relevant to us
- */
-static void cirrus_crtc_prepare(struct drm_crtc *crtc)
-{
-}
-
-static void cirrus_crtc_load_lut(struct drm_crtc *crtc)
-{
-	struct drm_device *dev = crtc->dev;
-	struct cirrus_device *cdev = dev->dev_private;
-	u16 *r, *g, *b;
-	int i;
-
-	if (!crtc->enabled)
-		return;
-
-	r = crtc->gamma_store;
-	g = r + crtc->gamma_size;
-	b = g + crtc->gamma_size;
-
-	for (i = 0; i < CIRRUS_LUT_SIZE; i++) {
-		/* VGA registers */
-		WREG8(PALETTE_INDEX, i);
-		WREG8(PALETTE_DATA, *r++ >> 8);
-		WREG8(PALETTE_DATA, *g++ >> 8);
-		WREG8(PALETTE_DATA, *b++ >> 8);
-	}
-}
-
-/*
- * This is called after a mode is programmed. It should reverse anything done
- * by the prepare function
- */
-static void cirrus_crtc_commit(struct drm_crtc *crtc)
-{
-	cirrus_crtc_load_lut(crtc);
-}
-
-/*
- * The core can pass us a set of gamma values to program. We actually only
- * use this for 8-bit mode so can't perform smooth fades on deeper modes,
- * but it's a requirement that we provide the function
- */
-static int cirrus_crtc_gamma_set(struct drm_crtc *crtc, u16 *red, u16 *green,
-				 u16 *blue, uint32_t size,
-				 struct drm_modeset_acquire_ctx *ctx)
-{
-	cirrus_crtc_load_lut(crtc);
-
-	return 0;
-}
-
-/* Simple cleanup function */
-static void cirrus_crtc_destroy(struct drm_crtc *crtc)
-{
-	struct cirrus_crtc *cirrus_crtc = to_cirrus_crtc(crtc);
-
-	drm_crtc_cleanup(crtc);
-	kfree(cirrus_crtc);
-}
-
-/* These provide the minimum set of functions required to handle a CRTC */
-static const struct drm_crtc_funcs cirrus_crtc_funcs = {
-	.gamma_set = cirrus_crtc_gamma_set,
-	.set_config = drm_crtc_helper_set_config,
-	.destroy = cirrus_crtc_destroy,
-};
-
-static const struct drm_crtc_helper_funcs cirrus_helper_funcs = {
-	.dpms = cirrus_crtc_dpms,
-	.mode_set = cirrus_crtc_mode_set,
-	.mode_set_base = cirrus_crtc_mode_set_base,
-	.prepare = cirrus_crtc_prepare,
-	.commit = cirrus_crtc_commit,
-};
-
-/* CRTC setup */
-static const uint32_t cirrus_formats_16[] = {
-	DRM_FORMAT_RGB565,
-};
-
-static const uint32_t cirrus_formats_24[] = {
-	DRM_FORMAT_RGB888,
-	DRM_FORMAT_RGB565,
-};
-
-static const uint32_t cirrus_formats_32[] = {
-	DRM_FORMAT_XRGB8888,
-	DRM_FORMAT_ARGB8888,
-	DRM_FORMAT_RGB888,
-	DRM_FORMAT_RGB565,
-};
-
-static struct drm_plane *cirrus_primary_plane(struct drm_device *dev)
-{
-	const uint32_t *formats;
-	uint32_t nformats;
-	struct drm_plane *primary;
-	int ret;
-
-	switch (cirrus_bpp) {
-	case 16:
-		formats = cirrus_formats_16;
-		nformats = ARRAY_SIZE(cirrus_formats_16);
-		break;
-	case 24:
-		formats = cirrus_formats_24;
-		nformats = ARRAY_SIZE(cirrus_formats_24);
-		break;
-	case 32:
-		formats = cirrus_formats_32;
-		nformats = ARRAY_SIZE(cirrus_formats_32);
-		break;
-	default:
-		return NULL;
-	}
-
-	primary = kzalloc(sizeof(*primary), GFP_KERNEL);
-	if (primary == NULL) {
-		DRM_DEBUG_KMS("Failed to allocate primary plane\n");
-		return NULL;
-	}
-
-	ret = drm_universal_plane_init(dev, primary, 0,
-				       &drm_primary_helper_funcs,
-				       formats, nformats,
-				       NULL,
-				       DRM_PLANE_TYPE_PRIMARY, NULL);
-	if (ret) {
-		kfree(primary);
-		primary = NULL;
-	}
-
-	return primary;
-}
-
-static void cirrus_crtc_init(struct drm_device *dev)
-{
-	struct cirrus_device *cdev = dev->dev_private;
-	struct cirrus_crtc *cirrus_crtc;
-	struct drm_plane *primary;
-
-	cirrus_crtc = kzalloc(sizeof(struct cirrus_crtc) +
-			      (CIRRUSFB_CONN_LIMIT * sizeof(struct drm_connector *)),
-			      GFP_KERNEL);
-
-	if (cirrus_crtc == NULL)
-		return;
-
-	primary = cirrus_primary_plane(dev);
-	if (primary == NULL) {
-		kfree(cirrus_crtc);
-		return;
-	}
-
-	drm_crtc_init_with_planes(dev, &cirrus_crtc->base,
-				  primary, NULL,
-				  &cirrus_crtc_funcs, NULL);
-
-	drm_mode_crtc_set_gamma_size(&cirrus_crtc->base, CIRRUS_LUT_SIZE);
-	cdev->mode_info.crtc = cirrus_crtc;
-
-	drm_crtc_helper_add(&cirrus_crtc->base, &cirrus_helper_funcs);
-}
-
-static void cirrus_encoder_mode_set(struct drm_encoder *encoder,
-				struct drm_display_mode *mode,
-				struct drm_display_mode *adjusted_mode)
-{
-}
-
-static void cirrus_encoder_dpms(struct drm_encoder *encoder, int state)
-{
-	return;
-}
-
-static void cirrus_encoder_prepare(struct drm_encoder *encoder)
-{
-}
-
-static void cirrus_encoder_commit(struct drm_encoder *encoder)
-{
-}
-
-static void cirrus_encoder_destroy(struct drm_encoder *encoder)
-{
-	struct cirrus_encoder *cirrus_encoder = to_cirrus_encoder(encoder);
-	drm_encoder_cleanup(encoder);
-	kfree(cirrus_encoder);
-}
-
-static const struct drm_encoder_helper_funcs cirrus_encoder_helper_funcs = {
-	.dpms = cirrus_encoder_dpms,
-	.mode_set = cirrus_encoder_mode_set,
-	.prepare = cirrus_encoder_prepare,
-	.commit = cirrus_encoder_commit,
-};
-
-static const struct drm_encoder_funcs cirrus_encoder_encoder_funcs = {
-	.destroy = cirrus_encoder_destroy,
-};
-
-static struct drm_encoder *cirrus_encoder_init(struct drm_device *dev)
-{
-	struct drm_encoder *encoder;
-	struct cirrus_encoder *cirrus_encoder;
-
-	cirrus_encoder = kzalloc(sizeof(struct cirrus_encoder), GFP_KERNEL);
-	if (!cirrus_encoder)
-		return NULL;
-
-	encoder = &cirrus_encoder->base;
-	encoder->possible_crtcs = 0x1;
-
-	drm_encoder_init(dev, encoder, &cirrus_encoder_encoder_funcs,
-			 DRM_MODE_ENCODER_DAC, NULL);
-	drm_encoder_helper_add(encoder, &cirrus_encoder_helper_funcs);
-
-	return encoder;
-}
-
-
-static int cirrus_vga_get_modes(struct drm_connector *connector)
-{
-	int count;
-
-	/* Just add a static list of modes */
-	if (cirrus_bpp <= 24) {
-		count = drm_add_modes_noedid(connector, 1280, 1024);
-		drm_set_preferred_mode(connector, 1024, 768);
-	} else {
-		count = drm_add_modes_noedid(connector, 800, 600);
-		drm_set_preferred_mode(connector, 800, 600);
-	}
-	return count;
-}
-
-static struct drm_encoder *cirrus_connector_best_encoder(struct drm_connector
-						  *connector)
-{
-	int enc_id = connector->encoder_ids[0];
-	/* pick the encoder ids */
-	if (enc_id)
-		return drm_encoder_find(connector->dev, NULL, enc_id);
-	return NULL;
-}
-
-static void cirrus_connector_destroy(struct drm_connector *connector)
-{
-	drm_connector_cleanup(connector);
-	kfree(connector);
-}
-
-static const struct drm_connector_helper_funcs cirrus_vga_connector_helper_funcs = {
-	.get_modes = cirrus_vga_get_modes,
-	.best_encoder = cirrus_connector_best_encoder,
-};
-
-static const struct drm_connector_funcs cirrus_vga_connector_funcs = {
-	.dpms = drm_helper_connector_dpms,
-	.fill_modes = drm_helper_probe_single_connector_modes,
-	.destroy = cirrus_connector_destroy,
-};
-
-static struct drm_connector *cirrus_vga_init(struct drm_device *dev)
-{
-	struct drm_connector *connector;
-	struct cirrus_connector *cirrus_connector;
-
-	cirrus_connector = kzalloc(sizeof(struct cirrus_connector), GFP_KERNEL);
-	if (!cirrus_connector)
-		return NULL;
-
-	connector = &cirrus_connector->base;
-
-	drm_connector_init(dev, connector,
-			   &cirrus_vga_connector_funcs, DRM_MODE_CONNECTOR_VGA);
-
-	drm_connector_helper_add(connector, &cirrus_vga_connector_helper_funcs);
-
-	drm_connector_register(connector);
-	return connector;
-}
-
-
-int cirrus_modeset_init(struct cirrus_device *cdev)
-{
-	struct drm_encoder *encoder;
-	struct drm_connector *connector;
-	int ret;
-
-	drm_mode_config_init(cdev->dev);
-	cdev->mode_info.mode_config_initialized = true;
-
-	cdev->dev->mode_config.max_width = CIRRUS_MAX_FB_WIDTH;
-	cdev->dev->mode_config.max_height = CIRRUS_MAX_FB_HEIGHT;
-
-	cdev->dev->mode_config.fb_base = cdev->mc.vram_base;
-	cdev->dev->mode_config.preferred_depth = cirrus_bpp;
-	/* don't prefer a shadow on virt GPU */
-	cdev->dev->mode_config.prefer_shadow = 0;
-
-	cirrus_crtc_init(cdev->dev);
-
-	encoder = cirrus_encoder_init(cdev->dev);
-	if (!encoder) {
-		DRM_ERROR("cirrus_encoder_init failed\n");
-		return -1;
-	}
-
-	connector = cirrus_vga_init(cdev->dev);
-	if (!connector) {
-		DRM_ERROR("cirrus_vga_init failed\n");
-		return -1;
-	}
-
-	drm_connector_attach_encoder(connector, encoder);
-
-	ret = cirrus_fbdev_init(cdev);
-	if (ret) {
-		DRM_ERROR("cirrus_fbdev_init failed\n");
-		return ret;
-	}
-
-	return 0;
-}
-
-void cirrus_modeset_fini(struct cirrus_device *cdev)
-{
-	cirrus_fbdev_fini(cdev);
-
-	if (cdev->mode_info.mode_config_initialized) {
-		drm_mode_config_cleanup(cdev->dev);
-		cdev->mode_info.mode_config_initialized = false;
-	}
-}
diff --git a/drivers/gpu/drm/cirrus/cirrus_ttm.c b/drivers/gpu/drm/cirrus/cirrus_ttm.c
index e075810b4bd4..e6b98467a428 100644
--- a/drivers/gpu/drm/cirrus/cirrus_ttm.c
+++ b/drivers/gpu/drm/cirrus/cirrus_ttm.c
@@ -178,7 +178,6 @@ int cirrus_mm_init(struct cirrus_device *cirrus)
 	ret = ttm_bo_device_init(&cirrus->ttm.bdev,
 				 &cirrus_bo_driver,
 				 dev->anon_inode->i_mapping,
-				 DRM_FILE_PAGE_OFFSET,
 				 true);
 	if (ret) {
 		DRM_ERROR("Error initialising bo driver; %d\n", ret);
@@ -331,13 +330,8 @@ int cirrus_bo_push_sysram(struct cirrus_bo *bo)
 
 int cirrus_mmap(struct file *filp, struct vm_area_struct *vma)
 {
-	struct drm_file *file_priv;
-	struct cirrus_device *cirrus;
+	struct drm_file *file_priv = filp->private_data;
+	struct cirrus_device *cirrus = file_priv->minor->dev->dev_private;
 
-	if (unlikely(vma->vm_pgoff < DRM_FILE_PAGE_OFFSET))
-		return -EINVAL;
-
-	file_priv = filp->private_data;
-	cirrus = file_priv->minor->dev->dev_private;
 	return ttm_bo_mmap(filp, vma, &cirrus->ttm.bdev);
 }
diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c
index 5eb40130fafb..f4924cb7f495 100644
--- a/drivers/gpu/drm/drm_atomic.c
+++ b/drivers/gpu/drm/drm_atomic.c
@@ -798,6 +798,50 @@ drm_atomic_get_private_obj_state(struct drm_atomic_state *state,
 EXPORT_SYMBOL(drm_atomic_get_private_obj_state);
 
 /**
+ * drm_atomic_get_old_private_obj_state
+ * @state: global atomic state object
+ * @obj: private_obj to grab
+ *
+ * This function returns the old private object state for the given private_obj,
+ * or NULL if the private_obj is not part of the global atomic state.
+ */
+struct drm_private_state *
+drm_atomic_get_old_private_obj_state(struct drm_atomic_state *state,
+				     struct drm_private_obj *obj)
+{
+	int i;
+
+	for (i = 0; i < state->num_private_objs; i++)
+		if (obj == state->private_objs[i].ptr)
+			return state->private_objs[i].old_state;
+
+	return NULL;
+}
+EXPORT_SYMBOL(drm_atomic_get_old_private_obj_state);
+
+/**
+ * drm_atomic_get_new_private_obj_state
+ * @state: global atomic state object
+ * @obj: private_obj to grab
+ *
+ * This function returns the new private object state for the given private_obj,
+ * or NULL if the private_obj is not part of the global atomic state.
+ */
+struct drm_private_state *
+drm_atomic_get_new_private_obj_state(struct drm_atomic_state *state,
+				     struct drm_private_obj *obj)
+{
+	int i;
+
+	for (i = 0; i < state->num_private_objs; i++)
+		if (obj == state->private_objs[i].ptr)
+			return state->private_objs[i].new_state;
+
+	return NULL;
+}
+EXPORT_SYMBOL(drm_atomic_get_new_private_obj_state);
+
+/**
  * drm_atomic_get_connector_state - get connector state
  * @state: global atomic state object
  * @connector: connector to get state object for
@@ -1236,4 +1280,3 @@ int drm_atomic_debugfs_init(struct drm_minor *minor)
 			minor->debugfs_root, minor);
 }
 #endif
-
diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c
index fbb76332cc9f..2e0cb4246cbd 100644
--- a/drivers/gpu/drm/drm_atomic_helper.c
+++ b/drivers/gpu/drm/drm_atomic_helper.c
@@ -495,7 +495,7 @@ mode_fixup(struct drm_atomic_state *state)
 static enum drm_mode_status mode_valid_path(struct drm_connector *connector,
 					    struct drm_encoder *encoder,
 					    struct drm_crtc *crtc,
-					    struct drm_display_mode *mode)
+					    const struct drm_display_mode *mode)
 {
 	enum drm_mode_status ret;
 
@@ -534,7 +534,7 @@ mode_valid(struct drm_atomic_state *state)
 		struct drm_crtc *crtc = conn_state->crtc;
 		struct drm_crtc_state *crtc_state;
 		enum drm_mode_status mode_status;
-		struct drm_display_mode *mode;
+		const struct drm_display_mode *mode;
 
 		if (!crtc || !encoder)
 			continue;
@@ -1751,7 +1751,7 @@ int drm_atomic_helper_commit(struct drm_device *dev,
 	 *
 	 * NOTE: Commit work has multiple phases, first hardware commit, then
 	 * cleanup. We want them to overlap, hence need system_unbound_wq to
-	 * make sure work items don't artifically stall on each another.
+	 * make sure work items don't artificially stall on each another.
 	 */
 
 	drm_atomic_state_get(state);
@@ -1785,7 +1785,7 @@ EXPORT_SYMBOL(drm_atomic_helper_commit);
  *
  * Asynchronous workers need to have sufficient parallelism to be able to run
  * different atomic commits on different CRTCs in parallel. The simplest way to
- * achive this is by running them on the &system_unbound_wq work queue. Note
+ * achieve this is by running them on the &system_unbound_wq work queue. Note
  * that drivers are not required to split up atomic commits and run an
  * individual commit in parallel - userspace is supposed to do that if it cares.
  * But it might be beneficial to do that for modesets, since those necessarily
@@ -2260,10 +2260,21 @@ EXPORT_SYMBOL(drm_atomic_helper_commit_cleanup_done);
 int drm_atomic_helper_prepare_planes(struct drm_device *dev,
 				     struct drm_atomic_state *state)
 {
+	struct drm_connector *connector;
+	struct drm_connector_state *new_conn_state;
 	struct drm_plane *plane;
 	struct drm_plane_state *new_plane_state;
 	int ret, i, j;
 
+	for_each_new_connector_in_state(state, connector, new_conn_state, i) {
+		if (!new_conn_state->writeback_job)
+			continue;
+
+		ret = drm_writeback_prepare_job(new_conn_state->writeback_job);
+		if (ret < 0)
+			return ret;
+	}
+
 	for_each_new_plane_in_state(state, plane, new_plane_state, i) {
 		const struct drm_plane_helper_funcs *funcs;
 
diff --git a/drivers/gpu/drm/drm_atomic_state_helper.c b/drivers/gpu/drm/drm_atomic_state_helper.c
index 4985384e51f6..59ffb6b9c745 100644
--- a/drivers/gpu/drm/drm_atomic_state_helper.c
+++ b/drivers/gpu/drm/drm_atomic_state_helper.c
@@ -30,6 +30,7 @@
 #include <drm/drm_connector.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_device.h>
+#include <drm/drm_writeback.h>
 
 #include <linux/slab.h>
 #include <linux/dma-fence.h>
@@ -412,6 +413,9 @@ __drm_atomic_helper_connector_destroy_state(struct drm_connector_state *state)
 
 	if (state->commit)
 		drm_crtc_commit_put(state->commit);
+
+	if (state->writeback_job)
+		drm_writeback_cleanup_job(state->writeback_job);
 }
 EXPORT_SYMBOL(__drm_atomic_helper_connector_destroy_state);
 
diff --git a/drivers/gpu/drm/drm_atomic_uapi.c b/drivers/gpu/drm/drm_atomic_uapi.c
index 0aabd401d3ca..428d82662dc4 100644
--- a/drivers/gpu/drm/drm_atomic_uapi.c
+++ b/drivers/gpu/drm/drm_atomic_uapi.c
@@ -512,8 +512,8 @@ drm_atomic_crtc_get_property(struct drm_crtc *crtc,
 }
 
 static int drm_atomic_plane_set_property(struct drm_plane *plane,
-		struct drm_plane_state *state, struct drm_property *property,
-		uint64_t val)
+		struct drm_plane_state *state, struct drm_file *file_priv,
+		struct drm_property *property, uint64_t val)
 {
 	struct drm_device *dev = plane->dev;
 	struct drm_mode_config *config = &dev->mode_config;
@@ -521,7 +521,8 @@ static int drm_atomic_plane_set_property(struct drm_plane *plane,
 	int ret;
 
 	if (property == config->prop_fb_id) {
-		struct drm_framebuffer *fb = drm_framebuffer_lookup(dev, NULL, val);
+		struct drm_framebuffer *fb;
+		fb = drm_framebuffer_lookup(dev, file_priv, val);
 		drm_atomic_set_fb_for_plane(state, fb);
 		if (fb)
 			drm_framebuffer_put(fb);
@@ -537,7 +538,9 @@ static int drm_atomic_plane_set_property(struct drm_plane *plane,
 			return -EINVAL;
 
 	} else if (property == config->prop_crtc_id) {
-		struct drm_crtc *crtc = drm_crtc_find(dev, NULL, val);
+		struct drm_crtc *crtc = drm_crtc_find(dev, file_priv, val);
+		if (val && !crtc)
+			return -EACCES;
 		return drm_atomic_set_crtc_for_plane(state, crtc);
 	} else if (property == config->prop_crtc_x) {
 		state->crtc_x = U642I64(val);
@@ -647,28 +650,15 @@ drm_atomic_plane_get_property(struct drm_plane *plane,
 	return 0;
 }
 
-static struct drm_writeback_job *
-drm_atomic_get_writeback_job(struct drm_connector_state *conn_state)
-{
-	WARN_ON(conn_state->connector->connector_type != DRM_MODE_CONNECTOR_WRITEBACK);
-
-	if (!conn_state->writeback_job)
-		conn_state->writeback_job =
-			kzalloc(sizeof(*conn_state->writeback_job), GFP_KERNEL);
-
-	return conn_state->writeback_job;
-}
-
 static int drm_atomic_set_writeback_fb_for_connector(
 		struct drm_connector_state *conn_state,
 		struct drm_framebuffer *fb)
 {
-	struct drm_writeback_job *job =
-		drm_atomic_get_writeback_job(conn_state);
-	if (!job)
-		return -ENOMEM;
+	int ret;
 
-	drm_framebuffer_assign(&job->fb, fb);
+	ret = drm_writeback_set_fb(conn_state, fb);
+	if (ret < 0)
+		return ret;
 
 	if (fb)
 		DRM_DEBUG_ATOMIC("Set [FB:%d] for connector state %p\n",
@@ -681,14 +671,16 @@ static int drm_atomic_set_writeback_fb_for_connector(
 }
 
 static int drm_atomic_connector_set_property(struct drm_connector *connector,
-		struct drm_connector_state *state, struct drm_property *property,
-		uint64_t val)
+		struct drm_connector_state *state, struct drm_file *file_priv,
+		struct drm_property *property, uint64_t val)
 {
 	struct drm_device *dev = connector->dev;
 	struct drm_mode_config *config = &dev->mode_config;
 
 	if (property == config->prop_crtc_id) {
-		struct drm_crtc *crtc = drm_crtc_find(dev, NULL, val);
+		struct drm_crtc *crtc = drm_crtc_find(dev, file_priv, val);
+		if (val && !crtc)
+			return -EACCES;
 		return drm_atomic_set_crtc_for_connector(state, crtc);
 	} else if (property == config->dpms_property) {
 		/* setting DPMS property requires special handling, which
@@ -746,9 +738,13 @@ static int drm_atomic_connector_set_property(struct drm_connector *connector,
 			return -EINVAL;
 		}
 		state->content_protection = val;
+	} else if (property == connector->colorspace_property) {
+		state->colorspace = val;
 	} else if (property == config->writeback_fb_id_property) {
-		struct drm_framebuffer *fb = drm_framebuffer_lookup(dev, NULL, val);
-		int ret = drm_atomic_set_writeback_fb_for_connector(state, fb);
+		struct drm_framebuffer *fb;
+		int ret;
+		fb = drm_framebuffer_lookup(dev, file_priv, val);
+		ret = drm_atomic_set_writeback_fb_for_connector(state, fb);
 		if (fb)
 			drm_framebuffer_put(fb);
 		return ret;
@@ -814,6 +810,8 @@ drm_atomic_connector_get_property(struct drm_connector *connector,
 		*val = state->picture_aspect_ratio;
 	} else if (property == config->content_type_property) {
 		*val = state->content_type;
+	} else if (property == connector->colorspace_property) {
+		*val = state->colorspace;
 	} else if (property == connector->scaling_mode_property) {
 		*val = state->scaling_mode;
 	} else if (property == connector->content_protection_property) {
@@ -943,6 +941,7 @@ out:
 }
 
 int drm_atomic_set_property(struct drm_atomic_state *state,
+			    struct drm_file *file_priv,
 			    struct drm_mode_object *obj,
 			    struct drm_property *prop,
 			    uint64_t prop_value)
@@ -965,7 +964,8 @@ int drm_atomic_set_property(struct drm_atomic_state *state,
 		}
 
 		ret = drm_atomic_connector_set_property(connector,
-				connector_state, prop, prop_value);
+				connector_state, file_priv,
+				prop, prop_value);
 		break;
 	}
 	case DRM_MODE_OBJECT_CRTC: {
@@ -993,7 +993,8 @@ int drm_atomic_set_property(struct drm_atomic_state *state,
 		}
 
 		ret = drm_atomic_plane_set_property(plane,
-				plane_state, prop, prop_value);
+				plane_state, file_priv,
+				prop, prop_value);
 		break;
 	}
 	default:
@@ -1158,19 +1159,17 @@ static int prepare_signaling(struct drm_device *dev,
 
 	for_each_new_connector_in_state(state, conn, conn_state, i) {
 		struct drm_writeback_connector *wb_conn;
-		struct drm_writeback_job *job;
 		struct drm_out_fence_state *f;
 		struct dma_fence *fence;
 		s32 __user *fence_ptr;
 
+		if (!conn_state->writeback_job)
+			continue;
+
 		fence_ptr = get_out_fence_for_connector(state, conn);
 		if (!fence_ptr)
 			continue;
 
-		job = drm_atomic_get_writeback_job(conn_state);
-		if (!job)
-			return -ENOMEM;
-
 		f = krealloc(*fence_state, sizeof(**fence_state) *
 			     (*num_fences + 1), GFP_KERNEL);
 		if (!f)
@@ -1192,7 +1191,7 @@ static int prepare_signaling(struct drm_device *dev,
 			return ret;
 		}
 
-		job->out_fence = fence;
+		conn_state->writeback_job->out_fence = fence;
 	}
 
 	/*
@@ -1365,8 +1364,8 @@ retry:
 				goto out;
 			}
 
-			ret = drm_atomic_set_property(state, obj, prop,
-						      prop_value);
+			ret = drm_atomic_set_property(state, file_priv,
+						      obj, prop, prop_value);
 			if (ret) {
 				drm_mode_object_put(obj);
 				goto out;
diff --git a/drivers/gpu/drm/drm_auth.c b/drivers/gpu/drm/drm_auth.c
index 1669c42c40ed..22c7a104b802 100644
--- a/drivers/gpu/drm/drm_auth.c
+++ b/drivers/gpu/drm/drm_auth.c
@@ -103,14 +103,11 @@ struct drm_master *drm_master_create(struct drm_device *dev)
 		return NULL;
 
 	kref_init(&master->refcount);
-	spin_lock_init(&master->lock.spinlock);
-	init_waitqueue_head(&master->lock.lock_queue);
+	drm_master_legacy_init(master);
 	idr_init(&master->magic_map);
 	master->dev = dev;
 
 	/* initialize the tree of output resource lessees */
-	master->lessor = NULL;
-	master->lessee_id = 0;
 	INIT_LIST_HEAD(&master->lessees);
 	INIT_LIST_HEAD(&master->lessee_list);
 	idr_init(&master->leases);
@@ -274,21 +271,7 @@ void drm_master_release(struct drm_file *file_priv)
 	if (!drm_is_current_master(file_priv))
 		goto out;
 
-	if (drm_core_check_feature(dev, DRIVER_LEGACY)) {
-		/*
-		 * Since the master is disappearing, so is the
-		 * possibility to lock.
-		 */
-		mutex_lock(&dev->struct_mutex);
-		if (master->lock.hw_lock) {
-			if (dev->sigdata.lock == master->lock.hw_lock)
-				dev->sigdata.lock = NULL;
-			master->lock.hw_lock = NULL;
-			master->lock.file_priv = NULL;
-			wake_up_interruptible_all(&master->lock.lock_queue);
-		}
-		mutex_unlock(&dev->struct_mutex);
-	}
+	drm_legacy_lock_master_cleanup(dev, master);
 
 	if (dev->master == file_priv->master)
 		drm_drop_master(dev, file_priv);
diff --git a/drivers/gpu/drm/drm_bufs.c b/drivers/gpu/drm/drm_bufs.c
index e407adb033e7..bfc419ed9d6c 100644
--- a/drivers/gpu/drm/drm_bufs.c
+++ b/drivers/gpu/drm/drm_bufs.c
@@ -584,6 +584,14 @@ void drm_legacy_master_rmmaps(struct drm_device *dev, struct drm_master *master)
 	mutex_unlock(&dev->struct_mutex);
 }
 
+void drm_legacy_rmmaps(struct drm_device *dev)
+{
+	struct drm_map_list *r_list, *list_temp;
+
+	list_for_each_entry_safe(r_list, list_temp, &dev->maplist, head)
+		drm_legacy_rmmap(dev, r_list->map);
+}
+
 /* The rmmap ioctl appears to be unnecessary.  All mappings are torn down on
  * the last close of the device, and this is necessary for cleanup when things
  * exit uncleanly.  Therefore, having userland manually remove mappings seems
diff --git a/drivers/gpu/drm/drm_client.c b/drivers/gpu/drm/drm_client.c
index 9b2bd28dde0a..f20d1dda3961 100644
--- a/drivers/gpu/drm/drm_client.c
+++ b/drivers/gpu/drm/drm_client.c
@@ -69,7 +69,8 @@ EXPORT_SYMBOL(drm_client_close);
  * @name: Client name
  * @funcs: DRM client functions (optional)
  *
- * This initialises the client and opens a &drm_file. Use drm_client_add() to complete the process.
+ * This initialises the client and opens a &drm_file.
+ * Use drm_client_register() to complete the process.
  * The caller needs to hold a reference on @dev before calling this function.
  * The client is freed when the &drm_device is unregistered. See drm_client_release().
  *
@@ -108,16 +109,16 @@ err_put_module:
 EXPORT_SYMBOL(drm_client_init);
 
 /**
- * drm_client_add - Add client to the device list
+ * drm_client_register - Register client
  * @client: DRM client
  *
  * Add the client to the &drm_device client list to activate its callbacks.
  * @client must be initialized by a call to drm_client_init(). After
- * drm_client_add() it is no longer permissible to call drm_client_release()
+ * drm_client_register() it is no longer permissible to call drm_client_release()
  * directly (outside the unregister callback), instead cleanup will happen
  * automatically on driver unload.
  */
-void drm_client_add(struct drm_client_dev *client)
+void drm_client_register(struct drm_client_dev *client)
 {
 	struct drm_device *dev = client->dev;
 
@@ -125,7 +126,7 @@ void drm_client_add(struct drm_client_dev *client)
 	list_add(&client->list, &dev->clientlist);
 	mutex_unlock(&dev->clientlist_mutex);
 }
-EXPORT_SYMBOL(drm_client_add);
+EXPORT_SYMBOL(drm_client_register);
 
 /**
  * drm_client_release - Release DRM client resources
diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c
index dd40eff0911c..b34c3d38bf15 100644
--- a/drivers/gpu/drm/drm_connector.c
+++ b/drivers/gpu/drm/drm_connector.c
@@ -245,6 +245,7 @@ int drm_connector_init(struct drm_device *dev,
 	INIT_LIST_HEAD(&connector->modes);
 	mutex_init(&connector->mutex);
 	connector->edid_blob_ptr = NULL;
+	connector->tile_blob_ptr = NULL;
 	connector->status = connector_status_unknown;
 	connector->display_info.panel_orientation =
 		DRM_MODE_PANEL_ORIENTATION_UNKNOWN;
@@ -272,6 +273,9 @@ int drm_connector_init(struct drm_device *dev,
 	drm_object_attach_property(&connector->base,
 				   config->non_desktop_property,
 				   0);
+	drm_object_attach_property(&connector->base,
+				   config->tile_property,
+				   0);
 
 	if (drm_core_check_feature(dev, DRIVER_ATOMIC)) {
 		drm_object_attach_property(&connector->base, config->prop_crtc_id, 0);
@@ -826,6 +830,33 @@ static struct drm_prop_enum_list drm_cp_enum_list[] = {
 };
 DRM_ENUM_NAME_FN(drm_get_content_protection_name, drm_cp_enum_list)
 
+static const struct drm_prop_enum_list hdmi_colorspaces[] = {
+	/* For Default case, driver will set the colorspace */
+	{ DRM_MODE_COLORIMETRY_DEFAULT, "Default" },
+	/* Standard Definition Colorimetry based on CEA 861 */
+	{ DRM_MODE_COLORIMETRY_SMPTE_170M_YCC, "SMPTE_170M_YCC" },
+	{ DRM_MODE_COLORIMETRY_BT709_YCC, "BT709_YCC" },
+	/* Standard Definition Colorimetry based on IEC 61966-2-4 */
+	{ DRM_MODE_COLORIMETRY_XVYCC_601, "XVYCC_601" },
+	/* High Definition Colorimetry based on IEC 61966-2-4 */
+	{ DRM_MODE_COLORIMETRY_XVYCC_709, "XVYCC_709" },
+	/* Colorimetry based on IEC 61966-2-1/Amendment 1 */
+	{ DRM_MODE_COLORIMETRY_SYCC_601, "SYCC_601" },
+	/* Colorimetry based on IEC 61966-2-5 [33] */
+	{ DRM_MODE_COLORIMETRY_OPYCC_601, "opYCC_601" },
+	/* Colorimetry based on IEC 61966-2-5 */
+	{ DRM_MODE_COLORIMETRY_OPRGB, "opRGB" },
+	/* Colorimetry based on ITU-R BT.2020 */
+	{ DRM_MODE_COLORIMETRY_BT2020_CYCC, "BT2020_CYCC" },
+	/* Colorimetry based on ITU-R BT.2020 */
+	{ DRM_MODE_COLORIMETRY_BT2020_RGB, "BT2020_RGB" },
+	/* Colorimetry based on ITU-R BT.2020 */
+	{ DRM_MODE_COLORIMETRY_BT2020_YCC, "BT2020_YCC" },
+	/* Added as part of Additional Colorimetry Extension in 861.G */
+	{ DRM_MODE_COLORIMETRY_DCI_P3_RGB_D65, "DCI-P3_RGB_D65" },
+	{ DRM_MODE_COLORIMETRY_DCI_P3_RGB_THEATER, "DCI-P3_RGB_Theater" },
+};
+
 /**
  * DOC: standard connector properties
  *
@@ -1385,12 +1416,6 @@ EXPORT_SYMBOL(drm_mode_create_scaling_mode_property);
  *
  *	The driver may place further restrictions within these minimum
  *	and maximum bounds.
- *
- *	The semantics for the vertical blank timestamp differ when
- *	variable refresh rate is active. The vertical blank timestamp
- *	is defined to be an estimate using the current mode's fixed
- *	refresh rate timings. The semantics for the page-flip event
- *	timestamp remain the same.
  */
 
 /**
@@ -1548,6 +1573,57 @@ int drm_mode_create_aspect_ratio_property(struct drm_device *dev)
 EXPORT_SYMBOL(drm_mode_create_aspect_ratio_property);
 
 /**
+ * DOC: standard connector properties
+ *
+ * Colorspace:
+ *     drm_mode_create_colorspace_property - create colorspace property
+ *     This property helps select a suitable colorspace based on the sink
+ *     capability. Modern sink devices support wider gamut like BT2020.
+ *     This helps switch to BT2020 mode if the BT2020 encoded video stream
+ *     is being played by the user, same for any other colorspace. Thereby
+ *     giving a good visual experience to users.
+ *
+ *     The expectation from userspace is that it should parse the EDID
+ *     and get supported colorspaces. Use this property and switch to the
+ *     one supported. Sink supported colorspaces should be retrieved by
+ *     userspace from EDID and driver will not explicitly expose them.
+ *
+ *     Basically the expectation from userspace is:
+ *      - Set up CRTC DEGAMMA/CTM/GAMMA to convert to some sink
+ *        colorspace
+ *      - Set this new property to let the sink know what it
+ *        converted the CRTC output to.
+ *      - This property is just to inform sink what colorspace
+ *        source is trying to drive.
+ *
+ * Called by a driver the first time it's needed, must be attached to desired
+ * connectors.
+ */
+int drm_mode_create_colorspace_property(struct drm_connector *connector)
+{
+	struct drm_device *dev = connector->dev;
+	struct drm_property *prop;
+
+	if (connector->connector_type == DRM_MODE_CONNECTOR_HDMIA ||
+	    connector->connector_type == DRM_MODE_CONNECTOR_HDMIB) {
+		prop = drm_property_create_enum(dev, DRM_MODE_PROP_ENUM,
+						"Colorspace",
+						hdmi_colorspaces,
+						ARRAY_SIZE(hdmi_colorspaces));
+		if (!prop)
+			return -ENOMEM;
+	} else {
+		DRM_DEBUG_KMS("Colorspace property not supported\n");
+		return 0;
+	}
+
+	connector->colorspace_property = prop;
+
+	return 0;
+}
+EXPORT_SYMBOL(drm_mode_create_colorspace_property);
+
+/**
  * drm_mode_create_content_type_property - create content type property
  * @dev: DRM device
  *
@@ -1634,6 +1710,8 @@ EXPORT_SYMBOL(drm_connector_set_path_property);
  * This looks up the tile information for a connector, and creates a
  * property for userspace to parse if it exists. The property is of
  * the form of 8 integers using ':' as a separator.
+ * This is used for dual port tiled displays with DisplayPort SST
+ * or DisplayPort MST connectors.
  *
  * Returns:
  * Zero on success, errno on failure.
@@ -1677,6 +1755,9 @@ EXPORT_SYMBOL(drm_connector_set_tile_property);
  *
  * This function creates a new blob modeset object and assigns its id to the
  * connector's edid property.
+ * Since we also parse tile information from EDID's displayID block, we also
+ * set the connector's tile property here. See drm_connector_set_tile_property()
+ * for more details.
  *
  * Returns:
  * Zero on success, negative errno on failure.
@@ -1718,7 +1799,9 @@ int drm_connector_update_edid_property(struct drm_connector *connector,
 	                                       edid,
 	                                       &connector->base,
 	                                       dev->mode_config.edid_property);
-	return ret;
+	if (ret)
+		return ret;
+	return drm_connector_set_tile_property(connector);
 }
 EXPORT_SYMBOL(drm_connector_update_edid_property);
 
diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index 7dabbaf033a1..790ba5941954 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -559,6 +559,10 @@ int drm_mode_setcrtc(struct drm_device *dev, void *data,
 
 	plane = crtc->primary;
 
+	/* allow disabling with the primary plane leased */
+	if (crtc_req->mode_valid && !drm_lease_held(file_priv, plane->base.id))
+		return -EACCES;
+
 	mutex_lock(&crtc->dev->mode_config.mutex);
 	DRM_MODESET_LOCK_ALL_BEGIN(dev, ctx,
 				   DRM_MODESET_ACQUIRE_INTERRUPTIBLE, ret);
diff --git a/drivers/gpu/drm/drm_crtc_internal.h b/drivers/gpu/drm/drm_crtc_internal.h
index 216f2a9ee3d4..0719a235d6cc 100644
--- a/drivers/gpu/drm/drm_crtc_internal.h
+++ b/drivers/gpu/drm/drm_crtc_internal.h
@@ -214,6 +214,7 @@ int drm_atomic_connector_commit_dpms(struct drm_atomic_state *state,
 				     struct drm_connector *connector,
 				     int mode);
 int drm_atomic_set_property(struct drm_atomic_state *state,
+			    struct drm_file *file_priv,
 			    struct drm_mode_object *obj,
 			    struct drm_property *prop,
 			    uint64_t prop_value);
diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c
index dc7ac0c60547..c630ed157994 100644
--- a/drivers/gpu/drm/drm_dp_mst_topology.c
+++ b/drivers/gpu/drm/drm_dp_mst_topology.c
@@ -3022,7 +3022,6 @@ struct edid *drm_dp_mst_get_edid(struct drm_connector *connector, struct drm_dp_
 		edid = drm_edid_duplicate(port->cached_edid);
 	else {
 		edid = drm_get_edid(connector, &port->aux.ddc);
-		drm_connector_set_tile_property(connector);
 	}
 	port->has_audio = drm_detect_monitor_audio(edid);
 	drm_dp_mst_topology_put_port(port);
diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index 05bbc2b622fc..862621494a93 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -286,6 +286,138 @@ void drm_minor_release(struct drm_minor *minor)
  * Note that the lifetime rules for &drm_device instance has still a lot of
  * historical baggage. Hence use the reference counting provided by
  * drm_dev_get() and drm_dev_put() only carefully.
+ *
+ * Display driver example
+ * ~~~~~~~~~~~~~~~~~~~~~~
+ *
+ * The following example shows a typical structure of a DRM display driver.
+ * The example focus on the probe() function and the other functions that is
+ * almost always present and serves as a demonstration of devm_drm_dev_init()
+ * usage with its accompanying drm_driver->release callback.
+ *
+ * .. code-block:: c
+ *
+ *	struct driver_device {
+ *		struct drm_device drm;
+ *		void *userspace_facing;
+ *		struct clk *pclk;
+ *	};
+ *
+ *	static void driver_drm_release(struct drm_device *drm)
+ *	{
+ *		struct driver_device *priv = container_of(...);
+ *
+ *		drm_mode_config_cleanup(drm);
+ *		drm_dev_fini(drm);
+ *		kfree(priv->userspace_facing);
+ *		kfree(priv);
+ *	}
+ *
+ *	static struct drm_driver driver_drm_driver = {
+ *		[...]
+ *		.release = driver_drm_release,
+ *	};
+ *
+ *	static int driver_probe(struct platform_device *pdev)
+ *	{
+ *		struct driver_device *priv;
+ *		struct drm_device *drm;
+ *		int ret;
+ *
+ *		[
+ *		  devm_kzalloc() can't be used here because the drm_device
+ *		  lifetime can exceed the device lifetime if driver unbind
+ *		  happens when userspace still has open file descriptors.
+ *		]
+ *		priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+ *		if (!priv)
+ *			return -ENOMEM;
+ *
+ *		drm = &priv->drm;
+ *
+ *		ret = devm_drm_dev_init(&pdev->dev, drm, &driver_drm_driver);
+ *		if (ret) {
+ *			kfree(drm);
+ *			return ret;
+ *		}
+ *
+ *		drm_mode_config_init(drm);
+ *
+ *		priv->userspace_facing = kzalloc(..., GFP_KERNEL);
+ *		if (!priv->userspace_facing)
+ *			return -ENOMEM;
+ *
+ *		priv->pclk = devm_clk_get(dev, "PCLK");
+ *		if (IS_ERR(priv->pclk))
+ *			return PTR_ERR(priv->pclk);
+ *
+ *		[ Further setup, display pipeline etc ]
+ *
+ *		platform_set_drvdata(pdev, drm);
+ *
+ *		drm_mode_config_reset(drm);
+ *
+ *		ret = drm_dev_register(drm);
+ *		if (ret)
+ *			return ret;
+ *
+ *		drm_fbdev_generic_setup(drm, 32);
+ *
+ *		return 0;
+ *	}
+ *
+ *	[ This function is called before the devm_ resources are released ]
+ *	static int driver_remove(struct platform_device *pdev)
+ *	{
+ *		struct drm_device *drm = platform_get_drvdata(pdev);
+ *
+ *		drm_dev_unregister(drm);
+ *		drm_atomic_helper_shutdown(drm)
+ *
+ *		return 0;
+ *	}
+ *
+ *	[ This function is called on kernel restart and shutdown ]
+ *	static void driver_shutdown(struct platform_device *pdev)
+ *	{
+ *		drm_atomic_helper_shutdown(platform_get_drvdata(pdev));
+ *	}
+ *
+ *	static int __maybe_unused driver_pm_suspend(struct device *dev)
+ *	{
+ *		return drm_mode_config_helper_suspend(dev_get_drvdata(dev));
+ *	}
+ *
+ *	static int __maybe_unused driver_pm_resume(struct device *dev)
+ *	{
+ *		drm_mode_config_helper_resume(dev_get_drvdata(dev));
+ *
+ *		return 0;
+ *	}
+ *
+ *	static const struct dev_pm_ops driver_pm_ops = {
+ *		SET_SYSTEM_SLEEP_PM_OPS(driver_pm_suspend, driver_pm_resume)
+ *	};
+ *
+ *	static struct platform_driver driver_driver = {
+ *		.driver = {
+ *			[...]
+ *			.pm = &driver_pm_ops,
+ *		},
+ *		.probe = driver_probe,
+ *		.remove = driver_remove,
+ *		.shutdown = driver_shutdown,
+ *	};
+ *	module_platform_driver(driver_driver);
+ *
+ * Drivers that want to support device unplugging (USB, DT overlay unload) should
+ * use drm_dev_unplug() instead of drm_dev_unregister(). The driver must protect
+ * regions that is accessing device resources to prevent use after they're
+ * released. This is done using drm_dev_enter() and drm_dev_exit(). There is one
+ * shortcoming however, drm_dev_unplug() marks the drm_device as unplugged before
+ * drm_atomic_helper_shutdown() is called. This means that if the disable code
+ * paths are protected, they will not run on regular driver module unload,
+ * possibily leaving the hardware enabled.
  */
 
 /**
@@ -376,7 +508,6 @@ void drm_dev_unplug(struct drm_device *dev)
 	synchronize_srcu(&drm_unplug_srcu);
 
 	drm_dev_unregister(dev);
-	drm_dev_put(dev);
 }
 EXPORT_SYMBOL(drm_dev_unplug);
 
@@ -453,6 +584,31 @@ static void drm_fs_inode_free(struct inode *inode)
 }
 
 /**
+ * DOC: component helper usage recommendations
+ *
+ * DRM drivers that drive hardware where a logical device consists of a pile of
+ * independent hardware blocks are recommended to use the :ref:`component helper
+ * library<component>`. For consistency and better options for code reuse the
+ * following guidelines apply:
+ *
+ *  - The entire device initialization procedure should be run from the
+ *    &component_master_ops.master_bind callback, starting with drm_dev_init(),
+ *    then binding all components with component_bind_all() and finishing with
+ *    drm_dev_register().
+ *
+ *  - The opaque pointer passed to all components through component_bind_all()
+ *    should point at &struct drm_device of the device instance, not some driver
+ *    specific private structure.
+ *
+ *  - The component helper fills the niche where further standardization of
+ *    interfaces is not practical. When there already is, or will be, a
+ *    standardized interface like &drm_bridge or &drm_panel, providing its own
+ *    functions to find such components at driver load time, like
+ *    drm_of_find_panel_or_bridge(), then the component helper should not be
+ *    used.
+ */
+
+/**
  * drm_dev_init - Initialise new DRM device
  * @dev: DRM device
  * @driver: DRM driver
@@ -497,26 +653,22 @@ int drm_dev_init(struct drm_device *dev,
 	BUG_ON(!parent);
 
 	kref_init(&dev->ref);
-	dev->dev = parent;
+	dev->dev = get_device(parent);
 	dev->driver = driver;
 
 	/* no per-device feature limits by default */
 	dev->driver_features = ~0u;
 
+	drm_legacy_init_members(dev);
 	INIT_LIST_HEAD(&dev->filelist);
 	INIT_LIST_HEAD(&dev->filelist_internal);
 	INIT_LIST_HEAD(&dev->clientlist);
-	INIT_LIST_HEAD(&dev->ctxlist);
-	INIT_LIST_HEAD(&dev->vmalist);
-	INIT_LIST_HEAD(&dev->maplist);
 	INIT_LIST_HEAD(&dev->vblank_event_list);
 
-	spin_lock_init(&dev->buf_lock);
 	spin_lock_init(&dev->event_lock);
 	mutex_init(&dev->struct_mutex);
 	mutex_init(&dev->filelist_mutex);
 	mutex_init(&dev->clientlist_mutex);
-	mutex_init(&dev->ctxlist_mutex);
 	mutex_init(&dev->master_mutex);
 
 	dev->anon_inode = drm_fs_inode_new();
@@ -536,7 +688,7 @@ int drm_dev_init(struct drm_device *dev,
 	if (ret)
 		goto err_minors;
 
-	ret = drm_ht_create(&dev->map_hash, 12);
+	ret = drm_legacy_create_map_hash(dev);
 	if (ret)
 		goto err_minors;
 
@@ -561,21 +713,61 @@ err_setunique:
 		drm_gem_destroy(dev);
 err_ctxbitmap:
 	drm_legacy_ctxbitmap_cleanup(dev);
-	drm_ht_remove(&dev->map_hash);
+	drm_legacy_remove_map_hash(dev);
 err_minors:
 	drm_minor_free(dev, DRM_MINOR_PRIMARY);
 	drm_minor_free(dev, DRM_MINOR_RENDER);
 	drm_fs_inode_free(dev->anon_inode);
 err_free:
+	put_device(dev->dev);
 	mutex_destroy(&dev->master_mutex);
-	mutex_destroy(&dev->ctxlist_mutex);
 	mutex_destroy(&dev->clientlist_mutex);
 	mutex_destroy(&dev->filelist_mutex);
 	mutex_destroy(&dev->struct_mutex);
+	drm_legacy_destroy_members(dev);
 	return ret;
 }
 EXPORT_SYMBOL(drm_dev_init);
 
+static void devm_drm_dev_init_release(void *data)
+{
+	drm_dev_put(data);
+}
+
+/**
+ * devm_drm_dev_init - Resource managed drm_dev_init()
+ * @parent: Parent device object
+ * @dev: DRM device
+ * @driver: DRM driver
+ *
+ * Managed drm_dev_init(). The DRM device initialized with this function is
+ * automatically put on driver detach using drm_dev_put(). You must supply a
+ * &drm_driver.release callback to control the finalization explicitly.
+ *
+ * RETURNS:
+ * 0 on success, or error code on failure.
+ */
+int devm_drm_dev_init(struct device *parent,
+		      struct drm_device *dev,
+		      struct drm_driver *driver)
+{
+	int ret;
+
+	if (WARN_ON(!parent || !driver->release))
+		return -EINVAL;
+
+	ret = drm_dev_init(dev, driver, parent);
+	if (ret)
+		return ret;
+
+	ret = devm_add_action(parent, devm_drm_dev_init_release, dev);
+	if (ret)
+		devm_drm_dev_init_release(dev);
+
+	return ret;
+}
+EXPORT_SYMBOL(devm_drm_dev_init);
+
 /**
  * drm_dev_fini - Finalize a dead DRM device
  * @dev: DRM device
@@ -596,17 +788,19 @@ void drm_dev_fini(struct drm_device *dev)
 		drm_gem_destroy(dev);
 
 	drm_legacy_ctxbitmap_cleanup(dev);
-	drm_ht_remove(&dev->map_hash);
+	drm_legacy_remove_map_hash(dev);
 	drm_fs_inode_free(dev->anon_inode);
 
 	drm_minor_free(dev, DRM_MINOR_PRIMARY);
 	drm_minor_free(dev, DRM_MINOR_RENDER);
 
+	put_device(dev->dev);
+
 	mutex_destroy(&dev->master_mutex);
-	mutex_destroy(&dev->ctxlist_mutex);
 	mutex_destroy(&dev->clientlist_mutex);
 	mutex_destroy(&dev->filelist_mutex);
 	mutex_destroy(&dev->struct_mutex);
+	drm_legacy_destroy_members(dev);
 	kfree(dev->unique);
 }
 EXPORT_SYMBOL(drm_dev_fini);
@@ -840,8 +1034,6 @@ EXPORT_SYMBOL(drm_dev_register);
  */
 void drm_dev_unregister(struct drm_device *dev)
 {
-	struct drm_map_list *r_list, *list_temp;
-
 	if (drm_core_check_feature(dev, DRIVER_LEGACY))
 		drm_lastclose(dev);
 
@@ -858,8 +1050,7 @@ void drm_dev_unregister(struct drm_device *dev)
 	if (dev->agp)
 		drm_pci_agp_destroy(dev);
 
-	list_for_each_entry_safe(r_list, list_temp, &dev->maplist, head)
-		drm_legacy_rmmap(dev, r_list->map);
+	drm_legacy_rmmaps(dev);
 
 	remove_compat_control_link(dev);
 	drm_minor_unregister(dev, DRM_MINOR_PRIMARY);
diff --git a/drivers/gpu/drm/drm_dsc.c b/drivers/gpu/drm/drm_dsc.c
index bce99f95c1a3..77f4e5ae4197 100644
--- a/drivers/gpu/drm/drm_dsc.c
+++ b/drivers/gpu/drm/drm_dsc.c
@@ -11,6 +11,7 @@
 #include <linux/init.h>
 #include <linux/errno.h>
 #include <linux/byteorder/generic.h>
+#include <drm/drm_print.h>
 #include <drm/drm_dp_helper.h>
 #include <drm/drm_dsc.h>
 
@@ -31,75 +32,74 @@
 /**
  * drm_dsc_dp_pps_header_init() - Initializes the PPS Header
  * for DisplayPort as per the DP 1.4 spec.
- * @pps_sdp: Secondary data packet for DSC Picture Parameter Set
- *           as defined in &struct drm_dsc_pps_infoframe
+ * @pps_header: Secondary data packet header for DSC Picture
+ *              Parameter Set as defined in &struct dp_sdp_header
  *
  * DP 1.4 spec defines the secondary data packet for sending the
  * picture parameter infoframes from the source to the sink.
- * This function populates the pps header defined in
- * &struct drm_dsc_pps_infoframe as per the header bytes defined
- * in &struct dp_sdp_header.
+ * This function populates the SDP header defined in
+ * &struct dp_sdp_header.
  */
-void drm_dsc_dp_pps_header_init(struct drm_dsc_pps_infoframe *pps_sdp)
+void drm_dsc_dp_pps_header_init(struct dp_sdp_header *pps_header)
 {
-	memset(&pps_sdp->pps_header, 0, sizeof(pps_sdp->pps_header));
+	memset(pps_header, 0, sizeof(*pps_header));
 
-	pps_sdp->pps_header.HB1 = DP_SDP_PPS;
-	pps_sdp->pps_header.HB2 = DP_SDP_PPS_HEADER_PAYLOAD_BYTES_MINUS_1;
+	pps_header->HB1 = DP_SDP_PPS;
+	pps_header->HB2 = DP_SDP_PPS_HEADER_PAYLOAD_BYTES_MINUS_1;
 }
 EXPORT_SYMBOL(drm_dsc_dp_pps_header_init);
 
 /**
- * drm_dsc_pps_infoframe_pack() - Populates the DSC PPS infoframe
+ * drm_dsc_pps_payload_pack() - Populates the DSC PPS
  *
- * @pps_sdp:
- * Secondary data packet for DSC Picture Parameter Set. This is defined
- * by &struct drm_dsc_pps_infoframe
+ * @pps_payload:
+ * Bitwise struct for DSC Picture Parameter Set. This is defined
+ * by &struct drm_dsc_picture_parameter_set
  * @dsc_cfg:
  * DSC Configuration data filled by driver as defined by
  * &struct drm_dsc_config
  *
- * DSC source device sends a secondary data packet filled with all the
- * picture parameter set (PPS) information required by the sink to decode
- * the compressed frame. Driver populates the dsC PPS infoframe using the DSC
- * configuration parameters in the order expected by the DSC Display Sink
- * device. For the DSC, the sink device expects the PPS payload in the big
- * endian format for the fields that span more than 1 byte.
+ * DSC source device sends a picture parameter set (PPS) containing the
+ * information required by the sink to decode the compressed frame. Driver
+ * populates the DSC PPS struct using the DSC configuration parameters in
+ * the order expected by the DSC Display Sink device. For the DSC, the sink
+ * device expects the PPS payload in big endian format for fields
+ * that span more than 1 byte.
  */
-void drm_dsc_pps_infoframe_pack(struct drm_dsc_pps_infoframe *pps_sdp,
+void drm_dsc_pps_payload_pack(struct drm_dsc_picture_parameter_set *pps_payload,
 				const struct drm_dsc_config *dsc_cfg)
 {
 	int i;
 
 	/* Protect against someone accidently changing struct size */
-	BUILD_BUG_ON(sizeof(pps_sdp->pps_payload) !=
+	BUILD_BUG_ON(sizeof(*pps_payload) !=
 		     DP_SDP_PPS_HEADER_PAYLOAD_BYTES_MINUS_1 + 1);
 
-	memset(&pps_sdp->pps_payload, 0, sizeof(pps_sdp->pps_payload));
+	memset(pps_payload, 0, sizeof(*pps_payload));
 
 	/* PPS 0 */
-	pps_sdp->pps_payload.dsc_version =
+	pps_payload->dsc_version =
 		dsc_cfg->dsc_version_minor |
 		dsc_cfg->dsc_version_major << DSC_PPS_VERSION_MAJOR_SHIFT;
 
 	/* PPS 1, 2 is 0 */
 
 	/* PPS 3 */
-	pps_sdp->pps_payload.pps_3 =
+	pps_payload->pps_3 =
 		dsc_cfg->line_buf_depth |
 		dsc_cfg->bits_per_component << DSC_PPS_BPC_SHIFT;
 
 	/* PPS 4 */
-	pps_sdp->pps_payload.pps_4 =
+	pps_payload->pps_4 =
 		((dsc_cfg->bits_per_pixel & DSC_PPS_BPP_HIGH_MASK) >>
 		 DSC_PPS_MSB_SHIFT) |
 		dsc_cfg->vbr_enable << DSC_PPS_VBR_EN_SHIFT |
-		dsc_cfg->enable422 << DSC_PPS_SIMPLE422_SHIFT |
+		dsc_cfg->simple_422 << DSC_PPS_SIMPLE422_SHIFT |
 		dsc_cfg->convert_rgb << DSC_PPS_CONVERT_RGB_SHIFT |
 		dsc_cfg->block_pred_enable << DSC_PPS_BLOCK_PRED_EN_SHIFT;
 
 	/* PPS 5 */
-	pps_sdp->pps_payload.bits_per_pixel_low =
+	pps_payload->bits_per_pixel_low =
 		(dsc_cfg->bits_per_pixel & DSC_PPS_LSB_MASK);
 
 	/*
@@ -110,103 +110,103 @@ void drm_dsc_pps_infoframe_pack(struct drm_dsc_pps_infoframe *pps_sdp,
 	 */
 
 	/* PPS 6, 7 */
-	pps_sdp->pps_payload.pic_height = cpu_to_be16(dsc_cfg->pic_height);
+	pps_payload->pic_height = cpu_to_be16(dsc_cfg->pic_height);
 
 	/* PPS 8, 9 */
-	pps_sdp->pps_payload.pic_width = cpu_to_be16(dsc_cfg->pic_width);
+	pps_payload->pic_width = cpu_to_be16(dsc_cfg->pic_width);
 
 	/* PPS 10, 11 */
-	pps_sdp->pps_payload.slice_height = cpu_to_be16(dsc_cfg->slice_height);
+	pps_payload->slice_height = cpu_to_be16(dsc_cfg->slice_height);
 
 	/* PPS 12, 13 */
-	pps_sdp->pps_payload.slice_width = cpu_to_be16(dsc_cfg->slice_width);
+	pps_payload->slice_width = cpu_to_be16(dsc_cfg->slice_width);
 
 	/* PPS 14, 15 */
-	pps_sdp->pps_payload.chunk_size = cpu_to_be16(dsc_cfg->slice_chunk_size);
+	pps_payload->chunk_size = cpu_to_be16(dsc_cfg->slice_chunk_size);
 
 	/* PPS 16 */
-	pps_sdp->pps_payload.initial_xmit_delay_high =
+	pps_payload->initial_xmit_delay_high =
 		((dsc_cfg->initial_xmit_delay &
 		  DSC_PPS_INIT_XMIT_DELAY_HIGH_MASK) >>
 		 DSC_PPS_MSB_SHIFT);
 
 	/* PPS 17 */
-	pps_sdp->pps_payload.initial_xmit_delay_low =
+	pps_payload->initial_xmit_delay_low =
 		(dsc_cfg->initial_xmit_delay & DSC_PPS_LSB_MASK);
 
 	/* PPS 18, 19 */
-	pps_sdp->pps_payload.initial_dec_delay =
+	pps_payload->initial_dec_delay =
 		cpu_to_be16(dsc_cfg->initial_dec_delay);
 
 	/* PPS 20 is 0 */
 
 	/* PPS 21 */
-	pps_sdp->pps_payload.initial_scale_value =
+	pps_payload->initial_scale_value =
 		dsc_cfg->initial_scale_value;
 
 	/* PPS 22, 23 */
-	pps_sdp->pps_payload.scale_increment_interval =
+	pps_payload->scale_increment_interval =
 		cpu_to_be16(dsc_cfg->scale_increment_interval);
 
 	/* PPS 24 */
-	pps_sdp->pps_payload.scale_decrement_interval_high =
+	pps_payload->scale_decrement_interval_high =
 		((dsc_cfg->scale_decrement_interval &
 		  DSC_PPS_SCALE_DEC_INT_HIGH_MASK) >>
 		 DSC_PPS_MSB_SHIFT);
 
 	/* PPS 25 */
-	pps_sdp->pps_payload.scale_decrement_interval_low =
+	pps_payload->scale_decrement_interval_low =
 		(dsc_cfg->scale_decrement_interval & DSC_PPS_LSB_MASK);
 
 	/* PPS 26[7:0], PPS 27[7:5] RESERVED */
 
 	/* PPS 27 */
-	pps_sdp->pps_payload.first_line_bpg_offset =
+	pps_payload->first_line_bpg_offset =
 		dsc_cfg->first_line_bpg_offset;
 
 	/* PPS 28, 29 */
-	pps_sdp->pps_payload.nfl_bpg_offset =
+	pps_payload->nfl_bpg_offset =
 		cpu_to_be16(dsc_cfg->nfl_bpg_offset);
 
 	/* PPS 30, 31 */
-	pps_sdp->pps_payload.slice_bpg_offset =
+	pps_payload->slice_bpg_offset =
 		cpu_to_be16(dsc_cfg->slice_bpg_offset);
 
 	/* PPS 32, 33 */
-	pps_sdp->pps_payload.initial_offset =
+	pps_payload->initial_offset =
 		cpu_to_be16(dsc_cfg->initial_offset);
 
 	/* PPS 34, 35 */
-	pps_sdp->pps_payload.final_offset = cpu_to_be16(dsc_cfg->final_offset);
+	pps_payload->final_offset = cpu_to_be16(dsc_cfg->final_offset);
 
 	/* PPS 36 */
-	pps_sdp->pps_payload.flatness_min_qp = dsc_cfg->flatness_min_qp;
+	pps_payload->flatness_min_qp = dsc_cfg->flatness_min_qp;
 
 	/* PPS 37 */
-	pps_sdp->pps_payload.flatness_max_qp = dsc_cfg->flatness_max_qp;
+	pps_payload->flatness_max_qp = dsc_cfg->flatness_max_qp;
 
 	/* PPS 38, 39 */
-	pps_sdp->pps_payload.rc_model_size =
+	pps_payload->rc_model_size =
 		cpu_to_be16(DSC_RC_MODEL_SIZE_CONST);
 
 	/* PPS 40 */
-	pps_sdp->pps_payload.rc_edge_factor = DSC_RC_EDGE_FACTOR_CONST;
+	pps_payload->rc_edge_factor = DSC_RC_EDGE_FACTOR_CONST;
 
 	/* PPS 41 */
-	pps_sdp->pps_payload.rc_quant_incr_limit0 =
+	pps_payload->rc_quant_incr_limit0 =
 		dsc_cfg->rc_quant_incr_limit0;
 
 	/* PPS 42 */
-	pps_sdp->pps_payload.rc_quant_incr_limit1 =
+	pps_payload->rc_quant_incr_limit1 =
 		dsc_cfg->rc_quant_incr_limit1;
 
 	/* PPS 43 */
-	pps_sdp->pps_payload.rc_tgt_offset = DSC_RC_TGT_OFFSET_LO_CONST |
+	pps_payload->rc_tgt_offset = DSC_RC_TGT_OFFSET_LO_CONST |
 		DSC_RC_TGT_OFFSET_HI_CONST << DSC_PPS_RC_TGT_OFFSET_HI_SHIFT;
 
 	/* PPS 44 - 57 */
 	for (i = 0; i < DSC_NUM_BUF_RANGES - 1; i++)
-		pps_sdp->pps_payload.rc_buf_thresh[i] =
+		pps_payload->rc_buf_thresh[i] =
 			dsc_cfg->rc_buf_thresh[i];
 
 	/* PPS 58 - 87 */
@@ -215,32 +215,181 @@ void drm_dsc_pps_infoframe_pack(struct drm_dsc_pps_infoframe *pps_sdp,
 	 * are as follows: Min_qp[15:11], max_qp[10:6], offset[5:0]
 	 */
 	for (i = 0; i < DSC_NUM_BUF_RANGES; i++) {
-		pps_sdp->pps_payload.rc_range_parameters[i] =
+		pps_payload->rc_range_parameters[i] =
 			((dsc_cfg->rc_range_params[i].range_min_qp <<
 			  DSC_PPS_RC_RANGE_MINQP_SHIFT) |
 			 (dsc_cfg->rc_range_params[i].range_max_qp <<
 			  DSC_PPS_RC_RANGE_MAXQP_SHIFT) |
 			 (dsc_cfg->rc_range_params[i].range_bpg_offset));
-		pps_sdp->pps_payload.rc_range_parameters[i] =
-			cpu_to_be16(pps_sdp->pps_payload.rc_range_parameters[i]);
+		pps_payload->rc_range_parameters[i] =
+			cpu_to_be16(pps_payload->rc_range_parameters[i]);
 	}
 
 	/* PPS 88 */
-	pps_sdp->pps_payload.native_422_420 = dsc_cfg->native_422 |
+	pps_payload->native_422_420 = dsc_cfg->native_422 |
 		dsc_cfg->native_420 << DSC_PPS_NATIVE_420_SHIFT;
 
 	/* PPS 89 */
-	pps_sdp->pps_payload.second_line_bpg_offset =
+	pps_payload->second_line_bpg_offset =
 		dsc_cfg->second_line_bpg_offset;
 
 	/* PPS 90, 91 */
-	pps_sdp->pps_payload.nsl_bpg_offset =
+	pps_payload->nsl_bpg_offset =
 		cpu_to_be16(dsc_cfg->nsl_bpg_offset);
 
 	/* PPS 92, 93 */
-	pps_sdp->pps_payload.second_line_offset_adj =
+	pps_payload->second_line_offset_adj =
 		cpu_to_be16(dsc_cfg->second_line_offset_adj);
 
 	/* PPS 94 - 127 are O */
 }
-EXPORT_SYMBOL(drm_dsc_pps_infoframe_pack);
+EXPORT_SYMBOL(drm_dsc_pps_payload_pack);
+
+/**
+ * drm_dsc_compute_rc_parameters() - Write rate control
+ * parameters to the dsc configuration defined in
+ * &struct drm_dsc_config in accordance with the DSC 1.2
+ * specification. Some configuration fields must be present
+ * beforehand.
+ *
+ * @vdsc_cfg:
+ * DSC Configuration data partially filled by driver
+ */
+int drm_dsc_compute_rc_parameters(struct drm_dsc_config *vdsc_cfg)
+{
+	unsigned long groups_per_line = 0;
+	unsigned long groups_total = 0;
+	unsigned long num_extra_mux_bits = 0;
+	unsigned long slice_bits = 0;
+	unsigned long hrd_delay = 0;
+	unsigned long final_scale = 0;
+	unsigned long rbs_min = 0;
+
+	if (vdsc_cfg->native_420 || vdsc_cfg->native_422) {
+		/* Number of groups used to code each line of a slice */
+		groups_per_line = DIV_ROUND_UP(vdsc_cfg->slice_width / 2,
+					       DSC_RC_PIXELS_PER_GROUP);
+
+		/* chunksize in Bytes */
+		vdsc_cfg->slice_chunk_size = DIV_ROUND_UP(vdsc_cfg->slice_width / 2 *
+							  vdsc_cfg->bits_per_pixel,
+							  (8 * 16));
+	} else {
+		/* Number of groups used to code each line of a slice */
+		groups_per_line = DIV_ROUND_UP(vdsc_cfg->slice_width,
+					       DSC_RC_PIXELS_PER_GROUP);
+
+		/* chunksize in Bytes */
+		vdsc_cfg->slice_chunk_size = DIV_ROUND_UP(vdsc_cfg->slice_width *
+							  vdsc_cfg->bits_per_pixel,
+							  (8 * 16));
+	}
+
+	if (vdsc_cfg->convert_rgb)
+		num_extra_mux_bits = 3 * (vdsc_cfg->mux_word_size +
+					  (4 * vdsc_cfg->bits_per_component + 4)
+					  - 2);
+	else if (vdsc_cfg->native_422)
+		num_extra_mux_bits = 4 * vdsc_cfg->mux_word_size +
+			(4 * vdsc_cfg->bits_per_component + 4) +
+			3 * (4 * vdsc_cfg->bits_per_component) - 2;
+	else
+		num_extra_mux_bits = 3 * vdsc_cfg->mux_word_size +
+			(4 * vdsc_cfg->bits_per_component + 4) +
+			2 * (4 * vdsc_cfg->bits_per_component) - 2;
+	/* Number of bits in one Slice */
+	slice_bits = 8 * vdsc_cfg->slice_chunk_size * vdsc_cfg->slice_height;
+
+	while ((num_extra_mux_bits > 0) &&
+	       ((slice_bits - num_extra_mux_bits) % vdsc_cfg->mux_word_size))
+		num_extra_mux_bits--;
+
+	if (groups_per_line < vdsc_cfg->initial_scale_value - 8)
+		vdsc_cfg->initial_scale_value = groups_per_line + 8;
+
+	/* scale_decrement_interval calculation according to DSC spec 1.11 */
+	if (vdsc_cfg->initial_scale_value > 8)
+		vdsc_cfg->scale_decrement_interval = groups_per_line /
+			(vdsc_cfg->initial_scale_value - 8);
+	else
+		vdsc_cfg->scale_decrement_interval = DSC_SCALE_DECREMENT_INTERVAL_MAX;
+
+	vdsc_cfg->final_offset = vdsc_cfg->rc_model_size -
+		(vdsc_cfg->initial_xmit_delay *
+		 vdsc_cfg->bits_per_pixel + 8) / 16 + num_extra_mux_bits;
+
+	if (vdsc_cfg->final_offset >= vdsc_cfg->rc_model_size) {
+		DRM_DEBUG_KMS("FinalOfs < RcModelSze for this InitialXmitDelay\n");
+		return -ERANGE;
+	}
+
+	final_scale = (vdsc_cfg->rc_model_size * 8) /
+		(vdsc_cfg->rc_model_size - vdsc_cfg->final_offset);
+	if (vdsc_cfg->slice_height > 1)
+		/*
+		 * NflBpgOffset is 16 bit value with 11 fractional bits
+		 * hence we multiply by 2^11 for preserving the
+		 * fractional part
+		 */
+		vdsc_cfg->nfl_bpg_offset = DIV_ROUND_UP((vdsc_cfg->first_line_bpg_offset << 11),
+							(vdsc_cfg->slice_height - 1));
+	else
+		vdsc_cfg->nfl_bpg_offset = 0;
+
+	/* 2^16 - 1 */
+	if (vdsc_cfg->nfl_bpg_offset > 65535) {
+		DRM_DEBUG_KMS("NflBpgOffset is too large for this slice height\n");
+		return -ERANGE;
+	}
+
+	/* Number of groups used to code the entire slice */
+	groups_total = groups_per_line * vdsc_cfg->slice_height;
+
+	/* slice_bpg_offset is 16 bit value with 11 fractional bits */
+	vdsc_cfg->slice_bpg_offset = DIV_ROUND_UP(((vdsc_cfg->rc_model_size -
+						    vdsc_cfg->initial_offset +
+						    num_extra_mux_bits) << 11),
+						  groups_total);
+
+	if (final_scale > 9) {
+		/*
+		 * ScaleIncrementInterval =
+		 * finaloffset/((NflBpgOffset + SliceBpgOffset)*8(finalscale - 1.125))
+		 * as (NflBpgOffset + SliceBpgOffset) has 11 bit fractional value,
+		 * we need divide by 2^11 from pstDscCfg values
+		 */
+		vdsc_cfg->scale_increment_interval =
+				(vdsc_cfg->final_offset * (1 << 11)) /
+				((vdsc_cfg->nfl_bpg_offset +
+				vdsc_cfg->slice_bpg_offset) *
+				(final_scale - 9));
+	} else {
+		/*
+		 * If finalScaleValue is less than or equal to 9, a value of 0 should
+		 * be used to disable the scale increment at the end of the slice
+		 */
+		vdsc_cfg->scale_increment_interval = 0;
+	}
+
+	if (vdsc_cfg->scale_increment_interval > 65535) {
+		DRM_DEBUG_KMS("ScaleIncrementInterval is large for slice height\n");
+		return -ERANGE;
+	}
+
+	/*
+	 * DSC spec mentions that bits_per_pixel specifies the target
+	 * bits/pixel (bpp) rate that is used by the encoder,
+	 * in steps of 1/16 of a bit per pixel
+	 */
+	rbs_min = vdsc_cfg->rc_model_size - vdsc_cfg->initial_offset +
+		DIV_ROUND_UP(vdsc_cfg->initial_xmit_delay *
+			     vdsc_cfg->bits_per_pixel, 16) +
+		groups_per_line * vdsc_cfg->first_line_bpg_offset;
+
+	hrd_delay = DIV_ROUND_UP((rbs_min * 16), vdsc_cfg->bits_per_pixel);
+	vdsc_cfg->rc_bits = (hrd_delay * vdsc_cfg->bits_per_pixel) / 16;
+	vdsc_cfg->initial_dec_delay = hrd_delay - vdsc_cfg->initial_xmit_delay;
+
+	return 0;
+}
+EXPORT_SYMBOL(drm_dsc_compute_rc_parameters);
diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
index 990b1909f9d7..649cfd8b4200 100644
--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -68,8 +68,6 @@
  * maximum size and use that.
  */
 #define EDID_QUIRK_DETAILED_USE_MAXIMUM_SIZE	(1 << 4)
-/* Monitor forgot to set the first detailed is preferred bit. */
-#define EDID_QUIRK_FIRST_DETAILED_PREFERRED	(1 << 5)
 /* use +hsync +vsync for detailed mode */
 #define EDID_QUIRK_DETAILED_SYNC_PP		(1 << 6)
 /* Force reduced-blanking timings for detailed modes */
@@ -107,8 +105,6 @@ static const struct edid_quirk {
 	{ "ACR", 44358, EDID_QUIRK_PREFER_LARGE_60 },
 	/* Acer F51 */
 	{ "API", 0x7602, EDID_QUIRK_PREFER_LARGE_60 },
-	/* Unknown Acer */
-	{ "ACR", 2423, EDID_QUIRK_FIRST_DETAILED_PREFERRED },
 
 	/* AEO model 0 reports 8 bpc, but is a 6 bpc panel */
 	{ "AEO", 0, EDID_QUIRK_FORCE_6BPC },
@@ -145,12 +141,6 @@ static const struct edid_quirk {
 	{ "LPL", 0, EDID_QUIRK_DETAILED_USE_MAXIMUM_SIZE },
 	{ "LPL", 0x2a00, EDID_QUIRK_DETAILED_USE_MAXIMUM_SIZE },
 
-	/* Philips 107p5 CRT */
-	{ "PHL", 57364, EDID_QUIRK_FIRST_DETAILED_PREFERRED },
-
-	/* Proview AY765C */
-	{ "PTS", 765, EDID_QUIRK_FIRST_DETAILED_PREFERRED },
-
 	/* Samsung SyncMaster 205BW.  Note: irony */
 	{ "SAM", 541, EDID_QUIRK_DETAILED_SYNC_PP },
 	/* Samsung SyncMaster 22[5-6]BW */
@@ -172,6 +162,25 @@ static const struct edid_quirk {
 	/* Rotel RSX-1058 forwards sink's EDID but only does HDMI 1.1*/
 	{ "ETR", 13896, EDID_QUIRK_FORCE_8BPC },
 
+	/* Valve Index Headset */
+	{ "VLV", 0x91a8, EDID_QUIRK_NON_DESKTOP },
+	{ "VLV", 0x91b0, EDID_QUIRK_NON_DESKTOP },
+	{ "VLV", 0x91b1, EDID_QUIRK_NON_DESKTOP },
+	{ "VLV", 0x91b2, EDID_QUIRK_NON_DESKTOP },
+	{ "VLV", 0x91b3, EDID_QUIRK_NON_DESKTOP },
+	{ "VLV", 0x91b4, EDID_QUIRK_NON_DESKTOP },
+	{ "VLV", 0x91b5, EDID_QUIRK_NON_DESKTOP },
+	{ "VLV", 0x91b6, EDID_QUIRK_NON_DESKTOP },
+	{ "VLV", 0x91b7, EDID_QUIRK_NON_DESKTOP },
+	{ "VLV", 0x91b8, EDID_QUIRK_NON_DESKTOP },
+	{ "VLV", 0x91b9, EDID_QUIRK_NON_DESKTOP },
+	{ "VLV", 0x91ba, EDID_QUIRK_NON_DESKTOP },
+	{ "VLV", 0x91bb, EDID_QUIRK_NON_DESKTOP },
+	{ "VLV", 0x91bc, EDID_QUIRK_NON_DESKTOP },
+	{ "VLV", 0x91bd, EDID_QUIRK_NON_DESKTOP },
+	{ "VLV", 0x91be, EDID_QUIRK_NON_DESKTOP },
+	{ "VLV", 0x91bf, EDID_QUIRK_NON_DESKTOP },
+
 	/* HTC Vive and Vive Pro VR Headsets */
 	{ "HVR", 0xaa01, EDID_QUIRK_NON_DESKTOP },
 	{ "HVR", 0xaa02, EDID_QUIRK_NON_DESKTOP },
@@ -193,6 +202,12 @@ static const struct edid_quirk {
 
 	/* Sony PlayStation VR Headset */
 	{ "SNY", 0x0704, EDID_QUIRK_NON_DESKTOP },
+
+	/* Sensics VR Headsets */
+	{ "SEN", 0x1019, EDID_QUIRK_NON_DESKTOP },
+
+	/* OSVR HDK and HDK2 VR Headsets */
+	{ "SVR", 0x1019, EDID_QUIRK_NON_DESKTOP },
 };
 
 /*
@@ -4924,6 +4939,76 @@ drm_hdmi_avi_infoframe_from_display_mode(struct hdmi_avi_infoframe *frame,
 }
 EXPORT_SYMBOL(drm_hdmi_avi_infoframe_from_display_mode);
 
+/* HDMI Colorspace Spec Definitions */
+#define FULL_COLORIMETRY_MASK		0x1FF
+#define NORMAL_COLORIMETRY_MASK		0x3
+#define EXTENDED_COLORIMETRY_MASK	0x7
+#define EXTENDED_ACE_COLORIMETRY_MASK	0xF
+
+#define C(x) ((x) << 0)
+#define EC(x) ((x) << 2)
+#define ACE(x) ((x) << 5)
+
+#define HDMI_COLORIMETRY_NO_DATA		0x0
+#define HDMI_COLORIMETRY_SMPTE_170M_YCC		(C(1) | EC(0) | ACE(0))
+#define HDMI_COLORIMETRY_BT709_YCC		(C(2) | EC(0) | ACE(0))
+#define HDMI_COLORIMETRY_XVYCC_601		(C(3) | EC(0) | ACE(0))
+#define HDMI_COLORIMETRY_XVYCC_709		(C(3) | EC(1) | ACE(0))
+#define HDMI_COLORIMETRY_SYCC_601		(C(3) | EC(2) | ACE(0))
+#define HDMI_COLORIMETRY_OPYCC_601		(C(3) | EC(3) | ACE(0))
+#define HDMI_COLORIMETRY_OPRGB			(C(3) | EC(4) | ACE(0))
+#define HDMI_COLORIMETRY_BT2020_CYCC		(C(3) | EC(5) | ACE(0))
+#define HDMI_COLORIMETRY_BT2020_RGB		(C(3) | EC(6) | ACE(0))
+#define HDMI_COLORIMETRY_BT2020_YCC		(C(3) | EC(6) | ACE(0))
+#define HDMI_COLORIMETRY_DCI_P3_RGB_D65		(C(3) | EC(7) | ACE(0))
+#define HDMI_COLORIMETRY_DCI_P3_RGB_THEATER	(C(3) | EC(7) | ACE(1))
+
+static const u32 hdmi_colorimetry_val[] = {
+	[DRM_MODE_COLORIMETRY_NO_DATA] = HDMI_COLORIMETRY_NO_DATA,
+	[DRM_MODE_COLORIMETRY_SMPTE_170M_YCC] = HDMI_COLORIMETRY_SMPTE_170M_YCC,
+	[DRM_MODE_COLORIMETRY_BT709_YCC] = HDMI_COLORIMETRY_BT709_YCC,
+	[DRM_MODE_COLORIMETRY_XVYCC_601] = HDMI_COLORIMETRY_XVYCC_601,
+	[DRM_MODE_COLORIMETRY_XVYCC_709] = HDMI_COLORIMETRY_XVYCC_709,
+	[DRM_MODE_COLORIMETRY_SYCC_601] = HDMI_COLORIMETRY_SYCC_601,
+	[DRM_MODE_COLORIMETRY_OPYCC_601] = HDMI_COLORIMETRY_OPYCC_601,
+	[DRM_MODE_COLORIMETRY_OPRGB] = HDMI_COLORIMETRY_OPRGB,
+	[DRM_MODE_COLORIMETRY_BT2020_CYCC] = HDMI_COLORIMETRY_BT2020_CYCC,
+	[DRM_MODE_COLORIMETRY_BT2020_RGB] = HDMI_COLORIMETRY_BT2020_RGB,
+	[DRM_MODE_COLORIMETRY_BT2020_YCC] = HDMI_COLORIMETRY_BT2020_YCC,
+};
+
+#undef C
+#undef EC
+#undef ACE
+
+/**
+ * drm_hdmi_avi_infoframe_colorspace() - fill the HDMI AVI infoframe
+ *                                       colorspace information
+ * @frame: HDMI AVI infoframe
+ * @conn_state: connector state
+ */
+void
+drm_hdmi_avi_infoframe_colorspace(struct hdmi_avi_infoframe *frame,
+				  const struct drm_connector_state *conn_state)
+{
+	u32 colorimetry_val;
+	u32 colorimetry_index = conn_state->colorspace & FULL_COLORIMETRY_MASK;
+
+	if (colorimetry_index >= ARRAY_SIZE(hdmi_colorimetry_val))
+		colorimetry_val = HDMI_COLORIMETRY_NO_DATA;
+	else
+		colorimetry_val = hdmi_colorimetry_val[colorimetry_index];
+
+	frame->colorimetry = colorimetry_val & NORMAL_COLORIMETRY_MASK;
+	/*
+	 * ToDo: Extend it for ACE formats as well. Modify the infoframe
+	 * structure and extend it in drivers/video/hdmi
+	 */
+	frame->extended_colorimetry = (colorimetry_val >> 2) &
+					EXTENDED_COLORIMETRY_MASK;
+}
+EXPORT_SYMBOL(drm_hdmi_avi_infoframe_colorspace);
+
 /**
  * drm_hdmi_avi_infoframe_quant_range() - fill the HDMI AVI infoframe
  *                                        quantization range information
diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index af2ab640cadb..498f95c3e81d 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -639,20 +639,19 @@ static struct sysrq_key_op sysrq_drm_fb_helper_restore_op = { };
 static void dpms_legacy(struct drm_fb_helper *fb_helper, int dpms_mode)
 {
 	struct drm_device *dev = fb_helper->dev;
-	struct drm_crtc *crtc;
 	struct drm_connector *connector;
+	struct drm_mode_set *modeset;
 	int i, j;
 
 	drm_modeset_lock_all(dev);
 	for (i = 0; i < fb_helper->crtc_count; i++) {
-		crtc = fb_helper->crtc_info[i].mode_set.crtc;
+		modeset = &fb_helper->crtc_info[i].mode_set;
 
-		if (!crtc->enabled)
+		if (!modeset->crtc->enabled)
 			continue;
 
-		/* Walk the connectors & encoders on this fb turning them on/off */
-		drm_fb_helper_for_each_connector(fb_helper, j) {
-			connector = fb_helper->connector_info[j]->connector;
+		for (j = 0; j < modeset->num_connectors; j++) {
+			connector = modeset->connectors[j];
 			connector->funcs->dpms(connector, dpms_mode);
 			drm_object_property_set_value(&connector->base,
 				dev->mode_config.dpms_property, dpms_mode);
@@ -934,6 +933,7 @@ struct fb_info *drm_fb_helper_alloc_fbi(struct drm_fb_helper *fb_helper)
 	}
 
 	fb_helper->fbdev = info;
+	info->skip_vt_switch = true;
 
 	return info;
 
@@ -1873,7 +1873,6 @@ static int drm_fb_helper_single_fb_probe(struct drm_fb_helper *fb_helper,
 	int crtc_count = 0;
 	int i;
 	struct drm_fb_helper_surface_size sizes;
-	int gamma_size = 0;
 	int best_depth = 0;
 
 	memset(&sizes, 0, sizeof(struct drm_fb_helper_surface_size));
@@ -1889,7 +1888,6 @@ static int drm_fb_helper_single_fb_probe(struct drm_fb_helper *fb_helper,
 	if (preferred_bpp != sizes.surface_bpp)
 		sizes.surface_depth = sizes.surface_bpp = preferred_bpp;
 
-	/* first up get a count of crtcs now in use and new min/maxes width/heights */
 	drm_fb_helper_for_each_connector(fb_helper, i) {
 		struct drm_fb_helper_connector *fb_helper_conn = fb_helper->connector_info[i];
 		struct drm_cmdline_mode *cmdline_mode;
@@ -1969,6 +1967,7 @@ static int drm_fb_helper_single_fb_probe(struct drm_fb_helper *fb_helper,
 		sizes.surface_depth = best_depth;
 	}
 
+	/* first up get a count of crtcs now in use and new min/maxes width/heights */
 	crtc_count = 0;
 	for (i = 0; i < fb_helper->crtc_count; i++) {
 		struct drm_display_mode *desired_mode;
@@ -1991,9 +1990,6 @@ static int drm_fb_helper_single_fb_probe(struct drm_fb_helper *fb_helper,
 		x = fb_helper->crtc_info[i].x;
 		y = fb_helper->crtc_info[i].y;
 
-		if (gamma_size == 0)
-			gamma_size = fb_helper->crtc_info[i].mode_set.crtc->gamma_size;
-
 		sizes.surface_width  = max_t(u32, desired_mode->hdisplay + x, sizes.surface_width);
 		sizes.surface_height = max_t(u32, desired_mode->vdisplay + y, sizes.surface_height);
 
@@ -2036,21 +2032,8 @@ static int drm_fb_helper_single_fb_probe(struct drm_fb_helper *fb_helper,
 	return 0;
 }
 
-/**
- * drm_fb_helper_fill_fix - initializes fixed fbdev information
- * @info: fbdev registered by the helper
- * @pitch: desired pitch
- * @depth: desired depth
- *
- * Helper to fill in the fixed fbdev information useful for a non-accelerated
- * fbdev emulations. Drivers which support acceleration methods which impose
- * additional constraints need to set up their own limits.
- *
- * Drivers should call this (or their equivalent setup code) from their
- * &drm_fb_helper_funcs.fb_probe callback.
- */
-void drm_fb_helper_fill_fix(struct fb_info *info, uint32_t pitch,
-			    uint32_t depth)
+static void drm_fb_helper_fill_fix(struct fb_info *info, uint32_t pitch,
+				   uint32_t depth)
 {
 	info->fix.type = FB_TYPE_PACKED_PIXELS;
 	info->fix.visual = depth == 8 ? FB_VISUAL_PSEUDOCOLOR :
@@ -2065,24 +2048,10 @@ void drm_fb_helper_fill_fix(struct fb_info *info, uint32_t pitch,
 
 	info->fix.line_length = pitch;
 }
-EXPORT_SYMBOL(drm_fb_helper_fill_fix);
 
-/**
- * drm_fb_helper_fill_var - initalizes variable fbdev information
- * @info: fbdev instance to set up
- * @fb_helper: fb helper instance to use as template
- * @fb_width: desired fb width
- * @fb_height: desired fb height
- *
- * Sets up the variable fbdev metainformation from the given fb helper instance
- * and the drm framebuffer allocated in &drm_fb_helper.fb.
- *
- * Drivers should call this (or their equivalent setup code) from their
- * &drm_fb_helper_funcs.fb_probe callback after having allocated the fbdev
- * backing storage framebuffer.
- */
-void drm_fb_helper_fill_var(struct fb_info *info, struct drm_fb_helper *fb_helper,
-			    uint32_t fb_width, uint32_t fb_height)
+static void drm_fb_helper_fill_var(struct fb_info *info,
+				   struct drm_fb_helper *fb_helper,
+				   uint32_t fb_width, uint32_t fb_height)
 {
 	struct drm_framebuffer *fb = fb_helper->fb;
 
@@ -2102,7 +2071,36 @@ void drm_fb_helper_fill_var(struct fb_info *info, struct drm_fb_helper *fb_helpe
 	info->var.xres = fb_width;
 	info->var.yres = fb_height;
 }
-EXPORT_SYMBOL(drm_fb_helper_fill_var);
+
+/**
+ * drm_fb_helper_fill_info - initializes fbdev information
+ * @info: fbdev instance to set up
+ * @fb_helper: fb helper instance to use as template
+ * @sizes: describes fbdev size and scanout surface size
+ *
+ * Sets up the variable and fixed fbdev metainformation from the given fb helper
+ * instance and the drm framebuffer allocated in &drm_fb_helper.fb.
+ *
+ * Drivers should call this (or their equivalent setup code) from their
+ * &drm_fb_helper_funcs.fb_probe callback after having allocated the fbdev
+ * backing storage framebuffer.
+ */
+void drm_fb_helper_fill_info(struct fb_info *info,
+			     struct drm_fb_helper *fb_helper,
+			     struct drm_fb_helper_surface_size *sizes)
+{
+	struct drm_framebuffer *fb = fb_helper->fb;
+
+	drm_fb_helper_fill_fix(info, fb->pitches[0], fb->format->depth);
+	drm_fb_helper_fill_var(info, fb_helper,
+			       sizes->fb_width, sizes->fb_height);
+
+	info->par = fb_helper;
+	snprintf(info->fix.id, sizeof(info->fix.id), "%sdrmfb",
+		 fb_helper->dev->driver->name);
+
+}
+EXPORT_SYMBOL(drm_fb_helper_fill_info);
 
 static int drm_fb_helper_probe_connector_modes(struct drm_fb_helper *fb_helper,
 						uint32_t maxX,
@@ -2561,6 +2559,195 @@ static void drm_setup_crtc_rotation(struct drm_fb_helper *fb_helper,
 	fb_helper->sw_rotations |= DRM_MODE_ROTATE_0;
 }
 
+static struct drm_fb_helper_crtc *
+drm_fb_helper_crtc(struct drm_fb_helper *fb_helper, struct drm_crtc *crtc)
+{
+	int i;
+
+	for (i = 0; i < fb_helper->crtc_count; i++)
+		if (fb_helper->crtc_info[i].mode_set.crtc == crtc)
+			return &fb_helper->crtc_info[i];
+
+	return NULL;
+}
+
+/* Try to read the BIOS display configuration and use it for the initial config */
+static bool drm_fb_helper_firmware_config(struct drm_fb_helper *fb_helper,
+					  struct drm_fb_helper_crtc **crtcs,
+					  struct drm_display_mode **modes,
+					  struct drm_fb_offset *offsets,
+					  bool *enabled, int width, int height)
+{
+	struct drm_device *dev = fb_helper->dev;
+	unsigned int count = min(fb_helper->connector_count, BITS_PER_LONG);
+	unsigned long conn_configured, conn_seq, mask;
+	int i, j;
+	bool *save_enabled;
+	bool fallback = true, ret = true;
+	int num_connectors_enabled = 0;
+	int num_connectors_detected = 0;
+	struct drm_modeset_acquire_ctx ctx;
+
+	if (!drm_drv_uses_atomic_modeset(dev))
+		return false;
+
+	save_enabled = kcalloc(count, sizeof(bool), GFP_KERNEL);
+	if (!save_enabled)
+		return false;
+
+	drm_modeset_acquire_init(&ctx, 0);
+
+	while (drm_modeset_lock_all_ctx(dev, &ctx) != 0)
+		drm_modeset_backoff(&ctx);
+
+	memcpy(save_enabled, enabled, count);
+	mask = GENMASK(count - 1, 0);
+	conn_configured = 0;
+retry:
+	conn_seq = conn_configured;
+	for (i = 0; i < count; i++) {
+		struct drm_fb_helper_connector *fb_conn;
+		struct drm_connector *connector;
+		struct drm_encoder *encoder;
+		struct drm_fb_helper_crtc *new_crtc;
+
+		fb_conn = fb_helper->connector_info[i];
+		connector = fb_conn->connector;
+
+		if (conn_configured & BIT(i))
+			continue;
+
+		if (conn_seq == 0 && !connector->has_tile)
+			continue;
+
+		if (connector->status == connector_status_connected)
+			num_connectors_detected++;
+
+		if (!enabled[i]) {
+			DRM_DEBUG_KMS("connector %s not enabled, skipping\n",
+				      connector->name);
+			conn_configured |= BIT(i);
+			continue;
+		}
+
+		if (connector->force == DRM_FORCE_OFF) {
+			DRM_DEBUG_KMS("connector %s is disabled by user, skipping\n",
+				      connector->name);
+			enabled[i] = false;
+			continue;
+		}
+
+		encoder = connector->state->best_encoder;
+		if (!encoder || WARN_ON(!connector->state->crtc)) {
+			if (connector->force > DRM_FORCE_OFF)
+				goto bail;
+
+			DRM_DEBUG_KMS("connector %s has no encoder or crtc, skipping\n",
+				      connector->name);
+			enabled[i] = false;
+			conn_configured |= BIT(i);
+			continue;
+		}
+
+		num_connectors_enabled++;
+
+		new_crtc = drm_fb_helper_crtc(fb_helper, connector->state->crtc);
+
+		/*
+		 * Make sure we're not trying to drive multiple connectors
+		 * with a single CRTC, since our cloning support may not
+		 * match the BIOS.
+		 */
+		for (j = 0; j < count; j++) {
+			if (crtcs[j] == new_crtc) {
+				DRM_DEBUG_KMS("fallback: cloned configuration\n");
+				goto bail;
+			}
+		}
+
+		DRM_DEBUG_KMS("looking for cmdline mode on connector %s\n",
+			      connector->name);
+
+		/* go for command line mode first */
+		modes[i] = drm_pick_cmdline_mode(fb_conn);
+
+		/* try for preferred next */
+		if (!modes[i]) {
+			DRM_DEBUG_KMS("looking for preferred mode on connector %s %d\n",
+				      connector->name, connector->has_tile);
+			modes[i] = drm_has_preferred_mode(fb_conn, width,
+							  height);
+		}
+
+		/* No preferred mode marked by the EDID? Are there any modes? */
+		if (!modes[i] && !list_empty(&connector->modes)) {
+			DRM_DEBUG_KMS("using first mode listed on connector %s\n",
+				      connector->name);
+			modes[i] = list_first_entry(&connector->modes,
+						    struct drm_display_mode,
+						    head);
+		}
+
+		/* last resort: use current mode */
+		if (!modes[i]) {
+			/*
+			 * IMPORTANT: We want to use the adjusted mode (i.e.
+			 * after the panel fitter upscaling) as the initial
+			 * config, not the input mode, which is what crtc->mode
+			 * usually contains. But since our current
+			 * code puts a mode derived from the post-pfit timings
+			 * into crtc->mode this works out correctly.
+			 *
+			 * This is crtc->mode and not crtc->state->mode for the
+			 * fastboot check to work correctly.
+			 */
+			DRM_DEBUG_KMS("looking for current mode on connector %s\n",
+				      connector->name);
+			modes[i] = &connector->state->crtc->mode;
+		}
+		crtcs[i] = new_crtc;
+
+		DRM_DEBUG_KMS("connector %s on [CRTC:%d:%s]: %dx%d%s\n",
+			      connector->name,
+			      connector->state->crtc->base.id,
+			      connector->state->crtc->name,
+			      modes[i]->hdisplay, modes[i]->vdisplay,
+			      modes[i]->flags & DRM_MODE_FLAG_INTERLACE ? "i" : "");
+
+		fallback = false;
+		conn_configured |= BIT(i);
+	}
+
+	if ((conn_configured & mask) != mask && conn_configured != conn_seq)
+		goto retry;
+
+	/*
+	 * If the BIOS didn't enable everything it could, fall back to have the
+	 * same user experiencing of lighting up as much as possible like the
+	 * fbdev helper library.
+	 */
+	if (num_connectors_enabled != num_connectors_detected &&
+	    num_connectors_enabled < dev->mode_config.num_crtc) {
+		DRM_DEBUG_KMS("fallback: Not all outputs enabled\n");
+		DRM_DEBUG_KMS("Enabled: %i, detected: %i\n", num_connectors_enabled,
+			      num_connectors_detected);
+		fallback = true;
+	}
+
+	if (fallback) {
+bail:
+		DRM_DEBUG_KMS("Not using firmware configuration\n");
+		memcpy(enabled, save_enabled, count);
+		ret = false;
+	}
+
+	drm_modeset_drop_locks(&ctx);
+	drm_modeset_acquire_fini(&ctx);
+
+	kfree(save_enabled);
+	return ret;
+}
+
 static void drm_setup_crtcs(struct drm_fb_helper *fb_helper,
 			    u32 width, u32 height)
 {
@@ -2593,10 +2780,8 @@ static void drm_setup_crtcs(struct drm_fb_helper *fb_helper,
 		DRM_DEBUG_KMS("No connectors reported connected with modes\n");
 	drm_enable_connectors(fb_helper, enabled);
 
-	if (!(fb_helper->funcs->initial_config &&
-	      fb_helper->funcs->initial_config(fb_helper, crtcs, modes,
-					       offsets,
-					       enabled, width, height))) {
+	if (!drm_fb_helper_firmware_config(fb_helper, crtcs, modes, offsets,
+					   enabled, width, height)) {
 		memset(modes, 0, fb_helper->connector_count*sizeof(modes[0]));
 		memset(crtcs, 0, fb_helper->connector_count*sizeof(crtcs[0]));
 		memset(offsets, 0, fb_helper->connector_count*sizeof(offsets[0]));
@@ -2780,9 +2965,8 @@ __drm_fb_helper_initial_config_and_unlock(struct drm_fb_helper *fb_helper,
  *
  * This function will call down into the &drm_fb_helper_funcs.fb_probe callback
  * to let the driver allocate and initialize the fbdev info structure and the
- * drm framebuffer used to back the fbdev. drm_fb_helper_fill_var() and
- * drm_fb_helper_fill_fix() are provided as helpers to setup simple default
- * values for the fbdev info structure.
+ * drm framebuffer used to back the fbdev. drm_fb_helper_fill_info() is provided
+ * as a helper to setup simple default values for the fbdev info structure.
  *
  * HANG DEBUGGING:
  *
@@ -3024,7 +3208,8 @@ static int drm_fbdev_fb_open(struct fb_info *info, int user)
 {
 	struct drm_fb_helper *fb_helper = info->par;
 
-	if (!try_module_get(fb_helper->dev->driver->fops->owner))
+	/* No need to take a ref for fbcon because it unbinds on unregister */
+	if (user && !try_module_get(fb_helper->dev->driver->fops->owner))
 		return -ENODEV;
 
 	return 0;
@@ -3034,7 +3219,8 @@ static int drm_fbdev_fb_release(struct fb_info *info, int user)
 {
 	struct drm_fb_helper *fb_helper = info->par;
 
-	module_put(fb_helper->dev->driver->fops->owner);
+	if (user)
+		module_put(fb_helper->dev->driver->fops->owner);
 
 	return 0;
 }
@@ -3149,7 +3335,6 @@ int drm_fb_helper_generic_probe(struct drm_fb_helper *fb_helper,
 	if (IS_ERR(fbi))
 		return PTR_ERR(fbi);
 
-	fbi->par = fb_helper;
 	fbi->fbops = &drm_fbdev_fb_ops;
 	fbi->screen_size = fb->height * fb->pitches[0];
 	fbi->fix.smem_len = fbi->screen_size;
@@ -3160,10 +3345,7 @@ int drm_fb_helper_generic_probe(struct drm_fb_helper *fb_helper,
 		fbi->fix.smem_start =
 			page_to_phys(virt_to_page(fbi->screen_buffer));
 #endif
-	strcpy(fbi->fix.id, "DRM emulated");
-
-	drm_fb_helper_fill_fix(fbi, fb->pitches[0], fb->format->depth);
-	drm_fb_helper_fill_var(fbi, fb_helper, sizes->fb_width, sizes->fb_height);
+	drm_fb_helper_fill_info(fbi, fb_helper, sizes);
 
 	if (fb->funcs->dirty) {
 		struct fb_ops *fbops;
@@ -3317,8 +3499,6 @@ int drm_fbdev_generic_setup(struct drm_device *dev, unsigned int preferred_bpp)
 		return ret;
 	}
 
-	drm_client_add(&fb_helper->client);
-
 	if (!preferred_bpp)
 		preferred_bpp = dev->mode_config.preferred_depth;
 	if (!preferred_bpp)
@@ -3329,6 +3509,8 @@ int drm_fbdev_generic_setup(struct drm_device *dev, unsigned int preferred_bpp)
 	if (ret)
 		DRM_DEV_DEBUG(dev->dev, "client hotplug ret=%d\n", ret);
 
+	drm_client_register(&fb_helper->client);
+
 	return 0;
 }
 EXPORT_SYMBOL(drm_fbdev_generic_setup);
diff --git a/drivers/gpu/drm/drm_file.c b/drivers/gpu/drm/drm_file.c
index 7caa3c7ed978..233f114d2186 100644
--- a/drivers/gpu/drm/drm_file.c
+++ b/drivers/gpu/drm/drm_file.c
@@ -128,7 +128,6 @@ struct drm_file *drm_file_alloc(struct drm_minor *minor)
 
 	/* for compatibility root is always authenticated */
 	file->authenticated = capable(CAP_SYS_ADMIN);
-	file->lock_count = 0;
 
 	INIT_LIST_HEAD(&file->lhead);
 	INIT_LIST_HEAD(&file->fbs);
@@ -425,30 +424,6 @@ static int drm_open_helper(struct file *filp, struct drm_minor *minor)
 	return 0;
 }
 
-static void drm_legacy_dev_reinit(struct drm_device *dev)
-{
-	if (dev->irq_enabled)
-		drm_irq_uninstall(dev);
-
-	mutex_lock(&dev->struct_mutex);
-
-	drm_legacy_agp_clear(dev);
-
-	drm_legacy_sg_cleanup(dev);
-	drm_legacy_vma_flush(dev);
-	drm_legacy_dma_takedown(dev);
-
-	mutex_unlock(&dev->struct_mutex);
-
-	dev->sigdata.lock = NULL;
-
-	dev->context_flag = 0;
-	dev->last_context = 0;
-	dev->if_version = 0;
-
-	DRM_DEBUG("lastclose completed\n");
-}
-
 void drm_lastclose(struct drm_device * dev)
 {
 	DRM_DEBUG("\n");
@@ -577,6 +552,7 @@ put_back_event:
 				file_priv->event_space -= length;
 				list_add(&e->link, &file_priv->event_list);
 				spin_unlock_irq(&dev->event_lock);
+				wake_up_interruptible(&file_priv->event_wait);
 				break;
 			}
 
diff --git a/drivers/gpu/drm/drm_format_helper.c b/drivers/gpu/drm/drm_format_helper.c
new file mode 100644
index 000000000000..a18da35145b7
--- /dev/null
+++ b/drivers/gpu/drm/drm_format_helper.c
@@ -0,0 +1,324 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2016 Noralf Trønnes
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/io.h>
+
+#include <drm/drm_format_helper.h>
+#include <drm/drm_framebuffer.h>
+#include <drm/drm_fourcc.h>
+#include <drm/drm_rect.h>
+
+static unsigned int clip_offset(struct drm_rect *clip,
+				unsigned int pitch, unsigned int cpp)
+{
+	return clip->y1 * pitch + clip->x1 * cpp;
+}
+
+/**
+ * drm_fb_memcpy - Copy clip buffer
+ * @dst: Destination buffer
+ * @vaddr: Source buffer
+ * @fb: DRM framebuffer
+ * @clip: Clip rectangle area to copy
+ *
+ * This function does not apply clipping on dst, i.e. the destination
+ * is a small buffer containing the clip rect only.
+ */
+void drm_fb_memcpy(void *dst, void *vaddr, struct drm_framebuffer *fb,
+		   struct drm_rect *clip)
+{
+	unsigned int cpp = drm_format_plane_cpp(fb->format->format, 0);
+	size_t len = (clip->x2 - clip->x1) * cpp;
+	unsigned int y, lines = clip->y2 - clip->y1;
+
+	vaddr += clip_offset(clip, fb->pitches[0], cpp);
+	for (y = 0; y < lines; y++) {
+		memcpy(dst, vaddr, len);
+		vaddr += fb->pitches[0];
+		dst += len;
+	}
+}
+EXPORT_SYMBOL(drm_fb_memcpy);
+
+/**
+ * drm_fb_memcpy_dstclip - Copy clip buffer
+ * @dst: Destination buffer (iomem)
+ * @vaddr: Source buffer
+ * @fb: DRM framebuffer
+ * @clip: Clip rectangle area to copy
+ *
+ * This function applies clipping on dst, i.e. the destination is a
+ * full (iomem) framebuffer but only the clip rect content is copied over.
+ */
+void drm_fb_memcpy_dstclip(void __iomem *dst, void *vaddr,
+			   struct drm_framebuffer *fb,
+			   struct drm_rect *clip)
+{
+	unsigned int cpp = drm_format_plane_cpp(fb->format->format, 0);
+	unsigned int offset = clip_offset(clip, fb->pitches[0], cpp);
+	size_t len = (clip->x2 - clip->x1) * cpp;
+	unsigned int y, lines = clip->y2 - clip->y1;
+
+	vaddr += offset;
+	dst += offset;
+	for (y = 0; y < lines; y++) {
+		memcpy_toio(dst, vaddr, len);
+		vaddr += fb->pitches[0];
+		dst += fb->pitches[0];
+	}
+}
+EXPORT_SYMBOL(drm_fb_memcpy_dstclip);
+
+/**
+ * drm_fb_swab16 - Swap bytes into clip buffer
+ * @dst: RGB565 destination buffer
+ * @vaddr: RGB565 source buffer
+ * @fb: DRM framebuffer
+ * @clip: Clip rectangle area to copy
+ */
+void drm_fb_swab16(u16 *dst, void *vaddr, struct drm_framebuffer *fb,
+		   struct drm_rect *clip)
+{
+	size_t len = (clip->x2 - clip->x1) * sizeof(u16);
+	unsigned int x, y;
+	u16 *src, *buf;
+
+	/*
+	 * The cma memory is write-combined so reads are uncached.
+	 * Speed up by fetching one line at a time.
+	 */
+	buf = kmalloc(len, GFP_KERNEL);
+	if (!buf)
+		return;
+
+	for (y = clip->y1; y < clip->y2; y++) {
+		src = vaddr + (y * fb->pitches[0]);
+		src += clip->x1;
+		memcpy(buf, src, len);
+		src = buf;
+		for (x = clip->x1; x < clip->x2; x++)
+			*dst++ = swab16(*src++);
+	}
+
+	kfree(buf);
+}
+EXPORT_SYMBOL(drm_fb_swab16);
+
+static void drm_fb_xrgb8888_to_rgb565_line(u16 *dbuf, u32 *sbuf,
+					   unsigned int pixels,
+					   bool swab)
+{
+	unsigned int x;
+	u16 val16;
+
+	for (x = 0; x < pixels; x++) {
+		val16 = ((sbuf[x] & 0x00F80000) >> 8) |
+			((sbuf[x] & 0x0000FC00) >> 5) |
+			((sbuf[x] & 0x000000F8) >> 3);
+		if (swab)
+			dbuf[x] = swab16(val16);
+		else
+			dbuf[x] = val16;
+	}
+}
+
+/**
+ * drm_fb_xrgb8888_to_rgb565 - Convert XRGB8888 to RGB565 clip buffer
+ * @dst: RGB565 destination buffer
+ * @vaddr: XRGB8888 source buffer
+ * @fb: DRM framebuffer
+ * @clip: Clip rectangle area to copy
+ * @swab: Swap bytes
+ *
+ * Drivers can use this function for RGB565 devices that don't natively
+ * support XRGB8888.
+ *
+ * This function does not apply clipping on dst, i.e. the destination
+ * is a small buffer containing the clip rect only.
+ */
+void drm_fb_xrgb8888_to_rgb565(void *dst, void *vaddr,
+			       struct drm_framebuffer *fb,
+			       struct drm_rect *clip, bool swab)
+{
+	size_t linepixels = clip->x2 - clip->x1;
+	size_t src_len = linepixels * sizeof(u32);
+	size_t dst_len = linepixels * sizeof(u16);
+	unsigned y, lines = clip->y2 - clip->y1;
+	void *sbuf;
+
+	/*
+	 * The cma memory is write-combined so reads are uncached.
+	 * Speed up by fetching one line at a time.
+	 */
+	sbuf = kmalloc(src_len, GFP_KERNEL);
+	if (!sbuf)
+		return;
+
+	vaddr += clip_offset(clip, fb->pitches[0], sizeof(u32));
+	for (y = 0; y < lines; y++) {
+		memcpy(sbuf, vaddr, src_len);
+		drm_fb_xrgb8888_to_rgb565_line(dst, sbuf, linepixels, swab);
+		vaddr += fb->pitches[0];
+		dst += dst_len;
+	}
+
+	kfree(sbuf);
+}
+EXPORT_SYMBOL(drm_fb_xrgb8888_to_rgb565);
+
+/**
+ * drm_fb_xrgb8888_to_rgb565_dstclip - Convert XRGB8888 to RGB565 clip buffer
+ * @dst: RGB565 destination buffer (iomem)
+ * @dst_pitch: destination buffer pitch
+ * @vaddr: XRGB8888 source buffer
+ * @fb: DRM framebuffer
+ * @clip: Clip rectangle area to copy
+ * @swab: Swap bytes
+ *
+ * Drivers can use this function for RGB565 devices that don't natively
+ * support XRGB8888.
+ *
+ * This function applies clipping on dst, i.e. the destination is a
+ * full (iomem) framebuffer but only the clip rect content is copied over.
+ */
+void drm_fb_xrgb8888_to_rgb565_dstclip(void __iomem *dst, unsigned int dst_pitch,
+				       void *vaddr, struct drm_framebuffer *fb,
+				       struct drm_rect *clip, bool swab)
+{
+	size_t linepixels = clip->x2 - clip->x1;
+	size_t dst_len = linepixels * sizeof(u16);
+	unsigned y, lines = clip->y2 - clip->y1;
+	void *dbuf;
+
+	dbuf = kmalloc(dst_len, GFP_KERNEL);
+	if (!dbuf)
+		return;
+
+	vaddr += clip_offset(clip, fb->pitches[0], sizeof(u32));
+	dst += clip_offset(clip, dst_pitch, sizeof(u16));
+	for (y = 0; y < lines; y++) {
+		drm_fb_xrgb8888_to_rgb565_line(dbuf, vaddr, linepixels, swab);
+		memcpy_toio(dst, dbuf, dst_len);
+		vaddr += fb->pitches[0];
+		dst += dst_len;
+	}
+
+	kfree(dbuf);
+}
+EXPORT_SYMBOL(drm_fb_xrgb8888_to_rgb565_dstclip);
+
+static void drm_fb_xrgb8888_to_rgb888_line(u8 *dbuf, u32 *sbuf,
+					   unsigned int pixels)
+{
+	unsigned int x;
+
+	for (x = 0; x < pixels; x++) {
+		*dbuf++ = (sbuf[x] & 0x000000FF) >>  0;
+		*dbuf++ = (sbuf[x] & 0x0000FF00) >>  8;
+		*dbuf++ = (sbuf[x] & 0x00FF0000) >> 16;
+	}
+}
+
+/**
+ * drm_fb_xrgb8888_to_rgb888_dstclip - Convert XRGB8888 to RGB888 clip buffer
+ * @dst: RGB565 destination buffer (iomem)
+ * @dst_pitch: destination buffer pitch
+ * @vaddr: XRGB8888 source buffer
+ * @fb: DRM framebuffer
+ * @clip: Clip rectangle area to copy
+ *
+ * Drivers can use this function for RGB888 devices that don't natively
+ * support XRGB8888.
+ *
+ * This function applies clipping on dst, i.e. the destination is a
+ * full (iomem) framebuffer but only the clip rect content is copied over.
+ */
+void drm_fb_xrgb8888_to_rgb888_dstclip(void __iomem *dst, unsigned int dst_pitch,
+				       void *vaddr, struct drm_framebuffer *fb,
+				       struct drm_rect *clip)
+{
+	size_t linepixels = clip->x2 - clip->x1;
+	size_t dst_len = linepixels * 3;
+	unsigned y, lines = clip->y2 - clip->y1;
+	void *dbuf;
+
+	dbuf = kmalloc(dst_len, GFP_KERNEL);
+	if (!dbuf)
+		return;
+
+	vaddr += clip_offset(clip, fb->pitches[0], sizeof(u32));
+	dst += clip_offset(clip, dst_pitch, sizeof(u16));
+	for (y = 0; y < lines; y++) {
+		drm_fb_xrgb8888_to_rgb888_line(dbuf, vaddr, linepixels);
+		memcpy_toio(dst, dbuf, dst_len);
+		vaddr += fb->pitches[0];
+		dst += dst_len;
+	}
+
+	kfree(dbuf);
+}
+EXPORT_SYMBOL(drm_fb_xrgb8888_to_rgb888_dstclip);
+
+/**
+ * drm_fb_xrgb8888_to_gray8 - Convert XRGB8888 to grayscale
+ * @dst: 8-bit grayscale destination buffer
+ * @vaddr: XRGB8888 source buffer
+ * @fb: DRM framebuffer
+ * @clip: Clip rectangle area to copy
+ *
+ * Drm doesn't have native monochrome or grayscale support.
+ * Such drivers can announce the commonly supported XR24 format to userspace
+ * and use this function to convert to the native format.
+ *
+ * Monochrome drivers will use the most significant bit,
+ * where 1 means foreground color and 0 background color.
+ *
+ * ITU BT.601 is used for the RGB -> luma (brightness) conversion.
+ */
+void drm_fb_xrgb8888_to_gray8(u8 *dst, void *vaddr, struct drm_framebuffer *fb,
+			       struct drm_rect *clip)
+{
+	unsigned int len = (clip->x2 - clip->x1) * sizeof(u32);
+	unsigned int x, y;
+	void *buf;
+	u32 *src;
+
+	if (WARN_ON(fb->format->format != DRM_FORMAT_XRGB8888))
+		return;
+	/*
+	 * The cma memory is write-combined so reads are uncached.
+	 * Speed up by fetching one line at a time.
+	 */
+	buf = kmalloc(len, GFP_KERNEL);
+	if (!buf)
+		return;
+
+	for (y = clip->y1; y < clip->y2; y++) {
+		src = vaddr + (y * fb->pitches[0]);
+		src += clip->x1;
+		memcpy(buf, src, len);
+		src = buf;
+		for (x = clip->x1; x < clip->x2; x++) {
+			u8 r = (*src & 0x00ff0000) >> 16;
+			u8 g = (*src & 0x0000ff00) >> 8;
+			u8 b =  *src & 0x000000ff;
+
+			/* ITU BT.601: Y = 0.299 R + 0.587 G + 0.114 B */
+			*dst++ = (3 * r + 6 * g + b) / 10;
+			src++;
+		}
+	}
+
+	kfree(buf);
+}
+EXPORT_SYMBOL(drm_fb_xrgb8888_to_gray8);
+
diff --git a/drivers/gpu/drm/drm_fourcc.c b/drivers/gpu/drm/drm_fourcc.c
index ba7e19d4336c..6ea55fb4526d 100644
--- a/drivers/gpu/drm/drm_fourcc.c
+++ b/drivers/gpu/drm/drm_fourcc.c
@@ -198,6 +198,10 @@ const struct drm_format_info *__drm_format_info(u32 format)
 		{ .format = DRM_FORMAT_ABGR8888,	.depth = 32, .num_planes = 1, .cpp = { 4, 0, 0 }, .hsub = 1, .vsub = 1, .has_alpha = true },
 		{ .format = DRM_FORMAT_RGBA8888,	.depth = 32, .num_planes = 1, .cpp = { 4, 0, 0 }, .hsub = 1, .vsub = 1, .has_alpha = true },
 		{ .format = DRM_FORMAT_BGRA8888,	.depth = 32, .num_planes = 1, .cpp = { 4, 0, 0 }, .hsub = 1, .vsub = 1, .has_alpha = true },
+		{ .format = DRM_FORMAT_XRGB16161616F,	.depth = 0,  .num_planes = 1, .cpp = { 8, 0, 0 }, .hsub = 1, .vsub = 1 },
+		{ .format = DRM_FORMAT_XBGR16161616F,	.depth = 0,  .num_planes = 1, .cpp = { 8, 0, 0 }, .hsub = 1, .vsub = 1 },
+		{ .format = DRM_FORMAT_ARGB16161616F,	.depth = 0,  .num_planes = 1, .cpp = { 8, 0, 0 }, .hsub = 1, .vsub = 1, .has_alpha = true },
+		{ .format = DRM_FORMAT_ABGR16161616F,	.depth = 0,  .num_planes = 1, .cpp = { 8, 0, 0 }, .hsub = 1, .vsub = 1, .has_alpha = true },
 		{ .format = DRM_FORMAT_RGB888_A8,	.depth = 32, .num_planes = 2, .cpp = { 3, 1, 0 }, .hsub = 1, .vsub = 1, .has_alpha = true },
 		{ .format = DRM_FORMAT_BGR888_A8,	.depth = 32, .num_planes = 2, .cpp = { 3, 1, 0 }, .hsub = 1, .vsub = 1, .has_alpha = true },
 		{ .format = DRM_FORMAT_XRGB8888_A8,	.depth = 32, .num_planes = 2, .cpp = { 4, 1, 0 }, .hsub = 1, .vsub = 1, .has_alpha = true },
@@ -225,7 +229,17 @@ const struct drm_format_info *__drm_format_info(u32 format)
 		{ .format = DRM_FORMAT_UYVY,		.depth = 0,  .num_planes = 1, .cpp = { 2, 0, 0 }, .hsub = 2, .vsub = 1, .is_yuv = true },
 		{ .format = DRM_FORMAT_VYUY,		.depth = 0,  .num_planes = 1, .cpp = { 2, 0, 0 }, .hsub = 2, .vsub = 1, .is_yuv = true },
 		{ .format = DRM_FORMAT_XYUV8888,	.depth = 0,  .num_planes = 1, .cpp = { 4, 0, 0 }, .hsub = 1, .vsub = 1, .is_yuv = true },
+		{ .format = DRM_FORMAT_VUY888,          .depth = 0,  .num_planes = 1, .cpp = { 3, 0, 0 }, .hsub = 1, .vsub = 1, .is_yuv = true },
 		{ .format = DRM_FORMAT_AYUV,		.depth = 0,  .num_planes = 1, .cpp = { 4, 0, 0 }, .hsub = 1, .vsub = 1, .has_alpha = true, .is_yuv = true },
+		{ .format = DRM_FORMAT_Y210,            .depth = 0,  .num_planes = 1, .cpp = { 4, 0, 0 }, .hsub = 2, .vsub = 1, .is_yuv = true },
+		{ .format = DRM_FORMAT_Y212,            .depth = 0,  .num_planes = 1, .cpp = { 4, 0, 0 }, .hsub = 2, .vsub = 1, .is_yuv = true },
+		{ .format = DRM_FORMAT_Y216,            .depth = 0,  .num_planes = 1, .cpp = { 4, 0, 0 }, .hsub = 2, .vsub = 1, .is_yuv = true },
+		{ .format = DRM_FORMAT_Y410,            .depth = 0,  .num_planes = 1, .cpp = { 4, 0, 0 }, .hsub = 1, .vsub = 1, .has_alpha = true, .is_yuv = true },
+		{ .format = DRM_FORMAT_Y412,            .depth = 0,  .num_planes = 1, .cpp = { 8, 0, 0 }, .hsub = 1, .vsub = 1, .has_alpha = true, .is_yuv = true },
+		{ .format = DRM_FORMAT_Y416,            .depth = 0,  .num_planes = 1, .cpp = { 8, 0, 0 }, .hsub = 1, .vsub = 1, .has_alpha = true, .is_yuv = true },
+		{ .format = DRM_FORMAT_XVYU2101010,	.depth = 0,  .num_planes = 1, .cpp = { 4, 0, 0 }, .hsub = 1, .vsub = 1, .is_yuv = true },
+		{ .format = DRM_FORMAT_XVYU12_16161616,	.depth = 0,  .num_planes = 1, .cpp = { 8, 0, 0 }, .hsub = 1, .vsub = 1, .is_yuv = true },
+		{ .format = DRM_FORMAT_XVYU16161616,	.depth = 0,  .num_planes = 1, .cpp = { 8, 0, 0 }, .hsub = 1, .vsub = 1, .is_yuv = true },
 		{ .format = DRM_FORMAT_Y0L0,		.depth = 0,  .num_planes = 1,
 		  .char_per_block = { 8, 0, 0 }, .block_w = { 2, 0, 0 }, .block_h = { 2, 0, 0 },
 		  .hsub = 2, .vsub = 2, .has_alpha = true, .is_yuv = true },
@@ -247,6 +261,19 @@ const struct drm_format_info *__drm_format_info(u32 format)
 		{ .format = DRM_FORMAT_P016,		.depth = 0,  .num_planes = 2,
 		  .char_per_block = { 2, 4, 0 }, .block_w = { 1, 0, 0 }, .block_h = { 1, 0, 0 },
 		  .hsub = 2, .vsub = 2, .is_yuv = true},
+		{ .format = DRM_FORMAT_P210,		.depth = 0,
+		  .num_planes = 2, .char_per_block = { 2, 4, 0 },
+		  .block_w = { 1, 0, 0 }, .block_h = { 1, 0, 0 }, .hsub = 2,
+		  .vsub = 1, .is_yuv = true },
+		{ .format = DRM_FORMAT_VUY101010,	.depth = 0,
+		  .num_planes = 1, .cpp = { 0, 0, 0 }, .hsub = 1, .vsub = 1,
+		  .is_yuv = true },
+		{ .format = DRM_FORMAT_YUV420_8BIT,     .depth = 0,
+		  .num_planes = 1, .cpp = { 0, 0, 0 }, .hsub = 2, .vsub = 2,
+		  .is_yuv = true },
+		{ .format = DRM_FORMAT_YUV420_10BIT,    .depth = 0,
+		  .num_planes = 1, .cpp = { 0, 0, 0 }, .hsub = 2, .vsub = 2,
+		  .is_yuv = true },
 	};
 
 	unsigned int i;
diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index d0b9f6a9953f..50de138c89e0 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -72,23 +72,6 @@
  * up at a later date, and as our interface with shmfs for memory allocation.
  */
 
-/*
- * We make up offsets for buffer objects so we can recognize them at
- * mmap time.
- */
-
-/* pgoff in mmap is an unsigned long, so we need to make sure that
- * the faked up offset will fit
- */
-
-#if BITS_PER_LONG == 64
-#define DRM_FILE_PAGE_OFFSET_START ((0xFFFFFFFFUL >> PAGE_SHIFT) + 1)
-#define DRM_FILE_PAGE_OFFSET_SIZE ((0xFFFFFFFFUL >> PAGE_SHIFT) * 16)
-#else
-#define DRM_FILE_PAGE_OFFSET_START ((0xFFFFFFFUL >> PAGE_SHIFT) + 1)
-#define DRM_FILE_PAGE_OFFSET_SIZE ((0xFFFFFFFUL >> PAGE_SHIFT) * 16)
-#endif
-
 /**
  * drm_gem_init - Initialize the GEM device fields
  * @dev: drm_devic structure to initialize
@@ -171,6 +154,10 @@ void drm_gem_private_object_init(struct drm_device *dev,
 	kref_init(&obj->refcount);
 	obj->handle_count = 0;
 	obj->size = size;
+	reservation_object_init(&obj->_resv);
+	if (!obj->resv)
+		obj->resv = &obj->_resv;
+
 	drm_vma_node_reset(&obj->vma_node);
 }
 EXPORT_SYMBOL(drm_gem_private_object_init);
@@ -659,6 +646,85 @@ void drm_gem_put_pages(struct drm_gem_object *obj, struct page **pages,
 }
 EXPORT_SYMBOL(drm_gem_put_pages);
 
+static int objects_lookup(struct drm_file *filp, u32 *handle, int count,
+			  struct drm_gem_object **objs)
+{
+	int i, ret = 0;
+	struct drm_gem_object *obj;
+
+	spin_lock(&filp->table_lock);
+
+	for (i = 0; i < count; i++) {
+		/* Check if we currently have a reference on the object */
+		obj = idr_find(&filp->object_idr, handle[i]);
+		if (!obj) {
+			ret = -ENOENT;
+			break;
+		}
+		drm_gem_object_get(obj);
+		objs[i] = obj;
+	}
+	spin_unlock(&filp->table_lock);
+
+	return ret;
+}
+
+/**
+ * drm_gem_objects_lookup - look up GEM objects from an array of handles
+ * @filp: DRM file private date
+ * @bo_handles: user pointer to array of userspace handle
+ * @count: size of handle array
+ * @objs_out: returned pointer to array of drm_gem_object pointers
+ *
+ * Takes an array of userspace handles and returns a newly allocated array of
+ * GEM objects.
+ *
+ * For a single handle lookup, use drm_gem_object_lookup().
+ *
+ * Returns:
+ *
+ * @objs filled in with GEM object pointers. Returned GEM objects need to be
+ * released with drm_gem_object_put(). -ENOENT is returned on a lookup
+ * failure. 0 is returned on success.
+ *
+ */
+int drm_gem_objects_lookup(struct drm_file *filp, void __user *bo_handles,
+			   int count, struct drm_gem_object ***objs_out)
+{
+	int ret;
+	u32 *handles;
+	struct drm_gem_object **objs;
+
+	if (!count)
+		return 0;
+
+	objs = kvmalloc_array(count, sizeof(struct drm_gem_object *),
+			     GFP_KERNEL | __GFP_ZERO);
+	if (!objs)
+		return -ENOMEM;
+
+	handles = kvmalloc_array(count, sizeof(u32), GFP_KERNEL);
+	if (!handles) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	if (copy_from_user(handles, bo_handles, count * sizeof(u32))) {
+		ret = -EFAULT;
+		DRM_DEBUG("Failed to copy in GEM handles\n");
+		goto out;
+	}
+
+	ret = objects_lookup(filp, handles, count, objs);
+	*objs_out = objs;
+
+out:
+	kvfree(handles);
+	return ret;
+
+}
+EXPORT_SYMBOL(drm_gem_objects_lookup);
+
 /**
  * drm_gem_object_lookup - look up a GEM object from its handle
  * @filp: DRM file private date
@@ -668,24 +734,56 @@ EXPORT_SYMBOL(drm_gem_put_pages);
  *
  * A reference to the object named by the handle if such exists on @filp, NULL
  * otherwise.
+ *
+ * If looking up an array of handles, use drm_gem_objects_lookup().
  */
 struct drm_gem_object *
 drm_gem_object_lookup(struct drm_file *filp, u32 handle)
 {
+	struct drm_gem_object *obj = NULL;
+
+	objects_lookup(filp, &handle, 1, &obj);
+	return obj;
+}
+EXPORT_SYMBOL(drm_gem_object_lookup);
+
+/**
+ * drm_gem_reservation_object_wait - Wait on GEM object's reservation's objects
+ * shared and/or exclusive fences.
+ * @filep: DRM file private date
+ * @handle: userspace handle
+ * @wait_all: if true, wait on all fences, else wait on just exclusive fence
+ * @timeout: timeout value in jiffies or zero to return immediately
+ *
+ * Returns:
+ *
+ * Returns -ERESTARTSYS if interrupted, 0 if the wait timed out, or
+ * greater than 0 on success.
+ */
+long drm_gem_reservation_object_wait(struct drm_file *filep, u32 handle,
+				    bool wait_all, unsigned long timeout)
+{
+	long ret;
 	struct drm_gem_object *obj;
 
-	spin_lock(&filp->table_lock);
+	obj = drm_gem_object_lookup(filep, handle);
+	if (!obj) {
+		DRM_DEBUG("Failed to look up GEM BO %d\n", handle);
+		return -EINVAL;
+	}
 
-	/* Check if we currently have a reference on the object */
-	obj = idr_find(&filp->object_idr, handle);
-	if (obj)
-		drm_gem_object_get(obj);
+	ret = reservation_object_wait_timeout_rcu(obj->resv, wait_all,
+						  true, timeout);
+	if (ret == 0)
+		ret = -ETIME;
+	else if (ret > 0)
+		ret = 0;
 
-	spin_unlock(&filp->table_lock);
+	drm_gem_object_put_unlocked(obj);
 
-	return obj;
+	return ret;
 }
-EXPORT_SYMBOL(drm_gem_object_lookup);
+EXPORT_SYMBOL(drm_gem_reservation_object_wait);
 
 /**
  * drm_gem_close_ioctl - implementation of the GEM_CLOSE ioctl
@@ -851,6 +949,7 @@ drm_gem_object_release(struct drm_gem_object *obj)
 	if (obj->filp)
 		fput(obj->filp);
 
+	reservation_object_fini(&obj->_resv);
 	drm_gem_free_mmap_offset(obj);
 }
 EXPORT_SYMBOL(drm_gem_object_release);
@@ -1190,3 +1289,174 @@ void drm_gem_vunmap(struct drm_gem_object *obj, void *vaddr)
 		obj->dev->driver->gem_prime_vunmap(obj, vaddr);
 }
 EXPORT_SYMBOL(drm_gem_vunmap);
+
+/**
+ * drm_gem_lock_reservations - Sets up the ww context and acquires
+ * the lock on an array of GEM objects.
+ *
+ * Once you've locked your reservations, you'll want to set up space
+ * for your shared fences (if applicable), submit your job, then
+ * drm_gem_unlock_reservations().
+ *
+ * @objs: drm_gem_objects to lock
+ * @count: Number of objects in @objs
+ * @acquire_ctx: struct ww_acquire_ctx that will be initialized as
+ * part of tracking this set of locked reservations.
+ */
+int
+drm_gem_lock_reservations(struct drm_gem_object **objs, int count,
+			  struct ww_acquire_ctx *acquire_ctx)
+{
+	int contended = -1;
+	int i, ret;
+
+	ww_acquire_init(acquire_ctx, &reservation_ww_class);
+
+retry:
+	if (contended != -1) {
+		struct drm_gem_object *obj = objs[contended];
+
+		ret = ww_mutex_lock_slow_interruptible(&obj->resv->lock,
+						       acquire_ctx);
+		if (ret) {
+			ww_acquire_done(acquire_ctx);
+			return ret;
+		}
+	}
+
+	for (i = 0; i < count; i++) {
+		if (i == contended)
+			continue;
+
+		ret = ww_mutex_lock_interruptible(&objs[i]->resv->lock,
+						  acquire_ctx);
+		if (ret) {
+			int j;
+
+			for (j = 0; j < i; j++)
+				ww_mutex_unlock(&objs[j]->resv->lock);
+
+			if (contended != -1 && contended >= i)
+				ww_mutex_unlock(&objs[contended]->resv->lock);
+
+			if (ret == -EDEADLK) {
+				contended = i;
+				goto retry;
+			}
+
+			ww_acquire_done(acquire_ctx);
+			return ret;
+		}
+	}
+
+	ww_acquire_done(acquire_ctx);
+
+	return 0;
+}
+EXPORT_SYMBOL(drm_gem_lock_reservations);
+
+void
+drm_gem_unlock_reservations(struct drm_gem_object **objs, int count,
+			    struct ww_acquire_ctx *acquire_ctx)
+{
+	int i;
+
+	for (i = 0; i < count; i++)
+		ww_mutex_unlock(&objs[i]->resv->lock);
+
+	ww_acquire_fini(acquire_ctx);
+}
+EXPORT_SYMBOL(drm_gem_unlock_reservations);
+
+/**
+ * drm_gem_fence_array_add - Adds the fence to an array of fences to be
+ * waited on, deduplicating fences from the same context.
+ *
+ * @fence_array: array of dma_fence * for the job to block on.
+ * @fence: the dma_fence to add to the list of dependencies.
+ *
+ * Returns:
+ * 0 on success, or an error on failing to expand the array.
+ */
+int drm_gem_fence_array_add(struct xarray *fence_array,
+			    struct dma_fence *fence)
+{
+	struct dma_fence *entry;
+	unsigned long index;
+	u32 id = 0;
+	int ret;
+
+	if (!fence)
+		return 0;
+
+	/* Deduplicate if we already depend on a fence from the same context.
+	 * This lets the size of the array of deps scale with the number of
+	 * engines involved, rather than the number of BOs.
+	 */
+	xa_for_each(fence_array, index, entry) {
+		if (entry->context != fence->context)
+			continue;
+
+		if (dma_fence_is_later(fence, entry)) {
+			dma_fence_put(entry);
+			xa_store(fence_array, index, fence, GFP_KERNEL);
+		} else {
+			dma_fence_put(fence);
+		}
+		return 0;
+	}
+
+	ret = xa_alloc(fence_array, &id, fence, xa_limit_32b, GFP_KERNEL);
+	if (ret != 0)
+		dma_fence_put(fence);
+
+	return ret;
+}
+EXPORT_SYMBOL(drm_gem_fence_array_add);
+
+/**
+ * drm_gem_fence_array_add_implicit - Adds the implicit dependencies tracked
+ * in the GEM object's reservation object to an array of dma_fences for use in
+ * scheduling a rendering job.
+ *
+ * This should be called after drm_gem_lock_reservations() on your array of
+ * GEM objects used in the job but before updating the reservations with your
+ * own fences.
+ *
+ * @fence_array: array of dma_fence * for the job to block on.
+ * @obj: the gem object to add new dependencies from.
+ * @write: whether the job might write the object (so we need to depend on
+ * shared fences in the reservation object).
+ */
+int drm_gem_fence_array_add_implicit(struct xarray *fence_array,
+				     struct drm_gem_object *obj,
+				     bool write)
+{
+	int ret;
+	struct dma_fence **fences;
+	unsigned int i, fence_count;
+
+	if (!write) {
+		struct dma_fence *fence =
+			reservation_object_get_excl_rcu(obj->resv);
+
+		return drm_gem_fence_array_add(fence_array, fence);
+	}
+
+	ret = reservation_object_get_fences_rcu(obj->resv, NULL,
+						&fence_count, &fences);
+	if (ret || !fence_count)
+		return ret;
+
+	for (i = 0; i < fence_count; i++) {
+		ret = drm_gem_fence_array_add(fence_array, fences[i]);
+		if (ret)
+			break;
+	}
+
+	for (; i < fence_count; i++)
+		dma_fence_put(fences[i]);
+	kfree(fences);
+	return ret;
+}
+EXPORT_SYMBOL(drm_gem_fence_array_add_implicit);
diff --git a/drivers/gpu/drm/drm_gem_cma_helper.c b/drivers/gpu/drm/drm_gem_cma_helper.c
index cc26625b4b33..e01ceed09e67 100644
--- a/drivers/gpu/drm/drm_gem_cma_helper.c
+++ b/drivers/gpu/drm/drm_gem_cma_helper.c
@@ -186,13 +186,13 @@ void drm_gem_cma_free_object(struct drm_gem_object *gem_obj)
 
 	cma_obj = to_drm_gem_cma_obj(gem_obj);
 
-	if (cma_obj->vaddr) {
-		dma_free_wc(gem_obj->dev->dev, cma_obj->base.size,
-			    cma_obj->vaddr, cma_obj->paddr);
-	} else if (gem_obj->import_attach) {
+	if (gem_obj->import_attach) {
 		if (cma_obj->vaddr)
 			dma_buf_vunmap(gem_obj->import_attach->dmabuf, cma_obj->vaddr);
 		drm_prime_gem_destroy(gem_obj, cma_obj->sgt);
+	} else if (cma_obj->vaddr) {
+		dma_free_wc(gem_obj->dev->dev, cma_obj->base.size,
+			    cma_obj->vaddr, cma_obj->paddr);
 	}
 
 	drm_gem_object_release(gem_obj);
diff --git a/drivers/gpu/drm/drm_gem_shmem_helper.c b/drivers/gpu/drm/drm_gem_shmem_helper.c
new file mode 100644
index 000000000000..1ee208c2c85e
--- /dev/null
+++ b/drivers/gpu/drm/drm_gem_shmem_helper.c
@@ -0,0 +1,625 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2018 Noralf Trønnes
+ */
+
+#include <linux/dma-buf.h>
+#include <linux/export.h>
+#include <linux/mutex.h>
+#include <linux/shmem_fs.h>
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+
+#include <drm/drm_device.h>
+#include <drm/drm_drv.h>
+#include <drm/drm_gem_shmem_helper.h>
+#include <drm/drm_prime.h>
+#include <drm/drm_print.h>
+
+/**
+ * DOC: overview
+ *
+ * This library provides helpers for GEM objects backed by shmem buffers
+ * allocated using anonymous pageable memory.
+ */
+
+static const struct drm_gem_object_funcs drm_gem_shmem_funcs = {
+	.free = drm_gem_shmem_free_object,
+	.print_info = drm_gem_shmem_print_info,
+	.pin = drm_gem_shmem_pin,
+	.unpin = drm_gem_shmem_unpin,
+	.get_sg_table = drm_gem_shmem_get_sg_table,
+	.vmap = drm_gem_shmem_vmap,
+	.vunmap = drm_gem_shmem_vunmap,
+	.vm_ops = &drm_gem_shmem_vm_ops,
+};
+
+/**
+ * drm_gem_shmem_create - Allocate an object with the given size
+ * @dev: DRM device
+ * @size: Size of the object to allocate
+ *
+ * This function creates a shmem GEM object.
+ *
+ * Returns:
+ * A struct drm_gem_shmem_object * on success or an ERR_PTR()-encoded negative
+ * error code on failure.
+ */
+struct drm_gem_shmem_object *drm_gem_shmem_create(struct drm_device *dev, size_t size)
+{
+	struct drm_gem_shmem_object *shmem;
+	struct drm_gem_object *obj;
+	int ret;
+
+	size = PAGE_ALIGN(size);
+
+	if (dev->driver->gem_create_object)
+		obj = dev->driver->gem_create_object(dev, size);
+	else
+		obj = kzalloc(sizeof(*shmem), GFP_KERNEL);
+	if (!obj)
+		return ERR_PTR(-ENOMEM);
+
+	if (!obj->funcs)
+		obj->funcs = &drm_gem_shmem_funcs;
+
+	ret = drm_gem_object_init(dev, obj, size);
+	if (ret)
+		goto err_free;
+
+	ret = drm_gem_create_mmap_offset(obj);
+	if (ret)
+		goto err_release;
+
+	shmem = to_drm_gem_shmem_obj(obj);
+	mutex_init(&shmem->pages_lock);
+	mutex_init(&shmem->vmap_lock);
+
+	/*
+	 * Our buffers are kept pinned, so allocating them
+	 * from the MOVABLE zone is a really bad idea, and
+	 * conflicts with CMA. See comments above new_inode()
+	 * why this is required _and_ expected if you're
+	 * going to pin these pages.
+	 */
+	mapping_set_gfp_mask(obj->filp->f_mapping, GFP_HIGHUSER |
+			     __GFP_RETRY_MAYFAIL | __GFP_NOWARN);
+
+	return shmem;
+
+err_release:
+	drm_gem_object_release(obj);
+err_free:
+	kfree(obj);
+
+	return ERR_PTR(ret);
+}
+EXPORT_SYMBOL_GPL(drm_gem_shmem_create);
+
+/**
+ * drm_gem_shmem_free_object - Free resources associated with a shmem GEM object
+ * @obj: GEM object to free
+ *
+ * This function cleans up the GEM object state and frees the memory used to
+ * store the object itself.
+ */
+void drm_gem_shmem_free_object(struct drm_gem_object *obj)
+{
+	struct drm_gem_shmem_object *shmem = to_drm_gem_shmem_obj(obj);
+
+	WARN_ON(shmem->vmap_use_count);
+
+	if (obj->import_attach) {
+		shmem->pages_use_count--;
+		drm_prime_gem_destroy(obj, shmem->sgt);
+		kvfree(shmem->pages);
+	} else {
+		if (shmem->sgt) {
+			dma_unmap_sg(obj->dev->dev, shmem->sgt->sgl,
+				     shmem->sgt->nents, DMA_BIDIRECTIONAL);
+
+			drm_gem_shmem_put_pages(shmem);
+			sg_free_table(shmem->sgt);
+			kfree(shmem->sgt);
+		}
+	}
+
+	WARN_ON(shmem->pages_use_count);
+
+	drm_gem_object_release(obj);
+	mutex_destroy(&shmem->pages_lock);
+	mutex_destroy(&shmem->vmap_lock);
+	kfree(shmem);
+}
+EXPORT_SYMBOL_GPL(drm_gem_shmem_free_object);
+
+static int drm_gem_shmem_get_pages_locked(struct drm_gem_shmem_object *shmem)
+{
+	struct drm_gem_object *obj = &shmem->base;
+	struct page **pages;
+
+	if (shmem->pages_use_count++ > 0)
+		return 0;
+
+	pages = drm_gem_get_pages(obj);
+	if (IS_ERR(pages)) {
+		DRM_DEBUG_KMS("Failed to get pages (%ld)\n", PTR_ERR(pages));
+		shmem->pages_use_count = 0;
+		return PTR_ERR(pages);
+	}
+
+	shmem->pages = pages;
+
+	return 0;
+}
+
+/*
+ * drm_gem_shmem_get_pages - Allocate backing pages for a shmem GEM object
+ * @shmem: shmem GEM object
+ *
+ * This function makes sure that backing pages exists for the shmem GEM object
+ * and increases the use count.
+ *
+ * Returns:
+ * 0 on success or a negative error code on failure.
+ */
+int drm_gem_shmem_get_pages(struct drm_gem_shmem_object *shmem)
+{
+	int ret;
+
+	ret = mutex_lock_interruptible(&shmem->pages_lock);
+	if (ret)
+		return ret;
+	ret = drm_gem_shmem_get_pages_locked(shmem);
+	mutex_unlock(&shmem->pages_lock);
+
+	return ret;
+}
+EXPORT_SYMBOL(drm_gem_shmem_get_pages);
+
+static void drm_gem_shmem_put_pages_locked(struct drm_gem_shmem_object *shmem)
+{
+	struct drm_gem_object *obj = &shmem->base;
+
+	if (WARN_ON_ONCE(!shmem->pages_use_count))
+		return;
+
+	if (--shmem->pages_use_count > 0)
+		return;
+
+	drm_gem_put_pages(obj, shmem->pages,
+			  shmem->pages_mark_dirty_on_put,
+			  shmem->pages_mark_accessed_on_put);
+	shmem->pages = NULL;
+}
+
+/*
+ * drm_gem_shmem_put_pages - Decrease use count on the backing pages for a shmem GEM object
+ * @shmem: shmem GEM object
+ *
+ * This function decreases the use count and puts the backing pages when use drops to zero.
+ */
+void drm_gem_shmem_put_pages(struct drm_gem_shmem_object *shmem)
+{
+	mutex_lock(&shmem->pages_lock);
+	drm_gem_shmem_put_pages_locked(shmem);
+	mutex_unlock(&shmem->pages_lock);
+}
+EXPORT_SYMBOL(drm_gem_shmem_put_pages);
+
+/**
+ * drm_gem_shmem_pin - Pin backing pages for a shmem GEM object
+ * @obj: GEM object
+ *
+ * This function makes sure the backing pages are pinned in memory while the
+ * buffer is exported.
+ *
+ * Returns:
+ * 0 on success or a negative error code on failure.
+ */
+int drm_gem_shmem_pin(struct drm_gem_object *obj)
+{
+	struct drm_gem_shmem_object *shmem = to_drm_gem_shmem_obj(obj);
+
+	return drm_gem_shmem_get_pages(shmem);
+}
+EXPORT_SYMBOL(drm_gem_shmem_pin);
+
+/**
+ * drm_gem_shmem_unpin - Unpin backing pages for a shmem GEM object
+ * @obj: GEM object
+ *
+ * This function removes the requirement that the backing pages are pinned in
+ * memory.
+ */
+void drm_gem_shmem_unpin(struct drm_gem_object *obj)
+{
+	struct drm_gem_shmem_object *shmem = to_drm_gem_shmem_obj(obj);
+
+	drm_gem_shmem_put_pages(shmem);
+}
+EXPORT_SYMBOL(drm_gem_shmem_unpin);
+
+static void *drm_gem_shmem_vmap_locked(struct drm_gem_shmem_object *shmem)
+{
+	struct drm_gem_object *obj = &shmem->base;
+	int ret;
+
+	if (shmem->vmap_use_count++ > 0)
+		return shmem->vaddr;
+
+	ret = drm_gem_shmem_get_pages(shmem);
+	if (ret)
+		goto err_zero_use;
+
+	if (obj->import_attach)
+		shmem->vaddr = dma_buf_vmap(obj->import_attach->dmabuf);
+	else
+		shmem->vaddr = vmap(shmem->pages, obj->size >> PAGE_SHIFT, VM_MAP, PAGE_KERNEL);
+
+	if (!shmem->vaddr) {
+		DRM_DEBUG_KMS("Failed to vmap pages\n");
+		ret = -ENOMEM;
+		goto err_put_pages;
+	}
+
+	return shmem->vaddr;
+
+err_put_pages:
+	drm_gem_shmem_put_pages(shmem);
+err_zero_use:
+	shmem->vmap_use_count = 0;
+
+	return ERR_PTR(ret);
+}
+
+/*
+ * drm_gem_shmem_vmap - Create a virtual mapping for a shmem GEM object
+ * @shmem: shmem GEM object
+ *
+ * This function makes sure that a virtual address exists for the buffer backing
+ * the shmem GEM object.
+ *
+ * Returns:
+ * 0 on success or a negative error code on failure.
+ */
+void *drm_gem_shmem_vmap(struct drm_gem_object *obj)
+{
+	struct drm_gem_shmem_object *shmem = to_drm_gem_shmem_obj(obj);
+	void *vaddr;
+	int ret;
+
+	ret = mutex_lock_interruptible(&shmem->vmap_lock);
+	if (ret)
+		return ERR_PTR(ret);
+	vaddr = drm_gem_shmem_vmap_locked(shmem);
+	mutex_unlock(&shmem->vmap_lock);
+
+	return vaddr;
+}
+EXPORT_SYMBOL(drm_gem_shmem_vmap);
+
+static void drm_gem_shmem_vunmap_locked(struct drm_gem_shmem_object *shmem)
+{
+	struct drm_gem_object *obj = &shmem->base;
+
+	if (WARN_ON_ONCE(!shmem->vmap_use_count))
+		return;
+
+	if (--shmem->vmap_use_count > 0)
+		return;
+
+	if (obj->import_attach)
+		dma_buf_vunmap(obj->import_attach->dmabuf, shmem->vaddr);
+	else
+		vunmap(shmem->vaddr);
+
+	shmem->vaddr = NULL;
+	drm_gem_shmem_put_pages(shmem);
+}
+
+/*
+ * drm_gem_shmem_vunmap - Unmap a virtual mapping fo a shmem GEM object
+ * @shmem: shmem GEM object
+ *
+ * This function removes the virtual address when use count drops to zero.
+ */
+void drm_gem_shmem_vunmap(struct drm_gem_object *obj, void *vaddr)
+{
+	struct drm_gem_shmem_object *shmem = to_drm_gem_shmem_obj(obj);
+
+	mutex_lock(&shmem->vmap_lock);
+	drm_gem_shmem_vunmap_locked(shmem);
+	mutex_unlock(&shmem->vmap_lock);
+}
+EXPORT_SYMBOL(drm_gem_shmem_vunmap);
+
+struct drm_gem_shmem_object *
+drm_gem_shmem_create_with_handle(struct drm_file *file_priv,
+				 struct drm_device *dev, size_t size,
+				 uint32_t *handle)
+{
+	struct drm_gem_shmem_object *shmem;
+	int ret;
+
+	shmem = drm_gem_shmem_create(dev, size);
+	if (IS_ERR(shmem))
+		return shmem;
+
+	/*
+	 * Allocate an id of idr table where the obj is registered
+	 * and handle has the id what user can see.
+	 */
+	ret = drm_gem_handle_create(file_priv, &shmem->base, handle);
+	/* drop reference from allocate - handle holds it now. */
+	drm_gem_object_put_unlocked(&shmem->base);
+	if (ret)
+		return ERR_PTR(ret);
+
+	return shmem;
+}
+EXPORT_SYMBOL(drm_gem_shmem_create_with_handle);
+
+/**
+ * drm_gem_shmem_dumb_create - Create a dumb shmem buffer object
+ * @file: DRM file structure to create the dumb buffer for
+ * @dev: DRM device
+ * @args: IOCTL data
+ *
+ * This function computes the pitch of the dumb buffer and rounds it up to an
+ * integer number of bytes per pixel. Drivers for hardware that doesn't have
+ * any additional restrictions on the pitch can directly use this function as
+ * their &drm_driver.dumb_create callback.
+ *
+ * For hardware with additional restrictions, drivers can adjust the fields
+ * set up by userspace before calling into this function.
+ *
+ * Returns:
+ * 0 on success or a negative error code on failure.
+ */
+int drm_gem_shmem_dumb_create(struct drm_file *file, struct drm_device *dev,
+			      struct drm_mode_create_dumb *args)
+{
+	u32 min_pitch = DIV_ROUND_UP(args->width * args->bpp, 8);
+	struct drm_gem_shmem_object *shmem;
+
+	if (!args->pitch || !args->size) {
+		args->pitch = min_pitch;
+		args->size = args->pitch * args->height;
+	} else {
+		/* ensure sane minimum values */
+		if (args->pitch < min_pitch)
+			args->pitch = min_pitch;
+		if (args->size < args->pitch * args->height)
+			args->size = args->pitch * args->height;
+	}
+
+	shmem = drm_gem_shmem_create_with_handle(file, dev, args->size, &args->handle);
+
+	return PTR_ERR_OR_ZERO(shmem);
+}
+EXPORT_SYMBOL_GPL(drm_gem_shmem_dumb_create);
+
+static vm_fault_t drm_gem_shmem_fault(struct vm_fault *vmf)
+{
+	struct vm_area_struct *vma = vmf->vma;
+	struct drm_gem_object *obj = vma->vm_private_data;
+	struct drm_gem_shmem_object *shmem = to_drm_gem_shmem_obj(obj);
+	loff_t num_pages = obj->size >> PAGE_SHIFT;
+	struct page *page;
+
+	if (vmf->pgoff >= num_pages || WARN_ON_ONCE(!shmem->pages))
+		return VM_FAULT_SIGBUS;
+
+	page = shmem->pages[vmf->pgoff];
+
+	return vmf_insert_page(vma, vmf->address, page);
+}
+
+static void drm_gem_shmem_vm_open(struct vm_area_struct *vma)
+{
+	struct drm_gem_object *obj = vma->vm_private_data;
+	struct drm_gem_shmem_object *shmem = to_drm_gem_shmem_obj(obj);
+	int ret;
+
+	ret = drm_gem_shmem_get_pages(shmem);
+	WARN_ON_ONCE(ret != 0);
+
+	drm_gem_vm_open(vma);
+}
+
+static void drm_gem_shmem_vm_close(struct vm_area_struct *vma)
+{
+	struct drm_gem_object *obj = vma->vm_private_data;
+	struct drm_gem_shmem_object *shmem = to_drm_gem_shmem_obj(obj);
+
+	drm_gem_shmem_put_pages(shmem);
+	drm_gem_vm_close(vma);
+}
+
+const struct vm_operations_struct drm_gem_shmem_vm_ops = {
+	.fault = drm_gem_shmem_fault,
+	.open = drm_gem_shmem_vm_open,
+	.close = drm_gem_shmem_vm_close,
+};
+EXPORT_SYMBOL_GPL(drm_gem_shmem_vm_ops);
+
+/**
+ * drm_gem_shmem_mmap - Memory-map a shmem GEM object
+ * @filp: File object
+ * @vma: VMA for the area to be mapped
+ *
+ * This function implements an augmented version of the GEM DRM file mmap
+ * operation for shmem objects. Drivers which employ the shmem helpers should
+ * use this function as their &file_operations.mmap handler in the DRM device file's
+ * file_operations structure.
+ *
+ * Instead of directly referencing this function, drivers should use the
+ * DEFINE_DRM_GEM_SHMEM_FOPS() macro.
+ *
+ * Returns:
+ * 0 on success or a negative error code on failure.
+ */
+int drm_gem_shmem_mmap(struct file *filp, struct vm_area_struct *vma)
+{
+	struct drm_gem_shmem_object *shmem;
+	int ret;
+
+	ret = drm_gem_mmap(filp, vma);
+	if (ret)
+		return ret;
+
+	shmem = to_drm_gem_shmem_obj(vma->vm_private_data);
+
+	ret = drm_gem_shmem_get_pages(shmem);
+	if (ret) {
+		drm_gem_vm_close(vma);
+		return ret;
+	}
+
+	/* VM_PFNMAP was set by drm_gem_mmap() */
+	vma->vm_flags &= ~VM_PFNMAP;
+	vma->vm_flags |= VM_MIXEDMAP;
+
+	/* Remove the fake offset */
+	vma->vm_pgoff -= drm_vma_node_start(&shmem->base.vma_node);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(drm_gem_shmem_mmap);
+
+/**
+ * drm_gem_shmem_print_info() - Print &drm_gem_shmem_object info for debugfs
+ * @p: DRM printer
+ * @indent: Tab indentation level
+ * @obj: GEM object
+ */
+void drm_gem_shmem_print_info(struct drm_printer *p, unsigned int indent,
+			      const struct drm_gem_object *obj)
+{
+	const struct drm_gem_shmem_object *shmem = to_drm_gem_shmem_obj(obj);
+
+	drm_printf_indent(p, indent, "pages_use_count=%u\n", shmem->pages_use_count);
+	drm_printf_indent(p, indent, "vmap_use_count=%u\n", shmem->vmap_use_count);
+	drm_printf_indent(p, indent, "vaddr=%p\n", shmem->vaddr);
+}
+EXPORT_SYMBOL(drm_gem_shmem_print_info);
+
+/**
+ * drm_gem_shmem_get_sg_table - Provide a scatter/gather table of pinned
+ *                              pages for a shmem GEM object
+ * @obj: GEM object
+ *
+ * This function exports a scatter/gather table suitable for PRIME usage by
+ * calling the standard DMA mapping API.
+ *
+ * Returns:
+ * A pointer to the scatter/gather table of pinned pages or NULL on failure.
+ */
+struct sg_table *drm_gem_shmem_get_sg_table(struct drm_gem_object *obj)
+{
+	struct drm_gem_shmem_object *shmem = to_drm_gem_shmem_obj(obj);
+
+	return drm_prime_pages_to_sg(shmem->pages, obj->size >> PAGE_SHIFT);
+}
+EXPORT_SYMBOL_GPL(drm_gem_shmem_get_sg_table);
+
+/**
+ * drm_gem_shmem_get_pages_sgt - Pin pages, dma map them, and return a
+ *				 scatter/gather table for a shmem GEM object.
+ * @obj: GEM object
+ *
+ * This function returns a scatter/gather table suitable for driver usage. If
+ * the sg table doesn't exist, the pages are pinned, dma-mapped, and a sg
+ * table created.
+ *
+ * Returns:
+ * A pointer to the scatter/gather table of pinned pages or errno on failure.
+ */
+struct sg_table *drm_gem_shmem_get_pages_sgt(struct drm_gem_object *obj)
+{
+	int ret;
+	struct drm_gem_shmem_object *shmem = to_drm_gem_shmem_obj(obj);
+	struct sg_table *sgt;
+
+	if (shmem->sgt)
+		return shmem->sgt;
+
+	WARN_ON(obj->import_attach);
+
+	ret = drm_gem_shmem_get_pages(shmem);
+	if (ret)
+		return ERR_PTR(ret);
+
+	sgt = drm_gem_shmem_get_sg_table(&shmem->base);
+	if (IS_ERR(sgt)) {
+		ret = PTR_ERR(sgt);
+		goto err_put_pages;
+	}
+	/* Map the pages for use by the h/w. */
+	dma_map_sg(obj->dev->dev, sgt->sgl, sgt->nents, DMA_BIDIRECTIONAL);
+
+	shmem->sgt = sgt;
+
+	return sgt;
+
+err_put_pages:
+	drm_gem_shmem_put_pages(shmem);
+	return ERR_PTR(ret);
+}
+EXPORT_SYMBOL_GPL(drm_gem_shmem_get_pages_sgt);
+
+/**
+ * drm_gem_shmem_prime_import_sg_table - Produce a shmem GEM object from
+ *                 another driver's scatter/gather table of pinned pages
+ * @dev: Device to import into
+ * @attach: DMA-BUF attachment
+ * @sgt: Scatter/gather table of pinned pages
+ *
+ * This function imports a scatter/gather table exported via DMA-BUF by
+ * another driver. Drivers that use the shmem helpers should set this as their
+ * &drm_driver.gem_prime_import_sg_table callback.
+ *
+ * Returns:
+ * A pointer to a newly created GEM object or an ERR_PTR-encoded negative
+ * error code on failure.
+ */
+struct drm_gem_object *
+drm_gem_shmem_prime_import_sg_table(struct drm_device *dev,
+				    struct dma_buf_attachment *attach,
+				    struct sg_table *sgt)
+{
+	size_t size = PAGE_ALIGN(attach->dmabuf->size);
+	size_t npages = size >> PAGE_SHIFT;
+	struct drm_gem_shmem_object *shmem;
+	int ret;
+
+	shmem = drm_gem_shmem_create(dev, size);
+	if (IS_ERR(shmem))
+		return ERR_CAST(shmem);
+
+	shmem->pages = kvmalloc_array(npages, sizeof(struct page *), GFP_KERNEL);
+	if (!shmem->pages) {
+		ret = -ENOMEM;
+		goto err_free_gem;
+	}
+
+	ret = drm_prime_sg_to_page_addr_arrays(sgt, shmem->pages, NULL, npages);
+	if (ret < 0)
+		goto err_free_array;
+
+	shmem->sgt = sgt;
+	shmem->pages_use_count = 1; /* Permanently pinned from our point of view */
+
+	DRM_DEBUG_PRIME("size = %zu\n", size);
+
+	return &shmem->base;
+
+err_free_array:
+	kvfree(shmem->pages);
+err_free_gem:
+	drm_gem_object_put_unlocked(&shmem->base);
+
+	return ERR_PTR(ret);
+}
+EXPORT_SYMBOL_GPL(drm_gem_shmem_prime_import_sg_table);
diff --git a/drivers/gpu/drm/drm_internal.h b/drivers/gpu/drm/drm_internal.h
index 251d67e04c2d..e19ac7ca602d 100644
--- a/drivers/gpu/drm/drm_internal.h
+++ b/drivers/gpu/drm/drm_internal.h
@@ -71,8 +71,10 @@ int drm_legacy_modeset_ctl_ioctl(struct drm_device *dev, void *data,
 /* drm_irq.c */
 
 /* IOCTLS */
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
 int drm_legacy_irq_control(struct drm_device *dev, void *data,
 			   struct drm_file *file_priv);
+#endif
 
 int drm_crtc_get_sequence_ioctl(struct drm_device *dev, void *data,
 				struct drm_file *filp);
@@ -180,12 +182,20 @@ int drm_syncobj_handle_to_fd_ioctl(struct drm_device *dev, void *data,
 				   struct drm_file *file_private);
 int drm_syncobj_fd_to_handle_ioctl(struct drm_device *dev, void *data,
 				   struct drm_file *file_private);
+int drm_syncobj_transfer_ioctl(struct drm_device *dev, void *data,
+			       struct drm_file *file_private);
 int drm_syncobj_wait_ioctl(struct drm_device *dev, void *data,
 			   struct drm_file *file_private);
+int drm_syncobj_timeline_wait_ioctl(struct drm_device *dev, void *data,
+				    struct drm_file *file_private);
 int drm_syncobj_reset_ioctl(struct drm_device *dev, void *data,
 			    struct drm_file *file_private);
 int drm_syncobj_signal_ioctl(struct drm_device *dev, void *data,
 			     struct drm_file *file_private);
+int drm_syncobj_timeline_signal_ioctl(struct drm_device *dev, void *data,
+				      struct drm_file *file_private);
+int drm_syncobj_query_ioctl(struct drm_device *dev, void *data,
+			    struct drm_file *file_private);
 
 /* drm_framebuffer.c */
 void drm_framebuffer_print_info(struct drm_printer *p, unsigned int indent,
diff --git a/drivers/gpu/drm/drm_ioc32.c b/drivers/gpu/drm/drm_ioc32.c
index 0e3043e08c69..374b372da58a 100644
--- a/drivers/gpu/drm/drm_ioc32.c
+++ b/drivers/gpu/drm/drm_ioc32.c
@@ -156,6 +156,7 @@ static int compat_drm_setunique(struct file *file, unsigned int cmd,
 	return -EINVAL;
 }
 
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
 typedef struct drm_map32 {
 	u32 offset;		/* Requested physical address (0 for SAREA) */
 	u32 size;		/* Requested physical size (bytes) */
@@ -239,6 +240,7 @@ static int compat_drm_rmmap(struct file *file, unsigned int cmd,
 	map.handle = compat_ptr(handle);
 	return drm_ioctl_kernel(file, drm_legacy_rmmap_ioctl, &map, DRM_AUTH);
 }
+#endif
 
 typedef struct drm_client32 {
 	int idx;	/* Which client desired? */
@@ -301,6 +303,7 @@ static int compat_drm_getstats(struct file *file, unsigned int cmd,
 	return 0;
 }
 
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
 typedef struct drm_buf_desc32 {
 	int count;		 /* Number of buffers of this size */
 	int size;		 /* Size in bytes */
@@ -604,6 +607,7 @@ static int compat_drm_dma(struct file *file, unsigned int cmd,
 
 	return 0;
 }
+#endif
 
 #if IS_ENABLED(CONFIG_AGP)
 typedef struct drm_agp_mode32 {
@@ -748,6 +752,7 @@ static int compat_drm_agp_unbind(struct file *file, unsigned int cmd,
 }
 #endif /* CONFIG_AGP */
 
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
 typedef struct drm_scatter_gather32 {
 	u32 size;	/**< In bytes -- will round to page boundary */
 	u32 handle;	/**< Used for mapping / unmapping */
@@ -788,7 +793,7 @@ static int compat_drm_sg_free(struct file *file, unsigned int cmd,
 	return drm_ioctl_kernel(file, drm_legacy_sg_free, &request,
 				DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY);
 }
-
+#endif
 #if defined(CONFIG_X86)
 typedef struct drm_update_draw32 {
 	drm_drawable_t handle;
@@ -903,10 +908,13 @@ static struct {
 #define DRM_IOCTL32_DEF(n, f) [DRM_IOCTL_NR(n##32)] = {.fn = f, .name = #n}
 	DRM_IOCTL32_DEF(DRM_IOCTL_VERSION, compat_drm_version),
 	DRM_IOCTL32_DEF(DRM_IOCTL_GET_UNIQUE, compat_drm_getunique),
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
 	DRM_IOCTL32_DEF(DRM_IOCTL_GET_MAP, compat_drm_getmap),
+#endif
 	DRM_IOCTL32_DEF(DRM_IOCTL_GET_CLIENT, compat_drm_getclient),
 	DRM_IOCTL32_DEF(DRM_IOCTL_GET_STATS, compat_drm_getstats),
 	DRM_IOCTL32_DEF(DRM_IOCTL_SET_UNIQUE, compat_drm_setunique),
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
 	DRM_IOCTL32_DEF(DRM_IOCTL_ADD_MAP, compat_drm_addmap),
 	DRM_IOCTL32_DEF(DRM_IOCTL_ADD_BUFS, compat_drm_addbufs),
 	DRM_IOCTL32_DEF(DRM_IOCTL_MARK_BUFS, compat_drm_markbufs),
@@ -918,6 +926,7 @@ static struct {
 	DRM_IOCTL32_DEF(DRM_IOCTL_GET_SAREA_CTX, compat_drm_getsareactx),
 	DRM_IOCTL32_DEF(DRM_IOCTL_RES_CTX, compat_drm_resctx),
 	DRM_IOCTL32_DEF(DRM_IOCTL_DMA, compat_drm_dma),
+#endif
 #if IS_ENABLED(CONFIG_AGP)
 	DRM_IOCTL32_DEF(DRM_IOCTL_AGP_ENABLE, compat_drm_agp_enable),
 	DRM_IOCTL32_DEF(DRM_IOCTL_AGP_INFO, compat_drm_agp_info),
@@ -926,8 +935,10 @@ static struct {
 	DRM_IOCTL32_DEF(DRM_IOCTL_AGP_BIND, compat_drm_agp_bind),
 	DRM_IOCTL32_DEF(DRM_IOCTL_AGP_UNBIND, compat_drm_agp_unbind),
 #endif
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
 	DRM_IOCTL32_DEF(DRM_IOCTL_SG_ALLOC, compat_drm_sg_alloc),
 	DRM_IOCTL32_DEF(DRM_IOCTL_SG_FREE, compat_drm_sg_free),
+#endif
 #if defined(CONFIG_X86) || defined(CONFIG_IA64)
 	DRM_IOCTL32_DEF(DRM_IOCTL_UPDATE_DRAW, compat_drm_update_draw),
 #endif
diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
index 687943df58e1..2263e3ddd822 100644
--- a/drivers/gpu/drm/drm_ioctl.c
+++ b/drivers/gpu/drm/drm_ioctl.c
@@ -245,6 +245,9 @@ static int drm_getcap(struct drm_device *dev, void *data, struct drm_file *file_
 	case DRM_CAP_SYNCOBJ:
 		req->value = drm_core_check_feature(dev, DRIVER_SYNCOBJ);
 		return 0;
+	case DRM_CAP_SYNCOBJ_TIMELINE:
+		req->value = drm_core_check_feature(dev, DRIVER_SYNCOBJ_TIMELINE);
+		return 0;
 	}
 
 	/* Other caps only work with KMS drivers */
@@ -508,13 +511,6 @@ int drm_version(struct drm_device *dev, void *data,
 	return err;
 }
 
-static inline bool
-drm_render_driver_and_ioctl(const struct drm_device *dev, u32 flags)
-{
-	return drm_core_check_feature(dev, DRIVER_RENDER) &&
-		(flags & DRM_RENDER_ALLOW);
-}
-
 /**
  * drm_ioctl_permit - Check ioctl permissions against caller
  *
@@ -529,19 +525,14 @@ drm_render_driver_and_ioctl(const struct drm_device *dev, u32 flags)
  */
 int drm_ioctl_permit(u32 flags, struct drm_file *file_priv)
 {
-	const struct drm_device *dev = file_priv->minor->dev;
-
 	/* ROOT_ONLY is only for CAP_SYS_ADMIN */
 	if (unlikely((flags & DRM_ROOT_ONLY) && !capable(CAP_SYS_ADMIN)))
 		return -EACCES;
 
-	/* AUTH is only for master ... */
-	if (unlikely((flags & DRM_AUTH) && drm_is_primary_client(file_priv))) {
-		/* authenticated ones, or render capable on DRM_RENDER_ALLOW. */
-		if (!file_priv->authenticated &&
-		    !drm_render_driver_and_ioctl(dev, flags))
-			return -EACCES;
-	}
+	/* AUTH is only for authenticated or render client */
+	if (unlikely((flags & DRM_AUTH) && !drm_is_render_client(file_priv) &&
+		     !file_priv->authenticated))
+		return -EACCES;
 
 	/* MASTER is only for master or control clients */
 	if (unlikely((flags & DRM_MASTER) &&
@@ -565,6 +556,12 @@ EXPORT_SYMBOL(drm_ioctl_permit);
 		.name = #ioctl			\
 	}
 
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
+#define DRM_LEGACY_IOCTL_DEF(ioctl, _func, _flags)  DRM_IOCTL_DEF(ioctl, _func, _flags)
+#else
+#define DRM_LEGACY_IOCTL_DEF(ioctl, _func, _flags) DRM_IOCTL_DEF(ioctl, drm_invalid_op, _flags)
+#endif
+
 /* Ioctl table */
 static const struct drm_ioctl_desc drm_ioctls[] = {
 	DRM_IOCTL_DEF(DRM_IOCTL_VERSION, drm_version,
@@ -572,7 +569,9 @@ static const struct drm_ioctl_desc drm_ioctls[] = {
 	DRM_IOCTL_DEF(DRM_IOCTL_GET_UNIQUE, drm_getunique, DRM_UNLOCKED),
 	DRM_IOCTL_DEF(DRM_IOCTL_GET_MAGIC, drm_getmagic, DRM_UNLOCKED),
 	DRM_IOCTL_DEF(DRM_IOCTL_IRQ_BUSID, drm_irq_by_busid, DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF(DRM_IOCTL_GET_MAP, drm_legacy_getmap_ioctl, DRM_UNLOCKED),
+
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_GET_MAP, drm_legacy_getmap_ioctl, DRM_UNLOCKED),
+
 	DRM_IOCTL_DEF(DRM_IOCTL_GET_CLIENT, drm_getclient, DRM_UNLOCKED),
 	DRM_IOCTL_DEF(DRM_IOCTL_GET_STATS, drm_getstats, DRM_UNLOCKED),
 	DRM_IOCTL_DEF(DRM_IOCTL_GET_CAP, drm_getcap, DRM_UNLOCKED|DRM_RENDER_ALLOW),
@@ -584,39 +583,38 @@ static const struct drm_ioctl_desc drm_ioctls[] = {
 	DRM_IOCTL_DEF(DRM_IOCTL_UNBLOCK, drm_noop, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
 	DRM_IOCTL_DEF(DRM_IOCTL_AUTH_MAGIC, drm_authmagic, DRM_UNLOCKED|DRM_MASTER),
 
-	DRM_IOCTL_DEF(DRM_IOCTL_ADD_MAP, drm_legacy_addmap_ioctl, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF(DRM_IOCTL_RM_MAP, drm_legacy_rmmap_ioctl, DRM_AUTH),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_ADD_MAP, drm_legacy_addmap_ioctl, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_RM_MAP, drm_legacy_rmmap_ioctl, DRM_AUTH),
 
-	DRM_IOCTL_DEF(DRM_IOCTL_SET_SAREA_CTX, drm_legacy_setsareactx, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF(DRM_IOCTL_GET_SAREA_CTX, drm_legacy_getsareactx, DRM_AUTH),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_SET_SAREA_CTX, drm_legacy_setsareactx, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_GET_SAREA_CTX, drm_legacy_getsareactx, DRM_AUTH),
 
 	DRM_IOCTL_DEF(DRM_IOCTL_SET_MASTER, drm_setmaster_ioctl, DRM_UNLOCKED|DRM_ROOT_ONLY),
 	DRM_IOCTL_DEF(DRM_IOCTL_DROP_MASTER, drm_dropmaster_ioctl, DRM_UNLOCKED|DRM_ROOT_ONLY),
 
-	DRM_IOCTL_DEF(DRM_IOCTL_ADD_CTX, drm_legacy_addctx, DRM_AUTH|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF(DRM_IOCTL_RM_CTX, drm_legacy_rmctx, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF(DRM_IOCTL_MOD_CTX, drm_noop, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF(DRM_IOCTL_GET_CTX, drm_legacy_getctx, DRM_AUTH),
-	DRM_IOCTL_DEF(DRM_IOCTL_SWITCH_CTX, drm_legacy_switchctx, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF(DRM_IOCTL_NEW_CTX, drm_legacy_newctx, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF(DRM_IOCTL_RES_CTX, drm_legacy_resctx, DRM_AUTH),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_ADD_CTX, drm_legacy_addctx, DRM_AUTH|DRM_ROOT_ONLY),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_RM_CTX, drm_legacy_rmctx, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_MOD_CTX, drm_noop, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_GET_CTX, drm_legacy_getctx, DRM_AUTH),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_SWITCH_CTX, drm_legacy_switchctx, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_NEW_CTX, drm_legacy_newctx, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_RES_CTX, drm_legacy_resctx, DRM_AUTH),
 
 	DRM_IOCTL_DEF(DRM_IOCTL_ADD_DRAW, drm_noop, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
 	DRM_IOCTL_DEF(DRM_IOCTL_RM_DRAW, drm_noop, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
 
-	DRM_IOCTL_DEF(DRM_IOCTL_LOCK, drm_legacy_lock, DRM_AUTH),
-	DRM_IOCTL_DEF(DRM_IOCTL_UNLOCK, drm_legacy_unlock, DRM_AUTH),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_LOCK, drm_legacy_lock, DRM_AUTH),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_UNLOCK, drm_legacy_unlock, DRM_AUTH),
 
 	DRM_IOCTL_DEF(DRM_IOCTL_FINISH, drm_noop, DRM_AUTH),
 
-	DRM_IOCTL_DEF(DRM_IOCTL_ADD_BUFS, drm_legacy_addbufs, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF(DRM_IOCTL_MARK_BUFS, drm_legacy_markbufs, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF(DRM_IOCTL_INFO_BUFS, drm_legacy_infobufs, DRM_AUTH),
-	DRM_IOCTL_DEF(DRM_IOCTL_MAP_BUFS, drm_legacy_mapbufs, DRM_AUTH),
-	DRM_IOCTL_DEF(DRM_IOCTL_FREE_BUFS, drm_legacy_freebufs, DRM_AUTH),
-	DRM_IOCTL_DEF(DRM_IOCTL_DMA, drm_legacy_dma_ioctl, DRM_AUTH),
-
-	DRM_IOCTL_DEF(DRM_IOCTL_CONTROL, drm_legacy_irq_control, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_ADD_BUFS, drm_legacy_addbufs, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_MARK_BUFS, drm_legacy_markbufs, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_INFO_BUFS, drm_legacy_infobufs, DRM_AUTH),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_MAP_BUFS, drm_legacy_mapbufs, DRM_AUTH),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_FREE_BUFS, drm_legacy_freebufs, DRM_AUTH),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_DMA, drm_legacy_dma_ioctl, DRM_AUTH),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_CONTROL, drm_legacy_irq_control, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
 
 #if IS_ENABLED(CONFIG_AGP)
 	DRM_IOCTL_DEF(DRM_IOCTL_AGP_ACQUIRE, drm_agp_acquire_ioctl, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
@@ -629,8 +627,8 @@ static const struct drm_ioctl_desc drm_ioctls[] = {
 	DRM_IOCTL_DEF(DRM_IOCTL_AGP_UNBIND, drm_agp_unbind_ioctl, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
 #endif
 
-	DRM_IOCTL_DEF(DRM_IOCTL_SG_ALLOC, drm_legacy_sg_alloc, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF(DRM_IOCTL_SG_FREE, drm_legacy_sg_free, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_SG_ALLOC, drm_legacy_sg_alloc, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_LEGACY_IOCTL_DEF(DRM_IOCTL_SG_FREE, drm_legacy_sg_free, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
 
 	DRM_IOCTL_DEF(DRM_IOCTL_WAIT_VBLANK, drm_wait_vblank_ioctl, DRM_UNLOCKED),
 
@@ -686,12 +684,20 @@ static const struct drm_ioctl_desc drm_ioctls[] = {
 		      DRM_UNLOCKED|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF(DRM_IOCTL_SYNCOBJ_FD_TO_HANDLE, drm_syncobj_fd_to_handle_ioctl,
 		      DRM_UNLOCKED|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF(DRM_IOCTL_SYNCOBJ_TRANSFER, drm_syncobj_transfer_ioctl,
+		      DRM_UNLOCKED|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF(DRM_IOCTL_SYNCOBJ_WAIT, drm_syncobj_wait_ioctl,
 		      DRM_UNLOCKED|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF(DRM_IOCTL_SYNCOBJ_TIMELINE_WAIT, drm_syncobj_timeline_wait_ioctl,
+		      DRM_UNLOCKED|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF(DRM_IOCTL_SYNCOBJ_RESET, drm_syncobj_reset_ioctl,
 		      DRM_UNLOCKED|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF(DRM_IOCTL_SYNCOBJ_SIGNAL, drm_syncobj_signal_ioctl,
 		      DRM_UNLOCKED|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF(DRM_IOCTL_SYNCOBJ_TIMELINE_SIGNAL, drm_syncobj_timeline_signal_ioctl,
+		      DRM_UNLOCKED|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF(DRM_IOCTL_SYNCOBJ_QUERY, drm_syncobj_query_ioctl,
+		      DRM_UNLOCKED|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF(DRM_IOCTL_CRTC_GET_SEQUENCE, drm_crtc_get_sequence_ioctl, DRM_UNLOCKED),
 	DRM_IOCTL_DEF(DRM_IOCTL_CRTC_QUEUE_SEQUENCE, drm_crtc_queue_sequence_ioctl, DRM_UNLOCKED),
 	DRM_IOCTL_DEF(DRM_IOCTL_MODE_CREATE_LEASE, drm_mode_create_lease_ioctl, DRM_MASTER|DRM_UNLOCKED),
diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index 9bd8908d5fd8..02f38cc9f468 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -213,6 +213,7 @@ int drm_irq_uninstall(struct drm_device *dev)
 }
 EXPORT_SYMBOL(drm_irq_uninstall);
 
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
 int drm_legacy_irq_control(struct drm_device *dev, void *data,
 			   struct drm_file *file_priv)
 {
@@ -253,3 +254,4 @@ int drm_legacy_irq_control(struct drm_device *dev, void *data,
 		return -EINVAL;
 	}
 }
+#endif
diff --git a/drivers/gpu/drm/drm_kms_helper_common.c b/drivers/gpu/drm/drm_kms_helper_common.c
index 93e2b30fe1a5..9c5ae825c507 100644
--- a/drivers/gpu/drm/drm_kms_helper_common.c
+++ b/drivers/gpu/drm/drm_kms_helper_common.c
@@ -39,7 +39,7 @@ MODULE_LICENSE("GPL and additional rights");
 /* Backward compatibility for drm_kms_helper.edid_firmware */
 static int edid_firmware_set(const char *val, const struct kernel_param *kp)
 {
-	DRM_NOTE("drm_kms_firmware.edid_firmware is deprecated, please use drm.edid_firmware intead.\n");
+	DRM_NOTE("drm_kms_firmware.edid_firmware is deprecated, please use drm.edid_firmware instead.\n");
 
 	return __drm_set_edid_firmware_path(val);
 }
diff --git a/drivers/gpu/drm/drm_lease.c b/drivers/gpu/drm/drm_lease.c
index 603b0bd9c5ce..694ff363a90b 100644
--- a/drivers/gpu/drm/drm_lease.c
+++ b/drivers/gpu/drm/drm_lease.c
@@ -111,7 +111,7 @@ static bool _drm_has_leased(struct drm_master *master, int id)
  */
 bool _drm_lease_held(struct drm_file *file_priv, int id)
 {
-	if (file_priv == NULL || file_priv->master == NULL)
+	if (!file_priv || !file_priv->master)
 		return true;
 
 	return _drm_lease_held_master(file_priv->master, id);
@@ -133,7 +133,7 @@ bool drm_lease_held(struct drm_file *file_priv, int id)
 	struct drm_master *master;
 	bool ret;
 
-	if (file_priv == NULL || file_priv->master == NULL)
+	if (!file_priv || !file_priv->master || !file_priv->master->lessor)
 		return true;
 
 	master = file_priv->master;
@@ -159,7 +159,7 @@ uint32_t drm_lease_filter_crtcs(struct drm_file *file_priv, uint32_t crtcs_in)
 	int count_in, count_out;
 	uint32_t crtcs_out = 0;
 
-	if (file_priv == NULL || file_priv->master == NULL)
+	if (!file_priv || !file_priv->master || !file_priv->master->lessor)
 		return crtcs_in;
 
 	master = file_priv->master;
@@ -220,8 +220,6 @@ static struct drm_master *drm_lease_create(struct drm_master *lessor, struct idr
 		error = 0;
 		if (!idr_find(&dev->mode_config.object_idr, object))
 			error = -ENOENT;
-		else if (!_drm_lease_held_master(lessor, object))
-			error = -EACCES;
 		else if (_drm_has_leased(lessor, object))
 			error = -EBUSY;
 
@@ -403,11 +401,6 @@ static int fill_object_idr(struct drm_device *dev,
 	/* step one - get references to all the mode objects
 	   and check for validity. */
 	for (o = 0; o < object_count; o++) {
-		if ((int) object_ids[o] < 0) {
-			ret = -EINVAL;
-			goto out_free_objects;
-		}
-
 		objects[o] = drm_mode_object_find(dev, lessor_priv,
 						  object_ids[o],
 						  DRM_MODE_OBJECT_ANY);
diff --git a/drivers/gpu/drm/drm_legacy.h b/drivers/gpu/drm/drm_legacy.h
index 280fbeb846ff..51f1fabfa145 100644
--- a/drivers/gpu/drm/drm_legacy.h
+++ b/drivers/gpu/drm/drm_legacy.h
@@ -42,11 +42,19 @@ struct drm_file;
 #define DRM_KERNEL_CONTEXT		0
 #define DRM_RESERVED_CONTEXTS		1
 
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
 void drm_legacy_ctxbitmap_init(struct drm_device *dev);
 void drm_legacy_ctxbitmap_cleanup(struct drm_device *dev);
-void drm_legacy_ctxbitmap_free(struct drm_device *dev, int ctx_handle);
 void drm_legacy_ctxbitmap_flush(struct drm_device *dev, struct drm_file *file);
+#else
+static inline void drm_legacy_ctxbitmap_init(struct drm_device *dev) {}
+static inline void drm_legacy_ctxbitmap_cleanup(struct drm_device *dev) {}
+static inline void drm_legacy_ctxbitmap_flush(struct drm_device *dev, struct drm_file *file) {}
+#endif
 
+void drm_legacy_ctxbitmap_free(struct drm_device *dev, int ctx_handle);
+
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
 int drm_legacy_resctx(struct drm_device *d, void *v, struct drm_file *f);
 int drm_legacy_addctx(struct drm_device *d, void *v, struct drm_file *f);
 int drm_legacy_getctx(struct drm_device *d, void *v, struct drm_file *f);
@@ -56,6 +64,7 @@ int drm_legacy_rmctx(struct drm_device *d, void *v, struct drm_file *f);
 
 int drm_legacy_setsareactx(struct drm_device *d, void *v, struct drm_file *f);
 int drm_legacy_getsareactx(struct drm_device *d, void *v, struct drm_file *f);
+#endif
 
 /*
  * Generic Buffer Management
@@ -63,16 +72,39 @@ int drm_legacy_getsareactx(struct drm_device *d, void *v, struct drm_file *f);
 
 #define DRM_MAP_HASH_OFFSET 0x10000000
 
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
+static inline int drm_legacy_create_map_hash(struct drm_device *dev)
+{
+	return drm_ht_create(&dev->map_hash, 12);
+}
+
+static inline void drm_legacy_remove_map_hash(struct drm_device *dev)
+{
+	drm_ht_remove(&dev->map_hash);
+}
+#else
+static inline int drm_legacy_create_map_hash(struct drm_device *dev)
+{
+	return 0;
+}
+
+static inline void drm_legacy_remove_map_hash(struct drm_device *dev) {}
+#endif
+
+
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
 int drm_legacy_getmap_ioctl(struct drm_device *dev, void *data,
 			    struct drm_file *file_priv);
 int drm_legacy_addmap_ioctl(struct drm_device *d, void *v, struct drm_file *f);
 int drm_legacy_rmmap_ioctl(struct drm_device *d, void *v, struct drm_file *f);
+
 int drm_legacy_addbufs(struct drm_device *d, void *v, struct drm_file *f);
 int drm_legacy_infobufs(struct drm_device *d, void *v, struct drm_file *f);
 int drm_legacy_markbufs(struct drm_device *d, void *v, struct drm_file *f);
 int drm_legacy_freebufs(struct drm_device *d, void *v, struct drm_file *f);
 int drm_legacy_mapbufs(struct drm_device *d, void *v, struct drm_file *f);
 int drm_legacy_dma_ioctl(struct drm_device *d, void *v, struct drm_file *f);
+#endif
 
 int __drm_legacy_infobufs(struct drm_device *, void *, int *,
 			  int (*)(void *, int, struct drm_buf_entry *));
@@ -81,7 +113,17 @@ int __drm_legacy_mapbufs(struct drm_device *, void *, int *,
 			  int (*)(void *, int, unsigned long, struct drm_buf *),
 			  struct drm_file *);
 
-#ifdef CONFIG_DRM_VM
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
+void drm_legacy_master_rmmaps(struct drm_device *dev,
+			      struct drm_master *master);
+void drm_legacy_rmmaps(struct drm_device *dev);
+#else
+static inline void drm_legacy_master_rmmaps(struct drm_device *dev,
+					    struct drm_master *master) {}
+static inline void drm_legacy_rmmaps(struct drm_device *dev) {}
+#endif
+
+#if IS_ENABLED(CONFIG_DRM_VM) && IS_ENABLED(CONFIG_DRM_LEGACY)
 void drm_legacy_vma_flush(struct drm_device *d);
 #else
 static inline void drm_legacy_vma_flush(struct drm_device *d)
@@ -103,23 +145,64 @@ struct drm_agp_mem {
 };
 
 /* drm_lock.c */
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
 int drm_legacy_lock(struct drm_device *d, void *v, struct drm_file *f);
 int drm_legacy_unlock(struct drm_device *d, void *v, struct drm_file *f);
 void drm_legacy_lock_release(struct drm_device *dev, struct file *filp);
+#else
+static inline void drm_legacy_lock_release(struct drm_device *dev, struct file *filp) {}
+#endif
 
 /* DMA support */
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
 int drm_legacy_dma_setup(struct drm_device *dev);
 void drm_legacy_dma_takedown(struct drm_device *dev);
+#else
+static inline int drm_legacy_dma_setup(struct drm_device *dev)
+{
+	return 0;
+}
+#endif
+
 void drm_legacy_free_buffer(struct drm_device *dev,
 			    struct drm_buf * buf);
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
 void drm_legacy_reclaim_buffers(struct drm_device *dev,
 				struct drm_file *filp);
+#else
+static inline void drm_legacy_reclaim_buffers(struct drm_device *dev,
+					      struct drm_file *filp) {}
+#endif
 
 /* Scatter Gather Support */
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
 void drm_legacy_sg_cleanup(struct drm_device *dev);
 int drm_legacy_sg_alloc(struct drm_device *dev, void *data,
 			struct drm_file *file_priv);
 int drm_legacy_sg_free(struct drm_device *dev, void *data,
 		       struct drm_file *file_priv);
+#endif
+
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
+void drm_legacy_init_members(struct drm_device *dev);
+void drm_legacy_destroy_members(struct drm_device *dev);
+void drm_legacy_dev_reinit(struct drm_device *dev);
+#else
+static inline void drm_legacy_init_members(struct drm_device *dev) {}
+static inline void drm_legacy_destroy_members(struct drm_device *dev) {}
+static inline void drm_legacy_dev_reinit(struct drm_device *dev) {}
+#endif
+
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
+void drm_legacy_lock_master_cleanup(struct drm_device *dev, struct drm_master *master);
+#else
+static inline void drm_legacy_lock_master_cleanup(struct drm_device *dev, struct drm_master *master) {}
+#endif
+
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
+void drm_master_legacy_init(struct drm_master *master);
+#else
+static inline void drm_master_legacy_init(struct drm_master *master) {}
+#endif
 
 #endif /* __DRM_LEGACY_H__ */
diff --git a/drivers/gpu/drm/drm_legacy_misc.c b/drivers/gpu/drm/drm_legacy_misc.c
new file mode 100644
index 000000000000..2fe786839ca8
--- /dev/null
+++ b/drivers/gpu/drm/drm_legacy_misc.c
@@ -0,0 +1,82 @@
+/**
+ * \file drm_legacy_misc.c
+ * Misc legacy support functions.
+ *
+ * \author Rickard E. (Rik) Faith <faith@valinux.com>
+ * \author Gareth Hughes <gareth@valinux.com>
+ */
+
+/*
+ * Created: Tue Feb  2 08:37:54 1999 by faith@valinux.com
+ *
+ * Copyright 1999 Precision Insight, Inc., Cedar Park, Texas.
+ * Copyright 2000 VA Linux Systems, Inc., Sunnyvale, California.
+ * All Rights Reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * VA LINUX SYSTEMS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#include <drm/drmP.h>
+#include "drm_internal.h"
+#include "drm_legacy.h"
+
+void drm_legacy_init_members(struct drm_device *dev)
+{
+	INIT_LIST_HEAD(&dev->ctxlist);
+	INIT_LIST_HEAD(&dev->vmalist);
+	INIT_LIST_HEAD(&dev->maplist);
+	spin_lock_init(&dev->buf_lock);
+	mutex_init(&dev->ctxlist_mutex);
+}
+
+void drm_legacy_destroy_members(struct drm_device *dev)
+{
+	mutex_destroy(&dev->ctxlist_mutex);
+}
+
+void drm_legacy_dev_reinit(struct drm_device *dev)
+{
+	if (dev->irq_enabled)
+		drm_irq_uninstall(dev);
+
+	mutex_lock(&dev->struct_mutex);
+
+	drm_legacy_agp_clear(dev);
+
+	drm_legacy_sg_cleanup(dev);
+	drm_legacy_vma_flush(dev);
+	drm_legacy_dma_takedown(dev);
+
+	mutex_unlock(&dev->struct_mutex);
+
+	dev->sigdata.lock = NULL;
+
+	dev->context_flag = 0;
+	dev->last_context = 0;
+	dev->if_version = 0;
+
+	DRM_DEBUG("lastclose completed\n");
+}
+
+void drm_master_legacy_init(struct drm_master *master)
+{
+	spin_lock_init(&master->lock.spinlock);
+	init_waitqueue_head(&master->lock.lock_queue);
+}
diff --git a/drivers/gpu/drm/drm_lock.c b/drivers/gpu/drm/drm_lock.c
index 67a1a2ca7174..b70058e77a28 100644
--- a/drivers/gpu/drm/drm_lock.c
+++ b/drivers/gpu/drm/drm_lock.c
@@ -347,3 +347,22 @@ void drm_legacy_lock_release(struct drm_device *dev, struct file *filp)
 				     _DRM_LOCKING_CONTEXT(file_priv->master->lock.hw_lock->lock));
 	}
 }
+
+void drm_legacy_lock_master_cleanup(struct drm_device *dev, struct drm_master *master)
+{
+	if (!drm_core_check_feature(dev, DRIVER_LEGACY))
+		return;
+
+	/*
+	 * Since the master is disappearing, so is the
+	 * possibility to lock.
+	 */	mutex_lock(&dev->struct_mutex);
+	if (master->lock.hw_lock) {
+		if (dev->sigdata.lock == master->lock.hw_lock)
+			dev->sigdata.lock = NULL;
+		master->lock.hw_lock = NULL;
+		master->lock.file_priv = NULL;
+		wake_up_interruptible_all(&master->lock.lock_queue);
+	}
+	mutex_unlock(&dev->struct_mutex);
+}
diff --git a/drivers/gpu/drm/drm_memory.c b/drivers/gpu/drm/drm_memory.c
index 40c4349cb939..132fef8ff1b6 100644
--- a/drivers/gpu/drm/drm_memory.c
+++ b/drivers/gpu/drm/drm_memory.c
@@ -35,6 +35,7 @@
 
 #include <linux/highmem.h>
 #include <linux/export.h>
+#include <xen/xen.h>
 #include <drm/drmP.h>
 #include "drm_legacy.h"
 
@@ -150,15 +151,34 @@ void drm_legacy_ioremapfree(struct drm_local_map *map, struct drm_device *dev)
 }
 EXPORT_SYMBOL(drm_legacy_ioremapfree);
 
-u64 drm_get_max_iomem(void)
+bool drm_need_swiotlb(int dma_bits)
 {
 	struct resource *tmp;
 	resource_size_t max_iomem = 0;
 
+	/*
+	 * Xen paravirtual hosts require swiotlb regardless of requested dma
+	 * transfer size.
+	 *
+	 * NOTE: Really, what it requires is use of the dma_alloc_coherent
+	 *       allocator used in ttm_dma_populate() instead of
+	 *       ttm_populate_and_map_pages(), which bounce buffers so much in
+	 *       Xen it leads to swiotlb buffer exhaustion.
+	 */
+	if (xen_pv_domain())
+		return true;
+
+	/*
+	 * Enforce dma_alloc_coherent when memory encryption is active as well
+	 * for the same reasons as for Xen paravirtual hosts.
+	 */
+	if (mem_encrypt_active())
+		return true;
+
 	for (tmp = iomem_resource.child; tmp; tmp = tmp->sibling) {
 		max_iomem = max(max_iomem,  tmp->end);
 	}
 
-	return max_iomem;
+	return max_iomem > ((u64)1 << dma_bits);
 }
-EXPORT_SYMBOL(drm_get_max_iomem);
+EXPORT_SYMBOL(drm_need_swiotlb);
diff --git a/drivers/gpu/drm/drm_mode_config.c b/drivers/gpu/drm/drm_mode_config.c
index 4a1c2023ccf0..1a346ae1599d 100644
--- a/drivers/gpu/drm/drm_mode_config.c
+++ b/drivers/gpu/drm/drm_mode_config.c
@@ -297,8 +297,9 @@ static int drm_mode_create_standard_properties(struct drm_device *dev)
 		return -ENOMEM;
 	dev->mode_config.prop_crtc_id = prop;
 
-	prop = drm_property_create(dev, DRM_MODE_PROP_BLOB, "FB_DAMAGE_CLIPS",
-				   0);
+	prop = drm_property_create(dev,
+			DRM_MODE_PROP_ATOMIC | DRM_MODE_PROP_BLOB,
+			"FB_DAMAGE_CLIPS", 0);
 	if (!prop)
 		return -ENOMEM;
 	dev->mode_config.prop_fb_damage_clips = prop;
diff --git a/drivers/gpu/drm/drm_mode_object.c b/drivers/gpu/drm/drm_mode_object.c
index a9005c1c2384..f32507e65b79 100644
--- a/drivers/gpu/drm/drm_mode_object.c
+++ b/drivers/gpu/drm/drm_mode_object.c
@@ -451,6 +451,7 @@ static int set_property_legacy(struct drm_mode_object *obj,
 }
 
 static int set_property_atomic(struct drm_mode_object *obj,
+			       struct drm_file *file_priv,
 			       struct drm_property *prop,
 			       uint64_t prop_value)
 {
@@ -477,7 +478,7 @@ retry:
 						       obj_to_connector(obj),
 						       prop_value);
 	} else {
-		ret = drm_atomic_set_property(state, obj, prop, prop_value);
+		ret = drm_atomic_set_property(state, file_priv, obj, prop, prop_value);
 		if (ret)
 			goto out;
 		ret = drm_atomic_commit(state);
@@ -520,7 +521,7 @@ int drm_mode_obj_set_property_ioctl(struct drm_device *dev, void *data,
 		goto out_unref;
 
 	if (drm_drv_uses_atomic_modeset(property->dev))
-		ret = set_property_atomic(arg_obj, property, arg->value);
+		ret = set_property_atomic(arg_obj, file_priv, property, arg->value);
 	else
 		ret = set_property_legacy(arg_obj, property, arg->value);
 
diff --git a/drivers/gpu/drm/drm_modes.c b/drivers/gpu/drm/drm_modes.c
index 869ac6f4671e..56f92a0bba62 100644
--- a/drivers/gpu/drm/drm_modes.c
+++ b/drivers/gpu/drm/drm_modes.c
@@ -655,22 +655,22 @@ EXPORT_SYMBOL_GPL(drm_display_mode_to_videomode);
  * @bus_flags: information about pixelclk, sync and DE polarity will be stored
  * here
  *
- * Sets DRM_BUS_FLAG_DE_(LOW|HIGH),  DRM_BUS_FLAG_PIXDATA_(POS|NEG)EDGE and
- * DISPLAY_FLAGS_SYNC_(POS|NEG)EDGE in @bus_flags according to DISPLAY_FLAGS
+ * Sets DRM_BUS_FLAG_DE_(LOW|HIGH),  DRM_BUS_FLAG_PIXDATA_DRIVE_(POS|NEG)EDGE
+ * and DISPLAY_FLAGS_SYNC_(POS|NEG)EDGE in @bus_flags according to DISPLAY_FLAGS
  * found in @vm
  */
 void drm_bus_flags_from_videomode(const struct videomode *vm, u32 *bus_flags)
 {
 	*bus_flags = 0;
 	if (vm->flags & DISPLAY_FLAGS_PIXDATA_POSEDGE)
-		*bus_flags |= DRM_BUS_FLAG_PIXDATA_POSEDGE;
+		*bus_flags |= DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE;
 	if (vm->flags & DISPLAY_FLAGS_PIXDATA_NEGEDGE)
-		*bus_flags |= DRM_BUS_FLAG_PIXDATA_NEGEDGE;
+		*bus_flags |= DRM_BUS_FLAG_PIXDATA_DRIVE_NEGEDGE;
 
 	if (vm->flags & DISPLAY_FLAGS_SYNC_POSEDGE)
-		*bus_flags |= DRM_BUS_FLAG_SYNC_POSEDGE;
+		*bus_flags |= DRM_BUS_FLAG_SYNC_DRIVE_POSEDGE;
 	if (vm->flags & DISPLAY_FLAGS_SYNC_NEGEDGE)
-		*bus_flags |= DRM_BUS_FLAG_SYNC_NEGEDGE;
+		*bus_flags |= DRM_BUS_FLAG_SYNC_DRIVE_NEGEDGE;
 
 	if (vm->flags & DISPLAY_FLAGS_DE_LOW)
 		*bus_flags |= DRM_BUS_FLAG_DE_LOW;
diff --git a/drivers/gpu/drm/drm_panel_orientation_quirks.c b/drivers/gpu/drm/drm_panel_orientation_quirks.c
index 52e445bb1aa5..521aff99b08a 100644
--- a/drivers/gpu/drm/drm_panel_orientation_quirks.c
+++ b/drivers/gpu/drm/drm_panel_orientation_quirks.c
@@ -80,6 +80,12 @@ static const struct drm_dmi_panel_orientation_data lcd800x1280_rightside_up = {
 	.orientation = DRM_MODE_PANEL_ORIENTATION_RIGHT_UP,
 };
 
+static const struct drm_dmi_panel_orientation_data lcd1200x1920_rightside_up = {
+	.width = 1200,
+	.height = 1920,
+	.orientation = DRM_MODE_PANEL_ORIENTATION_RIGHT_UP,
+};
+
 static const struct dmi_system_id orientation_data[] = {
 	{	/* Acer One 10 (S1003) */
 		.matches = {
@@ -148,6 +154,13 @@ static const struct dmi_system_id orientation_data[] = {
 		  DMI_EXACT_MATCH(DMI_PRODUCT_VERSION, "Lenovo MIIX 320-10ICR"),
 		},
 		.driver_data = (void *)&lcd800x1280_rightside_up,
+	}, {	/* Lenovo Ideapad D330 */
+		.matches = {
+		  DMI_EXACT_MATCH(DMI_SYS_VENDOR, "LENOVO"),
+		  DMI_EXACT_MATCH(DMI_PRODUCT_NAME, "81H3"),
+		  DMI_EXACT_MATCH(DMI_PRODUCT_VERSION, "Lenovo ideapad D330-10IGM"),
+		},
+		.driver_data = (void *)&lcd1200x1920_rightside_up,
 	}, {	/* VIOS LTH17 */
 		.matches = {
 		  DMI_EXACT_MATCH(DMI_SYS_VENDOR, "VIOS"),
diff --git a/drivers/gpu/drm/drm_plane.c b/drivers/gpu/drm/drm_plane.c
index 4cfb56893b7f..d6ad60ab0d38 100644
--- a/drivers/gpu/drm/drm_plane.c
+++ b/drivers/gpu/drm/drm_plane.c
@@ -960,6 +960,11 @@ retry:
 		if (ret)
 			goto out;
 
+		if (!drm_lease_held(file_priv, crtc->cursor->base.id)) {
+			ret = -EACCES;
+			goto out;
+		}
+
 		ret = drm_mode_cursor_universal(crtc, req, file_priv, &ctx);
 		goto out;
 	}
@@ -1062,6 +1067,9 @@ int drm_mode_page_flip_ioctl(struct drm_device *dev,
 
 	plane = crtc->primary;
 
+	if (!drm_lease_held(file_priv, plane->base.id))
+		return -EACCES;
+
 	if (crtc->funcs->page_flip_target) {
 		u32 current_vblank;
 		int r;
diff --git a/drivers/gpu/drm/drm_prime.c b/drivers/gpu/drm/drm_prime.c
index 231e3f6d5f41..dc079efb3b0f 100644
--- a/drivers/gpu/drm/drm_prime.c
+++ b/drivers/gpu/drm/drm_prime.c
@@ -504,6 +504,7 @@ struct dma_buf *drm_gem_prime_export(struct drm_device *dev,
 		.size = obj->size,
 		.flags = flags,
 		.priv = obj,
+		.resv = obj->resv,
 	};
 
 	if (dev->driver->gem_prime_res_obj)
diff --git a/drivers/gpu/drm/drm_print.c b/drivers/gpu/drm/drm_print.c
index 0e7fc3e7dfb4..f5cb0aabfe35 100644
--- a/drivers/gpu/drm/drm_print.c
+++ b/drivers/gpu/drm/drm_print.c
@@ -253,3 +253,31 @@ void drm_err(const char *format, ...)
 	va_end(args);
 }
 EXPORT_SYMBOL(drm_err);
+
+/**
+ * drm_print_regset32 - print the contents of registers to a
+ * &drm_printer stream.
+ *
+ * @p: the &drm printer
+ * @regset: the list of registers to print.
+ *
+ * Often in driver debug, it's useful to be able to either capture the
+ * contents of registers in the steady state using debugfs or at
+ * specific points during operation.  This lets the driver have a
+ * single list of registers for both.
+ */
+void drm_print_regset32(struct drm_printer *p, struct debugfs_regset32 *regset)
+{
+	int namelen = 0;
+	int i;
+
+	for (i = 0; i < regset->nregs; i++)
+		namelen = max(namelen, (int)strlen(regset->regs[i].name));
+
+	for (i = 0; i < regset->nregs; i++) {
+		drm_printf(p, "%*s = 0x%08x\n",
+			   namelen, regset->regs[i].name,
+			   readl(regset->base + regset->regs[i].offset));
+	}
+}
+EXPORT_SYMBOL(drm_print_regset32);
diff --git a/drivers/gpu/drm/drm_syncobj.c b/drivers/gpu/drm/drm_syncobj.c
index 8bdb4a3bd7bf..3d400905100b 100644
--- a/drivers/gpu/drm/drm_syncobj.c
+++ b/drivers/gpu/drm/drm_syncobj.c
@@ -61,6 +61,7 @@ struct syncobj_wait_entry {
 	struct task_struct *task;
 	struct dma_fence *fence;
 	struct dma_fence_cb fence_cb;
+	u64    point;
 };
 
 static void syncobj_wait_syncobj_func(struct drm_syncobj *syncobj,
@@ -95,6 +96,8 @@ EXPORT_SYMBOL(drm_syncobj_find);
 static void drm_syncobj_fence_add_wait(struct drm_syncobj *syncobj,
 				       struct syncobj_wait_entry *wait)
 {
+	struct dma_fence *fence;
+
 	if (wait->fence)
 		return;
 
@@ -103,11 +106,15 @@ static void drm_syncobj_fence_add_wait(struct drm_syncobj *syncobj,
 	 * have the lock, try one more time just to be sure we don't add a
 	 * callback when a fence has already been set.
 	 */
-	if (syncobj->fence)
-		wait->fence = dma_fence_get(
-			rcu_dereference_protected(syncobj->fence, 1));
-	else
+	fence = dma_fence_get(rcu_dereference_protected(syncobj->fence, 1));
+	if (!fence || dma_fence_chain_find_seqno(&fence, wait->point)) {
+		dma_fence_put(fence);
 		list_add_tail(&wait->node, &syncobj->cb_list);
+	} else if (!fence) {
+		wait->fence = dma_fence_get_stub();
+	} else {
+		wait->fence = fence;
+	}
 	spin_unlock(&syncobj->lock);
 }
 
@@ -123,6 +130,44 @@ static void drm_syncobj_remove_wait(struct drm_syncobj *syncobj,
 }
 
 /**
+ * drm_syncobj_add_point - add new timeline point to the syncobj
+ * @syncobj: sync object to add timeline point do
+ * @chain: chain node to use to add the point
+ * @fence: fence to encapsulate in the chain node
+ * @point: sequence number to use for the point
+ *
+ * Add the chain node as new timeline point to the syncobj.
+ */
+void drm_syncobj_add_point(struct drm_syncobj *syncobj,
+			   struct dma_fence_chain *chain,
+			   struct dma_fence *fence,
+			   uint64_t point)
+{
+	struct syncobj_wait_entry *cur, *tmp;
+	struct dma_fence *prev;
+
+	dma_fence_get(fence);
+
+	spin_lock(&syncobj->lock);
+
+	prev = drm_syncobj_fence_get(syncobj);
+	/* You are adding an unorder point to timeline, which could cause payload returned from query_ioctl is 0! */
+	if (prev && prev->seqno >= point)
+		DRM_ERROR("You are adding an unorder point to timeline!\n");
+	dma_fence_chain_init(chain, prev, fence, point);
+	rcu_assign_pointer(syncobj->fence, &chain->base);
+
+	list_for_each_entry_safe(cur, tmp, &syncobj->cb_list, node)
+		syncobj_wait_syncobj_func(syncobj, cur);
+	spin_unlock(&syncobj->lock);
+
+	/* Walk the chain once to trigger garbage collection */
+	dma_fence_chain_for_each(fence, prev);
+	dma_fence_put(prev);
+}
+EXPORT_SYMBOL(drm_syncobj_add_point);
+
+/**
  * drm_syncobj_replace_fence - replace fence in a sync object.
  * @syncobj: Sync object to replace fence in
  * @fence: fence to install in sync file.
@@ -145,10 +190,8 @@ void drm_syncobj_replace_fence(struct drm_syncobj *syncobj,
 	rcu_assign_pointer(syncobj->fence, fence);
 
 	if (fence != old_fence) {
-		list_for_each_entry_safe(cur, tmp, &syncobj->cb_list, node) {
-			list_del_init(&cur->node);
+		list_for_each_entry_safe(cur, tmp, &syncobj->cb_list, node)
 			syncobj_wait_syncobj_func(syncobj, cur);
-		}
 	}
 
 	spin_unlock(&syncobj->lock);
@@ -171,6 +214,8 @@ static void drm_syncobj_assign_null_handle(struct drm_syncobj *syncobj)
 	dma_fence_put(fence);
 }
 
+/* 5s default for wait submission */
+#define DRM_SYNCOBJ_WAIT_FOR_SUBMIT_TIMEOUT 5000000000ULL
 /**
  * drm_syncobj_find_fence - lookup and reference the fence in a sync object
  * @file_private: drm file private pointer
@@ -191,16 +236,58 @@ int drm_syncobj_find_fence(struct drm_file *file_private,
 			   struct dma_fence **fence)
 {
 	struct drm_syncobj *syncobj = drm_syncobj_find(file_private, handle);
-	int ret = 0;
+	struct syncobj_wait_entry wait;
+	u64 timeout = nsecs_to_jiffies64(DRM_SYNCOBJ_WAIT_FOR_SUBMIT_TIMEOUT);
+	int ret;
 
 	if (!syncobj)
 		return -ENOENT;
 
 	*fence = drm_syncobj_fence_get(syncobj);
-	if (!*fence) {
+	drm_syncobj_put(syncobj);
+
+	if (*fence) {
+		ret = dma_fence_chain_find_seqno(fence, point);
+		if (!ret)
+			return 0;
+		dma_fence_put(*fence);
+	} else {
 		ret = -EINVAL;
 	}
-	drm_syncobj_put(syncobj);
+
+	if (!(flags & DRM_SYNCOBJ_WAIT_FLAGS_WAIT_FOR_SUBMIT))
+		return ret;
+
+	memset(&wait, 0, sizeof(wait));
+	wait.task = current;
+	wait.point = point;
+	drm_syncobj_fence_add_wait(syncobj, &wait);
+
+	do {
+		set_current_state(TASK_INTERRUPTIBLE);
+		if (wait.fence) {
+			ret = 0;
+			break;
+		}
+                if (timeout == 0) {
+                        ret = -ETIME;
+                        break;
+                }
+
+		if (signal_pending(current)) {
+			ret = -ERESTARTSYS;
+			break;
+		}
+
+                timeout = schedule_timeout(timeout);
+	} while (1);
+
+	__set_current_state(TASK_RUNNING);
+	*fence = wait.fence;
+
+	if (wait.node.next)
+		drm_syncobj_remove_wait(syncobj, &wait);
+
 	return ret;
 }
 EXPORT_SYMBOL(drm_syncobj_find_fence);
@@ -592,6 +679,80 @@ drm_syncobj_fd_to_handle_ioctl(struct drm_device *dev, void *data,
 					&args->handle);
 }
 
+static int drm_syncobj_transfer_to_timeline(struct drm_file *file_private,
+					    struct drm_syncobj_transfer *args)
+{
+	struct drm_syncobj *timeline_syncobj = NULL;
+	struct dma_fence *fence;
+	struct dma_fence_chain *chain;
+	int ret;
+
+	timeline_syncobj = drm_syncobj_find(file_private, args->dst_handle);
+	if (!timeline_syncobj) {
+		return -ENOENT;
+	}
+	ret = drm_syncobj_find_fence(file_private, args->src_handle,
+				     args->src_point, args->flags,
+				     &fence);
+	if (ret)
+		goto err;
+	chain = kzalloc(sizeof(struct dma_fence_chain), GFP_KERNEL);
+	if (!chain) {
+		ret = -ENOMEM;
+		goto err1;
+	}
+	drm_syncobj_add_point(timeline_syncobj, chain, fence, args->dst_point);
+err1:
+	dma_fence_put(fence);
+err:
+	drm_syncobj_put(timeline_syncobj);
+
+	return ret;
+}
+
+static int
+drm_syncobj_transfer_to_binary(struct drm_file *file_private,
+			       struct drm_syncobj_transfer *args)
+{
+	struct drm_syncobj *binary_syncobj = NULL;
+	struct dma_fence *fence;
+	int ret;
+
+	binary_syncobj = drm_syncobj_find(file_private, args->dst_handle);
+	if (!binary_syncobj)
+		return -ENOENT;
+	ret = drm_syncobj_find_fence(file_private, args->src_handle,
+				     args->src_point, args->flags, &fence);
+	if (ret)
+		goto err;
+	drm_syncobj_replace_fence(binary_syncobj, fence);
+	dma_fence_put(fence);
+err:
+	drm_syncobj_put(binary_syncobj);
+
+	return ret;
+}
+int
+drm_syncobj_transfer_ioctl(struct drm_device *dev, void *data,
+			   struct drm_file *file_private)
+{
+	struct drm_syncobj_transfer *args = data;
+	int ret;
+
+	if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ_TIMELINE))
+		return -EOPNOTSUPP;
+
+	if (args->pad)
+		return -EINVAL;
+
+	if (args->dst_point)
+		ret = drm_syncobj_transfer_to_timeline(file_private, args);
+	else
+		ret = drm_syncobj_transfer_to_binary(file_private, args);
+
+	return ret;
+}
+
 static void syncobj_wait_fence_func(struct dma_fence *fence,
 				    struct dma_fence_cb *cb)
 {
@@ -604,13 +765,27 @@ static void syncobj_wait_fence_func(struct dma_fence *fence,
 static void syncobj_wait_syncobj_func(struct drm_syncobj *syncobj,
 				      struct syncobj_wait_entry *wait)
 {
+	struct dma_fence *fence;
+
 	/* This happens inside the syncobj lock */
-	wait->fence = dma_fence_get(rcu_dereference_protected(syncobj->fence,
-							      lockdep_is_held(&syncobj->lock)));
+	fence = rcu_dereference_protected(syncobj->fence,
+					  lockdep_is_held(&syncobj->lock));
+	dma_fence_get(fence);
+	if (!fence || dma_fence_chain_find_seqno(&fence, wait->point)) {
+		dma_fence_put(fence);
+		return;
+	} else if (!fence) {
+		wait->fence = dma_fence_get_stub();
+	} else {
+		wait->fence = fence;
+	}
+
 	wake_up_process(wait->task);
+	list_del_init(&wait->node);
 }
 
 static signed long drm_syncobj_array_wait_timeout(struct drm_syncobj **syncobjs,
+						  void __user *user_points,
 						  uint32_t count,
 						  uint32_t flags,
 						  signed long timeout,
@@ -618,12 +793,27 @@ static signed long drm_syncobj_array_wait_timeout(struct drm_syncobj **syncobjs,
 {
 	struct syncobj_wait_entry *entries;
 	struct dma_fence *fence;
+	uint64_t *points;
 	uint32_t signaled_count, i;
 
-	entries = kcalloc(count, sizeof(*entries), GFP_KERNEL);
-	if (!entries)
+	points = kmalloc_array(count, sizeof(*points), GFP_KERNEL);
+	if (points == NULL)
 		return -ENOMEM;
 
+	if (!user_points) {
+		memset(points, 0, count * sizeof(uint64_t));
+
+	} else if (copy_from_user(points, user_points,
+				  sizeof(uint64_t) * count)) {
+		timeout = -EFAULT;
+		goto err_free_points;
+	}
+
+	entries = kcalloc(count, sizeof(*entries), GFP_KERNEL);
+	if (!entries) {
+		timeout = -ENOMEM;
+		goto err_free_points;
+	}
 	/* Walk the list of sync objects and initialize entries.  We do
 	 * this up-front so that we can properly return -EINVAL if there is
 	 * a syncobj with a missing fence and then never have the chance of
@@ -631,9 +821,13 @@ static signed long drm_syncobj_array_wait_timeout(struct drm_syncobj **syncobjs,
 	 */
 	signaled_count = 0;
 	for (i = 0; i < count; ++i) {
+		struct dma_fence *fence;
+
 		entries[i].task = current;
-		entries[i].fence = drm_syncobj_fence_get(syncobjs[i]);
-		if (!entries[i].fence) {
+		entries[i].point = points[i];
+		fence = drm_syncobj_fence_get(syncobjs[i]);
+		if (!fence || dma_fence_chain_find_seqno(&fence, points[i])) {
+			dma_fence_put(fence);
 			if (flags & DRM_SYNCOBJ_WAIT_FLAGS_WAIT_FOR_SUBMIT) {
 				continue;
 			} else {
@@ -642,7 +836,13 @@ static signed long drm_syncobj_array_wait_timeout(struct drm_syncobj **syncobjs,
 			}
 		}
 
-		if (dma_fence_is_signaled(entries[i].fence)) {
+		if (fence)
+			entries[i].fence = fence;
+		else
+			entries[i].fence = dma_fence_get_stub();
+
+		if ((flags & DRM_SYNCOBJ_WAIT_FLAGS_WAIT_AVAILABLE) ||
+		    dma_fence_is_signaled(entries[i].fence)) {
 			if (signaled_count == 0 && idx)
 				*idx = i;
 			signaled_count++;
@@ -675,7 +875,8 @@ static signed long drm_syncobj_array_wait_timeout(struct drm_syncobj **syncobjs,
 			if (!fence)
 				continue;
 
-			if (dma_fence_is_signaled(fence) ||
+			if ((flags & DRM_SYNCOBJ_WAIT_FLAGS_WAIT_AVAILABLE) ||
+			    dma_fence_is_signaled(fence) ||
 			    (!entries[i].fence_cb.func &&
 			     dma_fence_add_callback(fence,
 						    &entries[i].fence_cb,
@@ -720,6 +921,9 @@ cleanup_entries:
 	}
 	kfree(entries);
 
+err_free_points:
+	kfree(points);
+
 	return timeout;
 }
 
@@ -730,7 +934,7 @@ cleanup_entries:
  *
  * Calculate the timeout in jiffies from an absolute time in sec/nsec.
  */
-static signed long drm_timeout_abs_to_jiffies(int64_t timeout_nsec)
+signed long drm_timeout_abs_to_jiffies(int64_t timeout_nsec)
 {
 	ktime_t abs_timeout, now;
 	u64 timeout_ns, timeout_jiffies64;
@@ -754,23 +958,38 @@ static signed long drm_timeout_abs_to_jiffies(int64_t timeout_nsec)
 
 	return timeout_jiffies64 + 1;
 }
+EXPORT_SYMBOL(drm_timeout_abs_to_jiffies);
 
 static int drm_syncobj_array_wait(struct drm_device *dev,
 				  struct drm_file *file_private,
 				  struct drm_syncobj_wait *wait,
-				  struct drm_syncobj **syncobjs)
+				  struct drm_syncobj_timeline_wait *timeline_wait,
+				  struct drm_syncobj **syncobjs, bool timeline)
 {
-	signed long timeout = drm_timeout_abs_to_jiffies(wait->timeout_nsec);
+	signed long timeout = 0;
 	uint32_t first = ~0;
 
-	timeout = drm_syncobj_array_wait_timeout(syncobjs,
-						 wait->count_handles,
-						 wait->flags,
-						 timeout, &first);
-	if (timeout < 0)
-		return timeout;
-
-	wait->first_signaled = first;
+	if (!timeline) {
+		timeout = drm_timeout_abs_to_jiffies(wait->timeout_nsec);
+		timeout = drm_syncobj_array_wait_timeout(syncobjs,
+							 NULL,
+							 wait->count_handles,
+							 wait->flags,
+							 timeout, &first);
+		if (timeout < 0)
+			return timeout;
+		wait->first_signaled = first;
+	} else {
+		timeout = drm_timeout_abs_to_jiffies(timeline_wait->timeout_nsec);
+		timeout = drm_syncobj_array_wait_timeout(syncobjs,
+							 u64_to_user_ptr(timeline_wait->points),
+							 timeline_wait->count_handles,
+							 timeline_wait->flags,
+							 timeout, &first);
+		if (timeout < 0)
+			return timeout;
+		timeline_wait->first_signaled = first;
+	}
 	return 0;
 }
 
@@ -856,13 +1075,48 @@ drm_syncobj_wait_ioctl(struct drm_device *dev, void *data,
 		return ret;
 
 	ret = drm_syncobj_array_wait(dev, file_private,
-				     args, syncobjs);
+				     args, NULL, syncobjs, false);
+
+	drm_syncobj_array_free(syncobjs, args->count_handles);
+
+	return ret;
+}
+
+int
+drm_syncobj_timeline_wait_ioctl(struct drm_device *dev, void *data,
+				struct drm_file *file_private)
+{
+	struct drm_syncobj_timeline_wait *args = data;
+	struct drm_syncobj **syncobjs;
+	int ret = 0;
+
+	if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ_TIMELINE))
+		return -EOPNOTSUPP;
+
+	if (args->flags & ~(DRM_SYNCOBJ_WAIT_FLAGS_WAIT_ALL |
+			    DRM_SYNCOBJ_WAIT_FLAGS_WAIT_FOR_SUBMIT |
+			    DRM_SYNCOBJ_WAIT_FLAGS_WAIT_AVAILABLE))
+		return -EINVAL;
+
+	if (args->count_handles == 0)
+		return -EINVAL;
+
+	ret = drm_syncobj_array_find(file_private,
+				     u64_to_user_ptr(args->handles),
+				     args->count_handles,
+				     &syncobjs);
+	if (ret < 0)
+		return ret;
+
+	ret = drm_syncobj_array_wait(dev, file_private,
+				     NULL, args, syncobjs, true);
 
 	drm_syncobj_array_free(syncobjs, args->count_handles);
 
 	return ret;
 }
 
+
 int
 drm_syncobj_reset_ioctl(struct drm_device *dev, void *data,
 			struct drm_file *file_private)
@@ -928,3 +1182,138 @@ drm_syncobj_signal_ioctl(struct drm_device *dev, void *data,
 
 	return ret;
 }
+
+int
+drm_syncobj_timeline_signal_ioctl(struct drm_device *dev, void *data,
+				  struct drm_file *file_private)
+{
+	struct drm_syncobj_timeline_array *args = data;
+	struct drm_syncobj **syncobjs;
+	struct dma_fence_chain **chains;
+	uint64_t *points;
+	uint32_t i, j;
+	int ret;
+
+	if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ_TIMELINE))
+		return -EOPNOTSUPP;
+
+	if (args->pad != 0)
+		return -EINVAL;
+
+	if (args->count_handles == 0)
+		return -EINVAL;
+
+	ret = drm_syncobj_array_find(file_private,
+				     u64_to_user_ptr(args->handles),
+				     args->count_handles,
+				     &syncobjs);
+	if (ret < 0)
+		return ret;
+
+	points = kmalloc_array(args->count_handles, sizeof(*points),
+			       GFP_KERNEL);
+	if (!points) {
+		ret = -ENOMEM;
+		goto out;
+	}
+	if (!u64_to_user_ptr(args->points)) {
+		memset(points, 0, args->count_handles * sizeof(uint64_t));
+	} else if (copy_from_user(points, u64_to_user_ptr(args->points),
+				  sizeof(uint64_t) * args->count_handles)) {
+		ret = -EFAULT;
+		goto err_points;
+	}
+
+	chains = kmalloc_array(args->count_handles, sizeof(void *), GFP_KERNEL);
+	if (!chains) {
+		ret = -ENOMEM;
+		goto err_points;
+	}
+	for (i = 0; i < args->count_handles; i++) {
+		chains[i] = kzalloc(sizeof(struct dma_fence_chain), GFP_KERNEL);
+		if (!chains[i]) {
+			for (j = 0; j < i; j++)
+				kfree(chains[j]);
+			ret = -ENOMEM;
+			goto err_chains;
+		}
+	}
+
+	for (i = 0; i < args->count_handles; i++) {
+		struct dma_fence *fence = dma_fence_get_stub();
+
+		drm_syncobj_add_point(syncobjs[i], chains[i],
+				      fence, points[i]);
+		dma_fence_put(fence);
+	}
+err_chains:
+	kfree(chains);
+err_points:
+	kfree(points);
+out:
+	drm_syncobj_array_free(syncobjs, args->count_handles);
+
+	return ret;
+}
+
+int drm_syncobj_query_ioctl(struct drm_device *dev, void *data,
+			    struct drm_file *file_private)
+{
+	struct drm_syncobj_timeline_array *args = data;
+	struct drm_syncobj **syncobjs;
+	uint64_t __user *points = u64_to_user_ptr(args->points);
+	uint32_t i;
+	int ret;
+
+	if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ_TIMELINE))
+		return -EOPNOTSUPP;
+
+	if (args->pad != 0)
+		return -EINVAL;
+
+	if (args->count_handles == 0)
+		return -EINVAL;
+
+	ret = drm_syncobj_array_find(file_private,
+				     u64_to_user_ptr(args->handles),
+				     args->count_handles,
+				     &syncobjs);
+	if (ret < 0)
+		return ret;
+
+	for (i = 0; i < args->count_handles; i++) {
+		struct dma_fence_chain *chain;
+		struct dma_fence *fence;
+		uint64_t point;
+
+		fence = drm_syncobj_fence_get(syncobjs[i]);
+		chain = to_dma_fence_chain(fence);
+		if (chain) {
+			struct dma_fence *iter, *last_signaled = NULL;
+
+			dma_fence_chain_for_each(iter, fence) {
+				if (!iter)
+					break;
+				dma_fence_put(last_signaled);
+				last_signaled = dma_fence_get(iter);
+				if (!to_dma_fence_chain(last_signaled)->prev_seqno)
+					/* It is most likely that timeline has
+					 * unorder points. */
+					break;
+			}
+			point = dma_fence_is_signaled(last_signaled) ?
+				last_signaled->seqno :
+				to_dma_fence_chain(last_signaled)->prev_seqno;
+			dma_fence_put(last_signaled);
+		} else {
+			point = 0;
+		}
+		ret = copy_to_user(&points[i], &point, sizeof(uint64_t));
+		ret = ret ? -EFAULT : 0;
+		if (ret)
+			break;
+	}
+	drm_syncobj_array_free(syncobjs, args->count_handles);
+
+	return ret;
+}
diff --git a/drivers/gpu/drm/drm_vm.c b/drivers/gpu/drm/drm_vm.c
index c3301046dfaa..10cf83d569e1 100644
--- a/drivers/gpu/drm/drm_vm.c
+++ b/drivers/gpu/drm/drm_vm.c
@@ -584,8 +584,8 @@ static int drm_mmap_locked(struct file *filp, struct vm_area_struct *vma)
 			vma->vm_ops = &drm_vm_ops;
 			break;
 		}
-		/* fall through to _DRM_FRAME_BUFFER... */
 #endif
+		/* fall through - to _DRM_FRAME_BUFFER... */
 	case _DRM_FRAME_BUFFER:
 	case _DRM_REGISTERS:
 		offset = drm_core_get_reg_ofs(dev);
@@ -610,7 +610,7 @@ static int drm_mmap_locked(struct file *filp, struct vm_area_struct *vma)
 		    vma->vm_end - vma->vm_start, vma->vm_page_prot))
 			return -EAGAIN;
 		vma->vm_page_prot = drm_dma_prot(map->type, vma);
-	/* fall through to _DRM_SHM */
+		/* fall through - to _DRM_SHM */
 	case _DRM_SHM:
 		vma->vm_ops = &drm_vm_shm_ops;
 		vma->vm_private_data = (void *)map;
@@ -646,6 +646,7 @@ int drm_legacy_mmap(struct file *filp, struct vm_area_struct *vma)
 }
 EXPORT_SYMBOL(drm_legacy_mmap);
 
+#if IS_ENABLED(CONFIG_DRM_LEGACY)
 void drm_legacy_vma_flush(struct drm_device *dev)
 {
 	struct drm_vma_entry *vma, *vma_temp;
@@ -656,3 +657,4 @@ void drm_legacy_vma_flush(struct drm_device *dev)
 		kfree(vma);
 	}
 }
+#endif
diff --git a/drivers/gpu/drm/drm_writeback.c b/drivers/gpu/drm/drm_writeback.c
index c20e6fe00cb3..79ac014701c8 100644
--- a/drivers/gpu/drm/drm_writeback.c
+++ b/drivers/gpu/drm/drm_writeback.c
@@ -239,14 +239,52 @@ fail:
 }
 EXPORT_SYMBOL(drm_writeback_connector_init);
 
+int drm_writeback_set_fb(struct drm_connector_state *conn_state,
+			 struct drm_framebuffer *fb)
+{
+	WARN_ON(conn_state->connector->connector_type != DRM_MODE_CONNECTOR_WRITEBACK);
+
+	if (!conn_state->writeback_job) {
+		conn_state->writeback_job =
+			kzalloc(sizeof(*conn_state->writeback_job), GFP_KERNEL);
+		if (!conn_state->writeback_job)
+			return -ENOMEM;
+
+		conn_state->writeback_job->connector =
+			drm_connector_to_writeback(conn_state->connector);
+	}
+
+	drm_framebuffer_assign(&conn_state->writeback_job->fb, fb);
+	return 0;
+}
+
+int drm_writeback_prepare_job(struct drm_writeback_job *job)
+{
+	struct drm_writeback_connector *connector = job->connector;
+	const struct drm_connector_helper_funcs *funcs =
+		connector->base.helper_private;
+	int ret;
+
+	if (funcs->prepare_writeback_job) {
+		ret = funcs->prepare_writeback_job(connector, job);
+		if (ret < 0)
+			return ret;
+	}
+
+	job->prepared = true;
+	return 0;
+}
+EXPORT_SYMBOL(drm_writeback_prepare_job);
+
 /**
  * drm_writeback_queue_job - Queue a writeback job for later signalling
  * @wb_connector: The writeback connector to queue a job on
- * @job: The job to queue
+ * @conn_state: The connector state containing the job to queue
  *
- * This function adds a job to the job_queue for a writeback connector. It
- * should be considered to take ownership of the writeback job, and so any other
- * references to the job must be cleared after calling this function.
+ * This function adds the job contained in @conn_state to the job_queue for a
+ * writeback connector. It takes ownership of the writeback job and sets the
+ * @conn_state->writeback_job to NULL, and so no access to the job may be
+ * performed by the caller after this function returns.
  *
  * Drivers must ensure that for a given writeback connector, jobs are queued in
  * exactly the same order as they will be completed by the hardware (and
@@ -258,16 +296,36 @@ EXPORT_SYMBOL(drm_writeback_connector_init);
  * See also: drm_writeback_signal_completion()
  */
 void drm_writeback_queue_job(struct drm_writeback_connector *wb_connector,
-			     struct drm_writeback_job *job)
+			     struct drm_connector_state *conn_state)
 {
+	struct drm_writeback_job *job;
 	unsigned long flags;
 
+	job = conn_state->writeback_job;
+	conn_state->writeback_job = NULL;
+
 	spin_lock_irqsave(&wb_connector->job_lock, flags);
 	list_add_tail(&job->list_entry, &wb_connector->job_queue);
 	spin_unlock_irqrestore(&wb_connector->job_lock, flags);
 }
 EXPORT_SYMBOL(drm_writeback_queue_job);
 
+void drm_writeback_cleanup_job(struct drm_writeback_job *job)
+{
+	struct drm_writeback_connector *connector = job->connector;
+	const struct drm_connector_helper_funcs *funcs =
+		connector->base.helper_private;
+
+	if (job->prepared && funcs->cleanup_writeback_job)
+		funcs->cleanup_writeback_job(connector, job);
+
+	if (job->fb)
+		drm_framebuffer_put(job->fb);
+
+	kfree(job);
+}
+EXPORT_SYMBOL(drm_writeback_cleanup_job);
+
 /*
  * @cleanup_work: deferred cleanup of a writeback job
  *
@@ -280,10 +338,9 @@ static void cleanup_work(struct work_struct *work)
 	struct drm_writeback_job *job = container_of(work,
 						     struct drm_writeback_job,
 						     cleanup_work);
-	drm_framebuffer_put(job->fb);
-	kfree(job);
-}
 
+	drm_writeback_cleanup_job(job);
+}
 
 /**
  * drm_writeback_signal_completion - Signal the completion of a writeback job
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_drv.c b/drivers/gpu/drm/etnaviv/etnaviv_drv.c
index 18c27f795cf6..7eb7cf9c3fa8 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_drv.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_drv.c
@@ -473,7 +473,6 @@ static struct drm_driver etnaviv_drm_driver = {
 	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
 	.gem_prime_export   = drm_gem_prime_export,
 	.gem_prime_import   = drm_gem_prime_import,
-	.gem_prime_res_obj  = etnaviv_gem_prime_res_obj,
 	.gem_prime_pin      = etnaviv_gem_prime_pin,
 	.gem_prime_unpin    = etnaviv_gem_prime_unpin,
 	.gem_prime_get_sg_table = etnaviv_gem_prime_get_sg_table,
@@ -515,6 +514,9 @@ static int etnaviv_bind(struct device *dev)
 	}
 	drm->dev_private = priv;
 
+	dev->dma_parms = &priv->dma_parms;
+	dma_set_max_seg_size(dev, SZ_2G);
+
 	mutex_init(&priv->gem_lock);
 	INIT_LIST_HEAD(&priv->gem_list);
 	priv->num_gpus = 0;
@@ -552,6 +554,8 @@ static void etnaviv_unbind(struct device *dev)
 
 	component_unbind_all(dev, drm);
 
+	dev->dma_parms = NULL;
+
 	drm->dev_private = NULL;
 	kfree(priv);
 
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_drv.h b/drivers/gpu/drm/etnaviv/etnaviv_drv.h
index a6a7ded37ef1..8798423705e1 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_drv.h
+++ b/drivers/gpu/drm/etnaviv/etnaviv_drv.h
@@ -42,6 +42,7 @@ struct etnaviv_file_private {
 
 struct etnaviv_drm_private {
 	int num_gpus;
+	struct device_dma_parameters dma_parms;
 	struct etnaviv_gpu *gpu[ETNA_MAX_PIPES];
 
 	/* list of GEM objects: */
@@ -60,7 +61,6 @@ void *etnaviv_gem_prime_vmap(struct drm_gem_object *obj);
 void etnaviv_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr);
 int etnaviv_gem_prime_mmap(struct drm_gem_object *obj,
 			   struct vm_area_struct *vma);
-struct reservation_object *etnaviv_gem_prime_res_obj(struct drm_gem_object *obj);
 struct drm_gem_object *etnaviv_gem_prime_import_sg_table(struct drm_device *dev,
 	struct dma_buf_attachment *attach, struct sg_table *sg);
 int etnaviv_gem_prime_pin(struct drm_gem_object *obj);
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem.c b/drivers/gpu/drm/etnaviv/etnaviv_gem.c
index 5c48915f492d..e8778ebb72e6 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_gem.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_gem.c
@@ -397,13 +397,13 @@ int etnaviv_gem_cpu_prep(struct drm_gem_object *obj, u32 op,
 	}
 
 	if (op & ETNA_PREP_NOSYNC) {
-		if (!reservation_object_test_signaled_rcu(etnaviv_obj->resv,
+		if (!reservation_object_test_signaled_rcu(obj->resv,
 							  write))
 			return -EBUSY;
 	} else {
 		unsigned long remain = etnaviv_timeout_to_jiffies(timeout);
 
-		ret = reservation_object_wait_timeout_rcu(etnaviv_obj->resv,
+		ret = reservation_object_wait_timeout_rcu(obj->resv,
 							  write, true, remain);
 		if (ret <= 0)
 			return ret == 0 ? -ETIMEDOUT : ret;
@@ -459,7 +459,7 @@ static void etnaviv_gem_describe_fence(struct dma_fence *fence,
 static void etnaviv_gem_describe(struct drm_gem_object *obj, struct seq_file *m)
 {
 	struct etnaviv_gem_object *etnaviv_obj = to_etnaviv_bo(obj);
-	struct reservation_object *robj = etnaviv_obj->resv;
+	struct reservation_object *robj = obj->resv;
 	struct reservation_object_list *fobj;
 	struct dma_fence *fence;
 	unsigned long off = drm_vma_node_start(&obj->vma_node);
@@ -549,8 +549,6 @@ void etnaviv_gem_free_object(struct drm_gem_object *obj)
 
 	drm_gem_free_mmap_offset(obj);
 	etnaviv_obj->ops->release(etnaviv_obj);
-	if (etnaviv_obj->resv == &etnaviv_obj->_resv)
-		reservation_object_fini(&etnaviv_obj->_resv);
 	drm_gem_object_release(obj);
 
 	kfree(etnaviv_obj);
@@ -596,12 +594,8 @@ static int etnaviv_gem_new_impl(struct drm_device *dev, u32 size, u32 flags,
 
 	etnaviv_obj->flags = flags;
 	etnaviv_obj->ops = ops;
-	if (robj) {
-		etnaviv_obj->resv = robj;
-	} else {
-		etnaviv_obj->resv = &etnaviv_obj->_resv;
-		reservation_object_init(&etnaviv_obj->_resv);
-	}
+	if (robj)
+		etnaviv_obj->base.resv = robj;
 
 	mutex_init(&etnaviv_obj->lock);
 	INIT_LIST_HEAD(&etnaviv_obj->vram_list);
@@ -628,24 +622,18 @@ int etnaviv_gem_new_handle(struct drm_device *dev, struct drm_file *file,
 	lockdep_set_class(&to_etnaviv_bo(obj)->lock, &etnaviv_shm_lock_class);
 
 	ret = drm_gem_object_init(dev, obj, size);
-	if (ret == 0) {
-		struct address_space *mapping;
-
-		/*
-		 * Our buffers are kept pinned, so allocating them
-		 * from the MOVABLE zone is a really bad idea, and
-		 * conflicts with CMA. See comments above new_inode()
-		 * why this is required _and_ expected if you're
-		 * going to pin these pages.
-		 */
-		mapping = obj->filp->f_mapping;
-		mapping_set_gfp_mask(mapping, GFP_HIGHUSER |
-				     __GFP_RETRY_MAYFAIL | __GFP_NOWARN);
-	}
-
 	if (ret)
 		goto fail;
 
+	/*
+	 * Our buffers are kept pinned, so allocating them from the MOVABLE
+	 * zone is a really bad idea, and conflicts with CMA. See comments
+	 * above new_inode() why this is required _and_ expected if you're
+	 * going to pin these pages.
+	 */
+	mapping_set_gfp_mask(obj->filp->f_mapping, GFP_HIGHUSER |
+			     __GFP_RETRY_MAYFAIL | __GFP_NOWARN);
+
 	etnaviv_gem_obj_add(dev, obj);
 
 	ret = drm_gem_handle_create(file, obj, handle);
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem.h b/drivers/gpu/drm/etnaviv/etnaviv_gem.h
index f0abb744ef95..753c458497d0 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_gem.h
+++ b/drivers/gpu/drm/etnaviv/etnaviv_gem.h
@@ -47,10 +47,6 @@ struct etnaviv_gem_object {
 	struct sg_table *sgt;
 	void *vaddr;
 
-	/* normally (resv == &_resv) except for imported bo's */
-	struct reservation_object *resv;
-	struct reservation_object _resv;
-
 	struct list_head vram_list;
 
 	/* cache maintenance */
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c b/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c
index f21529e635e3..00e8b6a817e3 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c
@@ -139,10 +139,3 @@ fail:
 
 	return ERR_PTR(ret);
 }
-
-struct reservation_object *etnaviv_gem_prime_res_obj(struct drm_gem_object *obj)
-{
-	struct etnaviv_gem_object *etnaviv_obj = to_etnaviv_bo(obj);
-
-	return etnaviv_obj->resv;
-}
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c b/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c
index b2fe3446bfbc..e054f09ac828 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c
@@ -108,9 +108,9 @@ out_unlock:
 static void submit_unlock_object(struct etnaviv_gem_submit *submit, int i)
 {
 	if (submit->bos[i].flags & BO_LOCKED) {
-		struct etnaviv_gem_object *etnaviv_obj = submit->bos[i].obj;
+		struct drm_gem_object *obj = &submit->bos[i].obj->base;
 
-		ww_mutex_unlock(&etnaviv_obj->resv->lock);
+		ww_mutex_unlock(&obj->resv->lock);
 		submit->bos[i].flags &= ~BO_LOCKED;
 	}
 }
@@ -122,7 +122,7 @@ static int submit_lock_objects(struct etnaviv_gem_submit *submit,
 
 retry:
 	for (i = 0; i < submit->nr_bos; i++) {
-		struct etnaviv_gem_object *etnaviv_obj = submit->bos[i].obj;
+		struct drm_gem_object *obj = &submit->bos[i].obj->base;
 
 		if (slow_locked == i)
 			slow_locked = -1;
@@ -130,7 +130,7 @@ retry:
 		contended = i;
 
 		if (!(submit->bos[i].flags & BO_LOCKED)) {
-			ret = ww_mutex_lock_interruptible(&etnaviv_obj->resv->lock,
+			ret = ww_mutex_lock_interruptible(&obj->resv->lock,
 							  ticket);
 			if (ret == -EALREADY)
 				DRM_ERROR("BO at index %u already on submit list\n",
@@ -153,12 +153,12 @@ fail:
 		submit_unlock_object(submit, slow_locked);
 
 	if (ret == -EDEADLK) {
-		struct etnaviv_gem_object *etnaviv_obj;
+		struct drm_gem_object *obj;
 
-		etnaviv_obj = submit->bos[contended].obj;
+		obj = &submit->bos[contended].obj->base;
 
 		/* we lost out in a seqno race, lock and retry.. */
-		ret = ww_mutex_lock_slow_interruptible(&etnaviv_obj->resv->lock,
+		ret = ww_mutex_lock_slow_interruptible(&obj->resv->lock,
 						       ticket);
 		if (!ret) {
 			submit->bos[contended].flags |= BO_LOCKED;
@@ -176,7 +176,7 @@ static int submit_fence_sync(struct etnaviv_gem_submit *submit)
 
 	for (i = 0; i < submit->nr_bos; i++) {
 		struct etnaviv_gem_submit_bo *bo = &submit->bos[i];
-		struct reservation_object *robj = bo->obj->resv;
+		struct reservation_object *robj = bo->obj->base.resv;
 
 		if (!(bo->flags & ETNA_SUBMIT_BO_WRITE)) {
 			ret = reservation_object_reserve_shared(robj, 1);
@@ -207,13 +207,13 @@ static void submit_attach_object_fences(struct etnaviv_gem_submit *submit)
 	int i;
 
 	for (i = 0; i < submit->nr_bos; i++) {
-		struct etnaviv_gem_object *etnaviv_obj = submit->bos[i].obj;
+		struct drm_gem_object *obj = &submit->bos[i].obj->base;
 
 		if (submit->bos[i].flags & ETNA_SUBMIT_BO_WRITE)
-			reservation_object_add_excl_fence(etnaviv_obj->resv,
+			reservation_object_add_excl_fence(obj->resv,
 							  submit->out_fence);
 		else
-			reservation_object_add_shared_fence(etnaviv_obj->resv,
+			reservation_object_add_shared_fence(obj->resv,
 							    submit->out_fence);
 
 		submit_unlock_object(submit, i);
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c
index 6904535475de..72d01e873160 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c
@@ -365,6 +365,7 @@ static void etnaviv_hw_identify(struct etnaviv_gpu *gpu)
 	dev_info(gpu->dev, "model: GC%x, revision: %x\n",
 		 gpu->identity.model, gpu->identity.revision);
 
+	gpu->idle_mask = ~VIVS_HI_IDLE_STATE_AXI_LP;
 	/*
 	 * If there is a match in the HWDB, we aren't interested in the
 	 * remaining register values, as they might be wrong.
@@ -412,7 +413,7 @@ static void etnaviv_hw_identify(struct etnaviv_gpu *gpu)
 	}
 
 	/* GC600 idle register reports zero bits where modules aren't present */
-	if (gpu->identity.model == chipModel_GC600) {
+	if (gpu->identity.model == chipModel_GC600)
 		gpu->idle_mask = VIVS_HI_IDLE_STATE_TX |
 				 VIVS_HI_IDLE_STATE_RA |
 				 VIVS_HI_IDLE_STATE_SE |
@@ -421,9 +422,6 @@ static void etnaviv_hw_identify(struct etnaviv_gpu *gpu)
 				 VIVS_HI_IDLE_STATE_PE |
 				 VIVS_HI_IDLE_STATE_DE |
 				 VIVS_HI_IDLE_STATE_FE;
-	} else {
-		gpu->idle_mask = ~VIVS_HI_IDLE_STATE_AXI_LP;
-	}
 
 	etnaviv_hw_specs(gpu);
 }
diff --git a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
index 5b4e0e8b23bc..73b318a7ef49 100644
--- a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
+++ b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
@@ -188,7 +188,7 @@ static void decon_setup_trigger(struct decon_context *ctx)
 
 	if (regmap_update_bits(ctx->sysreg, DSD_CFG_MUX,
 			       DSD_CFG_MUX_TE_UNMASK_GLOBAL, ~0))
-		DRM_ERROR("Cannot update sysreg.\n");
+		DRM_DEV_ERROR(ctx->dev, "Cannot update sysreg.\n");
 }
 
 static void decon_commit(struct exynos_drm_crtc *crtc)
@@ -356,7 +356,7 @@ static void decon_win_set_pixfmt(struct decon_context *ctx, unsigned int win,
 		break;
 	}
 
-	DRM_DEBUG_KMS("cpp = %u\n", fb->format->cpp[0]);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "cpp = %u\n", fb->format->cpp[0]);
 
 	/*
 	 * In case of exynos, setting dma-burst to 16Word causes permanent
@@ -561,8 +561,6 @@ static void decon_clear_channels(struct exynos_drm_crtc *crtc)
 	struct decon_context *ctx = crtc->ctx;
 	int win, i, ret;
 
-	DRM_DEBUG_KMS("%s\n", __FILE__);
-
 	for (i = 0; i < ARRAY_SIZE(decon_clks_name); i++) {
 		ret = clk_prepare_enable(ctx->clks[i]);
 		if (ret < 0)
diff --git a/drivers/gpu/drm/exynos/exynos7_drm_decon.c b/drivers/gpu/drm/exynos/exynos7_drm_decon.c
index 381aa3d60e37..0217ee9a118d 100644
--- a/drivers/gpu/drm/exynos/exynos7_drm_decon.c
+++ b/drivers/gpu/drm/exynos/exynos7_drm_decon.c
@@ -99,7 +99,7 @@ static void decon_wait_for_vblank(struct exynos_drm_crtc *crtc)
 	if (!wait_event_timeout(ctx->wait_vsync_queue,
 				!atomic_read(&ctx->wait_vsync_event),
 				HZ/20))
-		DRM_DEBUG_KMS("vblank wait timed out.\n");
+		DRM_DEV_DEBUG_KMS(ctx->dev, "vblank wait timed out.\n");
 }
 
 static void decon_clear_channels(struct exynos_drm_crtc *crtc)
@@ -107,8 +107,6 @@ static void decon_clear_channels(struct exynos_drm_crtc *crtc)
 	struct decon_context *ctx = crtc->ctx;
 	unsigned int win, ch_enabled = 0;
 
-	DRM_DEBUG_KMS("%s\n", __FILE__);
-
 	/* Check if any channel is enabled. */
 	for (win = 0; win < WINDOWS_NR; win++) {
 		u32 val = readl(ctx->regs + WINCON(win));
@@ -315,7 +313,7 @@ static void decon_win_set_pixfmt(struct decon_context *ctx, unsigned int win,
 		break;
 	}
 
-	DRM_DEBUG_KMS("cpp = %d\n", fb->format->cpp[0]);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "cpp = %d\n", fb->format->cpp[0]);
 
 	/*
 	 * In case of exynos, setting dma-burst to 16Word causes permanent
@@ -422,9 +420,9 @@ static void decon_update_plane(struct exynos_drm_crtc *crtc,
 	writel(state->src.x, ctx->regs + VIDW_OFFSET_X(win));
 	writel(state->src.y, ctx->regs + VIDW_OFFSET_Y(win));
 
-	DRM_DEBUG_KMS("start addr = 0x%lx\n",
+	DRM_DEV_DEBUG_KMS(ctx->dev, "start addr = 0x%lx\n",
 			(unsigned long)val);
-	DRM_DEBUG_KMS("ovl_width = %d, ovl_height = %d\n",
+	DRM_DEV_DEBUG_KMS(ctx->dev, "ovl_width = %d, ovl_height = %d\n",
 			state->crtc.w, state->crtc.h);
 
 	val = VIDOSDxA_TOPLEFT_X(state->crtc.x) |
@@ -442,7 +440,7 @@ static void decon_update_plane(struct exynos_drm_crtc *crtc,
 
 	writel(val, ctx->regs + VIDOSD_B(win));
 
-	DRM_DEBUG_KMS("osd pos: tx = %d, ty = %d, bx = %d, by = %d\n",
+	DRM_DEV_DEBUG_KMS(ctx->dev, "osd pos: tx = %d, ty = %d, bx = %d, by = %d\n",
 			state->crtc.x, state->crtc.y, last_x, last_y);
 
 	/* OSD alpha */
@@ -622,7 +620,7 @@ static int decon_bind(struct device *dev, struct device *master, void *data)
 
 	ret = decon_ctx_initialize(ctx, drm_dev);
 	if (ret) {
-		DRM_ERROR("decon_ctx_initialize failed.\n");
+		DRM_DEV_ERROR(dev, "decon_ctx_initialize failed.\n");
 		return ret;
 	}
 
@@ -802,25 +800,29 @@ static int exynos7_decon_resume(struct device *dev)
 
 	ret = clk_prepare_enable(ctx->pclk);
 	if (ret < 0) {
-		DRM_ERROR("Failed to prepare_enable the pclk [%d]\n", ret);
+		DRM_DEV_ERROR(dev, "Failed to prepare_enable the pclk [%d]\n",
+			      ret);
 		return ret;
 	}
 
 	ret = clk_prepare_enable(ctx->aclk);
 	if (ret < 0) {
-		DRM_ERROR("Failed to prepare_enable the aclk [%d]\n", ret);
+		DRM_DEV_ERROR(dev, "Failed to prepare_enable the aclk [%d]\n",
+			      ret);
 		return ret;
 	}
 
 	ret = clk_prepare_enable(ctx->eclk);
 	if  (ret < 0) {
-		DRM_ERROR("Failed to prepare_enable the eclk [%d]\n", ret);
+		DRM_DEV_ERROR(dev, "Failed to prepare_enable the eclk [%d]\n",
+			      ret);
 		return ret;
 	}
 
 	ret = clk_prepare_enable(ctx->vclk);
 	if  (ret < 0) {
-		DRM_ERROR("Failed to prepare_enable the vclk [%d]\n", ret);
+		DRM_DEV_ERROR(dev, "Failed to prepare_enable the vclk [%d]\n",
+			      ret);
 		return ret;
 	}
 
diff --git a/drivers/gpu/drm/exynos/exynos_dp.c b/drivers/gpu/drm/exynos/exynos_dp.c
index 471242a5e580..b0288cf85701 100644
--- a/drivers/gpu/drm/exynos/exynos_dp.c
+++ b/drivers/gpu/drm/exynos/exynos_dp.c
@@ -83,7 +83,8 @@ static int exynos_dp_get_modes(struct analogix_dp_plat_data *plat_data,
 
 	mode = drm_mode_create(connector->dev);
 	if (!mode) {
-		DRM_ERROR("failed to create a new display mode.\n");
+		DRM_DEV_ERROR(dp->dev,
+			      "failed to create a new display mode.\n");
 		return num_modes;
 	}
 
@@ -111,7 +112,8 @@ static int exynos_dp_bridge_attach(struct analogix_dp_plat_data *plat_data,
 	if (dp->ptn_bridge) {
 		ret = drm_bridge_attach(&dp->encoder, dp->ptn_bridge, bridge);
 		if (ret) {
-			DRM_ERROR("Failed to attach bridge to drm\n");
+			DRM_DEV_ERROR(dp->dev,
+				      "Failed to attach bridge to drm\n");
 			bridge->next = NULL;
 			return ret;
 		}
@@ -147,7 +149,8 @@ static int exynos_dp_dt_parse_panel(struct exynos_dp_device *dp)
 
 	ret = of_get_videomode(dp->dev->of_node, &dp->vm, OF_USE_NATIVE_MODE);
 	if (ret) {
-		DRM_ERROR("failed: of_get_videomode() : %d\n", ret);
+		DRM_DEV_ERROR(dp->dev,
+			      "failed: of_get_videomode() : %d\n", ret);
 		return ret;
 	}
 	return 0;
diff --git a/drivers/gpu/drm/exynos/exynos_drm_dma.c b/drivers/gpu/drm/exynos/exynos_drm_dma.c
index 3432c5ee9f0c..bef8bc3c8e00 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_dma.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_dma.c
@@ -62,7 +62,7 @@ static int drm_iommu_attach_device(struct drm_device *drm_dev,
 	int ret;
 
 	if (get_dma_ops(priv->dma_dev) != get_dma_ops(subdrv_dev)) {
-		DRM_ERROR("Device %s lacks support for IOMMU\n",
+		DRM_DEV_ERROR(subdrv_dev, "Device %s lacks support for IOMMU\n",
 			  dev_name(subdrv_dev));
 		return -EINVAL;
 	}
diff --git a/drivers/gpu/drm/exynos/exynos_drm_dpi.c b/drivers/gpu/drm/exynos/exynos_drm_dpi.c
index ae425c9a3f7b..6ea92173db9f 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_dpi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_dpi.c
@@ -77,7 +77,8 @@ static int exynos_dpi_get_modes(struct drm_connector *connector)
 
 		mode = drm_mode_create(connector->dev);
 		if (!mode) {
-			DRM_ERROR("failed to create a new display mode\n");
+			DRM_DEV_ERROR(ctx->dev,
+				      "failed to create a new display mode\n");
 			return 0;
 		}
 		drm_display_mode_from_videomode(ctx->vm, mode);
@@ -108,7 +109,8 @@ static int exynos_dpi_create_connector(struct drm_encoder *encoder)
 				 &exynos_dpi_connector_funcs,
 				 DRM_MODE_CONNECTOR_VGA);
 	if (ret) {
-		DRM_ERROR("failed to initialize connector with drm\n");
+		DRM_DEV_ERROR(ctx->dev,
+			      "failed to initialize connector with drm\n");
 		return ret;
 	}
 
@@ -213,7 +215,8 @@ int exynos_dpi_bind(struct drm_device *dev, struct drm_encoder *encoder)
 
 	ret = exynos_dpi_create_connector(encoder);
 	if (ret) {
-		DRM_ERROR("failed to create connector ret = %d\n", ret);
+		DRM_DEV_ERROR(encoder_to_dpi(encoder)->dev,
+			      "failed to create connector ret = %d\n", ret);
 		drm_encoder_cleanup(encoder);
 		return ret;
 	}
diff --git a/drivers/gpu/drm/exynos/exynos_drm_dsi.c b/drivers/gpu/drm/exynos/exynos_drm_dsi.c
index a4253dd55f86..63a4b5074a99 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_dsi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_dsi.c
@@ -1483,7 +1483,8 @@ static int exynos_dsi_create_connector(struct drm_encoder *encoder)
 	ret = drm_connector_init(drm, connector, &exynos_dsi_connector_funcs,
 				 DRM_MODE_CONNECTOR_DSI);
 	if (ret) {
-		DRM_ERROR("Failed to initialize connector with drm\n");
+		DRM_DEV_ERROR(dsi->dev,
+			      "Failed to initialize connector with drm\n");
 		return ret;
 	}
 
@@ -1527,7 +1528,9 @@ static int exynos_dsi_host_attach(struct mipi_dsi_host *host,
 		int ret = exynos_dsi_create_connector(encoder);
 
 		if (ret) {
-			DRM_ERROR("failed to create connector ret = %d\n", ret);
+			DRM_DEV_ERROR(dsi->dev,
+				      "failed to create connector ret = %d\n",
+				      ret);
 			drm_encoder_cleanup(encoder);
 			return ret;
 		}
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fb.c b/drivers/gpu/drm/exynos/exynos_drm_fb.c
index 1f11ab0f8e9d..832d22f57b4b 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fb.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fb.c
@@ -45,7 +45,8 @@ static int check_fb_gem_memory_type(struct drm_device *drm_dev,
 	 * supported without IOMMU.
 	 */
 	if (IS_NONCONTIG_BUFFER(flags)) {
-		DRM_ERROR("Non-contiguous GEM memory is not supported.\n");
+		DRM_DEV_ERROR(drm_dev->dev,
+			      "Non-contiguous GEM memory is not supported.\n");
 		return -EINVAL;
 	}
 
@@ -83,7 +84,8 @@ exynos_drm_framebuffer_init(struct drm_device *dev,
 
 	ret = drm_framebuffer_init(dev, fb, &exynos_drm_fb_funcs);
 	if (ret < 0) {
-		DRM_ERROR("failed to initialize framebuffer\n");
+		DRM_DEV_ERROR(dev->dev,
+			      "failed to initialize framebuffer\n");
 		goto err;
 	}
 
@@ -113,7 +115,8 @@ exynos_user_fb_create(struct drm_device *dev, struct drm_file *file_priv,
 		exynos_gem[i] = exynos_drm_gem_get(file_priv,
 						   mode_cmd->handles[i]);
 		if (!exynos_gem[i]) {
-			DRM_ERROR("failed to lookup gem object\n");
+			DRM_DEV_ERROR(dev->dev,
+				      "failed to lookup gem object\n");
 			ret = -ENOENT;
 			goto err;
 		}
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
index c30dd88cdb25..724cb52a374a 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
@@ -55,7 +55,7 @@ static int exynos_drm_fb_mmap(struct fb_info *info,
 			     exynos_gem->dma_addr, exynos_gem->size,
 			     exynos_gem->dma_attrs);
 	if (ret < 0) {
-		DRM_ERROR("failed to mmap.\n");
+		DRM_DEV_ERROR(to_dma_dev(helper->dev), "failed to mmap.\n");
 		return ret;
 	}
 
@@ -83,22 +83,22 @@ static int exynos_drm_fbdev_update(struct drm_fb_helper *helper,
 
 	fbi = drm_fb_helper_alloc_fbi(helper);
 	if (IS_ERR(fbi)) {
-		DRM_ERROR("failed to allocate fb info.\n");
+		DRM_DEV_ERROR(to_dma_dev(helper->dev),
+			      "failed to allocate fb info.\n");
 		return PTR_ERR(fbi);
 	}
 
-	fbi->par = helper;
 	fbi->fbops = &exynos_drm_fb_ops;
 
-	drm_fb_helper_fill_fix(fbi, fb->pitches[0], fb->format->depth);
-	drm_fb_helper_fill_var(fbi, helper, sizes->fb_width, sizes->fb_height);
+	drm_fb_helper_fill_info(fbi, helper, sizes);
 
 	nr_pages = exynos_gem->size >> PAGE_SHIFT;
 
 	exynos_gem->kvaddr = (void __iomem *) vmap(exynos_gem->pages, nr_pages,
 				VM_MAP, pgprot_writecombine(PAGE_KERNEL));
 	if (!exynos_gem->kvaddr) {
-		DRM_ERROR("failed to map pages to kernel space.\n");
+		DRM_DEV_ERROR(to_dma_dev(helper->dev),
+			      "failed to map pages to kernel space.\n");
 		return -EIO;
 	}
 
@@ -122,9 +122,10 @@ static int exynos_drm_fbdev_create(struct drm_fb_helper *helper,
 	unsigned long size;
 	int ret;
 
-	DRM_DEBUG_KMS("surface width(%d), height(%d) and bpp(%d\n",
-			sizes->surface_width, sizes->surface_height,
-			sizes->surface_bpp);
+	DRM_DEV_DEBUG_KMS(dev->dev,
+			  "surface width(%d), height(%d) and bpp(%d\n",
+			  sizes->surface_width, sizes->surface_height,
+			  sizes->surface_bpp);
 
 	mode_cmd.width = sizes->surface_width;
 	mode_cmd.height = sizes->surface_height;
@@ -154,7 +155,7 @@ static int exynos_drm_fbdev_create(struct drm_fb_helper *helper,
 	helper->fb =
 		exynos_drm_framebuffer_init(dev, &mode_cmd, &exynos_gem, 1);
 	if (IS_ERR(helper->fb)) {
-		DRM_ERROR("failed to create drm framebuffer.\n");
+		DRM_DEV_ERROR(dev->dev, "failed to create drm framebuffer.\n");
 		ret = PTR_ERR(helper->fb);
 		goto err_destroy_gem;
 	}
@@ -203,20 +204,23 @@ int exynos_drm_fbdev_init(struct drm_device *dev)
 
 	ret = drm_fb_helper_init(dev, helper, MAX_CONNECTOR);
 	if (ret < 0) {
-		DRM_ERROR("failed to initialize drm fb helper.\n");
+		DRM_DEV_ERROR(dev->dev,
+			      "failed to initialize drm fb helper.\n");
 		goto err_init;
 	}
 
 	ret = drm_fb_helper_single_add_all_connectors(helper);
 	if (ret < 0) {
-		DRM_ERROR("failed to register drm_fb_helper_connector.\n");
+		DRM_DEV_ERROR(dev->dev,
+			      "failed to register drm_fb_helper_connector.\n");
 		goto err_setup;
 
 	}
 
 	ret = drm_fb_helper_initial_config(helper, PREFERRED_BPP);
 	if (ret < 0) {
-		DRM_ERROR("failed to set up hw configuration.\n");
+		DRM_DEV_ERROR(dev->dev,
+			      "failed to set up hw configuration.\n");
 		goto err_setup;
 	}
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimc.c b/drivers/gpu/drm/exynos/exynos_drm_fimc.c
index 90dfea0aec4d..c50b0f9270a4 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fimc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fimc.c
@@ -186,7 +186,7 @@ static void fimc_handle_jpeg(struct fimc_context *ctx, bool enable)
 {
 	u32 cfg;
 
-	DRM_DEBUG_KMS("enable[%d]\n", enable);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "enable[%d]\n", enable);
 
 	cfg = fimc_read(ctx, EXYNOS_CIGCTRL);
 	if (enable)
@@ -201,7 +201,7 @@ static void fimc_mask_irq(struct fimc_context *ctx, bool enable)
 {
 	u32 cfg;
 
-	DRM_DEBUG_KMS("enable[%d]\n", enable);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "enable[%d]\n", enable);
 
 	cfg = fimc_read(ctx, EXYNOS_CIGCTRL);
 	if (enable) {
@@ -225,15 +225,16 @@ static bool fimc_check_ovf(struct fimc_context *ctx)
 	flag = EXYNOS_CISTATUS_OVFIY | EXYNOS_CISTATUS_OVFICB |
 		EXYNOS_CISTATUS_OVFICR;
 
-	DRM_DEBUG_KMS("flag[0x%x]\n", flag);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "flag[0x%x]\n", flag);
 
 	if (status & flag) {
 		fimc_set_bits(ctx, EXYNOS_CIWDOFST,
 			EXYNOS_CIWDOFST_CLROVFIY | EXYNOS_CIWDOFST_CLROVFICB |
 			EXYNOS_CIWDOFST_CLROVFICR);
 
-		dev_err(ctx->dev, "occurred overflow at %d, status 0x%x.\n",
-			ctx->id, status);
+		DRM_DEV_ERROR(ctx->dev,
+			      "occurred overflow at %d, status 0x%x.\n",
+			      ctx->id, status);
 		return true;
 	}
 
@@ -246,7 +247,7 @@ static bool fimc_check_frame_end(struct fimc_context *ctx)
 
 	cfg = fimc_read(ctx, EXYNOS_CISTATUS);
 
-	DRM_DEBUG_KMS("cfg[0x%x]\n", cfg);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "cfg[0x%x]\n", cfg);
 
 	if (!(cfg & EXYNOS_CISTATUS_FRAMEEND))
 		return false;
@@ -268,17 +269,17 @@ static int fimc_get_buf_id(struct fimc_context *ctx)
 	if (frame_cnt == 0)
 		frame_cnt = EXYNOS_CISTATUS2_GET_FRAMECOUNT_PRESENT(cfg);
 
-	DRM_DEBUG_KMS("present[%d]before[%d]\n",
-		EXYNOS_CISTATUS2_GET_FRAMECOUNT_PRESENT(cfg),
-		EXYNOS_CISTATUS2_GET_FRAMECOUNT_BEFORE(cfg));
+	DRM_DEV_DEBUG_KMS(ctx->dev, "present[%d]before[%d]\n",
+			  EXYNOS_CISTATUS2_GET_FRAMECOUNT_PRESENT(cfg),
+			  EXYNOS_CISTATUS2_GET_FRAMECOUNT_BEFORE(cfg));
 
 	if (frame_cnt == 0) {
-		DRM_ERROR("failed to get frame count.\n");
+		DRM_DEV_ERROR(ctx->dev, "failed to get frame count.\n");
 		return -EIO;
 	}
 
 	buf_id = frame_cnt - 1;
-	DRM_DEBUG_KMS("buf_id[%d]\n", buf_id);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "buf_id[%d]\n", buf_id);
 
 	return buf_id;
 }
@@ -287,7 +288,7 @@ static void fimc_handle_lastend(struct fimc_context *ctx, bool enable)
 {
 	u32 cfg;
 
-	DRM_DEBUG_KMS("enable[%d]\n", enable);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "enable[%d]\n", enable);
 
 	cfg = fimc_read(ctx, EXYNOS_CIOCTRL);
 	if (enable)
@@ -302,7 +303,7 @@ static void fimc_src_set_fmt_order(struct fimc_context *ctx, u32 fmt)
 {
 	u32 cfg;
 
-	DRM_DEBUG_KMS("fmt[0x%x]\n", fmt);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "fmt[0x%x]\n", fmt);
 
 	/* RGB */
 	cfg = fimc_read(ctx, EXYNOS_CISCCTRL);
@@ -367,7 +368,7 @@ static void fimc_src_set_fmt(struct fimc_context *ctx, u32 fmt, bool tiled)
 {
 	u32 cfg;
 
-	DRM_DEBUG_KMS("fmt[0x%x]\n", fmt);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "fmt[0x%x]\n", fmt);
 
 	cfg = fimc_read(ctx, EXYNOS_MSCTRL);
 	cfg &= ~EXYNOS_MSCTRL_INFORMAT_RGB;
@@ -420,7 +421,7 @@ static void fimc_src_set_transf(struct fimc_context *ctx, unsigned int rotation)
 	unsigned int degree = rotation & DRM_MODE_ROTATE_MASK;
 	u32 cfg1, cfg2;
 
-	DRM_DEBUG_KMS("rotation[%x]\n", rotation);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "rotation[%x]\n", rotation);
 
 	cfg1 = fimc_read(ctx, EXYNOS_MSCTRL);
 	cfg1 &= ~(EXYNOS_MSCTRL_FLIP_X_MIRROR |
@@ -478,10 +479,11 @@ static void fimc_set_window(struct fimc_context *ctx,
 	v1 = buf->rect.y;
 	v2 = buf->buf.height - buf->rect.h - buf->rect.y;
 
-	DRM_DEBUG_KMS("x[%d]y[%d]w[%d]h[%d]hsize[%d]vsize[%d]\n",
-		buf->rect.x, buf->rect.y, buf->rect.w, buf->rect.h,
-		real_width, buf->buf.height);
-	DRM_DEBUG_KMS("h1[%d]h2[%d]v1[%d]v2[%d]\n", h1, h2, v1, v2);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "x[%d]y[%d]w[%d]h[%d]hsize[%d]vsize[%d]\n",
+			  buf->rect.x, buf->rect.y, buf->rect.w, buf->rect.h,
+			  real_width, buf->buf.height);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "h1[%d]h2[%d]v1[%d]v2[%d]\n", h1, h2, v1,
+			  v2);
 
 	/*
 	 * set window offset 1, 2 size
@@ -506,7 +508,8 @@ static void fimc_src_set_size(struct fimc_context *ctx,
 	unsigned int real_width = buf->buf.pitch[0] / buf->format->cpp[0];
 	u32 cfg;
 
-	DRM_DEBUG_KMS("hsize[%d]vsize[%d]\n", real_width, buf->buf.height);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "hsize[%d]vsize[%d]\n", real_width,
+			  buf->buf.height);
 
 	/* original size */
 	cfg = (EXYNOS_ORGISIZE_HORIZONTAL(real_width) |
@@ -514,8 +517,8 @@ static void fimc_src_set_size(struct fimc_context *ctx,
 
 	fimc_write(ctx, cfg, EXYNOS_ORGISIZE);
 
-	DRM_DEBUG_KMS("x[%d]y[%d]w[%d]h[%d]\n", buf->rect.x, buf->rect.y,
-		buf->rect.w, buf->rect.h);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "x[%d]y[%d]w[%d]h[%d]\n", buf->rect.x,
+			  buf->rect.y, buf->rect.w, buf->rect.h);
 
 	/* set input DMA image size */
 	cfg = fimc_read(ctx, EXYNOS_CIREAL_ISIZE);
@@ -560,7 +563,7 @@ static void fimc_dst_set_fmt_order(struct fimc_context *ctx, u32 fmt)
 {
 	u32 cfg;
 
-	DRM_DEBUG_KMS("fmt[0x%x]\n", fmt);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "fmt[0x%x]\n", fmt);
 
 	/* RGB */
 	cfg = fimc_read(ctx, EXYNOS_CISCCTRL);
@@ -631,7 +634,7 @@ static void fimc_dst_set_fmt(struct fimc_context *ctx, u32 fmt, bool tiled)
 {
 	u32 cfg;
 
-	DRM_DEBUG_KMS("fmt[0x%x]\n", fmt);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "fmt[0x%x]\n", fmt);
 
 	cfg = fimc_read(ctx, EXYNOS_CIEXTEN);
 
@@ -691,7 +694,7 @@ static void fimc_dst_set_transf(struct fimc_context *ctx, unsigned int rotation)
 	unsigned int degree = rotation & DRM_MODE_ROTATE_MASK;
 	u32 cfg;
 
-	DRM_DEBUG_KMS("rotation[0x%x]\n", rotation);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "rotation[0x%x]\n", rotation);
 
 	cfg = fimc_read(ctx, EXYNOS_CITRGFMT);
 	cfg &= ~EXYNOS_CITRGFMT_FLIP_MASK;
@@ -775,19 +778,20 @@ static int fimc_set_prescaler(struct fimc_context *ctx, struct fimc_scaler *sc,
 
 	pre_dst_width = src_w >> hfactor;
 	pre_dst_height = src_h >> vfactor;
-	DRM_DEBUG_KMS("pre_dst_width[%d]pre_dst_height[%d]\n",
-		pre_dst_width, pre_dst_height);
-	DRM_DEBUG_KMS("hfactor[%d]vfactor[%d]\n", hfactor, vfactor);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "pre_dst_width[%d]pre_dst_height[%d]\n",
+			  pre_dst_width, pre_dst_height);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "hfactor[%d]vfactor[%d]\n", hfactor,
+			  vfactor);
 
 	sc->hratio = (src_w << 14) / (dst_w << hfactor);
 	sc->vratio = (src_h << 14) / (dst_h << vfactor);
 	sc->up_h = (dst_w >= src_w) ? true : false;
 	sc->up_v = (dst_h >= src_h) ? true : false;
-	DRM_DEBUG_KMS("hratio[%d]vratio[%d]up_h[%d]up_v[%d]\n",
-		sc->hratio, sc->vratio, sc->up_h, sc->up_v);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "hratio[%d]vratio[%d]up_h[%d]up_v[%d]\n",
+			  sc->hratio, sc->vratio, sc->up_h, sc->up_v);
 
 	shfactor = FIMC_SHFACTOR - (hfactor + vfactor);
-	DRM_DEBUG_KMS("shfactor[%d]\n", shfactor);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "shfactor[%d]\n", shfactor);
 
 	cfg = (EXYNOS_CISCPRERATIO_SHFACTOR(shfactor) |
 		EXYNOS_CISCPRERATIO_PREHORRATIO(1 << hfactor) |
@@ -805,10 +809,10 @@ static void fimc_set_scaler(struct fimc_context *ctx, struct fimc_scaler *sc)
 {
 	u32 cfg, cfg_ext;
 
-	DRM_DEBUG_KMS("range[%d]bypass[%d]up_h[%d]up_v[%d]\n",
-		sc->range, sc->bypass, sc->up_h, sc->up_v);
-	DRM_DEBUG_KMS("hratio[%d]vratio[%d]\n",
-		sc->hratio, sc->vratio);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "range[%d]bypass[%d]up_h[%d]up_v[%d]\n",
+			  sc->range, sc->bypass, sc->up_h, sc->up_v);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "hratio[%d]vratio[%d]\n",
+			  sc->hratio, sc->vratio);
 
 	cfg = fimc_read(ctx, EXYNOS_CISCCTRL);
 	cfg &= ~(EXYNOS_CISCCTRL_SCALERBYPASS |
@@ -846,7 +850,8 @@ static void fimc_dst_set_size(struct fimc_context *ctx,
 	unsigned int real_width = buf->buf.pitch[0] / buf->format->cpp[0];
 	u32 cfg, cfg_ext;
 
-	DRM_DEBUG_KMS("hsize[%d]vsize[%d]\n", real_width, buf->buf.height);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "hsize[%d]vsize[%d]\n", real_width,
+			  buf->buf.height);
 
 	/* original size */
 	cfg = (EXYNOS_ORGOSIZE_HORIZONTAL(real_width) |
@@ -854,8 +859,9 @@ static void fimc_dst_set_size(struct fimc_context *ctx,
 
 	fimc_write(ctx, cfg, EXYNOS_ORGOSIZE);
 
-	DRM_DEBUG_KMS("x[%d]y[%d]w[%d]h[%d]\n", buf->rect.x, buf->rect.y,
-		buf->rect.w, buf->rect.h);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "x[%d]y[%d]w[%d]h[%d]\n", buf->rect.x,
+			  buf->rect.y,
+			  buf->rect.w, buf->rect.h);
 
 	/* CSC ITU */
 	cfg = fimc_read(ctx, EXYNOS_CIGCTRL);
@@ -905,7 +911,7 @@ static void fimc_dst_set_buf_seq(struct fimc_context *ctx, u32 buf_id,
 	u32 buf_num;
 	u32 cfg;
 
-	DRM_DEBUG_KMS("buf_id[%d]enqueu[%d]\n", buf_id, enqueue);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "buf_id[%d]enqueu[%d]\n", buf_id, enqueue);
 
 	spin_lock_irqsave(&ctx->lock, flags);
 
@@ -945,7 +951,7 @@ static irqreturn_t fimc_irq_handler(int irq, void *dev_id)
 	struct fimc_context *ctx = dev_id;
 	int buf_id;
 
-	DRM_DEBUG_KMS("fimc id[%d]\n", ctx->id);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "fimc id[%d]\n", ctx->id);
 
 	fimc_clear_irq(ctx);
 	if (fimc_check_ovf(ctx))
@@ -958,7 +964,7 @@ static irqreturn_t fimc_irq_handler(int irq, void *dev_id)
 	if (buf_id < 0)
 		return IRQ_HANDLED;
 
-	DRM_DEBUG_KMS("buf_id[%d]\n", buf_id);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "buf_id[%d]\n", buf_id);
 
 	if (ctx->task) {
 		struct exynos_drm_ipp_task *task = ctx->task;
@@ -1128,9 +1134,10 @@ static int fimc_bind(struct device *dev, struct device *master, void *data)
 	struct exynos_drm_ipp *ipp = &ctx->ipp;
 
 	ctx->drm_dev = drm_dev;
+	ipp->drm_dev = drm_dev;
 	exynos_drm_register_dma(drm_dev, dev);
 
-	exynos_drm_ipp_register(drm_dev, ipp, &ipp_funcs,
+	exynos_drm_ipp_register(dev, ipp, &ipp_funcs,
 			DRM_EXYNOS_IPP_CAP_CROP | DRM_EXYNOS_IPP_CAP_ROTATE |
 			DRM_EXYNOS_IPP_CAP_SCALE | DRM_EXYNOS_IPP_CAP_CONVERT,
 			ctx->formats, ctx->num_formats, "fimc");
@@ -1147,7 +1154,7 @@ static void fimc_unbind(struct device *dev, struct device *master,
 	struct drm_device *drm_dev = data;
 	struct exynos_drm_ipp *ipp = &ctx->ipp;
 
-	exynos_drm_ipp_unregister(drm_dev, ipp);
+	exynos_drm_ipp_unregister(dev, ipp);
 	exynos_drm_unregister_dma(drm_dev, dev);
 }
 
@@ -1380,7 +1387,7 @@ static int fimc_runtime_suspend(struct device *dev)
 {
 	struct fimc_context *ctx = get_fimc_context(dev);
 
-	DRM_DEBUG_KMS("id[%d]\n", ctx->id);
+	DRM_DEV_DEBUG_KMS(dev, "id[%d]\n", ctx->id);
 	clk_disable_unprepare(ctx->clocks[FIMC_CLK_GATE]);
 	return 0;
 }
@@ -1389,7 +1396,7 @@ static int fimc_runtime_resume(struct device *dev)
 {
 	struct fimc_context *ctx = get_fimc_context(dev);
 
-	DRM_DEBUG_KMS("id[%d]\n", ctx->id);
+	DRM_DEV_DEBUG_KMS(dev, "id[%d]\n", ctx->id);
 	return clk_prepare_enable(ctx->clocks[FIMC_CLK_GATE]);
 }
 #endif
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
index 786a8ee6f10f..8039e1a3671d 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
@@ -315,7 +315,7 @@ static void fimd_wait_for_vblank(struct exynos_drm_crtc *crtc)
 	if (!wait_event_timeout(ctx->wait_vsync_queue,
 				!atomic_read(&ctx->wait_vsync_event),
 				HZ/20))
-		DRM_DEBUG_KMS("vblank wait timed out.\n");
+		DRM_DEV_DEBUG_KMS(ctx->dev, "vblank wait timed out.\n");
 }
 
 static void fimd_enable_video_output(struct fimd_context *ctx, unsigned int win,
@@ -350,8 +350,6 @@ static void fimd_clear_channels(struct exynos_drm_crtc *crtc)
 	struct fimd_context *ctx = crtc->ctx;
 	unsigned int win, ch_enabled = 0;
 
-	DRM_DEBUG_KMS("%s\n", __FILE__);
-
 	/* Hardware is in unknown state, so ensure it gets enabled properly */
 	pm_runtime_get_sync(ctx->dev);
 
@@ -400,7 +398,7 @@ static int fimd_atomic_check(struct exynos_drm_crtc *crtc,
 	u32 clkdiv;
 
 	if (mode->clock == 0) {
-		DRM_INFO("Mode has zero clock value.\n");
+		DRM_DEV_ERROR(ctx->dev, "Mode has zero clock value.\n");
 		return -EINVAL;
 	}
 
@@ -416,15 +414,17 @@ static int fimd_atomic_check(struct exynos_drm_crtc *crtc,
 
 	lcd_rate = clk_get_rate(ctx->lcd_clk);
 	if (2 * lcd_rate < ideal_clk) {
-		DRM_INFO("sclk_fimd clock too low(%lu) for requested pixel clock(%lu)\n",
-			 lcd_rate, ideal_clk);
+		DRM_DEV_ERROR(ctx->dev,
+			      "sclk_fimd clock too low(%lu) for requested pixel clock(%lu)\n",
+			      lcd_rate, ideal_clk);
 		return -EINVAL;
 	}
 
 	/* Find the clock divider value that gets us closest to ideal_clk */
 	clkdiv = DIV_ROUND_CLOSEST(lcd_rate, ideal_clk);
 	if (clkdiv >= 0x200) {
-		DRM_INFO("requested pixel clock(%lu) too low\n", ideal_clk);
+		DRM_DEV_ERROR(ctx->dev, "requested pixel clock(%lu) too low\n",
+			      ideal_clk);
 		return -EINVAL;
 	}
 
@@ -481,7 +481,8 @@ static void fimd_commit(struct exynos_drm_crtc *crtc)
 					driver_data->lcdblk_offset,
 					0x3 << driver_data->lcdblk_vt_shift,
 					0x1 << driver_data->lcdblk_vt_shift)) {
-			DRM_ERROR("Failed to update sysreg for I80 i/f.\n");
+			DRM_DEV_ERROR(ctx->dev,
+				      "Failed to update sysreg for I80 i/f.\n");
 			return;
 		}
 	} else {
@@ -525,7 +526,8 @@ static void fimd_commit(struct exynos_drm_crtc *crtc)
 				driver_data->lcdblk_offset,
 				0x1 << driver_data->lcdblk_bypass_shift,
 				0x1 << driver_data->lcdblk_bypass_shift)) {
-		DRM_ERROR("Failed to update sysreg for bypass setting.\n");
+		DRM_DEV_ERROR(ctx->dev,
+			      "Failed to update sysreg for bypass setting.\n");
 		return;
 	}
 
@@ -537,7 +539,8 @@ static void fimd_commit(struct exynos_drm_crtc *crtc)
 				driver_data->lcdblk_offset,
 				0x1 << driver_data->lcdblk_mic_bypass_shift,
 				0x1 << driver_data->lcdblk_mic_bypass_shift)) {
-		DRM_ERROR("Failed to update sysreg for bypass mic.\n");
+		DRM_DEV_ERROR(ctx->dev,
+			      "Failed to update sysreg for bypass mic.\n");
 		return;
 	}
 
@@ -814,10 +817,11 @@ static void fimd_update_plane(struct exynos_drm_crtc *crtc,
 	val = (unsigned long)(dma_addr + size);
 	writel(val, ctx->regs + VIDWx_BUF_END(win, 0));
 
-	DRM_DEBUG_KMS("start addr = 0x%lx, end addr = 0x%lx, size = 0x%lx\n",
-			(unsigned long)dma_addr, val, size);
-	DRM_DEBUG_KMS("ovl_width = %d, ovl_height = %d\n",
-			state->crtc.w, state->crtc.h);
+	DRM_DEV_DEBUG_KMS(ctx->dev,
+			  "start addr = 0x%lx, end addr = 0x%lx, size = 0x%lx\n",
+			  (unsigned long)dma_addr, val, size);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "ovl_width = %d, ovl_height = %d\n",
+			  state->crtc.w, state->crtc.h);
 
 	/* buffer size */
 	buf_offsize = pitch - (state->crtc.w * cpp);
@@ -847,8 +851,9 @@ static void fimd_update_plane(struct exynos_drm_crtc *crtc,
 
 	writel(val, ctx->regs + VIDOSD_B(win));
 
-	DRM_DEBUG_KMS("osd pos: tx = %d, ty = %d, bx = %d, by = %d\n",
-			state->crtc.x, state->crtc.y, last_x, last_y);
+	DRM_DEV_DEBUG_KMS(ctx->dev,
+			  "osd pos: tx = %d, ty = %d, bx = %d, by = %d\n",
+			  state->crtc.x, state->crtc.y, last_x, last_y);
 
 	/* OSD size */
 	if (win != 3 && win != 4) {
@@ -858,7 +863,8 @@ static void fimd_update_plane(struct exynos_drm_crtc *crtc,
 		val = state->crtc.w * state->crtc.h;
 		writel(val, ctx->regs + offset);
 
-		DRM_DEBUG_KMS("osd size = 0x%x\n", (unsigned int)val);
+		DRM_DEV_DEBUG_KMS(ctx->dev, "osd size = 0x%x\n",
+				  (unsigned int)val);
 	}
 
 	fimd_win_set_pixfmt(ctx, win, fb, state->src.w);
@@ -1252,13 +1258,17 @@ static int exynos_fimd_resume(struct device *dev)
 
 	ret = clk_prepare_enable(ctx->bus_clk);
 	if (ret < 0) {
-		DRM_ERROR("Failed to prepare_enable the bus clk [%d]\n", ret);
+		DRM_DEV_ERROR(dev,
+			      "Failed to prepare_enable the bus clk [%d]\n",
+			      ret);
 		return ret;
 	}
 
 	ret = clk_prepare_enable(ctx->lcd_clk);
 	if  (ret < 0) {
-		DRM_ERROR("Failed to prepare_enable the lcd clk [%d]\n", ret);
+		DRM_DEV_ERROR(dev,
+			      "Failed to prepare_enable the lcd clk [%d]\n",
+			      ret);
 		return ret;
 	}
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_g2d.c b/drivers/gpu/drm/exynos/exynos_drm_g2d.c
index 24c536d6d9cf..c20b3a759370 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_g2d.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_g2d.c
@@ -429,7 +429,7 @@ static dma_addr_t *g2d_userptr_get_dma_addr(struct g2d_data *g2d,
 	int ret;
 
 	if (!size) {
-		DRM_ERROR("invalid userptr size.\n");
+		DRM_DEV_ERROR(g2d->dev, "invalid userptr size.\n");
 		return ERR_PTR(-EINVAL);
 	}
 
@@ -482,7 +482,8 @@ static dma_addr_t *g2d_userptr_get_dma_addr(struct g2d_data *g2d,
 	ret = get_vaddr_frames(start, npages, FOLL_FORCE | FOLL_WRITE,
 		g2d_userptr->vec);
 	if (ret != npages) {
-		DRM_ERROR("failed to get user pages from userptr.\n");
+		DRM_DEV_ERROR(g2d->dev,
+			      "failed to get user pages from userptr.\n");
 		if (ret < 0)
 			goto err_destroy_framevec;
 		ret = -EFAULT;
@@ -503,7 +504,7 @@ static dma_addr_t *g2d_userptr_get_dma_addr(struct g2d_data *g2d,
 					frame_vector_pages(g2d_userptr->vec),
 					npages, offset, size, GFP_KERNEL);
 	if (ret < 0) {
-		DRM_ERROR("failed to get sgt from pages.\n");
+		DRM_DEV_ERROR(g2d->dev, "failed to get sgt from pages.\n");
 		goto err_free_sgt;
 	}
 
@@ -511,7 +512,7 @@ static dma_addr_t *g2d_userptr_get_dma_addr(struct g2d_data *g2d,
 
 	if (!dma_map_sg(to_dma_dev(g2d->drm_dev), sgt->sgl, sgt->nents,
 				DMA_BIDIRECTIONAL)) {
-		DRM_ERROR("failed to map sgt with dma region.\n");
+		DRM_DEV_ERROR(g2d->dev, "failed to map sgt with dma region.\n");
 		ret = -ENOMEM;
 		goto err_sg_free_table;
 	}
@@ -560,7 +561,7 @@ static void g2d_userptr_free_all(struct g2d_data *g2d, struct drm_file *filp)
 	g2d->current_pool = 0;
 }
 
-static enum g2d_reg_type g2d_get_reg_type(int reg_offset)
+static enum g2d_reg_type g2d_get_reg_type(struct g2d_data *g2d, int reg_offset)
 {
 	enum g2d_reg_type reg_type;
 
@@ -593,7 +594,8 @@ static enum g2d_reg_type g2d_get_reg_type(int reg_offset)
 		break;
 	default:
 		reg_type = REG_TYPE_NONE;
-		DRM_ERROR("Unknown register offset![%d]\n", reg_offset);
+		DRM_DEV_ERROR(g2d->dev, "Unknown register offset![%d]\n",
+			      reg_offset);
 		break;
 	}
 
@@ -627,9 +629,10 @@ static unsigned long g2d_get_buf_bpp(unsigned int format)
 	return bpp;
 }
 
-static bool g2d_check_buf_desc_is_valid(struct g2d_buf_desc *buf_desc,
-						enum g2d_reg_type reg_type,
-						unsigned long size)
+static bool g2d_check_buf_desc_is_valid(struct g2d_data *g2d,
+					struct g2d_buf_desc *buf_desc,
+					enum g2d_reg_type reg_type,
+					unsigned long size)
 {
 	int width, height;
 	unsigned long bpp, last_pos;
@@ -644,14 +647,15 @@ static bool g2d_check_buf_desc_is_valid(struct g2d_buf_desc *buf_desc,
 	/* This check also makes sure that right_x > left_x. */
 	width = (int)buf_desc->right_x - (int)buf_desc->left_x;
 	if (width < G2D_LEN_MIN || width > G2D_LEN_MAX) {
-		DRM_ERROR("width[%d] is out of range!\n", width);
+		DRM_DEV_ERROR(g2d->dev, "width[%d] is out of range!\n", width);
 		return false;
 	}
 
 	/* This check also makes sure that bottom_y > top_y. */
 	height = (int)buf_desc->bottom_y - (int)buf_desc->top_y;
 	if (height < G2D_LEN_MIN || height > G2D_LEN_MAX) {
-		DRM_ERROR("height[%d] is out of range!\n", height);
+		DRM_DEV_ERROR(g2d->dev,
+			      "height[%d] is out of range!\n", height);
 		return false;
 	}
 
@@ -670,8 +674,8 @@ static bool g2d_check_buf_desc_is_valid(struct g2d_buf_desc *buf_desc,
 	 */
 
 	if (last_pos >= size) {
-		DRM_ERROR("last engine access position [%lu] "
-			"is out of range [%lu]!\n", last_pos, size);
+		DRM_DEV_ERROR(g2d->dev, "last engine access position [%lu] "
+			      "is out of range [%lu]!\n", last_pos, size);
 		return false;
 	}
 
@@ -701,7 +705,7 @@ static int g2d_map_cmdlist_gem(struct g2d_data *g2d,
 		offset = cmdlist->data[reg_pos];
 		handle = cmdlist->data[reg_pos + 1];
 
-		reg_type = g2d_get_reg_type(offset);
+		reg_type = g2d_get_reg_type(g2d, offset);
 		if (reg_type == REG_TYPE_NONE) {
 			ret = -EFAULT;
 			goto err;
@@ -718,7 +722,7 @@ static int g2d_map_cmdlist_gem(struct g2d_data *g2d,
 				goto err;
 			}
 
-			if (!g2d_check_buf_desc_is_valid(buf_desc,
+			if (!g2d_check_buf_desc_is_valid(g2d, buf_desc,
 							 reg_type, exynos_gem->size)) {
 				exynos_drm_gem_put(exynos_gem);
 				ret = -EFAULT;
@@ -736,8 +740,9 @@ static int g2d_map_cmdlist_gem(struct g2d_data *g2d,
 				goto err;
 			}
 
-			if (!g2d_check_buf_desc_is_valid(buf_desc, reg_type,
-							g2d_userptr.size)) {
+			if (!g2d_check_buf_desc_is_valid(g2d, buf_desc,
+							 reg_type,
+							 g2d_userptr.size)) {
 				ret = -EFAULT;
 				goto err;
 			}
@@ -845,7 +850,7 @@ static void g2d_free_runqueue_node(struct g2d_data *g2d,
  *
  * Has to be called under runqueue lock.
  */
-static void g2d_remove_runqueue_nodes(struct g2d_data *g2d, struct drm_file* file)
+static void g2d_remove_runqueue_nodes(struct g2d_data *g2d, struct drm_file *file)
 {
 	struct g2d_runqueue_node *node, *n;
 
@@ -1044,7 +1049,7 @@ static int g2d_check_reg_offset(struct g2d_data *g2d,
 			if (!for_addr)
 				goto err;
 
-			reg_type = g2d_get_reg_type(reg_offset);
+			reg_type = g2d_get_reg_type(g2d, reg_offset);
 
 			/* check userptr buffer type. */
 			if ((cmdlist->data[index] & ~0x7fffffff) >> 31) {
@@ -1058,7 +1063,7 @@ static int g2d_check_reg_offset(struct g2d_data *g2d,
 			if (for_addr)
 				goto err;
 
-			reg_type = g2d_get_reg_type(reg_offset);
+			reg_type = g2d_get_reg_type(g2d, reg_offset);
 
 			buf_desc = &buf_info->descs[reg_type];
 			buf_desc->stride = cmdlist->data[index + 1];
@@ -1068,7 +1073,7 @@ static int g2d_check_reg_offset(struct g2d_data *g2d,
 			if (for_addr)
 				goto err;
 
-			reg_type = g2d_get_reg_type(reg_offset);
+			reg_type = g2d_get_reg_type(g2d, reg_offset);
 
 			buf_desc = &buf_info->descs[reg_type];
 			value = cmdlist->data[index + 1];
@@ -1080,7 +1085,7 @@ static int g2d_check_reg_offset(struct g2d_data *g2d,
 			if (for_addr)
 				goto err;
 
-			reg_type = g2d_get_reg_type(reg_offset);
+			reg_type = g2d_get_reg_type(g2d, reg_offset);
 
 			buf_desc = &buf_info->descs[reg_type];
 			value = cmdlist->data[index + 1];
@@ -1093,7 +1098,7 @@ static int g2d_check_reg_offset(struct g2d_data *g2d,
 			if (for_addr)
 				goto err;
 
-			reg_type = g2d_get_reg_type(reg_offset);
+			reg_type = g2d_get_reg_type(g2d, reg_offset);
 
 			buf_desc = &buf_info->descs[reg_type];
 			value = cmdlist->data[index + 1];
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c
index df66c383a877..a55f5ac41bf3 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gem.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c
@@ -29,7 +29,7 @@ static int exynos_drm_alloc_buf(struct exynos_drm_gem *exynos_gem)
 	int ret = -ENOMEM;
 
 	if (exynos_gem->dma_addr) {
-		DRM_DEBUG_KMS("already allocated.\n");
+		DRM_DEV_DEBUG_KMS(to_dma_dev(dev), "already allocated.\n");
 		return 0;
 	}
 
@@ -61,7 +61,7 @@ static int exynos_drm_alloc_buf(struct exynos_drm_gem *exynos_gem)
 	exynos_gem->pages = kvmalloc_array(nr_pages, sizeof(struct page *),
 			GFP_KERNEL | __GFP_ZERO);
 	if (!exynos_gem->pages) {
-		DRM_ERROR("failed to allocate pages.\n");
+		DRM_DEV_ERROR(to_dma_dev(dev), "failed to allocate pages.\n");
 		return -ENOMEM;
 	}
 
@@ -69,7 +69,7 @@ static int exynos_drm_alloc_buf(struct exynos_drm_gem *exynos_gem)
 					     &exynos_gem->dma_addr, GFP_KERNEL,
 					     exynos_gem->dma_attrs);
 	if (!exynos_gem->cookie) {
-		DRM_ERROR("failed to allocate buffer.\n");
+		DRM_DEV_ERROR(to_dma_dev(dev), "failed to allocate buffer.\n");
 		goto err_free;
 	}
 
@@ -77,20 +77,20 @@ static int exynos_drm_alloc_buf(struct exynos_drm_gem *exynos_gem)
 				    exynos_gem->dma_addr, exynos_gem->size,
 				    exynos_gem->dma_attrs);
 	if (ret < 0) {
-		DRM_ERROR("failed to get sgtable.\n");
+		DRM_DEV_ERROR(to_dma_dev(dev), "failed to get sgtable.\n");
 		goto err_dma_free;
 	}
 
 	if (drm_prime_sg_to_page_addr_arrays(&sgt, exynos_gem->pages, NULL,
 					     nr_pages)) {
-		DRM_ERROR("invalid sgtable.\n");
+		DRM_DEV_ERROR(to_dma_dev(dev), "invalid sgtable.\n");
 		ret = -EINVAL;
 		goto err_sgt_free;
 	}
 
 	sg_free_table(&sgt);
 
-	DRM_DEBUG_KMS("dma_addr(0x%lx), size(0x%lx)\n",
+	DRM_DEV_DEBUG_KMS(to_dma_dev(dev), "dma_addr(0x%lx), size(0x%lx)\n",
 			(unsigned long)exynos_gem->dma_addr, exynos_gem->size);
 
 	return 0;
@@ -111,11 +111,11 @@ static void exynos_drm_free_buf(struct exynos_drm_gem *exynos_gem)
 	struct drm_device *dev = exynos_gem->base.dev;
 
 	if (!exynos_gem->dma_addr) {
-		DRM_DEBUG_KMS("dma_addr is invalid.\n");
+		DRM_DEV_DEBUG_KMS(dev->dev, "dma_addr is invalid.\n");
 		return;
 	}
 
-	DRM_DEBUG_KMS("dma_addr(0x%lx), size(0x%lx)\n",
+	DRM_DEV_DEBUG_KMS(dev->dev, "dma_addr(0x%lx), size(0x%lx)\n",
 			(unsigned long)exynos_gem->dma_addr, exynos_gem->size);
 
 	dma_free_attrs(to_dma_dev(dev), exynos_gem->size, exynos_gem->cookie,
@@ -139,7 +139,7 @@ static int exynos_drm_gem_handle_create(struct drm_gem_object *obj,
 	if (ret)
 		return ret;
 
-	DRM_DEBUG_KMS("gem handle = 0x%x\n", *handle);
+	DRM_DEV_DEBUG_KMS(to_dma_dev(obj->dev), "gem handle = 0x%x\n", *handle);
 
 	/* drop reference from allocate - handle holds it now. */
 	drm_gem_object_put_unlocked(obj);
@@ -151,7 +151,8 @@ void exynos_drm_gem_destroy(struct exynos_drm_gem *exynos_gem)
 {
 	struct drm_gem_object *obj = &exynos_gem->base;
 
-	DRM_DEBUG_KMS("handle count = %d\n", obj->handle_count);
+	DRM_DEV_DEBUG_KMS(to_dma_dev(obj->dev), "handle count = %d\n",
+			  obj->handle_count);
 
 	/*
 	 * do not release memory region from exporter.
@@ -186,7 +187,7 @@ static struct exynos_drm_gem *exynos_drm_gem_init(struct drm_device *dev,
 
 	ret = drm_gem_object_init(dev, obj, size);
 	if (ret < 0) {
-		DRM_ERROR("failed to initialize gem object\n");
+		DRM_DEV_ERROR(dev->dev, "failed to initialize gem object\n");
 		kfree(exynos_gem);
 		return ERR_PTR(ret);
 	}
@@ -198,7 +199,7 @@ static struct exynos_drm_gem *exynos_drm_gem_init(struct drm_device *dev,
 		return ERR_PTR(ret);
 	}
 
-	DRM_DEBUG_KMS("created file object = %pK\n", obj->filp);
+	DRM_DEV_DEBUG_KMS(dev->dev, "created file object = %pK\n", obj->filp);
 
 	return exynos_gem;
 }
@@ -211,12 +212,13 @@ struct exynos_drm_gem *exynos_drm_gem_create(struct drm_device *dev,
 	int ret;
 
 	if (flags & ~(EXYNOS_BO_MASK)) {
-		DRM_ERROR("invalid GEM buffer flags: %u\n", flags);
+		DRM_DEV_ERROR(dev->dev,
+			      "invalid GEM buffer flags: %u\n", flags);
 		return ERR_PTR(-EINVAL);
 	}
 
 	if (!size) {
-		DRM_ERROR("invalid GEM buffer size: %lu\n", size);
+		DRM_DEV_ERROR(dev->dev, "invalid GEM buffer size: %lu\n", size);
 		return ERR_PTR(-EINVAL);
 	}
 
@@ -325,7 +327,7 @@ int exynos_drm_gem_get_ioctl(struct drm_device *dev, void *data,
 
 	obj = drm_gem_object_lookup(file_priv, args->handle);
 	if (!obj) {
-		DRM_ERROR("failed to lookup gem object.\n");
+		DRM_DEV_ERROR(dev->dev, "failed to lookup gem object.\n");
 		return -EINVAL;
 	}
 
@@ -408,7 +410,8 @@ static int exynos_drm_gem_mmap_obj(struct drm_gem_object *obj,
 	struct exynos_drm_gem *exynos_gem = to_exynos_gem(obj);
 	int ret;
 
-	DRM_DEBUG_KMS("flags = 0x%x\n", exynos_gem->flags);
+	DRM_DEV_DEBUG_KMS(to_dma_dev(obj->dev), "flags = 0x%x\n",
+			  exynos_gem->flags);
 
 	/* non-cachable as default. */
 	if (exynos_gem->flags & EXYNOS_BO_CACHABLE)
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gsc.c b/drivers/gpu/drm/exynos/exynos_drm_gsc.c
index f048d97fe9e2..0bfb5e9f6e91 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gsc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_gsc.c
@@ -395,7 +395,7 @@ static int gsc_sw_reset(struct gsc_context *ctx)
 	}
 
 	if (cfg) {
-		DRM_ERROR("failed to reset gsc h/w.\n");
+		DRM_DEV_ERROR(ctx->dev, "failed to reset gsc h/w.\n");
 		return -EBUSY;
 	}
 
@@ -422,8 +422,8 @@ static void gsc_handle_irq(struct gsc_context *ctx, bool enable,
 {
 	u32 cfg;
 
-	DRM_DEBUG_KMS("enable[%d]overflow[%d]level[%d]\n",
-			enable, overflow, done);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "enable[%d]overflow[%d]level[%d]\n",
+			  enable, overflow, done);
 
 	cfg = gsc_read(GSC_IRQ);
 	cfg |= (GSC_IRQ_OR_MASK | GSC_IRQ_FRMDONE_MASK);
@@ -451,7 +451,7 @@ static void gsc_src_set_fmt(struct gsc_context *ctx, u32 fmt, bool tiled)
 {
 	u32 cfg;
 
-	DRM_DEBUG_KMS("fmt[0x%x]\n", fmt);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "fmt[0x%x]\n", fmt);
 
 	cfg = gsc_read(GSC_IN_CON);
 	cfg &= ~(GSC_IN_RGB_TYPE_MASK | GSC_IN_YUV422_1P_ORDER_MASK |
@@ -638,7 +638,7 @@ static void gsc_dst_set_fmt(struct gsc_context *ctx, u32 fmt, bool tiled)
 {
 	u32 cfg;
 
-	DRM_DEBUG_KMS("fmt[0x%x]\n", fmt);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "fmt[0x%x]\n", fmt);
 
 	cfg = gsc_read(GSC_OUT_CON);
 	cfg &= ~(GSC_OUT_RGB_TYPE_MASK | GSC_OUT_YUV422_1P_ORDER_MASK |
@@ -706,12 +706,13 @@ static void gsc_dst_set_fmt(struct gsc_context *ctx, u32 fmt, bool tiled)
 	gsc_write(cfg, GSC_OUT_CON);
 }
 
-static int gsc_get_ratio_shift(u32 src, u32 dst, u32 *ratio)
+static int gsc_get_ratio_shift(struct gsc_context *ctx, u32 src, u32 dst,
+			       u32 *ratio)
 {
-	DRM_DEBUG_KMS("src[%d]dst[%d]\n", src, dst);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "src[%d]dst[%d]\n", src, dst);
 
 	if (src >= dst * 8) {
-		DRM_ERROR("failed to make ratio and shift.\n");
+		DRM_DEV_ERROR(ctx->dev, "failed to make ratio and shift.\n");
 		return -EINVAL;
 	} else if (src >= dst * 4)
 		*ratio = 4;
@@ -759,31 +760,31 @@ static int gsc_set_prescaler(struct gsc_context *ctx, struct gsc_scaler *sc,
 		dst_h = dst->h;
 	}
 
-	ret = gsc_get_ratio_shift(src_w, dst_w, &sc->pre_hratio);
+	ret = gsc_get_ratio_shift(ctx, src_w, dst_w, &sc->pre_hratio);
 	if (ret) {
-		dev_err(ctx->dev, "failed to get ratio horizontal.\n");
+		DRM_DEV_ERROR(ctx->dev, "failed to get ratio horizontal.\n");
 		return ret;
 	}
 
-	ret = gsc_get_ratio_shift(src_h, dst_h, &sc->pre_vratio);
+	ret = gsc_get_ratio_shift(ctx, src_h, dst_h, &sc->pre_vratio);
 	if (ret) {
-		dev_err(ctx->dev, "failed to get ratio vertical.\n");
+		DRM_DEV_ERROR(ctx->dev, "failed to get ratio vertical.\n");
 		return ret;
 	}
 
-	DRM_DEBUG_KMS("pre_hratio[%d]pre_vratio[%d]\n",
-		sc->pre_hratio, sc->pre_vratio);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "pre_hratio[%d]pre_vratio[%d]\n",
+			  sc->pre_hratio, sc->pre_vratio);
 
 	sc->main_hratio = (src_w << 16) / dst_w;
 	sc->main_vratio = (src_h << 16) / dst_h;
 
-	DRM_DEBUG_KMS("main_hratio[%ld]main_vratio[%ld]\n",
-		sc->main_hratio, sc->main_vratio);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "main_hratio[%ld]main_vratio[%ld]\n",
+			  sc->main_hratio, sc->main_vratio);
 
 	gsc_get_prescaler_shfactor(sc->pre_hratio, sc->pre_vratio,
 		&sc->pre_shfactor);
 
-	DRM_DEBUG_KMS("pre_shfactor[%d]\n", sc->pre_shfactor);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "pre_shfactor[%d]\n", sc->pre_shfactor);
 
 	cfg = (GSC_PRESC_SHFACTOR(sc->pre_shfactor) |
 		GSC_PRESC_H_RATIO(sc->pre_hratio) |
@@ -849,8 +850,8 @@ static void gsc_set_scaler(struct gsc_context *ctx, struct gsc_scaler *sc)
 {
 	u32 cfg;
 
-	DRM_DEBUG_KMS("main_hratio[%ld]main_vratio[%ld]\n",
-		sc->main_hratio, sc->main_vratio);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "main_hratio[%ld]main_vratio[%ld]\n",
+			  sc->main_hratio, sc->main_vratio);
 
 	gsc_set_h_coef(ctx, sc->main_hratio);
 	cfg = GSC_MAIN_H_RATIO_VALUE(sc->main_hratio);
@@ -916,7 +917,7 @@ static int gsc_dst_get_buf_seq(struct gsc_context *ctx)
 		if (cfg & (mask << i))
 			buf_num--;
 
-	DRM_DEBUG_KMS("buf_num[%d]\n", buf_num);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "buf_num[%d]\n", buf_num);
 
 	return buf_num;
 }
@@ -963,7 +964,7 @@ static int gsc_get_src_buf_index(struct gsc_context *ctx)
 	u32 cfg, curr_index, i;
 	u32 buf_id = GSC_MAX_SRC;
 
-	DRM_DEBUG_KMS("gsc id[%d]\n", ctx->id);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "gsc id[%d]\n", ctx->id);
 
 	cfg = gsc_read(GSC_IN_BASE_ADDR_Y_MASK);
 	curr_index = GSC_IN_CURR_GET_INDEX(cfg);
@@ -975,11 +976,11 @@ static int gsc_get_src_buf_index(struct gsc_context *ctx)
 		}
 	}
 
-	DRM_DEBUG_KMS("cfg[0x%x]curr_index[%d]buf_id[%d]\n", cfg,
-		curr_index, buf_id);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "cfg[0x%x]curr_index[%d]buf_id[%d]\n", cfg,
+			  curr_index, buf_id);
 
 	if (buf_id == GSC_MAX_SRC) {
-		DRM_ERROR("failed to get in buffer index.\n");
+		DRM_DEV_ERROR(ctx->dev, "failed to get in buffer index.\n");
 		return -EINVAL;
 	}
 
@@ -993,7 +994,7 @@ static int gsc_get_dst_buf_index(struct gsc_context *ctx)
 	u32 cfg, curr_index, i;
 	u32 buf_id = GSC_MAX_DST;
 
-	DRM_DEBUG_KMS("gsc id[%d]\n", ctx->id);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "gsc id[%d]\n", ctx->id);
 
 	cfg = gsc_read(GSC_OUT_BASE_ADDR_Y_MASK);
 	curr_index = GSC_OUT_CURR_GET_INDEX(cfg);
@@ -1006,14 +1007,14 @@ static int gsc_get_dst_buf_index(struct gsc_context *ctx)
 	}
 
 	if (buf_id == GSC_MAX_DST) {
-		DRM_ERROR("failed to get out buffer index.\n");
+		DRM_DEV_ERROR(ctx->dev, "failed to get out buffer index.\n");
 		return -EINVAL;
 	}
 
 	gsc_dst_set_buf_seq(ctx, buf_id, false);
 
-	DRM_DEBUG_KMS("cfg[0x%x]curr_index[%d]buf_id[%d]\n", cfg,
-		curr_index, buf_id);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "cfg[0x%x]curr_index[%d]buf_id[%d]\n", cfg,
+			  curr_index, buf_id);
 
 	return buf_id;
 }
@@ -1024,7 +1025,7 @@ static irqreturn_t gsc_irq_handler(int irq, void *dev_id)
 	u32 status;
 	int err = 0;
 
-	DRM_DEBUG_KMS("gsc id[%d]\n", ctx->id);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "gsc id[%d]\n", ctx->id);
 
 	status = gsc_read(GSC_IRQ);
 	if (status & GSC_IRQ_STATUS_OR_IRQ) {
@@ -1042,8 +1043,8 @@ static irqreturn_t gsc_irq_handler(int irq, void *dev_id)
 		src_buf_id = gsc_get_src_buf_index(ctx);
 		dst_buf_id = gsc_get_dst_buf_index(ctx);
 
-		DRM_DEBUG_KMS("buf_id_src[%d]buf_id_dst[%d]\n",	src_buf_id,
-			      dst_buf_id);
+		DRM_DEV_DEBUG_KMS(ctx->dev, "buf_id_src[%d]buf_id_dst[%d]\n",
+				  src_buf_id, dst_buf_id);
 
 		if (src_buf_id < 0 || dst_buf_id < 0)
 			err = -EINVAL;
@@ -1169,9 +1170,10 @@ static int gsc_bind(struct device *dev, struct device *master, void *data)
 	struct exynos_drm_ipp *ipp = &ctx->ipp;
 
 	ctx->drm_dev = drm_dev;
+	ctx->drm_dev = drm_dev;
 	exynos_drm_register_dma(drm_dev, dev);
 
-	exynos_drm_ipp_register(drm_dev, ipp, &ipp_funcs,
+	exynos_drm_ipp_register(dev, ipp, &ipp_funcs,
 			DRM_EXYNOS_IPP_CAP_CROP | DRM_EXYNOS_IPP_CAP_ROTATE |
 			DRM_EXYNOS_IPP_CAP_SCALE | DRM_EXYNOS_IPP_CAP_CONVERT,
 			ctx->formats, ctx->num_formats, "gsc");
@@ -1188,7 +1190,7 @@ static void gsc_unbind(struct device *dev, struct device *master,
 	struct drm_device *drm_dev = data;
 	struct exynos_drm_ipp *ipp = &ctx->ipp;
 
-	exynos_drm_ipp_unregister(drm_dev, ipp);
+	exynos_drm_ipp_unregister(dev, ipp);
 	exynos_drm_unregister_dma(drm_dev, dev);
 }
 
@@ -1324,7 +1326,7 @@ static int __maybe_unused gsc_runtime_suspend(struct device *dev)
 	struct gsc_context *ctx = get_gsc_context(dev);
 	int i;
 
-	DRM_DEBUG_KMS("id[%d]\n", ctx->id);
+	DRM_DEV_DEBUG_KMS(dev, "id[%d]\n", ctx->id);
 
 	for (i = ctx->num_clocks - 1; i >= 0; i--)
 		clk_disable_unprepare(ctx->clocks[i]);
@@ -1337,7 +1339,7 @@ static int __maybe_unused gsc_runtime_resume(struct device *dev)
 	struct gsc_context *ctx = get_gsc_context(dev);
 	int i, ret;
 
-	DRM_DEBUG_KMS("id[%d]\n", ctx->id);
+	DRM_DEV_DEBUG_KMS(dev, "id[%d]\n", ctx->id);
 
 	for (i = 0; i < ctx->num_clocks; i++) {
 		ret = clk_prepare_enable(ctx->clocks[i]);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_ipp.c b/drivers/gpu/drm/exynos/exynos_drm_ipp.c
index 23226a0212e8..c862099723a0 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_ipp.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_ipp.c
@@ -43,7 +43,7 @@ static LIST_HEAD(ipp_list);
  * Returns:
  * Zero on success, error code on failure.
  */
-int exynos_drm_ipp_register(struct drm_device *dev, struct exynos_drm_ipp *ipp,
+int exynos_drm_ipp_register(struct device *dev, struct exynos_drm_ipp *ipp,
 		const struct exynos_drm_ipp_funcs *funcs, unsigned int caps,
 		const struct exynos_drm_ipp_formats *formats,
 		unsigned int num_formats, const char *name)
@@ -67,7 +67,7 @@ int exynos_drm_ipp_register(struct drm_device *dev, struct exynos_drm_ipp *ipp,
 	list_add_tail(&ipp->head, &ipp_list);
 	ipp->id = num_ipp++;
 
-	DRM_DEBUG_DRIVER("Registered ipp %d\n", ipp->id);
+	DRM_DEV_DEBUG_DRIVER(dev, "Registered ipp %d\n", ipp->id);
 
 	return 0;
 }
@@ -77,7 +77,7 @@ int exynos_drm_ipp_register(struct drm_device *dev, struct exynos_drm_ipp *ipp,
  * @dev: DRM device
  * @ipp: ipp module
  */
-void exynos_drm_ipp_unregister(struct drm_device *dev,
+void exynos_drm_ipp_unregister(struct device *dev,
 			       struct exynos_drm_ipp *ipp)
 {
 	WARN_ON(ipp->task);
@@ -268,7 +268,7 @@ static inline struct exynos_drm_ipp_task *
 	task->src.rect.h = task->dst.rect.h = UINT_MAX;
 	task->transform.rotation = DRM_MODE_ROTATE_0;
 
-	DRM_DEBUG_DRIVER("Allocated task %pK\n", task);
+	DRM_DEV_DEBUG_DRIVER(task->dev, "Allocated task %pK\n", task);
 
 	return task;
 }
@@ -335,7 +335,9 @@ static int exynos_drm_ipp_task_set(struct exynos_drm_ipp_task *task,
 		size -= map[i].size;
 	}
 
-	DRM_DEBUG_DRIVER("Got task %pK configuration from userspace\n", task);
+	DRM_DEV_DEBUG_DRIVER(task->dev,
+			     "Got task %pK configuration from userspace\n",
+			     task);
 	return 0;
 }
 
@@ -389,12 +391,12 @@ static void exynos_drm_ipp_task_release_buf(struct exynos_drm_ipp_buffer *buf)
 static void exynos_drm_ipp_task_free(struct exynos_drm_ipp *ipp,
 				 struct exynos_drm_ipp_task *task)
 {
-	DRM_DEBUG_DRIVER("Freeing task %pK\n", task);
+	DRM_DEV_DEBUG_DRIVER(task->dev, "Freeing task %pK\n", task);
 
 	exynos_drm_ipp_task_release_buf(&task->src);
 	exynos_drm_ipp_task_release_buf(&task->dst);
 	if (task->event)
-		drm_event_cancel_free(ipp->dev, &task->event->base);
+		drm_event_cancel_free(ipp->drm_dev, &task->event->base);
 	kfree(task);
 }
 
@@ -553,8 +555,9 @@ static int exynos_drm_ipp_check_format(struct exynos_drm_ipp_task *task,
 			       buf == src ? DRM_EXYNOS_IPP_FORMAT_SOURCE :
 					    DRM_EXYNOS_IPP_FORMAT_DESTINATION);
 	if (!fmt) {
-		DRM_DEBUG_DRIVER("Task %pK: %s format not supported\n", task,
-				 buf == src ? "src" : "dst");
+		DRM_DEV_DEBUG_DRIVER(task->dev,
+				     "Task %pK: %s format not supported\n",
+				     task, buf == src ? "src" : "dst");
 		return -EINVAL;
 	}
 
@@ -603,7 +606,7 @@ static int exynos_drm_ipp_task_check(struct exynos_drm_ipp_task *task)
 	bool rotate = (rotation != DRM_MODE_ROTATE_0);
 	bool scale = false;
 
-	DRM_DEBUG_DRIVER("Checking task %pK\n", task);
+	DRM_DEV_DEBUG_DRIVER(task->dev, "Checking task %pK\n", task);
 
 	if (src->rect.w == UINT_MAX)
 		src->rect.w = src->buf.width;
@@ -618,8 +621,9 @@ static int exynos_drm_ipp_task_check(struct exynos_drm_ipp_task *task)
 	    src->rect.y + src->rect.h > (src->buf.height) ||
 	    dst->rect.x + dst->rect.w > (dst->buf.width) ||
 	    dst->rect.y + dst->rect.h > (dst->buf.height)) {
-		DRM_DEBUG_DRIVER("Task %pK: defined area is outside provided buffers\n",
-				 task);
+		DRM_DEV_DEBUG_DRIVER(task->dev,
+				     "Task %pK: defined area is outside provided buffers\n",
+				     task);
 		return -EINVAL;
 	}
 
@@ -635,7 +639,8 @@ static int exynos_drm_ipp_task_check(struct exynos_drm_ipp_task *task)
 	    (!(ipp->capabilities & DRM_EXYNOS_IPP_CAP_SCALE) && scale) ||
 	    (!(ipp->capabilities & DRM_EXYNOS_IPP_CAP_CONVERT) &&
 	     src->buf.fourcc != dst->buf.fourcc)) {
-		DRM_DEBUG_DRIVER("Task %pK: hw capabilities exceeded\n", task);
+		DRM_DEV_DEBUG_DRIVER(task->dev, "Task %pK: hw capabilities exceeded\n",
+				     task);
 		return -EINVAL;
 	}
 
@@ -647,7 +652,8 @@ static int exynos_drm_ipp_task_check(struct exynos_drm_ipp_task *task)
 	if (ret)
 		return ret;
 
-	DRM_DEBUG_DRIVER("Task %pK: all checks done.\n", task);
+	DRM_DEV_DEBUG_DRIVER(ipp->dev, "Task %pK: all checks done.\n",
+			     task);
 
 	return ret;
 }
@@ -658,20 +664,26 @@ static int exynos_drm_ipp_task_setup_buffers(struct exynos_drm_ipp_task *task,
 	struct exynos_drm_ipp_buffer *src = &task->src, *dst = &task->dst;
 	int ret = 0;
 
-	DRM_DEBUG_DRIVER("Setting buffer for task %pK\n", task);
+	DRM_DEV_DEBUG_DRIVER(task->dev, "Setting buffer for task %pK\n",
+			     task);
 
 	ret = exynos_drm_ipp_task_setup_buffer(src, filp);
 	if (ret) {
-		DRM_DEBUG_DRIVER("Task %pK: src buffer setup failed\n", task);
+		DRM_DEV_DEBUG_DRIVER(task->dev,
+				     "Task %pK: src buffer setup failed\n",
+				     task);
 		return ret;
 	}
 	ret = exynos_drm_ipp_task_setup_buffer(dst, filp);
 	if (ret) {
-		DRM_DEBUG_DRIVER("Task %pK: dst buffer setup failed\n", task);
+		DRM_DEV_DEBUG_DRIVER(task->dev,
+				     "Task %pK: dst buffer setup failed\n",
+				     task);
 		return ret;
 	}
 
-	DRM_DEBUG_DRIVER("Task %pK: buffers prepared.\n", task);
+	DRM_DEV_DEBUG_DRIVER(task->dev, "Task %pK: buffers prepared.\n",
+			     task);
 
 	return ret;
 }
@@ -691,7 +703,7 @@ static int exynos_drm_ipp_event_create(struct exynos_drm_ipp_task *task,
 	e->event.base.length = sizeof(e->event);
 	e->event.user_data = user_data;
 
-	ret = drm_event_reserve_init(task->dev, file_priv, &e->base,
+	ret = drm_event_reserve_init(task->ipp->drm_dev, file_priv, &e->base,
 				     &e->event.base);
 	if (ret)
 		goto free;
@@ -712,7 +724,7 @@ static void exynos_drm_ipp_event_send(struct exynos_drm_ipp_task *task)
 	task->event->event.tv_usec = now.tv_nsec / NSEC_PER_USEC;
 	task->event->event.sequence = atomic_inc_return(&task->ipp->sequence);
 
-	drm_send_event(task->dev, &task->event->base);
+	drm_send_event(task->ipp->drm_dev, &task->event->base);
 }
 
 static int exynos_drm_ipp_task_cleanup(struct exynos_drm_ipp_task *task)
@@ -749,7 +761,8 @@ void exynos_drm_ipp_task_done(struct exynos_drm_ipp_task *task, int ret)
 	struct exynos_drm_ipp *ipp = task->ipp;
 	unsigned long flags;
 
-	DRM_DEBUG_DRIVER("ipp: %d, task %pK done: %d\n", ipp->id, task, ret);
+	DRM_DEV_DEBUG_DRIVER(task->dev, "ipp: %d, task %pK done: %d\n",
+			     ipp->id, task, ret);
 
 	spin_lock_irqsave(&ipp->lock, flags);
 	if (ipp->task == task)
@@ -773,7 +786,8 @@ static void exynos_drm_ipp_next_task(struct exynos_drm_ipp *ipp)
 	unsigned long flags;
 	int ret;
 
-	DRM_DEBUG_DRIVER("ipp: %d, try to run new task\n", ipp->id);
+	DRM_DEV_DEBUG_DRIVER(ipp->dev, "ipp: %d, try to run new task\n",
+			     ipp->id);
 
 	spin_lock_irqsave(&ipp->lock, flags);
 
@@ -789,7 +803,9 @@ static void exynos_drm_ipp_next_task(struct exynos_drm_ipp *ipp)
 
 	spin_unlock_irqrestore(&ipp->lock, flags);
 
-	DRM_DEBUG_DRIVER("ipp: %d, selected task %pK to run\n", ipp->id, task);
+	DRM_DEV_DEBUG_DRIVER(ipp->dev,
+			     "ipp: %d, selected task %pK to run\n", ipp->id,
+			     task);
 
 	ret = ipp->funcs->commit(ipp, task);
 	if (ret)
@@ -897,15 +913,16 @@ int exynos_drm_ipp_commit_ioctl(struct drm_device *dev, void *data,
 	 * then freed after exynos_drm_ipp_task_done()
 	 */
 	if (arg->flags & DRM_EXYNOS_IPP_FLAG_NONBLOCK) {
-		DRM_DEBUG_DRIVER("ipp: %d, nonblocking processing task %pK\n",
-				 ipp->id, task);
+		DRM_DEV_DEBUG_DRIVER(ipp->dev,
+				     "ipp: %d, nonblocking processing task %pK\n",
+				     ipp->id, task);
 
 		task->flags |= DRM_EXYNOS_IPP_TASK_ASYNC;
 		exynos_drm_ipp_schedule_task(task->ipp, task);
 		ret = 0;
 	} else {
-		DRM_DEBUG_DRIVER("ipp: %d, processing task %pK\n", ipp->id,
-				 task);
+		DRM_DEV_DEBUG_DRIVER(ipp->dev, "ipp: %d, processing task %pK\n",
+				     ipp->id, task);
 		exynos_drm_ipp_schedule_task(ipp, task);
 		ret = wait_event_interruptible(ipp->done_wq,
 					task->flags & DRM_EXYNOS_IPP_TASK_DONE);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_ipp.h b/drivers/gpu/drm/exynos/exynos_drm_ipp.h
index 0b27d4a9bf94..5524c457a947 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_ipp.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_ipp.h
@@ -54,7 +54,8 @@ struct exynos_drm_ipp_funcs {
  * struct exynos_drm_ipp - central picture processor module structure
  */
 struct exynos_drm_ipp {
-	struct drm_device *dev;
+	struct drm_device *drm_dev;
+	struct device *dev;
 	struct list_head head;
 	unsigned int id;
 
@@ -85,7 +86,7 @@ struct exynos_drm_ipp_buffer {
  * has to be performed by the picture processor hardware module
  */
 struct exynos_drm_ipp_task {
-	struct drm_device *dev;
+	struct device *dev;
 	struct exynos_drm_ipp *ipp;
 	struct list_head head;
 
@@ -129,11 +130,11 @@ struct exynos_drm_ipp_formats {
 #define IPP_SCALE_LIMIT(val...)		\
 	.type = (DRM_EXYNOS_IPP_LIMIT_TYPE_SCALE), val
 
-int exynos_drm_ipp_register(struct drm_device *dev, struct exynos_drm_ipp *ipp,
+int exynos_drm_ipp_register(struct device *dev, struct exynos_drm_ipp *ipp,
 		const struct exynos_drm_ipp_funcs *funcs, unsigned int caps,
 		const struct exynos_drm_ipp_formats *formats,
 		unsigned int num_formats, const char *name);
-void exynos_drm_ipp_unregister(struct drm_device *dev,
+void exynos_drm_ipp_unregister(struct device *dev,
 			       struct exynos_drm_ipp *ipp);
 
 void exynos_drm_ipp_task_done(struct exynos_drm_ipp_task *task, int ret);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_mic.c b/drivers/gpu/drm/exynos/exynos_drm_mic.c
index dd02e8a323ef..d1c8411ae7d4 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_mic.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_mic.c
@@ -113,7 +113,8 @@ static void mic_set_path(struct exynos_mic *mic, bool enable)
 
 	ret = regmap_read(mic->sysreg, DSD_CFG_MUX, &val);
 	if (ret) {
-		DRM_ERROR("mic: Failed to read system register\n");
+		DRM_DEV_ERROR(mic->dev,
+			      "mic: Failed to read system register\n");
 		return;
 	}
 
@@ -129,7 +130,8 @@ static void mic_set_path(struct exynos_mic *mic, bool enable)
 
 	ret = regmap_write(mic->sysreg, DSD_CFG_MUX, val);
 	if (ret)
-		DRM_ERROR("mic: Failed to read system register\n");
+		DRM_DEV_ERROR(mic->dev,
+			      "mic: Failed to read system register\n");
 }
 
 static int mic_sw_reset(struct exynos_mic *mic)
@@ -190,7 +192,7 @@ static void mic_set_output_timing(struct exynos_mic *mic)
 	struct videomode vm = mic->vm;
 	u32 reg, bs_size_2d;
 
-	DRM_DEBUG("w: %u, h: %u\n", vm.hactive, vm.vactive);
+	DRM_DEV_DEBUG(mic->dev, "w: %u, h: %u\n", vm.hactive, vm.vactive);
 	bs_size_2d = ((vm.hactive >> 2) << 1) + (vm.vactive % 4);
 	reg = MIC_BS_SIZE_2D(bs_size_2d);
 	writel(reg, mic->reg + MIC_2D_OUTPUT_TIMING_2);
@@ -274,7 +276,7 @@ static void mic_pre_enable(struct drm_bridge *bridge)
 
 	ret = mic_sw_reset(mic);
 	if (ret) {
-		DRM_ERROR("Failed to reset\n");
+		DRM_DEV_ERROR(mic->dev, "Failed to reset\n");
 		goto turn_off;
 	}
 
@@ -354,8 +356,8 @@ static int exynos_mic_resume(struct device *dev)
 	for (i = 0; i < NUM_CLKS; i++) {
 		ret = clk_prepare_enable(mic->clks[i]);
 		if (ret < 0) {
-			DRM_ERROR("Failed to enable clock (%s)\n",
-							clk_names[i]);
+			DRM_DEV_ERROR(dev, "Failed to enable clock (%s)\n",
+				      clk_names[i]);
 			while (--i > -1)
 				clk_disable_unprepare(mic->clks[i]);
 			return ret;
@@ -380,7 +382,8 @@ static int exynos_mic_probe(struct platform_device *pdev)
 
 	mic = devm_kzalloc(dev, sizeof(*mic), GFP_KERNEL);
 	if (!mic) {
-		DRM_ERROR("mic: Failed to allocate memory for MIC object\n");
+		DRM_DEV_ERROR(dev,
+			      "mic: Failed to allocate memory for MIC object\n");
 		ret = -ENOMEM;
 		goto err;
 	}
@@ -389,12 +392,12 @@ static int exynos_mic_probe(struct platform_device *pdev)
 
 	ret = of_address_to_resource(dev->of_node, 0, &res);
 	if (ret) {
-		DRM_ERROR("mic: Failed to get mem region for MIC\n");
+		DRM_DEV_ERROR(dev, "mic: Failed to get mem region for MIC\n");
 		goto err;
 	}
 	mic->reg = devm_ioremap(dev, res.start, resource_size(&res));
 	if (!mic->reg) {
-		DRM_ERROR("mic: Failed to remap for MIC\n");
+		DRM_DEV_ERROR(dev, "mic: Failed to remap for MIC\n");
 		ret = -ENOMEM;
 		goto err;
 	}
@@ -402,7 +405,7 @@ static int exynos_mic_probe(struct platform_device *pdev)
 	mic->sysreg = syscon_regmap_lookup_by_phandle(dev->of_node,
 							"samsung,disp-syscon");
 	if (IS_ERR(mic->sysreg)) {
-		DRM_ERROR("mic: Failed to get system register.\n");
+		DRM_DEV_ERROR(dev, "mic: Failed to get system register.\n");
 		ret = PTR_ERR(mic->sysreg);
 		goto err;
 	}
@@ -410,8 +413,8 @@ static int exynos_mic_probe(struct platform_device *pdev)
 	for (i = 0; i < NUM_CLKS; i++) {
 		mic->clks[i] = devm_clk_get(dev, clk_names[i]);
 		if (IS_ERR(mic->clks[i])) {
-			DRM_ERROR("mic: Failed to get clock (%s)\n",
-								clk_names[i]);
+			DRM_DEV_ERROR(dev, "mic: Failed to get clock (%s)\n",
+				      clk_names[i]);
 			ret = PTR_ERR(mic->clks[i]);
 			goto err;
 		}
@@ -430,7 +433,7 @@ static int exynos_mic_probe(struct platform_device *pdev)
 	if (ret)
 		goto err_pm;
 
-	DRM_DEBUG_KMS("MIC has been probed\n");
+	DRM_DEV_DEBUG_KMS(dev, "MIC has been probed\n");
 
 	return 0;
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_plane.c b/drivers/gpu/drm/exynos/exynos_drm_plane.c
index df0508e0e49e..e18babb25170 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_plane.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_plane.c
@@ -119,9 +119,10 @@ static void exynos_plane_mode_set(struct exynos_drm_plane_state *exynos_state)
 	exynos_state->crtc.w = actual_w;
 	exynos_state->crtc.h = actual_h;
 
-	DRM_DEBUG_KMS("plane : offset_x/y(%d,%d), width/height(%d,%d)",
-			exynos_state->crtc.x, exynos_state->crtc.y,
-			exynos_state->crtc.w, exynos_state->crtc.h);
+	DRM_DEV_DEBUG_KMS(crtc->dev->dev,
+			  "plane : offset_x/y(%d,%d), width/height(%d,%d)",
+			  exynos_state->crtc.x, exynos_state->crtc.y,
+			  exynos_state->crtc.w, exynos_state->crtc.h);
 }
 
 static void exynos_drm_plane_reset(struct drm_plane *plane)
@@ -181,6 +182,7 @@ exynos_drm_plane_check_format(const struct exynos_drm_plane_config *config,
 			      struct exynos_drm_plane_state *state)
 {
 	struct drm_framebuffer *fb = state->base.fb;
+	struct drm_device *dev = fb->dev;
 
 	switch (fb->modifier) {
 	case DRM_FORMAT_MOD_SAMSUNG_64_32_TILE:
@@ -192,7 +194,7 @@ exynos_drm_plane_check_format(const struct exynos_drm_plane_config *config,
 		break;
 
 	default:
-		DRM_ERROR("unsupported pixel format modifier");
+		DRM_DEV_ERROR(dev->dev, "unsupported pixel format modifier");
 		return -ENOTSUPP;
 	}
 
@@ -203,6 +205,7 @@ static int
 exynos_drm_plane_check_size(const struct exynos_drm_plane_config *config,
 			    struct exynos_drm_plane_state *state)
 {
+	struct drm_crtc *crtc = state->base.crtc;
 	bool width_ok = false, height_ok = false;
 
 	if (config->capabilities & EXYNOS_DRM_PLANE_CAP_SCALE)
@@ -225,7 +228,7 @@ exynos_drm_plane_check_size(const struct exynos_drm_plane_config *config,
 	if (width_ok && height_ok)
 		return 0;
 
-	DRM_DEBUG_KMS("scaling mode is not supported");
+	DRM_DEV_DEBUG_KMS(crtc->dev->dev, "scaling mode is not supported");
 	return -ENOTSUPP;
 }
 
@@ -310,7 +313,7 @@ int exynos_plane_init(struct drm_device *dev,
 				       config->num_pixel_formats,
 				       NULL, config->type, NULL);
 	if (err) {
-		DRM_ERROR("failed to initialize plane\n");
+		DRM_DEV_ERROR(dev->dev, "failed to initialize plane\n");
 		return err;
 	}
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_rotator.c b/drivers/gpu/drm/exynos/exynos_drm_rotator.c
index 05abfed6f7f8..b6586fa95ad1 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_rotator.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_rotator.c
@@ -243,9 +243,10 @@ static int rotator_bind(struct device *dev, struct device *master, void *data)
 	struct exynos_drm_ipp *ipp = &rot->ipp;
 
 	rot->drm_dev = drm_dev;
+	ipp->drm_dev = drm_dev;
 	exynos_drm_register_dma(drm_dev, dev);
 
-	exynos_drm_ipp_register(drm_dev, ipp, &ipp_funcs,
+	exynos_drm_ipp_register(dev, ipp, &ipp_funcs,
 			   DRM_EXYNOS_IPP_CAP_CROP | DRM_EXYNOS_IPP_CAP_ROTATE,
 			   rot->formats, rot->num_formats, "rotator");
 
@@ -258,10 +259,9 @@ static void rotator_unbind(struct device *dev, struct device *master,
 			void *data)
 {
 	struct rot_context *rot = dev_get_drvdata(dev);
-	struct drm_device *drm_dev = data;
 	struct exynos_drm_ipp *ipp = &rot->ipp;
 
-	exynos_drm_ipp_unregister(drm_dev, ipp);
+	exynos_drm_ipp_unregister(dev, ipp);
 	exynos_drm_unregister_dma(rot->drm_dev, rot->dev);
 }
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_scaler.c b/drivers/gpu/drm/exynos/exynos_drm_scaler.c
index ed1dd1aec902..f1cbdd1e6e3c 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_scaler.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_scaler.c
@@ -451,9 +451,10 @@ static int scaler_bind(struct device *dev, struct device *master, void *data)
 	struct exynos_drm_ipp *ipp = &scaler->ipp;
 
 	scaler->drm_dev = drm_dev;
+	ipp->drm_dev = drm_dev;
 	exynos_drm_register_dma(drm_dev, dev);
 
-	exynos_drm_ipp_register(drm_dev, ipp, &ipp_funcs,
+	exynos_drm_ipp_register(dev, ipp, &ipp_funcs,
 			DRM_EXYNOS_IPP_CAP_CROP | DRM_EXYNOS_IPP_CAP_ROTATE |
 			DRM_EXYNOS_IPP_CAP_SCALE | DRM_EXYNOS_IPP_CAP_CONVERT,
 			scaler->scaler_data->formats,
@@ -468,10 +469,9 @@ static void scaler_unbind(struct device *dev, struct device *master,
 			void *data)
 {
 	struct scaler_context *scaler = dev_get_drvdata(dev);
-	struct drm_device *drm_dev = data;
 	struct exynos_drm_ipp *ipp = &scaler->ipp;
 
-	exynos_drm_ipp_unregister(drm_dev, ipp);
+	exynos_drm_ipp_unregister(dev, ipp);
 	exynos_drm_unregister_dma(scaler->drm_dev, scaler->dev);
 }
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_vidi.c b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
index 29f4c1932aed..44bcb2d60bb2 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_vidi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
@@ -40,8 +40,8 @@
 
 struct vidi_context {
 	struct drm_encoder		encoder;
-	struct platform_device		*pdev;
 	struct drm_device		*drm_dev;
+	struct device			*dev;
 	struct exynos_drm_crtc		*crtc;
 	struct drm_connector		connector;
 	struct exynos_drm_plane		planes[WINDOWS_NR];
@@ -123,7 +123,7 @@ static void vidi_update_plane(struct exynos_drm_crtc *crtc,
 		return;
 
 	addr = exynos_drm_fb_dma_addr(state->fb, 0);
-	DRM_DEBUG_KMS("dma_addr = %pad\n", &addr);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "dma_addr = %pad\n", &addr);
 }
 
 static void vidi_enable(struct exynos_drm_crtc *crtc)
@@ -205,11 +205,11 @@ static ssize_t vidi_store_connection(struct device *dev,
 
 	/* if raw_edid isn't same as fake data then it can't be tested. */
 	if (ctx->raw_edid != (struct edid *)fake_edid_info) {
-		DRM_DEBUG_KMS("edid data is not fake data.\n");
+		DRM_DEV_DEBUG_KMS(dev, "edid data is not fake data.\n");
 		return -EINVAL;
 	}
 
-	DRM_DEBUG_KMS("requested connection.\n");
+	DRM_DEV_DEBUG_KMS(dev, "requested connection.\n");
 
 	drm_helper_hpd_irq_event(ctx->drm_dev);
 
@@ -226,17 +226,20 @@ int vidi_connection_ioctl(struct drm_device *drm_dev, void *data,
 	struct drm_exynos_vidi_connection *vidi = data;
 
 	if (!vidi) {
-		DRM_DEBUG_KMS("user data for vidi is null.\n");
+		DRM_DEV_DEBUG_KMS(ctx->dev,
+				  "user data for vidi is null.\n");
 		return -EINVAL;
 	}
 
 	if (vidi->connection > 1) {
-		DRM_DEBUG_KMS("connection should be 0 or 1.\n");
+		DRM_DEV_DEBUG_KMS(ctx->dev,
+				  "connection should be 0 or 1.\n");
 		return -EINVAL;
 	}
 
 	if (ctx->connected == vidi->connection) {
-		DRM_DEBUG_KMS("same connection request.\n");
+		DRM_DEV_DEBUG_KMS(ctx->dev,
+				  "same connection request.\n");
 		return -EINVAL;
 	}
 
@@ -245,12 +248,14 @@ int vidi_connection_ioctl(struct drm_device *drm_dev, void *data,
 
 		raw_edid = (struct edid *)(unsigned long)vidi->edid;
 		if (!drm_edid_is_valid(raw_edid)) {
-			DRM_DEBUG_KMS("edid data is invalid.\n");
+			DRM_DEV_DEBUG_KMS(ctx->dev,
+					  "edid data is invalid.\n");
 			return -EINVAL;
 		}
 		ctx->raw_edid = drm_edid_duplicate(raw_edid);
 		if (!ctx->raw_edid) {
-			DRM_DEBUG_KMS("failed to allocate raw_edid.\n");
+			DRM_DEV_DEBUG_KMS(ctx->dev,
+					  "failed to allocate raw_edid.\n");
 			return -ENOMEM;
 		}
 	} else {
@@ -308,14 +313,14 @@ static int vidi_get_modes(struct drm_connector *connector)
 	 * to ctx->raw_edid through specific ioctl.
 	 */
 	if (!ctx->raw_edid) {
-		DRM_DEBUG_KMS("raw_edid is null.\n");
+		DRM_DEV_DEBUG_KMS(ctx->dev, "raw_edid is null.\n");
 		return -EFAULT;
 	}
 
 	edid_len = (1 + ctx->raw_edid->extensions) * EDID_LENGTH;
 	edid = kmemdup(ctx->raw_edid, edid_len, GFP_KERNEL);
 	if (!edid) {
-		DRM_DEBUG_KMS("failed to allocate edid\n");
+		DRM_DEV_DEBUG_KMS(ctx->dev, "failed to allocate edid\n");
 		return -ENOMEM;
 	}
 
@@ -339,7 +344,8 @@ static int vidi_create_connector(struct drm_encoder *encoder)
 	ret = drm_connector_init(ctx->drm_dev, connector,
 			&vidi_connector_funcs, DRM_MODE_CONNECTOR_VIRTUAL);
 	if (ret) {
-		DRM_ERROR("Failed to initialize connector with drm\n");
+		DRM_DEV_ERROR(ctx->dev,
+			      "Failed to initialize connector with drm\n");
 		return ret;
 	}
 
@@ -402,7 +408,7 @@ static int vidi_bind(struct device *dev, struct device *master, void *data)
 	ctx->crtc = exynos_drm_crtc_create(drm_dev, &exynos_plane->base,
 			EXYNOS_DISPLAY_TYPE_VIDI, &vidi_crtc_ops, ctx);
 	if (IS_ERR(ctx->crtc)) {
-		DRM_ERROR("failed to create crtc.\n");
+		DRM_DEV_ERROR(dev, "failed to create crtc.\n");
 		return PTR_ERR(ctx->crtc);
 	}
 
@@ -417,7 +423,8 @@ static int vidi_bind(struct device *dev, struct device *master, void *data)
 
 	ret = vidi_create_connector(encoder);
 	if (ret) {
-		DRM_ERROR("failed to create connector ret = %d\n", ret);
+		DRM_DEV_ERROR(dev, "failed to create connector ret = %d\n",
+			      ret);
 		drm_encoder_cleanup(encoder);
 		return ret;
 	}
@@ -441,13 +448,14 @@ static const struct component_ops vidi_component_ops = {
 static int vidi_probe(struct platform_device *pdev)
 {
 	struct vidi_context *ctx;
+	struct device *dev = &pdev->dev;
 	int ret;
 
-	ctx = devm_kzalloc(&pdev->dev, sizeof(*ctx), GFP_KERNEL);
+	ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL);
 	if (!ctx)
 		return -ENOMEM;
 
-	ctx->pdev = pdev;
+	ctx->dev = dev;
 
 	timer_setup(&ctx->timer, vidi_fake_vblank_timer, 0);
 
@@ -455,20 +463,21 @@ static int vidi_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, ctx);
 
-	ret = device_create_file(&pdev->dev, &dev_attr_connection);
+	ret = device_create_file(dev, &dev_attr_connection);
 	if (ret < 0) {
-		DRM_ERROR("failed to create connection sysfs.\n");
+		DRM_DEV_ERROR(dev,
+			      "failed to create connection sysfs.\n");
 		return ret;
 	}
 
-	ret = component_add(&pdev->dev, &vidi_component_ops);
+	ret = component_add(dev, &vidi_component_ops);
 	if (ret)
 		goto err_remove_file;
 
 	return ret;
 
 err_remove_file:
-	device_remove_file(&pdev->dev, &dev_attr_connection);
+	device_remove_file(dev, &dev_attr_connection);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index 8e2c02fc66e8..19c252f659dd 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -885,9 +885,9 @@ static int hdmi_get_modes(struct drm_connector *connector)
 		return -ENODEV;
 
 	hdata->dvi_mode = !drm_detect_hdmi_monitor(edid);
-	DRM_DEBUG_KMS("%s : width[%d] x height[%d]\n",
-		(hdata->dvi_mode ? "dvi monitor" : "hdmi monitor"),
-		edid->width_cm, edid->height_cm);
+	DRM_DEV_DEBUG_KMS(hdata->dev, "%s : width[%d] x height[%d]\n",
+			  (hdata->dvi_mode ? "dvi monitor" : "hdmi monitor"),
+			  edid->width_cm, edid->height_cm);
 
 	drm_connector_update_edid_property(connector, edid);
 	cec_notifier_set_phys_addr_from_edid(hdata->notifier, edid);
@@ -908,7 +908,8 @@ static int hdmi_find_phy_conf(struct hdmi_context *hdata, u32 pixel_clock)
 		if (confs->data[i].pixel_clock == pixel_clock)
 			return i;
 
-	DRM_DEBUG_KMS("Could not find phy config for %d\n", pixel_clock);
+	DRM_DEV_DEBUG_KMS(hdata->dev, "Could not find phy config for %d\n",
+			  pixel_clock);
 	return -EINVAL;
 }
 
@@ -918,10 +919,11 @@ static int hdmi_mode_valid(struct drm_connector *connector,
 	struct hdmi_context *hdata = connector_to_hdmi(connector);
 	int ret;
 
-	DRM_DEBUG_KMS("xres=%d, yres=%d, refresh=%d, intl=%d clock=%d\n",
-		mode->hdisplay, mode->vdisplay, mode->vrefresh,
-		(mode->flags & DRM_MODE_FLAG_INTERLACE) ? true :
-		false, mode->clock * 1000);
+	DRM_DEV_DEBUG_KMS(hdata->dev,
+			  "xres=%d, yres=%d, refresh=%d, intl=%d clock=%d\n",
+			  mode->hdisplay, mode->vdisplay, mode->vrefresh,
+			  (mode->flags & DRM_MODE_FLAG_INTERLACE) ? true :
+			  false, mode->clock * 1000);
 
 	ret = hdmi_find_phy_conf(hdata, mode->clock * 1000);
 	if (ret < 0)
@@ -947,7 +949,8 @@ static int hdmi_create_connector(struct drm_encoder *encoder)
 	ret = drm_connector_init(hdata->drm_dev, connector,
 			&hdmi_connector_funcs, DRM_MODE_CONNECTOR_HDMIA);
 	if (ret) {
-		DRM_ERROR("Failed to initialize connector with drm\n");
+		DRM_DEV_ERROR(hdata->dev,
+			      "Failed to initialize connector with drm\n");
 		return ret;
 	}
 
@@ -957,7 +960,7 @@ static int hdmi_create_connector(struct drm_encoder *encoder)
 	if (hdata->bridge) {
 		ret = drm_bridge_attach(encoder, hdata->bridge, NULL);
 		if (ret)
-			DRM_ERROR("Failed to attach bridge\n");
+			DRM_DEV_ERROR(hdata->dev, "Failed to attach bridge\n");
 	}
 
 	return ret;
@@ -1002,8 +1005,10 @@ static bool hdmi_mode_fixup(struct drm_encoder *encoder,
 			DRM_INFO("desired mode doesn't exist so\n");
 			DRM_INFO("use the most suitable mode among modes.\n");
 
-			DRM_DEBUG_KMS("Adjusted Mode: [%d]x[%d] [%d]Hz\n",
-				m->hdisplay, m->vdisplay, m->vrefresh);
+			DRM_DEV_DEBUG_KMS(dev->dev,
+					  "Adjusted Mode: [%d]x[%d] [%d]Hz\n",
+					  m->hdisplay, m->vdisplay,
+					  m->vrefresh);
 
 			drm_mode_copy(adjusted_mode, m);
 			break;
@@ -1169,13 +1174,15 @@ static void hdmiphy_wait_for_pll(struct hdmi_context *hdata)
 		u32 val = hdmi_reg_read(hdata, HDMI_PHY_STATUS);
 
 		if (val & HDMI_PHY_STATUS_READY) {
-			DRM_DEBUG_KMS("PLL stabilized after %d tries\n", tries);
+			DRM_DEV_DEBUG_KMS(hdata->dev,
+					  "PLL stabilized after %d tries\n",
+					  tries);
 			return;
 		}
 		usleep_range(10, 20);
 	}
 
-	DRM_ERROR("PLL could not reach steady state\n");
+	DRM_DEV_ERROR(hdata->dev, "PLL could not reach steady state\n");
 }
 
 static void hdmi_v13_mode_apply(struct hdmi_context *hdata)
@@ -1411,7 +1418,7 @@ static void hdmiphy_conf_apply(struct hdmi_context *hdata)
 
 	ret = hdmi_find_phy_conf(hdata, m->clock * 1000);
 	if (ret < 0) {
-		DRM_ERROR("failed to find hdmiphy conf\n");
+		DRM_DEV_ERROR(hdata->dev, "failed to find hdmiphy conf\n");
 		return;
 	}
 	phy_conf = hdata->drv_data->phy_confs.data[ret].conf;
@@ -1423,7 +1430,7 @@ static void hdmiphy_conf_apply(struct hdmi_context *hdata)
 	hdmiphy_enable_mode_set(hdata, true);
 	ret = hdmiphy_reg_write_buf(hdata, 0, phy_conf, 32);
 	if (ret) {
-		DRM_ERROR("failed to configure hdmiphy\n");
+		DRM_DEV_ERROR(hdata->dev, "failed to configure hdmiphy\n");
 		return;
 	}
 	hdmiphy_enable_mode_set(hdata, false);
@@ -1460,7 +1467,8 @@ static void hdmiphy_enable(struct hdmi_context *hdata)
 	pm_runtime_get_sync(hdata->dev);
 
 	if (regulator_bulk_enable(ARRAY_SIZE(supply), hdata->regul_bulk))
-		DRM_DEBUG_KMS("failed to enable regulator bulk\n");
+		DRM_DEV_DEBUG_KMS(hdata->dev,
+				  "failed to enable regulator bulk\n");
 
 	regmap_update_bits(hdata->pmureg, PMU_HDMI_PHY_CONTROL,
 			PMU_HDMI_PHY_ENABLE_BIT, 1);
@@ -1734,7 +1742,7 @@ static int hdmi_bridge_init(struct hdmi_context *hdata)
 	np = of_graph_get_remote_port_parent(ep);
 	of_node_put(ep);
 	if (!np) {
-		DRM_ERROR("failed to get remote port parent");
+		DRM_DEV_ERROR(dev, "failed to get remote port parent");
 		return -EINVAL;
 	}
 
@@ -1752,17 +1760,17 @@ static int hdmi_resources_init(struct hdmi_context *hdata)
 	struct device *dev = hdata->dev;
 	int i, ret;
 
-	DRM_DEBUG_KMS("HDMI resource init\n");
+	DRM_DEV_DEBUG_KMS(dev, "HDMI resource init\n");
 
 	hdata->hpd_gpio = devm_gpiod_get(dev, "hpd", GPIOD_IN);
 	if (IS_ERR(hdata->hpd_gpio)) {
-		DRM_ERROR("cannot get hpd gpio property\n");
+		DRM_DEV_ERROR(dev, "cannot get hpd gpio property\n");
 		return PTR_ERR(hdata->hpd_gpio);
 	}
 
 	hdata->irq = gpiod_to_irq(hdata->hpd_gpio);
 	if (hdata->irq < 0) {
-		DRM_ERROR("failed to get GPIO irq\n");
+		DRM_DEV_ERROR(dev, "failed to get GPIO irq\n");
 		return  hdata->irq;
 	}
 
@@ -1780,7 +1788,7 @@ static int hdmi_resources_init(struct hdmi_context *hdata)
 	ret = devm_regulator_bulk_get(dev, ARRAY_SIZE(supply), hdata->regul_bulk);
 	if (ret) {
 		if (ret != -EPROBE_DEFER)
-			DRM_ERROR("failed to get regulators\n");
+			DRM_DEV_ERROR(dev, "failed to get regulators\n");
 		return ret;
 	}
 
@@ -1792,7 +1800,8 @@ static int hdmi_resources_init(struct hdmi_context *hdata)
 
 		ret = regulator_enable(hdata->reg_hdmi_en);
 		if (ret) {
-			DRM_ERROR("failed to enable hdmi-en regulator\n");
+			DRM_DEV_ERROR(dev,
+				      "failed to enable hdmi-en regulator\n");
 			return ret;
 		}
 	}
@@ -1845,7 +1854,8 @@ static int hdmi_bind(struct device *dev, struct device *master, void *data)
 
 	ret = hdmi_create_connector(encoder);
 	if (ret) {
-		DRM_ERROR("failed to create connector ret = %d\n", ret);
+		DRM_DEV_ERROR(dev, "failed to create connector ret = %d\n",
+			      ret);
 		drm_encoder_cleanup(encoder);
 		return ret;
 	}
@@ -1875,7 +1885,8 @@ static int hdmi_get_ddc_adapter(struct hdmi_context *hdata)
 		np = of_parse_phandle(hdata->dev->of_node, "ddc", 0);
 
 	if (!np) {
-		DRM_ERROR("Failed to find ddc node in device tree\n");
+		DRM_DEV_ERROR(hdata->dev,
+			      "Failed to find ddc node in device tree\n");
 		return -ENODEV;
 	}
 
@@ -1902,7 +1913,8 @@ static int hdmi_get_phy_io(struct hdmi_context *hdata)
 	if (!np) {
 		np = of_parse_phandle(hdata->dev->of_node, "phy", 0);
 		if (!np) {
-			DRM_ERROR("Failed to find hdmiphy node in device tree\n");
+			DRM_DEV_ERROR(hdata->dev,
+				      "Failed to find hdmiphy node in device tree\n");
 			return -ENODEV;
 		}
 	}
@@ -1910,7 +1922,8 @@ static int hdmi_get_phy_io(struct hdmi_context *hdata)
 	if (hdata->drv_data->is_apb_phy) {
 		hdata->regs_hdmiphy = of_iomap(np, 0);
 		if (!hdata->regs_hdmiphy) {
-			DRM_ERROR("failed to ioremap hdmi phy\n");
+			DRM_DEV_ERROR(hdata->dev,
+				      "failed to ioremap hdmi phy\n");
 			ret = -ENOMEM;
 			goto out;
 		}
@@ -1951,7 +1964,7 @@ static int hdmi_probe(struct platform_device *pdev)
 	ret = hdmi_resources_init(hdata);
 	if (ret) {
 		if (ret != -EPROBE_DEFER)
-			DRM_ERROR("hdmi_resources_init failed\n");
+			DRM_DEV_ERROR(dev, "hdmi_resources_init failed\n");
 		return ret;
 	}
 
@@ -1977,14 +1990,14 @@ static int hdmi_probe(struct platform_device *pdev)
 			IRQF_TRIGGER_FALLING | IRQF_ONESHOT,
 			"hdmi", hdata);
 	if (ret) {
-		DRM_ERROR("failed to register hdmi interrupt\n");
+		DRM_DEV_ERROR(dev, "failed to register hdmi interrupt\n");
 		goto err_hdmiphy;
 	}
 
 	hdata->pmureg = syscon_regmap_lookup_by_phandle(dev->of_node,
 			"samsung,syscon-phandle");
 	if (IS_ERR(hdata->pmureg)) {
-		DRM_ERROR("syscon regmap lookup failed.\n");
+		DRM_DEV_ERROR(dev, "syscon regmap lookup failed.\n");
 		ret = -EPROBE_DEFER;
 		goto err_hdmiphy;
 	}
@@ -1993,7 +2006,7 @@ static int hdmi_probe(struct platform_device *pdev)
 		hdata->sysreg = syscon_regmap_lookup_by_phandle(dev->of_node,
 				"samsung,sysreg-phandle");
 		if (IS_ERR(hdata->sysreg)) {
-			DRM_ERROR("sysreg regmap lookup failed.\n");
+			DRM_DEV_ERROR(dev, "sysreg regmap lookup failed.\n");
 			ret = -EPROBE_DEFER;
 			goto err_hdmiphy;
 		}
diff --git a/drivers/gpu/drm/exynos/exynos_mixer.c b/drivers/gpu/drm/exynos/exynos_mixer.c
index f35e4ab55b27..b8415e53964d 100644
--- a/drivers/gpu/drm/exynos/exynos_mixer.c
+++ b/drivers/gpu/drm/exynos/exynos_mixer.c
@@ -228,8 +228,8 @@ static void mixer_regs_dump(struct mixer_context *ctx)
 {
 #define DUMPREG(reg_id) \
 do { \
-	DRM_DEBUG_KMS(#reg_id " = %08x\n", \
-		(u32)readl(ctx->mixer_regs + reg_id)); \
+	DRM_DEV_DEBUG_KMS(ctx->dev, #reg_id " = %08x\n", \
+			 (u32)readl(ctx->mixer_regs + reg_id)); \
 } while (0)
 
 	DUMPREG(MXR_STATUS);
@@ -260,8 +260,8 @@ static void vp_regs_dump(struct mixer_context *ctx)
 {
 #define DUMPREG(reg_id) \
 do { \
-	DRM_DEBUG_KMS(#reg_id " = %08x\n", \
-		(u32) readl(ctx->vp_regs + reg_id)); \
+	DRM_DEV_DEBUG_KMS(ctx->dev, #reg_id " = %08x\n", \
+			 (u32) readl(ctx->vp_regs + reg_id)); \
 } while (0)
 
 	DUMPREG(VP_ENABLE);
@@ -885,7 +885,8 @@ static int mixer_initialize(struct mixer_context *mixer_ctx,
 	/* acquire resources: regs, irqs, clocks */
 	ret = mixer_resources_init(mixer_ctx);
 	if (ret) {
-		DRM_ERROR("mixer_resources_init failed ret=%d\n", ret);
+		DRM_DEV_ERROR(mixer_ctx->dev,
+			      "mixer_resources_init failed ret=%d\n", ret);
 		return ret;
 	}
 
@@ -893,7 +894,8 @@ static int mixer_initialize(struct mixer_context *mixer_ctx,
 		/* acquire vp resources: regs, irqs, clocks */
 		ret = vp_resources_init(mixer_ctx);
 		if (ret) {
-			DRM_ERROR("vp_resources_init failed ret=%d\n", ret);
+			DRM_DEV_ERROR(mixer_ctx->dev,
+				      "vp_resources_init failed ret=%d\n", ret);
 			return ret;
 		}
 	}
@@ -952,7 +954,7 @@ static void mixer_update_plane(struct exynos_drm_crtc *crtc,
 {
 	struct mixer_context *mixer_ctx = crtc->ctx;
 
-	DRM_DEBUG_KMS("win: %d\n", plane->index);
+	DRM_DEV_DEBUG_KMS(mixer_ctx->dev, "win: %d\n", plane->index);
 
 	if (!test_bit(MXR_BIT_POWERED, &mixer_ctx->flags))
 		return;
@@ -969,7 +971,7 @@ static void mixer_disable_plane(struct exynos_drm_crtc *crtc,
 	struct mixer_context *mixer_ctx = crtc->ctx;
 	unsigned long flags;
 
-	DRM_DEBUG_KMS("win: %d\n", plane->index);
+	DRM_DEV_DEBUG_KMS(mixer_ctx->dev, "win: %d\n", plane->index);
 
 	if (!test_bit(MXR_BIT_POWERED, &mixer_ctx->flags))
 		return;
@@ -1046,8 +1048,9 @@ static int mixer_mode_valid(struct exynos_drm_crtc *crtc,
 	struct mixer_context *ctx = crtc->ctx;
 	u32 w = mode->hdisplay, h = mode->vdisplay;
 
-	DRM_DEBUG_KMS("xres=%d, yres=%d, refresh=%d, intl=%d\n", w, h,
-		mode->vrefresh, !!(mode->flags & DRM_MODE_FLAG_INTERLACE));
+	DRM_DEV_DEBUG_KMS(ctx->dev, "xres=%d, yres=%d, refresh=%d, intl=%d\n",
+			  w, h, mode->vrefresh,
+			  !!(mode->flags & DRM_MODE_FLAG_INTERLACE));
 
 	if (ctx->mxr_ver == MXR_VER_128_0_0_184)
 		return MODE_OK;
@@ -1227,7 +1230,7 @@ static int mixer_probe(struct platform_device *pdev)
 
 	ctx = devm_kzalloc(&pdev->dev, sizeof(*ctx), GFP_KERNEL);
 	if (!ctx) {
-		DRM_ERROR("failed to alloc mixer context.\n");
+		DRM_DEV_ERROR(dev, "failed to alloc mixer context.\n");
 		return -ENOMEM;
 	}
 
@@ -1282,27 +1285,33 @@ static int __maybe_unused exynos_mixer_resume(struct device *dev)
 
 	ret = clk_prepare_enable(ctx->mixer);
 	if (ret < 0) {
-		DRM_ERROR("Failed to prepare_enable the mixer clk [%d]\n", ret);
+		DRM_DEV_ERROR(ctx->dev,
+			      "Failed to prepare_enable the mixer clk [%d]\n",
+			      ret);
 		return ret;
 	}
 	ret = clk_prepare_enable(ctx->hdmi);
 	if (ret < 0) {
-		DRM_ERROR("Failed to prepare_enable the hdmi clk [%d]\n", ret);
+		DRM_DEV_ERROR(dev,
+			      "Failed to prepare_enable the hdmi clk [%d]\n",
+			      ret);
 		return ret;
 	}
 	if (test_bit(MXR_BIT_VP_ENABLED, &ctx->flags)) {
 		ret = clk_prepare_enable(ctx->vp);
 		if (ret < 0) {
-			DRM_ERROR("Failed to prepare_enable the vp clk [%d]\n",
-				  ret);
+			DRM_DEV_ERROR(dev,
+				      "Failed to prepare_enable the vp clk [%d]\n",
+				      ret);
 			return ret;
 		}
 		if (test_bit(MXR_BIT_HAS_SCLK, &ctx->flags)) {
 			ret = clk_prepare_enable(ctx->sclk_mixer);
 			if (ret < 0) {
-				DRM_ERROR("Failed to prepare_enable the " \
+				DRM_DEV_ERROR(dev,
+					   "Failed to prepare_enable the " \
 					   "sclk_mixer clk [%d]\n",
-					  ret);
+					   ret);
 				return ret;
 			}
 		}
diff --git a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_crtc.c b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_crtc.c
index bf256971063d..83c841b50272 100644
--- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_crtc.c
+++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_crtc.c
@@ -94,7 +94,7 @@ static void fsl_dcu_drm_crtc_mode_set_nofb(struct drm_crtc *crtc)
 	drm_display_mode_to_videomode(mode, &vm);
 
 	/* INV_PXCK as default (most display sample data on rising edge) */
-	if (!(con->display_info.bus_flags & DRM_BUS_FLAG_PIXDATA_POSEDGE))
+	if (!(con->display_info.bus_flags & DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE))
 		pol |= DCU_SYN_POL_INV_PXCK;
 
 	if (vm.flags & DISPLAY_FLAGS_HSYNC_LOW)
diff --git a/drivers/gpu/drm/gma500/framebuffer.c b/drivers/gpu/drm/gma500/framebuffer.c
index c934b3df1f81..a9d3a4a30ab8 100644
--- a/drivers/gpu/drm/gma500/framebuffer.c
+++ b/drivers/gpu/drm/gma500/framebuffer.c
@@ -389,7 +389,6 @@ static int psbfb_create(struct psb_fbdev *fbdev,
 		ret = PTR_ERR(info);
 		goto out;
 	}
-	info->par = fbdev;
 
 	mode_cmd.pixel_format = drm_mode_legacy_fb_format(bpp, depth);
 
@@ -402,9 +401,6 @@ static int psbfb_create(struct psb_fbdev *fbdev,
 
 	fbdev->psb_fb_helper.fb = fb;
 
-	drm_fb_helper_fill_fix(info, fb->pitches[0], fb->format->depth);
-	strcpy(info->fix.id, "psbdrmfb");
-
 	if (dev_priv->ops->accel_2d && pitch_lines > 8)	/* 2D engine */
 		info->fbops = &psbfb_ops;
 	else if (gtt_roll) {	/* GTT rolling seems best */
@@ -427,8 +423,7 @@ static int psbfb_create(struct psb_fbdev *fbdev,
 		info->apertures->ranges[0].size = dev_priv->gtt.stolen_size;
 	}
 
-	drm_fb_helper_fill_var(info, &fbdev->psb_fb_helper,
-				sizes->fb_width, sizes->fb_height);
+	drm_fb_helper_fill_info(info, &fbdev->psb_fb_helper, sizes);
 
 	info->fix.mmio_start = pci_resource_start(dev->pdev, 0);
 	info->fix.mmio_len = pci_resource_len(dev->pdev, 0);
diff --git a/drivers/gpu/drm/gma500/framebuffer.h b/drivers/gpu/drm/gma500/framebuffer.h
index 23dc3c5f8f0d..e8e6357f033b 100644
--- a/drivers/gpu/drm/gma500/framebuffer.h
+++ b/drivers/gpu/drm/gma500/framebuffer.h
@@ -34,7 +34,7 @@ struct psb_framebuffer {
 };
 
 struct psb_fbdev {
-	struct drm_fb_helper psb_fb_helper;
+	struct drm_fb_helper psb_fb_helper; /* must be first */
 	struct psb_framebuffer pfb;
 };
 
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h
index 3c168ae77b0c..0a381c22de26 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h
@@ -31,7 +31,7 @@ struct hibmc_framebuffer {
 };
 
 struct hibmc_fbdev {
-	struct drm_fb_helper helper;
+	struct drm_fb_helper helper; /* must be first */
 	struct hibmc_framebuffer *fb;
 	int size;
 };
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_fbdev.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_fbdev.c
index de9d7cc97e44..8026859aa07d 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_fbdev.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_fbdev.c
@@ -116,8 +116,6 @@ static int hibmc_drm_fb_create(struct drm_fb_helper *helper,
 		goto out_release_fbi;
 	}
 
-	info->par = hi_fbdev;
-
 	hi_fbdev->fb = hibmc_framebuffer_init(priv->dev, &mode_cmd, gobj);
 	if (IS_ERR(hi_fbdev->fb)) {
 		ret = PTR_ERR(hi_fbdev->fb);
@@ -129,14 +127,9 @@ static int hibmc_drm_fb_create(struct drm_fb_helper *helper,
 	priv->fbdev->size = size;
 	hi_fbdev->helper.fb = &hi_fbdev->fb->fb;
 
-	strcpy(info->fix.id, "hibmcdrmfb");
-
 	info->fbops = &hibmc_drm_fb_ops;
 
-	drm_fb_helper_fill_fix(info, hi_fbdev->fb->fb.pitches[0],
-			       hi_fbdev->fb->fb.format->depth);
-	drm_fb_helper_fill_var(info, &priv->fbdev->helper, sizes->fb_width,
-			       sizes->fb_height);
+	drm_fb_helper_fill_info(info, &priv->fbdev->helper, sizes);
 
 	info->screen_base = bo->kmap.virtual;
 	info->screen_size = size;
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
index d2cf7317930a..8c2f9b9cafb3 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
@@ -33,17 +33,10 @@ static enum drm_mode_status hibmc_connector_mode_valid(struct drm_connector *con
 	return MODE_OK;
 }
 
-static struct drm_encoder *
-hibmc_connector_best_encoder(struct drm_connector *connector)
-{
-	return drm_encoder_find(connector->dev, NULL, connector->encoder_ids[0]);
-}
-
 static const struct drm_connector_helper_funcs
 	hibmc_connector_helper_funcs = {
 	.get_modes = hibmc_connector_get_modes,
 	.mode_valid = hibmc_connector_mode_valid,
-	.best_encoder = hibmc_connector_best_encoder,
 };
 
 static const struct drm_connector_funcs hibmc_connector_funcs = {
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_ttm.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_ttm.c
index dd383267884c..6093c421daff 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_ttm.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_ttm.c
@@ -21,8 +21,6 @@
 
 #include "hibmc_drm_drv.h"
 
-#define DRM_FILE_PAGE_OFFSET (0x100000000ULL >> PAGE_SHIFT)
-
 static inline struct hibmc_drm_private *
 hibmc_bdev(struct ttm_bo_device *bd)
 {
@@ -191,7 +189,6 @@ int hibmc_mm_init(struct hibmc_drm_private *hibmc)
 	ret = ttm_bo_device_init(&hibmc->bdev,
 				 &hibmc_bo_driver,
 				 dev->anon_inode->i_mapping,
-				 DRM_FILE_PAGE_OFFSET,
 				 true);
 	if (ret) {
 		DRM_ERROR("error initializing bo driver: %d\n", ret);
@@ -322,14 +319,9 @@ int hibmc_bo_unpin(struct hibmc_bo *bo)
 
 int hibmc_mmap(struct file *filp, struct vm_area_struct *vma)
 {
-	struct drm_file *file_priv;
-	struct hibmc_drm_private *hibmc;
-
-	if (unlikely(vma->vm_pgoff < DRM_FILE_PAGE_OFFSET))
-		return -EINVAL;
+	struct drm_file *file_priv = filp->private_data;
+	struct hibmc_drm_private *hibmc = file_priv->minor->dev->dev_private;
 
-	file_priv = filp->private_data;
-	hibmc = file_priv->minor->dev->dev_private;
 	return ttm_bo_mmap(filp, vma, &hibmc->bdev);
 }
 
diff --git a/drivers/gpu/drm/i915/.gitignore b/drivers/gpu/drm/i915/.gitignore
new file mode 100644
index 000000000000..cff45d81f42f
--- /dev/null
+++ b/drivers/gpu/drm/i915/.gitignore
@@ -0,0 +1 @@
+header_test_*.c
diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index 1787e1299b1b..fbcb0904f4a8 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -32,10 +32,13 @@ CFLAGS_intel_fbdev.o = $(call cc-disable-warning, override-init)
 subdir-ccflags-y += \
 	$(call as-instr,movntdqa (%eax)$(comma)%xmm0,-DCONFIG_AS_MOVNTDQA)
 
+# Extra header tests
+include $(src)/Makefile.header-test
+
 # Please keep these build lists sorted!
 
 # core driver code
-i915-y := i915_drv.o \
+i915-y += i915_drv.o \
 	  i915_irq.o \
 	  i915_memcpy.o \
 	  i915_mm.o \
@@ -46,6 +49,7 @@ i915-y := i915_drv.o \
 	  i915_sw_fence.o \
 	  i915_syncmap.o \
 	  i915_sysfs.o \
+	  i915_user_extensions.o \
 	  intel_csr.o \
 	  intel_device_info.o \
 	  intel_pm.o \
@@ -77,6 +81,7 @@ i915-y += \
 	  i915_gem_tiling.o \
 	  i915_gem_userptr.o \
 	  i915_gemfs.o \
+	  i915_globals.o \
 	  i915_query.o \
 	  i915_request.o \
 	  i915_scheduler.o \
@@ -84,6 +89,7 @@ i915-y += \
 	  i915_trace_points.o \
 	  i915_vma.o \
 	  intel_breadcrumbs.o \
+	  intel_context.o \
 	  intel_engine_cs.o \
 	  intel_hangcheck.o \
 	  intel_lrc.o \
diff --git a/drivers/gpu/drm/i915/Makefile.header-test b/drivers/gpu/drm/i915/Makefile.header-test
new file mode 100644
index 000000000000..c1c391816fa7
--- /dev/null
+++ b/drivers/gpu/drm/i915/Makefile.header-test
@@ -0,0 +1,47 @@
+# SPDX-License-Identifier: MIT
+# Copyright © 2019 Intel Corporation
+
+# Test the headers are compilable as standalone units
+header_test := \
+	i915_active_types.h \
+	i915_gem_context_types.h \
+	i915_priolist_types.h \
+	i915_scheduler_types.h \
+	i915_timeline_types.h \
+	intel_atomic_plane.h \
+	intel_audio.h \
+	intel_cdclk.h \
+	intel_color.h \
+	intel_connector.h \
+	intel_context_types.h \
+	intel_crt.h \
+	intel_csr.h \
+	intel_ddi.h \
+	intel_dp.h \
+	intel_dvo.h \
+	intel_engine_types.h \
+	intel_fbc.h \
+	intel_fbdev.h \
+	intel_frontbuffer.h \
+	intel_hdcp.h \
+	intel_hdmi.h \
+	intel_lspcon.h \
+	intel_lvds.h \
+	intel_panel.h \
+	intel_pipe_crc.h \
+	intel_pm.h \
+	intel_psr.h \
+	intel_sdvo.h \
+	intel_sprite.h \
+	intel_tv.h \
+	intel_workarounds_types.h
+
+quiet_cmd_header_test = HDRTEST $@
+      cmd_header_test = echo "\#include \"$(<F)\"" > $@
+
+header_test_%.c: %.h
+	$(call cmd,header_test)
+
+i915-$(CONFIG_DRM_I915_WERROR) += $(foreach h,$(header_test),$(patsubst %.h,header_test_%.o,$(h)))
+
+clean-files += $(foreach h,$(header_test),$(patsubst %.h,header_test_%.c,$(h)))
diff --git a/drivers/gpu/drm/i915/gvt/Makefile b/drivers/gpu/drm/i915/gvt/Makefile
index 271fb46d4dd0..ea8324abc784 100644
--- a/drivers/gpu/drm/i915/gvt/Makefile
+++ b/drivers/gpu/drm/i915/gvt/Makefile
@@ -5,5 +5,5 @@ GVT_SOURCE := gvt.o aperture_gm.o handlers.o vgpu.o trace_points.o firmware.o \
 	execlist.o scheduler.o sched_policy.o mmio_context.o cmd_parser.o debugfs.o \
 	fb_decoder.o dmabuf.o page_track.o
 
-ccflags-y				+= -I$(src) -I$(src)/$(GVT_DIR)
+ccflags-y				+= -I $(srctree)/$(src) -I $(srctree)/$(src)/$(GVT_DIR)/
 i915-y					+= $(addprefix $(GVT_DIR)/, $(GVT_SOURCE))
diff --git a/drivers/gpu/drm/i915/gvt/cmd_parser.c b/drivers/gpu/drm/i915/gvt/cmd_parser.c
index 3592d04c33b2..ab002cfd3cab 100644
--- a/drivers/gpu/drm/i915/gvt/cmd_parser.c
+++ b/drivers/gpu/drm/i915/gvt/cmd_parser.c
@@ -391,12 +391,12 @@ struct cmd_info {
 #define F_POST_HANDLE	(1<<2)
 	u32 flag;
 
-#define R_RCS	(1 << RCS)
-#define R_VCS1  (1 << VCS)
-#define R_VCS2  (1 << VCS2)
+#define R_RCS	BIT(RCS0)
+#define R_VCS1  BIT(VCS0)
+#define R_VCS2  BIT(VCS1)
 #define R_VCS	(R_VCS1 | R_VCS2)
-#define R_BCS	(1 << BCS)
-#define R_VECS	(1 << VECS)
+#define R_BCS	BIT(BCS0)
+#define R_VECS	BIT(VECS0)
 #define R_ALL (R_RCS | R_VCS | R_BCS | R_VECS)
 	/* rings that support this cmd: BLT/RCS/VCS/VECS */
 	u16 rings;
@@ -558,7 +558,7 @@ static const struct decode_info decode_info_vebox = {
 };
 
 static const struct decode_info *ring_decode_info[I915_NUM_ENGINES][8] = {
-	[RCS] = {
+	[RCS0] = {
 		&decode_info_mi,
 		NULL,
 		NULL,
@@ -569,7 +569,7 @@ static const struct decode_info *ring_decode_info[I915_NUM_ENGINES][8] = {
 		NULL,
 	},
 
-	[VCS] = {
+	[VCS0] = {
 		&decode_info_mi,
 		NULL,
 		NULL,
@@ -580,7 +580,7 @@ static const struct decode_info *ring_decode_info[I915_NUM_ENGINES][8] = {
 		NULL,
 	},
 
-	[BCS] = {
+	[BCS0] = {
 		&decode_info_mi,
 		NULL,
 		&decode_info_2d,
@@ -591,7 +591,7 @@ static const struct decode_info *ring_decode_info[I915_NUM_ENGINES][8] = {
 		NULL,
 	},
 
-	[VECS] = {
+	[VECS0] = {
 		&decode_info_mi,
 		NULL,
 		NULL,
@@ -602,7 +602,7 @@ static const struct decode_info *ring_decode_info[I915_NUM_ENGINES][8] = {
 		NULL,
 	},
 
-	[VCS2] = {
+	[VCS1] = {
 		&decode_info_mi,
 		NULL,
 		NULL,
@@ -631,8 +631,7 @@ static inline const struct cmd_info *find_cmd_entry(struct intel_gvt *gvt,
 	struct cmd_entry *e;
 
 	hash_for_each_possible(gvt->cmd_table, e, hlist, opcode) {
-		if ((opcode == e->info->opcode) &&
-				(e->info->rings & (1 << ring_id)))
+		if (opcode == e->info->opcode && e->info->rings & BIT(ring_id))
 			return e->info;
 	}
 	return NULL;
@@ -943,15 +942,12 @@ static int cmd_handler_lri(struct parser_exec_state *s)
 	struct intel_gvt *gvt = s->vgpu->gvt;
 
 	for (i = 1; i < cmd_len; i += 2) {
-		if (IS_BROADWELL(gvt->dev_priv) &&
-				(s->ring_id != RCS)) {
-			if (s->ring_id == BCS &&
-					cmd_reg(s, i) ==
-					i915_mmio_reg_offset(DERRMR))
+		if (IS_BROADWELL(gvt->dev_priv) && s->ring_id != RCS0) {
+			if (s->ring_id == BCS0 &&
+			    cmd_reg(s, i) == i915_mmio_reg_offset(DERRMR))
 				ret |= 0;
 			else
-				ret |= (cmd_reg_inhibit(s, i)) ?
-					-EBADRQC : 0;
+				ret |= cmd_reg_inhibit(s, i) ? -EBADRQC : 0;
 		}
 		if (ret)
 			break;
@@ -1047,27 +1043,27 @@ struct cmd_interrupt_event {
 };
 
 static struct cmd_interrupt_event cmd_interrupt_events[] = {
-	[RCS] = {
+	[RCS0] = {
 		.pipe_control_notify = RCS_PIPE_CONTROL,
 		.mi_flush_dw = INTEL_GVT_EVENT_RESERVED,
 		.mi_user_interrupt = RCS_MI_USER_INTERRUPT,
 	},
-	[BCS] = {
+	[BCS0] = {
 		.pipe_control_notify = INTEL_GVT_EVENT_RESERVED,
 		.mi_flush_dw = BCS_MI_FLUSH_DW,
 		.mi_user_interrupt = BCS_MI_USER_INTERRUPT,
 	},
-	[VCS] = {
+	[VCS0] = {
 		.pipe_control_notify = INTEL_GVT_EVENT_RESERVED,
 		.mi_flush_dw = VCS_MI_FLUSH_DW,
 		.mi_user_interrupt = VCS_MI_USER_INTERRUPT,
 	},
-	[VCS2] = {
+	[VCS1] = {
 		.pipe_control_notify = INTEL_GVT_EVENT_RESERVED,
 		.mi_flush_dw = VCS2_MI_FLUSH_DW,
 		.mi_user_interrupt = VCS2_MI_USER_INTERRUPT,
 	},
-	[VECS] = {
+	[VECS0] = {
 		.pipe_control_notify = INTEL_GVT_EVENT_RESERVED,
 		.mi_flush_dw = VECS_MI_FLUSH_DW,
 		.mi_user_interrupt = VECS_MI_USER_INTERRUPT,
@@ -1081,6 +1077,7 @@ static int cmd_handler_pipe_control(struct parser_exec_state *s)
 	bool index_mode = false;
 	unsigned int post_sync;
 	int ret = 0;
+	u32 hws_pga, val;
 
 	post_sync = (cmd_val(s, 1) & PIPE_CONTROL_POST_SYNC_OP_MASK) >> 14;
 
@@ -1104,6 +1101,15 @@ static int cmd_handler_pipe_control(struct parser_exec_state *s)
 					index_mode = true;
 				ret |= cmd_address_audit(s, gma, sizeof(u64),
 						index_mode);
+				if (ret)
+					return ret;
+				if (index_mode) {
+					hws_pga = s->vgpu->hws_pga[s->ring_id];
+					gma = hws_pga + gma;
+					patch_value(s, cmd_ptr(s, 2), gma);
+					val = cmd_val(s, 1) & (~(1 << 21));
+					patch_value(s, cmd_ptr(s, 1), val);
+				}
 			}
 		}
 	}
@@ -1321,8 +1327,14 @@ static int gen8_update_plane_mmio_from_mi_display_flip(
 			      info->tile_val << 10);
 	}
 
-	vgpu_vreg_t(vgpu, PIPE_FRMCOUNT_G4X(info->pipe))++;
-	intel_vgpu_trigger_virtual_event(vgpu, info->event);
+	if (info->plane == PLANE_PRIMARY)
+		vgpu_vreg_t(vgpu, PIPE_FLIPCOUNT_G4X(info->pipe))++;
+
+	if (info->async_flip)
+		intel_vgpu_trigger_virtual_event(vgpu, info->event);
+	else
+		set_bit(info->event, vgpu->irq.flip_done_event[info->pipe]);
+
 	return 0;
 }
 
@@ -1567,6 +1579,7 @@ static int cmd_handler_mi_flush_dw(struct parser_exec_state *s)
 	unsigned long gma;
 	bool index_mode = false;
 	int ret = 0;
+	u32 hws_pga, val;
 
 	/* Check post-sync and ppgtt bit */
 	if (((cmd_val(s, 0) >> 14) & 0x3) && (cmd_val(s, 1) & (1 << 2))) {
@@ -1577,6 +1590,15 @@ static int cmd_handler_mi_flush_dw(struct parser_exec_state *s)
 		if (cmd_val(s, 0) & (1 << 21))
 			index_mode = true;
 		ret = cmd_address_audit(s, gma, sizeof(u64), index_mode);
+		if (ret)
+			return ret;
+		if (index_mode) {
+			hws_pga = s->vgpu->hws_pga[s->ring_id];
+			gma = hws_pga + gma;
+			patch_value(s, cmd_ptr(s, 1), gma);
+			val = cmd_val(s, 0) & (~(1 << 21));
+			patch_value(s, cmd_ptr(s, 0), val);
+		}
 	}
 	/* Check notify bit */
 	if ((cmd_val(s, 0) & (1 << 8)))
diff --git a/drivers/gpu/drm/i915/gvt/display.c b/drivers/gpu/drm/i915/gvt/display.c
index e3f9caa7839f..e1c313da6c00 100644
--- a/drivers/gpu/drm/i915/gvt/display.c
+++ b/drivers/gpu/drm/i915/gvt/display.c
@@ -407,7 +407,6 @@ static void emulate_vblank_on_pipe(struct intel_vgpu *vgpu, int pipe)
 		if (!pipe_is_enabled(vgpu, pipe))
 			continue;
 
-		vgpu_vreg_t(vgpu, PIPE_FLIPCOUNT_G4X(pipe))++;
 		intel_vgpu_trigger_virtual_event(vgpu, event);
 	}
 
diff --git a/drivers/gpu/drm/i915/gvt/dmabuf.c b/drivers/gpu/drm/i915/gvt/dmabuf.c
index 69a9a1b2ea4a..4e1e425189ba 100644
--- a/drivers/gpu/drm/i915/gvt/dmabuf.c
+++ b/drivers/gpu/drm/i915/gvt/dmabuf.c
@@ -153,7 +153,7 @@ static struct drm_i915_gem_object *vgpu_create_gem(struct drm_device *dev,
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct drm_i915_gem_object *obj;
 
-	obj = i915_gem_object_alloc(dev_priv);
+	obj = i915_gem_object_alloc();
 	if (obj == NULL)
 		return NULL;
 
diff --git a/drivers/gpu/drm/i915/gvt/execlist.c b/drivers/gpu/drm/i915/gvt/execlist.c
index 70494e394d2c..f21b8fb5b37e 100644
--- a/drivers/gpu/drm/i915/gvt/execlist.c
+++ b/drivers/gpu/drm/i915/gvt/execlist.c
@@ -47,17 +47,16 @@
 		((a)->lrca == (b)->lrca))
 
 static int context_switch_events[] = {
-	[RCS] = RCS_AS_CONTEXT_SWITCH,
-	[BCS] = BCS_AS_CONTEXT_SWITCH,
-	[VCS] = VCS_AS_CONTEXT_SWITCH,
-	[VCS2] = VCS2_AS_CONTEXT_SWITCH,
-	[VECS] = VECS_AS_CONTEXT_SWITCH,
+	[RCS0]  = RCS_AS_CONTEXT_SWITCH,
+	[BCS0]  = BCS_AS_CONTEXT_SWITCH,
+	[VCS0]  = VCS_AS_CONTEXT_SWITCH,
+	[VCS1]  = VCS2_AS_CONTEXT_SWITCH,
+	[VECS0] = VECS_AS_CONTEXT_SWITCH,
 };
 
-static int ring_id_to_context_switch_event(int ring_id)
+static int ring_id_to_context_switch_event(unsigned int ring_id)
 {
-	if (WARN_ON(ring_id < RCS ||
-		    ring_id >= ARRAY_SIZE(context_switch_events)))
+	if (WARN_ON(ring_id >= ARRAY_SIZE(context_switch_events)))
 		return -EINVAL;
 
 	return context_switch_events[ring_id];
@@ -411,7 +410,7 @@ static int complete_execlist_workload(struct intel_vgpu_workload *workload)
 	gvt_dbg_el("complete workload %p status %d\n", workload,
 			workload->status);
 
-	if (workload->status || (vgpu->resetting_eng & ENGINE_MASK(ring_id)))
+	if (workload->status || (vgpu->resetting_eng & BIT(ring_id)))
 		goto out;
 
 	if (!list_empty(workload_q_head(vgpu, ring_id))) {
@@ -527,12 +526,13 @@ static void init_vgpu_execlist(struct intel_vgpu *vgpu, int ring_id)
 	vgpu_vreg(vgpu, ctx_status_ptr_reg) = ctx_status_ptr.dw;
 }
 
-static void clean_execlist(struct intel_vgpu *vgpu, unsigned long engine_mask)
+static void clean_execlist(struct intel_vgpu *vgpu,
+			   intel_engine_mask_t engine_mask)
 {
-	unsigned int tmp;
 	struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
 	struct intel_engine_cs *engine;
 	struct intel_vgpu_submission *s = &vgpu->submission;
+	intel_engine_mask_t tmp;
 
 	for_each_engine_masked(engine, dev_priv, engine_mask, tmp) {
 		kfree(s->ring_scan_buffer[engine->id]);
@@ -542,18 +542,18 @@ static void clean_execlist(struct intel_vgpu *vgpu, unsigned long engine_mask)
 }
 
 static void reset_execlist(struct intel_vgpu *vgpu,
-		unsigned long engine_mask)
+			   intel_engine_mask_t engine_mask)
 {
 	struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
 	struct intel_engine_cs *engine;
-	unsigned int tmp;
+	intel_engine_mask_t tmp;
 
 	for_each_engine_masked(engine, dev_priv, engine_mask, tmp)
 		init_vgpu_execlist(vgpu, engine->id);
 }
 
 static int init_execlist(struct intel_vgpu *vgpu,
-			 unsigned long engine_mask)
+			 intel_engine_mask_t engine_mask)
 {
 	reset_execlist(vgpu, engine_mask);
 	return 0;
diff --git a/drivers/gpu/drm/i915/gvt/execlist.h b/drivers/gpu/drm/i915/gvt/execlist.h
index 714d709829a2..5ccc2c695848 100644
--- a/drivers/gpu/drm/i915/gvt/execlist.h
+++ b/drivers/gpu/drm/i915/gvt/execlist.h
@@ -180,6 +180,6 @@ int intel_vgpu_init_execlist(struct intel_vgpu *vgpu);
 int intel_vgpu_submit_execlist(struct intel_vgpu *vgpu, int ring_id);
 
 void intel_vgpu_reset_execlist(struct intel_vgpu *vgpu,
-		unsigned long engine_mask);
+			       intel_engine_mask_t engine_mask);
 
 #endif /*_GVT_EXECLIST_H_*/
diff --git a/drivers/gpu/drm/i915/gvt/gtt.c b/drivers/gpu/drm/i915/gvt/gtt.c
index 9814773882ec..c2f7d20f6346 100644
--- a/drivers/gpu/drm/i915/gvt/gtt.c
+++ b/drivers/gpu/drm/i915/gvt/gtt.c
@@ -2504,6 +2504,7 @@ static void clean_spt_oos(struct intel_gvt *gvt)
 	list_for_each_safe(pos, n, &gtt->oos_page_free_list_head) {
 		oos_page = container_of(pos, struct intel_vgpu_oos_page, list);
 		list_del(&oos_page->list);
+		free_page((unsigned long)oos_page->mem);
 		kfree(oos_page);
 	}
 }
@@ -2524,6 +2525,12 @@ static int setup_spt_oos(struct intel_gvt *gvt)
 			ret = -ENOMEM;
 			goto fail;
 		}
+		oos_page->mem = (void *)__get_free_pages(GFP_KERNEL, 0);
+		if (!oos_page->mem) {
+			ret = -ENOMEM;
+			kfree(oos_page);
+			goto fail;
+		}
 
 		INIT_LIST_HEAD(&oos_page->list);
 		INIT_LIST_HEAD(&oos_page->vm_list);
diff --git a/drivers/gpu/drm/i915/gvt/gtt.h b/drivers/gpu/drm/i915/gvt/gtt.h
index edb610dc5d86..32c573aea494 100644
--- a/drivers/gpu/drm/i915/gvt/gtt.h
+++ b/drivers/gpu/drm/i915/gvt/gtt.h
@@ -222,7 +222,7 @@ struct intel_vgpu_oos_page {
 	struct list_head list;
 	struct list_head vm_list;
 	int id;
-	unsigned char mem[I915_GTT_PAGE_SIZE];
+	void *mem;
 };
 
 #define GTT_ENTRY_NUM_IN_ONE_PAGE 512
diff --git a/drivers/gpu/drm/i915/gvt/gvt.h b/drivers/gpu/drm/i915/gvt/gvt.h
index 8bce09de4b82..f5a328b5290a 100644
--- a/drivers/gpu/drm/i915/gvt/gvt.h
+++ b/drivers/gpu/drm/i915/gvt/gvt.h
@@ -94,7 +94,6 @@ struct intel_vgpu_fence {
 
 struct intel_vgpu_mmio {
 	void *vreg;
-	void *sreg;
 };
 
 #define INTEL_GVT_MAX_BAR_NUM 4
@@ -111,11 +110,9 @@ struct intel_vgpu_cfg_space {
 
 #define vgpu_cfg_space(vgpu) ((vgpu)->cfg_space.virtual_cfg_space)
 
-#define INTEL_GVT_MAX_PIPE 4
-
 struct intel_vgpu_irq {
 	bool irq_warn_once[INTEL_GVT_EVENT_MAX];
-	DECLARE_BITMAP(flip_done_event[INTEL_GVT_MAX_PIPE],
+	DECLARE_BITMAP(flip_done_event[I915_MAX_PIPES],
 		       INTEL_GVT_EVENT_MAX);
 };
 
@@ -144,9 +141,9 @@ enum {
 
 struct intel_vgpu_submission_ops {
 	const char *name;
-	int (*init)(struct intel_vgpu *vgpu, unsigned long engine_mask);
-	void (*clean)(struct intel_vgpu *vgpu, unsigned long engine_mask);
-	void (*reset)(struct intel_vgpu *vgpu, unsigned long engine_mask);
+	int (*init)(struct intel_vgpu *vgpu, intel_engine_mask_t engine_mask);
+	void (*clean)(struct intel_vgpu *vgpu, intel_engine_mask_t engine_mask);
+	void (*reset)(struct intel_vgpu *vgpu, intel_engine_mask_t engine_mask);
 };
 
 struct intel_vgpu_submission {
@@ -449,10 +446,6 @@ void intel_vgpu_write_fence(struct intel_vgpu *vgpu,
 	(*(u64 *)(vgpu->mmio.vreg + i915_mmio_reg_offset(reg)))
 #define vgpu_vreg64(vgpu, offset) \
 	(*(u64 *)(vgpu->mmio.vreg + (offset)))
-#define vgpu_sreg_t(vgpu, reg) \
-	(*(u32 *)(vgpu->mmio.sreg + i915_mmio_reg_offset(reg)))
-#define vgpu_sreg(vgpu, offset) \
-	(*(u32 *)(vgpu->mmio.sreg + (offset)))
 
 #define for_each_active_vgpu(gvt, vgpu, id) \
 	idr_for_each_entry((&(gvt)->vgpu_idr), (vgpu), (id)) \
@@ -488,7 +481,7 @@ struct intel_vgpu *intel_gvt_create_vgpu(struct intel_gvt *gvt,
 void intel_gvt_destroy_vgpu(struct intel_vgpu *vgpu);
 void intel_gvt_release_vgpu(struct intel_vgpu *vgpu);
 void intel_gvt_reset_vgpu_locked(struct intel_vgpu *vgpu, bool dmlr,
-				 unsigned int engine_mask);
+				 intel_engine_mask_t engine_mask);
 void intel_gvt_reset_vgpu(struct intel_vgpu *vgpu);
 void intel_gvt_activate_vgpu(struct intel_vgpu *vgpu);
 void intel_gvt_deactivate_vgpu(struct intel_vgpu *vgpu);
diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c
index bc64b810e0d5..18f01eeb2510 100644
--- a/drivers/gpu/drm/i915/gvt/handlers.c
+++ b/drivers/gpu/drm/i915/gvt/handlers.c
@@ -311,7 +311,7 @@ static int mul_force_wake_write(struct intel_vgpu *vgpu,
 static int gdrst_mmio_write(struct intel_vgpu *vgpu, unsigned int offset,
 			    void *p_data, unsigned int bytes)
 {
-	unsigned int engine_mask = 0;
+	intel_engine_mask_t engine_mask = 0;
 	u32 data;
 
 	write_vreg(vgpu, offset, p_data, bytes);
@@ -323,25 +323,25 @@ static int gdrst_mmio_write(struct intel_vgpu *vgpu, unsigned int offset,
 	} else {
 		if (data & GEN6_GRDOM_RENDER) {
 			gvt_dbg_mmio("vgpu%d: request RCS reset\n", vgpu->id);
-			engine_mask |= (1 << RCS);
+			engine_mask |= BIT(RCS0);
 		}
 		if (data & GEN6_GRDOM_MEDIA) {
 			gvt_dbg_mmio("vgpu%d: request VCS reset\n", vgpu->id);
-			engine_mask |= (1 << VCS);
+			engine_mask |= BIT(VCS0);
 		}
 		if (data & GEN6_GRDOM_BLT) {
 			gvt_dbg_mmio("vgpu%d: request BCS Reset\n", vgpu->id);
-			engine_mask |= (1 << BCS);
+			engine_mask |= BIT(BCS0);
 		}
 		if (data & GEN6_GRDOM_VECS) {
 			gvt_dbg_mmio("vgpu%d: request VECS Reset\n", vgpu->id);
-			engine_mask |= (1 << VECS);
+			engine_mask |= BIT(VECS0);
 		}
 		if (data & GEN8_GRDOM_MEDIA2) {
 			gvt_dbg_mmio("vgpu%d: request VCS2 Reset\n", vgpu->id);
-			if (HAS_BSD2(vgpu->gvt->dev_priv))
-				engine_mask |= (1 << VCS2);
+			engine_mask |= BIT(VCS1);
 		}
+		engine_mask &= INTEL_INFO(vgpu->gvt->dev_priv)->engine_mask;
 	}
 
 	/* vgpu_lock already hold by emulate mmio r/w */
@@ -750,18 +750,19 @@ static int pri_surf_mmio_write(struct intel_vgpu *vgpu, unsigned int offset,
 		void *p_data, unsigned int bytes)
 {
 	struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
-	unsigned int index = DSPSURF_TO_PIPE(offset);
-	i915_reg_t surflive_reg = DSPSURFLIVE(index);
-	int flip_event[] = {
-		[PIPE_A] = PRIMARY_A_FLIP_DONE,
-		[PIPE_B] = PRIMARY_B_FLIP_DONE,
-		[PIPE_C] = PRIMARY_C_FLIP_DONE,
-	};
+	u32 pipe = DSPSURF_TO_PIPE(offset);
+	int event = SKL_FLIP_EVENT(pipe, PLANE_PRIMARY);
 
 	write_vreg(vgpu, offset, p_data, bytes);
-	vgpu_vreg_t(vgpu, surflive_reg) = vgpu_vreg(vgpu, offset);
+	vgpu_vreg_t(vgpu, DSPSURFLIVE(pipe)) = vgpu_vreg(vgpu, offset);
+
+	vgpu_vreg_t(vgpu, PIPE_FLIPCOUNT_G4X(pipe))++;
+
+	if (vgpu_vreg_t(vgpu, DSPCNTR(pipe)) & PLANE_CTL_ASYNC_FLIP)
+		intel_vgpu_trigger_virtual_event(vgpu, event);
+	else
+		set_bit(event, vgpu->irq.flip_done_event[pipe]);
 
-	set_bit(flip_event[index], vgpu->irq.flip_done_event[index]);
 	return 0;
 }
 
@@ -771,18 +772,42 @@ static int pri_surf_mmio_write(struct intel_vgpu *vgpu, unsigned int offset,
 static int spr_surf_mmio_write(struct intel_vgpu *vgpu, unsigned int offset,
 		void *p_data, unsigned int bytes)
 {
-	unsigned int index = SPRSURF_TO_PIPE(offset);
-	i915_reg_t surflive_reg = SPRSURFLIVE(index);
-	int flip_event[] = {
-		[PIPE_A] = SPRITE_A_FLIP_DONE,
-		[PIPE_B] = SPRITE_B_FLIP_DONE,
-		[PIPE_C] = SPRITE_C_FLIP_DONE,
-	};
+	u32 pipe = SPRSURF_TO_PIPE(offset);
+	int event = SKL_FLIP_EVENT(pipe, PLANE_SPRITE0);
+
+	write_vreg(vgpu, offset, p_data, bytes);
+	vgpu_vreg_t(vgpu, SPRSURFLIVE(pipe)) = vgpu_vreg(vgpu, offset);
+
+	if (vgpu_vreg_t(vgpu, SPRCTL(pipe)) & PLANE_CTL_ASYNC_FLIP)
+		intel_vgpu_trigger_virtual_event(vgpu, event);
+	else
+		set_bit(event, vgpu->irq.flip_done_event[pipe]);
+
+	return 0;
+}
+
+static int reg50080_mmio_write(struct intel_vgpu *vgpu,
+			       unsigned int offset, void *p_data,
+			       unsigned int bytes)
+{
+	struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
+	enum pipe pipe = REG_50080_TO_PIPE(offset);
+	enum plane_id plane = REG_50080_TO_PLANE(offset);
+	int event = SKL_FLIP_EVENT(pipe, plane);
 
 	write_vreg(vgpu, offset, p_data, bytes);
-	vgpu_vreg_t(vgpu, surflive_reg) = vgpu_vreg(vgpu, offset);
+	if (plane == PLANE_PRIMARY) {
+		vgpu_vreg_t(vgpu, DSPSURFLIVE(pipe)) = vgpu_vreg(vgpu, offset);
+		vgpu_vreg_t(vgpu, PIPE_FLIPCOUNT_G4X(pipe))++;
+	} else {
+		vgpu_vreg_t(vgpu, SPRSURFLIVE(pipe)) = vgpu_vreg(vgpu, offset);
+	}
+
+	if ((vgpu_vreg(vgpu, offset) & REG50080_FLIP_TYPE_MASK) == REG50080_FLIP_TYPE_ASYNC)
+		intel_vgpu_trigger_virtual_event(vgpu, event);
+	else
+		set_bit(event, vgpu->irq.flip_done_event[pipe]);
 
-	set_bit(flip_event[index], vgpu->irq.flip_done_event[index]);
 	return 0;
 }
 
@@ -1704,7 +1729,7 @@ static int ring_mode_mmio_write(struct intel_vgpu *vgpu, unsigned int offset,
 			return 0;
 
 		ret = intel_vgpu_select_submission_ops(vgpu,
-			       ENGINE_MASK(ring_id),
+			       BIT(ring_id),
 			       INTEL_VGPU_EXECLIST_SUBMISSION);
 		if (ret)
 			return ret;
@@ -1724,19 +1749,19 @@ static int gvt_reg_tlb_control_handler(struct intel_vgpu *vgpu,
 
 	switch (offset) {
 	case 0x4260:
-		id = RCS;
+		id = RCS0;
 		break;
 	case 0x4264:
-		id = VCS;
+		id = VCS0;
 		break;
 	case 0x4268:
-		id = VCS2;
+		id = VCS1;
 		break;
 	case 0x426c:
-		id = BCS;
+		id = BCS0;
 		break;
 	case 0x4270:
-		id = VECS;
+		id = VECS0;
 		break;
 	default:
 		return -EINVAL;
@@ -1793,7 +1818,7 @@ static int ring_reset_ctl_write(struct intel_vgpu *vgpu,
 	MMIO_F(prefix(BLT_RING_BASE), s, f, am, rm, d, r, w); \
 	MMIO_F(prefix(GEN6_BSD_RING_BASE), s, f, am, rm, d, r, w); \
 	MMIO_F(prefix(VEBOX_RING_BASE), s, f, am, rm, d, r, w); \
-	if (HAS_BSD2(dev_priv)) \
+	if (HAS_ENGINE(dev_priv, VCS1)) \
 		MMIO_F(prefix(GEN8_BSD2_RING_BASE), s, f, am, rm, d, r, w); \
 } while (0)
 
@@ -1848,7 +1873,7 @@ static int init_generic_mmio_info(struct intel_gvt *gvt)
 	MMIO_DH(GEN7_SC_INSTDONE, D_BDW_PLUS, mmio_read_from_hw, NULL);
 
 	MMIO_GM_RDR(_MMIO(0x2148), D_ALL, NULL, NULL);
-	MMIO_GM_RDR(CCID, D_ALL, NULL, NULL);
+	MMIO_GM_RDR(CCID(RENDER_RING_BASE), D_ALL, NULL, NULL);
 	MMIO_GM_RDR(_MMIO(0x12198), D_ALL, NULL, NULL);
 	MMIO_D(GEN7_CXT_SIZE, D_ALL);
 
@@ -1969,6 +1994,8 @@ static int init_generic_mmio_info(struct intel_gvt *gvt)
 	MMIO_DH(DSPSURF(PIPE_A), D_ALL, NULL, pri_surf_mmio_write);
 	MMIO_D(DSPOFFSET(PIPE_A), D_ALL);
 	MMIO_D(DSPSURFLIVE(PIPE_A), D_ALL);
+	MMIO_DH(REG_50080(PIPE_A, PLANE_PRIMARY), D_ALL, NULL,
+		reg50080_mmio_write);
 
 	MMIO_D(DSPCNTR(PIPE_B), D_ALL);
 	MMIO_D(DSPADDR(PIPE_B), D_ALL);
@@ -1978,6 +2005,8 @@ static int init_generic_mmio_info(struct intel_gvt *gvt)
 	MMIO_DH(DSPSURF(PIPE_B), D_ALL, NULL, pri_surf_mmio_write);
 	MMIO_D(DSPOFFSET(PIPE_B), D_ALL);
 	MMIO_D(DSPSURFLIVE(PIPE_B), D_ALL);
+	MMIO_DH(REG_50080(PIPE_B, PLANE_PRIMARY), D_ALL, NULL,
+		reg50080_mmio_write);
 
 	MMIO_D(DSPCNTR(PIPE_C), D_ALL);
 	MMIO_D(DSPADDR(PIPE_C), D_ALL);
@@ -1987,6 +2016,8 @@ static int init_generic_mmio_info(struct intel_gvt *gvt)
 	MMIO_DH(DSPSURF(PIPE_C), D_ALL, NULL, pri_surf_mmio_write);
 	MMIO_D(DSPOFFSET(PIPE_C), D_ALL);
 	MMIO_D(DSPSURFLIVE(PIPE_C), D_ALL);
+	MMIO_DH(REG_50080(PIPE_C, PLANE_PRIMARY), D_ALL, NULL,
+		reg50080_mmio_write);
 
 	MMIO_D(SPRCTL(PIPE_A), D_ALL);
 	MMIO_D(SPRLINOFF(PIPE_A), D_ALL);
@@ -2000,6 +2031,8 @@ static int init_generic_mmio_info(struct intel_gvt *gvt)
 	MMIO_D(SPROFFSET(PIPE_A), D_ALL);
 	MMIO_D(SPRSCALE(PIPE_A), D_ALL);
 	MMIO_D(SPRSURFLIVE(PIPE_A), D_ALL);
+	MMIO_DH(REG_50080(PIPE_A, PLANE_SPRITE0), D_ALL, NULL,
+		reg50080_mmio_write);
 
 	MMIO_D(SPRCTL(PIPE_B), D_ALL);
 	MMIO_D(SPRLINOFF(PIPE_B), D_ALL);
@@ -2013,6 +2046,8 @@ static int init_generic_mmio_info(struct intel_gvt *gvt)
 	MMIO_D(SPROFFSET(PIPE_B), D_ALL);
 	MMIO_D(SPRSCALE(PIPE_B), D_ALL);
 	MMIO_D(SPRSURFLIVE(PIPE_B), D_ALL);
+	MMIO_DH(REG_50080(PIPE_B, PLANE_SPRITE0), D_ALL, NULL,
+		reg50080_mmio_write);
 
 	MMIO_D(SPRCTL(PIPE_C), D_ALL);
 	MMIO_D(SPRLINOFF(PIPE_C), D_ALL);
@@ -2026,6 +2061,8 @@ static int init_generic_mmio_info(struct intel_gvt *gvt)
 	MMIO_D(SPROFFSET(PIPE_C), D_ALL);
 	MMIO_D(SPRSCALE(PIPE_C), D_ALL);
 	MMIO_D(SPRSURFLIVE(PIPE_C), D_ALL);
+	MMIO_DH(REG_50080(PIPE_C, PLANE_SPRITE0), D_ALL, NULL,
+		reg50080_mmio_write);
 
 	MMIO_D(HTOTAL(TRANSCODER_A), D_ALL);
 	MMIO_D(HBLANK(TRANSCODER_A), D_ALL);
@@ -2827,26 +2864,26 @@ static int init_skl_mmio_info(struct intel_gvt *gvt)
 
 	MMIO_DH(DBUF_CTL, D_SKL_PLUS, NULL, gen9_dbuf_ctl_mmio_write);
 
-	MMIO_D(_MMIO(0xa210), D_SKL_PLUS);
+	MMIO_D(GEN9_PG_ENABLE, D_SKL_PLUS);
 	MMIO_D(GEN9_MEDIA_PG_IDLE_HYSTERESIS, D_SKL_PLUS);
 	MMIO_D(GEN9_RENDER_PG_IDLE_HYSTERESIS, D_SKL_PLUS);
 	MMIO_DFH(GEN9_GAMT_ECO_REG_RW_IA, D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL);
-	MMIO_DH(_MMIO(0x4ddc), D_SKL_PLUS, NULL, NULL);
-	MMIO_DH(_MMIO(0x42080), D_SKL_PLUS, NULL, NULL);
-	MMIO_D(_MMIO(0x45504), D_SKL_PLUS);
-	MMIO_D(_MMIO(0x45520), D_SKL_PLUS);
-	MMIO_D(_MMIO(0x46000), D_SKL_PLUS);
-	MMIO_DH(_MMIO(0x46010), D_SKL_PLUS, NULL, skl_lcpll_write);
-	MMIO_DH(_MMIO(0x46014), D_SKL_PLUS, NULL, skl_lcpll_write);
-	MMIO_D(_MMIO(0x6C040), D_SKL_PLUS);
-	MMIO_D(_MMIO(0x6C048), D_SKL_PLUS);
-	MMIO_D(_MMIO(0x6C050), D_SKL_PLUS);
-	MMIO_D(_MMIO(0x6C044), D_SKL_PLUS);
-	MMIO_D(_MMIO(0x6C04C), D_SKL_PLUS);
-	MMIO_D(_MMIO(0x6C054), D_SKL_PLUS);
-	MMIO_D(_MMIO(0x6c058), D_SKL_PLUS);
-	MMIO_D(_MMIO(0x6c05c), D_SKL_PLUS);
-	MMIO_DH(_MMIO(0x6c060), D_SKL_PLUS, dpll_status_read, NULL);
+	MMIO_DH(MMCD_MISC_CTRL, D_SKL_PLUS, NULL, NULL);
+	MMIO_DH(CHICKEN_PAR1_1, D_SKL_PLUS, NULL, NULL);
+	MMIO_D(DC_STATE_EN, D_SKL_PLUS);
+	MMIO_D(DC_STATE_DEBUG, D_SKL_PLUS);
+	MMIO_D(CDCLK_CTL, D_SKL_PLUS);
+	MMIO_DH(LCPLL1_CTL, D_SKL_PLUS, NULL, skl_lcpll_write);
+	MMIO_DH(LCPLL2_CTL, D_SKL_PLUS, NULL, skl_lcpll_write);
+	MMIO_D(_MMIO(_DPLL1_CFGCR1), D_SKL_PLUS);
+	MMIO_D(_MMIO(_DPLL2_CFGCR1), D_SKL_PLUS);
+	MMIO_D(_MMIO(_DPLL3_CFGCR1), D_SKL_PLUS);
+	MMIO_D(_MMIO(_DPLL1_CFGCR2), D_SKL_PLUS);
+	MMIO_D(_MMIO(_DPLL2_CFGCR2), D_SKL_PLUS);
+	MMIO_D(_MMIO(_DPLL3_CFGCR2), D_SKL_PLUS);
+	MMIO_D(DPLL_CTRL1, D_SKL_PLUS);
+	MMIO_D(DPLL_CTRL2, D_SKL_PLUS);
+	MMIO_DH(DPLL_STATUS, D_SKL_PLUS, dpll_status_read, NULL);
 
 	MMIO_DH(SKL_PS_WIN_POS(PIPE_A, 0), D_SKL_PLUS, NULL, pf_write);
 	MMIO_DH(SKL_PS_WIN_POS(PIPE_A, 1), D_SKL_PLUS, NULL, pf_write);
@@ -2965,40 +3002,41 @@ static int init_skl_mmio_info(struct intel_gvt *gvt)
 	MMIO_DH(_MMIO(_REG_701C4(PIPE_C, 3)), D_SKL_PLUS, NULL, NULL);
 	MMIO_DH(_MMIO(_REG_701C4(PIPE_C, 4)), D_SKL_PLUS, NULL, NULL);
 
-	MMIO_D(_MMIO(0x70380), D_SKL_PLUS);
-	MMIO_D(_MMIO(0x71380), D_SKL_PLUS);
+	MMIO_D(_MMIO(_PLANE_CTL_3_A), D_SKL_PLUS);
+	MMIO_D(_MMIO(_PLANE_CTL_3_B), D_SKL_PLUS);
 	MMIO_D(_MMIO(0x72380), D_SKL_PLUS);
 	MMIO_D(_MMIO(0x7239c), D_SKL_PLUS);
-	MMIO_D(_MMIO(0x7039c), D_SKL_PLUS);
+	MMIO_D(_MMIO(_PLANE_SURF_3_A), D_SKL_PLUS);
 
-	MMIO_D(_MMIO(0x8f074), D_SKL_PLUS);
-	MMIO_D(_MMIO(0x8f004), D_SKL_PLUS);
-	MMIO_D(_MMIO(0x8f034), D_SKL_PLUS);
+	MMIO_D(CSR_SSP_BASE, D_SKL_PLUS);
+	MMIO_D(CSR_HTP_SKL, D_SKL_PLUS);
+	MMIO_D(CSR_LAST_WRITE, D_SKL_PLUS);
 
-	MMIO_D(_MMIO(0xb11c), D_SKL_PLUS);
+	MMIO_D(BDW_SCRATCH1, D_SKL_PLUS);
 
-	MMIO_D(_MMIO(0x51000), D_SKL_PLUS);
-	MMIO_D(_MMIO(0x6c00c), D_SKL_PLUS);
+	MMIO_D(SKL_DFSM, D_SKL_PLUS);
+	MMIO_D(DISPIO_CR_TX_BMU_CR0, D_SKL_PLUS);
 
-	MMIO_F(_MMIO(0xc800), 0x7f8, F_CMD_ACCESS, 0, 0, D_SKL_PLUS,
+	MMIO_F(GEN9_GFX_MOCS(0), 0x7f8, F_CMD_ACCESS, 0, 0, D_SKL_PLUS,
 		NULL, NULL);
-	MMIO_F(_MMIO(0xb020), 0x80, F_CMD_ACCESS, 0, 0, D_SKL_PLUS,
+	MMIO_F(GEN7_L3CNTLREG2, 0x80, F_CMD_ACCESS, 0, 0, D_SKL_PLUS,
 		NULL, NULL);
 
 	MMIO_D(RPM_CONFIG0, D_SKL_PLUS);
 	MMIO_D(_MMIO(0xd08), D_SKL_PLUS);
 	MMIO_D(RC6_LOCATION, D_SKL_PLUS);
-	MMIO_DFH(_MMIO(0x20e0), D_SKL_PLUS, F_MODE_MASK, NULL, NULL);
-	MMIO_DFH(_MMIO(0x20ec), D_SKL_PLUS, F_MODE_MASK | F_CMD_ACCESS,
+	MMIO_DFH(GEN7_FF_SLICE_CS_CHICKEN1, D_SKL_PLUS, F_MODE_MASK,
+		NULL, NULL);
+	MMIO_DFH(GEN9_CS_DEBUG_MODE1, D_SKL_PLUS, F_MODE_MASK | F_CMD_ACCESS,
 		NULL, NULL);
 
 	/* TRTT */
-	MMIO_DFH(_MMIO(0x4de0), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL);
-	MMIO_DFH(_MMIO(0x4de4), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL);
-	MMIO_DFH(_MMIO(0x4de8), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL);
-	MMIO_DFH(_MMIO(0x4dec), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL);
-	MMIO_DFH(_MMIO(0x4df0), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL);
-	MMIO_DFH(_MMIO(0x4df4), D_SKL_PLUS, F_CMD_ACCESS,
+	MMIO_DFH(TRVATTL3PTRDW(0), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL);
+	MMIO_DFH(TRVATTL3PTRDW(1), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL);
+	MMIO_DFH(TRVATTL3PTRDW(2), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL);
+	MMIO_DFH(TRVATTL3PTRDW(3), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL);
+	MMIO_DFH(TRVADR, D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL);
+	MMIO_DFH(TRTTE, D_SKL_PLUS, F_CMD_ACCESS,
 		NULL, gen9_trtte_write);
 	MMIO_DH(_MMIO(0x4dfc), D_SKL_PLUS, NULL, gen9_trtt_chicken_write);
 
@@ -3011,7 +3049,7 @@ static int init_skl_mmio_info(struct intel_gvt *gvt)
 	MMIO_DH(DMA_CTRL, D_SKL_PLUS, NULL, dma_ctrl_write);
 
 	MMIO_D(_MMIO(0x65900), D_SKL_PLUS);
-	MMIO_D(_MMIO(0x1082c0), D_SKL_PLUS);
+	MMIO_D(GEN6_STOLEN_RESERVED, D_SKL_PLUS);
 	MMIO_D(_MMIO(0x4068), D_SKL_PLUS);
 	MMIO_D(_MMIO(0x67054), D_SKL_PLUS);
 	MMIO_D(_MMIO(0x6e560), D_SKL_PLUS);
@@ -3042,8 +3080,8 @@ static int init_skl_mmio_info(struct intel_gvt *gvt)
 	MMIO_DFH(GEN9_WM_CHICKEN3, D_SKL_PLUS, F_MODE_MASK | F_CMD_ACCESS,
 		 NULL, NULL);
 
-	MMIO_D(_MMIO(0x4ab8), D_KBL | D_CFL);
-	MMIO_D(_MMIO(0x2248), D_SKL_PLUS);
+	MMIO_D(GAMT_CHKN_BIT_REG, D_KBL);
+	MMIO_D(GEN9_CTX_PREEMPT_REG, D_KBL | D_SKL);
 
 	return 0;
 }
@@ -3265,7 +3303,7 @@ void intel_gvt_clean_mmio_info(struct intel_gvt *gvt)
 /* Special MMIO blocks. */
 static struct gvt_mmio_block mmio_blocks[] = {
 	{D_SKL_PLUS, _MMIO(CSR_MMIO_START_RANGE), 0x3000, NULL, NULL},
-	{D_ALL, _MMIO(MCHBAR_MIRROR_BASE_SNB), 0x40000, NULL, NULL},
+	{D_ALL, MCHBAR_MIRROR_REG_BASE, 0x4000, NULL, NULL},
 	{D_ALL, _MMIO(VGT_PVINFO_PAGE), VGT_PVINFO_SIZE,
 		pvinfo_mmio_read, pvinfo_mmio_write},
 	{D_ALL, LGC_PALETTE(PIPE_A, 0), 1024, NULL, NULL},
@@ -3489,12 +3527,11 @@ int intel_vgpu_mmio_reg_rw(struct intel_vgpu *vgpu, unsigned int offset,
 		return mmio_info->read(vgpu, offset, pdata, bytes);
 	else {
 		u64 ro_mask = mmio_info->ro_mask;
-		u32 old_vreg = 0, old_sreg = 0;
+		u32 old_vreg = 0;
 		u64 data = 0;
 
 		if (intel_gvt_mmio_has_mode_mask(gvt, mmio_info->offset)) {
 			old_vreg = vgpu_vreg(vgpu, offset);
-			old_sreg = vgpu_sreg(vgpu, offset);
 		}
 
 		if (likely(!ro_mask))
@@ -3516,8 +3553,6 @@ int intel_vgpu_mmio_reg_rw(struct intel_vgpu *vgpu, unsigned int offset,
 
 			vgpu_vreg(vgpu, offset) = (old_vreg & ~mask)
 					| (vgpu_vreg(vgpu, offset) & mask);
-			vgpu_sreg(vgpu, offset) = (old_sreg & ~mask)
-					| (vgpu_sreg(vgpu, offset) & mask);
 		}
 	}
 
diff --git a/drivers/gpu/drm/i915/gvt/interrupt.c b/drivers/gpu/drm/i915/gvt/interrupt.c
index 67125c5eec6e..951681813230 100644
--- a/drivers/gpu/drm/i915/gvt/interrupt.c
+++ b/drivers/gpu/drm/i915/gvt/interrupt.c
@@ -536,7 +536,7 @@ static void gen8_init_irq(
 	SET_BIT_INFO(irq, 4, VCS_MI_FLUSH_DW, INTEL_GVT_IRQ_INFO_GT1);
 	SET_BIT_INFO(irq, 8, VCS_AS_CONTEXT_SWITCH, INTEL_GVT_IRQ_INFO_GT1);
 
-	if (HAS_BSD2(gvt->dev_priv)) {
+	if (HAS_ENGINE(gvt->dev_priv, VCS1)) {
 		SET_BIT_INFO(irq, 16, VCS2_MI_USER_INTERRUPT,
 			INTEL_GVT_IRQ_INFO_GT1);
 		SET_BIT_INFO(irq, 20, VCS2_MI_FLUSH_DW,
diff --git a/drivers/gpu/drm/i915/gvt/mmio.c b/drivers/gpu/drm/i915/gvt/mmio.c
index ed4df2f6d60b..a55178884d67 100644
--- a/drivers/gpu/drm/i915/gvt/mmio.c
+++ b/drivers/gpu/drm/i915/gvt/mmio.c
@@ -239,7 +239,6 @@ void intel_vgpu_reset_mmio(struct intel_vgpu *vgpu, bool dmlr)
 
 	if (dmlr) {
 		memcpy(vgpu->mmio.vreg, mmio, info->mmio_size);
-		memcpy(vgpu->mmio.sreg, mmio, info->mmio_size);
 
 		vgpu_vreg_t(vgpu, GEN6_GT_THREAD_STATUS_REG) = 0;
 
@@ -280,7 +279,6 @@ void intel_vgpu_reset_mmio(struct intel_vgpu *vgpu, bool dmlr)
 		 * touched
 		 */
 		memcpy(vgpu->mmio.vreg, mmio, GVT_GEN8_MMIO_RESET_OFFSET);
-		memcpy(vgpu->mmio.sreg, mmio, GVT_GEN8_MMIO_RESET_OFFSET);
 	}
 
 }
@@ -296,12 +294,10 @@ int intel_vgpu_init_mmio(struct intel_vgpu *vgpu)
 {
 	const struct intel_gvt_device_info *info = &vgpu->gvt->device_info;
 
-	vgpu->mmio.vreg = vzalloc(array_size(info->mmio_size, 2));
+	vgpu->mmio.vreg = vzalloc(info->mmio_size);
 	if (!vgpu->mmio.vreg)
 		return -ENOMEM;
 
-	vgpu->mmio.sreg = vgpu->mmio.vreg + info->mmio_size;
-
 	intel_vgpu_reset_mmio(vgpu, true);
 
 	return 0;
@@ -315,5 +311,5 @@ int intel_vgpu_init_mmio(struct intel_vgpu *vgpu)
 void intel_vgpu_clean_mmio(struct intel_vgpu *vgpu)
 {
 	vfree(vgpu->mmio.vreg);
-	vgpu->mmio.vreg = vgpu->mmio.sreg = NULL;
+	vgpu->mmio.vreg = NULL;
 }
diff --git a/drivers/gpu/drm/i915/gvt/mmio_context.c b/drivers/gpu/drm/i915/gvt/mmio_context.c
index 7902fb162d09..e7e14c842be4 100644
--- a/drivers/gpu/drm/i915/gvt/mmio_context.c
+++ b/drivers/gpu/drm/i915/gvt/mmio_context.c
@@ -41,103 +41,102 @@
 
 /* Raw offset is appened to each line for convenience. */
 static struct engine_mmio gen8_engine_mmio_list[] __cacheline_aligned = {
-	{RCS, GFX_MODE_GEN7, 0xffff, false}, /* 0x229c */
-	{RCS, GEN9_CTX_PREEMPT_REG, 0x0, false}, /* 0x2248 */
-	{RCS, HWSTAM, 0x0, false}, /* 0x2098 */
-	{RCS, INSTPM, 0xffff, true}, /* 0x20c0 */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 0), 0, false}, /* 0x24d0 */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 1), 0, false}, /* 0x24d4 */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 2), 0, false}, /* 0x24d8 */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 3), 0, false}, /* 0x24dc */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 4), 0, false}, /* 0x24e0 */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 5), 0, false}, /* 0x24e4 */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 6), 0, false}, /* 0x24e8 */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 7), 0, false}, /* 0x24ec */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 8), 0, false}, /* 0x24f0 */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 9), 0, false}, /* 0x24f4 */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 10), 0, false}, /* 0x24f8 */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 11), 0, false}, /* 0x24fc */
-	{RCS, CACHE_MODE_1, 0xffff, true}, /* 0x7004 */
-	{RCS, GEN7_GT_MODE, 0xffff, true}, /* 0x7008 */
-	{RCS, CACHE_MODE_0_GEN7, 0xffff, true}, /* 0x7000 */
-	{RCS, GEN7_COMMON_SLICE_CHICKEN1, 0xffff, true}, /* 0x7010 */
-	{RCS, HDC_CHICKEN0, 0xffff, true}, /* 0x7300 */
-	{RCS, VF_GUARDBAND, 0xffff, true}, /* 0x83a4 */
-
-	{BCS, RING_GFX_MODE(BLT_RING_BASE), 0xffff, false}, /* 0x2229c */
-	{BCS, RING_MI_MODE(BLT_RING_BASE), 0xffff, false}, /* 0x2209c */
-	{BCS, RING_INSTPM(BLT_RING_BASE), 0xffff, false}, /* 0x220c0 */
-	{BCS, RING_HWSTAM(BLT_RING_BASE), 0x0, false}, /* 0x22098 */
-	{BCS, RING_EXCC(BLT_RING_BASE), 0x0, false}, /* 0x22028 */
-	{RCS, INVALID_MMIO_REG, 0, false } /* Terminated */
+	{RCS0, GFX_MODE_GEN7, 0xffff, false}, /* 0x229c */
+	{RCS0, GEN9_CTX_PREEMPT_REG, 0x0, false}, /* 0x2248 */
+	{RCS0, HWSTAM, 0x0, false}, /* 0x2098 */
+	{RCS0, INSTPM, 0xffff, true}, /* 0x20c0 */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 0), 0, false}, /* 0x24d0 */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 1), 0, false}, /* 0x24d4 */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 2), 0, false}, /* 0x24d8 */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 3), 0, false}, /* 0x24dc */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 4), 0, false}, /* 0x24e0 */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 5), 0, false}, /* 0x24e4 */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 6), 0, false}, /* 0x24e8 */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 7), 0, false}, /* 0x24ec */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 8), 0, false}, /* 0x24f0 */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 9), 0, false}, /* 0x24f4 */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 10), 0, false}, /* 0x24f8 */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 11), 0, false}, /* 0x24fc */
+	{RCS0, CACHE_MODE_1, 0xffff, true}, /* 0x7004 */
+	{RCS0, GEN7_GT_MODE, 0xffff, true}, /* 0x7008 */
+	{RCS0, CACHE_MODE_0_GEN7, 0xffff, true}, /* 0x7000 */
+	{RCS0, GEN7_COMMON_SLICE_CHICKEN1, 0xffff, true}, /* 0x7010 */
+	{RCS0, HDC_CHICKEN0, 0xffff, true}, /* 0x7300 */
+	{RCS0, VF_GUARDBAND, 0xffff, true}, /* 0x83a4 */
+
+	{BCS0, RING_GFX_MODE(BLT_RING_BASE), 0xffff, false}, /* 0x2229c */
+	{BCS0, RING_MI_MODE(BLT_RING_BASE), 0xffff, false}, /* 0x2209c */
+	{BCS0, RING_INSTPM(BLT_RING_BASE), 0xffff, false}, /* 0x220c0 */
+	{BCS0, RING_HWSTAM(BLT_RING_BASE), 0x0, false}, /* 0x22098 */
+	{BCS0, RING_EXCC(BLT_RING_BASE), 0xffff, false}, /* 0x22028 */
+	{RCS0, INVALID_MMIO_REG, 0, false } /* Terminated */
 };
 
 static struct engine_mmio gen9_engine_mmio_list[] __cacheline_aligned = {
-	{RCS, GFX_MODE_GEN7, 0xffff, false}, /* 0x229c */
-	{RCS, GEN9_CTX_PREEMPT_REG, 0x0, false}, /* 0x2248 */
-	{RCS, HWSTAM, 0x0, false}, /* 0x2098 */
-	{RCS, INSTPM, 0xffff, true}, /* 0x20c0 */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 0), 0, false}, /* 0x24d0 */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 1), 0, false}, /* 0x24d4 */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 2), 0, false}, /* 0x24d8 */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 3), 0, false}, /* 0x24dc */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 4), 0, false}, /* 0x24e0 */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 5), 0, false}, /* 0x24e4 */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 6), 0, false}, /* 0x24e8 */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 7), 0, false}, /* 0x24ec */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 8), 0, false}, /* 0x24f0 */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 9), 0, false}, /* 0x24f4 */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 10), 0, false}, /* 0x24f8 */
-	{RCS, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 11), 0, false}, /* 0x24fc */
-	{RCS, CACHE_MODE_1, 0xffff, true}, /* 0x7004 */
-	{RCS, GEN7_GT_MODE, 0xffff, true}, /* 0x7008 */
-	{RCS, CACHE_MODE_0_GEN7, 0xffff, true}, /* 0x7000 */
-	{RCS, GEN7_COMMON_SLICE_CHICKEN1, 0xffff, true}, /* 0x7010 */
-	{RCS, HDC_CHICKEN0, 0xffff, true}, /* 0x7300 */
-	{RCS, VF_GUARDBAND, 0xffff, true}, /* 0x83a4 */
-
-	{RCS, GEN8_PRIVATE_PAT_LO, 0, false}, /* 0x40e0 */
-	{RCS, GEN8_PRIVATE_PAT_HI, 0, false}, /* 0x40e4 */
-	{RCS, GEN8_CS_CHICKEN1, 0xffff, true}, /* 0x2580 */
-	{RCS, COMMON_SLICE_CHICKEN2, 0xffff, true}, /* 0x7014 */
-	{RCS, GEN9_CS_DEBUG_MODE1, 0xffff, false}, /* 0x20ec */
-	{RCS, GEN8_L3SQCREG4, 0, false}, /* 0xb118 */
-	{RCS, GEN7_HALF_SLICE_CHICKEN1, 0xffff, true}, /* 0xe100 */
-	{RCS, HALF_SLICE_CHICKEN2, 0xffff, true}, /* 0xe180 */
-	{RCS, HALF_SLICE_CHICKEN3, 0xffff, true}, /* 0xe184 */
-	{RCS, GEN9_HALF_SLICE_CHICKEN5, 0xffff, true}, /* 0xe188 */
-	{RCS, GEN9_HALF_SLICE_CHICKEN7, 0xffff, true}, /* 0xe194 */
-	{RCS, GEN8_ROW_CHICKEN, 0xffff, true}, /* 0xe4f0 */
-	{RCS, TRVATTL3PTRDW(0), 0, false}, /* 0x4de0 */
-	{RCS, TRVATTL3PTRDW(1), 0, false}, /* 0x4de4 */
-	{RCS, TRNULLDETCT, 0, false}, /* 0x4de8 */
-	{RCS, TRINVTILEDETCT, 0, false}, /* 0x4dec */
-	{RCS, TRVADR, 0, false}, /* 0x4df0 */
-	{RCS, TRTTE, 0, false}, /* 0x4df4 */
-
-	{BCS, RING_GFX_MODE(BLT_RING_BASE), 0xffff, false}, /* 0x2229c */
-	{BCS, RING_MI_MODE(BLT_RING_BASE), 0xffff, false}, /* 0x2209c */
-	{BCS, RING_INSTPM(BLT_RING_BASE), 0xffff, false}, /* 0x220c0 */
-	{BCS, RING_HWSTAM(BLT_RING_BASE), 0x0, false}, /* 0x22098 */
-	{BCS, RING_EXCC(BLT_RING_BASE), 0x0, false}, /* 0x22028 */
-
-	{VCS2, RING_EXCC(GEN8_BSD2_RING_BASE), 0xffff, false}, /* 0x1c028 */
-
-	{VECS, RING_EXCC(VEBOX_RING_BASE), 0xffff, false}, /* 0x1a028 */
-
-	{RCS, GEN8_HDC_CHICKEN1, 0xffff, true}, /* 0x7304 */
-	{RCS, GEN9_CTX_PREEMPT_REG, 0x0, false}, /* 0x2248 */
-	{RCS, GEN7_UCGCTL4, 0x0, false}, /* 0x940c */
-	{RCS, GAMT_CHKN_BIT_REG, 0x0, false}, /* 0x4ab8 */
-
-	{RCS, GEN9_GAMT_ECO_REG_RW_IA, 0x0, false}, /* 0x4ab0 */
-	{RCS, GEN9_CSFE_CHICKEN1_RCS, 0xffff, false}, /* 0x20d4 */
-	{RCS, _MMIO(0x20D8), 0xffff, true}, /* 0x20d8 */
-
-	{RCS, GEN8_GARBCNTL, 0x0, false}, /* 0xb004 */
-	{RCS, GEN7_FF_THREAD_MODE, 0x0, false}, /* 0x20a0 */
-	{RCS, FF_SLICE_CS_CHICKEN2, 0xffff, false}, /* 0x20e4 */
-	{RCS, INVALID_MMIO_REG, 0, false } /* Terminated */
+	{RCS0, GFX_MODE_GEN7, 0xffff, false}, /* 0x229c */
+	{RCS0, GEN9_CTX_PREEMPT_REG, 0x0, false}, /* 0x2248 */
+	{RCS0, HWSTAM, 0x0, false}, /* 0x2098 */
+	{RCS0, INSTPM, 0xffff, true}, /* 0x20c0 */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 0), 0, false}, /* 0x24d0 */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 1), 0, false}, /* 0x24d4 */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 2), 0, false}, /* 0x24d8 */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 3), 0, false}, /* 0x24dc */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 4), 0, false}, /* 0x24e0 */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 5), 0, false}, /* 0x24e4 */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 6), 0, false}, /* 0x24e8 */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 7), 0, false}, /* 0x24ec */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 8), 0, false}, /* 0x24f0 */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 9), 0, false}, /* 0x24f4 */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 10), 0, false}, /* 0x24f8 */
+	{RCS0, RING_FORCE_TO_NONPRIV(RENDER_RING_BASE, 11), 0, false}, /* 0x24fc */
+	{RCS0, CACHE_MODE_1, 0xffff, true}, /* 0x7004 */
+	{RCS0, GEN7_GT_MODE, 0xffff, true}, /* 0x7008 */
+	{RCS0, CACHE_MODE_0_GEN7, 0xffff, true}, /* 0x7000 */
+	{RCS0, GEN7_COMMON_SLICE_CHICKEN1, 0xffff, true}, /* 0x7010 */
+	{RCS0, HDC_CHICKEN0, 0xffff, true}, /* 0x7300 */
+	{RCS0, VF_GUARDBAND, 0xffff, true}, /* 0x83a4 */
+
+	{RCS0, GEN8_PRIVATE_PAT_LO, 0, false}, /* 0x40e0 */
+	{RCS0, GEN8_PRIVATE_PAT_HI, 0, false}, /* 0x40e4 */
+	{RCS0, GEN8_CS_CHICKEN1, 0xffff, true}, /* 0x2580 */
+	{RCS0, COMMON_SLICE_CHICKEN2, 0xffff, true}, /* 0x7014 */
+	{RCS0, GEN9_CS_DEBUG_MODE1, 0xffff, false}, /* 0x20ec */
+	{RCS0, GEN8_L3SQCREG4, 0, false}, /* 0xb118 */
+	{RCS0, GEN7_HALF_SLICE_CHICKEN1, 0xffff, true}, /* 0xe100 */
+	{RCS0, HALF_SLICE_CHICKEN2, 0xffff, true}, /* 0xe180 */
+	{RCS0, HALF_SLICE_CHICKEN3, 0xffff, true}, /* 0xe184 */
+	{RCS0, GEN9_HALF_SLICE_CHICKEN5, 0xffff, true}, /* 0xe188 */
+	{RCS0, GEN9_HALF_SLICE_CHICKEN7, 0xffff, true}, /* 0xe194 */
+	{RCS0, GEN8_ROW_CHICKEN, 0xffff, true}, /* 0xe4f0 */
+	{RCS0, TRVATTL3PTRDW(0), 0, false}, /* 0x4de0 */
+	{RCS0, TRVATTL3PTRDW(1), 0, false}, /* 0x4de4 */
+	{RCS0, TRNULLDETCT, 0, false}, /* 0x4de8 */
+	{RCS0, TRINVTILEDETCT, 0, false}, /* 0x4dec */
+	{RCS0, TRVADR, 0, false}, /* 0x4df0 */
+	{RCS0, TRTTE, 0, false}, /* 0x4df4 */
+
+	{BCS0, RING_GFX_MODE(BLT_RING_BASE), 0xffff, false}, /* 0x2229c */
+	{BCS0, RING_MI_MODE(BLT_RING_BASE), 0xffff, false}, /* 0x2209c */
+	{BCS0, RING_INSTPM(BLT_RING_BASE), 0xffff, false}, /* 0x220c0 */
+	{BCS0, RING_HWSTAM(BLT_RING_BASE), 0x0, false}, /* 0x22098 */
+	{BCS0, RING_EXCC(BLT_RING_BASE), 0xffff, false}, /* 0x22028 */
+
+	{VCS1, RING_EXCC(GEN8_BSD2_RING_BASE), 0xffff, false}, /* 0x1c028 */
+
+	{VECS0, RING_EXCC(VEBOX_RING_BASE), 0xffff, false}, /* 0x1a028 */
+
+	{RCS0, GEN8_HDC_CHICKEN1, 0xffff, true}, /* 0x7304 */
+	{RCS0, GEN9_CTX_PREEMPT_REG, 0x0, false}, /* 0x2248 */
+	{RCS0, GEN7_UCGCTL4, 0x0, false}, /* 0x940c */
+	{RCS0, GAMT_CHKN_BIT_REG, 0x0, false}, /* 0x4ab8 */
+
+	{RCS0, GEN9_GAMT_ECO_REG_RW_IA, 0x0, false}, /* 0x4ab0 */
+	{RCS0, GEN9_CSFE_CHICKEN1_RCS, 0xffff, false}, /* 0x20d4 */
+
+	{RCS0, GEN8_GARBCNTL, 0x0, false}, /* 0xb004 */
+	{RCS0, GEN7_FF_THREAD_MODE, 0x0, false}, /* 0x20a0 */
+	{RCS0, FF_SLICE_CS_CHICKEN2, 0xffff, false}, /* 0x20e4 */
+	{RCS0, INVALID_MMIO_REG, 0, false } /* Terminated */
 };
 
 static struct {
@@ -150,11 +149,11 @@ static void load_render_mocs(struct drm_i915_private *dev_priv)
 {
 	i915_reg_t offset;
 	u32 regs[] = {
-		[RCS] = 0xc800,
-		[VCS] = 0xc900,
-		[VCS2] = 0xca00,
-		[BCS] = 0xcc00,
-		[VECS] = 0xcb00,
+		[RCS0]  = 0xc800,
+		[VCS0]  = 0xc900,
+		[VCS1]  = 0xca00,
+		[BCS0]  = 0xcc00,
+		[VECS0] = 0xcb00,
 	};
 	int ring_id, i;
 
@@ -302,7 +301,7 @@ int intel_vgpu_restore_inhibit_context(struct intel_vgpu *vgpu,
 		goto out;
 
 	/* no MOCS register in context except render engine */
-	if (req->engine->id != RCS)
+	if (req->engine->id != RCS0)
 		goto out;
 
 	ret = restore_render_mocs_control_for_inhibit(vgpu, req);
@@ -328,15 +327,16 @@ out:
 static void handle_tlb_pending_event(struct intel_vgpu *vgpu, int ring_id)
 {
 	struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
+	struct intel_uncore *uncore = &dev_priv->uncore;
 	struct intel_vgpu_submission *s = &vgpu->submission;
 	enum forcewake_domains fw;
 	i915_reg_t reg;
 	u32 regs[] = {
-		[RCS] = 0x4260,
-		[VCS] = 0x4264,
-		[VCS2] = 0x4268,
-		[BCS] = 0x426c,
-		[VECS] = 0x4270,
+		[RCS0]  = 0x4260,
+		[VCS0]  = 0x4264,
+		[VCS1]  = 0x4268,
+		[BCS0]  = 0x426c,
+		[VECS0] = 0x4270,
 	};
 
 	if (WARN_ON(ring_id >= ARRAY_SIZE(regs)))
@@ -352,21 +352,21 @@ static void handle_tlb_pending_event(struct intel_vgpu *vgpu, int ring_id)
 	 * otherwise device can go to RC6 state and interrupt invalidation
 	 * process
 	 */
-	fw = intel_uncore_forcewake_for_reg(dev_priv, reg,
+	fw = intel_uncore_forcewake_for_reg(uncore, reg,
 					    FW_REG_READ | FW_REG_WRITE);
-	if (ring_id == RCS && (INTEL_GEN(dev_priv) >= 9))
+	if (ring_id == RCS0 && INTEL_GEN(dev_priv) >= 9)
 		fw |= FORCEWAKE_RENDER;
 
-	intel_uncore_forcewake_get(dev_priv, fw);
+	intel_uncore_forcewake_get(uncore, fw);
 
-	I915_WRITE_FW(reg, 0x1);
+	intel_uncore_write_fw(uncore, reg, 0x1);
 
-	if (wait_for_atomic((I915_READ_FW(reg) == 0), 50))
+	if (wait_for_atomic((intel_uncore_read_fw(uncore, reg) == 0), 50))
 		gvt_vgpu_err("timeout in invalidate ring (%d) tlb\n", ring_id);
 	else
 		vgpu_vreg_t(vgpu, reg) = 0;
 
-	intel_uncore_forcewake_put(dev_priv, fw);
+	intel_uncore_forcewake_put(uncore, fw);
 
 	gvt_dbg_core("invalidate TLB for ring %d\n", ring_id);
 }
@@ -379,11 +379,11 @@ static void switch_mocs(struct intel_vgpu *pre, struct intel_vgpu *next,
 	u32 old_v, new_v;
 
 	u32 regs[] = {
-		[RCS] = 0xc800,
-		[VCS] = 0xc900,
-		[VCS2] = 0xca00,
-		[BCS] = 0xcc00,
-		[VECS] = 0xcb00,
+		[RCS0]  = 0xc800,
+		[VCS0]  = 0xc900,
+		[VCS1]  = 0xca00,
+		[BCS0]  = 0xcc00,
+		[VECS0] = 0xcb00,
 	};
 	int i;
 
@@ -391,8 +391,10 @@ static void switch_mocs(struct intel_vgpu *pre, struct intel_vgpu *next,
 	if (WARN_ON(ring_id >= ARRAY_SIZE(regs)))
 		return;
 
-	if ((IS_KABYLAKE(dev_priv)  || IS_BROXTON(dev_priv)
-		|| IS_COFFEELAKE(dev_priv)) && ring_id == RCS)
+	if (ring_id == RCS0 &&
+	    (IS_KABYLAKE(dev_priv) ||
+	     IS_BROXTON(dev_priv) ||
+	     IS_COFFEELAKE(dev_priv)))
 		return;
 
 	if (!pre && !gen9_render_mocs.initialized)
@@ -415,7 +417,7 @@ static void switch_mocs(struct intel_vgpu *pre, struct intel_vgpu *next,
 		offset.reg += 4;
 	}
 
-	if (ring_id == RCS) {
+	if (ring_id == RCS0) {
 		l3_offset.reg = 0xb020;
 		for (i = 0; i < GEN9_MOCS_SIZE / 2; i++) {
 			if (pre)
@@ -493,7 +495,8 @@ static void switch_mmio(struct intel_vgpu *pre,
 			 * itself.
 			 */
 			if (mmio->in_context &&
-			    !is_inhibit_context(&s->shadow_ctx->__engine[ring_id]))
+			    !is_inhibit_context(intel_context_lookup(s->shadow_ctx,
+								     dev_priv->engine[ring_id])))
 				continue;
 
 			if (mmio->mask)
@@ -550,9 +553,9 @@ void intel_gvt_switch_mmio(struct intel_vgpu *pre,
 	 * performace for batch mmio read/write, so we need
 	 * handle forcewake mannually.
 	 */
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL);
 	switch_mmio(pre, next, ring_id);
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 }
 
 /**
diff --git a/drivers/gpu/drm/i915/gvt/reg.h b/drivers/gpu/drm/i915/gvt/reg.h
index 428d252344f1..3de5b643b266 100644
--- a/drivers/gpu/drm/i915/gvt/reg.h
+++ b/drivers/gpu/drm/i915/gvt/reg.h
@@ -60,6 +60,37 @@
 #define _REG_701C0(pipe, plane) (0x701c0 + pipe * 0x1000 + (plane - 1) * 0x100)
 #define _REG_701C4(pipe, plane) (0x701c4 + pipe * 0x1000 + (plane - 1) * 0x100)
 
+#define SKL_FLIP_EVENT(pipe, plane) (PRIMARY_A_FLIP_DONE + (plane) * 3 + (pipe))
+
+#define PLANE_CTL_ASYNC_FLIP		(1 << 9)
+#define REG50080_FLIP_TYPE_MASK	0x3
+#define REG50080_FLIP_TYPE_ASYNC	0x1
+
+#define REG_50080(_pipe, _plane) ({ \
+	typeof(_pipe) (p) = (_pipe); \
+	typeof(_plane) (q) = (_plane); \
+	(((p) == PIPE_A) ? (((q) == PLANE_PRIMARY) ? (_MMIO(0x50080)) : \
+		(_MMIO(0x50090))) : \
+	(((p) == PIPE_B) ? (((q) == PLANE_PRIMARY) ? (_MMIO(0x50088)) : \
+		(_MMIO(0x50098))) : \
+	(((p) == PIPE_C) ? (((q) == PLANE_PRIMARY) ? (_MMIO(0x5008C)) : \
+		(_MMIO(0x5009C))) : \
+		(_MMIO(0x50080))))); })
+
+#define REG_50080_TO_PIPE(_reg) ({ \
+	typeof(_reg) (reg) = (_reg); \
+	(((reg) == 0x50080 || (reg) == 0x50090) ? (PIPE_A) : \
+	(((reg) == 0x50088 || (reg) == 0x50098) ? (PIPE_B) : \
+	(((reg) == 0x5008C || (reg) == 0x5009C) ? (PIPE_C) : \
+	(INVALID_PIPE)))); })
+
+#define REG_50080_TO_PLANE(_reg) ({ \
+	typeof(_reg) (reg) = (_reg); \
+	(((reg) == 0x50080 || (reg) == 0x50088 || (reg) == 0x5008C) ? \
+		(PLANE_PRIMARY) : \
+	(((reg) == 0x50090 || (reg) == 0x50098 || (reg) == 0x5009C) ? \
+		(PLANE_SPRITE0) : (I915_MAX_PLANES))); })
+
 #define GFX_MODE_BIT_SET_IN_MASK(val, bit) \
 		((((bit) & 0xffff0000) == 0) && !!((val) & (((bit) << 16))))
 
@@ -95,4 +126,7 @@
 #define RING_GFX_MODE(base)	_MMIO((base) + 0x29c)
 #define VF_GUARDBAND		_MMIO(0x83a4)
 
+/* define the effective range of MCHBAR register on Sandybridge+ */
+#define MCHBAR_MIRROR_REG_BASE	_MMIO(MCHBAR_MIRROR_BASE_SNB + 0x4000)
+
 #endif
diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c
index 05b953793316..8998fa5ab198 100644
--- a/drivers/gpu/drm/i915/gvt/scheduler.c
+++ b/drivers/gpu/drm/i915/gvt/scheduler.c
@@ -93,7 +93,7 @@ static void sr_oa_regs(struct intel_vgpu_workload *workload,
 		i915_mmio_reg_offset(EU_PERF_CNTL6),
 	};
 
-	if (workload->ring_id != RCS)
+	if (workload->ring_id != RCS0)
 		return;
 
 	if (save) {
@@ -149,7 +149,7 @@ static int populate_shadow_context(struct intel_vgpu_workload *workload)
 	COPY_REG_MASKED(ctx_ctrl);
 	COPY_REG(ctx_timestamp);
 
-	if (ring_id == RCS) {
+	if (ring_id == RCS0) {
 		COPY_REG(bb_per_ctx_ptr);
 		COPY_REG(rcs_indirect_ctx);
 		COPY_REG(rcs_indirect_ctx_offset);
@@ -177,7 +177,7 @@ static int populate_shadow_context(struct intel_vgpu_workload *workload)
 
 	context_page_num = context_page_num >> PAGE_SHIFT;
 
-	if (IS_BROADWELL(gvt->dev_priv) && ring_id == RCS)
+	if (IS_BROADWELL(gvt->dev_priv) && ring_id == RCS0)
 		context_page_num = 19;
 
 	i = 2;
@@ -434,8 +434,7 @@ int intel_gvt_scan_and_shadow_workload(struct intel_vgpu_workload *workload)
 	if (ret)
 		goto err_unpin;
 
-	if ((workload->ring_id == RCS) &&
-	    (workload->wa_ctx.indirect_ctx.size != 0)) {
+	if (workload->ring_id == RCS0 && workload->wa_ctx.indirect_ctx.size) {
 		ret = intel_gvt_scan_and_shadow_wa_ctx(&workload->wa_ctx);
 		if (ret)
 			goto err_shadow;
@@ -803,7 +802,7 @@ static void update_guest_context(struct intel_vgpu_workload *workload)
 	context_page_num = rq->engine->context_size;
 	context_page_num = context_page_num >> PAGE_SHIFT;
 
-	if (IS_BROADWELL(gvt->dev_priv) && rq->engine->id == RCS)
+	if (IS_BROADWELL(gvt->dev_priv) && rq->engine->id == RCS0)
 		context_page_num = 19;
 
 	i = 2;
@@ -851,13 +850,13 @@ static void update_guest_context(struct intel_vgpu_workload *workload)
 }
 
 void intel_vgpu_clean_workloads(struct intel_vgpu *vgpu,
-				unsigned long engine_mask)
+				intel_engine_mask_t engine_mask)
 {
 	struct intel_vgpu_submission *s = &vgpu->submission;
 	struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
 	struct intel_engine_cs *engine;
 	struct intel_vgpu_workload *pos, *n;
-	unsigned int tmp;
+	intel_engine_mask_t tmp;
 
 	/* free the unsubmited workloads in the queues. */
 	for_each_engine_masked(engine, dev_priv, engine_mask, tmp) {
@@ -903,8 +902,8 @@ static void complete_current_workload(struct intel_gvt *gvt, int ring_id)
 				workload->status = 0;
 		}
 
-		if (!workload->status && !(vgpu->resetting_eng &
-					   ENGINE_MASK(ring_id))) {
+		if (!workload->status &&
+		    !(vgpu->resetting_eng & BIT(ring_id))) {
 			update_guest_context(workload);
 
 			for_each_set_bit(event, workload->pending_events,
@@ -927,7 +926,7 @@ static void complete_current_workload(struct intel_gvt *gvt, int ring_id)
 
 	list_del_init(&workload->list);
 
-	if (workload->status || (vgpu->resetting_eng & ENGINE_MASK(ring_id))) {
+	if (workload->status || vgpu->resetting_eng & BIT(ring_id)) {
 		/* if workload->status is not successful means HW GPU
 		 * has occurred GPU hang or something wrong with i915/GVT,
 		 * and GVT won't inject context switch interrupt to guest.
@@ -941,7 +940,7 @@ static void complete_current_workload(struct intel_gvt *gvt, int ring_id)
 		 * cleaned up during the resetting process later, so doing
 		 * the workload clean up here doesn't have any impact.
 		 **/
-		intel_vgpu_clean_workloads(vgpu, ENGINE_MASK(ring_id));
+		intel_vgpu_clean_workloads(vgpu, BIT(ring_id));
 	}
 
 	workload->complete(workload);
@@ -1001,7 +1000,7 @@ static int workload_thread(void *priv)
 				workload->ring_id, workload);
 
 		if (need_force_wake)
-			intel_uncore_forcewake_get(gvt->dev_priv,
+			intel_uncore_forcewake_get(&gvt->dev_priv->uncore,
 					FORCEWAKE_ALL);
 
 		ret = dispatch_workload(workload);
@@ -1023,7 +1022,7 @@ complete:
 		complete_current_workload(gvt, ring_id);
 
 		if (need_force_wake)
-			intel_uncore_forcewake_put(gvt->dev_priv,
+			intel_uncore_forcewake_put(&gvt->dev_priv->uncore,
 					FORCEWAKE_ALL);
 
 		intel_runtime_pm_put_unchecked(gvt->dev_priv);
@@ -1114,9 +1113,9 @@ i915_context_ppgtt_root_restore(struct intel_vgpu_submission *s)
 	struct i915_hw_ppgtt *i915_ppgtt = s->shadow_ctx->ppgtt;
 	int i;
 
-	if (i915_vm_is_48bit(&i915_ppgtt->vm))
+	if (i915_vm_is_4lvl(&i915_ppgtt->vm)) {
 		px_dma(&i915_ppgtt->pml4) = s->i915_context_pml4;
-	else {
+	} else {
 		for (i = 0; i < GEN8_3LVL_PDPES; i++)
 			px_dma(i915_ppgtt->pdp.page_directory[i]) =
 						s->i915_context_pdps[i];
@@ -1150,7 +1149,7 @@ void intel_vgpu_clean_submission(struct intel_vgpu *vgpu)
  *
  */
 void intel_vgpu_reset_submission(struct intel_vgpu *vgpu,
-		unsigned long engine_mask)
+				 intel_engine_mask_t engine_mask)
 {
 	struct intel_vgpu_submission *s = &vgpu->submission;
 
@@ -1167,7 +1166,7 @@ i915_context_ppgtt_root_save(struct intel_vgpu_submission *s)
 	struct i915_hw_ppgtt *i915_ppgtt = s->shadow_ctx->ppgtt;
 	int i;
 
-	if (i915_vm_is_48bit(&i915_ppgtt->vm))
+	if (i915_vm_is_4lvl(&i915_ppgtt->vm))
 		s->i915_context_pml4 = px_dma(&i915_ppgtt->pml4);
 	else {
 		for (i = 0; i < GEN8_3LVL_PDPES; i++)
@@ -1240,7 +1239,7 @@ out_shadow_ctx:
  *
  */
 int intel_vgpu_select_submission_ops(struct intel_vgpu *vgpu,
-				     unsigned long engine_mask,
+				     intel_engine_mask_t engine_mask,
 				     unsigned int interface)
 {
 	struct intel_vgpu_submission *s = &vgpu->submission;
@@ -1450,7 +1449,7 @@ intel_vgpu_create_workload(struct intel_vgpu *vgpu, int ring_id,
 	workload->rb_start = start;
 	workload->rb_ctl = ctl;
 
-	if (ring_id == RCS) {
+	if (ring_id == RCS0) {
 		intel_gvt_hypervisor_read_gpa(vgpu, ring_context_gpa +
 			RING_CTX_OFF(bb_per_ctx_ptr.val), &per_ctx, 4);
 		intel_gvt_hypervisor_read_gpa(vgpu, ring_context_gpa +
diff --git a/drivers/gpu/drm/i915/gvt/scheduler.h b/drivers/gpu/drm/i915/gvt/scheduler.h
index 0635b2c4bed7..90c6756f5453 100644
--- a/drivers/gpu/drm/i915/gvt/scheduler.h
+++ b/drivers/gpu/drm/i915/gvt/scheduler.h
@@ -142,12 +142,12 @@ void intel_gvt_wait_vgpu_idle(struct intel_vgpu *vgpu);
 int intel_vgpu_setup_submission(struct intel_vgpu *vgpu);
 
 void intel_vgpu_reset_submission(struct intel_vgpu *vgpu,
-				 unsigned long engine_mask);
+				 intel_engine_mask_t engine_mask);
 
 void intel_vgpu_clean_submission(struct intel_vgpu *vgpu);
 
 int intel_vgpu_select_submission_ops(struct intel_vgpu *vgpu,
-				     unsigned long engine_mask,
+				     intel_engine_mask_t engine_mask,
 				     unsigned int interface);
 
 extern const struct intel_vgpu_submission_ops
@@ -160,6 +160,6 @@ intel_vgpu_create_workload(struct intel_vgpu *vgpu, int ring_id,
 void intel_vgpu_destroy_workload(struct intel_vgpu_workload *workload);
 
 void intel_vgpu_clean_workloads(struct intel_vgpu *vgpu,
-				unsigned long engine_mask);
+				intel_engine_mask_t engine_mask);
 
 #endif
diff --git a/drivers/gpu/drm/i915/gvt/vgpu.c b/drivers/gpu/drm/i915/gvt/vgpu.c
index 720e2b10adaa..44ce3c2b9ac1 100644
--- a/drivers/gpu/drm/i915/gvt/vgpu.c
+++ b/drivers/gpu/drm/i915/gvt/vgpu.c
@@ -44,7 +44,7 @@ void populate_pvinfo_page(struct intel_vgpu *vgpu)
 	vgpu_vreg_t(vgpu, vgtif_reg(display_ready)) = 0;
 	vgpu_vreg_t(vgpu, vgtif_reg(vgt_id)) = vgpu->id;
 
-	vgpu_vreg_t(vgpu, vgtif_reg(vgt_caps)) = VGT_CAPS_FULL_48BIT_PPGTT;
+	vgpu_vreg_t(vgpu, vgtif_reg(vgt_caps)) = VGT_CAPS_FULL_PPGTT;
 	vgpu_vreg_t(vgpu, vgtif_reg(vgt_caps)) |= VGT_CAPS_HWSP_EMULATION;
 	vgpu_vreg_t(vgpu, vgtif_reg(vgt_caps)) |= VGT_CAPS_HUGE_GTT;
 
@@ -526,11 +526,11 @@ struct intel_vgpu *intel_gvt_create_vgpu(struct intel_gvt *gvt,
  * GPU engines. For FLR, engine_mask is ignored.
  */
 void intel_gvt_reset_vgpu_locked(struct intel_vgpu *vgpu, bool dmlr,
-				 unsigned int engine_mask)
+				 intel_engine_mask_t engine_mask)
 {
 	struct intel_gvt *gvt = vgpu->gvt;
 	struct intel_gvt_workload_scheduler *scheduler = &gvt->scheduler;
-	unsigned int resetting_eng = dmlr ? ALL_ENGINES : engine_mask;
+	intel_engine_mask_t resetting_eng = dmlr ? ALL_ENGINES : engine_mask;
 
 	gvt_dbg_core("------------------------------------------\n");
 	gvt_dbg_core("resseting vgpu%d, dmlr %d, engine_mask %08x\n",
diff --git a/drivers/gpu/drm/i915/i915_active.c b/drivers/gpu/drm/i915/i915_active.c
index db7bb5bd5add..863ae12707ba 100644
--- a/drivers/gpu/drm/i915/i915_active.c
+++ b/drivers/gpu/drm/i915/i915_active.c
@@ -6,6 +6,7 @@
 
 #include "i915_drv.h"
 #include "i915_active.h"
+#include "i915_globals.h"
 
 #define BKL(ref) (&(ref)->i915->drm.struct_mutex)
 
@@ -17,6 +18,7 @@
  * nodes from a local slab cache to hopefully reduce the fragmentation.
  */
 static struct i915_global_active {
+	struct i915_global base;
 	struct kmem_cache *slab_cache;
 } global;
 
@@ -285,16 +287,27 @@ void i915_active_retire_noop(struct i915_active_request *active,
 #include "selftests/i915_active.c"
 #endif
 
+static void i915_global_active_shrink(void)
+{
+	kmem_cache_shrink(global.slab_cache);
+}
+
+static void i915_global_active_exit(void)
+{
+	kmem_cache_destroy(global.slab_cache);
+}
+
+static struct i915_global_active global = { {
+	.shrink = i915_global_active_shrink,
+	.exit = i915_global_active_exit,
+} };
+
 int __init i915_global_active_init(void)
 {
 	global.slab_cache = KMEM_CACHE(active_node, SLAB_HWCACHE_ALIGN);
 	if (!global.slab_cache)
 		return -ENOMEM;
 
+	i915_global_register(&global.base);
 	return 0;
 }
-
-void __exit i915_global_active_exit(void)
-{
-	kmem_cache_destroy(global.slab_cache);
-}
diff --git a/drivers/gpu/drm/i915/i915_active.h b/drivers/gpu/drm/i915/i915_active.h
index 12b5c1d287d1..7d758719ce39 100644
--- a/drivers/gpu/drm/i915/i915_active.h
+++ b/drivers/gpu/drm/i915/i915_active.h
@@ -108,19 +108,6 @@ i915_active_request_set_retire_fn(struct i915_active_request *active,
 	active->retire = fn ?: i915_active_retire_noop;
 }
 
-static inline struct i915_request *
-__i915_active_request_peek(const struct i915_active_request *active)
-{
-	/*
-	 * Inside the error capture (running with the driver in an unknown
-	 * state), we want to bend the rules slightly (a lot).
-	 *
-	 * Work is in progress to make it safer, in the meantime this keeps
-	 * the known issue from spamming the logs.
-	 */
-	return rcu_dereference_protected(active->request, 1);
-}
-
 /**
  * i915_active_request_raw - return the active request
  * @active - the active tracker
@@ -419,7 +406,4 @@ void i915_active_fini(struct i915_active *ref);
 static inline void i915_active_fini(struct i915_active *ref) { }
 #endif
 
-int i915_global_active_init(void);
-void i915_global_active_exit(void);
-
 #endif /* _I915_ACTIVE_H_ */
diff --git a/drivers/gpu/drm/i915/i915_cmd_parser.c b/drivers/gpu/drm/i915/i915_cmd_parser.c
index 33e8eed64423..503d548a55f7 100644
--- a/drivers/gpu/drm/i915/i915_cmd_parser.c
+++ b/drivers/gpu/drm/i915/i915_cmd_parser.c
@@ -868,8 +868,8 @@ void intel_engine_init_cmd_parser(struct intel_engine_cs *engine)
 	if (!IS_GEN(engine->i915, 7))
 		return;
 
-	switch (engine->id) {
-	case RCS:
+	switch (engine->class) {
+	case RENDER_CLASS:
 		if (IS_HASWELL(engine->i915)) {
 			cmd_tables = hsw_render_ring_cmds;
 			cmd_table_count =
@@ -889,12 +889,12 @@ void intel_engine_init_cmd_parser(struct intel_engine_cs *engine)
 
 		engine->get_cmd_length_mask = gen7_render_get_cmd_length_mask;
 		break;
-	case VCS:
+	case VIDEO_DECODE_CLASS:
 		cmd_tables = gen7_video_cmds;
 		cmd_table_count = ARRAY_SIZE(gen7_video_cmds);
 		engine->get_cmd_length_mask = gen7_bsd_get_cmd_length_mask;
 		break;
-	case BCS:
+	case COPY_ENGINE_CLASS:
 		if (IS_HASWELL(engine->i915)) {
 			cmd_tables = hsw_blt_ring_cmds;
 			cmd_table_count = ARRAY_SIZE(hsw_blt_ring_cmds);
@@ -913,14 +913,14 @@ void intel_engine_init_cmd_parser(struct intel_engine_cs *engine)
 
 		engine->get_cmd_length_mask = gen7_blt_get_cmd_length_mask;
 		break;
-	case VECS:
+	case VIDEO_ENHANCEMENT_CLASS:
 		cmd_tables = hsw_vebox_cmds;
 		cmd_table_count = ARRAY_SIZE(hsw_vebox_cmds);
 		/* VECS can use the same length_mask function as VCS */
 		engine->get_cmd_length_mask = gen7_bsd_get_cmd_length_mask;
 		break;
 	default:
-		MISSING_CASE(engine->id);
+		MISSING_CASE(engine->class);
 		return;
 	}
 
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index f6f6e5b78e97..5823ffb17821 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -26,14 +26,21 @@
  *
  */
 
-#include <linux/sort.h>
 #include <linux/sched/mm.h>
+#include <linux/sort.h>
+
 #include <drm/drm_debugfs.h>
 #include <drm/drm_fourcc.h>
-#include "intel_drv.h"
-#include "intel_guc_submission.h"
 
 #include "i915_reset.h"
+#include "intel_dp.h"
+#include "intel_drv.h"
+#include "intel_fbc.h"
+#include "intel_guc_submission.h"
+#include "intel_hdcp.h"
+#include "intel_hdmi.h"
+#include "intel_pm.h"
+#include "intel_psr.h"
 
 static inline struct drm_i915_private *node_to_i915(struct drm_info_node *node)
 {
@@ -388,12 +395,9 @@ static void print_context_stats(struct seq_file *m,
 	struct i915_gem_context *ctx;
 
 	list_for_each_entry(ctx, &i915->contexts.list, link) {
-		struct intel_engine_cs *engine;
-		enum intel_engine_id id;
-
-		for_each_engine(engine, i915, id) {
-			struct intel_context *ce = to_intel_context(ctx, engine);
+		struct intel_context *ce;
 
+		list_for_each_entry(ce, &ctx->active_engines, active_link) {
 			if (ce->state)
 				per_file_stats(0, ce->state->obj, &kstats);
 			if (ce->ring)
@@ -412,9 +416,8 @@ static void print_context_stats(struct seq_file *m,
 
 			rcu_read_lock();
 			task = pid_task(ctx->pid ?: file->pid, PIDTYPE_PID);
-			snprintf(name, sizeof(name), "%s/%d",
-				 task ? task->comm : "<unknown>",
-				 ctx->user_handle);
+			snprintf(name, sizeof(name), "%s",
+				 task ? task->comm : "<unknown>");
 			rcu_read_unlock();
 
 			print_file_stats(m, name, stats);
@@ -830,11 +833,11 @@ static int i915_interrupt_info(struct seq_file *m, void *data)
 
 	} else if (!HAS_PCH_SPLIT(dev_priv)) {
 		seq_printf(m, "Interrupt enable:    %08x\n",
-			   I915_READ(IER));
+			   I915_READ(GEN2_IER));
 		seq_printf(m, "Interrupt identity:  %08x\n",
-			   I915_READ(IIR));
+			   I915_READ(GEN2_IIR));
 		seq_printf(m, "Interrupt mask:      %08x\n",
-			   I915_READ(IMR));
+			   I915_READ(GEN2_IMR));
 		for_each_pipe(dev_priv, pipe)
 			seq_printf(m, "Pipe %c stat:         %08x\n",
 				   pipe_name(pipe),
@@ -884,7 +887,7 @@ static int i915_interrupt_info(struct seq_file *m, void *data)
 		for_each_engine(engine, dev_priv, id) {
 			seq_printf(m,
 				   "Graphics Interrupt mask (%s):	%08x\n",
-				   engine->name, I915_READ_IMR(engine));
+				   engine->name, ENGINE_READ(engine, RING_IMR));
 		}
 	}
 
@@ -1097,7 +1100,7 @@ static int i915_frequency_info(struct seq_file *m, void *unused)
 		}
 
 		/* RPSTAT1 is in the GT power well */
-		intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+		intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL);
 
 		reqf = I915_READ(GEN6_RPNSWREQ);
 		if (INTEL_GEN(dev_priv) >= 9)
@@ -1125,7 +1128,7 @@ static int i915_frequency_info(struct seq_file *m, void *unused)
 		cagf = intel_gpu_freq(dev_priv,
 				      intel_get_cagf(dev_priv, rpstat));
 
-		intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+		intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 
 		if (INTEL_GEN(dev_priv) >= 11) {
 			pm_ier = I915_READ(GEN11_GPM_WGBOXPERF_INTR_ENABLE);
@@ -1281,14 +1284,11 @@ static int i915_hangcheck_info(struct seq_file *m, void *unused)
 	intel_wakeref_t wakeref;
 	enum intel_engine_id id;
 
+	seq_printf(m, "Reset flags: %lx\n", dev_priv->gpu_error.flags);
 	if (test_bit(I915_WEDGED, &dev_priv->gpu_error.flags))
-		seq_puts(m, "Wedged\n");
+		seq_puts(m, "\tWedged\n");
 	if (test_bit(I915_RESET_BACKOFF, &dev_priv->gpu_error.flags))
-		seq_puts(m, "Reset in progress: struct_mutex backoff\n");
-	if (waitqueue_active(&dev_priv->gpu_error.wait_queue))
-		seq_puts(m, "Waiter holding struct mutex\n");
-	if (waitqueue_active(&dev_priv->gpu_error.reset_queue))
-		seq_puts(m, "struct_mutex blocked for reset\n");
+		seq_puts(m, "\tDevice (global) reset in progress\n");
 
 	if (!i915_modparams.enable_hangcheck) {
 		seq_puts(m, "Hangcheck disabled\n");
@@ -1298,10 +1298,10 @@ static int i915_hangcheck_info(struct seq_file *m, void *unused)
 	with_intel_runtime_pm(dev_priv, wakeref) {
 		for_each_engine(engine, dev_priv, id) {
 			acthd[id] = intel_engine_get_active_head(engine);
-			seqno[id] = intel_engine_get_seqno(engine);
+			seqno[id] = intel_engine_get_hangcheck_seqno(engine);
 		}
 
-		intel_engine_get_instdone(dev_priv->engine[RCS], &instdone);
+		intel_engine_get_instdone(dev_priv->engine[RCS0], &instdone);
 	}
 
 	if (timer_pending(&dev_priv->gpu_error.hangcheck_work.timer))
@@ -1318,8 +1318,9 @@ static int i915_hangcheck_info(struct seq_file *m, void *unused)
 	for_each_engine(engine, dev_priv, id) {
 		seq_printf(m, "%s:\n", engine->name);
 		seq_printf(m, "\tseqno = %x [current %x, last %x], %dms ago\n",
-			   engine->hangcheck.seqno, seqno[id],
-			   intel_engine_last_submit(engine),
+			   engine->hangcheck.last_seqno,
+			   seqno[id],
+			   engine->hangcheck.next_seqno,
 			   jiffies_to_msecs(jiffies -
 					    engine->hangcheck.action_timestamp));
 
@@ -1327,7 +1328,7 @@ static int i915_hangcheck_info(struct seq_file *m, void *unused)
 			   (long long)engine->hangcheck.acthd,
 			   (long long)acthd[id]);
 
-		if (engine->id == RCS) {
+		if (engine->id == RCS0) {
 			seq_puts(m, "\tinstdone read =\n");
 
 			i915_instdone_info(dev_priv, m, &instdone);
@@ -1419,13 +1420,14 @@ static int ironlake_drpc_info(struct seq_file *m)
 static int i915_forcewake_domains(struct seq_file *m, void *data)
 {
 	struct drm_i915_private *i915 = node_to_i915(m->private);
+	struct intel_uncore *uncore = &i915->uncore;
 	struct intel_uncore_forcewake_domain *fw_domain;
 	unsigned int tmp;
 
 	seq_printf(m, "user.bypass_count = %u\n",
-		   i915->uncore.user_forcewake.count);
+		   uncore->user_forcewake.count);
 
-	for_each_fw_domain(fw_domain, i915, tmp)
+	for_each_fw_domain(fw_domain, uncore, tmp)
 		seq_printf(m, "%s.wake_count = %u\n",
 			   intel_uncore_forcewake_domain_to_str(fw_domain->id),
 			   READ_ONCE(fw_domain->wake_count));
@@ -1882,9 +1884,7 @@ static int i915_context_status(struct seq_file *m, void *unused)
 {
 	struct drm_i915_private *dev_priv = node_to_i915(m->private);
 	struct drm_device *dev = &dev_priv->drm;
-	struct intel_engine_cs *engine;
 	struct i915_gem_context *ctx;
-	enum intel_engine_id id;
 	int ret;
 
 	ret = mutex_lock_interruptible(&dev->struct_mutex);
@@ -1892,6 +1892,8 @@ static int i915_context_status(struct seq_file *m, void *unused)
 		return ret;
 
 	list_for_each_entry(ctx, &dev_priv->contexts.list, link) {
+		struct intel_context *ce;
+
 		seq_puts(m, "HW context ");
 		if (!list_empty(&ctx->hw_id_link))
 			seq_printf(m, "%x [pin %u]", ctx->hw_id,
@@ -1914,11 +1916,8 @@ static int i915_context_status(struct seq_file *m, void *unused)
 		seq_putc(m, ctx->remap_slice ? 'R' : 'r');
 		seq_putc(m, '\n');
 
-		for_each_engine(engine, dev_priv, id) {
-			struct intel_context *ce =
-				to_intel_context(ctx, engine);
-
-			seq_printf(m, "%s: ", engine->name);
+		list_for_each_entry(ce, &ctx->active_engines, active_link) {
+			seq_printf(m, "%s: ", ce->engine->name);
 			if (ce->state)
 				describe_obj(m, ce->state->obj);
 			if (ce->ring)
@@ -2023,11 +2022,9 @@ static const char *rps_power_to_str(unsigned int power)
 static int i915_rps_boost_info(struct seq_file *m, void *data)
 {
 	struct drm_i915_private *dev_priv = node_to_i915(m->private);
-	struct drm_device *dev = &dev_priv->drm;
 	struct intel_rps *rps = &dev_priv->gt_pm.rps;
 	u32 act_freq = rps->cur_freq;
 	intel_wakeref_t wakeref;
-	struct drm_file *file;
 
 	with_intel_runtime_pm_if_in_use(dev_priv, wakeref) {
 		if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {
@@ -2061,22 +2058,7 @@ static int i915_rps_boost_info(struct seq_file *m, void *data)
 		   intel_gpu_freq(dev_priv, rps->efficient_freq),
 		   intel_gpu_freq(dev_priv, rps->boost_freq));
 
-	mutex_lock(&dev->filelist_mutex);
-	list_for_each_entry_reverse(file, &dev->filelist, lhead) {
-		struct drm_i915_file_private *file_priv = file->driver_priv;
-		struct task_struct *task;
-
-		rcu_read_lock();
-		task = pid_task(file->pid, PIDTYPE_PID);
-		seq_printf(m, "%s [%d]: %d boosts\n",
-			   task ? task->comm : "<unknown>",
-			   task ? task->pid : -1,
-			   atomic_read(&file_priv->rps_client.boosts));
-		rcu_read_unlock();
-	}
-	seq_printf(m, "Kernel (anonymous) boosts: %d\n",
-		   atomic_read(&rps->boosts));
-	mutex_unlock(&dev->filelist_mutex);
+	seq_printf(m, "Wait boosts: %d\n", atomic_read(&rps->boosts));
 
 	if (INTEL_GEN(dev_priv) >= 6 &&
 	    rps->enabled &&
@@ -2084,12 +2066,12 @@ static int i915_rps_boost_info(struct seq_file *m, void *data)
 		u32 rpup, rpupei;
 		u32 rpdown, rpdownei;
 
-		intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+		intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL);
 		rpup = I915_READ_FW(GEN6_RP_CUR_UP) & GEN6_RP_EI_MASK;
 		rpupei = I915_READ_FW(GEN6_RP_CUR_UP_EI) & GEN6_RP_EI_MASK;
 		rpdown = I915_READ_FW(GEN6_RP_CUR_DOWN) & GEN6_RP_EI_MASK;
 		rpdownei = I915_READ_FW(GEN6_RP_CUR_DOWN_EI) & GEN6_RP_EI_MASK;
-		intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+		intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 
 		seq_printf(m, "\nRPS Autotuning (current \"%s\" window):\n",
 			   rps_power_to_str(rps->power.mode));
@@ -2112,8 +2094,8 @@ static int i915_llc(struct seq_file *m, void *data)
 	const bool edram = INTEL_GEN(dev_priv) > 8;
 
 	seq_printf(m, "LLC: %s\n", yesno(HAS_LLC(dev_priv)));
-	seq_printf(m, "%s: %lluMB\n", edram ? "eDRAM" : "eLLC",
-		   intel_uncore_edram_size(dev_priv)/1024/1024);
+	seq_printf(m, "%s: %uMB\n", edram ? "eDRAM" : "eLLC",
+		   dev_priv->edram_size_mb);
 
 	return 0;
 }
@@ -2270,7 +2252,7 @@ static int i915_guc_stage_pool(struct seq_file *m, void *data)
 	const struct intel_guc *guc = &dev_priv->guc;
 	struct guc_stage_desc *desc = guc->stage_desc_pool_vaddr;
 	struct intel_guc_client *client = guc->execbuf_client;
-	unsigned int tmp;
+	intel_engine_mask_t tmp;
 	int index;
 
 	if (!USES_GUC_SUBMISSION(dev_priv))
@@ -2607,7 +2589,6 @@ static int
 i915_edp_psr_debug_set(void *data, u64 val)
 {
 	struct drm_i915_private *dev_priv = data;
-	struct drm_modeset_acquire_ctx ctx;
 	intel_wakeref_t wakeref;
 	int ret;
 
@@ -2618,18 +2599,7 @@ i915_edp_psr_debug_set(void *data, u64 val)
 
 	wakeref = intel_runtime_pm_get(dev_priv);
 
-	drm_modeset_acquire_init(&ctx, DRM_MODESET_ACQUIRE_INTERRUPTIBLE);
-
-retry:
-	ret = intel_psr_set_debugfs_mode(dev_priv, &ctx, val);
-	if (ret == -EDEADLK) {
-		ret = drm_modeset_backoff(&ctx);
-		if (!ret)
-			goto retry;
-	}
-
-	drm_modeset_drop_locks(&ctx);
-	drm_modeset_acquire_fini(&ctx);
+	ret = intel_psr_debug_set(dev_priv, val);
 
 	intel_runtime_pm_put(dev_priv, wakeref);
 
@@ -2686,8 +2656,7 @@ static int i915_runtime_pm_status(struct seq_file *m, void *unused)
 	seq_printf(m, "Runtime power status: %s\n",
 		   enableddisabled(!dev_priv->power_domains.wakeref));
 
-	seq_printf(m, "GPU idle: %s (epoch %u)\n",
-		   yesno(!dev_priv->gt.awake), dev_priv->gt.epoch);
+	seq_printf(m, "GPU idle: %s\n", yesno(!dev_priv->gt.awake));
 	seq_printf(m, "IRQs disabled: %s\n",
 		   yesno(!intel_irqs_enabled(dev_priv)));
 #ifdef CONFIG_PM
@@ -2904,7 +2873,6 @@ static void intel_connector_info(struct seq_file *m,
 	if (connector->status == connector_status_disconnected)
 		return;
 
-	seq_printf(m, "\tname: %s\n", connector->display_info.name);
 	seq_printf(m, "\tphysical dimensions: %dx%dmm\n",
 		   connector->display_info.width_mm,
 		   connector->display_info.height_mm);
@@ -3123,8 +3091,7 @@ static int i915_engine_info(struct seq_file *m, void *unused)
 
 	wakeref = intel_runtime_pm_get(dev_priv);
 
-	seq_printf(m, "GT awake? %s (epoch %u)\n",
-		   yesno(dev_priv->gt.awake), dev_priv->gt.epoch);
+	seq_printf(m, "GT awake? %s\n", yesno(dev_priv->gt.awake));
 	seq_printf(m, "Global active requests: %d\n",
 		   dev_priv->gt.active_requests);
 	seq_printf(m, "CS timestamp frequency: %u kHz\n",
@@ -3211,7 +3178,7 @@ static int i915_shared_dplls_info(struct seq_file *m, void *unused)
 static int i915_wa_registers(struct seq_file *m, void *unused)
 {
 	struct drm_i915_private *i915 = node_to_i915(m->private);
-	const struct i915_wa_list *wal = &i915->engine[RCS]->ctx_wa_list;
+	const struct i915_wa_list *wal = &i915->engine[RCS0]->ctx_wa_list;
 	struct i915_wa *wa;
 	unsigned int i;
 
@@ -3865,11 +3832,18 @@ static const struct file_operations i915_cur_wm_latency_fops = {
 static int
 i915_wedged_get(void *data, u64 *val)
 {
-	struct drm_i915_private *dev_priv = data;
-
-	*val = i915_terminally_wedged(&dev_priv->gpu_error);
+	int ret = i915_terminally_wedged(data);
 
-	return 0;
+	switch (ret) {
+	case -EIO:
+		*val = 1;
+		return 0;
+	case 0:
+		*val = 0;
+		return 0;
+	default:
+		return ret;
+	}
 }
 
 static int
@@ -3877,16 +3851,9 @@ i915_wedged_set(void *data, u64 val)
 {
 	struct drm_i915_private *i915 = data;
 
-	/*
-	 * There is no safeguard against this debugfs entry colliding
-	 * with the hangcheck calling same i915_handle_error() in
-	 * parallel, causing an explosion. For now we assume that the
-	 * test harness is responsible enough not to inject gpu hangs
-	 * while it is writing to 'i915_wedged'
-	 */
-
-	if (i915_reset_backoff(&i915->gpu_error))
-		return -EAGAIN;
+	/* Flush any previous reset before applying for a new one */
+	wait_event(i915->gpu_error.reset_queue,
+		   !test_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags));
 
 	i915_handle_error(i915, val, I915_ERROR_CAPTURE,
 			  "Manually set wedged engine mask = %llx", val);
@@ -3927,12 +3894,9 @@ static int
 i915_drop_caches_set(void *data, u64 val)
 {
 	struct drm_i915_private *i915 = data;
-	intel_wakeref_t wakeref;
-	int ret = 0;
 
 	DRM_DEBUG("Dropping caches: 0x%08llx [0x%08llx]\n",
 		  val, val & DROP_ALL);
-	wakeref = intel_runtime_pm_get(i915);
 
 	if (val & DROP_RESET_ACTIVE &&
 	    wait_for(intel_engines_are_idle(i915), I915_IDLE_ENGINES_TIMEOUT))
@@ -3941,9 +3905,11 @@ i915_drop_caches_set(void *data, u64 val)
 	/* No need to check and wait for gpu resets, only libdrm auto-restarts
 	 * on ioctls on -EAGAIN. */
 	if (val & (DROP_ACTIVE | DROP_RETIRE | DROP_RESET_SEQNO)) {
+		int ret;
+
 		ret = mutex_lock_interruptible(&i915->drm.struct_mutex);
 		if (ret)
-			goto out;
+			return ret;
 
 		if (val & DROP_ACTIVE)
 			ret = i915_gem_wait_for_idle(i915,
@@ -3957,7 +3923,7 @@ i915_drop_caches_set(void *data, u64 val)
 		mutex_unlock(&i915->drm.struct_mutex);
 	}
 
-	if (val & DROP_RESET_ACTIVE && i915_terminally_wedged(&i915->gpu_error))
+	if (val & DROP_RESET_ACTIVE && i915_terminally_wedged(i915))
 		i915_handle_error(i915, ALL_ENGINES, 0, NULL);
 
 	fs_reclaim_acquire(GFP_KERNEL);
@@ -3982,10 +3948,7 @@ i915_drop_caches_set(void *data, u64 val)
 	if (val & DROP_FREED)
 		i915_gem_drain_freed_objects(i915);
 
-out:
-	intel_runtime_pm_put(i915, wakeref);
-
-	return ret;
+	return 0;
 }
 
 DEFINE_SIMPLE_ATTRIBUTE(i915_drop_caches_fops,
@@ -4293,7 +4256,7 @@ static int i915_forcewake_open(struct inode *inode, struct file *file)
 		return 0;
 
 	file->private_data = (void *)(uintptr_t)intel_runtime_pm_get(i915);
-	intel_uncore_forcewake_user_get(i915);
+	intel_uncore_forcewake_user_get(&i915->uncore);
 
 	return 0;
 }
@@ -4305,7 +4268,7 @@ static int i915_forcewake_release(struct inode *inode, struct file *file)
 	if (INTEL_GEN(i915) < 6)
 		return 0;
 
-	intel_uncore_forcewake_user_put(i915);
+	intel_uncore_forcewake_user_put(&i915->uncore);
 	intel_runtime_pm_put(i915,
 			     (intel_wakeref_t)(uintptr_t)file->private_data);
 
@@ -4858,6 +4821,8 @@ static int i915_dsc_fec_support_show(struct seq_file *m, void *data)
 			   yesno(crtc_state->dsc_params.compression_enable));
 		seq_printf(m, "DSC_Sink_Support: %s\n",
 			   yesno(drm_dp_sink_supports_dsc(intel_dp->dsc_dpcd)));
+		seq_printf(m, "Force_DSC_Enable: %s\n",
+			   yesno(intel_dp->force_dsc_en));
 		if (!intel_dp_is_edp(intel_dp))
 			seq_printf(m, "FEC_Sink_Support: %s\n",
 				   yesno(drm_dp_sink_supports_fec(intel_dp->fec_capable)));
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 9df65d386d11..1ad88e6d7c04 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -48,12 +48,19 @@
 #include <drm/i915_drm.h>
 
 #include "i915_drv.h"
-#include "i915_trace.h"
 #include "i915_pmu.h"
-#include "i915_reset.h"
 #include "i915_query.h"
+#include "i915_reset.h"
+#include "i915_trace.h"
 #include "i915_vgpu.h"
+#include "intel_audio.h"
+#include "intel_cdclk.h"
+#include "intel_csr.h"
+#include "intel_dp.h"
 #include "intel_drv.h"
+#include "intel_fbdev.h"
+#include "intel_pm.h"
+#include "intel_sprite.h"
 #include "intel_uc.h"
 #include "intel_workarounds.h"
 
@@ -188,6 +195,11 @@ intel_pch_type(const struct drm_i915_private *dev_priv, unsigned short id)
 		DRM_DEBUG_KMS("Found Cannon Lake LP PCH (CNP-LP)\n");
 		WARN_ON(!IS_CANNONLAKE(dev_priv) && !IS_COFFEELAKE(dev_priv));
 		return PCH_CNP;
+	case INTEL_PCH_CMP_DEVICE_ID_TYPE:
+		DRM_DEBUG_KMS("Found Comet Lake PCH (CMP)\n");
+		WARN_ON(!IS_COFFEELAKE(dev_priv));
+		/* CometPoint is CNP Compatible */
+		return PCH_CNP;
 	case INTEL_PCH_ICP_DEVICE_ID_TYPE:
 		DRM_DEBUG_KMS("Found Ice Lake PCH\n");
 		WARN_ON(!IS_ICELAKE(dev_priv));
@@ -219,20 +231,20 @@ intel_virt_detect_pch(const struct drm_i915_private *dev_priv)
 	 * make an educated guess as to which PCH is really there.
 	 */
 
-	if (IS_GEN(dev_priv, 5))
-		id = INTEL_PCH_IBX_DEVICE_ID_TYPE;
-	else if (IS_GEN(dev_priv, 6) || IS_IVYBRIDGE(dev_priv))
-		id = INTEL_PCH_CPT_DEVICE_ID_TYPE;
+	if (IS_ICELAKE(dev_priv))
+		id = INTEL_PCH_ICP_DEVICE_ID_TYPE;
+	else if (IS_CANNONLAKE(dev_priv) || IS_COFFEELAKE(dev_priv))
+		id = INTEL_PCH_CNP_DEVICE_ID_TYPE;
+	else if (IS_KABYLAKE(dev_priv) || IS_SKYLAKE(dev_priv))
+		id = INTEL_PCH_SPT_DEVICE_ID_TYPE;
 	else if (IS_HSW_ULT(dev_priv) || IS_BDW_ULT(dev_priv))
 		id = INTEL_PCH_LPT_LP_DEVICE_ID_TYPE;
 	else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
 		id = INTEL_PCH_LPT_DEVICE_ID_TYPE;
-	else if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv))
-		id = INTEL_PCH_SPT_DEVICE_ID_TYPE;
-	else if (IS_COFFEELAKE(dev_priv) || IS_CANNONLAKE(dev_priv))
-		id = INTEL_PCH_CNP_DEVICE_ID_TYPE;
-	else if (IS_ICELAKE(dev_priv))
-		id = INTEL_PCH_ICP_DEVICE_ID_TYPE;
+	else if (IS_GEN(dev_priv, 6) || IS_IVYBRIDGE(dev_priv))
+		id = INTEL_PCH_CPT_DEVICE_ID_TYPE;
+	else if (IS_GEN(dev_priv, 5))
+		id = INTEL_PCH_IBX_DEVICE_ID_TYPE;
 
 	if (id)
 		DRM_DEBUG_KMS("Assuming PCH ID %04x\n", id);
@@ -330,16 +342,16 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data,
 		value = dev_priv->overlay ? 1 : 0;
 		break;
 	case I915_PARAM_HAS_BSD:
-		value = !!dev_priv->engine[VCS];
+		value = !!dev_priv->engine[VCS0];
 		break;
 	case I915_PARAM_HAS_BLT:
-		value = !!dev_priv->engine[BCS];
+		value = !!dev_priv->engine[BCS0];
 		break;
 	case I915_PARAM_HAS_VEBOX:
-		value = !!dev_priv->engine[VECS];
+		value = !!dev_priv->engine[VECS0];
 		break;
 	case I915_PARAM_HAS_BSD2:
-		value = !!dev_priv->engine[VCS2];
+		value = !!dev_priv->engine[VCS1];
 		break;
 	case I915_PARAM_HAS_LLC:
 		value = HAS_LLC(dev_priv);
@@ -348,10 +360,10 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data,
 		value = HAS_WT(dev_priv);
 		break;
 	case I915_PARAM_HAS_ALIASING_PPGTT:
-		value = min_t(int, INTEL_PPGTT(dev_priv), I915_GEM_PPGTT_FULL);
+		value = INTEL_PPGTT(dev_priv);
 		break;
 	case I915_PARAM_HAS_SEMAPHORES:
-		value = 0;
+		value = !!(dev_priv->caps.scheduler & I915_SCHEDULER_CAP_SEMAPHORES);
 		break;
 	case I915_PARAM_HAS_SECURE_BATCHES:
 		value = capable(CAP_SYS_ADMIN);
@@ -714,8 +726,7 @@ static int i915_load_modeset_init(struct drm_device *dev)
 	return 0;
 
 cleanup_gem:
-	if (i915_gem_suspend(dev_priv))
-		DRM_ERROR("failed to idle hardware; continuing to unload!\n");
+	i915_gem_suspend(dev_priv);
 	i915_gem_fini(dev_priv);
 cleanup_modeset:
 	intel_modeset_cleanup(dev);
@@ -864,15 +875,19 @@ static int i915_driver_init_early(struct drm_i915_private *dev_priv)
 	if (i915_inject_load_failure())
 		return -ENODEV;
 
+	intel_device_info_subplatform_init(dev_priv);
+
+	intel_uncore_init_early(&dev_priv->uncore);
+
 	spin_lock_init(&dev_priv->irq_lock);
 	spin_lock_init(&dev_priv->gpu_error.lock);
 	mutex_init(&dev_priv->backlight_lock);
-	spin_lock_init(&dev_priv->uncore.lock);
 
 	mutex_init(&dev_priv->sb_lock);
 	mutex_init(&dev_priv->av_mutex);
 	mutex_init(&dev_priv->wm.wm_mutex);
 	mutex_init(&dev_priv->pps_mutex);
+	mutex_init(&dev_priv->hdcp_comp_mutex);
 
 	i915_memcpy_init_early(dev_priv);
 	intel_runtime_pm_init_early(dev_priv);
@@ -930,46 +945,6 @@ static void i915_driver_cleanup_early(struct drm_i915_private *dev_priv)
 	i915_engines_cleanup(dev_priv);
 }
 
-static int i915_mmio_setup(struct drm_i915_private *dev_priv)
-{
-	struct pci_dev *pdev = dev_priv->drm.pdev;
-	int mmio_bar;
-	int mmio_size;
-
-	mmio_bar = IS_GEN(dev_priv, 2) ? 1 : 0;
-	/*
-	 * Before gen4, the registers and the GTT are behind different BARs.
-	 * However, from gen4 onwards, the registers and the GTT are shared
-	 * in the same BAR, so we want to restrict this ioremap from
-	 * clobbering the GTT which we want ioremap_wc instead. Fortunately,
-	 * the register BAR remains the same size for all the earlier
-	 * generations up to Ironlake.
-	 */
-	if (INTEL_GEN(dev_priv) < 5)
-		mmio_size = 512 * 1024;
-	else
-		mmio_size = 2 * 1024 * 1024;
-	dev_priv->regs = pci_iomap(pdev, mmio_bar, mmio_size);
-	if (dev_priv->regs == NULL) {
-		DRM_ERROR("failed to map registers\n");
-
-		return -EIO;
-	}
-
-	/* Try to make sure MCHBAR is enabled before poking at it */
-	intel_setup_mchbar(dev_priv);
-
-	return 0;
-}
-
-static void i915_mmio_cleanup(struct drm_i915_private *dev_priv)
-{
-	struct pci_dev *pdev = dev_priv->drm.pdev;
-
-	intel_teardown_mchbar(dev_priv);
-	pci_iounmap(pdev, dev_priv->regs);
-}
-
 /**
  * i915_driver_init_mmio - setup device MMIO
  * @dev_priv: device private
@@ -989,15 +964,16 @@ static int i915_driver_init_mmio(struct drm_i915_private *dev_priv)
 	if (i915_get_bridge_dev(dev_priv))
 		return -EIO;
 
-	ret = i915_mmio_setup(dev_priv);
+	ret = intel_uncore_init_mmio(&dev_priv->uncore);
 	if (ret < 0)
 		goto err_bridge;
 
-	intel_uncore_init(dev_priv);
+	/* Try to make sure MCHBAR is enabled before poking at it */
+	intel_setup_mchbar(dev_priv);
 
 	intel_device_info_init_mmio(dev_priv);
 
-	intel_uncore_prune(dev_priv);
+	intel_uncore_prune_mmio_domains(&dev_priv->uncore);
 
 	intel_uc_init_mmio(dev_priv);
 
@@ -1010,8 +986,8 @@ static int i915_driver_init_mmio(struct drm_i915_private *dev_priv)
 	return 0;
 
 err_uncore:
-	intel_uncore_fini(dev_priv);
-	i915_mmio_cleanup(dev_priv);
+	intel_teardown_mchbar(dev_priv);
+	intel_uncore_fini_mmio(&dev_priv->uncore);
 err_bridge:
 	pci_dev_put(dev_priv->bridge_dev);
 
@@ -1024,8 +1000,8 @@ err_bridge:
  */
 static void i915_driver_cleanup_mmio(struct drm_i915_private *dev_priv)
 {
-	intel_uncore_fini(dev_priv);
-	i915_mmio_cleanup(dev_priv);
+	intel_teardown_mchbar(dev_priv);
+	intel_uncore_fini_mmio(&dev_priv->uncore);
 	pci_dev_put(dev_priv->bridge_dev);
 }
 
@@ -1034,110 +1010,180 @@ static void intel_sanitize_options(struct drm_i915_private *dev_priv)
 	intel_gvt_sanitize_options(dev_priv);
 }
 
-static enum dram_rank skl_get_dimm_rank(u8 size, u32 rank)
+#define DRAM_TYPE_STR(type) [INTEL_DRAM_ ## type] = #type
+
+static const char *intel_dram_type_str(enum intel_dram_type type)
 {
-	if (size == 0)
-		return I915_DRAM_RANK_INVALID;
-	if (rank == SKL_DRAM_RANK_SINGLE)
-		return I915_DRAM_RANK_SINGLE;
-	else if (rank == SKL_DRAM_RANK_DUAL)
-		return I915_DRAM_RANK_DUAL;
+	static const char * const str[] = {
+		DRAM_TYPE_STR(UNKNOWN),
+		DRAM_TYPE_STR(DDR3),
+		DRAM_TYPE_STR(DDR4),
+		DRAM_TYPE_STR(LPDDR3),
+		DRAM_TYPE_STR(LPDDR4),
+	};
+
+	if (type >= ARRAY_SIZE(str))
+		type = INTEL_DRAM_UNKNOWN;
 
-	return I915_DRAM_RANK_INVALID;
+	return str[type];
 }
 
-static bool
-skl_is_16gb_dimm(enum dram_rank rank, u8 size, u8 width)
+#undef DRAM_TYPE_STR
+
+static int intel_dimm_num_devices(const struct dram_dimm_info *dimm)
 {
-	if (rank == I915_DRAM_RANK_SINGLE && width == 8 && size == 16)
-		return true;
-	else if (rank == I915_DRAM_RANK_DUAL && width == 8 && size == 32)
-		return true;
-	else if (rank == SKL_DRAM_RANK_SINGLE && width == 16 && size == 8)
-		return true;
-	else if (rank == SKL_DRAM_RANK_DUAL && width == 16 && size == 16)
-		return true;
+	return dimm->ranks * 64 / (dimm->width ?: 1);
+}
 
-	return false;
+/* Returns total GB for the whole DIMM */
+static int skl_get_dimm_size(u16 val)
+{
+	return val & SKL_DRAM_SIZE_MASK;
 }
 
-static int
-skl_dram_get_channel_info(struct dram_channel_info *ch, u32 val)
+static int skl_get_dimm_width(u16 val)
 {
-	u32 tmp_l, tmp_s;
-	u32 s_val = val >> SKL_DRAM_S_SHIFT;
+	if (skl_get_dimm_size(val) == 0)
+		return 0;
 
-	if (!val)
-		return -EINVAL;
+	switch (val & SKL_DRAM_WIDTH_MASK) {
+	case SKL_DRAM_WIDTH_X8:
+	case SKL_DRAM_WIDTH_X16:
+	case SKL_DRAM_WIDTH_X32:
+		val = (val & SKL_DRAM_WIDTH_MASK) >> SKL_DRAM_WIDTH_SHIFT;
+		return 8 << val;
+	default:
+		MISSING_CASE(val);
+		return 0;
+	}
+}
+
+static int skl_get_dimm_ranks(u16 val)
+{
+	if (skl_get_dimm_size(val) == 0)
+		return 0;
+
+	val = (val & SKL_DRAM_RANK_MASK) >> SKL_DRAM_RANK_SHIFT;
+
+	return val + 1;
+}
+
+/* Returns total GB for the whole DIMM */
+static int cnl_get_dimm_size(u16 val)
+{
+	return (val & CNL_DRAM_SIZE_MASK) / 2;
+}
+
+static int cnl_get_dimm_width(u16 val)
+{
+	if (cnl_get_dimm_size(val) == 0)
+		return 0;
+
+	switch (val & CNL_DRAM_WIDTH_MASK) {
+	case CNL_DRAM_WIDTH_X8:
+	case CNL_DRAM_WIDTH_X16:
+	case CNL_DRAM_WIDTH_X32:
+		val = (val & CNL_DRAM_WIDTH_MASK) >> CNL_DRAM_WIDTH_SHIFT;
+		return 8 << val;
+	default:
+		MISSING_CASE(val);
+		return 0;
+	}
+}
+
+static int cnl_get_dimm_ranks(u16 val)
+{
+	if (cnl_get_dimm_size(val) == 0)
+		return 0;
+
+	val = (val & CNL_DRAM_RANK_MASK) >> CNL_DRAM_RANK_SHIFT;
+
+	return val + 1;
+}
 
-	tmp_l = val & SKL_DRAM_SIZE_MASK;
-	tmp_s = s_val & SKL_DRAM_SIZE_MASK;
+static bool
+skl_is_16gb_dimm(const struct dram_dimm_info *dimm)
+{
+	/* Convert total GB to Gb per DRAM device */
+	return 8 * dimm->size / (intel_dimm_num_devices(dimm) ?: 1) == 16;
+}
+
+static void
+skl_dram_get_dimm_info(struct drm_i915_private *dev_priv,
+		       struct dram_dimm_info *dimm,
+		       int channel, char dimm_name, u16 val)
+{
+	if (INTEL_GEN(dev_priv) >= 10) {
+		dimm->size = cnl_get_dimm_size(val);
+		dimm->width = cnl_get_dimm_width(val);
+		dimm->ranks = cnl_get_dimm_ranks(val);
+	} else {
+		dimm->size = skl_get_dimm_size(val);
+		dimm->width = skl_get_dimm_width(val);
+		dimm->ranks = skl_get_dimm_ranks(val);
+	}
+
+	DRM_DEBUG_KMS("CH%u DIMM %c size: %u GB, width: X%u, ranks: %u, 16Gb DIMMs: %s\n",
+		      channel, dimm_name, dimm->size, dimm->width, dimm->ranks,
+		      yesno(skl_is_16gb_dimm(dimm)));
+}
+
+static int
+skl_dram_get_channel_info(struct drm_i915_private *dev_priv,
+			  struct dram_channel_info *ch,
+			  int channel, u32 val)
+{
+	skl_dram_get_dimm_info(dev_priv, &ch->dimm_l,
+			       channel, 'L', val & 0xffff);
+	skl_dram_get_dimm_info(dev_priv, &ch->dimm_s,
+			       channel, 'S', val >> 16);
 
-	if (tmp_l == 0 && tmp_s == 0)
+	if (ch->dimm_l.size == 0 && ch->dimm_s.size == 0) {
+		DRM_DEBUG_KMS("CH%u not populated\n", channel);
 		return -EINVAL;
+	}
 
-	ch->l_info.size = tmp_l;
-	ch->s_info.size = tmp_s;
-
-	tmp_l = (val & SKL_DRAM_WIDTH_MASK) >> SKL_DRAM_WIDTH_SHIFT;
-	tmp_s = (s_val & SKL_DRAM_WIDTH_MASK) >> SKL_DRAM_WIDTH_SHIFT;
-	ch->l_info.width = (1 << tmp_l) * 8;
-	ch->s_info.width = (1 << tmp_s) * 8;
-
-	tmp_l = val & SKL_DRAM_RANK_MASK;
-	tmp_s = s_val & SKL_DRAM_RANK_MASK;
-	ch->l_info.rank = skl_get_dimm_rank(ch->l_info.size, tmp_l);
-	ch->s_info.rank = skl_get_dimm_rank(ch->s_info.size, tmp_s);
-
-	if (ch->l_info.rank == I915_DRAM_RANK_DUAL ||
-	    ch->s_info.rank == I915_DRAM_RANK_DUAL)
-		ch->rank = I915_DRAM_RANK_DUAL;
-	else if (ch->l_info.rank == I915_DRAM_RANK_SINGLE &&
-		 ch->s_info.rank == I915_DRAM_RANK_SINGLE)
-		ch->rank = I915_DRAM_RANK_DUAL;
+	if (ch->dimm_l.ranks == 2 || ch->dimm_s.ranks == 2)
+		ch->ranks = 2;
+	else if (ch->dimm_l.ranks == 1 && ch->dimm_s.ranks == 1)
+		ch->ranks = 2;
 	else
-		ch->rank = I915_DRAM_RANK_SINGLE;
+		ch->ranks = 1;
 
-	ch->is_16gb_dimm = skl_is_16gb_dimm(ch->l_info.rank, ch->l_info.size,
-					    ch->l_info.width) ||
-			   skl_is_16gb_dimm(ch->s_info.rank, ch->s_info.size,
-					    ch->s_info.width);
+	ch->is_16gb_dimm =
+		skl_is_16gb_dimm(&ch->dimm_l) ||
+		skl_is_16gb_dimm(&ch->dimm_s);
 
-	DRM_DEBUG_KMS("(size:width:rank) L(%dGB:X%d:%s) S(%dGB:X%d:%s)\n",
-		      ch->l_info.size, ch->l_info.width,
-		      ch->l_info.rank ? "dual" : "single",
-		      ch->s_info.size, ch->s_info.width,
-		      ch->s_info.rank ? "dual" : "single");
+	DRM_DEBUG_KMS("CH%u ranks: %u, 16Gb DIMMs: %s\n",
+		      channel, ch->ranks, yesno(ch->is_16gb_dimm));
 
 	return 0;
 }
 
 static bool
-intel_is_dram_symmetric(u32 val_ch0, u32 val_ch1,
-			struct dram_channel_info *ch0)
+intel_is_dram_symmetric(const struct dram_channel_info *ch0,
+			const struct dram_channel_info *ch1)
 {
-	return (val_ch0 == val_ch1 &&
-		(ch0->s_info.size == 0 ||
-		 (ch0->l_info.size == ch0->s_info.size &&
-		  ch0->l_info.width == ch0->s_info.width &&
-		  ch0->l_info.rank == ch0->s_info.rank)));
+	return !memcmp(ch0, ch1, sizeof(*ch0)) &&
+		(ch0->dimm_s.size == 0 ||
+		 !memcmp(&ch0->dimm_l, &ch0->dimm_s, sizeof(ch0->dimm_l)));
 }
 
 static int
 skl_dram_get_channels_info(struct drm_i915_private *dev_priv)
 {
 	struct dram_info *dram_info = &dev_priv->dram_info;
-	struct dram_channel_info ch0, ch1;
-	u32 val_ch0, val_ch1;
+	struct dram_channel_info ch0 = {}, ch1 = {};
+	u32 val;
 	int ret;
 
-	val_ch0 = I915_READ(SKL_MAD_DIMM_CH0_0_0_0_MCHBAR_MCMAIN);
-	ret = skl_dram_get_channel_info(&ch0, val_ch0);
+	val = I915_READ(SKL_MAD_DIMM_CH0_0_0_0_MCHBAR_MCMAIN);
+	ret = skl_dram_get_channel_info(dev_priv, &ch0, 0, val);
 	if (ret == 0)
 		dram_info->num_channels++;
 
-	val_ch1 = I915_READ(SKL_MAD_DIMM_CH1_0_0_0_MCHBAR_MCMAIN);
-	ret = skl_dram_get_channel_info(&ch1, val_ch1);
+	val = I915_READ(SKL_MAD_DIMM_CH1_0_0_0_MCHBAR_MCMAIN);
+	ret = skl_dram_get_channel_info(dev_priv, &ch1, 1, val);
 	if (ret == 0)
 		dram_info->num_channels++;
 
@@ -1151,28 +1197,47 @@ skl_dram_get_channels_info(struct drm_i915_private *dev_priv)
 	 * will be same as if single rank memory, so consider single rank
 	 * memory.
 	 */
-	if (ch0.rank == I915_DRAM_RANK_SINGLE ||
-	    ch1.rank == I915_DRAM_RANK_SINGLE)
-		dram_info->rank = I915_DRAM_RANK_SINGLE;
+	if (ch0.ranks == 1 || ch1.ranks == 1)
+		dram_info->ranks = 1;
 	else
-		dram_info->rank = max(ch0.rank, ch1.rank);
+		dram_info->ranks = max(ch0.ranks, ch1.ranks);
 
-	if (dram_info->rank == I915_DRAM_RANK_INVALID) {
+	if (dram_info->ranks == 0) {
 		DRM_INFO("couldn't get memory rank information\n");
 		return -EINVAL;
 	}
 
 	dram_info->is_16gb_dimm = ch0.is_16gb_dimm || ch1.is_16gb_dimm;
 
-	dev_priv->dram_info.symmetric_memory = intel_is_dram_symmetric(val_ch0,
-								       val_ch1,
-								       &ch0);
+	dram_info->symmetric_memory = intel_is_dram_symmetric(&ch0, &ch1);
 
-	DRM_DEBUG_KMS("memory configuration is %sSymmetric memory\n",
-		      dev_priv->dram_info.symmetric_memory ? "" : "not ");
+	DRM_DEBUG_KMS("Memory configuration is symmetric? %s\n",
+		      yesno(dram_info->symmetric_memory));
 	return 0;
 }
 
+static enum intel_dram_type
+skl_get_dram_type(struct drm_i915_private *dev_priv)
+{
+	u32 val;
+
+	val = I915_READ(SKL_MAD_INTER_CHANNEL_0_0_0_MCHBAR_MCMAIN);
+
+	switch (val & SKL_DRAM_DDR_TYPE_MASK) {
+	case SKL_DRAM_DDR_TYPE_DDR3:
+		return INTEL_DRAM_DDR3;
+	case SKL_DRAM_DDR_TYPE_DDR4:
+		return INTEL_DRAM_DDR4;
+	case SKL_DRAM_DDR_TYPE_LPDDR3:
+		return INTEL_DRAM_LPDDR3;
+	case SKL_DRAM_DDR_TYPE_LPDDR4:
+		return INTEL_DRAM_LPDDR4;
+	default:
+		MISSING_CASE(val);
+		return INTEL_DRAM_UNKNOWN;
+	}
+}
+
 static int
 skl_get_dram_info(struct drm_i915_private *dev_priv)
 {
@@ -1180,6 +1245,9 @@ skl_get_dram_info(struct drm_i915_private *dev_priv)
 	u32 mem_freq_khz, val;
 	int ret;
 
+	dram_info->type = skl_get_dram_type(dev_priv);
+	DRM_DEBUG_KMS("DRAM type: %s\n", intel_dram_type_str(dram_info->type));
+
 	ret = skl_dram_get_channels_info(dev_priv);
 	if (ret)
 		return ret;
@@ -1200,6 +1268,85 @@ skl_get_dram_info(struct drm_i915_private *dev_priv)
 	return 0;
 }
 
+/* Returns Gb per DRAM device */
+static int bxt_get_dimm_size(u32 val)
+{
+	switch (val & BXT_DRAM_SIZE_MASK) {
+	case BXT_DRAM_SIZE_4GBIT:
+		return 4;
+	case BXT_DRAM_SIZE_6GBIT:
+		return 6;
+	case BXT_DRAM_SIZE_8GBIT:
+		return 8;
+	case BXT_DRAM_SIZE_12GBIT:
+		return 12;
+	case BXT_DRAM_SIZE_16GBIT:
+		return 16;
+	default:
+		MISSING_CASE(val);
+		return 0;
+	}
+}
+
+static int bxt_get_dimm_width(u32 val)
+{
+	if (!bxt_get_dimm_size(val))
+		return 0;
+
+	val = (val & BXT_DRAM_WIDTH_MASK) >> BXT_DRAM_WIDTH_SHIFT;
+
+	return 8 << val;
+}
+
+static int bxt_get_dimm_ranks(u32 val)
+{
+	if (!bxt_get_dimm_size(val))
+		return 0;
+
+	switch (val & BXT_DRAM_RANK_MASK) {
+	case BXT_DRAM_RANK_SINGLE:
+		return 1;
+	case BXT_DRAM_RANK_DUAL:
+		return 2;
+	default:
+		MISSING_CASE(val);
+		return 0;
+	}
+}
+
+static enum intel_dram_type bxt_get_dimm_type(u32 val)
+{
+	if (!bxt_get_dimm_size(val))
+		return INTEL_DRAM_UNKNOWN;
+
+	switch (val & BXT_DRAM_TYPE_MASK) {
+	case BXT_DRAM_TYPE_DDR3:
+		return INTEL_DRAM_DDR3;
+	case BXT_DRAM_TYPE_LPDDR3:
+		return INTEL_DRAM_LPDDR3;
+	case BXT_DRAM_TYPE_DDR4:
+		return INTEL_DRAM_DDR4;
+	case BXT_DRAM_TYPE_LPDDR4:
+		return INTEL_DRAM_LPDDR4;
+	default:
+		MISSING_CASE(val);
+		return INTEL_DRAM_UNKNOWN;
+	}
+}
+
+static void bxt_get_dimm_info(struct dram_dimm_info *dimm,
+			      u32 val)
+{
+	dimm->width = bxt_get_dimm_width(val);
+	dimm->ranks = bxt_get_dimm_ranks(val);
+
+	/*
+	 * Size in register is Gb per DRAM device. Convert to total
+	 * GB to match the way we report this for non-LP platforms.
+	 */
+	dimm->size = bxt_get_dimm_size(val) * intel_dimm_num_devices(dimm) / 8;
+}
+
 static int
 bxt_get_dram_info(struct drm_i915_private *dev_priv)
 {
@@ -1228,57 +1375,44 @@ bxt_get_dram_info(struct drm_i915_private *dev_priv)
 	 * Now read each DUNIT8/9/10/11 to check the rank of each dimms.
 	 */
 	for (i = BXT_D_CR_DRP0_DUNIT_START; i <= BXT_D_CR_DRP0_DUNIT_END; i++) {
-		u8 size, width;
-		enum dram_rank rank;
-		u32 tmp;
+		struct dram_dimm_info dimm;
+		enum intel_dram_type type;
 
 		val = I915_READ(BXT_D_CR_DRP0_DUNIT(i));
 		if (val == 0xFFFFFFFF)
 			continue;
 
 		dram_info->num_channels++;
-		tmp = val & BXT_DRAM_RANK_MASK;
-
-		if (tmp == BXT_DRAM_RANK_SINGLE)
-			rank = I915_DRAM_RANK_SINGLE;
-		else if (tmp == BXT_DRAM_RANK_DUAL)
-			rank = I915_DRAM_RANK_DUAL;
-		else
-			rank = I915_DRAM_RANK_INVALID;
-
-		tmp = val & BXT_DRAM_SIZE_MASK;
-		if (tmp == BXT_DRAM_SIZE_4GB)
-			size = 4;
-		else if (tmp == BXT_DRAM_SIZE_6GB)
-			size = 6;
-		else if (tmp == BXT_DRAM_SIZE_8GB)
-			size = 8;
-		else if (tmp == BXT_DRAM_SIZE_12GB)
-			size = 12;
-		else if (tmp == BXT_DRAM_SIZE_16GB)
-			size = 16;
-		else
-			size = 0;
-
-		tmp = (val & BXT_DRAM_WIDTH_MASK) >> BXT_DRAM_WIDTH_SHIFT;
-		width = (1 << tmp) * 8;
-		DRM_DEBUG_KMS("dram size:%dGB width:X%d rank:%s\n", size,
-			      width, rank == I915_DRAM_RANK_SINGLE ? "single" :
-			      rank == I915_DRAM_RANK_DUAL ? "dual" : "unknown");
+
+		bxt_get_dimm_info(&dimm, val);
+		type = bxt_get_dimm_type(val);
+
+		WARN_ON(type != INTEL_DRAM_UNKNOWN &&
+			dram_info->type != INTEL_DRAM_UNKNOWN &&
+			dram_info->type != type);
+
+		DRM_DEBUG_KMS("CH%u DIMM size: %u GB, width: X%u, ranks: %u, type: %s\n",
+			      i - BXT_D_CR_DRP0_DUNIT_START,
+			      dimm.size, dimm.width, dimm.ranks,
+			      intel_dram_type_str(type));
 
 		/*
 		 * If any of the channel is single rank channel,
 		 * worst case output will be same as if single rank
 		 * memory, so consider single rank memory.
 		 */
-		if (dram_info->rank == I915_DRAM_RANK_INVALID)
-			dram_info->rank = rank;
-		else if (rank == I915_DRAM_RANK_SINGLE)
-			dram_info->rank = I915_DRAM_RANK_SINGLE;
+		if (dram_info->ranks == 0)
+			dram_info->ranks = dimm.ranks;
+		else if (dimm.ranks == 1)
+			dram_info->ranks = 1;
+
+		if (type != INTEL_DRAM_UNKNOWN)
+			dram_info->type = type;
 	}
 
-	if (dram_info->rank == I915_DRAM_RANK_INVALID) {
-		DRM_INFO("couldn't get memory rank information\n");
+	if (dram_info->type == INTEL_DRAM_UNKNOWN ||
+	    dram_info->ranks == 0) {
+		DRM_INFO("couldn't get memory information\n");
 		return -EINVAL;
 	}
 
@@ -1290,14 +1424,8 @@ static void
 intel_get_dram_info(struct drm_i915_private *dev_priv)
 {
 	struct dram_info *dram_info = &dev_priv->dram_info;
-	char bandwidth_str[32];
 	int ret;
 
-	dram_info->valid = false;
-	dram_info->rank = I915_DRAM_RANK_INVALID;
-	dram_info->bandwidth_kbps = 0;
-	dram_info->num_channels = 0;
-
 	/*
 	 * Assume 16Gb DIMMs are present until proven otherwise.
 	 * This is only used for the level 0 watermark latency
@@ -1305,28 +1433,61 @@ intel_get_dram_info(struct drm_i915_private *dev_priv)
 	 */
 	dram_info->is_16gb_dimm = !IS_GEN9_LP(dev_priv);
 
-	if (INTEL_GEN(dev_priv) < 9 || IS_GEMINILAKE(dev_priv))
+	if (INTEL_GEN(dev_priv) < 9)
 		return;
 
-	/* Need to calculate bandwidth only for Gen9 */
-	if (IS_BROXTON(dev_priv))
+	if (IS_GEN9_LP(dev_priv))
 		ret = bxt_get_dram_info(dev_priv);
-	else if (IS_GEN(dev_priv, 9))
-		ret = skl_get_dram_info(dev_priv);
 	else
-		ret = skl_dram_get_channels_info(dev_priv);
+		ret = skl_get_dram_info(dev_priv);
 	if (ret)
 		return;
 
-	if (dram_info->bandwidth_kbps)
-		sprintf(bandwidth_str, "%d KBps", dram_info->bandwidth_kbps);
+	DRM_DEBUG_KMS("DRAM bandwidth: %u kBps, channels: %u\n",
+		      dram_info->bandwidth_kbps,
+		      dram_info->num_channels);
+
+	DRM_DEBUG_KMS("DRAM ranks: %u, 16Gb DIMMs: %s\n",
+		      dram_info->ranks, yesno(dram_info->is_16gb_dimm));
+}
+
+static u32 gen9_edram_size_mb(struct drm_i915_private *dev_priv, u32 cap)
+{
+	const unsigned int ways[8] = { 4, 8, 12, 16, 16, 16, 16, 16 };
+	const unsigned int sets[4] = { 1, 1, 2, 2 };
+
+	return EDRAM_NUM_BANKS(cap) *
+		ways[EDRAM_WAYS_IDX(cap)] *
+		sets[EDRAM_SETS_IDX(cap)];
+}
+
+static void edram_detect(struct drm_i915_private *dev_priv)
+{
+	u32 edram_cap = 0;
+
+	if (!(IS_HASWELL(dev_priv) ||
+	      IS_BROADWELL(dev_priv) ||
+	      INTEL_GEN(dev_priv) >= 9))
+		return;
+
+	edram_cap = __raw_uncore_read32(&dev_priv->uncore, HSW_EDRAM_CAP);
+
+	/* NB: We can't write IDICR yet because we don't have gt funcs set up */
+
+	if (!(edram_cap & EDRAM_ENABLED))
+		return;
+
+	/*
+	 * The needed capability bits for size calculation are not there with
+	 * pre gen9 so return 128MB always.
+	 */
+	if (INTEL_GEN(dev_priv) < 9)
+		dev_priv->edram_size_mb = 128;
 	else
-		sprintf(bandwidth_str, "unknown");
-	DRM_DEBUG_KMS("DRAM bandwidth:%s, total-channels: %u\n",
-		      bandwidth_str, dram_info->num_channels);
-	DRM_DEBUG_KMS("DRAM rank: %s rank 16GB-dimm:%s\n",
-		      (dram_info->rank == I915_DRAM_RANK_DUAL) ?
-		      "dual" : "single", yesno(dram_info->is_16gb_dimm));
+		dev_priv->edram_size_mb =
+			gen9_edram_size_mb(dev_priv, edram_cap);
+
+	DRM_INFO("Found %uMB of eDRAM\n", dev_priv->edram_size_mb);
 }
 
 /**
@@ -1348,7 +1509,7 @@ static int i915_driver_init_hw(struct drm_i915_private *dev_priv)
 
 	if (HAS_PPGTT(dev_priv)) {
 		if (intel_vgpu_active(dev_priv) &&
-		    !intel_vgpu_has_full_48bit_ppgtt(dev_priv)) {
+		    !intel_vgpu_has_full_ppgtt(dev_priv)) {
 			i915_report_error(dev_priv,
 					  "incompatible vGPU found, support for isolated ppGTT required\n");
 			return -ENXIO;
@@ -1371,6 +1532,9 @@ static int i915_driver_init_hw(struct drm_i915_private *dev_priv)
 
 	intel_sanitize_options(dev_priv);
 
+	/* needs to be done before ggtt probe */
+	edram_detect(dev_priv);
+
 	i915_perf_init(dev_priv);
 
 	ret = i915_ggtt_probe_hw(dev_priv);
@@ -1606,10 +1770,12 @@ static void i915_welcome_messages(struct drm_i915_private *dev_priv)
 	if (drm_debug & DRM_UT_DRIVER) {
 		struct drm_printer p = drm_debug_printer("i915 device info:");
 
-		drm_printf(&p, "pciid=0x%04x rev=0x%02x platform=%s gen=%i\n",
+		drm_printf(&p, "pciid=0x%04x rev=0x%02x platform=%s (subplatform=0x%x) gen=%i\n",
 			   INTEL_DEVID(dev_priv),
 			   INTEL_REVID(dev_priv),
 			   intel_platform_name(INTEL_INFO(dev_priv)->platform),
+			   intel_subplatform(RUNTIME_INFO(dev_priv),
+					     INTEL_INFO(dev_priv)->platform),
 			   INTEL_GEN(dev_priv));
 
 		intel_device_info_dump_flags(INTEL_INFO(dev_priv), &p);
@@ -1652,8 +1818,6 @@ i915_driver_create(struct pci_dev *pdev, const struct pci_device_id *ent)
 	memcpy(device_info, match_info, sizeof(*device_info));
 	RUNTIME_INFO(i915)->device_id = pdev->device;
 
-	BUILD_BUG_ON(INTEL_MAX_PLATFORMS >
-		     BITS_PER_TYPE(device_info->platform_mask));
 	BUG_ON(device_info->gen > BITS_PER_TYPE(device_info->gen_mask));
 
 	return i915;
@@ -1750,11 +1914,17 @@ void i915_driver_unload(struct drm_device *dev)
 
 	i915_driver_unregister(dev_priv);
 
+	/*
+	 * After unregistering the device to prevent any new users, cancel
+	 * all in-flight requests so that we can quickly unbind the active
+	 * resources.
+	 */
+	i915_gem_set_wedged(dev_priv);
+
 	/* Flush any external code that still may be under the RCU lock */
 	synchronize_rcu();
 
-	if (i915_gem_suspend(dev_priv))
-		DRM_ERROR("failed to idle hardware; continuing to unload!\n");
+	i915_gem_suspend(dev_priv);
 
 	drm_atomic_helper_shutdown(dev);
 
@@ -1862,7 +2032,6 @@ static bool suspend_to_idle(struct drm_i915_private *dev_priv)
 static int i915_drm_prepare(struct drm_device *dev)
 {
 	struct drm_i915_private *i915 = to_i915(dev);
-	int err;
 
 	/*
 	 * NB intel_display_suspend() may issue new requests after we've
@@ -1870,12 +2039,9 @@ static int i915_drm_prepare(struct drm_device *dev)
 	 * split out that work and pull it forward so that after point,
 	 * the GPU is not woken again.
 	 */
-	err = i915_gem_suspend(i915);
-	if (err)
-		dev_err(&i915->drm.pdev->dev,
-			"GEM idle failed, suspend/resume might fail\n");
+	i915_gem_suspend(i915);
 
-	return err;
+	return 0;
 }
 
 static int i915_drm_suspend(struct drm_device *dev)
@@ -1945,7 +2111,7 @@ static int i915_drm_suspend_late(struct drm_device *dev, bool hibernation)
 
 	i915_gem_suspend_late(dev_priv);
 
-	intel_uncore_suspend(dev_priv);
+	intel_uncore_suspend(&dev_priv->uncore);
 
 	intel_power_domains_suspend(dev_priv,
 				    get_suspend_mode(dev_priv, hibernation));
@@ -2141,7 +2307,9 @@ static int i915_drm_resume_early(struct drm_device *dev)
 		DRM_ERROR("Resume prepare failed: %d, continuing anyway\n",
 			  ret);
 
-	intel_uncore_resume_early(dev_priv);
+	intel_uncore_resume_early(&dev_priv->uncore);
+
+	i915_check_and_clear_faults(dev_priv);
 
 	if (INTEL_GEN(dev_priv) >= 11 || IS_GEN9_LP(dev_priv)) {
 		gen9_sanitize_dc_state(dev_priv);
@@ -2545,7 +2713,7 @@ int vlv_force_gfx_clock(struct drm_i915_private *dev_priv, bool force_on)
 	if (!force_on)
 		return 0;
 
-	err = intel_wait_for_register(dev_priv,
+	err = intel_wait_for_register(&dev_priv->uncore,
 				      VLV_GTLC_SURVIVABILITY_REG,
 				      VLV_GFX_CLK_STATUS_BIT,
 				      VLV_GFX_CLK_STATUS_BIT,
@@ -2711,7 +2879,7 @@ static int intel_runtime_suspend(struct device *kdev)
 
 	intel_runtime_pm_disable_interrupts(dev_priv);
 
-	intel_uncore_suspend(dev_priv);
+	intel_uncore_suspend(&dev_priv->uncore);
 
 	ret = 0;
 	if (INTEL_GEN(dev_priv) >= 11) {
@@ -2728,7 +2896,7 @@ static int intel_runtime_suspend(struct device *kdev)
 
 	if (ret) {
 		DRM_ERROR("Runtime suspend failed, disabling it (%d)\n", ret);
-		intel_uncore_runtime_resume(dev_priv);
+		intel_uncore_runtime_resume(&dev_priv->uncore);
 
 		intel_runtime_pm_enable_interrupts(dev_priv);
 
@@ -2745,7 +2913,7 @@ static int intel_runtime_suspend(struct device *kdev)
 	enable_rpm_wakeref_asserts(dev_priv);
 	intel_runtime_pm_cleanup(dev_priv);
 
-	if (intel_uncore_arm_unclaimed_mmio_detection(dev_priv))
+	if (intel_uncore_arm_unclaimed_mmio_detection(&dev_priv->uncore))
 		DRM_ERROR("Unclaimed access detected prior to suspending\n");
 
 	dev_priv->runtime_pm.suspended = true;
@@ -2773,7 +2941,7 @@ static int intel_runtime_suspend(struct device *kdev)
 		intel_opregion_notify_adapter(dev_priv, PCI_D1);
 	}
 
-	assert_forcewakes_inactive(dev_priv);
+	assert_forcewakes_inactive(&dev_priv->uncore);
 
 	if (!IS_VALLEYVIEW(dev_priv) && !IS_CHERRYVIEW(dev_priv))
 		intel_hpd_poll_init(dev_priv);
@@ -2799,7 +2967,7 @@ static int intel_runtime_resume(struct device *kdev)
 
 	intel_opregion_notify_adapter(dev_priv, PCI_D0);
 	dev_priv->runtime_pm.suspended = false;
-	if (intel_uncore_unclaimed_mmio(dev_priv))
+	if (intel_uncore_unclaimed_mmio(&dev_priv->uncore))
 		DRM_DEBUG_DRIVER("Unclaimed access during suspend, bios?\n");
 
 	if (INTEL_GEN(dev_priv) >= 11) {
@@ -2825,7 +2993,7 @@ static int intel_runtime_resume(struct device *kdev)
 		ret = vlv_resume_prepare(dev_priv, true);
 	}
 
-	intel_uncore_runtime_resume(dev_priv);
+	intel_uncore_runtime_resume(&dev_priv->uncore);
 
 	intel_runtime_pm_enable_interrupts(dev_priv);
 
@@ -2969,7 +3137,7 @@ static const struct drm_ioctl_desc i915_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(I915_SET_SPRITE_COLORKEY, intel_sprite_set_colorkey_ioctl, DRM_MASTER),
 	DRM_IOCTL_DEF_DRV(I915_GET_SPRITE_COLORKEY, drm_noop, DRM_MASTER),
 	DRM_IOCTL_DEF_DRV(I915_GEM_WAIT, i915_gem_wait_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GEM_CONTEXT_CREATE, i915_gem_context_create_ioctl, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_CONTEXT_CREATE_EXT, i915_gem_context_create_ioctl, DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(I915_GEM_CONTEXT_DESTROY, i915_gem_context_destroy_ioctl, DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(I915_REG_READ, i915_reg_read_ioctl, DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(I915_GET_RESET_STATS, i915_gem_context_reset_stats_ioctl, DRM_RENDER_ALLOW),
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index a67a63b5aa84..066fd2a12851 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -55,6 +55,7 @@
 #include <drm/drm_util.h>
 #include <drm/drm_dsc.h>
 #include <drm/drm_connector.h>
+#include <drm/i915_mei_hdcp_interface.h>
 
 #include "i915_fixed.h"
 #include "i915_params.h"
@@ -65,13 +66,14 @@
 #include "intel_device_info.h"
 #include "intel_display.h"
 #include "intel_dpll_mgr.h"
+#include "intel_frontbuffer.h"
 #include "intel_lrc.h"
 #include "intel_opregion.h"
 #include "intel_ringbuffer.h"
+#include "intel_uc.h"
 #include "intel_uncore.h"
 #include "intel_wopcm.h"
 #include "intel_workarounds.h"
-#include "intel_uc.h"
 
 #include "i915_gem.h"
 #include "i915_gem_context.h"
@@ -91,8 +93,8 @@
 
 #define DRIVER_NAME		"i915"
 #define DRIVER_DESC		"Intel Graphics"
-#define DRIVER_DATE		"20190207"
-#define DRIVER_TIMESTAMP	1549572331
+#define DRIVER_DATE		"20190417"
+#define DRIVER_TIMESTAMP	1555492067
 
 /* Use I915_STATE_WARN(x) and I915_STATE_WARN_ON() (rather than WARN() and
  * WARN_ON()) for hw state sanity checks to check for unexpected conditions
@@ -215,11 +217,12 @@ struct drm_i915_file_private {
  */
 #define DRM_I915_THROTTLE_JIFFIES msecs_to_jiffies(20)
 	} mm;
+
 	struct idr context_idr;
+	struct mutex context_idr_lock; /* guards context_idr */
 
-	struct intel_rps_client {
-		atomic_t boosts;
-	} rps_client;
+	struct idr vm_idr;
+	struct mutex vm_idr_lock; /* guards vm_idr */
 
 	unsigned int bsd_engine;
 
@@ -280,7 +283,8 @@ struct drm_i915_display_funcs {
 	void (*get_cdclk)(struct drm_i915_private *dev_priv,
 			  struct intel_cdclk_state *cdclk_state);
 	void (*set_cdclk)(struct drm_i915_private *dev_priv,
-			  const struct intel_cdclk_state *cdclk_state);
+			  const struct intel_cdclk_state *cdclk_state,
+			  enum pipe pipe);
 	int (*get_fifo_size)(struct drm_i915_private *dev_priv,
 			     enum i9xx_plane_id i9xx_plane);
 	int (*compute_pipe_wm)(struct intel_crtc_state *cstate);
@@ -323,6 +327,7 @@ struct drm_i915_display_funcs {
 	/* display clock increase/decrease */
 	/* pll clock increase/decrease */
 
+	int (*color_check)(struct intel_crtc_state *crtc_state);
 	/*
 	 * Program double buffered color management registers during
 	 * vblank evasion. The registers should then latch during the
@@ -371,14 +376,6 @@ enum i915_cache_level {
 
 #define I915_COLOR_UNEVICTABLE (-1) /* a non-vma sharing the address space */
 
-enum fb_op_origin {
-	ORIGIN_GTT,
-	ORIGIN_CPU,
-	ORIGIN_CS,
-	ORIGIN_FLIP,
-	ORIGIN_DIRTYFB,
-};
-
 struct intel_fbc {
 	/* This is always the inner lock when overlapping with struct_mutex and
 	 * it's the outer lock when overlapping with stolen_lock. */
@@ -508,7 +505,7 @@ struct i915_psr {
 
 	u32 debug;
 	bool sink_support;
-	bool prepared, enabled;
+	bool enabled;
 	struct intel_dp *dp;
 	enum pipe pipe;
 	bool active;
@@ -526,16 +523,22 @@ struct i915_psr {
 	u16 su_x_granularity;
 };
 
+/*
+ * Sorted by south display engine compatibility.
+ * If the new PCH comes with a south display engine that is not
+ * inherited from the latest item, please do not add it to the
+ * end. Instead, add it right after its "parent" PCH.
+ */
 enum intel_pch {
+	PCH_NOP = -1,	/* PCH without south display */
 	PCH_NONE = 0,	/* No PCH present */
 	PCH_IBX,	/* Ibexpeak PCH */
 	PCH_CPT,	/* Cougarpoint/Pantherpoint PCH */
 	PCH_LPT,	/* Lynxpoint/Wildcatpoint PCH */
 	PCH_SPT,        /* Sunrisepoint PCH */
 	PCH_KBP,        /* Kaby Lake PCH */
-	PCH_CNP,        /* Cannon Lake PCH */
+	PCH_CNP,        /* Cannon/Comet Lake PCH */
 	PCH_ICP,	/* Ice Lake PCH */
-	PCH_NOP,	/* PCH without south display */
 };
 
 enum intel_sbi_destination {
@@ -949,6 +952,7 @@ struct ddi_vbt_port_info {
 #define HDMI_LEVEL_SHIFT_UNKNOWN	0xff
 	u8 hdmi_level_shift;
 
+	u8 present:1;
 	u8 supports_dvi:1;
 	u8 supports_hdmi:1;
 	u8 supports_dp:1;
@@ -1009,6 +1013,7 @@ struct intel_vbt_data {
 		enum psr_lines_to_wait lines_to_wait;
 		int tp1_wakeup_time_us;
 		int tp2_tp3_wakeup_time_us;
+		int psr2_tp2_tp3_wakeup_time_us;
 	} psr;
 
 	struct {
@@ -1130,6 +1135,7 @@ struct skl_wm_level {
 	u16 plane_res_b;
 	u8 plane_res_l;
 	bool plane_en;
+	bool ignore_lines;
 };
 
 /* Stores plane specific WM parameters */
@@ -1200,7 +1206,11 @@ enum intel_pipe_crc_source {
 	INTEL_PIPE_CRC_SOURCE_NONE,
 	INTEL_PIPE_CRC_SOURCE_PLANE1,
 	INTEL_PIPE_CRC_SOURCE_PLANE2,
-	INTEL_PIPE_CRC_SOURCE_PF,
+	INTEL_PIPE_CRC_SOURCE_PLANE3,
+	INTEL_PIPE_CRC_SOURCE_PLANE4,
+	INTEL_PIPE_CRC_SOURCE_PLANE5,
+	INTEL_PIPE_CRC_SOURCE_PLANE6,
+	INTEL_PIPE_CRC_SOURCE_PLANE7,
 	INTEL_PIPE_CRC_SOURCE_PIPE,
 	/* TV/DP on pre-gen5/vlv can't use the pipe source. */
 	INTEL_PIPE_CRC_SOURCE_TV,
@@ -1468,13 +1478,6 @@ struct intel_cdclk_state {
 struct drm_i915_private {
 	struct drm_device drm;
 
-	struct kmem_cache *objects;
-	struct kmem_cache *vmas;
-	struct kmem_cache *luts;
-	struct kmem_cache *requests;
-	struct kmem_cache *dependencies;
-	struct kmem_cache *priorities;
-
 	const struct intel_device_info __info; /* Use INTEL_INFO() to access. */
 	struct intel_runtime_info __runtime; /* Use RUNTIME_INFO() to access. */
 	struct intel_driver_caps caps;
@@ -1503,8 +1506,6 @@ struct drm_i915_private {
 	 */
 	resource_size_t stolen_usable_size;	/* Total size minus reserved ranges */
 
-	void __iomem *regs;
-
 	struct intel_uncore uncore;
 
 	struct i915_virtual_gpu vgpu;
@@ -1622,6 +1623,8 @@ struct drm_i915_private {
 		struct intel_cdclk_state actual;
 		/* The current hardware cdclk state */
 		struct intel_cdclk_state hw;
+
+		int force_min_cdclk;
 	} cdclk;
 
 	/**
@@ -1700,8 +1703,11 @@ struct drm_i915_private {
 
 	struct intel_l3_parity l3_parity;
 
-	/* Cannot be determined by PCIID. You must always read a register. */
-	u32 edram_cap;
+	/*
+	 * edram size in MB.
+	 * Cannot be determined by PCIID. You must always read a register.
+	 */
+	u32 edram_size_mb;
 
 	/*
 	 * Protects RPS/RC6 register access and PCU communication.
@@ -1741,6 +1747,7 @@ struct drm_i915_private {
 	 *
 	 */
 	struct mutex av_mutex;
+	int audio_power_refcount;
 
 	struct {
 		struct mutex mutex;
@@ -1831,13 +1838,16 @@ struct drm_i915_private {
 		bool valid;
 		bool is_16gb_dimm;
 		u8 num_channels;
-		enum dram_rank {
-			I915_DRAM_RANK_INVALID = 0,
-			I915_DRAM_RANK_SINGLE,
-			I915_DRAM_RANK_DUAL
-		} rank;
+		u8 ranks;
 		u32 bandwidth_kbps;
 		bool symmetric_memory;
+		enum intel_dram_type {
+			INTEL_DRAM_UNKNOWN,
+			INTEL_DRAM_DDR3,
+			INTEL_DRAM_DDR4,
+			INTEL_DRAM_LPDDR3,
+			INTEL_DRAM_LPDDR4
+		} type;
 	} dram_info;
 
 	struct i915_runtime_pm runtime_pm;
@@ -1985,7 +1995,6 @@ struct drm_i915_private {
 
 	/* Abstract the submission mechanism (legacy ringbuffer or execlists) away */
 	struct {
-		void (*resume)(struct drm_i915_private *);
 		void (*cleanup_engine)(struct intel_engine_cs *engine);
 
 		struct i915_gt_timelines {
@@ -1997,6 +2006,7 @@ struct drm_i915_private {
 			struct list_head hwsp_free_list;
 		} timelines;
 
+		intel_engine_mask_t active_engines;
 		struct list_head active_rings;
 		struct list_head closed_vma;
 		u32 active_requests;
@@ -2011,12 +2021,6 @@ struct drm_i915_private {
 		intel_wakeref_t awake;
 
 		/**
-		 * The number of times we have woken up.
-		 */
-		unsigned int epoch;
-#define I915_EPOCH_INVALID 0
-
-		/**
 		 * We leave the user IRQ off as much as possible,
 		 * but this means that requests will finish and never
 		 * be retired once the system goes idle. Set a timer to
@@ -2039,6 +2043,14 @@ struct drm_i915_private {
 		struct i915_vma *scratch;
 	} gt;
 
+	/* For i945gm vblank irq vs. C3 workaround */
+	struct {
+		struct work_struct work;
+		struct pm_qos_request pm_qos;
+		u8 c3_disable_latency;
+		u8 enabled;
+	} i945gm_vblank;
+
 	/* perform PHY state sanity checks? */
 	bool chv_phy_assert[2];
 
@@ -2055,18 +2067,25 @@ struct drm_i915_private {
 
 	struct i915_pmu pmu;
 
+	struct i915_hdcp_comp_master *hdcp_master;
+	bool hdcp_comp_added;
+
+	/* Mutex to protect the above hdcp component related values. */
+	struct mutex hdcp_comp_mutex;
+
 	/*
 	 * NOTE: This is the dri1/ums dungeon, don't add stuff here. Your patch
 	 * will be rejected. Instead look for a better place.
 	 */
 };
 
+struct dram_dimm_info {
+	u8 size, width, ranks;
+};
+
 struct dram_channel_info {
-	struct info {
-		u8 size, width;
-		enum dram_rank rank;
-	} l_info, s_info;
-	enum dram_rank rank;
+	struct dram_dimm_info dimm_l, dimm_s;
+	u8 ranks;
 	bool is_16gb_dimm;
 };
 
@@ -2095,6 +2114,11 @@ static inline struct drm_i915_private *huc_to_i915(struct intel_huc *huc)
 	return container_of(huc, struct drm_i915_private, huc);
 }
 
+static inline struct drm_i915_private *uncore_to_i915(struct intel_uncore *uncore)
+{
+	return container_of(uncore, struct drm_i915_private, uncore);
+}
+
 /* Simple iterator over all initialised engines */
 #define for_each_engine(engine__, dev_priv__, id__) \
 	for ((id__) = 0; \
@@ -2104,7 +2128,7 @@ static inline struct drm_i915_private *huc_to_i915(struct intel_huc *huc)
 
 /* Iterator over subset of engines selected by mask */
 #define for_each_engine_masked(engine__, dev_priv__, mask__, tmp__) \
-	for ((tmp__) = (mask__) & INTEL_INFO(dev_priv__)->ring_mask; \
+	for ((tmp__) = (mask__) & INTEL_INFO(dev_priv__)->engine_mask; \
 	     (tmp__) ? \
 	     ((engine__) = (dev_priv__)->engine[__mask_next_bit(tmp__)]), 1 : \
 	     0;)
@@ -2274,7 +2298,69 @@ static inline unsigned int i915_sg_segment_size(void)
 #define IS_REVID(p, since, until) \
 	(INTEL_REVID(p) >= (since) && INTEL_REVID(p) <= (until))
 
-#define IS_PLATFORM(dev_priv, p) (INTEL_INFO(dev_priv)->platform_mask & BIT(p))
+static __always_inline unsigned int
+__platform_mask_index(const struct intel_runtime_info *info,
+		      enum intel_platform p)
+{
+	const unsigned int pbits =
+		BITS_PER_TYPE(info->platform_mask[0]) - INTEL_SUBPLATFORM_BITS;
+
+	/* Expand the platform_mask array if this fails. */
+	BUILD_BUG_ON(INTEL_MAX_PLATFORMS >
+		     pbits * ARRAY_SIZE(info->platform_mask));
+
+	return p / pbits;
+}
+
+static __always_inline unsigned int
+__platform_mask_bit(const struct intel_runtime_info *info,
+		    enum intel_platform p)
+{
+	const unsigned int pbits =
+		BITS_PER_TYPE(info->platform_mask[0]) - INTEL_SUBPLATFORM_BITS;
+
+	return p % pbits + INTEL_SUBPLATFORM_BITS;
+}
+
+static inline u32
+intel_subplatform(const struct intel_runtime_info *info, enum intel_platform p)
+{
+	const unsigned int pi = __platform_mask_index(info, p);
+
+	return info->platform_mask[pi] & INTEL_SUBPLATFORM_BITS;
+}
+
+static __always_inline bool
+IS_PLATFORM(const struct drm_i915_private *i915, enum intel_platform p)
+{
+	const struct intel_runtime_info *info = RUNTIME_INFO(i915);
+	const unsigned int pi = __platform_mask_index(info, p);
+	const unsigned int pb = __platform_mask_bit(info, p);
+
+	BUILD_BUG_ON(!__builtin_constant_p(p));
+
+	return info->platform_mask[pi] & BIT(pb);
+}
+
+static __always_inline bool
+IS_SUBPLATFORM(const struct drm_i915_private *i915,
+	       enum intel_platform p, unsigned int s)
+{
+	const struct intel_runtime_info *info = RUNTIME_INFO(i915);
+	const unsigned int pi = __platform_mask_index(info, p);
+	const unsigned int pb = __platform_mask_bit(info, p);
+	const unsigned int msb = BITS_PER_TYPE(info->platform_mask[0]) - 1;
+	const u32 mask = info->platform_mask[pi];
+
+	BUILD_BUG_ON(!__builtin_constant_p(p));
+	BUILD_BUG_ON(!__builtin_constant_p(s));
+	BUILD_BUG_ON((s) >= INTEL_SUBPLATFORM_BITS);
+
+	/* Shift and test on the MSB position so sign flag can be used. */
+	return ((mask << (msb - pb)) & (mask << (msb - s))) & BIT(msb);
+}
+
+#define IS_MOBILE(dev_priv)	(INTEL_INFO(dev_priv)->is_mobile)
 
 #define IS_I830(dev_priv)	IS_PLATFORM(dev_priv, INTEL_I830)
 #define IS_I845G(dev_priv)	IS_PLATFORM(dev_priv, INTEL_I845G)
@@ -2289,11 +2375,11 @@ static inline unsigned int i915_sg_segment_size(void)
 #define IS_G45(dev_priv)	IS_PLATFORM(dev_priv, INTEL_G45)
 #define IS_GM45(dev_priv)	IS_PLATFORM(dev_priv, INTEL_GM45)
 #define IS_G4X(dev_priv)	(IS_G45(dev_priv) || IS_GM45(dev_priv))
-#define IS_PINEVIEW_G(dev_priv)	(INTEL_DEVID(dev_priv) == 0xa001)
-#define IS_PINEVIEW_M(dev_priv)	(INTEL_DEVID(dev_priv) == 0xa011)
 #define IS_PINEVIEW(dev_priv)	IS_PLATFORM(dev_priv, INTEL_PINEVIEW)
 #define IS_G33(dev_priv)	IS_PLATFORM(dev_priv, INTEL_G33)
-#define IS_IRONLAKE_M(dev_priv)	(INTEL_DEVID(dev_priv) == 0x0046)
+#define IS_IRONLAKE(dev_priv)	IS_PLATFORM(dev_priv, INTEL_IRONLAKE)
+#define IS_IRONLAKE_M(dev_priv) \
+	(IS_PLATFORM(dev_priv, INTEL_IRONLAKE) && IS_MOBILE(dev_priv))
 #define IS_IVYBRIDGE(dev_priv)	IS_PLATFORM(dev_priv, INTEL_IVYBRIDGE)
 #define IS_IVB_GT1(dev_priv)	(IS_IVYBRIDGE(dev_priv) && \
 				 INTEL_INFO(dev_priv)->gt == 1)
@@ -2308,46 +2394,35 @@ static inline unsigned int i915_sg_segment_size(void)
 #define IS_COFFEELAKE(dev_priv)	IS_PLATFORM(dev_priv, INTEL_COFFEELAKE)
 #define IS_CANNONLAKE(dev_priv)	IS_PLATFORM(dev_priv, INTEL_CANNONLAKE)
 #define IS_ICELAKE(dev_priv)	IS_PLATFORM(dev_priv, INTEL_ICELAKE)
-#define IS_MOBILE(dev_priv)	(INTEL_INFO(dev_priv)->is_mobile)
+#define IS_ELKHARTLAKE(dev_priv)	IS_PLATFORM(dev_priv, INTEL_ELKHARTLAKE)
 #define IS_HSW_EARLY_SDV(dev_priv) (IS_HASWELL(dev_priv) && \
 				    (INTEL_DEVID(dev_priv) & 0xFF00) == 0x0C00)
-#define IS_BDW_ULT(dev_priv)	(IS_BROADWELL(dev_priv) && \
-				 ((INTEL_DEVID(dev_priv) & 0xf) == 0x6 ||	\
-				 (INTEL_DEVID(dev_priv) & 0xf) == 0xb ||	\
-				 (INTEL_DEVID(dev_priv) & 0xf) == 0xe))
-/* ULX machines are also considered ULT. */
-#define IS_BDW_ULX(dev_priv)	(IS_BROADWELL(dev_priv) && \
-				 (INTEL_DEVID(dev_priv) & 0xf) == 0xe)
+#define IS_BDW_ULT(dev_priv) \
+	IS_SUBPLATFORM(dev_priv, INTEL_BROADWELL, INTEL_SUBPLATFORM_ULT)
+#define IS_BDW_ULX(dev_priv) \
+	IS_SUBPLATFORM(dev_priv, INTEL_BROADWELL, INTEL_SUBPLATFORM_ULX)
 #define IS_BDW_GT3(dev_priv)	(IS_BROADWELL(dev_priv) && \
 				 INTEL_INFO(dev_priv)->gt == 3)
-#define IS_HSW_ULT(dev_priv)	(IS_HASWELL(dev_priv) && \
-				 (INTEL_DEVID(dev_priv) & 0xFF00) == 0x0A00)
+#define IS_HSW_ULT(dev_priv) \
+	IS_SUBPLATFORM(dev_priv, INTEL_HASWELL, INTEL_SUBPLATFORM_ULT)
 #define IS_HSW_GT3(dev_priv)	(IS_HASWELL(dev_priv) && \
 				 INTEL_INFO(dev_priv)->gt == 3)
 #define IS_HSW_GT1(dev_priv)	(IS_HASWELL(dev_priv) && \
 				 INTEL_INFO(dev_priv)->gt == 1)
 /* ULX machines are also considered ULT. */
-#define IS_HSW_ULX(dev_priv)	(INTEL_DEVID(dev_priv) == 0x0A0E || \
-				 INTEL_DEVID(dev_priv) == 0x0A1E)
-#define IS_SKL_ULT(dev_priv)	(INTEL_DEVID(dev_priv) == 0x1906 || \
-				 INTEL_DEVID(dev_priv) == 0x1913 || \
-				 INTEL_DEVID(dev_priv) == 0x1916 || \
-				 INTEL_DEVID(dev_priv) == 0x1921 || \
-				 INTEL_DEVID(dev_priv) == 0x1926)
-#define IS_SKL_ULX(dev_priv)	(INTEL_DEVID(dev_priv) == 0x190E || \
-				 INTEL_DEVID(dev_priv) == 0x1915 || \
-				 INTEL_DEVID(dev_priv) == 0x191E)
-#define IS_KBL_ULT(dev_priv)	(INTEL_DEVID(dev_priv) == 0x5906 || \
-				 INTEL_DEVID(dev_priv) == 0x5913 || \
-				 INTEL_DEVID(dev_priv) == 0x5916 || \
-				 INTEL_DEVID(dev_priv) == 0x5921 || \
-				 INTEL_DEVID(dev_priv) == 0x5926)
-#define IS_KBL_ULX(dev_priv)	(INTEL_DEVID(dev_priv) == 0x590E || \
-				 INTEL_DEVID(dev_priv) == 0x5915 || \
-				 INTEL_DEVID(dev_priv) == 0x591E)
-#define IS_AML_ULX(dev_priv)	(INTEL_DEVID(dev_priv) == 0x591C || \
-				 INTEL_DEVID(dev_priv) == 0x87C0 || \
-				 INTEL_DEVID(dev_priv) == 0x87CA)
+#define IS_HSW_ULX(dev_priv) \
+	IS_SUBPLATFORM(dev_priv, INTEL_HASWELL, INTEL_SUBPLATFORM_ULX)
+#define IS_SKL_ULT(dev_priv) \
+	IS_SUBPLATFORM(dev_priv, INTEL_SKYLAKE, INTEL_SUBPLATFORM_ULT)
+#define IS_SKL_ULX(dev_priv) \
+	IS_SUBPLATFORM(dev_priv, INTEL_SKYLAKE, INTEL_SUBPLATFORM_ULX)
+#define IS_KBL_ULT(dev_priv) \
+	IS_SUBPLATFORM(dev_priv, INTEL_KABYLAKE, INTEL_SUBPLATFORM_ULT)
+#define IS_KBL_ULX(dev_priv) \
+	IS_SUBPLATFORM(dev_priv, INTEL_KABYLAKE, INTEL_SUBPLATFORM_ULX)
+#define IS_AML_ULX(dev_priv) \
+	(IS_SUBPLATFORM(dev_priv, INTEL_KABYLAKE, INTEL_SUBPLATFORM_AML) || \
+	 IS_SUBPLATFORM(dev_priv, INTEL_COFFEELAKE, INTEL_SUBPLATFORM_AML))
 #define IS_SKL_GT2(dev_priv)	(IS_SKYLAKE(dev_priv) && \
 				 INTEL_INFO(dev_priv)->gt == 2)
 #define IS_SKL_GT3(dev_priv)	(IS_SKYLAKE(dev_priv) && \
@@ -2358,16 +2433,16 @@ static inline unsigned int i915_sg_segment_size(void)
 				 INTEL_INFO(dev_priv)->gt == 2)
 #define IS_KBL_GT3(dev_priv)	(IS_KABYLAKE(dev_priv) && \
 				 INTEL_INFO(dev_priv)->gt == 3)
-#define IS_CFL_ULT(dev_priv)	(IS_COFFEELAKE(dev_priv) && \
-				 (INTEL_DEVID(dev_priv) & 0x00F0) == 0x00A0)
+#define IS_CFL_ULT(dev_priv) \
+	IS_SUBPLATFORM(dev_priv, INTEL_COFFEELAKE, INTEL_SUBPLATFORM_ULT)
 #define IS_CFL_GT2(dev_priv)	(IS_COFFEELAKE(dev_priv) && \
 				 INTEL_INFO(dev_priv)->gt == 2)
 #define IS_CFL_GT3(dev_priv)	(IS_COFFEELAKE(dev_priv) && \
 				 INTEL_INFO(dev_priv)->gt == 3)
-#define IS_CNL_WITH_PORT_F(dev_priv)   (IS_CANNONLAKE(dev_priv) && \
-					(INTEL_DEVID(dev_priv) & 0x0004) == 0x0004)
-#define IS_ICL_WITH_PORT_F(dev_priv)   (IS_ICELAKE(dev_priv) && \
-					INTEL_DEVID(dev_priv) != 0x8A51)
+#define IS_CNL_WITH_PORT_F(dev_priv) \
+	IS_SUBPLATFORM(dev_priv, INTEL_CANNONLAKE, INTEL_SUBPLATFORM_PORTF)
+#define IS_ICL_WITH_PORT_F(dev_priv) \
+	IS_SUBPLATFORM(dev_priv, INTEL_ICELAKE, INTEL_SUBPLATFORM_PORTF)
 
 #define IS_ALPHA_SUPPORT(intel_info) ((intel_info)->is_alpha_support)
 
@@ -2426,28 +2501,22 @@ static inline unsigned int i915_sg_segment_size(void)
 #define IS_GEN9_LP(dev_priv)	(IS_GEN(dev_priv, 9) && IS_LP(dev_priv))
 #define IS_GEN9_BC(dev_priv)	(IS_GEN(dev_priv, 9) && !IS_LP(dev_priv))
 
-#define ENGINE_MASK(id)	BIT(id)
-#define RENDER_RING	ENGINE_MASK(RCS)
-#define BSD_RING	ENGINE_MASK(VCS)
-#define BLT_RING	ENGINE_MASK(BCS)
-#define VEBOX_RING	ENGINE_MASK(VECS)
-#define BSD2_RING	ENGINE_MASK(VCS2)
-#define BSD3_RING	ENGINE_MASK(VCS3)
-#define BSD4_RING	ENGINE_MASK(VCS4)
-#define VEBOX2_RING	ENGINE_MASK(VECS2)
-#define ALL_ENGINES	(~0)
-
-#define HAS_ENGINE(dev_priv, id) \
-	(!!(INTEL_INFO(dev_priv)->ring_mask & ENGINE_MASK(id)))
-
-#define HAS_BSD(dev_priv)	HAS_ENGINE(dev_priv, VCS)
-#define HAS_BSD2(dev_priv)	HAS_ENGINE(dev_priv, VCS2)
-#define HAS_BLT(dev_priv)	HAS_ENGINE(dev_priv, BCS)
-#define HAS_VEBOX(dev_priv)	HAS_ENGINE(dev_priv, VECS)
+#define HAS_ENGINE(dev_priv, id) (INTEL_INFO(dev_priv)->engine_mask & BIT(id))
+
+#define ENGINE_INSTANCES_MASK(dev_priv, first, count) ({		\
+	unsigned int first__ = (first);					\
+	unsigned int count__ = (count);					\
+	(INTEL_INFO(dev_priv)->engine_mask &				\
+	 GENMASK(first__ + count__ - 1, first__)) >> first__;		\
+})
+#define VDBOX_MASK(dev_priv) \
+	ENGINE_INSTANCES_MASK(dev_priv, VCS0, I915_MAX_VCS)
+#define VEBOX_MASK(dev_priv) \
+	ENGINE_INSTANCES_MASK(dev_priv, VECS0, I915_MAX_VECS)
 
 #define HAS_LLC(dev_priv)	(INTEL_INFO(dev_priv)->has_llc)
 #define HAS_SNOOP(dev_priv)	(INTEL_INFO(dev_priv)->has_snoop)
-#define HAS_EDRAM(dev_priv)	(!!((dev_priv)->edram_cap & EDRAM_ENABLED))
+#define HAS_EDRAM(dev_priv)	((dev_priv)->edram_size_mb)
 #define HAS_WT(dev_priv)	((IS_HASWELL(dev_priv) || \
 				 IS_BROADWELL(dev_priv)) && HAS_EDRAM(dev_priv))
 
@@ -2462,13 +2531,11 @@ static inline unsigned int i915_sg_segment_size(void)
 
 #define HAS_EXECLISTS(dev_priv) HAS_LOGICAL_RING_CONTEXTS(dev_priv)
 
-#define INTEL_PPGTT(dev_priv) (INTEL_INFO(dev_priv)->ppgtt)
+#define INTEL_PPGTT(dev_priv) (INTEL_INFO(dev_priv)->ppgtt_type)
 #define HAS_PPGTT(dev_priv) \
 	(INTEL_PPGTT(dev_priv) != INTEL_PPGTT_NONE)
 #define HAS_FULL_PPGTT(dev_priv) \
 	(INTEL_PPGTT(dev_priv) >= INTEL_PPGTT_FULL)
-#define HAS_FULL_48BIT_PPGTT(dev_priv)	\
-	(INTEL_PPGTT(dev_priv) >= INTEL_PPGTT_FULL_4LVL)
 
 #define HAS_PAGE_SIZES(dev_priv, sizes) ({ \
 	GEM_BUG_ON((sizes) == 0); \
@@ -2512,6 +2579,7 @@ static inline unsigned int i915_sg_segment_size(void)
 #define HAS_DDI(dev_priv)		 (INTEL_INFO(dev_priv)->display.has_ddi)
 #define HAS_FPGA_DBG_UNCLAIMED(dev_priv) (INTEL_INFO(dev_priv)->has_fpga_dbg)
 #define HAS_PSR(dev_priv)		 (INTEL_INFO(dev_priv)->display.has_psr)
+#define HAS_TRANSCODER_EDP(dev_priv)	 (INTEL_INFO(dev_priv)->trans_offsets[TRANSCODER_EDP] != 0)
 
 #define HAS_RC6(dev_priv)		 (INTEL_INFO(dev_priv)->has_rc6)
 #define HAS_RC6p(dev_priv)		 (INTEL_INFO(dev_priv)->has_rc6p)
@@ -2558,6 +2626,7 @@ static inline unsigned int i915_sg_segment_size(void)
 #define INTEL_PCH_KBP_DEVICE_ID_TYPE		0xA280
 #define INTEL_PCH_CNP_DEVICE_ID_TYPE		0xA300
 #define INTEL_PCH_CNP_LP_DEVICE_ID_TYPE		0x9D80
+#define INTEL_PCH_CMP_DEVICE_ID_TYPE		0x0280
 #define INTEL_PCH_ICP_DEVICE_ID_TYPE		0x3480
 #define INTEL_PCH_P2X_DEVICE_ID_TYPE		0x7100
 #define INTEL_PCH_P3X_DEVICE_ID_TYPE		0x7000
@@ -2567,8 +2636,6 @@ static inline unsigned int i915_sg_segment_size(void)
 #define INTEL_PCH_ID(dev_priv) ((dev_priv)->pch_id)
 #define HAS_PCH_ICP(dev_priv) (INTEL_PCH_TYPE(dev_priv) == PCH_ICP)
 #define HAS_PCH_CNP(dev_priv) (INTEL_PCH_TYPE(dev_priv) == PCH_CNP)
-#define HAS_PCH_CNP_LP(dev_priv) \
-	(INTEL_PCH_ID(dev_priv) == INTEL_PCH_CNP_LP_DEVICE_ID_TYPE)
 #define HAS_PCH_KBP(dev_priv) (INTEL_PCH_TYPE(dev_priv) == PCH_KBP)
 #define HAS_PCH_SPT(dev_priv) (INTEL_PCH_TYPE(dev_priv) == PCH_SPT)
 #define HAS_PCH_LPT(dev_priv) (INTEL_PCH_TYPE(dev_priv) == PCH_LPT)
@@ -2800,8 +2867,6 @@ void i915_gem_load_init_fences(struct drm_i915_private *dev_priv);
 int i915_gem_freeze(struct drm_i915_private *dev_priv);
 int i915_gem_freeze_late(struct drm_i915_private *dev_priv);
 
-void *i915_gem_object_alloc(struct drm_i915_private *dev_priv);
-void i915_gem_object_free(struct drm_i915_gem_object *obj);
 void i915_gem_object_init(struct drm_i915_gem_object *obj,
 			 const struct drm_i915_gem_object_ops *ops);
 struct drm_i915_gem_object *
@@ -2844,6 +2909,7 @@ static inline void i915_gem_drain_workqueue(struct drm_i915_private *i915)
 	int pass = 2;
 	do {
 		rcu_barrier();
+		i915_gem_drain_freed_objects(i915);
 		drain_workqueue(i915->wq);
 	} while (--pass);
 }
@@ -2974,6 +3040,14 @@ i915_coherent_map_type(struct drm_i915_private *i915)
 void *__must_check i915_gem_object_pin_map(struct drm_i915_gem_object *obj,
 					   enum i915_map_type type);
 
+void __i915_gem_object_flush_map(struct drm_i915_gem_object *obj,
+				 unsigned long offset,
+				 unsigned long size);
+static inline void i915_gem_object_flush_map(struct drm_i915_gem_object *obj)
+{
+	__i915_gem_object_flush_map(obj, 0, obj->base.size);
+}
+
 /**
  * i915_gem_object_unpin_map - releases an earlier mapping
  * @obj: the object to unmap
@@ -3002,7 +3076,12 @@ i915_gem_obj_finish_shmem_access(struct drm_i915_gem_object *obj)
 	i915_gem_object_unpin_pages(obj);
 }
 
-int __must_check i915_mutex_lock_interruptible(struct drm_device *dev);
+static inline int __must_check
+i915_mutex_lock_interruptible(struct drm_device *dev)
+{
+	return mutex_lock_interruptible(&dev->struct_mutex);
+}
+
 int i915_gem_dumb_create(struct drm_file *file_priv,
 			 struct drm_device *dev,
 			 struct drm_mode_create_dumb *args);
@@ -3016,22 +3095,14 @@ void i915_gem_track_fb(struct drm_i915_gem_object *old,
 
 int __must_check i915_gem_set_global_seqno(struct drm_device *dev, u32 seqno);
 
-struct i915_request *
-i915_gem_find_active_request(struct intel_engine_cs *engine);
-
-static inline bool i915_reset_backoff(struct i915_gpu_error *error)
-{
-	return unlikely(test_bit(I915_RESET_BACKOFF, &error->flags));
-}
-
-static inline bool i915_terminally_wedged(struct i915_gpu_error *error)
+static inline bool __i915_wedged(struct i915_gpu_error *error)
 {
 	return unlikely(test_bit(I915_WEDGED, &error->flags));
 }
 
-static inline bool i915_reset_backoff_or_wedged(struct i915_gpu_error *error)
+static inline bool i915_reset_failed(struct drm_i915_private *i915)
 {
-	return i915_reset_backoff(error) | i915_terminally_wedged(error);
+	return __i915_wedged(&i915->gpu_error);
 }
 
 static inline u32 i915_reset_count(struct i915_gpu_error *error)
@@ -3056,14 +3127,13 @@ void i915_gem_fini(struct drm_i915_private *dev_priv);
 void i915_gem_cleanup_engines(struct drm_i915_private *dev_priv);
 int i915_gem_wait_for_idle(struct drm_i915_private *dev_priv,
 			   unsigned int flags, long timeout);
-int __must_check i915_gem_suspend(struct drm_i915_private *dev_priv);
+void i915_gem_suspend(struct drm_i915_private *dev_priv);
 void i915_gem_suspend_late(struct drm_i915_private *dev_priv);
 void i915_gem_resume(struct drm_i915_private *dev_priv);
 vm_fault_t i915_gem_fault(struct vm_fault *vmf);
 int i915_gem_object_wait(struct drm_i915_gem_object *obj,
 			 unsigned int flags,
-			 long timeout,
-			 struct intel_rps_client *rps);
+			 long timeout);
 int i915_gem_object_wait_priority(struct drm_i915_gem_object *obj,
 				  unsigned int flags,
 				  const struct i915_sched_attr *attr);
@@ -3106,7 +3176,6 @@ struct drm_i915_fence_reg *
 i915_reserve_fence(struct drm_i915_private *dev_priv);
 void i915_unreserve_fence(struct drm_i915_fence_reg *fence);
 
-void i915_gem_revoke_fences(struct drm_i915_private *dev_priv);
 void i915_gem_restore_fences(struct drm_i915_private *dev_priv);
 
 void i915_gem_detect_bit_6_swizzle(struct drm_i915_private *dev_priv);
@@ -3142,7 +3211,7 @@ int i915_perf_add_config_ioctl(struct drm_device *dev, void *data,
 int i915_perf_remove_config_ioctl(struct drm_device *dev, void *data,
 				  struct drm_file *file);
 void i915_oa_init_reg_state(struct intel_engine_cs *engine,
-			    struct i915_gem_context *ctx,
+			    struct intel_context *ce,
 			    u32 *reg_state);
 
 /* i915_gem_evict.c */
@@ -3457,18 +3526,21 @@ static inline u64 intel_rc6_residency_us(struct drm_i915_private *dev_priv,
 	return DIV_ROUND_UP_ULL(intel_rc6_residency_ns(dev_priv, reg), 1000);
 }
 
-#define I915_READ8(reg)		dev_priv->uncore.funcs.mmio_readb(dev_priv, (reg), true)
-#define I915_WRITE8(reg, val)	dev_priv->uncore.funcs.mmio_writeb(dev_priv, (reg), (val), true)
+#define __I915_REG_OP(op__, dev_priv__, ...) \
+	intel_uncore_##op__(&(dev_priv__)->uncore, __VA_ARGS__)
+
+#define I915_READ8(reg__)	  __I915_REG_OP(read8, dev_priv, (reg__))
+#define I915_WRITE8(reg__, val__) __I915_REG_OP(write8, dev_priv, (reg__), (val__))
 
-#define I915_READ16(reg)	dev_priv->uncore.funcs.mmio_readw(dev_priv, (reg), true)
-#define I915_WRITE16(reg, val)	dev_priv->uncore.funcs.mmio_writew(dev_priv, (reg), (val), true)
-#define I915_READ16_NOTRACE(reg)	dev_priv->uncore.funcs.mmio_readw(dev_priv, (reg), false)
-#define I915_WRITE16_NOTRACE(reg, val)	dev_priv->uncore.funcs.mmio_writew(dev_priv, (reg), (val), false)
+#define I915_READ16(reg__)	   __I915_REG_OP(read16, dev_priv, (reg__))
+#define I915_WRITE16(reg__, val__) __I915_REG_OP(write16, dev_priv, (reg__), (val__))
+#define I915_READ16_NOTRACE(reg__)	   __I915_REG_OP(read16_notrace, dev_priv, (reg__))
+#define I915_WRITE16_NOTRACE(reg__, val__) __I915_REG_OP(write16_notrace, dev_priv, (reg__), (val__))
 
-#define I915_READ(reg)		dev_priv->uncore.funcs.mmio_readl(dev_priv, (reg), true)
-#define I915_WRITE(reg, val)	dev_priv->uncore.funcs.mmio_writel(dev_priv, (reg), (val), true)
-#define I915_READ_NOTRACE(reg)		dev_priv->uncore.funcs.mmio_readl(dev_priv, (reg), false)
-#define I915_WRITE_NOTRACE(reg, val)	dev_priv->uncore.funcs.mmio_writel(dev_priv, (reg), (val), false)
+#define I915_READ(reg__)	 __I915_REG_OP(read, dev_priv, (reg__))
+#define I915_WRITE(reg__, val__) __I915_REG_OP(write, dev_priv, (reg__), (val__))
+#define I915_READ_NOTRACE(reg__)	 __I915_REG_OP(read_notrace, dev_priv, (reg__))
+#define I915_WRITE_NOTRACE(reg__, val__) __I915_REG_OP(write_notrace, dev_priv, (reg__), (val__))
 
 /* Be very careful with read/write 64-bit values. On 32-bit machines, they
  * will be implemented using 2 32-bit writes in an arbitrary order with
@@ -3484,46 +3556,12 @@ static inline u64 intel_rc6_residency_us(struct drm_i915_private *dev_priv,
  *
  * You have been warned.
  */
-#define I915_READ64(reg)	dev_priv->uncore.funcs.mmio_readq(dev_priv, (reg), true)
-
-#define I915_READ64_2x32(lower_reg, upper_reg) ({			\
-	u32 upper, lower, old_upper, loop = 0;				\
-	upper = I915_READ(upper_reg);					\
-	do {								\
-		old_upper = upper;					\
-		lower = I915_READ(lower_reg);				\
-		upper = I915_READ(upper_reg);				\
-	} while (upper != old_upper && loop++ < 2);			\
-	(u64)upper << 32 | lower; })
-
-#define POSTING_READ(reg)	(void)I915_READ_NOTRACE(reg)
-#define POSTING_READ16(reg)	(void)I915_READ16_NOTRACE(reg)
-
-#define __raw_read(x, s) \
-static inline uint##x##_t __raw_i915_read##x(const struct drm_i915_private *dev_priv, \
-					     i915_reg_t reg) \
-{ \
-	return read##s(dev_priv->regs + i915_mmio_reg_offset(reg)); \
-}
-
-#define __raw_write(x, s) \
-static inline void __raw_i915_write##x(const struct drm_i915_private *dev_priv, \
-				       i915_reg_t reg, uint##x##_t val) \
-{ \
-	write##s(val, dev_priv->regs + i915_mmio_reg_offset(reg)); \
-}
-__raw_read(8, b)
-__raw_read(16, w)
-__raw_read(32, l)
-__raw_read(64, q)
-
-__raw_write(8, b)
-__raw_write(16, w)
-__raw_write(32, l)
-__raw_write(64, q)
+#define I915_READ64(reg__)	__I915_REG_OP(read64, dev_priv, (reg__))
+#define I915_READ64_2x32(lower_reg__, upper_reg__) \
+	__I915_REG_OP(read64_2x32, dev_priv, (lower_reg__), (upper_reg__))
 
-#undef __raw_read
-#undef __raw_write
+#define POSTING_READ(reg__)	__I915_REG_OP(posting_read, dev_priv, (reg__))
+#define POSTING_READ16(reg__)	__I915_REG_OP(posting_read16, dev_priv, (reg__))
 
 /* These are untraced mmio-accessors that are only valid to be used inside
  * critical sections, such as inside IRQ handlers, where forcewake is explicitly
@@ -3551,10 +3589,10 @@ __raw_write(64, q)
  * therefore generally be serialised, by either the dev_priv->uncore.lock or
  * a more localised lock guarding all access to that bank of registers.
  */
-#define I915_READ_FW(reg__) __raw_i915_read32(dev_priv, (reg__))
-#define I915_WRITE_FW(reg__, val__) __raw_i915_write32(dev_priv, (reg__), (val__))
-#define I915_WRITE64_FW(reg__, val__) __raw_i915_write64(dev_priv, (reg__), (val__))
-#define POSTING_READ_FW(reg__) (void)I915_READ_FW(reg__)
+#define I915_READ_FW(reg__) __I915_REG_OP(read_fw, dev_priv, (reg__))
+#define I915_WRITE_FW(reg__, val__) __I915_REG_OP(write_fw, dev_priv, (reg__), (val__))
+#define I915_WRITE64_FW(reg__, val__) __I915_REG_OP(write64_fw, dev_priv, (reg__), (val__))
+#define POSTING_READ_FW(reg__) __I915_REG_OP(posting_read_fw, dev_priv, (reg__))
 
 /* "Broadcast RGB" property */
 #define INTEL_BROADCAST_RGB_AUTO 0
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 3c724cc949a5..ad01c92aaf74 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -42,6 +42,7 @@
 #include "i915_drv.h"
 #include "i915_gem_clflush.h"
 #include "i915_gemfs.h"
+#include "i915_globals.h"
 #include "i915_reset.h"
 #include "i915_trace.h"
 #include "i915_vgpu.h"
@@ -49,6 +50,7 @@
 #include "intel_drv.h"
 #include "intel_frontbuffer.h"
 #include "intel_mocs.h"
+#include "intel_pm.h"
 #include "intel_workarounds.h"
 
 static void i915_gem_flush_free_objects(struct drm_i915_private *i915);
@@ -100,48 +102,7 @@ static void i915_gem_info_remove_obj(struct drm_i915_private *dev_priv,
 	spin_unlock(&dev_priv->mm.object_stat_lock);
 }
 
-static int
-i915_gem_wait_for_error(struct i915_gpu_error *error)
-{
-	int ret;
-
-	might_sleep();
-
-	/*
-	 * Only wait 10 seconds for the gpu reset to complete to avoid hanging
-	 * userspace. If it takes that long something really bad is going on and
-	 * we should simply try to bail out and fail as gracefully as possible.
-	 */
-	ret = wait_event_interruptible_timeout(error->reset_queue,
-					       !i915_reset_backoff(error),
-					       I915_RESET_TIMEOUT);
-	if (ret == 0) {
-		DRM_ERROR("Timed out waiting for the gpu reset to complete\n");
-		return -EIO;
-	} else if (ret < 0) {
-		return ret;
-	} else {
-		return 0;
-	}
-}
-
-int i915_mutex_lock_interruptible(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	int ret;
-
-	ret = i915_gem_wait_for_error(&dev_priv->gpu_error);
-	if (ret)
-		return ret;
-
-	ret = mutex_lock_interruptible(&dev->struct_mutex);
-	if (ret)
-		return ret;
-
-	return 0;
-}
-
-static u32 __i915_gem_park(struct drm_i915_private *i915)
+static void __i915_gem_park(struct drm_i915_private *i915)
 {
 	intel_wakeref_t wakeref;
 
@@ -152,9 +113,7 @@ static u32 __i915_gem_park(struct drm_i915_private *i915)
 	GEM_BUG_ON(!list_empty(&i915->gt.active_rings));
 
 	if (!i915->gt.awake)
-		return I915_EPOCH_INVALID;
-
-	GEM_BUG_ON(i915->gt.epoch == I915_EPOCH_INVALID);
+		return;
 
 	/*
 	 * Be paranoid and flush a concurrent interrupt to make sure
@@ -183,7 +142,7 @@ static u32 __i915_gem_park(struct drm_i915_private *i915)
 
 	intel_display_power_put(i915, POWER_DOMAIN_GT_IRQ, wakeref);
 
-	return i915->gt.epoch;
+	i915_globals_park();
 }
 
 void i915_gem_park(struct drm_i915_private *i915)
@@ -225,8 +184,7 @@ void i915_gem_unpark(struct drm_i915_private *i915)
 	i915->gt.awake = intel_display_power_get(i915, POWER_DOMAIN_GT_IRQ);
 	GEM_BUG_ON(!i915->gt.awake);
 
-	if (unlikely(++i915->gt.epoch == 0)) /* keep 0 as invalid */
-		i915->gt.epoch = 1;
+	i915_globals_unpark();
 
 	intel_enable_gt_powersave(i915);
 	i915_update_gfx_val(i915);
@@ -351,7 +309,7 @@ static void __start_cpu_write(struct drm_i915_gem_object *obj)
 		obj->cache_dirty = true;
 }
 
-static void
+void
 __i915_gem_object_release_shmem(struct drm_i915_gem_object *obj,
 				struct sg_table *pages,
 				bool needs_clflush)
@@ -459,8 +417,7 @@ int i915_gem_object_unbind(struct drm_i915_gem_object *obj)
 static long
 i915_gem_object_wait_fence(struct dma_fence *fence,
 			   unsigned int flags,
-			   long timeout,
-			   struct intel_rps_client *rps_client)
+			   long timeout)
 {
 	struct i915_request *rq;
 
@@ -478,27 +435,6 @@ i915_gem_object_wait_fence(struct dma_fence *fence,
 	if (i915_request_completed(rq))
 		goto out;
 
-	/*
-	 * This client is about to stall waiting for the GPU. In many cases
-	 * this is undesirable and limits the throughput of the system, as
-	 * many clients cannot continue processing user input/output whilst
-	 * blocked. RPS autotuning may take tens of milliseconds to respond
-	 * to the GPU load and thus incurs additional latency for the client.
-	 * We can circumvent that by promoting the GPU frequency to maximum
-	 * before we wait. This makes the GPU throttle up much more quickly
-	 * (good for benchmarks and user experience, e.g. window animations),
-	 * but at a cost of spending more power processing the workload
-	 * (bad for battery). Not all clients even want their results
-	 * immediately and for them we should just let the GPU select its own
-	 * frequency to maximise efficiency. To prevent a single client from
-	 * forcing the clocks too high for the whole system, we only allow
-	 * each client to waitboost once in a busy period.
-	 */
-	if (rps_client && !i915_request_started(rq)) {
-		if (INTEL_GEN(rq->i915) >= 6)
-			gen6_rps_boost(rq, rps_client);
-	}
-
 	timeout = i915_request_wait(rq, flags, timeout);
 
 out:
@@ -511,8 +447,7 @@ out:
 static long
 i915_gem_object_wait_reservation(struct reservation_object *resv,
 				 unsigned int flags,
-				 long timeout,
-				 struct intel_rps_client *rps_client)
+				 long timeout)
 {
 	unsigned int seq = __read_seqcount_begin(&resv->seq);
 	struct dma_fence *excl;
@@ -530,8 +465,7 @@ i915_gem_object_wait_reservation(struct reservation_object *resv,
 
 		for (i = 0; i < count; i++) {
 			timeout = i915_gem_object_wait_fence(shared[i],
-							     flags, timeout,
-							     rps_client);
+							     flags, timeout);
 			if (timeout < 0)
 				break;
 
@@ -557,8 +491,7 @@ i915_gem_object_wait_reservation(struct reservation_object *resv,
 	}
 
 	if (excl && timeout >= 0)
-		timeout = i915_gem_object_wait_fence(excl, flags, timeout,
-						     rps_client);
+		timeout = i915_gem_object_wait_fence(excl, flags, timeout);
 
 	dma_fence_put(excl);
 
@@ -652,30 +585,19 @@ i915_gem_object_wait_priority(struct drm_i915_gem_object *obj,
  * @obj: i915 gem object
  * @flags: how to wait (under a lock, for all rendering or just for writes etc)
  * @timeout: how long to wait
- * @rps_client: client (user process) to charge for any waitboosting
  */
 int
 i915_gem_object_wait(struct drm_i915_gem_object *obj,
 		     unsigned int flags,
-		     long timeout,
-		     struct intel_rps_client *rps_client)
+		     long timeout)
 {
 	might_sleep();
 	GEM_BUG_ON(timeout < 0);
 
-	timeout = i915_gem_object_wait_reservation(obj->resv,
-						   flags, timeout,
-						   rps_client);
+	timeout = i915_gem_object_wait_reservation(obj->resv, flags, timeout);
 	return timeout < 0 ? timeout : 0;
 }
 
-static struct intel_rps_client *to_rps_client(struct drm_file *file)
-{
-	struct drm_i915_file_private *fpriv = file->driver_priv;
-
-	return &fpriv->rps_client;
-}
-
 static int
 i915_gem_phys_pwrite(struct drm_i915_gem_object *obj,
 		     struct drm_i915_gem_pwrite *args,
@@ -698,28 +620,18 @@ i915_gem_phys_pwrite(struct drm_i915_gem_object *obj,
 	return 0;
 }
 
-void *i915_gem_object_alloc(struct drm_i915_private *dev_priv)
-{
-	return kmem_cache_zalloc(dev_priv->objects, GFP_KERNEL);
-}
-
-void i915_gem_object_free(struct drm_i915_gem_object *obj)
-{
-	struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
-	kmem_cache_free(dev_priv->objects, obj);
-}
-
 static int
 i915_gem_create(struct drm_file *file,
 		struct drm_i915_private *dev_priv,
-		u64 size,
+		u64 *size_p,
 		u32 *handle_p)
 {
 	struct drm_i915_gem_object *obj;
-	int ret;
 	u32 handle;
+	u64 size;
+	int ret;
 
-	size = roundup(size, PAGE_SIZE);
+	size = round_up(*size_p, PAGE_SIZE);
 	if (size == 0)
 		return -EINVAL;
 
@@ -735,6 +647,7 @@ i915_gem_create(struct drm_file *file,
 		return ret;
 
 	*handle_p = handle;
+	*size_p = size;
 	return 0;
 }
 
@@ -747,7 +660,7 @@ i915_gem_dumb_create(struct drm_file *file,
 	args->pitch = ALIGN(args->width * DIV_ROUND_UP(args->bpp, 8), 64);
 	args->size = args->pitch * args->height;
 	return i915_gem_create(file, to_i915(dev),
-			       args->size, &args->handle);
+			       &args->size, &args->handle);
 }
 
 static bool gpu_write_needs_clflush(struct drm_i915_gem_object *obj)
@@ -772,7 +685,7 @@ i915_gem_create_ioctl(struct drm_device *dev, void *data,
 	i915_gem_flush_free_objects(dev_priv);
 
 	return i915_gem_create(file, dev_priv,
-			       args->size, &args->handle);
+			       &args->size, &args->handle);
 }
 
 static inline enum fb_op_origin
@@ -881,8 +794,7 @@ int i915_gem_obj_prepare_shmem_read(struct drm_i915_gem_object *obj,
 	ret = i915_gem_object_wait(obj,
 				   I915_WAIT_INTERRUPTIBLE |
 				   I915_WAIT_LOCKED,
-				   MAX_SCHEDULE_TIMEOUT,
-				   NULL);
+				   MAX_SCHEDULE_TIMEOUT);
 	if (ret)
 		return ret;
 
@@ -934,8 +846,7 @@ int i915_gem_obj_prepare_shmem_write(struct drm_i915_gem_object *obj,
 				   I915_WAIT_INTERRUPTIBLE |
 				   I915_WAIT_LOCKED |
 				   I915_WAIT_ALL,
-				   MAX_SCHEDULE_TIMEOUT,
-				   NULL);
+				   MAX_SCHEDULE_TIMEOUT);
 	if (ret)
 		return ret;
 
@@ -1197,8 +1108,7 @@ i915_gem_pread_ioctl(struct drm_device *dev, void *data,
 
 	ret = i915_gem_object_wait(obj,
 				   I915_WAIT_INTERRUPTIBLE,
-				   MAX_SCHEDULE_TIMEOUT,
-				   to_rps_client(file));
+				   MAX_SCHEDULE_TIMEOUT);
 	if (ret)
 		goto out;
 
@@ -1497,8 +1407,7 @@ i915_gem_pwrite_ioctl(struct drm_device *dev, void *data,
 	ret = i915_gem_object_wait(obj,
 				   I915_WAIT_INTERRUPTIBLE |
 				   I915_WAIT_ALL,
-				   MAX_SCHEDULE_TIMEOUT,
-				   to_rps_client(file));
+				   MAX_SCHEDULE_TIMEOUT);
 	if (ret)
 		goto err;
 
@@ -1578,17 +1487,37 @@ i915_gem_set_domain_ioctl(struct drm_device *dev, void *data,
 	if ((write_domain | read_domains) & I915_GEM_GPU_DOMAINS)
 		return -EINVAL;
 
-	/* Having something in the write domain implies it's in the read
+	/*
+	 * Having something in the write domain implies it's in the read
 	 * domain, and only that read domain.  Enforce that in the request.
 	 */
-	if (write_domain != 0 && read_domains != write_domain)
+	if (write_domain && read_domains != write_domain)
 		return -EINVAL;
 
+	if (!read_domains)
+		return 0;
+
 	obj = i915_gem_object_lookup(file, args->handle);
 	if (!obj)
 		return -ENOENT;
 
-	/* Try to flush the object off the GPU without holding the lock.
+	/*
+	 * Already in the desired write domain? Nothing for us to do!
+	 *
+	 * We apply a little bit of cunning here to catch a broader set of
+	 * no-ops. If obj->write_domain is set, we must be in the same
+	 * obj->read_domains, and only that domain. Therefore, if that
+	 * obj->write_domain matches the request read_domains, we are
+	 * already in the same read/write domain and can skip the operation,
+	 * without having to further check the requested write_domain.
+	 */
+	if (READ_ONCE(obj->write_domain) == read_domains) {
+		err = 0;
+		goto out;
+	}
+
+	/*
+	 * Try to flush the object off the GPU without holding the lock.
 	 * We will repeat the flush holding the lock in the normal manner
 	 * to catch cases where we are gazumped.
 	 */
@@ -1596,8 +1525,7 @@ i915_gem_set_domain_ioctl(struct drm_device *dev, void *data,
 				   I915_WAIT_INTERRUPTIBLE |
 				   I915_WAIT_PRIORITY |
 				   (write_domain ? I915_WAIT_ALL : 0),
-				   MAX_SCHEDULE_TIMEOUT,
-				   to_rps_client(file));
+				   MAX_SCHEDULE_TIMEOUT);
 	if (err)
 		goto out;
 
@@ -1808,6 +1736,9 @@ static unsigned int tile_row_pages(const struct drm_i915_gem_object *obj)
  * 2 - Recognise WC as a separate cache domain so that we can flush the
  *     delayed writes via GTT before performing direct access via WC.
  *
+ * 3 - Remove implicit set-domain(GTT) and synchronisation on initial
+ *     pagefault; swapin remains transparent.
+ *
  * Restrictions:
  *
  *  * snoopable objects cannot be accessed via the GTT. It can cause machine
@@ -1835,7 +1766,7 @@ static unsigned int tile_row_pages(const struct drm_i915_gem_object *obj)
  */
 int i915_gem_mmap_gtt_version(void)
 {
-	return 2;
+	return 3;
 }
 
 static inline struct i915_ggtt_view
@@ -1891,6 +1822,7 @@ vm_fault_t i915_gem_fault(struct vm_fault *vmf)
 	intel_wakeref_t wakeref;
 	struct i915_vma *vma;
 	pgoff_t page_offset;
+	int srcu;
 	int ret;
 
 	/* Sanity check that we allow writing into this object */
@@ -1902,27 +1834,21 @@ vm_fault_t i915_gem_fault(struct vm_fault *vmf)
 
 	trace_i915_gem_object_fault(obj, page_offset, true, write);
 
-	/* Try to flush the object off the GPU first without holding the lock.
-	 * Upon acquiring the lock, we will perform our sanity checks and then
-	 * repeat the flush holding the lock in the normal manner to catch cases
-	 * where we are gazumped.
-	 */
-	ret = i915_gem_object_wait(obj,
-				   I915_WAIT_INTERRUPTIBLE,
-				   MAX_SCHEDULE_TIMEOUT,
-				   NULL);
-	if (ret)
-		goto err;
-
 	ret = i915_gem_object_pin_pages(obj);
 	if (ret)
 		goto err;
 
 	wakeref = intel_runtime_pm_get(dev_priv);
 
+	srcu = i915_reset_trylock(dev_priv);
+	if (srcu < 0) {
+		ret = srcu;
+		goto err_rpm;
+	}
+
 	ret = i915_mutex_lock_interruptible(dev);
 	if (ret)
-		goto err_rpm;
+		goto err_reset;
 
 	/* Access to snoopable pages through the GTT is incoherent. */
 	if (obj->cache_level != I915_CACHE_NONE && !HAS_LLC(dev_priv)) {
@@ -1930,7 +1856,6 @@ vm_fault_t i915_gem_fault(struct vm_fault *vmf)
 		goto err_unlock;
 	}
 
-
 	/* Now pin it into the GTT as needed */
 	vma = i915_gem_object_ggtt_pin(obj, NULL, 0, 0,
 				       PIN_MAPPABLE |
@@ -1964,10 +1889,6 @@ vm_fault_t i915_gem_fault(struct vm_fault *vmf)
 		goto err_unlock;
 	}
 
-	ret = i915_gem_object_set_to_gtt_domain(obj, write);
-	if (ret)
-		goto err_unpin;
-
 	ret = i915_vma_pin_fence(vma);
 	if (ret)
 		goto err_unpin;
@@ -1995,6 +1916,8 @@ err_unpin:
 	__i915_vma_unpin(vma);
 err_unlock:
 	mutex_unlock(&dev->struct_mutex);
+err_reset:
+	i915_reset_unlock(dev_priv, srcu);
 err_rpm:
 	intel_runtime_pm_put(dev_priv, wakeref);
 	i915_gem_object_unpin_pages(obj);
@@ -2007,7 +1930,7 @@ err:
 		 * fail). But any other -EIO isn't ours (e.g. swap in failure)
 		 * and so needs to be reported.
 		 */
-		if (!i915_terminally_wedged(&dev_priv->gpu_error))
+		if (!i915_terminally_wedged(dev_priv))
 			return VM_FAULT_SIGBUS;
 		/* else: fall through */
 	case -EAGAIN:
@@ -2280,7 +2203,6 @@ i915_gem_object_put_pages_gtt(struct drm_i915_gem_object *obj,
 	struct page *page;
 
 	__i915_gem_object_release_shmem(obj, pages, true);
-
 	i915_gem_gtt_finish_pages(obj, pages);
 
 	if (i915_gem_object_needs_bit17_swizzle(obj))
@@ -2488,7 +2410,7 @@ rebuild_st:
 		do {
 			cond_resched();
 			page = shmem_read_mapping_page_gfp(mapping, i, gfp);
-			if (likely(!IS_ERR(page)))
+			if (!IS_ERR(page))
 				break;
 
 			if (!*s) {
@@ -2622,6 +2544,14 @@ void __i915_gem_object_set_pages(struct drm_i915_gem_object *obj,
 
 	lockdep_assert_held(&obj->mm.lock);
 
+	/* Make the pages coherent with the GPU (flushing any swapin). */
+	if (obj->cache_dirty) {
+		obj->write_domain = 0;
+		if (i915_gem_object_has_struct_page(obj))
+			drm_clflush_sg(pages);
+		obj->cache_dirty = false;
+	}
+
 	obj->mm.get_page.sg_pos = pages->sgl;
 	obj->mm.get_page.sg_idx = 0;
 
@@ -2823,6 +2753,33 @@ err_unlock:
 	goto out_unlock;
 }
 
+void __i915_gem_object_flush_map(struct drm_i915_gem_object *obj,
+				 unsigned long offset,
+				 unsigned long size)
+{
+	enum i915_map_type has_type;
+	void *ptr;
+
+	GEM_BUG_ON(!i915_gem_object_has_pinned_pages(obj));
+	GEM_BUG_ON(range_overflows_t(typeof(obj->base.size),
+				     offset, size, obj->base.size));
+
+	obj->mm.dirty = true;
+
+	if (obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_WRITE)
+		return;
+
+	ptr = page_unpack_bits(obj->mm.mapping, &has_type);
+	if (has_type == I915_MAP_WC)
+		return;
+
+	drm_clflush_virt_range(ptr + offset, size);
+	if (size == obj->base.size) {
+		obj->write_domain &= ~I915_GEM_DOMAIN_CPU;
+		obj->cache_dirty = false;
+	}
+}
+
 static int
 i915_gem_object_pwrite_gtt(struct drm_i915_gem_object *obj,
 			   const struct drm_i915_gem_pwrite *arg)
@@ -2832,7 +2789,11 @@ i915_gem_object_pwrite_gtt(struct drm_i915_gem_object *obj,
 	u64 remain, offset;
 	unsigned int pg;
 
-	/* Before we instantiate/pin the backing store for our use, we
+	/* Caller already validated user args */
+	GEM_BUG_ON(!access_ok(user_data, arg->size));
+
+	/*
+	 * Before we instantiate/pin the backing store for our use, we
 	 * can prepopulate the shmemfs filp efficiently using a write into
 	 * the pagecache. We avoid the penalty of instantiating all the
 	 * pages, important if the user is just writing to a few and never
@@ -2846,7 +2807,8 @@ i915_gem_object_pwrite_gtt(struct drm_i915_gem_object *obj,
 	if (obj->mm.madv != I915_MADV_WILLNEED)
 		return -EFAULT;
 
-	/* Before the pages are instantiated the object is treated as being
+	/*
+	 * Before the pages are instantiated the object is treated as being
 	 * in the CPU domain. The pages will be clflushed as required before
 	 * use, and we can freely write into the pages directly. If userspace
 	 * races pwrite with any other operation; corruption will ensue -
@@ -2862,20 +2824,32 @@ i915_gem_object_pwrite_gtt(struct drm_i915_gem_object *obj,
 		struct page *page;
 		void *data, *vaddr;
 		int err;
+		char c;
 
 		len = PAGE_SIZE - pg;
 		if (len > remain)
 			len = remain;
 
+		/* Prefault the user page to reduce potential recursion */
+		err = __get_user(c, user_data);
+		if (err)
+			return err;
+
+		err = __get_user(c, user_data + len - 1);
+		if (err)
+			return err;
+
 		err = pagecache_write_begin(obj->base.filp, mapping,
 					    offset, len, 0,
 					    &page, &data);
 		if (err < 0)
 			return err;
 
-		vaddr = kmap(page);
-		unwritten = copy_from_user(vaddr + pg, user_data, len);
-		kunmap(page);
+		vaddr = kmap_atomic(page);
+		unwritten = __copy_from_user_inatomic(vaddr + pg,
+						      user_data,
+						      len);
+		kunmap_atomic(vaddr);
 
 		err = pagecache_write_end(obj->base.filp, mapping,
 					  offset, len, len - unwritten,
@@ -2883,8 +2857,9 @@ i915_gem_object_pwrite_gtt(struct drm_i915_gem_object *obj,
 		if (err < 0)
 			return err;
 
+		/* We don't handle -EFAULT, leave it to the caller to check */
 		if (unwritten)
-			return -EFAULT;
+			return -ENODEV;
 
 		remain -= len;
 		user_data += len;
@@ -2895,51 +2870,6 @@ i915_gem_object_pwrite_gtt(struct drm_i915_gem_object *obj,
 	return 0;
 }
 
-static bool match_ring(struct i915_request *rq)
-{
-	struct drm_i915_private *dev_priv = rq->i915;
-	u32 ring = I915_READ(RING_START(rq->engine->mmio_base));
-
-	return ring == i915_ggtt_offset(rq->ring->vma);
-}
-
-struct i915_request *
-i915_gem_find_active_request(struct intel_engine_cs *engine)
-{
-	struct i915_request *request, *active = NULL;
-	unsigned long flags;
-
-	/*
-	 * We are called by the error capture, reset and to dump engine
-	 * state at random points in time. In particular, note that neither is
-	 * crucially ordered with an interrupt. After a hang, the GPU is dead
-	 * and we assume that no more writes can happen (we waited long enough
-	 * for all writes that were in transaction to be flushed) - adding an
-	 * extra delay for a recent interrupt is pointless. Hence, we do
-	 * not need an engine->irq_seqno_barrier() before the seqno reads.
-	 * At all other times, we must assume the GPU is still running, but
-	 * we only care about the snapshot of this moment.
-	 */
-	spin_lock_irqsave(&engine->timeline.lock, flags);
-	list_for_each_entry(request, &engine->timeline.requests, link) {
-		if (i915_request_completed(request))
-			continue;
-
-		if (!i915_request_started(request))
-			break;
-
-		/* More than one preemptible request may match! */
-		if (!match_ring(request))
-			break;
-
-		active = request;
-		break;
-	}
-	spin_unlock_irqrestore(&engine->timeline.lock, flags);
-
-	return active;
-}
-
 static void
 i915_gem_retire_work_handler(struct work_struct *work)
 {
@@ -2964,180 +2894,105 @@ i915_gem_retire_work_handler(struct work_struct *work)
 				   round_jiffies_up_relative(HZ));
 }
 
-static void shrink_caches(struct drm_i915_private *i915)
+static bool switch_to_kernel_context_sync(struct drm_i915_private *i915,
+					  unsigned long mask)
 {
-	/*
-	 * kmem_cache_shrink() discards empty slabs and reorders partially
-	 * filled slabs to prioritise allocating from the mostly full slabs,
-	 * with the aim of reducing fragmentation.
-	 */
-	kmem_cache_shrink(i915->priorities);
-	kmem_cache_shrink(i915->dependencies);
-	kmem_cache_shrink(i915->requests);
-	kmem_cache_shrink(i915->luts);
-	kmem_cache_shrink(i915->vmas);
-	kmem_cache_shrink(i915->objects);
-}
-
-struct sleep_rcu_work {
-	union {
-		struct rcu_head rcu;
-		struct work_struct work;
-	};
-	struct drm_i915_private *i915;
-	unsigned int epoch;
-};
+	bool result = true;
 
-static inline bool
-same_epoch(struct drm_i915_private *i915, unsigned int epoch)
-{
 	/*
-	 * There is a small chance that the epoch wrapped since we started
-	 * sleeping. If we assume that epoch is at least a u32, then it will
-	 * take at least 2^32 * 100ms for it to wrap, or about 326 years.
+	 * Even if we fail to switch, give whatever is running a small chance
+	 * to save itself before we report the failure. Yes, this may be a
+	 * false positive due to e.g. ENOMEM, caveat emptor!
 	 */
-	return epoch == READ_ONCE(i915->gt.epoch);
-}
-
-static void __sleep_work(struct work_struct *work)
-{
-	struct sleep_rcu_work *s = container_of(work, typeof(*s), work);
-	struct drm_i915_private *i915 = s->i915;
-	unsigned int epoch = s->epoch;
-
-	kfree(s);
-	if (same_epoch(i915, epoch))
-		shrink_caches(i915);
-}
+	if (i915_gem_switch_to_kernel_context(i915, mask))
+		result = false;
 
-static void __sleep_rcu(struct rcu_head *rcu)
-{
-	struct sleep_rcu_work *s = container_of(rcu, typeof(*s), rcu);
-	struct drm_i915_private *i915 = s->i915;
-
-	destroy_rcu_head(&s->rcu);
+	if (i915_gem_wait_for_idle(i915,
+				   I915_WAIT_LOCKED |
+				   I915_WAIT_FOR_IDLE_BOOST,
+				   I915_GEM_IDLE_TIMEOUT))
+		result = false;
+
+	if (!result) {
+		if (i915_modparams.reset) { /* XXX hide warning from gem_eio */
+			dev_err(i915->drm.dev,
+				"Failed to idle engines, declaring wedged!\n");
+			GEM_TRACE_DUMP();
+		}
 
-	if (same_epoch(i915, s->epoch)) {
-		INIT_WORK(&s->work, __sleep_work);
-		queue_work(i915->wq, &s->work);
-	} else {
-		kfree(s);
+		/* Forcibly cancel outstanding work and leave the gpu quiet. */
+		i915_gem_set_wedged(i915);
 	}
-}
 
-static inline bool
-new_requests_since_last_retire(const struct drm_i915_private *i915)
-{
-	return (READ_ONCE(i915->gt.active_requests) ||
-		work_pending(&i915->gt.idle_work.work));
+	i915_retire_requests(i915); /* ensure we flush after wedging */
+	return result;
 }
 
-static void assert_kernel_context_is_current(struct drm_i915_private *i915)
+static bool load_power_context(struct drm_i915_private *i915)
 {
-	struct intel_engine_cs *engine;
-	enum intel_engine_id id;
+	/* Force loading the kernel context on all engines */
+	if (!switch_to_kernel_context_sync(i915, ALL_ENGINES))
+		return false;
 
-	if (i915_terminally_wedged(&i915->gpu_error))
-		return;
+	/*
+	 * Immediately park the GPU so that we enable powersaving and
+	 * treat it as idle. The next time we issue a request, we will
+	 * unpark and start using the engine->pinned_default_state, otherwise
+	 * it is in limbo and an early reset may fail.
+	 */
+	__i915_gem_park(i915);
 
-	GEM_BUG_ON(i915->gt.active_requests);
-	for_each_engine(engine, i915, id) {
-		GEM_BUG_ON(__i915_active_request_peek(&engine->timeline.last_request));
-		GEM_BUG_ON(engine->last_retired_context !=
-			   to_intel_context(i915->kernel_context, engine));
-	}
+	return true;
 }
 
 static void
 i915_gem_idle_work_handler(struct work_struct *work)
 {
-	struct drm_i915_private *dev_priv =
-		container_of(work, typeof(*dev_priv), gt.idle_work.work);
-	unsigned int epoch = I915_EPOCH_INVALID;
+	struct drm_i915_private *i915 =
+		container_of(work, typeof(*i915), gt.idle_work.work);
 	bool rearm_hangcheck;
 
-	if (!READ_ONCE(dev_priv->gt.awake))
+	if (!READ_ONCE(i915->gt.awake))
 		return;
 
-	if (READ_ONCE(dev_priv->gt.active_requests))
+	if (READ_ONCE(i915->gt.active_requests))
 		return;
 
-	/*
-	 * Flush out the last user context, leaving only the pinned
-	 * kernel context resident. When we are idling on the kernel_context,
-	 * no more new requests (with a context switch) are emitted and we
-	 * can finally rest. A consequence is that the idle work handler is
-	 * always called at least twice before idling (and if the system is
-	 * idle that implies a round trip through the retire worker).
-	 */
-	mutex_lock(&dev_priv->drm.struct_mutex);
-	i915_gem_switch_to_kernel_context(dev_priv);
-	mutex_unlock(&dev_priv->drm.struct_mutex);
-
-	GEM_TRACE("active_requests=%d (after switch-to-kernel-context)\n",
-		  READ_ONCE(dev_priv->gt.active_requests));
-
-	/*
-	 * Wait for last execlists context complete, but bail out in case a
-	 * new request is submitted. As we don't trust the hardware, we
-	 * continue on if the wait times out. This is necessary to allow
-	 * the machine to suspend even if the hardware dies, and we will
-	 * try to recover in resume (after depriving the hardware of power,
-	 * it may be in a better mmod).
-	 */
-	__wait_for(if (new_requests_since_last_retire(dev_priv)) return,
-		   intel_engines_are_idle(dev_priv),
-		   I915_IDLE_ENGINES_TIMEOUT * 1000,
-		   10, 500);
-
 	rearm_hangcheck =
-		cancel_delayed_work_sync(&dev_priv->gpu_error.hangcheck_work);
+		cancel_delayed_work_sync(&i915->gpu_error.hangcheck_work);
 
-	if (!mutex_trylock(&dev_priv->drm.struct_mutex)) {
+	if (!mutex_trylock(&i915->drm.struct_mutex)) {
 		/* Currently busy, come back later */
-		mod_delayed_work(dev_priv->wq,
-				 &dev_priv->gt.idle_work,
+		mod_delayed_work(i915->wq,
+				 &i915->gt.idle_work,
 				 msecs_to_jiffies(50));
 		goto out_rearm;
 	}
 
 	/*
-	 * New request retired after this work handler started, extend active
-	 * period until next instance of the work.
+	 * Flush out the last user context, leaving only the pinned
+	 * kernel context resident. Should anything unfortunate happen
+	 * while we are idle (such as the GPU being power cycled), no users
+	 * will be harmed.
 	 */
-	if (new_requests_since_last_retire(dev_priv))
-		goto out_unlock;
+	if (!work_pending(&i915->gt.idle_work.work) &&
+	    !i915->gt.active_requests) {
+		++i915->gt.active_requests; /* don't requeue idle */
 
-	epoch = __i915_gem_park(dev_priv);
+		switch_to_kernel_context_sync(i915, i915->gt.active_engines);
 
-	assert_kernel_context_is_current(dev_priv);
+		if (!--i915->gt.active_requests) {
+			__i915_gem_park(i915);
+			rearm_hangcheck = false;
+		}
+	}
 
-	rearm_hangcheck = false;
-out_unlock:
-	mutex_unlock(&dev_priv->drm.struct_mutex);
+	mutex_unlock(&i915->drm.struct_mutex);
 
 out_rearm:
 	if (rearm_hangcheck) {
-		GEM_BUG_ON(!dev_priv->gt.awake);
-		i915_queue_hangcheck(dev_priv);
-	}
-
-	/*
-	 * When we are idle, it is an opportune time to reap our caches.
-	 * However, we have many objects that utilise RCU and the ordered
-	 * i915->wq that this work is executing on. To try and flush any
-	 * pending frees now we are idle, we first wait for an RCU grace
-	 * period, and then queue a task (that will run last on the wq) to
-	 * shrink and re-optimize the caches.
-	 */
-	if (same_epoch(dev_priv, epoch)) {
-		struct sleep_rcu_work *s = kmalloc(sizeof(*s), GFP_KERNEL);
-		if (s) {
-			init_rcu_head(&s->rcu);
-			s->i915 = dev_priv;
-			s->epoch = epoch;
-			call_rcu(&s->rcu, __sleep_rcu);
-		}
+		GEM_BUG_ON(!i915->gt.awake);
+		i915_queue_hangcheck(i915);
 	}
 }
 
@@ -3171,7 +3026,7 @@ void i915_gem_close_object(struct drm_gem_object *gem, struct drm_file *file)
 		list_del(&lut->obj_link);
 		list_del(&lut->ctx_link);
 
-		kmem_cache_free(i915->luts, lut);
+		i915_lut_handle_free(lut);
 		__i915_gem_object_release_unless_active(obj);
 	}
 
@@ -3234,8 +3089,7 @@ i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 				   I915_WAIT_INTERRUPTIBLE |
 				   I915_WAIT_PRIORITY |
 				   I915_WAIT_ALL,
-				   to_wait_timeout(args->timeout_ns),
-				   to_rps_client(file));
+				   to_wait_timeout(args->timeout_ns));
 
 	if (args->timeout_ns > 0) {
 		args->timeout_ns -= ktime_to_ns(ktime_sub(ktime_get(), start));
@@ -3304,7 +3158,7 @@ wait_for_timelines(struct drm_i915_private *i915,
 		 * stalls, so allow the gpu to boost to maximum clocks.
 		 */
 		if (flags & I915_WAIT_FOR_IDLE_BOOST)
-			gen6_rps_boost(rq, NULL);
+			gen6_rps_boost(rq);
 
 		timeout = i915_request_wait(rq, flags, timeout);
 		i915_request_put(rq);
@@ -3340,19 +3194,11 @@ int i915_gem_wait_for_idle(struct drm_i915_private *i915,
 
 		lockdep_assert_held(&i915->drm.struct_mutex);
 
-		if (GEM_SHOW_DEBUG() && !timeout) {
-			/* Presume that timeout was non-zero to begin with! */
-			dev_warn(&i915->drm.pdev->dev,
-				 "Missed idle-completion interrupt!\n");
-			GEM_TRACE_DUMP();
-		}
-
 		err = wait_for_engines(i915);
 		if (err)
 			return err;
 
 		i915_retire_requests(i915);
-		GEM_BUG_ON(i915->gt.active_requests);
 	}
 
 	return 0;
@@ -3399,8 +3245,7 @@ i915_gem_object_set_to_wc_domain(struct drm_i915_gem_object *obj, bool write)
 				   I915_WAIT_INTERRUPTIBLE |
 				   I915_WAIT_LOCKED |
 				   (write ? I915_WAIT_ALL : 0),
-				   MAX_SCHEDULE_TIMEOUT,
-				   NULL);
+				   MAX_SCHEDULE_TIMEOUT);
 	if (ret)
 		return ret;
 
@@ -3462,8 +3307,7 @@ i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, bool write)
 				   I915_WAIT_INTERRUPTIBLE |
 				   I915_WAIT_LOCKED |
 				   (write ? I915_WAIT_ALL : 0),
-				   MAX_SCHEDULE_TIMEOUT,
-				   NULL);
+				   MAX_SCHEDULE_TIMEOUT);
 	if (ret)
 		return ret;
 
@@ -3578,8 +3422,7 @@ restart:
 					   I915_WAIT_INTERRUPTIBLE |
 					   I915_WAIT_LOCKED |
 					   I915_WAIT_ALL,
-					   MAX_SCHEDULE_TIMEOUT,
-					   NULL);
+					   MAX_SCHEDULE_TIMEOUT);
 		if (ret)
 			return ret;
 
@@ -3717,8 +3560,7 @@ int i915_gem_set_caching_ioctl(struct drm_device *dev, void *data,
 
 	ret = i915_gem_object_wait(obj,
 				   I915_WAIT_INTERRUPTIBLE,
-				   MAX_SCHEDULE_TIMEOUT,
-				   to_rps_client(file));
+				   MAX_SCHEDULE_TIMEOUT);
 	if (ret)
 		goto out;
 
@@ -3844,8 +3686,7 @@ i915_gem_object_set_to_cpu_domain(struct drm_i915_gem_object *obj, bool write)
 				   I915_WAIT_INTERRUPTIBLE |
 				   I915_WAIT_LOCKED |
 				   (write ? I915_WAIT_ALL : 0),
-				   MAX_SCHEDULE_TIMEOUT,
-				   NULL);
+				   MAX_SCHEDULE_TIMEOUT);
 	if (ret)
 		return ret;
 
@@ -3891,8 +3732,9 @@ i915_gem_ring_throttle(struct drm_device *dev, struct drm_file *file)
 	long ret;
 
 	/* ABI: return -EIO if already wedged */
-	if (i915_terminally_wedged(&dev_priv->gpu_error))
-		return -EIO;
+	ret = i915_terminally_wedged(dev_priv);
+	if (ret)
+		return ret;
 
 	spin_lock(&file_priv->mm.lock);
 	list_for_each_entry(request, &file_priv->mm.request_list, client_link) {
@@ -3968,7 +3810,7 @@ i915_gem_object_ggtt_pin(struct drm_i915_gem_object *obj,
 	}
 
 	vma = i915_vma_instance(obj, vm, view);
-	if (unlikely(IS_ERR(vma)))
+	if (IS_ERR(vma))
 		return vma;
 
 	if (i915_vma_misplaced(vma, size, alignment, flags)) {
@@ -4000,22 +3842,19 @@ i915_gem_object_ggtt_pin(struct drm_i915_gem_object *obj,
 	return vma;
 }
 
-static __always_inline unsigned int __busy_read_flag(unsigned int id)
+static __always_inline u32 __busy_read_flag(u8 id)
 {
-	/* Note that we could alias engines in the execbuf API, but
-	 * that would be very unwise as it prevents userspace from
-	 * fine control over engine selection. Ahem.
-	 *
-	 * This should be something like EXEC_MAX_ENGINE instead of
-	 * I915_NUM_ENGINES.
-	 */
-	BUILD_BUG_ON(I915_NUM_ENGINES > 16);
-	return 0x10000 << id;
+	if (id == (u8)I915_ENGINE_CLASS_INVALID)
+		return 0xffff0000u;
+
+	GEM_BUG_ON(id >= 16);
+	return 0x10000u << id;
 }
 
-static __always_inline unsigned int __busy_write_id(unsigned int id)
+static __always_inline u32 __busy_write_id(u8 id)
 {
-	/* The uABI guarantees an active writer is also amongst the read
+	/*
+	 * The uABI guarantees an active writer is also amongst the read
 	 * engines. This would be true if we accessed the activity tracking
 	 * under the lock, but as we perform the lookup of the object and
 	 * its activity locklessly we can not guarantee that the last_write
@@ -4023,16 +3862,19 @@ static __always_inline unsigned int __busy_write_id(unsigned int id)
 	 * last_read - hence we always set both read and write busy for
 	 * last_write.
 	 */
-	return id | __busy_read_flag(id);
+	if (id == (u8)I915_ENGINE_CLASS_INVALID)
+		return 0xffffffffu;
+
+	return (id + 1) | __busy_read_flag(id);
 }
 
 static __always_inline unsigned int
-__busy_set_if_active(const struct dma_fence *fence,
-		     unsigned int (*flag)(unsigned int id))
+__busy_set_if_active(const struct dma_fence *fence, u32 (*flag)(u8 id))
 {
-	struct i915_request *rq;
+	const struct i915_request *rq;
 
-	/* We have to check the current hw status of the fence as the uABI
+	/*
+	 * We have to check the current hw status of the fence as the uABI
 	 * guarantees forward progress. We could rely on the idle worker
 	 * to eventually flush us, but to minimise latency just ask the
 	 * hardware.
@@ -4043,11 +3885,13 @@ __busy_set_if_active(const struct dma_fence *fence,
 		return 0;
 
 	/* opencode to_request() in order to avoid const warnings */
-	rq = container_of(fence, struct i915_request, fence);
+	rq = container_of(fence, const struct i915_request, fence);
 	if (i915_request_completed(rq))
 		return 0;
 
-	return flag(rq->engine->uabi_id);
+	/* Beware type-expansion follies! */
+	BUILD_BUG_ON(!typecheck(u8, rq->engine->uabi_class));
+	return flag(rq->engine->uabi_class);
 }
 
 static __always_inline unsigned int
@@ -4081,7 +3925,8 @@ i915_gem_busy_ioctl(struct drm_device *dev, void *data,
 	if (!obj)
 		goto out;
 
-	/* A discrepancy here is that we do not report the status of
+	/*
+	 * A discrepancy here is that we do not report the status of
 	 * non-i915 fences, i.e. even though we may report the object as idle,
 	 * a call to set-domain may still stall waiting for foreign rendering.
 	 * This also means that wait-ioctl may report an object as busy,
@@ -4281,7 +4126,7 @@ i915_gem_object_create(struct drm_i915_private *dev_priv, u64 size)
 	if (overflows_type(size, obj->base.size))
 		return ERR_PTR(-E2BIG);
 
-	obj = i915_gem_object_alloc(dev_priv);
+	obj = i915_gem_object_alloc();
 	if (obj == NULL)
 		return ERR_PTR(-ENOMEM);
 
@@ -4414,7 +4259,7 @@ static void __i915_gem_free_objects(struct drm_i915_private *i915,
 		drm_gem_object_release(&obj->base);
 		i915_gem_info_remove_obj(i915, obj->base.size);
 
-		kfree(obj->bit_17);
+		bitmap_free(obj->bit_17);
 		i915_gem_object_free(obj);
 
 		GEM_BUG_ON(!atomic_read(&i915->mm.free_count));
@@ -4537,7 +4382,7 @@ void i915_gem_sanitize(struct drm_i915_private *i915)
 	GEM_TRACE("\n");
 
 	wakeref = intel_runtime_pm_get(i915);
-	intel_uncore_forcewake_get(i915, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(&i915->uncore, FORCEWAKE_ALL);
 
 	/*
 	 * As we have just resumed the machine and woken the device up from
@@ -4545,7 +4390,7 @@ void i915_gem_sanitize(struct drm_i915_private *i915)
 	 * back to defaults, recovering from whatever wedged state we left it
 	 * in and so worth trying to use the device once more.
 	 */
-	if (i915_terminally_wedged(&i915->gpu_error))
+	if (i915_terminally_wedged(i915))
 		i915_gem_unset_wedged(i915);
 
 	/*
@@ -4558,7 +4403,7 @@ void i915_gem_sanitize(struct drm_i915_private *i915)
 	 */
 	intel_engines_sanitize(i915, false);
 
-	intel_uncore_forcewake_put(i915, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&i915->uncore, FORCEWAKE_ALL);
 	intel_runtime_pm_put(i915, wakeref);
 
 	mutex_lock(&i915->drm.struct_mutex);
@@ -4566,15 +4411,13 @@ void i915_gem_sanitize(struct drm_i915_private *i915)
 	mutex_unlock(&i915->drm.struct_mutex);
 }
 
-int i915_gem_suspend(struct drm_i915_private *i915)
+void i915_gem_suspend(struct drm_i915_private *i915)
 {
 	intel_wakeref_t wakeref;
-	int ret;
 
 	GEM_TRACE("\n");
 
 	wakeref = intel_runtime_pm_get(i915);
-	intel_suspend_gt_powersave(i915);
 
 	flush_workqueue(i915->wq);
 
@@ -4589,22 +4432,7 @@ int i915_gem_suspend(struct drm_i915_private *i915)
 	 * state. Fortunately, the kernel_context is disposable and we do
 	 * not rely on its state.
 	 */
-	if (!i915_terminally_wedged(&i915->gpu_error)) {
-		ret = i915_gem_switch_to_kernel_context(i915);
-		if (ret)
-			goto err_unlock;
-
-		ret = i915_gem_wait_for_idle(i915,
-					     I915_WAIT_INTERRUPTIBLE |
-					     I915_WAIT_LOCKED |
-					     I915_WAIT_FOR_IDLE_BOOST,
-					     MAX_SCHEDULE_TIMEOUT);
-		if (ret && ret != -EIO)
-			goto err_unlock;
-
-		assert_kernel_context_is_current(i915);
-	}
-	i915_retire_requests(i915); /* ensure we flush after wedging */
+	switch_to_kernel_context_sync(i915, i915->gt.active_engines);
 
 	mutex_unlock(&i915->drm.struct_mutex);
 	i915_reset_flush(i915);
@@ -4617,23 +4445,15 @@ int i915_gem_suspend(struct drm_i915_private *i915)
 	 */
 	drain_delayed_work(&i915->gt.idle_work);
 
-	intel_uc_suspend(i915);
-
 	/*
 	 * Assert that we successfully flushed all the work and
 	 * reset the GPU back to its idle, low power state.
 	 */
-	WARN_ON(i915->gt.awake);
-	if (WARN_ON(!intel_engines_are_idle(i915)))
-		i915_gem_set_wedged(i915); /* no hope, discard everything */
+	GEM_BUG_ON(i915->gt.awake);
 
-	intel_runtime_pm_put(i915, wakeref);
-	return 0;
+	intel_uc_suspend(i915);
 
-err_unlock:
-	mutex_unlock(&i915->drm.struct_mutex);
 	intel_runtime_pm_put(i915, wakeref);
-	return ret;
 }
 
 void i915_gem_suspend_late(struct drm_i915_private *i915)
@@ -4683,7 +4503,7 @@ void i915_gem_resume(struct drm_i915_private *i915)
 	WARN_ON(i915->gt.awake);
 
 	mutex_lock(&i915->drm.struct_mutex);
-	intel_uncore_forcewake_get(i915, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(&i915->uncore, FORCEWAKE_ALL);
 
 	i915_gem_restore_gtt_mappings(i915);
 	i915_gem_restore_fences(i915);
@@ -4693,7 +4513,7 @@ void i915_gem_resume(struct drm_i915_private *i915)
 	 * guarantee that the context image is complete. So let's just reset
 	 * it and start again.
 	 */
-	i915->gt.resume(i915);
+	intel_gt_resume(i915);
 
 	if (i915_gem_init_hw(i915))
 		goto err_wedged;
@@ -4701,17 +4521,18 @@ void i915_gem_resume(struct drm_i915_private *i915)
 	intel_uc_resume(i915);
 
 	/* Always reload a context for powersaving. */
-	if (i915_gem_switch_to_kernel_context(i915))
+	if (!load_power_context(i915))
 		goto err_wedged;
 
 out_unlock:
-	intel_uncore_forcewake_put(i915, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&i915->uncore, FORCEWAKE_ALL);
 	mutex_unlock(&i915->drm.struct_mutex);
 	return;
 
 err_wedged:
-	if (!i915_terminally_wedged(&i915->gpu_error)) {
-		DRM_ERROR("failed to re-initialize GPU, declaring wedged!\n");
+	if (!i915_reset_failed(i915)) {
+		dev_err(i915->drm.dev,
+			"Failed to re-initialize GPU, declaring it wedged!\n");
 		i915_gem_set_wedged(i915);
 	}
 	goto out_unlock;
@@ -4781,6 +4602,8 @@ static int __i915_gem_restart_engines(void *data)
 		}
 	}
 
+	intel_engines_set_scheduler_caps(i915);
+
 	return 0;
 }
 
@@ -4791,7 +4614,7 @@ int i915_gem_init_hw(struct drm_i915_private *dev_priv)
 	dev_priv->gt.last_init_time = ktime_get();
 
 	/* Double layer security blanket, see i915_gem_init() */
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL);
 
 	if (HAS_EDRAM(dev_priv) && INTEL_GEN(dev_priv) < 9)
 		I915_WRITE(HSW_IDICR, I915_READ(HSW_IDICR) | IDIHASHMSK(0xf));
@@ -4816,10 +4639,9 @@ int i915_gem_init_hw(struct drm_i915_private *dev_priv)
 	init_unused_rings(dev_priv);
 
 	BUG_ON(!dev_priv->kernel_context);
-	if (i915_terminally_wedged(&dev_priv->gpu_error)) {
-		ret = -EIO;
+	ret = i915_terminally_wedged(dev_priv);
+	if (ret)
 		goto out;
-	}
 
 	ret = i915_ppgtt_init_hw(dev_priv);
 	if (ret) {
@@ -4847,14 +4669,14 @@ int i915_gem_init_hw(struct drm_i915_private *dev_priv)
 	if (ret)
 		goto cleanup_uc;
 
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 
 	return 0;
 
 cleanup_uc:
 	intel_uc_fini_hw(dev_priv);
 out:
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 
 	return ret;
 }
@@ -4864,7 +4686,7 @@ static int __intel_engines_record_defaults(struct drm_i915_private *i915)
 	struct i915_gem_context *ctx;
 	struct intel_engine_cs *engine;
 	enum intel_engine_id id;
-	int err;
+	int err = 0;
 
 	/*
 	 * As we reset the gpu during very early sanitisation, the current
@@ -4897,36 +4719,27 @@ static int __intel_engines_record_defaults(struct drm_i915_private *i915)
 			goto err_active;
 	}
 
-	err = i915_gem_switch_to_kernel_context(i915);
-	if (err)
-		goto err_active;
-
-	if (i915_gem_wait_for_idle(i915, I915_WAIT_LOCKED, HZ / 5)) {
-		i915_gem_set_wedged(i915);
-		err = -EIO; /* Caller will declare us wedged */
+	/* Flush the default context image to memory, and enable powersaving. */
+	if (!load_power_context(i915)) {
+		err = -EIO;
 		goto err_active;
 	}
 
-	assert_kernel_context_is_current(i915);
-
-	/*
-	 * Immediately park the GPU so that we enable powersaving and
-	 * treat it as idle. The next time we issue a request, we will
-	 * unpark and start using the engine->pinned_default_state, otherwise
-	 * it is in limbo and an early reset may fail.
-	 */
-	__i915_gem_park(i915);
-
 	for_each_engine(engine, i915, id) {
+		struct intel_context *ce;
 		struct i915_vma *state;
 		void *vaddr;
 
-		GEM_BUG_ON(to_intel_context(ctx, engine)->pin_count);
+		ce = intel_context_lookup(ctx, engine);
+		if (!ce)
+			continue;
 
-		state = to_intel_context(ctx, engine)->state;
+		state = ce->state;
 		if (!state)
 			continue;
 
+		GEM_BUG_ON(intel_context_is_pinned(ce));
+
 		/*
 		 * As we will hold a reference to the logical state, it will
 		 * not be torn down with the context, and importantly the
@@ -4944,6 +4757,8 @@ static int __intel_engines_record_defaults(struct drm_i915_private *i915)
 			goto err_active;
 
 		engine->default_state = i915_gem_object_get(state->obj);
+		i915_gem_object_set_cache_coherency(engine->default_state,
+						    I915_CACHE_LLC);
 
 		/* Check we can acquire the image of the context state */
 		vaddr = i915_gem_object_pin_map(engine->default_state,
@@ -4982,19 +4797,10 @@ out_ctx:
 err_active:
 	/*
 	 * If we have to abandon now, we expect the engines to be idle
-	 * and ready to be torn-down. First try to flush any remaining
-	 * request, ensure we are pointing at the kernel context and
-	 * then remove it.
+	 * and ready to be torn-down. The quickest way we can accomplish
+	 * this is by declaring ourselves wedged.
 	 */
-	if (WARN_ON(i915_gem_switch_to_kernel_context(i915)))
-		goto out_ctx;
-
-	if (WARN_ON(i915_gem_wait_for_idle(i915,
-					   I915_WAIT_LOCKED,
-					   MAX_SCHEDULE_TIMEOUT)))
-		goto out_ctx;
-
-	i915_gem_contexts_lost(i915);
+	i915_gem_set_wedged(i915);
 	goto out_ctx;
 }
 
@@ -5047,13 +4853,10 @@ int i915_gem_init(struct drm_i915_private *dev_priv)
 
 	dev_priv->mm.unordered_timeline = dma_fence_context_alloc(1);
 
-	if (HAS_LOGICAL_RING_CONTEXTS(dev_priv)) {
-		dev_priv->gt.resume = intel_lr_context_resume;
+	if (HAS_LOGICAL_RING_CONTEXTS(dev_priv))
 		dev_priv->gt.cleanup_engine = intel_logical_ring_cleanup;
-	} else {
-		dev_priv->gt.resume = intel_legacy_submission_resume;
+	else
 		dev_priv->gt.cleanup_engine = intel_engine_cleanup;
-	}
 
 	i915_timelines_init(dev_priv);
 
@@ -5076,7 +4879,7 @@ int i915_gem_init(struct drm_i915_private *dev_priv)
 	 * just magically go away.
 	 */
 	mutex_lock(&dev_priv->drm.struct_mutex);
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL);
 
 	ret = i915_gem_init_ggtt(dev_priv);
 	if (ret) {
@@ -5138,7 +4941,7 @@ int i915_gem_init(struct drm_i915_private *dev_priv)
 		goto err_init_hw;
 	}
 
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 	mutex_unlock(&dev_priv->drm.struct_mutex);
 
 	return 0;
@@ -5152,7 +4955,7 @@ int i915_gem_init(struct drm_i915_private *dev_priv)
 err_init_hw:
 	mutex_unlock(&dev_priv->drm.struct_mutex);
 
-	WARN_ON(i915_gem_suspend(dev_priv));
+	i915_gem_suspend(dev_priv);
 	i915_gem_suspend_late(dev_priv);
 
 	i915_gem_drain_workqueue(dev_priv);
@@ -5173,7 +4976,7 @@ err_scratch:
 	i915_gem_fini_scratch(dev_priv);
 err_ggtt:
 err_unlock:
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 	mutex_unlock(&dev_priv->drm.struct_mutex);
 
 err_uc_misc:
@@ -5192,7 +4995,7 @@ err_uc_misc:
 		 * wedged. But we only want to do this where the GPU is angry,
 		 * for all other failure, such as an allocation failure, bail.
 		 */
-		if (!i915_terminally_wedged(&dev_priv->gpu_error)) {
+		if (!i915_reset_failed(dev_priv)) {
 			i915_load_error(dev_priv,
 					"Failed to initialize GPU, declaring it wedged!\n");
 			i915_gem_set_wedged(dev_priv);
@@ -5305,36 +5108,7 @@ static void i915_gem_init__mm(struct drm_i915_private *i915)
 
 int i915_gem_init_early(struct drm_i915_private *dev_priv)
 {
-	int err = -ENOMEM;
-
-	dev_priv->objects = KMEM_CACHE(drm_i915_gem_object, SLAB_HWCACHE_ALIGN);
-	if (!dev_priv->objects)
-		goto err_out;
-
-	dev_priv->vmas = KMEM_CACHE(i915_vma, SLAB_HWCACHE_ALIGN);
-	if (!dev_priv->vmas)
-		goto err_objects;
-
-	dev_priv->luts = KMEM_CACHE(i915_lut_handle, 0);
-	if (!dev_priv->luts)
-		goto err_vmas;
-
-	dev_priv->requests = KMEM_CACHE(i915_request,
-					SLAB_HWCACHE_ALIGN |
-					SLAB_RECLAIM_ACCOUNT |
-					SLAB_TYPESAFE_BY_RCU);
-	if (!dev_priv->requests)
-		goto err_luts;
-
-	dev_priv->dependencies = KMEM_CACHE(i915_dependency,
-					    SLAB_HWCACHE_ALIGN |
-					    SLAB_RECLAIM_ACCOUNT);
-	if (!dev_priv->dependencies)
-		goto err_requests;
-
-	dev_priv->priorities = KMEM_CACHE(i915_priolist, SLAB_HWCACHE_ALIGN);
-	if (!dev_priv->priorities)
-		goto err_dependencies;
+	int err;
 
 	INIT_LIST_HEAD(&dev_priv->gt.active_rings);
 	INIT_LIST_HEAD(&dev_priv->gt.closed_vma);
@@ -5348,6 +5122,7 @@ int i915_gem_init_early(struct drm_i915_private *dev_priv)
 	init_waitqueue_head(&dev_priv->gpu_error.wait_queue);
 	init_waitqueue_head(&dev_priv->gpu_error.reset_queue);
 	mutex_init(&dev_priv->gpu_error.wedge_mutex);
+	init_srcu_struct(&dev_priv->gpu_error.reset_backoff_srcu);
 
 	atomic_set(&dev_priv->mm.bsd_engine_dispatch_index, 0);
 
@@ -5358,19 +5133,6 @@ int i915_gem_init_early(struct drm_i915_private *dev_priv)
 		DRM_NOTE("Unable to create a private tmpfs mount, hugepage support will be disabled(%d).\n", err);
 
 	return 0;
-
-err_dependencies:
-	kmem_cache_destroy(dev_priv->dependencies);
-err_requests:
-	kmem_cache_destroy(dev_priv->requests);
-err_luts:
-	kmem_cache_destroy(dev_priv->luts);
-err_vmas:
-	kmem_cache_destroy(dev_priv->vmas);
-err_objects:
-	kmem_cache_destroy(dev_priv->objects);
-err_out:
-	return err;
 }
 
 void i915_gem_cleanup_early(struct drm_i915_private *dev_priv)
@@ -5380,15 +5142,7 @@ void i915_gem_cleanup_early(struct drm_i915_private *dev_priv)
 	GEM_BUG_ON(atomic_read(&dev_priv->mm.free_count));
 	WARN_ON(dev_priv->mm.object_count);
 
-	kmem_cache_destroy(dev_priv->priorities);
-	kmem_cache_destroy(dev_priv->dependencies);
-	kmem_cache_destroy(dev_priv->requests);
-	kmem_cache_destroy(dev_priv->luts);
-	kmem_cache_destroy(dev_priv->vmas);
-	kmem_cache_destroy(dev_priv->objects);
-
-	/* And ensure that our DESTROY_BY_RCU slabs are truly destroyed */
-	rcu_barrier();
+	cleanup_srcu_struct(&dev_priv->gpu_error.reset_backoff_srcu);
 
 	i915_gemfs_fini(dev_priv);
 }
diff --git a/drivers/gpu/drm/i915/i915_gem.h b/drivers/gpu/drm/i915/i915_gem.h
index b0e4b976880c..9074eb1e843f 100644
--- a/drivers/gpu/drm/i915/i915_gem.h
+++ b/drivers/gpu/drm/i915/i915_gem.h
@@ -73,14 +73,14 @@ struct drm_i915_private;
 #define GEM_TRACE_DUMP_ON(expr) BUILD_BUG_ON_INVALID(expr)
 #endif
 
-#define I915_NUM_ENGINES 8
+#define I915_GEM_IDLE_TIMEOUT (HZ / 5)
 
 void i915_gem_park(struct drm_i915_private *i915);
 void i915_gem_unpark(struct drm_i915_private *i915);
 
 static inline void __tasklet_disable_sync_once(struct tasklet_struct *t)
 {
-	if (atomic_inc_return(&t->count) == 1)
+	if (!atomic_fetch_inc(&t->count))
 		tasklet_unlock_wait(t);
 }
 
@@ -89,4 +89,9 @@ static inline bool __tasklet_is_enabled(const struct tasklet_struct *t)
 	return !atomic_read(&t->count);
 }
 
+static inline bool __tasklet_enable(struct tasklet_struct *t)
+{
+	return atomic_dec_and_test(&t->count);
+}
+
 #endif /* __I915_GEM_H__ */
diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
index 280813a4bf82..dd728b26b5aa 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -88,12 +88,32 @@
 #include <linux/log2.h>
 #include <drm/i915_drm.h>
 #include "i915_drv.h"
+#include "i915_globals.h"
 #include "i915_trace.h"
+#include "i915_user_extensions.h"
 #include "intel_lrc_reg.h"
 #include "intel_workarounds.h"
 
+#define I915_CONTEXT_CREATE_FLAGS_SINGLE_TIMELINE (1 << 1)
+#define I915_CONTEXT_PARAM_VM 0x9
+
 #define ALL_L3_SLICES(dev) (1 << NUM_L3_SLICES(dev)) - 1
 
+static struct i915_global_gem_context {
+	struct i915_global base;
+	struct kmem_cache *slab_luts;
+} global;
+
+struct i915_lut_handle *i915_lut_handle_alloc(void)
+{
+	return kmem_cache_alloc(global.slab_luts, GFP_KERNEL);
+}
+
+void i915_lut_handle_free(struct i915_lut_handle *lut)
+{
+	return kmem_cache_free(global.slab_luts, lut);
+}
+
 static void lut_close(struct i915_gem_context *ctx)
 {
 	struct i915_lut_handle *lut, *ln;
@@ -102,14 +122,17 @@ static void lut_close(struct i915_gem_context *ctx)
 
 	list_for_each_entry_safe(lut, ln, &ctx->handles_list, ctx_link) {
 		list_del(&lut->obj_link);
-		kmem_cache_free(ctx->i915->luts, lut);
+		i915_lut_handle_free(lut);
 	}
+	INIT_LIST_HEAD(&ctx->handles_list);
 
 	rcu_read_lock();
 	radix_tree_for_each_slot(slot, &ctx->handles_vma, &iter, 0) {
 		struct i915_vma *vma = rcu_dereference_raw(*slot);
 
 		radix_tree_iter_delete(&ctx->handles_vma, &iter, slot);
+
+		vma->open_count--;
 		__i915_gem_object_release_unless_active(vma->obj);
 	}
 	rcu_read_unlock();
@@ -206,25 +229,26 @@ static void release_hw_id(struct i915_gem_context *ctx)
 
 static void i915_gem_context_free(struct i915_gem_context *ctx)
 {
-	unsigned int n;
+	struct intel_context *it, *n;
 
 	lockdep_assert_held(&ctx->i915->drm.struct_mutex);
 	GEM_BUG_ON(!i915_gem_context_is_closed(ctx));
+	GEM_BUG_ON(!list_empty(&ctx->active_engines));
 
 	release_hw_id(ctx);
 	i915_ppgtt_put(ctx->ppgtt);
 
-	for (n = 0; n < ARRAY_SIZE(ctx->__engine); n++) {
-		struct intel_context *ce = &ctx->__engine[n];
+	rbtree_postorder_for_each_entry_safe(it, n, &ctx->hw_contexts, node)
+		intel_context_put(it);
 
-		if (ce->ops)
-			ce->ops->destroy(ce);
-	}
+	if (ctx->timeline)
+		i915_timeline_put(ctx->timeline);
 
 	kfree(ctx->name);
 	put_pid(ctx->pid);
 
 	list_del(&ctx->link);
+	mutex_destroy(&ctx->mutex);
 
 	kfree_rcu(ctx, rcu);
 }
@@ -291,8 +315,6 @@ static void context_close(struct i915_gem_context *ctx)
 	 * the ppgtt).
 	 */
 	lut_close(ctx);
-	if (ctx->ppgtt)
-		i915_ppgtt_close(&ctx->ppgtt->vm);
 
 	ctx->file_priv = ERR_PTR(-EBADF);
 	i915_gem_context_put(ctx);
@@ -307,7 +329,7 @@ static u32 default_desc_template(const struct drm_i915_private *i915,
 	desc = GEN8_CTX_VALID | GEN8_CTX_PRIVILEGE;
 
 	address_mode = INTEL_LEGACY_32B_CONTEXT;
-	if (ppgtt && i915_vm_is_48bit(&ppgtt->vm))
+	if (ppgtt && i915_vm_is_4lvl(&ppgtt->vm))
 		address_mode = INTEL_LEGACY_64B_CONTEXT;
 	desc |= address_mode << GEN8_CTX_ADDRESSING_MODE_SHIFT;
 
@@ -322,134 +344,115 @@ static u32 default_desc_template(const struct drm_i915_private *i915,
 	return desc;
 }
 
-static void intel_context_retire(struct i915_active_request *active,
-				 struct i915_request *rq)
-{
-	struct intel_context *ce =
-		container_of(active, typeof(*ce), active_tracker);
-
-	intel_context_unpin(ce);
-}
-
-void
-intel_context_init(struct intel_context *ce,
-		   struct i915_gem_context *ctx,
-		   struct intel_engine_cs *engine)
-{
-	ce->gem_context = ctx;
-
-	INIT_LIST_HEAD(&ce->signal_link);
-	INIT_LIST_HEAD(&ce->signals);
-
-	/* Use the whole device by default */
-	ce->sseu = intel_device_default_sseu(ctx->i915);
-
-	i915_active_request_init(&ce->active_tracker,
-				 NULL, intel_context_retire);
-}
-
 static struct i915_gem_context *
-__create_hw_context(struct drm_i915_private *dev_priv,
-		    struct drm_i915_file_private *file_priv)
+__create_context(struct drm_i915_private *dev_priv)
 {
 	struct i915_gem_context *ctx;
-	unsigned int n;
-	int ret;
+	int i;
 
 	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
-	if (ctx == NULL)
+	if (!ctx)
 		return ERR_PTR(-ENOMEM);
 
 	kref_init(&ctx->ref);
 	list_add_tail(&ctx->link, &dev_priv->contexts.list);
 	ctx->i915 = dev_priv;
 	ctx->sched.priority = I915_USER_PRIORITY(I915_PRIORITY_NORMAL);
+	INIT_LIST_HEAD(&ctx->active_engines);
+	mutex_init(&ctx->mutex);
 
-	for (n = 0; n < ARRAY_SIZE(ctx->__engine); n++)
-		intel_context_init(&ctx->__engine[n], ctx, dev_priv->engine[n]);
+	ctx->hw_contexts = RB_ROOT;
+	spin_lock_init(&ctx->hw_contexts_lock);
 
 	INIT_RADIX_TREE(&ctx->handles_vma, GFP_KERNEL);
 	INIT_LIST_HEAD(&ctx->handles_list);
 	INIT_LIST_HEAD(&ctx->hw_id_link);
 
-	/* Default context will never have a file_priv */
-	ret = DEFAULT_CONTEXT_HANDLE;
-	if (file_priv) {
-		ret = idr_alloc(&file_priv->context_idr, ctx,
-				DEFAULT_CONTEXT_HANDLE, 0, GFP_KERNEL);
-		if (ret < 0)
-			goto err_lut;
-	}
-	ctx->user_handle = ret;
-
-	ctx->file_priv = file_priv;
-	if (file_priv) {
-		ctx->pid = get_task_pid(current, PIDTYPE_PID);
-		ctx->name = kasprintf(GFP_KERNEL, "%s[%d]/%x",
-				      current->comm,
-				      pid_nr(ctx->pid),
-				      ctx->user_handle);
-		if (!ctx->name) {
-			ret = -ENOMEM;
-			goto err_pid;
-		}
-	}
-
 	/* NB: Mark all slices as needing a remap so that when the context first
 	 * loads it will restore whatever remap state already exists. If there
 	 * is no remap info, it will be a NOP. */
 	ctx->remap_slice = ALL_L3_SLICES(dev_priv);
 
 	i915_gem_context_set_bannable(ctx);
+	i915_gem_context_set_recoverable(ctx);
+
 	ctx->ring_size = 4 * PAGE_SIZE;
 	ctx->desc_template =
 		default_desc_template(dev_priv, dev_priv->mm.aliasing_ppgtt);
 
+	for (i = 0; i < ARRAY_SIZE(ctx->hang_timestamp); i++)
+		ctx->hang_timestamp[i] = jiffies - CONTEXT_FAST_HANG_JIFFIES;
+
 	return ctx;
+}
 
-err_pid:
-	put_pid(ctx->pid);
-	idr_remove(&file_priv->context_idr, ctx->user_handle);
-err_lut:
-	context_close(ctx);
-	return ERR_PTR(ret);
+static struct i915_hw_ppgtt *
+__set_ppgtt(struct i915_gem_context *ctx, struct i915_hw_ppgtt *ppgtt)
+{
+	struct i915_hw_ppgtt *old = ctx->ppgtt;
+
+	ctx->ppgtt = i915_ppgtt_get(ppgtt);
+	ctx->desc_template = default_desc_template(ctx->i915, ppgtt);
+
+	return old;
 }
 
-static void __destroy_hw_context(struct i915_gem_context *ctx,
-				 struct drm_i915_file_private *file_priv)
+static void __assign_ppgtt(struct i915_gem_context *ctx,
+			   struct i915_hw_ppgtt *ppgtt)
 {
-	idr_remove(&file_priv->context_idr, ctx->user_handle);
-	context_close(ctx);
+	if (ppgtt == ctx->ppgtt)
+		return;
+
+	ppgtt = __set_ppgtt(ctx, ppgtt);
+	if (ppgtt)
+		i915_ppgtt_put(ppgtt);
 }
 
 static struct i915_gem_context *
-i915_gem_create_context(struct drm_i915_private *dev_priv,
-			struct drm_i915_file_private *file_priv)
+i915_gem_create_context(struct drm_i915_private *dev_priv, unsigned int flags)
 {
 	struct i915_gem_context *ctx;
 
 	lockdep_assert_held(&dev_priv->drm.struct_mutex);
 
+	BUILD_BUG_ON(I915_CONTEXT_CREATE_FLAGS_SINGLE_TIMELINE &
+		     ~I915_CONTEXT_CREATE_FLAGS_UNKNOWN);
+	if (flags & I915_CONTEXT_CREATE_FLAGS_SINGLE_TIMELINE &&
+	    !HAS_EXECLISTS(dev_priv))
+		return ERR_PTR(-EINVAL);
+
 	/* Reap the most stale context */
 	contexts_free_first(dev_priv);
 
-	ctx = __create_hw_context(dev_priv, file_priv);
+	ctx = __create_context(dev_priv);
 	if (IS_ERR(ctx))
 		return ctx;
 
 	if (HAS_FULL_PPGTT(dev_priv)) {
 		struct i915_hw_ppgtt *ppgtt;
 
-		ppgtt = i915_ppgtt_create(dev_priv, file_priv);
+		ppgtt = i915_ppgtt_create(dev_priv);
 		if (IS_ERR(ppgtt)) {
 			DRM_DEBUG_DRIVER("PPGTT setup failed (%ld)\n",
 					 PTR_ERR(ppgtt));
-			__destroy_hw_context(ctx, file_priv);
+			context_close(ctx);
 			return ERR_CAST(ppgtt);
 		}
 
-		ctx->ppgtt = ppgtt;
-		ctx->desc_template = default_desc_template(dev_priv, ppgtt);
+		__assign_ppgtt(ctx, ppgtt);
+		i915_ppgtt_put(ppgtt);
+	}
+
+	if (flags & I915_CONTEXT_CREATE_FLAGS_SINGLE_TIMELINE) {
+		struct i915_timeline *timeline;
+
+		timeline = i915_timeline_create(dev_priv, NULL);
+		if (IS_ERR(timeline)) {
+			context_close(ctx);
+			return ERR_CAST(timeline);
+		}
+
+		ctx->timeline = timeline;
 	}
 
 	trace_i915_context_create(ctx);
@@ -480,10 +483,17 @@ i915_gem_context_create_gvt(struct drm_device *dev)
 	if (ret)
 		return ERR_PTR(ret);
 
-	ctx = i915_gem_create_context(to_i915(dev), NULL);
+	ctx = i915_gem_create_context(to_i915(dev), 0);
 	if (IS_ERR(ctx))
 		goto out;
 
+	ret = i915_gem_context_pin_hw_id(ctx);
+	if (ret) {
+		context_close(ctx);
+		ctx = ERR_PTR(ret);
+		goto out;
+	}
+
 	ctx->file_priv = ERR_PTR(-EBADF);
 	i915_gem_context_set_closed(ctx); /* not user accessible */
 	i915_gem_context_clear_bannable(ctx);
@@ -516,7 +526,7 @@ i915_gem_context_create_kernel(struct drm_i915_private *i915, int prio)
 	struct i915_gem_context *ctx;
 	int err;
 
-	ctx = i915_gem_create_context(i915, NULL);
+	ctx = i915_gem_create_context(i915, 0);
 	if (IS_ERR(ctx))
 		return ctx;
 
@@ -552,7 +562,7 @@ static void init_contexts(struct drm_i915_private *i915)
 
 static bool needs_preempt_context(struct drm_i915_private *i915)
 {
-	return HAS_LOGICAL_RING_PREEMPTION(i915);
+	return HAS_EXECLISTS(i915);
 }
 
 int i915_gem_contexts_init(struct drm_i915_private *dev_priv)
@@ -563,7 +573,7 @@ int i915_gem_contexts_init(struct drm_i915_private *dev_priv)
 	GEM_BUG_ON(dev_priv->kernel_context);
 	GEM_BUG_ON(dev_priv->preempt_context);
 
-	intel_engine_init_ctx_wa(dev_priv->engine[RCS]);
+	intel_engine_init_ctx_wa(dev_priv->engine[RCS0]);
 	init_contexts(dev_priv);
 
 	/* lowest priority; idle task */
@@ -624,31 +634,87 @@ void i915_gem_contexts_fini(struct drm_i915_private *i915)
 
 static int context_idr_cleanup(int id, void *p, void *data)
 {
-	struct i915_gem_context *ctx = p;
+	context_close(p);
+	return 0;
+}
 
-	context_close(ctx);
+static int vm_idr_cleanup(int id, void *p, void *data)
+{
+	i915_ppgtt_put(p);
 	return 0;
 }
 
+static int gem_context_register(struct i915_gem_context *ctx,
+				struct drm_i915_file_private *fpriv)
+{
+	int ret;
+
+	ctx->file_priv = fpriv;
+	if (ctx->ppgtt)
+		ctx->ppgtt->vm.file = fpriv;
+
+	ctx->pid = get_task_pid(current, PIDTYPE_PID);
+	ctx->name = kasprintf(GFP_KERNEL, "%s[%d]",
+			      current->comm, pid_nr(ctx->pid));
+	if (!ctx->name) {
+		ret = -ENOMEM;
+		goto err_pid;
+	}
+
+	/* And finally expose ourselves to userspace via the idr */
+	mutex_lock(&fpriv->context_idr_lock);
+	ret = idr_alloc(&fpriv->context_idr, ctx, 0, 0, GFP_KERNEL);
+	mutex_unlock(&fpriv->context_idr_lock);
+	if (ret >= 0)
+		goto out;
+
+	kfree(fetch_and_zero(&ctx->name));
+err_pid:
+	put_pid(fetch_and_zero(&ctx->pid));
+out:
+	return ret;
+}
+
 int i915_gem_context_open(struct drm_i915_private *i915,
 			  struct drm_file *file)
 {
 	struct drm_i915_file_private *file_priv = file->driver_priv;
 	struct i915_gem_context *ctx;
+	int err;
+
+	mutex_init(&file_priv->context_idr_lock);
+	mutex_init(&file_priv->vm_idr_lock);
 
 	idr_init(&file_priv->context_idr);
+	idr_init_base(&file_priv->vm_idr, 1);
 
 	mutex_lock(&i915->drm.struct_mutex);
-	ctx = i915_gem_create_context(i915, file_priv);
+	ctx = i915_gem_create_context(i915, 0);
 	mutex_unlock(&i915->drm.struct_mutex);
 	if (IS_ERR(ctx)) {
-		idr_destroy(&file_priv->context_idr);
-		return PTR_ERR(ctx);
+		err = PTR_ERR(ctx);
+		goto err;
 	}
 
+	err = gem_context_register(ctx, file_priv);
+	if (err < 0)
+		goto err_ctx;
+
 	GEM_BUG_ON(i915_gem_context_is_kernel(ctx));
+	GEM_BUG_ON(err > 0);
 
 	return 0;
+
+err_ctx:
+	mutex_lock(&i915->drm.struct_mutex);
+	context_close(ctx);
+	mutex_unlock(&i915->drm.struct_mutex);
+err:
+	idr_destroy(&file_priv->vm_idr);
+	idr_destroy(&file_priv->context_idr);
+	mutex_destroy(&file_priv->vm_idr_lock);
+	mutex_destroy(&file_priv->context_idr_lock);
+	return err;
 }
 
 void i915_gem_context_close(struct drm_file *file)
@@ -659,6 +725,100 @@ void i915_gem_context_close(struct drm_file *file)
 
 	idr_for_each(&file_priv->context_idr, context_idr_cleanup, NULL);
 	idr_destroy(&file_priv->context_idr);
+	mutex_destroy(&file_priv->context_idr_lock);
+
+	idr_for_each(&file_priv->vm_idr, vm_idr_cleanup, NULL);
+	idr_destroy(&file_priv->vm_idr);
+	mutex_destroy(&file_priv->vm_idr_lock);
+}
+
+int i915_gem_vm_create_ioctl(struct drm_device *dev, void *data,
+			     struct drm_file *file)
+{
+	struct drm_i915_private *i915 = to_i915(dev);
+	struct drm_i915_gem_vm_control *args = data;
+	struct drm_i915_file_private *file_priv = file->driver_priv;
+	struct i915_hw_ppgtt *ppgtt;
+	int err;
+
+	if (!HAS_FULL_PPGTT(i915))
+		return -ENODEV;
+
+	if (args->flags)
+		return -EINVAL;
+
+	ppgtt = i915_ppgtt_create(i915);
+	if (IS_ERR(ppgtt))
+		return PTR_ERR(ppgtt);
+
+	ppgtt->vm.file = file_priv;
+
+	if (args->extensions) {
+		err = i915_user_extensions(u64_to_user_ptr(args->extensions),
+					   NULL, 0,
+					   ppgtt);
+		if (err)
+			goto err_put;
+	}
+
+	err = mutex_lock_interruptible(&file_priv->vm_idr_lock);
+	if (err)
+		goto err_put;
+
+	err = idr_alloc(&file_priv->vm_idr, ppgtt, 0, 0, GFP_KERNEL);
+	if (err < 0)
+		goto err_unlock;
+
+	GEM_BUG_ON(err == 0); /* reserved for default/unassigned ppgtt */
+	ppgtt->user_handle = err;
+
+	mutex_unlock(&file_priv->vm_idr_lock);
+
+	args->vm_id = err;
+	return 0;
+
+err_unlock:
+	mutex_unlock(&file_priv->vm_idr_lock);
+err_put:
+	i915_ppgtt_put(ppgtt);
+	return err;
+}
+
+int i915_gem_vm_destroy_ioctl(struct drm_device *dev, void *data,
+			      struct drm_file *file)
+{
+	struct drm_i915_file_private *file_priv = file->driver_priv;
+	struct drm_i915_gem_vm_control *args = data;
+	struct i915_hw_ppgtt *ppgtt;
+	int err;
+	u32 id;
+
+	if (args->flags)
+		return -EINVAL;
+
+	if (args->extensions)
+		return -EINVAL;
+
+	id = args->vm_id;
+	if (!id)
+		return -ENOENT;
+
+	err = mutex_lock_interruptible(&file_priv->vm_idr_lock);
+	if (err)
+		return err;
+
+	ppgtt = idr_remove(&file_priv->vm_idr, id);
+	if (ppgtt) {
+		GEM_BUG_ON(ppgtt->user_handle != id);
+		ppgtt->user_handle = 0;
+	}
+
+	mutex_unlock(&file_priv->vm_idr_lock);
+	if (!ppgtt)
+		return -ENOENT;
+
+	i915_ppgtt_put(ppgtt);
+	return 0;
 }
 
 static struct i915_request *
@@ -671,10 +831,9 @@ last_request_on_engine(struct i915_timeline *timeline,
 
 	rq = i915_active_request_raw(&timeline->last_request,
 				     &engine->i915->drm.struct_mutex);
-	if (rq && rq->engine == engine) {
-		GEM_TRACE("last request for %s on engine %s: %llx:%llu\n",
-			  timeline->name, engine->name,
-			  rq->fence.context, rq->fence.seqno);
+	if (rq && rq->engine->mask & engine->mask) {
+		GEM_TRACE("last request on engine %s: %llx:%llu\n",
+			  engine->name, rq->fence.context, rq->fence.seqno);
 		GEM_BUG_ON(rq->timeline != timeline);
 		return rq;
 	}
@@ -682,81 +841,104 @@ last_request_on_engine(struct i915_timeline *timeline,
 	return NULL;
 }
 
-static bool engine_has_kernel_context_barrier(struct intel_engine_cs *engine)
+struct context_barrier_task {
+	struct i915_active base;
+	void (*task)(void *data);
+	void *data;
+};
+
+static void cb_retire(struct i915_active *base)
+{
+	struct context_barrier_task *cb = container_of(base, typeof(*cb), base);
+
+	if (cb->task)
+		cb->task(cb->data);
+
+	i915_active_fini(&cb->base);
+	kfree(cb);
+}
+
+I915_SELFTEST_DECLARE(static intel_engine_mask_t context_barrier_inject_fault);
+static int context_barrier_task(struct i915_gem_context *ctx,
+				intel_engine_mask_t engines,
+				int (*emit)(struct i915_request *rq, void *data),
+				void (*task)(void *data),
+				void *data)
 {
-	struct drm_i915_private *i915 = engine->i915;
-	const struct intel_context * const ce =
-		to_intel_context(i915->kernel_context, engine);
-	struct i915_timeline *barrier = ce->ring->timeline;
-	struct intel_ring *ring;
-	bool any_active = false;
+	struct drm_i915_private *i915 = ctx->i915;
+	struct context_barrier_task *cb;
+	struct intel_context *ce, *next;
+	intel_wakeref_t wakeref;
+	int err = 0;
 
 	lockdep_assert_held(&i915->drm.struct_mutex);
-	list_for_each_entry(ring, &i915->gt.active_rings, active_link) {
-		struct i915_request *rq;
+	GEM_BUG_ON(!task);
 
-		rq = last_request_on_engine(ring->timeline, engine);
-		if (!rq)
-			continue;
+	cb = kmalloc(sizeof(*cb), GFP_KERNEL);
+	if (!cb)
+		return -ENOMEM;
 
-		any_active = true;
+	i915_active_init(i915, &cb->base, cb_retire);
+	i915_active_acquire(&cb->base);
 
-		if (rq->hw_context == ce)
+	wakeref = intel_runtime_pm_get(i915);
+	rbtree_postorder_for_each_entry_safe(ce, next, &ctx->hw_contexts, node) {
+		struct intel_engine_cs *engine = ce->engine;
+		struct i915_request *rq;
+
+		if (!(engine->mask & engines))
 			continue;
 
-		/*
-		 * Was this request submitted after the previous
-		 * switch-to-kernel-context?
-		 */
-		if (!i915_timeline_sync_is_later(barrier, &rq->fence)) {
-			GEM_TRACE("%s needs barrier for %llx:%lld\n",
-				  ring->timeline->name,
-				  rq->fence.context,
-				  rq->fence.seqno);
-			return false;
+		if (I915_SELFTEST_ONLY(context_barrier_inject_fault &
+				       engine->mask)) {
+			err = -ENXIO;
+			break;
+		}
+
+		rq = i915_request_alloc(engine, ctx);
+		if (IS_ERR(rq)) {
+			err = PTR_ERR(rq);
+			break;
 		}
 
-		GEM_TRACE("%s has barrier after %llx:%lld\n",
-			  ring->timeline->name,
-			  rq->fence.context,
-			  rq->fence.seqno);
+		err = 0;
+		if (emit)
+			err = emit(rq, data);
+		if (err == 0)
+			err = i915_active_ref(&cb->base, rq->fence.context, rq);
+
+		i915_request_add(rq);
+		if (err)
+			break;
 	}
+	intel_runtime_pm_put(i915, wakeref);
 
-	/*
-	 * If any other timeline was still active and behind the last barrier,
-	 * then our last switch-to-kernel-context must still be queued and
-	 * will run last (leaving the engine in the kernel context when it
-	 * eventually idles).
-	 */
-	if (any_active)
-		return true;
+	cb->task = err ? NULL : task; /* caller needs to unwind instead */
+	cb->data = data;
 
-	/* The engine is idle; check that it is idling in the kernel context. */
-	return engine->last_retired_context == ce;
+	i915_active_release(&cb->base);
+
+	return err;
 }
 
-int i915_gem_switch_to_kernel_context(struct drm_i915_private *i915)
+int i915_gem_switch_to_kernel_context(struct drm_i915_private *i915,
+				      intel_engine_mask_t mask)
 {
 	struct intel_engine_cs *engine;
-	enum intel_engine_id id;
 
 	GEM_TRACE("awake?=%s\n", yesno(i915->gt.awake));
 
 	lockdep_assert_held(&i915->drm.struct_mutex);
 	GEM_BUG_ON(!i915->kernel_context);
 
-	i915_retire_requests(i915);
+	/* Inoperable, so presume the GPU is safely pointing into the void! */
+	if (i915_terminally_wedged(i915))
+		return 0;
 
-	for_each_engine(engine, i915, id) {
+	for_each_engine_masked(engine, i915, mask, mask) {
 		struct intel_ring *ring;
 		struct i915_request *rq;
 
-		GEM_BUG_ON(!to_intel_context(i915->kernel_context, engine));
-		if (engine_has_kernel_context_barrier(engine))
-			continue;
-
-		GEM_TRACE("emit barrier on %s\n", engine->name);
-
 		rq = i915_request_alloc(engine, i915->kernel_context);
 		if (IS_ERR(rq))
 			return PTR_ERR(rq);
@@ -779,7 +961,6 @@ int i915_gem_switch_to_kernel_context(struct drm_i915_private *i915)
 			i915_sw_fence_await_sw_fence_gfp(&rq->submit,
 							 &prev->submit,
 							 I915_FENCE_GFP);
-			i915_timeline_sync_set(rq->timeline, &prev->fence);
 		}
 
 		i915_request_add(rq);
@@ -788,183 +969,173 @@ int i915_gem_switch_to_kernel_context(struct drm_i915_private *i915)
 	return 0;
 }
 
-static bool client_is_banned(struct drm_i915_file_private *file_priv)
+static int get_ppgtt(struct drm_i915_file_private *file_priv,
+		     struct i915_gem_context *ctx,
+		     struct drm_i915_gem_context_param *args)
 {
-	return atomic_read(&file_priv->ban_score) >= I915_CLIENT_SCORE_BANNED;
-}
-
-int i915_gem_context_create_ioctl(struct drm_device *dev, void *data,
-				  struct drm_file *file)
-{
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct drm_i915_gem_context_create *args = data;
-	struct drm_i915_file_private *file_priv = file->driver_priv;
-	struct i915_gem_context *ctx;
+	struct i915_hw_ppgtt *ppgtt;
 	int ret;
 
-	if (!DRIVER_CAPS(dev_priv)->has_logical_contexts)
+	return -EINVAL; /* nothing to see here; please move along */
+
+	if (!ctx->ppgtt)
 		return -ENODEV;
 
-	if (args->pad != 0)
-		return -EINVAL;
+	/* XXX rcu acquire? */
+	ret = mutex_lock_interruptible(&ctx->i915->drm.struct_mutex);
+	if (ret)
+		return ret;
 
-	if (client_is_banned(file_priv)) {
-		DRM_DEBUG("client %s[%d] banned from creating ctx\n",
-			  current->comm,
-			  pid_nr(get_task_pid(current, PIDTYPE_PID)));
+	ppgtt = i915_ppgtt_get(ctx->ppgtt);
+	mutex_unlock(&ctx->i915->drm.struct_mutex);
 
-		return -EIO;
+	ret = mutex_lock_interruptible(&file_priv->vm_idr_lock);
+	if (ret)
+		goto err_put;
+
+	if (!ppgtt->user_handle) {
+		ret = idr_alloc(&file_priv->vm_idr, ppgtt, 0, 0, GFP_KERNEL);
+		GEM_BUG_ON(!ret);
+		if (ret < 0)
+			goto err_unlock;
+
+		ppgtt->user_handle = ret;
+		i915_ppgtt_get(ppgtt);
 	}
 
-	ret = i915_mutex_lock_interruptible(dev);
-	if (ret)
-		return ret;
+	args->size = 0;
+	args->value = ppgtt->user_handle;
 
-	ctx = i915_gem_create_context(dev_priv, file_priv);
-	mutex_unlock(&dev->struct_mutex);
-	if (IS_ERR(ctx))
-		return PTR_ERR(ctx);
+	ret = 0;
+err_unlock:
+	mutex_unlock(&file_priv->vm_idr_lock);
+err_put:
+	i915_ppgtt_put(ppgtt);
+	return ret;
+}
 
-	GEM_BUG_ON(i915_gem_context_is_kernel(ctx));
+static void set_ppgtt_barrier(void *data)
+{
+	struct i915_hw_ppgtt *old = data;
 
-	args->ctx_id = ctx->user_handle;
-	DRM_DEBUG("HW context %d created\n", args->ctx_id);
+	if (INTEL_GEN(old->vm.i915) < 8)
+		gen6_ppgtt_unpin_all(old);
 
-	return 0;
+	i915_ppgtt_put(old);
 }
 
-int i915_gem_context_destroy_ioctl(struct drm_device *dev, void *data,
-				   struct drm_file *file)
+static int emit_ppgtt_update(struct i915_request *rq, void *data)
 {
-	struct drm_i915_gem_context_destroy *args = data;
-	struct drm_i915_file_private *file_priv = file->driver_priv;
-	struct i915_gem_context *ctx;
-	int ret;
+	struct i915_hw_ppgtt *ppgtt = rq->gem_context->ppgtt;
+	struct intel_engine_cs *engine = rq->engine;
+	u32 base = engine->mmio_base;
+	u32 *cs;
+	int i;
 
-	if (args->pad != 0)
-		return -EINVAL;
+	if (i915_vm_is_4lvl(&ppgtt->vm)) {
+		const dma_addr_t pd_daddr = px_dma(&ppgtt->pml4);
 
-	if (args->ctx_id == DEFAULT_CONTEXT_HANDLE)
-		return -ENOENT;
+		cs = intel_ring_begin(rq, 6);
+		if (IS_ERR(cs))
+			return PTR_ERR(cs);
 
-	ctx = i915_gem_context_lookup(file_priv, args->ctx_id);
-	if (!ctx)
-		return -ENOENT;
+		*cs++ = MI_LOAD_REGISTER_IMM(2);
 
-	ret = mutex_lock_interruptible(&dev->struct_mutex);
-	if (ret)
-		goto out;
+		*cs++ = i915_mmio_reg_offset(GEN8_RING_PDP_UDW(base, 0));
+		*cs++ = upper_32_bits(pd_daddr);
+		*cs++ = i915_mmio_reg_offset(GEN8_RING_PDP_LDW(base, 0));
+		*cs++ = lower_32_bits(pd_daddr);
 
-	__destroy_hw_context(ctx, file_priv);
-	mutex_unlock(&dev->struct_mutex);
+		*cs++ = MI_NOOP;
+		intel_ring_advance(rq, cs);
+	} else if (HAS_LOGICAL_RING_CONTEXTS(engine->i915)) {
+		cs = intel_ring_begin(rq, 4 * GEN8_3LVL_PDPES + 2);
+		if (IS_ERR(cs))
+			return PTR_ERR(cs);
+
+		*cs++ = MI_LOAD_REGISTER_IMM(2 * GEN8_3LVL_PDPES);
+		for (i = GEN8_3LVL_PDPES; i--; ) {
+			const dma_addr_t pd_daddr = i915_page_dir_dma_addr(ppgtt, i);
+
+			*cs++ = i915_mmio_reg_offset(GEN8_RING_PDP_UDW(base, i));
+			*cs++ = upper_32_bits(pd_daddr);
+			*cs++ = i915_mmio_reg_offset(GEN8_RING_PDP_LDW(base, i));
+			*cs++ = lower_32_bits(pd_daddr);
+		}
+		*cs++ = MI_NOOP;
+		intel_ring_advance(rq, cs);
+	} else {
+		/* ppGTT is not part of the legacy context image */
+		gen6_ppgtt_pin(ppgtt);
+	}
 
-out:
-	i915_gem_context_put(ctx);
 	return 0;
 }
 
-static int get_sseu(struct i915_gem_context *ctx,
-		    struct drm_i915_gem_context_param *args)
+static int set_ppgtt(struct drm_i915_file_private *file_priv,
+		     struct i915_gem_context *ctx,
+		     struct drm_i915_gem_context_param *args)
 {
-	struct drm_i915_gem_context_param_sseu user_sseu;
-	struct intel_engine_cs *engine;
-	struct intel_context *ce;
-	int ret;
-
-	if (args->size == 0)
-		goto out;
-	else if (args->size < sizeof(user_sseu))
-		return -EINVAL;
-
-	if (copy_from_user(&user_sseu, u64_to_user_ptr(args->value),
-			   sizeof(user_sseu)))
-		return -EFAULT;
+	struct i915_hw_ppgtt *ppgtt, *old;
+	int err;
 
-	if (user_sseu.flags || user_sseu.rsvd)
-		return -EINVAL;
+	return -EINVAL; /* nothing to see here; please move along */
 
-	engine = intel_engine_lookup_user(ctx->i915,
-					  user_sseu.engine_class,
-					  user_sseu.engine_instance);
-	if (!engine)
+	if (args->size)
 		return -EINVAL;
 
-	/* Only use for mutex here is to serialize get_param and set_param. */
-	ret = mutex_lock_interruptible(&ctx->i915->drm.struct_mutex);
-	if (ret)
-		return ret;
-
-	ce = to_intel_context(ctx, engine);
-
-	user_sseu.slice_mask = ce->sseu.slice_mask;
-	user_sseu.subslice_mask = ce->sseu.subslice_mask;
-	user_sseu.min_eus_per_subslice = ce->sseu.min_eus_per_subslice;
-	user_sseu.max_eus_per_subslice = ce->sseu.max_eus_per_subslice;
+	if (!ctx->ppgtt)
+		return -ENODEV;
 
-	mutex_unlock(&ctx->i915->drm.struct_mutex);
+	if (upper_32_bits(args->value))
+		return -ENOENT;
 
-	if (copy_to_user(u64_to_user_ptr(args->value), &user_sseu,
-			 sizeof(user_sseu)))
-		return -EFAULT;
+	err = mutex_lock_interruptible(&file_priv->vm_idr_lock);
+	if (err)
+		return err;
 
-out:
-	args->size = sizeof(user_sseu);
+	ppgtt = idr_find(&file_priv->vm_idr, args->value);
+	if (ppgtt) {
+		GEM_BUG_ON(ppgtt->user_handle != args->value);
+		i915_ppgtt_get(ppgtt);
+	}
+	mutex_unlock(&file_priv->vm_idr_lock);
+	if (!ppgtt)
+		return -ENOENT;
 
-	return 0;
-}
+	err = mutex_lock_interruptible(&ctx->i915->drm.struct_mutex);
+	if (err)
+		goto out;
 
-int i915_gem_context_getparam_ioctl(struct drm_device *dev, void *data,
-				    struct drm_file *file)
-{
-	struct drm_i915_file_private *file_priv = file->driver_priv;
-	struct drm_i915_gem_context_param *args = data;
-	struct i915_gem_context *ctx;
-	int ret = 0;
+	if (ppgtt == ctx->ppgtt)
+		goto unlock;
 
-	ctx = i915_gem_context_lookup(file_priv, args->ctx_id);
-	if (!ctx)
-		return -ENOENT;
+	/* Teardown the existing obj:vma cache, it will have to be rebuilt. */
+	lut_close(ctx);
 
-	switch (args->param) {
-	case I915_CONTEXT_PARAM_BAN_PERIOD:
-		ret = -EINVAL;
-		break;
-	case I915_CONTEXT_PARAM_NO_ZEROMAP:
-		args->size = 0;
-		args->value = test_bit(UCONTEXT_NO_ZEROMAP, &ctx->user_flags);
-		break;
-	case I915_CONTEXT_PARAM_GTT_SIZE:
-		args->size = 0;
+	old = __set_ppgtt(ctx, ppgtt);
 
-		if (ctx->ppgtt)
-			args->value = ctx->ppgtt->vm.total;
-		else if (to_i915(dev)->mm.aliasing_ppgtt)
-			args->value = to_i915(dev)->mm.aliasing_ppgtt->vm.total;
-		else
-			args->value = to_i915(dev)->ggtt.vm.total;
-		break;
-	case I915_CONTEXT_PARAM_NO_ERROR_CAPTURE:
-		args->size = 0;
-		args->value = i915_gem_context_no_error_capture(ctx);
-		break;
-	case I915_CONTEXT_PARAM_BANNABLE:
-		args->size = 0;
-		args->value = i915_gem_context_is_bannable(ctx);
-		break;
-	case I915_CONTEXT_PARAM_PRIORITY:
-		args->size = 0;
-		args->value = ctx->sched.priority >> I915_USER_PRIORITY_SHIFT;
-		break;
-	case I915_CONTEXT_PARAM_SSEU:
-		ret = get_sseu(ctx, args);
-		break;
-	default:
-		ret = -EINVAL;
-		break;
+	/*
+	 * We need to flush any requests using the current ppgtt before
+	 * we release it as the requests do not hold a reference themselves,
+	 * only indirectly through the context.
+	 */
+	err = context_barrier_task(ctx, ALL_ENGINES,
+				   emit_ppgtt_update,
+				   set_ppgtt_barrier,
+				   old);
+	if (err) {
+		ctx->ppgtt = old;
+		ctx->desc_template = default_desc_template(ctx->i915, old);
+		i915_ppgtt_put(ppgtt);
 	}
 
-	i915_gem_context_put(ctx);
-	return ret;
+unlock:
+	mutex_unlock(&ctx->i915->drm.struct_mutex);
+
+out:
+	i915_ppgtt_put(ppgtt);
+	return err;
 }
 
 static int gen8_emit_rpcs_config(struct i915_request *rq,
@@ -993,39 +1164,35 @@ static int gen8_emit_rpcs_config(struct i915_request *rq,
 }
 
 static int
-gen8_modify_rpcs_gpu(struct intel_context *ce,
-		     struct intel_engine_cs *engine,
-		     struct intel_sseu sseu)
+gen8_modify_rpcs(struct intel_context *ce, struct intel_sseu sseu)
 {
-	struct drm_i915_private *i915 = engine->i915;
-	struct i915_request *rq, *prev;
+	struct drm_i915_private *i915 = ce->engine->i915;
+	struct i915_request *rq;
 	intel_wakeref_t wakeref;
 	int ret;
 
-	GEM_BUG_ON(!ce->pin_count);
+	lockdep_assert_held(&ce->pin_mutex);
 
-	lockdep_assert_held(&i915->drm.struct_mutex);
+	/*
+	 * If the context is not idle, we have to submit an ordered request to
+	 * modify its context image via the kernel context (writing to our own
+	 * image, or into the registers directory, does not stick). Pristine
+	 * and idle contexts will be configured on pinning.
+	 */
+	if (!intel_context_is_pinned(ce))
+		return 0;
 
 	/* Submitting requests etc needs the hw awake. */
 	wakeref = intel_runtime_pm_get(i915);
 
-	rq = i915_request_alloc(engine, i915->kernel_context);
+	rq = i915_request_alloc(ce->engine, i915->kernel_context);
 	if (IS_ERR(rq)) {
 		ret = PTR_ERR(rq);
 		goto out_put;
 	}
 
 	/* Queue this switch after all other activity by this context. */
-	prev = i915_active_request_raw(&ce->ring->timeline->last_request,
-				       &i915->drm.struct_mutex);
-	if (prev && !i915_request_completed(prev)) {
-		ret = i915_request_await_dma_fence(rq, &prev->fence);
-		if (ret < 0)
-			goto out_add;
-	}
-
-	/* Order all following requests to be after. */
-	ret = i915_timeline_set_barrier(ce->ring->timeline, rq);
+	ret = i915_active_request_set(&ce->ring->timeline->last_request, rq);
 	if (ret)
 		goto out_add;
 
@@ -1057,27 +1224,26 @@ __i915_gem_context_reconfigure_sseu(struct i915_gem_context *ctx,
 				    struct intel_engine_cs *engine,
 				    struct intel_sseu sseu)
 {
-	struct intel_context *ce = to_intel_context(ctx, engine);
+	struct intel_context *ce;
 	int ret = 0;
 
 	GEM_BUG_ON(INTEL_GEN(ctx->i915) < 8);
-	GEM_BUG_ON(engine->id != RCS);
+	GEM_BUG_ON(engine->id != RCS0);
+
+	ce = intel_context_pin_lock(ctx, engine);
+	if (IS_ERR(ce))
+		return PTR_ERR(ce);
 
 	/* Nothing to do if unmodified. */
 	if (!memcmp(&ce->sseu, &sseu, sizeof(sseu)))
-		return 0;
-
-	/*
-	 * If context is not idle we have to submit an ordered request to modify
-	 * its context image via the kernel context. Pristine and idle contexts
-	 * will be configured on pinning.
-	 */
-	if (ce->pin_count)
-		ret = gen8_modify_rpcs_gpu(ce, engine, sseu);
+		goto unlock;
 
+	ret = gen8_modify_rpcs(ce, sseu);
 	if (!ret)
 		ce->sseu = sseu;
 
+unlock:
+	intel_context_pin_unlock(ce);
 	return ret;
 }
 
@@ -1220,8 +1386,8 @@ static int set_sseu(struct i915_gem_context *ctx,
 		return -EINVAL;
 
 	engine = intel_engine_lookup_user(i915,
-					  user_sseu.engine_class,
-					  user_sseu.engine_instance);
+					  user_sseu.engine.engine_class,
+					  user_sseu.engine.engine_instance);
 	if (!engine)
 		return -EINVAL;
 
@@ -1242,22 +1408,13 @@ static int set_sseu(struct i915_gem_context *ctx,
 	return 0;
 }
 
-int i915_gem_context_setparam_ioctl(struct drm_device *dev, void *data,
-				    struct drm_file *file)
+static int ctx_setparam(struct drm_i915_file_private *fpriv,
+			struct i915_gem_context *ctx,
+			struct drm_i915_gem_context_param *args)
 {
-	struct drm_i915_file_private *file_priv = file->driver_priv;
-	struct drm_i915_gem_context_param *args = data;
-	struct i915_gem_context *ctx;
 	int ret = 0;
 
-	ctx = i915_gem_context_lookup(file_priv, args->ctx_id);
-	if (!ctx)
-		return -ENOENT;
-
 	switch (args->param) {
-	case I915_CONTEXT_PARAM_BAN_PERIOD:
-		ret = -EINVAL;
-		break;
 	case I915_CONTEXT_PARAM_NO_ZEROMAP:
 		if (args->size)
 			ret = -EINVAL;
@@ -1266,6 +1423,7 @@ int i915_gem_context_setparam_ioctl(struct drm_device *dev, void *data,
 		else
 			clear_bit(UCONTEXT_NO_ZEROMAP, &ctx->user_flags);
 		break;
+
 	case I915_CONTEXT_PARAM_NO_ERROR_CAPTURE:
 		if (args->size)
 			ret = -EINVAL;
@@ -1274,6 +1432,7 @@ int i915_gem_context_setparam_ioctl(struct drm_device *dev, void *data,
 		else
 			i915_gem_context_clear_no_error_capture(ctx);
 		break;
+
 	case I915_CONTEXT_PARAM_BANNABLE:
 		if (args->size)
 			ret = -EINVAL;
@@ -1285,13 +1444,22 @@ int i915_gem_context_setparam_ioctl(struct drm_device *dev, void *data,
 			i915_gem_context_clear_bannable(ctx);
 		break;
 
+	case I915_CONTEXT_PARAM_RECOVERABLE:
+		if (args->size)
+			ret = -EINVAL;
+		else if (args->value)
+			i915_gem_context_set_recoverable(ctx);
+		else
+			i915_gem_context_clear_recoverable(ctx);
+		break;
+
 	case I915_CONTEXT_PARAM_PRIORITY:
 		{
 			s64 priority = args->value;
 
 			if (args->size)
 				ret = -EINVAL;
-			else if (!(to_i915(dev)->caps.scheduler & I915_SCHEDULER_CAP_PRIORITY))
+			else if (!(ctx->i915->caps.scheduler & I915_SCHEDULER_CAP_PRIORITY))
 				ret = -ENODEV;
 			else if (priority > I915_CONTEXT_MAX_USER_PRIORITY ||
 				 priority < I915_CONTEXT_MIN_USER_PRIORITY)
@@ -1304,14 +1472,266 @@ int i915_gem_context_setparam_ioctl(struct drm_device *dev, void *data,
 					I915_USER_PRIORITY(priority);
 		}
 		break;
+
 	case I915_CONTEXT_PARAM_SSEU:
 		ret = set_sseu(ctx, args);
 		break;
+
+	case I915_CONTEXT_PARAM_VM:
+		ret = set_ppgtt(fpriv, ctx, args);
+		break;
+
+	case I915_CONTEXT_PARAM_BAN_PERIOD:
 	default:
 		ret = -EINVAL;
 		break;
 	}
 
+	return ret;
+}
+
+struct create_ext {
+	struct i915_gem_context *ctx;
+	struct drm_i915_file_private *fpriv;
+};
+
+static int create_setparam(struct i915_user_extension __user *ext, void *data)
+{
+	struct drm_i915_gem_context_create_ext_setparam local;
+	const struct create_ext *arg = data;
+
+	if (copy_from_user(&local, ext, sizeof(local)))
+		return -EFAULT;
+
+	if (local.param.ctx_id)
+		return -EINVAL;
+
+	return ctx_setparam(arg->fpriv, arg->ctx, &local.param);
+}
+
+static const i915_user_extension_fn create_extensions[] = {
+	[I915_CONTEXT_CREATE_EXT_SETPARAM] = create_setparam,
+};
+
+static bool client_is_banned(struct drm_i915_file_private *file_priv)
+{
+	return atomic_read(&file_priv->ban_score) >= I915_CLIENT_SCORE_BANNED;
+}
+
+int i915_gem_context_create_ioctl(struct drm_device *dev, void *data,
+				  struct drm_file *file)
+{
+	struct drm_i915_private *i915 = to_i915(dev);
+	struct drm_i915_gem_context_create_ext *args = data;
+	struct create_ext ext_data;
+	int ret;
+
+	if (!DRIVER_CAPS(i915)->has_logical_contexts)
+		return -ENODEV;
+
+	if (args->flags & I915_CONTEXT_CREATE_FLAGS_UNKNOWN)
+		return -EINVAL;
+
+	ret = i915_terminally_wedged(i915);
+	if (ret)
+		return ret;
+
+	ext_data.fpriv = file->driver_priv;
+	if (client_is_banned(ext_data.fpriv)) {
+		DRM_DEBUG("client %s[%d] banned from creating ctx\n",
+			  current->comm,
+			  pid_nr(get_task_pid(current, PIDTYPE_PID)));
+		return -EIO;
+	}
+
+	ret = i915_mutex_lock_interruptible(dev);
+	if (ret)
+		return ret;
+
+	ext_data.ctx = i915_gem_create_context(i915, args->flags);
+	mutex_unlock(&dev->struct_mutex);
+	if (IS_ERR(ext_data.ctx))
+		return PTR_ERR(ext_data.ctx);
+
+	if (args->flags & I915_CONTEXT_CREATE_FLAGS_USE_EXTENSIONS) {
+		ret = i915_user_extensions(u64_to_user_ptr(args->extensions),
+					   create_extensions,
+					   ARRAY_SIZE(create_extensions),
+					   &ext_data);
+		if (ret)
+			goto err_ctx;
+	}
+
+	ret = gem_context_register(ext_data.ctx, ext_data.fpriv);
+	if (ret < 0)
+		goto err_ctx;
+
+	args->ctx_id = ret;
+	DRM_DEBUG("HW context %d created\n", args->ctx_id);
+
+	return 0;
+
+err_ctx:
+	mutex_lock(&dev->struct_mutex);
+	context_close(ext_data.ctx);
+	mutex_unlock(&dev->struct_mutex);
+	return ret;
+}
+
+int i915_gem_context_destroy_ioctl(struct drm_device *dev, void *data,
+				   struct drm_file *file)
+{
+	struct drm_i915_gem_context_destroy *args = data;
+	struct drm_i915_file_private *file_priv = file->driver_priv;
+	struct i915_gem_context *ctx;
+
+	if (args->pad != 0)
+		return -EINVAL;
+
+	if (!args->ctx_id)
+		return -ENOENT;
+
+	if (mutex_lock_interruptible(&file_priv->context_idr_lock))
+		return -EINTR;
+
+	ctx = idr_remove(&file_priv->context_idr, args->ctx_id);
+	mutex_unlock(&file_priv->context_idr_lock);
+	if (!ctx)
+		return -ENOENT;
+
+	mutex_lock(&dev->struct_mutex);
+	context_close(ctx);
+	mutex_unlock(&dev->struct_mutex);
+
+	return 0;
+}
+
+static int get_sseu(struct i915_gem_context *ctx,
+		    struct drm_i915_gem_context_param *args)
+{
+	struct drm_i915_gem_context_param_sseu user_sseu;
+	struct intel_engine_cs *engine;
+	struct intel_context *ce;
+
+	if (args->size == 0)
+		goto out;
+	else if (args->size < sizeof(user_sseu))
+		return -EINVAL;
+
+	if (copy_from_user(&user_sseu, u64_to_user_ptr(args->value),
+			   sizeof(user_sseu)))
+		return -EFAULT;
+
+	if (user_sseu.flags || user_sseu.rsvd)
+		return -EINVAL;
+
+	engine = intel_engine_lookup_user(ctx->i915,
+					  user_sseu.engine.engine_class,
+					  user_sseu.engine.engine_instance);
+	if (!engine)
+		return -EINVAL;
+
+	ce = intel_context_pin_lock(ctx, engine); /* serialises with set_sseu */
+	if (IS_ERR(ce))
+		return PTR_ERR(ce);
+
+	user_sseu.slice_mask = ce->sseu.slice_mask;
+	user_sseu.subslice_mask = ce->sseu.subslice_mask;
+	user_sseu.min_eus_per_subslice = ce->sseu.min_eus_per_subslice;
+	user_sseu.max_eus_per_subslice = ce->sseu.max_eus_per_subslice;
+
+	intel_context_pin_unlock(ce);
+
+	if (copy_to_user(u64_to_user_ptr(args->value), &user_sseu,
+			 sizeof(user_sseu)))
+		return -EFAULT;
+
+out:
+	args->size = sizeof(user_sseu);
+
+	return 0;
+}
+
+int i915_gem_context_getparam_ioctl(struct drm_device *dev, void *data,
+				    struct drm_file *file)
+{
+	struct drm_i915_file_private *file_priv = file->driver_priv;
+	struct drm_i915_gem_context_param *args = data;
+	struct i915_gem_context *ctx;
+	int ret = 0;
+
+	ctx = i915_gem_context_lookup(file_priv, args->ctx_id);
+	if (!ctx)
+		return -ENOENT;
+
+	switch (args->param) {
+	case I915_CONTEXT_PARAM_NO_ZEROMAP:
+		args->size = 0;
+		args->value = test_bit(UCONTEXT_NO_ZEROMAP, &ctx->user_flags);
+		break;
+
+	case I915_CONTEXT_PARAM_GTT_SIZE:
+		args->size = 0;
+		if (ctx->ppgtt)
+			args->value = ctx->ppgtt->vm.total;
+		else if (to_i915(dev)->mm.aliasing_ppgtt)
+			args->value = to_i915(dev)->mm.aliasing_ppgtt->vm.total;
+		else
+			args->value = to_i915(dev)->ggtt.vm.total;
+		break;
+
+	case I915_CONTEXT_PARAM_NO_ERROR_CAPTURE:
+		args->size = 0;
+		args->value = i915_gem_context_no_error_capture(ctx);
+		break;
+
+	case I915_CONTEXT_PARAM_BANNABLE:
+		args->size = 0;
+		args->value = i915_gem_context_is_bannable(ctx);
+		break;
+
+	case I915_CONTEXT_PARAM_RECOVERABLE:
+		args->size = 0;
+		args->value = i915_gem_context_is_recoverable(ctx);
+		break;
+
+	case I915_CONTEXT_PARAM_PRIORITY:
+		args->size = 0;
+		args->value = ctx->sched.priority >> I915_USER_PRIORITY_SHIFT;
+		break;
+
+	case I915_CONTEXT_PARAM_SSEU:
+		ret = get_sseu(ctx, args);
+		break;
+
+	case I915_CONTEXT_PARAM_VM:
+		ret = get_ppgtt(file_priv, ctx, args);
+		break;
+
+	case I915_CONTEXT_PARAM_BAN_PERIOD:
+	default:
+		ret = -EINVAL;
+		break;
+	}
+
+	i915_gem_context_put(ctx);
+	return ret;
+}
+
+int i915_gem_context_setparam_ioctl(struct drm_device *dev, void *data,
+				    struct drm_file *file)
+{
+	struct drm_i915_file_private *file_priv = file->driver_priv;
+	struct drm_i915_gem_context_param *args = data;
+	struct i915_gem_context *ctx;
+	int ret;
+
+	ctx = i915_gem_context_lookup(file_priv, args->ctx_id);
+	if (!ctx)
+		return -ENOENT;
+
+	ret = ctx_setparam(file_priv, ctx, args);
+
 	i915_gem_context_put(ctx);
 	return ret;
 }
@@ -1385,3 +1805,28 @@ out_unlock:
 #include "selftests/mock_context.c"
 #include "selftests/i915_gem_context.c"
 #endif
+
+static void i915_global_gem_context_shrink(void)
+{
+	kmem_cache_shrink(global.slab_luts);
+}
+
+static void i915_global_gem_context_exit(void)
+{
+	kmem_cache_destroy(global.slab_luts);
+}
+
+static struct i915_global_gem_context global = { {
+	.shrink = i915_global_gem_context_shrink,
+	.exit = i915_global_gem_context_exit,
+} };
+
+int __init i915_global_gem_context_init(void)
+{
+	global.slab_luts = KMEM_CACHE(i915_lut_handle, 0);
+	if (!global.slab_luts)
+		return -ENOMEM;
+
+	i915_global_register(&global.base);
+	return 0;
+}
diff --git a/drivers/gpu/drm/i915/i915_gem_context.h b/drivers/gpu/drm/i915/i915_gem_context.h
index ca150a764c24..23dcb01bfd82 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.h
+++ b/drivers/gpu/drm/i915/i915_gem_context.h
@@ -25,210 +25,16 @@
 #ifndef __I915_GEM_CONTEXT_H__
 #define __I915_GEM_CONTEXT_H__
 
-#include <linux/bitops.h>
-#include <linux/list.h>
-#include <linux/radix-tree.h>
+#include "i915_gem_context_types.h"
 
 #include "i915_gem.h"
 #include "i915_scheduler.h"
+#include "intel_context.h"
 #include "intel_device_info.h"
 
-struct pid;
-
 struct drm_device;
 struct drm_file;
 
-struct drm_i915_private;
-struct drm_i915_file_private;
-struct i915_hw_ppgtt;
-struct i915_request;
-struct i915_vma;
-struct intel_ring;
-
-#define DEFAULT_CONTEXT_HANDLE 0
-
-struct intel_context;
-
-struct intel_context_ops {
-	void (*unpin)(struct intel_context *ce);
-	void (*destroy)(struct intel_context *ce);
-};
-
-/*
- * Powergating configuration for a particular (context,engine).
- */
-struct intel_sseu {
-	u8 slice_mask;
-	u8 subslice_mask;
-	u8 min_eus_per_subslice;
-	u8 max_eus_per_subslice;
-};
-
-/**
- * struct i915_gem_context - client state
- *
- * The struct i915_gem_context represents the combined view of the driver and
- * logical hardware state for a particular client.
- */
-struct i915_gem_context {
-	/** i915: i915 device backpointer */
-	struct drm_i915_private *i915;
-
-	/** file_priv: owning file descriptor */
-	struct drm_i915_file_private *file_priv;
-
-	/**
-	 * @ppgtt: unique address space (GTT)
-	 *
-	 * In full-ppgtt mode, each context has its own address space ensuring
-	 * complete seperation of one client from all others.
-	 *
-	 * In other modes, this is a NULL pointer with the expectation that
-	 * the caller uses the shared global GTT.
-	 */
-	struct i915_hw_ppgtt *ppgtt;
-
-	/**
-	 * @pid: process id of creator
-	 *
-	 * Note that who created the context may not be the principle user,
-	 * as the context may be shared across a local socket. However,
-	 * that should only affect the default context, all contexts created
-	 * explicitly by the client are expected to be isolated.
-	 */
-	struct pid *pid;
-
-	/**
-	 * @name: arbitrary name
-	 *
-	 * A name is constructed for the context from the creator's process
-	 * name, pid and user handle in order to uniquely identify the
-	 * context in messages.
-	 */
-	const char *name;
-
-	/** link: place with &drm_i915_private.context_list */
-	struct list_head link;
-	struct llist_node free_link;
-
-	/**
-	 * @ref: reference count
-	 *
-	 * A reference to a context is held by both the client who created it
-	 * and on each request submitted to the hardware using the request
-	 * (to ensure the hardware has access to the state until it has
-	 * finished all pending writes). See i915_gem_context_get() and
-	 * i915_gem_context_put() for access.
-	 */
-	struct kref ref;
-
-	/**
-	 * @rcu: rcu_head for deferred freeing.
-	 */
-	struct rcu_head rcu;
-
-	/**
-	 * @user_flags: small set of booleans controlled by the user
-	 */
-	unsigned long user_flags;
-#define UCONTEXT_NO_ZEROMAP		0
-#define UCONTEXT_NO_ERROR_CAPTURE	1
-#define UCONTEXT_BANNABLE		2
-
-	/**
-	 * @flags: small set of booleans
-	 */
-	unsigned long flags;
-#define CONTEXT_BANNED			0
-#define CONTEXT_CLOSED			1
-#define CONTEXT_FORCE_SINGLE_SUBMISSION	2
-
-	/**
-	 * @hw_id: - unique identifier for the context
-	 *
-	 * The hardware needs to uniquely identify the context for a few
-	 * functions like fault reporting, PASID, scheduling. The
-	 * &drm_i915_private.context_hw_ida is used to assign a unqiue
-	 * id for the lifetime of the context.
-	 *
-	 * @hw_id_pin_count: - number of times this context had been pinned
-	 * for use (should be, at most, once per engine).
-	 *
-	 * @hw_id_link: - all contexts with an assigned id are tracked
-	 * for possible repossession.
-	 */
-	unsigned int hw_id;
-	atomic_t hw_id_pin_count;
-	struct list_head hw_id_link;
-
-	/**
-	 * @user_handle: userspace identifier
-	 *
-	 * A unique per-file identifier is generated from
-	 * &drm_i915_file_private.contexts.
-	 */
-	u32 user_handle;
-
-	struct i915_sched_attr sched;
-
-	/** engine: per-engine logical HW state */
-	struct intel_context {
-		struct i915_gem_context *gem_context;
-		struct intel_engine_cs *active;
-		struct list_head signal_link;
-		struct list_head signals;
-		struct i915_vma *state;
-		struct intel_ring *ring;
-		u32 *lrc_reg_state;
-		u64 lrc_desc;
-		int pin_count;
-
-		/**
-		 * active_tracker: Active tracker for the external rq activity
-		 * on this intel_context object.
-		 */
-		struct i915_active_request active_tracker;
-
-		const struct intel_context_ops *ops;
-
-		/** sseu: Control eu/slice partitioning */
-		struct intel_sseu sseu;
-	} __engine[I915_NUM_ENGINES];
-
-	/** ring_size: size for allocating the per-engine ring buffer */
-	u32 ring_size;
-	/** desc_template: invariant fields for the HW context descriptor */
-	u32 desc_template;
-
-	/** guilty_count: How many times this context has caused a GPU hang. */
-	atomic_t guilty_count;
-	/**
-	 * @active_count: How many times this context was active during a GPU
-	 * hang, but did not cause it.
-	 */
-	atomic_t active_count;
-
-#define CONTEXT_SCORE_GUILTY		10
-#define CONTEXT_SCORE_BAN_THRESHOLD	40
-	/** ban_score: Accumulated score of all hangs caused by this context. */
-	atomic_t ban_score;
-
-	/** remap_slice: Bitmask of cache lines that need remapping */
-	u8 remap_slice;
-
-	/** handles_vma: rbtree to look up our context specific obj/vma for
-	 * the user handle. (user handles are per fd, but the binding is
-	 * per vm, which may be one per context or shared with the global GTT)
-	 */
-	struct radix_tree_root handles_vma;
-
-	/** handles_list: reverse list of all the rbtree entries in use for
-	 * this context, which allows us to free all the allocations on
-	 * context close.
-	 */
-	struct list_head handles_list;
-};
-
 static inline bool i915_gem_context_is_closed(const struct i915_gem_context *ctx)
 {
 	return test_bit(CONTEXT_CLOSED, &ctx->flags);
@@ -270,6 +76,21 @@ static inline void i915_gem_context_clear_bannable(struct i915_gem_context *ctx)
 	clear_bit(UCONTEXT_BANNABLE, &ctx->user_flags);
 }
 
+static inline bool i915_gem_context_is_recoverable(const struct i915_gem_context *ctx)
+{
+	return test_bit(UCONTEXT_RECOVERABLE, &ctx->user_flags);
+}
+
+static inline void i915_gem_context_set_recoverable(struct i915_gem_context *ctx)
+{
+	set_bit(UCONTEXT_RECOVERABLE, &ctx->user_flags);
+}
+
+static inline void i915_gem_context_clear_recoverable(struct i915_gem_context *ctx)
+{
+	clear_bit(UCONTEXT_RECOVERABLE, &ctx->user_flags);
+}
+
 static inline bool i915_gem_context_is_banned(const struct i915_gem_context *ctx)
 {
 	return test_bit(CONTEXT_BANNED, &ctx->flags);
@@ -305,45 +126,11 @@ static inline void i915_gem_context_unpin_hw_id(struct i915_gem_context *ctx)
 	atomic_dec(&ctx->hw_id_pin_count);
 }
 
-static inline bool i915_gem_context_is_default(const struct i915_gem_context *c)
-{
-	return c->user_handle == DEFAULT_CONTEXT_HANDLE;
-}
-
 static inline bool i915_gem_context_is_kernel(struct i915_gem_context *ctx)
 {
 	return !ctx->file_priv;
 }
 
-static inline struct intel_context *
-to_intel_context(struct i915_gem_context *ctx,
-		 const struct intel_engine_cs *engine)
-{
-	return &ctx->__engine[engine->id];
-}
-
-static inline struct intel_context *
-intel_context_pin(struct i915_gem_context *ctx, struct intel_engine_cs *engine)
-{
-	return engine->context_pin(engine, ctx);
-}
-
-static inline void __intel_context_pin(struct intel_context *ce)
-{
-	GEM_BUG_ON(!ce->pin_count);
-	ce->pin_count++;
-}
-
-static inline void intel_context_unpin(struct intel_context *ce)
-{
-	GEM_BUG_ON(!ce->pin_count);
-	if (--ce->pin_count)
-		return;
-
-	GEM_BUG_ON(!ce->ops);
-	ce->ops->unpin(ce);
-}
-
 /* i915_gem_context.c */
 int __must_check i915_gem_contexts_init(struct drm_i915_private *dev_priv);
 void i915_gem_contexts_lost(struct drm_i915_private *dev_priv);
@@ -354,12 +141,18 @@ int i915_gem_context_open(struct drm_i915_private *i915,
 void i915_gem_context_close(struct drm_file *file);
 
 int i915_switch_context(struct i915_request *rq);
-int i915_gem_switch_to_kernel_context(struct drm_i915_private *dev_priv);
+int i915_gem_switch_to_kernel_context(struct drm_i915_private *i915,
+				      intel_engine_mask_t engine_mask);
 
 void i915_gem_context_release(struct kref *ctx_ref);
 struct i915_gem_context *
 i915_gem_context_create_gvt(struct drm_device *dev);
 
+int i915_gem_vm_create_ioctl(struct drm_device *dev, void *data,
+			     struct drm_file *file);
+int i915_gem_vm_destroy_ioctl(struct drm_device *dev, void *data,
+			      struct drm_file *file);
+
 int i915_gem_context_create_ioctl(struct drm_device *dev, void *data,
 				  struct drm_file *file);
 int i915_gem_context_destroy_ioctl(struct drm_device *dev, void *data,
@@ -386,8 +179,7 @@ static inline void i915_gem_context_put(struct i915_gem_context *ctx)
 	kref_put(&ctx->ref, i915_gem_context_release);
 }
 
-void intel_context_init(struct intel_context *ce,
-			struct i915_gem_context *ctx,
-			struct intel_engine_cs *engine);
+struct i915_lut_handle *i915_lut_handle_alloc(void);
+void i915_lut_handle_free(struct i915_lut_handle *lut);
 
 #endif /* !__I915_GEM_CONTEXT_H__ */
diff --git a/drivers/gpu/drm/i915/i915_gem_context_types.h b/drivers/gpu/drm/i915/i915_gem_context_types.h
new file mode 100644
index 000000000000..e2ec58b10fb2
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_gem_context_types.h
@@ -0,0 +1,175 @@
+/*
+ * SPDX-License-Identifier: MIT
+ *
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __I915_GEM_CONTEXT_TYPES_H__
+#define __I915_GEM_CONTEXT_TYPES_H__
+
+#include <linux/atomic.h>
+#include <linux/list.h>
+#include <linux/llist.h>
+#include <linux/kref.h>
+#include <linux/mutex.h>
+#include <linux/radix-tree.h>
+#include <linux/rbtree.h>
+#include <linux/rcupdate.h>
+#include <linux/types.h>
+
+#include "i915_scheduler.h"
+#include "intel_context_types.h"
+
+struct pid;
+
+struct drm_i915_private;
+struct drm_i915_file_private;
+struct i915_hw_ppgtt;
+struct i915_timeline;
+struct intel_ring;
+
+/**
+ * struct i915_gem_context - client state
+ *
+ * The struct i915_gem_context represents the combined view of the driver and
+ * logical hardware state for a particular client.
+ */
+struct i915_gem_context {
+	/** i915: i915 device backpointer */
+	struct drm_i915_private *i915;
+
+	/** file_priv: owning file descriptor */
+	struct drm_i915_file_private *file_priv;
+
+	struct i915_timeline *timeline;
+
+	/**
+	 * @ppgtt: unique address space (GTT)
+	 *
+	 * In full-ppgtt mode, each context has its own address space ensuring
+	 * complete seperation of one client from all others.
+	 *
+	 * In other modes, this is a NULL pointer with the expectation that
+	 * the caller uses the shared global GTT.
+	 */
+	struct i915_hw_ppgtt *ppgtt;
+
+	/**
+	 * @pid: process id of creator
+	 *
+	 * Note that who created the context may not be the principle user,
+	 * as the context may be shared across a local socket. However,
+	 * that should only affect the default context, all contexts created
+	 * explicitly by the client are expected to be isolated.
+	 */
+	struct pid *pid;
+
+	/**
+	 * @name: arbitrary name
+	 *
+	 * A name is constructed for the context from the creator's process
+	 * name, pid and user handle in order to uniquely identify the
+	 * context in messages.
+	 */
+	const char *name;
+
+	/** link: place with &drm_i915_private.context_list */
+	struct list_head link;
+	struct llist_node free_link;
+
+	/**
+	 * @ref: reference count
+	 *
+	 * A reference to a context is held by both the client who created it
+	 * and on each request submitted to the hardware using the request
+	 * (to ensure the hardware has access to the state until it has
+	 * finished all pending writes). See i915_gem_context_get() and
+	 * i915_gem_context_put() for access.
+	 */
+	struct kref ref;
+
+	/**
+	 * @rcu: rcu_head for deferred freeing.
+	 */
+	struct rcu_head rcu;
+
+	/**
+	 * @user_flags: small set of booleans controlled by the user
+	 */
+	unsigned long user_flags;
+#define UCONTEXT_NO_ZEROMAP		0
+#define UCONTEXT_NO_ERROR_CAPTURE	1
+#define UCONTEXT_BANNABLE		2
+#define UCONTEXT_RECOVERABLE		3
+
+	/**
+	 * @flags: small set of booleans
+	 */
+	unsigned long flags;
+#define CONTEXT_BANNED			0
+#define CONTEXT_CLOSED			1
+#define CONTEXT_FORCE_SINGLE_SUBMISSION	2
+
+	/**
+	 * @hw_id: - unique identifier for the context
+	 *
+	 * The hardware needs to uniquely identify the context for a few
+	 * functions like fault reporting, PASID, scheduling. The
+	 * &drm_i915_private.context_hw_ida is used to assign a unqiue
+	 * id for the lifetime of the context.
+	 *
+	 * @hw_id_pin_count: - number of times this context had been pinned
+	 * for use (should be, at most, once per engine).
+	 *
+	 * @hw_id_link: - all contexts with an assigned id are tracked
+	 * for possible repossession.
+	 */
+	unsigned int hw_id;
+	atomic_t hw_id_pin_count;
+	struct list_head hw_id_link;
+
+	struct list_head active_engines;
+	struct mutex mutex;
+
+	struct i915_sched_attr sched;
+
+	/** hw_contexts: per-engine logical HW state */
+	struct rb_root hw_contexts;
+	spinlock_t hw_contexts_lock;
+
+	/** ring_size: size for allocating the per-engine ring buffer */
+	u32 ring_size;
+	/** desc_template: invariant fields for the HW context descriptor */
+	u32 desc_template;
+
+	/** guilty_count: How many times this context has caused a GPU hang. */
+	atomic_t guilty_count;
+	/**
+	 * @active_count: How many times this context was active during a GPU
+	 * hang, but did not cause it.
+	 */
+	atomic_t active_count;
+
+	/**
+	 * @hang_timestamp: The last time(s) this context caused a GPU hang
+	 */
+	unsigned long hang_timestamp[2];
+#define CONTEXT_FAST_HANG_JIFFIES (120 * HZ) /* 3 hangs within 120s? Banned! */
+
+	/** remap_slice: Bitmask of cache lines that need remapping */
+	u8 remap_slice;
+
+	/** handles_vma: rbtree to look up our context specific obj/vma for
+	 * the user handle. (user handles are per fd, but the binding is
+	 * per vm, which may be one per context or shared with the global GTT)
+	 */
+	struct radix_tree_root handles_vma;
+
+	/** handles_list: reverse list of all the rbtree entries in use for
+	 * this context, which allows us to free all the allocations on
+	 * context close.
+	 */
+	struct list_head handles_list;
+};
+
+#endif /* __I915_GEM_CONTEXT_TYPES_H__ */
diff --git a/drivers/gpu/drm/i915/i915_gem_dmabuf.c b/drivers/gpu/drm/i915/i915_gem_dmabuf.c
index 02f7298bfe57..5a101a9462d8 100644
--- a/drivers/gpu/drm/i915/i915_gem_dmabuf.c
+++ b/drivers/gpu/drm/i915/i915_gem_dmabuf.c
@@ -107,6 +107,7 @@ static void i915_gem_dmabuf_vunmap(struct dma_buf *dma_buf, void *vaddr)
 {
 	struct drm_i915_gem_object *obj = dma_buf_to_obj(dma_buf);
 
+	i915_gem_object_flush_map(obj);
 	i915_gem_object_unpin_map(obj);
 }
 
@@ -300,7 +301,7 @@ struct drm_gem_object *i915_gem_prime_import(struct drm_device *dev,
 
 	get_dma_buf(dma_buf);
 
-	obj = i915_gem_object_alloc(to_i915(dev));
+	obj = i915_gem_object_alloc();
 	if (obj == NULL) {
 		ret = -ENOMEM;
 		goto fail_detach;
diff --git a/drivers/gpu/drm/i915/i915_gem_evict.c b/drivers/gpu/drm/i915/i915_gem_evict.c
index 68d74c50ac39..060f5903544a 100644
--- a/drivers/gpu/drm/i915/i915_gem_evict.c
+++ b/drivers/gpu/drm/i915/i915_gem_evict.c
@@ -38,31 +38,21 @@ I915_SELFTEST_DECLARE(static struct igt_evict_ctl {
 
 static bool ggtt_is_idle(struct drm_i915_private *i915)
 {
-       struct intel_engine_cs *engine;
-       enum intel_engine_id id;
-
-       if (i915->gt.active_requests)
-	       return false;
-
-       for_each_engine(engine, i915, id) {
-	       if (!intel_engine_has_kernel_context(engine))
-		       return false;
-       }
-
-       return true;
+	return !i915->gt.active_requests;
 }
 
 static int ggtt_flush(struct drm_i915_private *i915)
 {
 	int err;
 
-	/* Not everything in the GGTT is tracked via vma (otherwise we
+	/*
+	 * Not everything in the GGTT is tracked via vma (otherwise we
 	 * could evict as required with minimal stalling) so we are forced
 	 * to idle the GPU and explicitly retire outstanding requests in
 	 * the hopes that we can then remove contexts and the like only
 	 * bound by their active reference.
 	 */
-	err = i915_gem_switch_to_kernel_context(i915);
+	err = i915_gem_switch_to_kernel_context(i915, i915->gt.active_engines);
 	if (err)
 		return err;
 
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 16f80a448820..c83d2a195d15 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -794,8 +794,8 @@ static int eb_wait_for_ring(const struct i915_execbuffer *eb)
 	 * keeping all of their resources pinned.
 	 */
 
-	ce = to_intel_context(eb->ctx, eb->engine);
-	if (!ce->ring) /* first use, assume empty! */
+	ce = intel_context_lookup(eb->ctx, eb->engine);
+	if (!ce || !ce->ring) /* first use, assume empty! */
 		return 0;
 
 	rq = __eb_wait_for_ring(ce->ring);
@@ -849,12 +849,12 @@ static int eb_lookup_vmas(struct i915_execbuffer *eb)
 		}
 
 		vma = i915_vma_instance(obj, eb->vm, NULL);
-		if (unlikely(IS_ERR(vma))) {
+		if (IS_ERR(vma)) {
 			err = PTR_ERR(vma);
 			goto err_obj;
 		}
 
-		lut = kmem_cache_alloc(eb->i915->luts, GFP_KERNEL);
+		lut = i915_lut_handle_alloc();
 		if (unlikely(!lut)) {
 			err = -ENOMEM;
 			goto err_obj;
@@ -862,7 +862,7 @@ static int eb_lookup_vmas(struct i915_execbuffer *eb)
 
 		err = radix_tree_insert(handles_vma, handle, vma);
 		if (unlikely(err)) {
-			kmem_cache_free(eb->i915->luts, lut);
+			i915_lut_handle_free(lut);
 			goto err_obj;
 		}
 
@@ -1001,7 +1001,10 @@ static void reloc_gpu_flush(struct reloc_cache *cache)
 {
 	GEM_BUG_ON(cache->rq_size >= cache->rq->batch->obj->base.size / sizeof(u32));
 	cache->rq_cmd[cache->rq_size] = MI_BATCH_BUFFER_END;
+
+	__i915_gem_object_flush_map(cache->rq->batch->obj, 0, cache->rq_size);
 	i915_gem_object_unpin_map(cache->rq->batch->obj);
+
 	i915_gem_chipset_flush(cache->rq->i915);
 
 	i915_request_add(cache->rq);
@@ -1214,10 +1217,6 @@ static int __reloc_gpu_alloc(struct i915_execbuffer *eb,
 	if (IS_ERR(cmd))
 		return PTR_ERR(cmd);
 
-	err = i915_gem_object_set_to_wc_domain(obj, false);
-	if (err)
-		goto err_unmap;
-
 	batch = i915_vma_instance(obj, vma->vm, NULL);
 	if (IS_ERR(batch)) {
 		err = PTR_ERR(batch);
@@ -1958,7 +1957,7 @@ static int i915_reset_gen7_sol_offsets(struct i915_request *rq)
 	u32 *cs;
 	int i;
 
-	if (!IS_GEN(rq->i915, 7) || rq->engine->id != RCS) {
+	if (!IS_GEN(rq->i915, 7) || rq->engine->id != RCS0) {
 		DRM_DEBUG("sol reset is gen7/rcs only\n");
 		return -EINVAL;
 	}
@@ -2083,11 +2082,11 @@ gen8_dispatch_bsd_engine(struct drm_i915_private *dev_priv,
 #define I915_USER_RINGS (4)
 
 static const enum intel_engine_id user_ring_map[I915_USER_RINGS + 1] = {
-	[I915_EXEC_DEFAULT]	= RCS,
-	[I915_EXEC_RENDER]	= RCS,
-	[I915_EXEC_BLT]		= BCS,
-	[I915_EXEC_BSD]		= VCS,
-	[I915_EXEC_VEBOX]	= VECS
+	[I915_EXEC_DEFAULT]	= RCS0,
+	[I915_EXEC_RENDER]	= RCS0,
+	[I915_EXEC_BLT]		= BCS0,
+	[I915_EXEC_BSD]		= VCS0,
+	[I915_EXEC_VEBOX]	= VECS0
 };
 
 static struct intel_engine_cs *
@@ -2110,7 +2109,7 @@ eb_select_engine(struct drm_i915_private *dev_priv,
 		return NULL;
 	}
 
-	if (user_ring_id == I915_EXEC_BSD && HAS_BSD2(dev_priv)) {
+	if (user_ring_id == I915_EXEC_BSD && HAS_ENGINE(dev_priv, VCS1)) {
 		unsigned int bsd_idx = args->flags & I915_EXEC_BSD_MASK;
 
 		if (bsd_idx == I915_EXEC_BSD_DEFAULT) {
@@ -2313,10 +2312,6 @@ i915_gem_do_execbuffer(struct drm_device *dev,
 	if (args->flags & I915_EXEC_IS_PINNED)
 		eb.batch_flags |= I915_DISPATCH_PINNED;
 
-	eb.engine = eb_select_engine(eb.i915, file, args);
-	if (!eb.engine)
-		return -EINVAL;
-
 	if (args->flags & I915_EXEC_FENCE_IN) {
 		in_fence = sync_file_get_fence(lower_32_bits(args->rsvd2));
 		if (!in_fence)
@@ -2341,6 +2336,12 @@ i915_gem_do_execbuffer(struct drm_device *dev,
 	if (unlikely(err))
 		goto err_destroy;
 
+	eb.engine = eb_select_engine(eb.i915, file, args);
+	if (!eb.engine) {
+		err = -EINVAL;
+		goto err_engine;
+	}
+
 	/*
 	 * Take a local wakeref for preparing to dispatch the execbuf as
 	 * we expect to access the hardware fairly frequently in the
@@ -2506,6 +2507,7 @@ err_unlock:
 	mutex_unlock(&dev->struct_mutex);
 err_rpm:
 	intel_runtime_pm_put(eb.i915, wakeref);
+err_engine:
 	i915_gem_context_put(eb.ctx);
 err_destroy:
 	eb_destroy(&eb);
diff --git a/drivers/gpu/drm/i915/i915_gem_fence_reg.c b/drivers/gpu/drm/i915/i915_gem_fence_reg.c
index e037e94792f3..3084f52e3372 100644
--- a/drivers/gpu/drm/i915/i915_gem_fence_reg.c
+++ b/drivers/gpu/drm/i915/i915_gem_fence_reg.c
@@ -210,6 +210,7 @@ static int fence_update(struct drm_i915_fence_reg *fence,
 			struct i915_vma *vma)
 {
 	intel_wakeref_t wakeref;
+	struct i915_vma *old;
 	int ret;
 
 	if (vma) {
@@ -229,49 +230,55 @@ static int fence_update(struct drm_i915_fence_reg *fence,
 			return ret;
 	}
 
-	if (fence->vma) {
-		struct i915_vma *old = fence->vma;
-
+	old = xchg(&fence->vma, NULL);
+	if (old) {
 		ret = i915_active_request_retire(&old->last_fence,
 					     &old->obj->base.dev->struct_mutex);
-		if (ret)
+		if (ret) {
+			fence->vma = old;
 			return ret;
+		}
 
 		i915_vma_flush_writes(old);
-	}
 
-	if (fence->vma && fence->vma != vma) {
-		/* Ensure that all userspace CPU access is completed before
+		/*
+		 * Ensure that all userspace CPU access is completed before
 		 * stealing the fence.
 		 */
-		GEM_BUG_ON(fence->vma->fence != fence);
-		i915_vma_revoke_mmap(fence->vma);
-
-		fence->vma->fence = NULL;
-		fence->vma = NULL;
+		if (old != vma) {
+			GEM_BUG_ON(old->fence != fence);
+			i915_vma_revoke_mmap(old);
+			old->fence = NULL;
+		}
 
 		list_move(&fence->link, &fence->i915->mm.fence_list);
 	}
 
-	/* We only need to update the register itself if the device is awake.
+	/*
+	 * We only need to update the register itself if the device is awake.
 	 * If the device is currently powered down, we will defer the write
 	 * to the runtime resume, see i915_gem_restore_fences().
+	 *
+	 * This only works for removing the fence register, on acquisition
+	 * the caller must hold the rpm wakeref. The fence register must
+	 * be cleared before we can use any other fences to ensure that
+	 * the new fences do not overlap the elided clears, confusing HW.
 	 */
 	wakeref = intel_runtime_pm_get_if_in_use(fence->i915);
-	if (wakeref) {
-		fence_write(fence, vma);
-		intel_runtime_pm_put(fence->i915, wakeref);
+	if (!wakeref) {
+		GEM_BUG_ON(vma);
+		return 0;
 	}
 
-	if (vma) {
-		if (fence->vma != vma) {
-			vma->fence = fence;
-			fence->vma = vma;
-		}
+	WRITE_ONCE(fence->vma, vma);
+	fence_write(fence, vma);
 
+	if (vma) {
+		vma->fence = fence;
 		list_move_tail(&fence->link, &fence->i915->mm.fence_list);
 	}
 
+	intel_runtime_pm_put(fence->i915, wakeref);
 	return 0;
 }
 
@@ -436,32 +443,6 @@ void i915_unreserve_fence(struct drm_i915_fence_reg *fence)
 }
 
 /**
- * i915_gem_revoke_fences - revoke fence state
- * @dev_priv: i915 device private
- *
- * Removes all GTT mmappings via the fence registers. This forces any user
- * of the fence to reacquire that fence before continuing with their access.
- * One use is during GPU reset where the fence register is lost and we need to
- * revoke concurrent userspace access via GTT mmaps until the hardware has been
- * reset and the fence registers have been restored.
- */
-void i915_gem_revoke_fences(struct drm_i915_private *dev_priv)
-{
-	int i;
-
-	lockdep_assert_held(&dev_priv->drm.struct_mutex);
-
-	for (i = 0; i < dev_priv->num_fence_regs; i++) {
-		struct drm_i915_fence_reg *fence = &dev_priv->fence_regs[i];
-
-		GEM_BUG_ON(fence->vma && fence->vma->fence != fence);
-
-		if (fence->vma)
-			i915_vma_revoke_mmap(fence->vma);
-	}
-}
-
-/**
  * i915_gem_restore_fences - restore fence state
  * @dev_priv: i915 device private
  *
@@ -473,9 +454,10 @@ void i915_gem_restore_fences(struct drm_i915_private *dev_priv)
 {
 	int i;
 
+	rcu_read_lock(); /* keep obj alive as we dereference */
 	for (i = 0; i < dev_priv->num_fence_regs; i++) {
 		struct drm_i915_fence_reg *reg = &dev_priv->fence_regs[i];
-		struct i915_vma *vma = reg->vma;
+		struct i915_vma *vma = READ_ONCE(reg->vma);
 
 		GEM_BUG_ON(vma && vma->fence != reg);
 
@@ -483,18 +465,12 @@ void i915_gem_restore_fences(struct drm_i915_private *dev_priv)
 		 * Commit delayed tiling changes if we have an object still
 		 * attached to the fence, otherwise just clear the fence.
 		 */
-		if (vma && !i915_gem_object_is_tiled(vma->obj)) {
-			GEM_BUG_ON(!reg->dirty);
-			GEM_BUG_ON(i915_vma_has_userfault(vma));
-
-			list_move(&reg->link, &dev_priv->mm.fence_list);
-			vma->fence = NULL;
+		if (vma && !i915_gem_object_is_tiled(vma->obj))
 			vma = NULL;
-		}
 
 		fence_write(reg, vma);
-		reg->vma = vma;
 	}
+	rcu_read_unlock();
 }
 
 /**
@@ -609,8 +585,38 @@ i915_gem_detect_bit_6_swizzle(struct drm_i915_private *dev_priv)
 		 */
 		swizzle_x = I915_BIT_6_SWIZZLE_NONE;
 		swizzle_y = I915_BIT_6_SWIZZLE_NONE;
-	} else if (IS_MOBILE(dev_priv) ||
-		   IS_I915G(dev_priv) || IS_I945G(dev_priv)) {
+	} else if (IS_G45(dev_priv) || IS_I965G(dev_priv) || IS_G33(dev_priv)) {
+		/* The 965, G33, and newer, have a very flexible memory
+		 * configuration.  It will enable dual-channel mode
+		 * (interleaving) on as much memory as it can, and the GPU
+		 * will additionally sometimes enable different bit 6
+		 * swizzling for tiled objects from the CPU.
+		 *
+		 * Here's what I found on the G965:
+		 *    slot fill         memory size  swizzling
+		 * 0A   0B   1A   1B    1-ch   2-ch
+		 * 512  0    0    0     512    0     O
+		 * 512  0    512  0     16     1008  X
+		 * 512  0    0    512   16     1008  X
+		 * 0    512  0    512   16     1008  X
+		 * 1024 1024 1024 0     2048   1024  O
+		 *
+		 * We could probably detect this based on either the DRB
+		 * matching, which was the case for the swizzling required in
+		 * the table above, or from the 1-ch value being less than
+		 * the minimum size of a rank.
+		 *
+		 * Reports indicate that the swizzling actually
+		 * varies depending upon page placement inside the
+		 * channels, i.e. we see swizzled pages where the
+		 * banks of memory are paired and unswizzled on the
+		 * uneven portion, so leave that as unknown.
+		 */
+		if (I915_READ16(C0DRB3) == I915_READ16(C1DRB3)) {
+			swizzle_x = I915_BIT_6_SWIZZLE_9_10;
+			swizzle_y = I915_BIT_6_SWIZZLE_9;
+		}
+	} else {
 		u32 dcc;
 
 		/* On 9xx chipsets, channel interleave by the CPU is
@@ -660,37 +666,6 @@ i915_gem_detect_bit_6_swizzle(struct drm_i915_private *dev_priv)
 			swizzle_x = I915_BIT_6_SWIZZLE_UNKNOWN;
 			swizzle_y = I915_BIT_6_SWIZZLE_UNKNOWN;
 		}
-	} else {
-		/* The 965, G33, and newer, have a very flexible memory
-		 * configuration.  It will enable dual-channel mode
-		 * (interleaving) on as much memory as it can, and the GPU
-		 * will additionally sometimes enable different bit 6
-		 * swizzling for tiled objects from the CPU.
-		 *
-		 * Here's what I found on the G965:
-		 *    slot fill         memory size  swizzling
-		 * 0A   0B   1A   1B    1-ch   2-ch
-		 * 512  0    0    0     512    0     O
-		 * 512  0    512  0     16     1008  X
-		 * 512  0    0    512   16     1008  X
-		 * 0    512  0    512   16     1008  X
-		 * 1024 1024 1024 0     2048   1024  O
-		 *
-		 * We could probably detect this based on either the DRB
-		 * matching, which was the case for the swizzling required in
-		 * the table above, or from the 1-ch value being less than
-		 * the minimum size of a rank.
-		 *
-		 * Reports indicate that the swizzling actually
-		 * varies depending upon page placement inside the
-		 * channels, i.e. we see swizzled pages where the
-		 * banks of memory are paired and unswizzled on the
-		 * uneven portion, so leave that as unknown.
-		 */
-		if (I915_READ16(C0DRB3) == I915_READ16(C1DRB3)) {
-			swizzle_x = I915_BIT_6_SWIZZLE_9_10;
-			swizzle_y = I915_BIT_6_SWIZZLE_9;
-		}
 	}
 
 	if (swizzle_x == I915_BIT_6_SWIZZLE_UNKNOWN ||
@@ -790,8 +765,7 @@ i915_gem_object_save_bit_17_swizzle(struct drm_i915_gem_object *obj,
 	int i;
 
 	if (obj->bit_17 == NULL) {
-		obj->bit_17 = kcalloc(BITS_TO_LONGS(page_count),
-				      sizeof(long), GFP_KERNEL);
+		obj->bit_17 = bitmap_zalloc(page_count, GFP_KERNEL);
 		if (obj->bit_17 == NULL) {
 			DRM_ERROR("Failed to allocate memory for bit 17 "
 				  "record\n");
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index d646d37eec2f..8f460cc4cc1f 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -584,7 +584,7 @@ setup_scratch_page(struct i915_address_space *vm, gfp_t gfp)
 	 * for all.
 	 */
 	size = I915_GTT_PAGE_SIZE_4K;
-	if (i915_vm_is_48bit(vm) &&
+	if (i915_vm_is_4lvl(vm) &&
 	    HAS_PAGE_SIZES(vm->i915, I915_GTT_PAGE_SIZE_64K)) {
 		size = I915_GTT_PAGE_SIZE_64K;
 		gfp |= __GFP_NOWARN;
@@ -613,7 +613,7 @@ setup_scratch_page(struct i915_address_space *vm, gfp_t gfp)
 
 		vm->scratch_page.page = page;
 		vm->scratch_page.daddr = addr;
-		vm->scratch_page.order = order;
+		vm->scratch_order = order;
 		return 0;
 
 unmap_page:
@@ -632,10 +632,11 @@ skip:
 static void cleanup_scratch_page(struct i915_address_space *vm)
 {
 	struct i915_page_dma *p = &vm->scratch_page;
+	int order = vm->scratch_order;
 
-	dma_unmap_page(vm->dma, p->daddr, BIT(p->order) << PAGE_SHIFT,
+	dma_unmap_page(vm->dma, p->daddr, BIT(order) << PAGE_SHIFT,
 		       PCI_DMA_BIDIRECTIONAL);
-	__free_pages(p->page, p->order);
+	__free_pages(p->page, order);
 }
 
 static struct i915_page_table *alloc_pt(struct i915_address_space *vm)
@@ -726,18 +727,13 @@ static void __pdp_fini(struct i915_page_directory_pointer *pdp)
 	pdp->page_directory = NULL;
 }
 
-static inline bool use_4lvl(const struct i915_address_space *vm)
-{
-	return i915_vm_is_48bit(vm);
-}
-
 static struct i915_page_directory_pointer *
 alloc_pdp(struct i915_address_space *vm)
 {
 	struct i915_page_directory_pointer *pdp;
 	int ret = -ENOMEM;
 
-	GEM_BUG_ON(!use_4lvl(vm));
+	GEM_BUG_ON(!i915_vm_is_4lvl(vm));
 
 	pdp = kzalloc(sizeof(*pdp), GFP_KERNEL);
 	if (!pdp)
@@ -766,7 +762,7 @@ static void free_pdp(struct i915_address_space *vm,
 {
 	__pdp_fini(pdp);
 
-	if (!use_4lvl(vm))
+	if (!i915_vm_is_4lvl(vm))
 		return;
 
 	cleanup_px(vm, pdp);
@@ -791,14 +787,15 @@ static void gen8_initialize_pml4(struct i915_address_space *vm,
 	memset_p((void **)pml4->pdps, vm->scratch_pdp, GEN8_PML4ES_PER_PML4);
 }
 
-/* PDE TLBs are a pain to invalidate on GEN8+. When we modify
+/*
+ * PDE TLBs are a pain to invalidate on GEN8+. When we modify
  * the page table structures, we mark them dirty so that
  * context switching/execlist queuing code takes extra steps
  * to ensure that tlbs are flushed.
  */
 static void mark_tlbs_dirty(struct i915_hw_ppgtt *ppgtt)
 {
-	ppgtt->pd_dirty_rings = INTEL_INFO(ppgtt->vm.i915)->ring_mask;
+	ppgtt->pd_dirty_engines = ALL_ENGINES;
 }
 
 /* Removes entries from a single page table, releasing it if it's empty.
@@ -809,8 +806,6 @@ static bool gen8_ppgtt_clear_pt(const struct i915_address_space *vm,
 				u64 start, u64 length)
 {
 	unsigned int num_entries = gen8_pte_count(start, length);
-	unsigned int pte = gen8_pte_index(start);
-	unsigned int pte_end = pte + num_entries;
 	gen8_pte_t *vaddr;
 
 	GEM_BUG_ON(num_entries > pt->used_ptes);
@@ -820,8 +815,7 @@ static bool gen8_ppgtt_clear_pt(const struct i915_address_space *vm,
 		return true;
 
 	vaddr = kmap_atomic_px(pt);
-	while (pte < pte_end)
-		vaddr[pte++] = vm->scratch_pte;
+	memset64(vaddr + gen8_pte_index(start), vm->scratch_pte, num_entries);
 	kunmap_atomic(vaddr);
 
 	return false;
@@ -872,7 +866,7 @@ static void gen8_ppgtt_set_pdpe(struct i915_address_space *vm,
 	gen8_ppgtt_pdpe_t *vaddr;
 
 	pdp->page_directory[pdpe] = pd;
-	if (!use_4lvl(vm))
+	if (!i915_vm_is_4lvl(vm))
 		return;
 
 	vaddr = kmap_atomic_px(pdp);
@@ -937,7 +931,7 @@ static void gen8_ppgtt_clear_4lvl(struct i915_address_space *vm,
 	struct i915_page_directory_pointer *pdp;
 	unsigned int pml4e;
 
-	GEM_BUG_ON(!use_4lvl(vm));
+	GEM_BUG_ON(!i915_vm_is_4lvl(vm));
 
 	gen8_for_each_pml4e(pdp, pml4, start, length, pml4e) {
 		GEM_BUG_ON(pdp == vm->scratch_pdp);
@@ -1219,7 +1213,7 @@ static int gen8_init_scratch(struct i915_address_space *vm)
 
 		GEM_BUG_ON(!clone->has_read_only);
 
-		vm->scratch_page.order = clone->scratch_page.order;
+		vm->scratch_order = clone->scratch_order;
 		vm->scratch_pte = clone->scratch_pte;
 		vm->scratch_pt  = clone->scratch_pt;
 		vm->scratch_pd  = clone->scratch_pd;
@@ -1234,7 +1228,7 @@ static int gen8_init_scratch(struct i915_address_space *vm)
 	vm->scratch_pte =
 		gen8_pte_encode(vm->scratch_page.daddr,
 				I915_CACHE_LLC,
-				PTE_READ_ONLY);
+				vm->has_read_only);
 
 	vm->scratch_pt = alloc_pt(vm);
 	if (IS_ERR(vm->scratch_pt)) {
@@ -1248,7 +1242,7 @@ static int gen8_init_scratch(struct i915_address_space *vm)
 		goto free_pt;
 	}
 
-	if (use_4lvl(vm)) {
+	if (i915_vm_is_4lvl(vm)) {
 		vm->scratch_pdp = alloc_pdp(vm);
 		if (IS_ERR(vm->scratch_pdp)) {
 			ret = PTR_ERR(vm->scratch_pdp);
@@ -1258,7 +1252,7 @@ static int gen8_init_scratch(struct i915_address_space *vm)
 
 	gen8_initialize_pt(vm, vm->scratch_pt);
 	gen8_initialize_pd(vm, vm->scratch_pd);
-	if (use_4lvl(vm))
+	if (i915_vm_is_4lvl(vm))
 		gen8_initialize_pdp(vm, vm->scratch_pdp);
 
 	return 0;
@@ -1280,7 +1274,7 @@ static int gen8_ppgtt_notify_vgt(struct i915_hw_ppgtt *ppgtt, bool create)
 	enum vgt_g2v_type msg;
 	int i;
 
-	if (use_4lvl(vm)) {
+	if (i915_vm_is_4lvl(vm)) {
 		const u64 daddr = px_dma(&ppgtt->pml4);
 
 		I915_WRITE(vgtif_reg(pdp[0].lo), lower_32_bits(daddr));
@@ -1310,7 +1304,7 @@ static void gen8_free_scratch(struct i915_address_space *vm)
 	if (!vm->scratch_page.daddr)
 		return;
 
-	if (use_4lvl(vm))
+	if (i915_vm_is_4lvl(vm))
 		free_pdp(vm, vm->scratch_pdp);
 	free_pd(vm, vm->scratch_pd);
 	free_pt(vm, vm->scratch_pt);
@@ -1356,7 +1350,7 @@ static void gen8_ppgtt_cleanup(struct i915_address_space *vm)
 	if (intel_vgpu_active(dev_priv))
 		gen8_ppgtt_notify_vgt(ppgtt, false);
 
-	if (use_4lvl(vm))
+	if (i915_vm_is_4lvl(vm))
 		gen8_ppgtt_cleanup_4lvl(ppgtt);
 	else
 		gen8_ppgtt_cleanup_3lvl(&ppgtt->vm, &ppgtt->pdp);
@@ -1519,6 +1513,23 @@ unwind:
 	return -ENOMEM;
 }
 
+static void ppgtt_init(struct drm_i915_private *i915,
+		       struct i915_hw_ppgtt *ppgtt)
+{
+	kref_init(&ppgtt->ref);
+
+	ppgtt->vm.i915 = i915;
+	ppgtt->vm.dma = &i915->drm.pdev->dev;
+	ppgtt->vm.total = BIT_ULL(INTEL_INFO(i915)->ppgtt_size);
+
+	i915_address_space_init(&ppgtt->vm, VM_CLASS_PPGTT);
+
+	ppgtt->vm.vma_ops.bind_vma    = ppgtt_bind_vma;
+	ppgtt->vm.vma_ops.unbind_vma  = ppgtt_unbind_vma;
+	ppgtt->vm.vma_ops.set_pages   = ppgtt_set_pages;
+	ppgtt->vm.vma_ops.clear_pages = clear_pages;
+}
+
 /*
  * GEN8 legacy ppgtt programming is accomplished through a max 4 PDP registers
  * with a net effect resembling a 2-level page table in normal x86 terms. Each
@@ -1535,19 +1546,15 @@ static struct i915_hw_ppgtt *gen8_ppgtt_create(struct drm_i915_private *i915)
 	if (!ppgtt)
 		return ERR_PTR(-ENOMEM);
 
-	kref_init(&ppgtt->ref);
+	ppgtt_init(i915, ppgtt);
 
-	ppgtt->vm.i915 = i915;
-	ppgtt->vm.dma = &i915->drm.pdev->dev;
-
-	ppgtt->vm.total = HAS_FULL_48BIT_PPGTT(i915) ?
-		1ULL << 48 :
-		1ULL << 32;
-
-	/* From bdw, there is support for read-only pages in the PPGTT. */
-	ppgtt->vm.has_read_only = true;
-
-	i915_address_space_init(&ppgtt->vm, VM_CLASS_PPGTT);
+	/*
+	 * From bdw, there is hw support for read-only pages in the PPGTT.
+	 *
+	 * Gen11 has HSDES#:1807136187 unresolved. Disable ro support
+	 * for now.
+	 */
+	ppgtt->vm.has_read_only = INTEL_GEN(i915) != 11;
 
 	/* There are only few exceptions for gen >=6. chv and bxt.
 	 * And we are not sure about the latter so play safe for now.
@@ -1559,7 +1566,7 @@ static struct i915_hw_ppgtt *gen8_ppgtt_create(struct drm_i915_private *i915)
 	if (err)
 		goto err_free;
 
-	if (use_4lvl(&ppgtt->vm)) {
+	if (i915_vm_is_4lvl(&ppgtt->vm)) {
 		err = setup_px(&ppgtt->vm, &ppgtt->pml4);
 		if (err)
 			goto err_scratch;
@@ -1592,11 +1599,6 @@ static struct i915_hw_ppgtt *gen8_ppgtt_create(struct drm_i915_private *i915)
 
 	ppgtt->vm.cleanup = gen8_ppgtt_cleanup;
 
-	ppgtt->vm.vma_ops.bind_vma    = ppgtt_bind_vma;
-	ppgtt->vm.vma_ops.unbind_vma  = ppgtt_unbind_vma;
-	ppgtt->vm.vma_ops.set_pages   = ppgtt_set_pages;
-	ppgtt->vm.vma_ops.clear_pages = clear_pages;
-
 	return ppgtt;
 
 err_scratch:
@@ -1672,8 +1674,7 @@ static void gen6_ppgtt_clear_range(struct i915_address_space *vm,
 
 	while (num_entries) {
 		struct i915_page_table *pt = ppgtt->base.pd.page_table[pde++];
-		const unsigned int end = min(pte + num_entries, GEN6_PTES);
-		const unsigned int count = end - pte;
+		const unsigned int count = min(num_entries, GEN6_PTES - pte);
 		gen6_pte_t *vaddr;
 
 		GEM_BUG_ON(pt == vm->scratch_pt);
@@ -1693,9 +1694,7 @@ static void gen6_ppgtt_clear_range(struct i915_address_space *vm,
 		 */
 
 		vaddr = kmap_atomic_px(pt);
-		do {
-			vaddr[pte++] = scratch_pte;
-		} while (pte < end);
+		memset32(vaddr + pte, scratch_pte, count);
 		kunmap_atomic(vaddr);
 
 		pte = 0;
@@ -1913,7 +1912,7 @@ static struct i915_vma *pd_vma_create(struct gen6_hw_ppgtt *ppgtt, int size)
 	GEM_BUG_ON(!IS_ALIGNED(size, I915_GTT_PAGE_SIZE));
 	GEM_BUG_ON(size > ggtt->vm.total);
 
-	vma = kmem_cache_zalloc(i915->vmas, GFP_KERNEL);
+	vma = i915_vma_alloc();
 	if (!vma)
 		return ERR_PTR(-ENOMEM);
 
@@ -1943,6 +1942,8 @@ int gen6_ppgtt_pin(struct i915_hw_ppgtt *base)
 	struct gen6_hw_ppgtt *ppgtt = to_gen6_ppgtt(base);
 	int err;
 
+	GEM_BUG_ON(ppgtt->base.vm.closed);
+
 	/*
 	 * Workaround the limited maximum vma->pin_count and the aliasing_ppgtt
 	 * which will be pinned into every active context.
@@ -1981,6 +1982,17 @@ void gen6_ppgtt_unpin(struct i915_hw_ppgtt *base)
 	i915_vma_unpin(ppgtt->vma);
 }
 
+void gen6_ppgtt_unpin_all(struct i915_hw_ppgtt *base)
+{
+	struct gen6_hw_ppgtt *ppgtt = to_gen6_ppgtt(base);
+
+	if (!ppgtt->pin_count)
+		return;
+
+	ppgtt->pin_count = 0;
+	i915_vma_unpin(ppgtt->vma);
+}
+
 static struct i915_hw_ppgtt *gen6_ppgtt_create(struct drm_i915_private *i915)
 {
 	struct i915_ggtt * const ggtt = &i915->ggtt;
@@ -1991,25 +2003,13 @@ static struct i915_hw_ppgtt *gen6_ppgtt_create(struct drm_i915_private *i915)
 	if (!ppgtt)
 		return ERR_PTR(-ENOMEM);
 
-	kref_init(&ppgtt->base.ref);
-
-	ppgtt->base.vm.i915 = i915;
-	ppgtt->base.vm.dma = &i915->drm.pdev->dev;
-
-	ppgtt->base.vm.total = I915_PDES * GEN6_PTES * I915_GTT_PAGE_SIZE;
-
-	i915_address_space_init(&ppgtt->base.vm, VM_CLASS_PPGTT);
+	ppgtt_init(i915, &ppgtt->base);
 
 	ppgtt->base.vm.allocate_va_range = gen6_alloc_va_range;
 	ppgtt->base.vm.clear_range = gen6_ppgtt_clear_range;
 	ppgtt->base.vm.insert_entries = gen6_ppgtt_insert_entries;
 	ppgtt->base.vm.cleanup = gen6_ppgtt_cleanup;
 
-	ppgtt->base.vm.vma_ops.bind_vma    = ppgtt_bind_vma;
-	ppgtt->base.vm.vma_ops.unbind_vma  = ppgtt_unbind_vma;
-	ppgtt->base.vm.vma_ops.set_pages   = ppgtt_set_pages;
-	ppgtt->base.vm.vma_ops.clear_pages = clear_pages;
-
 	ppgtt->base.vm.pte_encode = ggtt->vm.pte_encode;
 
 	err = gen6_ppgtt_init_scratch(ppgtt);
@@ -2087,8 +2087,7 @@ __hw_ppgtt_create(struct drm_i915_private *i915)
 }
 
 struct i915_hw_ppgtt *
-i915_ppgtt_create(struct drm_i915_private *i915,
-		  struct drm_i915_file_private *fpriv)
+i915_ppgtt_create(struct drm_i915_private *i915)
 {
 	struct i915_hw_ppgtt *ppgtt;
 
@@ -2096,19 +2095,11 @@ i915_ppgtt_create(struct drm_i915_private *i915,
 	if (IS_ERR(ppgtt))
 		return ppgtt;
 
-	ppgtt->vm.file = fpriv;
-
 	trace_i915_ppgtt_create(&ppgtt->vm);
 
 	return ppgtt;
 }
 
-void i915_ppgtt_close(struct i915_address_space *vm)
-{
-	GEM_BUG_ON(vm->closed);
-	vm->closed = true;
-}
-
 static void ppgtt_destroy_vma(struct i915_address_space *vm)
 {
 	struct list_head *phases[] = {
@@ -2675,7 +2666,7 @@ int i915_gem_init_aliasing_ppgtt(struct drm_i915_private *i915)
 	struct i915_hw_ppgtt *ppgtt;
 	int err;
 
-	ppgtt = i915_ppgtt_create(i915, ERR_PTR(-EPERM));
+	ppgtt = i915_ppgtt_create(i915);
 	if (IS_ERR(ppgtt))
 		return PTR_ERR(ppgtt);
 
@@ -3701,7 +3692,7 @@ i915_get_ggtt_vma_pages(struct i915_vma *vma)
 	}
 
 	ret = 0;
-	if (unlikely(IS_ERR(vma->pages))) {
+	if (IS_ERR(vma->pages)) {
 		ret = PTR_ERR(vma->pages);
 		vma->pages = NULL;
 		DRM_ERROR("Failed to get pages for VMA view type %u (%d)!\n",
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h
index 03ade71b8d9a..f597f35b109b 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.h
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.h
@@ -213,7 +213,6 @@ struct i915_vma;
 
 struct i915_page_dma {
 	struct page *page;
-	int order;
 	union {
 		dma_addr_t daddr;
 
@@ -293,6 +292,7 @@ struct i915_address_space {
 #define VM_CLASS_PPGTT 1
 
 	u64 scratch_pte;
+	int scratch_order;
 	struct i915_page_dma scratch_page;
 	struct i915_page_table *scratch_pt;
 	struct i915_page_directory *scratch_pd;
@@ -348,7 +348,7 @@ struct i915_address_space {
 #define i915_is_ggtt(vm) ((vm)->is_ggtt)
 
 static inline bool
-i915_vm_is_48bit(const struct i915_address_space *vm)
+i915_vm_is_4lvl(const struct i915_address_space *vm)
 {
 	return (vm->total - 1) >> 32;
 }
@@ -356,7 +356,7 @@ i915_vm_is_48bit(const struct i915_address_space *vm)
 static inline bool
 i915_vm_has_scratch_64K(struct i915_address_space *vm)
 {
-	return vm->scratch_page.order == get_order(I915_GTT_PAGE_SIZE_64K);
+	return vm->scratch_order == get_order(I915_GTT_PAGE_SIZE_64K);
 }
 
 /* The Graphics Translation Table is the way in which GEN hardware translates a
@@ -390,12 +390,14 @@ struct i915_hw_ppgtt {
 	struct i915_address_space vm;
 	struct kref ref;
 
-	unsigned long pd_dirty_rings;
+	intel_engine_mask_t pd_dirty_engines;
 	union {
 		struct i915_pml4 pml4;		/* GEN8+ & 48b PPGTT */
 		struct i915_page_directory_pointer pdp;	/* GEN8+ */
 		struct i915_page_directory pd;		/* GEN6-7 */
 	};
+
+	u32 user_handle;
 };
 
 struct gen6_hw_ppgtt {
@@ -488,7 +490,7 @@ static inline u32 gen6_pde_index(u32 addr)
 static inline unsigned int
 i915_pdpes_per_pdp(const struct i915_address_space *vm)
 {
-	if (i915_vm_is_48bit(vm))
+	if (i915_vm_is_4lvl(vm))
 		return GEN8_PML4ES_PER_PML4;
 
 	return GEN8_3LVL_PDPES;
@@ -603,15 +605,16 @@ int i915_gem_init_ggtt(struct drm_i915_private *dev_priv);
 void i915_ggtt_cleanup_hw(struct drm_i915_private *dev_priv);
 
 int i915_ppgtt_init_hw(struct drm_i915_private *dev_priv);
+
+struct i915_hw_ppgtt *i915_ppgtt_create(struct drm_i915_private *dev_priv);
 void i915_ppgtt_release(struct kref *kref);
-struct i915_hw_ppgtt *i915_ppgtt_create(struct drm_i915_private *dev_priv,
-					struct drm_i915_file_private *fpriv);
-void i915_ppgtt_close(struct i915_address_space *vm);
-static inline void i915_ppgtt_get(struct i915_hw_ppgtt *ppgtt)
+
+static inline struct i915_hw_ppgtt *i915_ppgtt_get(struct i915_hw_ppgtt *ppgtt)
 {
-	if (ppgtt)
-		kref_get(&ppgtt->ref);
+	kref_get(&ppgtt->ref);
+	return ppgtt;
 }
+
 static inline void i915_ppgtt_put(struct i915_hw_ppgtt *ppgtt)
 {
 	if (ppgtt)
@@ -620,6 +623,7 @@ static inline void i915_ppgtt_put(struct i915_hw_ppgtt *ppgtt)
 
 int gen6_ppgtt_pin(struct i915_hw_ppgtt *base);
 void gen6_ppgtt_unpin(struct i915_hw_ppgtt *base);
+void gen6_ppgtt_unpin_all(struct i915_hw_ppgtt *base);
 
 void i915_check_and_clear_faults(struct drm_i915_private *dev_priv);
 void i915_gem_suspend_gtt_mappings(struct drm_i915_private *dev_priv);
diff --git a/drivers/gpu/drm/i915/i915_gem_internal.c b/drivers/gpu/drm/i915/i915_gem_internal.c
index fddde1033e74..ab627ed1269c 100644
--- a/drivers/gpu/drm/i915/i915_gem_internal.c
+++ b/drivers/gpu/drm/i915/i915_gem_internal.c
@@ -193,7 +193,7 @@ i915_gem_object_create_internal(struct drm_i915_private *i915,
 	if (overflows_type(size, obj->base.size))
 		return ERR_PTR(-E2BIG);
 
-	obj = i915_gem_object_alloc(i915);
+	obj = i915_gem_object_alloc();
 	if (!obj)
 		return ERR_PTR(-ENOMEM);
 
diff --git a/drivers/gpu/drm/i915/i915_gem_object.c b/drivers/gpu/drm/i915/i915_gem_object.c
index aab8cdd80e6d..ac6a5ab84586 100644
--- a/drivers/gpu/drm/i915/i915_gem_object.c
+++ b/drivers/gpu/drm/i915/i915_gem_object.c
@@ -24,6 +24,22 @@
 
 #include "i915_drv.h"
 #include "i915_gem_object.h"
+#include "i915_globals.h"
+
+static struct i915_global_object {
+	struct i915_global base;
+	struct kmem_cache *slab_objects;
+} global;
+
+struct drm_i915_gem_object *i915_gem_object_alloc(void)
+{
+	return kmem_cache_zalloc(global.slab_objects, GFP_KERNEL);
+}
+
+void i915_gem_object_free(struct drm_i915_gem_object *obj)
+{
+	return kmem_cache_free(global.slab_objects, obj);
+}
 
 /**
  * Mark up the object's coherency levels for a given cache_level
@@ -46,3 +62,29 @@ void i915_gem_object_set_cache_coherency(struct drm_i915_gem_object *obj,
 	obj->cache_dirty =
 		!(obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_WRITE);
 }
+
+static void i915_global_objects_shrink(void)
+{
+	kmem_cache_shrink(global.slab_objects);
+}
+
+static void i915_global_objects_exit(void)
+{
+	kmem_cache_destroy(global.slab_objects);
+}
+
+static struct i915_global_object global = { {
+	.shrink = i915_global_objects_shrink,
+	.exit = i915_global_objects_exit,
+} };
+
+int __init i915_global_objects_init(void)
+{
+	global.slab_objects =
+		KMEM_CACHE(drm_i915_gem_object, SLAB_HWCACHE_ALIGN);
+	if (!global.slab_objects)
+		return -ENOMEM;
+
+	i915_global_register(&global.base);
+	return 0;
+}
diff --git a/drivers/gpu/drm/i915/i915_gem_object.h b/drivers/gpu/drm/i915/i915_gem_object.h
index fab040331cdb..ca93a40c0c87 100644
--- a/drivers/gpu/drm/i915/i915_gem_object.h
+++ b/drivers/gpu/drm/i915/i915_gem_object.h
@@ -304,6 +304,9 @@ to_intel_bo(struct drm_gem_object *gem)
 	return container_of(gem, struct drm_i915_gem_object, base);
 }
 
+struct drm_i915_gem_object *i915_gem_object_alloc(void);
+void i915_gem_object_free(struct drm_i915_gem_object *obj);
+
 /**
  * i915_gem_object_lookup_rcu - look up a temporary GEM object from its handle
  * @filp: DRM file private date
@@ -499,5 +502,8 @@ void i915_gem_object_set_cache_coherency(struct drm_i915_gem_object *obj,
 					 unsigned int cache_level);
 void i915_gem_object_flush_if_display(struct drm_i915_gem_object *obj);
 
-#endif
+void __i915_gem_object_release_shmem(struct drm_i915_gem_object *obj,
+				     struct sg_table *pages,
+				     bool needs_clflush);
 
+#endif
diff --git a/drivers/gpu/drm/i915/i915_gem_render_state.c b/drivers/gpu/drm/i915/i915_gem_render_state.c
index 90baf9086d0a..9440024c763f 100644
--- a/drivers/gpu/drm/i915/i915_gem_render_state.c
+++ b/drivers/gpu/drm/i915/i915_gem_render_state.c
@@ -42,7 +42,7 @@ struct intel_render_state {
 static const struct intel_renderstate_rodata *
 render_state_get_rodata(const struct intel_engine_cs *engine)
 {
-	if (engine->id != RCS)
+	if (engine->id != RCS0)
 		return NULL;
 
 	switch (INTEL_GEN(engine->i915)) {
@@ -164,7 +164,7 @@ static int render_state_setup(struct intel_render_state *so,
 		drm_clflush_virt_range(d, i * sizeof(u32));
 	kunmap_atomic(d);
 
-	ret = i915_gem_object_set_to_gtt_domain(so->obj, false);
+	ret = 0;
 out:
 	i915_gem_obj_finish_shmem_access(so->obj);
 	return ret;
diff --git a/drivers/gpu/drm/i915/i915_gem_stolen.c b/drivers/gpu/drm/i915/i915_gem_stolen.c
index 74a9661479ca..0a8082cfc761 100644
--- a/drivers/gpu/drm/i915/i915_gem_stolen.c
+++ b/drivers/gpu/drm/i915/i915_gem_stolen.c
@@ -565,7 +565,7 @@ _i915_gem_object_create_stolen(struct drm_i915_private *dev_priv,
 	struct drm_i915_gem_object *obj;
 	unsigned int cache_level;
 
-	obj = i915_gem_object_alloc(dev_priv);
+	obj = i915_gem_object_alloc();
 	if (obj == NULL)
 		return NULL;
 
diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c b/drivers/gpu/drm/i915/i915_gem_tiling.c
index 16cc9ddbce34..a9b5329dae3b 100644
--- a/drivers/gpu/drm/i915/i915_gem_tiling.c
+++ b/drivers/gpu/drm/i915/i915_gem_tiling.c
@@ -301,11 +301,11 @@ i915_gem_object_set_tiling(struct drm_i915_gem_object *obj,
 	/* Try to preallocate memory required to save swizzling on put-pages */
 	if (i915_gem_object_needs_bit17_swizzle(obj)) {
 		if (!obj->bit_17) {
-			obj->bit_17 = kcalloc(BITS_TO_LONGS(obj->base.size >> PAGE_SHIFT),
-					      sizeof(long), GFP_KERNEL);
+			obj->bit_17 = bitmap_zalloc(obj->base.size >> PAGE_SHIFT,
+						    GFP_KERNEL);
 		}
 	} else {
-		kfree(obj->bit_17);
+		bitmap_free(obj->bit_17);
 		obj->bit_17 = NULL;
 	}
 
diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c
index 1d3f9a31ad61..215bf3fef10c 100644
--- a/drivers/gpu/drm/i915/i915_gem_userptr.c
+++ b/drivers/gpu/drm/i915/i915_gem_userptr.c
@@ -673,9 +673,7 @@ i915_gem_userptr_put_pages(struct drm_i915_gem_object *obj,
 	if (!pages)
 		return;
 
-	if (obj->mm.madv != I915_MADV_WILLNEED)
-		obj->mm.dirty = false;
-
+	__i915_gem_object_release_shmem(obj, pages, true);
 	i915_gem_gtt_finish_pages(obj, pages);
 
 	for_each_sgt_page(page, sgt_iter, pages) {
@@ -795,7 +793,7 @@ i915_gem_userptr_ioctl(struct drm_device *dev,
 			return -ENODEV;
 	}
 
-	obj = i915_gem_object_alloc(dev_priv);
+	obj = i915_gem_object_alloc();
 	if (obj == NULL)
 		return -ENOMEM;
 
diff --git a/drivers/gpu/drm/i915/i915_globals.c b/drivers/gpu/drm/i915/i915_globals.c
new file mode 100644
index 000000000000..81e5c2ce336b
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_globals.c
@@ -0,0 +1,125 @@
+/*
+ * SPDX-License-Identifier: MIT
+ *
+ * Copyright © 2019 Intel Corporation
+ */
+
+#include <linux/slab.h>
+#include <linux/workqueue.h>
+
+#include "i915_active.h"
+#include "i915_gem_context.h"
+#include "i915_gem_object.h"
+#include "i915_globals.h"
+#include "i915_request.h"
+#include "i915_scheduler.h"
+#include "i915_vma.h"
+
+static LIST_HEAD(globals);
+
+static atomic_t active;
+static atomic_t epoch;
+static struct park_work {
+	struct rcu_work work;
+	int epoch;
+} park;
+
+static void i915_globals_shrink(void)
+{
+	struct i915_global *global;
+
+	/*
+	 * kmem_cache_shrink() discards empty slabs and reorders partially
+	 * filled slabs to prioritise allocating from the mostly full slabs,
+	 * with the aim of reducing fragmentation.
+	 */
+	list_for_each_entry(global, &globals, link)
+		global->shrink();
+}
+
+static void __i915_globals_park(struct work_struct *work)
+{
+	/* Confirm nothing woke up in the last grace period */
+	if (park.epoch == atomic_read(&epoch))
+		i915_globals_shrink();
+}
+
+void __init i915_global_register(struct i915_global *global)
+{
+	GEM_BUG_ON(!global->shrink);
+	GEM_BUG_ON(!global->exit);
+
+	list_add_tail(&global->link, &globals);
+}
+
+static void __i915_globals_cleanup(void)
+{
+	struct i915_global *global, *next;
+
+	list_for_each_entry_safe_reverse(global, next, &globals, link)
+		global->exit();
+}
+
+static __initconst int (* const initfn[])(void) = {
+	i915_global_active_init,
+	i915_global_context_init,
+	i915_global_gem_context_init,
+	i915_global_objects_init,
+	i915_global_request_init,
+	i915_global_scheduler_init,
+	i915_global_vma_init,
+};
+
+int __init i915_globals_init(void)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(initfn); i++) {
+		int err;
+
+		err = initfn[i]();
+		if (err) {
+			__i915_globals_cleanup();
+			return err;
+		}
+	}
+
+	INIT_RCU_WORK(&park.work, __i915_globals_park);
+	return 0;
+}
+
+void i915_globals_park(void)
+{
+	/*
+	 * Defer shrinking the global slab caches (and other work) until
+	 * after a RCU grace period has completed with no activity. This
+	 * is to try and reduce the latency impact on the consumers caused
+	 * by us shrinking the caches the same time as they are trying to
+	 * allocate, with the assumption being that if we idle long enough
+	 * for an RCU grace period to elapse since the last use, it is likely
+	 * to be longer until we need the caches again.
+	 */
+	if (!atomic_dec_and_test(&active))
+		return;
+
+	park.epoch = atomic_inc_return(&epoch);
+	queue_rcu_work(system_wq, &park.work);
+}
+
+void i915_globals_unpark(void)
+{
+	atomic_inc(&epoch);
+	atomic_inc(&active);
+}
+
+void __exit i915_globals_exit(void)
+{
+	/* Flush any residual park_work */
+	atomic_inc(&epoch);
+	flush_rcu_work(&park.work);
+
+	__i915_globals_cleanup();
+
+	/* And ensure that our DESTROY_BY_RCU slabs are truly destroyed */
+	rcu_barrier();
+}
diff --git a/drivers/gpu/drm/i915/i915_globals.h b/drivers/gpu/drm/i915/i915_globals.h
new file mode 100644
index 000000000000..04c1ce107fc0
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_globals.h
@@ -0,0 +1,35 @@
+/*
+ * SPDX-License-Identifier: MIT
+ *
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef _I915_GLOBALS_H_
+#define _I915_GLOBALS_H_
+
+typedef void (*i915_global_func_t)(void);
+
+struct i915_global {
+	struct list_head link;
+
+	i915_global_func_t shrink;
+	i915_global_func_t exit;
+};
+
+void i915_global_register(struct i915_global *global);
+
+int i915_globals_init(void);
+void i915_globals_park(void);
+void i915_globals_unpark(void);
+void i915_globals_exit(void);
+
+/* constructors */
+int i915_global_active_init(void);
+int i915_global_context_init(void);
+int i915_global_gem_context_init(void);
+int i915_global_objects_init(void);
+int i915_global_request_init(void);
+int i915_global_scheduler_init(void);
+int i915_global_vma_init(void);
+
+#endif /* _I915_GLOBALS_H_ */
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index aa6791255252..f51ff683dd2e 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -380,19 +380,16 @@ static void print_error_buffers(struct drm_i915_error_state_buf *m,
 	err_printf(m, "%s [%d]:\n", name, count);
 
 	while (count--) {
-		err_printf(m, "    %08x_%08x %8u %02x %02x %02x",
+		err_printf(m, "    %08x_%08x %8u %02x %02x",
 			   upper_32_bits(err->gtt_offset),
 			   lower_32_bits(err->gtt_offset),
 			   err->size,
 			   err->read_domains,
-			   err->write_domain,
-			   err->wseqno);
+			   err->write_domain);
 		err_puts(m, tiling_flag(err->tiling));
 		err_puts(m, dirty_flag(err->dirty));
 		err_puts(m, purgeable_flag(err->purgeable));
 		err_puts(m, err->userptr ? " userptr" : "");
-		err_puts(m, err->engine != -1 ? " " : "");
-		err_puts(m, engine_name(m->i915, err->engine));
 		err_puts(m, i915_cache_level_str(m->i915, err->cache_level));
 
 		if (err->name)
@@ -414,7 +411,7 @@ static void error_print_instdone(struct drm_i915_error_state_buf *m,
 	err_printf(m, "  INSTDONE: 0x%08x\n",
 		   ee->instdone.instdone);
 
-	if (ee->engine_id != RCS || INTEL_GEN(m->i915) <= 3)
+	if (ee->engine_id != RCS0 || INTEL_GEN(m->i915) <= 3)
 		return;
 
 	err_printf(m, "  SC_INSTDONE: 0x%08x\n",
@@ -434,11 +431,6 @@ static void error_print_instdone(struct drm_i915_error_state_buf *m,
 			   ee->instdone.row[slice][subslice]);
 }
 
-static const char *bannable(const struct drm_i915_error_context *ctx)
-{
-	return ctx->bannable ? "" : " (unbannable)";
-}
-
 static void error_print_request(struct drm_i915_error_state_buf *m,
 				const char *prefix,
 				const struct drm_i915_error_request *erq,
@@ -447,9 +439,8 @@ static void error_print_request(struct drm_i915_error_state_buf *m,
 	if (!erq->seqno)
 		return;
 
-	err_printf(m, "%s pid %d, ban score %d, seqno %8x:%08x%s%s, prio %d, emitted %dms, start %08x, head %08x, tail %08x\n",
-		   prefix, erq->pid, erq->ban_score,
-		   erq->context, erq->seqno,
+	err_printf(m, "%s pid %d, seqno %8x:%08x%s%s, prio %d, emitted %dms, start %08x, head %08x, tail %08x\n",
+		   prefix, erq->pid, erq->context, erq->seqno,
 		   test_bit(DMA_FENCE_FLAG_SIGNALED_BIT,
 			    &erq->flags) ? "!" : "",
 		   test_bit(DMA_FENCE_FLAG_ENABLE_SIGNAL_BIT,
@@ -463,10 +454,9 @@ static void error_print_context(struct drm_i915_error_state_buf *m,
 				const char *header,
 				const struct drm_i915_error_context *ctx)
 {
-	err_printf(m, "%s%s[%d] user_handle %d hw_id %d, prio %d, ban score %d%s guilty %d active %d\n",
-		   header, ctx->comm, ctx->pid, ctx->handle, ctx->hw_id,
-		   ctx->sched_attr.priority, ctx->ban_score, bannable(ctx),
-		   ctx->guilty, ctx->active);
+	err_printf(m, "%s%s[%d] hw_id %d, prio %d, guilty %d active %d\n",
+		   header, ctx->comm, ctx->pid, ctx->hw_id,
+		   ctx->sched_attr.priority, ctx->guilty, ctx->active);
 }
 
 static void error_print_engine(struct drm_i915_error_state_buf *m,
@@ -512,13 +502,6 @@ static void error_print_engine(struct drm_i915_error_state_buf *m,
 	if (INTEL_GEN(m->i915) >= 6) {
 		err_printf(m, "  RC PSMI: 0x%08x\n", ee->rc_psmi);
 		err_printf(m, "  FAULT_REG: 0x%08x\n", ee->fault_reg);
-		err_printf(m, "  SYNC_0: 0x%08x\n",
-			   ee->semaphore_mboxes[0]);
-		err_printf(m, "  SYNC_1: 0x%08x\n",
-			   ee->semaphore_mboxes[1]);
-		if (HAS_VEBOX(m->i915))
-			err_printf(m, "  SYNC_2: 0x%08x\n",
-				   ee->semaphore_mboxes[2]);
 	}
 	if (HAS_PPGTT(m->i915)) {
 		err_printf(m, "  GFX_MODE: 0x%08x\n", ee->vm_info.gfx_mode);
@@ -533,8 +516,6 @@ static void error_print_engine(struct drm_i915_error_state_buf *m,
 				   ee->vm_info.pp_dir_base);
 		}
 	}
-	err_printf(m, "  seqno: 0x%08x\n", ee->seqno);
-	err_printf(m, "  last_seqno: 0x%08x\n", ee->last_seqno);
 	err_printf(m, "  ring->head: 0x%08x\n", ee->cpu_ring_head);
 	err_printf(m, "  ring->tail: 0x%08x\n", ee->cpu_ring_tail);
 	err_printf(m, "  hangcheck timestamp: %dms (%lu%s)\n",
@@ -688,16 +669,17 @@ static void __err_print_to_sgl(struct drm_i915_error_state_buf *m,
 		if (!error->engine[i].context.pid)
 			continue;
 
-		err_printf(m, "Active process (on ring %s): %s [%d], score %d%s\n",
+		err_printf(m, "Active process (on ring %s): %s [%d]\n",
 			   engine_name(m->i915, i),
 			   error->engine[i].context.comm,
-			   error->engine[i].context.pid,
-			   error->engine[i].context.ban_score,
-			   bannable(&error->engine[i].context));
+			   error->engine[i].context.pid);
 	}
 	err_printf(m, "Reset count: %u\n", error->reset_count);
 	err_printf(m, "Suspend count: %u\n", error->suspend_count);
 	err_printf(m, "Platform: %s\n", intel_platform_name(error->device_info.platform));
+	err_printf(m, "Subplatform: 0x%x\n",
+		   intel_subplatform(&error->runtime_info,
+				     error->device_info.platform));
 	err_print_pciid(m, m->i915);
 
 	err_printf(m, "IOMMU enabled?: %d\n", error->iommu);
@@ -779,13 +761,9 @@ static void __err_print_to_sgl(struct drm_i915_error_state_buf *m,
 		if (obj) {
 			err_puts(m, m->i915->engine[i]->name);
 			if (ee->context.pid)
-				err_printf(m, " (submitted by %s [%d], ctx %d [%d], score %d%s)",
+				err_printf(m, " (submitted by %s [%d])",
 					   ee->context.comm,
-					   ee->context.pid,
-					   ee->context.handle,
-					   ee->context.hw_id,
-					   ee->context.ban_score,
-					   bannable(&ee->context));
+					   ee->context.pid);
 			err_printf(m, " --- gtt_offset = 0x%08x %08x\n",
 				   upper_32_bits(obj->gtt_offset),
 				   lower_32_bits(obj->gtt_offset));
@@ -1061,27 +1039,6 @@ i915_error_object_create(struct drm_i915_private *i915,
 	return dst;
 }
 
-/* The error capture is special as tries to run underneath the normal
- * locking rules - so we use the raw version of the i915_active_request lookup.
- */
-static inline u32
-__active_get_seqno(struct i915_active_request *active)
-{
-	struct i915_request *request;
-
-	request = __i915_active_request_peek(active);
-	return request ? request->global_seqno : 0;
-}
-
-static inline int
-__active_get_engine_id(struct i915_active_request *active)
-{
-	struct i915_request *request;
-
-	request = __i915_active_request_peek(active);
-	return request ? request->engine->id : -1;
-}
-
 static void capture_bo(struct drm_i915_error_buffer *err,
 		       struct i915_vma *vma)
 {
@@ -1090,9 +1047,6 @@ static void capture_bo(struct drm_i915_error_buffer *err,
 	err->size = obj->base.size;
 	err->name = obj->base.name;
 
-	err->wseqno = __active_get_seqno(&obj->frontbuffer_write);
-	err->engine = __active_get_engine_id(&obj->frontbuffer_write);
-
 	err->gtt_offset = vma->node.start;
 	err->read_domains = obj->read_domains;
 	err->write_domain = obj->write_domain;
@@ -1142,7 +1096,7 @@ static u32 capture_error_bo(struct drm_i915_error_buffer *err,
  * It's only a small step better than a random number in its current form.
  */
 static u32 i915_error_generate_code(struct i915_gpu_state *error,
-				    unsigned long engine_mask)
+				    intel_engine_mask_t engine_mask)
 {
 	/*
 	 * IPEHR would be an ideal way to detect errors, as it's the gross
@@ -1178,18 +1132,6 @@ static void gem_record_fences(struct i915_gpu_state *error)
 	error->nfence = i;
 }
 
-static void gen6_record_semaphore_state(struct intel_engine_cs *engine,
-					struct drm_i915_error_engine *ee)
-{
-	struct drm_i915_private *dev_priv = engine->i915;
-
-	ee->semaphore_mboxes[0] = I915_READ(RING_SYNC_0(engine->mmio_base));
-	ee->semaphore_mboxes[1] = I915_READ(RING_SYNC_1(engine->mmio_base));
-	if (HAS_VEBOX(dev_priv))
-		ee->semaphore_mboxes[2] =
-			I915_READ(RING_SYNC_2(engine->mmio_base));
-}
-
 static void error_record_engine_registers(struct i915_gpu_state *error,
 					  struct intel_engine_cs *engine,
 					  struct drm_i915_error_engine *ee)
@@ -1197,44 +1139,40 @@ static void error_record_engine_registers(struct i915_gpu_state *error,
 	struct drm_i915_private *dev_priv = engine->i915;
 
 	if (INTEL_GEN(dev_priv) >= 6) {
-		ee->rc_psmi = I915_READ(RING_PSMI_CTL(engine->mmio_base));
-		if (INTEL_GEN(dev_priv) >= 8) {
+		ee->rc_psmi = ENGINE_READ(engine, RING_PSMI_CTL);
+		if (INTEL_GEN(dev_priv) >= 8)
 			ee->fault_reg = I915_READ(GEN8_RING_FAULT_REG);
-		} else {
-			gen6_record_semaphore_state(engine, ee);
+		else
 			ee->fault_reg = I915_READ(RING_FAULT_REG(engine));
-		}
 	}
 
 	if (INTEL_GEN(dev_priv) >= 4) {
-		ee->faddr = I915_READ(RING_DMA_FADD(engine->mmio_base));
-		ee->ipeir = I915_READ(RING_IPEIR(engine->mmio_base));
-		ee->ipehr = I915_READ(RING_IPEHR(engine->mmio_base));
-		ee->instps = I915_READ(RING_INSTPS(engine->mmio_base));
-		ee->bbaddr = I915_READ(RING_BBADDR(engine->mmio_base));
+		ee->faddr = ENGINE_READ(engine, RING_DMA_FADD);
+		ee->ipeir = ENGINE_READ(engine, RING_IPEIR);
+		ee->ipehr = ENGINE_READ(engine, RING_IPEHR);
+		ee->instps = ENGINE_READ(engine, RING_INSTPS);
+		ee->bbaddr = ENGINE_READ(engine, RING_BBADDR);
 		if (INTEL_GEN(dev_priv) >= 8) {
-			ee->faddr |= (u64) I915_READ(RING_DMA_FADD_UDW(engine->mmio_base)) << 32;
-			ee->bbaddr |= (u64) I915_READ(RING_BBADDR_UDW(engine->mmio_base)) << 32;
+			ee->faddr |= (u64)ENGINE_READ(engine, RING_DMA_FADD_UDW) << 32;
+			ee->bbaddr |= (u64)ENGINE_READ(engine, RING_BBADDR_UDW) << 32;
 		}
-		ee->bbstate = I915_READ(RING_BBSTATE(engine->mmio_base));
+		ee->bbstate = ENGINE_READ(engine, RING_BBSTATE);
 	} else {
-		ee->faddr = I915_READ(DMA_FADD_I8XX);
-		ee->ipeir = I915_READ(IPEIR);
-		ee->ipehr = I915_READ(IPEHR);
+		ee->faddr = ENGINE_READ(engine, DMA_FADD_I8XX);
+		ee->ipeir = ENGINE_READ(engine, IPEIR);
+		ee->ipehr = ENGINE_READ(engine, IPEHR);
 	}
 
 	intel_engine_get_instdone(engine, &ee->instdone);
 
-	ee->instpm = I915_READ(RING_INSTPM(engine->mmio_base));
+	ee->instpm = ENGINE_READ(engine, RING_INSTPM);
 	ee->acthd = intel_engine_get_active_head(engine);
-	ee->seqno = intel_engine_get_seqno(engine);
-	ee->last_seqno = intel_engine_last_submit(engine);
-	ee->start = I915_READ_START(engine);
-	ee->head = I915_READ_HEAD(engine);
-	ee->tail = I915_READ_TAIL(engine);
-	ee->ctl = I915_READ_CTL(engine);
+	ee->start = ENGINE_READ(engine, RING_START);
+	ee->head = ENGINE_READ(engine, RING_HEAD);
+	ee->tail = ENGINE_READ(engine, RING_TAIL);
+	ee->ctl = ENGINE_READ(engine, RING_CTL);
 	if (INTEL_GEN(dev_priv) > 2)
-		ee->mode = I915_READ_MODE(engine);
+		ee->mode = ENGINE_READ(engine, RING_MI_MODE);
 
 	if (!HWS_NEEDS_PHYSICAL(dev_priv)) {
 		i915_reg_t mmio;
@@ -1242,16 +1180,17 @@ static void error_record_engine_registers(struct i915_gpu_state *error,
 		if (IS_GEN(dev_priv, 7)) {
 			switch (engine->id) {
 			default:
-			case RCS:
+				MISSING_CASE(engine->id);
+			case RCS0:
 				mmio = RENDER_HWS_PGA_GEN7;
 				break;
-			case BCS:
+			case BCS0:
 				mmio = BLT_HWS_PGA_GEN7;
 				break;
-			case VCS:
+			case VCS0:
 				mmio = BSD_HWS_PGA_GEN7;
 				break;
-			case VECS:
+			case VECS0:
 				mmio = VEBOX_HWS_PGA_GEN7;
 				break;
 			}
@@ -1276,20 +1215,23 @@ static void error_record_engine_registers(struct i915_gpu_state *error,
 
 		ee->vm_info.gfx_mode = I915_READ(RING_MODE_GEN7(engine));
 
-		if (IS_GEN(dev_priv, 6))
+		if (IS_GEN(dev_priv, 6)) {
 			ee->vm_info.pp_dir_base =
-				I915_READ(RING_PP_DIR_BASE_READ(engine));
-		else if (IS_GEN(dev_priv, 7))
+				ENGINE_READ(engine, RING_PP_DIR_BASE_READ);
+		} else if (IS_GEN(dev_priv, 7)) {
 			ee->vm_info.pp_dir_base =
-				I915_READ(RING_PP_DIR_BASE(engine));
-		else if (INTEL_GEN(dev_priv) >= 8)
+				ENGINE_READ(engine, RING_PP_DIR_BASE);
+		} else if (INTEL_GEN(dev_priv) >= 8) {
+			u32 base = engine->mmio_base;
+
 			for (i = 0; i < 4; i++) {
 				ee->vm_info.pdp[i] =
-					I915_READ(GEN8_RING_PDP_UDW(engine, i));
+					I915_READ(GEN8_RING_PDP_UDW(base, i));
 				ee->vm_info.pdp[i] <<= 32;
 				ee->vm_info.pdp[i] |=
-					I915_READ(GEN8_RING_PDP_LDW(engine, i));
+					I915_READ(GEN8_RING_PDP_LDW(base, i));
 			}
+		}
 	}
 }
 
@@ -1299,10 +1241,9 @@ static void record_request(struct i915_request *request,
 	struct i915_gem_context *ctx = request->gem_context;
 
 	erq->flags = request->fence.flags;
-	erq->context = ctx->hw_id;
+	erq->context = request->fence.context;
+	erq->seqno = request->fence.seqno;
 	erq->sched_attr = request->sched.attr;
-	erq->ban_score = atomic_read(&ctx->ban_score);
-	erq->seqno = request->global_seqno;
 	erq->jiffies = request->emitted_jiffies;
 	erq->start = i915_ggtt_offset(request->ring->vma);
 	erq->head = request->head;
@@ -1393,11 +1334,8 @@ static void record_context(struct drm_i915_error_context *e,
 		rcu_read_unlock();
 	}
 
-	e->handle = ctx->user_handle;
 	e->hw_id = ctx->hw_id;
 	e->sched_attr = ctx->sched;
-	e->ban_score = atomic_read(&ctx->ban_score);
-	e->bannable = i915_gem_context_is_bannable(ctx);
 	e->guilty = atomic_read(&ctx->guilty_count);
 	e->active = atomic_read(&ctx->active_count);
 }
@@ -1476,7 +1414,7 @@ static void gem_record_rings(struct i915_gpu_state *error)
 		error_record_engine_registers(error, engine, ee);
 		error_record_engine_execlists(engine, ee);
 
-		request = i915_gem_find_active_request(engine);
+		request = intel_engine_find_active_request(engine);
 		if (request) {
 			struct i915_gem_context *ctx = request->gem_context;
 			struct intel_ring *ring;
@@ -1669,7 +1607,7 @@ static void capture_reg_state(struct i915_gpu_state *error)
 	}
 
 	if (INTEL_GEN(dev_priv) >= 5)
-		error->ccid = I915_READ(CCID);
+		error->ccid = I915_READ(CCID(RENDER_RING_BASE));
 
 	/* 3: Feature specific registers */
 	if (IS_GEN_RANGE(dev_priv, 6, 7)) {
@@ -1697,16 +1635,17 @@ static void capture_reg_state(struct i915_gpu_state *error)
 		error->gtier[0] = I915_READ(GTIER);
 		error->ngtier = 1;
 	} else if (IS_GEN(dev_priv, 2)) {
-		error->ier = I915_READ16(IER);
+		error->ier = I915_READ16(GEN2_IER);
 	} else if (!IS_VALLEYVIEW(dev_priv)) {
-		error->ier = I915_READ(IER);
+		error->ier = I915_READ(GEN2_IER);
 	}
 	error->eir = I915_READ(EIR);
 	error->pgtbl_er = I915_READ(PGTBL_ER);
 }
 
 static const char *
-error_msg(struct i915_gpu_state *error, unsigned long engines, const char *msg)
+error_msg(struct i915_gpu_state *error,
+	  intel_engine_mask_t engines, const char *msg)
 {
 	int len;
 	int i;
@@ -1716,7 +1655,7 @@ error_msg(struct i915_gpu_state *error, unsigned long engines, const char *msg)
 			engines &= ~BIT(i);
 
 	len = scnprintf(error->error_msg, sizeof(error->error_msg),
-			"GPU HANG: ecode %d:%lx:0x%08x",
+			"GPU HANG: ecode %d:%x:0x%08x",
 			INTEL_GEN(error->i915), engines,
 			i915_error_generate_code(error, engines));
 	if (engines) {
@@ -1855,7 +1794,7 @@ i915_capture_gpu_state(struct drm_i915_private *i915)
  * to pick up.
  */
 void i915_capture_error_state(struct drm_i915_private *i915,
-			      unsigned long engine_mask,
+			      intel_engine_mask_t engine_mask,
 			      const char *msg)
 {
 	static bool warned;
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.h b/drivers/gpu/drm/i915/i915_gpu_error.h
index 53b1f22dd365..5dc761e85d9d 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.h
+++ b/drivers/gpu/drm/i915/i915_gpu_error.h
@@ -94,8 +94,6 @@ struct i915_gpu_state {
 		u32 cpu_ring_head;
 		u32 cpu_ring_tail;
 
-		u32 last_seqno;
-
 		/* Register state */
 		u32 start;
 		u32 tail;
@@ -108,24 +106,19 @@ struct i915_gpu_state {
 		u32 bbstate;
 		u32 instpm;
 		u32 instps;
-		u32 seqno;
 		u64 bbaddr;
 		u64 acthd;
 		u32 fault_reg;
 		u64 faddr;
 		u32 rc_psmi; /* sleep state */
-		u32 semaphore_mboxes[I915_NUM_ENGINES - 1];
 		struct intel_instdone instdone;
 
 		struct drm_i915_error_context {
 			char comm[TASK_COMM_LEN];
 			pid_t pid;
-			u32 handle;
 			u32 hw_id;
-			int ban_score;
 			int active;
 			int guilty;
-			bool bannable;
 			struct i915_sched_attr sched_attr;
 		} context;
 
@@ -149,7 +142,6 @@ struct i915_gpu_state {
 			long jiffies;
 			pid_t pid;
 			u32 context;
-			int ban_score;
 			u32 seqno;
 			u32 start;
 			u32 head;
@@ -170,7 +162,6 @@ struct i915_gpu_state {
 	struct drm_i915_error_buffer {
 		u32 size;
 		u32 name;
-		u32 wseqno;
 		u64 gtt_offset;
 		u32 read_domains;
 		u32 write_domain;
@@ -179,7 +170,6 @@ struct i915_gpu_state {
 		u32 dirty:1;
 		u32 purgeable:1;
 		u32 userptr:1;
-		s32 engine:4;
 		u32 cache_level:3;
 	} *active_bo[I915_NUM_ENGINES], *pinned_bo;
 	u32 active_bo_count[I915_NUM_ENGINES], pinned_bo_count;
@@ -205,38 +195,12 @@ struct i915_gpu_error {
 	atomic_t pending_fb_pin;
 
 	/**
-	 * State variable controlling the reset flow and count
-	 *
-	 * This is a counter which gets incremented when reset is triggered,
-	 *
-	 * Before the reset commences, the I915_RESET_BACKOFF bit is set
-	 * meaning that any waiters holding onto the struct_mutex should
-	 * relinquish the lock immediately in order for the reset to start.
-	 *
-	 * If reset is not completed successfully, the I915_WEDGE bit is
-	 * set meaning that hardware is terminally sour and there is no
-	 * recovery. All waiters on the reset_queue will be woken when
-	 * that happens.
-	 *
-	 * This counter is used by the wait_seqno code to notice that reset
-	 * event happened and it needs to restart the entire ioctl (since most
-	 * likely the seqno it waited for won't ever signal anytime soon).
-	 *
-	 * This is important for lock-free wait paths, where no contended lock
-	 * naturally enforces the correct ordering between the bail-out of the
-	 * waiter and the gpu reset work code.
-	 */
-	unsigned long reset_count;
-
-	/**
 	 * flags: Control various stages of the GPU reset
 	 *
-	 * #I915_RESET_BACKOFF - When we start a reset, we want to stop any
-	 * other users acquiring the struct_mutex. To do this we set the
-	 * #I915_RESET_BACKOFF bit in the error flags when we detect a reset
-	 * and then check for that bit before acquiring the struct_mutex (in
-	 * i915_mutex_lock_interruptible()?). I915_RESET_BACKOFF serves a
-	 * secondary role in preventing two concurrent global reset attempts.
+	 * #I915_RESET_BACKOFF - When we start a global reset, we need to
+	 * serialise with any other users attempting to do the same, and
+	 * any global resources that may be clobber by the reset (such as
+	 * FENCE registers).
 	 *
 	 * #I915_RESET_ENGINE[num_engines] - Since the driver doesn't need to
 	 * acquire the struct_mutex to reset an engine, we need an explicit
@@ -255,6 +219,9 @@ struct i915_gpu_error {
 #define I915_RESET_ENGINE	2
 #define I915_WEDGED		(BITS_PER_LONG - 1)
 
+	/** Number of times the device has been reset (global) */
+	u32 reset_count;
+
 	/** Number of times an engine has been reset */
 	u32 reset_engine_count[I915_NUM_ENGINES];
 
@@ -272,6 +239,8 @@ struct i915_gpu_error {
 	 */
 	wait_queue_head_t reset_queue;
 
+	struct srcu_struct reset_backoff_srcu;
+
 	struct i915_gpu_restart *restart;
 };
 
@@ -294,7 +263,7 @@ void i915_error_printf(struct drm_i915_error_state_buf *e, const char *f, ...);
 
 struct i915_gpu_state *i915_capture_gpu_state(struct drm_i915_private *i915);
 void i915_capture_error_state(struct drm_i915_private *dev_priv,
-			      unsigned long engine_mask,
+			      intel_engine_mask_t engine_mask,
 			      const char *error_msg);
 
 static inline struct i915_gpu_state *
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 441d2674b272..b92cfd69134b 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -28,15 +28,19 @@
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
-#include <linux/sysrq.h>
-#include <linux/slab.h>
 #include <linux/circ_buf.h>
-#include <drm/drm_irq.h>
+#include <linux/cpuidle.h>
+#include <linux/slab.h>
+#include <linux/sysrq.h>
+
 #include <drm/drm_drv.h>
+#include <drm/drm_irq.h>
 #include <drm/i915_drm.h>
+
 #include "i915_drv.h"
 #include "i915_trace.h"
 #include "intel_drv.h"
+#include "intel_psr.h"
 
 /**
  * DOC: interrupt handling
@@ -132,92 +136,120 @@ static const u32 hpd_icp[HPD_NUM_PINS] = {
 	[HPD_PORT_F] = SDE_TC4_HOTPLUG_ICP
 };
 
-/* IIR can theoretically queue up two events. Be paranoid. */
-#define GEN8_IRQ_RESET_NDX(type, which) do { \
-	I915_WRITE(GEN8_##type##_IMR(which), 0xffffffff); \
-	POSTING_READ(GEN8_##type##_IMR(which)); \
-	I915_WRITE(GEN8_##type##_IER(which), 0); \
-	I915_WRITE(GEN8_##type##_IIR(which), 0xffffffff); \
-	POSTING_READ(GEN8_##type##_IIR(which)); \
-	I915_WRITE(GEN8_##type##_IIR(which), 0xffffffff); \
-	POSTING_READ(GEN8_##type##_IIR(which)); \
-} while (0)
-
-#define GEN3_IRQ_RESET(type) do { \
-	I915_WRITE(type##IMR, 0xffffffff); \
-	POSTING_READ(type##IMR); \
-	I915_WRITE(type##IER, 0); \
-	I915_WRITE(type##IIR, 0xffffffff); \
-	POSTING_READ(type##IIR); \
-	I915_WRITE(type##IIR, 0xffffffff); \
-	POSTING_READ(type##IIR); \
-} while (0)
-
-#define GEN2_IRQ_RESET(type) do { \
-	I915_WRITE16(type##IMR, 0xffff); \
-	POSTING_READ16(type##IMR); \
-	I915_WRITE16(type##IER, 0); \
-	I915_WRITE16(type##IIR, 0xffff); \
-	POSTING_READ16(type##IIR); \
-	I915_WRITE16(type##IIR, 0xffff); \
-	POSTING_READ16(type##IIR); \
-} while (0)
+static void gen3_irq_reset(struct intel_uncore *uncore, i915_reg_t imr,
+			   i915_reg_t iir, i915_reg_t ier)
+{
+	intel_uncore_write(uncore, imr, 0xffffffff);
+	intel_uncore_posting_read(uncore, imr);
+
+	intel_uncore_write(uncore, ier, 0);
+
+	/* IIR can theoretically queue up two events. Be paranoid. */
+	intel_uncore_write(uncore, iir, 0xffffffff);
+	intel_uncore_posting_read(uncore, iir);
+	intel_uncore_write(uncore, iir, 0xffffffff);
+	intel_uncore_posting_read(uncore, iir);
+}
+
+static void gen2_irq_reset(struct intel_uncore *uncore)
+{
+	intel_uncore_write16(uncore, GEN2_IMR, 0xffff);
+	intel_uncore_posting_read16(uncore, GEN2_IMR);
+
+	intel_uncore_write16(uncore, GEN2_IER, 0);
+
+	/* IIR can theoretically queue up two events. Be paranoid. */
+	intel_uncore_write16(uncore, GEN2_IIR, 0xffff);
+	intel_uncore_posting_read16(uncore, GEN2_IIR);
+	intel_uncore_write16(uncore, GEN2_IIR, 0xffff);
+	intel_uncore_posting_read16(uncore, GEN2_IIR);
+}
+
+#define GEN8_IRQ_RESET_NDX(uncore, type, which) \
+({ \
+	unsigned int which_ = which; \
+	gen3_irq_reset((uncore), GEN8_##type##_IMR(which_), \
+		       GEN8_##type##_IIR(which_), GEN8_##type##_IER(which_)); \
+})
+
+#define GEN3_IRQ_RESET(uncore, type) \
+	gen3_irq_reset((uncore), type##IMR, type##IIR, type##IER)
+
+#define GEN2_IRQ_RESET(uncore) \
+	gen2_irq_reset(uncore)
 
 /*
  * We should clear IMR at preinstall/uninstall, and just check at postinstall.
  */
-static void gen3_assert_iir_is_zero(struct drm_i915_private *dev_priv,
-				    i915_reg_t reg)
+static void gen3_assert_iir_is_zero(struct intel_uncore *uncore, i915_reg_t reg)
 {
-	u32 val = I915_READ(reg);
+	u32 val = intel_uncore_read(uncore, reg);
 
 	if (val == 0)
 		return;
 
 	WARN(1, "Interrupt register 0x%x is not zero: 0x%08x\n",
 	     i915_mmio_reg_offset(reg), val);
-	I915_WRITE(reg, 0xffffffff);
-	POSTING_READ(reg);
-	I915_WRITE(reg, 0xffffffff);
-	POSTING_READ(reg);
+	intel_uncore_write(uncore, reg, 0xffffffff);
+	intel_uncore_posting_read(uncore, reg);
+	intel_uncore_write(uncore, reg, 0xffffffff);
+	intel_uncore_posting_read(uncore, reg);
 }
 
-static void gen2_assert_iir_is_zero(struct drm_i915_private *dev_priv,
-				    i915_reg_t reg)
+static void gen2_assert_iir_is_zero(struct intel_uncore *uncore)
 {
-	u16 val = I915_READ16(reg);
+	u16 val = intel_uncore_read16(uncore, GEN2_IIR);
 
 	if (val == 0)
 		return;
 
 	WARN(1, "Interrupt register 0x%x is not zero: 0x%08x\n",
-	     i915_mmio_reg_offset(reg), val);
-	I915_WRITE16(reg, 0xffff);
-	POSTING_READ16(reg);
-	I915_WRITE16(reg, 0xffff);
-	POSTING_READ16(reg);
-}
-
-#define GEN8_IRQ_INIT_NDX(type, which, imr_val, ier_val) do { \
-	gen3_assert_iir_is_zero(dev_priv, GEN8_##type##_IIR(which)); \
-	I915_WRITE(GEN8_##type##_IER(which), (ier_val)); \
-	I915_WRITE(GEN8_##type##_IMR(which), (imr_val)); \
-	POSTING_READ(GEN8_##type##_IMR(which)); \
-} while (0)
-
-#define GEN3_IRQ_INIT(type, imr_val, ier_val) do { \
-	gen3_assert_iir_is_zero(dev_priv, type##IIR); \
-	I915_WRITE(type##IER, (ier_val)); \
-	I915_WRITE(type##IMR, (imr_val)); \
-	POSTING_READ(type##IMR); \
-} while (0)
-
-#define GEN2_IRQ_INIT(type, imr_val, ier_val) do { \
-	gen2_assert_iir_is_zero(dev_priv, type##IIR); \
-	I915_WRITE16(type##IER, (ier_val)); \
-	I915_WRITE16(type##IMR, (imr_val)); \
-	POSTING_READ16(type##IMR); \
-} while (0)
+	     i915_mmio_reg_offset(GEN2_IIR), val);
+	intel_uncore_write16(uncore, GEN2_IIR, 0xffff);
+	intel_uncore_posting_read16(uncore, GEN2_IIR);
+	intel_uncore_write16(uncore, GEN2_IIR, 0xffff);
+	intel_uncore_posting_read16(uncore, GEN2_IIR);
+}
+
+static void gen3_irq_init(struct intel_uncore *uncore,
+			  i915_reg_t imr, u32 imr_val,
+			  i915_reg_t ier, u32 ier_val,
+			  i915_reg_t iir)
+{
+	gen3_assert_iir_is_zero(uncore, iir);
+
+	intel_uncore_write(uncore, ier, ier_val);
+	intel_uncore_write(uncore, imr, imr_val);
+	intel_uncore_posting_read(uncore, imr);
+}
+
+static void gen2_irq_init(struct intel_uncore *uncore,
+			  u32 imr_val, u32 ier_val)
+{
+	gen2_assert_iir_is_zero(uncore);
+
+	intel_uncore_write16(uncore, GEN2_IER, ier_val);
+	intel_uncore_write16(uncore, GEN2_IMR, imr_val);
+	intel_uncore_posting_read16(uncore, GEN2_IMR);
+}
+
+#define GEN8_IRQ_INIT_NDX(uncore, type, which, imr_val, ier_val) \
+({ \
+	unsigned int which_ = which; \
+	gen3_irq_init((uncore), \
+		      GEN8_##type##_IMR(which_), imr_val, \
+		      GEN8_##type##_IER(which_), ier_val, \
+		      GEN8_##type##_IIR(which_)); \
+})
+
+#define GEN3_IRQ_INIT(uncore, type, imr_val, ier_val) \
+	gen3_irq_init((uncore), \
+		      type##IMR, imr_val, \
+		      type##IER, ier_val, \
+		      type##IIR)
+
+#define GEN2_IRQ_INIT(uncore, imr_val, ier_val) \
+	gen2_irq_init((uncore), imr_val, ier_val)
 
 static void gen6_rps_irq_handler(struct drm_i915_private *dev_priv, u32 pm_iir);
 static void gen9_guc_irq_handler(struct drm_i915_private *dev_priv, u32 pm_iir);
@@ -268,7 +300,7 @@ static bool gen11_reset_one_iir(struct drm_i915_private * const i915,
 				const unsigned int bank,
 				const unsigned int bit)
 {
-	void __iomem * const regs = i915->regs;
+	void __iomem * const regs = i915->uncore.regs;
 	u32 dw;
 
 	lockdep_assert_held(&i915->irq_lock);
@@ -365,24 +397,41 @@ static i915_reg_t gen6_pm_iir(struct drm_i915_private *dev_priv)
 	return INTEL_GEN(dev_priv) >= 8 ? GEN8_GT_IIR(2) : GEN6_PMIIR;
 }
 
-static i915_reg_t gen6_pm_imr(struct drm_i915_private *dev_priv)
+static void write_pm_imr(struct drm_i915_private *dev_priv)
 {
-	if (INTEL_GEN(dev_priv) >= 11)
-		return GEN11_GPM_WGBOXPERF_INTR_MASK;
-	else if (INTEL_GEN(dev_priv) >= 8)
-		return GEN8_GT_IMR(2);
-	else
-		return GEN6_PMIMR;
+	i915_reg_t reg;
+	u32 mask = dev_priv->pm_imr;
+
+	if (INTEL_GEN(dev_priv) >= 11) {
+		reg = GEN11_GPM_WGBOXPERF_INTR_MASK;
+		/* pm is in upper half */
+		mask = mask << 16;
+	} else if (INTEL_GEN(dev_priv) >= 8) {
+		reg = GEN8_GT_IMR(2);
+	} else {
+		reg = GEN6_PMIMR;
+	}
+
+	I915_WRITE(reg, mask);
+	POSTING_READ(reg);
 }
 
-static i915_reg_t gen6_pm_ier(struct drm_i915_private *dev_priv)
+static void write_pm_ier(struct drm_i915_private *dev_priv)
 {
-	if (INTEL_GEN(dev_priv) >= 11)
-		return GEN11_GPM_WGBOXPERF_INTR_ENABLE;
-	else if (INTEL_GEN(dev_priv) >= 8)
-		return GEN8_GT_IER(2);
-	else
-		return GEN6_PMIER;
+	i915_reg_t reg;
+	u32 mask = dev_priv->pm_ier;
+
+	if (INTEL_GEN(dev_priv) >= 11) {
+		reg = GEN11_GPM_WGBOXPERF_INTR_ENABLE;
+		/* pm is in upper half */
+		mask = mask << 16;
+	} else if (INTEL_GEN(dev_priv) >= 8) {
+		reg = GEN8_GT_IER(2);
+	} else {
+		reg = GEN6_PMIER;
+	}
+
+	I915_WRITE(reg, mask);
 }
 
 /**
@@ -407,8 +456,7 @@ static void snb_update_pm_irq(struct drm_i915_private *dev_priv,
 
 	if (new_val != dev_priv->pm_imr) {
 		dev_priv->pm_imr = new_val;
-		I915_WRITE(gen6_pm_imr(dev_priv), dev_priv->pm_imr);
-		POSTING_READ(gen6_pm_imr(dev_priv));
+		write_pm_imr(dev_priv);
 	}
 }
 
@@ -449,7 +497,7 @@ static void gen6_enable_pm_irq(struct drm_i915_private *dev_priv, u32 enable_mas
 	lockdep_assert_held(&dev_priv->irq_lock);
 
 	dev_priv->pm_ier |= enable_mask;
-	I915_WRITE(gen6_pm_ier(dev_priv), dev_priv->pm_ier);
+	write_pm_ier(dev_priv);
 	gen6_unmask_pm_irq(dev_priv, enable_mask);
 	/* unmask_pm_irq provides an implicit barrier (POSTING_READ) */
 }
@@ -460,7 +508,7 @@ static void gen6_disable_pm_irq(struct drm_i915_private *dev_priv, u32 disable_m
 
 	dev_priv->pm_ier &= ~disable_mask;
 	__gen6_mask_pm_irq(dev_priv, disable_mask);
-	I915_WRITE(gen6_pm_ier(dev_priv), dev_priv->pm_ier);
+	write_pm_ier(dev_priv);
 	/* though a barrier is missing here, but don't really need a one */
 }
 
@@ -748,13 +796,21 @@ void i915_disable_pipestat(struct drm_i915_private *dev_priv,
 	POSTING_READ(reg);
 }
 
+static bool i915_has_asle(struct drm_i915_private *dev_priv)
+{
+	if (!dev_priv->opregion.asle)
+		return false;
+
+	return IS_PINEVIEW(dev_priv) || IS_MOBILE(dev_priv);
+}
+
 /**
  * i915_enable_asle_pipestat - enable ASLE pipestat for OpRegion
  * @dev_priv: i915 device private
  */
 static void i915_enable_asle_pipestat(struct drm_i915_private *dev_priv)
 {
-	if (!dev_priv->opregion.asle || !IS_MOBILE(dev_priv))
+	if (!i915_has_asle(dev_priv))
 		return;
 
 	spin_lock_irq(&dev_priv->irq_lock);
@@ -1288,6 +1344,18 @@ static void gen6_pm_rps_work(struct work_struct *work)
 
 	rps->last_adj = adj;
 
+	/*
+	 * Limit deboosting and boosting to keep ourselves at the extremes
+	 * when in the respective power modes (i.e. slowly decrease frequencies
+	 * while in the HIGH_POWER zone and slowly increase frequencies while
+	 * in the LOW_POWER zone). On idle, we will hit the timeout and drop
+	 * to the next level quickly, and conversely if busy we expect to
+	 * hit a waitboost and rapidly switch into max power.
+	 */
+	if ((adj < 0 && rps->power.mode == HIGH_POWER) ||
+	    (adj > 0 && rps->power.mode == LOW_POWER))
+		rps->last_adj = 0;
+
 	/* sysfs frequency interfaces may have snuck in while servicing the
 	 * interrupt
 	 */
@@ -1415,20 +1483,20 @@ static void ilk_gt_irq_handler(struct drm_i915_private *dev_priv,
 			       u32 gt_iir)
 {
 	if (gt_iir & GT_RENDER_USER_INTERRUPT)
-		intel_engine_breadcrumbs_irq(dev_priv->engine[RCS]);
+		intel_engine_breadcrumbs_irq(dev_priv->engine[RCS0]);
 	if (gt_iir & ILK_BSD_USER_INTERRUPT)
-		intel_engine_breadcrumbs_irq(dev_priv->engine[VCS]);
+		intel_engine_breadcrumbs_irq(dev_priv->engine[VCS0]);
 }
 
 static void snb_gt_irq_handler(struct drm_i915_private *dev_priv,
 			       u32 gt_iir)
 {
 	if (gt_iir & GT_RENDER_USER_INTERRUPT)
-		intel_engine_breadcrumbs_irq(dev_priv->engine[RCS]);
+		intel_engine_breadcrumbs_irq(dev_priv->engine[RCS0]);
 	if (gt_iir & GT_BSD_USER_INTERRUPT)
-		intel_engine_breadcrumbs_irq(dev_priv->engine[VCS]);
+		intel_engine_breadcrumbs_irq(dev_priv->engine[VCS0]);
 	if (gt_iir & GT_BLT_USER_INTERRUPT)
-		intel_engine_breadcrumbs_irq(dev_priv->engine[BCS]);
+		intel_engine_breadcrumbs_irq(dev_priv->engine[BCS0]);
 
 	if (gt_iir & (GT_BLT_CS_ERROR_INTERRUPT |
 		      GT_BSD_CS_ERROR_INTERRUPT |
@@ -1449,7 +1517,7 @@ gen8_cs_irq_handler(struct intel_engine_cs *engine, u32 iir)
 
 	if (iir & GT_RENDER_USER_INTERRUPT) {
 		intel_engine_breadcrumbs_irq(engine);
-		tasklet |= USES_GUC_SUBMISSION(engine->i915);
+		tasklet |= intel_engine_needs_breadcrumb_tasklet(engine);
 	}
 
 	if (tasklet)
@@ -1459,12 +1527,12 @@ gen8_cs_irq_handler(struct intel_engine_cs *engine, u32 iir)
 static void gen8_gt_irq_ack(struct drm_i915_private *i915,
 			    u32 master_ctl, u32 gt_iir[4])
 {
-	void __iomem * const regs = i915->regs;
+	void __iomem * const regs = i915->uncore.regs;
 
 #define GEN8_GT_IRQS (GEN8_GT_RCS_IRQ | \
 		      GEN8_GT_BCS_IRQ | \
+		      GEN8_GT_VCS0_IRQ | \
 		      GEN8_GT_VCS1_IRQ | \
-		      GEN8_GT_VCS2_IRQ | \
 		      GEN8_GT_VECS_IRQ | \
 		      GEN8_GT_PM_IRQ | \
 		      GEN8_GT_GUC_IRQ)
@@ -1475,7 +1543,7 @@ static void gen8_gt_irq_ack(struct drm_i915_private *i915,
 			raw_reg_write(regs, GEN8_GT_IIR(0), gt_iir[0]);
 	}
 
-	if (master_ctl & (GEN8_GT_VCS1_IRQ | GEN8_GT_VCS2_IRQ)) {
+	if (master_ctl & (GEN8_GT_VCS0_IRQ | GEN8_GT_VCS1_IRQ)) {
 		gt_iir[1] = raw_reg_read(regs, GEN8_GT_IIR(1));
 		if (likely(gt_iir[1]))
 			raw_reg_write(regs, GEN8_GT_IIR(1), gt_iir[1]);
@@ -1498,21 +1566,21 @@ static void gen8_gt_irq_handler(struct drm_i915_private *i915,
 				u32 master_ctl, u32 gt_iir[4])
 {
 	if (master_ctl & (GEN8_GT_RCS_IRQ | GEN8_GT_BCS_IRQ)) {
-		gen8_cs_irq_handler(i915->engine[RCS],
+		gen8_cs_irq_handler(i915->engine[RCS0],
 				    gt_iir[0] >> GEN8_RCS_IRQ_SHIFT);
-		gen8_cs_irq_handler(i915->engine[BCS],
+		gen8_cs_irq_handler(i915->engine[BCS0],
 				    gt_iir[0] >> GEN8_BCS_IRQ_SHIFT);
 	}
 
-	if (master_ctl & (GEN8_GT_VCS1_IRQ | GEN8_GT_VCS2_IRQ)) {
-		gen8_cs_irq_handler(i915->engine[VCS],
+	if (master_ctl & (GEN8_GT_VCS0_IRQ | GEN8_GT_VCS1_IRQ)) {
+		gen8_cs_irq_handler(i915->engine[VCS0],
+				    gt_iir[1] >> GEN8_VCS0_IRQ_SHIFT);
+		gen8_cs_irq_handler(i915->engine[VCS1],
 				    gt_iir[1] >> GEN8_VCS1_IRQ_SHIFT);
-		gen8_cs_irq_handler(i915->engine[VCS2],
-				    gt_iir[1] >> GEN8_VCS2_IRQ_SHIFT);
 	}
 
 	if (master_ctl & GEN8_GT_VECS_IRQ) {
-		gen8_cs_irq_handler(i915->engine[VECS],
+		gen8_cs_irq_handler(i915->engine[VECS0],
 				    gt_iir[3] >> GEN8_VECS_IRQ_SHIFT);
 	}
 
@@ -1693,7 +1761,9 @@ static void display_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
 {
 	struct intel_pipe_crc *pipe_crc = &dev_priv->pipe_crc[pipe];
 	struct intel_crtc *crtc = intel_get_crtc_for_pipe(dev_priv, pipe);
-	u32 crcs[5];
+	u32 crcs[5] = { crc0, crc1, crc2, crc3, crc4 };
+
+	trace_intel_pipe_crc(crtc, crcs);
 
 	spin_lock(&pipe_crc->lock);
 	/*
@@ -1712,11 +1782,6 @@ static void display_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
 	}
 	spin_unlock(&pipe_crc->lock);
 
-	crcs[0] = crc0;
-	crcs[1] = crc1;
-	crcs[2] = crc2;
-	crcs[3] = crc3;
-	crcs[4] = crc4;
 	drm_crtc_add_crc_entry(&crtc->base, true,
 				drm_crtc_accurate_vblank_count(&crtc->base),
 				crcs);
@@ -1775,6 +1840,25 @@ static void i9xx_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
 /* The RPS events need forcewake, so we add them to a work queue and mask their
  * IMR bits until the work is done. Other interrupts can be processed without
  * the work queue. */
+static void gen11_rps_irq_handler(struct drm_i915_private *i915, u32 pm_iir)
+{
+	struct intel_rps *rps = &i915->gt_pm.rps;
+	const u32 events = i915->pm_rps_events & pm_iir;
+
+	lockdep_assert_held(&i915->irq_lock);
+
+	if (unlikely(!events))
+		return;
+
+	gen6_mask_pm_irq(i915, events);
+
+	if (!rps->interrupts_enabled)
+		return;
+
+	rps->pm_iir |= events;
+	schedule_work(&rps->work);
+}
+
 static void gen6_rps_irq_handler(struct drm_i915_private *dev_priv, u32 pm_iir)
 {
 	struct intel_rps *rps = &dev_priv->gt_pm.rps;
@@ -1792,13 +1876,11 @@ static void gen6_rps_irq_handler(struct drm_i915_private *dev_priv, u32 pm_iir)
 	if (INTEL_GEN(dev_priv) >= 8)
 		return;
 
-	if (HAS_VEBOX(dev_priv)) {
-		if (pm_iir & PM_VEBOX_USER_INTERRUPT)
-			intel_engine_breadcrumbs_irq(dev_priv->engine[VECS]);
+	if (pm_iir & PM_VEBOX_USER_INTERRUPT)
+		intel_engine_breadcrumbs_irq(dev_priv->engine[VECS0]);
 
-		if (pm_iir & PM_VEBOX_CS_ERROR_INTERRUPT)
-			DRM_DEBUG("Command parser error, pm_iir 0x%08x\n", pm_iir);
-	}
+	if (pm_iir & PM_VEBOX_CS_ERROR_INTERRUPT)
+		DRM_DEBUG("Command parser error, pm_iir 0x%08x\n", pm_iir);
 }
 
 static void gen9_guc_irq_handler(struct drm_i915_private *dev_priv, u32 gt_iir)
@@ -2667,6 +2749,25 @@ static void gen11_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 iir)
 		DRM_ERROR("Unexpected DE HPD interrupt 0x%08x\n", iir);
 }
 
+static u32 gen8_de_port_aux_mask(struct drm_i915_private *dev_priv)
+{
+	u32 mask = GEN8_AUX_CHANNEL_A;
+
+	if (INTEL_GEN(dev_priv) >= 9)
+		mask |= GEN9_AUX_CHANNEL_B |
+			GEN9_AUX_CHANNEL_C |
+			GEN9_AUX_CHANNEL_D;
+
+	if (IS_CNL_WITH_PORT_F(dev_priv))
+		mask |= CNL_AUX_CHANNEL_F;
+
+	if (INTEL_GEN(dev_priv) >= 11)
+		mask |= ICL_AUX_CHANNEL_E |
+			CNL_AUX_CHANNEL_F;
+
+	return mask;
+}
+
 static irqreturn_t
 gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
 {
@@ -2722,20 +2823,7 @@ gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
 			I915_WRITE(GEN8_DE_PORT_IIR, iir);
 			ret = IRQ_HANDLED;
 
-			tmp_mask = GEN8_AUX_CHANNEL_A;
-			if (INTEL_GEN(dev_priv) >= 9)
-				tmp_mask |= GEN9_AUX_CHANNEL_B |
-					    GEN9_AUX_CHANNEL_C |
-					    GEN9_AUX_CHANNEL_D;
-
-			if (INTEL_GEN(dev_priv) >= 11)
-				tmp_mask |= ICL_AUX_CHANNEL_E;
-
-			if (IS_CNL_WITH_PORT_F(dev_priv) ||
-			    INTEL_GEN(dev_priv) >= 11)
-				tmp_mask |= CNL_AUX_CHANNEL_F;
-
-			if (iir & tmp_mask) {
+			if (iir & gen8_de_port_aux_mask(dev_priv)) {
 				dp_aux_irq_handler(dev_priv);
 				found = true;
 			}
@@ -2816,11 +2904,9 @@ gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
 			I915_WRITE(SDEIIR, iir);
 			ret = IRQ_HANDLED;
 
-			if (HAS_PCH_ICP(dev_priv))
+			if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
 				icp_irq_handler(dev_priv, iir);
-			else if (HAS_PCH_SPT(dev_priv) ||
-				 HAS_PCH_KBP(dev_priv) ||
-				 HAS_PCH_CNP(dev_priv))
+			else if (INTEL_PCH_TYPE(dev_priv) >= PCH_SPT)
 				spt_irq_handler(dev_priv, iir);
 			else
 				cpt_irq_handler(dev_priv, iir);
@@ -2857,7 +2943,7 @@ static inline void gen8_master_intr_enable(void __iomem * const regs)
 static irqreturn_t gen8_irq_handler(int irq, void *arg)
 {
 	struct drm_i915_private *dev_priv = to_i915(arg);
-	void __iomem * const regs = dev_priv->regs;
+	void __iomem * const regs = dev_priv->uncore.regs;
 	u32 master_ctl;
 	u32 gt_iir[4];
 
@@ -2891,7 +2977,7 @@ static u32
 gen11_gt_engine_identity(struct drm_i915_private * const i915,
 			 const unsigned int bank, const unsigned int bit)
 {
-	void __iomem * const regs = i915->regs;
+	void __iomem * const regs = i915->uncore.regs;
 	u32 timeout_ts;
 	u32 ident;
 
@@ -2926,7 +3012,7 @@ gen11_other_irq_handler(struct drm_i915_private * const i915,
 			const u8 instance, const u16 iir)
 {
 	if (instance == OTHER_GTPM_INSTANCE)
-		return gen6_rps_irq_handler(i915, iir);
+		return gen11_rps_irq_handler(i915, iir);
 
 	WARN_ONCE(1, "unhandled other interrupt instance=0x%x, iir=0x%x\n",
 		  instance, iir);
@@ -2975,7 +3061,7 @@ static void
 gen11_gt_bank_handler(struct drm_i915_private * const i915,
 		      const unsigned int bank)
 {
-	void __iomem * const regs = i915->regs;
+	void __iomem * const regs = i915->uncore.regs;
 	unsigned long intr_dw;
 	unsigned int bit;
 
@@ -2983,14 +3069,8 @@ gen11_gt_bank_handler(struct drm_i915_private * const i915,
 
 	intr_dw = raw_reg_read(regs, GEN11_GT_INTR_DW(bank));
 
-	if (unlikely(!intr_dw)) {
-		DRM_ERROR("GT_INTR_DW%u blank!\n", bank);
-		return;
-	}
-
 	for_each_set_bit(bit, &intr_dw, 32) {
-		const u32 ident = gen11_gt_engine_identity(i915,
-							   bank, bit);
+		const u32 ident = gen11_gt_engine_identity(i915, bank, bit);
 
 		gen11_gt_identity_handler(i915, ident);
 	}
@@ -3018,7 +3098,7 @@ gen11_gt_irq_handler(struct drm_i915_private * const i915,
 static u32
 gen11_gu_misc_irq_ack(struct drm_i915_private *dev_priv, const u32 master_ctl)
 {
-	void __iomem * const regs = dev_priv->regs;
+	void __iomem * const regs = dev_priv->uncore.regs;
 	u32 iir;
 
 	if (!(master_ctl & GEN11_GU_MISC_IRQ))
@@ -3059,7 +3139,7 @@ static inline void gen11_master_intr_enable(void __iomem * const regs)
 static irqreturn_t gen11_irq_handler(int irq, void *arg)
 {
 	struct drm_i915_private * const i915 = to_i915(arg);
-	void __iomem * const regs = i915->regs;
+	void __iomem * const regs = i915->uncore.regs;
 	u32 master_ctl;
 	u32 gu_misc_iir;
 
@@ -3112,6 +3192,16 @@ static int i8xx_enable_vblank(struct drm_device *dev, unsigned int pipe)
 	return 0;
 }
 
+static int i945gm_enable_vblank(struct drm_device *dev, unsigned int pipe)
+{
+	struct drm_i915_private *dev_priv = to_i915(dev);
+
+	if (dev_priv->i945gm_vblank.enabled++ == 0)
+		schedule_work(&dev_priv->i945gm_vblank.work);
+
+	return i8xx_enable_vblank(dev, pipe);
+}
+
 static int i965_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
@@ -3176,6 +3266,16 @@ static void i8xx_disable_vblank(struct drm_device *dev, unsigned int pipe)
 	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 }
 
+static void i945gm_disable_vblank(struct drm_device *dev, unsigned int pipe)
+{
+	struct drm_i915_private *dev_priv = to_i915(dev);
+
+	i8xx_disable_vblank(dev, pipe);
+
+	if (--dev_priv->i945gm_vblank.enabled == 0)
+		schedule_work(&dev_priv->i945gm_vblank.work);
+}
+
 static void i965_disable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
@@ -3209,12 +3309,68 @@ static void gen8_disable_vblank(struct drm_device *dev, unsigned int pipe)
 	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 }
 
+static void i945gm_vblank_work_func(struct work_struct *work)
+{
+	struct drm_i915_private *dev_priv =
+		container_of(work, struct drm_i915_private, i945gm_vblank.work);
+
+	/*
+	 * Vblank interrupts fail to wake up the device from C3,
+	 * hence we want to prevent C3 usage while vblank interrupts
+	 * are enabled.
+	 */
+	pm_qos_update_request(&dev_priv->i945gm_vblank.pm_qos,
+			      READ_ONCE(dev_priv->i945gm_vblank.enabled) ?
+			      dev_priv->i945gm_vblank.c3_disable_latency :
+			      PM_QOS_DEFAULT_VALUE);
+}
+
+static int cstate_disable_latency(const char *name)
+{
+	const struct cpuidle_driver *drv;
+	int i;
+
+	drv = cpuidle_get_driver();
+	if (!drv)
+		return 0;
+
+	for (i = 0; i < drv->state_count; i++) {
+		const struct cpuidle_state *state = &drv->states[i];
+
+		if (!strcmp(state->name, name))
+			return state->exit_latency ?
+				state->exit_latency - 1 : 0;
+	}
+
+	return 0;
+}
+
+static void i945gm_vblank_work_init(struct drm_i915_private *dev_priv)
+{
+	INIT_WORK(&dev_priv->i945gm_vblank.work,
+		  i945gm_vblank_work_func);
+
+	dev_priv->i945gm_vblank.c3_disable_latency =
+		cstate_disable_latency("C3");
+	pm_qos_add_request(&dev_priv->i945gm_vblank.pm_qos,
+			   PM_QOS_CPU_DMA_LATENCY,
+			   PM_QOS_DEFAULT_VALUE);
+}
+
+static void i945gm_vblank_work_fini(struct drm_i915_private *dev_priv)
+{
+	cancel_work_sync(&dev_priv->i945gm_vblank.work);
+	pm_qos_remove_request(&dev_priv->i945gm_vblank.pm_qos);
+}
+
 static void ibx_irq_reset(struct drm_i915_private *dev_priv)
 {
+	struct intel_uncore *uncore = &dev_priv->uncore;
+
 	if (HAS_PCH_NOP(dev_priv))
 		return;
 
-	GEN3_IRQ_RESET(SDE);
+	GEN3_IRQ_RESET(uncore, SDE);
 
 	if (HAS_PCH_CPT(dev_priv) || HAS_PCH_LPT(dev_priv))
 		I915_WRITE(SERR_INT, 0xffffffff);
@@ -3242,13 +3398,17 @@ static void ibx_irq_pre_postinstall(struct drm_device *dev)
 
 static void gen5_gt_irq_reset(struct drm_i915_private *dev_priv)
 {
-	GEN3_IRQ_RESET(GT);
+	struct intel_uncore *uncore = &dev_priv->uncore;
+
+	GEN3_IRQ_RESET(uncore, GT);
 	if (INTEL_GEN(dev_priv) >= 6)
-		GEN3_IRQ_RESET(GEN6_PM);
+		GEN3_IRQ_RESET(uncore, GEN6_PM);
 }
 
 static void vlv_display_irq_reset(struct drm_i915_private *dev_priv)
 {
+	struct intel_uncore *uncore = &dev_priv->uncore;
+
 	if (IS_CHERRYVIEW(dev_priv))
 		I915_WRITE(DPINVGTT, DPINVGTT_STATUS_MASK_CHV);
 	else
@@ -3259,12 +3419,14 @@ static void vlv_display_irq_reset(struct drm_i915_private *dev_priv)
 
 	i9xx_pipestat_irq_reset(dev_priv);
 
-	GEN3_IRQ_RESET(VLV_);
+	GEN3_IRQ_RESET(uncore, VLV_);
 	dev_priv->irq_mask = ~0u;
 }
 
 static void vlv_display_irq_postinstall(struct drm_i915_private *dev_priv)
 {
+	struct intel_uncore *uncore = &dev_priv->uncore;
+
 	u32 pipestat_mask;
 	u32 enable_mask;
 	enum pipe pipe;
@@ -3289,7 +3451,7 @@ static void vlv_display_irq_postinstall(struct drm_i915_private *dev_priv)
 
 	dev_priv->irq_mask = ~enable_mask;
 
-	GEN3_IRQ_INIT(VLV_, dev_priv->irq_mask, enable_mask);
+	GEN3_IRQ_INIT(uncore, VLV_, dev_priv->irq_mask, enable_mask);
 }
 
 /* drm_dma.h hooks
@@ -3297,8 +3459,9 @@ static void vlv_display_irq_postinstall(struct drm_i915_private *dev_priv)
 static void ironlake_irq_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_uncore *uncore = &dev_priv->uncore;
 
-	GEN3_IRQ_RESET(DE);
+	GEN3_IRQ_RESET(uncore, DE);
 	if (IS_GEN(dev_priv, 7))
 		I915_WRITE(GEN7_ERR_INT, 0xffffffff);
 
@@ -3329,18 +3492,21 @@ static void valleyview_irq_reset(struct drm_device *dev)
 
 static void gen8_gt_irq_reset(struct drm_i915_private *dev_priv)
 {
-	GEN8_IRQ_RESET_NDX(GT, 0);
-	GEN8_IRQ_RESET_NDX(GT, 1);
-	GEN8_IRQ_RESET_NDX(GT, 2);
-	GEN8_IRQ_RESET_NDX(GT, 3);
+	struct intel_uncore *uncore = &dev_priv->uncore;
+
+	GEN8_IRQ_RESET_NDX(uncore, GT, 0);
+	GEN8_IRQ_RESET_NDX(uncore, GT, 1);
+	GEN8_IRQ_RESET_NDX(uncore, GT, 2);
+	GEN8_IRQ_RESET_NDX(uncore, GT, 3);
 }
 
 static void gen8_irq_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_uncore *uncore = &dev_priv->uncore;
 	int pipe;
 
-	gen8_master_intr_disable(dev_priv->regs);
+	gen8_master_intr_disable(dev_priv->uncore.regs);
 
 	gen8_gt_irq_reset(dev_priv);
 
@@ -3350,11 +3516,11 @@ static void gen8_irq_reset(struct drm_device *dev)
 	for_each_pipe(dev_priv, pipe)
 		if (intel_display_power_is_enabled(dev_priv,
 						   POWER_DOMAIN_PIPE(pipe)))
-			GEN8_IRQ_RESET_NDX(DE_PIPE, pipe);
+			GEN8_IRQ_RESET_NDX(uncore, DE_PIPE, pipe);
 
-	GEN3_IRQ_RESET(GEN8_DE_PORT_);
-	GEN3_IRQ_RESET(GEN8_DE_MISC_);
-	GEN3_IRQ_RESET(GEN8_PCU_);
+	GEN3_IRQ_RESET(uncore, GEN8_DE_PORT_);
+	GEN3_IRQ_RESET(uncore, GEN8_DE_MISC_);
+	GEN3_IRQ_RESET(uncore, GEN8_PCU_);
 
 	if (HAS_PCH_SPLIT(dev_priv))
 		ibx_irq_reset(dev_priv);
@@ -3380,9 +3546,10 @@ static void gen11_gt_irq_reset(struct drm_i915_private *dev_priv)
 static void gen11_irq_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_uncore *uncore = &dev_priv->uncore;
 	int pipe;
 
-	gen11_master_intr_disable(dev_priv->regs);
+	gen11_master_intr_disable(dev_priv->uncore.regs);
 
 	gen11_gt_irq_reset(dev_priv);
 
@@ -3394,21 +3561,23 @@ static void gen11_irq_reset(struct drm_device *dev)
 	for_each_pipe(dev_priv, pipe)
 		if (intel_display_power_is_enabled(dev_priv,
 						   POWER_DOMAIN_PIPE(pipe)))
-			GEN8_IRQ_RESET_NDX(DE_PIPE, pipe);
+			GEN8_IRQ_RESET_NDX(uncore, DE_PIPE, pipe);
 
-	GEN3_IRQ_RESET(GEN8_DE_PORT_);
-	GEN3_IRQ_RESET(GEN8_DE_MISC_);
-	GEN3_IRQ_RESET(GEN11_DE_HPD_);
-	GEN3_IRQ_RESET(GEN11_GU_MISC_);
-	GEN3_IRQ_RESET(GEN8_PCU_);
+	GEN3_IRQ_RESET(uncore, GEN8_DE_PORT_);
+	GEN3_IRQ_RESET(uncore, GEN8_DE_MISC_);
+	GEN3_IRQ_RESET(uncore, GEN11_DE_HPD_);
+	GEN3_IRQ_RESET(uncore, GEN11_GU_MISC_);
+	GEN3_IRQ_RESET(uncore, GEN8_PCU_);
 
-	if (HAS_PCH_ICP(dev_priv))
-		GEN3_IRQ_RESET(SDE);
+	if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
+		GEN3_IRQ_RESET(uncore, SDE);
 }
 
 void gen8_irq_power_well_post_enable(struct drm_i915_private *dev_priv,
 				     u8 pipe_mask)
 {
+	struct intel_uncore *uncore = &dev_priv->uncore;
+
 	u32 extra_ier = GEN8_PIPE_VBLANK | GEN8_PIPE_FIFO_UNDERRUN;
 	enum pipe pipe;
 
@@ -3420,7 +3589,7 @@ void gen8_irq_power_well_post_enable(struct drm_i915_private *dev_priv,
 	}
 
 	for_each_pipe_masked(dev_priv, pipe, pipe_mask)
-		GEN8_IRQ_INIT_NDX(DE_PIPE, pipe,
+		GEN8_IRQ_INIT_NDX(uncore, DE_PIPE, pipe,
 				  dev_priv->de_irq_mask[pipe],
 				  ~dev_priv->de_irq_mask[pipe] | extra_ier);
 
@@ -3430,6 +3599,7 @@ void gen8_irq_power_well_post_enable(struct drm_i915_private *dev_priv,
 void gen8_irq_power_well_pre_disable(struct drm_i915_private *dev_priv,
 				     u8 pipe_mask)
 {
+	struct intel_uncore *uncore = &dev_priv->uncore;
 	enum pipe pipe;
 
 	spin_lock_irq(&dev_priv->irq_lock);
@@ -3440,7 +3610,7 @@ void gen8_irq_power_well_pre_disable(struct drm_i915_private *dev_priv,
 	}
 
 	for_each_pipe_masked(dev_priv, pipe, pipe_mask)
-		GEN8_IRQ_RESET_NDX(DE_PIPE, pipe);
+		GEN8_IRQ_RESET_NDX(uncore, DE_PIPE, pipe);
 
 	spin_unlock_irq(&dev_priv->irq_lock);
 
@@ -3451,13 +3621,14 @@ void gen8_irq_power_well_pre_disable(struct drm_i915_private *dev_priv,
 static void cherryview_irq_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_uncore *uncore = &dev_priv->uncore;
 
 	I915_WRITE(GEN8_MASTER_IRQ, 0);
 	POSTING_READ(GEN8_MASTER_IRQ);
 
 	gen8_gt_irq_reset(dev_priv);
 
-	GEN3_IRQ_RESET(GEN8_PCU_);
+	GEN3_IRQ_RESET(uncore, GEN8_PCU_);
 
 	spin_lock_irq(&dev_priv->irq_lock);
 	if (dev_priv->display_irqs_enabled)
@@ -3583,7 +3754,7 @@ static void gen11_hpd_irq_setup(struct drm_i915_private *dev_priv)
 
 	gen11_hpd_detection_setup(dev_priv);
 
-	if (HAS_PCH_ICP(dev_priv))
+	if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
 		icp_hpd_irq_setup(dev_priv);
 }
 
@@ -3729,7 +3900,7 @@ static void ibx_irq_postinstall(struct drm_device *dev)
 	else
 		mask = SDE_GMBUS_CPT;
 
-	gen3_assert_iir_is_zero(dev_priv, SDEIIR);
+	gen3_assert_iir_is_zero(&dev_priv->uncore, SDEIIR);
 	I915_WRITE(SDEIMR, ~mask);
 
 	if (HAS_PCH_IBX(dev_priv) || HAS_PCH_CPT(dev_priv) ||
@@ -3742,6 +3913,7 @@ static void ibx_irq_postinstall(struct drm_device *dev)
 static void gen5_gt_irq_postinstall(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_uncore *uncore = &dev_priv->uncore;
 	u32 pm_irqs, gt_irqs;
 
 	pm_irqs = gt_irqs = 0;
@@ -3760,26 +3932,27 @@ static void gen5_gt_irq_postinstall(struct drm_device *dev)
 		gt_irqs |= GT_BLT_USER_INTERRUPT | GT_BSD_USER_INTERRUPT;
 	}
 
-	GEN3_IRQ_INIT(GT, dev_priv->gt_irq_mask, gt_irqs);
+	GEN3_IRQ_INIT(uncore, GT, dev_priv->gt_irq_mask, gt_irqs);
 
 	if (INTEL_GEN(dev_priv) >= 6) {
 		/*
 		 * RPS interrupts will get enabled/disabled on demand when RPS
 		 * itself is enabled/disabled.
 		 */
-		if (HAS_VEBOX(dev_priv)) {
+		if (HAS_ENGINE(dev_priv, VECS0)) {
 			pm_irqs |= PM_VEBOX_USER_INTERRUPT;
 			dev_priv->pm_ier |= PM_VEBOX_USER_INTERRUPT;
 		}
 
 		dev_priv->pm_imr = 0xffffffff;
-		GEN3_IRQ_INIT(GEN6_PM, dev_priv->pm_imr, pm_irqs);
+		GEN3_IRQ_INIT(uncore, GEN6_PM, dev_priv->pm_imr, pm_irqs);
 	}
 }
 
 static int ironlake_irq_postinstall(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_uncore *uncore = &dev_priv->uncore;
 	u32 display_mask, extra_mask;
 
 	if (INTEL_GEN(dev_priv) >= 7) {
@@ -3798,7 +3971,7 @@ static int ironlake_irq_postinstall(struct drm_device *dev)
 	}
 
 	if (IS_HASWELL(dev_priv)) {
-		gen3_assert_iir_is_zero(dev_priv, EDP_PSR_IIR);
+		gen3_assert_iir_is_zero(uncore, EDP_PSR_IIR);
 		intel_psr_irq_control(dev_priv, dev_priv->psr.debug);
 		display_mask |= DE_EDP_PSR_INT_HSW;
 	}
@@ -3807,7 +3980,8 @@ static int ironlake_irq_postinstall(struct drm_device *dev)
 
 	ibx_irq_pre_postinstall(dev);
 
-	GEN3_IRQ_INIT(DE, dev_priv->irq_mask, display_mask | extra_mask);
+	GEN3_IRQ_INIT(uncore, DE, dev_priv->irq_mask,
+		      display_mask | extra_mask);
 
 	gen5_gt_irq_postinstall(dev);
 
@@ -3877,35 +4051,42 @@ static int valleyview_irq_postinstall(struct drm_device *dev)
 
 static void gen8_gt_irq_postinstall(struct drm_i915_private *dev_priv)
 {
+	struct intel_uncore *uncore = &dev_priv->uncore;
+
 	/* These are interrupts we'll toggle with the ring mask register */
 	u32 gt_interrupts[] = {
-		GT_RENDER_USER_INTERRUPT << GEN8_RCS_IRQ_SHIFT |
-			GT_CONTEXT_SWITCH_INTERRUPT << GEN8_RCS_IRQ_SHIFT |
-			GT_RENDER_USER_INTERRUPT << GEN8_BCS_IRQ_SHIFT |
-			GT_CONTEXT_SWITCH_INTERRUPT << GEN8_BCS_IRQ_SHIFT,
-		GT_RENDER_USER_INTERRUPT << GEN8_VCS1_IRQ_SHIFT |
-			GT_CONTEXT_SWITCH_INTERRUPT << GEN8_VCS1_IRQ_SHIFT |
-			GT_RENDER_USER_INTERRUPT << GEN8_VCS2_IRQ_SHIFT |
-			GT_CONTEXT_SWITCH_INTERRUPT << GEN8_VCS2_IRQ_SHIFT,
+		(GT_RENDER_USER_INTERRUPT << GEN8_RCS_IRQ_SHIFT |
+		 GT_CONTEXT_SWITCH_INTERRUPT << GEN8_RCS_IRQ_SHIFT |
+		 GT_RENDER_USER_INTERRUPT << GEN8_BCS_IRQ_SHIFT |
+		 GT_CONTEXT_SWITCH_INTERRUPT << GEN8_BCS_IRQ_SHIFT),
+
+		(GT_RENDER_USER_INTERRUPT << GEN8_VCS0_IRQ_SHIFT |
+		 GT_CONTEXT_SWITCH_INTERRUPT << GEN8_VCS0_IRQ_SHIFT |
+		 GT_RENDER_USER_INTERRUPT << GEN8_VCS1_IRQ_SHIFT |
+		 GT_CONTEXT_SWITCH_INTERRUPT << GEN8_VCS1_IRQ_SHIFT),
+
 		0,
-		GT_RENDER_USER_INTERRUPT << GEN8_VECS_IRQ_SHIFT |
-			GT_CONTEXT_SWITCH_INTERRUPT << GEN8_VECS_IRQ_SHIFT
-		};
+
+		(GT_RENDER_USER_INTERRUPT << GEN8_VECS_IRQ_SHIFT |
+		 GT_CONTEXT_SWITCH_INTERRUPT << GEN8_VECS_IRQ_SHIFT)
+	};
 
 	dev_priv->pm_ier = 0x0;
 	dev_priv->pm_imr = ~dev_priv->pm_ier;
-	GEN8_IRQ_INIT_NDX(GT, 0, ~gt_interrupts[0], gt_interrupts[0]);
-	GEN8_IRQ_INIT_NDX(GT, 1, ~gt_interrupts[1], gt_interrupts[1]);
+	GEN8_IRQ_INIT_NDX(uncore, GT, 0, ~gt_interrupts[0], gt_interrupts[0]);
+	GEN8_IRQ_INIT_NDX(uncore, GT, 1, ~gt_interrupts[1], gt_interrupts[1]);
 	/*
 	 * RPS interrupts will get enabled/disabled on demand when RPS itself
 	 * is enabled/disabled. Same wil be the case for GuC interrupts.
 	 */
-	GEN8_IRQ_INIT_NDX(GT, 2, dev_priv->pm_imr, dev_priv->pm_ier);
-	GEN8_IRQ_INIT_NDX(GT, 3, ~gt_interrupts[3], gt_interrupts[3]);
+	GEN8_IRQ_INIT_NDX(uncore, GT, 2, dev_priv->pm_imr, dev_priv->pm_ier);
+	GEN8_IRQ_INIT_NDX(uncore, GT, 3, ~gt_interrupts[3], gt_interrupts[3]);
 }
 
 static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 {
+	struct intel_uncore *uncore = &dev_priv->uncore;
+
 	u32 de_pipe_masked = GEN8_PIPE_CDCLK_CRC_DONE;
 	u32 de_pipe_enables;
 	u32 de_port_masked = GEN8_AUX_CHANNEL_A;
@@ -3941,7 +4122,7 @@ static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 	else if (IS_BROADWELL(dev_priv))
 		de_port_enables |= GEN8_PORT_DP_A_HOTPLUG;
 
-	gen3_assert_iir_is_zero(dev_priv, EDP_PSR_IIR);
+	gen3_assert_iir_is_zero(uncore, EDP_PSR_IIR);
 	intel_psr_irq_control(dev_priv, dev_priv->psr.debug);
 
 	for_each_pipe(dev_priv, pipe) {
@@ -3949,20 +4130,21 @@ static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 
 		if (intel_display_power_is_enabled(dev_priv,
 				POWER_DOMAIN_PIPE(pipe)))
-			GEN8_IRQ_INIT_NDX(DE_PIPE, pipe,
+			GEN8_IRQ_INIT_NDX(uncore, DE_PIPE, pipe,
 					  dev_priv->de_irq_mask[pipe],
 					  de_pipe_enables);
 	}
 
-	GEN3_IRQ_INIT(GEN8_DE_PORT_, ~de_port_masked, de_port_enables);
-	GEN3_IRQ_INIT(GEN8_DE_MISC_, ~de_misc_masked, de_misc_masked);
+	GEN3_IRQ_INIT(uncore, GEN8_DE_PORT_, ~de_port_masked, de_port_enables);
+	GEN3_IRQ_INIT(uncore, GEN8_DE_MISC_, ~de_misc_masked, de_misc_masked);
 
 	if (INTEL_GEN(dev_priv) >= 11) {
 		u32 de_hpd_masked = 0;
 		u32 de_hpd_enables = GEN11_DE_TC_HOTPLUG_MASK |
 				     GEN11_DE_TBT_HOTPLUG_MASK;
 
-		GEN3_IRQ_INIT(GEN11_DE_HPD_, ~de_hpd_masked, de_hpd_enables);
+		GEN3_IRQ_INIT(uncore, GEN11_DE_HPD_, ~de_hpd_masked,
+			      de_hpd_enables);
 		gen11_hpd_detection_setup(dev_priv);
 	} else if (IS_GEN9_LP(dev_priv)) {
 		bxt_hpd_detection_setup(dev_priv);
@@ -3984,7 +4166,7 @@ static int gen8_irq_postinstall(struct drm_device *dev)
 	if (HAS_PCH_SPLIT(dev_priv))
 		ibx_irq_postinstall(dev);
 
-	gen8_master_intr_enable(dev_priv->regs);
+	gen8_master_intr_enable(dev_priv->uncore.regs);
 
 	return 0;
 }
@@ -4025,7 +4207,7 @@ static void icp_irq_postinstall(struct drm_device *dev)
 	I915_WRITE(SDEIER, 0xffffffff);
 	POSTING_READ(SDEIER);
 
-	gen3_assert_iir_is_zero(dev_priv, SDEIIR);
+	gen3_assert_iir_is_zero(&dev_priv->uncore, SDEIIR);
 	I915_WRITE(SDEIMR, ~mask);
 
 	icp_hpd_detection_setup(dev_priv);
@@ -4034,19 +4216,20 @@ static void icp_irq_postinstall(struct drm_device *dev)
 static int gen11_irq_postinstall(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_uncore *uncore = &dev_priv->uncore;
 	u32 gu_misc_masked = GEN11_GU_MISC_GSE;
 
-	if (HAS_PCH_ICP(dev_priv))
+	if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
 		icp_irq_postinstall(dev);
 
 	gen11_gt_irq_postinstall(dev_priv);
 	gen8_de_irq_postinstall(dev_priv);
 
-	GEN3_IRQ_INIT(GEN11_GU_MISC_, ~gu_misc_masked, gu_misc_masked);
+	GEN3_IRQ_INIT(uncore, GEN11_GU_MISC_, ~gu_misc_masked, gu_misc_masked);
 
 	I915_WRITE(GEN11_DISPLAY_INT_CTL, GEN11_DISPLAY_IRQ_ENABLE);
 
-	gen11_master_intr_enable(dev_priv->regs);
+	gen11_master_intr_enable(dev_priv->uncore.regs);
 	POSTING_READ(GEN11_GFX_MSTR_IRQ);
 
 	return 0;
@@ -4072,15 +4255,17 @@ static int cherryview_irq_postinstall(struct drm_device *dev)
 static void i8xx_irq_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_uncore *uncore = &dev_priv->uncore;
 
 	i9xx_pipestat_irq_reset(dev_priv);
 
-	GEN2_IRQ_RESET();
+	GEN2_IRQ_RESET(uncore);
 }
 
 static int i8xx_irq_postinstall(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_uncore *uncore = &dev_priv->uncore;
 	u16 enable_mask;
 
 	I915_WRITE16(EMR, ~(I915_ERROR_PAGE_TABLE |
@@ -4098,7 +4283,7 @@ static int i8xx_irq_postinstall(struct drm_device *dev)
 		I915_MASTER_ERROR_INTERRUPT |
 		I915_USER_INTERRUPT;
 
-	GEN2_IRQ_INIT(, dev_priv->irq_mask, enable_mask);
+	GEN2_IRQ_INIT(uncore, dev_priv->irq_mask, enable_mask);
 
 	/* Interrupt setup is already guaranteed to be single-threaded, this is
 	 * just to make the assert_spin_locked check happy. */
@@ -4202,7 +4387,7 @@ static irqreturn_t i8xx_irq_handler(int irq, void *arg)
 		u16 eir = 0, eir_stuck = 0;
 		u16 iir;
 
-		iir = I915_READ16(IIR);
+		iir = I915_READ16(GEN2_IIR);
 		if (iir == 0)
 			break;
 
@@ -4215,10 +4400,10 @@ static irqreturn_t i8xx_irq_handler(int irq, void *arg)
 		if (iir & I915_MASTER_ERROR_INTERRUPT)
 			i8xx_error_irq_ack(dev_priv, &eir, &eir_stuck);
 
-		I915_WRITE16(IIR, iir);
+		I915_WRITE16(GEN2_IIR, iir);
 
 		if (iir & I915_USER_INTERRUPT)
-			intel_engine_breadcrumbs_irq(dev_priv->engine[RCS]);
+			intel_engine_breadcrumbs_irq(dev_priv->engine[RCS0]);
 
 		if (iir & I915_MASTER_ERROR_INTERRUPT)
 			i8xx_error_irq_handler(dev_priv, eir, eir_stuck);
@@ -4234,6 +4419,7 @@ static irqreturn_t i8xx_irq_handler(int irq, void *arg)
 static void i915_irq_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_uncore *uncore = &dev_priv->uncore;
 
 	if (I915_HAS_HOTPLUG(dev_priv)) {
 		i915_hotplug_interrupt_update(dev_priv, 0xffffffff, 0);
@@ -4242,12 +4428,13 @@ static void i915_irq_reset(struct drm_device *dev)
 
 	i9xx_pipestat_irq_reset(dev_priv);
 
-	GEN3_IRQ_RESET();
+	GEN3_IRQ_RESET(uncore, GEN2_);
 }
 
 static int i915_irq_postinstall(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_uncore *uncore = &dev_priv->uncore;
 	u32 enable_mask;
 
 	I915_WRITE(EMR, ~(I915_ERROR_PAGE_TABLE |
@@ -4274,7 +4461,7 @@ static int i915_irq_postinstall(struct drm_device *dev)
 		dev_priv->irq_mask &= ~I915_DISPLAY_PORT_INTERRUPT;
 	}
 
-	GEN3_IRQ_INIT(, dev_priv->irq_mask, enable_mask);
+	GEN3_IRQ_INIT(uncore, GEN2_, dev_priv->irq_mask, enable_mask);
 
 	/* Interrupt setup is already guaranteed to be single-threaded, this is
 	 * just to make the assert_spin_locked check happy. */
@@ -4306,7 +4493,7 @@ static irqreturn_t i915_irq_handler(int irq, void *arg)
 		u32 hotplug_status = 0;
 		u32 iir;
 
-		iir = I915_READ(IIR);
+		iir = I915_READ(GEN2_IIR);
 		if (iir == 0)
 			break;
 
@@ -4323,10 +4510,10 @@ static irqreturn_t i915_irq_handler(int irq, void *arg)
 		if (iir & I915_MASTER_ERROR_INTERRUPT)
 			i9xx_error_irq_ack(dev_priv, &eir, &eir_stuck);
 
-		I915_WRITE(IIR, iir);
+		I915_WRITE(GEN2_IIR, iir);
 
 		if (iir & I915_USER_INTERRUPT)
-			intel_engine_breadcrumbs_irq(dev_priv->engine[RCS]);
+			intel_engine_breadcrumbs_irq(dev_priv->engine[RCS0]);
 
 		if (iir & I915_MASTER_ERROR_INTERRUPT)
 			i9xx_error_irq_handler(dev_priv, eir, eir_stuck);
@@ -4345,18 +4532,20 @@ static irqreturn_t i915_irq_handler(int irq, void *arg)
 static void i965_irq_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_uncore *uncore = &dev_priv->uncore;
 
 	i915_hotplug_interrupt_update(dev_priv, 0xffffffff, 0);
 	I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT));
 
 	i9xx_pipestat_irq_reset(dev_priv);
 
-	GEN3_IRQ_RESET();
+	GEN3_IRQ_RESET(uncore, GEN2_);
 }
 
 static int i965_irq_postinstall(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_uncore *uncore = &dev_priv->uncore;
 	u32 enable_mask;
 	u32 error_mask;
 
@@ -4394,7 +4583,7 @@ static int i965_irq_postinstall(struct drm_device *dev)
 	if (IS_G4X(dev_priv))
 		enable_mask |= I915_BSD_USER_INTERRUPT;
 
-	GEN3_IRQ_INIT(, dev_priv->irq_mask, enable_mask);
+	GEN3_IRQ_INIT(uncore, GEN2_, dev_priv->irq_mask, enable_mask);
 
 	/* Interrupt setup is already guaranteed to be single-threaded, this is
 	 * just to make the assert_spin_locked check happy. */
@@ -4452,7 +4641,7 @@ static irqreturn_t i965_irq_handler(int irq, void *arg)
 		u32 hotplug_status = 0;
 		u32 iir;
 
-		iir = I915_READ(IIR);
+		iir = I915_READ(GEN2_IIR);
 		if (iir == 0)
 			break;
 
@@ -4468,13 +4657,13 @@ static irqreturn_t i965_irq_handler(int irq, void *arg)
 		if (iir & I915_MASTER_ERROR_INTERRUPT)
 			i9xx_error_irq_ack(dev_priv, &eir, &eir_stuck);
 
-		I915_WRITE(IIR, iir);
+		I915_WRITE(GEN2_IIR, iir);
 
 		if (iir & I915_USER_INTERRUPT)
-			intel_engine_breadcrumbs_irq(dev_priv->engine[RCS]);
+			intel_engine_breadcrumbs_irq(dev_priv->engine[RCS0]);
 
 		if (iir & I915_BSD_USER_INTERRUPT)
-			intel_engine_breadcrumbs_irq(dev_priv->engine[VCS]);
+			intel_engine_breadcrumbs_irq(dev_priv->engine[VCS0]);
 
 		if (iir & I915_MASTER_ERROR_INTERRUPT)
 			i9xx_error_irq_handler(dev_priv, eir, eir_stuck);
@@ -4503,6 +4692,9 @@ void intel_irq_init(struct drm_i915_private *dev_priv)
 	struct intel_rps *rps = &dev_priv->gt_pm.rps;
 	int i;
 
+	if (IS_I945GM(dev_priv))
+		i945gm_vblank_work_init(dev_priv);
+
 	intel_hpd_init_work(dev_priv);
 
 	INIT_WORK(&rps->work, gen6_pm_rps_work);
@@ -4523,6 +4715,10 @@ void intel_irq_init(struct drm_i915_private *dev_priv)
 					   GEN6_PM_RP_DOWN_THRESHOLD |
 					   GEN6_PM_RP_DOWN_TIMEOUT);
 
+	/* We share the register with other engine */
+	if (INTEL_GEN(dev_priv) > 9)
+		GEM_WARN_ON(dev_priv->pm_rps_events & 0xffff0000);
+
 	rps->pm_intrmsk_mbz = 0;
 
 	/*
@@ -4542,13 +4738,7 @@ void intel_irq_init(struct drm_i915_private *dev_priv)
 	else if (INTEL_GEN(dev_priv) >= 3)
 		dev->driver->get_vblank_counter = i915_get_vblank_counter;
 
-	/*
-	 * Opt out of the vblank disable timer on everything except gen2.
-	 * Gen2 doesn't have a hardware frame counter and so depends on
-	 * vblank interrupts to produce sane vblank seuquence numbers.
-	 */
-	if (!IS_GEN(dev_priv, 2))
-		dev->vblank_disable_immediate = true;
+	dev->vblank_disable_immediate = true;
 
 	/* Most platforms treat the display irq block as an always-on
 	 * power domain. vlv/chv can disable it at runtime and need
@@ -4605,8 +4795,7 @@ void intel_irq_init(struct drm_i915_private *dev_priv)
 		dev->driver->disable_vblank = gen8_disable_vblank;
 		if (IS_GEN9_LP(dev_priv))
 			dev_priv->display.hpd_irq_setup = bxt_hpd_irq_setup;
-		else if (HAS_PCH_SPT(dev_priv) || HAS_PCH_KBP(dev_priv) ||
-			 HAS_PCH_CNP(dev_priv))
+		else if (INTEL_PCH_TYPE(dev_priv) >= PCH_SPT)
 			dev_priv->display.hpd_irq_setup = spt_hpd_irq_setup;
 		else
 			dev_priv->display.hpd_irq_setup = ilk_hpd_irq_setup;
@@ -4626,6 +4815,13 @@ void intel_irq_init(struct drm_i915_private *dev_priv)
 			dev->driver->irq_uninstall = i8xx_irq_reset;
 			dev->driver->enable_vblank = i8xx_enable_vblank;
 			dev->driver->disable_vblank = i8xx_disable_vblank;
+		} else if (IS_I945GM(dev_priv)) {
+			dev->driver->irq_preinstall = i915_irq_reset;
+			dev->driver->irq_postinstall = i915_irq_postinstall;
+			dev->driver->irq_uninstall = i915_irq_reset;
+			dev->driver->irq_handler = i915_irq_handler;
+			dev->driver->enable_vblank = i945gm_enable_vblank;
+			dev->driver->disable_vblank = i945gm_disable_vblank;
 		} else if (IS_GEN(dev_priv, 3)) {
 			dev->driver->irq_preinstall = i915_irq_reset;
 			dev->driver->irq_postinstall = i915_irq_postinstall;
@@ -4656,6 +4852,9 @@ void intel_irq_fini(struct drm_i915_private *i915)
 {
 	int i;
 
+	if (IS_I945GM(i915))
+		i945gm_vblank_work_fini(i915);
+
 	for (i = 0; i < MAX_L3_SLICES; ++i)
 		kfree(i915->l3_parity.remap_info[i]);
 }
diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c
index 66f82f3f050f..f893c2cbce15 100644
--- a/drivers/gpu/drm/i915/i915_pci.c
+++ b/drivers/gpu/drm/i915/i915_pci.c
@@ -28,14 +28,45 @@
 
 #include <drm/drm_drv.h>
 
-#include "i915_active.h"
 #include "i915_drv.h"
+#include "i915_globals.h"
 #include "i915_selftest.h"
+#include "intel_fbdev.h"
 
-#define PLATFORM(x) .platform = (x), .platform_mask = BIT(x)
+#define PLATFORM(x) .platform = (x)
 #define GEN(x) .gen = (x), .gen_mask = BIT((x) - 1)
 
-#define GEN_DEFAULT_PIPEOFFSETS \
+#define I845_PIPE_OFFSETS \
+	.pipe_offsets = { \
+		[TRANSCODER_A] = PIPE_A_OFFSET,	\
+	}, \
+	.trans_offsets = { \
+		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
+	}
+
+#define I9XX_PIPE_OFFSETS \
+	.pipe_offsets = { \
+		[TRANSCODER_A] = PIPE_A_OFFSET,	\
+		[TRANSCODER_B] = PIPE_B_OFFSET, \
+	}, \
+	.trans_offsets = { \
+		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
+		[TRANSCODER_B] = TRANSCODER_B_OFFSET, \
+	}
+
+#define IVB_PIPE_OFFSETS \
+	.pipe_offsets = { \
+		[TRANSCODER_A] = PIPE_A_OFFSET,	\
+		[TRANSCODER_B] = PIPE_B_OFFSET, \
+		[TRANSCODER_C] = PIPE_C_OFFSET, \
+	}, \
+	.trans_offsets = { \
+		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
+		[TRANSCODER_B] = TRANSCODER_B_OFFSET, \
+		[TRANSCODER_C] = TRANSCODER_C_OFFSET, \
+	}
+
+#define HSW_PIPE_OFFSETS \
 	.pipe_offsets = { \
 		[TRANSCODER_A] = PIPE_A_OFFSET,	\
 		[TRANSCODER_B] = PIPE_B_OFFSET, \
@@ -49,7 +80,7 @@
 		[TRANSCODER_EDP] = TRANSCODER_EDP_OFFSET, \
 	}
 
-#define GEN_CHV_PIPEOFFSETS \
+#define CHV_PIPE_OFFSETS \
 	.pipe_offsets = { \
 		[TRANSCODER_A] = PIPE_A_OFFSET, \
 		[TRANSCODER_B] = PIPE_B_OFFSET, \
@@ -61,21 +92,48 @@
 		[TRANSCODER_C] = CHV_TRANSCODER_C_OFFSET, \
 	}
 
-#define CURSOR_OFFSETS \
-	.cursor_offsets = { CURSOR_A_OFFSET, CURSOR_B_OFFSET, CHV_CURSOR_C_OFFSET }
+#define I845_CURSOR_OFFSETS \
+	.cursor_offsets = { \
+		[PIPE_A] = CURSOR_A_OFFSET, \
+	}
+
+#define I9XX_CURSOR_OFFSETS \
+	.cursor_offsets = { \
+		[PIPE_A] = CURSOR_A_OFFSET, \
+		[PIPE_B] = CURSOR_B_OFFSET, \
+	}
+
+#define CHV_CURSOR_OFFSETS \
+	.cursor_offsets = { \
+		[PIPE_A] = CURSOR_A_OFFSET, \
+		[PIPE_B] = CURSOR_B_OFFSET, \
+		[PIPE_C] = CHV_CURSOR_C_OFFSET, \
+	}
 
 #define IVB_CURSOR_OFFSETS \
-	.cursor_offsets = { CURSOR_A_OFFSET, IVB_CURSOR_B_OFFSET, IVB_CURSOR_C_OFFSET }
+	.cursor_offsets = { \
+		[PIPE_A] = CURSOR_A_OFFSET, \
+		[PIPE_B] = IVB_CURSOR_B_OFFSET, \
+		[PIPE_C] = IVB_CURSOR_C_OFFSET, \
+	}
 
-#define BDW_COLORS \
-	.color = { .degamma_lut_size = 512, .gamma_lut_size = 512 }
+#define I9XX_COLORS \
+	.color = { .gamma_lut_size = 256 }
+#define I965_COLORS \
+	.color = { .gamma_lut_size = 129, \
+		   .gamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \
+	}
+#define ILK_COLORS \
+	.color = { .gamma_lut_size = 1024 }
+#define IVB_COLORS \
+	.color = { .degamma_lut_size = 1024, .gamma_lut_size = 1024 }
 #define CHV_COLORS \
 	.color = { .degamma_lut_size = 65, .gamma_lut_size = 257, \
 		   .degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \
 		   .gamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \
 	}
 #define GLK_COLORS \
-	.color = { .degamma_lut_size = 0, .gamma_lut_size = 1024, \
+	.color = { .degamma_lut_size = 33, .gamma_lut_size = 1024, \
 		   .degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING | \
 					DRM_COLOR_LUT_EQUAL_CHANNELS, \
 	}
@@ -85,7 +143,26 @@
 #define GEN_DEFAULT_PAGE_SIZES \
 	.page_sizes = I915_GTT_PAGE_SIZE_4K
 
-#define GEN2_FEATURES \
+#define I830_FEATURES \
+	GEN(2), \
+	.is_mobile = 1, \
+	.num_pipes = 2, \
+	.display.has_overlay = 1, \
+	.display.cursor_needs_physical = 1, \
+	.display.overlay_needs_physical = 1, \
+	.display.has_gmch = 1, \
+	.gpu_reset_clobbers_display = true, \
+	.hws_needs_physical = 1, \
+	.unfenced_needs_alignment = 1, \
+	.engine_mask = BIT(RCS0), \
+	.has_snoop = true, \
+	.has_coherent_ggtt = false, \
+	I9XX_PIPE_OFFSETS, \
+	I9XX_CURSOR_OFFSETS, \
+	I9XX_COLORS, \
+	GEN_DEFAULT_PAGE_SIZES
+
+#define I845_FEATURES \
 	GEN(2), \
 	.num_pipes = 1, \
 	.display.has_overlay = 1, \
@@ -94,37 +171,32 @@
 	.gpu_reset_clobbers_display = true, \
 	.hws_needs_physical = 1, \
 	.unfenced_needs_alignment = 1, \
-	.ring_mask = RENDER_RING, \
+	.engine_mask = BIT(RCS0), \
 	.has_snoop = true, \
 	.has_coherent_ggtt = false, \
-	GEN_DEFAULT_PIPEOFFSETS, \
-	GEN_DEFAULT_PAGE_SIZES, \
-	CURSOR_OFFSETS
+	I845_PIPE_OFFSETS, \
+	I845_CURSOR_OFFSETS, \
+	I9XX_COLORS, \
+	GEN_DEFAULT_PAGE_SIZES
 
 static const struct intel_device_info intel_i830_info = {
-	GEN2_FEATURES,
+	I830_FEATURES,
 	PLATFORM(INTEL_I830),
-	.is_mobile = 1,
-	.display.cursor_needs_physical = 1,
-	.num_pipes = 2, /* legal, last one wins */
 };
 
 static const struct intel_device_info intel_i845g_info = {
-	GEN2_FEATURES,
+	I845_FEATURES,
 	PLATFORM(INTEL_I845G),
 };
 
 static const struct intel_device_info intel_i85x_info = {
-	GEN2_FEATURES,
+	I830_FEATURES,
 	PLATFORM(INTEL_I85X),
-	.is_mobile = 1,
-	.num_pipes = 2, /* legal, last one wins */
-	.display.cursor_needs_physical = 1,
 	.display.has_fbc = 1,
 };
 
 static const struct intel_device_info intel_i865g_info = {
-	GEN2_FEATURES,
+	I845_FEATURES,
 	PLATFORM(INTEL_I865G),
 };
 
@@ -133,12 +205,13 @@ static const struct intel_device_info intel_i865g_info = {
 	.num_pipes = 2, \
 	.display.has_gmch = 1, \
 	.gpu_reset_clobbers_display = true, \
-	.ring_mask = RENDER_RING, \
+	.engine_mask = BIT(RCS0), \
 	.has_snoop = true, \
 	.has_coherent_ggtt = true, \
-	GEN_DEFAULT_PIPEOFFSETS, \
-	GEN_DEFAULT_PAGE_SIZES, \
-	CURSOR_OFFSETS
+	I9XX_PIPE_OFFSETS, \
+	I9XX_CURSOR_OFFSETS, \
+	I9XX_COLORS, \
+	GEN_DEFAULT_PAGE_SIZES
 
 static const struct intel_device_info intel_i915g_info = {
 	GEN3_FEATURES,
@@ -196,7 +269,14 @@ static const struct intel_device_info intel_g33_info = {
 	.display.has_overlay = 1,
 };
 
-static const struct intel_device_info intel_pineview_info = {
+static const struct intel_device_info intel_pineview_g_info = {
+	GEN3_FEATURES,
+	PLATFORM(INTEL_PINEVIEW),
+	.display.has_hotplug = 1,
+	.display.has_overlay = 1,
+};
+
+static const struct intel_device_info intel_pineview_m_info = {
 	GEN3_FEATURES,
 	PLATFORM(INTEL_PINEVIEW),
 	.is_mobile = 1,
@@ -210,12 +290,13 @@ static const struct intel_device_info intel_pineview_info = {
 	.display.has_hotplug = 1, \
 	.display.has_gmch = 1, \
 	.gpu_reset_clobbers_display = true, \
-	.ring_mask = RENDER_RING, \
+	.engine_mask = BIT(RCS0), \
 	.has_snoop = true, \
 	.has_coherent_ggtt = true, \
-	GEN_DEFAULT_PIPEOFFSETS, \
-	GEN_DEFAULT_PAGE_SIZES, \
-	CURSOR_OFFSETS
+	I9XX_PIPE_OFFSETS, \
+	I9XX_CURSOR_OFFSETS, \
+	I965_COLORS, \
+	GEN_DEFAULT_PAGE_SIZES
 
 static const struct intel_device_info intel_i965g_info = {
 	GEN4_FEATURES,
@@ -239,7 +320,7 @@ static const struct intel_device_info intel_i965gm_info = {
 static const struct intel_device_info intel_g45_info = {
 	GEN4_FEATURES,
 	PLATFORM(INTEL_G45),
-	.ring_mask = RENDER_RING | BSD_RING,
+	.engine_mask = BIT(RCS0) | BIT(VCS0),
 	.gpu_reset_clobbers_display = false,
 };
 
@@ -249,7 +330,7 @@ static const struct intel_device_info intel_gm45_info = {
 	.is_mobile = 1,
 	.display.has_fbc = 1,
 	.display.supports_tv = 1,
-	.ring_mask = RENDER_RING | BSD_RING,
+	.engine_mask = BIT(RCS0) | BIT(VCS0),
 	.gpu_reset_clobbers_display = false,
 };
 
@@ -257,14 +338,15 @@ static const struct intel_device_info intel_gm45_info = {
 	GEN(5), \
 	.num_pipes = 2, \
 	.display.has_hotplug = 1, \
-	.ring_mask = RENDER_RING | BSD_RING, \
+	.engine_mask = BIT(RCS0) | BIT(VCS0), \
 	.has_snoop = true, \
 	.has_coherent_ggtt = true, \
 	/* ilk does support rc6, but we do not implement [power] contexts */ \
 	.has_rc6 = 0, \
-	GEN_DEFAULT_PIPEOFFSETS, \
-	GEN_DEFAULT_PAGE_SIZES, \
-	CURSOR_OFFSETS
+	I9XX_PIPE_OFFSETS, \
+	I9XX_CURSOR_OFFSETS, \
+	ILK_COLORS, \
+	GEN_DEFAULT_PAGE_SIZES
 
 static const struct intel_device_info intel_ironlake_d_info = {
 	GEN5_FEATURES,
@@ -283,15 +365,17 @@ static const struct intel_device_info intel_ironlake_m_info = {
 	.num_pipes = 2, \
 	.display.has_hotplug = 1, \
 	.display.has_fbc = 1, \
-	.ring_mask = RENDER_RING | BSD_RING | BLT_RING, \
+	.engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0), \
 	.has_coherent_ggtt = true, \
 	.has_llc = 1, \
 	.has_rc6 = 1, \
 	.has_rc6p = 1, \
-	.ppgtt = INTEL_PPGTT_ALIASING, \
-	GEN_DEFAULT_PIPEOFFSETS, \
-	GEN_DEFAULT_PAGE_SIZES, \
-	CURSOR_OFFSETS
+	.ppgtt_type = INTEL_PPGTT_ALIASING, \
+	.ppgtt_size = 31, \
+	I9XX_PIPE_OFFSETS, \
+	I9XX_CURSOR_OFFSETS, \
+	ILK_COLORS, \
+	GEN_DEFAULT_PAGE_SIZES
 
 #define SNB_D_PLATFORM \
 	GEN6_FEATURES, \
@@ -328,15 +412,17 @@ static const struct intel_device_info intel_sandybridge_m_gt2_info = {
 	.num_pipes = 3, \
 	.display.has_hotplug = 1, \
 	.display.has_fbc = 1, \
-	.ring_mask = RENDER_RING | BSD_RING | BLT_RING, \
+	.engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0), \
 	.has_coherent_ggtt = true, \
 	.has_llc = 1, \
 	.has_rc6 = 1, \
 	.has_rc6p = 1, \
-	.ppgtt = INTEL_PPGTT_FULL, \
-	GEN_DEFAULT_PIPEOFFSETS, \
-	GEN_DEFAULT_PAGE_SIZES, \
-	IVB_CURSOR_OFFSETS
+	.ppgtt_type = INTEL_PPGTT_FULL, \
+	.ppgtt_size = 31, \
+	IVB_PIPE_OFFSETS, \
+	IVB_CURSOR_OFFSETS, \
+	IVB_COLORS, \
+	GEN_DEFAULT_PAGE_SIZES
 
 #define IVB_D_PLATFORM \
 	GEN7_FEATURES, \
@@ -386,24 +472,27 @@ static const struct intel_device_info intel_valleyview_info = {
 	.has_rc6 = 1,
 	.display.has_gmch = 1,
 	.display.has_hotplug = 1,
-	.ppgtt = INTEL_PPGTT_FULL,
+	.ppgtt_type = INTEL_PPGTT_FULL,
+	.ppgtt_size = 31,
 	.has_snoop = true,
 	.has_coherent_ggtt = false,
-	.ring_mask = RENDER_RING | BSD_RING | BLT_RING,
+	.engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0),
 	.display_mmio_offset = VLV_DISPLAY_BASE,
+	I9XX_PIPE_OFFSETS,
+	I9XX_CURSOR_OFFSETS,
+	I965_COLORS,
 	GEN_DEFAULT_PAGE_SIZES,
-	GEN_DEFAULT_PIPEOFFSETS,
-	CURSOR_OFFSETS
 };
 
 #define G75_FEATURES  \
 	GEN7_FEATURES, \
-	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING, \
+	.engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0), \
 	.display.has_ddi = 1, \
 	.has_fpga_dbg = 1, \
 	.display.has_psr = 1, \
 	.display.has_dp_mst = 1, \
 	.has_rc6p = 0 /* RC6p removed-by HSW */, \
+	HSW_PIPE_OFFSETS, \
 	.has_runtime_pm = 1
 
 #define HSW_PLATFORM \
@@ -429,11 +518,11 @@ static const struct intel_device_info intel_haswell_gt3_info = {
 #define GEN8_FEATURES \
 	G75_FEATURES, \
 	GEN(8), \
-	BDW_COLORS, \
 	.page_sizes = I915_GTT_PAGE_SIZE_4K | \
 		      I915_GTT_PAGE_SIZE_2M, \
 	.has_logical_ring_contexts = 1, \
-	.ppgtt = INTEL_PPGTT_FULL_4LVL, \
+	.ppgtt_type = INTEL_PPGTT_FULL, \
+	.ppgtt_size = 48, \
 	.has_64bit_reloc = 1, \
 	.has_reset_engine = 1
 
@@ -462,7 +551,8 @@ static const struct intel_device_info intel_broadwell_rsvd_info = {
 static const struct intel_device_info intel_broadwell_gt3_info = {
 	BDW_PLATFORM,
 	.gt = 3,
-	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING | BSD2_RING,
+	.engine_mask =
+		BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS1),
 };
 
 static const struct intel_device_info intel_cherryview_info = {
@@ -471,21 +561,22 @@ static const struct intel_device_info intel_cherryview_info = {
 	.num_pipes = 3,
 	.display.has_hotplug = 1,
 	.is_lp = 1,
-	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING,
+	.engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0),
 	.has_64bit_reloc = 1,
 	.has_runtime_pm = 1,
 	.has_rc6 = 1,
 	.has_logical_ring_contexts = 1,
 	.display.has_gmch = 1,
-	.ppgtt = INTEL_PPGTT_FULL,
+	.ppgtt_type = INTEL_PPGTT_FULL,
+	.ppgtt_size = 32,
 	.has_reset_engine = 1,
 	.has_snoop = true,
 	.has_coherent_ggtt = false,
 	.display_mmio_offset = VLV_DISPLAY_BASE,
-	GEN_DEFAULT_PAGE_SIZES,
-	GEN_CHV_PIPEOFFSETS,
-	CURSOR_OFFSETS,
+	CHV_PIPE_OFFSETS,
+	CHV_CURSOR_OFFSETS,
 	CHV_COLORS,
+	GEN_DEFAULT_PAGE_SIZES,
 };
 
 #define GEN9_DEFAULT_PAGE_SIZES \
@@ -521,7 +612,8 @@ static const struct intel_device_info intel_skylake_gt2_info = {
 
 #define SKL_GT3_PLUS_PLATFORM \
 	SKL_PLATFORM, \
-	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING | BSD2_RING
+	.engine_mask = \
+		BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS1)
 
 
 static const struct intel_device_info intel_skylake_gt3_info = {
@@ -538,7 +630,7 @@ static const struct intel_device_info intel_skylake_gt4_info = {
 	GEN(9), \
 	.is_lp = 1, \
 	.display.has_hotplug = 1, \
-	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING, \
+	.engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0), \
 	.num_pipes = 3, \
 	.has_64bit_reloc = 1, \
 	.display.has_ddi = 1, \
@@ -552,15 +644,16 @@ static const struct intel_device_info intel_skylake_gt4_info = {
 	.has_logical_ring_contexts = 1, \
 	.has_logical_ring_preemption = 1, \
 	.has_guc = 1, \
-	.ppgtt = INTEL_PPGTT_FULL_4LVL, \
+	.ppgtt_type = INTEL_PPGTT_FULL, \
+	.ppgtt_size = 48, \
 	.has_reset_engine = 1, \
 	.has_snoop = true, \
 	.has_coherent_ggtt = false, \
 	.display.has_ipc = 1, \
-	GEN9_DEFAULT_PAGE_SIZES, \
-	GEN_DEFAULT_PIPEOFFSETS, \
+	HSW_PIPE_OFFSETS, \
 	IVB_CURSOR_OFFSETS, \
-	BDW_COLORS
+	IVB_COLORS, \
+	GEN9_DEFAULT_PAGE_SIZES
 
 static const struct intel_device_info intel_broxton_info = {
 	GEN9_LP_FEATURES,
@@ -592,7 +685,8 @@ static const struct intel_device_info intel_kabylake_gt2_info = {
 static const struct intel_device_info intel_kabylake_gt3_info = {
 	KBL_PLATFORM,
 	.gt = 3,
-	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING | BSD2_RING,
+	.engine_mask =
+		BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS1),
 };
 
 #define CFL_PLATFORM \
@@ -612,7 +706,8 @@ static const struct intel_device_info intel_coffeelake_gt2_info = {
 static const struct intel_device_info intel_coffeelake_gt3_info = {
 	CFL_PLATFORM,
 	.gt = 3,
-	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING | BSD2_RING,
+	.engine_mask =
+		BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS1),
 };
 
 #define GEN10_FEATURES \
@@ -648,13 +743,22 @@ static const struct intel_device_info intel_cannonlake_info = {
 	}, \
 	GEN(11), \
 	.ddb_size = 2048, \
-	.has_logical_ring_elsq = 1
+	.has_logical_ring_elsq = 1, \
+	.color = { .degamma_lut_size = 33, .gamma_lut_size = 1024 }
 
 static const struct intel_device_info intel_icelake_11_info = {
 	GEN11_FEATURES,
 	PLATFORM(INTEL_ICELAKE),
+	.engine_mask =
+		BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0) | BIT(VCS2),
+};
+
+static const struct intel_device_info intel_elkhartlake_info = {
+	GEN11_FEATURES,
+	PLATFORM(INTEL_ELKHARTLAKE),
 	.is_alpha_support = 1,
-	.ring_mask = RENDER_RING | BLT_RING | VEBOX_RING | BSD_RING | BSD3_RING,
+	.engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VCS0),
+	.ppgtt_size = 36,
 };
 
 #undef GEN
@@ -680,7 +784,8 @@ static const struct pci_device_id pciidlist[] = {
 	INTEL_I965GM_IDS(&intel_i965gm_info),
 	INTEL_GM45_IDS(&intel_gm45_info),
 	INTEL_G45_IDS(&intel_g45_info),
-	INTEL_PINEVIEW_IDS(&intel_pineview_info),
+	INTEL_PINEVIEW_G_IDS(&intel_pineview_g_info),
+	INTEL_PINEVIEW_M_IDS(&intel_pineview_m_info),
 	INTEL_IRONLAKE_D_IDS(&intel_ironlake_d_info),
 	INTEL_IRONLAKE_M_IDS(&intel_ironlake_m_info),
 	INTEL_SNB_D_GT1_IDS(&intel_sandybridge_d_gt1_info),
@@ -722,8 +827,11 @@ static const struct pci_device_id pciidlist[] = {
 	INTEL_WHL_U_GT2_IDS(&intel_coffeelake_gt2_info),
 	INTEL_AML_CFL_GT2_IDS(&intel_coffeelake_gt2_info),
 	INTEL_WHL_U_GT3_IDS(&intel_coffeelake_gt3_info),
+	INTEL_CML_GT1_IDS(&intel_coffeelake_gt1_info),
+	INTEL_CML_GT2_IDS(&intel_coffeelake_gt2_info),
 	INTEL_CNL_IDS(&intel_cannonlake_info),
 	INTEL_ICL_11_IDS(&intel_icelake_11_info),
+	INTEL_EHL_IDS(&intel_elkhartlake_info),
 	{0, 0, 0}
 };
 MODULE_DEVICE_TABLE(pci, pciidlist);
@@ -801,7 +909,9 @@ static int __init i915_init(void)
 	bool use_kms = true;
 	int err;
 
-	i915_global_active_init();
+	err = i915_globals_init();
+	if (err)
+		return err;
 
 	err = i915_mock_selftests();
 	if (err)
@@ -834,7 +944,7 @@ static void __exit i915_exit(void)
 		return;
 
 	pci_unregister_driver(&i915_pci_driver);
-	i915_global_active_exit();
+	i915_globals_exit();
 }
 
 module_init(i915_init);
diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c
index 9ebf99f3d8d3..39a4804091d7 100644
--- a/drivers/gpu/drm/i915/i915_perf.c
+++ b/drivers/gpu/drm/i915/i915_perf.c
@@ -1202,7 +1202,7 @@ static int i915_oa_read(struct i915_perf_stream *stream,
 static struct intel_context *oa_pin_context(struct drm_i915_private *i915,
 					    struct i915_gem_context *ctx)
 {
-	struct intel_engine_cs *engine = i915->engine[RCS];
+	struct intel_engine_cs *engine = i915->engine[RCS0];
 	struct intel_context *ce;
 	int ret;
 
@@ -1364,7 +1364,7 @@ static void i915_oa_stream_destroy(struct i915_perf_stream *stream)
 
 	free_oa_buffer(dev_priv);
 
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 	intel_runtime_pm_put(dev_priv, stream->wakeref);
 
 	if (stream->ctx)
@@ -1509,9 +1509,7 @@ static int alloc_oa_buffer(struct drm_i915_private *dev_priv)
 		goto unlock;
 	}
 
-	ret = i915_gem_object_set_cache_level(bo, I915_CACHE_LLC);
-	if (ret)
-		goto err_unref;
+	i915_gem_object_set_cache_coherency(bo, I915_CACHE_LLC);
 
 	/* PreHSW required 512K alignment, HSW requires 16M */
 	vma = i915_gem_object_ggtt_pin(bo, NULL, 0, SZ_16M, 0);
@@ -1629,13 +1627,14 @@ static void hsw_disable_metric_set(struct drm_i915_private *dev_priv)
  * It's fine to put out-of-date values into these per-context registers
  * in the case that the OA unit has been disabled.
  */
-static void gen8_update_reg_state_unlocked(struct i915_gem_context *ctx,
-					   u32 *reg_state,
-					   const struct i915_oa_config *oa_config)
+static void
+gen8_update_reg_state_unlocked(struct intel_context *ce,
+			       u32 *reg_state,
+			       const struct i915_oa_config *oa_config)
 {
-	struct drm_i915_private *dev_priv = ctx->i915;
-	u32 ctx_oactxctrl = dev_priv->perf.oa.ctx_oactxctrl_offset;
-	u32 ctx_flexeu0 = dev_priv->perf.oa.ctx_flexeu0_offset;
+	struct drm_i915_private *i915 = ce->gem_context->i915;
+	u32 ctx_oactxctrl = i915->perf.oa.ctx_oactxctrl_offset;
+	u32 ctx_flexeu0 = i915->perf.oa.ctx_flexeu0_offset;
 	/* The MMIO offsets for Flex EU registers aren't contiguous */
 	i915_reg_t flex_regs[] = {
 		EU_PERF_CNTL0,
@@ -1649,8 +1648,8 @@ static void gen8_update_reg_state_unlocked(struct i915_gem_context *ctx,
 	int i;
 
 	CTX_REG(reg_state, ctx_oactxctrl, GEN8_OACTXCONTROL,
-		(dev_priv->perf.oa.period_exponent << GEN8_OA_TIMER_PERIOD_SHIFT) |
-		(dev_priv->perf.oa.periodic ? GEN8_OA_TIMER_ENABLE : 0) |
+		(i915->perf.oa.period_exponent << GEN8_OA_TIMER_PERIOD_SHIFT) |
+		(i915->perf.oa.periodic ? GEN8_OA_TIMER_ENABLE : 0) |
 		GEN8_OA_COUNTER_RESUME);
 
 	for (i = 0; i < ARRAY_SIZE(flex_regs); i++) {
@@ -1678,10 +1677,9 @@ static void gen8_update_reg_state_unlocked(struct i915_gem_context *ctx,
 		CTX_REG(reg_state, state_offset, flex_regs[i], value);
 	}
 
-	CTX_REG(reg_state, CTX_R_PWR_CLK_STATE, GEN8_R_PWR_CLK_STATE,
-		gen8_make_rpcs(dev_priv,
-			       &to_intel_context(ctx,
-						 dev_priv->engine[RCS])->sseu));
+	CTX_REG(reg_state,
+		CTX_R_PWR_CLK_STATE, GEN8_R_PWR_CLK_STATE,
+		gen8_make_rpcs(i915, &ce->sseu));
 }
 
 /*
@@ -1711,7 +1709,7 @@ static void gen8_update_reg_state_unlocked(struct i915_gem_context *ctx,
 static int gen8_configure_all_contexts(struct drm_i915_private *dev_priv,
 				       const struct i915_oa_config *oa_config)
 {
-	struct intel_engine_cs *engine = dev_priv->engine[RCS];
+	struct intel_engine_cs *engine = dev_priv->engine[RCS0];
 	unsigned int map_type = i915_coherent_map_type(dev_priv);
 	struct i915_gem_context *ctx;
 	struct i915_request *rq;
@@ -1740,11 +1738,11 @@ static int gen8_configure_all_contexts(struct drm_i915_private *dev_priv,
 
 	/* Update all contexts now that we've stalled the submission. */
 	list_for_each_entry(ctx, &dev_priv->contexts.list, link) {
-		struct intel_context *ce = to_intel_context(ctx, engine);
+		struct intel_context *ce = intel_context_lookup(ctx, engine);
 		u32 *regs;
 
 		/* OA settings will be set upon first use */
-		if (!ce->state)
+		if (!ce || !ce->state)
 			continue;
 
 		regs = i915_gem_object_pin_map(ce->state->obj, map_type);
@@ -1754,7 +1752,7 @@ static int gen8_configure_all_contexts(struct drm_i915_private *dev_priv,
 		ce->state->obj->mm.dirty = true;
 		regs += LRC_STATE_PN * PAGE_SIZE / sizeof(*regs);
 
-		gen8_update_reg_state_unlocked(ctx, regs, oa_config);
+		gen8_update_reg_state_unlocked(ce, regs, oa_config);
 
 		i915_gem_object_unpin_map(ce->state->obj);
 	}
@@ -1922,10 +1920,10 @@ static void i915_oa_stream_enable(struct i915_perf_stream *stream)
 
 static void gen7_oa_disable(struct i915_perf_stream *stream)
 {
-	struct drm_i915_private *dev_priv = stream->dev_priv;
+	struct intel_uncore *uncore = &stream->dev_priv->uncore;
 
-	I915_WRITE(GEN7_OACONTROL, 0);
-	if (intel_wait_for_register(dev_priv,
+	intel_uncore_write(uncore, GEN7_OACONTROL, 0);
+	if (intel_wait_for_register(uncore,
 				    GEN7_OACONTROL, GEN7_OACONTROL_ENABLE, 0,
 				    50))
 		DRM_ERROR("wait for OA to be disabled timed out\n");
@@ -1933,10 +1931,10 @@ static void gen7_oa_disable(struct i915_perf_stream *stream)
 
 static void gen8_oa_disable(struct i915_perf_stream *stream)
 {
-	struct drm_i915_private *dev_priv = stream->dev_priv;
+	struct intel_uncore *uncore = &stream->dev_priv->uncore;
 
-	I915_WRITE(GEN8_OACONTROL, 0);
-	if (intel_wait_for_register(dev_priv,
+	intel_uncore_write(uncore, GEN8_OACONTROL, 0);
+	if (intel_wait_for_register(uncore,
 				    GEN8_OACONTROL, GEN8_OA_COUNTER_ENABLE, 0,
 				    50))
 		DRM_ERROR("wait for OA to be disabled timed out\n");
@@ -2093,7 +2091,7 @@ static int i915_oa_stream_init(struct i915_perf_stream *stream,
 	 *   references will effectively disable RC6.
 	 */
 	stream->wakeref = intel_runtime_pm_get(dev_priv);
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL);
 
 	ret = alloc_oa_buffer(dev_priv);
 	if (ret)
@@ -2127,7 +2125,7 @@ err_lock:
 err_oa_buf_alloc:
 	put_oa_config(dev_priv, stream->oa_config);
 
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 	intel_runtime_pm_put(dev_priv, stream->wakeref);
 
 err_config:
@@ -2138,17 +2136,17 @@ err_config:
 }
 
 void i915_oa_init_reg_state(struct intel_engine_cs *engine,
-			    struct i915_gem_context *ctx,
-			    u32 *reg_state)
+			    struct intel_context *ce,
+			    u32 *regs)
 {
 	struct i915_perf_stream *stream;
 
-	if (engine->id != RCS)
+	if (engine->class != RENDER_CLASS)
 		return;
 
 	stream = engine->i915->perf.oa.exclusive_stream;
 	if (stream)
-		gen8_update_reg_state_unlocked(ctx, reg_state, stream->oa_config);
+		gen8_update_reg_state_unlocked(ce, regs, stream->oa_config);
 }
 
 /**
@@ -2881,12 +2879,24 @@ void i915_perf_register(struct drm_i915_private *dev_priv)
 
 	sysfs_attr_init(&dev_priv->perf.oa.test_config.sysfs_metric_id.attr);
 
-	if (IS_HASWELL(dev_priv)) {
-		i915_perf_load_test_config_hsw(dev_priv);
-	} else if (IS_BROADWELL(dev_priv)) {
-		i915_perf_load_test_config_bdw(dev_priv);
-	} else if (IS_CHERRYVIEW(dev_priv)) {
-		i915_perf_load_test_config_chv(dev_priv);
+	if (INTEL_GEN(dev_priv) >= 11) {
+		i915_perf_load_test_config_icl(dev_priv);
+	} else if (IS_CANNONLAKE(dev_priv)) {
+		i915_perf_load_test_config_cnl(dev_priv);
+	} else if (IS_COFFEELAKE(dev_priv)) {
+		if (IS_CFL_GT2(dev_priv))
+			i915_perf_load_test_config_cflgt2(dev_priv);
+		if (IS_CFL_GT3(dev_priv))
+			i915_perf_load_test_config_cflgt3(dev_priv);
+	} else if (IS_GEMINILAKE(dev_priv)) {
+		i915_perf_load_test_config_glk(dev_priv);
+	} else if (IS_KABYLAKE(dev_priv)) {
+		if (IS_KBL_GT2(dev_priv))
+			i915_perf_load_test_config_kblgt2(dev_priv);
+		else if (IS_KBL_GT3(dev_priv))
+			i915_perf_load_test_config_kblgt3(dev_priv);
+	} else if (IS_BROXTON(dev_priv)) {
+		i915_perf_load_test_config_bxt(dev_priv);
 	} else if (IS_SKYLAKE(dev_priv)) {
 		if (IS_SKL_GT2(dev_priv))
 			i915_perf_load_test_config_sklgt2(dev_priv);
@@ -2894,25 +2904,13 @@ void i915_perf_register(struct drm_i915_private *dev_priv)
 			i915_perf_load_test_config_sklgt3(dev_priv);
 		else if (IS_SKL_GT4(dev_priv))
 			i915_perf_load_test_config_sklgt4(dev_priv);
-	} else if (IS_BROXTON(dev_priv)) {
-		i915_perf_load_test_config_bxt(dev_priv);
-	} else if (IS_KABYLAKE(dev_priv)) {
-		if (IS_KBL_GT2(dev_priv))
-			i915_perf_load_test_config_kblgt2(dev_priv);
-		else if (IS_KBL_GT3(dev_priv))
-			i915_perf_load_test_config_kblgt3(dev_priv);
-	} else if (IS_GEMINILAKE(dev_priv)) {
-		i915_perf_load_test_config_glk(dev_priv);
-	} else if (IS_COFFEELAKE(dev_priv)) {
-		if (IS_CFL_GT2(dev_priv))
-			i915_perf_load_test_config_cflgt2(dev_priv);
-		if (IS_CFL_GT3(dev_priv))
-			i915_perf_load_test_config_cflgt3(dev_priv);
-	} else if (IS_CANNONLAKE(dev_priv)) {
-		i915_perf_load_test_config_cnl(dev_priv);
-	} else if (IS_ICELAKE(dev_priv)) {
-		i915_perf_load_test_config_icl(dev_priv);
-	}
+	} else if (IS_CHERRYVIEW(dev_priv)) {
+		i915_perf_load_test_config_chv(dev_priv);
+	} else if (IS_BROADWELL(dev_priv)) {
+		i915_perf_load_test_config_bdw(dev_priv);
+	} else if (IS_HASWELL(dev_priv)) {
+		i915_perf_load_test_config_hsw(dev_priv);
+}
 
 	if (dev_priv->perf.oa.test_config.id == 0)
 		goto sysfs_error;
diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c
index b745c49a5af6..46a52da3db29 100644
--- a/drivers/gpu/drm/i915/i915_pmu.c
+++ b/drivers/gpu/drm/i915/i915_pmu.c
@@ -102,7 +102,7 @@ static bool pmu_needs_timer(struct drm_i915_private *i915, bool gpu_active)
 	 *
 	 * Use RCS as proxy for all engines.
 	 */
-	else if (intel_engine_supports_stats(i915->engine[RCS]))
+	else if (intel_engine_supports_stats(i915->engine[RCS0]))
 		enable &= ~BIT(I915_SAMPLE_BUSY);
 
 	/*
@@ -149,14 +149,6 @@ void i915_pmu_gt_unparked(struct drm_i915_private *i915)
 	spin_unlock_irq(&i915->pmu.lock);
 }
 
-static bool grab_forcewake(struct drm_i915_private *i915, bool fw)
-{
-	if (!fw)
-		intel_uncore_forcewake_get(i915, FORCEWAKE_ALL);
-
-	return true;
-}
-
 static void
 add_sample(struct i915_pmu_sample *sample, u32 val)
 {
@@ -169,49 +161,48 @@ engines_sample(struct drm_i915_private *dev_priv, unsigned int period_ns)
 	struct intel_engine_cs *engine;
 	enum intel_engine_id id;
 	intel_wakeref_t wakeref;
-	bool fw = false;
+	unsigned long flags;
 
 	if ((dev_priv->pmu.enable & ENGINE_SAMPLE_MASK) == 0)
 		return;
 
-	if (!dev_priv->gt.awake)
-		return;
-
-	wakeref = intel_runtime_pm_get_if_in_use(dev_priv);
+	wakeref = 0;
+	if (READ_ONCE(dev_priv->gt.awake))
+		wakeref = intel_runtime_pm_get_if_in_use(dev_priv);
 	if (!wakeref)
 		return;
 
+	spin_lock_irqsave(&dev_priv->uncore.lock, flags);
 	for_each_engine(engine, dev_priv, id) {
-		u32 current_seqno = intel_engine_get_seqno(engine);
-		u32 last_seqno = intel_engine_last_submit(engine);
+		struct intel_engine_pmu *pmu = &engine->pmu;
+		bool busy;
 		u32 val;
 
-		val = !i915_seqno_passed(current_seqno, last_seqno);
-
-		if (val)
-			add_sample(&engine->pmu.sample[I915_SAMPLE_BUSY],
-				   period_ns);
-
-		if (val && (engine->pmu.enable &
-		    (BIT(I915_SAMPLE_WAIT) | BIT(I915_SAMPLE_SEMA)))) {
-			fw = grab_forcewake(dev_priv, fw);
-
-			val = I915_READ_FW(RING_CTL(engine->mmio_base));
-		} else {
-			val = 0;
-		}
+		val = I915_READ_FW(RING_CTL(engine->mmio_base));
+		if (val == 0) /* powerwell off => engine idle */
+			continue;
 
 		if (val & RING_WAIT)
-			add_sample(&engine->pmu.sample[I915_SAMPLE_WAIT],
-				   period_ns);
-
+			add_sample(&pmu->sample[I915_SAMPLE_WAIT], period_ns);
 		if (val & RING_WAIT_SEMAPHORE)
-			add_sample(&engine->pmu.sample[I915_SAMPLE_SEMA],
-				   period_ns);
-	}
+			add_sample(&pmu->sample[I915_SAMPLE_SEMA], period_ns);
 
-	if (fw)
-		intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+		/*
+		 * While waiting on a semaphore or event, MI_MODE reports the
+		 * ring as idle. However, previously using the seqno, and with
+		 * execlists sampling, we account for the ring waiting as the
+		 * engine being busy. Therefore, we record the sample as being
+		 * busy if either waiting or !idle.
+		 */
+		busy = val & (RING_WAIT_SEMAPHORE | RING_WAIT);
+		if (!busy) {
+			val = I915_READ_FW(RING_MI_MODE(engine->mmio_base));
+			busy = !(val & MODE_IDLE);
+		}
+		if (busy)
+			add_sample(&pmu->sample[I915_SAMPLE_BUSY], period_ns);
+	}
+	spin_unlock_irqrestore(&dev_priv->uncore.lock, flags);
 
 	intel_runtime_pm_put(dev_priv, wakeref);
 }
diff --git a/drivers/gpu/drm/i915/i915_priolist_types.h b/drivers/gpu/drm/i915/i915_priolist_types.h
new file mode 100644
index 000000000000..cc44ebd3b553
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_priolist_types.h
@@ -0,0 +1,42 @@
+/*
+ * SPDX-License-Identifier: MIT
+ *
+ * Copyright © 2018 Intel Corporation
+ */
+
+#ifndef _I915_PRIOLIST_TYPES_H_
+#define _I915_PRIOLIST_TYPES_H_
+
+#include <linux/list.h>
+#include <linux/rbtree.h>
+
+#include <uapi/drm/i915_drm.h>
+
+enum {
+	I915_PRIORITY_MIN = I915_CONTEXT_MIN_USER_PRIORITY - 1,
+	I915_PRIORITY_NORMAL = I915_CONTEXT_DEFAULT_PRIORITY,
+	I915_PRIORITY_MAX = I915_CONTEXT_MAX_USER_PRIORITY + 1,
+
+	I915_PRIORITY_INVALID = INT_MIN
+};
+
+#define I915_USER_PRIORITY_SHIFT 3
+#define I915_USER_PRIORITY(x) ((x) << I915_USER_PRIORITY_SHIFT)
+
+#define I915_PRIORITY_COUNT BIT(I915_USER_PRIORITY_SHIFT)
+#define I915_PRIORITY_MASK (I915_PRIORITY_COUNT - 1)
+
+#define I915_PRIORITY_WAIT		((u8)BIT(0))
+#define I915_PRIORITY_NEWCLIENT		((u8)BIT(1))
+#define I915_PRIORITY_NOSEMAPHORE	((u8)BIT(2))
+
+#define __NO_PREEMPTION (I915_PRIORITY_WAIT)
+
+struct i915_priolist {
+	struct list_head requests[I915_PRIORITY_COUNT];
+	struct rb_node node;
+	unsigned long used;
+	int priority;
+};
+
+#endif /* _I915_PRIOLIST_TYPES_H_ */
diff --git a/drivers/gpu/drm/i915/i915_pvinfo.h b/drivers/gpu/drm/i915/i915_pvinfo.h
index eeaa3d506d95..969e514916ab 100644
--- a/drivers/gpu/drm/i915/i915_pvinfo.h
+++ b/drivers/gpu/drm/i915/i915_pvinfo.h
@@ -52,7 +52,7 @@ enum vgt_g2v_type {
 /*
  * VGT capabilities type
  */
-#define VGT_CAPS_FULL_48BIT_PPGTT	BIT(2)
+#define VGT_CAPS_FULL_PPGTT		BIT(2)
 #define VGT_CAPS_HWSP_EMULATION		BIT(3)
 #define VGT_CAPS_HUGE_GTT		BIT(4)
 
diff --git a/drivers/gpu/drm/i915/i915_query.c b/drivers/gpu/drm/i915/i915_query.c
index cbcb957b7141..782183b78f49 100644
--- a/drivers/gpu/drm/i915/i915_query.c
+++ b/drivers/gpu/drm/i915/i915_query.c
@@ -10,12 +10,34 @@
 #include "i915_query.h"
 #include <uapi/drm/i915_drm.h>
 
+static int copy_query_item(void *query_hdr, size_t query_sz,
+			   u32 total_length,
+			   struct drm_i915_query_item *query_item)
+{
+	if (query_item->length == 0)
+		return total_length;
+
+	if (query_item->length < total_length)
+		return -EINVAL;
+
+	if (copy_from_user(query_hdr, u64_to_user_ptr(query_item->data_ptr),
+			   query_sz))
+		return -EFAULT;
+
+	if (!access_ok(u64_to_user_ptr(query_item->data_ptr),
+		       total_length))
+		return -EFAULT;
+
+	return 0;
+}
+
 static int query_topology_info(struct drm_i915_private *dev_priv,
 			       struct drm_i915_query_item *query_item)
 {
 	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
 	struct drm_i915_query_topology_info topo;
 	u32 slice_length, subslice_length, eu_length, total_length;
+	int ret;
 
 	if (query_item->flags != 0)
 		return -EINVAL;
@@ -33,23 +55,14 @@ static int query_topology_info(struct drm_i915_private *dev_priv,
 
 	total_length = sizeof(topo) + slice_length + subslice_length + eu_length;
 
-	if (query_item->length == 0)
-		return total_length;
-
-	if (query_item->length < total_length)
-		return -EINVAL;
-
-	if (copy_from_user(&topo, u64_to_user_ptr(query_item->data_ptr),
-			   sizeof(topo)))
-		return -EFAULT;
+	ret = copy_query_item(&topo, sizeof(topo), total_length,
+			      query_item);
+	if (ret != 0)
+		return ret;
 
 	if (topo.flags != 0)
 		return -EINVAL;
 
-	if (!access_ok(u64_to_user_ptr(query_item->data_ptr),
-		       total_length))
-		return -EFAULT;
-
 	memset(&topo, 0, sizeof(topo));
 	topo.max_slices = sseu->max_slices;
 	topo.max_subslices = sseu->max_subslices;
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 047855dd8c6b..b74824f0b5b1 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -25,6 +25,9 @@
 #ifndef _I915_REG_H_
 #define _I915_REG_H_
 
+#include <linux/bitfield.h>
+#include <linux/bits.h>
+
 /**
  * DOC: The i915 register macro definition style guide
  *
@@ -59,15 +62,13 @@
  * significant to least significant bit. Indent the register content macros
  * using two extra spaces between ``#define`` and the macro name.
  *
- * For bit fields, define a ``_MASK`` and a ``_SHIFT`` macro. Define bit field
- * contents so that they are already shifted in place, and can be directly
- * OR'd. For convenience, function-like macros may be used to define bit fields,
- * but do note that the macros may be needed to read as well as write the
- * register contents.
+ * Define bit fields using ``REG_GENMASK(h, l)``. Define bit field contents
+ * using ``REG_FIELD_PREP(mask, value)``. This will define the values already
+ * shifted in place, so they can be directly OR'd together. For convenience,
+ * function-like macros may be used to define bit fields, but do note that the
+ * macros may be needed to read as well as write the register contents.
  *
- * Define bits using ``(1 << N)`` instead of ``BIT(N)``. We may change this in
- * the future, but this is the prevailing style. Do **not** add ``_BIT`` suffix
- * to the name.
+ * Define bits using ``REG_BIT(N)``. Do **not** add ``_BIT`` suffix to the name.
  *
  * Group the register and its contents together without blank lines, separate
  * from other registers and their contents with one blank line.
@@ -105,17 +106,78 @@
  *  #define _FOO_A                      0xf000
  *  #define _FOO_B                      0xf001
  *  #define FOO(pipe)                   _MMIO_PIPE(pipe, _FOO_A, _FOO_B)
- *  #define   FOO_ENABLE                (1 << 31)
- *  #define   FOO_MODE_MASK             (0xf << 16)
- *  #define   FOO_MODE_SHIFT            16
- *  #define   FOO_MODE_BAR              (0 << 16)
- *  #define   FOO_MODE_BAZ              (1 << 16)
- *  #define   FOO_MODE_QUX_SNB          (2 << 16)
+ *  #define   FOO_ENABLE                REG_BIT(31)
+ *  #define   FOO_MODE_MASK             REG_GENMASK(19, 16)
+ *  #define   FOO_MODE_BAR              REG_FIELD_PREP(FOO_MODE_MASK, 0)
+ *  #define   FOO_MODE_BAZ              REG_FIELD_PREP(FOO_MODE_MASK, 1)
+ *  #define   FOO_MODE_QUX_SNB          REG_FIELD_PREP(FOO_MODE_MASK, 2)
  *
  *  #define BAR                         _MMIO(0xb000)
  *  #define GEN8_BAR                    _MMIO(0xb888)
  */
 
+/**
+ * REG_BIT() - Prepare a u32 bit value
+ * @__n: 0-based bit number
+ *
+ * Local wrapper for BIT() to force u32, with compile time checks.
+ *
+ * @return: Value with bit @__n set.
+ */
+#define REG_BIT(__n)							\
+	((u32)(BIT(__n) +						\
+	       BUILD_BUG_ON_ZERO(__builtin_constant_p(__n) &&		\
+				 ((__n) < 0 || (__n) > 31))))
+
+/**
+ * REG_GENMASK() - Prepare a continuous u32 bitmask
+ * @__high: 0-based high bit
+ * @__low: 0-based low bit
+ *
+ * Local wrapper for GENMASK() to force u32, with compile time checks.
+ *
+ * @return: Continuous bitmask from @__high to @__low, inclusive.
+ */
+#define REG_GENMASK(__high, __low)					\
+	((u32)(GENMASK(__high, __low) +					\
+	       BUILD_BUG_ON_ZERO(__builtin_constant_p(__high) &&	\
+				 __builtin_constant_p(__low) &&		\
+				 ((__low) < 0 || (__high) > 31 || (__low) > (__high)))))
+
+/*
+ * Local integer constant expression version of is_power_of_2().
+ */
+#define IS_POWER_OF_2(__x)		((__x) && (((__x) & ((__x) - 1)) == 0))
+
+/**
+ * REG_FIELD_PREP() - Prepare a u32 bitfield value
+ * @__mask: shifted mask defining the field's length and position
+ * @__val: value to put in the field
+
+ * Local copy of FIELD_PREP() to generate an integer constant expression, force
+ * u32 and for consistency with REG_FIELD_GET(), REG_BIT() and REG_GENMASK().
+ *
+ * @return: @__val masked and shifted into the field defined by @__mask.
+ */
+#define REG_FIELD_PREP(__mask, __val)						\
+	((u32)((((typeof(__mask))(__val) << __bf_shf(__mask)) & (__mask)) +	\
+	       BUILD_BUG_ON_ZERO(!__is_constexpr(__mask)) +		\
+	       BUILD_BUG_ON_ZERO((__mask) == 0 || (__mask) > U32_MAX) +		\
+	       BUILD_BUG_ON_ZERO(!IS_POWER_OF_2((__mask) + (1ULL << __bf_shf(__mask)))) + \
+	       BUILD_BUG_ON_ZERO(__builtin_choose_expr(__is_constexpr(__val), (~((__mask) >> __bf_shf(__mask)) & (__val)), 0))))
+
+/**
+ * REG_FIELD_GET() - Extract a u32 bitfield value
+ * @__mask: shifted mask defining the field's length and position
+ * @__val: value to extract the bitfield value from
+ *
+ * Local wrapper for FIELD_GET() to force u32 and for consistency with
+ * REG_FIELD_PREP(), REG_BIT() and REG_GENMASK().
+ *
+ * @return: Masked and shifted value of the field defined by @__mask in @__val.
+ */
+#define REG_FIELD_GET(__mask, __val)	((u32)FIELD_GET(__mask, __val))
+
 typedef struct {
 	u32 reg;
 } i915_reg_t;
@@ -210,14 +272,14 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg)
 
 /* Engine ID */
 
-#define RCS_HW		0
-#define VCS_HW		1
-#define BCS_HW		2
-#define VECS_HW		3
-#define VCS2_HW		4
-#define VCS3_HW		6
-#define VCS4_HW		7
-#define VECS2_HW	12
+#define RCS0_HW		0
+#define VCS0_HW		1
+#define BCS0_HW		2
+#define VECS0_HW	3
+#define VCS1_HW		4
+#define VCS2_HW		6
+#define VCS3_HW		7
+#define VECS1_HW	12
 
 /* Engine class */
 
@@ -372,13 +434,13 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg)
 #define GEN11_VECS_SFC_USAGE(engine)		_MMIO((engine)->mmio_base + 0x2014)
 #define   GEN11_VECS_SFC_USAGE_BIT		(1 << 0)
 
-#define RING_PP_DIR_BASE(engine)	_MMIO((engine)->mmio_base + 0x228)
-#define RING_PP_DIR_BASE_READ(engine)	_MMIO((engine)->mmio_base + 0x518)
-#define RING_PP_DIR_DCLV(engine)	_MMIO((engine)->mmio_base + 0x220)
+#define RING_PP_DIR_BASE(base)		_MMIO((base) + 0x228)
+#define RING_PP_DIR_BASE_READ(base)	_MMIO((base) + 0x518)
+#define RING_PP_DIR_DCLV(base)		_MMIO((base) + 0x220)
 #define   PP_DIR_DCLV_2G		0xffffffff
 
-#define GEN8_RING_PDP_UDW(engine, n)	_MMIO((engine)->mmio_base + 0x270 + (n) * 8 + 4)
-#define GEN8_RING_PDP_LDW(engine, n)	_MMIO((engine)->mmio_base + 0x270 + (n) * 8)
+#define GEN8_RING_PDP_UDW(base, n)	_MMIO((base) + 0x270 + (n) * 8 + 4)
+#define GEN8_RING_PDP_LDW(base, n)	_MMIO((base) + 0x270 + (n) * 8)
 
 #define GEN8_R_PWR_CLK_STATE		_MMIO(0x20C8)
 #define   GEN8_RPCS_ENABLE		(1 << 31)
@@ -1044,7 +1106,32 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg)
 /* See configdb bunit SB addr map */
 #define BUNIT_REG_BISOC				0x11
 
-#define PUNIT_REG_DSPFREQ			0x36
+/* PUNIT_REG_*SSPM0 */
+#define   _SSPM0_SSC(val)			((val) << 0)
+#define   SSPM0_SSC_MASK			_SSPM0_SSC(0x3)
+#define   SSPM0_SSC_PWR_ON			_SSPM0_SSC(0x0)
+#define   SSPM0_SSC_CLK_GATE			_SSPM0_SSC(0x1)
+#define   SSPM0_SSC_RESET			_SSPM0_SSC(0x2)
+#define   SSPM0_SSC_PWR_GATE			_SSPM0_SSC(0x3)
+#define   _SSPM0_SSS(val)			((val) << 24)
+#define   SSPM0_SSS_MASK			_SSPM0_SSS(0x3)
+#define   SSPM0_SSS_PWR_ON			_SSPM0_SSS(0x0)
+#define   SSPM0_SSS_CLK_GATE			_SSPM0_SSS(0x1)
+#define   SSPM0_SSS_RESET			_SSPM0_SSS(0x2)
+#define   SSPM0_SSS_PWR_GATE			_SSPM0_SSS(0x3)
+
+/* PUNIT_REG_*SSPM1 */
+#define   SSPM1_FREQSTAT_SHIFT			24
+#define   SSPM1_FREQSTAT_MASK			(0x1f << SSPM1_FREQSTAT_SHIFT)
+#define   SSPM1_FREQGUAR_SHIFT			8
+#define   SSPM1_FREQGUAR_MASK			(0x1f << SSPM1_FREQGUAR_SHIFT)
+#define   SSPM1_FREQ_SHIFT			0
+#define   SSPM1_FREQ_MASK			(0x1f << SSPM1_FREQ_SHIFT)
+
+#define PUNIT_REG_VEDSSPM0			0x32
+#define PUNIT_REG_VEDSSPM1			0x33
+
+#define PUNIT_REG_DSPSSPM			0x36
 #define   DSPFREQSTAT_SHIFT_CHV			24
 #define   DSPFREQSTAT_MASK_CHV			(0x1f << DSPFREQSTAT_SHIFT_CHV)
 #define   DSPFREQGUAR_SHIFT_CHV			8
@@ -1069,6 +1156,9 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg)
 #define   DP_SSS_RESET(pipe)			_DP_SSS(0x2, (pipe))
 #define   DP_SSS_PWR_GATE(pipe)			_DP_SSS(0x3, (pipe))
 
+#define PUNIT_REG_ISPSSPM0			0x39
+#define PUNIT_REG_ISPSSPM1			0x3a
+
 /*
  * i915_power_well_id:
  *
@@ -1860,13 +1950,13 @@ enum i915_power_well_id {
 #define _CNL_PORT_TX_DW4_LN1_AE		0x1624D0
 #define CNL_PORT_TX_DW4_GRP(port)	_MMIO(_CNL_PORT_TX_DW_GRP(4, (port)))
 #define CNL_PORT_TX_DW4_LN0(port)	_MMIO(_CNL_PORT_TX_DW_LN0(4, (port)))
-#define CNL_PORT_TX_DW4_LN(port, ln)   _MMIO(_CNL_PORT_TX_DW_LN0(4, (port)) + \
+#define CNL_PORT_TX_DW4_LN(ln, port)   _MMIO(_CNL_PORT_TX_DW_LN0(4, (port)) + \
 					   ((ln) * (_CNL_PORT_TX_DW4_LN1_AE - \
 						    _CNL_PORT_TX_DW4_LN0_AE)))
 #define ICL_PORT_TX_DW4_AUX(port)	_MMIO(_ICL_PORT_TX_DW_AUX(4, port))
 #define ICL_PORT_TX_DW4_GRP(port)	_MMIO(_ICL_PORT_TX_DW_GRP(4, port))
 #define ICL_PORT_TX_DW4_LN0(port)	_MMIO(_ICL_PORT_TX_DW_LN(4, 0, port))
-#define ICL_PORT_TX_DW4_LN(port, ln)	_MMIO(_ICL_PORT_TX_DW_LN(4, ln, port))
+#define ICL_PORT_TX_DW4_LN(ln, port)	_MMIO(_ICL_PORT_TX_DW_LN(4, ln, port))
 #define   LOADGEN_SELECT		(1 << 31)
 #define   POST_CURSOR_1(x)		((x) << 12)
 #define   POST_CURSOR_1_MASK		(0x3F << 12)
@@ -1893,11 +1983,11 @@ enum i915_power_well_id {
 #define ICL_PORT_TX_DW7_AUX(port)	_MMIO(_ICL_PORT_TX_DW_AUX(7, port))
 #define ICL_PORT_TX_DW7_GRP(port)	_MMIO(_ICL_PORT_TX_DW_GRP(7, port))
 #define ICL_PORT_TX_DW7_LN0(port)	_MMIO(_ICL_PORT_TX_DW_LN(7, 0, port))
-#define ICL_PORT_TX_DW7_LN(port, ln)	_MMIO(_ICL_PORT_TX_DW_LN(7, ln, port))
+#define ICL_PORT_TX_DW7_LN(ln, port)	_MMIO(_ICL_PORT_TX_DW_LN(7, ln, port))
 #define   N_SCALAR(x)			((x) << 24)
 #define   N_SCALAR_MASK			(0x7F << 24)
 
-#define MG_PHY_PORT_LN(port, ln, ln0p1, ln0p2, ln1p1) \
+#define MG_PHY_PORT_LN(ln, port, ln0p1, ln0p2, ln1p1) \
 	_MMIO(_PORT((port) - PORT_C, ln0p1, ln0p2) + (ln) * ((ln1p1) - (ln0p1)))
 
 #define MG_TX_LINK_PARAMS_TX1LN0_PORT1		0x16812C
@@ -1908,8 +1998,8 @@ enum i915_power_well_id {
 #define MG_TX_LINK_PARAMS_TX1LN1_PORT3		0x16A52C
 #define MG_TX_LINK_PARAMS_TX1LN0_PORT4		0x16B12C
 #define MG_TX_LINK_PARAMS_TX1LN1_PORT4		0x16B52C
-#define MG_TX1_LINK_PARAMS(port, ln) \
-	MG_PHY_PORT_LN(port, ln, MG_TX_LINK_PARAMS_TX1LN0_PORT1, \
+#define MG_TX1_LINK_PARAMS(ln, port) \
+	MG_PHY_PORT_LN(ln, port, MG_TX_LINK_PARAMS_TX1LN0_PORT1, \
 				 MG_TX_LINK_PARAMS_TX1LN0_PORT2, \
 				 MG_TX_LINK_PARAMS_TX1LN1_PORT1)
 
@@ -1921,8 +2011,8 @@ enum i915_power_well_id {
 #define MG_TX_LINK_PARAMS_TX2LN1_PORT3		0x16A4AC
 #define MG_TX_LINK_PARAMS_TX2LN0_PORT4		0x16B0AC
 #define MG_TX_LINK_PARAMS_TX2LN1_PORT4		0x16B4AC
-#define MG_TX2_LINK_PARAMS(port, ln) \
-	MG_PHY_PORT_LN(port, ln, MG_TX_LINK_PARAMS_TX2LN0_PORT1, \
+#define MG_TX2_LINK_PARAMS(ln, port) \
+	MG_PHY_PORT_LN(ln, port, MG_TX_LINK_PARAMS_TX2LN0_PORT1, \
 				 MG_TX_LINK_PARAMS_TX2LN0_PORT2, \
 				 MG_TX_LINK_PARAMS_TX2LN1_PORT1)
 #define   CRI_USE_FS32			(1 << 5)
@@ -1935,8 +2025,8 @@ enum i915_power_well_id {
 #define MG_TX_PISO_READLOAD_TX1LN1_PORT3		0x16A54C
 #define MG_TX_PISO_READLOAD_TX1LN0_PORT4		0x16B14C
 #define MG_TX_PISO_READLOAD_TX1LN1_PORT4		0x16B54C
-#define MG_TX1_PISO_READLOAD(port, ln) \
-	MG_PHY_PORT_LN(port, ln, MG_TX_PISO_READLOAD_TX1LN0_PORT1, \
+#define MG_TX1_PISO_READLOAD(ln, port) \
+	MG_PHY_PORT_LN(ln, port, MG_TX_PISO_READLOAD_TX1LN0_PORT1, \
 				 MG_TX_PISO_READLOAD_TX1LN0_PORT2, \
 				 MG_TX_PISO_READLOAD_TX1LN1_PORT1)
 
@@ -1948,8 +2038,8 @@ enum i915_power_well_id {
 #define MG_TX_PISO_READLOAD_TX2LN1_PORT3		0x16A4CC
 #define MG_TX_PISO_READLOAD_TX2LN0_PORT4		0x16B0CC
 #define MG_TX_PISO_READLOAD_TX2LN1_PORT4		0x16B4CC
-#define MG_TX2_PISO_READLOAD(port, ln) \
-	MG_PHY_PORT_LN(port, ln, MG_TX_PISO_READLOAD_TX2LN0_PORT1, \
+#define MG_TX2_PISO_READLOAD(ln, port) \
+	MG_PHY_PORT_LN(ln, port, MG_TX_PISO_READLOAD_TX2LN0_PORT1, \
 				 MG_TX_PISO_READLOAD_TX2LN0_PORT2, \
 				 MG_TX_PISO_READLOAD_TX2LN1_PORT1)
 #define   CRI_CALCINIT					(1 << 1)
@@ -1962,8 +2052,8 @@ enum i915_power_well_id {
 #define MG_TX_SWINGCTRL_TX1LN1_PORT3		0x16A548
 #define MG_TX_SWINGCTRL_TX1LN0_PORT4		0x16B148
 #define MG_TX_SWINGCTRL_TX1LN1_PORT4		0x16B548
-#define MG_TX1_SWINGCTRL(port, ln) \
-	MG_PHY_PORT_LN(port, ln, MG_TX_SWINGCTRL_TX1LN0_PORT1, \
+#define MG_TX1_SWINGCTRL(ln, port) \
+	MG_PHY_PORT_LN(ln, port, MG_TX_SWINGCTRL_TX1LN0_PORT1, \
 				 MG_TX_SWINGCTRL_TX1LN0_PORT2, \
 				 MG_TX_SWINGCTRL_TX1LN1_PORT1)
 
@@ -1975,8 +2065,8 @@ enum i915_power_well_id {
 #define MG_TX_SWINGCTRL_TX2LN1_PORT3		0x16A4C8
 #define MG_TX_SWINGCTRL_TX2LN0_PORT4		0x16B0C8
 #define MG_TX_SWINGCTRL_TX2LN1_PORT4		0x16B4C8
-#define MG_TX2_SWINGCTRL(port, ln) \
-	MG_PHY_PORT_LN(port, ln, MG_TX_SWINGCTRL_TX2LN0_PORT1, \
+#define MG_TX2_SWINGCTRL(ln, port) \
+	MG_PHY_PORT_LN(ln, port, MG_TX_SWINGCTRL_TX2LN0_PORT1, \
 				 MG_TX_SWINGCTRL_TX2LN0_PORT2, \
 				 MG_TX_SWINGCTRL_TX2LN1_PORT1)
 #define   CRI_TXDEEMPH_OVERRIDE_17_12(x)		((x) << 0)
@@ -1990,8 +2080,8 @@ enum i915_power_well_id {
 #define MG_TX_DRVCTRL_TX1LN1_TXPORT3			0x16A544
 #define MG_TX_DRVCTRL_TX1LN0_TXPORT4			0x16B144
 #define MG_TX_DRVCTRL_TX1LN1_TXPORT4			0x16B544
-#define MG_TX1_DRVCTRL(port, ln) \
-	MG_PHY_PORT_LN(port, ln, MG_TX_DRVCTRL_TX1LN0_TXPORT1, \
+#define MG_TX1_DRVCTRL(ln, port) \
+	MG_PHY_PORT_LN(ln, port, MG_TX_DRVCTRL_TX1LN0_TXPORT1, \
 				 MG_TX_DRVCTRL_TX1LN0_TXPORT2, \
 				 MG_TX_DRVCTRL_TX1LN1_TXPORT1)
 
@@ -2003,8 +2093,8 @@ enum i915_power_well_id {
 #define MG_TX_DRVCTRL_TX2LN1_PORT3			0x16A4C4
 #define MG_TX_DRVCTRL_TX2LN0_PORT4			0x16B0C4
 #define MG_TX_DRVCTRL_TX2LN1_PORT4			0x16B4C4
-#define MG_TX2_DRVCTRL(port, ln) \
-	MG_PHY_PORT_LN(port, ln, MG_TX_DRVCTRL_TX2LN0_PORT1, \
+#define MG_TX2_DRVCTRL(ln, port) \
+	MG_PHY_PORT_LN(ln, port, MG_TX_DRVCTRL_TX2LN0_PORT1, \
 				 MG_TX_DRVCTRL_TX2LN0_PORT2, \
 				 MG_TX_DRVCTRL_TX2LN1_PORT1)
 #define   CRI_TXDEEMPH_OVERRIDE_11_6(x)			((x) << 24)
@@ -2023,8 +2113,8 @@ enum i915_power_well_id {
 #define MG_CLKHUB_LN1_PORT3			0x16A79C
 #define MG_CLKHUB_LN0_PORT4			0x16B39C
 #define MG_CLKHUB_LN1_PORT4			0x16B79C
-#define MG_CLKHUB(port, ln) \
-	MG_PHY_PORT_LN(port, ln, MG_CLKHUB_LN0_PORT1, \
+#define MG_CLKHUB(ln, port) \
+	MG_PHY_PORT_LN(ln, port, MG_CLKHUB_LN0_PORT1, \
 				 MG_CLKHUB_LN0_PORT2, \
 				 MG_CLKHUB_LN1_PORT1)
 #define   CFG_LOW_RATE_LKREN_EN				(1 << 11)
@@ -2037,8 +2127,8 @@ enum i915_power_well_id {
 #define MG_TX_DCC_TX1LN1_PORT3			0x16A510
 #define MG_TX_DCC_TX1LN0_PORT4			0x16B110
 #define MG_TX_DCC_TX1LN1_PORT4			0x16B510
-#define MG_TX1_DCC(port, ln) \
-	MG_PHY_PORT_LN(port, ln, MG_TX_DCC_TX1LN0_PORT1, \
+#define MG_TX1_DCC(ln, port) \
+	MG_PHY_PORT_LN(ln, port, MG_TX_DCC_TX1LN0_PORT1, \
 				 MG_TX_DCC_TX1LN0_PORT2, \
 				 MG_TX_DCC_TX1LN1_PORT1)
 #define MG_TX_DCC_TX2LN0_PORT1			0x168090
@@ -2049,8 +2139,8 @@ enum i915_power_well_id {
 #define MG_TX_DCC_TX2LN1_PORT3			0x16A490
 #define MG_TX_DCC_TX2LN0_PORT4			0x16B090
 #define MG_TX_DCC_TX2LN1_PORT4			0x16B490
-#define MG_TX2_DCC(port, ln) \
-	MG_PHY_PORT_LN(port, ln, MG_TX_DCC_TX2LN0_PORT1, \
+#define MG_TX2_DCC(ln, port) \
+	MG_PHY_PORT_LN(ln, port, MG_TX_DCC_TX2LN0_PORT1, \
 				 MG_TX_DCC_TX2LN0_PORT2, \
 				 MG_TX_DCC_TX2LN1_PORT1)
 #define   CFG_AMI_CK_DIV_OVERRIDE_VAL(x)	((x) << 25)
@@ -2065,8 +2155,8 @@ enum i915_power_well_id {
 #define MG_DP_MODE_LN1_ACU_PORT3			0x16A7A0
 #define MG_DP_MODE_LN0_ACU_PORT4			0x16B3A0
 #define MG_DP_MODE_LN1_ACU_PORT4			0x16B7A0
-#define MG_DP_MODE(port, ln)	\
-	MG_PHY_PORT_LN(port, ln, MG_DP_MODE_LN0_ACU_PORT1, \
+#define MG_DP_MODE(ln, port)	\
+	MG_PHY_PORT_LN(ln, port, MG_DP_MODE_LN0_ACU_PORT1, \
 				 MG_DP_MODE_LN0_ACU_PORT2, \
 				 MG_DP_MODE_LN1_ACU_PORT1)
 #define   MG_DP_MODE_CFG_DP_X2_MODE			(1 << 7)
@@ -2356,8 +2446,10 @@ enum i915_power_well_id {
 #define RING_HWS_PGA(base)	_MMIO((base) + 0x80)
 #define RING_HWS_PGA_GEN6(base)	_MMIO((base) + 0x2080)
 #define RING_RESET_CTL(base)	_MMIO((base) + 0xd0)
-#define   RESET_CTL_REQUEST_RESET  (1 << 0)
-#define   RESET_CTL_READY_TO_RESET (1 << 1)
+#define   RESET_CTL_CAT_ERROR	   REG_BIT(2)
+#define   RESET_CTL_READY_TO_RESET REG_BIT(1)
+#define   RESET_CTL_REQUEST_RESET  REG_BIT(0)
+
 #define RING_SEMA_WAIT_POLL(base) _MMIO((base) + 0x24c)
 
 #define HSW_GTT_CACHE_EN	_MMIO(0x4024)
@@ -2478,12 +2570,12 @@ enum i915_power_well_id {
 #define HWS_START_ADDRESS_SHIFT	4
 #define PWRCTXA		_MMIO(0x2088) /* 965GM+ only */
 #define   PWRCTX_EN	(1 << 0)
-#define IPEIR		_MMIO(0x2088)
-#define IPEHR		_MMIO(0x208c)
+#define IPEIR(base)	_MMIO((base) + 0x88)
+#define IPEHR(base)	_MMIO((base) + 0x8c)
 #define GEN2_INSTDONE	_MMIO(0x2090)
 #define NOPID		_MMIO(0x2094)
 #define HWSTAM		_MMIO(0x2098)
-#define DMA_FADD_I8XX	_MMIO(0x20d0)
+#define DMA_FADD_I8XX(base)	_MMIO((base) + 0xd0)
 #define RING_BBSTATE(base)	_MMIO((base) + 0x110)
 #define   RING_BB_PPGTT		(1 << 5)
 #define RING_SBBADDR(base)	_MMIO((base) + 0x114) /* hsw+ */
@@ -2623,10 +2715,10 @@ enum i915_power_well_id {
 #define VLV_GU_CTL0	_MMIO(VLV_DISPLAY_BASE + 0x2030)
 #define VLV_GU_CTL1	_MMIO(VLV_DISPLAY_BASE + 0x2034)
 #define SCPD0		_MMIO(0x209c) /* 915+ only */
-#define IER		_MMIO(0x20a0)
-#define IIR		_MMIO(0x20a4)
-#define IMR		_MMIO(0x20a8)
-#define ISR		_MMIO(0x20ac)
+#define GEN2_IER	_MMIO(0x20a0)
+#define GEN2_IIR	_MMIO(0x20a4)
+#define GEN2_IMR	_MMIO(0x20a8)
+#define GEN2_ISR	_MMIO(0x20ac)
 #define VLV_GUNIT_CLOCK_GATE	_MMIO(VLV_DISPLAY_BASE + 0x2060)
 #define   GINT_DIS		(1 << 22)
 #define   GCFG_DIS		(1 << 8)
@@ -2657,7 +2749,7 @@ enum i915_power_well_id {
 #define   INSTPM_FORCE_ORDERING				(1 << 7) /* GEN6+ */
 #define   INSTPM_TLB_INVALIDATE	(1 << 9)
 #define   INSTPM_SYNC_FLUSH	(1 << 5)
-#define ACTHD	        _MMIO(0x20c8)
+#define ACTHD(base)	_MMIO((base) + 0xc8)
 #define MEM_MODE	_MMIO(0x20cc)
 #define   MEM_DISPLAY_B_TRICKLE_FEED_DISABLE (1 << 3) /* 830 only */
 #define   MEM_DISPLAY_A_TRICKLE_FEED_DISABLE (1 << 2) /* 830/845 only */
@@ -3857,7 +3949,7 @@ enum i915_power_well_id {
 /*
  * Logical Context regs
  */
-#define CCID				_MMIO(0x2180)
+#define CCID(base)			_MMIO((base) + 0x180)
 #define   CCID_EN			BIT(0)
 #define   CCID_EXTENDED_STATE_RESTORE	BIT(2)
 #define   CCID_EXTENDED_STATE_SAVE	BIT(3)
@@ -3989,6 +4081,15 @@ enum {
 /* Pipe A CRC regs */
 #define _PIPE_CRC_CTL_A			0x60050
 #define   PIPE_CRC_ENABLE		(1 << 31)
+/* skl+ source selection */
+#define   PIPE_CRC_SOURCE_PLANE_1_SKL	(0 << 28)
+#define   PIPE_CRC_SOURCE_PLANE_2_SKL	(2 << 28)
+#define   PIPE_CRC_SOURCE_DMUX_SKL	(4 << 28)
+#define   PIPE_CRC_SOURCE_PLANE_3_SKL	(6 << 28)
+#define   PIPE_CRC_SOURCE_PLANE_4_SKL	(7 << 28)
+#define   PIPE_CRC_SOURCE_PLANE_5_SKL	(5 << 28)
+#define   PIPE_CRC_SOURCE_PLANE_6_SKL	(3 << 28)
+#define   PIPE_CRC_SOURCE_PLANE_7_SKL	(1 << 28)
 /* ivb+ source selection */
 #define   PIPE_CRC_SOURCE_PRIMARY_IVB	(0 << 29)
 #define   PIPE_CRC_SOURCE_SPRITE_IVB	(1 << 29)
@@ -4110,42 +4211,6 @@ enum {
 #define PIPESRC(trans)		_MMIO_TRANS2(trans, _PIPEASRC)
 #define PIPE_MULT(trans)	_MMIO_TRANS2(trans, _PIPE_MULT_A)
 
-/* VLV eDP PSR registers */
-#define _PSRCTLA				(VLV_DISPLAY_BASE + 0x60090)
-#define _PSRCTLB				(VLV_DISPLAY_BASE + 0x61090)
-#define  VLV_EDP_PSR_ENABLE			(1 << 0)
-#define  VLV_EDP_PSR_RESET			(1 << 1)
-#define  VLV_EDP_PSR_MODE_MASK			(7 << 2)
-#define  VLV_EDP_PSR_MODE_HW_TIMER		(1 << 3)
-#define  VLV_EDP_PSR_MODE_SW_TIMER		(1 << 2)
-#define  VLV_EDP_PSR_SINGLE_FRAME_UPDATE	(1 << 7)
-#define  VLV_EDP_PSR_ACTIVE_ENTRY		(1 << 8)
-#define  VLV_EDP_PSR_SRC_TRANSMITTER_STATE	(1 << 9)
-#define  VLV_EDP_PSR_DBL_FRAME			(1 << 10)
-#define  VLV_EDP_PSR_FRAME_COUNT_MASK		(0xff << 16)
-#define  VLV_EDP_PSR_IDLE_FRAME_SHIFT		16
-#define VLV_PSRCTL(pipe)	_MMIO_PIPE(pipe, _PSRCTLA, _PSRCTLB)
-
-#define _VSCSDPA			(VLV_DISPLAY_BASE + 0x600a0)
-#define _VSCSDPB			(VLV_DISPLAY_BASE + 0x610a0)
-#define  VLV_EDP_PSR_SDP_FREQ_MASK	(3 << 30)
-#define  VLV_EDP_PSR_SDP_FREQ_ONCE	(1 << 31)
-#define  VLV_EDP_PSR_SDP_FREQ_EVFRAME	(1 << 30)
-#define VLV_VSCSDP(pipe)	_MMIO_PIPE(pipe, _VSCSDPA, _VSCSDPB)
-
-#define _PSRSTATA			(VLV_DISPLAY_BASE + 0x60094)
-#define _PSRSTATB			(VLV_DISPLAY_BASE + 0x61094)
-#define  VLV_EDP_PSR_LAST_STATE_MASK	(7 << 3)
-#define  VLV_EDP_PSR_CURR_STATE_MASK	7
-#define  VLV_EDP_PSR_DISABLED		(0 << 0)
-#define  VLV_EDP_PSR_INACTIVE		(1 << 0)
-#define  VLV_EDP_PSR_IN_TRANS_TO_ACTIVE	(2 << 0)
-#define  VLV_EDP_PSR_ACTIVE_NORFB_UP	(3 << 0)
-#define  VLV_EDP_PSR_ACTIVE_SF_UPDATE	(4 << 0)
-#define  VLV_EDP_PSR_EXIT		(5 << 0)
-#define  VLV_EDP_PSR_IN_TRANS		(1 << 7)
-#define VLV_PSRSTAT(pipe)	_MMIO_PIPE(pipe, _PSRSTATA, _PSRSTATB)
-
 /* HSW+ eDP PSR registers */
 #define HSW_EDP_PSR_BASE	0x64800
 #define BDW_EDP_PSR_BASE	0x6f800
@@ -4168,6 +4233,7 @@ enum {
 #define   EDP_PSR_TP2_TP3_TIME_100us		(1 << 8)
 #define   EDP_PSR_TP2_TP3_TIME_2500us		(2 << 8)
 #define   EDP_PSR_TP2_TP3_TIME_0us		(3 << 8)
+#define   EDP_PSR_TP4_TIME_0US			(3 << 6) /* ICL+ */
 #define   EDP_PSR_TP1_TIME_500us		(0 << 4)
 #define   EDP_PSR_TP1_TIME_100us		(1 << 4)
 #define   EDP_PSR_TP1_TIME_2500us		(2 << 4)
@@ -4612,13 +4678,14 @@ enum {
 #define   VIDEO_DIP_ENABLE		(1 << 31)
 #define   VIDEO_DIP_PORT(port)		((port) << 29)
 #define   VIDEO_DIP_PORT_MASK		(3 << 29)
-#define   VIDEO_DIP_ENABLE_GCP		(1 << 25)
+#define   VIDEO_DIP_ENABLE_GCP		(1 << 25) /* ilk+ */
 #define   VIDEO_DIP_ENABLE_AVI		(1 << 21)
 #define   VIDEO_DIP_ENABLE_VENDOR	(2 << 21)
-#define   VIDEO_DIP_ENABLE_GAMUT	(4 << 21)
+#define   VIDEO_DIP_ENABLE_GAMUT	(4 << 21) /* ilk+ */
 #define   VIDEO_DIP_ENABLE_SPD		(8 << 21)
 #define   VIDEO_DIP_SELECT_AVI		(0 << 19)
 #define   VIDEO_DIP_SELECT_VENDOR	(1 << 19)
+#define   VIDEO_DIP_SELECT_GAMUT	(2 << 19)
 #define   VIDEO_DIP_SELECT_SPD		(3 << 19)
 #define   VIDEO_DIP_SELECT_MASK		(3 << 19)
 #define   VIDEO_DIP_FREQ_ONCE		(0 << 16)
@@ -4653,18 +4720,17 @@ enum {
 
 #define _PP_STATUS			0x61200
 #define PP_STATUS(pps_idx)		_MMIO_PPS(pps_idx, _PP_STATUS)
-#define   PP_ON				(1 << 31)
+#define   PP_ON				REG_BIT(31)
 
 #define _PP_CONTROL_1			0xc7204
 #define _PP_CONTROL_2			0xc7304
 #define ICP_PP_CONTROL(x)		_MMIO(((x) == 1) ? _PP_CONTROL_1 : \
 					      _PP_CONTROL_2)
-#define  POWER_CYCLE_DELAY_MASK	(0x1f << 4)
-#define  POWER_CYCLE_DELAY_SHIFT	4
-#define  VDD_OVERRIDE_FORCE		(1 << 3)
-#define  BACKLIGHT_ENABLE		(1 << 2)
-#define  PWR_DOWN_ON_RESET		(1 << 1)
-#define  PWR_STATE_TARGET		(1 << 0)
+#define  POWER_CYCLE_DELAY_MASK		REG_GENMASK(8, 4)
+#define  VDD_OVERRIDE_FORCE		REG_BIT(3)
+#define  BACKLIGHT_ENABLE		REG_BIT(2)
+#define  PWR_DOWN_ON_RESET		REG_BIT(1)
+#define  PWR_STATE_TARGET		REG_BIT(0)
 /*
  * Indicates that all dependencies of the panel are on:
  *
@@ -4672,62 +4738,53 @@ enum {
  * - pipe enabled
  * - LVDS/DVOB/DVOC on
  */
-#define   PP_READY			(1 << 30)
-#define   PP_SEQUENCE_NONE		(0 << 28)
-#define   PP_SEQUENCE_POWER_UP		(1 << 28)
-#define   PP_SEQUENCE_POWER_DOWN	(2 << 28)
-#define   PP_SEQUENCE_MASK		(3 << 28)
-#define   PP_SEQUENCE_SHIFT		28
-#define   PP_CYCLE_DELAY_ACTIVE		(1 << 27)
-#define   PP_SEQUENCE_STATE_MASK	0x0000000f
-#define   PP_SEQUENCE_STATE_OFF_IDLE	(0x0 << 0)
-#define   PP_SEQUENCE_STATE_OFF_S0_1	(0x1 << 0)
-#define   PP_SEQUENCE_STATE_OFF_S0_2	(0x2 << 0)
-#define   PP_SEQUENCE_STATE_OFF_S0_3	(0x3 << 0)
-#define   PP_SEQUENCE_STATE_ON_IDLE	(0x8 << 0)
-#define   PP_SEQUENCE_STATE_ON_S1_0	(0x9 << 0)
-#define   PP_SEQUENCE_STATE_ON_S1_2	(0xa << 0)
-#define   PP_SEQUENCE_STATE_ON_S1_3	(0xb << 0)
-#define   PP_SEQUENCE_STATE_RESET	(0xf << 0)
+#define   PP_READY			REG_BIT(30)
+#define   PP_SEQUENCE_MASK		REG_GENMASK(29, 28)
+#define   PP_SEQUENCE_NONE		REG_FIELD_PREP(PP_SEQUENCE_MASK, 0)
+#define   PP_SEQUENCE_POWER_UP		REG_FIELD_PREP(PP_SEQUENCE_MASK, 1)
+#define   PP_SEQUENCE_POWER_DOWN	REG_FIELD_PREP(PP_SEQUENCE_MASK, 2)
+#define   PP_CYCLE_DELAY_ACTIVE		REG_BIT(27)
+#define   PP_SEQUENCE_STATE_MASK	REG_GENMASK(3, 0)
+#define   PP_SEQUENCE_STATE_OFF_IDLE	REG_FIELD_PREP(PP_SEQUENCE_STATE_MASK, 0x0)
+#define   PP_SEQUENCE_STATE_OFF_S0_1	REG_FIELD_PREP(PP_SEQUENCE_STATE_MASK, 0x1)
+#define   PP_SEQUENCE_STATE_OFF_S0_2	REG_FIELD_PREP(PP_SEQUENCE_STATE_MASK, 0x2)
+#define   PP_SEQUENCE_STATE_OFF_S0_3	REG_FIELD_PREP(PP_SEQUENCE_STATE_MASK, 0x3)
+#define   PP_SEQUENCE_STATE_ON_IDLE	REG_FIELD_PREP(PP_SEQUENCE_STATE_MASK, 0x8)
+#define   PP_SEQUENCE_STATE_ON_S1_1	REG_FIELD_PREP(PP_SEQUENCE_STATE_MASK, 0x9)
+#define   PP_SEQUENCE_STATE_ON_S1_2	REG_FIELD_PREP(PP_SEQUENCE_STATE_MASK, 0xa)
+#define   PP_SEQUENCE_STATE_ON_S1_3	REG_FIELD_PREP(PP_SEQUENCE_STATE_MASK, 0xb)
+#define   PP_SEQUENCE_STATE_RESET	REG_FIELD_PREP(PP_SEQUENCE_STATE_MASK, 0xf)
 
 #define _PP_CONTROL			0x61204
 #define PP_CONTROL(pps_idx)		_MMIO_PPS(pps_idx, _PP_CONTROL)
-#define  PANEL_UNLOCK_REGS		(0xabcd << 16)
-#define  PANEL_UNLOCK_MASK		(0xffff << 16)
-#define  BXT_POWER_CYCLE_DELAY_MASK	0x1f0
-#define  BXT_POWER_CYCLE_DELAY_SHIFT	4
-#define  EDP_FORCE_VDD			(1 << 3)
-#define  EDP_BLC_ENABLE			(1 << 2)
-#define  PANEL_POWER_RESET		(1 << 1)
-#define  PANEL_POWER_ON			(1 << 0)
+#define  PANEL_UNLOCK_MASK		REG_GENMASK(31, 16)
+#define  PANEL_UNLOCK_REGS		REG_FIELD_PREP(PANEL_UNLOCK_MASK, 0xabcd)
+#define  BXT_POWER_CYCLE_DELAY_MASK	REG_GENMASK(8, 4)
+#define  EDP_FORCE_VDD			REG_BIT(3)
+#define  EDP_BLC_ENABLE			REG_BIT(2)
+#define  PANEL_POWER_RESET		REG_BIT(1)
+#define  PANEL_POWER_ON			REG_BIT(0)
 
 #define _PP_ON_DELAYS			0x61208
 #define PP_ON_DELAYS(pps_idx)		_MMIO_PPS(pps_idx, _PP_ON_DELAYS)
-#define  PANEL_PORT_SELECT_SHIFT	30
-#define  PANEL_PORT_SELECT_MASK		(3 << 30)
-#define  PANEL_PORT_SELECT_LVDS		(0 << 30)
-#define  PANEL_PORT_SELECT_DPA		(1 << 30)
-#define  PANEL_PORT_SELECT_DPC		(2 << 30)
-#define  PANEL_PORT_SELECT_DPD		(3 << 30)
-#define  PANEL_PORT_SELECT_VLV(port)	((port) << 30)
-#define  PANEL_POWER_UP_DELAY_MASK	0x1fff0000
-#define  PANEL_POWER_UP_DELAY_SHIFT	16
-#define  PANEL_LIGHT_ON_DELAY_MASK	0x1fff
-#define  PANEL_LIGHT_ON_DELAY_SHIFT	0
+#define  PANEL_PORT_SELECT_MASK		REG_GENMASK(31, 30)
+#define  PANEL_PORT_SELECT_LVDS		REG_FIELD_PREP(PANEL_PORT_SELECT_MASK, 0)
+#define  PANEL_PORT_SELECT_DPA		REG_FIELD_PREP(PANEL_PORT_SELECT_MASK, 1)
+#define  PANEL_PORT_SELECT_DPC		REG_FIELD_PREP(PANEL_PORT_SELECT_MASK, 2)
+#define  PANEL_PORT_SELECT_DPD		REG_FIELD_PREP(PANEL_PORT_SELECT_MASK, 3)
+#define  PANEL_PORT_SELECT_VLV(port)	REG_FIELD_PREP(PANEL_PORT_SELECT_MASK, port)
+#define  PANEL_POWER_UP_DELAY_MASK	REG_GENMASK(28, 16)
+#define  PANEL_LIGHT_ON_DELAY_MASK	REG_GENMASK(12, 0)
 
 #define _PP_OFF_DELAYS			0x6120C
 #define PP_OFF_DELAYS(pps_idx)		_MMIO_PPS(pps_idx, _PP_OFF_DELAYS)
-#define  PANEL_POWER_DOWN_DELAY_MASK	0x1fff0000
-#define  PANEL_POWER_DOWN_DELAY_SHIFT	16
-#define  PANEL_LIGHT_OFF_DELAY_MASK	0x1fff
-#define  PANEL_LIGHT_OFF_DELAY_SHIFT	0
+#define  PANEL_POWER_DOWN_DELAY_MASK	REG_GENMASK(28, 16)
+#define  PANEL_LIGHT_OFF_DELAY_MASK	REG_GENMASK(12, 0)
 
 #define _PP_DIVISOR			0x61210
 #define PP_DIVISOR(pps_idx)		_MMIO_PPS(pps_idx, _PP_DIVISOR)
-#define  PP_REFERENCE_DIVIDER_MASK	0xffffff00
-#define  PP_REFERENCE_DIVIDER_SHIFT	8
-#define  PANEL_POWER_CYCLE_DELAY_MASK	0x1f
-#define  PANEL_POWER_CYCLE_DELAY_SHIFT	0
+#define  PP_REFERENCE_DIVIDER_MASK	REG_GENMASK(31, 8)
+#define  PANEL_POWER_CYCLE_DELAY_MASK	REG_GENMASK(4, 0)
 
 /* Panel fitting */
 #define PFIT_CONTROL	_MMIO(DISPLAY_MMIO_BASE(dev_priv) + 0x61230)
@@ -5590,9 +5647,15 @@ enum {
 #define   PIPECONF_SINGLE_WIDE	0
 #define   PIPECONF_PIPE_UNLOCKED 0
 #define   PIPECONF_PIPE_LOCKED	(1 << 25)
-#define   PIPECONF_PALETTE	0
-#define   PIPECONF_GAMMA		(1 << 24)
 #define   PIPECONF_FORCE_BORDER	(1 << 25)
+#define   PIPECONF_GAMMA_MODE_MASK_I9XX	(1 << 24) /* gmch */
+#define   PIPECONF_GAMMA_MODE_MASK_ILK	(3 << 24) /* ilk-ivb */
+#define   PIPECONF_GAMMA_MODE_8BIT	(0 << 24) /* gmch,ilk-ivb */
+#define   PIPECONF_GAMMA_MODE_10BIT	(1 << 24) /* gmch,ilk-ivb */
+#define   PIPECONF_GAMMA_MODE_12BIT	(2 << 24) /* ilk-ivb */
+#define   PIPECONF_GAMMA_MODE_SPLIT	(3 << 24) /* ivb */
+#define   PIPECONF_GAMMA_MODE(x)	((x) << 24) /* pass in GAMMA_MODE_MODE_* */
+#define   PIPECONF_GAMMA_MODE_SHIFT	24
 #define   PIPECONF_INTERLACE_MASK	(7 << 21)
 #define   PIPECONF_INTERLACE_MASK_HSW	(3 << 21)
 /* Note that pre-gen3 does not support interlaced display directly. Panel
@@ -5698,6 +5761,10 @@ enum {
 #define PIPEFRAMEPIXEL(pipe)	_MMIO_PIPE2(pipe, _PIPEAFRAMEPIXEL)
 #define PIPESTAT(pipe)		_MMIO_PIPE2(pipe, _PIPEASTAT)
 
+#define  _PIPEAGCMAX           0x70010
+#define  _PIPEBGCMAX           0x71010
+#define PIPEGCMAX(pipe, i)     _MMIO_PIPE2(pipe, _PIPEAGCMAX + (i) * 4)
+
 #define _PIPE_MISC_A			0x70030
 #define _PIPE_MISC_B			0x71030
 #define   PIPEMISC_YUV420_ENABLE	(1 << 27)
@@ -5998,6 +6065,7 @@ enum {
 #define _CUR_WM_TRANS_A_0	0x70168
 #define _CUR_WM_TRANS_B_0	0x71168
 #define   PLANE_WM_EN		(1 << 31)
+#define   PLANE_WM_IGNORE_LINES	(1 << 30)
 #define   PLANE_WM_LINES_SHIFT	14
 #define   PLANE_WM_LINES_MASK	0x1f
 #define   PLANE_WM_BLOCKS_MASK	0x7ff /* skl+: 10 bits, icl+ 11 bits */
@@ -6124,7 +6192,7 @@ enum {
 #define   MCURSOR_PIPE_SELECT_SHIFT	28
 #define   MCURSOR_PIPE_SELECT(pipe)	((pipe) << 28)
 #define   MCURSOR_GAMMA_ENABLE  (1 << 26)
-#define   MCURSOR_PIPE_CSC_ENABLE (1 << 24)
+#define   MCURSOR_PIPE_CSC_ENABLE (1 << 24) /* ilk+ */
 #define   MCURSOR_ROTATE_180	(1 << 15)
 #define   MCURSOR_TRICKLE_FEED_DISABLE	(1 << 14)
 #define _CURABASE		0x70084
@@ -6179,7 +6247,7 @@ enum {
 #define   DISPPLANE_RGBA888			(0xf << 26)
 #define   DISPPLANE_STEREO_ENABLE		(1 << 25)
 #define   DISPPLANE_STEREO_DISABLE		0
-#define   DISPPLANE_PIPE_CSC_ENABLE		(1 << 24)
+#define   DISPPLANE_PIPE_CSC_ENABLE		(1 << 24) /* ilk+ */
 #define   DISPPLANE_SEL_PIPE_SHIFT		24
 #define   DISPPLANE_SEL_PIPE_MASK		(3 << DISPPLANE_SEL_PIPE_SHIFT)
 #define   DISPPLANE_SEL_PIPE(pipe)		((pipe) << DISPPLANE_SEL_PIPE_SHIFT)
@@ -6557,13 +6625,22 @@ enum {
 #define   PLANE_CTL_FORMAT_YUV422		(0 << 24)
 #define   PLANE_CTL_FORMAT_NV12			(1 << 24)
 #define   PLANE_CTL_FORMAT_XRGB_2101010		(2 << 24)
+#define   PLANE_CTL_FORMAT_P010			(3 << 24)
 #define   PLANE_CTL_FORMAT_XRGB_8888		(4 << 24)
+#define   PLANE_CTL_FORMAT_P012			(5 << 24)
 #define   PLANE_CTL_FORMAT_XRGB_16161616F	(6 << 24)
+#define   PLANE_CTL_FORMAT_P016			(7 << 24)
 #define   PLANE_CTL_FORMAT_AYUV			(8 << 24)
 #define   PLANE_CTL_FORMAT_INDEXED		(12 << 24)
 #define   PLANE_CTL_FORMAT_RGB_565		(14 << 24)
 #define   ICL_PLANE_CTL_FORMAT_MASK		(0x1f << 23)
 #define   PLANE_CTL_PIPE_CSC_ENABLE		(1 << 23) /* Pre-GLK */
+#define   PLANE_CTL_FORMAT_Y210                 (1 << 23)
+#define   PLANE_CTL_FORMAT_Y212                 (3 << 23)
+#define   PLANE_CTL_FORMAT_Y216                 (5 << 23)
+#define   PLANE_CTL_FORMAT_Y410                 (7 << 23)
+#define   PLANE_CTL_FORMAT_Y412                 (9 << 23)
+#define   PLANE_CTL_FORMAT_Y416                 (0xb << 23)
 #define   PLANE_CTL_KEY_ENABLE_MASK		(0x3 << 21)
 #define   PLANE_CTL_KEY_ENABLE_SOURCE		(1 << 21)
 #define   PLANE_CTL_KEY_ENABLE_DESTINATION	(2 << 21)
@@ -7102,14 +7179,25 @@ enum {
 #define _LGC_PALETTE_B           0x4a800
 #define LGC_PALETTE(pipe, i) _MMIO(_PIPE(pipe, _LGC_PALETTE_A, _LGC_PALETTE_B) + (i) * 4)
 
+/* ilk/snb precision palette */
+#define _PREC_PALETTE_A           0x4b000
+#define _PREC_PALETTE_B           0x4c000
+#define PREC_PALETTE(pipe, i) _MMIO(_PIPE(pipe, _PREC_PALETTE_A, _PREC_PALETTE_B) + (i) * 4)
+
+#define  _PREC_PIPEAGCMAX              0x4d000
+#define  _PREC_PIPEBGCMAX              0x4d010
+#define PREC_PIPEGCMAX(pipe, i)        _MMIO(_PIPE(pipe, _PIPEAGCMAX, _PIPEBGCMAX) + (i) * 4)
+
 #define _GAMMA_MODE_A		0x4a480
 #define _GAMMA_MODE_B		0x4ac80
 #define GAMMA_MODE(pipe) _MMIO_PIPE(pipe, _GAMMA_MODE_A, _GAMMA_MODE_B)
-#define GAMMA_MODE_MODE_MASK	(3 << 0)
-#define GAMMA_MODE_MODE_8BIT	(0 << 0)
-#define GAMMA_MODE_MODE_10BIT	(1 << 0)
-#define GAMMA_MODE_MODE_12BIT	(2 << 0)
-#define GAMMA_MODE_MODE_SPLIT	(3 << 0)
+#define  PRE_CSC_GAMMA_ENABLE	(1 << 31)
+#define  POST_CSC_GAMMA_ENABLE	(1 << 30)
+#define  GAMMA_MODE_MODE_MASK	(3 << 0)
+#define  GAMMA_MODE_MODE_8BIT	(0 << 0)
+#define  GAMMA_MODE_MODE_10BIT	(1 << 0)
+#define  GAMMA_MODE_MODE_12BIT	(2 << 0)
+#define  GAMMA_MODE_MODE_SPLIT	(3 << 0)
 
 /* DMC/CSR */
 #define CSR_PROGRAM(i)		_MMIO(0x80000 + (i) * 4)
@@ -7204,8 +7292,8 @@ enum {
 #define  GEN8_GT_VECS_IRQ		(1 << 6)
 #define  GEN8_GT_GUC_IRQ		(1 << 5)
 #define  GEN8_GT_PM_IRQ			(1 << 4)
-#define  GEN8_GT_VCS2_IRQ		(1 << 3)
-#define  GEN8_GT_VCS1_IRQ		(1 << 2)
+#define  GEN8_GT_VCS1_IRQ		(1 << 3) /* NB: VCS2 in bspec! */
+#define  GEN8_GT_VCS0_IRQ		(1 << 2) /* NB: VCS1 in bpsec! */
 #define  GEN8_GT_BCS_IRQ		(1 << 1)
 #define  GEN8_GT_RCS_IRQ		(1 << 0)
 
@@ -7226,8 +7314,8 @@ enum {
 
 #define GEN8_RCS_IRQ_SHIFT 0
 #define GEN8_BCS_IRQ_SHIFT 16
-#define GEN8_VCS1_IRQ_SHIFT 0
-#define GEN8_VCS2_IRQ_SHIFT 16
+#define GEN8_VCS0_IRQ_SHIFT 0  /* NB: VCS1 in bspec! */
+#define GEN8_VCS1_IRQ_SHIFT 16 /* NB: VCS2 in bpsec! */
 #define GEN8_VECS_IRQ_SHIFT 0
 #define GEN8_WD_IRQ_SHIFT 16
 
@@ -7613,13 +7701,13 @@ enum {
 #define  GEN9_LBS_SLA_RETRY_TIMER_DECREMENT_ENABLE	(1 << 2)
 
 /*GEN11 chicken */
-#define _PIPEA_CHICKEN			0x70038
-#define _PIPEB_CHICKEN			0x71038
-#define _PIPEC_CHICKEN			0x72038
-#define  PER_PIXEL_ALPHA_BYPASS_EN	(1 << 7)
-#define  PM_FILL_MAINTAIN_DBUF_FULLNESS	(1 << 0)
-#define PIPE_CHICKEN(pipe)		_MMIO_PIPE(pipe, _PIPEA_CHICKEN,\
-						   _PIPEB_CHICKEN)
+#define _PIPEA_CHICKEN				0x70038
+#define _PIPEB_CHICKEN				0x71038
+#define _PIPEC_CHICKEN				0x72038
+#define PIPE_CHICKEN(pipe)			_MMIO_PIPE(pipe, _PIPEA_CHICKEN,\
+							   _PIPEB_CHICKEN)
+#define   PIXEL_ROUNDING_TRUNC_FB_PASSTHRU 	(1 << 15)
+#define   PER_PIXEL_ALPHA_BYPASS_EN		(1 << 7)
 
 /* PCH */
 
@@ -8089,10 +8177,11 @@ enum {
 #define _ICL_VIDEO_DIP_PPS_ECC_B	0x613D4
 
 #define HSW_TVIDEO_DIP_CTL(trans)		_MMIO_TRANS2(trans, _HSW_VIDEO_DIP_CTL_A)
+#define HSW_TVIDEO_DIP_GCP(trans)		_MMIO_TRANS2(trans, _HSW_VIDEO_DIP_GCP_A)
 #define HSW_TVIDEO_DIP_AVI_DATA(trans, i)	_MMIO_TRANS2(trans, _HSW_VIDEO_DIP_AVI_DATA_A + (i) * 4)
 #define HSW_TVIDEO_DIP_VS_DATA(trans, i)	_MMIO_TRANS2(trans, _HSW_VIDEO_DIP_VS_DATA_A + (i) * 4)
 #define HSW_TVIDEO_DIP_SPD_DATA(trans, i)	_MMIO_TRANS2(trans, _HSW_VIDEO_DIP_SPD_DATA_A + (i) * 4)
-#define HSW_TVIDEO_DIP_GCP(trans)		_MMIO_TRANS2(trans, _HSW_VIDEO_DIP_GCP_A)
+#define HSW_TVIDEO_DIP_GMP_DATA(trans, i)	_MMIO_TRANS2(trans, _HSW_VIDEO_DIP_GMP_DATA_A + (i) * 4)
 #define HSW_TVIDEO_DIP_VSC_DATA(trans, i)	_MMIO_TRANS2(trans, _HSW_VIDEO_DIP_VSC_DATA_A + (i) * 4)
 #define ICL_VIDEO_DIP_PPS_DATA(trans, i)	_MMIO_TRANS2(trans, _ICL_VIDEO_DIP_PPS_DATA_A + (i) * 4)
 #define ICL_VIDEO_DIP_PPS_ECC(trans, i)		_MMIO_TRANS2(trans, _ICL_VIDEO_DIP_PPS_ECC_A + (i) * 4)
@@ -8600,8 +8689,9 @@ enum {
 #define GEN9_MEDIA_PG_IDLE_HYSTERESIS		_MMIO(0xA0C4)
 #define GEN9_RENDER_PG_IDLE_HYSTERESIS		_MMIO(0xA0C8)
 #define GEN9_PG_ENABLE				_MMIO(0xA210)
-#define GEN9_RENDER_PG_ENABLE			(1 << 0)
-#define GEN9_MEDIA_PG_ENABLE			(1 << 1)
+#define GEN9_RENDER_PG_ENABLE			REG_BIT(0)
+#define GEN9_MEDIA_PG_ENABLE			REG_BIT(1)
+#define GEN11_MEDIA_SAMPLER_PG_ENABLE		REG_BIT(2)
 #define GEN8_PUSHBUS_CONTROL			_MMIO(0xA248)
 #define GEN8_PUSHBUS_ENABLE			_MMIO(0xA250)
 #define GEN8_PUSHBUS_SHIFT			_MMIO(0xA25C)
@@ -8616,6 +8706,11 @@ enum {
 #define GEN6_PMIER				_MMIO(0x4402C)
 #define  GEN6_PM_MBOX_EVENT			(1 << 25)
 #define  GEN6_PM_THERMAL_EVENT			(1 << 24)
+
+/*
+ * For Gen11 these are in the upper word of the GPM_WGBOXPERF
+ * registers. Shifting is handled on accessing the imr and ier.
+ */
 #define  GEN6_PM_RP_DOWN_TIMEOUT		(1 << 6)
 #define  GEN6_PM_RP_UP_THRESHOLD		(1 << 5)
 #define  GEN6_PM_RP_DOWN_THRESHOLD		(1 << 4)
@@ -9741,7 +9836,7 @@ enum skl_power_gate {
 #define  DPLL_CFGCR1_KDIV(x)		((x) << 6)
 #define  DPLL_CFGCR1_KDIV_1		(1 << 6)
 #define  DPLL_CFGCR1_KDIV_2		(2 << 6)
-#define  DPLL_CFGCR1_KDIV_4		(4 << 6)
+#define  DPLL_CFGCR1_KDIV_3		(4 << 6)
 #define  DPLL_CFGCR1_PDIV_MASK		(0xf << 2)
 #define  DPLL_CFGCR1_PDIV_SHIFT		(2)
 #define  DPLL_CFGCR1_PDIV(x)		((x) << 2)
@@ -9810,16 +9905,29 @@ enum skl_power_gate {
 #define  BXT_DRAM_WIDTH_X64			(0x3 << 4)
 #define  BXT_DRAM_SIZE_MASK			(0x7 << 6)
 #define  BXT_DRAM_SIZE_SHIFT			6
-#define  BXT_DRAM_SIZE_4GB			(0x0 << 6)
-#define  BXT_DRAM_SIZE_6GB			(0x1 << 6)
-#define  BXT_DRAM_SIZE_8GB			(0x2 << 6)
-#define  BXT_DRAM_SIZE_12GB			(0x3 << 6)
-#define  BXT_DRAM_SIZE_16GB			(0x4 << 6)
+#define  BXT_DRAM_SIZE_4GBIT			(0x0 << 6)
+#define  BXT_DRAM_SIZE_6GBIT			(0x1 << 6)
+#define  BXT_DRAM_SIZE_8GBIT			(0x2 << 6)
+#define  BXT_DRAM_SIZE_12GBIT			(0x3 << 6)
+#define  BXT_DRAM_SIZE_16GBIT			(0x4 << 6)
+#define  BXT_DRAM_TYPE_MASK			(0x7 << 22)
+#define  BXT_DRAM_TYPE_SHIFT			22
+#define  BXT_DRAM_TYPE_DDR3			(0x0 << 22)
+#define  BXT_DRAM_TYPE_LPDDR3			(0x1 << 22)
+#define  BXT_DRAM_TYPE_LPDDR4			(0x2 << 22)
+#define  BXT_DRAM_TYPE_DDR4			(0x4 << 22)
 
 #define SKL_MEMORY_FREQ_MULTIPLIER_HZ		266666666
 #define SKL_MC_BIOS_DATA_0_0_0_MCHBAR_PCU	_MMIO(MCHBAR_MIRROR_BASE_SNB + 0x5E04)
 #define  SKL_REQ_DATA_MASK			(0xF << 0)
 
+#define SKL_MAD_INTER_CHANNEL_0_0_0_MCHBAR_MCMAIN _MMIO(MCHBAR_MIRROR_BASE_SNB + 0x5000)
+#define  SKL_DRAM_DDR_TYPE_MASK			(0x3 << 0)
+#define  SKL_DRAM_DDR_TYPE_DDR4			(0 << 0)
+#define  SKL_DRAM_DDR_TYPE_DDR3			(1 << 0)
+#define  SKL_DRAM_DDR_TYPE_LPDDR3		(2 << 0)
+#define  SKL_DRAM_DDR_TYPE_LPDDR4		(3 << 0)
+
 #define SKL_MAD_DIMM_CH0_0_0_0_MCHBAR_MCMAIN	_MMIO(MCHBAR_MIRROR_BASE_SNB + 0x500C)
 #define SKL_MAD_DIMM_CH1_0_0_0_MCHBAR_MCMAIN	_MMIO(MCHBAR_MIRROR_BASE_SNB + 0x5010)
 #define  SKL_DRAM_S_SHIFT			16
@@ -9831,8 +9939,21 @@ enum skl_power_gate {
 #define  SKL_DRAM_WIDTH_X32			(0x2 << 8)
 #define  SKL_DRAM_RANK_MASK			(0x1 << 10)
 #define  SKL_DRAM_RANK_SHIFT			10
-#define  SKL_DRAM_RANK_SINGLE			(0x0 << 10)
-#define  SKL_DRAM_RANK_DUAL			(0x1 << 10)
+#define  SKL_DRAM_RANK_1			(0x0 << 10)
+#define  SKL_DRAM_RANK_2			(0x1 << 10)
+#define  SKL_DRAM_RANK_MASK			(0x1 << 10)
+#define  CNL_DRAM_SIZE_MASK			0x7F
+#define  CNL_DRAM_WIDTH_MASK			(0x3 << 7)
+#define  CNL_DRAM_WIDTH_SHIFT			7
+#define  CNL_DRAM_WIDTH_X8			(0x0 << 7)
+#define  CNL_DRAM_WIDTH_X16			(0x1 << 7)
+#define  CNL_DRAM_WIDTH_X32			(0x2 << 7)
+#define  CNL_DRAM_RANK_MASK			(0x3 << 9)
+#define  CNL_DRAM_RANK_SHIFT			9
+#define  CNL_DRAM_RANK_1			(0x0 << 9)
+#define  CNL_DRAM_RANK_2			(0x1 << 9)
+#define  CNL_DRAM_RANK_3			(0x2 << 9)
+#define  CNL_DRAM_RANK_4			(0x3 << 9)
 
 /* Please see hsw_read_dcomp() and hsw_write_dcomp() before using this register,
  * since on HSW we can't write to it using I915_WRITE. */
@@ -9877,10 +9998,14 @@ enum skl_power_gate {
 #define _PIPE_A_CSC_COEFF_BU	0x4901c
 #define _PIPE_A_CSC_COEFF_RV_GV	0x49020
 #define _PIPE_A_CSC_COEFF_BV	0x49024
+
 #define _PIPE_A_CSC_MODE	0x49028
-#define   CSC_BLACK_SCREEN_OFFSET	(1 << 2)
-#define   CSC_POSITION_BEFORE_GAMMA	(1 << 1)
-#define   CSC_MODE_YUV_TO_RGB		(1 << 0)
+#define  ICL_CSC_ENABLE			(1 << 31)
+#define  ICL_OUTPUT_CSC_ENABLE		(1 << 30)
+#define  CSC_BLACK_SCREEN_OFFSET	(1 << 2)
+#define  CSC_POSITION_BEFORE_GAMMA	(1 << 1)
+#define  CSC_MODE_YUV_TO_RGB		(1 << 0)
+
 #define _PIPE_A_CSC_PREOFF_HI	0x49030
 #define _PIPE_A_CSC_PREOFF_ME	0x49034
 #define _PIPE_A_CSC_PREOFF_LO	0x49038
@@ -9916,6 +10041,70 @@ enum skl_power_gate {
 #define PIPE_CSC_POSTOFF_ME(pipe)	_MMIO_PIPE(pipe, _PIPE_A_CSC_POSTOFF_ME, _PIPE_B_CSC_POSTOFF_ME)
 #define PIPE_CSC_POSTOFF_LO(pipe)	_MMIO_PIPE(pipe, _PIPE_A_CSC_POSTOFF_LO, _PIPE_B_CSC_POSTOFF_LO)
 
+/* Pipe Output CSC */
+#define _PIPE_A_OUTPUT_CSC_COEFF_RY_GY	0x49050
+#define _PIPE_A_OUTPUT_CSC_COEFF_BY	0x49054
+#define _PIPE_A_OUTPUT_CSC_COEFF_RU_GU	0x49058
+#define _PIPE_A_OUTPUT_CSC_COEFF_BU	0x4905c
+#define _PIPE_A_OUTPUT_CSC_COEFF_RV_GV	0x49060
+#define _PIPE_A_OUTPUT_CSC_COEFF_BV	0x49064
+#define _PIPE_A_OUTPUT_CSC_PREOFF_HI	0x49068
+#define _PIPE_A_OUTPUT_CSC_PREOFF_ME	0x4906c
+#define _PIPE_A_OUTPUT_CSC_PREOFF_LO	0x49070
+#define _PIPE_A_OUTPUT_CSC_POSTOFF_HI	0x49074
+#define _PIPE_A_OUTPUT_CSC_POSTOFF_ME	0x49078
+#define _PIPE_A_OUTPUT_CSC_POSTOFF_LO	0x4907c
+
+#define _PIPE_B_OUTPUT_CSC_COEFF_RY_GY	0x49150
+#define _PIPE_B_OUTPUT_CSC_COEFF_BY	0x49154
+#define _PIPE_B_OUTPUT_CSC_COEFF_RU_GU	0x49158
+#define _PIPE_B_OUTPUT_CSC_COEFF_BU	0x4915c
+#define _PIPE_B_OUTPUT_CSC_COEFF_RV_GV	0x49160
+#define _PIPE_B_OUTPUT_CSC_COEFF_BV	0x49164
+#define _PIPE_B_OUTPUT_CSC_PREOFF_HI	0x49168
+#define _PIPE_B_OUTPUT_CSC_PREOFF_ME	0x4916c
+#define _PIPE_B_OUTPUT_CSC_PREOFF_LO	0x49170
+#define _PIPE_B_OUTPUT_CSC_POSTOFF_HI	0x49174
+#define _PIPE_B_OUTPUT_CSC_POSTOFF_ME	0x49178
+#define _PIPE_B_OUTPUT_CSC_POSTOFF_LO	0x4917c
+
+#define PIPE_CSC_OUTPUT_COEFF_RY_GY(pipe)	_MMIO_PIPE(pipe,\
+							   _PIPE_A_OUTPUT_CSC_COEFF_RY_GY,\
+							   _PIPE_B_OUTPUT_CSC_COEFF_RY_GY)
+#define PIPE_CSC_OUTPUT_COEFF_BY(pipe)		_MMIO_PIPE(pipe, \
+							   _PIPE_A_OUTPUT_CSC_COEFF_BY, \
+							   _PIPE_B_OUTPUT_CSC_COEFF_BY)
+#define PIPE_CSC_OUTPUT_COEFF_RU_GU(pipe)	_MMIO_PIPE(pipe, \
+							   _PIPE_A_OUTPUT_CSC_COEFF_RU_GU, \
+							   _PIPE_B_OUTPUT_CSC_COEFF_RU_GU)
+#define PIPE_CSC_OUTPUT_COEFF_BU(pipe)		_MMIO_PIPE(pipe, \
+							   _PIPE_A_OUTPUT_CSC_COEFF_BU, \
+							   _PIPE_B_OUTPUT_CSC_COEFF_BU)
+#define PIPE_CSC_OUTPUT_COEFF_RV_GV(pipe)	_MMIO_PIPE(pipe, \
+							   _PIPE_A_OUTPUT_CSC_COEFF_RV_GV, \
+							   _PIPE_B_OUTPUT_CSC_COEFF_RV_GV)
+#define PIPE_CSC_OUTPUT_COEFF_BV(pipe)		_MMIO_PIPE(pipe, \
+							   _PIPE_A_OUTPUT_CSC_COEFF_BV, \
+							   _PIPE_B_OUTPUT_CSC_COEFF_BV)
+#define PIPE_CSC_OUTPUT_PREOFF_HI(pipe)		_MMIO_PIPE(pipe, \
+							   _PIPE_A_OUTPUT_CSC_PREOFF_HI, \
+							   _PIPE_B_OUTPUT_CSC_PREOFF_HI)
+#define PIPE_CSC_OUTPUT_PREOFF_ME(pipe)		_MMIO_PIPE(pipe, \
+							   _PIPE_A_OUTPUT_CSC_PREOFF_ME, \
+							   _PIPE_B_OUTPUT_CSC_PREOFF_ME)
+#define PIPE_CSC_OUTPUT_PREOFF_LO(pipe)		_MMIO_PIPE(pipe, \
+							   _PIPE_A_OUTPUT_CSC_PREOFF_LO, \
+							   _PIPE_B_OUTPUT_CSC_PREOFF_LO)
+#define PIPE_CSC_OUTPUT_POSTOFF_HI(pipe)	_MMIO_PIPE(pipe, \
+							   _PIPE_A_OUTPUT_CSC_POSTOFF_HI, \
+							   _PIPE_B_OUTPUT_CSC_POSTOFF_HI)
+#define PIPE_CSC_OUTPUT_POSTOFF_ME(pipe)	_MMIO_PIPE(pipe, \
+							   _PIPE_A_OUTPUT_CSC_POSTOFF_ME, \
+							   _PIPE_B_OUTPUT_CSC_POSTOFF_ME)
+#define PIPE_CSC_OUTPUT_POSTOFF_LO(pipe)	_MMIO_PIPE(pipe, \
+							   _PIPE_A_OUTPUT_CSC_POSTOFF_LO, \
+							   _PIPE_B_OUTPUT_CSC_POSTOFF_LO)
+
 /* pipe degamma/gamma LUTs on IVB+ */
 #define _PAL_PREC_INDEX_A	0x4A400
 #define _PAL_PREC_INDEX_B	0x4AC00
@@ -9924,6 +10113,7 @@ enum skl_power_gate {
 #define   PAL_PREC_SPLIT_MODE		(1 << 31)
 #define   PAL_PREC_AUTO_INCREMENT	(1 << 15)
 #define   PAL_PREC_INDEX_VALUE_MASK	(0x3ff << 0)
+#define   PAL_PREC_INDEX_VALUE(x)	((x) << 0)
 #define _PAL_PREC_DATA_A	0x4A404
 #define _PAL_PREC_DATA_B	0x4AC04
 #define _PAL_PREC_DATA_C	0x4B404
@@ -9941,6 +10131,7 @@ enum skl_power_gate {
 #define PREC_PAL_DATA(pipe)		_MMIO_PIPE(pipe, _PAL_PREC_DATA_A, _PAL_PREC_DATA_B)
 #define PREC_PAL_GC_MAX(pipe, i)	_MMIO(_PIPE(pipe, _PAL_PREC_GC_MAX_A, _PAL_PREC_GC_MAX_B) + (i) * 4)
 #define PREC_PAL_EXT_GC_MAX(pipe, i)	_MMIO(_PIPE(pipe, _PAL_PREC_EXT_GC_MAX_A, _PAL_PREC_EXT_GC_MAX_B) + (i) * 4)
+#define PREC_PAL_EXT2_GC_MAX(pipe, i)	_MMIO(_PIPE(pipe, _PAL_PREC_EXT2_GC_MAX_A, _PAL_PREC_EXT2_GC_MAX_B) + (i) * 4)
 
 #define _PRE_CSC_GAMC_INDEX_A	0x4A484
 #define _PRE_CSC_GAMC_INDEX_B	0x4AC84
diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c
index c2a5c48c7541..b836721d3b13 100644
--- a/drivers/gpu/drm/i915/i915_request.c
+++ b/drivers/gpu/drm/i915/i915_request.c
@@ -22,15 +22,31 @@
  *
  */
 
-#include <linux/prefetch.h>
 #include <linux/dma-fence-array.h>
+#include <linux/irq_work.h>
+#include <linux/prefetch.h>
 #include <linux/sched.h>
 #include <linux/sched/clock.h>
 #include <linux/sched/signal.h>
 
-#include "i915_drv.h"
 #include "i915_active.h"
+#include "i915_drv.h"
+#include "i915_globals.h"
 #include "i915_reset.h"
+#include "intel_pm.h"
+
+struct execute_cb {
+	struct list_head link;
+	struct irq_work work;
+	struct i915_sw_fence *fence;
+};
+
+static struct i915_global_request {
+	struct i915_global base;
+	struct kmem_cache *slab_requests;
+	struct kmem_cache *slab_dependencies;
+	struct kmem_cache *slab_execute_cbs;
+} global;
 
 static const char *i915_fence_get_driver_name(struct dma_fence *fence)
 {
@@ -51,7 +67,7 @@ static const char *i915_fence_get_timeline_name(struct dma_fence *fence)
 	if (test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, &fence->flags))
 		return "signaled";
 
-	return to_request(fence)->timeline->name;
+	return to_request(fence)->gem_context->name ?: "[i915]";
 }
 
 static bool i915_fence_signaled(struct dma_fence *fence)
@@ -68,7 +84,9 @@ static signed long i915_fence_wait(struct dma_fence *fence,
 				   bool interruptible,
 				   signed long timeout)
 {
-	return i915_request_wait(to_request(fence), interruptible, timeout);
+	return i915_request_wait(to_request(fence),
+				 interruptible | I915_WAIT_PRIORITY,
+				 timeout);
 }
 
 static void i915_fence_release(struct dma_fence *fence)
@@ -83,8 +101,9 @@ static void i915_fence_release(struct dma_fence *fence)
 	 * caught trying to reuse dead objects.
 	 */
 	i915_sw_fence_fini(&rq->submit);
+	i915_sw_fence_fini(&rq->semaphore);
 
-	kmem_cache_free(rq->i915->requests, rq);
+	kmem_cache_free(global.slab_requests, rq);
 }
 
 const struct dma_fence_ops i915_fence_ops = {
@@ -150,7 +169,6 @@ static void advance_ring(struct i915_request *request)
 		 * is just about to be. Either works, if we miss the last two
 		 * noops - they are safe to be replayed on a reset.
 		 */
-		GEM_TRACE("marking %s as inactive\n", ring->timeline->name);
 		tail = READ_ONCE(request->tail);
 		list_del(&ring->active_link);
 	} else {
@@ -177,12 +195,10 @@ static void free_capture_list(struct i915_request *request)
 static void __retire_engine_request(struct intel_engine_cs *engine,
 				    struct i915_request *rq)
 {
-	GEM_TRACE("%s(%s) fence %llx:%lld, global=%d, current %d:%d\n",
+	GEM_TRACE("%s(%s) fence %llx:%lld, current %d\n",
 		  __func__, engine->name,
 		  rq->fence.context, rq->fence.seqno,
-		  rq->global_seqno,
-		  hwsp_seqno(rq),
-		  intel_engine_get_seqno(engine));
+		  hwsp_seqno(rq));
 
 	GEM_BUG_ON(!i915_request_completed(rq));
 
@@ -241,12 +257,10 @@ static void i915_request_retire(struct i915_request *request)
 {
 	struct i915_active_request *active, *next;
 
-	GEM_TRACE("%s fence %llx:%lld, global=%d, current %d:%d\n",
+	GEM_TRACE("%s fence %llx:%lld, current %d\n",
 		  request->engine->name,
 		  request->fence.context, request->fence.seqno,
-		  request->global_seqno,
-		  hwsp_seqno(request),
-		  intel_engine_get_seqno(request->engine));
+		  hwsp_seqno(request));
 
 	lockdep_assert_held(&request->i915->drm.struct_mutex);
 	GEM_BUG_ON(!i915_sw_fence_signaled(&request->submit));
@@ -288,15 +302,13 @@ static void i915_request_retire(struct i915_request *request)
 
 	i915_request_remove_from_client(request);
 
-	/* Retirement decays the ban score as it is a sign of ctx progress */
-	atomic_dec_if_positive(&request->gem_context->ban_score);
 	intel_context_unpin(request->hw_context);
 
 	__retire_engine_upto(request->engine, request);
 
 	unreserve_gt(request->i915);
 
-	i915_sched_node_fini(request->i915, &request->sched);
+	i915_sched_node_fini(&request->sched);
 	i915_request_put(request);
 }
 
@@ -305,12 +317,10 @@ void i915_request_retire_upto(struct i915_request *rq)
 	struct intel_ring *ring = rq->ring;
 	struct i915_request *tmp;
 
-	GEM_TRACE("%s fence %llx:%lld, global=%d, current %d:%d\n",
+	GEM_TRACE("%s fence %llx:%lld, current %d\n",
 		  rq->engine->name,
 		  rq->fence.context, rq->fence.seqno,
-		  rq->global_seqno,
-		  hwsp_seqno(rq),
-		  intel_engine_get_seqno(rq->engine));
+		  hwsp_seqno(rq));
 
 	lockdep_assert_held(&rq->i915->drm.struct_mutex);
 	GEM_BUG_ON(!i915_request_completed(rq));
@@ -326,9 +336,67 @@ void i915_request_retire_upto(struct i915_request *rq)
 	} while (tmp != rq);
 }
 
-static u32 timeline_get_seqno(struct i915_timeline *tl)
+static void irq_execute_cb(struct irq_work *wrk)
+{
+	struct execute_cb *cb = container_of(wrk, typeof(*cb), work);
+
+	i915_sw_fence_complete(cb->fence);
+	kmem_cache_free(global.slab_execute_cbs, cb);
+}
+
+static void __notify_execute_cb(struct i915_request *rq)
+{
+	struct execute_cb *cb;
+
+	lockdep_assert_held(&rq->lock);
+
+	if (list_empty(&rq->execute_cb))
+		return;
+
+	list_for_each_entry(cb, &rq->execute_cb, link)
+		irq_work_queue(&cb->work);
+
+	/*
+	 * XXX Rollback on __i915_request_unsubmit()
+	 *
+	 * In the future, perhaps when we have an active time-slicing scheduler,
+	 * it will be interesting to unsubmit parallel execution and remove
+	 * busywaits from the GPU until their master is restarted. This is
+	 * quite hairy, we have to carefully rollback the fence and do a
+	 * preempt-to-idle cycle on the target engine, all the while the
+	 * master execute_cb may refire.
+	 */
+	INIT_LIST_HEAD(&rq->execute_cb);
+}
+
+static int
+i915_request_await_execution(struct i915_request *rq,
+			     struct i915_request *signal,
+			     gfp_t gfp)
 {
-	return tl->seqno += 1 + tl->has_initial_breadcrumb;
+	struct execute_cb *cb;
+
+	if (i915_request_is_active(signal))
+		return 0;
+
+	cb = kmem_cache_alloc(global.slab_execute_cbs, gfp);
+	if (!cb)
+		return -ENOMEM;
+
+	cb->fence = &rq->submit;
+	i915_sw_fence_await(cb->fence);
+	init_irq_work(&cb->work, irq_execute_cb);
+
+	spin_lock_irq(&signal->lock);
+	if (i915_request_is_active(signal)) {
+		i915_sw_fence_complete(cb->fence);
+		kmem_cache_free(global.slab_execute_cbs, cb);
+	} else {
+		list_add_tail(&cb->link, &signal->execute_cb);
+	}
+	spin_unlock_irq(&signal->lock);
+
+	return 0;
 }
 
 static void move_to_timeline(struct i915_request *request,
@@ -342,42 +410,33 @@ static void move_to_timeline(struct i915_request *request,
 	spin_unlock(&request->timeline->lock);
 }
 
-static u32 next_global_seqno(struct i915_timeline *tl)
-{
-	if (!++tl->seqno)
-		++tl->seqno;
-	return tl->seqno;
-}
-
 void __i915_request_submit(struct i915_request *request)
 {
 	struct intel_engine_cs *engine = request->engine;
-	u32 seqno;
 
-	GEM_TRACE("%s fence %llx:%lld -> global=%d, current %d:%d\n",
+	GEM_TRACE("%s fence %llx:%lld -> current %d\n",
 		  engine->name,
 		  request->fence.context, request->fence.seqno,
-		  engine->timeline.seqno + 1,
-		  hwsp_seqno(request),
-		  intel_engine_get_seqno(engine));
+		  hwsp_seqno(request));
 
 	GEM_BUG_ON(!irqs_disabled());
 	lockdep_assert_held(&engine->timeline.lock);
 
-	GEM_BUG_ON(request->global_seqno);
-
-	seqno = next_global_seqno(&engine->timeline);
-	GEM_BUG_ON(!seqno);
-	GEM_BUG_ON(intel_engine_signaled(engine, seqno));
+	if (i915_gem_context_is_banned(request->gem_context))
+		i915_request_skip(request, -EIO);
 
 	/* We may be recursing from the signal callback of another i915 fence */
 	spin_lock_nested(&request->lock, SINGLE_DEPTH_NESTING);
+
 	GEM_BUG_ON(test_bit(I915_FENCE_FLAG_ACTIVE, &request->fence.flags));
 	set_bit(I915_FENCE_FLAG_ACTIVE, &request->fence.flags);
-	request->global_seqno = seqno;
+
 	if (test_bit(DMA_FENCE_FLAG_ENABLE_SIGNAL_BIT, &request->fence.flags) &&
 	    !i915_request_enable_breadcrumb(request))
 		intel_engine_queue_breadcrumbs(engine);
+
+	__notify_execute_cb(request);
+
 	spin_unlock(&request->lock);
 
 	engine->emit_fini_breadcrumb(request,
@@ -406,12 +465,10 @@ void __i915_request_unsubmit(struct i915_request *request)
 {
 	struct intel_engine_cs *engine = request->engine;
 
-	GEM_TRACE("%s fence %llx:%lld <- global=%d, current %d:%d\n",
+	GEM_TRACE("%s fence %llx:%lld, current %d\n",
 		  engine->name,
 		  request->fence.context, request->fence.seqno,
-		  request->global_seqno,
-		  hwsp_seqno(request),
-		  intel_engine_get_seqno(engine));
+		  hwsp_seqno(request));
 
 	GEM_BUG_ON(!irqs_disabled());
 	lockdep_assert_held(&engine->timeline.lock);
@@ -420,18 +477,25 @@ void __i915_request_unsubmit(struct i915_request *request)
 	 * Only unwind in reverse order, required so that the per-context list
 	 * is kept in seqno/ring order.
 	 */
-	GEM_BUG_ON(!request->global_seqno);
-	GEM_BUG_ON(request->global_seqno != engine->timeline.seqno);
-	GEM_BUG_ON(intel_engine_has_completed(engine, request->global_seqno));
-	engine->timeline.seqno--;
 
 	/* We may be recursing from the signal callback of another i915 fence */
 	spin_lock_nested(&request->lock, SINGLE_DEPTH_NESTING);
-	request->global_seqno = 0;
+
+	/*
+	 * As we do not allow WAIT to preempt inflight requests,
+	 * once we have executed a request, along with triggering
+	 * any execution callbacks, we must preserve its ordering
+	 * within the non-preemptible FIFO.
+	 */
+	BUILD_BUG_ON(__NO_PREEMPTION & ~I915_PRIORITY_MASK); /* only internal */
+	request->sched.attr.priority |= __NO_PREEMPTION;
+
 	if (test_bit(DMA_FENCE_FLAG_ENABLE_SIGNAL_BIT, &request->fence.flags))
 		i915_request_cancel_breadcrumb(request);
+
 	GEM_BUG_ON(!test_bit(I915_FENCE_FLAG_ACTIVE, &request->fence.flags));
 	clear_bit(I915_FENCE_FLAG_ACTIVE, &request->fence.flags);
+
 	spin_unlock(&request->lock);
 
 	/* Transfer back from the global per-engine timeline to per-context */
@@ -489,6 +553,36 @@ submit_notify(struct i915_sw_fence *fence, enum i915_sw_fence_notify state)
 	return NOTIFY_DONE;
 }
 
+static int __i915_sw_fence_call
+semaphore_notify(struct i915_sw_fence *fence, enum i915_sw_fence_notify state)
+{
+	struct i915_request *request =
+		container_of(fence, typeof(*request), semaphore);
+
+	switch (state) {
+	case FENCE_COMPLETE:
+		/*
+		 * We only check a small portion of our dependencies
+		 * and so cannot guarantee that there remains no
+		 * semaphore chain across all. Instead of opting
+		 * for the full NOSEMAPHORE boost, we go for the
+		 * smaller (but still preempting) boost of
+		 * NEWCLIENT. This will be enough to boost over
+		 * a busywaiting request (as that cannot be
+		 * NEWCLIENT) without accidentally boosting
+		 * a busywait over real work elsewhere.
+		 */
+		i915_schedule_bump_priority(request, I915_PRIORITY_NEWCLIENT);
+		break;
+
+	case FENCE_FREE:
+		i915_request_put(request);
+		break;
+	}
+
+	return NOTIFY_DONE;
+}
+
 static void ring_retire_requests(struct intel_ring *ring)
 {
 	struct i915_request *rq, *rn;
@@ -518,12 +612,7 @@ i915_request_alloc_slow(struct intel_context *ce)
 	ring_retire_requests(ring);
 
 out:
-	return kmem_cache_alloc(ce->gem_context->i915->requests, GFP_KERNEL);
-}
-
-static int add_timeline_barrier(struct i915_request *rq)
-{
-	return i915_request_await_active_request(rq, &rq->timeline->barrier);
+	return kmem_cache_alloc(global.slab_requests, GFP_KERNEL);
 }
 
 /**
@@ -539,8 +628,10 @@ struct i915_request *
 i915_request_alloc(struct intel_engine_cs *engine, struct i915_gem_context *ctx)
 {
 	struct drm_i915_private *i915 = engine->i915;
-	struct i915_request *rq;
 	struct intel_context *ce;
+	struct i915_timeline *tl;
+	struct i915_request *rq;
+	u32 seqno;
 	int ret;
 
 	lockdep_assert_held(&i915->drm.struct_mutex);
@@ -556,8 +647,9 @@ i915_request_alloc(struct intel_engine_cs *engine, struct i915_gem_context *ctx)
 	 * ABI: Before userspace accesses the GPU (e.g. execbuffer), report
 	 * EIO if the GPU is already wedged.
 	 */
-	if (i915_terminally_wedged(&i915->gpu_error))
-		return ERR_PTR(-EIO);
+	ret = i915_terminally_wedged(i915);
+	if (ret)
+		return ERR_PTR(ret);
 
 	/*
 	 * Pinning the contexts may generate requests in order to acquire
@@ -569,6 +661,7 @@ i915_request_alloc(struct intel_engine_cs *engine, struct i915_gem_context *ctx)
 		return ERR_CAST(ce);
 
 	reserve_gt(i915);
+	mutex_lock(&ce->ring->timeline->mutex);
 
 	/* Move our oldest request to the slab-cache (if not in use!) */
 	rq = list_first_entry(&ce->ring->request_list, typeof(*rq), ring_link);
@@ -605,7 +698,7 @@ i915_request_alloc(struct intel_engine_cs *engine, struct i915_gem_context *ctx)
 	 *
 	 * Do not use kmem_cache_zalloc() here!
 	 */
-	rq = kmem_cache_alloc(i915->requests,
+	rq = kmem_cache_alloc(global.slab_requests,
 			      GFP_KERNEL | __GFP_RETRY_MAYFAIL | __GFP_NOWARN);
 	if (unlikely(!rq)) {
 		rq = i915_request_alloc_slow(ce);
@@ -615,32 +708,36 @@ i915_request_alloc(struct intel_engine_cs *engine, struct i915_gem_context *ctx)
 		}
 	}
 
-	rq->rcustate = get_state_synchronize_rcu();
-
 	INIT_LIST_HEAD(&rq->active_list);
+	INIT_LIST_HEAD(&rq->execute_cb);
+
+	tl = ce->ring->timeline;
+	ret = i915_timeline_get_seqno(tl, rq, &seqno);
+	if (ret)
+		goto err_free;
+
 	rq->i915 = i915;
 	rq->engine = engine;
 	rq->gem_context = ctx;
 	rq->hw_context = ce;
 	rq->ring = ce->ring;
-	rq->timeline = ce->ring->timeline;
+	rq->timeline = tl;
 	GEM_BUG_ON(rq->timeline == &engine->timeline);
-	rq->hwsp_seqno = rq->timeline->hwsp_seqno;
+	rq->hwsp_seqno = tl->hwsp_seqno;
+	rq->hwsp_cacheline = tl->hwsp_cacheline;
+	rq->rcustate = get_state_synchronize_rcu(); /* acts as smp_mb() */
 
 	spin_lock_init(&rq->lock);
-	dma_fence_init(&rq->fence,
-		       &i915_fence_ops,
-		       &rq->lock,
-		       rq->timeline->fence_context,
-		       timeline_get_seqno(rq->timeline));
+	dma_fence_init(&rq->fence, &i915_fence_ops, &rq->lock,
+		       tl->fence_context, seqno);
 
 	/* We bump the ref for the fence chain */
 	i915_sw_fence_init(&i915_request_get(rq)->submit, submit_notify);
+	i915_sw_fence_init(&i915_request_get(rq)->semaphore, semaphore_notify);
 
 	i915_sched_node_init(&rq->sched);
 
 	/* No zalloc, must clear what we need by hand */
-	rq->global_seqno = 0;
 	rq->file_priv = NULL;
 	rq->batch = NULL;
 	rq->capture_list = NULL;
@@ -668,10 +765,6 @@ i915_request_alloc(struct intel_engine_cs *engine, struct i915_gem_context *ctx)
 	 */
 	rq->head = rq->ring->emit;
 
-	ret = add_timeline_barrier(rq);
-	if (ret)
-		goto err_unwind;
-
 	ret = engine->request_alloc(rq);
 	if (ret)
 		goto err_unwind;
@@ -682,7 +775,10 @@ i915_request_alloc(struct intel_engine_cs *engine, struct i915_gem_context *ctx)
 	rq->infix = rq->ring->emit; /* end of header; start of user payload */
 
 	/* Check that we didn't interrupt ourselves with a new request */
+	lockdep_assert_held(&rq->timeline->mutex);
 	GEM_BUG_ON(rq->timeline->seqno != rq->fence.seqno);
+	rq->cookie = lockdep_pin_lock(&rq->timeline->mutex);
+
 	return rq;
 
 err_unwind:
@@ -693,14 +789,76 @@ err_unwind:
 	GEM_BUG_ON(!list_empty(&rq->sched.signalers_list));
 	GEM_BUG_ON(!list_empty(&rq->sched.waiters_list));
 
-	kmem_cache_free(i915->requests, rq);
+err_free:
+	kmem_cache_free(global.slab_requests, rq);
 err_unreserve:
+	mutex_unlock(&ce->ring->timeline->mutex);
 	unreserve_gt(i915);
 	intel_context_unpin(ce);
 	return ERR_PTR(ret);
 }
 
 static int
+emit_semaphore_wait(struct i915_request *to,
+		    struct i915_request *from,
+		    gfp_t gfp)
+{
+	u32 hwsp_offset;
+	u32 *cs;
+	int err;
+
+	GEM_BUG_ON(!from->timeline->has_initial_breadcrumb);
+	GEM_BUG_ON(INTEL_GEN(to->i915) < 8);
+
+	/* Just emit the first semaphore we see as request space is limited. */
+	if (to->sched.semaphores & from->engine->mask)
+		return i915_sw_fence_await_dma_fence(&to->submit,
+						     &from->fence, 0,
+						     I915_FENCE_GFP);
+
+	err = i915_sw_fence_await_dma_fence(&to->semaphore,
+					    &from->fence, 0,
+					    I915_FENCE_GFP);
+	if (err < 0)
+		return err;
+
+	/* We need to pin the signaler's HWSP until we are finished reading. */
+	err = i915_timeline_read_hwsp(from, to, &hwsp_offset);
+	if (err)
+		return err;
+
+	/* Only submit our spinner after the signaler is running! */
+	err = i915_request_await_execution(to, from, gfp);
+	if (err)
+		return err;
+
+	cs = intel_ring_begin(to, 4);
+	if (IS_ERR(cs))
+		return PTR_ERR(cs);
+
+	/*
+	 * Using greater-than-or-equal here means we have to worry
+	 * about seqno wraparound. To side step that issue, we swap
+	 * the timeline HWSP upon wrapping, so that everyone listening
+	 * for the old (pre-wrap) values do not see the much smaller
+	 * (post-wrap) values than they were expecting (and so wait
+	 * forever).
+	 */
+	*cs++ = MI_SEMAPHORE_WAIT |
+		MI_SEMAPHORE_GLOBAL_GTT |
+		MI_SEMAPHORE_POLL |
+		MI_SEMAPHORE_SAD_GTE_SDD;
+	*cs++ = from->fence.seqno;
+	*cs++ = hwsp_offset;
+	*cs++ = 0;
+
+	intel_ring_advance(to, cs);
+	to->sched.semaphores |= from->engine->mask;
+	to->sched.flags |= I915_SCHED_HAS_SEMAPHORE_CHAIN;
+	return 0;
+}
+
+static int
 i915_request_await_request(struct i915_request *to, struct i915_request *from)
 {
 	int ret;
@@ -712,9 +870,7 @@ i915_request_await_request(struct i915_request *to, struct i915_request *from)
 		return 0;
 
 	if (to->engine->schedule) {
-		ret = i915_sched_node_add_dependency(to->i915,
-						     &to->sched,
-						     &from->sched);
+		ret = i915_sched_node_add_dependency(&to->sched, &from->sched);
 		if (ret < 0)
 			return ret;
 	}
@@ -723,6 +879,9 @@ i915_request_await_request(struct i915_request *to, struct i915_request *from)
 		ret = i915_sw_fence_await_sw_fence_gfp(&to->submit,
 						       &from->submit,
 						       I915_FENCE_GFP);
+	} else if (intel_engine_has_semaphores(to->engine) &&
+		   to->gem_context->sched.priority >= I915_PRIORITY_NORMAL) {
+		ret = emit_semaphore_wait(to, from, I915_FENCE_GFP);
 	} else {
 		ret = i915_sw_fence_await_dma_fence(&to->submit,
 						    &from->fence, 0,
@@ -873,6 +1032,60 @@ void i915_request_skip(struct i915_request *rq, int error)
 	memset(vaddr + head, 0, rq->postfix - head);
 }
 
+static struct i915_request *
+__i915_request_add_to_timeline(struct i915_request *rq)
+{
+	struct i915_timeline *timeline = rq->timeline;
+	struct i915_request *prev;
+
+	/*
+	 * Dependency tracking and request ordering along the timeline
+	 * is special cased so that we can eliminate redundant ordering
+	 * operations while building the request (we know that the timeline
+	 * itself is ordered, and here we guarantee it).
+	 *
+	 * As we know we will need to emit tracking along the timeline,
+	 * we embed the hooks into our request struct -- at the cost of
+	 * having to have specialised no-allocation interfaces (which will
+	 * be beneficial elsewhere).
+	 *
+	 * A second benefit to open-coding i915_request_await_request is
+	 * that we can apply a slight variant of the rules specialised
+	 * for timelines that jump between engines (such as virtual engines).
+	 * If we consider the case of virtual engine, we must emit a dma-fence
+	 * to prevent scheduling of the second request until the first is
+	 * complete (to maximise our greedy late load balancing) and this
+	 * precludes optimising to use semaphores serialisation of a single
+	 * timeline across engines.
+	 */
+	prev = i915_active_request_raw(&timeline->last_request,
+				       &rq->i915->drm.struct_mutex);
+	if (prev && !i915_request_completed(prev)) {
+		if (is_power_of_2(prev->engine->mask | rq->engine->mask))
+			i915_sw_fence_await_sw_fence(&rq->submit,
+						     &prev->submit,
+						     &rq->submitq);
+		else
+			__i915_sw_fence_await_dma_fence(&rq->submit,
+							&prev->fence,
+							&rq->dmaq);
+		if (rq->engine->schedule)
+			__i915_sched_node_add_dependency(&rq->sched,
+							 &prev->sched,
+							 &rq->dep,
+							 0);
+	}
+
+	spin_lock_irq(&timeline->lock);
+	list_add_tail(&rq->link, &timeline->requests);
+	spin_unlock_irq(&timeline->lock);
+
+	GEM_BUG_ON(timeline->seqno != rq->fence.seqno);
+	__i915_active_request_set(&timeline->last_request, rq);
+
+	return prev;
+}
+
 /*
  * NB: This function is not allowed to fail. Doing so would mean the the
  * request is not being tracked for completion but the work itself is
@@ -889,7 +1102,9 @@ void i915_request_add(struct i915_request *request)
 	GEM_TRACE("%s fence %llx:%lld\n",
 		  engine->name, request->fence.context, request->fence.seqno);
 
-	lockdep_assert_held(&request->i915->drm.struct_mutex);
+	lockdep_assert_held(&request->timeline->mutex);
+	lockdep_unpin_lock(&request->timeline->mutex, request->cookie);
+
 	trace_i915_request_add(request);
 
 	/*
@@ -917,37 +1132,12 @@ void i915_request_add(struct i915_request *request)
 	GEM_BUG_ON(IS_ERR(cs));
 	request->postfix = intel_ring_offset(request, cs);
 
-	/*
-	 * Seal the request and mark it as pending execution. Note that
-	 * we may inspect this state, without holding any locks, during
-	 * hangcheck. Hence we apply the barrier to ensure that we do not
-	 * see a more recent value in the hws than we are tracking.
-	 */
-
-	prev = i915_active_request_raw(&timeline->last_request,
-				       &request->i915->drm.struct_mutex);
-	if (prev && !i915_request_completed(prev)) {
-		i915_sw_fence_await_sw_fence(&request->submit, &prev->submit,
-					     &request->submitq);
-		if (engine->schedule)
-			__i915_sched_node_add_dependency(&request->sched,
-							 &prev->sched,
-							 &request->dep,
-							 0);
-	}
-
-	spin_lock_irq(&timeline->lock);
-	list_add_tail(&request->link, &timeline->requests);
-	spin_unlock_irq(&timeline->lock);
-
-	GEM_BUG_ON(timeline->seqno != request->fence.seqno);
-	__i915_active_request_set(&timeline->last_request, request);
+	prev = __i915_request_add_to_timeline(request);
 
 	list_add_tail(&request->ring_link, &ring->request_list);
-	if (list_is_first(&request->ring_link, &ring->request_list)) {
-		GEM_TRACE("marking %s as active\n", ring->timeline->name);
+	if (list_is_first(&request->ring_link, &ring->request_list))
 		list_add(&ring->active_link, &request->i915->gt.active_rings);
-	}
+	request->i915->gt.active_engines |= request->engine->mask;
 	request->emitted_jiffies = jiffies;
 
 	/*
@@ -962,11 +1152,27 @@ void i915_request_add(struct i915_request *request)
 	 * run at the earliest possible convenience.
 	 */
 	local_bh_disable();
+	i915_sw_fence_commit(&request->semaphore);
 	rcu_read_lock(); /* RCU serialisation for set-wedged protection */
 	if (engine->schedule) {
 		struct i915_sched_attr attr = request->gem_context->sched;
 
 		/*
+		 * Boost actual workloads past semaphores!
+		 *
+		 * With semaphores we spin on one engine waiting for another,
+		 * simply to reduce the latency of starting our work when
+		 * the signaler completes. However, if there is any other
+		 * work that we could be doing on this engine instead, that
+		 * is better utilisation and will reduce the overall duration
+		 * of the current work. To avoid PI boosting a semaphore
+		 * far in the distance past over useful work, we keep a history
+		 * of any semaphore use along our dependency chain.
+		 */
+		if (!(request->sched.flags & I915_SCHED_HAS_SEMAPHORE_CHAIN))
+			attr.priority |= I915_PRIORITY_NOSEMAPHORE;
+
+		/*
 		 * Boost priorities to new clients (new request flows).
 		 *
 		 * Allow interactive/synchronous clients to jump ahead of
@@ -1000,6 +1206,8 @@ void i915_request_add(struct i915_request *request)
 	 */
 	if (prev && i915_request_completed(prev))
 		i915_request_retire_upto(prev);
+
+	mutex_unlock(&request->timeline->mutex);
 }
 
 static unsigned long local_clock_us(unsigned int *cpu)
@@ -1136,8 +1344,25 @@ long i915_request_wait(struct i915_request *rq,
 	if (__i915_spin_request(rq, state, 5))
 		goto out;
 
-	if (flags & I915_WAIT_PRIORITY)
+	/*
+	 * This client is about to stall waiting for the GPU. In many cases
+	 * this is undesirable and limits the throughput of the system, as
+	 * many clients cannot continue processing user input/output whilst
+	 * blocked. RPS autotuning may take tens of milliseconds to respond
+	 * to the GPU load and thus incurs additional latency for the client.
+	 * We can circumvent that by promoting the GPU frequency to maximum
+	 * before we sleep. This makes the GPU throttle up much more quickly
+	 * (good for benchmarks and user experience, e.g. window animations),
+	 * but at a cost of spending more power processing the workload
+	 * (bad for battery).
+	 */
+	if (flags & I915_WAIT_PRIORITY) {
+		if (!i915_request_started(rq) && INTEL_GEN(rq->i915) >= 6)
+			gen6_rps_boost(rq);
+		local_bh_disable(); /* suspend tasklets for reprioritisation */
 		i915_schedule_bump_priority(rq, I915_PRIORITY_WAIT);
+		local_bh_enable(); /* kick tasklets en masse */
+	}
 
 	wait.tsk = current;
 	if (dma_fence_add_callback(&rq->fence, &wait.cb, request_wait_wake))
@@ -1179,11 +1404,66 @@ void i915_retire_requests(struct drm_i915_private *i915)
 	if (!i915->gt.active_requests)
 		return;
 
-	list_for_each_entry_safe(ring, tmp, &i915->gt.active_rings, active_link)
+	list_for_each_entry_safe(ring, tmp,
+				 &i915->gt.active_rings, active_link) {
+		intel_ring_get(ring); /* last rq holds reference! */
 		ring_retire_requests(ring);
+		intel_ring_put(ring);
+	}
 }
 
 #if IS_ENABLED(CONFIG_DRM_I915_SELFTEST)
 #include "selftests/mock_request.c"
 #include "selftests/i915_request.c"
 #endif
+
+static void i915_global_request_shrink(void)
+{
+	kmem_cache_shrink(global.slab_dependencies);
+	kmem_cache_shrink(global.slab_execute_cbs);
+	kmem_cache_shrink(global.slab_requests);
+}
+
+static void i915_global_request_exit(void)
+{
+	kmem_cache_destroy(global.slab_dependencies);
+	kmem_cache_destroy(global.slab_execute_cbs);
+	kmem_cache_destroy(global.slab_requests);
+}
+
+static struct i915_global_request global = { {
+	.shrink = i915_global_request_shrink,
+	.exit = i915_global_request_exit,
+} };
+
+int __init i915_global_request_init(void)
+{
+	global.slab_requests = KMEM_CACHE(i915_request,
+					  SLAB_HWCACHE_ALIGN |
+					  SLAB_RECLAIM_ACCOUNT |
+					  SLAB_TYPESAFE_BY_RCU);
+	if (!global.slab_requests)
+		return -ENOMEM;
+
+	global.slab_execute_cbs = KMEM_CACHE(execute_cb,
+					     SLAB_HWCACHE_ALIGN |
+					     SLAB_RECLAIM_ACCOUNT |
+					     SLAB_TYPESAFE_BY_RCU);
+	if (!global.slab_execute_cbs)
+		goto err_requests;
+
+	global.slab_dependencies = KMEM_CACHE(i915_dependency,
+					      SLAB_HWCACHE_ALIGN |
+					      SLAB_RECLAIM_ACCOUNT);
+	if (!global.slab_dependencies)
+		goto err_execute_cbs;
+
+	i915_global_register(&global.base);
+	return 0;
+
+err_execute_cbs:
+	kmem_cache_destroy(global.slab_execute_cbs);
+err_requests:
+	kmem_cache_destroy(global.slab_requests);
+	return -ENOMEM;
+}
diff --git a/drivers/gpu/drm/i915/i915_request.h b/drivers/gpu/drm/i915/i915_request.h
index 40f3e8dcbdd5..a982664618c2 100644
--- a/drivers/gpu/drm/i915/i915_request.h
+++ b/drivers/gpu/drm/i915/i915_request.h
@@ -26,9 +26,11 @@
 #define I915_REQUEST_H
 
 #include <linux/dma-fence.h>
+#include <linux/lockdep.h>
 
 #include "i915_gem.h"
 #include "i915_scheduler.h"
+#include "i915_selftest.h"
 #include "i915_sw_fence.h"
 
 #include <uapi/drm/i915_drm.h>
@@ -37,6 +39,7 @@ struct drm_file;
 struct drm_i915_gem_object;
 struct i915_request;
 struct i915_timeline;
+struct i915_timeline_cacheline;
 
 struct i915_capture_list {
 	struct i915_capture_list *next;
@@ -119,6 +122,15 @@ struct i915_request {
 	unsigned long rcustate;
 
 	/*
+	 * We pin the timeline->mutex while constructing the request to
+	 * ensure that no caller accidentally drops it during construction.
+	 * The timeline->mutex must be held to ensure that only this caller
+	 * can use the ring and manipulate the associated timeline during
+	 * construction.
+	 */
+	struct pin_cookie cookie;
+
+	/*
 	 * Fences for the various phases in the request's lifetime.
 	 *
 	 * The submit fence is used to await upon all of the request's
@@ -126,7 +138,12 @@ struct i915_request {
 	 * It is used by the driver to then queue the request for execution.
 	 */
 	struct i915_sw_fence submit;
-	wait_queue_entry_t submitq;
+	union {
+		wait_queue_entry_t submitq;
+		struct i915_sw_dma_fence_cb dmaq;
+	};
+	struct list_head execute_cb;
+	struct i915_sw_fence semaphore;
 
 	/*
 	 * A list of everyone we wait upon, and everyone who waits upon us.
@@ -147,13 +164,15 @@ struct i915_request {
 	 */
 	const u32 *hwsp_seqno;
 
-	/**
-	 * GEM sequence number associated with this request on the
-	 * global execution timeline. It is zero when the request is not
-	 * on the HW queue (i.e. not on the engine timeline list).
-	 * Its value is guarded by the timeline spinlock.
+	/*
+	 * If we need to access the timeline's seqno for this request in
+	 * another request, we need to keep a read reference to this associated
+	 * cacheline, so that we do not free and recycle it before the foreign
+	 * observers have completed. Hence, we keep a pointer to the cacheline
+	 * inside the timeline's HWSP vma, but it is only valid while this
+	 * request has not completed and guarded by the timeline mutex.
 	 */
-	u32 global_seqno;
+	struct i915_timeline_cacheline *hwsp_cacheline;
 
 	/** Position in the ring of the start of the request */
 	u32 head;
@@ -204,6 +223,11 @@ struct i915_request {
 	struct drm_i915_file_private *file_priv;
 	/** file_priv list entry for this request */
 	struct list_head client_link;
+
+	I915_SELFTEST_DECLARE(struct {
+		struct list_head link;
+		unsigned long delay;
+	} mock;)
 };
 
 #define I915_FENCE_GFP (GFP_KERNEL | __GFP_RETRY_MAYFAIL | __GFP_NOWARN)
@@ -247,30 +271,6 @@ i915_request_put(struct i915_request *rq)
 	dma_fence_put(&rq->fence);
 }
 
-/**
- * i915_request_global_seqno - report the current global seqno
- * @request - the request
- *
- * A request is assigned a global seqno only when it is on the hardware
- * execution queue. The global seqno can be used to maintain a list of
- * requests on the same engine in retirement order, for example for
- * constructing a priority queue for waiting. Prior to its execution, or
- * if it is subsequently removed in the event of preemption, its global
- * seqno is zero. As both insertion and removal from the execution queue
- * may operate in IRQ context, it is not guarded by the usual struct_mutex
- * BKL. Instead those relying on the global seqno must be prepared for its
- * value to change between reads. Only when the request is complete can
- * the global seqno be stable (due to the memory barriers on submitting
- * the commands to the hardware to write the breadcrumb, if the HWS shows
- * that it has passed the global seqno and the global seqno is unchanged
- * after the read, it is indeed complete).
- */
-static inline u32
-i915_request_global_seqno(const struct i915_request *request)
-{
-	return READ_ONCE(request->global_seqno);
-}
-
 int i915_request_await_object(struct i915_request *to,
 			      struct drm_i915_gem_object *obj,
 			      bool write);
@@ -358,10 +358,27 @@ static inline bool __i915_request_has_started(const struct i915_request *rq)
  * i915_request_started - check if the request has begun being executed
  * @rq: the request
  *
- * Returns true if the request has been submitted to hardware, and the hardware
- * has advanced passed the end of the previous request and so should be either
- * currently processing the request (though it may be preempted and so
- * not necessarily the next request to complete) or have completed the request.
+ * If the timeline is not using initial breadcrumbs, a request is
+ * considered started if the previous request on its timeline (i.e.
+ * context) has been signaled.
+ *
+ * If the timeline is using semaphores, it will also be emitting an
+ * "initial breadcrumb" after the semaphores are complete and just before
+ * it began executing the user payload. A request can therefore be active
+ * on the HW and not yet started as it is still busywaiting on its
+ * dependencies (via HW semaphores).
+ *
+ * If the request has started, its dependencies will have been signaled
+ * (either by fences or by semaphores) and it will have begun processing
+ * the user payload.
+ *
+ * However, even if a request has started, it may have been preempted and
+ * so no longer active, or it may have already completed.
+ *
+ * See also i915_request_is_active().
+ *
+ * Returns true if the request has begun executing the user payload, or
+ * has completed:
  */
 static inline bool i915_request_started(const struct i915_request *rq)
 {
diff --git a/drivers/gpu/drm/i915/i915_reset.c b/drivers/gpu/drm/i915/i915_reset.c
index 0e0ddf2e6815..677d59304e78 100644
--- a/drivers/gpu/drm/i915/i915_reset.c
+++ b/drivers/gpu/drm/i915/i915_reset.c
@@ -18,28 +18,39 @@
 /* XXX How to handle concurrent GGTT updates using tiling registers? */
 #define RESET_UNDER_STOP_MACHINE 0
 
+static void rmw_set(struct intel_uncore *uncore, i915_reg_t reg, u32 set)
+{
+	intel_uncore_rmw(uncore, reg, 0, set);
+}
+
+static void rmw_clear(struct intel_uncore *uncore, i915_reg_t reg, u32 clr)
+{
+	intel_uncore_rmw(uncore, reg, clr, 0);
+}
+
+static void rmw_set_fw(struct intel_uncore *uncore, i915_reg_t reg, u32 set)
+{
+	intel_uncore_rmw_fw(uncore, reg, 0, set);
+}
+
+static void rmw_clear_fw(struct intel_uncore *uncore, i915_reg_t reg, u32 clr)
+{
+	intel_uncore_rmw_fw(uncore, reg, clr, 0);
+}
+
 static void engine_skip_context(struct i915_request *rq)
 {
 	struct intel_engine_cs *engine = rq->engine;
 	struct i915_gem_context *hung_ctx = rq->gem_context;
-	struct i915_timeline *timeline = rq->timeline;
 
 	lockdep_assert_held(&engine->timeline.lock);
-	GEM_BUG_ON(timeline == &engine->timeline);
-
-	spin_lock(&timeline->lock);
-
-	if (i915_request_is_active(rq)) {
-		list_for_each_entry_continue(rq,
-					     &engine->timeline.requests, link)
-			if (rq->gem_context == hung_ctx)
-				i915_request_skip(rq, -EIO);
-	}
 
-	list_for_each_entry(rq, &timeline->requests, link)
-		i915_request_skip(rq, -EIO);
+	if (!i915_request_is_active(rq))
+		return;
 
-	spin_unlock(&timeline->lock);
+	list_for_each_entry_continue(rq, &engine->timeline.requests, link)
+		if (rq->gem_context == hung_ctx)
+			i915_request_skip(rq, -EIO);
 }
 
 static void client_mark_guilty(struct drm_i915_file_private *file_priv,
@@ -68,23 +79,29 @@ static void client_mark_guilty(struct drm_i915_file_private *file_priv,
 
 static bool context_mark_guilty(struct i915_gem_context *ctx)
 {
-	unsigned int score;
-	bool banned, bannable;
+	unsigned long prev_hang;
+	bool banned;
+	int i;
 
 	atomic_inc(&ctx->guilty_count);
 
-	bannable = i915_gem_context_is_bannable(ctx);
-	score = atomic_add_return(CONTEXT_SCORE_GUILTY, &ctx->ban_score);
-	banned = score >= CONTEXT_SCORE_BAN_THRESHOLD;
-
-	/* Cool contexts don't accumulate client ban score */
-	if (!bannable)
+	/* Cool contexts are too cool to be banned! (Used for reset testing.) */
+	if (!i915_gem_context_is_bannable(ctx))
 		return false;
 
+	/* Record the timestamp for the last N hangs */
+	prev_hang = ctx->hang_timestamp[0];
+	for (i = 0; i < ARRAY_SIZE(ctx->hang_timestamp) - 1; i++)
+		ctx->hang_timestamp[i] = ctx->hang_timestamp[i + 1];
+	ctx->hang_timestamp[i] = jiffies;
+
+	/* If we have hung N+1 times in rapid succession, we ban the context! */
+	banned = !i915_gem_context_is_recoverable(ctx);
+	if (time_before(jiffies, prev_hang + CONTEXT_FAST_HANG_JIFFIES))
+		banned = true;
 	if (banned) {
-		DRM_DEBUG_DRIVER("context %s: guilty %d, score %u, banned\n",
-				 ctx->name, atomic_read(&ctx->guilty_count),
-				 score);
+		DRM_DEBUG_DRIVER("context %s: guilty %d, banned\n",
+				 ctx->name, atomic_read(&ctx->guilty_count));
 		i915_gem_context_set_banned(ctx);
 	}
 
@@ -101,6 +118,12 @@ static void context_mark_innocent(struct i915_gem_context *ctx)
 
 void i915_reset_request(struct i915_request *rq, bool guilty)
 {
+	GEM_TRACE("%s rq=%llx:%lld, guilty? %s\n",
+		  rq->engine->name,
+		  rq->fence.context,
+		  rq->fence.seqno,
+		  yesno(guilty));
+
 	lockdep_assert_held(&rq->engine->timeline.lock);
 	GEM_BUG_ON(i915_request_completed(rq));
 
@@ -116,38 +139,43 @@ void i915_reset_request(struct i915_request *rq, bool guilty)
 
 static void gen3_stop_engine(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = engine->i915;
+	struct intel_uncore *uncore = engine->uncore;
 	const u32 base = engine->mmio_base;
 
+	GEM_TRACE("%s\n", engine->name);
+
 	if (intel_engine_stop_cs(engine))
-		DRM_DEBUG_DRIVER("%s: timed out on STOP_RING\n", engine->name);
+		GEM_TRACE("%s: timed out on STOP_RING\n", engine->name);
 
-	I915_WRITE_FW(RING_HEAD(base), I915_READ_FW(RING_TAIL(base)));
-	POSTING_READ_FW(RING_HEAD(base)); /* paranoia */
+	intel_uncore_write_fw(uncore,
+			      RING_HEAD(base),
+			      intel_uncore_read_fw(uncore, RING_TAIL(base)));
+	intel_uncore_posting_read_fw(uncore, RING_HEAD(base)); /* paranoia */
 
-	I915_WRITE_FW(RING_HEAD(base), 0);
-	I915_WRITE_FW(RING_TAIL(base), 0);
-	POSTING_READ_FW(RING_TAIL(base));
+	intel_uncore_write_fw(uncore, RING_HEAD(base), 0);
+	intel_uncore_write_fw(uncore, RING_TAIL(base), 0);
+	intel_uncore_posting_read_fw(uncore, RING_TAIL(base));
 
 	/* The ring must be empty before it is disabled */
-	I915_WRITE_FW(RING_CTL(base), 0);
+	intel_uncore_write_fw(uncore, RING_CTL(base), 0);
 
 	/* Check acts as a post */
-	if (I915_READ_FW(RING_HEAD(base)) != 0)
-		DRM_DEBUG_DRIVER("%s: ring head not parked\n",
-				 engine->name);
+	if (intel_uncore_read_fw(uncore, RING_HEAD(base)))
+		GEM_TRACE("%s: ring head [%x] not parked\n",
+			  engine->name,
+			  intel_uncore_read_fw(uncore, RING_HEAD(base)));
 }
 
 static void i915_stop_engines(struct drm_i915_private *i915,
-			      unsigned int engine_mask)
+			      intel_engine_mask_t engine_mask)
 {
 	struct intel_engine_cs *engine;
-	enum intel_engine_id id;
+	intel_engine_mask_t tmp;
 
 	if (INTEL_GEN(i915) < 3)
 		return;
 
-	for_each_engine_masked(engine, i915, engine_mask, id)
+	for_each_engine_masked(engine, i915, engine_mask, tmp)
 		gen3_stop_engine(engine);
 }
 
@@ -160,7 +188,7 @@ static bool i915_in_reset(struct pci_dev *pdev)
 }
 
 static int i915_do_reset(struct drm_i915_private *i915,
-			 unsigned int engine_mask,
+			 intel_engine_mask_t engine_mask,
 			 unsigned int retry)
 {
 	struct pci_dev *pdev = i915->drm.pdev;
@@ -189,7 +217,7 @@ static bool g4x_reset_complete(struct pci_dev *pdev)
 }
 
 static int g33_do_reset(struct drm_i915_private *i915,
-			unsigned int engine_mask,
+			intel_engine_mask_t engine_mask,
 			unsigned int retry)
 {
 	struct pci_dev *pdev = i915->drm.pdev;
@@ -198,17 +226,17 @@ static int g33_do_reset(struct drm_i915_private *i915,
 	return wait_for_atomic(g4x_reset_complete(pdev), 50);
 }
 
-static int g4x_do_reset(struct drm_i915_private *dev_priv,
-			unsigned int engine_mask,
+static int g4x_do_reset(struct drm_i915_private *i915,
+			intel_engine_mask_t engine_mask,
 			unsigned int retry)
 {
-	struct pci_dev *pdev = dev_priv->drm.pdev;
+	struct pci_dev *pdev = i915->drm.pdev;
+	struct intel_uncore *uncore = &i915->uncore;
 	int ret;
 
 	/* WaVcpClkGateDisableForMediaReset:ctg,elk */
-	I915_WRITE_FW(VDECCLK_GATE_D,
-		      I915_READ(VDECCLK_GATE_D) | VCP_UNIT_CLOCK_GATE_DISABLE);
-	POSTING_READ_FW(VDECCLK_GATE_D);
+	rmw_set_fw(uncore, VDECCLK_GATE_D, VCP_UNIT_CLOCK_GATE_DISABLE);
+	intel_uncore_posting_read_fw(uncore, VDECCLK_GATE_D);
 
 	pci_write_config_byte(pdev, I915_GDRST,
 			      GRDOM_MEDIA | GRDOM_RESET_ENABLE);
@@ -229,21 +257,22 @@ static int g4x_do_reset(struct drm_i915_private *dev_priv,
 out:
 	pci_write_config_byte(pdev, I915_GDRST, 0);
 
-	I915_WRITE_FW(VDECCLK_GATE_D,
-		      I915_READ(VDECCLK_GATE_D) & ~VCP_UNIT_CLOCK_GATE_DISABLE);
-	POSTING_READ_FW(VDECCLK_GATE_D);
+	rmw_clear_fw(uncore, VDECCLK_GATE_D, VCP_UNIT_CLOCK_GATE_DISABLE);
+	intel_uncore_posting_read_fw(uncore, VDECCLK_GATE_D);
 
 	return ret;
 }
 
-static int ironlake_do_reset(struct drm_i915_private *dev_priv,
-			     unsigned int engine_mask,
+static int ironlake_do_reset(struct drm_i915_private *i915,
+			     intel_engine_mask_t engine_mask,
 			     unsigned int retry)
 {
+	struct intel_uncore *uncore = &i915->uncore;
 	int ret;
 
-	I915_WRITE_FW(ILK_GDSR, ILK_GRDOM_RENDER | ILK_GRDOM_RESET_ENABLE);
-	ret = __intel_wait_for_register_fw(dev_priv, ILK_GDSR,
+	intel_uncore_write_fw(uncore, ILK_GDSR,
+			      ILK_GRDOM_RENDER | ILK_GRDOM_RESET_ENABLE);
+	ret = __intel_wait_for_register_fw(uncore, ILK_GDSR,
 					   ILK_GRDOM_RESET_ENABLE, 0,
 					   5000, 0,
 					   NULL);
@@ -252,8 +281,9 @@ static int ironlake_do_reset(struct drm_i915_private *dev_priv,
 		goto out;
 	}
 
-	I915_WRITE_FW(ILK_GDSR, ILK_GRDOM_MEDIA | ILK_GRDOM_RESET_ENABLE);
-	ret = __intel_wait_for_register_fw(dev_priv, ILK_GDSR,
+	intel_uncore_write_fw(uncore, ILK_GDSR,
+			      ILK_GRDOM_MEDIA | ILK_GRDOM_RESET_ENABLE);
+	ret = __intel_wait_for_register_fw(uncore, ILK_GDSR,
 					   ILK_GRDOM_RESET_ENABLE, 0,
 					   5000, 0,
 					   NULL);
@@ -263,15 +293,16 @@ static int ironlake_do_reset(struct drm_i915_private *dev_priv,
 	}
 
 out:
-	I915_WRITE_FW(ILK_GDSR, 0);
-	POSTING_READ_FW(ILK_GDSR);
+	intel_uncore_write_fw(uncore, ILK_GDSR, 0);
+	intel_uncore_posting_read_fw(uncore, ILK_GDSR);
 	return ret;
 }
 
 /* Reset the hardware domains (GENX_GRDOM_*) specified by mask */
-static int gen6_hw_domain_reset(struct drm_i915_private *dev_priv,
+static int gen6_hw_domain_reset(struct drm_i915_private *i915,
 				u32 hw_domain_mask)
 {
+	struct intel_uncore *uncore = &i915->uncore;
 	int err;
 
 	/*
@@ -279,10 +310,10 @@ static int gen6_hw_domain_reset(struct drm_i915_private *dev_priv,
 	 * for fifo space for the write or forcewake the chip for
 	 * the read
 	 */
-	I915_WRITE_FW(GEN6_GDRST, hw_domain_mask);
+	intel_uncore_write_fw(uncore, GEN6_GDRST, hw_domain_mask);
 
 	/* Wait for the device to ack the reset requests */
-	err = __intel_wait_for_register_fw(dev_priv,
+	err = __intel_wait_for_register_fw(uncore,
 					   GEN6_GDRST, hw_domain_mask, 0,
 					   500, 0,
 					   NULL);
@@ -294,36 +325,38 @@ static int gen6_hw_domain_reset(struct drm_i915_private *dev_priv,
 }
 
 static int gen6_reset_engines(struct drm_i915_private *i915,
-			      unsigned int engine_mask,
+			      intel_engine_mask_t engine_mask,
 			      unsigned int retry)
 {
 	struct intel_engine_cs *engine;
-	const u32 hw_engine_mask[I915_NUM_ENGINES] = {
-		[RCS] = GEN6_GRDOM_RENDER,
-		[BCS] = GEN6_GRDOM_BLT,
-		[VCS] = GEN6_GRDOM_MEDIA,
-		[VCS2] = GEN8_GRDOM_MEDIA2,
-		[VECS] = GEN6_GRDOM_VECS,
+	const u32 hw_engine_mask[] = {
+		[RCS0]  = GEN6_GRDOM_RENDER,
+		[BCS0]  = GEN6_GRDOM_BLT,
+		[VCS0]  = GEN6_GRDOM_MEDIA,
+		[VCS1]  = GEN8_GRDOM_MEDIA2,
+		[VECS0] = GEN6_GRDOM_VECS,
 	};
 	u32 hw_mask;
 
 	if (engine_mask == ALL_ENGINES) {
 		hw_mask = GEN6_GRDOM_FULL;
 	} else {
-		unsigned int tmp;
+		intel_engine_mask_t tmp;
 
 		hw_mask = 0;
-		for_each_engine_masked(engine, i915, engine_mask, tmp)
+		for_each_engine_masked(engine, i915, engine_mask, tmp) {
+			GEM_BUG_ON(engine->id >= ARRAY_SIZE(hw_engine_mask));
 			hw_mask |= hw_engine_mask[engine->id];
+		}
 	}
 
 	return gen6_hw_domain_reset(i915, hw_mask);
 }
 
-static u32 gen11_lock_sfc(struct drm_i915_private *dev_priv,
-			  struct intel_engine_cs *engine)
+static u32 gen11_lock_sfc(struct intel_engine_cs *engine)
 {
-	u8 vdbox_sfc_access = RUNTIME_INFO(dev_priv)->vdbox_sfc_access;
+	struct intel_uncore *uncore = engine->uncore;
+	u8 vdbox_sfc_access = RUNTIME_INFO(engine->i915)->vdbox_sfc_access;
 	i915_reg_t sfc_forced_lock, sfc_forced_lock_ack;
 	u32 sfc_forced_lock_bit, sfc_forced_lock_ack_bit;
 	i915_reg_t sfc_usage;
@@ -370,10 +403,9 @@ static u32 gen11_lock_sfc(struct drm_i915_private *dev_priv,
 	 * ends up being locked to the engine we want to reset, we have to reset
 	 * it as well (we will unlock it once the reset sequence is completed).
 	 */
-	I915_WRITE_FW(sfc_forced_lock,
-		      I915_READ_FW(sfc_forced_lock) | sfc_forced_lock_bit);
+	rmw_set_fw(uncore, sfc_forced_lock, sfc_forced_lock_bit);
 
-	if (__intel_wait_for_register_fw(dev_priv,
+	if (__intel_wait_for_register_fw(uncore,
 					 sfc_forced_lock_ack,
 					 sfc_forced_lock_ack_bit,
 					 sfc_forced_lock_ack_bit,
@@ -382,16 +414,16 @@ static u32 gen11_lock_sfc(struct drm_i915_private *dev_priv,
 		return 0;
 	}
 
-	if (I915_READ_FW(sfc_usage) & sfc_usage_bit)
+	if (intel_uncore_read_fw(uncore, sfc_usage) & sfc_usage_bit)
 		return sfc_reset_bit;
 
 	return 0;
 }
 
-static void gen11_unlock_sfc(struct drm_i915_private *dev_priv,
-			     struct intel_engine_cs *engine)
+static void gen11_unlock_sfc(struct intel_engine_cs *engine)
 {
-	u8 vdbox_sfc_access = RUNTIME_INFO(dev_priv)->vdbox_sfc_access;
+	struct intel_uncore *uncore = engine->uncore;
+	u8 vdbox_sfc_access = RUNTIME_INFO(engine->i915)->vdbox_sfc_access;
 	i915_reg_t sfc_forced_lock;
 	u32 sfc_forced_lock_bit;
 
@@ -413,38 +445,36 @@ static void gen11_unlock_sfc(struct drm_i915_private *dev_priv,
 		return;
 	}
 
-	I915_WRITE_FW(sfc_forced_lock,
-		      I915_READ_FW(sfc_forced_lock) & ~sfc_forced_lock_bit);
+	rmw_clear_fw(uncore, sfc_forced_lock, sfc_forced_lock_bit);
 }
 
 static int gen11_reset_engines(struct drm_i915_private *i915,
-			       unsigned int engine_mask,
+			       intel_engine_mask_t engine_mask,
 			       unsigned int retry)
 {
-	const u32 hw_engine_mask[I915_NUM_ENGINES] = {
-		[RCS] = GEN11_GRDOM_RENDER,
-		[BCS] = GEN11_GRDOM_BLT,
-		[VCS] = GEN11_GRDOM_MEDIA,
-		[VCS2] = GEN11_GRDOM_MEDIA2,
-		[VCS3] = GEN11_GRDOM_MEDIA3,
-		[VCS4] = GEN11_GRDOM_MEDIA4,
-		[VECS] = GEN11_GRDOM_VECS,
-		[VECS2] = GEN11_GRDOM_VECS2,
+	const u32 hw_engine_mask[] = {
+		[RCS0]  = GEN11_GRDOM_RENDER,
+		[BCS0]  = GEN11_GRDOM_BLT,
+		[VCS0]  = GEN11_GRDOM_MEDIA,
+		[VCS1]  = GEN11_GRDOM_MEDIA2,
+		[VCS2]  = GEN11_GRDOM_MEDIA3,
+		[VCS3]  = GEN11_GRDOM_MEDIA4,
+		[VECS0] = GEN11_GRDOM_VECS,
+		[VECS1] = GEN11_GRDOM_VECS2,
 	};
 	struct intel_engine_cs *engine;
-	unsigned int tmp;
+	intel_engine_mask_t tmp;
 	u32 hw_mask;
 	int ret;
 
-	BUILD_BUG_ON(VECS2 + 1 != I915_NUM_ENGINES);
-
 	if (engine_mask == ALL_ENGINES) {
 		hw_mask = GEN11_GRDOM_FULL;
 	} else {
 		hw_mask = 0;
 		for_each_engine_masked(engine, i915, engine_mask, tmp) {
+			GEM_BUG_ON(engine->id >= ARRAY_SIZE(hw_engine_mask));
 			hw_mask |= hw_engine_mask[engine->id];
-			hw_mask |= gen11_lock_sfc(i915, engine);
+			hw_mask |= gen11_lock_sfc(engine);
 		}
 	}
 
@@ -452,46 +482,62 @@ static int gen11_reset_engines(struct drm_i915_private *i915,
 
 	if (engine_mask != ALL_ENGINES)
 		for_each_engine_masked(engine, i915, engine_mask, tmp)
-			gen11_unlock_sfc(i915, engine);
+			gen11_unlock_sfc(engine);
 
 	return ret;
 }
 
 static int gen8_engine_reset_prepare(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = engine->i915;
+	struct intel_uncore *uncore = engine->uncore;
+	const i915_reg_t reg = RING_RESET_CTL(engine->mmio_base);
+	u32 request, mask, ack;
 	int ret;
 
-	I915_WRITE_FW(RING_RESET_CTL(engine->mmio_base),
-		      _MASKED_BIT_ENABLE(RESET_CTL_REQUEST_RESET));
+	ack = intel_uncore_read_fw(uncore, reg);
+	if (ack & RESET_CTL_CAT_ERROR) {
+		/*
+		 * For catastrophic errors, ready-for-reset sequence
+		 * needs to be bypassed: HAS#396813
+		 */
+		request = RESET_CTL_CAT_ERROR;
+		mask = RESET_CTL_CAT_ERROR;
+
+		/* Catastrophic errors need to be cleared by HW */
+		ack = 0;
+	} else if (!(ack & RESET_CTL_READY_TO_RESET)) {
+		request = RESET_CTL_REQUEST_RESET;
+		mask = RESET_CTL_READY_TO_RESET;
+		ack = RESET_CTL_READY_TO_RESET;
+	} else {
+		return 0;
+	}
 
-	ret = __intel_wait_for_register_fw(dev_priv,
-					   RING_RESET_CTL(engine->mmio_base),
-					   RESET_CTL_READY_TO_RESET,
-					   RESET_CTL_READY_TO_RESET,
-					   700, 0,
-					   NULL);
+	intel_uncore_write_fw(uncore, reg, _MASKED_BIT_ENABLE(request));
+	ret = __intel_wait_for_register_fw(uncore, reg, mask, ack,
+					   700, 0, NULL);
 	if (ret)
-		DRM_ERROR("%s: reset request timeout\n", engine->name);
+		DRM_ERROR("%s reset request timed out: {request: %08x, RESET_CTL: %08x}\n",
+			  engine->name, request,
+			  intel_uncore_read_fw(uncore, reg));
 
 	return ret;
 }
 
 static void gen8_engine_reset_cancel(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = engine->i915;
-
-	I915_WRITE_FW(RING_RESET_CTL(engine->mmio_base),
-		      _MASKED_BIT_DISABLE(RESET_CTL_REQUEST_RESET));
+	intel_uncore_write_fw(engine->uncore,
+			      RING_RESET_CTL(engine->mmio_base),
+			      _MASKED_BIT_DISABLE(RESET_CTL_REQUEST_RESET));
 }
 
 static int gen8_reset_engines(struct drm_i915_private *i915,
-			      unsigned int engine_mask,
+			      intel_engine_mask_t engine_mask,
 			      unsigned int retry)
 {
 	struct intel_engine_cs *engine;
 	const bool reset_non_ready = retry >= 1;
-	unsigned int tmp;
+	intel_engine_mask_t tmp;
 	int ret;
 
 	for_each_engine_masked(engine, i915, engine_mask, tmp) {
@@ -527,14 +573,11 @@ skip_reset:
 }
 
 typedef int (*reset_func)(struct drm_i915_private *,
-			  unsigned int engine_mask,
+			  intel_engine_mask_t engine_mask,
 			  unsigned int retry);
 
 static reset_func intel_get_gpu_reset(struct drm_i915_private *i915)
 {
-	if (!i915_modparams.reset)
-		return NULL;
-
 	if (INTEL_GEN(i915) >= 8)
 		return gen8_reset_engines;
 	else if (INTEL_GEN(i915) >= 6)
@@ -551,7 +594,8 @@ static reset_func intel_get_gpu_reset(struct drm_i915_private *i915)
 		return NULL;
 }
 
-int intel_gpu_reset(struct drm_i915_private *i915, unsigned int engine_mask)
+int intel_gpu_reset(struct drm_i915_private *i915,
+		    intel_engine_mask_t engine_mask)
 {
 	const int retries = engine_mask == ALL_ENGINES ? RESET_MAX_RETRIES : 1;
 	reset_func reset;
@@ -566,7 +610,7 @@ int intel_gpu_reset(struct drm_i915_private *i915, unsigned int engine_mask)
 	 * If the power well sleeps during the reset, the reset
 	 * request may be dropped and never completes (causing -EIO).
 	 */
-	intel_uncore_forcewake_get(i915, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(&i915->uncore, FORCEWAKE_ALL);
 	for (retry = 0; ret == -ETIMEDOUT && retry < retries; retry++) {
 		/*
 		 * We stop engines, otherwise we might get failed reset and a
@@ -582,14 +626,15 @@ int intel_gpu_reset(struct drm_i915_private *i915, unsigned int engine_mask)
 		 *
 		 * FIXME: Wa for more modern gens needs to be validated
 		 */
-		i915_stop_engines(i915, engine_mask);
+		if (retry)
+			i915_stop_engines(i915, engine_mask);
 
 		GEM_TRACE("engine_mask=%x\n", engine_mask);
 		preempt_disable();
 		ret = reset(i915, engine_mask, retry);
 		preempt_enable();
 	}
-	intel_uncore_forcewake_put(i915, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&i915->uncore, FORCEWAKE_ALL);
 
 	return ret;
 }
@@ -599,6 +644,9 @@ bool intel_has_gpu_reset(struct drm_i915_private *i915)
 	if (USES_GUC(i915))
 		return false;
 
+	if (!i915_modparams.reset)
+		return NULL;
+
 	return intel_get_gpu_reset(i915);
 }
 
@@ -615,9 +663,9 @@ int intel_reset_guc(struct drm_i915_private *i915)
 
 	GEM_BUG_ON(!HAS_GUC(i915));
 
-	intel_uncore_forcewake_get(i915, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(&i915->uncore, FORCEWAKE_ALL);
 	ret = gen6_hw_domain_reset(i915, guc_domain);
-	intel_uncore_forcewake_put(i915, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&i915->uncore, FORCEWAKE_ALL);
 
 	return ret;
 }
@@ -635,10 +683,36 @@ static void reset_prepare_engine(struct intel_engine_cs *engine)
 	 * written to the powercontext is undefined and so we may lose
 	 * GPU state upon resume, i.e. fail to restart after a reset.
 	 */
-	intel_uncore_forcewake_get(engine->i915, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(engine->uncore, FORCEWAKE_ALL);
 	engine->reset.prepare(engine);
 }
 
+static void revoke_mmaps(struct drm_i915_private *i915)
+{
+	int i;
+
+	for (i = 0; i < i915->num_fence_regs; i++) {
+		struct drm_vma_offset_node *node;
+		struct i915_vma *vma;
+		u64 vma_offset;
+
+		vma = READ_ONCE(i915->fence_regs[i].vma);
+		if (!vma)
+			continue;
+
+		if (!i915_vma_has_userfault(vma))
+			continue;
+
+		GEM_BUG_ON(vma->fence != &i915->fence_regs[i]);
+		node = &vma->obj->base.vma_node;
+		vma_offset = vma->ggtt_view.partial.offset << PAGE_SHIFT;
+		unmap_mapping_range(i915->drm.anon_inode->i_mapping,
+				    drm_vma_node_offset_addr(node) + vma_offset,
+				    vma->size,
+				    1);
+	}
+}
+
 static void reset_prepare(struct drm_i915_private *i915)
 {
 	struct intel_engine_cs *engine;
@@ -647,10 +721,16 @@ static void reset_prepare(struct drm_i915_private *i915)
 	for_each_engine(engine, i915, id)
 		reset_prepare_engine(engine);
 
-	intel_uc_sanitize(i915);
+	intel_uc_reset_prepare(i915);
+}
+
+static void gt_revoke(struct drm_i915_private *i915)
+{
+	revoke_mmaps(i915);
 }
 
-static int gt_reset(struct drm_i915_private *i915, unsigned int stalled_mask)
+static int gt_reset(struct drm_i915_private *i915,
+		    intel_engine_mask_t stalled_mask)
 {
 	struct intel_engine_cs *engine;
 	enum intel_engine_id id;
@@ -665,7 +745,7 @@ static int gt_reset(struct drm_i915_private *i915, unsigned int stalled_mask)
 		return err;
 
 	for_each_engine(engine, i915, id)
-		intel_engine_reset(engine, stalled_mask & ENGINE_MASK(id));
+		intel_engine_reset(engine, stalled_mask & engine->mask);
 
 	i915_gem_restore_fences(i915);
 
@@ -675,7 +755,7 @@ static int gt_reset(struct drm_i915_private *i915, unsigned int stalled_mask)
 static void reset_finish_engine(struct intel_engine_cs *engine)
 {
 	engine->reset.finish(engine);
-	intel_uncore_forcewake_put(engine->i915, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(engine->uncore, FORCEWAKE_ALL);
 }
 
 struct i915_gpu_restart {
@@ -722,8 +802,10 @@ static void reset_finish(struct drm_i915_private *i915)
 	struct intel_engine_cs *engine;
 	enum intel_engine_id id;
 
-	for_each_engine(engine, i915, id)
+	for_each_engine(engine, i915, id) {
 		reset_finish_engine(engine);
+		intel_engine_signal_breadcrumbs(engine);
+	}
 }
 
 static void reset_restart(struct drm_i915_private *i915)
@@ -761,23 +843,19 @@ static void nop_submit_request(struct i915_request *request)
 	spin_lock_irqsave(&engine->timeline.lock, flags);
 	__i915_request_submit(request);
 	i915_request_mark_complete(request);
-	intel_engine_write_global_seqno(engine, request->global_seqno);
 	spin_unlock_irqrestore(&engine->timeline.lock, flags);
 
 	intel_engine_queue_breadcrumbs(engine);
 }
 
-void i915_gem_set_wedged(struct drm_i915_private *i915)
+static void __i915_gem_set_wedged(struct drm_i915_private *i915)
 {
 	struct i915_gpu_error *error = &i915->gpu_error;
 	struct intel_engine_cs *engine;
 	enum intel_engine_id id;
 
-	mutex_lock(&error->wedge_mutex);
-	if (test_bit(I915_WEDGED, &error->flags)) {
-		mutex_unlock(&error->wedge_mutex);
+	if (test_bit(I915_WEDGED, &error->flags))
 		return;
-	}
 
 	if (GEM_SHOW_DEBUG() && !intel_engines_are_idle(i915)) {
 		struct drm_printer p = drm_debug_printer(__func__);
@@ -793,11 +871,10 @@ void i915_gem_set_wedged(struct drm_i915_private *i915)
 	 * rolling the global seqno forward (since this would complete requests
 	 * for which we haven't set the fence error to EIO yet).
 	 */
-	for_each_engine(engine, i915, id)
-		reset_prepare_engine(engine);
+	reset_prepare(i915);
 
 	/* Even if the GPU reset fails, it should still stop the engines */
-	if (INTEL_GEN(i915) >= 5)
+	if (!INTEL_INFO(i915)->gpu_reset_clobbers_display)
 		intel_gpu_reset(i915, ALL_ENGINES);
 
 	for_each_engine(engine, i915, id) {
@@ -811,31 +888,35 @@ void i915_gem_set_wedged(struct drm_i915_private *i915)
 	 * either this call here to intel_engine_write_global_seqno, or the one
 	 * in nop_submit_request.
 	 */
-	synchronize_rcu();
+	synchronize_rcu_expedited();
 
 	/* Mark all executing requests as skipped */
 	for_each_engine(engine, i915, id)
 		engine->cancel_requests(engine);
 
-	for_each_engine(engine, i915, id) {
-		reset_finish_engine(engine);
-		intel_engine_signal_breadcrumbs(engine);
-	}
+	reset_finish(i915);
 
 	smp_mb__before_atomic();
 	set_bit(I915_WEDGED, &error->flags);
 
 	GEM_TRACE("end\n");
-	mutex_unlock(&error->wedge_mutex);
+}
 
-	wake_up_all(&error->reset_queue);
+void i915_gem_set_wedged(struct drm_i915_private *i915)
+{
+	struct i915_gpu_error *error = &i915->gpu_error;
+	intel_wakeref_t wakeref;
+
+	mutex_lock(&error->wedge_mutex);
+	with_intel_runtime_pm(i915, wakeref)
+		__i915_gem_set_wedged(i915);
+	mutex_unlock(&error->wedge_mutex);
 }
 
-bool i915_gem_unset_wedged(struct drm_i915_private *i915)
+static bool __i915_gem_unset_wedged(struct drm_i915_private *i915)
 {
 	struct i915_gpu_error *error = &i915->gpu_error;
 	struct i915_timeline *tl;
-	bool ret = false;
 
 	if (!test_bit(I915_WEDGED, &error->flags))
 		return true;
@@ -843,8 +924,6 @@ bool i915_gem_unset_wedged(struct drm_i915_private *i915)
 	if (!i915->gt.scratch) /* Never full initialised, recovery impossible */
 		return false;
 
-	mutex_lock(&error->wedge_mutex);
-
 	GEM_TRACE("start\n");
 
 	/*
@@ -860,30 +939,20 @@ bool i915_gem_unset_wedged(struct drm_i915_private *i915)
 	mutex_lock(&i915->gt.timelines.mutex);
 	list_for_each_entry(tl, &i915->gt.timelines.active_list, link) {
 		struct i915_request *rq;
-		long timeout;
 
 		rq = i915_active_request_get_unlocked(&tl->last_request);
 		if (!rq)
 			continue;
 
 		/*
-		 * We can't use our normal waiter as we want to
-		 * avoid recursively trying to handle the current
-		 * reset. The basic dma_fence_default_wait() installs
-		 * a callback for dma_fence_signal(), which is
-		 * triggered by our nop handler (indirectly, the
-		 * callback enables the signaler thread which is
-		 * woken by the nop_submit_request() advancing the seqno
-		 * and when the seqno passes the fence, the signaler
-		 * then signals the fence waking us up).
+		 * All internal dependencies (i915_requests) will have
+		 * been flushed by the set-wedge, but we may be stuck waiting
+		 * for external fences. These should all be capped to 10s
+		 * (I915_FENCE_TIMEOUT) so this wait should not be unbounded
+		 * in the worst case.
 		 */
-		timeout = dma_fence_default_wait(&rq->fence, true,
-						 MAX_SCHEDULE_TIMEOUT);
+		dma_fence_default_wait(&rq->fence, false, MAX_SCHEDULE_TIMEOUT);
 		i915_request_put(rq);
-		if (timeout < 0) {
-			mutex_unlock(&i915->gt.timelines.mutex);
-			goto unlock;
-		}
 	}
 	mutex_unlock(&i915->gt.timelines.mutex);
 
@@ -904,57 +973,38 @@ bool i915_gem_unset_wedged(struct drm_i915_private *i915)
 
 	smp_mb__before_atomic(); /* complete takeover before enabling execbuf */
 	clear_bit(I915_WEDGED, &i915->gpu_error.flags);
-	ret = true;
-unlock:
-	mutex_unlock(&i915->gpu_error.wedge_mutex);
 
-	return ret;
+	return true;
 }
 
-struct __i915_reset {
-	struct drm_i915_private *i915;
-	unsigned int stalled_mask;
-};
-
-static int __i915_reset__BKL(void *data)
+bool i915_gem_unset_wedged(struct drm_i915_private *i915)
 {
-	struct __i915_reset *arg = data;
-	int err;
+	struct i915_gpu_error *error = &i915->gpu_error;
+	bool result;
 
-	err = intel_gpu_reset(arg->i915, ALL_ENGINES);
-	if (err)
-		return err;
+	mutex_lock(&error->wedge_mutex);
+	result = __i915_gem_unset_wedged(i915);
+	mutex_unlock(&error->wedge_mutex);
 
-	return gt_reset(arg->i915, arg->stalled_mask);
+	return result;
 }
 
-#if RESET_UNDER_STOP_MACHINE
-/*
- * XXX An alternative to using stop_machine would be to park only the
- * processes that have a GGTT mmap. By remote parking the threads (SIGSTOP)
- * we should be able to prevent their memmory accesses via the lost fence
- * registers over the course of the reset without the potential recursive
- * of mutexes between the pagefault handler and reset.
- *
- * See igt/gem_mmap_gtt/hang
- */
-#define __do_reset(fn, arg) stop_machine(fn, arg, NULL)
-#else
-#define __do_reset(fn, arg) fn(arg)
-#endif
-
-static int do_reset(struct drm_i915_private *i915, unsigned int stalled_mask)
+static int do_reset(struct drm_i915_private *i915,
+		    intel_engine_mask_t stalled_mask)
 {
-	struct __i915_reset arg = { i915, stalled_mask };
 	int err, i;
 
-	err = __do_reset(__i915_reset__BKL, &arg);
+	gt_revoke(i915);
+
+	err = intel_gpu_reset(i915, ALL_ENGINES);
 	for (i = 0; err && i < RESET_MAX_RETRIES; i++) {
-		msleep(100);
-		err = __do_reset(__i915_reset__BKL, &arg);
+		msleep(10 * (i + 1));
+		err = intel_gpu_reset(i915, ALL_ENGINES);
 	}
+	if (err)
+		return err;
 
-	return err;
+	return gt_reset(i915, stalled_mask);
 }
 
 /**
@@ -966,8 +1016,6 @@ static int do_reset(struct drm_i915_private *i915, unsigned int stalled_mask)
  * Reset the chip.  Useful if a hang is detected. Marks the device as wedged
  * on failure.
  *
- * Caller must hold the struct_mutex.
- *
  * Procedure is fairly simple:
  *   - reset the chip using the reset reg
  *   - re-init context state
@@ -977,7 +1025,7 @@ static int do_reset(struct drm_i915_private *i915, unsigned int stalled_mask)
  *   - re-init display
  */
 void i915_reset(struct drm_i915_private *i915,
-		unsigned int stalled_mask,
+		intel_engine_mask_t stalled_mask,
 		const char *reason)
 {
 	struct i915_gpu_error *error = &i915->gpu_error;
@@ -990,7 +1038,7 @@ void i915_reset(struct drm_i915_private *i915,
 	GEM_BUG_ON(!test_bit(I915_RESET_BACKOFF, &error->flags));
 
 	/* Clear any previous failed attempts at recovery. Time to try again. */
-	if (!i915_gem_unset_wedged(i915))
+	if (!__i915_gem_unset_wedged(i915))
 		return;
 
 	if (reason)
@@ -1007,11 +1055,17 @@ void i915_reset(struct drm_i915_private *i915,
 		goto error;
 	}
 
+	if (INTEL_INFO(i915)->gpu_reset_clobbers_display)
+		intel_runtime_pm_disable_interrupts(i915);
+
 	if (do_reset(i915, stalled_mask)) {
 		dev_err(i915->drm.dev, "Failed to reset chip\n");
 		goto taint;
 	}
 
+	if (INTEL_INFO(i915)->gpu_reset_clobbers_display)
+		intel_runtime_pm_enable_interrupts(i915);
+
 	intel_overlay_reset(i915);
 
 	/*
@@ -1033,7 +1087,7 @@ void i915_reset(struct drm_i915_private *i915,
 
 finish:
 	reset_finish(i915);
-	if (!i915_terminally_wedged(error))
+	if (!__i915_wedged(error))
 		reset_restart(i915);
 	return;
 
@@ -1052,14 +1106,14 @@ taint:
 	 */
 	add_taint(TAINT_WARN, LOCKDEP_STILL_OK);
 error:
-	i915_gem_set_wedged(i915);
+	__i915_gem_set_wedged(i915);
 	goto finish;
 }
 
 static inline int intel_gt_reset_engine(struct drm_i915_private *i915,
 					struct intel_engine_cs *engine)
 {
-	return intel_gpu_reset(i915, intel_engine_flag(engine));
+	return intel_gpu_reset(i915, engine->mask);
 }
 
 /**
@@ -1144,7 +1198,12 @@ static void i915_reset_device(struct drm_i915_private *i915,
 	i915_wedge_on_timeout(&w, i915, 5 * HZ) {
 		intel_prepare_reset(i915);
 
+		/* Flush everyone using a resource about to be clobbered */
+		synchronize_srcu_expedited(&error->reset_backoff_srcu);
+
+		mutex_lock(&error->wedge_mutex);
 		i915_reset(i915, engine_mask, reason);
+		mutex_unlock(&error->wedge_mutex);
 
 		intel_finish_reset(i915);
 	}
@@ -1153,44 +1212,50 @@ static void i915_reset_device(struct drm_i915_private *i915,
 		kobject_uevent_env(kobj, KOBJ_CHANGE, reset_done_event);
 }
 
-void i915_clear_error_registers(struct drm_i915_private *dev_priv)
+static void clear_register(struct intel_uncore *uncore, i915_reg_t reg)
 {
+	intel_uncore_rmw(uncore, reg, 0, 0);
+}
+
+void i915_clear_error_registers(struct drm_i915_private *i915)
+{
+	struct intel_uncore *uncore = &i915->uncore;
 	u32 eir;
 
-	if (!IS_GEN(dev_priv, 2))
-		I915_WRITE(PGTBL_ER, I915_READ(PGTBL_ER));
+	if (!IS_GEN(i915, 2))
+		clear_register(uncore, PGTBL_ER);
 
-	if (INTEL_GEN(dev_priv) < 4)
-		I915_WRITE(IPEIR, I915_READ(IPEIR));
+	if (INTEL_GEN(i915) < 4)
+		clear_register(uncore, IPEIR(RENDER_RING_BASE));
 	else
-		I915_WRITE(IPEIR_I965, I915_READ(IPEIR_I965));
+		clear_register(uncore, IPEIR_I965);
 
-	I915_WRITE(EIR, I915_READ(EIR));
-	eir = I915_READ(EIR);
+	clear_register(uncore, EIR);
+	eir = intel_uncore_read(uncore, EIR);
 	if (eir) {
 		/*
 		 * some errors might have become stuck,
 		 * mask them.
 		 */
 		DRM_DEBUG_DRIVER("EIR stuck: 0x%08x, masking\n", eir);
-		I915_WRITE(EMR, I915_READ(EMR) | eir);
-		I915_WRITE(IIR, I915_MASTER_ERROR_INTERRUPT);
+		rmw_set(uncore, EMR, eir);
+		intel_uncore_write(uncore, GEN2_IIR,
+				   I915_MASTER_ERROR_INTERRUPT);
 	}
 
-	if (INTEL_GEN(dev_priv) >= 8) {
-		I915_WRITE(GEN8_RING_FAULT_REG,
-			   I915_READ(GEN8_RING_FAULT_REG) & ~RING_FAULT_VALID);
-		POSTING_READ(GEN8_RING_FAULT_REG);
-	} else if (INTEL_GEN(dev_priv) >= 6) {
+	if (INTEL_GEN(i915) >= 8) {
+		rmw_clear(uncore, GEN8_RING_FAULT_REG, RING_FAULT_VALID);
+		intel_uncore_posting_read(uncore, GEN8_RING_FAULT_REG);
+	} else if (INTEL_GEN(i915) >= 6) {
 		struct intel_engine_cs *engine;
 		enum intel_engine_id id;
 
-		for_each_engine(engine, dev_priv, id) {
-			I915_WRITE(RING_FAULT_REG(engine),
-				   I915_READ(RING_FAULT_REG(engine)) &
-				   ~RING_FAULT_VALID);
+		for_each_engine(engine, i915, id) {
+			rmw_clear(uncore,
+				  RING_FAULT_REG(engine), RING_FAULT_VALID);
+			intel_uncore_posting_read(uncore,
+						  RING_FAULT_REG(engine));
 		}
-		POSTING_READ(RING_FAULT_REG(dev_priv->engine[RCS]));
 	}
 }
 
@@ -1208,13 +1273,14 @@ void i915_clear_error_registers(struct drm_i915_private *dev_priv)
  * of a ring dump etc.).
  */
 void i915_handle_error(struct drm_i915_private *i915,
-		       u32 engine_mask,
+		       intel_engine_mask_t engine_mask,
 		       unsigned long flags,
 		       const char *fmt, ...)
 {
+	struct i915_gpu_error *error = &i915->gpu_error;
 	struct intel_engine_cs *engine;
 	intel_wakeref_t wakeref;
-	unsigned int tmp;
+	intel_engine_mask_t tmp;
 	char error_msg[80];
 	char *msg = NULL;
 
@@ -1237,7 +1303,7 @@ void i915_handle_error(struct drm_i915_private *i915,
 	 */
 	wakeref = intel_runtime_pm_get(i915);
 
-	engine_mask &= INTEL_INFO(i915)->ring_mask;
+	engine_mask &= INTEL_INFO(i915)->engine_mask;
 
 	if (flags & I915_ERROR_CAPTURE) {
 		i915_capture_error_state(i915, engine_mask, msg);
@@ -1248,20 +1314,19 @@ void i915_handle_error(struct drm_i915_private *i915,
 	 * Try engine reset when available. We fall back to full reset if
 	 * single reset fails.
 	 */
-	if (intel_has_reset_engine(i915) &&
-	    !i915_terminally_wedged(&i915->gpu_error)) {
+	if (intel_has_reset_engine(i915) && !__i915_wedged(error)) {
 		for_each_engine_masked(engine, i915, engine_mask, tmp) {
 			BUILD_BUG_ON(I915_RESET_MODESET >= I915_RESET_ENGINE);
 			if (test_and_set_bit(I915_RESET_ENGINE + engine->id,
-					     &i915->gpu_error.flags))
+					     &error->flags))
 				continue;
 
 			if (i915_reset_engine(engine, msg) == 0)
-				engine_mask &= ~intel_engine_flag(engine);
+				engine_mask &= ~engine->mask;
 
 			clear_bit(I915_RESET_ENGINE + engine->id,
-				  &i915->gpu_error.flags);
-			wake_up_bit(&i915->gpu_error.flags,
+				  &error->flags);
+			wake_up_bit(&error->flags,
 				    I915_RESET_ENGINE + engine->id);
 		}
 	}
@@ -1270,18 +1335,20 @@ void i915_handle_error(struct drm_i915_private *i915,
 		goto out;
 
 	/* Full reset needs the mutex, stop any other user trying to do so. */
-	if (test_and_set_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags)) {
-		wait_event(i915->gpu_error.reset_queue,
-			   !test_bit(I915_RESET_BACKOFF,
-				     &i915->gpu_error.flags));
-		goto out;
+	if (test_and_set_bit(I915_RESET_BACKOFF, &error->flags)) {
+		wait_event(error->reset_queue,
+			   !test_bit(I915_RESET_BACKOFF, &error->flags));
+		goto out; /* piggy-back on the other reset */
 	}
 
+	/* Make sure i915_reset_trylock() sees the I915_RESET_BACKOFF */
+	synchronize_rcu_expedited();
+
 	/* Prevent any other reset-engine attempt. */
 	for_each_engine(engine, i915, tmp) {
 		while (test_and_set_bit(I915_RESET_ENGINE + engine->id,
-					&i915->gpu_error.flags))
-			wait_on_bit(&i915->gpu_error.flags,
+					&error->flags))
+			wait_on_bit(&error->flags,
 				    I915_RESET_ENGINE + engine->id,
 				    TASK_UNINTERRUPTIBLE);
 	}
@@ -1290,16 +1357,74 @@ void i915_handle_error(struct drm_i915_private *i915,
 
 	for_each_engine(engine, i915, tmp) {
 		clear_bit(I915_RESET_ENGINE + engine->id,
-			  &i915->gpu_error.flags);
+			  &error->flags);
 	}
 
-	clear_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags);
-	wake_up_all(&i915->gpu_error.reset_queue);
+	clear_bit(I915_RESET_BACKOFF, &error->flags);
+	wake_up_all(&error->reset_queue);
 
 out:
 	intel_runtime_pm_put(i915, wakeref);
 }
 
+int i915_reset_trylock(struct drm_i915_private *i915)
+{
+	struct i915_gpu_error *error = &i915->gpu_error;
+	int srcu;
+
+	might_lock(&error->reset_backoff_srcu);
+	might_sleep();
+
+	rcu_read_lock();
+	while (test_bit(I915_RESET_BACKOFF, &error->flags)) {
+		rcu_read_unlock();
+
+		if (wait_event_interruptible(error->reset_queue,
+					     !test_bit(I915_RESET_BACKOFF,
+						       &error->flags)))
+			return -EINTR;
+
+		rcu_read_lock();
+	}
+	srcu = srcu_read_lock(&error->reset_backoff_srcu);
+	rcu_read_unlock();
+
+	return srcu;
+}
+
+void i915_reset_unlock(struct drm_i915_private *i915, int tag)
+__releases(&i915->gpu_error.reset_backoff_srcu)
+{
+	struct i915_gpu_error *error = &i915->gpu_error;
+
+	srcu_read_unlock(&error->reset_backoff_srcu, tag);
+}
+
+int i915_terminally_wedged(struct drm_i915_private *i915)
+{
+	struct i915_gpu_error *error = &i915->gpu_error;
+
+	might_sleep();
+
+	if (!__i915_wedged(error))
+		return 0;
+
+	/* Reset still in progress? Maybe we will recover? */
+	if (!test_bit(I915_RESET_BACKOFF, &error->flags))
+		return -EIO;
+
+	/* XXX intel_reset_finish() still takes struct_mutex!!! */
+	if (mutex_is_locked(&i915->drm.struct_mutex))
+		return -EAGAIN;
+
+	if (wait_event_interruptible(error->reset_queue,
+				     !test_bit(I915_RESET_BACKOFF,
+					       &error->flags)))
+		return -EINTR;
+
+	return __i915_wedged(error) ? -EIO : 0;
+}
+
 bool i915_reset_flush(struct drm_i915_private *i915)
 {
 	int err;
diff --git a/drivers/gpu/drm/i915/i915_reset.h b/drivers/gpu/drm/i915/i915_reset.h
index f2d347f319df..3c0450289b8f 100644
--- a/drivers/gpu/drm/i915/i915_reset.h
+++ b/drivers/gpu/drm/i915/i915_reset.h
@@ -9,14 +9,18 @@
 
 #include <linux/compiler.h>
 #include <linux/types.h>
+#include <linux/srcu.h>
+
+#include "intel_engine_types.h"
 
 struct drm_i915_private;
+struct i915_request;
 struct intel_engine_cs;
 struct intel_guc;
 
 __printf(4, 5)
 void i915_handle_error(struct drm_i915_private *i915,
-		       u32 engine_mask,
+		       intel_engine_mask_t engine_mask,
 		       unsigned long flags,
 		       const char *fmt, ...);
 #define I915_ERROR_CAPTURE BIT(0)
@@ -24,7 +28,7 @@ void i915_handle_error(struct drm_i915_private *i915,
 void i915_clear_error_registers(struct drm_i915_private *i915);
 
 void i915_reset(struct drm_i915_private *i915,
-		unsigned int stalled_mask,
+		intel_engine_mask_t stalled_mask,
 		const char *reason);
 int i915_reset_engine(struct intel_engine_cs *engine,
 		      const char *reason);
@@ -32,10 +36,16 @@ int i915_reset_engine(struct intel_engine_cs *engine,
 void i915_reset_request(struct i915_request *rq, bool guilty);
 bool i915_reset_flush(struct drm_i915_private *i915);
 
+int __must_check i915_reset_trylock(struct drm_i915_private *i915);
+void i915_reset_unlock(struct drm_i915_private *i915, int tag);
+
+int i915_terminally_wedged(struct drm_i915_private *i915);
+
 bool intel_has_gpu_reset(struct drm_i915_private *i915);
 bool intel_has_reset_engine(struct drm_i915_private *i915);
 
-int intel_gpu_reset(struct drm_i915_private *i915, u32 engine_mask);
+int intel_gpu_reset(struct drm_i915_private *i915,
+		    intel_engine_mask_t engine_mask);
 
 int intel_reset_guc(struct drm_i915_private *i915);
 
diff --git a/drivers/gpu/drm/i915/i915_scheduler.c b/drivers/gpu/drm/i915/i915_scheduler.c
index 8bc042551692..39bc4f54e272 100644
--- a/drivers/gpu/drm/i915/i915_scheduler.c
+++ b/drivers/gpu/drm/i915/i915_scheduler.c
@@ -7,9 +7,16 @@
 #include <linux/mutex.h>
 
 #include "i915_drv.h"
+#include "i915_globals.h"
 #include "i915_request.h"
 #include "i915_scheduler.h"
 
+static struct i915_global_scheduler {
+	struct i915_global base;
+	struct kmem_cache *slab_dependencies;
+	struct kmem_cache *slab_priorities;
+} global;
+
 static DEFINE_SPINLOCK(schedule_lock);
 
 static const struct i915_request *
@@ -18,6 +25,11 @@ node_to_request(const struct i915_sched_node *node)
 	return container_of(node, const struct i915_request, sched);
 }
 
+static inline bool node_started(const struct i915_sched_node *node)
+{
+	return i915_request_started(node_to_request(node));
+}
+
 static inline bool node_signaled(const struct i915_sched_node *node)
 {
 	return i915_request_completed(node_to_request(node));
@@ -29,19 +41,20 @@ void i915_sched_node_init(struct i915_sched_node *node)
 	INIT_LIST_HEAD(&node->waiters_list);
 	INIT_LIST_HEAD(&node->link);
 	node->attr.priority = I915_PRIORITY_INVALID;
+	node->semaphores = 0;
+	node->flags = 0;
 }
 
 static struct i915_dependency *
-i915_dependency_alloc(struct drm_i915_private *i915)
+i915_dependency_alloc(void)
 {
-	return kmem_cache_alloc(i915->dependencies, GFP_KERNEL);
+	return kmem_cache_alloc(global.slab_dependencies, GFP_KERNEL);
 }
 
 static void
-i915_dependency_free(struct drm_i915_private *i915,
-		     struct i915_dependency *dep)
+i915_dependency_free(struct i915_dependency *dep)
 {
-	kmem_cache_free(i915->dependencies, dep);
+	kmem_cache_free(global.slab_dependencies, dep);
 }
 
 bool __i915_sched_node_add_dependency(struct i915_sched_node *node,
@@ -51,7 +64,7 @@ bool __i915_sched_node_add_dependency(struct i915_sched_node *node,
 {
 	bool ret = false;
 
-	spin_lock(&schedule_lock);
+	spin_lock_irq(&schedule_lock);
 
 	if (!node_signaled(signal)) {
 		INIT_LIST_HEAD(&dep->dfs_link);
@@ -60,39 +73,42 @@ bool __i915_sched_node_add_dependency(struct i915_sched_node *node,
 		dep->signaler = signal;
 		dep->flags = flags;
 
+		/* Keep track of whether anyone on this chain has a semaphore */
+		if (signal->flags & I915_SCHED_HAS_SEMAPHORE_CHAIN &&
+		    !node_started(signal))
+			node->flags |= I915_SCHED_HAS_SEMAPHORE_CHAIN;
+
 		ret = true;
 	}
 
-	spin_unlock(&schedule_lock);
+	spin_unlock_irq(&schedule_lock);
 
 	return ret;
 }
 
-int i915_sched_node_add_dependency(struct drm_i915_private *i915,
-				   struct i915_sched_node *node,
+int i915_sched_node_add_dependency(struct i915_sched_node *node,
 				   struct i915_sched_node *signal)
 {
 	struct i915_dependency *dep;
 
-	dep = i915_dependency_alloc(i915);
+	dep = i915_dependency_alloc();
 	if (!dep)
 		return -ENOMEM;
 
 	if (!__i915_sched_node_add_dependency(node, signal, dep,
 					      I915_DEPENDENCY_ALLOC))
-		i915_dependency_free(i915, dep);
+		i915_dependency_free(dep);
 
 	return 0;
 }
 
-void i915_sched_node_fini(struct drm_i915_private *i915,
-			  struct i915_sched_node *node)
+void i915_sched_node_fini(struct i915_sched_node *node)
 {
 	struct i915_dependency *dep, *tmp;
 
 	GEM_BUG_ON(!list_empty(&node->link));
 
-	spin_lock(&schedule_lock);
+	spin_lock_irq(&schedule_lock);
 
 	/*
 	 * Everyone we depended upon (the fences we wait to be signaled)
@@ -106,7 +122,7 @@ void i915_sched_node_fini(struct drm_i915_private *i915,
 
 		list_del(&dep->wait_link);
 		if (dep->flags & I915_DEPENDENCY_ALLOC)
-			i915_dependency_free(i915, dep);
+			i915_dependency_free(dep);
 	}
 
 	/* Remove ourselves from everyone who depends upon us */
@@ -116,10 +132,10 @@ void i915_sched_node_fini(struct drm_i915_private *i915,
 
 		list_del(&dep->signal_link);
 		if (dep->flags & I915_DEPENDENCY_ALLOC)
-			i915_dependency_free(i915, dep);
+			i915_dependency_free(dep);
 	}
 
-	spin_unlock(&schedule_lock);
+	spin_unlock_irq(&schedule_lock);
 }
 
 static inline struct i915_priolist *to_priolist(struct rb_node *rb)
@@ -193,7 +209,7 @@ find_priolist:
 	if (prio == I915_PRIORITY_NORMAL) {
 		p = &execlists->default_priolist;
 	} else {
-		p = kmem_cache_alloc(engine->i915->priorities, GFP_ATOMIC);
+		p = kmem_cache_alloc(global.slab_priorities, GFP_ATOMIC);
 		/* Convert an allocation failure to a priority bump */
 		if (unlikely(!p)) {
 			prio = I915_PRIORITY_NORMAL; /* recurses just once */
@@ -301,6 +317,10 @@ static void __i915_schedule(struct i915_request *rq,
 	list_for_each_entry(dep, &dfs, dfs_link) {
 		struct i915_sched_node *node = dep->signaler;
 
+		/* If we are already flying, we know we have no signalers */
+		if (node_started(node))
+			continue;
+
 		/*
 		 * Within an engine, there can be no cycle, but we may
 		 * refer to the same dependency chain multiple times
@@ -313,7 +333,6 @@ static void __i915_schedule(struct i915_request *rq,
 			if (node_signaled(p->signaler))
 				continue;
 
-			GEM_BUG_ON(p->signaler->attr.priority < node->attr.priority);
 			if (prio > READ_ONCE(p->signaler->attr.priority))
 				list_move_tail(&p->dfs_link, &dfs);
 		}
@@ -337,7 +356,7 @@ static void __i915_schedule(struct i915_request *rq,
 
 	memset(&cache, 0, sizeof(cache));
 	engine = rq->engine;
-	spin_lock_irq(&engine->timeline.lock);
+	spin_lock(&engine->timeline.lock);
 
 	/* Fifo and depth-first replacement ensure our deps execute before us */
 	list_for_each_entry_safe_reverse(dep, p, &dfs, dfs_link) {
@@ -388,30 +407,73 @@ static void __i915_schedule(struct i915_request *rq,
 		tasklet_hi_schedule(&engine->execlists.tasklet);
 	}
 
-	spin_unlock_irq(&engine->timeline.lock);
+	spin_unlock(&engine->timeline.lock);
 }
 
 void i915_schedule(struct i915_request *rq, const struct i915_sched_attr *attr)
 {
-	spin_lock(&schedule_lock);
+	spin_lock_irq(&schedule_lock);
 	__i915_schedule(rq, attr);
-	spin_unlock(&schedule_lock);
+	spin_unlock_irq(&schedule_lock);
 }
 
 void i915_schedule_bump_priority(struct i915_request *rq, unsigned int bump)
 {
 	struct i915_sched_attr attr;
+	unsigned long flags;
 
 	GEM_BUG_ON(bump & ~I915_PRIORITY_MASK);
 
 	if (READ_ONCE(rq->sched.attr.priority) == I915_PRIORITY_INVALID)
 		return;
 
-	spin_lock_bh(&schedule_lock);
+	spin_lock_irqsave(&schedule_lock, flags);
 
 	attr = rq->sched.attr;
 	attr.priority |= bump;
 	__i915_schedule(rq, &attr);
 
-	spin_unlock_bh(&schedule_lock);
+	spin_unlock_irqrestore(&schedule_lock, flags);
+}
+
+void __i915_priolist_free(struct i915_priolist *p)
+{
+	kmem_cache_free(global.slab_priorities, p);
+}
+
+static void i915_global_scheduler_shrink(void)
+{
+	kmem_cache_shrink(global.slab_dependencies);
+	kmem_cache_shrink(global.slab_priorities);
+}
+
+static void i915_global_scheduler_exit(void)
+{
+	kmem_cache_destroy(global.slab_dependencies);
+	kmem_cache_destroy(global.slab_priorities);
+}
+
+static struct i915_global_scheduler global = { {
+	.shrink = i915_global_scheduler_shrink,
+	.exit = i915_global_scheduler_exit,
+} };
+
+int __init i915_global_scheduler_init(void)
+{
+	global.slab_dependencies = KMEM_CACHE(i915_dependency,
+					      SLAB_HWCACHE_ALIGN);
+	if (!global.slab_dependencies)
+		return -ENOMEM;
+
+	global.slab_priorities = KMEM_CACHE(i915_priolist,
+					    SLAB_HWCACHE_ALIGN);
+	if (!global.slab_priorities)
+		goto err_priorities;
+
+	i915_global_register(&global.base);
+	return 0;
+
+err_priorities:
+	kmem_cache_destroy(global.slab_priorities);
+	return -ENOMEM;
 }
diff --git a/drivers/gpu/drm/i915/i915_scheduler.h b/drivers/gpu/drm/i915/i915_scheduler.h
index dbe9cb7ecd82..07d243acf553 100644
--- a/drivers/gpu/drm/i915/i915_scheduler.h
+++ b/drivers/gpu/drm/i915/i915_scheduler.h
@@ -8,80 +8,22 @@
 #define _I915_SCHEDULER_H_
 
 #include <linux/bitops.h>
+#include <linux/list.h>
 #include <linux/kernel.h>
 
-#include <uapi/drm/i915_drm.h>
+#include "i915_scheduler_types.h"
 
-struct drm_i915_private;
-struct i915_request;
-struct intel_engine_cs;
+#define priolist_for_each_request(it, plist, idx) \
+	for (idx = 0; idx < ARRAY_SIZE((plist)->requests); idx++) \
+		list_for_each_entry(it, &(plist)->requests[idx], sched.link)
 
-enum {
-	I915_PRIORITY_MIN = I915_CONTEXT_MIN_USER_PRIORITY - 1,
-	I915_PRIORITY_NORMAL = I915_CONTEXT_DEFAULT_PRIORITY,
-	I915_PRIORITY_MAX = I915_CONTEXT_MAX_USER_PRIORITY + 1,
-
-	I915_PRIORITY_INVALID = INT_MIN
-};
-
-#define I915_USER_PRIORITY_SHIFT 2
-#define I915_USER_PRIORITY(x) ((x) << I915_USER_PRIORITY_SHIFT)
-
-#define I915_PRIORITY_COUNT BIT(I915_USER_PRIORITY_SHIFT)
-#define I915_PRIORITY_MASK (I915_PRIORITY_COUNT - 1)
-
-#define I915_PRIORITY_WAIT	((u8)BIT(0))
-#define I915_PRIORITY_NEWCLIENT	((u8)BIT(1))
-
-struct i915_sched_attr {
-	/**
-	 * @priority: execution and service priority
-	 *
-	 * All clients are equal, but some are more equal than others!
-	 *
-	 * Requests from a context with a greater (more positive) value of
-	 * @priority will be executed before those with a lower @priority
-	 * value, forming a simple QoS.
-	 *
-	 * The &drm_i915_private.kernel_context is assigned the lowest priority.
-	 */
-	int priority;
-};
-
-/*
- * "People assume that time is a strict progression of cause to effect, but
- * actually, from a nonlinear, non-subjective viewpoint, it's more like a big
- * ball of wibbly-wobbly, timey-wimey ... stuff." -The Doctor, 2015
- *
- * Requests exist in a complex web of interdependencies. Each request
- * has to wait for some other request to complete before it is ready to be run
- * (e.g. we have to wait until the pixels have been rendering into a texture
- * before we can copy from it). We track the readiness of a request in terms
- * of fences, but we also need to keep the dependency tree for the lifetime
- * of the request (beyond the life of an individual fence). We use the tree
- * at various points to reorder the requests whilst keeping the requests
- * in order with respect to their various dependencies.
- *
- * There is no active component to the "scheduler". As we know the dependency
- * DAG of each request, we are able to insert it into a sorted queue when it
- * is ready, and are able to reorder its portion of the graph to accommodate
- * dynamic priority changes.
- */
-struct i915_sched_node {
-	struct list_head signalers_list; /* those before us, we depend upon */
-	struct list_head waiters_list; /* those after us, they depend upon us */
-	struct list_head link;
-	struct i915_sched_attr attr;
-};
-
-struct i915_dependency {
-	struct i915_sched_node *signaler;
-	struct list_head signal_link;
-	struct list_head wait_link;
-	struct list_head dfs_link;
-	unsigned long flags;
-#define I915_DEPENDENCY_ALLOC BIT(0)
-};
+#define priolist_for_each_request_consume(it, n, plist, idx) \
+	for (; \
+	     (plist)->used ? (idx = __ffs((plist)->used)), 1 : 0; \
+	     (plist)->used &= ~BIT(idx)) \
+		list_for_each_entry_safe(it, n, \
+					 &(plist)->requests[idx], \
+					 sched.link)
 
 void i915_sched_node_init(struct i915_sched_node *node);
 
@@ -90,12 +32,10 @@ bool __i915_sched_node_add_dependency(struct i915_sched_node *node,
 				      struct i915_dependency *dep,
 				      unsigned long flags);
 
-int i915_sched_node_add_dependency(struct drm_i915_private *i915,
-				   struct i915_sched_node *node,
+int i915_sched_node_add_dependency(struct i915_sched_node *node,
 				   struct i915_sched_node *signal);
 
-void i915_sched_node_fini(struct drm_i915_private *i915,
-			  struct i915_sched_node *node);
+void i915_sched_node_fini(struct i915_sched_node *node);
 
 void i915_schedule(struct i915_request *request,
 		   const struct i915_sched_attr *attr);
@@ -105,4 +45,11 @@ void i915_schedule_bump_priority(struct i915_request *rq, unsigned int bump);
 struct list_head *
 i915_sched_lookup_priolist(struct intel_engine_cs *engine, int prio);
 
+void __i915_priolist_free(struct i915_priolist *p);
+static inline void i915_priolist_free(struct i915_priolist *p)
+{
+	if (p->priority != I915_PRIORITY_NORMAL)
+		__i915_priolist_free(p);
+}
+
 #endif /* _I915_SCHEDULER_H_ */
diff --git a/drivers/gpu/drm/i915/i915_scheduler_types.h b/drivers/gpu/drm/i915/i915_scheduler_types.h
new file mode 100644
index 000000000000..f1af3916a808
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_scheduler_types.h
@@ -0,0 +1,72 @@
+/*
+ * SPDX-License-Identifier: MIT
+ *
+ * Copyright © 2018 Intel Corporation
+ */
+
+#ifndef _I915_SCHEDULER_TYPES_H_
+#define _I915_SCHEDULER_TYPES_H_
+
+#include <linux/list.h>
+
+#include "i915_priolist_types.h"
+#include "intel_engine_types.h"
+
+struct drm_i915_private;
+struct i915_request;
+struct intel_engine_cs;
+
+struct i915_sched_attr {
+	/**
+	 * @priority: execution and service priority
+	 *
+	 * All clients are equal, but some are more equal than others!
+	 *
+	 * Requests from a context with a greater (more positive) value of
+	 * @priority will be executed before those with a lower @priority
+	 * value, forming a simple QoS.
+	 *
+	 * The &drm_i915_private.kernel_context is assigned the lowest priority.
+	 */
+	int priority;
+};
+
+/*
+ * "People assume that time is a strict progression of cause to effect, but
+ * actually, from a nonlinear, non-subjective viewpoint, it's more like a big
+ * ball of wibbly-wobbly, timey-wimey ... stuff." -The Doctor, 2015
+ *
+ * Requests exist in a complex web of interdependencies. Each request
+ * has to wait for some other request to complete before it is ready to be run
+ * (e.g. we have to wait until the pixels have been rendering into a texture
+ * before we can copy from it). We track the readiness of a request in terms
+ * of fences, but we also need to keep the dependency tree for the lifetime
+ * of the request (beyond the life of an individual fence). We use the tree
+ * at various points to reorder the requests whilst keeping the requests
+ * in order with respect to their various dependencies.
+ *
+ * There is no active component to the "scheduler". As we know the dependency
+ * DAG of each request, we are able to insert it into a sorted queue when it
+ * is ready, and are able to reorder its portion of the graph to accommodate
+ * dynamic priority changes.
+ */
+struct i915_sched_node {
+	struct list_head signalers_list; /* those before us, we depend upon */
+	struct list_head waiters_list; /* those after us, they depend upon us */
+	struct list_head link;
+	struct i915_sched_attr attr;
+	unsigned int flags;
+#define I915_SCHED_HAS_SEMAPHORE_CHAIN	BIT(0)
+	intel_engine_mask_t semaphores;
+};
+
+struct i915_dependency {
+	struct i915_sched_node *signaler;
+	struct list_head signal_link;
+	struct list_head wait_link;
+	struct list_head dfs_link;
+	unsigned long flags;
+#define I915_DEPENDENCY_ALLOC BIT(0)
+};
+
+#endif /* _I915_SCHEDULER_TYPES_H_ */
diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c
index d2f2a9c2fabd..95f3dab1b229 100644
--- a/drivers/gpu/drm/i915/i915_suspend.c
+++ b/drivers/gpu/drm/i915/i915_suspend.c
@@ -25,8 +25,10 @@
  */
 
 #include <drm/i915_drm.h>
-#include "intel_drv.h"
+
 #include "i915_reg.h"
+#include "intel_drv.h"
+#include "intel_fbc.h"
 
 static void i915_save_display(struct drm_i915_private *dev_priv)
 {
diff --git a/drivers/gpu/drm/i915/i915_sw_fence.c b/drivers/gpu/drm/i915/i915_sw_fence.c
index 7c58b049ecb5..5387aafd3424 100644
--- a/drivers/gpu/drm/i915/i915_sw_fence.c
+++ b/drivers/gpu/drm/i915/i915_sw_fence.c
@@ -192,7 +192,7 @@ static void __i915_sw_fence_complete(struct i915_sw_fence *fence,
 	__i915_sw_fence_notify(fence, FENCE_FREE);
 }
 
-static void i915_sw_fence_complete(struct i915_sw_fence *fence)
+void i915_sw_fence_complete(struct i915_sw_fence *fence)
 {
 	debug_fence_assert(fence);
 
@@ -202,7 +202,7 @@ static void i915_sw_fence_complete(struct i915_sw_fence *fence)
 	__i915_sw_fence_complete(fence, NULL);
 }
 
-static void i915_sw_fence_await(struct i915_sw_fence *fence)
+void i915_sw_fence_await(struct i915_sw_fence *fence)
 {
 	debug_fence_assert(fence);
 	WARN_ON(atomic_inc_return(&fence->pending) <= 1);
@@ -359,11 +359,6 @@ int i915_sw_fence_await_sw_fence_gfp(struct i915_sw_fence *fence,
 	return __i915_sw_fence_await_sw_fence(fence, signaler, NULL, gfp);
 }
 
-struct i915_sw_dma_fence_cb {
-	struct dma_fence_cb base;
-	struct i915_sw_fence *fence;
-};
-
 struct i915_sw_dma_fence_cb_timer {
 	struct i915_sw_dma_fence_cb base;
 	struct dma_fence *dma;
@@ -480,6 +475,40 @@ int i915_sw_fence_await_dma_fence(struct i915_sw_fence *fence,
 	return ret;
 }
 
+static void __dma_i915_sw_fence_wake(struct dma_fence *dma,
+				     struct dma_fence_cb *data)
+{
+	struct i915_sw_dma_fence_cb *cb = container_of(data, typeof(*cb), base);
+
+	i915_sw_fence_complete(cb->fence);
+}
+
+int __i915_sw_fence_await_dma_fence(struct i915_sw_fence *fence,
+				    struct dma_fence *dma,
+				    struct i915_sw_dma_fence_cb *cb)
+{
+	int ret;
+
+	debug_fence_assert(fence);
+
+	if (dma_fence_is_signaled(dma))
+		return 0;
+
+	cb->fence = fence;
+	i915_sw_fence_await(fence);
+
+	ret = dma_fence_add_callback(dma, &cb->base, __dma_i915_sw_fence_wake);
+	if (ret == 0) {
+		ret = 1;
+	} else {
+		i915_sw_fence_complete(fence);
+		if (ret == -ENOENT) /* fence already signaled */
+			ret = 0;
+	}
+
+	return ret;
+}
+
 int i915_sw_fence_await_reservation(struct i915_sw_fence *fence,
 				    struct reservation_object *resv,
 				    const struct dma_fence_ops *exclude,
diff --git a/drivers/gpu/drm/i915/i915_sw_fence.h b/drivers/gpu/drm/i915/i915_sw_fence.h
index 0e055ea0179f..9cb5c3b307a6 100644
--- a/drivers/gpu/drm/i915/i915_sw_fence.h
+++ b/drivers/gpu/drm/i915/i915_sw_fence.h
@@ -9,14 +9,13 @@
 #ifndef _I915_SW_FENCE_H_
 #define _I915_SW_FENCE_H_
 
+#include <linux/dma-fence.h>
 #include <linux/gfp.h>
 #include <linux/kref.h>
 #include <linux/notifier.h> /* for NOTIFY_DONE */
 #include <linux/wait.h>
 
 struct completion;
-struct dma_fence;
-struct dma_fence_ops;
 struct reservation_object;
 
 struct i915_sw_fence {
@@ -68,10 +67,20 @@ int i915_sw_fence_await_sw_fence(struct i915_sw_fence *fence,
 int i915_sw_fence_await_sw_fence_gfp(struct i915_sw_fence *fence,
 				     struct i915_sw_fence *after,
 				     gfp_t gfp);
+
+struct i915_sw_dma_fence_cb {
+	struct dma_fence_cb base;
+	struct i915_sw_fence *fence;
+};
+
+int __i915_sw_fence_await_dma_fence(struct i915_sw_fence *fence,
+				    struct dma_fence *dma,
+				    struct i915_sw_dma_fence_cb *cb);
 int i915_sw_fence_await_dma_fence(struct i915_sw_fence *fence,
 				  struct dma_fence *dma,
 				  unsigned long timeout,
 				  gfp_t gfp);
+
 int i915_sw_fence_await_reservation(struct i915_sw_fence *fence,
 				    struct reservation_object *resv,
 				    const struct dma_fence_ops *exclude,
@@ -79,6 +88,9 @@ int i915_sw_fence_await_reservation(struct i915_sw_fence *fence,
 				    unsigned long timeout,
 				    gfp_t gfp);
 
+void i915_sw_fence_await(struct i915_sw_fence *fence);
+void i915_sw_fence_complete(struct i915_sw_fence *fence);
+
 static inline bool i915_sw_fence_signaled(const struct i915_sw_fence *fence)
 {
 	return atomic_read(&fence->pending) <= 0;
diff --git a/drivers/gpu/drm/i915/i915_timeline.c b/drivers/gpu/drm/i915/i915_timeline.c
index b2202d2e58a2..5fbea0892f33 100644
--- a/drivers/gpu/drm/i915/i915_timeline.c
+++ b/drivers/gpu/drm/i915/i915_timeline.c
@@ -6,19 +6,32 @@
 
 #include "i915_drv.h"
 
-#include "i915_timeline.h"
+#include "i915_active.h"
 #include "i915_syncmap.h"
+#include "i915_timeline.h"
+
+#define ptr_set_bit(ptr, bit) ((typeof(ptr))((unsigned long)(ptr) | BIT(bit)))
+#define ptr_test_bit(ptr, bit) ((unsigned long)(ptr) & BIT(bit))
 
 struct i915_timeline_hwsp {
-	struct i915_vma *vma;
+	struct i915_gt_timelines *gt;
 	struct list_head free_link;
+	struct i915_vma *vma;
 	u64 free_bitmap;
 };
 
-static inline struct i915_timeline_hwsp *
-i915_timeline_hwsp(const struct i915_timeline *tl)
+struct i915_timeline_cacheline {
+	struct i915_active active;
+	struct i915_timeline_hwsp *hwsp;
+	void *vaddr;
+#define CACHELINE_BITS 6
+#define CACHELINE_FREE CACHELINE_BITS
+};
+
+static inline struct drm_i915_private *
+hwsp_to_i915(struct i915_timeline_hwsp *hwsp)
 {
-	return tl->hwsp_ggtt->private;
+	return container_of(hwsp->gt, struct drm_i915_private, gt.timelines);
 }
 
 static struct i915_vma *__hwsp_alloc(struct drm_i915_private *i915)
@@ -71,6 +84,7 @@ hwsp_alloc(struct i915_timeline *timeline, unsigned int *cacheline)
 		vma->private = hwsp;
 		hwsp->vma = vma;
 		hwsp->free_bitmap = ~0ull;
+		hwsp->gt = gt;
 
 		spin_lock(&gt->hwsp_lock);
 		list_add(&hwsp->free_link, &gt->hwsp_free_list);
@@ -88,14 +102,9 @@ hwsp_alloc(struct i915_timeline *timeline, unsigned int *cacheline)
 	return hwsp->vma;
 }
 
-static void hwsp_free(struct i915_timeline *timeline)
+static void __idle_hwsp_free(struct i915_timeline_hwsp *hwsp, int cacheline)
 {
-	struct i915_gt_timelines *gt = &timeline->i915->gt.timelines;
-	struct i915_timeline_hwsp *hwsp;
-
-	hwsp = i915_timeline_hwsp(timeline);
-	if (!hwsp) /* leave global HWSP alone! */
-		return;
+	struct i915_gt_timelines *gt = hwsp->gt;
 
 	spin_lock(&gt->hwsp_lock);
 
@@ -103,7 +112,8 @@ static void hwsp_free(struct i915_timeline *timeline)
 	if (!hwsp->free_bitmap)
 		list_add_tail(&hwsp->free_link, &gt->hwsp_free_list);
 
-	hwsp->free_bitmap |= BIT_ULL(timeline->hwsp_offset / CACHELINE_BYTES);
+	GEM_BUG_ON(cacheline >= BITS_PER_TYPE(hwsp->free_bitmap));
+	hwsp->free_bitmap |= BIT_ULL(cacheline);
 
 	/* And if no one is left using it, give the page back to the system */
 	if (hwsp->free_bitmap == ~0ull) {
@@ -115,9 +125,78 @@ static void hwsp_free(struct i915_timeline *timeline)
 	spin_unlock(&gt->hwsp_lock);
 }
 
+static void __idle_cacheline_free(struct i915_timeline_cacheline *cl)
+{
+	GEM_BUG_ON(!i915_active_is_idle(&cl->active));
+
+	i915_gem_object_unpin_map(cl->hwsp->vma->obj);
+	i915_vma_put(cl->hwsp->vma);
+	__idle_hwsp_free(cl->hwsp, ptr_unmask_bits(cl->vaddr, CACHELINE_BITS));
+
+	i915_active_fini(&cl->active);
+	kfree(cl);
+}
+
+static void __cacheline_retire(struct i915_active *active)
+{
+	struct i915_timeline_cacheline *cl =
+		container_of(active, typeof(*cl), active);
+
+	i915_vma_unpin(cl->hwsp->vma);
+	if (ptr_test_bit(cl->vaddr, CACHELINE_FREE))
+		__idle_cacheline_free(cl);
+}
+
+static struct i915_timeline_cacheline *
+cacheline_alloc(struct i915_timeline_hwsp *hwsp, unsigned int cacheline)
+{
+	struct i915_timeline_cacheline *cl;
+	void *vaddr;
+
+	GEM_BUG_ON(cacheline >= BIT(CACHELINE_BITS));
+
+	cl = kmalloc(sizeof(*cl), GFP_KERNEL);
+	if (!cl)
+		return ERR_PTR(-ENOMEM);
+
+	vaddr = i915_gem_object_pin_map(hwsp->vma->obj, I915_MAP_WB);
+	if (IS_ERR(vaddr)) {
+		kfree(cl);
+		return ERR_CAST(vaddr);
+	}
+
+	i915_vma_get(hwsp->vma);
+	cl->hwsp = hwsp;
+	cl->vaddr = page_pack_bits(vaddr, cacheline);
+
+	i915_active_init(hwsp_to_i915(hwsp), &cl->active, __cacheline_retire);
+
+	return cl;
+}
+
+static void cacheline_acquire(struct i915_timeline_cacheline *cl)
+{
+	if (cl && i915_active_acquire(&cl->active))
+		__i915_vma_pin(cl->hwsp->vma);
+}
+
+static void cacheline_release(struct i915_timeline_cacheline *cl)
+{
+	if (cl)
+		i915_active_release(&cl->active);
+}
+
+static void cacheline_free(struct i915_timeline_cacheline *cl)
+{
+	GEM_BUG_ON(ptr_test_bit(cl->vaddr, CACHELINE_FREE));
+	cl->vaddr = ptr_set_bit(cl->vaddr, CACHELINE_FREE);
+
+	if (i915_active_is_idle(&cl->active))
+		__idle_cacheline_free(cl);
+}
+
 int i915_timeline_init(struct drm_i915_private *i915,
 		       struct i915_timeline *timeline,
-		       const char *name,
 		       struct i915_vma *hwsp)
 {
 	void *vaddr;
@@ -133,37 +212,47 @@ int i915_timeline_init(struct drm_i915_private *i915,
 	BUILD_BUG_ON(KSYNCMAP < I915_NUM_ENGINES);
 
 	timeline->i915 = i915;
-	timeline->name = name;
 	timeline->pin_count = 0;
 	timeline->has_initial_breadcrumb = !hwsp;
+	timeline->hwsp_cacheline = NULL;
 
-	timeline->hwsp_offset = I915_GEM_HWS_SEQNO_ADDR;
 	if (!hwsp) {
+		struct i915_timeline_cacheline *cl;
 		unsigned int cacheline;
 
 		hwsp = hwsp_alloc(timeline, &cacheline);
 		if (IS_ERR(hwsp))
 			return PTR_ERR(hwsp);
 
+		cl = cacheline_alloc(hwsp->private, cacheline);
+		if (IS_ERR(cl)) {
+			__idle_hwsp_free(hwsp->private, cacheline);
+			return PTR_ERR(cl);
+		}
+
+		timeline->hwsp_cacheline = cl;
 		timeline->hwsp_offset = cacheline * CACHELINE_BYTES;
-	}
-	timeline->hwsp_ggtt = i915_vma_get(hwsp);
 
-	vaddr = i915_gem_object_pin_map(hwsp->obj, I915_MAP_WB);
-	if (IS_ERR(vaddr)) {
-		hwsp_free(timeline);
-		i915_vma_put(hwsp);
-		return PTR_ERR(vaddr);
+		vaddr = page_mask_bits(cl->vaddr);
+	} else {
+		timeline->hwsp_offset = I915_GEM_HWS_SEQNO_ADDR;
+
+		vaddr = i915_gem_object_pin_map(hwsp->obj, I915_MAP_WB);
+		if (IS_ERR(vaddr))
+			return PTR_ERR(vaddr);
 	}
 
 	timeline->hwsp_seqno =
 		memset(vaddr + timeline->hwsp_offset, 0, CACHELINE_BYTES);
 
+	timeline->hwsp_ggtt = i915_vma_get(hwsp);
+	GEM_BUG_ON(timeline->hwsp_offset >= hwsp->size);
+
 	timeline->fence_context = dma_fence_context_alloc(1);
 
 	spin_lock_init(&timeline->lock);
+	mutex_init(&timeline->mutex);
 
-	INIT_ACTIVE_REQUEST(&timeline->barrier);
 	INIT_ACTIVE_REQUEST(&timeline->last_request);
 	INIT_LIST_HEAD(&timeline->requests);
 
@@ -236,18 +325,19 @@ void i915_timeline_fini(struct i915_timeline *timeline)
 {
 	GEM_BUG_ON(timeline->pin_count);
 	GEM_BUG_ON(!list_empty(&timeline->requests));
-	GEM_BUG_ON(i915_active_request_isset(&timeline->barrier));
 
 	i915_syncmap_free(&timeline->sync);
-	hwsp_free(timeline);
 
-	i915_gem_object_unpin_map(timeline->hwsp_ggtt->obj);
+	if (timeline->hwsp_cacheline)
+		cacheline_free(timeline->hwsp_cacheline);
+	else
+		i915_gem_object_unpin_map(timeline->hwsp_ggtt->obj);
+
 	i915_vma_put(timeline->hwsp_ggtt);
 }
 
 struct i915_timeline *
 i915_timeline_create(struct drm_i915_private *i915,
-		     const char *name,
 		     struct i915_vma *global_hwsp)
 {
 	struct i915_timeline *timeline;
@@ -257,7 +347,7 @@ i915_timeline_create(struct drm_i915_private *i915,
 	if (!timeline)
 		return ERR_PTR(-ENOMEM);
 
-	err = i915_timeline_init(i915, timeline, name, global_hwsp);
+	err = i915_timeline_init(i915, timeline, global_hwsp);
 	if (err) {
 		kfree(timeline);
 		return ERR_PTR(err);
@@ -284,6 +374,7 @@ int i915_timeline_pin(struct i915_timeline *tl)
 		i915_ggtt_offset(tl->hwsp_ggtt) +
 		offset_in_page(tl->hwsp_offset);
 
+	cacheline_acquire(tl->hwsp_cacheline);
 	timeline_add_to_active(tl);
 
 	return 0;
@@ -293,6 +384,157 @@ unpin:
 	return err;
 }
 
+static u32 timeline_advance(struct i915_timeline *tl)
+{
+	GEM_BUG_ON(!tl->pin_count);
+	GEM_BUG_ON(tl->seqno & tl->has_initial_breadcrumb);
+
+	return tl->seqno += 1 + tl->has_initial_breadcrumb;
+}
+
+static void timeline_rollback(struct i915_timeline *tl)
+{
+	tl->seqno -= 1 + tl->has_initial_breadcrumb;
+}
+
+static noinline int
+__i915_timeline_get_seqno(struct i915_timeline *tl,
+			  struct i915_request *rq,
+			  u32 *seqno)
+{
+	struct i915_timeline_cacheline *cl;
+	unsigned int cacheline;
+	struct i915_vma *vma;
+	void *vaddr;
+	int err;
+
+	/*
+	 * If there is an outstanding GPU reference to this cacheline,
+	 * such as it being sampled by a HW semaphore on another timeline,
+	 * we cannot wraparound our seqno value (the HW semaphore does
+	 * a strict greater-than-or-equals compare, not i915_seqno_passed).
+	 * So if the cacheline is still busy, we must detach ourselves
+	 * from it and leave it inflight alongside its users.
+	 *
+	 * However, if nobody is watching and we can guarantee that nobody
+	 * will, we could simply reuse the same cacheline.
+	 *
+	 * if (i915_active_request_is_signaled(&tl->last_request) &&
+	 *     i915_active_is_signaled(&tl->hwsp_cacheline->active))
+	 *	return 0;
+	 *
+	 * That seems unlikely for a busy timeline that needed to wrap in
+	 * the first place, so just replace the cacheline.
+	 */
+
+	vma = hwsp_alloc(tl, &cacheline);
+	if (IS_ERR(vma)) {
+		err = PTR_ERR(vma);
+		goto err_rollback;
+	}
+
+	err = i915_vma_pin(vma, 0, 0, PIN_GLOBAL | PIN_HIGH);
+	if (err) {
+		__idle_hwsp_free(vma->private, cacheline);
+		goto err_rollback;
+	}
+
+	cl = cacheline_alloc(vma->private, cacheline);
+	if (IS_ERR(cl)) {
+		err = PTR_ERR(cl);
+		__idle_hwsp_free(vma->private, cacheline);
+		goto err_unpin;
+	}
+	GEM_BUG_ON(cl->hwsp->vma != vma);
+
+	/*
+	 * Attach the old cacheline to the current request, so that we only
+	 * free it after the current request is retired, which ensures that
+	 * all writes into the cacheline from previous requests are complete.
+	 */
+	err = i915_active_ref(&tl->hwsp_cacheline->active,
+			      tl->fence_context, rq);
+	if (err)
+		goto err_cacheline;
+
+	cacheline_release(tl->hwsp_cacheline); /* ownership now xfered to rq */
+	cacheline_free(tl->hwsp_cacheline);
+
+	i915_vma_unpin(tl->hwsp_ggtt); /* binding kept alive by old cacheline */
+	i915_vma_put(tl->hwsp_ggtt);
+
+	tl->hwsp_ggtt = i915_vma_get(vma);
+
+	vaddr = page_mask_bits(cl->vaddr);
+	tl->hwsp_offset = cacheline * CACHELINE_BYTES;
+	tl->hwsp_seqno =
+		memset(vaddr + tl->hwsp_offset, 0, CACHELINE_BYTES);
+
+	tl->hwsp_offset += i915_ggtt_offset(vma);
+
+	cacheline_acquire(cl);
+	tl->hwsp_cacheline = cl;
+
+	*seqno = timeline_advance(tl);
+	GEM_BUG_ON(i915_seqno_passed(*tl->hwsp_seqno, *seqno));
+	return 0;
+
+err_cacheline:
+	cacheline_free(cl);
+err_unpin:
+	i915_vma_unpin(vma);
+err_rollback:
+	timeline_rollback(tl);
+	return err;
+}
+
+int i915_timeline_get_seqno(struct i915_timeline *tl,
+			    struct i915_request *rq,
+			    u32 *seqno)
+{
+	*seqno = timeline_advance(tl);
+
+	/* Replace the HWSP on wraparound for HW semaphores */
+	if (unlikely(!*seqno && tl->hwsp_cacheline))
+		return __i915_timeline_get_seqno(tl, rq, seqno);
+
+	return 0;
+}
+
+static int cacheline_ref(struct i915_timeline_cacheline *cl,
+			 struct i915_request *rq)
+{
+	return i915_active_ref(&cl->active, rq->fence.context, rq);
+}
+
+int i915_timeline_read_hwsp(struct i915_request *from,
+			    struct i915_request *to,
+			    u32 *hwsp)
+{
+	struct i915_timeline_cacheline *cl = from->hwsp_cacheline;
+	struct i915_timeline *tl = from->timeline;
+	int err;
+
+	GEM_BUG_ON(to->timeline == tl);
+
+	mutex_lock_nested(&tl->mutex, SINGLE_DEPTH_NESTING);
+	err = i915_request_completed(from);
+	if (!err)
+		err = cacheline_ref(cl, to);
+	if (!err) {
+		if (likely(cl == tl->hwsp_cacheline)) {
+			*hwsp = tl->hwsp_offset;
+		} else { /* across a seqno wrap, recover the original offset */
+			*hwsp = i915_ggtt_offset(cl->hwsp->vma) +
+				ptr_unmask_bits(cl->vaddr, CACHELINE_BITS) *
+				CACHELINE_BYTES;
+		}
+	}
+	mutex_unlock(&tl->mutex);
+
+	return err;
+}
+
 void i915_timeline_unpin(struct i915_timeline *tl)
 {
 	GEM_BUG_ON(!tl->pin_count);
@@ -300,6 +542,7 @@ void i915_timeline_unpin(struct i915_timeline *tl)
 		return;
 
 	timeline_remove_from_active(tl);
+	cacheline_release(tl->hwsp_cacheline);
 
 	/*
 	 * Since this timeline is idle, all bariers upon which we were waiting
diff --git a/drivers/gpu/drm/i915/i915_timeline.h b/drivers/gpu/drm/i915/i915_timeline.h
index 7bec7d2e45bf..27668a1a69a3 100644
--- a/drivers/gpu/drm/i915/i915_timeline.h
+++ b/drivers/gpu/drm/i915/i915_timeline.h
@@ -25,76 +25,14 @@
 #ifndef I915_TIMELINE_H
 #define I915_TIMELINE_H
 
-#include <linux/list.h>
-#include <linux/kref.h>
+#include <linux/lockdep.h>
 
 #include "i915_active.h"
-#include "i915_request.h"
 #include "i915_syncmap.h"
-#include "i915_utils.h"
-
-struct i915_vma;
-struct i915_timeline_hwsp;
-
-struct i915_timeline {
-	u64 fence_context;
-	u32 seqno;
-
-	spinlock_t lock;
-#define TIMELINE_CLIENT 0 /* default subclass */
-#define TIMELINE_ENGINE 1
-
-	unsigned int pin_count;
-	const u32 *hwsp_seqno;
-	struct i915_vma *hwsp_ggtt;
-	u32 hwsp_offset;
-
-	bool has_initial_breadcrumb;
-
-	/**
-	 * List of breadcrumbs associated with GPU requests currently
-	 * outstanding.
-	 */
-	struct list_head requests;
-
-	/* Contains an RCU guarded pointer to the last request. No reference is
-	 * held to the request, users must carefully acquire a reference to
-	 * the request using i915_active_request_get_request_rcu(), or hold the
-	 * struct_mutex.
-	 */
-	struct i915_active_request last_request;
-
-	/**
-	 * We track the most recent seqno that we wait on in every context so
-	 * that we only have to emit a new await and dependency on a more
-	 * recent sync point. As the contexts may be executed out-of-order, we
-	 * have to track each individually and can not rely on an absolute
-	 * global_seqno. When we know that all tracked fences are completed
-	 * (i.e. when the driver is idle), we know that the syncmap is
-	 * redundant and we can discard it without loss of generality.
-	 */
-	struct i915_syncmap *sync;
-
-	/**
-	 * Barrier provides the ability to serialize ordering between different
-	 * timelines.
-	 *
-	 * Users can call i915_timeline_set_barrier which will make all
-	 * subsequent submissions to this timeline be executed only after the
-	 * barrier has been completed.
-	 */
-	struct i915_active_request barrier;
-
-	struct list_head link;
-	const char *name;
-	struct drm_i915_private *i915;
-
-	struct kref kref;
-};
+#include "i915_timeline_types.h"
 
 int i915_timeline_init(struct drm_i915_private *i915,
 		       struct i915_timeline *tl,
-		       const char *name,
 		       struct i915_vma *hwsp);
 void i915_timeline_fini(struct i915_timeline *tl);
 
@@ -119,7 +57,6 @@ i915_timeline_set_subclass(struct i915_timeline *timeline,
 
 struct i915_timeline *
 i915_timeline_create(struct drm_i915_private *i915,
-		     const char *name,
 		     struct i915_vma *global_hwsp);
 
 static inline struct i915_timeline *
@@ -160,25 +97,17 @@ static inline bool i915_timeline_sync_is_later(struct i915_timeline *tl,
 }
 
 int i915_timeline_pin(struct i915_timeline *tl);
+int i915_timeline_get_seqno(struct i915_timeline *tl,
+			    struct i915_request *rq,
+			    u32 *seqno);
 void i915_timeline_unpin(struct i915_timeline *tl);
 
+int i915_timeline_read_hwsp(struct i915_request *from,
+			    struct i915_request *until,
+			    u32 *hwsp_offset);
+
 void i915_timelines_init(struct drm_i915_private *i915);
 void i915_timelines_park(struct drm_i915_private *i915);
 void i915_timelines_fini(struct drm_i915_private *i915);
 
-/**
- * i915_timeline_set_barrier - orders submission between different timelines
- * @timeline: timeline to set the barrier on
- * @rq: request after which new submissions can proceed
- *
- * Sets the passed in request as the serialization point for all subsequent
- * submissions on @timeline. Subsequent requests will not be submitted to GPU
- * until the barrier has been completed.
- */
-static inline int
-i915_timeline_set_barrier(struct i915_timeline *tl, struct i915_request *rq)
-{
-	return i915_active_request_set(&tl->barrier, rq);
-}
-
 #endif
diff --git a/drivers/gpu/drm/i915/i915_timeline_types.h b/drivers/gpu/drm/i915/i915_timeline_types.h
new file mode 100644
index 000000000000..5256a0b5c5f7
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_timeline_types.h
@@ -0,0 +1,70 @@
+/*
+ * SPDX-License-Identifier: MIT
+ *
+ * Copyright © 2016 Intel Corporation
+ */
+
+#ifndef __I915_TIMELINE_TYPES_H__
+#define __I915_TIMELINE_TYPES_H__
+
+#include <linux/list.h>
+#include <linux/kref.h>
+#include <linux/mutex.h>
+#include <linux/types.h>
+
+#include "i915_active_types.h"
+
+struct drm_i915_private;
+struct i915_vma;
+struct i915_timeline_cacheline;
+struct i915_syncmap;
+
+struct i915_timeline {
+	u64 fence_context;
+	u32 seqno;
+
+	spinlock_t lock;
+#define TIMELINE_CLIENT 0 /* default subclass */
+#define TIMELINE_ENGINE 1
+	struct mutex mutex; /* protects the flow of requests */
+
+	unsigned int pin_count;
+	const u32 *hwsp_seqno;
+	struct i915_vma *hwsp_ggtt;
+	u32 hwsp_offset;
+
+	struct i915_timeline_cacheline *hwsp_cacheline;
+
+	bool has_initial_breadcrumb;
+
+	/**
+	 * List of breadcrumbs associated with GPU requests currently
+	 * outstanding.
+	 */
+	struct list_head requests;
+
+	/* Contains an RCU guarded pointer to the last request. No reference is
+	 * held to the request, users must carefully acquire a reference to
+	 * the request using i915_active_request_get_request_rcu(), or hold the
+	 * struct_mutex.
+	 */
+	struct i915_active_request last_request;
+
+	/**
+	 * We track the most recent seqno that we wait on in every context so
+	 * that we only have to emit a new await and dependency on a more
+	 * recent sync point. As the contexts may be executed out-of-order, we
+	 * have to track each individually and can not rely on an absolute
+	 * global_seqno. When we know that all tracked fences are completed
+	 * (i.e. when the driver is idle), we know that the syncmap is
+	 * redundant and we can discard it without loss of generality.
+	 */
+	struct i915_syncmap *sync;
+
+	struct list_head link;
+	struct drm_i915_private *i915;
+
+	struct kref kref;
+};
+
+#endif /* __I915_TIMELINE_TYPES_H__ */
diff --git a/drivers/gpu/drm/i915/i915_trace.h b/drivers/gpu/drm/i915/i915_trace.h
index eab313c3163c..12893304c8f8 100644
--- a/drivers/gpu/drm/i915/i915_trace.h
+++ b/drivers/gpu/drm/i915/i915_trace.h
@@ -18,6 +18,87 @@
 
 /* watermark/fifo updates */
 
+TRACE_EVENT(intel_pipe_enable,
+	    TP_PROTO(struct drm_i915_private *dev_priv, enum pipe pipe),
+	    TP_ARGS(dev_priv, pipe),
+
+	    TP_STRUCT__entry(
+			     __array(u32, frame, 3)
+			     __array(u32, scanline, 3)
+			     __field(enum pipe, pipe)
+			     ),
+
+	    TP_fast_assign(
+			   enum pipe _pipe;
+			   for_each_pipe(dev_priv, _pipe) {
+				   __entry->frame[_pipe] =
+					   dev_priv->drm.driver->get_vblank_counter(&dev_priv->drm, _pipe);
+				   __entry->scanline[_pipe] =
+					   intel_get_crtc_scanline(intel_get_crtc_for_pipe(dev_priv, _pipe));
+			   }
+			   __entry->pipe = pipe;
+			   ),
+
+	    TP_printk("pipe %c enable, pipe A: frame=%u, scanline=%u, pipe B: frame=%u, scanline=%u, pipe C: frame=%u, scanline=%u",
+		      pipe_name(__entry->pipe),
+		      __entry->frame[PIPE_A], __entry->scanline[PIPE_A],
+		      __entry->frame[PIPE_B], __entry->scanline[PIPE_B],
+		      __entry->frame[PIPE_C], __entry->scanline[PIPE_C])
+);
+
+TRACE_EVENT(intel_pipe_disable,
+	    TP_PROTO(struct drm_i915_private *dev_priv, enum pipe pipe),
+	    TP_ARGS(dev_priv, pipe),
+
+	    TP_STRUCT__entry(
+			     __array(u32, frame, 3)
+			     __array(u32, scanline, 3)
+			     __field(enum pipe, pipe)
+			     ),
+
+	    TP_fast_assign(
+			   enum pipe _pipe;
+			   for_each_pipe(dev_priv, _pipe) {
+				   __entry->frame[_pipe] =
+					   dev_priv->drm.driver->get_vblank_counter(&dev_priv->drm, _pipe);
+				   __entry->scanline[_pipe] =
+					   intel_get_crtc_scanline(intel_get_crtc_for_pipe(dev_priv, _pipe));
+			   }
+			   __entry->pipe = pipe;
+			   ),
+
+	    TP_printk("pipe %c disable, pipe A: frame=%u, scanline=%u, pipe B: frame=%u, scanline=%u, pipe C: frame=%u, scanline=%u",
+		      pipe_name(__entry->pipe),
+		      __entry->frame[PIPE_A], __entry->scanline[PIPE_A],
+		      __entry->frame[PIPE_B], __entry->scanline[PIPE_B],
+		      __entry->frame[PIPE_C], __entry->scanline[PIPE_C])
+);
+
+TRACE_EVENT(intel_pipe_crc,
+	    TP_PROTO(struct intel_crtc *crtc, const u32 *crcs),
+	    TP_ARGS(crtc, crcs),
+
+	    TP_STRUCT__entry(
+			     __field(enum pipe, pipe)
+			     __field(u32, frame)
+			     __field(u32, scanline)
+			     __array(u32, crcs, 5)
+			     ),
+
+	    TP_fast_assign(
+			   __entry->pipe = crtc->pipe;
+			   __entry->frame = crtc->base.dev->driver->get_vblank_counter(crtc->base.dev,
+										       crtc->pipe);
+			   __entry->scanline = intel_get_crtc_scanline(crtc);
+			   memcpy(__entry->crcs, crcs, sizeof(__entry->crcs));
+			   ),
+
+	    TP_printk("pipe %c, frame=%u, scanline=%u crc=%08x %08x %08x %08x %08x",
+		      pipe_name(__entry->pipe), __entry->frame, __entry->scanline,
+		      __entry->crcs[0], __entry->crcs[1], __entry->crcs[2],
+		      __entry->crcs[3], __entry->crcs[4])
+);
+
 TRACE_EVENT(intel_cpu_fifo_underrun,
 	    TP_PROTO(struct drm_i915_private *dev_priv, enum pipe pipe),
 	    TP_ARGS(dev_priv, pipe),
@@ -627,7 +708,6 @@ DECLARE_EVENT_CLASS(i915_request,
 			     __field(u16, class)
 			     __field(u16, instance)
 			     __field(u32, seqno)
-			     __field(u32, global)
 			     ),
 
 	    TP_fast_assign(
@@ -637,13 +717,11 @@ DECLARE_EVENT_CLASS(i915_request,
 			   __entry->instance = rq->engine->instance;
 			   __entry->ctx = rq->fence.context;
 			   __entry->seqno = rq->fence.seqno;
-			   __entry->global = rq->global_seqno;
 			   ),
 
-	    TP_printk("dev=%u, engine=%u:%u, hw_id=%u, ctx=%llu, seqno=%u, global=%u",
+	    TP_printk("dev=%u, engine=%u:%u, hw_id=%u, ctx=%llu, seqno=%u",
 		      __entry->dev, __entry->class, __entry->instance,
-		      __entry->hw_id, __entry->ctx, __entry->seqno,
-		      __entry->global)
+		      __entry->hw_id, __entry->ctx, __entry->seqno)
 );
 
 DEFINE_EVENT(i915_request, i915_request_add,
@@ -673,7 +751,6 @@ TRACE_EVENT(i915_request_in,
 			     __field(u16, class)
 			     __field(u16, instance)
 			     __field(u32, seqno)
-			     __field(u32, global_seqno)
 			     __field(u32, port)
 			     __field(u32, prio)
 			    ),
@@ -685,15 +762,14 @@ TRACE_EVENT(i915_request_in,
 			   __entry->instance = rq->engine->instance;
 			   __entry->ctx = rq->fence.context;
 			   __entry->seqno = rq->fence.seqno;
-			   __entry->global_seqno = rq->global_seqno;
 			   __entry->prio = rq->sched.attr.priority;
 			   __entry->port = port;
 			   ),
 
-	    TP_printk("dev=%u, engine=%u:%u, hw_id=%u, ctx=%llu, seqno=%u, prio=%u, global=%u, port=%u",
+	    TP_printk("dev=%u, engine=%u:%u, hw_id=%u, ctx=%llu, seqno=%u, prio=%u, port=%u",
 		      __entry->dev, __entry->class, __entry->instance,
 		      __entry->hw_id, __entry->ctx, __entry->seqno,
-		      __entry->prio, __entry->global_seqno, __entry->port)
+		      __entry->prio, __entry->port)
 );
 
 TRACE_EVENT(i915_request_out,
@@ -707,7 +783,6 @@ TRACE_EVENT(i915_request_out,
 			     __field(u16, class)
 			     __field(u16, instance)
 			     __field(u32, seqno)
-			     __field(u32, global_seqno)
 			     __field(u32, completed)
 			    ),
 
@@ -718,14 +793,13 @@ TRACE_EVENT(i915_request_out,
 			   __entry->instance = rq->engine->instance;
 			   __entry->ctx = rq->fence.context;
 			   __entry->seqno = rq->fence.seqno;
-			   __entry->global_seqno = rq->global_seqno;
 			   __entry->completed = i915_request_completed(rq);
 			   ),
 
-		    TP_printk("dev=%u, engine=%u:%u, hw_id=%u, ctx=%llu, seqno=%u, global=%u, completed?=%u",
+		    TP_printk("dev=%u, engine=%u:%u, hw_id=%u, ctx=%llu, seqno=%u, completed?=%u",
 			      __entry->dev, __entry->class, __entry->instance,
 			      __entry->hw_id, __entry->ctx, __entry->seqno,
-			      __entry->global_seqno, __entry->completed)
+			      __entry->completed)
 );
 
 #else
@@ -768,7 +842,6 @@ TRACE_EVENT(i915_request_wait_begin,
 			     __field(u16, class)
 			     __field(u16, instance)
 			     __field(u32, seqno)
-			     __field(u32, global)
 			     __field(unsigned int, flags)
 			     ),
 
@@ -785,14 +858,13 @@ TRACE_EVENT(i915_request_wait_begin,
 			   __entry->instance = rq->engine->instance;
 			   __entry->ctx = rq->fence.context;
 			   __entry->seqno = rq->fence.seqno;
-			   __entry->global = rq->global_seqno;
 			   __entry->flags = flags;
 			   ),
 
-	    TP_printk("dev=%u, engine=%u:%u, hw_id=%u, ctx=%llu, seqno=%u, global=%u, blocking=%u, flags=0x%x",
+	    TP_printk("dev=%u, engine=%u:%u, hw_id=%u, ctx=%llu, seqno=%u, blocking=%u, flags=0x%x",
 		      __entry->dev, __entry->class, __entry->instance,
 		      __entry->hw_id, __entry->ctx, __entry->seqno,
-		      __entry->global, !!(__entry->flags & I915_WAIT_LOCKED),
+		      !!(__entry->flags & I915_WAIT_LOCKED),
 		      __entry->flags)
 );
 
diff --git a/drivers/gpu/drm/i915/i915_user_extensions.c b/drivers/gpu/drm/i915/i915_user_extensions.c
new file mode 100644
index 000000000000..c822d0aafd2d
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_user_extensions.c
@@ -0,0 +1,61 @@
+/*
+ * SPDX-License-Identifier: MIT
+ *
+ * Copyright © 2018 Intel Corporation
+ */
+
+#include <linux/nospec.h>
+#include <linux/sched/signal.h>
+#include <linux/uaccess.h>
+
+#include <uapi/drm/i915_drm.h>
+
+#include "i915_user_extensions.h"
+#include "i915_utils.h"
+
+int i915_user_extensions(struct i915_user_extension __user *ext,
+			 const i915_user_extension_fn *tbl,
+			 unsigned int count,
+			 void *data)
+{
+	unsigned int stackdepth = 512;
+
+	while (ext) {
+		int i, err;
+		u32 name;
+		u64 next;
+
+		if (!stackdepth--) /* recursion vs useful flexibility */
+			return -E2BIG;
+
+		err = check_user_mbz(&ext->flags);
+		if (err)
+			return err;
+
+		for (i = 0; i < ARRAY_SIZE(ext->rsvd); i++) {
+			err = check_user_mbz(&ext->rsvd[i]);
+			if (err)
+				return err;
+		}
+
+		if (get_user(name, &ext->name))
+			return -EFAULT;
+
+		err = -EINVAL;
+		if (name < count) {
+			name = array_index_nospec(name, count);
+			if (tbl[name])
+				err = tbl[name](ext, data);
+		}
+		if (err)
+			return err;
+
+		if (get_user(next, &ext->next_extension) ||
+		    overflows_type(next, ext))
+			return -EFAULT;
+
+		ext = u64_to_user_ptr(next);
+	}
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/i915/i915_user_extensions.h b/drivers/gpu/drm/i915/i915_user_extensions.h
new file mode 100644
index 000000000000..a14bf6bba9a1
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_user_extensions.h
@@ -0,0 +1,20 @@
+/*
+ * SPDX-License-Identifier: MIT
+ *
+ * Copyright © 2018 Intel Corporation
+ */
+
+#ifndef I915_USER_EXTENSIONS_H
+#define I915_USER_EXTENSIONS_H
+
+struct i915_user_extension;
+
+typedef int (*i915_user_extension_fn)(struct i915_user_extension __user *ext,
+				      void *data);
+
+int i915_user_extensions(struct i915_user_extension __user *ext,
+			 const i915_user_extension_fn *tbl,
+			 unsigned int count,
+			 void *data);
+
+#endif /* I915_USER_EXTENSIONS_H */
diff --git a/drivers/gpu/drm/i915/i915_utils.h b/drivers/gpu/drm/i915/i915_utils.h
index 540e20eb032c..2dbe8933b50a 100644
--- a/drivers/gpu/drm/i915/i915_utils.h
+++ b/drivers/gpu/drm/i915/i915_utils.h
@@ -105,6 +105,37 @@
 	__T;								\
 })
 
+/*
+ * container_of_user: Extract the superclass from a pointer to a member.
+ *
+ * Exactly like container_of() with the exception that it plays nicely
+ * with sparse for __user @ptr.
+ */
+#define container_of_user(ptr, type, member) ({				\
+	void __user *__mptr = (void __user *)(ptr);			\
+	BUILD_BUG_ON_MSG(!__same_type(*(ptr), ((type *)0)->member) &&	\
+			 !__same_type(*(ptr), void),			\
+			 "pointer type mismatch in container_of()");	\
+	((type __user *)(__mptr - offsetof(type, member))); })
+
+/*
+ * check_user_mbz: Check that a user value exists and is zero
+ *
+ * Frequently in our uABI we reserve space for future extensions, and
+ * two ensure that userspace is prepared we enforce that space must
+ * be zero. (Then any future extension can safely assume a default value
+ * of 0.)
+ *
+ * check_user_mbz() combines checking that the user pointer is accessible
+ * and that the contained value is zero.
+ *
+ * Returns: -EFAULT if not accessible, -EINVAL if !zero, or 0 on success.
+ */
+#define check_user_mbz(U) ({						\
+	typeof(*(U)) mbz__;						\
+	get_user(mbz__, (U)) ? -EFAULT : mbz__ ? -EINVAL : 0;		\
+})
+
 static inline u64 ptr_to_u64(const void *ptr)
 {
 	return (uintptr_t)ptr;
diff --git a/drivers/gpu/drm/i915/i915_vgpu.c b/drivers/gpu/drm/i915/i915_vgpu.c
index 869cf4a3b6de..94d3992b599d 100644
--- a/drivers/gpu/drm/i915/i915_vgpu.c
+++ b/drivers/gpu/drm/i915/i915_vgpu.c
@@ -60,30 +60,31 @@
  */
 void i915_check_vgpu(struct drm_i915_private *dev_priv)
 {
+	struct intel_uncore *uncore = &dev_priv->uncore;
 	u64 magic;
 	u16 version_major;
 
 	BUILD_BUG_ON(sizeof(struct vgt_if) != VGT_PVINFO_SIZE);
 
-	magic = __raw_i915_read64(dev_priv, vgtif_reg(magic));
+	magic = __raw_uncore_read64(uncore, vgtif_reg(magic));
 	if (magic != VGT_MAGIC)
 		return;
 
-	version_major = __raw_i915_read16(dev_priv, vgtif_reg(version_major));
+	version_major = __raw_uncore_read16(uncore, vgtif_reg(version_major));
 	if (version_major < VGT_VERSION_MAJOR) {
 		DRM_INFO("VGT interface version mismatch!\n");
 		return;
 	}
 
-	dev_priv->vgpu.caps = __raw_i915_read32(dev_priv, vgtif_reg(vgt_caps));
+	dev_priv->vgpu.caps = __raw_uncore_read32(uncore, vgtif_reg(vgt_caps));
 
 	dev_priv->vgpu.active = true;
 	DRM_INFO("Virtual GPU for Intel GVT-g detected.\n");
 }
 
-bool intel_vgpu_has_full_48bit_ppgtt(struct drm_i915_private *dev_priv)
+bool intel_vgpu_has_full_ppgtt(struct drm_i915_private *dev_priv)
 {
-	return dev_priv->vgpu.caps & VGT_CAPS_FULL_48BIT_PPGTT;
+	return dev_priv->vgpu.caps & VGT_CAPS_FULL_PPGTT;
 }
 
 struct _balloon_info_ {
diff --git a/drivers/gpu/drm/i915/i915_vgpu.h b/drivers/gpu/drm/i915/i915_vgpu.h
index 551acc390046..ebe1b7bced98 100644
--- a/drivers/gpu/drm/i915/i915_vgpu.h
+++ b/drivers/gpu/drm/i915/i915_vgpu.h
@@ -28,7 +28,7 @@
 
 void i915_check_vgpu(struct drm_i915_private *dev_priv);
 
-bool intel_vgpu_has_full_48bit_ppgtt(struct drm_i915_private *dev_priv);
+bool intel_vgpu_has_full_ppgtt(struct drm_i915_private *dev_priv);
 
 static inline bool
 intel_vgpu_has_hwsp_emulation(struct drm_i915_private *dev_priv)
diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c
index 41b5bcb803cb..961268f66c63 100644
--- a/drivers/gpu/drm/i915/i915_vma.c
+++ b/drivers/gpu/drm/i915/i915_vma.c
@@ -25,11 +25,27 @@
 #include "i915_vma.h"
 
 #include "i915_drv.h"
+#include "i915_globals.h"
 #include "intel_ringbuffer.h"
 #include "intel_frontbuffer.h"
 
 #include <drm/drm_gem.h>
 
+static struct i915_global_vma {
+	struct i915_global base;
+	struct kmem_cache *slab_vmas;
+} global;
+
+struct i915_vma *i915_vma_alloc(void)
+{
+	return kmem_cache_zalloc(global.slab_vmas, GFP_KERNEL);
+}
+
+void i915_vma_free(struct i915_vma *vma)
+{
+	return kmem_cache_free(global.slab_vmas, vma);
+}
+
 #if IS_ENABLED(CONFIG_DRM_I915_ERRLOG_GEM) && IS_ENABLED(CONFIG_DRM_DEBUG_MM)
 
 #include <linux/stackdepot.h>
@@ -112,7 +128,7 @@ vma_create(struct drm_i915_gem_object *obj,
 	/* The aliasing_ppgtt should never be used directly! */
 	GEM_BUG_ON(vm == &vm->i915->mm.aliasing_ppgtt->vm);
 
-	vma = kmem_cache_zalloc(vm->i915->vmas, GFP_KERNEL);
+	vma = i915_vma_alloc();
 	if (vma == NULL)
 		return ERR_PTR(-ENOMEM);
 
@@ -187,7 +203,7 @@ vma_create(struct drm_i915_gem_object *obj,
 		cmp = i915_vma_compare(pos, vm, view);
 		if (cmp == 0) {
 			spin_unlock(&obj->vma.lock);
-			kmem_cache_free(vm->i915->vmas, vma);
+			i915_vma_free(vma);
 			return pos;
 		}
 
@@ -219,7 +235,7 @@ vma_create(struct drm_i915_gem_object *obj,
 	return vma;
 
 err_vma:
-	kmem_cache_free(vm->i915->vmas, vma);
+	i915_vma_free(vma);
 	return ERR_PTR(-E2BIG);
 }
 
@@ -800,8 +816,6 @@ void i915_vma_reopen(struct i915_vma *vma)
 
 static void __i915_vma_destroy(struct i915_vma *vma)
 {
-	struct drm_i915_private *i915 = vma->vm->i915;
-
 	GEM_BUG_ON(vma->node.allocated);
 	GEM_BUG_ON(vma->fence);
 
@@ -822,7 +836,7 @@ static void __i915_vma_destroy(struct i915_vma *vma)
 
 	i915_active_fini(&vma->active);
 
-	kmem_cache_free(i915->vmas, vma);
+	i915_vma_free(vma);
 }
 
 void i915_vma_destroy(struct i915_vma *vma)
@@ -1038,3 +1052,28 @@ unpin:
 #if IS_ENABLED(CONFIG_DRM_I915_SELFTEST)
 #include "selftests/i915_vma.c"
 #endif
+
+static void i915_global_vma_shrink(void)
+{
+	kmem_cache_shrink(global.slab_vmas);
+}
+
+static void i915_global_vma_exit(void)
+{
+	kmem_cache_destroy(global.slab_vmas);
+}
+
+static struct i915_global_vma global = { {
+	.shrink = i915_global_vma_shrink,
+	.exit = i915_global_vma_exit,
+} };
+
+int __init i915_global_vma_init(void)
+{
+	global.slab_vmas = KMEM_CACHE(i915_vma, SLAB_HWCACHE_ALIGN);
+	if (!global.slab_vmas)
+		return -ENOMEM;
+
+	i915_global_register(&global.base);
+	return 0;
+}
diff --git a/drivers/gpu/drm/i915/i915_vma.h b/drivers/gpu/drm/i915/i915_vma.h
index 7c742027f866..6eab70953a57 100644
--- a/drivers/gpu/drm/i915/i915_vma.h
+++ b/drivers/gpu/drm/i915/i915_vma.h
@@ -440,4 +440,7 @@ void i915_vma_parked(struct drm_i915_private *i915);
 	list_for_each_entry(V, &(OBJ)->vma.list, obj_link)		\
 		for_each_until(!i915_vma_is_ggtt(V))
 
+struct i915_vma *i915_vma_alloc(void);
+void i915_vma_free(struct i915_vma *vma);
+
 #endif
diff --git a/drivers/gpu/drm/i915/icl_dsi.c b/drivers/gpu/drm/i915/icl_dsi.c
index 641e0778fa9c..9d962ea1e635 100644
--- a/drivers/gpu/drm/i915/icl_dsi.c
+++ b/drivers/gpu/drm/i915/icl_dsi.c
@@ -25,9 +25,13 @@
  *   Jani Nikula <jani.nikula@intel.com>
  */
 
-#include <drm/drm_mipi_dsi.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_mipi_dsi.h>
+
+#include "intel_connector.h"
+#include "intel_ddi.h"
 #include "intel_dsi.h"
+#include "intel_panel.h"
 
 static inline int header_credits_available(struct drm_i915_private *dev_priv,
 					   enum transcoder dsi_trans)
@@ -246,13 +250,13 @@ static void dsi_program_swing_and_deemphasis(struct intel_encoder *encoder)
 
 		for (lane = 0; lane <= 3; lane++) {
 			/* Bspec: must not use GRP register for write */
-			tmp = I915_READ(ICL_PORT_TX_DW4_LN(port, lane));
+			tmp = I915_READ(ICL_PORT_TX_DW4_LN(lane, port));
 			tmp &= ~(POST_CURSOR_1_MASK | POST_CURSOR_2_MASK |
 				 CURSOR_COEFF_MASK);
 			tmp |= POST_CURSOR_1(0x0);
 			tmp |= POST_CURSOR_2(0x0);
 			tmp |= CURSOR_COEFF(0x3f);
-			I915_WRITE(ICL_PORT_TX_DW4_LN(port, lane), tmp);
+			I915_WRITE(ICL_PORT_TX_DW4_LN(lane, port), tmp);
 		}
 	}
 }
@@ -399,11 +403,11 @@ static void gen11_dsi_config_phy_lanes_sequence(struct intel_encoder *encoder)
 		tmp &= ~LOADGEN_SELECT;
 		I915_WRITE(ICL_PORT_TX_DW4_AUX(port), tmp);
 		for (lane = 0; lane <= 3; lane++) {
-			tmp = I915_READ(ICL_PORT_TX_DW4_LN(port, lane));
+			tmp = I915_READ(ICL_PORT_TX_DW4_LN(lane, port));
 			tmp &= ~LOADGEN_SELECT;
 			if (lane != 2)
 				tmp |= LOADGEN_SELECT;
-			I915_WRITE(ICL_PORT_TX_DW4_LN(port, lane), tmp);
+			I915_WRITE(ICL_PORT_TX_DW4_LN(lane, port), tmp);
 		}
 	}
 
@@ -876,7 +880,8 @@ static void gen11_dsi_enable_transcoder(struct intel_encoder *encoder)
 		I915_WRITE(PIPECONF(dsi_trans), tmp);
 
 		/* wait for transcoder to be enabled */
-		if (intel_wait_for_register(dev_priv, PIPECONF(dsi_trans),
+		if (intel_wait_for_register(&dev_priv->uncore,
+					    PIPECONF(dsi_trans),
 					    I965_PIPECONF_ACTIVE,
 					    I965_PIPECONF_ACTIVE, 10))
 			DRM_ERROR("DSI transcoder not enabled\n");
@@ -1054,7 +1059,8 @@ static void gen11_dsi_disable_transcoder(struct intel_encoder *encoder)
 		I915_WRITE(PIPECONF(dsi_trans), tmp);
 
 		/* wait for transcoder to be disabled */
-		if (intel_wait_for_register(dev_priv, PIPECONF(dsi_trans),
+		if (intel_wait_for_register(&dev_priv->uncore,
+					    PIPECONF(dsi_trans),
 					    I965_PIPECONF_ACTIVE, 0, 50))
 			DRM_ERROR("DSI trancoder not disabled\n");
 	}
@@ -1146,13 +1152,11 @@ static void gen11_dsi_disable_io_power(struct intel_encoder *encoder)
 		intel_wakeref_t wakeref;
 
 		wakeref = fetch_and_zero(&intel_dsi->io_wakeref[port]);
-		if (wakeref) {
-			intel_display_power_put(dev_priv,
-						port == PORT_A ?
-						POWER_DOMAIN_PORT_DDI_A_IO :
-						POWER_DOMAIN_PORT_DDI_B_IO,
-						wakeref);
-		}
+		intel_display_power_put(dev_priv,
+					port == PORT_A ?
+					POWER_DOMAIN_PORT_DDI_A_IO :
+					POWER_DOMAIN_PORT_DDI_B_IO,
+					wakeref);
 	}
 
 	/* set mode to DDI */
@@ -1194,11 +1198,10 @@ static void gen11_dsi_get_config(struct intel_encoder *encoder,
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
-	u32 pll_id;
 
 	/* FIXME: adapt icl_ddi_clock_get() for DSI and use that? */
-	pll_id = intel_get_shared_dpll_id(dev_priv, pipe_config->shared_dpll);
-	pipe_config->port_clock = cnl_calc_wrpll_link(dev_priv, pll_id);
+	pipe_config->port_clock =
+		cnl_calc_wrpll_link(dev_priv, &pipe_config->dpll_hw_state);
 	pipe_config->base.adjusted_mode.crtc_clock = intel_dsi->pclk;
 	pipe_config->output_types |= BIT(INTEL_OUTPUT_DSI);
 }
@@ -1367,7 +1370,7 @@ void icl_dsi_init(struct drm_i915_private *dev_priv)
 	struct intel_encoder *encoder;
 	struct intel_connector *intel_connector;
 	struct drm_connector *connector;
-	struct drm_display_mode *scan, *fixed_mode = NULL;
+	struct drm_display_mode *fixed_mode;
 	enum port port;
 
 	if (!intel_bios_is_dsi_present(dev_priv, &port))
@@ -1417,15 +1420,8 @@ void icl_dsi_init(struct drm_i915_private *dev_priv)
 	/* attach connector to encoder */
 	intel_connector_attach_encoder(intel_connector, encoder);
 
-	/* fill mode info from VBT */
 	mutex_lock(&dev->mode_config.mutex);
-	intel_dsi_vbt_get_modes(intel_dsi);
-	list_for_each_entry(scan, &connector->probed_modes, head) {
-		if (scan->type & DRM_MODE_TYPE_PREFERRED) {
-			fixed_mode = drm_mode_duplicate(dev, scan);
-			break;
-		}
-	}
+	fixed_mode = intel_panel_vbt_fixed_mode(intel_connector);
 	mutex_unlock(&dev->mode_config.mutex);
 
 	if (!fixed_mode) {
@@ -1433,12 +1429,9 @@ void icl_dsi_init(struct drm_i915_private *dev_priv)
 		goto err;
 	}
 
-	connector->display_info.width_mm = fixed_mode->width_mm;
-	connector->display_info.height_mm = fixed_mode->height_mm;
 	intel_panel_init(&intel_connector->panel, fixed_mode, NULL);
 	intel_panel_setup_backlight(connector, INVALID_PIPE);
 
-
 	if (dev_priv->vbt.dsi.config->dual_link)
 		intel_dsi->ports = BIT(PORT_A) | BIT(PORT_B);
 	else
diff --git a/drivers/gpu/drm/i915/intel_atomic.c b/drivers/gpu/drm/i915/intel_atomic.c
index 7cf9290ea34a..8c8fae32ec50 100644
--- a/drivers/gpu/drm/i915/intel_atomic.c
+++ b/drivers/gpu/drm/i915/intel_atomic.c
@@ -35,6 +35,8 @@
 #include <drm/drm_plane_helper.h>
 
 #include "intel_drv.h"
+#include "intel_hdcp.h"
+#include "intel_sprite.h"
 
 /**
  * intel_digital_connector_atomic_get_property - hook for connector->atomic_get_property.
@@ -126,6 +128,7 @@ int intel_digital_connector_atomic_check(struct drm_connector *conn,
 	 */
 	if (new_conn_state->force_audio != old_conn_state->force_audio ||
 	    new_conn_state->broadcast_rgb != old_conn_state->broadcast_rgb ||
+	    new_conn_state->base.colorspace != old_conn_state->base.colorspace ||
 	    new_conn_state->base.picture_aspect_ratio != old_conn_state->base.picture_aspect_ratio ||
 	    new_conn_state->base.content_type != old_conn_state->base.content_type ||
 	    new_conn_state->base.scaling_mode != old_conn_state->base.scaling_mode)
@@ -234,10 +237,11 @@ static void intel_atomic_setup_scaler(struct intel_crtc_scaler_state *scaler_sta
 	if (plane_state && plane_state->base.fb &&
 	    plane_state->base.fb->format->is_yuv &&
 	    plane_state->base.fb->format->num_planes > 1) {
+		struct intel_plane *plane = to_intel_plane(plane_state->base.plane);
 		if (IS_GEN(dev_priv, 9) &&
 		    !IS_GEMINILAKE(dev_priv)) {
 			mode = SKL_PS_SCALER_MODE_NV12;
-		} else if (icl_is_hdr_plane(to_intel_plane(plane_state->base.plane))) {
+		} else if (icl_is_hdr_plane(dev_priv, plane->id)) {
 			/*
 			 * On gen11+'s HDR planes we only use the scaler for
 			 * scaling. They have a dedicated chroma upsampler, so
diff --git a/drivers/gpu/drm/i915/intel_atomic_plane.c b/drivers/gpu/drm/i915/intel_atomic_plane.c
index db0965904439..d11681d71add 100644
--- a/drivers/gpu/drm/i915/intel_atomic_plane.c
+++ b/drivers/gpu/drm/i915/intel_atomic_plane.c
@@ -35,7 +35,10 @@
 #include <drm/drm_fourcc.h>
 #include <drm/drm_plane_helper.h>
 
+#include "intel_atomic_plane.h"
 #include "intel_drv.h"
+#include "intel_pm.h"
+#include "intel_sprite.h"
 
 struct intel_plane *intel_plane_alloc(void)
 {
@@ -121,6 +124,7 @@ int intel_plane_atomic_check_with_state(const struct intel_crtc_state *old_crtc_
 
 	new_crtc_state->active_planes &= ~BIT(plane->id);
 	new_crtc_state->nv12_planes &= ~BIT(plane->id);
+	new_crtc_state->c8_planes &= ~BIT(plane->id);
 	new_plane_state->base.visible = false;
 
 	if (!new_plane_state->base.crtc && !old_plane_state->base.crtc)
@@ -135,9 +139,13 @@ int intel_plane_atomic_check_with_state(const struct intel_crtc_state *old_crtc_
 		new_crtc_state->active_planes |= BIT(plane->id);
 
 	if (new_plane_state->base.visible &&
-	    new_plane_state->base.fb->format->format == DRM_FORMAT_NV12)
+	    is_planar_yuv_format(new_plane_state->base.fb->format->format))
 		new_crtc_state->nv12_planes |= BIT(plane->id);
 
+	if (new_plane_state->base.visible &&
+	    new_plane_state->base.fb->format->format == DRM_FORMAT_C8)
+		new_crtc_state->c8_planes |= BIT(plane->id);
+
 	if (new_plane_state->base.visible || old_plane_state->base.visible)
 		new_crtc_state->update_planes |= BIT(plane->id);
 
@@ -214,6 +222,35 @@ skl_next_plane_to_commit(struct intel_atomic_state *state,
 	return NULL;
 }
 
+void intel_update_plane(struct intel_plane *plane,
+			const struct intel_crtc_state *crtc_state,
+			const struct intel_plane_state *plane_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
+
+	trace_intel_update_plane(&plane->base, crtc);
+	plane->update_plane(plane, crtc_state, plane_state);
+}
+
+void intel_update_slave(struct intel_plane *plane,
+			const struct intel_crtc_state *crtc_state,
+			const struct intel_plane_state *plane_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
+
+	trace_intel_update_plane(&plane->base, crtc);
+	plane->update_slave(plane, crtc_state, plane_state);
+}
+
+void intel_disable_plane(struct intel_plane *plane,
+			 const struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
+
+	trace_intel_disable_plane(&plane->base, crtc);
+	plane->disable_plane(plane, crtc_state);
+}
+
 void skl_update_planes_on_crtc(struct intel_atomic_state *state,
 			       struct intel_crtc *crtc)
 {
@@ -238,8 +275,7 @@ void skl_update_planes_on_crtc(struct intel_atomic_state *state,
 			intel_atomic_get_new_plane_state(state, plane);
 
 		if (new_plane_state->base.visible) {
-			trace_intel_update_plane(&plane->base, crtc);
-			plane->update_plane(plane, new_crtc_state, new_plane_state);
+			intel_update_plane(plane, new_crtc_state, new_plane_state);
 		} else if (new_plane_state->slave) {
 			struct intel_plane *master =
 				new_plane_state->linked_plane;
@@ -256,11 +292,9 @@ void skl_update_planes_on_crtc(struct intel_atomic_state *state,
 			new_plane_state =
 				intel_atomic_get_new_plane_state(state, master);
 
-			trace_intel_update_plane(&plane->base, crtc);
-			plane->update_slave(plane, new_crtc_state, new_plane_state);
+			intel_update_slave(plane, new_crtc_state, new_plane_state);
 		} else {
-			trace_intel_disable_plane(&plane->base, crtc);
-			plane->disable_plane(plane, new_crtc_state);
+			intel_disable_plane(plane, new_crtc_state);
 		}
 	}
 }
@@ -280,13 +314,10 @@ void i9xx_update_planes_on_crtc(struct intel_atomic_state *state,
 		    !(update_mask & BIT(plane->id)))
 			continue;
 
-		if (new_plane_state->base.visible) {
-			trace_intel_update_plane(&plane->base, crtc);
-			plane->update_plane(plane, new_crtc_state, new_plane_state);
-		} else {
-			trace_intel_disable_plane(&plane->base, crtc);
-			plane->disable_plane(plane, new_crtc_state);
-		}
+		if (new_plane_state->base.visible)
+			intel_update_plane(plane, new_crtc_state, new_plane_state);
+		else
+			intel_disable_plane(plane, new_crtc_state);
 	}
 }
 
diff --git a/drivers/gpu/drm/i915/intel_atomic_plane.h b/drivers/gpu/drm/i915/intel_atomic_plane.h
new file mode 100644
index 000000000000..14678620440f
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_atomic_plane.h
@@ -0,0 +1,40 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_ATOMIC_PLANE_H__
+#define __INTEL_ATOMIC_PLANE_H__
+
+struct drm_plane;
+struct intel_atomic_state;
+struct intel_crtc;
+struct intel_crtc_state;
+struct intel_plane;
+struct intel_plane_state;
+
+extern const struct drm_plane_helper_funcs intel_plane_helper_funcs;
+
+void intel_update_plane(struct intel_plane *plane,
+			const struct intel_crtc_state *crtc_state,
+			const struct intel_plane_state *plane_state);
+void intel_update_slave(struct intel_plane *plane,
+			const struct intel_crtc_state *crtc_state,
+			const struct intel_plane_state *plane_state);
+void intel_disable_plane(struct intel_plane *plane,
+			 const struct intel_crtc_state *crtc_state);
+struct intel_plane *intel_plane_alloc(void);
+void intel_plane_free(struct intel_plane *plane);
+struct drm_plane_state *intel_plane_duplicate_state(struct drm_plane *plane);
+void intel_plane_destroy_state(struct drm_plane *plane,
+			       struct drm_plane_state *state);
+void skl_update_planes_on_crtc(struct intel_atomic_state *state,
+			       struct intel_crtc *crtc);
+void i9xx_update_planes_on_crtc(struct intel_atomic_state *state,
+				struct intel_crtc *crtc);
+int intel_plane_atomic_check_with_state(const struct intel_crtc_state *old_crtc_state,
+					struct intel_crtc_state *crtc_state,
+					const struct intel_plane_state *old_plane_state,
+					struct intel_plane_state *intel_state);
+
+#endif /* __INTEL_ATOMIC_PLANE_H__ */
diff --git a/drivers/gpu/drm/i915/intel_audio.c b/drivers/gpu/drm/i915/intel_audio.c
index 5104c6bbd66f..bca4cc025d3d 100644
--- a/drivers/gpu/drm/i915/intel_audio.c
+++ b/drivers/gpu/drm/i915/intel_audio.c
@@ -21,14 +21,16 @@
  * DEALINGS IN THE SOFTWARE.
  */
 
-#include <linux/kernel.h>
 #include <linux/component.h>
+#include <linux/kernel.h>
+
+#include <drm/drm_edid.h>
 #include <drm/i915_component.h>
 #include <drm/intel_lpe_audio.h>
-#include "intel_drv.h"
 
-#include <drm/drm_edid.h>
 #include "i915_drv.h"
+#include "intel_audio.h"
+#include "intel_drv.h"
 
 /**
  * DOC: High Definition Audio over HDMI and Display Port
@@ -741,27 +743,91 @@ void intel_init_audio_hooks(struct drm_i915_private *dev_priv)
 	}
 }
 
-static void i915_audio_component_get_power(struct device *kdev)
+static void glk_force_audio_cdclk(struct drm_i915_private *dev_priv,
+				  bool enable)
 {
-	intel_display_power_get(kdev_to_i915(kdev), POWER_DOMAIN_AUDIO);
+	struct drm_modeset_acquire_ctx ctx;
+	struct drm_atomic_state *state;
+	int ret;
+
+	drm_modeset_acquire_init(&ctx, 0);
+	state = drm_atomic_state_alloc(&dev_priv->drm);
+	if (WARN_ON(!state))
+		return;
+
+	state->acquire_ctx = &ctx;
+
+retry:
+	to_intel_atomic_state(state)->cdclk.force_min_cdclk_changed = true;
+	to_intel_atomic_state(state)->cdclk.force_min_cdclk =
+		enable ? 2 * 96000 : 0;
+
+	/*
+	 * Protects dev_priv->cdclk.force_min_cdclk
+	 * Need to lock this here in case we have no active pipes
+	 * and thus wouldn't lock it during the commit otherwise.
+	 */
+	ret = drm_modeset_lock(&dev_priv->drm.mode_config.connection_mutex,
+			       &ctx);
+	if (!ret)
+		ret = drm_atomic_commit(state);
+
+	if (ret == -EDEADLK) {
+		drm_atomic_state_clear(state);
+		drm_modeset_backoff(&ctx);
+		goto retry;
+	}
+
+	WARN_ON(ret);
+
+	drm_atomic_state_put(state);
+
+	drm_modeset_drop_locks(&ctx);
+	drm_modeset_acquire_fini(&ctx);
 }
 
-static void i915_audio_component_put_power(struct device *kdev)
+static unsigned long i915_audio_component_get_power(struct device *kdev)
 {
-	intel_display_power_put_unchecked(kdev_to_i915(kdev),
-					  POWER_DOMAIN_AUDIO);
+	struct drm_i915_private *dev_priv = kdev_to_i915(kdev);
+	intel_wakeref_t ret;
+
+	/* Catch potential impedance mismatches before they occur! */
+	BUILD_BUG_ON(sizeof(intel_wakeref_t) > sizeof(unsigned long));
+
+	ret = intel_display_power_get(dev_priv, POWER_DOMAIN_AUDIO);
+
+	/* Force CDCLK to 2*BCLK as long as we need audio to be powered. */
+	if (dev_priv->audio_power_refcount++ == 0)
+		if (IS_CANNONLAKE(dev_priv) || IS_GEMINILAKE(dev_priv))
+			glk_force_audio_cdclk(dev_priv, true);
+
+	return ret;
+}
+
+static void i915_audio_component_put_power(struct device *kdev,
+					   unsigned long cookie)
+{
+	struct drm_i915_private *dev_priv = kdev_to_i915(kdev);
+
+	/* Stop forcing CDCLK to 2*BCLK if no need for audio to be powered. */
+	if (--dev_priv->audio_power_refcount == 0)
+		if (IS_CANNONLAKE(dev_priv) || IS_GEMINILAKE(dev_priv))
+			glk_force_audio_cdclk(dev_priv, false);
+
+	intel_display_power_put(dev_priv, POWER_DOMAIN_AUDIO, cookie);
 }
 
 static void i915_audio_component_codec_wake_override(struct device *kdev,
 						     bool enable)
 {
 	struct drm_i915_private *dev_priv = kdev_to_i915(kdev);
+	unsigned long cookie;
 	u32 tmp;
 
 	if (!IS_GEN(dev_priv, 9))
 		return;
 
-	i915_audio_component_get_power(kdev);
+	cookie = i915_audio_component_get_power(kdev);
 
 	/*
 	 * Enable/disable generating the codec wake signal, overriding the
@@ -779,7 +845,7 @@ static void i915_audio_component_codec_wake_override(struct device *kdev,
 		usleep_range(1000, 1500);
 	}
 
-	i915_audio_component_put_power(kdev);
+	i915_audio_component_put_power(kdev, cookie);
 }
 
 /* Get CDCLK in kHz  */
@@ -850,12 +916,13 @@ static int i915_audio_component_sync_audio_rate(struct device *kdev, int port,
 	struct i915_audio_component *acomp = dev_priv->audio_component;
 	struct intel_encoder *encoder;
 	struct intel_crtc *crtc;
+	unsigned long cookie;
 	int err = 0;
 
 	if (!HAS_DDI(dev_priv))
 		return 0;
 
-	i915_audio_component_get_power(kdev);
+	cookie = i915_audio_component_get_power(kdev);
 	mutex_lock(&dev_priv->av_mutex);
 
 	/* 1. get the pipe */
@@ -875,7 +942,7 @@ static int i915_audio_component_sync_audio_rate(struct device *kdev, int port,
 
  unlock:
 	mutex_unlock(&dev_priv->av_mutex);
-	i915_audio_component_put_power(kdev);
+	i915_audio_component_put_power(kdev, cookie);
 	return err;
 }
 
@@ -980,7 +1047,7 @@ static const struct component_ops i915_audio_component_bind_ops = {
  * We ignore any error during registration and continue with reduced
  * functionality (i.e. without HDMI audio).
  */
-void i915_audio_component_init(struct drm_i915_private *dev_priv)
+static void i915_audio_component_init(struct drm_i915_private *dev_priv)
 {
 	int ret;
 
@@ -1003,7 +1070,7 @@ void i915_audio_component_init(struct drm_i915_private *dev_priv)
  * Deregisters the audio component, breaking any existing binding to the
  * corresponding snd_hda_intel driver's master component.
  */
-void i915_audio_component_cleanup(struct drm_i915_private *dev_priv)
+static void i915_audio_component_cleanup(struct drm_i915_private *dev_priv)
 {
 	if (!dev_priv->audio_component_registered)
 		return;
diff --git a/drivers/gpu/drm/i915/intel_audio.h b/drivers/gpu/drm/i915/intel_audio.h
new file mode 100644
index 000000000000..a3657c7a7ba2
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_audio.h
@@ -0,0 +1,24 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_AUDIO_H__
+#define __INTEL_AUDIO_H__
+
+struct drm_connector_state;
+struct drm_i915_private;
+struct intel_crtc_state;
+struct intel_encoder;
+
+void intel_init_audio_hooks(struct drm_i915_private *dev_priv);
+void intel_audio_codec_enable(struct intel_encoder *encoder,
+			      const struct intel_crtc_state *crtc_state,
+			      const struct drm_connector_state *conn_state);
+void intel_audio_codec_disable(struct intel_encoder *encoder,
+			       const struct intel_crtc_state *old_crtc_state,
+			       const struct drm_connector_state *old_conn_state);
+void intel_audio_init(struct drm_i915_private *dev_priv);
+void intel_audio_deinit(struct drm_i915_private *dev_priv);
+
+#endif /* __INTEL_AUDIO_H__ */
diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c
index 4364f42cac6b..1dc8d03ff127 100644
--- a/drivers/gpu/drm/i915/intel_bios.c
+++ b/drivers/gpu/drm/i915/intel_bios.c
@@ -760,6 +760,31 @@ parse_psr(struct drm_i915_private *dev_priv, const struct bdb_header *bdb)
 		dev_priv->vbt.psr.tp1_wakeup_time_us = psr_table->tp1_wakeup_time * 100;
 		dev_priv->vbt.psr.tp2_tp3_wakeup_time_us = psr_table->tp2_tp3_wakeup_time * 100;
 	}
+
+	if (bdb->version >= 226) {
+		u32 wakeup_time = psr_table->psr2_tp2_tp3_wakeup_time;
+
+		wakeup_time = (wakeup_time >> (2 * panel_type)) & 0x3;
+		switch (wakeup_time) {
+		case 0:
+			wakeup_time = 500;
+			break;
+		case 1:
+			wakeup_time = 100;
+			break;
+		case 3:
+			wakeup_time = 50;
+			break;
+		default:
+		case 2:
+			wakeup_time = 2500;
+			break;
+		}
+		dev_priv->vbt.psr.psr2_tp2_tp3_wakeup_time_us = wakeup_time;
+	} else {
+		/* Reusing PSR1 wakeup time for PSR2 in older VBTs */
+		dev_priv->vbt.psr.psr2_tp2_tp3_wakeup_time_us = dev_priv->vbt.psr.tp2_tp3_wakeup_time_us;
+	}
 }
 
 static void parse_dsi_backlight_ports(struct drm_i915_private *dev_priv,
@@ -1222,10 +1247,11 @@ static void sanitize_ddc_pin(struct drm_i915_private *dev_priv,
 	if (!info->alternate_ddc_pin)
 		return;
 
-	for_each_port_masked(p, (1 << port) - 1) {
+	for (p = PORT_A; p < I915_MAX_PORTS; p++) {
 		struct ddi_vbt_port_info *i = &dev_priv->vbt.ddi_port_info[p];
 
-		if (info->alternate_ddc_pin != i->alternate_ddc_pin)
+		if (p == port || !i->present ||
+		    info->alternate_ddc_pin != i->alternate_ddc_pin)
 			continue;
 
 		DRM_DEBUG_KMS("port %c trying to use the same DDC pin (0x%x) as port %c, "
@@ -1239,8 +1265,8 @@ static void sanitize_ddc_pin(struct drm_i915_private *dev_priv,
 		 * port. Otherwise they share the same ddc bin and
 		 * system couldn't communicate with them separately.
 		 *
-		 * Due to parsing the ports in alphabetical order,
-		 * a higher port will always clobber a lower one.
+		 * Due to parsing the ports in child device order,
+		 * a later device will always clobber an earlier one.
 		 */
 		i->supports_dvi = false;
 		i->supports_hdmi = false;
@@ -1258,10 +1284,11 @@ static void sanitize_aux_ch(struct drm_i915_private *dev_priv,
 	if (!info->alternate_aux_channel)
 		return;
 
-	for_each_port_masked(p, (1 << port) - 1) {
+	for (p = PORT_A; p < I915_MAX_PORTS; p++) {
 		struct ddi_vbt_port_info *i = &dev_priv->vbt.ddi_port_info[p];
 
-		if (info->alternate_aux_channel != i->alternate_aux_channel)
+		if (p == port || !i->present ||
+		    info->alternate_aux_channel != i->alternate_aux_channel)
 			continue;
 
 		DRM_DEBUG_KMS("port %c trying to use the same AUX CH (0x%x) as port %c, "
@@ -1275,8 +1302,8 @@ static void sanitize_aux_ch(struct drm_i915_private *dev_priv,
 		 * port. Otherwise they share the same aux channel
 		 * and system couldn't communicate with them separately.
 		 *
-		 * Due to parsing the ports in alphabetical order,
-		 * a higher port will always clobber a lower one.
+		 * Due to parsing the ports in child device order,
+		 * a later device will always clobber an earlier one.
 		 */
 		i->supports_dp = false;
 		i->alternate_aux_channel = 0;
@@ -1324,48 +1351,57 @@ static u8 map_ddc_pin(struct drm_i915_private *dev_priv, u8 vbt_pin)
 	return 0;
 }
 
-static void parse_ddi_port(struct drm_i915_private *dev_priv, enum port port,
-			   u8 bdb_version)
+static enum port dvo_port_to_port(u8 dvo_port)
 {
-	struct child_device_config *it, *child = NULL;
-	struct ddi_vbt_port_info *info = &dev_priv->vbt.ddi_port_info[port];
-	int i, j;
-	bool is_dvi, is_hdmi, is_dp, is_edp, is_crt;
-	/* Each DDI port can have more than one value on the "DVO Port" field,
+	/*
+	 * Each DDI port can have more than one value on the "DVO Port" field,
 	 * so look for all the possible values for each port.
 	 */
-	int dvo_ports[][3] = {
-		{DVO_PORT_HDMIA, DVO_PORT_DPA, -1},
-		{DVO_PORT_HDMIB, DVO_PORT_DPB, -1},
-		{DVO_PORT_HDMIC, DVO_PORT_DPC, -1},
-		{DVO_PORT_HDMID, DVO_PORT_DPD, -1},
-		{DVO_PORT_CRT, DVO_PORT_HDMIE, DVO_PORT_DPE},
-		{DVO_PORT_HDMIF, DVO_PORT_DPF, -1},
+	static const int dvo_ports[][3] = {
+		[PORT_A] = { DVO_PORT_HDMIA, DVO_PORT_DPA, -1},
+		[PORT_B] = { DVO_PORT_HDMIB, DVO_PORT_DPB, -1},
+		[PORT_C] = { DVO_PORT_HDMIC, DVO_PORT_DPC, -1},
+		[PORT_D] = { DVO_PORT_HDMID, DVO_PORT_DPD, -1},
+		[PORT_E] = { DVO_PORT_CRT, DVO_PORT_HDMIE, DVO_PORT_DPE},
+		[PORT_F] = { DVO_PORT_HDMIF, DVO_PORT_DPF, -1},
 	};
+	enum port port;
+	int i;
 
-	/*
-	 * Find the first child device to reference the port, report if more
-	 * than one found.
-	 */
-	for (i = 0; i < dev_priv->vbt.child_dev_num; i++) {
-		it = dev_priv->vbt.child_dev + i;
-
-		for (j = 0; j < 3; j++) {
-			if (dvo_ports[port][j] == -1)
+	for (port = PORT_A; port < ARRAY_SIZE(dvo_ports); port++) {
+		for (i = 0; i < ARRAY_SIZE(dvo_ports[port]); i++) {
+			if (dvo_ports[port][i] == -1)
 				break;
 
-			if (it->dvo_port == dvo_ports[port][j]) {
-				if (child) {
-					DRM_DEBUG_KMS("More than one child device for port %c in VBT, using the first.\n",
-						      port_name(port));
-				} else {
-					child = it;
-				}
-			}
+			if (dvo_port == dvo_ports[port][i])
+				return port;
 		}
 	}
-	if (!child)
+
+	return PORT_NONE;
+}
+
+static void parse_ddi_port(struct drm_i915_private *dev_priv,
+			   const struct child_device_config *child,
+			   u8 bdb_version)
+{
+	struct ddi_vbt_port_info *info;
+	bool is_dvi, is_hdmi, is_dp, is_edp, is_crt;
+	enum port port;
+
+	port = dvo_port_to_port(child->dvo_port);
+	if (port == PORT_NONE)
+		return;
+
+	info = &dev_priv->vbt.ddi_port_info[port];
+
+	if (info->present) {
+		DRM_DEBUG_KMS("More than one child device for port %c in VBT, using the first.\n",
+			      port_name(port));
 		return;
+	}
+
+	info->present = true;
 
 	is_dvi = child->device_type & DEVICE_TYPE_TMDS_DVI_SIGNALING;
 	is_dp = child->device_type & DEVICE_TYPE_DISPLAYPORT_OUTPUT;
@@ -1498,19 +1534,20 @@ static void parse_ddi_port(struct drm_i915_private *dev_priv, enum port port,
 
 static void parse_ddi_ports(struct drm_i915_private *dev_priv, u8 bdb_version)
 {
-	enum port port;
+	const struct child_device_config *child;
+	int i;
 
 	if (!HAS_DDI(dev_priv) && !IS_CHERRYVIEW(dev_priv))
 		return;
 
-	if (!dev_priv->vbt.child_dev_num)
-		return;
-
 	if (bdb_version < 155)
 		return;
 
-	for (port = PORT_A; port < I915_MAX_PORTS; port++)
-		parse_ddi_port(dev_priv, port, bdb_version);
+	for (i = 0; i < dev_priv->vbt.child_dev_num; i++) {
+		child = dev_priv->vbt.child_dev + i;
+
+		parse_ddi_port(dev_priv, child, bdb_version);
+	}
 }
 
 static void
@@ -2094,8 +2131,8 @@ bool intel_bios_is_dsi_present(struct drm_i915_private *dev_priv,
 		dvo_port = child->dvo_port;
 
 		if (dvo_port == DVO_PORT_MIPIA ||
-		    (dvo_port == DVO_PORT_MIPIB && IS_ICELAKE(dev_priv)) ||
-		    (dvo_port == DVO_PORT_MIPIC && !IS_ICELAKE(dev_priv))) {
+		    (dvo_port == DVO_PORT_MIPIB && INTEL_GEN(dev_priv) >= 11) ||
+		    (dvo_port == DVO_PORT_MIPIC && INTEL_GEN(dev_priv) < 11)) {
 			if (port)
 				*port = dvo_port - DVO_PORT_MIPIA;
 			return true;
diff --git a/drivers/gpu/drm/i915/intel_breadcrumbs.c b/drivers/gpu/drm/i915/intel_breadcrumbs.c
index 09ed90c0ba00..3cbffd400b1b 100644
--- a/drivers/gpu/drm/i915/intel_breadcrumbs.c
+++ b/drivers/gpu/drm/i915/intel_breadcrumbs.c
@@ -27,8 +27,6 @@
 
 #include "i915_drv.h"
 
-#define task_asleep(tsk) ((tsk)->state & TASK_NORMAL && !(tsk)->on_rq)
-
 static void irq_enable(struct intel_engine_cs *engine)
 {
 	if (!engine->irq_enable)
@@ -82,7 +80,7 @@ static inline bool __request_completed(const struct i915_request *rq)
 	return i915_seqno_passed(__hwsp_seqno(rq), rq->fence.seqno);
 }
 
-bool intel_engine_breadcrumbs_irq(struct intel_engine_cs *engine)
+void intel_engine_breadcrumbs_irq(struct intel_engine_cs *engine)
 {
 	struct intel_breadcrumbs *b = &engine->breadcrumbs;
 	struct intel_context *ce, *cn;
@@ -146,19 +144,13 @@ bool intel_engine_breadcrumbs_irq(struct intel_engine_cs *engine)
 		dma_fence_signal(&rq->fence);
 		i915_request_put(rq);
 	}
-
-	return !list_empty(&signal);
 }
 
-bool intel_engine_signal_breadcrumbs(struct intel_engine_cs *engine)
+void intel_engine_signal_breadcrumbs(struct intel_engine_cs *engine)
 {
-	bool result;
-
 	local_irq_disable();
-	result = intel_engine_breadcrumbs_irq(engine);
+	intel_engine_breadcrumbs_irq(engine);
 	local_irq_enable();
-
-	return result;
 }
 
 static void signal_irq_work(struct irq_work *work)
diff --git a/drivers/gpu/drm/i915/intel_cdclk.c b/drivers/gpu/drm/i915/intel_cdclk.c
index 15ba950dee00..ae40a8679314 100644
--- a/drivers/gpu/drm/i915/intel_cdclk.c
+++ b/drivers/gpu/drm/i915/intel_cdclk.c
@@ -21,6 +21,7 @@
  * DEALINGS IN THE SOFTWARE.
  */
 
+#include "intel_cdclk.h"
 #include "intel_drv.h"
 
 /**
@@ -234,7 +235,8 @@ static unsigned int intel_hpll_vco(struct drm_i915_private *dev_priv)
 	else
 		return 0;
 
-	tmp = I915_READ(IS_MOBILE(dev_priv) ? HPLLVCO_MOBILE : HPLLVCO);
+	tmp = I915_READ(IS_PINEVIEW(dev_priv) || IS_MOBILE(dev_priv) ?
+			HPLLVCO_MOBILE : HPLLVCO);
 
 	vco = vco_table[tmp & 0x7];
 	if (vco == 0)
@@ -468,7 +470,7 @@ static void vlv_get_cdclk(struct drm_i915_private *dev_priv,
 					       cdclk_state->vco);
 
 	mutex_lock(&dev_priv->pcu_lock);
-	val = vlv_punit_read(dev_priv, PUNIT_REG_DSPFREQ);
+	val = vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM);
 	mutex_unlock(&dev_priv->pcu_lock);
 
 	if (IS_VALLEYVIEW(dev_priv))
@@ -516,7 +518,8 @@ static void vlv_program_pfi_credits(struct drm_i915_private *dev_priv)
 }
 
 static void vlv_set_cdclk(struct drm_i915_private *dev_priv,
-			  const struct intel_cdclk_state *cdclk_state)
+			  const struct intel_cdclk_state *cdclk_state,
+			  enum pipe pipe)
 {
 	int cdclk = cdclk_state->cdclk;
 	u32 val, cmd = cdclk_state->voltage_level;
@@ -543,11 +546,11 @@ static void vlv_set_cdclk(struct drm_i915_private *dev_priv,
 	wakeref = intel_display_power_get(dev_priv, POWER_DOMAIN_PIPE_A);
 
 	mutex_lock(&dev_priv->pcu_lock);
-	val = vlv_punit_read(dev_priv, PUNIT_REG_DSPFREQ);
+	val = vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM);
 	val &= ~DSPFREQGUAR_MASK;
 	val |= (cmd << DSPFREQGUAR_SHIFT);
-	vlv_punit_write(dev_priv, PUNIT_REG_DSPFREQ, val);
-	if (wait_for((vlv_punit_read(dev_priv, PUNIT_REG_DSPFREQ) &
+	vlv_punit_write(dev_priv, PUNIT_REG_DSPSSPM, val);
+	if (wait_for((vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM) &
 		      DSPFREQSTAT_MASK) == (cmd << DSPFREQSTAT_SHIFT),
 		     50)) {
 		DRM_ERROR("timed out waiting for CDclk change\n");
@@ -598,7 +601,8 @@ static void vlv_set_cdclk(struct drm_i915_private *dev_priv,
 }
 
 static void chv_set_cdclk(struct drm_i915_private *dev_priv,
-			  const struct intel_cdclk_state *cdclk_state)
+			  const struct intel_cdclk_state *cdclk_state,
+			  enum pipe pipe)
 {
 	int cdclk = cdclk_state->cdclk;
 	u32 val, cmd = cdclk_state->voltage_level;
@@ -624,11 +628,11 @@ static void chv_set_cdclk(struct drm_i915_private *dev_priv,
 	wakeref = intel_display_power_get(dev_priv, POWER_DOMAIN_PIPE_A);
 
 	mutex_lock(&dev_priv->pcu_lock);
-	val = vlv_punit_read(dev_priv, PUNIT_REG_DSPFREQ);
+	val = vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM);
 	val &= ~DSPFREQGUAR_MASK_CHV;
 	val |= (cmd << DSPFREQGUAR_SHIFT_CHV);
-	vlv_punit_write(dev_priv, PUNIT_REG_DSPFREQ, val);
-	if (wait_for((vlv_punit_read(dev_priv, PUNIT_REG_DSPFREQ) &
+	vlv_punit_write(dev_priv, PUNIT_REG_DSPSSPM, val);
+	if (wait_for((vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM) &
 		      DSPFREQSTAT_MASK_CHV) == (cmd << DSPFREQSTAT_SHIFT_CHV),
 		     50)) {
 		DRM_ERROR("timed out waiting for CDclk change\n");
@@ -697,7 +701,8 @@ static void bdw_get_cdclk(struct drm_i915_private *dev_priv,
 }
 
 static void bdw_set_cdclk(struct drm_i915_private *dev_priv,
-			  const struct intel_cdclk_state *cdclk_state)
+			  const struct intel_cdclk_state *cdclk_state,
+			  enum pipe pipe)
 {
 	int cdclk = cdclk_state->cdclk;
 	u32 val;
@@ -964,7 +969,7 @@ static void skl_dpll0_enable(struct drm_i915_private *dev_priv, int vco)
 
 	I915_WRITE(LCPLL1_CTL, I915_READ(LCPLL1_CTL) | LCPLL_PLL_ENABLE);
 
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    LCPLL1_CTL, LCPLL_PLL_LOCK, LCPLL_PLL_LOCK,
 				    5))
 		DRM_ERROR("DPLL0 not locked\n");
@@ -978,16 +983,17 @@ static void skl_dpll0_enable(struct drm_i915_private *dev_priv, int vco)
 static void skl_dpll0_disable(struct drm_i915_private *dev_priv)
 {
 	I915_WRITE(LCPLL1_CTL, I915_READ(LCPLL1_CTL) & ~LCPLL_PLL_ENABLE);
-	if (intel_wait_for_register(dev_priv,
-				   LCPLL1_CTL, LCPLL_PLL_LOCK, 0,
-				   1))
+	if (intel_wait_for_register(&dev_priv->uncore,
+				    LCPLL1_CTL, LCPLL_PLL_LOCK, 0,
+				    1))
 		DRM_ERROR("Couldn't disable DPLL0\n");
 
 	dev_priv->cdclk.hw.vco = 0;
 }
 
 static void skl_set_cdclk(struct drm_i915_private *dev_priv,
-			  const struct intel_cdclk_state *cdclk_state)
+			  const struct intel_cdclk_state *cdclk_state,
+			  enum pipe pipe)
 {
 	int cdclk = cdclk_state->cdclk;
 	int vco = cdclk_state->vco;
@@ -1123,16 +1129,7 @@ sanitize:
 	dev_priv->cdclk.hw.vco = -1;
 }
 
-/**
- * skl_init_cdclk - Initialize CDCLK on SKL
- * @dev_priv: i915 device
- *
- * Initialize CDCLK for SKL and derivatives. This is generally
- * done only during the display core initialization sequence,
- * after which the DMC will take care of turning CDCLK off/on
- * as needed.
- */
-void skl_init_cdclk(struct drm_i915_private *dev_priv)
+static void skl_init_cdclk(struct drm_i915_private *dev_priv)
 {
 	struct intel_cdclk_state cdclk_state;
 
@@ -1158,17 +1155,10 @@ void skl_init_cdclk(struct drm_i915_private *dev_priv)
 	cdclk_state.cdclk = skl_calc_cdclk(0, cdclk_state.vco);
 	cdclk_state.voltage_level = skl_calc_voltage_level(cdclk_state.cdclk);
 
-	skl_set_cdclk(dev_priv, &cdclk_state);
+	skl_set_cdclk(dev_priv, &cdclk_state, INVALID_PIPE);
 }
 
-/**
- * skl_uninit_cdclk - Uninitialize CDCLK on SKL
- * @dev_priv: i915 device
- *
- * Uninitialize CDCLK for SKL and derivatives. This is done only
- * during the display core uninitialization sequence.
- */
-void skl_uninit_cdclk(struct drm_i915_private *dev_priv)
+static void skl_uninit_cdclk(struct drm_i915_private *dev_priv)
 {
 	struct intel_cdclk_state cdclk_state = dev_priv->cdclk.hw;
 
@@ -1176,7 +1166,7 @@ void skl_uninit_cdclk(struct drm_i915_private *dev_priv)
 	cdclk_state.vco = 0;
 	cdclk_state.voltage_level = skl_calc_voltage_level(cdclk_state.cdclk);
 
-	skl_set_cdclk(dev_priv, &cdclk_state);
+	skl_set_cdclk(dev_priv, &cdclk_state, INVALID_PIPE);
 }
 
 static int bxt_calc_cdclk(int min_cdclk)
@@ -1323,7 +1313,7 @@ static void bxt_de_pll_disable(struct drm_i915_private *dev_priv)
 	I915_WRITE(BXT_DE_PLL_ENABLE, 0);
 
 	/* Timeout 200us */
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    BXT_DE_PLL_ENABLE, BXT_DE_PLL_LOCK, 0,
 				    1))
 		DRM_ERROR("timeout waiting for DE PLL unlock\n");
@@ -1344,7 +1334,7 @@ static void bxt_de_pll_enable(struct drm_i915_private *dev_priv, int vco)
 	I915_WRITE(BXT_DE_PLL_ENABLE, BXT_DE_PLL_PLL_ENABLE);
 
 	/* Timeout 200us */
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    BXT_DE_PLL_ENABLE,
 				    BXT_DE_PLL_LOCK,
 				    BXT_DE_PLL_LOCK,
@@ -1355,7 +1345,8 @@ static void bxt_de_pll_enable(struct drm_i915_private *dev_priv, int vco)
 }
 
 static void bxt_set_cdclk(struct drm_i915_private *dev_priv,
-			  const struct intel_cdclk_state *cdclk_state)
+			  const struct intel_cdclk_state *cdclk_state,
+			  enum pipe pipe)
 {
 	int cdclk = cdclk_state->cdclk;
 	int vco = cdclk_state->vco;
@@ -1408,11 +1399,10 @@ static void bxt_set_cdclk(struct drm_i915_private *dev_priv,
 		bxt_de_pll_enable(dev_priv, vco);
 
 	val = divider | skl_cdclk_decimal(cdclk);
-	/*
-	 * FIXME if only the cd2x divider needs changing, it could be done
-	 * without shutting off the pipe (if only one pipe is active).
-	 */
-	val |= BXT_CDCLK_CD2X_PIPE_NONE;
+	if (pipe == INVALID_PIPE)
+		val |= BXT_CDCLK_CD2X_PIPE_NONE;
+	else
+		val |= BXT_CDCLK_CD2X_PIPE(pipe);
 	/*
 	 * Disable SSA Precharge when CD clock frequency < 500 MHz,
 	 * enable otherwise.
@@ -1421,6 +1411,9 @@ static void bxt_set_cdclk(struct drm_i915_private *dev_priv,
 		val |= BXT_CDCLK_SSA_PRECHARGE_ENABLE;
 	I915_WRITE(CDCLK_CTL, val);
 
+	if (pipe != INVALID_PIPE)
+		intel_wait_for_vblank(dev_priv, pipe);
+
 	mutex_lock(&dev_priv->pcu_lock);
 	/*
 	 * The timeout isn't specified, the 2ms used here is based on
@@ -1490,16 +1483,7 @@ sanitize:
 	dev_priv->cdclk.hw.vco = -1;
 }
 
-/**
- * bxt_init_cdclk - Initialize CDCLK on BXT
- * @dev_priv: i915 device
- *
- * Initialize CDCLK for BXT and derivatives. This is generally
- * done only during the display core initialization sequence,
- * after which the DMC will take care of turning CDCLK off/on
- * as needed.
- */
-void bxt_init_cdclk(struct drm_i915_private *dev_priv)
+static void bxt_init_cdclk(struct drm_i915_private *dev_priv)
 {
 	struct intel_cdclk_state cdclk_state;
 
@@ -1525,17 +1509,10 @@ void bxt_init_cdclk(struct drm_i915_private *dev_priv)
 	}
 	cdclk_state.voltage_level = bxt_calc_voltage_level(cdclk_state.cdclk);
 
-	bxt_set_cdclk(dev_priv, &cdclk_state);
+	bxt_set_cdclk(dev_priv, &cdclk_state, INVALID_PIPE);
 }
 
-/**
- * bxt_uninit_cdclk - Uninitialize CDCLK on BXT
- * @dev_priv: i915 device
- *
- * Uninitialize CDCLK for BXT and derivatives. This is done only
- * during the display core uninitialization sequence.
- */
-void bxt_uninit_cdclk(struct drm_i915_private *dev_priv)
+static void bxt_uninit_cdclk(struct drm_i915_private *dev_priv)
 {
 	struct intel_cdclk_state cdclk_state = dev_priv->cdclk.hw;
 
@@ -1543,7 +1520,7 @@ void bxt_uninit_cdclk(struct drm_i915_private *dev_priv)
 	cdclk_state.vco = 0;
 	cdclk_state.voltage_level = bxt_calc_voltage_level(cdclk_state.cdclk);
 
-	bxt_set_cdclk(dev_priv, &cdclk_state);
+	bxt_set_cdclk(dev_priv, &cdclk_state, INVALID_PIPE);
 }
 
 static int cnl_calc_cdclk(int min_cdclk)
@@ -1663,7 +1640,8 @@ static void cnl_cdclk_pll_enable(struct drm_i915_private *dev_priv, int vco)
 }
 
 static void cnl_set_cdclk(struct drm_i915_private *dev_priv,
-			  const struct intel_cdclk_state *cdclk_state)
+			  const struct intel_cdclk_state *cdclk_state,
+			  enum pipe pipe)
 {
 	int cdclk = cdclk_state->cdclk;
 	int vco = cdclk_state->vco;
@@ -1704,13 +1682,15 @@ static void cnl_set_cdclk(struct drm_i915_private *dev_priv,
 		cnl_cdclk_pll_enable(dev_priv, vco);
 
 	val = divider | skl_cdclk_decimal(cdclk);
-	/*
-	 * FIXME if only the cd2x divider needs changing, it could be done
-	 * without shutting off the pipe (if only one pipe is active).
-	 */
-	val |= BXT_CDCLK_CD2X_PIPE_NONE;
+	if (pipe == INVALID_PIPE)
+		val |= BXT_CDCLK_CD2X_PIPE_NONE;
+	else
+		val |= BXT_CDCLK_CD2X_PIPE(pipe);
 	I915_WRITE(CDCLK_CTL, val);
 
+	if (pipe != INVALID_PIPE)
+		intel_wait_for_vblank(dev_priv, pipe);
+
 	/* inform PCU of the change */
 	mutex_lock(&dev_priv->pcu_lock);
 	sandybridge_pcode_write(dev_priv, SKL_PCODE_CDCLK_CONTROL,
@@ -1847,7 +1827,8 @@ static int icl_calc_cdclk_pll_vco(struct drm_i915_private *dev_priv, int cdclk)
 }
 
 static void icl_set_cdclk(struct drm_i915_private *dev_priv,
-			  const struct intel_cdclk_state *cdclk_state)
+			  const struct intel_cdclk_state *cdclk_state,
+			  enum pipe pipe)
 {
 	unsigned int cdclk = cdclk_state->cdclk;
 	unsigned int vco = cdclk_state->vco;
@@ -1872,6 +1853,11 @@ static void icl_set_cdclk(struct drm_i915_private *dev_priv,
 	if (dev_priv->cdclk.hw.vco != vco)
 		cnl_cdclk_pll_enable(dev_priv, vco);
 
+	/*
+	 * On ICL CD2X_DIV can only be 1, so we'll never end up changing the
+	 * divider here synchronized to a pipe while CDCLK is on, nor will we
+	 * need the corresponding vblank wait.
+	 */
 	I915_WRITE(CDCLK_CTL, ICL_CDCLK_CD2X_PIPE_NONE |
 			      skl_cdclk_decimal(cdclk));
 
@@ -1959,16 +1945,7 @@ out:
 		icl_calc_voltage_level(cdclk_state->cdclk);
 }
 
-/**
- * icl_init_cdclk - Initialize CDCLK on ICL
- * @dev_priv: i915 device
- *
- * Initialize CDCLK for ICL. This consists mainly of initializing
- * dev_priv->cdclk.hw and sanitizing the state of the hardware if needed. This
- * is generally done only during the display core initialization sequence, after
- * which the DMC will take care of turning CDCLK off/on as needed.
- */
-void icl_init_cdclk(struct drm_i915_private *dev_priv)
+static void icl_init_cdclk(struct drm_i915_private *dev_priv)
 {
 	struct intel_cdclk_state sanitized_state;
 	u32 val;
@@ -2002,17 +1979,10 @@ sanitize:
 	sanitized_state.voltage_level =
 				icl_calc_voltage_level(sanitized_state.cdclk);
 
-	icl_set_cdclk(dev_priv, &sanitized_state);
+	icl_set_cdclk(dev_priv, &sanitized_state, INVALID_PIPE);
 }
 
-/**
- * icl_uninit_cdclk - Uninitialize CDCLK on ICL
- * @dev_priv: i915 device
- *
- * Uninitialize CDCLK for ICL. This is done only during the display core
- * uninitialization sequence.
- */
-void icl_uninit_cdclk(struct drm_i915_private *dev_priv)
+static void icl_uninit_cdclk(struct drm_i915_private *dev_priv)
 {
 	struct intel_cdclk_state cdclk_state = dev_priv->cdclk.hw;
 
@@ -2020,19 +1990,10 @@ void icl_uninit_cdclk(struct drm_i915_private *dev_priv)
 	cdclk_state.vco = 0;
 	cdclk_state.voltage_level = icl_calc_voltage_level(cdclk_state.cdclk);
 
-	icl_set_cdclk(dev_priv, &cdclk_state);
+	icl_set_cdclk(dev_priv, &cdclk_state, INVALID_PIPE);
 }
 
-/**
- * cnl_init_cdclk - Initialize CDCLK on CNL
- * @dev_priv: i915 device
- *
- * Initialize CDCLK for CNL. This is generally
- * done only during the display core initialization sequence,
- * after which the DMC will take care of turning CDCLK off/on
- * as needed.
- */
-void cnl_init_cdclk(struct drm_i915_private *dev_priv)
+static void cnl_init_cdclk(struct drm_i915_private *dev_priv)
 {
 	struct intel_cdclk_state cdclk_state;
 
@@ -2048,17 +2009,10 @@ void cnl_init_cdclk(struct drm_i915_private *dev_priv)
 	cdclk_state.vco = cnl_cdclk_pll_vco(dev_priv, cdclk_state.cdclk);
 	cdclk_state.voltage_level = cnl_calc_voltage_level(cdclk_state.cdclk);
 
-	cnl_set_cdclk(dev_priv, &cdclk_state);
+	cnl_set_cdclk(dev_priv, &cdclk_state, INVALID_PIPE);
 }
 
-/**
- * cnl_uninit_cdclk - Uninitialize CDCLK on CNL
- * @dev_priv: i915 device
- *
- * Uninitialize CDCLK for CNL. This is done only
- * during the display core uninitialization sequence.
- */
-void cnl_uninit_cdclk(struct drm_i915_private *dev_priv)
+static void cnl_uninit_cdclk(struct drm_i915_private *dev_priv)
 {
 	struct intel_cdclk_state cdclk_state = dev_priv->cdclk.hw;
 
@@ -2066,7 +2020,47 @@ void cnl_uninit_cdclk(struct drm_i915_private *dev_priv)
 	cdclk_state.vco = 0;
 	cdclk_state.voltage_level = cnl_calc_voltage_level(cdclk_state.cdclk);
 
-	cnl_set_cdclk(dev_priv, &cdclk_state);
+	cnl_set_cdclk(dev_priv, &cdclk_state, INVALID_PIPE);
+}
+
+/**
+ * intel_cdclk_init - Initialize CDCLK
+ * @i915: i915 device
+ *
+ * Initialize CDCLK. This consists mainly of initializing dev_priv->cdclk.hw and
+ * sanitizing the state of the hardware if needed. This is generally done only
+ * during the display core initialization sequence, after which the DMC will
+ * take care of turning CDCLK off/on as needed.
+ */
+void intel_cdclk_init(struct drm_i915_private *i915)
+{
+	if (INTEL_GEN(i915) >= 11)
+		icl_init_cdclk(i915);
+	else if (IS_CANNONLAKE(i915))
+		cnl_init_cdclk(i915);
+	else if (IS_GEN9_BC(i915))
+		skl_init_cdclk(i915);
+	else if (IS_GEN9_LP(i915))
+		bxt_init_cdclk(i915);
+}
+
+/**
+ * intel_cdclk_uninit - Uninitialize CDCLK
+ * @i915: i915 device
+ *
+ * Uninitialize CDCLK. This is done only during the display core
+ * uninitialization sequence.
+ */
+void intel_cdclk_uninit(struct drm_i915_private *i915)
+{
+	if (INTEL_GEN(i915) >= 11)
+		icl_uninit_cdclk(i915);
+	else if (IS_CANNONLAKE(i915))
+		cnl_uninit_cdclk(i915);
+	else if (IS_GEN9_BC(i915))
+		skl_uninit_cdclk(i915);
+	else if (IS_GEN9_LP(i915))
+		bxt_uninit_cdclk(i915);
 }
 
 /**
@@ -2086,6 +2080,28 @@ bool intel_cdclk_needs_modeset(const struct intel_cdclk_state *a,
 }
 
 /**
+ * intel_cdclk_needs_cd2x_update - Determine if two CDCLK states require a cd2x divider update
+ * @dev_priv: Not a CDCLK state, it's the drm_i915_private!
+ * @a: first CDCLK state
+ * @b: second CDCLK state
+ *
+ * Returns:
+ * True if the CDCLK states require just a cd2x divider update, false if not.
+ */
+bool intel_cdclk_needs_cd2x_update(struct drm_i915_private *dev_priv,
+				   const struct intel_cdclk_state *a,
+				   const struct intel_cdclk_state *b)
+{
+	/* Older hw doesn't have the capability */
+	if (INTEL_GEN(dev_priv) < 10 && !IS_GEN9_LP(dev_priv))
+		return false;
+
+	return a->cdclk != b->cdclk &&
+		a->vco == b->vco &&
+		a->ref == b->ref;
+}
+
+/**
  * intel_cdclk_changed - Determine if two CDCLK states are different
  * @a: first CDCLK state
  * @b: second CDCLK state
@@ -2100,6 +2116,26 @@ bool intel_cdclk_changed(const struct intel_cdclk_state *a,
 		a->voltage_level != b->voltage_level;
 }
 
+/**
+ * intel_cdclk_swap_state - make atomic CDCLK configuration effective
+ * @state: atomic state
+ *
+ * This is the CDCLK version of drm_atomic_helper_swap_state() since the
+ * helper does not handle driver-specific global state.
+ *
+ * Similarly to the atomic helpers this function does a complete swap,
+ * i.e. it also puts the old state into @state. This is used by the commit
+ * code to determine how CDCLK has changed (for instance did it increase or
+ * decrease).
+ */
+void intel_cdclk_swap_state(struct intel_atomic_state *state)
+{
+	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
+
+	swap(state->cdclk.logical, dev_priv->cdclk.logical);
+	swap(state->cdclk.actual, dev_priv->cdclk.actual);
+}
+
 void intel_dump_cdclk_state(const struct intel_cdclk_state *cdclk_state,
 			    const char *context)
 {
@@ -2113,12 +2149,14 @@ void intel_dump_cdclk_state(const struct intel_cdclk_state *cdclk_state,
  * intel_set_cdclk - Push the CDCLK state to the hardware
  * @dev_priv: i915 device
  * @cdclk_state: new CDCLK state
+ * @pipe: pipe with which to synchronize the update
  *
  * Program the hardware based on the passed in CDCLK state,
  * if necessary.
  */
-void intel_set_cdclk(struct drm_i915_private *dev_priv,
-		     const struct intel_cdclk_state *cdclk_state)
+static void intel_set_cdclk(struct drm_i915_private *dev_priv,
+			    const struct intel_cdclk_state *cdclk_state,
+			    enum pipe pipe)
 {
 	if (!intel_cdclk_changed(&dev_priv->cdclk.hw, cdclk_state))
 		return;
@@ -2128,7 +2166,7 @@ void intel_set_cdclk(struct drm_i915_private *dev_priv,
 
 	intel_dump_cdclk_state(cdclk_state, "Changing CDCLK to");
 
-	dev_priv->display.set_cdclk(dev_priv, cdclk_state);
+	dev_priv->display.set_cdclk(dev_priv, cdclk_state, pipe);
 
 	if (WARN(intel_cdclk_changed(&dev_priv->cdclk.hw, cdclk_state),
 		 "cdclk state doesn't match!\n")) {
@@ -2137,6 +2175,46 @@ void intel_set_cdclk(struct drm_i915_private *dev_priv,
 	}
 }
 
+/**
+ * intel_set_cdclk_pre_plane_update - Push the CDCLK state to the hardware
+ * @dev_priv: i915 device
+ * @old_state: old CDCLK state
+ * @new_state: new CDCLK state
+ * @pipe: pipe with which to synchronize the update
+ *
+ * Program the hardware before updating the HW plane state based on the passed
+ * in CDCLK state, if necessary.
+ */
+void
+intel_set_cdclk_pre_plane_update(struct drm_i915_private *dev_priv,
+				 const struct intel_cdclk_state *old_state,
+				 const struct intel_cdclk_state *new_state,
+				 enum pipe pipe)
+{
+	if (pipe == INVALID_PIPE || old_state->cdclk <= new_state->cdclk)
+		intel_set_cdclk(dev_priv, new_state, pipe);
+}
+
+/**
+ * intel_set_cdclk_post_plane_update - Push the CDCLK state to the hardware
+ * @dev_priv: i915 device
+ * @old_state: old CDCLK state
+ * @new_state: new CDCLK state
+ * @pipe: pipe with which to synchronize the update
+ *
+ * Program the hardware after updating the HW plane state based on the passed
+ * in CDCLK state, if necessary.
+ */
+void
+intel_set_cdclk_post_plane_update(struct drm_i915_private *dev_priv,
+				  const struct intel_cdclk_state *old_state,
+				  const struct intel_cdclk_state *new_state,
+				  enum pipe pipe)
+{
+	if (pipe != INVALID_PIPE && old_state->cdclk > new_state->cdclk)
+		intel_set_cdclk(dev_priv, new_state, pipe);
+}
+
 static int intel_pixel_rate_to_cdclk(struct drm_i915_private *dev_priv,
 				     int pixel_rate)
 {
@@ -2187,19 +2265,8 @@ int intel_crtc_compute_min_cdclk(const struct intel_crtc_state *crtc_state)
 	/*
 	 * According to BSpec, "The CD clock frequency must be at least twice
 	 * the frequency of the Azalia BCLK." and BCLK is 96 MHz by default.
-	 *
-	 * FIXME: Check the actual, not default, BCLK being used.
-	 *
-	 * FIXME: This does not depend on ->has_audio because the higher CDCLK
-	 * is required for audio probe, also when there are no audio capable
-	 * displays connected at probe time. This leads to unnecessarily high
-	 * CDCLK when audio is not required.
-	 *
-	 * FIXME: This limit is only applied when there are displays connected
-	 * at probe time. If we probe without displays, we'll still end up using
-	 * the platform minimum CDCLK, failing audio probe.
 	 */
-	if (INTEL_GEN(dev_priv) >= 9)
+	if (crtc_state->has_audio && INTEL_GEN(dev_priv) >= 9)
 		min_cdclk = max(2 * 96000, min_cdclk);
 
 	/*
@@ -2239,7 +2306,7 @@ static int intel_compute_min_cdclk(struct drm_atomic_state *state)
 		intel_state->min_cdclk[i] = min_cdclk;
 	}
 
-	min_cdclk = 0;
+	min_cdclk = intel_state->cdclk.force_min_cdclk;
 	for_each_pipe(dev_priv, pipe)
 		min_cdclk = max(intel_state->min_cdclk[pipe], min_cdclk);
 
@@ -2300,7 +2367,8 @@ static int vlv_modeset_calc_cdclk(struct drm_atomic_state *state)
 		vlv_calc_voltage_level(dev_priv, cdclk);
 
 	if (!intel_state->active_crtcs) {
-		cdclk = vlv_calc_cdclk(dev_priv, 0);
+		cdclk = vlv_calc_cdclk(dev_priv,
+				       intel_state->cdclk.force_min_cdclk);
 
 		intel_state->cdclk.actual.cdclk = cdclk;
 		intel_state->cdclk.actual.voltage_level =
@@ -2333,7 +2401,7 @@ static int bdw_modeset_calc_cdclk(struct drm_atomic_state *state)
 		bdw_calc_voltage_level(cdclk);
 
 	if (!intel_state->active_crtcs) {
-		cdclk = bdw_calc_cdclk(0);
+		cdclk = bdw_calc_cdclk(intel_state->cdclk.force_min_cdclk);
 
 		intel_state->cdclk.actual.cdclk = cdclk;
 		intel_state->cdclk.actual.voltage_level =
@@ -2405,7 +2473,7 @@ static int skl_modeset_calc_cdclk(struct drm_atomic_state *state)
 		skl_calc_voltage_level(cdclk);
 
 	if (!intel_state->active_crtcs) {
-		cdclk = skl_calc_cdclk(0, vco);
+		cdclk = skl_calc_cdclk(intel_state->cdclk.force_min_cdclk, vco);
 
 		intel_state->cdclk.actual.vco = vco;
 		intel_state->cdclk.actual.cdclk = cdclk;
@@ -2444,10 +2512,10 @@ static int bxt_modeset_calc_cdclk(struct drm_atomic_state *state)
 
 	if (!intel_state->active_crtcs) {
 		if (IS_GEMINILAKE(dev_priv)) {
-			cdclk = glk_calc_cdclk(0);
+			cdclk = glk_calc_cdclk(intel_state->cdclk.force_min_cdclk);
 			vco = glk_de_pll_vco(dev_priv, cdclk);
 		} else {
-			cdclk = bxt_calc_cdclk(0);
+			cdclk = bxt_calc_cdclk(intel_state->cdclk.force_min_cdclk);
 			vco = bxt_de_pll_vco(dev_priv, cdclk);
 		}
 
@@ -2483,7 +2551,7 @@ static int cnl_modeset_calc_cdclk(struct drm_atomic_state *state)
 		    cnl_compute_min_voltage_level(intel_state));
 
 	if (!intel_state->active_crtcs) {
-		cdclk = cnl_calc_cdclk(0);
+		cdclk = cnl_calc_cdclk(intel_state->cdclk.force_min_cdclk);
 		vco = cnl_cdclk_pll_vco(dev_priv, cdclk);
 
 		intel_state->cdclk.actual.vco = vco;
@@ -2519,7 +2587,7 @@ static int icl_modeset_calc_cdclk(struct drm_atomic_state *state)
 		    cnl_compute_min_voltage_level(intel_state));
 
 	if (!intel_state->active_crtcs) {
-		cdclk = icl_calc_cdclk(0, ref);
+		cdclk = icl_calc_cdclk(intel_state->cdclk.force_min_cdclk, ref);
 		vco = icl_calc_cdclk_pll_vco(dev_priv, cdclk);
 
 		intel_state->cdclk.actual.vco = vco;
@@ -2560,7 +2628,7 @@ static int intel_compute_max_dotclk(struct drm_i915_private *dev_priv)
  */
 void intel_update_max_cdclk(struct drm_i915_private *dev_priv)
 {
-	if (IS_ICELAKE(dev_priv)) {
+	if (INTEL_GEN(dev_priv) >= 11) {
 		if (dev_priv->cdclk.hw.ref == 24000)
 			dev_priv->max_cdclk_freq = 648000;
 		else
@@ -2668,7 +2736,7 @@ static int cnp_rawclk(struct drm_i915_private *dev_priv)
 
 		rawclk |= CNP_RAWCLK_DEN(DIV_ROUND_CLOSEST(numerator * 1000,
 							   fraction) - 1);
-		if (HAS_PCH_ICP(dev_priv))
+		if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
 			rawclk |= ICP_RAWCLK_NUM(numerator);
 	}
 
@@ -2723,7 +2791,7 @@ static int g4x_hrawclk(struct drm_i915_private *dev_priv)
  */
 void intel_update_rawclk(struct drm_i915_private *dev_priv)
 {
-	if (HAS_PCH_CNP(dev_priv) || HAS_PCH_ICP(dev_priv))
+	if (INTEL_PCH_TYPE(dev_priv) >= PCH_CNP)
 		dev_priv->rawclk_freq = cnp_rawclk(dev_priv);
 	else if (HAS_PCH_SPLIT(dev_priv))
 		dev_priv->rawclk_freq = pch_rawclk(dev_priv);
@@ -2744,18 +2812,13 @@ void intel_update_rawclk(struct drm_i915_private *dev_priv)
  */
 void intel_init_cdclk_hooks(struct drm_i915_private *dev_priv)
 {
-	if (IS_CHERRYVIEW(dev_priv)) {
-		dev_priv->display.set_cdclk = chv_set_cdclk;
-		dev_priv->display.modeset_calc_cdclk =
-			vlv_modeset_calc_cdclk;
-	} else if (IS_VALLEYVIEW(dev_priv)) {
-		dev_priv->display.set_cdclk = vlv_set_cdclk;
-		dev_priv->display.modeset_calc_cdclk =
-			vlv_modeset_calc_cdclk;
-	} else if (IS_BROADWELL(dev_priv)) {
-		dev_priv->display.set_cdclk = bdw_set_cdclk;
+	if (INTEL_GEN(dev_priv) >= 11) {
+		dev_priv->display.set_cdclk = icl_set_cdclk;
+		dev_priv->display.modeset_calc_cdclk = icl_modeset_calc_cdclk;
+	} else if (IS_CANNONLAKE(dev_priv)) {
+		dev_priv->display.set_cdclk = cnl_set_cdclk;
 		dev_priv->display.modeset_calc_cdclk =
-			bdw_modeset_calc_cdclk;
+			cnl_modeset_calc_cdclk;
 	} else if (IS_GEN9_LP(dev_priv)) {
 		dev_priv->display.set_cdclk = bxt_set_cdclk;
 		dev_priv->display.modeset_calc_cdclk =
@@ -2764,23 +2827,28 @@ void intel_init_cdclk_hooks(struct drm_i915_private *dev_priv)
 		dev_priv->display.set_cdclk = skl_set_cdclk;
 		dev_priv->display.modeset_calc_cdclk =
 			skl_modeset_calc_cdclk;
-	} else if (IS_CANNONLAKE(dev_priv)) {
-		dev_priv->display.set_cdclk = cnl_set_cdclk;
+	} else if (IS_BROADWELL(dev_priv)) {
+		dev_priv->display.set_cdclk = bdw_set_cdclk;
 		dev_priv->display.modeset_calc_cdclk =
-			cnl_modeset_calc_cdclk;
-	} else if (IS_ICELAKE(dev_priv)) {
-		dev_priv->display.set_cdclk = icl_set_cdclk;
-		dev_priv->display.modeset_calc_cdclk = icl_modeset_calc_cdclk;
+			bdw_modeset_calc_cdclk;
+	} else if (IS_CHERRYVIEW(dev_priv)) {
+		dev_priv->display.set_cdclk = chv_set_cdclk;
+		dev_priv->display.modeset_calc_cdclk =
+			vlv_modeset_calc_cdclk;
+	} else if (IS_VALLEYVIEW(dev_priv)) {
+		dev_priv->display.set_cdclk = vlv_set_cdclk;
+		dev_priv->display.modeset_calc_cdclk =
+			vlv_modeset_calc_cdclk;
 	}
 
-	if (IS_ICELAKE(dev_priv))
+	if (INTEL_GEN(dev_priv) >= 11)
 		dev_priv->display.get_cdclk = icl_get_cdclk;
 	else if (IS_CANNONLAKE(dev_priv))
 		dev_priv->display.get_cdclk = cnl_get_cdclk;
-	else if (IS_GEN9_BC(dev_priv))
-		dev_priv->display.get_cdclk = skl_get_cdclk;
 	else if (IS_GEN9_LP(dev_priv))
 		dev_priv->display.get_cdclk = bxt_get_cdclk;
+	else if (IS_GEN9_BC(dev_priv))
+		dev_priv->display.get_cdclk = skl_get_cdclk;
 	else if (IS_BROADWELL(dev_priv))
 		dev_priv->display.get_cdclk = bdw_get_cdclk;
 	else if (IS_HASWELL(dev_priv))
diff --git a/drivers/gpu/drm/i915/intel_cdclk.h b/drivers/gpu/drm/i915/intel_cdclk.h
new file mode 100644
index 000000000000..4d6f7f5f8930
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_cdclk.h
@@ -0,0 +1,46 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_CDCLK_H__
+#define __INTEL_CDCLK_H__
+
+#include <linux/types.h>
+
+#include "intel_display.h"
+
+struct drm_i915_private;
+struct intel_atomic_state;
+struct intel_cdclk_state;
+struct intel_crtc_state;
+
+int intel_crtc_compute_min_cdclk(const struct intel_crtc_state *crtc_state);
+void intel_cdclk_init(struct drm_i915_private *i915);
+void intel_cdclk_uninit(struct drm_i915_private *i915);
+void intel_init_cdclk_hooks(struct drm_i915_private *dev_priv);
+void intel_update_max_cdclk(struct drm_i915_private *dev_priv);
+void intel_update_cdclk(struct drm_i915_private *dev_priv);
+void intel_update_rawclk(struct drm_i915_private *dev_priv);
+bool intel_cdclk_needs_cd2x_update(struct drm_i915_private *dev_priv,
+				   const struct intel_cdclk_state *a,
+				   const struct intel_cdclk_state *b);
+bool intel_cdclk_needs_modeset(const struct intel_cdclk_state *a,
+			       const struct intel_cdclk_state *b);
+bool intel_cdclk_changed(const struct intel_cdclk_state *a,
+			 const struct intel_cdclk_state *b);
+void intel_cdclk_swap_state(struct intel_atomic_state *state);
+void
+intel_set_cdclk_pre_plane_update(struct drm_i915_private *dev_priv,
+				 const struct intel_cdclk_state *old_state,
+				 const struct intel_cdclk_state *new_state,
+				 enum pipe pipe);
+void
+intel_set_cdclk_post_plane_update(struct drm_i915_private *dev_priv,
+				  const struct intel_cdclk_state *old_state,
+				  const struct intel_cdclk_state *new_state,
+				  enum pipe pipe);
+void intel_dump_cdclk_state(const struct intel_cdclk_state *cdclk_state,
+			    const char *context);
+
+#endif /* __INTEL_CDCLK_H__ */
diff --git a/drivers/gpu/drm/i915/intel_color.c b/drivers/gpu/drm/i915/intel_color.c
index 71a1f12c6b2a..9093daabc290 100644
--- a/drivers/gpu/drm/i915/intel_color.c
+++ b/drivers/gpu/drm/i915/intel_color.c
@@ -22,6 +22,7 @@
  *
  */
 
+#include "intel_color.h"
 #include "intel_drv.h"
 
 #define CTM_COEFF_SIGN	(1ULL << 63)
@@ -40,23 +41,6 @@
 #define CTM_COEFF_ABS(coeff)		((coeff) & (CTM_COEFF_SIGN - 1))
 
 #define LEGACY_LUT_LENGTH		256
-
-/* Post offset values for RGB->YCBCR conversion */
-#define POSTOFF_RGB_TO_YUV_HI 0x800
-#define POSTOFF_RGB_TO_YUV_ME 0x100
-#define POSTOFF_RGB_TO_YUV_LO 0x800
-
-/*
- * These values are direct register values specified in the Bspec,
- * for RGB->YUV conversion matrix (colorspace BT709)
- */
-#define CSC_RGB_TO_YUV_RU_GU 0x2ba809d8
-#define CSC_RGB_TO_YUV_BU 0x37e80000
-#define CSC_RGB_TO_YUV_RY_GY 0x1e089cc0
-#define CSC_RGB_TO_YUV_BY 0xb5280000
-#define CSC_RGB_TO_YUV_RV_GV 0xbce89ad8
-#define CSC_RGB_TO_YUV_BV 0x1e080000
-
 /*
  * Extract the CSC coefficient from a CTM coefficient (in U32.32 fixed point
  * format). This macro takes the coefficient we want transformed and the
@@ -69,10 +53,45 @@
 #define ILK_CSC_COEFF_FP(coeff, fbits)	\
 	(clamp_val(((coeff) >> (32 - (fbits) - 3)) + 4, 0, 0xfff) & 0xff8)
 
-#define ILK_CSC_COEFF_LIMITED_RANGE	\
-	ILK_CSC_COEFF_FP(CTM_COEFF_LIMITED_RANGE, 9)
-#define ILK_CSC_COEFF_1_0		\
-	((7 << 12) | ILK_CSC_COEFF_FP(CTM_COEFF_1_0, 8))
+#define ILK_CSC_COEFF_LIMITED_RANGE 0x0dc0
+#define ILK_CSC_COEFF_1_0 0x7800
+
+#define ILK_CSC_POSTOFF_LIMITED_RANGE (16 * (1 << 12) / 255)
+
+static const u16 ilk_csc_off_zero[3] = {};
+
+static const u16 ilk_csc_coeff_identity[9] = {
+	ILK_CSC_COEFF_1_0, 0, 0,
+	0, ILK_CSC_COEFF_1_0, 0,
+	0, 0, ILK_CSC_COEFF_1_0,
+};
+
+static const u16 ilk_csc_postoff_limited_range[3] = {
+	ILK_CSC_POSTOFF_LIMITED_RANGE,
+	ILK_CSC_POSTOFF_LIMITED_RANGE,
+	ILK_CSC_POSTOFF_LIMITED_RANGE,
+};
+
+static const u16 ilk_csc_coeff_limited_range[9] = {
+	ILK_CSC_COEFF_LIMITED_RANGE, 0, 0,
+	0, ILK_CSC_COEFF_LIMITED_RANGE, 0,
+	0, 0, ILK_CSC_COEFF_LIMITED_RANGE,
+};
+
+/*
+ * These values are direct register values specified in the Bspec,
+ * for RGB->YUV conversion matrix (colorspace BT709)
+ */
+static const u16 ilk_csc_coeff_rgb_to_ycbcr[9] = {
+	0x1e08, 0x9cc0, 0xb528,
+	0x2ba8, 0x09d8, 0x37e8,
+	0xbce8, 0x9ad8, 0x1e08,
+};
+
+/* Post offset values for RGB->YCBCR conversion */
+static const u16 ilk_csc_postoff_rgb_to_ycbcr[3] = {
+	0x0800, 0x0100, 0x0800,
+};
 
 static bool lut_is_legacy(const struct drm_property_blob *lut)
 {
@@ -113,145 +132,188 @@ static u64 *ctm_mult_by_limited(u64 *result, const u64 *input)
 	return result;
 }
 
-static void ilk_load_ycbcr_conversion_matrix(struct intel_crtc *crtc)
+static void ilk_update_pipe_csc(struct intel_crtc *crtc,
+				const u16 preoff[3],
+				const u16 coeff[9],
+				const u16 postoff[3])
 {
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
 
-	I915_WRITE(PIPE_CSC_PREOFF_HI(pipe), 0);
-	I915_WRITE(PIPE_CSC_PREOFF_ME(pipe), 0);
-	I915_WRITE(PIPE_CSC_PREOFF_LO(pipe), 0);
+	I915_WRITE(PIPE_CSC_PREOFF_HI(pipe), preoff[0]);
+	I915_WRITE(PIPE_CSC_PREOFF_ME(pipe), preoff[1]);
+	I915_WRITE(PIPE_CSC_PREOFF_LO(pipe), preoff[2]);
 
-	I915_WRITE(PIPE_CSC_COEFF_RU_GU(pipe), CSC_RGB_TO_YUV_RU_GU);
-	I915_WRITE(PIPE_CSC_COEFF_BU(pipe), CSC_RGB_TO_YUV_BU);
+	I915_WRITE(PIPE_CSC_COEFF_RY_GY(pipe), coeff[0] << 16 | coeff[1]);
+	I915_WRITE(PIPE_CSC_COEFF_BY(pipe), coeff[2] << 16);
 
-	I915_WRITE(PIPE_CSC_COEFF_RY_GY(pipe), CSC_RGB_TO_YUV_RY_GY);
-	I915_WRITE(PIPE_CSC_COEFF_BY(pipe), CSC_RGB_TO_YUV_BY);
+	I915_WRITE(PIPE_CSC_COEFF_RU_GU(pipe), coeff[3] << 16 | coeff[4]);
+	I915_WRITE(PIPE_CSC_COEFF_BU(pipe), coeff[5] << 16);
 
-	I915_WRITE(PIPE_CSC_COEFF_RV_GV(pipe), CSC_RGB_TO_YUV_RV_GV);
-	I915_WRITE(PIPE_CSC_COEFF_BV(pipe), CSC_RGB_TO_YUV_BV);
+	I915_WRITE(PIPE_CSC_COEFF_RV_GV(pipe), coeff[6] << 16 | coeff[7]);
+	I915_WRITE(PIPE_CSC_COEFF_BV(pipe), coeff[8] << 16);
 
-	I915_WRITE(PIPE_CSC_POSTOFF_HI(pipe), POSTOFF_RGB_TO_YUV_HI);
-	I915_WRITE(PIPE_CSC_POSTOFF_ME(pipe), POSTOFF_RGB_TO_YUV_ME);
-	I915_WRITE(PIPE_CSC_POSTOFF_LO(pipe), POSTOFF_RGB_TO_YUV_LO);
-	I915_WRITE(PIPE_CSC_MODE(pipe), 0);
+	if (INTEL_GEN(dev_priv) >= 7) {
+		I915_WRITE(PIPE_CSC_POSTOFF_HI(pipe), postoff[0]);
+		I915_WRITE(PIPE_CSC_POSTOFF_ME(pipe), postoff[1]);
+		I915_WRITE(PIPE_CSC_POSTOFF_LO(pipe), postoff[2]);
+	}
 }
 
-static void ilk_load_csc_matrix(const struct intel_crtc_state *crtc_state)
+static void icl_update_output_csc(struct intel_crtc *crtc,
+				  const u16 preoff[3],
+				  const u16 coeff[9],
+				  const u16 postoff[3])
 {
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
-	bool limited_color_range = false;
 	enum pipe pipe = crtc->pipe;
-	u16 coeffs[9] = {};
-	int i;
+
+	I915_WRITE(PIPE_CSC_OUTPUT_PREOFF_HI(pipe), preoff[0]);
+	I915_WRITE(PIPE_CSC_OUTPUT_PREOFF_ME(pipe), preoff[1]);
+	I915_WRITE(PIPE_CSC_OUTPUT_PREOFF_LO(pipe), preoff[2]);
+
+	I915_WRITE(PIPE_CSC_OUTPUT_COEFF_RY_GY(pipe), coeff[0] << 16 | coeff[1]);
+	I915_WRITE(PIPE_CSC_OUTPUT_COEFF_BY(pipe), coeff[2] << 16);
+
+	I915_WRITE(PIPE_CSC_OUTPUT_COEFF_RU_GU(pipe), coeff[3] << 16 | coeff[4]);
+	I915_WRITE(PIPE_CSC_OUTPUT_COEFF_BU(pipe), coeff[5] << 16);
+
+	I915_WRITE(PIPE_CSC_OUTPUT_COEFF_RV_GV(pipe), coeff[6] << 16 | coeff[7]);
+	I915_WRITE(PIPE_CSC_OUTPUT_COEFF_BV(pipe), coeff[8] << 16);
+
+	I915_WRITE(PIPE_CSC_OUTPUT_POSTOFF_HI(pipe), postoff[0]);
+	I915_WRITE(PIPE_CSC_OUTPUT_POSTOFF_ME(pipe), postoff[1]);
+	I915_WRITE(PIPE_CSC_OUTPUT_POSTOFF_LO(pipe), postoff[2]);
+}
+
+static bool ilk_csc_limited_range(const struct intel_crtc_state *crtc_state)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc_state->base.crtc->dev);
 
 	/*
 	 * FIXME if there's a gamma LUT after the CSC, we should
 	 * do the range compression using the gamma LUT instead.
 	 */
-	if (INTEL_GEN(dev_priv) >= 8 || IS_HASWELL(dev_priv))
-		limited_color_range = crtc_state->limited_color_range;
+	return crtc_state->limited_color_range &&
+		(IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv) ||
+		 IS_GEN_RANGE(dev_priv, 9, 10));
+}
 
-	if (crtc_state->output_format == INTEL_OUTPUT_FORMAT_YCBCR420 ||
-	    crtc_state->output_format == INTEL_OUTPUT_FORMAT_YCBCR444) {
-		ilk_load_ycbcr_conversion_matrix(crtc);
-		return;
-	} else if (crtc_state->base.ctm) {
-		struct drm_color_ctm *ctm = crtc_state->base.ctm->data;
-		const u64 *input;
-		u64 temp[9];
+static void ilk_csc_convert_ctm(const struct intel_crtc_state *crtc_state,
+				u16 coeffs[9])
+{
+	const struct drm_color_ctm *ctm = crtc_state->base.ctm->data;
+	const u64 *input;
+	u64 temp[9];
+	int i;
 
-		if (limited_color_range)
-			input = ctm_mult_by_limited(temp, ctm->matrix);
-		else
-			input = ctm->matrix;
+	if (ilk_csc_limited_range(crtc_state))
+		input = ctm_mult_by_limited(temp, ctm->matrix);
+	else
+		input = ctm->matrix;
+
+	/*
+	 * Convert fixed point S31.32 input to format supported by the
+	 * hardware.
+	 */
+	for (i = 0; i < 9; i++) {
+		u64 abs_coeff = ((1ULL << 63) - 1) & input[i];
 
 		/*
-		 * Convert fixed point S31.32 input to format supported by the
+		 * Clamp input value to min/max supported by
 		 * hardware.
 		 */
-		for (i = 0; i < ARRAY_SIZE(coeffs); i++) {
-			u64 abs_coeff = ((1ULL << 63) - 1) & input[i];
-
-			/*
-			 * Clamp input value to min/max supported by
-			 * hardware.
-			 */
-			abs_coeff = clamp_val(abs_coeff, 0, CTM_COEFF_4_0 - 1);
-
-			/* sign bit */
-			if (CTM_COEFF_NEGATIVE(input[i]))
-				coeffs[i] |= 1 << 15;
-
-			if (abs_coeff < CTM_COEFF_0_125)
-				coeffs[i] |= (3 << 12) |
-					ILK_CSC_COEFF_FP(abs_coeff, 12);
-			else if (abs_coeff < CTM_COEFF_0_25)
-				coeffs[i] |= (2 << 12) |
-					ILK_CSC_COEFF_FP(abs_coeff, 11);
-			else if (abs_coeff < CTM_COEFF_0_5)
-				coeffs[i] |= (1 << 12) |
-					ILK_CSC_COEFF_FP(abs_coeff, 10);
-			else if (abs_coeff < CTM_COEFF_1_0)
-				coeffs[i] |= ILK_CSC_COEFF_FP(abs_coeff, 9);
-			else if (abs_coeff < CTM_COEFF_2_0)
-				coeffs[i] |= (7 << 12) |
-					ILK_CSC_COEFF_FP(abs_coeff, 8);
-			else
-				coeffs[i] |= (6 << 12) |
-					ILK_CSC_COEFF_FP(abs_coeff, 7);
-		}
-	} else {
-		/*
-		 * Load an identity matrix if no coefficients are provided.
-		 *
-		 * TODO: Check what kind of values actually come out of the
-		 * pipe with these coeff/postoff values and adjust to get the
-		 * best accuracy. Perhaps we even need to take the bpc value
-		 * into consideration.
-		 */
-		for (i = 0; i < 3; i++) {
-			if (limited_color_range)
-				coeffs[i * 3 + i] =
-					ILK_CSC_COEFF_LIMITED_RANGE;
-			else
-				coeffs[i * 3 + i] = ILK_CSC_COEFF_1_0;
-		}
+		abs_coeff = clamp_val(abs_coeff, 0, CTM_COEFF_4_0 - 1);
+
+		coeffs[i] = 0;
+
+		/* sign bit */
+		if (CTM_COEFF_NEGATIVE(input[i]))
+			coeffs[i] |= 1 << 15;
+
+		if (abs_coeff < CTM_COEFF_0_125)
+			coeffs[i] |= (3 << 12) |
+				ILK_CSC_COEFF_FP(abs_coeff, 12);
+		else if (abs_coeff < CTM_COEFF_0_25)
+			coeffs[i] |= (2 << 12) |
+				ILK_CSC_COEFF_FP(abs_coeff, 11);
+		else if (abs_coeff < CTM_COEFF_0_5)
+			coeffs[i] |= (1 << 12) |
+				ILK_CSC_COEFF_FP(abs_coeff, 10);
+		else if (abs_coeff < CTM_COEFF_1_0)
+			coeffs[i] |= ILK_CSC_COEFF_FP(abs_coeff, 9);
+		else if (abs_coeff < CTM_COEFF_2_0)
+			coeffs[i] |= (7 << 12) |
+				ILK_CSC_COEFF_FP(abs_coeff, 8);
+		else
+			coeffs[i] |= (6 << 12) |
+				ILK_CSC_COEFF_FP(abs_coeff, 7);
 	}
+}
 
-	I915_WRITE(PIPE_CSC_COEFF_RY_GY(pipe), coeffs[0] << 16 | coeffs[1]);
-	I915_WRITE(PIPE_CSC_COEFF_BY(pipe), coeffs[2] << 16);
-
-	I915_WRITE(PIPE_CSC_COEFF_RU_GU(pipe), coeffs[3] << 16 | coeffs[4]);
-	I915_WRITE(PIPE_CSC_COEFF_BU(pipe), coeffs[5] << 16);
-
-	I915_WRITE(PIPE_CSC_COEFF_RV_GV(pipe), coeffs[6] << 16 | coeffs[7]);
-	I915_WRITE(PIPE_CSC_COEFF_BV(pipe), coeffs[8] << 16);
+static void ilk_load_csc_matrix(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	bool limited_color_range = ilk_csc_limited_range(crtc_state);
 
-	I915_WRITE(PIPE_CSC_PREOFF_HI(pipe), 0);
-	I915_WRITE(PIPE_CSC_PREOFF_ME(pipe), 0);
-	I915_WRITE(PIPE_CSC_PREOFF_LO(pipe), 0);
+	if (crtc_state->base.ctm) {
+		u16 coeff[9];
+
+		ilk_csc_convert_ctm(crtc_state, coeff);
+		ilk_update_pipe_csc(crtc, ilk_csc_off_zero, coeff,
+				    limited_color_range ?
+				    ilk_csc_postoff_limited_range :
+				    ilk_csc_off_zero);
+	} else if (crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB) {
+		ilk_update_pipe_csc(crtc, ilk_csc_off_zero,
+				    ilk_csc_coeff_rgb_to_ycbcr,
+				    ilk_csc_postoff_rgb_to_ycbcr);
+	} else if (limited_color_range) {
+		ilk_update_pipe_csc(crtc, ilk_csc_off_zero,
+				    ilk_csc_coeff_limited_range,
+				    ilk_csc_postoff_limited_range);
+	} else if (crtc_state->csc_enable) {
+		/*
+		 * On GLK+ both pipe CSC and degamma LUT are controlled
+		 * by csc_enable. Hence for the cases where the degama
+		 * LUT is needed but CSC is not we need to load an
+		 * identity matrix.
+		 */
+		WARN_ON(!IS_CANNONLAKE(dev_priv) && !IS_GEMINILAKE(dev_priv));
 
-	if (INTEL_GEN(dev_priv) > 6) {
-		u16 postoff = 0;
+		ilk_update_pipe_csc(crtc, ilk_csc_off_zero,
+				    ilk_csc_coeff_identity,
+				    ilk_csc_off_zero);
+	}
 
-		if (limited_color_range)
-			postoff = (16 * (1 << 12) / 255) & 0x1fff;
+	I915_WRITE(PIPE_CSC_MODE(crtc->pipe), crtc_state->csc_mode);
+}
 
-		I915_WRITE(PIPE_CSC_POSTOFF_HI(pipe), postoff);
-		I915_WRITE(PIPE_CSC_POSTOFF_ME(pipe), postoff);
-		I915_WRITE(PIPE_CSC_POSTOFF_LO(pipe), postoff);
+static void icl_load_csc_matrix(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 
-		I915_WRITE(PIPE_CSC_MODE(pipe), 0);
-	} else {
-		u32 mode = CSC_MODE_YUV_TO_RGB;
+	if (crtc_state->base.ctm) {
+		u16 coeff[9];
 
-		if (limited_color_range)
-			mode |= CSC_BLACK_SCREEN_OFFSET;
+		ilk_csc_convert_ctm(crtc_state, coeff);
+		ilk_update_pipe_csc(crtc, ilk_csc_off_zero,
+				    coeff, ilk_csc_off_zero);
+	}
 
-		I915_WRITE(PIPE_CSC_MODE(pipe), mode);
+	if (crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB) {
+		icl_update_output_csc(crtc, ilk_csc_off_zero,
+				      ilk_csc_coeff_rgb_to_ycbcr,
+				      ilk_csc_postoff_rgb_to_ycbcr);
+	} else if (crtc_state->limited_color_range) {
+		icl_update_output_csc(crtc, ilk_csc_off_zero,
+				      ilk_csc_coeff_limited_range,
+				      ilk_csc_postoff_limited_range);
 	}
+
+	I915_WRITE(PIPE_CSC_MODE(crtc->pipe), crtc_state->csc_mode);
 }
 
 /*
@@ -262,7 +324,6 @@ static void cherryview_load_csc_matrix(const struct intel_crtc_state *crtc_state
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
-	u32 mode;
 
 	if (crtc_state->base.ctm) {
 		const struct drm_color_ctm *ctm = crtc_state->base.ctm->data;
@@ -296,12 +357,30 @@ static void cherryview_load_csc_matrix(const struct intel_crtc_state *crtc_state
 		I915_WRITE(CGM_PIPE_CSC_COEFF8(pipe), coeffs[8]);
 	}
 
-	mode = (crtc_state->base.ctm ? CGM_PIPE_MODE_CSC : 0);
-	if (!crtc_state_is_legacy_gamma(crtc_state)) {
-		mode |= (crtc_state->base.degamma_lut ? CGM_PIPE_MODE_DEGAMMA : 0) |
-			(crtc_state->base.gamma_lut ? CGM_PIPE_MODE_GAMMA : 0);
-	}
-	I915_WRITE(CGM_PIPE_MODE(pipe), mode);
+	I915_WRITE(CGM_PIPE_MODE(pipe), crtc_state->cgm_mode);
+}
+
+/* i965+ "10.6" bit interpolated format "even DW" (low 8 bits) */
+static u32 i965_lut_10p6_ldw(const struct drm_color_lut *color)
+{
+	return (color->red & 0xff) << 16 |
+		(color->green & 0xff) << 8 |
+		(color->blue & 0xff);
+}
+
+/* i965+ "10.6" interpolated format "odd DW" (high 8 bits) */
+static u32 i965_lut_10p6_udw(const struct drm_color_lut *color)
+{
+	return (color->red >> 8) << 16 |
+		(color->green >> 8) << 8 |
+		(color->blue >> 8);
+}
+
+static u32 ilk_lut_10(const struct drm_color_lut *color)
+{
+	return drm_color_lut_extract(color->red, 10) << 20 |
+		drm_color_lut_extract(color->green, 10) << 10 |
+		drm_color_lut_extract(color->blue, 10);
 }
 
 /* Loads the legacy palette/gamma unit for the CRTC. */
@@ -334,15 +413,6 @@ static void i9xx_load_luts_internal(const struct intel_crtc_state *crtc_state,
 			else
 				I915_WRITE(LGC_PALETTE(pipe, i), word);
 		}
-	} else {
-		for (i = 0; i < 256; i++) {
-			u32 word = (i << 16) | (i << 8) | i;
-
-			if (HAS_GMCH(dev_priv))
-				I915_WRITE(PALETTE(pipe, i), word);
-			else
-				I915_WRITE(LGC_PALETTE(pipe, i), word);
-		}
 	}
 }
 
@@ -351,6 +421,34 @@ static void i9xx_load_luts(const struct intel_crtc_state *crtc_state)
 	i9xx_load_luts_internal(crtc_state, crtc_state->base.gamma_lut);
 }
 
+static void i9xx_color_commit(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	enum pipe pipe = crtc->pipe;
+	u32 val;
+
+	val = I915_READ(PIPECONF(pipe));
+	val &= ~PIPECONF_GAMMA_MODE_MASK_I9XX;
+	val |= PIPECONF_GAMMA_MODE(crtc_state->gamma_mode);
+	I915_WRITE(PIPECONF(pipe), val);
+}
+
+static void ilk_color_commit(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	enum pipe pipe = crtc->pipe;
+	u32 val;
+
+	val = I915_READ(PIPECONF(pipe));
+	val &= ~PIPECONF_GAMMA_MODE_MASK_ILK;
+	val |= PIPECONF_GAMMA_MODE(crtc_state->gamma_mode);
+	I915_WRITE(PIPECONF(pipe), val);
+
+	ilk_load_csc_matrix(crtc_state);
+}
+
 static void hsw_color_commit(const struct intel_crtc_state *crtc_state)
 {
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
@@ -361,106 +459,219 @@ static void hsw_color_commit(const struct intel_crtc_state *crtc_state)
 	ilk_load_csc_matrix(crtc_state);
 }
 
-static void bdw_load_degamma_lut(const struct intel_crtc_state *crtc_state)
+static void skl_color_commit(const struct intel_crtc_state *crtc_state)
 {
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
-	const struct drm_property_blob *degamma_lut = crtc_state->base.degamma_lut;
-	u32 i, lut_size = INTEL_INFO(dev_priv)->color.degamma_lut_size;
 	enum pipe pipe = crtc->pipe;
+	u32 val = 0;
 
-	I915_WRITE(PREC_PAL_INDEX(pipe),
-		   PAL_PREC_SPLIT_MODE | PAL_PREC_AUTO_INCREMENT);
+	/*
+	 * We don't (yet) allow userspace to control the pipe background color,
+	 * so force it to black, but apply pipe gamma and CSC appropriately
+	 * so that its handling will match how we program our planes.
+	 */
+	if (crtc_state->gamma_enable)
+		val |= SKL_BOTTOM_COLOR_GAMMA_ENABLE;
+	if (crtc_state->csc_enable)
+		val |= SKL_BOTTOM_COLOR_CSC_ENABLE;
+	I915_WRITE(SKL_BOTTOM_COLOR(pipe), val);
 
-	if (degamma_lut) {
-		const struct drm_color_lut *lut = degamma_lut->data;
+	I915_WRITE(GAMMA_MODE(crtc->pipe), crtc_state->gamma_mode);
 
-		for (i = 0; i < lut_size; i++) {
-			u32 word =
-			drm_color_lut_extract(lut[i].red, 10) << 20 |
-			drm_color_lut_extract(lut[i].green, 10) << 10 |
-			drm_color_lut_extract(lut[i].blue, 10);
+	if (INTEL_GEN(dev_priv) >= 11)
+		icl_load_csc_matrix(crtc_state);
+	else
+		ilk_load_csc_matrix(crtc_state);
+}
 
-			I915_WRITE(PREC_PAL_DATA(pipe), word);
-		}
-	} else {
-		for (i = 0; i < lut_size; i++) {
-			u32 v = (i * ((1 << 10) - 1)) / (lut_size - 1);
+static void i965_load_lut_10p6(struct intel_crtc *crtc,
+			       const struct drm_property_blob *blob)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	const struct drm_color_lut *lut = blob->data;
+	int i, lut_size = drm_color_lut_size(blob);
+	enum pipe pipe = crtc->pipe;
 
-			I915_WRITE(PREC_PAL_DATA(pipe),
-				   (v << 20) | (v << 10) | v);
-		}
+	for (i = 0; i < lut_size - 1; i++) {
+		I915_WRITE(PALETTE(pipe, 2 * i + 0),
+			   i965_lut_10p6_ldw(&lut[i]));
+		I915_WRITE(PALETTE(pipe, 2 * i + 1),
+			   i965_lut_10p6_udw(&lut[i]));
 	}
+
+	I915_WRITE(PIPEGCMAX(pipe, 0), lut[i].red);
+	I915_WRITE(PIPEGCMAX(pipe, 1), lut[i].green);
+	I915_WRITE(PIPEGCMAX(pipe, 2), lut[i].blue);
 }
 
-static void bdw_load_gamma_lut(const struct intel_crtc_state *crtc_state, u32 offset)
+static void i965_load_luts(const struct intel_crtc_state *crtc_state)
 {
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	const struct drm_property_blob *gamma_lut = crtc_state->base.gamma_lut;
-	u32 i, lut_size = INTEL_INFO(dev_priv)->color.gamma_lut_size;
+
+	if (crtc_state->gamma_mode == GAMMA_MODE_MODE_8BIT)
+		i9xx_load_luts(crtc_state);
+	else
+		i965_load_lut_10p6(crtc, gamma_lut);
+}
+
+static void ilk_load_lut_10(struct intel_crtc *crtc,
+			    const struct drm_property_blob *blob)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	const struct drm_color_lut *lut = blob->data;
+	int i, lut_size = drm_color_lut_size(blob);
 	enum pipe pipe = crtc->pipe;
 
-	WARN_ON(offset & ~PAL_PREC_INDEX_VALUE_MASK);
+	for (i = 0; i < lut_size; i++)
+		I915_WRITE(PREC_PALETTE(pipe, i), ilk_lut_10(&lut[i]));
+}
 
-	I915_WRITE(PREC_PAL_INDEX(pipe),
-		   (offset ? PAL_PREC_SPLIT_MODE : 0) |
-		   PAL_PREC_AUTO_INCREMENT |
-		   offset);
+static void ilk_load_luts(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
+	const struct drm_property_blob *gamma_lut = crtc_state->base.gamma_lut;
 
-	if (gamma_lut) {
-		const struct drm_color_lut *lut = gamma_lut->data;
+	if (crtc_state->gamma_mode == GAMMA_MODE_MODE_8BIT)
+		i9xx_load_luts(crtc_state);
+	else
+		ilk_load_lut_10(crtc, gamma_lut);
+}
 
-		for (i = 0; i < lut_size; i++) {
-			u32 word =
-			(drm_color_lut_extract(lut[i].red, 10) << 20) |
-			(drm_color_lut_extract(lut[i].green, 10) << 10) |
-			drm_color_lut_extract(lut[i].blue, 10);
+static int ivb_lut_10_size(u32 prec_index)
+{
+	if (prec_index & PAL_PREC_SPLIT_MODE)
+		return 512;
+	else
+		return 1024;
+}
 
-			I915_WRITE(PREC_PAL_DATA(pipe), word);
-		}
+/*
+ * IVB/HSW Bspec / PAL_PREC_INDEX:
+ * "Restriction : Index auto increment mode is not
+ *  supported and must not be enabled."
+ */
+static void ivb_load_lut_10(struct intel_crtc *crtc,
+			    const struct drm_property_blob *blob,
+			    u32 prec_index)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	int hw_lut_size = ivb_lut_10_size(prec_index);
+	const struct drm_color_lut *lut = blob->data;
+	int i, lut_size = drm_color_lut_size(blob);
+	enum pipe pipe = crtc->pipe;
 
-		/* Program the max register to clamp values > 1.0. */
-		i = lut_size - 1;
-		I915_WRITE(PREC_PAL_GC_MAX(pipe, 0),
-			   drm_color_lut_extract(lut[i].red, 16));
-		I915_WRITE(PREC_PAL_GC_MAX(pipe, 1),
-			   drm_color_lut_extract(lut[i].green, 16));
-		I915_WRITE(PREC_PAL_GC_MAX(pipe, 2),
-			   drm_color_lut_extract(lut[i].blue, 16));
-	} else {
-		for (i = 0; i < lut_size; i++) {
-			u32 v = (i * ((1 << 10) - 1)) / (lut_size - 1);
+	for (i = 0; i < hw_lut_size; i++) {
+		/* We discard half the user entries in split gamma mode */
+		const struct drm_color_lut *entry =
+			&lut[i * (lut_size - 1) / (hw_lut_size - 1)];
 
-			I915_WRITE(PREC_PAL_DATA(pipe),
-				   (v << 20) | (v << 10) | v);
-		}
+		I915_WRITE(PREC_PAL_INDEX(pipe), prec_index++);
+		I915_WRITE(PREC_PAL_DATA(pipe), ilk_lut_10(entry));
+	}
+
+	/*
+	 * Reset the index, otherwise it prevents the legacy palette to be
+	 * written properly.
+	 */
+	I915_WRITE(PREC_PAL_INDEX(pipe), 0);
+}
 
-		I915_WRITE(PREC_PAL_GC_MAX(pipe, 0), (1 << 16) - 1);
-		I915_WRITE(PREC_PAL_GC_MAX(pipe, 1), (1 << 16) - 1);
-		I915_WRITE(PREC_PAL_GC_MAX(pipe, 2), (1 << 16) - 1);
+/* On BDW+ the index auto increment mode actually works */
+static void bdw_load_lut_10(struct intel_crtc *crtc,
+			    const struct drm_property_blob *blob,
+			    u32 prec_index)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	int hw_lut_size = ivb_lut_10_size(prec_index);
+	const struct drm_color_lut *lut = blob->data;
+	int i, lut_size = drm_color_lut_size(blob);
+	enum pipe pipe = crtc->pipe;
+
+	I915_WRITE(PREC_PAL_INDEX(pipe), prec_index |
+		   PAL_PREC_AUTO_INCREMENT);
+
+	for (i = 0; i < hw_lut_size; i++) {
+		/* We discard half the user entries in split gamma mode */
+		const struct drm_color_lut *entry =
+			&lut[i * (lut_size - 1) / (hw_lut_size - 1)];
+
+		I915_WRITE(PREC_PAL_DATA(pipe), ilk_lut_10(entry));
 	}
+
+	/*
+	 * Reset the index, otherwise it prevents the legacy palette to be
+	 * written properly.
+	 */
+	I915_WRITE(PREC_PAL_INDEX(pipe), 0);
 }
 
-/* Loads the palette/gamma unit for the CRTC on Broadwell+. */
-static void broadwell_load_luts(const struct intel_crtc_state *crtc_state)
+static void ivb_load_lut_10_max(struct intel_crtc *crtc)
 {
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
 
-	if (crtc_state_is_legacy_gamma(crtc_state)) {
+	/* Program the max register to clamp values > 1.0. */
+	I915_WRITE(PREC_PAL_EXT_GC_MAX(pipe, 0), 1 << 16);
+	I915_WRITE(PREC_PAL_EXT_GC_MAX(pipe, 1), 1 << 16);
+	I915_WRITE(PREC_PAL_EXT_GC_MAX(pipe, 2), 1 << 16);
+
+	/*
+	 * Program the gc max 2 register to clamp values > 1.0.
+	 * ToDo: Extend the ABI to be able to program values
+	 * from 3.0 to 7.0
+	 */
+	if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv)) {
+		I915_WRITE(PREC_PAL_EXT2_GC_MAX(pipe, 0), 1 << 16);
+		I915_WRITE(PREC_PAL_EXT2_GC_MAX(pipe, 1), 1 << 16);
+		I915_WRITE(PREC_PAL_EXT2_GC_MAX(pipe, 2), 1 << 16);
+	}
+}
+
+static void ivb_load_luts(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
+	const struct drm_property_blob *gamma_lut = crtc_state->base.gamma_lut;
+	const struct drm_property_blob *degamma_lut = crtc_state->base.degamma_lut;
+
+	if (crtc_state->gamma_mode == GAMMA_MODE_MODE_8BIT) {
 		i9xx_load_luts(crtc_state);
+	} else if (crtc_state->gamma_mode == GAMMA_MODE_MODE_SPLIT) {
+		ivb_load_lut_10(crtc, degamma_lut, PAL_PREC_SPLIT_MODE |
+				PAL_PREC_INDEX_VALUE(0));
+		ivb_load_lut_10_max(crtc);
+		ivb_load_lut_10(crtc, gamma_lut, PAL_PREC_SPLIT_MODE |
+				PAL_PREC_INDEX_VALUE(512));
 	} else {
-		bdw_load_degamma_lut(crtc_state);
-		bdw_load_gamma_lut(crtc_state,
-				   INTEL_INFO(dev_priv)->color.degamma_lut_size);
+		const struct drm_property_blob *blob = gamma_lut ?: degamma_lut;
 
-		/*
-		 * Reset the index, otherwise it prevents the legacy palette to be
-		 * written properly.
-		 */
-		I915_WRITE(PREC_PAL_INDEX(pipe), 0);
+		ivb_load_lut_10(crtc, blob,
+				PAL_PREC_INDEX_VALUE(0));
+		ivb_load_lut_10_max(crtc);
+	}
+}
+
+static void bdw_load_luts(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
+	const struct drm_property_blob *gamma_lut = crtc_state->base.gamma_lut;
+	const struct drm_property_blob *degamma_lut = crtc_state->base.degamma_lut;
+
+	if (crtc_state->gamma_mode == GAMMA_MODE_MODE_8BIT) {
+		i9xx_load_luts(crtc_state);
+	} else if (crtc_state->gamma_mode == GAMMA_MODE_MODE_SPLIT) {
+		bdw_load_lut_10(crtc, degamma_lut, PAL_PREC_SPLIT_MODE |
+				PAL_PREC_INDEX_VALUE(0));
+		ivb_load_lut_10_max(crtc);
+		bdw_load_lut_10(crtc, gamma_lut, PAL_PREC_SPLIT_MODE |
+				PAL_PREC_INDEX_VALUE(512));
+	} else {
+		const struct drm_property_blob *blob = gamma_lut ?: degamma_lut;
+
+		bdw_load_lut_10(crtc, blob,
+				PAL_PREC_INDEX_VALUE(0));
+		ivb_load_lut_10_max(crtc);
 	}
 }
 
@@ -469,7 +680,8 @@ static void glk_load_degamma_lut(const struct intel_crtc_state *crtc_state)
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
-	const u32 lut_size = 33;
+	const u32 lut_size = INTEL_INFO(dev_priv)->color.degamma_lut_size;
+	const struct drm_color_lut *lut = crtc_state->base.degamma_lut->data;
 	u32 i;
 
 	/*
@@ -480,39 +692,95 @@ static void glk_load_degamma_lut(const struct intel_crtc_state *crtc_state)
 	I915_WRITE(PRE_CSC_GAMC_INDEX(pipe), 0);
 	I915_WRITE(PRE_CSC_GAMC_INDEX(pipe), PRE_CSC_GAMC_AUTO_INCREMENT);
 
+	for (i = 0; i < lut_size; i++) {
+		/*
+		 * First 33 entries represent range from 0 to 1.0
+		 * 34th and 35th entry will represent extended range
+		 * inputs 3.0 and 7.0 respectively, currently clamped
+		 * at 1.0. Since the precision is 16bit, the user
+		 * value can be directly filled to register.
+		 * The pipe degamma table in GLK+ onwards doesn't
+		 * support different values per channel, so this just
+		 * programs green value which will be equal to Red and
+		 * Blue into the lut registers.
+		 * ToDo: Extend to max 7.0. Enable 32 bit input value
+		 * as compared to just 16 to achieve this.
+		 */
+		I915_WRITE(PRE_CSC_GAMC_DATA(pipe), lut[i].green);
+	}
+
+	/* Clamp values > 1.0. */
+	while (i++ < 35)
+		I915_WRITE(PRE_CSC_GAMC_DATA(pipe), 1 << 16);
+}
+
+static void glk_load_degamma_lut_linear(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	enum pipe pipe = crtc->pipe;
+	const u32 lut_size = INTEL_INFO(dev_priv)->color.degamma_lut_size;
+	u32 i;
+
 	/*
-	 *  FIXME: The pipe degamma table in geminilake doesn't support
-	 *  different values per channel, so this just loads a linear table.
+	 * When setting the auto-increment bit, the hardware seems to
+	 * ignore the index bits, so we need to reset it to index 0
+	 * separately.
 	 */
+	I915_WRITE(PRE_CSC_GAMC_INDEX(pipe), 0);
+	I915_WRITE(PRE_CSC_GAMC_INDEX(pipe), PRE_CSC_GAMC_AUTO_INCREMENT);
+
 	for (i = 0; i < lut_size; i++) {
-		u32 v = (i * (1 << 16)) / (lut_size - 1);
+		u32 v = (i << 16) / (lut_size - 1);
 
 		I915_WRITE(PRE_CSC_GAMC_DATA(pipe), v);
 	}
 
 	/* Clamp values > 1.0. */
 	while (i++ < 35)
-		I915_WRITE(PRE_CSC_GAMC_DATA(pipe), (1 << 16));
+		I915_WRITE(PRE_CSC_GAMC_DATA(pipe), 1 << 16);
 }
 
 static void glk_load_luts(const struct intel_crtc_state *crtc_state)
 {
+	const struct drm_property_blob *gamma_lut = crtc_state->base.gamma_lut;
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
-	enum pipe pipe = crtc->pipe;
 
-	glk_load_degamma_lut(crtc_state);
+	/*
+	 * On GLK+ both pipe CSC and degamma LUT are controlled
+	 * by csc_enable. Hence for the cases where the CSC is
+	 * needed but degamma LUT is not we need to load a
+	 * linear degamma LUT. In fact we'll just always load
+	 * the degama LUT so that we don't have to reload
+	 * it every time the pipe CSC is being enabled.
+	 */
+	if (crtc_state->base.degamma_lut)
+		glk_load_degamma_lut(crtc_state);
+	else
+		glk_load_degamma_lut_linear(crtc_state);
 
-	if (crtc_state_is_legacy_gamma(crtc_state)) {
+	if (crtc_state->gamma_mode == GAMMA_MODE_MODE_8BIT) {
 		i9xx_load_luts(crtc_state);
 	} else {
-		bdw_load_gamma_lut(crtc_state, 0);
+		bdw_load_lut_10(crtc, gamma_lut, PAL_PREC_INDEX_VALUE(0));
+		ivb_load_lut_10_max(crtc);
+	}
+}
 
-		/*
-		 * Reset the index, otherwise it prevents the legacy palette to be
-		 * written properly.
-		 */
-		I915_WRITE(PREC_PAL_INDEX(pipe), 0);
+static void icl_load_luts(const struct intel_crtc_state *crtc_state)
+{
+	const struct drm_property_blob *gamma_lut = crtc_state->base.gamma_lut;
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
+
+	if (crtc_state->base.degamma_lut)
+		glk_load_degamma_lut(crtc_state);
+
+	if ((crtc_state->gamma_mode & GAMMA_MODE_MODE_MASK) ==
+	    GAMMA_MODE_MODE_8BIT) {
+		i9xx_load_luts(crtc_state);
+	} else {
+		bdw_load_lut_10(crtc, gamma_lut, PAL_PREC_INDEX_VALUE(0));
+		ivb_load_lut_10_max(crtc);
 	}
 }
 
@@ -527,7 +795,7 @@ static void cherryview_load_luts(const struct intel_crtc_state *crtc_state)
 	cherryview_load_csc_matrix(crtc_state);
 
 	if (crtc_state_is_legacy_gamma(crtc_state)) {
-		i9xx_load_luts_internal(crtc_state, gamma_lut);
+		i9xx_load_luts(crtc_state);
 		return;
 	}
 
@@ -566,12 +834,6 @@ static void cherryview_load_luts(const struct intel_crtc_state *crtc_state)
 			I915_WRITE(CGM_PIPE_GAMMA(pipe, i, 1), word1);
 		}
 	}
-
-	/*
-	 * Also program a linear LUT in the legacy block (behind the
-	 * CGM block).
-	 */
-	i9xx_load_luts_internal(crtc_state, NULL);
 }
 
 void intel_color_load_luts(const struct intel_crtc_state *crtc_state)
@@ -585,8 +847,64 @@ void intel_color_commit(const struct intel_crtc_state *crtc_state)
 {
 	struct drm_i915_private *dev_priv = to_i915(crtc_state->base.crtc->dev);
 
-	if (dev_priv->display.color_commit)
-		dev_priv->display.color_commit(crtc_state);
+	dev_priv->display.color_commit(crtc_state);
+}
+
+int intel_color_check(struct intel_crtc_state *crtc_state)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc_state->base.crtc->dev);
+
+	return dev_priv->display.color_check(crtc_state);
+}
+
+static bool need_plane_update(struct intel_plane *plane,
+			      const struct intel_crtc_state *crtc_state)
+{
+	struct drm_i915_private *dev_priv = to_i915(plane->base.dev);
+
+	/*
+	 * On pre-SKL the pipe gamma enable and pipe csc enable for
+	 * the pipe bottom color are configured via the primary plane.
+	 * We have to reconfigure that even if the plane is inactive.
+	 */
+	return crtc_state->active_planes & BIT(plane->id) ||
+		(INTEL_GEN(dev_priv) < 9 &&
+		 plane->id == PLANE_PRIMARY);
+}
+
+static int
+intel_color_add_affected_planes(struct intel_crtc_state *new_crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(new_crtc_state->base.crtc);
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_atomic_state *state =
+		to_intel_atomic_state(new_crtc_state->base.state);
+	const struct intel_crtc_state *old_crtc_state =
+		intel_atomic_get_old_crtc_state(state, crtc);
+	struct intel_plane *plane;
+
+	if (!new_crtc_state->base.active ||
+	    drm_atomic_crtc_needs_modeset(&new_crtc_state->base))
+		return 0;
+
+	if (new_crtc_state->gamma_enable == old_crtc_state->gamma_enable &&
+	    new_crtc_state->csc_enable == old_crtc_state->csc_enable)
+		return 0;
+
+	for_each_intel_plane_on_crtc(&dev_priv->drm, crtc, plane) {
+		struct intel_plane_state *plane_state;
+
+		if (!need_plane_update(plane, new_crtc_state))
+			continue;
+
+		plane_state = intel_atomic_get_plane_state(state, plane);
+		if (IS_ERR(plane_state))
+			return PTR_ERR(plane_state);
+
+		new_crtc_state->update_planes |= BIT(plane->id);
+	}
+
+	return 0;
 }
 
 static int check_lut_size(const struct drm_property_blob *lut, int expected)
@@ -606,7 +924,7 @@ static int check_lut_size(const struct drm_property_blob *lut, int expected)
 	return 0;
 }
 
-int intel_color_check(struct intel_crtc_state *crtc_state)
+static int check_luts(const struct intel_crtc_state *crtc_state)
 {
 	struct drm_i915_private *dev_priv = to_i915(crtc_state->base.crtc->dev);
 	const struct drm_property_blob *gamma_lut = crtc_state->base.gamma_lut;
@@ -614,17 +932,19 @@ int intel_color_check(struct intel_crtc_state *crtc_state)
 	int gamma_length, degamma_length;
 	u32 gamma_tests, degamma_tests;
 
+	/* Always allow legacy gamma LUT with no further checking. */
+	if (crtc_state_is_legacy_gamma(crtc_state))
+		return 0;
+
+	/* C8 relies on its palette being stored in the legacy LUT */
+	if (crtc_state->c8_planes)
+		return -EINVAL;
+
 	degamma_length = INTEL_INFO(dev_priv)->color.degamma_lut_size;
 	gamma_length = INTEL_INFO(dev_priv)->color.gamma_lut_size;
 	degamma_tests = INTEL_INFO(dev_priv)->color.degamma_lut_tests;
 	gamma_tests = INTEL_INFO(dev_priv)->color.gamma_lut_tests;
 
-	/* Always allow legacy gamma LUT with no further checking. */
-	if (crtc_state_is_legacy_gamma(crtc_state)) {
-		crtc_state->gamma_mode = GAMMA_MODE_MODE_8BIT;
-		return 0;
-	}
-
 	if (check_lut_size(degamma_lut, degamma_length) ||
 	    check_lut_size(gamma_lut, gamma_length))
 		return -EINVAL;
@@ -633,12 +953,270 @@ int intel_color_check(struct intel_crtc_state *crtc_state)
 	    drm_color_lut_check(gamma_lut, gamma_tests))
 		return -EINVAL;
 
-	if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv))
-		crtc_state->gamma_mode = GAMMA_MODE_MODE_10BIT;
-	else if (INTEL_GEN(dev_priv) >= 9 || IS_BROADWELL(dev_priv))
-		crtc_state->gamma_mode = GAMMA_MODE_MODE_SPLIT;
+	return 0;
+}
+
+static u32 i9xx_gamma_mode(struct intel_crtc_state *crtc_state)
+{
+	if (!crtc_state->gamma_enable ||
+	    crtc_state_is_legacy_gamma(crtc_state))
+		return GAMMA_MODE_MODE_8BIT;
 	else
-		crtc_state->gamma_mode = GAMMA_MODE_MODE_8BIT;
+		return GAMMA_MODE_MODE_10BIT; /* i965+ only */
+}
+
+static int i9xx_color_check(struct intel_crtc_state *crtc_state)
+{
+	int ret;
+
+	ret = check_luts(crtc_state);
+	if (ret)
+		return ret;
+
+	crtc_state->gamma_enable =
+		crtc_state->base.gamma_lut &&
+		!crtc_state->c8_planes;
+
+	crtc_state->gamma_mode = i9xx_gamma_mode(crtc_state);
+
+	ret = intel_color_add_affected_planes(crtc_state);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static u32 chv_cgm_mode(const struct intel_crtc_state *crtc_state)
+{
+	u32 cgm_mode = 0;
+
+	if (crtc_state_is_legacy_gamma(crtc_state))
+		return 0;
+
+	if (crtc_state->base.degamma_lut)
+		cgm_mode |= CGM_PIPE_MODE_DEGAMMA;
+	if (crtc_state->base.ctm)
+		cgm_mode |= CGM_PIPE_MODE_CSC;
+	if (crtc_state->base.gamma_lut)
+		cgm_mode |= CGM_PIPE_MODE_GAMMA;
+
+	return cgm_mode;
+}
+
+/*
+ * CHV color pipeline:
+ * u0.10 -> CGM degamma -> u0.14 -> CGM csc -> u0.14 -> CGM gamma ->
+ * u0.10 -> WGC csc -> u0.10 -> pipe gamma -> u0.10
+ *
+ * We always bypass the WGC csc and use the CGM csc
+ * instead since it has degamma and better precision.
+ */
+static int chv_color_check(struct intel_crtc_state *crtc_state)
+{
+	int ret;
+
+	ret = check_luts(crtc_state);
+	if (ret)
+		return ret;
+
+	/*
+	 * Pipe gamma will be used only for the legacy LUT.
+	 * Otherwise we bypass it and use the CGM gamma instead.
+	 */
+	crtc_state->gamma_enable =
+		crtc_state_is_legacy_gamma(crtc_state) &&
+		!crtc_state->c8_planes;
+
+	crtc_state->gamma_mode = GAMMA_MODE_MODE_8BIT;
+
+	crtc_state->cgm_mode = chv_cgm_mode(crtc_state);
+
+	ret = intel_color_add_affected_planes(crtc_state);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static u32 ilk_gamma_mode(const struct intel_crtc_state *crtc_state)
+{
+	if (!crtc_state->gamma_enable ||
+	    crtc_state_is_legacy_gamma(crtc_state))
+		return GAMMA_MODE_MODE_8BIT;
+	else
+		return GAMMA_MODE_MODE_10BIT;
+}
+
+static int ilk_color_check(struct intel_crtc_state *crtc_state)
+{
+	int ret;
+
+	ret = check_luts(crtc_state);
+	if (ret)
+		return ret;
+
+	crtc_state->gamma_enable =
+		crtc_state->base.gamma_lut &&
+		!crtc_state->c8_planes;
+
+	/*
+	 * We don't expose the ctm on ilk/snb currently,
+	 * nor do we enable YCbCr output. Also RGB limited
+	 * range output is handled by the hw automagically.
+	 */
+	crtc_state->csc_enable = false;
+
+	crtc_state->gamma_mode = ilk_gamma_mode(crtc_state);
+
+	crtc_state->csc_mode = 0;
+
+	ret = intel_color_add_affected_planes(crtc_state);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static u32 ivb_gamma_mode(const struct intel_crtc_state *crtc_state)
+{
+	if (!crtc_state->gamma_enable ||
+	    crtc_state_is_legacy_gamma(crtc_state))
+		return GAMMA_MODE_MODE_8BIT;
+	else if (crtc_state->base.gamma_lut &&
+		 crtc_state->base.degamma_lut)
+		return GAMMA_MODE_MODE_SPLIT;
+	else
+		return GAMMA_MODE_MODE_10BIT;
+}
+
+static u32 ivb_csc_mode(const struct intel_crtc_state *crtc_state)
+{
+	bool limited_color_range = ilk_csc_limited_range(crtc_state);
+
+	/*
+	 * CSC comes after the LUT in degamma, RGB->YCbCr,
+	 * and RGB full->limited range mode.
+	 */
+	if (crtc_state->base.degamma_lut ||
+	    crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB ||
+	    limited_color_range)
+		return 0;
+
+	return CSC_POSITION_BEFORE_GAMMA;
+}
+
+static int ivb_color_check(struct intel_crtc_state *crtc_state)
+{
+	bool limited_color_range = ilk_csc_limited_range(crtc_state);
+	int ret;
+
+	ret = check_luts(crtc_state);
+	if (ret)
+		return ret;
+
+	crtc_state->gamma_enable =
+		(crtc_state->base.gamma_lut ||
+		 crtc_state->base.degamma_lut) &&
+		!crtc_state->c8_planes;
+
+	crtc_state->csc_enable =
+		crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB ||
+		crtc_state->base.ctm || limited_color_range;
+
+	crtc_state->gamma_mode = ivb_gamma_mode(crtc_state);
+
+	crtc_state->csc_mode = ivb_csc_mode(crtc_state);
+
+	ret = intel_color_add_affected_planes(crtc_state);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static u32 glk_gamma_mode(const struct intel_crtc_state *crtc_state)
+{
+	if (!crtc_state->gamma_enable ||
+	    crtc_state_is_legacy_gamma(crtc_state))
+		return GAMMA_MODE_MODE_8BIT;
+	else
+		return GAMMA_MODE_MODE_10BIT;
+}
+
+static int glk_color_check(struct intel_crtc_state *crtc_state)
+{
+	int ret;
+
+	ret = check_luts(crtc_state);
+	if (ret)
+		return ret;
+
+	crtc_state->gamma_enable =
+		crtc_state->base.gamma_lut &&
+		!crtc_state->c8_planes;
+
+	/* On GLK+ degamma LUT is controlled by csc_enable */
+	crtc_state->csc_enable =
+		crtc_state->base.degamma_lut ||
+		crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB ||
+		crtc_state->base.ctm || crtc_state->limited_color_range;
+
+	crtc_state->gamma_mode = glk_gamma_mode(crtc_state);
+
+	crtc_state->csc_mode = 0;
+
+	ret = intel_color_add_affected_planes(crtc_state);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static u32 icl_gamma_mode(const struct intel_crtc_state *crtc_state)
+{
+	u32 gamma_mode = 0;
+
+	if (crtc_state->base.degamma_lut)
+		gamma_mode |= PRE_CSC_GAMMA_ENABLE;
+
+	if (crtc_state->base.gamma_lut &&
+	    !crtc_state->c8_planes)
+		gamma_mode |= POST_CSC_GAMMA_ENABLE;
+
+	if (!crtc_state->base.gamma_lut ||
+	    crtc_state_is_legacy_gamma(crtc_state))
+		gamma_mode |= GAMMA_MODE_MODE_8BIT;
+	else
+		gamma_mode |= GAMMA_MODE_MODE_10BIT;
+
+	return gamma_mode;
+}
+
+static u32 icl_csc_mode(const struct intel_crtc_state *crtc_state)
+{
+	u32 csc_mode = 0;
+
+	if (crtc_state->base.ctm)
+		csc_mode |= ICL_CSC_ENABLE;
+
+	if (crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB ||
+	    crtc_state->limited_color_range)
+		csc_mode |= ICL_OUTPUT_CSC_ENABLE;
+
+	return csc_mode;
+}
+
+static int icl_color_check(struct intel_crtc_state *crtc_state)
+{
+	int ret;
+
+	ret = check_luts(crtc_state);
+	if (ret)
+		return ret;
+
+	crtc_state->gamma_mode = icl_gamma_mode(crtc_state);
+
+	crtc_state->csc_mode = icl_csc_mode(crtc_state);
 
 	return 0;
 }
@@ -646,30 +1224,55 @@ int intel_color_check(struct intel_crtc_state *crtc_state)
 void intel_color_init(struct intel_crtc *crtc)
 {
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	bool has_ctm = INTEL_INFO(dev_priv)->color.degamma_lut_size != 0;
 
 	drm_mode_crtc_set_gamma_size(&crtc->base, 256);
 
-	if (IS_CHERRYVIEW(dev_priv)) {
-		dev_priv->display.load_luts = cherryview_load_luts;
-	} else if (IS_HASWELL(dev_priv)) {
-		dev_priv->display.load_luts = i9xx_load_luts;
-		dev_priv->display.color_commit = hsw_color_commit;
-	} else if (IS_BROADWELL(dev_priv) || IS_GEN9_BC(dev_priv) ||
-		   IS_BROXTON(dev_priv)) {
-		dev_priv->display.load_luts = broadwell_load_luts;
-		dev_priv->display.color_commit = hsw_color_commit;
-	} else if (IS_GEMINILAKE(dev_priv) || IS_CANNONLAKE(dev_priv)) {
-		dev_priv->display.load_luts = glk_load_luts;
-		dev_priv->display.color_commit = hsw_color_commit;
+	if (HAS_GMCH(dev_priv)) {
+		if (IS_CHERRYVIEW(dev_priv)) {
+			dev_priv->display.color_check = chv_color_check;
+			dev_priv->display.color_commit = i9xx_color_commit;
+			dev_priv->display.load_luts = cherryview_load_luts;
+		} else if (INTEL_GEN(dev_priv) >= 4) {
+			dev_priv->display.color_check = i9xx_color_check;
+			dev_priv->display.color_commit = i9xx_color_commit;
+			dev_priv->display.load_luts = i965_load_luts;
+		} else {
+			dev_priv->display.color_check = i9xx_color_check;
+			dev_priv->display.color_commit = i9xx_color_commit;
+			dev_priv->display.load_luts = i9xx_load_luts;
+		}
 	} else {
-		dev_priv->display.load_luts = i9xx_load_luts;
+		if (INTEL_GEN(dev_priv) >= 11)
+			dev_priv->display.color_check = icl_color_check;
+		else if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv))
+			dev_priv->display.color_check = glk_color_check;
+		else if (INTEL_GEN(dev_priv) >= 7)
+			dev_priv->display.color_check = ivb_color_check;
+		else
+			dev_priv->display.color_check = ilk_color_check;
+
+		if (INTEL_GEN(dev_priv) >= 9)
+			dev_priv->display.color_commit = skl_color_commit;
+		else if (IS_BROADWELL(dev_priv) || IS_HASWELL(dev_priv))
+			dev_priv->display.color_commit = hsw_color_commit;
+		else
+			dev_priv->display.color_commit = ilk_color_commit;
+
+		if (INTEL_GEN(dev_priv) >= 11)
+			dev_priv->display.load_luts = icl_load_luts;
+		else if (IS_CANNONLAKE(dev_priv) || IS_GEMINILAKE(dev_priv))
+			dev_priv->display.load_luts = glk_load_luts;
+		else if (INTEL_GEN(dev_priv) >= 8)
+			dev_priv->display.load_luts = bdw_load_luts;
+		else if (INTEL_GEN(dev_priv) >= 7)
+			dev_priv->display.load_luts = ivb_load_luts;
+		else
+			dev_priv->display.load_luts = ilk_load_luts;
 	}
 
-	/* Enable color management support when we have degamma & gamma LUTs. */
-	if (INTEL_INFO(dev_priv)->color.degamma_lut_size != 0 &&
-	    INTEL_INFO(dev_priv)->color.gamma_lut_size != 0)
-		drm_crtc_enable_color_mgmt(&crtc->base,
-					   INTEL_INFO(dev_priv)->color.degamma_lut_size,
-					   true,
-					   INTEL_INFO(dev_priv)->color.gamma_lut_size);
+	drm_crtc_enable_color_mgmt(&crtc->base,
+				   INTEL_INFO(dev_priv)->color.degamma_lut_size,
+				   has_ctm,
+				   INTEL_INFO(dev_priv)->color.gamma_lut_size);
 }
diff --git a/drivers/gpu/drm/i915/intel_color.h b/drivers/gpu/drm/i915/intel_color.h
new file mode 100644
index 000000000000..b8a3ce609587
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_color.h
@@ -0,0 +1,17 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_COLOR_H__
+#define __INTEL_COLOR_H__
+
+struct intel_crtc_state;
+struct intel_crtc;
+
+void intel_color_init(struct intel_crtc *crtc);
+int intel_color_check(struct intel_crtc_state *crtc_state);
+void intel_color_commit(const struct intel_crtc_state *crtc_state);
+void intel_color_load_luts(const struct intel_crtc_state *crtc_state);
+
+#endif /* __INTEL_COLOR_H__ */
diff --git a/drivers/gpu/drm/i915/intel_combo_phy.c b/drivers/gpu/drm/i915/intel_combo_phy.c
index 3d0271cebf99..2bf4359d7e41 100644
--- a/drivers/gpu/drm/i915/intel_combo_phy.c
+++ b/drivers/gpu/drm/i915/intel_combo_phy.c
@@ -239,7 +239,8 @@ void icl_combo_phys_uninit(struct drm_i915_private *dev_priv)
 	for_each_combo_port_reverse(dev_priv, port) {
 		u32 val;
 
-		if (!icl_combo_phy_verify_state(dev_priv, port))
+		if (port == PORT_A &&
+		    !icl_combo_phy_verify_state(dev_priv, port))
 			DRM_WARN("Port %c combo PHY HW state changed unexpectedly\n",
 				 port_name(port));
 
diff --git a/drivers/gpu/drm/i915/intel_connector.c b/drivers/gpu/drm/i915/intel_connector.c
index ee16758747c5..073b6c3ab7cc 100644
--- a/drivers/gpu/drm/i915/intel_connector.c
+++ b/drivers/gpu/drm/i915/intel_connector.c
@@ -23,12 +23,17 @@
  * DEALINGS IN THE SOFTWARE.
  */
 
-#include <linux/slab.h>
 #include <linux/i2c.h>
+#include <linux/slab.h>
+
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_edid.h>
-#include "intel_drv.h"
+
 #include "i915_drv.h"
+#include "intel_connector.h"
+#include "intel_drv.h"
+#include "intel_hdcp.h"
+#include "intel_panel.h"
 
 int intel_connector_init(struct intel_connector *connector)
 {
@@ -88,6 +93,8 @@ void intel_connector_destroy(struct drm_connector *connector)
 
 	kfree(intel_connector->detect_edid);
 
+	intel_hdcp_cleanup(intel_connector);
+
 	if (!IS_ERR_OR_NULL(intel_connector->edid))
 		kfree(intel_connector->edid);
 
@@ -265,3 +272,11 @@ intel_attach_aspect_ratio_property(struct drm_connector *connector)
 			connector->dev->mode_config.aspect_ratio_property,
 			DRM_MODE_PICTURE_ASPECT_NONE);
 }
+
+void
+intel_attach_colorspace_property(struct drm_connector *connector)
+{
+	if (!drm_mode_create_colorspace_property(connector))
+		drm_object_attach_property(&connector->base,
+					   connector->colorspace_property, 0);
+}
diff --git a/drivers/gpu/drm/i915/intel_connector.h b/drivers/gpu/drm/i915/intel_connector.h
new file mode 100644
index 000000000000..93a7375c8196
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_connector.h
@@ -0,0 +1,35 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_CONNECTOR_H__
+#define __INTEL_CONNECTOR_H__
+
+#include "intel_display.h"
+
+struct drm_connector;
+struct edid;
+struct i2c_adapter;
+struct intel_connector;
+struct intel_encoder;
+
+int intel_connector_init(struct intel_connector *connector);
+struct intel_connector *intel_connector_alloc(void);
+void intel_connector_free(struct intel_connector *connector);
+void intel_connector_destroy(struct drm_connector *connector);
+int intel_connector_register(struct drm_connector *connector);
+void intel_connector_unregister(struct drm_connector *connector);
+void intel_connector_attach_encoder(struct intel_connector *connector,
+				    struct intel_encoder *encoder);
+bool intel_connector_get_hw_state(struct intel_connector *connector);
+enum pipe intel_connector_get_pipe(struct intel_connector *connector);
+int intel_connector_update_modes(struct drm_connector *connector,
+				 struct edid *edid);
+int intel_ddc_get_modes(struct drm_connector *c, struct i2c_adapter *adapter);
+void intel_attach_force_audio_property(struct drm_connector *connector);
+void intel_attach_broadcast_rgb_property(struct drm_connector *connector);
+void intel_attach_aspect_ratio_property(struct drm_connector *connector);
+void intel_attach_colorspace_property(struct drm_connector *connector);
+
+#endif /* __INTEL_CONNECTOR_H__ */
diff --git a/drivers/gpu/drm/i915/intel_context.c b/drivers/gpu/drm/i915/intel_context.c
new file mode 100644
index 000000000000..8931e0fee873
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_context.c
@@ -0,0 +1,269 @@
+/*
+ * SPDX-License-Identifier: MIT
+ *
+ * Copyright © 2019 Intel Corporation
+ */
+
+#include "i915_drv.h"
+#include "i915_gem_context.h"
+#include "i915_globals.h"
+#include "intel_context.h"
+#include "intel_ringbuffer.h"
+
+static struct i915_global_context {
+	struct i915_global base;
+	struct kmem_cache *slab_ce;
+} global;
+
+struct intel_context *intel_context_alloc(void)
+{
+	return kmem_cache_zalloc(global.slab_ce, GFP_KERNEL);
+}
+
+void intel_context_free(struct intel_context *ce)
+{
+	kmem_cache_free(global.slab_ce, ce);
+}
+
+struct intel_context *
+intel_context_lookup(struct i915_gem_context *ctx,
+		     struct intel_engine_cs *engine)
+{
+	struct intel_context *ce = NULL;
+	struct rb_node *p;
+
+	spin_lock(&ctx->hw_contexts_lock);
+	p = ctx->hw_contexts.rb_node;
+	while (p) {
+		struct intel_context *this =
+			rb_entry(p, struct intel_context, node);
+
+		if (this->engine == engine) {
+			GEM_BUG_ON(this->gem_context != ctx);
+			ce = this;
+			break;
+		}
+
+		if (this->engine < engine)
+			p = p->rb_right;
+		else
+			p = p->rb_left;
+	}
+	spin_unlock(&ctx->hw_contexts_lock);
+
+	return ce;
+}
+
+struct intel_context *
+__intel_context_insert(struct i915_gem_context *ctx,
+		       struct intel_engine_cs *engine,
+		       struct intel_context *ce)
+{
+	struct rb_node **p, *parent;
+	int err = 0;
+
+	spin_lock(&ctx->hw_contexts_lock);
+
+	parent = NULL;
+	p = &ctx->hw_contexts.rb_node;
+	while (*p) {
+		struct intel_context *this;
+
+		parent = *p;
+		this = rb_entry(parent, struct intel_context, node);
+
+		if (this->engine == engine) {
+			err = -EEXIST;
+			ce = this;
+			break;
+		}
+
+		if (this->engine < engine)
+			p = &parent->rb_right;
+		else
+			p = &parent->rb_left;
+	}
+	if (!err) {
+		rb_link_node(&ce->node, parent, p);
+		rb_insert_color(&ce->node, &ctx->hw_contexts);
+	}
+
+	spin_unlock(&ctx->hw_contexts_lock);
+
+	return ce;
+}
+
+void __intel_context_remove(struct intel_context *ce)
+{
+	struct i915_gem_context *ctx = ce->gem_context;
+
+	spin_lock(&ctx->hw_contexts_lock);
+	rb_erase(&ce->node, &ctx->hw_contexts);
+	spin_unlock(&ctx->hw_contexts_lock);
+}
+
+static struct intel_context *
+intel_context_instance(struct i915_gem_context *ctx,
+		       struct intel_engine_cs *engine)
+{
+	struct intel_context *ce, *pos;
+
+	ce = intel_context_lookup(ctx, engine);
+	if (likely(ce))
+		return ce;
+
+	ce = intel_context_alloc();
+	if (!ce)
+		return ERR_PTR(-ENOMEM);
+
+	intel_context_init(ce, ctx, engine);
+
+	pos = __intel_context_insert(ctx, engine, ce);
+	if (unlikely(pos != ce)) /* Beaten! Use their HW context instead */
+		intel_context_free(ce);
+
+	GEM_BUG_ON(intel_context_lookup(ctx, engine) != pos);
+	return pos;
+}
+
+struct intel_context *
+intel_context_pin_lock(struct i915_gem_context *ctx,
+		       struct intel_engine_cs *engine)
+	__acquires(ce->pin_mutex)
+{
+	struct intel_context *ce;
+
+	ce = intel_context_instance(ctx, engine);
+	if (IS_ERR(ce))
+		return ce;
+
+	if (mutex_lock_interruptible(&ce->pin_mutex))
+		return ERR_PTR(-EINTR);
+
+	return ce;
+}
+
+struct intel_context *
+intel_context_pin(struct i915_gem_context *ctx,
+		  struct intel_engine_cs *engine)
+{
+	struct intel_context *ce;
+	int err;
+
+	ce = intel_context_instance(ctx, engine);
+	if (IS_ERR(ce))
+		return ce;
+
+	if (likely(atomic_inc_not_zero(&ce->pin_count)))
+		return ce;
+
+	if (mutex_lock_interruptible(&ce->pin_mutex))
+		return ERR_PTR(-EINTR);
+
+	if (likely(!atomic_read(&ce->pin_count))) {
+		err = ce->ops->pin(ce);
+		if (err)
+			goto err;
+
+		i915_gem_context_get(ctx);
+		GEM_BUG_ON(ce->gem_context != ctx);
+
+		mutex_lock(&ctx->mutex);
+		list_add(&ce->active_link, &ctx->active_engines);
+		mutex_unlock(&ctx->mutex);
+
+		intel_context_get(ce);
+		smp_mb__before_atomic(); /* flush pin before it is visible */
+	}
+
+	atomic_inc(&ce->pin_count);
+	GEM_BUG_ON(!intel_context_is_pinned(ce)); /* no overflow! */
+
+	mutex_unlock(&ce->pin_mutex);
+	return ce;
+
+err:
+	mutex_unlock(&ce->pin_mutex);
+	return ERR_PTR(err);
+}
+
+void intel_context_unpin(struct intel_context *ce)
+{
+	if (likely(atomic_add_unless(&ce->pin_count, -1, 1)))
+		return;
+
+	/* We may be called from inside intel_context_pin() to evict another */
+	intel_context_get(ce);
+	mutex_lock_nested(&ce->pin_mutex, SINGLE_DEPTH_NESTING);
+
+	if (likely(atomic_dec_and_test(&ce->pin_count))) {
+		ce->ops->unpin(ce);
+
+		mutex_lock(&ce->gem_context->mutex);
+		list_del(&ce->active_link);
+		mutex_unlock(&ce->gem_context->mutex);
+
+		i915_gem_context_put(ce->gem_context);
+		intel_context_put(ce);
+	}
+
+	mutex_unlock(&ce->pin_mutex);
+	intel_context_put(ce);
+}
+
+static void intel_context_retire(struct i915_active_request *active,
+				 struct i915_request *rq)
+{
+	struct intel_context *ce =
+		container_of(active, typeof(*ce), active_tracker);
+
+	intel_context_unpin(ce);
+}
+
+void
+intel_context_init(struct intel_context *ce,
+		   struct i915_gem_context *ctx,
+		   struct intel_engine_cs *engine)
+{
+	kref_init(&ce->ref);
+
+	ce->gem_context = ctx;
+	ce->engine = engine;
+	ce->ops = engine->cops;
+
+	INIT_LIST_HEAD(&ce->signal_link);
+	INIT_LIST_HEAD(&ce->signals);
+
+	mutex_init(&ce->pin_mutex);
+
+	/* Use the whole device by default */
+	ce->sseu = intel_device_default_sseu(ctx->i915);
+
+	i915_active_request_init(&ce->active_tracker,
+				 NULL, intel_context_retire);
+}
+
+static void i915_global_context_shrink(void)
+{
+	kmem_cache_shrink(global.slab_ce);
+}
+
+static void i915_global_context_exit(void)
+{
+	kmem_cache_destroy(global.slab_ce);
+}
+
+static struct i915_global_context global = { {
+	.shrink = i915_global_context_shrink,
+	.exit = i915_global_context_exit,
+} };
+
+int __init i915_global_context_init(void)
+{
+	global.slab_ce = KMEM_CACHE(intel_context, SLAB_HWCACHE_ALIGN);
+	if (!global.slab_ce)
+		return -ENOMEM;
+
+	i915_global_register(&global.base);
+	return 0;
+}
diff --git a/drivers/gpu/drm/i915/intel_context.h b/drivers/gpu/drm/i915/intel_context.h
new file mode 100644
index 000000000000..ebc861b1a49e
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_context.h
@@ -0,0 +1,87 @@
+/*
+ * SPDX-License-Identifier: MIT
+ *
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_CONTEXT_H__
+#define __INTEL_CONTEXT_H__
+
+#include <linux/lockdep.h>
+
+#include "intel_context_types.h"
+#include "intel_engine_types.h"
+
+struct intel_context *intel_context_alloc(void);
+void intel_context_free(struct intel_context *ce);
+
+void intel_context_init(struct intel_context *ce,
+			struct i915_gem_context *ctx,
+			struct intel_engine_cs *engine);
+
+/**
+ * intel_context_lookup - Find the matching HW context for this (ctx, engine)
+ * @ctx - the parent GEM context
+ * @engine - the target HW engine
+ *
+ * May return NULL if the HW context hasn't been instantiated (i.e. unused).
+ */
+struct intel_context *
+intel_context_lookup(struct i915_gem_context *ctx,
+		     struct intel_engine_cs *engine);
+
+/**
+ * intel_context_pin_lock - Stablises the 'pinned' status of the HW context
+ * @ctx - the parent GEM context
+ * @engine - the target HW engine
+ *
+ * Acquire a lock on the pinned status of the HW context, such that the context
+ * can neither be bound to the GPU or unbound whilst the lock is held, i.e.
+ * intel_context_is_pinned() remains stable.
+ */
+struct intel_context *
+intel_context_pin_lock(struct i915_gem_context *ctx,
+		       struct intel_engine_cs *engine);
+
+static inline bool
+intel_context_is_pinned(struct intel_context *ce)
+{
+	return atomic_read(&ce->pin_count);
+}
+
+static inline void intel_context_pin_unlock(struct intel_context *ce)
+__releases(ce->pin_mutex)
+{
+	mutex_unlock(&ce->pin_mutex);
+}
+
+struct intel_context *
+__intel_context_insert(struct i915_gem_context *ctx,
+		       struct intel_engine_cs *engine,
+		       struct intel_context *ce);
+void
+__intel_context_remove(struct intel_context *ce);
+
+struct intel_context *
+intel_context_pin(struct i915_gem_context *ctx, struct intel_engine_cs *engine);
+
+static inline void __intel_context_pin(struct intel_context *ce)
+{
+	GEM_BUG_ON(!intel_context_is_pinned(ce));
+	atomic_inc(&ce->pin_count);
+}
+
+void intel_context_unpin(struct intel_context *ce);
+
+static inline struct intel_context *intel_context_get(struct intel_context *ce)
+{
+	kref_get(&ce->ref);
+	return ce;
+}
+
+static inline void intel_context_put(struct intel_context *ce)
+{
+	kref_put(&ce->ref, ce->ops->destroy);
+}
+
+#endif /* __INTEL_CONTEXT_H__ */
diff --git a/drivers/gpu/drm/i915/intel_context_types.h b/drivers/gpu/drm/i915/intel_context_types.h
new file mode 100644
index 000000000000..68b4ca1611e0
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_context_types.h
@@ -0,0 +1,74 @@
+/*
+ * SPDX-License-Identifier: MIT
+ *
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_CONTEXT_TYPES__
+#define __INTEL_CONTEXT_TYPES__
+
+#include <linux/kref.h>
+#include <linux/list.h>
+#include <linux/mutex.h>
+#include <linux/rbtree.h>
+#include <linux/types.h>
+
+#include "i915_active_types.h"
+
+struct i915_gem_context;
+struct i915_vma;
+struct intel_context;
+struct intel_ring;
+
+struct intel_context_ops {
+	int (*pin)(struct intel_context *ce);
+	void (*unpin)(struct intel_context *ce);
+
+	void (*reset)(struct intel_context *ce);
+	void (*destroy)(struct kref *kref);
+};
+
+/*
+ * Powergating configuration for a particular (context,engine).
+ */
+struct intel_sseu {
+	u8 slice_mask;
+	u8 subslice_mask;
+	u8 min_eus_per_subslice;
+	u8 max_eus_per_subslice;
+};
+
+struct intel_context {
+	struct kref ref;
+
+	struct i915_gem_context *gem_context;
+	struct intel_engine_cs *engine;
+	struct intel_engine_cs *active;
+
+	struct list_head active_link;
+	struct list_head signal_link;
+	struct list_head signals;
+
+	struct i915_vma *state;
+	struct intel_ring *ring;
+
+	u32 *lrc_reg_state;
+	u64 lrc_desc;
+
+	atomic_t pin_count;
+	struct mutex pin_mutex; /* guards pinning and associated on-gpuing */
+
+	/**
+	 * active_tracker: Active tracker for the external rq activity
+	 * on this intel_context object.
+	 */
+	struct i915_active_request active_tracker;
+
+	const struct intel_context_ops *ops;
+	struct rb_node node;
+
+	/** sseu: Control eu/slice partitioning */
+	struct intel_sseu sseu;
+};
+
+#endif /* __INTEL_CONTEXT_TYPES__ */
diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c
index 3716b2ee362f..b665c370111b 100644
--- a/drivers/gpu/drm/i915/intel_crt.c
+++ b/drivers/gpu/drm/i915/intel_crt.c
@@ -27,13 +27,18 @@
 #include <linux/dmi.h>
 #include <linux/i2c.h>
 #include <linux/slab.h>
+
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_probe_helper.h>
-#include "intel_drv.h"
 #include <drm/i915_drm.h>
+
 #include "i915_drv.h"
+#include "intel_connector.h"
+#include "intel_crt.h"
+#include "intel_ddi.h"
+#include "intel_drv.h"
 
 /* Here's the desired hotplug mode */
 #define ADPA_HOTPLUG_BITS (ADPA_CRT_HOTPLUG_PERIOD_128 |		\
@@ -435,7 +440,7 @@ static bool intel_ironlake_crt_detect_hotplug(struct drm_connector *connector)
 
 		I915_WRITE(crt->adpa_reg, adpa);
 
-		if (intel_wait_for_register(dev_priv,
+		if (intel_wait_for_register(&dev_priv->uncore,
 					    crt->adpa_reg,
 					    ADPA_CRT_HOTPLUG_FORCE_TRIGGER, 0,
 					    1000))
@@ -489,7 +494,7 @@ static bool valleyview_crt_detect_hotplug(struct drm_connector *connector)
 
 	I915_WRITE(crt->adpa_reg, adpa);
 
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    crt->adpa_reg,
 				    ADPA_CRT_HOTPLUG_FORCE_TRIGGER, 0,
 				    1000)) {
@@ -542,7 +547,7 @@ static bool intel_crt_detect_hotplug(struct drm_connector *connector)
 					      CRT_HOTPLUG_FORCE_DETECT,
 					      CRT_HOTPLUG_FORCE_DETECT);
 		/* wait for FORCE_DETECT to go off */
-		if (intel_wait_for_register(dev_priv, PORT_HOTPLUG_EN,
+		if (intel_wait_for_register(&dev_priv->uncore, PORT_HOTPLUG_EN,
 					    CRT_HOTPLUG_FORCE_DETECT, 0,
 					    1000))
 			DRM_DEBUG_KMS("timed out waiting for FORCE_DETECT to go off");
diff --git a/drivers/gpu/drm/i915/intel_crt.h b/drivers/gpu/drm/i915/intel_crt.h
new file mode 100644
index 000000000000..1b3fba359efc
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_crt.h
@@ -0,0 +1,21 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_CRT_H__
+#define __INTEL_CRT_H__
+
+#include "i915_reg.h"
+
+enum pipe;
+struct drm_encoder;
+struct drm_i915_private;
+struct drm_i915_private;
+
+bool intel_crt_port_enabled(struct drm_i915_private *dev_priv,
+			    i915_reg_t adpa_reg, enum pipe *pipe);
+void intel_crt_init(struct drm_i915_private *dev_priv);
+void intel_crt_reset(struct drm_encoder *encoder);
+
+#endif /* __INTEL_CRT_H__ */
diff --git a/drivers/gpu/drm/i915/intel_csr.c b/drivers/gpu/drm/i915/intel_csr.c
index e8ac04c33e29..f43c2a2563a5 100644
--- a/drivers/gpu/drm/i915/intel_csr.c
+++ b/drivers/gpu/drm/i915/intel_csr.c
@@ -21,9 +21,12 @@
  * IN THE SOFTWARE.
  *
  */
+
 #include <linux/firmware.h>
+
 #include "i915_drv.h"
 #include "i915_reg.h"
+#include "intel_csr.h"
 
 /**
  * DOC: csr support for dmc
@@ -486,7 +489,7 @@ void intel_csr_ucode_init(struct drm_i915_private *dev_priv)
 	if (INTEL_GEN(dev_priv) >= 12) {
 		/* Allow to load fw via parameter using the last known size */
 		csr->max_fw_size = GEN12_CSR_MAX_FW_SIZE;
-	} else if (IS_ICELAKE(dev_priv)) {
+	} else if (IS_GEN(dev_priv, 11)) {
 		csr->fw_path = ICL_CSR_PATH;
 		csr->required_version = ICL_CSR_VERSION_REQUIRED;
 		csr->max_fw_size = ICL_CSR_MAX_FW_SIZE;
diff --git a/drivers/gpu/drm/i915/intel_csr.h b/drivers/gpu/drm/i915/intel_csr.h
new file mode 100644
index 000000000000..17a32c1e8a35
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_csr.h
@@ -0,0 +1,17 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_CSR_H__
+#define __INTEL_CSR_H__
+
+struct drm_i915_private;
+
+void intel_csr_ucode_init(struct drm_i915_private *i915);
+void intel_csr_load_program(struct drm_i915_private *i915);
+void intel_csr_ucode_fini(struct drm_i915_private *i915);
+void intel_csr_ucode_suspend(struct drm_i915_private *i915);
+void intel_csr_ucode_resume(struct drm_i915_private *i915);
+
+#endif /* __INTEL_CSR_H__ */
diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 98cea1f4b3bf..f181c26f62fd 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -26,9 +26,19 @@
  */
 
 #include <drm/drm_scdc_helper.h>
+
 #include "i915_drv.h"
+#include "intel_audio.h"
+#include "intel_connector.h"
+#include "intel_ddi.h"
+#include "intel_dp.h"
 #include "intel_drv.h"
 #include "intel_dsi.h"
+#include "intel_hdcp.h"
+#include "intel_hdmi.h"
+#include "intel_lspcon.h"
+#include "intel_panel.h"
+#include "intel_psr.h"
 
 struct ddi_buf_trans {
 	u32 trans1;	/* balance leg enable, de-emph level */
@@ -851,7 +861,7 @@ static int intel_ddi_hdmi_level(struct drm_i915_private *dev_priv, enum port por
 
 	level = dev_priv->vbt.ddi_port_info[port].hdmi_level_shift;
 
-	if (IS_ICELAKE(dev_priv)) {
+	if (INTEL_GEN(dev_priv) >= 11) {
 		if (intel_port_is_combophy(dev_priv, port))
 			icl_get_combo_buf_trans(dev_priv, port, INTEL_OUTPUT_HDMI,
 						0, &n_entries);
@@ -1240,24 +1250,15 @@ static int hsw_ddi_calc_wrpll_link(struct drm_i915_private *dev_priv,
 	return (refclk * n * 100) / (p * r);
 }
 
-static int skl_calc_wrpll_link(struct drm_i915_private *dev_priv,
-			       enum intel_dpll_id pll_id)
+static int skl_calc_wrpll_link(const struct intel_dpll_hw_state *pll_state)
 {
-	i915_reg_t cfgcr1_reg, cfgcr2_reg;
-	u32 cfgcr1_val, cfgcr2_val;
 	u32 p0, p1, p2, dco_freq;
 
-	cfgcr1_reg = DPLL_CFGCR1(pll_id);
-	cfgcr2_reg = DPLL_CFGCR2(pll_id);
-
-	cfgcr1_val = I915_READ(cfgcr1_reg);
-	cfgcr2_val = I915_READ(cfgcr2_reg);
-
-	p0 = cfgcr2_val & DPLL_CFGCR2_PDIV_MASK;
-	p2 = cfgcr2_val & DPLL_CFGCR2_KDIV_MASK;
+	p0 = pll_state->cfgcr2 & DPLL_CFGCR2_PDIV_MASK;
+	p2 = pll_state->cfgcr2 & DPLL_CFGCR2_KDIV_MASK;
 
-	if (cfgcr2_val &  DPLL_CFGCR2_QDIV_MODE(1))
-		p1 = (cfgcr2_val & DPLL_CFGCR2_QDIV_RATIO_MASK) >> 8;
+	if (pll_state->cfgcr2 &  DPLL_CFGCR2_QDIV_MODE(1))
+		p1 = (pll_state->cfgcr2 & DPLL_CFGCR2_QDIV_RATIO_MASK) >> 8;
 	else
 		p1 = 1;
 
@@ -1292,10 +1293,11 @@ static int skl_calc_wrpll_link(struct drm_i915_private *dev_priv,
 		break;
 	}
 
-	dco_freq = (cfgcr1_val & DPLL_CFGCR1_DCO_INTEGER_MASK) * 24 * 1000;
+	dco_freq = (pll_state->cfgcr1 & DPLL_CFGCR1_DCO_INTEGER_MASK)
+		* 24 * 1000;
 
-	dco_freq += (((cfgcr1_val & DPLL_CFGCR1_DCO_FRACTION_MASK) >> 9) * 24 *
-		1000) / 0x8000;
+	dco_freq += (((pll_state->cfgcr1 & DPLL_CFGCR1_DCO_FRACTION_MASK) >> 9)
+		     * 24 * 1000) / 0x8000;
 
 	if (WARN_ON(p0 == 0 || p1 == 0 || p2 == 0))
 		return 0;
@@ -1304,24 +1306,15 @@ static int skl_calc_wrpll_link(struct drm_i915_private *dev_priv,
 }
 
 int cnl_calc_wrpll_link(struct drm_i915_private *dev_priv,
-			enum intel_dpll_id pll_id)
+			struct intel_dpll_hw_state *pll_state)
 {
-	u32 cfgcr0, cfgcr1;
 	u32 p0, p1, p2, dco_freq, ref_clock;
 
-	if (INTEL_GEN(dev_priv) >= 11) {
-		cfgcr0 = I915_READ(ICL_DPLL_CFGCR0(pll_id));
-		cfgcr1 = I915_READ(ICL_DPLL_CFGCR1(pll_id));
-	} else {
-		cfgcr0 = I915_READ(CNL_DPLL_CFGCR0(pll_id));
-		cfgcr1 = I915_READ(CNL_DPLL_CFGCR1(pll_id));
-	}
-
-	p0 = cfgcr1 & DPLL_CFGCR1_PDIV_MASK;
-	p2 = cfgcr1 & DPLL_CFGCR1_KDIV_MASK;
+	p0 = pll_state->cfgcr1 & DPLL_CFGCR1_PDIV_MASK;
+	p2 = pll_state->cfgcr1 & DPLL_CFGCR1_KDIV_MASK;
 
-	if (cfgcr1 & DPLL_CFGCR1_QDIV_MODE(1))
-		p1 = (cfgcr1 & DPLL_CFGCR1_QDIV_RATIO_MASK) >>
+	if (pll_state->cfgcr1 & DPLL_CFGCR1_QDIV_MODE(1))
+		p1 = (pll_state->cfgcr1 & DPLL_CFGCR1_QDIV_RATIO_MASK) >>
 			DPLL_CFGCR1_QDIV_RATIO_SHIFT;
 	else
 		p1 = 1;
@@ -1349,16 +1342,17 @@ int cnl_calc_wrpll_link(struct drm_i915_private *dev_priv,
 	case DPLL_CFGCR1_KDIV_2:
 		p2 = 2;
 		break;
-	case DPLL_CFGCR1_KDIV_4:
-		p2 = 4;
+	case DPLL_CFGCR1_KDIV_3:
+		p2 = 3;
 		break;
 	}
 
 	ref_clock = cnl_hdmi_pll_ref_clock(dev_priv);
 
-	dco_freq = (cfgcr0 & DPLL_CFGCR0_DCO_INTEGER_MASK) * ref_clock;
+	dco_freq = (pll_state->cfgcr0 & DPLL_CFGCR0_DCO_INTEGER_MASK)
+		* ref_clock;
 
-	dco_freq += (((cfgcr0 & DPLL_CFGCR0_DCO_FRACTION_MASK) >>
+	dco_freq += (((pll_state->cfgcr0 & DPLL_CFGCR0_DCO_FRACTION_MASK) >>
 		      DPLL_CFGCR0_DCO_FRACTION_SHIFT) * ref_clock) / 0x8000;
 
 	if (WARN_ON(p0 == 0 || p1 == 0 || p2 == 0))
@@ -1390,25 +1384,21 @@ static int icl_calc_tbt_pll_link(struct drm_i915_private *dev_priv,
 }
 
 static int icl_calc_mg_pll_link(struct drm_i915_private *dev_priv,
-				enum port port)
+				const struct intel_dpll_hw_state *pll_state)
 {
-	enum tc_port tc_port = intel_port_to_tc(dev_priv, port);
-	u32 mg_pll_div0, mg_clktop_hsclkctl;
-	u32 m1, m2_int, m2_frac, div1, div2, refclk;
+	u32 m1, m2_int, m2_frac, div1, div2, ref_clock;
 	u64 tmp;
 
-	refclk = dev_priv->cdclk.hw.ref;
-
-	mg_pll_div0 = I915_READ(MG_PLL_DIV0(tc_port));
-	mg_clktop_hsclkctl = I915_READ(MG_CLKTOP2_HSCLKCTL(tc_port));
+	ref_clock = dev_priv->cdclk.hw.ref;
 
-	m1 = I915_READ(MG_PLL_DIV1(tc_port)) & MG_PLL_DIV1_FBPREDIV_MASK;
-	m2_int = mg_pll_div0 & MG_PLL_DIV0_FBDIV_INT_MASK;
-	m2_frac = (mg_pll_div0 & MG_PLL_DIV0_FRACNEN_H) ?
-		  (mg_pll_div0 & MG_PLL_DIV0_FBDIV_FRAC_MASK) >>
-		  MG_PLL_DIV0_FBDIV_FRAC_SHIFT : 0;
+	m1 = pll_state->mg_pll_div1 & MG_PLL_DIV1_FBPREDIV_MASK;
+	m2_int = pll_state->mg_pll_div0 & MG_PLL_DIV0_FBDIV_INT_MASK;
+	m2_frac = (pll_state->mg_pll_div0 & MG_PLL_DIV0_FRACNEN_H) ?
+		(pll_state->mg_pll_div0 & MG_PLL_DIV0_FBDIV_FRAC_MASK) >>
+		MG_PLL_DIV0_FBDIV_FRAC_SHIFT : 0;
 
-	switch (mg_clktop_hsclkctl & MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_MASK) {
+	switch (pll_state->mg_clktop2_hsclkctl &
+		MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_MASK) {
 	case MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_2:
 		div1 = 2;
 		break;
@@ -1422,12 +1412,14 @@ static int icl_calc_mg_pll_link(struct drm_i915_private *dev_priv,
 		div1 = 7;
 		break;
 	default:
-		MISSING_CASE(mg_clktop_hsclkctl);
+		MISSING_CASE(pll_state->mg_clktop2_hsclkctl);
 		return 0;
 	}
 
-	div2 = (mg_clktop_hsclkctl & MG_CLKTOP2_HSCLKCTL_DSDIV_RATIO_MASK) >>
+	div2 = (pll_state->mg_clktop2_hsclkctl &
+		MG_CLKTOP2_HSCLKCTL_DSDIV_RATIO_MASK) >>
 		MG_CLKTOP2_HSCLKCTL_DSDIV_RATIO_SHIFT;
+
 	/* div2 value of 0 is same as 1 means no div */
 	if (div2 == 0)
 		div2 = 1;
@@ -1436,8 +1428,8 @@ static int icl_calc_mg_pll_link(struct drm_i915_private *dev_priv,
 	 * Adjust the original formula to delay the division by 2^22 in order to
 	 * minimize possible rounding errors.
 	 */
-	tmp = (u64)m1 * m2_int * refclk +
-	      (((u64)m1 * m2_frac * refclk) >> 22);
+	tmp = (u64)m1 * m2_int * ref_clock +
+	      (((u64)m1 * m2_frac * ref_clock) >> 22);
 	tmp = div_u64(tmp, 5 * div1 * div2);
 
 	return tmp;
@@ -1471,25 +1463,24 @@ static void icl_ddi_clock_get(struct intel_encoder *encoder,
 			      struct intel_crtc_state *pipe_config)
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_dpll_hw_state *pll_state = &pipe_config->dpll_hw_state;
 	enum port port = encoder->port;
-	int link_clock = 0;
-	u32 pll_id;
+	int link_clock;
 
-	pll_id = intel_get_shared_dpll_id(dev_priv, pipe_config->shared_dpll);
 	if (intel_port_is_combophy(dev_priv, port)) {
-		if (intel_crtc_has_type(pipe_config, INTEL_OUTPUT_HDMI))
-			link_clock = cnl_calc_wrpll_link(dev_priv, pll_id);
-		else
-			link_clock = icl_calc_dp_combo_pll_link(dev_priv,
-								pll_id);
+		link_clock = cnl_calc_wrpll_link(dev_priv, pll_state);
 	} else {
+		enum intel_dpll_id pll_id = intel_get_shared_dpll_id(dev_priv,
+						pipe_config->shared_dpll);
+
 		if (pll_id == DPLL_ID_ICL_TBTPLL)
 			link_clock = icl_calc_tbt_pll_link(dev_priv, port);
 		else
-			link_clock = icl_calc_mg_pll_link(dev_priv, port);
+			link_clock = icl_calc_mg_pll_link(dev_priv, pll_state);
 	}
 
 	pipe_config->port_clock = link_clock;
+
 	ddi_dotclock_get(pipe_config);
 }
 
@@ -1497,18 +1488,13 @@ static void cnl_ddi_clock_get(struct intel_encoder *encoder,
 			      struct intel_crtc_state *pipe_config)
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
-	int link_clock = 0;
-	u32 cfgcr0;
-	enum intel_dpll_id pll_id;
-
-	pll_id = intel_get_shared_dpll_id(dev_priv, pipe_config->shared_dpll);
+	struct intel_dpll_hw_state *pll_state = &pipe_config->dpll_hw_state;
+	int link_clock;
 
-	cfgcr0 = I915_READ(CNL_DPLL_CFGCR0(pll_id));
-
-	if (cfgcr0 & DPLL_CFGCR0_HDMI_MODE) {
-		link_clock = cnl_calc_wrpll_link(dev_priv, pll_id);
+	if (pll_state->cfgcr0 & DPLL_CFGCR0_HDMI_MODE) {
+		link_clock = cnl_calc_wrpll_link(dev_priv, pll_state);
 	} else {
-		link_clock = cfgcr0 & DPLL_CFGCR0_LINK_RATE_MASK;
+		link_clock = pll_state->cfgcr0 & DPLL_CFGCR0_LINK_RATE_MASK;
 
 		switch (link_clock) {
 		case DPLL_CFGCR0_LINK_RATE_810:
@@ -1548,22 +1534,20 @@ static void cnl_ddi_clock_get(struct intel_encoder *encoder,
 }
 
 static void skl_ddi_clock_get(struct intel_encoder *encoder,
-				struct intel_crtc_state *pipe_config)
+			      struct intel_crtc_state *pipe_config)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
-	int link_clock = 0;
-	u32 dpll_ctl1;
-	enum intel_dpll_id pll_id;
-
-	pll_id = intel_get_shared_dpll_id(dev_priv, pipe_config->shared_dpll);
+	struct intel_dpll_hw_state *pll_state = &pipe_config->dpll_hw_state;
+	int link_clock;
 
-	dpll_ctl1 = I915_READ(DPLL_CTRL1);
-
-	if (dpll_ctl1 & DPLL_CTRL1_HDMI_MODE(pll_id)) {
-		link_clock = skl_calc_wrpll_link(dev_priv, pll_id);
+	/*
+	 * ctrl1 register is already shifted for each pll, just use 0 to get
+	 * the internal shift for each field
+	 */
+	if (pll_state->ctrl1 & DPLL_CTRL1_HDMI_MODE(0)) {
+		link_clock = skl_calc_wrpll_link(pll_state);
 	} else {
-		link_clock = dpll_ctl1 & DPLL_CTRL1_LINK_RATE_MASK(pll_id);
-		link_clock >>= DPLL_CTRL1_LINK_RATE_SHIFT(pll_id);
+		link_clock = pll_state->ctrl1 & DPLL_CTRL1_LINK_RATE_MASK(0);
+		link_clock >>= DPLL_CTRL1_LINK_RATE_SHIFT(0);
 
 		switch (link_clock) {
 		case DPLL_CTRL1_LINK_RATE_810:
@@ -1643,24 +1627,17 @@ static void hsw_ddi_clock_get(struct intel_encoder *encoder,
 	ddi_dotclock_get(pipe_config);
 }
 
-static int bxt_calc_pll_link(struct intel_crtc_state *crtc_state)
+static int bxt_calc_pll_link(const struct intel_dpll_hw_state *pll_state)
 {
-	struct intel_dpll_hw_state *state;
 	struct dpll clock;
 
-	/* For DDI ports we always use a shared PLL. */
-	if (WARN_ON(!crtc_state->shared_dpll))
-		return 0;
-
-	state = &crtc_state->dpll_hw_state;
-
 	clock.m1 = 2;
-	clock.m2 = (state->pll0 & PORT_PLL_M2_MASK) << 22;
-	if (state->pll3 & PORT_PLL_M2_FRAC_ENABLE)
-		clock.m2 |= state->pll2 & PORT_PLL_M2_FRAC_MASK;
-	clock.n = (state->pll1 & PORT_PLL_N_MASK) >> PORT_PLL_N_SHIFT;
-	clock.p1 = (state->ebb0 & PORT_PLL_P1_MASK) >> PORT_PLL_P1_SHIFT;
-	clock.p2 = (state->ebb0 & PORT_PLL_P2_MASK) >> PORT_PLL_P2_SHIFT;
+	clock.m2 = (pll_state->pll0 & PORT_PLL_M2_MASK) << 22;
+	if (pll_state->pll3 & PORT_PLL_M2_FRAC_ENABLE)
+		clock.m2 |= pll_state->pll2 & PORT_PLL_M2_FRAC_MASK;
+	clock.n = (pll_state->pll1 & PORT_PLL_N_MASK) >> PORT_PLL_N_SHIFT;
+	clock.p1 = (pll_state->ebb0 & PORT_PLL_P1_MASK) >> PORT_PLL_P1_SHIFT;
+	clock.p2 = (pll_state->ebb0 & PORT_PLL_P2_MASK) >> PORT_PLL_P2_SHIFT;
 
 	return chv_calc_dpll_params(100000, &clock);
 }
@@ -1668,7 +1645,8 @@ static int bxt_calc_pll_link(struct intel_crtc_state *crtc_state)
 static void bxt_ddi_clock_get(struct intel_encoder *encoder,
 			      struct intel_crtc_state *pipe_config)
 {
-	pipe_config->port_clock = bxt_calc_pll_link(pipe_config);
+	pipe_config->port_clock =
+		bxt_calc_pll_link(&pipe_config->dpll_hw_state);
 
 	ddi_dotclock_get(pipe_config);
 }
@@ -1678,7 +1656,7 @@ static void intel_ddi_clock_get(struct intel_encoder *encoder,
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 
-	if (IS_ICELAKE(dev_priv))
+	if (INTEL_GEN(dev_priv) >= 11)
 		icl_ddi_clock_get(encoder, pipe_config);
 	else if (IS_CANNONLAKE(dev_priv))
 		cnl_ddi_clock_get(encoder, pipe_config);
@@ -1911,7 +1889,7 @@ bool intel_ddi_connector_get_hw_state(struct intel_connector *intel_connector)
 		goto out;
 	}
 
-	if (port == PORT_A)
+	if (HAS_TRANSCODER_EDP(dev_priv) && port == PORT_A)
 		cpu_transcoder = TRANSCODER_EDP;
 	else
 		cpu_transcoder = (enum transcoder) pipe;
@@ -1973,7 +1951,7 @@ static void intel_ddi_get_encoder_pipes(struct intel_encoder *encoder,
 	if (!(tmp & DDI_BUF_CTL_ENABLE))
 		goto out;
 
-	if (port == PORT_A) {
+	if (HAS_TRANSCODER_EDP(dev_priv) && port == PORT_A) {
 		tmp = I915_READ(TRANS_DDI_FUNC_CTL(TRANSCODER_EDP));
 
 		switch (tmp & TRANS_DDI_EDP_INPUT_MASK) {
@@ -2224,7 +2202,7 @@ u8 intel_ddi_dp_voltage_max(struct intel_encoder *encoder)
 	enum port port = encoder->port;
 	int n_entries;
 
-	if (IS_ICELAKE(dev_priv)) {
+	if (INTEL_GEN(dev_priv) >= 11) {
 		if (intel_port_is_combophy(dev_priv, port))
 			icl_get_combo_buf_trans(dev_priv, port, encoder->type,
 						intel_dp->link_rate, &n_entries);
@@ -2316,13 +2294,13 @@ static void cnl_ddi_vswing_program(struct intel_encoder *encoder,
 	/* Program PORT_TX_DW4 */
 	/* We cannot write to GRP. It would overrite individual loadgen */
 	for (ln = 0; ln < 4; ln++) {
-		val = I915_READ(CNL_PORT_TX_DW4_LN(port, ln));
+		val = I915_READ(CNL_PORT_TX_DW4_LN(ln, port));
 		val &= ~(POST_CURSOR_1_MASK | POST_CURSOR_2_MASK |
 			 CURSOR_COEFF_MASK);
 		val |= POST_CURSOR_1(ddi_translations[level].dw4_post_cursor_1);
 		val |= POST_CURSOR_2(ddi_translations[level].dw4_post_cursor_2);
 		val |= CURSOR_COEFF(ddi_translations[level].dw4_cursor_coeff);
-		I915_WRITE(CNL_PORT_TX_DW4_LN(port, ln), val);
+		I915_WRITE(CNL_PORT_TX_DW4_LN(ln, port), val);
 	}
 
 	/* Program PORT_TX_DW5 */
@@ -2378,14 +2356,14 @@ static void cnl_ddi_vswing_sequence(struct intel_encoder *encoder,
 	 * > 6 GHz (LN0=0, LN1=0, LN2=0, LN3=0)
 	 */
 	for (ln = 0; ln <= 3; ln++) {
-		val = I915_READ(CNL_PORT_TX_DW4_LN(port, ln));
+		val = I915_READ(CNL_PORT_TX_DW4_LN(ln, port));
 		val &= ~LOADGEN_SELECT;
 
 		if ((rate <= 600000 && width == 4 && ln >= 1)  ||
 		    (rate <= 600000 && width < 4 && (ln == 1 || ln == 2))) {
 			val |= LOADGEN_SELECT;
 		}
-		I915_WRITE(CNL_PORT_TX_DW4_LN(port, ln), val);
+		I915_WRITE(CNL_PORT_TX_DW4_LN(ln, port), val);
 	}
 
 	/* 3. Set PORT_CL_DW5 SUS Clock Config to 11b */
@@ -2447,13 +2425,13 @@ static void icl_ddi_combo_vswing_program(struct drm_i915_private *dev_priv,
 	/* Program PORT_TX_DW4 */
 	/* We cannot write to GRP. It would overwrite individual loadgen. */
 	for (ln = 0; ln <= 3; ln++) {
-		val = I915_READ(ICL_PORT_TX_DW4_LN(port, ln));
+		val = I915_READ(ICL_PORT_TX_DW4_LN(ln, port));
 		val &= ~(POST_CURSOR_1_MASK | POST_CURSOR_2_MASK |
 			 CURSOR_COEFF_MASK);
 		val |= POST_CURSOR_1(ddi_translations[level].dw4_post_cursor_1);
 		val |= POST_CURSOR_2(ddi_translations[level].dw4_post_cursor_2);
 		val |= CURSOR_COEFF(ddi_translations[level].dw4_cursor_coeff);
-		I915_WRITE(ICL_PORT_TX_DW4_LN(port, ln), val);
+		I915_WRITE(ICL_PORT_TX_DW4_LN(ln, port), val);
 	}
 
 	/* Program PORT_TX_DW7 */
@@ -2504,14 +2482,14 @@ static void icl_combo_phy_ddi_vswing_sequence(struct intel_encoder *encoder,
 	 * > 6 GHz (LN0=0, LN1=0, LN2=0, LN3=0)
 	 */
 	for (ln = 0; ln <= 3; ln++) {
-		val = I915_READ(ICL_PORT_TX_DW4_LN(port, ln));
+		val = I915_READ(ICL_PORT_TX_DW4_LN(ln, port));
 		val &= ~LOADGEN_SELECT;
 
 		if ((rate <= 600000 && width == 4 && ln >= 1) ||
 		    (rate <= 600000 && width < 4 && (ln == 1 || ln == 2))) {
 			val |= LOADGEN_SELECT;
 		}
-		I915_WRITE(ICL_PORT_TX_DW4_LN(port, ln), val);
+		I915_WRITE(ICL_PORT_TX_DW4_LN(ln, port), val);
 	}
 
 	/* 3. Set PORT_CL_DW5 SUS Clock Config to 11b */
@@ -2554,33 +2532,33 @@ static void icl_mg_phy_ddi_vswing_sequence(struct intel_encoder *encoder,
 
 	/* Set MG_TX_LINK_PARAMS cri_use_fs32 to 0. */
 	for (ln = 0; ln < 2; ln++) {
-		val = I915_READ(MG_TX1_LINK_PARAMS(port, ln));
+		val = I915_READ(MG_TX1_LINK_PARAMS(ln, port));
 		val &= ~CRI_USE_FS32;
-		I915_WRITE(MG_TX1_LINK_PARAMS(port, ln), val);
+		I915_WRITE(MG_TX1_LINK_PARAMS(ln, port), val);
 
-		val = I915_READ(MG_TX2_LINK_PARAMS(port, ln));
+		val = I915_READ(MG_TX2_LINK_PARAMS(ln, port));
 		val &= ~CRI_USE_FS32;
-		I915_WRITE(MG_TX2_LINK_PARAMS(port, ln), val);
+		I915_WRITE(MG_TX2_LINK_PARAMS(ln, port), val);
 	}
 
 	/* Program MG_TX_SWINGCTRL with values from vswing table */
 	for (ln = 0; ln < 2; ln++) {
-		val = I915_READ(MG_TX1_SWINGCTRL(port, ln));
+		val = I915_READ(MG_TX1_SWINGCTRL(ln, port));
 		val &= ~CRI_TXDEEMPH_OVERRIDE_17_12_MASK;
 		val |= CRI_TXDEEMPH_OVERRIDE_17_12(
 			ddi_translations[level].cri_txdeemph_override_17_12);
-		I915_WRITE(MG_TX1_SWINGCTRL(port, ln), val);
+		I915_WRITE(MG_TX1_SWINGCTRL(ln, port), val);
 
-		val = I915_READ(MG_TX2_SWINGCTRL(port, ln));
+		val = I915_READ(MG_TX2_SWINGCTRL(ln, port));
 		val &= ~CRI_TXDEEMPH_OVERRIDE_17_12_MASK;
 		val |= CRI_TXDEEMPH_OVERRIDE_17_12(
 			ddi_translations[level].cri_txdeemph_override_17_12);
-		I915_WRITE(MG_TX2_SWINGCTRL(port, ln), val);
+		I915_WRITE(MG_TX2_SWINGCTRL(ln, port), val);
 	}
 
 	/* Program MG_TX_DRVCTRL with values from vswing table */
 	for (ln = 0; ln < 2; ln++) {
-		val = I915_READ(MG_TX1_DRVCTRL(port, ln));
+		val = I915_READ(MG_TX1_DRVCTRL(ln, port));
 		val &= ~(CRI_TXDEEMPH_OVERRIDE_11_6_MASK |
 			 CRI_TXDEEMPH_OVERRIDE_5_0_MASK);
 		val |= CRI_TXDEEMPH_OVERRIDE_5_0(
@@ -2588,9 +2566,9 @@ static void icl_mg_phy_ddi_vswing_sequence(struct intel_encoder *encoder,
 			CRI_TXDEEMPH_OVERRIDE_11_6(
 				ddi_translations[level].cri_txdeemph_override_11_6) |
 			CRI_TXDEEMPH_OVERRIDE_EN;
-		I915_WRITE(MG_TX1_DRVCTRL(port, ln), val);
+		I915_WRITE(MG_TX1_DRVCTRL(ln, port), val);
 
-		val = I915_READ(MG_TX2_DRVCTRL(port, ln));
+		val = I915_READ(MG_TX2_DRVCTRL(ln, port));
 		val &= ~(CRI_TXDEEMPH_OVERRIDE_11_6_MASK |
 			 CRI_TXDEEMPH_OVERRIDE_5_0_MASK);
 		val |= CRI_TXDEEMPH_OVERRIDE_5_0(
@@ -2598,7 +2576,7 @@ static void icl_mg_phy_ddi_vswing_sequence(struct intel_encoder *encoder,
 			CRI_TXDEEMPH_OVERRIDE_11_6(
 				ddi_translations[level].cri_txdeemph_override_11_6) |
 			CRI_TXDEEMPH_OVERRIDE_EN;
-		I915_WRITE(MG_TX2_DRVCTRL(port, ln), val);
+		I915_WRITE(MG_TX2_DRVCTRL(ln, port), val);
 
 		/* FIXME: Program CRI_LOADGEN_SEL after the spec is updated */
 	}
@@ -2609,17 +2587,17 @@ static void icl_mg_phy_ddi_vswing_sequence(struct intel_encoder *encoder,
 	 * values from table for which TX1 and TX2 enabled.
 	 */
 	for (ln = 0; ln < 2; ln++) {
-		val = I915_READ(MG_CLKHUB(port, ln));
+		val = I915_READ(MG_CLKHUB(ln, port));
 		if (link_clock < 300000)
 			val |= CFG_LOW_RATE_LKREN_EN;
 		else
 			val &= ~CFG_LOW_RATE_LKREN_EN;
-		I915_WRITE(MG_CLKHUB(port, ln), val);
+		I915_WRITE(MG_CLKHUB(ln, port), val);
 	}
 
 	/* Program the MG_TX_DCC<LN, port being used> based on the link frequency */
 	for (ln = 0; ln < 2; ln++) {
-		val = I915_READ(MG_TX1_DCC(port, ln));
+		val = I915_READ(MG_TX1_DCC(ln, port));
 		val &= ~CFG_AMI_CK_DIV_OVERRIDE_VAL_MASK;
 		if (link_clock <= 500000) {
 			val &= ~CFG_AMI_CK_DIV_OVERRIDE_EN;
@@ -2627,9 +2605,9 @@ static void icl_mg_phy_ddi_vswing_sequence(struct intel_encoder *encoder,
 			val |= CFG_AMI_CK_DIV_OVERRIDE_EN |
 				CFG_AMI_CK_DIV_OVERRIDE_VAL(1);
 		}
-		I915_WRITE(MG_TX1_DCC(port, ln), val);
+		I915_WRITE(MG_TX1_DCC(ln, port), val);
 
-		val = I915_READ(MG_TX2_DCC(port, ln));
+		val = I915_READ(MG_TX2_DCC(ln, port));
 		val &= ~CFG_AMI_CK_DIV_OVERRIDE_VAL_MASK;
 		if (link_clock <= 500000) {
 			val &= ~CFG_AMI_CK_DIV_OVERRIDE_EN;
@@ -2637,18 +2615,18 @@ static void icl_mg_phy_ddi_vswing_sequence(struct intel_encoder *encoder,
 			val |= CFG_AMI_CK_DIV_OVERRIDE_EN |
 				CFG_AMI_CK_DIV_OVERRIDE_VAL(1);
 		}
-		I915_WRITE(MG_TX2_DCC(port, ln), val);
+		I915_WRITE(MG_TX2_DCC(ln, port), val);
 	}
 
 	/* Program MG_TX_PISO_READLOAD with values from vswing table */
 	for (ln = 0; ln < 2; ln++) {
-		val = I915_READ(MG_TX1_PISO_READLOAD(port, ln));
+		val = I915_READ(MG_TX1_PISO_READLOAD(ln, port));
 		val |= CRI_CALCINIT;
-		I915_WRITE(MG_TX1_PISO_READLOAD(port, ln), val);
+		I915_WRITE(MG_TX1_PISO_READLOAD(ln, port), val);
 
-		val = I915_READ(MG_TX2_PISO_READLOAD(port, ln));
+		val = I915_READ(MG_TX2_PISO_READLOAD(ln, port));
 		val |= CRI_CALCINIT;
-		I915_WRITE(MG_TX2_PISO_READLOAD(port, ln), val);
+		I915_WRITE(MG_TX2_PISO_READLOAD(ln, port), val);
 	}
 }
 
@@ -2697,7 +2675,7 @@ u32 bxt_signal_levels(struct intel_dp *intel_dp)
 	struct intel_encoder *encoder = &dport->base;
 	int level = intel_ddi_dp_level(intel_dp);
 
-	if (IS_ICELAKE(dev_priv))
+	if (INTEL_GEN(dev_priv) >= 11)
 		icl_ddi_vswing_sequence(encoder, intel_dp->link_rate,
 					level, encoder->type);
 	else if (IS_CANNONLAKE(dev_priv))
@@ -2866,7 +2844,7 @@ static void intel_ddi_clk_select(struct intel_encoder *encoder,
 
 	mutex_lock(&dev_priv->dpll_lock);
 
-	if (IS_ICELAKE(dev_priv)) {
+	if (INTEL_GEN(dev_priv) >= 11) {
 		if (!intel_port_is_combophy(dev_priv, port))
 			I915_WRITE(DDI_CLK_SEL(port),
 				   icl_pll_to_ddi_clk_sel(encoder, crtc_state));
@@ -2908,7 +2886,7 @@ static void intel_ddi_clk_disable(struct intel_encoder *encoder)
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	enum port port = encoder->port;
 
-	if (IS_ICELAKE(dev_priv)) {
+	if (INTEL_GEN(dev_priv) >= 11) {
 		if (!intel_port_is_combophy(dev_priv, port))
 			I915_WRITE(DDI_CLK_SEL(port), DDI_CLK_SEL_NONE);
 	} else if (IS_CANNONLAKE(dev_priv)) {
@@ -2927,21 +2905,20 @@ static void icl_enable_phy_clock_gating(struct intel_digital_port *dig_port)
 	struct drm_i915_private *dev_priv = to_i915(dig_port->base.base.dev);
 	enum port port = dig_port->base.port;
 	enum tc_port tc_port = intel_port_to_tc(dev_priv, port);
-	i915_reg_t mg_regs[2] = { MG_DP_MODE(port, 0), MG_DP_MODE(port, 1) };
 	u32 val;
-	int i;
+	int ln;
 
 	if (tc_port == PORT_TC_NONE)
 		return;
 
-	for (i = 0; i < ARRAY_SIZE(mg_regs); i++) {
-		val = I915_READ(mg_regs[i]);
+	for (ln = 0; ln < 2; ln++) {
+		val = I915_READ(MG_DP_MODE(ln, port));
 		val |= MG_DP_MODE_CFG_TR2PWR_GATING |
 		       MG_DP_MODE_CFG_TRPWR_GATING |
 		       MG_DP_MODE_CFG_CLNPWR_GATING |
 		       MG_DP_MODE_CFG_DIGPWR_GATING |
 		       MG_DP_MODE_CFG_GAONPWR_GATING;
-		I915_WRITE(mg_regs[i], val);
+		I915_WRITE(MG_DP_MODE(ln, port), val);
 	}
 
 	val = I915_READ(MG_MISC_SUS0(tc_port));
@@ -2960,21 +2937,20 @@ static void icl_disable_phy_clock_gating(struct intel_digital_port *dig_port)
 	struct drm_i915_private *dev_priv = to_i915(dig_port->base.base.dev);
 	enum port port = dig_port->base.port;
 	enum tc_port tc_port = intel_port_to_tc(dev_priv, port);
-	i915_reg_t mg_regs[2] = { MG_DP_MODE(port, 0), MG_DP_MODE(port, 1) };
 	u32 val;
-	int i;
+	int ln;
 
 	if (tc_port == PORT_TC_NONE)
 		return;
 
-	for (i = 0; i < ARRAY_SIZE(mg_regs); i++) {
-		val = I915_READ(mg_regs[i]);
+	for (ln = 0; ln < 2; ln++) {
+		val = I915_READ(MG_DP_MODE(ln, port));
 		val &= ~(MG_DP_MODE_CFG_TR2PWR_GATING |
 			 MG_DP_MODE_CFG_TRPWR_GATING |
 			 MG_DP_MODE_CFG_CLNPWR_GATING |
 			 MG_DP_MODE_CFG_DIGPWR_GATING |
 			 MG_DP_MODE_CFG_GAONPWR_GATING);
-		I915_WRITE(mg_regs[i], val);
+		I915_WRITE(MG_DP_MODE(ln, port), val);
 	}
 
 	val = I915_READ(MG_MISC_SUS0(tc_port));
@@ -2998,8 +2974,8 @@ static void icl_program_mg_dp_mode(struct intel_digital_port *intel_dig_port)
 	if (tc_port == PORT_TC_NONE || intel_dig_port->tc_type == TC_PORT_TBT)
 		return;
 
-	ln0 = I915_READ(MG_DP_MODE(port, 0));
-	ln1 = I915_READ(MG_DP_MODE(port, 1));
+	ln0 = I915_READ(MG_DP_MODE(0, port));
+	ln1 = I915_READ(MG_DP_MODE(1, port));
 
 	switch (intel_dig_port->tc_type) {
 	case TC_PORT_TYPEC:
@@ -3049,8 +3025,8 @@ static void icl_program_mg_dp_mode(struct intel_digital_port *intel_dig_port)
 		return;
 	}
 
-	I915_WRITE(MG_DP_MODE(port, 0), ln0);
-	I915_WRITE(MG_DP_MODE(port, 1), ln1);
+	I915_WRITE(MG_DP_MODE(0, port), ln0);
+	I915_WRITE(MG_DP_MODE(1, port), ln1);
 }
 
 static void intel_dp_sink_set_fec_ready(struct intel_dp *intel_dp,
@@ -3077,7 +3053,7 @@ static void intel_ddi_enable_fec(struct intel_encoder *encoder,
 	val |= DP_TP_CTL_FEC_ENABLE;
 	I915_WRITE(DP_TP_CTL(port), val);
 
-	if (intel_wait_for_register(dev_priv, DP_TP_STATUS(port),
+	if (intel_wait_for_register(&dev_priv->uncore, DP_TP_STATUS(port),
 				    DP_TP_STATUS_FEC_ENABLE_LIVE,
 				    DP_TP_STATUS_FEC_ENABLE_LIVE,
 				    1))
@@ -3125,7 +3101,7 @@ static void intel_ddi_pre_enable_dp(struct intel_encoder *encoder,
 	icl_program_mg_dp_mode(dig_port);
 	icl_disable_phy_clock_gating(dig_port);
 
-	if (IS_ICELAKE(dev_priv))
+	if (INTEL_GEN(dev_priv) >= 11)
 		icl_ddi_vswing_sequence(encoder, crtc_state->port_clock,
 					level, encoder->type);
 	else if (IS_CANNONLAKE(dev_priv))
@@ -3174,7 +3150,7 @@ static void intel_ddi_pre_enable_hdmi(struct intel_encoder *encoder,
 	icl_program_mg_dp_mode(dig_port);
 	icl_disable_phy_clock_gating(dig_port);
 
-	if (IS_ICELAKE(dev_priv))
+	if (INTEL_GEN(dev_priv) >= 11)
 		icl_ddi_vswing_sequence(encoder, crtc_state->port_clock,
 					level, INTEL_OUTPUT_HDMI);
 	else if (IS_CANNONLAKE(dev_priv))
@@ -3555,7 +3531,9 @@ static void intel_ddi_update_pipe_dp(struct intel_encoder *encoder,
 {
 	struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
 
-	intel_psr_enable(intel_dp, crtc_state);
+	intel_ddi_set_pipe_settings(crtc_state);
+
+	intel_psr_update(intel_dp, crtc_state);
 	intel_edp_drrs_enable(intel_dp, crtc_state);
 
 	intel_panel_update_backlight(encoder, crtc_state, conn_state);
@@ -3710,7 +3688,7 @@ static bool intel_ddi_is_audio_enabled(struct drm_i915_private *dev_priv,
 void intel_ddi_compute_min_voltage_level(struct drm_i915_private *dev_priv,
 					 struct intel_crtc_state *crtc_state)
 {
-	if (IS_ICELAKE(dev_priv) && crtc_state->port_clock > 594000)
+	if (INTEL_GEN(dev_priv) >= 11 && crtc_state->port_clock > 594000)
 		crtc_state->min_voltage_level = 1;
 	else if (IS_CANNONLAKE(dev_priv) && crtc_state->port_clock > 594000)
 		crtc_state->min_voltage_level = 2;
@@ -3763,7 +3741,10 @@ void intel_ddi_get_config(struct intel_encoder *encoder,
 		pipe_config->has_hdmi_sink = true;
 		intel_dig_port = enc_to_dig_port(&encoder->base);
 
-		if (intel_dig_port->infoframe_enabled(encoder, pipe_config))
+		pipe_config->infoframes.enable |=
+			intel_hdmi_infoframes_enabled(encoder, pipe_config);
+
+		if (pipe_config->infoframes.enable)
 			pipe_config->has_infoframe = true;
 
 		if (temp & TRANS_DDI_HDMI_SCRAMBLING)
@@ -3827,6 +3808,18 @@ void intel_ddi_get_config(struct intel_encoder *encoder,
 			bxt_ddi_phy_get_lane_lat_optim_mask(encoder);
 
 	intel_ddi_compute_min_voltage_level(dev_priv, pipe_config);
+
+	intel_hdmi_read_gcp_infoframe(encoder, pipe_config);
+
+	intel_read_infoframe(encoder, pipe_config,
+			     HDMI_INFOFRAME_TYPE_AVI,
+			     &pipe_config->infoframes.avi);
+	intel_read_infoframe(encoder, pipe_config,
+			     HDMI_INFOFRAME_TYPE_SPD,
+			     &pipe_config->infoframes.spd);
+	intel_read_infoframe(encoder, pipe_config,
+			     HDMI_INFOFRAME_TYPE_VENDOR,
+			     &pipe_config->infoframes.hdmi);
 }
 
 static enum intel_output_type
@@ -3855,7 +3848,7 @@ static int intel_ddi_compute_config(struct intel_encoder *encoder,
 	enum port port = encoder->port;
 	int ret;
 
-	if (port == PORT_A)
+	if (HAS_TRANSCODER_EDP(dev_priv) && port == PORT_A)
 		pipe_config->cpu_transcoder = TRANSCODER_EDP;
 
 	if (intel_crtc_has_type(pipe_config, INTEL_OUTPUT_HDMI))
@@ -3959,15 +3952,7 @@ static int modeset_pipe(struct drm_crtc *crtc,
 		goto out;
 	}
 
-	crtc_state->mode_changed = true;
-
-	ret = drm_atomic_add_affected_connectors(state, crtc);
-	if (ret)
-		goto out;
-
-	ret = drm_atomic_add_affected_planes(state, crtc);
-	if (ret)
-		goto out;
+	crtc_state->connectors_changed = true;
 
 	ret = drm_atomic_commit(state);
 out:
diff --git a/drivers/gpu/drm/i915/intel_ddi.h b/drivers/gpu/drm/i915/intel_ddi.h
new file mode 100644
index 000000000000..9cf69175942e
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_ddi.h
@@ -0,0 +1,53 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_DDI_H__
+#define __INTEL_DDI_H__
+
+#include <drm/i915_drm.h>
+
+#include "intel_display.h"
+
+struct drm_connector_state;
+struct drm_i915_private;
+struct intel_connector;
+struct intel_crtc;
+struct intel_crtc_state;
+struct intel_dp;
+struct intel_dpll_hw_state;
+struct intel_encoder;
+
+void intel_ddi_fdi_post_disable(struct intel_encoder *intel_encoder,
+				const struct intel_crtc_state *old_crtc_state,
+				const struct drm_connector_state *old_conn_state);
+void hsw_fdi_link_train(struct intel_crtc *crtc,
+			const struct intel_crtc_state *crtc_state);
+void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port);
+bool intel_ddi_get_hw_state(struct intel_encoder *encoder, enum pipe *pipe);
+void intel_ddi_enable_transcoder_func(const struct intel_crtc_state *crtc_state);
+void intel_ddi_disable_transcoder_func(const struct intel_crtc_state *crtc_state);
+void intel_ddi_enable_pipe_clock(const struct intel_crtc_state *crtc_state);
+void intel_ddi_disable_pipe_clock(const  struct intel_crtc_state *crtc_state);
+void intel_ddi_set_pipe_settings(const struct intel_crtc_state *crtc_state);
+void intel_ddi_prepare_link_retrain(struct intel_dp *intel_dp);
+bool intel_ddi_connector_get_hw_state(struct intel_connector *intel_connector);
+void intel_ddi_get_config(struct intel_encoder *encoder,
+			  struct intel_crtc_state *pipe_config);
+void intel_ddi_set_vc_payload_alloc(const struct intel_crtc_state *crtc_state,
+				    bool state);
+void intel_ddi_compute_min_voltage_level(struct drm_i915_private *dev_priv,
+					 struct intel_crtc_state *crtc_state);
+u32 bxt_signal_levels(struct intel_dp *intel_dp);
+u32 ddi_signal_levels(struct intel_dp *intel_dp);
+u8 intel_ddi_dp_voltage_max(struct intel_encoder *encoder);
+u8 intel_ddi_dp_pre_emphasis_max(struct intel_encoder *encoder,
+				 u8 voltage_swing);
+int intel_ddi_toggle_hdcp_signalling(struct intel_encoder *intel_encoder,
+				     bool enable);
+void icl_sanitize_encoder_pll_mapping(struct intel_encoder *encoder);
+int cnl_calc_wrpll_link(struct drm_i915_private *dev_priv,
+			struct intel_dpll_hw_state *state);
+
+#endif /* __INTEL_DDI_H__ */
diff --git a/drivers/gpu/drm/i915/intel_device_info.c b/drivers/gpu/drm/i915/intel_device_info.c
index 855a5074ad77..6af480b95bc6 100644
--- a/drivers/gpu/drm/i915/intel_device_info.c
+++ b/drivers/gpu/drm/i915/intel_device_info.c
@@ -57,6 +57,7 @@ static const char * const platform_names[] = {
 	PLATFORM_NAME(COFFEELAKE),
 	PLATFORM_NAME(CANNONLAKE),
 	PLATFORM_NAME(ICELAKE),
+	PLATFORM_NAME(ELKHARTLAKE),
 };
 #undef PLATFORM_NAME
 
@@ -155,9 +156,15 @@ static void gen11_sseu_info_init(struct drm_i915_private *dev_priv)
 	u8 eu_en;
 	int s;
 
-	sseu->max_slices = 1;
-	sseu->max_subslices = 8;
-	sseu->max_eus_per_subslice = 8;
+	if (IS_ELKHARTLAKE(dev_priv)) {
+		sseu->max_slices = 1;
+		sseu->max_subslices = 4;
+		sseu->max_eus_per_subslice = 8;
+	} else {
+		sseu->max_slices = 1;
+		sseu->max_subslices = 8;
+		sseu->max_eus_per_subslice = 8;
+	}
 
 	s_en = I915_READ(GEN11_GT_SLICE_ENABLE) & GEN11_GT_S_ENA_MASK;
 	ss_en = ~I915_READ(GEN11_GT_SUBSLICE_DISABLE);
@@ -707,6 +714,99 @@ static u32 read_timestamp_frequency(struct drm_i915_private *dev_priv)
 	return 0;
 }
 
+#undef INTEL_VGA_DEVICE
+#define INTEL_VGA_DEVICE(id, info) (id)
+
+static const u16 subplatform_ult_ids[] = {
+	INTEL_HSW_ULT_GT1_IDS(0),
+	INTEL_HSW_ULT_GT2_IDS(0),
+	INTEL_HSW_ULT_GT3_IDS(0),
+	INTEL_BDW_ULT_GT1_IDS(0),
+	INTEL_BDW_ULT_GT2_IDS(0),
+	INTEL_BDW_ULT_GT3_IDS(0),
+	INTEL_BDW_ULT_RSVD_IDS(0),
+	INTEL_SKL_ULT_GT1_IDS(0),
+	INTEL_SKL_ULT_GT2_IDS(0),
+	INTEL_SKL_ULT_GT3_IDS(0),
+	INTEL_KBL_ULT_GT1_IDS(0),
+	INTEL_KBL_ULT_GT2_IDS(0),
+	INTEL_KBL_ULT_GT3_IDS(0),
+	INTEL_CFL_U_GT2_IDS(0),
+	INTEL_CFL_U_GT3_IDS(0),
+	INTEL_WHL_U_GT1_IDS(0),
+	INTEL_WHL_U_GT2_IDS(0),
+	INTEL_WHL_U_GT3_IDS(0)
+};
+
+static const u16 subplatform_ulx_ids[] = {
+	INTEL_HSW_ULX_GT1_IDS(0),
+	INTEL_HSW_ULX_GT2_IDS(0),
+	INTEL_BDW_ULX_GT1_IDS(0),
+	INTEL_BDW_ULX_GT2_IDS(0),
+	INTEL_BDW_ULX_GT3_IDS(0),
+	INTEL_BDW_ULX_RSVD_IDS(0),
+	INTEL_SKL_ULX_GT1_IDS(0),
+	INTEL_SKL_ULX_GT2_IDS(0),
+	INTEL_KBL_ULX_GT1_IDS(0),
+	INTEL_KBL_ULX_GT2_IDS(0)
+};
+
+static const u16 subplatform_aml_ids[] = {
+	INTEL_AML_KBL_GT2_IDS(0),
+	INTEL_AML_CFL_GT2_IDS(0)
+};
+
+static const u16 subplatform_portf_ids[] = {
+	INTEL_CNL_PORT_F_IDS(0),
+	INTEL_ICL_PORT_F_IDS(0)
+};
+
+static bool find_devid(u16 id, const u16 *p, unsigned int num)
+{
+	for (; num; num--, p++) {
+		if (*p == id)
+			return true;
+	}
+
+	return false;
+}
+
+void intel_device_info_subplatform_init(struct drm_i915_private *i915)
+{
+	const struct intel_device_info *info = INTEL_INFO(i915);
+	const struct intel_runtime_info *rinfo = RUNTIME_INFO(i915);
+	const unsigned int pi = __platform_mask_index(rinfo, info->platform);
+	const unsigned int pb = __platform_mask_bit(rinfo, info->platform);
+	u16 devid = INTEL_DEVID(i915);
+	u32 mask = 0;
+
+	/* Make sure IS_<platform> checks are working. */
+	RUNTIME_INFO(i915)->platform_mask[pi] = BIT(pb);
+
+	/* Find and mark subplatform bits based on the PCI device id. */
+	if (find_devid(devid, subplatform_ult_ids,
+		       ARRAY_SIZE(subplatform_ult_ids))) {
+		mask = BIT(INTEL_SUBPLATFORM_ULT);
+	} else if (find_devid(devid, subplatform_ulx_ids,
+			      ARRAY_SIZE(subplatform_ulx_ids))) {
+		mask = BIT(INTEL_SUBPLATFORM_ULX);
+		if (IS_HASWELL(i915) || IS_BROADWELL(i915)) {
+			/* ULX machines are also considered ULT. */
+			mask |= BIT(INTEL_SUBPLATFORM_ULT);
+		}
+	} else if (find_devid(devid, subplatform_aml_ids,
+			      ARRAY_SIZE(subplatform_aml_ids))) {
+		mask = BIT(INTEL_SUBPLATFORM_AML);
+	} else if (find_devid(devid, subplatform_portf_ids,
+			      ARRAY_SIZE(subplatform_portf_ids))) {
+		mask = BIT(INTEL_SUBPLATFORM_PORTF);
+	}
+
+	GEM_BUG_ON(mask & ~INTEL_SUBPLATFORM_BITS);
+
+	RUNTIME_INFO(i915)->platform_mask[pi] |= mask;
+}
+
 /**
  * intel_device_info_runtime_init - initialize runtime info
  * @dev_priv: the i915 device
@@ -738,9 +838,9 @@ void intel_device_info_runtime_init(struct drm_i915_private *dev_priv)
 		runtime->num_scalers[PIPE_C] = 1;
 	}
 
-	BUILD_BUG_ON(I915_NUM_ENGINES > BITS_PER_TYPE(intel_ring_mask_t));
+	BUILD_BUG_ON(BITS_PER_TYPE(intel_engine_mask_t) < I915_NUM_ENGINES);
 
-	if (IS_GEN(dev_priv, 11))
+	if (INTEL_GEN(dev_priv) >= 11)
 		for_each_pipe(dev_priv, pipe)
 			runtime->num_sprites[pipe] = 6;
 	else if (IS_GEN(dev_priv, 10) || IS_GEMINILAKE(dev_priv))
@@ -844,7 +944,7 @@ void intel_device_info_runtime_init(struct drm_i915_private *dev_priv)
 
 	if (IS_GEN(dev_priv, 6) && intel_vtd_active()) {
 		DRM_INFO("Disabling ppGTT for VT-d support\n");
-		info->ppgtt = INTEL_PPGTT_NONE;
+		info->ppgtt_type = INTEL_PPGTT_NONE;
 	}
 
 	/* Initialize command stream timestamp frequency */
@@ -871,23 +971,24 @@ void intel_device_info_init_mmio(struct drm_i915_private *dev_priv)
 	unsigned int logical_vdbox = 0;
 	unsigned int i;
 	u32 media_fuse;
+	u16 vdbox_mask;
+	u16 vebox_mask;
 
 	if (INTEL_GEN(dev_priv) < 11)
 		return;
 
 	media_fuse = ~I915_READ(GEN11_GT_VEBOX_VDBOX_DISABLE);
 
-	RUNTIME_INFO(dev_priv)->vdbox_enable = media_fuse & GEN11_GT_VDBOX_DISABLE_MASK;
-	RUNTIME_INFO(dev_priv)->vebox_enable = (media_fuse & GEN11_GT_VEBOX_DISABLE_MASK) >>
-		GEN11_GT_VEBOX_DISABLE_SHIFT;
+	vdbox_mask = media_fuse & GEN11_GT_VDBOX_DISABLE_MASK;
+	vebox_mask = (media_fuse & GEN11_GT_VEBOX_DISABLE_MASK) >>
+		      GEN11_GT_VEBOX_DISABLE_SHIFT;
 
-	DRM_DEBUG_DRIVER("vdbox enable: %04x\n", RUNTIME_INFO(dev_priv)->vdbox_enable);
 	for (i = 0; i < I915_MAX_VCS; i++) {
 		if (!HAS_ENGINE(dev_priv, _VCS(i)))
 			continue;
 
-		if (!(BIT(i) & RUNTIME_INFO(dev_priv)->vdbox_enable)) {
-			info->ring_mask &= ~ENGINE_MASK(_VCS(i));
+		if (!(BIT(i) & vdbox_mask)) {
+			info->engine_mask &= ~BIT(_VCS(i));
 			DRM_DEBUG_DRIVER("vcs%u fused off\n", i);
 			continue;
 		}
@@ -899,15 +1000,20 @@ void intel_device_info_init_mmio(struct drm_i915_private *dev_priv)
 		if (logical_vdbox++ % 2 == 0)
 			RUNTIME_INFO(dev_priv)->vdbox_sfc_access |= BIT(i);
 	}
+	DRM_DEBUG_DRIVER("vdbox enable: %04x, instances: %04lx\n",
+			 vdbox_mask, VDBOX_MASK(dev_priv));
+	GEM_BUG_ON(vdbox_mask != VDBOX_MASK(dev_priv));
 
-	DRM_DEBUG_DRIVER("vebox enable: %04x\n", RUNTIME_INFO(dev_priv)->vebox_enable);
 	for (i = 0; i < I915_MAX_VECS; i++) {
 		if (!HAS_ENGINE(dev_priv, _VECS(i)))
 			continue;
 
-		if (!(BIT(i) & RUNTIME_INFO(dev_priv)->vebox_enable)) {
-			info->ring_mask &= ~ENGINE_MASK(_VECS(i));
+		if (!(BIT(i) & vebox_mask)) {
+			info->engine_mask &= ~BIT(_VECS(i));
 			DRM_DEBUG_DRIVER("vecs%u fused off\n", i);
 		}
 	}
+	DRM_DEBUG_DRIVER("vebox enable: %04x, instances: %04lx\n",
+			 vebox_mask, VEBOX_MASK(dev_priv));
+	GEM_BUG_ON(vebox_mask != VEBOX_MASK(dev_priv));
 }
diff --git a/drivers/gpu/drm/i915/intel_device_info.h b/drivers/gpu/drm/i915/intel_device_info.h
index e8b8661df746..0e579f158016 100644
--- a/drivers/gpu/drm/i915/intel_device_info.h
+++ b/drivers/gpu/drm/i915/intel_device_info.h
@@ -27,6 +27,7 @@
 
 #include <uapi/drm/i915_drm.h>
 
+#include "intel_engine_types.h"
 #include "intel_display.h"
 
 struct drm_printer;
@@ -73,14 +74,29 @@ enum intel_platform {
 	INTEL_CANNONLAKE,
 	/* gen11 */
 	INTEL_ICELAKE,
+	INTEL_ELKHARTLAKE,
 	INTEL_MAX_PLATFORMS
 };
 
-enum intel_ppgtt {
+/*
+ * Subplatform bits share the same namespace per parent platform. In other words
+ * it is fine for the same bit to be used on multiple parent platforms.
+ */
+
+#define INTEL_SUBPLATFORM_BITS (3)
+
+/* HSW/BDW/SKL/KBL/CFL */
+#define INTEL_SUBPLATFORM_ULT	(0)
+#define INTEL_SUBPLATFORM_ULX	(1)
+#define INTEL_SUBPLATFORM_AML	(2)
+
+/* CNL/ICL */
+#define INTEL_SUBPLATFORM_PORTF	(0)
+
+enum intel_ppgtt_type {
 	INTEL_PPGTT_NONE = I915_GEM_PPGTT_NONE,
 	INTEL_PPGTT_ALIASING = I915_GEM_PPGTT_ALIASING,
 	INTEL_PPGTT_FULL = I915_GEM_PPGTT_FULL,
-	INTEL_PPGTT_FULL_4LVL,
 };
 
 #define DEV_INFO_FOR_EACH_FLAG(func) \
@@ -150,19 +166,18 @@ struct sseu_dev_info {
 	u8 eu_mask[GEN_MAX_SLICES * GEN_MAX_SUBSLICES];
 };
 
-typedef u8 intel_ring_mask_t;
-
 struct intel_device_info {
 	u16 gen_mask;
 
 	u8 gen;
 	u8 gt; /* GT number, 0 if undefined */
-	intel_ring_mask_t ring_mask; /* Rings supported by the HW */
+	intel_engine_mask_t engine_mask; /* Engines supported by the HW */
 
 	enum intel_platform platform;
-	u32 platform_mask;
 
-	enum intel_ppgtt ppgtt;
+	enum intel_ppgtt_type ppgtt_type;
+	unsigned int ppgtt_size; /* log2, e.g. 31/32/48 bits */
+
 	unsigned int page_sizes; /* page sizes supported by the HW */
 
 	u32 display_mmio_offset;
@@ -195,22 +210,28 @@ struct intel_device_info {
 };
 
 struct intel_runtime_info {
+	/*
+	 * Platform mask is used for optimizing or-ed IS_PLATFORM calls into
+	 * into single runtime conditionals, and also to provide groundwork
+	 * for future per platform, or per SKU build optimizations.
+	 *
+	 * Array can be extended when necessary if the corresponding
+	 * BUILD_BUG_ON is hit.
+	 */
+	u32 platform_mask[2];
+
 	u16 device_id;
 
 	u8 num_sprites[I915_MAX_PIPES];
 	u8 num_scalers[I915_MAX_PIPES];
 
-	u8 num_rings;
+	u8 num_engines;
 
 	/* Slice/subslice/EU info */
 	struct sseu_dev_info sseu;
 
 	u32 cs_timestamp_frequency_khz;
 
-	/* Enabled (not fused off) media engine bitmasks. */
-	u8 vdbox_enable;
-	u8 vebox_enable;
-
 	/* Media engine access to SFC per instance */
 	u8 vdbox_sfc_access;
 };
@@ -269,6 +290,7 @@ static inline void sseu_set_eus(struct sseu_dev_info *sseu,
 
 const char *intel_platform_name(enum intel_platform platform);
 
+void intel_device_info_subplatform_init(struct drm_i915_private *dev_priv);
 void intel_device_info_runtime_init(struct drm_i915_private *dev_priv);
 void intel_device_info_dump_flags(const struct intel_device_info *info,
 				  struct drm_printer *p);
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 421aac80a838..3bd40a4a6739 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -46,19 +46,29 @@
 
 #include "i915_drv.h"
 #include "i915_gem_clflush.h"
+#include "i915_reset.h"
 #include "i915_trace.h"
+#include "intel_atomic_plane.h"
+#include "intel_color.h"
+#include "intel_cdclk.h"
+#include "intel_crt.h"
+#include "intel_ddi.h"
+#include "intel_dp.h"
 #include "intel_drv.h"
 #include "intel_dsi.h"
+#include "intel_dvo.h"
+#include "intel_fbc.h"
+#include "intel_fbdev.h"
 #include "intel_frontbuffer.h"
-
-#include "intel_drv.h"
-#include "intel_dsi.h"
-#include "intel_frontbuffer.h"
-
-#include "i915_drv.h"
-#include "i915_gem_clflush.h"
-#include "i915_reset.h"
-#include "i915_trace.h"
+#include "intel_hdcp.h"
+#include "intel_hdmi.h"
+#include "intel_lvds.h"
+#include "intel_pipe_crc.h"
+#include "intel_pm.h"
+#include "intel_psr.h"
+#include "intel_sdvo.h"
+#include "intel_sprite.h"
+#include "intel_tv.h"
 
 /* Primary plane formats for gen <= 3 */
 static const u32 i8xx_primary_formats[] = {
@@ -115,8 +125,8 @@ static void vlv_prepare_pll(struct intel_crtc *crtc,
 			    const struct intel_crtc_state *pipe_config);
 static void chv_prepare_pll(struct intel_crtc *crtc,
 			    const struct intel_crtc_state *pipe_config);
-static void intel_begin_crtc_commit(struct drm_crtc *, struct drm_crtc_state *);
-static void intel_finish_crtc_commit(struct drm_crtc *, struct drm_crtc_state *);
+static void intel_begin_crtc_commit(struct intel_atomic_state *, struct intel_crtc *);
+static void intel_finish_crtc_commit(struct intel_atomic_state *, struct intel_crtc *);
 static void intel_crtc_init_scalers(struct intel_crtc *crtc,
 				    struct intel_crtc_state *crtc_state);
 static void skylake_pfit_enable(const struct intel_crtc_state *crtc_state);
@@ -467,10 +477,11 @@ static const struct intel_limit intel_limits_bxt = {
 };
 
 static void
-skl_wa_clkgate(struct drm_i915_private *dev_priv, int pipe, bool enable)
+skl_wa_827(struct drm_i915_private *dev_priv, int pipe, bool enable)
 {
 	if (enable)
 		I915_WRITE(CLKGATE_DIS_PSL(pipe),
+			   I915_READ(CLKGATE_DIS_PSL(pipe)) |
 			   DUPS1_GATING_DIS | DUPS2_GATING_DIS);
 	else
 		I915_WRITE(CLKGATE_DIS_PSL(pipe),
@@ -595,7 +606,7 @@ i9xx_select_p2_div(const struct intel_limit *limit,
 		   const struct intel_crtc_state *crtc_state,
 		   int target)
 {
-	struct drm_device *dev = crtc_state->base.crtc->dev;
+	struct drm_i915_private *dev_priv = to_i915(crtc_state->base.crtc->dev);
 
 	if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_LVDS)) {
 		/*
@@ -603,7 +614,7 @@ i9xx_select_p2_div(const struct intel_limit *limit,
 		 * We haven't figured out how to reliably set up different
 		 * single/dual channel state, if we even can.
 		 */
-		if (intel_is_dual_link_lvds(dev))
+		if (intel_is_dual_link_lvds(dev_priv))
 			return limit->p2.p2_fast;
 		else
 			return limit->p2.p2_slow;
@@ -951,14 +962,15 @@ chv_find_best_dpll(const struct intel_limit *limit,
 	return found;
 }
 
-bool bxt_find_best_dpll(struct intel_crtc_state *crtc_state, int target_clock,
+bool bxt_find_best_dpll(struct intel_crtc_state *crtc_state,
 			struct dpll *best_clock)
 {
 	int refclk = 100000;
 	const struct intel_limit *limit = &intel_limits_bxt;
 
 	return chv_find_best_dpll(limit, crtc_state,
-				  target_clock, refclk, NULL, best_clock);
+				  crtc_state->port_clock, refclk,
+				  NULL, best_clock);
 }
 
 bool intel_crtc_active(struct intel_crtc *crtc)
@@ -1039,7 +1051,7 @@ intel_wait_for_pipe_off(const struct intel_crtc_state *old_crtc_state)
 		i915_reg_t reg = PIPECONF(cpu_transcoder);
 
 		/* Wait for the Pipe State to go off */
-		if (intel_wait_for_register(dev_priv,
+		if (intel_wait_for_register(&dev_priv->uncore,
 					    reg, I965_PIPECONF_ACTIVE, 0,
 					    100))
 			WARN(1, "pipe_off wait timed out\n");
@@ -1345,7 +1357,7 @@ static void _vlv_enable_pll(struct intel_crtc *crtc,
 	POSTING_READ(DPLL(pipe));
 	udelay(150);
 
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    DPLL(pipe),
 				    DPLL_LOCK_VLV,
 				    DPLL_LOCK_VLV,
@@ -1398,7 +1410,7 @@ static void _chv_enable_pll(struct intel_crtc *crtc,
 	I915_WRITE(DPLL(pipe), pipe_config->dpll_hw_state.dpll);
 
 	/* Check PLL is locked */
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    DPLL(pipe), DPLL_LOCK_VLV, DPLL_LOCK_VLV,
 				    1))
 		DRM_ERROR("PLL %d failed to lock\n", pipe);
@@ -1441,17 +1453,12 @@ static void chv_enable_pll(struct intel_crtc *crtc,
 	}
 }
 
-static int intel_num_dvo_pipes(struct drm_i915_private *dev_priv)
+static bool i9xx_has_pps(struct drm_i915_private *dev_priv)
 {
-	struct intel_crtc *crtc;
-	int count = 0;
-
-	for_each_intel_crtc(&dev_priv->drm, crtc) {
-		count += crtc->base.state->active &&
-			intel_crtc_has_type(crtc->config, INTEL_OUTPUT_DVO);
-	}
+	if (IS_I830(dev_priv))
+		return false;
 
-	return count;
+	return IS_PINEVIEW(dev_priv) || IS_MOBILE(dev_priv);
 }
 
 static void i9xx_enable_pll(struct intel_crtc *crtc,
@@ -1465,29 +1472,15 @@ static void i9xx_enable_pll(struct intel_crtc *crtc,
 	assert_pipe_disabled(dev_priv, crtc->pipe);
 
 	/* PLL is protected by panel, make sure we can write it */
-	if (IS_MOBILE(dev_priv) && !IS_I830(dev_priv))
+	if (i9xx_has_pps(dev_priv))
 		assert_panel_unlocked(dev_priv, crtc->pipe);
 
-	/* Enable DVO 2x clock on both PLLs if necessary */
-	if (IS_I830(dev_priv) && intel_num_dvo_pipes(dev_priv) > 0) {
-		/*
-		 * It appears to be important that we don't enable this
-		 * for the current pipe before otherwise configuring the
-		 * PLL. No idea how this should be handled if multiple
-		 * DVO outputs are enabled simultaneosly.
-		 */
-		dpll |= DPLL_DVO_2X_MODE;
-		I915_WRITE(DPLL(!crtc->pipe),
-			   I915_READ(DPLL(!crtc->pipe)) | DPLL_DVO_2X_MODE);
-	}
-
 	/*
 	 * Apparently we need to have VGA mode enabled prior to changing
 	 * the P1/P2 dividers. Otherwise the DPLL will keep using the old
 	 * dividers, even though the register value does change.
 	 */
-	I915_WRITE(reg, 0);
-
+	I915_WRITE(reg, dpll & ~DPLL_VGA_MODE_DIS);
 	I915_WRITE(reg, dpll);
 
 	/* Wait for the clocks to stabilize. */
@@ -1520,16 +1513,6 @@ static void i9xx_disable_pll(const struct intel_crtc_state *crtc_state)
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
 
-	/* Disable DVO 2x clock on both PLLs if necessary */
-	if (IS_I830(dev_priv) &&
-	    intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DVO) &&
-	    !intel_num_dvo_pipes(dev_priv)) {
-		I915_WRITE(DPLL(PIPE_B),
-			   I915_READ(DPLL(PIPE_B)) & ~DPLL_DVO_2X_MODE);
-		I915_WRITE(DPLL(PIPE_A),
-			   I915_READ(DPLL(PIPE_A)) & ~DPLL_DVO_2X_MODE);
-	}
-
 	/* Don't disable pipe or pipe PLLs if needed */
 	if (IS_I830(dev_priv))
 		return;
@@ -1608,7 +1591,7 @@ void vlv_wait_port_ready(struct drm_i915_private *dev_priv,
 		BUG();
 	}
 
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    dpll_reg, port_mask, expected_mask,
 				    1000))
 		WARN(1, "timed out waiting for port %c ready: got 0x%x, expected 0x%x\n",
@@ -1658,17 +1641,18 @@ static void ironlake_enable_pch_transcoder(const struct intel_crtc_state *crtc_s
 	}
 
 	val &= ~TRANS_INTERLACE_MASK;
-	if ((pipeconf_val & PIPECONF_INTERLACE_MASK) == PIPECONF_INTERLACED_ILK)
+	if ((pipeconf_val & PIPECONF_INTERLACE_MASK) == PIPECONF_INTERLACED_ILK) {
 		if (HAS_PCH_IBX(dev_priv) &&
 		    intel_crtc_has_type(crtc_state, INTEL_OUTPUT_SDVO))
 			val |= TRANS_LEGACY_INTERLACED_ILK;
 		else
 			val |= TRANS_INTERLACED;
-	else
+	} else {
 		val |= TRANS_PROGRESSIVE;
+	}
 
 	I915_WRITE(reg, val | TRANS_ENABLE);
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    reg, TRANS_STATE_ENABLE, TRANS_STATE_ENABLE,
 				    100))
 		DRM_ERROR("failed to enable transcoder %c\n", pipe_name(pipe));
@@ -1698,7 +1682,7 @@ static void lpt_enable_pch_transcoder(struct drm_i915_private *dev_priv,
 		val |= TRANS_PROGRESSIVE;
 
 	I915_WRITE(LPT_TRANSCONF, val);
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    LPT_TRANSCONF,
 				    TRANS_STATE_ENABLE,
 				    TRANS_STATE_ENABLE,
@@ -1724,7 +1708,7 @@ static void ironlake_disable_pch_transcoder(struct drm_i915_private *dev_priv,
 	val &= ~TRANS_ENABLE;
 	I915_WRITE(reg, val);
 	/* wait for PCH transcoder off, transcoder state */
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    reg, TRANS_STATE_ENABLE, 0,
 				    50))
 		DRM_ERROR("failed to disable transcoder %c\n", pipe_name(pipe));
@@ -1746,7 +1730,7 @@ void lpt_disable_pch_transcoder(struct drm_i915_private *dev_priv)
 	val &= ~TRANS_ENABLE;
 	I915_WRITE(LPT_TRANSCONF, val);
 	/* wait for PCH transcoder off, transcoder state */
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    LPT_TRANSCONF, TRANS_STATE_ENABLE, 0,
 				    50))
 		DRM_ERROR("Failed to disable PCH transcoder\n");
@@ -1830,6 +1814,8 @@ static void intel_enable_pipe(const struct intel_crtc_state *new_crtc_state)
 		/* FIXME: assert CPU port conditions for SNB+ */
 	}
 
+	trace_intel_pipe_enable(dev_priv, pipe);
+
 	reg = PIPECONF(cpu_transcoder);
 	val = I915_READ(reg);
 	if (val & PIPECONF_ENABLE) {
@@ -1869,6 +1855,8 @@ static void intel_disable_pipe(const struct intel_crtc_state *old_crtc_state)
 	 */
 	assert_planes_disabled(crtc);
 
+	trace_intel_pipe_disable(dev_priv, pipe);
+
 	reg = PIPECONF(cpu_transcoder);
 	val = I915_READ(reg);
 	if ((val & PIPECONF_ENABLE) == 0)
@@ -2677,6 +2665,24 @@ int skl_format_to_fourcc(int format, bool rgb_order, bool alpha)
 		return DRM_FORMAT_RGB565;
 	case PLANE_CTL_FORMAT_NV12:
 		return DRM_FORMAT_NV12;
+	case PLANE_CTL_FORMAT_P010:
+		return DRM_FORMAT_P010;
+	case PLANE_CTL_FORMAT_P012:
+		return DRM_FORMAT_P012;
+	case PLANE_CTL_FORMAT_P016:
+		return DRM_FORMAT_P016;
+	case PLANE_CTL_FORMAT_Y210:
+		return DRM_FORMAT_Y210;
+	case PLANE_CTL_FORMAT_Y212:
+		return DRM_FORMAT_Y212;
+	case PLANE_CTL_FORMAT_Y216:
+		return DRM_FORMAT_Y216;
+	case PLANE_CTL_FORMAT_Y410:
+		return DRM_FORMAT_XVYU2101010;
+	case PLANE_CTL_FORMAT_Y412:
+		return DRM_FORMAT_XVYU12_16161616;
+	case PLANE_CTL_FORMAT_Y416:
+		return DRM_FORMAT_XVYU16161616;
 	default:
 	case PLANE_CTL_FORMAT_XRGB_8888:
 		if (rgb_order) {
@@ -2695,6 +2701,18 @@ int skl_format_to_fourcc(int format, bool rgb_order, bool alpha)
 			return DRM_FORMAT_XBGR2101010;
 		else
 			return DRM_FORMAT_XRGB2101010;
+	case PLANE_CTL_FORMAT_XRGB_16161616F:
+		if (rgb_order) {
+			if (alpha)
+				return DRM_FORMAT_ABGR16161616F;
+			else
+				return DRM_FORMAT_XBGR16161616F;
+		} else {
+			if (alpha)
+				return DRM_FORMAT_ARGB16161616F;
+			else
+				return DRM_FORMAT_XRGB16161616F;
+		}
 	}
 }
 
@@ -2825,8 +2843,7 @@ static void intel_plane_disable_noatomic(struct intel_crtc *crtc,
 	if (plane->id == PLANE_PRIMARY)
 		intel_pre_disable_primary_noatomic(&crtc->base);
 
-	trace_intel_disable_plane(&plane->base, crtc);
-	plane->disable_plane(plane, crtc_state);
+	intel_disable_plane(plane, crtc_state);
 }
 
 static void
@@ -3176,7 +3193,7 @@ int skl_check_plane_surface(struct intel_plane_state *plane_state)
 	 * Handle the AUX surface first since
 	 * the main surface setup depends on it.
 	 */
-	if (fb->format->format == DRM_FORMAT_NV12) {
+	if (is_planar_yuv_format(fb->format->format)) {
 		ret = skl_check_nv12_aux_surface(plane_state);
 		if (ret)
 			return ret;
@@ -3230,9 +3247,10 @@ static u32 i9xx_plane_ctl_crtc(const struct intel_crtc_state *crtc_state)
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	u32 dspcntr = 0;
 
-	dspcntr |= DISPPLANE_GAMMA_ENABLE;
+	if (crtc_state->gamma_enable)
+		dspcntr |= DISPPLANE_GAMMA_ENABLE;
 
-	if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
+	if (crtc_state->csc_enable)
 		dspcntr |= DISPPLANE_PIPE_CSC_ENABLE;
 
 	if (INTEL_GEN(dev_priv) < 5)
@@ -3459,7 +3477,7 @@ static void i9xx_disable_plane(struct intel_plane *plane,
 	 *
 	 * On pre-g4x there is no way to gamma correct the
 	 * pipe bottom color but we'll keep on doing this
-	 * anyway.
+	 * anyway so that the crtc state readout works correctly.
 	 */
 	dspcntr = i9xx_plane_ctl_crtc(crtc_state);
 
@@ -3590,6 +3608,12 @@ static u32 skl_plane_ctl_format(u32 pixel_format)
 		return PLANE_CTL_FORMAT_XRGB_2101010;
 	case DRM_FORMAT_XBGR2101010:
 		return PLANE_CTL_ORDER_RGBX | PLANE_CTL_FORMAT_XRGB_2101010;
+	case DRM_FORMAT_XBGR16161616F:
+	case DRM_FORMAT_ABGR16161616F:
+		return PLANE_CTL_FORMAT_XRGB_16161616F | PLANE_CTL_ORDER_RGBX;
+	case DRM_FORMAT_XRGB16161616F:
+	case DRM_FORMAT_ARGB16161616F:
+		return PLANE_CTL_FORMAT_XRGB_16161616F;
 	case DRM_FORMAT_YUYV:
 		return PLANE_CTL_FORMAT_YUV422 | PLANE_CTL_YUV422_YUYV;
 	case DRM_FORMAT_YVYU:
@@ -3600,6 +3624,24 @@ static u32 skl_plane_ctl_format(u32 pixel_format)
 		return PLANE_CTL_FORMAT_YUV422 | PLANE_CTL_YUV422_VYUY;
 	case DRM_FORMAT_NV12:
 		return PLANE_CTL_FORMAT_NV12;
+	case DRM_FORMAT_P010:
+		return PLANE_CTL_FORMAT_P010;
+	case DRM_FORMAT_P012:
+		return PLANE_CTL_FORMAT_P012;
+	case DRM_FORMAT_P016:
+		return PLANE_CTL_FORMAT_P016;
+	case DRM_FORMAT_Y210:
+		return PLANE_CTL_FORMAT_Y210;
+	case DRM_FORMAT_Y212:
+		return PLANE_CTL_FORMAT_Y212;
+	case DRM_FORMAT_Y216:
+		return PLANE_CTL_FORMAT_Y216;
+	case DRM_FORMAT_XVYU2101010:
+		return PLANE_CTL_FORMAT_Y410;
+	case DRM_FORMAT_XVYU12_16161616:
+		return PLANE_CTL_FORMAT_Y412;
+	case DRM_FORMAT_XVYU16161616:
+		return PLANE_CTL_FORMAT_Y416;
 	default:
 		MISSING_CASE(pixel_format);
 	}
@@ -3710,8 +3752,11 @@ u32 skl_plane_ctl_crtc(const struct intel_crtc_state *crtc_state)
 	if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv))
 		return plane_ctl;
 
-	plane_ctl |= PLANE_CTL_PIPE_GAMMA_ENABLE;
-	plane_ctl |= PLANE_CTL_PIPE_CSC_ENABLE;
+	if (crtc_state->gamma_enable)
+		plane_ctl |= PLANE_CTL_PIPE_GAMMA_ENABLE;
+
+	if (crtc_state->csc_enable)
+		plane_ctl |= PLANE_CTL_PIPE_CSC_ENABLE;
 
 	return plane_ctl;
 }
@@ -3763,8 +3808,11 @@ u32 glk_plane_color_ctl_crtc(const struct intel_crtc_state *crtc_state)
 	if (INTEL_GEN(dev_priv) >= 11)
 		return plane_color_ctl;
 
-	plane_color_ctl |= PLANE_COLOR_PIPE_GAMMA_ENABLE;
-	plane_color_ctl |= PLANE_COLOR_PIPE_CSC_ENABLE;
+	if (crtc_state->gamma_enable)
+		plane_color_ctl |= PLANE_COLOR_PIPE_GAMMA_ENABLE;
+
+	if (crtc_state->csc_enable)
+		plane_color_ctl |= PLANE_COLOR_PIPE_CSC_ENABLE;
 
 	return plane_color_ctl;
 }
@@ -3772,6 +3820,8 @@ u32 glk_plane_color_ctl_crtc(const struct intel_crtc_state *crtc_state)
 u32 glk_plane_color_ctl(const struct intel_crtc_state *crtc_state,
 			const struct intel_plane_state *plane_state)
 {
+	struct drm_i915_private *dev_priv =
+		to_i915(plane_state->base.plane->dev);
 	const struct drm_framebuffer *fb = plane_state->base.fb;
 	struct intel_plane *plane = to_intel_plane(plane_state->base.plane);
 	u32 plane_color_ctl = 0;
@@ -3779,7 +3829,7 @@ u32 glk_plane_color_ctl(const struct intel_crtc_state *crtc_state,
 	plane_color_ctl |= PLANE_COLOR_PLANE_GAMMA_DISABLE;
 	plane_color_ctl |= glk_plane_color_ctl_alpha(plane_state);
 
-	if (fb->format->is_yuv && !icl_is_hdr_plane(plane)) {
+	if (fb->format->is_yuv && !icl_is_hdr_plane(dev_priv, plane->id)) {
 		if (plane_state->base.color_encoding == DRM_COLOR_YCBCR_BT709)
 			plane_color_ctl |= PLANE_COLOR_CSC_MODE_YUV709_TO_RGB709;
 		else
@@ -3921,9 +3971,6 @@ void intel_finish_reset(struct drm_i915_private *dev_priv)
 		 * The display has been reset as well,
 		 * so need a full re-initialization.
 		 */
-		intel_runtime_pm_disable_interrupts(dev_priv);
-		intel_runtime_pm_enable_interrupts(dev_priv);
-
 		intel_pps_unlock_regs_wa(dev_priv);
 		intel_modeset_init_hw(dev);
 		intel_init_clock_gating(dev_priv);
@@ -3963,13 +4010,13 @@ static void icl_set_pipe_chicken(struct intel_crtc *crtc)
 	 * and rounding for per-pixel values 00 and 0xff
 	 */
 	tmp |= PER_PIXEL_ALPHA_BYPASS_EN;
-
 	/*
-	 * W/A for underruns with linear/X-tiled with
-	 * WM1+ disabled.
+	 * Display WA # 1605353570: icl
+	 * Set the pixel rounding bit to 1 for allowing
+	 * passthrough of Frame buffer pixels unmodified
+	 * across pipe
 	 */
-	tmp |= PM_FILL_MAINTAIN_DBUF_FULLNESS;
-
+	tmp |= PIXEL_ROUNDING_TRUNC_FB_PASSTHRU;
 	I915_WRITE(PIPE_CHICKEN(pipe), tmp);
 }
 
@@ -4008,16 +4055,6 @@ static void intel_update_pipe_config(const struct intel_crtc_state *old_crtc_sta
 			ironlake_pfit_disable(old_crtc_state);
 	}
 
-	/*
-	 * We don't (yet) allow userspace to control the pipe background color,
-	 * so force it to black, but apply pipe gamma and CSC so that its
-	 * handling will match how we program our planes.
-	 */
-	if (INTEL_GEN(dev_priv) >= 9)
-		I915_WRITE(SKL_BOTTOM_COLOR(crtc->pipe),
-			   SKL_BOTTOM_COLOR_GAMMA_ENABLE |
-			   SKL_BOTTOM_COLOR_CSC_ENABLE);
-
 	if (INTEL_GEN(dev_priv) >= 11)
 		icl_set_pipe_chicken(crtc);
 }
@@ -5036,19 +5073,19 @@ skl_update_scaler(struct intel_crtc_state *crtc_state, bool force_detach,
 		return 0;
 	}
 
-	if (format && format->format == DRM_FORMAT_NV12 &&
+	if (format && is_planar_yuv_format(format->format) &&
 	    (src_h < SKL_MIN_YUV_420_SRC_H || src_w < SKL_MIN_YUV_420_SRC_W)) {
-		DRM_DEBUG_KMS("NV12: src dimensions not met\n");
+		DRM_DEBUG_KMS("Planar YUV: src dimensions not met\n");
 		return -EINVAL;
 	}
 
 	/* range checks */
 	if (src_w < SKL_MIN_SRC_W || src_h < SKL_MIN_SRC_H ||
 	    dst_w < SKL_MIN_DST_W || dst_h < SKL_MIN_DST_H ||
-	    (IS_GEN(dev_priv, 11) &&
+	    (INTEL_GEN(dev_priv) >= 11 &&
 	     (src_w > ICL_MAX_SRC_W || src_h > ICL_MAX_SRC_H ||
 	      dst_w > ICL_MAX_DST_W || dst_h > ICL_MAX_DST_H)) ||
-	    (!IS_GEN(dev_priv, 11) &&
+	    (INTEL_GEN(dev_priv) < 11 &&
 	     (src_w > SKL_MAX_SRC_W || src_h > SKL_MAX_SRC_H ||
 	      dst_w > SKL_MAX_DST_W || dst_h > SKL_MAX_DST_H)))	{
 		DRM_DEBUG_KMS("scaler_user index %u.%u: src %ux%u dst %ux%u "
@@ -5105,14 +5142,15 @@ static int skl_update_scaler_plane(struct intel_crtc_state *crtc_state,
 {
 	struct intel_plane *intel_plane =
 		to_intel_plane(plane_state->base.plane);
+	struct drm_i915_private *dev_priv = to_i915(intel_plane->base.dev);
 	struct drm_framebuffer *fb = plane_state->base.fb;
 	int ret;
 	bool force_detach = !fb || !plane_state->base.visible;
 	bool need_scaler = false;
 
 	/* Pre-gen11 and SDR planes always need a scaler for planar formats. */
-	if (!icl_is_hdr_plane(intel_plane) &&
-	    fb && fb->format->format == DRM_FORMAT_NV12)
+	if (!icl_is_hdr_plane(dev_priv, intel_plane->id) &&
+	    fb && is_planar_yuv_format(fb->format->format))
 		need_scaler = true;
 
 	ret = skl_update_scaler(crtc_state, force_detach,
@@ -5144,11 +5182,24 @@ static int skl_update_scaler_plane(struct intel_crtc_state *crtc_state,
 	case DRM_FORMAT_ARGB8888:
 	case DRM_FORMAT_XRGB2101010:
 	case DRM_FORMAT_XBGR2101010:
+	case DRM_FORMAT_XBGR16161616F:
+	case DRM_FORMAT_ABGR16161616F:
+	case DRM_FORMAT_XRGB16161616F:
+	case DRM_FORMAT_ARGB16161616F:
 	case DRM_FORMAT_YUYV:
 	case DRM_FORMAT_YVYU:
 	case DRM_FORMAT_UYVY:
 	case DRM_FORMAT_VYUY:
 	case DRM_FORMAT_NV12:
+	case DRM_FORMAT_P010:
+	case DRM_FORMAT_P012:
+	case DRM_FORMAT_P016:
+	case DRM_FORMAT_Y210:
+	case DRM_FORMAT_Y212:
+	case DRM_FORMAT_Y216:
+	case DRM_FORMAT_XVYU2101010:
+	case DRM_FORMAT_XVYU12_16161616:
+	case DRM_FORMAT_XVYU16161616:
 		break;
 	default:
 		DRM_DEBUG_KMS("[PLANE:%d:%s] FB:%d unsupported scaling format 0x%x\n",
@@ -5259,7 +5310,7 @@ void hsw_enable_ips(const struct intel_crtc_state *crtc_state)
 		 * and don't wait for vblanks until the end of crtc_enable, then
 		 * the HW state readout code will complain that the expected
 		 * IPS_CTL value is not the one we read. */
-		if (intel_wait_for_register(dev_priv,
+		if (intel_wait_for_register(&dev_priv->uncore,
 					    IPS_CTL, IPS_ENABLE, IPS_ENABLE,
 					    50))
 			DRM_ERROR("Timed out waiting for IPS enable\n");
@@ -5284,7 +5335,7 @@ void hsw_disable_ips(const struct intel_crtc_state *crtc_state)
 		 * 42ms timeout value leads to occasional timeouts so use 100ms
 		 * instead.
 		 */
-		if (intel_wait_for_register(dev_priv,
+		if (intel_wait_for_register(&dev_priv->uncore,
 					    IPS_CTL, IPS_ENABLE, 0,
 					    100))
 			DRM_ERROR("Timed out waiting for IPS disable\n");
@@ -5490,7 +5541,7 @@ static void intel_post_plane_update(struct intel_crtc_state *old_crtc_state)
 	/* Display WA 827 */
 	if (needs_nv12_wa(dev_priv, old_crtc_state) &&
 	    !needs_nv12_wa(dev_priv, pipe_config)) {
-		skl_wa_clkgate(dev_priv, crtc->pipe, false);
+		skl_wa_827(dev_priv, crtc->pipe, false);
 	}
 }
 
@@ -5529,7 +5580,7 @@ static void intel_pre_plane_update(struct intel_crtc_state *old_crtc_state,
 	/* Display WA 827 */
 	if (!needs_nv12_wa(dev_priv, old_crtc_state) &&
 	    needs_nv12_wa(dev_priv, pipe_config)) {
-		skl_wa_clkgate(dev_priv, crtc->pipe, true);
+		skl_wa_827(dev_priv, crtc->pipe, true);
 	}
 
 	/*
@@ -5603,7 +5654,7 @@ static void intel_crtc_disable_planes(struct intel_atomic_state *state,
 		    !(update_mask & BIT(plane->id)))
 			continue;
 
-		plane->disable_plane(plane, new_crtc_state);
+		intel_disable_plane(plane, new_crtc_state);
 
 		if (old_plane_state->base.visible)
 			fb_bits |= plane->frontbuffer_bit;
@@ -5754,6 +5805,14 @@ static void intel_encoders_update_pipe(struct drm_crtc *crtc,
 	}
 }
 
+static void intel_disable_primary_plane(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
+	struct intel_plane *plane = to_intel_plane(crtc->base.primary);
+
+	plane->disable_plane(plane, crtc_state);
+}
+
 static void ironlake_crtc_enable(struct intel_crtc_state *pipe_config,
 				 struct drm_atomic_state *old_state)
 {
@@ -5819,6 +5878,8 @@ static void ironlake_crtc_enable(struct intel_crtc_state *pipe_config,
 	 */
 	intel_color_load_luts(pipe_config);
 	intel_color_commit(pipe_config);
+	/* update DSPCNTR to configure gamma for pipe bottom color */
+	intel_disable_primary_plane(pipe_config);
 
 	if (dev_priv->display.initial_watermarks != NULL)
 		dev_priv->display.initial_watermarks(old_intel_state, pipe_config);
@@ -5947,6 +6008,9 @@ static void haswell_crtc_enable(struct intel_crtc_state *pipe_config,
 	 */
 	intel_color_load_luts(pipe_config);
 	intel_color_commit(pipe_config);
+	/* update DSPCNTR to configure gamma/csc for pipe bottom color */
+	if (INTEL_GEN(dev_priv) < 9)
+		intel_disable_primary_plane(pipe_config);
 
 	if (INTEL_GEN(dev_priv) >= 11)
 		icl_set_pipe_chicken(intel_crtc);
@@ -6127,7 +6191,10 @@ bool intel_port_is_combophy(struct drm_i915_private *dev_priv, enum port port)
 	if (port == PORT_NONE)
 		return false;
 
-	if (IS_ICELAKE(dev_priv))
+	if (IS_ELKHARTLAKE(dev_priv))
+		return port <= PORT_C;
+
+	if (INTEL_GEN(dev_priv) >= 11)
 		return port <= PORT_B;
 
 	return false;
@@ -6135,7 +6202,7 @@ bool intel_port_is_combophy(struct drm_i915_private *dev_priv, enum port port)
 
 bool intel_port_is_tc(struct drm_i915_private *dev_priv, enum port port)
 {
-	if (IS_ICELAKE(dev_priv))
+	if (INTEL_GEN(dev_priv) >= 11 && !IS_ELKHARTLAKE(dev_priv))
 		return port >= PORT_C && port <= PORT_F;
 
 	return false;
@@ -6304,6 +6371,8 @@ static void valleyview_crtc_enable(struct intel_crtc_state *pipe_config,
 
 	intel_color_load_luts(pipe_config);
 	intel_color_commit(pipe_config);
+	/* update DSPCNTR to configure gamma for pipe bottom color */
+	intel_disable_primary_plane(pipe_config);
 
 	dev_priv->display.initial_watermarks(old_intel_state,
 					     pipe_config);
@@ -6361,6 +6430,8 @@ static void i9xx_crtc_enable(struct intel_crtc_state *pipe_config,
 
 	intel_color_load_luts(pipe_config);
 	intel_color_commit(pipe_config);
+	/* update DSPCNTR to configure gamma for pipe bottom color */
+	intel_disable_primary_plane(pipe_config);
 
 	if (dev_priv->display.initial_watermarks != NULL)
 		dev_priv->display.initial_watermarks(old_intel_state,
@@ -6743,7 +6814,13 @@ static bool hsw_compute_ips_config(struct intel_crtc_state *crtc_state)
 	if (!hsw_crtc_state_ips_capable(crtc_state))
 		return false;
 
-	if (crtc_state->ips_force_disable)
+	/*
+	 * When IPS gets enabled, the pipe CRC changes. Since IPS gets
+	 * enabled and disabled dynamically based on package C states,
+	 * user space can't make reliable use of the CRCs, so let's just
+	 * completely disable it.
+	 */
+	if (crtc_state->crc_enabled)
 		return false;
 
 	/* IPS should be fine as long as at least one plane is enabled. */
@@ -6818,8 +6895,7 @@ static void intel_crtc_compute_pixel_rate(struct intel_crtc_state *crtc_state)
 static int intel_crtc_compute_config(struct intel_crtc *crtc,
 				     struct intel_crtc_state *pipe_config)
 {
-	struct drm_device *dev = crtc->base.dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	const struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
 	int clock_limit = dev_priv->max_dotclk_freq;
 
@@ -6869,7 +6945,7 @@ static int intel_crtc_compute_config(struct intel_crtc *crtc,
 		}
 
 		if (intel_crtc_has_type(pipe_config, INTEL_OUTPUT_LVDS) &&
-		    intel_is_dual_link_lvds(dev)) {
+		    intel_is_dual_link_lvds(dev_priv)) {
 			DRM_DEBUG_KMS("Odd pipe source width not supported with dual link LVDS\n");
 			return -EINVAL;
 		}
@@ -7486,7 +7562,19 @@ static void i8xx_compute_dpll(struct intel_crtc *crtc,
 			dpll |= PLL_P2_DIVIDE_BY_4;
 	}
 
-	if (!IS_I830(dev_priv) &&
+	/*
+	 * Bspec:
+	 * "[Almador Errata}: For the correct operation of the muxed DVO pins
+	 *  (GDEVSELB/I2Cdata, GIRDBY/I2CClk) and (GFRAMEB/DVI_Data,
+	 *  GTRDYB/DVI_Clk): Bit 31 (DPLL VCO Enable) and Bit 30 (2X Clock
+	 *  Enable) must be set to “1” in both the DPLL A Control Register
+	 *  (06014h-06017h) and DPLL B Control Register (06018h-0601Bh)."
+	 *
+	 * For simplicity We simply keep both bits always enabled in
+	 * both DPLLS. The spec says we should disable the DVO 2X clock
+	 * when not needed, but this seems to work fine in practice.
+	 */
+	if (IS_I830(dev_priv) ||
 	    intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DVO))
 		dpll |= DPLL_DVO_2X_MODE;
 
@@ -7694,13 +7782,16 @@ static void i9xx_set_pipeconf(const struct intel_crtc_state *crtc_state)
 			pipeconf |= PIPECONF_INTERLACE_W_FIELD_INDICATION;
 		else
 			pipeconf |= PIPECONF_INTERLACE_W_SYNC_SHIFT;
-	} else
+	} else {
 		pipeconf |= PIPECONF_PROGRESSIVE;
+	}
 
 	if ((IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) &&
 	     crtc_state->limited_color_range)
 		pipeconf |= PIPECONF_COLOR_RANGE_SELECT;
 
+	pipeconf |= PIPECONF_GAMMA_MODE(crtc_state->gamma_mode);
+
 	I915_WRITE(PIPECONF(crtc->pipe), pipeconf);
 	POSTING_READ(PIPECONF(crtc->pipe));
 }
@@ -7744,8 +7835,7 @@ static int i8xx_crtc_compute_clock(struct intel_crtc *crtc,
 static int g4x_crtc_compute_clock(struct intel_crtc *crtc,
 				  struct intel_crtc_state *crtc_state)
 {
-	struct drm_device *dev = crtc->base.dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	const struct intel_limit *limit;
 	int refclk = 96000;
 
@@ -7758,7 +7848,7 @@ static int g4x_crtc_compute_clock(struct intel_crtc *crtc,
 			DRM_DEBUG_KMS("using SSC reference clock of %d kHz\n", refclk);
 		}
 
-		if (intel_is_dual_link_lvds(dev))
+		if (intel_is_dual_link_lvds(dev_priv))
 			limit = &intel_limits_g4x_dual_channel_lvds;
 		else
 			limit = &intel_limits_g4x_single_channel_lvds;
@@ -7894,14 +7984,22 @@ static int vlv_crtc_compute_clock(struct intel_crtc *crtc,
 	return 0;
 }
 
+static bool i9xx_has_pfit(struct drm_i915_private *dev_priv)
+{
+	if (IS_I830(dev_priv))
+		return false;
+
+	return INTEL_GEN(dev_priv) >= 4 ||
+		IS_PINEVIEW(dev_priv) || IS_MOBILE(dev_priv);
+}
+
 static void i9xx_get_pfit_config(struct intel_crtc *crtc,
 				 struct intel_crtc_state *pipe_config)
 {
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	u32 tmp;
 
-	if (INTEL_GEN(dev_priv) <= 3 &&
-	    (IS_I830(dev_priv) || !IS_MOBILE(dev_priv)))
+	if (!i9xx_has_pfit(dev_priv))
 		return;
 
 	tmp = I915_READ(PFIT_CONTROL);
@@ -8108,6 +8206,24 @@ static void intel_get_crtc_ycbcr_config(struct intel_crtc *crtc,
 	pipe_config->output_format = output;
 }
 
+static void i9xx_get_pipe_color_config(struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
+	struct intel_plane *plane = to_intel_plane(crtc->base.primary);
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	enum i9xx_plane_id i9xx_plane = plane->i9xx_plane;
+	u32 tmp;
+
+	tmp = I915_READ(DSPCNTR(i9xx_plane));
+
+	if (tmp & DISPPLANE_GAMMA_ENABLE)
+		crtc_state->gamma_enable = true;
+
+	if (!HAS_GMCH(dev_priv) &&
+	    tmp & DISPPLANE_PIPE_CSC_ENABLE)
+		crtc_state->csc_enable = true;
+}
+
 static bool i9xx_get_pipe_config(struct intel_crtc *crtc,
 				 struct intel_crtc_state *pipe_config)
 {
@@ -8153,6 +8269,14 @@ static bool i9xx_get_pipe_config(struct intel_crtc *crtc,
 	    (tmp & PIPECONF_COLOR_RANGE_SELECT))
 		pipe_config->limited_color_range = true;
 
+	pipe_config->gamma_mode = (tmp & PIPECONF_GAMMA_MODE_MASK_I9XX) >>
+		PIPECONF_GAMMA_MODE_SHIFT;
+
+	if (IS_CHERRYVIEW(dev_priv))
+		pipe_config->cgm_mode = I915_READ(CGM_PIPE_MODE(crtc->pipe));
+
+	i9xx_get_pipe_color_config(pipe_config);
+
 	if (INTEL_GEN(dev_priv) < 4)
 		pipe_config->double_wide = tmp & PIPECONF_DOUBLE_WIDE;
 
@@ -8185,14 +8309,6 @@ static bool i9xx_get_pipe_config(struct intel_crtc *crtc,
 	}
 	pipe_config->dpll_hw_state.dpll = I915_READ(DPLL(crtc->pipe));
 	if (!IS_VALLEYVIEW(dev_priv) && !IS_CHERRYVIEW(dev_priv)) {
-		/*
-		 * DPLL_DVO_2X_MODE must be enabled for both DPLLs
-		 * on 830. Filter it out here so that we don't
-		 * report errors due to that.
-		 */
-		if (IS_I830(dev_priv))
-			pipe_config->dpll_hw_state.dpll &= ~DPLL_DVO_2X_MODE;
-
 		pipe_config->dpll_hw_state.fp0 = I915_READ(FP0(crtc->pipe));
 		pipe_config->dpll_hw_state.fp1 = I915_READ(FP1(crtc->pipe));
 	} else {
@@ -8692,6 +8808,8 @@ static void ironlake_set_pipeconf(const struct intel_crtc_state *crtc_state)
 	if (crtc_state->limited_color_range)
 		val |= PIPECONF_COLOR_RANGE_SELECT;
 
+	val |= PIPECONF_GAMMA_MODE(crtc_state->gamma_mode);
+
 	I915_WRITE(PIPECONF(pipe), val);
 	POSTING_READ(PIPECONF(pipe));
 }
@@ -8772,13 +8890,11 @@ static bool ironlake_needs_fb_cb_tune(struct dpll *dpll, int factor)
 	return i9xx_dpll_compute_m(dpll) < factor * dpll->n;
 }
 
-static void ironlake_compute_dpll(struct intel_crtc *intel_crtc,
+static void ironlake_compute_dpll(struct intel_crtc *crtc,
 				  struct intel_crtc_state *crtc_state,
 				  struct dpll *reduced_clock)
 {
-	struct drm_crtc *crtc = &intel_crtc->base;
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	u32 dpll, fp, fp2;
 	int factor;
 
@@ -8787,10 +8903,12 @@ static void ironlake_compute_dpll(struct intel_crtc *intel_crtc,
 	if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_LVDS)) {
 		if ((intel_panel_use_ssc(dev_priv) &&
 		     dev_priv->vbt.lvds_ssc_freq == 100000) ||
-		    (HAS_PCH_IBX(dev_priv) && intel_is_dual_link_lvds(dev)))
+		    (HAS_PCH_IBX(dev_priv) &&
+		     intel_is_dual_link_lvds(dev_priv)))
 			factor = 25;
-	} else if (crtc_state->sdvo_tv_clock)
+	} else if (crtc_state->sdvo_tv_clock) {
 		factor = 20;
+	}
 
 	fp = i9xx_dpll_compute_fp(&crtc_state->dpll);
 
@@ -8877,8 +8995,7 @@ static void ironlake_compute_dpll(struct intel_crtc *intel_crtc,
 static int ironlake_crtc_compute_clock(struct intel_crtc *crtc,
 				       struct intel_crtc_state *crtc_state)
 {
-	struct drm_device *dev = crtc->base.dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	const struct intel_limit *limit;
 	int refclk = 120000;
 
@@ -8896,7 +9013,7 @@ static int ironlake_crtc_compute_clock(struct intel_crtc *crtc,
 			refclk = dev_priv->vbt.lvds_ssc_freq;
 		}
 
-		if (intel_is_dual_link_lvds(dev)) {
+		if (intel_is_dual_link_lvds(dev_priv)) {
 			if (refclk == 100000)
 				limit = &intel_limits_ironlake_dual_lvds_100m;
 			else
@@ -8920,7 +9037,7 @@ static int ironlake_crtc_compute_clock(struct intel_crtc *crtc,
 
 	ironlake_compute_dpll(crtc, crtc_state, NULL);
 
-	if (!intel_get_shared_dpll(crtc, crtc_state, NULL)) {
+	if (!intel_get_shared_dpll(crtc_state, NULL)) {
 		DRM_DEBUG_KMS("failed to find PLL for pipe %c\n",
 			      pipe_name(crtc->pipe));
 		return -EINVAL;
@@ -9226,6 +9343,13 @@ static bool ironlake_get_pipe_config(struct intel_crtc *crtc,
 	if (tmp & PIPECONF_COLOR_RANGE_SELECT)
 		pipe_config->limited_color_range = true;
 
+	pipe_config->gamma_mode = (tmp & PIPECONF_GAMMA_MODE_MASK_ILK) >>
+		PIPECONF_GAMMA_MODE_SHIFT;
+
+	pipe_config->csc_mode = I915_READ(PIPE_CSC_MODE(crtc->pipe));
+
+	i9xx_get_pipe_color_config(pipe_config);
+
 	if (I915_READ(PCH_TRANSCONF(crtc->pipe)) & TRANS_ENABLE) {
 		struct intel_shared_dpll *pll;
 		enum intel_dpll_id pll_id;
@@ -9371,7 +9495,8 @@ static void hsw_disable_lcpll(struct drm_i915_private *dev_priv,
 	I915_WRITE(LCPLL_CTL, val);
 	POSTING_READ(LCPLL_CTL);
 
-	if (intel_wait_for_register(dev_priv, LCPLL_CTL, LCPLL_PLL_LOCK, 0, 1))
+	if (intel_wait_for_register(&dev_priv->uncore,
+				    LCPLL_CTL, LCPLL_PLL_LOCK, 0, 1))
 		DRM_ERROR("LCPLL still locked\n");
 
 	val = hsw_read_dcomp(dev_priv);
@@ -9409,7 +9534,7 @@ static void hsw_restore_lcpll(struct drm_i915_private *dev_priv)
 	 * Make sure we're not on PC8 state before disabling PC8, otherwise
 	 * we'll hang the machine. To prevent PC8 state, just enable force_wake.
 	 */
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL);
 
 	if (val & LCPLL_POWER_DOWN_ALLOW) {
 		val &= ~LCPLL_POWER_DOWN_ALLOW;
@@ -9426,7 +9551,7 @@ static void hsw_restore_lcpll(struct drm_i915_private *dev_priv)
 	val &= ~LCPLL_PLL_DISABLE;
 	I915_WRITE(LCPLL_CTL, val);
 
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    LCPLL_CTL, LCPLL_PLL_LOCK, LCPLL_PLL_LOCK,
 				    5))
 		DRM_ERROR("LCPLL not locked yet\n");
@@ -9441,7 +9566,7 @@ static void hsw_restore_lcpll(struct drm_i915_private *dev_priv)
 			DRM_ERROR("Switching back to LCPLL failed\n");
 	}
 
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 
 	intel_update_cdclk(dev_priv);
 	intel_dump_cdclk_state(&dev_priv->cdclk.hw, "Current CDCLK");
@@ -9510,11 +9635,11 @@ static int haswell_crtc_compute_clock(struct intel_crtc *crtc,
 		to_intel_atomic_state(crtc_state->base.state);
 
 	if (!intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DSI) ||
-	    IS_ICELAKE(dev_priv)) {
+	    INTEL_GEN(dev_priv) >= 11) {
 		struct intel_encoder *encoder =
 			intel_get_crtc_new_encoder(state, crtc_state);
 
-		if (!intel_get_shared_dpll(crtc, crtc_state, encoder)) {
+		if (!intel_get_shared_dpll(crtc_state, encoder)) {
 			DRM_DEBUG_KMS("failed to find PLL for pipe %c\n",
 				      pipe_name(crtc->pipe));
 			return -EINVAL;
@@ -9552,9 +9677,6 @@ static void icelake_get_ddi_pll(struct drm_i915_private *dev_priv,
 		temp = I915_READ(DPCLKA_CFGCR0_ICL) &
 		       DPCLKA_CFGCR0_DDI_CLK_SEL_MASK(port);
 		id = temp >> DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(port);
-
-		if (WARN_ON(!intel_dpll_is_combophy(id)))
-			return;
 	} else if (intel_port_is_tc(dev_priv, port)) {
 		id = icl_tc_port_to_pll_id(intel_port_to_tc(dev_priv, port));
 	} else {
@@ -9643,20 +9765,25 @@ static void haswell_get_ddi_pll(struct drm_i915_private *dev_priv,
 
 static bool hsw_get_transcoder_state(struct intel_crtc *crtc,
 				     struct intel_crtc_state *pipe_config,
-				     u64 *power_domain_mask)
+				     u64 *power_domain_mask,
+				     intel_wakeref_t *wakerefs)
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	enum intel_display_power_domain power_domain;
-	unsigned long panel_transcoder_mask = BIT(TRANSCODER_EDP);
+	unsigned long panel_transcoder_mask = 0;
 	unsigned long enabled_panel_transcoders = 0;
 	enum transcoder panel_transcoder;
+	intel_wakeref_t wf;
 	u32 tmp;
 
-	if (IS_ICELAKE(dev_priv))
+	if (INTEL_GEN(dev_priv) >= 11)
 		panel_transcoder_mask |=
 			BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1);
 
+	if (HAS_TRANSCODER_EDP(dev_priv))
+		panel_transcoder_mask |= BIT(TRANSCODER_EDP);
+
 	/*
 	 * The pipe->transcoder mapping is fixed with the exception of the eDP
 	 * and DSI transcoders handled below.
@@ -9713,10 +9840,13 @@ static bool hsw_get_transcoder_state(struct intel_crtc *crtc,
 		enabled_panel_transcoders != BIT(TRANSCODER_EDP));
 
 	power_domain = POWER_DOMAIN_TRANSCODER(pipe_config->cpu_transcoder);
-	if (!intel_display_power_get_if_enabled(dev_priv, power_domain))
+	WARN_ON(*power_domain_mask & BIT_ULL(power_domain));
+
+	wf = intel_display_power_get_if_enabled(dev_priv, power_domain);
+	if (!wf)
 		return false;
 
-	WARN_ON(*power_domain_mask & BIT_ULL(power_domain));
+	wakerefs[power_domain] = wf;
 	*power_domain_mask |= BIT_ULL(power_domain);
 
 	tmp = I915_READ(PIPECONF(pipe_config->cpu_transcoder));
@@ -9726,13 +9856,15 @@ static bool hsw_get_transcoder_state(struct intel_crtc *crtc,
 
 static bool bxt_get_dsi_transcoder_state(struct intel_crtc *crtc,
 					 struct intel_crtc_state *pipe_config,
-					 u64 *power_domain_mask)
+					 u64 *power_domain_mask,
+					 intel_wakeref_t *wakerefs)
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	enum intel_display_power_domain power_domain;
-	enum port port;
 	enum transcoder cpu_transcoder;
+	intel_wakeref_t wf;
+	enum port port;
 	u32 tmp;
 
 	for_each_port_masked(port, BIT(PORT_A) | BIT(PORT_C)) {
@@ -9742,10 +9874,13 @@ static bool bxt_get_dsi_transcoder_state(struct intel_crtc *crtc,
 			cpu_transcoder = TRANSCODER_DSI_C;
 
 		power_domain = POWER_DOMAIN_TRANSCODER(cpu_transcoder);
-		if (!intel_display_power_get_if_enabled(dev_priv, power_domain))
+		WARN_ON(*power_domain_mask & BIT_ULL(power_domain));
+
+		wf = intel_display_power_get_if_enabled(dev_priv, power_domain);
+		if (!wf)
 			continue;
 
-		WARN_ON(*power_domain_mask & BIT_ULL(power_domain));
+		wakerefs[power_domain] = wf;
 		*power_domain_mask |= BIT_ULL(power_domain);
 
 		/*
@@ -9786,7 +9921,7 @@ static void haswell_get_ddi_port_state(struct intel_crtc *crtc,
 
 	port = (tmp & TRANS_DDI_PORT_MASK) >> TRANS_DDI_PORT_SHIFT;
 
-	if (IS_ICELAKE(dev_priv))
+	if (INTEL_GEN(dev_priv) >= 11)
 		icelake_get_ddi_pll(dev_priv, port, pipe_config);
 	else if (IS_CANNONLAKE(dev_priv))
 		cannonlake_get_ddi_pll(dev_priv, port, pipe_config);
@@ -9824,6 +9959,7 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc,
 				    struct intel_crtc_state *pipe_config)
 {
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	intel_wakeref_t wakerefs[POWER_DOMAIN_NUM], wf;
 	enum intel_display_power_domain power_domain;
 	u64 power_domain_mask;
 	bool active;
@@ -9831,16 +9967,21 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc,
 	intel_crtc_init_scalers(crtc, pipe_config);
 
 	power_domain = POWER_DOMAIN_PIPE(crtc->pipe);
-	if (!intel_display_power_get_if_enabled(dev_priv, power_domain))
+	wf = intel_display_power_get_if_enabled(dev_priv, power_domain);
+	if (!wf)
 		return false;
+
+	wakerefs[power_domain] = wf;
 	power_domain_mask = BIT_ULL(power_domain);
 
 	pipe_config->shared_dpll = NULL;
 
-	active = hsw_get_transcoder_state(crtc, pipe_config, &power_domain_mask);
+	active = hsw_get_transcoder_state(crtc, pipe_config,
+					  &power_domain_mask, wakerefs);
 
 	if (IS_GEN9_LP(dev_priv) &&
-	    bxt_get_dsi_transcoder_state(crtc, pipe_config, &power_domain_mask)) {
+	    bxt_get_dsi_transcoder_state(crtc, pipe_config,
+					 &power_domain_mask, wakerefs)) {
 		WARN_ON(active);
 		active = true;
 	}
@@ -9849,7 +9990,7 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc,
 		goto out;
 
 	if (!transcoder_is_dsi(pipe_config->cpu_transcoder) ||
-	    IS_ICELAKE(dev_priv)) {
+	    INTEL_GEN(dev_priv) >= 11) {
 		haswell_get_ddi_port_state(crtc, pipe_config);
 		intel_get_pipe_timings(crtc, pipe_config);
 	}
@@ -9857,12 +9998,28 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc,
 	intel_get_pipe_src_size(crtc, pipe_config);
 	intel_get_crtc_ycbcr_config(crtc, pipe_config);
 
-	pipe_config->gamma_mode =
-		I915_READ(GAMMA_MODE(crtc->pipe)) & GAMMA_MODE_MODE_MASK;
+	pipe_config->gamma_mode = I915_READ(GAMMA_MODE(crtc->pipe));
+
+	pipe_config->csc_mode = I915_READ(PIPE_CSC_MODE(crtc->pipe));
+
+	if (INTEL_GEN(dev_priv) >= 9) {
+		u32 tmp = I915_READ(SKL_BOTTOM_COLOR(crtc->pipe));
+
+		if (tmp & SKL_BOTTOM_COLOR_GAMMA_ENABLE)
+			pipe_config->gamma_enable = true;
+
+		if (tmp & SKL_BOTTOM_COLOR_CSC_ENABLE)
+			pipe_config->csc_enable = true;
+	} else {
+		i9xx_get_pipe_color_config(pipe_config);
+	}
 
 	power_domain = POWER_DOMAIN_PIPE_PANEL_FITTER(crtc->pipe);
-	if (intel_display_power_get_if_enabled(dev_priv, power_domain)) {
-		WARN_ON(power_domain_mask & BIT_ULL(power_domain));
+	WARN_ON(power_domain_mask & BIT_ULL(power_domain));
+
+	wf = intel_display_power_get_if_enabled(dev_priv, power_domain);
+	if (wf) {
+		wakerefs[power_domain] = wf;
 		power_domain_mask |= BIT_ULL(power_domain);
 
 		if (INTEL_GEN(dev_priv) >= 9)
@@ -9894,7 +10051,8 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc,
 
 out:
 	for_each_power_domain(power_domain, power_domain_mask)
-		intel_display_power_put_unchecked(dev_priv, power_domain);
+		intel_display_power_put(dev_priv,
+					power_domain, wakerefs[power_domain]);
 
 	return active;
 }
@@ -10030,7 +10188,12 @@ i845_cursor_max_stride(struct intel_plane *plane,
 
 static u32 i845_cursor_ctl_crtc(const struct intel_crtc_state *crtc_state)
 {
-	return CURSOR_GAMMA_ENABLE;
+	u32 cntl = 0;
+
+	if (crtc_state->gamma_enable)
+		cntl |= CURSOR_GAMMA_ENABLE;
+
+	return cntl;
 }
 
 static u32 i845_cursor_ctl(const struct intel_crtc_state *crtc_state,
@@ -10184,9 +10347,10 @@ static u32 i9xx_cursor_ctl_crtc(const struct intel_crtc_state *crtc_state)
 	if (INTEL_GEN(dev_priv) >= 11)
 		return cntl;
 
-	cntl |= MCURSOR_GAMMA_ENABLE;
+	if (crtc_state->gamma_enable)
+		cntl = MCURSOR_GAMMA_ENABLE;
 
-	if (HAS_DDI(dev_priv))
+	if (crtc_state->csc_enable)
 		cntl |= MCURSOR_PIPE_CSC_ENABLE;
 
 	if (INTEL_GEN(dev_priv) < 5 && !IS_G4X(dev_priv))
@@ -11134,7 +11298,7 @@ static int icl_check_nv12_planes(struct intel_crtc_state *crtc_state)
 		}
 
 		if (!linked_state) {
-			DRM_DEBUG_KMS("Need %d free Y planes for NV12\n",
+			DRM_DEBUG_KMS("Need %d free Y planes for planar YUV\n",
 				      hweight8(crtc_state->nv12_planes));
 
 			return -EINVAL;
@@ -11175,16 +11339,11 @@ static int intel_crtc_atomic_check(struct drm_crtc *crtc,
 			return ret;
 	}
 
-	if (mode_changed || crtc_state->color_mgmt_changed) {
+	if (mode_changed || pipe_config->update_pipe ||
+	    crtc_state->color_mgmt_changed) {
 		ret = intel_color_check(pipe_config);
 		if (ret)
 			return ret;
-
-		/*
-		 * Changing color management on Intel hardware is
-		 * handled as part of planes update.
-		 */
-		crtc_state->planes_changed = true;
 	}
 
 	ret = 0;
@@ -11355,6 +11514,16 @@ intel_dump_m_n_config(struct intel_crtc_state *pipe_config, char *id,
 		      m_n->link_m, m_n->link_n, m_n->tu);
 }
 
+static void
+intel_dump_infoframe(struct drm_i915_private *dev_priv,
+		     const union hdmi_infoframe *frame)
+{
+	if ((drm_debug & DRM_UT_KMS) == 0)
+		return;
+
+	hdmi_infoframe_log(KERN_DEBUG, dev_priv->drm.dev, frame);
+}
+
 #define OUTPUT_TYPE(x) [INTEL_OUTPUT_ ## x] = #x
 
 static const char * const output_type_str[] = {
@@ -11458,6 +11627,22 @@ static void intel_dump_pipe_config(struct intel_crtc *crtc,
 	DRM_DEBUG_KMS("audio: %i, infoframes: %i\n",
 		      pipe_config->has_audio, pipe_config->has_infoframe);
 
+	DRM_DEBUG_KMS("infoframes enabled: 0x%x\n",
+		      pipe_config->infoframes.enable);
+
+	if (pipe_config->infoframes.enable &
+	    intel_hdmi_infoframe_enable(HDMI_PACKET_TYPE_GENERAL_CONTROL))
+		DRM_DEBUG_KMS("GCP: 0x%x\n", pipe_config->infoframes.gcp);
+	if (pipe_config->infoframes.enable &
+	    intel_hdmi_infoframe_enable(HDMI_INFOFRAME_TYPE_AVI))
+		intel_dump_infoframe(dev_priv, &pipe_config->infoframes.avi);
+	if (pipe_config->infoframes.enable &
+	    intel_hdmi_infoframe_enable(HDMI_INFOFRAME_TYPE_SPD))
+		intel_dump_infoframe(dev_priv, &pipe_config->infoframes.spd);
+	if (pipe_config->infoframes.enable &
+	    intel_hdmi_infoframe_enable(HDMI_INFOFRAME_TYPE_VENDOR))
+		intel_dump_infoframe(dev_priv, &pipe_config->infoframes.hdmi);
+
 	DRM_DEBUG_KMS("requested mode:\n");
 	drm_mode_debug_printmodeline(&pipe_config->base.mode);
 	DRM_DEBUG_KMS("adjusted mode:\n");
@@ -11606,7 +11791,7 @@ clear_intel_crtc_state(struct intel_crtc_state *crtc_state)
 	saved_state->shared_dpll = crtc_state->shared_dpll;
 	saved_state->dpll_hw_state = crtc_state->dpll_hw_state;
 	saved_state->pch_pfit.force_thru = crtc_state->pch_pfit.force_thru;
-	saved_state->ips_force_disable = crtc_state->ips_force_disable;
+	saved_state->crc_enabled = crtc_state->crc_enabled;
 	if (IS_G4X(dev_priv) ||
 	    IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
 		saved_state->wm = crtc_state->wm;
@@ -11825,6 +12010,37 @@ intel_compare_link_m_n(const struct intel_link_m_n *m_n,
 	return false;
 }
 
+static bool
+intel_compare_infoframe(const union hdmi_infoframe *a,
+			const union hdmi_infoframe *b)
+{
+	return memcmp(a, b, sizeof(*a)) == 0;
+}
+
+static void
+pipe_config_infoframe_err(struct drm_i915_private *dev_priv,
+			  bool adjust, const char *name,
+			  const union hdmi_infoframe *a,
+			  const union hdmi_infoframe *b)
+{
+	if (adjust) {
+		if ((drm_debug & DRM_UT_KMS) == 0)
+			return;
+
+		drm_dbg(DRM_UT_KMS, "mismatch in %s infoframe", name);
+		drm_dbg(DRM_UT_KMS, "expected:");
+		hdmi_infoframe_log(KERN_DEBUG, dev_priv->drm.dev, a);
+		drm_dbg(DRM_UT_KMS, "found");
+		hdmi_infoframe_log(KERN_DEBUG, dev_priv->drm.dev, b);
+	} else {
+		drm_err("mismatch in %s infoframe", name);
+		drm_err("expected:");
+		hdmi_infoframe_log(KERN_ERR, dev_priv->drm.dev, a);
+		drm_err("found");
+		hdmi_infoframe_log(KERN_ERR, dev_priv->drm.dev, b);
+	}
+}
+
 static void __printf(3, 4)
 pipe_config_err(bool adjust, const char *name, const char *format, ...)
 {
@@ -12008,7 +12224,17 @@ intel_pipe_config_compare(struct drm_i915_private *dev_priv,
 	} \
 } while (0)
 
-#define PIPE_CONF_QUIRK(quirk)	\
+#define PIPE_CONF_CHECK_INFOFRAME(name) do { \
+	if (!intel_compare_infoframe(&current_config->infoframes.name, \
+				     &pipe_config->infoframes.name)) { \
+		pipe_config_infoframe_err(dev_priv, adjust, __stringify(name), \
+					  &current_config->infoframes.name, \
+					  &pipe_config->infoframes.name); \
+		ret = false; \
+	} \
+} while (0)
+
+#define PIPE_CONF_QUIRK(quirk) \
 	((current_config->quirks | pipe_config->quirks) & (quirk))
 
 	PIPE_CONF_CHECK_I(cpu_transcoder);
@@ -12089,6 +12315,14 @@ intel_pipe_config_compare(struct drm_i915_private *dev_priv,
 
 		PIPE_CONF_CHECK_I(scaler_state.scaler_id);
 		PIPE_CONF_CHECK_CLOCK_FUZZY(pixel_rate);
+
+		PIPE_CONF_CHECK_X(gamma_mode);
+		if (IS_CHERRYVIEW(dev_priv))
+			PIPE_CONF_CHECK_X(cgm_mode);
+		else
+			PIPE_CONF_CHECK_X(csc_mode);
+		PIPE_CONF_CHECK_BOOL(gamma_enable);
+		PIPE_CONF_CHECK_BOOL(csc_enable);
 	}
 
 	PIPE_CONF_CHECK_BOOL(double_wide);
@@ -12137,6 +12371,12 @@ intel_pipe_config_compare(struct drm_i915_private *dev_priv,
 
 	PIPE_CONF_CHECK_I(min_voltage_level);
 
+	PIPE_CONF_CHECK_X(infoframes.enable);
+	PIPE_CONF_CHECK_X(infoframes.gcp);
+	PIPE_CONF_CHECK_INFOFRAME(avi);
+	PIPE_CONF_CHECK_INFOFRAME(spd);
+	PIPE_CONF_CHECK_INFOFRAME(hdmi);
+
 #undef PIPE_CONF_CHECK_X
 #undef PIPE_CONF_CHECK_I
 #undef PIPE_CONF_CHECK_BOOL
@@ -12171,12 +12411,15 @@ static void verify_wm_state(struct drm_crtc *crtc,
 			    struct drm_crtc_state *new_state)
 {
 	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
-	struct skl_ddb_allocation hw_ddb, *sw_ddb;
-	struct skl_pipe_wm hw_wm, *sw_wm;
-	struct skl_plane_wm *hw_plane_wm, *sw_plane_wm;
+	struct skl_hw_state {
+		struct skl_ddb_entry ddb_y[I915_MAX_PLANES];
+		struct skl_ddb_entry ddb_uv[I915_MAX_PLANES];
+		struct skl_ddb_allocation ddb;
+		struct skl_pipe_wm wm;
+	} *hw;
+	struct skl_ddb_allocation *sw_ddb;
+	struct skl_pipe_wm *sw_wm;
 	struct skl_ddb_entry *hw_ddb_entry, *sw_ddb_entry;
-	struct skl_ddb_entry hw_ddb_y[I915_MAX_PLANES];
-	struct skl_ddb_entry hw_ddb_uv[I915_MAX_PLANES];
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	const enum pipe pipe = intel_crtc->pipe;
 	int plane, level, max_level = ilk_wm_max_level(dev_priv);
@@ -12184,22 +12427,29 @@ static void verify_wm_state(struct drm_crtc *crtc,
 	if (INTEL_GEN(dev_priv) < 9 || !new_state->active)
 		return;
 
-	skl_pipe_wm_get_hw_state(intel_crtc, &hw_wm);
+	hw = kzalloc(sizeof(*hw), GFP_KERNEL);
+	if (!hw)
+		return;
+
+	skl_pipe_wm_get_hw_state(intel_crtc, &hw->wm);
 	sw_wm = &to_intel_crtc_state(new_state)->wm.skl.optimal;
 
-	skl_pipe_ddb_get_hw_state(intel_crtc, hw_ddb_y, hw_ddb_uv);
+	skl_pipe_ddb_get_hw_state(intel_crtc, hw->ddb_y, hw->ddb_uv);
 
-	skl_ddb_get_hw_state(dev_priv, &hw_ddb);
+	skl_ddb_get_hw_state(dev_priv, &hw->ddb);
 	sw_ddb = &dev_priv->wm.skl_hw.ddb;
 
-	if (INTEL_GEN(dev_priv) >= 11)
-		if (hw_ddb.enabled_slices != sw_ddb->enabled_slices)
-			DRM_ERROR("mismatch in DBUF Slices (expected %u, got %u)\n",
-				  sw_ddb->enabled_slices,
-				  hw_ddb.enabled_slices);
+	if (INTEL_GEN(dev_priv) >= 11 &&
+	    hw->ddb.enabled_slices != sw_ddb->enabled_slices)
+		DRM_ERROR("mismatch in DBUF Slices (expected %u, got %u)\n",
+			  sw_ddb->enabled_slices,
+			  hw->ddb.enabled_slices);
+
 	/* planes */
 	for_each_universal_plane(dev_priv, pipe, plane) {
-		hw_plane_wm = &hw_wm.planes[plane];
+		struct skl_plane_wm *hw_plane_wm, *sw_plane_wm;
+
+		hw_plane_wm = &hw->wm.planes[plane];
 		sw_plane_wm = &sw_wm->planes[plane];
 
 		/* Watermarks */
@@ -12231,7 +12481,7 @@ static void verify_wm_state(struct drm_crtc *crtc,
 		}
 
 		/* DDB */
-		hw_ddb_entry = &hw_ddb_y[plane];
+		hw_ddb_entry = &hw->ddb_y[plane];
 		sw_ddb_entry = &to_intel_crtc_state(new_state)->wm.skl.plane_ddb_y[plane];
 
 		if (!skl_ddb_entry_equal(hw_ddb_entry, sw_ddb_entry)) {
@@ -12249,7 +12499,9 @@ static void verify_wm_state(struct drm_crtc *crtc,
 	 * once the plane becomes visible, we can skip this check
 	 */
 	if (1) {
-		hw_plane_wm = &hw_wm.planes[PLANE_CURSOR];
+		struct skl_plane_wm *hw_plane_wm, *sw_plane_wm;
+
+		hw_plane_wm = &hw->wm.planes[PLANE_CURSOR];
 		sw_plane_wm = &sw_wm->planes[PLANE_CURSOR];
 
 		/* Watermarks */
@@ -12281,7 +12533,7 @@ static void verify_wm_state(struct drm_crtc *crtc,
 		}
 
 		/* DDB */
-		hw_ddb_entry = &hw_ddb_y[PLANE_CURSOR];
+		hw_ddb_entry = &hw->ddb_y[PLANE_CURSOR];
 		sw_ddb_entry = &to_intel_crtc_state(new_state)->wm.skl.plane_ddb_y[PLANE_CURSOR];
 
 		if (!skl_ddb_entry_equal(hw_ddb_entry, sw_ddb_entry)) {
@@ -12291,6 +12543,8 @@ static void verify_wm_state(struct drm_crtc *crtc,
 				  hw_ddb_entry->start, hw_ddb_entry->end);
 		}
 	}
+
+	kfree(hw);
 }
 
 static void
@@ -12447,7 +12701,8 @@ intel_verify_planes(struct intel_atomic_state *state)
 
 	for_each_new_intel_plane_in_state(state, plane,
 					  plane_state, i)
-		assert_plane(plane, plane_state->base.visible);
+		assert_plane(plane, plane_state->slave ||
+			     plane_state->base.visible);
 }
 
 static void
@@ -12769,10 +13024,16 @@ static int intel_modeset_checks(struct drm_atomic_state *state)
 		return -EINVAL;
 	}
 
+	/* keep the current setting */
+	if (!intel_state->cdclk.force_min_cdclk_changed)
+		intel_state->cdclk.force_min_cdclk =
+			dev_priv->cdclk.force_min_cdclk;
+
 	intel_state->modeset = true;
 	intel_state->active_crtcs = dev_priv->active_crtcs;
 	intel_state->cdclk.logical = dev_priv->cdclk.logical;
 	intel_state->cdclk.actual = dev_priv->cdclk.actual;
+	intel_state->cdclk.pipe = INVALID_PIPE;
 
 	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
 		if (new_crtc_state->active)
@@ -12792,6 +13053,8 @@ static int intel_modeset_checks(struct drm_atomic_state *state)
 	 * adjusted_mode bits in the crtc directly.
 	 */
 	if (dev_priv->display.modeset_calc_cdclk) {
+		enum pipe pipe;
+
 		ret = dev_priv->display.modeset_calc_cdclk(state);
 		if (ret < 0)
 			return ret;
@@ -12808,12 +13071,36 @@ static int intel_modeset_checks(struct drm_atomic_state *state)
 				return ret;
 		}
 
+		if (is_power_of_2(intel_state->active_crtcs)) {
+			struct drm_crtc *crtc;
+			struct drm_crtc_state *crtc_state;
+
+			pipe = ilog2(intel_state->active_crtcs);
+			crtc = &intel_get_crtc_for_pipe(dev_priv, pipe)->base;
+			crtc_state = drm_atomic_get_new_crtc_state(state, crtc);
+			if (crtc_state && needs_modeset(crtc_state))
+				pipe = INVALID_PIPE;
+		} else {
+			pipe = INVALID_PIPE;
+		}
+
 		/* All pipes must be switched off while we change the cdclk. */
-		if (intel_cdclk_needs_modeset(&dev_priv->cdclk.actual,
-					      &intel_state->cdclk.actual)) {
+		if (pipe != INVALID_PIPE &&
+		    intel_cdclk_needs_cd2x_update(dev_priv,
+						  &dev_priv->cdclk.actual,
+						  &intel_state->cdclk.actual)) {
+			ret = intel_lock_all_pipes(state);
+			if (ret < 0)
+				return ret;
+
+			intel_state->cdclk.pipe = pipe;
+		} else if (intel_cdclk_needs_modeset(&dev_priv->cdclk.actual,
+						     &intel_state->cdclk.actual)) {
 			ret = intel_modeset_all_pipes(state);
 			if (ret < 0)
 				return ret;
+
+			intel_state->cdclk.pipe = INVALID_PIPE;
 		}
 
 		DRM_DEBUG_KMS("New cdclk calculated to be logical %u kHz, actual %u kHz\n",
@@ -12822,8 +13109,6 @@ static int intel_modeset_checks(struct drm_atomic_state *state)
 		DRM_DEBUG_KMS("New voltage level calculated to be logical %u, actual %u\n",
 			      intel_state->cdclk.logical.voltage_level,
 			      intel_state->cdclk.actual.voltage_level);
-	} else {
-		to_intel_atomic_state(state)->cdclk.logical = dev_priv->cdclk.logical;
 	}
 
 	intel_modeset_clear_plls(state);
@@ -12864,7 +13149,7 @@ static int intel_atomic_check(struct drm_device *dev,
 	struct drm_crtc *crtc;
 	struct drm_crtc_state *old_crtc_state, *crtc_state;
 	int ret, i;
-	bool any_ms = false;
+	bool any_ms = intel_state->cdclk.force_min_cdclk_changed;
 
 	/* Catch I915_MODE_FLAG_INHERITED */
 	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state,
@@ -12989,14 +13274,14 @@ static void intel_update_crtc(struct drm_crtc *crtc,
 	else if (new_plane_state)
 		intel_fbc_enable(intel_crtc, pipe_config, new_plane_state);
 
-	intel_begin_crtc_commit(crtc, old_crtc_state);
+	intel_begin_crtc_commit(to_intel_atomic_state(state), intel_crtc);
 
 	if (INTEL_GEN(dev_priv) >= 9)
 		skl_update_planes_on_crtc(to_intel_atomic_state(state), intel_crtc);
 	else
 		i9xx_update_planes_on_crtc(to_intel_atomic_state(state), intel_crtc);
 
-	intel_finish_crtc_commit(crtc, old_crtc_state);
+	intel_finish_crtc_commit(to_intel_atomic_state(state), intel_crtc);
 }
 
 static void intel_update_crtcs(struct drm_atomic_state *state)
@@ -13224,7 +13509,10 @@ static void intel_atomic_commit_tail(struct drm_atomic_state *state)
 	if (intel_state->modeset) {
 		drm_atomic_helper_update_legacy_modeset_state(state->dev, state);
 
-		intel_set_cdclk(dev_priv, &dev_priv->cdclk.actual);
+		intel_set_cdclk_pre_plane_update(dev_priv,
+						 &intel_state->cdclk.actual,
+						 &dev_priv->cdclk.actual,
+						 intel_state->cdclk.pipe);
 
 		/*
 		 * SKL workaround: bspec recommends we disable the SAGV when we
@@ -13253,6 +13541,12 @@ static void intel_atomic_commit_tail(struct drm_atomic_state *state)
 	/* Now enable the clocks, plane, pipe, and connectors that we set up. */
 	dev_priv->display.update_crtcs(state);
 
+	if (intel_state->modeset)
+		intel_set_cdclk_post_plane_update(dev_priv,
+						  &intel_state->cdclk.actual,
+						  &dev_priv->cdclk.actual,
+						  intel_state->cdclk.pipe);
+
 	/* FIXME: We should call drm_atomic_helper_commit_hw_done() here
 	 * already, but still need the state for the delayed optimization. To
 	 * fix this:
@@ -13313,7 +13607,7 @@ static void intel_atomic_commit_tail(struct drm_atomic_state *state)
 		 * so enable debugging for the next modeset - and hope we catch
 		 * the culprit.
 		 */
-		intel_uncore_arm_unclaimed_mmio_detection(dev_priv);
+		intel_uncore_arm_unclaimed_mmio_detection(&dev_priv->uncore);
 		intel_display_power_put(dev_priv, POWER_DOMAIN_MODESET, wakeref);
 	}
 
@@ -13454,8 +13748,10 @@ static int intel_atomic_commit(struct drm_device *dev,
 		       intel_state->min_voltage_level,
 		       sizeof(intel_state->min_voltage_level));
 		dev_priv->active_crtcs = intel_state->active_crtcs;
-		dev_priv->cdclk.logical = intel_state->cdclk.logical;
-		dev_priv->cdclk.actual = intel_state->cdclk.actual;
+		dev_priv->cdclk.force_min_cdclk =
+			intel_state->cdclk.force_min_cdclk;
+
+		intel_cdclk_swap_state(intel_state);
 	}
 
 	drm_atomic_state_get(state);
@@ -13506,7 +13802,7 @@ static int do_rps_boost(struct wait_queue_entry *_wait,
 	 * vblank without our intervention, so leave RPS alone.
 	 */
 	if (!i915_request_started(rq))
-		gen6_rps_boost(rq, NULL);
+		gen6_rps_boost(rq);
 	i915_request_put(rq);
 
 	drm_crtc_vblank_put(wait->crtc);
@@ -13767,7 +14063,7 @@ skl_max_scale(const struct intel_crtc_state *crtc_state,
 	 *            or
 	 *    cdclk/crtc_clock
 	 */
-	mult = pixel_format == DRM_FORMAT_NV12 ? 2 : 3;
+	mult = is_planar_yuv_format(pixel_format) ? 2 : 3;
 	tmpclk1 = (1 << 16) * mult - 1;
 	tmpclk2 = (1 << 8) * ((max_dotclk << 8) / crtc_clock);
 	max_scale = min(tmpclk1, tmpclk2);
@@ -13775,39 +14071,35 @@ skl_max_scale(const struct intel_crtc_state *crtc_state,
 	return max_scale;
 }
 
-static void intel_begin_crtc_commit(struct drm_crtc *crtc,
-				    struct drm_crtc_state *old_crtc_state)
+static void intel_begin_crtc_commit(struct intel_atomic_state *state,
+				    struct intel_crtc *crtc)
 {
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct intel_crtc_state *old_intel_cstate =
-		to_intel_crtc_state(old_crtc_state);
-	struct intel_atomic_state *old_intel_state =
-		to_intel_atomic_state(old_crtc_state->state);
-	struct intel_crtc_state *intel_cstate =
-		intel_atomic_get_new_crtc_state(old_intel_state, intel_crtc);
-	bool modeset = needs_modeset(&intel_cstate->base);
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_crtc_state *old_crtc_state =
+		intel_atomic_get_old_crtc_state(state, crtc);
+	struct intel_crtc_state *new_crtc_state =
+		intel_atomic_get_new_crtc_state(state, crtc);
+	bool modeset = needs_modeset(&new_crtc_state->base);
 
 	/* Perform vblank evasion around commit operation */
-	intel_pipe_update_start(intel_cstate);
+	intel_pipe_update_start(new_crtc_state);
 
 	if (modeset)
 		goto out;
 
-	if (intel_cstate->base.color_mgmt_changed ||
-	    intel_cstate->update_pipe)
-		intel_color_commit(intel_cstate);
+	if (new_crtc_state->base.color_mgmt_changed ||
+	    new_crtc_state->update_pipe)
+		intel_color_commit(new_crtc_state);
 
-	if (intel_cstate->update_pipe)
-		intel_update_pipe_config(old_intel_cstate, intel_cstate);
+	if (new_crtc_state->update_pipe)
+		intel_update_pipe_config(old_crtc_state, new_crtc_state);
 	else if (INTEL_GEN(dev_priv) >= 9)
-		skl_detach_scalers(intel_cstate);
+		skl_detach_scalers(new_crtc_state);
 
 out:
 	if (dev_priv->display.atomic_update_watermarks)
-		dev_priv->display.atomic_update_watermarks(old_intel_state,
-							   intel_cstate);
+		dev_priv->display.atomic_update_watermarks(state,
+							   new_crtc_state);
 }
 
 void intel_crtc_arm_fifo_underrun(struct intel_crtc *crtc,
@@ -13826,21 +14118,20 @@ void intel_crtc_arm_fifo_underrun(struct intel_crtc *crtc,
 	}
 }
 
-static void intel_finish_crtc_commit(struct drm_crtc *crtc,
-				     struct drm_crtc_state *old_crtc_state)
+static void intel_finish_crtc_commit(struct intel_atomic_state *state,
+				     struct intel_crtc *crtc)
 {
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct intel_atomic_state *old_intel_state =
-		to_intel_atomic_state(old_crtc_state->state);
+	struct intel_crtc_state *old_crtc_state =
+		intel_atomic_get_old_crtc_state(state, crtc);
 	struct intel_crtc_state *new_crtc_state =
-		intel_atomic_get_new_crtc_state(old_intel_state, intel_crtc);
+		intel_atomic_get_new_crtc_state(state, crtc);
 
 	intel_pipe_update_end(new_crtc_state);
 
 	if (new_crtc_state->update_pipe &&
 	    !needs_modeset(&new_crtc_state->base) &&
-	    old_crtc_state->mode.private_flags & I915_MODE_FLAG_INHERITED)
-		intel_crtc_arm_fifo_underrun(intel_crtc, new_crtc_state);
+	    old_crtc_state->base.mode.private_flags & I915_MODE_FLAG_INHERITED)
+		intel_crtc_arm_fifo_underrun(crtc, new_crtc_state);
 }
 
 /**
@@ -14039,14 +14330,11 @@ intel_legacy_cursor_update(struct drm_plane *plane,
 	 */
 	crtc_state->active_planes = new_crtc_state->active_planes;
 
-	if (plane->state->visible) {
-		trace_intel_update_plane(plane, to_intel_crtc(crtc));
-		intel_plane->update_plane(intel_plane, crtc_state,
-					  to_intel_plane_state(plane->state));
-	} else {
-		trace_intel_disable_plane(plane, to_intel_crtc(crtc));
-		intel_plane->disable_plane(intel_plane, crtc_state);
-	}
+	if (plane->state->visible)
+		intel_update_plane(intel_plane, crtc_state,
+				   to_intel_plane_state(plane->state));
+	else
+		intel_disable_plane(intel_plane, crtc_state);
 
 	intel_plane_unpin_fb(to_intel_plane_state(old_plane_state));
 
@@ -14496,7 +14784,12 @@ static void intel_setup_outputs(struct drm_i915_private *dev_priv)
 	if (!HAS_DISPLAY(dev_priv))
 		return;
 
-	if (IS_ICELAKE(dev_priv)) {
+	if (IS_ELKHARTLAKE(dev_priv)) {
+		intel_ddi_init(dev_priv, PORT_A);
+		intel_ddi_init(dev_priv, PORT_B);
+		intel_ddi_init(dev_priv, PORT_C);
+		icl_dsi_init(dev_priv);
+	} else if (INTEL_GEN(dev_priv) >= 11) {
 		intel_ddi_init(dev_priv, PORT_A);
 		intel_ddi_init(dev_priv, PORT_B);
 		intel_ddi_init(dev_priv, PORT_C);
@@ -15397,6 +15690,8 @@ int intel_modeset_init(struct drm_device *dev)
 	intel_update_czclk(dev_priv);
 	intel_modeset_init_hw(dev);
 
+	intel_hdcp_component_init(dev_priv);
+
 	if (dev_priv->max_cdclk_freq == 0)
 		intel_update_max_cdclk(dev_priv);
 
@@ -15472,7 +15767,7 @@ void i830_enable_pipe(struct drm_i915_private *dev_priv, enum pipe pipe)
 		      pipe_name(pipe), clock.vco, clock.dot);
 
 	fp = i9xx_dpll_compute_fp(&clock);
-	dpll = (I915_READ(DPLL(pipe)) & DPLL_DVO_2X_MODE) |
+	dpll = DPLL_DVO_2X_MODE |
 		DPLL_VGA_MODE_DIS |
 		((clock.p1 - 2) << DPLL_FPA01_P1_POST_DIV_SHIFT) |
 		PLL_P2_DIVIDE_BY_4 |
@@ -16254,6 +16549,8 @@ void intel_modeset_cleanup(struct drm_device *dev)
 	/* flush any delayed tasks or pending work */
 	flush_scheduled_work();
 
+	intel_hdcp_component_fini(dev_priv);
+
 	drm_mode_config_cleanup(dev);
 
 	intel_overlay_cleanup(dev_priv);
@@ -16300,8 +16597,6 @@ struct intel_display_error_state {
 
 	u32 power_well_driver;
 
-	int num_transcoders;
-
 	struct intel_cursor_error_state {
 		u32 control;
 		u32 position;
@@ -16326,6 +16621,7 @@ struct intel_display_error_state {
 	} plane[I915_MAX_PIPES];
 
 	struct intel_transcoder_error_state {
+		bool available;
 		bool power_domain_on;
 		enum transcoder cpu_transcoder;
 
@@ -16352,6 +16648,8 @@ intel_display_capture_error_state(struct drm_i915_private *dev_priv)
 	};
 	int i;
 
+	BUILD_BUG_ON(ARRAY_SIZE(transcoders) != ARRAY_SIZE(error->transcoder));
+
 	if (!HAS_DISPLAY(dev_priv))
 		return NULL;
 
@@ -16392,14 +16690,13 @@ intel_display_capture_error_state(struct drm_i915_private *dev_priv)
 			error->pipe[i].stat = I915_READ(PIPESTAT(i));
 	}
 
-	/* Note: this does not include DSI transcoders. */
-	error->num_transcoders = INTEL_INFO(dev_priv)->num_pipes;
-	if (HAS_DDI(dev_priv))
-		error->num_transcoders++; /* Account for eDP. */
-
-	for (i = 0; i < error->num_transcoders; i++) {
+	for (i = 0; i < ARRAY_SIZE(error->transcoder); i++) {
 		enum transcoder cpu_transcoder = transcoders[i];
 
+		if (!INTEL_INFO(dev_priv)->trans_offsets[cpu_transcoder])
+			continue;
+
+		error->transcoder[i].available = true;
 		error->transcoder[i].power_domain_on =
 			__intel_display_power_is_enabled(dev_priv,
 				POWER_DOMAIN_TRANSCODER(cpu_transcoder));
@@ -16463,7 +16760,10 @@ intel_display_print_error_state(struct drm_i915_error_state_buf *m,
 		err_printf(m, "  BASE: %08x\n", error->cursor[i].base);
 	}
 
-	for (i = 0; i < error->num_transcoders; i++) {
+	for (i = 0; i < ARRAY_SIZE(error->transcoder); i++) {
+		if (!error->transcoder[i].available)
+			continue;
+
 		err_printf(m, "CPU transcoder: %s\n",
 			   transcoder_name(error->transcoder[i].cpu_transcoder));
 		err_printf(m, "  Power: %s\n",
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 48da4a969a0a..560274d1c50b 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -25,22 +25,34 @@
  *
  */
 
-#include <linux/i2c.h>
-#include <linux/slab.h>
 #include <linux/export.h>
-#include <linux/types.h>
+#include <linux/i2c.h>
 #include <linux/notifier.h>
 #include <linux/reboot.h>
+#include <linux/slab.h>
+#include <linux/types.h>
 #include <asm/byteorder.h>
+
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_dp_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_hdcp.h>
 #include <drm/drm_probe_helper.h>
-#include "intel_drv.h"
 #include <drm/i915_drm.h>
+
 #include "i915_drv.h"
+#include "intel_audio.h"
+#include "intel_connector.h"
+#include "intel_ddi.h"
+#include "intel_dp.h"
+#include "intel_drv.h"
+#include "intel_hdcp.h"
+#include "intel_hdmi.h"
+#include "intel_lspcon.h"
+#include "intel_lvds.h"
+#include "intel_panel.h"
+#include "intel_psr.h"
 
 #define DP_DPRX_ESI_LEN 14
 
@@ -949,8 +961,11 @@ static void intel_pps_get_registers(struct intel_dp *intel_dp,
 	regs->pp_stat = PP_STATUS(pps_idx);
 	regs->pp_on = PP_ON_DELAYS(pps_idx);
 	regs->pp_off = PP_OFF_DELAYS(pps_idx);
-	if (!IS_GEN9_LP(dev_priv) && !HAS_PCH_CNP(dev_priv) &&
-	    !HAS_PCH_ICP(dev_priv))
+
+	/* Cycle delay moved from PP_DIVISOR to PP_CONTROL */
+	if (IS_GEN9_LP(dev_priv) || INTEL_PCH_TYPE(dev_priv) >= PCH_CNP)
+		regs->pp_div = INVALID_MMIO_REG;
+	else
 		regs->pp_div = PP_DIVISOR(pps_idx);
 }
 
@@ -1720,12 +1735,6 @@ void intel_dp_compute_rate(struct intel_dp *intel_dp, int port_clock,
 	}
 }
 
-struct link_config_limits {
-	int min_clock, max_clock;
-	int min_lane_count, max_lane_count;
-	int min_bpp, max_bpp;
-};
-
 static bool intel_dp_source_supports_fec(struct intel_dp *intel_dp,
 					 const struct intel_crtc_state *pipe_config)
 {
@@ -1788,7 +1797,7 @@ static int intel_dp_compute_bpp(struct intel_dp *intel_dp,
 }
 
 /* Adjust link config limits based on compliance test requests. */
-static void
+void
 intel_dp_adjust_compliance_config(struct intel_dp *intel_dp,
 				  struct intel_crtc_state *pipe_config,
 				  struct link_config_limits *limits)
@@ -1972,6 +1981,14 @@ static int intel_dp_dsc_compute_config(struct intel_dp *intel_dp,
 	return 0;
 }
 
+int intel_dp_min_bpp(const struct intel_crtc_state *crtc_state)
+{
+	if (crtc_state->output_format == INTEL_OUTPUT_FORMAT_RGB)
+		return 6 * 3;
+	else
+		return 8 * 3;
+}
+
 static int
 intel_dp_compute_link_config(struct intel_encoder *encoder,
 			     struct intel_crtc_state *pipe_config,
@@ -1995,7 +2012,7 @@ intel_dp_compute_link_config(struct intel_encoder *encoder,
 	limits.min_lane_count = 1;
 	limits.max_lane_count = intel_dp_max_lane_count(intel_dp);
 
-	limits.min_bpp = 6 * 3;
+	limits.min_bpp = intel_dp_min_bpp(pipe_config);
 	limits.max_bpp = intel_dp_compute_bpp(intel_dp, pipe_config);
 
 	if (intel_dp_is_edp(intel_dp)) {
@@ -2058,6 +2075,29 @@ intel_dp_compute_link_config(struct intel_encoder *encoder,
 	return 0;
 }
 
+bool intel_dp_limited_color_range(const struct intel_crtc_state *crtc_state,
+				  const struct drm_connector_state *conn_state)
+{
+	const struct intel_digital_connector_state *intel_conn_state =
+		to_intel_digital_connector_state(conn_state);
+	const struct drm_display_mode *adjusted_mode =
+		&crtc_state->base.adjusted_mode;
+
+	if (intel_conn_state->broadcast_rgb == INTEL_BROADCAST_RGB_AUTO) {
+		/*
+		 * See:
+		 * CEA-861-E - 5.1 Default Encoding Parameters
+		 * VESA DisplayPort Ver.1.2a - 5.1.1.1 Video Colorimetry
+		 */
+		return crtc_state->pipe_bpp != 18 &&
+			drm_default_rgb_quant_range(adjusted_mode) ==
+			HDMI_QUANTIZATION_RANGE_LIMITED;
+	} else {
+		return intel_conn_state->broadcast_rgb ==
+			INTEL_BROADCAST_RGB_LIMITED;
+	}
+}
+
 int
 intel_dp_compute_config(struct intel_encoder *encoder,
 			struct intel_crtc_state *pipe_config,
@@ -2074,7 +2114,7 @@ intel_dp_compute_config(struct intel_encoder *encoder,
 		to_intel_digital_connector_state(conn_state);
 	bool constant_n = drm_dp_has_quirk(&intel_dp->desc,
 					   DP_DPCD_QUIRK_CONSTANT_N);
-	int ret;
+	int ret, output_bpp;
 
 	if (HAS_PCH_SPLIT(dev_priv) && !HAS_DDI(dev_priv) && port != PORT_A)
 		pipe_config->has_pch_encoder = true;
@@ -2123,40 +2163,25 @@ intel_dp_compute_config(struct intel_encoder *encoder,
 	if (ret < 0)
 		return ret;
 
-	if (intel_conn_state->broadcast_rgb == INTEL_BROADCAST_RGB_AUTO) {
-		/*
-		 * See:
-		 * CEA-861-E - 5.1 Default Encoding Parameters
-		 * VESA DisplayPort Ver.1.2a - 5.1.1.1 Video Colorimetry
-		 */
-		pipe_config->limited_color_range =
-			pipe_config->pipe_bpp != 18 &&
-			drm_default_rgb_quant_range(adjusted_mode) ==
-			HDMI_QUANTIZATION_RANGE_LIMITED;
-	} else {
-		pipe_config->limited_color_range =
-			intel_conn_state->broadcast_rgb == INTEL_BROADCAST_RGB_LIMITED;
-	}
+	pipe_config->limited_color_range =
+		intel_dp_limited_color_range(pipe_config, conn_state);
 
-	if (!pipe_config->dsc_params.compression_enable)
-		intel_link_compute_m_n(pipe_config->pipe_bpp,
-				       pipe_config->lane_count,
-				       adjusted_mode->crtc_clock,
-				       pipe_config->port_clock,
-				       &pipe_config->dp_m_n,
-				       constant_n);
+	if (pipe_config->dsc_params.compression_enable)
+		output_bpp = pipe_config->dsc_params.compressed_bpp;
 	else
-		intel_link_compute_m_n(pipe_config->dsc_params.compressed_bpp,
-				       pipe_config->lane_count,
-				       adjusted_mode->crtc_clock,
-				       pipe_config->port_clock,
-				       &pipe_config->dp_m_n,
-				       constant_n);
+		output_bpp = pipe_config->pipe_bpp;
+
+	intel_link_compute_m_n(output_bpp,
+			       pipe_config->lane_count,
+			       adjusted_mode->crtc_clock,
+			       pipe_config->port_clock,
+			       &pipe_config->dp_m_n,
+			       constant_n);
 
 	if (intel_connector->panel.downclock_mode != NULL &&
 		dev_priv->drrs.type == SEAMLESS_DRRS_SUPPORT) {
 			pipe_config->has_drrs = true;
-			intel_link_compute_m_n(pipe_config->pipe_bpp,
+			intel_link_compute_m_n(output_bpp,
 					       pipe_config->lane_count,
 					       intel_connector->panel.downclock_mode->clock,
 					       pipe_config->port_clock,
@@ -2296,7 +2321,7 @@ static void wait_panel_status(struct intel_dp *intel_dp,
 			I915_READ(pp_stat_reg),
 			I915_READ(pp_ctrl_reg));
 
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    pp_stat_reg, mask, value,
 				    5000))
 		DRM_ERROR("Panel status timeout: status %08x control %08x\n",
@@ -3885,7 +3910,7 @@ void intel_dp_set_idle_link_train(struct intel_dp *intel_dp)
 	if (port == PORT_A)
 		return;
 
-	if (intel_wait_for_register(dev_priv,DP_TP_STATUS(port),
+	if (intel_wait_for_register(&dev_priv->uncore, DP_TP_STATUS(port),
 				    DP_TP_STATUS_IDLE_DONE,
 				    DP_TP_STATUS_IDLE_DONE,
 				    1))
@@ -4731,7 +4756,7 @@ static void intel_dp_check_service_irq(struct intel_dp *intel_dp)
 		intel_dp_handle_test_request(intel_dp);
 
 	if (val & DP_CP_IRQ)
-		intel_hdcp_check_link(intel_dp->attached_connector);
+		intel_hdcp_handle_cp_irq(intel_dp->attached_connector);
 
 	if (val & DP_SINK_SPECIFIC_IRQ)
 		DRM_DEBUG_DRIVER("Sink specific irq unhandled\n");
@@ -5574,6 +5599,18 @@ void intel_dp_encoder_suspend(struct intel_encoder *intel_encoder)
 		edp_panel_vdd_off_sync(intel_dp);
 }
 
+static void intel_dp_hdcp_wait_for_cp_irq(struct intel_hdcp *hdcp, int timeout)
+{
+	long ret;
+
+#define C (hdcp->cp_irq_count_cached != atomic_read(&hdcp->cp_irq_count))
+	ret = wait_event_interruptible_timeout(hdcp->cp_irq_queue, C,
+					       msecs_to_jiffies(timeout));
+
+	if (!ret)
+		DRM_DEBUG_KMS("Timedout at waiting for CP_IRQ\n");
+}
+
 static
 int intel_dp_hdcp_write_an_aksv(struct intel_digital_port *intel_dig_port,
 				u8 *an)
@@ -5798,6 +5835,336 @@ int intel_dp_hdcp_capable(struct intel_digital_port *intel_dig_port,
 	return 0;
 }
 
+struct hdcp2_dp_errata_stream_type {
+	u8	msg_id;
+	u8	stream_type;
+} __packed;
+
+static struct hdcp2_dp_msg_data {
+	u8 msg_id;
+	u32 offset;
+	bool msg_detectable;
+	u32 timeout;
+	u32 timeout2; /* Added for non_paired situation */
+	} hdcp2_msg_data[] = {
+		{HDCP_2_2_AKE_INIT, DP_HDCP_2_2_AKE_INIT_OFFSET, false, 0, 0},
+		{HDCP_2_2_AKE_SEND_CERT, DP_HDCP_2_2_AKE_SEND_CERT_OFFSET,
+				false, HDCP_2_2_CERT_TIMEOUT_MS, 0},
+		{HDCP_2_2_AKE_NO_STORED_KM, DP_HDCP_2_2_AKE_NO_STORED_KM_OFFSET,
+				false, 0, 0},
+		{HDCP_2_2_AKE_STORED_KM, DP_HDCP_2_2_AKE_STORED_KM_OFFSET,
+				false, 0, 0},
+		{HDCP_2_2_AKE_SEND_HPRIME, DP_HDCP_2_2_AKE_SEND_HPRIME_OFFSET,
+				true, HDCP_2_2_HPRIME_PAIRED_TIMEOUT_MS,
+				HDCP_2_2_HPRIME_NO_PAIRED_TIMEOUT_MS},
+		{HDCP_2_2_AKE_SEND_PAIRING_INFO,
+				DP_HDCP_2_2_AKE_SEND_PAIRING_INFO_OFFSET, true,
+				HDCP_2_2_PAIRING_TIMEOUT_MS, 0},
+		{HDCP_2_2_LC_INIT, DP_HDCP_2_2_LC_INIT_OFFSET, false, 0, 0},
+		{HDCP_2_2_LC_SEND_LPRIME, DP_HDCP_2_2_LC_SEND_LPRIME_OFFSET,
+				false, HDCP_2_2_DP_LPRIME_TIMEOUT_MS, 0},
+		{HDCP_2_2_SKE_SEND_EKS, DP_HDCP_2_2_SKE_SEND_EKS_OFFSET, false,
+				0, 0},
+		{HDCP_2_2_REP_SEND_RECVID_LIST,
+				DP_HDCP_2_2_REP_SEND_RECVID_LIST_OFFSET, true,
+				HDCP_2_2_RECVID_LIST_TIMEOUT_MS, 0},
+		{HDCP_2_2_REP_SEND_ACK, DP_HDCP_2_2_REP_SEND_ACK_OFFSET, false,
+				0, 0},
+		{HDCP_2_2_REP_STREAM_MANAGE,
+				DP_HDCP_2_2_REP_STREAM_MANAGE_OFFSET, false,
+				0, 0},
+		{HDCP_2_2_REP_STREAM_READY, DP_HDCP_2_2_REP_STREAM_READY_OFFSET,
+				false, HDCP_2_2_STREAM_READY_TIMEOUT_MS, 0},
+/* local define to shovel this through the write_2_2 interface */
+#define HDCP_2_2_ERRATA_DP_STREAM_TYPE	50
+		{HDCP_2_2_ERRATA_DP_STREAM_TYPE,
+				DP_HDCP_2_2_REG_STREAM_TYPE_OFFSET, false,
+				0, 0},
+		};
+
+static inline
+int intel_dp_hdcp2_read_rx_status(struct intel_digital_port *intel_dig_port,
+				  u8 *rx_status)
+{
+	ssize_t ret;
+
+	ret = drm_dp_dpcd_read(&intel_dig_port->dp.aux,
+			       DP_HDCP_2_2_REG_RXSTATUS_OFFSET, rx_status,
+			       HDCP_2_2_DP_RXSTATUS_LEN);
+	if (ret != HDCP_2_2_DP_RXSTATUS_LEN) {
+		DRM_DEBUG_KMS("Read bstatus from DP/AUX failed (%zd)\n", ret);
+		return ret >= 0 ? -EIO : ret;
+	}
+
+	return 0;
+}
+
+static
+int hdcp2_detect_msg_availability(struct intel_digital_port *intel_dig_port,
+				  u8 msg_id, bool *msg_ready)
+{
+	u8 rx_status;
+	int ret;
+
+	*msg_ready = false;
+	ret = intel_dp_hdcp2_read_rx_status(intel_dig_port, &rx_status);
+	if (ret < 0)
+		return ret;
+
+	switch (msg_id) {
+	case HDCP_2_2_AKE_SEND_HPRIME:
+		if (HDCP_2_2_DP_RXSTATUS_H_PRIME(rx_status))
+			*msg_ready = true;
+		break;
+	case HDCP_2_2_AKE_SEND_PAIRING_INFO:
+		if (HDCP_2_2_DP_RXSTATUS_PAIRING(rx_status))
+			*msg_ready = true;
+		break;
+	case HDCP_2_2_REP_SEND_RECVID_LIST:
+		if (HDCP_2_2_DP_RXSTATUS_READY(rx_status))
+			*msg_ready = true;
+		break;
+	default:
+		DRM_ERROR("Unidentified msg_id: %d\n", msg_id);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static ssize_t
+intel_dp_hdcp2_wait_for_msg(struct intel_digital_port *intel_dig_port,
+			    struct hdcp2_dp_msg_data *hdcp2_msg_data)
+{
+	struct intel_dp *dp = &intel_dig_port->dp;
+	struct intel_hdcp *hdcp = &dp->attached_connector->hdcp;
+	u8 msg_id = hdcp2_msg_data->msg_id;
+	int ret, timeout;
+	bool msg_ready = false;
+
+	if (msg_id == HDCP_2_2_AKE_SEND_HPRIME && !hdcp->is_paired)
+		timeout = hdcp2_msg_data->timeout2;
+	else
+		timeout = hdcp2_msg_data->timeout;
+
+	/*
+	 * There is no way to detect the CERT, LPRIME and STREAM_READY
+	 * availability. So Wait for timeout and read the msg.
+	 */
+	if (!hdcp2_msg_data->msg_detectable) {
+		mdelay(timeout);
+		ret = 0;
+	} else {
+		/*
+		 * As we want to check the msg availability at timeout, Ignoring
+		 * the timeout at wait for CP_IRQ.
+		 */
+		intel_dp_hdcp_wait_for_cp_irq(hdcp, timeout);
+		ret = hdcp2_detect_msg_availability(intel_dig_port,
+						    msg_id, &msg_ready);
+		if (!msg_ready)
+			ret = -ETIMEDOUT;
+	}
+
+	if (ret)
+		DRM_DEBUG_KMS("msg_id %d, ret %d, timeout(mSec): %d\n",
+			      hdcp2_msg_data->msg_id, ret, timeout);
+
+	return ret;
+}
+
+static struct hdcp2_dp_msg_data *get_hdcp2_dp_msg_data(u8 msg_id)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(hdcp2_msg_data); i++)
+		if (hdcp2_msg_data[i].msg_id == msg_id)
+			return &hdcp2_msg_data[i];
+
+	return NULL;
+}
+
+static
+int intel_dp_hdcp2_write_msg(struct intel_digital_port *intel_dig_port,
+			     void *buf, size_t size)
+{
+	struct intel_dp *dp = &intel_dig_port->dp;
+	struct intel_hdcp *hdcp = &dp->attached_connector->hdcp;
+	unsigned int offset;
+	u8 *byte = buf;
+	ssize_t ret, bytes_to_write, len;
+	struct hdcp2_dp_msg_data *hdcp2_msg_data;
+
+	hdcp2_msg_data = get_hdcp2_dp_msg_data(*byte);
+	if (!hdcp2_msg_data)
+		return -EINVAL;
+
+	offset = hdcp2_msg_data->offset;
+
+	/* No msg_id in DP HDCP2.2 msgs */
+	bytes_to_write = size - 1;
+	byte++;
+
+	hdcp->cp_irq_count_cached = atomic_read(&hdcp->cp_irq_count);
+
+	while (bytes_to_write) {
+		len = bytes_to_write > DP_AUX_MAX_PAYLOAD_BYTES ?
+				DP_AUX_MAX_PAYLOAD_BYTES : bytes_to_write;
+
+		ret = drm_dp_dpcd_write(&intel_dig_port->dp.aux,
+					offset, (void *)byte, len);
+		if (ret < 0)
+			return ret;
+
+		bytes_to_write -= ret;
+		byte += ret;
+		offset += ret;
+	}
+
+	return size;
+}
+
+static
+ssize_t get_receiver_id_list_size(struct intel_digital_port *intel_dig_port)
+{
+	u8 rx_info[HDCP_2_2_RXINFO_LEN];
+	u32 dev_cnt;
+	ssize_t ret;
+
+	ret = drm_dp_dpcd_read(&intel_dig_port->dp.aux,
+			       DP_HDCP_2_2_REG_RXINFO_OFFSET,
+			       (void *)rx_info, HDCP_2_2_RXINFO_LEN);
+	if (ret != HDCP_2_2_RXINFO_LEN)
+		return ret >= 0 ? -EIO : ret;
+
+	dev_cnt = (HDCP_2_2_DEV_COUNT_HI(rx_info[0]) << 4 |
+		   HDCP_2_2_DEV_COUNT_LO(rx_info[1]));
+
+	if (dev_cnt > HDCP_2_2_MAX_DEVICE_COUNT)
+		dev_cnt = HDCP_2_2_MAX_DEVICE_COUNT;
+
+	ret = sizeof(struct hdcp2_rep_send_receiverid_list) -
+		HDCP_2_2_RECEIVER_IDS_MAX_LEN +
+		(dev_cnt * HDCP_2_2_RECEIVER_ID_LEN);
+
+	return ret;
+}
+
+static
+int intel_dp_hdcp2_read_msg(struct intel_digital_port *intel_dig_port,
+			    u8 msg_id, void *buf, size_t size)
+{
+	unsigned int offset;
+	u8 *byte = buf;
+	ssize_t ret, bytes_to_recv, len;
+	struct hdcp2_dp_msg_data *hdcp2_msg_data;
+
+	hdcp2_msg_data = get_hdcp2_dp_msg_data(msg_id);
+	if (!hdcp2_msg_data)
+		return -EINVAL;
+	offset = hdcp2_msg_data->offset;
+
+	ret = intel_dp_hdcp2_wait_for_msg(intel_dig_port, hdcp2_msg_data);
+	if (ret < 0)
+		return ret;
+
+	if (msg_id == HDCP_2_2_REP_SEND_RECVID_LIST) {
+		ret = get_receiver_id_list_size(intel_dig_port);
+		if (ret < 0)
+			return ret;
+
+		size = ret;
+	}
+	bytes_to_recv = size - 1;
+
+	/* DP adaptation msgs has no msg_id */
+	byte++;
+
+	while (bytes_to_recv) {
+		len = bytes_to_recv > DP_AUX_MAX_PAYLOAD_BYTES ?
+		      DP_AUX_MAX_PAYLOAD_BYTES : bytes_to_recv;
+
+		ret = drm_dp_dpcd_read(&intel_dig_port->dp.aux, offset,
+				       (void *)byte, len);
+		if (ret < 0) {
+			DRM_DEBUG_KMS("msg_id %d, ret %zd\n", msg_id, ret);
+			return ret;
+		}
+
+		bytes_to_recv -= ret;
+		byte += ret;
+		offset += ret;
+	}
+	byte = buf;
+	*byte = msg_id;
+
+	return size;
+}
+
+static
+int intel_dp_hdcp2_config_stream_type(struct intel_digital_port *intel_dig_port,
+				      bool is_repeater, u8 content_type)
+{
+	struct hdcp2_dp_errata_stream_type stream_type_msg;
+
+	if (is_repeater)
+		return 0;
+
+	/*
+	 * Errata for DP: As Stream type is used for encryption, Receiver
+	 * should be communicated with stream type for the decryption of the
+	 * content.
+	 * Repeater will be communicated with stream type as a part of it's
+	 * auth later in time.
+	 */
+	stream_type_msg.msg_id = HDCP_2_2_ERRATA_DP_STREAM_TYPE;
+	stream_type_msg.stream_type = content_type;
+
+	return intel_dp_hdcp2_write_msg(intel_dig_port, &stream_type_msg,
+					sizeof(stream_type_msg));
+}
+
+static
+int intel_dp_hdcp2_check_link(struct intel_digital_port *intel_dig_port)
+{
+	u8 rx_status;
+	int ret;
+
+	ret = intel_dp_hdcp2_read_rx_status(intel_dig_port, &rx_status);
+	if (ret)
+		return ret;
+
+	if (HDCP_2_2_DP_RXSTATUS_REAUTH_REQ(rx_status))
+		ret = HDCP_REAUTH_REQUEST;
+	else if (HDCP_2_2_DP_RXSTATUS_LINK_FAILED(rx_status))
+		ret = HDCP_LINK_INTEGRITY_FAILURE;
+	else if (HDCP_2_2_DP_RXSTATUS_READY(rx_status))
+		ret = HDCP_TOPOLOGY_CHANGE;
+
+	return ret;
+}
+
+static
+int intel_dp_hdcp2_capable(struct intel_digital_port *intel_dig_port,
+			   bool *capable)
+{
+	u8 rx_caps[3];
+	int ret;
+
+	*capable = false;
+	ret = drm_dp_dpcd_read(&intel_dig_port->dp.aux,
+			       DP_HDCP_2_2_REG_RX_CAPS_OFFSET,
+			       rx_caps, HDCP_2_2_RXCAPS_LEN);
+	if (ret != HDCP_2_2_RXCAPS_LEN)
+		return ret >= 0 ? -EIO : ret;
+
+	if (rx_caps[0] == HDCP_2_2_RX_CAPS_VERSION_VAL &&
+	    HDCP_2_2_DP_HDCP_CAPABLE(rx_caps[2]))
+		*capable = true;
+
+	return 0;
+}
+
 static const struct intel_hdcp_shim intel_dp_hdcp_shim = {
 	.write_an_aksv = intel_dp_hdcp_write_an_aksv,
 	.read_bksv = intel_dp_hdcp_read_bksv,
@@ -5810,6 +6177,12 @@ static const struct intel_hdcp_shim intel_dp_hdcp_shim = {
 	.toggle_signalling = intel_dp_hdcp_toggle_signalling,
 	.check_link = intel_dp_hdcp_check_link,
 	.hdcp_capable = intel_dp_hdcp_capable,
+	.write_2_2_msg = intel_dp_hdcp2_write_msg,
+	.read_2_2_msg = intel_dp_hdcp2_read_msg,
+	.config_stream_type = intel_dp_hdcp2_config_stream_type,
+	.check_2_2_link = intel_dp_hdcp2_check_link,
+	.hdcp_2_2_capable = intel_dp_hdcp2_capable,
+	.protocol = HDCP_PROTOCOL_DP,
 };
 
 static void intel_edp_panel_vdd_sanitize(struct intel_dp *intel_dp)
@@ -6023,43 +6396,34 @@ static void
 intel_pps_readout_hw_state(struct intel_dp *intel_dp, struct edp_power_seq *seq)
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
-	u32 pp_on, pp_off, pp_div = 0, pp_ctl = 0;
+	u32 pp_on, pp_off, pp_ctl;
 	struct pps_registers regs;
 
 	intel_pps_get_registers(intel_dp, &regs);
 
-	/* Workaround: Need to write PP_CONTROL with the unlock key as
-	 * the very first thing. */
 	pp_ctl = ironlake_get_pp_control(intel_dp);
 
+	/* Ensure PPS is unlocked */
+	if (!HAS_DDI(dev_priv))
+		I915_WRITE(regs.pp_ctrl, pp_ctl);
+
 	pp_on = I915_READ(regs.pp_on);
 	pp_off = I915_READ(regs.pp_off);
-	if (!IS_GEN9_LP(dev_priv) && !HAS_PCH_CNP(dev_priv) &&
-	    !HAS_PCH_ICP(dev_priv)) {
-		I915_WRITE(regs.pp_ctrl, pp_ctl);
-		pp_div = I915_READ(regs.pp_div);
-	}
 
 	/* Pull timing values out of registers */
-	seq->t1_t3 = (pp_on & PANEL_POWER_UP_DELAY_MASK) >>
-		     PANEL_POWER_UP_DELAY_SHIFT;
+	seq->t1_t3 = REG_FIELD_GET(PANEL_POWER_UP_DELAY_MASK, pp_on);
+	seq->t8 = REG_FIELD_GET(PANEL_LIGHT_ON_DELAY_MASK, pp_on);
+	seq->t9 = REG_FIELD_GET(PANEL_LIGHT_OFF_DELAY_MASK, pp_off);
+	seq->t10 = REG_FIELD_GET(PANEL_POWER_DOWN_DELAY_MASK, pp_off);
 
-	seq->t8 = (pp_on & PANEL_LIGHT_ON_DELAY_MASK) >>
-		  PANEL_LIGHT_ON_DELAY_SHIFT;
+	if (i915_mmio_reg_valid(regs.pp_div)) {
+		u32 pp_div;
 
-	seq->t9 = (pp_off & PANEL_LIGHT_OFF_DELAY_MASK) >>
-		  PANEL_LIGHT_OFF_DELAY_SHIFT;
-
-	seq->t10 = (pp_off & PANEL_POWER_DOWN_DELAY_MASK) >>
-		   PANEL_POWER_DOWN_DELAY_SHIFT;
+		pp_div = I915_READ(regs.pp_div);
 
-	if (IS_GEN9_LP(dev_priv) || HAS_PCH_CNP(dev_priv) ||
-	    HAS_PCH_ICP(dev_priv)) {
-		seq->t11_t12 = ((pp_ctl & BXT_POWER_CYCLE_DELAY_MASK) >>
-				BXT_POWER_CYCLE_DELAY_SHIFT) * 1000;
+		seq->t11_t12 = REG_FIELD_GET(PANEL_POWER_CYCLE_DELAY_MASK, pp_div) * 1000;
 	} else {
-		seq->t11_t12 = ((pp_div & PANEL_POWER_CYCLE_DELAY_MASK) >>
-		       PANEL_POWER_CYCLE_DELAY_SHIFT) * 1000;
+		seq->t11_t12 = REG_FIELD_GET(BXT_POWER_CYCLE_DELAY_MASK, pp_ctl) * 1000;
 	}
 }
 
@@ -6184,7 +6548,7 @@ intel_dp_init_panel_power_sequencer_registers(struct intel_dp *intel_dp,
 					      bool force_disable_vdd)
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
-	u32 pp_on, pp_off, pp_div, port_sel = 0;
+	u32 pp_on, pp_off, port_sel = 0;
 	int div = dev_priv->rawclk_freq / 1000;
 	struct pps_registers regs;
 	enum port port = dp_to_dig_port(intel_dp)->base.port;
@@ -6219,23 +6583,10 @@ intel_dp_init_panel_power_sequencer_registers(struct intel_dp *intel_dp,
 		I915_WRITE(regs.pp_ctrl, pp);
 	}
 
-	pp_on = (seq->t1_t3 << PANEL_POWER_UP_DELAY_SHIFT) |
-		(seq->t8 << PANEL_LIGHT_ON_DELAY_SHIFT);
-	pp_off = (seq->t9 << PANEL_LIGHT_OFF_DELAY_SHIFT) |
-		 (seq->t10 << PANEL_POWER_DOWN_DELAY_SHIFT);
-	/* Compute the divisor for the pp clock, simply match the Bspec
-	 * formula. */
-	if (IS_GEN9_LP(dev_priv) || HAS_PCH_CNP(dev_priv) ||
-	    HAS_PCH_ICP(dev_priv)) {
-		pp_div = I915_READ(regs.pp_ctrl);
-		pp_div &= ~BXT_POWER_CYCLE_DELAY_MASK;
-		pp_div |= (DIV_ROUND_UP(seq->t11_t12, 1000)
-				<< BXT_POWER_CYCLE_DELAY_SHIFT);
-	} else {
-		pp_div = ((100 * div)/2 - 1) << PP_REFERENCE_DIVIDER_SHIFT;
-		pp_div |= (DIV_ROUND_UP(seq->t11_t12, 1000)
-				<< PANEL_POWER_CYCLE_DELAY_SHIFT);
-	}
+	pp_on = REG_FIELD_PREP(PANEL_POWER_UP_DELAY_MASK, seq->t1_t3) |
+		REG_FIELD_PREP(PANEL_LIGHT_ON_DELAY_MASK, seq->t8);
+	pp_off = REG_FIELD_PREP(PANEL_LIGHT_OFF_DELAY_MASK, seq->t9) |
+		REG_FIELD_PREP(PANEL_POWER_DOWN_DELAY_MASK, seq->t10);
 
 	/* Haswell doesn't have any port selection bits for the panel
 	 * power sequencer any more. */
@@ -6262,19 +6613,29 @@ intel_dp_init_panel_power_sequencer_registers(struct intel_dp *intel_dp,
 
 	I915_WRITE(regs.pp_on, pp_on);
 	I915_WRITE(regs.pp_off, pp_off);
-	if (IS_GEN9_LP(dev_priv) || HAS_PCH_CNP(dev_priv) ||
-	    HAS_PCH_ICP(dev_priv))
-		I915_WRITE(regs.pp_ctrl, pp_div);
-	else
-		I915_WRITE(regs.pp_div, pp_div);
+
+	/*
+	 * Compute the divisor for the pp clock, simply match the Bspec formula.
+	 */
+	if (i915_mmio_reg_valid(regs.pp_div)) {
+		I915_WRITE(regs.pp_div,
+			   REG_FIELD_PREP(PP_REFERENCE_DIVIDER_MASK, (100 * div) / 2 - 1) |
+			   REG_FIELD_PREP(PANEL_POWER_CYCLE_DELAY_MASK, DIV_ROUND_UP(seq->t11_t12, 1000)));
+	} else {
+		u32 pp_ctl;
+
+		pp_ctl = I915_READ(regs.pp_ctrl);
+		pp_ctl &= ~BXT_POWER_CYCLE_DELAY_MASK;
+		pp_ctl |= REG_FIELD_PREP(BXT_POWER_CYCLE_DELAY_MASK, DIV_ROUND_UP(seq->t11_t12, 1000));
+		I915_WRITE(regs.pp_ctrl, pp_ctl);
+	}
 
 	DRM_DEBUG_KMS("panel power sequencer register settings: PP_ON %#x, PP_OFF %#x, PP_DIV %#x\n",
 		      I915_READ(regs.pp_on),
 		      I915_READ(regs.pp_off),
-		      (IS_GEN9_LP(dev_priv) || HAS_PCH_CNP(dev_priv)  ||
-		       HAS_PCH_ICP(dev_priv)) ?
-		      (I915_READ(regs.pp_ctrl) & BXT_POWER_CYCLE_DELAY_MASK) :
-		      I915_READ(regs.pp_div));
+		      i915_mmio_reg_valid(regs.pp_div) ?
+		      I915_READ(regs.pp_div) :
+		      (I915_READ(regs.pp_ctrl) & BXT_POWER_CYCLE_DELAY_MASK));
 }
 
 static void intel_dp_pps_init(struct intel_dp *intel_dp)
@@ -6645,9 +7006,7 @@ intel_dp_drrs_init(struct intel_connector *connector,
 		return NULL;
 	}
 
-	downclock_mode = intel_find_panel_downclock(dev_priv, fixed_mode,
-						    &connector->base);
-
+	downclock_mode = intel_panel_edid_downclock_mode(connector, fixed_mode);
 	if (!downclock_mode) {
 		DRM_DEBUG_KMS("Downclock mode is not found. DRRS not supported\n");
 		return NULL;
@@ -6669,7 +7028,6 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp,
 	struct drm_display_mode *fixed_mode = NULL;
 	struct drm_display_mode *downclock_mode = NULL;
 	bool has_dpcd;
-	struct drm_display_mode *scan;
 	enum pipe pipe = INVALID_PIPE;
 	intel_wakeref_t wakeref;
 	struct edid *edid;
@@ -6685,7 +7043,7 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp,
 	 * eDP and LVDS bail out early in this case to prevent interfering
 	 * with an already powered-on LVDS power sequencer.
 	 */
-	if (intel_get_lvds_encoder(&dev_priv->drm)) {
+	if (intel_get_lvds_encoder(dev_priv)) {
 		WARN_ON(!(HAS_PCH_IBX(dev_priv) || HAS_PCH_CPT(dev_priv)));
 		DRM_INFO("LVDS was detected, not registering eDP\n");
 
@@ -6722,26 +7080,13 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp,
 	}
 	intel_connector->edid = edid;
 
-	/* prefer fixed mode from EDID if available */
-	list_for_each_entry(scan, &connector->probed_modes, head) {
-		if ((scan->type & DRM_MODE_TYPE_PREFERRED)) {
-			fixed_mode = drm_mode_duplicate(dev, scan);
-			downclock_mode = intel_dp_drrs_init(
-						intel_connector, fixed_mode);
-			break;
-		}
-	}
+	fixed_mode = intel_panel_edid_fixed_mode(intel_connector);
+	if (fixed_mode)
+		downclock_mode = intel_dp_drrs_init(intel_connector, fixed_mode);
 
 	/* fallback to VBT if available for eDP */
-	if (!fixed_mode && dev_priv->vbt.lfp_lvds_vbt_mode) {
-		fixed_mode = drm_mode_duplicate(dev,
-					dev_priv->vbt.lfp_lvds_vbt_mode);
-		if (fixed_mode) {
-			fixed_mode->type |= DRM_MODE_TYPE_PREFERRED;
-			connector->display_info.width_mm = fixed_mode->width_mm;
-			connector->display_info.height_mm = fixed_mode->height_mm;
-		}
-	}
+	if (!fixed_mode)
+		fixed_mode = intel_panel_vbt_fixed_mode(intel_connector);
 	mutex_unlock(&dev->mode_config.mutex);
 
 	if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {
diff --git a/drivers/gpu/drm/i915/intel_dp.h b/drivers/gpu/drm/i915/intel_dp.h
new file mode 100644
index 000000000000..5e9e8d13de6e
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_dp.h
@@ -0,0 +1,122 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_DP_H__
+#define __INTEL_DP_H__
+
+#include <linux/types.h>
+
+#include <drm/i915_drm.h>
+
+#include "i915_reg.h"
+
+enum pipe;
+struct drm_connector_state;
+struct drm_encoder;
+struct drm_i915_private;
+struct drm_modeset_acquire_ctx;
+struct intel_connector;
+struct intel_crtc_state;
+struct intel_digital_port;
+struct intel_dp;
+struct intel_encoder;
+
+struct link_config_limits {
+	int min_clock, max_clock;
+	int min_lane_count, max_lane_count;
+	int min_bpp, max_bpp;
+};
+
+void intel_dp_adjust_compliance_config(struct intel_dp *intel_dp,
+				       struct intel_crtc_state *pipe_config,
+				       struct link_config_limits *limits);
+bool intel_dp_limited_color_range(const struct intel_crtc_state *crtc_state,
+				  const struct drm_connector_state *conn_state);
+int intel_dp_min_bpp(const struct intel_crtc_state *crtc_state);
+bool intel_dp_port_enabled(struct drm_i915_private *dev_priv,
+			   i915_reg_t dp_reg, enum port port,
+			   enum pipe *pipe);
+bool intel_dp_init(struct drm_i915_private *dev_priv, i915_reg_t output_reg,
+		   enum port port);
+bool intel_dp_init_connector(struct intel_digital_port *intel_dig_port,
+			     struct intel_connector *intel_connector);
+void intel_dp_set_link_params(struct intel_dp *intel_dp,
+			      int link_rate, u8 lane_count,
+			      bool link_mst);
+int intel_dp_get_link_train_fallback_values(struct intel_dp *intel_dp,
+					    int link_rate, u8 lane_count);
+int intel_dp_retrain_link(struct intel_encoder *encoder,
+			  struct drm_modeset_acquire_ctx *ctx);
+void intel_dp_sink_dpms(struct intel_dp *intel_dp, int mode);
+void intel_dp_sink_set_decompression_state(struct intel_dp *intel_dp,
+					   const struct intel_crtc_state *crtc_state,
+					   bool enable);
+void intel_dp_encoder_reset(struct drm_encoder *encoder);
+void intel_dp_encoder_suspend(struct intel_encoder *intel_encoder);
+void intel_dp_encoder_flush_work(struct drm_encoder *encoder);
+int intel_dp_compute_config(struct intel_encoder *encoder,
+			    struct intel_crtc_state *pipe_config,
+			    struct drm_connector_state *conn_state);
+bool intel_dp_is_edp(struct intel_dp *intel_dp);
+bool intel_dp_is_port_edp(struct drm_i915_private *dev_priv, enum port port);
+enum irqreturn intel_dp_hpd_pulse(struct intel_digital_port *intel_dig_port,
+				  bool long_hpd);
+void intel_edp_backlight_on(const struct intel_crtc_state *crtc_state,
+			    const struct drm_connector_state *conn_state);
+void intel_edp_backlight_off(const struct drm_connector_state *conn_state);
+void intel_edp_panel_vdd_on(struct intel_dp *intel_dp);
+void intel_edp_panel_on(struct intel_dp *intel_dp);
+void intel_edp_panel_off(struct intel_dp *intel_dp);
+void intel_dp_mst_suspend(struct drm_i915_private *dev_priv);
+void intel_dp_mst_resume(struct drm_i915_private *dev_priv);
+int intel_dp_max_link_rate(struct intel_dp *intel_dp);
+int intel_dp_max_lane_count(struct intel_dp *intel_dp);
+int intel_dp_rate_select(struct intel_dp *intel_dp, int rate);
+void intel_power_sequencer_reset(struct drm_i915_private *dev_priv);
+u32 intel_dp_pack_aux(const u8 *src, int src_bytes);
+
+void intel_edp_drrs_enable(struct intel_dp *intel_dp,
+			   const struct intel_crtc_state *crtc_state);
+void intel_edp_drrs_disable(struct intel_dp *intel_dp,
+			    const struct intel_crtc_state *crtc_state);
+void intel_edp_drrs_invalidate(struct drm_i915_private *dev_priv,
+			       unsigned int frontbuffer_bits);
+void intel_edp_drrs_flush(struct drm_i915_private *dev_priv,
+			  unsigned int frontbuffer_bits);
+
+void
+intel_dp_program_link_training_pattern(struct intel_dp *intel_dp,
+				       u8 dp_train_pat);
+void
+intel_dp_set_signal_levels(struct intel_dp *intel_dp);
+void intel_dp_set_idle_link_train(struct intel_dp *intel_dp);
+u8
+intel_dp_voltage_max(struct intel_dp *intel_dp);
+u8
+intel_dp_pre_emphasis_max(struct intel_dp *intel_dp, u8 voltage_swing);
+void intel_dp_compute_rate(struct intel_dp *intel_dp, int port_clock,
+			   u8 *link_bw, u8 *rate_select);
+bool intel_dp_source_supports_hbr2(struct intel_dp *intel_dp);
+bool intel_dp_source_supports_hbr3(struct intel_dp *intel_dp);
+bool
+intel_dp_get_link_status(struct intel_dp *intel_dp, u8 *link_status);
+u16 intel_dp_dsc_get_output_bpp(int link_clock, u8 lane_count,
+				int mode_clock, int mode_hdisplay);
+u8 intel_dp_dsc_get_slice_count(struct intel_dp *intel_dp, int mode_clock,
+				int mode_hdisplay);
+
+bool intel_dp_read_dpcd(struct intel_dp *intel_dp);
+int intel_dp_link_required(int pixel_clock, int bpp);
+int intel_dp_max_data_rate(int max_link_clock, int max_lanes);
+bool intel_digital_port_connected(struct intel_encoder *encoder);
+void icl_tc_phy_disconnect(struct drm_i915_private *dev_priv,
+			   struct intel_digital_port *dig_port);
+
+static inline unsigned int intel_dp_unused_lane_mask(int lane_count)
+{
+	return ~((1 << lane_count) - 1) & 0xf;
+}
+
+#endif /* __INTEL_DP_H__ */
diff --git a/drivers/gpu/drm/i915/intel_dp_link_training.c b/drivers/gpu/drm/i915/intel_dp_link_training.c
index b59c87daa4f7..54b069333e2f 100644
--- a/drivers/gpu/drm/i915/intel_dp_link_training.c
+++ b/drivers/gpu/drm/i915/intel_dp_link_training.c
@@ -21,6 +21,7 @@
  * IN THE SOFTWARE.
  */
 
+#include "intel_dp.h"
 #include "intel_drv.h"
 
 static void
diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c
index fb67cd931117..8839eaea8371 100644
--- a/drivers/gpu/drm/i915/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/intel_dp_mst.c
@@ -23,78 +23,119 @@
  *
  */
 
-#include "i915_drv.h"
-#include "intel_drv.h"
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_probe_helper.h>
 
+#include "i915_drv.h"
+#include "intel_audio.h"
+#include "intel_connector.h"
+#include "intel_ddi.h"
+#include "intel_dp.h"
+#include "intel_drv.h"
+
+static int intel_dp_mst_compute_link_config(struct intel_encoder *encoder,
+					    struct intel_crtc_state *crtc_state,
+					    struct drm_connector_state *conn_state,
+					    struct link_config_limits *limits)
+{
+	struct drm_atomic_state *state = crtc_state->base.state;
+	struct intel_dp_mst_encoder *intel_mst = enc_to_mst(&encoder->base);
+	struct intel_dp *intel_dp = &intel_mst->primary->dp;
+	struct intel_connector *connector =
+		to_intel_connector(conn_state->connector);
+	const struct drm_display_mode *adjusted_mode =
+		&crtc_state->base.adjusted_mode;
+	void *port = connector->port;
+	bool constant_n = drm_dp_has_quirk(&intel_dp->desc,
+					   DP_DPCD_QUIRK_CONSTANT_N);
+	int bpp, slots = -EINVAL;
+
+	crtc_state->lane_count = limits->max_lane_count;
+	crtc_state->port_clock = limits->max_clock;
+
+	for (bpp = limits->max_bpp; bpp >= limits->min_bpp; bpp -= 2 * 3) {
+		crtc_state->pipe_bpp = bpp;
+
+		crtc_state->pbn = drm_dp_calc_pbn_mode(adjusted_mode->crtc_clock,
+						       crtc_state->pipe_bpp);
+
+		slots = drm_dp_atomic_find_vcpi_slots(state, &intel_dp->mst_mgr,
+						      port, crtc_state->pbn);
+		if (slots == -EDEADLK)
+			return slots;
+		if (slots >= 0)
+			break;
+	}
+
+	if (slots < 0) {
+		DRM_DEBUG_KMS("failed finding vcpi slots:%d\n", slots);
+		return slots;
+	}
+
+	intel_link_compute_m_n(crtc_state->pipe_bpp,
+			       crtc_state->lane_count,
+			       adjusted_mode->crtc_clock,
+			       crtc_state->port_clock,
+			       &crtc_state->dp_m_n,
+			       constant_n);
+	crtc_state->dp_m_n.tu = slots;
+
+	return 0;
+}
+
 static int intel_dp_mst_compute_config(struct intel_encoder *encoder,
 				       struct intel_crtc_state *pipe_config,
 				       struct drm_connector_state *conn_state)
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_dp_mst_encoder *intel_mst = enc_to_mst(&encoder->base);
-	struct intel_digital_port *intel_dig_port = intel_mst->primary;
-	struct intel_dp *intel_dp = &intel_dig_port->dp;
-	struct drm_connector *connector = conn_state->connector;
-	void *port = to_intel_connector(connector)->port;
-	struct drm_atomic_state *state = pipe_config->base.state;
-	struct drm_crtc *crtc = pipe_config->base.crtc;
-	struct drm_crtc_state *old_crtc_state =
-		drm_atomic_get_old_crtc_state(state, crtc);
-	int bpp;
-	int lane_count, slots =
-		to_intel_crtc_state(old_crtc_state)->dp_m_n.tu;
-	const struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
-	int mst_pbn;
-	bool constant_n = drm_dp_has_quirk(&intel_dp->desc,
-					   DP_DPCD_QUIRK_CONSTANT_N);
+	struct intel_dp *intel_dp = &intel_mst->primary->dp;
+	struct intel_connector *connector =
+		to_intel_connector(conn_state->connector);
+	struct intel_digital_connector_state *intel_conn_state =
+		to_intel_digital_connector_state(conn_state);
+	const struct drm_display_mode *adjusted_mode =
+		&pipe_config->base.adjusted_mode;
+	void *port = connector->port;
+	struct link_config_limits limits;
+	int ret;
 
 	if (adjusted_mode->flags & DRM_MODE_FLAG_DBLSCAN)
 		return -EINVAL;
 
 	pipe_config->output_format = INTEL_OUTPUT_FORMAT_RGB;
 	pipe_config->has_pch_encoder = false;
-	bpp = 24;
-	if (intel_dp->compliance.test_data.bpc) {
-		bpp = intel_dp->compliance.test_data.bpc * 3;
-		DRM_DEBUG_KMS("Setting pipe bpp to %d\n",
-			      bpp);
-	}
+
+	if (intel_conn_state->force_audio == HDMI_AUDIO_AUTO)
+		pipe_config->has_audio =
+			drm_dp_mst_port_has_audio(&intel_dp->mst_mgr, port);
+	else
+		pipe_config->has_audio =
+			intel_conn_state->force_audio == HDMI_AUDIO_ON;
+
 	/*
 	 * for MST we always configure max link bw - the spec doesn't
 	 * seem to suggest we should do otherwise.
 	 */
-	lane_count = intel_dp_max_lane_count(intel_dp);
-
-	pipe_config->lane_count = lane_count;
-
-	pipe_config->pipe_bpp = bpp;
+	limits.min_clock =
+	limits.max_clock = intel_dp_max_link_rate(intel_dp);
 
-	pipe_config->port_clock = intel_dp_max_link_rate(intel_dp);
+	limits.min_lane_count =
+	limits.max_lane_count = intel_dp_max_lane_count(intel_dp);
 
-	if (drm_dp_mst_port_has_audio(&intel_dp->mst_mgr, port))
-		pipe_config->has_audio = true;
+	limits.min_bpp = intel_dp_min_bpp(pipe_config);
+	limits.max_bpp = pipe_config->pipe_bpp;
 
-	mst_pbn = drm_dp_calc_pbn_mode(adjusted_mode->crtc_clock, bpp);
-	pipe_config->pbn = mst_pbn;
+	intel_dp_adjust_compliance_config(intel_dp, pipe_config, &limits);
 
-	slots = drm_dp_atomic_find_vcpi_slots(state, &intel_dp->mst_mgr, port,
-					      mst_pbn);
-	if (slots < 0) {
-		DRM_DEBUG_KMS("failed finding vcpi slots:%d\n",
-			      slots);
-		return slots;
-	}
-
-	intel_link_compute_m_n(bpp, lane_count,
-			       adjusted_mode->crtc_clock,
-			       pipe_config->port_clock,
-			       &pipe_config->dp_m_n,
-			       constant_n);
+	ret = intel_dp_mst_compute_link_config(encoder, pipe_config,
+					       conn_state, &limits);
+	if (ret)
+		return ret;
 
-	pipe_config->dp_m_n.tu = slots;
+	pipe_config->limited_color_range =
+		intel_dp_limited_color_range(pipe_config, conn_state);
 
 	if (IS_GEN9_LP(dev_priv))
 		pipe_config->lane_lat_optim_mask =
@@ -117,7 +158,11 @@ intel_dp_mst_atomic_check(struct drm_connector *connector,
 	struct drm_crtc *new_crtc = new_conn_state->crtc;
 	struct drm_crtc_state *crtc_state;
 	struct drm_dp_mst_topology_mgr *mgr;
-	int ret = 0;
+	int ret;
+
+	ret = intel_digital_connector_atomic_check(connector, new_conn_state);
+	if (ret)
+		return ret;
 
 	if (!old_conn_state->crtc)
 		return 0;
@@ -289,7 +334,7 @@ static void intel_mst_enable_dp(struct intel_encoder *encoder,
 
 	DRM_DEBUG_KMS("active links %d\n", intel_dp->active_mst_links);
 
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    DP_TP_STATUS(port),
 				    DP_TP_STATUS_ACT_SENT,
 				    DP_TP_STATUS_ACT_SENT,
@@ -354,11 +399,13 @@ intel_dp_mst_detect(struct drm_connector *connector, bool force)
 static const struct drm_connector_funcs intel_dp_mst_connector_funcs = {
 	.detect = intel_dp_mst_detect,
 	.fill_modes = drm_helper_probe_single_connector_modes,
+	.atomic_get_property = intel_digital_connector_atomic_get_property,
+	.atomic_set_property = intel_digital_connector_atomic_set_property,
 	.late_register = intel_connector_register,
 	.early_unregister = intel_connector_unregister,
 	.destroy = intel_connector_destroy,
 	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
-	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
+	.atomic_duplicate_state = intel_digital_connector_duplicate_state,
 };
 
 static int intel_dp_mst_get_modes(struct drm_connector *connector)
@@ -373,7 +420,6 @@ intel_dp_mst_mode_valid(struct drm_connector *connector,
 	struct intel_connector *intel_connector = to_intel_connector(connector);
 	struct intel_dp *intel_dp = intel_connector->mst_port;
 	int max_dotclk = to_i915(connector->dev)->max_dotclk_freq;
-	int bpp = 24; /* MST uses fixed bpp */
 	int max_rate, mode_rate, max_lanes, max_link_clock;
 
 	if (drm_connector_is_unregistered(connector))
@@ -386,7 +432,7 @@ intel_dp_mst_mode_valid(struct drm_connector *connector,
 	max_lanes = intel_dp_max_lane_count(intel_dp);
 
 	max_rate = intel_dp_max_data_rate(max_link_clock, max_lanes);
-	mode_rate = intel_dp_link_required(mode->clock, bpp);
+	mode_rate = intel_dp_link_required(mode->clock, 18);
 
 	/* TODO - validate mode against available PBN for link */
 	if (mode->clock < 10000)
@@ -487,6 +533,10 @@ static struct drm_connector *intel_dp_add_mst_connector(struct drm_dp_mst_topolo
 	if (ret)
 		goto err;
 
+	intel_attach_force_audio_property(connector);
+	intel_attach_broadcast_rgb_property(connector);
+	drm_connector_attach_max_bpc_property(connector, 6, 12);
+
 	return connector;
 
 err:
diff --git a/drivers/gpu/drm/i915/intel_dpio_phy.c b/drivers/gpu/drm/i915/intel_dpio_phy.c
index 95cb8b154f87..ab4ac7158b79 100644
--- a/drivers/gpu/drm/i915/intel_dpio_phy.c
+++ b/drivers/gpu/drm/i915/intel_dpio_phy.c
@@ -21,6 +21,7 @@
  * DEALINGS IN THE SOFTWARE.
  */
 
+#include "intel_dp.h"
 #include "intel_drv.h"
 
 /**
@@ -341,7 +342,7 @@ static u32 bxt_get_grc(struct drm_i915_private *dev_priv, enum dpio_phy phy)
 static void bxt_phy_wait_grc_done(struct drm_i915_private *dev_priv,
 				  enum dpio_phy phy)
 {
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    BXT_PORT_REF_DW3(phy),
 				    GRC_DONE, GRC_DONE,
 				    10))
@@ -383,7 +384,8 @@ static void _bxt_ddi_phy_init(struct drm_i915_private *dev_priv,
 	 * The flag should get set in 100us according to the HW team, but
 	 * use 1ms due to occasional timeouts observed with that.
 	 */
-	if (intel_wait_for_register_fw(dev_priv, BXT_PORT_CL1CM_DW0(phy),
+	if (intel_wait_for_register_fw(&dev_priv->uncore,
+				       BXT_PORT_CL1CM_DW0(phy),
 				       PHY_RESERVED | PHY_POWER_GOOD,
 				       PHY_POWER_GOOD,
 				       1))
diff --git a/drivers/gpu/drm/i915/intel_dpll_mgr.c b/drivers/gpu/drm/i915/intel_dpll_mgr.c
index 0a42d11c4c33..e01c057ce50b 100644
--- a/drivers/gpu/drm/i915/intel_dpll_mgr.c
+++ b/drivers/gpu/drm/i915/intel_dpll_mgr.c
@@ -241,11 +241,11 @@ out:
 }
 
 static struct intel_shared_dpll *
-intel_find_shared_dpll(struct intel_crtc *crtc,
-		       struct intel_crtc_state *crtc_state,
+intel_find_shared_dpll(struct intel_crtc_state *crtc_state,
 		       enum intel_dpll_id range_min,
 		       enum intel_dpll_id range_max)
 {
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	struct intel_shared_dpll *pll, *unused_pll = NULL;
 	struct intel_shared_dpll_state *shared_dpll;
@@ -420,9 +420,10 @@ static void ibx_pch_dpll_disable(struct drm_i915_private *dev_priv,
 }
 
 static struct intel_shared_dpll *
-ibx_get_dpll(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state,
+ibx_get_dpll(struct intel_crtc_state *crtc_state,
 	     struct intel_encoder *encoder)
 {
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	struct intel_shared_dpll *pll;
 	enum intel_dpll_id i;
@@ -436,7 +437,7 @@ ibx_get_dpll(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state,
 			      crtc->base.base.id, crtc->base.name,
 			      pll->info->name);
 	} else {
-		pll = intel_find_shared_dpll(crtc, crtc_state,
+		pll = intel_find_shared_dpll(crtc_state,
 					     DPLL_ID_PCH_PLL_A,
 					     DPLL_ID_PCH_PLL_B);
 	}
@@ -764,15 +765,13 @@ hsw_ddi_calculate_wrpll(int clock /* in Hz */,
 	*r2_out = best.r2;
 }
 
-static struct intel_shared_dpll *hsw_ddi_hdmi_get_dpll(int clock,
-						       struct intel_crtc *crtc,
-						       struct intel_crtc_state *crtc_state)
+static struct intel_shared_dpll *hsw_ddi_hdmi_get_dpll(struct intel_crtc_state *crtc_state)
 {
 	struct intel_shared_dpll *pll;
 	u32 val;
 	unsigned int p, n2, r2;
 
-	hsw_ddi_calculate_wrpll(clock * 1000, &r2, &n2, &p);
+	hsw_ddi_calculate_wrpll(crtc_state->port_clock * 1000, &r2, &n2, &p);
 
 	val = WRPLL_PLL_ENABLE | WRPLL_PLL_LCPLL |
 	      WRPLL_DIVIDER_REFERENCE(r2) | WRPLL_DIVIDER_FEEDBACK(n2) |
@@ -780,7 +779,7 @@ static struct intel_shared_dpll *hsw_ddi_hdmi_get_dpll(int clock,
 
 	crtc_state->dpll_hw_state.wrpll = val;
 
-	pll = intel_find_shared_dpll(crtc, crtc_state,
+	pll = intel_find_shared_dpll(crtc_state,
 				     DPLL_ID_WRPLL1, DPLL_ID_WRPLL2);
 
 	if (!pll)
@@ -790,11 +789,12 @@ static struct intel_shared_dpll *hsw_ddi_hdmi_get_dpll(int clock,
 }
 
 static struct intel_shared_dpll *
-hsw_ddi_dp_get_dpll(struct intel_encoder *encoder, int clock)
+hsw_ddi_dp_get_dpll(struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct drm_i915_private *dev_priv = to_i915(crtc_state->base.crtc->dev);
 	struct intel_shared_dpll *pll;
 	enum intel_dpll_id pll_id;
+	int clock = crtc_state->port_clock;
 
 	switch (clock / 2) {
 	case 81000:
@@ -820,19 +820,18 @@ hsw_ddi_dp_get_dpll(struct intel_encoder *encoder, int clock)
 }
 
 static struct intel_shared_dpll *
-hsw_get_dpll(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state,
+hsw_get_dpll(struct intel_crtc_state *crtc_state,
 	     struct intel_encoder *encoder)
 {
 	struct intel_shared_dpll *pll;
-	int clock = crtc_state->port_clock;
 
 	memset(&crtc_state->dpll_hw_state, 0,
 	       sizeof(crtc_state->dpll_hw_state));
 
 	if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI)) {
-		pll = hsw_ddi_hdmi_get_dpll(clock, crtc, crtc_state);
+		pll = hsw_ddi_hdmi_get_dpll(crtc_state);
 	} else if (intel_crtc_has_dp_encoder(crtc_state)) {
-		pll = hsw_ddi_dp_get_dpll(encoder, clock);
+		pll = hsw_ddi_dp_get_dpll(crtc_state);
 	} else if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_ANALOG)) {
 		if (WARN_ON(crtc_state->port_clock / 2 != 135000))
 			return NULL;
@@ -840,7 +839,7 @@ hsw_get_dpll(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state,
 		crtc_state->dpll_hw_state.spll =
 			SPLL_PLL_ENABLE | SPLL_PLL_FREQ_1350MHz | SPLL_PLL_SSC;
 
-		pll = intel_find_shared_dpll(crtc, crtc_state,
+		pll = intel_find_shared_dpll(crtc_state,
 					     DPLL_ID_SPLL, DPLL_ID_SPLL);
 	} else {
 		return NULL;
@@ -961,7 +960,7 @@ static void skl_ddi_pll_enable(struct drm_i915_private *dev_priv,
 	I915_WRITE(regs[id].ctl,
 		   I915_READ(regs[id].ctl) | LCPLL_PLL_ENABLE);
 
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    DPLL_STATUS,
 				    DPLL_LOCK(id),
 				    DPLL_LOCK(id),
@@ -1308,9 +1307,7 @@ skip_remaining_dividers:
 	return true;
 }
 
-static bool skl_ddi_hdmi_pll_dividers(struct intel_crtc *crtc,
-				      struct intel_crtc_state *crtc_state,
-				      int clock)
+static bool skl_ddi_hdmi_pll_dividers(struct intel_crtc_state *crtc_state)
 {
 	u32 ctrl1, cfgcr1, cfgcr2;
 	struct skl_wrpll_params wrpll_params = { 0, };
@@ -1323,7 +1320,8 @@ static bool skl_ddi_hdmi_pll_dividers(struct intel_crtc *crtc,
 
 	ctrl1 |= DPLL_CTRL1_HDMI_MODE(0);
 
-	if (!skl_ddi_calculate_wrpll(clock * 1000, &wrpll_params))
+	if (!skl_ddi_calculate_wrpll(crtc_state->port_clock * 1000,
+				     &wrpll_params))
 		return false;
 
 	cfgcr1 = DPLL_CFGCR1_FREQ_ENABLE |
@@ -1346,8 +1344,7 @@ static bool skl_ddi_hdmi_pll_dividers(struct intel_crtc *crtc,
 }
 
 static bool
-skl_ddi_dp_set_dpll_hw_state(int clock,
-			     struct intel_dpll_hw_state *dpll_hw_state)
+skl_ddi_dp_set_dpll_hw_state(struct intel_crtc_state *crtc_state)
 {
 	u32 ctrl1;
 
@@ -1356,7 +1353,7 @@ skl_ddi_dp_set_dpll_hw_state(int clock,
 	 * as the DPLL id in this function.
 	 */
 	ctrl1 = DPLL_CTRL1_OVERRIDE(0);
-	switch (clock / 2) {
+	switch (crtc_state->port_clock / 2) {
 	case 81000:
 		ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_810, 0);
 		break;
@@ -1378,44 +1375,43 @@ skl_ddi_dp_set_dpll_hw_state(int clock,
 		break;
 	}
 
-	dpll_hw_state->ctrl1 = ctrl1;
+	memset(&crtc_state->dpll_hw_state, 0,
+	       sizeof(crtc_state->dpll_hw_state));
+
+	crtc_state->dpll_hw_state.ctrl1 = ctrl1;
+
 	return true;
 }
 
 static struct intel_shared_dpll *
-skl_get_dpll(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state,
+skl_get_dpll(struct intel_crtc_state *crtc_state,
 	     struct intel_encoder *encoder)
 {
 	struct intel_shared_dpll *pll;
-	int clock = crtc_state->port_clock;
 	bool bret;
-	struct intel_dpll_hw_state dpll_hw_state;
-
-	memset(&dpll_hw_state, 0, sizeof(dpll_hw_state));
 
 	if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI)) {
-		bret = skl_ddi_hdmi_pll_dividers(crtc, crtc_state, clock);
+		bret = skl_ddi_hdmi_pll_dividers(crtc_state);
 		if (!bret) {
 			DRM_DEBUG_KMS("Could not get HDMI pll dividers.\n");
 			return NULL;
 		}
 	} else if (intel_crtc_has_dp_encoder(crtc_state)) {
-		bret = skl_ddi_dp_set_dpll_hw_state(clock, &dpll_hw_state);
+		bret = skl_ddi_dp_set_dpll_hw_state(crtc_state);
 		if (!bret) {
 			DRM_DEBUG_KMS("Could not set DP dpll HW state.\n");
 			return NULL;
 		}
-		crtc_state->dpll_hw_state = dpll_hw_state;
 	} else {
 		return NULL;
 	}
 
 	if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_EDP))
-		pll = intel_find_shared_dpll(crtc, crtc_state,
+		pll = intel_find_shared_dpll(crtc_state,
 					     DPLL_ID_SKL_DPLL0,
 					     DPLL_ID_SKL_DPLL0);
 	else
-		pll = intel_find_shared_dpll(crtc, crtc_state,
+		pll = intel_find_shared_dpll(crtc_state,
 					     DPLL_ID_SKL_DPLL1,
 					     DPLL_ID_SKL_DPLL3);
 	if (!pll)
@@ -1692,10 +1688,10 @@ static const struct bxt_clk_div bxt_dp_clk_val[] = {
 };
 
 static bool
-bxt_ddi_hdmi_pll_dividers(struct intel_crtc *intel_crtc,
-			  struct intel_crtc_state *crtc_state, int clock,
+bxt_ddi_hdmi_pll_dividers(struct intel_crtc_state *crtc_state,
 			  struct bxt_clk_div *clk_div)
 {
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
 	struct dpll best_clock;
 
 	/* Calculate HDMI div */
@@ -1703,9 +1699,10 @@ bxt_ddi_hdmi_pll_dividers(struct intel_crtc *intel_crtc,
 	 * FIXME: tie the following calculation into
 	 * i9xx_crtc_compute_clock
 	 */
-	if (!bxt_find_best_dpll(crtc_state, clock, &best_clock)) {
+	if (!bxt_find_best_dpll(crtc_state, &best_clock)) {
 		DRM_DEBUG_DRIVER("no PLL dividers found for clock %d pipe %c\n",
-				 clock, pipe_name(intel_crtc->pipe));
+				 crtc_state->port_clock,
+				 pipe_name(crtc->pipe));
 		return false;
 	}
 
@@ -1722,8 +1719,10 @@ bxt_ddi_hdmi_pll_dividers(struct intel_crtc *intel_crtc,
 	return true;
 }
 
-static void bxt_ddi_dp_pll_dividers(int clock, struct bxt_clk_div *clk_div)
+static void bxt_ddi_dp_pll_dividers(struct intel_crtc_state *crtc_state,
+				    struct bxt_clk_div *clk_div)
 {
+	int clock = crtc_state->port_clock;
 	int i;
 
 	*clk_div = bxt_dp_clk_val[0];
@@ -1737,14 +1736,17 @@ static void bxt_ddi_dp_pll_dividers(int clock, struct bxt_clk_div *clk_div)
 	clk_div->vco = clock * 10 / 2 * clk_div->p1 * clk_div->p2;
 }
 
-static bool bxt_ddi_set_dpll_hw_state(int clock,
-			  struct bxt_clk_div *clk_div,
-			  struct intel_dpll_hw_state *dpll_hw_state)
+static bool bxt_ddi_set_dpll_hw_state(struct intel_crtc_state *crtc_state,
+				      const struct bxt_clk_div *clk_div)
 {
+	struct intel_dpll_hw_state *dpll_hw_state = &crtc_state->dpll_hw_state;
+	int clock = crtc_state->port_clock;
 	int vco = clk_div->vco;
 	u32 prop_coef, int_coef, gain_ctl, targ_cnt;
 	u32 lanestagger;
 
+	memset(dpll_hw_state, 0, sizeof(*dpll_hw_state));
+
 	if (vco >= 6200000 && vco <= 6700000) {
 		prop_coef = 4;
 		int_coef = 9;
@@ -1804,55 +1806,45 @@ static bool bxt_ddi_set_dpll_hw_state(int clock,
 }
 
 static bool
-bxt_ddi_dp_set_dpll_hw_state(int clock,
-			     struct intel_dpll_hw_state *dpll_hw_state)
+bxt_ddi_dp_set_dpll_hw_state(struct intel_crtc_state *crtc_state)
 {
-	struct bxt_clk_div clk_div = {0};
+	struct bxt_clk_div clk_div = {};
 
-	bxt_ddi_dp_pll_dividers(clock, &clk_div);
+	bxt_ddi_dp_pll_dividers(crtc_state, &clk_div);
 
-	return bxt_ddi_set_dpll_hw_state(clock, &clk_div, dpll_hw_state);
+	return bxt_ddi_set_dpll_hw_state(crtc_state, &clk_div);
 }
 
 static bool
-bxt_ddi_hdmi_set_dpll_hw_state(struct intel_crtc *intel_crtc,
-			       struct intel_crtc_state *crtc_state, int clock,
-			       struct intel_dpll_hw_state *dpll_hw_state)
+bxt_ddi_hdmi_set_dpll_hw_state(struct intel_crtc_state *crtc_state)
 {
-	struct bxt_clk_div clk_div = { };
+	struct bxt_clk_div clk_div = {};
 
-	bxt_ddi_hdmi_pll_dividers(intel_crtc, crtc_state, clock, &clk_div);
+	bxt_ddi_hdmi_pll_dividers(crtc_state, &clk_div);
 
-	return bxt_ddi_set_dpll_hw_state(clock, &clk_div, dpll_hw_state);
+	return bxt_ddi_set_dpll_hw_state(crtc_state, &clk_div);
 }
 
 static struct intel_shared_dpll *
-bxt_get_dpll(struct intel_crtc *crtc,
-		struct intel_crtc_state *crtc_state,
-		struct intel_encoder *encoder)
+bxt_get_dpll(struct intel_crtc_state *crtc_state,
+	     struct intel_encoder *encoder)
 {
-	struct intel_dpll_hw_state dpll_hw_state = { };
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	struct intel_shared_dpll *pll;
-	int i, clock = crtc_state->port_clock;
+	enum intel_dpll_id id;
 
 	if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI) &&
-	    !bxt_ddi_hdmi_set_dpll_hw_state(crtc, crtc_state, clock,
-					    &dpll_hw_state))
+	    !bxt_ddi_hdmi_set_dpll_hw_state(crtc_state))
 		return NULL;
 
 	if (intel_crtc_has_dp_encoder(crtc_state) &&
-	    !bxt_ddi_dp_set_dpll_hw_state(clock, &dpll_hw_state))
+	    !bxt_ddi_dp_set_dpll_hw_state(crtc_state))
 		return NULL;
 
-	memset(&crtc_state->dpll_hw_state, 0,
-	       sizeof(crtc_state->dpll_hw_state));
-
-	crtc_state->dpll_hw_state = dpll_hw_state;
-
 	/* 1:1 mapping between ports and PLLs */
-	i = (enum intel_dpll_id) encoder->port;
-	pll = intel_get_shared_dpll_by_id(dev_priv, i);
+	id = (enum intel_dpll_id) encoder->port;
+	pll = intel_get_shared_dpll_by_id(dev_priv, id);
 
 	DRM_DEBUG_KMS("[CRTC:%d:%s] using pre-allocated %s\n",
 		      crtc->base.base.id, crtc->base.name, pll->info->name);
@@ -1911,8 +1903,7 @@ static void intel_ddi_pll_init(struct drm_device *dev)
 struct intel_dpll_mgr {
 	const struct dpll_info *dpll_info;
 
-	struct intel_shared_dpll *(*get_dpll)(struct intel_crtc *crtc,
-					      struct intel_crtc_state *crtc_state,
+	struct intel_shared_dpll *(*get_dpll)(struct intel_crtc_state *crtc_state,
 					      struct intel_encoder *encoder);
 
 	void (*dump_hw_state)(struct drm_i915_private *dev_priv,
@@ -1986,7 +1977,7 @@ static void cnl_ddi_pll_enable(struct drm_i915_private *dev_priv,
 	I915_WRITE(CNL_DPLL_ENABLE(id), val);
 
 	/* 2. Wait for DPLL power state enabled in DPLL_ENABLE. */
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    CNL_DPLL_ENABLE(id),
 				    PLL_POWER_STATE,
 				    PLL_POWER_STATE,
@@ -2027,7 +2018,7 @@ static void cnl_ddi_pll_enable(struct drm_i915_private *dev_priv,
 	I915_WRITE(CNL_DPLL_ENABLE(id), val);
 
 	/* 7. Wait for PLL lock status in DPLL_ENABLE. */
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    CNL_DPLL_ENABLE(id),
 				    PLL_LOCK,
 				    PLL_LOCK,
@@ -2075,7 +2066,7 @@ static void cnl_ddi_pll_disable(struct drm_i915_private *dev_priv,
 	I915_WRITE(CNL_DPLL_ENABLE(id), val);
 
 	/* 4. Wait for PLL not locked status in DPLL_ENABLE. */
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    CNL_DPLL_ENABLE(id),
 				    PLL_LOCK,
 				    0,
@@ -2097,7 +2088,7 @@ static void cnl_ddi_pll_disable(struct drm_i915_private *dev_priv,
 	I915_WRITE(CNL_DPLL_ENABLE(id), val);
 
 	/* 7. Wait for DPLL power state disabled in DPLL_ENABLE. */
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    CNL_DPLL_ENABLE(id),
 				    PLL_POWER_STATE,
 				    0,
@@ -2242,11 +2233,11 @@ int cnl_hdmi_pll_ref_clock(struct drm_i915_private *dev_priv)
 }
 
 static bool
-cnl_ddi_calculate_wrpll(int clock,
-			struct drm_i915_private *dev_priv,
+cnl_ddi_calculate_wrpll(struct intel_crtc_state *crtc_state,
 			struct skl_wrpll_params *wrpll_params)
 {
-	u32 afe_clock = clock * 5;
+	struct drm_i915_private *dev_priv = to_i915(crtc_state->base.crtc->dev);
+	u32 afe_clock = crtc_state->port_clock * 5;
 	u32 ref_clock;
 	u32 dco_min = 7998000;
 	u32 dco_max = 10000000;
@@ -2282,23 +2273,20 @@ cnl_ddi_calculate_wrpll(int clock,
 
 	ref_clock = cnl_hdmi_pll_ref_clock(dev_priv);
 
-	cnl_wrpll_params_populate(wrpll_params, best_dco, ref_clock, pdiv, qdiv,
-				  kdiv);
+	cnl_wrpll_params_populate(wrpll_params, best_dco, ref_clock,
+				  pdiv, qdiv, kdiv);
 
 	return true;
 }
 
-static bool cnl_ddi_hdmi_pll_dividers(struct intel_crtc *crtc,
-				      struct intel_crtc_state *crtc_state,
-				      int clock)
+static bool cnl_ddi_hdmi_pll_dividers(struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	u32 cfgcr0, cfgcr1;
 	struct skl_wrpll_params wrpll_params = { 0, };
 
 	cfgcr0 = DPLL_CFGCR0_HDMI_MODE;
 
-	if (!cnl_ddi_calculate_wrpll(clock, dev_priv, &wrpll_params))
+	if (!cnl_ddi_calculate_wrpll(crtc_state, &wrpll_params))
 		return false;
 
 	cfgcr0 |= DPLL_CFGCR0_DCO_FRACTION(wrpll_params.dco_fraction) |
@@ -2319,14 +2307,13 @@ static bool cnl_ddi_hdmi_pll_dividers(struct intel_crtc *crtc,
 }
 
 static bool
-cnl_ddi_dp_set_dpll_hw_state(int clock,
-			     struct intel_dpll_hw_state *dpll_hw_state)
+cnl_ddi_dp_set_dpll_hw_state(struct intel_crtc_state *crtc_state)
 {
 	u32 cfgcr0;
 
 	cfgcr0 = DPLL_CFGCR0_SSC_ENABLE;
 
-	switch (clock / 2) {
+	switch (crtc_state->port_clock / 2) {
 	case 81000:
 		cfgcr0 |= DPLL_CFGCR0_LINK_RATE_810;
 		break;
@@ -2356,41 +2343,40 @@ cnl_ddi_dp_set_dpll_hw_state(int clock,
 		break;
 	}
 
-	dpll_hw_state->cfgcr0 = cfgcr0;
+	memset(&crtc_state->dpll_hw_state, 0,
+	       sizeof(crtc_state->dpll_hw_state));
+
+	crtc_state->dpll_hw_state.cfgcr0 = cfgcr0;
+
 	return true;
 }
 
 static struct intel_shared_dpll *
-cnl_get_dpll(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state,
+cnl_get_dpll(struct intel_crtc_state *crtc_state,
 	     struct intel_encoder *encoder)
 {
 	struct intel_shared_dpll *pll;
-	int clock = crtc_state->port_clock;
 	bool bret;
-	struct intel_dpll_hw_state dpll_hw_state;
-
-	memset(&dpll_hw_state, 0, sizeof(dpll_hw_state));
 
 	if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI)) {
-		bret = cnl_ddi_hdmi_pll_dividers(crtc, crtc_state, clock);
+		bret = cnl_ddi_hdmi_pll_dividers(crtc_state);
 		if (!bret) {
 			DRM_DEBUG_KMS("Could not get HDMI pll dividers.\n");
 			return NULL;
 		}
 	} else if (intel_crtc_has_dp_encoder(crtc_state)) {
-		bret = cnl_ddi_dp_set_dpll_hw_state(clock, &dpll_hw_state);
+		bret = cnl_ddi_dp_set_dpll_hw_state(crtc_state);
 		if (!bret) {
 			DRM_DEBUG_KMS("Could not set DP dpll HW state.\n");
 			return NULL;
 		}
-		crtc_state->dpll_hw_state = dpll_hw_state;
 	} else {
 		DRM_DEBUG_KMS("Skip DPLL setup for output_types 0x%x\n",
 			      crtc_state->output_types);
 		return NULL;
 	}
 
-	pll = intel_find_shared_dpll(crtc, crtc_state,
+	pll = intel_find_shared_dpll(crtc_state,
 				     DPLL_ID_SKL_DPLL0,
 				     DPLL_ID_SKL_DPLL2);
 	if (!pll) {
@@ -2431,47 +2417,69 @@ static const struct intel_dpll_mgr cnl_pll_mgr = {
 	.dump_hw_state = cnl_dump_hw_state,
 };
 
+struct icl_combo_pll_params {
+	int clock;
+	struct skl_wrpll_params wrpll;
+};
+
 /*
  * These values alrea already adjusted: they're the bits we write to the
  * registers, not the logical values.
  */
-static const struct skl_wrpll_params icl_dp_combo_pll_24MHz_values[] = {
-	{ .dco_integer = 0x151, .dco_fraction = 0x4000,		/* [0]: 5.4 */
-	  .pdiv = 0x2 /* 3 */, .kdiv = 1, .qdiv_mode = 0, .qdiv_ratio = 0},
-	{ .dco_integer = 0x151, .dco_fraction = 0x4000,		/* [1]: 2.7 */
-	  .pdiv = 0x2 /* 3 */, .kdiv = 2, .qdiv_mode = 0, .qdiv_ratio = 0},
-	{ .dco_integer = 0x151, .dco_fraction = 0x4000,		/* [2]: 1.62 */
-	  .pdiv = 0x4 /* 5 */, .kdiv = 2, .qdiv_mode = 0, .qdiv_ratio = 0},
-	{ .dco_integer = 0x151, .dco_fraction = 0x4000,		/* [3]: 3.24 */
-	  .pdiv = 0x4 /* 5 */, .kdiv = 1, .qdiv_mode = 0, .qdiv_ratio = 0},
-	{ .dco_integer = 0x168, .dco_fraction = 0x0000,		/* [4]: 2.16 */
-	  .pdiv = 0x1 /* 2 */, .kdiv = 2, .qdiv_mode = 1, .qdiv_ratio = 2},
-	{ .dco_integer = 0x168, .dco_fraction = 0x0000,		/* [5]: 4.32 */
-	  .pdiv = 0x1 /* 2 */, .kdiv = 2, .qdiv_mode = 0, .qdiv_ratio = 0},
-	{ .dco_integer = 0x195, .dco_fraction = 0x0000,		/* [6]: 6.48 */
-	  .pdiv = 0x2 /* 3 */, .kdiv = 1, .qdiv_mode = 0, .qdiv_ratio = 0},
-	{ .dco_integer = 0x151, .dco_fraction = 0x4000,		/* [7]: 8.1 */
-	  .pdiv = 0x1 /* 2 */, .kdiv = 1, .qdiv_mode = 0, .qdiv_ratio = 0},
+static const struct icl_combo_pll_params icl_dp_combo_pll_24MHz_values[] = {
+	{ 540000,
+	  { .dco_integer = 0x151, .dco_fraction = 0x4000,		/* [0]: 5.4 */
+	    .pdiv = 0x2 /* 3 */, .kdiv = 1, .qdiv_mode = 0, .qdiv_ratio = 0, }, },
+	{ 270000,
+	  { .dco_integer = 0x151, .dco_fraction = 0x4000,		/* [1]: 2.7 */
+	    .pdiv = 0x2 /* 3 */, .kdiv = 2, .qdiv_mode = 0, .qdiv_ratio = 0, }, },
+	{ 162000,
+	  { .dco_integer = 0x151, .dco_fraction = 0x4000,		/* [2]: 1.62 */
+	    .pdiv = 0x4 /* 5 */, .kdiv = 2, .qdiv_mode = 0, .qdiv_ratio = 0, }, },
+	{ 324000,
+	  { .dco_integer = 0x151, .dco_fraction = 0x4000,		/* [3]: 3.24 */
+	    .pdiv = 0x4 /* 5 */, .kdiv = 1, .qdiv_mode = 0, .qdiv_ratio = 0, }, },
+	{ 216000,
+	  { .dco_integer = 0x168, .dco_fraction = 0x0000,		/* [4]: 2.16 */
+	    .pdiv = 0x1 /* 2 */, .kdiv = 2, .qdiv_mode = 1, .qdiv_ratio = 2, }, },
+	{ 432000,
+	  { .dco_integer = 0x168, .dco_fraction = 0x0000,		/* [5]: 4.32 */
+	    .pdiv = 0x1 /* 2 */, .kdiv = 2, .qdiv_mode = 0, .qdiv_ratio = 0, }, },
+	{ 648000,
+	  { .dco_integer = 0x195, .dco_fraction = 0x0000,		/* [6]: 6.48 */
+	    .pdiv = 0x2 /* 3 */, .kdiv = 1, .qdiv_mode = 0, .qdiv_ratio = 0, }, },
+	{ 810000,
+	  { .dco_integer = 0x151, .dco_fraction = 0x4000,		/* [7]: 8.1 */
+	    .pdiv = 0x1 /* 2 */, .kdiv = 1, .qdiv_mode = 0, .qdiv_ratio = 0, }, },
 };
 
+
 /* Also used for 38.4 MHz values. */
-static const struct skl_wrpll_params icl_dp_combo_pll_19_2MHz_values[] = {
-	{ .dco_integer = 0x1A5, .dco_fraction = 0x7000,		/* [0]: 5.4 */
-	  .pdiv = 0x2 /* 3 */, .kdiv = 1, .qdiv_mode = 0, .qdiv_ratio = 0},
-	{ .dco_integer = 0x1A5, .dco_fraction = 0x7000,		/* [1]: 2.7 */
-	  .pdiv = 0x2 /* 3 */, .kdiv = 2, .qdiv_mode = 0, .qdiv_ratio = 0},
-	{ .dco_integer = 0x1A5, .dco_fraction = 0x7000,		/* [2]: 1.62 */
-	  .pdiv = 0x4 /* 5 */, .kdiv = 2, .qdiv_mode = 0, .qdiv_ratio = 0},
-	{ .dco_integer = 0x1A5, .dco_fraction = 0x7000,		/* [3]: 3.24 */
-	  .pdiv = 0x4 /* 5 */, .kdiv = 1, .qdiv_mode = 0, .qdiv_ratio = 0},
-	{ .dco_integer = 0x1C2, .dco_fraction = 0x0000,		/* [4]: 2.16 */
-	  .pdiv = 0x1 /* 2 */, .kdiv = 2, .qdiv_mode = 1, .qdiv_ratio = 2},
-	{ .dco_integer = 0x1C2, .dco_fraction = 0x0000,		/* [5]: 4.32 */
-	  .pdiv = 0x1 /* 2 */, .kdiv = 2, .qdiv_mode = 0, .qdiv_ratio = 0},
-	{ .dco_integer = 0x1FA, .dco_fraction = 0x2000,		/* [6]: 6.48 */
-	  .pdiv = 0x2 /* 3 */, .kdiv = 1, .qdiv_mode = 0, .qdiv_ratio = 0},
-	{ .dco_integer = 0x1A5, .dco_fraction = 0x7000,		/* [7]: 8.1 */
-	  .pdiv = 0x1 /* 2 */, .kdiv = 1, .qdiv_mode = 0, .qdiv_ratio = 0},
+static const struct icl_combo_pll_params icl_dp_combo_pll_19_2MHz_values[] = {
+	{ 540000,
+	  { .dco_integer = 0x1A5, .dco_fraction = 0x7000,		/* [0]: 5.4 */
+	    .pdiv = 0x2 /* 3 */, .kdiv = 1, .qdiv_mode = 0, .qdiv_ratio = 0, }, },
+	{ 270000,
+	  { .dco_integer = 0x1A5, .dco_fraction = 0x7000,		/* [1]: 2.7 */
+	    .pdiv = 0x2 /* 3 */, .kdiv = 2, .qdiv_mode = 0, .qdiv_ratio = 0, }, },
+	{ 162000,
+	  { .dco_integer = 0x1A5, .dco_fraction = 0x7000,		/* [2]: 1.62 */
+	    .pdiv = 0x4 /* 5 */, .kdiv = 2, .qdiv_mode = 0, .qdiv_ratio = 0, }, },
+	{ 324000,
+	  { .dco_integer = 0x1A5, .dco_fraction = 0x7000,		/* [3]: 3.24 */
+	    .pdiv = 0x4 /* 5 */, .kdiv = 1, .qdiv_mode = 0, .qdiv_ratio = 0, }, },
+	{ 216000,
+	  { .dco_integer = 0x1C2, .dco_fraction = 0x0000,		/* [4]: 2.16 */
+	    .pdiv = 0x1 /* 2 */, .kdiv = 2, .qdiv_mode = 1, .qdiv_ratio = 2, }, },
+	{ 432000,
+	  { .dco_integer = 0x1C2, .dco_fraction = 0x0000,		/* [5]: 4.32 */
+	    .pdiv = 0x1 /* 2 */, .kdiv = 2, .qdiv_mode = 0, .qdiv_ratio = 0, }, },
+	{ 648000,
+	  { .dco_integer = 0x1FA, .dco_fraction = 0x2000,		/* [6]: 6.48 */
+	    .pdiv = 0x2 /* 3 */, .kdiv = 1, .qdiv_mode = 0, .qdiv_ratio = 0, }, },
+	{ 810000,
+	  { .dco_integer = 0x1A5, .dco_fraction = 0x7000,		/* [7]: 8.1 */
+	    .pdiv = 0x1 /* 2 */, .kdiv = 1, .qdiv_mode = 0, .qdiv_ratio = 0, }, },
 };
 
 static const struct skl_wrpll_params icl_tbt_pll_24MHz_values = {
@@ -2484,72 +2492,53 @@ static const struct skl_wrpll_params icl_tbt_pll_19_2MHz_values = {
 	.pdiv = 0x4 /* 5 */, .kdiv = 1, .qdiv_mode = 0, .qdiv_ratio = 0,
 };
 
-static bool icl_calc_dp_combo_pll(struct drm_i915_private *dev_priv, int clock,
+static bool icl_calc_dp_combo_pll(struct intel_crtc_state *crtc_state,
 				  struct skl_wrpll_params *pll_params)
 {
-	const struct skl_wrpll_params *params;
-
-	params = dev_priv->cdclk.hw.ref == 24000 ?
-			icl_dp_combo_pll_24MHz_values :
-			icl_dp_combo_pll_19_2MHz_values;
+	struct drm_i915_private *dev_priv = to_i915(crtc_state->base.crtc->dev);
+	const struct icl_combo_pll_params *params =
+		dev_priv->cdclk.hw.ref == 24000 ?
+		icl_dp_combo_pll_24MHz_values :
+		icl_dp_combo_pll_19_2MHz_values;
+	int clock = crtc_state->port_clock;
+	int i;
 
-	switch (clock) {
-	case 540000:
-		*pll_params = params[0];
-		break;
-	case 270000:
-		*pll_params = params[1];
-		break;
-	case 162000:
-		*pll_params = params[2];
-		break;
-	case 324000:
-		*pll_params = params[3];
-		break;
-	case 216000:
-		*pll_params = params[4];
-		break;
-	case 432000:
-		*pll_params = params[5];
-		break;
-	case 648000:
-		*pll_params = params[6];
-		break;
-	case 810000:
-		*pll_params = params[7];
-		break;
-	default:
-		MISSING_CASE(clock);
-		return false;
+	for (i = 0; i < ARRAY_SIZE(icl_dp_combo_pll_24MHz_values); i++) {
+		if (clock == params[i].clock) {
+			*pll_params = params[i].wrpll;
+			return true;
+		}
 	}
 
-	return true;
+	MISSING_CASE(clock);
+	return false;
 }
 
-static bool icl_calc_tbt_pll(struct drm_i915_private *dev_priv, int clock,
+static bool icl_calc_tbt_pll(struct intel_crtc_state *crtc_state,
 			     struct skl_wrpll_params *pll_params)
 {
+	struct drm_i915_private *dev_priv = to_i915(crtc_state->base.crtc->dev);
+
 	*pll_params = dev_priv->cdclk.hw.ref == 24000 ?
 			icl_tbt_pll_24MHz_values : icl_tbt_pll_19_2MHz_values;
 	return true;
 }
 
 static bool icl_calc_dpll_state(struct intel_crtc_state *crtc_state,
-				struct intel_encoder *encoder, int clock,
-				struct intel_dpll_hw_state *pll_state)
+				struct intel_encoder *encoder)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct drm_i915_private *dev_priv = to_i915(crtc_state->base.crtc->dev);
 	u32 cfgcr0, cfgcr1;
 	struct skl_wrpll_params pll_params = { 0 };
 	bool ret;
 
 	if (intel_port_is_tc(dev_priv, encoder->port))
-		ret = icl_calc_tbt_pll(dev_priv, clock, &pll_params);
+		ret = icl_calc_tbt_pll(crtc_state, &pll_params);
 	else if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI) ||
 		 intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DSI))
-		ret = cnl_ddi_calculate_wrpll(clock, dev_priv, &pll_params);
+		ret = cnl_ddi_calculate_wrpll(crtc_state, &pll_params);
 	else
-		ret = icl_calc_dp_combo_pll(dev_priv, clock, &pll_params);
+		ret = icl_calc_dp_combo_pll(crtc_state, &pll_params);
 
 	if (!ret)
 		return false;
@@ -2563,82 +2552,16 @@ static bool icl_calc_dpll_state(struct intel_crtc_state *crtc_state,
 		 DPLL_CFGCR1_PDIV(pll_params.pdiv) |
 		 DPLL_CFGCR1_CENTRAL_FREQ_8400;
 
-	pll_state->cfgcr0 = cfgcr0;
-	pll_state->cfgcr1 = cfgcr1;
-	return true;
-}
-
-int icl_calc_dp_combo_pll_link(struct drm_i915_private *dev_priv,
-			       u32 pll_id)
-{
-	u32 cfgcr0, cfgcr1;
-	u32 pdiv, kdiv, qdiv_mode, qdiv_ratio, dco_integer, dco_fraction;
-	const struct skl_wrpll_params *params;
-	int index, n_entries, link_clock;
-
-	/* Read back values from DPLL CFGCR registers */
-	cfgcr0 = I915_READ(ICL_DPLL_CFGCR0(pll_id));
-	cfgcr1 = I915_READ(ICL_DPLL_CFGCR1(pll_id));
-
-	dco_integer = cfgcr0 & DPLL_CFGCR0_DCO_INTEGER_MASK;
-	dco_fraction = (cfgcr0 & DPLL_CFGCR0_DCO_FRACTION_MASK) >>
-		DPLL_CFGCR0_DCO_FRACTION_SHIFT;
-	pdiv = (cfgcr1 & DPLL_CFGCR1_PDIV_MASK) >> DPLL_CFGCR1_PDIV_SHIFT;
-	kdiv = (cfgcr1 & DPLL_CFGCR1_KDIV_MASK) >> DPLL_CFGCR1_KDIV_SHIFT;
-	qdiv_mode = (cfgcr1 & DPLL_CFGCR1_QDIV_MODE(1)) >>
-		DPLL_CFGCR1_QDIV_MODE_SHIFT;
-	qdiv_ratio = (cfgcr1 & DPLL_CFGCR1_QDIV_RATIO_MASK) >>
-		DPLL_CFGCR1_QDIV_RATIO_SHIFT;
-
-	params = dev_priv->cdclk.hw.ref == 24000 ?
-		icl_dp_combo_pll_24MHz_values :
-		icl_dp_combo_pll_19_2MHz_values;
-	n_entries = ARRAY_SIZE(icl_dp_combo_pll_24MHz_values);
-
-	for (index = 0; index < n_entries; index++) {
-		if (dco_integer == params[index].dco_integer &&
-		    dco_fraction == params[index].dco_fraction &&
-		    pdiv == params[index].pdiv &&
-		    kdiv == params[index].kdiv &&
-		    qdiv_mode == params[index].qdiv_mode &&
-		    qdiv_ratio == params[index].qdiv_ratio)
-			break;
-	}
+	memset(&crtc_state->dpll_hw_state, 0,
+	       sizeof(crtc_state->dpll_hw_state));
 
-	/* Map PLL Index to Link Clock */
-	switch (index) {
-	default:
-		MISSING_CASE(index);
-		/* fall through */
-	case 0:
-		link_clock = 540000;
-		break;
-	case 1:
-		link_clock = 270000;
-		break;
-	case 2:
-		link_clock = 162000;
-		break;
-	case 3:
-		link_clock = 324000;
-		break;
-	case 4:
-		link_clock = 216000;
-		break;
-	case 5:
-		link_clock = 432000;
-		break;
-	case 6:
-		link_clock = 648000;
-		break;
-	case 7:
-		link_clock = 810000;
-		break;
-	}
+	crtc_state->dpll_hw_state.cfgcr0 = cfgcr0;
+	crtc_state->dpll_hw_state.cfgcr1 = cfgcr1;
 
-	return link_clock;
+	return true;
 }
 
+
 static enum tc_port icl_pll_id_to_tc_port(enum intel_dpll_id id)
 {
 	return id - DPLL_ID_ICL_MGPLL1;
@@ -2649,11 +2572,6 @@ enum intel_dpll_id icl_tc_port_to_pll_id(enum tc_port tc_port)
 	return tc_port + DPLL_ID_ICL_MGPLL1;
 }
 
-bool intel_dpll_is_combophy(enum intel_dpll_id id)
-{
-	return id == DPLL_ID_ICL_DPLL0 || id == DPLL_ID_ICL_DPLL1;
-}
-
 static bool icl_mg_pll_find_divisors(int clock_khz, bool is_dp, bool use_ssc,
 				     u32 *target_dco_khz,
 				     struct intel_dpll_hw_state *state)
@@ -2728,12 +2646,12 @@ static bool icl_mg_pll_find_divisors(int clock_khz, bool is_dp, bool use_ssc,
  * The specification for this function uses real numbers, so the math had to be
  * adapted to integer-only calculation, that's why it looks so different.
  */
-static bool icl_calc_mg_pll_state(struct intel_crtc_state *crtc_state,
-				  struct intel_encoder *encoder, int clock,
-				  struct intel_dpll_hw_state *pll_state)
+static bool icl_calc_mg_pll_state(struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct drm_i915_private *dev_priv = to_i915(crtc_state->base.crtc->dev);
+	struct intel_dpll_hw_state *pll_state = &crtc_state->dpll_hw_state;
 	int refclk_khz = dev_priv->cdclk.hw.ref;
+	int clock = crtc_state->port_clock;
 	u32 dco_khz, m1div, m2div_int, m2div_rem, m2div_frac;
 	u32 iref_ndiv, iref_trim, iref_pulse_w;
 	u32 prop_coeff, int_coeff;
@@ -2743,6 +2661,8 @@ static bool icl_calc_mg_pll_state(struct intel_crtc_state *crtc_state,
 	bool use_ssc = false;
 	bool is_dp = !intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI);
 
+	memset(pll_state, 0, sizeof(*pll_state));
+
 	if (!icl_mg_pll_find_divisors(clock, is_dp, use_ssc, &dco_khz,
 				      pll_state)) {
 		DRM_DEBUG_KMS("Failed to find divisors for clock %d\n", clock);
@@ -2892,23 +2812,20 @@ static bool icl_calc_mg_pll_state(struct intel_crtc_state *crtc_state,
 }
 
 static struct intel_shared_dpll *
-icl_get_dpll(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state,
+icl_get_dpll(struct intel_crtc_state *crtc_state,
 	     struct intel_encoder *encoder)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct drm_i915_private *dev_priv = to_i915(crtc_state->base.crtc->dev);
 	struct intel_digital_port *intel_dig_port;
 	struct intel_shared_dpll *pll;
-	struct intel_dpll_hw_state pll_state = {};
 	enum port port = encoder->port;
 	enum intel_dpll_id min, max;
-	int clock = crtc_state->port_clock;
 	bool ret;
 
 	if (intel_port_is_combophy(dev_priv, port)) {
 		min = DPLL_ID_ICL_DPLL0;
 		max = DPLL_ID_ICL_DPLL1;
-		ret = icl_calc_dpll_state(crtc_state, encoder, clock,
-					  &pll_state);
+		ret = icl_calc_dpll_state(crtc_state, encoder);
 	} else if (intel_port_is_tc(dev_priv, port)) {
 		if (encoder->type == INTEL_OUTPUT_DP_MST) {
 			struct intel_dp_mst_encoder *mst_encoder;
@@ -2922,16 +2839,14 @@ icl_get_dpll(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state,
 		if (intel_dig_port->tc_type == TC_PORT_TBT) {
 			min = DPLL_ID_ICL_TBTPLL;
 			max = min;
-			ret = icl_calc_dpll_state(crtc_state, encoder, clock,
-						  &pll_state);
+			ret = icl_calc_dpll_state(crtc_state, encoder);
 		} else {
 			enum tc_port tc_port;
 
 			tc_port = intel_port_to_tc(dev_priv, port);
 			min = icl_tc_port_to_pll_id(tc_port);
 			max = min;
-			ret = icl_calc_mg_pll_state(crtc_state, encoder, clock,
-						    &pll_state);
+			ret = icl_calc_mg_pll_state(crtc_state);
 		}
 	} else {
 		MISSING_CASE(port);
@@ -2943,9 +2858,8 @@ icl_get_dpll(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state,
 		return NULL;
 	}
 
-	crtc_state->dpll_hw_state = pll_state;
 
-	pll = intel_find_shared_dpll(crtc, crtc_state, min, max);
+	pll = intel_find_shared_dpll(crtc_state, min, max);
 	if (!pll) {
 		DRM_DEBUG_KMS("No PLL selected\n");
 		return NULL;
@@ -2956,19 +2870,72 @@ icl_get_dpll(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state,
 	return pll;
 }
 
-static i915_reg_t icl_pll_id_to_enable_reg(enum intel_dpll_id id)
+static bool mg_pll_get_hw_state(struct drm_i915_private *dev_priv,
+				struct intel_shared_dpll *pll,
+				struct intel_dpll_hw_state *hw_state)
 {
-	if (intel_dpll_is_combophy(id))
-		return CNL_DPLL_ENABLE(id);
-	else if (id == DPLL_ID_ICL_TBTPLL)
-		return TBT_PLL_ENABLE;
+	const enum intel_dpll_id id = pll->info->id;
+	enum tc_port tc_port = icl_pll_id_to_tc_port(id);
+	intel_wakeref_t wakeref;
+	bool ret = false;
+	u32 val;
+
+	wakeref = intel_display_power_get_if_enabled(dev_priv,
+						     POWER_DOMAIN_PLLS);
+	if (!wakeref)
+		return false;
 
-	return MG_PLL_ENABLE(icl_pll_id_to_tc_port(id));
+	val = I915_READ(MG_PLL_ENABLE(tc_port));
+	if (!(val & PLL_ENABLE))
+		goto out;
+
+	hw_state->mg_refclkin_ctl = I915_READ(MG_REFCLKIN_CTL(tc_port));
+	hw_state->mg_refclkin_ctl &= MG_REFCLKIN_CTL_OD_2_MUX_MASK;
+
+	hw_state->mg_clktop2_coreclkctl1 =
+		I915_READ(MG_CLKTOP2_CORECLKCTL1(tc_port));
+	hw_state->mg_clktop2_coreclkctl1 &=
+		MG_CLKTOP2_CORECLKCTL1_A_DIVRATIO_MASK;
+
+	hw_state->mg_clktop2_hsclkctl =
+		I915_READ(MG_CLKTOP2_HSCLKCTL(tc_port));
+	hw_state->mg_clktop2_hsclkctl &=
+		MG_CLKTOP2_HSCLKCTL_TLINEDRV_CLKSEL_MASK |
+		MG_CLKTOP2_HSCLKCTL_CORE_INPUTSEL_MASK |
+		MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_MASK |
+		MG_CLKTOP2_HSCLKCTL_DSDIV_RATIO_MASK;
+
+	hw_state->mg_pll_div0 = I915_READ(MG_PLL_DIV0(tc_port));
+	hw_state->mg_pll_div1 = I915_READ(MG_PLL_DIV1(tc_port));
+	hw_state->mg_pll_lf = I915_READ(MG_PLL_LF(tc_port));
+	hw_state->mg_pll_frac_lock = I915_READ(MG_PLL_FRAC_LOCK(tc_port));
+	hw_state->mg_pll_ssc = I915_READ(MG_PLL_SSC(tc_port));
+
+	hw_state->mg_pll_bias = I915_READ(MG_PLL_BIAS(tc_port));
+	hw_state->mg_pll_tdc_coldst_bias =
+		I915_READ(MG_PLL_TDC_COLDST_BIAS(tc_port));
+
+	if (dev_priv->cdclk.hw.ref == 38400) {
+		hw_state->mg_pll_tdc_coldst_bias_mask = MG_PLL_TDC_COLDST_COLDSTART;
+		hw_state->mg_pll_bias_mask = 0;
+	} else {
+		hw_state->mg_pll_tdc_coldst_bias_mask = -1U;
+		hw_state->mg_pll_bias_mask = -1U;
+	}
+
+	hw_state->mg_pll_tdc_coldst_bias &= hw_state->mg_pll_tdc_coldst_bias_mask;
+	hw_state->mg_pll_bias &= hw_state->mg_pll_bias_mask;
+
+	ret = true;
+out:
+	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS, wakeref);
+	return ret;
 }
 
 static bool icl_pll_get_hw_state(struct drm_i915_private *dev_priv,
 				 struct intel_shared_dpll *pll,
-				 struct intel_dpll_hw_state *hw_state)
+				 struct intel_dpll_hw_state *hw_state,
+				 i915_reg_t enable_reg)
 {
 	const enum intel_dpll_id id = pll->info->id;
 	intel_wakeref_t wakeref;
@@ -2980,54 +2947,12 @@ static bool icl_pll_get_hw_state(struct drm_i915_private *dev_priv,
 	if (!wakeref)
 		return false;
 
-	val = I915_READ(icl_pll_id_to_enable_reg(id));
+	val = I915_READ(enable_reg);
 	if (!(val & PLL_ENABLE))
 		goto out;
 
-	if (intel_dpll_is_combophy(id) ||
-	    id == DPLL_ID_ICL_TBTPLL) {
-		hw_state->cfgcr0 = I915_READ(ICL_DPLL_CFGCR0(id));
-		hw_state->cfgcr1 = I915_READ(ICL_DPLL_CFGCR1(id));
-	} else {
-		enum tc_port tc_port = icl_pll_id_to_tc_port(id);
-
-		hw_state->mg_refclkin_ctl = I915_READ(MG_REFCLKIN_CTL(tc_port));
-		hw_state->mg_refclkin_ctl &= MG_REFCLKIN_CTL_OD_2_MUX_MASK;
-
-		hw_state->mg_clktop2_coreclkctl1 =
-			I915_READ(MG_CLKTOP2_CORECLKCTL1(tc_port));
-		hw_state->mg_clktop2_coreclkctl1 &=
-			MG_CLKTOP2_CORECLKCTL1_A_DIVRATIO_MASK;
-
-		hw_state->mg_clktop2_hsclkctl =
-			I915_READ(MG_CLKTOP2_HSCLKCTL(tc_port));
-		hw_state->mg_clktop2_hsclkctl &=
-			MG_CLKTOP2_HSCLKCTL_TLINEDRV_CLKSEL_MASK |
-			MG_CLKTOP2_HSCLKCTL_CORE_INPUTSEL_MASK |
-			MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_MASK |
-			MG_CLKTOP2_HSCLKCTL_DSDIV_RATIO_MASK;
-
-		hw_state->mg_pll_div0 = I915_READ(MG_PLL_DIV0(tc_port));
-		hw_state->mg_pll_div1 = I915_READ(MG_PLL_DIV1(tc_port));
-		hw_state->mg_pll_lf = I915_READ(MG_PLL_LF(tc_port));
-		hw_state->mg_pll_frac_lock = I915_READ(MG_PLL_FRAC_LOCK(tc_port));
-		hw_state->mg_pll_ssc = I915_READ(MG_PLL_SSC(tc_port));
-
-		hw_state->mg_pll_bias = I915_READ(MG_PLL_BIAS(tc_port));
-		hw_state->mg_pll_tdc_coldst_bias =
-			I915_READ(MG_PLL_TDC_COLDST_BIAS(tc_port));
-
-		if (dev_priv->cdclk.hw.ref == 38400) {
-			hw_state->mg_pll_tdc_coldst_bias_mask = MG_PLL_TDC_COLDST_COLDSTART;
-			hw_state->mg_pll_bias_mask = 0;
-		} else {
-			hw_state->mg_pll_tdc_coldst_bias_mask = -1U;
-			hw_state->mg_pll_bias_mask = -1U;
-		}
-
-		hw_state->mg_pll_tdc_coldst_bias &= hw_state->mg_pll_tdc_coldst_bias_mask;
-		hw_state->mg_pll_bias &= hw_state->mg_pll_bias_mask;
-	}
+	hw_state->cfgcr0 = I915_READ(ICL_DPLL_CFGCR0(id));
+	hw_state->cfgcr1 = I915_READ(ICL_DPLL_CFGCR1(id));
 
 	ret = true;
 out:
@@ -3035,6 +2960,21 @@ out:
 	return ret;
 }
 
+static bool combo_pll_get_hw_state(struct drm_i915_private *dev_priv,
+				   struct intel_shared_dpll *pll,
+				   struct intel_dpll_hw_state *hw_state)
+{
+	return icl_pll_get_hw_state(dev_priv, pll, hw_state,
+				    CNL_DPLL_ENABLE(pll->info->id));
+}
+
+static bool tbt_pll_get_hw_state(struct drm_i915_private *dev_priv,
+				 struct intel_shared_dpll *pll,
+				 struct intel_dpll_hw_state *hw_state)
+{
+	return icl_pll_get_hw_state(dev_priv, pll, hw_state, TBT_PLL_ENABLE);
+}
+
 static void icl_dpll_write(struct drm_i915_private *dev_priv,
 			   struct intel_shared_dpll *pll)
 {
@@ -3096,11 +3036,10 @@ static void icl_mg_pll_write(struct drm_i915_private *dev_priv,
 	POSTING_READ(MG_PLL_TDC_COLDST_BIAS(tc_port));
 }
 
-static void icl_pll_enable(struct drm_i915_private *dev_priv,
-			   struct intel_shared_dpll *pll)
+static void icl_pll_power_enable(struct drm_i915_private *dev_priv,
+				 struct intel_shared_dpll *pll,
+				 i915_reg_t enable_reg)
 {
-	const enum intel_dpll_id id = pll->info->id;
-	i915_reg_t enable_reg = icl_pll_id_to_enable_reg(id);
 	u32 val;
 
 	val = I915_READ(enable_reg);
@@ -3111,37 +3050,90 @@ static void icl_pll_enable(struct drm_i915_private *dev_priv,
 	 * The spec says we need to "wait" but it also says it should be
 	 * immediate.
 	 */
-	if (intel_wait_for_register(dev_priv, enable_reg, PLL_POWER_STATE,
-				    PLL_POWER_STATE, 1))
-		DRM_ERROR("PLL %d Power not enabled\n", id);
+	if (intel_wait_for_register(&dev_priv->uncore, enable_reg,
+				    PLL_POWER_STATE, PLL_POWER_STATE, 1))
+		DRM_ERROR("PLL %d Power not enabled\n", pll->info->id);
+}
 
-	if (intel_dpll_is_combophy(id) || id == DPLL_ID_ICL_TBTPLL)
-		icl_dpll_write(dev_priv, pll);
-	else
-		icl_mg_pll_write(dev_priv, pll);
+static void icl_pll_enable(struct drm_i915_private *dev_priv,
+			   struct intel_shared_dpll *pll,
+			   i915_reg_t enable_reg)
+{
+	u32 val;
+
+	val = I915_READ(enable_reg);
+	val |= PLL_ENABLE;
+	I915_WRITE(enable_reg, val);
+
+	/* Timeout is actually 600us. */
+	if (intel_wait_for_register(&dev_priv->uncore, enable_reg,
+				    PLL_LOCK, PLL_LOCK, 1))
+		DRM_ERROR("PLL %d not locked\n", pll->info->id);
+}
+
+static void combo_pll_enable(struct drm_i915_private *dev_priv,
+			     struct intel_shared_dpll *pll)
+{
+	i915_reg_t enable_reg = CNL_DPLL_ENABLE(pll->info->id);
+
+	icl_pll_power_enable(dev_priv, pll, enable_reg);
+
+	icl_dpll_write(dev_priv, pll);
 
 	/*
 	 * DVFS pre sequence would be here, but in our driver the cdclk code
 	 * paths should already be setting the appropriate voltage, hence we do
-	 * nothign here.
+	 * nothing here.
 	 */
 
-	val = I915_READ(enable_reg);
-	val |= PLL_ENABLE;
-	I915_WRITE(enable_reg, val);
+	icl_pll_enable(dev_priv, pll, enable_reg);
 
-	if (intel_wait_for_register(dev_priv, enable_reg, PLL_LOCK, PLL_LOCK,
-				    1)) /* 600us actually. */
-		DRM_ERROR("PLL %d not locked\n", id);
+	/* DVFS post sequence would be here. See the comment above. */
+}
+
+static void tbt_pll_enable(struct drm_i915_private *dev_priv,
+			   struct intel_shared_dpll *pll)
+{
+	icl_pll_power_enable(dev_priv, pll, TBT_PLL_ENABLE);
+
+	icl_dpll_write(dev_priv, pll);
+
+	/*
+	 * DVFS pre sequence would be here, but in our driver the cdclk code
+	 * paths should already be setting the appropriate voltage, hence we do
+	 * nothing here.
+	 */
+
+	icl_pll_enable(dev_priv, pll, TBT_PLL_ENABLE);
+
+	/* DVFS post sequence would be here. See the comment above. */
+}
+
+static void mg_pll_enable(struct drm_i915_private *dev_priv,
+			  struct intel_shared_dpll *pll)
+{
+	i915_reg_t enable_reg =
+		MG_PLL_ENABLE(icl_pll_id_to_tc_port(pll->info->id));
+
+	icl_pll_power_enable(dev_priv, pll, enable_reg);
+
+	icl_mg_pll_write(dev_priv, pll);
+
+	/*
+	 * DVFS pre sequence would be here, but in our driver the cdclk code
+	 * paths should already be setting the appropriate voltage, hence we do
+	 * nothing here.
+	 */
+
+	icl_pll_enable(dev_priv, pll, enable_reg);
 
 	/* DVFS post sequence would be here. See the comment above. */
 }
 
 static void icl_pll_disable(struct drm_i915_private *dev_priv,
-			    struct intel_shared_dpll *pll)
+			    struct intel_shared_dpll *pll,
+			    i915_reg_t enable_reg)
 {
-	const enum intel_dpll_id id = pll->info->id;
-	i915_reg_t enable_reg = icl_pll_id_to_enable_reg(id);
 	u32 val;
 
 	/* The first steps are done by intel_ddi_post_disable(). */
@@ -3157,8 +3149,9 @@ static void icl_pll_disable(struct drm_i915_private *dev_priv,
 	I915_WRITE(enable_reg, val);
 
 	/* Timeout is actually 1us. */
-	if (intel_wait_for_register(dev_priv, enable_reg, PLL_LOCK, 0, 1))
-		DRM_ERROR("PLL %d locked\n", id);
+	if (intel_wait_for_register(&dev_priv->uncore,
+				    enable_reg, PLL_LOCK, 0, 1))
+		DRM_ERROR("PLL %d locked\n", pll->info->id);
 
 	/* DVFS post sequence would be here. See the comment above. */
 
@@ -3170,9 +3163,30 @@ static void icl_pll_disable(struct drm_i915_private *dev_priv,
 	 * The spec says we need to "wait" but it also says it should be
 	 * immediate.
 	 */
-	if (intel_wait_for_register(dev_priv, enable_reg, PLL_POWER_STATE, 0,
-				    1))
-		DRM_ERROR("PLL %d Power not disabled\n", id);
+	if (intel_wait_for_register(&dev_priv->uncore,
+				    enable_reg, PLL_POWER_STATE, 0, 1))
+		DRM_ERROR("PLL %d Power not disabled\n", pll->info->id);
+}
+
+static void combo_pll_disable(struct drm_i915_private *dev_priv,
+			      struct intel_shared_dpll *pll)
+{
+	icl_pll_disable(dev_priv, pll, CNL_DPLL_ENABLE(pll->info->id));
+}
+
+static void tbt_pll_disable(struct drm_i915_private *dev_priv,
+			    struct intel_shared_dpll *pll)
+{
+	icl_pll_disable(dev_priv, pll, TBT_PLL_ENABLE);
+}
+
+static void mg_pll_disable(struct drm_i915_private *dev_priv,
+			   struct intel_shared_dpll *pll)
+{
+	i915_reg_t enable_reg =
+		MG_PLL_ENABLE(icl_pll_id_to_tc_port(pll->info->id));
+
+	icl_pll_disable(dev_priv, pll, enable_reg);
 }
 
 static void icl_dump_hw_state(struct drm_i915_private *dev_priv,
@@ -3197,20 +3211,32 @@ static void icl_dump_hw_state(struct drm_i915_private *dev_priv,
 		      hw_state->mg_pll_tdc_coldst_bias);
 }
 
-static const struct intel_shared_dpll_funcs icl_pll_funcs = {
-	.enable = icl_pll_enable,
-	.disable = icl_pll_disable,
-	.get_hw_state = icl_pll_get_hw_state,
+static const struct intel_shared_dpll_funcs combo_pll_funcs = {
+	.enable = combo_pll_enable,
+	.disable = combo_pll_disable,
+	.get_hw_state = combo_pll_get_hw_state,
+};
+
+static const struct intel_shared_dpll_funcs tbt_pll_funcs = {
+	.enable = tbt_pll_enable,
+	.disable = tbt_pll_disable,
+	.get_hw_state = tbt_pll_get_hw_state,
+};
+
+static const struct intel_shared_dpll_funcs mg_pll_funcs = {
+	.enable = mg_pll_enable,
+	.disable = mg_pll_disable,
+	.get_hw_state = mg_pll_get_hw_state,
 };
 
 static const struct dpll_info icl_plls[] = {
-	{ "DPLL 0",   &icl_pll_funcs, DPLL_ID_ICL_DPLL0,  0 },
-	{ "DPLL 1",   &icl_pll_funcs, DPLL_ID_ICL_DPLL1,  0 },
-	{ "TBT PLL",  &icl_pll_funcs, DPLL_ID_ICL_TBTPLL, 0 },
-	{ "MG PLL 1", &icl_pll_funcs, DPLL_ID_ICL_MGPLL1, 0 },
-	{ "MG PLL 2", &icl_pll_funcs, DPLL_ID_ICL_MGPLL2, 0 },
-	{ "MG PLL 3", &icl_pll_funcs, DPLL_ID_ICL_MGPLL3, 0 },
-	{ "MG PLL 4", &icl_pll_funcs, DPLL_ID_ICL_MGPLL4, 0 },
+	{ "DPLL 0",   &combo_pll_funcs, DPLL_ID_ICL_DPLL0,  0 },
+	{ "DPLL 1",   &combo_pll_funcs, DPLL_ID_ICL_DPLL1,  0 },
+	{ "TBT PLL",  &tbt_pll_funcs, DPLL_ID_ICL_TBTPLL, 0 },
+	{ "MG PLL 1", &mg_pll_funcs, DPLL_ID_ICL_MGPLL1, 0 },
+	{ "MG PLL 2", &mg_pll_funcs, DPLL_ID_ICL_MGPLL2, 0 },
+	{ "MG PLL 3", &mg_pll_funcs, DPLL_ID_ICL_MGPLL3, 0 },
+	{ "MG PLL 4", &mg_pll_funcs, DPLL_ID_ICL_MGPLL4, 0 },
 	{ },
 };
 
@@ -3220,6 +3246,18 @@ static const struct intel_dpll_mgr icl_pll_mgr = {
 	.dump_hw_state = icl_dump_hw_state,
 };
 
+static const struct dpll_info ehl_plls[] = {
+	{ "DPLL 0", &combo_pll_funcs, DPLL_ID_ICL_DPLL0, 0 },
+	{ "DPLL 1", &combo_pll_funcs, DPLL_ID_ICL_DPLL1, 0 },
+	{ },
+};
+
+static const struct intel_dpll_mgr ehl_pll_mgr = {
+	.dpll_info = ehl_plls,
+	.get_dpll = icl_get_dpll,
+	.dump_hw_state = icl_dump_hw_state,
+};
+
 /**
  * intel_shared_dpll_init - Initialize shared DPLLs
  * @dev: drm device
@@ -3233,7 +3271,9 @@ void intel_shared_dpll_init(struct drm_device *dev)
 	const struct dpll_info *dpll_info;
 	int i;
 
-	if (IS_ICELAKE(dev_priv))
+	if (IS_ELKHARTLAKE(dev_priv))
+		dpll_mgr = &ehl_pll_mgr;
+	else if (INTEL_GEN(dev_priv) >= 11)
 		dpll_mgr = &icl_pll_mgr;
 	else if (IS_CANNONLAKE(dev_priv))
 		dpll_mgr = &cnl_pll_mgr;
@@ -3271,31 +3311,29 @@ void intel_shared_dpll_init(struct drm_device *dev)
 
 /**
  * intel_get_shared_dpll - get a shared DPLL for CRTC and encoder combination
- * @crtc: CRTC
- * @crtc_state: atomic state for @crtc
+ * @crtc_state: atomic state for the crtc
  * @encoder: encoder
  *
  * Find an appropriate DPLL for the given CRTC and encoder combination. A
- * reference from the @crtc to the returned pll is registered in the atomic
- * state. That configuration is made effective by calling
+ * reference from the @crtc_state to the returned pll is registered in the
+ * atomic state. That configuration is made effective by calling
  * intel_shared_dpll_swap_state(). The reference should be released by calling
  * intel_release_shared_dpll().
  *
  * Returns:
- * A shared DPLL to be used by @crtc and @encoder with the given @crtc_state.
+ * A shared DPLL to be used by @crtc_state and @encoder.
  */
 struct intel_shared_dpll *
-intel_get_shared_dpll(struct intel_crtc *crtc,
-		      struct intel_crtc_state *crtc_state,
+intel_get_shared_dpll(struct intel_crtc_state *crtc_state,
 		      struct intel_encoder *encoder)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct drm_i915_private *dev_priv = to_i915(crtc_state->base.crtc->dev);
 	const struct intel_dpll_mgr *dpll_mgr = dev_priv->dpll_mgr;
 
 	if (WARN_ON(!dpll_mgr))
 		return NULL;
 
-	return dpll_mgr->get_dpll(crtc, crtc_state, encoder);
+	return dpll_mgr->get_dpll(crtc_state, encoder);
 }
 
 /**
diff --git a/drivers/gpu/drm/i915/intel_dpll_mgr.h b/drivers/gpu/drm/i915/intel_dpll_mgr.h
index 40e8391a92f2..bd8124cc81ed 100644
--- a/drivers/gpu/drm/i915/intel_dpll_mgr.h
+++ b/drivers/gpu/drm/i915/intel_dpll_mgr.h
@@ -327,8 +327,7 @@ void assert_shared_dpll(struct drm_i915_private *dev_priv,
 			bool state);
 #define assert_shared_dpll_enabled(d, p) assert_shared_dpll(d, p, true)
 #define assert_shared_dpll_disabled(d, p) assert_shared_dpll(d, p, false)
-struct intel_shared_dpll *intel_get_shared_dpll(struct intel_crtc *crtc,
-						struct intel_crtc_state *state,
+struct intel_shared_dpll *intel_get_shared_dpll(struct intel_crtc_state *state,
 						struct intel_encoder *encoder);
 void intel_release_shared_dpll(struct intel_shared_dpll *dpll,
 			       struct intel_crtc *crtc,
@@ -341,8 +340,6 @@ void intel_shared_dpll_init(struct drm_device *dev);
 
 void intel_dpll_dump_hw_state(struct drm_i915_private *dev_priv,
 			      struct intel_dpll_hw_state *hw_state);
-int icl_calc_dp_combo_pll_link(struct drm_i915_private *dev_priv,
-			       u32 pll_id);
 int cnl_hdmi_pll_ref_clock(struct drm_i915_private *dev_priv);
 enum intel_dpll_id icl_tc_port_to_pll_id(enum tc_port tc_port);
 bool intel_dpll_is_combophy(enum intel_dpll_id id);
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index d5660ac1b0d6..a38b9cff5cd0 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -27,22 +27,24 @@
 
 #include <linux/async.h>
 #include <linux/i2c.h>
-#include <linux/hdmi.h>
 #include <linux/sched/clock.h>
 #include <linux/stackdepot.h>
-#include <drm/i915_drm.h>
-#include "i915_drv.h"
+
+#include <drm/drm_atomic.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_encoder.h>
-#include <drm/drm_fb_helper.h>
 #include <drm/drm_dp_dual_mode_helper.h>
 #include <drm/drm_dp_mst_helper.h>
+#include <drm/drm_encoder.h>
+#include <drm/drm_fb_helper.h>
 #include <drm/drm_probe_helper.h>
 #include <drm/drm_rect.h>
 #include <drm/drm_vblank.h>
-#include <drm/drm_atomic.h>
+#include <drm/i915_drm.h>
+#include <drm/i915_mei_hdcp_interface.h>
 #include <media/cec-notifier.h>
 
+#include "i915_drv.h"
+
 struct drm_printer;
 
 /**
@@ -325,6 +327,13 @@ struct intel_panel {
 
 struct intel_digital_port;
 
+enum check_link_response {
+	HDCP_LINK_PROTECTED	= 0,
+	HDCP_TOPOLOGY_CHANGE,
+	HDCP_LINK_INTEGRITY_FAILURE,
+	HDCP_REAUTH_REQUEST
+};
+
 /*
  * This structure serves as a translation layer between the generic HDCP code
  * and the bus-specific code. What that means is that HDCP over HDMI differs
@@ -397,6 +406,32 @@ struct intel_hdcp_shim {
 	/* Detects panel's hdcp capability. This is optional for HDMI. */
 	int (*hdcp_capable)(struct intel_digital_port *intel_dig_port,
 			    bool *hdcp_capable);
+
+	/* HDCP adaptation(DP/HDMI) required on the port */
+	enum hdcp_wired_protocol protocol;
+
+	/* Detects whether sink is HDCP2.2 capable */
+	int (*hdcp_2_2_capable)(struct intel_digital_port *intel_dig_port,
+				bool *capable);
+
+	/* Write HDCP2.2 messages */
+	int (*write_2_2_msg)(struct intel_digital_port *intel_dig_port,
+			     void *buf, size_t size);
+
+	/* Read HDCP2.2 messages */
+	int (*read_2_2_msg)(struct intel_digital_port *intel_dig_port,
+			    u8 msg_id, void *buf, size_t size);
+
+	/*
+	 * Implementation of DP HDCP2.2 Errata for the communication of stream
+	 * type to Receivers. In DP HDCP2.2 Stream type is one of the input to
+	 * the HDCP2.2 Cipher for En/De-Cryption. Not applicable for HDMI.
+	 */
+	int (*config_stream_type)(struct intel_digital_port *intel_dig_port,
+				  bool is_repeater, u8 type);
+
+	/* HDCP2.2 Link Integrity Check */
+	int (*check_2_2_link)(struct intel_digital_port *intel_dig_port);
 };
 
 struct intel_hdcp {
@@ -406,6 +441,50 @@ struct intel_hdcp {
 	u64 value;
 	struct delayed_work check_work;
 	struct work_struct prop_work;
+
+	/* HDCP1.4 Encryption status */
+	bool hdcp_encrypted;
+
+	/* HDCP2.2 related definitions */
+	/* Flag indicates whether this connector supports HDCP2.2 or not. */
+	bool hdcp2_supported;
+
+	/* HDCP2.2 Encryption status */
+	bool hdcp2_encrypted;
+
+	/*
+	 * Content Stream Type defined by content owner. TYPE0(0x0) content can
+	 * flow in the link protected by HDCP2.2 or HDCP1.4, where as TYPE1(0x1)
+	 * content can flow only through a link protected by HDCP2.2.
+	 */
+	u8 content_type;
+	struct hdcp_port_data port_data;
+
+	bool is_paired;
+	bool is_repeater;
+
+	/*
+	 * Count of ReceiverID_List received. Initialized to 0 at AKE_INIT.
+	 * Incremented after processing the RepeaterAuth_Send_ReceiverID_List.
+	 * When it rolls over re-auth has to be triggered.
+	 */
+	u32 seq_num_v;
+
+	/*
+	 * Count of RepeaterAuth_Stream_Manage msg propagated.
+	 * Initialized to 0 on AKE_INIT. Incremented after every successful
+	 * transmission of RepeaterAuth_Stream_Manage message. When it rolls
+	 * over re-Auth has to be triggered.
+	 */
+	u32 seq_num_m;
+
+	/*
+	 * Work queue to signal the CP_IRQ. Used for the waiters to read the
+	 * available information from HDCP DP sink.
+	 */
+	wait_queue_head_t cp_irq_queue;
+	atomic_t cp_irq_count;
+	int cp_irq_count_cached;
 };
 
 struct intel_connector {
@@ -480,6 +559,11 @@ struct intel_atomic_state {
 		 * state only when all crtc's are DPMS off.
 		 */
 		struct intel_cdclk_state actual;
+
+		int force_min_cdclk;
+		bool force_min_cdclk_changed;
+		/* pipe to which cd2x update is synchronized */
+		enum pipe pipe;
 	} cdclk;
 
 	bool dpll_set, modeset;
@@ -923,7 +1007,8 @@ struct intel_crtc_state {
 	struct intel_link_m_n fdi_m_n;
 
 	bool ips_enabled;
-	bool ips_force_disable;
+
+	bool crc_enabled;
 
 	bool enable_fbc;
 
@@ -944,13 +1029,30 @@ struct intel_crtc_state {
 	/* Gamma mode programmed on the pipe */
 	u32 gamma_mode;
 
+	union {
+		/* CSC mode programmed on the pipe */
+		u32 csc_mode;
+
+		/* CHV CGM mode */
+		u32 cgm_mode;
+	};
+
 	/* bitmask of visible planes (enum plane_id) */
 	u8 active_planes;
 	u8 nv12_planes;
+	u8 c8_planes;
 
 	/* bitmask of planes that will be updated during the commit */
 	u8 update_planes;
 
+	struct {
+		u32 enable;
+		u32 gcp;
+		union hdmi_infoframe avi;
+		union hdmi_infoframe spd;
+		union hdmi_infoframe hdmi;
+	} infoframes;
+
 	/* HDMI scrambling status */
 	bool hdmi_scrambling;
 
@@ -963,6 +1065,12 @@ struct intel_crtc_state {
 	/* Output down scaling is done in LSPCON device */
 	bool lspcon_downsampling;
 
+	/* enable pipe gamma? */
+	bool gamma_enable;
+
+	/* enable pipe csc? */
+	bool csc_enable;
+
 	/* Display Stream compression state */
 	struct {
 		bool compression_enable;
@@ -991,9 +1099,6 @@ struct intel_crtc {
 
 	struct intel_crtc_state *config;
 
-	/* global reset count when the last flip was submitted */
-	unsigned int reset_count;
-
 	/* Access to these should be protected by dev_priv->irq_lock. */
 	bool cpu_fifo_underrun_disabled;
 	bool pch_fifo_underrun_disabled;
@@ -1262,11 +1367,15 @@ struct intel_digital_port {
 				const struct intel_crtc_state *crtc_state,
 				unsigned int type,
 				const void *frame, ssize_t len);
+	void (*read_infoframe)(struct intel_encoder *encoder,
+			       const struct intel_crtc_state *crtc_state,
+			       unsigned int type,
+			       void *frame, ssize_t len);
 	void (*set_infoframes)(struct intel_encoder *encoder,
 			       bool enable,
 			       const struct intel_crtc_state *crtc_state,
 			       const struct drm_connector_state *conn_state);
-	bool (*infoframe_enabled)(struct intel_encoder *encoder,
+	u32 (*infoframes_enabled)(struct intel_encoder *encoder,
 				  const struct intel_crtc_state *pipe_config);
 };
 
@@ -1494,6 +1603,7 @@ void gen11_reset_rps_interrupts(struct drm_i915_private *dev_priv);
 void gen6_reset_rps_interrupts(struct drm_i915_private *dev_priv);
 void gen6_enable_rps_interrupts(struct drm_i915_private *dev_priv);
 void gen6_disable_rps_interrupts(struct drm_i915_private *dev_priv);
+void gen6_rps_reset_ei(struct drm_i915_private *dev_priv);
 
 static inline u32 gen6_sanitize_rps_pm_mask(const struct drm_i915_private *i915,
 					    u32 mask)
@@ -1521,85 +1631,8 @@ void gen9_reset_guc_interrupts(struct drm_i915_private *dev_priv);
 void gen9_enable_guc_interrupts(struct drm_i915_private *dev_priv);
 void gen9_disable_guc_interrupts(struct drm_i915_private *dev_priv);
 
-/* intel_crt.c */
-bool intel_crt_port_enabled(struct drm_i915_private *dev_priv,
-			    i915_reg_t adpa_reg, enum pipe *pipe);
-void intel_crt_init(struct drm_i915_private *dev_priv);
-void intel_crt_reset(struct drm_encoder *encoder);
-
-/* intel_ddi.c */
-void intel_ddi_fdi_post_disable(struct intel_encoder *intel_encoder,
-				const struct intel_crtc_state *old_crtc_state,
-				const struct drm_connector_state *old_conn_state);
-void hsw_fdi_link_train(struct intel_crtc *crtc,
-			const struct intel_crtc_state *crtc_state);
-void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port);
-bool intel_ddi_get_hw_state(struct intel_encoder *encoder, enum pipe *pipe);
-void intel_ddi_enable_transcoder_func(const struct intel_crtc_state *crtc_state);
-void intel_ddi_disable_transcoder_func(const struct intel_crtc_state *crtc_state);
-void intel_ddi_enable_pipe_clock(const struct intel_crtc_state *crtc_state);
-void intel_ddi_disable_pipe_clock(const  struct intel_crtc_state *crtc_state);
-void intel_ddi_set_pipe_settings(const struct intel_crtc_state *crtc_state);
-void intel_ddi_prepare_link_retrain(struct intel_dp *intel_dp);
-bool intel_ddi_connector_get_hw_state(struct intel_connector *intel_connector);
-void intel_ddi_get_config(struct intel_encoder *encoder,
-			  struct intel_crtc_state *pipe_config);
-
-void intel_ddi_set_vc_payload_alloc(const struct intel_crtc_state *crtc_state,
-				    bool state);
-void intel_ddi_compute_min_voltage_level(struct drm_i915_private *dev_priv,
-					 struct intel_crtc_state *crtc_state);
-u32 bxt_signal_levels(struct intel_dp *intel_dp);
-u32 ddi_signal_levels(struct intel_dp *intel_dp);
-u8 intel_ddi_dp_voltage_max(struct intel_encoder *encoder);
-u8 intel_ddi_dp_pre_emphasis_max(struct intel_encoder *encoder,
-				 u8 voltage_swing);
-int intel_ddi_toggle_hdcp_signalling(struct intel_encoder *intel_encoder,
-				     bool enable);
-void icl_sanitize_encoder_pll_mapping(struct intel_encoder *encoder);
-int cnl_calc_wrpll_link(struct drm_i915_private *dev_priv,
-			enum intel_dpll_id pll_id);
-
-unsigned int intel_fb_align_height(const struct drm_framebuffer *fb,
-				   int color_plane, unsigned int height);
-
-/* intel_audio.c */
-void intel_init_audio_hooks(struct drm_i915_private *dev_priv);
-void intel_audio_codec_enable(struct intel_encoder *encoder,
-			      const struct intel_crtc_state *crtc_state,
-			      const struct drm_connector_state *conn_state);
-void intel_audio_codec_disable(struct intel_encoder *encoder,
-			       const struct intel_crtc_state *old_crtc_state,
-			       const struct drm_connector_state *old_conn_state);
-void i915_audio_component_init(struct drm_i915_private *dev_priv);
-void i915_audio_component_cleanup(struct drm_i915_private *dev_priv);
-void intel_audio_init(struct drm_i915_private *dev_priv);
-void intel_audio_deinit(struct drm_i915_private *dev_priv);
-
-/* intel_cdclk.c */
-int intel_crtc_compute_min_cdclk(const struct intel_crtc_state *crtc_state);
-void skl_init_cdclk(struct drm_i915_private *dev_priv);
-void skl_uninit_cdclk(struct drm_i915_private *dev_priv);
-void cnl_init_cdclk(struct drm_i915_private *dev_priv);
-void cnl_uninit_cdclk(struct drm_i915_private *dev_priv);
-void bxt_init_cdclk(struct drm_i915_private *dev_priv);
-void bxt_uninit_cdclk(struct drm_i915_private *dev_priv);
-void icl_init_cdclk(struct drm_i915_private *dev_priv);
-void icl_uninit_cdclk(struct drm_i915_private *dev_priv);
-void intel_init_cdclk_hooks(struct drm_i915_private *dev_priv);
-void intel_update_max_cdclk(struct drm_i915_private *dev_priv);
-void intel_update_cdclk(struct drm_i915_private *dev_priv);
-void intel_update_rawclk(struct drm_i915_private *dev_priv);
-bool intel_cdclk_needs_modeset(const struct intel_cdclk_state *a,
-			       const struct intel_cdclk_state *b);
-bool intel_cdclk_changed(const struct intel_cdclk_state *a,
-			 const struct intel_cdclk_state *b);
-void intel_set_cdclk(struct drm_i915_private *dev_priv,
-		     const struct intel_cdclk_state *cdclk_state);
-void intel_dump_cdclk_state(const struct intel_cdclk_state *cdclk_state,
-			    const char *context);
-
 /* intel_display.c */
+void intel_plane_destroy(struct drm_plane *plane);
 void i830_enable_pipe(struct drm_i915_private *dev_priv, enum pipe pipe);
 void i830_disable_pipe(struct drm_i915_private *dev_priv, enum pipe pipe);
 enum pipe intel_crtc_pch_transcoder(struct intel_crtc *crtc);
@@ -1614,6 +1647,8 @@ void intel_init_display_hooks(struct drm_i915_private *dev_priv);
 unsigned int intel_fb_xy_to_linear(int x, int y,
 				   const struct intel_plane_state *state,
 				   int plane);
+unsigned int intel_fb_align_height(const struct drm_framebuffer *fb,
+				   int color_plane, unsigned int height);
 void intel_add_fb_offsets(int *x, int *y,
 			  const struct intel_plane_state *state, int plane);
 unsigned int intel_rotation_info_size(const struct intel_rotation_info *rot_info);
@@ -1740,7 +1775,7 @@ void intel_dp_get_m_n(struct intel_crtc *crtc,
 void intel_dp_set_m_n(const struct intel_crtc_state *crtc_state,
 		      enum link_m_n_set m_n);
 int intel_dotclock_calculate(int link_freq, const struct intel_link_m_n *m_n);
-bool bxt_find_best_dpll(struct intel_crtc_state *crtc_state, int target_clock,
+bool bxt_find_best_dpll(struct intel_crtc_state *crtc_state,
 			struct dpll *best_clock);
 int chv_calc_dpll_params(int refclk, struct dpll *pll_clock);
 
@@ -1781,106 +1816,9 @@ unsigned int i9xx_plane_max_stride(struct intel_plane *plane,
 				   u32 pixel_format, u64 modifier,
 				   unsigned int rotation);
 
-/* intel_connector.c */
-int intel_connector_init(struct intel_connector *connector);
-struct intel_connector *intel_connector_alloc(void);
-void intel_connector_free(struct intel_connector *connector);
-void intel_connector_destroy(struct drm_connector *connector);
-int intel_connector_register(struct drm_connector *connector);
-void intel_connector_unregister(struct drm_connector *connector);
-void intel_connector_attach_encoder(struct intel_connector *connector,
-				    struct intel_encoder *encoder);
-bool intel_connector_get_hw_state(struct intel_connector *connector);
-enum pipe intel_connector_get_pipe(struct intel_connector *connector);
-int intel_connector_update_modes(struct drm_connector *connector,
-				 struct edid *edid);
-int intel_ddc_get_modes(struct drm_connector *c, struct i2c_adapter *adapter);
-void intel_attach_force_audio_property(struct drm_connector *connector);
-void intel_attach_broadcast_rgb_property(struct drm_connector *connector);
-void intel_attach_aspect_ratio_property(struct drm_connector *connector);
-
-/* intel_csr.c */
-void intel_csr_ucode_init(struct drm_i915_private *);
-void intel_csr_load_program(struct drm_i915_private *);
-void intel_csr_ucode_fini(struct drm_i915_private *);
-void intel_csr_ucode_suspend(struct drm_i915_private *);
-void intel_csr_ucode_resume(struct drm_i915_private *);
-
-/* intel_dp.c */
-bool intel_dp_port_enabled(struct drm_i915_private *dev_priv,
-			   i915_reg_t dp_reg, enum port port,
-			   enum pipe *pipe);
-bool intel_dp_init(struct drm_i915_private *dev_priv, i915_reg_t output_reg,
-		   enum port port);
-bool intel_dp_init_connector(struct intel_digital_port *intel_dig_port,
-			     struct intel_connector *intel_connector);
-void intel_dp_set_link_params(struct intel_dp *intel_dp,
-			      int link_rate, u8 lane_count,
-			      bool link_mst);
-int intel_dp_get_link_train_fallback_values(struct intel_dp *intel_dp,
-					    int link_rate, u8 lane_count);
+/* intel_dp_link_training.c */
 void intel_dp_start_link_train(struct intel_dp *intel_dp);
 void intel_dp_stop_link_train(struct intel_dp *intel_dp);
-int intel_dp_retrain_link(struct intel_encoder *encoder,
-			  struct drm_modeset_acquire_ctx *ctx);
-void intel_dp_sink_dpms(struct intel_dp *intel_dp, int mode);
-void intel_dp_sink_set_decompression_state(struct intel_dp *intel_dp,
-					   const struct intel_crtc_state *crtc_state,
-					   bool enable);
-void intel_dp_encoder_reset(struct drm_encoder *encoder);
-void intel_dp_encoder_suspend(struct intel_encoder *intel_encoder);
-void intel_dp_encoder_flush_work(struct drm_encoder *encoder);
-int intel_dp_compute_config(struct intel_encoder *encoder,
-			    struct intel_crtc_state *pipe_config,
-			    struct drm_connector_state *conn_state);
-bool intel_dp_is_edp(struct intel_dp *intel_dp);
-bool intel_dp_is_port_edp(struct drm_i915_private *dev_priv, enum port port);
-enum irqreturn intel_dp_hpd_pulse(struct intel_digital_port *intel_dig_port,
-				  bool long_hpd);
-void intel_edp_backlight_on(const struct intel_crtc_state *crtc_state,
-			    const struct drm_connector_state *conn_state);
-void intel_edp_backlight_off(const struct drm_connector_state *conn_state);
-void intel_edp_panel_vdd_on(struct intel_dp *intel_dp);
-void intel_edp_panel_on(struct intel_dp *intel_dp);
-void intel_edp_panel_off(struct intel_dp *intel_dp);
-void intel_dp_mst_suspend(struct drm_i915_private *dev_priv);
-void intel_dp_mst_resume(struct drm_i915_private *dev_priv);
-int intel_dp_max_link_rate(struct intel_dp *intel_dp);
-int intel_dp_max_lane_count(struct intel_dp *intel_dp);
-int intel_dp_rate_select(struct intel_dp *intel_dp, int rate);
-void intel_dp_hot_plug(struct intel_encoder *intel_encoder);
-void intel_power_sequencer_reset(struct drm_i915_private *dev_priv);
-u32 intel_dp_pack_aux(const u8 *src, int src_bytes);
-void intel_plane_destroy(struct drm_plane *plane);
-void intel_edp_drrs_enable(struct intel_dp *intel_dp,
-			   const struct intel_crtc_state *crtc_state);
-void intel_edp_drrs_disable(struct intel_dp *intel_dp,
-			    const struct intel_crtc_state *crtc_state);
-void intel_edp_drrs_invalidate(struct drm_i915_private *dev_priv,
-			       unsigned int frontbuffer_bits);
-void intel_edp_drrs_flush(struct drm_i915_private *dev_priv,
-			  unsigned int frontbuffer_bits);
-
-void
-intel_dp_program_link_training_pattern(struct intel_dp *intel_dp,
-				       u8 dp_train_pat);
-void
-intel_dp_set_signal_levels(struct intel_dp *intel_dp);
-void intel_dp_set_idle_link_train(struct intel_dp *intel_dp);
-u8
-intel_dp_voltage_max(struct intel_dp *intel_dp);
-u8
-intel_dp_pre_emphasis_max(struct intel_dp *intel_dp, u8 voltage_swing);
-void intel_dp_compute_rate(struct intel_dp *intel_dp, int port_clock,
-			   u8 *link_bw, u8 *rate_select);
-bool intel_dp_source_supports_hbr2(struct intel_dp *intel_dp);
-bool intel_dp_source_supports_hbr3(struct intel_dp *intel_dp);
-bool
-intel_dp_get_link_status(struct intel_dp *intel_dp, u8 link_status[DP_LINK_STATUS_SIZE]);
-u16 intel_dp_dsc_get_output_bpp(int link_clock, u8 lane_count,
-				int mode_clock, int mode_hdisplay);
-u8 intel_dp_dsc_get_slice_count(struct intel_dp *intel_dp, int mode_clock,
-				int mode_hdisplay);
 
 /* intel_vdsc.c */
 int intel_dp_compute_dsc_params(struct intel_dp *intel_dp,
@@ -1888,18 +1826,6 @@ int intel_dp_compute_dsc_params(struct intel_dp *intel_dp,
 enum intel_display_power_domain
 intel_dsc_power_domain(const struct intel_crtc_state *crtc_state);
 
-static inline unsigned int intel_dp_unused_lane_mask(int lane_count)
-{
-	return ~((1 << lane_count) - 1) & 0xf;
-}
-
-bool intel_dp_read_dpcd(struct intel_dp *intel_dp);
-int intel_dp_link_required(int pixel_clock, int bpp);
-int intel_dp_max_data_rate(int max_link_clock, int max_lanes);
-bool intel_digital_port_connected(struct intel_encoder *encoder);
-void icl_tc_phy_disconnect(struct drm_i915_private *dev_priv,
-			   struct intel_digital_port *dig_port);
-
 /* intel_dp_aux_backlight.c */
 int intel_dp_aux_init_backlight_funcs(struct intel_connector *intel_connector);
 
@@ -1915,100 +1841,11 @@ void icl_dsi_init(struct drm_i915_private *dev_priv);
 /* intel_dsi_dcs_backlight.c */
 int intel_dsi_dcs_init_backlight_funcs(struct intel_connector *intel_connector);
 
-/* intel_dvo.c */
-void intel_dvo_init(struct drm_i915_private *dev_priv);
 /* intel_hotplug.c */
 void intel_hpd_poll_init(struct drm_i915_private *dev_priv);
 bool intel_encoder_hotplug(struct intel_encoder *encoder,
 			   struct intel_connector *connector);
 
-/* legacy fbdev emulation in intel_fbdev.c */
-#ifdef CONFIG_DRM_FBDEV_EMULATION
-extern int intel_fbdev_init(struct drm_device *dev);
-extern void intel_fbdev_initial_config_async(struct drm_device *dev);
-extern void intel_fbdev_unregister(struct drm_i915_private *dev_priv);
-extern void intel_fbdev_fini(struct drm_i915_private *dev_priv);
-extern void intel_fbdev_set_suspend(struct drm_device *dev, int state, bool synchronous);
-extern void intel_fbdev_output_poll_changed(struct drm_device *dev);
-extern void intel_fbdev_restore_mode(struct drm_device *dev);
-#else
-static inline int intel_fbdev_init(struct drm_device *dev)
-{
-	return 0;
-}
-
-static inline void intel_fbdev_initial_config_async(struct drm_device *dev)
-{
-}
-
-static inline void intel_fbdev_unregister(struct drm_i915_private *dev_priv)
-{
-}
-
-static inline void intel_fbdev_fini(struct drm_i915_private *dev_priv)
-{
-}
-
-static inline void intel_fbdev_set_suspend(struct drm_device *dev, int state, bool synchronous)
-{
-}
-
-static inline void intel_fbdev_output_poll_changed(struct drm_device *dev)
-{
-}
-
-static inline void intel_fbdev_restore_mode(struct drm_device *dev)
-{
-}
-#endif
-
-/* intel_fbc.c */
-void intel_fbc_choose_crtc(struct drm_i915_private *dev_priv,
-			   struct intel_atomic_state *state);
-bool intel_fbc_is_active(struct drm_i915_private *dev_priv);
-void intel_fbc_pre_update(struct intel_crtc *crtc,
-			  struct intel_crtc_state *crtc_state,
-			  struct intel_plane_state *plane_state);
-void intel_fbc_post_update(struct intel_crtc *crtc);
-void intel_fbc_init(struct drm_i915_private *dev_priv);
-void intel_fbc_init_pipe_state(struct drm_i915_private *dev_priv);
-void intel_fbc_enable(struct intel_crtc *crtc,
-		      struct intel_crtc_state *crtc_state,
-		      struct intel_plane_state *plane_state);
-void intel_fbc_disable(struct intel_crtc *crtc);
-void intel_fbc_global_disable(struct drm_i915_private *dev_priv);
-void intel_fbc_invalidate(struct drm_i915_private *dev_priv,
-			  unsigned int frontbuffer_bits,
-			  enum fb_op_origin origin);
-void intel_fbc_flush(struct drm_i915_private *dev_priv,
-		     unsigned int frontbuffer_bits, enum fb_op_origin origin);
-void intel_fbc_cleanup_cfb(struct drm_i915_private *dev_priv);
-void intel_fbc_handle_fifo_underrun_irq(struct drm_i915_private *dev_priv);
-int intel_fbc_reset_underrun(struct drm_i915_private *dev_priv);
-
-/* intel_hdmi.c */
-void intel_hdmi_init(struct drm_i915_private *dev_priv, i915_reg_t hdmi_reg,
-		     enum port port);
-void intel_hdmi_init_connector(struct intel_digital_port *intel_dig_port,
-			       struct intel_connector *intel_connector);
-struct intel_hdmi *enc_to_intel_hdmi(struct drm_encoder *encoder);
-int intel_hdmi_compute_config(struct intel_encoder *encoder,
-			      struct intel_crtc_state *pipe_config,
-			      struct drm_connector_state *conn_state);
-bool intel_hdmi_handle_sink_scrambling(struct intel_encoder *encoder,
-				       struct drm_connector *connector,
-				       bool high_tmds_clock_ratio,
-				       bool scrambling);
-void intel_dp_dual_mode_set_tmds_output(struct intel_hdmi *hdmi, bool enable);
-void intel_infoframe_init(struct intel_digital_port *intel_dig_port);
-
-/* intel_lvds.c */
-bool intel_lvds_port_enabled(struct drm_i915_private *dev_priv,
-			     i915_reg_t lvds_reg, enum pipe *pipe);
-void intel_lvds_init(struct drm_i915_private *dev_priv);
-struct intel_encoder *intel_get_lvds_encoder(struct drm_device *dev);
-bool intel_is_dual_link_lvds(struct drm_device *dev);
-
 /* intel_overlay.c */
 void intel_overlay_setup(struct drm_i915_private *dev_priv);
 void intel_overlay_cleanup(struct drm_i915_private *dev_priv);
@@ -2019,86 +1856,6 @@ int intel_overlay_attrs_ioctl(struct drm_device *dev, void *data,
 			      struct drm_file *file_priv);
 void intel_overlay_reset(struct drm_i915_private *dev_priv);
 
-
-/* intel_panel.c */
-int intel_panel_init(struct intel_panel *panel,
-		     struct drm_display_mode *fixed_mode,
-		     struct drm_display_mode *downclock_mode);
-void intel_panel_fini(struct intel_panel *panel);
-void intel_fixed_panel_mode(const struct drm_display_mode *fixed_mode,
-			    struct drm_display_mode *adjusted_mode);
-void intel_pch_panel_fitting(struct intel_crtc *crtc,
-			     struct intel_crtc_state *pipe_config,
-			     int fitting_mode);
-void intel_gmch_panel_fitting(struct intel_crtc *crtc,
-			      struct intel_crtc_state *pipe_config,
-			      int fitting_mode);
-void intel_panel_set_backlight_acpi(const struct drm_connector_state *conn_state,
-				    u32 level, u32 max);
-int intel_panel_setup_backlight(struct drm_connector *connector,
-				enum pipe pipe);
-void intel_panel_enable_backlight(const struct intel_crtc_state *crtc_state,
-				  const struct drm_connector_state *conn_state);
-void intel_panel_update_backlight(struct intel_encoder *encoder,
-				  const struct intel_crtc_state *crtc_state,
-				  const struct drm_connector_state *conn_state);
-void intel_panel_disable_backlight(const struct drm_connector_state *old_conn_state);
-extern struct drm_display_mode *intel_find_panel_downclock(
-				struct drm_i915_private *dev_priv,
-				struct drm_display_mode *fixed_mode,
-				struct drm_connector *connector);
-
-#if IS_ENABLED(CONFIG_BACKLIGHT_CLASS_DEVICE)
-int intel_backlight_device_register(struct intel_connector *connector);
-void intel_backlight_device_unregister(struct intel_connector *connector);
-#else /* CONFIG_BACKLIGHT_CLASS_DEVICE */
-static inline int intel_backlight_device_register(struct intel_connector *connector)
-{
-	return 0;
-}
-static inline void intel_backlight_device_unregister(struct intel_connector *connector)
-{
-}
-#endif /* CONFIG_BACKLIGHT_CLASS_DEVICE */
-
-/* intel_hdcp.c */
-void intel_hdcp_atomic_check(struct drm_connector *connector,
-			     struct drm_connector_state *old_state,
-			     struct drm_connector_state *new_state);
-int intel_hdcp_init(struct intel_connector *connector,
-		    const struct intel_hdcp_shim *hdcp_shim);
-int intel_hdcp_enable(struct intel_connector *connector);
-int intel_hdcp_disable(struct intel_connector *connector);
-int intel_hdcp_check_link(struct intel_connector *connector);
-bool is_hdcp_supported(struct drm_i915_private *dev_priv, enum port port);
-bool intel_hdcp_capable(struct intel_connector *connector);
-
-/* intel_psr.c */
-#define CAN_PSR(dev_priv) (HAS_PSR(dev_priv) && dev_priv->psr.sink_support)
-void intel_psr_init_dpcd(struct intel_dp *intel_dp);
-void intel_psr_enable(struct intel_dp *intel_dp,
-		      const struct intel_crtc_state *crtc_state);
-void intel_psr_disable(struct intel_dp *intel_dp,
-		      const struct intel_crtc_state *old_crtc_state);
-int intel_psr_set_debugfs_mode(struct drm_i915_private *dev_priv,
-			       struct drm_modeset_acquire_ctx *ctx,
-			       u64 value);
-void intel_psr_invalidate(struct drm_i915_private *dev_priv,
-			  unsigned frontbuffer_bits,
-			  enum fb_op_origin origin);
-void intel_psr_flush(struct drm_i915_private *dev_priv,
-		     unsigned frontbuffer_bits,
-		     enum fb_op_origin origin);
-void intel_psr_init(struct drm_i915_private *dev_priv);
-void intel_psr_compute_config(struct intel_dp *intel_dp,
-			      struct intel_crtc_state *crtc_state);
-void intel_psr_irq_control(struct drm_i915_private *dev_priv, u32 debug);
-void intel_psr_irq_handler(struct drm_i915_private *dev_priv, u32 psr_iir);
-void intel_psr_short_pulse(struct intel_dp *intel_dp);
-int intel_psr_wait_for_idle(const struct intel_crtc_state *new_crtc_state,
-			    u32 *out_value);
-bool intel_psr_enabled(struct intel_dp *intel_dp);
-
 /* intel_quirks.c */
 void intel_init_quirks(struct drm_i915_private *dev_priv);
 
@@ -2153,20 +1910,26 @@ void icl_dbuf_slices_update(struct drm_i915_private *dev_priv,
 			    u8 req_slices);
 
 static inline void
-assert_rpm_device_not_suspended(struct drm_i915_private *i915)
+assert_rpm_device_not_suspended(struct i915_runtime_pm *rpm)
 {
-	WARN_ONCE(i915->runtime_pm.suspended,
+	WARN_ONCE(rpm->suspended,
 		  "Device suspended during HW access\n");
 }
 
 static inline void
-assert_rpm_wakelock_held(struct drm_i915_private *i915)
+__assert_rpm_wakelock_held(struct i915_runtime_pm *rpm)
 {
-	assert_rpm_device_not_suspended(i915);
-	WARN_ONCE(!atomic_read(&i915->runtime_pm.wakeref_count),
+	assert_rpm_device_not_suspended(rpm);
+	WARN_ONCE(!atomic_read(&rpm->wakeref_count),
 		  "RPM wakelock ref not held during HW access");
 }
 
+static inline void
+assert_rpm_wakelock_held(struct drm_i915_private *i915)
+{
+	__assert_rpm_wakelock_held(&i915->runtime_pm);
+}
+
 /**
  * disable_rpm_wakeref_asserts - disable the RPM assert checks
  * @i915: i915 device instance
@@ -2242,101 +2005,6 @@ void chv_phy_powergate_lanes(struct intel_encoder *encoder,
 bool chv_phy_powergate_ch(struct drm_i915_private *dev_priv, enum dpio_phy phy,
 			  enum dpio_channel ch, bool override);
 
-
-/* intel_pm.c */
-void intel_init_clock_gating(struct drm_i915_private *dev_priv);
-void intel_suspend_hw(struct drm_i915_private *dev_priv);
-int ilk_wm_max_level(const struct drm_i915_private *dev_priv);
-void intel_update_watermarks(struct intel_crtc *crtc);
-void intel_init_pm(struct drm_i915_private *dev_priv);
-void intel_init_clock_gating_hooks(struct drm_i915_private *dev_priv);
-void intel_pm_setup(struct drm_i915_private *dev_priv);
-void intel_gpu_ips_init(struct drm_i915_private *dev_priv);
-void intel_gpu_ips_teardown(void);
-void intel_init_gt_powersave(struct drm_i915_private *dev_priv);
-void intel_cleanup_gt_powersave(struct drm_i915_private *dev_priv);
-void intel_sanitize_gt_powersave(struct drm_i915_private *dev_priv);
-void intel_enable_gt_powersave(struct drm_i915_private *dev_priv);
-void intel_disable_gt_powersave(struct drm_i915_private *dev_priv);
-void intel_suspend_gt_powersave(struct drm_i915_private *dev_priv);
-void gen6_rps_busy(struct drm_i915_private *dev_priv);
-void gen6_rps_reset_ei(struct drm_i915_private *dev_priv);
-void gen6_rps_idle(struct drm_i915_private *dev_priv);
-void gen6_rps_boost(struct i915_request *rq, struct intel_rps_client *rps);
-void g4x_wm_get_hw_state(struct drm_i915_private *dev_priv);
-void vlv_wm_get_hw_state(struct drm_i915_private *dev_priv);
-void ilk_wm_get_hw_state(struct drm_i915_private *dev_priv);
-void skl_wm_get_hw_state(struct drm_i915_private *dev_priv);
-void skl_pipe_ddb_get_hw_state(struct intel_crtc *crtc,
-			       struct skl_ddb_entry *ddb_y,
-			       struct skl_ddb_entry *ddb_uv);
-void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv,
-			  struct skl_ddb_allocation *ddb /* out */);
-void skl_pipe_wm_get_hw_state(struct intel_crtc *crtc,
-			      struct skl_pipe_wm *out);
-void g4x_wm_sanitize(struct drm_i915_private *dev_priv);
-void vlv_wm_sanitize(struct drm_i915_private *dev_priv);
-bool intel_can_enable_sagv(struct drm_atomic_state *state);
-int intel_enable_sagv(struct drm_i915_private *dev_priv);
-int intel_disable_sagv(struct drm_i915_private *dev_priv);
-bool skl_wm_level_equals(const struct skl_wm_level *l1,
-			 const struct skl_wm_level *l2);
-bool skl_ddb_allocation_overlaps(const struct skl_ddb_entry *ddb,
-				 const struct skl_ddb_entry entries[],
-				 int num_entries, int ignore_idx);
-void skl_write_plane_wm(struct intel_plane *plane,
-			const struct intel_crtc_state *crtc_state);
-void skl_write_cursor_wm(struct intel_plane *plane,
-			 const struct intel_crtc_state *crtc_state);
-bool ilk_disable_lp_wm(struct drm_device *dev);
-int skl_check_pipe_max_pixel_rate(struct intel_crtc *intel_crtc,
-				  struct intel_crtc_state *cstate);
-void intel_init_ipc(struct drm_i915_private *dev_priv);
-void intel_enable_ipc(struct drm_i915_private *dev_priv);
-
-/* intel_sdvo.c */
-bool intel_sdvo_port_enabled(struct drm_i915_private *dev_priv,
-			     i915_reg_t sdvo_reg, enum pipe *pipe);
-bool intel_sdvo_init(struct drm_i915_private *dev_priv,
-		     i915_reg_t reg, enum port port);
-
-
-/* intel_sprite.c */
-int intel_usecs_to_scanlines(const struct drm_display_mode *adjusted_mode,
-			     int usecs);
-struct intel_plane *intel_sprite_plane_create(struct drm_i915_private *dev_priv,
-					      enum pipe pipe, int plane);
-int intel_sprite_set_colorkey_ioctl(struct drm_device *dev, void *data,
-				    struct drm_file *file_priv);
-void intel_pipe_update_start(const struct intel_crtc_state *new_crtc_state);
-void intel_pipe_update_end(struct intel_crtc_state *new_crtc_state);
-int intel_plane_check_stride(const struct intel_plane_state *plane_state);
-int intel_plane_check_src_coordinates(struct intel_plane_state *plane_state);
-int chv_plane_check_rotation(const struct intel_plane_state *plane_state);
-struct intel_plane *
-skl_universal_plane_create(struct drm_i915_private *dev_priv,
-			   enum pipe pipe, enum plane_id plane_id);
-
-static inline bool icl_is_nv12_y_plane(enum plane_id id)
-{
-	/* Don't need to do a gen check, these planes are only available on gen11 */
-	if (id == PLANE_SPRITE4 || id == PLANE_SPRITE5)
-		return true;
-
-	return false;
-}
-
-static inline bool icl_is_hdr_plane(struct intel_plane *plane)
-{
-	if (INTEL_GEN(to_i915(plane->base.dev)) < 11)
-		return false;
-
-	return plane->id < PLANE_SPRITE2;
-}
-
-/* intel_tv.c */
-void intel_tv_init(struct drm_i915_private *dev_priv);
-
 /* intel_atomic.c */
 int intel_digital_connector_atomic_get_property(struct drm_connector *connector,
 						const struct drm_connector_state *state,
@@ -2373,64 +2041,4 @@ int intel_atomic_setup_scalers(struct drm_i915_private *dev_priv,
 			       struct intel_crtc *intel_crtc,
 			       struct intel_crtc_state *crtc_state);
 
-/* intel_atomic_plane.c */
-struct intel_plane *intel_plane_alloc(void);
-void intel_plane_free(struct intel_plane *plane);
-struct drm_plane_state *intel_plane_duplicate_state(struct drm_plane *plane);
-void intel_plane_destroy_state(struct drm_plane *plane,
-			       struct drm_plane_state *state);
-extern const struct drm_plane_helper_funcs intel_plane_helper_funcs;
-void skl_update_planes_on_crtc(struct intel_atomic_state *state,
-			       struct intel_crtc *crtc);
-void i9xx_update_planes_on_crtc(struct intel_atomic_state *state,
-				struct intel_crtc *crtc);
-int intel_plane_atomic_check_with_state(const struct intel_crtc_state *old_crtc_state,
-					struct intel_crtc_state *crtc_state,
-					const struct intel_plane_state *old_plane_state,
-					struct intel_plane_state *intel_state);
-
-/* intel_color.c */
-void intel_color_init(struct intel_crtc *crtc);
-int intel_color_check(struct intel_crtc_state *crtc_state);
-void intel_color_commit(const struct intel_crtc_state *crtc_state);
-void intel_color_load_luts(const struct intel_crtc_state *crtc_state);
-
-/* intel_lspcon.c */
-bool lspcon_init(struct intel_digital_port *intel_dig_port);
-void lspcon_resume(struct intel_lspcon *lspcon);
-void lspcon_wait_pcon_mode(struct intel_lspcon *lspcon);
-void lspcon_write_infoframe(struct intel_encoder *encoder,
-			    const struct intel_crtc_state *crtc_state,
-			    unsigned int type,
-			    const void *buf, ssize_t len);
-void lspcon_set_infoframes(struct intel_encoder *encoder,
-			   bool enable,
-			   const struct intel_crtc_state *crtc_state,
-			   const struct drm_connector_state *conn_state);
-bool lspcon_infoframe_enabled(struct intel_encoder *encoder,
-			      const struct intel_crtc_state *pipe_config);
-void lspcon_ycbcr420_config(struct drm_connector *connector,
-			    struct intel_crtc_state *crtc_state);
-
-/* intel_pipe_crc.c */
-#ifdef CONFIG_DEBUG_FS
-int intel_crtc_set_crc_source(struct drm_crtc *crtc, const char *source_name);
-int intel_crtc_verify_crc_source(struct drm_crtc *crtc,
-				 const char *source_name, size_t *values_cnt);
-const char *const *intel_crtc_get_crc_sources(struct drm_crtc *crtc,
-					      size_t *count);
-void intel_crtc_disable_pipe_crc(struct intel_crtc *crtc);
-void intel_crtc_enable_pipe_crc(struct intel_crtc *crtc);
-#else
-#define intel_crtc_set_crc_source NULL
-#define intel_crtc_verify_crc_source NULL
-#define intel_crtc_get_crc_sources NULL
-static inline void intel_crtc_disable_pipe_crc(struct intel_crtc *crtc)
-{
-}
-
-static inline void intel_crtc_enable_pipe_crc(struct intel_crtc *crtc)
-{
-}
-#endif
 #endif /* __INTEL_DRV_H__ */
diff --git a/drivers/gpu/drm/i915/intel_dsi.h b/drivers/gpu/drm/i915/intel_dsi.h
index a9a19778dc7f..705a609050c0 100644
--- a/drivers/gpu/drm/i915/intel_dsi.h
+++ b/drivers/gpu/drm/i915/intel_dsi.h
@@ -189,7 +189,6 @@ void bxt_dsi_reset_clocks(struct intel_encoder *encoder, enum port port);
 
 /* intel_dsi_vbt.c */
 bool intel_dsi_vbt_init(struct intel_dsi *intel_dsi, u16 panel_id);
-int intel_dsi_vbt_get_modes(struct intel_dsi *intel_dsi);
 void intel_dsi_vbt_exec_sequence(struct intel_dsi *intel_dsi,
 				 enum mipi_seq seq_id);
 void intel_dsi_msleep(struct intel_dsi *intel_dsi, int msec);
diff --git a/drivers/gpu/drm/i915/intel_dsi_vbt.c b/drivers/gpu/drm/i915/intel_dsi_vbt.c
index 06a11c35a784..3074448446bc 100644
--- a/drivers/gpu/drm/i915/intel_dsi_vbt.c
+++ b/drivers/gpu/drm/i915/intel_dsi_vbt.c
@@ -194,7 +194,7 @@ static const u8 *mipi_exec_send_packet(struct intel_dsi *intel_dsi,
 		break;
 	}
 
-	if (!IS_ICELAKE(dev_priv))
+	if (INTEL_GEN(dev_priv) < 11)
 		vlv_dsi_wait_for_fifo_empty(intel_dsi, port);
 
 out:
@@ -365,7 +365,7 @@ static const u8 *mipi_exec_gpio(struct intel_dsi *intel_dsi, const u8 *data)
 	/* pull up/down */
 	value = *data++ & 1;
 
-	if (IS_ICELAKE(dev_priv))
+	if (INTEL_GEN(dev_priv) >= 11)
 		icl_exec_gpio(dev_priv, gpio_source, gpio_index, value);
 	else if (IS_VALLEYVIEW(dev_priv))
 		vlv_exec_gpio(dev_priv, gpio_source, gpio_number, value);
@@ -532,24 +532,6 @@ void intel_dsi_msleep(struct intel_dsi *intel_dsi, int msec)
 	msleep(msec);
 }
 
-int intel_dsi_vbt_get_modes(struct intel_dsi *intel_dsi)
-{
-	struct intel_connector *connector = intel_dsi->attached_connector;
-	struct drm_device *dev = intel_dsi->base.base.dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct drm_display_mode *mode;
-
-	mode = drm_mode_duplicate(dev, dev_priv->vbt.lfp_lvds_vbt_mode);
-	if (!mode)
-		return 0;
-
-	mode->type |= DRM_MODE_TYPE_PREFERRED;
-
-	drm_mode_probed_add(&connector->base, mode);
-
-	return 1;
-}
-
 #define ICL_PREPARE_CNT_MAX	0x7
 #define ICL_CLK_ZERO_CNT_MAX	0xf
 #define ICL_TRAIL_CNT_MAX	0x7
@@ -890,7 +872,7 @@ bool intel_dsi_vbt_init(struct intel_dsi *intel_dsi, u16 panel_id)
 
 	intel_dsi->burst_mode_ratio = burst_mode_ratio;
 
-	if (IS_ICELAKE(dev_priv))
+	if (INTEL_GEN(dev_priv) >= 11)
 		icl_dphy_param_init(intel_dsi);
 	else
 		vlv_dphy_param_init(intel_dsi);
diff --git a/drivers/gpu/drm/i915/intel_dvo.c b/drivers/gpu/drm/i915/intel_dvo.c
index a6c82482a841..adef81c8cccb 100644
--- a/drivers/gpu/drm/i915/intel_dvo.c
+++ b/drivers/gpu/drm/i915/intel_dvo.c
@@ -24,14 +24,20 @@
  * Authors:
  *	Eric Anholt <eric@anholt.net>
  */
+
 #include <linux/i2c.h>
 #include <linux/slab.h>
+
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
-#include "intel_drv.h"
 #include <drm/i915_drm.h>
-#include "i915_drv.h"
+
 #include "dvo.h"
+#include "i915_drv.h"
+#include "intel_connector.h"
+#include "intel_drv.h"
+#include "intel_dvo.h"
+#include "intel_panel.h"
 
 #define SIL164_ADDR	0x38
 #define CH7xxx_ADDR	0x76
diff --git a/drivers/gpu/drm/i915/intel_dvo.h b/drivers/gpu/drm/i915/intel_dvo.h
new file mode 100644
index 000000000000..3ed0fdf8efff
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_dvo.h
@@ -0,0 +1,13 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_DVO_H__
+#define __INTEL_DVO_H__
+
+struct drm_i915_private;
+
+void intel_dvo_init(struct drm_i915_private *dev_priv);
+
+#endif /* __INTEL_DVO_H__ */
diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c b/drivers/gpu/drm/i915/intel_engine_cs.c
index 49fa43ff02ba..eea9bec04f1b 100644
--- a/drivers/gpu/drm/i915/intel_engine_cs.c
+++ b/drivers/gpu/drm/i915/intel_engine_cs.c
@@ -84,7 +84,6 @@ static const struct engine_class_info intel_engine_classes[] = {
 #define MAX_MMIO_BASES 3
 struct engine_info {
 	unsigned int hw_id;
-	unsigned int uabi_id;
 	u8 class;
 	u8 instance;
 	/* mmio bases table *must* be sorted in reverse gen order */
@@ -95,27 +94,24 @@ struct engine_info {
 };
 
 static const struct engine_info intel_engines[] = {
-	[RCS] = {
-		.hw_id = RCS_HW,
-		.uabi_id = I915_EXEC_RENDER,
+	[RCS0] = {
+		.hw_id = RCS0_HW,
 		.class = RENDER_CLASS,
 		.instance = 0,
 		.mmio_bases = {
 			{ .gen = 1, .base = RENDER_RING_BASE }
 		},
 	},
-	[BCS] = {
-		.hw_id = BCS_HW,
-		.uabi_id = I915_EXEC_BLT,
+	[BCS0] = {
+		.hw_id = BCS0_HW,
 		.class = COPY_ENGINE_CLASS,
 		.instance = 0,
 		.mmio_bases = {
 			{ .gen = 6, .base = BLT_RING_BASE }
 		},
 	},
-	[VCS] = {
-		.hw_id = VCS_HW,
-		.uabi_id = I915_EXEC_BSD,
+	[VCS0] = {
+		.hw_id = VCS0_HW,
 		.class = VIDEO_DECODE_CLASS,
 		.instance = 0,
 		.mmio_bases = {
@@ -124,9 +120,8 @@ static const struct engine_info intel_engines[] = {
 			{ .gen = 4, .base = BSD_RING_BASE }
 		},
 	},
-	[VCS2] = {
-		.hw_id = VCS2_HW,
-		.uabi_id = I915_EXEC_BSD,
+	[VCS1] = {
+		.hw_id = VCS1_HW,
 		.class = VIDEO_DECODE_CLASS,
 		.instance = 1,
 		.mmio_bases = {
@@ -134,27 +129,24 @@ static const struct engine_info intel_engines[] = {
 			{ .gen = 8, .base = GEN8_BSD2_RING_BASE }
 		},
 	},
-	[VCS3] = {
-		.hw_id = VCS3_HW,
-		.uabi_id = I915_EXEC_BSD,
+	[VCS2] = {
+		.hw_id = VCS2_HW,
 		.class = VIDEO_DECODE_CLASS,
 		.instance = 2,
 		.mmio_bases = {
 			{ .gen = 11, .base = GEN11_BSD3_RING_BASE }
 		},
 	},
-	[VCS4] = {
-		.hw_id = VCS4_HW,
-		.uabi_id = I915_EXEC_BSD,
+	[VCS3] = {
+		.hw_id = VCS3_HW,
 		.class = VIDEO_DECODE_CLASS,
 		.instance = 3,
 		.mmio_bases = {
 			{ .gen = 11, .base = GEN11_BSD4_RING_BASE }
 		},
 	},
-	[VECS] = {
-		.hw_id = VECS_HW,
-		.uabi_id = I915_EXEC_VEBOX,
+	[VECS0] = {
+		.hw_id = VECS0_HW,
 		.class = VIDEO_ENHANCEMENT_CLASS,
 		.instance = 0,
 		.mmio_bases = {
@@ -162,9 +154,8 @@ static const struct engine_info intel_engines[] = {
 			{ .gen = 7, .base = VEBOX_RING_BASE }
 		},
 	},
-	[VECS2] = {
-		.hw_id = VECS2_HW,
-		.uabi_id = I915_EXEC_VEBOX,
+	[VECS1] = {
+		.hw_id = VECS1_HW,
 		.class = VIDEO_ENHANCEMENT_CLASS,
 		.instance = 1,
 		.mmio_bases = {
@@ -264,21 +255,17 @@ static void __sprint_engine_name(char *name, const struct engine_info *info)
 
 void intel_engine_set_hwsp_writemask(struct intel_engine_cs *engine, u32 mask)
 {
-	struct drm_i915_private *dev_priv = engine->i915;
-	i915_reg_t hwstam;
-
 	/*
 	 * Though they added more rings on g4x/ilk, they did not add
 	 * per-engine HWSTAM until gen6.
 	 */
-	if (INTEL_GEN(dev_priv) < 6 && engine->class != RENDER_CLASS)
+	if (INTEL_GEN(engine->i915) < 6 && engine->class != RENDER_CLASS)
 		return;
 
-	hwstam = RING_HWSTAM(engine->mmio_base);
-	if (INTEL_GEN(dev_priv) >= 3)
-		I915_WRITE(hwstam, mask);
+	if (INTEL_GEN(engine->i915) >= 3)
+		ENGINE_WRITE(engine, RING_HWSTAM, mask);
 	else
-		I915_WRITE16(hwstam, mask);
+		ENGINE_WRITE16(engine, RING_HWSTAM, mask);
 }
 
 static void intel_engine_sanitize_mmio(struct intel_engine_cs *engine)
@@ -313,15 +300,18 @@ intel_engine_setup(struct drm_i915_private *dev_priv,
 	if (!engine)
 		return -ENOMEM;
 
+	BUILD_BUG_ON(BITS_PER_TYPE(engine->mask) < I915_NUM_ENGINES);
+
 	engine->id = id;
+	engine->mask = BIT(id);
 	engine->i915 = dev_priv;
+	engine->uncore = &dev_priv->uncore;
 	__sprint_engine_name(engine->name, info);
 	engine->hw_id = engine->guc_id = info->hw_id;
 	engine->mmio_base = __engine_mmio_base(dev_priv, info->mmio_bases);
 	engine->class = info->class;
 	engine->instance = info->instance;
 
-	engine->uabi_id = info->uabi_id;
 	engine->uabi_class = intel_engine_classes[info->class].uabi_class;
 
 	engine->context_size = __intel_engine_context_size(dev_priv,
@@ -355,15 +345,15 @@ intel_engine_setup(struct drm_i915_private *dev_priv,
 int intel_engines_init_mmio(struct drm_i915_private *dev_priv)
 {
 	struct intel_device_info *device_info = mkwrite_device_info(dev_priv);
-	const unsigned int ring_mask = INTEL_INFO(dev_priv)->ring_mask;
+	const unsigned int engine_mask = INTEL_INFO(dev_priv)->engine_mask;
 	struct intel_engine_cs *engine;
 	enum intel_engine_id id;
 	unsigned int mask = 0;
 	unsigned int i;
 	int err;
 
-	WARN_ON(ring_mask == 0);
-	WARN_ON(ring_mask &
+	WARN_ON(engine_mask == 0);
+	WARN_ON(engine_mask &
 		GENMASK(BITS_PER_TYPE(mask) - 1, I915_NUM_ENGINES));
 
 	if (i915_inject_load_failure())
@@ -377,7 +367,7 @@ int intel_engines_init_mmio(struct drm_i915_private *dev_priv)
 		if (err)
 			goto cleanup;
 
-		mask |= ENGINE_MASK(i);
+		mask |= BIT(i);
 	}
 
 	/*
@@ -385,16 +375,16 @@ int intel_engines_init_mmio(struct drm_i915_private *dev_priv)
 	 * are added to the driver by a warning and disabling the forgotten
 	 * engines.
 	 */
-	if (WARN_ON(mask != ring_mask))
-		device_info->ring_mask = mask;
+	if (WARN_ON(mask != engine_mask))
+		device_info->engine_mask = mask;
 
 	/* We always presume we have at least RCS available for later probing */
-	if (WARN_ON(!HAS_ENGINE(dev_priv, RCS))) {
+	if (WARN_ON(!HAS_ENGINE(dev_priv, RCS0))) {
 		err = -ENODEV;
 		goto cleanup;
 	}
 
-	RUNTIME_INFO(dev_priv)->num_rings = hweight32(mask);
+	RUNTIME_INFO(dev_priv)->num_engines = hweight32(mask);
 
 	i915_check_and_clear_faults(dev_priv);
 
@@ -455,12 +445,6 @@ cleanup:
 	return err;
 }
 
-void intel_engine_write_global_seqno(struct intel_engine_cs *engine, u32 seqno)
-{
-	intel_write_status_page(engine, I915_GEM_HWS_INDEX, seqno);
-	GEM_BUG_ON(intel_engine_get_seqno(engine) != seqno);
-}
-
 static void intel_engine_init_batch_pool(struct intel_engine_cs *engine)
 {
 	i915_gem_batch_pool_init(&engine->batch_pool, engine);
@@ -541,9 +525,7 @@ static int init_status_page(struct intel_engine_cs *engine)
 		return PTR_ERR(obj);
 	}
 
-	ret = i915_gem_object_set_cache_level(obj, I915_CACHE_LLC);
-	if (ret)
-		goto err;
+	i915_gem_object_set_cache_coherency(obj, I915_CACHE_LLC);
 
 	vma = i915_vma_instance(obj, &engine->i915->ggtt.vm, NULL);
 	if (IS_ERR(vma)) {
@@ -594,7 +576,6 @@ int intel_engine_setup_common(struct intel_engine_cs *engine)
 
 	err = i915_timeline_init(engine->i915,
 				 &engine->timeline,
-				 engine->name,
 				 engine->status_page.vma);
 	if (err)
 		goto err_hwsp;
@@ -614,10 +595,44 @@ err_hwsp:
 	return err;
 }
 
-static void __intel_context_unpin(struct i915_gem_context *ctx,
-				  struct intel_engine_cs *engine)
+void intel_engines_set_scheduler_caps(struct drm_i915_private *i915)
 {
-	intel_context_unpin(to_intel_context(ctx, engine));
+	static const struct {
+		u8 engine;
+		u8 sched;
+	} map[] = {
+#define MAP(x, y) { ilog2(I915_ENGINE_HAS_##x), ilog2(I915_SCHEDULER_CAP_##y) }
+		MAP(PREEMPTION, PREEMPTION),
+		MAP(SEMAPHORES, SEMAPHORES),
+#undef MAP
+	};
+	struct intel_engine_cs *engine;
+	enum intel_engine_id id;
+	u32 enabled, disabled;
+
+	enabled = 0;
+	disabled = 0;
+	for_each_engine(engine, i915, id) { /* all engines must agree! */
+		int i;
+
+		if (engine->schedule)
+			enabled |= (I915_SCHEDULER_CAP_ENABLED |
+				    I915_SCHEDULER_CAP_PRIORITY);
+		else
+			disabled |= (I915_SCHEDULER_CAP_ENABLED |
+				     I915_SCHEDULER_CAP_PRIORITY);
+
+		for (i = 0; i < ARRAY_SIZE(map); i++) {
+			if (engine->flags & BIT(map[i].engine))
+				enabled |= BIT(map[i].sched);
+			else
+				disabled |= BIT(map[i].sched);
+		}
+	}
+
+	i915->caps.scheduler = enabled & ~disabled;
+	if (!(i915->caps.scheduler & I915_SCHEDULER_CAP_ENABLED))
+		i915->caps.scheduler = 0;
 }
 
 struct measure_breadcrumb {
@@ -639,7 +654,7 @@ static int measure_breadcrumb_dw(struct intel_engine_cs *engine)
 		return -ENOMEM;
 
 	if (i915_timeline_init(engine->i915,
-			       &frame->timeline, "measure",
+			       &frame->timeline,
 			       engine->status_page.vma))
 		goto out_frame;
 
@@ -670,6 +685,20 @@ out_frame:
 	return dw;
 }
 
+static int pin_context(struct i915_gem_context *ctx,
+		       struct intel_engine_cs *engine,
+		       struct intel_context **out)
+{
+	struct intel_context *ce;
+
+	ce = intel_context_pin(ctx, engine);
+	if (IS_ERR(ce))
+		return PTR_ERR(ce);
+
+	*out = ce;
+	return 0;
+}
+
 /**
  * intel_engines_init_common - initialize cengine state which might require hw access
  * @engine: Engine to initialize.
@@ -684,11 +713,8 @@ out_frame:
 int intel_engine_init_common(struct intel_engine_cs *engine)
 {
 	struct drm_i915_private *i915 = engine->i915;
-	struct intel_context *ce;
 	int ret;
 
-	engine->set_default_submission(engine);
-
 	/* We may need to do things with the shrinker which
 	 * require us to immediately switch back to the default
 	 * context. This can cause a problem as pinning the
@@ -696,39 +722,61 @@ int intel_engine_init_common(struct intel_engine_cs *engine)
 	 * be available. To avoid this we always pin the default
 	 * context.
 	 */
-	ce = intel_context_pin(i915->kernel_context, engine);
-	if (IS_ERR(ce))
-		return PTR_ERR(ce);
+	ret = pin_context(i915->kernel_context, engine,
+			  &engine->kernel_context);
+	if (ret)
+		return ret;
 
 	/*
 	 * Similarly the preempt context must always be available so that
-	 * we can interrupt the engine at any time.
+	 * we can interrupt the engine at any time. However, as preemption
+	 * is optional, we allow it to fail.
 	 */
-	if (i915->preempt_context) {
-		ce = intel_context_pin(i915->preempt_context, engine);
-		if (IS_ERR(ce)) {
-			ret = PTR_ERR(ce);
-			goto err_unpin_kernel;
-		}
-	}
+	if (i915->preempt_context)
+		pin_context(i915->preempt_context, engine,
+			    &engine->preempt_context);
 
 	ret = measure_breadcrumb_dw(engine);
 	if (ret < 0)
-		goto err_unpin_preempt;
+		goto err_unpin;
 
 	engine->emit_fini_breadcrumb_dw = ret;
 
-	return 0;
+	engine->set_default_submission(engine);
 
-err_unpin_preempt:
-	if (i915->preempt_context)
-		__intel_context_unpin(i915->preempt_context, engine);
+	return 0;
 
-err_unpin_kernel:
-	__intel_context_unpin(i915->kernel_context, engine);
+err_unpin:
+	if (engine->preempt_context)
+		intel_context_unpin(engine->preempt_context);
+	intel_context_unpin(engine->kernel_context);
 	return ret;
 }
 
+void intel_gt_resume(struct drm_i915_private *i915)
+{
+	struct intel_engine_cs *engine;
+	enum intel_engine_id id;
+
+	/*
+	 * After resume, we may need to poke into the pinned kernel
+	 * contexts to paper over any damage caused by the sudden suspend.
+	 * Only the kernel contexts should remain pinned over suspend,
+	 * allowing us to fixup the user contexts on their first pin.
+	 */
+	for_each_engine(engine, i915, id) {
+		struct intel_context *ce;
+
+		ce = engine->kernel_context;
+		if (ce)
+			ce->ops->reset(ce);
+
+		ce = engine->preempt_context;
+		if (ce)
+			ce->ops->reset(ce);
+	}
+}
+
 /**
  * intel_engines_cleanup_common - cleans up the engine state created by
  *                                the common initiailizers.
@@ -738,8 +786,6 @@ err_unpin_kernel:
  */
 void intel_engine_cleanup_common(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *i915 = engine->i915;
-
 	cleanup_status_page(engine);
 
 	intel_engine_fini_breadcrumbs(engine);
@@ -749,9 +795,9 @@ void intel_engine_cleanup_common(struct intel_engine_cs *engine)
 	if (engine->default_state)
 		i915_gem_object_put(engine->default_state);
 
-	if (i915->preempt_context)
-		__intel_context_unpin(i915->preempt_context, engine);
-	__intel_context_unpin(i915->kernel_context, engine);
+	if (engine->preempt_context)
+		intel_context_unpin(engine->preempt_context);
+	intel_context_unpin(engine->kernel_context);
 
 	i915_timeline_fini(&engine->timeline);
 
@@ -762,50 +808,48 @@ void intel_engine_cleanup_common(struct intel_engine_cs *engine)
 
 u64 intel_engine_get_active_head(const struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = engine->i915;
+	struct drm_i915_private *i915 = engine->i915;
+
 	u64 acthd;
 
-	if (INTEL_GEN(dev_priv) >= 8)
-		acthd = I915_READ64_2x32(RING_ACTHD(engine->mmio_base),
-					 RING_ACTHD_UDW(engine->mmio_base));
-	else if (INTEL_GEN(dev_priv) >= 4)
-		acthd = I915_READ(RING_ACTHD(engine->mmio_base));
+	if (INTEL_GEN(i915) >= 8)
+		acthd = ENGINE_READ64(engine, RING_ACTHD, RING_ACTHD_UDW);
+	else if (INTEL_GEN(i915) >= 4)
+		acthd = ENGINE_READ(engine, RING_ACTHD);
 	else
-		acthd = I915_READ(ACTHD);
+		acthd = ENGINE_READ(engine, ACTHD);
 
 	return acthd;
 }
 
 u64 intel_engine_get_last_batch_head(const struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = engine->i915;
 	u64 bbaddr;
 
-	if (INTEL_GEN(dev_priv) >= 8)
-		bbaddr = I915_READ64_2x32(RING_BBADDR(engine->mmio_base),
-					  RING_BBADDR_UDW(engine->mmio_base));
+	if (INTEL_GEN(engine->i915) >= 8)
+		bbaddr = ENGINE_READ64(engine, RING_BBADDR, RING_BBADDR_UDW);
 	else
-		bbaddr = I915_READ(RING_BBADDR(engine->mmio_base));
+		bbaddr = ENGINE_READ(engine, RING_BBADDR);
 
 	return bbaddr;
 }
 
 int intel_engine_stop_cs(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = engine->i915;
+	struct intel_uncore *uncore = engine->uncore;
 	const u32 base = engine->mmio_base;
 	const i915_reg_t mode = RING_MI_MODE(base);
 	int err;
 
-	if (INTEL_GEN(dev_priv) < 3)
+	if (INTEL_GEN(engine->i915) < 3)
 		return -ENODEV;
 
 	GEM_TRACE("%s\n", engine->name);
 
-	I915_WRITE_FW(mode, _MASKED_BIT_ENABLE(STOP_RING));
+	intel_uncore_write_fw(uncore, mode, _MASKED_BIT_ENABLE(STOP_RING));
 
 	err = 0;
-	if (__intel_wait_for_register_fw(dev_priv,
+	if (__intel_wait_for_register_fw(uncore,
 					 mode, MODE_IDLE, MODE_IDLE,
 					 1000, 0,
 					 NULL)) {
@@ -814,19 +858,16 @@ int intel_engine_stop_cs(struct intel_engine_cs *engine)
 	}
 
 	/* A final mmio read to let GPU writes be hopefully flushed to memory */
-	POSTING_READ_FW(mode);
+	intel_uncore_posting_read_fw(uncore, mode);
 
 	return err;
 }
 
 void intel_engine_cancel_stop_cs(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = engine->i915;
-
 	GEM_TRACE("%s\n", engine->name);
 
-	I915_WRITE_FW(RING_MI_MODE(engine->mmio_base),
-		      _MASKED_BIT_DISABLE(STOP_RING));
+	ENGINE_WRITE_FW(engine, RING_MI_MODE, _MASKED_BIT_DISABLE(STOP_RING));
 }
 
 const char *i915_cache_level_str(struct drm_i915_private *i915, int type)
@@ -863,6 +904,7 @@ static inline u32
 read_subslice_reg(struct drm_i915_private *dev_priv, int slice,
 		  int subslice, i915_reg_t reg)
 {
+	struct intel_uncore *uncore = &dev_priv->uncore;
 	u32 mcr_slice_subslice_mask;
 	u32 mcr_slice_subslice_select;
 	u32 default_mcr_s_ss_select;
@@ -884,33 +926,33 @@ read_subslice_reg(struct drm_i915_private *dev_priv, int slice,
 
 	default_mcr_s_ss_select = intel_calculate_mcr_s_ss_select(dev_priv);
 
-	fw_domains = intel_uncore_forcewake_for_reg(dev_priv, reg,
+	fw_domains = intel_uncore_forcewake_for_reg(uncore, reg,
 						    FW_REG_READ);
-	fw_domains |= intel_uncore_forcewake_for_reg(dev_priv,
+	fw_domains |= intel_uncore_forcewake_for_reg(uncore,
 						     GEN8_MCR_SELECTOR,
 						     FW_REG_READ | FW_REG_WRITE);
 
-	spin_lock_irq(&dev_priv->uncore.lock);
-	intel_uncore_forcewake_get__locked(dev_priv, fw_domains);
+	spin_lock_irq(&uncore->lock);
+	intel_uncore_forcewake_get__locked(uncore, fw_domains);
 
-	mcr = I915_READ_FW(GEN8_MCR_SELECTOR);
+	mcr = intel_uncore_read_fw(uncore, GEN8_MCR_SELECTOR);
 
 	WARN_ON_ONCE((mcr & mcr_slice_subslice_mask) !=
 		     default_mcr_s_ss_select);
 
 	mcr &= ~mcr_slice_subslice_mask;
 	mcr |= mcr_slice_subslice_select;
-	I915_WRITE_FW(GEN8_MCR_SELECTOR, mcr);
+	intel_uncore_write_fw(uncore, GEN8_MCR_SELECTOR, mcr);
 
-	ret = I915_READ_FW(reg);
+	ret = intel_uncore_read_fw(uncore, reg);
 
 	mcr &= ~mcr_slice_subslice_mask;
 	mcr |= default_mcr_s_ss_select;
 
-	I915_WRITE_FW(GEN8_MCR_SELECTOR, mcr);
+	intel_uncore_write_fw(uncore, GEN8_MCR_SELECTOR, mcr);
 
-	intel_uncore_forcewake_put__locked(dev_priv, fw_domains);
-	spin_unlock_irq(&dev_priv->uncore.lock);
+	intel_uncore_forcewake_put__locked(uncore, fw_domains);
+	spin_unlock_irq(&uncore->lock);
 
 	return ret;
 }
@@ -920,6 +962,7 @@ void intel_engine_get_instdone(struct intel_engine_cs *engine,
 			       struct intel_instdone *instdone)
 {
 	struct drm_i915_private *dev_priv = engine->i915;
+	struct intel_uncore *uncore = engine->uncore;
 	u32 mmio_base = engine->mmio_base;
 	int slice;
 	int subslice;
@@ -928,12 +971,14 @@ void intel_engine_get_instdone(struct intel_engine_cs *engine,
 
 	switch (INTEL_GEN(dev_priv)) {
 	default:
-		instdone->instdone = I915_READ(RING_INSTDONE(mmio_base));
+		instdone->instdone =
+			intel_uncore_read(uncore, RING_INSTDONE(mmio_base));
 
-		if (engine->id != RCS)
+		if (engine->id != RCS0)
 			break;
 
-		instdone->slice_common = I915_READ(GEN7_SC_INSTDONE);
+		instdone->slice_common =
+			intel_uncore_read(uncore, GEN7_SC_INSTDONE);
 		for_each_instdone_slice_subslice(dev_priv, slice, subslice) {
 			instdone->sampler[slice][subslice] =
 				read_subslice_reg(dev_priv, slice, subslice,
@@ -944,28 +989,33 @@ void intel_engine_get_instdone(struct intel_engine_cs *engine,
 		}
 		break;
 	case 7:
-		instdone->instdone = I915_READ(RING_INSTDONE(mmio_base));
+		instdone->instdone =
+			intel_uncore_read(uncore, RING_INSTDONE(mmio_base));
 
-		if (engine->id != RCS)
+		if (engine->id != RCS0)
 			break;
 
-		instdone->slice_common = I915_READ(GEN7_SC_INSTDONE);
-		instdone->sampler[0][0] = I915_READ(GEN7_SAMPLER_INSTDONE);
-		instdone->row[0][0] = I915_READ(GEN7_ROW_INSTDONE);
+		instdone->slice_common =
+			intel_uncore_read(uncore, GEN7_SC_INSTDONE);
+		instdone->sampler[0][0] =
+			intel_uncore_read(uncore, GEN7_SAMPLER_INSTDONE);
+		instdone->row[0][0] =
+			intel_uncore_read(uncore, GEN7_ROW_INSTDONE);
 
 		break;
 	case 6:
 	case 5:
 	case 4:
-		instdone->instdone = I915_READ(RING_INSTDONE(mmio_base));
-
-		if (engine->id == RCS)
+		instdone->instdone =
+			intel_uncore_read(uncore, RING_INSTDONE(mmio_base));
+		if (engine->id == RCS0)
 			/* HACK: Using the wrong struct member */
-			instdone->slice_common = I915_READ(GEN4_INSTDONE1);
+			instdone->slice_common =
+				intel_uncore_read(uncore, GEN4_INSTDONE1);
 		break;
 	case 3:
 	case 2:
-		instdone->instdone = I915_READ(GEN2_INSTDONE);
+		instdone->instdone = intel_uncore_read(uncore, GEN2_INSTDONE);
 		break;
 	}
 }
@@ -985,12 +1035,13 @@ static bool ring_is_idle(struct intel_engine_cs *engine)
 		return true;
 
 	/* First check that no commands are left in the ring */
-	if ((I915_READ_HEAD(engine) & HEAD_ADDR) !=
-	    (I915_READ_TAIL(engine) & TAIL_ADDR))
+	if ((ENGINE_READ(engine, RING_HEAD) & HEAD_ADDR) !=
+	    (ENGINE_READ(engine, RING_TAIL) & TAIL_ADDR))
 		idle = false;
 
 	/* No bit for gen2, so assume the CS parser is idle */
-	if (INTEL_GEN(dev_priv) > 2 && !(I915_READ_MODE(engine) & MODE_IDLE))
+	if (INTEL_GEN(dev_priv) > 2 &&
+	    !(ENGINE_READ(engine, RING_MI_MODE) & MODE_IDLE))
 		idle = false;
 
 	intel_runtime_pm_put(dev_priv, wakeref);
@@ -1007,16 +1058,10 @@ static bool ring_is_idle(struct intel_engine_cs *engine)
  */
 bool intel_engine_is_idle(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = engine->i915;
-
 	/* More white lies, if wedged, hw state is inconsistent */
-	if (i915_terminally_wedged(&dev_priv->gpu_error))
+	if (i915_reset_failed(engine->i915))
 		return true;
 
-	/* Any inflight/incomplete requests? */
-	if (!intel_engine_signaled(engine, intel_engine_last_submit(engine)))
-		return false;
-
 	/* Waiting to drain ELSP? */
 	if (READ_ONCE(engine->execlists.active)) {
 		struct tasklet_struct *t = &engine->execlists.tasklet;
@@ -1045,7 +1090,7 @@ bool intel_engine_is_idle(struct intel_engine_cs *engine)
 	return ring_is_idle(engine);
 }
 
-bool intel_engines_are_idle(struct drm_i915_private *dev_priv)
+bool intel_engines_are_idle(struct drm_i915_private *i915)
 {
 	struct intel_engine_cs *engine;
 	enum intel_engine_id id;
@@ -1054,10 +1099,14 @@ bool intel_engines_are_idle(struct drm_i915_private *dev_priv)
 	 * If the driver is wedged, HW state may be very inconsistent and
 	 * report that it is still busy, even though we have stopped using it.
 	 */
-	if (i915_terminally_wedged(&dev_priv->gpu_error))
+	if (i915_reset_failed(i915))
 		return true;
 
-	for_each_engine(engine, dev_priv, id) {
+	/* Already parked (and passed an idleness test); must still be idle */
+	if (!READ_ONCE(i915->gt.awake))
+		return true;
+
+	for_each_engine(engine, i915, id) {
 		if (!intel_engine_is_idle(engine))
 			return false;
 	}
@@ -1065,34 +1114,6 @@ bool intel_engines_are_idle(struct drm_i915_private *dev_priv)
 	return true;
 }
 
-/**
- * intel_engine_has_kernel_context:
- * @engine: the engine
- *
- * Returns true if the last context to be executed on this engine, or has been
- * executed if the engine is already idle, is the kernel context
- * (#i915.kernel_context).
- */
-bool intel_engine_has_kernel_context(const struct intel_engine_cs *engine)
-{
-	const struct intel_context *kernel_context =
-		to_intel_context(engine->i915->kernel_context, engine);
-	struct i915_request *rq;
-
-	lockdep_assert_held(&engine->i915->drm.struct_mutex);
-
-	/*
-	 * Check the last context seen by the engine. If active, it will be
-	 * the last request that remains in the timeline. When idle, it is
-	 * the last executed context as tracked by retirement.
-	 */
-	rq = __i915_active_request_peek(&engine->timeline.last_request);
-	if (rq)
-		return rq->hw_context == kernel_context;
-	else
-		return engine->last_retired_context == kernel_context;
-}
-
 void intel_engines_reset_default_submission(struct drm_i915_private *i915)
 {
 	struct intel_engine_cs *engine;
@@ -1180,6 +1201,8 @@ void intel_engines_park(struct drm_i915_private *i915)
 		i915_gem_batch_pool_fini(&engine->batch_pool);
 		engine->execlists.no_priolist = false;
 	}
+
+	i915->gt.active_engines = 0;
 }
 
 /**
@@ -1283,15 +1306,14 @@ static void print_request(struct drm_printer *m,
 
 	x = print_sched_attr(rq->i915, &rq->sched.attr, buf, x, sizeof(buf));
 
-	drm_printf(m, "%s%x%s%s [%llx:%llx]%s @ %dms: %s\n",
+	drm_printf(m, "%s %llx:%llx%s%s %s @ %dms: %s\n",
 		   prefix,
-		   rq->global_seqno,
+		   rq->fence.context, rq->fence.seqno,
 		   i915_request_completed(rq) ? "!" :
 		   i915_request_started(rq) ? "*" :
 		   "",
 		   test_bit(DMA_FENCE_FLAG_ENABLE_SIGNAL_BIT,
 			    &rq->fence.flags) ?  "+" : "",
-		   rq->fence.context, rq->fence.seqno,
 		   buf,
 		   jiffies_to_msecs(jiffies - rq->emitted_jiffies),
 		   name);
@@ -1334,25 +1356,26 @@ static void intel_engine_print_registers(const struct intel_engine_cs *engine,
 		&engine->execlists;
 	u64 addr;
 
-	if (engine->id == RCS && IS_GEN_RANGE(dev_priv, 4, 7))
-		drm_printf(m, "\tCCID: 0x%08x\n", I915_READ(CCID));
+	if (engine->id == RCS0 && IS_GEN_RANGE(dev_priv, 4, 7))
+		drm_printf(m, "\tCCID: 0x%08x\n", ENGINE_READ(engine, CCID));
 	drm_printf(m, "\tRING_START: 0x%08x\n",
-		   I915_READ(RING_START(engine->mmio_base)));
+		   ENGINE_READ(engine, RING_START));
 	drm_printf(m, "\tRING_HEAD:  0x%08x\n",
-		   I915_READ(RING_HEAD(engine->mmio_base)) & HEAD_ADDR);
+		   ENGINE_READ(engine, RING_HEAD) & HEAD_ADDR);
 	drm_printf(m, "\tRING_TAIL:  0x%08x\n",
-		   I915_READ(RING_TAIL(engine->mmio_base)) & TAIL_ADDR);
+		   ENGINE_READ(engine, RING_TAIL) & TAIL_ADDR);
 	drm_printf(m, "\tRING_CTL:   0x%08x%s\n",
-		   I915_READ(RING_CTL(engine->mmio_base)),
-		   I915_READ(RING_CTL(engine->mmio_base)) & (RING_WAIT | RING_WAIT_SEMAPHORE) ? " [waiting]" : "");
+		   ENGINE_READ(engine, RING_CTL),
+		   ENGINE_READ(engine, RING_CTL) & (RING_WAIT | RING_WAIT_SEMAPHORE) ? " [waiting]" : "");
 	if (INTEL_GEN(engine->i915) > 2) {
 		drm_printf(m, "\tRING_MODE:  0x%08x%s\n",
-			   I915_READ(RING_MI_MODE(engine->mmio_base)),
-			   I915_READ(RING_MI_MODE(engine->mmio_base)) & (MODE_IDLE) ? " [idle]" : "");
+			   ENGINE_READ(engine, RING_MI_MODE),
+			   ENGINE_READ(engine, RING_MI_MODE) & (MODE_IDLE) ? " [idle]" : "");
 	}
 
 	if (INTEL_GEN(dev_priv) >= 6) {
-		drm_printf(m, "\tRING_IMR: %08x\n", I915_READ_IMR(engine));
+		drm_printf(m, "\tRING_IMR: %08x\n",
+			   ENGINE_READ(engine, RING_IMR));
 	}
 
 	addr = intel_engine_get_active_head(engine);
@@ -1362,57 +1385,53 @@ static void intel_engine_print_registers(const struct intel_engine_cs *engine,
 	drm_printf(m, "\tBBADDR: 0x%08x_%08x\n",
 		   upper_32_bits(addr), lower_32_bits(addr));
 	if (INTEL_GEN(dev_priv) >= 8)
-		addr = I915_READ64_2x32(RING_DMA_FADD(engine->mmio_base),
-					RING_DMA_FADD_UDW(engine->mmio_base));
+		addr = ENGINE_READ64(engine, RING_DMA_FADD, RING_DMA_FADD_UDW);
 	else if (INTEL_GEN(dev_priv) >= 4)
-		addr = I915_READ(RING_DMA_FADD(engine->mmio_base));
+		addr = ENGINE_READ(engine, RING_DMA_FADD);
 	else
-		addr = I915_READ(DMA_FADD_I8XX);
+		addr = ENGINE_READ(engine, DMA_FADD_I8XX);
 	drm_printf(m, "\tDMA_FADDR: 0x%08x_%08x\n",
 		   upper_32_bits(addr), lower_32_bits(addr));
 	if (INTEL_GEN(dev_priv) >= 4) {
 		drm_printf(m, "\tIPEIR: 0x%08x\n",
-			   I915_READ(RING_IPEIR(engine->mmio_base)));
+			   ENGINE_READ(engine, RING_IPEIR));
 		drm_printf(m, "\tIPEHR: 0x%08x\n",
-			   I915_READ(RING_IPEHR(engine->mmio_base)));
+			   ENGINE_READ(engine, RING_IPEHR));
 	} else {
-		drm_printf(m, "\tIPEIR: 0x%08x\n", I915_READ(IPEIR));
-		drm_printf(m, "\tIPEHR: 0x%08x\n", I915_READ(IPEHR));
+		drm_printf(m, "\tIPEIR: 0x%08x\n", ENGINE_READ(engine, IPEIR));
+		drm_printf(m, "\tIPEHR: 0x%08x\n", ENGINE_READ(engine, IPEHR));
 	}
 
 	if (HAS_EXECLISTS(dev_priv)) {
 		const u32 *hws =
 			&engine->status_page.addr[I915_HWS_CSB_BUF0_INDEX];
+		const u8 num_entries = execlists->csb_size;
 		unsigned int idx;
 		u8 read, write;
 
-		drm_printf(m, "\tExeclist status: 0x%08x %08x\n",
-			   I915_READ(RING_EXECLIST_STATUS_LO(engine)),
-			   I915_READ(RING_EXECLIST_STATUS_HI(engine)));
+		drm_printf(m, "\tExeclist status: 0x%08x %08x, entries %u\n",
+			   ENGINE_READ(engine, RING_EXECLIST_STATUS_LO),
+			   ENGINE_READ(engine, RING_EXECLIST_STATUS_HI),
+			   num_entries);
 
 		read = execlists->csb_head;
 		write = READ_ONCE(*execlists->csb_write);
 
-		drm_printf(m, "\tExeclist CSB read %d, write %d [mmio:%d], tasklet queued? %s (%s)\n",
+		drm_printf(m, "\tExeclist CSB read %d, write %d, tasklet queued? %s (%s)\n",
 			   read, write,
-			   GEN8_CSB_WRITE_PTR(I915_READ(RING_CONTEXT_STATUS_PTR(engine))),
 			   yesno(test_bit(TASKLET_STATE_SCHED,
 					  &engine->execlists.tasklet.state)),
 			   enableddisabled(!atomic_read(&engine->execlists.tasklet.count)));
-		if (read >= GEN8_CSB_ENTRIES)
+		if (read >= num_entries)
 			read = 0;
-		if (write >= GEN8_CSB_ENTRIES)
+		if (write >= num_entries)
 			write = 0;
 		if (read > write)
-			write += GEN8_CSB_ENTRIES;
+			write += num_entries;
 		while (read < write) {
-			idx = ++read % GEN8_CSB_ENTRIES;
-			drm_printf(m, "\tExeclist CSB[%d]: 0x%08x [mmio:0x%08x], context: %d [mmio:%d]\n",
-				   idx,
-				   hws[idx * 2],
-				   I915_READ(RING_CONTEXT_STATUS_BUF_LO(engine, idx)),
-				   hws[idx * 2 + 1],
-				   I915_READ(RING_CONTEXT_STATUS_BUF_HI(engine, idx)));
+			idx = ++read % num_entries;
+			drm_printf(m, "\tExeclist CSB[%d]: 0x%08x, context: %d\n",
+				   idx, hws[idx * 2], hws[idx * 2 + 1]);
 		}
 
 		rcu_read_lock();
@@ -1425,10 +1444,11 @@ static void intel_engine_print_registers(const struct intel_engine_cs *engine,
 				char hdr[80];
 
 				snprintf(hdr, sizeof(hdr),
-					 "\t\tELSP[%d] count=%d, ring:{start:%08x, hwsp:%08x}, rq: ",
+					 "\t\tELSP[%d] count=%d, ring:{start:%08x, hwsp:%08x, seqno:%08x}, rq: ",
 					 idx, count,
 					 i915_ggtt_offset(rq->ring->vma),
-					 rq->timeline->hwsp_offset);
+					 rq->timeline->hwsp_offset,
+					 hwsp_seqno(rq));
 				print_request(m, rq, hdr);
 			} else {
 				drm_printf(m, "\t\tELSP[%d] idle\n", idx);
@@ -1438,11 +1458,11 @@ static void intel_engine_print_registers(const struct intel_engine_cs *engine,
 		rcu_read_unlock();
 	} else if (INTEL_GEN(dev_priv) > 6) {
 		drm_printf(m, "\tPP_DIR_BASE: 0x%08x\n",
-			   I915_READ(RING_PP_DIR_BASE(engine)));
+			   ENGINE_READ(engine, RING_PP_DIR_BASE));
 		drm_printf(m, "\tPP_DIR_BASE_READ: 0x%08x\n",
-			   I915_READ(RING_PP_DIR_BASE_READ(engine)));
+			   ENGINE_READ(engine, RING_PP_DIR_BASE_READ));
 		drm_printf(m, "\tPP_DIR_DCLV: 0x%08x\n",
-			   I915_READ(RING_PP_DIR_DCLV(engine)));
+			   ENGINE_READ(engine, RING_PP_DIR_DCLV));
 	}
 }
 
@@ -1495,13 +1515,12 @@ void intel_engine_dump(struct intel_engine_cs *engine,
 		va_end(ap);
 	}
 
-	if (i915_terminally_wedged(&engine->i915->gpu_error))
+	if (i915_reset_failed(engine->i915))
 		drm_printf(m, "*** WEDGED ***\n");
 
-	drm_printf(m, "\tcurrent seqno %x, last %x, hangcheck %x [%d ms]\n",
-		   intel_engine_get_seqno(engine),
-		   intel_engine_last_submit(engine),
-		   engine->hangcheck.seqno,
+	drm_printf(m, "\tHangcheck %x:%x [%d ms]\n",
+		   engine->hangcheck.last_seqno,
+		   engine->hangcheck.next_seqno,
 		   jiffies_to_msecs(jiffies - engine->hangcheck.action_timestamp));
 	drm_printf(m, "\tReset count: %d (global %d)\n",
 		   i915_reset_engine_count(error, engine),
@@ -1521,7 +1540,7 @@ void intel_engine_dump(struct intel_engine_cs *engine,
 	if (&rq->link != &engine->timeline.requests)
 		print_request(m, rq, "\t\tlast   ");
 
-	rq = i915_gem_find_active_request(engine);
+	rq = intel_engine_find_active_request(engine);
 	if (rq) {
 		print_request(m, rq, "\t\tactive ");
 
@@ -1688,6 +1707,50 @@ void intel_disable_engine_stats(struct intel_engine_cs *engine)
 	write_sequnlock_irqrestore(&engine->stats.lock, flags);
 }
 
+static bool match_ring(struct i915_request *rq)
+{
+	u32 ring = ENGINE_READ(rq->engine, RING_START);
+
+	return ring == i915_ggtt_offset(rq->ring->vma);
+}
+
+struct i915_request *
+intel_engine_find_active_request(struct intel_engine_cs *engine)
+{
+	struct i915_request *request, *active = NULL;
+	unsigned long flags;
+
+	/*
+	 * We are called by the error capture, reset and to dump engine
+	 * state at random points in time. In particular, note that neither is
+	 * crucially ordered with an interrupt. After a hang, the GPU is dead
+	 * and we assume that no more writes can happen (we waited long enough
+	 * for all writes that were in transaction to be flushed) - adding an
+	 * extra delay for a recent interrupt is pointless. Hence, we do
+	 * not need an engine->irq_seqno_barrier() before the seqno reads.
+	 * At all other times, we must assume the GPU is still running, but
+	 * we only care about the snapshot of this moment.
+	 */
+	spin_lock_irqsave(&engine->timeline.lock, flags);
+	list_for_each_entry(request, &engine->timeline.requests, link) {
+		if (i915_request_completed(request))
+			continue;
+
+		if (!i915_request_started(request))
+			break;
+
+		/* More than one preemptible request may match! */
+		if (!match_ring(request))
+			break;
+
+		active = request;
+		break;
+	}
+	spin_unlock_irqrestore(&engine->timeline.lock, flags);
+
+	return active;
+}
+
 #if IS_ENABLED(CONFIG_DRM_I915_SELFTEST)
 #include "selftests/mock_engine.c"
 #include "selftests/intel_engine_cs.c"
diff --git a/drivers/gpu/drm/i915/intel_engine_types.h b/drivers/gpu/drm/i915/intel_engine_types.h
new file mode 100644
index 000000000000..1f970c76b6a6
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_engine_types.h
@@ -0,0 +1,546 @@
+/*
+ * SPDX-License-Identifier: MIT
+ *
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_ENGINE_TYPES__
+#define __INTEL_ENGINE_TYPES__
+
+#include <linux/hashtable.h>
+#include <linux/irq_work.h>
+#include <linux/kref.h>
+#include <linux/list.h>
+#include <linux/types.h>
+
+#include "i915_gem.h"
+#include "i915_priolist_types.h"
+#include "i915_selftest.h"
+#include "i915_timeline_types.h"
+#include "intel_workarounds_types.h"
+
+#include "i915_gem_batch_pool.h"
+#include "i915_pmu.h"
+
+#define I915_MAX_SLICES	3
+#define I915_MAX_SUBSLICES 8
+
+#define I915_CMD_HASH_ORDER 9
+
+struct dma_fence;
+struct drm_i915_reg_table;
+struct i915_gem_context;
+struct i915_request;
+struct i915_sched_attr;
+struct intel_uncore;
+
+typedef u8 intel_engine_mask_t;
+#define ALL_ENGINES ((intel_engine_mask_t)~0ul)
+
+struct intel_hw_status_page {
+	struct i915_vma *vma;
+	u32 *addr;
+};
+
+struct intel_instdone {
+	u32 instdone;
+	/* The following exist only in the RCS engine */
+	u32 slice_common;
+	u32 sampler[I915_MAX_SLICES][I915_MAX_SUBSLICES];
+	u32 row[I915_MAX_SLICES][I915_MAX_SUBSLICES];
+};
+
+struct intel_engine_hangcheck {
+	u64 acthd;
+	u32 last_seqno;
+	u32 next_seqno;
+	unsigned long action_timestamp;
+	struct intel_instdone instdone;
+};
+
+struct intel_ring {
+	struct kref ref;
+	struct i915_vma *vma;
+	void *vaddr;
+
+	struct i915_timeline *timeline;
+	struct list_head request_list;
+	struct list_head active_link;
+
+	u32 head;
+	u32 tail;
+	u32 emit;
+
+	u32 space;
+	u32 size;
+	u32 effective_size;
+};
+
+/*
+ * we use a single page to load ctx workarounds so all of these
+ * values are referred in terms of dwords
+ *
+ * struct i915_wa_ctx_bb:
+ *  offset: specifies batch starting position, also helpful in case
+ *    if we want to have multiple batches at different offsets based on
+ *    some criteria. It is not a requirement at the moment but provides
+ *    an option for future use.
+ *  size: size of the batch in DWORDS
+ */
+struct i915_ctx_workarounds {
+	struct i915_wa_ctx_bb {
+		u32 offset;
+		u32 size;
+	} indirect_ctx, per_ctx;
+	struct i915_vma *vma;
+};
+
+#define I915_MAX_VCS	4
+#define I915_MAX_VECS	2
+
+/*
+ * Engine IDs definitions.
+ * Keep instances of the same type engine together.
+ */
+enum intel_engine_id {
+	RCS0 = 0,
+	BCS0,
+	VCS0,
+	VCS1,
+	VCS2,
+	VCS3,
+#define _VCS(n) (VCS0 + (n))
+	VECS0,
+	VECS1,
+#define _VECS(n) (VECS0 + (n))
+	I915_NUM_ENGINES
+};
+
+struct st_preempt_hang {
+	struct completion completion;
+	unsigned int count;
+	bool inject_hang;
+};
+
+/**
+ * struct intel_engine_execlists - execlist submission queue and port state
+ *
+ * The struct intel_engine_execlists represents the combined logical state of
+ * driver and the hardware state for execlist mode of submission.
+ */
+struct intel_engine_execlists {
+	/**
+	 * @tasklet: softirq tasklet for bottom handler
+	 */
+	struct tasklet_struct tasklet;
+
+	/**
+	 * @default_priolist: priority list for I915_PRIORITY_NORMAL
+	 */
+	struct i915_priolist default_priolist;
+
+	/**
+	 * @no_priolist: priority lists disabled
+	 */
+	bool no_priolist;
+
+	/**
+	 * @submit_reg: gen-specific execlist submission register
+	 * set to the ExecList Submission Port (elsp) register pre-Gen11 and to
+	 * the ExecList Submission Queue Contents register array for Gen11+
+	 */
+	u32 __iomem *submit_reg;
+
+	/**
+	 * @ctrl_reg: the enhanced execlists control register, used to load the
+	 * submit queue on the HW and to request preemptions to idle
+	 */
+	u32 __iomem *ctrl_reg;
+
+	/**
+	 * @port: execlist port states
+	 *
+	 * For each hardware ELSP (ExecList Submission Port) we keep
+	 * track of the last request and the number of times we submitted
+	 * that port to hw. We then count the number of times the hw reports
+	 * a context completion or preemption. As only one context can
+	 * be active on hw, we limit resubmission of context to port[0]. This
+	 * is called Lite Restore, of the context.
+	 */
+	struct execlist_port {
+		/**
+		 * @request_count: combined request and submission count
+		 */
+		struct i915_request *request_count;
+#define EXECLIST_COUNT_BITS 2
+#define port_request(p) ptr_mask_bits((p)->request_count, EXECLIST_COUNT_BITS)
+#define port_count(p) ptr_unmask_bits((p)->request_count, EXECLIST_COUNT_BITS)
+#define port_pack(rq, count) ptr_pack_bits(rq, count, EXECLIST_COUNT_BITS)
+#define port_unpack(p, count) ptr_unpack_bits((p)->request_count, count, EXECLIST_COUNT_BITS)
+#define port_set(p, packed) ((p)->request_count = (packed))
+#define port_isset(p) ((p)->request_count)
+#define port_index(p, execlists) ((p) - (execlists)->port)
+
+		/**
+		 * @context_id: context ID for port
+		 */
+		GEM_DEBUG_DECL(u32 context_id);
+
+#define EXECLIST_MAX_PORTS 2
+	} port[EXECLIST_MAX_PORTS];
+
+	/**
+	 * @active: is the HW active? We consider the HW as active after
+	 * submitting any context for execution and until we have seen the
+	 * last context completion event. After that, we do not expect any
+	 * more events until we submit, and so can park the HW.
+	 *
+	 * As we have a small number of different sources from which we feed
+	 * the HW, we track the state of each inside a single bitfield.
+	 */
+	unsigned int active;
+#define EXECLISTS_ACTIVE_USER 0
+#define EXECLISTS_ACTIVE_PREEMPT 1
+#define EXECLISTS_ACTIVE_HWACK 2
+
+	/**
+	 * @port_mask: number of execlist ports - 1
+	 */
+	unsigned int port_mask;
+
+	/**
+	 * @queue_priority_hint: Highest pending priority.
+	 *
+	 * When we add requests into the queue, or adjust the priority of
+	 * executing requests, we compute the maximum priority of those
+	 * pending requests. We can then use this value to determine if
+	 * we need to preempt the executing requests to service the queue.
+	 * However, since the we may have recorded the priority of an inflight
+	 * request we wanted to preempt but since completed, at the time of
+	 * dequeuing the priority hint may no longer may match the highest
+	 * available request priority.
+	 */
+	int queue_priority_hint;
+
+	/**
+	 * @queue: queue of requests, in priority lists
+	 */
+	struct rb_root_cached queue;
+
+	/**
+	 * @csb_write: control register for Context Switch buffer
+	 *
+	 * Note this register may be either mmio or HWSP shadow.
+	 */
+	u32 *csb_write;
+
+	/**
+	 * @csb_status: status array for Context Switch buffer
+	 *
+	 * Note these register may be either mmio or HWSP shadow.
+	 */
+	u32 *csb_status;
+
+	/**
+	 * @preempt_complete_status: expected CSB upon completing preemption
+	 */
+	u32 preempt_complete_status;
+
+	/**
+	 * @csb_size: context status buffer FIFO size
+	 */
+	u8 csb_size;
+
+	/**
+	 * @csb_head: context status buffer head
+	 */
+	u8 csb_head;
+
+	I915_SELFTEST_DECLARE(struct st_preempt_hang preempt_hang;)
+};
+
+#define INTEL_ENGINE_CS_MAX_NAME 8
+
+struct intel_engine_cs {
+	struct drm_i915_private *i915;
+	struct intel_uncore *uncore;
+	char name[INTEL_ENGINE_CS_MAX_NAME];
+
+	enum intel_engine_id id;
+	unsigned int hw_id;
+	unsigned int guc_id;
+	intel_engine_mask_t mask;
+
+	u8 uabi_class;
+
+	u8 class;
+	u8 instance;
+	u32 context_size;
+	u32 mmio_base;
+
+	struct intel_ring *buffer;
+
+	struct i915_timeline timeline;
+
+	struct intel_context *kernel_context; /* pinned */
+	struct intel_context *preempt_context; /* pinned; optional */
+
+	struct drm_i915_gem_object *default_state;
+	void *pinned_default_state;
+
+	/* Rather than have every client wait upon all user interrupts,
+	 * with the herd waking after every interrupt and each doing the
+	 * heavyweight seqno dance, we delegate the task (of being the
+	 * bottom-half of the user interrupt) to the first client. After
+	 * every interrupt, we wake up one client, who does the heavyweight
+	 * coherent seqno read and either goes back to sleep (if incomplete),
+	 * or wakes up all the completed clients in parallel, before then
+	 * transferring the bottom-half status to the next client in the queue.
+	 *
+	 * Compared to walking the entire list of waiters in a single dedicated
+	 * bottom-half, we reduce the latency of the first waiter by avoiding
+	 * a context switch, but incur additional coherent seqno reads when
+	 * following the chain of request breadcrumbs. Since it is most likely
+	 * that we have a single client waiting on each seqno, then reducing
+	 * the overhead of waking that client is much preferred.
+	 */
+	struct intel_breadcrumbs {
+		spinlock_t irq_lock;
+		struct list_head signalers;
+
+		struct irq_work irq_work; /* for use from inside irq_lock */
+
+		unsigned int irq_enabled;
+
+		bool irq_armed;
+	} breadcrumbs;
+
+	struct intel_engine_pmu {
+		/**
+		 * @enable: Bitmask of enable sample events on this engine.
+		 *
+		 * Bits correspond to sample event types, for instance
+		 * I915_SAMPLE_QUEUED is bit 0 etc.
+		 */
+		u32 enable;
+		/**
+		 * @enable_count: Reference count for the enabled samplers.
+		 *
+		 * Index number corresponds to @enum drm_i915_pmu_engine_sample.
+		 */
+		unsigned int enable_count[I915_ENGINE_SAMPLE_COUNT];
+		/**
+		 * @sample: Counter values for sampling events.
+		 *
+		 * Our internal timer stores the current counters in this field.
+		 *
+		 * Index number corresponds to @enum drm_i915_pmu_engine_sample.
+		 */
+		struct i915_pmu_sample sample[I915_ENGINE_SAMPLE_COUNT];
+	} pmu;
+
+	/*
+	 * A pool of objects to use as shadow copies of client batch buffers
+	 * when the command parser is enabled. Prevents the client from
+	 * modifying the batch contents after software parsing.
+	 */
+	struct i915_gem_batch_pool batch_pool;
+
+	struct intel_hw_status_page status_page;
+	struct i915_ctx_workarounds wa_ctx;
+	struct i915_wa_list ctx_wa_list;
+	struct i915_wa_list wa_list;
+	struct i915_wa_list whitelist;
+
+	u32             irq_keep_mask; /* always keep these interrupts */
+	u32		irq_enable_mask; /* bitmask to enable ring interrupt */
+	void		(*irq_enable)(struct intel_engine_cs *engine);
+	void		(*irq_disable)(struct intel_engine_cs *engine);
+
+	int		(*init_hw)(struct intel_engine_cs *engine);
+
+	struct {
+		void (*prepare)(struct intel_engine_cs *engine);
+		void (*reset)(struct intel_engine_cs *engine, bool stalled);
+		void (*finish)(struct intel_engine_cs *engine);
+	} reset;
+
+	void		(*park)(struct intel_engine_cs *engine);
+	void		(*unpark)(struct intel_engine_cs *engine);
+
+	void		(*set_default_submission)(struct intel_engine_cs *engine);
+
+	const struct intel_context_ops *cops;
+
+	int		(*request_alloc)(struct i915_request *rq);
+	int		(*init_context)(struct i915_request *rq);
+
+	int		(*emit_flush)(struct i915_request *request, u32 mode);
+#define EMIT_INVALIDATE	BIT(0)
+#define EMIT_FLUSH	BIT(1)
+#define EMIT_BARRIER	(EMIT_INVALIDATE | EMIT_FLUSH)
+	int		(*emit_bb_start)(struct i915_request *rq,
+					 u64 offset, u32 length,
+					 unsigned int dispatch_flags);
+#define I915_DISPATCH_SECURE BIT(0)
+#define I915_DISPATCH_PINNED BIT(1)
+	int		 (*emit_init_breadcrumb)(struct i915_request *rq);
+	u32		*(*emit_fini_breadcrumb)(struct i915_request *rq,
+						 u32 *cs);
+	unsigned int	emit_fini_breadcrumb_dw;
+
+	/* Pass the request to the hardware queue (e.g. directly into
+	 * the legacy ringbuffer or to the end of an execlist).
+	 *
+	 * This is called from an atomic context with irqs disabled; must
+	 * be irq safe.
+	 */
+	void		(*submit_request)(struct i915_request *rq);
+
+	/*
+	 * Call when the priority on a request has changed and it and its
+	 * dependencies may need rescheduling. Note the request itself may
+	 * not be ready to run!
+	 */
+	void		(*schedule)(struct i915_request *request,
+				    const struct i915_sched_attr *attr);
+
+	/*
+	 * Cancel all requests on the hardware, or queued for execution.
+	 * This should only cancel the ready requests that have been
+	 * submitted to the engine (via the engine->submit_request callback).
+	 * This is called when marking the device as wedged.
+	 */
+	void		(*cancel_requests)(struct intel_engine_cs *engine);
+
+	void		(*cleanup)(struct intel_engine_cs *engine);
+
+	struct intel_engine_execlists execlists;
+
+	/* Contexts are pinned whilst they are active on the GPU. The last
+	 * context executed remains active whilst the GPU is idle - the
+	 * switch away and write to the context object only occurs on the
+	 * next execution.  Contexts are only unpinned on retirement of the
+	 * following request ensuring that we can always write to the object
+	 * on the context switch even after idling. Across suspend, we switch
+	 * to the kernel context and trash it as the save may not happen
+	 * before the hardware is powered down.
+	 */
+	struct intel_context *last_retired_context;
+
+	/* status_notifier: list of callbacks for context-switch changes */
+	struct atomic_notifier_head context_status_notifier;
+
+	struct intel_engine_hangcheck hangcheck;
+
+#define I915_ENGINE_NEEDS_CMD_PARSER BIT(0)
+#define I915_ENGINE_SUPPORTS_STATS   BIT(1)
+#define I915_ENGINE_HAS_PREEMPTION   BIT(2)
+#define I915_ENGINE_HAS_SEMAPHORES   BIT(3)
+#define I915_ENGINE_NEEDS_BREADCRUMB_TASKLET BIT(4)
+	unsigned int flags;
+
+	/*
+	 * Table of commands the command parser needs to know about
+	 * for this engine.
+	 */
+	DECLARE_HASHTABLE(cmd_hash, I915_CMD_HASH_ORDER);
+
+	/*
+	 * Table of registers allowed in commands that read/write registers.
+	 */
+	const struct drm_i915_reg_table *reg_tables;
+	int reg_table_count;
+
+	/*
+	 * Returns the bitmask for the length field of the specified command.
+	 * Return 0 for an unrecognized/invalid command.
+	 *
+	 * If the command parser finds an entry for a command in the engine's
+	 * cmd_tables, it gets the command's length based on the table entry.
+	 * If not, it calls this function to determine the per-engine length
+	 * field encoding for the command (i.e. different opcode ranges use
+	 * certain bits to encode the command length in the header).
+	 */
+	u32 (*get_cmd_length_mask)(u32 cmd_header);
+
+	struct {
+		/**
+		 * @lock: Lock protecting the below fields.
+		 */
+		seqlock_t lock;
+		/**
+		 * @enabled: Reference count indicating number of listeners.
+		 */
+		unsigned int enabled;
+		/**
+		 * @active: Number of contexts currently scheduled in.
+		 */
+		unsigned int active;
+		/**
+		 * @enabled_at: Timestamp when busy stats were enabled.
+		 */
+		ktime_t enabled_at;
+		/**
+		 * @start: Timestamp of the last idle to active transition.
+		 *
+		 * Idle is defined as active == 0, active is active > 0.
+		 */
+		ktime_t start;
+		/**
+		 * @total: Total time this engine was busy.
+		 *
+		 * Accumulated time not counting the most recent block in cases
+		 * where engine is currently busy (active > 0).
+		 */
+		ktime_t total;
+	} stats;
+};
+
+static inline bool
+intel_engine_needs_cmd_parser(const struct intel_engine_cs *engine)
+{
+	return engine->flags & I915_ENGINE_NEEDS_CMD_PARSER;
+}
+
+static inline bool
+intel_engine_supports_stats(const struct intel_engine_cs *engine)
+{
+	return engine->flags & I915_ENGINE_SUPPORTS_STATS;
+}
+
+static inline bool
+intel_engine_has_preemption(const struct intel_engine_cs *engine)
+{
+	return engine->flags & I915_ENGINE_HAS_PREEMPTION;
+}
+
+static inline bool
+intel_engine_has_semaphores(const struct intel_engine_cs *engine)
+{
+	return engine->flags & I915_ENGINE_HAS_SEMAPHORES;
+}
+
+static inline bool
+intel_engine_needs_breadcrumb_tasklet(const struct intel_engine_cs *engine)
+{
+	return engine->flags & I915_ENGINE_NEEDS_BREADCRUMB_TASKLET;
+}
+
+#define instdone_slice_mask(dev_priv__) \
+	(IS_GEN(dev_priv__, 7) ? \
+	 1 : RUNTIME_INFO(dev_priv__)->sseu.slice_mask)
+
+#define instdone_subslice_mask(dev_priv__) \
+	(IS_GEN(dev_priv__, 7) ? \
+	 1 : RUNTIME_INFO(dev_priv__)->sseu.subslice_mask[0])
+
+#define for_each_instdone_slice_subslice(dev_priv__, slice__, subslice__) \
+	for ((slice__) = 0, (subslice__) = 0; \
+	     (slice__) < I915_MAX_SLICES; \
+	     (subslice__) = ((subslice__) + 1) < I915_MAX_SUBSLICES ? (subslice__) + 1 : 0, \
+	       (slice__) += ((subslice__) == 0)) \
+		for_each_if((BIT(slice__) & instdone_slice_mask(dev_priv__)) && \
+			    (BIT(subslice__) & instdone_subslice_mask(dev_priv__)))
+
+#endif /* __INTEL_ENGINE_TYPES_H__ */
diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c
index 656e684e7c9a..c805a0966395 100644
--- a/drivers/gpu/drm/i915/intel_fbc.c
+++ b/drivers/gpu/drm/i915/intel_fbc.c
@@ -40,8 +40,10 @@
 
 #include <drm/drm_fourcc.h>
 
-#include "intel_drv.h"
 #include "i915_drv.h"
+#include "intel_drv.h"
+#include "intel_fbc.h"
+#include "intel_frontbuffer.h"
 
 static inline bool fbc_supported(struct drm_i915_private *dev_priv)
 {
@@ -108,7 +110,7 @@ static void i8xx_fbc_deactivate(struct drm_i915_private *dev_priv)
 	I915_WRITE(FBC_CONTROL, fbc_ctl);
 
 	/* Wait for compressing bit to clear */
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    FBC_STATUS, FBC_STAT_COMPRESSING, 0,
 				    10)) {
 		DRM_DEBUG_KMS("FBC idle timed out\n");
diff --git a/drivers/gpu/drm/i915/intel_fbc.h b/drivers/gpu/drm/i915/intel_fbc.h
new file mode 100644
index 000000000000..50272eda8d43
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_fbc.h
@@ -0,0 +1,42 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_FBC_H__
+#define __INTEL_FBC_H__
+
+#include <linux/types.h>
+
+#include "intel_frontbuffer.h"
+
+struct drm_i915_private;
+struct intel_atomic_state;
+struct intel_crtc;
+struct intel_crtc_state;
+struct intel_plane_state;
+
+void intel_fbc_choose_crtc(struct drm_i915_private *dev_priv,
+			   struct intel_atomic_state *state);
+bool intel_fbc_is_active(struct drm_i915_private *dev_priv);
+void intel_fbc_pre_update(struct intel_crtc *crtc,
+			  struct intel_crtc_state *crtc_state,
+			  struct intel_plane_state *plane_state);
+void intel_fbc_post_update(struct intel_crtc *crtc);
+void intel_fbc_init(struct drm_i915_private *dev_priv);
+void intel_fbc_init_pipe_state(struct drm_i915_private *dev_priv);
+void intel_fbc_enable(struct intel_crtc *crtc,
+		      struct intel_crtc_state *crtc_state,
+		      struct intel_plane_state *plane_state);
+void intel_fbc_disable(struct intel_crtc *crtc);
+void intel_fbc_global_disable(struct drm_i915_private *dev_priv);
+void intel_fbc_invalidate(struct drm_i915_private *dev_priv,
+			  unsigned int frontbuffer_bits,
+			  enum fb_op_origin origin);
+void intel_fbc_flush(struct drm_i915_private *dev_priv,
+		     unsigned int frontbuffer_bits, enum fb_op_origin origin);
+void intel_fbc_cleanup_cfb(struct drm_i915_private *dev_priv);
+void intel_fbc_handle_fifo_underrun_irq(struct drm_i915_private *dev_priv);
+int intel_fbc_reset_underrun(struct drm_i915_private *dev_priv);
+
+#endif /* __INTEL_FBC_H__ */
diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c
index 376ffe842e26..89db71996148 100644
--- a/drivers/gpu/drm/i915/intel_fbdev.c
+++ b/drivers/gpu/drm/i915/intel_fbdev.c
@@ -25,26 +25,27 @@
  */
 
 #include <linux/async.h>
-#include <linux/module.h>
-#include <linux/kernel.h>
 #include <linux/console.h>
+#include <linux/delay.h>
 #include <linux/errno.h>
-#include <linux/string.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
 #include <linux/mm.h>
-#include <linux/tty.h>
+#include <linux/module.h>
+#include <linux/string.h>
 #include <linux/sysrq.h>
-#include <linux/delay.h>
-#include <linux/init.h>
+#include <linux/tty.h>
 #include <linux/vga_switcheroo.h>
 
 #include <drm/drm_crtc.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_fourcc.h>
+#include <drm/i915_drm.h>
 
+#include "i915_drv.h"
 #include "intel_drv.h"
+#include "intel_fbdev.h"
 #include "intel_frontbuffer.h"
-#include <drm/i915_drm.h>
-#include "i915_drv.h"
 
 static void intel_fbdev_invalidate(struct intel_fbdev *ifbdev)
 {
@@ -235,12 +236,8 @@ static int intelfb_create(struct drm_fb_helper *helper,
 		goto out_unpin;
 	}
 
-	info->par = helper;
-
 	ifbdev->helper.fb = fb;
 
-	strcpy(info->fix.id, "inteldrmfb");
-
 	info->fbops = &intelfb_ops;
 
 	/* setup aperture base/size for vesafb takeover */
@@ -259,11 +256,7 @@ static int intelfb_create(struct drm_fb_helper *helper,
 	info->screen_base = vaddr;
 	info->screen_size = vma->node.size;
 
-	/* This driver doesn't need a VT switch to restore the mode on resume */
-	info->skip_vt_switch = true;
-
-	drm_fb_helper_fill_fix(info, fb->pitches[0], fb->format->depth);
-	drm_fb_helper_fill_var(info, &ifbdev->helper, sizes->fb_width, sizes->fb_height);
+	drm_fb_helper_fill_info(info, &ifbdev->helper, sizes);
 
 	/* If the object is shmemfs backed, it will have given us zeroed pages.
 	 * If the object is stolen however, it will be full of whatever
@@ -292,223 +285,7 @@ out_unlock:
 	return ret;
 }
 
-static struct drm_fb_helper_crtc *
-intel_fb_helper_crtc(struct drm_fb_helper *fb_helper, struct drm_crtc *crtc)
-{
-	int i;
-
-	for (i = 0; i < fb_helper->crtc_count; i++)
-		if (fb_helper->crtc_info[i].mode_set.crtc == crtc)
-			return &fb_helper->crtc_info[i];
-
-	return NULL;
-}
-
-/*
- * Try to read the BIOS display configuration and use it for the initial
- * fb configuration.
- *
- * The BIOS or boot loader will generally create an initial display
- * configuration for us that includes some set of active pipes and displays.
- * This routine tries to figure out which pipes and connectors are active
- * and stuffs them into the crtcs and modes array given to us by the
- * drm_fb_helper code.
- *
- * The overall sequence is:
- *   intel_fbdev_init - from driver load
- *     intel_fbdev_init_bios - initialize the intel_fbdev using BIOS data
- *     drm_fb_helper_init - build fb helper structs
- *     drm_fb_helper_single_add_all_connectors - more fb helper structs
- *   intel_fbdev_initial_config - apply the config
- *     drm_fb_helper_initial_config - call ->probe then register_framebuffer()
- *         drm_setup_crtcs - build crtc config for fbdev
- *           intel_fb_initial_config - find active connectors etc
- *         drm_fb_helper_single_fb_probe - set up fbdev
- *           intelfb_create - re-use or alloc fb, build out fbdev structs
- *
- * Note that we don't make special consideration whether we could actually
- * switch to the selected modes without a full modeset. E.g. when the display
- * is in VGA mode we need to recalculate watermarks and set a new high-res
- * framebuffer anyway.
- */
-static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper,
-				    struct drm_fb_helper_crtc **crtcs,
-				    struct drm_display_mode **modes,
-				    struct drm_fb_offset *offsets,
-				    bool *enabled, int width, int height)
-{
-	struct drm_i915_private *dev_priv = to_i915(fb_helper->dev);
-	unsigned long conn_configured, conn_seq, mask;
-	unsigned int count = min(fb_helper->connector_count, BITS_PER_LONG);
-	int i, j;
-	bool *save_enabled;
-	bool fallback = true, ret = true;
-	int num_connectors_enabled = 0;
-	int num_connectors_detected = 0;
-	struct drm_modeset_acquire_ctx ctx;
-
-	save_enabled = kcalloc(count, sizeof(bool), GFP_KERNEL);
-	if (!save_enabled)
-		return false;
-
-	drm_modeset_acquire_init(&ctx, 0);
-
-	while (drm_modeset_lock_all_ctx(fb_helper->dev, &ctx) != 0)
-		drm_modeset_backoff(&ctx);
-
-	memcpy(save_enabled, enabled, count);
-	mask = GENMASK(count - 1, 0);
-	conn_configured = 0;
-retry:
-	conn_seq = conn_configured;
-	for (i = 0; i < count; i++) {
-		struct drm_fb_helper_connector *fb_conn;
-		struct drm_connector *connector;
-		struct drm_encoder *encoder;
-		struct drm_fb_helper_crtc *new_crtc;
-
-		fb_conn = fb_helper->connector_info[i];
-		connector = fb_conn->connector;
-
-		if (conn_configured & BIT(i))
-			continue;
-
-		if (conn_seq == 0 && !connector->has_tile)
-			continue;
-
-		if (connector->status == connector_status_connected)
-			num_connectors_detected++;
-
-		if (!enabled[i]) {
-			DRM_DEBUG_KMS("connector %s not enabled, skipping\n",
-				      connector->name);
-			conn_configured |= BIT(i);
-			continue;
-		}
-
-		if (connector->force == DRM_FORCE_OFF) {
-			DRM_DEBUG_KMS("connector %s is disabled by user, skipping\n",
-				      connector->name);
-			enabled[i] = false;
-			continue;
-		}
-
-		encoder = connector->state->best_encoder;
-		if (!encoder || WARN_ON(!connector->state->crtc)) {
-			if (connector->force > DRM_FORCE_OFF)
-				goto bail;
-
-			DRM_DEBUG_KMS("connector %s has no encoder or crtc, skipping\n",
-				      connector->name);
-			enabled[i] = false;
-			conn_configured |= BIT(i);
-			continue;
-		}
-
-		num_connectors_enabled++;
-
-		new_crtc = intel_fb_helper_crtc(fb_helper,
-						connector->state->crtc);
-
-		/*
-		 * Make sure we're not trying to drive multiple connectors
-		 * with a single CRTC, since our cloning support may not
-		 * match the BIOS.
-		 */
-		for (j = 0; j < count; j++) {
-			if (crtcs[j] == new_crtc) {
-				DRM_DEBUG_KMS("fallback: cloned configuration\n");
-				goto bail;
-			}
-		}
-
-		DRM_DEBUG_KMS("looking for cmdline mode on connector %s\n",
-			      connector->name);
-
-		/* go for command line mode first */
-		modes[i] = drm_pick_cmdline_mode(fb_conn);
-
-		/* try for preferred next */
-		if (!modes[i]) {
-			DRM_DEBUG_KMS("looking for preferred mode on connector %s %d\n",
-				      connector->name, connector->has_tile);
-			modes[i] = drm_has_preferred_mode(fb_conn, width,
-							  height);
-		}
-
-		/* No preferred mode marked by the EDID? Are there any modes? */
-		if (!modes[i] && !list_empty(&connector->modes)) {
-			DRM_DEBUG_KMS("using first mode listed on connector %s\n",
-				      connector->name);
-			modes[i] = list_first_entry(&connector->modes,
-						    struct drm_display_mode,
-						    head);
-		}
-
-		/* last resort: use current mode */
-		if (!modes[i]) {
-			/*
-			 * IMPORTANT: We want to use the adjusted mode (i.e.
-			 * after the panel fitter upscaling) as the initial
-			 * config, not the input mode, which is what crtc->mode
-			 * usually contains. But since our current
-			 * code puts a mode derived from the post-pfit timings
-			 * into crtc->mode this works out correctly.
-			 *
-			 * This is crtc->mode and not crtc->state->mode for the
-			 * fastboot check to work correctly. crtc_state->mode has
-			 * I915_MODE_FLAG_INHERITED, which we clear to force check
-			 * state.
-			 */
-			DRM_DEBUG_KMS("looking for current mode on connector %s\n",
-				      connector->name);
-			modes[i] = &connector->state->crtc->mode;
-		}
-		crtcs[i] = new_crtc;
-
-		DRM_DEBUG_KMS("connector %s on [CRTC:%d:%s]: %dx%d%s\n",
-			      connector->name,
-			      connector->state->crtc->base.id,
-			      connector->state->crtc->name,
-			      modes[i]->hdisplay, modes[i]->vdisplay,
-			      modes[i]->flags & DRM_MODE_FLAG_INTERLACE ? "i" :"");
-
-		fallback = false;
-		conn_configured |= BIT(i);
-	}
-
-	if ((conn_configured & mask) != mask && conn_configured != conn_seq)
-		goto retry;
-
-	/*
-	 * If the BIOS didn't enable everything it could, fall back to have the
-	 * same user experiencing of lighting up as much as possible like the
-	 * fbdev helper library.
-	 */
-	if (num_connectors_enabled != num_connectors_detected &&
-	    num_connectors_enabled < INTEL_INFO(dev_priv)->num_pipes) {
-		DRM_DEBUG_KMS("fallback: Not all outputs enabled\n");
-		DRM_DEBUG_KMS("Enabled: %i, detected: %i\n", num_connectors_enabled,
-			      num_connectors_detected);
-		fallback = true;
-	}
-
-	if (fallback) {
-bail:
-		DRM_DEBUG_KMS("Not using firmware configuration\n");
-		memcpy(enabled, save_enabled, count);
-		ret = false;
-	}
-
-	drm_modeset_drop_locks(&ctx);
-	drm_modeset_acquire_fini(&ctx);
-
-	kfree(save_enabled);
-	return ret;
-}
-
 static const struct drm_fb_helper_funcs intel_fb_helper_funcs = {
-	.initial_config = intel_fb_initial_config,
 	.fb_probe = intelfb_create,
 };
 
diff --git a/drivers/gpu/drm/i915/intel_fbdev.h b/drivers/gpu/drm/i915/intel_fbdev.h
new file mode 100644
index 000000000000..de7c84250eb5
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_fbdev.h
@@ -0,0 +1,53 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_FBDEV_H__
+#define __INTEL_FBDEV_H__
+
+#include <linux/types.h>
+
+struct drm_device;
+struct drm_i915_private;
+
+#ifdef CONFIG_DRM_FBDEV_EMULATION
+int intel_fbdev_init(struct drm_device *dev);
+void intel_fbdev_initial_config_async(struct drm_device *dev);
+void intel_fbdev_unregister(struct drm_i915_private *dev_priv);
+void intel_fbdev_fini(struct drm_i915_private *dev_priv);
+void intel_fbdev_set_suspend(struct drm_device *dev, int state, bool synchronous);
+void intel_fbdev_output_poll_changed(struct drm_device *dev);
+void intel_fbdev_restore_mode(struct drm_device *dev);
+#else
+static inline int intel_fbdev_init(struct drm_device *dev)
+{
+	return 0;
+}
+
+static inline void intel_fbdev_initial_config_async(struct drm_device *dev)
+{
+}
+
+static inline void intel_fbdev_unregister(struct drm_i915_private *dev_priv)
+{
+}
+
+static inline void intel_fbdev_fini(struct drm_i915_private *dev_priv)
+{
+}
+
+static inline void intel_fbdev_set_suspend(struct drm_device *dev, int state, bool synchronous)
+{
+}
+
+static inline void intel_fbdev_output_poll_changed(struct drm_device *dev)
+{
+}
+
+static inline void intel_fbdev_restore_mode(struct drm_device *dev)
+{
+}
+#endif
+
+#endif /* __INTEL_FBDEV_H__ */
diff --git a/drivers/gpu/drm/i915/intel_fifo_underrun.c b/drivers/gpu/drm/i915/intel_fifo_underrun.c
index f33de4be4b89..74c8b0528294 100644
--- a/drivers/gpu/drm/i915/intel_fifo_underrun.c
+++ b/drivers/gpu/drm/i915/intel_fifo_underrun.c
@@ -27,6 +27,7 @@
 
 #include "i915_drv.h"
 #include "intel_drv.h"
+#include "intel_fbc.h"
 
 /**
  * DOC: fifo underrun handling
diff --git a/drivers/gpu/drm/i915/intel_frontbuffer.c b/drivers/gpu/drm/i915/intel_frontbuffer.c
index 16f253deaf8d..aa34e33b6087 100644
--- a/drivers/gpu/drm/i915/intel_frontbuffer.c
+++ b/drivers/gpu/drm/i915/intel_frontbuffer.c
@@ -61,9 +61,12 @@
  */
 
 
+#include "i915_drv.h"
+#include "intel_dp.h"
 #include "intel_drv.h"
+#include "intel_fbc.h"
 #include "intel_frontbuffer.h"
-#include "i915_drv.h"
+#include "intel_psr.h"
 
 void __intel_fb_obj_invalidate(struct drm_i915_gem_object *obj,
 			       enum fb_op_origin origin,
diff --git a/drivers/gpu/drm/i915/intel_frontbuffer.h b/drivers/gpu/drm/i915/intel_frontbuffer.h
index 63cd9a753a72..d5894666f658 100644
--- a/drivers/gpu/drm/i915/intel_frontbuffer.h
+++ b/drivers/gpu/drm/i915/intel_frontbuffer.h
@@ -24,9 +24,19 @@
 #ifndef __INTEL_FRONTBUFFER_H__
 #define __INTEL_FRONTBUFFER_H__
 
+#include "i915_gem_object.h"
+
 struct drm_i915_private;
 struct drm_i915_gem_object;
 
+enum fb_op_origin {
+	ORIGIN_GTT,
+	ORIGIN_CPU,
+	ORIGIN_CS,
+	ORIGIN_FLIP,
+	ORIGIN_DIRTYFB,
+};
+
 void intel_frontbuffer_flip_prepare(struct drm_i915_private *dev_priv,
 				    unsigned frontbuffer_bits);
 void intel_frontbuffer_flip_complete(struct drm_i915_private *dev_priv,
diff --git a/drivers/gpu/drm/i915/intel_gpu_commands.h b/drivers/gpu/drm/i915/intel_gpu_commands.h
index b96a31bc1080..a34ece53a771 100644
--- a/drivers/gpu/drm/i915/intel_gpu_commands.h
+++ b/drivers/gpu/drm/i915/intel_gpu_commands.h
@@ -105,8 +105,13 @@
 #define MI_SEMAPHORE_SIGNAL	MI_INSTR(0x1b, 0) /* GEN8+ */
 #define   MI_SEMAPHORE_TARGET(engine)	((engine)<<15)
 #define MI_SEMAPHORE_WAIT	MI_INSTR(0x1c, 2) /* GEN8+ */
-#define   MI_SEMAPHORE_POLL		(1<<15)
-#define   MI_SEMAPHORE_SAD_GTE_SDD	(1<<12)
+#define   MI_SEMAPHORE_POLL		(1 << 15)
+#define   MI_SEMAPHORE_SAD_GT_SDD	(0 << 12)
+#define   MI_SEMAPHORE_SAD_GTE_SDD	(1 << 12)
+#define   MI_SEMAPHORE_SAD_LT_SDD	(2 << 12)
+#define   MI_SEMAPHORE_SAD_LTE_SDD	(3 << 12)
+#define   MI_SEMAPHORE_SAD_EQ_SDD	(4 << 12)
+#define   MI_SEMAPHORE_SAD_NEQ_SDD	(5 << 12)
 #define MI_STORE_DWORD_IMM	MI_INSTR(0x20, 1)
 #define MI_STORE_DWORD_IMM_GEN4	MI_INSTR(0x20, 2)
 #define   MI_MEM_VIRTUAL	(1 << 22) /* 945,g33,965 */
diff --git a/drivers/gpu/drm/i915/intel_guc.c b/drivers/gpu/drm/i915/intel_guc.c
index 8660af3fd755..3aabfa2d9198 100644
--- a/drivers/gpu/drm/i915/intel_guc.c
+++ b/drivers/gpu/drm/i915/intel_guc.c
@@ -54,7 +54,7 @@ void intel_guc_init_send_regs(struct intel_guc *guc)
 	BUILD_BUG_ON(GUC_MAX_MMIO_MSG_LEN > SOFT_SCRATCH_COUNT);
 
 	for (i = 0; i < guc->send_regs.count; i++) {
-		fw_domains |= intel_uncore_forcewake_for_reg(dev_priv,
+		fw_domains |= intel_uncore_forcewake_for_reg(&dev_priv->uncore,
 					guc_send_reg(guc, i),
 					FW_REG_READ | FW_REG_WRITE);
 	}
@@ -203,11 +203,19 @@ int intel_guc_init(struct intel_guc *guc)
 		goto err_log;
 	GEM_BUG_ON(!guc->ads_vma);
 
+	if (HAS_GUC_CT(dev_priv)) {
+		ret = intel_guc_ct_init(&guc->ct);
+		if (ret)
+			goto err_ads;
+	}
+
 	/* We need to notify the guc whenever we change the GGTT */
 	i915_ggtt_enable_guc(dev_priv);
 
 	return 0;
 
+err_ads:
+	intel_guc_ads_destroy(guc);
 err_log:
 	intel_guc_log_destroy(&guc->log);
 err_shared:
@@ -222,6 +230,10 @@ void intel_guc_fini(struct intel_guc *guc)
 	struct drm_i915_private *dev_priv = guc_to_i915(guc);
 
 	i915_ggtt_disable_guc(dev_priv);
+
+	if (HAS_GUC_CT(dev_priv))
+		intel_guc_ct_fini(&guc->ct);
+
 	intel_guc_ads_destroy(guc);
 	intel_guc_log_destroy(&guc->log);
 	guc_shared_data_destroy(guc);
@@ -357,14 +369,14 @@ void intel_guc_init_params(struct intel_guc *guc)
 	 * they are power context saved so it's ok to release forcewake
 	 * when we are done here and take it again at xfer time.
 	 */
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_BLITTER);
+	intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_BLITTER);
 
 	I915_WRITE(SOFT_SCRATCH(0), 0);
 
 	for (i = 0; i < GUC_CTL_MAX_DWORDS; i++)
 		I915_WRITE(SOFT_SCRATCH(1 + i), params[i]);
 
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_BLITTER);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_BLITTER);
 }
 
 int intel_guc_send_nop(struct intel_guc *guc, const u32 *action, u32 len,
@@ -386,6 +398,7 @@ int intel_guc_send_mmio(struct intel_guc *guc, const u32 *action, u32 len,
 			u32 *response_buf, u32 response_buf_size)
 {
 	struct drm_i915_private *dev_priv = guc_to_i915(guc);
+	struct intel_uncore *uncore = &dev_priv->uncore;
 	u32 status;
 	int i;
 	int ret;
@@ -402,12 +415,12 @@ int intel_guc_send_mmio(struct intel_guc *guc, const u32 *action, u32 len,
 		*action != INTEL_GUC_ACTION_DEREGISTER_COMMAND_TRANSPORT_BUFFER);
 
 	mutex_lock(&guc->send_mutex);
-	intel_uncore_forcewake_get(dev_priv, guc->send_regs.fw_domains);
+	intel_uncore_forcewake_get(uncore, guc->send_regs.fw_domains);
 
 	for (i = 0; i < len; i++)
-		I915_WRITE(guc_send_reg(guc, i), action[i]);
+		intel_uncore_write(uncore, guc_send_reg(guc, i), action[i]);
 
-	POSTING_READ(guc_send_reg(guc, i - 1));
+	intel_uncore_posting_read(uncore, guc_send_reg(guc, i - 1));
 
 	intel_guc_notify(guc);
 
@@ -415,7 +428,7 @@ int intel_guc_send_mmio(struct intel_guc *guc, const u32 *action, u32 len,
 	 * No GuC command should ever take longer than 10ms.
 	 * Fast commands should still complete in 10us.
 	 */
-	ret = __intel_wait_for_register_fw(dev_priv,
+	ret = __intel_wait_for_register_fw(uncore,
 					   guc_send_reg(guc, 0),
 					   INTEL_GUC_MSG_TYPE_MASK,
 					   INTEL_GUC_MSG_TYPE_RESPONSE <<
@@ -442,7 +455,7 @@ int intel_guc_send_mmio(struct intel_guc *guc, const u32 *action, u32 len,
 	ret = INTEL_GUC_MSG_TO_DATA(status);
 
 out:
-	intel_uncore_forcewake_put(dev_priv, guc->send_regs.fw_domains);
+	intel_uncore_forcewake_put(uncore, guc->send_regs.fw_domains);
 	mutex_unlock(&guc->send_mutex);
 
 	return ret;
@@ -472,17 +485,25 @@ void intel_guc_to_host_event_handler_mmio(struct intel_guc *guc)
 	spin_unlock(&guc->irq_lock);
 	enable_rpm_wakeref_asserts(dev_priv);
 
-	intel_guc_to_host_process_recv_msg(guc, msg);
+	intel_guc_to_host_process_recv_msg(guc, &msg, 1);
 }
 
-void intel_guc_to_host_process_recv_msg(struct intel_guc *guc, u32 msg)
+int intel_guc_to_host_process_recv_msg(struct intel_guc *guc,
+				       const u32 *payload, u32 len)
 {
+	u32 msg;
+
+	if (unlikely(!len))
+		return -EPROTO;
+
 	/* Make sure to handle only enabled messages */
-	msg &= guc->msg_enabled_mask;
+	msg = payload[0] & guc->msg_enabled_mask;
 
 	if (msg & (INTEL_GUC_RECV_MSG_FLUSH_LOG_BUFFER |
 		   INTEL_GUC_RECV_MSG_CRASH_DUMP_POSTED))
 		intel_guc_log_handle_flush_event(&guc->log);
+
+	return 0;
 }
 
 int intel_guc_sample_forcewake(struct intel_guc *guc)
@@ -544,7 +565,7 @@ static int guc_sleep_state_action(struct intel_guc *guc,
 	if (ret)
 		return ret;
 
-	ret = __intel_wait_for_register(dev_priv, SOFT_SCRATCH(14),
+	ret = __intel_wait_for_register(&dev_priv->uncore, SOFT_SCRATCH(14),
 					INTEL_GUC_SLEEP_STATE_INVALID_MASK,
 					0, 0, 10, &status);
 	if (ret)
diff --git a/drivers/gpu/drm/i915/intel_guc.h b/drivers/gpu/drm/i915/intel_guc.h
index 744220296653..2c59ff8d9f39 100644
--- a/drivers/gpu/drm/i915/intel_guc.h
+++ b/drivers/gpu/drm/i915/intel_guc.h
@@ -32,6 +32,7 @@
 #include "intel_guc_log.h"
 #include "intel_guc_reg.h"
 #include "intel_uc_fw.h"
+#include "i915_utils.h"
 #include "i915_vma.h"
 
 struct guc_preempt_work {
@@ -164,7 +165,8 @@ int intel_guc_send_mmio(struct intel_guc *guc, const u32 *action, u32 len,
 void intel_guc_to_host_event_handler(struct intel_guc *guc);
 void intel_guc_to_host_event_handler_nop(struct intel_guc *guc);
 void intel_guc_to_host_event_handler_mmio(struct intel_guc *guc);
-void intel_guc_to_host_process_recv_msg(struct intel_guc *guc, u32 msg);
+int intel_guc_to_host_process_recv_msg(struct intel_guc *guc,
+				       const u32 *payload, u32 len);
 int intel_guc_sample_forcewake(struct intel_guc *guc);
 int intel_guc_auth_huc(struct intel_guc *guc, u32 rsa_offset);
 int intel_guc_suspend(struct intel_guc *guc);
diff --git a/drivers/gpu/drm/i915/intel_guc_ads.c b/drivers/gpu/drm/i915/intel_guc_ads.c
index f0db62887f50..bec62f34b15a 100644
--- a/drivers/gpu/drm/i915/intel_guc_ads.c
+++ b/drivers/gpu/drm/i915/intel_guc_ads.c
@@ -121,8 +121,7 @@ int intel_guc_ads_create(struct intel_guc *guc)
 	 * to find it. Note that we have to skip our header (1 page),
 	 * because our GuC shared data is there.
 	 */
-	kernel_ctx_vma = to_intel_context(dev_priv->kernel_context,
-					  dev_priv->engine[RCS])->state;
+	kernel_ctx_vma = dev_priv->engine[RCS0]->kernel_context->state;
 	blob->ads.golden_context_lrca =
 		intel_guc_ggtt_offset(guc, kernel_ctx_vma) + skipped_offset;
 
diff --git a/drivers/gpu/drm/i915/intel_guc_ct.c b/drivers/gpu/drm/i915/intel_guc_ct.c
index a52883e9146f..dde1dc0d6e69 100644
--- a/drivers/gpu/drm/i915/intel_guc_ct.c
+++ b/drivers/gpu/drm/i915/intel_guc_ct.c
@@ -140,11 +140,6 @@ static int guc_action_deregister_ct_buffer(struct intel_guc *guc,
 	return err;
 }
 
-static bool ctch_is_open(struct intel_guc_ct_channel *ctch)
-{
-	return ctch->vma != NULL;
-}
-
 static int ctch_init(struct intel_guc *guc,
 		     struct intel_guc_ct_channel *ctch)
 {
@@ -214,25 +209,21 @@ err_out:
 static void ctch_fini(struct intel_guc *guc,
 		      struct intel_guc_ct_channel *ctch)
 {
+	GEM_BUG_ON(ctch->enabled);
+
 	i915_vma_unpin_and_release(&ctch->vma, I915_VMA_RELEASE_MAP);
 }
 
-static int ctch_open(struct intel_guc *guc,
-		     struct intel_guc_ct_channel *ctch)
+static int ctch_enable(struct intel_guc *guc,
+		       struct intel_guc_ct_channel *ctch)
 {
 	u32 base;
 	int err;
 	int i;
 
-	CT_DEBUG_DRIVER("CT: channel %d reopen=%s\n",
-			ctch->owner, yesno(ctch_is_open(ctch)));
+	GEM_BUG_ON(!ctch->vma);
 
-	if (!ctch->vma) {
-		err = ctch_init(guc, ctch);
-		if (unlikely(err))
-			goto err_out;
-		GEM_BUG_ON(!ctch->vma);
-	}
+	GEM_BUG_ON(ctch->enabled);
 
 	/* vma should be already allocated and map'ed */
 	base = intel_guc_ggtt_offset(guc, ctch->vma);
@@ -255,7 +246,7 @@ static int ctch_open(struct intel_guc *guc,
 					    base + PAGE_SIZE/4 * CTB_RECV,
 					    INTEL_GUC_CT_BUFFER_TYPE_RECV);
 	if (unlikely(err))
-		goto err_fini;
+		goto err_out;
 
 	err = guc_action_register_ct_buffer(guc,
 					    base + PAGE_SIZE/4 * CTB_SEND,
@@ -263,23 +254,25 @@ static int ctch_open(struct intel_guc *guc,
 	if (unlikely(err))
 		goto err_deregister;
 
+	ctch->enabled = true;
+
 	return 0;
 
 err_deregister:
 	guc_action_deregister_ct_buffer(guc,
 					ctch->owner,
 					INTEL_GUC_CT_BUFFER_TYPE_RECV);
-err_fini:
-	ctch_fini(guc, ctch);
 err_out:
 	DRM_ERROR("CT: can't open channel %d; err=%d\n", ctch->owner, err);
 	return err;
 }
 
-static void ctch_close(struct intel_guc *guc,
-		       struct intel_guc_ct_channel *ctch)
+static void ctch_disable(struct intel_guc *guc,
+			 struct intel_guc_ct_channel *ctch)
 {
-	GEM_BUG_ON(!ctch_is_open(ctch));
+	GEM_BUG_ON(!ctch->enabled);
+
+	ctch->enabled = false;
 
 	guc_action_deregister_ct_buffer(guc,
 					ctch->owner,
@@ -287,7 +280,6 @@ static void ctch_close(struct intel_guc *guc,
 	guc_action_deregister_ct_buffer(guc,
 					ctch->owner,
 					INTEL_GUC_CT_BUFFER_TYPE_RECV);
-	ctch_fini(guc, ctch);
 }
 
 static u32 ctch_get_next_fence(struct intel_guc_ct_channel *ctch)
@@ -481,7 +473,7 @@ static int ctch_send(struct intel_guc_ct *ct,
 	u32 fence;
 	int err;
 
-	GEM_BUG_ON(!ctch_is_open(ctch));
+	GEM_BUG_ON(!ctch->enabled);
 	GEM_BUG_ON(!len);
 	GEM_BUG_ON(len & ~GUC_CT_MSG_LEN_MASK);
 	GEM_BUG_ON(!response_buf && response_buf_size);
@@ -709,14 +701,15 @@ static void ct_process_request(struct intel_guc_ct *ct,
 			       u32 action, u32 len, const u32 *payload)
 {
 	struct intel_guc *guc = ct_to_guc(ct);
+	int ret;
 
 	CT_DEBUG_DRIVER("CT: request %x %*ph\n", action, 4 * len, payload);
 
 	switch (action) {
 	case INTEL_GUC_ACTION_DEFAULT:
-		if (unlikely(len < 1))
+		ret = intel_guc_to_host_process_recv_msg(guc, payload, len);
+		if (unlikely(ret))
 			goto fail_unexpected;
-		intel_guc_to_host_process_recv_msg(guc, *payload);
 		break;
 
 	default:
@@ -817,7 +810,7 @@ static void ct_process_host_channel(struct intel_guc_ct *ct)
 	u32 msg[GUC_CT_MSG_LEN_MASK + 1]; /* one extra dw for the header */
 	int err = 0;
 
-	if (!ctch_is_open(ctch))
+	if (!ctch->enabled)
 		return;
 
 	do {
@@ -849,6 +842,51 @@ static void intel_guc_to_host_event_handler_ct(struct intel_guc *guc)
 }
 
 /**
+ * intel_guc_ct_init - Init CT communication
+ * @ct: pointer to CT struct
+ *
+ * Allocate memory required for communication via
+ * the CT channel.
+ *
+ * Shall only be called for platforms with HAS_GUC_CT.
+ *
+ * Return: 0 on success, a negative errno code on failure.
+ */
+int intel_guc_ct_init(struct intel_guc_ct *ct)
+{
+	struct intel_guc *guc = ct_to_guc(ct);
+	struct intel_guc_ct_channel *ctch = &ct->host_channel;
+	int err;
+
+	err = ctch_init(guc, ctch);
+	if (unlikely(err)) {
+		DRM_ERROR("CT: can't open channel %d; err=%d\n",
+			  ctch->owner, err);
+		return err;
+	}
+
+	GEM_BUG_ON(!ctch->vma);
+	return 0;
+}
+
+/**
+ * intel_guc_ct_fini - Fini CT communication
+ * @ct: pointer to CT struct
+ *
+ * Deallocate memory required for communication via
+ * the CT channel.
+ *
+ * Shall only be called for platforms with HAS_GUC_CT.
+ */
+void intel_guc_ct_fini(struct intel_guc_ct *ct)
+{
+	struct intel_guc *guc = ct_to_guc(ct);
+	struct intel_guc_ct_channel *ctch = &ct->host_channel;
+
+	ctch_fini(guc, ctch);
+}
+
+/**
  * intel_guc_ct_enable - Enable buffer based command transport.
  * @ct: pointer to CT struct
  *
@@ -865,7 +903,10 @@ int intel_guc_ct_enable(struct intel_guc_ct *ct)
 
 	GEM_BUG_ON(!HAS_GUC_CT(i915));
 
-	err = ctch_open(guc, ctch);
+	if (ctch->enabled)
+		return 0;
+
+	err = ctch_enable(guc, ctch);
 	if (unlikely(err))
 		return err;
 
@@ -890,10 +931,10 @@ void intel_guc_ct_disable(struct intel_guc_ct *ct)
 
 	GEM_BUG_ON(!HAS_GUC_CT(i915));
 
-	if (!ctch_is_open(ctch))
+	if (!ctch->enabled)
 		return;
 
-	ctch_close(guc, ctch);
+	ctch_disable(guc, ctch);
 
 	/* Disable send */
 	guc->send = intel_guc_send_nop;
diff --git a/drivers/gpu/drm/i915/intel_guc_ct.h b/drivers/gpu/drm/i915/intel_guc_ct.h
index d774895ab143..f5e7f0663304 100644
--- a/drivers/gpu/drm/i915/intel_guc_ct.h
+++ b/drivers/gpu/drm/i915/intel_guc_ct.h
@@ -66,6 +66,7 @@ struct intel_guc_ct_channel {
 	struct intel_guc_ct_buffer ctbs[2];
 	u32 owner;
 	u32 next_fence;
+	bool enabled;
 };
 
 /** Holds all command transport channels.
@@ -90,6 +91,8 @@ struct intel_guc_ct {
 };
 
 void intel_guc_ct_init_early(struct intel_guc_ct *ct);
+int intel_guc_ct_init(struct intel_guc_ct *ct);
+void intel_guc_ct_fini(struct intel_guc_ct *ct);
 int intel_guc_ct_enable(struct intel_guc_ct *ct);
 void intel_guc_ct_disable(struct intel_guc_ct *ct);
 
diff --git a/drivers/gpu/drm/i915/intel_guc_fw.c b/drivers/gpu/drm/i915/intel_guc_fw.c
index 13ff7003c6be..792a551450c7 100644
--- a/drivers/gpu/drm/i915/intel_guc_fw.c
+++ b/drivers/gpu/drm/i915/intel_guc_fw.c
@@ -241,7 +241,7 @@ static int guc_fw_xfer(struct intel_uc_fw *guc_fw, struct i915_vma *vma)
 
 	GEM_BUG_ON(guc_fw->type != INTEL_UC_FW_TYPE_GUC);
 
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL);
 
 	guc_prepare_xfer(guc);
 
@@ -254,7 +254,7 @@ static int guc_fw_xfer(struct intel_uc_fw *guc_fw, struct i915_vma *vma)
 
 	ret = guc_xfer_ucode(guc, vma);
 
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/i915/intel_guc_log.c b/drivers/gpu/drm/i915/intel_guc_log.c
index 806fdfd7c78a..7146524264dd 100644
--- a/drivers/gpu/drm/i915/intel_guc_log.c
+++ b/drivers/gpu/drm/i915/intel_guc_log.c
@@ -620,7 +620,12 @@ void intel_guc_log_relay_flush(struct intel_guc_log *log)
 
 void intel_guc_log_relay_close(struct intel_guc_log *log)
 {
+	struct intel_guc *guc = log_to_guc(log);
+	struct drm_i915_private *i915 = guc_to_i915(guc);
+
 	guc_log_disable_flush_events(log);
+	synchronize_irq(i915->drm.irq);
+
 	flush_work(&log->relay.flush_work);
 
 	mutex_lock(&log->relay.lock);
diff --git a/drivers/gpu/drm/i915/intel_guc_submission.c b/drivers/gpu/drm/i915/intel_guc_submission.c
index 8bc8aa54aa35..37f60cb8e9e1 100644
--- a/drivers/gpu/drm/i915/intel_guc_submission.c
+++ b/drivers/gpu/drm/i915/intel_guc_submission.c
@@ -382,7 +382,7 @@ static void guc_stage_desc_init(struct intel_guc_client *client)
 	desc->db_id = client->doorbell_id;
 
 	for_each_engine_masked(engine, dev_priv, client->engines, tmp) {
-		struct intel_context *ce = to_intel_context(ctx, engine);
+		struct intel_context *ce = intel_context_lookup(ctx, engine);
 		u32 guc_engine_id = engine->guc_id;
 		struct guc_execlist_context *lrc = &desc->lrc[guc_engine_id];
 
@@ -393,7 +393,7 @@ static void guc_stage_desc_init(struct intel_guc_client *client)
 		 * for now who owns a GuC client. But for future owner of GuC
 		 * client, need to make sure lrc is pinned prior to enter here.
 		 */
-		if (!ce->state)
+		if (!ce || !ce->state)
 			break;	/* XXX: continue? */
 
 		/*
@@ -535,7 +535,7 @@ static void guc_add_request(struct intel_guc *guc, struct i915_request *rq)
 	spin_lock(&client->wq_lock);
 
 	guc_wq_item_append(client, engine->guc_id, ctx_desc,
-			   ring_tail, rq->global_seqno);
+			   ring_tail, rq->fence.seqno);
 	guc_ring_doorbell(client);
 
 	client->submissions[engine->id] += 1;
@@ -567,7 +567,7 @@ static void inject_preempt_context(struct work_struct *work)
 					     preempt_work[engine->id]);
 	struct intel_guc_client *client = guc->preempt_client;
 	struct guc_stage_desc *stage_desc = __get_stage_desc(client);
-	struct intel_context *ce = to_intel_context(client->owner, engine);
+	struct intel_context *ce = engine->preempt_context;
 	u32 data[7];
 
 	if (!ce->ring->emit) { /* recreate upon load/resume */
@@ -575,7 +575,7 @@ static void inject_preempt_context(struct work_struct *work)
 		u32 *cs;
 
 		cs = ce->ring->vaddr;
-		if (engine->id == RCS) {
+		if (engine->class == RENDER_CLASS) {
 			cs = gen8_emit_ggtt_write_rcs(cs,
 						      GUC_PREEMPT_FINISHED,
 						      addr,
@@ -583,7 +583,8 @@ static void inject_preempt_context(struct work_struct *work)
 		} else {
 			cs = gen8_emit_ggtt_write(cs,
 						  GUC_PREEMPT_FINISHED,
-						  addr);
+						  addr,
+						  0);
 			*cs++ = MI_NOOP;
 			*cs++ = MI_NOOP;
 		}
@@ -649,9 +650,10 @@ static void wait_for_guc_preempt_report(struct intel_engine_cs *engine)
 	struct guc_ctx_report *report =
 		&data->preempt_ctx_report[engine->guc_id];
 
-	WARN_ON(wait_for_atomic(report->report_return_status ==
-				INTEL_GUC_REPORT_STATUS_COMPLETE,
-				GUC_PREEMPT_POSTPROCESS_DELAY_MS));
+	if (wait_for_atomic(report->report_return_status ==
+			    INTEL_GUC_REPORT_STATUS_COMPLETE,
+			    GUC_PREEMPT_POSTPROCESS_DELAY_MS))
+		DRM_ERROR("Timed out waiting for GuC preemption report\n");
 	/*
 	 * GuC is expecting that we're also going to clear the affected context
 	 * counter, let's also reset the return status to not depend on GuC
@@ -720,7 +722,7 @@ static inline int rq_prio(const struct i915_request *rq)
 
 static inline int port_prio(const struct execlist_port *port)
 {
-	return rq_prio(port_request(port));
+	return rq_prio(port_request(port)) | __NO_PREEMPTION;
 }
 
 static bool __guc_dequeue(struct intel_engine_cs *engine)
@@ -781,8 +783,7 @@ static bool __guc_dequeue(struct intel_engine_cs *engine)
 		}
 
 		rb_erase_cached(&p->node, &execlists->queue);
-		if (p->priority != I915_PRIORITY_NORMAL)
-			kmem_cache_free(engine->i915->priorities, p);
+		i915_priolist_free(p);
 	}
 done:
 	execlists->queue_priority_hint =
@@ -871,6 +872,104 @@ static void guc_reset_prepare(struct intel_engine_cs *engine)
 		flush_workqueue(engine->i915->guc.preempt_wq);
 }
 
+static void guc_reset(struct intel_engine_cs *engine, bool stalled)
+{
+	struct intel_engine_execlists * const execlists = &engine->execlists;
+	struct i915_request *rq;
+	unsigned long flags;
+
+	spin_lock_irqsave(&engine->timeline.lock, flags);
+
+	execlists_cancel_port_requests(execlists);
+
+	/* Push back any incomplete requests for replay after the reset. */
+	rq = execlists_unwind_incomplete_requests(execlists);
+	if (!rq)
+		goto out_unlock;
+
+	if (!i915_request_started(rq))
+		stalled = false;
+
+	i915_reset_request(rq, stalled);
+	intel_lr_context_reset(engine, rq->hw_context, rq->head, stalled);
+
+out_unlock:
+	spin_unlock_irqrestore(&engine->timeline.lock, flags);
+}
+
+static void guc_cancel_requests(struct intel_engine_cs *engine)
+{
+	struct intel_engine_execlists * const execlists = &engine->execlists;
+	struct i915_request *rq, *rn;
+	struct rb_node *rb;
+	unsigned long flags;
+
+	GEM_TRACE("%s\n", engine->name);
+
+	/*
+	 * Before we call engine->cancel_requests(), we should have exclusive
+	 * access to the submission state. This is arranged for us by the
+	 * caller disabling the interrupt generation, the tasklet and other
+	 * threads that may then access the same state, giving us a free hand
+	 * to reset state. However, we still need to let lockdep be aware that
+	 * we know this state may be accessed in hardirq context, so we
+	 * disable the irq around this manipulation and we want to keep
+	 * the spinlock focused on its duties and not accidentally conflate
+	 * coverage to the submission's irq state. (Similarly, although we
+	 * shouldn't need to disable irq around the manipulation of the
+	 * submission's irq state, we also wish to remind ourselves that
+	 * it is irq state.)
+	 */
+	spin_lock_irqsave(&engine->timeline.lock, flags);
+
+	/* Cancel the requests on the HW and clear the ELSP tracker. */
+	execlists_cancel_port_requests(execlists);
+
+	/* Mark all executing requests as skipped. */
+	list_for_each_entry(rq, &engine->timeline.requests, link) {
+		if (!i915_request_signaled(rq))
+			dma_fence_set_error(&rq->fence, -EIO);
+
+		i915_request_mark_complete(rq);
+	}
+
+	/* Flush the queued requests to the timeline list (for retiring). */
+	while ((rb = rb_first_cached(&execlists->queue))) {
+		struct i915_priolist *p = to_priolist(rb);
+		int i;
+
+		priolist_for_each_request_consume(rq, rn, p, i) {
+			list_del_init(&rq->sched.link);
+			__i915_request_submit(rq);
+			dma_fence_set_error(&rq->fence, -EIO);
+			i915_request_mark_complete(rq);
+		}
+
+		rb_erase_cached(&p->node, &execlists->queue);
+		i915_priolist_free(p);
+	}
+
+	/* Remaining _unready_ requests will be nop'ed when submitted */
+
+	execlists->queue_priority_hint = INT_MIN;
+	execlists->queue = RB_ROOT_CACHED;
+	GEM_BUG_ON(port_isset(execlists->port));
+
+	spin_unlock_irqrestore(&engine->timeline.lock, flags);
+}
+
+static void guc_reset_finish(struct intel_engine_cs *engine)
+{
+	struct intel_engine_execlists * const execlists = &engine->execlists;
+
+	if (__tasklet_enable(&execlists->tasklet))
+		/* And kick in case we missed a new request submission. */
+		tasklet_hi_schedule(&execlists->tasklet);
+
+	GEM_TRACE("%s: depth->%d\n", engine->name,
+		  atomic_read(&execlists->tasklet.count));
+}
+
 /*
  * Everything below here is concerned with setup & teardown, and is
  * therefore not part of the somewhat time-critical batch-submission
@@ -1031,7 +1130,7 @@ static int guc_clients_create(struct intel_guc *guc)
 	GEM_BUG_ON(guc->preempt_client);
 
 	client = guc_client_alloc(dev_priv,
-				  INTEL_INFO(dev_priv)->ring_mask,
+				  INTEL_INFO(dev_priv)->engine_mask,
 				  GUC_CLIENT_PRIORITY_KMD_NORMAL,
 				  dev_priv->kernel_context);
 	if (IS_ERR(client)) {
@@ -1042,7 +1141,7 @@ static int guc_clients_create(struct intel_guc *guc)
 
 	if (dev_priv->preempt_context) {
 		client = guc_client_alloc(dev_priv,
-					  INTEL_INFO(dev_priv)->ring_mask,
+					  INTEL_INFO(dev_priv)->engine_mask,
 					  GUC_CLIENT_PRIORITY_KMD_HIGH,
 					  dev_priv->preempt_context);
 		if (IS_ERR(client)) {
@@ -1262,10 +1361,12 @@ static void guc_interrupts_release(struct drm_i915_private *dev_priv)
 static void guc_submission_park(struct intel_engine_cs *engine)
 {
 	intel_engine_unpin_breadcrumbs_irq(engine);
+	engine->flags &= ~I915_ENGINE_NEEDS_BREADCRUMB_TASKLET;
 }
 
 static void guc_submission_unpark(struct intel_engine_cs *engine)
 {
+	engine->flags |= I915_ENGINE_NEEDS_BREADCRUMB_TASKLET;
 	intel_engine_pin_breadcrumbs_irq(engine);
 }
 
@@ -1290,6 +1391,10 @@ static void guc_set_default_submission(struct intel_engine_cs *engine)
 	engine->unpark = guc_submission_unpark;
 
 	engine->reset.prepare = guc_reset_prepare;
+	engine->reset.reset = guc_reset;
+	engine->reset.finish = guc_reset_finish;
+
+	engine->cancel_requests = guc_cancel_requests;
 
 	engine->flags &= ~I915_ENGINE_SUPPORTS_STATS;
 }
diff --git a/drivers/gpu/drm/i915/intel_guc_submission.h b/drivers/gpu/drm/i915/intel_guc_submission.h
index 169c54568340..aa5e6749c925 100644
--- a/drivers/gpu/drm/i915/intel_guc_submission.h
+++ b/drivers/gpu/drm/i915/intel_guc_submission.h
@@ -29,6 +29,7 @@
 
 #include "i915_gem.h"
 #include "i915_selftest.h"
+#include "intel_engine_types.h"
 
 struct drm_i915_private;
 
diff --git a/drivers/gpu/drm/i915/intel_hangcheck.c b/drivers/gpu/drm/i915/intel_hangcheck.c
index a219c796e56d..3d51ed1428d4 100644
--- a/drivers/gpu/drm/i915/intel_hangcheck.c
+++ b/drivers/gpu/drm/i915/intel_hangcheck.c
@@ -56,7 +56,7 @@ static bool subunits_stuck(struct intel_engine_cs *engine)
 	int slice;
 	int subslice;
 
-	if (engine->id != RCS)
+	if (engine->id != RCS0)
 		return true;
 
 	intel_engine_get_instdone(engine, &instdone);
@@ -118,11 +118,11 @@ engine_stuck(struct intel_engine_cs *engine, u64 acthd)
 	 * and break the hang. This should work on
 	 * all but the second generation chipsets.
 	 */
-	tmp = I915_READ_CTL(engine);
+	tmp = ENGINE_READ(engine, RING_CTL);
 	if (tmp & RING_WAIT) {
-		i915_handle_error(dev_priv, BIT(engine->id), 0,
+		i915_handle_error(dev_priv, engine->mask, 0,
 				  "stuck wait on %s", engine->name);
-		I915_WRITE_CTL(engine, tmp);
+		ENGINE_WRITE(engine, RING_CTL, tmp);
 		return ENGINE_WAIT_KICK;
 	}
 
@@ -133,21 +133,21 @@ static void hangcheck_load_sample(struct intel_engine_cs *engine,
 				  struct hangcheck *hc)
 {
 	hc->acthd = intel_engine_get_active_head(engine);
-	hc->seqno = intel_engine_get_seqno(engine);
+	hc->seqno = intel_engine_get_hangcheck_seqno(engine);
 }
 
 static void hangcheck_store_sample(struct intel_engine_cs *engine,
 				   const struct hangcheck *hc)
 {
 	engine->hangcheck.acthd = hc->acthd;
-	engine->hangcheck.seqno = hc->seqno;
+	engine->hangcheck.last_seqno = hc->seqno;
 }
 
 static enum intel_engine_hangcheck_action
 hangcheck_get_action(struct intel_engine_cs *engine,
 		     const struct hangcheck *hc)
 {
-	if (engine->hangcheck.seqno != hc->seqno)
+	if (engine->hangcheck.last_seqno != hc->seqno)
 		return ENGINE_ACTIVE_SEQNO;
 
 	if (intel_engine_is_idle(engine))
@@ -221,8 +221,8 @@ static void hangcheck_declare_hang(struct drm_i915_private *i915,
 				   unsigned int stuck)
 {
 	struct intel_engine_cs *engine;
+	intel_engine_mask_t tmp;
 	char msg[80];
-	unsigned int tmp;
 	int len;
 
 	/* If some rings hung but others were still busy, only
@@ -263,14 +263,14 @@ static void i915_hangcheck_elapsed(struct work_struct *work)
 	if (!READ_ONCE(dev_priv->gt.awake))
 		return;
 
-	if (i915_terminally_wedged(&dev_priv->gpu_error))
+	if (i915_terminally_wedged(dev_priv))
 		return;
 
 	/* As enabling the GPU requires fairly extensive mmio access,
 	 * periodically arm the mmio checker to see if we are triggering
 	 * any invalid access.
 	 */
-	intel_uncore_arm_unclaimed_mmio_detection(dev_priv);
+	intel_uncore_arm_unclaimed_mmio_detection(&dev_priv->uncore);
 
 	for_each_engine(engine, dev_priv, id) {
 		struct hangcheck hc;
@@ -282,13 +282,13 @@ static void i915_hangcheck_elapsed(struct work_struct *work)
 		hangcheck_store_sample(engine, &hc);
 
 		if (hc.stalled) {
-			hung |= intel_engine_flag(engine);
+			hung |= engine->mask;
 			if (hc.action != ENGINE_DEAD)
-				stuck |= intel_engine_flag(engine);
+				stuck |= engine->mask;
 		}
 
 		if (hc.wedged)
-			wedged |= intel_engine_flag(engine);
+			wedged |= engine->mask;
 	}
 
 	if (GEM_SHOW_DEBUG() && (hung | stuck)) {
diff --git a/drivers/gpu/drm/i915/intel_hdcp.c b/drivers/gpu/drm/i915/intel_hdcp.c
index ce7ba3a9c000..99b007169c49 100644
--- a/drivers/gpu/drm/i915/intel_hdcp.c
+++ b/drivers/gpu/drm/i915/intel_hdcp.c
@@ -6,15 +6,20 @@
  * Sean Paul <seanpaul@chromium.org>
  */
 
-#include <drm/drm_hdcp.h>
+#include <linux/component.h>
 #include <linux/i2c.h>
 #include <linux/random.h>
 
-#include "intel_drv.h"
+#include <drm/drm_hdcp.h>
+#include <drm/i915_component.h>
+
 #include "i915_reg.h"
+#include "intel_drv.h"
+#include "intel_hdcp.h"
 
 #define KEY_LOAD_TRIES	5
 #define ENCRYPT_STATUS_CHANGE_TIMEOUT_MS	50
+#define HDCP2_LC_RETRY_CNT			3
 
 static
 bool intel_hdcp_is_ksv_valid(u8 *ksv)
@@ -72,6 +77,52 @@ bool intel_hdcp_capable(struct intel_connector *connector)
 	return capable;
 }
 
+/* Is HDCP2.2 capable on Platform and Sink */
+static bool intel_hdcp2_capable(struct intel_connector *connector)
+{
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct intel_digital_port *intel_dig_port = conn_to_dig_port(connector);
+	struct intel_hdcp *hdcp = &connector->hdcp;
+	bool capable = false;
+
+	/* I915 support for HDCP2.2 */
+	if (!hdcp->hdcp2_supported)
+		return false;
+
+	/* MEI interface is solid */
+	mutex_lock(&dev_priv->hdcp_comp_mutex);
+	if (!dev_priv->hdcp_comp_added ||  !dev_priv->hdcp_master) {
+		mutex_unlock(&dev_priv->hdcp_comp_mutex);
+		return false;
+	}
+	mutex_unlock(&dev_priv->hdcp_comp_mutex);
+
+	/* Sink's capability for HDCP2.2 */
+	hdcp->shim->hdcp_2_2_capable(intel_dig_port, &capable);
+
+	return capable;
+}
+
+static inline bool intel_hdcp_in_use(struct intel_connector *connector)
+{
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	enum port port = connector->encoder->port;
+	u32 reg;
+
+	reg = I915_READ(PORT_HDCP_STATUS(port));
+	return reg & HDCP_STATUS_ENC;
+}
+
+static inline bool intel_hdcp2_in_use(struct intel_connector *connector)
+{
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	enum port port = connector->encoder->port;
+	u32 reg;
+
+	reg = I915_READ(HDCP2_STATUS_DDI(port));
+	return reg & LINK_ENCRYPTION_STATUS;
+}
+
 static int intel_hdcp_poll_ksv_fifo(struct intel_digital_port *intel_dig_port,
 				    const struct intel_hdcp_shim *shim)
 {
@@ -176,7 +227,7 @@ static int intel_hdcp_load_keys(struct drm_i915_private *dev_priv)
 	}
 
 	/* Wait for the keys to load (500us) */
-	ret = __intel_wait_for_register(dev_priv, HDCP_KEY_STATUS,
+	ret = __intel_wait_for_register(&dev_priv->uncore, HDCP_KEY_STATUS,
 					HDCP_KEY_LOAD_DONE, HDCP_KEY_LOAD_DONE,
 					10, 1, &val);
 	if (ret)
@@ -194,7 +245,7 @@ static int intel_hdcp_load_keys(struct drm_i915_private *dev_priv)
 static int intel_write_sha_text(struct drm_i915_private *dev_priv, u32 sha_text)
 {
 	I915_WRITE(HDCP_SHA_TEXT, sha_text);
-	if (intel_wait_for_register(dev_priv, HDCP_REP_CTL,
+	if (intel_wait_for_register(&dev_priv->uncore, HDCP_REP_CTL,
 				    HDCP_SHA1_READY, HDCP_SHA1_READY, 1)) {
 		DRM_ERROR("Timed out waiting for SHA1 ready\n");
 		return -ETIMEDOUT;
@@ -425,7 +476,7 @@ int intel_hdcp_validate_v_prime(struct intel_digital_port *intel_dig_port,
 
 	/* Tell the HW we're done with the hash and wait for it to ACK */
 	I915_WRITE(HDCP_REP_CTL, rep_ctl | HDCP_SHA1_COMPLETE_HASH);
-	if (intel_wait_for_register(dev_priv, HDCP_REP_CTL,
+	if (intel_wait_for_register(&dev_priv->uncore, HDCP_REP_CTL,
 				    HDCP_SHA1_COMPLETE,
 				    HDCP_SHA1_COMPLETE, 1)) {
 		DRM_ERROR("Timed out waiting for SHA1 complete\n");
@@ -555,7 +606,7 @@ static int intel_hdcp_auth(struct intel_digital_port *intel_dig_port,
 	I915_WRITE(PORT_HDCP_CONF(port), HDCP_CONF_CAPTURE_AN);
 
 	/* Wait for An to be acquired */
-	if (intel_wait_for_register(dev_priv, PORT_HDCP_STATUS(port),
+	if (intel_wait_for_register(&dev_priv->uncore, PORT_HDCP_STATUS(port),
 				    HDCP_STATUS_AN_READY,
 				    HDCP_STATUS_AN_READY, 1)) {
 		DRM_ERROR("Timed out waiting for An\n");
@@ -636,7 +687,7 @@ static int intel_hdcp_auth(struct intel_digital_port *intel_dig_port,
 	}
 
 	/* Wait for encryption confirmation */
-	if (intel_wait_for_register(dev_priv, PORT_HDCP_STATUS(port),
+	if (intel_wait_for_register(&dev_priv->uncore, PORT_HDCP_STATUS(port),
 				    HDCP_STATUS_ENC, HDCP_STATUS_ENC,
 				    ENCRYPT_STATUS_CHANGE_TIMEOUT_MS)) {
 		DRM_ERROR("Timed out waiting for encryption\n");
@@ -666,8 +717,10 @@ static int _intel_hdcp_disable(struct intel_connector *connector)
 	DRM_DEBUG_KMS("[%s:%d] HDCP is being disabled...\n",
 		      connector->base.name, connector->base.base.id);
 
+	hdcp->hdcp_encrypted = false;
 	I915_WRITE(PORT_HDCP_CONF(port), 0);
-	if (intel_wait_for_register(dev_priv, PORT_HDCP_STATUS(port), ~0, 0,
+	if (intel_wait_for_register(&dev_priv->uncore,
+				    PORT_HDCP_STATUS(port), ~0, 0,
 				    ENCRYPT_STATUS_CHANGE_TIMEOUT_MS)) {
 		DRM_ERROR("Failed to disable HDCP, timeout clearing status\n");
 		return -ETIMEDOUT;
@@ -711,8 +764,10 @@ static int _intel_hdcp_enable(struct intel_connector *connector)
 	/* Incase of authentication failures, HDCP spec expects reauth. */
 	for (i = 0; i < tries; i++) {
 		ret = intel_hdcp_auth(conn_to_dig_port(connector), hdcp->shim);
-		if (!ret)
+		if (!ret) {
+			hdcp->hdcp_encrypted = true;
 			return 0;
+		}
 
 		DRM_DEBUG_KMS("HDCP Auth failure (%d)\n", ret);
 
@@ -730,16 +785,64 @@ struct intel_connector *intel_hdcp_to_connector(struct intel_hdcp *hdcp)
 	return container_of(hdcp, struct intel_connector, hdcp);
 }
 
-static void intel_hdcp_check_work(struct work_struct *work)
+/* Implements Part 3 of the HDCP authorization procedure */
+static int intel_hdcp_check_link(struct intel_connector *connector)
 {
-	struct intel_hdcp *hdcp = container_of(to_delayed_work(work),
-					       struct intel_hdcp,
-					       check_work);
-	struct intel_connector *connector = intel_hdcp_to_connector(hdcp);
+	struct intel_hdcp *hdcp = &connector->hdcp;
+	struct drm_i915_private *dev_priv = connector->base.dev->dev_private;
+	struct intel_digital_port *intel_dig_port = conn_to_dig_port(connector);
+	enum port port = intel_dig_port->base.port;
+	int ret = 0;
 
-	if (!intel_hdcp_check_link(connector))
-		schedule_delayed_work(&hdcp->check_work,
-				      DRM_HDCP_CHECK_PERIOD_MS);
+	mutex_lock(&hdcp->mutex);
+
+	/* Check_link valid only when HDCP1.4 is enabled */
+	if (hdcp->value != DRM_MODE_CONTENT_PROTECTION_ENABLED ||
+	    !hdcp->hdcp_encrypted) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	if (WARN_ON(!intel_hdcp_in_use(connector))) {
+		DRM_ERROR("%s:%d HDCP link stopped encryption,%x\n",
+			  connector->base.name, connector->base.base.id,
+			  I915_READ(PORT_HDCP_STATUS(port)));
+		ret = -ENXIO;
+		hdcp->value = DRM_MODE_CONTENT_PROTECTION_DESIRED;
+		schedule_work(&hdcp->prop_work);
+		goto out;
+	}
+
+	if (hdcp->shim->check_link(intel_dig_port)) {
+		if (hdcp->value != DRM_MODE_CONTENT_PROTECTION_UNDESIRED) {
+			hdcp->value = DRM_MODE_CONTENT_PROTECTION_ENABLED;
+			schedule_work(&hdcp->prop_work);
+		}
+		goto out;
+	}
+
+	DRM_DEBUG_KMS("[%s:%d] HDCP link failed, retrying authentication\n",
+		      connector->base.name, connector->base.base.id);
+
+	ret = _intel_hdcp_disable(connector);
+	if (ret) {
+		DRM_ERROR("Failed to disable hdcp (%d)\n", ret);
+		hdcp->value = DRM_MODE_CONTENT_PROTECTION_DESIRED;
+		schedule_work(&hdcp->prop_work);
+		goto out;
+	}
+
+	ret = _intel_hdcp_enable(connector);
+	if (ret) {
+		DRM_ERROR("Failed to enable hdcp (%d)\n", ret);
+		hdcp->value = DRM_MODE_CONTENT_PROTECTION_DESIRED;
+		schedule_work(&hdcp->prop_work);
+		goto out;
+	}
+
+out:
+	mutex_unlock(&hdcp->mutex);
+	return ret;
 }
 
 static void intel_hdcp_prop_work(struct work_struct *work)
@@ -773,14 +876,929 @@ bool is_hdcp_supported(struct drm_i915_private *dev_priv, enum port port)
 	return INTEL_GEN(dev_priv) >= 9 && port < PORT_E;
 }
 
+static int
+hdcp2_prepare_ake_init(struct intel_connector *connector,
+		       struct hdcp2_ake_init *ake_data)
+{
+	struct hdcp_port_data *data = &connector->hdcp.port_data;
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct i915_hdcp_comp_master *comp;
+	int ret;
+
+	mutex_lock(&dev_priv->hdcp_comp_mutex);
+	comp = dev_priv->hdcp_master;
+
+	if (!comp || !comp->ops) {
+		mutex_unlock(&dev_priv->hdcp_comp_mutex);
+		return -EINVAL;
+	}
+
+	ret = comp->ops->initiate_hdcp2_session(comp->mei_dev, data, ake_data);
+	if (ret)
+		DRM_DEBUG_KMS("Prepare_ake_init failed. %d\n", ret);
+	mutex_unlock(&dev_priv->hdcp_comp_mutex);
+
+	return ret;
+}
+
+static int
+hdcp2_verify_rx_cert_prepare_km(struct intel_connector *connector,
+				struct hdcp2_ake_send_cert *rx_cert,
+				bool *paired,
+				struct hdcp2_ake_no_stored_km *ek_pub_km,
+				size_t *msg_sz)
+{
+	struct hdcp_port_data *data = &connector->hdcp.port_data;
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct i915_hdcp_comp_master *comp;
+	int ret;
+
+	mutex_lock(&dev_priv->hdcp_comp_mutex);
+	comp = dev_priv->hdcp_master;
+
+	if (!comp || !comp->ops) {
+		mutex_unlock(&dev_priv->hdcp_comp_mutex);
+		return -EINVAL;
+	}
+
+	ret = comp->ops->verify_receiver_cert_prepare_km(comp->mei_dev, data,
+							 rx_cert, paired,
+							 ek_pub_km, msg_sz);
+	if (ret < 0)
+		DRM_DEBUG_KMS("Verify rx_cert failed. %d\n", ret);
+	mutex_unlock(&dev_priv->hdcp_comp_mutex);
+
+	return ret;
+}
+
+static int hdcp2_verify_hprime(struct intel_connector *connector,
+			       struct hdcp2_ake_send_hprime *rx_hprime)
+{
+	struct hdcp_port_data *data = &connector->hdcp.port_data;
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct i915_hdcp_comp_master *comp;
+	int ret;
+
+	mutex_lock(&dev_priv->hdcp_comp_mutex);
+	comp = dev_priv->hdcp_master;
+
+	if (!comp || !comp->ops) {
+		mutex_unlock(&dev_priv->hdcp_comp_mutex);
+		return -EINVAL;
+	}
+
+	ret = comp->ops->verify_hprime(comp->mei_dev, data, rx_hprime);
+	if (ret < 0)
+		DRM_DEBUG_KMS("Verify hprime failed. %d\n", ret);
+	mutex_unlock(&dev_priv->hdcp_comp_mutex);
+
+	return ret;
+}
+
+static int
+hdcp2_store_pairing_info(struct intel_connector *connector,
+			 struct hdcp2_ake_send_pairing_info *pairing_info)
+{
+	struct hdcp_port_data *data = &connector->hdcp.port_data;
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct i915_hdcp_comp_master *comp;
+	int ret;
+
+	mutex_lock(&dev_priv->hdcp_comp_mutex);
+	comp = dev_priv->hdcp_master;
+
+	if (!comp || !comp->ops) {
+		mutex_unlock(&dev_priv->hdcp_comp_mutex);
+		return -EINVAL;
+	}
+
+	ret = comp->ops->store_pairing_info(comp->mei_dev, data, pairing_info);
+	if (ret < 0)
+		DRM_DEBUG_KMS("Store pairing info failed. %d\n", ret);
+	mutex_unlock(&dev_priv->hdcp_comp_mutex);
+
+	return ret;
+}
+
+static int
+hdcp2_prepare_lc_init(struct intel_connector *connector,
+		      struct hdcp2_lc_init *lc_init)
+{
+	struct hdcp_port_data *data = &connector->hdcp.port_data;
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct i915_hdcp_comp_master *comp;
+	int ret;
+
+	mutex_lock(&dev_priv->hdcp_comp_mutex);
+	comp = dev_priv->hdcp_master;
+
+	if (!comp || !comp->ops) {
+		mutex_unlock(&dev_priv->hdcp_comp_mutex);
+		return -EINVAL;
+	}
+
+	ret = comp->ops->initiate_locality_check(comp->mei_dev, data, lc_init);
+	if (ret < 0)
+		DRM_DEBUG_KMS("Prepare lc_init failed. %d\n", ret);
+	mutex_unlock(&dev_priv->hdcp_comp_mutex);
+
+	return ret;
+}
+
+static int
+hdcp2_verify_lprime(struct intel_connector *connector,
+		    struct hdcp2_lc_send_lprime *rx_lprime)
+{
+	struct hdcp_port_data *data = &connector->hdcp.port_data;
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct i915_hdcp_comp_master *comp;
+	int ret;
+
+	mutex_lock(&dev_priv->hdcp_comp_mutex);
+	comp = dev_priv->hdcp_master;
+
+	if (!comp || !comp->ops) {
+		mutex_unlock(&dev_priv->hdcp_comp_mutex);
+		return -EINVAL;
+	}
+
+	ret = comp->ops->verify_lprime(comp->mei_dev, data, rx_lprime);
+	if (ret < 0)
+		DRM_DEBUG_KMS("Verify L_Prime failed. %d\n", ret);
+	mutex_unlock(&dev_priv->hdcp_comp_mutex);
+
+	return ret;
+}
+
+static int hdcp2_prepare_skey(struct intel_connector *connector,
+			      struct hdcp2_ske_send_eks *ske_data)
+{
+	struct hdcp_port_data *data = &connector->hdcp.port_data;
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct i915_hdcp_comp_master *comp;
+	int ret;
+
+	mutex_lock(&dev_priv->hdcp_comp_mutex);
+	comp = dev_priv->hdcp_master;
+
+	if (!comp || !comp->ops) {
+		mutex_unlock(&dev_priv->hdcp_comp_mutex);
+		return -EINVAL;
+	}
+
+	ret = comp->ops->get_session_key(comp->mei_dev, data, ske_data);
+	if (ret < 0)
+		DRM_DEBUG_KMS("Get session key failed. %d\n", ret);
+	mutex_unlock(&dev_priv->hdcp_comp_mutex);
+
+	return ret;
+}
+
+static int
+hdcp2_verify_rep_topology_prepare_ack(struct intel_connector *connector,
+				      struct hdcp2_rep_send_receiverid_list
+								*rep_topology,
+				      struct hdcp2_rep_send_ack *rep_send_ack)
+{
+	struct hdcp_port_data *data = &connector->hdcp.port_data;
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct i915_hdcp_comp_master *comp;
+	int ret;
+
+	mutex_lock(&dev_priv->hdcp_comp_mutex);
+	comp = dev_priv->hdcp_master;
+
+	if (!comp || !comp->ops) {
+		mutex_unlock(&dev_priv->hdcp_comp_mutex);
+		return -EINVAL;
+	}
+
+	ret = comp->ops->repeater_check_flow_prepare_ack(comp->mei_dev, data,
+							 rep_topology,
+							 rep_send_ack);
+	if (ret < 0)
+		DRM_DEBUG_KMS("Verify rep topology failed. %d\n", ret);
+	mutex_unlock(&dev_priv->hdcp_comp_mutex);
+
+	return ret;
+}
+
+static int
+hdcp2_verify_mprime(struct intel_connector *connector,
+		    struct hdcp2_rep_stream_ready *stream_ready)
+{
+	struct hdcp_port_data *data = &connector->hdcp.port_data;
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct i915_hdcp_comp_master *comp;
+	int ret;
+
+	mutex_lock(&dev_priv->hdcp_comp_mutex);
+	comp = dev_priv->hdcp_master;
+
+	if (!comp || !comp->ops) {
+		mutex_unlock(&dev_priv->hdcp_comp_mutex);
+		return -EINVAL;
+	}
+
+	ret = comp->ops->verify_mprime(comp->mei_dev, data, stream_ready);
+	if (ret < 0)
+		DRM_DEBUG_KMS("Verify mprime failed. %d\n", ret);
+	mutex_unlock(&dev_priv->hdcp_comp_mutex);
+
+	return ret;
+}
+
+static int hdcp2_authenticate_port(struct intel_connector *connector)
+{
+	struct hdcp_port_data *data = &connector->hdcp.port_data;
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct i915_hdcp_comp_master *comp;
+	int ret;
+
+	mutex_lock(&dev_priv->hdcp_comp_mutex);
+	comp = dev_priv->hdcp_master;
+
+	if (!comp || !comp->ops) {
+		mutex_unlock(&dev_priv->hdcp_comp_mutex);
+		return -EINVAL;
+	}
+
+	ret = comp->ops->enable_hdcp_authentication(comp->mei_dev, data);
+	if (ret < 0)
+		DRM_DEBUG_KMS("Enable hdcp auth failed. %d\n", ret);
+	mutex_unlock(&dev_priv->hdcp_comp_mutex);
+
+	return ret;
+}
+
+static int hdcp2_close_mei_session(struct intel_connector *connector)
+{
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct i915_hdcp_comp_master *comp;
+	int ret;
+
+	mutex_lock(&dev_priv->hdcp_comp_mutex);
+	comp = dev_priv->hdcp_master;
+
+	if (!comp || !comp->ops) {
+		mutex_unlock(&dev_priv->hdcp_comp_mutex);
+		return -EINVAL;
+	}
+
+	ret = comp->ops->close_hdcp_session(comp->mei_dev,
+					     &connector->hdcp.port_data);
+	mutex_unlock(&dev_priv->hdcp_comp_mutex);
+
+	return ret;
+}
+
+static int hdcp2_deauthenticate_port(struct intel_connector *connector)
+{
+	return hdcp2_close_mei_session(connector);
+}
+
+/* Authentication flow starts from here */
+static int hdcp2_authentication_key_exchange(struct intel_connector *connector)
+{
+	struct intel_digital_port *intel_dig_port = conn_to_dig_port(connector);
+	struct intel_hdcp *hdcp = &connector->hdcp;
+	union {
+		struct hdcp2_ake_init ake_init;
+		struct hdcp2_ake_send_cert send_cert;
+		struct hdcp2_ake_no_stored_km no_stored_km;
+		struct hdcp2_ake_send_hprime send_hprime;
+		struct hdcp2_ake_send_pairing_info pairing_info;
+	} msgs;
+	const struct intel_hdcp_shim *shim = hdcp->shim;
+	size_t size;
+	int ret;
+
+	/* Init for seq_num */
+	hdcp->seq_num_v = 0;
+	hdcp->seq_num_m = 0;
+
+	ret = hdcp2_prepare_ake_init(connector, &msgs.ake_init);
+	if (ret < 0)
+		return ret;
+
+	ret = shim->write_2_2_msg(intel_dig_port, &msgs.ake_init,
+				  sizeof(msgs.ake_init));
+	if (ret < 0)
+		return ret;
+
+	ret = shim->read_2_2_msg(intel_dig_port, HDCP_2_2_AKE_SEND_CERT,
+				 &msgs.send_cert, sizeof(msgs.send_cert));
+	if (ret < 0)
+		return ret;
+
+	if (msgs.send_cert.rx_caps[0] != HDCP_2_2_RX_CAPS_VERSION_VAL)
+		return -EINVAL;
+
+	hdcp->is_repeater = HDCP_2_2_RX_REPEATER(msgs.send_cert.rx_caps[2]);
+
+	/*
+	 * Here msgs.no_stored_km will hold msgs corresponding to the km
+	 * stored also.
+	 */
+	ret = hdcp2_verify_rx_cert_prepare_km(connector, &msgs.send_cert,
+					      &hdcp->is_paired,
+					      &msgs.no_stored_km, &size);
+	if (ret < 0)
+		return ret;
+
+	ret = shim->write_2_2_msg(intel_dig_port, &msgs.no_stored_km, size);
+	if (ret < 0)
+		return ret;
+
+	ret = shim->read_2_2_msg(intel_dig_port, HDCP_2_2_AKE_SEND_HPRIME,
+				 &msgs.send_hprime, sizeof(msgs.send_hprime));
+	if (ret < 0)
+		return ret;
+
+	ret = hdcp2_verify_hprime(connector, &msgs.send_hprime);
+	if (ret < 0)
+		return ret;
+
+	if (!hdcp->is_paired) {
+		/* Pairing is required */
+		ret = shim->read_2_2_msg(intel_dig_port,
+					 HDCP_2_2_AKE_SEND_PAIRING_INFO,
+					 &msgs.pairing_info,
+					 sizeof(msgs.pairing_info));
+		if (ret < 0)
+			return ret;
+
+		ret = hdcp2_store_pairing_info(connector, &msgs.pairing_info);
+		if (ret < 0)
+			return ret;
+		hdcp->is_paired = true;
+	}
+
+	return 0;
+}
+
+static int hdcp2_locality_check(struct intel_connector *connector)
+{
+	struct intel_digital_port *intel_dig_port = conn_to_dig_port(connector);
+	struct intel_hdcp *hdcp = &connector->hdcp;
+	union {
+		struct hdcp2_lc_init lc_init;
+		struct hdcp2_lc_send_lprime send_lprime;
+	} msgs;
+	const struct intel_hdcp_shim *shim = hdcp->shim;
+	int tries = HDCP2_LC_RETRY_CNT, ret, i;
+
+	for (i = 0; i < tries; i++) {
+		ret = hdcp2_prepare_lc_init(connector, &msgs.lc_init);
+		if (ret < 0)
+			continue;
+
+		ret = shim->write_2_2_msg(intel_dig_port, &msgs.lc_init,
+				      sizeof(msgs.lc_init));
+		if (ret < 0)
+			continue;
+
+		ret = shim->read_2_2_msg(intel_dig_port,
+					 HDCP_2_2_LC_SEND_LPRIME,
+					 &msgs.send_lprime,
+					 sizeof(msgs.send_lprime));
+		if (ret < 0)
+			continue;
+
+		ret = hdcp2_verify_lprime(connector, &msgs.send_lprime);
+		if (!ret)
+			break;
+	}
+
+	return ret;
+}
+
+static int hdcp2_session_key_exchange(struct intel_connector *connector)
+{
+	struct intel_digital_port *intel_dig_port = conn_to_dig_port(connector);
+	struct intel_hdcp *hdcp = &connector->hdcp;
+	struct hdcp2_ske_send_eks send_eks;
+	int ret;
+
+	ret = hdcp2_prepare_skey(connector, &send_eks);
+	if (ret < 0)
+		return ret;
+
+	ret = hdcp->shim->write_2_2_msg(intel_dig_port, &send_eks,
+					sizeof(send_eks));
+	if (ret < 0)
+		return ret;
+
+	return 0;
+}
+
+static
+int hdcp2_propagate_stream_management_info(struct intel_connector *connector)
+{
+	struct intel_digital_port *intel_dig_port = conn_to_dig_port(connector);
+	struct intel_hdcp *hdcp = &connector->hdcp;
+	union {
+		struct hdcp2_rep_stream_manage stream_manage;
+		struct hdcp2_rep_stream_ready stream_ready;
+	} msgs;
+	const struct intel_hdcp_shim *shim = hdcp->shim;
+	int ret;
+
+	/* Prepare RepeaterAuth_Stream_Manage msg */
+	msgs.stream_manage.msg_id = HDCP_2_2_REP_STREAM_MANAGE;
+	drm_hdcp2_u32_to_seq_num(msgs.stream_manage.seq_num_m, hdcp->seq_num_m);
+
+	/* K no of streams is fixed as 1. Stored as big-endian. */
+	msgs.stream_manage.k = cpu_to_be16(1);
+
+	/* For HDMI this is forced to be 0x0. For DP SST also this is 0x0. */
+	msgs.stream_manage.streams[0].stream_id = 0;
+	msgs.stream_manage.streams[0].stream_type = hdcp->content_type;
+
+	/* Send it to Repeater */
+	ret = shim->write_2_2_msg(intel_dig_port, &msgs.stream_manage,
+				  sizeof(msgs.stream_manage));
+	if (ret < 0)
+		return ret;
+
+	ret = shim->read_2_2_msg(intel_dig_port, HDCP_2_2_REP_STREAM_READY,
+				 &msgs.stream_ready, sizeof(msgs.stream_ready));
+	if (ret < 0)
+		return ret;
+
+	hdcp->port_data.seq_num_m = hdcp->seq_num_m;
+	hdcp->port_data.streams[0].stream_type = hdcp->content_type;
+
+	ret = hdcp2_verify_mprime(connector, &msgs.stream_ready);
+	if (ret < 0)
+		return ret;
+
+	hdcp->seq_num_m++;
+
+	if (hdcp->seq_num_m > HDCP_2_2_SEQ_NUM_MAX) {
+		DRM_DEBUG_KMS("seq_num_m roll over.\n");
+		return -1;
+	}
+
+	return 0;
+}
+
+static
+int hdcp2_authenticate_repeater_topology(struct intel_connector *connector)
+{
+	struct intel_digital_port *intel_dig_port = conn_to_dig_port(connector);
+	struct intel_hdcp *hdcp = &connector->hdcp;
+	union {
+		struct hdcp2_rep_send_receiverid_list recvid_list;
+		struct hdcp2_rep_send_ack rep_ack;
+	} msgs;
+	const struct intel_hdcp_shim *shim = hdcp->shim;
+	u8 *rx_info;
+	u32 seq_num_v;
+	int ret;
+
+	ret = shim->read_2_2_msg(intel_dig_port, HDCP_2_2_REP_SEND_RECVID_LIST,
+				 &msgs.recvid_list, sizeof(msgs.recvid_list));
+	if (ret < 0)
+		return ret;
+
+	rx_info = msgs.recvid_list.rx_info;
+
+	if (HDCP_2_2_MAX_CASCADE_EXCEEDED(rx_info[1]) ||
+	    HDCP_2_2_MAX_DEVS_EXCEEDED(rx_info[1])) {
+		DRM_DEBUG_KMS("Topology Max Size Exceeded\n");
+		return -EINVAL;
+	}
+
+	/* Converting and Storing the seq_num_v to local variable as DWORD */
+	seq_num_v = drm_hdcp2_seq_num_to_u32(msgs.recvid_list.seq_num_v);
+
+	if (seq_num_v < hdcp->seq_num_v) {
+		/* Roll over of the seq_num_v from repeater. Reauthenticate. */
+		DRM_DEBUG_KMS("Seq_num_v roll over.\n");
+		return -EINVAL;
+	}
+
+	ret = hdcp2_verify_rep_topology_prepare_ack(connector,
+						    &msgs.recvid_list,
+						    &msgs.rep_ack);
+	if (ret < 0)
+		return ret;
+
+	hdcp->seq_num_v = seq_num_v;
+	ret = shim->write_2_2_msg(intel_dig_port, &msgs.rep_ack,
+				  sizeof(msgs.rep_ack));
+	if (ret < 0)
+		return ret;
+
+	return 0;
+}
+
+static int hdcp2_authenticate_repeater(struct intel_connector *connector)
+{
+	int ret;
+
+	ret = hdcp2_authenticate_repeater_topology(connector);
+	if (ret < 0)
+		return ret;
+
+	return hdcp2_propagate_stream_management_info(connector);
+}
+
+static int hdcp2_authenticate_sink(struct intel_connector *connector)
+{
+	struct intel_digital_port *intel_dig_port = conn_to_dig_port(connector);
+	struct intel_hdcp *hdcp = &connector->hdcp;
+	const struct intel_hdcp_shim *shim = hdcp->shim;
+	int ret;
+
+	ret = hdcp2_authentication_key_exchange(connector);
+	if (ret < 0) {
+		DRM_DEBUG_KMS("AKE Failed. Err : %d\n", ret);
+		return ret;
+	}
+
+	ret = hdcp2_locality_check(connector);
+	if (ret < 0) {
+		DRM_DEBUG_KMS("Locality Check failed. Err : %d\n", ret);
+		return ret;
+	}
+
+	ret = hdcp2_session_key_exchange(connector);
+	if (ret < 0) {
+		DRM_DEBUG_KMS("SKE Failed. Err : %d\n", ret);
+		return ret;
+	}
+
+	if (shim->config_stream_type) {
+		ret = shim->config_stream_type(intel_dig_port,
+					       hdcp->is_repeater,
+					       hdcp->content_type);
+		if (ret < 0)
+			return ret;
+	}
+
+	if (hdcp->is_repeater) {
+		ret = hdcp2_authenticate_repeater(connector);
+		if (ret < 0) {
+			DRM_DEBUG_KMS("Repeater Auth Failed. Err: %d\n", ret);
+			return ret;
+		}
+	}
+
+	hdcp->port_data.streams[0].stream_type = hdcp->content_type;
+	ret = hdcp2_authenticate_port(connector);
+	if (ret < 0)
+		return ret;
+
+	return ret;
+}
+
+static int hdcp2_enable_encryption(struct intel_connector *connector)
+{
+	struct intel_digital_port *intel_dig_port = conn_to_dig_port(connector);
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct intel_hdcp *hdcp = &connector->hdcp;
+	enum port port = connector->encoder->port;
+	int ret;
+
+	WARN_ON(I915_READ(HDCP2_STATUS_DDI(port)) & LINK_ENCRYPTION_STATUS);
+
+	if (hdcp->shim->toggle_signalling) {
+		ret = hdcp->shim->toggle_signalling(intel_dig_port, true);
+		if (ret) {
+			DRM_ERROR("Failed to enable HDCP signalling. %d\n",
+				  ret);
+			return ret;
+		}
+	}
+
+	if (I915_READ(HDCP2_STATUS_DDI(port)) & LINK_AUTH_STATUS) {
+		/* Link is Authenticated. Now set for Encryption */
+		I915_WRITE(HDCP2_CTL_DDI(port),
+			   I915_READ(HDCP2_CTL_DDI(port)) |
+			   CTL_LINK_ENCRYPTION_REQ);
+	}
+
+	ret = intel_wait_for_register(&dev_priv->uncore, HDCP2_STATUS_DDI(port),
+				      LINK_ENCRYPTION_STATUS,
+				      LINK_ENCRYPTION_STATUS,
+				      ENCRYPT_STATUS_CHANGE_TIMEOUT_MS);
+
+	return ret;
+}
+
+static int hdcp2_disable_encryption(struct intel_connector *connector)
+{
+	struct intel_digital_port *intel_dig_port = conn_to_dig_port(connector);
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct intel_hdcp *hdcp = &connector->hdcp;
+	enum port port = connector->encoder->port;
+	int ret;
+
+	WARN_ON(!(I915_READ(HDCP2_STATUS_DDI(port)) & LINK_ENCRYPTION_STATUS));
+
+	I915_WRITE(HDCP2_CTL_DDI(port),
+		   I915_READ(HDCP2_CTL_DDI(port)) & ~CTL_LINK_ENCRYPTION_REQ);
+
+	ret = intel_wait_for_register(&dev_priv->uncore, HDCP2_STATUS_DDI(port),
+				      LINK_ENCRYPTION_STATUS, 0x0,
+				      ENCRYPT_STATUS_CHANGE_TIMEOUT_MS);
+	if (ret == -ETIMEDOUT)
+		DRM_DEBUG_KMS("Disable Encryption Timedout");
+
+	if (hdcp->shim->toggle_signalling) {
+		ret = hdcp->shim->toggle_signalling(intel_dig_port, false);
+		if (ret) {
+			DRM_ERROR("Failed to disable HDCP signalling. %d\n",
+				  ret);
+			return ret;
+		}
+	}
+
+	return ret;
+}
+
+static int hdcp2_authenticate_and_encrypt(struct intel_connector *connector)
+{
+	int ret, i, tries = 3;
+
+	for (i = 0; i < tries; i++) {
+		ret = hdcp2_authenticate_sink(connector);
+		if (!ret)
+			break;
+
+		/* Clearing the mei hdcp session */
+		DRM_DEBUG_KMS("HDCP2.2 Auth %d of %d Failed.(%d)\n",
+			      i + 1, tries, ret);
+		if (hdcp2_deauthenticate_port(connector) < 0)
+			DRM_DEBUG_KMS("Port deauth failed.\n");
+	}
+
+	if (i != tries) {
+		/*
+		 * Ensuring the required 200mSec min time interval between
+		 * Session Key Exchange and encryption.
+		 */
+		msleep(HDCP_2_2_DELAY_BEFORE_ENCRYPTION_EN);
+		ret = hdcp2_enable_encryption(connector);
+		if (ret < 0) {
+			DRM_DEBUG_KMS("Encryption Enable Failed.(%d)\n", ret);
+			if (hdcp2_deauthenticate_port(connector) < 0)
+				DRM_DEBUG_KMS("Port deauth failed.\n");
+		}
+	}
+
+	return ret;
+}
+
+static int _intel_hdcp2_enable(struct intel_connector *connector)
+{
+	struct intel_hdcp *hdcp = &connector->hdcp;
+	int ret;
+
+	DRM_DEBUG_KMS("[%s:%d] HDCP2.2 is being enabled. Type: %d\n",
+		      connector->base.name, connector->base.base.id,
+		      hdcp->content_type);
+
+	ret = hdcp2_authenticate_and_encrypt(connector);
+	if (ret) {
+		DRM_DEBUG_KMS("HDCP2 Type%d  Enabling Failed. (%d)\n",
+			      hdcp->content_type, ret);
+		return ret;
+	}
+
+	DRM_DEBUG_KMS("[%s:%d] HDCP2.2 is enabled. Type %d\n",
+		      connector->base.name, connector->base.base.id,
+		      hdcp->content_type);
+
+	hdcp->hdcp2_encrypted = true;
+	return 0;
+}
+
+static int _intel_hdcp2_disable(struct intel_connector *connector)
+{
+	int ret;
+
+	DRM_DEBUG_KMS("[%s:%d] HDCP2.2 is being Disabled\n",
+		      connector->base.name, connector->base.base.id);
+
+	ret = hdcp2_disable_encryption(connector);
+
+	if (hdcp2_deauthenticate_port(connector) < 0)
+		DRM_DEBUG_KMS("Port deauth failed.\n");
+
+	connector->hdcp.hdcp2_encrypted = false;
+
+	return ret;
+}
+
+/* Implements the Link Integrity Check for HDCP2.2 */
+static int intel_hdcp2_check_link(struct intel_connector *connector)
+{
+	struct intel_digital_port *intel_dig_port = conn_to_dig_port(connector);
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct intel_hdcp *hdcp = &connector->hdcp;
+	enum port port = connector->encoder->port;
+	int ret = 0;
+
+	mutex_lock(&hdcp->mutex);
+
+	/* hdcp2_check_link is expected only when HDCP2.2 is Enabled */
+	if (hdcp->value != DRM_MODE_CONTENT_PROTECTION_ENABLED ||
+	    !hdcp->hdcp2_encrypted) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	if (WARN_ON(!intel_hdcp2_in_use(connector))) {
+		DRM_ERROR("HDCP2.2 link stopped the encryption, %x\n",
+			  I915_READ(HDCP2_STATUS_DDI(port)));
+		ret = -ENXIO;
+		hdcp->value = DRM_MODE_CONTENT_PROTECTION_DESIRED;
+		schedule_work(&hdcp->prop_work);
+		goto out;
+	}
+
+	ret = hdcp->shim->check_2_2_link(intel_dig_port);
+	if (ret == HDCP_LINK_PROTECTED) {
+		if (hdcp->value != DRM_MODE_CONTENT_PROTECTION_UNDESIRED) {
+			hdcp->value = DRM_MODE_CONTENT_PROTECTION_ENABLED;
+			schedule_work(&hdcp->prop_work);
+		}
+		goto out;
+	}
+
+	if (ret == HDCP_TOPOLOGY_CHANGE) {
+		if (hdcp->value == DRM_MODE_CONTENT_PROTECTION_UNDESIRED)
+			goto out;
+
+		DRM_DEBUG_KMS("HDCP2.2 Downstream topology change\n");
+		ret = hdcp2_authenticate_repeater_topology(connector);
+		if (!ret) {
+			hdcp->value = DRM_MODE_CONTENT_PROTECTION_ENABLED;
+			schedule_work(&hdcp->prop_work);
+			goto out;
+		}
+		DRM_DEBUG_KMS("[%s:%d] Repeater topology auth failed.(%d)\n",
+			      connector->base.name, connector->base.base.id,
+			      ret);
+	} else {
+		DRM_DEBUG_KMS("[%s:%d] HDCP2.2 link failed, retrying auth\n",
+			      connector->base.name, connector->base.base.id);
+	}
+
+	ret = _intel_hdcp2_disable(connector);
+	if (ret) {
+		DRM_ERROR("[%s:%d] Failed to disable hdcp2.2 (%d)\n",
+			  connector->base.name, connector->base.base.id, ret);
+		hdcp->value = DRM_MODE_CONTENT_PROTECTION_DESIRED;
+		schedule_work(&hdcp->prop_work);
+		goto out;
+	}
+
+	ret = _intel_hdcp2_enable(connector);
+	if (ret) {
+		DRM_DEBUG_KMS("[%s:%d] Failed to enable hdcp2.2 (%d)\n",
+			      connector->base.name, connector->base.base.id,
+			      ret);
+		hdcp->value = DRM_MODE_CONTENT_PROTECTION_DESIRED;
+		schedule_work(&hdcp->prop_work);
+		goto out;
+	}
+
+out:
+	mutex_unlock(&hdcp->mutex);
+	return ret;
+}
+
+static void intel_hdcp_check_work(struct work_struct *work)
+{
+	struct intel_hdcp *hdcp = container_of(to_delayed_work(work),
+					       struct intel_hdcp,
+					       check_work);
+	struct intel_connector *connector = intel_hdcp_to_connector(hdcp);
+
+	if (!intel_hdcp2_check_link(connector))
+		schedule_delayed_work(&hdcp->check_work,
+				      DRM_HDCP2_CHECK_PERIOD_MS);
+	else if (!intel_hdcp_check_link(connector))
+		schedule_delayed_work(&hdcp->check_work,
+				      DRM_HDCP_CHECK_PERIOD_MS);
+}
+
+static int i915_hdcp_component_bind(struct device *i915_kdev,
+				    struct device *mei_kdev, void *data)
+{
+	struct drm_i915_private *dev_priv = kdev_to_i915(i915_kdev);
+
+	DRM_DEBUG("I915 HDCP comp bind\n");
+	mutex_lock(&dev_priv->hdcp_comp_mutex);
+	dev_priv->hdcp_master = (struct i915_hdcp_comp_master *)data;
+	dev_priv->hdcp_master->mei_dev = mei_kdev;
+	mutex_unlock(&dev_priv->hdcp_comp_mutex);
+
+	return 0;
+}
+
+static void i915_hdcp_component_unbind(struct device *i915_kdev,
+				       struct device *mei_kdev, void *data)
+{
+	struct drm_i915_private *dev_priv = kdev_to_i915(i915_kdev);
+
+	DRM_DEBUG("I915 HDCP comp unbind\n");
+	mutex_lock(&dev_priv->hdcp_comp_mutex);
+	dev_priv->hdcp_master = NULL;
+	mutex_unlock(&dev_priv->hdcp_comp_mutex);
+}
+
+static const struct component_ops i915_hdcp_component_ops = {
+	.bind   = i915_hdcp_component_bind,
+	.unbind = i915_hdcp_component_unbind,
+};
+
+static inline int initialize_hdcp_port_data(struct intel_connector *connector)
+{
+	struct intel_hdcp *hdcp = &connector->hdcp;
+	struct hdcp_port_data *data = &hdcp->port_data;
+
+	data->port = connector->encoder->port;
+	data->port_type = (u8)HDCP_PORT_TYPE_INTEGRATED;
+	data->protocol = (u8)hdcp->shim->protocol;
+
+	data->k = 1;
+	if (!data->streams)
+		data->streams = kcalloc(data->k,
+					sizeof(struct hdcp2_streamid_type),
+					GFP_KERNEL);
+	if (!data->streams) {
+		DRM_ERROR("Out of Memory\n");
+		return -ENOMEM;
+	}
+
+	data->streams[0].stream_id = 0;
+	data->streams[0].stream_type = hdcp->content_type;
+
+	return 0;
+}
+
+static bool is_hdcp2_supported(struct drm_i915_private *dev_priv)
+{
+	if (!IS_ENABLED(CONFIG_INTEL_MEI_HDCP))
+		return false;
+
+	return (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv) ||
+		IS_KABYLAKE(dev_priv));
+}
+
+void intel_hdcp_component_init(struct drm_i915_private *dev_priv)
+{
+	int ret;
+
+	if (!is_hdcp2_supported(dev_priv))
+		return;
+
+	mutex_lock(&dev_priv->hdcp_comp_mutex);
+	WARN_ON(dev_priv->hdcp_comp_added);
+
+	dev_priv->hdcp_comp_added = true;
+	mutex_unlock(&dev_priv->hdcp_comp_mutex);
+	ret = component_add_typed(dev_priv->drm.dev, &i915_hdcp_component_ops,
+				  I915_COMPONENT_HDCP);
+	if (ret < 0) {
+		DRM_DEBUG_KMS("Failed at component add(%d)\n", ret);
+		mutex_lock(&dev_priv->hdcp_comp_mutex);
+		dev_priv->hdcp_comp_added = false;
+		mutex_unlock(&dev_priv->hdcp_comp_mutex);
+		return;
+	}
+}
+
+static void intel_hdcp2_init(struct intel_connector *connector)
+{
+	struct intel_hdcp *hdcp = &connector->hdcp;
+	int ret;
+
+	ret = initialize_hdcp_port_data(connector);
+	if (ret) {
+		DRM_DEBUG_KMS("Mei hdcp data init failed\n");
+		return;
+	}
+
+	hdcp->hdcp2_supported = true;
+}
+
 int intel_hdcp_init(struct intel_connector *connector,
 		    const struct intel_hdcp_shim *shim)
 {
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
 	struct intel_hdcp *hdcp = &connector->hdcp;
 	int ret;
 
-	ret = drm_connector_attach_content_protection_property(
-			&connector->base);
+	if (!shim)
+		return -EINVAL;
+
+	ret = drm_connector_attach_content_protection_property(&connector->base);
 	if (ret)
 		return ret;
 
@@ -788,28 +1806,47 @@ int intel_hdcp_init(struct intel_connector *connector,
 	mutex_init(&hdcp->mutex);
 	INIT_DELAYED_WORK(&hdcp->check_work, intel_hdcp_check_work);
 	INIT_WORK(&hdcp->prop_work, intel_hdcp_prop_work);
+
+	if (is_hdcp2_supported(dev_priv))
+		intel_hdcp2_init(connector);
+	init_waitqueue_head(&hdcp->cp_irq_queue);
+
 	return 0;
 }
 
 int intel_hdcp_enable(struct intel_connector *connector)
 {
 	struct intel_hdcp *hdcp = &connector->hdcp;
-	int ret;
+	unsigned long check_link_interval = DRM_HDCP_CHECK_PERIOD_MS;
+	int ret = -EINVAL;
 
 	if (!hdcp->shim)
 		return -ENOENT;
 
 	mutex_lock(&hdcp->mutex);
+	WARN_ON(hdcp->value == DRM_MODE_CONTENT_PROTECTION_ENABLED);
 
-	ret = _intel_hdcp_enable(connector);
-	if (ret)
-		goto out;
+	/*
+	 * Considering that HDCP2.2 is more secure than HDCP1.4, If the setup
+	 * is capable of HDCP2.2, it is preferred to use HDCP2.2.
+	 */
+	if (intel_hdcp2_capable(connector)) {
+		ret = _intel_hdcp2_enable(connector);
+		if (!ret)
+			check_link_interval = DRM_HDCP2_CHECK_PERIOD_MS;
+	}
+
+	/* When HDCP2.2 fails, HDCP1.4 will be attempted */
+	if (ret && intel_hdcp_capable(connector)) {
+		ret = _intel_hdcp_enable(connector);
+	}
+
+	if (!ret) {
+		schedule_delayed_work(&hdcp->check_work, check_link_interval);
+		hdcp->value = DRM_MODE_CONTENT_PROTECTION_ENABLED;
+		schedule_work(&hdcp->prop_work);
+	}
 
-	hdcp->value = DRM_MODE_CONTENT_PROTECTION_ENABLED;
-	schedule_work(&hdcp->prop_work);
-	schedule_delayed_work(&hdcp->check_work,
-			      DRM_HDCP_CHECK_PERIOD_MS);
-out:
 	mutex_unlock(&hdcp->mutex);
 	return ret;
 }
@@ -826,7 +1863,10 @@ int intel_hdcp_disable(struct intel_connector *connector)
 
 	if (hdcp->value != DRM_MODE_CONTENT_PROTECTION_UNDESIRED) {
 		hdcp->value = DRM_MODE_CONTENT_PROTECTION_UNDESIRED;
-		ret = _intel_hdcp_disable(connector);
+		if (hdcp->hdcp2_encrypted)
+			ret = _intel_hdcp2_disable(connector);
+		else if (hdcp->hdcp_encrypted)
+			ret = _intel_hdcp_disable(connector);
 	}
 
 	mutex_unlock(&hdcp->mutex);
@@ -834,6 +1874,30 @@ int intel_hdcp_disable(struct intel_connector *connector)
 	return ret;
 }
 
+void intel_hdcp_component_fini(struct drm_i915_private *dev_priv)
+{
+	mutex_lock(&dev_priv->hdcp_comp_mutex);
+	if (!dev_priv->hdcp_comp_added) {
+		mutex_unlock(&dev_priv->hdcp_comp_mutex);
+		return;
+	}
+
+	dev_priv->hdcp_comp_added = false;
+	mutex_unlock(&dev_priv->hdcp_comp_mutex);
+
+	component_del(dev_priv->drm.dev, &i915_hdcp_component_ops);
+}
+
+void intel_hdcp_cleanup(struct intel_connector *connector)
+{
+	if (!connector->hdcp.shim)
+		return;
+
+	mutex_lock(&connector->hdcp.mutex);
+	kfree(connector->hdcp.port_data.streams);
+	mutex_unlock(&connector->hdcp.mutex);
+}
+
 void intel_hdcp_atomic_check(struct drm_connector *connector,
 			     struct drm_connector_state *old_state,
 			     struct drm_connector_state *new_state)
@@ -867,61 +1931,16 @@ void intel_hdcp_atomic_check(struct drm_connector *connector,
 	crtc_state->mode_changed = true;
 }
 
-/* Implements Part 3 of the HDCP authorization procedure */
-int intel_hdcp_check_link(struct intel_connector *connector)
+/* Handles the CP_IRQ raised from the DP HDCP sink */
+void intel_hdcp_handle_cp_irq(struct intel_connector *connector)
 {
 	struct intel_hdcp *hdcp = &connector->hdcp;
-	struct drm_i915_private *dev_priv = connector->base.dev->dev_private;
-	struct intel_digital_port *intel_dig_port = conn_to_dig_port(connector);
-	enum port port = intel_dig_port->base.port;
-	int ret = 0;
 
 	if (!hdcp->shim)
-		return -ENOENT;
-
-	mutex_lock(&hdcp->mutex);
-
-	if (hdcp->value == DRM_MODE_CONTENT_PROTECTION_UNDESIRED)
-		goto out;
-
-	if (!(I915_READ(PORT_HDCP_STATUS(port)) & HDCP_STATUS_ENC)) {
-		DRM_ERROR("%s:%d HDCP check failed: link is not encrypted,%x\n",
-			  connector->base.name, connector->base.base.id,
-			  I915_READ(PORT_HDCP_STATUS(port)));
-		ret = -ENXIO;
-		hdcp->value = DRM_MODE_CONTENT_PROTECTION_DESIRED;
-		schedule_work(&hdcp->prop_work);
-		goto out;
-	}
-
-	if (hdcp->shim->check_link(intel_dig_port)) {
-		if (hdcp->value != DRM_MODE_CONTENT_PROTECTION_UNDESIRED) {
-			hdcp->value = DRM_MODE_CONTENT_PROTECTION_ENABLED;
-			schedule_work(&hdcp->prop_work);
-		}
-		goto out;
-	}
-
-	DRM_DEBUG_KMS("[%s:%d] HDCP link failed, retrying authentication\n",
-		      connector->base.name, connector->base.base.id);
-
-	ret = _intel_hdcp_disable(connector);
-	if (ret) {
-		DRM_ERROR("Failed to disable hdcp (%d)\n", ret);
-		hdcp->value = DRM_MODE_CONTENT_PROTECTION_DESIRED;
-		schedule_work(&hdcp->prop_work);
-		goto out;
-	}
+		return;
 
-	ret = _intel_hdcp_enable(connector);
-	if (ret) {
-		DRM_DEBUG_KMS("Failed to enable hdcp (%d)\n", ret);
-		hdcp->value = DRM_MODE_CONTENT_PROTECTION_DESIRED;
-		schedule_work(&hdcp->prop_work);
-		goto out;
-	}
+	atomic_inc(&connector->hdcp.cp_irq_count);
+	wake_up_all(&connector->hdcp.cp_irq_queue);
 
-out:
-	mutex_unlock(&hdcp->mutex);
-	return ret;
+	schedule_delayed_work(&hdcp->check_work, 0);
 }
diff --git a/drivers/gpu/drm/i915/intel_hdcp.h b/drivers/gpu/drm/i915/intel_hdcp.h
new file mode 100644
index 000000000000..a75f25f09d39
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_hdcp.h
@@ -0,0 +1,33 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_HDCP_H__
+#define __INTEL_HDCP_H__
+
+#include <linux/types.h>
+
+#include <drm/i915_drm.h>
+
+struct drm_connector;
+struct drm_connector_state;
+struct drm_i915_private;
+struct intel_connector;
+struct intel_hdcp_shim;
+
+void intel_hdcp_atomic_check(struct drm_connector *connector,
+			     struct drm_connector_state *old_state,
+			     struct drm_connector_state *new_state);
+int intel_hdcp_init(struct intel_connector *connector,
+		    const struct intel_hdcp_shim *hdcp_shim);
+int intel_hdcp_enable(struct intel_connector *connector);
+int intel_hdcp_disable(struct intel_connector *connector);
+bool is_hdcp_supported(struct drm_i915_private *dev_priv, enum port port);
+bool intel_hdcp_capable(struct intel_connector *connector);
+void intel_hdcp_component_init(struct drm_i915_private *dev_priv);
+void intel_hdcp_component_fini(struct drm_i915_private *dev_priv);
+void intel_hdcp_cleanup(struct intel_connector *connector);
+void intel_hdcp_handle_cp_irq(struct intel_connector *connector);
+
+#endif /* __INTEL_HDCP_H__ */
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index a46bffe2b288..34be2cfd0ec8 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -26,19 +26,30 @@
  *	Jesse Barnes <jesse.barnes@intel.com>
  */
 
-#include <linux/i2c.h>
-#include <linux/slab.h>
 #include <linux/delay.h>
 #include <linux/hdmi.h>
+#include <linux/i2c.h>
+#include <linux/slab.h>
+
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_hdcp.h>
 #include <drm/drm_scdc_helper.h>
-#include "intel_drv.h"
 #include <drm/i915_drm.h>
 #include <drm/intel_lpe_audio.h>
+
 #include "i915_drv.h"
+#include "intel_audio.h"
+#include "intel_connector.h"
+#include "intel_ddi.h"
+#include "intel_dp.h"
+#include "intel_drv.h"
+#include "intel_hdcp.h"
+#include "intel_hdmi.h"
+#include "intel_lspcon.h"
+#include "intel_sdvo.h"
+#include "intel_panel.h"
 
 static struct drm_device *intel_hdmi_to_dev(struct intel_hdmi *intel_hdmi)
 {
@@ -82,6 +93,8 @@ static struct intel_hdmi *intel_attached_hdmi(struct drm_connector *connector)
 static u32 g4x_infoframe_index(unsigned int type)
 {
 	switch (type) {
+	case HDMI_PACKET_TYPE_GAMUT_METADATA:
+		return VIDEO_DIP_SELECT_GAMUT;
 	case HDMI_INFOFRAME_TYPE_AVI:
 		return VIDEO_DIP_SELECT_AVI;
 	case HDMI_INFOFRAME_TYPE_SPD:
@@ -97,6 +110,12 @@ static u32 g4x_infoframe_index(unsigned int type)
 static u32 g4x_infoframe_enable(unsigned int type)
 {
 	switch (type) {
+	case HDMI_PACKET_TYPE_GENERAL_CONTROL:
+		return VIDEO_DIP_ENABLE_GCP;
+	case HDMI_PACKET_TYPE_GAMUT_METADATA:
+		return VIDEO_DIP_ENABLE_GAMUT;
+	case DP_SDP_VSC:
+		return 0;
 	case HDMI_INFOFRAME_TYPE_AVI:
 		return VIDEO_DIP_ENABLE_AVI;
 	case HDMI_INFOFRAME_TYPE_SPD:
@@ -112,6 +131,10 @@ static u32 g4x_infoframe_enable(unsigned int type)
 static u32 hsw_infoframe_enable(unsigned int type)
 {
 	switch (type) {
+	case HDMI_PACKET_TYPE_GENERAL_CONTROL:
+		return VIDEO_DIP_ENABLE_GCP_HSW;
+	case HDMI_PACKET_TYPE_GAMUT_METADATA:
+		return VIDEO_DIP_ENABLE_GMP_HSW;
 	case DP_SDP_VSC:
 		return VIDEO_DIP_ENABLE_VSC_HSW;
 	case DP_SDP_PPS:
@@ -135,6 +158,8 @@ hsw_dip_data_reg(struct drm_i915_private *dev_priv,
 		 int i)
 {
 	switch (type) {
+	case HDMI_PACKET_TYPE_GAMUT_METADATA:
+		return HSW_TVIDEO_DIP_GMP_DATA(cpu_transcoder, i);
 	case DP_SDP_VSC:
 		return HSW_TVIDEO_DIP_VSC_DATA(cpu_transcoder, i);
 	case DP_SDP_PPS:
@@ -198,17 +223,37 @@ static void g4x_write_infoframe(struct intel_encoder *encoder,
 	POSTING_READ(VIDEO_DIP_CTL);
 }
 
-static bool g4x_infoframe_enabled(struct intel_encoder *encoder,
+static void g4x_read_infoframe(struct intel_encoder *encoder,
+			       const struct intel_crtc_state *crtc_state,
+			       unsigned int type,
+			       void *frame, ssize_t len)
+{
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	u32 val, *data = frame;
+	int i;
+
+	val = I915_READ(VIDEO_DIP_CTL);
+
+	val &= ~(VIDEO_DIP_SELECT_MASK | 0xf); /* clear DIP data offset */
+	val |= g4x_infoframe_index(type);
+
+	I915_WRITE(VIDEO_DIP_CTL, val);
+
+	for (i = 0; i < len; i += 4)
+		*data++ = I915_READ(VIDEO_DIP_DATA);
+}
+
+static u32 g4x_infoframes_enabled(struct intel_encoder *encoder,
 				  const struct intel_crtc_state *pipe_config)
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	u32 val = I915_READ(VIDEO_DIP_CTL);
 
 	if ((val & VIDEO_DIP_ENABLE) == 0)
-		return false;
+		return 0;
 
 	if ((val & VIDEO_DIP_PORT_MASK) != VIDEO_DIP_PORT(encoder->port))
-		return false;
+		return 0;
 
 	return val & (VIDEO_DIP_ENABLE_AVI |
 		      VIDEO_DIP_ENABLE_VENDOR | VIDEO_DIP_ENABLE_SPD);
@@ -251,7 +296,28 @@ static void ibx_write_infoframe(struct intel_encoder *encoder,
 	POSTING_READ(reg);
 }
 
-static bool ibx_infoframe_enabled(struct intel_encoder *encoder,
+static void ibx_read_infoframe(struct intel_encoder *encoder,
+			       const struct intel_crtc_state *crtc_state,
+			       unsigned int type,
+			       void *frame, ssize_t len)
+{
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
+	u32 val, *data = frame;
+	int i;
+
+	val = I915_READ(TVIDEO_DIP_CTL(crtc->pipe));
+
+	val &= ~(VIDEO_DIP_SELECT_MASK | 0xf); /* clear DIP data offset */
+	val |= g4x_infoframe_index(type);
+
+	I915_WRITE(TVIDEO_DIP_CTL(crtc->pipe), val);
+
+	for (i = 0; i < len; i += 4)
+		*data++ = I915_READ(TVIDEO_DIP_DATA(crtc->pipe));
+}
+
+static u32 ibx_infoframes_enabled(struct intel_encoder *encoder,
 				  const struct intel_crtc_state *pipe_config)
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
@@ -260,10 +326,10 @@ static bool ibx_infoframe_enabled(struct intel_encoder *encoder,
 	u32 val = I915_READ(reg);
 
 	if ((val & VIDEO_DIP_ENABLE) == 0)
-		return false;
+		return 0;
 
 	if ((val & VIDEO_DIP_PORT_MASK) != VIDEO_DIP_PORT(encoder->port))
-		return false;
+		return 0;
 
 	return val & (VIDEO_DIP_ENABLE_AVI |
 		      VIDEO_DIP_ENABLE_VENDOR | VIDEO_DIP_ENABLE_GAMUT |
@@ -310,7 +376,28 @@ static void cpt_write_infoframe(struct intel_encoder *encoder,
 	POSTING_READ(reg);
 }
 
-static bool cpt_infoframe_enabled(struct intel_encoder *encoder,
+static void cpt_read_infoframe(struct intel_encoder *encoder,
+			       const struct intel_crtc_state *crtc_state,
+			       unsigned int type,
+			       void *frame, ssize_t len)
+{
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
+	u32 val, *data = frame;
+	int i;
+
+	val = I915_READ(TVIDEO_DIP_CTL(crtc->pipe));
+
+	val &= ~(VIDEO_DIP_SELECT_MASK | 0xf); /* clear DIP data offset */
+	val |= g4x_infoframe_index(type);
+
+	I915_WRITE(TVIDEO_DIP_CTL(crtc->pipe), val);
+
+	for (i = 0; i < len; i += 4)
+		*data++ = I915_READ(TVIDEO_DIP_DATA(crtc->pipe));
+}
+
+static u32 cpt_infoframes_enabled(struct intel_encoder *encoder,
 				  const struct intel_crtc_state *pipe_config)
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
@@ -318,7 +405,7 @@ static bool cpt_infoframe_enabled(struct intel_encoder *encoder,
 	u32 val = I915_READ(TVIDEO_DIP_CTL(pipe));
 
 	if ((val & VIDEO_DIP_ENABLE) == 0)
-		return false;
+		return 0;
 
 	return val & (VIDEO_DIP_ENABLE_AVI |
 		      VIDEO_DIP_ENABLE_VENDOR | VIDEO_DIP_ENABLE_GAMUT |
@@ -362,7 +449,28 @@ static void vlv_write_infoframe(struct intel_encoder *encoder,
 	POSTING_READ(reg);
 }
 
-static bool vlv_infoframe_enabled(struct intel_encoder *encoder,
+static void vlv_read_infoframe(struct intel_encoder *encoder,
+			       const struct intel_crtc_state *crtc_state,
+			       unsigned int type,
+			       void *frame, ssize_t len)
+{
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
+	u32 val, *data = frame;
+	int i;
+
+	val = I915_READ(VLV_TVIDEO_DIP_CTL(crtc->pipe));
+
+	val &= ~(VIDEO_DIP_SELECT_MASK | 0xf); /* clear DIP data offset */
+	val |= g4x_infoframe_index(type);
+
+	I915_WRITE(VLV_TVIDEO_DIP_CTL(crtc->pipe), val);
+
+	for (i = 0; i < len; i += 4)
+		*data++ = I915_READ(VLV_TVIDEO_DIP_DATA(crtc->pipe));
+}
+
+static u32 vlv_infoframes_enabled(struct intel_encoder *encoder,
 				  const struct intel_crtc_state *pipe_config)
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
@@ -370,10 +478,10 @@ static bool vlv_infoframe_enabled(struct intel_encoder *encoder,
 	u32 val = I915_READ(VLV_TVIDEO_DIP_CTL(pipe));
 
 	if ((val & VIDEO_DIP_ENABLE) == 0)
-		return false;
+		return 0;
 
 	if ((val & VIDEO_DIP_PORT_MASK) != VIDEO_DIP_PORT(encoder->port))
-		return false;
+		return 0;
 
 	return val & (VIDEO_DIP_ENABLE_AVI |
 		      VIDEO_DIP_ENABLE_VENDOR | VIDEO_DIP_ENABLE_GAMUT |
@@ -413,7 +521,24 @@ static void hsw_write_infoframe(struct intel_encoder *encoder,
 	POSTING_READ(ctl_reg);
 }
 
-static bool hsw_infoframe_enabled(struct intel_encoder *encoder,
+static void hsw_read_infoframe(struct intel_encoder *encoder,
+			       const struct intel_crtc_state *crtc_state,
+			       unsigned int type,
+			       void *frame, ssize_t len)
+{
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
+	u32 val, *data = frame;
+	int i;
+
+	val = I915_READ(HSW_TVIDEO_DIP_CTL(cpu_transcoder));
+
+	for (i = 0; i < len; i += 4)
+		*data++ = I915_READ(hsw_dip_data_reg(dev_priv, cpu_transcoder,
+						     type, i >> 2));
+}
+
+static u32 hsw_infoframes_enabled(struct intel_encoder *encoder,
 				  const struct intel_crtc_state *pipe_config)
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
@@ -424,6 +549,53 @@ static bool hsw_infoframe_enabled(struct intel_encoder *encoder,
 		      VIDEO_DIP_ENABLE_GMP_HSW | VIDEO_DIP_ENABLE_SPD_HSW);
 }
 
+static const u8 infoframe_type_to_idx[] = {
+	HDMI_PACKET_TYPE_GENERAL_CONTROL,
+	HDMI_PACKET_TYPE_GAMUT_METADATA,
+	DP_SDP_VSC,
+	HDMI_INFOFRAME_TYPE_AVI,
+	HDMI_INFOFRAME_TYPE_SPD,
+	HDMI_INFOFRAME_TYPE_VENDOR,
+};
+
+u32 intel_hdmi_infoframe_enable(unsigned int type)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(infoframe_type_to_idx); i++) {
+		if (infoframe_type_to_idx[i] == type)
+			return BIT(i);
+	}
+
+	return 0;
+}
+
+u32 intel_hdmi_infoframes_enabled(struct intel_encoder *encoder,
+				  const struct intel_crtc_state *crtc_state)
+{
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_digital_port *dig_port = enc_to_dig_port(&encoder->base);
+	u32 val, ret = 0;
+	int i;
+
+	val = dig_port->infoframes_enabled(encoder, crtc_state);
+
+	/* map from hardware bits to dip idx */
+	for (i = 0; i < ARRAY_SIZE(infoframe_type_to_idx); i++) {
+		unsigned int type = infoframe_type_to_idx[i];
+
+		if (HAS_DDI(dev_priv)) {
+			if (val & hsw_infoframe_enable(type))
+				ret |= BIT(i);
+		} else {
+			if (val & g4x_infoframe_enable(type))
+				ret |= BIT(i);
+		}
+	}
+
+	return ret;
+}
+
 /*
  * The data we write to the DIP data buffer registers is 1 byte bigger than the
  * HDMI infoframe size because of an ECC/reserved byte at position 3 (starting
@@ -443,15 +615,23 @@ static bool hsw_infoframe_enabled(struct intel_encoder *encoder,
  */
 static void intel_write_infoframe(struct intel_encoder *encoder,
 				  const struct intel_crtc_state *crtc_state,
-				  union hdmi_infoframe *frame)
+				  enum hdmi_infoframe_type type,
+				  const union hdmi_infoframe *frame)
 {
 	struct intel_digital_port *intel_dig_port = enc_to_dig_port(&encoder->base);
 	u8 buffer[VIDEO_DIP_DATA_SIZE];
 	ssize_t len;
 
+	if ((crtc_state->infoframes.enable &
+	     intel_hdmi_infoframe_enable(type)) == 0)
+		return;
+
+	if (WARN_ON(frame->any.type != type))
+		return;
+
 	/* see comment above for the reason for this offset */
-	len = hdmi_infoframe_pack(frame, buffer + 1, sizeof(buffer) - 1);
-	if (len < 0)
+	len = hdmi_infoframe_pack_only(frame, buffer + 1, sizeof(buffer) - 1);
+	if (WARN_ON(len < 0))
 		return;
 
 	/* Insert the 'hole' (see big comment above) at position 3 */
@@ -459,84 +639,143 @@ static void intel_write_infoframe(struct intel_encoder *encoder,
 	buffer[3] = 0;
 	len++;
 
-	intel_dig_port->write_infoframe(encoder,
-					crtc_state,
-					frame->any.type, buffer, len);
+	intel_dig_port->write_infoframe(encoder, crtc_state, type, buffer, len);
 }
 
-static void intel_hdmi_set_avi_infoframe(struct intel_encoder *encoder,
-					 const struct intel_crtc_state *crtc_state,
-					 const struct drm_connector_state *conn_state)
+void intel_read_infoframe(struct intel_encoder *encoder,
+			  const struct intel_crtc_state *crtc_state,
+			  enum hdmi_infoframe_type type,
+			  union hdmi_infoframe *frame)
+{
+	struct intel_digital_port *intel_dig_port = enc_to_dig_port(&encoder->base);
+	u8 buffer[VIDEO_DIP_DATA_SIZE];
+	int ret;
+
+	if ((crtc_state->infoframes.enable &
+	     intel_hdmi_infoframe_enable(type)) == 0)
+		return;
+
+	intel_dig_port->read_infoframe(encoder, crtc_state,
+				       type, buffer, sizeof(buffer));
+
+	/* Fill the 'hole' (see big comment above) at position 3 */
+	memmove(&buffer[1], &buffer[0], 3);
+
+	/* see comment above for the reason for this offset */
+	ret = hdmi_infoframe_unpack(frame, buffer + 1, sizeof(buffer) - 1);
+	if (ret) {
+		DRM_DEBUG_KMS("Failed to unpack infoframe type 0x%02x\n", type);
+		return;
+	}
+
+	if (frame->any.type != type)
+		DRM_DEBUG_KMS("Found the wrong infoframe type 0x%x (expected 0x%02x)\n",
+			      frame->any.type, type);
+}
+
+static bool
+intel_hdmi_compute_avi_infoframe(struct intel_encoder *encoder,
+				 struct intel_crtc_state *crtc_state,
+				 struct drm_connector_state *conn_state)
 {
+	struct hdmi_avi_infoframe *frame = &crtc_state->infoframes.avi.avi;
 	const struct drm_display_mode *adjusted_mode =
 		&crtc_state->base.adjusted_mode;
-	union hdmi_infoframe frame;
+	struct drm_connector *connector = conn_state->connector;
 	int ret;
 
-	ret = drm_hdmi_avi_infoframe_from_display_mode(&frame.avi,
-						       conn_state->connector,
+	if (!crtc_state->has_infoframe)
+		return true;
+
+	crtc_state->infoframes.enable |=
+		intel_hdmi_infoframe_enable(HDMI_INFOFRAME_TYPE_AVI);
+
+	ret = drm_hdmi_avi_infoframe_from_display_mode(frame, connector,
 						       adjusted_mode);
-	if (ret < 0) {
-		DRM_ERROR("couldn't fill AVI infoframe\n");
-		return;
-	}
+	if (ret)
+		return false;
 
 	if (crtc_state->output_format == INTEL_OUTPUT_FORMAT_YCBCR420)
-		frame.avi.colorspace = HDMI_COLORSPACE_YUV420;
+		frame->colorspace = HDMI_COLORSPACE_YUV420;
 	else if (crtc_state->output_format == INTEL_OUTPUT_FORMAT_YCBCR444)
-		frame.avi.colorspace = HDMI_COLORSPACE_YUV444;
+		frame->colorspace = HDMI_COLORSPACE_YUV444;
 	else
-		frame.avi.colorspace = HDMI_COLORSPACE_RGB;
+		frame->colorspace = HDMI_COLORSPACE_RGB;
+
+	drm_hdmi_avi_infoframe_colorspace(frame, conn_state);
 
-	drm_hdmi_avi_infoframe_quant_range(&frame.avi,
-					   conn_state->connector,
+	drm_hdmi_avi_infoframe_quant_range(frame, connector,
 					   adjusted_mode,
 					   crtc_state->limited_color_range ?
 					   HDMI_QUANTIZATION_RANGE_LIMITED :
 					   HDMI_QUANTIZATION_RANGE_FULL);
 
-	drm_hdmi_avi_infoframe_content_type(&frame.avi,
-					    conn_state);
+	drm_hdmi_avi_infoframe_content_type(frame, conn_state);
 
 	/* TODO: handle pixel repetition for YCBCR420 outputs */
-	intel_write_infoframe(encoder, crtc_state,
-			      &frame);
+
+	ret = hdmi_avi_infoframe_check(frame);
+	if (WARN_ON(ret))
+		return false;
+
+	return true;
 }
 
-static void intel_hdmi_set_spd_infoframe(struct intel_encoder *encoder,
-					 const struct intel_crtc_state *crtc_state)
+static bool
+intel_hdmi_compute_spd_infoframe(struct intel_encoder *encoder,
+				 struct intel_crtc_state *crtc_state,
+				 struct drm_connector_state *conn_state)
 {
-	union hdmi_infoframe frame;
+	struct hdmi_spd_infoframe *frame = &crtc_state->infoframes.spd.spd;
 	int ret;
 
-	ret = hdmi_spd_infoframe_init(&frame.spd, "Intel", "Integrated gfx");
-	if (ret < 0) {
-		DRM_ERROR("couldn't fill SPD infoframe\n");
-		return;
-	}
+	if (!crtc_state->has_infoframe)
+		return true;
 
-	frame.spd.sdi = HDMI_SPD_SDI_PC;
+	crtc_state->infoframes.enable |=
+		intel_hdmi_infoframe_enable(HDMI_INFOFRAME_TYPE_SPD);
 
-	intel_write_infoframe(encoder, crtc_state,
-			      &frame);
+	ret = hdmi_spd_infoframe_init(frame, "Intel", "Integrated gfx");
+	if (WARN_ON(ret))
+		return false;
+
+	frame->sdi = HDMI_SPD_SDI_PC;
+
+	ret = hdmi_spd_infoframe_check(frame);
+	if (WARN_ON(ret))
+		return false;
+
+	return true;
 }
 
-static void
-intel_hdmi_set_hdmi_infoframe(struct intel_encoder *encoder,
-			      const struct intel_crtc_state *crtc_state,
-			      const struct drm_connector_state *conn_state)
-{
-	union hdmi_infoframe frame;
+static bool
+intel_hdmi_compute_hdmi_infoframe(struct intel_encoder *encoder,
+				  struct intel_crtc_state *crtc_state,
+				  struct drm_connector_state *conn_state)
+{
+	struct hdmi_vendor_infoframe *frame =
+		&crtc_state->infoframes.hdmi.vendor.hdmi;
+	const struct drm_display_info *info =
+		&conn_state->connector->display_info;
 	int ret;
 
-	ret = drm_hdmi_vendor_infoframe_from_display_mode(&frame.vendor.hdmi,
+	if (!crtc_state->has_infoframe || !info->has_hdmi_infoframe)
+		return true;
+
+	crtc_state->infoframes.enable |=
+		intel_hdmi_infoframe_enable(HDMI_INFOFRAME_TYPE_VENDOR);
+
+	ret = drm_hdmi_vendor_infoframe_from_display_mode(frame,
 							  conn_state->connector,
 							  &crtc_state->base.adjusted_mode);
-	if (ret < 0)
-		return;
+	if (WARN_ON(ret))
+		return false;
 
-	intel_write_infoframe(encoder, crtc_state,
-			      &frame);
+	ret = hdmi_vendor_infoframe_check(frame);
+	if (WARN_ON(ret))
+		return false;
+
+	return true;
 }
 
 static void g4x_set_infoframes(struct intel_encoder *encoder,
@@ -596,9 +835,15 @@ static void g4x_set_infoframes(struct intel_encoder *encoder,
 	I915_WRITE(reg, val);
 	POSTING_READ(reg);
 
-	intel_hdmi_set_avi_infoframe(encoder, crtc_state, conn_state);
-	intel_hdmi_set_spd_infoframe(encoder, crtc_state);
-	intel_hdmi_set_hdmi_infoframe(encoder, crtc_state, conn_state);
+	intel_write_infoframe(encoder, crtc_state,
+			      HDMI_INFOFRAME_TYPE_AVI,
+			      &crtc_state->infoframes.avi);
+	intel_write_infoframe(encoder, crtc_state,
+			      HDMI_INFOFRAME_TYPE_SPD,
+			      &crtc_state->infoframes.spd);
+	intel_write_infoframe(encoder, crtc_state,
+			      HDMI_INFOFRAME_TYPE_VENDOR,
+			      &crtc_state->infoframes.hdmi);
 }
 
 static bool hdmi_sink_is_deep_color(const struct drm_connector_state *conn_state)
@@ -664,7 +909,10 @@ static bool intel_hdmi_set_gcp_infoframe(struct intel_encoder *encoder,
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
 	i915_reg_t reg;
-	u32 val = 0;
+
+	if ((crtc_state->infoframes.enable &
+	     intel_hdmi_infoframe_enable(HDMI_PACKET_TYPE_GENERAL_CONTROL)) == 0)
+		return false;
 
 	if (HAS_DDI(dev_priv))
 		reg = HSW_TVIDEO_DIP_GCP(crtc_state->cpu_transcoder);
@@ -675,18 +923,54 @@ static bool intel_hdmi_set_gcp_infoframe(struct intel_encoder *encoder,
 	else
 		return false;
 
+	I915_WRITE(reg, crtc_state->infoframes.gcp);
+
+	return true;
+}
+
+void intel_hdmi_read_gcp_infoframe(struct intel_encoder *encoder,
+				   struct intel_crtc_state *crtc_state)
+{
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
+	i915_reg_t reg;
+
+	if ((crtc_state->infoframes.enable &
+	     intel_hdmi_infoframe_enable(HDMI_PACKET_TYPE_GENERAL_CONTROL)) == 0)
+		return;
+
+	if (HAS_DDI(dev_priv))
+		reg = HSW_TVIDEO_DIP_GCP(crtc_state->cpu_transcoder);
+	else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
+		reg = VLV_TVIDEO_DIP_GCP(crtc->pipe);
+	else if (HAS_PCH_SPLIT(dev_priv))
+		reg = TVIDEO_DIP_GCP(crtc->pipe);
+	else
+		return;
+
+	crtc_state->infoframes.gcp = I915_READ(reg);
+}
+
+static void intel_hdmi_compute_gcp_infoframe(struct intel_encoder *encoder,
+					     struct intel_crtc_state *crtc_state,
+					     struct drm_connector_state *conn_state)
+{
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+
+	if (IS_G4X(dev_priv) || !crtc_state->has_infoframe)
+		return;
+
+	crtc_state->infoframes.enable |=
+		intel_hdmi_infoframe_enable(HDMI_PACKET_TYPE_GENERAL_CONTROL);
+
 	/* Indicate color depth whenever the sink supports deep color */
 	if (hdmi_sink_is_deep_color(conn_state))
-		val |= GCP_COLOR_INDICATION;
+		crtc_state->infoframes.gcp |= GCP_COLOR_INDICATION;
 
 	/* Enable default_phase whenever the display mode is suitably aligned */
 	if (gcp_default_phase_possible(crtc_state->pipe_bpp,
 				       &crtc_state->base.adjusted_mode))
-		val |= GCP_DEFAULT_PHASE_ENABLE;
-
-	I915_WRITE(reg, val);
-
-	return val != 0;
+		crtc_state->infoframes.gcp |= GCP_DEFAULT_PHASE_ENABLE;
 }
 
 static void ibx_set_infoframes(struct intel_encoder *encoder,
@@ -737,9 +1021,15 @@ static void ibx_set_infoframes(struct intel_encoder *encoder,
 	I915_WRITE(reg, val);
 	POSTING_READ(reg);
 
-	intel_hdmi_set_avi_infoframe(encoder, crtc_state, conn_state);
-	intel_hdmi_set_spd_infoframe(encoder, crtc_state);
-	intel_hdmi_set_hdmi_infoframe(encoder, crtc_state, conn_state);
+	intel_write_infoframe(encoder, crtc_state,
+			      HDMI_INFOFRAME_TYPE_AVI,
+			      &crtc_state->infoframes.avi);
+	intel_write_infoframe(encoder, crtc_state,
+			      HDMI_INFOFRAME_TYPE_SPD,
+			      &crtc_state->infoframes.spd);
+	intel_write_infoframe(encoder, crtc_state,
+			      HDMI_INFOFRAME_TYPE_VENDOR,
+			      &crtc_state->infoframes.hdmi);
 }
 
 static void cpt_set_infoframes(struct intel_encoder *encoder,
@@ -780,9 +1070,15 @@ static void cpt_set_infoframes(struct intel_encoder *encoder,
 	I915_WRITE(reg, val);
 	POSTING_READ(reg);
 
-	intel_hdmi_set_avi_infoframe(encoder, crtc_state, conn_state);
-	intel_hdmi_set_spd_infoframe(encoder, crtc_state);
-	intel_hdmi_set_hdmi_infoframe(encoder, crtc_state, conn_state);
+	intel_write_infoframe(encoder, crtc_state,
+			      HDMI_INFOFRAME_TYPE_AVI,
+			      &crtc_state->infoframes.avi);
+	intel_write_infoframe(encoder, crtc_state,
+			      HDMI_INFOFRAME_TYPE_SPD,
+			      &crtc_state->infoframes.spd);
+	intel_write_infoframe(encoder, crtc_state,
+			      HDMI_INFOFRAME_TYPE_VENDOR,
+			      &crtc_state->infoframes.hdmi);
 }
 
 static void vlv_set_infoframes(struct intel_encoder *encoder,
@@ -832,9 +1128,15 @@ static void vlv_set_infoframes(struct intel_encoder *encoder,
 	I915_WRITE(reg, val);
 	POSTING_READ(reg);
 
-	intel_hdmi_set_avi_infoframe(encoder, crtc_state, conn_state);
-	intel_hdmi_set_spd_infoframe(encoder, crtc_state);
-	intel_hdmi_set_hdmi_infoframe(encoder, crtc_state, conn_state);
+	intel_write_infoframe(encoder, crtc_state,
+			      HDMI_INFOFRAME_TYPE_AVI,
+			      &crtc_state->infoframes.avi);
+	intel_write_infoframe(encoder, crtc_state,
+			      HDMI_INFOFRAME_TYPE_SPD,
+			      &crtc_state->infoframes.spd);
+	intel_write_infoframe(encoder, crtc_state,
+			      HDMI_INFOFRAME_TYPE_VENDOR,
+			      &crtc_state->infoframes.hdmi);
 }
 
 static void hsw_set_infoframes(struct intel_encoder *encoder,
@@ -865,9 +1167,15 @@ static void hsw_set_infoframes(struct intel_encoder *encoder,
 	I915_WRITE(reg, val);
 	POSTING_READ(reg);
 
-	intel_hdmi_set_avi_infoframe(encoder, crtc_state, conn_state);
-	intel_hdmi_set_spd_infoframe(encoder, crtc_state);
-	intel_hdmi_set_hdmi_infoframe(encoder, crtc_state, conn_state);
+	intel_write_infoframe(encoder, crtc_state,
+			      HDMI_INFOFRAME_TYPE_AVI,
+			      &crtc_state->infoframes.avi);
+	intel_write_infoframe(encoder, crtc_state,
+			      HDMI_INFOFRAME_TYPE_SPD,
+			      &crtc_state->infoframes.spd);
+	intel_write_infoframe(encoder, crtc_state,
+			      HDMI_INFOFRAME_TYPE_VENDOR,
+			      &crtc_state->infoframes.hdmi);
 }
 
 void intel_dp_dual_mode_set_tmds_output(struct intel_hdmi *hdmi, bool enable)
@@ -1073,10 +1381,44 @@ int intel_hdmi_hdcp_read_v_prime_part(struct intel_digital_port *intel_dig_port,
 	return ret;
 }
 
+static int kbl_repositioning_enc_en_signal(struct intel_connector *connector)
+{
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct intel_digital_port *intel_dig_port = conn_to_dig_port(connector);
+	struct drm_crtc *crtc = connector->base.state->crtc;
+	struct intel_crtc *intel_crtc = container_of(crtc,
+						     struct intel_crtc, base);
+	u32 scanline;
+	int ret;
+
+	for (;;) {
+		scanline = I915_READ(PIPEDSL(intel_crtc->pipe));
+		if (scanline > 100 && scanline < 200)
+			break;
+		usleep_range(25, 50);
+	}
+
+	ret = intel_ddi_toggle_hdcp_signalling(&intel_dig_port->base, false);
+	if (ret) {
+		DRM_ERROR("Disable HDCP signalling failed (%d)\n", ret);
+		return ret;
+	}
+	ret = intel_ddi_toggle_hdcp_signalling(&intel_dig_port->base, true);
+	if (ret) {
+		DRM_ERROR("Enable HDCP signalling failed (%d)\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
 static
 int intel_hdmi_hdcp_toggle_signalling(struct intel_digital_port *intel_dig_port,
 				      bool enable)
 {
+	struct intel_hdmi *hdmi = &intel_dig_port->hdmi;
+	struct intel_connector *connector = hdmi->attached_connector;
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
 	int ret;
 
 	if (!enable)
@@ -1088,6 +1430,14 @@ int intel_hdmi_hdcp_toggle_signalling(struct intel_digital_port *intel_dig_port,
 			  enable ? "Enable" : "Disable", ret);
 		return ret;
 	}
+
+	/*
+	 * WA: To fix incorrect positioning of the window of
+	 * opportunity and enc_en signalling in KABYLAKE.
+	 */
+	if (IS_KABYLAKE(dev_priv) && enable)
+		return kbl_repositioning_enc_en_signal(connector);
+
 	return 0;
 }
 
@@ -1119,6 +1469,190 @@ bool intel_hdmi_hdcp_check_link(struct intel_digital_port *intel_dig_port)
 	return true;
 }
 
+static struct hdcp2_hdmi_msg_data {
+	u8 msg_id;
+	u32 timeout;
+	u32 timeout2;
+	} hdcp2_msg_data[] = {
+		{HDCP_2_2_AKE_INIT, 0, 0},
+		{HDCP_2_2_AKE_SEND_CERT, HDCP_2_2_CERT_TIMEOUT_MS, 0},
+		{HDCP_2_2_AKE_NO_STORED_KM, 0, 0},
+		{HDCP_2_2_AKE_STORED_KM, 0, 0},
+		{HDCP_2_2_AKE_SEND_HPRIME, HDCP_2_2_HPRIME_PAIRED_TIMEOUT_MS,
+				HDCP_2_2_HPRIME_NO_PAIRED_TIMEOUT_MS},
+		{HDCP_2_2_AKE_SEND_PAIRING_INFO, HDCP_2_2_PAIRING_TIMEOUT_MS,
+				0},
+		{HDCP_2_2_LC_INIT, 0, 0},
+		{HDCP_2_2_LC_SEND_LPRIME, HDCP_2_2_HDMI_LPRIME_TIMEOUT_MS, 0},
+		{HDCP_2_2_SKE_SEND_EKS, 0, 0},
+		{HDCP_2_2_REP_SEND_RECVID_LIST,
+				HDCP_2_2_RECVID_LIST_TIMEOUT_MS, 0},
+		{HDCP_2_2_REP_SEND_ACK, 0, 0},
+		{HDCP_2_2_REP_STREAM_MANAGE, 0, 0},
+		{HDCP_2_2_REP_STREAM_READY, HDCP_2_2_STREAM_READY_TIMEOUT_MS,
+				0},
+	};
+
+static
+int intel_hdmi_hdcp2_read_rx_status(struct intel_digital_port *intel_dig_port,
+				    u8 *rx_status)
+{
+	return intel_hdmi_hdcp_read(intel_dig_port,
+				    HDCP_2_2_HDMI_REG_RXSTATUS_OFFSET,
+				    rx_status,
+				    HDCP_2_2_HDMI_RXSTATUS_LEN);
+}
+
+static int get_hdcp2_msg_timeout(u8 msg_id, bool is_paired)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(hdcp2_msg_data); i++)
+		if (hdcp2_msg_data[i].msg_id == msg_id &&
+		    (msg_id != HDCP_2_2_AKE_SEND_HPRIME || is_paired))
+			return hdcp2_msg_data[i].timeout;
+		else if (hdcp2_msg_data[i].msg_id == msg_id)
+			return hdcp2_msg_data[i].timeout2;
+
+	return -EINVAL;
+}
+
+static inline
+int hdcp2_detect_msg_availability(struct intel_digital_port *intel_digital_port,
+				  u8 msg_id, bool *msg_ready,
+				  ssize_t *msg_sz)
+{
+	u8 rx_status[HDCP_2_2_HDMI_RXSTATUS_LEN];
+	int ret;
+
+	ret = intel_hdmi_hdcp2_read_rx_status(intel_digital_port, rx_status);
+	if (ret < 0) {
+		DRM_DEBUG_KMS("rx_status read failed. Err %d\n", ret);
+		return ret;
+	}
+
+	*msg_sz = ((HDCP_2_2_HDMI_RXSTATUS_MSG_SZ_HI(rx_status[1]) << 8) |
+		  rx_status[0]);
+
+	if (msg_id == HDCP_2_2_REP_SEND_RECVID_LIST)
+		*msg_ready = (HDCP_2_2_HDMI_RXSTATUS_READY(rx_status[1]) &&
+			     *msg_sz);
+	else
+		*msg_ready = *msg_sz;
+
+	return 0;
+}
+
+static ssize_t
+intel_hdmi_hdcp2_wait_for_msg(struct intel_digital_port *intel_dig_port,
+			      u8 msg_id, bool paired)
+{
+	bool msg_ready = false;
+	int timeout, ret;
+	ssize_t msg_sz = 0;
+
+	timeout = get_hdcp2_msg_timeout(msg_id, paired);
+	if (timeout < 0)
+		return timeout;
+
+	ret = __wait_for(ret = hdcp2_detect_msg_availability(intel_dig_port,
+							     msg_id, &msg_ready,
+							     &msg_sz),
+			 !ret && msg_ready && msg_sz, timeout * 1000,
+			 1000, 5 * 1000);
+	if (ret)
+		DRM_DEBUG_KMS("msg_id: %d, ret: %d, timeout: %d\n",
+			      msg_id, ret, timeout);
+
+	return ret ? ret : msg_sz;
+}
+
+static
+int intel_hdmi_hdcp2_write_msg(struct intel_digital_port *intel_dig_port,
+			       void *buf, size_t size)
+{
+	unsigned int offset;
+
+	offset = HDCP_2_2_HDMI_REG_WR_MSG_OFFSET;
+	return intel_hdmi_hdcp_write(intel_dig_port, offset, buf, size);
+}
+
+static
+int intel_hdmi_hdcp2_read_msg(struct intel_digital_port *intel_dig_port,
+			      u8 msg_id, void *buf, size_t size)
+{
+	struct intel_hdmi *hdmi = &intel_dig_port->hdmi;
+	struct intel_hdcp *hdcp = &hdmi->attached_connector->hdcp;
+	unsigned int offset;
+	ssize_t ret;
+
+	ret = intel_hdmi_hdcp2_wait_for_msg(intel_dig_port, msg_id,
+					    hdcp->is_paired);
+	if (ret < 0)
+		return ret;
+
+	/*
+	 * Available msg size should be equal to or lesser than the
+	 * available buffer.
+	 */
+	if (ret > size) {
+		DRM_DEBUG_KMS("msg_sz(%zd) is more than exp size(%zu)\n",
+			      ret, size);
+		return -1;
+	}
+
+	offset = HDCP_2_2_HDMI_REG_RD_MSG_OFFSET;
+	ret = intel_hdmi_hdcp_read(intel_dig_port, offset, buf, ret);
+	if (ret)
+		DRM_DEBUG_KMS("Failed to read msg_id: %d(%zd)\n", msg_id, ret);
+
+	return ret;
+}
+
+static
+int intel_hdmi_hdcp2_check_link(struct intel_digital_port *intel_dig_port)
+{
+	u8 rx_status[HDCP_2_2_HDMI_RXSTATUS_LEN];
+	int ret;
+
+	ret = intel_hdmi_hdcp2_read_rx_status(intel_dig_port, rx_status);
+	if (ret)
+		return ret;
+
+	/*
+	 * Re-auth request and Link Integrity Failures are represented by
+	 * same bit. i.e reauth_req.
+	 */
+	if (HDCP_2_2_HDMI_RXSTATUS_REAUTH_REQ(rx_status[1]))
+		ret = HDCP_REAUTH_REQUEST;
+	else if (HDCP_2_2_HDMI_RXSTATUS_READY(rx_status[1]))
+		ret = HDCP_TOPOLOGY_CHANGE;
+
+	return ret;
+}
+
+static
+int intel_hdmi_hdcp2_capable(struct intel_digital_port *intel_dig_port,
+			     bool *capable)
+{
+	u8 hdcp2_version;
+	int ret;
+
+	*capable = false;
+	ret = intel_hdmi_hdcp_read(intel_dig_port, HDCP_2_2_HDMI_REG_VER_OFFSET,
+				   &hdcp2_version, sizeof(hdcp2_version));
+	if (!ret && hdcp2_version & HDCP_2_2_HDMI_SUPPORT_MASK)
+		*capable = true;
+
+	return ret;
+}
+
+static inline
+enum hdcp_wired_protocol intel_hdmi_hdcp2_protocol(void)
+{
+	return HDCP_PROTOCOL_HDMI;
+}
+
 static const struct intel_hdcp_shim intel_hdmi_hdcp_shim = {
 	.write_an_aksv = intel_hdmi_hdcp_write_an_aksv,
 	.read_bksv = intel_hdmi_hdcp_read_bksv,
@@ -1130,6 +1664,11 @@ static const struct intel_hdcp_shim intel_hdmi_hdcp_shim = {
 	.read_v_prime_part = intel_hdmi_hdcp_read_v_prime_part,
 	.toggle_signalling = intel_hdmi_hdcp_toggle_signalling,
 	.check_link = intel_hdmi_hdcp_check_link,
+	.write_2_2_msg = intel_hdmi_hdcp2_write_msg,
+	.read_2_2_msg = intel_hdmi_hdcp2_read_msg,
+	.check_2_2_link	= intel_hdmi_hdcp2_check_link,
+	.hdcp_2_2_capable = intel_hdmi_hdcp2_capable,
+	.protocol = HDCP_PROTOCOL_HDMI,
 };
 
 static void intel_hdmi_prepare(struct intel_encoder *encoder,
@@ -1195,7 +1734,6 @@ static void intel_hdmi_get_config(struct intel_encoder *encoder,
 				  struct intel_crtc_state *pipe_config)
 {
 	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&encoder->base);
-	struct intel_digital_port *intel_dig_port = hdmi_to_dig_port(intel_hdmi);
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	u32 tmp, flags = 0;
@@ -1218,7 +1756,10 @@ static void intel_hdmi_get_config(struct intel_encoder *encoder,
 	if (tmp & HDMI_MODE_SELECT_HDMI)
 		pipe_config->has_hdmi_sink = true;
 
-	if (intel_dig_port->infoframe_enabled(encoder, pipe_config))
+	pipe_config->infoframes.enable |=
+		intel_hdmi_infoframes_enabled(encoder, pipe_config);
+
+	if (pipe_config->infoframes.enable)
 		pipe_config->has_infoframe = true;
 
 	if (tmp & SDVO_AUDIO_ENABLE)
@@ -1241,6 +1782,18 @@ static void intel_hdmi_get_config(struct intel_encoder *encoder,
 	pipe_config->base.adjusted_mode.crtc_clock = dotclock;
 
 	pipe_config->lane_count = 4;
+
+	intel_hdmi_read_gcp_infoframe(encoder, pipe_config);
+
+	intel_read_infoframe(encoder, pipe_config,
+			     HDMI_INFOFRAME_TYPE_AVI,
+			     &pipe_config->infoframes.avi);
+	intel_read_infoframe(encoder, pipe_config,
+			     HDMI_INFOFRAME_TYPE_SPD,
+			     &pipe_config->infoframes.spd);
+	intel_read_infoframe(encoder, pipe_config,
+			     HDMI_INFOFRAME_TYPE_VENDOR,
+			     &pipe_config->infoframes.hdmi);
 }
 
 static void intel_enable_hdmi_audio(struct intel_encoder *encoder,
@@ -1654,7 +2207,7 @@ static bool hdmi_deep_color_possible(const struct intel_crtc_state *crtc_state,
 
 	/* Display Wa_1405510057:icl */
 	if (crtc_state->output_format == INTEL_OUTPUT_FORMAT_YCBCR420 &&
-	    bpc == 10 && IS_ICELAKE(dev_priv) &&
+	    bpc == 10 && INTEL_GEN(dev_priv) >= 11 &&
 	    (adjusted_mode->crtc_hblank_end -
 	     adjusted_mode->crtc_hblank_start) % 8 == 2)
 		return false;
@@ -1812,6 +2365,23 @@ int intel_hdmi_compute_config(struct intel_encoder *encoder,
 		}
 	}
 
+	intel_hdmi_compute_gcp_infoframe(encoder, pipe_config, conn_state);
+
+	if (!intel_hdmi_compute_avi_infoframe(encoder, pipe_config, conn_state)) {
+		DRM_DEBUG_KMS("bad AVI infoframe\n");
+		return -EINVAL;
+	}
+
+	if (!intel_hdmi_compute_spd_infoframe(encoder, pipe_config, conn_state)) {
+		DRM_DEBUG_KMS("bad SPD infoframe\n");
+		return -EINVAL;
+	}
+
+	if (!intel_hdmi_compute_hdmi_infoframe(encoder, pipe_config, conn_state)) {
+		DRM_DEBUG_KMS("bad HDMI infoframe\n");
+		return -EINVAL;
+	}
+
 	return 0;
 }
 
@@ -1931,7 +2501,7 @@ intel_hdmi_detect(struct drm_connector *connector, bool force)
 
 	wakeref = intel_display_power_get(dev_priv, POWER_DOMAIN_GMBUS);
 
-	if (IS_ICELAKE(dev_priv) &&
+	if (INTEL_GEN(dev_priv) >= 11 &&
 	    !intel_digital_port_connected(encoder))
 		goto out;
 
@@ -2133,10 +2703,21 @@ static void
 intel_hdmi_add_properties(struct intel_hdmi *intel_hdmi, struct drm_connector *connector)
 {
 	struct drm_i915_private *dev_priv = to_i915(connector->dev);
+	struct intel_digital_port *intel_dig_port =
+				hdmi_to_dig_port(intel_hdmi);
 
 	intel_attach_force_audio_property(connector);
 	intel_attach_broadcast_rgb_property(connector);
 	intel_attach_aspect_ratio_property(connector);
+
+	/*
+	 * Attach Colorspace property for Non LSPCON based device
+	 * ToDo: This needs to be extended for LSPCON implementation
+	 * as well. Will be implemented separately.
+	 */
+	if (!intel_dig_port->lspcon.active)
+		intel_attach_colorspace_property(connector);
+
 	drm_connector_attach_content_type_property(connector);
 	connector->state->picture_aspect_ratio = HDMI_PICTURE_ASPECT_NONE;
 
@@ -2321,14 +2902,14 @@ static u8 intel_hdmi_ddc_pin(struct drm_i915_private *dev_priv,
 		return info->alternate_ddc_pin;
 	}
 
-	if (IS_CHERRYVIEW(dev_priv))
-		ddc_pin = chv_port_to_ddc_pin(dev_priv, port);
-	else if (IS_GEN9_LP(dev_priv))
-		ddc_pin = bxt_port_to_ddc_pin(dev_priv, port);
+	if (HAS_PCH_ICP(dev_priv))
+		ddc_pin = icl_port_to_ddc_pin(dev_priv, port);
 	else if (HAS_PCH_CNP(dev_priv))
 		ddc_pin = cnp_port_to_ddc_pin(dev_priv, port);
-	else if (HAS_PCH_ICP(dev_priv))
-		ddc_pin = icl_port_to_ddc_pin(dev_priv, port);
+	else if (IS_GEN9_LP(dev_priv))
+		ddc_pin = bxt_port_to_ddc_pin(dev_priv, port);
+	else if (IS_CHERRYVIEW(dev_priv))
+		ddc_pin = chv_port_to_ddc_pin(dev_priv, port);
 	else
 		ddc_pin = g4x_port_to_ddc_pin(dev_priv, port);
 
@@ -2345,33 +2926,36 @@ void intel_infoframe_init(struct intel_digital_port *intel_dig_port)
 
 	if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {
 		intel_dig_port->write_infoframe = vlv_write_infoframe;
+		intel_dig_port->read_infoframe = vlv_read_infoframe;
 		intel_dig_port->set_infoframes = vlv_set_infoframes;
-		intel_dig_port->infoframe_enabled = vlv_infoframe_enabled;
+		intel_dig_port->infoframes_enabled = vlv_infoframes_enabled;
 	} else if (IS_G4X(dev_priv)) {
 		intel_dig_port->write_infoframe = g4x_write_infoframe;
+		intel_dig_port->read_infoframe = g4x_read_infoframe;
 		intel_dig_port->set_infoframes = g4x_set_infoframes;
-		intel_dig_port->infoframe_enabled = g4x_infoframe_enabled;
+		intel_dig_port->infoframes_enabled = g4x_infoframes_enabled;
 	} else if (HAS_DDI(dev_priv)) {
 		if (intel_dig_port->lspcon.active) {
-			intel_dig_port->write_infoframe =
-					lspcon_write_infoframe;
+			intel_dig_port->write_infoframe = lspcon_write_infoframe;
+			intel_dig_port->read_infoframe = lspcon_read_infoframe;
 			intel_dig_port->set_infoframes = lspcon_set_infoframes;
-			intel_dig_port->infoframe_enabled =
-						lspcon_infoframe_enabled;
+			intel_dig_port->infoframes_enabled = lspcon_infoframes_enabled;
 		} else {
-			intel_dig_port->set_infoframes = hsw_set_infoframes;
-			intel_dig_port->infoframe_enabled =
-						hsw_infoframe_enabled;
 			intel_dig_port->write_infoframe = hsw_write_infoframe;
+			intel_dig_port->read_infoframe = hsw_read_infoframe;
+			intel_dig_port->set_infoframes = hsw_set_infoframes;
+			intel_dig_port->infoframes_enabled = hsw_infoframes_enabled;
 		}
 	} else if (HAS_PCH_IBX(dev_priv)) {
 		intel_dig_port->write_infoframe = ibx_write_infoframe;
+		intel_dig_port->read_infoframe = ibx_read_infoframe;
 		intel_dig_port->set_infoframes = ibx_set_infoframes;
-		intel_dig_port->infoframe_enabled = ibx_infoframe_enabled;
+		intel_dig_port->infoframes_enabled = ibx_infoframes_enabled;
 	} else {
 		intel_dig_port->write_infoframe = cpt_write_infoframe;
+		intel_dig_port->read_infoframe = cpt_read_infoframe;
 		intel_dig_port->set_infoframes = cpt_set_infoframes;
-		intel_dig_port->infoframe_enabled = cpt_infoframe_enabled;
+		intel_dig_port->infoframes_enabled = cpt_infoframes_enabled;
 	}
 }
 
@@ -2417,6 +3001,9 @@ void intel_hdmi_init_connector(struct intel_digital_port *intel_dig_port,
 
 	intel_hdmi_add_properties(intel_hdmi, connector);
 
+	intel_connector_attach_encoder(intel_connector, intel_encoder);
+	intel_hdmi->attached_connector = intel_connector;
+
 	if (is_hdcp_supported(dev_priv, port)) {
 		int ret = intel_hdcp_init(intel_connector,
 					  &intel_hdmi_hdcp_shim);
@@ -2424,9 +3011,6 @@ void intel_hdmi_init_connector(struct intel_digital_port *intel_dig_port,
 			DRM_DEBUG_KMS("HDCP init failed, skipping.\n");
 	}
 
-	intel_connector_attach_encoder(intel_connector, intel_encoder);
-	intel_hdmi->attached_connector = intel_connector;
-
 	/* For G4X desktop chip, PEG_BAND_GAP_DATA 3:0 must first be written
 	 * 0xd.  Failure to do so will result in spurious interrupts being
 	 * generated on the port when a cable is not attached.
diff --git a/drivers/gpu/drm/i915/intel_hdmi.h b/drivers/gpu/drm/i915/intel_hdmi.h
new file mode 100644
index 000000000000..106c2e0bc3c9
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_hdmi.h
@@ -0,0 +1,51 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_HDMI_H__
+#define __INTEL_HDMI_H__
+
+#include <linux/hdmi.h>
+#include <linux/types.h>
+
+#include <drm/i915_drm.h>
+
+#include "i915_reg.h"
+
+struct drm_connector;
+struct drm_encoder;
+struct drm_i915_private;
+struct intel_connector;
+struct intel_digital_port;
+struct intel_encoder;
+struct intel_crtc_state;
+struct intel_hdmi;
+struct drm_connector_state;
+union hdmi_infoframe;
+
+void intel_hdmi_init(struct drm_i915_private *dev_priv, i915_reg_t hdmi_reg,
+		     enum port port);
+void intel_hdmi_init_connector(struct intel_digital_port *intel_dig_port,
+			       struct intel_connector *intel_connector);
+struct intel_hdmi *enc_to_intel_hdmi(struct drm_encoder *encoder);
+int intel_hdmi_compute_config(struct intel_encoder *encoder,
+			      struct intel_crtc_state *pipe_config,
+			      struct drm_connector_state *conn_state);
+bool intel_hdmi_handle_sink_scrambling(struct intel_encoder *encoder,
+				       struct drm_connector *connector,
+				       bool high_tmds_clock_ratio,
+				       bool scrambling);
+void intel_dp_dual_mode_set_tmds_output(struct intel_hdmi *hdmi, bool enable);
+void intel_infoframe_init(struct intel_digital_port *intel_dig_port);
+u32 intel_hdmi_infoframes_enabled(struct intel_encoder *encoder,
+				  const struct intel_crtc_state *crtc_state);
+u32 intel_hdmi_infoframe_enable(unsigned int type);
+void intel_hdmi_read_gcp_infoframe(struct intel_encoder *encoder,
+				   struct intel_crtc_state *crtc_state);
+void intel_read_infoframe(struct intel_encoder *encoder,
+			  const struct intel_crtc_state *crtc_state,
+			  enum hdmi_infoframe_type type,
+			  union hdmi_infoframe *frame);
+
+#endif /* __INTEL_HDMI_H__ */
diff --git a/drivers/gpu/drm/i915/intel_huc.c b/drivers/gpu/drm/i915/intel_huc.c
index 9bd1c9002c2a..94c04f16a2ad 100644
--- a/drivers/gpu/drm/i915/intel_huc.c
+++ b/drivers/gpu/drm/i915/intel_huc.c
@@ -79,7 +79,7 @@ int intel_huc_auth(struct intel_huc *huc)
 	}
 
 	/* Check authentication status, it should be done by now */
-	ret = __intel_wait_for_register(i915,
+	ret = __intel_wait_for_register(&i915->uncore,
 					HUC_STATUS2,
 					HUC_FW_VERIFIED,
 					HUC_FW_VERIFIED,
diff --git a/drivers/gpu/drm/i915/intel_huc_fw.c b/drivers/gpu/drm/i915/intel_huc_fw.c
index 7d7bfc7f7ca7..68d47c105939 100644
--- a/drivers/gpu/drm/i915/intel_huc_fw.c
+++ b/drivers/gpu/drm/i915/intel_huc_fw.c
@@ -106,41 +106,46 @@ static int huc_fw_xfer(struct intel_uc_fw *huc_fw, struct i915_vma *vma)
 {
 	struct intel_huc *huc = container_of(huc_fw, struct intel_huc, fw);
 	struct drm_i915_private *dev_priv = huc_to_i915(huc);
+	struct intel_uncore *uncore = &dev_priv->uncore;
 	unsigned long offset = 0;
 	u32 size;
 	int ret;
 
 	GEM_BUG_ON(huc_fw->type != INTEL_UC_FW_TYPE_HUC);
 
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(uncore, FORCEWAKE_ALL);
 
 	/* Set the source address for the uCode */
 	offset = intel_guc_ggtt_offset(&dev_priv->guc, vma) +
 		 huc_fw->header_offset;
-	I915_WRITE(DMA_ADDR_0_LOW, lower_32_bits(offset));
-	I915_WRITE(DMA_ADDR_0_HIGH, upper_32_bits(offset) & 0xFFFF);
+	intel_uncore_write(uncore, DMA_ADDR_0_LOW,
+			   lower_32_bits(offset));
+	intel_uncore_write(uncore, DMA_ADDR_0_HIGH,
+			   upper_32_bits(offset) & 0xFFFF);
 
-	/* Hardware doesn't look at destination address for HuC. Set it to 0,
+	/*
+	 * Hardware doesn't look at destination address for HuC. Set it to 0,
 	 * but still program the correct address space.
 	 */
-	I915_WRITE(DMA_ADDR_1_LOW, 0);
-	I915_WRITE(DMA_ADDR_1_HIGH, DMA_ADDRESS_SPACE_WOPCM);
+	intel_uncore_write(uncore, DMA_ADDR_1_LOW, 0);
+	intel_uncore_write(uncore, DMA_ADDR_1_HIGH, DMA_ADDRESS_SPACE_WOPCM);
 
 	size = huc_fw->header_size + huc_fw->ucode_size;
-	I915_WRITE(DMA_COPY_SIZE, size);
+	intel_uncore_write(uncore, DMA_COPY_SIZE, size);
 
 	/* Start the DMA */
-	I915_WRITE(DMA_CTRL, _MASKED_BIT_ENABLE(HUC_UKERNEL | START_DMA));
+	intel_uncore_write(uncore, DMA_CTRL,
+			   _MASKED_BIT_ENABLE(HUC_UKERNEL | START_DMA));
 
 	/* Wait for DMA to finish */
-	ret = intel_wait_for_register_fw(dev_priv, DMA_CTRL, START_DMA, 0, 100);
+	ret = intel_wait_for_register_fw(uncore, DMA_CTRL, START_DMA, 0, 100);
 
 	DRM_DEBUG_DRIVER("HuC DMA transfer wait over with ret %d\n", ret);
 
 	/* Disable the bits once DMA is over */
-	I915_WRITE(DMA_CTRL, _MASKED_BIT_DISABLE(HUC_UKERNEL));
+	intel_uncore_write(uncore, DMA_CTRL, _MASKED_BIT_DISABLE(HUC_UKERNEL));
 
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(uncore, FORCEWAKE_ALL);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/i915/intel_i2c.c b/drivers/gpu/drm/i915/intel_i2c.c
index 5a733e711355..422685d120e9 100644
--- a/drivers/gpu/drm/i915/intel_i2c.c
+++ b/drivers/gpu/drm/i915/intel_i2c.c
@@ -348,7 +348,7 @@ gmbus_wait_idle(struct drm_i915_private *dev_priv)
 	add_wait_queue(&dev_priv->gmbus_wait_queue, &wait);
 	I915_WRITE_FW(GMBUS4, irq_enable);
 
-	ret = intel_wait_for_register_fw(dev_priv,
+	ret = intel_wait_for_register_fw(&dev_priv->uncore,
 					 GMBUS2, GMBUS_ACTIVE, 0,
 					 10);
 
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 5e98fd79bd9d..4e0a351bfbca 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -164,20 +164,15 @@
 #define WA_TAIL_DWORDS 2
 #define WA_TAIL_BYTES (sizeof(u32) * WA_TAIL_DWORDS)
 
-static int execlists_context_deferred_alloc(struct i915_gem_context *ctx,
-					    struct intel_engine_cs *engine,
-					    struct intel_context *ce);
+#define ACTIVE_PRIORITY (I915_PRIORITY_NEWCLIENT | I915_PRIORITY_NOSEMAPHORE)
+
+static int execlists_context_deferred_alloc(struct intel_context *ce,
+					    struct intel_engine_cs *engine);
 static void execlists_init_reg_state(u32 *reg_state,
-				     struct i915_gem_context *ctx,
+				     struct intel_context *ce,
 				     struct intel_engine_cs *engine,
 				     struct intel_ring *ring);
 
-static inline u32 intel_hws_seqno_address(struct intel_engine_cs *engine)
-{
-	return (i915_ggtt_offset(engine->status_page.vma) +
-		I915_GEM_HWS_INDEX_ADDR);
-}
-
 static inline struct i915_priolist *to_priolist(struct rb_node *rb)
 {
 	return rb_entry(rb, struct i915_priolist, node);
@@ -188,6 +183,34 @@ static inline int rq_prio(const struct i915_request *rq)
 	return rq->sched.attr.priority;
 }
 
+static int effective_prio(const struct i915_request *rq)
+{
+	int prio = rq_prio(rq);
+
+	/*
+	 * On unwinding the active request, we give it a priority bump
+	 * equivalent to a freshly submitted request. This protects it from
+	 * being gazumped again, but it would be preferable if we didn't
+	 * let it be gazumped in the first place!
+	 *
+	 * See __unwind_incomplete_requests()
+	 */
+	if (~prio & ACTIVE_PRIORITY && __i915_request_has_started(rq)) {
+		/*
+		 * After preemption, we insert the active request at the
+		 * end of the new priority level. This means that we will be
+		 * _lower_ priority than the preemptee all things equal (and
+		 * so the preemption is valid), so adjust our comparison
+		 * accordingly.
+		 */
+		prio |= ACTIVE_PRIORITY;
+		prio--;
+	}
+
+	/* Restrict mere WAIT boosts from triggering preemption */
+	return prio | __NO_PREEMPTION;
+}
+
 static int queue_prio(const struct intel_engine_execlists *execlists)
 {
 	struct i915_priolist *p;
@@ -208,9 +231,9 @@ static int queue_prio(const struct intel_engine_execlists *execlists)
 static inline bool need_preempt(const struct intel_engine_cs *engine,
 				const struct i915_request *rq)
 {
-	const int last_prio = rq_prio(rq);
+	int last_prio;
 
-	if (!intel_engine_has_preemption(engine))
+	if (!engine->preempt_context)
 		return false;
 
 	if (i915_request_completed(rq))
@@ -228,6 +251,7 @@ static inline bool need_preempt(const struct intel_engine_cs *engine,
 	 * preempt. If that hint is stale or we may be trying to preempt
 	 * ourselves, ignore the request.
 	 */
+	last_prio = effective_prio(rq);
 	if (!__execlists_need_preempt(engine->execlists.queue_priority_hint,
 				      last_prio))
 		return false;
@@ -254,12 +278,11 @@ static inline bool need_preempt(const struct intel_engine_cs *engine,
 }
 
 __maybe_unused static inline bool
-assert_priority_queue(const struct intel_engine_execlists *execlists,
-		      const struct i915_request *prev,
+assert_priority_queue(const struct i915_request *prev,
 		      const struct i915_request *next)
 {
-	if (!prev)
-		return true;
+	const struct intel_engine_execlists *execlists =
+		&prev->engine->execlists;
 
 	/*
 	 * Without preemption, the prev may refer to the still active element
@@ -300,11 +323,10 @@ assert_priority_queue(const struct intel_engine_execlists *execlists,
  * engine info, SW context ID and SW counter need to form a unique number
  * (Context ID) per lrc.
  */
-static void
-intel_lr_context_descriptor_update(struct i915_gem_context *ctx,
-				   struct intel_engine_cs *engine,
-				   struct intel_context *ce)
+static u64
+lrc_descriptor(struct intel_context *ce, struct intel_engine_cs *engine)
 {
+	struct i915_gem_context *ctx = ce->gem_context;
 	u64 desc;
 
 	BUILD_BUG_ON(MAX_CONTEXT_HW_ID > (BIT(GEN8_CTX_ID_WIDTH)));
@@ -322,7 +344,7 @@ intel_lr_context_descriptor_update(struct i915_gem_context *ctx,
 	 * Consider updating oa_get_render_ctx_id in i915_perf.c when changing
 	 * anything below.
 	 */
-	if (INTEL_GEN(ctx->i915) >= 11) {
+	if (INTEL_GEN(engine->i915) >= 11) {
 		GEM_BUG_ON(ctx->hw_id >= BIT(GEN11_SW_CTX_ID_WIDTH));
 		desc |= (u64)ctx->hw_id << GEN11_SW_CTX_ID_SHIFT;
 								/* bits 37-47 */
@@ -339,7 +361,7 @@ intel_lr_context_descriptor_update(struct i915_gem_context *ctx,
 		desc |= (u64)ctx->hw_id << GEN8_CTX_ID_SHIFT;	/* bits 32-52 */
 	}
 
-	ce->lrc_desc = desc;
+	return desc;
 }
 
 static void unwind_wa_tail(struct i915_request *rq)
@@ -353,7 +375,7 @@ __unwind_incomplete_requests(struct intel_engine_cs *engine)
 {
 	struct i915_request *rq, *rn, *active = NULL;
 	struct list_head *uninitialized_var(pl);
-	int prio = I915_PRIORITY_INVALID | I915_PRIORITY_NEWCLIENT;
+	int prio = I915_PRIORITY_INVALID | ACTIVE_PRIORITY;
 
 	lockdep_assert_held(&engine->timeline.lock);
 
@@ -384,9 +406,21 @@ __unwind_incomplete_requests(struct intel_engine_cs *engine)
 	 * The active request is now effectively the start of a new client
 	 * stream, so give it the equivalent small priority bump to prevent
 	 * it being gazumped a second time by another peer.
+	 *
+	 * Note we have to be careful not to apply a priority boost to a request
+	 * still spinning on its semaphores. If the request hasn't started, that
+	 * means it is still waiting for its dependencies to be signaled, and
+	 * if we apply a priority boost to this request, we will boost it past
+	 * its signalers and so break PI.
+	 *
+	 * One consequence of this preemption boost is that we may jump
+	 * over lesser priorities (such as I915_PRIORITY_WAIT), effectively
+	 * making those priorities non-preemptible. They will be moved forward
+	 * in the priority queue, but they will not gain immediate access to
+	 * the GPU.
 	 */
-	if (!(prio & I915_PRIORITY_NEWCLIENT)) {
-		prio |= I915_PRIORITY_NEWCLIENT;
+	if (~prio & ACTIVE_PRIORITY && __i915_request_has_started(active)) {
+		prio |= ACTIVE_PRIORITY;
 		active->sched.attr.priority = prio;
 		list_move_tail(&active->sched.link,
 			       i915_sched_lookup_priolist(engine, prio));
@@ -395,13 +429,13 @@ __unwind_incomplete_requests(struct intel_engine_cs *engine)
 	return active;
 }
 
-void
+struct i915_request *
 execlists_unwind_incomplete_requests(struct intel_engine_execlists *execlists)
 {
 	struct intel_engine_cs *engine =
 		container_of(execlists, typeof(*engine), execlists);
 
-	__unwind_incomplete_requests(engine);
+	return __unwind_incomplete_requests(engine);
 }
 
 static inline void
@@ -523,13 +557,11 @@ static void execlists_submit_ports(struct intel_engine_cs *engine)
 			desc = execlists_update_context(rq);
 			GEM_DEBUG_EXEC(port[n].context_id = upper_32_bits(desc));
 
-			GEM_TRACE("%s in[%d]:  ctx=%d.%d, global=%d (fence %llx:%lld) (current %d:%d), prio=%d\n",
+			GEM_TRACE("%s in[%d]:  ctx=%d.%d, fence %llx:%lld (current %d), prio=%d\n",
 				  engine->name, n,
 				  port[n].context_id, count,
-				  rq->global_seqno,
 				  rq->fence.context, rq->fence.seqno,
 				  hwsp_seqno(rq),
-				  intel_engine_get_seqno(engine),
 				  rq_prio(rq));
 		} else {
 			GEM_BUG_ON(!n);
@@ -564,6 +596,17 @@ static bool can_merge_ctx(const struct intel_context *prev,
 	return true;
 }
 
+static bool can_merge_rq(const struct i915_request *prev,
+			 const struct i915_request *next)
+{
+	GEM_BUG_ON(!assert_priority_queue(prev, next));
+
+	if (!can_merge_ctx(prev->hw_context, next->hw_context))
+		return false;
+
+	return true;
+}
+
 static void port_assign(struct execlist_port *port, struct i915_request *rq)
 {
 	GEM_BUG_ON(rq == port_request(port));
@@ -577,8 +620,7 @@ static void port_assign(struct execlist_port *port, struct i915_request *rq)
 static void inject_preempt_context(struct intel_engine_cs *engine)
 {
 	struct intel_engine_execlists *execlists = &engine->execlists;
-	struct intel_context *ce =
-		to_intel_context(engine->i915->preempt_context, engine);
+	struct intel_context *ce = engine->preempt_context;
 	unsigned int n;
 
 	GEM_BUG_ON(execlists->preempt_complete_status !=
@@ -716,8 +758,6 @@ static void execlists_dequeue(struct intel_engine_cs *engine)
 		int i;
 
 		priolist_for_each_request_consume(rq, rn, p, i) {
-			GEM_BUG_ON(!assert_priority_queue(execlists, last, rq));
-
 			/*
 			 * Can we combine this request with the current port?
 			 * It has to be the same context/ringbuffer and not
@@ -729,8 +769,7 @@ static void execlists_dequeue(struct intel_engine_cs *engine)
 			 * second request, and so we never need to tell the
 			 * hardware about the first.
 			 */
-			if (last &&
-			    !can_merge_ctx(rq->hw_context, last->hw_context)) {
+			if (last && !can_merge_rq(last, rq)) {
 				/*
 				 * If we are on the second port and cannot
 				 * combine this request with the last, then we
@@ -740,6 +779,14 @@ static void execlists_dequeue(struct intel_engine_cs *engine)
 					goto done;
 
 				/*
+				 * We must not populate both ELSP[] with the
+				 * same LRCA, i.e. we must submit 2 different
+				 * contexts if we submit 2 ELSP.
+				 */
+				if (last->hw_context == rq->hw_context)
+					goto done;
+
+				/*
 				 * If GVT overrides us we only ever submit
 				 * port[0], leaving port[1] empty. Note that we
 				 * also have to be careful that we don't queue
@@ -750,7 +797,6 @@ static void execlists_dequeue(struct intel_engine_cs *engine)
 				    ctx_single_port_submission(rq->hw_context))
 					goto done;
 
-				GEM_BUG_ON(last->hw_context == rq->hw_context);
 
 				if (submit)
 					port_assign(port, last);
@@ -769,8 +815,7 @@ static void execlists_dequeue(struct intel_engine_cs *engine)
 		}
 
 		rb_erase_cached(&p->node, &execlists->queue);
-		if (p->priority != I915_PRIORITY_NORMAL)
-			kmem_cache_free(engine->i915->priorities, p);
+		i915_priolist_free(p);
 	}
 
 done:
@@ -790,8 +835,7 @@ done:
 	 * request triggering preemption on the next dequeue (or subsequent
 	 * interrupt for secondary ports).
 	 */
-	execlists->queue_priority_hint =
-		port != execlists->port ? rq_prio(last) : INT_MIN;
+	execlists->queue_priority_hint = queue_prio(execlists);
 
 	if (submit) {
 		port_assign(port, last);
@@ -821,13 +865,11 @@ execlists_cancel_port_requests(struct intel_engine_execlists * const execlists)
 	while (num_ports-- && port_isset(port)) {
 		struct i915_request *rq = port_request(port);
 
-		GEM_TRACE("%s:port%u global=%d (fence %llx:%lld), (current %d:%d)\n",
+		GEM_TRACE("%s:port%u fence %llx:%lld, (current %d)\n",
 			  rq->engine->name,
 			  (unsigned int)(port - execlists->port),
-			  rq->global_seqno,
 			  rq->fence.context, rq->fence.seqno,
-			  hwsp_seqno(rq),
-			  intel_engine_get_seqno(rq->engine));
+			  hwsp_seqno(rq));
 
 		GEM_BUG_ON(!execlists->active);
 		execlists_context_schedule_out(rq,
@@ -851,104 +893,6 @@ invalidate_csb_entries(const u32 *first, const u32 *last)
 	clflush((void *)last);
 }
 
-static void reset_csb_pointers(struct intel_engine_execlists *execlists)
-{
-	const unsigned int reset_value = GEN8_CSB_ENTRIES - 1;
-
-	/*
-	 * After a reset, the HW starts writing into CSB entry [0]. We
-	 * therefore have to set our HEAD pointer back one entry so that
-	 * the *first* entry we check is entry 0. To complicate this further,
-	 * as we don't wait for the first interrupt after reset, we have to
-	 * fake the HW write to point back to the last entry so that our
-	 * inline comparison of our cached head position against the last HW
-	 * write works even before the first interrupt.
-	 */
-	execlists->csb_head = reset_value;
-	WRITE_ONCE(*execlists->csb_write, reset_value);
-
-	invalidate_csb_entries(&execlists->csb_status[0],
-			       &execlists->csb_status[GEN8_CSB_ENTRIES - 1]);
-}
-
-static void nop_submission_tasklet(unsigned long data)
-{
-	/* The driver is wedged; don't process any more events. */
-}
-
-static void execlists_cancel_requests(struct intel_engine_cs *engine)
-{
-	struct intel_engine_execlists * const execlists = &engine->execlists;
-	struct i915_request *rq, *rn;
-	struct rb_node *rb;
-	unsigned long flags;
-
-	GEM_TRACE("%s current %d\n",
-		  engine->name, intel_engine_get_seqno(engine));
-
-	/*
-	 * Before we call engine->cancel_requests(), we should have exclusive
-	 * access to the submission state. This is arranged for us by the
-	 * caller disabling the interrupt generation, the tasklet and other
-	 * threads that may then access the same state, giving us a free hand
-	 * to reset state. However, we still need to let lockdep be aware that
-	 * we know this state may be accessed in hardirq context, so we
-	 * disable the irq around this manipulation and we want to keep
-	 * the spinlock focused on its duties and not accidentally conflate
-	 * coverage to the submission's irq state. (Similarly, although we
-	 * shouldn't need to disable irq around the manipulation of the
-	 * submission's irq state, we also wish to remind ourselves that
-	 * it is irq state.)
-	 */
-	spin_lock_irqsave(&engine->timeline.lock, flags);
-
-	/* Cancel the requests on the HW and clear the ELSP tracker. */
-	execlists_cancel_port_requests(execlists);
-	execlists_user_end(execlists);
-
-	/* Mark all executing requests as skipped. */
-	list_for_each_entry(rq, &engine->timeline.requests, link) {
-		GEM_BUG_ON(!rq->global_seqno);
-
-		if (!i915_request_signaled(rq))
-			dma_fence_set_error(&rq->fence, -EIO);
-
-		i915_request_mark_complete(rq);
-	}
-
-	/* Flush the queued requests to the timeline list (for retiring). */
-	while ((rb = rb_first_cached(&execlists->queue))) {
-		struct i915_priolist *p = to_priolist(rb);
-		int i;
-
-		priolist_for_each_request_consume(rq, rn, p, i) {
-			list_del_init(&rq->sched.link);
-			__i915_request_submit(rq);
-			dma_fence_set_error(&rq->fence, -EIO);
-			i915_request_mark_complete(rq);
-		}
-
-		rb_erase_cached(&p->node, &execlists->queue);
-		if (p->priority != I915_PRIORITY_NORMAL)
-			kmem_cache_free(engine->i915->priorities, p);
-	}
-
-	intel_write_status_page(engine,
-				I915_GEM_HWS_INDEX,
-				intel_engine_last_submit(engine));
-
-	/* Remaining _unready_ requests will be nop'ed when submitted */
-
-	execlists->queue_priority_hint = INT_MIN;
-	execlists->queue = RB_ROOT_CACHED;
-	GEM_BUG_ON(port_isset(execlists->port));
-
-	GEM_BUG_ON(__tasklet_is_enabled(&execlists->tasklet));
-	execlists->tasklet.func = nop_submission_tasklet;
-
-	spin_unlock_irqrestore(&engine->timeline.lock, flags);
-}
-
 static inline bool
 reset_in_progress(const struct intel_engine_execlists *execlists)
 {
@@ -960,6 +904,7 @@ static void process_csb(struct intel_engine_cs *engine)
 	struct intel_engine_execlists * const execlists = &engine->execlists;
 	struct execlist_port *port = execlists->port;
 	const u32 * const buf = execlists->csb_status;
+	const u8 num_entries = execlists->csb_size;
 	u8 head, tail;
 
 	lockdep_assert_held(&engine->timeline.lock);
@@ -995,7 +940,7 @@ static void process_csb(struct intel_engine_cs *engine)
 		unsigned int status;
 		unsigned int count;
 
-		if (++head == GEN8_CSB_ENTRIES)
+		if (++head == num_entries)
 			head = 0;
 
 		/*
@@ -1052,14 +997,12 @@ static void process_csb(struct intel_engine_cs *engine)
 						EXECLISTS_ACTIVE_USER));
 
 		rq = port_unpack(port, &count);
-		GEM_TRACE("%s out[0]: ctx=%d.%d, global=%d (fence %llx:%lld) (current %d:%d), prio=%d\n",
+		GEM_TRACE("%s out[0]: ctx=%d.%d, fence %llx:%lld (current %d), prio=%d\n",
 			  engine->name,
 			  port->context_id, count,
-			  rq ? rq->global_seqno : 0,
 			  rq ? rq->fence.context : 0,
 			  rq ? rq->fence.seqno : 0,
 			  rq ? hwsp_seqno(rq) : 0,
-			  intel_engine_get_seqno(engine),
 			  rq ? rq_prio(rq) : 0);
 
 		/* Check the context/desc id for this event matches */
@@ -1119,7 +1062,7 @@ static void process_csb(struct intel_engine_cs *engine)
 	 * the wash as hardware, working or not, will need to do the
 	 * invalidation before.
 	 */
-	invalidate_csb_entries(&buf[0], &buf[GEN8_CSB_ENTRIES - 1]);
+	invalidate_csb_entries(&buf[0], &buf[num_entries - 1]);
 }
 
 static void __execlists_submission_tasklet(struct intel_engine_cs *const engine)
@@ -1196,19 +1139,50 @@ static void execlists_submit_request(struct i915_request *request)
 	spin_unlock_irqrestore(&engine->timeline.lock, flags);
 }
 
-static void execlists_context_destroy(struct intel_context *ce)
+static void __execlists_context_fini(struct intel_context *ce)
 {
-	GEM_BUG_ON(ce->pin_count);
-
-	if (!ce->state)
-		return;
-
-	intel_ring_free(ce->ring);
+	intel_ring_put(ce->ring);
 
 	GEM_BUG_ON(i915_gem_object_is_active(ce->state->obj));
 	i915_gem_object_put(ce->state->obj);
 }
 
+static void execlists_context_destroy(struct kref *kref)
+{
+	struct intel_context *ce = container_of(kref, typeof(*ce), ref);
+
+	GEM_BUG_ON(intel_context_is_pinned(ce));
+
+	if (ce->state)
+		__execlists_context_fini(ce);
+
+	intel_context_free(ce);
+}
+
+static int __context_pin(struct i915_vma *vma)
+{
+	unsigned int flags;
+	int err;
+
+	flags = PIN_GLOBAL | PIN_HIGH;
+	flags |= PIN_OFFSET_BIAS | i915_ggtt_pin_bias(vma);
+
+	err = i915_vma_pin(vma, 0, 0, flags);
+	if (err)
+		return err;
+
+	vma->obj->pin_global++;
+	vma->obj->mm.dirty = true;
+
+	return 0;
+}
+
+static void __context_unpin(struct i915_vma *vma)
+{
+	vma->obj->pin_global--;
+	__i915_vma_unpin(vma);
+}
+
 static void execlists_context_unpin(struct intel_context *ce)
 {
 	struct intel_engine_cs *engine;
@@ -1237,41 +1211,19 @@ static void execlists_context_unpin(struct intel_context *ce)
 
 	intel_ring_unpin(ce->ring);
 
-	ce->state->obj->pin_global--;
 	i915_gem_object_unpin_map(ce->state->obj);
-	i915_vma_unpin(ce->state);
-
-	i915_gem_context_put(ce->gem_context);
-}
-
-static int __context_pin(struct i915_gem_context *ctx, struct i915_vma *vma)
-{
-	unsigned int flags;
-	int err;
-
-	/*
-	 * Clear this page out of any CPU caches for coherent swap-in/out.
-	 * We only want to do this on the first bind so that we do not stall
-	 * on an active context (which by nature is already on the GPU).
-	 */
-	if (!(vma->flags & I915_VMA_GLOBAL_BIND)) {
-		err = i915_gem_object_set_to_wc_domain(vma->obj, true);
-		if (err)
-			return err;
-	}
-
-	flags = PIN_GLOBAL | PIN_HIGH;
-	flags |= PIN_OFFSET_BIAS | i915_ggtt_pin_bias(vma);
-
-	return i915_vma_pin(vma, 0, 0, flags);
+	__context_unpin(ce->state);
 }
 
 static void
-__execlists_update_reg_state(struct intel_engine_cs *engine,
-			     struct intel_context *ce)
+__execlists_update_reg_state(struct intel_context *ce,
+			     struct intel_engine_cs *engine)
 {
-	u32 *regs = ce->lrc_reg_state;
 	struct intel_ring *ring = ce->ring;
+	u32 *regs = ce->lrc_reg_state;
+
+	GEM_BUG_ON(!intel_ring_offset_valid(ring, ring->head));
+	GEM_BUG_ON(!intel_ring_offset_valid(ring, ring->tail));
 
 	regs[CTX_RING_BUFFER_START + 1] = i915_ggtt_offset(ring->vma);
 	regs[CTX_RING_HEAD + 1] = ring->head;
@@ -1279,29 +1231,30 @@ __execlists_update_reg_state(struct intel_engine_cs *engine,
 
 	/* RPCS */
 	if (engine->class == RENDER_CLASS)
-		regs[CTX_R_PWR_CLK_STATE + 1] = gen8_make_rpcs(engine->i915,
-							       &ce->sseu);
+		regs[CTX_R_PWR_CLK_STATE + 1] =
+			gen8_make_rpcs(engine->i915, &ce->sseu);
 }
 
-static struct intel_context *
-__execlists_context_pin(struct intel_engine_cs *engine,
-			struct i915_gem_context *ctx,
-			struct intel_context *ce)
+static int
+__execlists_context_pin(struct intel_context *ce,
+			struct intel_engine_cs *engine)
 {
 	void *vaddr;
 	int ret;
 
-	ret = execlists_context_deferred_alloc(ctx, engine, ce);
+	GEM_BUG_ON(!ce->gem_context->ppgtt);
+
+	ret = execlists_context_deferred_alloc(ce, engine);
 	if (ret)
 		goto err;
 	GEM_BUG_ON(!ce->state);
 
-	ret = __context_pin(ctx, ce->state);
+	ret = __context_pin(ce->state);
 	if (ret)
 		goto err;
 
 	vaddr = i915_gem_object_pin_map(ce->state->obj,
-					i915_coherent_map_type(ctx->i915) |
+					i915_coherent_map_type(engine->i915) |
 					I915_MAP_OVERRIDE);
 	if (IS_ERR(vaddr)) {
 		ret = PTR_ERR(vaddr);
@@ -1312,55 +1265,60 @@ __execlists_context_pin(struct intel_engine_cs *engine,
 	if (ret)
 		goto unpin_map;
 
-	ret = i915_gem_context_pin_hw_id(ctx);
+	ret = i915_gem_context_pin_hw_id(ce->gem_context);
 	if (ret)
 		goto unpin_ring;
 
-	intel_lr_context_descriptor_update(ctx, engine, ce);
-
-	GEM_BUG_ON(!intel_ring_offset_valid(ce->ring, ce->ring->head));
-
+	ce->lrc_desc = lrc_descriptor(ce, engine);
 	ce->lrc_reg_state = vaddr + LRC_STATE_PN * PAGE_SIZE;
+	__execlists_update_reg_state(ce, engine);
 
-	__execlists_update_reg_state(engine, ce);
-
-	ce->state->obj->pin_global++;
-	i915_gem_context_get(ctx);
-	return ce;
+	return 0;
 
 unpin_ring:
 	intel_ring_unpin(ce->ring);
 unpin_map:
 	i915_gem_object_unpin_map(ce->state->obj);
 unpin_vma:
-	__i915_vma_unpin(ce->state);
+	__context_unpin(ce->state);
 err:
-	ce->pin_count = 0;
-	return ERR_PTR(ret);
+	return ret;
 }
 
-static const struct intel_context_ops execlists_context_ops = {
-	.unpin = execlists_context_unpin,
-	.destroy = execlists_context_destroy,
-};
-
-static struct intel_context *
-execlists_context_pin(struct intel_engine_cs *engine,
-		      struct i915_gem_context *ctx)
+static int execlists_context_pin(struct intel_context *ce)
 {
-	struct intel_context *ce = to_intel_context(ctx, engine);
-
-	lockdep_assert_held(&ctx->i915->drm.struct_mutex);
-	GEM_BUG_ON(!ctx->ppgtt);
+	return __execlists_context_pin(ce, ce->engine);
+}
 
-	if (likely(ce->pin_count++))
-		return ce;
-	GEM_BUG_ON(!ce->pin_count); /* no overflow please! */
+static void execlists_context_reset(struct intel_context *ce)
+{
+	/*
+	 * Because we emit WA_TAIL_DWORDS there may be a disparity
+	 * between our bookkeeping in ce->ring->head and ce->ring->tail and
+	 * that stored in context. As we only write new commands from
+	 * ce->ring->tail onwards, everything before that is junk. If the GPU
+	 * starts reading from its RING_HEAD from the context, it may try to
+	 * execute that junk and die.
+	 *
+	 * The contexts that are stilled pinned on resume belong to the
+	 * kernel, and are local to each engine. All other contexts will
+	 * have their head/tail sanitized upon pinning before use, so they
+	 * will never see garbage,
+	 *
+	 * So to avoid that we reset the context images upon resume. For
+	 * simplicity, we just zero everything out.
+	 */
+	intel_ring_reset(ce->ring, 0);
+	__execlists_update_reg_state(ce, ce->engine);
+}
 
-	ce->ops = &execlists_context_ops;
+static const struct intel_context_ops execlists_context_ops = {
+	.pin = execlists_context_pin,
+	.unpin = execlists_context_unpin,
 
-	return __execlists_context_pin(engine, ctx, ce);
-}
+	.reset = execlists_context_reset,
+	.destroy = execlists_context_destroy,
+};
 
 static int gen8_emit_init_breadcrumb(struct i915_request *rq)
 {
@@ -1387,6 +1345,10 @@ static int gen8_emit_init_breadcrumb(struct i915_request *rq)
 	*cs++ = rq->fence.seqno - 1;
 
 	intel_ring_advance(rq, cs);
+
+	/* Record the updated position of the request's payload */
+	rq->infix = intel_ring_offset(rq, cs);
+
 	return 0;
 }
 
@@ -1424,10 +1386,11 @@ static int emit_pdps(struct i915_request *rq)
 	*cs++ = MI_LOAD_REGISTER_IMM(2 * GEN8_3LVL_PDPES) | MI_LRI_FORCE_POSTED;
 	for (i = GEN8_3LVL_PDPES; i--; ) {
 		const dma_addr_t pd_daddr = i915_page_dir_dma_addr(ppgtt, i);
+		u32 base = engine->mmio_base;
 
-		*cs++ = i915_mmio_reg_offset(GEN8_RING_PDP_UDW(engine, i));
+		*cs++ = i915_mmio_reg_offset(GEN8_RING_PDP_UDW(base, i));
 		*cs++ = upper_32_bits(pd_daddr);
-		*cs++ = i915_mmio_reg_offset(GEN8_RING_PDP_LDW(engine, i));
+		*cs++ = i915_mmio_reg_offset(GEN8_RING_PDP_LDW(base, i));
 		*cs++ = lower_32_bits(pd_daddr);
 	}
 	*cs++ = MI_NOOP;
@@ -1447,7 +1410,7 @@ static int execlists_request_alloc(struct i915_request *request)
 {
 	int ret;
 
-	GEM_BUG_ON(!request->hw_context->pin_count);
+	GEM_BUG_ON(!intel_context_is_pinned(request->hw_context));
 
 	/*
 	 * Flush enough space to reduce the likelihood of waiting after
@@ -1465,7 +1428,7 @@ static int execlists_request_alloc(struct i915_request *request)
 	 */
 
 	/* Unconditionally invalidate GPU caches and TLBs. */
-	if (i915_vm_is_48bit(&request->gem_context->ppgtt->vm))
+	if (i915_vm_is_4lvl(&request->gem_context->ppgtt->vm))
 		ret = request->engine->emit_flush(request, EMIT_INVALIDATE);
 	else
 		ret = emit_pdps(request);
@@ -1732,7 +1695,7 @@ static int intel_init_workaround_bb(struct intel_engine_cs *engine)
 	unsigned int i;
 	int ret;
 
-	if (GEM_DEBUG_WARN_ON(engine->id != RCS))
+	if (GEM_DEBUG_WARN_ON(engine->id != RCS0))
 		return -EINVAL;
 
 	switch (INTEL_GEN(engine->i915)) {
@@ -1796,17 +1759,9 @@ static void enable_execlists(struct intel_engine_cs *engine)
 
 	intel_engine_set_hwsp_writemask(engine, ~0u); /* HWSTAM */
 
-	/*
-	 * Make sure we're not enabling the new 12-deep CSB
-	 * FIFO as that requires a slightly updated handling
-	 * in the ctx switch irq. Since we're currently only
-	 * using only 2 elements of the enhanced execlists the
-	 * deeper FIFO it's not needed and it's not worth adding
-	 * more statements to the irq handler to support it.
-	 */
 	if (INTEL_GEN(dev_priv) >= 11)
 		I915_WRITE(RING_MODE_GEN7(engine),
-			   _MASKED_BIT_DISABLE(GEN11_GFX_DISABLE_LEGACY_MODE));
+			   _MASKED_BIT_ENABLE(GEN11_GFX_DISABLE_LEGACY_MODE));
 	else
 		I915_WRITE(RING_MODE_GEN7(engine),
 			   _MASKED_BIT_ENABLE(GFX_RUN_LIST_ENABLE));
@@ -1872,20 +1827,72 @@ static void execlists_reset_prepare(struct intel_engine_cs *engine)
 	__tasklet_disable_sync_once(&execlists->tasklet);
 	GEM_BUG_ON(!reset_in_progress(execlists));
 
+	intel_engine_stop_cs(engine);
+
 	/* And flush any current direct submission. */
 	spin_lock_irqsave(&engine->timeline.lock, flags);
-	process_csb(engine); /* drain preemption events */
 	spin_unlock_irqrestore(&engine->timeline.lock, flags);
 }
 
-static void execlists_reset(struct intel_engine_cs *engine, bool stalled)
+static bool lrc_regs_ok(const struct i915_request *rq)
+{
+	const struct intel_ring *ring = rq->ring;
+	const u32 *regs = rq->hw_context->lrc_reg_state;
+
+	/* Quick spot check for the common signs of context corruption */
+
+	if (regs[CTX_RING_BUFFER_CONTROL + 1] !=
+	    (RING_CTL_SIZE(ring->size) | RING_VALID))
+		return false;
+
+	if (regs[CTX_RING_BUFFER_START + 1] != i915_ggtt_offset(ring->vma))
+		return false;
+
+	return true;
+}
+
+static void reset_csb_pointers(struct intel_engine_execlists *execlists)
+{
+	const unsigned int reset_value = execlists->csb_size - 1;
+
+	/*
+	 * After a reset, the HW starts writing into CSB entry [0]. We
+	 * therefore have to set our HEAD pointer back one entry so that
+	 * the *first* entry we check is entry 0. To complicate this further,
+	 * as we don't wait for the first interrupt after reset, we have to
+	 * fake the HW write to point back to the last entry so that our
+	 * inline comparison of our cached head position against the last HW
+	 * write works even before the first interrupt.
+	 */
+	execlists->csb_head = reset_value;
+	WRITE_ONCE(*execlists->csb_write, reset_value);
+	wmb(); /* Make sure this is visible to HW (paranoia?) */
+
+	invalidate_csb_entries(&execlists->csb_status[0],
+			       &execlists->csb_status[reset_value]);
+}
+
+static void __execlists_reset(struct intel_engine_cs *engine, bool stalled)
 {
 	struct intel_engine_execlists * const execlists = &engine->execlists;
+	struct intel_context *ce;
 	struct i915_request *rq;
-	unsigned long flags;
 	u32 *regs;
 
-	spin_lock_irqsave(&engine->timeline.lock, flags);
+	process_csb(engine); /* drain preemption events */
+
+	/* Following the reset, we need to reload the CSB read/write pointers */
+	reset_csb_pointers(&engine->execlists);
+
+	/*
+	 * Save the currently executing context, even if we completed
+	 * its request, it was still running at the time of the
+	 * reset and will have been clobbered.
+	 */
+	if (!port_isset(execlists->port))
+		goto out_clear;
+
+	ce = port_request(execlists->port)->hw_context;
 
 	/*
 	 * Catch up with any missed context-switch interrupts.
@@ -1900,17 +1907,28 @@ static void execlists_reset(struct intel_engine_cs *engine, bool stalled)
 
 	/* Push back any incomplete requests for replay after the reset. */
 	rq = __unwind_incomplete_requests(engine);
+	if (!rq)
+		goto out_replay;
 
-	/* Following the reset, we need to reload the CSB read/write pointers */
-	reset_csb_pointers(&engine->execlists);
+	if (rq->hw_context != ce) { /* caught just before a CS event */
+		rq = NULL;
+		goto out_replay;
+	}
 
-	GEM_TRACE("%s seqno=%d, current=%d, stalled? %s\n",
-		  engine->name,
-		  rq ? rq->global_seqno : 0,
-		  intel_engine_get_seqno(engine),
-		  yesno(stalled));
-	if (!rq)
-		goto out_unlock;
+	/*
+	 * If this request hasn't started yet, e.g. it is waiting on a
+	 * semaphore, we need to avoid skipping the request or else we
+	 * break the signaling chain. However, if the context is corrupt
+	 * the request will not restart and we will be stuck with a wedged
+	 * device. It is quite often the case that if we issue a reset
+	 * while the GPU is loading the context image, that the context
+	 * image becomes corrupt.
+	 *
+	 * Otherwise, if we have not started yet, the request should replay
+	 * perfectly and we do not need to flag the result as being erroneous.
+	 */
+	if (!i915_request_started(rq) && lrc_regs_ok(rq))
+		goto out_replay;
 
 	/*
 	 * If the request was innocent, we leave the request in the ELSP
@@ -1924,8 +1942,8 @@ static void execlists_reset(struct intel_engine_cs *engine, bool stalled)
 	 * image back to the expected values to skip over the guilty request.
 	 */
 	i915_reset_request(rq, stalled);
-	if (!stalled)
-		goto out_unlock;
+	if (!stalled && lrc_regs_ok(rq))
+		goto out_replay;
 
 	/*
 	 * We want a simple context + ring to execute the breadcrumb update.
@@ -1935,21 +1953,103 @@ static void execlists_reset(struct intel_engine_cs *engine, bool stalled)
 	 * future request will be after userspace has had the opportunity
 	 * to recreate its own state.
 	 */
-	regs = rq->hw_context->lrc_reg_state;
+	regs = ce->lrc_reg_state;
 	if (engine->pinned_default_state) {
 		memcpy(regs, /* skip restoring the vanilla PPHWSP */
 		       engine->pinned_default_state + LRC_STATE_PN * PAGE_SIZE,
 		       engine->context_size - PAGE_SIZE);
 	}
+	execlists_init_reg_state(regs, ce, engine, ce->ring);
 
-	/* Move the RING_HEAD onto the breadcrumb, past the hanging batch */
-	rq->ring->head = intel_ring_wrap(rq->ring, rq->postfix);
-	intel_ring_update_space(rq->ring);
+	/* Rerun the request; its payload has been neutered (if guilty). */
+out_replay:
+	ce->ring->head =
+		rq ? intel_ring_wrap(ce->ring, rq->head) : ce->ring->tail;
+	intel_ring_update_space(ce->ring);
+	__execlists_update_reg_state(ce, engine);
+
+out_clear:
+	execlists_clear_all_active(execlists);
+}
+
+static void execlists_reset(struct intel_engine_cs *engine, bool stalled)
+{
+	unsigned long flags;
+
+	GEM_TRACE("%s\n", engine->name);
+
+	spin_lock_irqsave(&engine->timeline.lock, flags);
 
-	execlists_init_reg_state(regs, rq->gem_context, engine, rq->ring);
-	__execlists_update_reg_state(engine, rq->hw_context);
+	__execlists_reset(engine, stalled);
+
+	spin_unlock_irqrestore(&engine->timeline.lock, flags);
+}
+
+static void nop_submission_tasklet(unsigned long data)
+{
+	/* The driver is wedged; don't process any more events. */
+}
+
+static void execlists_cancel_requests(struct intel_engine_cs *engine)
+{
+	struct intel_engine_execlists * const execlists = &engine->execlists;
+	struct i915_request *rq, *rn;
+	struct rb_node *rb;
+	unsigned long flags;
+
+	GEM_TRACE("%s\n", engine->name);
+
+	/*
+	 * Before we call engine->cancel_requests(), we should have exclusive
+	 * access to the submission state. This is arranged for us by the
+	 * caller disabling the interrupt generation, the tasklet and other
+	 * threads that may then access the same state, giving us a free hand
+	 * to reset state. However, we still need to let lockdep be aware that
+	 * we know this state may be accessed in hardirq context, so we
+	 * disable the irq around this manipulation and we want to keep
+	 * the spinlock focused on its duties and not accidentally conflate
+	 * coverage to the submission's irq state. (Similarly, although we
+	 * shouldn't need to disable irq around the manipulation of the
+	 * submission's irq state, we also wish to remind ourselves that
+	 * it is irq state.)
+	 */
+	spin_lock_irqsave(&engine->timeline.lock, flags);
+
+	__execlists_reset(engine, true);
+
+	/* Mark all executing requests as skipped. */
+	list_for_each_entry(rq, &engine->timeline.requests, link) {
+		if (!i915_request_signaled(rq))
+			dma_fence_set_error(&rq->fence, -EIO);
+
+		i915_request_mark_complete(rq);
+	}
+
+	/* Flush the queued requests to the timeline list (for retiring). */
+	while ((rb = rb_first_cached(&execlists->queue))) {
+		struct i915_priolist *p = to_priolist(rb);
+		int i;
+
+		priolist_for_each_request_consume(rq, rn, p, i) {
+			list_del_init(&rq->sched.link);
+			__i915_request_submit(rq);
+			dma_fence_set_error(&rq->fence, -EIO);
+			i915_request_mark_complete(rq);
+		}
+
+		rb_erase_cached(&p->node, &execlists->queue);
+		i915_priolist_free(p);
+	}
+
+	/* Remaining _unready_ requests will be nop'ed when submitted */
+
+	execlists->queue_priority_hint = INT_MIN;
+	execlists->queue = RB_ROOT_CACHED;
+	GEM_BUG_ON(port_isset(execlists->port));
+
+	GEM_BUG_ON(__tasklet_is_enabled(&execlists->tasklet));
+	execlists->tasklet.func = nop_submission_tasklet;
 
-out_unlock:
 	spin_unlock_irqrestore(&engine->timeline.lock, flags);
 }
 
@@ -1961,13 +2061,14 @@ static void execlists_reset_finish(struct intel_engine_cs *engine)
 	 * After a GPU reset, we may have requests to replay. Do so now while
 	 * we still have the forcewake to be sure that the GPU is not allowed
 	 * to sleep before we restart and reload a context.
-	 *
 	 */
 	GEM_BUG_ON(!reset_in_progress(execlists));
 	if (!RB_EMPTY_ROOT(&execlists->queue.rb_root))
 		execlists->tasklet.func(execlists->tasklet.data);
 
-	tasklet_enable(&execlists->tasklet);
+	if (__tasklet_enable(&execlists->tasklet))
+		/* And kick in case we missed a new request submission. */
+		tasklet_hi_schedule(&execlists->tasklet);
 	GEM_TRACE("%s: depth->%d\n", engine->name,
 		  atomic_read(&execlists->tasklet.count));
 }
@@ -1978,7 +2079,7 @@ static int gen8_emit_bb_start(struct i915_request *rq,
 {
 	u32 *cs;
 
-	cs = intel_ring_begin(rq, 6);
+	cs = intel_ring_begin(rq, 4);
 	if (IS_ERR(cs))
 		return PTR_ERR(cs);
 
@@ -1989,19 +2090,37 @@ static int gen8_emit_bb_start(struct i915_request *rq,
 	 * particular all the gen that do not need the w/a at all!), if we
 	 * took care to make sure that on every switch into this context
 	 * (both ordinary and for preemption) that arbitrartion was enabled
-	 * we would be fine. However, there doesn't seem to be a downside to
-	 * being paranoid and making sure it is set before each batch and
-	 * every context-switch.
-	 *
-	 * Note that if we fail to enable arbitration before the request
-	 * is complete, then we do not see the context-switch interrupt and
-	 * the engine hangs (with RING_HEAD == RING_TAIL).
-	 *
-	 * That satisfies both the GPGPU w/a and our heavy-handed paranoia.
+	 * we would be fine.  However, for gen8 there is another w/a that
+	 * requires us to not preempt inside GPGPU execution, so we keep
+	 * arbitration disabled for gen8 batches. Arbitration will be
+	 * re-enabled before we close the request
+	 * (engine->emit_fini_breadcrumb).
 	 */
+	*cs++ = MI_ARB_ON_OFF | MI_ARB_DISABLE;
+
+	/* FIXME(BDW+): Address space and security selectors. */
+	*cs++ = MI_BATCH_BUFFER_START_GEN8 |
+		(flags & I915_DISPATCH_SECURE ? 0 : BIT(8));
+	*cs++ = lower_32_bits(offset);
+	*cs++ = upper_32_bits(offset);
+
+	intel_ring_advance(rq, cs);
+
+	return 0;
+}
+
+static int gen9_emit_bb_start(struct i915_request *rq,
+			      u64 offset, u32 len,
+			      const unsigned int flags)
+{
+	u32 *cs;
+
+	cs = intel_ring_begin(rq, 6);
+	if (IS_ERR(cs))
+		return PTR_ERR(cs);
+
 	*cs++ = MI_ARB_ON_OFF | MI_ARB_ENABLE;
 
-	/* FIXME(BDW): Address space and security selectors. */
 	*cs++ = MI_BATCH_BUFFER_START_GEN8 |
 		(flags & I915_DISPATCH_SECURE ? 0 : BIT(8));
 	*cs++ = lower_32_bits(offset);
@@ -2017,16 +2136,14 @@ static int gen8_emit_bb_start(struct i915_request *rq,
 
 static void gen8_logical_ring_enable_irq(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = engine->i915;
-	I915_WRITE_IMR(engine,
-		       ~(engine->irq_enable_mask | engine->irq_keep_mask));
-	POSTING_READ_FW(RING_IMR(engine->mmio_base));
+	ENGINE_WRITE(engine, RING_IMR,
+		     ~(engine->irq_enable_mask | engine->irq_keep_mask));
+	ENGINE_POSTING_READ(engine, RING_IMR);
 }
 
 static void gen8_logical_ring_disable_irq(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = engine->i915;
-	I915_WRITE_IMR(engine, ~engine->irq_keep_mask);
+	ENGINE_WRITE(engine, RING_IMR, ~engine->irq_keep_mask);
 }
 
 static int gen8_emit_flush(struct i915_request *request, u32 mode)
@@ -2148,16 +2265,16 @@ static u32 *gen8_emit_wa_tail(struct i915_request *request, u32 *cs)
 
 static u32 *gen8_emit_fini_breadcrumb(struct i915_request *request, u32 *cs)
 {
-	/* w/a: bit 5 needs to be zero for MI_FLUSH_DW address. */
-	BUILD_BUG_ON(I915_GEM_HWS_INDEX_ADDR & (1 << 5));
-
 	cs = gen8_emit_ggtt_write(cs,
 				  request->fence.seqno,
-				  request->timeline->hwsp_offset);
+				  request->timeline->hwsp_offset,
+				  0);
 
 	cs = gen8_emit_ggtt_write(cs,
-				  request->global_seqno,
-				  intel_hws_seqno_address(request->engine));
+				  intel_engine_next_hangcheck_seqno(request->engine),
+				  I915_GEM_HWS_HANGCHECK_ADDR,
+				  MI_FLUSH_DW_STORE_INDEX);
+
 
 	*cs++ = MI_USER_INTERRUPT;
 	*cs++ = MI_ARB_ON_OFF | MI_ARB_ENABLE;
@@ -2180,9 +2297,9 @@ static u32 *gen8_emit_fini_breadcrumb_rcs(struct i915_request *request, u32 *cs)
 				      PIPE_CONTROL_CS_STALL);
 
 	cs = gen8_emit_ggtt_write_rcs(cs,
-				      request->global_seqno,
-				      intel_hws_seqno_address(request->engine),
-				      PIPE_CONTROL_CS_STALL);
+				      intel_engine_next_hangcheck_seqno(request->engine),
+				      I915_GEM_HWS_HANGCHECK_ADDR,
+				      PIPE_CONTROL_STORE_DATA_INDEX);
 
 	*cs++ = MI_USER_INTERRUPT;
 	*cs++ = MI_ARB_ON_OFF | MI_ARB_ENABLE;
@@ -2231,7 +2348,7 @@ void intel_logical_ring_cleanup(struct intel_engine_cs *engine)
 	dev_priv = engine->i915;
 
 	if (engine->buffer) {
-		WARN_ON((I915_READ_MODE(engine) & MODE_IDLE) == 0);
+		WARN_ON((ENGINE_READ(engine, RING_MI_MODE) & MODE_IDLE) == 0);
 	}
 
 	if (engine->cleanup)
@@ -2254,19 +2371,18 @@ void intel_execlists_set_default_submission(struct intel_engine_cs *engine)
 	engine->execlists.tasklet.func = execlists_submission_tasklet;
 
 	engine->reset.prepare = execlists_reset_prepare;
+	engine->reset.reset = execlists_reset;
+	engine->reset.finish = execlists_reset_finish;
 
 	engine->park = NULL;
 	engine->unpark = NULL;
 
 	engine->flags |= I915_ENGINE_SUPPORTS_STATS;
-	if (engine->i915->preempt_context)
+	if (!intel_vgpu_active(engine->i915))
+		engine->flags |= I915_ENGINE_HAS_SEMAPHORES;
+	if (engine->preempt_context &&
+	    HAS_LOGICAL_RING_PREEMPTION(engine->i915))
 		engine->flags |= I915_ENGINE_HAS_PREEMPTION;
-
-	engine->i915->caps.scheduler =
-		I915_SCHEDULER_CAP_ENABLED |
-		I915_SCHEDULER_CAP_PRIORITY;
-	if (intel_engine_has_preemption(engine))
-		engine->i915->caps.scheduler |= I915_SCHEDULER_CAP_PREEMPTION;
 }
 
 static void
@@ -2279,7 +2395,7 @@ logical_ring_default_vfuncs(struct intel_engine_cs *engine)
 	engine->reset.reset = execlists_reset;
 	engine->reset.finish = execlists_reset_finish;
 
-	engine->context_pin = execlists_context_pin;
+	engine->cops = &execlists_context_ops;
 	engine->request_alloc = execlists_request_alloc;
 
 	engine->emit_flush = gen8_emit_flush;
@@ -2299,7 +2415,10 @@ logical_ring_default_vfuncs(struct intel_engine_cs *engine)
 		 * until a more refined solution exists.
 		 */
 	}
-	engine->emit_bb_start = gen8_emit_bb_start;
+	if (IS_GEN(engine->i915, 8))
+		engine->emit_bb_start = gen8_emit_bb_start;
+	else
+		engine->emit_bb_start = gen9_emit_bb_start;
 }
 
 static inline void
@@ -2309,11 +2428,11 @@ logical_ring_default_irqs(struct intel_engine_cs *engine)
 
 	if (INTEL_GEN(engine->i915) < 11) {
 		const u8 irq_shifts[] = {
-			[RCS]  = GEN8_RCS_IRQ_SHIFT,
-			[BCS]  = GEN8_BCS_IRQ_SHIFT,
-			[VCS]  = GEN8_VCS1_IRQ_SHIFT,
-			[VCS2] = GEN8_VCS2_IRQ_SHIFT,
-			[VECS] = GEN8_VECS_IRQ_SHIFT,
+			[RCS0]  = GEN8_RCS_IRQ_SHIFT,
+			[BCS0]  = GEN8_BCS_IRQ_SHIFT,
+			[VCS0]  = GEN8_VCS0_IRQ_SHIFT,
+			[VCS1]  = GEN8_VCS1_IRQ_SHIFT,
+			[VECS0] = GEN8_VECS_IRQ_SHIFT,
 		};
 
 		shift = irq_shifts[engine->id];
@@ -2348,6 +2467,7 @@ static int logical_ring_init(struct intel_engine_cs *engine)
 {
 	struct drm_i915_private *i915 = engine->i915;
 	struct intel_engine_execlists * const execlists = &engine->execlists;
+	u32 base = engine->mmio_base;
 	int ret;
 
 	ret = intel_engine_init_common(engine);
@@ -2357,23 +2477,19 @@ static int logical_ring_init(struct intel_engine_cs *engine)
 	intel_engine_init_workarounds(engine);
 
 	if (HAS_LOGICAL_RING_ELSQ(i915)) {
-		execlists->submit_reg = i915->regs +
-			i915_mmio_reg_offset(RING_EXECLIST_SQ_CONTENTS(engine));
-		execlists->ctrl_reg = i915->regs +
-			i915_mmio_reg_offset(RING_EXECLIST_CONTROL(engine));
+		execlists->submit_reg = i915->uncore.regs +
+			i915_mmio_reg_offset(RING_EXECLIST_SQ_CONTENTS(base));
+		execlists->ctrl_reg = i915->uncore.regs +
+			i915_mmio_reg_offset(RING_EXECLIST_CONTROL(base));
 	} else {
-		execlists->submit_reg = i915->regs +
-			i915_mmio_reg_offset(RING_ELSP(engine));
+		execlists->submit_reg = i915->uncore.regs +
+			i915_mmio_reg_offset(RING_ELSP(base));
 	}
 
 	execlists->preempt_complete_status = ~0u;
-	if (i915->preempt_context) {
-		struct intel_context *ce =
-			to_intel_context(i915->preempt_context, engine);
-
+	if (engine->preempt_context)
 		execlists->preempt_complete_status =
-			upper_32_bits(ce->lrc_desc);
-	}
+			upper_32_bits(engine->preempt_context->lrc_desc);
 
 	execlists->csb_status =
 		&engine->status_page.addr[I915_HWS_CSB_BUF0_INDEX];
@@ -2381,6 +2497,11 @@ static int logical_ring_init(struct intel_engine_cs *engine)
 	execlists->csb_write =
 		&engine->status_page.addr[intel_hws_csb_write_index(i915)];
 
+	if (INTEL_GEN(engine->i915) < 11)
+		execlists->csb_size = GEN8_CSB_ENTRIES;
+	else
+		execlists->csb_size = GEN11_CSB_ENTRIES;
+
 	reset_csb_pointers(execlists);
 
 	return 0;
@@ -2592,13 +2713,13 @@ static u32 intel_lr_indirect_ctx_offset(struct intel_engine_cs *engine)
 }
 
 static void execlists_init_reg_state(u32 *regs,
-				     struct i915_gem_context *ctx,
+				     struct intel_context *ce,
 				     struct intel_engine_cs *engine,
 				     struct intel_ring *ring)
 {
-	struct drm_i915_private *dev_priv = engine->i915;
-	u32 base = engine->mmio_base;
+	struct i915_hw_ppgtt *ppgtt = ce->gem_context->ppgtt;
 	bool rcs = engine->class == RENDER_CLASS;
+	u32 base = engine->mmio_base;
 
 	/* A context is actually a big batch buffer with several
 	 * MI_LOAD_REGISTER_IMM commands followed by (reg, value) pairs. The
@@ -2610,10 +2731,10 @@ static void execlists_init_reg_state(u32 *regs,
 	regs[CTX_LRI_HEADER_0] = MI_LOAD_REGISTER_IMM(rcs ? 14 : 11) |
 				 MI_LRI_FORCE_POSTED;
 
-	CTX_REG(regs, CTX_CONTEXT_CONTROL, RING_CONTEXT_CONTROL(engine),
+	CTX_REG(regs, CTX_CONTEXT_CONTROL, RING_CONTEXT_CONTROL(base),
 		_MASKED_BIT_DISABLE(CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT) |
 		_MASKED_BIT_ENABLE(CTX_CTRL_INHIBIT_SYN_CTX_SWITCH));
-	if (INTEL_GEN(dev_priv) < 11) {
+	if (INTEL_GEN(engine->i915) < 11) {
 		regs[CTX_CONTEXT_CONTROL + 1] |=
 			_MASKED_BIT_DISABLE(CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT |
 					    CTX_CTRL_RS_CTX_ENABLE);
@@ -2659,42 +2780,42 @@ static void execlists_init_reg_state(u32 *regs,
 
 	CTX_REG(regs, CTX_CTX_TIMESTAMP, RING_CTX_TIMESTAMP(base), 0);
 	/* PDP values well be assigned later if needed */
-	CTX_REG(regs, CTX_PDP3_UDW, GEN8_RING_PDP_UDW(engine, 3), 0);
-	CTX_REG(regs, CTX_PDP3_LDW, GEN8_RING_PDP_LDW(engine, 3), 0);
-	CTX_REG(regs, CTX_PDP2_UDW, GEN8_RING_PDP_UDW(engine, 2), 0);
-	CTX_REG(regs, CTX_PDP2_LDW, GEN8_RING_PDP_LDW(engine, 2), 0);
-	CTX_REG(regs, CTX_PDP1_UDW, GEN8_RING_PDP_UDW(engine, 1), 0);
-	CTX_REG(regs, CTX_PDP1_LDW, GEN8_RING_PDP_LDW(engine, 1), 0);
-	CTX_REG(regs, CTX_PDP0_UDW, GEN8_RING_PDP_UDW(engine, 0), 0);
-	CTX_REG(regs, CTX_PDP0_LDW, GEN8_RING_PDP_LDW(engine, 0), 0);
-
-	if (i915_vm_is_48bit(&ctx->ppgtt->vm)) {
+	CTX_REG(regs, CTX_PDP3_UDW, GEN8_RING_PDP_UDW(base, 3), 0);
+	CTX_REG(regs, CTX_PDP3_LDW, GEN8_RING_PDP_LDW(base, 3), 0);
+	CTX_REG(regs, CTX_PDP2_UDW, GEN8_RING_PDP_UDW(base, 2), 0);
+	CTX_REG(regs, CTX_PDP2_LDW, GEN8_RING_PDP_LDW(base, 2), 0);
+	CTX_REG(regs, CTX_PDP1_UDW, GEN8_RING_PDP_UDW(base, 1), 0);
+	CTX_REG(regs, CTX_PDP1_LDW, GEN8_RING_PDP_LDW(base, 1), 0);
+	CTX_REG(regs, CTX_PDP0_UDW, GEN8_RING_PDP_UDW(base, 0), 0);
+	CTX_REG(regs, CTX_PDP0_LDW, GEN8_RING_PDP_LDW(base, 0), 0);
+
+	if (i915_vm_is_4lvl(&ppgtt->vm)) {
 		/* 64b PPGTT (48bit canonical)
 		 * PDP0_DESCRIPTOR contains the base address to PML4 and
 		 * other PDP Descriptors are ignored.
 		 */
-		ASSIGN_CTX_PML4(ctx->ppgtt, regs);
+		ASSIGN_CTX_PML4(ppgtt, regs);
 	} else {
-		ASSIGN_CTX_PDP(ctx->ppgtt, regs, 3);
-		ASSIGN_CTX_PDP(ctx->ppgtt, regs, 2);
-		ASSIGN_CTX_PDP(ctx->ppgtt, regs, 1);
-		ASSIGN_CTX_PDP(ctx->ppgtt, regs, 0);
+		ASSIGN_CTX_PDP(ppgtt, regs, 3);
+		ASSIGN_CTX_PDP(ppgtt, regs, 2);
+		ASSIGN_CTX_PDP(ppgtt, regs, 1);
+		ASSIGN_CTX_PDP(ppgtt, regs, 0);
 	}
 
 	if (rcs) {
 		regs[CTX_LRI_HEADER_2] = MI_LOAD_REGISTER_IMM(1);
 		CTX_REG(regs, CTX_R_PWR_CLK_STATE, GEN8_R_PWR_CLK_STATE, 0);
 
-		i915_oa_init_reg_state(engine, ctx, regs);
+		i915_oa_init_reg_state(engine, ce, regs);
 	}
 
 	regs[CTX_END] = MI_BATCH_BUFFER_END;
-	if (INTEL_GEN(dev_priv) >= 10)
+	if (INTEL_GEN(engine->i915) >= 10)
 		regs[CTX_END] |= BIT(0);
 }
 
 static int
-populate_lr_context(struct i915_gem_context *ctx,
+populate_lr_context(struct intel_context *ce,
 		    struct drm_i915_gem_object *ctx_obj,
 		    struct intel_engine_cs *engine,
 		    struct intel_ring *ring)
@@ -2703,19 +2824,12 @@ populate_lr_context(struct i915_gem_context *ctx,
 	u32 *regs;
 	int ret;
 
-	ret = i915_gem_object_set_to_cpu_domain(ctx_obj, true);
-	if (ret) {
-		DRM_DEBUG_DRIVER("Could not set to CPU domain\n");
-		return ret;
-	}
-
 	vaddr = i915_gem_object_pin_map(ctx_obj, I915_MAP_WB);
 	if (IS_ERR(vaddr)) {
 		ret = PTR_ERR(vaddr);
 		DRM_DEBUG_DRIVER("Could not map object pages! (%d)\n", ret);
 		return ret;
 	}
-	ctx_obj->mm.dirty = true;
 
 	if (engine->default_state) {
 		/*
@@ -2740,23 +2854,35 @@ populate_lr_context(struct i915_gem_context *ctx,
 	/* The second page of the context object contains some fields which must
 	 * be set up prior to the first execution. */
 	regs = vaddr + LRC_STATE_PN * PAGE_SIZE;
-	execlists_init_reg_state(regs, ctx, engine, ring);
+	execlists_init_reg_state(regs, ce, engine, ring);
 	if (!engine->default_state)
 		regs[CTX_CONTEXT_CONTROL + 1] |=
 			_MASKED_BIT_ENABLE(CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT);
-	if (ctx == ctx->i915->preempt_context && INTEL_GEN(engine->i915) < 11)
+	if (ce->gem_context == engine->i915->preempt_context &&
+	    INTEL_GEN(engine->i915) < 11)
 		regs[CTX_CONTEXT_CONTROL + 1] |=
 			_MASKED_BIT_ENABLE(CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT |
 					   CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT);
 
+	ret = 0;
 err_unpin_ctx:
+	__i915_gem_object_flush_map(ctx_obj,
+				    LRC_HEADER_PAGES * PAGE_SIZE,
+				    engine->context_size);
 	i915_gem_object_unpin_map(ctx_obj);
 	return ret;
 }
 
-static int execlists_context_deferred_alloc(struct i915_gem_context *ctx,
-					    struct intel_engine_cs *engine,
-					    struct intel_context *ce)
+static struct i915_timeline *get_timeline(struct i915_gem_context *ctx)
+{
+	if (ctx->timeline)
+		return i915_timeline_get(ctx->timeline);
+	else
+		return i915_timeline_create(ctx->i915, NULL);
+}
+
+static int execlists_context_deferred_alloc(struct intel_context *ce,
+					    struct intel_engine_cs *engine)
 {
 	struct drm_i915_gem_object *ctx_obj;
 	struct i915_vma *vma;
@@ -2776,30 +2902,32 @@ static int execlists_context_deferred_alloc(struct i915_gem_context *ctx,
 	 */
 	context_size += LRC_HEADER_PAGES * PAGE_SIZE;
 
-	ctx_obj = i915_gem_object_create(ctx->i915, context_size);
+	ctx_obj = i915_gem_object_create(engine->i915, context_size);
 	if (IS_ERR(ctx_obj))
 		return PTR_ERR(ctx_obj);
 
-	vma = i915_vma_instance(ctx_obj, &ctx->i915->ggtt.vm, NULL);
+	vma = i915_vma_instance(ctx_obj, &engine->i915->ggtt.vm, NULL);
 	if (IS_ERR(vma)) {
 		ret = PTR_ERR(vma);
 		goto error_deref_obj;
 	}
 
-	timeline = i915_timeline_create(ctx->i915, ctx->name, NULL);
+	timeline = get_timeline(ce->gem_context);
 	if (IS_ERR(timeline)) {
 		ret = PTR_ERR(timeline);
 		goto error_deref_obj;
 	}
 
-	ring = intel_engine_create_ring(engine, timeline, ctx->ring_size);
+	ring = intel_engine_create_ring(engine,
+					timeline,
+					ce->gem_context->ring_size);
 	i915_timeline_put(timeline);
 	if (IS_ERR(ring)) {
 		ret = PTR_ERR(ring);
 		goto error_deref_obj;
 	}
 
-	ret = populate_lr_context(ctx, ctx_obj, engine, ring);
+	ret = populate_lr_context(ce, ctx_obj, engine, ring);
 	if (ret) {
 		DRM_DEBUG_DRIVER("Failed to populate LRC: %d\n", ret);
 		goto error_ring_free;
@@ -2811,45 +2939,12 @@ static int execlists_context_deferred_alloc(struct i915_gem_context *ctx,
 	return 0;
 
 error_ring_free:
-	intel_ring_free(ring);
+	intel_ring_put(ring);
 error_deref_obj:
 	i915_gem_object_put(ctx_obj);
 	return ret;
 }
 
-void intel_lr_context_resume(struct drm_i915_private *i915)
-{
-	struct intel_engine_cs *engine;
-	struct i915_gem_context *ctx;
-	enum intel_engine_id id;
-
-	/*
-	 * Because we emit WA_TAIL_DWORDS there may be a disparity
-	 * between our bookkeeping in ce->ring->head and ce->ring->tail and
-	 * that stored in context. As we only write new commands from
-	 * ce->ring->tail onwards, everything before that is junk. If the GPU
-	 * starts reading from its RING_HEAD from the context, it may try to
-	 * execute that junk and die.
-	 *
-	 * So to avoid that we reset the context images upon resume. For
-	 * simplicity, we just zero everything out.
-	 */
-	list_for_each_entry(ctx, &i915->contexts.list, link) {
-		for_each_engine(engine, i915, id) {
-			struct intel_context *ce =
-				to_intel_context(ctx, engine);
-
-			if (!ce->state)
-				continue;
-
-			intel_ring_reset(ce->ring, 0);
-
-			if (ce->pin_count) /* otherwise done in context_pin */
-				__execlists_update_reg_state(engine, ce);
-		}
-	}
-}
-
 void intel_execlists_show_requests(struct intel_engine_cs *engine,
 				   struct drm_printer *m,
 				   void (*show_request)(struct drm_printer *m,
@@ -2910,6 +3005,37 @@ void intel_execlists_show_requests(struct intel_engine_cs *engine,
 	spin_unlock_irqrestore(&engine->timeline.lock, flags);
 }
 
+void intel_lr_context_reset(struct intel_engine_cs *engine,
+			    struct intel_context *ce,
+			    u32 head,
+			    bool scrub)
+{
+	/*
+	 * We want a simple context + ring to execute the breadcrumb update.
+	 * We cannot rely on the context being intact across the GPU hang,
+	 * so clear it and rebuild just what we need for the breadcrumb.
+	 * All pending requests for this context will be zapped, and any
+	 * future request will be after userspace has had the opportunity
+	 * to recreate its own state.
+	 */
+	if (scrub) {
+		u32 *regs = ce->lrc_reg_state;
+
+		if (engine->pinned_default_state) {
+			memcpy(regs, /* skip restoring the vanilla PPHWSP */
+			       engine->pinned_default_state + LRC_STATE_PN * PAGE_SIZE,
+			       engine->context_size - PAGE_SIZE);
+		}
+		execlists_init_reg_state(regs, ce, engine, ce->ring);
+	}
+
+	/* Rerun the request; its payload has been neutered (if guilty). */
+	ce->ring->head = head;
+	intel_ring_update_space(ce->ring);
+
+	__execlists_update_reg_state(ce, engine);
+}
+
 #if IS_ENABLED(CONFIG_DRM_I915_SELFTEST)
 #include "selftests/intel_lrc.c"
 #endif
diff --git a/drivers/gpu/drm/i915/intel_lrc.h b/drivers/gpu/drm/i915/intel_lrc.h
index f1aec8a6986f..84aa230ea27b 100644
--- a/drivers/gpu/drm/i915/intel_lrc.h
+++ b/drivers/gpu/drm/i915/intel_lrc.h
@@ -28,20 +28,18 @@
 #include "i915_gem_context.h"
 
 /* Execlists regs */
-#define RING_ELSP(engine)			_MMIO((engine)->mmio_base + 0x230)
-#define RING_EXECLIST_STATUS_LO(engine)		_MMIO((engine)->mmio_base + 0x234)
-#define RING_EXECLIST_STATUS_HI(engine)		_MMIO((engine)->mmio_base + 0x234 + 4)
-#define RING_CONTEXT_CONTROL(engine)		_MMIO((engine)->mmio_base + 0x244)
+#define RING_ELSP(base)				_MMIO((base) + 0x230)
+#define RING_EXECLIST_STATUS_LO(base)		_MMIO((base) + 0x234)
+#define RING_EXECLIST_STATUS_HI(base)		_MMIO((base) + 0x234 + 4)
+#define RING_CONTEXT_CONTROL(base)		_MMIO((base) + 0x244)
 #define	  CTX_CTRL_INHIBIT_SYN_CTX_SWITCH	(1 << 3)
 #define	  CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT	(1 << 0)
-#define   CTX_CTRL_RS_CTX_ENABLE                (1 << 1)
+#define   CTX_CTRL_RS_CTX_ENABLE		(1 << 1)
 #define	  CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT	(1 << 2)
-#define RING_CONTEXT_STATUS_BUF_BASE(engine)	_MMIO((engine)->mmio_base + 0x370)
-#define RING_CONTEXT_STATUS_BUF_LO(engine, i)	_MMIO((engine)->mmio_base + 0x370 + (i) * 8)
-#define RING_CONTEXT_STATUS_BUF_HI(engine, i)	_MMIO((engine)->mmio_base + 0x370 + (i) * 8 + 4)
-#define RING_CONTEXT_STATUS_PTR(engine)		_MMIO((engine)->mmio_base + 0x3a0)
-#define RING_EXECLIST_SQ_CONTENTS(engine)	_MMIO((engine)->mmio_base + 0x510)
-#define RING_EXECLIST_CONTROL(engine)		_MMIO((engine)->mmio_base + 0x550)
+#define RING_CONTEXT_STATUS_PTR(base)		_MMIO((base) + 0x3a0)
+#define RING_EXECLIST_SQ_CONTENTS(base)		_MMIO((base) + 0x510)
+#define RING_EXECLIST_CONTROL(base)		_MMIO((base) + 0x550)
+
 #define	  EL_CTRL_LOAD				(1 << 0)
 
 /* The docs specify that the write pointer wraps around after 5h, "After status
@@ -55,10 +53,11 @@
 #define GEN8_CSB_PTR_MASK 0x7
 #define GEN8_CSB_READ_PTR_MASK (GEN8_CSB_PTR_MASK << 8)
 #define GEN8_CSB_WRITE_PTR_MASK (GEN8_CSB_PTR_MASK << 0)
-#define GEN8_CSB_WRITE_PTR(csb_status) \
-	(((csb_status) & GEN8_CSB_WRITE_PTR_MASK) >> 0)
-#define GEN8_CSB_READ_PTR(csb_status) \
-	(((csb_status) & GEN8_CSB_READ_PTR_MASK) >> 8)
+
+#define GEN11_CSB_ENTRIES 12
+#define GEN11_CSB_PTR_MASK 0xf
+#define GEN11_CSB_READ_PTR_MASK (GEN11_CSB_PTR_MASK << 8)
+#define GEN11_CSB_WRITE_PTR_MASK (GEN11_CSB_PTR_MASK << 0)
 
 enum {
 	INTEL_CONTEXT_SCHEDULE_IN = 0,
@@ -102,9 +101,13 @@ struct drm_printer;
 struct drm_i915_private;
 struct i915_gem_context;
 
-void intel_lr_context_resume(struct drm_i915_private *dev_priv);
 void intel_execlists_set_default_submission(struct intel_engine_cs *engine);
 
+void intel_lr_context_reset(struct intel_engine_cs *engine,
+			    struct intel_context *ce,
+			    u32 head,
+			    bool scrub);
+
 void intel_execlists_show_requests(struct intel_engine_cs *engine,
 				   struct drm_printer *m,
 				   void (*show_request)(struct drm_printer *m,
diff --git a/drivers/gpu/drm/i915/intel_lspcon.c b/drivers/gpu/drm/i915/intel_lspcon.c
index 322bdddda164..7028d0cf3bb1 100644
--- a/drivers/gpu/drm/i915/intel_lspcon.c
+++ b/drivers/gpu/drm/i915/intel_lspcon.c
@@ -22,10 +22,14 @@
  *
  *
  */
-#include <drm/drm_edid.h>
+
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_dp_dual_mode_helper.h>
+#include <drm/drm_edid.h>
+
+#include "intel_dp.h"
 #include "intel_drv.h"
+#include "intel_lspcon.h"
 
 /* LSPCON OUI Vendor ID(signatures) */
 #define LSPCON_VENDOR_PARADE_OUI 0x001CF8
@@ -452,6 +456,14 @@ void lspcon_write_infoframe(struct intel_encoder *encoder,
 	DRM_DEBUG_DRIVER("AVI infoframes updated successfully\n");
 }
 
+void lspcon_read_infoframe(struct intel_encoder *encoder,
+			   const struct intel_crtc_state *crtc_state,
+			   unsigned int type,
+			   void *frame, ssize_t len)
+{
+	/* FIXME implement this */
+}
+
 void lspcon_set_infoframes(struct intel_encoder *encoder,
 			   bool enable,
 			   const struct intel_crtc_state *crtc_state,
@@ -470,6 +482,8 @@ void lspcon_set_infoframes(struct intel_encoder *encoder,
 		return;
 	}
 
+	/* FIXME precompute infoframes */
+
 	ret = drm_hdmi_avi_infoframe_from_display_mode(&frame.avi,
 						       conn_state->connector,
 						       adjusted_mode);
@@ -504,9 +518,10 @@ void lspcon_set_infoframes(struct intel_encoder *encoder,
 				  buf, ret);
 }
 
-bool lspcon_infoframe_enabled(struct intel_encoder *encoder,
+u32 lspcon_infoframes_enabled(struct intel_encoder *encoder,
 			      const struct intel_crtc_state *pipe_config)
 {
+	/* FIXME actually read this from the hw */
 	return enc_to_intel_lspcon(&encoder->base)->active;
 }
 
diff --git a/drivers/gpu/drm/i915/intel_lspcon.h b/drivers/gpu/drm/i915/intel_lspcon.h
new file mode 100644
index 000000000000..37cfddf8a9c5
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_lspcon.h
@@ -0,0 +1,38 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_LSPCON_H__
+#define __INTEL_LSPCON_H__
+
+#include <linux/types.h>
+
+struct drm_connector;
+struct drm_connector_state;
+struct intel_crtc_state;
+struct intel_digital_port;
+struct intel_encoder;
+struct intel_lspcon;
+
+bool lspcon_init(struct intel_digital_port *intel_dig_port);
+void lspcon_resume(struct intel_lspcon *lspcon);
+void lspcon_wait_pcon_mode(struct intel_lspcon *lspcon);
+void lspcon_write_infoframe(struct intel_encoder *encoder,
+			    const struct intel_crtc_state *crtc_state,
+			    unsigned int type,
+			    const void *buf, ssize_t len);
+void lspcon_read_infoframe(struct intel_encoder *encoder,
+			   const struct intel_crtc_state *crtc_state,
+			   unsigned int type,
+			   void *frame, ssize_t len);
+void lspcon_set_infoframes(struct intel_encoder *encoder,
+			   bool enable,
+			   const struct intel_crtc_state *crtc_state,
+			   const struct drm_connector_state *conn_state);
+u32 lspcon_infoframes_enabled(struct intel_encoder *encoder,
+			      const struct intel_crtc_state *pipe_config);
+void lspcon_ycbcr420_config(struct drm_connector *connector,
+			    struct intel_crtc_state *crtc_state);
+
+#endif /* __INTEL_LSPCON_H__ */
diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index b4aa49768e90..51d1d59c1619 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -28,17 +28,22 @@
  */
 
 #include <acpi/button.h>
+#include <linux/acpi.h>
 #include <linux/dmi.h>
 #include <linux/i2c.h>
 #include <linux/slab.h>
 #include <linux/vga_switcheroo.h>
+
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_edid.h>
-#include "intel_drv.h"
 #include <drm/i915_drm.h>
+
 #include "i915_drv.h"
-#include <linux/acpi.h>
+#include "intel_connector.h"
+#include "intel_drv.h"
+#include "intel_lvds.h"
+#include "intel_panel.h"
 
 /* Private structure for the integrated LVDS support */
 struct intel_lvds_pps {
@@ -152,24 +157,17 @@ static void intel_lvds_pps_get_hw_state(struct drm_i915_private *dev_priv,
 	pps->powerdown_on_reset = I915_READ(PP_CONTROL(0)) & PANEL_POWER_RESET;
 
 	val = I915_READ(PP_ON_DELAYS(0));
-	pps->port = (val & PANEL_PORT_SELECT_MASK) >>
-		    PANEL_PORT_SELECT_SHIFT;
-	pps->t1_t2 = (val & PANEL_POWER_UP_DELAY_MASK) >>
-		     PANEL_POWER_UP_DELAY_SHIFT;
-	pps->t5 = (val & PANEL_LIGHT_ON_DELAY_MASK) >>
-		  PANEL_LIGHT_ON_DELAY_SHIFT;
+	pps->port = REG_FIELD_GET(PANEL_PORT_SELECT_MASK, val);
+	pps->t1_t2 = REG_FIELD_GET(PANEL_POWER_UP_DELAY_MASK, val);
+	pps->t5 = REG_FIELD_GET(PANEL_LIGHT_ON_DELAY_MASK, val);
 
 	val = I915_READ(PP_OFF_DELAYS(0));
-	pps->t3 = (val & PANEL_POWER_DOWN_DELAY_MASK) >>
-		  PANEL_POWER_DOWN_DELAY_SHIFT;
-	pps->tx = (val & PANEL_LIGHT_OFF_DELAY_MASK) >>
-		  PANEL_LIGHT_OFF_DELAY_SHIFT;
+	pps->t3 = REG_FIELD_GET(PANEL_POWER_DOWN_DELAY_MASK, val);
+	pps->tx = REG_FIELD_GET(PANEL_LIGHT_OFF_DELAY_MASK, val);
 
 	val = I915_READ(PP_DIVISOR(0));
-	pps->divider = (val & PP_REFERENCE_DIVIDER_MASK) >>
-		       PP_REFERENCE_DIVIDER_SHIFT;
-	val = (val & PANEL_POWER_CYCLE_DELAY_MASK) >>
-	      PANEL_POWER_CYCLE_DELAY_SHIFT;
+	pps->divider = REG_FIELD_GET(PP_REFERENCE_DIVIDER_MASK, val);
+	val = REG_FIELD_GET(PANEL_POWER_CYCLE_DELAY_MASK, val);
 	/*
 	 * Remove the BSpec specified +1 (100ms) offset that accounts for a
 	 * too short power-cycle delay due to the asynchronous programming of
@@ -209,16 +207,19 @@ static void intel_lvds_pps_init_hw(struct drm_i915_private *dev_priv,
 		val |= PANEL_POWER_RESET;
 	I915_WRITE(PP_CONTROL(0), val);
 
-	I915_WRITE(PP_ON_DELAYS(0), (pps->port << PANEL_PORT_SELECT_SHIFT) |
-				    (pps->t1_t2 << PANEL_POWER_UP_DELAY_SHIFT) |
-				    (pps->t5 << PANEL_LIGHT_ON_DELAY_SHIFT));
-	I915_WRITE(PP_OFF_DELAYS(0), (pps->t3 << PANEL_POWER_DOWN_DELAY_SHIFT) |
-				     (pps->tx << PANEL_LIGHT_OFF_DELAY_SHIFT));
+	I915_WRITE(PP_ON_DELAYS(0),
+		   REG_FIELD_PREP(PANEL_PORT_SELECT_MASK, pps->port) |
+		   REG_FIELD_PREP(PANEL_POWER_UP_DELAY_MASK, pps->t1_t2) |
+		   REG_FIELD_PREP(PANEL_LIGHT_ON_DELAY_MASK, pps->t5));
+
+	I915_WRITE(PP_OFF_DELAYS(0),
+		   REG_FIELD_PREP(PANEL_POWER_DOWN_DELAY_MASK, pps->t3) |
+		   REG_FIELD_PREP(PANEL_LIGHT_OFF_DELAY_MASK, pps->tx));
 
-	val = pps->divider << PP_REFERENCE_DIVIDER_SHIFT;
-	val |= (DIV_ROUND_UP(pps->t4, 1000) + 1) <<
-	       PANEL_POWER_CYCLE_DELAY_SHIFT;
-	I915_WRITE(PP_DIVISOR(0), val);
+	I915_WRITE(PP_DIVISOR(0),
+		   REG_FIELD_PREP(PP_REFERENCE_DIVIDER_MASK, pps->divider) |
+		   REG_FIELD_PREP(PANEL_POWER_CYCLE_DELAY_MASK,
+				  DIV_ROUND_UP(pps->t4, 1000) + 1));
 }
 
 static void intel_pre_enable_lvds(struct intel_encoder *encoder,
@@ -315,7 +316,8 @@ static void intel_enable_lvds(struct intel_encoder *encoder,
 	I915_WRITE(PP_CONTROL(0), I915_READ(PP_CONTROL(0)) | PANEL_POWER_ON);
 	POSTING_READ(lvds_encoder->reg);
 
-	if (intel_wait_for_register(dev_priv, PP_STATUS(0), PP_ON, PP_ON, 5000))
+	if (intel_wait_for_register(&dev_priv->uncore,
+				    PP_STATUS(0), PP_ON, PP_ON, 5000))
 		DRM_ERROR("timed out waiting for panel to power on\n");
 
 	intel_panel_enable_backlight(pipe_config, conn_state);
@@ -329,7 +331,8 @@ static void intel_disable_lvds(struct intel_encoder *encoder,
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 
 	I915_WRITE(PP_CONTROL(0), I915_READ(PP_CONTROL(0)) & ~PANEL_POWER_ON);
-	if (intel_wait_for_register(dev_priv, PP_STATUS(0), PP_ON, 0, 1000))
+	if (intel_wait_for_register(&dev_priv->uncore,
+				    PP_STATUS(0), PP_ON, 0, 1000))
 		DRM_ERROR("timed out waiting for panel to power off\n");
 
 	I915_WRITE(lvds_encoder->reg, I915_READ(lvds_encoder->reg) & ~LVDS_PORT_EN);
@@ -746,20 +749,21 @@ static const struct dmi_system_id intel_dual_link_lvds[] = {
 	{ }	/* terminating entry */
 };
 
-struct intel_encoder *intel_get_lvds_encoder(struct drm_device *dev)
+struct intel_encoder *intel_get_lvds_encoder(struct drm_i915_private *dev_priv)
 {
-	struct intel_encoder *intel_encoder;
+	struct intel_encoder *encoder;
 
-	for_each_intel_encoder(dev, intel_encoder)
-		if (intel_encoder->type == INTEL_OUTPUT_LVDS)
-			return intel_encoder;
+	for_each_intel_encoder(&dev_priv->drm, encoder) {
+		if (encoder->type == INTEL_OUTPUT_LVDS)
+			return encoder;
+	}
 
 	return NULL;
 }
 
-bool intel_is_dual_link_lvds(struct drm_device *dev)
+bool intel_is_dual_link_lvds(struct drm_i915_private *dev_priv)
 {
-	struct intel_encoder *encoder = intel_get_lvds_encoder(dev);
+	struct intel_encoder *encoder = intel_get_lvds_encoder(dev_priv);
 
 	return encoder && to_lvds_encoder(&encoder->base)->is_dual_link;
 }
@@ -813,7 +817,6 @@ void intel_lvds_init(struct drm_i915_private *dev_priv)
 	struct intel_connector *intel_connector;
 	struct drm_connector *connector;
 	struct drm_encoder *encoder;
-	struct drm_display_mode *scan; /* *modes, *bios_mode; */
 	struct drm_display_mode *fixed_mode = NULL;
 	struct drm_display_mode *downclock_mode = NULL;
 	struct edid *edid;
@@ -952,30 +955,14 @@ void intel_lvds_init(struct drm_i915_private *dev_priv)
 	}
 	intel_connector->edid = edid;
 
-	list_for_each_entry(scan, &connector->probed_modes, head) {
-		if (scan->type & DRM_MODE_TYPE_PREFERRED) {
-			DRM_DEBUG_KMS("using preferred mode from EDID: ");
-			drm_mode_debug_printmodeline(scan);
-
-			fixed_mode = drm_mode_duplicate(dev, scan);
-			if (fixed_mode)
-				goto out;
-		}
-	}
+	fixed_mode = intel_panel_edid_fixed_mode(intel_connector);
+	if (fixed_mode)
+		goto out;
 
 	/* Failed to get EDID, what about VBT? */
-	if (dev_priv->vbt.lfp_lvds_vbt_mode) {
-		DRM_DEBUG_KMS("using mode from VBT: ");
-		drm_mode_debug_printmodeline(dev_priv->vbt.lfp_lvds_vbt_mode);
-
-		fixed_mode = drm_mode_duplicate(dev, dev_priv->vbt.lfp_lvds_vbt_mode);
-		if (fixed_mode) {
-			fixed_mode->type |= DRM_MODE_TYPE_PREFERRED;
-			connector->display_info.width_mm = fixed_mode->width_mm;
-			connector->display_info.height_mm = fixed_mode->height_mm;
-			goto out;
-		}
-	}
+	fixed_mode = intel_panel_vbt_fixed_mode(intel_connector);
+	if (fixed_mode)
+		goto out;
 
 	/*
 	 * If we didn't get EDID, try checking if the panel is already turned
diff --git a/drivers/gpu/drm/i915/intel_lvds.h b/drivers/gpu/drm/i915/intel_lvds.h
new file mode 100644
index 000000000000..bc9c8b84ba2f
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_lvds.h
@@ -0,0 +1,22 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_LVDS_H__
+#define __INTEL_LVDS_H__
+
+#include <linux/types.h>
+
+#include "i915_reg.h"
+
+enum pipe;
+struct drm_i915_private;
+
+bool intel_lvds_port_enabled(struct drm_i915_private *dev_priv,
+			     i915_reg_t lvds_reg, enum pipe *pipe);
+void intel_lvds_init(struct drm_i915_private *dev_priv);
+struct intel_encoder *intel_get_lvds_encoder(struct drm_i915_private *dev_priv);
+bool intel_is_dual_link_lvds(struct drm_i915_private *dev_priv);
+
+#endif /* __INTEL_LVDS_H__ */
diff --git a/drivers/gpu/drm/i915/intel_mocs.c b/drivers/gpu/drm/i915/intel_mocs.c
index 331e7a678fb7..274ba78500c0 100644
--- a/drivers/gpu/drm/i915/intel_mocs.c
+++ b/drivers/gpu/drm/i915/intel_mocs.c
@@ -252,7 +252,7 @@ static bool get_mocs_settings(struct drm_i915_private *dev_priv,
 {
 	bool result = false;
 
-	if (IS_ICELAKE(dev_priv)) {
+	if (INTEL_GEN(dev_priv) >= 11) {
 		table->size  = ARRAY_SIZE(icelake_mocs_table);
 		table->table = icelake_mocs_table;
 		table->n_entries = GEN11_NUM_MOCS_ENTRIES;
@@ -288,17 +288,17 @@ static bool get_mocs_settings(struct drm_i915_private *dev_priv,
 static i915_reg_t mocs_register(enum intel_engine_id engine_id, int index)
 {
 	switch (engine_id) {
-	case RCS:
+	case RCS0:
 		return GEN9_GFX_MOCS(index);
-	case VCS:
+	case VCS0:
 		return GEN9_MFX0_MOCS(index);
-	case BCS:
+	case BCS0:
 		return GEN9_BLT_MOCS(index);
-	case VECS:
+	case VECS0:
 		return GEN9_VEBOX_MOCS(index);
-	case VCS2:
+	case VCS1:
 		return GEN9_MFX1_MOCS(index);
-	case VCS3:
+	case VCS2:
 		return GEN11_MFX2_MOCS(index);
 	default:
 		MISSING_CASE(engine_id);
diff --git a/drivers/gpu/drm/i915/intel_opregion.c b/drivers/gpu/drm/i915/intel_opregion.c
index 5e00ee9270b5..8fa1159d097f 100644
--- a/drivers/gpu/drm/i915/intel_opregion.c
+++ b/drivers/gpu/drm/i915/intel_opregion.c
@@ -32,9 +32,10 @@
 
 #include <drm/i915_drm.h>
 
-#include "intel_opregion.h"
 #include "i915_drv.h"
 #include "intel_drv.h"
+#include "intel_opregion.h"
+#include "intel_panel.h"
 
 #define OPREGION_HEADER_OFFSET 0
 #define OPREGION_ACPI_OFFSET   0x100
diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c
index c0df1dbb0069..eb317759b5d3 100644
--- a/drivers/gpu/drm/i915/intel_overlay.c
+++ b/drivers/gpu/drm/i915/intel_overlay.c
@@ -236,7 +236,7 @@ static int intel_overlay_do_wait_request(struct intel_overlay *overlay,
 static struct i915_request *alloc_request(struct intel_overlay *overlay)
 {
 	struct drm_i915_private *dev_priv = overlay->i915;
-	struct intel_engine_cs *engine = dev_priv->engine[RCS];
+	struct intel_engine_cs *engine = dev_priv->engine[RCS0];
 
 	return i915_request_alloc(engine, dev_priv->kernel_context);
 }
@@ -446,7 +446,7 @@ static int intel_overlay_release_old_vid(struct intel_overlay *overlay)
 	if (!overlay->old_vma)
 		return 0;
 
-	if (I915_READ(ISR) & I915_OVERLAY_PLANE_FLIP_PENDING_INTERRUPT) {
+	if (I915_READ(GEN2_ISR) & I915_OVERLAY_PLANE_FLIP_PENDING_INTERRUPT) {
 		/* synchronous slowpath */
 		struct i915_request *rq;
 
@@ -1430,7 +1430,7 @@ intel_overlay_capture_error_state(struct drm_i915_private *dev_priv)
 		return NULL;
 
 	error->dovsta = I915_READ(DOVSTA);
-	error->isr = I915_READ(ISR);
+	error->isr = I915_READ(GEN2_ISR);
 	error->base = overlay->flip_addr;
 
 	memcpy_fromio(&error->regs, overlay->regs, sizeof(error->regs));
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index beca98d2b035..4ab4ce6569e7 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -33,7 +33,10 @@
 #include <linux/kernel.h>
 #include <linux/moduleparam.h>
 #include <linux/pwm.h>
+
+#include "intel_connector.h"
 #include "intel_drv.h"
+#include "intel_panel.h"
 
 #define CRC_PMIC_PWM_PERIOD_NS	21333
 
@@ -46,27 +49,26 @@ intel_fixed_panel_mode(const struct drm_display_mode *fixed_mode,
 	drm_mode_set_crtcinfo(adjusted_mode, 0);
 }
 
-/**
- * intel_find_panel_downclock - find the reduced downclock for LVDS in EDID
- * @dev_priv: i915 device instance
- * @fixed_mode : panel native mode
- * @connector: LVDS/eDP connector
- *
- * Return downclock_avail
- * Find the reduced downclock for LVDS/eDP in EDID.
- */
-struct drm_display_mode *
-intel_find_panel_downclock(struct drm_i915_private *dev_priv,
-			struct drm_display_mode *fixed_mode,
-			struct drm_connector *connector)
+static bool is_downclock_mode(const struct drm_display_mode *downclock_mode,
+			      const struct drm_display_mode *fixed_mode)
 {
-	struct drm_display_mode *scan, *tmp_mode;
-	int temp_downclock;
+	return drm_mode_match(downclock_mode, fixed_mode,
+			      DRM_MODE_MATCH_TIMINGS |
+			      DRM_MODE_MATCH_FLAGS |
+			      DRM_MODE_MATCH_3D_FLAGS) &&
+		downclock_mode->clock < fixed_mode->clock;
+}
 
-	temp_downclock = fixed_mode->clock;
-	tmp_mode = NULL;
+struct drm_display_mode *
+intel_panel_edid_downclock_mode(struct intel_connector *connector,
+				const struct drm_display_mode *fixed_mode)
+{
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	const struct drm_display_mode *scan, *best_mode = NULL;
+	struct drm_display_mode *downclock_mode;
+	int best_clock = fixed_mode->clock;
 
-	list_for_each_entry(scan, &connector->probed_modes, head) {
+	list_for_each_entry(scan, &connector->base.probed_modes, head) {
 		/*
 		 * If one mode has the same resolution with the fixed_panel
 		 * mode while they have the different refresh rate, it means
@@ -74,29 +76,98 @@ intel_find_panel_downclock(struct drm_i915_private *dev_priv,
 		 * case we can set the different FPx0/1 to dynamically select
 		 * between low and high frequency.
 		 */
-		if (scan->hdisplay == fixed_mode->hdisplay &&
-		    scan->hsync_start == fixed_mode->hsync_start &&
-		    scan->hsync_end == fixed_mode->hsync_end &&
-		    scan->htotal == fixed_mode->htotal &&
-		    scan->vdisplay == fixed_mode->vdisplay &&
-		    scan->vsync_start == fixed_mode->vsync_start &&
-		    scan->vsync_end == fixed_mode->vsync_end &&
-		    scan->vtotal == fixed_mode->vtotal) {
-			if (scan->clock < temp_downclock) {
-				/*
-				 * The downclock is already found. But we
-				 * expect to find the lower downclock.
-				 */
-				temp_downclock = scan->clock;
-				tmp_mode = scan;
-			}
+		if (is_downclock_mode(scan, fixed_mode) &&
+		    scan->clock < best_clock) {
+			/*
+			 * The downclock is already found. But we
+			 * expect to find the lower downclock.
+			 */
+			best_clock = scan->clock;
+			best_mode = scan;
 		}
 	}
 
-	if (temp_downclock < fixed_mode->clock)
-		return drm_mode_duplicate(&dev_priv->drm, tmp_mode);
-	else
+	if (!best_mode)
+		return NULL;
+
+	downclock_mode = drm_mode_duplicate(&dev_priv->drm, best_mode);
+	if (!downclock_mode)
+		return NULL;
+
+	DRM_DEBUG_KMS("[CONNECTOR:%d:%s] using downclock mode from EDID: ",
+		      connector->base.base.id, connector->base.name);
+	drm_mode_debug_printmodeline(downclock_mode);
+
+	return downclock_mode;
+}
+
+struct drm_display_mode *
+intel_panel_edid_fixed_mode(struct intel_connector *connector)
+{
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	const struct drm_display_mode *scan;
+	struct drm_display_mode *fixed_mode;
+
+	if (list_empty(&connector->base.probed_modes))
+		return NULL;
+
+	/* prefer fixed mode from EDID if available */
+	list_for_each_entry(scan, &connector->base.probed_modes, head) {
+		if ((scan->type & DRM_MODE_TYPE_PREFERRED) == 0)
+			continue;
+
+		fixed_mode = drm_mode_duplicate(&dev_priv->drm, scan);
+		if (!fixed_mode)
+			return NULL;
+
+		DRM_DEBUG_KMS("[CONNECTOR:%d:%s] using preferred mode from EDID: ",
+			      connector->base.base.id, connector->base.name);
+		drm_mode_debug_printmodeline(fixed_mode);
+
+		return fixed_mode;
+	}
+
+	scan = list_first_entry(&connector->base.probed_modes,
+				typeof(*scan), head);
+
+	fixed_mode = drm_mode_duplicate(&dev_priv->drm, scan);
+	if (!fixed_mode)
 		return NULL;
+
+	fixed_mode->type |= DRM_MODE_TYPE_PREFERRED;
+
+	DRM_DEBUG_KMS("[CONNECTOR:%d:%s] using first mode from EDID: ",
+		      connector->base.base.id, connector->base.name);
+	drm_mode_debug_printmodeline(fixed_mode);
+
+	return fixed_mode;
+}
+
+struct drm_display_mode *
+intel_panel_vbt_fixed_mode(struct intel_connector *connector)
+{
+	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_display_info *info = &connector->base.display_info;
+	struct drm_display_mode *fixed_mode;
+
+	if (!dev_priv->vbt.lfp_lvds_vbt_mode)
+		return NULL;
+
+	fixed_mode = drm_mode_duplicate(&dev_priv->drm,
+					dev_priv->vbt.lfp_lvds_vbt_mode);
+	if (!fixed_mode)
+		return NULL;
+
+	fixed_mode->type |= DRM_MODE_TYPE_PREFERRED;
+
+	DRM_DEBUG_KMS("[CONNECTOR:%d:%s] using mode from VBT: ",
+		      connector->base.base.id, connector->base.name);
+	drm_mode_debug_printmodeline(fixed_mode);
+
+	info->width_mm = fixed_mode->width_mm;
+	info->height_mm = fixed_mode->height_mm;
+
+	return fixed_mode;
 }
 
 /* adjusted_mode has been preset to be the panel's fixed mode */
@@ -1894,15 +1965,14 @@ intel_panel_init_backlight_funcs(struct intel_panel *panel)
 		panel->backlight.set = bxt_set_backlight;
 		panel->backlight.get = bxt_get_backlight;
 		panel->backlight.hz_to_pwm = bxt_hz_to_pwm;
-	} else if (HAS_PCH_CNP(dev_priv) || HAS_PCH_ICP(dev_priv)) {
+	} else if (INTEL_PCH_TYPE(dev_priv) >= PCH_CNP) {
 		panel->backlight.setup = cnp_setup_backlight;
 		panel->backlight.enable = cnp_enable_backlight;
 		panel->backlight.disable = cnp_disable_backlight;
 		panel->backlight.set = bxt_set_backlight;
 		panel->backlight.get = bxt_get_backlight;
 		panel->backlight.hz_to_pwm = cnp_hz_to_pwm;
-	} else if (HAS_PCH_LPT(dev_priv) || HAS_PCH_SPT(dev_priv) ||
-		   HAS_PCH_KBP(dev_priv)) {
+	} else if (INTEL_PCH_TYPE(dev_priv) >= PCH_LPT) {
 		panel->backlight.setup = lpt_setup_backlight;
 		panel->backlight.enable = lpt_enable_backlight;
 		panel->backlight.disable = lpt_disable_backlight;
diff --git a/drivers/gpu/drm/i915/intel_panel.h b/drivers/gpu/drm/i915/intel_panel.h
new file mode 100644
index 000000000000..cedeea443336
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_panel.h
@@ -0,0 +1,65 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_PANEL_H__
+#define __INTEL_PANEL_H__
+
+#include <linux/types.h>
+
+#include "intel_display.h"
+
+struct drm_connector;
+struct drm_connector_state;
+struct drm_display_mode;
+struct intel_connector;
+struct intel_crtc;
+struct intel_crtc_state;
+struct intel_encoder;
+struct intel_panel;
+
+int intel_panel_init(struct intel_panel *panel,
+		     struct drm_display_mode *fixed_mode,
+		     struct drm_display_mode *downclock_mode);
+void intel_panel_fini(struct intel_panel *panel);
+void intel_fixed_panel_mode(const struct drm_display_mode *fixed_mode,
+			    struct drm_display_mode *adjusted_mode);
+void intel_pch_panel_fitting(struct intel_crtc *crtc,
+			     struct intel_crtc_state *pipe_config,
+			     int fitting_mode);
+void intel_gmch_panel_fitting(struct intel_crtc *crtc,
+			      struct intel_crtc_state *pipe_config,
+			      int fitting_mode);
+void intel_panel_set_backlight_acpi(const struct drm_connector_state *conn_state,
+				    u32 level, u32 max);
+int intel_panel_setup_backlight(struct drm_connector *connector,
+				enum pipe pipe);
+void intel_panel_enable_backlight(const struct intel_crtc_state *crtc_state,
+				  const struct drm_connector_state *conn_state);
+void intel_panel_update_backlight(struct intel_encoder *encoder,
+				  const struct intel_crtc_state *crtc_state,
+				  const struct drm_connector_state *conn_state);
+void intel_panel_disable_backlight(const struct drm_connector_state *old_conn_state);
+struct drm_display_mode *
+intel_panel_edid_downclock_mode(struct intel_connector *connector,
+				const struct drm_display_mode *fixed_mode);
+struct drm_display_mode *
+intel_panel_edid_fixed_mode(struct intel_connector *connector);
+struct drm_display_mode *
+intel_panel_vbt_fixed_mode(struct intel_connector *connector);
+
+#if IS_ENABLED(CONFIG_BACKLIGHT_CLASS_DEVICE)
+int intel_backlight_device_register(struct intel_connector *connector);
+void intel_backlight_device_unregister(struct intel_connector *connector);
+#else /* CONFIG_BACKLIGHT_CLASS_DEVICE */
+static inline int intel_backlight_device_register(struct intel_connector *connector)
+{
+	return 0;
+}
+static inline void intel_backlight_device_unregister(struct intel_connector *connector)
+{
+}
+#endif /* CONFIG_BACKLIGHT_CLASS_DEVICE */
+
+#endif /* __INTEL_PANEL_H__ */
diff --git a/drivers/gpu/drm/i915/intel_pipe_crc.c b/drivers/gpu/drm/i915/intel_pipe_crc.c
index a8554dc4f196..e94b5b1bc1b7 100644
--- a/drivers/gpu/drm/i915/intel_pipe_crc.c
+++ b/drivers/gpu/drm/i915/intel_pipe_crc.c
@@ -24,23 +24,29 @@
  *
  */
 
-#include <linux/seq_file.h>
 #include <linux/circ_buf.h>
 #include <linux/ctype.h>
 #include <linux/debugfs.h>
+#include <linux/seq_file.h>
+
 #include "intel_drv.h"
+#include "intel_pipe_crc.h"
 
 static const char * const pipe_crc_sources[] = {
-	"none",
-	"plane1",
-	"plane2",
-	"pf",
-	"pipe",
-	"TV",
-	"DP-B",
-	"DP-C",
-	"DP-D",
-	"auto",
+	[INTEL_PIPE_CRC_SOURCE_NONE] = "none",
+	[INTEL_PIPE_CRC_SOURCE_PLANE1] = "plane1",
+	[INTEL_PIPE_CRC_SOURCE_PLANE2] = "plane2",
+	[INTEL_PIPE_CRC_SOURCE_PLANE3] = "plane3",
+	[INTEL_PIPE_CRC_SOURCE_PLANE4] = "plane4",
+	[INTEL_PIPE_CRC_SOURCE_PLANE5] = "plane5",
+	[INTEL_PIPE_CRC_SOURCE_PLANE6] = "plane6",
+	[INTEL_PIPE_CRC_SOURCE_PLANE7] = "plane7",
+	[INTEL_PIPE_CRC_SOURCE_PIPE] = "pipe",
+	[INTEL_PIPE_CRC_SOURCE_TV] = "TV",
+	[INTEL_PIPE_CRC_SOURCE_DP_B] = "DP-B",
+	[INTEL_PIPE_CRC_SOURCE_DP_C] = "DP-C",
+	[INTEL_PIPE_CRC_SOURCE_DP_D] = "DP-D",
+	[INTEL_PIPE_CRC_SOURCE_AUTO] = "auto",
 };
 
 static int i8xx_pipe_crc_ctl_reg(enum intel_pipe_crc_source *source,
@@ -192,8 +198,6 @@ static int i9xx_pipe_crc_ctl_reg(struct drm_i915_private *dev_priv,
 				 enum intel_pipe_crc_source *source,
 				 u32 *val)
 {
-	bool need_stable_symbols = false;
-
 	if (*source == INTEL_PIPE_CRC_SOURCE_AUTO) {
 		int ret = i9xx_pipe_crc_auto_source(dev_priv, pipe, source);
 		if (ret)
@@ -209,56 +213,23 @@ static int i9xx_pipe_crc_ctl_reg(struct drm_i915_private *dev_priv,
 			return -EINVAL;
 		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_TV_PRE;
 		break;
-	case INTEL_PIPE_CRC_SOURCE_DP_B:
-		if (!IS_G4X(dev_priv))
-			return -EINVAL;
-		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_DP_B_G4X;
-		need_stable_symbols = true;
-		break;
-	case INTEL_PIPE_CRC_SOURCE_DP_C:
-		if (!IS_G4X(dev_priv))
-			return -EINVAL;
-		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_DP_C_G4X;
-		need_stable_symbols = true;
-		break;
-	case INTEL_PIPE_CRC_SOURCE_DP_D:
-		if (!IS_G4X(dev_priv))
-			return -EINVAL;
-		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_DP_D_G4X;
-		need_stable_symbols = true;
-		break;
 	case INTEL_PIPE_CRC_SOURCE_NONE:
 		*val = 0;
 		break;
 	default:
+		/*
+		 * The DP CRC source doesn't work on g4x.
+		 * It can be made to work to some degree by selecting
+		 * the correct CRC source before the port is enabled,
+		 * and not touching the CRC source bits again until
+		 * the port is disabled. But even then the bits
+		 * eventually get stuck and a reboot is needed to get
+		 * working CRCs on the pipe again. Let's simply
+		 * refuse to use DP CRCs on g4x.
+		 */
 		return -EINVAL;
 	}
 
-	/*
-	 * When the pipe CRC tap point is after the transcoders we need
-	 * to tweak symbol-level features to produce a deterministic series of
-	 * symbols for a given frame. We need to reset those features only once
-	 * a frame (instead of every nth symbol):
-	 *   - DC-balance: used to ensure a better clock recovery from the data
-	 *     link (SDVO)
-	 *   - DisplayPort scrambling: used for EMI reduction
-	 */
-	if (need_stable_symbols) {
-		u32 tmp = I915_READ(PORT_DFT2_G4X);
-
-		WARN_ON(!IS_G4X(dev_priv));
-
-		I915_WRITE(PORT_DFT_I9XX,
-			   I915_READ(PORT_DFT_I9XX) | DC_BALANCE_RESET);
-
-		if (pipe == PIPE_A)
-			tmp |= PIPE_A_SCRAMBLE_RESET;
-		else
-			tmp |= PIPE_B_SCRAMBLE_RESET;
-
-		I915_WRITE(PORT_DFT2_G4X, tmp);
-	}
-
 	return 0;
 }
 
@@ -283,24 +254,6 @@ static void vlv_undo_pipe_scramble_reset(struct drm_i915_private *dev_priv,
 	if (!(tmp & PIPE_SCRAMBLE_RESET_MASK))
 		tmp &= ~DC_BALANCE_RESET_VLV;
 	I915_WRITE(PORT_DFT2_G4X, tmp);
-
-}
-
-static void g4x_undo_pipe_scramble_reset(struct drm_i915_private *dev_priv,
-					 enum pipe pipe)
-{
-	u32 tmp = I915_READ(PORT_DFT2_G4X);
-
-	if (pipe == PIPE_A)
-		tmp &= ~PIPE_A_SCRAMBLE_RESET;
-	else
-		tmp &= ~PIPE_B_SCRAMBLE_RESET;
-	I915_WRITE(PORT_DFT2_G4X, tmp);
-
-	if (!(tmp & PIPE_SCRAMBLE_RESET_MASK)) {
-		I915_WRITE(PORT_DFT_I9XX,
-			   I915_READ(PORT_DFT_I9XX) & ~DC_BALANCE_RESET);
-	}
 }
 
 static int ilk_pipe_crc_ctl_reg(enum intel_pipe_crc_source *source,
@@ -329,19 +282,18 @@ static int ilk_pipe_crc_ctl_reg(enum intel_pipe_crc_source *source,
 	return 0;
 }
 
-static void hsw_pipe_A_crc_wa(struct drm_i915_private *dev_priv,
-			      bool enable)
+static void
+intel_crtc_crc_setup_workarounds(struct intel_crtc *crtc, bool enable)
 {
-	struct drm_device *dev = &dev_priv->drm;
-	struct intel_crtc *crtc = intel_get_crtc_for_pipe(dev_priv, PIPE_A);
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	struct intel_crtc_state *pipe_config;
 	struct drm_atomic_state *state;
 	struct drm_modeset_acquire_ctx ctx;
-	int ret = 0;
+	int ret;
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-	state = drm_atomic_state_alloc(dev);
+	state = drm_atomic_state_alloc(&dev_priv->drm);
 	if (!state) {
 		ret = -ENOMEM;
 		goto unlock;
@@ -356,17 +308,10 @@ retry:
 		goto put_state;
 	}
 
-	if (HAS_IPS(dev_priv)) {
-		/*
-		 * When IPS gets enabled, the pipe CRC changes. Since IPS gets
-		 * enabled and disabled dynamically based on package C states,
-		 * user space can't make reliable use of the CRCs, so let's just
-		 * completely disable it.
-		 */
-		pipe_config->ips_force_disable = enable;
-	}
+	pipe_config->base.mode_changed = pipe_config->has_psr;
+	pipe_config->crc_enabled = enable;
 
-	if (IS_HASWELL(dev_priv)) {
+	if (IS_HASWELL(dev_priv) && crtc->pipe == PIPE_A) {
 		pipe_config->pch_pfit.force_thru = enable;
 		if (pipe_config->cpu_transcoder == TRANSCODER_EDP &&
 		    pipe_config->pch_pfit.enabled != enable)
@@ -392,11 +337,10 @@ unlock:
 static int ivb_pipe_crc_ctl_reg(struct drm_i915_private *dev_priv,
 				enum pipe pipe,
 				enum intel_pipe_crc_source *source,
-				u32 *val,
-				bool set_wa)
+				u32 *val)
 {
 	if (*source == INTEL_PIPE_CRC_SOURCE_AUTO)
-		*source = INTEL_PIPE_CRC_SOURCE_PF;
+		*source = INTEL_PIPE_CRC_SOURCE_PIPE;
 
 	switch (*source) {
 	case INTEL_PIPE_CRC_SOURCE_PLANE1:
@@ -405,11 +349,7 @@ static int ivb_pipe_crc_ctl_reg(struct drm_i915_private *dev_priv,
 	case INTEL_PIPE_CRC_SOURCE_PLANE2:
 		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_SPRITE_IVB;
 		break;
-	case INTEL_PIPE_CRC_SOURCE_PF:
-		if (set_wa && (IS_HASWELL(dev_priv) ||
-		     IS_BROADWELL(dev_priv)) && pipe == PIPE_A)
-			hsw_pipe_A_crc_wa(dev_priv, true);
-
+	case INTEL_PIPE_CRC_SOURCE_PIPE:
 		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_PF_IVB;
 		break;
 	case INTEL_PIPE_CRC_SOURCE_NONE:
@@ -422,10 +362,52 @@ static int ivb_pipe_crc_ctl_reg(struct drm_i915_private *dev_priv,
 	return 0;
 }
 
+static int skl_pipe_crc_ctl_reg(struct drm_i915_private *dev_priv,
+				enum pipe pipe,
+				enum intel_pipe_crc_source *source,
+				u32 *val)
+{
+	if (*source == INTEL_PIPE_CRC_SOURCE_AUTO)
+		*source = INTEL_PIPE_CRC_SOURCE_PIPE;
+
+	switch (*source) {
+	case INTEL_PIPE_CRC_SOURCE_PLANE1:
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_PLANE_1_SKL;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_PLANE2:
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_PLANE_2_SKL;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_PLANE3:
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_PLANE_3_SKL;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_PLANE4:
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_PLANE_4_SKL;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_PLANE5:
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_PLANE_5_SKL;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_PLANE6:
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_PLANE_6_SKL;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_PLANE7:
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_PLANE_7_SKL;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_PIPE:
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_DMUX_SKL;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_NONE:
+		*val = 0;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 static int get_new_crc_ctl_reg(struct drm_i915_private *dev_priv,
 			       enum pipe pipe,
-			       enum intel_pipe_crc_source *source, u32 *val,
-			       bool set_wa)
+			       enum intel_pipe_crc_source *source, u32 *val)
 {
 	if (IS_GEN(dev_priv, 2))
 		return i8xx_pipe_crc_ctl_reg(source, val);
@@ -435,8 +417,10 @@ static int get_new_crc_ctl_reg(struct drm_i915_private *dev_priv,
 		return vlv_pipe_crc_ctl_reg(dev_priv, pipe, source, val);
 	else if (IS_GEN_RANGE(dev_priv, 5, 6))
 		return ilk_pipe_crc_ctl_reg(source, val);
+	else if (INTEL_GEN(dev_priv) < 9)
+		return ivb_pipe_crc_ctl_reg(dev_priv, pipe, source, val);
 	else
-		return ivb_pipe_crc_ctl_reg(dev_priv, pipe, source, val, set_wa);
+		return skl_pipe_crc_ctl_reg(dev_priv, pipe, source, val);
 }
 
 static int
@@ -486,9 +470,6 @@ static int i9xx_crc_source_valid(struct drm_i915_private *dev_priv,
 	switch (source) {
 	case INTEL_PIPE_CRC_SOURCE_PIPE:
 	case INTEL_PIPE_CRC_SOURCE_TV:
-	case INTEL_PIPE_CRC_SOURCE_DP_B:
-	case INTEL_PIPE_CRC_SOURCE_DP_C:
-	case INTEL_PIPE_CRC_SOURCE_DP_D:
 	case INTEL_PIPE_CRC_SOURCE_NONE:
 		return 0;
 	default:
@@ -532,7 +513,25 @@ static int ivb_crc_source_valid(struct drm_i915_private *dev_priv,
 	case INTEL_PIPE_CRC_SOURCE_PIPE:
 	case INTEL_PIPE_CRC_SOURCE_PLANE1:
 	case INTEL_PIPE_CRC_SOURCE_PLANE2:
-	case INTEL_PIPE_CRC_SOURCE_PF:
+	case INTEL_PIPE_CRC_SOURCE_NONE:
+		return 0;
+	default:
+		return -EINVAL;
+	}
+}
+
+static int skl_crc_source_valid(struct drm_i915_private *dev_priv,
+				const enum intel_pipe_crc_source source)
+{
+	switch (source) {
+	case INTEL_PIPE_CRC_SOURCE_PIPE:
+	case INTEL_PIPE_CRC_SOURCE_PLANE1:
+	case INTEL_PIPE_CRC_SOURCE_PLANE2:
+	case INTEL_PIPE_CRC_SOURCE_PLANE3:
+	case INTEL_PIPE_CRC_SOURCE_PLANE4:
+	case INTEL_PIPE_CRC_SOURCE_PLANE5:
+	case INTEL_PIPE_CRC_SOURCE_PLANE6:
+	case INTEL_PIPE_CRC_SOURCE_PLANE7:
 	case INTEL_PIPE_CRC_SOURCE_NONE:
 		return 0;
 	default:
@@ -552,8 +551,10 @@ intel_is_valid_crc_source(struct drm_i915_private *dev_priv,
 		return vlv_crc_source_valid(dev_priv, source);
 	else if (IS_GEN_RANGE(dev_priv, 5, 6))
 		return ilk_crc_source_valid(dev_priv, source);
-	else
+	else if (INTEL_GEN(dev_priv) < 9)
 		return ivb_crc_source_valid(dev_priv, source);
+	else
+		return skl_crc_source_valid(dev_priv, source);
 }
 
 const char *const *intel_crtc_get_crc_sources(struct drm_crtc *crtc,
@@ -592,6 +593,7 @@ int intel_crtc_set_crc_source(struct drm_crtc *crtc, const char *source_name)
 	intel_wakeref_t wakeref;
 	u32 val = 0; /* shut up gcc */
 	int ret = 0;
+	bool enable;
 
 	if (display_crc_ctl_parse_source(source_name, &source) < 0) {
 		DRM_DEBUG_DRIVER("unknown source %s\n", source_name);
@@ -605,7 +607,11 @@ int intel_crtc_set_crc_source(struct drm_crtc *crtc, const char *source_name)
 		return -EIO;
 	}
 
-	ret = get_new_crc_ctl_reg(dev_priv, crtc->index, &source, &val, true);
+	enable = source != INTEL_PIPE_CRC_SOURCE_NONE;
+	if (enable)
+		intel_crtc_crc_setup_workarounds(to_intel_crtc(crtc), true);
+
+	ret = get_new_crc_ctl_reg(dev_priv, crtc->index, &source, &val);
 	if (ret != 0)
 		goto out;
 
@@ -614,18 +620,16 @@ int intel_crtc_set_crc_source(struct drm_crtc *crtc, const char *source_name)
 	POSTING_READ(PIPE_CRC_CTL(crtc->index));
 
 	if (!source) {
-		if (IS_G4X(dev_priv))
-			g4x_undo_pipe_scramble_reset(dev_priv, crtc->index);
-		else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
+		if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
 			vlv_undo_pipe_scramble_reset(dev_priv, crtc->index);
-		else if ((IS_HASWELL(dev_priv) ||
-			  IS_BROADWELL(dev_priv)) && crtc->index == PIPE_A)
-			hsw_pipe_A_crc_wa(dev_priv, false);
 	}
 
 	pipe_crc->skipped = 0;
 
 out:
+	if (!enable)
+		intel_crtc_crc_setup_workarounds(to_intel_crtc(crtc), false);
+
 	intel_display_power_put(dev_priv, power_domain, wakeref);
 
 	return ret;
@@ -641,7 +645,7 @@ void intel_crtc_enable_pipe_crc(struct intel_crtc *intel_crtc)
 	if (!crtc->crc.opened)
 		return;
 
-	if (get_new_crc_ctl_reg(dev_priv, crtc->index, &pipe_crc->source, &val, false) < 0)
+	if (get_new_crc_ctl_reg(dev_priv, crtc->index, &pipe_crc->source, &val) < 0)
 		return;
 
 	/* Don't need pipe_crc->lock here, IRQs are not generated. */
diff --git a/drivers/gpu/drm/i915/intel_pipe_crc.h b/drivers/gpu/drm/i915/intel_pipe_crc.h
new file mode 100644
index 000000000000..81eaf1854788
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_pipe_crc.h
@@ -0,0 +1,35 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_PIPE_CRC_H__
+#define __INTEL_PIPE_CRC_H__
+
+#include <linux/types.h>
+
+struct drm_crtc;
+struct intel_crtc;
+
+#ifdef CONFIG_DEBUG_FS
+int intel_crtc_set_crc_source(struct drm_crtc *crtc, const char *source_name);
+int intel_crtc_verify_crc_source(struct drm_crtc *crtc,
+				 const char *source_name, size_t *values_cnt);
+const char *const *intel_crtc_get_crc_sources(struct drm_crtc *crtc,
+					      size_t *count);
+void intel_crtc_disable_pipe_crc(struct intel_crtc *crtc);
+void intel_crtc_enable_pipe_crc(struct intel_crtc *crtc);
+#else
+#define intel_crtc_set_crc_source NULL
+#define intel_crtc_verify_crc_source NULL
+#define intel_crtc_get_crc_sources NULL
+static inline void intel_crtc_disable_pipe_crc(struct intel_crtc *crtc)
+{
+}
+
+static inline void intel_crtc_enable_pipe_crc(struct intel_crtc *crtc)
+{
+}
+#endif
+
+#endif /* __INTEL_PIPE_CRC_H__ */
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 54307f1df6cf..44be676fabd6 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -35,6 +35,9 @@
 
 #include "i915_drv.h"
 #include "intel_drv.h"
+#include "intel_fbc.h"
+#include "intel_pm.h"
+#include "intel_sprite.h"
 #include "../../../platform/x86/intel_ips.h"
 
 /**
@@ -338,12 +341,12 @@ static void chv_set_memory_pm5(struct drm_i915_private *dev_priv, bool enable)
 
 	mutex_lock(&dev_priv->pcu_lock);
 
-	val = vlv_punit_read(dev_priv, PUNIT_REG_DSPFREQ);
+	val = vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM);
 	if (enable)
 		val |= DSP_MAXFIFO_PM5_ENABLE;
 	else
 		val &= ~DSP_MAXFIFO_PM5_ENABLE;
-	vlv_punit_write(dev_priv, PUNIT_REG_DSPFREQ, val);
+	vlv_punit_write(dev_priv, PUNIT_REG_DSPSSPM, val);
 
 	mutex_unlock(&dev_priv->pcu_lock);
 }
@@ -850,7 +853,7 @@ static void pineview_update_wm(struct intel_crtc *unused_crtc)
 	u32 reg;
 	unsigned int wm;
 
-	latency = intel_get_cxsr_latency(IS_PINEVIEW_G(dev_priv),
+	latency = intel_get_cxsr_latency(!IS_MOBILE(dev_priv),
 					 dev_priv->is_ddr3,
 					 dev_priv->fsb_freq,
 					 dev_priv->mem_freq);
@@ -3624,7 +3627,12 @@ static u8 intel_enabled_dbuf_slices_num(struct drm_i915_private *dev_priv)
 	if (INTEL_GEN(dev_priv) < 11)
 		return enabled_slices;
 
-	if (I915_READ(DBUF_CTL_S2) & DBUF_POWER_STATE)
+	/*
+	 * FIXME: for now we'll only ever use 1 slice; pretend that we have
+	 * only that 1 slice enabled until we have a proper way for on-demand
+	 * toggling of the second slice.
+	 */
+	if (0 && I915_READ(DBUF_CTL_S2) & DBUF_POWER_STATE)
 		enabled_slices++;
 
 	return enabled_slices;
@@ -3919,12 +3927,43 @@ skl_ddb_get_pipe_allocation_limits(struct drm_i915_private *dev_priv,
 	alloc->end = ddb_size * (width_before_pipe + pipe_width) / total_width;
 }
 
-static unsigned int skl_cursor_allocation(int num_active)
+static int skl_compute_wm_params(const struct intel_crtc_state *crtc_state,
+				 int width, const struct drm_format_info *format,
+				 u64 modifier, unsigned int rotation,
+				 u32 plane_pixel_rate, struct skl_wm_params *wp,
+				 int color_plane);
+static void skl_compute_plane_wm(const struct intel_crtc_state *cstate,
+				 int level,
+				 const struct skl_wm_params *wp,
+				 const struct skl_wm_level *result_prev,
+				 struct skl_wm_level *result /* out */);
+
+static unsigned int
+skl_cursor_allocation(const struct intel_crtc_state *crtc_state,
+		      int num_active)
 {
-	if (num_active == 1)
-		return 32;
+	struct drm_i915_private *dev_priv = to_i915(crtc_state->base.crtc->dev);
+	int level, max_level = ilk_wm_max_level(dev_priv);
+	struct skl_wm_level wm = {};
+	int ret, min_ddb_alloc = 0;
+	struct skl_wm_params wp;
+
+	ret = skl_compute_wm_params(crtc_state, 256,
+				    drm_format_info(DRM_FORMAT_ARGB8888),
+				    DRM_FORMAT_MOD_LINEAR,
+				    DRM_MODE_ROTATE_0,
+				    crtc_state->pixel_rate, &wp, 0);
+	WARN_ON(ret);
 
-	return 8;
+	for (level = 0; level <= max_level; level++) {
+		skl_compute_plane_wm(crtc_state, level, &wp, &wm, &wm);
+		if (wm.min_ddb_alloc == U16_MAX)
+			break;
+
+		min_ddb_alloc = wm.min_ddb_alloc;
+	}
+
+	return max(num_active == 1 ? 32 : 8, min_ddb_alloc);
 }
 
 static void skl_ddb_entry_init_from_hw(struct drm_i915_private *dev_priv,
@@ -3970,7 +4009,7 @@ skl_ddb_get_hw_plane_state(struct drm_i915_private *dev_priv,
 		val = I915_READ(PLANE_BUF_CFG(pipe, plane_id));
 		val2 = I915_READ(PLANE_NV12_BUF_CFG(pipe, plane_id));
 
-		if (fourcc == DRM_FORMAT_NV12)
+		if (is_planar_yuv_format(fourcc))
 			swap(val, val2);
 
 		skl_ddb_entry_init_from_hw(dev_priv, ddb_y, val);
@@ -4180,7 +4219,7 @@ skl_plane_relative_data_rate(const struct intel_crtc_state *cstate,
 
 	if (intel_plane->id == PLANE_CURSOR)
 		return 0;
-	if (plane == 1 && format != DRM_FORMAT_NV12)
+	if (plane == 1 && !is_planar_yuv_format(format))
 		return 0;
 
 	/*
@@ -4192,7 +4231,7 @@ skl_plane_relative_data_rate(const struct intel_crtc_state *cstate,
 	height = drm_rect_height(&intel_pstate->base.src) >> 16;
 
 	/* UV plane does 1/2 pixel sub-sampling */
-	if (plane == 1 && format == DRM_FORMAT_NV12) {
+	if (plane == 1 && is_planar_yuv_format(format)) {
 		width /= 2;
 		height /= 2;
 	}
@@ -4308,7 +4347,6 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *cstate,
 	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct skl_ddb_entry *alloc = &cstate->wm.skl.ddb;
-	struct skl_plane_wm *wm;
 	u16 alloc_size, start = 0;
 	u16 total[I915_MAX_PLANES] = {};
 	u16 uv_total[I915_MAX_PLANES] = {};
@@ -4349,7 +4387,7 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *cstate,
 		return 0;
 
 	/* Allocate fixed number of blocks for cursor. */
-	total[PLANE_CURSOR] = skl_cursor_allocation(num_active);
+	total[PLANE_CURSOR] = skl_cursor_allocation(cstate, num_active);
 	alloc_size -= total[PLANE_CURSOR];
 	cstate->wm.skl.plane_ddb_y[PLANE_CURSOR].start =
 		alloc->end - total[PLANE_CURSOR];
@@ -4365,15 +4403,23 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *cstate,
 	for (level = ilk_wm_max_level(dev_priv); level >= 0; level--) {
 		blocks = 0;
 		for_each_plane_id_on_crtc(intel_crtc, plane_id) {
-			if (plane_id == PLANE_CURSOR)
+			const struct skl_plane_wm *wm =
+				&cstate->wm.skl.optimal.planes[plane_id];
+
+			if (plane_id == PLANE_CURSOR) {
+				if (WARN_ON(wm->wm[level].min_ddb_alloc >
+					    total[PLANE_CURSOR])) {
+					blocks = U32_MAX;
+					break;
+				}
 				continue;
+			}
 
-			wm = &cstate->wm.skl.optimal.planes[plane_id];
 			blocks += wm->wm[level].min_ddb_alloc;
 			blocks += wm->uv_wm[level].min_ddb_alloc;
 		}
 
-		if (blocks < alloc_size) {
+		if (blocks <= alloc_size) {
 			alloc_size -= blocks;
 			break;
 		}
@@ -4392,6 +4438,8 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *cstate,
 	 * proportional to its relative data rate.
 	 */
 	for_each_plane_id_on_crtc(intel_crtc, plane_id) {
+		const struct skl_plane_wm *wm =
+			&cstate->wm.skl.optimal.planes[plane_id];
 		u64 rate;
 		u16 extra;
 
@@ -4405,8 +4453,6 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *cstate,
 		if (total_data_rate == 0)
 			break;
 
-		wm = &cstate->wm.skl.optimal.planes[plane_id];
-
 		rate = plane_data_rate[plane_id];
 		extra = min_t(u16, alloc_size,
 			      DIV64_U64_ROUND_UP(alloc_size * rate,
@@ -4431,14 +4477,14 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *cstate,
 	/* Set the actual DDB start/end points for each plane */
 	start = alloc->start;
 	for_each_plane_id_on_crtc(intel_crtc, plane_id) {
-		struct skl_ddb_entry *plane_alloc, *uv_plane_alloc;
+		struct skl_ddb_entry *plane_alloc =
+			&cstate->wm.skl.plane_ddb_y[plane_id];
+		struct skl_ddb_entry *uv_plane_alloc =
+			&cstate->wm.skl.plane_ddb_uv[plane_id];
 
 		if (plane_id == PLANE_CURSOR)
 			continue;
 
-		plane_alloc = &cstate->wm.skl.plane_ddb_y[plane_id];
-		uv_plane_alloc = &cstate->wm.skl.plane_ddb_uv[plane_id];
-
 		/* Gen11+ uses a separate plane for UV watermarks */
 		WARN_ON(INTEL_GEN(dev_priv) >= 11 && uv_total[plane_id]);
 
@@ -4464,8 +4510,35 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *cstate,
 	 */
 	for (level++; level <= ilk_wm_max_level(dev_priv); level++) {
 		for_each_plane_id_on_crtc(intel_crtc, plane_id) {
-			wm = &cstate->wm.skl.optimal.planes[plane_id];
-			memset(&wm->wm[level], 0, sizeof(wm->wm[level]));
+			struct skl_plane_wm *wm =
+				&cstate->wm.skl.optimal.planes[plane_id];
+
+			/*
+			 * We only disable the watermarks for each plane if
+			 * they exceed the ddb allocation of said plane. This
+			 * is done so that we don't end up touching cursor
+			 * watermarks needlessly when some other plane reduces
+			 * our max possible watermark level.
+			 *
+			 * Bspec has this to say about the PLANE_WM enable bit:
+			 * "All the watermarks at this level for all enabled
+			 *  planes must be enabled before the level will be used."
+			 * So this is actually safe to do.
+			 */
+			if (wm->wm[level].min_ddb_alloc > total[plane_id] ||
+			    wm->uv_wm[level].min_ddb_alloc > uv_total[plane_id])
+				memset(&wm->wm[level], 0, sizeof(wm->wm[level]));
+
+			/*
+			 * Wa_1408961008:icl, ehl
+			 * Underruns with WM1+ disabled
+			 */
+			if (IS_GEN(dev_priv, 11) &&
+			    level == 1 && wm->wm[0].plane_en) {
+				wm->wm[level].plane_res_b = wm->wm[0].plane_res_b;
+				wm->wm[level].plane_res_l = wm->wm[0].plane_res_l;
+				wm->wm[level].ignore_lines = wm->wm[0].ignore_lines;
+			}
 		}
 	}
 
@@ -4474,7 +4547,9 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *cstate,
 	 * don't have enough DDB blocks for it.
 	 */
 	for_each_plane_id_on_crtc(intel_crtc, plane_id) {
-		wm = &cstate->wm.skl.optimal.planes[plane_id];
+		struct skl_plane_wm *wm =
+			&cstate->wm.skl.optimal.planes[plane_id];
+
 		if (wm->trans_wm.plane_res_b >= total[plane_id])
 			memset(&wm->trans_wm, 0, sizeof(wm->trans_wm));
 	}
@@ -4568,57 +4643,45 @@ skl_adjusted_plane_pixel_rate(const struct intel_crtc_state *cstate,
 }
 
 static int
-skl_compute_plane_wm_params(const struct intel_crtc_state *cstate,
-			    const struct intel_plane_state *intel_pstate,
-			    struct skl_wm_params *wp, int color_plane)
+skl_compute_wm_params(const struct intel_crtc_state *crtc_state,
+		      int width, const struct drm_format_info *format,
+		      u64 modifier, unsigned int rotation,
+		      u32 plane_pixel_rate, struct skl_wm_params *wp,
+		      int color_plane)
 {
-	struct intel_plane *plane = to_intel_plane(intel_pstate->base.plane);
-	struct drm_i915_private *dev_priv = to_i915(plane->base.dev);
-	const struct drm_plane_state *pstate = &intel_pstate->base;
-	const struct drm_framebuffer *fb = pstate->fb;
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc);
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	u32 interm_pbpl;
 
-	/* only NV12 format has two planes */
-	if (color_plane == 1 && fb->format->format != DRM_FORMAT_NV12) {
-		DRM_DEBUG_KMS("Non NV12 format have single plane\n");
+	/* only planar format has two planes */
+	if (color_plane == 1 && !is_planar_yuv_format(format->format)) {
+		DRM_DEBUG_KMS("Non planar format have single plane\n");
 		return -EINVAL;
 	}
 
-	wp->y_tiled = fb->modifier == I915_FORMAT_MOD_Y_TILED ||
-		      fb->modifier == I915_FORMAT_MOD_Yf_TILED ||
-		      fb->modifier == I915_FORMAT_MOD_Y_TILED_CCS ||
-		      fb->modifier == I915_FORMAT_MOD_Yf_TILED_CCS;
-	wp->x_tiled = fb->modifier == I915_FORMAT_MOD_X_TILED;
-	wp->rc_surface = fb->modifier == I915_FORMAT_MOD_Y_TILED_CCS ||
-			 fb->modifier == I915_FORMAT_MOD_Yf_TILED_CCS;
-	wp->is_planar = fb->format->format == DRM_FORMAT_NV12;
-
-	if (plane->id == PLANE_CURSOR) {
-		wp->width = intel_pstate->base.crtc_w;
-	} else {
-		/*
-		 * Src coordinates are already rotated by 270 degrees for
-		 * the 90/270 degree plane rotation cases (to match the
-		 * GTT mapping), hence no need to account for rotation here.
-		 */
-		wp->width = drm_rect_width(&intel_pstate->base.src) >> 16;
-	}
+	wp->y_tiled = modifier == I915_FORMAT_MOD_Y_TILED ||
+		      modifier == I915_FORMAT_MOD_Yf_TILED ||
+		      modifier == I915_FORMAT_MOD_Y_TILED_CCS ||
+		      modifier == I915_FORMAT_MOD_Yf_TILED_CCS;
+	wp->x_tiled = modifier == I915_FORMAT_MOD_X_TILED;
+	wp->rc_surface = modifier == I915_FORMAT_MOD_Y_TILED_CCS ||
+			 modifier == I915_FORMAT_MOD_Yf_TILED_CCS;
+	wp->is_planar = is_planar_yuv_format(format->format);
 
+	wp->width = width;
 	if (color_plane == 1 && wp->is_planar)
 		wp->width /= 2;
 
-	wp->cpp = fb->format->cpp[color_plane];
-	wp->plane_pixel_rate = skl_adjusted_plane_pixel_rate(cstate,
-							     intel_pstate);
+	wp->cpp = format->cpp[color_plane];
+	wp->plane_pixel_rate = plane_pixel_rate;
 
 	if (INTEL_GEN(dev_priv) >= 11 &&
-	    fb->modifier == I915_FORMAT_MOD_Yf_TILED && wp->cpp == 1)
+	    modifier == I915_FORMAT_MOD_Yf_TILED  && wp->cpp == 1)
 		wp->dbuf_block_size = 256;
 	else
 		wp->dbuf_block_size = 512;
 
-	if (drm_rotation_90_or_270(pstate->rotation)) {
-
+	if (drm_rotation_90_or_270(rotation)) {
 		switch (wp->cpp) {
 		case 1:
 			wp->y_min_scanlines = 16;
@@ -4663,12 +4726,40 @@ skl_compute_plane_wm_params(const struct intel_crtc_state *cstate,
 
 	wp->y_tile_minimum = mul_u32_fixed16(wp->y_min_scanlines,
 					     wp->plane_blocks_per_line);
+
 	wp->linetime_us = fixed16_to_u32_round_up(
-					intel_get_linetime_us(cstate));
+					intel_get_linetime_us(crtc_state));
 
 	return 0;
 }
 
+static int
+skl_compute_plane_wm_params(const struct intel_crtc_state *crtc_state,
+			    const struct intel_plane_state *plane_state,
+			    struct skl_wm_params *wp, int color_plane)
+{
+	struct intel_plane *plane = to_intel_plane(plane_state->base.plane);
+	const struct drm_framebuffer *fb = plane_state->base.fb;
+	int width;
+
+	if (plane->id == PLANE_CURSOR) {
+		width = plane_state->base.crtc_w;
+	} else {
+		/*
+		 * Src coordinates are already rotated by 270 degrees for
+		 * the 90/270 degree plane rotation cases (to match the
+		 * GTT mapping), hence no need to account for rotation here.
+		 */
+		width = drm_rect_width(&plane_state->base.src) >> 16;
+	}
+
+	return skl_compute_wm_params(crtc_state, width,
+				     fb->format, fb->modifier,
+				     plane_state->base.rotation,
+				     skl_adjusted_plane_pixel_rate(crtc_state, plane_state),
+				     wp, color_plane);
+}
+
 static bool skl_wm_has_lines(struct drm_i915_private *dev_priv, int level)
 {
 	if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv))
@@ -4679,14 +4770,12 @@ static bool skl_wm_has_lines(struct drm_i915_private *dev_priv, int level)
 }
 
 static void skl_compute_plane_wm(const struct intel_crtc_state *cstate,
-				 const struct intel_plane_state *intel_pstate,
 				 int level,
 				 const struct skl_wm_params *wp,
 				 const struct skl_wm_level *result_prev,
 				 struct skl_wm_level *result /* out */)
 {
-	struct drm_i915_private *dev_priv =
-		to_i915(intel_pstate->base.plane->dev);
+	struct drm_i915_private *dev_priv = to_i915(cstate->base.crtc->dev);
 	u32 latency = dev_priv->wm.skl_latency[level];
 	uint_fixed_16_16_t method1, method2;
 	uint_fixed_16_16_t selected_result;
@@ -4805,19 +4894,17 @@ static void skl_compute_plane_wm(const struct intel_crtc_state *cstate,
 
 static void
 skl_compute_wm_levels(const struct intel_crtc_state *cstate,
-		      const struct intel_plane_state *intel_pstate,
 		      const struct skl_wm_params *wm_params,
 		      struct skl_wm_level *levels)
 {
-	struct drm_i915_private *dev_priv =
-		to_i915(intel_pstate->base.plane->dev);
+	struct drm_i915_private *dev_priv = to_i915(cstate->base.crtc->dev);
 	int level, max_level = ilk_wm_max_level(dev_priv);
 	struct skl_wm_level *result_prev = &levels[0];
 
 	for (level = 0; level <= max_level; level++) {
 		struct skl_wm_level *result = &levels[level];
 
-		skl_compute_plane_wm(cstate, intel_pstate, level, wm_params,
+		skl_compute_plane_wm(cstate, level, wm_params,
 				     result_prev, result);
 
 		result_prev = result;
@@ -4914,7 +5001,7 @@ static int skl_build_plane_wm_single(struct intel_crtc_state *crtc_state,
 	if (ret)
 		return ret;
 
-	skl_compute_wm_levels(crtc_state, plane_state, &wm_params, wm->wm);
+	skl_compute_wm_levels(crtc_state, &wm_params, wm->wm);
 	skl_compute_transition_wm(crtc_state, &wm_params, wm);
 
 	return 0;
@@ -4936,13 +5023,12 @@ static int skl_build_plane_wm_uv(struct intel_crtc_state *crtc_state,
 	if (ret)
 		return ret;
 
-	skl_compute_wm_levels(crtc_state, plane_state, &wm_params, wm->uv_wm);
+	skl_compute_wm_levels(crtc_state, &wm_params, wm->uv_wm);
 
 	return 0;
 }
 
-static int skl_build_plane_wm(struct skl_pipe_wm *pipe_wm,
-			      struct intel_crtc_state *crtc_state,
+static int skl_build_plane_wm(struct intel_crtc_state *crtc_state,
 			      const struct intel_plane_state *plane_state)
 {
 	struct intel_plane *plane = to_intel_plane(plane_state->base.plane);
@@ -4968,8 +5054,7 @@ static int skl_build_plane_wm(struct skl_pipe_wm *pipe_wm,
 	return 0;
 }
 
-static int icl_build_plane_wm(struct skl_pipe_wm *pipe_wm,
-			      struct intel_crtc_state *crtc_state,
+static int icl_build_plane_wm(struct intel_crtc_state *crtc_state,
 			      const struct intel_plane_state *plane_state)
 {
 	enum plane_id plane_id = to_intel_plane(plane_state->base.plane)->id;
@@ -5006,10 +5091,10 @@ static int icl_build_plane_wm(struct skl_pipe_wm *pipe_wm,
 	return 0;
 }
 
-static int skl_build_pipe_wm(struct intel_crtc_state *cstate,
-			     struct skl_pipe_wm *pipe_wm)
+static int skl_build_pipe_wm(struct intel_crtc_state *cstate)
 {
 	struct drm_i915_private *dev_priv = to_i915(cstate->base.crtc->dev);
+	struct skl_pipe_wm *pipe_wm = &cstate->wm.skl.optimal;
 	struct drm_crtc_state *crtc_state = &cstate->base;
 	struct drm_plane *plane;
 	const struct drm_plane_state *pstate;
@@ -5026,11 +5111,9 @@ static int skl_build_pipe_wm(struct intel_crtc_state *cstate,
 						to_intel_plane_state(pstate);
 
 		if (INTEL_GEN(dev_priv) >= 11)
-			ret = icl_build_plane_wm(pipe_wm,
-						 cstate, intel_pstate);
+			ret = icl_build_plane_wm(cstate, intel_pstate);
 		else
-			ret = skl_build_plane_wm(pipe_wm,
-						 cstate, intel_pstate);
+			ret = skl_build_plane_wm(cstate, intel_pstate);
 		if (ret)
 			return ret;
 	}
@@ -5056,11 +5139,12 @@ static void skl_write_wm_level(struct drm_i915_private *dev_priv,
 {
 	u32 val = 0;
 
-	if (level->plane_en) {
+	if (level->plane_en)
 		val |= PLANE_WM_EN;
-		val |= level->plane_res_b;
-		val |= level->plane_res_l << PLANE_WM_LINES_SHIFT;
-	}
+	if (level->ignore_lines)
+		val |= PLANE_WM_IGNORE_LINES;
+	val |= level->plane_res_b;
+	val |= level->plane_res_l << PLANE_WM_LINES_SHIFT;
 
 	I915_WRITE_FW(reg, val);
 }
@@ -5126,6 +5210,7 @@ bool skl_wm_level_equals(const struct skl_wm_level *l1,
 			 const struct skl_wm_level *l2)
 {
 	return l1->plane_en == l2->plane_en &&
+		l1->ignore_lines == l2->ignore_lines &&
 		l1->plane_res_l == l2->plane_res_l &&
 		l1->plane_res_b == l2->plane_res_b;
 }
@@ -5169,7 +5254,7 @@ static inline bool skl_ddb_entries_overlap(const struct skl_ddb_entry *a,
 }
 
 bool skl_ddb_allocation_overlaps(const struct skl_ddb_entry *ddb,
-				 const struct skl_ddb_entry entries[],
+				 const struct skl_ddb_entry *entries,
 				 int num_entries, int ignore_idx)
 {
 	int i;
@@ -5183,23 +5268,6 @@ bool skl_ddb_allocation_overlaps(const struct skl_ddb_entry *ddb,
 	return false;
 }
 
-static int skl_update_pipe_wm(struct intel_crtc_state *cstate,
-			      const struct skl_pipe_wm *old_pipe_wm,
-			      struct skl_pipe_wm *pipe_wm, /* out */
-			      bool *changed /* out */)
-{
-	struct intel_crtc *crtc = to_intel_crtc(cstate->base.crtc);
-	int ret;
-
-	ret = skl_build_pipe_wm(cstate, pipe_wm);
-	if (ret)
-		return ret;
-
-	*changed = !skl_pipe_wm_equals(crtc, old_pipe_wm, pipe_wm);
-
-	return 0;
-}
-
 static u32
 pipes_modified(struct intel_atomic_state *state)
 {
@@ -5269,6 +5337,11 @@ skl_compute_ddb(struct intel_atomic_state *state)
 	return 0;
 }
 
+static char enast(bool enable)
+{
+	return enable ? '*' : ' ';
+}
+
 static void
 skl_print_wm_changes(struct intel_atomic_state *state)
 {
@@ -5279,8 +5352,16 @@ skl_print_wm_changes(struct intel_atomic_state *state)
 	struct intel_crtc *crtc;
 	int i;
 
+	if ((drm_debug & DRM_UT_KMS) == 0)
+		return;
+
 	for_each_oldnew_intel_crtc_in_state(state, crtc, old_crtc_state,
 					    new_crtc_state, i) {
+		const struct skl_pipe_wm *old_pipe_wm, *new_pipe_wm;
+
+		old_pipe_wm = &old_crtc_state->wm.skl.optimal;
+		new_pipe_wm = &new_crtc_state->wm.skl.optimal;
+
 		for_each_intel_plane_on_crtc(&dev_priv->drm, crtc, plane) {
 			enum plane_id plane_id = plane->id;
 			const struct skl_ddb_entry *old, *new;
@@ -5291,10 +5372,86 @@ skl_print_wm_changes(struct intel_atomic_state *state)
 			if (skl_ddb_entry_equal(old, new))
 				continue;
 
-			DRM_DEBUG_KMS("[PLANE:%d:%s] ddb (%d - %d) -> (%d - %d)\n",
+			DRM_DEBUG_KMS("[PLANE:%d:%s] ddb (%4d - %4d) -> (%4d - %4d), size %4d -> %4d\n",
+				      plane->base.base.id, plane->base.name,
+				      old->start, old->end, new->start, new->end,
+				      skl_ddb_entry_size(old), skl_ddb_entry_size(new));
+		}
+
+		for_each_intel_plane_on_crtc(&dev_priv->drm, crtc, plane) {
+			enum plane_id plane_id = plane->id;
+			const struct skl_plane_wm *old_wm, *new_wm;
+
+			old_wm = &old_pipe_wm->planes[plane_id];
+			new_wm = &new_pipe_wm->planes[plane_id];
+
+			if (skl_plane_wm_equals(dev_priv, old_wm, new_wm))
+				continue;
+
+			DRM_DEBUG_KMS("[PLANE:%d:%s]   level %cwm0,%cwm1,%cwm2,%cwm3,%cwm4,%cwm5,%cwm6,%cwm7,%ctwm"
+				      " -> %cwm0,%cwm1,%cwm2,%cwm3,%cwm4,%cwm5,%cwm6,%cwm7,%ctwm\n",
+				      plane->base.base.id, plane->base.name,
+				      enast(old_wm->wm[0].plane_en), enast(old_wm->wm[1].plane_en),
+				      enast(old_wm->wm[2].plane_en), enast(old_wm->wm[3].plane_en),
+				      enast(old_wm->wm[4].plane_en), enast(old_wm->wm[5].plane_en),
+				      enast(old_wm->wm[6].plane_en), enast(old_wm->wm[7].plane_en),
+				      enast(old_wm->trans_wm.plane_en),
+				      enast(new_wm->wm[0].plane_en), enast(new_wm->wm[1].plane_en),
+				      enast(new_wm->wm[2].plane_en), enast(new_wm->wm[3].plane_en),
+				      enast(new_wm->wm[4].plane_en), enast(new_wm->wm[5].plane_en),
+				      enast(new_wm->wm[6].plane_en), enast(new_wm->wm[7].plane_en),
+				      enast(new_wm->trans_wm.plane_en));
+
+			DRM_DEBUG_KMS("[PLANE:%d:%s]   lines %c%3d,%c%3d,%c%3d,%c%3d,%c%3d,%c%3d,%c%3d,%c%3d,%c%3d"
+				      " -> %c%3d,%c%3d,%c%3d,%c%3d,%c%3d,%c%3d,%c%3d,%c%3d,%c%3d\n",
 				      plane->base.base.id, plane->base.name,
-				      old->start, old->end,
-				      new->start, new->end);
+				      enast(old_wm->wm[0].ignore_lines), old_wm->wm[0].plane_res_l,
+				      enast(old_wm->wm[1].ignore_lines), old_wm->wm[1].plane_res_l,
+				      enast(old_wm->wm[2].ignore_lines), old_wm->wm[2].plane_res_l,
+				      enast(old_wm->wm[3].ignore_lines), old_wm->wm[3].plane_res_l,
+				      enast(old_wm->wm[4].ignore_lines), old_wm->wm[4].plane_res_l,
+				      enast(old_wm->wm[5].ignore_lines), old_wm->wm[5].plane_res_l,
+				      enast(old_wm->wm[6].ignore_lines), old_wm->wm[6].plane_res_l,
+				      enast(old_wm->wm[7].ignore_lines), old_wm->wm[7].plane_res_l,
+				      enast(old_wm->trans_wm.ignore_lines), old_wm->trans_wm.plane_res_l,
+
+				      enast(new_wm->wm[0].ignore_lines), new_wm->wm[0].plane_res_l,
+				      enast(new_wm->wm[1].ignore_lines), new_wm->wm[1].plane_res_l,
+				      enast(new_wm->wm[2].ignore_lines), new_wm->wm[2].plane_res_l,
+				      enast(new_wm->wm[3].ignore_lines), new_wm->wm[3].plane_res_l,
+				      enast(new_wm->wm[4].ignore_lines), new_wm->wm[4].plane_res_l,
+				      enast(new_wm->wm[5].ignore_lines), new_wm->wm[5].plane_res_l,
+				      enast(new_wm->wm[6].ignore_lines), new_wm->wm[6].plane_res_l,
+				      enast(new_wm->wm[7].ignore_lines), new_wm->wm[7].plane_res_l,
+				      enast(new_wm->trans_wm.ignore_lines), new_wm->trans_wm.plane_res_l);
+
+			DRM_DEBUG_KMS("[PLANE:%d:%s]  blocks %4d,%4d,%4d,%4d,%4d,%4d,%4d,%4d,%4d"
+				      " -> %4d,%4d,%4d,%4d,%4d,%4d,%4d,%4d,%4d\n",
+				      plane->base.base.id, plane->base.name,
+				      old_wm->wm[0].plane_res_b, old_wm->wm[1].plane_res_b,
+				      old_wm->wm[2].plane_res_b, old_wm->wm[3].plane_res_b,
+				      old_wm->wm[4].plane_res_b, old_wm->wm[5].plane_res_b,
+				      old_wm->wm[6].plane_res_b, old_wm->wm[7].plane_res_b,
+				      old_wm->trans_wm.plane_res_b,
+				      new_wm->wm[0].plane_res_b, new_wm->wm[1].plane_res_b,
+				      new_wm->wm[2].plane_res_b, new_wm->wm[3].plane_res_b,
+				      new_wm->wm[4].plane_res_b, new_wm->wm[5].plane_res_b,
+				      new_wm->wm[6].plane_res_b, new_wm->wm[7].plane_res_b,
+				      new_wm->trans_wm.plane_res_b);
+
+			DRM_DEBUG_KMS("[PLANE:%d:%s] min_ddb %4d,%4d,%4d,%4d,%4d,%4d,%4d,%4d,%4d"
+				      " -> %4d,%4d,%4d,%4d,%4d,%4d,%4d,%4d,%4d\n",
+				      plane->base.base.id, plane->base.name,
+				      old_wm->wm[0].min_ddb_alloc, old_wm->wm[1].min_ddb_alloc,
+				      old_wm->wm[2].min_ddb_alloc, old_wm->wm[3].min_ddb_alloc,
+				      old_wm->wm[4].min_ddb_alloc, old_wm->wm[5].min_ddb_alloc,
+				      old_wm->wm[6].min_ddb_alloc, old_wm->wm[7].min_ddb_alloc,
+				      old_wm->trans_wm.min_ddb_alloc,
+				      new_wm->wm[0].min_ddb_alloc, new_wm->wm[1].min_ddb_alloc,
+				      new_wm->wm[2].min_ddb_alloc, new_wm->wm[3].min_ddb_alloc,
+				      new_wm->wm[4].min_ddb_alloc, new_wm->wm[5].min_ddb_alloc,
+				      new_wm->wm[6].min_ddb_alloc, new_wm->wm[7].min_ddb_alloc,
+				      new_wm->trans_wm.min_ddb_alloc);
 		}
 	}
 }
@@ -5449,10 +5606,9 @@ static int
 skl_compute_wm(struct intel_atomic_state *state)
 {
 	struct intel_crtc *crtc;
-	struct intel_crtc_state *cstate;
+	struct intel_crtc_state *new_crtc_state;
 	struct intel_crtc_state *old_crtc_state;
 	struct skl_ddb_values *results = &state->wm_results;
-	struct skl_pipe_wm *pipe_wm;
 	bool changed = false;
 	int ret, i;
 
@@ -5470,12 +5626,8 @@ skl_compute_wm(struct intel_atomic_state *state)
 	 * pipe allocations had to change.
 	 */
 	for_each_oldnew_intel_crtc_in_state(state, crtc, old_crtc_state,
-					    cstate, i) {
-		const struct skl_pipe_wm *old_pipe_wm =
-			&old_crtc_state->wm.skl.optimal;
-
-		pipe_wm = &cstate->wm.skl.optimal;
-		ret = skl_update_pipe_wm(cstate, old_pipe_wm, pipe_wm, &changed);
+					    new_crtc_state, i) {
+		ret = skl_build_pipe_wm(new_crtc_state);
 		if (ret)
 			return ret;
 
@@ -5483,7 +5635,9 @@ skl_compute_wm(struct intel_atomic_state *state)
 		if (ret)
 			return ret;
 
-		if (changed)
+		if (!skl_pipe_wm_equals(crtc,
+					&old_crtc_state->wm.skl.optimal,
+					&new_crtc_state->wm.skl.optimal))
 			results->dirty_pipes |= drm_crtc_mask(&crtc->base);
 	}
 
@@ -5609,6 +5763,7 @@ static inline void skl_wm_level_from_reg_val(u32 val,
 					     struct skl_wm_level *level)
 {
 	level->plane_en = val & PLANE_WM_EN;
+	level->ignore_lines = val & PLANE_WM_IGNORE_LINES;
 	level->plane_res_b = val & PLANE_WM_BLOCKS_MASK;
 	level->plane_res_l = (val >> PLANE_WM_LINES_SHIFT) &
 		PLANE_WM_LINES_MASK;
@@ -5986,7 +6141,7 @@ void vlv_wm_get_hw_state(struct drm_i915_private *dev_priv)
 	if (IS_CHERRYVIEW(dev_priv)) {
 		mutex_lock(&dev_priv->pcu_lock);
 
-		val = vlv_punit_read(dev_priv, PUNIT_REG_DSPFREQ);
+		val = vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM);
 		if (val & DSP_MAXFIFO_PM5_ENABLE)
 			wm->level = VLV_WM_LEVEL_PM5;
 
@@ -6451,7 +6606,7 @@ static void rps_set_power(struct drm_i915_private *dev_priv, int new_power)
 				       ei_down * threshold_down / 100));
 
 	I915_WRITE(GEN6_RP_CONTROL,
-		   GEN6_RP_MEDIA_TURBO |
+		   (INTEL_GEN(dev_priv) > 9 ? 0 : GEN6_RP_MEDIA_TURBO) |
 		   GEN6_RP_MEDIA_HW_NORMAL_MODE |
 		   GEN6_RP_MEDIA_IS_GFX |
 		   GEN6_RP_ENABLE |
@@ -6629,9 +6784,9 @@ static void vlv_set_rps_idle(struct drm_i915_private *dev_priv)
 	 * punit into committing the voltage change) as that takes a lot less
 	 * power than the render powerwell.
 	 */
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_MEDIA);
+	intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_MEDIA);
 	err = valleyview_set_rps(dev_priv, val);
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_MEDIA);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_MEDIA);
 
 	if (err)
 		DRM_ERROR("Failed to set RPS for idle\n");
@@ -6691,8 +6846,7 @@ void gen6_rps_idle(struct drm_i915_private *dev_priv)
 	mutex_unlock(&dev_priv->pcu_lock);
 }
 
-void gen6_rps_boost(struct i915_request *rq,
-		    struct intel_rps_client *rps_client)
+void gen6_rps_boost(struct i915_request *rq)
 {
 	struct intel_rps *rps = &rq->i915->gt_pm.rps;
 	unsigned long flags;
@@ -6721,7 +6875,7 @@ void gen6_rps_boost(struct i915_request *rq,
 	if (READ_ONCE(rps->cur_freq) < rps->boost_freq)
 		schedule_work(&rps->work);
 
-	atomic_inc(rps_client ? &rps_client->boosts : &rps->boosts);
+	atomic_inc(&rps->boosts);
 }
 
 int intel_set_rps(struct drm_i915_private *dev_priv, u8 val)
@@ -6782,11 +6936,11 @@ static void valleyview_disable_rc6(struct drm_i915_private *dev_priv)
 {
 	/* We're doing forcewake before Disabling RC6,
 	 * This what the BIOS expects when going into suspend */
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL);
 
 	I915_WRITE(GEN6_RC_CONTROL, 0);
 
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 }
 
 static void valleyview_disable_rps(struct drm_i915_private *dev_priv)
@@ -6945,7 +7099,7 @@ static void reset_rps(struct drm_i915_private *dev_priv,
 /* See the Gen9_GT_PM_Programming_Guide doc for the below */
 static void gen9_enable_rps(struct drm_i915_private *dev_priv)
 {
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL);
 
 	/* Program defaults and thresholds for RPS */
 	if (IS_GEN(dev_priv, 9))
@@ -6963,7 +7117,79 @@ static void gen9_enable_rps(struct drm_i915_private *dev_priv)
 	 * RP_INTERRUPT_LIMITS & RPNSWREQ registers */
 	reset_rps(dev_priv, gen6_set_rps);
 
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
+}
+
+static void gen11_enable_rc6(struct drm_i915_private *dev_priv)
+{
+	struct intel_engine_cs *engine;
+	enum intel_engine_id id;
+
+	/* 1a: Software RC state - RC0 */
+	I915_WRITE(GEN6_RC_STATE, 0);
+
+	/*
+	 * 1b: Get forcewake during program sequence. Although the driver
+	 * hasn't enabled a state yet where we need forcewake, BIOS may have.
+	 */
+	intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL);
+
+	/* 2a: Disable RC states. */
+	I915_WRITE(GEN6_RC_CONTROL, 0);
+
+	/* 2b: Program RC6 thresholds.*/
+	I915_WRITE(GEN6_RC6_WAKE_RATE_LIMIT, 54 << 16 | 85);
+	I915_WRITE(GEN10_MEDIA_WAKE_RATE_LIMIT, 150);
+
+	I915_WRITE(GEN6_RC_EVALUATION_INTERVAL, 125000); /* 12500 * 1280ns */
+	I915_WRITE(GEN6_RC_IDLE_HYSTERSIS, 25); /* 25 * 1280ns */
+	for_each_engine(engine, dev_priv, id)
+		I915_WRITE(RING_MAX_IDLE(engine->mmio_base), 10);
+
+	if (HAS_GUC(dev_priv))
+		I915_WRITE(GUC_MAX_IDLE_COUNT, 0xA);
+
+	I915_WRITE(GEN6_RC_SLEEP, 0);
+
+	I915_WRITE(GEN6_RC6_THRESHOLD, 50000); /* 50/125ms per EI */
+
+	/*
+	 * 2c: Program Coarse Power Gating Policies.
+	 *
+	 * Bspec's guidance is to use 25us (really 25 * 1280ns) here. What we
+	 * use instead is a more conservative estimate for the maximum time
+	 * it takes us to service a CS interrupt and submit a new ELSP - that
+	 * is the time which the GPU is idle waiting for the CPU to select the
+	 * next request to execute. If the idle hysteresis is less than that
+	 * interrupt service latency, the hardware will automatically gate
+	 * the power well and we will then incur the wake up cost on top of
+	 * the service latency. A similar guide from intel_pstate is that we
+	 * do not want the enable hysteresis to less than the wakeup latency.
+	 *
+	 * igt/gem_exec_nop/sequential provides a rough estimate for the
+	 * service latency, and puts it around 10us for Broadwell (and other
+	 * big core) and around 40us for Broxton (and other low power cores).
+	 * [Note that for legacy ringbuffer submission, this is less than 1us!]
+	 * However, the wakeup latency on Broxton is closer to 100us. To be
+	 * conservative, we have to factor in a context switch on top (due
+	 * to ksoftirqd).
+	 */
+	I915_WRITE(GEN9_MEDIA_PG_IDLE_HYSTERESIS, 250);
+	I915_WRITE(GEN9_RENDER_PG_IDLE_HYSTERESIS, 250);
+
+	/* 3a: Enable RC6 */
+	I915_WRITE(GEN6_RC_CONTROL,
+		   GEN6_RC_CTL_HW_ENABLE |
+		   GEN6_RC_CTL_RC6_ENABLE |
+		   GEN6_RC_CTL_EI_MODE(1));
+
+	/* 3b: Enable Coarse Power Gating only when RC6 is enabled. */
+	I915_WRITE(GEN9_PG_ENABLE,
+		   GEN9_RENDER_PG_ENABLE |
+		   GEN9_MEDIA_PG_ENABLE |
+		   GEN11_MEDIA_SAMPLER_PG_ENABLE);
+
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 }
 
 static void gen9_enable_rc6(struct drm_i915_private *dev_priv)
@@ -6977,7 +7203,7 @@ static void gen9_enable_rc6(struct drm_i915_private *dev_priv)
 
 	/* 1b: Get forcewake during program sequence. Although the driver
 	 * hasn't enabled a state yet where we need forcewake, BIOS may have.*/
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL);
 
 	/* 2a: Disable RC states. */
 	I915_WRITE(GEN6_RC_CONTROL, 0);
@@ -7054,7 +7280,7 @@ static void gen9_enable_rc6(struct drm_i915_private *dev_priv)
 		I915_WRITE(GEN9_PG_ENABLE,
 			   GEN9_RENDER_PG_ENABLE | GEN9_MEDIA_PG_ENABLE);
 
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 }
 
 static void gen8_enable_rc6(struct drm_i915_private *dev_priv)
@@ -7067,7 +7293,7 @@ static void gen8_enable_rc6(struct drm_i915_private *dev_priv)
 
 	/* 1b: Get forcewake during program sequence. Although the driver
 	 * hasn't enabled a state yet where we need forcewake, BIOS may have.*/
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL);
 
 	/* 2a: Disable RC states. */
 	I915_WRITE(GEN6_RC_CONTROL, 0);
@@ -7088,14 +7314,14 @@ static void gen8_enable_rc6(struct drm_i915_private *dev_priv)
 		   GEN7_RC_CTL_TO_MODE |
 		   GEN6_RC_CTL_RC6_ENABLE);
 
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 }
 
 static void gen8_enable_rps(struct drm_i915_private *dev_priv)
 {
 	struct intel_rps *rps = &dev_priv->gt_pm.rps;
 
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL);
 
 	/* 1 Program defaults and thresholds for RPS*/
 	I915_WRITE(GEN6_RPNSWREQ,
@@ -7128,7 +7354,7 @@ static void gen8_enable_rps(struct drm_i915_private *dev_priv)
 
 	reset_rps(dev_priv, gen6_set_rps);
 
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 }
 
 static void gen6_enable_rc6(struct drm_i915_private *dev_priv)
@@ -7148,7 +7374,7 @@ static void gen6_enable_rc6(struct drm_i915_private *dev_priv)
 		I915_WRITE(GTFIFODBG, gtfifodbg);
 	}
 
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL);
 
 	/* disable the counters and set deterministic thresholds */
 	I915_WRITE(GEN6_RC_CONTROL, 0);
@@ -7196,7 +7422,7 @@ static void gen6_enable_rc6(struct drm_i915_private *dev_priv)
 			DRM_ERROR("Couldn't fix incorrect rc6 voltage\n");
 	}
 
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 }
 
 static void gen6_enable_rps(struct drm_i915_private *dev_priv)
@@ -7207,7 +7433,7 @@ static void gen6_enable_rps(struct drm_i915_private *dev_priv)
 	 * Perhaps there might be some value in exposing these to
 	 * userspace...
 	 */
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL);
 
 	/* Power down if completely idle for over 50ms */
 	I915_WRITE(GEN6_RP_DOWN_TIMEOUT, 50000);
@@ -7215,7 +7441,7 @@ static void gen6_enable_rps(struct drm_i915_private *dev_priv)
 
 	reset_rps(dev_priv, gen6_set_rps);
 
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 }
 
 static void gen6_update_ring_freq(struct drm_i915_private *dev_priv)
@@ -7638,7 +7864,7 @@ static void cherryview_enable_rc6(struct drm_i915_private *dev_priv)
 
 	/* 1a & 1b: Get forcewake during program sequence. Although the driver
 	 * hasn't enabled a state yet where we need forcewake, BIOS may have.*/
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL);
 
 	/*  Disable RC states. */
 	I915_WRITE(GEN6_RC_CONTROL, 0);
@@ -7670,14 +7896,14 @@ static void cherryview_enable_rc6(struct drm_i915_private *dev_priv)
 		rc6_mode = GEN7_RC_CTL_TO_MODE;
 	I915_WRITE(GEN6_RC_CONTROL, rc6_mode);
 
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 }
 
 static void cherryview_enable_rps(struct drm_i915_private *dev_priv)
 {
 	u32 val;
 
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL);
 
 	/* 1: Program defaults and thresholds for RPS*/
 	I915_WRITE(GEN6_RP_DOWN_TIMEOUT, 1000000);
@@ -7712,7 +7938,7 @@ static void cherryview_enable_rps(struct drm_i915_private *dev_priv)
 
 	reset_rps(dev_priv, valleyview_set_rps);
 
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 }
 
 static void valleyview_enable_rc6(struct drm_i915_private *dev_priv)
@@ -7730,7 +7956,7 @@ static void valleyview_enable_rc6(struct drm_i915_private *dev_priv)
 		I915_WRITE(GTFIFODBG, gtfifodbg);
 	}
 
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL);
 
 	/*  Disable RC states. */
 	I915_WRITE(GEN6_RC_CONTROL, 0);
@@ -7755,14 +7981,14 @@ static void valleyview_enable_rc6(struct drm_i915_private *dev_priv)
 	I915_WRITE(GEN6_RC_CONTROL,
 		   GEN7_RC_CTL_TO_MODE | VLV_RC_CTL_CTX_RST_PARALLEL);
 
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 }
 
 static void valleyview_enable_rps(struct drm_i915_private *dev_priv)
 {
 	u32 val;
 
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL);
 
 	I915_WRITE(GEN6_RP_DOWN_TIMEOUT, 1000000);
 	I915_WRITE(GEN6_RP_UP_THRESHOLD, 59400);
@@ -7796,7 +8022,7 @@ static void valleyview_enable_rps(struct drm_i915_private *dev_priv)
 
 	reset_rps(dev_priv, valleyview_set_rps);
 
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 }
 
 static unsigned long intel_pxfreq(u32 vidfreq)
@@ -8037,14 +8263,14 @@ unsigned long i915_gfx_val(struct drm_i915_private *dev_priv)
 	return val;
 }
 
-static struct drm_i915_private *i915_mch_dev;
+static struct drm_i915_private __rcu *i915_mch_dev;
 
 static struct drm_i915_private *mchdev_get(void)
 {
 	struct drm_i915_private *i915;
 
 	rcu_read_lock();
-	i915 = i915_mch_dev;
+	i915 = rcu_dereference(i915_mch_dev);
 	if (!kref_get_unless_zero(&i915->drm.ref))
 		i915 = NULL;
 	rcu_read_unlock();
@@ -8343,22 +8569,6 @@ void intel_cleanup_gt_powersave(struct drm_i915_private *dev_priv)
 		pm_runtime_put(&dev_priv->drm.pdev->dev);
 }
 
-/**
- * intel_suspend_gt_powersave - suspend PM work and helper threads
- * @dev_priv: i915 device
- *
- * We don't want to disable RC6 or other features here, we just want
- * to make sure any work we've queued has finished and won't bother
- * us while we're suspended.
- */
-void intel_suspend_gt_powersave(struct drm_i915_private *dev_priv)
-{
-	if (INTEL_GEN(dev_priv) < 6)
-		return;
-
-	/* gen6_rps_idle() will be called later to disable interrupts */
-}
-
 void intel_sanitize_gt_powersave(struct drm_i915_private *dev_priv)
 {
 	dev_priv->gt_pm.rps.enabled = true; /* force RPS disabling */
@@ -8458,6 +8668,8 @@ static void intel_enable_rc6(struct drm_i915_private *dev_priv)
 		cherryview_enable_rc6(dev_priv);
 	else if (IS_VALLEYVIEW(dev_priv))
 		valleyview_enable_rc6(dev_priv);
+	else if (INTEL_GEN(dev_priv) >= 11)
+		gen11_enable_rc6(dev_priv);
 	else if (INTEL_GEN(dev_priv) >= 9)
 		gen9_enable_rc6(dev_priv);
 	else if (IS_BROADWELL(dev_priv))
@@ -9361,7 +9573,7 @@ static void nop_init_clock_gating(struct drm_i915_private *dev_priv)
  */
 void intel_init_clock_gating_hooks(struct drm_i915_private *dev_priv)
 {
-	if (IS_ICELAKE(dev_priv))
+	if (IS_GEN(dev_priv, 11))
 		dev_priv->display.init_clock_gating = icl_init_clock_gating;
 	else if (IS_CANNONLAKE(dev_priv))
 		dev_priv->display.init_clock_gating = cnl_init_clock_gating;
@@ -9454,7 +9666,7 @@ void intel_init_pm(struct drm_i915_private *dev_priv)
 		dev_priv->display.initial_watermarks = g4x_initial_watermarks;
 		dev_priv->display.optimize_watermarks = g4x_optimize_watermarks;
 	} else if (IS_PINEVIEW(dev_priv)) {
-		if (!intel_get_cxsr_latency(IS_PINEVIEW_G(dev_priv),
+		if (!intel_get_cxsr_latency(!IS_MOBILE(dev_priv),
 					    dev_priv->is_ddr3,
 					    dev_priv->fsb_freq,
 					    dev_priv->mem_freq)) {
@@ -9552,7 +9764,7 @@ int sandybridge_pcode_read(struct drm_i915_private *dev_priv, u32 mbox, u32 *val
 	I915_WRITE_FW(GEN6_PCODE_DATA1, 0);
 	I915_WRITE_FW(GEN6_PCODE_MAILBOX, GEN6_PCODE_READY | mbox);
 
-	if (__intel_wait_for_register_fw(dev_priv,
+	if (__intel_wait_for_register_fw(&dev_priv->uncore,
 					 GEN6_PCODE_MAILBOX, GEN6_PCODE_READY, 0,
 					 500, 0, NULL)) {
 		DRM_ERROR("timeout waiting for pcode read (from mbox %x) to finish for %ps\n",
@@ -9600,7 +9812,7 @@ int sandybridge_pcode_write_timeout(struct drm_i915_private *dev_priv,
 	I915_WRITE_FW(GEN6_PCODE_DATA1, 0);
 	I915_WRITE_FW(GEN6_PCODE_MAILBOX, GEN6_PCODE_READY | mbox);
 
-	if (__intel_wait_for_register_fw(dev_priv,
+	if (__intel_wait_for_register_fw(&dev_priv->uncore,
 					 GEN6_PCODE_MAILBOX, GEN6_PCODE_READY, 0,
 					 fast_timeout_us, slow_timeout_ms,
 					 NULL)) {
@@ -9824,6 +10036,7 @@ static u64 vlv_residency_raw(struct drm_i915_private *dev_priv,
 u64 intel_rc6_residency_ns(struct drm_i915_private *dev_priv,
 			   const i915_reg_t reg)
 {
+	struct intel_uncore *uncore = &dev_priv->uncore;
 	u64 time_hw, prev_hw, overflow_hw;
 	unsigned int fw_domains;
 	unsigned long flags;
@@ -9845,10 +10058,10 @@ u64 intel_rc6_residency_ns(struct drm_i915_private *dev_priv,
 	if (WARN_ON_ONCE(i >= ARRAY_SIZE(dev_priv->gt_pm.rc6.cur_residency)))
 		return 0;
 
-	fw_domains = intel_uncore_forcewake_for_reg(dev_priv, reg, FW_REG_READ);
+	fw_domains = intel_uncore_forcewake_for_reg(uncore, reg, FW_REG_READ);
 
-	spin_lock_irqsave(&dev_priv->uncore.lock, flags);
-	intel_uncore_forcewake_get__locked(dev_priv, fw_domains);
+	spin_lock_irqsave(&uncore->lock, flags);
+	intel_uncore_forcewake_get__locked(uncore, fw_domains);
 
 	/* On VLV and CHV, residency time is in CZ units rather than 1.28us */
 	if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {
@@ -9867,7 +10080,7 @@ u64 intel_rc6_residency_ns(struct drm_i915_private *dev_priv,
 		}
 
 		overflow_hw = BIT_ULL(32);
-		time_hw = I915_READ_FW(reg);
+		time_hw = intel_uncore_read_fw(uncore, reg);
 	}
 
 	/*
@@ -9889,8 +10102,8 @@ u64 intel_rc6_residency_ns(struct drm_i915_private *dev_priv,
 	time_hw += dev_priv->gt_pm.rc6.cur_residency[i];
 	dev_priv->gt_pm.rc6.cur_residency[i] = time_hw;
 
-	intel_uncore_forcewake_put__locked(dev_priv, fw_domains);
-	spin_unlock_irqrestore(&dev_priv->uncore.lock, flags);
+	intel_uncore_forcewake_put__locked(uncore, fw_domains);
+	spin_unlock_irqrestore(&uncore->lock, flags);
 
 	return mul_u64_u32_div(time_hw, mul, div);
 }
diff --git a/drivers/gpu/drm/i915/intel_pm.h b/drivers/gpu/drm/i915/intel_pm.h
new file mode 100644
index 000000000000..674a3f0f16a7
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_pm.h
@@ -0,0 +1,71 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_PM_H__
+#define __INTEL_PM_H__
+
+#include <linux/types.h>
+
+struct drm_atomic_state;
+struct drm_device;
+struct drm_i915_private;
+struct i915_request;
+struct intel_crtc;
+struct intel_crtc_state;
+struct intel_plane;
+struct skl_ddb_allocation;
+struct skl_ddb_entry;
+struct skl_pipe_wm;
+struct skl_wm_level;
+
+void intel_init_clock_gating(struct drm_i915_private *dev_priv);
+void intel_suspend_hw(struct drm_i915_private *dev_priv);
+int ilk_wm_max_level(const struct drm_i915_private *dev_priv);
+void intel_update_watermarks(struct intel_crtc *crtc);
+void intel_init_pm(struct drm_i915_private *dev_priv);
+void intel_init_clock_gating_hooks(struct drm_i915_private *dev_priv);
+void intel_pm_setup(struct drm_i915_private *dev_priv);
+void intel_gpu_ips_init(struct drm_i915_private *dev_priv);
+void intel_gpu_ips_teardown(void);
+void intel_init_gt_powersave(struct drm_i915_private *dev_priv);
+void intel_cleanup_gt_powersave(struct drm_i915_private *dev_priv);
+void intel_sanitize_gt_powersave(struct drm_i915_private *dev_priv);
+void intel_enable_gt_powersave(struct drm_i915_private *dev_priv);
+void intel_disable_gt_powersave(struct drm_i915_private *dev_priv);
+void gen6_rps_busy(struct drm_i915_private *dev_priv);
+void gen6_rps_idle(struct drm_i915_private *dev_priv);
+void gen6_rps_boost(struct i915_request *rq);
+void g4x_wm_get_hw_state(struct drm_i915_private *dev_priv);
+void vlv_wm_get_hw_state(struct drm_i915_private *dev_priv);
+void ilk_wm_get_hw_state(struct drm_i915_private *dev_priv);
+void skl_wm_get_hw_state(struct drm_i915_private *dev_priv);
+void skl_pipe_ddb_get_hw_state(struct intel_crtc *crtc,
+			       struct skl_ddb_entry *ddb_y,
+			       struct skl_ddb_entry *ddb_uv);
+void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv,
+			  struct skl_ddb_allocation *ddb /* out */);
+void skl_pipe_wm_get_hw_state(struct intel_crtc *crtc,
+			      struct skl_pipe_wm *out);
+void g4x_wm_sanitize(struct drm_i915_private *dev_priv);
+void vlv_wm_sanitize(struct drm_i915_private *dev_priv);
+bool intel_can_enable_sagv(struct drm_atomic_state *state);
+int intel_enable_sagv(struct drm_i915_private *dev_priv);
+int intel_disable_sagv(struct drm_i915_private *dev_priv);
+bool skl_wm_level_equals(const struct skl_wm_level *l1,
+			 const struct skl_wm_level *l2);
+bool skl_ddb_allocation_overlaps(const struct skl_ddb_entry *ddb,
+				 const struct skl_ddb_entry *entries,
+				 int num_entries, int ignore_idx);
+void skl_write_plane_wm(struct intel_plane *plane,
+			const struct intel_crtc_state *crtc_state);
+void skl_write_cursor_wm(struct intel_plane *plane,
+			 const struct intel_crtc_state *crtc_state);
+bool ilk_disable_lp_wm(struct drm_device *dev);
+int skl_check_pipe_max_pixel_rate(struct intel_crtc *intel_crtc,
+				  struct intel_crtc_state *cstate);
+void intel_init_ipc(struct drm_i915_private *dev_priv);
+void intel_enable_ipc(struct drm_i915_private *dev_priv);
+
+#endif /* __INTEL_PM_H__ */
diff --git a/drivers/gpu/drm/i915/intel_psr.c b/drivers/gpu/drm/i915/intel_psr.c
index 84a0fb981561..963663ba0edf 100644
--- a/drivers/gpu/drm/i915/intel_psr.c
+++ b/drivers/gpu/drm/i915/intel_psr.c
@@ -21,6 +21,14 @@
  * DEALINGS IN THE SOFTWARE.
  */
 
+#include <drm/drm_atomic_helper.h>
+
+#include "i915_drv.h"
+#include "intel_dp.h"
+#include "intel_drv.h"
+#include "intel_psr.h"
+#include "intel_sprite.h"
+
 /**
  * DOC: Panel Self Refresh (PSR/SRD)
  *
@@ -51,10 +59,6 @@
  * must be correctly synchronized/cancelled when shutting down the pipe."
  */
 
-
-#include "intel_drv.h"
-#include "i915_drv.h"
-
 static bool psr_global_enabled(u32 debug)
 {
 	switch (debug & I915_PSR_DEBUG_MODE_MASK) {
@@ -78,9 +82,6 @@ static bool intel_psr2_enabled(struct drm_i915_private *dev_priv,
 	case I915_PSR_DEBUG_DISABLE:
 	case I915_PSR_DEBUG_FORCE_PSR1:
 		return false;
-	case I915_PSR_DEBUG_DEFAULT:
-		if (i915_modparams.enable_psr <= 0)
-			return false;
 	default:
 		return crtc_state->has_psr2;
 	}
@@ -435,32 +436,16 @@ static void intel_psr_enable_sink(struct intel_dp *intel_dp)
 	drm_dp_dpcd_writeb(&intel_dp->aux, DP_SET_POWER, DP_SET_POWER_D0);
 }
 
-static void hsw_activate_psr1(struct intel_dp *intel_dp)
+static u32 intel_psr1_get_tp_time(struct intel_dp *intel_dp)
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
-	u32 max_sleep_time = 0x1f;
-	u32 val = EDP_PSR_ENABLE;
+	u32 val = 0;
 
-	/* Let's use 6 as the minimum to cover all known cases including the
-	 * off-by-one issue that HW has in some cases.
-	 */
-	int idle_frames = max(6, dev_priv->vbt.psr.idle_frames);
-
-	/* sink_sync_latency of 8 means source has to wait for more than 8
-	 * frames, we'll go with 9 frames for now
-	 */
-	idle_frames = max(idle_frames, dev_priv->psr.sink_sync_latency + 1);
-	val |= idle_frames << EDP_PSR_IDLE_FRAME_SHIFT;
-
-	val |= max_sleep_time << EDP_PSR_MAX_SLEEP_TIME_SHIFT;
-	if (IS_HASWELL(dev_priv))
-		val |= EDP_PSR_MIN_LINK_ENTRY_TIME_8_LINES;
-
-	if (dev_priv->psr.link_standby)
-		val |= EDP_PSR_LINK_STANDBY;
+	if (INTEL_GEN(dev_priv) >= 11)
+		val |= EDP_PSR_TP4_TIME_0US;
 
 	if (dev_priv->vbt.psr.tp1_wakeup_time_us == 0)
-		val |=  EDP_PSR_TP1_TIME_0us;
+		val |= EDP_PSR_TP1_TIME_0us;
 	else if (dev_priv->vbt.psr.tp1_wakeup_time_us <= 100)
 		val |= EDP_PSR_TP1_TIME_100us;
 	else if (dev_priv->vbt.psr.tp1_wakeup_time_us <= 500)
@@ -469,7 +454,7 @@ static void hsw_activate_psr1(struct intel_dp *intel_dp)
 		val |= EDP_PSR_TP1_TIME_2500us;
 
 	if (dev_priv->vbt.psr.tp2_tp3_wakeup_time_us == 0)
-		val |=  EDP_PSR_TP2_TP3_TIME_0us;
+		val |= EDP_PSR_TP2_TP3_TIME_0us;
 	else if (dev_priv->vbt.psr.tp2_tp3_wakeup_time_us <= 100)
 		val |= EDP_PSR_TP2_TP3_TIME_100us;
 	else if (dev_priv->vbt.psr.tp2_tp3_wakeup_time_us <= 500)
@@ -483,6 +468,35 @@ static void hsw_activate_psr1(struct intel_dp *intel_dp)
 	else
 		val |= EDP_PSR_TP1_TP2_SEL;
 
+	return val;
+}
+
+static void hsw_activate_psr1(struct intel_dp *intel_dp)
+{
+	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+	u32 max_sleep_time = 0x1f;
+	u32 val = EDP_PSR_ENABLE;
+
+	/* Let's use 6 as the minimum to cover all known cases including the
+	 * off-by-one issue that HW has in some cases.
+	 */
+	int idle_frames = max(6, dev_priv->vbt.psr.idle_frames);
+
+	/* sink_sync_latency of 8 means source has to wait for more than 8
+	 * frames, we'll go with 9 frames for now
+	 */
+	idle_frames = max(idle_frames, dev_priv->psr.sink_sync_latency + 1);
+	val |= idle_frames << EDP_PSR_IDLE_FRAME_SHIFT;
+
+	val |= max_sleep_time << EDP_PSR_MAX_SLEEP_TIME_SHIFT;
+	if (IS_HASWELL(dev_priv))
+		val |= EDP_PSR_MIN_LINK_ENTRY_TIME_8_LINES;
+
+	if (dev_priv->psr.link_standby)
+		val |= EDP_PSR_LINK_STANDBY;
+
+	val |= intel_psr1_get_tp_time(intel_dp);
+
 	if (INTEL_GEN(dev_priv) >= 8)
 		val |= EDP_PSR_CRC_ENABLE;
 
@@ -509,16 +523,22 @@ static void hsw_activate_psr2(struct intel_dp *intel_dp)
 
 	val |= EDP_PSR2_FRAME_BEFORE_SU(dev_priv->psr.sink_sync_latency + 1);
 
-	if (dev_priv->vbt.psr.tp2_tp3_wakeup_time_us >= 0 &&
-	    dev_priv->vbt.psr.tp2_tp3_wakeup_time_us <= 50)
+	if (dev_priv->vbt.psr.psr2_tp2_tp3_wakeup_time_us >= 0 &&
+	    dev_priv->vbt.psr.psr2_tp2_tp3_wakeup_time_us <= 50)
 		val |= EDP_PSR2_TP2_TIME_50us;
-	else if (dev_priv->vbt.psr.tp2_tp3_wakeup_time_us <= 100)
+	else if (dev_priv->vbt.psr.psr2_tp2_tp3_wakeup_time_us <= 100)
 		val |= EDP_PSR2_TP2_TIME_100us;
-	else if (dev_priv->vbt.psr.tp2_tp3_wakeup_time_us <= 500)
+	else if (dev_priv->vbt.psr.psr2_tp2_tp3_wakeup_time_us <= 500)
 		val |= EDP_PSR2_TP2_TIME_500us;
 	else
 		val |= EDP_PSR2_TP2_TIME_2500us;
 
+	/*
+	 * PSR2 HW is incorrectly using EDP_PSR_TP1_TP3_SEL and BSpec is
+	 * recommending keep this bit unset while PSR2 is enabled.
+	 */
+	I915_WRITE(EDP_PSR_CTL, 0);
+
 	I915_WRITE(EDP_PSR2_CTL, val);
 }
 
@@ -530,11 +550,6 @@ static bool intel_psr2_config_valid(struct intel_dp *intel_dp,
 	int crtc_vdisplay = crtc_state->base.adjusted_mode.crtc_vdisplay;
 	int psr_max_h = 0, psr_max_v = 0;
 
-	/*
-	 * FIXME psr2_support is messed up. It's both computed
-	 * dynamically during PSR enable, and extracted from sink
-	 * caps during eDP detection.
-	 */
 	if (!dev_priv->psr.sink_psr2_support)
 		return false;
 
@@ -575,6 +590,11 @@ static bool intel_psr2_config_valid(struct intel_dp *intel_dp,
 		return false;
 	}
 
+	if (crtc_state->crc_enabled) {
+		DRM_DEBUG_KMS("PSR2 not enabled because it would inhibit pipe CRC calculation\n");
+		return false;
+	}
+
 	return true;
 }
 
@@ -610,9 +630,8 @@ void intel_psr_compute_config(struct intel_dp *intel_dp,
 		return;
 	}
 
-	if (IS_HASWELL(dev_priv) &&
-	    adjusted_mode->flags & DRM_MODE_FLAG_INTERLACE) {
-		DRM_DEBUG_KMS("PSR condition failed: Interlaced is Enabled\n");
+	if (adjusted_mode->flags & DRM_MODE_FLAG_INTERLACE) {
+		DRM_DEBUG_KMS("PSR condition failed: Interlaced mode enabled\n");
 		return;
 	}
 
@@ -718,8 +737,11 @@ static void intel_psr_enable_locked(struct drm_i915_private *dev_priv,
 {
 	struct intel_dp *intel_dp = dev_priv->psr.dp;
 
-	if (dev_priv->psr.enabled)
-		return;
+	WARN_ON(dev_priv->psr.enabled);
+
+	dev_priv->psr.psr2_enabled = intel_psr2_enabled(dev_priv, crtc_state);
+	dev_priv->psr.busy_frontbuffer_bits = 0;
+	dev_priv->psr.pipe = to_intel_crtc(crtc_state->base.crtc)->pipe;
 
 	DRM_DEBUG_KMS("Enabling PSR%s\n",
 		      dev_priv->psr.psr2_enabled ? "2" : "1");
@@ -752,20 +774,13 @@ void intel_psr_enable(struct intel_dp *intel_dp,
 	WARN_ON(dev_priv->drrs.dp);
 
 	mutex_lock(&dev_priv->psr.lock);
-	if (dev_priv->psr.prepared) {
-		DRM_DEBUG_KMS("PSR already in use\n");
+
+	if (!psr_global_enabled(dev_priv->psr.debug)) {
+		DRM_DEBUG_KMS("PSR disabled by flag\n");
 		goto unlock;
 	}
 
-	dev_priv->psr.psr2_enabled = intel_psr2_enabled(dev_priv, crtc_state);
-	dev_priv->psr.busy_frontbuffer_bits = 0;
-	dev_priv->psr.prepared = true;
-	dev_priv->psr.pipe = to_intel_crtc(crtc_state->base.crtc)->pipe;
-
-	if (psr_global_enabled(dev_priv->psr.debug))
-		intel_psr_enable_locked(dev_priv, crtc_state);
-	else
-		DRM_DEBUG_KMS("PSR disabled by flag\n");
+	intel_psr_enable_locked(dev_priv, crtc_state);
 
 unlock:
 	mutex_unlock(&dev_priv->psr.lock);
@@ -819,8 +834,8 @@ static void intel_psr_disable_locked(struct intel_dp *intel_dp)
 	}
 
 	/* Wait till PSR is idle */
-	if (intel_wait_for_register(dev_priv, psr_status, psr_status_mask, 0,
-				    2000))
+	if (intel_wait_for_register(&dev_priv->uncore,
+				    psr_status, psr_status_mask, 0, 2000))
 		DRM_ERROR("Timed out waiting PSR idle state\n");
 
 	/* Disable PSR on Sink */
@@ -848,18 +863,69 @@ void intel_psr_disable(struct intel_dp *intel_dp,
 		return;
 
 	mutex_lock(&dev_priv->psr.lock);
-	if (!dev_priv->psr.prepared) {
-		mutex_unlock(&dev_priv->psr.lock);
-		return;
-	}
 
 	intel_psr_disable_locked(intel_dp);
 
-	dev_priv->psr.prepared = false;
 	mutex_unlock(&dev_priv->psr.lock);
 	cancel_work_sync(&dev_priv->psr.work);
 }
 
+static void psr_force_hw_tracking_exit(struct drm_i915_private *dev_priv)
+{
+	/*
+	 * Display WA #0884: all
+	 * This documented WA for bxt can be safely applied
+	 * broadly so we can force HW tracking to exit PSR
+	 * instead of disabling and re-enabling.
+	 * Workaround tells us to write 0 to CUR_SURFLIVE_A,
+	 * but it makes more sense write to the current active
+	 * pipe.
+	 */
+	I915_WRITE(CURSURFLIVE(dev_priv->psr.pipe), 0);
+}
+
+/**
+ * intel_psr_update - Update PSR state
+ * @intel_dp: Intel DP
+ * @crtc_state: new CRTC state
+ *
+ * This functions will update PSR states, disabling, enabling or switching PSR
+ * version when executing fastsets. For full modeset, intel_psr_disable() and
+ * intel_psr_enable() should be called instead.
+ */
+void intel_psr_update(struct intel_dp *intel_dp,
+		      const struct intel_crtc_state *crtc_state)
+{
+	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+	struct i915_psr *psr = &dev_priv->psr;
+	bool enable, psr2_enable;
+
+	if (!CAN_PSR(dev_priv) || READ_ONCE(psr->dp) != intel_dp)
+		return;
+
+	mutex_lock(&dev_priv->psr.lock);
+
+	enable = crtc_state->has_psr && psr_global_enabled(psr->debug);
+	psr2_enable = intel_psr2_enabled(dev_priv, crtc_state);
+
+	if (enable == psr->enabled && psr2_enable == psr->psr2_enabled) {
+		/* Force a PSR exit when enabling CRC to avoid CRC timeouts */
+		if (crtc_state->crc_enabled && psr->enabled)
+			psr_force_hw_tracking_exit(dev_priv);
+
+		goto unlock;
+	}
+
+	if (psr->enabled)
+		intel_psr_disable_locked(intel_dp);
+
+	if (enable)
+		intel_psr_enable_locked(dev_priv, crtc_state);
+
+unlock:
+	mutex_unlock(&dev_priv->psr.lock);
+}
+
 /**
  * intel_psr_wait_for_idle - wait for PSR1 to idle
  * @new_crtc_state: new CRTC state
@@ -890,7 +956,7 @@ int intel_psr_wait_for_idle(const struct intel_crtc_state *new_crtc_state,
 	 * defensive enough to cover everything.
 	 */
 
-	return __intel_wait_for_register(dev_priv, EDP_PSR_STATUS,
+	return __intel_wait_for_register(&dev_priv->uncore, EDP_PSR_STATUS,
 					 EDP_PSR_STATUS_STATE_MASK,
 					 EDP_PSR_STATUS_STATE_IDLE, 2, 50,
 					 out_value);
@@ -915,7 +981,7 @@ static bool __psr_wait_for_idle_locked(struct drm_i915_private *dev_priv)
 
 	mutex_unlock(&dev_priv->psr.lock);
 
-	err = intel_wait_for_register(dev_priv, reg, mask, 0, 50);
+	err = intel_wait_for_register(&dev_priv->uncore, reg, mask, 0, 50);
 	if (err)
 		DRM_ERROR("Timed out waiting for PSR Idle for re-enable\n");
 
@@ -924,36 +990,63 @@ static bool __psr_wait_for_idle_locked(struct drm_i915_private *dev_priv)
 	return err == 0 && dev_priv->psr.enabled;
 }
 
-static bool switching_psr(struct drm_i915_private *dev_priv,
-			  struct intel_crtc_state *crtc_state,
-			  u32 mode)
+static int intel_psr_fastset_force(struct drm_i915_private *dev_priv)
 {
-	/* Can't switch psr state anyway if PSR2 is not supported. */
-	if (!crtc_state || !crtc_state->has_psr2)
-		return false;
+	struct drm_device *dev = &dev_priv->drm;
+	struct drm_modeset_acquire_ctx ctx;
+	struct drm_atomic_state *state;
+	struct drm_crtc *crtc;
+	int err;
 
-	if (dev_priv->psr.psr2_enabled && mode == I915_PSR_DEBUG_FORCE_PSR1)
-		return true;
+	state = drm_atomic_state_alloc(dev);
+	if (!state)
+		return -ENOMEM;
 
-	if (!dev_priv->psr.psr2_enabled && mode != I915_PSR_DEBUG_FORCE_PSR1)
-		return true;
+	drm_modeset_acquire_init(&ctx, DRM_MODESET_ACQUIRE_INTERRUPTIBLE);
+	state->acquire_ctx = &ctx;
+
+retry:
+	drm_for_each_crtc(crtc, dev) {
+		struct drm_crtc_state *crtc_state;
+		struct intel_crtc_state *intel_crtc_state;
+
+		crtc_state = drm_atomic_get_crtc_state(state, crtc);
+		if (IS_ERR(crtc_state)) {
+			err = PTR_ERR(crtc_state);
+			goto error;
+		}
+
+		intel_crtc_state = to_intel_crtc_state(crtc_state);
 
-	return false;
+		if (crtc_state->active && intel_crtc_state->has_psr) {
+			/* Mark mode as changed to trigger a pipe->update() */
+			crtc_state->mode_changed = true;
+			break;
+		}
+	}
+
+	err = drm_atomic_commit(state);
+
+error:
+	if (err == -EDEADLK) {
+		drm_atomic_state_clear(state);
+		err = drm_modeset_backoff(&ctx);
+		if (!err)
+			goto retry;
+	}
+
+	drm_modeset_drop_locks(&ctx);
+	drm_modeset_acquire_fini(&ctx);
+	drm_atomic_state_put(state);
+
+	return err;
 }
 
-int intel_psr_set_debugfs_mode(struct drm_i915_private *dev_priv,
-			       struct drm_modeset_acquire_ctx *ctx,
-			       u64 val)
+int intel_psr_debug_set(struct drm_i915_private *dev_priv, u64 val)
 {
-	struct drm_device *dev = &dev_priv->drm;
-	struct drm_connector_state *conn_state;
-	struct intel_crtc_state *crtc_state = NULL;
-	struct drm_crtc_commit *commit;
-	struct drm_crtc *crtc;
-	struct intel_dp *dp;
+	const u32 mode = val & I915_PSR_DEBUG_MODE_MASK;
+	u32 old_mode;
 	int ret;
-	bool enable;
-	u32 mode = val & I915_PSR_DEBUG_MODE_MASK;
 
 	if (val & ~(I915_PSR_DEBUG_IRQ | I915_PSR_DEBUG_MODE_MASK) ||
 	    mode > I915_PSR_DEBUG_FORCE_PSR1) {
@@ -961,49 +1054,19 @@ int intel_psr_set_debugfs_mode(struct drm_i915_private *dev_priv,
 		return -EINVAL;
 	}
 
-	ret = drm_modeset_lock(&dev->mode_config.connection_mutex, ctx);
-	if (ret)
-		return ret;
-
-	/* dev_priv->psr.dp should be set once and then never touched again. */
-	dp = READ_ONCE(dev_priv->psr.dp);
-	conn_state = dp->attached_connector->base.state;
-	crtc = conn_state->crtc;
-	if (crtc) {
-		ret = drm_modeset_lock(&crtc->mutex, ctx);
-		if (ret)
-			return ret;
-
-		crtc_state = to_intel_crtc_state(crtc->state);
-		commit = crtc_state->base.commit;
-	} else {
-		commit = conn_state->commit;
-	}
-	if (commit) {
-		ret = wait_for_completion_interruptible(&commit->hw_done);
-		if (ret)
-			return ret;
-	}
-
 	ret = mutex_lock_interruptible(&dev_priv->psr.lock);
 	if (ret)
 		return ret;
 
-	enable = psr_global_enabled(val);
-
-	if (!enable || switching_psr(dev_priv, crtc_state, mode))
-		intel_psr_disable_locked(dev_priv->psr.dp);
-
+	old_mode = dev_priv->psr.debug & I915_PSR_DEBUG_MODE_MASK;
 	dev_priv->psr.debug = val;
-	if (crtc)
-		dev_priv->psr.psr2_enabled = intel_psr2_enabled(dev_priv, crtc_state);
-
 	intel_psr_irq_control(dev_priv, dev_priv->psr.debug);
 
-	if (dev_priv->psr.prepared && enable)
-		intel_psr_enable_locked(dev_priv, crtc_state);
-
 	mutex_unlock(&dev_priv->psr.lock);
+
+	if (old_mode != mode)
+		ret = intel_psr_fastset_force(dev_priv);
+
 	return ret;
 }
 
@@ -1121,18 +1184,8 @@ void intel_psr_flush(struct drm_i915_private *dev_priv,
 	dev_priv->psr.busy_frontbuffer_bits &= ~frontbuffer_bits;
 
 	/* By definition flush = invalidate + flush */
-	if (frontbuffer_bits) {
-		/*
-		 * Display WA #0884: all
-		 * This documented WA for bxt can be safely applied
-		 * broadly so we can force HW tracking to exit PSR
-		 * instead of disabling and re-enabling.
-		 * Workaround tells us to write 0 to CUR_SURFLIVE_A,
-		 * but it makes more sense write to the current active
-		 * pipe.
-		 */
-		I915_WRITE(CURSURFLIVE(dev_priv->psr.pipe), 0);
-	}
+	if (frontbuffer_bits)
+		psr_force_hw_tracking_exit(dev_priv);
 
 	if (!dev_priv->psr.active && !dev_priv->psr.busy_frontbuffer_bits)
 		schedule_work(&dev_priv->psr.work);
@@ -1176,7 +1229,6 @@ void intel_psr_init(struct drm_i915_private *dev_priv)
 	if (val) {
 		DRM_DEBUG_KMS("PSR interruption error set\n");
 		dev_priv->psr.sink_not_reliable = true;
-		return;
 	}
 
 	/* Set link_standby x link_off defaults */
diff --git a/drivers/gpu/drm/i915/intel_psr.h b/drivers/gpu/drm/i915/intel_psr.h
new file mode 100644
index 000000000000..dc818826f36d
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_psr.h
@@ -0,0 +1,40 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_PSR_H__
+#define __INTEL_PSR_H__
+
+#include "intel_frontbuffer.h"
+
+struct drm_i915_private;
+struct intel_crtc_state;
+struct intel_dp;
+
+#define CAN_PSR(dev_priv) (HAS_PSR(dev_priv) && dev_priv->psr.sink_support)
+void intel_psr_init_dpcd(struct intel_dp *intel_dp);
+void intel_psr_enable(struct intel_dp *intel_dp,
+		      const struct intel_crtc_state *crtc_state);
+void intel_psr_disable(struct intel_dp *intel_dp,
+		       const struct intel_crtc_state *old_crtc_state);
+void intel_psr_update(struct intel_dp *intel_dp,
+		      const struct intel_crtc_state *crtc_state);
+int intel_psr_debug_set(struct drm_i915_private *dev_priv, u64 value);
+void intel_psr_invalidate(struct drm_i915_private *dev_priv,
+			  unsigned frontbuffer_bits,
+			  enum fb_op_origin origin);
+void intel_psr_flush(struct drm_i915_private *dev_priv,
+		     unsigned frontbuffer_bits,
+		     enum fb_op_origin origin);
+void intel_psr_init(struct drm_i915_private *dev_priv);
+void intel_psr_compute_config(struct intel_dp *intel_dp,
+			      struct intel_crtc_state *crtc_state);
+void intel_psr_irq_control(struct drm_i915_private *dev_priv, u32 debug);
+void intel_psr_irq_handler(struct drm_i915_private *dev_priv, u32 psr_iir);
+void intel_psr_short_pulse(struct intel_dp *intel_dp);
+int intel_psr_wait_for_idle(const struct intel_crtc_state *new_crtc_state,
+			    u32 *out_value);
+bool intel_psr_enabled(struct intel_dp *intel_dp);
+
+#endif /* __INTEL_PSR_H__ */
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 7f841dba87b3..029fd8ec1857 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -43,12 +43,6 @@
  */
 #define LEGACY_REQUEST_SIZE 200
 
-static inline u32 intel_hws_seqno_address(struct intel_engine_cs *engine)
-{
-	return (i915_ggtt_offset(engine->status_page.vma) +
-		I915_GEM_HWS_INDEX_ADDR);
-}
-
 unsigned int intel_ring_update_space(struct intel_ring *ring)
 {
 	unsigned int space;
@@ -317,9 +311,9 @@ static u32 *gen6_rcs_emit_breadcrumb(struct i915_request *rq, u32 *cs)
 	*cs++ = rq->fence.seqno;
 
 	*cs++ = GFX_OP_PIPE_CONTROL(4);
-	*cs++ = PIPE_CONTROL_QW_WRITE | PIPE_CONTROL_CS_STALL;
-	*cs++ = intel_hws_seqno_address(rq->engine) | PIPE_CONTROL_GLOBAL_GTT;
-	*cs++ = rq->global_seqno;
+	*cs++ = PIPE_CONTROL_QW_WRITE | PIPE_CONTROL_STORE_DATA_INDEX;
+	*cs++ = I915_GEM_HWS_HANGCHECK_ADDR | PIPE_CONTROL_GLOBAL_GTT;
+	*cs++ = intel_engine_next_hangcheck_seqno(rq->engine);
 
 	*cs++ = MI_USER_INTERRUPT;
 	*cs++ = MI_NOOP;
@@ -424,10 +418,10 @@ static u32 *gen7_rcs_emit_breadcrumb(struct i915_request *rq, u32 *cs)
 
 	*cs++ = GFX_OP_PIPE_CONTROL(4);
 	*cs++ = (PIPE_CONTROL_QW_WRITE |
-		 PIPE_CONTROL_GLOBAL_GTT_IVB |
-		 PIPE_CONTROL_CS_STALL);
-	*cs++ = intel_hws_seqno_address(rq->engine);
-	*cs++ = rq->global_seqno;
+		 PIPE_CONTROL_STORE_DATA_INDEX |
+		 PIPE_CONTROL_GLOBAL_GTT_IVB);
+	*cs++ = I915_GEM_HWS_HANGCHECK_ADDR;
+	*cs++ = intel_engine_next_hangcheck_seqno(rq->engine);
 
 	*cs++ = MI_USER_INTERRUPT;
 	*cs++ = MI_NOOP;
@@ -448,8 +442,8 @@ static u32 *gen6_xcs_emit_breadcrumb(struct i915_request *rq, u32 *cs)
 	*cs++ = rq->fence.seqno;
 
 	*cs++ = MI_FLUSH_DW | MI_FLUSH_DW_OP_STOREDW | MI_FLUSH_DW_STORE_INDEX;
-	*cs++ = I915_GEM_HWS_INDEX_ADDR | MI_FLUSH_DW_USE_GTT;
-	*cs++ = rq->global_seqno;
+	*cs++ = I915_GEM_HWS_HANGCHECK_ADDR | MI_FLUSH_DW_USE_GTT;
+	*cs++ = intel_engine_next_hangcheck_seqno(rq->engine);
 
 	*cs++ = MI_USER_INTERRUPT;
 	*cs++ = MI_NOOP;
@@ -473,8 +467,8 @@ static u32 *gen7_xcs_emit_breadcrumb(struct i915_request *rq, u32 *cs)
 	*cs++ = rq->fence.seqno;
 
 	*cs++ = MI_FLUSH_DW | MI_FLUSH_DW_OP_STOREDW | MI_FLUSH_DW_STORE_INDEX;
-	*cs++ = I915_GEM_HWS_INDEX_ADDR | MI_FLUSH_DW_USE_GTT;
-	*cs++ = rq->global_seqno;
+	*cs++ = I915_GEM_HWS_HANGCHECK_ADDR | MI_FLUSH_DW_USE_GTT;
+	*cs++ = intel_engine_next_hangcheck_seqno(rq->engine);
 
 	for (i = 0; i < GEN7_XCS_WA; i++) {
 		*cs++ = MI_STORE_DWORD_INDEX;
@@ -554,16 +548,17 @@ static void set_hwsp(struct intel_engine_cs *engine, u32 offset)
 		 */
 		default:
 			GEM_BUG_ON(engine->id);
-		case RCS:
+			/* fallthrough */
+		case RCS0:
 			hwsp = RENDER_HWS_PGA_GEN7;
 			break;
-		case BCS:
+		case BCS0:
 			hwsp = BLT_HWS_PGA_GEN7;
 			break;
-		case VCS:
+		case VCS0:
 			hwsp = BSD_HWS_PGA_GEN7;
 			break;
-		case VECS:
+		case VECS0:
 			hwsp = VEBOX_HWS_PGA_GEN7;
 			break;
 		}
@@ -580,19 +575,19 @@ static void set_hwsp(struct intel_engine_cs *engine, u32 offset)
 static void flush_cs_tlb(struct intel_engine_cs *engine)
 {
 	struct drm_i915_private *dev_priv = engine->i915;
-	i915_reg_t instpm = RING_INSTPM(engine->mmio_base);
 
 	if (!IS_GEN_RANGE(dev_priv, 6, 7))
 		return;
 
 	/* ring should be idle before issuing a sync flush*/
-	WARN_ON((I915_READ_MODE(engine) & MODE_IDLE) == 0);
-
-	I915_WRITE(instpm,
-		   _MASKED_BIT_ENABLE(INSTPM_TLB_INVALIDATE |
-				      INSTPM_SYNC_FLUSH));
-	if (intel_wait_for_register(dev_priv,
-				    instpm, INSTPM_SYNC_FLUSH, 0,
+	WARN_ON((ENGINE_READ(engine, RING_MI_MODE) & MODE_IDLE) == 0);
+
+	ENGINE_WRITE(engine, RING_INSTPM,
+		     _MASKED_BIT_ENABLE(INSTPM_TLB_INVALIDATE |
+					INSTPM_SYNC_FLUSH));
+	if (intel_wait_for_register(engine->uncore,
+				    RING_INSTPM(engine->mmio_base),
+				    INSTPM_SYNC_FLUSH, 0,
 				    1000))
 		DRM_ERROR("%s: wait for SyncFlush to complete for TLB invalidation timed out\n",
 			  engine->name);
@@ -611,32 +606,36 @@ static bool stop_ring(struct intel_engine_cs *engine)
 	struct drm_i915_private *dev_priv = engine->i915;
 
 	if (INTEL_GEN(dev_priv) > 2) {
-		I915_WRITE_MODE(engine, _MASKED_BIT_ENABLE(STOP_RING));
-		if (intel_wait_for_register(dev_priv,
+		ENGINE_WRITE(engine,
+			     RING_MI_MODE, _MASKED_BIT_ENABLE(STOP_RING));
+		if (intel_wait_for_register(engine->uncore,
 					    RING_MI_MODE(engine->mmio_base),
 					    MODE_IDLE,
 					    MODE_IDLE,
 					    1000)) {
 			DRM_ERROR("%s : timed out trying to stop ring\n",
 				  engine->name);
-			/* Sometimes we observe that the idle flag is not
+
+			/*
+			 * Sometimes we observe that the idle flag is not
 			 * set even though the ring is empty. So double
 			 * check before giving up.
 			 */
-			if (I915_READ_HEAD(engine) != I915_READ_TAIL(engine))
+			if (ENGINE_READ(engine, RING_HEAD) !=
+			    ENGINE_READ(engine, RING_TAIL))
 				return false;
 		}
 	}
 
-	I915_WRITE_HEAD(engine, I915_READ_TAIL(engine));
+	ENGINE_WRITE(engine, RING_HEAD, ENGINE_READ(engine, RING_TAIL));
 
-	I915_WRITE_HEAD(engine, 0);
-	I915_WRITE_TAIL(engine, 0);
+	ENGINE_WRITE(engine, RING_HEAD, 0);
+	ENGINE_WRITE(engine, RING_TAIL, 0);
 
 	/* The ring must be empty before it is disabled */
-	I915_WRITE_CTL(engine, 0);
+	ENGINE_WRITE(engine, RING_CTL, 0);
 
-	return (I915_READ_HEAD(engine) & HEAD_ADDR) == 0;
+	return (ENGINE_READ(engine, RING_HEAD) & HEAD_ADDR) == 0;
 }
 
 static int init_ring_common(struct intel_engine_cs *engine)
@@ -645,26 +644,26 @@ static int init_ring_common(struct intel_engine_cs *engine)
 	struct intel_ring *ring = engine->buffer;
 	int ret = 0;
 
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(engine->uncore, FORCEWAKE_ALL);
 
 	if (!stop_ring(engine)) {
 		/* G45 ring initialization often fails to reset head to zero */
 		DRM_DEBUG_DRIVER("%s head not reset to zero "
 				"ctl %08x head %08x tail %08x start %08x\n",
 				engine->name,
-				I915_READ_CTL(engine),
-				I915_READ_HEAD(engine),
-				I915_READ_TAIL(engine),
-				I915_READ_START(engine));
+				ENGINE_READ(engine, RING_CTL),
+				ENGINE_READ(engine, RING_HEAD),
+				ENGINE_READ(engine, RING_TAIL),
+				ENGINE_READ(engine, RING_START));
 
 		if (!stop_ring(engine)) {
 			DRM_ERROR("failed to set %s head to zero "
 				  "ctl %08x head %08x tail %08x start %08x\n",
 				  engine->name,
-				  I915_READ_CTL(engine),
-				  I915_READ_HEAD(engine),
-				  I915_READ_TAIL(engine),
-				  I915_READ_START(engine));
+				  ENGINE_READ(engine, RING_CTL),
+				  ENGINE_READ(engine, RING_HEAD),
+				  ENGINE_READ(engine, RING_TAIL),
+				  ENGINE_READ(engine, RING_START));
 			ret = -EIO;
 			goto out;
 		}
@@ -678,18 +677,18 @@ static int init_ring_common(struct intel_engine_cs *engine)
 	intel_engine_reset_breadcrumbs(engine);
 
 	/* Enforce ordering by reading HEAD register back */
-	I915_READ_HEAD(engine);
+	ENGINE_READ(engine, RING_HEAD);
 
 	/* Initialize the ring. This must happen _after_ we've cleared the ring
 	 * registers with the above sequence (the readback of the HEAD registers
 	 * also enforces ordering), otherwise the hw might lose the new ring
 	 * register values. */
-	I915_WRITE_START(engine, i915_ggtt_offset(ring->vma));
+	ENGINE_WRITE(engine, RING_START, i915_ggtt_offset(ring->vma));
 
 	/* WaClearRingBufHeadRegAtInit:ctg,elk */
-	if (I915_READ_HEAD(engine))
+	if (ENGINE_READ(engine, RING_HEAD))
 		DRM_DEBUG_DRIVER("%s initialization failed [head=%08x], fudging\n",
-				 engine->name, I915_READ_HEAD(engine));
+				 engine->name, ENGINE_READ(engine, RING_HEAD));
 
 	/* Check that the ring offsets point within the ring! */
 	GEM_BUG_ON(!intel_ring_offset_valid(ring, ring->head));
@@ -697,42 +696,44 @@ static int init_ring_common(struct intel_engine_cs *engine)
 	intel_ring_update_space(ring);
 
 	/* First wake the ring up to an empty/idle ring */
-	I915_WRITE_HEAD(engine, ring->head);
-	I915_WRITE_TAIL(engine, ring->head);
-	(void)I915_READ_TAIL(engine);
+	ENGINE_WRITE(engine, RING_HEAD, ring->head);
+	ENGINE_WRITE(engine, RING_TAIL, ring->head);
+	ENGINE_POSTING_READ(engine, RING_TAIL);
 
-	I915_WRITE_CTL(engine, RING_CTL_SIZE(ring->size) | RING_VALID);
+	ENGINE_WRITE(engine, RING_CTL, RING_CTL_SIZE(ring->size) | RING_VALID);
 
 	/* If the head is still not zero, the ring is dead */
-	if (intel_wait_for_register(dev_priv, RING_CTL(engine->mmio_base),
+	if (intel_wait_for_register(engine->uncore,
+				    RING_CTL(engine->mmio_base),
 				    RING_VALID, RING_VALID,
 				    50)) {
 		DRM_ERROR("%s initialization failed "
 			  "ctl %08x (valid? %d) head %08x [%08x] tail %08x [%08x] start %08x [expected %08x]\n",
 			  engine->name,
-			  I915_READ_CTL(engine),
-			  I915_READ_CTL(engine) & RING_VALID,
-			  I915_READ_HEAD(engine), ring->head,
-			  I915_READ_TAIL(engine), ring->tail,
-			  I915_READ_START(engine),
+			  ENGINE_READ(engine, RING_CTL),
+			  ENGINE_READ(engine, RING_CTL) & RING_VALID,
+			  ENGINE_READ(engine, RING_HEAD), ring->head,
+			  ENGINE_READ(engine, RING_TAIL), ring->tail,
+			  ENGINE_READ(engine, RING_START),
 			  i915_ggtt_offset(ring->vma));
 		ret = -EIO;
 		goto out;
 	}
 
 	if (INTEL_GEN(dev_priv) > 2)
-		I915_WRITE_MODE(engine, _MASKED_BIT_DISABLE(STOP_RING));
+		ENGINE_WRITE(engine,
+			     RING_MI_MODE, _MASKED_BIT_DISABLE(STOP_RING));
 
 	/* Now awake, let it get started */
 	if (ring->tail != ring->head) {
-		I915_WRITE_TAIL(engine, ring->tail);
-		(void)I915_READ_TAIL(engine);
+		ENGINE_WRITE(engine, RING_TAIL, ring->tail);
+		ENGINE_POSTING_READ(engine, RING_TAIL);
 	}
 
 	/* Papering over lost _interrupts_ immediately following the restart */
 	intel_engine_queue_breadcrumbs(engine);
 out:
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(engine->uncore, FORCEWAKE_ALL);
 
 	return ret;
 }
@@ -758,11 +759,6 @@ static void reset_ring(struct intel_engine_cs *engine, bool stalled)
 		}
 	}
 
-	GEM_TRACE("%s seqno=%d, current=%d, stalled? %s\n",
-		  engine->name,
-		  rq ? rq->global_seqno : 0,
-		  intel_engine_get_seqno(engine),
-		  yesno(stalled));
 	/*
 	 * The guilty request will get skipped on a hung engine.
 	 *
@@ -878,7 +874,7 @@ static int init_render_ring(struct intel_engine_cs *engine)
 		I915_WRITE(INSTPM, _MASKED_BIT_ENABLE(INSTPM_FORCE_ORDERING));
 
 	if (INTEL_GEN(dev_priv) >= 6)
-		I915_WRITE_IMR(engine, ~engine->irq_keep_mask);
+		ENGINE_WRITE(engine, RING_IMR, ~engine->irq_keep_mask);
 
 	return 0;
 }
@@ -892,18 +888,12 @@ static void cancel_requests(struct intel_engine_cs *engine)
 
 	/* Mark all submitted requests as skipped. */
 	list_for_each_entry(request, &engine->timeline.requests, link) {
-		GEM_BUG_ON(!request->global_seqno);
-
 		if (!i915_request_signaled(request))
 			dma_fence_set_error(&request->fence, -EIO);
 
 		i915_request_mark_complete(request);
 	}
 
-	intel_write_status_page(engine,
-				I915_GEM_HWS_INDEX,
-				intel_engine_last_submit(engine));
-
 	/* Remaining _unready_ requests will be nop'ed when submitted */
 
 	spin_unlock_irqrestore(&engine->timeline.lock, flags);
@@ -911,12 +901,10 @@ static void cancel_requests(struct intel_engine_cs *engine)
 
 static void i9xx_submit_request(struct i915_request *request)
 {
-	struct drm_i915_private *dev_priv = request->i915;
-
 	i915_request_submit(request);
 
-	I915_WRITE_TAIL(request->engine,
-			intel_ring_set_tail(request->ring, request->tail));
+	ENGINE_WRITE(request->engine, RING_TAIL,
+		     intel_ring_set_tail(request->ring, request->tail));
 }
 
 static u32 *i9xx_emit_breadcrumb(struct i915_request *rq, u32 *cs)
@@ -931,8 +919,8 @@ static u32 *i9xx_emit_breadcrumb(struct i915_request *rq, u32 *cs)
 	*cs++ = rq->fence.seqno;
 
 	*cs++ = MI_STORE_DWORD_INDEX;
-	*cs++ = I915_GEM_HWS_INDEX_ADDR;
-	*cs++ = rq->global_seqno;
+	*cs++ = I915_GEM_HWS_HANGCHECK_ADDR;
+	*cs++ = intel_engine_next_hangcheck_seqno(rq->engine);
 
 	*cs++ = MI_USER_INTERRUPT;
 
@@ -953,14 +941,14 @@ static u32 *gen5_emit_breadcrumb(struct i915_request *rq, u32 *cs)
 	*cs++ = MI_FLUSH;
 
 	*cs++ = MI_STORE_DWORD_INDEX;
-	*cs++ = I915_GEM_HWS_SEQNO_ADDR;
-	*cs++ = rq->fence.seqno;
+	*cs++ = I915_GEM_HWS_HANGCHECK_ADDR;
+	*cs++ = intel_engine_next_hangcheck_seqno(rq->engine);
 
 	BUILD_BUG_ON(GEN5_WA_STORES < 1);
 	for (i = 0; i < GEN5_WA_STORES; i++) {
 		*cs++ = MI_STORE_DWORD_INDEX;
-		*cs++ = I915_GEM_HWS_INDEX_ADDR;
-		*cs++ = rq->global_seqno;
+		*cs++ = I915_GEM_HWS_SEQNO_ADDR;
+		*cs++ = rq->fence.seqno;
 	}
 
 	*cs++ = MI_USER_INTERRUPT;
@@ -988,20 +976,16 @@ gen5_irq_disable(struct intel_engine_cs *engine)
 static void
 i9xx_irq_enable(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = engine->i915;
-
-	dev_priv->irq_mask &= ~engine->irq_enable_mask;
-	I915_WRITE(IMR, dev_priv->irq_mask);
-	POSTING_READ_FW(RING_IMR(engine->mmio_base));
+	engine->i915->irq_mask &= ~engine->irq_enable_mask;
+	intel_uncore_write(engine->uncore, GEN2_IMR, engine->i915->irq_mask);
+	intel_uncore_posting_read_fw(engine->uncore, GEN2_IMR);
 }
 
 static void
 i9xx_irq_disable(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = engine->i915;
-
-	dev_priv->irq_mask |= engine->irq_enable_mask;
-	I915_WRITE(IMR, dev_priv->irq_mask);
+	engine->i915->irq_mask |= engine->irq_enable_mask;
+	intel_uncore_write(engine->uncore, GEN2_IMR, engine->i915->irq_mask);
 }
 
 static void
@@ -1010,7 +994,7 @@ i8xx_irq_enable(struct intel_engine_cs *engine)
 	struct drm_i915_private *dev_priv = engine->i915;
 
 	dev_priv->irq_mask &= ~engine->irq_enable_mask;
-	I915_WRITE16(IMR, dev_priv->irq_mask);
+	I915_WRITE16(GEN2_IMR, dev_priv->irq_mask);
 	POSTING_READ16(RING_IMR(engine->mmio_base));
 }
 
@@ -1020,7 +1004,7 @@ i8xx_irq_disable(struct intel_engine_cs *engine)
 	struct drm_i915_private *dev_priv = engine->i915;
 
 	dev_priv->irq_mask |= engine->irq_enable_mask;
-	I915_WRITE16(IMR, dev_priv->irq_mask);
+	I915_WRITE16(GEN2_IMR, dev_priv->irq_mask);
 }
 
 static int
@@ -1041,47 +1025,38 @@ bsd_ring_flush(struct i915_request *rq, u32 mode)
 static void
 gen6_irq_enable(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = engine->i915;
-
-	I915_WRITE_IMR(engine,
-		       ~(engine->irq_enable_mask |
-			 engine->irq_keep_mask));
+	ENGINE_WRITE(engine, RING_IMR,
+		     ~(engine->irq_enable_mask | engine->irq_keep_mask));
 
 	/* Flush/delay to ensure the RING_IMR is active before the GT IMR */
-	POSTING_READ_FW(RING_IMR(engine->mmio_base));
+	ENGINE_POSTING_READ(engine, RING_IMR);
 
-	gen5_enable_gt_irq(dev_priv, engine->irq_enable_mask);
+	gen5_enable_gt_irq(engine->i915, engine->irq_enable_mask);
 }
 
 static void
 gen6_irq_disable(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = engine->i915;
-
-	I915_WRITE_IMR(engine, ~engine->irq_keep_mask);
-	gen5_disable_gt_irq(dev_priv, engine->irq_enable_mask);
+	ENGINE_WRITE(engine, RING_IMR, ~engine->irq_keep_mask);
+	gen5_disable_gt_irq(engine->i915, engine->irq_enable_mask);
 }
 
 static void
 hsw_vebox_irq_enable(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = engine->i915;
-
-	I915_WRITE_IMR(engine, ~engine->irq_enable_mask);
+	ENGINE_WRITE(engine, RING_IMR, ~engine->irq_enable_mask);
 
 	/* Flush/delay to ensure the RING_IMR is active before the GT IMR */
-	POSTING_READ_FW(RING_IMR(engine->mmio_base));
+	ENGINE_POSTING_READ(engine, RING_IMR);
 
-	gen6_unmask_pm_irq(dev_priv, engine->irq_enable_mask);
+	gen6_unmask_pm_irq(engine->i915, engine->irq_enable_mask);
 }
 
 static void
 hsw_vebox_irq_disable(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = engine->i915;
-
-	I915_WRITE_IMR(engine, ~0);
-	gen6_mask_pm_irq(dev_priv, engine->irq_enable_mask);
+	ENGINE_WRITE(engine, RING_IMR, ~0);
+	gen6_mask_pm_irq(engine->i915, engine->irq_enable_mask);
 }
 
 static int
@@ -1211,15 +1186,6 @@ int intel_ring_pin(struct intel_ring *ring)
 	else
 		flags |= PIN_HIGH;
 
-	if (!(vma->flags & I915_VMA_GLOBAL_BIND)) {
-		if (flags & PIN_MAPPABLE || map == I915_MAP_WC)
-			ret = i915_gem_object_set_to_gtt_domain(vma->obj, true);
-		else
-			ret = i915_gem_object_set_to_cpu_domain(vma->obj, true);
-		if (unlikely(ret))
-			goto unpin_timeline;
-	}
-
 	ret = i915_vma_pin(vma, 0, 0, flags);
 	if (unlikely(ret))
 		goto unpin_timeline;
@@ -1323,6 +1289,7 @@ intel_engine_create_ring(struct intel_engine_cs *engine,
 	if (!ring)
 		return ERR_PTR(-ENOMEM);
 
+	kref_init(&ring->ref);
 	INIT_LIST_HEAD(&ring->request_list);
 	ring->timeline = i915_timeline_get(timeline);
 
@@ -1347,9 +1314,9 @@ intel_engine_create_ring(struct intel_engine_cs *engine,
 	return ring;
 }
 
-void
-intel_ring_free(struct intel_ring *ring)
+void intel_ring_free(struct kref *ref)
 {
+	struct intel_ring *ring = container_of(ref, typeof(*ring), ref);
 	struct drm_i915_gem_object *obj = ring->vma->obj;
 
 	i915_vma_close(ring->vma);
@@ -1359,17 +1326,24 @@ intel_ring_free(struct intel_ring *ring)
 	kfree(ring);
 }
 
-static void intel_ring_context_destroy(struct intel_context *ce)
+static void __ring_context_fini(struct intel_context *ce)
 {
-	GEM_BUG_ON(ce->pin_count);
-
-	if (!ce->state)
-		return;
-
 	GEM_BUG_ON(i915_gem_object_is_active(ce->state->obj));
 	i915_gem_object_put(ce->state->obj);
 }
 
+static void ring_context_destroy(struct kref *ref)
+{
+	struct intel_context *ce = container_of(ref, typeof(*ce), ref);
+
+	GEM_BUG_ON(intel_context_is_pinned(ce));
+
+	if (ce->state)
+		__ring_context_fini(ce);
+
+	intel_context_free(ce);
+}
+
 static int __context_pin_ppgtt(struct i915_gem_context *ctx)
 {
 	struct i915_hw_ppgtt *ppgtt;
@@ -1400,17 +1374,6 @@ static int __context_pin(struct intel_context *ce)
 	if (!vma)
 		return 0;
 
-	/*
-	 * Clear this page out of any CPU caches for coherent swap-in/out.
-	 * We only want to do this on the first bind so that we do not stall
-	 * on an active context (which by nature is already on the GPU).
-	 */
-	if (!(vma->flags & I915_VMA_GLOBAL_BIND)) {
-		err = i915_gem_object_set_to_gtt_domain(vma->obj, true);
-		if (err)
-			return err;
-	}
-
 	err = i915_vma_pin(vma, 0, 0, PIN_GLOBAL | PIN_HIGH);
 	if (err)
 		return err;
@@ -1420,6 +1383,7 @@ static int __context_pin(struct intel_context *ce)
 	 * it cannot reclaim the object until we release it.
 	 */
 	vma->obj->pin_global++;
+	vma->obj->mm.dirty = true;
 
 	return 0;
 }
@@ -1436,12 +1400,10 @@ static void __context_unpin(struct intel_context *ce)
 	i915_vma_unpin(vma);
 }
 
-static void intel_ring_context_unpin(struct intel_context *ce)
+static void ring_context_unpin(struct intel_context *ce)
 {
 	__context_unpin_ppgtt(ce->gem_context);
 	__context_unpin(ce);
-
-	i915_gem_context_put(ce->gem_context);
 }
 
 static struct i915_vma *
@@ -1456,6 +1418,24 @@ alloc_context_vma(struct intel_engine_cs *engine)
 	if (IS_ERR(obj))
 		return ERR_CAST(obj);
 
+	/*
+	 * Try to make the context utilize L3 as well as LLC.
+	 *
+	 * On VLV we don't have L3 controls in the PTEs so we
+	 * shouldn't touch the cache level, especially as that
+	 * would make the object snooped which might have a
+	 * negative performance impact.
+	 *
+	 * Snooping is required on non-llc platforms in execlist
+	 * mode, but since all GGTT accesses use PAT entry 0 we
+	 * get snooping anyway regardless of cache_level.
+	 *
+	 * This is only applicable for Ivy Bridge devices since
+	 * later platforms don't have L3 control bits in the PTE.
+	 */
+	if (IS_IVYBRIDGE(i915))
+		i915_gem_object_set_cache_coherency(obj, I915_CACHE_L3_LLC);
+
 	if (engine->default_state) {
 		void *defaults, *vaddr;
 
@@ -1473,29 +1453,10 @@ alloc_context_vma(struct intel_engine_cs *engine)
 		}
 
 		memcpy(vaddr, defaults, engine->context_size);
-
 		i915_gem_object_unpin_map(engine->default_state);
-		i915_gem_object_unpin_map(obj);
-	}
 
-	/*
-	 * Try to make the context utilize L3 as well as LLC.
-	 *
-	 * On VLV we don't have L3 controls in the PTEs so we
-	 * shouldn't touch the cache level, especially as that
-	 * would make the object snooped which might have a
-	 * negative performance impact.
-	 *
-	 * Snooping is required on non-llc platforms in execlist
-	 * mode, but since all GGTT accesses use PAT entry 0 we
-	 * get snooping anyway regardless of cache_level.
-	 *
-	 * This is only applicable for Ivy Bridge devices since
-	 * later platforms don't have L3 control bits in the PTE.
-	 */
-	if (IS_IVYBRIDGE(i915)) {
-		/* Ignore any error, regard it as a simple optimisation */
-		i915_gem_object_set_cache_level(obj, I915_CACHE_L3_LLC);
+		i915_gem_object_flush_map(obj);
+		i915_gem_object_unpin_map(obj);
 	}
 
 	vma = i915_vma_instance(obj, &i915->ggtt.vm, NULL);
@@ -1513,69 +1474,52 @@ err_obj:
 	return ERR_PTR(err);
 }
 
-static struct intel_context *
-__ring_context_pin(struct intel_engine_cs *engine,
-		   struct i915_gem_context *ctx,
-		   struct intel_context *ce)
+static int ring_context_pin(struct intel_context *ce)
 {
+	struct intel_engine_cs *engine = ce->engine;
 	int err;
 
+	/* One ringbuffer to rule them all */
+	GEM_BUG_ON(!engine->buffer);
+	ce->ring = engine->buffer;
+
 	if (!ce->state && engine->context_size) {
 		struct i915_vma *vma;
 
 		vma = alloc_context_vma(engine);
-		if (IS_ERR(vma)) {
-			err = PTR_ERR(vma);
-			goto err;
-		}
+		if (IS_ERR(vma))
+			return PTR_ERR(vma);
 
 		ce->state = vma;
 	}
 
 	err = __context_pin(ce);
 	if (err)
-		goto err;
+		return err;
 
 	err = __context_pin_ppgtt(ce->gem_context);
 	if (err)
 		goto err_unpin;
 
-	i915_gem_context_get(ctx);
-
-	/* One ringbuffer to rule them all */
-	GEM_BUG_ON(!engine->buffer);
-	ce->ring = engine->buffer;
-
-	return ce;
+	return 0;
 
 err_unpin:
 	__context_unpin(ce);
-err:
-	ce->pin_count = 0;
-	return ERR_PTR(err);
+	return err;
 }
 
-static const struct intel_context_ops ring_context_ops = {
-	.unpin = intel_ring_context_unpin,
-	.destroy = intel_ring_context_destroy,
-};
-
-static struct intel_context *
-intel_ring_context_pin(struct intel_engine_cs *engine,
-		       struct i915_gem_context *ctx)
+static void ring_context_reset(struct intel_context *ce)
 {
-	struct intel_context *ce = to_intel_context(ctx, engine);
-
-	lockdep_assert_held(&ctx->i915->drm.struct_mutex);
-
-	if (likely(ce->pin_count++))
-		return ce;
-	GEM_BUG_ON(!ce->pin_count); /* no overflow please! */
+	intel_ring_reset(ce->ring, 0);
+}
 
-	ce->ops = &ring_context_ops;
+static const struct intel_context_ops ring_context_ops = {
+	.pin = ring_context_pin,
+	.unpin = ring_context_unpin,
 
-	return __ring_context_pin(engine, ctx, ce);
-}
+	.reset = ring_context_reset,
+	.destroy = ring_context_destroy,
+};
 
 static int intel_init_ring_buffer(struct intel_engine_cs *engine)
 {
@@ -1587,9 +1531,7 @@ static int intel_init_ring_buffer(struct intel_engine_cs *engine)
 	if (err)
 		return err;
 
-	timeline = i915_timeline_create(engine->i915,
-					engine->name,
-					engine->status_page.vma);
+	timeline = i915_timeline_create(engine->i915, engine->status_page.vma);
 	if (IS_ERR(timeline)) {
 		err = PTR_ERR(timeline);
 		goto err;
@@ -1621,7 +1563,7 @@ static int intel_init_ring_buffer(struct intel_engine_cs *engine)
 err_unpin:
 	intel_ring_unpin(ring);
 err_ring:
-	intel_ring_free(ring);
+	intel_ring_put(ring);
 err:
 	intel_engine_cleanup_common(engine);
 	return err;
@@ -1632,10 +1574,10 @@ void intel_engine_cleanup(struct intel_engine_cs *engine)
 	struct drm_i915_private *dev_priv = engine->i915;
 
 	WARN_ON(INTEL_GEN(dev_priv) > 2 &&
-		(I915_READ_MODE(engine) & MODE_IDLE) == 0);
+		(ENGINE_READ(engine, RING_MI_MODE) & MODE_IDLE) == 0);
 
 	intel_ring_unpin(engine->buffer);
-	intel_ring_free(engine->buffer);
+	intel_ring_put(engine->buffer);
 
 	if (engine->cleanup)
 		engine->cleanup(engine);
@@ -1646,16 +1588,6 @@ void intel_engine_cleanup(struct intel_engine_cs *engine)
 	kfree(engine);
 }
 
-void intel_legacy_submission_resume(struct drm_i915_private *dev_priv)
-{
-	struct intel_engine_cs *engine;
-	enum intel_engine_id id;
-
-	/* Restart from the beginning of the rings for convenience */
-	for_each_engine(engine, dev_priv, id)
-		intel_ring_reset(engine->buffer, 0);
-}
-
 static int load_pd_dir(struct i915_request *rq,
 		       const struct i915_hw_ppgtt *ppgtt)
 {
@@ -1667,11 +1599,11 @@ static int load_pd_dir(struct i915_request *rq,
 		return PTR_ERR(cs);
 
 	*cs++ = MI_LOAD_REGISTER_IMM(1);
-	*cs++ = i915_mmio_reg_offset(RING_PP_DIR_DCLV(engine));
+	*cs++ = i915_mmio_reg_offset(RING_PP_DIR_DCLV(engine->mmio_base));
 	*cs++ = PP_DIR_DCLV_2G;
 
 	*cs++ = MI_LOAD_REGISTER_IMM(1);
-	*cs++ = i915_mmio_reg_offset(RING_PP_DIR_BASE(engine));
+	*cs++ = i915_mmio_reg_offset(RING_PP_DIR_BASE(engine->mmio_base));
 	*cs++ = ppgtt->pd.base.ggtt_offset << 10;
 
 	intel_ring_advance(rq, cs);
@@ -1690,7 +1622,7 @@ static int flush_pd_dir(struct i915_request *rq)
 
 	/* Stall until the page table load is complete */
 	*cs++ = MI_STORE_REGISTER_MEM | MI_SRM_LRM_GLOBAL_GTT;
-	*cs++ = i915_mmio_reg_offset(RING_PP_DIR_BASE(engine));
+	*cs++ = i915_mmio_reg_offset(RING_PP_DIR_BASE(engine->mmio_base));
 	*cs++ = i915_scratch_offset(rq->i915);
 	*cs++ = MI_NOOP;
 
@@ -1703,8 +1635,8 @@ static inline int mi_set_context(struct i915_request *rq, u32 flags)
 	struct drm_i915_private *i915 = rq->i915;
 	struct intel_engine_cs *engine = rq->engine;
 	enum intel_engine_id id;
-	const int num_rings =
-		IS_HSW_GT1(i915) ? RUNTIME_INFO(i915)->num_rings - 1 : 0;
+	const int num_engines =
+		IS_HSW_GT1(i915) ? RUNTIME_INFO(i915)->num_engines - 1 : 0;
 	bool force_restore = false;
 	int len;
 	u32 *cs;
@@ -1718,7 +1650,7 @@ static inline int mi_set_context(struct i915_request *rq, u32 flags)
 
 	len = 4;
 	if (IS_GEN(i915, 7))
-		len += 2 + (num_rings ? 4*num_rings + 6 : 0);
+		len += 2 + (num_engines ? 4 * num_engines + 6 : 0);
 	if (flags & MI_FORCE_RESTORE) {
 		GEM_BUG_ON(flags & MI_RESTORE_INHIBIT);
 		flags &= ~MI_FORCE_RESTORE;
@@ -1733,10 +1665,10 @@ static inline int mi_set_context(struct i915_request *rq, u32 flags)
 	/* WaProgramMiArbOnOffAroundMiSetContext:ivb,vlv,hsw,bdw,chv */
 	if (IS_GEN(i915, 7)) {
 		*cs++ = MI_ARB_ON_OFF | MI_ARB_DISABLE;
-		if (num_rings) {
+		if (num_engines) {
 			struct intel_engine_cs *signaller;
 
-			*cs++ = MI_LOAD_REGISTER_IMM(num_rings);
+			*cs++ = MI_LOAD_REGISTER_IMM(num_engines);
 			for_each_engine(signaller, i915, id) {
 				if (signaller == engine)
 					continue;
@@ -1763,8 +1695,7 @@ static inline int mi_set_context(struct i915_request *rq, u32 flags)
 		 * placeholder we use to flush other contexts.
 		 */
 		*cs++ = MI_SET_CONTEXT;
-		*cs++ = i915_ggtt_offset(to_intel_context(i915->kernel_context,
-							  engine)->state) |
+		*cs++ = i915_ggtt_offset(engine->kernel_context->state) |
 			MI_MM_SPACE_GTT |
 			MI_RESTORE_INHIBIT;
 	}
@@ -1779,11 +1710,11 @@ static inline int mi_set_context(struct i915_request *rq, u32 flags)
 	*cs++ = MI_NOOP;
 
 	if (IS_GEN(i915, 7)) {
-		if (num_rings) {
+		if (num_engines) {
 			struct intel_engine_cs *signaller;
 			i915_reg_t last_reg = {}; /* keep gcc quiet */
 
-			*cs++ = MI_LOAD_REGISTER_IMM(num_rings);
+			*cs++ = MI_LOAD_REGISTER_IMM(num_engines);
 			for_each_engine(signaller, i915, id) {
 				if (signaller == engine)
 					continue;
@@ -1861,7 +1792,7 @@ static int switch_context(struct i915_request *rq)
 		 * explanation.
 		 */
 		loops = 1;
-		if (engine->id == BCS && IS_VALLEYVIEW(engine->i915))
+		if (engine->id == BCS0 && IS_VALLEYVIEW(engine->i915))
 			loops = 32;
 
 		do {
@@ -1870,15 +1801,15 @@ static int switch_context(struct i915_request *rq)
 				goto err;
 		} while (--loops);
 
-		if (intel_engine_flag(engine) & ppgtt->pd_dirty_rings) {
-			unwind_mm = intel_engine_flag(engine);
-			ppgtt->pd_dirty_rings &= ~unwind_mm;
+		if (ppgtt->pd_dirty_engines & engine->mask) {
+			unwind_mm = engine->mask;
+			ppgtt->pd_dirty_engines &= ~unwind_mm;
 			hw_flags = MI_FORCE_RESTORE;
 		}
 	}
 
 	if (rq->hw_context->state) {
-		GEM_BUG_ON(engine->id != RCS);
+		GEM_BUG_ON(engine->id != RCS0);
 
 		/*
 		 * The kernel context(s) is treated as pure scratch and is not
@@ -1938,7 +1869,7 @@ static int switch_context(struct i915_request *rq)
 
 err_mm:
 	if (unwind_mm)
-		ppgtt->pd_dirty_rings |= unwind_mm;
+		ppgtt->pd_dirty_engines |= unwind_mm;
 err:
 	return ret;
 }
@@ -1947,7 +1878,7 @@ static int ring_request_alloc(struct i915_request *request)
 {
 	int ret;
 
-	GEM_BUG_ON(!request->hw_context->pin_count);
+	GEM_BUG_ON(!intel_context_is_pinned(request->hw_context));
 	GEM_BUG_ON(request->timeline->has_initial_breadcrumb);
 
 	/*
@@ -2108,23 +2039,23 @@ int intel_ring_cacheline_align(struct i915_request *rq)
 
 static void gen6_bsd_submit_request(struct i915_request *request)
 {
-	struct drm_i915_private *dev_priv = request->i915;
+	struct intel_uncore *uncore = request->engine->uncore;
 
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(uncore, FORCEWAKE_ALL);
 
        /* Every tail move must follow the sequence below */
 
 	/* Disable notification that the ring is IDLE. The GT
 	 * will then assume that it is busy and bring it out of rc6.
 	 */
-	I915_WRITE_FW(GEN6_BSD_SLEEP_PSMI_CONTROL,
-		      _MASKED_BIT_ENABLE(GEN6_BSD_SLEEP_MSG_DISABLE));
+	intel_uncore_write_fw(uncore, GEN6_BSD_SLEEP_PSMI_CONTROL,
+			      _MASKED_BIT_ENABLE(GEN6_BSD_SLEEP_MSG_DISABLE));
 
 	/* Clear the context id. Here be magic! */
-	I915_WRITE64_FW(GEN6_BSD_RNCID, 0x0);
+	intel_uncore_write64_fw(uncore, GEN6_BSD_RNCID, 0x0);
 
 	/* Wait for the ring not to be idle, i.e. for it to wake up. */
-	if (__intel_wait_for_register_fw(dev_priv,
+	if (__intel_wait_for_register_fw(uncore,
 					 GEN6_BSD_SLEEP_PSMI_CONTROL,
 					 GEN6_BSD_SLEEP_INDICATOR,
 					 0,
@@ -2137,10 +2068,10 @@ static void gen6_bsd_submit_request(struct i915_request *request)
 	/* Let the ring send IDLE messages to the GT again,
 	 * and so let it sleep to conserve power when idle.
 	 */
-	I915_WRITE_FW(GEN6_BSD_SLEEP_PSMI_CONTROL,
-		      _MASKED_BIT_DISABLE(GEN6_BSD_SLEEP_MSG_DISABLE));
+	intel_uncore_write_fw(uncore, GEN6_BSD_SLEEP_PSMI_CONTROL,
+			      _MASKED_BIT_DISABLE(GEN6_BSD_SLEEP_MSG_DISABLE));
 
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(uncore, FORCEWAKE_ALL);
 }
 
 static int mi_flush_dw(struct i915_request *rq, u32 flags)
@@ -2282,7 +2213,7 @@ static void intel_ring_default_vfuncs(struct drm_i915_private *dev_priv,
 	engine->reset.reset = reset_ring;
 	engine->reset.finish = reset_finish;
 
-	engine->context_pin = intel_ring_context_pin;
+	engine->cops = &ring_context_ops;
 	engine->request_alloc = ring_request_alloc;
 
 	/*
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
index 710ffb221775..72c7c337ace9 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.h
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
@@ -6,22 +6,20 @@
 
 #include <linux/hashtable.h>
 #include <linux/irq_work.h>
+#include <linux/random.h>
 #include <linux/seqlock.h>
 
 #include "i915_gem_batch_pool.h"
-
-#include "i915_reg.h"
 #include "i915_pmu.h"
+#include "i915_reg.h"
 #include "i915_request.h"
 #include "i915_selftest.h"
 #include "i915_timeline.h"
+#include "intel_engine_types.h"
 #include "intel_gpu_commands.h"
 #include "intel_workarounds.h"
 
 struct drm_printer;
-struct i915_sched_attr;
-
-#define I915_CMD_HASH_ORDER 9
 
 /* Early gen2 devices have a cacheline of just 32 bytes, using 64 is overkill,
  * but keeps the logic simple. Indeed, the whole purpose of this macro is just
@@ -31,28 +29,44 @@ struct i915_sched_attr;
 #define CACHELINE_BYTES 64
 #define CACHELINE_DWORDS (CACHELINE_BYTES / sizeof(u32))
 
-struct intel_hw_status_page {
-	struct i915_vma *vma;
-	u32 *addr;
-};
+/*
+ * The register defines to be used with the following macros need to accept a
+ * base param, e.g:
+ *
+ * REG_FOO(base) _MMIO((base) + <relative offset>)
+ * ENGINE_READ(engine, REG_FOO);
+ *
+ * register arrays are to be defined and accessed as follows:
+ *
+ * REG_BAR(base, i) _MMIO((base) + <relative offset> + (i) * <shift>)
+ * ENGINE_READ_IDX(engine, REG_BAR, i)
+ */
+
+#define __ENGINE_REG_OP(op__, engine__, ...) \
+	intel_uncore_##op__((engine__)->uncore, __VA_ARGS__)
 
-#define I915_READ_TAIL(engine) I915_READ(RING_TAIL((engine)->mmio_base))
-#define I915_WRITE_TAIL(engine, val) I915_WRITE(RING_TAIL((engine)->mmio_base), val)
+#define __ENGINE_READ_OP(op__, engine__, reg__) \
+	__ENGINE_REG_OP(op__, (engine__), reg__((engine__)->mmio_base))
 
-#define I915_READ_START(engine) I915_READ(RING_START((engine)->mmio_base))
-#define I915_WRITE_START(engine, val) I915_WRITE(RING_START((engine)->mmio_base), val)
+#define ENGINE_READ16(...)	__ENGINE_READ_OP(read16, __VA_ARGS__)
+#define ENGINE_READ(...)	__ENGINE_READ_OP(read, __VA_ARGS__)
+#define ENGINE_READ_FW(...)	__ENGINE_READ_OP(read_fw, __VA_ARGS__)
+#define ENGINE_POSTING_READ(...) __ENGINE_READ_OP(posting_read, __VA_ARGS__)
 
-#define I915_READ_HEAD(engine)  I915_READ(RING_HEAD((engine)->mmio_base))
-#define I915_WRITE_HEAD(engine, val) I915_WRITE(RING_HEAD((engine)->mmio_base), val)
+#define ENGINE_READ64(engine__, lower_reg__, upper_reg__) \
+	__ENGINE_REG_OP(read64_2x32, (engine__), \
+			lower_reg__((engine__)->mmio_base), \
+			upper_reg__((engine__)->mmio_base))
 
-#define I915_READ_CTL(engine) I915_READ(RING_CTL((engine)->mmio_base))
-#define I915_WRITE_CTL(engine, val) I915_WRITE(RING_CTL((engine)->mmio_base), val)
+#define ENGINE_READ_IDX(engine__, reg__, idx__) \
+	__ENGINE_REG_OP(read, (engine__), reg__((engine__)->mmio_base, (idx__)))
 
-#define I915_READ_IMR(engine) I915_READ(RING_IMR((engine)->mmio_base))
-#define I915_WRITE_IMR(engine, val) I915_WRITE(RING_IMR((engine)->mmio_base), val)
+#define __ENGINE_WRITE_OP(op__, engine__, reg__, val__) \
+	__ENGINE_REG_OP(op__, (engine__), reg__((engine__)->mmio_base), (val__))
 
-#define I915_READ_MODE(engine) I915_READ(RING_MI_MODE((engine)->mmio_base))
-#define I915_WRITE_MODE(engine, val) I915_WRITE(RING_MI_MODE((engine)->mmio_base), val)
+#define ENGINE_WRITE16(...)	__ENGINE_WRITE_OP(write16, __VA_ARGS__)
+#define ENGINE_WRITE(...)	__ENGINE_WRITE_OP(write, __VA_ARGS__)
+#define ENGINE_WRITE_FW(...)	__ENGINE_WRITE_OP(write_fw, __VA_ARGS__)
 
 /* seqno size is actually only a uint32, but since we plan to use MI_FLUSH_DW to
  * do the writes, and that must have qw aligned offsets, simply pretend it's 8b.
@@ -90,506 +104,7 @@ hangcheck_action_to_str(const enum intel_engine_hangcheck_action a)
 	return "unknown";
 }
 
-#define I915_MAX_SLICES	3
-#define I915_MAX_SUBSLICES 8
-
-#define instdone_slice_mask(dev_priv__) \
-	(IS_GEN(dev_priv__, 7) ? \
-	 1 : RUNTIME_INFO(dev_priv__)->sseu.slice_mask)
-
-#define instdone_subslice_mask(dev_priv__) \
-	(IS_GEN(dev_priv__, 7) ? \
-	 1 : RUNTIME_INFO(dev_priv__)->sseu.subslice_mask[0])
-
-#define for_each_instdone_slice_subslice(dev_priv__, slice__, subslice__) \
-	for ((slice__) = 0, (subslice__) = 0; \
-	     (slice__) < I915_MAX_SLICES; \
-	     (subslice__) = ((subslice__) + 1) < I915_MAX_SUBSLICES ? (subslice__) + 1 : 0, \
-	       (slice__) += ((subslice__) == 0)) \
-		for_each_if((BIT(slice__) & instdone_slice_mask(dev_priv__)) && \
-			    (BIT(subslice__) & instdone_subslice_mask(dev_priv__)))
-
-struct intel_instdone {
-	u32 instdone;
-	/* The following exist only in the RCS engine */
-	u32 slice_common;
-	u32 sampler[I915_MAX_SLICES][I915_MAX_SUBSLICES];
-	u32 row[I915_MAX_SLICES][I915_MAX_SUBSLICES];
-};
-
-struct intel_engine_hangcheck {
-	u64 acthd;
-	u32 seqno;
-	unsigned long action_timestamp;
-	struct intel_instdone instdone;
-};
-
-struct intel_ring {
-	struct i915_vma *vma;
-	void *vaddr;
-
-	struct i915_timeline *timeline;
-	struct list_head request_list;
-	struct list_head active_link;
-
-	u32 head;
-	u32 tail;
-	u32 emit;
-
-	u32 space;
-	u32 size;
-	u32 effective_size;
-};
-
-struct i915_gem_context;
-struct drm_i915_reg_table;
-
-/*
- * we use a single page to load ctx workarounds so all of these
- * values are referred in terms of dwords
- *
- * struct i915_wa_ctx_bb:
- *  offset: specifies batch starting position, also helpful in case
- *    if we want to have multiple batches at different offsets based on
- *    some criteria. It is not a requirement at the moment but provides
- *    an option for future use.
- *  size: size of the batch in DWORDS
- */
-struct i915_ctx_workarounds {
-	struct i915_wa_ctx_bb {
-		u32 offset;
-		u32 size;
-	} indirect_ctx, per_ctx;
-	struct i915_vma *vma;
-};
-
-struct i915_request;
-
-#define I915_MAX_VCS	4
-#define I915_MAX_VECS	2
-
-/*
- * Engine IDs definitions.
- * Keep instances of the same type engine together.
- */
-enum intel_engine_id {
-	RCS = 0,
-	BCS,
-	VCS,
-	VCS2,
-	VCS3,
-	VCS4,
-#define _VCS(n) (VCS + (n))
-	VECS,
-	VECS2
-#define _VECS(n) (VECS + (n))
-};
-
-struct i915_priolist {
-	struct list_head requests[I915_PRIORITY_COUNT];
-	struct rb_node node;
-	unsigned long used;
-	int priority;
-};
-
-#define priolist_for_each_request(it, plist, idx) \
-	for (idx = 0; idx < ARRAY_SIZE((plist)->requests); idx++) \
-		list_for_each_entry(it, &(plist)->requests[idx], sched.link)
-
-#define priolist_for_each_request_consume(it, n, plist, idx) \
-	for (; (idx = ffs((plist)->used)); (plist)->used &= ~BIT(idx - 1)) \
-		list_for_each_entry_safe(it, n, \
-					 &(plist)->requests[idx - 1], \
-					 sched.link)
-
-struct st_preempt_hang {
-	struct completion completion;
-	unsigned int count;
-	bool inject_hang;
-};
-
-/**
- * struct intel_engine_execlists - execlist submission queue and port state
- *
- * The struct intel_engine_execlists represents the combined logical state of
- * driver and the hardware state for execlist mode of submission.
- */
-struct intel_engine_execlists {
-	/**
-	 * @tasklet: softirq tasklet for bottom handler
-	 */
-	struct tasklet_struct tasklet;
-
-	/**
-	 * @default_priolist: priority list for I915_PRIORITY_NORMAL
-	 */
-	struct i915_priolist default_priolist;
-
-	/**
-	 * @no_priolist: priority lists disabled
-	 */
-	bool no_priolist;
-
-	/**
-	 * @submit_reg: gen-specific execlist submission register
-	 * set to the ExecList Submission Port (elsp) register pre-Gen11 and to
-	 * the ExecList Submission Queue Contents register array for Gen11+
-	 */
-	u32 __iomem *submit_reg;
-
-	/**
-	 * @ctrl_reg: the enhanced execlists control register, used to load the
-	 * submit queue on the HW and to request preemptions to idle
-	 */
-	u32 __iomem *ctrl_reg;
-
-	/**
-	 * @port: execlist port states
-	 *
-	 * For each hardware ELSP (ExecList Submission Port) we keep
-	 * track of the last request and the number of times we submitted
-	 * that port to hw. We then count the number of times the hw reports
-	 * a context completion or preemption. As only one context can
-	 * be active on hw, we limit resubmission of context to port[0]. This
-	 * is called Lite Restore, of the context.
-	 */
-	struct execlist_port {
-		/**
-		 * @request_count: combined request and submission count
-		 */
-		struct i915_request *request_count;
-#define EXECLIST_COUNT_BITS 2
-#define port_request(p) ptr_mask_bits((p)->request_count, EXECLIST_COUNT_BITS)
-#define port_count(p) ptr_unmask_bits((p)->request_count, EXECLIST_COUNT_BITS)
-#define port_pack(rq, count) ptr_pack_bits(rq, count, EXECLIST_COUNT_BITS)
-#define port_unpack(p, count) ptr_unpack_bits((p)->request_count, count, EXECLIST_COUNT_BITS)
-#define port_set(p, packed) ((p)->request_count = (packed))
-#define port_isset(p) ((p)->request_count)
-#define port_index(p, execlists) ((p) - (execlists)->port)
-
-		/**
-		 * @context_id: context ID for port
-		 */
-		GEM_DEBUG_DECL(u32 context_id);
-
-#define EXECLIST_MAX_PORTS 2
-	} port[EXECLIST_MAX_PORTS];
-
-	/**
-	 * @active: is the HW active? We consider the HW as active after
-	 * submitting any context for execution and until we have seen the
-	 * last context completion event. After that, we do not expect any
-	 * more events until we submit, and so can park the HW.
-	 *
-	 * As we have a small number of different sources from which we feed
-	 * the HW, we track the state of each inside a single bitfield.
-	 */
-	unsigned int active;
-#define EXECLISTS_ACTIVE_USER 0
-#define EXECLISTS_ACTIVE_PREEMPT 1
-#define EXECLISTS_ACTIVE_HWACK 2
-
-	/**
-	 * @port_mask: number of execlist ports - 1
-	 */
-	unsigned int port_mask;
-
-	/**
-	 * @queue_priority_hint: Highest pending priority.
-	 *
-	 * When we add requests into the queue, or adjust the priority of
-	 * executing requests, we compute the maximum priority of those
-	 * pending requests. We can then use this value to determine if
-	 * we need to preempt the executing requests to service the queue.
-	 * However, since the we may have recorded the priority of an inflight
-	 * request we wanted to preempt but since completed, at the time of
-	 * dequeuing the priority hint may no longer may match the highest
-	 * available request priority.
-	 */
-	int queue_priority_hint;
-
-	/**
-	 * @queue: queue of requests, in priority lists
-	 */
-	struct rb_root_cached queue;
-
-	/**
-	 * @csb_write: control register for Context Switch buffer
-	 *
-	 * Note this register may be either mmio or HWSP shadow.
-	 */
-	u32 *csb_write;
-
-	/**
-	 * @csb_status: status array for Context Switch buffer
-	 *
-	 * Note these register may be either mmio or HWSP shadow.
-	 */
-	u32 *csb_status;
-
-	/**
-	 * @preempt_complete_status: expected CSB upon completing preemption
-	 */
-	u32 preempt_complete_status;
-
-	/**
-	 * @csb_head: context status buffer head
-	 */
-	u8 csb_head;
-
-	I915_SELFTEST_DECLARE(struct st_preempt_hang preempt_hang;)
-};
-
-#define INTEL_ENGINE_CS_MAX_NAME 8
-
-struct intel_engine_cs {
-	struct drm_i915_private *i915;
-	char name[INTEL_ENGINE_CS_MAX_NAME];
-
-	enum intel_engine_id id;
-	unsigned int hw_id;
-	unsigned int guc_id;
-
-	u8 uabi_id;
-	u8 uabi_class;
-
-	u8 class;
-	u8 instance;
-	u32 context_size;
-	u32 mmio_base;
-
-	struct intel_ring *buffer;
-
-	struct i915_timeline timeline;
-
-	struct drm_i915_gem_object *default_state;
-	void *pinned_default_state;
-
-	/* Rather than have every client wait upon all user interrupts,
-	 * with the herd waking after every interrupt and each doing the
-	 * heavyweight seqno dance, we delegate the task (of being the
-	 * bottom-half of the user interrupt) to the first client. After
-	 * every interrupt, we wake up one client, who does the heavyweight
-	 * coherent seqno read and either goes back to sleep (if incomplete),
-	 * or wakes up all the completed clients in parallel, before then
-	 * transferring the bottom-half status to the next client in the queue.
-	 *
-	 * Compared to walking the entire list of waiters in a single dedicated
-	 * bottom-half, we reduce the latency of the first waiter by avoiding
-	 * a context switch, but incur additional coherent seqno reads when
-	 * following the chain of request breadcrumbs. Since it is most likely
-	 * that we have a single client waiting on each seqno, then reducing
-	 * the overhead of waking that client is much preferred.
-	 */
-	struct intel_breadcrumbs {
-		spinlock_t irq_lock;
-		struct list_head signalers;
-
-		struct irq_work irq_work; /* for use from inside irq_lock */
-
-		unsigned int irq_enabled;
-
-		bool irq_armed;
-	} breadcrumbs;
-
-	struct {
-		/**
-		 * @enable: Bitmask of enable sample events on this engine.
-		 *
-		 * Bits correspond to sample event types, for instance
-		 * I915_SAMPLE_QUEUED is bit 0 etc.
-		 */
-		u32 enable;
-		/**
-		 * @enable_count: Reference count for the enabled samplers.
-		 *
-		 * Index number corresponds to @enum drm_i915_pmu_engine_sample.
-		 */
-		unsigned int enable_count[I915_ENGINE_SAMPLE_COUNT];
-		/**
-		 * @sample: Counter values for sampling events.
-		 *
-		 * Our internal timer stores the current counters in this field.
-		 *
-		 * Index number corresponds to @enum drm_i915_pmu_engine_sample.
-		 */
-		struct i915_pmu_sample sample[I915_ENGINE_SAMPLE_COUNT];
-	} pmu;
-
-	/*
-	 * A pool of objects to use as shadow copies of client batch buffers
-	 * when the command parser is enabled. Prevents the client from
-	 * modifying the batch contents after software parsing.
-	 */
-	struct i915_gem_batch_pool batch_pool;
-
-	struct intel_hw_status_page status_page;
-	struct i915_ctx_workarounds wa_ctx;
-	struct i915_wa_list ctx_wa_list;
-	struct i915_wa_list wa_list;
-	struct i915_wa_list whitelist;
-
-	u32             irq_keep_mask; /* always keep these interrupts */
-	u32		irq_enable_mask; /* bitmask to enable ring interrupt */
-	void		(*irq_enable)(struct intel_engine_cs *engine);
-	void		(*irq_disable)(struct intel_engine_cs *engine);
-
-	int		(*init_hw)(struct intel_engine_cs *engine);
-
-	struct {
-		void (*prepare)(struct intel_engine_cs *engine);
-		void (*reset)(struct intel_engine_cs *engine, bool stalled);
-		void (*finish)(struct intel_engine_cs *engine);
-	} reset;
-
-	void		(*park)(struct intel_engine_cs *engine);
-	void		(*unpark)(struct intel_engine_cs *engine);
-
-	void		(*set_default_submission)(struct intel_engine_cs *engine);
-
-	struct intel_context *(*context_pin)(struct intel_engine_cs *engine,
-					     struct i915_gem_context *ctx);
-
-	int		(*request_alloc)(struct i915_request *rq);
-	int		(*init_context)(struct i915_request *rq);
-
-	int		(*emit_flush)(struct i915_request *request, u32 mode);
-#define EMIT_INVALIDATE	BIT(0)
-#define EMIT_FLUSH	BIT(1)
-#define EMIT_BARRIER	(EMIT_INVALIDATE | EMIT_FLUSH)
-	int		(*emit_bb_start)(struct i915_request *rq,
-					 u64 offset, u32 length,
-					 unsigned int dispatch_flags);
-#define I915_DISPATCH_SECURE BIT(0)
-#define I915_DISPATCH_PINNED BIT(1)
-	int		 (*emit_init_breadcrumb)(struct i915_request *rq);
-	u32		*(*emit_fini_breadcrumb)(struct i915_request *rq,
-						 u32 *cs);
-	unsigned int	emit_fini_breadcrumb_dw;
-
-	/* Pass the request to the hardware queue (e.g. directly into
-	 * the legacy ringbuffer or to the end of an execlist).
-	 *
-	 * This is called from an atomic context with irqs disabled; must
-	 * be irq safe.
-	 */
-	void		(*submit_request)(struct i915_request *rq);
-
-	/*
-	 * Call when the priority on a request has changed and it and its
-	 * dependencies may need rescheduling. Note the request itself may
-	 * not be ready to run!
-	 */
-	void		(*schedule)(struct i915_request *request,
-				    const struct i915_sched_attr *attr);
-
-	/*
-	 * Cancel all requests on the hardware, or queued for execution.
-	 * This should only cancel the ready requests that have been
-	 * submitted to the engine (via the engine->submit_request callback).
-	 * This is called when marking the device as wedged.
-	 */
-	void		(*cancel_requests)(struct intel_engine_cs *engine);
-
-	void		(*cleanup)(struct intel_engine_cs *engine);
-
-	struct intel_engine_execlists execlists;
-
-	/* Contexts are pinned whilst they are active on the GPU. The last
-	 * context executed remains active whilst the GPU is idle - the
-	 * switch away and write to the context object only occurs on the
-	 * next execution.  Contexts are only unpinned on retirement of the
-	 * following request ensuring that we can always write to the object
-	 * on the context switch even after idling. Across suspend, we switch
-	 * to the kernel context and trash it as the save may not happen
-	 * before the hardware is powered down.
-	 */
-	struct intel_context *last_retired_context;
-
-	/* status_notifier: list of callbacks for context-switch changes */
-	struct atomic_notifier_head context_status_notifier;
-
-	struct intel_engine_hangcheck hangcheck;
-
-#define I915_ENGINE_NEEDS_CMD_PARSER BIT(0)
-#define I915_ENGINE_SUPPORTS_STATS   BIT(1)
-#define I915_ENGINE_HAS_PREEMPTION   BIT(2)
-	unsigned int flags;
-
-	/*
-	 * Table of commands the command parser needs to know about
-	 * for this engine.
-	 */
-	DECLARE_HASHTABLE(cmd_hash, I915_CMD_HASH_ORDER);
-
-	/*
-	 * Table of registers allowed in commands that read/write registers.
-	 */
-	const struct drm_i915_reg_table *reg_tables;
-	int reg_table_count;
-
-	/*
-	 * Returns the bitmask for the length field of the specified command.
-	 * Return 0 for an unrecognized/invalid command.
-	 *
-	 * If the command parser finds an entry for a command in the engine's
-	 * cmd_tables, it gets the command's length based on the table entry.
-	 * If not, it calls this function to determine the per-engine length
-	 * field encoding for the command (i.e. different opcode ranges use
-	 * certain bits to encode the command length in the header).
-	 */
-	u32 (*get_cmd_length_mask)(u32 cmd_header);
-
-	struct {
-		/**
-		 * @lock: Lock protecting the below fields.
-		 */
-		seqlock_t lock;
-		/**
-		 * @enabled: Reference count indicating number of listeners.
-		 */
-		unsigned int enabled;
-		/**
-		 * @active: Number of contexts currently scheduled in.
-		 */
-		unsigned int active;
-		/**
-		 * @enabled_at: Timestamp when busy stats were enabled.
-		 */
-		ktime_t enabled_at;
-		/**
-		 * @start: Timestamp of the last idle to active transition.
-		 *
-		 * Idle is defined as active == 0, active is active > 0.
-		 */
-		ktime_t start;
-		/**
-		 * @total: Total time this engine was busy.
-		 *
-		 * Accumulated time not counting the most recent block in cases
-		 * where engine is currently busy (active > 0).
-		 */
-		ktime_t total;
-	} stats;
-};
-
-static inline bool
-intel_engine_needs_cmd_parser(const struct intel_engine_cs *engine)
-{
-	return engine->flags & I915_ENGINE_NEEDS_CMD_PARSER;
-}
-
-static inline bool
-intel_engine_supports_stats(const struct intel_engine_cs *engine)
-{
-	return engine->flags & I915_ENGINE_SUPPORTS_STATS;
-}
-
-static inline bool
-intel_engine_has_preemption(const struct intel_engine_cs *engine)
-{
-	return engine->flags & I915_ENGINE_HAS_PREEMPTION;
-}
+void intel_engines_set_scheduler_caps(struct drm_i915_private *i915);
 
 static inline bool __execlists_need_preempt(int prio, int last)
 {
@@ -650,7 +165,7 @@ void execlists_user_end(struct intel_engine_execlists *execlists);
 void
 execlists_cancel_port_requests(struct intel_engine_execlists * const execlists);
 
-void
+struct i915_request *
 execlists_unwind_incomplete_requests(struct intel_engine_execlists *execlists);
 
 static inline unsigned int
@@ -674,12 +189,6 @@ execlists_port_complete(struct intel_engine_execlists * const execlists,
 	return port;
 }
 
-static inline unsigned int
-intel_engine_flag(const struct intel_engine_cs *engine)
-{
-	return BIT(engine->id);
-}
-
 static inline u32
 intel_read_status_page(const struct intel_engine_cs *engine, int reg)
 {
@@ -722,10 +231,10 @@ intel_write_status_page(struct intel_engine_cs *engine, int reg, u32 value)
  *
  * The area from dword 0x30 to 0x3ff is available for driver usage.
  */
-#define I915_GEM_HWS_INDEX		0x30
-#define I915_GEM_HWS_INDEX_ADDR		(I915_GEM_HWS_INDEX * sizeof(u32))
 #define I915_GEM_HWS_PREEMPT		0x32
 #define I915_GEM_HWS_PREEMPT_ADDR	(I915_GEM_HWS_PREEMPT * sizeof(u32))
+#define I915_GEM_HWS_HANGCHECK		0x34
+#define I915_GEM_HWS_HANGCHECK_ADDR	(I915_GEM_HWS_HANGCHECK * sizeof(u32))
 #define I915_GEM_HWS_SEQNO		0x40
 #define I915_GEM_HWS_SEQNO_ADDR		(I915_GEM_HWS_SEQNO * sizeof(u32))
 #define I915_GEM_HWS_SCRATCH		0x80
@@ -743,13 +252,22 @@ int intel_ring_pin(struct intel_ring *ring);
 void intel_ring_reset(struct intel_ring *ring, u32 tail);
 unsigned int intel_ring_update_space(struct intel_ring *ring);
 void intel_ring_unpin(struct intel_ring *ring);
-void intel_ring_free(struct intel_ring *ring);
+void intel_ring_free(struct kref *ref);
+
+static inline struct intel_ring *intel_ring_get(struct intel_ring *ring)
+{
+	kref_get(&ring->ref);
+	return ring;
+}
+
+static inline void intel_ring_put(struct intel_ring *ring)
+{
+	kref_put(&ring->ref, intel_ring_free);
+}
 
 void intel_engine_stop(struct intel_engine_cs *engine);
 void intel_engine_cleanup(struct intel_engine_cs *engine);
 
-void intel_legacy_submission_resume(struct drm_i915_private *dev_priv);
-
 int __must_check intel_ring_cacheline_align(struct i915_request *rq);
 
 u32 __must_check *intel_ring_begin(struct i915_request *rq, unsigned int n);
@@ -844,8 +362,6 @@ __intel_ring_space(unsigned int head, unsigned int tail, unsigned int size)
 	return (head - tail - CACHELINE_BYTES) & (size - 1);
 }
 
-void intel_engine_write_global_seqno(struct intel_engine_cs *engine, u32 seqno);
-
 int intel_engine_setup_common(struct intel_engine_cs *engine);
 int intel_engine_init_common(struct intel_engine_cs *engine);
 void intel_engine_cleanup_common(struct intel_engine_cs *engine);
@@ -863,44 +379,6 @@ void intel_engine_set_hwsp_writemask(struct intel_engine_cs *engine, u32 mask);
 u64 intel_engine_get_active_head(const struct intel_engine_cs *engine);
 u64 intel_engine_get_last_batch_head(const struct intel_engine_cs *engine);
 
-static inline u32 intel_engine_last_submit(struct intel_engine_cs *engine)
-{
-	/*
-	 * We are only peeking at the tail of the submit queue (and not the
-	 * queue itself) in order to gain a hint as to the current active
-	 * state of the engine. Callers are not expected to be taking
-	 * engine->timeline->lock, nor are they expected to be concerned
-	 * wtih serialising this hint with anything, so document it as
-	 * a hint and nothing more.
-	 */
-	return READ_ONCE(engine->timeline.seqno);
-}
-
-static inline u32 intel_engine_get_seqno(struct intel_engine_cs *engine)
-{
-	return intel_read_status_page(engine, I915_GEM_HWS_INDEX);
-}
-
-static inline bool intel_engine_signaled(struct intel_engine_cs *engine,
-					 u32 seqno)
-{
-	return i915_seqno_passed(intel_engine_get_seqno(engine), seqno);
-}
-
-static inline bool intel_engine_has_completed(struct intel_engine_cs *engine,
-					      u32 seqno)
-{
-	GEM_BUG_ON(!seqno);
-	return intel_engine_signaled(engine, seqno);
-}
-
-static inline bool intel_engine_has_started(struct intel_engine_cs *engine,
-					    u32 seqno)
-{
-	GEM_BUG_ON(!seqno);
-	return intel_engine_signaled(engine, seqno - 1);
-}
-
 void intel_engine_get_instdone(struct intel_engine_cs *engine,
 			       struct intel_instdone *instdone);
 
@@ -910,7 +388,7 @@ void intel_engine_fini_breadcrumbs(struct intel_engine_cs *engine);
 void intel_engine_pin_breadcrumbs_irq(struct intel_engine_cs *engine);
 void intel_engine_unpin_breadcrumbs_irq(struct intel_engine_cs *engine);
 
-bool intel_engine_signal_breadcrumbs(struct intel_engine_cs *engine);
+void intel_engine_signal_breadcrumbs(struct intel_engine_cs *engine);
 void intel_engine_disarm_breadcrumbs(struct intel_engine_cs *engine);
 
 static inline void
@@ -919,7 +397,7 @@ intel_engine_queue_breadcrumbs(struct intel_engine_cs *engine)
 	irq_work_queue(&engine->breadcrumbs.irq_work);
 }
 
-bool intel_engine_breadcrumbs_irq(struct intel_engine_cs *engine);
+void intel_engine_breadcrumbs_irq(struct intel_engine_cs *engine);
 
 void intel_engine_reset_breadcrumbs(struct intel_engine_cs *engine);
 void intel_engine_fini_breadcrumbs(struct intel_engine_cs *engine);
@@ -960,14 +438,14 @@ gen8_emit_ggtt_write_rcs(u32 *cs, u32 value, u32 gtt_offset, u32 flags)
 }
 
 static inline u32 *
-gen8_emit_ggtt_write(u32 *cs, u32 value, u32 gtt_offset)
+gen8_emit_ggtt_write(u32 *cs, u32 value, u32 gtt_offset, u32 flags)
 {
 	/* w/a: bit 5 needs to be zero for MI_FLUSH_DW address. */
 	GEM_BUG_ON(gtt_offset & (1 << 5));
 	/* Offset should be aligned to 8 bytes for both (QW/DW) write types */
 	GEM_BUG_ON(!IS_ALIGNED(gtt_offset, 8));
 
-	*cs++ = (MI_FLUSH_DW + 1) | MI_FLUSH_DW_OP_STOREDW;
+	*cs++ = (MI_FLUSH_DW + 1) | MI_FLUSH_DW_OP_STOREDW | flags;
 	*cs++ = gtt_offset | MI_FLUSH_DW_USE_GTT;
 	*cs++ = 0;
 	*cs++ = value;
@@ -983,11 +461,11 @@ static inline void intel_engine_reset(struct intel_engine_cs *engine,
 }
 
 void intel_engines_sanitize(struct drm_i915_private *i915, bool force);
+void intel_gt_resume(struct drm_i915_private *i915);
 
 bool intel_engine_is_idle(struct intel_engine_cs *engine);
 bool intel_engines_are_idle(struct drm_i915_private *dev_priv);
 
-bool intel_engine_has_kernel_context(const struct intel_engine_cs *engine);
 void intel_engine_lost_context(struct intel_engine_cs *engine);
 
 void intel_engines_park(struct drm_i915_private *i915);
@@ -1066,6 +544,9 @@ void intel_disable_engine_stats(struct intel_engine_cs *engine);
 
 ktime_t intel_engine_get_busy_time(struct intel_engine_cs *engine);
 
+struct i915_request *
+intel_engine_find_active_request(struct intel_engine_cs *engine);
+
 #if IS_ENABLED(CONFIG_DRM_I915_SELFTEST)
 
 static inline bool inject_preempt_hang(struct intel_engine_execlists *execlists)
@@ -1086,4 +567,17 @@ static inline bool inject_preempt_hang(struct intel_engine_execlists *execlists)
 
 #endif
 
+static inline u32
+intel_engine_next_hangcheck_seqno(struct intel_engine_cs *engine)
+{
+	return engine->hangcheck.next_seqno =
+		next_pseudo_random32(engine->hangcheck.next_seqno);
+}
+
+static inline u32
+intel_engine_get_hangcheck_seqno(struct intel_engine_cs *engine)
+{
+	return intel_read_status_page(engine, I915_GEM_HWS_HANGCHECK);
+}
+
 #endif /* _INTEL_RINGBUFFER_H_ */
diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
index 20c4434474e3..6150e35bf7b5 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.c
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
@@ -32,6 +32,10 @@
 #include <drm/drm_print.h>
 
 #include "i915_drv.h"
+#include "intel_cdclk.h"
+#include "intel_crt.h"
+#include "intel_csr.h"
+#include "intel_dp.h"
 #include "intel_drv.h"
 
 /**
@@ -147,7 +151,7 @@ static void cancel_intel_runtime_pm_wakeref(struct drm_i915_private *i915,
 		 rpm->debug.count, atomic_read(&rpm->wakeref_count))) {
 		char *buf;
 
-		buf = kmalloc(PAGE_SIZE, GFP_KERNEL);
+		buf = kmalloc(PAGE_SIZE, GFP_NOWAIT | __GFP_NOWARN);
 		if (!buf)
 			return;
 
@@ -183,7 +187,7 @@ __print_intel_runtime_pm_wakeref(struct drm_printer *p,
 	unsigned long i;
 	char *buf;
 
-	buf = kmalloc(PAGE_SIZE, GFP_KERNEL);
+	buf = kmalloc(PAGE_SIZE, GFP_NOWAIT | __GFP_NOWARN);
 	if (!buf)
 		return;
 
@@ -267,7 +271,9 @@ void print_intel_runtime_pm_wakeref(struct drm_i915_private *i915,
 		if (dbg.count <= alloc)
 			break;
 
-		s = krealloc(dbg.owners, dbg.count * sizeof(*s), GFP_KERNEL);
+		s = krealloc(dbg.owners,
+			     dbg.count * sizeof(*s),
+			     GFP_NOWAIT | __GFP_NOWARN);
 		if (!s)
 			goto out;
 
@@ -554,7 +560,7 @@ static void hsw_wait_for_power_well_enable(struct drm_i915_private *dev_priv,
 	int pw_idx = power_well->desc->hsw.idx;
 
 	/* Timeout for PW1:10 us, AUX:not specified, other PWs:20 us. */
-	WARN_ON(intel_wait_for_register(dev_priv,
+	WARN_ON(intel_wait_for_register(&dev_priv->uncore,
 					regs->driver,
 					HSW_PWR_WELL_CTL_STATE(pw_idx),
 					HSW_PWR_WELL_CTL_STATE(pw_idx),
@@ -609,7 +615,7 @@ static void gen9_wait_for_power_well_fuses(struct drm_i915_private *dev_priv,
 					   enum skl_power_gate pg)
 {
 	/* Timeout 5us for PG#0, for other PGs 1us */
-	WARN_ON(intel_wait_for_register(dev_priv, SKL_FUSE_STATUS,
+	WARN_ON(intel_wait_for_register(&dev_priv->uncore, SKL_FUSE_STATUS,
 					SKL_FUSE_PG_DIST_STATUS(pg),
 					SKL_FUSE_PG_DIST_STATUS(pg), 1));
 }
@@ -1510,7 +1516,7 @@ static void assert_chv_phy_status(struct drm_i915_private *dev_priv)
 	 * The PHY may be busy with some initial calibration and whatnot,
 	 * so the power state can take a while to actually change.
 	 */
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    DISPLAY_PHY_STATUS,
 				    phy_status_mask,
 				    phy_status,
@@ -1545,7 +1551,7 @@ static void chv_dpio_cmn_power_well_enable(struct drm_i915_private *dev_priv,
 	vlv_set_power_well(dev_priv, power_well, true);
 
 	/* Poll for phypwrgood signal */
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    DISPLAY_PHY_STATUS,
 				    PHY_POWERGOOD(phy),
 				    PHY_POWERGOOD(phy),
@@ -1749,7 +1755,7 @@ static bool chv_pipe_power_well_enabled(struct drm_i915_private *dev_priv,
 
 	mutex_lock(&dev_priv->pcu_lock);
 
-	state = vlv_punit_read(dev_priv, PUNIT_REG_DSPFREQ) & DP_SSS_MASK(pipe);
+	state = vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM) & DP_SSS_MASK(pipe);
 	/*
 	 * We only ever set the power-on and power-gate states, anything
 	 * else is unexpected.
@@ -1761,7 +1767,7 @@ static bool chv_pipe_power_well_enabled(struct drm_i915_private *dev_priv,
 	 * A transient state at this point would mean some unexpected party
 	 * is poking at the power controls too.
 	 */
-	ctrl = vlv_punit_read(dev_priv, PUNIT_REG_DSPFREQ) & DP_SSC_MASK(pipe);
+	ctrl = vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM) & DP_SSC_MASK(pipe);
 	WARN_ON(ctrl << 16 != state);
 
 	mutex_unlock(&dev_priv->pcu_lock);
@@ -1782,20 +1788,20 @@ static void chv_set_pipe_power_well(struct drm_i915_private *dev_priv,
 	mutex_lock(&dev_priv->pcu_lock);
 
 #define COND \
-	((vlv_punit_read(dev_priv, PUNIT_REG_DSPFREQ) & DP_SSS_MASK(pipe)) == state)
+	((vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM) & DP_SSS_MASK(pipe)) == state)
 
 	if (COND)
 		goto out;
 
-	ctrl = vlv_punit_read(dev_priv, PUNIT_REG_DSPFREQ);
+	ctrl = vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM);
 	ctrl &= ~DP_SSC_MASK(pipe);
 	ctrl |= enable ? DP_SSC_PWR_ON(pipe) : DP_SSC_PWR_GATE(pipe);
-	vlv_punit_write(dev_priv, PUNIT_REG_DSPFREQ, ctrl);
+	vlv_punit_write(dev_priv, PUNIT_REG_DSPSSPM, ctrl);
 
 	if (wait_for(COND, 100))
 		DRM_ERROR("timeout setting power well state %08x (%08x)\n",
 			  state,
-			  vlv_punit_read(dev_priv, PUNIT_REG_DSPFREQ));
+			  vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM));
 
 #undef COND
 
@@ -3431,7 +3437,7 @@ int intel_power_domains_init(struct drm_i915_private *dev_priv)
 	 * The enabling order will be from lower to higher indexed wells,
 	 * the disabling order is reversed.
 	 */
-	if (IS_ICELAKE(dev_priv)) {
+	if (IS_GEN(dev_priv, 11)) {
 		err = set_power_wells(power_domains, icl_power_wells);
 	} else if (IS_CANNONLAKE(dev_priv)) {
 		err = set_power_wells(power_domains, cnl_power_wells);
@@ -3565,7 +3571,11 @@ static void icl_dbuf_enable(struct drm_i915_private *dev_priv)
 	    !(I915_READ(DBUF_CTL_S2) & DBUF_POWER_STATE))
 		DRM_ERROR("DBuf power enable timeout\n");
 	else
-		dev_priv->wm.skl_hw.ddb.enabled_slices = 2;
+		/*
+		 * FIXME: for now pretend that we only have 1 slice, see
+		 * intel_enabled_dbuf_slices_num().
+		 */
+		dev_priv->wm.skl_hw.ddb.enabled_slices = 1;
 }
 
 static void icl_dbuf_disable(struct drm_i915_private *dev_priv)
@@ -3580,7 +3590,11 @@ static void icl_dbuf_disable(struct drm_i915_private *dev_priv)
 	    (I915_READ(DBUF_CTL_S2) & DBUF_POWER_STATE))
 		DRM_ERROR("DBuf power disable timeout!\n");
 	else
-		dev_priv->wm.skl_hw.ddb.enabled_slices = 0;
+		/*
+		 * FIXME: for now pretend that the first slice is always
+		 * enabled, see intel_enabled_dbuf_slices_num().
+		 */
+		dev_priv->wm.skl_hw.ddb.enabled_slices = 1;
 }
 
 static void icl_mbus_init(struct drm_i915_private *dev_priv)
@@ -3641,7 +3655,7 @@ static void skl_display_core_init(struct drm_i915_private *dev_priv,
 
 	mutex_unlock(&power_domains->lock);
 
-	skl_init_cdclk(dev_priv);
+	intel_cdclk_init(dev_priv);
 
 	gen9_dbuf_enable(dev_priv);
 
@@ -3658,7 +3672,7 @@ static void skl_display_core_uninit(struct drm_i915_private *dev_priv)
 
 	gen9_dbuf_disable(dev_priv);
 
-	skl_uninit_cdclk(dev_priv);
+	intel_cdclk_uninit(dev_priv);
 
 	/* The spec doesn't call for removing the reset handshake flag */
 	/* disable PG1 and Misc I/O */
@@ -3703,7 +3717,7 @@ void bxt_display_core_init(struct drm_i915_private *dev_priv,
 
 	mutex_unlock(&power_domains->lock);
 
-	bxt_init_cdclk(dev_priv);
+	intel_cdclk_init(dev_priv);
 
 	gen9_dbuf_enable(dev_priv);
 
@@ -3720,7 +3734,7 @@ void bxt_display_core_uninit(struct drm_i915_private *dev_priv)
 
 	gen9_dbuf_disable(dev_priv);
 
-	bxt_uninit_cdclk(dev_priv);
+	intel_cdclk_uninit(dev_priv);
 
 	/* The spec doesn't call for removing the reset handshake flag */
 
@@ -3762,7 +3776,7 @@ static void cnl_display_core_init(struct drm_i915_private *dev_priv, bool resume
 	mutex_unlock(&power_domains->lock);
 
 	/* 5. Enable CD clock */
-	cnl_init_cdclk(dev_priv);
+	intel_cdclk_init(dev_priv);
 
 	/* 6. Enable DBUF */
 	gen9_dbuf_enable(dev_priv);
@@ -3784,7 +3798,7 @@ static void cnl_display_core_uninit(struct drm_i915_private *dev_priv)
 	gen9_dbuf_disable(dev_priv);
 
 	/* 3. Disable CD clock */
-	cnl_uninit_cdclk(dev_priv);
+	intel_cdclk_uninit(dev_priv);
 
 	/*
 	 * 4. Disable Power Well 1 (PG1).
@@ -3826,7 +3840,7 @@ void icl_display_core_init(struct drm_i915_private *dev_priv,
 	mutex_unlock(&power_domains->lock);
 
 	/* 5. Enable CDCLK. */
-	icl_init_cdclk(dev_priv);
+	intel_cdclk_init(dev_priv);
 
 	/* 6. Enable DBUF. */
 	icl_dbuf_enable(dev_priv);
@@ -3851,7 +3865,7 @@ void icl_display_core_uninit(struct drm_i915_private *dev_priv)
 	icl_dbuf_disable(dev_priv);
 
 	/* 3. Disable CD clock */
-	icl_uninit_cdclk(dev_priv);
+	intel_cdclk_uninit(dev_priv);
 
 	/*
 	 * 4. Disable Power Well 1 (PG1).
@@ -3982,6 +3996,36 @@ static void vlv_cmnlane_wa(struct drm_i915_private *dev_priv)
 	cmn->desc->ops->disable(dev_priv, cmn);
 }
 
+static bool vlv_punit_is_power_gated(struct drm_i915_private *dev_priv, u32 reg0)
+{
+	bool ret;
+
+	mutex_lock(&dev_priv->pcu_lock);
+	ret = (vlv_punit_read(dev_priv, reg0) & SSPM0_SSC_MASK) == SSPM0_SSC_PWR_GATE;
+	mutex_unlock(&dev_priv->pcu_lock);
+
+	return ret;
+}
+
+static void assert_ved_power_gated(struct drm_i915_private *dev_priv)
+{
+	WARN(!vlv_punit_is_power_gated(dev_priv, PUNIT_REG_VEDSSPM0),
+	     "VED not power gated\n");
+}
+
+static void assert_isp_power_gated(struct drm_i915_private *dev_priv)
+{
+	static const struct pci_device_id isp_ids[] = {
+		{PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x0f38)},
+		{PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x22b8)},
+		{}
+	};
+
+	WARN(!pci_dev_present(isp_ids) &&
+	     !vlv_punit_is_power_gated(dev_priv, PUNIT_REG_ISPSSPM0),
+	     "ISP not power gated\n");
+}
+
 static void intel_power_domains_verify_state(struct drm_i915_private *dev_priv);
 
 /**
@@ -4006,7 +4050,7 @@ void intel_power_domains_init_hw(struct drm_i915_private *i915, bool resume)
 
 	power_domains->initializing = true;
 
-	if (IS_ICELAKE(i915)) {
+	if (INTEL_GEN(i915) >= 11) {
 		icl_display_core_init(i915, resume);
 	} else if (IS_CANNONLAKE(i915)) {
 		cnl_display_core_init(i915, resume);
@@ -4018,10 +4062,13 @@ void intel_power_domains_init_hw(struct drm_i915_private *i915, bool resume)
 		mutex_lock(&power_domains->lock);
 		chv_phy_control_init(i915);
 		mutex_unlock(&power_domains->lock);
+		assert_isp_power_gated(i915);
 	} else if (IS_VALLEYVIEW(i915)) {
 		mutex_lock(&power_domains->lock);
 		vlv_cmnlane_wa(i915);
 		mutex_unlock(&power_domains->lock);
+		assert_ved_power_gated(i915);
+		assert_isp_power_gated(i915);
 	} else if (IS_IVYBRIDGE(i915) || INTEL_GEN(i915) >= 7) {
 		intel_pch_reset_handshake(i915, !HAS_PCH_NOP(i915));
 	}
@@ -4151,7 +4198,7 @@ void intel_power_domains_suspend(struct drm_i915_private *i915,
 		intel_power_domains_verify_state(i915);
 	}
 
-	if (IS_ICELAKE(i915))
+	if (INTEL_GEN(i915) >= 11)
 		icl_display_core_uninit(i915);
 	else if (IS_CANNONLAKE(i915))
 		cnl_display_core_uninit(i915);
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index e7b0884ba5a5..0e3d91d9ef13 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -25,16 +25,23 @@
  * Authors:
  *	Eric Anholt <eric@anholt.net>
  */
-#include <linux/i2c.h>
-#include <linux/slab.h>
+
 #include <linux/delay.h>
 #include <linux/export.h>
+#include <linux/i2c.h>
+#include <linux/slab.h>
+
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_edid.h>
-#include "intel_drv.h"
 #include <drm/i915_drm.h>
+
 #include "i915_drv.h"
+#include "intel_connector.h"
+#include "intel_drv.h"
+#include "intel_hdmi.h"
+#include "intel_panel.h"
+#include "intel_sdvo.h"
 #include "intel_sdvo_regs.h"
 
 #define SDVO_TMDS_MASK (SDVO_OUTPUT_TMDS0 | SDVO_OUTPUT_TMDS1)
@@ -978,34 +985,109 @@ static bool intel_sdvo_write_infoframe(struct intel_sdvo *intel_sdvo,
 				    &tx_rate, 1);
 }
 
-static bool intel_sdvo_set_avi_infoframe(struct intel_sdvo *intel_sdvo,
-					 const struct intel_crtc_state *pipe_config,
-					 const struct drm_connector_state *conn_state)
+static ssize_t intel_sdvo_read_infoframe(struct intel_sdvo *intel_sdvo,
+					 unsigned int if_index,
+					 u8 *data, unsigned int length)
 {
+	u8 set_buf_index[2] = { if_index, 0 };
+	u8 hbuf_size, tx_rate, av_split;
+	int i;
+
+	if (!intel_sdvo_get_value(intel_sdvo,
+				  SDVO_CMD_GET_HBUF_AV_SPLIT,
+				  &av_split, 1))
+		return -ENXIO;
+
+	if (av_split < if_index)
+		return 0;
+
+	if (!intel_sdvo_get_value(intel_sdvo,
+				  SDVO_CMD_GET_HBUF_TXRATE,
+				  &tx_rate, 1))
+		return -ENXIO;
+
+	if (tx_rate == SDVO_HBUF_TX_DISABLED)
+		return 0;
+
+	if (!intel_sdvo_set_value(intel_sdvo,
+				  SDVO_CMD_SET_HBUF_INDEX,
+				  set_buf_index, 2))
+		return -ENXIO;
+
+	if (!intel_sdvo_get_value(intel_sdvo, SDVO_CMD_GET_HBUF_INFO,
+				  &hbuf_size, 1))
+		return -ENXIO;
+
+	/* Buffer size is 0 based, hooray! */
+	hbuf_size++;
+
+	DRM_DEBUG_KMS("reading sdvo hbuf: %i, hbuf_size %i, hbuf_size: %i\n",
+		      if_index, length, hbuf_size);
+
+	hbuf_size = min_t(unsigned int, length, hbuf_size);
+
+	for (i = 0; i < hbuf_size; i += 8) {
+		if (!intel_sdvo_write_cmd(intel_sdvo, SDVO_CMD_GET_HBUF_DATA, NULL, 0))
+			return -ENXIO;
+		if (!intel_sdvo_read_response(intel_sdvo, &data[i],
+					      min_t(unsigned int, 8, hbuf_size - i)))
+			return -ENXIO;
+	}
+
+	return hbuf_size;
+}
+
+static bool intel_sdvo_compute_avi_infoframe(struct intel_sdvo *intel_sdvo,
+					     struct intel_crtc_state *crtc_state,
+					     struct drm_connector_state *conn_state)
+{
+	struct hdmi_avi_infoframe *frame = &crtc_state->infoframes.avi.avi;
 	const struct drm_display_mode *adjusted_mode =
-		&pipe_config->base.adjusted_mode;
-	u8 sdvo_data[HDMI_INFOFRAME_SIZE(AVI)];
-	union hdmi_infoframe frame;
+		&crtc_state->base.adjusted_mode;
 	int ret;
-	ssize_t len;
 
-	ret = drm_hdmi_avi_infoframe_from_display_mode(&frame.avi,
+	if (!crtc_state->has_hdmi_sink)
+		return true;
+
+	crtc_state->infoframes.enable |=
+		intel_hdmi_infoframe_enable(HDMI_INFOFRAME_TYPE_AVI);
+
+	ret = drm_hdmi_avi_infoframe_from_display_mode(frame,
 						       conn_state->connector,
 						       adjusted_mode);
-	if (ret < 0) {
-		DRM_ERROR("couldn't fill AVI infoframe\n");
+	if (ret)
 		return false;
-	}
 
-	drm_hdmi_avi_infoframe_quant_range(&frame.avi,
+	drm_hdmi_avi_infoframe_quant_range(frame,
 					   conn_state->connector,
 					   adjusted_mode,
-					   pipe_config->limited_color_range ?
+					   crtc_state->limited_color_range ?
 					   HDMI_QUANTIZATION_RANGE_LIMITED :
 					   HDMI_QUANTIZATION_RANGE_FULL);
 
-	len = hdmi_infoframe_pack(&frame, sdvo_data, sizeof(sdvo_data));
-	if (len < 0)
+	ret = hdmi_avi_infoframe_check(frame);
+	if (WARN_ON(ret))
+		return false;
+
+	return true;
+}
+
+static bool intel_sdvo_set_avi_infoframe(struct intel_sdvo *intel_sdvo,
+					 const struct intel_crtc_state *crtc_state)
+{
+	u8 sdvo_data[HDMI_INFOFRAME_SIZE(AVI)];
+	const union hdmi_infoframe *frame = &crtc_state->infoframes.avi;
+	ssize_t len;
+
+	if ((crtc_state->infoframes.enable &
+	     intel_hdmi_infoframe_enable(HDMI_INFOFRAME_TYPE_AVI)) == 0)
+		return true;
+
+	if (WARN_ON(frame->any.type != HDMI_INFOFRAME_TYPE_AVI))
+		return false;
+
+	len = hdmi_infoframe_pack_only(frame, sdvo_data, sizeof(sdvo_data));
+	if (WARN_ON(len < 0))
 		return false;
 
 	return intel_sdvo_write_infoframe(intel_sdvo, SDVO_HBUF_INDEX_AVI_IF,
@@ -1013,6 +1095,40 @@ static bool intel_sdvo_set_avi_infoframe(struct intel_sdvo *intel_sdvo,
 					  sdvo_data, sizeof(sdvo_data));
 }
 
+static void intel_sdvo_get_avi_infoframe(struct intel_sdvo *intel_sdvo,
+					 struct intel_crtc_state *crtc_state)
+{
+	u8 sdvo_data[HDMI_INFOFRAME_SIZE(AVI)];
+	union hdmi_infoframe *frame = &crtc_state->infoframes.avi;
+	ssize_t len;
+	int ret;
+
+	if (!crtc_state->has_hdmi_sink)
+		return;
+
+	len = intel_sdvo_read_infoframe(intel_sdvo, SDVO_HBUF_INDEX_AVI_IF,
+					sdvo_data, sizeof(sdvo_data));
+	if (len < 0) {
+		DRM_DEBUG_KMS("failed to read AVI infoframe\n");
+		return;
+	} else if (len == 0) {
+		return;
+	}
+
+	crtc_state->infoframes.enable |=
+		intel_hdmi_infoframe_enable(HDMI_INFOFRAME_TYPE_AVI);
+
+	ret = hdmi_infoframe_unpack(frame, sdvo_data, sizeof(sdvo_data));
+	if (ret) {
+		DRM_DEBUG_KMS("Failed to unpack AVI infoframe\n");
+		return;
+	}
+
+	if (frame->any.type != HDMI_INFOFRAME_TYPE_AVI)
+		DRM_DEBUG_KMS("Found the wrong infoframe type 0x%x (expected 0x%02x)\n",
+			      frame->any.type, HDMI_INFOFRAME_TYPE_AVI);
+}
+
 static bool intel_sdvo_set_tv_format(struct intel_sdvo *intel_sdvo,
 				     const struct drm_connector_state *conn_state)
 {
@@ -1193,6 +1309,12 @@ static int intel_sdvo_compute_config(struct intel_encoder *encoder,
 	if (intel_sdvo_connector->is_hdmi)
 		adjusted_mode->picture_aspect_ratio = conn_state->picture_aspect_ratio;
 
+	if (!intel_sdvo_compute_avi_infoframe(intel_sdvo,
+					      pipe_config, conn_state)) {
+		DRM_DEBUG_KMS("bad AVI infoframe\n");
+		return -EINVAL;
+	}
+
 	return 0;
 }
 
@@ -1315,8 +1437,7 @@ static void intel_sdvo_pre_enable(struct intel_encoder *intel_encoder,
 		intel_sdvo_set_encode(intel_sdvo, SDVO_ENCODE_HDMI);
 		intel_sdvo_set_colorimetry(intel_sdvo,
 					   SDVO_COLORIMETRY_RGB256);
-		intel_sdvo_set_avi_infoframe(intel_sdvo,
-					     crtc_state, conn_state);
+		intel_sdvo_set_avi_infoframe(intel_sdvo, crtc_state);
 	} else
 		intel_sdvo_set_encode(intel_sdvo, SDVO_ENCODE_DVI);
 
@@ -1507,6 +1628,10 @@ static void intel_sdvo_get_config(struct intel_encoder *encoder,
 		}
 	}
 
+	WARN(encoder_pixel_multiplier != pipe_config->pixel_multiplier,
+	     "SDVO pixel multiplier mismatch, port: %i, encoder: %i\n",
+	     pipe_config->pixel_multiplier, encoder_pixel_multiplier);
+
 	if (sdvox & HDMI_COLOR_RANGE_16_235)
 		pipe_config->limited_color_range = true;
 
@@ -1519,9 +1644,7 @@ static void intel_sdvo_get_config(struct intel_encoder *encoder,
 			pipe_config->has_hdmi_sink = true;
 	}
 
-	WARN(encoder_pixel_multiplier != pipe_config->pixel_multiplier,
-	     "SDVO pixel multiplier mismatch, port: %i, encoder: %i\n",
-	     pipe_config->pixel_multiplier, encoder_pixel_multiplier);
+	intel_sdvo_get_avi_infoframe(intel_sdvo, pipe_config);
 }
 
 static void intel_disable_sdvo(struct intel_encoder *encoder,
diff --git a/drivers/gpu/drm/i915/intel_sdvo.h b/drivers/gpu/drm/i915/intel_sdvo.h
new file mode 100644
index 000000000000..c9e05bcdd141
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_sdvo.h
@@ -0,0 +1,23 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_SDVO_H__
+#define __INTEL_SDVO_H__
+
+#include <linux/types.h>
+
+#include <drm/i915_drm.h>
+
+#include "i915_reg.h"
+
+struct drm_i915_private;
+enum pipe;
+
+bool intel_sdvo_port_enabled(struct drm_i915_private *dev_priv,
+			     i915_reg_t sdvo_reg, enum pipe *pipe);
+bool intel_sdvo_init(struct drm_i915_private *dev_priv,
+		     i915_reg_t reg, enum port port);
+
+#endif /* __INTEL_SDVO_H__ */
diff --git a/drivers/gpu/drm/i915/intel_sideband.c b/drivers/gpu/drm/i915/intel_sideband.c
index 75c872bb8cc9..57de41b1f989 100644
--- a/drivers/gpu/drm/i915/intel_sideband.c
+++ b/drivers/gpu/drm/i915/intel_sideband.c
@@ -51,7 +51,7 @@ static int vlv_sideband_rw(struct drm_i915_private *dev_priv, u32 devfn,
 
 	WARN_ON(!mutex_is_locked(&dev_priv->sb_lock));
 
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    VLV_IOSF_DOORBELL_REQ, IOSF_SB_BUSY, 0,
 				    5)) {
 		DRM_DEBUG_DRIVER("IOSF sideband idle wait (%s) timed out\n",
@@ -63,7 +63,7 @@ static int vlv_sideband_rw(struct drm_i915_private *dev_priv, u32 devfn,
 	I915_WRITE(VLV_IOSF_DATA, is_read ? 0 : *val);
 	I915_WRITE(VLV_IOSF_DOORBELL_REQ, cmd);
 
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    VLV_IOSF_DOORBELL_REQ, IOSF_SB_BUSY, 0,
 				    5)) {
 		DRM_DEBUG_DRIVER("IOSF sideband finish wait (%s) timed out\n",
@@ -208,7 +208,7 @@ u32 intel_sbi_read(struct drm_i915_private *dev_priv, u16 reg,
 	u32 value = 0;
 	WARN_ON(!mutex_is_locked(&dev_priv->sb_lock));
 
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    SBI_CTL_STAT, SBI_BUSY, 0,
 				    100)) {
 		DRM_ERROR("timeout waiting for SBI to become ready\n");
@@ -224,7 +224,7 @@ u32 intel_sbi_read(struct drm_i915_private *dev_priv, u16 reg,
 		value = SBI_CTL_DEST_MPHY | SBI_CTL_OP_IORD;
 	I915_WRITE(SBI_CTL_STAT, value | SBI_BUSY);
 
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    SBI_CTL_STAT,
 				    SBI_BUSY,
 				    0,
@@ -248,7 +248,7 @@ void intel_sbi_write(struct drm_i915_private *dev_priv, u16 reg, u32 value,
 
 	WARN_ON(!mutex_is_locked(&dev_priv->sb_lock));
 
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    SBI_CTL_STAT, SBI_BUSY, 0,
 				    100)) {
 		DRM_ERROR("timeout waiting for SBI to become ready\n");
@@ -264,7 +264,7 @@ void intel_sbi_write(struct drm_i915_private *dev_priv, u16 reg, u32 value,
 		tmp = SBI_CTL_DEST_MPHY | SBI_CTL_OP_IOWR;
 	I915_WRITE(SBI_CTL_STAT, SBI_BUSY | tmp);
 
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    SBI_CTL_STAT,
 				    SBI_BUSY,
 				    0,
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index b56a1a9ad01d..2913e89280d7 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -29,17 +29,36 @@
  * registers; newer ones are much simpler and we can use the new DRM plane
  * support.
  */
+
+#include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_color_mgmt.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_fourcc.h>
-#include <drm/drm_rect.h>
-#include <drm/drm_atomic.h>
 #include <drm/drm_plane_helper.h>
-#include "intel_drv.h"
-#include "intel_frontbuffer.h"
+#include <drm/drm_rect.h>
 #include <drm/i915_drm.h>
+
 #include "i915_drv.h"
-#include <drm/drm_color_mgmt.h>
+#include "intel_atomic_plane.h"
+#include "intel_drv.h"
+#include "intel_frontbuffer.h"
+#include "intel_pm.h"
+#include "intel_psr.h"
+#include "intel_sprite.h"
+
+bool is_planar_yuv_format(u32 pixelformat)
+{
+	switch (pixelformat) {
+	case DRM_FORMAT_NV12:
+	case DRM_FORMAT_P010:
+	case DRM_FORMAT_P012:
+	case DRM_FORMAT_P016:
+		return true;
+	default:
+		return false;
+	}
+}
 
 int intel_usecs_to_scanlines(const struct drm_display_mode *adjusted_mode,
 			     int usecs)
@@ -256,7 +275,8 @@ int intel_plane_check_src_coordinates(struct intel_plane_state *plane_state)
 {
 	const struct drm_framebuffer *fb = plane_state->base.fb;
 	struct drm_rect *src = &plane_state->base.src;
-	u32 src_x, src_y, src_w, src_h;
+	u32 src_x, src_y, src_w, src_h, hsub, vsub;
+	bool rotated = drm_rotation_90_or_270(plane_state->base.rotation);
 
 	/*
 	 * Hardware doesn't handle subpixel coordinates.
@@ -274,18 +294,26 @@ int intel_plane_check_src_coordinates(struct intel_plane_state *plane_state)
 	src->y1 = src_y << 16;
 	src->y2 = (src_y + src_h) << 16;
 
-	if (fb->format->is_yuv &&
-	    (src_x & 1 || src_w & 1)) {
-		DRM_DEBUG_KMS("src x/w (%u, %u) must be a multiple of 2 for YUV planes\n",
-			      src_x, src_w);
+	if (!fb->format->is_yuv)
+		return 0;
+
+	/* YUV specific checks */
+	if (!rotated) {
+		hsub = fb->format->hsub;
+		vsub = fb->format->vsub;
+	} else {
+		hsub = vsub = max(fb->format->hsub, fb->format->vsub);
+	}
+
+	if (src_x % hsub || src_w % hsub) {
+		DRM_DEBUG_KMS("src x/w (%u, %u) must be a multiple of %u for %sYUV planes\n",
+			      src_x, src_w, hsub, rotated ? "rotated " : "");
 		return -EINVAL;
 	}
 
-	if (fb->format->is_yuv &&
-	    fb->format->num_planes > 1 &&
-	    (src_y & 1 || src_h & 1)) {
-		DRM_DEBUG_KMS("src y/h (%u, %u) must be a multiple of 2 for planar YUV planes\n",
-			      src_y, src_h);
+	if (src_y % vsub || src_h % vsub) {
+		DRM_DEBUG_KMS("src y/h (%u, %u) must be a multiple of %u for %sYUV planes\n",
+			      src_y, src_h, vsub, rotated ? "rotated " : "");
 		return -EINVAL;
 	}
 
@@ -335,8 +363,8 @@ skl_program_scaler(struct intel_plane *plane,
 				      0, INT_MAX);
 
 	/* TODO: handle sub-pixel coordinates */
-	if (plane_state->base.fb->format->format == DRM_FORMAT_NV12 &&
-	    !icl_is_hdr_plane(plane)) {
+	if (is_planar_yuv_format(plane_state->base.fb->format->format) &&
+	    !icl_is_hdr_plane(dev_priv, plane->id)) {
 		y_hphase = skl_scaler_calc_phase(1, hscale, false);
 		y_vphase = skl_scaler_calc_phase(1, vscale, false);
 
@@ -518,7 +546,7 @@ skl_program_plane(struct intel_plane *plane,
 	I915_WRITE_FW(PLANE_AUX_DIST(pipe, plane_id),
 		      (plane_state->color_plane[1].offset - surf_addr) | aux_stride);
 
-	if (icl_is_hdr_plane(plane)) {
+	if (icl_is_hdr_plane(dev_priv, plane_id)) {
 		u32 cus_ctl = 0;
 
 		if (linked) {
@@ -542,7 +570,7 @@ skl_program_plane(struct intel_plane *plane,
 	if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv))
 		I915_WRITE_FW(PLANE_COLOR_CTL(pipe, plane_id), plane_color_ctl);
 
-	if (fb->format->is_yuv && icl_is_hdr_plane(plane))
+	if (fb->format->is_yuv && icl_is_hdr_plane(dev_priv, plane_id))
 		icl_program_input_csc(plane, crtc_state, plane_state);
 
 	skl_write_plane_wm(plane, crtc_state);
@@ -609,6 +637,9 @@ skl_disable_plane(struct intel_plane *plane,
 
 	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
 
+	if (icl_is_hdr_plane(dev_priv, plane_id))
+		I915_WRITE_FW(PLANE_CUS_CTL(pipe, plane_id), 0);
+
 	skl_write_plane_wm(plane, crtc_state);
 
 	I915_WRITE_FW(PLANE_CTL(pipe, plane_id), 0);
@@ -741,7 +772,12 @@ vlv_update_clrc(const struct intel_plane_state *plane_state)
 
 static u32 vlv_sprite_ctl_crtc(const struct intel_crtc_state *crtc_state)
 {
-	return SP_GAMMA_ENABLE;
+	u32 sprctl = 0;
+
+	if (crtc_state->gamma_enable)
+		sprctl |= SP_GAMMA_ENABLE;
+
+	return sprctl;
 }
 
 static u32 vlv_sprite_ctl(const struct intel_crtc_state *crtc_state,
@@ -916,12 +952,12 @@ vlv_plane_get_hw_state(struct intel_plane *plane,
 
 static u32 ivb_sprite_ctl_crtc(const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc_state->base.crtc->dev);
 	u32 sprctl = 0;
 
-	sprctl |= SPRITE_GAMMA_ENABLE;
+	if (crtc_state->gamma_enable)
+		sprctl |= SPRITE_GAMMA_ENABLE;
 
-	if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
+	if (crtc_state->csc_enable)
 		sprctl |= SPRITE_PIPE_CSC_ENABLE;
 
 	return sprctl;
@@ -1107,7 +1143,15 @@ g4x_sprite_max_stride(struct intel_plane *plane,
 
 static u32 g4x_sprite_ctl_crtc(const struct intel_crtc_state *crtc_state)
 {
-	return DVS_GAMMA_ENABLE;
+	u32 dvscntr = 0;
+
+	if (crtc_state->gamma_enable)
+		dvscntr |= DVS_GAMMA_ENABLE;
+
+	if (crtc_state->csc_enable)
+		dvscntr |= DVS_PIPE_CSC_ENABLE;
+
+	return dvscntr;
 }
 
 static u32 g4x_sprite_ctl(const struct intel_crtc_state *crtc_state,
@@ -1482,8 +1526,6 @@ static int skl_plane_check_fb(const struct intel_crtc_state *crtc_state,
 		/*
 		 * 90/270 is not allowed with RGB64 16:16:16:16 and
 		 * Indexed 8-bit. RGB 16-bit 5:6:5 is allowed gen11 onwards.
-		 * TBD: Add RGB64 case once its added in supported format
-		 * list.
 		 */
 		switch (fb->format->format) {
 		case DRM_FORMAT_RGB565:
@@ -1491,6 +1533,15 @@ static int skl_plane_check_fb(const struct intel_crtc_state *crtc_state,
 				break;
 			/* fall through */
 		case DRM_FORMAT_C8:
+		case DRM_FORMAT_XRGB16161616F:
+		case DRM_FORMAT_XBGR16161616F:
+		case DRM_FORMAT_ARGB16161616F:
+		case DRM_FORMAT_ABGR16161616F:
+		case DRM_FORMAT_Y210:
+		case DRM_FORMAT_Y212:
+		case DRM_FORMAT_Y216:
+		case DRM_FORMAT_XVYU12_16161616:
+		case DRM_FORMAT_XVYU16161616:
 			DRM_DEBUG_KMS("Unsupported pixel format %s for 90/270!\n",
 				      drm_get_format_name(fb->format->format,
 							  &format_name));
@@ -1551,10 +1602,10 @@ static int skl_plane_check_nv12_rotation(const struct intel_plane_state *plane_s
 	int src_w = drm_rect_width(&plane_state->base.src) >> 16;
 
 	/* Display WA #1106 */
-	if (fb->format->format == DRM_FORMAT_NV12 && src_w & 3 &&
+	if (is_planar_yuv_format(fb->format->format) && src_w & 3 &&
 	    (rotation == DRM_MODE_ROTATE_270 ||
 	     rotation == (DRM_MODE_REFLECT_X | DRM_MODE_ROTATE_90))) {
-		DRM_DEBUG_KMS("src width must be multiple of 4 for rotated NV12\n");
+		DRM_DEBUG_KMS("src width must be multiple of 4 for rotated planar YUV\n");
 		return -EINVAL;
 	}
 
@@ -1790,6 +1841,52 @@ static const u32 skl_plane_formats[] = {
 	DRM_FORMAT_VYUY,
 };
 
+static const u32 icl_plane_formats[] = {
+	DRM_FORMAT_C8,
+	DRM_FORMAT_RGB565,
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_XBGR8888,
+	DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_ABGR8888,
+	DRM_FORMAT_XRGB2101010,
+	DRM_FORMAT_XBGR2101010,
+	DRM_FORMAT_YUYV,
+	DRM_FORMAT_YVYU,
+	DRM_FORMAT_UYVY,
+	DRM_FORMAT_VYUY,
+	DRM_FORMAT_Y210,
+	DRM_FORMAT_Y212,
+	DRM_FORMAT_Y216,
+	DRM_FORMAT_XVYU2101010,
+	DRM_FORMAT_XVYU12_16161616,
+	DRM_FORMAT_XVYU16161616,
+};
+
+static const u32 icl_hdr_plane_formats[] = {
+	DRM_FORMAT_C8,
+	DRM_FORMAT_RGB565,
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_XBGR8888,
+	DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_ABGR8888,
+	DRM_FORMAT_XRGB2101010,
+	DRM_FORMAT_XBGR2101010,
+	DRM_FORMAT_XRGB16161616F,
+	DRM_FORMAT_XBGR16161616F,
+	DRM_FORMAT_ARGB16161616F,
+	DRM_FORMAT_ABGR16161616F,
+	DRM_FORMAT_YUYV,
+	DRM_FORMAT_YVYU,
+	DRM_FORMAT_UYVY,
+	DRM_FORMAT_VYUY,
+	DRM_FORMAT_Y210,
+	DRM_FORMAT_Y212,
+	DRM_FORMAT_Y216,
+	DRM_FORMAT_XVYU2101010,
+	DRM_FORMAT_XVYU12_16161616,
+	DRM_FORMAT_XVYU16161616,
+};
+
 static const u32 skl_planar_formats[] = {
 	DRM_FORMAT_C8,
 	DRM_FORMAT_RGB565,
@@ -1806,6 +1903,79 @@ static const u32 skl_planar_formats[] = {
 	DRM_FORMAT_NV12,
 };
 
+static const u32 glk_planar_formats[] = {
+	DRM_FORMAT_C8,
+	DRM_FORMAT_RGB565,
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_XBGR8888,
+	DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_ABGR8888,
+	DRM_FORMAT_XRGB2101010,
+	DRM_FORMAT_XBGR2101010,
+	DRM_FORMAT_YUYV,
+	DRM_FORMAT_YVYU,
+	DRM_FORMAT_UYVY,
+	DRM_FORMAT_VYUY,
+	DRM_FORMAT_NV12,
+	DRM_FORMAT_P010,
+	DRM_FORMAT_P012,
+	DRM_FORMAT_P016,
+};
+
+static const u32 icl_planar_formats[] = {
+	DRM_FORMAT_C8,
+	DRM_FORMAT_RGB565,
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_XBGR8888,
+	DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_ABGR8888,
+	DRM_FORMAT_XRGB2101010,
+	DRM_FORMAT_XBGR2101010,
+	DRM_FORMAT_YUYV,
+	DRM_FORMAT_YVYU,
+	DRM_FORMAT_UYVY,
+	DRM_FORMAT_VYUY,
+	DRM_FORMAT_NV12,
+	DRM_FORMAT_P010,
+	DRM_FORMAT_P012,
+	DRM_FORMAT_P016,
+	DRM_FORMAT_Y210,
+	DRM_FORMAT_Y212,
+	DRM_FORMAT_Y216,
+	DRM_FORMAT_XVYU2101010,
+	DRM_FORMAT_XVYU12_16161616,
+	DRM_FORMAT_XVYU16161616,
+};
+
+static const u32 icl_hdr_planar_formats[] = {
+	DRM_FORMAT_C8,
+	DRM_FORMAT_RGB565,
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_XBGR8888,
+	DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_ABGR8888,
+	DRM_FORMAT_XRGB2101010,
+	DRM_FORMAT_XBGR2101010,
+	DRM_FORMAT_XRGB16161616F,
+	DRM_FORMAT_XBGR16161616F,
+	DRM_FORMAT_ARGB16161616F,
+	DRM_FORMAT_ABGR16161616F,
+	DRM_FORMAT_YUYV,
+	DRM_FORMAT_YVYU,
+	DRM_FORMAT_UYVY,
+	DRM_FORMAT_VYUY,
+	DRM_FORMAT_NV12,
+	DRM_FORMAT_P010,
+	DRM_FORMAT_P012,
+	DRM_FORMAT_P016,
+	DRM_FORMAT_Y210,
+	DRM_FORMAT_Y212,
+	DRM_FORMAT_Y216,
+	DRM_FORMAT_XVYU2101010,
+	DRM_FORMAT_XVYU12_16161616,
+	DRM_FORMAT_XVYU16161616,
+};
+
 static const u64 skl_plane_format_modifiers_noccs[] = {
 	I915_FORMAT_MOD_Yf_TILED,
 	I915_FORMAT_MOD_Y_TILED,
@@ -1945,10 +2115,23 @@ static bool skl_plane_format_mod_supported(struct drm_plane *_plane,
 	case DRM_FORMAT_UYVY:
 	case DRM_FORMAT_VYUY:
 	case DRM_FORMAT_NV12:
+	case DRM_FORMAT_P010:
+	case DRM_FORMAT_P012:
+	case DRM_FORMAT_P016:
+	case DRM_FORMAT_XVYU2101010:
 		if (modifier == I915_FORMAT_MOD_Yf_TILED)
 			return true;
 		/* fall through */
 	case DRM_FORMAT_C8:
+	case DRM_FORMAT_XBGR16161616F:
+	case DRM_FORMAT_ABGR16161616F:
+	case DRM_FORMAT_XRGB16161616F:
+	case DRM_FORMAT_ARGB16161616F:
+	case DRM_FORMAT_Y210:
+	case DRM_FORMAT_Y212:
+	case DRM_FORMAT_Y216:
+	case DRM_FORMAT_XVYU12_16161616:
+	case DRM_FORMAT_XVYU16161616:
 		if (modifier == DRM_FORMAT_MOD_LINEAR ||
 		    modifier == I915_FORMAT_MOD_X_TILED ||
 		    modifier == I915_FORMAT_MOD_Y_TILED)
@@ -2085,8 +2268,25 @@ skl_universal_plane_create(struct drm_i915_private *dev_priv,
 		plane->update_slave = icl_update_slave;
 
 	if (skl_plane_has_planar(dev_priv, pipe, plane_id)) {
-		formats = skl_planar_formats;
-		num_formats = ARRAY_SIZE(skl_planar_formats);
+		if (icl_is_hdr_plane(dev_priv, plane_id)) {
+			formats = icl_hdr_planar_formats;
+			num_formats = ARRAY_SIZE(icl_hdr_planar_formats);
+		} else if (INTEL_GEN(dev_priv) >= 11) {
+			formats = icl_planar_formats;
+			num_formats = ARRAY_SIZE(icl_planar_formats);
+		} else if (INTEL_GEN(dev_priv) == 10 || IS_GEMINILAKE(dev_priv)) {
+			formats = glk_planar_formats;
+			num_formats = ARRAY_SIZE(glk_planar_formats);
+		} else {
+			formats = skl_planar_formats;
+			num_formats = ARRAY_SIZE(skl_planar_formats);
+		}
+	} else if (icl_is_hdr_plane(dev_priv, plane_id)) {
+		formats = icl_hdr_plane_formats;
+		num_formats = ARRAY_SIZE(icl_hdr_plane_formats);
+	} else if (INTEL_GEN(dev_priv) >= 11) {
+		formats = icl_plane_formats;
+		num_formats = ARRAY_SIZE(icl_plane_formats);
 	} else {
 		formats = skl_plane_formats;
 		num_formats = ARRAY_SIZE(skl_plane_formats);
diff --git a/drivers/gpu/drm/i915/intel_sprite.h b/drivers/gpu/drm/i915/intel_sprite.h
new file mode 100644
index 000000000000..84be8686be16
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_sprite.h
@@ -0,0 +1,55 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_SPRITE_H__
+#define __INTEL_SPRITE_H__
+
+#include <linux/types.h>
+
+#include "i915_drv.h"
+#include "intel_display.h"
+
+struct drm_device;
+struct drm_display_mode;
+struct drm_file;
+struct drm_i915_private;
+struct intel_crtc_state;
+struct intel_plane_state;
+
+bool is_planar_yuv_format(u32 pixelformat);
+int intel_usecs_to_scanlines(const struct drm_display_mode *adjusted_mode,
+			     int usecs);
+struct intel_plane *intel_sprite_plane_create(struct drm_i915_private *dev_priv,
+					      enum pipe pipe, int plane);
+int intel_sprite_set_colorkey_ioctl(struct drm_device *dev, void *data,
+				    struct drm_file *file_priv);
+void intel_pipe_update_start(const struct intel_crtc_state *new_crtc_state);
+void intel_pipe_update_end(struct intel_crtc_state *new_crtc_state);
+int intel_plane_check_stride(const struct intel_plane_state *plane_state);
+int intel_plane_check_src_coordinates(struct intel_plane_state *plane_state);
+int chv_plane_check_rotation(const struct intel_plane_state *plane_state);
+struct intel_plane *
+skl_universal_plane_create(struct drm_i915_private *dev_priv,
+			   enum pipe pipe, enum plane_id plane_id);
+
+static inline bool icl_is_nv12_y_plane(enum plane_id id)
+{
+	/* Don't need to do a gen check, these planes are only available on gen11 */
+	if (id == PLANE_SPRITE4 || id == PLANE_SPRITE5)
+		return true;
+
+	return false;
+}
+
+static inline bool icl_is_hdr_plane(struct drm_i915_private *dev_priv,
+				    enum plane_id plane_id)
+{
+	if (INTEL_GEN(dev_priv) < 11)
+		return false;
+
+	return plane_id < PLANE_SPRITE2;
+}
+
+#endif /* __INTEL_SPRITE_H__ */
diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c
index 3924c4944e1f..5dbba33f4202 100644
--- a/drivers/gpu/drm/i915/intel_tv.c
+++ b/drivers/gpu/drm/i915/intel_tv.c
@@ -33,9 +33,12 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_edid.h>
-#include "intel_drv.h"
 #include <drm/i915_drm.h>
+
 #include "i915_drv.h"
+#include "intel_connector.h"
+#include "intel_drv.h"
+#include "intel_tv.h"
 
 enum tv_margin {
 	TV_MARGIN_LEFT, TV_MARGIN_TOP,
diff --git a/drivers/gpu/drm/i915/intel_tv.h b/drivers/gpu/drm/i915/intel_tv.h
new file mode 100644
index 000000000000..44518575ec5c
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_tv.h
@@ -0,0 +1,13 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2019 Intel Corporation
+ */
+
+#ifndef __INTEL_TV_H__
+#define __INTEL_TV_H__
+
+struct drm_i915_private;
+
+void intel_tv_init(struct drm_i915_private *dev_priv);
+
+#endif /* __INTEL_TV_H__ */
diff --git a/drivers/gpu/drm/i915/intel_uc.c b/drivers/gpu/drm/i915/intel_uc.c
index e711eb3268bc..25b80ffe71ad 100644
--- a/drivers/gpu/drm/i915/intel_uc.c
+++ b/drivers/gpu/drm/i915/intel_uc.c
@@ -332,8 +332,6 @@ void intel_uc_sanitize(struct drm_i915_private *i915)
 
 	GEM_BUG_ON(!HAS_GUC(i915));
 
-	guc_disable_communication(guc);
-
 	intel_huc_sanitize(huc);
 	intel_guc_sanitize(guc);
 
@@ -377,7 +375,7 @@ int intel_uc_init_hw(struct drm_i915_private *i915)
 
 		intel_guc_init_params(guc);
 		ret = intel_guc_fw_upload(guc);
-		if (ret == 0 || ret != -ETIMEDOUT)
+		if (ret == 0)
 			break;
 
 		DRM_DEBUG_DRIVER("GuC fw load failed: %d; will reset and "
@@ -451,6 +449,23 @@ void intel_uc_fini_hw(struct drm_i915_private *i915)
 	guc_disable_communication(guc);
 }
 
+/**
+ * intel_uc_reset_prepare - Prepare for reset
+ * @i915: device private
+ *
+ * Preparing for full gpu reset.
+ */
+void intel_uc_reset_prepare(struct drm_i915_private *i915)
+{
+	struct intel_guc *guc = &i915->guc;
+
+	if (!USES_GUC(i915))
+		return;
+
+	guc_disable_communication(guc);
+	intel_uc_sanitize(i915);
+}
+
 int intel_uc_suspend(struct drm_i915_private *i915)
 {
 	struct intel_guc *guc = &i915->guc;
@@ -468,7 +483,7 @@ int intel_uc_suspend(struct drm_i915_private *i915)
 		return err;
 	}
 
-	gen9_disable_guc_interrupts(i915);
+	guc_disable_communication(guc);
 
 	return 0;
 }
@@ -484,7 +499,7 @@ int intel_uc_resume(struct drm_i915_private *i915)
 	if (guc->fw.load_status != INTEL_UC_FIRMWARE_SUCCESS)
 		return 0;
 
-	gen9_enable_guc_interrupts(i915);
+	guc_enable_communication(guc);
 
 	err = intel_guc_resume(guc);
 	if (err) {
diff --git a/drivers/gpu/drm/i915/intel_uc.h b/drivers/gpu/drm/i915/intel_uc.h
index 870faf9011b9..c14729786652 100644
--- a/drivers/gpu/drm/i915/intel_uc.h
+++ b/drivers/gpu/drm/i915/intel_uc.h
@@ -38,6 +38,7 @@ int intel_uc_init_hw(struct drm_i915_private *dev_priv);
 void intel_uc_fini_hw(struct drm_i915_private *dev_priv);
 int intel_uc_init(struct drm_i915_private *dev_priv);
 void intel_uc_fini(struct drm_i915_private *dev_priv);
+void intel_uc_reset_prepare(struct drm_i915_private *i915);
 int intel_uc_suspend(struct drm_i915_private *dev_priv);
 int intel_uc_resume(struct drm_i915_private *dev_priv);
 
diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index 75646a1e0051..d1d51e1121e2 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -21,17 +21,18 @@
  * IN THE SOFTWARE.
  */
 
+#include <linux/pm_runtime.h>
+#include <asm/iosf_mbi.h>
+
 #include "i915_drv.h"
-#include "intel_drv.h"
 #include "i915_vgpu.h"
-
-#include <asm/iosf_mbi.h>
-#include <linux/pm_runtime.h>
+#include "intel_drv.h"
+#include "intel_pm.h"
 
 #define FORCEWAKE_ACK_TIMEOUT_MS 50
 #define GT_FIFO_TIMEOUT_MS	 10
 
-#define __raw_posting_read(dev_priv__, reg__) (void)__raw_i915_read32((dev_priv__), (reg__))
+#define __raw_posting_read(...) ((void)__raw_uncore_read32(__VA_ARGS__))
 
 static const char * const forcewake_domain_names[] = {
 	"render",
@@ -58,16 +59,20 @@ intel_uncore_forcewake_domain_to_str(const enum forcewake_domain_id id)
 	return "unknown";
 }
 
+#define fw_ack(d) readl((d)->reg_ack)
+#define fw_set(d, val) writel(_MASKED_BIT_ENABLE((val)), (d)->reg_set)
+#define fw_clear(d, val) writel(_MASKED_BIT_DISABLE((val)), (d)->reg_set)
+
 static inline void
-fw_domain_reset(struct drm_i915_private *i915,
-		const struct intel_uncore_forcewake_domain *d)
+fw_domain_reset(const struct intel_uncore_forcewake_domain *d)
 {
 	/*
 	 * We don't really know if the powerwell for the forcewake domain we are
 	 * trying to reset here does exist at this point (engines could be fused
 	 * off in ICL+), so no waiting for acks
 	 */
-	__raw_i915_write32(i915, d->reg_set, i915->uncore.fw_reset);
+	/* WaRsClearFWBitsAtReset:bdw,skl */
+	fw_clear(d, 0xffff);
 }
 
 static inline void
@@ -81,36 +86,32 @@ fw_domain_arm_timer(struct intel_uncore_forcewake_domain *d)
 }
 
 static inline int
-__wait_for_ack(const struct drm_i915_private *i915,
-	       const struct intel_uncore_forcewake_domain *d,
+__wait_for_ack(const struct intel_uncore_forcewake_domain *d,
 	       const u32 ack,
 	       const u32 value)
 {
-	return wait_for_atomic((__raw_i915_read32(i915, d->reg_ack) & ack) == value,
+	return wait_for_atomic((fw_ack(d) & ack) == value,
 			       FORCEWAKE_ACK_TIMEOUT_MS);
 }
 
 static inline int
-wait_ack_clear(const struct drm_i915_private *i915,
-	       const struct intel_uncore_forcewake_domain *d,
+wait_ack_clear(const struct intel_uncore_forcewake_domain *d,
 	       const u32 ack)
 {
-	return __wait_for_ack(i915, d, ack, 0);
+	return __wait_for_ack(d, ack, 0);
 }
 
 static inline int
-wait_ack_set(const struct drm_i915_private *i915,
-	     const struct intel_uncore_forcewake_domain *d,
+wait_ack_set(const struct intel_uncore_forcewake_domain *d,
 	     const u32 ack)
 {
-	return __wait_for_ack(i915, d, ack, ack);
+	return __wait_for_ack(d, ack, ack);
 }
 
 static inline void
-fw_domain_wait_ack_clear(const struct drm_i915_private *i915,
-			 const struct intel_uncore_forcewake_domain *d)
+fw_domain_wait_ack_clear(const struct intel_uncore_forcewake_domain *d)
 {
-	if (wait_ack_clear(i915, d, FORCEWAKE_KERNEL))
+	if (wait_ack_clear(d, FORCEWAKE_KERNEL))
 		DRM_ERROR("%s: timed out waiting for forcewake ack to clear.\n",
 			  intel_uncore_forcewake_domain_to_str(d->id));
 }
@@ -121,8 +122,7 @@ enum ack_type {
 };
 
 static int
-fw_domain_wait_ack_with_fallback(const struct drm_i915_private *i915,
-				 const struct intel_uncore_forcewake_domain *d,
+fw_domain_wait_ack_with_fallback(const struct intel_uncore_forcewake_domain *d,
 				 const enum ack_type type)
 {
 	const u32 ack_bit = FORCEWAKE_KERNEL;
@@ -146,129 +146,122 @@ fw_domain_wait_ack_with_fallback(const struct drm_i915_private *i915,
 
 	pass = 1;
 	do {
-		wait_ack_clear(i915, d, FORCEWAKE_KERNEL_FALLBACK);
+		wait_ack_clear(d, FORCEWAKE_KERNEL_FALLBACK);
 
-		__raw_i915_write32(i915, d->reg_set,
-				   _MASKED_BIT_ENABLE(FORCEWAKE_KERNEL_FALLBACK));
+		fw_set(d, FORCEWAKE_KERNEL_FALLBACK);
 		/* Give gt some time to relax before the polling frenzy */
 		udelay(10 * pass);
-		wait_ack_set(i915, d, FORCEWAKE_KERNEL_FALLBACK);
+		wait_ack_set(d, FORCEWAKE_KERNEL_FALLBACK);
 
-		ack_detected = (__raw_i915_read32(i915, d->reg_ack) & ack_bit) == value;
+		ack_detected = (fw_ack(d) & ack_bit) == value;
 
-		__raw_i915_write32(i915, d->reg_set,
-				   _MASKED_BIT_DISABLE(FORCEWAKE_KERNEL_FALLBACK));
+		fw_clear(d, FORCEWAKE_KERNEL_FALLBACK);
 	} while (!ack_detected && pass++ < 10);
 
 	DRM_DEBUG_DRIVER("%s had to use fallback to %s ack, 0x%x (passes %u)\n",
 			 intel_uncore_forcewake_domain_to_str(d->id),
 			 type == ACK_SET ? "set" : "clear",
-			 __raw_i915_read32(i915, d->reg_ack),
+			 fw_ack(d),
 			 pass);
 
 	return ack_detected ? 0 : -ETIMEDOUT;
 }
 
 static inline void
-fw_domain_wait_ack_clear_fallback(const struct drm_i915_private *i915,
-				  const struct intel_uncore_forcewake_domain *d)
+fw_domain_wait_ack_clear_fallback(const struct intel_uncore_forcewake_domain *d)
 {
-	if (likely(!wait_ack_clear(i915, d, FORCEWAKE_KERNEL)))
+	if (likely(!wait_ack_clear(d, FORCEWAKE_KERNEL)))
 		return;
 
-	if (fw_domain_wait_ack_with_fallback(i915, d, ACK_CLEAR))
-		fw_domain_wait_ack_clear(i915, d);
+	if (fw_domain_wait_ack_with_fallback(d, ACK_CLEAR))
+		fw_domain_wait_ack_clear(d);
 }
 
 static inline void
-fw_domain_get(struct drm_i915_private *i915,
-	      const struct intel_uncore_forcewake_domain *d)
+fw_domain_get(const struct intel_uncore_forcewake_domain *d)
 {
-	__raw_i915_write32(i915, d->reg_set, i915->uncore.fw_set);
+	fw_set(d, FORCEWAKE_KERNEL);
 }
 
 static inline void
-fw_domain_wait_ack_set(const struct drm_i915_private *i915,
-		       const struct intel_uncore_forcewake_domain *d)
+fw_domain_wait_ack_set(const struct intel_uncore_forcewake_domain *d)
 {
-	if (wait_ack_set(i915, d, FORCEWAKE_KERNEL))
+	if (wait_ack_set(d, FORCEWAKE_KERNEL))
 		DRM_ERROR("%s: timed out waiting for forcewake ack request.\n",
 			  intel_uncore_forcewake_domain_to_str(d->id));
 }
 
 static inline void
-fw_domain_wait_ack_set_fallback(const struct drm_i915_private *i915,
-				const struct intel_uncore_forcewake_domain *d)
+fw_domain_wait_ack_set_fallback(const struct intel_uncore_forcewake_domain *d)
 {
-	if (likely(!wait_ack_set(i915, d, FORCEWAKE_KERNEL)))
+	if (likely(!wait_ack_set(d, FORCEWAKE_KERNEL)))
 		return;
 
-	if (fw_domain_wait_ack_with_fallback(i915, d, ACK_SET))
-		fw_domain_wait_ack_set(i915, d);
+	if (fw_domain_wait_ack_with_fallback(d, ACK_SET))
+		fw_domain_wait_ack_set(d);
 }
 
 static inline void
-fw_domain_put(const struct drm_i915_private *i915,
-	      const struct intel_uncore_forcewake_domain *d)
+fw_domain_put(const struct intel_uncore_forcewake_domain *d)
 {
-	__raw_i915_write32(i915, d->reg_set, i915->uncore.fw_clear);
+	fw_clear(d, FORCEWAKE_KERNEL);
 }
 
 static void
-fw_domains_get(struct drm_i915_private *i915, enum forcewake_domains fw_domains)
+fw_domains_get(struct intel_uncore *uncore, enum forcewake_domains fw_domains)
 {
 	struct intel_uncore_forcewake_domain *d;
 	unsigned int tmp;
 
-	GEM_BUG_ON(fw_domains & ~i915->uncore.fw_domains);
+	GEM_BUG_ON(fw_domains & ~uncore->fw_domains);
 
-	for_each_fw_domain_masked(d, fw_domains, i915, tmp) {
-		fw_domain_wait_ack_clear(i915, d);
-		fw_domain_get(i915, d);
+	for_each_fw_domain_masked(d, fw_domains, uncore, tmp) {
+		fw_domain_wait_ack_clear(d);
+		fw_domain_get(d);
 	}
 
-	for_each_fw_domain_masked(d, fw_domains, i915, tmp)
-		fw_domain_wait_ack_set(i915, d);
+	for_each_fw_domain_masked(d, fw_domains, uncore, tmp)
+		fw_domain_wait_ack_set(d);
 
-	i915->uncore.fw_domains_active |= fw_domains;
+	uncore->fw_domains_active |= fw_domains;
 }
 
 static void
-fw_domains_get_with_fallback(struct drm_i915_private *i915,
+fw_domains_get_with_fallback(struct intel_uncore *uncore,
 			     enum forcewake_domains fw_domains)
 {
 	struct intel_uncore_forcewake_domain *d;
 	unsigned int tmp;
 
-	GEM_BUG_ON(fw_domains & ~i915->uncore.fw_domains);
+	GEM_BUG_ON(fw_domains & ~uncore->fw_domains);
 
-	for_each_fw_domain_masked(d, fw_domains, i915, tmp) {
-		fw_domain_wait_ack_clear_fallback(i915, d);
-		fw_domain_get(i915, d);
+	for_each_fw_domain_masked(d, fw_domains, uncore, tmp) {
+		fw_domain_wait_ack_clear_fallback(d);
+		fw_domain_get(d);
 	}
 
-	for_each_fw_domain_masked(d, fw_domains, i915, tmp)
-		fw_domain_wait_ack_set_fallback(i915, d);
+	for_each_fw_domain_masked(d, fw_domains, uncore, tmp)
+		fw_domain_wait_ack_set_fallback(d);
 
-	i915->uncore.fw_domains_active |= fw_domains;
+	uncore->fw_domains_active |= fw_domains;
 }
 
 static void
-fw_domains_put(struct drm_i915_private *i915, enum forcewake_domains fw_domains)
+fw_domains_put(struct intel_uncore *uncore, enum forcewake_domains fw_domains)
 {
 	struct intel_uncore_forcewake_domain *d;
 	unsigned int tmp;
 
-	GEM_BUG_ON(fw_domains & ~i915->uncore.fw_domains);
+	GEM_BUG_ON(fw_domains & ~uncore->fw_domains);
 
-	for_each_fw_domain_masked(d, fw_domains, i915, tmp)
-		fw_domain_put(i915, d);
+	for_each_fw_domain_masked(d, fw_domains, uncore, tmp)
+		fw_domain_put(d);
 
-	i915->uncore.fw_domains_active &= ~fw_domains;
+	uncore->fw_domains_active &= ~fw_domains;
 }
 
 static void
-fw_domains_reset(struct drm_i915_private *i915,
+fw_domains_reset(struct intel_uncore *uncore,
 		 enum forcewake_domains fw_domains)
 {
 	struct intel_uncore_forcewake_domain *d;
@@ -277,61 +270,61 @@ fw_domains_reset(struct drm_i915_private *i915,
 	if (!fw_domains)
 		return;
 
-	GEM_BUG_ON(fw_domains & ~i915->uncore.fw_domains);
+	GEM_BUG_ON(fw_domains & ~uncore->fw_domains);
 
-	for_each_fw_domain_masked(d, fw_domains, i915, tmp)
-		fw_domain_reset(i915, d);
+	for_each_fw_domain_masked(d, fw_domains, uncore, tmp)
+		fw_domain_reset(d);
 }
 
-static inline u32 gt_thread_status(struct drm_i915_private *dev_priv)
+static inline u32 gt_thread_status(struct intel_uncore *uncore)
 {
 	u32 val;
 
-	val = __raw_i915_read32(dev_priv, GEN6_GT_THREAD_STATUS_REG);
+	val = __raw_uncore_read32(uncore, GEN6_GT_THREAD_STATUS_REG);
 	val &= GEN6_GT_THREAD_STATUS_CORE_MASK;
 
 	return val;
 }
 
-static void __gen6_gt_wait_for_thread_c0(struct drm_i915_private *dev_priv)
+static void __gen6_gt_wait_for_thread_c0(struct intel_uncore *uncore)
 {
 	/*
 	 * w/a for a sporadic read returning 0 by waiting for the GT
 	 * thread to wake up.
 	 */
-	WARN_ONCE(wait_for_atomic_us(gt_thread_status(dev_priv) == 0, 5000),
+	WARN_ONCE(wait_for_atomic_us(gt_thread_status(uncore) == 0, 5000),
 		  "GT thread status wait timed out\n");
 }
 
-static void fw_domains_get_with_thread_status(struct drm_i915_private *dev_priv,
+static void fw_domains_get_with_thread_status(struct intel_uncore *uncore,
 					      enum forcewake_domains fw_domains)
 {
-	fw_domains_get(dev_priv, fw_domains);
+	fw_domains_get(uncore, fw_domains);
 
 	/* WaRsForcewakeWaitTC0:snb,ivb,hsw,bdw,vlv */
-	__gen6_gt_wait_for_thread_c0(dev_priv);
+	__gen6_gt_wait_for_thread_c0(uncore);
 }
 
-static inline u32 fifo_free_entries(struct drm_i915_private *dev_priv)
+static inline u32 fifo_free_entries(struct intel_uncore *uncore)
 {
-	u32 count = __raw_i915_read32(dev_priv, GTFIFOCTL);
+	u32 count = __raw_uncore_read32(uncore, GTFIFOCTL);
 
 	return count & GT_FIFO_FREE_ENTRIES_MASK;
 }
 
-static void __gen6_gt_wait_for_fifo(struct drm_i915_private *dev_priv)
+static void __gen6_gt_wait_for_fifo(struct intel_uncore *uncore)
 {
 	u32 n;
 
 	/* On VLV, FIFO will be shared by both SW and HW.
 	 * So, we need to read the FREE_ENTRIES everytime */
-	if (IS_VALLEYVIEW(dev_priv))
-		n = fifo_free_entries(dev_priv);
+	if (IS_VALLEYVIEW(uncore_to_i915(uncore)))
+		n = fifo_free_entries(uncore);
 	else
-		n = dev_priv->uncore.fifo_count;
+		n = uncore->fifo_count;
 
 	if (n <= GT_FIFO_NUM_RESERVED_ENTRIES) {
-		if (wait_for_atomic((n = fifo_free_entries(dev_priv)) >
+		if (wait_for_atomic((n = fifo_free_entries(uncore)) >
 				    GT_FIFO_NUM_RESERVED_ENTRIES,
 				    GT_FIFO_TIMEOUT_MS)) {
 			DRM_DEBUG("GT_FIFO timeout, entries: %u\n", n);
@@ -339,7 +332,7 @@ static void __gen6_gt_wait_for_fifo(struct drm_i915_private *dev_priv)
 		}
 	}
 
-	dev_priv->uncore.fifo_count = n - 1;
+	uncore->fifo_count = n - 1;
 }
 
 static enum hrtimer_restart
@@ -347,30 +340,29 @@ intel_uncore_fw_release_timer(struct hrtimer *timer)
 {
 	struct intel_uncore_forcewake_domain *domain =
 	       container_of(timer, struct intel_uncore_forcewake_domain, timer);
-	struct drm_i915_private *dev_priv =
-		container_of(domain, struct drm_i915_private, uncore.fw_domain[domain->id]);
+	struct intel_uncore *uncore = forcewake_domain_to_uncore(domain);
 	unsigned long irqflags;
 
-	assert_rpm_device_not_suspended(dev_priv);
+	assert_rpm_device_not_suspended(uncore->rpm);
 
 	if (xchg(&domain->active, false))
 		return HRTIMER_RESTART;
 
-	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
+	spin_lock_irqsave(&uncore->lock, irqflags);
 	if (WARN_ON(domain->wake_count == 0))
 		domain->wake_count++;
 
 	if (--domain->wake_count == 0)
-		dev_priv->uncore.funcs.force_wake_put(dev_priv, domain->mask);
+		uncore->funcs.force_wake_put(uncore, domain->mask);
 
-	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
+	spin_unlock_irqrestore(&uncore->lock, irqflags);
 
 	return HRTIMER_NORESTART;
 }
 
 /* Note callers must have acquired the PUNIT->PMIC bus, before calling this. */
 static unsigned int
-intel_uncore_forcewake_reset(struct drm_i915_private *dev_priv)
+intel_uncore_forcewake_reset(struct intel_uncore *uncore)
 {
 	unsigned long irqflags;
 	struct intel_uncore_forcewake_domain *domain;
@@ -388,7 +380,7 @@ intel_uncore_forcewake_reset(struct drm_i915_private *dev_priv)
 
 		active_domains = 0;
 
-		for_each_fw_domain(domain, dev_priv, tmp) {
+		for_each_fw_domain(domain, uncore, tmp) {
 			smp_store_mb(domain->active, false);
 			if (hrtimer_cancel(&domain->timer) == 0)
 				continue;
@@ -396,9 +388,9 @@ intel_uncore_forcewake_reset(struct drm_i915_private *dev_priv)
 			intel_uncore_fw_release_timer(&domain->timer);
 		}
 
-		spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
+		spin_lock_irqsave(&uncore->lock, irqflags);
 
-		for_each_fw_domain(domain, dev_priv, tmp) {
+		for_each_fw_domain(domain, uncore, tmp) {
 			if (hrtimer_active(&domain->timer))
 				active_domains |= domain->mask;
 		}
@@ -411,185 +403,134 @@ intel_uncore_forcewake_reset(struct drm_i915_private *dev_priv)
 			break;
 		}
 
-		spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
+		spin_unlock_irqrestore(&uncore->lock, irqflags);
 		cond_resched();
 	}
 
 	WARN_ON(active_domains);
 
-	fw = dev_priv->uncore.fw_domains_active;
+	fw = uncore->fw_domains_active;
 	if (fw)
-		dev_priv->uncore.funcs.force_wake_put(dev_priv, fw);
+		uncore->funcs.force_wake_put(uncore, fw);
 
-	fw_domains_reset(dev_priv, dev_priv->uncore.fw_domains);
-	assert_forcewakes_inactive(dev_priv);
+	fw_domains_reset(uncore, uncore->fw_domains);
+	assert_forcewakes_inactive(uncore);
 
-	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
+	spin_unlock_irqrestore(&uncore->lock, irqflags);
 
 	return fw; /* track the lost user forcewake domains */
 }
 
-static u64 gen9_edram_size(struct drm_i915_private *dev_priv)
-{
-	const unsigned int ways[8] = { 4, 8, 12, 16, 16, 16, 16, 16 };
-	const unsigned int sets[4] = { 1, 1, 2, 2 };
-	const u32 cap = dev_priv->edram_cap;
-
-	return EDRAM_NUM_BANKS(cap) *
-		ways[EDRAM_WAYS_IDX(cap)] *
-		sets[EDRAM_SETS_IDX(cap)] *
-		1024 * 1024;
-}
-
-u64 intel_uncore_edram_size(struct drm_i915_private *dev_priv)
-{
-	if (!HAS_EDRAM(dev_priv))
-		return 0;
-
-	/* The needed capability bits for size calculation
-	 * are not there with pre gen9 so return 128MB always.
-	 */
-	if (INTEL_GEN(dev_priv) < 9)
-		return 128 * 1024 * 1024;
-
-	return gen9_edram_size(dev_priv);
-}
-
-static void intel_uncore_edram_detect(struct drm_i915_private *dev_priv)
-{
-	if (IS_HASWELL(dev_priv) ||
-	    IS_BROADWELL(dev_priv) ||
-	    INTEL_GEN(dev_priv) >= 9) {
-		dev_priv->edram_cap = __raw_i915_read32(dev_priv,
-							HSW_EDRAM_CAP);
-
-		/* NB: We can't write IDICR yet because we do not have gt funcs
-		 * set up */
-	} else {
-		dev_priv->edram_cap = 0;
-	}
-
-	if (HAS_EDRAM(dev_priv))
-		DRM_INFO("Found %lluMB of eDRAM\n",
-			 intel_uncore_edram_size(dev_priv) / (1024 * 1024));
-}
-
 static bool
-fpga_check_for_unclaimed_mmio(struct drm_i915_private *dev_priv)
+fpga_check_for_unclaimed_mmio(struct intel_uncore *uncore)
 {
 	u32 dbg;
 
-	dbg = __raw_i915_read32(dev_priv, FPGA_DBG);
+	dbg = __raw_uncore_read32(uncore, FPGA_DBG);
 	if (likely(!(dbg & FPGA_DBG_RM_NOCLAIM)))
 		return false;
 
-	__raw_i915_write32(dev_priv, FPGA_DBG, FPGA_DBG_RM_NOCLAIM);
+	__raw_uncore_write32(uncore, FPGA_DBG, FPGA_DBG_RM_NOCLAIM);
 
 	return true;
 }
 
 static bool
-vlv_check_for_unclaimed_mmio(struct drm_i915_private *dev_priv)
+vlv_check_for_unclaimed_mmio(struct intel_uncore *uncore)
 {
 	u32 cer;
 
-	cer = __raw_i915_read32(dev_priv, CLAIM_ER);
+	cer = __raw_uncore_read32(uncore, CLAIM_ER);
 	if (likely(!(cer & (CLAIM_ER_OVERFLOW | CLAIM_ER_CTR_MASK))))
 		return false;
 
-	__raw_i915_write32(dev_priv, CLAIM_ER, CLAIM_ER_CLR);
+	__raw_uncore_write32(uncore, CLAIM_ER, CLAIM_ER_CLR);
 
 	return true;
 }
 
 static bool
-gen6_check_for_fifo_debug(struct drm_i915_private *dev_priv)
+gen6_check_for_fifo_debug(struct intel_uncore *uncore)
 {
 	u32 fifodbg;
 
-	fifodbg = __raw_i915_read32(dev_priv, GTFIFODBG);
+	fifodbg = __raw_uncore_read32(uncore, GTFIFODBG);
 
 	if (unlikely(fifodbg)) {
 		DRM_DEBUG_DRIVER("GTFIFODBG = 0x08%x\n", fifodbg);
-		__raw_i915_write32(dev_priv, GTFIFODBG, fifodbg);
+		__raw_uncore_write32(uncore, GTFIFODBG, fifodbg);
 	}
 
 	return fifodbg;
 }
 
 static bool
-check_for_unclaimed_mmio(struct drm_i915_private *dev_priv)
+check_for_unclaimed_mmio(struct intel_uncore *uncore)
 {
 	bool ret = false;
 
-	if (HAS_FPGA_DBG_UNCLAIMED(dev_priv))
-		ret |= fpga_check_for_unclaimed_mmio(dev_priv);
+	if (intel_uncore_has_fpga_dbg_unclaimed(uncore))
+		ret |= fpga_check_for_unclaimed_mmio(uncore);
 
-	if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
-		ret |= vlv_check_for_unclaimed_mmio(dev_priv);
+	if (intel_uncore_has_dbg_unclaimed(uncore))
+		ret |= vlv_check_for_unclaimed_mmio(uncore);
 
-	if (IS_GEN_RANGE(dev_priv, 6, 7))
-		ret |= gen6_check_for_fifo_debug(dev_priv);
+	if (intel_uncore_has_fifo(uncore))
+		ret |= gen6_check_for_fifo_debug(uncore);
 
 	return ret;
 }
 
-static void __intel_uncore_early_sanitize(struct drm_i915_private *dev_priv,
+static void __intel_uncore_early_sanitize(struct intel_uncore *uncore,
 					  unsigned int restore_forcewake)
 {
 	/* clear out unclaimed reg detection bit */
-	if (check_for_unclaimed_mmio(dev_priv))
+	if (check_for_unclaimed_mmio(uncore))
 		DRM_DEBUG("unclaimed mmio detected on uncore init, clearing\n");
 
 	/* WaDisableShadowRegForCpd:chv */
-	if (IS_CHERRYVIEW(dev_priv)) {
-		__raw_i915_write32(dev_priv, GTFIFOCTL,
-				   __raw_i915_read32(dev_priv, GTFIFOCTL) |
-				   GT_FIFO_CTL_BLOCK_ALL_POLICY_STALL |
-				   GT_FIFO_CTL_RC6_POLICY_STALL);
+	if (IS_CHERRYVIEW(uncore_to_i915(uncore))) {
+		__raw_uncore_write32(uncore, GTFIFOCTL,
+				     __raw_uncore_read32(uncore, GTFIFOCTL) |
+				     GT_FIFO_CTL_BLOCK_ALL_POLICY_STALL |
+				     GT_FIFO_CTL_RC6_POLICY_STALL);
 	}
 
 	iosf_mbi_punit_acquire();
-	intel_uncore_forcewake_reset(dev_priv);
+	intel_uncore_forcewake_reset(uncore);
 	if (restore_forcewake) {
-		spin_lock_irq(&dev_priv->uncore.lock);
-		dev_priv->uncore.funcs.force_wake_get(dev_priv,
-						      restore_forcewake);
-
-		if (IS_GEN_RANGE(dev_priv, 6, 7))
-			dev_priv->uncore.fifo_count =
-				fifo_free_entries(dev_priv);
-		spin_unlock_irq(&dev_priv->uncore.lock);
+		spin_lock_irq(&uncore->lock);
+		uncore->funcs.force_wake_get(uncore, restore_forcewake);
+
+		if (intel_uncore_has_fifo(uncore))
+			uncore->fifo_count = fifo_free_entries(uncore);
+		spin_unlock_irq(&uncore->lock);
 	}
 	iosf_mbi_punit_release();
 }
 
-void intel_uncore_suspend(struct drm_i915_private *dev_priv)
+void intel_uncore_suspend(struct intel_uncore *uncore)
 {
 	iosf_mbi_punit_acquire();
 	iosf_mbi_unregister_pmic_bus_access_notifier_unlocked(
-		&dev_priv->uncore.pmic_bus_access_nb);
-	dev_priv->uncore.fw_domains_saved =
-		intel_uncore_forcewake_reset(dev_priv);
+		&uncore->pmic_bus_access_nb);
+	uncore->fw_domains_saved = intel_uncore_forcewake_reset(uncore);
 	iosf_mbi_punit_release();
 }
 
-void intel_uncore_resume_early(struct drm_i915_private *dev_priv)
+void intel_uncore_resume_early(struct intel_uncore *uncore)
 {
 	unsigned int restore_forcewake;
 
-	restore_forcewake = fetch_and_zero(&dev_priv->uncore.fw_domains_saved);
-	__intel_uncore_early_sanitize(dev_priv, restore_forcewake);
+	restore_forcewake = fetch_and_zero(&uncore->fw_domains_saved);
+	__intel_uncore_early_sanitize(uncore, restore_forcewake);
 
-	iosf_mbi_register_pmic_bus_access_notifier(
-		&dev_priv->uncore.pmic_bus_access_nb);
-	i915_check_and_clear_faults(dev_priv);
+	iosf_mbi_register_pmic_bus_access_notifier(&uncore->pmic_bus_access_nb);
 }
 
-void intel_uncore_runtime_resume(struct drm_i915_private *dev_priv)
+void intel_uncore_runtime_resume(struct intel_uncore *uncore)
 {
-	iosf_mbi_register_pmic_bus_access_notifier(
-		&dev_priv->uncore.pmic_bus_access_nb);
+	iosf_mbi_register_pmic_bus_access_notifier(&uncore->pmic_bus_access_nb);
 }
 
 void intel_uncore_sanitize(struct drm_i915_private *dev_priv)
@@ -598,15 +539,15 @@ void intel_uncore_sanitize(struct drm_i915_private *dev_priv)
 	intel_sanitize_gt_powersave(dev_priv);
 }
 
-static void __intel_uncore_forcewake_get(struct drm_i915_private *dev_priv,
+static void __intel_uncore_forcewake_get(struct intel_uncore *uncore,
 					 enum forcewake_domains fw_domains)
 {
 	struct intel_uncore_forcewake_domain *domain;
 	unsigned int tmp;
 
-	fw_domains &= dev_priv->uncore.fw_domains;
+	fw_domains &= uncore->fw_domains;
 
-	for_each_fw_domain_masked(domain, fw_domains, dev_priv, tmp) {
+	for_each_fw_domain_masked(domain, fw_domains, uncore, tmp) {
 		if (domain->wake_count++) {
 			fw_domains &= ~domain->mask;
 			domain->active = true;
@@ -614,12 +555,12 @@ static void __intel_uncore_forcewake_get(struct drm_i915_private *dev_priv,
 	}
 
 	if (fw_domains)
-		dev_priv->uncore.funcs.force_wake_get(dev_priv, fw_domains);
+		uncore->funcs.force_wake_get(uncore, fw_domains);
 }
 
 /**
  * intel_uncore_forcewake_get - grab forcewake domain references
- * @dev_priv: i915 device instance
+ * @uncore: the intel_uncore structure
  * @fw_domains: forcewake domains to get reference on
  *
  * This function can be used get GT's forcewake domain references.
@@ -630,100 +571,100 @@ static void __intel_uncore_forcewake_get(struct drm_i915_private *dev_priv,
  * call to intel_unforce_forcewake_put(). Usually caller wants all the domains
  * to be kept awake so the @fw_domains would be then FORCEWAKE_ALL.
  */
-void intel_uncore_forcewake_get(struct drm_i915_private *dev_priv,
+void intel_uncore_forcewake_get(struct intel_uncore *uncore,
 				enum forcewake_domains fw_domains)
 {
 	unsigned long irqflags;
 
-	if (!dev_priv->uncore.funcs.force_wake_get)
+	if (!uncore->funcs.force_wake_get)
 		return;
 
-	assert_rpm_wakelock_held(dev_priv);
+	__assert_rpm_wakelock_held(uncore->rpm);
 
-	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
-	__intel_uncore_forcewake_get(dev_priv, fw_domains);
-	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
+	spin_lock_irqsave(&uncore->lock, irqflags);
+	__intel_uncore_forcewake_get(uncore, fw_domains);
+	spin_unlock_irqrestore(&uncore->lock, irqflags);
 }
 
 /**
  * intel_uncore_forcewake_user_get - claim forcewake on behalf of userspace
- * @dev_priv: i915 device instance
+ * @uncore: the intel_uncore structure
  *
  * This function is a wrapper around intel_uncore_forcewake_get() to acquire
  * the GT powerwell and in the process disable our debugging for the
  * duration of userspace's bypass.
  */
-void intel_uncore_forcewake_user_get(struct drm_i915_private *dev_priv)
+void intel_uncore_forcewake_user_get(struct intel_uncore *uncore)
 {
-	spin_lock_irq(&dev_priv->uncore.lock);
-	if (!dev_priv->uncore.user_forcewake.count++) {
-		intel_uncore_forcewake_get__locked(dev_priv, FORCEWAKE_ALL);
+	spin_lock_irq(&uncore->lock);
+	if (!uncore->user_forcewake.count++) {
+		intel_uncore_forcewake_get__locked(uncore, FORCEWAKE_ALL);
 
 		/* Save and disable mmio debugging for the user bypass */
-		dev_priv->uncore.user_forcewake.saved_mmio_check =
-			dev_priv->uncore.unclaimed_mmio_check;
-		dev_priv->uncore.user_forcewake.saved_mmio_debug =
+		uncore->user_forcewake.saved_mmio_check =
+			uncore->unclaimed_mmio_check;
+		uncore->user_forcewake.saved_mmio_debug =
 			i915_modparams.mmio_debug;
 
-		dev_priv->uncore.unclaimed_mmio_check = 0;
+		uncore->unclaimed_mmio_check = 0;
 		i915_modparams.mmio_debug = 0;
 	}
-	spin_unlock_irq(&dev_priv->uncore.lock);
+	spin_unlock_irq(&uncore->lock);
 }
 
 /**
  * intel_uncore_forcewake_user_put - release forcewake on behalf of userspace
- * @dev_priv: i915 device instance
+ * @uncore: the intel_uncore structure
  *
  * This function complements intel_uncore_forcewake_user_get() and releases
  * the GT powerwell taken on behalf of the userspace bypass.
  */
-void intel_uncore_forcewake_user_put(struct drm_i915_private *dev_priv)
+void intel_uncore_forcewake_user_put(struct intel_uncore *uncore)
 {
-	spin_lock_irq(&dev_priv->uncore.lock);
-	if (!--dev_priv->uncore.user_forcewake.count) {
-		if (intel_uncore_unclaimed_mmio(dev_priv))
-			dev_info(dev_priv->drm.dev,
+	spin_lock_irq(&uncore->lock);
+	if (!--uncore->user_forcewake.count) {
+		if (intel_uncore_unclaimed_mmio(uncore))
+			dev_info(uncore_to_i915(uncore)->drm.dev,
 				 "Invalid mmio detected during user access\n");
 
-		dev_priv->uncore.unclaimed_mmio_check =
-			dev_priv->uncore.user_forcewake.saved_mmio_check;
+		uncore->unclaimed_mmio_check =
+			uncore->user_forcewake.saved_mmio_check;
 		i915_modparams.mmio_debug =
-			dev_priv->uncore.user_forcewake.saved_mmio_debug;
+			uncore->user_forcewake.saved_mmio_debug;
 
-		intel_uncore_forcewake_put__locked(dev_priv, FORCEWAKE_ALL);
+		intel_uncore_forcewake_put__locked(uncore, FORCEWAKE_ALL);
 	}
-	spin_unlock_irq(&dev_priv->uncore.lock);
+	spin_unlock_irq(&uncore->lock);
 }
 
 /**
  * intel_uncore_forcewake_get__locked - grab forcewake domain references
- * @dev_priv: i915 device instance
+ * @uncore: the intel_uncore structure
  * @fw_domains: forcewake domains to get reference on
  *
  * See intel_uncore_forcewake_get(). This variant places the onus
  * on the caller to explicitly handle the dev_priv->uncore.lock spinlock.
  */
-void intel_uncore_forcewake_get__locked(struct drm_i915_private *dev_priv,
+void intel_uncore_forcewake_get__locked(struct intel_uncore *uncore,
 					enum forcewake_domains fw_domains)
 {
-	lockdep_assert_held(&dev_priv->uncore.lock);
+	lockdep_assert_held(&uncore->lock);
 
-	if (!dev_priv->uncore.funcs.force_wake_get)
+	if (!uncore->funcs.force_wake_get)
 		return;
 
-	__intel_uncore_forcewake_get(dev_priv, fw_domains);
+	__intel_uncore_forcewake_get(uncore, fw_domains);
 }
 
-static void __intel_uncore_forcewake_put(struct drm_i915_private *dev_priv,
+static void __intel_uncore_forcewake_put(struct intel_uncore *uncore,
 					 enum forcewake_domains fw_domains)
 {
 	struct intel_uncore_forcewake_domain *domain;
 	unsigned int tmp;
 
-	fw_domains &= dev_priv->uncore.fw_domains;
+	fw_domains &= uncore->fw_domains;
 
-	for_each_fw_domain_masked(domain, fw_domains, dev_priv, tmp) {
+	for_each_fw_domain_masked(domain, fw_domains, uncore, tmp) {
 		if (WARN_ON(domain->wake_count == 0))
 			continue;
 
@@ -738,66 +679,66 @@ static void __intel_uncore_forcewake_put(struct drm_i915_private *dev_priv,
 
 /**
  * intel_uncore_forcewake_put - release a forcewake domain reference
- * @dev_priv: i915 device instance
+ * @uncore: the intel_uncore structure
  * @fw_domains: forcewake domains to put references
  *
  * This function drops the device-level forcewakes for specified
  * domains obtained by intel_uncore_forcewake_get().
  */
-void intel_uncore_forcewake_put(struct drm_i915_private *dev_priv,
+void intel_uncore_forcewake_put(struct intel_uncore *uncore,
 				enum forcewake_domains fw_domains)
 {
 	unsigned long irqflags;
 
-	if (!dev_priv->uncore.funcs.force_wake_put)
+	if (!uncore->funcs.force_wake_put)
 		return;
 
-	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
-	__intel_uncore_forcewake_put(dev_priv, fw_domains);
-	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
+	spin_lock_irqsave(&uncore->lock, irqflags);
+	__intel_uncore_forcewake_put(uncore, fw_domains);
+	spin_unlock_irqrestore(&uncore->lock, irqflags);
 }
 
 /**
  * intel_uncore_forcewake_put__locked - grab forcewake domain references
- * @dev_priv: i915 device instance
+ * @uncore: the intel_uncore structure
  * @fw_domains: forcewake domains to get reference on
  *
  * See intel_uncore_forcewake_put(). This variant places the onus
  * on the caller to explicitly handle the dev_priv->uncore.lock spinlock.
  */
-void intel_uncore_forcewake_put__locked(struct drm_i915_private *dev_priv,
+void intel_uncore_forcewake_put__locked(struct intel_uncore *uncore,
 					enum forcewake_domains fw_domains)
 {
-	lockdep_assert_held(&dev_priv->uncore.lock);
+	lockdep_assert_held(&uncore->lock);
 
-	if (!dev_priv->uncore.funcs.force_wake_put)
+	if (!uncore->funcs.force_wake_put)
 		return;
 
-	__intel_uncore_forcewake_put(dev_priv, fw_domains);
+	__intel_uncore_forcewake_put(uncore, fw_domains);
 }
 
-void assert_forcewakes_inactive(struct drm_i915_private *dev_priv)
+void assert_forcewakes_inactive(struct intel_uncore *uncore)
 {
-	if (!dev_priv->uncore.funcs.force_wake_get)
+	if (!uncore->funcs.force_wake_get)
 		return;
 
-	WARN(dev_priv->uncore.fw_domains_active,
+	WARN(uncore->fw_domains_active,
 	     "Expected all fw_domains to be inactive, but %08x are still on\n",
-	     dev_priv->uncore.fw_domains_active);
+	     uncore->fw_domains_active);
 }
 
-void assert_forcewakes_active(struct drm_i915_private *dev_priv,
+void assert_forcewakes_active(struct intel_uncore *uncore,
 			      enum forcewake_domains fw_domains)
 {
-	if (!dev_priv->uncore.funcs.force_wake_get)
+	if (!uncore->funcs.force_wake_get)
 		return;
 
-	assert_rpm_wakelock_held(dev_priv);
+	__assert_rpm_wakelock_held(uncore->rpm);
 
-	fw_domains &= dev_priv->uncore.fw_domains;
-	WARN(fw_domains & ~dev_priv->uncore.fw_domains_active,
+	fw_domains &= uncore->fw_domains;
+	WARN(fw_domains & ~uncore->fw_domains_active,
 	     "Expected %08x fw_domains to be active, but %08x are off\n",
-	     fw_domains, fw_domains & ~dev_priv->uncore.fw_domains_active);
+	     fw_domains, fw_domains & ~uncore->fw_domains_active);
 }
 
 /* We give fast paths for the really cool registers */
@@ -806,7 +747,7 @@ void assert_forcewakes_active(struct drm_i915_private *dev_priv,
 #define GEN11_NEEDS_FORCE_WAKE(reg) \
 	((reg) < 0x40000 || ((reg) >= 0x1c0000 && (reg) < 0x1dc000))
 
-#define __gen6_reg_read_fw_domains(offset) \
+#define __gen6_reg_read_fw_domains(uncore, offset) \
 ({ \
 	enum forcewake_domains __fwd; \
 	if (NEEDS_FORCE_WAKE(offset)) \
@@ -846,13 +787,13 @@ static int fw_range_cmp(u32 offset, const struct intel_forcewake_range *entry)
 })
 
 static enum forcewake_domains
-find_fw_domain(struct drm_i915_private *dev_priv, u32 offset)
+find_fw_domain(struct intel_uncore *uncore, u32 offset)
 {
 	const struct intel_forcewake_range *entry;
 
 	entry = BSEARCH(offset,
-			dev_priv->uncore.fw_domains_table,
-			dev_priv->uncore.fw_domains_table_entries,
+			uncore->fw_domains_table,
+			uncore->fw_domains_table_entries,
 			fw_range_cmp);
 
 	if (!entry)
@@ -864,11 +805,11 @@ find_fw_domain(struct drm_i915_private *dev_priv, u32 offset)
 	 * translate it here to the list of available domains.
 	 */
 	if (entry->domains == FORCEWAKE_ALL)
-		return dev_priv->uncore.fw_domains;
+		return uncore->fw_domains;
 
-	WARN(entry->domains & ~dev_priv->uncore.fw_domains,
+	WARN(entry->domains & ~uncore->fw_domains,
 	     "Uninitialized forcewake domain(s) 0x%x accessed at 0x%x\n",
-	     entry->domains & ~dev_priv->uncore.fw_domains, offset);
+	     entry->domains & ~uncore->fw_domains, offset);
 
 	return entry->domains;
 }
@@ -892,19 +833,19 @@ static const struct intel_forcewake_range __vlv_fw_ranges[] = {
 	GEN_FW_RANGE(0x30000, 0x3ffff, FORCEWAKE_MEDIA),
 };
 
-#define __fwtable_reg_read_fw_domains(offset) \
+#define __fwtable_reg_read_fw_domains(uncore, offset) \
 ({ \
 	enum forcewake_domains __fwd = 0; \
 	if (NEEDS_FORCE_WAKE((offset))) \
-		__fwd = find_fw_domain(dev_priv, offset); \
+		__fwd = find_fw_domain(uncore, offset); \
 	__fwd; \
 })
 
-#define __gen11_fwtable_reg_read_fw_domains(offset) \
+#define __gen11_fwtable_reg_read_fw_domains(uncore, offset) \
 ({ \
 	enum forcewake_domains __fwd = 0; \
 	if (GEN11_NEEDS_FORCE_WAKE((offset))) \
-		__fwd = find_fw_domain(dev_priv, offset); \
+		__fwd = find_fw_domain(uncore, offset); \
 	__fwd; \
 })
 
@@ -956,7 +897,7 @@ static bool is_gen##x##_shadowed(u32 offset) \
 __is_genX_shadowed(8)
 __is_genX_shadowed(11)
 
-#define __gen8_reg_write_fw_domains(offset) \
+#define __gen8_reg_write_fw_domains(uncore, offset) \
 ({ \
 	enum forcewake_domains __fwd; \
 	if (NEEDS_FORCE_WAKE(offset) && !is_gen8_shadowed(offset)) \
@@ -986,19 +927,19 @@ static const struct intel_forcewake_range __chv_fw_ranges[] = {
 	GEN_FW_RANGE(0x30000, 0x37fff, FORCEWAKE_MEDIA),
 };
 
-#define __fwtable_reg_write_fw_domains(offset) \
+#define __fwtable_reg_write_fw_domains(uncore, offset) \
 ({ \
 	enum forcewake_domains __fwd = 0; \
 	if (NEEDS_FORCE_WAKE((offset)) && !is_gen8_shadowed(offset)) \
-		__fwd = find_fw_domain(dev_priv, offset); \
+		__fwd = find_fw_domain(uncore, offset); \
 	__fwd; \
 })
 
-#define __gen11_fwtable_reg_write_fw_domains(offset) \
+#define __gen11_fwtable_reg_write_fw_domains(uncore, offset) \
 ({ \
 	enum forcewake_domains __fwd = 0; \
 	if (GEN11_NEEDS_FORCE_WAKE((offset)) && !is_gen11_shadowed(offset)) \
-		__fwd = find_fw_domain(dev_priv, offset); \
+		__fwd = find_fw_domain(uncore, offset); \
 	__fwd; \
 })
 
@@ -1073,21 +1014,21 @@ static const struct intel_forcewake_range __gen11_fw_ranges[] = {
 };
 
 static void
-ilk_dummy_write(struct drm_i915_private *dev_priv)
+ilk_dummy_write(struct intel_uncore *uncore)
 {
 	/* WaIssueDummyWriteToWakeupFromRC6:ilk Issue a dummy write to wake up
 	 * the chip from rc6 before touching it for real. MI_MODE is masked,
 	 * hence harmless to write 0 into. */
-	__raw_i915_write32(dev_priv, MI_MODE, 0);
+	__raw_uncore_write32(uncore, MI_MODE, 0);
 }
 
 static void
-__unclaimed_reg_debug(struct drm_i915_private *dev_priv,
+__unclaimed_reg_debug(struct intel_uncore *uncore,
 		      const i915_reg_t reg,
 		      const bool read,
 		      const bool before)
 {
-	if (WARN(check_for_unclaimed_mmio(dev_priv) && !before,
+	if (WARN(check_for_unclaimed_mmio(uncore) && !before,
 		 "Unclaimed %s register 0x%x\n",
 		 read ? "read from" : "write to",
 		 i915_mmio_reg_offset(reg)))
@@ -1096,7 +1037,7 @@ __unclaimed_reg_debug(struct drm_i915_private *dev_priv,
 }
 
 static inline void
-unclaimed_reg_debug(struct drm_i915_private *dev_priv,
+unclaimed_reg_debug(struct intel_uncore *uncore,
 		    const i915_reg_t reg,
 		    const bool read,
 		    const bool before)
@@ -1104,12 +1045,12 @@ unclaimed_reg_debug(struct drm_i915_private *dev_priv,
 	if (likely(!i915_modparams.mmio_debug))
 		return;
 
-	__unclaimed_reg_debug(dev_priv, reg, read, before);
+	__unclaimed_reg_debug(uncore, reg, read, before);
 }
 
 #define GEN2_READ_HEADER(x) \
 	u##x val = 0; \
-	assert_rpm_wakelock_held(dev_priv);
+	__assert_rpm_wakelock_held(uncore->rpm);
 
 #define GEN2_READ_FOOTER \
 	trace_i915_reg_rw(false, reg, val, sizeof(val), trace); \
@@ -1117,18 +1058,18 @@ unclaimed_reg_debug(struct drm_i915_private *dev_priv,
 
 #define __gen2_read(x) \
 static u##x \
-gen2_read##x(struct drm_i915_private *dev_priv, i915_reg_t reg, bool trace) { \
+gen2_read##x(struct intel_uncore *uncore, i915_reg_t reg, bool trace) { \
 	GEN2_READ_HEADER(x); \
-	val = __raw_i915_read##x(dev_priv, reg); \
+	val = __raw_uncore_read##x(uncore, reg); \
 	GEN2_READ_FOOTER; \
 }
 
 #define __gen5_read(x) \
 static u##x \
-gen5_read##x(struct drm_i915_private *dev_priv, i915_reg_t reg, bool trace) { \
+gen5_read##x(struct intel_uncore *uncore, i915_reg_t reg, bool trace) { \
 	GEN2_READ_HEADER(x); \
-	ilk_dummy_write(dev_priv); \
-	val = __raw_i915_read##x(dev_priv, reg); \
+	ilk_dummy_write(uncore); \
+	val = __raw_uncore_read##x(uncore, reg); \
 	GEN2_READ_FOOTER; \
 }
 
@@ -1151,53 +1092,53 @@ __gen2_read(64)
 	u32 offset = i915_mmio_reg_offset(reg); \
 	unsigned long irqflags; \
 	u##x val = 0; \
-	assert_rpm_wakelock_held(dev_priv); \
-	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags); \
-	unclaimed_reg_debug(dev_priv, reg, true, true)
+	__assert_rpm_wakelock_held(uncore->rpm); \
+	spin_lock_irqsave(&uncore->lock, irqflags); \
+	unclaimed_reg_debug(uncore, reg, true, true)
 
 #define GEN6_READ_FOOTER \
-	unclaimed_reg_debug(dev_priv, reg, true, false); \
-	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags); \
+	unclaimed_reg_debug(uncore, reg, true, false); \
+	spin_unlock_irqrestore(&uncore->lock, irqflags); \
 	trace_i915_reg_rw(false, reg, val, sizeof(val), trace); \
 	return val
 
-static noinline void ___force_wake_auto(struct drm_i915_private *dev_priv,
+static noinline void ___force_wake_auto(struct intel_uncore *uncore,
 					enum forcewake_domains fw_domains)
 {
 	struct intel_uncore_forcewake_domain *domain;
 	unsigned int tmp;
 
-	GEM_BUG_ON(fw_domains & ~dev_priv->uncore.fw_domains);
+	GEM_BUG_ON(fw_domains & ~uncore->fw_domains);
 
-	for_each_fw_domain_masked(domain, fw_domains, dev_priv, tmp)
+	for_each_fw_domain_masked(domain, fw_domains, uncore, tmp)
 		fw_domain_arm_timer(domain);
 
-	dev_priv->uncore.funcs.force_wake_get(dev_priv, fw_domains);
+	uncore->funcs.force_wake_get(uncore, fw_domains);
 }
 
-static inline void __force_wake_auto(struct drm_i915_private *dev_priv,
+static inline void __force_wake_auto(struct intel_uncore *uncore,
 				     enum forcewake_domains fw_domains)
 {
 	if (WARN_ON(!fw_domains))
 		return;
 
 	/* Turn on all requested but inactive supported forcewake domains. */
-	fw_domains &= dev_priv->uncore.fw_domains;
-	fw_domains &= ~dev_priv->uncore.fw_domains_active;
+	fw_domains &= uncore->fw_domains;
+	fw_domains &= ~uncore->fw_domains_active;
 
 	if (fw_domains)
-		___force_wake_auto(dev_priv, fw_domains);
+		___force_wake_auto(uncore, fw_domains);
 }
 
 #define __gen_read(func, x) \
 static u##x \
-func##_read##x(struct drm_i915_private *dev_priv, i915_reg_t reg, bool trace) { \
+func##_read##x(struct intel_uncore *uncore, i915_reg_t reg, bool trace) { \
 	enum forcewake_domains fw_engine; \
 	GEN6_READ_HEADER(x); \
-	fw_engine = __##func##_reg_read_fw_domains(offset); \
+	fw_engine = __##func##_reg_read_fw_domains(uncore, offset); \
 	if (fw_engine) \
-		__force_wake_auto(dev_priv, fw_engine); \
-	val = __raw_i915_read##x(dev_priv, reg); \
+		__force_wake_auto(uncore, fw_engine); \
+	val = __raw_uncore_read##x(uncore, reg); \
 	GEN6_READ_FOOTER; \
 }
 #define __gen6_read(x) __gen_read(gen6, x)
@@ -1225,24 +1166,24 @@ __gen6_read(64)
 
 #define GEN2_WRITE_HEADER \
 	trace_i915_reg_rw(true, reg, val, sizeof(val), trace); \
-	assert_rpm_wakelock_held(dev_priv); \
+	__assert_rpm_wakelock_held(uncore->rpm); \
 
 #define GEN2_WRITE_FOOTER
 
 #define __gen2_write(x) \
 static void \
-gen2_write##x(struct drm_i915_private *dev_priv, i915_reg_t reg, u##x val, bool trace) { \
+gen2_write##x(struct intel_uncore *uncore, i915_reg_t reg, u##x val, bool trace) { \
 	GEN2_WRITE_HEADER; \
-	__raw_i915_write##x(dev_priv, reg, val); \
+	__raw_uncore_write##x(uncore, reg, val); \
 	GEN2_WRITE_FOOTER; \
 }
 
 #define __gen5_write(x) \
 static void \
-gen5_write##x(struct drm_i915_private *dev_priv, i915_reg_t reg, u##x val, bool trace) { \
+gen5_write##x(struct intel_uncore *uncore, i915_reg_t reg, u##x val, bool trace) { \
 	GEN2_WRITE_HEADER; \
-	ilk_dummy_write(dev_priv); \
-	__raw_i915_write##x(dev_priv, reg, val); \
+	ilk_dummy_write(uncore); \
+	__raw_uncore_write##x(uncore, reg, val); \
 	GEN2_WRITE_FOOTER; \
 }
 
@@ -1263,33 +1204,33 @@ __gen2_write(32)
 	u32 offset = i915_mmio_reg_offset(reg); \
 	unsigned long irqflags; \
 	trace_i915_reg_rw(true, reg, val, sizeof(val), trace); \
-	assert_rpm_wakelock_held(dev_priv); \
-	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags); \
-	unclaimed_reg_debug(dev_priv, reg, false, true)
+	__assert_rpm_wakelock_held(uncore->rpm); \
+	spin_lock_irqsave(&uncore->lock, irqflags); \
+	unclaimed_reg_debug(uncore, reg, false, true)
 
 #define GEN6_WRITE_FOOTER \
-	unclaimed_reg_debug(dev_priv, reg, false, false); \
-	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags)
+	unclaimed_reg_debug(uncore, reg, false, false); \
+	spin_unlock_irqrestore(&uncore->lock, irqflags)
 
 #define __gen6_write(x) \
 static void \
-gen6_write##x(struct drm_i915_private *dev_priv, i915_reg_t reg, u##x val, bool trace) { \
+gen6_write##x(struct intel_uncore *uncore, i915_reg_t reg, u##x val, bool trace) { \
 	GEN6_WRITE_HEADER; \
 	if (NEEDS_FORCE_WAKE(offset)) \
-		__gen6_gt_wait_for_fifo(dev_priv); \
-	__raw_i915_write##x(dev_priv, reg, val); \
+		__gen6_gt_wait_for_fifo(uncore); \
+	__raw_uncore_write##x(uncore, reg, val); \
 	GEN6_WRITE_FOOTER; \
 }
 
 #define __gen_write(func, x) \
 static void \
-func##_write##x(struct drm_i915_private *dev_priv, i915_reg_t reg, u##x val, bool trace) { \
+func##_write##x(struct intel_uncore *uncore, i915_reg_t reg, u##x val, bool trace) { \
 	enum forcewake_domains fw_engine; \
 	GEN6_WRITE_HEADER; \
-	fw_engine = __##func##_reg_write_fw_domains(offset); \
+	fw_engine = __##func##_reg_write_fw_domains(uncore, offset); \
 	if (fw_engine) \
-		__force_wake_auto(dev_priv, fw_engine); \
-	__raw_i915_write##x(dev_priv, reg, val); \
+		__force_wake_auto(uncore, fw_engine); \
+	__raw_uncore_write##x(uncore, reg, val); \
 	GEN6_WRITE_FOOTER; \
 }
 #define __gen8_write(x) __gen_write(gen8, x)
@@ -1316,23 +1257,23 @@ __gen6_write(32)
 #undef GEN6_WRITE_FOOTER
 #undef GEN6_WRITE_HEADER
 
-#define ASSIGN_WRITE_MMIO_VFUNCS(i915, x) \
+#define ASSIGN_WRITE_MMIO_VFUNCS(uncore, x) \
 do { \
-	(i915)->uncore.funcs.mmio_writeb = x##_write8; \
-	(i915)->uncore.funcs.mmio_writew = x##_write16; \
-	(i915)->uncore.funcs.mmio_writel = x##_write32; \
+	(uncore)->funcs.mmio_writeb = x##_write8; \
+	(uncore)->funcs.mmio_writew = x##_write16; \
+	(uncore)->funcs.mmio_writel = x##_write32; \
 } while (0)
 
-#define ASSIGN_READ_MMIO_VFUNCS(i915, x) \
+#define ASSIGN_READ_MMIO_VFUNCS(uncore, x) \
 do { \
-	(i915)->uncore.funcs.mmio_readb = x##_read8; \
-	(i915)->uncore.funcs.mmio_readw = x##_read16; \
-	(i915)->uncore.funcs.mmio_readl = x##_read32; \
-	(i915)->uncore.funcs.mmio_readq = x##_read64; \
+	(uncore)->funcs.mmio_readb = x##_read8; \
+	(uncore)->funcs.mmio_readw = x##_read16; \
+	(uncore)->funcs.mmio_readl = x##_read32; \
+	(uncore)->funcs.mmio_readq = x##_read64; \
 } while (0)
 
 
-static void fw_domain_init(struct drm_i915_private *dev_priv,
+static void fw_domain_init(struct intel_uncore *uncore,
 			   enum forcewake_domain_id domain_id,
 			   i915_reg_t reg_set,
 			   i915_reg_t reg_ack)
@@ -1342,7 +1283,7 @@ static void fw_domain_init(struct drm_i915_private *dev_priv,
 	if (WARN_ON(domain_id >= FW_DOMAIN_ID_COUNT))
 		return;
 
-	d = &dev_priv->uncore.fw_domain[domain_id];
+	d = &uncore->fw_domain[domain_id];
 
 	WARN_ON(d->wake_count);
 
@@ -1350,8 +1291,8 @@ static void fw_domain_init(struct drm_i915_private *dev_priv,
 	WARN_ON(!i915_mmio_reg_valid(reg_ack));
 
 	d->wake_count = 0;
-	d->reg_set = reg_set;
-	d->reg_ack = reg_ack;
+	d->reg_set = uncore->regs + i915_mmio_reg_offset(reg_set);
+	d->reg_ack = uncore->regs + i915_mmio_reg_offset(reg_ack);
 
 	d->id = domain_id;
 
@@ -1371,12 +1312,12 @@ static void fw_domain_init(struct drm_i915_private *dev_priv,
 	hrtimer_init(&d->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
 	d->timer.function = intel_uncore_fw_release_timer;
 
-	dev_priv->uncore.fw_domains |= BIT(domain_id);
+	uncore->fw_domains |= BIT(domain_id);
 
-	fw_domain_reset(dev_priv, d);
+	fw_domain_reset(d);
 }
 
-static void fw_domain_fini(struct drm_i915_private *dev_priv,
+static void fw_domain_fini(struct intel_uncore *uncore,
 			   enum forcewake_domain_id domain_id)
 {
 	struct intel_uncore_forcewake_domain *d;
@@ -1384,85 +1325,76 @@ static void fw_domain_fini(struct drm_i915_private *dev_priv,
 	if (WARN_ON(domain_id >= FW_DOMAIN_ID_COUNT))
 		return;
 
-	d = &dev_priv->uncore.fw_domain[domain_id];
+	d = &uncore->fw_domain[domain_id];
 
 	WARN_ON(d->wake_count);
 	WARN_ON(hrtimer_cancel(&d->timer));
 	memset(d, 0, sizeof(*d));
 
-	dev_priv->uncore.fw_domains &= ~BIT(domain_id);
+	uncore->fw_domains &= ~BIT(domain_id);
 }
 
-static void intel_uncore_fw_domains_init(struct drm_i915_private *dev_priv)
+static void intel_uncore_fw_domains_init(struct intel_uncore *uncore)
 {
-	if (INTEL_GEN(dev_priv) <= 5 || intel_vgpu_active(dev_priv))
-		return;
+	struct drm_i915_private *i915 = uncore_to_i915(uncore);
 
-	if (IS_GEN(dev_priv, 6)) {
-		dev_priv->uncore.fw_reset = 0;
-		dev_priv->uncore.fw_set = FORCEWAKE_KERNEL;
-		dev_priv->uncore.fw_clear = 0;
-	} else {
-		/* WaRsClearFWBitsAtReset:bdw,skl */
-		dev_priv->uncore.fw_reset = _MASKED_BIT_DISABLE(0xffff);
-		dev_priv->uncore.fw_set = _MASKED_BIT_ENABLE(FORCEWAKE_KERNEL);
-		dev_priv->uncore.fw_clear = _MASKED_BIT_DISABLE(FORCEWAKE_KERNEL);
-	}
+	if (!intel_uncore_has_forcewake(uncore))
+		return;
 
-	if (INTEL_GEN(dev_priv) >= 11) {
+	if (INTEL_GEN(i915) >= 11) {
 		int i;
 
-		dev_priv->uncore.funcs.force_wake_get =
+		uncore->funcs.force_wake_get =
 			fw_domains_get_with_fallback;
-		dev_priv->uncore.funcs.force_wake_put = fw_domains_put;
-		fw_domain_init(dev_priv, FW_DOMAIN_ID_RENDER,
+		uncore->funcs.force_wake_put = fw_domains_put;
+		fw_domain_init(uncore, FW_DOMAIN_ID_RENDER,
 			       FORCEWAKE_RENDER_GEN9,
 			       FORCEWAKE_ACK_RENDER_GEN9);
-		fw_domain_init(dev_priv, FW_DOMAIN_ID_BLITTER,
+		fw_domain_init(uncore, FW_DOMAIN_ID_BLITTER,
 			       FORCEWAKE_BLITTER_GEN9,
 			       FORCEWAKE_ACK_BLITTER_GEN9);
 		for (i = 0; i < I915_MAX_VCS; i++) {
-			if (!HAS_ENGINE(dev_priv, _VCS(i)))
+			if (!HAS_ENGINE(i915, _VCS(i)))
 				continue;
 
-			fw_domain_init(dev_priv, FW_DOMAIN_ID_MEDIA_VDBOX0 + i,
+			fw_domain_init(uncore, FW_DOMAIN_ID_MEDIA_VDBOX0 + i,
 				       FORCEWAKE_MEDIA_VDBOX_GEN11(i),
 				       FORCEWAKE_ACK_MEDIA_VDBOX_GEN11(i));
 		}
 		for (i = 0; i < I915_MAX_VECS; i++) {
-			if (!HAS_ENGINE(dev_priv, _VECS(i)))
+			if (!HAS_ENGINE(i915, _VECS(i)))
 				continue;
 
-			fw_domain_init(dev_priv, FW_DOMAIN_ID_MEDIA_VEBOX0 + i,
+			fw_domain_init(uncore, FW_DOMAIN_ID_MEDIA_VEBOX0 + i,
 				       FORCEWAKE_MEDIA_VEBOX_GEN11(i),
 				       FORCEWAKE_ACK_MEDIA_VEBOX_GEN11(i));
 		}
-	} else if (IS_GEN_RANGE(dev_priv, 9, 10)) {
-		dev_priv->uncore.funcs.force_wake_get =
+	} else if (IS_GEN_RANGE(i915, 9, 10)) {
+		uncore->funcs.force_wake_get =
 			fw_domains_get_with_fallback;
-		dev_priv->uncore.funcs.force_wake_put = fw_domains_put;
-		fw_domain_init(dev_priv, FW_DOMAIN_ID_RENDER,
+		uncore->funcs.force_wake_put = fw_domains_put;
+		fw_domain_init(uncore, FW_DOMAIN_ID_RENDER,
 			       FORCEWAKE_RENDER_GEN9,
 			       FORCEWAKE_ACK_RENDER_GEN9);
-		fw_domain_init(dev_priv, FW_DOMAIN_ID_BLITTER,
+		fw_domain_init(uncore, FW_DOMAIN_ID_BLITTER,
 			       FORCEWAKE_BLITTER_GEN9,
 			       FORCEWAKE_ACK_BLITTER_GEN9);
-		fw_domain_init(dev_priv, FW_DOMAIN_ID_MEDIA,
+		fw_domain_init(uncore, FW_DOMAIN_ID_MEDIA,
 			       FORCEWAKE_MEDIA_GEN9, FORCEWAKE_ACK_MEDIA_GEN9);
-	} else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {
-		dev_priv->uncore.funcs.force_wake_get = fw_domains_get;
-		dev_priv->uncore.funcs.force_wake_put = fw_domains_put;
-		fw_domain_init(dev_priv, FW_DOMAIN_ID_RENDER,
+	} else if (IS_VALLEYVIEW(i915) || IS_CHERRYVIEW(i915)) {
+		uncore->funcs.force_wake_get = fw_domains_get;
+		uncore->funcs.force_wake_put = fw_domains_put;
+		fw_domain_init(uncore, FW_DOMAIN_ID_RENDER,
 			       FORCEWAKE_VLV, FORCEWAKE_ACK_VLV);
-		fw_domain_init(dev_priv, FW_DOMAIN_ID_MEDIA,
+		fw_domain_init(uncore, FW_DOMAIN_ID_MEDIA,
 			       FORCEWAKE_MEDIA_VLV, FORCEWAKE_ACK_MEDIA_VLV);
-	} else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) {
-		dev_priv->uncore.funcs.force_wake_get =
+	} else if (IS_HASWELL(i915) || IS_BROADWELL(i915)) {
+		uncore->funcs.force_wake_get =
 			fw_domains_get_with_thread_status;
-		dev_priv->uncore.funcs.force_wake_put = fw_domains_put;
-		fw_domain_init(dev_priv, FW_DOMAIN_ID_RENDER,
+		uncore->funcs.force_wake_put = fw_domains_put;
+		fw_domain_init(uncore, FW_DOMAIN_ID_RENDER,
 			       FORCEWAKE_MT, FORCEWAKE_ACK_HSW);
-	} else if (IS_IVYBRIDGE(dev_priv)) {
+	} else if (IS_IVYBRIDGE(i915)) {
 		u32 ecobus;
 
 		/* IVB configs may use multi-threaded forcewake */
@@ -1474,9 +1406,9 @@ static void intel_uncore_fw_domains_init(struct drm_i915_private *dev_priv)
 		 * (correctly) interpreted by the test below as MT
 		 * forcewake being disabled.
 		 */
-		dev_priv->uncore.funcs.force_wake_get =
+		uncore->funcs.force_wake_get =
 			fw_domains_get_with_thread_status;
-		dev_priv->uncore.funcs.force_wake_put = fw_domains_put;
+		uncore->funcs.force_wake_put = fw_domains_put;
 
 		/* We need to init first for ECOBUS access and then
 		 * determine later if we want to reinit, in case of MT access is
@@ -1485,41 +1417,41 @@ static void intel_uncore_fw_domains_init(struct drm_i915_private *dev_priv)
 		 * before the ecobus check.
 		 */
 
-		__raw_i915_write32(dev_priv, FORCEWAKE, 0);
-		__raw_posting_read(dev_priv, ECOBUS);
+		__raw_uncore_write32(uncore, FORCEWAKE, 0);
+		__raw_posting_read(uncore, ECOBUS);
 
-		fw_domain_init(dev_priv, FW_DOMAIN_ID_RENDER,
+		fw_domain_init(uncore, FW_DOMAIN_ID_RENDER,
 			       FORCEWAKE_MT, FORCEWAKE_MT_ACK);
 
-		spin_lock_irq(&dev_priv->uncore.lock);
-		fw_domains_get_with_thread_status(dev_priv, FORCEWAKE_RENDER);
-		ecobus = __raw_i915_read32(dev_priv, ECOBUS);
-		fw_domains_put(dev_priv, FORCEWAKE_RENDER);
-		spin_unlock_irq(&dev_priv->uncore.lock);
+		spin_lock_irq(&uncore->lock);
+		fw_domains_get_with_thread_status(uncore, FORCEWAKE_RENDER);
+		ecobus = __raw_uncore_read32(uncore, ECOBUS);
+		fw_domains_put(uncore, FORCEWAKE_RENDER);
+		spin_unlock_irq(&uncore->lock);
 
 		if (!(ecobus & FORCEWAKE_MT_ENABLE)) {
 			DRM_INFO("No MT forcewake available on Ivybridge, this can result in issues\n");
 			DRM_INFO("when using vblank-synced partial screen updates.\n");
-			fw_domain_init(dev_priv, FW_DOMAIN_ID_RENDER,
+			fw_domain_init(uncore, FW_DOMAIN_ID_RENDER,
 				       FORCEWAKE, FORCEWAKE_ACK);
 		}
-	} else if (IS_GEN(dev_priv, 6)) {
-		dev_priv->uncore.funcs.force_wake_get =
+	} else if (IS_GEN(i915, 6)) {
+		uncore->funcs.force_wake_get =
 			fw_domains_get_with_thread_status;
-		dev_priv->uncore.funcs.force_wake_put = fw_domains_put;
-		fw_domain_init(dev_priv, FW_DOMAIN_ID_RENDER,
+		uncore->funcs.force_wake_put = fw_domains_put;
+		fw_domain_init(uncore, FW_DOMAIN_ID_RENDER,
 			       FORCEWAKE, FORCEWAKE_ACK);
 	}
 
 	/* All future platforms are expected to require complex power gating */
-	WARN_ON(dev_priv->uncore.fw_domains == 0);
+	WARN_ON(uncore->fw_domains == 0);
 }
 
-#define ASSIGN_FW_DOMAINS_TABLE(d) \
+#define ASSIGN_FW_DOMAINS_TABLE(uncore, d) \
 { \
-	dev_priv->uncore.fw_domains_table = \
+	(uncore)->fw_domains_table = \
 			(struct intel_forcewake_range *)(d); \
-	dev_priv->uncore.fw_domains_table_entries = ARRAY_SIZE((d)); \
+	(uncore)->fw_domains_table_entries = ARRAY_SIZE((d)); \
 }
 
 static int i915_pmic_bus_access_notifier(struct notifier_block *nb,
@@ -1544,66 +1476,132 @@ static int i915_pmic_bus_access_notifier(struct notifier_block *nb,
 		 * the access.
 		 */
 		disable_rpm_wakeref_asserts(dev_priv);
-		intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+		intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL);
 		enable_rpm_wakeref_asserts(dev_priv);
 		break;
 	case MBI_PMIC_BUS_ACCESS_END:
-		intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+		intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 		break;
 	}
 
 	return NOTIFY_OK;
 }
 
-void intel_uncore_init(struct drm_i915_private *dev_priv)
+static int uncore_mmio_setup(struct intel_uncore *uncore)
 {
-	i915_check_vgpu(dev_priv);
+	struct drm_i915_private *i915 = uncore_to_i915(uncore);
+	struct pci_dev *pdev = i915->drm.pdev;
+	int mmio_bar;
+	int mmio_size;
 
-	intel_uncore_edram_detect(dev_priv);
-	intel_uncore_fw_domains_init(dev_priv);
-	__intel_uncore_early_sanitize(dev_priv, 0);
+	mmio_bar = IS_GEN(i915, 2) ? 1 : 0;
+	/*
+	 * Before gen4, the registers and the GTT are behind different BARs.
+	 * However, from gen4 onwards, the registers and the GTT are shared
+	 * in the same BAR, so we want to restrict this ioremap from
+	 * clobbering the GTT which we want ioremap_wc instead. Fortunately,
+	 * the register BAR remains the same size for all the earlier
+	 * generations up to Ironlake.
+	 */
+	if (INTEL_GEN(i915) < 5)
+		mmio_size = 512 * 1024;
+	else
+		mmio_size = 2 * 1024 * 1024;
+	uncore->regs = pci_iomap(pdev, mmio_bar, mmio_size);
+	if (uncore->regs == NULL) {
+		DRM_ERROR("failed to map registers\n");
+
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static void uncore_mmio_cleanup(struct intel_uncore *uncore)
+{
+	struct drm_i915_private *i915 = uncore_to_i915(uncore);
+	struct pci_dev *pdev = i915->drm.pdev;
+
+	pci_iounmap(pdev, uncore->regs);
+}
+
+void intel_uncore_init_early(struct intel_uncore *uncore)
+{
+	spin_lock_init(&uncore->lock);
+}
+
+int intel_uncore_init_mmio(struct intel_uncore *uncore)
+{
+	struct drm_i915_private *i915 = uncore_to_i915(uncore);
+	int ret;
+
+	ret = uncore_mmio_setup(uncore);
+	if (ret)
+		return ret;
+
+	i915_check_vgpu(i915);
+
+	if (INTEL_GEN(i915) > 5 && !intel_vgpu_active(i915))
+		uncore->flags |= UNCORE_HAS_FORCEWAKE;
 
-	dev_priv->uncore.unclaimed_mmio_check = 1;
-	dev_priv->uncore.pmic_bus_access_nb.notifier_call =
+	intel_uncore_fw_domains_init(uncore);
+	__intel_uncore_early_sanitize(uncore, 0);
+
+	uncore->unclaimed_mmio_check = 1;
+	uncore->pmic_bus_access_nb.notifier_call =
 		i915_pmic_bus_access_notifier;
 
-	if (IS_GEN_RANGE(dev_priv, 2, 4) || intel_vgpu_active(dev_priv)) {
-		ASSIGN_WRITE_MMIO_VFUNCS(dev_priv, gen2);
-		ASSIGN_READ_MMIO_VFUNCS(dev_priv, gen2);
-	} else if (IS_GEN(dev_priv, 5)) {
-		ASSIGN_WRITE_MMIO_VFUNCS(dev_priv, gen5);
-		ASSIGN_READ_MMIO_VFUNCS(dev_priv, gen5);
-	} else if (IS_GEN_RANGE(dev_priv, 6, 7)) {
-		ASSIGN_WRITE_MMIO_VFUNCS(dev_priv, gen6);
-
-		if (IS_VALLEYVIEW(dev_priv)) {
-			ASSIGN_FW_DOMAINS_TABLE(__vlv_fw_ranges);
-			ASSIGN_READ_MMIO_VFUNCS(dev_priv, fwtable);
+	uncore->rpm = &i915->runtime_pm;
+
+	if (!intel_uncore_has_forcewake(uncore)) {
+		if (IS_GEN(i915, 5)) {
+			ASSIGN_WRITE_MMIO_VFUNCS(uncore, gen5);
+			ASSIGN_READ_MMIO_VFUNCS(uncore, gen5);
+		} else {
+			ASSIGN_WRITE_MMIO_VFUNCS(uncore, gen2);
+			ASSIGN_READ_MMIO_VFUNCS(uncore, gen2);
+		}
+	} else if (IS_GEN_RANGE(i915, 6, 7)) {
+		ASSIGN_WRITE_MMIO_VFUNCS(uncore, gen6);
+
+		if (IS_VALLEYVIEW(i915)) {
+			ASSIGN_FW_DOMAINS_TABLE(uncore, __vlv_fw_ranges);
+			ASSIGN_READ_MMIO_VFUNCS(uncore, fwtable);
 		} else {
-			ASSIGN_READ_MMIO_VFUNCS(dev_priv, gen6);
+			ASSIGN_READ_MMIO_VFUNCS(uncore, gen6);
 		}
-	} else if (IS_GEN(dev_priv, 8)) {
-		if (IS_CHERRYVIEW(dev_priv)) {
-			ASSIGN_FW_DOMAINS_TABLE(__chv_fw_ranges);
-			ASSIGN_WRITE_MMIO_VFUNCS(dev_priv, fwtable);
-			ASSIGN_READ_MMIO_VFUNCS(dev_priv, fwtable);
+	} else if (IS_GEN(i915, 8)) {
+		if (IS_CHERRYVIEW(i915)) {
+			ASSIGN_FW_DOMAINS_TABLE(uncore, __chv_fw_ranges);
+			ASSIGN_WRITE_MMIO_VFUNCS(uncore, fwtable);
+			ASSIGN_READ_MMIO_VFUNCS(uncore, fwtable);
 
 		} else {
-			ASSIGN_WRITE_MMIO_VFUNCS(dev_priv, gen8);
-			ASSIGN_READ_MMIO_VFUNCS(dev_priv, gen6);
+			ASSIGN_WRITE_MMIO_VFUNCS(uncore, gen8);
+			ASSIGN_READ_MMIO_VFUNCS(uncore, gen6);
 		}
-	} else if (IS_GEN_RANGE(dev_priv, 9, 10)) {
-		ASSIGN_FW_DOMAINS_TABLE(__gen9_fw_ranges);
-		ASSIGN_WRITE_MMIO_VFUNCS(dev_priv, fwtable);
-		ASSIGN_READ_MMIO_VFUNCS(dev_priv, fwtable);
+	} else if (IS_GEN_RANGE(i915, 9, 10)) {
+		ASSIGN_FW_DOMAINS_TABLE(uncore, __gen9_fw_ranges);
+		ASSIGN_WRITE_MMIO_VFUNCS(uncore, fwtable);
+		ASSIGN_READ_MMIO_VFUNCS(uncore, fwtable);
 	} else {
-		ASSIGN_FW_DOMAINS_TABLE(__gen11_fw_ranges);
-		ASSIGN_WRITE_MMIO_VFUNCS(dev_priv, gen11_fwtable);
-		ASSIGN_READ_MMIO_VFUNCS(dev_priv, gen11_fwtable);
+		ASSIGN_FW_DOMAINS_TABLE(uncore, __gen11_fw_ranges);
+		ASSIGN_WRITE_MMIO_VFUNCS(uncore, gen11_fwtable);
+		ASSIGN_READ_MMIO_VFUNCS(uncore, gen11_fwtable);
 	}
 
-	iosf_mbi_register_pmic_bus_access_notifier(
-		&dev_priv->uncore.pmic_bus_access_nb);
+	if (HAS_FPGA_DBG_UNCLAIMED(i915))
+		uncore->flags |= UNCORE_HAS_FPGA_DBG_UNCLAIMED;
+
+	if (IS_VALLEYVIEW(i915) || IS_CHERRYVIEW(i915))
+		uncore->flags |= UNCORE_HAS_DBG_UNCLAIMED;
+
+	if (IS_GEN_RANGE(i915, 6, 7))
+		uncore->flags |= UNCORE_HAS_FIFO;
+
+	iosf_mbi_register_pmic_bus_access_notifier(&uncore->pmic_bus_access_nb);
+
+	return 0;
 }
 
 /*
@@ -1611,45 +1609,48 @@ void intel_uncore_init(struct drm_i915_private *dev_priv)
  * the forcewake domains. Prune them, to make sure they only reference existing
  * engines.
  */
-void intel_uncore_prune(struct drm_i915_private *dev_priv)
+void intel_uncore_prune_mmio_domains(struct intel_uncore *uncore)
 {
-	if (INTEL_GEN(dev_priv) >= 11) {
-		enum forcewake_domains fw_domains = dev_priv->uncore.fw_domains;
+	struct drm_i915_private *i915 = uncore_to_i915(uncore);
+
+	if (INTEL_GEN(i915) >= 11) {
+		enum forcewake_domains fw_domains = uncore->fw_domains;
 		enum forcewake_domain_id domain_id;
 		int i;
 
 		for (i = 0; i < I915_MAX_VCS; i++) {
 			domain_id = FW_DOMAIN_ID_MEDIA_VDBOX0 + i;
 
-			if (HAS_ENGINE(dev_priv, _VCS(i)))
+			if (HAS_ENGINE(i915, _VCS(i)))
 				continue;
 
 			if (fw_domains & BIT(domain_id))
-				fw_domain_fini(dev_priv, domain_id);
+				fw_domain_fini(uncore, domain_id);
 		}
 
 		for (i = 0; i < I915_MAX_VECS; i++) {
 			domain_id = FW_DOMAIN_ID_MEDIA_VEBOX0 + i;
 
-			if (HAS_ENGINE(dev_priv, _VECS(i)))
+			if (HAS_ENGINE(i915, _VECS(i)))
 				continue;
 
 			if (fw_domains & BIT(domain_id))
-				fw_domain_fini(dev_priv, domain_id);
+				fw_domain_fini(uncore, domain_id);
 		}
 	}
 }
 
-void intel_uncore_fini(struct drm_i915_private *dev_priv)
+void intel_uncore_fini_mmio(struct intel_uncore *uncore)
 {
 	/* Paranoia: make sure we have disabled everything before we exit. */
-	intel_uncore_sanitize(dev_priv);
+	intel_uncore_sanitize(uncore_to_i915(uncore));
 
 	iosf_mbi_punit_acquire();
 	iosf_mbi_unregister_pmic_bus_access_notifier_unlocked(
-		&dev_priv->uncore.pmic_bus_access_nb);
-	intel_uncore_forcewake_reset(dev_priv);
+		&uncore->pmic_bus_access_nb);
+	intel_uncore_forcewake_reset(uncore);
 	iosf_mbi_punit_release();
+	uncore_mmio_cleanup(uncore);
 }
 
 static const struct reg_whitelist {
@@ -1717,7 +1718,7 @@ int i915_reg_read_ioctl(struct drm_device *dev,
 
 /**
  * __intel_wait_for_register_fw - wait until register matches expected state
- * @dev_priv: the i915 device
+ * @uncore: the struct intel_uncore
  * @reg: the register to read
  * @mask: mask to apply to register value
  * @value: expected value
@@ -1741,7 +1742,7 @@ int i915_reg_read_ioctl(struct drm_device *dev,
  *
  * Returns 0 if the register matches the desired condition, or -ETIMEOUT.
  */
-int __intel_wait_for_register_fw(struct drm_i915_private *dev_priv,
+int __intel_wait_for_register_fw(struct intel_uncore *uncore,
 				 i915_reg_t reg,
 				 u32 mask,
 				 u32 value,
@@ -1750,7 +1751,7 @@ int __intel_wait_for_register_fw(struct drm_i915_private *dev_priv,
 				 u32 *out_value)
 {
 	u32 uninitialized_var(reg_value);
-#define done (((reg_value = I915_READ_FW(reg)) & mask) == value)
+#define done (((reg_value = intel_uncore_read_fw(uncore, reg)) & mask) == value)
 	int ret;
 
 	/* Catch any overuse of this function */
@@ -1772,7 +1773,7 @@ int __intel_wait_for_register_fw(struct drm_i915_private *dev_priv,
 
 /**
  * __intel_wait_for_register - wait until register matches expected state
- * @dev_priv: the i915 device
+ * @uncore: the struct intel_uncore
  * @reg: the register to read
  * @mask: mask to apply to register value
  * @value: expected value
@@ -1789,33 +1790,34 @@ int __intel_wait_for_register_fw(struct drm_i915_private *dev_priv,
  *
  * Returns 0 if the register matches the desired condition, or -ETIMEOUT.
  */
-int __intel_wait_for_register(struct drm_i915_private *dev_priv,
-			    i915_reg_t reg,
-			    u32 mask,
-			    u32 value,
-			    unsigned int fast_timeout_us,
-			    unsigned int slow_timeout_ms,
-			    u32 *out_value)
+int __intel_wait_for_register(struct intel_uncore *uncore,
+			      i915_reg_t reg,
+			      u32 mask,
+			      u32 value,
+			      unsigned int fast_timeout_us,
+			      unsigned int slow_timeout_ms,
+			      u32 *out_value)
 {
 	unsigned fw =
-		intel_uncore_forcewake_for_reg(dev_priv, reg, FW_REG_READ);
+		intel_uncore_forcewake_for_reg(uncore, reg, FW_REG_READ);
 	u32 reg_value;
 	int ret;
 
 	might_sleep_if(slow_timeout_ms);
 
-	spin_lock_irq(&dev_priv->uncore.lock);
-	intel_uncore_forcewake_get__locked(dev_priv, fw);
+	spin_lock_irq(&uncore->lock);
+	intel_uncore_forcewake_get__locked(uncore, fw);
 
-	ret = __intel_wait_for_register_fw(dev_priv,
+	ret = __intel_wait_for_register_fw(uncore,
 					   reg, mask, value,
 					   fast_timeout_us, 0, &reg_value);
 
-	intel_uncore_forcewake_put__locked(dev_priv, fw);
-	spin_unlock_irq(&dev_priv->uncore.lock);
+	intel_uncore_forcewake_put__locked(uncore, fw);
+	spin_unlock_irq(&uncore->lock);
 
 	if (ret && slow_timeout_ms)
-		ret = __wait_for(reg_value = I915_READ_NOTRACE(reg),
+		ret = __wait_for(reg_value = intel_uncore_read_notrace(uncore,
+								       reg),
 				 (reg_value & mask) == value,
 				 slow_timeout_ms * 1000, 10, 1000);
 
@@ -1828,82 +1830,90 @@ int __intel_wait_for_register(struct drm_i915_private *dev_priv,
 	return ret;
 }
 
-bool intel_uncore_unclaimed_mmio(struct drm_i915_private *dev_priv)
+bool intel_uncore_unclaimed_mmio(struct intel_uncore *uncore)
 {
-	return check_for_unclaimed_mmio(dev_priv);
+	return check_for_unclaimed_mmio(uncore);
 }
 
 bool
-intel_uncore_arm_unclaimed_mmio_detection(struct drm_i915_private *dev_priv)
+intel_uncore_arm_unclaimed_mmio_detection(struct intel_uncore *uncore)
 {
 	bool ret = false;
 
-	spin_lock_irq(&dev_priv->uncore.lock);
+	spin_lock_irq(&uncore->lock);
 
-	if (unlikely(dev_priv->uncore.unclaimed_mmio_check <= 0))
+	if (unlikely(uncore->unclaimed_mmio_check <= 0))
 		goto out;
 
-	if (unlikely(intel_uncore_unclaimed_mmio(dev_priv))) {
+	if (unlikely(intel_uncore_unclaimed_mmio(uncore))) {
 		if (!i915_modparams.mmio_debug) {
 			DRM_DEBUG("Unclaimed register detected, "
 				  "enabling oneshot unclaimed register reporting. "
 				  "Please use i915.mmio_debug=N for more information.\n");
 			i915_modparams.mmio_debug++;
 		}
-		dev_priv->uncore.unclaimed_mmio_check--;
+		uncore->unclaimed_mmio_check--;
 		ret = true;
 	}
 
 out:
-	spin_unlock_irq(&dev_priv->uncore.lock);
+	spin_unlock_irq(&uncore->lock);
 
 	return ret;
 }
 
 static enum forcewake_domains
-intel_uncore_forcewake_for_read(struct drm_i915_private *dev_priv,
+intel_uncore_forcewake_for_read(struct intel_uncore *uncore,
 				i915_reg_t reg)
 {
+	struct drm_i915_private *i915 = uncore_to_i915(uncore);
 	u32 offset = i915_mmio_reg_offset(reg);
 	enum forcewake_domains fw_domains;
 
-	if (INTEL_GEN(dev_priv) >= 11) {
-		fw_domains = __gen11_fwtable_reg_read_fw_domains(offset);
-	} else if (HAS_FWTABLE(dev_priv)) {
-		fw_domains = __fwtable_reg_read_fw_domains(offset);
-	} else if (INTEL_GEN(dev_priv) >= 6) {
-		fw_domains = __gen6_reg_read_fw_domains(offset);
+	if (INTEL_GEN(i915) >= 11) {
+		fw_domains = __gen11_fwtable_reg_read_fw_domains(uncore, offset);
+	} else if (HAS_FWTABLE(i915)) {
+		fw_domains = __fwtable_reg_read_fw_domains(uncore, offset);
+	} else if (INTEL_GEN(i915) >= 6) {
+		fw_domains = __gen6_reg_read_fw_domains(uncore, offset);
 	} else {
-		WARN_ON(!IS_GEN_RANGE(dev_priv, 2, 5));
+		/* on devices with FW we expect to hit one of the above cases */
+		if (intel_uncore_has_forcewake(uncore))
+			MISSING_CASE(INTEL_GEN(i915));
+
 		fw_domains = 0;
 	}
 
-	WARN_ON(fw_domains & ~dev_priv->uncore.fw_domains);
+	WARN_ON(fw_domains & ~uncore->fw_domains);
 
 	return fw_domains;
 }
 
 static enum forcewake_domains
-intel_uncore_forcewake_for_write(struct drm_i915_private *dev_priv,
+intel_uncore_forcewake_for_write(struct intel_uncore *uncore,
 				 i915_reg_t reg)
 {
+	struct drm_i915_private *i915 = uncore_to_i915(uncore);
 	u32 offset = i915_mmio_reg_offset(reg);
 	enum forcewake_domains fw_domains;
 
-	if (INTEL_GEN(dev_priv) >= 11) {
-		fw_domains = __gen11_fwtable_reg_write_fw_domains(offset);
-	} else if (HAS_FWTABLE(dev_priv) && !IS_VALLEYVIEW(dev_priv)) {
-		fw_domains = __fwtable_reg_write_fw_domains(offset);
-	} else if (IS_GEN(dev_priv, 8)) {
-		fw_domains = __gen8_reg_write_fw_domains(offset);
-	} else if (IS_GEN_RANGE(dev_priv, 6, 7)) {
+	if (INTEL_GEN(i915) >= 11) {
+		fw_domains = __gen11_fwtable_reg_write_fw_domains(uncore, offset);
+	} else if (HAS_FWTABLE(i915) && !IS_VALLEYVIEW(i915)) {
+		fw_domains = __fwtable_reg_write_fw_domains(uncore, offset);
+	} else if (IS_GEN(i915, 8)) {
+		fw_domains = __gen8_reg_write_fw_domains(uncore, offset);
+	} else if (IS_GEN_RANGE(i915, 6, 7)) {
 		fw_domains = FORCEWAKE_RENDER;
 	} else {
-		WARN_ON(!IS_GEN_RANGE(dev_priv, 2, 5));
+		/* on devices with FW we expect to hit one of the above cases */
+		if (intel_uncore_has_forcewake(uncore))
+			MISSING_CASE(INTEL_GEN(i915));
+
 		fw_domains = 0;
 	}
 
-	WARN_ON(fw_domains & ~dev_priv->uncore.fw_domains);
+	WARN_ON(fw_domains & ~uncore->fw_domains);
 
 	return fw_domains;
 }
@@ -1911,7 +1921,7 @@ intel_uncore_forcewake_for_write(struct drm_i915_private *dev_priv,
 /**
  * intel_uncore_forcewake_for_reg - which forcewake domains are needed to access
  * 				    a register
- * @dev_priv: pointer to struct drm_i915_private
+ * @uncore: pointer to struct intel_uncore
  * @reg: register in question
  * @op: operation bitmask of FW_REG_READ and/or FW_REG_WRITE
  *
@@ -1923,21 +1933,21 @@ intel_uncore_forcewake_for_write(struct drm_i915_private *dev_priv,
  * callers to do FIFO management on their own or risk losing writes.
  */
 enum forcewake_domains
-intel_uncore_forcewake_for_reg(struct drm_i915_private *dev_priv,
+intel_uncore_forcewake_for_reg(struct intel_uncore *uncore,
 			       i915_reg_t reg, unsigned int op)
 {
 	enum forcewake_domains fw_domains = 0;
 
 	WARN_ON(!op);
 
-	if (intel_vgpu_active(dev_priv))
+	if (!intel_uncore_has_forcewake(uncore))
 		return 0;
 
 	if (op & FW_REG_READ)
-		fw_domains = intel_uncore_forcewake_for_read(dev_priv, reg);
+		fw_domains = intel_uncore_forcewake_for_read(uncore, reg);
 
 	if (op & FW_REG_WRITE)
-		fw_domains |= intel_uncore_forcewake_for_write(dev_priv, reg);
+		fw_domains |= intel_uncore_forcewake_for_write(uncore, reg);
 
 	return fw_domains;
 }
diff --git a/drivers/gpu/drm/i915/intel_uncore.h b/drivers/gpu/drm/i915/intel_uncore.h
index e5e157d288de..d6af3de70121 100644
--- a/drivers/gpu/drm/i915/intel_uncore.h
+++ b/drivers/gpu/drm/i915/intel_uncore.h
@@ -28,10 +28,13 @@
 #include <linux/spinlock.h>
 #include <linux/notifier.h>
 #include <linux/hrtimer.h>
+#include <linux/io-64-nonatomic-lo-hi.h>
 
 #include "i915_reg.h"
 
 struct drm_i915_private;
+struct i915_runtime_pm;
+struct intel_uncore;
 
 enum forcewake_domain_id {
 	FW_DOMAIN_ID_RENDER = 0,
@@ -62,25 +65,25 @@ enum forcewake_domains {
 };
 
 struct intel_uncore_funcs {
-	void (*force_wake_get)(struct drm_i915_private *dev_priv,
+	void (*force_wake_get)(struct intel_uncore *uncore,
 			       enum forcewake_domains domains);
-	void (*force_wake_put)(struct drm_i915_private *dev_priv,
+	void (*force_wake_put)(struct intel_uncore *uncore,
 			       enum forcewake_domains domains);
 
-	u8 (*mmio_readb)(struct drm_i915_private *dev_priv,
+	u8 (*mmio_readb)(struct intel_uncore *uncore,
 			 i915_reg_t r, bool trace);
-	u16 (*mmio_readw)(struct drm_i915_private *dev_priv,
+	u16 (*mmio_readw)(struct intel_uncore *uncore,
 			  i915_reg_t r, bool trace);
-	u32 (*mmio_readl)(struct drm_i915_private *dev_priv,
+	u32 (*mmio_readl)(struct intel_uncore *uncore,
 			  i915_reg_t r, bool trace);
-	u64 (*mmio_readq)(struct drm_i915_private *dev_priv,
+	u64 (*mmio_readq)(struct intel_uncore *uncore,
 			  i915_reg_t r, bool trace);
 
-	void (*mmio_writeb)(struct drm_i915_private *dev_priv,
+	void (*mmio_writeb)(struct intel_uncore *uncore,
 			    i915_reg_t r, u8 val, bool trace);
-	void (*mmio_writew)(struct drm_i915_private *dev_priv,
+	void (*mmio_writew)(struct intel_uncore *uncore,
 			    i915_reg_t r, u16 val, bool trace);
-	void (*mmio_writel)(struct drm_i915_private *dev_priv,
+	void (*mmio_writel)(struct intel_uncore *uncore,
 			    i915_reg_t r, u32 val, bool trace);
 };
 
@@ -92,8 +95,18 @@ struct intel_forcewake_range {
 };
 
 struct intel_uncore {
+	void __iomem *regs;
+
+	struct i915_runtime_pm *rpm;
+
 	spinlock_t lock; /** lock is also taken in irq contexts. */
 
+	unsigned int flags;
+#define UNCORE_HAS_FORCEWAKE		BIT(0)
+#define UNCORE_HAS_FPGA_DBG_UNCLAIMED	BIT(1)
+#define UNCORE_HAS_DBG_UNCLAIMED	BIT(2)
+#define UNCORE_HAS_FIFO			BIT(3)
+
 	const struct intel_forcewake_range *fw_domains_table;
 	unsigned int fw_domains_table_entries;
 
@@ -106,18 +119,14 @@ struct intel_uncore {
 	enum forcewake_domains fw_domains_active;
 	enum forcewake_domains fw_domains_saved; /* user domains saved for S3 */
 
-	u32 fw_set;
-	u32 fw_clear;
-	u32 fw_reset;
-
 	struct intel_uncore_forcewake_domain {
 		enum forcewake_domain_id id;
 		enum forcewake_domains mask;
 		unsigned int wake_count;
 		bool active;
 		struct hrtimer timer;
-		i915_reg_t reg_set;
-		i915_reg_t reg_ack;
+		u32 __iomem *reg_set;
+		u32 __iomem *reg_ack;
 	} fw_domain[FW_DOMAIN_ID_COUNT];
 
 	struct {
@@ -131,86 +140,257 @@ struct intel_uncore {
 };
 
 /* Iterate over initialised fw domains */
-#define for_each_fw_domain_masked(domain__, mask__, dev_priv__, tmp__) \
+#define for_each_fw_domain_masked(domain__, mask__, uncore__, tmp__) \
 	for (tmp__ = (mask__); \
-	     tmp__ ? (domain__ = &(dev_priv__)->uncore.fw_domain[__mask_next_bit(tmp__)]), 1 : 0;)
+	     tmp__ ? (domain__ = &(uncore__)->fw_domain[__mask_next_bit(tmp__)]), 1 : 0;)
+
+#define for_each_fw_domain(domain__, uncore__, tmp__) \
+	for_each_fw_domain_masked(domain__, (uncore__)->fw_domains, uncore__, tmp__)
+
+static inline struct intel_uncore *
+forcewake_domain_to_uncore(const struct intel_uncore_forcewake_domain *d)
+{
+	return container_of(d, struct intel_uncore, fw_domain[d->id]);
+}
 
-#define for_each_fw_domain(domain__, dev_priv__, tmp__) \
-	for_each_fw_domain_masked(domain__, (dev_priv__)->uncore.fw_domains, dev_priv__, tmp__)
+static inline bool
+intel_uncore_has_forcewake(const struct intel_uncore *uncore)
+{
+	return uncore->flags & UNCORE_HAS_FORCEWAKE;
+}
 
+static inline bool
+intel_uncore_has_fpga_dbg_unclaimed(const struct intel_uncore *uncore)
+{
+	return uncore->flags & UNCORE_HAS_FPGA_DBG_UNCLAIMED;
+}
+
+static inline bool
+intel_uncore_has_dbg_unclaimed(const struct intel_uncore *uncore)
+{
+	return uncore->flags & UNCORE_HAS_DBG_UNCLAIMED;
+}
+
+static inline bool
+intel_uncore_has_fifo(const struct intel_uncore *uncore)
+{
+	return uncore->flags & UNCORE_HAS_FIFO;
+}
 
 void intel_uncore_sanitize(struct drm_i915_private *dev_priv);
-void intel_uncore_init(struct drm_i915_private *dev_priv);
-void intel_uncore_prune(struct drm_i915_private *dev_priv);
-bool intel_uncore_unclaimed_mmio(struct drm_i915_private *dev_priv);
-bool intel_uncore_arm_unclaimed_mmio_detection(struct drm_i915_private *dev_priv);
-void intel_uncore_fini(struct drm_i915_private *dev_priv);
-void intel_uncore_suspend(struct drm_i915_private *dev_priv);
-void intel_uncore_resume_early(struct drm_i915_private *dev_priv);
-void intel_uncore_runtime_resume(struct drm_i915_private *dev_priv);
-
-u64 intel_uncore_edram_size(struct drm_i915_private *dev_priv);
-void assert_forcewakes_inactive(struct drm_i915_private *dev_priv);
-void assert_forcewakes_active(struct drm_i915_private *dev_priv,
+void intel_uncore_init_early(struct intel_uncore *uncore);
+int intel_uncore_init_mmio(struct intel_uncore *uncore);
+void intel_uncore_prune_mmio_domains(struct intel_uncore *uncore);
+bool intel_uncore_unclaimed_mmio(struct intel_uncore *uncore);
+bool intel_uncore_arm_unclaimed_mmio_detection(struct intel_uncore *uncore);
+void intel_uncore_fini_mmio(struct intel_uncore *uncore);
+void intel_uncore_suspend(struct intel_uncore *uncore);
+void intel_uncore_resume_early(struct intel_uncore *uncore);
+void intel_uncore_runtime_resume(struct intel_uncore *uncore);
+
+void assert_forcewakes_inactive(struct intel_uncore *uncore);
+void assert_forcewakes_active(struct intel_uncore *uncore,
 			      enum forcewake_domains fw_domains);
 const char *intel_uncore_forcewake_domain_to_str(const enum forcewake_domain_id id);
 
 enum forcewake_domains
-intel_uncore_forcewake_for_reg(struct drm_i915_private *dev_priv,
+intel_uncore_forcewake_for_reg(struct intel_uncore *uncore,
 			       i915_reg_t reg, unsigned int op);
 #define FW_REG_READ  (1)
 #define FW_REG_WRITE (2)
 
-void intel_uncore_forcewake_get(struct drm_i915_private *dev_priv,
+void intel_uncore_forcewake_get(struct intel_uncore *uncore,
 				enum forcewake_domains domains);
-void intel_uncore_forcewake_put(struct drm_i915_private *dev_priv,
+void intel_uncore_forcewake_put(struct intel_uncore *uncore,
 				enum forcewake_domains domains);
 /* Like above but the caller must manage the uncore.lock itself.
  * Must be used with I915_READ_FW and friends.
  */
-void intel_uncore_forcewake_get__locked(struct drm_i915_private *dev_priv,
+void intel_uncore_forcewake_get__locked(struct intel_uncore *uncore,
 					enum forcewake_domains domains);
-void intel_uncore_forcewake_put__locked(struct drm_i915_private *dev_priv,
+void intel_uncore_forcewake_put__locked(struct intel_uncore *uncore,
 					enum forcewake_domains domains);
 
-void intel_uncore_forcewake_user_get(struct drm_i915_private *dev_priv);
-void intel_uncore_forcewake_user_put(struct drm_i915_private *dev_priv);
+void intel_uncore_forcewake_user_get(struct intel_uncore *uncore);
+void intel_uncore_forcewake_user_put(struct intel_uncore *uncore);
 
-int __intel_wait_for_register(struct drm_i915_private *dev_priv,
+int __intel_wait_for_register(struct intel_uncore *uncore,
 			      i915_reg_t reg,
 			      u32 mask,
 			      u32 value,
 			      unsigned int fast_timeout_us,
 			      unsigned int slow_timeout_ms,
 			      u32 *out_value);
-static inline
-int intel_wait_for_register(struct drm_i915_private *dev_priv,
-			    i915_reg_t reg,
-			    u32 mask,
-			    u32 value,
-			    unsigned int timeout_ms)
+static inline int
+intel_wait_for_register(struct intel_uncore *uncore,
+			i915_reg_t reg,
+			u32 mask,
+			u32 value,
+			unsigned int timeout_ms)
 {
-	return __intel_wait_for_register(dev_priv, reg, mask, value, 2,
+	return __intel_wait_for_register(uncore, reg, mask, value, 2,
 					 timeout_ms, NULL);
 }
-int __intel_wait_for_register_fw(struct drm_i915_private *dev_priv,
+
+int __intel_wait_for_register_fw(struct intel_uncore *uncore,
 				 i915_reg_t reg,
 				 u32 mask,
 				 u32 value,
 				 unsigned int fast_timeout_us,
 				 unsigned int slow_timeout_ms,
 				 u32 *out_value);
-static inline
-int intel_wait_for_register_fw(struct drm_i915_private *dev_priv,
-			       i915_reg_t reg,
-			       u32 mask,
-			       u32 value,
+static inline int
+intel_wait_for_register_fw(struct intel_uncore *uncore,
+			   i915_reg_t reg,
+			   u32 mask,
+			   u32 value,
 			       unsigned int timeout_ms)
 {
-	return __intel_wait_for_register_fw(dev_priv, reg, mask, value,
+	return __intel_wait_for_register_fw(uncore, reg, mask, value,
 					    2, timeout_ms, NULL);
 }
 
+/* register access functions */
+#define __raw_read(x__, s__) \
+static inline u##x__ __raw_uncore_read##x__(const struct intel_uncore *uncore, \
+					    i915_reg_t reg) \
+{ \
+	return read##s__(uncore->regs + i915_mmio_reg_offset(reg)); \
+}
+
+#define __raw_write(x__, s__) \
+static inline void __raw_uncore_write##x__(const struct intel_uncore *uncore, \
+					   i915_reg_t reg, u##x__ val) \
+{ \
+	write##s__(val, uncore->regs + i915_mmio_reg_offset(reg)); \
+}
+__raw_read(8, b)
+__raw_read(16, w)
+__raw_read(32, l)
+__raw_read(64, q)
+
+__raw_write(8, b)
+__raw_write(16, w)
+__raw_write(32, l)
+__raw_write(64, q)
+
+#undef __raw_read
+#undef __raw_write
+
+#define __uncore_read(name__, x__, s__, trace__) \
+static inline u##x__ intel_uncore_##name__(struct intel_uncore *uncore, \
+					   i915_reg_t reg) \
+{ \
+	return uncore->funcs.mmio_read##s__(uncore, reg, (trace__)); \
+}
+
+#define __uncore_write(name__, x__, s__, trace__) \
+static inline void intel_uncore_##name__(struct intel_uncore *uncore, \
+					 i915_reg_t reg, u##x__ val) \
+{ \
+	uncore->funcs.mmio_write##s__(uncore, reg, val, (trace__)); \
+}
+
+__uncore_read(read8, 8, b, true)
+__uncore_read(read16, 16, w, true)
+__uncore_read(read, 32, l, true)
+__uncore_read(read16_notrace, 16, w, false)
+__uncore_read(read_notrace, 32, l, false)
+
+__uncore_write(write8, 8, b, true)
+__uncore_write(write16, 16, w, true)
+__uncore_write(write, 32, l, true)
+__uncore_write(write_notrace, 32, l, false)
+
+/* Be very careful with read/write 64-bit values. On 32-bit machines, they
+ * will be implemented using 2 32-bit writes in an arbitrary order with
+ * an arbitrary delay between them. This can cause the hardware to
+ * act upon the intermediate value, possibly leading to corruption and
+ * machine death. For this reason we do not support I915_WRITE64, or
+ * uncore->funcs.mmio_writeq.
+ *
+ * When reading a 64-bit value as two 32-bit values, the delay may cause
+ * the two reads to mismatch, e.g. a timestamp overflowing. Also note that
+ * occasionally a 64-bit register does not actually support a full readq
+ * and must be read using two 32-bit reads.
+ *
+ * You have been warned.
+ */
+__uncore_read(read64, 64, q, true)
+
+static inline u64
+intel_uncore_read64_2x32(struct intel_uncore *uncore,
+			 i915_reg_t lower_reg, i915_reg_t upper_reg)
+{
+	u32 upper, lower, old_upper, loop = 0;
+	upper = intel_uncore_read(uncore, upper_reg);
+	do {
+		old_upper = upper;
+		lower = intel_uncore_read(uncore, lower_reg);
+		upper = intel_uncore_read(uncore, upper_reg);
+	} while (upper != old_upper && loop++ < 2);
+	return (u64)upper << 32 | lower;
+}
+
+#define intel_uncore_posting_read(...) ((void)intel_uncore_read_notrace(__VA_ARGS__))
+#define intel_uncore_posting_read16(...) ((void)intel_uncore_read16_notrace(__VA_ARGS__))
+
+#undef __uncore_read
+#undef __uncore_write
+
+/* These are untraced mmio-accessors that are only valid to be used inside
+ * critical sections, such as inside IRQ handlers, where forcewake is explicitly
+ * controlled.
+ *
+ * Think twice, and think again, before using these.
+ *
+ * As an example, these accessors can possibly be used between:
+ *
+ * spin_lock_irq(&uncore->lock);
+ * intel_uncore_forcewake_get__locked();
+ *
+ * and
+ *
+ * intel_uncore_forcewake_put__locked();
+ * spin_unlock_irq(&uncore->lock);
+ *
+ *
+ * Note: some registers may not need forcewake held, so
+ * intel_uncore_forcewake_{get,put} can be omitted, see
+ * intel_uncore_forcewake_for_reg().
+ *
+ * Certain architectures will die if the same cacheline is concurrently accessed
+ * by different clients (e.g. on Ivybridge). Access to registers should
+ * therefore generally be serialised, by either the dev_priv->uncore.lock or
+ * a more localised lock guarding all access to that bank of registers.
+ */
+#define intel_uncore_read_fw(...) __raw_uncore_read32(__VA_ARGS__)
+#define intel_uncore_write_fw(...) __raw_uncore_write32(__VA_ARGS__)
+#define intel_uncore_write64_fw(...) __raw_uncore_write64(__VA_ARGS__)
+#define intel_uncore_posting_read_fw(...) ((void)intel_uncore_read_fw(__VA_ARGS__))
+
+static inline void intel_uncore_rmw(struct intel_uncore *uncore,
+				    i915_reg_t reg, u32 clear, u32 set)
+{
+	u32 val;
+
+	val = intel_uncore_read(uncore, reg);
+	val &= ~clear;
+	val |= set;
+	intel_uncore_write(uncore, reg, val);
+}
+
+static inline void intel_uncore_rmw_fw(struct intel_uncore *uncore,
+				       i915_reg_t reg, u32 clear, u32 set)
+{
+	u32 val;
+
+	val = intel_uncore_read_fw(uncore, reg);
+	val &= ~clear;
+	val |= set;
+	intel_uncore_write_fw(uncore, reg, val);
+}
+
 #define raw_reg_read(base, reg) \
 	readl(base + i915_mmio_reg_offset(reg))
 #define raw_reg_write(base, reg, value) \
diff --git a/drivers/gpu/drm/i915/intel_vbt_defs.h b/drivers/gpu/drm/i915/intel_vbt_defs.h
index bf3662ad5fed..fdbbb9a53804 100644
--- a/drivers/gpu/drm/i915/intel_vbt_defs.h
+++ b/drivers/gpu/drm/i915/intel_vbt_defs.h
@@ -772,6 +772,9 @@ struct psr_table {
 	/* TP wake up time in multiple of 100 */
 	u16 tp1_wakeup_time;
 	u16 tp2_tp3_wakeup_time;
+
+	/* PSR2 TP2/TP3 wakeup time for 16 panels */
+	u32 psr2_tp2_tp3_wakeup_time;
 } __packed;
 
 struct bdb_psr {
diff --git a/drivers/gpu/drm/i915/intel_vdsc.c b/drivers/gpu/drm/i915/intel_vdsc.c
index 23abf03736e7..3f9921ba4a76 100644
--- a/drivers/gpu/drm/i915/intel_vdsc.c
+++ b/drivers/gpu/drm/i915/intel_vdsc.c
@@ -317,129 +317,6 @@ static int get_column_index_for_rc_params(u8 bits_per_component)
 	}
 }
 
-static int intel_compute_rc_parameters(struct drm_dsc_config *vdsc_cfg)
-{
-	unsigned long groups_per_line = 0;
-	unsigned long groups_total = 0;
-	unsigned long num_extra_mux_bits = 0;
-	unsigned long slice_bits = 0;
-	unsigned long hrd_delay = 0;
-	unsigned long final_scale = 0;
-	unsigned long rbs_min = 0;
-
-	/* Number of groups used to code each line of a slice */
-	groups_per_line = DIV_ROUND_UP(vdsc_cfg->slice_width,
-				       DSC_RC_PIXELS_PER_GROUP);
-
-	/* chunksize in Bytes */
-	vdsc_cfg->slice_chunk_size = DIV_ROUND_UP(vdsc_cfg->slice_width *
-						  vdsc_cfg->bits_per_pixel,
-						  (8 * 16));
-
-	if (vdsc_cfg->convert_rgb)
-		num_extra_mux_bits = 3 * (vdsc_cfg->mux_word_size +
-					  (4 * vdsc_cfg->bits_per_component + 4)
-					  - 2);
-	else
-		num_extra_mux_bits = 3 * vdsc_cfg->mux_word_size +
-			(4 * vdsc_cfg->bits_per_component + 4) +
-			2 * (4 * vdsc_cfg->bits_per_component) - 2;
-	/* Number of bits in one Slice */
-	slice_bits = 8 * vdsc_cfg->slice_chunk_size * vdsc_cfg->slice_height;
-
-	while ((num_extra_mux_bits > 0) &&
-	       ((slice_bits - num_extra_mux_bits) % vdsc_cfg->mux_word_size))
-		num_extra_mux_bits--;
-
-	if (groups_per_line < vdsc_cfg->initial_scale_value - 8)
-		vdsc_cfg->initial_scale_value = groups_per_line + 8;
-
-	/* scale_decrement_interval calculation according to DSC spec 1.11 */
-	if (vdsc_cfg->initial_scale_value > 8)
-		vdsc_cfg->scale_decrement_interval = groups_per_line /
-			(vdsc_cfg->initial_scale_value - 8);
-	else
-		vdsc_cfg->scale_decrement_interval = DSC_SCALE_DECREMENT_INTERVAL_MAX;
-
-	vdsc_cfg->final_offset = vdsc_cfg->rc_model_size -
-		(vdsc_cfg->initial_xmit_delay *
-		 vdsc_cfg->bits_per_pixel + 8) / 16 + num_extra_mux_bits;
-
-	if (vdsc_cfg->final_offset >= vdsc_cfg->rc_model_size) {
-		DRM_DEBUG_KMS("FinalOfs < RcModelSze for this InitialXmitDelay\n");
-		return -ERANGE;
-	}
-
-	final_scale = (vdsc_cfg->rc_model_size * 8) /
-		(vdsc_cfg->rc_model_size - vdsc_cfg->final_offset);
-	if (vdsc_cfg->slice_height > 1)
-		/*
-		 * NflBpgOffset is 16 bit value with 11 fractional bits
-		 * hence we multiply by 2^11 for preserving the
-		 * fractional part
-		 */
-		vdsc_cfg->nfl_bpg_offset = DIV_ROUND_UP((vdsc_cfg->first_line_bpg_offset << 11),
-							(vdsc_cfg->slice_height - 1));
-	else
-		vdsc_cfg->nfl_bpg_offset = 0;
-
-	/* 2^16 - 1 */
-	if (vdsc_cfg->nfl_bpg_offset > 65535) {
-		DRM_DEBUG_KMS("NflBpgOffset is too large for this slice height\n");
-		return -ERANGE;
-	}
-
-	/* Number of groups used to code the entire slice */
-	groups_total = groups_per_line * vdsc_cfg->slice_height;
-
-	/* slice_bpg_offset is 16 bit value with 11 fractional bits */
-	vdsc_cfg->slice_bpg_offset = DIV_ROUND_UP(((vdsc_cfg->rc_model_size -
-						    vdsc_cfg->initial_offset +
-						    num_extra_mux_bits) << 11),
-						  groups_total);
-
-	if (final_scale > 9) {
-		/*
-		 * ScaleIncrementInterval =
-		 * finaloffset/((NflBpgOffset + SliceBpgOffset)*8(finalscale - 1.125))
-		 * as (NflBpgOffset + SliceBpgOffset) has 11 bit fractional value,
-		 * we need divide by 2^11 from pstDscCfg values
-		 */
-		vdsc_cfg->scale_increment_interval =
-				(vdsc_cfg->final_offset * (1 << 11)) /
-				((vdsc_cfg->nfl_bpg_offset +
-				vdsc_cfg->slice_bpg_offset) *
-				(final_scale - 9));
-	} else {
-		/*
-		 * If finalScaleValue is less than or equal to 9, a value of 0 should
-		 * be used to disable the scale increment at the end of the slice
-		 */
-		vdsc_cfg->scale_increment_interval = 0;
-	}
-
-	if (vdsc_cfg->scale_increment_interval > 65535) {
-		DRM_DEBUG_KMS("ScaleIncrementInterval is large for slice height\n");
-		return -ERANGE;
-	}
-
-	/*
-	 * DSC spec mentions that bits_per_pixel specifies the target
-	 * bits/pixel (bpp) rate that is used by the encoder,
-	 * in steps of 1/16 of a bit per pixel
-	 */
-	rbs_min = vdsc_cfg->rc_model_size - vdsc_cfg->initial_offset +
-		DIV_ROUND_UP(vdsc_cfg->initial_xmit_delay *
-			     vdsc_cfg->bits_per_pixel, 16) +
-		groups_per_line * vdsc_cfg->first_line_bpg_offset;
-
-	hrd_delay = DIV_ROUND_UP((rbs_min * 16), vdsc_cfg->bits_per_pixel);
-	vdsc_cfg->rc_bits = (hrd_delay * vdsc_cfg->bits_per_pixel) / 16;
-	vdsc_cfg->initial_dec_delay = hrd_delay - vdsc_cfg->initial_xmit_delay;
-
-	return 0;
-}
-
 int intel_dp_compute_dsc_params(struct intel_dp *intel_dp,
 				struct intel_crtc_state *pipe_config)
 {
@@ -491,7 +368,7 @@ int intel_dp_compute_dsc_params(struct intel_dp *intel_dp,
 			DSC_1_1_MAX_LINEBUF_DEPTH_BITS : line_buf_depth;
 
 	/* Gen 11 does not support YCbCr */
-	vdsc_cfg->enable422 = false;
+	vdsc_cfg->simple_422 = false;
 	/* Gen 11 does not support VBR */
 	vdsc_cfg->vbr_enable = false;
 	vdsc_cfg->block_pred_enable =
@@ -574,7 +451,7 @@ int intel_dp_compute_dsc_params(struct intel_dp *intel_dp,
 	vdsc_cfg->initial_scale_value = (vdsc_cfg->rc_model_size << 3) /
 		(vdsc_cfg->rc_model_size - vdsc_cfg->initial_offset);
 
-	return intel_compute_rc_parameters(vdsc_cfg);
+	return drm_dsc_compute_rc_parameters(vdsc_cfg);
 }
 
 enum intel_display_power_domain
@@ -618,7 +495,7 @@ static void intel_configure_pps_for_dsc_encoder(struct intel_encoder *encoder,
 		pps_val |= DSC_BLOCK_PREDICTION;
 	if (vdsc_cfg->convert_rgb)
 		pps_val |= DSC_COLOR_SPACE_CONVERSION;
-	if (vdsc_cfg->enable422)
+	if (vdsc_cfg->simple_422)
 		pps_val |= DSC_422_ENABLE;
 	if (vdsc_cfg->vbr_enable)
 		pps_val |= DSC_VBR_ENABLE;
@@ -1004,10 +881,10 @@ static void intel_dp_write_dsc_pps_sdp(struct intel_encoder *encoder,
 	struct drm_dsc_pps_infoframe dp_dsc_pps_sdp;
 
 	/* Prepare DP SDP PPS header as per DP 1.4 spec, Table 2-123 */
-	drm_dsc_dp_pps_header_init(&dp_dsc_pps_sdp);
+	drm_dsc_dp_pps_header_init(&dp_dsc_pps_sdp.pps_header);
 
 	/* Fill the PPS payload bytes as per DSC spec 1.2 Table 4-1 */
-	drm_dsc_pps_infoframe_pack(&dp_dsc_pps_sdp, vdsc_cfg);
+	drm_dsc_pps_payload_pack(&dp_dsc_pps_sdp.pps_payload, vdsc_cfg);
 
 	intel_dig_port->write_infoframe(encoder, crtc_state,
 					DP_SDP_PPS, &dp_dsc_pps_sdp,
diff --git a/drivers/gpu/drm/i915/intel_workarounds.c b/drivers/gpu/drm/i915/intel_workarounds.c
index 15f4a6dee5aa..9682dd575152 100644
--- a/drivers/gpu/drm/i915/intel_workarounds.c
+++ b/drivers/gpu/drm/i915/intel_workarounds.c
@@ -541,10 +541,6 @@ static void icl_ctx_workarounds_init(struct intel_engine_cs *engine)
 		WA_SET_BIT_MASKED(GEN7_ROW_CHICKEN2,
 				  GEN11_TDL_CLOCK_GATING_FIX_DISABLE);
 
-	/* WaEnableStateCacheRedirectToCS:icl */
-	WA_SET_BIT_MASKED(GEN9_SLICE_COMMON_ECO_CHICKEN1,
-			  GEN11_STATE_CACHE_REDIRECT_TO_CS);
-
 	/* Wa_2006665173:icl (pre-prod) */
 	if (IS_ICL_REVID(i915, ICL_REVID_A0, ICL_REVID_A0))
 		WA_SET_BIT_MASKED(GEN11_COMMON_SLICE_CHICKEN3,
@@ -555,6 +551,11 @@ static void icl_ctx_workarounds_init(struct intel_engine_cs *engine)
 			   GEN10_CACHE_MODE_SS,
 			   0, /* write-only, so skip validation */
 			   _MASKED_BIT_ENABLE(FLOAT_BLEND_OPTIMIZATION_ENABLE));
+
+	/* WaDisableGPGPUMidThreadPreemption:icl */
+	WA_SET_FIELD_MASKED(GEN8_CS_CHICKEN1,
+			    GEN9_PREEMPT_GPGPU_LEVEL_MASK,
+			    GEN9_PREEMPT_GPGPU_THREAD_GROUP_LEVEL);
 }
 
 void intel_engine_init_ctx_wa(struct intel_engine_cs *engine)
@@ -564,26 +565,26 @@ void intel_engine_init_ctx_wa(struct intel_engine_cs *engine)
 
 	wa_init_start(wal, "context");
 
-	if (INTEL_GEN(i915) < 8)
-		return;
-	else if (IS_BROADWELL(i915))
-		bdw_ctx_workarounds_init(engine);
-	else if (IS_CHERRYVIEW(i915))
-		chv_ctx_workarounds_init(engine);
-	else if (IS_SKYLAKE(i915))
-		skl_ctx_workarounds_init(engine);
-	else if (IS_BROXTON(i915))
-		bxt_ctx_workarounds_init(engine);
-	else if (IS_KABYLAKE(i915))
-		kbl_ctx_workarounds_init(engine);
-	else if (IS_GEMINILAKE(i915))
-		glk_ctx_workarounds_init(engine);
-	else if (IS_COFFEELAKE(i915))
-		cfl_ctx_workarounds_init(engine);
+	if (IS_GEN(i915, 11))
+		icl_ctx_workarounds_init(engine);
 	else if (IS_CANNONLAKE(i915))
 		cnl_ctx_workarounds_init(engine);
-	else if (IS_ICELAKE(i915))
-		icl_ctx_workarounds_init(engine);
+	else if (IS_COFFEELAKE(i915))
+		cfl_ctx_workarounds_init(engine);
+	else if (IS_GEMINILAKE(i915))
+		glk_ctx_workarounds_init(engine);
+	else if (IS_KABYLAKE(i915))
+		kbl_ctx_workarounds_init(engine);
+	else if (IS_BROXTON(i915))
+		bxt_ctx_workarounds_init(engine);
+	else if (IS_SKYLAKE(i915))
+		skl_ctx_workarounds_init(engine);
+	else if (IS_CHERRYVIEW(i915))
+		chv_ctx_workarounds_init(engine);
+	else if (IS_BROADWELL(i915))
+		bdw_ctx_workarounds_init(engine);
+	else if (INTEL_GEN(i915) < 8)
+		return;
 	else
 		MISSING_CASE(INTEL_GEN(i915));
 
@@ -724,9 +725,9 @@ cfl_gt_workarounds_init(struct drm_i915_private *i915, struct i915_wa_list *wal)
 }
 
 static void
-wa_init_mcr(struct drm_i915_private *dev_priv, struct i915_wa_list *wal)
+wa_init_mcr(struct drm_i915_private *i915, struct i915_wa_list *wal)
 {
-	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
+	const struct sseu_dev_info *sseu = &RUNTIME_INFO(i915)->sseu;
 	u32 mcr_slice_subslice_mask;
 
 	/*
@@ -742,14 +743,15 @@ wa_init_mcr(struct drm_i915_private *dev_priv, struct i915_wa_list *wal)
 	 * something more complex that requires checking the range of every
 	 * MMIO read).
 	 */
-	if (INTEL_GEN(dev_priv) >= 10 &&
+	if (INTEL_GEN(i915) >= 10 &&
 	    is_power_of_2(sseu->slice_mask)) {
 		/*
 		 * read FUSE3 for enabled L3 Bank IDs, if L3 Bank matches
 		 * enabled subslice, no need to redirect MCR packet
 		 */
 		u32 slice = fls(sseu->slice_mask);
-		u32 fuse3 = I915_READ(GEN10_MIRROR_FUSE3);
+		u32 fuse3 =
+			intel_uncore_read(&i915->uncore, GEN10_MIRROR_FUSE3);
 		u8 ss_mask = sseu->subslice_mask[slice];
 
 		u8 enabled_mask = (ss_mask | ss_mask >>
@@ -763,7 +765,7 @@ wa_init_mcr(struct drm_i915_private *dev_priv, struct i915_wa_list *wal)
 		WARN_ON((enabled_mask & disabled_mask) != enabled_mask);
 	}
 
-	if (INTEL_GEN(dev_priv) >= 11)
+	if (INTEL_GEN(i915) >= 11)
 		mcr_slice_subslice_mask = GEN11_MCR_SLICE_MASK |
 					  GEN11_MCR_SUBSLICE_MASK;
 	else
@@ -783,7 +785,7 @@ wa_init_mcr(struct drm_i915_private *dev_priv, struct i915_wa_list *wal)
 	wa_write_masked_or(wal,
 			   GEN8_MCR_SELECTOR,
 			   mcr_slice_subslice_mask,
-			   intel_calculate_mcr_s_ss_select(dev_priv));
+			   intel_calculate_mcr_s_ss_select(i915));
 }
 
 static void
@@ -862,26 +864,22 @@ icl_gt_workarounds_init(struct drm_i915_private *i915, struct i915_wa_list *wal)
 static void
 gt_init_workarounds(struct drm_i915_private *i915, struct i915_wa_list *wal)
 {
-	if (INTEL_GEN(i915) < 8)
-		return;
-	else if (IS_BROADWELL(i915))
-		return;
-	else if (IS_CHERRYVIEW(i915))
-		return;
-	else if (IS_SKYLAKE(i915))
-		skl_gt_workarounds_init(i915, wal);
-	else if (IS_BROXTON(i915))
-		bxt_gt_workarounds_init(i915, wal);
-	else if (IS_KABYLAKE(i915))
-		kbl_gt_workarounds_init(i915, wal);
-	else if (IS_GEMINILAKE(i915))
-		glk_gt_workarounds_init(i915, wal);
-	else if (IS_COFFEELAKE(i915))
-		cfl_gt_workarounds_init(i915, wal);
+	if (IS_GEN(i915, 11))
+		icl_gt_workarounds_init(i915, wal);
 	else if (IS_CANNONLAKE(i915))
 		cnl_gt_workarounds_init(i915, wal);
-	else if (IS_ICELAKE(i915))
-		icl_gt_workarounds_init(i915, wal);
+	else if (IS_COFFEELAKE(i915))
+		cfl_gt_workarounds_init(i915, wal);
+	else if (IS_GEMINILAKE(i915))
+		glk_gt_workarounds_init(i915, wal);
+	else if (IS_KABYLAKE(i915))
+		kbl_gt_workarounds_init(i915, wal);
+	else if (IS_BROXTON(i915))
+		bxt_gt_workarounds_init(i915, wal);
+	else if (IS_SKYLAKE(i915))
+		skl_gt_workarounds_init(i915, wal);
+	else if (INTEL_GEN(i915) <= 8)
+		return;
 	else
 		MISSING_CASE(INTEL_GEN(i915));
 }
@@ -896,15 +894,14 @@ void intel_gt_init_workarounds(struct drm_i915_private *i915)
 }
 
 static enum forcewake_domains
-wal_get_fw_for_rmw(struct drm_i915_private *dev_priv,
-		   const struct i915_wa_list *wal)
+wal_get_fw_for_rmw(struct intel_uncore *uncore, const struct i915_wa_list *wal)
 {
 	enum forcewake_domains fw = 0;
 	struct i915_wa *wa;
 	unsigned int i;
 
 	for (i = 0, wa = wal->list; i < wal->count; i++, wa++)
-		fw |= intel_uncore_forcewake_for_reg(dev_priv,
+		fw |= intel_uncore_forcewake_for_reg(uncore,
 						     wa->reg,
 						     FW_REG_READ |
 						     FW_REG_WRITE);
@@ -913,7 +910,7 @@ wal_get_fw_for_rmw(struct drm_i915_private *dev_priv,
 }
 
 static void
-wa_list_apply(struct drm_i915_private *dev_priv, const struct i915_wa_list *wal)
+wa_list_apply(struct intel_uncore *uncore, const struct i915_wa_list *wal)
 {
 	enum forcewake_domains fw;
 	unsigned long flags;
@@ -923,27 +920,22 @@ wa_list_apply(struct drm_i915_private *dev_priv, const struct i915_wa_list *wal)
 	if (!wal->count)
 		return;
 
-	fw = wal_get_fw_for_rmw(dev_priv, wal);
+	fw = wal_get_fw_for_rmw(uncore, wal);
 
-	spin_lock_irqsave(&dev_priv->uncore.lock, flags);
-	intel_uncore_forcewake_get__locked(dev_priv, fw);
+	spin_lock_irqsave(&uncore->lock, flags);
+	intel_uncore_forcewake_get__locked(uncore, fw);
 
 	for (i = 0, wa = wal->list; i < wal->count; i++, wa++) {
-		u32 val = I915_READ_FW(wa->reg);
-
-		val &= ~wa->mask;
-		val |= wa->val;
-
-		I915_WRITE_FW(wa->reg, val);
+		intel_uncore_rmw_fw(uncore, wa->reg, wa->mask, wa->val);
 	}
 
-	intel_uncore_forcewake_put__locked(dev_priv, fw);
-	spin_unlock_irqrestore(&dev_priv->uncore.lock, flags);
+	intel_uncore_forcewake_put__locked(uncore, fw);
+	spin_unlock_irqrestore(&uncore->lock, flags);
 }
 
-void intel_gt_apply_workarounds(struct drm_i915_private *dev_priv)
+void intel_gt_apply_workarounds(struct drm_i915_private *i915)
 {
-	wa_list_apply(dev_priv, &dev_priv->gt_wa_list);
+	wa_list_apply(&i915->uncore, &i915->gt_wa_list);
 }
 
 static bool
@@ -960,7 +952,7 @@ wa_verify(const struct i915_wa *wa, u32 cur, const char *name, const char *from)
 	return true;
 }
 
-static bool wa_list_verify(struct drm_i915_private *dev_priv,
+static bool wa_list_verify(struct intel_uncore *uncore,
 			   const struct i915_wa_list *wal,
 			   const char *from)
 {
@@ -969,15 +961,17 @@ static bool wa_list_verify(struct drm_i915_private *dev_priv,
 	bool ok = true;
 
 	for (i = 0, wa = wal->list; i < wal->count; i++, wa++)
-		ok &= wa_verify(wa, I915_READ(wa->reg), wal->name, from);
+		ok &= wa_verify(wa,
+				intel_uncore_read(uncore, wa->reg),
+				wal->name, from);
 
 	return ok;
 }
 
-bool intel_gt_verify_workarounds(struct drm_i915_private *dev_priv,
+bool intel_gt_verify_workarounds(struct drm_i915_private *i915,
 				 const char *from)
 {
-	return wa_list_verify(dev_priv, &dev_priv->gt_wa_list, from);
+	return wa_list_verify(&i915->uncore, &i915->gt_wa_list, from);
 }
 
 static void
@@ -1052,6 +1046,9 @@ static void icl_whitelist_build(struct i915_wa_list *w)
 
 	/* WaAllowUMDToModifySamplerMode:icl */
 	whitelist_reg(w, GEN10_SAMPLER_MODE);
+
+	/* WaEnableStateCacheRedirectToCS:icl */
+	whitelist_reg(w, GEN9_SLICE_COMMON_ECO_CHICKEN1);
 }
 
 void intel_engine_init_whitelist(struct intel_engine_cs *engine)
@@ -1059,30 +1056,26 @@ void intel_engine_init_whitelist(struct intel_engine_cs *engine)
 	struct drm_i915_private *i915 = engine->i915;
 	struct i915_wa_list *w = &engine->whitelist;
 
-	GEM_BUG_ON(engine->id != RCS);
+	GEM_BUG_ON(engine->id != RCS0);
 
 	wa_init_start(w, "whitelist");
 
-	if (INTEL_GEN(i915) < 8)
-		return;
-	else if (IS_BROADWELL(i915))
-		return;
-	else if (IS_CHERRYVIEW(i915))
-		return;
-	else if (IS_SKYLAKE(i915))
-		skl_whitelist_build(w);
-	else if (IS_BROXTON(i915))
-		bxt_whitelist_build(w);
-	else if (IS_KABYLAKE(i915))
-		kbl_whitelist_build(w);
-	else if (IS_GEMINILAKE(i915))
-		glk_whitelist_build(w);
-	else if (IS_COFFEELAKE(i915))
-		cfl_whitelist_build(w);
+	if (IS_GEN(i915, 11))
+		icl_whitelist_build(w);
 	else if (IS_CANNONLAKE(i915))
 		cnl_whitelist_build(w);
-	else if (IS_ICELAKE(i915))
-		icl_whitelist_build(w);
+	else if (IS_COFFEELAKE(i915))
+		cfl_whitelist_build(w);
+	else if (IS_GEMINILAKE(i915))
+		glk_whitelist_build(w);
+	else if (IS_KABYLAKE(i915))
+		kbl_whitelist_build(w);
+	else if (IS_BROXTON(i915))
+		bxt_whitelist_build(w);
+	else if (IS_SKYLAKE(i915))
+		skl_whitelist_build(w);
+	else if (INTEL_GEN(i915) <= 8)
+		return;
 	else
 		MISSING_CASE(INTEL_GEN(i915));
 
@@ -1091,8 +1084,8 @@ void intel_engine_init_whitelist(struct intel_engine_cs *engine)
 
 void intel_engine_apply_whitelist(struct intel_engine_cs *engine)
 {
-	struct drm_i915_private *dev_priv = engine->i915;
 	const struct i915_wa_list *wal = &engine->whitelist;
+	struct intel_uncore *uncore = engine->uncore;
 	const u32 base = engine->mmio_base;
 	struct i915_wa *wa;
 	unsigned int i;
@@ -1101,13 +1094,15 @@ void intel_engine_apply_whitelist(struct intel_engine_cs *engine)
 		return;
 
 	for (i = 0, wa = wal->list; i < wal->count; i++, wa++)
-		I915_WRITE(RING_FORCE_TO_NONPRIV(base, i),
-			   i915_mmio_reg_offset(wa->reg));
+		intel_uncore_write(uncore,
+				   RING_FORCE_TO_NONPRIV(base, i),
+				   i915_mmio_reg_offset(wa->reg));
 
 	/* And clear the rest just in case of garbage */
 	for (; i < RING_MAX_NONPRIV_SLOTS; i++)
-		I915_WRITE(RING_FORCE_TO_NONPRIV(base, i),
-			   i915_mmio_reg_offset(RING_NOPID(base)));
+		intel_uncore_write(uncore,
+				   RING_FORCE_TO_NONPRIV(base, i),
+				   i915_mmio_reg_offset(RING_NOPID(base)));
 }
 
 static void
@@ -1115,7 +1110,7 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal)
 {
 	struct drm_i915_private *i915 = engine->i915;
 
-	if (IS_ICELAKE(i915)) {
+	if (IS_GEN(i915, 11)) {
 		/* This is not an Wa. Enable for better image quality */
 		wa_masked_en(wal,
 			     _3D_CHICKEN3,
@@ -1170,8 +1165,8 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal)
 				    GEN7_DISABLE_SAMPLER_PREFETCH);
 	}
 
-	if (IS_GEN(i915, 9) || IS_CANNONLAKE(i915)) {
-		/* WaEnablePreemptionGranularityControlByUMD:skl,bxt,kbl,cfl,cnl */
+	if (IS_GEN_RANGE(i915, 9, 11)) {
+		/* FtrPerCtxtPreemptionGranularityControl:skl,bxt,kbl,cfl,cnl,icl */
 		wa_masked_en(wal,
 			     GEN7_FF_SLICE_CS_CHICKEN1,
 			     GEN9_FFSC_PERCTX_PREEMPT_CTRL);
@@ -1236,7 +1231,7 @@ engine_init_workarounds(struct intel_engine_cs *engine, struct i915_wa_list *wal
 	if (I915_SELFTEST_ONLY(INTEL_GEN(engine->i915) < 8))
 		return;
 
-	if (engine->id == RCS)
+	if (engine->id == RCS0)
 		rcs_engine_wa_init(engine, wal);
 	else
 		xcs_engine_wa_init(engine, wal);
@@ -1256,7 +1251,7 @@ void intel_engine_init_workarounds(struct intel_engine_cs *engine)
 
 void intel_engine_apply_workarounds(struct intel_engine_cs *engine)
 {
-	wa_list_apply(engine->i915, &engine->wa_list);
+	wa_list_apply(engine->uncore, &engine->wa_list);
 }
 
 #if IS_ENABLED(CONFIG_DRM_I915_SELFTEST)
diff --git a/drivers/gpu/drm/i915/intel_workarounds.h b/drivers/gpu/drm/i915/intel_workarounds.h
index 7c734714b05e..34eee5ec511e 100644
--- a/drivers/gpu/drm/i915/intel_workarounds.h
+++ b/drivers/gpu/drm/i915/intel_workarounds.h
@@ -9,18 +9,7 @@
 
 #include <linux/slab.h>
 
-struct i915_wa {
-	i915_reg_t	  reg;
-	u32		  mask;
-	u32		  val;
-};
-
-struct i915_wa_list {
-	const char	*name;
-	struct i915_wa	*list;
-	unsigned int	count;
-	unsigned int	wa_count;
-};
+#include "intel_workarounds_types.h"
 
 static inline void intel_wa_list_free(struct i915_wa_list *wal)
 {
@@ -31,9 +20,9 @@ static inline void intel_wa_list_free(struct i915_wa_list *wal)
 void intel_engine_init_ctx_wa(struct intel_engine_cs *engine);
 int intel_engine_emit_ctx_wa(struct i915_request *rq);
 
-void intel_gt_init_workarounds(struct drm_i915_private *dev_priv);
-void intel_gt_apply_workarounds(struct drm_i915_private *dev_priv);
-bool intel_gt_verify_workarounds(struct drm_i915_private *dev_priv,
+void intel_gt_init_workarounds(struct drm_i915_private *i915);
+void intel_gt_apply_workarounds(struct drm_i915_private *i915);
+bool intel_gt_verify_workarounds(struct drm_i915_private *i915,
 				 const char *from);
 
 void intel_engine_init_whitelist(struct intel_engine_cs *engine);
diff --git a/drivers/gpu/drm/i915/intel_workarounds_types.h b/drivers/gpu/drm/i915/intel_workarounds_types.h
new file mode 100644
index 000000000000..30918da180ff
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_workarounds_types.h
@@ -0,0 +1,27 @@
+/*
+ * SPDX-License-Identifier: MIT
+ *
+ * Copyright © 2014-2018 Intel Corporation
+ */
+
+#ifndef __INTEL_WORKAROUNDS_TYPES_H__
+#define __INTEL_WORKAROUNDS_TYPES_H__
+
+#include <linux/types.h>
+
+#include "i915_reg.h"
+
+struct i915_wa {
+	i915_reg_t	  reg;
+	u32		  mask;
+	u32		  val;
+};
+
+struct i915_wa_list {
+	const char	*name;
+	struct i915_wa	*list;
+	unsigned int	count;
+	unsigned int	wa_count;
+};
+
+#endif /* __INTEL_WORKAROUNDS_TYPES_H__ */
diff --git a/drivers/gpu/drm/i915/selftests/huge_gem_object.c b/drivers/gpu/drm/i915/selftests/huge_gem_object.c
index 391f3d9ffdf1..419fd4d6a8f0 100644
--- a/drivers/gpu/drm/i915/selftests/huge_gem_object.c
+++ b/drivers/gpu/drm/i915/selftests/huge_gem_object.c
@@ -122,7 +122,7 @@ huge_gem_object(struct drm_i915_private *i915,
 	if (overflows_type(dma_size, obj->base.size))
 		return ERR_PTR(-E2BIG);
 
-	obj = i915_gem_object_alloc(i915);
+	obj = i915_gem_object_alloc();
 	if (!obj)
 		return ERR_PTR(-ENOMEM);
 
diff --git a/drivers/gpu/drm/i915/selftests/huge_pages.c b/drivers/gpu/drm/i915/selftests/huge_pages.c
index a9a2fa35876f..90721b54e7ae 100644
--- a/drivers/gpu/drm/i915/selftests/huge_pages.c
+++ b/drivers/gpu/drm/i915/selftests/huge_pages.c
@@ -171,7 +171,7 @@ huge_pages_object(struct drm_i915_private *i915,
 	if (overflows_type(size, obj->base.size))
 		return ERR_PTR(-E2BIG);
 
-	obj = i915_gem_object_alloc(i915);
+	obj = i915_gem_object_alloc();
 	if (!obj)
 		return ERR_PTR(-ENOMEM);
 
@@ -320,7 +320,7 @@ fake_huge_pages_object(struct drm_i915_private *i915, u64 size, bool single)
 	if (overflows_type(size, obj->base.size))
 		return ERR_PTR(-E2BIG);
 
-	obj = i915_gem_object_alloc(i915);
+	obj = i915_gem_object_alloc();
 	if (!obj)
 		return ERR_PTR(-ENOMEM);
 
@@ -908,10 +908,6 @@ gpu_write_dw(struct i915_vma *vma, u64 offset, u32 val)
 	if (IS_ERR(obj))
 		return ERR_CAST(obj);
 
-	err = i915_gem_object_set_to_wc_domain(obj, true);
-	if (err)
-		goto err;
-
 	cmd = i915_gem_object_pin_map(obj, I915_MAP_WC);
 	if (IS_ERR(cmd)) {
 		err = PTR_ERR(cmd);
@@ -1449,7 +1445,7 @@ static int igt_ppgtt_pin_update(void *arg)
 	 * huge-gtt-pages.
 	 */
 
-	if (!ppgtt || !i915_vm_is_48bit(&ppgtt->vm)) {
+	if (!ppgtt || !i915_vm_is_4lvl(&ppgtt->vm)) {
 		pr_info("48b PPGTT not supported, skipping\n");
 		return 0;
 	}
@@ -1535,7 +1531,7 @@ static int igt_ppgtt_pin_update(void *arg)
 	 * land in the now stale 2M page.
 	 */
 
-	err = gpu_write(vma, ctx, dev_priv->engine[RCS], 0, 0xdeadbeaf);
+	err = gpu_write(vma, ctx, dev_priv->engine[RCS0], 0, 0xdeadbeaf);
 	if (err)
 		goto out_unpin;
 
@@ -1584,6 +1580,7 @@ static int igt_tmpfs_fallback(void *arg)
 	}
 	*vaddr = 0xdeadbeaf;
 
+	__i915_gem_object_flush_map(obj, 0, 64);
 	i915_gem_object_unpin_map(obj);
 
 	vma = i915_vma_instance(obj, vm, NULL);
@@ -1653,7 +1650,7 @@ static int igt_shrink_thp(void *arg)
 	if (err)
 		goto out_unpin;
 
-	err = gpu_write(vma, ctx, i915->engine[RCS], 0, 0xdeadbeaf);
+	err = gpu_write(vma, ctx, i915->engine[RCS0], 0, 0xdeadbeaf);
 	if (err)
 		goto out_unpin;
 
@@ -1709,16 +1706,17 @@ int i915_gem_huge_page_mock_selftests(void)
 		return -ENOMEM;
 
 	/* Pretend to be a device which supports the 48b PPGTT */
-	mkwrite_device_info(dev_priv)->ppgtt = INTEL_PPGTT_FULL_4LVL;
+	mkwrite_device_info(dev_priv)->ppgtt_type = INTEL_PPGTT_FULL;
+	mkwrite_device_info(dev_priv)->ppgtt_size = 48;
 
 	mutex_lock(&dev_priv->drm.struct_mutex);
-	ppgtt = i915_ppgtt_create(dev_priv, ERR_PTR(-ENODEV));
+	ppgtt = i915_ppgtt_create(dev_priv);
 	if (IS_ERR(ppgtt)) {
 		err = PTR_ERR(ppgtt);
 		goto out_unlock;
 	}
 
-	if (!i915_vm_is_48bit(&ppgtt->vm)) {
+	if (!i915_vm_is_4lvl(&ppgtt->vm)) {
 		pr_err("failed to create 48b PPGTT\n");
 		err = -EINVAL;
 		goto out_close;
@@ -1734,7 +1732,6 @@ int i915_gem_huge_page_mock_selftests(void)
 	err = i915_subtests(tests, ppgtt);
 
 out_close:
-	i915_ppgtt_close(&ppgtt->vm);
 	i915_ppgtt_put(ppgtt);
 
 out_unlock:
@@ -1764,7 +1761,7 @@ int i915_gem_huge_page_live_selftests(struct drm_i915_private *dev_priv)
 		return 0;
 	}
 
-	if (i915_terminally_wedged(&dev_priv->gpu_error))
+	if (i915_terminally_wedged(dev_priv))
 		return 0;
 
 	file = mock_file(dev_priv);
diff --git a/drivers/gpu/drm/i915/selftests/i915_active.c b/drivers/gpu/drm/i915/selftests/i915_active.c
index 337b1f98b923..27d8f853111b 100644
--- a/drivers/gpu/drm/i915/selftests/i915_active.c
+++ b/drivers/gpu/drm/i915/selftests/i915_active.c
@@ -150,7 +150,7 @@ int i915_active_live_selftests(struct drm_i915_private *i915)
 		SUBTEST(live_active_retire),
 	};
 
-	if (i915_terminally_wedged(&i915->gpu_error))
+	if (i915_terminally_wedged(i915))
 		return 0;
 
 	return i915_subtests(tests, i915);
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem.c b/drivers/gpu/drm/i915/selftests/i915_gem.c
index e77b7ed449ae..6fd70d326468 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem.c
@@ -84,14 +84,9 @@ static void simulate_hibernate(struct drm_i915_private *i915)
 
 static int pm_prepare(struct drm_i915_private *i915)
 {
-	int err = 0;
-
-	if (i915_gem_suspend(i915)) {
-		pr_err("i915_gem_suspend failed\n");
-		err = -EINVAL;
-	}
+	i915_gem_suspend(i915);
 
-	return err;
+	return 0;
 }
 
 static void pm_suspend(struct drm_i915_private *i915)
@@ -220,5 +215,8 @@ int i915_gem_live_selftests(struct drm_i915_private *i915)
 		SUBTEST(igt_gem_hibernate),
 	};
 
+	if (i915_terminally_wedged(i915))
+		return 0;
+
 	return i915_subtests(tests, i915);
 }
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_coherency.c b/drivers/gpu/drm/i915/selftests/i915_gem_coherency.c
index fd89a5a33c1a..e43630b40fce 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_coherency.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_coherency.c
@@ -202,7 +202,7 @@ static int gpu_set(struct drm_i915_gem_object *obj,
 	if (IS_ERR(vma))
 		return PTR_ERR(vma);
 
-	rq = i915_request_alloc(i915->engine[RCS], i915->kernel_context);
+	rq = i915_request_alloc(i915->engine[RCS0], i915->kernel_context);
 	if (IS_ERR(rq)) {
 		i915_vma_unpin(vma);
 		return PTR_ERR(rq);
@@ -248,15 +248,15 @@ static bool always_valid(struct drm_i915_private *i915)
 
 static bool needs_fence_registers(struct drm_i915_private *i915)
 {
-	return !i915_terminally_wedged(&i915->gpu_error);
+	return !i915_terminally_wedged(i915);
 }
 
 static bool needs_mi_store_dword(struct drm_i915_private *i915)
 {
-	if (i915_terminally_wedged(&i915->gpu_error))
+	if (i915_terminally_wedged(i915))
 		return false;
 
-	return intel_engine_can_store_dword(i915->engine[RCS]);
+	return intel_engine_can_store_dword(i915->engine[RCS0]);
 }
 
 static const struct igt_coherency_mode {
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_context.c b/drivers/gpu/drm/i915/selftests/i915_gem_context.c
index 7eb58a9d1319..4e1b6efc6b22 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_context.c
@@ -76,7 +76,7 @@ static int live_nop_switch(void *arg)
 	}
 
 	for (n = 0; n < nctx; n++) {
-		ctx[n] = i915_gem_create_context(i915, file->driver_priv);
+		ctx[n] = live_context(i915, file);
 		if (IS_ERR(ctx[n])) {
 			err = PTR_ERR(ctx[n]);
 			goto out_unlock;
@@ -220,6 +220,7 @@ gpu_fill_dw(struct i915_vma *vma, u64 offset, unsigned long count, u32 value)
 		offset += PAGE_SIZE;
 	}
 	*cmd = MI_BATCH_BUFFER_END;
+	i915_gem_object_flush_map(obj);
 	i915_gem_object_unpin_map(obj);
 
 	err = i915_gem_object_set_to_gtt_domain(obj, false);
@@ -372,7 +373,8 @@ static int cpu_fill(struct drm_i915_gem_object *obj, u32 value)
 	return 0;
 }
 
-static int cpu_check(struct drm_i915_gem_object *obj, unsigned int max)
+static noinline int cpu_check(struct drm_i915_gem_object *obj,
+			      unsigned int idx, unsigned int max)
 {
 	unsigned int n, m, needs_flush;
 	int err;
@@ -390,8 +392,10 @@ static int cpu_check(struct drm_i915_gem_object *obj, unsigned int max)
 
 		for (m = 0; m < max; m++) {
 			if (map[m] != m) {
-				pr_err("Invalid value at page %d, offset %d: found %x expected %x\n",
-				       n, m, map[m], m);
+				pr_err("%pS: Invalid value at object %d page %d/%ld, offset %d/%d: found %x expected %x\n",
+				       __builtin_return_address(0), idx,
+				       n, real_page_count(obj), m, max,
+				       map[m], m);
 				err = -EINVAL;
 				goto out_unmap;
 			}
@@ -399,8 +403,9 @@ static int cpu_check(struct drm_i915_gem_object *obj, unsigned int max)
 
 		for (; m < DW_PER_PAGE; m++) {
 			if (map[m] != STACK_MAGIC) {
-				pr_err("Invalid value at page %d, offset %d: found %x expected %x\n",
-				       n, m, map[m], STACK_MAGIC);
+				pr_err("%pS: Invalid value at object %d page %d, offset %d: found %x expected %x (uninitialised)\n",
+				       __builtin_return_address(0), idx, n, m,
+				       map[m], STACK_MAGIC);
 				err = -EINVAL;
 				goto out_unmap;
 			}
@@ -478,12 +483,8 @@ static unsigned long max_dwords(struct drm_i915_gem_object *obj)
 static int igt_ctx_exec(void *arg)
 {
 	struct drm_i915_private *i915 = arg;
-	struct drm_i915_gem_object *obj = NULL;
-	unsigned long ncontexts, ndwords, dw;
-	struct igt_live_test t;
-	struct drm_file *file;
-	IGT_TIMEOUT(end_time);
-	LIST_HEAD(objects);
+	struct intel_engine_cs *engine;
+	enum intel_engine_id id;
 	int err = -ENODEV;
 
 	/*
@@ -495,44 +496,167 @@ static int igt_ctx_exec(void *arg)
 	if (!DRIVER_CAPS(i915)->has_logical_contexts)
 		return 0;
 
+	for_each_engine(engine, i915, id) {
+		struct drm_i915_gem_object *obj = NULL;
+		unsigned long ncontexts, ndwords, dw;
+		struct igt_live_test t;
+		struct drm_file *file;
+		IGT_TIMEOUT(end_time);
+		LIST_HEAD(objects);
+
+		if (!intel_engine_can_store_dword(engine))
+			continue;
+
+		if (!engine->context_size)
+			continue; /* No logical context support in HW */
+
+		file = mock_file(i915);
+		if (IS_ERR(file))
+			return PTR_ERR(file);
+
+		mutex_lock(&i915->drm.struct_mutex);
+
+		err = igt_live_test_begin(&t, i915, __func__, engine->name);
+		if (err)
+			goto out_unlock;
+
+		ncontexts = 0;
+		ndwords = 0;
+		dw = 0;
+		while (!time_after(jiffies, end_time)) {
+			struct i915_gem_context *ctx;
+			intel_wakeref_t wakeref;
+
+			ctx = live_context(i915, file);
+			if (IS_ERR(ctx)) {
+				err = PTR_ERR(ctx);
+				goto out_unlock;
+			}
+
+			if (!obj) {
+				obj = create_test_object(ctx, file, &objects);
+				if (IS_ERR(obj)) {
+					err = PTR_ERR(obj);
+					goto out_unlock;
+				}
+			}
+
+			with_intel_runtime_pm(i915, wakeref)
+				err = gpu_fill(obj, ctx, engine, dw);
+			if (err) {
+				pr_err("Failed to fill dword %lu [%lu/%lu] with gpu (%s) in ctx %u [full-ppgtt? %s], err=%d\n",
+				       ndwords, dw, max_dwords(obj),
+				       engine->name, ctx->hw_id,
+				       yesno(!!ctx->ppgtt), err);
+				goto out_unlock;
+			}
+
+			if (++dw == max_dwords(obj)) {
+				obj = NULL;
+				dw = 0;
+			}
+
+			ndwords++;
+			ncontexts++;
+		}
+
+		pr_info("Submitted %lu contexts to %s, filling %lu dwords\n",
+			ncontexts, engine->name, ndwords);
+
+		ncontexts = dw = 0;
+		list_for_each_entry(obj, &objects, st_link) {
+			unsigned int rem =
+				min_t(unsigned int, ndwords - dw, max_dwords(obj));
+
+			err = cpu_check(obj, ncontexts++, rem);
+			if (err)
+				break;
+
+			dw += rem;
+		}
+
+out_unlock:
+		if (igt_live_test_end(&t))
+			err = -EIO;
+		mutex_unlock(&i915->drm.struct_mutex);
+
+		mock_file_free(i915, file);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int igt_shared_ctx_exec(void *arg)
+{
+	struct drm_i915_private *i915 = arg;
+	struct i915_gem_context *parent;
+	struct intel_engine_cs *engine;
+	enum intel_engine_id id;
+	struct igt_live_test t;
+	struct drm_file *file;
+	int err = 0;
+
+	/*
+	 * Create a few different contexts with the same mm and write
+	 * through each ctx using the GPU making sure those writes end
+	 * up in the expected pages of our obj.
+	 */
+	if (!DRIVER_CAPS(i915)->has_logical_contexts)
+		return 0;
+
 	file = mock_file(i915);
 	if (IS_ERR(file))
 		return PTR_ERR(file);
 
 	mutex_lock(&i915->drm.struct_mutex);
 
+	parent = live_context(i915, file);
+	if (IS_ERR(parent)) {
+		err = PTR_ERR(parent);
+		goto out_unlock;
+	}
+
+	if (!parent->ppgtt) { /* not full-ppgtt; nothing to share */
+		err = 0;
+		goto out_unlock;
+	}
+
 	err = igt_live_test_begin(&t, i915, __func__, "");
 	if (err)
 		goto out_unlock;
 
-	ncontexts = 0;
-	ndwords = 0;
-	dw = 0;
-	while (!time_after(jiffies, end_time)) {
-		struct intel_engine_cs *engine;
-		struct i915_gem_context *ctx;
-		unsigned int id;
+	for_each_engine(engine, i915, id) {
+		unsigned long ncontexts, ndwords, dw;
+		struct drm_i915_gem_object *obj = NULL;
+		IGT_TIMEOUT(end_time);
+		LIST_HEAD(objects);
 
-		ctx = i915_gem_create_context(i915, file->driver_priv);
-		if (IS_ERR(ctx)) {
-			err = PTR_ERR(ctx);
-			goto out_unlock;
-		}
+		if (!intel_engine_can_store_dword(engine))
+			continue;
 
-		for_each_engine(engine, i915, id) {
+		dw = 0;
+		ndwords = 0;
+		ncontexts = 0;
+		while (!time_after(jiffies, end_time)) {
+			struct i915_gem_context *ctx;
 			intel_wakeref_t wakeref;
 
-			if (!engine->context_size)
-				continue; /* No logical context support in HW */
+			ctx = kernel_context(i915);
+			if (IS_ERR(ctx)) {
+				err = PTR_ERR(ctx);
+				goto out_test;
+			}
 
-			if (!intel_engine_can_store_dword(engine))
-				continue;
+			__assign_ppgtt(ctx, parent->ppgtt);
 
 			if (!obj) {
-				obj = create_test_object(ctx, file, &objects);
+				obj = create_test_object(parent, file, &objects);
 				if (IS_ERR(obj)) {
 					err = PTR_ERR(obj);
-					goto out_unlock;
+					kernel_context_close(ctx);
+					goto out_test;
 				}
 			}
 
@@ -544,35 +668,39 @@ static int igt_ctx_exec(void *arg)
 				       ndwords, dw, max_dwords(obj),
 				       engine->name, ctx->hw_id,
 				       yesno(!!ctx->ppgtt), err);
-				goto out_unlock;
+				kernel_context_close(ctx);
+				goto out_test;
 			}
 
 			if (++dw == max_dwords(obj)) {
 				obj = NULL;
 				dw = 0;
 			}
+
 			ndwords++;
+			ncontexts++;
+
+			kernel_context_close(ctx);
 		}
-		ncontexts++;
-	}
-	pr_info("Submitted %lu contexts (across %u engines), filling %lu dwords\n",
-		ncontexts, RUNTIME_INFO(i915)->num_rings, ndwords);
+		pr_info("Submitted %lu contexts to %s, filling %lu dwords\n",
+			ncontexts, engine->name, ndwords);
 
-	dw = 0;
-	list_for_each_entry(obj, &objects, st_link) {
-		unsigned int rem =
-			min_t(unsigned int, ndwords - dw, max_dwords(obj));
+		ncontexts = dw = 0;
+		list_for_each_entry(obj, &objects, st_link) {
+			unsigned int rem =
+				min_t(unsigned int, ndwords - dw, max_dwords(obj));
 
-		err = cpu_check(obj, rem);
-		if (err)
-			break;
+			err = cpu_check(obj, ncontexts++, rem);
+			if (err)
+				goto out_test;
 
-		dw += rem;
+			dw += rem;
+		}
 	}
-
-out_unlock:
+out_test:
 	if (igt_live_test_end(&t))
 		err = -EIO;
+out_unlock:
 	mutex_unlock(&i915->drm.struct_mutex);
 
 	mock_file_free(i915, file);
@@ -604,12 +732,9 @@ static struct i915_vma *rpcs_query_batch(struct i915_vma *vma)
 	*cmd++ = upper_32_bits(vma->node.start);
 	*cmd = MI_BATCH_BUFFER_END;
 
+	__i915_gem_object_flush_map(obj, 0, 64);
 	i915_gem_object_unpin_map(obj);
 
-	err = i915_gem_object_set_to_gtt_domain(obj, false);
-	if (err)
-		goto err;
-
 	vma = i915_vma_instance(obj, vma->vm, NULL);
 	if (IS_ERR(vma)) {
 		err = PTR_ERR(vma);
@@ -923,7 +1048,7 @@ __igt_ctx_sseu(struct drm_i915_private *i915,
 	       unsigned int flags)
 {
 	struct intel_sseu default_sseu = intel_device_default_sseu(i915);
-	struct intel_engine_cs *engine = i915->engine[RCS];
+	struct intel_engine_cs *engine = i915->engine[RCS0];
 	struct drm_i915_gem_object *obj;
 	struct i915_gem_context *ctx;
 	struct intel_sseu pg_sseu;
@@ -962,11 +1087,12 @@ __igt_ctx_sseu(struct drm_i915_private *i915,
 
 	mutex_lock(&i915->drm.struct_mutex);
 
-	ctx = i915_gem_create_context(i915, file->driver_priv);
+	ctx = live_context(i915, file);
 	if (IS_ERR(ctx)) {
 		ret = PTR_ERR(ctx);
 		goto out_unlock;
 	}
+	i915_gem_context_clear_bannable(ctx); /* to reset and beyond! */
 
 	obj = i915_gem_object_create_internal(i915, PAGE_SIZE);
 	if (IS_ERR(obj)) {
@@ -1047,7 +1173,7 @@ static int igt_ctx_readonly(void *arg)
 	struct drm_i915_gem_object *obj = NULL;
 	struct i915_gem_context *ctx;
 	struct i915_hw_ppgtt *ppgtt;
-	unsigned long ndwords, dw;
+	unsigned long idx, ndwords, dw;
 	struct igt_live_test t;
 	struct drm_file *file;
 	I915_RND_STATE(prng);
@@ -1071,7 +1197,7 @@ static int igt_ctx_readonly(void *arg)
 	if (err)
 		goto out_unlock;
 
-	ctx = i915_gem_create_context(i915, file->driver_priv);
+	ctx = live_context(i915, file);
 	if (IS_ERR(ctx)) {
 		err = PTR_ERR(ctx);
 		goto out_unlock;
@@ -1125,9 +1251,10 @@ static int igt_ctx_readonly(void *arg)
 		}
 	}
 	pr_info("Submitted %lu dwords (across %u engines)\n",
-		ndwords, RUNTIME_INFO(i915)->num_rings);
+		ndwords, RUNTIME_INFO(i915)->num_engines);
 
 	dw = 0;
+	idx = 0;
 	list_for_each_entry(obj, &objects, st_link) {
 		unsigned int rem =
 			min_t(unsigned int, ndwords - dw, max_dwords(obj));
@@ -1137,7 +1264,7 @@ static int igt_ctx_readonly(void *arg)
 		if (i915_gem_object_is_readonly(obj))
 			num_writes = 0;
 
-		err = cpu_check(obj, num_writes);
+		err = cpu_check(obj, idx++, num_writes);
 		if (err)
 			break;
 
@@ -1201,12 +1328,9 @@ static int write_to_scratch(struct i915_gem_context *ctx,
 	}
 	*cmd++ = value;
 	*cmd = MI_BATCH_BUFFER_END;
+	__i915_gem_object_flush_map(obj, 0, 64);
 	i915_gem_object_unpin_map(obj);
 
-	err = i915_gem_object_set_to_gtt_domain(obj, false);
-	if (err)
-		goto err;
-
 	vma = i915_vma_instance(obj, &ctx->ppgtt->vm, NULL);
 	if (IS_ERR(vma)) {
 		err = PTR_ERR(vma);
@@ -1298,11 +1422,9 @@ static int read_from_scratch(struct i915_gem_context *ctx,
 		*cmd++ = result;
 	}
 	*cmd = MI_BATCH_BUFFER_END;
-	i915_gem_object_unpin_map(obj);
 
-	err = i915_gem_object_set_to_gtt_domain(obj, false);
-	if (err)
-		goto err;
+	i915_gem_object_flush_map(obj);
+	i915_gem_object_unpin_map(obj);
 
 	vma = i915_vma_instance(obj, &ctx->ppgtt->vm, NULL);
 	if (IS_ERR(vma)) {
@@ -1396,13 +1518,13 @@ static int igt_vm_isolation(void *arg)
 	if (err)
 		goto out_unlock;
 
-	ctx_a = i915_gem_create_context(i915, file->driver_priv);
+	ctx_a = live_context(i915, file);
 	if (IS_ERR(ctx_a)) {
 		err = PTR_ERR(ctx_a);
 		goto out_unlock;
 	}
 
-	ctx_b = i915_gem_create_context(i915, file->driver_priv);
+	ctx_b = live_context(i915, file);
 	if (IS_ERR(ctx_b)) {
 		err = PTR_ERR(ctx_b);
 		goto out_unlock;
@@ -1432,7 +1554,7 @@ static int igt_vm_isolation(void *arg)
 
 			div64_u64_rem(i915_prandom_u64_state(&prng),
 				      vm_total, &offset);
-			offset &= ~sizeof(u32);
+			offset &= -sizeof(u32);
 			offset += I915_GTT_PAGE_SIZE;
 
 			err = write_to_scratch(ctx_a, engine,
@@ -1458,7 +1580,7 @@ static int igt_vm_isolation(void *arg)
 		count += this;
 	}
 	pr_info("Checked %lu scratch offsets across %d engines\n",
-		count, RUNTIME_INFO(i915)->num_rings);
+		count, RUNTIME_INFO(i915)->num_engines);
 
 out_rpm:
 	intel_runtime_pm_put(i915, wakeref);
@@ -1472,10 +1594,10 @@ out_unlock:
 }
 
 static __maybe_unused const char *
-__engine_name(struct drm_i915_private *i915, unsigned int engines)
+__engine_name(struct drm_i915_private *i915, intel_engine_mask_t engines)
 {
 	struct intel_engine_cs *engine;
-	unsigned int tmp;
+	intel_engine_mask_t tmp;
 
 	if (engines == ALL_ENGINES)
 		return "all";
@@ -1488,67 +1610,60 @@ __engine_name(struct drm_i915_private *i915, unsigned int engines)
 
 static int __igt_switch_to_kernel_context(struct drm_i915_private *i915,
 					  struct i915_gem_context *ctx,
-					  unsigned int engines)
+					  intel_engine_mask_t engines)
 {
 	struct intel_engine_cs *engine;
-	unsigned int tmp;
-	int err;
+	intel_engine_mask_t tmp;
+	int pass;
 
 	GEM_TRACE("Testing %s\n", __engine_name(i915, engines));
-	for_each_engine_masked(engine, i915, engines, tmp) {
-		struct i915_request *rq;
+	for (pass = 0; pass < 4; pass++) { /* Once busy; once idle; repeat */
+		bool from_idle = pass & 1;
+		int err;
 
-		rq = i915_request_alloc(engine, ctx);
-		if (IS_ERR(rq))
-			return PTR_ERR(rq);
+		if (!from_idle) {
+			for_each_engine_masked(engine, i915, engines, tmp) {
+				struct i915_request *rq;
 
-		i915_request_add(rq);
-	}
+				rq = i915_request_alloc(engine, ctx);
+				if (IS_ERR(rq))
+					return PTR_ERR(rq);
 
-	err = i915_gem_switch_to_kernel_context(i915);
-	if (err)
-		return err;
-
-	for_each_engine_masked(engine, i915, engines, tmp) {
-		if (!engine_has_kernel_context_barrier(engine)) {
-			pr_err("kernel context not last on engine %s!\n",
-			       engine->name);
-			return -EINVAL;
+				i915_request_add(rq);
+			}
 		}
-	}
 
-	err = i915_gem_wait_for_idle(i915,
-				     I915_WAIT_LOCKED,
-				     MAX_SCHEDULE_TIMEOUT);
-	if (err)
-		return err;
+		err = i915_gem_switch_to_kernel_context(i915,
+							i915->gt.active_engines);
+		if (err)
+			return err;
 
-	GEM_BUG_ON(i915->gt.active_requests);
-	for_each_engine_masked(engine, i915, engines, tmp) {
-		if (engine->last_retired_context->gem_context != i915->kernel_context) {
-			pr_err("engine %s not idling in kernel context!\n",
-			       engine->name);
+		if (!from_idle) {
+			err = i915_gem_wait_for_idle(i915,
+						     I915_WAIT_LOCKED,
+						     MAX_SCHEDULE_TIMEOUT);
+			if (err)
+				return err;
+		}
+
+		if (i915->gt.active_requests) {
+			pr_err("%d active requests remain after switching to kernel context, pass %d (%s) on %s engine%s\n",
+			       i915->gt.active_requests,
+			       pass, from_idle ? "idle" : "busy",
+			       __engine_name(i915, engines),
+			       is_power_of_2(engines) ? "" : "s");
 			return -EINVAL;
 		}
-	}
 
-	err = i915_gem_switch_to_kernel_context(i915);
-	if (err)
-		return err;
+		/* XXX Bonus points for proving we are the kernel context! */
 
-	if (i915->gt.active_requests) {
-		pr_err("switch-to-kernel-context emitted %d requests even though it should already be idling in the kernel context\n",
-		       i915->gt.active_requests);
-		return -EINVAL;
+		mutex_unlock(&i915->drm.struct_mutex);
+		drain_delayed_work(&i915->gt.idle_work);
+		mutex_lock(&i915->drm.struct_mutex);
 	}
 
-	for_each_engine_masked(engine, i915, engines, tmp) {
-		if (!intel_engine_has_kernel_context(engine)) {
-			pr_err("kernel context not last on engine %s!\n",
-			       engine->name);
-			return -EINVAL;
-		}
-	}
+	if (igt_flush_test(i915, I915_WAIT_LOCKED))
+		return -EIO;
 
 	return 0;
 }
@@ -1592,8 +1707,6 @@ static int igt_switch_to_kernel_context(void *arg)
 
 out_unlock:
 	GEM_TRACE_DUMP_ON(err);
-	if (igt_flush_test(i915, I915_WAIT_LOCKED))
-		err = -EIO;
 
 	intel_runtime_pm_put(i915, wakeref);
 	mutex_unlock(&i915->drm.struct_mutex);
@@ -1602,10 +1715,117 @@ out_unlock:
 	return err;
 }
 
+static void mock_barrier_task(void *data)
+{
+	unsigned int *counter = data;
+
+	++*counter;
+}
+
+static int mock_context_barrier(void *arg)
+{
+#undef pr_fmt
+#define pr_fmt(x) "context_barrier_task():" # x
+	struct drm_i915_private *i915 = arg;
+	struct i915_gem_context *ctx;
+	struct i915_request *rq;
+	intel_wakeref_t wakeref;
+	unsigned int counter;
+	int err;
+
+	/*
+	 * The context barrier provides us with a callback after it emits
+	 * a request; useful for retiring old state after loading new.
+	 */
+
+	mutex_lock(&i915->drm.struct_mutex);
+
+	ctx = mock_context(i915, "mock");
+	if (!ctx) {
+		err = -ENOMEM;
+		goto unlock;
+	}
+
+	counter = 0;
+	err = context_barrier_task(ctx, 0,
+				   NULL, mock_barrier_task, &counter);
+	if (err) {
+		pr_err("Failed at line %d, err=%d\n", __LINE__, err);
+		goto out;
+	}
+	if (counter == 0) {
+		pr_err("Did not retire immediately with 0 engines\n");
+		err = -EINVAL;
+		goto out;
+	}
+
+	counter = 0;
+	err = context_barrier_task(ctx, ALL_ENGINES,
+				   NULL, mock_barrier_task, &counter);
+	if (err) {
+		pr_err("Failed at line %d, err=%d\n", __LINE__, err);
+		goto out;
+	}
+	if (counter == 0) {
+		pr_err("Did not retire immediately for all inactive engines\n");
+		err = -EINVAL;
+		goto out;
+	}
+
+	rq = ERR_PTR(-ENODEV);
+	with_intel_runtime_pm(i915, wakeref)
+		rq = i915_request_alloc(i915->engine[RCS0], ctx);
+	if (IS_ERR(rq)) {
+		pr_err("Request allocation failed!\n");
+		goto out;
+	}
+	i915_request_add(rq);
+	GEM_BUG_ON(list_empty(&ctx->active_engines));
+
+	counter = 0;
+	context_barrier_inject_fault = BIT(RCS0);
+	err = context_barrier_task(ctx, ALL_ENGINES,
+				   NULL, mock_barrier_task, &counter);
+	context_barrier_inject_fault = 0;
+	if (err == -ENXIO)
+		err = 0;
+	else
+		pr_err("Did not hit fault injection!\n");
+	if (counter != 0) {
+		pr_err("Invoked callback on error!\n");
+		err = -EIO;
+	}
+	if (err)
+		goto out;
+
+	counter = 0;
+	err = context_barrier_task(ctx, ALL_ENGINES,
+				   NULL, mock_barrier_task, &counter);
+	if (err) {
+		pr_err("Failed at line %d, err=%d\n", __LINE__, err);
+		goto out;
+	}
+	mock_device_flush(i915);
+	if (counter == 0) {
+		pr_err("Did not retire on each active engines\n");
+		err = -EINVAL;
+		goto out;
+	}
+
+out:
+	mock_context_close(ctx);
+unlock:
+	mutex_unlock(&i915->drm.struct_mutex);
+	return err;
+#undef pr_fmt
+#define pr_fmt(x) x
+}
+
 int i915_gem_context_mock_selftests(void)
 {
 	static const struct i915_subtest tests[] = {
 		SUBTEST(igt_switch_to_kernel_context),
+		SUBTEST(mock_context_barrier),
 	};
 	struct drm_i915_private *i915;
 	int err;
@@ -1628,10 +1848,11 @@ int i915_gem_context_live_selftests(struct drm_i915_private *dev_priv)
 		SUBTEST(igt_ctx_exec),
 		SUBTEST(igt_ctx_readonly),
 		SUBTEST(igt_ctx_sseu),
+		SUBTEST(igt_shared_ctx_exec),
 		SUBTEST(igt_vm_isolation),
 	};
 
-	if (i915_terminally_wedged(&dev_priv->gpu_error))
+	if (i915_terminally_wedged(dev_priv))
 		return 0;
 
 	return i915_subtests(tests, dev_priv);
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_dmabuf.c b/drivers/gpu/drm/i915/selftests/i915_gem_dmabuf.c
index a7055b12e53c..2b943ee246c9 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_dmabuf.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_dmabuf.c
@@ -315,6 +315,7 @@ static int igt_dmabuf_export_kmap(void *arg)
 		goto err;
 	}
 	memset(ptr + PAGE_SIZE, 0xaa, PAGE_SIZE);
+	i915_gem_object_flush_map(obj);
 	i915_gem_object_unpin_map(obj);
 
 	ptr = dma_buf_kmap(dmabuf, 1);
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_evict.c b/drivers/gpu/drm/i915/selftests/i915_gem_evict.c
index b9b0ea4e2404..89766688e420 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_evict.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_evict.c
@@ -274,7 +274,7 @@ static int igt_evict_for_cache_color(void *arg)
 		err = PTR_ERR(obj);
 		goto cleanup;
 	}
-	i915_gem_object_set_cache_level(obj, I915_CACHE_LLC);
+	i915_gem_object_set_cache_coherency(obj, I915_CACHE_LLC);
 	quirk_add(obj, &objects);
 
 	vma = i915_gem_object_ggtt_pin(obj, NULL, 0, 0,
@@ -290,7 +290,7 @@ static int igt_evict_for_cache_color(void *arg)
 		err = PTR_ERR(obj);
 		goto cleanup;
 	}
-	i915_gem_object_set_cache_level(obj, I915_CACHE_LLC);
+	i915_gem_object_set_cache_coherency(obj, I915_CACHE_LLC);
 	quirk_add(obj, &objects);
 
 	/* Neighbouring; same colour - should fit */
@@ -547,7 +547,7 @@ int i915_gem_evict_live_selftests(struct drm_i915_private *i915)
 		SUBTEST(igt_evict_contexts),
 	};
 
-	if (i915_terminally_wedged(&i915->gpu_error))
+	if (i915_terminally_wedged(i915))
 		return 0;
 
 	return i915_subtests(tests, i915);
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
index 3850ef4a5ec8..9cca66e4420a 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
@@ -120,7 +120,7 @@ fake_dma_object(struct drm_i915_private *i915, u64 size)
 	if (overflows_type(size, obj->base.size))
 		return ERR_PTR(-E2BIG);
 
-	obj = i915_gem_object_alloc(i915);
+	obj = i915_gem_object_alloc();
 	if (!obj)
 		goto err;
 
@@ -1010,7 +1010,7 @@ static int exercise_ppgtt(struct drm_i915_private *dev_priv,
 		return PTR_ERR(file);
 
 	mutex_lock(&dev_priv->drm.struct_mutex);
-	ppgtt = i915_ppgtt_create(dev_priv, file->driver_priv);
+	ppgtt = i915_ppgtt_create(dev_priv);
 	if (IS_ERR(ppgtt)) {
 		err = PTR_ERR(ppgtt);
 		goto out_unlock;
@@ -1020,7 +1020,6 @@ static int exercise_ppgtt(struct drm_i915_private *dev_priv,
 
 	err = func(dev_priv, &ppgtt->vm, 0, ppgtt->vm.total, end_time);
 
-	i915_ppgtt_close(&ppgtt->vm);
 	i915_ppgtt_put(ppgtt);
 out_unlock:
 	mutex_unlock(&dev_priv->drm.struct_mutex);
@@ -1681,25 +1680,31 @@ int i915_gem_gtt_mock_selftests(void)
 		SUBTEST(igt_gtt_insert),
 	};
 	struct drm_i915_private *i915;
-	struct i915_ggtt ggtt;
+	struct i915_ggtt *ggtt;
 	int err;
 
 	i915 = mock_gem_device();
 	if (!i915)
 		return -ENOMEM;
 
-	mock_init_ggtt(i915, &ggtt);
+	ggtt = kmalloc(sizeof(*ggtt), GFP_KERNEL);
+	if (!ggtt) {
+		err = -ENOMEM;
+		goto out_put;
+	}
+	mock_init_ggtt(i915, ggtt);
 
 	mutex_lock(&i915->drm.struct_mutex);
-	err = i915_subtests(tests, &ggtt);
+	err = i915_subtests(tests, ggtt);
 	mock_device_flush(i915);
 	mutex_unlock(&i915->drm.struct_mutex);
 
 	i915_gem_drain_freed_objects(i915);
 
-	mock_fini_ggtt(&ggtt);
+	mock_fini_ggtt(ggtt);
+	kfree(ggtt);
+out_put:
 	drm_dev_put(&i915->drm);
-
 	return err;
 }
 
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_object.c b/drivers/gpu/drm/i915/selftests/i915_gem_object.c
index 395ae878e0f7..971148fbe6f5 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_object.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_object.c
@@ -468,7 +468,7 @@ static int make_obj_busy(struct drm_i915_gem_object *obj)
 	if (err)
 		return err;
 
-	rq = i915_request_alloc(i915->engine[RCS], i915->kernel_context);
+	rq = i915_request_alloc(i915->engine[RCS0], i915->kernel_context);
 	if (IS_ERR(rq)) {
 		i915_vma_unpin(vma);
 		return PTR_ERR(rq);
@@ -583,7 +583,7 @@ static int igt_mmap_offset_exhaustion(void *arg)
 	for (loop = 0; loop < 3; loop++) {
 		intel_wakeref_t wakeref;
 
-		if (i915_terminally_wedged(&i915->gpu_error))
+		if (i915_terminally_wedged(i915))
 			break;
 
 		obj = i915_gem_object_create_internal(i915, PAGE_SIZE);
diff --git a/drivers/gpu/drm/i915/selftests/i915_request.c b/drivers/gpu/drm/i915/selftests/i915_request.c
index 6733dc5b6b4c..e6ffe2240126 100644
--- a/drivers/gpu/drm/i915/selftests/i915_request.c
+++ b/drivers/gpu/drm/i915/selftests/i915_request.c
@@ -42,7 +42,7 @@ static int igt_add_request(void *arg)
 	/* Basic preliminary test to create a request and let it loose! */
 
 	mutex_lock(&i915->drm.struct_mutex);
-	request = mock_request(i915->engine[RCS],
+	request = mock_request(i915->engine[RCS0],
 			       i915->kernel_context,
 			       HZ / 10);
 	if (!request)
@@ -66,7 +66,7 @@ static int igt_wait_request(void *arg)
 	/* Submit a request, then wait upon it */
 
 	mutex_lock(&i915->drm.struct_mutex);
-	request = mock_request(i915->engine[RCS], i915->kernel_context, T);
+	request = mock_request(i915->engine[RCS0], i915->kernel_context, T);
 	if (!request) {
 		err = -ENOMEM;
 		goto out_unlock;
@@ -136,19 +136,17 @@ static int igt_fence_wait(void *arg)
 	/* Submit a request, treat it as a fence and wait upon it */
 
 	mutex_lock(&i915->drm.struct_mutex);
-	request = mock_request(i915->engine[RCS], i915->kernel_context, T);
+	request = mock_request(i915->engine[RCS0], i915->kernel_context, T);
 	if (!request) {
 		err = -ENOMEM;
 		goto out_locked;
 	}
-	mutex_unlock(&i915->drm.struct_mutex); /* safe as we are single user */
 
 	if (dma_fence_wait_timeout(&request->fence, false, T) != -ETIME) {
 		pr_err("fence wait success before submit (expected timeout)!\n");
-		goto out_device;
+		goto out_locked;
 	}
 
-	mutex_lock(&i915->drm.struct_mutex);
 	i915_request_add(request);
 	mutex_unlock(&i915->drm.struct_mutex);
 
@@ -195,7 +193,7 @@ static int igt_request_rewind(void *arg)
 
 	mutex_lock(&i915->drm.struct_mutex);
 	ctx[0] = mock_context(i915, "A");
-	request = mock_request(i915->engine[RCS], ctx[0], 2 * HZ);
+	request = mock_request(i915->engine[RCS0], ctx[0], 2 * HZ);
 	if (!request) {
 		err = -ENOMEM;
 		goto err_context_0;
@@ -205,7 +203,7 @@ static int igt_request_rewind(void *arg)
 	i915_request_add(request);
 
 	ctx[1] = mock_context(i915, "B");
-	vip = mock_request(i915->engine[RCS], ctx[1], 0);
+	vip = mock_request(i915->engine[RCS0], ctx[1], 0);
 	if (!vip) {
 		err = -ENOMEM;
 		goto err_context_1;
@@ -226,8 +224,7 @@ static int igt_request_rewind(void *arg)
 	mutex_unlock(&i915->drm.struct_mutex);
 
 	if (i915_request_wait(vip, 0, HZ) == -ETIME) {
-		pr_err("timed out waiting for high priority request, vip.seqno=%d, current seqno=%d\n",
-		       vip->global_seqno, intel_engine_get_seqno(i915->engine[RCS]));
+		pr_err("timed out waiting for high priority request\n");
 		goto err;
 	}
 
@@ -418,7 +415,7 @@ static int mock_breadcrumbs_smoketest(void *arg)
 {
 	struct drm_i915_private *i915 = arg;
 	struct smoketest t = {
-		.engine = i915->engine[RCS],
+		.engine = i915->engine[RCS0],
 		.ncontexts = 1024,
 		.max_batch = 1024,
 		.request_alloc = __mock_request_alloc
@@ -622,13 +619,11 @@ static struct i915_vma *empty_batch(struct drm_i915_private *i915)
 	}
 
 	*cmd = MI_BATCH_BUFFER_END;
-	i915_gem_chipset_flush(i915);
 
+	__i915_gem_object_flush_map(obj, 0, 64);
 	i915_gem_object_unpin_map(obj);
 
-	err = i915_gem_object_set_to_gtt_domain(obj, false);
-	if (err)
-		goto err;
+	i915_gem_chipset_flush(i915);
 
 	vma = i915_vma_instance(obj, &i915->ggtt.vm, NULL);
 	if (IS_ERR(vma)) {
@@ -780,10 +775,6 @@ static struct i915_vma *recursive_batch(struct drm_i915_private *i915)
 	if (err)
 		goto err;
 
-	err = i915_gem_object_set_to_wc_domain(obj, true);
-	if (err)
-		goto err;
-
 	cmd = i915_gem_object_pin_map(obj, I915_MAP_WC);
 	if (IS_ERR(cmd)) {
 		err = PTR_ERR(cmd);
@@ -802,10 +793,12 @@ static struct i915_vma *recursive_batch(struct drm_i915_private *i915)
 		*cmd++ = lower_32_bits(vma->node.start);
 	}
 	*cmd++ = MI_BATCH_BUFFER_END; /* terminate early in case of error */
-	i915_gem_chipset_flush(i915);
 
+	__i915_gem_object_flush_map(obj, 0, 64);
 	i915_gem_object_unpin_map(obj);
 
+	i915_gem_chipset_flush(i915);
+
 	return vma;
 
 err:
@@ -1219,7 +1212,7 @@ out_flush:
 		num_fences += atomic_long_read(&t[id].num_fences);
 	}
 	pr_info("Completed %lu waits for %lu fences across %d engines and %d cpus\n",
-		num_waits, num_fences, RUNTIME_INFO(i915)->num_rings, ncpus);
+		num_waits, num_fences, RUNTIME_INFO(i915)->num_engines, ncpus);
 
 	mutex_lock(&i915->drm.struct_mutex);
 	ret = igt_live_test_end(&live) ?: ret;
@@ -1246,7 +1239,7 @@ int i915_request_live_selftests(struct drm_i915_private *i915)
 		SUBTEST(live_breadcrumbs_smoketest),
 	};
 
-	if (i915_terminally_wedged(&i915->gpu_error))
+	if (i915_terminally_wedged(i915))
 		return 0;
 
 	return i915_subtests(tests, i915);
diff --git a/drivers/gpu/drm/i915/selftests/i915_selftest.c b/drivers/gpu/drm/i915/selftests/i915_selftest.c
index 10ef0e636a24..b18eaefef798 100644
--- a/drivers/gpu/drm/i915/selftests/i915_selftest.c
+++ b/drivers/gpu/drm/i915/selftests/i915_selftest.c
@@ -133,7 +133,7 @@ static int __run_selftests(const char *name,
 		if (signal_pending(current))
 			return -EINTR;
 
-		pr_debug(DRIVER_NAME ": Running %s\n", st->name);
+		pr_info(DRIVER_NAME ": Running %s\n", st->name);
 		if (data)
 			err = st->live(data);
 		else
@@ -255,7 +255,7 @@ int __i915_subtests(const char *caller,
 		if (!apply_subtest_filter(caller, st->name))
 			continue;
 
-		pr_debug(DRIVER_NAME ": Running %s/%s\n", caller, st->name);
+		pr_info(DRIVER_NAME ": Running %s/%s\n", caller, st->name);
 		GEM_TRACE("Running %s/%s\n", caller, st->name);
 
 		err = st->func(data);
diff --git a/drivers/gpu/drm/i915/selftests/i915_sw_fence.c b/drivers/gpu/drm/i915/selftests/i915_sw_fence.c
index cdbc8f134e5e..cbf45d85cbff 100644
--- a/drivers/gpu/drm/i915/selftests/i915_sw_fence.c
+++ b/drivers/gpu/drm/i915/selftests/i915_sw_fence.c
@@ -571,21 +571,27 @@ static int test_timer(void *arg)
 	unsigned long target, delay;
 	struct timed_fence tf;
 
+	preempt_disable();
 	timed_fence_init(&tf, target = jiffies);
 	if (!i915_sw_fence_done(&tf.fence)) {
 		pr_err("Fence with immediate expiration not signaled\n");
 		goto err;
 	}
+	preempt_enable();
 	timed_fence_fini(&tf);
 
 	for_each_prime_number(delay, i915_selftest.timeout_jiffies/2) {
+		preempt_disable();
 		timed_fence_init(&tf, target = jiffies + delay);
 		if (i915_sw_fence_done(&tf.fence)) {
 			pr_err("Fence with future expiration (%lu jiffies) already signaled\n", delay);
 			goto err;
 		}
+		preempt_enable();
 
 		i915_sw_fence_wait(&tf.fence);
+
+		preempt_disable();
 		if (!i915_sw_fence_done(&tf.fence)) {
 			pr_err("Fence not signaled after wait\n");
 			goto err;
@@ -595,13 +601,14 @@ static int test_timer(void *arg)
 			       target, jiffies);
 			goto err;
 		}
-
+		preempt_enable();
 		timed_fence_fini(&tf);
 	}
 
 	return 0;
 
 err:
+	preempt_enable();
 	timed_fence_fini(&tf);
 	return -EINVAL;
 }
diff --git a/drivers/gpu/drm/i915/selftests/i915_timeline.c b/drivers/gpu/drm/i915/selftests/i915_timeline.c
index 12ea69b1a1e5..bd96afcadfe7 100644
--- a/drivers/gpu/drm/i915/selftests/i915_timeline.c
+++ b/drivers/gpu/drm/i915/selftests/i915_timeline.c
@@ -64,7 +64,7 @@ static int __mock_hwsp_timeline(struct mock_hwsp_freelist *state,
 		unsigned long cacheline;
 		int err;
 
-		tl = i915_timeline_create(state->i915, "mock", NULL);
+		tl = i915_timeline_create(state->i915, NULL);
 		if (IS_ERR(tl))
 			return PTR_ERR(tl);
 
@@ -476,7 +476,7 @@ checked_i915_timeline_create(struct drm_i915_private *i915)
 {
 	struct i915_timeline *tl;
 
-	tl = i915_timeline_create(i915, "live", NULL);
+	tl = i915_timeline_create(i915, NULL);
 	if (IS_ERR(tl))
 		return tl;
 
@@ -641,6 +641,118 @@ out:
 #undef NUM_TIMELINES
 }
 
+static int live_hwsp_wrap(void *arg)
+{
+	struct drm_i915_private *i915 = arg;
+	struct intel_engine_cs *engine;
+	struct i915_timeline *tl;
+	enum intel_engine_id id;
+	intel_wakeref_t wakeref;
+	int err = 0;
+
+	/*
+	 * Across a seqno wrap, we need to keep the old cacheline alive for
+	 * foreign GPU references.
+	 */
+
+	mutex_lock(&i915->drm.struct_mutex);
+	wakeref = intel_runtime_pm_get(i915);
+
+	tl = i915_timeline_create(i915, NULL);
+	if (IS_ERR(tl)) {
+		err = PTR_ERR(tl);
+		goto out_rpm;
+	}
+	if (!tl->has_initial_breadcrumb || !tl->hwsp_cacheline)
+		goto out_free;
+
+	err = i915_timeline_pin(tl);
+	if (err)
+		goto out_free;
+
+	for_each_engine(engine, i915, id) {
+		const u32 *hwsp_seqno[2];
+		struct i915_request *rq;
+		u32 seqno[2];
+
+		if (!intel_engine_can_store_dword(engine))
+			continue;
+
+		rq = i915_request_alloc(engine, i915->kernel_context);
+		if (IS_ERR(rq)) {
+			err = PTR_ERR(rq);
+			goto out;
+		}
+
+		tl->seqno = -4u;
+
+		err = i915_timeline_get_seqno(tl, rq, &seqno[0]);
+		if (err) {
+			i915_request_add(rq);
+			goto out;
+		}
+		pr_debug("seqno[0]:%08x, hwsp_offset:%08x\n",
+			 seqno[0], tl->hwsp_offset);
+
+		err = emit_ggtt_store_dw(rq, tl->hwsp_offset, seqno[0]);
+		if (err) {
+			i915_request_add(rq);
+			goto out;
+		}
+		hwsp_seqno[0] = tl->hwsp_seqno;
+
+		err = i915_timeline_get_seqno(tl, rq, &seqno[1]);
+		if (err) {
+			i915_request_add(rq);
+			goto out;
+		}
+		pr_debug("seqno[1]:%08x, hwsp_offset:%08x\n",
+			 seqno[1], tl->hwsp_offset);
+
+		err = emit_ggtt_store_dw(rq, tl->hwsp_offset, seqno[1]);
+		if (err) {
+			i915_request_add(rq);
+			goto out;
+		}
+		hwsp_seqno[1] = tl->hwsp_seqno;
+
+		/* With wrap should come a new hwsp */
+		GEM_BUG_ON(seqno[1] >= seqno[0]);
+		GEM_BUG_ON(hwsp_seqno[0] == hwsp_seqno[1]);
+
+		i915_request_add(rq);
+
+		if (i915_request_wait(rq, I915_WAIT_LOCKED, HZ / 5) < 0) {
+			pr_err("Wait for timeline writes timed out!\n");
+			err = -EIO;
+			goto out;
+		}
+
+		if (*hwsp_seqno[0] != seqno[0] || *hwsp_seqno[1] != seqno[1]) {
+			pr_err("Bad timeline values: found (%x, %x), expected (%x, %x)\n",
+			       *hwsp_seqno[0], *hwsp_seqno[1],
+			       seqno[0], seqno[1]);
+			err = -EINVAL;
+			goto out;
+		}
+
+		i915_retire_requests(i915); /* recycle HWSP */
+	}
+
+out:
+	if (igt_flush_test(i915, I915_WAIT_LOCKED))
+		err = -EIO;
+
+	i915_timeline_unpin(tl);
+out_free:
+	i915_timeline_put(tl);
+out_rpm:
+	intel_runtime_pm_put(i915, wakeref);
+	mutex_unlock(&i915->drm.struct_mutex);
+
+	return err;
+}
+
 static int live_hwsp_recycle(void *arg)
 {
 	struct drm_i915_private *i915 = arg;
@@ -723,7 +835,11 @@ int i915_timeline_live_selftests(struct drm_i915_private *i915)
 		SUBTEST(live_hwsp_recycle),
 		SUBTEST(live_hwsp_engine),
 		SUBTEST(live_hwsp_alternate),
+		SUBTEST(live_hwsp_wrap),
 	};
 
+	if (i915_terminally_wedged(i915))
+		return 0;
+
 	return i915_subtests(tests, i915);
 }
diff --git a/drivers/gpu/drm/i915/selftests/i915_vma.c b/drivers/gpu/drm/i915/selftests/i915_vma.c
index cf1de82741fa..fc594b030f5a 100644
--- a/drivers/gpu/drm/i915/selftests/i915_vma.c
+++ b/drivers/gpu/drm/i915/selftests/i915_vma.c
@@ -725,24 +725,30 @@ int i915_vma_mock_selftests(void)
 		SUBTEST(igt_vma_partial),
 	};
 	struct drm_i915_private *i915;
-	struct i915_ggtt ggtt;
+	struct i915_ggtt *ggtt;
 	int err;
 
 	i915 = mock_gem_device();
 	if (!i915)
 		return -ENOMEM;
 
-	mock_init_ggtt(i915, &ggtt);
+	ggtt = kmalloc(sizeof(*ggtt), GFP_KERNEL);
+	if (!ggtt) {
+		err = -ENOMEM;
+		goto out_put;
+	}
+	mock_init_ggtt(i915, ggtt);
 
 	mutex_lock(&i915->drm.struct_mutex);
-	err = i915_subtests(tests, &ggtt);
+	err = i915_subtests(tests, ggtt);
 	mock_device_flush(i915);
 	mutex_unlock(&i915->drm.struct_mutex);
 
 	i915_gem_drain_freed_objects(i915);
 
-	mock_fini_ggtt(&ggtt);
+	mock_fini_ggtt(ggtt);
+	kfree(ggtt);
+out_put:
 	drm_dev_put(&i915->drm);
-
 	return err;
 }
diff --git a/drivers/gpu/drm/i915/selftests/igt_flush_test.c b/drivers/gpu/drm/i915/selftests/igt_flush_test.c
index af66e3d4e23a..94aee4071a66 100644
--- a/drivers/gpu/drm/i915/selftests/igt_flush_test.c
+++ b/drivers/gpu/drm/i915/selftests/igt_flush_test.c
@@ -14,7 +14,7 @@ int igt_flush_test(struct drm_i915_private *i915, unsigned int flags)
 	cond_resched();
 
 	if (flags & I915_WAIT_LOCKED &&
-	    i915_gem_switch_to_kernel_context(i915)) {
+	    i915_gem_switch_to_kernel_context(i915, i915->gt.active_engines)) {
 		pr_err("Failed to switch back to kernel context; declaring wedged\n");
 		i915_gem_set_wedged(i915);
 	}
@@ -29,5 +29,5 @@ int igt_flush_test(struct drm_i915_private *i915, unsigned int flags)
 		i915_gem_set_wedged(i915);
 	}
 
-	return i915_terminally_wedged(&i915->gpu_error) ? -EIO : 0;
+	return i915_terminally_wedged(i915);
 }
diff --git a/drivers/gpu/drm/i915/selftests/igt_spinner.c b/drivers/gpu/drm/i915/selftests/igt_spinner.c
index 9ebd9225684e..16890dfe74c0 100644
--- a/drivers/gpu/drm/i915/selftests/igt_spinner.c
+++ b/drivers/gpu/drm/i915/selftests/igt_spinner.c
@@ -29,7 +29,7 @@ int igt_spinner_init(struct igt_spinner *spin, struct drm_i915_private *i915)
 		goto err_hws;
 	}
 
-	i915_gem_object_set_cache_level(spin->hws, I915_CACHE_LLC);
+	i915_gem_object_set_cache_coherency(spin->hws, I915_CACHE_LLC);
 	vaddr = i915_gem_object_pin_map(spin->hws, I915_MAP_WB);
 	if (IS_ERR(vaddr)) {
 		err = PTR_ERR(vaddr);
@@ -144,6 +144,13 @@ igt_spinner_create_request(struct igt_spinner *spin,
 
 	i915_gem_chipset_flush(spin->i915);
 
+	if (engine->emit_init_breadcrumb &&
+	    rq->timeline->has_initial_breadcrumb) {
+		err = engine->emit_init_breadcrumb(rq);
+		if (err)
+			goto cancel_rq;
+	}
+
 	err = engine->emit_bb_start(rq, vma->node.start, PAGE_SIZE, 0);
 
 cancel_rq:
diff --git a/drivers/gpu/drm/i915/selftests/intel_guc.c b/drivers/gpu/drm/i915/selftests/intel_guc.c
index c5e0a0e98fcb..b05a21eaa8f4 100644
--- a/drivers/gpu/drm/i915/selftests/intel_guc.c
+++ b/drivers/gpu/drm/i915/selftests/intel_guc.c
@@ -111,7 +111,7 @@ static int validate_client(struct intel_guc_client *client,
 			dev_priv->preempt_context : dev_priv->kernel_context;
 
 	if (client->owner != ctx_owner ||
-	    client->engines != INTEL_INFO(dev_priv)->ring_mask ||
+	    client->engines != INTEL_INFO(dev_priv)->engine_mask ||
 	    client->priority != client_priority ||
 	    client->doorbell_id == GUC_DOORBELL_INVALID)
 		return -EINVAL;
@@ -261,7 +261,7 @@ static int igt_guc_doorbells(void *arg)
 
 	for (i = 0; i < ATTEMPTS; i++) {
 		clients[i] = guc_client_alloc(dev_priv,
-					      INTEL_INFO(dev_priv)->ring_mask,
+					      INTEL_INFO(dev_priv)->engine_mask,
 					      i % GUC_CLIENT_PRIORITY_NUM,
 					      dev_priv->kernel_context);
 
diff --git a/drivers/gpu/drm/i915/selftests/intel_hangcheck.c b/drivers/gpu/drm/i915/selftests/intel_hangcheck.c
index 7b6f3bea9ef8..050bd1e19e02 100644
--- a/drivers/gpu/drm/i915/selftests/intel_hangcheck.c
+++ b/drivers/gpu/drm/i915/selftests/intel_hangcheck.c
@@ -56,6 +56,8 @@ static int hang_init(struct hang *h, struct drm_i915_private *i915)
 	if (IS_ERR(h->ctx))
 		return PTR_ERR(h->ctx);
 
+	GEM_BUG_ON(i915_gem_context_is_bannable(h->ctx));
+
 	h->hws = i915_gem_object_create_internal(i915, PAGE_SIZE);
 	if (IS_ERR(h->hws)) {
 		err = PTR_ERR(h->hws);
@@ -68,7 +70,7 @@ static int hang_init(struct hang *h, struct drm_i915_private *i915)
 		goto err_hws;
 	}
 
-	i915_gem_object_set_cache_level(h->hws, I915_CACHE_LLC);
+	i915_gem_object_set_cache_coherency(h->hws, I915_CACHE_LLC);
 	vaddr = i915_gem_object_pin_map(h->hws, I915_MAP_WB);
 	if (IS_ERR(vaddr)) {
 		err = PTR_ERR(vaddr);
@@ -242,6 +244,12 @@ hang_create_request(struct hang *h, struct intel_engine_cs *engine)
 	*batch++ = MI_BATCH_BUFFER_END; /* not reached */
 	i915_gem_chipset_flush(h->i915);
 
+	if (rq->engine->emit_init_breadcrumb) {
+		err = rq->engine->emit_init_breadcrumb(rq);
+		if (err)
+			goto cancel_rq;
+	}
+
 	flags = 0;
 	if (INTEL_GEN(vm->i915) <= 5)
 		flags |= I915_DISPATCH_SECURE;
@@ -334,7 +342,7 @@ static int igt_hang_sanitycheck(void *arg)
 			timeout = i915_request_wait(rq,
 						    I915_WAIT_LOCKED,
 						    MAX_SCHEDULE_TIMEOUT);
-		if (i915_terminally_wedged(&i915->gpu_error))
+		if (i915_reset_failed(i915))
 			timeout = -EIO;
 
 		i915_request_put(rq);
@@ -375,7 +383,7 @@ static int igt_global_reset(void *arg)
 
 	igt_global_reset_unlock(i915);
 
-	if (i915_terminally_wedged(&i915->gpu_error))
+	if (i915_reset_failed(i915))
 		err = -EIO;
 
 	return err;
@@ -393,15 +401,13 @@ static int igt_wedged_reset(void *arg)
 
 	i915_gem_set_wedged(i915);
 
-	mutex_lock(&i915->drm.struct_mutex);
-	GEM_BUG_ON(!i915_terminally_wedged(&i915->gpu_error));
+	GEM_BUG_ON(!i915_reset_failed(i915));
 	i915_reset(i915, ALL_ENGINES, NULL);
-	mutex_unlock(&i915->drm.struct_mutex);
 
 	intel_runtime_pm_put(i915, wakeref);
 	igt_global_reset_unlock(i915);
 
-	return i915_terminally_wedged(&i915->gpu_error) ? -EIO : 0;
+	return i915_reset_failed(i915) ? -EIO : 0;
 }
 
 static bool wait_for_idle(struct intel_engine_cs *engine)
@@ -409,6 +415,222 @@ static bool wait_for_idle(struct intel_engine_cs *engine)
 	return wait_for(intel_engine_is_idle(engine), IGT_IDLE_TIMEOUT) == 0;
 }
 
+static int igt_reset_nop(void *arg)
+{
+	struct drm_i915_private *i915 = arg;
+	struct intel_engine_cs *engine;
+	struct i915_gem_context *ctx;
+	unsigned int reset_count, count;
+	enum intel_engine_id id;
+	intel_wakeref_t wakeref;
+	struct drm_file *file;
+	IGT_TIMEOUT(end_time);
+	int err = 0;
+
+	/* Check that we can reset during non-user portions of requests */
+
+	file = mock_file(i915);
+	if (IS_ERR(file))
+		return PTR_ERR(file);
+
+	mutex_lock(&i915->drm.struct_mutex);
+	ctx = live_context(i915, file);
+	mutex_unlock(&i915->drm.struct_mutex);
+	if (IS_ERR(ctx)) {
+		err = PTR_ERR(ctx);
+		goto out;
+	}
+
+	i915_gem_context_clear_bannable(ctx);
+	wakeref = intel_runtime_pm_get(i915);
+	reset_count = i915_reset_count(&i915->gpu_error);
+	count = 0;
+	do {
+		mutex_lock(&i915->drm.struct_mutex);
+		for_each_engine(engine, i915, id) {
+			int i;
+
+			for (i = 0; i < 16; i++) {
+				struct i915_request *rq;
+
+				rq = i915_request_alloc(engine, ctx);
+				if (IS_ERR(rq)) {
+					err = PTR_ERR(rq);
+					break;
+				}
+
+				i915_request_add(rq);
+			}
+		}
+		mutex_unlock(&i915->drm.struct_mutex);
+
+		igt_global_reset_lock(i915);
+		i915_reset(i915, ALL_ENGINES, NULL);
+		igt_global_reset_unlock(i915);
+		if (i915_reset_failed(i915)) {
+			err = -EIO;
+			break;
+		}
+
+		if (i915_reset_count(&i915->gpu_error) !=
+		    reset_count + ++count) {
+			pr_err("Full GPU reset not recorded!\n");
+			err = -EINVAL;
+			break;
+		}
+
+		if (!i915_reset_flush(i915)) {
+			struct drm_printer p =
+				drm_info_printer(i915->drm.dev);
+
+			pr_err("%s failed to idle after reset\n",
+			       engine->name);
+			intel_engine_dump(engine, &p,
+					  "%s\n", engine->name);
+
+			err = -EIO;
+			break;
+		}
+
+		err = igt_flush_test(i915, 0);
+		if (err)
+			break;
+	} while (time_before(jiffies, end_time));
+	pr_info("%s: %d resets\n", __func__, count);
+
+	mutex_lock(&i915->drm.struct_mutex);
+	err = igt_flush_test(i915, I915_WAIT_LOCKED);
+	mutex_unlock(&i915->drm.struct_mutex);
+
+	intel_runtime_pm_put(i915, wakeref);
+
+out:
+	mock_file_free(i915, file);
+	if (i915_reset_failed(i915))
+		err = -EIO;
+	return err;
+}
+
+static int igt_reset_nop_engine(void *arg)
+{
+	struct drm_i915_private *i915 = arg;
+	struct intel_engine_cs *engine;
+	struct i915_gem_context *ctx;
+	enum intel_engine_id id;
+	intel_wakeref_t wakeref;
+	struct drm_file *file;
+	int err = 0;
+
+	/* Check that we can engine-reset during non-user portions */
+
+	if (!intel_has_reset_engine(i915))
+		return 0;
+
+	file = mock_file(i915);
+	if (IS_ERR(file))
+		return PTR_ERR(file);
+
+	mutex_lock(&i915->drm.struct_mutex);
+	ctx = live_context(i915, file);
+	mutex_unlock(&i915->drm.struct_mutex);
+	if (IS_ERR(ctx)) {
+		err = PTR_ERR(ctx);
+		goto out;
+	}
+
+	i915_gem_context_clear_bannable(ctx);
+	wakeref = intel_runtime_pm_get(i915);
+	for_each_engine(engine, i915, id) {
+		unsigned int reset_count, reset_engine_count;
+		unsigned int count;
+		IGT_TIMEOUT(end_time);
+
+		reset_count = i915_reset_count(&i915->gpu_error);
+		reset_engine_count = i915_reset_engine_count(&i915->gpu_error,
+							     engine);
+		count = 0;
+
+		set_bit(I915_RESET_ENGINE + id, &i915->gpu_error.flags);
+		do {
+			int i;
+
+			if (!wait_for_idle(engine)) {
+				pr_err("%s failed to idle before reset\n",
+				       engine->name);
+				err = -EIO;
+				break;
+			}
+
+			mutex_lock(&i915->drm.struct_mutex);
+			for (i = 0; i < 16; i++) {
+				struct i915_request *rq;
+
+				rq = i915_request_alloc(engine, ctx);
+				if (IS_ERR(rq)) {
+					err = PTR_ERR(rq);
+					break;
+				}
+
+				i915_request_add(rq);
+			}
+			mutex_unlock(&i915->drm.struct_mutex);
+
+			err = i915_reset_engine(engine, NULL);
+			if (err) {
+				pr_err("i915_reset_engine failed\n");
+				break;
+			}
+
+			if (i915_reset_count(&i915->gpu_error) != reset_count) {
+				pr_err("Full GPU reset recorded! (engine reset expected)\n");
+				err = -EINVAL;
+				break;
+			}
+
+			if (i915_reset_engine_count(&i915->gpu_error, engine) !=
+			    reset_engine_count + ++count) {
+				pr_err("%s engine reset not recorded!\n",
+				       engine->name);
+				err = -EINVAL;
+				break;
+			}
+
+			if (!i915_reset_flush(i915)) {
+				struct drm_printer p =
+					drm_info_printer(i915->drm.dev);
+
+				pr_err("%s failed to idle after reset\n",
+				       engine->name);
+				intel_engine_dump(engine, &p,
+						  "%s\n", engine->name);
+
+				err = -EIO;
+				break;
+			}
+		} while (time_before(jiffies, end_time));
+		clear_bit(I915_RESET_ENGINE + id, &i915->gpu_error.flags);
+		pr_info("%s(%s): %d resets\n", __func__, engine->name, count);
+
+		if (err)
+			break;
+
+		err = igt_flush_test(i915, 0);
+		if (err)
+			break;
+	}
+
+	mutex_lock(&i915->drm.struct_mutex);
+	err = igt_flush_test(i915, I915_WAIT_LOCKED);
+	mutex_unlock(&i915->drm.struct_mutex);
+
+	intel_runtime_pm_put(i915, wakeref);
+out:
+	mock_file_free(i915, file);
+	if (i915_reset_failed(i915))
+		err = -EIO;
+	return err;
+}
+
 static int __igt_reset_engine(struct drm_i915_private *i915, bool active)
 {
 	struct intel_engine_cs *engine;
@@ -523,7 +745,7 @@ static int __igt_reset_engine(struct drm_i915_private *i915, bool active)
 			break;
 	}
 
-	if (i915_terminally_wedged(&i915->gpu_error))
+	if (i915_reset_failed(i915))
 		err = -EIO;
 
 	if (active) {
@@ -565,11 +787,10 @@ static int active_request_put(struct i915_request *rq)
 		return 0;
 
 	if (i915_request_wait(rq, 0, 5 * HZ) < 0) {
-		GEM_TRACE("%s timed out waiting for completion of fence %llx:%lld, seqno %d.\n",
+		GEM_TRACE("%s timed out waiting for completion of fence %llx:%lld\n",
 			  rq->engine->name,
 			  rq->fence.context,
-			  rq->fence.seqno,
-			  i915_request_global_seqno(rq));
+			  rq->fence.seqno);
 		GEM_TRACE_DUMP();
 
 		i915_gem_set_wedged(rq->i915);
@@ -762,7 +983,23 @@ static int __igt_reset_engines(struct drm_i915_private *i915,
 			count++;
 
 			if (rq) {
-				i915_request_wait(rq, 0, MAX_SCHEDULE_TIMEOUT);
+				if (i915_request_wait(rq, 0, HZ / 5) < 0) {
+					struct drm_printer p =
+						drm_info_printer(i915->drm.dev);
+
+					pr_err("i915_reset_engine(%s:%s):"
+					       " failed to complete request after reset\n",
+					       engine->name, test_name);
+					intel_engine_dump(engine, &p,
+							  "%s\n", engine->name);
+					i915_request_put(rq);
+
+					GEM_TRACE_DUMP();
+					i915_gem_set_wedged(i915);
+					err = -EIO;
+					break;
+				}
+
 				i915_request_put(rq);
 			}
 
@@ -837,7 +1074,7 @@ unwind:
 			break;
 	}
 
-	if (i915_terminally_wedged(&i915->gpu_error))
+	if (i915_reset_failed(i915))
 		err = -EIO;
 
 	if (flags & TEST_ACTIVE) {
@@ -887,7 +1124,8 @@ static int igt_reset_engines(void *arg)
 	return 0;
 }
 
-static u32 fake_hangcheck(struct drm_i915_private *i915, u32 mask)
+static u32 fake_hangcheck(struct drm_i915_private *i915,
+			  intel_engine_mask_t mask)
 {
 	u32 count = i915_reset_count(&i915->gpu_error);
 
@@ -905,7 +1143,7 @@ static int igt_reset_wait(void *arg)
 	long timeout;
 	int err;
 
-	if (!intel_engine_can_store_dword(i915->engine[RCS]))
+	if (!intel_engine_can_store_dword(i915->engine[RCS0]))
 		return 0;
 
 	/* Check that we detect a stuck waiter and issue a reset */
@@ -917,7 +1155,7 @@ static int igt_reset_wait(void *arg)
 	if (err)
 		goto unlock;
 
-	rq = hang_create_request(&h, i915->engine[RCS]);
+	rq = hang_create_request(&h, i915->engine[RCS0]);
 	if (IS_ERR(rq)) {
 		err = PTR_ERR(rq);
 		goto fini;
@@ -963,7 +1201,7 @@ unlock:
 	mutex_unlock(&i915->drm.struct_mutex);
 	igt_global_reset_unlock(i915);
 
-	if (i915_terminally_wedged(&i915->gpu_error))
+	if (i915_reset_failed(i915))
 		return -EIO;
 
 	return err;
@@ -1034,13 +1272,11 @@ static int __igt_reset_evict_vma(struct drm_i915_private *i915,
 	struct hang h;
 	int err;
 
-	if (!intel_engine_can_store_dword(i915->engine[RCS]))
+	if (!intel_engine_can_store_dword(i915->engine[RCS0]))
 		return 0;
 
 	/* Check that we can recover an unbind stuck on a hanging request */
 
-	igt_global_reset_lock(i915);
-
 	mutex_lock(&i915->drm.struct_mutex);
 	err = hang_init(&h, i915);
 	if (err)
@@ -1066,7 +1302,7 @@ static int __igt_reset_evict_vma(struct drm_i915_private *i915,
 		goto out_obj;
 	}
 
-	rq = hang_create_request(&h, i915->engine[RCS]);
+	rq = hang_create_request(&h, i915->engine[RCS0]);
 	if (IS_ERR(rq)) {
 		err = PTR_ERR(rq);
 		goto out_obj;
@@ -1138,7 +1374,9 @@ static int __igt_reset_evict_vma(struct drm_i915_private *i915,
 	}
 
 out_reset:
-	fake_hangcheck(rq->i915, intel_engine_flag(rq->engine));
+	igt_global_reset_lock(i915);
+	fake_hangcheck(rq->i915, rq->engine->mask);
+	igt_global_reset_unlock(i915);
 
 	if (tsk) {
 		struct igt_wedge_me w;
@@ -1159,9 +1397,8 @@ fini:
 	hang_fini(&h);
 unlock:
 	mutex_unlock(&i915->drm.struct_mutex);
-	igt_global_reset_unlock(i915);
 
-	if (i915_terminally_wedged(&i915->gpu_error))
+	if (i915_reset_failed(i915))
 		return -EIO;
 
 	return err;
@@ -1317,7 +1554,7 @@ static int igt_reset_queue(void *arg)
 				goto fini;
 			}
 
-			reset_count = fake_hangcheck(i915, ENGINE_MASK(id));
+			reset_count = fake_hangcheck(i915, BIT(id));
 
 			if (prev->fence.error != -EIO) {
 				pr_err("GPU reset not recorded on hanging request [fence.error=%d]!\n",
@@ -1367,7 +1604,7 @@ unlock:
 	mutex_unlock(&i915->drm.struct_mutex);
 	igt_global_reset_unlock(i915);
 
-	if (i915_terminally_wedged(&i915->gpu_error))
+	if (i915_reset_failed(i915))
 		return -EIO;
 
 	return err;
@@ -1376,7 +1613,7 @@ unlock:
 static int igt_handle_error(void *arg)
 {
 	struct drm_i915_private *i915 = arg;
-	struct intel_engine_cs *engine = i915->engine[RCS];
+	struct intel_engine_cs *engine = i915->engine[RCS0];
 	struct hang h;
 	struct i915_request *rq;
 	struct i915_gpu_state *error;
@@ -1423,7 +1660,7 @@ static int igt_handle_error(void *arg)
 	/* Temporarily disable error capture */
 	error = xchg(&i915->gpu_error.first_error, (void *)-1);
 
-	i915_handle_error(i915, ENGINE_MASK(engine->id), 0, NULL);
+	i915_handle_error(i915, engine->mask, 0, NULL);
 
 	xchg(&i915->gpu_error.first_error, error);
 
@@ -1547,7 +1784,7 @@ static int igt_atomic_reset_engine(struct intel_engine_cs *engine,
 			i915_request_wait(rq,
 					  I915_WAIT_LOCKED,
 					  MAX_SCHEDULE_TIMEOUT);
-		if (i915_terminally_wedged(&i915->gpu_error))
+		if (i915_reset_failed(i915))
 			err = -EIO;
 	}
 
@@ -1586,7 +1823,7 @@ static int igt_atomic_reset(void *arg)
 
 	/* Flush any requests before we get started and check basics */
 	force_reset(i915);
-	if (i915_terminally_wedged(&i915->gpu_error))
+	if (i915_reset_failed(i915))
 		goto unlock;
 
 	if (intel_has_gpu_reset(i915)) {
@@ -1642,6 +1879,8 @@ int intel_hangcheck_live_selftests(struct drm_i915_private *i915)
 		SUBTEST(igt_global_reset), /* attempt to recover GPU first */
 		SUBTEST(igt_wedged_reset),
 		SUBTEST(igt_hang_sanitycheck),
+		SUBTEST(igt_reset_nop),
+		SUBTEST(igt_reset_nop_engine),
 		SUBTEST(igt_reset_idle_engine),
 		SUBTEST(igt_reset_active_engine),
 		SUBTEST(igt_reset_engines),
@@ -1660,7 +1899,7 @@ int intel_hangcheck_live_selftests(struct drm_i915_private *i915)
 	if (!intel_has_gpu_reset(i915))
 		return 0;
 
-	if (i915_terminally_wedged(&i915->gpu_error))
+	if (i915_terminally_wedged(i915))
 		return -EIO; /* we're long past hope of a successful reset */
 
 	wakeref = intel_runtime_pm_get(i915);
diff --git a/drivers/gpu/drm/i915/selftests/intel_lrc.c b/drivers/gpu/drm/i915/selftests/intel_lrc.c
index 58144e024751..fbee030db940 100644
--- a/drivers/gpu/drm/i915/selftests/intel_lrc.c
+++ b/drivers/gpu/drm/i915/selftests/intel_lrc.c
@@ -10,6 +10,7 @@
 
 #include "../i915_selftest.h"
 #include "igt_flush_test.h"
+#include "igt_live_test.h"
 #include "igt_spinner.h"
 #include "i915_random.h"
 
@@ -75,6 +76,185 @@ err_unlock:
 	return err;
 }
 
+static int live_busywait_preempt(void *arg)
+{
+	struct drm_i915_private *i915 = arg;
+	struct i915_gem_context *ctx_hi, *ctx_lo;
+	struct intel_engine_cs *engine;
+	struct drm_i915_gem_object *obj;
+	struct i915_vma *vma;
+	enum intel_engine_id id;
+	intel_wakeref_t wakeref;
+	int err = -ENOMEM;
+	u32 *map;
+
+	/*
+	 * Verify that even without HAS_LOGICAL_RING_PREEMPTION, we can
+	 * preempt the busywaits used to synchronise between rings.
+	 */
+
+	mutex_lock(&i915->drm.struct_mutex);
+	wakeref = intel_runtime_pm_get(i915);
+
+	ctx_hi = kernel_context(i915);
+	if (!ctx_hi)
+		goto err_unlock;
+	ctx_hi->sched.priority = INT_MAX;
+
+	ctx_lo = kernel_context(i915);
+	if (!ctx_lo)
+		goto err_ctx_hi;
+	ctx_lo->sched.priority = INT_MIN;
+
+	obj = i915_gem_object_create_internal(i915, PAGE_SIZE);
+	if (IS_ERR(obj)) {
+		err = PTR_ERR(obj);
+		goto err_ctx_lo;
+	}
+
+	map = i915_gem_object_pin_map(obj, I915_MAP_WC);
+	if (IS_ERR(map)) {
+		err = PTR_ERR(map);
+		goto err_obj;
+	}
+
+	vma = i915_vma_instance(obj, &i915->ggtt.vm, NULL);
+	if (IS_ERR(vma)) {
+		err = PTR_ERR(vma);
+		goto err_map;
+	}
+
+	err = i915_vma_pin(vma, 0, 0, PIN_GLOBAL);
+	if (err)
+		goto err_map;
+
+	for_each_engine(engine, i915, id) {
+		struct i915_request *lo, *hi;
+		struct igt_live_test t;
+		u32 *cs;
+
+		if (!intel_engine_can_store_dword(engine))
+			continue;
+
+		if (igt_live_test_begin(&t, i915, __func__, engine->name)) {
+			err = -EIO;
+			goto err_vma;
+		}
+
+		/*
+		 * We create two requests. The low priority request
+		 * busywaits on a semaphore (inside the ringbuffer where
+		 * is should be preemptible) and the high priority requests
+		 * uses a MI_STORE_DWORD_IMM to update the semaphore value
+		 * allowing the first request to complete. If preemption
+		 * fails, we hang instead.
+		 */
+
+		lo = i915_request_alloc(engine, ctx_lo);
+		if (IS_ERR(lo)) {
+			err = PTR_ERR(lo);
+			goto err_vma;
+		}
+
+		cs = intel_ring_begin(lo, 8);
+		if (IS_ERR(cs)) {
+			err = PTR_ERR(cs);
+			i915_request_add(lo);
+			goto err_vma;
+		}
+
+		*cs++ = MI_STORE_DWORD_IMM_GEN4 | MI_USE_GGTT;
+		*cs++ = i915_ggtt_offset(vma);
+		*cs++ = 0;
+		*cs++ = 1;
+
+		/* XXX Do we need a flush + invalidate here? */
+
+		*cs++ = MI_SEMAPHORE_WAIT |
+			MI_SEMAPHORE_GLOBAL_GTT |
+			MI_SEMAPHORE_POLL |
+			MI_SEMAPHORE_SAD_EQ_SDD;
+		*cs++ = 0;
+		*cs++ = i915_ggtt_offset(vma);
+		*cs++ = 0;
+
+		intel_ring_advance(lo, cs);
+		i915_request_add(lo);
+
+		if (wait_for(READ_ONCE(*map), 10)) {
+			err = -ETIMEDOUT;
+			goto err_vma;
+		}
+
+		/* Low priority request should be busywaiting now */
+		if (i915_request_wait(lo, I915_WAIT_LOCKED, 1) != -ETIME) {
+			pr_err("%s: Busywaiting request did not!\n",
+			       engine->name);
+			err = -EIO;
+			goto err_vma;
+		}
+
+		hi = i915_request_alloc(engine, ctx_hi);
+		if (IS_ERR(hi)) {
+			err = PTR_ERR(hi);
+			goto err_vma;
+		}
+
+		cs = intel_ring_begin(hi, 4);
+		if (IS_ERR(cs)) {
+			err = PTR_ERR(cs);
+			i915_request_add(hi);
+			goto err_vma;
+		}
+
+		*cs++ = MI_STORE_DWORD_IMM_GEN4 | MI_USE_GGTT;
+		*cs++ = i915_ggtt_offset(vma);
+		*cs++ = 0;
+		*cs++ = 0;
+
+		intel_ring_advance(hi, cs);
+		i915_request_add(hi);
+
+		if (i915_request_wait(lo, I915_WAIT_LOCKED, HZ / 5) < 0) {
+			struct drm_printer p = drm_info_printer(i915->drm.dev);
+
+			pr_err("%s: Failed to preempt semaphore busywait!\n",
+			       engine->name);
+
+			intel_engine_dump(engine, &p, "%s\n", engine->name);
+			GEM_TRACE_DUMP();
+
+			i915_gem_set_wedged(i915);
+			err = -EIO;
+			goto err_vma;
+		}
+		GEM_BUG_ON(READ_ONCE(*map));
+
+		if (igt_live_test_end(&t)) {
+			err = -EIO;
+			goto err_vma;
+		}
+	}
+
+	err = 0;
+err_vma:
+	i915_vma_unpin(vma);
+err_map:
+	i915_gem_object_unpin_map(obj);
+err_obj:
+	i915_gem_object_put(obj);
+err_ctx_lo:
+	kernel_context_close(ctx_lo);
+err_ctx_hi:
+	kernel_context_close(ctx_hi);
+err_unlock:
+	if (igt_flush_test(i915, I915_WAIT_LOCKED))
+		err = -EIO;
+	intel_runtime_pm_put(i915, wakeref);
+	mutex_unlock(&i915->drm.struct_mutex);
+	return err;
+}
+
 static int live_preempt(void *arg)
 {
 	struct drm_i915_private *i915 = arg;
@@ -88,6 +268,9 @@ static int live_preempt(void *arg)
 	if (!HAS_LOGICAL_RING_PREEMPTION(i915))
 		return 0;
 
+	if (!(i915->caps.scheduler & I915_SCHEDULER_CAP_PREEMPTION))
+		pr_err("Logical preemption supported, but not exposed\n");
+
 	mutex_lock(&i915->drm.struct_mutex);
 	wakeref = intel_runtime_pm_get(i915);
 
@@ -110,8 +293,17 @@ static int live_preempt(void *arg)
 		I915_USER_PRIORITY(I915_CONTEXT_MIN_USER_PRIORITY);
 
 	for_each_engine(engine, i915, id) {
+		struct igt_live_test t;
 		struct i915_request *rq;
 
+		if (!intel_engine_has_preemption(engine))
+			continue;
+
+		if (igt_live_test_begin(&t, i915, __func__, engine->name)) {
+			err = -EIO;
+			goto err_ctx_lo;
+		}
+
 		rq = igt_spinner_create_request(&spin_lo, ctx_lo, engine,
 						MI_ARB_CHECK);
 		if (IS_ERR(rq)) {
@@ -147,7 +339,8 @@ static int live_preempt(void *arg)
 
 		igt_spinner_end(&spin_hi);
 		igt_spinner_end(&spin_lo);
-		if (igt_flush_test(i915, I915_WAIT_LOCKED)) {
+
+		if (igt_live_test_end(&t)) {
 			err = -EIO;
 			goto err_ctx_lo;
 		}
@@ -201,8 +394,17 @@ static int live_late_preempt(void *arg)
 		goto err_ctx_hi;
 
 	for_each_engine(engine, i915, id) {
+		struct igt_live_test t;
 		struct i915_request *rq;
 
+		if (!intel_engine_has_preemption(engine))
+			continue;
+
+		if (igt_live_test_begin(&t, i915, __func__, engine->name)) {
+			err = -EIO;
+			goto err_ctx_lo;
+		}
+
 		rq = igt_spinner_create_request(&spin_lo, ctx_lo, engine,
 						MI_ARB_CHECK);
 		if (IS_ERR(rq)) {
@@ -241,7 +443,8 @@ static int live_late_preempt(void *arg)
 
 		igt_spinner_end(&spin_hi);
 		igt_spinner_end(&spin_lo);
-		if (igt_flush_test(i915, I915_WAIT_LOCKED)) {
+
+		if (igt_live_test_end(&t)) {
 			err = -EIO;
 			goto err_ctx_lo;
 		}
@@ -335,6 +538,9 @@ static int live_suppress_self_preempt(void *arg)
 		struct i915_request *rq_a, *rq_b;
 		int depth;
 
+		if (!intel_engine_has_preemption(engine))
+			continue;
+
 		engine->execlists.preempt_hang.count = 0;
 
 		rq_a = igt_spinner_create_request(&a.spin,
@@ -407,6 +613,171 @@ err_wedged:
 	goto err_client_b;
 }
 
+static int __i915_sw_fence_call
+dummy_notify(struct i915_sw_fence *fence, enum i915_sw_fence_notify state)
+{
+	return NOTIFY_DONE;
+}
+
+static struct i915_request *dummy_request(struct intel_engine_cs *engine)
+{
+	struct i915_request *rq;
+
+	rq = kzalloc(sizeof(*rq), GFP_KERNEL);
+	if (!rq)
+		return NULL;
+
+	INIT_LIST_HEAD(&rq->active_list);
+	rq->engine = engine;
+
+	i915_sched_node_init(&rq->sched);
+
+	/* mark this request as permanently incomplete */
+	rq->fence.seqno = 1;
+	BUILD_BUG_ON(sizeof(rq->fence.seqno) != 8); /* upper 32b == 0 */
+	rq->hwsp_seqno = (u32 *)&rq->fence.seqno + 1;
+	GEM_BUG_ON(i915_request_completed(rq));
+
+	i915_sw_fence_init(&rq->submit, dummy_notify);
+	i915_sw_fence_commit(&rq->submit);
+
+	return rq;
+}
+
+static void dummy_request_free(struct i915_request *dummy)
+{
+	i915_request_mark_complete(dummy);
+	i915_sched_node_fini(&dummy->sched);
+	i915_sw_fence_fini(&dummy->submit);
+
+	dma_fence_free(&dummy->fence);
+}
+
+static int live_suppress_wait_preempt(void *arg)
+{
+	struct drm_i915_private *i915 = arg;
+	struct preempt_client client[4];
+	struct intel_engine_cs *engine;
+	enum intel_engine_id id;
+	intel_wakeref_t wakeref;
+	int err = -ENOMEM;
+	int i;
+
+	/*
+	 * Waiters are given a little priority nudge, but not enough
+	 * to actually cause any preemption. Double check that we do
+	 * not needlessly generate preempt-to-idle cycles.
+	 */
+
+	if (!HAS_LOGICAL_RING_PREEMPTION(i915))
+		return 0;
+
+	mutex_lock(&i915->drm.struct_mutex);
+	wakeref = intel_runtime_pm_get(i915);
+
+	if (preempt_client_init(i915, &client[0])) /* ELSP[0] */
+		goto err_unlock;
+	if (preempt_client_init(i915, &client[1])) /* ELSP[1] */
+		goto err_client_0;
+	if (preempt_client_init(i915, &client[2])) /* head of queue */
+		goto err_client_1;
+	if (preempt_client_init(i915, &client[3])) /* bystander */
+		goto err_client_2;
+
+	for_each_engine(engine, i915, id) {
+		int depth;
+
+		if (!intel_engine_has_preemption(engine))
+			continue;
+
+		if (!engine->emit_init_breadcrumb)
+			continue;
+
+		for (depth = 0; depth < ARRAY_SIZE(client); depth++) {
+			struct i915_request *rq[ARRAY_SIZE(client)];
+			struct i915_request *dummy;
+
+			engine->execlists.preempt_hang.count = 0;
+
+			dummy = dummy_request(engine);
+			if (!dummy)
+				goto err_client_3;
+
+			for (i = 0; i < ARRAY_SIZE(client); i++) {
+				rq[i] = igt_spinner_create_request(&client[i].spin,
+								   client[i].ctx, engine,
+								   MI_NOOP);
+				if (IS_ERR(rq[i])) {
+					err = PTR_ERR(rq[i]);
+					goto err_wedged;
+				}
+
+				/* Disable NEWCLIENT promotion */
+				__i915_active_request_set(&rq[i]->timeline->last_request,
+							  dummy);
+				i915_request_add(rq[i]);
+			}
+
+			dummy_request_free(dummy);
+
+			GEM_BUG_ON(i915_request_completed(rq[0]));
+			if (!igt_wait_for_spinner(&client[0].spin, rq[0])) {
+				pr_err("%s: First client failed to start\n",
+				       engine->name);
+				goto err_wedged;
+			}
+			GEM_BUG_ON(!i915_request_started(rq[0]));
+
+			if (i915_request_wait(rq[depth],
+					      I915_WAIT_LOCKED |
+					      I915_WAIT_PRIORITY,
+					      1) != -ETIME) {
+				pr_err("%s: Waiter depth:%d completed!\n",
+				       engine->name, depth);
+				goto err_wedged;
+			}
+
+			for (i = 0; i < ARRAY_SIZE(client); i++)
+				igt_spinner_end(&client[i].spin);
+
+			if (igt_flush_test(i915, I915_WAIT_LOCKED))
+				goto err_wedged;
+
+			if (engine->execlists.preempt_hang.count) {
+				pr_err("%s: Preemption recorded x%d, depth %d; should have been suppressed!\n",
+				       engine->name,
+				       engine->execlists.preempt_hang.count,
+				       depth);
+				err = -EINVAL;
+				goto err_client_3;
+			}
+		}
+	}
+
+	err = 0;
+err_client_3:
+	preempt_client_fini(&client[3]);
+err_client_2:
+	preempt_client_fini(&client[2]);
+err_client_1:
+	preempt_client_fini(&client[1]);
+err_client_0:
+	preempt_client_fini(&client[0]);
+err_unlock:
+	if (igt_flush_test(i915, I915_WAIT_LOCKED))
+		err = -EIO;
+	intel_runtime_pm_put(i915, wakeref);
+	mutex_unlock(&i915->drm.struct_mutex);
+	return err;
+
+err_wedged:
+	for (i = 0; i < ARRAY_SIZE(client); i++)
+		igt_spinner_end(&client[i].spin);
+	i915_gem_set_wedged(i915);
+	err = -EIO;
+	goto err_client_3;
+}
+
 static int live_chain_preempt(void *arg)
 {
 	struct drm_i915_private *i915 = arg;
@@ -438,11 +809,39 @@ static int live_chain_preempt(void *arg)
 		struct i915_sched_attr attr = {
 			.priority = I915_USER_PRIORITY(I915_PRIORITY_MAX),
 		};
-		int count, i;
+		struct igt_live_test t;
+		struct i915_request *rq;
+		int ring_size, count, i;
 
-		for_each_prime_number_from(count, 1, 32) { /* must fit ring! */
-			struct i915_request *rq;
+		if (!intel_engine_has_preemption(engine))
+			continue;
 
+		rq = igt_spinner_create_request(&lo.spin,
+						lo.ctx, engine,
+						MI_ARB_CHECK);
+		if (IS_ERR(rq))
+			goto err_wedged;
+		i915_request_add(rq);
+
+		ring_size = rq->wa_tail - rq->head;
+		if (ring_size < 0)
+			ring_size += rq->ring->size;
+		ring_size = rq->ring->size / ring_size;
+		pr_debug("%s(%s): Using maximum of %d requests\n",
+			 __func__, engine->name, ring_size);
+
+		igt_spinner_end(&lo.spin);
+		if (i915_request_wait(rq, I915_WAIT_LOCKED, HZ / 2) < 0) {
+			pr_err("Timed out waiting to flush %s\n", engine->name);
+			goto err_wedged;
+		}
+
+		if (igt_live_test_begin(&t, i915, __func__, engine->name)) {
+			err = -EIO;
+			goto err_wedged;
+		}
+
+		for_each_prime_number_from(count, 1, ring_size) {
 			rq = igt_spinner_create_request(&hi.spin,
 							hi.ctx, engine,
 							MI_ARB_CHECK);
@@ -484,6 +883,26 @@ static int live_chain_preempt(void *arg)
 				goto err_wedged;
 			}
 			igt_spinner_end(&lo.spin);
+
+			rq = i915_request_alloc(engine, lo.ctx);
+			if (IS_ERR(rq))
+				goto err_wedged;
+			i915_request_add(rq);
+			if (i915_request_wait(rq, I915_WAIT_LOCKED, HZ / 5) < 0) {
+				struct drm_printer p =
+					drm_info_printer(i915->drm.dev);
+
+				pr_err("Failed to flush low priority chain of %d requests\n",
+				       count);
+				intel_engine_dump(engine, &p,
+						  "%s\n", engine->name);
+				goto err_wedged;
+			}
+		}
+
+		if (igt_live_test_end(&t)) {
+			err = -EIO;
+			goto err_wedged;
 		}
 	}
 
@@ -767,7 +1186,7 @@ static int smoke_crescendo(struct preempt_smoke *smoke, unsigned int flags)
 
 	pr_info("Submitted %lu crescendo:%x requests across %d engines and %d contexts\n",
 		count, flags,
-		RUNTIME_INFO(smoke->i915)->num_rings, smoke->ncontext);
+		RUNTIME_INFO(smoke->i915)->num_engines, smoke->ncontext);
 	return 0;
 }
 
@@ -795,7 +1214,7 @@ static int smoke_random(struct preempt_smoke *smoke, unsigned int flags)
 
 	pr_info("Submitted %lu random:%x requests across %d engines and %d contexts\n",
 		count, flags,
-		RUNTIME_INFO(smoke->i915)->num_rings, smoke->ncontext);
+		RUNTIME_INFO(smoke->i915)->num_engines, smoke->ncontext);
 	return 0;
 }
 
@@ -808,6 +1227,7 @@ static int live_preempt_smoke(void *arg)
 	};
 	const unsigned int phase[] = { 0, BATCH };
 	intel_wakeref_t wakeref;
+	struct igt_live_test t;
 	int err = -ENOMEM;
 	u32 *cs;
 	int n;
@@ -838,11 +1258,13 @@ static int live_preempt_smoke(void *arg)
 	for (n = 0; n < PAGE_SIZE / sizeof(*cs) - 1; n++)
 		cs[n] = MI_ARB_CHECK;
 	cs[n] = MI_BATCH_BUFFER_END;
+	i915_gem_object_flush_map(smoke.batch);
 	i915_gem_object_unpin_map(smoke.batch);
 
-	err = i915_gem_object_set_to_gtt_domain(smoke.batch, false);
-	if (err)
+	if (igt_live_test_begin(&t, smoke.i915, __func__, "all")) {
+		err = -EIO;
 		goto err_batch;
+	}
 
 	for (n = 0; n < smoke.ncontext; n++) {
 		smoke.contexts[n] = kernel_context(smoke.i915);
@@ -861,7 +1283,7 @@ static int live_preempt_smoke(void *arg)
 	}
 
 err_ctx:
-	if (igt_flush_test(smoke.i915, I915_WAIT_LOCKED))
+	if (igt_live_test_end(&t))
 		err = -EIO;
 
 	for (n = 0; n < smoke.ncontext; n++) {
@@ -884,9 +1306,11 @@ int intel_execlists_live_selftests(struct drm_i915_private *i915)
 {
 	static const struct i915_subtest tests[] = {
 		SUBTEST(live_sanitycheck),
+		SUBTEST(live_busywait_preempt),
 		SUBTEST(live_preempt),
 		SUBTEST(live_late_preempt),
 		SUBTEST(live_suppress_self_preempt),
+		SUBTEST(live_suppress_wait_preempt),
 		SUBTEST(live_chain_preempt),
 		SUBTEST(live_preempt_hang),
 		SUBTEST(live_preempt_smoke),
@@ -895,7 +1319,7 @@ int intel_execlists_live_selftests(struct drm_i915_private *i915)
 	if (!HAS_EXECLISTS(i915))
 		return 0;
 
-	if (i915_terminally_wedged(&i915->gpu_error))
+	if (i915_terminally_wedged(i915))
 		return 0;
 
 	return i915_subtests(tests, i915);
diff --git a/drivers/gpu/drm/i915/selftests/intel_uncore.c b/drivers/gpu/drm/i915/selftests/intel_uncore.c
index 81d9d31042a9..e0d7ebecb215 100644
--- a/drivers/gpu/drm/i915/selftests/intel_uncore.c
+++ b/drivers/gpu/drm/i915/selftests/intel_uncore.c
@@ -119,9 +119,143 @@ int intel_uncore_mock_selftests(void)
 	return 0;
 }
 
-static int intel_uncore_check_forcewake_domains(struct drm_i915_private *dev_priv)
+static int live_forcewake_ops(void *arg)
+{
+	static const struct reg {
+		const char *name;
+		unsigned long platforms;
+		unsigned int offset;
+	} registers[] = {
+		{
+			"RING_START",
+			INTEL_GEN_MASK(6, 7),
+			0x38,
+		},
+		{
+			"RING_MI_MODE",
+			INTEL_GEN_MASK(8, BITS_PER_LONG),
+			0x9c,
+		}
+	};
+	const struct reg *r;
+	struct drm_i915_private *i915 = arg;
+	struct intel_uncore_forcewake_domain *domain;
+	struct intel_uncore *uncore = &i915->uncore;
+	struct intel_engine_cs *engine;
+	enum intel_engine_id id;
+	intel_wakeref_t wakeref;
+	unsigned int tmp;
+	int err = 0;
+
+	GEM_BUG_ON(i915->gt.awake);
+
+	/* vlv/chv with their pcu behave differently wrt reads */
+	if (IS_VALLEYVIEW(i915) || IS_CHERRYVIEW(i915)) {
+		pr_debug("PCU fakes forcewake badly; skipping\n");
+		return 0;
+	}
+
+	/*
+	 * Not quite as reliable across the gen as one would hope.
+	 *
+	 * Either our theory of operation is incorrect, or there remain
+	 * external parties interfering with the powerwells.
+	 *
+	 * https://bugs.freedesktop.org/show_bug.cgi?id=110210
+	 */
+	if (!IS_ENABLED(CONFIG_DRM_I915_SELFTEST_BROKEN))
+		return 0;
+
+	/* We have to pick carefully to get the exact behaviour we need */
+	for (r = registers; r->name; r++)
+		if (r->platforms & INTEL_INFO(i915)->gen_mask)
+			break;
+	if (!r->name) {
+		pr_debug("Forcewaked register not known for %s; skipping\n",
+			 intel_platform_name(INTEL_INFO(i915)->platform));
+		return 0;
+	}
+
+	wakeref = intel_runtime_pm_get(i915);
+
+	for_each_fw_domain(domain, uncore, tmp) {
+		smp_store_mb(domain->active, false);
+		if (!hrtimer_cancel(&domain->timer))
+			continue;
+
+		intel_uncore_fw_release_timer(&domain->timer);
+	}
+
+	for_each_engine(engine, i915, id) {
+		i915_reg_t mmio = _MMIO(engine->mmio_base + r->offset);
+		u32 __iomem *reg = uncore->regs + engine->mmio_base + r->offset;
+		enum forcewake_domains fw_domains;
+		u32 val;
+
+		if (!engine->default_state)
+			continue;
+
+		fw_domains = intel_uncore_forcewake_for_reg(uncore, mmio,
+							    FW_REG_READ);
+		if (!fw_domains)
+			continue;
+
+		for_each_fw_domain_masked(domain, fw_domains, uncore, tmp) {
+			if (!domain->wake_count)
+				continue;
+
+			pr_err("fw_domain %s still active, aborting test!\n",
+			       intel_uncore_forcewake_domain_to_str(domain->id));
+			err = -EINVAL;
+			goto out_rpm;
+		}
+
+		intel_uncore_forcewake_get(uncore, fw_domains);
+		val = readl(reg);
+		intel_uncore_forcewake_put(uncore, fw_domains);
+
+		/* Flush the forcewake release (delayed onto a timer) */
+		for_each_fw_domain_masked(domain, fw_domains, uncore, tmp) {
+			smp_store_mb(domain->active, false);
+			if (hrtimer_cancel(&domain->timer))
+				intel_uncore_fw_release_timer(&domain->timer);
+
+			preempt_disable();
+			err = wait_ack_clear(domain, FORCEWAKE_KERNEL);
+			preempt_enable();
+			if (err) {
+				pr_err("Failed to clear fw_domain %s\n",
+				       intel_uncore_forcewake_domain_to_str(domain->id));
+				goto out_rpm;
+			}
+		}
+
+		if (!val) {
+			pr_err("%s:%s was zero while fw was held!\n",
+			       engine->name, r->name);
+			err = -EINVAL;
+			goto out_rpm;
+		}
+
+		/* We then expect the read to return 0 outside of the fw */
+		if (wait_for(readl(reg) == 0, 100)) {
+			pr_err("%s:%s=%0x, fw_domains 0x%x still up after 100ms!\n",
+			       engine->name, r->name, readl(reg), fw_domains);
+			err = -ETIMEDOUT;
+			goto out_rpm;
+		}
+	}
+
+out_rpm:
+	intel_runtime_pm_put(i915, wakeref);
+	return err;
+}
+
+static int live_forcewake_domains(void *arg)
 {
 #define FW_RANGE 0x40000
+	struct drm_i915_private *dev_priv = arg;
+	struct intel_uncore *uncore = &dev_priv->uncore;
 	unsigned long *valid;
 	u32 offset;
 	int err;
@@ -137,48 +271,52 @@ static int intel_uncore_check_forcewake_domains(struct drm_i915_private *dev_pri
 	if (!IS_ENABLED(CONFIG_DRM_I915_SELFTEST_BROKEN))
 		return 0;
 
-	valid = kcalloc(BITS_TO_LONGS(FW_RANGE), sizeof(*valid),
-			GFP_KERNEL);
+	valid = bitmap_zalloc(FW_RANGE, GFP_KERNEL);
 	if (!valid)
 		return -ENOMEM;
 
-	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_get(uncore, FORCEWAKE_ALL);
 
-	check_for_unclaimed_mmio(dev_priv);
+	check_for_unclaimed_mmio(uncore);
 	for (offset = 0; offset < FW_RANGE; offset += 4) {
 		i915_reg_t reg = { offset };
 
 		(void)I915_READ_FW(reg);
-		if (!check_for_unclaimed_mmio(dev_priv))
+		if (!check_for_unclaimed_mmio(uncore))
 			set_bit(offset, valid);
 	}
 
-	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_uncore_forcewake_put(uncore, FORCEWAKE_ALL);
 
 	err = 0;
 	for_each_set_bit(offset, valid, FW_RANGE) {
 		i915_reg_t reg = { offset };
 
 		iosf_mbi_punit_acquire();
-		intel_uncore_forcewake_reset(dev_priv);
+		intel_uncore_forcewake_reset(uncore);
 		iosf_mbi_punit_release();
 
-		check_for_unclaimed_mmio(dev_priv);
+		check_for_unclaimed_mmio(uncore);
 
 		(void)I915_READ(reg);
-		if (check_for_unclaimed_mmio(dev_priv)) {
+		if (check_for_unclaimed_mmio(uncore)) {
 			pr_err("Unclaimed mmio read to register 0x%04x\n",
 			       offset);
 			err = -EINVAL;
 		}
 	}
 
-	kfree(valid);
+	bitmap_free(valid);
 	return err;
 }
 
 int intel_uncore_live_selftests(struct drm_i915_private *i915)
 {
+	static const struct i915_subtest tests[] = {
+		SUBTEST(live_forcewake_ops),
+		SUBTEST(live_forcewake_domains),
+	};
+
 	int err;
 
 	/* Confirm the table we load is still valid */
@@ -188,9 +326,5 @@ int intel_uncore_live_selftests(struct drm_i915_private *i915)
 	if (err)
 		return err;
 
-	err = intel_uncore_check_forcewake_domains(i915);
-	if (err)
-		return err;
-
-	return 0;
+	return i915_subtests(tests, i915);
 }
diff --git a/drivers/gpu/drm/i915/selftests/intel_workarounds.c b/drivers/gpu/drm/i915/selftests/intel_workarounds.c
index b15c4f26c593..567b6f8dae86 100644
--- a/drivers/gpu/drm/i915/selftests/intel_workarounds.c
+++ b/drivers/gpu/drm/i915/selftests/intel_workarounds.c
@@ -12,6 +12,14 @@
 #include "igt_spinner.h"
 #include "igt_wedge_me.h"
 #include "mock_context.h"
+#include "mock_drm.h"
+
+static const struct wo_register {
+	enum intel_platform platform;
+	u32 reg;
+} wo_registers[] = {
+	{ INTEL_GEMINILAKE, 0x731c }
+};
 
 #define REF_NAME_MAX (INTEL_ENGINE_CS_MAX_NAME + 4)
 struct wa_lists {
@@ -74,7 +82,7 @@ read_nonprivs(struct i915_gem_context *ctx, struct intel_engine_cs *engine)
 	if (IS_ERR(result))
 		return result;
 
-	i915_gem_object_set_cache_level(result, I915_CACHE_LLC);
+	i915_gem_object_set_cache_coherency(result, I915_CACHE_LLC);
 
 	cs = i915_gem_object_pin_map(result, I915_MAP_WB);
 	if (IS_ERR(cs)) {
@@ -82,6 +90,7 @@ read_nonprivs(struct i915_gem_context *ctx, struct intel_engine_cs *engine)
 		goto err_obj;
 	}
 	memset(cs, 0xc5, PAGE_SIZE);
+	i915_gem_object_flush_map(result);
 	i915_gem_object_unpin_map(result);
 
 	vma = i915_vma_instance(result, &engine->i915->ggtt.vm, NULL);
@@ -181,7 +190,7 @@ static int check_whitelist(struct i915_gem_context *ctx,
 	err = 0;
 	igt_wedge_on_timeout(&wedge, ctx->i915, HZ / 5) /* a safety net! */
 		err = i915_gem_object_set_to_cpu_domain(results, false);
-	if (i915_terminally_wedged(&ctx->i915->gpu_error))
+	if (i915_terminally_wedged(ctx->i915))
 		err = -EIO;
 	if (err)
 		goto out_put;
@@ -214,7 +223,7 @@ out_put:
 
 static int do_device_reset(struct intel_engine_cs *engine)
 {
-	i915_reset(engine->i915, ENGINE_MASK(engine->id), "live_workarounds");
+	i915_reset(engine->i915, engine->mask, "live_workarounds");
 	return 0;
 }
 
@@ -236,15 +245,11 @@ switch_to_scratch_context(struct intel_engine_cs *engine,
 	if (IS_ERR(ctx))
 		return PTR_ERR(ctx);
 
+	GEM_BUG_ON(i915_gem_context_is_bannable(ctx));
+
 	rq = ERR_PTR(-ENODEV);
-	with_intel_runtime_pm(engine->i915, wakeref) {
-		if (spin)
-			rq = igt_spinner_create_request(spin,
-							ctx, engine,
-							MI_NOOP);
-		else
-			rq = i915_request_alloc(engine, ctx);
-	}
+	with_intel_runtime_pm(engine->i915, wakeref)
+		rq = igt_spinner_create_request(spin, ctx, engine, MI_NOOP);
 
 	kernel_context_close(ctx);
 
@@ -273,7 +278,6 @@ static int check_whitelist_across_reset(struct intel_engine_cs *engine,
 					const char *name)
 {
 	struct drm_i915_private *i915 = engine->i915;
-	bool want_spin = reset == do_engine_reset;
 	struct i915_gem_context *ctx;
 	struct igt_spinner spin;
 	intel_wakeref_t wakeref;
@@ -282,11 +286,9 @@ static int check_whitelist_across_reset(struct intel_engine_cs *engine,
 	pr_info("Checking %d whitelisted registers (RING_NONPRIV) [%s]\n",
 		engine->whitelist.count, name);
 
-	if (want_spin) {
-		err = igt_spinner_init(&spin, i915);
-		if (err)
-			return err;
-	}
+	err = igt_spinner_init(&spin, i915);
+	if (err)
+		return err;
 
 	ctx = kernel_context(i915);
 	if (IS_ERR(ctx))
@@ -298,17 +300,15 @@ static int check_whitelist_across_reset(struct intel_engine_cs *engine,
 		goto out;
 	}
 
-	err = switch_to_scratch_context(engine, want_spin ? &spin : NULL);
+	err = switch_to_scratch_context(engine, &spin);
 	if (err)
 		goto out;
 
 	with_intel_runtime_pm(i915, wakeref)
 		err = reset(engine);
 
-	if (want_spin) {
-		igt_spinner_end(&spin);
-		igt_spinner_fini(&spin);
-	}
+	igt_spinner_end(&spin);
+	igt_spinner_fini(&spin);
 
 	if (err) {
 		pr_err("%s reset failed\n", name);
@@ -340,10 +340,379 @@ out:
 	return err;
 }
 
+static struct i915_vma *create_scratch(struct i915_gem_context *ctx)
+{
+	struct drm_i915_gem_object *obj;
+	struct i915_vma *vma;
+	void *ptr;
+	int err;
+
+	obj = i915_gem_object_create_internal(ctx->i915, PAGE_SIZE);
+	if (IS_ERR(obj))
+		return ERR_CAST(obj);
+
+	i915_gem_object_set_cache_coherency(obj, I915_CACHE_LLC);
+
+	ptr = i915_gem_object_pin_map(obj, I915_MAP_WB);
+	if (IS_ERR(ptr)) {
+		err = PTR_ERR(ptr);
+		goto err_obj;
+	}
+	memset(ptr, 0xc5, PAGE_SIZE);
+	i915_gem_object_flush_map(obj);
+	i915_gem_object_unpin_map(obj);
+
+	vma = i915_vma_instance(obj, &ctx->ppgtt->vm, NULL);
+	if (IS_ERR(vma)) {
+		err = PTR_ERR(vma);
+		goto err_obj;
+	}
+
+	err = i915_vma_pin(vma, 0, 0, PIN_USER);
+	if (err)
+		goto err_obj;
+
+	err = i915_gem_object_set_to_cpu_domain(obj, false);
+	if (err)
+		goto err_obj;
+
+	return vma;
+
+err_obj:
+	i915_gem_object_put(obj);
+	return ERR_PTR(err);
+}
+
+static struct i915_vma *create_batch(struct i915_gem_context *ctx)
+{
+	struct drm_i915_gem_object *obj;
+	struct i915_vma *vma;
+	int err;
+
+	obj = i915_gem_object_create_internal(ctx->i915, 16 * PAGE_SIZE);
+	if (IS_ERR(obj))
+		return ERR_CAST(obj);
+
+	vma = i915_vma_instance(obj, &ctx->ppgtt->vm, NULL);
+	if (IS_ERR(vma)) {
+		err = PTR_ERR(vma);
+		goto err_obj;
+	}
+
+	err = i915_vma_pin(vma, 0, 0, PIN_USER);
+	if (err)
+		goto err_obj;
+
+	err = i915_gem_object_set_to_wc_domain(obj, true);
+	if (err)
+		goto err_obj;
+
+	return vma;
+
+err_obj:
+	i915_gem_object_put(obj);
+	return ERR_PTR(err);
+}
+
+static u32 reg_write(u32 old, u32 new, u32 rsvd)
+{
+	if (rsvd == 0x0000ffff) {
+		old &= ~(new >> 16);
+		old |= new & (new >> 16);
+	} else {
+		old &= ~rsvd;
+		old |= new & rsvd;
+	}
+
+	return old;
+}
+
+static bool wo_register(struct intel_engine_cs *engine, u32 reg)
+{
+	enum intel_platform platform = INTEL_INFO(engine->i915)->platform;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(wo_registers); i++) {
+		if (wo_registers[i].platform == platform &&
+		    wo_registers[i].reg == reg)
+			return true;
+	}
+
+	return false;
+}
+
+static int check_dirty_whitelist(struct i915_gem_context *ctx,
+				 struct intel_engine_cs *engine)
+{
+	const u32 values[] = {
+		0x00000000,
+		0x01010101,
+		0x10100101,
+		0x03030303,
+		0x30300303,
+		0x05050505,
+		0x50500505,
+		0x0f0f0f0f,
+		0xf00ff00f,
+		0x10101010,
+		0xf0f01010,
+		0x30303030,
+		0xa0a03030,
+		0x50505050,
+		0xc0c05050,
+		0xf0f0f0f0,
+		0x11111111,
+		0x33333333,
+		0x55555555,
+		0x0000ffff,
+		0x00ff00ff,
+		0xff0000ff,
+		0xffff00ff,
+		0xffffffff,
+	};
+	struct i915_vma *scratch;
+	struct i915_vma *batch;
+	int err = 0, i, v;
+	u32 *cs, *results;
+
+	scratch = create_scratch(ctx);
+	if (IS_ERR(scratch))
+		return PTR_ERR(scratch);
+
+	batch = create_batch(ctx);
+	if (IS_ERR(batch)) {
+		err = PTR_ERR(batch);
+		goto out_scratch;
+	}
+
+	for (i = 0; i < engine->whitelist.count; i++) {
+		u32 reg = i915_mmio_reg_offset(engine->whitelist.list[i].reg);
+		u64 addr = scratch->node.start;
+		struct i915_request *rq;
+		u32 srm, lrm, rsvd;
+		u32 expect;
+		int idx;
+
+		if (wo_register(engine, reg))
+			continue;
+
+		srm = MI_STORE_REGISTER_MEM;
+		lrm = MI_LOAD_REGISTER_MEM;
+		if (INTEL_GEN(ctx->i915) >= 8)
+			lrm++, srm++;
+
+		pr_debug("%s: Writing garbage to %x\n",
+			 engine->name, reg);
+
+		cs = i915_gem_object_pin_map(batch->obj, I915_MAP_WC);
+		if (IS_ERR(cs)) {
+			err = PTR_ERR(cs);
+			goto out_batch;
+		}
+
+		/* SRM original */
+		*cs++ = srm;
+		*cs++ = reg;
+		*cs++ = lower_32_bits(addr);
+		*cs++ = upper_32_bits(addr);
+
+		idx = 1;
+		for (v = 0; v < ARRAY_SIZE(values); v++) {
+			/* LRI garbage */
+			*cs++ = MI_LOAD_REGISTER_IMM(1);
+			*cs++ = reg;
+			*cs++ = values[v];
+
+			/* SRM result */
+			*cs++ = srm;
+			*cs++ = reg;
+			*cs++ = lower_32_bits(addr + sizeof(u32) * idx);
+			*cs++ = upper_32_bits(addr + sizeof(u32) * idx);
+			idx++;
+		}
+		for (v = 0; v < ARRAY_SIZE(values); v++) {
+			/* LRI garbage */
+			*cs++ = MI_LOAD_REGISTER_IMM(1);
+			*cs++ = reg;
+			*cs++ = ~values[v];
+
+			/* SRM result */
+			*cs++ = srm;
+			*cs++ = reg;
+			*cs++ = lower_32_bits(addr + sizeof(u32) * idx);
+			*cs++ = upper_32_bits(addr + sizeof(u32) * idx);
+			idx++;
+		}
+		GEM_BUG_ON(idx * sizeof(u32) > scratch->size);
+
+		/* LRM original -- don't leave garbage in the context! */
+		*cs++ = lrm;
+		*cs++ = reg;
+		*cs++ = lower_32_bits(addr);
+		*cs++ = upper_32_bits(addr);
+
+		*cs++ = MI_BATCH_BUFFER_END;
+
+		i915_gem_object_flush_map(batch->obj);
+		i915_gem_object_unpin_map(batch->obj);
+		i915_gem_chipset_flush(ctx->i915);
+
+		rq = i915_request_alloc(engine, ctx);
+		if (IS_ERR(rq)) {
+			err = PTR_ERR(rq);
+			goto out_batch;
+		}
+
+		if (engine->emit_init_breadcrumb) { /* Be nice if we hang */
+			err = engine->emit_init_breadcrumb(rq);
+			if (err)
+				goto err_request;
+		}
+
+		err = engine->emit_bb_start(rq,
+					    batch->node.start, PAGE_SIZE,
+					    0);
+		if (err)
+			goto err_request;
+
+err_request:
+		i915_request_add(rq);
+		if (err)
+			goto out_batch;
+
+		if (i915_request_wait(rq, I915_WAIT_LOCKED, HZ / 5) < 0) {
+			pr_err("%s: Futzing %x timedout; cancelling test\n",
+			       engine->name, reg);
+			i915_gem_set_wedged(ctx->i915);
+			err = -EIO;
+			goto out_batch;
+		}
+
+		results = i915_gem_object_pin_map(scratch->obj, I915_MAP_WB);
+		if (IS_ERR(results)) {
+			err = PTR_ERR(results);
+			goto out_batch;
+		}
+
+		GEM_BUG_ON(values[ARRAY_SIZE(values) - 1] != 0xffffffff);
+		rsvd = results[ARRAY_SIZE(values)]; /* detect write masking */
+		if (!rsvd) {
+			pr_err("%s: Unable to write to whitelisted register %x\n",
+			       engine->name, reg);
+			err = -EINVAL;
+			goto out_unpin;
+		}
+
+		expect = results[0];
+		idx = 1;
+		for (v = 0; v < ARRAY_SIZE(values); v++) {
+			expect = reg_write(expect, values[v], rsvd);
+			if (results[idx] != expect)
+				err++;
+			idx++;
+		}
+		for (v = 0; v < ARRAY_SIZE(values); v++) {
+			expect = reg_write(expect, ~values[v], rsvd);
+			if (results[idx] != expect)
+				err++;
+			idx++;
+		}
+		if (err) {
+			pr_err("%s: %d mismatch between values written to whitelisted register [%x], and values read back!\n",
+			       engine->name, err, reg);
+
+			pr_info("%s: Whitelisted register: %x, original value %08x, rsvd %08x\n",
+				engine->name, reg, results[0], rsvd);
+
+			expect = results[0];
+			idx = 1;
+			for (v = 0; v < ARRAY_SIZE(values); v++) {
+				u32 w = values[v];
+
+				expect = reg_write(expect, w, rsvd);
+				pr_info("Wrote %08x, read %08x, expect %08x\n",
+					w, results[idx], expect);
+				idx++;
+			}
+			for (v = 0; v < ARRAY_SIZE(values); v++) {
+				u32 w = ~values[v];
+
+				expect = reg_write(expect, w, rsvd);
+				pr_info("Wrote %08x, read %08x, expect %08x\n",
+					w, results[idx], expect);
+				idx++;
+			}
+
+			err = -EINVAL;
+		}
+out_unpin:
+		i915_gem_object_unpin_map(scratch->obj);
+		if (err)
+			break;
+	}
+
+	if (igt_flush_test(ctx->i915, I915_WAIT_LOCKED))
+		err = -EIO;
+out_batch:
+	i915_vma_unpin_and_release(&batch, 0);
+out_scratch:
+	i915_vma_unpin_and_release(&scratch, 0);
+	return err;
+}
+
+static int live_dirty_whitelist(void *arg)
+{
+	struct drm_i915_private *i915 = arg;
+	struct intel_engine_cs *engine;
+	struct i915_gem_context *ctx;
+	enum intel_engine_id id;
+	intel_wakeref_t wakeref;
+	struct drm_file *file;
+	int err = 0;
+
+	/* Can the user write to the whitelisted registers? */
+
+	if (INTEL_GEN(i915) < 7) /* minimum requirement for LRI, SRM, LRM */
+		return 0;
+
+	wakeref = intel_runtime_pm_get(i915);
+
+	mutex_unlock(&i915->drm.struct_mutex);
+	file = mock_file(i915);
+	mutex_lock(&i915->drm.struct_mutex);
+	if (IS_ERR(file)) {
+		err = PTR_ERR(file);
+		goto out_rpm;
+	}
+
+	ctx = live_context(i915, file);
+	if (IS_ERR(ctx)) {
+		err = PTR_ERR(ctx);
+		goto out_file;
+	}
+
+	for_each_engine(engine, i915, id) {
+		if (engine->whitelist.count == 0)
+			continue;
+
+		err = check_dirty_whitelist(ctx, engine);
+		if (err)
+			goto out_file;
+	}
+
+out_file:
+	mutex_unlock(&i915->drm.struct_mutex);
+	mock_file_free(i915, file);
+	mutex_lock(&i915->drm.struct_mutex);
+out_rpm:
+	intel_runtime_pm_put(i915, wakeref);
+	return err;
+}
+
 static int live_reset_whitelist(void *arg)
 {
 	struct drm_i915_private *i915 = arg;
-	struct intel_engine_cs *engine = i915->engine[RCS];
+	struct intel_engine_cs *engine = i915->engine[RCS0];
 	int err = 0;
 
 	/* If we reset the gpu, we should not lose the RING_NONPRIV */
@@ -381,10 +750,11 @@ static bool verify_gt_engine_wa(struct drm_i915_private *i915,
 	enum intel_engine_id id;
 	bool ok = true;
 
-	ok &= wa_list_verify(i915, &lists->gt_wa_list, str);
+	ok &= wa_list_verify(&i915->uncore, &lists->gt_wa_list, str);
 
 	for_each_engine(engine, i915, id)
-		ok &= wa_list_verify(i915, &lists->engine[id].wa_list, str);
+		ok &= wa_list_verify(engine->uncore,
+				     &lists->engine[id].wa_list, str);
 
 	return ok;
 }
@@ -513,13 +883,14 @@ err:
 int intel_workarounds_live_selftests(struct drm_i915_private *i915)
 {
 	static const struct i915_subtest tests[] = {
+		SUBTEST(live_dirty_whitelist),
 		SUBTEST(live_reset_whitelist),
 		SUBTEST(live_gpu_reset_gt_engine_workarounds),
 		SUBTEST(live_engine_reset_gt_engine_workarounds),
 	};
 	int err;
 
-	if (i915_terminally_wedged(&i915->gpu_error))
+	if (i915_terminally_wedged(i915))
 		return 0;
 
 	mutex_lock(&i915->drm.struct_mutex);
diff --git a/drivers/gpu/drm/i915/selftests/mock_context.c b/drivers/gpu/drm/i915/selftests/mock_context.c
index b646cdcdd602..0426093bf1d9 100644
--- a/drivers/gpu/drm/i915/selftests/mock_context.c
+++ b/drivers/gpu/drm/i915/selftests/mock_context.c
@@ -30,7 +30,6 @@ mock_context(struct drm_i915_private *i915,
 	     const char *name)
 {
 	struct i915_gem_context *ctx;
-	unsigned int n;
 	int ret;
 
 	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
@@ -41,25 +40,31 @@ mock_context(struct drm_i915_private *i915,
 	INIT_LIST_HEAD(&ctx->link);
 	ctx->i915 = i915;
 
+	ctx->hw_contexts = RB_ROOT;
+	spin_lock_init(&ctx->hw_contexts_lock);
+
 	INIT_RADIX_TREE(&ctx->handles_vma, GFP_KERNEL);
 	INIT_LIST_HEAD(&ctx->handles_list);
 	INIT_LIST_HEAD(&ctx->hw_id_link);
-
-	for (n = 0; n < ARRAY_SIZE(ctx->__engine); n++)
-		intel_context_init(&ctx->__engine[n], ctx, i915->engine[n]);
+	INIT_LIST_HEAD(&ctx->active_engines);
+	mutex_init(&ctx->mutex);
 
 	ret = i915_gem_context_pin_hw_id(ctx);
 	if (ret < 0)
 		goto err_handles;
 
 	if (name) {
+		struct i915_hw_ppgtt *ppgtt;
+
 		ctx->name = kstrdup(name, GFP_KERNEL);
 		if (!ctx->name)
 			goto err_put;
 
-		ctx->ppgtt = mock_ppgtt(i915, name);
-		if (!ctx->ppgtt)
+		ppgtt = mock_ppgtt(i915, name);
+		if (!ppgtt)
 			goto err_put;
+
+		__set_ppgtt(ctx, ppgtt);
 	}
 
 	return ctx;
@@ -87,9 +92,24 @@ void mock_init_contexts(struct drm_i915_private *i915)
 struct i915_gem_context *
 live_context(struct drm_i915_private *i915, struct drm_file *file)
 {
+	struct i915_gem_context *ctx;
+	int err;
+
 	lockdep_assert_held(&i915->drm.struct_mutex);
 
-	return i915_gem_create_context(i915, file->driver_priv);
+	ctx = i915_gem_create_context(i915, 0);
+	if (IS_ERR(ctx))
+		return ctx;
+
+	err = gem_context_register(ctx, file->driver_priv);
+	if (err < 0)
+		goto err_ctx;
+
+	return ctx;
+
+err_ctx:
+	context_close(ctx);
+	return ERR_PTR(err);
 }
 
 struct i915_gem_context *
diff --git a/drivers/gpu/drm/i915/selftests/mock_engine.c b/drivers/gpu/drm/i915/selftests/mock_engine.c
index 08f0cab02e0f..61a8206ed677 100644
--- a/drivers/gpu/drm/i915/selftests/mock_engine.c
+++ b/drivers/gpu/drm/i915/selftests/mock_engine.c
@@ -50,13 +50,12 @@ static struct intel_ring *mock_ring(struct intel_engine_cs *engine)
 	if (!ring)
 		return NULL;
 
-	if (i915_timeline_init(engine->i915,
-			       &ring->timeline, engine->name,
-			       NULL)) {
+	if (i915_timeline_init(engine->i915, &ring->timeline, NULL)) {
 		kfree(ring);
 		return NULL;
 	}
 
+	kref_init(&ring->base.ref);
 	ring->base.size = sz;
 	ring->base.effective_size = sz;
 	ring->base.vaddr = (void *)(ring + 1);
@@ -76,28 +75,26 @@ static void mock_ring_free(struct intel_ring *base)
 	kfree(ring);
 }
 
-static struct mock_request *first_request(struct mock_engine *engine)
+static struct i915_request *first_request(struct mock_engine *engine)
 {
 	return list_first_entry_or_null(&engine->hw_queue,
-					struct mock_request,
-					link);
+					struct i915_request,
+					mock.link);
 }
 
-static void advance(struct mock_request *request)
+static void advance(struct i915_request *request)
 {
-	list_del_init(&request->link);
-	intel_engine_write_global_seqno(request->base.engine,
-					request->base.global_seqno);
-	i915_request_mark_complete(&request->base);
-	GEM_BUG_ON(!i915_request_completed(&request->base));
+	list_del_init(&request->mock.link);
+	i915_request_mark_complete(request);
+	GEM_BUG_ON(!i915_request_completed(request));
 
-	intel_engine_queue_breadcrumbs(request->base.engine);
+	intel_engine_queue_breadcrumbs(request->engine);
 }
 
 static void hw_delay_complete(struct timer_list *t)
 {
 	struct mock_engine *engine = from_timer(engine, t, hw_delay);
-	struct mock_request *request;
+	struct i915_request *request;
 	unsigned long flags;
 
 	spin_lock_irqsave(&engine->hw_lock, flags);
@@ -112,8 +109,9 @@ static void hw_delay_complete(struct timer_list *t)
 	 * requeue the timer for the next delayed request.
 	 */
 	while ((request = first_request(engine))) {
-		if (request->delay) {
-			mod_timer(&engine->hw_delay, jiffies + request->delay);
+		if (request->mock.delay) {
+			mod_timer(&engine->hw_delay,
+				  jiffies + request->mock.delay);
 			break;
 		}
 
@@ -126,55 +124,43 @@ static void hw_delay_complete(struct timer_list *t)
 static void mock_context_unpin(struct intel_context *ce)
 {
 	mock_timeline_unpin(ce->ring->timeline);
-	i915_gem_context_put(ce->gem_context);
 }
 
-static void mock_context_destroy(struct intel_context *ce)
+static void mock_context_destroy(struct kref *ref)
 {
-	GEM_BUG_ON(ce->pin_count);
+	struct intel_context *ce = container_of(ref, typeof(*ce), ref);
+
+	GEM_BUG_ON(intel_context_is_pinned(ce));
 
 	if (ce->ring)
 		mock_ring_free(ce->ring);
-}
 
-static const struct intel_context_ops mock_context_ops = {
-	.unpin = mock_context_unpin,
-	.destroy = mock_context_destroy,
-};
+	intel_context_free(ce);
+}
 
-static struct intel_context *
-mock_context_pin(struct intel_engine_cs *engine,
-		 struct i915_gem_context *ctx)
+static int mock_context_pin(struct intel_context *ce)
 {
-	struct intel_context *ce = to_intel_context(ctx, engine);
-	int err = -ENOMEM;
-
-	if (ce->pin_count++)
-		return ce;
-
 	if (!ce->ring) {
-		ce->ring = mock_ring(engine);
+		ce->ring = mock_ring(ce->engine);
 		if (!ce->ring)
-			goto err;
+			return -ENOMEM;
 	}
 
 	mock_timeline_pin(ce->ring->timeline);
+	return 0;
+}
 
-	ce->ops = &mock_context_ops;
-	i915_gem_context_get(ctx);
-	return ce;
+static const struct intel_context_ops mock_context_ops = {
+	.pin = mock_context_pin,
+	.unpin = mock_context_unpin,
 
-err:
-	ce->pin_count = 0;
-	return ERR_PTR(err);
-}
+	.destroy = mock_context_destroy,
+};
 
 static int mock_request_alloc(struct i915_request *request)
 {
-	struct mock_request *mock = container_of(request, typeof(*mock), base);
-
-	INIT_LIST_HEAD(&mock->link);
-	mock->delay = 0;
+	INIT_LIST_HEAD(&request->mock.link);
+	request->mock.delay = 0;
 
 	return 0;
 }
@@ -192,25 +178,55 @@ static u32 *mock_emit_breadcrumb(struct i915_request *request, u32 *cs)
 
 static void mock_submit_request(struct i915_request *request)
 {
-	struct mock_request *mock = container_of(request, typeof(*mock), base);
 	struct mock_engine *engine =
 		container_of(request->engine, typeof(*engine), base);
 	unsigned long flags;
 
 	i915_request_submit(request);
-	GEM_BUG_ON(!request->global_seqno);
 
 	spin_lock_irqsave(&engine->hw_lock, flags);
-	list_add_tail(&mock->link, &engine->hw_queue);
-	if (mock->link.prev == &engine->hw_queue) {
-		if (mock->delay)
-			mod_timer(&engine->hw_delay, jiffies + mock->delay);
+	list_add_tail(&request->mock.link, &engine->hw_queue);
+	if (list_is_first(&request->mock.link, &engine->hw_queue)) {
+		if (request->mock.delay)
+			mod_timer(&engine->hw_delay,
+				  jiffies + request->mock.delay);
 		else
-			advance(mock);
+			advance(request);
 	}
 	spin_unlock_irqrestore(&engine->hw_lock, flags);
 }
 
+static void mock_reset_prepare(struct intel_engine_cs *engine)
+{
+}
+
+static void mock_reset(struct intel_engine_cs *engine, bool stalled)
+{
+	GEM_BUG_ON(stalled);
+}
+
+static void mock_reset_finish(struct intel_engine_cs *engine)
+{
+}
+
+static void mock_cancel_requests(struct intel_engine_cs *engine)
+{
+	struct i915_request *request;
+	unsigned long flags;
+
+	spin_lock_irqsave(&engine->timeline.lock, flags);
+
+	/* Mark all submitted requests as skipped. */
+	list_for_each_entry(request, &engine->timeline.requests, sched.link) {
+		if (!i915_request_signaled(request))
+			dma_fence_set_error(&request->fence, -EIO);
+
+		i915_request_mark_complete(request);
+	}
+
+	spin_unlock_irqrestore(&engine->timeline.lock, flags);
+}
+
 struct intel_engine_cs *mock_engine(struct drm_i915_private *i915,
 				    const char *name,
 				    int id)
@@ -227,18 +243,21 @@ struct intel_engine_cs *mock_engine(struct drm_i915_private *i915,
 	engine->base.i915 = i915;
 	snprintf(engine->base.name, sizeof(engine->base.name), "%s", name);
 	engine->base.id = id;
+	engine->base.mask = BIT(id);
 	engine->base.status_page.addr = (void *)(engine + 1);
 
-	engine->base.context_pin = mock_context_pin;
+	engine->base.cops = &mock_context_ops;
 	engine->base.request_alloc = mock_request_alloc;
 	engine->base.emit_flush = mock_emit_flush;
 	engine->base.emit_fini_breadcrumb = mock_emit_breadcrumb;
 	engine->base.submit_request = mock_submit_request;
 
-	if (i915_timeline_init(i915,
-			       &engine->base.timeline,
-			       engine->base.name,
-			       NULL))
+	engine->base.reset.prepare = mock_reset_prepare;
+	engine->base.reset.reset = mock_reset;
+	engine->base.reset.finish = mock_reset_finish;
+	engine->base.cancel_requests = mock_cancel_requests;
+
+	if (i915_timeline_init(i915, &engine->base.timeline, NULL))
 		goto err_free;
 	i915_timeline_set_subclass(&engine->base.timeline, TIMELINE_ENGINE);
 
@@ -249,7 +268,8 @@ struct intel_engine_cs *mock_engine(struct drm_i915_private *i915,
 	timer_setup(&engine->hw_delay, hw_delay_complete, 0);
 	INIT_LIST_HEAD(&engine->hw_queue);
 
-	if (IS_ERR(intel_context_pin(i915->kernel_context, &engine->base)))
+	if (pin_context(i915->kernel_context, &engine->base,
+			&engine->base.kernel_context))
 		goto err_breadcrumbs;
 
 	return &engine->base;
@@ -266,19 +286,18 @@ void mock_engine_flush(struct intel_engine_cs *engine)
 {
 	struct mock_engine *mock =
 		container_of(engine, typeof(*mock), base);
-	struct mock_request *request, *rn;
+	struct i915_request *request, *rn;
 
 	del_timer_sync(&mock->hw_delay);
 
 	spin_lock_irq(&mock->hw_lock);
-	list_for_each_entry_safe(request, rn, &mock->hw_queue, link)
+	list_for_each_entry_safe(request, rn, &mock->hw_queue, mock.link)
 		advance(request);
 	spin_unlock_irq(&mock->hw_lock);
 }
 
 void mock_engine_reset(struct intel_engine_cs *engine)
 {
-	intel_engine_write_global_seqno(engine, 0);
 }
 
 void mock_engine_free(struct intel_engine_cs *engine)
@@ -293,7 +312,7 @@ void mock_engine_free(struct intel_engine_cs *engine)
 	if (ce)
 		intel_context_unpin(ce);
 
-	__intel_context_unpin(engine->i915->kernel_context, engine);
+	intel_context_unpin(engine->kernel_context);
 
 	intel_engine_fini_breadcrumbs(engine);
 	i915_timeline_fini(&engine->timeline);
diff --git a/drivers/gpu/drm/i915/selftests/mock_gem_device.c b/drivers/gpu/drm/i915/selftests/mock_gem_device.c
index 14ae46fda49f..60bbf8b4df40 100644
--- a/drivers/gpu/drm/i915/selftests/mock_gem_device.c
+++ b/drivers/gpu/drm/i915/selftests/mock_gem_device.c
@@ -79,12 +79,6 @@ static void mock_device_release(struct drm_device *dev)
 
 	destroy_workqueue(i915->wq);
 
-	kmem_cache_destroy(i915->priorities);
-	kmem_cache_destroy(i915->dependencies);
-	kmem_cache_destroy(i915->requests);
-	kmem_cache_destroy(i915->vmas);
-	kmem_cache_destroy(i915->objects);
-
 	i915_gemfs_fini(i915);
 
 	drm_mode_config_cleanup(&i915->drm);
@@ -115,6 +109,10 @@ static void mock_retire_work_handler(struct work_struct *work)
 
 static void mock_idle_work_handler(struct work_struct *work)
 {
+	struct drm_i915_private *i915 =
+		container_of(work, typeof(*i915), gt.idle_work.work);
+
+	i915->gt.active_engines = 0;
 }
 
 static int pm_domain_resume(struct device *dev)
@@ -184,11 +182,12 @@ struct drm_i915_private *mock_gem_device(void)
 		I915_GTT_PAGE_SIZE_64K |
 		I915_GTT_PAGE_SIZE_2M;
 
-	mock_uncore_init(i915);
+	mock_uncore_init(&i915->uncore);
 	i915_gem_init__mm(i915);
 
 	init_waitqueue_head(&i915->gpu_error.wait_queue);
 	init_waitqueue_head(&i915->gpu_error.reset_queue);
+	init_srcu_struct(&i915->gpu_error.reset_backoff_srcu);
 	mutex_init(&i915->gpu_error.wedge_mutex);
 
 	i915->wq = alloc_ordered_workqueue("mock", 0);
@@ -202,31 +201,6 @@ struct drm_i915_private *mock_gem_device(void)
 
 	i915->gt.awake = true;
 
-	i915->objects = KMEM_CACHE(mock_object, SLAB_HWCACHE_ALIGN);
-	if (!i915->objects)
-		goto err_wq;
-
-	i915->vmas = KMEM_CACHE(i915_vma, SLAB_HWCACHE_ALIGN);
-	if (!i915->vmas)
-		goto err_objects;
-
-	i915->requests = KMEM_CACHE(mock_request,
-				    SLAB_HWCACHE_ALIGN |
-				    SLAB_RECLAIM_ACCOUNT |
-				    SLAB_TYPESAFE_BY_RCU);
-	if (!i915->requests)
-		goto err_vmas;
-
-	i915->dependencies = KMEM_CACHE(i915_dependency,
-					SLAB_HWCACHE_ALIGN |
-					SLAB_RECLAIM_ACCOUNT);
-	if (!i915->dependencies)
-		goto err_requests;
-
-	i915->priorities = KMEM_CACHE(i915_priolist, SLAB_HWCACHE_ALIGN);
-	if (!i915->priorities)
-		goto err_dependencies;
-
 	i915_timelines_init(i915);
 
 	INIT_LIST_HEAD(&i915->gt.active_rings);
@@ -236,13 +210,13 @@ struct drm_i915_private *mock_gem_device(void)
 
 	mock_init_ggtt(i915, &i915->ggtt);
 
-	mkwrite_device_info(i915)->ring_mask = BIT(0);
+	mkwrite_device_info(i915)->engine_mask = BIT(0);
 	i915->kernel_context = mock_context(i915, NULL);
 	if (!i915->kernel_context)
 		goto err_unlock;
 
-	i915->engine[RCS] = mock_engine(i915, "mock", RCS);
-	if (!i915->engine[RCS])
+	i915->engine[RCS0] = mock_engine(i915, "mock", RCS0);
+	if (!i915->engine[RCS0])
 		goto err_context;
 
 	mutex_unlock(&i915->drm.struct_mutex);
@@ -256,16 +230,6 @@ err_context:
 err_unlock:
 	mutex_unlock(&i915->drm.struct_mutex);
 	i915_timelines_fini(i915);
-	kmem_cache_destroy(i915->priorities);
-err_dependencies:
-	kmem_cache_destroy(i915->dependencies);
-err_requests:
-	kmem_cache_destroy(i915->requests);
-err_vmas:
-	kmem_cache_destroy(i915->vmas);
-err_objects:
-	kmem_cache_destroy(i915->objects);
-err_wq:
 	destroy_workqueue(i915->wq);
 err_drv:
 	drm_mode_config_cleanup(&i915->drm);
diff --git a/drivers/gpu/drm/i915/selftests/mock_request.c b/drivers/gpu/drm/i915/selftests/mock_request.c
index 0dc29e242597..d1a7c9608712 100644
--- a/drivers/gpu/drm/i915/selftests/mock_request.c
+++ b/drivers/gpu/drm/i915/selftests/mock_request.c
@@ -31,29 +31,25 @@ mock_request(struct intel_engine_cs *engine,
 	     unsigned long delay)
 {
 	struct i915_request *request;
-	struct mock_request *mock;
 
 	/* NB the i915->requests slab cache is enlarged to fit mock_request */
 	request = i915_request_alloc(engine, context);
 	if (IS_ERR(request))
 		return NULL;
 
-	mock = container_of(request, typeof(*mock), base);
-	mock->delay = delay;
-
-	return &mock->base;
+	request->mock.delay = delay;
+	return request;
 }
 
 bool mock_cancel_request(struct i915_request *request)
 {
-	struct mock_request *mock = container_of(request, typeof(*mock), base);
 	struct mock_engine *engine =
 		container_of(request->engine, typeof(*engine), base);
 	bool was_queued;
 
 	spin_lock_irq(&engine->hw_lock);
-	was_queued = !list_empty(&mock->link);
-	list_del_init(&mock->link);
+	was_queued = !list_empty(&request->mock.link);
+	list_del_init(&request->mock.link);
 	spin_unlock_irq(&engine->hw_lock);
 
 	if (was_queued)
diff --git a/drivers/gpu/drm/i915/selftests/mock_request.h b/drivers/gpu/drm/i915/selftests/mock_request.h
index 995fb728380c..4acf0211df20 100644
--- a/drivers/gpu/drm/i915/selftests/mock_request.h
+++ b/drivers/gpu/drm/i915/selftests/mock_request.h
@@ -29,13 +29,6 @@
 
 #include "../i915_request.h"
 
-struct mock_request {
-	struct i915_request base;
-
-	struct list_head link;
-	unsigned long delay;
-};
-
 struct i915_request *
 mock_request(struct intel_engine_cs *engine,
 	     struct i915_gem_context *context,
diff --git a/drivers/gpu/drm/i915/selftests/mock_timeline.c b/drivers/gpu/drm/i915/selftests/mock_timeline.c
index d2de9ece2118..e084476469ef 100644
--- a/drivers/gpu/drm/i915/selftests/mock_timeline.c
+++ b/drivers/gpu/drm/i915/selftests/mock_timeline.c
@@ -14,8 +14,8 @@ void mock_timeline_init(struct i915_timeline *timeline, u64 context)
 	timeline->fence_context = context;
 
 	spin_lock_init(&timeline->lock);
+	mutex_init(&timeline->mutex);
 
-	INIT_ACTIVE_REQUEST(&timeline->barrier);
 	INIT_ACTIVE_REQUEST(&timeline->last_request);
 	INIT_LIST_HEAD(&timeline->requests);
 
diff --git a/drivers/gpu/drm/i915/selftests/mock_uncore.c b/drivers/gpu/drm/i915/selftests/mock_uncore.c
index 8ef14c7e5e38..ff8999c63a12 100644
--- a/drivers/gpu/drm/i915/selftests/mock_uncore.c
+++ b/drivers/gpu/drm/i915/selftests/mock_uncore.c
@@ -26,21 +26,21 @@
 
 #define __nop_write(x) \
 static void \
-nop_write##x(struct drm_i915_private *dev_priv, i915_reg_t reg, u##x val, bool trace) { }
+nop_write##x(struct intel_uncore *uncore, i915_reg_t reg, u##x val, bool trace) { }
 __nop_write(8)
 __nop_write(16)
 __nop_write(32)
 
 #define __nop_read(x) \
 static u##x \
-nop_read##x(struct drm_i915_private *dev_priv, i915_reg_t reg, bool trace) { return 0; }
+nop_read##x(struct intel_uncore *uncore, i915_reg_t reg, bool trace) { return 0; }
 __nop_read(8)
 __nop_read(16)
 __nop_read(32)
 __nop_read(64)
 
-void mock_uncore_init(struct drm_i915_private *i915)
+void mock_uncore_init(struct intel_uncore *uncore)
 {
-	ASSIGN_WRITE_MMIO_VFUNCS(i915, nop);
-	ASSIGN_READ_MMIO_VFUNCS(i915, nop);
+	ASSIGN_WRITE_MMIO_VFUNCS(uncore, nop);
+	ASSIGN_READ_MMIO_VFUNCS(uncore, nop);
 }
diff --git a/drivers/gpu/drm/i915/selftests/mock_uncore.h b/drivers/gpu/drm/i915/selftests/mock_uncore.h
index d79aa3ca4d51..dacb36b5ffcd 100644
--- a/drivers/gpu/drm/i915/selftests/mock_uncore.h
+++ b/drivers/gpu/drm/i915/selftests/mock_uncore.h
@@ -25,6 +25,6 @@
 #ifndef __MOCK_UNCORE_H
 #define __MOCK_UNCORE_H
 
-void mock_uncore_init(struct drm_i915_private *i915);
+void mock_uncore_init(struct intel_uncore *uncore);
 
 #endif /* !__MOCK_UNCORE_H */
diff --git a/drivers/gpu/drm/i915/vlv_dsi.c b/drivers/gpu/drm/i915/vlv_dsi.c
index 31c93c3ccd00..e0b1ec821960 100644
--- a/drivers/gpu/drm/i915/vlv_dsi.c
+++ b/drivers/gpu/drm/i915/vlv_dsi.c
@@ -23,16 +23,20 @@
  * Author: Jani Nikula <jani.nikula@intel.com>
  */
 
+#include <linux/gpio/consumer.h>
+#include <linux/slab.h>
+
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_edid.h>
-#include <drm/i915_drm.h>
 #include <drm/drm_mipi_dsi.h>
-#include <linux/slab.h>
-#include <linux/gpio/consumer.h>
+#include <drm/i915_drm.h>
+
 #include "i915_drv.h"
+#include "intel_connector.h"
 #include "intel_drv.h"
 #include "intel_dsi.h"
+#include "intel_panel.h"
 
 /* return pixels in terms of txbyteclkhs */
 static u16 txbyteclkhs(u16 pixels, int bpp, int lane_count,
@@ -78,7 +82,7 @@ void vlv_dsi_wait_for_fifo_empty(struct intel_dsi *intel_dsi, enum port port)
 	mask = LP_CTRL_FIFO_EMPTY | HS_CTRL_FIFO_EMPTY |
 		LP_DATA_FIFO_EMPTY | HS_DATA_FIFO_EMPTY;
 
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    MIPI_GEN_FIFO_STAT(port), mask, mask,
 				    100))
 		DRM_ERROR("DPI FIFOs are not empty\n");
@@ -148,7 +152,7 @@ static ssize_t intel_dsi_host_transfer(struct mipi_dsi_host *host,
 
 	/* note: this is never true for reads */
 	if (packet.payload_length) {
-		if (intel_wait_for_register(dev_priv,
+		if (intel_wait_for_register(&dev_priv->uncore,
 					    MIPI_GEN_FIFO_STAT(port),
 					    data_mask, 0,
 					    50))
@@ -162,7 +166,7 @@ static ssize_t intel_dsi_host_transfer(struct mipi_dsi_host *host,
 		I915_WRITE(MIPI_INTR_STAT(port), GEN_READ_DATA_AVAIL);
 	}
 
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    MIPI_GEN_FIFO_STAT(port),
 				    ctrl_mask, 0,
 				    50)) {
@@ -174,7 +178,7 @@ static ssize_t intel_dsi_host_transfer(struct mipi_dsi_host *host,
 	/* ->rx_len is set only for reads */
 	if (msg->rx_len) {
 		data_mask = GEN_READ_DATA_AVAIL;
-		if (intel_wait_for_register(dev_priv,
+		if (intel_wait_for_register(&dev_priv->uncore,
 					    MIPI_INTR_STAT(port),
 					    data_mask, data_mask,
 					    50))
@@ -234,7 +238,7 @@ static int dpi_send_cmd(struct intel_dsi *intel_dsi, u32 cmd, bool hs,
 	I915_WRITE(MIPI_DPI_CONTROL(port), cmd);
 
 	mask = SPL_PKT_SENT_INTERRUPT;
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    MIPI_INTR_STAT(port), mask, mask,
 				    100))
 		DRM_ERROR("Video mode command 0x%08x send failed.\n", cmd);
@@ -375,16 +379,18 @@ static bool glk_dsi_enable_io(struct intel_encoder *encoder)
 
 	/* Wait for Pwr ACK */
 	for_each_dsi_port(port, intel_dsi->ports) {
-		if (intel_wait_for_register(dev_priv,
-				MIPI_CTRL(port), GLK_MIPIIO_PORT_POWERED,
-				GLK_MIPIIO_PORT_POWERED, 20))
+		if (intel_wait_for_register(&dev_priv->uncore,
+					    MIPI_CTRL(port),
+					    GLK_MIPIIO_PORT_POWERED,
+					    GLK_MIPIIO_PORT_POWERED,
+					    20))
 			DRM_ERROR("MIPIO port is powergated\n");
 	}
 
 	/* Check for cold boot scenario */
 	for_each_dsi_port(port, intel_dsi->ports) {
-		cold_boot |= !(I915_READ(MIPI_DEVICE_READY(port)) &
-							DEVICE_READY);
+		cold_boot |=
+			!(I915_READ(MIPI_DEVICE_READY(port)) & DEVICE_READY);
 	}
 
 	return cold_boot;
@@ -399,9 +405,11 @@ static void glk_dsi_device_ready(struct intel_encoder *encoder)
 
 	/* Wait for MIPI PHY status bit to set */
 	for_each_dsi_port(port, intel_dsi->ports) {
-		if (intel_wait_for_register(dev_priv,
-				MIPI_CTRL(port), GLK_PHY_STATUS_PORT_READY,
-				GLK_PHY_STATUS_PORT_READY, 20))
+		if (intel_wait_for_register(&dev_priv->uncore,
+					    MIPI_CTRL(port),
+					    GLK_PHY_STATUS_PORT_READY,
+					    GLK_PHY_STATUS_PORT_READY,
+					    20))
 			DRM_ERROR("PHY is not ON\n");
 	}
 
@@ -425,8 +433,11 @@ static void glk_dsi_device_ready(struct intel_encoder *encoder)
 			I915_WRITE(MIPI_DEVICE_READY(port), val);
 
 			/* Wait for ULPS active */
-			if (intel_wait_for_register(dev_priv,
-				MIPI_CTRL(port), GLK_ULPS_NOT_ACTIVE, 0, 20))
+			if (intel_wait_for_register(&dev_priv->uncore,
+						    MIPI_CTRL(port),
+						    GLK_ULPS_NOT_ACTIVE,
+						    0,
+						    20))
 				DRM_ERROR("ULPS not active\n");
 
 			/* Exit ULPS */
@@ -449,17 +460,21 @@ static void glk_dsi_device_ready(struct intel_encoder *encoder)
 
 	/* Wait for Stop state */
 	for_each_dsi_port(port, intel_dsi->ports) {
-		if (intel_wait_for_register(dev_priv,
-				MIPI_CTRL(port), GLK_DATA_LANE_STOP_STATE,
-				GLK_DATA_LANE_STOP_STATE, 20))
+		if (intel_wait_for_register(&dev_priv->uncore,
+					    MIPI_CTRL(port),
+					    GLK_DATA_LANE_STOP_STATE,
+					    GLK_DATA_LANE_STOP_STATE,
+					    20))
 			DRM_ERROR("Date lane not in STOP state\n");
 	}
 
 	/* Wait for AFE LATCH */
 	for_each_dsi_port(port, intel_dsi->ports) {
-		if (intel_wait_for_register(dev_priv,
-				BXT_MIPI_PORT_CTRL(port), AFE_LATCHOUT,
-				AFE_LATCHOUT, 20))
+		if (intel_wait_for_register(&dev_priv->uncore,
+					    BXT_MIPI_PORT_CTRL(port),
+					    AFE_LATCHOUT,
+					    AFE_LATCHOUT,
+					    20))
 			DRM_ERROR("D-PHY not entering LP-11 state\n");
 	}
 }
@@ -559,7 +574,7 @@ static void glk_dsi_enter_low_power_mode(struct intel_encoder *encoder)
 
 	/* Wait for MIPI PHY status bit to unset */
 	for_each_dsi_port(port, intel_dsi->ports) {
-		if (intel_wait_for_register(dev_priv,
+		if (intel_wait_for_register(&dev_priv->uncore,
 					    MIPI_CTRL(port),
 					    GLK_PHY_STATUS_PORT_READY, 0, 20))
 			DRM_ERROR("PHY is not turning OFF\n");
@@ -567,7 +582,7 @@ static void glk_dsi_enter_low_power_mode(struct intel_encoder *encoder)
 
 	/* Wait for Pwr ACK bit to unset */
 	for_each_dsi_port(port, intel_dsi->ports) {
-		if (intel_wait_for_register(dev_priv,
+		if (intel_wait_for_register(&dev_priv->uncore,
 					    MIPI_CTRL(port),
 					    GLK_MIPIIO_PORT_POWERED, 0, 20))
 			DRM_ERROR("MIPI IO Port is not powergated\n");
@@ -588,7 +603,7 @@ static void glk_dsi_disable_mipi_io(struct intel_encoder *encoder)
 
 	/* Wait for MIPI PHY status bit to unset */
 	for_each_dsi_port(port, intel_dsi->ports) {
-		if (intel_wait_for_register(dev_priv,
+		if (intel_wait_for_register(&dev_priv->uncore,
 					    MIPI_CTRL(port),
 					    GLK_PHY_STATUS_PORT_READY, 0, 20))
 			DRM_ERROR("PHY is not turning OFF\n");
@@ -638,7 +653,7 @@ static void vlv_dsi_clear_device_ready(struct intel_encoder *encoder)
 		 * Port A only. MIPI Port C has no similar bit for checking.
 		 */
 		if ((IS_GEN9_LP(dev_priv) || port == PORT_A) &&
-		    intel_wait_for_register(dev_priv,
+		    intel_wait_for_register(&dev_priv->uncore,
 					    port_ctrl, AFE_LATCHOUT, 0,
 					    30))
 			DRM_ERROR("DSI LP not going Low\n");
@@ -1682,7 +1697,7 @@ void vlv_dsi_init(struct drm_i915_private *dev_priv)
 	struct drm_encoder *encoder;
 	struct intel_connector *intel_connector;
 	struct drm_connector *connector;
-	struct drm_display_mode *scan, *fixed_mode = NULL;
+	struct drm_display_mode *fixed_mode;
 	enum port port;
 
 	DRM_DEBUG_KMS("\n");
@@ -1793,13 +1808,7 @@ void vlv_dsi_init(struct drm_i915_private *dev_priv)
 	intel_connector_attach_encoder(intel_connector, intel_encoder);
 
 	mutex_lock(&dev->mode_config.mutex);
-	intel_dsi_vbt_get_modes(intel_dsi);
-	list_for_each_entry(scan, &connector->probed_modes, head) {
-		if ((scan->type & DRM_MODE_TYPE_PREFERRED)) {
-			fixed_mode = drm_mode_duplicate(dev, scan);
-			break;
-		}
-	}
+	fixed_mode = intel_panel_vbt_fixed_mode(intel_connector);
 	mutex_unlock(&dev->mode_config.mutex);
 
 	if (!fixed_mode) {
@@ -1807,9 +1816,6 @@ void vlv_dsi_init(struct drm_i915_private *dev_priv)
 		goto err;
 	}
 
-	connector->display_info.width_mm = fixed_mode->width_mm;
-	connector->display_info.height_mm = fixed_mode->height_mm;
-
 	intel_panel_init(&intel_connector->panel, fixed_mode, NULL);
 	intel_panel_setup_backlight(connector, INVALID_PIPE);
 
diff --git a/drivers/gpu/drm/i915/vlv_dsi_pll.c b/drivers/gpu/drm/i915/vlv_dsi_pll.c
index 954d5a8c4fa7..5e7b1fb2db5d 100644
--- a/drivers/gpu/drm/i915/vlv_dsi_pll.c
+++ b/drivers/gpu/drm/i915/vlv_dsi_pll.c
@@ -244,7 +244,7 @@ void bxt_dsi_pll_disable(struct intel_encoder *encoder)
 	 * PLL lock should deassert within 200us.
 	 * Wait up to 1ms before timing out.
 	 */
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    BXT_DSI_PLL_ENABLE,
 				    BXT_DSI_PLL_LOCKED,
 				    0,
@@ -528,7 +528,7 @@ void bxt_dsi_pll_enable(struct intel_encoder *encoder,
 	I915_WRITE(BXT_DSI_PLL_ENABLE, val);
 
 	/* Timeout and fail if PLL not locked */
-	if (intel_wait_for_register(dev_priv,
+	if (intel_wait_for_register(&dev_priv->uncore,
 				    BXT_DSI_PLL_ENABLE,
 				    BXT_DSI_PLL_LOCKED,
 				    BXT_DSI_PLL_LOCKED,
diff --git a/drivers/gpu/drm/imx/imx-drm-core.c b/drivers/gpu/drm/imx/imx-drm-core.c
index c935cbe059a7..3e8bece620df 100644
--- a/drivers/gpu/drm/imx/imx-drm-core.c
+++ b/drivers/gpu/drm/imx/imx-drm-core.c
@@ -185,7 +185,7 @@ static int compare_of(struct device *dev, void *data)
 	}
 
 	/* Special case for LDB, one device for two channels */
-	if (of_node_cmp(np->name, "lvds-channel") == 0) {
+	if (of_node_name_eq(np, "lvds-channel")) {
 		np = of_get_parent(np);
 		of_node_put(np);
 	}
diff --git a/drivers/gpu/drm/imx/ipuv3-crtc.c b/drivers/gpu/drm/imx/ipuv3-crtc.c
index 54011df8c2e8..9cc1d678674f 100644
--- a/drivers/gpu/drm/imx/ipuv3-crtc.c
+++ b/drivers/gpu/drm/imx/ipuv3-crtc.c
@@ -295,7 +295,7 @@ static void ipu_crtc_mode_set_nofb(struct drm_crtc *crtc)
 	sig_cfg.enable_pol = !(imx_crtc_state->bus_flags & DRM_BUS_FLAG_DE_LOW);
 	/* Default to driving pixel data on negative clock edges */
 	sig_cfg.clk_pol = !!(imx_crtc_state->bus_flags &
-			     DRM_BUS_FLAG_PIXDATA_POSEDGE);
+			     DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE);
 	sig_cfg.bus_format = imx_crtc_state->bus_format;
 	sig_cfg.v_to_h_sync = 0;
 	sig_cfg.hsync_pin = imx_crtc_state->di_hsync_pin;
diff --git a/drivers/gpu/drm/lima/Kconfig b/drivers/gpu/drm/lima/Kconfig
new file mode 100644
index 000000000000..bb4ddc6bb0a6
--- /dev/null
+++ b/drivers/gpu/drm/lima/Kconfig
@@ -0,0 +1,13 @@
+# SPDX-License-Identifier: GPL-2.0 OR MIT
+# Copyright 2017-2019 Qiang Yu <yuq825@gmail.com>
+
+config DRM_LIMA
+       tristate "LIMA (DRM support for ARM Mali 400/450 GPU)"
+       depends on DRM
+       depends on ARM || ARM64 || COMPILE_TEST
+       depends on MMU
+       depends on COMMON_CLK
+       depends on OF
+       select DRM_SCHED
+       help
+         DRM driver for ARM Mali 400/450 GPUs.
diff --git a/drivers/gpu/drm/lima/Makefile b/drivers/gpu/drm/lima/Makefile
new file mode 100644
index 000000000000..38cc70281ba5
--- /dev/null
+++ b/drivers/gpu/drm/lima/Makefile
@@ -0,0 +1,21 @@
+# SPDX-License-Identifier: GPL-2.0 OR MIT
+# Copyright 2017-2019 Qiang Yu <yuq825@gmail.com>
+
+lima-y := \
+	lima_drv.o \
+	lima_device.o \
+	lima_pmu.o \
+	lima_l2_cache.o \
+	lima_mmu.o \
+	lima_gp.o \
+	lima_pp.o \
+	lima_gem.o \
+	lima_vm.o \
+	lima_sched.o \
+	lima_ctx.o \
+	lima_gem_prime.o \
+	lima_dlbu.o \
+	lima_bcast.o \
+	lima_object.o
+
+obj-$(CONFIG_DRM_LIMA) += lima.o
diff --git a/drivers/gpu/drm/lima/lima_bcast.c b/drivers/gpu/drm/lima/lima_bcast.c
new file mode 100644
index 000000000000..288398027bfa
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_bcast.c
@@ -0,0 +1,47 @@
+// SPDX-License-Identifier: GPL-2.0 OR MIT
+/* Copyright 2018-2019 Qiang Yu <yuq825@gmail.com> */
+
+#include <linux/io.h>
+#include <linux/device.h>
+
+#include "lima_device.h"
+#include "lima_bcast.h"
+#include "lima_regs.h"
+
+#define bcast_write(reg, data) writel(data, ip->iomem + reg)
+#define bcast_read(reg) readl(ip->iomem + reg)
+
+void lima_bcast_enable(struct lima_device *dev, int num_pp)
+{
+	struct lima_sched_pipe *pipe = dev->pipe + lima_pipe_pp;
+	struct lima_ip *ip = dev->ip + lima_ip_bcast;
+	int i, mask = bcast_read(LIMA_BCAST_BROADCAST_MASK) & 0xffff0000;
+
+	for (i = 0; i < num_pp; i++) {
+		struct lima_ip *pp = pipe->processor[i];
+
+		mask |= 1 << (pp->id - lima_ip_pp0);
+	}
+
+	bcast_write(LIMA_BCAST_BROADCAST_MASK, mask);
+}
+
+int lima_bcast_init(struct lima_ip *ip)
+{
+	int i, mask = 0;
+
+	for (i = lima_ip_pp0; i <= lima_ip_pp7; i++) {
+		if (ip->dev->ip[i].present)
+			mask |= 1 << (i - lima_ip_pp0);
+	}
+
+	bcast_write(LIMA_BCAST_BROADCAST_MASK, mask << 16);
+	bcast_write(LIMA_BCAST_INTERRUPT_MASK, mask);
+	return 0;
+}
+
+void lima_bcast_fini(struct lima_ip *ip)
+{
+
+}
+
diff --git a/drivers/gpu/drm/lima/lima_bcast.h b/drivers/gpu/drm/lima/lima_bcast.h
new file mode 100644
index 000000000000..c47e58563d0a
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_bcast.h
@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: GPL-2.0 OR MIT */
+/* Copyright 2018-2019 Qiang Yu <yuq825@gmail.com> */
+
+#ifndef __LIMA_BCAST_H__
+#define __LIMA_BCAST_H__
+
+struct lima_ip;
+
+int lima_bcast_init(struct lima_ip *ip);
+void lima_bcast_fini(struct lima_ip *ip);
+
+void lima_bcast_enable(struct lima_device *dev, int num_pp);
+
+#endif
diff --git a/drivers/gpu/drm/lima/lima_ctx.c b/drivers/gpu/drm/lima/lima_ctx.c
new file mode 100644
index 000000000000..22fff6caa961
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_ctx.c
@@ -0,0 +1,98 @@
+// SPDX-License-Identifier: GPL-2.0 OR MIT
+/* Copyright 2018-2019 Qiang Yu <yuq825@gmail.com> */
+
+#include <linux/slab.h>
+
+#include "lima_device.h"
+#include "lima_ctx.h"
+
+int lima_ctx_create(struct lima_device *dev, struct lima_ctx_mgr *mgr, u32 *id)
+{
+	struct lima_ctx *ctx;
+	int i, err;
+
+	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+	if (!ctx)
+		return -ENOMEM;
+	ctx->dev = dev;
+	kref_init(&ctx->refcnt);
+
+	for (i = 0; i < lima_pipe_num; i++) {
+		err = lima_sched_context_init(dev->pipe + i, ctx->context + i, &ctx->guilty);
+		if (err)
+			goto err_out0;
+	}
+
+	err = xa_alloc(&mgr->handles, id, ctx, xa_limit_32b, GFP_KERNEL);
+	if (err < 0)
+		goto err_out0;
+
+	return 0;
+
+err_out0:
+	for (i--; i >= 0; i--)
+		lima_sched_context_fini(dev->pipe + i, ctx->context + i);
+	kfree(ctx);
+	return err;
+}
+
+static void lima_ctx_do_release(struct kref *ref)
+{
+	struct lima_ctx *ctx = container_of(ref, struct lima_ctx, refcnt);
+	int i;
+
+	for (i = 0; i < lima_pipe_num; i++)
+		lima_sched_context_fini(ctx->dev->pipe + i, ctx->context + i);
+	kfree(ctx);
+}
+
+int lima_ctx_free(struct lima_ctx_mgr *mgr, u32 id)
+{
+	struct lima_ctx *ctx;
+	int ret = 0;
+
+	mutex_lock(&mgr->lock);
+	ctx = xa_erase(&mgr->handles, id);
+	if (ctx)
+		kref_put(&ctx->refcnt, lima_ctx_do_release);
+	else
+		ret = -EINVAL;
+	mutex_unlock(&mgr->lock);
+	return ret;
+}
+
+struct lima_ctx *lima_ctx_get(struct lima_ctx_mgr *mgr, u32 id)
+{
+	struct lima_ctx *ctx;
+
+	mutex_lock(&mgr->lock);
+	ctx = xa_load(&mgr->handles, id);
+	if (ctx)
+		kref_get(&ctx->refcnt);
+	mutex_unlock(&mgr->lock);
+	return ctx;
+}
+
+void lima_ctx_put(struct lima_ctx *ctx)
+{
+	kref_put(&ctx->refcnt, lima_ctx_do_release);
+}
+
+void lima_ctx_mgr_init(struct lima_ctx_mgr *mgr)
+{
+	mutex_init(&mgr->lock);
+	xa_init_flags(&mgr->handles, XA_FLAGS_ALLOC);
+}
+
+void lima_ctx_mgr_fini(struct lima_ctx_mgr *mgr)
+{
+	struct lima_ctx *ctx;
+	unsigned long id;
+
+	xa_for_each(&mgr->handles, id, ctx) {
+		kref_put(&ctx->refcnt, lima_ctx_do_release);
+	}
+
+	xa_destroy(&mgr->handles);
+	mutex_destroy(&mgr->lock);
+}
diff --git a/drivers/gpu/drm/lima/lima_ctx.h b/drivers/gpu/drm/lima/lima_ctx.h
new file mode 100644
index 000000000000..6154e5c9bfe4
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_ctx.h
@@ -0,0 +1,30 @@
+/* SPDX-License-Identifier: GPL-2.0 OR MIT */
+/* Copyright 2018-2019 Qiang Yu <yuq825@gmail.com> */
+
+#ifndef __LIMA_CTX_H__
+#define __LIMA_CTX_H__
+
+#include <linux/xarray.h>
+
+#include "lima_device.h"
+
+struct lima_ctx {
+	struct kref refcnt;
+	struct lima_device *dev;
+	struct lima_sched_context context[lima_pipe_num];
+	atomic_t guilty;
+};
+
+struct lima_ctx_mgr {
+	struct mutex lock;
+	struct xarray handles;
+};
+
+int lima_ctx_create(struct lima_device *dev, struct lima_ctx_mgr *mgr, u32 *id);
+int lima_ctx_free(struct lima_ctx_mgr *mgr, u32 id);
+struct lima_ctx *lima_ctx_get(struct lima_ctx_mgr *mgr, u32 id);
+void lima_ctx_put(struct lima_ctx *ctx);
+void lima_ctx_mgr_init(struct lima_ctx_mgr *mgr);
+void lima_ctx_mgr_fini(struct lima_ctx_mgr *mgr);
+
+#endif
diff --git a/drivers/gpu/drm/lima/lima_device.c b/drivers/gpu/drm/lima/lima_device.c
new file mode 100644
index 000000000000..570d0e93f9a9
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_device.c
@@ -0,0 +1,385 @@
+// SPDX-License-Identifier: GPL-2.0 OR MIT
+/* Copyright 2017-2019 Qiang Yu <yuq825@gmail.com> */
+
+#include <linux/regulator/consumer.h>
+#include <linux/reset.h>
+#include <linux/clk.h>
+#include <linux/dma-mapping.h>
+#include <linux/platform_device.h>
+
+#include "lima_device.h"
+#include "lima_gp.h"
+#include "lima_pp.h"
+#include "lima_mmu.h"
+#include "lima_pmu.h"
+#include "lima_l2_cache.h"
+#include "lima_dlbu.h"
+#include "lima_bcast.h"
+#include "lima_vm.h"
+
+struct lima_ip_desc {
+	char *name;
+	char *irq_name;
+	bool must_have[lima_gpu_num];
+	int offset[lima_gpu_num];
+
+	int (*init)(struct lima_ip *ip);
+	void (*fini)(struct lima_ip *ip);
+};
+
+#define LIMA_IP_DESC(ipname, mst0, mst1, off0, off1, func, irq) \
+	[lima_ip_##ipname] = { \
+		.name = #ipname, \
+		.irq_name = irq, \
+		.must_have = { \
+			[lima_gpu_mali400] = mst0, \
+			[lima_gpu_mali450] = mst1, \
+		}, \
+		.offset = { \
+			[lima_gpu_mali400] = off0, \
+			[lima_gpu_mali450] = off1, \
+		}, \
+		.init = lima_##func##_init, \
+		.fini = lima_##func##_fini, \
+	}
+
+static struct lima_ip_desc lima_ip_desc[lima_ip_num] = {
+	LIMA_IP_DESC(pmu,         false, false, 0x02000, 0x02000, pmu,      "pmu"),
+	LIMA_IP_DESC(l2_cache0,   true,  true,  0x01000, 0x10000, l2_cache, NULL),
+	LIMA_IP_DESC(l2_cache1,   false, true,  -1,      0x01000, l2_cache, NULL),
+	LIMA_IP_DESC(l2_cache2,   false, false, -1,      0x11000, l2_cache, NULL),
+	LIMA_IP_DESC(gp,          true,  true,  0x00000, 0x00000, gp,       "gp"),
+	LIMA_IP_DESC(pp0,         true,  true,  0x08000, 0x08000, pp,       "pp0"),
+	LIMA_IP_DESC(pp1,         false, false, 0x0A000, 0x0A000, pp,       "pp1"),
+	LIMA_IP_DESC(pp2,         false, false, 0x0C000, 0x0C000, pp,       "pp2"),
+	LIMA_IP_DESC(pp3,         false, false, 0x0E000, 0x0E000, pp,       "pp3"),
+	LIMA_IP_DESC(pp4,         false, false, -1,      0x28000, pp,       "pp4"),
+	LIMA_IP_DESC(pp5,         false, false, -1,      0x2A000, pp,       "pp5"),
+	LIMA_IP_DESC(pp6,         false, false, -1,      0x2C000, pp,       "pp6"),
+	LIMA_IP_DESC(pp7,         false, false, -1,      0x2E000, pp,       "pp7"),
+	LIMA_IP_DESC(gpmmu,       true,  true,  0x03000, 0x03000, mmu,      "gpmmu"),
+	LIMA_IP_DESC(ppmmu0,      true,  true,  0x04000, 0x04000, mmu,      "ppmmu0"),
+	LIMA_IP_DESC(ppmmu1,      false, false, 0x05000, 0x05000, mmu,      "ppmmu1"),
+	LIMA_IP_DESC(ppmmu2,      false, false, 0x06000, 0x06000, mmu,      "ppmmu2"),
+	LIMA_IP_DESC(ppmmu3,      false, false, 0x07000, 0x07000, mmu,      "ppmmu3"),
+	LIMA_IP_DESC(ppmmu4,      false, false, -1,      0x1C000, mmu,      "ppmmu4"),
+	LIMA_IP_DESC(ppmmu5,      false, false, -1,      0x1D000, mmu,      "ppmmu5"),
+	LIMA_IP_DESC(ppmmu6,      false, false, -1,      0x1E000, mmu,      "ppmmu6"),
+	LIMA_IP_DESC(ppmmu7,      false, false, -1,      0x1F000, mmu,      "ppmmu7"),
+	LIMA_IP_DESC(dlbu,        false, true,  -1,      0x14000, dlbu,     NULL),
+	LIMA_IP_DESC(bcast,       false, true,  -1,      0x13000, bcast,    NULL),
+	LIMA_IP_DESC(pp_bcast,    false, true,  -1,      0x16000, pp_bcast, "pp"),
+	LIMA_IP_DESC(ppmmu_bcast, false, true,  -1,      0x15000, mmu,      NULL),
+};
+
+const char *lima_ip_name(struct lima_ip *ip)
+{
+	return lima_ip_desc[ip->id].name;
+}
+
+static int lima_clk_init(struct lima_device *dev)
+{
+	int err;
+	unsigned long bus_rate, gpu_rate;
+
+	dev->clk_bus = devm_clk_get(dev->dev, "bus");
+	if (IS_ERR(dev->clk_bus)) {
+		dev_err(dev->dev, "get bus clk failed %ld\n", PTR_ERR(dev->clk_bus));
+		return PTR_ERR(dev->clk_bus);
+	}
+
+	dev->clk_gpu = devm_clk_get(dev->dev, "core");
+	if (IS_ERR(dev->clk_gpu)) {
+		dev_err(dev->dev, "get core clk failed %ld\n", PTR_ERR(dev->clk_gpu));
+		return PTR_ERR(dev->clk_gpu);
+	}
+
+	bus_rate = clk_get_rate(dev->clk_bus);
+	dev_info(dev->dev, "bus rate = %lu\n", bus_rate);
+
+	gpu_rate = clk_get_rate(dev->clk_gpu);
+	dev_info(dev->dev, "mod rate = %lu", gpu_rate);
+
+	err = clk_prepare_enable(dev->clk_bus);
+	if (err)
+		return err;
+
+	err = clk_prepare_enable(dev->clk_gpu);
+	if (err)
+		goto error_out0;
+
+	dev->reset = devm_reset_control_get_optional(dev->dev, NULL);
+	if (IS_ERR(dev->reset)) {
+		err = PTR_ERR(dev->reset);
+		goto error_out1;
+	} else if (dev->reset != NULL) {
+		err = reset_control_deassert(dev->reset);
+		if (err)
+			goto error_out1;
+	}
+
+	return 0;
+
+error_out1:
+	clk_disable_unprepare(dev->clk_gpu);
+error_out0:
+	clk_disable_unprepare(dev->clk_bus);
+	return err;
+}
+
+static void lima_clk_fini(struct lima_device *dev)
+{
+	if (dev->reset != NULL)
+		reset_control_assert(dev->reset);
+	clk_disable_unprepare(dev->clk_gpu);
+	clk_disable_unprepare(dev->clk_bus);
+}
+
+static int lima_regulator_init(struct lima_device *dev)
+{
+	int ret;
+
+	dev->regulator = devm_regulator_get_optional(dev->dev, "mali");
+	if (IS_ERR(dev->regulator)) {
+		ret = PTR_ERR(dev->regulator);
+		dev->regulator = NULL;
+		if (ret == -ENODEV)
+			return 0;
+		dev_err(dev->dev, "failed to get regulator: %d\n", ret);
+		return ret;
+	}
+
+	ret = regulator_enable(dev->regulator);
+	if (ret < 0) {
+		dev_err(dev->dev, "failed to enable regulator: %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static void lima_regulator_fini(struct lima_device *dev)
+{
+	if (dev->regulator)
+		regulator_disable(dev->regulator);
+}
+
+static int lima_init_ip(struct lima_device *dev, int index)
+{
+	struct lima_ip_desc *desc = lima_ip_desc + index;
+	struct lima_ip *ip = dev->ip + index;
+	int offset = desc->offset[dev->id];
+	bool must = desc->must_have[dev->id];
+	int err;
+
+	if (offset < 0)
+		return 0;
+
+	ip->dev = dev;
+	ip->id = index;
+	ip->iomem = dev->iomem + offset;
+	if (desc->irq_name) {
+		err = platform_get_irq_byname(dev->pdev, desc->irq_name);
+		if (err < 0)
+			goto out;
+		ip->irq = err;
+	}
+
+	err = desc->init(ip);
+	if (!err) {
+		ip->present = true;
+		return 0;
+	}
+
+out:
+	return must ? err : 0;
+}
+
+static void lima_fini_ip(struct lima_device *ldev, int index)
+{
+	struct lima_ip_desc *desc = lima_ip_desc + index;
+	struct lima_ip *ip = ldev->ip + index;
+
+	if (ip->present)
+		desc->fini(ip);
+}
+
+static int lima_init_gp_pipe(struct lima_device *dev)
+{
+	struct lima_sched_pipe *pipe = dev->pipe + lima_pipe_gp;
+	int err;
+
+	err = lima_sched_pipe_init(pipe, "gp");
+	if (err)
+		return err;
+
+	pipe->l2_cache[pipe->num_l2_cache++] = dev->ip + lima_ip_l2_cache0;
+	pipe->mmu[pipe->num_mmu++] = dev->ip + lima_ip_gpmmu;
+	pipe->processor[pipe->num_processor++] = dev->ip + lima_ip_gp;
+
+	err = lima_gp_pipe_init(dev);
+	if (err) {
+		lima_sched_pipe_fini(pipe);
+		return err;
+	}
+
+	return 0;
+}
+
+static void lima_fini_gp_pipe(struct lima_device *dev)
+{
+	struct lima_sched_pipe *pipe = dev->pipe + lima_pipe_gp;
+
+	lima_gp_pipe_fini(dev);
+	lima_sched_pipe_fini(pipe);
+}
+
+static int lima_init_pp_pipe(struct lima_device *dev)
+{
+	struct lima_sched_pipe *pipe = dev->pipe + lima_pipe_pp;
+	int err, i;
+
+	err = lima_sched_pipe_init(pipe, "pp");
+	if (err)
+		return err;
+
+	for (i = 0; i < LIMA_SCHED_PIPE_MAX_PROCESSOR; i++) {
+		struct lima_ip *pp = dev->ip + lima_ip_pp0 + i;
+		struct lima_ip *ppmmu = dev->ip + lima_ip_ppmmu0 + i;
+		struct lima_ip *l2_cache;
+
+		if (dev->id == lima_gpu_mali400)
+			l2_cache = dev->ip + lima_ip_l2_cache0;
+		else
+			l2_cache = dev->ip + lima_ip_l2_cache1 + (i >> 2);
+
+		if (pp->present && ppmmu->present && l2_cache->present) {
+			pipe->mmu[pipe->num_mmu++] = ppmmu;
+			pipe->processor[pipe->num_processor++] = pp;
+			if (!pipe->l2_cache[i >> 2])
+				pipe->l2_cache[pipe->num_l2_cache++] = l2_cache;
+		}
+	}
+
+	if (dev->ip[lima_ip_bcast].present) {
+		pipe->bcast_processor = dev->ip + lima_ip_pp_bcast;
+		pipe->bcast_mmu = dev->ip + lima_ip_ppmmu_bcast;
+	}
+
+	err = lima_pp_pipe_init(dev);
+	if (err) {
+		lima_sched_pipe_fini(pipe);
+		return err;
+	}
+
+	return 0;
+}
+
+static void lima_fini_pp_pipe(struct lima_device *dev)
+{
+	struct lima_sched_pipe *pipe = dev->pipe + lima_pipe_pp;
+
+	lima_pp_pipe_fini(dev);
+	lima_sched_pipe_fini(pipe);
+}
+
+int lima_device_init(struct lima_device *ldev)
+{
+	int err, i;
+	struct resource *res;
+
+	dma_set_coherent_mask(ldev->dev, DMA_BIT_MASK(32));
+
+	err = lima_clk_init(ldev);
+	if (err) {
+		dev_err(ldev->dev, "clk init fail %d\n", err);
+		return err;
+	}
+
+	err = lima_regulator_init(ldev);
+	if (err) {
+		dev_err(ldev->dev, "regulator init fail %d\n", err);
+		goto err_out0;
+	}
+
+	ldev->empty_vm = lima_vm_create(ldev);
+	if (!ldev->empty_vm) {
+		err = -ENOMEM;
+		goto err_out1;
+	}
+
+	ldev->va_start = 0;
+	if (ldev->id == lima_gpu_mali450) {
+		ldev->va_end = LIMA_VA_RESERVE_START;
+		ldev->dlbu_cpu = dma_alloc_wc(
+			ldev->dev, LIMA_PAGE_SIZE,
+			&ldev->dlbu_dma, GFP_KERNEL);
+		if (!ldev->dlbu_cpu) {
+			err = -ENOMEM;
+			goto err_out2;
+		}
+	} else
+		ldev->va_end = LIMA_VA_RESERVE_END;
+
+	res = platform_get_resource(ldev->pdev, IORESOURCE_MEM, 0);
+	ldev->iomem = devm_ioremap_resource(ldev->dev, res);
+	if (IS_ERR(ldev->iomem)) {
+		dev_err(ldev->dev, "fail to ioremap iomem\n");
+		err = PTR_ERR(ldev->iomem);
+		goto err_out3;
+	}
+
+	for (i = 0; i < lima_ip_num; i++) {
+		err = lima_init_ip(ldev, i);
+		if (err)
+			goto err_out4;
+	}
+
+	err = lima_init_gp_pipe(ldev);
+	if (err)
+		goto err_out4;
+
+	err = lima_init_pp_pipe(ldev);
+	if (err)
+		goto err_out5;
+
+	return 0;
+
+err_out5:
+	lima_fini_gp_pipe(ldev);
+err_out4:
+	while (--i >= 0)
+		lima_fini_ip(ldev, i);
+err_out3:
+	if (ldev->dlbu_cpu)
+		dma_free_wc(ldev->dev, LIMA_PAGE_SIZE,
+			    ldev->dlbu_cpu, ldev->dlbu_dma);
+err_out2:
+	lima_vm_put(ldev->empty_vm);
+err_out1:
+	lima_regulator_fini(ldev);
+err_out0:
+	lima_clk_fini(ldev);
+	return err;
+}
+
+void lima_device_fini(struct lima_device *ldev)
+{
+	int i;
+
+	lima_fini_pp_pipe(ldev);
+	lima_fini_gp_pipe(ldev);
+
+	for (i = lima_ip_num - 1; i >= 0; i--)
+		lima_fini_ip(ldev, i);
+
+	if (ldev->dlbu_cpu)
+		dma_free_wc(ldev->dev, LIMA_PAGE_SIZE,
+			    ldev->dlbu_cpu, ldev->dlbu_dma);
+
+	lima_vm_put(ldev->empty_vm);
+
+	lima_regulator_fini(ldev);
+
+	lima_clk_fini(ldev);
+}
diff --git a/drivers/gpu/drm/lima/lima_device.h b/drivers/gpu/drm/lima/lima_device.h
new file mode 100644
index 000000000000..31158d86271c
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_device.h
@@ -0,0 +1,131 @@
+/* SPDX-License-Identifier: GPL-2.0 OR MIT */
+/* Copyright 2018-2019 Qiang Yu <yuq825@gmail.com> */
+
+#ifndef __LIMA_DEVICE_H__
+#define __LIMA_DEVICE_H__
+
+#include <drm/drm_device.h>
+#include <linux/delay.h>
+
+#include "lima_sched.h"
+
+enum lima_gpu_id {
+	lima_gpu_mali400 = 0,
+	lima_gpu_mali450,
+	lima_gpu_num,
+};
+
+enum lima_ip_id {
+	lima_ip_pmu,
+	lima_ip_gpmmu,
+	lima_ip_ppmmu0,
+	lima_ip_ppmmu1,
+	lima_ip_ppmmu2,
+	lima_ip_ppmmu3,
+	lima_ip_ppmmu4,
+	lima_ip_ppmmu5,
+	lima_ip_ppmmu6,
+	lima_ip_ppmmu7,
+	lima_ip_gp,
+	lima_ip_pp0,
+	lima_ip_pp1,
+	lima_ip_pp2,
+	lima_ip_pp3,
+	lima_ip_pp4,
+	lima_ip_pp5,
+	lima_ip_pp6,
+	lima_ip_pp7,
+	lima_ip_l2_cache0,
+	lima_ip_l2_cache1,
+	lima_ip_l2_cache2,
+	lima_ip_dlbu,
+	lima_ip_bcast,
+	lima_ip_pp_bcast,
+	lima_ip_ppmmu_bcast,
+	lima_ip_num,
+};
+
+struct lima_device;
+
+struct lima_ip {
+	struct lima_device *dev;
+	enum lima_ip_id id;
+	bool present;
+
+	void __iomem *iomem;
+	int irq;
+
+	union {
+		/* gp/pp */
+		bool async_reset;
+		/* l2 cache */
+		spinlock_t lock;
+	} data;
+};
+
+enum lima_pipe_id {
+	lima_pipe_gp,
+	lima_pipe_pp,
+	lima_pipe_num,
+};
+
+struct lima_device {
+	struct device *dev;
+	struct drm_device *ddev;
+	struct platform_device *pdev;
+
+	enum lima_gpu_id id;
+	u32 gp_version;
+	u32 pp_version;
+	int num_pp;
+
+	void __iomem *iomem;
+	struct clk *clk_bus;
+	struct clk *clk_gpu;
+	struct reset_control *reset;
+	struct regulator *regulator;
+
+	struct lima_ip ip[lima_ip_num];
+	struct lima_sched_pipe pipe[lima_pipe_num];
+
+	struct lima_vm *empty_vm;
+	uint64_t va_start;
+	uint64_t va_end;
+
+	u32 *dlbu_cpu;
+	dma_addr_t dlbu_dma;
+};
+
+static inline struct lima_device *
+to_lima_dev(struct drm_device *dev)
+{
+	return dev->dev_private;
+}
+
+int lima_device_init(struct lima_device *ldev);
+void lima_device_fini(struct lima_device *ldev);
+
+const char *lima_ip_name(struct lima_ip *ip);
+
+typedef int (*lima_poll_func_t)(struct lima_ip *);
+
+static inline int lima_poll_timeout(struct lima_ip *ip, lima_poll_func_t func,
+				    int sleep_us, int timeout_us)
+{
+	ktime_t timeout = ktime_add_us(ktime_get(), timeout_us);
+
+	might_sleep_if(sleep_us);
+	while (1) {
+		if (func(ip))
+			return 0;
+
+		if (timeout_us && ktime_compare(ktime_get(), timeout) > 0)
+			return -ETIMEDOUT;
+
+		if (sleep_us)
+			usleep_range((sleep_us >> 2) + 1, sleep_us);
+	}
+	return 0;
+}
+
+#endif
diff --git a/drivers/gpu/drm/lima/lima_dlbu.c b/drivers/gpu/drm/lima/lima_dlbu.c
new file mode 100644
index 000000000000..8399ceffb94b
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_dlbu.c
@@ -0,0 +1,58 @@
+// SPDX-License-Identifier: GPL-2.0 OR MIT
+/* Copyright 2018-2019 Qiang Yu <yuq825@gmail.com> */
+
+#include <linux/io.h>
+#include <linux/device.h>
+
+#include "lima_device.h"
+#include "lima_dlbu.h"
+#include "lima_vm.h"
+#include "lima_regs.h"
+
+#define dlbu_write(reg, data) writel(data, ip->iomem + reg)
+#define dlbu_read(reg) readl(ip->iomem + reg)
+
+void lima_dlbu_enable(struct lima_device *dev, int num_pp)
+{
+	struct lima_sched_pipe *pipe = dev->pipe + lima_pipe_pp;
+	struct lima_ip *ip = dev->ip + lima_ip_dlbu;
+	int i, mask = 0;
+
+	for (i = 0; i < num_pp; i++) {
+		struct lima_ip *pp = pipe->processor[i];
+
+		mask |= 1 << (pp->id - lima_ip_pp0);
+	}
+
+	dlbu_write(LIMA_DLBU_PP_ENABLE_MASK, mask);
+}
+
+void lima_dlbu_disable(struct lima_device *dev)
+{
+	struct lima_ip *ip = dev->ip + lima_ip_dlbu;
+
+	dlbu_write(LIMA_DLBU_PP_ENABLE_MASK, 0);
+}
+
+void lima_dlbu_set_reg(struct lima_ip *ip, u32 *reg)
+{
+	dlbu_write(LIMA_DLBU_TLLIST_VBASEADDR, reg[0]);
+	dlbu_write(LIMA_DLBU_FB_DIM, reg[1]);
+	dlbu_write(LIMA_DLBU_TLLIST_CONF, reg[2]);
+	dlbu_write(LIMA_DLBU_START_TILE_POS, reg[3]);
+}
+
+int lima_dlbu_init(struct lima_ip *ip)
+{
+	struct lima_device *dev = ip->dev;
+
+	dlbu_write(LIMA_DLBU_MASTER_TLLIST_PHYS_ADDR, dev->dlbu_dma | 1);
+	dlbu_write(LIMA_DLBU_MASTER_TLLIST_VADDR, LIMA_VA_RESERVE_DLBU);
+
+	return 0;
+}
+
+void lima_dlbu_fini(struct lima_ip *ip)
+{
+
+}
diff --git a/drivers/gpu/drm/lima/lima_dlbu.h b/drivers/gpu/drm/lima/lima_dlbu.h
new file mode 100644
index 000000000000..16f877984466
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_dlbu.h
@@ -0,0 +1,18 @@
+/* SPDX-License-Identifier: GPL-2.0 OR MIT */
+/* Copyright 2018-2019 Qiang Yu <yuq825@gmail.com> */
+
+#ifndef __LIMA_DLBU_H__
+#define __LIMA_DLBU_H__
+
+struct lima_ip;
+struct lima_device;
+
+void lima_dlbu_enable(struct lima_device *dev, int num_pp);
+void lima_dlbu_disable(struct lima_device *dev);
+
+void lima_dlbu_set_reg(struct lima_ip *ip, u32 *reg);
+
+int lima_dlbu_init(struct lima_ip *ip);
+void lima_dlbu_fini(struct lima_ip *ip);
+
+#endif
diff --git a/drivers/gpu/drm/lima/lima_drv.c b/drivers/gpu/drm/lima/lima_drv.c
new file mode 100644
index 000000000000..f9a281a62083
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_drv.c
@@ -0,0 +1,376 @@
+// SPDX-License-Identifier: GPL-2.0 OR MIT
+/* Copyright 2017-2019 Qiang Yu <yuq825@gmail.com> */
+
+#include <linux/module.h>
+#include <linux/of_platform.h>
+#include <linux/uaccess.h>
+#include <linux/slab.h>
+#include <drm/drm_ioctl.h>
+#include <drm/drm_drv.h>
+#include <drm/drm_prime.h>
+#include <drm/lima_drm.h>
+
+#include "lima_drv.h"
+#include "lima_gem.h"
+#include "lima_gem_prime.h"
+#include "lima_vm.h"
+
+int lima_sched_timeout_ms;
+
+MODULE_PARM_DESC(sched_timeout_ms, "task run timeout in ms (0 = no timeout (default))");
+module_param_named(sched_timeout_ms, lima_sched_timeout_ms, int, 0444);
+
+static int lima_ioctl_get_param(struct drm_device *dev, void *data, struct drm_file *file)
+{
+	struct drm_lima_get_param *args = data;
+	struct lima_device *ldev = to_lima_dev(dev);
+
+	if (args->pad)
+		return -EINVAL;
+
+	switch (args->param) {
+	case DRM_LIMA_PARAM_GPU_ID:
+		switch (ldev->id) {
+		case lima_gpu_mali400:
+			args->value = DRM_LIMA_PARAM_GPU_ID_MALI400;
+			break;
+		case lima_gpu_mali450:
+			args->value = DRM_LIMA_PARAM_GPU_ID_MALI450;
+			break;
+		default:
+			args->value = DRM_LIMA_PARAM_GPU_ID_UNKNOWN;
+			break;
+		}
+		break;
+
+	case DRM_LIMA_PARAM_NUM_PP:
+		args->value = ldev->pipe[lima_pipe_pp].num_processor;
+		break;
+
+	case DRM_LIMA_PARAM_GP_VERSION:
+		args->value = ldev->gp_version;
+		break;
+
+	case DRM_LIMA_PARAM_PP_VERSION:
+		args->value = ldev->pp_version;
+		break;
+
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int lima_ioctl_gem_create(struct drm_device *dev, void *data, struct drm_file *file)
+{
+	struct drm_lima_gem_create *args = data;
+
+	if (args->pad)
+		return -EINVAL;
+
+	if (args->flags)
+		return -EINVAL;
+
+	if (args->size == 0)
+		return -EINVAL;
+
+	return lima_gem_create_handle(dev, file, args->size, args->flags, &args->handle);
+}
+
+static int lima_ioctl_gem_info(struct drm_device *dev, void *data, struct drm_file *file)
+{
+	struct drm_lima_gem_info *args = data;
+
+	return lima_gem_get_info(file, args->handle, &args->va, &args->offset);
+}
+
+static int lima_ioctl_gem_submit(struct drm_device *dev, void *data, struct drm_file *file)
+{
+	struct drm_lima_gem_submit *args = data;
+	struct lima_device *ldev = to_lima_dev(dev);
+	struct lima_drm_priv *priv = file->driver_priv;
+	struct drm_lima_gem_submit_bo *bos;
+	struct lima_sched_pipe *pipe;
+	struct lima_sched_task *task;
+	struct lima_ctx *ctx;
+	struct lima_submit submit = {0};
+	size_t size;
+	int err = 0;
+
+	if (args->pipe >= lima_pipe_num || args->nr_bos == 0)
+		return -EINVAL;
+
+	if (args->flags & ~(LIMA_SUBMIT_FLAG_EXPLICIT_FENCE))
+		return -EINVAL;
+
+	pipe = ldev->pipe + args->pipe;
+	if (args->frame_size != pipe->frame_size)
+		return -EINVAL;
+
+	bos = kvcalloc(args->nr_bos, sizeof(*submit.bos) + sizeof(*submit.lbos), GFP_KERNEL);
+	if (!bos)
+		return -ENOMEM;
+
+	size = args->nr_bos * sizeof(*submit.bos);
+	if (copy_from_user(bos, u64_to_user_ptr(args->bos), size)) {
+		err = -EFAULT;
+		goto out0;
+	}
+
+	task = kmem_cache_zalloc(pipe->task_slab, GFP_KERNEL);
+	if (!task) {
+		err = -ENOMEM;
+		goto out0;
+	}
+
+	task->frame = task + 1;
+	if (copy_from_user(task->frame, u64_to_user_ptr(args->frame), args->frame_size)) {
+		err = -EFAULT;
+		goto out1;
+	}
+
+	err = pipe->task_validate(pipe, task);
+	if (err)
+		goto out1;
+
+	ctx = lima_ctx_get(&priv->ctx_mgr, args->ctx);
+	if (!ctx) {
+		err = -ENOENT;
+		goto out1;
+	}
+
+	submit.pipe = args->pipe;
+	submit.bos = bos;
+	submit.lbos = (void *)bos + size;
+	submit.nr_bos = args->nr_bos;
+	submit.task = task;
+	submit.ctx = ctx;
+	submit.flags = args->flags;
+	submit.in_sync[0] = args->in_sync[0];
+	submit.in_sync[1] = args->in_sync[1];
+	submit.out_sync = args->out_sync;
+
+	err = lima_gem_submit(file, &submit);
+
+	lima_ctx_put(ctx);
+out1:
+	if (err)
+		kmem_cache_free(pipe->task_slab, task);
+out0:
+	kvfree(bos);
+	return err;
+}
+
+static int lima_ioctl_gem_wait(struct drm_device *dev, void *data, struct drm_file *file)
+{
+	struct drm_lima_gem_wait *args = data;
+
+	if (args->op & ~(LIMA_GEM_WAIT_READ|LIMA_GEM_WAIT_WRITE))
+		return -EINVAL;
+
+	return lima_gem_wait(file, args->handle, args->op, args->timeout_ns);
+}
+
+static int lima_ioctl_ctx_create(struct drm_device *dev, void *data, struct drm_file *file)
+{
+	struct drm_lima_ctx_create *args = data;
+	struct lima_drm_priv *priv = file->driver_priv;
+	struct lima_device *ldev = to_lima_dev(dev);
+
+	if (args->_pad)
+		return -EINVAL;
+
+	return lima_ctx_create(ldev, &priv->ctx_mgr, &args->id);
+}
+
+static int lima_ioctl_ctx_free(struct drm_device *dev, void *data, struct drm_file *file)
+{
+	struct drm_lima_ctx_create *args = data;
+	struct lima_drm_priv *priv = file->driver_priv;
+
+	if (args->_pad)
+		return -EINVAL;
+
+	return lima_ctx_free(&priv->ctx_mgr, args->id);
+}
+
+static int lima_drm_driver_open(struct drm_device *dev, struct drm_file *file)
+{
+	int err;
+	struct lima_drm_priv *priv;
+	struct lima_device *ldev = to_lima_dev(dev);
+
+	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	priv->vm = lima_vm_create(ldev);
+	if (!priv->vm) {
+		err = -ENOMEM;
+		goto err_out0;
+	}
+
+	lima_ctx_mgr_init(&priv->ctx_mgr);
+
+	file->driver_priv = priv;
+	return 0;
+
+err_out0:
+	kfree(priv);
+	return err;
+}
+
+static void lima_drm_driver_postclose(struct drm_device *dev, struct drm_file *file)
+{
+	struct lima_drm_priv *priv = file->driver_priv;
+
+	lima_ctx_mgr_fini(&priv->ctx_mgr);
+	lima_vm_put(priv->vm);
+	kfree(priv);
+}
+
+static const struct drm_ioctl_desc lima_drm_driver_ioctls[] = {
+	DRM_IOCTL_DEF_DRV(LIMA_GET_PARAM, lima_ioctl_get_param, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(LIMA_GEM_CREATE, lima_ioctl_gem_create, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(LIMA_GEM_INFO, lima_ioctl_gem_info, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(LIMA_GEM_SUBMIT, lima_ioctl_gem_submit, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(LIMA_GEM_WAIT, lima_ioctl_gem_wait, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(LIMA_CTX_CREATE, lima_ioctl_ctx_create, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(LIMA_CTX_FREE, lima_ioctl_ctx_free, DRM_AUTH|DRM_RENDER_ALLOW),
+};
+
+static const struct file_operations lima_drm_driver_fops = {
+	.owner              = THIS_MODULE,
+	.open               = drm_open,
+	.release            = drm_release,
+	.unlocked_ioctl     = drm_ioctl,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl       = drm_compat_ioctl,
+#endif
+	.mmap               = lima_gem_mmap,
+};
+
+static struct drm_driver lima_drm_driver = {
+	.driver_features    = DRIVER_RENDER | DRIVER_GEM | DRIVER_PRIME | DRIVER_SYNCOBJ,
+	.open               = lima_drm_driver_open,
+	.postclose          = lima_drm_driver_postclose,
+	.ioctls             = lima_drm_driver_ioctls,
+	.num_ioctls         = ARRAY_SIZE(lima_drm_driver_ioctls),
+	.fops               = &lima_drm_driver_fops,
+	.gem_free_object_unlocked = lima_gem_free_object,
+	.gem_open_object    = lima_gem_object_open,
+	.gem_close_object   = lima_gem_object_close,
+	.gem_vm_ops         = &lima_gem_vm_ops,
+	.name               = "lima",
+	.desc               = "lima DRM",
+	.date               = "20190217",
+	.major              = 1,
+	.minor              = 0,
+	.patchlevel         = 0,
+
+	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
+	.gem_prime_import_sg_table = lima_gem_prime_import_sg_table,
+	.prime_handle_to_fd = drm_gem_prime_handle_to_fd,
+	.gem_prime_get_sg_table = lima_gem_prime_get_sg_table,
+	.gem_prime_mmap = lima_gem_prime_mmap,
+};
+
+static int lima_pdev_probe(struct platform_device *pdev)
+{
+	struct lima_device *ldev;
+	struct drm_device *ddev;
+	int err;
+
+	err = lima_sched_slab_init();
+	if (err)
+		return err;
+
+	ldev = devm_kzalloc(&pdev->dev, sizeof(*ldev), GFP_KERNEL);
+	if (!ldev) {
+		err = -ENOMEM;
+		goto err_out0;
+	}
+
+	ldev->pdev = pdev;
+	ldev->dev = &pdev->dev;
+	ldev->id = (enum lima_gpu_id)of_device_get_match_data(&pdev->dev);
+
+	platform_set_drvdata(pdev, ldev);
+
+	/* Allocate and initialize the DRM device. */
+	ddev = drm_dev_alloc(&lima_drm_driver, &pdev->dev);
+	if (IS_ERR(ddev))
+		return PTR_ERR(ddev);
+
+	ddev->dev_private = ldev;
+	ldev->ddev = ddev;
+
+	err = lima_device_init(ldev);
+	if (err) {
+		dev_err(&pdev->dev, "Fatal error during GPU init\n");
+		goto err_out1;
+	}
+
+	/*
+	 * Register the DRM device with the core and the connectors with
+	 * sysfs.
+	 */
+	err = drm_dev_register(ddev, 0);
+	if (err < 0)
+		goto err_out2;
+
+	return 0;
+
+err_out2:
+	lima_device_fini(ldev);
+err_out1:
+	drm_dev_put(ddev);
+err_out0:
+	lima_sched_slab_fini();
+	return err;
+}
+
+static int lima_pdev_remove(struct platform_device *pdev)
+{
+	struct lima_device *ldev = platform_get_drvdata(pdev);
+	struct drm_device *ddev = ldev->ddev;
+
+	drm_dev_unregister(ddev);
+	lima_device_fini(ldev);
+	drm_dev_put(ddev);
+	lima_sched_slab_fini();
+	return 0;
+}
+
+static const struct of_device_id dt_match[] = {
+	{ .compatible = "arm,mali-400", .data = (void *)lima_gpu_mali400 },
+	{ .compatible = "arm,mali-450", .data = (void *)lima_gpu_mali450 },
+	{}
+};
+MODULE_DEVICE_TABLE(of, dt_match);
+
+static struct platform_driver lima_platform_driver = {
+	.probe      = lima_pdev_probe,
+	.remove     = lima_pdev_remove,
+	.driver     = {
+		.name   = "lima",
+		.of_match_table = dt_match,
+	},
+};
+
+static int __init lima_init(void)
+{
+	return platform_driver_register(&lima_platform_driver);
+}
+module_init(lima_init);
+
+static void __exit lima_exit(void)
+{
+	platform_driver_unregister(&lima_platform_driver);
+}
+module_exit(lima_exit);
+
+MODULE_AUTHOR("Lima Project Developers");
+MODULE_DESCRIPTION("Lima DRM Driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/gpu/drm/lima/lima_drv.h b/drivers/gpu/drm/lima/lima_drv.h
new file mode 100644
index 000000000000..69c7344715c9
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_drv.h
@@ -0,0 +1,45 @@
+/* SPDX-License-Identifier: GPL-2.0 OR MIT */
+/* Copyright 2017-2019 Qiang Yu <yuq825@gmail.com> */
+
+#ifndef __LIMA_DRV_H__
+#define __LIMA_DRV_H__
+
+#include <drm/drm_file.h>
+
+#include "lima_ctx.h"
+
+extern int lima_sched_timeout_ms;
+
+struct lima_vm;
+struct lima_bo;
+struct lima_sched_task;
+
+struct drm_lima_gem_submit_bo;
+
+struct lima_drm_priv {
+	struct lima_vm *vm;
+	struct lima_ctx_mgr ctx_mgr;
+};
+
+struct lima_submit {
+	struct lima_ctx *ctx;
+	int pipe;
+	u32 flags;
+
+	struct drm_lima_gem_submit_bo *bos;
+	struct lima_bo **lbos;
+	u32 nr_bos;
+
+	u32 in_sync[2];
+	u32 out_sync;
+
+	struct lima_sched_task *task;
+};
+
+static inline struct lima_drm_priv *
+to_lima_drm_priv(struct drm_file *file)
+{
+	return file->driver_priv;
+}
+
+#endif
diff --git a/drivers/gpu/drm/lima/lima_gem.c b/drivers/gpu/drm/lima/lima_gem.c
new file mode 100644
index 000000000000..477c0f766663
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_gem.c
@@ -0,0 +1,349 @@
+// SPDX-License-Identifier: GPL-2.0 OR MIT
+/* Copyright 2017-2019 Qiang Yu <yuq825@gmail.com> */
+
+#include <linux/mm.h>
+#include <linux/sync_file.h>
+#include <linux/pfn_t.h>
+
+#include <drm/drm_file.h>
+#include <drm/drm_syncobj.h>
+#include <drm/drm_utils.h>
+
+#include <drm/lima_drm.h>
+
+#include "lima_drv.h"
+#include "lima_gem.h"
+#include "lima_gem_prime.h"
+#include "lima_vm.h"
+#include "lima_object.h"
+
+int lima_gem_create_handle(struct drm_device *dev, struct drm_file *file,
+			   u32 size, u32 flags, u32 *handle)
+{
+	int err;
+	struct lima_bo *bo;
+	struct lima_device *ldev = to_lima_dev(dev);
+
+	bo = lima_bo_create(ldev, size, flags, NULL, NULL);
+	if (IS_ERR(bo))
+		return PTR_ERR(bo);
+
+	err = drm_gem_handle_create(file, &bo->gem, handle);
+
+	/* drop reference from allocate - handle holds it now */
+	drm_gem_object_put_unlocked(&bo->gem);
+
+	return err;
+}
+
+void lima_gem_free_object(struct drm_gem_object *obj)
+{
+	struct lima_bo *bo = to_lima_bo(obj);
+
+	if (!list_empty(&bo->va))
+		dev_err(obj->dev->dev, "lima gem free bo still has va\n");
+
+	lima_bo_destroy(bo);
+}
+
+int lima_gem_object_open(struct drm_gem_object *obj, struct drm_file *file)
+{
+	struct lima_bo *bo = to_lima_bo(obj);
+	struct lima_drm_priv *priv = to_lima_drm_priv(file);
+	struct lima_vm *vm = priv->vm;
+
+	return lima_vm_bo_add(vm, bo, true);
+}
+
+void lima_gem_object_close(struct drm_gem_object *obj, struct drm_file *file)
+{
+	struct lima_bo *bo = to_lima_bo(obj);
+	struct lima_drm_priv *priv = to_lima_drm_priv(file);
+	struct lima_vm *vm = priv->vm;
+
+	lima_vm_bo_del(vm, bo);
+}
+
+int lima_gem_get_info(struct drm_file *file, u32 handle, u32 *va, u64 *offset)
+{
+	struct drm_gem_object *obj;
+	struct lima_bo *bo;
+	struct lima_drm_priv *priv = to_lima_drm_priv(file);
+	struct lima_vm *vm = priv->vm;
+	int err;
+
+	obj = drm_gem_object_lookup(file, handle);
+	if (!obj)
+		return -ENOENT;
+
+	bo = to_lima_bo(obj);
+
+	*va = lima_vm_get_va(vm, bo);
+
+	err = drm_gem_create_mmap_offset(obj);
+	if (!err)
+		*offset = drm_vma_node_offset_addr(&obj->vma_node);
+
+	drm_gem_object_put_unlocked(obj);
+	return err;
+}
+
+static vm_fault_t lima_gem_fault(struct vm_fault *vmf)
+{
+	struct vm_area_struct *vma = vmf->vma;
+	struct drm_gem_object *obj = vma->vm_private_data;
+	struct lima_bo *bo = to_lima_bo(obj);
+	pfn_t pfn;
+	pgoff_t pgoff;
+
+	/* We don't use vmf->pgoff since that has the fake offset: */
+	pgoff = (vmf->address - vma->vm_start) >> PAGE_SHIFT;
+	pfn = __pfn_to_pfn_t(page_to_pfn(bo->pages[pgoff]), PFN_DEV);
+
+	return vmf_insert_mixed(vma, vmf->address, pfn);
+}
+
+const struct vm_operations_struct lima_gem_vm_ops = {
+	.fault = lima_gem_fault,
+	.open = drm_gem_vm_open,
+	.close = drm_gem_vm_close,
+};
+
+void lima_set_vma_flags(struct vm_area_struct *vma)
+{
+	pgprot_t prot = vm_get_page_prot(vma->vm_flags);
+
+	vma->vm_flags |= VM_MIXEDMAP;
+	vma->vm_flags &= ~VM_PFNMAP;
+	vma->vm_page_prot = pgprot_writecombine(prot);
+}
+
+int lima_gem_mmap(struct file *filp, struct vm_area_struct *vma)
+{
+	int ret;
+
+	ret = drm_gem_mmap(filp, vma);
+	if (ret)
+		return ret;
+
+	lima_set_vma_flags(vma);
+	return 0;
+}
+
+static int lima_gem_sync_bo(struct lima_sched_task *task, struct lima_bo *bo,
+			    bool write, bool explicit)
+{
+	int err = 0;
+
+	if (!write) {
+		err = reservation_object_reserve_shared(bo->gem.resv, 1);
+		if (err)
+			return err;
+	}
+
+	/* explicit sync use user passed dep fence */
+	if (explicit)
+		return 0;
+
+	return drm_gem_fence_array_add_implicit(&task->deps, &bo->gem, write);
+}
+
+static int lima_gem_lock_bos(struct lima_bo **bos, u32 nr_bos,
+			     struct ww_acquire_ctx *ctx)
+{
+	int i, ret = 0, contended, slow_locked = -1;
+
+	ww_acquire_init(ctx, &reservation_ww_class);
+
+retry:
+	for (i = 0; i < nr_bos; i++) {
+		if (i == slow_locked) {
+			slow_locked = -1;
+			continue;
+		}
+
+		ret = ww_mutex_lock_interruptible(&bos[i]->gem.resv->lock, ctx);
+		if (ret < 0) {
+			contended = i;
+			goto err;
+		}
+	}
+
+	ww_acquire_done(ctx);
+	return 0;
+
+err:
+	for (i--; i >= 0; i--)
+		ww_mutex_unlock(&bos[i]->gem.resv->lock);
+
+	if (slow_locked >= 0)
+		ww_mutex_unlock(&bos[slow_locked]->gem.resv->lock);
+
+	if (ret == -EDEADLK) {
+		/* we lost out in a seqno race, lock and retry.. */
+		ret = ww_mutex_lock_slow_interruptible(
+			&bos[contended]->gem.resv->lock, ctx);
+		if (!ret) {
+			slow_locked = contended;
+			goto retry;
+		}
+	}
+	ww_acquire_fini(ctx);
+
+	return ret;
+}
+
+static void lima_gem_unlock_bos(struct lima_bo **bos, u32 nr_bos,
+				struct ww_acquire_ctx *ctx)
+{
+	int i;
+
+	for (i = 0; i < nr_bos; i++)
+		ww_mutex_unlock(&bos[i]->gem.resv->lock);
+	ww_acquire_fini(ctx);
+}
+
+static int lima_gem_add_deps(struct drm_file *file, struct lima_submit *submit)
+{
+	int i, err;
+
+	for (i = 0; i < ARRAY_SIZE(submit->in_sync); i++) {
+		struct dma_fence *fence = NULL;
+
+		if (!submit->in_sync[i])
+			continue;
+
+		err = drm_syncobj_find_fence(file, submit->in_sync[i],
+					     0, 0, &fence);
+		if (err)
+			return err;
+
+		err = drm_gem_fence_array_add(&submit->task->deps, fence);
+		if (err) {
+			dma_fence_put(fence);
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+int lima_gem_submit(struct drm_file *file, struct lima_submit *submit)
+{
+	int i, err = 0;
+	struct ww_acquire_ctx ctx;
+	struct lima_drm_priv *priv = to_lima_drm_priv(file);
+	struct lima_vm *vm = priv->vm;
+	struct drm_syncobj *out_sync = NULL;
+	struct dma_fence *fence;
+	struct lima_bo **bos = submit->lbos;
+
+	if (submit->out_sync) {
+		out_sync = drm_syncobj_find(file, submit->out_sync);
+		if (!out_sync)
+			return -ENOENT;
+	}
+
+	for (i = 0; i < submit->nr_bos; i++) {
+		struct drm_gem_object *obj;
+		struct lima_bo *bo;
+
+		obj = drm_gem_object_lookup(file, submit->bos[i].handle);
+		if (!obj) {
+			err = -ENOENT;
+			goto err_out0;
+		}
+
+		bo = to_lima_bo(obj);
+
+		/* increase refcnt of gpu va map to prevent unmapped when executing,
+		 * will be decreased when task done
+		 */
+		err = lima_vm_bo_add(vm, bo, false);
+		if (err) {
+			drm_gem_object_put_unlocked(obj);
+			goto err_out0;
+		}
+
+		bos[i] = bo;
+	}
+
+	err = lima_gem_lock_bos(bos, submit->nr_bos, &ctx);
+	if (err)
+		goto err_out0;
+
+	err = lima_sched_task_init(
+		submit->task, submit->ctx->context + submit->pipe,
+		bos, submit->nr_bos, vm);
+	if (err)
+		goto err_out1;
+
+	err = lima_gem_add_deps(file, submit);
+	if (err)
+		goto err_out2;
+
+	for (i = 0; i < submit->nr_bos; i++) {
+		err = lima_gem_sync_bo(
+			submit->task, bos[i],
+			submit->bos[i].flags & LIMA_SUBMIT_BO_WRITE,
+			submit->flags & LIMA_SUBMIT_FLAG_EXPLICIT_FENCE);
+		if (err)
+			goto err_out2;
+	}
+
+	fence = lima_sched_context_queue_task(
+		submit->ctx->context + submit->pipe, submit->task);
+
+	for (i = 0; i < submit->nr_bos; i++) {
+		if (submit->bos[i].flags & LIMA_SUBMIT_BO_WRITE)
+			reservation_object_add_excl_fence(bos[i]->gem.resv, fence);
+		else
+			reservation_object_add_shared_fence(bos[i]->gem.resv, fence);
+	}
+
+	lima_gem_unlock_bos(bos, submit->nr_bos, &ctx);
+
+	for (i = 0; i < submit->nr_bos; i++)
+		drm_gem_object_put_unlocked(&bos[i]->gem);
+
+	if (out_sync) {
+		drm_syncobj_replace_fence(out_sync, fence);
+		drm_syncobj_put(out_sync);
+	}
+
+	dma_fence_put(fence);
+
+	return 0;
+
+err_out2:
+	lima_sched_task_fini(submit->task);
+err_out1:
+	lima_gem_unlock_bos(bos, submit->nr_bos, &ctx);
+err_out0:
+	for (i = 0; i < submit->nr_bos; i++) {
+		if (!bos[i])
+			break;
+		lima_vm_bo_del(vm, bos[i]);
+		drm_gem_object_put_unlocked(&bos[i]->gem);
+	}
+	if (out_sync)
+		drm_syncobj_put(out_sync);
+	return err;
+}
+
+int lima_gem_wait(struct drm_file *file, u32 handle, u32 op, s64 timeout_ns)
+{
+	bool write = op & LIMA_GEM_WAIT_WRITE;
+	long ret, timeout;
+
+	if (!op)
+		return 0;
+
+	timeout = drm_timeout_abs_to_jiffies(timeout_ns);
+
+	ret = drm_gem_reservation_object_wait(file, handle, write, timeout);
+	if (ret == 0)
+		ret = timeout ? -ETIMEDOUT : -EBUSY;
+
+	return ret;
+}
diff --git a/drivers/gpu/drm/lima/lima_gem.h b/drivers/gpu/drm/lima/lima_gem.h
new file mode 100644
index 000000000000..556111a01135
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_gem.h
@@ -0,0 +1,25 @@
+/* SPDX-License-Identifier: GPL-2.0 OR MIT */
+/* Copyright 2017-2019 Qiang Yu <yuq825@gmail.com> */
+
+#ifndef __LIMA_GEM_H__
+#define __LIMA_GEM_H__
+
+struct lima_bo;
+struct lima_submit;
+
+extern const struct vm_operations_struct lima_gem_vm_ops;
+
+struct lima_bo *lima_gem_create_bo(struct drm_device *dev, u32 size, u32 flags);
+int lima_gem_create_handle(struct drm_device *dev, struct drm_file *file,
+			   u32 size, u32 flags, u32 *handle);
+void lima_gem_free_object(struct drm_gem_object *obj);
+int lima_gem_object_open(struct drm_gem_object *obj, struct drm_file *file);
+void lima_gem_object_close(struct drm_gem_object *obj, struct drm_file *file);
+int lima_gem_get_info(struct drm_file *file, u32 handle, u32 *va, u64 *offset);
+int lima_gem_mmap(struct file *filp, struct vm_area_struct *vma);
+int lima_gem_submit(struct drm_file *file, struct lima_submit *submit);
+int lima_gem_wait(struct drm_file *file, u32 handle, u32 op, s64 timeout_ns);
+
+void lima_set_vma_flags(struct vm_area_struct *vma);
+
+#endif
diff --git a/drivers/gpu/drm/lima/lima_gem_prime.c b/drivers/gpu/drm/lima/lima_gem_prime.c
new file mode 100644
index 000000000000..9c6d9f1dba55
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_gem_prime.c
@@ -0,0 +1,47 @@
+// SPDX-License-Identifier: GPL-2.0 OR MIT
+/* Copyright 2018-2019 Qiang Yu <yuq825@gmail.com> */
+
+#include <linux/dma-buf.h>
+#include <drm/drm_prime.h>
+#include <drm/drm_drv.h>
+#include <drm/drm_file.h>
+
+#include "lima_device.h"
+#include "lima_object.h"
+#include "lima_gem.h"
+#include "lima_gem_prime.h"
+
+struct drm_gem_object *lima_gem_prime_import_sg_table(
+	struct drm_device *dev, struct dma_buf_attachment *attach,
+	struct sg_table *sgt)
+{
+	struct lima_device *ldev = to_lima_dev(dev);
+	struct lima_bo *bo;
+
+	bo = lima_bo_create(ldev, attach->dmabuf->size, 0, sgt,
+			    attach->dmabuf->resv);
+	if (IS_ERR(bo))
+		return ERR_CAST(bo);
+
+	return &bo->gem;
+}
+
+struct sg_table *lima_gem_prime_get_sg_table(struct drm_gem_object *obj)
+{
+	struct lima_bo *bo = to_lima_bo(obj);
+	int npages = obj->size >> PAGE_SHIFT;
+
+	return drm_prime_pages_to_sg(bo->pages, npages);
+}
+
+int lima_gem_prime_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma)
+{
+	int ret;
+
+	ret = drm_gem_mmap_obj(obj, obj->size, vma);
+	if (ret)
+		return ret;
+
+	lima_set_vma_flags(vma);
+	return 0;
+}
diff --git a/drivers/gpu/drm/lima/lima_gem_prime.h b/drivers/gpu/drm/lima/lima_gem_prime.h
new file mode 100644
index 000000000000..34b4d35c21e3
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_gem_prime.h
@@ -0,0 +1,13 @@
+/* SPDX-License-Identifier: GPL-2.0 OR MIT */
+/* Copyright 2018-2019 Qiang Yu <yuq825@gmail.com> */
+
+#ifndef __LIMA_GEM_PRIME_H__
+#define __LIMA_GEM_PRIME_H__
+
+struct drm_gem_object *lima_gem_prime_import_sg_table(
+	struct drm_device *dev, struct dma_buf_attachment *attach,
+	struct sg_table *sgt);
+struct sg_table *lima_gem_prime_get_sg_table(struct drm_gem_object *obj);
+int lima_gem_prime_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma);
+
+#endif
diff --git a/drivers/gpu/drm/lima/lima_gp.c b/drivers/gpu/drm/lima/lima_gp.c
new file mode 100644
index 000000000000..ccf49faedebf
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_gp.c
@@ -0,0 +1,283 @@
+// SPDX-License-Identifier: GPL-2.0 OR MIT
+/* Copyright 2017-2019 Qiang Yu <yuq825@gmail.com> */
+
+#include <linux/interrupt.h>
+#include <linux/iopoll.h>
+#include <linux/device.h>
+#include <linux/slab.h>
+
+#include <drm/lima_drm.h>
+
+#include "lima_device.h"
+#include "lima_gp.h"
+#include "lima_regs.h"
+
+#define gp_write(reg, data) writel(data, ip->iomem + reg)
+#define gp_read(reg) readl(ip->iomem + reg)
+
+static irqreturn_t lima_gp_irq_handler(int irq, void *data)
+{
+	struct lima_ip *ip = data;
+	struct lima_device *dev = ip->dev;
+	struct lima_sched_pipe *pipe = dev->pipe + lima_pipe_gp;
+	u32 state = gp_read(LIMA_GP_INT_STAT);
+	u32 status = gp_read(LIMA_GP_STATUS);
+	bool done = false;
+
+	/* for shared irq case */
+	if (!state)
+		return IRQ_NONE;
+
+	if (state & LIMA_GP_IRQ_MASK_ERROR) {
+		dev_err(dev->dev, "gp error irq state=%x status=%x\n",
+			state, status);
+
+		/* mask all interrupts before hard reset */
+		gp_write(LIMA_GP_INT_MASK, 0);
+
+		pipe->error = true;
+		done = true;
+	} else {
+		bool valid = state & (LIMA_GP_IRQ_VS_END_CMD_LST |
+				      LIMA_GP_IRQ_PLBU_END_CMD_LST);
+		bool active = status & (LIMA_GP_STATUS_VS_ACTIVE |
+					LIMA_GP_STATUS_PLBU_ACTIVE);
+		done = valid && !active;
+	}
+
+	gp_write(LIMA_GP_INT_CLEAR, state);
+
+	if (done)
+		lima_sched_pipe_task_done(pipe);
+
+	return IRQ_HANDLED;
+}
+
+static void lima_gp_soft_reset_async(struct lima_ip *ip)
+{
+	if (ip->data.async_reset)
+		return;
+
+	gp_write(LIMA_GP_INT_MASK, 0);
+	gp_write(LIMA_GP_INT_CLEAR, LIMA_GP_IRQ_RESET_COMPLETED);
+	gp_write(LIMA_GP_CMD, LIMA_GP_CMD_SOFT_RESET);
+	ip->data.async_reset = true;
+}
+
+static int lima_gp_soft_reset_async_wait(struct lima_ip *ip)
+{
+	struct lima_device *dev = ip->dev;
+	int err;
+	u32 v;
+
+	if (!ip->data.async_reset)
+		return 0;
+
+	err = readl_poll_timeout(ip->iomem + LIMA_GP_INT_RAWSTAT, v,
+				 v & LIMA_GP_IRQ_RESET_COMPLETED,
+				 0, 100);
+	if (err) {
+		dev_err(dev->dev, "gp soft reset time out\n");
+		return err;
+	}
+
+	gp_write(LIMA_GP_INT_CLEAR, LIMA_GP_IRQ_MASK_ALL);
+	gp_write(LIMA_GP_INT_MASK, LIMA_GP_IRQ_MASK_USED);
+
+	ip->data.async_reset = false;
+	return 0;
+}
+
+static int lima_gp_task_validate(struct lima_sched_pipe *pipe,
+				 struct lima_sched_task *task)
+{
+	struct drm_lima_gp_frame *frame = task->frame;
+	u32 *f = frame->frame;
+	(void)pipe;
+
+	if (f[LIMA_GP_VSCL_START_ADDR >> 2] >
+	    f[LIMA_GP_VSCL_END_ADDR >> 2] ||
+	    f[LIMA_GP_PLBUCL_START_ADDR >> 2] >
+	    f[LIMA_GP_PLBUCL_END_ADDR >> 2] ||
+	    f[LIMA_GP_PLBU_ALLOC_START_ADDR >> 2] >
+	    f[LIMA_GP_PLBU_ALLOC_END_ADDR >> 2])
+		return -EINVAL;
+
+	if (f[LIMA_GP_VSCL_START_ADDR >> 2] ==
+	    f[LIMA_GP_VSCL_END_ADDR >> 2] &&
+	    f[LIMA_GP_PLBUCL_START_ADDR >> 2] ==
+	    f[LIMA_GP_PLBUCL_END_ADDR >> 2])
+		return -EINVAL;
+
+	return 0;
+}
+
+static void lima_gp_task_run(struct lima_sched_pipe *pipe,
+			     struct lima_sched_task *task)
+{
+	struct lima_ip *ip = pipe->processor[0];
+	struct drm_lima_gp_frame *frame = task->frame;
+	u32 *f = frame->frame;
+	u32 cmd = 0;
+	int i;
+
+	if (f[LIMA_GP_VSCL_START_ADDR >> 2] !=
+	    f[LIMA_GP_VSCL_END_ADDR >> 2])
+		cmd |= LIMA_GP_CMD_START_VS;
+	if (f[LIMA_GP_PLBUCL_START_ADDR >> 2] !=
+	    f[LIMA_GP_PLBUCL_END_ADDR >> 2])
+		cmd |= LIMA_GP_CMD_START_PLBU;
+
+	/* before any hw ops, wait last success task async soft reset */
+	lima_gp_soft_reset_async_wait(ip);
+
+	for (i = 0; i < LIMA_GP_FRAME_REG_NUM; i++)
+		writel(f[i], ip->iomem + LIMA_GP_VSCL_START_ADDR + i * 4);
+
+	gp_write(LIMA_GP_CMD, LIMA_GP_CMD_UPDATE_PLBU_ALLOC);
+	gp_write(LIMA_GP_CMD, cmd);
+}
+
+static int lima_gp_hard_reset_poll(struct lima_ip *ip)
+{
+	gp_write(LIMA_GP_PERF_CNT_0_LIMIT, 0xC01A0000);
+	return gp_read(LIMA_GP_PERF_CNT_0_LIMIT) == 0xC01A0000;
+}
+
+static int lima_gp_hard_reset(struct lima_ip *ip)
+{
+	struct lima_device *dev = ip->dev;
+	int ret;
+
+	gp_write(LIMA_GP_PERF_CNT_0_LIMIT, 0xC0FFE000);
+	gp_write(LIMA_GP_INT_MASK, 0);
+	gp_write(LIMA_GP_CMD, LIMA_GP_CMD_RESET);
+	ret = lima_poll_timeout(ip, lima_gp_hard_reset_poll, 10, 100);
+	if (ret) {
+		dev_err(dev->dev, "gp hard reset timeout\n");
+		return ret;
+	}
+
+	gp_write(LIMA_GP_PERF_CNT_0_LIMIT, 0);
+	gp_write(LIMA_GP_INT_CLEAR, LIMA_GP_IRQ_MASK_ALL);
+	gp_write(LIMA_GP_INT_MASK, LIMA_GP_IRQ_MASK_USED);
+	return 0;
+}
+
+static void lima_gp_task_fini(struct lima_sched_pipe *pipe)
+{
+	lima_gp_soft_reset_async(pipe->processor[0]);
+}
+
+static void lima_gp_task_error(struct lima_sched_pipe *pipe)
+{
+	struct lima_ip *ip = pipe->processor[0];
+
+	dev_err(ip->dev->dev, "gp task error int_state=%x status=%x\n",
+		gp_read(LIMA_GP_INT_STAT), gp_read(LIMA_GP_STATUS));
+
+	lima_gp_hard_reset(ip);
+}
+
+static void lima_gp_task_mmu_error(struct lima_sched_pipe *pipe)
+{
+	lima_sched_pipe_task_done(pipe);
+}
+
+static void lima_gp_print_version(struct lima_ip *ip)
+{
+	u32 version, major, minor;
+	char *name;
+
+	version = gp_read(LIMA_GP_VERSION);
+	major = (version >> 8) & 0xFF;
+	minor = version & 0xFF;
+	switch (version >> 16) {
+	case 0xA07:
+	    name = "mali200";
+		break;
+	case 0xC07:
+		name = "mali300";
+		break;
+	case 0xB07:
+		name = "mali400";
+		break;
+	case 0xD07:
+		name = "mali450";
+		break;
+	default:
+		name = "unknown";
+		break;
+	}
+	dev_info(ip->dev->dev, "%s - %s version major %d minor %d\n",
+		 lima_ip_name(ip), name, major, minor);
+}
+
+static struct kmem_cache *lima_gp_task_slab;
+static int lima_gp_task_slab_refcnt;
+
+int lima_gp_init(struct lima_ip *ip)
+{
+	struct lima_device *dev = ip->dev;
+	int err;
+
+	lima_gp_print_version(ip);
+
+	ip->data.async_reset = false;
+	lima_gp_soft_reset_async(ip);
+	err = lima_gp_soft_reset_async_wait(ip);
+	if (err)
+		return err;
+
+	err = devm_request_irq(dev->dev, ip->irq, lima_gp_irq_handler,
+			       IRQF_SHARED, lima_ip_name(ip), ip);
+	if (err) {
+		dev_err(dev->dev, "gp %s fail to request irq\n",
+			lima_ip_name(ip));
+		return err;
+	}
+
+	dev->gp_version = gp_read(LIMA_GP_VERSION);
+
+	return 0;
+}
+
+void lima_gp_fini(struct lima_ip *ip)
+{
+
+}
+
+int lima_gp_pipe_init(struct lima_device *dev)
+{
+	int frame_size = sizeof(struct drm_lima_gp_frame);
+	struct lima_sched_pipe *pipe = dev->pipe + lima_pipe_gp;
+
+	if (!lima_gp_task_slab) {
+		lima_gp_task_slab = kmem_cache_create_usercopy(
+			"lima_gp_task", sizeof(struct lima_sched_task) + frame_size,
+			0, SLAB_HWCACHE_ALIGN, sizeof(struct lima_sched_task),
+			frame_size, NULL);
+		if (!lima_gp_task_slab)
+			return -ENOMEM;
+	}
+	lima_gp_task_slab_refcnt++;
+
+	pipe->frame_size = frame_size;
+	pipe->task_slab = lima_gp_task_slab;
+
+	pipe->task_validate = lima_gp_task_validate;
+	pipe->task_run = lima_gp_task_run;
+	pipe->task_fini = lima_gp_task_fini;
+	pipe->task_error = lima_gp_task_error;
+	pipe->task_mmu_error = lima_gp_task_mmu_error;
+
+	return 0;
+}
+
+void lima_gp_pipe_fini(struct lima_device *dev)
+{
+	if (!--lima_gp_task_slab_refcnt) {
+		kmem_cache_destroy(lima_gp_task_slab);
+		lima_gp_task_slab = NULL;
+	}
+}
diff --git a/drivers/gpu/drm/lima/lima_gp.h b/drivers/gpu/drm/lima/lima_gp.h
new file mode 100644
index 000000000000..516e5c1babbb
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_gp.h
@@ -0,0 +1,16 @@
+/* SPDX-License-Identifier: GPL-2.0 OR MIT */
+/* Copyright 2017-2019 Qiang Yu <yuq825@gmail.com> */
+
+#ifndef __LIMA_GP_H__
+#define __LIMA_GP_H__
+
+struct lima_ip;
+struct lima_device;
+
+int lima_gp_init(struct lima_ip *ip);
+void lima_gp_fini(struct lima_ip *ip);
+
+int lima_gp_pipe_init(struct lima_device *dev);
+void lima_gp_pipe_fini(struct lima_device *dev);
+
+#endif
diff --git a/drivers/gpu/drm/lima/lima_l2_cache.c b/drivers/gpu/drm/lima/lima_l2_cache.c
new file mode 100644
index 000000000000..6873a7af5a5c
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_l2_cache.c
@@ -0,0 +1,80 @@
+// SPDX-License-Identifier: GPL-2.0 OR MIT
+/* Copyright 2017-2019 Qiang Yu <yuq825@gmail.com> */
+
+#include <linux/iopoll.h>
+#include <linux/device.h>
+
+#include "lima_device.h"
+#include "lima_l2_cache.h"
+#include "lima_regs.h"
+
+#define l2_cache_write(reg, data) writel(data, ip->iomem + reg)
+#define l2_cache_read(reg) readl(ip->iomem + reg)
+
+static int lima_l2_cache_wait_idle(struct lima_ip *ip)
+{
+	struct lima_device *dev = ip->dev;
+	int err;
+	u32 v;
+
+	err = readl_poll_timeout(ip->iomem + LIMA_L2_CACHE_STATUS, v,
+				 !(v & LIMA_L2_CACHE_STATUS_COMMAND_BUSY),
+				 0, 1000);
+	if (err) {
+		dev_err(dev->dev, "l2 cache wait command timeout\n");
+		return err;
+	}
+	return 0;
+}
+
+int lima_l2_cache_flush(struct lima_ip *ip)
+{
+	int ret;
+
+	spin_lock(&ip->data.lock);
+	l2_cache_write(LIMA_L2_CACHE_COMMAND, LIMA_L2_CACHE_COMMAND_CLEAR_ALL);
+	ret = lima_l2_cache_wait_idle(ip);
+	spin_unlock(&ip->data.lock);
+	return ret;
+}
+
+int lima_l2_cache_init(struct lima_ip *ip)
+{
+	int i, err;
+	u32 size;
+	struct lima_device *dev = ip->dev;
+
+	/* l2_cache2 only exists when one of PP4-7 present */
+	if (ip->id == lima_ip_l2_cache2) {
+		for (i = lima_ip_pp4; i <= lima_ip_pp7; i++) {
+			if (dev->ip[i].present)
+				break;
+		}
+		if (i > lima_ip_pp7)
+			return -ENODEV;
+	}
+
+	spin_lock_init(&ip->data.lock);
+
+	size = l2_cache_read(LIMA_L2_CACHE_SIZE);
+	dev_info(dev->dev, "l2 cache %uK, %u-way, %ubyte cache line, %ubit external bus\n",
+		 1 << (((size >> 16) & 0xff) - 10),
+		 1 << ((size >> 8) & 0xff),
+		 1 << (size & 0xff),
+		 1 << ((size >> 24) & 0xff));
+
+	err = lima_l2_cache_flush(ip);
+	if (err)
+		return err;
+
+	l2_cache_write(LIMA_L2_CACHE_ENABLE,
+		       LIMA_L2_CACHE_ENABLE_ACCESS|LIMA_L2_CACHE_ENABLE_READ_ALLOCATE);
+	l2_cache_write(LIMA_L2_CACHE_MAX_READS, 0x1c);
+
+	return 0;
+}
+
+void lima_l2_cache_fini(struct lima_ip *ip)
+{
+
+}
diff --git a/drivers/gpu/drm/lima/lima_l2_cache.h b/drivers/gpu/drm/lima/lima_l2_cache.h
new file mode 100644
index 000000000000..c63fb676ff14
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_l2_cache.h
@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: GPL-2.0 OR MIT */
+/* Copyright 2017-2019 Qiang Yu <yuq825@gmail.com> */
+
+#ifndef __LIMA_L2_CACHE_H__
+#define __LIMA_L2_CACHE_H__
+
+struct lima_ip;
+
+int lima_l2_cache_init(struct lima_ip *ip);
+void lima_l2_cache_fini(struct lima_ip *ip);
+
+int lima_l2_cache_flush(struct lima_ip *ip);
+
+#endif
diff --git a/drivers/gpu/drm/lima/lima_mmu.c b/drivers/gpu/drm/lima/lima_mmu.c
new file mode 100644
index 000000000000..8e1651d6a61f
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_mmu.c
@@ -0,0 +1,142 @@
+// SPDX-License-Identifier: GPL-2.0 OR MIT
+/* Copyright 2017-2019 Qiang Yu <yuq825@gmail.com> */
+
+#include <linux/interrupt.h>
+#include <linux/iopoll.h>
+#include <linux/device.h>
+
+#include "lima_device.h"
+#include "lima_mmu.h"
+#include "lima_vm.h"
+#include "lima_object.h"
+#include "lima_regs.h"
+
+#define mmu_write(reg, data) writel(data, ip->iomem + reg)
+#define mmu_read(reg) readl(ip->iomem + reg)
+
+#define lima_mmu_send_command(cmd, addr, val, cond)	     \
+({							     \
+	int __ret;					     \
+							     \
+	mmu_write(LIMA_MMU_COMMAND, cmd);		     \
+	__ret = readl_poll_timeout(ip->iomem + (addr), val,  \
+				  cond, 0, 100);	     \
+	if (__ret)					     \
+		dev_err(dev->dev,			     \
+			"mmu command %x timeout\n", cmd);    \
+	__ret;						     \
+})
+
+static irqreturn_t lima_mmu_irq_handler(int irq, void *data)
+{
+	struct lima_ip *ip = data;
+	struct lima_device *dev = ip->dev;
+	u32 status = mmu_read(LIMA_MMU_INT_STATUS);
+	struct lima_sched_pipe *pipe;
+
+	/* for shared irq case */
+	if (!status)
+		return IRQ_NONE;
+
+	if (status & LIMA_MMU_INT_PAGE_FAULT) {
+		u32 fault = mmu_read(LIMA_MMU_PAGE_FAULT_ADDR);
+
+		dev_err(dev->dev, "mmu page fault at 0x%x from bus id %d of type %s on %s\n",
+			fault, LIMA_MMU_STATUS_BUS_ID(status),
+			status & LIMA_MMU_STATUS_PAGE_FAULT_IS_WRITE ? "write" : "read",
+			lima_ip_name(ip));
+	}
+
+	if (status & LIMA_MMU_INT_READ_BUS_ERROR)
+		dev_err(dev->dev, "mmu %s irq bus error\n", lima_ip_name(ip));
+
+	/* mask all interrupts before resume */
+	mmu_write(LIMA_MMU_INT_MASK, 0);
+	mmu_write(LIMA_MMU_INT_CLEAR, status);
+
+	pipe = dev->pipe + (ip->id == lima_ip_gpmmu ? lima_pipe_gp : lima_pipe_pp);
+	lima_sched_pipe_mmu_error(pipe);
+
+	return IRQ_HANDLED;
+}
+
+int lima_mmu_init(struct lima_ip *ip)
+{
+	struct lima_device *dev = ip->dev;
+	int err;
+	u32 v;
+
+	if (ip->id == lima_ip_ppmmu_bcast)
+		return 0;
+
+	mmu_write(LIMA_MMU_DTE_ADDR, 0xCAFEBABE);
+	if (mmu_read(LIMA_MMU_DTE_ADDR) != 0xCAFEB000) {
+		dev_err(dev->dev, "mmu %s dte write test fail\n", lima_ip_name(ip));
+		return -EIO;
+	}
+
+	mmu_write(LIMA_MMU_COMMAND, LIMA_MMU_COMMAND_HARD_RESET);
+	err = lima_mmu_send_command(LIMA_MMU_COMMAND_HARD_RESET,
+				    LIMA_MMU_DTE_ADDR, v, v == 0);
+	if (err)
+		return err;
+
+	err = devm_request_irq(dev->dev, ip->irq, lima_mmu_irq_handler,
+			       IRQF_SHARED, lima_ip_name(ip), ip);
+	if (err) {
+		dev_err(dev->dev, "mmu %s fail to request irq\n", lima_ip_name(ip));
+		return err;
+	}
+
+	mmu_write(LIMA_MMU_INT_MASK, LIMA_MMU_INT_PAGE_FAULT | LIMA_MMU_INT_READ_BUS_ERROR);
+	mmu_write(LIMA_MMU_DTE_ADDR, dev->empty_vm->pd.dma);
+	return lima_mmu_send_command(LIMA_MMU_COMMAND_ENABLE_PAGING,
+				     LIMA_MMU_STATUS, v,
+				     v & LIMA_MMU_STATUS_PAGING_ENABLED);
+}
+
+void lima_mmu_fini(struct lima_ip *ip)
+{
+
+}
+
+void lima_mmu_switch_vm(struct lima_ip *ip, struct lima_vm *vm)
+{
+	struct lima_device *dev = ip->dev;
+	u32 v;
+
+	lima_mmu_send_command(LIMA_MMU_COMMAND_ENABLE_STALL,
+			      LIMA_MMU_STATUS, v,
+			      v & LIMA_MMU_STATUS_STALL_ACTIVE);
+
+	if (vm)
+		mmu_write(LIMA_MMU_DTE_ADDR, vm->pd.dma);
+
+	/* flush the TLB */
+	mmu_write(LIMA_MMU_COMMAND, LIMA_MMU_COMMAND_ZAP_CACHE);
+
+	lima_mmu_send_command(LIMA_MMU_COMMAND_DISABLE_STALL,
+			      LIMA_MMU_STATUS, v,
+			      !(v & LIMA_MMU_STATUS_STALL_ACTIVE));
+}
+
+void lima_mmu_page_fault_resume(struct lima_ip *ip)
+{
+	struct lima_device *dev = ip->dev;
+	u32 status = mmu_read(LIMA_MMU_STATUS);
+	u32 v;
+
+	if (status & LIMA_MMU_STATUS_PAGE_FAULT_ACTIVE) {
+		dev_info(dev->dev, "mmu resume\n");
+
+		mmu_write(LIMA_MMU_INT_MASK, 0);
+		mmu_write(LIMA_MMU_DTE_ADDR, 0xCAFEBABE);
+		lima_mmu_send_command(LIMA_MMU_COMMAND_HARD_RESET,
+				      LIMA_MMU_DTE_ADDR, v, v == 0);
+		mmu_write(LIMA_MMU_INT_MASK, LIMA_MMU_INT_PAGE_FAULT | LIMA_MMU_INT_READ_BUS_ERROR);
+		mmu_write(LIMA_MMU_DTE_ADDR, dev->empty_vm->pd.dma);
+		lima_mmu_send_command(LIMA_MMU_COMMAND_ENABLE_PAGING,
+				      LIMA_MMU_STATUS, v,
+				      v & LIMA_MMU_STATUS_PAGING_ENABLED);
+	}
+}
diff --git a/drivers/gpu/drm/lima/lima_mmu.h b/drivers/gpu/drm/lima/lima_mmu.h
new file mode 100644
index 000000000000..8c78319bcc8e
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_mmu.h
@@ -0,0 +1,16 @@
+/* SPDX-License-Identifier: GPL-2.0 OR MIT */
+/* Copyright 2017-2019 Qiang Yu <yuq825@gmail.com> */
+
+#ifndef __LIMA_MMU_H__
+#define __LIMA_MMU_H__
+
+struct lima_ip;
+struct lima_vm;
+
+int lima_mmu_init(struct lima_ip *ip);
+void lima_mmu_fini(struct lima_ip *ip);
+
+void lima_mmu_switch_vm(struct lima_ip *ip, struct lima_vm *vm);
+void lima_mmu_page_fault_resume(struct lima_ip *ip);
+
+#endif
diff --git a/drivers/gpu/drm/lima/lima_object.c b/drivers/gpu/drm/lima/lima_object.c
new file mode 100644
index 000000000000..5c41f859a72f
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_object.c
@@ -0,0 +1,122 @@
+// SPDX-License-Identifier: GPL-2.0 OR MIT
+/* Copyright 2018-2019 Qiang Yu <yuq825@gmail.com> */
+
+#include <drm/drm_prime.h>
+#include <linux/pagemap.h>
+#include <linux/dma-mapping.h>
+
+#include "lima_object.h"
+
+void lima_bo_destroy(struct lima_bo *bo)
+{
+	if (bo->sgt) {
+		kfree(bo->pages);
+		drm_prime_gem_destroy(&bo->gem, bo->sgt);
+	} else {
+		if (bo->pages_dma_addr) {
+			int i, npages = bo->gem.size >> PAGE_SHIFT;
+
+			for (i = 0; i < npages; i++) {
+				if (bo->pages_dma_addr[i])
+					dma_unmap_page(bo->gem.dev->dev,
+						       bo->pages_dma_addr[i],
+						       PAGE_SIZE, DMA_BIDIRECTIONAL);
+			}
+		}
+
+		if (bo->pages)
+			drm_gem_put_pages(&bo->gem, bo->pages, true, true);
+	}
+
+	kfree(bo->pages_dma_addr);
+	drm_gem_object_release(&bo->gem);
+	kfree(bo);
+}
+
+static struct lima_bo *lima_bo_create_struct(struct lima_device *dev, u32 size, u32 flags,
+					     struct reservation_object *resv)
+{
+	struct lima_bo *bo;
+	int err;
+
+	size = PAGE_ALIGN(size);
+
+	bo = kzalloc(sizeof(*bo), GFP_KERNEL);
+	if (!bo)
+		return ERR_PTR(-ENOMEM);
+
+	mutex_init(&bo->lock);
+	INIT_LIST_HEAD(&bo->va);
+	bo->gem.resv = resv;
+
+	err = drm_gem_object_init(dev->ddev, &bo->gem, size);
+	if (err) {
+		kfree(bo);
+		return ERR_PTR(err);
+	}
+
+	return bo;
+}
+
+struct lima_bo *lima_bo_create(struct lima_device *dev, u32 size,
+			       u32 flags, struct sg_table *sgt,
+			       struct reservation_object *resv)
+{
+	int i, err;
+	size_t npages;
+	struct lima_bo *bo, *ret;
+
+	bo = lima_bo_create_struct(dev, size, flags, resv);
+	if (IS_ERR(bo))
+		return bo;
+
+	npages = bo->gem.size >> PAGE_SHIFT;
+
+	bo->pages_dma_addr = kcalloc(npages, sizeof(dma_addr_t), GFP_KERNEL);
+	if (!bo->pages_dma_addr) {
+		ret = ERR_PTR(-ENOMEM);
+		goto err_out;
+	}
+
+	if (sgt) {
+		bo->sgt = sgt;
+
+		bo->pages = kcalloc(npages, sizeof(*bo->pages), GFP_KERNEL);
+		if (!bo->pages) {
+			ret = ERR_PTR(-ENOMEM);
+			goto err_out;
+		}
+
+		err = drm_prime_sg_to_page_addr_arrays(
+			sgt, bo->pages, bo->pages_dma_addr, npages);
+		if (err) {
+			ret = ERR_PTR(err);
+			goto err_out;
+		}
+	} else {
+		mapping_set_gfp_mask(bo->gem.filp->f_mapping, GFP_DMA32);
+		bo->pages = drm_gem_get_pages(&bo->gem);
+		if (IS_ERR(bo->pages)) {
+			ret = ERR_CAST(bo->pages);
+			bo->pages = NULL;
+			goto err_out;
+		}
+
+		for (i = 0; i < npages; i++) {
+			dma_addr_t addr = dma_map_page(dev->dev, bo->pages[i], 0,
+						       PAGE_SIZE, DMA_BIDIRECTIONAL);
+			if (dma_mapping_error(dev->dev, addr)) {
+				ret = ERR_PTR(-EFAULT);
+				goto err_out;
+			}
+			bo->pages_dma_addr[i] = addr;
+		}
+
+	}
+
+	return bo;
+
+err_out:
+	lima_bo_destroy(bo);
+	return ret;
+}
diff --git a/drivers/gpu/drm/lima/lima_object.h b/drivers/gpu/drm/lima/lima_object.h
new file mode 100644
index 000000000000..6738724afb7b
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_object.h
@@ -0,0 +1,36 @@
+/* SPDX-License-Identifier: GPL-2.0 OR MIT */
+/* Copyright 2018-2019 Qiang Yu <yuq825@gmail.com> */
+
+#ifndef __LIMA_OBJECT_H__
+#define __LIMA_OBJECT_H__
+
+#include <drm/drm_gem.h>
+
+#include "lima_device.h"
+
+struct lima_bo {
+	struct drm_gem_object gem;
+
+	struct page **pages;
+	dma_addr_t *pages_dma_addr;
+	struct sg_table *sgt;
+	void *vaddr;
+
+	struct mutex lock;
+	struct list_head va;
+};
+
+static inline struct lima_bo *
+to_lima_bo(struct drm_gem_object *obj)
+{
+	return container_of(obj, struct lima_bo, gem);
+}
+
+struct lima_bo *lima_bo_create(struct lima_device *dev, u32 size,
+			       u32 flags, struct sg_table *sgt,
+			       struct reservation_object *resv);
+void lima_bo_destroy(struct lima_bo *bo);
+void *lima_bo_vmap(struct lima_bo *bo);
+void lima_bo_vunmap(struct lima_bo *bo);
+
+#endif
diff --git a/drivers/gpu/drm/lima/lima_pmu.c b/drivers/gpu/drm/lima/lima_pmu.c
new file mode 100644
index 000000000000..571f6d661581
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_pmu.c
@@ -0,0 +1,60 @@
+// SPDX-License-Identifier: GPL-2.0 OR MIT
+/* Copyright 2017-2019 Qiang Yu <yuq825@gmail.com> */
+
+#include <linux/iopoll.h>
+#include <linux/device.h>
+
+#include "lima_device.h"
+#include "lima_pmu.h"
+#include "lima_regs.h"
+
+#define pmu_write(reg, data) writel(data, ip->iomem + reg)
+#define pmu_read(reg) readl(ip->iomem + reg)
+
+static int lima_pmu_wait_cmd(struct lima_ip *ip)
+{
+	struct lima_device *dev = ip->dev;
+	int err;
+	u32 v;
+
+	err = readl_poll_timeout(ip->iomem + LIMA_PMU_INT_RAWSTAT,
+				 v, v & LIMA_PMU_INT_CMD_MASK,
+				 100, 100000);
+	if (err) {
+		dev_err(dev->dev, "timeout wait pmd cmd\n");
+		return err;
+	}
+
+	pmu_write(LIMA_PMU_INT_CLEAR, LIMA_PMU_INT_CMD_MASK);
+	return 0;
+}
+
+int lima_pmu_init(struct lima_ip *ip)
+{
+	int err;
+	u32 stat;
+
+	pmu_write(LIMA_PMU_INT_MASK, 0);
+
+	/* If this value is too low, when in high GPU clk freq,
+	 * GPU will be in unstable state.
+	 */
+	pmu_write(LIMA_PMU_SW_DELAY, 0xffff);
+
+	/* status reg 1=off 0=on */
+	stat = pmu_read(LIMA_PMU_STATUS);
+
+	/* power up all ip */
+	if (stat) {
+		pmu_write(LIMA_PMU_POWER_UP, stat);
+		err = lima_pmu_wait_cmd(ip);
+		if (err)
+			return err;
+	}
+	return 0;
+}
+
+void lima_pmu_fini(struct lima_ip *ip)
+{
+
+}
diff --git a/drivers/gpu/drm/lima/lima_pmu.h b/drivers/gpu/drm/lima/lima_pmu.h
new file mode 100644
index 000000000000..a2a18775eb07
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_pmu.h
@@ -0,0 +1,12 @@
+/* SPDX-License-Identifier: GPL-2.0 OR MIT */
+/* Copyright 2017-2019 Qiang Yu <yuq825@gmail.com> */
+
+#ifndef __LIMA_PMU_H__
+#define __LIMA_PMU_H__
+
+struct lima_ip;
+
+int lima_pmu_init(struct lima_ip *ip);
+void lima_pmu_fini(struct lima_ip *ip);
+
+#endif
diff --git a/drivers/gpu/drm/lima/lima_pp.c b/drivers/gpu/drm/lima/lima_pp.c
new file mode 100644
index 000000000000..d29721e177bf
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_pp.c
@@ -0,0 +1,427 @@
+// SPDX-License-Identifier: GPL-2.0 OR MIT
+/* Copyright 2017-2019 Qiang Yu <yuq825@gmail.com> */
+
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/device.h>
+#include <linux/slab.h>
+
+#include <drm/lima_drm.h>
+
+#include "lima_device.h"
+#include "lima_pp.h"
+#include "lima_dlbu.h"
+#include "lima_bcast.h"
+#include "lima_vm.h"
+#include "lima_regs.h"
+
+#define pp_write(reg, data) writel(data, ip->iomem + reg)
+#define pp_read(reg) readl(ip->iomem + reg)
+
+static void lima_pp_handle_irq(struct lima_ip *ip, u32 state)
+{
+	struct lima_device *dev = ip->dev;
+	struct lima_sched_pipe *pipe = dev->pipe + lima_pipe_pp;
+
+	if (state & LIMA_PP_IRQ_MASK_ERROR) {
+		u32 status = pp_read(LIMA_PP_STATUS);
+
+		dev_err(dev->dev, "pp error irq state=%x status=%x\n",
+			state, status);
+
+		pipe->error = true;
+
+		/* mask all interrupts before hard reset */
+		pp_write(LIMA_PP_INT_MASK, 0);
+	}
+
+	pp_write(LIMA_PP_INT_CLEAR, state);
+}
+
+static irqreturn_t lima_pp_irq_handler(int irq, void *data)
+{
+	struct lima_ip *ip = data;
+	struct lima_device *dev = ip->dev;
+	struct lima_sched_pipe *pipe = dev->pipe + lima_pipe_pp;
+	u32 state = pp_read(LIMA_PP_INT_STATUS);
+
+	/* for shared irq case */
+	if (!state)
+		return IRQ_NONE;
+
+	lima_pp_handle_irq(ip, state);
+
+	if (atomic_dec_and_test(&pipe->task))
+		lima_sched_pipe_task_done(pipe);
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t lima_pp_bcast_irq_handler(int irq, void *data)
+{
+	int i;
+	irqreturn_t ret = IRQ_NONE;
+	struct lima_ip *pp_bcast = data;
+	struct lima_device *dev = pp_bcast->dev;
+	struct lima_sched_pipe *pipe = dev->pipe + lima_pipe_pp;
+	struct drm_lima_m450_pp_frame *frame = pipe->current_task->frame;
+
+	for (i = 0; i < frame->num_pp; i++) {
+		struct lima_ip *ip = pipe->processor[i];
+		u32 status, state;
+
+		if (pipe->done & (1 << i))
+			continue;
+
+		/* status read first in case int state change in the middle
+		 * which may miss the interrupt handling
+		 */
+		status = pp_read(LIMA_PP_STATUS);
+		state = pp_read(LIMA_PP_INT_STATUS);
+
+		if (state) {
+			lima_pp_handle_irq(ip, state);
+			ret = IRQ_HANDLED;
+		} else {
+			if (status & LIMA_PP_STATUS_RENDERING_ACTIVE)
+				continue;
+		}
+
+		pipe->done |= (1 << i);
+		if (atomic_dec_and_test(&pipe->task))
+			lima_sched_pipe_task_done(pipe);
+	}
+
+	return ret;
+}
+
+static void lima_pp_soft_reset_async(struct lima_ip *ip)
+{
+	if (ip->data.async_reset)
+		return;
+
+	pp_write(LIMA_PP_INT_MASK, 0);
+	pp_write(LIMA_PP_INT_RAWSTAT, LIMA_PP_IRQ_MASK_ALL);
+	pp_write(LIMA_PP_CTRL, LIMA_PP_CTRL_SOFT_RESET);
+	ip->data.async_reset = true;
+}
+
+static int lima_pp_soft_reset_poll(struct lima_ip *ip)
+{
+	return !(pp_read(LIMA_PP_STATUS) & LIMA_PP_STATUS_RENDERING_ACTIVE) &&
+		pp_read(LIMA_PP_INT_RAWSTAT) == LIMA_PP_IRQ_RESET_COMPLETED;
+}
+
+static int lima_pp_soft_reset_async_wait_one(struct lima_ip *ip)
+{
+	struct lima_device *dev = ip->dev;
+	int ret;
+
+	ret = lima_poll_timeout(ip, lima_pp_soft_reset_poll, 0, 100);
+	if (ret) {
+		dev_err(dev->dev, "pp %s reset time out\n", lima_ip_name(ip));
+		return ret;
+	}
+
+	pp_write(LIMA_PP_INT_CLEAR, LIMA_PP_IRQ_MASK_ALL);
+	pp_write(LIMA_PP_INT_MASK, LIMA_PP_IRQ_MASK_USED);
+	return 0;
+}
+
+static int lima_pp_soft_reset_async_wait(struct lima_ip *ip)
+{
+	int i, err = 0;
+
+	if (!ip->data.async_reset)
+		return 0;
+
+	if (ip->id == lima_ip_pp_bcast) {
+		struct lima_device *dev = ip->dev;
+		struct lima_sched_pipe *pipe = dev->pipe + lima_pipe_pp;
+		struct drm_lima_m450_pp_frame *frame = pipe->current_task->frame;
+
+		for (i = 0; i < frame->num_pp; i++)
+			err |= lima_pp_soft_reset_async_wait_one(pipe->processor[i]);
+	} else
+		err = lima_pp_soft_reset_async_wait_one(ip);
+
+	ip->data.async_reset = false;
+	return err;
+}
+
+static void lima_pp_write_frame(struct lima_ip *ip, u32 *frame, u32 *wb)
+{
+	int i, j, n = 0;
+
+	for (i = 0; i < LIMA_PP_FRAME_REG_NUM; i++)
+		writel(frame[i], ip->iomem + LIMA_PP_FRAME + i * 4);
+
+	for (i = 0; i < 3; i++) {
+		for (j = 0; j < LIMA_PP_WB_REG_NUM; j++)
+			writel(wb[n++], ip->iomem + LIMA_PP_WB(i) + j * 4);
+	}
+}
+
+static int lima_pp_hard_reset_poll(struct lima_ip *ip)
+{
+	pp_write(LIMA_PP_PERF_CNT_0_LIMIT, 0xC01A0000);
+	return pp_read(LIMA_PP_PERF_CNT_0_LIMIT) == 0xC01A0000;
+}
+
+static int lima_pp_hard_reset(struct lima_ip *ip)
+{
+	struct lima_device *dev = ip->dev;
+	int ret;
+
+	pp_write(LIMA_PP_PERF_CNT_0_LIMIT, 0xC0FFE000);
+	pp_write(LIMA_PP_INT_MASK, 0);
+	pp_write(LIMA_PP_CTRL, LIMA_PP_CTRL_FORCE_RESET);
+	ret = lima_poll_timeout(ip, lima_pp_hard_reset_poll, 10, 100);
+	if (ret) {
+		dev_err(dev->dev, "pp hard reset timeout\n");
+		return ret;
+	}
+
+	pp_write(LIMA_PP_PERF_CNT_0_LIMIT, 0);
+	pp_write(LIMA_PP_INT_CLEAR, LIMA_PP_IRQ_MASK_ALL);
+	pp_write(LIMA_PP_INT_MASK, LIMA_PP_IRQ_MASK_USED);
+	return 0;
+}
+
+static void lima_pp_print_version(struct lima_ip *ip)
+{
+	u32 version, major, minor;
+	char *name;
+
+	version = pp_read(LIMA_PP_VERSION);
+	major = (version >> 8) & 0xFF;
+	minor = version & 0xFF;
+	switch (version >> 16) {
+	case 0xC807:
+	    name = "mali200";
+		break;
+	case 0xCE07:
+		name = "mali300";
+		break;
+	case 0xCD07:
+		name = "mali400";
+		break;
+	case 0xCF07:
+		name = "mali450";
+		break;
+	default:
+		name = "unknown";
+		break;
+	}
+	dev_info(ip->dev->dev, "%s - %s version major %d minor %d\n",
+		 lima_ip_name(ip), name, major, minor);
+}
+
+int lima_pp_init(struct lima_ip *ip)
+{
+	struct lima_device *dev = ip->dev;
+	int err;
+
+	lima_pp_print_version(ip);
+
+	ip->data.async_reset = false;
+	lima_pp_soft_reset_async(ip);
+	err = lima_pp_soft_reset_async_wait(ip);
+	if (err)
+		return err;
+
+	err = devm_request_irq(dev->dev, ip->irq, lima_pp_irq_handler,
+			       IRQF_SHARED, lima_ip_name(ip), ip);
+	if (err) {
+		dev_err(dev->dev, "pp %s fail to request irq\n",
+			lima_ip_name(ip));
+		return err;
+	}
+
+	dev->pp_version = pp_read(LIMA_PP_VERSION);
+
+	return 0;
+}
+
+void lima_pp_fini(struct lima_ip *ip)
+{
+
+}
+
+int lima_pp_bcast_init(struct lima_ip *ip)
+{
+	struct lima_device *dev = ip->dev;
+	int err;
+
+	err = devm_request_irq(dev->dev, ip->irq, lima_pp_bcast_irq_handler,
+			       IRQF_SHARED, lima_ip_name(ip), ip);
+	if (err) {
+		dev_err(dev->dev, "pp %s fail to request irq\n",
+			lima_ip_name(ip));
+		return err;
+	}
+
+	return 0;
+}
+
+void lima_pp_bcast_fini(struct lima_ip *ip)
+{
+
+}
+
+static int lima_pp_task_validate(struct lima_sched_pipe *pipe,
+				 struct lima_sched_task *task)
+{
+	u32 num_pp;
+
+	if (pipe->bcast_processor) {
+		struct drm_lima_m450_pp_frame *f = task->frame;
+
+		num_pp = f->num_pp;
+
+		if (f->_pad)
+			return -EINVAL;
+	} else {
+		struct drm_lima_m400_pp_frame *f = task->frame;
+
+		num_pp = f->num_pp;
+	}
+
+	if (num_pp == 0 || num_pp > pipe->num_processor)
+		return -EINVAL;
+
+	return 0;
+}
+
+static void lima_pp_task_run(struct lima_sched_pipe *pipe,
+			     struct lima_sched_task *task)
+{
+	if (pipe->bcast_processor) {
+		struct drm_lima_m450_pp_frame *frame = task->frame;
+		struct lima_device *dev = pipe->bcast_processor->dev;
+		struct lima_ip *ip = pipe->bcast_processor;
+		int i;
+
+		pipe->done = 0;
+		atomic_set(&pipe->task, frame->num_pp);
+
+		if (frame->use_dlbu) {
+			lima_dlbu_enable(dev, frame->num_pp);
+
+			frame->frame[LIMA_PP_FRAME >> 2] = LIMA_VA_RESERVE_DLBU;
+			lima_dlbu_set_reg(dev->ip + lima_ip_dlbu, frame->dlbu_regs);
+		} else
+			lima_dlbu_disable(dev);
+
+		lima_bcast_enable(dev, frame->num_pp);
+
+		lima_pp_soft_reset_async_wait(ip);
+
+		lima_pp_write_frame(ip, frame->frame, frame->wb);
+
+		for (i = 0; i < frame->num_pp; i++) {
+			struct lima_ip *ip = pipe->processor[i];
+
+			pp_write(LIMA_PP_STACK, frame->fragment_stack_address[i]);
+			if (!frame->use_dlbu)
+				pp_write(LIMA_PP_FRAME, frame->plbu_array_address[i]);
+		}
+
+		pp_write(LIMA_PP_CTRL, LIMA_PP_CTRL_START_RENDERING);
+	} else {
+		struct drm_lima_m400_pp_frame *frame = task->frame;
+		int i;
+
+		atomic_set(&pipe->task, frame->num_pp);
+
+		for (i = 0; i < frame->num_pp; i++) {
+			struct lima_ip *ip = pipe->processor[i];
+
+			frame->frame[LIMA_PP_FRAME >> 2] =
+				frame->plbu_array_address[i];
+			frame->frame[LIMA_PP_STACK >> 2] =
+				frame->fragment_stack_address[i];
+
+			lima_pp_soft_reset_async_wait(ip);
+
+			lima_pp_write_frame(ip, frame->frame, frame->wb);
+
+			pp_write(LIMA_PP_CTRL, LIMA_PP_CTRL_START_RENDERING);
+		}
+	}
+}
+
+static void lima_pp_task_fini(struct lima_sched_pipe *pipe)
+{
+	if (pipe->bcast_processor)
+		lima_pp_soft_reset_async(pipe->bcast_processor);
+	else {
+		int i;
+
+		for (i = 0; i < pipe->num_processor; i++)
+			lima_pp_soft_reset_async(pipe->processor[i]);
+	}
+}
+
+static void lima_pp_task_error(struct lima_sched_pipe *pipe)
+{
+	int i;
+
+	for (i = 0; i < pipe->num_processor; i++) {
+		struct lima_ip *ip = pipe->processor[i];
+
+		dev_err(ip->dev->dev, "pp task error %d int_state=%x status=%x\n",
+			i, pp_read(LIMA_PP_INT_STATUS), pp_read(LIMA_PP_STATUS));
+
+		lima_pp_hard_reset(ip);
+	}
+}
+
+static void lima_pp_task_mmu_error(struct lima_sched_pipe *pipe)
+{
+	if (atomic_dec_and_test(&pipe->task))
+		lima_sched_pipe_task_done(pipe);
+}
+
+static struct kmem_cache *lima_pp_task_slab;
+static int lima_pp_task_slab_refcnt;
+
+int lima_pp_pipe_init(struct lima_device *dev)
+{
+	int frame_size;
+	struct lima_sched_pipe *pipe = dev->pipe + lima_pipe_pp;
+
+	if (dev->id == lima_gpu_mali400)
+		frame_size = sizeof(struct drm_lima_m400_pp_frame);
+	else
+		frame_size = sizeof(struct drm_lima_m450_pp_frame);
+
+	if (!lima_pp_task_slab) {
+		lima_pp_task_slab = kmem_cache_create_usercopy(
+			"lima_pp_task", sizeof(struct lima_sched_task) + frame_size,
+			0, SLAB_HWCACHE_ALIGN, sizeof(struct lima_sched_task),
+			frame_size, NULL);
+		if (!lima_pp_task_slab)
+			return -ENOMEM;
+	}
+	lima_pp_task_slab_refcnt++;
+
+	pipe->frame_size = frame_size;
+	pipe->task_slab = lima_pp_task_slab;
+
+	pipe->task_validate = lima_pp_task_validate;
+	pipe->task_run = lima_pp_task_run;
+	pipe->task_fini = lima_pp_task_fini;
+	pipe->task_error = lima_pp_task_error;
+	pipe->task_mmu_error = lima_pp_task_mmu_error;
+
+	return 0;
+}
+
+void lima_pp_pipe_fini(struct lima_device *dev)
+{
+	if (!--lima_pp_task_slab_refcnt) {
+		kmem_cache_destroy(lima_pp_task_slab);
+		lima_pp_task_slab = NULL;
+	}
+}
diff --git a/drivers/gpu/drm/lima/lima_pp.h b/drivers/gpu/drm/lima/lima_pp.h
new file mode 100644
index 000000000000..bf60c77b2633
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_pp.h
@@ -0,0 +1,19 @@
+/* SPDX-License-Identifier: GPL-2.0 OR MIT */
+/* Copyright 2017-2019 Qiang Yu <yuq825@gmail.com> */
+
+#ifndef __LIMA_PP_H__
+#define __LIMA_PP_H__
+
+struct lima_ip;
+struct lima_device;
+
+int lima_pp_init(struct lima_ip *ip);
+void lima_pp_fini(struct lima_ip *ip);
+
+int lima_pp_bcast_init(struct lima_ip *ip);
+void lima_pp_bcast_fini(struct lima_ip *ip);
+
+int lima_pp_pipe_init(struct lima_device *dev);
+void lima_pp_pipe_fini(struct lima_device *dev);
+
+#endif
diff --git a/drivers/gpu/drm/lima/lima_regs.h b/drivers/gpu/drm/lima/lima_regs.h
new file mode 100644
index 000000000000..ace8ecefbe90
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_regs.h
@@ -0,0 +1,298 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright 2010-2017 ARM Limited. All rights reserved.
+ * Copyright 2017-2019 Qiang Yu <yuq825@gmail.com>
+ */
+
+#ifndef __LIMA_REGS_H__
+#define __LIMA_REGS_H__
+
+/* This file's register definition is collected from the
+ * official ARM Mali Utgard GPU kernel driver source code
+ */
+
+/* PMU regs */
+#define LIMA_PMU_POWER_UP                  0x00
+#define LIMA_PMU_POWER_DOWN                0x04
+#define   LIMA_PMU_POWER_GP0_MASK          BIT(0)
+#define   LIMA_PMU_POWER_L2_MASK           BIT(1)
+#define   LIMA_PMU_POWER_PP_MASK(i)        BIT(2 + i)
+
+/*
+ * On Mali450 each block automatically starts up its corresponding L2
+ * and the PPs are not fully independent controllable.
+ * Instead PP0, PP1-3 and PP4-7 can be turned on or off.
+ */
+#define   LIMA450_PMU_POWER_PP0_MASK       BIT(1)
+#define   LIMA450_PMU_POWER_PP13_MASK      BIT(2)
+#define   LIMA450_PMU_POWER_PP47_MASK      BIT(3)
+
+#define LIMA_PMU_STATUS                    0x08
+#define LIMA_PMU_INT_MASK                  0x0C
+#define LIMA_PMU_INT_RAWSTAT               0x10
+#define LIMA_PMU_INT_CLEAR                 0x18
+#define   LIMA_PMU_INT_CMD_MASK            BIT(0)
+#define LIMA_PMU_SW_DELAY                  0x1C
+
+/* L2 cache regs */
+#define LIMA_L2_CACHE_SIZE                   0x0004
+#define LIMA_L2_CACHE_STATUS                 0x0008
+#define   LIMA_L2_CACHE_STATUS_COMMAND_BUSY  BIT(0)
+#define   LIMA_L2_CACHE_STATUS_DATA_BUSY     BIT(1)
+#define LIMA_L2_CACHE_COMMAND                0x0010
+#define   LIMA_L2_CACHE_COMMAND_CLEAR_ALL    BIT(0)
+#define LIMA_L2_CACHE_CLEAR_PAGE             0x0014
+#define LIMA_L2_CACHE_MAX_READS              0x0018
+#define LIMA_L2_CACHE_ENABLE                 0x001C
+#define   LIMA_L2_CACHE_ENABLE_ACCESS        BIT(0)
+#define   LIMA_L2_CACHE_ENABLE_READ_ALLOCATE BIT(1)
+#define LIMA_L2_CACHE_PERFCNT_SRC0           0x0020
+#define LIMA_L2_CACHE_PERFCNT_VAL0           0x0024
+#define LIMA_L2_CACHE_PERFCNT_SRC1           0x0028
+#define LIMA_L2_CACHE_ERFCNT_VAL1            0x002C
+
+/* GP regs */
+#define LIMA_GP_VSCL_START_ADDR                0x00
+#define LIMA_GP_VSCL_END_ADDR                  0x04
+#define LIMA_GP_PLBUCL_START_ADDR              0x08
+#define LIMA_GP_PLBUCL_END_ADDR                0x0c
+#define LIMA_GP_PLBU_ALLOC_START_ADDR          0x10
+#define LIMA_GP_PLBU_ALLOC_END_ADDR            0x14
+#define LIMA_GP_CMD                            0x20
+#define   LIMA_GP_CMD_START_VS                 BIT(0)
+#define   LIMA_GP_CMD_START_PLBU               BIT(1)
+#define   LIMA_GP_CMD_UPDATE_PLBU_ALLOC        BIT(4)
+#define   LIMA_GP_CMD_RESET                    BIT(5)
+#define   LIMA_GP_CMD_FORCE_HANG               BIT(6)
+#define   LIMA_GP_CMD_STOP_BUS                 BIT(9)
+#define   LIMA_GP_CMD_SOFT_RESET               BIT(10)
+#define LIMA_GP_INT_RAWSTAT                    0x24
+#define LIMA_GP_INT_CLEAR                      0x28
+#define LIMA_GP_INT_MASK                       0x2C
+#define LIMA_GP_INT_STAT                       0x30
+#define   LIMA_GP_IRQ_VS_END_CMD_LST           BIT(0)
+#define   LIMA_GP_IRQ_PLBU_END_CMD_LST         BIT(1)
+#define   LIMA_GP_IRQ_PLBU_OUT_OF_MEM          BIT(2)
+#define   LIMA_GP_IRQ_VS_SEM_IRQ               BIT(3)
+#define   LIMA_GP_IRQ_PLBU_SEM_IRQ             BIT(4)
+#define   LIMA_GP_IRQ_HANG                     BIT(5)
+#define   LIMA_GP_IRQ_FORCE_HANG               BIT(6)
+#define   LIMA_GP_IRQ_PERF_CNT_0_LIMIT         BIT(7)
+#define   LIMA_GP_IRQ_PERF_CNT_1_LIMIT         BIT(8)
+#define   LIMA_GP_IRQ_WRITE_BOUND_ERR          BIT(9)
+#define   LIMA_GP_IRQ_SYNC_ERROR               BIT(10)
+#define   LIMA_GP_IRQ_AXI_BUS_ERROR            BIT(11)
+#define   LIMA_GP_IRQ_AXI_BUS_STOPPED          BIT(12)
+#define   LIMA_GP_IRQ_VS_INVALID_CMD           BIT(13)
+#define   LIMA_GP_IRQ_PLB_INVALID_CMD          BIT(14)
+#define   LIMA_GP_IRQ_RESET_COMPLETED          BIT(19)
+#define   LIMA_GP_IRQ_SEMAPHORE_UNDERFLOW      BIT(20)
+#define   LIMA_GP_IRQ_SEMAPHORE_OVERFLOW       BIT(21)
+#define   LIMA_GP_IRQ_PTR_ARRAY_OUT_OF_BOUNDS  BIT(22)
+#define LIMA_GP_WRITE_BOUND_LOW                0x34
+#define LIMA_GP_PERF_CNT_0_ENABLE              0x3C
+#define LIMA_GP_PERF_CNT_1_ENABLE              0x40
+#define LIMA_GP_PERF_CNT_0_SRC                 0x44
+#define LIMA_GP_PERF_CNT_1_SRC                 0x48
+#define LIMA_GP_PERF_CNT_0_VALUE               0x4C
+#define LIMA_GP_PERF_CNT_1_VALUE               0x50
+#define LIMA_GP_PERF_CNT_0_LIMIT               0x54
+#define LIMA_GP_STATUS                         0x68
+#define   LIMA_GP_STATUS_VS_ACTIVE             BIT(1)
+#define   LIMA_GP_STATUS_BUS_STOPPED           BIT(2)
+#define   LIMA_GP_STATUS_PLBU_ACTIVE           BIT(3)
+#define   LIMA_GP_STATUS_BUS_ERROR             BIT(6)
+#define   LIMA_GP_STATUS_WRITE_BOUND_ERR       BIT(8)
+#define LIMA_GP_VERSION                        0x6C
+#define LIMA_GP_VSCL_START_ADDR_READ           0x80
+#define LIMA_GP_PLBCL_START_ADDR_READ          0x84
+#define LIMA_GP_CONTR_AXI_BUS_ERROR_STAT       0x94
+
+#define LIMA_GP_IRQ_MASK_ALL		   \
+	(				   \
+	 LIMA_GP_IRQ_VS_END_CMD_LST      | \
+	 LIMA_GP_IRQ_PLBU_END_CMD_LST    | \
+	 LIMA_GP_IRQ_PLBU_OUT_OF_MEM     | \
+	 LIMA_GP_IRQ_VS_SEM_IRQ          | \
+	 LIMA_GP_IRQ_PLBU_SEM_IRQ        | \
+	 LIMA_GP_IRQ_HANG                | \
+	 LIMA_GP_IRQ_FORCE_HANG          | \
+	 LIMA_GP_IRQ_PERF_CNT_0_LIMIT    | \
+	 LIMA_GP_IRQ_PERF_CNT_1_LIMIT    | \
+	 LIMA_GP_IRQ_WRITE_BOUND_ERR     | \
+	 LIMA_GP_IRQ_SYNC_ERROR          | \
+	 LIMA_GP_IRQ_AXI_BUS_ERROR       | \
+	 LIMA_GP_IRQ_AXI_BUS_STOPPED     | \
+	 LIMA_GP_IRQ_VS_INVALID_CMD      | \
+	 LIMA_GP_IRQ_PLB_INVALID_CMD     | \
+	 LIMA_GP_IRQ_RESET_COMPLETED     | \
+	 LIMA_GP_IRQ_SEMAPHORE_UNDERFLOW | \
+	 LIMA_GP_IRQ_SEMAPHORE_OVERFLOW  | \
+	 LIMA_GP_IRQ_PTR_ARRAY_OUT_OF_BOUNDS)
+
+#define LIMA_GP_IRQ_MASK_ERROR             \
+	(                                  \
+	 LIMA_GP_IRQ_PLBU_OUT_OF_MEM     | \
+	 LIMA_GP_IRQ_FORCE_HANG          | \
+	 LIMA_GP_IRQ_WRITE_BOUND_ERR     | \
+	 LIMA_GP_IRQ_SYNC_ERROR          | \
+	 LIMA_GP_IRQ_AXI_BUS_ERROR       | \
+	 LIMA_GP_IRQ_VS_INVALID_CMD      | \
+	 LIMA_GP_IRQ_PLB_INVALID_CMD     | \
+	 LIMA_GP_IRQ_SEMAPHORE_UNDERFLOW | \
+	 LIMA_GP_IRQ_SEMAPHORE_OVERFLOW  | \
+	 LIMA_GP_IRQ_PTR_ARRAY_OUT_OF_BOUNDS)
+
+#define LIMA_GP_IRQ_MASK_USED		   \
+	(				   \
+	 LIMA_GP_IRQ_VS_END_CMD_LST      | \
+	 LIMA_GP_IRQ_PLBU_END_CMD_LST    | \
+	 LIMA_GP_IRQ_MASK_ERROR)
+
+/* PP regs */
+#define LIMA_PP_FRAME                        0x0000
+#define LIMA_PP_RSW			     0x0004
+#define LIMA_PP_STACK			     0x0030
+#define LIMA_PP_STACK_SIZE		     0x0034
+#define LIMA_PP_ORIGIN_OFFSET_X	             0x0040
+#define LIMA_PP_WB(i)                        (0x0100 * (i + 1))
+#define   LIMA_PP_WB_SOURCE_SELECT           0x0000
+#define	  LIMA_PP_WB_SOURCE_ADDR             0x0004
+
+#define LIMA_PP_VERSION                      0x1000
+#define LIMA_PP_CURRENT_REND_LIST_ADDR       0x1004
+#define LIMA_PP_STATUS                       0x1008
+#define   LIMA_PP_STATUS_RENDERING_ACTIVE    BIT(0)
+#define   LIMA_PP_STATUS_BUS_STOPPED         BIT(4)
+#define LIMA_PP_CTRL                         0x100c
+#define   LIMA_PP_CTRL_STOP_BUS              BIT(0)
+#define   LIMA_PP_CTRL_FLUSH_CACHES          BIT(3)
+#define   LIMA_PP_CTRL_FORCE_RESET           BIT(5)
+#define   LIMA_PP_CTRL_START_RENDERING       BIT(6)
+#define   LIMA_PP_CTRL_SOFT_RESET            BIT(7)
+#define LIMA_PP_INT_RAWSTAT                  0x1020
+#define LIMA_PP_INT_CLEAR                    0x1024
+#define LIMA_PP_INT_MASK                     0x1028
+#define LIMA_PP_INT_STATUS                   0x102c
+#define   LIMA_PP_IRQ_END_OF_FRAME           BIT(0)
+#define   LIMA_PP_IRQ_END_OF_TILE            BIT(1)
+#define   LIMA_PP_IRQ_HANG                   BIT(2)
+#define   LIMA_PP_IRQ_FORCE_HANG             BIT(3)
+#define   LIMA_PP_IRQ_BUS_ERROR              BIT(4)
+#define   LIMA_PP_IRQ_BUS_STOP               BIT(5)
+#define   LIMA_PP_IRQ_CNT_0_LIMIT            BIT(6)
+#define   LIMA_PP_IRQ_CNT_1_LIMIT            BIT(7)
+#define   LIMA_PP_IRQ_WRITE_BOUNDARY_ERROR   BIT(8)
+#define   LIMA_PP_IRQ_INVALID_PLIST_COMMAND  BIT(9)
+#define   LIMA_PP_IRQ_CALL_STACK_UNDERFLOW   BIT(10)
+#define   LIMA_PP_IRQ_CALL_STACK_OVERFLOW    BIT(11)
+#define   LIMA_PP_IRQ_RESET_COMPLETED        BIT(12)
+#define LIMA_PP_WRITE_BOUNDARY_LOW           0x1044
+#define LIMA_PP_BUS_ERROR_STATUS             0x1050
+#define LIMA_PP_PERF_CNT_0_ENABLE            0x1080
+#define LIMA_PP_PERF_CNT_0_SRC               0x1084
+#define LIMA_PP_PERF_CNT_0_LIMIT             0x1088
+#define LIMA_PP_PERF_CNT_0_VALUE             0x108c
+#define LIMA_PP_PERF_CNT_1_ENABLE            0x10a0
+#define LIMA_PP_PERF_CNT_1_SRC               0x10a4
+#define LIMA_PP_PERF_CNT_1_LIMIT             0x10a8
+#define LIMA_PP_PERF_CNT_1_VALUE             0x10ac
+#define LIMA_PP_PERFMON_CONTR                0x10b0
+#define LIMA_PP_PERFMON_BASE                 0x10b4
+
+#define LIMA_PP_IRQ_MASK_ALL                 \
+	(                                    \
+	 LIMA_PP_IRQ_END_OF_FRAME          | \
+	 LIMA_PP_IRQ_END_OF_TILE           | \
+	 LIMA_PP_IRQ_HANG                  | \
+	 LIMA_PP_IRQ_FORCE_HANG            | \
+	 LIMA_PP_IRQ_BUS_ERROR             | \
+	 LIMA_PP_IRQ_BUS_STOP              | \
+	 LIMA_PP_IRQ_CNT_0_LIMIT           | \
+	 LIMA_PP_IRQ_CNT_1_LIMIT           | \
+	 LIMA_PP_IRQ_WRITE_BOUNDARY_ERROR  | \
+	 LIMA_PP_IRQ_INVALID_PLIST_COMMAND | \
+	 LIMA_PP_IRQ_CALL_STACK_UNDERFLOW  | \
+	 LIMA_PP_IRQ_CALL_STACK_OVERFLOW   | \
+	 LIMA_PP_IRQ_RESET_COMPLETED)
+
+#define LIMA_PP_IRQ_MASK_ERROR               \
+	(                                    \
+	 LIMA_PP_IRQ_FORCE_HANG            | \
+	 LIMA_PP_IRQ_BUS_ERROR             | \
+	 LIMA_PP_IRQ_WRITE_BOUNDARY_ERROR  | \
+	 LIMA_PP_IRQ_INVALID_PLIST_COMMAND | \
+	 LIMA_PP_IRQ_CALL_STACK_UNDERFLOW  | \
+	 LIMA_PP_IRQ_CALL_STACK_OVERFLOW)
+
+#define LIMA_PP_IRQ_MASK_USED                \
+	(                                    \
+	 LIMA_PP_IRQ_END_OF_FRAME          | \
+	 LIMA_PP_IRQ_MASK_ERROR)
+
+/* MMU regs */
+#define LIMA_MMU_DTE_ADDR                     0x0000
+#define LIMA_MMU_STATUS                       0x0004
+#define   LIMA_MMU_STATUS_PAGING_ENABLED      BIT(0)
+#define   LIMA_MMU_STATUS_PAGE_FAULT_ACTIVE   BIT(1)
+#define   LIMA_MMU_STATUS_STALL_ACTIVE        BIT(2)
+#define   LIMA_MMU_STATUS_IDLE                BIT(3)
+#define   LIMA_MMU_STATUS_REPLAY_BUFFER_EMPTY BIT(4)
+#define   LIMA_MMU_STATUS_PAGE_FAULT_IS_WRITE BIT(5)
+#define   LIMA_MMU_STATUS_BUS_ID(x)           ((x >> 6) & 0x1F)
+#define LIMA_MMU_COMMAND                      0x0008
+#define   LIMA_MMU_COMMAND_ENABLE_PAGING      0x00
+#define   LIMA_MMU_COMMAND_DISABLE_PAGING     0x01
+#define   LIMA_MMU_COMMAND_ENABLE_STALL       0x02
+#define   LIMA_MMU_COMMAND_DISABLE_STALL      0x03
+#define   LIMA_MMU_COMMAND_ZAP_CACHE          0x04
+#define   LIMA_MMU_COMMAND_PAGE_FAULT_DONE    0x05
+#define   LIMA_MMU_COMMAND_HARD_RESET         0x06
+#define LIMA_MMU_PAGE_FAULT_ADDR              0x000C
+#define LIMA_MMU_ZAP_ONE_LINE                 0x0010
+#define LIMA_MMU_INT_RAWSTAT                  0x0014
+#define LIMA_MMU_INT_CLEAR                    0x0018
+#define LIMA_MMU_INT_MASK                     0x001C
+#define   LIMA_MMU_INT_PAGE_FAULT             BIT(0)
+#define   LIMA_MMU_INT_READ_BUS_ERROR         BIT(1)
+#define LIMA_MMU_INT_STATUS                   0x0020
+
+#define LIMA_VM_FLAG_PRESENT          BIT(0)
+#define LIMA_VM_FLAG_READ_PERMISSION  BIT(1)
+#define LIMA_VM_FLAG_WRITE_PERMISSION BIT(2)
+#define LIMA_VM_FLAG_OVERRIDE_CACHE   BIT(3)
+#define LIMA_VM_FLAG_WRITE_CACHEABLE  BIT(4)
+#define LIMA_VM_FLAG_WRITE_ALLOCATE   BIT(5)
+#define LIMA_VM_FLAG_WRITE_BUFFERABLE BIT(6)
+#define LIMA_VM_FLAG_READ_CACHEABLE   BIT(7)
+#define LIMA_VM_FLAG_READ_ALLOCATE    BIT(8)
+#define LIMA_VM_FLAG_MASK             0x1FF
+
+#define LIMA_VM_FLAGS_CACHE (			 \
+		LIMA_VM_FLAG_PRESENT |		 \
+		LIMA_VM_FLAG_READ_PERMISSION |	 \
+		LIMA_VM_FLAG_WRITE_PERMISSION |	 \
+		LIMA_VM_FLAG_OVERRIDE_CACHE |	 \
+		LIMA_VM_FLAG_WRITE_CACHEABLE |	 \
+		LIMA_VM_FLAG_WRITE_BUFFERABLE |	 \
+		LIMA_VM_FLAG_READ_CACHEABLE |	 \
+		LIMA_VM_FLAG_READ_ALLOCATE)
+
+#define LIMA_VM_FLAGS_UNCACHE (			\
+		LIMA_VM_FLAG_PRESENT |		\
+		LIMA_VM_FLAG_READ_PERMISSION |	\
+		LIMA_VM_FLAG_WRITE_PERMISSION)
+
+/* DLBU regs */
+#define LIMA_DLBU_MASTER_TLLIST_PHYS_ADDR  0x0000
+#define	LIMA_DLBU_MASTER_TLLIST_VADDR      0x0004
+#define	LIMA_DLBU_TLLIST_VBASEADDR         0x0008
+#define	LIMA_DLBU_FB_DIM                   0x000C
+#define	LIMA_DLBU_TLLIST_CONF              0x0010
+#define	LIMA_DLBU_START_TILE_POS           0x0014
+#define	LIMA_DLBU_PP_ENABLE_MASK           0x0018
+
+/* BCAST regs */
+#define LIMA_BCAST_BROADCAST_MASK    0x0
+#define LIMA_BCAST_INTERRUPT_MASK    0x4
+
+#endif
diff --git a/drivers/gpu/drm/lima/lima_sched.c b/drivers/gpu/drm/lima/lima_sched.c
new file mode 100644
index 000000000000..d53bd45f8d96
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_sched.c
@@ -0,0 +1,362 @@
+// SPDX-License-Identifier: GPL-2.0 OR MIT
+/* Copyright 2017-2019 Qiang Yu <yuq825@gmail.com> */
+
+#include <linux/kthread.h>
+#include <linux/slab.h>
+#include <linux/xarray.h>
+
+#include "lima_drv.h"
+#include "lima_sched.h"
+#include "lima_vm.h"
+#include "lima_mmu.h"
+#include "lima_l2_cache.h"
+#include "lima_object.h"
+
+struct lima_fence {
+	struct dma_fence base;
+	struct lima_sched_pipe *pipe;
+};
+
+static struct kmem_cache *lima_fence_slab;
+static int lima_fence_slab_refcnt;
+
+int lima_sched_slab_init(void)
+{
+	if (!lima_fence_slab) {
+		lima_fence_slab = kmem_cache_create(
+			"lima_fence", sizeof(struct lima_fence), 0,
+			SLAB_HWCACHE_ALIGN, NULL);
+		if (!lima_fence_slab)
+			return -ENOMEM;
+	}
+
+	lima_fence_slab_refcnt++;
+	return 0;
+}
+
+void lima_sched_slab_fini(void)
+{
+	if (!--lima_fence_slab_refcnt) {
+		kmem_cache_destroy(lima_fence_slab);
+		lima_fence_slab = NULL;
+	}
+}
+
+static inline struct lima_fence *to_lima_fence(struct dma_fence *fence)
+{
+	return container_of(fence, struct lima_fence, base);
+}
+
+static const char *lima_fence_get_driver_name(struct dma_fence *fence)
+{
+	return "lima";
+}
+
+static const char *lima_fence_get_timeline_name(struct dma_fence *fence)
+{
+	struct lima_fence *f = to_lima_fence(fence);
+
+	return f->pipe->base.name;
+}
+
+static void lima_fence_release_rcu(struct rcu_head *rcu)
+{
+	struct dma_fence *f = container_of(rcu, struct dma_fence, rcu);
+	struct lima_fence *fence = to_lima_fence(f);
+
+	kmem_cache_free(lima_fence_slab, fence);
+}
+
+static void lima_fence_release(struct dma_fence *fence)
+{
+	struct lima_fence *f = to_lima_fence(fence);
+
+	call_rcu(&f->base.rcu, lima_fence_release_rcu);
+}
+
+static const struct dma_fence_ops lima_fence_ops = {
+	.get_driver_name = lima_fence_get_driver_name,
+	.get_timeline_name = lima_fence_get_timeline_name,
+	.release = lima_fence_release,
+};
+
+static struct lima_fence *lima_fence_create(struct lima_sched_pipe *pipe)
+{
+	struct lima_fence *fence;
+
+	fence = kmem_cache_zalloc(lima_fence_slab, GFP_KERNEL);
+	if (!fence)
+		return NULL;
+
+	fence->pipe = pipe;
+	dma_fence_init(&fence->base, &lima_fence_ops, &pipe->fence_lock,
+		       pipe->fence_context, ++pipe->fence_seqno);
+
+	return fence;
+}
+
+static inline struct lima_sched_task *to_lima_task(struct drm_sched_job *job)
+{
+	return container_of(job, struct lima_sched_task, base);
+}
+
+static inline struct lima_sched_pipe *to_lima_pipe(struct drm_gpu_scheduler *sched)
+{
+	return container_of(sched, struct lima_sched_pipe, base);
+}
+
+int lima_sched_task_init(struct lima_sched_task *task,
+			 struct lima_sched_context *context,
+			 struct lima_bo **bos, int num_bos,
+			 struct lima_vm *vm)
+{
+	int err, i;
+
+	task->bos = kmemdup(bos, sizeof(*bos) * num_bos, GFP_KERNEL);
+	if (!task->bos)
+		return -ENOMEM;
+
+	for (i = 0; i < num_bos; i++)
+		drm_gem_object_get(&bos[i]->gem);
+
+	err = drm_sched_job_init(&task->base, &context->base, vm);
+	if (err) {
+		kfree(task->bos);
+		return err;
+	}
+
+	task->num_bos = num_bos;
+	task->vm = lima_vm_get(vm);
+
+	xa_init_flags(&task->deps, XA_FLAGS_ALLOC);
+
+	return 0;
+}
+
+void lima_sched_task_fini(struct lima_sched_task *task)
+{
+	struct dma_fence *fence;
+	unsigned long index;
+	int i;
+
+	drm_sched_job_cleanup(&task->base);
+
+	xa_for_each(&task->deps, index, fence) {
+		dma_fence_put(fence);
+	}
+	xa_destroy(&task->deps);
+
+	if (task->bos) {
+		for (i = 0; i < task->num_bos; i++)
+			drm_gem_object_put_unlocked(&task->bos[i]->gem);
+		kfree(task->bos);
+	}
+
+	lima_vm_put(task->vm);
+}
+
+int lima_sched_context_init(struct lima_sched_pipe *pipe,
+			    struct lima_sched_context *context,
+			    atomic_t *guilty)
+{
+	struct drm_sched_rq *rq = pipe->base.sched_rq + DRM_SCHED_PRIORITY_NORMAL;
+
+	return drm_sched_entity_init(&context->base, &rq, 1, guilty);
+}
+
+void lima_sched_context_fini(struct lima_sched_pipe *pipe,
+			     struct lima_sched_context *context)
+{
+	drm_sched_entity_fini(&context->base);
+}
+
+struct dma_fence *lima_sched_context_queue_task(struct lima_sched_context *context,
+						struct lima_sched_task *task)
+{
+	struct dma_fence *fence = dma_fence_get(&task->base.s_fence->finished);
+
+	drm_sched_entity_push_job(&task->base, &context->base);
+	return fence;
+}
+
+static struct dma_fence *lima_sched_dependency(struct drm_sched_job *job,
+					       struct drm_sched_entity *entity)
+{
+	struct lima_sched_task *task = to_lima_task(job);
+
+	if (!xa_empty(&task->deps))
+		return xa_erase(&task->deps, task->last_dep++);
+
+	return NULL;
+}
+
+static struct dma_fence *lima_sched_run_job(struct drm_sched_job *job)
+{
+	struct lima_sched_task *task = to_lima_task(job);
+	struct lima_sched_pipe *pipe = to_lima_pipe(job->sched);
+	struct lima_fence *fence;
+	struct dma_fence *ret;
+	struct lima_vm *vm = NULL, *last_vm = NULL;
+	int i;
+
+	/* after GPU reset */
+	if (job->s_fence->finished.error < 0)
+		return NULL;
+
+	fence = lima_fence_create(pipe);
+	if (!fence)
+		return NULL;
+	task->fence = &fence->base;
+
+	/* for caller usage of the fence, otherwise irq handler
+	 * may consume the fence before caller use it
+	 */
+	ret = dma_fence_get(task->fence);
+
+	pipe->current_task = task;
+
+	/* this is needed for MMU to work correctly, otherwise GP/PP
+	 * will hang or page fault for unknown reason after running for
+	 * a while.
+	 *
+	 * Need to investigate:
+	 * 1. is it related to TLB
+	 * 2. how much performance will be affected by L2 cache flush
+	 * 3. can we reduce the calling of this function because all
+	 *    GP/PP use the same L2 cache on mali400
+	 *
+	 * TODO:
+	 * 1. move this to task fini to save some wait time?
+	 * 2. when GP/PP use different l2 cache, need PP wait GP l2
+	 *    cache flush?
+	 */
+	for (i = 0; i < pipe->num_l2_cache; i++)
+		lima_l2_cache_flush(pipe->l2_cache[i]);
+
+	if (task->vm != pipe->current_vm) {
+		vm = lima_vm_get(task->vm);
+		last_vm = pipe->current_vm;
+		pipe->current_vm = task->vm;
+	}
+
+	if (pipe->bcast_mmu)
+		lima_mmu_switch_vm(pipe->bcast_mmu, vm);
+	else {
+		for (i = 0; i < pipe->num_mmu; i++)
+			lima_mmu_switch_vm(pipe->mmu[i], vm);
+	}
+
+	if (last_vm)
+		lima_vm_put(last_vm);
+
+	pipe->error = false;
+	pipe->task_run(pipe, task);
+
+	return task->fence;
+}
+
+static void lima_sched_handle_error_task(struct lima_sched_pipe *pipe,
+					 struct lima_sched_task *task)
+{
+	drm_sched_stop(&pipe->base);
+
+	if (task)
+		drm_sched_increase_karma(&task->base);
+
+	pipe->task_error(pipe);
+
+	if (pipe->bcast_mmu)
+		lima_mmu_page_fault_resume(pipe->bcast_mmu);
+	else {
+		int i;
+
+		for (i = 0; i < pipe->num_mmu; i++)
+			lima_mmu_page_fault_resume(pipe->mmu[i]);
+	}
+
+	if (pipe->current_vm)
+		lima_vm_put(pipe->current_vm);
+
+	pipe->current_vm = NULL;
+	pipe->current_task = NULL;
+
+	drm_sched_resubmit_jobs(&pipe->base);
+	drm_sched_start(&pipe->base, true);
+}
+
+static void lima_sched_timedout_job(struct drm_sched_job *job)
+{
+	struct lima_sched_pipe *pipe = to_lima_pipe(job->sched);
+	struct lima_sched_task *task = to_lima_task(job);
+
+	DRM_ERROR("lima job timeout\n");
+
+	lima_sched_handle_error_task(pipe, task);
+}
+
+static void lima_sched_free_job(struct drm_sched_job *job)
+{
+	struct lima_sched_task *task = to_lima_task(job);
+	struct lima_sched_pipe *pipe = to_lima_pipe(job->sched);
+	struct lima_vm *vm = task->vm;
+	struct lima_bo **bos = task->bos;
+	int i;
+
+	dma_fence_put(task->fence);
+
+	for (i = 0; i < task->num_bos; i++)
+		lima_vm_bo_del(vm, bos[i]);
+
+	lima_sched_task_fini(task);
+	kmem_cache_free(pipe->task_slab, task);
+}
+
+static const struct drm_sched_backend_ops lima_sched_ops = {
+	.dependency = lima_sched_dependency,
+	.run_job = lima_sched_run_job,
+	.timedout_job = lima_sched_timedout_job,
+	.free_job = lima_sched_free_job,
+};
+
+static void lima_sched_error_work(struct work_struct *work)
+{
+	struct lima_sched_pipe *pipe =
+		container_of(work, struct lima_sched_pipe, error_work);
+	struct lima_sched_task *task = pipe->current_task;
+
+	lima_sched_handle_error_task(pipe, task);
+}
+
+int lima_sched_pipe_init(struct lima_sched_pipe *pipe, const char *name)
+{
+	long timeout;
+
+	if (lima_sched_timeout_ms <= 0)
+		timeout = MAX_SCHEDULE_TIMEOUT;
+	else
+		timeout = msecs_to_jiffies(lima_sched_timeout_ms);
+
+	pipe->fence_context = dma_fence_context_alloc(1);
+	spin_lock_init(&pipe->fence_lock);
+
+	INIT_WORK(&pipe->error_work, lima_sched_error_work);
+
+	return drm_sched_init(&pipe->base, &lima_sched_ops, 1, 0, timeout, name);
+}
+
+void lima_sched_pipe_fini(struct lima_sched_pipe *pipe)
+{
+	drm_sched_fini(&pipe->base);
+}
+
+void lima_sched_pipe_task_done(struct lima_sched_pipe *pipe)
+{
+	if (pipe->error)
+		schedule_work(&pipe->error_work);
+	else {
+		struct lima_sched_task *task = pipe->current_task;
+
+		pipe->task_fini(pipe);
+		dma_fence_signal(task->fence);
+	}
+}
diff --git a/drivers/gpu/drm/lima/lima_sched.h b/drivers/gpu/drm/lima/lima_sched.h
new file mode 100644
index 000000000000..928af91c1118
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_sched.h
@@ -0,0 +1,102 @@
+/* SPDX-License-Identifier: GPL-2.0 OR MIT */
+/* Copyright 2017-2019 Qiang Yu <yuq825@gmail.com> */
+
+#ifndef __LIMA_SCHED_H__
+#define __LIMA_SCHED_H__
+
+#include <drm/gpu_scheduler.h>
+
+struct lima_vm;
+
+struct lima_sched_task {
+	struct drm_sched_job base;
+
+	struct lima_vm *vm;
+	void *frame;
+
+	struct xarray deps;
+	unsigned long last_dep;
+
+	struct lima_bo **bos;
+	int num_bos;
+
+	/* pipe fence */
+	struct dma_fence *fence;
+};
+
+struct lima_sched_context {
+	struct drm_sched_entity base;
+};
+
+#define LIMA_SCHED_PIPE_MAX_MMU       8
+#define LIMA_SCHED_PIPE_MAX_L2_CACHE  2
+#define LIMA_SCHED_PIPE_MAX_PROCESSOR 8
+
+struct lima_ip;
+
+struct lima_sched_pipe {
+	struct drm_gpu_scheduler base;
+
+	u64 fence_context;
+	u32 fence_seqno;
+	spinlock_t fence_lock;
+
+	struct lima_sched_task *current_task;
+	struct lima_vm *current_vm;
+
+	struct lima_ip *mmu[LIMA_SCHED_PIPE_MAX_MMU];
+	int num_mmu;
+
+	struct lima_ip *l2_cache[LIMA_SCHED_PIPE_MAX_L2_CACHE];
+	int num_l2_cache;
+
+	struct lima_ip *processor[LIMA_SCHED_PIPE_MAX_PROCESSOR];
+	int num_processor;
+
+	struct lima_ip *bcast_processor;
+	struct lima_ip *bcast_mmu;
+
+	u32 done;
+	bool error;
+	atomic_t task;
+
+	int frame_size;
+	struct kmem_cache *task_slab;
+
+	int (*task_validate)(struct lima_sched_pipe *pipe, struct lima_sched_task *task);
+	void (*task_run)(struct lima_sched_pipe *pipe, struct lima_sched_task *task);
+	void (*task_fini)(struct lima_sched_pipe *pipe);
+	void (*task_error)(struct lima_sched_pipe *pipe);
+	void (*task_mmu_error)(struct lima_sched_pipe *pipe);
+
+	struct work_struct error_work;
+};
+
+int lima_sched_task_init(struct lima_sched_task *task,
+			 struct lima_sched_context *context,
+			 struct lima_bo **bos, int num_bos,
+			 struct lima_vm *vm);
+void lima_sched_task_fini(struct lima_sched_task *task);
+
+int lima_sched_context_init(struct lima_sched_pipe *pipe,
+			    struct lima_sched_context *context,
+			    atomic_t *guilty);
+void lima_sched_context_fini(struct lima_sched_pipe *pipe,
+			     struct lima_sched_context *context);
+struct dma_fence *lima_sched_context_queue_task(struct lima_sched_context *context,
+						struct lima_sched_task *task);
+
+int lima_sched_pipe_init(struct lima_sched_pipe *pipe, const char *name);
+void lima_sched_pipe_fini(struct lima_sched_pipe *pipe);
+void lima_sched_pipe_task_done(struct lima_sched_pipe *pipe);
+
+static inline void lima_sched_pipe_mmu_error(struct lima_sched_pipe *pipe)
+{
+	pipe->error = true;
+	pipe->task_mmu_error(pipe);
+}
+
+int lima_sched_slab_init(void);
+void lima_sched_slab_fini(void);
+
+#endif
diff --git a/drivers/gpu/drm/lima/lima_vm.c b/drivers/gpu/drm/lima/lima_vm.c
new file mode 100644
index 000000000000..19e88ca16527
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_vm.c
@@ -0,0 +1,282 @@
+// SPDX-License-Identifier: GPL-2.0 OR MIT
+/* Copyright 2017-2019 Qiang Yu <yuq825@gmail.com> */
+
+#include <linux/slab.h>
+#include <linux/dma-mapping.h>
+
+#include "lima_device.h"
+#include "lima_vm.h"
+#include "lima_object.h"
+#include "lima_regs.h"
+
+struct lima_bo_va {
+	struct list_head list;
+	unsigned int ref_count;
+
+	struct drm_mm_node node;
+
+	struct lima_vm *vm;
+};
+
+#define LIMA_VM_PD_SHIFT 22
+#define LIMA_VM_PT_SHIFT 12
+#define LIMA_VM_PB_SHIFT (LIMA_VM_PD_SHIFT + LIMA_VM_NUM_PT_PER_BT_SHIFT)
+#define LIMA_VM_BT_SHIFT LIMA_VM_PT_SHIFT
+
+#define LIMA_VM_PT_MASK ((1 << LIMA_VM_PD_SHIFT) - 1)
+#define LIMA_VM_BT_MASK ((1 << LIMA_VM_PB_SHIFT) - 1)
+
+#define LIMA_PDE(va) (va >> LIMA_VM_PD_SHIFT)
+#define LIMA_PTE(va) ((va & LIMA_VM_PT_MASK) >> LIMA_VM_PT_SHIFT)
+#define LIMA_PBE(va) (va >> LIMA_VM_PB_SHIFT)
+#define LIMA_BTE(va) ((va & LIMA_VM_BT_MASK) >> LIMA_VM_BT_SHIFT)
+
+
+static void lima_vm_unmap_page_table(struct lima_vm *vm, u32 start, u32 end)
+{
+	u32 addr;
+
+	for (addr = start; addr <= end; addr += LIMA_PAGE_SIZE) {
+		u32 pbe = LIMA_PBE(addr);
+		u32 bte = LIMA_BTE(addr);
+
+		vm->bts[pbe].cpu[bte] = 0;
+	}
+}
+
+static int lima_vm_map_page_table(struct lima_vm *vm, dma_addr_t *dma,
+				  u32 start, u32 end)
+{
+	u64 addr;
+	int i = 0;
+
+	for (addr = start; addr <= end; addr += LIMA_PAGE_SIZE) {
+		u32 pbe = LIMA_PBE(addr);
+		u32 bte = LIMA_BTE(addr);
+
+		if (!vm->bts[pbe].cpu) {
+			dma_addr_t pts;
+			u32 *pd;
+			int j;
+
+			vm->bts[pbe].cpu = dma_alloc_wc(
+				vm->dev->dev, LIMA_PAGE_SIZE << LIMA_VM_NUM_PT_PER_BT_SHIFT,
+				&vm->bts[pbe].dma, GFP_KERNEL | __GFP_ZERO);
+			if (!vm->bts[pbe].cpu) {
+				if (addr != start)
+					lima_vm_unmap_page_table(vm, start, addr - 1);
+				return -ENOMEM;
+			}
+
+			pts = vm->bts[pbe].dma;
+			pd = vm->pd.cpu + (pbe << LIMA_VM_NUM_PT_PER_BT_SHIFT);
+			for (j = 0; j < LIMA_VM_NUM_PT_PER_BT; j++) {
+				pd[j] = pts | LIMA_VM_FLAG_PRESENT;
+				pts += LIMA_PAGE_SIZE;
+			}
+		}
+
+		vm->bts[pbe].cpu[bte] = dma[i++] | LIMA_VM_FLAGS_CACHE;
+	}
+
+	return 0;
+}
+
+static struct lima_bo_va *
+lima_vm_bo_find(struct lima_vm *vm, struct lima_bo *bo)
+{
+	struct lima_bo_va *bo_va, *ret = NULL;
+
+	list_for_each_entry(bo_va, &bo->va, list) {
+		if (bo_va->vm == vm) {
+			ret = bo_va;
+			break;
+		}
+	}
+
+	return ret;
+}
+
+int lima_vm_bo_add(struct lima_vm *vm, struct lima_bo *bo, bool create)
+{
+	struct lima_bo_va *bo_va;
+	int err;
+
+	mutex_lock(&bo->lock);
+
+	bo_va = lima_vm_bo_find(vm, bo);
+	if (bo_va) {
+		bo_va->ref_count++;
+		mutex_unlock(&bo->lock);
+		return 0;
+	}
+
+	/* should not create new bo_va if not asked by caller */
+	if (!create) {
+		mutex_unlock(&bo->lock);
+		return -ENOENT;
+	}
+
+	bo_va = kzalloc(sizeof(*bo_va), GFP_KERNEL);
+	if (!bo_va) {
+		err = -ENOMEM;
+		goto err_out0;
+	}
+
+	bo_va->vm = vm;
+	bo_va->ref_count = 1;
+
+	mutex_lock(&vm->lock);
+
+	err = drm_mm_insert_node(&vm->mm, &bo_va->node, bo->gem.size);
+	if (err)
+		goto err_out1;
+
+	err = lima_vm_map_page_table(vm, bo->pages_dma_addr, bo_va->node.start,
+				     bo_va->node.start + bo_va->node.size - 1);
+	if (err)
+		goto err_out2;
+
+	mutex_unlock(&vm->lock);
+
+	list_add_tail(&bo_va->list, &bo->va);
+
+	mutex_unlock(&bo->lock);
+	return 0;
+
+err_out2:
+	drm_mm_remove_node(&bo_va->node);
+err_out1:
+	mutex_unlock(&vm->lock);
+	kfree(bo_va);
+err_out0:
+	mutex_unlock(&bo->lock);
+	return err;
+}
+
+void lima_vm_bo_del(struct lima_vm *vm, struct lima_bo *bo)
+{
+	struct lima_bo_va *bo_va;
+
+	mutex_lock(&bo->lock);
+
+	bo_va = lima_vm_bo_find(vm, bo);
+	if (--bo_va->ref_count > 0) {
+		mutex_unlock(&bo->lock);
+		return;
+	}
+
+	mutex_lock(&vm->lock);
+
+	lima_vm_unmap_page_table(vm, bo_va->node.start,
+				 bo_va->node.start + bo_va->node.size - 1);
+
+	drm_mm_remove_node(&bo_va->node);
+
+	mutex_unlock(&vm->lock);
+
+	list_del(&bo_va->list);
+
+	mutex_unlock(&bo->lock);
+
+	kfree(bo_va);
+}
+
+u32 lima_vm_get_va(struct lima_vm *vm, struct lima_bo *bo)
+{
+	struct lima_bo_va *bo_va;
+	u32 ret;
+
+	mutex_lock(&bo->lock);
+
+	bo_va = lima_vm_bo_find(vm, bo);
+	ret = bo_va->node.start;
+
+	mutex_unlock(&bo->lock);
+
+	return ret;
+}
+
+struct lima_vm *lima_vm_create(struct lima_device *dev)
+{
+	struct lima_vm *vm;
+
+	vm = kzalloc(sizeof(*vm), GFP_KERNEL);
+	if (!vm)
+		return NULL;
+
+	vm->dev = dev;
+	mutex_init(&vm->lock);
+	kref_init(&vm->refcount);
+
+	vm->pd.cpu = dma_alloc_wc(dev->dev, LIMA_PAGE_SIZE, &vm->pd.dma,
+				  GFP_KERNEL | __GFP_ZERO);
+	if (!vm->pd.cpu)
+		goto err_out0;
+
+	if (dev->dlbu_cpu) {
+		int err = lima_vm_map_page_table(
+			vm, &dev->dlbu_dma, LIMA_VA_RESERVE_DLBU,
+			LIMA_VA_RESERVE_DLBU + LIMA_PAGE_SIZE - 1);
+		if (err)
+			goto err_out1;
+	}
+
+	drm_mm_init(&vm->mm, dev->va_start, dev->va_end - dev->va_start);
+
+	return vm;
+
+err_out1:
+	dma_free_wc(dev->dev, LIMA_PAGE_SIZE, vm->pd.cpu, vm->pd.dma);
+err_out0:
+	kfree(vm);
+	return NULL;
+}
+
+void lima_vm_release(struct kref *kref)
+{
+	struct lima_vm *vm = container_of(kref, struct lima_vm, refcount);
+	int i;
+
+	drm_mm_takedown(&vm->mm);
+
+	for (i = 0; i < LIMA_VM_NUM_BT; i++) {
+		if (vm->bts[i].cpu)
+			dma_free_wc(vm->dev->dev, LIMA_PAGE_SIZE << LIMA_VM_NUM_PT_PER_BT_SHIFT,
+				    vm->bts[i].cpu, vm->bts[i].dma);
+	}
+
+	if (vm->pd.cpu)
+		dma_free_wc(vm->dev->dev, LIMA_PAGE_SIZE, vm->pd.cpu, vm->pd.dma);
+
+	kfree(vm);
+}
+
+void lima_vm_print(struct lima_vm *vm)
+{
+	int i, j, k;
+	u32 *pd, *pt;
+
+	if (!vm->pd.cpu)
+		return;
+
+	pd = vm->pd.cpu;
+	for (i = 0; i < LIMA_VM_NUM_BT; i++) {
+		if (!vm->bts[i].cpu)
+			continue;
+
+		pt = vm->bts[i].cpu;
+		for (j = 0; j < LIMA_VM_NUM_PT_PER_BT; j++) {
+			int idx = (i << LIMA_VM_NUM_PT_PER_BT_SHIFT) + j;
+
+			printk(KERN_INFO "lima vm pd %03x:%08x\n", idx, pd[idx]);
+
+			for (k = 0; k < LIMA_PAGE_ENT_NUM; k++) {
+				u32 pte = *pt++;
+
+				if (pte)
+					printk(KERN_INFO "  pt %03x:%08x\n", k, pte);
+			}
+		}
+	}
+}
diff --git a/drivers/gpu/drm/lima/lima_vm.h b/drivers/gpu/drm/lima/lima_vm.h
new file mode 100644
index 000000000000..caee2f8a29b4
--- /dev/null
+++ b/drivers/gpu/drm/lima/lima_vm.h
@@ -0,0 +1,62 @@
+/* SPDX-License-Identifier: GPL-2.0 OR MIT */
+/* Copyright 2017-2019 Qiang Yu <yuq825@gmail.com> */
+
+#ifndef __LIMA_VM_H__
+#define __LIMA_VM_H__
+
+#include <drm/drm_mm.h>
+#include <linux/kref.h>
+
+#define LIMA_PAGE_SIZE    4096
+#define LIMA_PAGE_MASK    (LIMA_PAGE_SIZE - 1)
+#define LIMA_PAGE_ENT_NUM (LIMA_PAGE_SIZE / sizeof(u32))
+
+#define LIMA_VM_NUM_PT_PER_BT_SHIFT 3
+#define LIMA_VM_NUM_PT_PER_BT (1 << LIMA_VM_NUM_PT_PER_BT_SHIFT)
+#define LIMA_VM_NUM_BT (LIMA_PAGE_ENT_NUM >> LIMA_VM_NUM_PT_PER_BT_SHIFT)
+
+#define LIMA_VA_RESERVE_START  0xFFF00000
+#define LIMA_VA_RESERVE_DLBU   LIMA_VA_RESERVE_START
+#define LIMA_VA_RESERVE_END    0x100000000
+
+struct lima_device;
+
+struct lima_vm_page {
+	u32 *cpu;
+	dma_addr_t dma;
+};
+
+struct lima_vm {
+	struct mutex lock;
+	struct kref refcount;
+
+	struct drm_mm mm;
+
+	struct lima_device *dev;
+
+	struct lima_vm_page pd;
+	struct lima_vm_page bts[LIMA_VM_NUM_BT];
+};
+
+int lima_vm_bo_add(struct lima_vm *vm, struct lima_bo *bo, bool create);
+void lima_vm_bo_del(struct lima_vm *vm, struct lima_bo *bo);
+
+u32 lima_vm_get_va(struct lima_vm *vm, struct lima_bo *bo);
+
+struct lima_vm *lima_vm_create(struct lima_device *dev);
+void lima_vm_release(struct kref *kref);
+
+static inline struct lima_vm *lima_vm_get(struct lima_vm *vm)
+{
+	kref_get(&vm->refcount);
+	return vm;
+}
+
+static inline void lima_vm_put(struct lima_vm *vm)
+{
+	kref_put(&vm->refcount, lima_vm_release);
+}
+
+void lima_vm_print(struct lima_vm *vm);
+
+#endif
diff --git a/drivers/gpu/drm/meson/Makefile b/drivers/gpu/drm/meson/Makefile
index 7709f2fbb9f7..d4ea82fc493b 100644
--- a/drivers/gpu/drm/meson/Makefile
+++ b/drivers/gpu/drm/meson/Makefile
@@ -1,5 +1,5 @@
 meson-drm-y := meson_drv.o meson_plane.o meson_crtc.o meson_venc_cvbs.o
-meson-drm-y += meson_viu.o meson_vpp.o meson_venc.o meson_vclk.o meson_canvas.o meson_overlay.o
+meson-drm-y += meson_viu.o meson_vpp.o meson_venc.o meson_vclk.o meson_overlay.o
 
 obj-$(CONFIG_DRM_MESON) += meson-drm.o
 obj-$(CONFIG_DRM_MESON_DW_HDMI) += meson_dw_hdmi.o
diff --git a/drivers/gpu/drm/meson/meson_canvas.c b/drivers/gpu/drm/meson/meson_canvas.c
deleted file mode 100644
index 5de11aa7c775..000000000000
--- a/drivers/gpu/drm/meson/meson_canvas.c
+++ /dev/null
@@ -1,73 +0,0 @@
-/*
- * Copyright (C) 2016 BayLibre, SAS
- * Author: Neil Armstrong <narmstrong@baylibre.com>
- * Copyright (C) 2015 Amlogic, Inc. All rights reserved.
- * Copyright (C) 2014 Endless Mobile
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License as
- * published by the Free Software Foundation; either version 2 of the
- * License, or (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, see <http://www.gnu.org/licenses/>.
- */
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include "meson_drv.h"
-#include "meson_canvas.h"
-#include "meson_registers.h"
-
-/**
- * DOC: Canvas
- *
- * CANVAS is a memory zone where physical memory frames information
- * are stored for the VIU to scanout.
- */
-
-/* DMC Registers */
-#define DMC_CAV_LUT_DATAL	0x48 /* 0x12 offset in data sheet */
-#define CANVAS_WIDTH_LBIT	29
-#define CANVAS_WIDTH_LWID       3
-#define DMC_CAV_LUT_DATAH	0x4c /* 0x13 offset in data sheet */
-#define CANVAS_WIDTH_HBIT       0
-#define CANVAS_HEIGHT_BIT       9
-#define CANVAS_BLKMODE_BIT      24
-#define CANVAS_ENDIAN_BIT	26
-#define DMC_CAV_LUT_ADDR	0x50 /* 0x14 offset in data sheet */
-#define CANVAS_LUT_WR_EN        (0x2 << 8)
-#define CANVAS_LUT_RD_EN        (0x1 << 8)
-
-void meson_canvas_setup(struct meson_drm *priv,
-			uint32_t canvas_index, uint32_t addr,
-			uint32_t stride, uint32_t height,
-			unsigned int wrap,
-			unsigned int blkmode,
-			unsigned int endian)
-{
-	unsigned int val;
-
-	regmap_write(priv->dmc, DMC_CAV_LUT_DATAL,
-		(((addr + 7) >> 3)) |
-		(((stride + 7) >> 3) << CANVAS_WIDTH_LBIT));
-
-	regmap_write(priv->dmc, DMC_CAV_LUT_DATAH,
-		((((stride + 7) >> 3) >> CANVAS_WIDTH_LWID) <<
-						CANVAS_WIDTH_HBIT) |
-		(height << CANVAS_HEIGHT_BIT) |
-		(wrap << 22) |
-		(blkmode << CANVAS_BLKMODE_BIT) |
-		(endian << CANVAS_ENDIAN_BIT));
-
-	regmap_write(priv->dmc, DMC_CAV_LUT_ADDR,
-			CANVAS_LUT_WR_EN | canvas_index);
-
-	/* Force a read-back to make sure everything is flushed. */
-	regmap_read(priv->dmc, DMC_CAV_LUT_DATAH, &val);
-}
diff --git a/drivers/gpu/drm/meson/meson_canvas.h b/drivers/gpu/drm/meson/meson_canvas.h
deleted file mode 100644
index 85dbf26e2826..000000000000
--- a/drivers/gpu/drm/meson/meson_canvas.h
+++ /dev/null
@@ -1,51 +0,0 @@
-/*
- * Copyright (C) 2016 BayLibre, SAS
- * Author: Neil Armstrong <narmstrong@baylibre.com>
- * Copyright (C) 2014 Endless Mobile
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License as
- * published by the Free Software Foundation; either version 2 of the
- * License, or (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, see <http://www.gnu.org/licenses/>.
- */
-
-/* Canvas LUT Memory */
-
-#ifndef __MESON_CANVAS_H
-#define __MESON_CANVAS_H
-
-#define MESON_CANVAS_ID_OSD1	0x4e
-#define MESON_CANVAS_ID_VD1_0	0x60
-#define MESON_CANVAS_ID_VD1_1	0x61
-#define MESON_CANVAS_ID_VD1_2	0x62
-
-/* Canvas configuration. */
-#define MESON_CANVAS_WRAP_NONE	0x00
-#define	MESON_CANVAS_WRAP_X	0x01
-#define	MESON_CANVAS_WRAP_Y	0x02
-
-#define	MESON_CANVAS_BLKMODE_LINEAR	0x00
-#define	MESON_CANVAS_BLKMODE_32x32	0x01
-#define	MESON_CANVAS_BLKMODE_64x64	0x02
-
-#define MESON_CANVAS_ENDIAN_SWAP16	0x1
-#define MESON_CANVAS_ENDIAN_SWAP32	0x3
-#define MESON_CANVAS_ENDIAN_SWAP64	0x7
-#define MESON_CANVAS_ENDIAN_SWAP128	0xf
-
-void meson_canvas_setup(struct meson_drm *priv,
-			uint32_t canvas_index, uint32_t addr,
-			uint32_t stride, uint32_t height,
-			unsigned int wrap,
-			unsigned int blkmode,
-			unsigned int endian);
-
-#endif /* __MESON_CANVAS_H */
diff --git a/drivers/gpu/drm/meson/meson_crtc.c b/drivers/gpu/drm/meson/meson_crtc.c
index 43e29984f8b1..5579f8ac3e3f 100644
--- a/drivers/gpu/drm/meson/meson_crtc.c
+++ b/drivers/gpu/drm/meson/meson_crtc.c
@@ -37,15 +37,19 @@
 #include "meson_venc.h"
 #include "meson_vpp.h"
 #include "meson_viu.h"
-#include "meson_canvas.h"
 #include "meson_registers.h"
 
+#define MESON_G12A_VIU_OFFSET	0x5ec0
+
 /* CRTC definition */
 
 struct meson_crtc {
 	struct drm_crtc base;
 	struct drm_pending_vblank_event *event;
 	struct meson_drm *priv;
+	void (*enable_osd1)(struct meson_drm *priv);
+	void (*enable_vd1)(struct meson_drm *priv);
+	unsigned int viu_offset;
 };
 #define to_meson_crtc(x) container_of(x, struct meson_crtc, base)
 
@@ -81,6 +85,44 @@ static const struct drm_crtc_funcs meson_crtc_funcs = {
 
 };
 
+static void meson_g12a_crtc_atomic_enable(struct drm_crtc *crtc,
+					  struct drm_crtc_state *old_state)
+{
+	struct meson_crtc *meson_crtc = to_meson_crtc(crtc);
+	struct drm_crtc_state *crtc_state = crtc->state;
+	struct meson_drm *priv = meson_crtc->priv;
+
+	DRM_DEBUG_DRIVER("\n");
+
+	if (!crtc_state) {
+		DRM_ERROR("Invalid crtc_state\n");
+		return;
+	}
+
+	/* VD1 Preblend vertical start/end */
+	writel(FIELD_PREP(GENMASK(11, 0), 2303),
+	       priv->io_base + _REG(VPP_PREBLEND_VD1_V_START_END));
+
+	/* Setup Blender */
+	writel(crtc_state->mode.hdisplay |
+	       crtc_state->mode.vdisplay << 16,
+	       priv->io_base + _REG(VPP_POSTBLEND_H_SIZE));
+
+	writel_relaxed(0 << 16 |
+			(crtc_state->mode.hdisplay - 1),
+			priv->io_base + _REG(VPP_OSD1_BLD_H_SCOPE));
+	writel_relaxed(0 << 16 |
+			(crtc_state->mode.vdisplay - 1),
+			priv->io_base + _REG(VPP_OSD1_BLD_V_SCOPE));
+	writel_relaxed(crtc_state->mode.hdisplay << 16 |
+			crtc_state->mode.vdisplay,
+			priv->io_base + _REG(VPP_OUT_H_V_SIZE));
+
+	drm_crtc_vblank_on(crtc);
+
+	priv->viu.osd1_enabled = true;
+}
+
 static void meson_crtc_atomic_enable(struct drm_crtc *crtc,
 				     struct drm_crtc_state *old_state)
 {
@@ -111,6 +153,31 @@ static void meson_crtc_atomic_enable(struct drm_crtc *crtc,
 	priv->viu.osd1_enabled = true;
 }
 
+static void meson_g12a_crtc_atomic_disable(struct drm_crtc *crtc,
+					   struct drm_crtc_state *old_state)
+{
+	struct meson_crtc *meson_crtc = to_meson_crtc(crtc);
+	struct meson_drm *priv = meson_crtc->priv;
+
+	DRM_DEBUG_DRIVER("\n");
+
+	drm_crtc_vblank_off(crtc);
+
+	priv->viu.osd1_enabled = false;
+	priv->viu.osd1_commit = false;
+
+	priv->viu.vd1_enabled = false;
+	priv->viu.vd1_commit = false;
+
+	if (crtc->state->event && !crtc->state->active) {
+		spin_lock_irq(&crtc->dev->event_lock);
+		drm_crtc_send_vblank_event(crtc, crtc->state->event);
+		spin_unlock_irq(&crtc->dev->event_lock);
+
+		crtc->state->event = NULL;
+	}
+}
+
 static void meson_crtc_atomic_disable(struct drm_crtc *crtc,
 				      struct drm_crtc_state *old_state)
 {
@@ -174,6 +241,53 @@ static const struct drm_crtc_helper_funcs meson_crtc_helper_funcs = {
 	.atomic_disable	= meson_crtc_atomic_disable,
 };
 
+static const struct drm_crtc_helper_funcs meson_g12a_crtc_helper_funcs = {
+	.atomic_begin	= meson_crtc_atomic_begin,
+	.atomic_flush	= meson_crtc_atomic_flush,
+	.atomic_enable	= meson_g12a_crtc_atomic_enable,
+	.atomic_disable	= meson_g12a_crtc_atomic_disable,
+};
+
+static void meson_crtc_enable_osd1(struct meson_drm *priv)
+{
+	writel_bits_relaxed(VPP_OSD1_POSTBLEND, VPP_OSD1_POSTBLEND,
+			    priv->io_base + _REG(VPP_MISC));
+}
+
+static void meson_g12a_crtc_enable_osd1(struct meson_drm *priv)
+{
+	writel_relaxed(priv->viu.osd_blend_din0_scope_h,
+		       priv->io_base +
+		       _REG(VIU_OSD_BLEND_DIN0_SCOPE_H));
+	writel_relaxed(priv->viu.osd_blend_din0_scope_v,
+		       priv->io_base +
+		       _REG(VIU_OSD_BLEND_DIN0_SCOPE_V));
+	writel_relaxed(priv->viu.osb_blend0_size,
+		       priv->io_base +
+		       _REG(VIU_OSD_BLEND_BLEND0_SIZE));
+	writel_relaxed(priv->viu.osb_blend1_size,
+		       priv->io_base +
+		       _REG(VIU_OSD_BLEND_BLEND1_SIZE));
+}
+
+static void meson_crtc_enable_vd1(struct meson_drm *priv)
+{
+	writel_bits_relaxed(VPP_VD1_PREBLEND | VPP_VD1_POSTBLEND |
+			    VPP_COLOR_MNG_ENABLE,
+			    VPP_VD1_PREBLEND | VPP_VD1_POSTBLEND |
+			    VPP_COLOR_MNG_ENABLE,
+			    priv->io_base + _REG(VPP_MISC));
+}
+
+static void meson_g12a_crtc_enable_vd1(struct meson_drm *priv)
+{
+	writel_relaxed(((1 << 16) | /* post bld premult*/
+			(1 << 8) | /* post src */
+			(1 << 4) | /* pre bld premult*/
+			(1 << 0)),
+			priv->io_base + _REG(VD1_BLEND_SRC_CTRL));
+}
+
 void meson_crtc_irq(struct meson_drm *priv)
 {
 	struct meson_crtc *meson_crtc = to_meson_crtc(priv->crtc);
@@ -214,20 +328,14 @@ void meson_crtc_irq(struct meson_drm *priv)
 		writel_relaxed(priv->viu.osd_sc_v_ctrl0,
 				priv->io_base + _REG(VPP_OSD_VSC_CTRL0));
 
-		if (priv->canvas)
-			meson_canvas_config(priv->canvas, priv->canvas_id_osd1,
-				priv->viu.osd1_addr, priv->viu.osd1_stride,
-				priv->viu.osd1_height, MESON_CANVAS_WRAP_NONE,
-				MESON_CANVAS_BLKMODE_LINEAR, 0);
-		else
-			meson_canvas_setup(priv, MESON_CANVAS_ID_OSD1,
+		meson_canvas_config(priv->canvas, priv->canvas_id_osd1,
 				priv->viu.osd1_addr, priv->viu.osd1_stride,
 				priv->viu.osd1_height, MESON_CANVAS_WRAP_NONE,
 				MESON_CANVAS_BLKMODE_LINEAR, 0);
 
 		/* Enable OSD1 */
-		writel_bits_relaxed(VPP_OSD1_POSTBLEND, VPP_OSD1_POSTBLEND,
-				    priv->io_base + _REG(VPP_MISC));
+		if (meson_crtc->enable_osd1)
+			meson_crtc->enable_osd1(priv);
 
 		priv->viu.osd1_commit = false;
 	}
@@ -237,147 +345,164 @@ void meson_crtc_irq(struct meson_drm *priv)
 
 		switch (priv->viu.vd1_planes) {
 		case 3:
-			if (priv->canvas)
-				meson_canvas_config(priv->canvas,
-						    priv->canvas_id_vd1_2,
-						    priv->viu.vd1_addr2,
-						    priv->viu.vd1_stride2,
-						    priv->viu.vd1_height2,
-						    MESON_CANVAS_WRAP_NONE,
-						    MESON_CANVAS_BLKMODE_LINEAR,
-						    MESON_CANVAS_ENDIAN_SWAP64);
-			else
-				meson_canvas_setup(priv, MESON_CANVAS_ID_VD1_2,
-						   priv->viu.vd1_addr2,
-						   priv->viu.vd1_stride2,
-						   priv->viu.vd1_height2,
-						   MESON_CANVAS_WRAP_NONE,
-						   MESON_CANVAS_BLKMODE_LINEAR,
-						   MESON_CANVAS_ENDIAN_SWAP64);
+			meson_canvas_config(priv->canvas,
+					    priv->canvas_id_vd1_2,
+					    priv->viu.vd1_addr2,
+					    priv->viu.vd1_stride2,
+					    priv->viu.vd1_height2,
+					    MESON_CANVAS_WRAP_NONE,
+					    MESON_CANVAS_BLKMODE_LINEAR,
+					    MESON_CANVAS_ENDIAN_SWAP64);
 		/* fallthrough */
 		case 2:
-			if (priv->canvas)
-				meson_canvas_config(priv->canvas,
-						    priv->canvas_id_vd1_1,
-						    priv->viu.vd1_addr1,
-						    priv->viu.vd1_stride1,
-						    priv->viu.vd1_height1,
-						    MESON_CANVAS_WRAP_NONE,
-						    MESON_CANVAS_BLKMODE_LINEAR,
-						    MESON_CANVAS_ENDIAN_SWAP64);
-			else
-				meson_canvas_setup(priv, MESON_CANVAS_ID_VD1_1,
-						   priv->viu.vd1_addr2,
-						   priv->viu.vd1_stride2,
-						   priv->viu.vd1_height2,
-						   MESON_CANVAS_WRAP_NONE,
-						   MESON_CANVAS_BLKMODE_LINEAR,
-						   MESON_CANVAS_ENDIAN_SWAP64);
+			meson_canvas_config(priv->canvas,
+					    priv->canvas_id_vd1_1,
+					    priv->viu.vd1_addr1,
+					    priv->viu.vd1_stride1,
+					    priv->viu.vd1_height1,
+					    MESON_CANVAS_WRAP_NONE,
+					    MESON_CANVAS_BLKMODE_LINEAR,
+					    MESON_CANVAS_ENDIAN_SWAP64);
 		/* fallthrough */
 		case 1:
-			if (priv->canvas)
-				meson_canvas_config(priv->canvas,
-						    priv->canvas_id_vd1_0,
-						    priv->viu.vd1_addr0,
-						    priv->viu.vd1_stride0,
-						    priv->viu.vd1_height0,
-						    MESON_CANVAS_WRAP_NONE,
-						    MESON_CANVAS_BLKMODE_LINEAR,
-						    MESON_CANVAS_ENDIAN_SWAP64);
-			else
-				meson_canvas_setup(priv, MESON_CANVAS_ID_VD1_0,
-						   priv->viu.vd1_addr2,
-						   priv->viu.vd1_stride2,
-						   priv->viu.vd1_height2,
-						   MESON_CANVAS_WRAP_NONE,
-						   MESON_CANVAS_BLKMODE_LINEAR,
-						   MESON_CANVAS_ENDIAN_SWAP64);
+			meson_canvas_config(priv->canvas,
+					    priv->canvas_id_vd1_0,
+					    priv->viu.vd1_addr0,
+					    priv->viu.vd1_stride0,
+					    priv->viu.vd1_height0,
+					    MESON_CANVAS_WRAP_NONE,
+					    MESON_CANVAS_BLKMODE_LINEAR,
+					    MESON_CANVAS_ENDIAN_SWAP64);
 		};
 
 		writel_relaxed(priv->viu.vd1_if0_gen_reg,
-				priv->io_base + _REG(VD1_IF0_GEN_REG));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_GEN_REG));
 		writel_relaxed(priv->viu.vd1_if0_gen_reg,
-				priv->io_base + _REG(VD2_IF0_GEN_REG));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD2_IF0_GEN_REG));
 		writel_relaxed(priv->viu.vd1_if0_gen_reg2,
-				priv->io_base + _REG(VD1_IF0_GEN_REG2));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_GEN_REG2));
 		writel_relaxed(priv->viu.viu_vd1_fmt_ctrl,
-				priv->io_base + _REG(VIU_VD1_FMT_CTRL));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VIU_VD1_FMT_CTRL));
 		writel_relaxed(priv->viu.viu_vd1_fmt_ctrl,
-				priv->io_base + _REG(VIU_VD2_FMT_CTRL));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VIU_VD2_FMT_CTRL));
 		writel_relaxed(priv->viu.viu_vd1_fmt_w,
-				priv->io_base + _REG(VIU_VD1_FMT_W));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VIU_VD1_FMT_W));
 		writel_relaxed(priv->viu.viu_vd1_fmt_w,
-				priv->io_base + _REG(VIU_VD2_FMT_W));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VIU_VD2_FMT_W));
 		writel_relaxed(priv->viu.vd1_if0_canvas0,
-				priv->io_base + _REG(VD1_IF0_CANVAS0));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_CANVAS0));
 		writel_relaxed(priv->viu.vd1_if0_canvas0,
-				priv->io_base + _REG(VD1_IF0_CANVAS1));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_CANVAS1));
 		writel_relaxed(priv->viu.vd1_if0_canvas0,
-				priv->io_base + _REG(VD2_IF0_CANVAS0));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD2_IF0_CANVAS0));
 		writel_relaxed(priv->viu.vd1_if0_canvas0,
-				priv->io_base + _REG(VD2_IF0_CANVAS1));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD2_IF0_CANVAS1));
 		writel_relaxed(priv->viu.vd1_if0_luma_x0,
-				priv->io_base + _REG(VD1_IF0_LUMA_X0));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_LUMA_X0));
 		writel_relaxed(priv->viu.vd1_if0_luma_x0,
-				priv->io_base + _REG(VD1_IF0_LUMA_X1));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_LUMA_X1));
 		writel_relaxed(priv->viu.vd1_if0_luma_x0,
-				priv->io_base + _REG(VD2_IF0_LUMA_X0));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD2_IF0_LUMA_X0));
 		writel_relaxed(priv->viu.vd1_if0_luma_x0,
-				priv->io_base + _REG(VD2_IF0_LUMA_X1));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD2_IF0_LUMA_X1));
 		writel_relaxed(priv->viu.vd1_if0_luma_y0,
-				priv->io_base + _REG(VD1_IF0_LUMA_Y0));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_LUMA_Y0));
 		writel_relaxed(priv->viu.vd1_if0_luma_y0,
-				priv->io_base + _REG(VD1_IF0_LUMA_Y1));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_LUMA_Y1));
 		writel_relaxed(priv->viu.vd1_if0_luma_y0,
-				priv->io_base + _REG(VD2_IF0_LUMA_Y0));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD2_IF0_LUMA_Y0));
 		writel_relaxed(priv->viu.vd1_if0_luma_y0,
-				priv->io_base + _REG(VD2_IF0_LUMA_Y1));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD2_IF0_LUMA_Y1));
 		writel_relaxed(priv->viu.vd1_if0_chroma_x0,
-				priv->io_base + _REG(VD1_IF0_CHROMA_X0));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_CHROMA_X0));
 		writel_relaxed(priv->viu.vd1_if0_chroma_x0,
-				priv->io_base + _REG(VD1_IF0_CHROMA_X1));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_CHROMA_X1));
 		writel_relaxed(priv->viu.vd1_if0_chroma_x0,
-				priv->io_base + _REG(VD2_IF0_CHROMA_X0));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD2_IF0_CHROMA_X0));
 		writel_relaxed(priv->viu.vd1_if0_chroma_x0,
-				priv->io_base + _REG(VD2_IF0_CHROMA_X1));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD2_IF0_CHROMA_X1));
 		writel_relaxed(priv->viu.vd1_if0_chroma_y0,
-				priv->io_base + _REG(VD1_IF0_CHROMA_Y0));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_CHROMA_Y0));
 		writel_relaxed(priv->viu.vd1_if0_chroma_y0,
-				priv->io_base + _REG(VD1_IF0_CHROMA_Y1));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_CHROMA_Y1));
 		writel_relaxed(priv->viu.vd1_if0_chroma_y0,
-				priv->io_base + _REG(VD2_IF0_CHROMA_Y0));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD2_IF0_CHROMA_Y0));
 		writel_relaxed(priv->viu.vd1_if0_chroma_y0,
-				priv->io_base + _REG(VD2_IF0_CHROMA_Y1));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD2_IF0_CHROMA_Y1));
 		writel_relaxed(priv->viu.vd1_if0_repeat_loop,
-				priv->io_base + _REG(VD1_IF0_RPT_LOOP));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_RPT_LOOP));
 		writel_relaxed(priv->viu.vd1_if0_repeat_loop,
-				priv->io_base + _REG(VD2_IF0_RPT_LOOP));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD2_IF0_RPT_LOOP));
 		writel_relaxed(priv->viu.vd1_if0_luma0_rpt_pat,
-				priv->io_base + _REG(VD1_IF0_LUMA0_RPT_PAT));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_LUMA0_RPT_PAT));
 		writel_relaxed(priv->viu.vd1_if0_luma0_rpt_pat,
-				priv->io_base + _REG(VD2_IF0_LUMA0_RPT_PAT));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD2_IF0_LUMA0_RPT_PAT));
 		writel_relaxed(priv->viu.vd1_if0_luma0_rpt_pat,
-				priv->io_base + _REG(VD1_IF0_LUMA1_RPT_PAT));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_LUMA1_RPT_PAT));
 		writel_relaxed(priv->viu.vd1_if0_luma0_rpt_pat,
-				priv->io_base + _REG(VD2_IF0_LUMA1_RPT_PAT));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD2_IF0_LUMA1_RPT_PAT));
 		writel_relaxed(priv->viu.vd1_if0_chroma0_rpt_pat,
-				priv->io_base + _REG(VD1_IF0_CHROMA0_RPT_PAT));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_CHROMA0_RPT_PAT));
 		writel_relaxed(priv->viu.vd1_if0_chroma0_rpt_pat,
-				priv->io_base + _REG(VD2_IF0_CHROMA0_RPT_PAT));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD2_IF0_CHROMA0_RPT_PAT));
 		writel_relaxed(priv->viu.vd1_if0_chroma0_rpt_pat,
-				priv->io_base + _REG(VD1_IF0_CHROMA1_RPT_PAT));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_CHROMA1_RPT_PAT));
 		writel_relaxed(priv->viu.vd1_if0_chroma0_rpt_pat,
-				priv->io_base + _REG(VD2_IF0_CHROMA1_RPT_PAT));
-		writel_relaxed(0, priv->io_base + _REG(VD1_IF0_LUMA_PSEL));
-		writel_relaxed(0, priv->io_base + _REG(VD1_IF0_CHROMA_PSEL));
-		writel_relaxed(0, priv->io_base + _REG(VD2_IF0_LUMA_PSEL));
-		writel_relaxed(0, priv->io_base + _REG(VD2_IF0_CHROMA_PSEL));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD2_IF0_CHROMA1_RPT_PAT));
+		writel_relaxed(0, priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_LUMA_PSEL));
+		writel_relaxed(0, priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_CHROMA_PSEL));
+		writel_relaxed(0, priv->io_base + meson_crtc->viu_offset +
+				_REG(VD2_IF0_LUMA_PSEL));
+		writel_relaxed(0, priv->io_base + meson_crtc->viu_offset +
+				_REG(VD2_IF0_CHROMA_PSEL));
 		writel_relaxed(priv->viu.vd1_range_map_y,
-				priv->io_base + _REG(VD1_IF0_RANGE_MAP_Y));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_RANGE_MAP_Y));
 		writel_relaxed(priv->viu.vd1_range_map_cb,
-				priv->io_base + _REG(VD1_IF0_RANGE_MAP_CB));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_RANGE_MAP_CB));
 		writel_relaxed(priv->viu.vd1_range_map_cr,
-				priv->io_base + _REG(VD1_IF0_RANGE_MAP_CR));
+				priv->io_base + meson_crtc->viu_offset +
+				_REG(VD1_IF0_RANGE_MAP_CR));
 		writel_relaxed(0x78404,
 				priv->io_base + _REG(VPP_SC_MISC));
 		writel_relaxed(priv->viu.vpp_pic_in_height,
@@ -423,11 +548,8 @@ void meson_crtc_irq(struct meson_drm *priv)
 		writel_relaxed(0x42, priv->io_base + _REG(VPP_SCALE_COEF_IDX));
 
 		/* Enable VD1 */
-		writel_bits_relaxed(VPP_VD1_PREBLEND | VPP_VD1_POSTBLEND |
-				    VPP_COLOR_MNG_ENABLE,
-				    VPP_VD1_PREBLEND | VPP_VD1_POSTBLEND |
-				    VPP_COLOR_MNG_ENABLE,
-				    priv->io_base + _REG(VPP_MISC));
+		if (meson_crtc->enable_vd1)
+			meson_crtc->enable_vd1(priv);
 
 		priv->viu.vd1_commit = false;
 	}
@@ -464,7 +586,16 @@ int meson_crtc_create(struct meson_drm *priv)
 		return ret;
 	}
 
-	drm_crtc_helper_add(crtc, &meson_crtc_helper_funcs);
+	if (meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu")) {
+		meson_crtc->enable_osd1 = meson_g12a_crtc_enable_osd1;
+		meson_crtc->enable_vd1 = meson_g12a_crtc_enable_vd1;
+		meson_crtc->viu_offset = MESON_G12A_VIU_OFFSET;
+		drm_crtc_helper_add(crtc, &meson_g12a_crtc_helper_funcs);
+	} else {
+		meson_crtc->enable_osd1 = meson_crtc_enable_osd1;
+		meson_crtc->enable_vd1 = meson_crtc_enable_vd1;
+		drm_crtc_helper_add(crtc, &meson_crtc_helper_funcs);
+	}
 
 	priv->crtc = crtc;
 
diff --git a/drivers/gpu/drm/meson/meson_drv.c b/drivers/gpu/drm/meson/meson_drv.c
index 8a4ebcb6405c..72b01e6be0d9 100644
--- a/drivers/gpu/drm/meson/meson_drv.c
+++ b/drivers/gpu/drm/meson/meson_drv.c
@@ -48,7 +48,6 @@
 #include "meson_vpp.h"
 #include "meson_viu.h"
 #include "meson_venc.h"
-#include "meson_canvas.h"
 #include "meson_registers.h"
 
 #define DRIVER_NAME "meson"
@@ -91,6 +90,18 @@ static irqreturn_t meson_irq(int irq, void *arg)
 	return IRQ_HANDLED;
 }
 
+static int meson_dumb_create(struct drm_file *file, struct drm_device *dev,
+			     struct drm_mode_create_dumb *args)
+{
+	/*
+	 * We need 64bytes aligned stride, and PAGE aligned size
+	 */
+	args->pitch = ALIGN(DIV_ROUND_UP(args->width * args->bpp, 8), SZ_64);
+	args->size = PAGE_ALIGN(args->pitch * args->height);
+
+	return drm_gem_cma_dumb_create_internal(file, dev, args);
+}
+
 DEFINE_DRM_GEM_CMA_FOPS(fops);
 
 static struct drm_driver meson_driver = {
@@ -113,7 +124,7 @@ static struct drm_driver meson_driver = {
 	.gem_prime_mmap		= drm_gem_cma_prime_mmap,
 
 	/* GEM Ops */
-	.dumb_create		= drm_gem_cma_dumb_create,
+	.dumb_create		= meson_dumb_create,
 	.gem_free_object_unlocked = drm_gem_cma_free_object,
 	.gem_vm_ops		= &drm_gem_cma_vm_ops,
 
@@ -231,50 +242,31 @@ static int meson_drv_bind_master(struct device *dev, bool has_components)
 	}
 
 	priv->canvas = meson_canvas_get(dev);
-	if (!IS_ERR(priv->canvas)) {
-		ret = meson_canvas_alloc(priv->canvas, &priv->canvas_id_osd1);
-		if (ret)
-			goto free_drm;
-		ret = meson_canvas_alloc(priv->canvas, &priv->canvas_id_vd1_0);
-		if (ret) {
-			meson_canvas_free(priv->canvas, priv->canvas_id_osd1);
-			goto free_drm;
-		}
-		ret = meson_canvas_alloc(priv->canvas, &priv->canvas_id_vd1_1);
-		if (ret) {
-			meson_canvas_free(priv->canvas, priv->canvas_id_osd1);
-			meson_canvas_free(priv->canvas, priv->canvas_id_vd1_0);
-			goto free_drm;
-		}
-		ret = meson_canvas_alloc(priv->canvas, &priv->canvas_id_vd1_2);
-		if (ret) {
-			meson_canvas_free(priv->canvas, priv->canvas_id_osd1);
-			meson_canvas_free(priv->canvas, priv->canvas_id_vd1_0);
-			meson_canvas_free(priv->canvas, priv->canvas_id_vd1_1);
-			goto free_drm;
-		}
-	} else {
-		priv->canvas = NULL;
-
-		res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "dmc");
-		if (!res) {
-			ret = -EINVAL;
-			goto free_drm;
-		}
-		/* Simply ioremap since it may be a shared register zone */
-		regs = devm_ioremap(dev, res->start, resource_size(res));
-		if (!regs) {
-			ret = -EADDRNOTAVAIL;
-			goto free_drm;
-		}
+	if (IS_ERR(priv->canvas)) {
+		ret = PTR_ERR(priv->canvas);
+		goto free_drm;
+	}
 
-		priv->dmc = devm_regmap_init_mmio(dev, regs,
-						  &meson_regmap_config);
-		if (IS_ERR(priv->dmc)) {
-			dev_err(&pdev->dev, "Couldn't create the DMC regmap\n");
-			ret = PTR_ERR(priv->dmc);
-			goto free_drm;
-		}
+	ret = meson_canvas_alloc(priv->canvas, &priv->canvas_id_osd1);
+	if (ret)
+		goto free_drm;
+	ret = meson_canvas_alloc(priv->canvas, &priv->canvas_id_vd1_0);
+	if (ret) {
+		meson_canvas_free(priv->canvas, priv->canvas_id_osd1);
+		goto free_drm;
+	}
+	ret = meson_canvas_alloc(priv->canvas, &priv->canvas_id_vd1_1);
+	if (ret) {
+		meson_canvas_free(priv->canvas, priv->canvas_id_osd1);
+		meson_canvas_free(priv->canvas, priv->canvas_id_vd1_0);
+		goto free_drm;
+	}
+	ret = meson_canvas_alloc(priv->canvas, &priv->canvas_id_vd1_2);
+	if (ret) {
+		meson_canvas_free(priv->canvas, priv->canvas_id_osd1);
+		meson_canvas_free(priv->canvas, priv->canvas_id_vd1_0);
+		meson_canvas_free(priv->canvas, priv->canvas_id_vd1_1);
+		goto free_drm;
 	}
 
 	priv->vsync_irq = platform_get_irq(pdev, 0);
@@ -467,6 +459,7 @@ static const struct of_device_id dt_match[] = {
 	{ .compatible = "amlogic,meson-gxbb-vpu" },
 	{ .compatible = "amlogic,meson-gxl-vpu" },
 	{ .compatible = "amlogic,meson-gxm-vpu" },
+	{ .compatible = "amlogic,meson-g12a-vpu" },
 	{}
 };
 MODULE_DEVICE_TABLE(of, dt_match);
diff --git a/drivers/gpu/drm/meson/meson_drv.h b/drivers/gpu/drm/meson/meson_drv.h
index 4dccf4cd042a..9614baa836b9 100644
--- a/drivers/gpu/drm/meson/meson_drv.h
+++ b/drivers/gpu/drm/meson/meson_drv.h
@@ -29,7 +29,6 @@ struct meson_drm {
 	struct device *dev;
 	void __iomem *io_base;
 	struct regmap *hhi;
-	struct regmap *dmc;
 	int vsync_irq;
 
 	struct meson_canvas *canvas;
@@ -63,6 +62,10 @@ struct meson_drm {
 		uint32_t osd_sc_h_phase_step;
 		uint32_t osd_sc_h_ctrl0;
 		uint32_t osd_sc_v_ctrl0;
+		uint32_t osd_blend_din0_scope_h;
+		uint32_t osd_blend_din0_scope_v;
+		uint32_t osb_blend0_size;
+		uint32_t osb_blend1_size;
 
 		bool vd1_enabled;
 		bool vd1_commit;
diff --git a/drivers/gpu/drm/meson/meson_dw_hdmi.c b/drivers/gpu/drm/meson/meson_dw_hdmi.c
index 563953ec6ad0..779da21143b9 100644
--- a/drivers/gpu/drm/meson/meson_dw_hdmi.c
+++ b/drivers/gpu/drm/meson/meson_dw_hdmi.c
@@ -20,6 +20,7 @@
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/component.h>
+#include <linux/of_device.h>
 #include <linux/of_graph.h>
 #include <linux/reset.h>
 #include <linux/clk.h>
@@ -105,6 +106,7 @@
 #define HDMITX_TOP_ADDR_REG	0x0
 #define HDMITX_TOP_DATA_REG	0x4
 #define HDMITX_TOP_CTRL_REG	0x8
+#define HDMITX_TOP_G12A_OFFSET	0x8000
 
 /* Controller Communication Channel */
 #define HDMITX_DWC_ADDR_REG	0x10
@@ -118,6 +120,8 @@
 #define HHI_HDMI_PHY_CNTL1	0x3a4 /* 0xe9 */
 #define HHI_HDMI_PHY_CNTL2	0x3a8 /* 0xea */
 #define HHI_HDMI_PHY_CNTL3	0x3ac /* 0xeb */
+#define HHI_HDMI_PHY_CNTL4	0x3b0 /* 0xec */
+#define HHI_HDMI_PHY_CNTL5	0x3b4 /* 0xed */
 
 static DEFINE_SPINLOCK(reg_lock);
 
@@ -127,12 +131,26 @@ enum meson_venc_source {
 	MESON_VENC_SOURCE_ENCP = 2,
 };
 
+struct meson_dw_hdmi;
+
+struct meson_dw_hdmi_data {
+	unsigned int	(*top_read)(struct meson_dw_hdmi *dw_hdmi,
+				    unsigned int addr);
+	void		(*top_write)(struct meson_dw_hdmi *dw_hdmi,
+				     unsigned int addr, unsigned int data);
+	unsigned int	(*dwc_read)(struct meson_dw_hdmi *dw_hdmi,
+				    unsigned int addr);
+	void		(*dwc_write)(struct meson_dw_hdmi *dw_hdmi,
+				     unsigned int addr, unsigned int data);
+};
+
 struct meson_dw_hdmi {
 	struct drm_encoder encoder;
 	struct dw_hdmi_plat_data dw_plat_data;
 	struct meson_drm *priv;
 	struct device *dev;
 	void __iomem *hdmitx;
+	const struct meson_dw_hdmi_data *data;
 	struct reset_control *hdmitx_apb;
 	struct reset_control *hdmitx_ctrl;
 	struct reset_control *hdmitx_phy;
@@ -174,6 +192,12 @@ static unsigned int dw_hdmi_top_read(struct meson_dw_hdmi *dw_hdmi,
 	return data;
 }
 
+static unsigned int dw_hdmi_g12a_top_read(struct meson_dw_hdmi *dw_hdmi,
+					  unsigned int addr)
+{
+	return readl(dw_hdmi->hdmitx + HDMITX_TOP_G12A_OFFSET + (addr << 2));
+}
+
 static inline void dw_hdmi_top_write(struct meson_dw_hdmi *dw_hdmi,
 				     unsigned int addr, unsigned int data)
 {
@@ -191,18 +215,24 @@ static inline void dw_hdmi_top_write(struct meson_dw_hdmi *dw_hdmi,
 	spin_unlock_irqrestore(&reg_lock, flags);
 }
 
+static inline void dw_hdmi_g12a_top_write(struct meson_dw_hdmi *dw_hdmi,
+					  unsigned int addr, unsigned int data)
+{
+	writel(data, dw_hdmi->hdmitx + HDMITX_TOP_G12A_OFFSET + (addr << 2));
+}
+
 /* Helper to change specific bits in PHY registers */
 static inline void dw_hdmi_top_write_bits(struct meson_dw_hdmi *dw_hdmi,
 					  unsigned int addr,
 					  unsigned int mask,
 					  unsigned int val)
 {
-	unsigned int data = dw_hdmi_top_read(dw_hdmi, addr);
+	unsigned int data = dw_hdmi->data->top_read(dw_hdmi, addr);
 
 	data &= ~mask;
 	data |= val;
 
-	dw_hdmi_top_write(dw_hdmi, addr, data);
+	dw_hdmi->data->top_write(dw_hdmi, addr, data);
 }
 
 static unsigned int dw_hdmi_dwc_read(struct meson_dw_hdmi *dw_hdmi,
@@ -226,6 +256,12 @@ static unsigned int dw_hdmi_dwc_read(struct meson_dw_hdmi *dw_hdmi,
 	return data;
 }
 
+static unsigned int dw_hdmi_g12a_dwc_read(struct meson_dw_hdmi *dw_hdmi,
+					  unsigned int addr)
+{
+	return readb(dw_hdmi->hdmitx + addr);
+}
+
 static inline void dw_hdmi_dwc_write(struct meson_dw_hdmi *dw_hdmi,
 				     unsigned int addr, unsigned int data)
 {
@@ -243,18 +279,24 @@ static inline void dw_hdmi_dwc_write(struct meson_dw_hdmi *dw_hdmi,
 	spin_unlock_irqrestore(&reg_lock, flags);
 }
 
+static inline void dw_hdmi_g12a_dwc_write(struct meson_dw_hdmi *dw_hdmi,
+					  unsigned int addr, unsigned int data)
+{
+	writeb(data, dw_hdmi->hdmitx + addr);
+}
+
 /* Helper to change specific bits in controller registers */
 static inline void dw_hdmi_dwc_write_bits(struct meson_dw_hdmi *dw_hdmi,
 					  unsigned int addr,
 					  unsigned int mask,
 					  unsigned int val)
 {
-	unsigned int data = dw_hdmi_dwc_read(dw_hdmi, addr);
+	unsigned int data = dw_hdmi->data->dwc_read(dw_hdmi, addr);
 
 	data &= ~mask;
 	data |= val;
 
-	dw_hdmi_dwc_write(dw_hdmi, addr, data);
+	dw_hdmi->data->dwc_write(dw_hdmi, addr, data);
 }
 
 /* Bridge */
@@ -300,6 +342,24 @@ static void meson_hdmi_phy_setup_mode(struct meson_dw_hdmi *dw_hdmi,
 			regmap_write(priv->hhi, HHI_HDMI_PHY_CNTL0, 0x33632122);
 			regmap_write(priv->hhi, HHI_HDMI_PHY_CNTL3, 0x2000115b);
 		}
+	} else if (dw_hdmi_is_compatible(dw_hdmi,
+					 "amlogic,meson-g12a-dw-hdmi")) {
+		if (pixel_clock >= 371250) {
+			/* 5.94Gbps, 3.7125Gbps */
+			regmap_write(priv->hhi, HHI_HDMI_PHY_CNTL0, 0x37eb65c4);
+			regmap_write(priv->hhi, HHI_HDMI_PHY_CNTL3, 0x2ab0ff3b);
+			regmap_write(priv->hhi, HHI_HDMI_PHY_CNTL5, 0x0000080b);
+		} else if (pixel_clock >= 297000) {
+			/* 2.97Gbps */
+			regmap_write(priv->hhi, HHI_HDMI_PHY_CNTL0, 0x33eb6262);
+			regmap_write(priv->hhi, HHI_HDMI_PHY_CNTL3, 0x2ab0ff3b);
+			regmap_write(priv->hhi, HHI_HDMI_PHY_CNTL5, 0x00000003);
+		} else {
+			/* 1.485Gbps, and below */
+			regmap_write(priv->hhi, HHI_HDMI_PHY_CNTL0, 0x33eb4242);
+			regmap_write(priv->hhi, HHI_HDMI_PHY_CNTL3, 0x2ab0ff3b);
+			regmap_write(priv->hhi, HHI_HDMI_PHY_CNTL5, 0x00000003);
+		}
 	}
 }
 
@@ -375,7 +435,7 @@ static int dw_hdmi_phy_init(struct dw_hdmi *hdmi, void *data,
 	regmap_update_bits(priv->hhi, HHI_MEM_PD_REG0, 0xff << 8, 0);
 
 	/* Bring out of reset */
-	dw_hdmi_top_write(dw_hdmi, HDMITX_TOP_SW_RESET,  0);
+	dw_hdmi->data->top_write(dw_hdmi, HDMITX_TOP_SW_RESET,  0);
 
 	/* Enable internal pixclk, tmds_clk, spdif_clk, i2s_clk, cecclk */
 	dw_hdmi_top_write_bits(dw_hdmi, HDMITX_TOP_CLK_CNTL,
@@ -384,24 +444,25 @@ static int dw_hdmi_phy_init(struct dw_hdmi *hdmi, void *data,
 			       0x3 << 4, 0x3 << 4);
 
 	/* Enable normal output to PHY */
-	dw_hdmi_top_write(dw_hdmi, HDMITX_TOP_BIST_CNTL, BIT(12));
+	dw_hdmi->data->top_write(dw_hdmi, HDMITX_TOP_BIST_CNTL, BIT(12));
 
 	/* TMDS pattern setup (TOFIX Handle the YUV420 case) */
 	if (mode->clock > 340000) {
-		dw_hdmi_top_write(dw_hdmi, HDMITX_TOP_TMDS_CLK_PTTN_01, 0);
-		dw_hdmi_top_write(dw_hdmi, HDMITX_TOP_TMDS_CLK_PTTN_23,
+		dw_hdmi->data->top_write(dw_hdmi, HDMITX_TOP_TMDS_CLK_PTTN_01,
+				  0);
+		dw_hdmi->data->top_write(dw_hdmi, HDMITX_TOP_TMDS_CLK_PTTN_23,
 				  0x03ff03ff);
 	} else {
-		dw_hdmi_top_write(dw_hdmi, HDMITX_TOP_TMDS_CLK_PTTN_01,
+		dw_hdmi->data->top_write(dw_hdmi, HDMITX_TOP_TMDS_CLK_PTTN_01,
 				  0x001f001f);
-		dw_hdmi_top_write(dw_hdmi, HDMITX_TOP_TMDS_CLK_PTTN_23,
+		dw_hdmi->data->top_write(dw_hdmi, HDMITX_TOP_TMDS_CLK_PTTN_23,
 				  0x001f001f);
 	}
 
 	/* Load TMDS pattern */
-	dw_hdmi_top_write(dw_hdmi, HDMITX_TOP_TMDS_CLK_PTTN_CNTL, 0x1);
+	dw_hdmi->data->top_write(dw_hdmi, HDMITX_TOP_TMDS_CLK_PTTN_CNTL, 0x1);
 	msleep(20);
-	dw_hdmi_top_write(dw_hdmi, HDMITX_TOP_TMDS_CLK_PTTN_CNTL, 0x2);
+	dw_hdmi->data->top_write(dw_hdmi, HDMITX_TOP_TMDS_CLK_PTTN_CNTL, 0x2);
 
 	/* Setup PHY parameters */
 	meson_hdmi_phy_setup_mode(dw_hdmi, mode);
@@ -412,7 +473,8 @@ static int dw_hdmi_phy_init(struct dw_hdmi *hdmi, void *data,
 
 	/* BIT_INVERT */
 	if (dw_hdmi_is_compatible(dw_hdmi, "amlogic,meson-gxl-dw-hdmi") ||
-	    dw_hdmi_is_compatible(dw_hdmi, "amlogic,meson-gxm-dw-hdmi"))
+	    dw_hdmi_is_compatible(dw_hdmi, "amlogic,meson-gxm-dw-hdmi") ||
+	    dw_hdmi_is_compatible(dw_hdmi, "amlogic,meson-g12a-dw-hdmi"))
 		regmap_update_bits(priv->hhi, HHI_HDMI_PHY_CNTL1,
 				   BIT(17), 0);
 	else
@@ -480,7 +542,7 @@ static enum drm_connector_status dw_hdmi_read_hpd(struct dw_hdmi *hdmi,
 {
 	struct meson_dw_hdmi *dw_hdmi = (struct meson_dw_hdmi *)data;
 
-	return !!dw_hdmi_top_read(dw_hdmi, HDMITX_TOP_STAT0) ?
+	return !!dw_hdmi->data->top_read(dw_hdmi, HDMITX_TOP_STAT0) ?
 		connector_status_connected : connector_status_disconnected;
 }
 
@@ -490,11 +552,11 @@ static void dw_hdmi_setup_hpd(struct dw_hdmi *hdmi,
 	struct meson_dw_hdmi *dw_hdmi = (struct meson_dw_hdmi *)data;
 
 	/* Setup HPD Filter */
-	dw_hdmi_top_write(dw_hdmi, HDMITX_TOP_HPD_FILTER,
+	dw_hdmi->data->top_write(dw_hdmi, HDMITX_TOP_HPD_FILTER,
 			  (0xa << 12) | 0xa0);
 
 	/* Clear interrupts */
-	dw_hdmi_top_write(dw_hdmi, HDMITX_TOP_INTR_STAT_CLR,
+	dw_hdmi->data->top_write(dw_hdmi, HDMITX_TOP_INTR_STAT_CLR,
 			  HDMITX_TOP_INTR_HPD_RISE | HDMITX_TOP_INTR_HPD_FALL);
 
 	/* Unmask interrupts */
@@ -515,8 +577,8 @@ static irqreturn_t dw_hdmi_top_irq(int irq, void *dev_id)
 	struct meson_dw_hdmi *dw_hdmi = dev_id;
 	u32 stat;
 
-	stat = dw_hdmi_top_read(dw_hdmi, HDMITX_TOP_INTR_STAT);
-	dw_hdmi_top_write(dw_hdmi, HDMITX_TOP_INTR_STAT_CLR, stat);
+	stat = dw_hdmi->data->top_read(dw_hdmi, HDMITX_TOP_INTR_STAT);
+	dw_hdmi->data->top_write(dw_hdmi, HDMITX_TOP_INTR_STAT_CLR, stat);
 
 	/* HPD Events, handle in the threaded interrupt handler */
 	if (stat & (HDMITX_TOP_INTR_HPD_RISE | HDMITX_TOP_INTR_HPD_FALL)) {
@@ -686,7 +748,9 @@ static const struct drm_encoder_helper_funcs
 static int meson_dw_hdmi_reg_read(void *context, unsigned int reg,
 				  unsigned int *result)
 {
-	*result = dw_hdmi_dwc_read(context, reg);
+	struct meson_dw_hdmi *dw_hdmi = context;
+
+	*result = dw_hdmi->data->dwc_read(dw_hdmi, reg);
 
 	return 0;
 
@@ -695,7 +759,9 @@ static int meson_dw_hdmi_reg_read(void *context, unsigned int reg,
 static int meson_dw_hdmi_reg_write(void *context, unsigned int reg,
 				   unsigned int val)
 {
-	dw_hdmi_dwc_write(context, reg, val);
+	struct meson_dw_hdmi *dw_hdmi = context;
+
+	dw_hdmi->data->dwc_write(dw_hdmi, reg, val);
 
 	return 0;
 }
@@ -709,6 +775,20 @@ static const struct regmap_config meson_dw_hdmi_regmap_config = {
 	.fast_io = true,
 };
 
+static const struct meson_dw_hdmi_data meson_dw_hdmi_gx_data = {
+	.top_read = dw_hdmi_top_read,
+	.top_write = dw_hdmi_top_write,
+	.dwc_read = dw_hdmi_dwc_read,
+	.dwc_write = dw_hdmi_dwc_write,
+};
+
+static const struct meson_dw_hdmi_data meson_dw_hdmi_g12a_data = {
+	.top_read = dw_hdmi_g12a_top_read,
+	.top_write = dw_hdmi_g12a_top_write,
+	.dwc_read = dw_hdmi_g12a_dwc_read,
+	.dwc_write = dw_hdmi_g12a_dwc_write,
+};
+
 static bool meson_hdmi_connector_is_available(struct device *dev)
 {
 	struct device_node *ep, *remote;
@@ -735,6 +815,7 @@ static int meson_dw_hdmi_bind(struct device *dev, struct device *master,
 				void *data)
 {
 	struct platform_device *pdev = to_platform_device(dev);
+	const struct meson_dw_hdmi_data *match;
 	struct meson_dw_hdmi *meson_dw_hdmi;
 	struct drm_device *drm = data;
 	struct meson_drm *priv = drm->dev_private;
@@ -751,6 +832,12 @@ static int meson_dw_hdmi_bind(struct device *dev, struct device *master,
 		return -ENODEV;
 	}
 
+	match = of_device_get_match_data(&pdev->dev);
+	if (!match) {
+		dev_err(&pdev->dev, "failed to get match data\n");
+		return -ENODEV;
+	}
+
 	meson_dw_hdmi = devm_kzalloc(dev, sizeof(*meson_dw_hdmi),
 				     GFP_KERNEL);
 	if (!meson_dw_hdmi)
@@ -758,6 +845,7 @@ static int meson_dw_hdmi_bind(struct device *dev, struct device *master,
 
 	meson_dw_hdmi->priv = priv;
 	meson_dw_hdmi->dev = dev;
+	meson_dw_hdmi->data = match;
 	dw_plat_data = &meson_dw_hdmi->dw_plat_data;
 	encoder = &meson_dw_hdmi->encoder;
 
@@ -858,24 +946,28 @@ static int meson_dw_hdmi_bind(struct device *dev, struct device *master,
 	reset_control_reset(meson_dw_hdmi->hdmitx_phy);
 
 	/* Enable APB3 fail on error */
-	writel_bits_relaxed(BIT(15), BIT(15),
-			    meson_dw_hdmi->hdmitx + HDMITX_TOP_CTRL_REG);
-	writel_bits_relaxed(BIT(15), BIT(15),
-			    meson_dw_hdmi->hdmitx + HDMITX_DWC_CTRL_REG);
+	if (!meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu")) {
+		writel_bits_relaxed(BIT(15), BIT(15),
+				    meson_dw_hdmi->hdmitx + HDMITX_TOP_CTRL_REG);
+		writel_bits_relaxed(BIT(15), BIT(15),
+				    meson_dw_hdmi->hdmitx + HDMITX_DWC_CTRL_REG);
+	}
 
 	/* Bring out of reset */
-	dw_hdmi_top_write(meson_dw_hdmi, HDMITX_TOP_SW_RESET,  0);
+	meson_dw_hdmi->data->top_write(meson_dw_hdmi,
+				       HDMITX_TOP_SW_RESET,  0);
 
 	msleep(20);
 
-	dw_hdmi_top_write(meson_dw_hdmi, HDMITX_TOP_CLK_CNTL, 0xff);
+	meson_dw_hdmi->data->top_write(meson_dw_hdmi,
+				       HDMITX_TOP_CLK_CNTL, 0xff);
 
 	/* Enable HDMI-TX Interrupt */
-	dw_hdmi_top_write(meson_dw_hdmi, HDMITX_TOP_INTR_STAT_CLR,
-			  HDMITX_TOP_INTR_CORE);
+	meson_dw_hdmi->data->top_write(meson_dw_hdmi, HDMITX_TOP_INTR_STAT_CLR,
+				       HDMITX_TOP_INTR_CORE);
 
-	dw_hdmi_top_write(meson_dw_hdmi, HDMITX_TOP_INTR_MASKN,
-			  HDMITX_TOP_INTR_CORE);
+	meson_dw_hdmi->data->top_write(meson_dw_hdmi, HDMITX_TOP_INTR_MASKN,
+				       HDMITX_TOP_INTR_CORE);
 
 	/* Bridge / Connector */
 
@@ -924,9 +1016,14 @@ static int meson_dw_hdmi_remove(struct platform_device *pdev)
 }
 
 static const struct of_device_id meson_dw_hdmi_of_table[] = {
-	{ .compatible = "amlogic,meson-gxbb-dw-hdmi" },
-	{ .compatible = "amlogic,meson-gxl-dw-hdmi" },
-	{ .compatible = "amlogic,meson-gxm-dw-hdmi" },
+	{ .compatible = "amlogic,meson-gxbb-dw-hdmi",
+	  .data = &meson_dw_hdmi_gx_data },
+	{ .compatible = "amlogic,meson-gxl-dw-hdmi",
+	  .data = &meson_dw_hdmi_gx_data },
+	{ .compatible = "amlogic,meson-gxm-dw-hdmi",
+	  .data = &meson_dw_hdmi_gx_data },
+	{ .compatible = "amlogic,meson-g12a-dw-hdmi",
+	  .data = &meson_dw_hdmi_g12a_data },
 	{ }
 };
 MODULE_DEVICE_TABLE(of, meson_dw_hdmi_of_table);
diff --git a/drivers/gpu/drm/meson/meson_dw_hdmi.h b/drivers/gpu/drm/meson/meson_dw_hdmi.h
index 0b81183125e3..03e2f0c1a2d5 100644
--- a/drivers/gpu/drm/meson/meson_dw_hdmi.h
+++ b/drivers/gpu/drm/meson/meson_dw_hdmi.h
@@ -21,9 +21,12 @@
 #define __MESON_DW_HDMI_H
 
 /*
- * Bit 7 RW Reserved. Default 1.
- * Bit 6 RW Reserved. Default 1.
- * Bit 5 RW Reserved. Default 1.
+ * Bit 15-10: RW Reserved. Default 1 starting from G12A
+ * Bit 9 RW sw_reset_i2c starting from G12A
+ * Bit 8 RW sw_reset_axiarb starting from G12A
+ * Bit 7 RW Reserved. Default 1, sw_reset_emp starting from G12A
+ * Bit 6 RW Reserved. Default 1, sw_reset_flt starting from G12A
+ * Bit 5 RW Reserved. Default 1, sw_reset_hdcp22 starting from G12A
  * Bit 4 RW sw_reset_phyif: PHY interface. 1=Apply reset; 0=Release from reset.
  *     Default 1.
  * Bit 3 RW sw_reset_intr: interrupt module. 1=Apply reset;
@@ -39,12 +42,16 @@
 #define HDMITX_TOP_SW_RESET                     (0x000)
 
 /*
+ * Bit 31 RW free_clk_en: 0=Enable clock gating for power saving; 1= Disable
  * Bit 12 RW i2s_ws_inv:1=Invert i2s_ws; 0=No invert. Default 0.
  * Bit 11 RW i2s_clk_inv: 1=Invert i2s_clk; 0=No invert. Default 0.
  * Bit 10 RW spdif_clk_inv: 1=Invert spdif_clk; 0=No invert. Default 0.
  * Bit 9 RW tmds_clk_inv: 1=Invert tmds_clk; 0=No invert. Default 0.
  * Bit 8 RW pixel_clk_inv: 1=Invert pixel_clk; 0=No invert. Default 0.
- * Bit 4 RW cec_clk_en: 1=enable cec_clk; 0=disable. Default 0.
+ * Bit 7 RW hdcp22_skpclk_en: starting from G12A, 1=enable; 0=disable
+ * Bit 6 RW hdcp22_esmclk_en: starting from G12A, 1=enable; 0=disable
+ * Bit 5 RW hdcp22_tmdsclk_en: starting from G12A, 1=enable; 0=disable
+ * Bit 4 RW cec_clk_en: 1=enable cec_clk; 0=disable. Default 0. Reserved for G12A
  * Bit 3 RW i2s_clk_en: 1=enable i2s_clk; 0=disable. Default 0.
  * Bit 2 RW spdif_clk_en: 1=enable spdif_clk; 0=disable. Default 0.
  * Bit 1 RW tmds_clk_en: 1=enable tmds_clk;  0=disable. Default 0.
@@ -53,6 +60,8 @@
 #define HDMITX_TOP_CLK_CNTL                     (0x001)
 
 /*
+ * Bit 31:28 RW rxsense_glitch_width: starting from G12A
+ * Bit 27:16 RW rxsense_valid_width: starting from G12A
  * Bit 11: 0 RW hpd_valid_width: filter out width <= M*1024.    Default 0.
  * Bit 15:12 RW hpd_glitch_width: filter out glitch <= N.       Default 0.
  */
@@ -61,6 +70,9 @@
 /*
  * intr_maskn: MASK_N, one bit per interrupt source.
  *     1=Enable interrupt source; 0=Disable interrupt source. Default 0.
+ * [  7] rxsense_fall starting from G12A
+ * [  6] rxsense_rise starting from G12A
+ * [  5] err_i2c_timeout starting from G12A
  * [  4] hdcp22_rndnum_err
  * [  3] nonce_rfrsh_rise
  * [  2] hpd_fall_intr
@@ -73,6 +85,9 @@
  * Bit 30: 0 RW intr_stat: For each bit, write 1 to manually set the interrupt
  *     bit, read back the interrupt status.
  * Bit    31 R  IP interrupt status
+ * Bit     7 RW rxsense_fall starting from G12A
+ * Bit     6 RW rxsense_rise starting from G12A
+ * Bit     5 RW err_i2c_timeout starting from G12A
  * Bit     2 RW hpd_fall
  * Bit     1 RW hpd_rise
  * Bit     0 RW IP interrupt
@@ -80,6 +95,9 @@
 #define HDMITX_TOP_INTR_STAT                    (0x004)
 
 /*
+ * [7]    rxsense_fall starting from G12A
+ * [6]    rxsense_rise starting from G12A
+ * [5]    err_i2c_timeout starting from G12A
  * [4]	  hdcp22_rndnum_err
  * [3]	  nonce_rfrsh_rise
  * [2]	  hpd_fall
@@ -91,6 +109,8 @@
 #define HDMITX_TOP_INTR_CORE		BIT(0)
 #define HDMITX_TOP_INTR_HPD_RISE	BIT(1)
 #define HDMITX_TOP_INTR_HPD_FALL	BIT(2)
+#define HDMITX_TOP_INTR_RXSENSE_RISE	BIT(6)
+#define HDMITX_TOP_INTR_RXSENSE_FALL	BIT(7)
 
 /* Bit 14:12 RW tmds_sel: 3'b000=Output zero; 3'b001=Output normal TMDS data;
  *     3'b010=Output PRBS data; 3'b100=Output shift pattern. Default 0.
@@ -140,7 +160,9 @@
  */
 #define HDMITX_TOP_REVOCMEM_STAT                (0x00D)
 
-/* Bit     0 R  filtered HPD status. */
+/* Bit	   1 R	filtered RxSense status
+ * Bit     0 R  filtered HPD status.
+ */
 #define HDMITX_TOP_STAT0                        (0x00E)
 
 #endif /* __MESON_DW_HDMI_H */
diff --git a/drivers/gpu/drm/meson/meson_overlay.c b/drivers/gpu/drm/meson/meson_overlay.c
index 691a9fd16b36..bdbf925ff3e8 100644
--- a/drivers/gpu/drm/meson/meson_overlay.c
+++ b/drivers/gpu/drm/meson/meson_overlay.c
@@ -22,7 +22,6 @@
 #include "meson_overlay.h"
 #include "meson_vpp.h"
 #include "meson_viu.h"
-#include "meson_canvas.h"
 #include "meson_registers.h"
 
 /* VD1_IF0_GEN_REG */
@@ -350,13 +349,6 @@ static void meson_overlay_atomic_update(struct drm_plane *plane,
 
 	DRM_DEBUG_DRIVER("\n");
 
-	/* Fallback is canvas provider is not available */
-	if (!priv->canvas) {
-		priv->canvas_id_vd1_0 = MESON_CANVAS_ID_VD1_0;
-		priv->canvas_id_vd1_1 = MESON_CANVAS_ID_VD1_1;
-		priv->canvas_id_vd1_2 = MESON_CANVAS_ID_VD1_2;
-	}
-
 	interlace_mode = state->crtc->mode.flags & DRM_MODE_FLAG_INTERLACE;
 
 	spin_lock_irqsave(&priv->drm->event_lock, flags);
@@ -524,8 +516,14 @@ static void meson_overlay_atomic_disable(struct drm_plane *plane,
 	priv->viu.vd1_enabled = false;
 
 	/* Disable VD1 */
-	writel_bits_relaxed(VPP_VD1_POSTBLEND | VPP_VD1_PREBLEND, 0,
-			    priv->io_base + _REG(VPP_MISC));
+	if (meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu")) {
+		writel_relaxed(0, priv->io_base + _REG(VD1_BLEND_SRC_CTRL));
+		writel_relaxed(0, priv->io_base + _REG(VD2_BLEND_SRC_CTRL));
+		writel_relaxed(0, priv->io_base + _REG(VD1_IF0_GEN_REG + 0x17b0));
+		writel_relaxed(0, priv->io_base + _REG(VD2_IF0_GEN_REG + 0x17b0));
+	} else
+		writel_bits_relaxed(VPP_VD1_POSTBLEND | VPP_VD1_PREBLEND, 0,
+				    priv->io_base + _REG(VPP_MISC));
 
 }
 
diff --git a/drivers/gpu/drm/meson/meson_plane.c b/drivers/gpu/drm/meson/meson_plane.c
index 6119a0224278..bf8f1fab63aa 100644
--- a/drivers/gpu/drm/meson/meson_plane.c
+++ b/drivers/gpu/drm/meson/meson_plane.c
@@ -38,7 +38,6 @@
 #include "meson_plane.h"
 #include "meson_vpp.h"
 #include "meson_viu.h"
-#include "meson_canvas.h"
 #include "meson_registers.h"
 
 /* OSD_SCI_WH_M1 */
@@ -148,10 +147,7 @@ static void meson_plane_atomic_update(struct drm_plane *plane,
 				   (0xFF << OSD_GLOBAL_ALPHA_SHIFT) |
 				   OSD_BLK0_ENABLE;
 
-	if (priv->canvas)
-		canvas_id_osd1 = priv->canvas_id_osd1;
-	else
-		canvas_id_osd1 = MESON_CANVAS_ID_OSD1;
+	canvas_id_osd1 = priv->canvas_id_osd1;
 
 	/* Set up BLK0 to point to the right canvas */
 	priv->viu.osd1_blk0_cfg[0] = ((canvas_id_osd1 << OSD_CANVAS_SEL) |
@@ -298,6 +294,13 @@ static void meson_plane_atomic_update(struct drm_plane *plane,
 	priv->viu.osd1_blk0_cfg[3] = ((dest.x2 - 1) << 16) | dest.x1;
 	priv->viu.osd1_blk0_cfg[4] = ((dest.y2 - 1) << 16) | dest.y1;
 
+	if (meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu")) {
+		priv->viu.osd_blend_din0_scope_h = ((dest.x2 - 1) << 16) | dest.x1;
+		priv->viu.osd_blend_din0_scope_v = ((dest.y2 - 1) << 16) | dest.y1;
+		priv->viu.osb_blend0_size = dst_h << 16 | dst_w;
+		priv->viu.osb_blend1_size = dst_h << 16 | dst_w;
+	}
+
 	/* Update Canvas with buffer address */
 	gem = drm_fb_cma_get_gem_obj(fb, 0);
 
@@ -324,8 +327,12 @@ static void meson_plane_atomic_disable(struct drm_plane *plane,
 	struct meson_drm *priv = meson_plane->priv;
 
 	/* Disable OSD1 */
-	writel_bits_relaxed(VPP_OSD1_POSTBLEND, 0,
-			    priv->io_base + _REG(VPP_MISC));
+	if (meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu"))
+		writel_bits_relaxed(BIT(0) | BIT(21), 0,
+			priv->io_base + _REG(VIU_OSD1_CTRL_STAT));
+	else
+		writel_bits_relaxed(VPP_OSD1_POSTBLEND, 0,
+				    priv->io_base + _REG(VPP_MISC));
 
 	meson_plane->enabled = false;
 
diff --git a/drivers/gpu/drm/meson/meson_registers.h b/drivers/gpu/drm/meson/meson_registers.h
index 5c7e02c703bc..cfaf90501bb1 100644
--- a/drivers/gpu/drm/meson/meson_registers.h
+++ b/drivers/gpu/drm/meson/meson_registers.h
@@ -216,6 +216,29 @@
 #define VIU_OSD2_FIFO_CTRL_STAT 0x1a4b
 #define VIU_OSD2_TEST_RDDATA 0x1a4c
 #define VIU_OSD2_PROT_CTRL 0x1a4e
+#define VIU_OSD2_MALI_UNPACK_CTRL 0x1abd
+#define VIU_OSD2_DIMM_CTRL 0x1acf
+
+#define VIU_OSD3_CTRL_STAT 0x3d80
+#define VIU_OSD3_CTRL_STAT2 0x3d81
+#define VIU_OSD3_COLOR_ADDR 0x3d82
+#define VIU_OSD3_COLOR 0x3d83
+#define VIU_OSD3_TCOLOR_AG0 0x3d84
+#define VIU_OSD3_TCOLOR_AG1 0x3d85
+#define VIU_OSD3_TCOLOR_AG2 0x3d86
+#define VIU_OSD3_TCOLOR_AG3 0x3d87
+#define VIU_OSD3_BLK0_CFG_W0 0x3d88
+#define VIU_OSD3_BLK0_CFG_W1 0x3d8c
+#define VIU_OSD3_BLK0_CFG_W2 0x3d90
+#define VIU_OSD3_BLK0_CFG_W3 0x3d94
+#define VIU_OSD3_BLK0_CFG_W4 0x3d98
+#define VIU_OSD3_BLK1_CFG_W4 0x3d99
+#define VIU_OSD3_BLK2_CFG_W4 0x3d9a
+#define VIU_OSD3_FIFO_CTRL_STAT 0x3d9c
+#define VIU_OSD3_TEST_RDDATA 0x3d9d
+#define VIU_OSD3_PROT_CTRL 0x3d9e
+#define VIU_OSD3_MALI_UNPACK_CTRL 0x3d9f
+#define VIU_OSD3_DIMM_CTRL 0x3da0
 
 #define VD1_IF0_GEN_REG 0x1a50
 #define VD1_IF0_CANVAS0 0x1a51
@@ -287,6 +310,27 @@
 #define VIU_OSD1_MATRIX_COEF31_32 0x1a9e
 #define VIU_OSD1_MATRIX_COEF40_41 0x1a9f
 #define VD1_IF0_GEN_REG3 0x1aa7
+
+#define VIU_OSD_BLENDO_H_START_END 0x1aa9
+#define VIU_OSD_BLENDO_V_START_END 0x1aaa
+#define VIU_OSD_BLEND_GEN_CTRL0 0x1aab
+#define VIU_OSD_BLEND_GEN_CTRL1 0x1aac
+#define VIU_OSD_BLEND_DUMMY_DATA 0x1aad
+#define VIU_OSD_BLEND_CURRENT_XY 0x1aae
+
+#define VIU_OSD2_MATRIX_CTRL 0x1ab0
+#define VIU_OSD2_MATRIX_COEF00_01 0x1ab1
+#define VIU_OSD2_MATRIX_COEF02_10 0x1ab2
+#define VIU_OSD2_MATRIX_COEF11_12 0x1ab3
+#define VIU_OSD2_MATRIX_COEF20_21 0x1ab4
+#define VIU_OSD2_MATRIX_COEF22 0x1ab5
+#define VIU_OSD2_MATRIX_OFFSET0_1 0x1ab6
+#define VIU_OSD2_MATRIX_OFFSET2 0x1ab7
+#define VIU_OSD2_MATRIX_PRE_OFFSET0_1 0x1ab8
+#define VIU_OSD2_MATRIX_PRE_OFFSET2 0x1ab9
+#define VIU_OSD2_MATRIX_PROBE_COLOR 0x1aba
+#define VIU_OSD2_MATRIX_HL_COLOR 0x1abb
+#define VIU_OSD2_MATRIX_PROBE_POS 0x1abc
 #define VIU_OSD1_EOTF_CTL 0x1ad4
 #define VIU_OSD1_EOTF_COEF00_01 0x1ad5
 #define VIU_OSD1_EOTF_COEF02_10 0x1ad6
@@ -481,6 +525,82 @@
 #define VPP_OSD_SCALE_COEF 0x1dcd
 #define VPP_INT_LINE_NUM 0x1dce
 
+#define VPP_WRAP_OSD1_MATRIX_COEF00_01 0x3d60
+#define VPP_WRAP_OSD1_MATRIX_COEF02_10 0x3d61
+#define VPP_WRAP_OSD1_MATRIX_COEF11_12 0x3d62
+#define VPP_WRAP_OSD1_MATRIX_COEF20_21 0x3d63
+#define VPP_WRAP_OSD1_MATRIX_COEF22 0x3d64
+#define VPP_WRAP_OSD1_MATRIX_COEF13_14 0x3d65
+#define VPP_WRAP_OSD1_MATRIX_COEF23_24 0x3d66
+#define VPP_WRAP_OSD1_MATRIX_COEF15_25 0x3d67
+#define VPP_WRAP_OSD1_MATRIX_CLIP 0x3d68
+#define VPP_WRAP_OSD1_MATRIX_OFFSET0_1 0x3d69
+#define VPP_WRAP_OSD1_MATRIX_OFFSET2 0x3d6a
+#define VPP_WRAP_OSD1_MATRIX_PRE_OFFSET0_1 0x3d6b
+#define VPP_WRAP_OSD1_MATRIX_PRE_OFFSET2 0x3d6c
+#define VPP_WRAP_OSD1_MATRIX_EN_CTRL 0x3d6d
+
+#define VPP_WRAP_OSD2_MATRIX_COEF00_01 0x3d70
+#define VPP_WRAP_OSD2_MATRIX_COEF02_10 0x3d71
+#define VPP_WRAP_OSD2_MATRIX_COEF11_12 0x3d72
+#define VPP_WRAP_OSD2_MATRIX_COEF20_21 0x3d73
+#define VPP_WRAP_OSD2_MATRIX_COEF22 0x3d74
+#define VPP_WRAP_OSD2_MATRIX_COEF13_14 0x3d75
+#define VPP_WRAP_OSD2_MATRIX_COEF23_24 0x3d76
+#define VPP_WRAP_OSD2_MATRIX_COEF15_25 0x3d77
+#define VPP_WRAP_OSD2_MATRIX_CLIP 0x3d78
+#define VPP_WRAP_OSD2_MATRIX_OFFSET0_1 0x3d79
+#define VPP_WRAP_OSD2_MATRIX_OFFSET2 0x3d7a
+#define VPP_WRAP_OSD2_MATRIX_PRE_OFFSET0_1 0x3d7b
+#define VPP_WRAP_OSD2_MATRIX_PRE_OFFSET2 0x3d7c
+#define VPP_WRAP_OSD2_MATRIX_EN_CTRL 0x3d7d
+
+#define VPP_WRAP_OSD3_MATRIX_COEF00_01 0x3db0
+#define VPP_WRAP_OSD3_MATRIX_COEF02_10 0x3db1
+#define VPP_WRAP_OSD3_MATRIX_COEF11_12 0x3db2
+#define VPP_WRAP_OSD3_MATRIX_COEF20_21 0x3db3
+#define VPP_WRAP_OSD3_MATRIX_COEF22 0x3db4
+#define VPP_WRAP_OSD3_MATRIX_COEF13_14 0x3db5
+#define VPP_WRAP_OSD3_MATRIX_COEF23_24 0x3db6
+#define VPP_WRAP_OSD3_MATRIX_COEF15_25 0x3db7
+#define VPP_WRAP_OSD3_MATRIX_CLIP 0x3db8
+#define VPP_WRAP_OSD3_MATRIX_OFFSET0_1 0x3db9
+#define VPP_WRAP_OSD3_MATRIX_OFFSET2 0x3dba
+#define VPP_WRAP_OSD3_MATRIX_PRE_OFFSET0_1 0x3dbb
+#define VPP_WRAP_OSD3_MATRIX_PRE_OFFSET2 0x3dbc
+#define VPP_WRAP_OSD3_MATRIX_EN_CTRL 0x3dbd
+
+/* osd2 scaler */
+#define OSD2_VSC_PHASE_STEP 0x3d00
+#define OSD2_VSC_INI_PHASE 0x3d01
+#define OSD2_VSC_CTRL0 0x3d02
+#define OSD2_HSC_PHASE_STEP 0x3d03
+#define OSD2_HSC_INI_PHASE 0x3d04
+#define OSD2_HSC_CTRL0 0x3d05
+#define OSD2_HSC_INI_PAT_CTRL 0x3d06
+#define OSD2_SC_DUMMY_DATA 0x3d07
+#define OSD2_SC_CTRL0 0x3d08
+#define OSD2_SCI_WH_M1 0x3d09
+#define OSD2_SCO_H_START_END 0x3d0a
+#define OSD2_SCO_V_START_END 0x3d0b
+#define OSD2_SCALE_COEF_IDX 0x3d18
+#define OSD2_SCALE_COEF 0x3d19
+
+/* osd34 scaler */
+#define OSD34_SCALE_COEF_IDX 0x3d1e
+#define OSD34_SCALE_COEF 0x3d1f
+#define OSD34_VSC_PHASE_STEP 0x3d20
+#define OSD34_VSC_INI_PHASE 0x3d21
+#define OSD34_VSC_CTRL0 0x3d22
+#define OSD34_HSC_PHASE_STEP 0x3d23
+#define OSD34_HSC_INI_PHASE 0x3d24
+#define OSD34_HSC_CTRL0 0x3d25
+#define OSD34_HSC_INI_PAT_CTRL 0x3d26
+#define OSD34_SC_DUMMY_DATA 0x3d27
+#define OSD34_SC_CTRL0 0x3d28
+#define OSD34_SCI_WH_M1 0x3d29
+#define OSD34_SCO_H_START_END 0x3d2a
+#define OSD34_SCO_V_START_END 0x3d2b
 /* viu2 */
 #define VIU2_ADDR_START 0x1e00
 #define VIU2_ADDR_END 0x1eff
@@ -1400,4 +1520,131 @@
 #define OSDSR_YBIC_VCOEF0 0x3149
 #define OSDSR_CBIC_VCOEF0 0x314a
 
+/* osd afbcd on gxtvbb */
+#define OSD1_AFBCD_ENABLE 0x31a0
+#define OSD1_AFBCD_MODE 0x31a1
+#define OSD1_AFBCD_SIZE_IN 0x31a2
+#define OSD1_AFBCD_HDR_PTR 0x31a3
+#define OSD1_AFBCD_FRAME_PTR 0x31a4
+#define OSD1_AFBCD_CHROMA_PTR 0x31a5
+#define OSD1_AFBCD_CONV_CTRL 0x31a6
+#define OSD1_AFBCD_STATUS 0x31a8
+#define OSD1_AFBCD_PIXEL_HSCOPE 0x31a9
+#define OSD1_AFBCD_PIXEL_VSCOPE 0x31aa
+#define VIU_MISC_CTRL1 0x1a07
+
+/* add for gxm and 962e dv core2 */
+#define DOLBY_CORE2A_SWAP_CTRL1	0x3434
+#define DOLBY_CORE2A_SWAP_CTRL2	0x3435
+
+/* osd afbc on g12a */
+#define VPU_MAFBC_BLOCK_ID 0x3a00
+#define VPU_MAFBC_IRQ_RAW_STATUS 0x3a01
+#define VPU_MAFBC_IRQ_CLEAR 0x3a02
+#define VPU_MAFBC_IRQ_MASK 0x3a03
+#define VPU_MAFBC_IRQ_STATUS 0x3a04
+#define VPU_MAFBC_COMMAND 0x3a05
+#define VPU_MAFBC_STATUS 0x3a06
+#define VPU_MAFBC_SURFACE_CFG 0x3a07
+
+/* osd afbc on g12a */
+#define VPU_MAFBC_HEADER_BUF_ADDR_LOW_S0 0x3a10
+#define VPU_MAFBC_HEADER_BUF_ADDR_HIGH_S0 0x3a11
+#define VPU_MAFBC_FORMAT_SPECIFIER_S0 0x3a12
+#define VPU_MAFBC_BUFFER_WIDTH_S0 0x3a13
+#define VPU_MAFBC_BUFFER_HEIGHT_S0 0x3a14
+#define VPU_MAFBC_BOUNDING_BOX_X_START_S0 0x3a15
+#define VPU_MAFBC_BOUNDING_BOX_X_END_S0 0x3a16
+#define VPU_MAFBC_BOUNDING_BOX_Y_START_S0 0x3a17
+#define VPU_MAFBC_BOUNDING_BOX_Y_END_S0 0x3a18
+#define VPU_MAFBC_OUTPUT_BUF_ADDR_LOW_S0 0x3a19
+#define VPU_MAFBC_OUTPUT_BUF_ADDR_HIGH_S0 0x3a1a
+#define VPU_MAFBC_OUTPUT_BUF_STRIDE_S0 0x3a1b
+#define VPU_MAFBC_PREFETCH_CFG_S0 0x3a1c
+
+#define VPU_MAFBC_HEADER_BUF_ADDR_LOW_S1 0x3a30
+#define VPU_MAFBC_HEADER_BUF_ADDR_HIGH_S1 0x3a31
+#define VPU_MAFBC_FORMAT_SPECIFIER_S1 0x3a32
+#define VPU_MAFBC_BUFFER_WIDTH_S1 0x3a33
+#define VPU_MAFBC_BUFFER_HEIGHT_S1 0x3a34
+#define VPU_MAFBC_BOUNDING_BOX_X_START_S1 0x3a35
+#define VPU_MAFBC_BOUNDING_BOX_X_END_S1 0x3a36
+#define VPU_MAFBC_BOUNDING_BOX_Y_START_S1 0x3a37
+#define VPU_MAFBC_BOUNDING_BOX_Y_END_S1 0x3a38
+#define VPU_MAFBC_OUTPUT_BUF_ADDR_LOW_S1 0x3a39
+#define VPU_MAFBC_OUTPUT_BUF_ADDR_HIGH_S1 0x3a3a
+#define VPU_MAFBC_OUTPUT_BUF_STRIDE_S1 0x3a3b
+#define VPU_MAFBC_PREFETCH_CFG_S1 0x3a3c
+
+#define VPU_MAFBC_HEADER_BUF_ADDR_LOW_S2 0x3a50
+#define VPU_MAFBC_HEADER_BUF_ADDR_HIGH_S2 0x3a51
+#define VPU_MAFBC_FORMAT_SPECIFIER_S2 0x3a52
+#define VPU_MAFBC_BUFFER_WIDTH_S2 0x3a53
+#define VPU_MAFBC_BUFFER_HEIGHT_S2 0x3a54
+#define VPU_MAFBC_BOUNDING_BOX_X_START_S2 0x3a55
+#define VPU_MAFBC_BOUNDING_BOX_X_END_S2 0x3a56
+#define VPU_MAFBC_BOUNDING_BOX_Y_START_S2 0x3a57
+#define VPU_MAFBC_BOUNDING_BOX_Y_END_S2 0x3a58
+#define VPU_MAFBC_OUTPUT_BUF_ADDR_LOW_S2 0x3a59
+#define VPU_MAFBC_OUTPUT_BUF_ADDR_HIGH_S2 0x3a5a
+#define VPU_MAFBC_OUTPUT_BUF_STRIDE_S2 0x3a5b
+#define VPU_MAFBC_PREFETCH_CFG_S2 0x3a5c
+
+#define VPU_MAFBC_HEADER_BUF_ADDR_LOW_S3 0x3a70
+#define VPU_MAFBC_HEADER_BUF_ADDR_HIGH_S3 0x3a71
+#define VPU_MAFBC_FORMAT_SPECIFIER_S3 0x3a72
+#define VPU_MAFBC_BUFFER_WIDTH_S3 0x3a73
+#define VPU_MAFBC_BUFFER_HEIGHT_S3 0x3a74
+#define VPU_MAFBC_BOUNDING_BOX_X_START_S3 0x3a75
+#define VPU_MAFBC_BOUNDING_BOX_X_END_S3 0x3a76
+#define VPU_MAFBC_BOUNDING_BOX_Y_START_S3 0x3a77
+#define VPU_MAFBC_BOUNDING_BOX_Y_END_S3 0x3a78
+#define VPU_MAFBC_OUTPUT_BUF_ADDR_LOW_S3 0x3a79
+#define VPU_MAFBC_OUTPUT_BUF_ADDR_HIGH_S3 0x3a7a
+#define VPU_MAFBC_OUTPUT_BUF_STRIDE_S3 0x3a7b
+#define VPU_MAFBC_PREFETCH_CFG_S3 0x3a7c
+
+#define DOLBY_PATH_CTRL 0x1a0c
+#define OSD_PATH_MISC_CTRL 0x1a0e
+#define MALI_AFBCD_TOP_CTRL 0x1a0f
+
+#define VIU_OSD_BLEND_CTRL 0x39b0
+#define VIU_OSD_BLEND_CTRL1 0x39c0
+#define VIU_OSD_BLEND_DIN0_SCOPE_H 0x39b1
+#define VIU_OSD_BLEND_DIN0_SCOPE_V 0x39b2
+#define VIU_OSD_BLEND_DIN1_SCOPE_H 0x39b3
+#define VIU_OSD_BLEND_DIN1_SCOPE_V 0x39b4
+#define VIU_OSD_BLEND_DIN2_SCOPE_H 0x39b5
+#define VIU_OSD_BLEND_DIN2_SCOPE_V 0x39b6
+#define VIU_OSD_BLEND_DIN3_SCOPE_H 0x39b7
+#define VIU_OSD_BLEND_DIN3_SCOPE_V 0x39b8
+#define VIU_OSD_BLEND_DUMMY_DATA0 0x39b9
+#define VIU_OSD_BLEND_DUMMY_ALPHA 0x39ba
+#define VIU_OSD_BLEND_BLEND0_SIZE 0x39bb
+#define VIU_OSD_BLEND_BLEND1_SIZE 0x39bc
+#define VIU_OSD_BLEND_RO_CURRENT_XY 0x39bf
+
+#define VPP_OUT_H_V_SIZE 0x1da5
+
+#define VPP_VD2_HDR_IN_SIZE 0x1df0
+#define VPP_OSD1_IN_SIZE 0x1df1
+#define VPP_GCLK_CTRL2 0x1df2
+#define VD2_PPS_DUMMY_DATA 0x1df4
+#define VPP_OSD1_BLD_H_SCOPE 0x1df5
+#define VPP_OSD1_BLD_V_SCOPE 0x1df6
+#define VPP_OSD2_BLD_H_SCOPE 0x1df7
+#define VPP_OSD2_BLD_V_SCOPE 0x1df8
+#define VPP_WRBAK_CTRL 0x1df9
+#define VPP_SLEEP_CTRL 0x1dfa
+#define VD1_BLEND_SRC_CTRL 0x1dfb
+#define VD2_BLEND_SRC_CTRL 0x1dfc
+#define OSD1_BLEND_SRC_CTRL 0x1dfd
+#define OSD2_BLEND_SRC_CTRL 0x1dfe
+
+#define VPP_POST_BLEND_BLEND_DUMMY_DATA 0x3968
+#define VPP_POST_BLEND_DUMMY_ALPHA 0x3969
+#define VPP_RDARB_MODE 0x3978
+#define VPP_RDARB_REQEN_SLV 0x3979
+#define VPU_RDARB_MODE_L2C1 0x279d
+
 #endif /* __MESON_REGISTERS_H */
diff --git a/drivers/gpu/drm/meson/meson_vclk.c b/drivers/gpu/drm/meson/meson_vclk.c
index f6ba35a405f8..b39034745444 100644
--- a/drivers/gpu/drm/meson/meson_vclk.c
+++ b/drivers/gpu/drm/meson/meson_vclk.c
@@ -113,9 +113,12 @@
 #define HHI_HDMI_PLL_CNTL4	0x32C /* 0xcb offset in data sheet */
 #define HHI_HDMI_PLL_CNTL5	0x330 /* 0xcc offset in data sheet */
 #define HHI_HDMI_PLL_CNTL6	0x334 /* 0xcd offset in data sheet */
+#define HHI_HDMI_PLL_CNTL7	0x338 /* 0xce offset in data sheet */
 
 #define HDMI_PLL_RESET		BIT(28)
+#define HDMI_PLL_RESET_G12A	BIT(29)
 #define HDMI_PLL_LOCK		BIT(31)
+#define HDMI_PLL_LOCK_G12A	(3 << 30)
 
 #define FREQ_1000_1001(_freq)	DIV_ROUND_CLOSEST(_freq * 1000, 1001)
 
@@ -257,6 +260,10 @@ static void meson_venci_cvbs_clock_config(struct meson_drm *priv)
 		regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL5, 0x71486980);
 		regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL6, 0x00000e55);
 		regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL, 0x4800023d);
+
+		/* Poll for lock bit */
+		regmap_read_poll_timeout(priv->hhi, HHI_HDMI_PLL_CNTL, val,
+					 (val & HDMI_PLL_LOCK), 10, 0);
 	} else if (meson_vpu_is_compatible(priv, "amlogic,meson-gxm-vpu") ||
 		   meson_vpu_is_compatible(priv, "amlogic,meson-gxl-vpu")) {
 		regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL, 0x4000027b);
@@ -271,11 +278,26 @@ static void meson_venci_cvbs_clock_config(struct meson_drm *priv)
 					HDMI_PLL_RESET, HDMI_PLL_RESET);
 		regmap_update_bits(priv->hhi, HHI_HDMI_PLL_CNTL,
 					HDMI_PLL_RESET, 0);
-	}
 
-	/* Poll for lock bit */
-	regmap_read_poll_timeout(priv->hhi, HHI_HDMI_PLL_CNTL, val,
-				 (val & HDMI_PLL_LOCK), 10, 0);
+		/* Poll for lock bit */
+		regmap_read_poll_timeout(priv->hhi, HHI_HDMI_PLL_CNTL, val,
+					 (val & HDMI_PLL_LOCK), 10, 0);
+	} else if (meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu")) {
+		regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL, 0x1a0504f7);
+		regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL2, 0x00010000);
+		regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL3, 0x00000000);
+		regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL4, 0x6a28dc00);
+		regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL5, 0x65771290);
+		regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL6, 0x39272000);
+		regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL7, 0x56540000);
+		regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL, 0x3a0504f7);
+		regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL, 0x1a0504f7);
+
+		/* Poll for lock bit */
+		regmap_read_poll_timeout(priv->hhi, HHI_HDMI_PLL_CNTL, val,
+			((val & HDMI_PLL_LOCK_G12A) == HDMI_PLL_LOCK_G12A),
+			10, 0);
+	}
 
 	/* Disable VCLK2 */
 	regmap_update_bits(priv->hhi, HHI_VIID_CLK_CNTL, VCLK2_EN, 0);
@@ -288,8 +310,13 @@ static void meson_venci_cvbs_clock_config(struct meson_drm *priv)
 				VCLK2_DIV_MASK, (55 - 1));
 
 	/* select vid_pll for vclk2 */
-	regmap_update_bits(priv->hhi, HHI_VIID_CLK_CNTL,
-				VCLK2_SEL_MASK, (4 << VCLK2_SEL_SHIFT));
+	if (meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu"))
+		regmap_update_bits(priv->hhi, HHI_VIID_CLK_CNTL,
+					VCLK2_SEL_MASK, (0 << VCLK2_SEL_SHIFT));
+	else
+		regmap_update_bits(priv->hhi, HHI_VIID_CLK_CNTL,
+					VCLK2_SEL_MASK, (4 << VCLK2_SEL_SHIFT));
+
 	/* enable vclk2 gate */
 	regmap_update_bits(priv->hhi, HHI_VIID_CLK_CNTL, VCLK2_EN, VCLK2_EN);
 
@@ -396,8 +423,8 @@ struct meson_vclk_params {
 	},
 	[MESON_VCLK_HDMI_297000] = {
 		.pixel_freq = 297000,
-		.pll_base_freq = 2970000,
-		.pll_od1 = 1,
+		.pll_base_freq = 5940000,
+		.pll_od1 = 2,
 		.pll_od2 = 1,
 		.pll_od3 = 1,
 		.vid_pll_div = VID_PLL_DIV_5,
@@ -476,32 +503,80 @@ void meson_hdmi_pll_set_params(struct meson_drm *priv, unsigned int m,
 		/* Poll for lock bit */
 		regmap_read_poll_timeout(priv->hhi, HHI_HDMI_PLL_CNTL, val,
 				(val & HDMI_PLL_LOCK), 10, 0);
+	} else if (meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu")) {
+		regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL, 0x0b3a0400 | m);
+
+		/* Enable and reset */
+		regmap_update_bits(priv->hhi, HHI_HDMI_PLL_CNTL,
+				   0x3 << 28, 0x3 << 28);
+
+		regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL2, frac);
+		regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL3, 0x00000000);
+
+		/* G12A HDMI PLL Needs specific parameters for 5.4GHz */
+		if (m >= 0xf7) {
+			regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL4, 0xea68dc00);
+			regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL5, 0x65771290);
+			regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL6, 0x39272000);
+			regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL7, 0x55540000);
+		} else {
+			regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL4, 0x0a691c00);
+			regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL5, 0x33771290);
+			regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL6, 0x39270000);
+			regmap_write(priv->hhi, HHI_HDMI_PLL_CNTL7, 0x50540000);
+		}
+
+		do {
+			/* Reset PLL */
+			regmap_update_bits(priv->hhi, HHI_HDMI_PLL_CNTL,
+					HDMI_PLL_RESET_G12A, HDMI_PLL_RESET_G12A);
+
+			/* UN-Reset PLL */
+			regmap_update_bits(priv->hhi, HHI_HDMI_PLL_CNTL,
+					HDMI_PLL_RESET_G12A, 0);
+
+			/* Poll for lock bits */
+			if (!regmap_read_poll_timeout(priv->hhi,
+						      HHI_HDMI_PLL_CNTL, val,
+						      ((val & HDMI_PLL_LOCK_G12A)
+						        == HDMI_PLL_LOCK_G12A),
+						      10, 100))
+				break;
+		} while(1);
 	}
 
 	if (meson_vpu_is_compatible(priv, "amlogic,meson-gxbb-vpu"))
 		regmap_update_bits(priv->hhi, HHI_HDMI_PLL_CNTL2,
 				3 << 16, pll_od_to_reg(od1) << 16);
 	else if (meson_vpu_is_compatible(priv, "amlogic,meson-gxm-vpu") ||
-			meson_vpu_is_compatible(priv, "amlogic,meson-gxl-vpu"))
+		 meson_vpu_is_compatible(priv, "amlogic,meson-gxl-vpu"))
 		regmap_update_bits(priv->hhi, HHI_HDMI_PLL_CNTL3,
 				3 << 21, pll_od_to_reg(od1) << 21);
+	else if (meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu"))
+		regmap_update_bits(priv->hhi, HHI_HDMI_PLL_CNTL,
+				3 << 16, pll_od_to_reg(od1) << 16);
 
 	if (meson_vpu_is_compatible(priv, "amlogic,meson-gxbb-vpu"))
 		regmap_update_bits(priv->hhi, HHI_HDMI_PLL_CNTL2,
 				3 << 22, pll_od_to_reg(od2) << 22);
 	else if (meson_vpu_is_compatible(priv, "amlogic,meson-gxm-vpu") ||
-			meson_vpu_is_compatible(priv, "amlogic,meson-gxl-vpu"))
+		 meson_vpu_is_compatible(priv, "amlogic,meson-gxl-vpu"))
 		regmap_update_bits(priv->hhi, HHI_HDMI_PLL_CNTL3,
 				3 << 23, pll_od_to_reg(od2) << 23);
+	else if (meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu"))
+		regmap_update_bits(priv->hhi, HHI_HDMI_PLL_CNTL,
+				3 << 18, pll_od_to_reg(od2) << 18);
 
 	if (meson_vpu_is_compatible(priv, "amlogic,meson-gxbb-vpu"))
 		regmap_update_bits(priv->hhi, HHI_HDMI_PLL_CNTL2,
 				3 << 18, pll_od_to_reg(od3) << 18);
 	else if (meson_vpu_is_compatible(priv, "amlogic,meson-gxm-vpu") ||
-			meson_vpu_is_compatible(priv, "amlogic,meson-gxl-vpu"))
+		 meson_vpu_is_compatible(priv, "amlogic,meson-gxl-vpu"))
 		regmap_update_bits(priv->hhi, HHI_HDMI_PLL_CNTL3,
 				3 << 19, pll_od_to_reg(od3) << 19);
-
+	else if (meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu"))
+		regmap_update_bits(priv->hhi, HHI_HDMI_PLL_CNTL,
+				3 << 20, pll_od_to_reg(od3) << 20);
 }
 
 #define XTAL_FREQ 24000
@@ -518,6 +593,7 @@ static unsigned int meson_hdmi_pll_get_m(struct meson_drm *priv,
 
 #define HDMI_FRAC_MAX_GXBB	4096
 #define HDMI_FRAC_MAX_GXL	1024
+#define HDMI_FRAC_MAX_G12A	131072
 
 static unsigned int meson_hdmi_pll_get_frac(struct meson_drm *priv,
 					    unsigned int m,
@@ -534,6 +610,9 @@ static unsigned int meson_hdmi_pll_get_frac(struct meson_drm *priv,
 		parent_freq *= 2;
 	}
 
+	if (meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu"))
+		frac_max = HDMI_FRAC_MAX_G12A;
+
 	/* We can have a perfect match !*/
 	if (pll_freq / m == parent_freq &&
 	    pll_freq % m == 0)
@@ -559,7 +638,8 @@ static bool meson_hdmi_pll_validate_params(struct meson_drm *priv,
 		if (frac >= HDMI_FRAC_MAX_GXBB)
 			return false;
 	} else if (meson_vpu_is_compatible(priv, "amlogic,meson-gxm-vpu") ||
-		   meson_vpu_is_compatible(priv, "amlogic,meson-gxl-vpu")) {
+		   meson_vpu_is_compatible(priv, "amlogic,meson-gxl-vpu") ||
+		   meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu")) {
 		/* Empiric supported min/max dividers */
 		if (m < 106 || m > 247)
 			return false;
@@ -714,6 +794,23 @@ static void meson_vclk_set(struct meson_drm *priv, unsigned int pll_base_freq,
 		}
 
 		meson_hdmi_pll_set_params(priv, m, frac, od1, od2, od3);
+	} else if (meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu")) {
+		switch (pll_base_freq) {
+		case 2970000:
+			m = 0x7b;
+			frac = vic_alternate_clock ? 0x140b4 : 0x18000;
+			break;
+		case 4320000:
+			m = vic_alternate_clock ? 0xb3 : 0xb4;
+			frac = vic_alternate_clock ? 0x1a3ee : 0;
+			break;
+		case 5940000:
+			m = 0xf7;
+			frac = vic_alternate_clock ? 0x8148 : 0x10000;
+			break;
+		}
+
+		meson_hdmi_pll_set_params(priv, m, frac, od1, od2, od3);
 	}
 
 	/* Setup vid_pll divider */
diff --git a/drivers/gpu/drm/meson/meson_venc.c b/drivers/gpu/drm/meson/meson_venc.c
index 66d73a932d19..6faca7313339 100644
--- a/drivers/gpu/drm/meson/meson_venc.c
+++ b/drivers/gpu/drm/meson/meson_venc.c
@@ -73,7 +73,9 @@
 /* HHI Registers */
 #define HHI_GCLK_MPEG2		0x148 /* 0x52 offset in data sheet */
 #define HHI_VDAC_CNTL0		0x2F4 /* 0xbd offset in data sheet */
+#define HHI_VDAC_CNTL0_G12A	0x2EC /* 0xbd offset in data sheet */
 #define HHI_VDAC_CNTL1		0x2F8 /* 0xbe offset in data sheet */
+#define HHI_VDAC_CNTL1_G12A	0x2F0 /* 0xbe offset in data sheet */
 #define HHI_HDMI_PHY_CNTL0	0x3a0 /* 0xe8 offset in data sheet */
 
 struct meson_cvbs_enci_mode meson_cvbs_enci_pal = {
@@ -1675,8 +1677,13 @@ void meson_venc_disable_vsync(struct meson_drm *priv)
 void meson_venc_init(struct meson_drm *priv)
 {
 	/* Disable CVBS VDAC */
-	regmap_write(priv->hhi, HHI_VDAC_CNTL0, 0);
-	regmap_write(priv->hhi, HHI_VDAC_CNTL1, 8);
+	if (meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu")) {
+		regmap_write(priv->hhi, HHI_VDAC_CNTL0_G12A, 0);
+		regmap_write(priv->hhi, HHI_VDAC_CNTL1_G12A, 8);
+	} else {
+		regmap_write(priv->hhi, HHI_VDAC_CNTL0, 0);
+		regmap_write(priv->hhi, HHI_VDAC_CNTL1, 8);
+	}
 
 	/* Power Down Dacs */
 	writel_relaxed(0xff, priv->io_base + _REG(VENC_VDAC_SETTING));
diff --git a/drivers/gpu/drm/meson/meson_venc_cvbs.c b/drivers/gpu/drm/meson/meson_venc_cvbs.c
index d622d817b6df..2c5341c881c4 100644
--- a/drivers/gpu/drm/meson/meson_venc_cvbs.c
+++ b/drivers/gpu/drm/meson/meson_venc_cvbs.c
@@ -37,7 +37,9 @@
 
 /* HHI VDAC Registers */
 #define HHI_VDAC_CNTL0		0x2F4 /* 0xbd offset in data sheet */
+#define HHI_VDAC_CNTL0_G12A	0x2EC /* 0xbd offset in data sheet */
 #define HHI_VDAC_CNTL1		0x2F8 /* 0xbe offset in data sheet */
+#define HHI_VDAC_CNTL1_G12A	0x2F0 /* 0xbe offset in data sheet */
 
 struct meson_venc_cvbs {
 	struct drm_encoder	encoder;
@@ -166,8 +168,13 @@ static void meson_venc_cvbs_encoder_disable(struct drm_encoder *encoder)
 	struct meson_drm *priv = meson_venc_cvbs->priv;
 
 	/* Disable CVBS VDAC */
-	regmap_write(priv->hhi, HHI_VDAC_CNTL0, 0);
-	regmap_write(priv->hhi, HHI_VDAC_CNTL1, 8);
+	if (meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu")) {
+		regmap_write(priv->hhi, HHI_VDAC_CNTL0_G12A, 0);
+		regmap_write(priv->hhi, HHI_VDAC_CNTL1_G12A, 0);
+	} else {
+		regmap_write(priv->hhi, HHI_VDAC_CNTL0, 0);
+		regmap_write(priv->hhi, HHI_VDAC_CNTL1, 8);
+	}
 }
 
 static void meson_venc_cvbs_encoder_enable(struct drm_encoder *encoder)
@@ -179,13 +186,17 @@ static void meson_venc_cvbs_encoder_enable(struct drm_encoder *encoder)
 	/* VDAC0 source is not from ATV */
 	writel_bits_relaxed(BIT(5), 0, priv->io_base + _REG(VENC_VDAC_DACSEL0));
 
-	if (meson_vpu_is_compatible(priv, "amlogic,meson-gxbb-vpu"))
+	if (meson_vpu_is_compatible(priv, "amlogic,meson-gxbb-vpu")) {
 		regmap_write(priv->hhi, HHI_VDAC_CNTL0, 1);
-	else if (meson_vpu_is_compatible(priv, "amlogic,meson-gxm-vpu") ||
-		 meson_vpu_is_compatible(priv, "amlogic,meson-gxl-vpu"))
+		regmap_write(priv->hhi, HHI_VDAC_CNTL1, 0);
+	} else if (meson_vpu_is_compatible(priv, "amlogic,meson-gxm-vpu") ||
+		 meson_vpu_is_compatible(priv, "amlogic,meson-gxl-vpu")) {
 		regmap_write(priv->hhi, HHI_VDAC_CNTL0, 0xf0001);
-
-	regmap_write(priv->hhi, HHI_VDAC_CNTL1, 0);
+		regmap_write(priv->hhi, HHI_VDAC_CNTL1, 0);
+	} else if (meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu")) {
+		regmap_write(priv->hhi, HHI_VDAC_CNTL0_G12A, 0x906001);
+		regmap_write(priv->hhi, HHI_VDAC_CNTL1_G12A, 0);
+	}
 }
 
 static void meson_venc_cvbs_encoder_mode_set(struct drm_encoder *encoder,
diff --git a/drivers/gpu/drm/meson/meson_viu.c b/drivers/gpu/drm/meson/meson_viu.c
index e46e05f50bad..b59072342cae 100644
--- a/drivers/gpu/drm/meson/meson_viu.c
+++ b/drivers/gpu/drm/meson/meson_viu.c
@@ -25,7 +25,6 @@
 #include "meson_viu.h"
 #include "meson_vpp.h"
 #include "meson_venc.h"
-#include "meson_canvas.h"
 #include "meson_registers.h"
 
 /**
@@ -91,8 +90,36 @@ static int eotf_bypass_coeff[EOTF_COEFF_SIZE] = {
 	EOTF_COEFF_RIGHTSHIFT /* right shift */
 };
 
-void meson_viu_set_osd_matrix(struct meson_drm *priv,
-			      enum viu_matrix_sel_e m_select,
+static void meson_viu_set_g12a_osd1_matrix(struct meson_drm *priv,
+					   int *m, bool csc_on)
+{
+	/* VPP WRAP OSD1 matrix */
+	writel(((m[0] & 0xfff) << 16) | (m[1] & 0xfff),
+		priv->io_base + _REG(VPP_WRAP_OSD1_MATRIX_PRE_OFFSET0_1));
+	writel(m[2] & 0xfff,
+		priv->io_base + _REG(VPP_WRAP_OSD1_MATRIX_PRE_OFFSET2));
+	writel(((m[3] & 0x1fff) << 16) | (m[4] & 0x1fff),
+		priv->io_base + _REG(VPP_WRAP_OSD1_MATRIX_COEF00_01));
+	writel(((m[5] & 0x1fff) << 16) | (m[6] & 0x1fff),
+		priv->io_base + _REG(VPP_WRAP_OSD1_MATRIX_COEF02_10));
+	writel(((m[7] & 0x1fff) << 16) | (m[8] & 0x1fff),
+		priv->io_base + _REG(VPP_WRAP_OSD1_MATRIX_COEF11_12));
+	writel(((m[9] & 0x1fff) << 16) | (m[10] & 0x1fff),
+		priv->io_base + _REG(VPP_WRAP_OSD1_MATRIX_COEF20_21));
+	writel((m[11] & 0x1fff) << 16,
+		priv->io_base +	_REG(VPP_WRAP_OSD1_MATRIX_COEF22));
+
+	writel(((m[18] & 0xfff) << 16) | (m[19] & 0xfff),
+		priv->io_base + _REG(VPP_WRAP_OSD1_MATRIX_OFFSET0_1));
+	writel(m[20] & 0xfff,
+		priv->io_base + _REG(VPP_WRAP_OSD1_MATRIX_OFFSET2));
+
+	writel_bits_relaxed(BIT(0), csc_on ? BIT(0) : 0,
+		priv->io_base + _REG(VPP_WRAP_OSD1_MATRIX_EN_CTRL));
+}
+
+static void meson_viu_set_osd_matrix(struct meson_drm *priv,
+				     enum viu_matrix_sel_e m_select,
 			      int *m, bool csc_on)
 {
 	if (m_select == VIU_MATRIX_OSD) {
@@ -160,10 +187,10 @@ void meson_viu_set_osd_matrix(struct meson_drm *priv,
 #define OSD_EOTF_LUT_SIZE 33
 #define OSD_OETF_LUT_SIZE 41
 
-void meson_viu_set_osd_lut(struct meson_drm *priv, enum viu_lut_sel_e lut_sel,
-			   unsigned int *r_map, unsigned int *g_map,
-			   unsigned int *b_map,
-			   bool csc_on)
+static void
+meson_viu_set_osd_lut(struct meson_drm *priv, enum viu_lut_sel_e lut_sel,
+		      unsigned int *r_map, unsigned int *g_map,
+		      unsigned int *b_map, bool csc_on)
 {
 	unsigned int addr_port;
 	unsigned int data_port;
@@ -337,14 +364,24 @@ void meson_viu_init(struct meson_drm *priv)
 	if (meson_vpu_is_compatible(priv, "amlogic,meson-gxm-vpu") ||
 	    meson_vpu_is_compatible(priv, "amlogic,meson-gxl-vpu"))
 		meson_viu_load_matrix(priv);
+	else if (meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu"))
+		meson_viu_set_g12a_osd1_matrix(priv, RGB709_to_YUV709l_coeff,
+					       true);
 
 	/* Initialize OSD1 fifo control register */
 	reg = BIT(0) |	/* Urgent DDR request priority */
-	      (4 << 5) | /* hold_fifo_lines */
-	      (3 << 10) | /* burst length 64 */
-	      (32 << 12) | /* fifo_depth_val: 32*8=256 */
-	      (2 << 22) | /* 4 words in 1 burst */
-	      (2 << 24);
+	      (4 << 5); /* hold_fifo_lines */
+	if (meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu"))
+		reg |= (1 << 10) | /* burst length 32 */
+		       (32 << 12) | /* fifo_depth_val: 32*8=256 */
+		       (2 << 22) | /* 4 words in 1 burst */
+		       (2 << 24) |
+		       (1 << 31);
+	else
+		reg |= (3 << 10) | /* burst length 64 */
+		       (32 << 12) | /* fifo_depth_val: 32*8=256 */
+		       (2 << 22) | /* 4 words in 1 burst */
+		       (2 << 24);
 	writel_relaxed(reg, priv->io_base + _REG(VIU_OSD1_FIFO_CTRL_STAT));
 	writel_relaxed(reg, priv->io_base + _REG(VIU_OSD2_FIFO_CTRL_STAT));
 
@@ -370,6 +407,30 @@ void meson_viu_init(struct meson_drm *priv)
 	writel_relaxed(0x00FF00C0,
 			priv->io_base + _REG(VD2_IF0_LUMA_FIFO_SIZE));
 
+	if (meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu")) {
+		writel_relaxed(4 << 29 |
+				1 << 27 |
+				1 << 26 | /* blend_din0 input to blend0 */
+				1 << 25 | /* blend1_dout to blend2 */
+				1 << 24 | /* blend1_din3 input to blend1 */
+				1 << 20 |
+				0 << 16 |
+				1,
+				priv->io_base + _REG(VIU_OSD_BLEND_CTRL));
+		writel_relaxed(3 << 8 |
+				1 << 20,
+				priv->io_base + _REG(OSD1_BLEND_SRC_CTRL));
+		writel_relaxed(1 << 20,
+				priv->io_base + _REG(OSD2_BLEND_SRC_CTRL));
+		writel_relaxed(0, priv->io_base + _REG(VD1_BLEND_SRC_CTRL));
+		writel_relaxed(0, priv->io_base + _REG(VD2_BLEND_SRC_CTRL));
+		writel_relaxed(0,
+				priv->io_base + _REG(VIU_OSD_BLEND_DUMMY_DATA0));
+		writel_relaxed(0,
+				priv->io_base + _REG(VIU_OSD_BLEND_DUMMY_ALPHA));
+		writel_bits_relaxed(0x3 << 2, 0x3 << 2,
+				priv->io_base + _REG(DOLBY_PATH_CTRL));
+	}
 
 	priv->viu.osd1_enabled = false;
 	priv->viu.osd1_commit = false;
diff --git a/drivers/gpu/drm/meson/meson_vpp.c b/drivers/gpu/drm/meson/meson_vpp.c
index f9efb431e953..8c52a3455ef4 100644
--- a/drivers/gpu/drm/meson/meson_vpp.c
+++ b/drivers/gpu/drm/meson/meson_vpp.c
@@ -112,32 +112,39 @@ void meson_vpp_init(struct meson_drm *priv)
 		writel_relaxed(0x20000, priv->io_base + _REG(VPP_DOLBY_CTRL));
 		writel_relaxed(0x1020080,
 				priv->io_base + _REG(VPP_DUMMY_DATA1));
-	}
+	} else if (meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu"))
+		writel_relaxed(0xf, priv->io_base + _REG(DOLBY_PATH_CTRL));
 
 	/* Initialize vpu fifo control registers */
-	writel_relaxed(readl_relaxed(priv->io_base + _REG(VPP_OFIFO_SIZE)) |
-			0x77f, priv->io_base + _REG(VPP_OFIFO_SIZE));
+	if (meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu"))
+		writel_relaxed(0xfff << 20 | 0x1000,
+			       priv->io_base + _REG(VPP_OFIFO_SIZE));
+	else
+		writel_relaxed(readl_relaxed(priv->io_base + _REG(VPP_OFIFO_SIZE)) |
+				0x77f, priv->io_base + _REG(VPP_OFIFO_SIZE));
 	writel_relaxed(0x08080808, priv->io_base + _REG(VPP_HOLD_LINES));
 
-	/* Turn off preblend */
-	writel_bits_relaxed(VPP_PREBLEND_ENABLE, 0,
-			    priv->io_base + _REG(VPP_MISC));
-
-	/* Turn off POSTBLEND */
-	writel_bits_relaxed(VPP_POSTBLEND_ENABLE, 0,
-			    priv->io_base + _REG(VPP_MISC));
-
-	/* Force all planes off */
-	writel_bits_relaxed(VPP_OSD1_POSTBLEND | VPP_OSD2_POSTBLEND |
-			    VPP_VD1_POSTBLEND | VPP_VD2_POSTBLEND |
-			    VPP_VD1_PREBLEND | VPP_VD2_PREBLEND, 0,
-			    priv->io_base + _REG(VPP_MISC));
-
-	/* Setup default VD settings */
-	writel_relaxed(4096,
-			priv->io_base + _REG(VPP_PREBLEND_VD1_H_START_END));
-	writel_relaxed(4096,
-			priv->io_base + _REG(VPP_BLEND_VD2_H_START_END));
+	if (!meson_vpu_is_compatible(priv, "amlogic,meson-g12a-vpu")) {
+		/* Turn off preblend */
+		writel_bits_relaxed(VPP_PREBLEND_ENABLE, 0,
+				    priv->io_base + _REG(VPP_MISC));
+
+		/* Turn off POSTBLEND */
+		writel_bits_relaxed(VPP_POSTBLEND_ENABLE, 0,
+				    priv->io_base + _REG(VPP_MISC));
+
+		/* Force all planes off */
+		writel_bits_relaxed(VPP_OSD1_POSTBLEND | VPP_OSD2_POSTBLEND |
+				    VPP_VD1_POSTBLEND | VPP_VD2_POSTBLEND |
+				    VPP_VD1_PREBLEND | VPP_VD2_PREBLEND, 0,
+				    priv->io_base + _REG(VPP_MISC));
+
+		/* Setup default VD settings */
+		writel_relaxed(4096,
+				priv->io_base + _REG(VPP_PREBLEND_VD1_H_START_END));
+		writel_relaxed(4096,
+				priv->io_base + _REG(VPP_BLEND_VD2_H_START_END));
+	}
 
 	/* Disable Scalers */
 	writel_relaxed(0, priv->io_base + _REG(VPP_OSD_SC_CTRL0));
diff --git a/drivers/gpu/drm/mgag200/mgag200_drv.h b/drivers/gpu/drm/mgag200/mgag200_drv.h
index 0aaedc554879..8c31e4422cae 100644
--- a/drivers/gpu/drm/mgag200/mgag200_drv.h
+++ b/drivers/gpu/drm/mgag200/mgag200_drv.h
@@ -113,7 +113,7 @@ struct mga_framebuffer {
 };
 
 struct mga_fbdev {
-	struct drm_fb_helper helper;
+	struct drm_fb_helper helper; /* must be first */
 	struct mga_framebuffer mfb;
 	void *sysram;
 	int size;
@@ -269,7 +269,6 @@ mgag200_dumb_mmap_offset(struct drm_file *file,
 struct mga_i2c_chan *mgag200_i2c_create(struct drm_device *dev);
 void mgag200_i2c_destroy(struct mga_i2c_chan *i2c);
 
-#define DRM_FILE_PAGE_OFFSET (0x100000000ULL >> PAGE_SHIFT)
 void mgag200_ttm_placement(struct mgag200_bo *bo, int domain);
 
 static inline int mgag200_bo_reserve(struct mgag200_bo *bo, bool no_wait)
diff --git a/drivers/gpu/drm/mgag200/mgag200_fb.c b/drivers/gpu/drm/mgag200/mgag200_fb.c
index 6893934b26c0..5b7e64cac004 100644
--- a/drivers/gpu/drm/mgag200/mgag200_fb.c
+++ b/drivers/gpu/drm/mgag200/mgag200_fb.c
@@ -195,8 +195,6 @@ static int mgag200fb_create(struct drm_fb_helper *helper,
 		goto err_alloc_fbi;
 	}
 
-	info->par = mfbdev;
-
 	ret = mgag200_framebuffer_init(dev, &mfbdev->mfb, &mode_cmd, gobj);
 	if (ret)
 		goto err_alloc_fbi;
@@ -209,17 +207,13 @@ static int mgag200fb_create(struct drm_fb_helper *helper,
 	/* setup helper */
 	mfbdev->helper.fb = fb;
 
-	strcpy(info->fix.id, "mgadrmfb");
-
 	info->fbops = &mgag200fb_ops;
 
 	/* setup aperture base/size for vesafb takeover */
 	info->apertures->ranges[0].base = mdev->dev->mode_config.fb_base;
 	info->apertures->ranges[0].size = mdev->mc.vram_size;
 
-	drm_fb_helper_fill_fix(info, fb->pitches[0], fb->format->depth);
-	drm_fb_helper_fill_var(info, &mfbdev->helper, sizes->fb_width,
-			       sizes->fb_height);
+	drm_fb_helper_fill_info(info, &mfbdev->helper, sizes);
 
 	info->screen_base = sysram;
 	info->screen_size = size;
diff --git a/drivers/gpu/drm/mgag200/mgag200_ttm.c b/drivers/gpu/drm/mgag200/mgag200_ttm.c
index d96a9b32455e..bd42365a8aa8 100644
--- a/drivers/gpu/drm/mgag200/mgag200_ttm.c
+++ b/drivers/gpu/drm/mgag200/mgag200_ttm.c
@@ -178,7 +178,6 @@ int mgag200_mm_init(struct mga_device *mdev)
 	ret = ttm_bo_device_init(&mdev->ttm.bdev,
 				 &mgag200_bo_driver,
 				 dev->anon_inode->i_mapping,
-				 DRM_FILE_PAGE_OFFSET,
 				 true);
 	if (ret) {
 		DRM_ERROR("Error initialising bo driver; %d\n", ret);
@@ -345,13 +344,8 @@ int mgag200_bo_push_sysram(struct mgag200_bo *bo)
 
 int mgag200_mmap(struct file *filp, struct vm_area_struct *vma)
 {
-	struct drm_file *file_priv;
-	struct mga_device *mdev;
+	struct drm_file *file_priv = filp->private_data;
+	struct mga_device *mdev = file_priv->minor->dev->dev_private;
 
-	if (unlikely(vma->vm_pgoff < DRM_FILE_PAGE_OFFSET))
-		return -EINVAL;
-
-	file_priv = filp->private_data;
-	mdev = file_priv->minor->dev->dev_private;
 	return ttm_bo_mmap(filp, vma, &mdev->ttm.bdev);
 }
diff --git a/drivers/gpu/drm/msm/Kconfig b/drivers/gpu/drm/msm/Kconfig
index 78c9e5a5e793..9f2029eca39f 100644
--- a/drivers/gpu/drm/msm/Kconfig
+++ b/drivers/gpu/drm/msm/Kconfig
@@ -21,6 +21,11 @@ config DRM_MSM
 	help
 	  DRM/KMS driver for MSM/snapdragon.
 
+config DRM_MSM_GPU_STATE
+	bool
+	depends on DRM_MSM && (DEBUG_FS || DEV_COREDUMP)
+	default y
+
 config DRM_MSM_REGISTER_LOGGING
 	bool "MSM DRM register logging"
 	depends on DRM_MSM
diff --git a/drivers/gpu/drm/msm/Makefile b/drivers/gpu/drm/msm/Makefile
index 56a70c74af4e..7a05cbf2f820 100644
--- a/drivers/gpu/drm/msm/Makefile
+++ b/drivers/gpu/drm/msm/Makefile
@@ -1,7 +1,7 @@
 # SPDX-License-Identifier: GPL-2.0
-ccflags-y := -Idrivers/gpu/drm/msm
-ccflags-y += -Idrivers/gpu/drm/msm/disp/dpu1
-ccflags-$(CONFIG_DRM_MSM_DSI) += -Idrivers/gpu/drm/msm/dsi
+ccflags-y := -I $(srctree)/$(src)
+ccflags-y += -I $(srctree)/$(src)/disp/dpu1
+ccflags-$(CONFIG_DRM_MSM_DSI) += -I $(srctree)/$(src)/dsi
 
 msm-y := \
 	adreno/adreno_device.o \
@@ -15,7 +15,6 @@ msm-y := \
 	adreno/a6xx_gpu.o \
 	adreno/a6xx_gmu.o \
 	adreno/a6xx_hfi.o \
-	adreno/a6xx_gpu_state.o \
 	hdmi/hdmi.o \
 	hdmi/hdmi_audio.o \
 	hdmi/hdmi_bridge.o \
@@ -96,6 +95,8 @@ msm-y := \
 
 msm-$(CONFIG_DEBUG_FS) += adreno/a5xx_debugfs.o
 
+msm-$(CONFIG_DRM_MSM_GPU_STATE)	+= adreno/a6xx_gpu_state.o
+
 msm-$(CONFIG_DRM_FBDEV_EMULATION) += msm_fbdev.o
 msm-$(CONFIG_COMMON_CLK) += disp/mdp4/mdp4_lvds_pll.o
 msm-$(CONFIG_COMMON_CLK) += hdmi/hdmi_pll_8960.o
diff --git a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c
index d5f5e56422f5..e5fcefa49f19 100644
--- a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c
@@ -15,9 +15,6 @@
 #include <linux/types.h>
 #include <linux/cpumask.h>
 #include <linux/qcom_scm.h>
-#include <linux/dma-mapping.h>
-#include <linux/of_address.h>
-#include <linux/soc/qcom/mdt_loader.h>
 #include <linux/pm_opp.h>
 #include <linux/nvmem-consumer.h>
 #include <linux/slab.h>
@@ -30,94 +27,6 @@ static void a5xx_dump(struct msm_gpu *gpu);
 
 #define GPU_PAS_ID 13
 
-static int zap_shader_load_mdt(struct msm_gpu *gpu, const char *fwname)
-{
-	struct device *dev = &gpu->pdev->dev;
-	const struct firmware *fw;
-	struct device_node *np;
-	struct resource r;
-	phys_addr_t mem_phys;
-	ssize_t mem_size;
-	void *mem_region = NULL;
-	int ret;
-
-	if (!IS_ENABLED(CONFIG_ARCH_QCOM))
-		return -EINVAL;
-
-	np = of_get_child_by_name(dev->of_node, "zap-shader");
-	if (!np)
-		return -ENODEV;
-
-	np = of_parse_phandle(np, "memory-region", 0);
-	if (!np)
-		return -EINVAL;
-
-	ret = of_address_to_resource(np, 0, &r);
-	if (ret)
-		return ret;
-
-	mem_phys = r.start;
-	mem_size = resource_size(&r);
-
-	/* Request the MDT file for the firmware */
-	fw = adreno_request_fw(to_adreno_gpu(gpu), fwname);
-	if (IS_ERR(fw)) {
-		DRM_DEV_ERROR(dev, "Unable to load %s\n", fwname);
-		return PTR_ERR(fw);
-	}
-
-	/* Figure out how much memory we need */
-	mem_size = qcom_mdt_get_size(fw);
-	if (mem_size < 0) {
-		ret = mem_size;
-		goto out;
-	}
-
-	/* Allocate memory for the firmware image */
-	mem_region = memremap(mem_phys, mem_size,  MEMREMAP_WC);
-	if (!mem_region) {
-		ret = -ENOMEM;
-		goto out;
-	}
-
-	/*
-	 * Load the rest of the MDT
-	 *
-	 * Note that we could be dealing with two different paths, since
-	 * with upstream linux-firmware it would be in a qcom/ subdir..
-	 * adreno_request_fw() handles this, but qcom_mdt_load() does
-	 * not.  But since we've already gotten thru adreno_request_fw()
-	 * we know which of the two cases it is:
-	 */
-	if (to_adreno_gpu(gpu)->fwloc == FW_LOCATION_LEGACY) {
-		ret = qcom_mdt_load(dev, fw, fwname, GPU_PAS_ID,
-				mem_region, mem_phys, mem_size, NULL);
-	} else {
-		char *newname;
-
-		newname = kasprintf(GFP_KERNEL, "qcom/%s", fwname);
-
-		ret = qcom_mdt_load(dev, fw, newname, GPU_PAS_ID,
-				mem_region, mem_phys, mem_size, NULL);
-		kfree(newname);
-	}
-	if (ret)
-		goto out;
-
-	/* Send the image to the secure world */
-	ret = qcom_scm_pas_auth_and_reset(GPU_PAS_ID);
-	if (ret)
-		DRM_DEV_ERROR(dev, "Unable to authorize the image\n");
-
-out:
-	if (mem_region)
-		memunmap(mem_region);
-
-	release_firmware(fw);
-
-	return ret;
-}
-
 static void a5xx_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
@@ -563,8 +472,6 @@ static int a5xx_zap_shader_resume(struct msm_gpu *gpu)
 static int a5xx_zap_shader_init(struct msm_gpu *gpu)
 {
 	static bool loaded;
-	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
-	struct platform_device *pdev = gpu->pdev;
 	int ret;
 
 	/*
@@ -574,23 +481,9 @@ static int a5xx_zap_shader_init(struct msm_gpu *gpu)
 	if (loaded)
 		return a5xx_zap_shader_resume(gpu);
 
-	/* We need SCM to be able to load the firmware */
-	if (!qcom_scm_is_available()) {
-		DRM_DEV_ERROR(&pdev->dev, "SCM is not available\n");
-		return -EPROBE_DEFER;
-	}
-
-	/* Each GPU has a target specific zap shader firmware name to use */
-	if (!adreno_gpu->info->zapfw) {
-		DRM_DEV_ERROR(&pdev->dev,
-			"Zap shader firmware file not specified for this target\n");
-		return -ENODEV;
-	}
-
-	ret = zap_shader_load_mdt(gpu, adreno_gpu->info->zapfw);
+	ret = adreno_zap_shader_load(gpu, GPU_PAS_ID);
 
 	loaded = !ret;
-
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
index d1662a75c7ec..9155dafae2a9 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
@@ -3,12 +3,31 @@
 
 #include <linux/clk.h>
 #include <linux/interconnect.h>
+#include <linux/pm_domain.h>
 #include <linux/pm_opp.h>
 #include <soc/qcom/cmd-db.h>
 
 #include "a6xx_gpu.h"
 #include "a6xx_gmu.xml.h"
 
+static void a6xx_gmu_fault(struct a6xx_gmu *gmu)
+{
+	struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu);
+	struct adreno_gpu *adreno_gpu = &a6xx_gpu->base;
+	struct msm_gpu *gpu = &adreno_gpu->base;
+	struct drm_device *dev = gpu->dev;
+	struct msm_drm_private *priv = dev->dev_private;
+
+	/* FIXME: add a banner here */
+	gmu->hung = true;
+
+	/* Turn off the hangcheck timer while we are resetting */
+	del_timer(&gpu->hangcheck_timer);
+
+	/* Queue the GPU handler because we need to treat this as a recovery */
+	queue_work(priv->wq, &gpu->recover_work);
+}
+
 static irqreturn_t a6xx_gmu_irq(int irq, void *data)
 {
 	struct a6xx_gmu *gmu = data;
@@ -20,8 +39,7 @@ static irqreturn_t a6xx_gmu_irq(int irq, void *data)
 	if (status & A6XX_GMU_AO_HOST_INTERRUPT_STATUS_WDOG_BITE) {
 		dev_err_ratelimited(gmu->dev, "GMU watchdog expired\n");
 
-		/* Temporary until we can recover safely */
-		BUG();
+		a6xx_gmu_fault(gmu);
 	}
 
 	if (status &  A6XX_GMU_AO_HOST_INTERRUPT_STATUS_HOST_AHB_BUS_ERROR)
@@ -45,8 +63,7 @@ static irqreturn_t a6xx_hfi_irq(int irq, void *data)
 	if (status & A6XX_GMU_GMU2HOST_INTR_INFO_CM3_FAULT) {
 		dev_err_ratelimited(gmu->dev, "GMU firmware fault\n");
 
-		/* Temporary until we can recover safely */
-		BUG();
+		a6xx_gmu_fault(gmu);
 	}
 
 	return IRQ_HANDLED;
@@ -165,10 +182,8 @@ static bool a6xx_gmu_check_idle_level(struct a6xx_gmu *gmu)
 }
 
 /* Wait for the GMU to get to its most idle state */
-int a6xx_gmu_wait_for_idle(struct a6xx_gpu *a6xx_gpu)
+int a6xx_gmu_wait_for_idle(struct a6xx_gmu *gmu)
 {
-	struct a6xx_gmu *gmu = &a6xx_gpu->gmu;
-
 	return spin_until(a6xx_gmu_check_idle_level(gmu));
 }
 
@@ -567,7 +582,7 @@ static int a6xx_gmu_fw_start(struct a6xx_gmu *gmu, unsigned int state)
 		if (!rpmh_init) {
 			a6xx_gmu_rpmh_init(gmu);
 			rpmh_init = true;
-		} else if (state != GMU_RESET) {
+		} else {
 			ret = a6xx_rpmh_start(gmu);
 			if (ret)
 				return ret;
@@ -633,20 +648,6 @@ static int a6xx_gmu_fw_start(struct a6xx_gmu *gmu, unsigned int state)
 	 A6XX_GMU_AO_HOST_INTERRUPT_STATUS_HOST_AHB_BUS_ERROR | \
 	 A6XX_GMU_AO_HOST_INTERRUPT_STATUS_FENCE_ERR)
 
-static void a6xx_gmu_irq_enable(struct a6xx_gmu *gmu)
-{
-	gmu_write(gmu, REG_A6XX_GMU_AO_HOST_INTERRUPT_CLR, ~0);
-	gmu_write(gmu, REG_A6XX_GMU_GMU2HOST_INTR_CLR, ~0);
-
-	gmu_write(gmu, REG_A6XX_GMU_AO_HOST_INTERRUPT_MASK,
-		~A6XX_GMU_IRQ_MASK);
-	gmu_write(gmu, REG_A6XX_GMU_GMU2HOST_INTR_MASK,
-		~A6XX_HFI_IRQ_MASK);
-
-	enable_irq(gmu->gmu_irq);
-	enable_irq(gmu->hfi_irq);
-}
-
 static void a6xx_gmu_irq_disable(struct a6xx_gmu *gmu)
 {
 	disable_irq(gmu->gmu_irq);
@@ -656,21 +657,10 @@ static void a6xx_gmu_irq_disable(struct a6xx_gmu *gmu)
 	gmu_write(gmu, REG_A6XX_GMU_GMU2HOST_INTR_MASK, ~0);
 }
 
-int a6xx_gmu_reset(struct a6xx_gpu *a6xx_gpu)
+static void a6xx_gmu_rpmh_off(struct a6xx_gmu *gmu)
 {
-	struct a6xx_gmu *gmu = &a6xx_gpu->gmu;
-	int ret;
 	u32 val;
 
-	/* Flush all the queues */
-	a6xx_hfi_stop(gmu);
-
-	/* Stop the interrupts */
-	a6xx_gmu_irq_disable(gmu);
-
-	/* Force off SPTP in case the GMU is managing it */
-	a6xx_sptprac_disable(gmu);
-
 	/* Make sure there are no outstanding RPMh votes */
 	gmu_poll_timeout(gmu, REG_A6XX_RSCC_TCS0_DRV0_STATUS, val,
 		(val & 1), 100, 10000);
@@ -680,37 +670,22 @@ int a6xx_gmu_reset(struct a6xx_gpu *a6xx_gpu)
 		(val & 1), 100, 10000);
 	gmu_poll_timeout(gmu, REG_A6XX_RSCC_TCS3_DRV0_STATUS, val,
 		(val & 1), 100, 1000);
+}
 
-	/* Force off the GX GSDC */
-	regulator_force_disable(gmu->gx);
-
-	/* Disable the resources */
-	clk_bulk_disable_unprepare(gmu->nr_clocks, gmu->clocks);
-	pm_runtime_put_sync(gmu->dev);
-
-	/* Re-enable the resources */
-	pm_runtime_get_sync(gmu->dev);
-
-	/* Use a known rate to bring up the GMU */
-	clk_set_rate(gmu->core_clk, 200000000);
-	ret = clk_bulk_prepare_enable(gmu->nr_clocks, gmu->clocks);
-	if (ret)
-		goto out;
-
-	a6xx_gmu_irq_enable(gmu);
-
-	ret = a6xx_gmu_fw_start(gmu, GMU_RESET);
-	if (!ret)
-		ret = a6xx_hfi_start(gmu, GMU_COLD_BOOT);
+/* Force the GMU off in case it isn't responsive */
+static void a6xx_gmu_force_off(struct a6xx_gmu *gmu)
+{
+	/* Flush all the queues */
+	a6xx_hfi_stop(gmu);
 
-	/* Set the GPU back to the highest power frequency */
-	__a6xx_gmu_set_freq(gmu, gmu->nr_gpu_freqs - 1);
+	/* Stop the interrupts */
+	a6xx_gmu_irq_disable(gmu);
 
-out:
-	if (ret)
-		a6xx_gmu_clear_oob(gmu, GMU_OOB_BOOT_SLUMBER);
+	/* Force off SPTP in case the GMU is managing it */
+	a6xx_sptprac_disable(gmu);
 
-	return ret;
+	/* Make sure there are no outstanding RPMh votes */
+	a6xx_gmu_rpmh_off(gmu);
 }
 
 int a6xx_gmu_resume(struct a6xx_gpu *a6xx_gpu)
@@ -723,19 +698,26 @@ int a6xx_gmu_resume(struct a6xx_gpu *a6xx_gpu)
 	if (WARN(!gmu->mmio, "The GMU is not set up yet\n"))
 		return 0;
 
+	gmu->hung = false;
+
 	/* Turn on the resources */
 	pm_runtime_get_sync(gmu->dev);
 
 	/* Use a known rate to bring up the GMU */
 	clk_set_rate(gmu->core_clk, 200000000);
 	ret = clk_bulk_prepare_enable(gmu->nr_clocks, gmu->clocks);
-	if (ret)
-		goto out;
+	if (ret) {
+		pm_runtime_put(gmu->dev);
+		return ret;
+	}
 
 	/* Set the bus quota to a reasonable value for boot */
 	icc_set_bw(gpu->icc_path, 0, MBps_to_icc(3072));
 
-	a6xx_gmu_irq_enable(gmu);
+	/* Enable the GMU interrupt */
+	gmu_write(gmu, REG_A6XX_GMU_AO_HOST_INTERRUPT_CLR, ~0);
+	gmu_write(gmu, REG_A6XX_GMU_AO_HOST_INTERRUPT_MASK, ~A6XX_GMU_IRQ_MASK);
+	enable_irq(gmu->gmu_irq);
 
 	/* Check to see if we are doing a cold or warm boot */
 	status = gmu_read(gmu, REG_A6XX_GMU_GENERAL_7) == 1 ?
@@ -746,14 +728,35 @@ int a6xx_gmu_resume(struct a6xx_gpu *a6xx_gpu)
 		goto out;
 
 	ret = a6xx_hfi_start(gmu, status);
+	if (ret)
+		goto out;
+
+	/*
+	 * Turn on the GMU firmware fault interrupt after we know the boot
+	 * sequence is successful
+	 */
+	gmu_write(gmu, REG_A6XX_GMU_GMU2HOST_INTR_CLR, ~0);
+	gmu_write(gmu, REG_A6XX_GMU_GMU2HOST_INTR_MASK, ~A6XX_HFI_IRQ_MASK);
+	enable_irq(gmu->hfi_irq);
 
 	/* Set the GPU to the highest power frequency */
 	__a6xx_gmu_set_freq(gmu, gmu->nr_gpu_freqs - 1);
 
+	/*
+	 * "enable" the GX power domain which won't actually do anything but it
+	 * will make sure that the refcounting is correct in case we need to
+	 * bring down the GX after a GMU failure
+	 */
+	if (!IS_ERR(gmu->gxpd))
+		pm_runtime_get(gmu->gxpd);
+
 out:
-	/* Make sure to turn off the boot OOB request on error */
-	if (ret)
-		a6xx_gmu_clear_oob(gmu, GMU_OOB_BOOT_SLUMBER);
+	/* On failure, shut down the GMU to leave it in a good state */
+	if (ret) {
+		disable_irq(gmu->gmu_irq);
+		a6xx_rpmh_stop(gmu);
+		pm_runtime_put(gmu->dev);
+	}
 
 	return ret;
 }
@@ -773,11 +776,12 @@ bool a6xx_gmu_isidle(struct a6xx_gmu *gmu)
 	return true;
 }
 
-int a6xx_gmu_stop(struct a6xx_gpu *a6xx_gpu)
+/* Gracefully try to shut down the GMU and by extension the GPU */
+static void a6xx_gmu_shutdown(struct a6xx_gmu *gmu)
 {
+	struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu);
 	struct adreno_gpu *adreno_gpu = &a6xx_gpu->base;
 	struct msm_gpu *gpu = &adreno_gpu->base;
-	struct a6xx_gmu *gmu = &a6xx_gpu->gmu;
 	u32 val;
 
 	/*
@@ -787,10 +791,19 @@ int a6xx_gmu_stop(struct a6xx_gpu *a6xx_gpu)
 	val = gmu_read(gmu, REG_A6XX_GPU_GMU_CX_GMU_RPMH_POWER_STATE);
 
 	if (val != 0xf) {
-		int ret = a6xx_gmu_wait_for_idle(a6xx_gpu);
+		int ret = a6xx_gmu_wait_for_idle(gmu);
 
-		/* Temporary until we can recover safely */
-		BUG_ON(ret);
+		/* If the GMU isn't responding assume it is hung */
+		if (ret) {
+			a6xx_gmu_force_off(gmu);
+			return;
+		}
+
+		/* Clear the VBIF pipe before shutting down */
+		gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0xf);
+		spin_until((gpu_read(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL1) & 0xf)
+			== 0xf);
+		gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0);
 
 		/* tell the GMU we want to slumber */
 		a6xx_gmu_notify_slumber(gmu);
@@ -822,10 +835,37 @@ int a6xx_gmu_stop(struct a6xx_gpu *a6xx_gpu)
 
 	/* Tell RPMh to power off the GPU */
 	a6xx_rpmh_stop(gmu);
+}
+
+
+int a6xx_gmu_stop(struct a6xx_gpu *a6xx_gpu)
+{
+	struct a6xx_gmu *gmu = &a6xx_gpu->gmu;
+	struct msm_gpu *gpu = &a6xx_gpu->base.base;
+
+	if (!pm_runtime_active(gmu->dev))
+		return 0;
+
+	/*
+	 * Force the GMU off if we detected a hang, otherwise try to shut it
+	 * down gracefully
+	 */
+	if (gmu->hung)
+		a6xx_gmu_force_off(gmu);
+	else
+		a6xx_gmu_shutdown(gmu);
 
 	/* Remove the bus vote */
 	icc_set_bw(gpu->icc_path, 0, 0);
 
+	/*
+	 * Make sure the GX domain is off before turning off the GMU (CX)
+	 * domain. Usually the GMU does this but only if the shutdown sequence
+	 * was successful
+	 */
+	if (!IS_ERR(gmu->gxpd))
+		pm_runtime_put_sync(gmu->gxpd);
+
 	clk_bulk_disable_unprepare(gmu->nr_clocks, gmu->clocks);
 
 	pm_runtime_put_sync(gmu->dev);
@@ -948,25 +988,20 @@ static int a6xx_gmu_memory_probe(struct a6xx_gmu *gmu)
 }
 
 /* Return the 'arc-level' for the given frequency */
-static u32 a6xx_gmu_get_arc_level(struct device *dev, unsigned long freq)
+static unsigned int a6xx_gmu_get_arc_level(struct device *dev,
+					   unsigned long freq)
 {
 	struct dev_pm_opp *opp;
-	struct device_node *np;
-	u32 val = 0;
+	unsigned int val;
 
 	if (!freq)
 		return 0;
 
-	opp  = dev_pm_opp_find_freq_exact(dev, freq, true);
+	opp = dev_pm_opp_find_freq_exact(dev, freq, true);
 	if (IS_ERR(opp))
 		return 0;
 
-	np = dev_pm_opp_get_of_node(opp);
-
-	if (np) {
-		of_property_read_u32(np, "opp-level", &val);
-		of_node_put(np);
-	}
+	val = dev_pm_opp_get_level(opp);
 
 	dev_pm_opp_put(opp);
 
@@ -1002,7 +1037,7 @@ static int a6xx_gmu_rpmh_arc_votes_init(struct device *dev, u32 *votes,
 	/* Construct a vote for each frequency */
 	for (i = 0; i < freqs_count; i++) {
 		u8 pindex = 0, sindex = 0;
-		u32 level = a6xx_gmu_get_arc_level(dev, freqs[i]);
+		unsigned int level = a6xx_gmu_get_arc_level(dev, freqs[i]);
 
 		/* Get the primary index that matches the arc level */
 		for (j = 0; j < pri_count; j++) {
@@ -1195,9 +1230,15 @@ void a6xx_gmu_remove(struct a6xx_gpu *a6xx_gpu)
 	if (IS_ERR_OR_NULL(gmu->mmio))
 		return;
 
-	pm_runtime_disable(gmu->dev);
 	a6xx_gmu_stop(a6xx_gpu);
 
+	pm_runtime_disable(gmu->dev);
+
+	if (!IS_ERR(gmu->gxpd)) {
+		pm_runtime_disable(gmu->gxpd);
+		dev_pm_domain_detach(gmu->gxpd, false);
+	}
+
 	a6xx_gmu_irq_disable(gmu);
 	a6xx_gmu_memory_free(gmu, gmu->hfi);
 
@@ -1223,7 +1264,6 @@ int a6xx_gmu_probe(struct a6xx_gpu *a6xx_gpu, struct device_node *node)
 	gmu->idle_level = GMU_IDLE_STATE_ACTIVE;
 
 	pm_runtime_enable(gmu->dev);
-	gmu->gx = devm_regulator_get(gmu->dev, "vdd");
 
 	/* Get the list of clocks */
 	ret = a6xx_gmu_clocks_probe(gmu);
@@ -1257,6 +1297,12 @@ int a6xx_gmu_probe(struct a6xx_gpu *a6xx_gpu, struct device_node *node)
 	if (gmu->hfi_irq < 0 || gmu->gmu_irq < 0)
 		goto err;
 
+	/*
+	 * Get a link to the GX power domain to reset the GPU in case of GMU
+	 * crash
+	 */
+	gmu->gxpd = dev_pm_domain_attach_by_name(gmu->dev, "gx");
+
 	/* Get the power levels for the GMU and GPU */
 	a6xx_gmu_pwrlevels_probe(gmu);
 
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
index c721d9165d8e..bedd8e6a63aa 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
@@ -27,9 +27,6 @@ struct a6xx_gmu_bo {
 /* the GMU is coming up for the first time or back from a power collapse */
 #define GMU_COLD_BOOT 1
 
-/* The GMU is being soft reset after a fault */
-#define GMU_RESET 2
-
 /*
  * These define the level of control that the GMU has - the higher the number
  * the more things that the GMU hardware controls on its own.
@@ -52,11 +49,11 @@ struct a6xx_gmu {
 	int hfi_irq;
 	int gmu_irq;
 
-	struct regulator *gx;
-
 	struct iommu_domain *domain;
 	u64 uncached_iova_base;
 
+	struct device *gxpd;
+
 	int idle_level;
 
 	struct a6xx_gmu_bo *hfi;
@@ -78,7 +75,7 @@ struct a6xx_gmu {
 
 	struct a6xx_hfi_queue queues[2];
 
-	struct tasklet_struct hfi_tasklet;
+	bool hung;
 };
 
 static inline u32 gmu_read(struct a6xx_gmu *gmu, u32 offset)
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index fefe773c989e..e74dce474250 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -10,6 +10,8 @@
 
 #include <linux/devfreq.h>
 
+#define GPU_PAS_ID 13
+
 static inline bool _a6xx_check_idle(struct msm_gpu *gpu)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
@@ -343,6 +345,20 @@ static int a6xx_ucode_init(struct msm_gpu *gpu)
 	return 0;
 }
 
+static int a6xx_zap_shader_init(struct msm_gpu *gpu)
+{
+	static bool loaded;
+	int ret;
+
+	if (loaded)
+		return 0;
+
+	ret = adreno_zap_shader_load(gpu, GPU_PAS_ID);
+
+	loaded = !ret;
+	return ret;
+}
+
 #define A6XX_INT_MASK (A6XX_RBBM_INT_0_MASK_CP_AHB_ERROR | \
 	  A6XX_RBBM_INT_0_MASK_RBBM_ATB_ASYNCFIFO_OVERFLOW | \
 	  A6XX_RBBM_INT_0_MASK_CP_HW_ERROR | \
@@ -491,7 +507,28 @@ static int a6xx_hw_init(struct msm_gpu *gpu)
 	if (ret)
 		goto out;
 
-	gpu_write(gpu, REG_A6XX_RBBM_SECVID_TRUST_CNTL, 0x0);
+	/*
+	 * Try to load a zap shader into the secure world. If successful
+	 * we can use the CP to switch out of secure mode. If not then we
+	 * have no resource but to try to switch ourselves out manually. If we
+	 * guessed wrong then access to the RBBM_SECVID_TRUST_CNTL register will
+	 * be blocked and a permissions violation will soon follow.
+	 */
+	ret = a6xx_zap_shader_init(gpu);
+	if (!ret) {
+		OUT_PKT7(gpu->rb[0], CP_SET_SECURE_MODE, 1);
+		OUT_RING(gpu->rb[0], 0x00000000);
+
+		a6xx_flush(gpu, gpu->rb[0]);
+		if (!a6xx_idle(gpu, gpu->rb[0]))
+			return -EINVAL;
+	} else {
+		/* Print a warning so if we die, we know why */
+		dev_warn_once(gpu->dev->dev,
+			"Zap shader not enabled - using SECVID_TRUST_CNTL instead\n");
+		gpu_write(gpu, REG_A6XX_RBBM_SECVID_TRUST_CNTL, 0x0);
+		ret = 0;
+	}
 
 out:
 	/*
@@ -678,13 +715,15 @@ static int a6xx_pm_resume(struct msm_gpu *gpu)
 	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
 	int ret;
 
-	ret = a6xx_gmu_resume(a6xx_gpu);
-
 	gpu->needs_hw_init = true;
 
+	ret = a6xx_gmu_resume(a6xx_gpu);
+	if (ret)
+		return ret;
+
 	msm_gpu_resume_devfreq(gpu);
 
-	return ret;
+	return 0;
 }
 
 static int a6xx_pm_suspend(struct msm_gpu *gpu)
@@ -694,18 +733,6 @@ static int a6xx_pm_suspend(struct msm_gpu *gpu)
 
 	devfreq_suspend_device(gpu->devfreq.devfreq);
 
-	/*
-	 * Make sure the GMU is idle before continuing (because some transitions
-	 * may use VBIF
-	 */
-	a6xx_gmu_wait_for_idle(a6xx_gpu);
-
-	/* Clear the VBIF pipe before shutting down */
-	/* FIXME: This accesses the GPU - do we need to make sure it is on? */
-	gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0xf);
-	spin_until((gpu_read(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL1) & 0xf) == 0xf);
-	gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0);
-
 	return a6xx_gmu_stop(a6xx_gpu);
 }
 
@@ -781,14 +808,16 @@ static const struct adreno_gpu_funcs funcs = {
 		.active_ring = a6xx_active_ring,
 		.irq = a6xx_irq,
 		.destroy = a6xx_destroy,
-#if defined(CONFIG_DEBUG_FS) || defined(CONFIG_DEV_COREDUMP)
+#if defined(CONFIG_DRM_MSM_GPU_STATE)
 		.show = a6xx_show,
 #endif
 		.gpu_busy = a6xx_gpu_busy,
 		.gpu_get_freq = a6xx_gmu_get_freq,
 		.gpu_set_freq = a6xx_gmu_set_freq,
+#if defined(CONFIG_DRM_MSM_GPU_STATE)
 		.gpu_state_get = a6xx_gpu_state_get,
 		.gpu_state_put = a6xx_gpu_state_put,
+#endif
 	},
 	.get_timestamp = a6xx_get_timestamp,
 };
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
index 528a4cfe07cd..b46279eb18c5 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
@@ -46,9 +46,8 @@ struct a6xx_gpu {
 int a6xx_gmu_resume(struct a6xx_gpu *gpu);
 int a6xx_gmu_stop(struct a6xx_gpu *gpu);
 
-int a6xx_gmu_wait_for_idle(struct a6xx_gpu *gpu);
+int a6xx_gmu_wait_for_idle(struct a6xx_gmu *gmu);
 
-int a6xx_gmu_reset(struct a6xx_gpu *a6xx_gpu);
 bool a6xx_gmu_isidle(struct a6xx_gmu *gmu);
 
 int a6xx_gmu_set_oob(struct a6xx_gmu *gmu, enum a6xx_gmu_oob_state state);
diff --git a/drivers/gpu/drm/msm/adreno/adreno_device.c b/drivers/gpu/drm/msm/adreno/adreno_device.c
index 714ed6505e47..b907245d3d96 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_device.c
+++ b/drivers/gpu/drm/msm/adreno/adreno_device.c
@@ -155,6 +155,7 @@ static const struct adreno_info gpulist[] = {
 		.gmem = SZ_1M,
 		.inactive_period = DRM_MSM_INACTIVE_PERIOD,
 		.init = a6xx_gpu_init,
+		.zapfw = "a630_zap.mdt",
 	},
 };
 
@@ -229,6 +230,7 @@ struct msm_gpu *adreno_load_gpu(struct drm_device *dev)
 
 	ret = pm_runtime_get_sync(&pdev->dev);
 	if (ret < 0) {
+		pm_runtime_put_sync(&pdev->dev);
 		DRM_DEV_ERROR(dev->dev, "Couldn't power up the GPU: %d\n", ret);
 		return NULL;
 	}
diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
index 27898475cdf4..6f7f4114afcf 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
@@ -19,13 +19,148 @@
 
 #include <linux/ascii85.h>
 #include <linux/interconnect.h>
+#include <linux/qcom_scm.h>
 #include <linux/kernel.h>
+#include <linux/of_address.h>
 #include <linux/pm_opp.h>
 #include <linux/slab.h>
+#include <linux/soc/qcom/mdt_loader.h>
 #include "adreno_gpu.h"
 #include "msm_gem.h"
 #include "msm_mmu.h"
 
+static bool zap_available = true;
+
+static int zap_shader_load_mdt(struct msm_gpu *gpu, const char *fwname,
+		u32 pasid)
+{
+	struct device *dev = &gpu->pdev->dev;
+	const struct firmware *fw;
+	struct device_node *np, *mem_np;
+	struct resource r;
+	phys_addr_t mem_phys;
+	ssize_t mem_size;
+	void *mem_region = NULL;
+	int ret;
+
+	if (!IS_ENABLED(CONFIG_ARCH_QCOM)) {
+		zap_available = false;
+		return -EINVAL;
+	}
+
+	np = of_get_child_by_name(dev->of_node, "zap-shader");
+	if (!np) {
+		zap_available = false;
+		return -ENODEV;
+	}
+
+	mem_np = of_parse_phandle(np, "memory-region", 0);
+	of_node_put(np);
+	if (!mem_np) {
+		zap_available = false;
+		return -EINVAL;
+	}
+
+	ret = of_address_to_resource(mem_np, 0, &r);
+	of_node_put(mem_np);
+	if (ret)
+		return ret;
+
+	mem_phys = r.start;
+	mem_size = resource_size(&r);
+
+	/* Request the MDT file for the firmware */
+	fw = adreno_request_fw(to_adreno_gpu(gpu), fwname);
+	if (IS_ERR(fw)) {
+		DRM_DEV_ERROR(dev, "Unable to load %s\n", fwname);
+		return PTR_ERR(fw);
+	}
+
+	/* Figure out how much memory we need */
+	mem_size = qcom_mdt_get_size(fw);
+	if (mem_size < 0) {
+		ret = mem_size;
+		goto out;
+	}
+
+	/* Allocate memory for the firmware image */
+	mem_region = memremap(mem_phys, mem_size,  MEMREMAP_WC);
+	if (!mem_region) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	/*
+	 * Load the rest of the MDT
+	 *
+	 * Note that we could be dealing with two different paths, since
+	 * with upstream linux-firmware it would be in a qcom/ subdir..
+	 * adreno_request_fw() handles this, but qcom_mdt_load() does
+	 * not.  But since we've already gotten through adreno_request_fw()
+	 * we know which of the two cases it is:
+	 */
+	if (to_adreno_gpu(gpu)->fwloc == FW_LOCATION_LEGACY) {
+		ret = qcom_mdt_load(dev, fw, fwname, pasid,
+				mem_region, mem_phys, mem_size, NULL);
+	} else {
+		char *newname;
+
+		newname = kasprintf(GFP_KERNEL, "qcom/%s", fwname);
+
+		ret = qcom_mdt_load(dev, fw, newname, pasid,
+				mem_region, mem_phys, mem_size, NULL);
+		kfree(newname);
+	}
+	if (ret)
+		goto out;
+
+	/* Send the image to the secure world */
+	ret = qcom_scm_pas_auth_and_reset(pasid);
+
+	/*
+	 * If the scm call returns -EOPNOTSUPP we assume that this target
+	 * doesn't need/support the zap shader so quietly fail
+	 */
+	if (ret == -EOPNOTSUPP)
+		zap_available = false;
+	else if (ret)
+		DRM_DEV_ERROR(dev, "Unable to authorize the image\n");
+
+out:
+	if (mem_region)
+		memunmap(mem_region);
+
+	release_firmware(fw);
+
+	return ret;
+}
+
+int adreno_zap_shader_load(struct msm_gpu *gpu, u32 pasid)
+{
+	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
+	struct platform_device *pdev = gpu->pdev;
+
+	/* Short cut if we determine the zap shader isn't available/needed */
+	if (!zap_available)
+		return -ENODEV;
+
+	/* We need SCM to be able to load the firmware */
+	if (!qcom_scm_is_available()) {
+		DRM_DEV_ERROR(&pdev->dev, "SCM is not available\n");
+		return -EPROBE_DEFER;
+	}
+
+	/* Each GPU has a target specific zap shader firmware name to use */
+	if (!adreno_gpu->info->zapfw) {
+		zap_available = false;
+		DRM_DEV_ERROR(&pdev->dev,
+			"Zap shader firmware file not specified for this target\n");
+		return -ENODEV;
+	}
+
+	return zap_shader_load_mdt(gpu, adreno_gpu->info->zapfw, pasid);
+}
+
 int adreno_get_param(struct msm_gpu *gpu, uint32_t param, uint64_t *value)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
@@ -63,6 +198,12 @@ int adreno_get_param(struct msm_gpu *gpu, uint32_t param, uint64_t *value)
 	case MSM_PARAM_NR_RINGS:
 		*value = gpu->nr_rings;
 		return 0;
+	case MSM_PARAM_PP_PGTABLE:
+		*value = 0;
+		return 0;
+	case MSM_PARAM_FAULTS:
+		*value = gpu->global_faults;
+		return 0;
 	default:
 		DBG("%s: invalid param: %u", gpu->name, param);
 		return -EINVAL;
diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
index 5db459bc28a7..0925606ec9b5 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
@@ -252,6 +252,12 @@ void adreno_gpu_state_destroy(struct msm_gpu_state *state);
 int adreno_gpu_state_get(struct msm_gpu *gpu, struct msm_gpu_state *state);
 int adreno_gpu_state_put(struct msm_gpu_state *state);
 
+/*
+ * For a5xx and a6xx targets load the zap shader that is used to pull the GPU
+ * out of secure mode
+ */
+int adreno_zap_shader_load(struct msm_gpu *gpu, u32 pasid);
+
 /* ringbuffer helpers (the parts that are adreno specific) */
 
 static inline void
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
index b776fca571f3..dfdfa766da8f 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
@@ -46,6 +46,9 @@
 #define LEFT_MIXER 0
 #define RIGHT_MIXER 1
 
+/* timeout in ms waiting for frame done */
+#define DPU_CRTC_FRAME_DONE_TIMEOUT_MS	60
+
 static struct dpu_kms *_dpu_crtc_get_kms(struct drm_crtc *crtc)
 {
 	struct msm_drm_private *priv = crtc->dev->dev_private;
@@ -425,65 +428,6 @@ void dpu_crtc_complete_commit(struct drm_crtc *crtc,
 	trace_dpu_crtc_complete_commit(DRMID(crtc));
 }
 
-static void _dpu_crtc_setup_mixer_for_encoder(
-		struct drm_crtc *crtc,
-		struct drm_encoder *enc)
-{
-	struct dpu_crtc_state *cstate = to_dpu_crtc_state(crtc->state);
-	struct dpu_kms *dpu_kms = _dpu_crtc_get_kms(crtc);
-	struct dpu_rm *rm = &dpu_kms->rm;
-	struct dpu_crtc_mixer *mixer;
-	struct dpu_hw_ctl *last_valid_ctl = NULL;
-	int i;
-	struct dpu_rm_hw_iter lm_iter, ctl_iter;
-
-	dpu_rm_init_hw_iter(&lm_iter, enc->base.id, DPU_HW_BLK_LM);
-	dpu_rm_init_hw_iter(&ctl_iter, enc->base.id, DPU_HW_BLK_CTL);
-
-	/* Set up all the mixers and ctls reserved by this encoder */
-	for (i = cstate->num_mixers; i < ARRAY_SIZE(cstate->mixers); i++) {
-		mixer = &cstate->mixers[i];
-
-		if (!dpu_rm_get_hw(rm, &lm_iter))
-			break;
-		mixer->hw_lm = (struct dpu_hw_mixer *)lm_iter.hw;
-
-		/* CTL may be <= LMs, if <, multiple LMs controlled by 1 CTL */
-		if (!dpu_rm_get_hw(rm, &ctl_iter)) {
-			DPU_DEBUG("no ctl assigned to lm %d, using previous\n",
-					mixer->hw_lm->idx - LM_0);
-			mixer->lm_ctl = last_valid_ctl;
-		} else {
-			mixer->lm_ctl = (struct dpu_hw_ctl *)ctl_iter.hw;
-			last_valid_ctl = mixer->lm_ctl;
-		}
-
-		/* Shouldn't happen, mixers are always >= ctls */
-		if (!mixer->lm_ctl) {
-			DPU_ERROR("no valid ctls found for lm %d\n",
-					mixer->hw_lm->idx - LM_0);
-			return;
-		}
-
-		cstate->num_mixers++;
-		DPU_DEBUG("setup mixer %d: lm %d\n",
-				i, mixer->hw_lm->idx - LM_0);
-		DPU_DEBUG("setup mixer %d: ctl %d\n",
-				i, mixer->lm_ctl->idx - CTL_0);
-	}
-}
-
-static void _dpu_crtc_setup_mixers(struct drm_crtc *crtc)
-{
-	struct drm_encoder *enc;
-
-	WARN_ON(!drm_modeset_is_locked(&crtc->mutex));
-
-	/* Check for mixers on all encoders attached to this crtc */
-	drm_for_each_encoder_mask(enc, crtc->dev, crtc->state->encoder_mask)
-		_dpu_crtc_setup_mixer_for_encoder(crtc, enc);
-}
-
 static void _dpu_crtc_setup_lm_bounds(struct drm_crtc *crtc,
 		struct drm_crtc_state *state)
 {
@@ -533,10 +477,7 @@ static void dpu_crtc_atomic_begin(struct drm_crtc *crtc,
 	dev = crtc->dev;
 	smmu_state = &dpu_crtc->smmu_state;
 
-	if (!cstate->num_mixers) {
-		_dpu_crtc_setup_mixers(crtc);
-		_dpu_crtc_setup_lm_bounds(crtc, crtc->state);
-	}
+	_dpu_crtc_setup_lm_bounds(crtc, crtc->state);
 
 	if (dpu_crtc->event) {
 		WARN_ON(dpu_crtc->event);
@@ -683,7 +624,7 @@ static int _dpu_crtc_wait_for_frame_done(struct drm_crtc *crtc)
 
 	DPU_ATRACE_BEGIN("frame done completion wait");
 	ret = wait_for_completion_timeout(&dpu_crtc->frame_done_comp,
-			msecs_to_jiffies(DPU_FRAME_DONE_TIMEOUT));
+			msecs_to_jiffies(DPU_CRTC_FRAME_DONE_TIMEOUT_MS));
 	if (!ret) {
 		DRM_ERROR("frame done wait timed out, ret:%d\n", ret);
 		rc = -ETIMEDOUT;
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
index 5aa3307f3f0c..82bf16d61a45 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
@@ -69,6 +69,9 @@
 
 #define MAX_VDISPLAY_SPLIT 1080
 
+/* timeout in frames waiting for frame done */
+#define DPU_ENCODER_FRAME_DONE_TIMEOUT_FRAMES 5
+
 /**
  * enum dpu_enc_rc_events - events for resource control state machine
  * @DPU_ENC_RC_EVENT_KICKOFF:
@@ -158,7 +161,7 @@ enum dpu_enc_rc_states {
  *				Bit0 = phys_encs[0] etc.
  * @crtc_frame_event_cb:	callback handler for frame event
  * @crtc_frame_event_cb_data:	callback handler private data
- * @frame_done_timeout:		frame done timeout in Hz
+ * @frame_done_timeout_ms:	frame done timeout in ms
  * @frame_done_timer:		watchdog timer for frame done event
  * @vsync_event_timer:		vsync timer
  * @disp_info:			local copy of msm_display_info struct
@@ -196,7 +199,7 @@ struct dpu_encoder_virt {
 	void (*crtc_frame_event_cb)(void *, u32 event);
 	void *crtc_frame_event_cb_data;
 
-	atomic_t frame_done_timeout;
+	atomic_t frame_done_timeout_ms;
 	struct timer_list frame_done_timer;
 	struct timer_list vsync_event_timer;
 
@@ -520,8 +523,8 @@ static void _dpu_encoder_adjust_mode(struct drm_connector *connector,
 
 	list_for_each_entry(cur_mode, &connector->modes, head) {
 		if (cur_mode->vdisplay == adj_mode->vdisplay &&
-			cur_mode->hdisplay == adj_mode->hdisplay &&
-			cur_mode->vrefresh == adj_mode->vrefresh) {
+		    cur_mode->hdisplay == adj_mode->hdisplay &&
+		    drm_mode_vrefresh(cur_mode) == drm_mode_vrefresh(adj_mode)) {
 			adj_mode->private = cur_mode->private;
 			adj_mode->private_flags |= cur_mode->private_flags;
 		}
@@ -959,10 +962,14 @@ static void dpu_encoder_virt_mode_set(struct drm_encoder *drm_enc,
 	struct dpu_kms *dpu_kms;
 	struct list_head *connector_list;
 	struct drm_connector *conn = NULL, *conn_iter;
-	struct dpu_rm_hw_iter pp_iter, ctl_iter;
+	struct drm_crtc *drm_crtc;
+	struct dpu_crtc_state *cstate;
+	struct dpu_rm_hw_iter hw_iter;
 	struct msm_display_topology topology;
 	struct dpu_hw_ctl *hw_ctl[MAX_CHANNELS_PER_ENC] = { NULL };
-	int i = 0, ret;
+	struct dpu_hw_mixer *hw_lm[MAX_CHANNELS_PER_ENC] = { NULL };
+	int num_lm = 0, num_ctl = 0;
+	int i, j, ret;
 
 	if (!drm_enc) {
 		DPU_ERROR("invalid encoder\n");
@@ -990,10 +997,14 @@ static void dpu_encoder_virt_mode_set(struct drm_encoder *drm_enc,
 		return;
 	}
 
+	drm_for_each_crtc(drm_crtc, drm_enc->dev)
+		if (drm_crtc->state->encoder_mask & drm_encoder_mask(drm_enc))
+			break;
+
 	topology = dpu_encoder_get_topology(dpu_enc, dpu_kms, adj_mode);
 
 	/* Reserve dynamic resources now. Indicating non-AtomicTest phase */
-	ret = dpu_rm_reserve(&dpu_kms->rm, drm_enc, drm_enc->crtc->state,
+	ret = dpu_rm_reserve(&dpu_kms->rm, drm_enc, drm_crtc->state,
 			     topology, false);
 	if (ret) {
 		DPU_ERROR_ENC(dpu_enc,
@@ -1001,21 +1012,41 @@ static void dpu_encoder_virt_mode_set(struct drm_encoder *drm_enc,
 		return;
 	}
 
-	dpu_rm_init_hw_iter(&pp_iter, drm_enc->base.id, DPU_HW_BLK_PINGPONG);
+	dpu_rm_init_hw_iter(&hw_iter, drm_enc->base.id, DPU_HW_BLK_PINGPONG);
 	for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) {
 		dpu_enc->hw_pp[i] = NULL;
-		if (!dpu_rm_get_hw(&dpu_kms->rm, &pp_iter))
+		if (!dpu_rm_get_hw(&dpu_kms->rm, &hw_iter))
+			break;
+		dpu_enc->hw_pp[i] = (struct dpu_hw_pingpong *) hw_iter.hw;
+	}
+
+	dpu_rm_init_hw_iter(&hw_iter, drm_enc->base.id, DPU_HW_BLK_CTL);
+	for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) {
+		if (!dpu_rm_get_hw(&dpu_kms->rm, &hw_iter))
 			break;
-		dpu_enc->hw_pp[i] = (struct dpu_hw_pingpong *) pp_iter.hw;
+		hw_ctl[i] = (struct dpu_hw_ctl *)hw_iter.hw;
+		num_ctl++;
 	}
 
-	dpu_rm_init_hw_iter(&ctl_iter, drm_enc->base.id, DPU_HW_BLK_CTL);
+	dpu_rm_init_hw_iter(&hw_iter, drm_enc->base.id, DPU_HW_BLK_LM);
 	for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) {
-		if (!dpu_rm_get_hw(&dpu_kms->rm, &ctl_iter))
+		if (!dpu_rm_get_hw(&dpu_kms->rm, &hw_iter))
 			break;
-		hw_ctl[i] = (struct dpu_hw_ctl *)ctl_iter.hw;
+		hw_lm[i] = (struct dpu_hw_mixer *)hw_iter.hw;
+		num_lm++;
 	}
 
+	cstate = to_dpu_crtc_state(drm_crtc->state);
+
+	for (i = 0; i < num_lm; i++) {
+		int ctl_idx = (i < num_ctl) ? i : (num_ctl-1);
+
+		cstate->mixers[i].hw_lm = hw_lm[i];
+		cstate->mixers[i].lm_ctl = hw_ctl[ctl_idx];
+	}
+
+	cstate->num_mixers = num_lm;
+
 	for (i = 0; i < dpu_enc->num_phys_encs; i++) {
 		struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i];
 
@@ -1023,18 +1054,38 @@ static void dpu_encoder_virt_mode_set(struct drm_encoder *drm_enc,
 			if (!dpu_enc->hw_pp[i]) {
 				DPU_ERROR_ENC(dpu_enc, "no pp block assigned"
 					     "at idx: %d\n", i);
-				return;
+				goto error;
 			}
 
 			if (!hw_ctl[i]) {
 				DPU_ERROR_ENC(dpu_enc, "no ctl block assigned"
 					     "at idx: %d\n", i);
-				return;
+				goto error;
 			}
 
 			phys->hw_pp = dpu_enc->hw_pp[i];
 			phys->hw_ctl = hw_ctl[i];
 
+			dpu_rm_init_hw_iter(&hw_iter, drm_enc->base.id,
+					    DPU_HW_BLK_INTF);
+			for (j = 0; j < MAX_CHANNELS_PER_ENC; j++) {
+				struct dpu_hw_intf *hw_intf;
+
+				if (!dpu_rm_get_hw(&dpu_kms->rm, &hw_iter))
+					break;
+
+				hw_intf = (struct dpu_hw_intf *)hw_iter.hw;
+				if (hw_intf->idx == phys->intf_idx)
+					phys->hw_intf = hw_intf;
+			}
+
+			if (!phys->hw_intf) {
+				DPU_ERROR_ENC(dpu_enc,
+					      "no intf block assigned at idx: %d\n",
+					      i);
+				goto error;
+			}
+
 			phys->connector = conn->state->connector;
 			if (phys->ops.mode_set)
 				phys->ops.mode_set(phys, mode, adj_mode);
@@ -1042,6 +1093,9 @@ static void dpu_encoder_virt_mode_set(struct drm_encoder *drm_enc,
 	}
 
 	dpu_enc->mode_set_complete = true;
+
+error:
+	dpu_rm_release(&dpu_kms->rm, drm_enc);
 }
 
 static void _dpu_encoder_virt_enable_helper(struct drm_encoder *drm_enc)
@@ -1182,7 +1236,7 @@ static void dpu_encoder_virt_disable(struct drm_encoder *drm_enc)
 	}
 
 	/* after phys waits for frame-done, should be no more frames pending */
-	if (atomic_xchg(&dpu_enc->frame_done_timeout, 0)) {
+	if (atomic_xchg(&dpu_enc->frame_done_timeout_ms, 0)) {
 		DPU_ERROR("enc%d timeout pending\n", drm_enc->base.id);
 		del_timer_sync(&dpu_enc->frame_done_timer);
 	}
@@ -1339,7 +1393,7 @@ static void dpu_encoder_frame_done_callback(
 		}
 
 		if (!dpu_enc->frame_busy_mask[0]) {
-			atomic_set(&dpu_enc->frame_done_timeout, 0);
+			atomic_set(&dpu_enc->frame_done_timeout_ms, 0);
 			del_timer(&dpu_enc->frame_done_timer);
 
 			dpu_encoder_resource_control(drm_enc,
@@ -1547,8 +1601,14 @@ static void _dpu_encoder_kickoff_phys(struct dpu_encoder_virt *dpu_enc,
 		if (!ctl)
 			continue;
 
-		if (phys->split_role != ENC_ROLE_SLAVE)
+		/*
+		 * This is cleared in frame_done worker, which isn't invoked
+		 * for async commits. So don't set this for async, since it'll
+		 * roll over to the next commit.
+		 */
+		if (!async && phys->split_role != ENC_ROLE_SLAVE)
 			set_bit(i, dpu_enc->frame_busy_mask);
+
 		if (!phys->ops.needs_single_flush ||
 				!phys->ops.needs_single_flush(phys))
 			_dpu_encoder_trigger_flush(&dpu_enc->base, phys, 0x0,
@@ -1800,11 +1860,20 @@ void dpu_encoder_kickoff(struct drm_encoder *drm_enc, bool async)
 
 	trace_dpu_enc_kickoff(DRMID(drm_enc));
 
-	atomic_set(&dpu_enc->frame_done_timeout,
-			DPU_FRAME_DONE_TIMEOUT * 1000 /
-			drm_enc->crtc->state->adjusted_mode.vrefresh);
-	mod_timer(&dpu_enc->frame_done_timer, jiffies +
-		((atomic_read(&dpu_enc->frame_done_timeout) * HZ) / 1000));
+	/*
+	 * Asynchronous frames don't handle FRAME_DONE events. As such, they
+	 * shouldn't enable the frame_done watchdog since it will always time
+	 * out.
+	 */
+	if (!async) {
+		unsigned long timeout_ms;
+		timeout_ms = DPU_ENCODER_FRAME_DONE_TIMEOUT_FRAMES * 1000 /
+			drm_mode_vrefresh(&drm_enc->crtc->state->adjusted_mode);
+
+		atomic_set(&dpu_enc->frame_done_timeout_ms, timeout_ms);
+		mod_timer(&dpu_enc->frame_done_timer,
+			  jiffies + msecs_to_jiffies(timeout_ms));
+	}
 
 	/* All phys encs are ready to go, trigger the kickoff */
 	_dpu_encoder_kickoff_phys(dpu_enc, async);
@@ -2124,7 +2193,7 @@ static void dpu_encoder_frame_done_timeout(struct timer_list *t)
 		DRM_DEBUG_KMS("id:%u invalid timeout frame_busy_mask=%lu\n",
 			      DRMID(drm_enc), dpu_enc->frame_busy_mask[0]);
 		return;
-	} else if (!atomic_xchg(&dpu_enc->frame_done_timeout, 0)) {
+	} else if (!atomic_xchg(&dpu_enc->frame_done_timeout_ms, 0)) {
 		DRM_DEBUG_KMS("id:%u invalid timeout\n", DRMID(drm_enc));
 		return;
 	}
@@ -2170,7 +2239,7 @@ int dpu_encoder_setup(struct drm_device *dev, struct drm_encoder *enc,
 
 	spin_lock_init(&dpu_enc->enc_spinlock);
 
-	atomic_set(&dpu_enc->frame_done_timeout, 0);
+	atomic_set(&dpu_enc->frame_done_timeout_ms, 0);
 	timer_setup(&dpu_enc->frame_done_timer,
 			dpu_encoder_frame_done_timeout, 0);
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h
index db94f3d3bea3..97fb868a4ef6 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h
@@ -200,6 +200,7 @@ struct dpu_encoder_irq {
  * @hw_mdptop:		Hardware interface to the top registers
  * @hw_ctl:		Hardware interface to the ctl registers
  * @hw_pp:		Hardware interface to the ping pong registers
+ * @hw_intf:		Hardware interface to the intf registers
  * @dpu_kms:		Pointer to the dpu_kms top level
  * @cached_mode:	DRM mode cached at mode_set time, acted on in enable
  * @enabled:		Whether the encoder has enabled and running a mode
@@ -228,6 +229,7 @@ struct dpu_encoder_phys {
 	struct dpu_hw_mdp *hw_mdptop;
 	struct dpu_hw_ctl *hw_ctl;
 	struct dpu_hw_pingpong *hw_pp;
+	struct dpu_hw_intf *hw_intf;
 	struct dpu_kms *dpu_kms;
 	struct drm_display_mode cached_mode;
 	enum dpu_enc_split_role split_role;
@@ -251,19 +253,6 @@ static inline int dpu_encoder_phys_inc_pending(struct dpu_encoder_phys *phys)
 }
 
 /**
- * struct dpu_encoder_phys_vid - sub-class of dpu_encoder_phys to handle video
- *	mode specific operations
- * @base:	Baseclass physical encoder structure
- * @hw_intf:	Hardware interface to the intf registers
- * @timing_params: Current timing parameter
- */
-struct dpu_encoder_phys_vid {
-	struct dpu_encoder_phys base;
-	struct dpu_hw_intf *hw_intf;
-	struct intf_timing_params timing_params;
-};
-
-/**
  * struct dpu_encoder_phys_cmd - sub-class of dpu_encoder_phys to handle command
  *	mode specific operations
  * @base:	Baseclass physical encoder structure
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
index a399e1edd313..973737fb5c9f 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
@@ -404,7 +404,8 @@ static void dpu_encoder_phys_cmd_tearcheck_config(
 		return;
 	}
 
-	tc_cfg.vsync_count = vsync_hz / (mode->vtotal * mode->vrefresh);
+	tc_cfg.vsync_count = vsync_hz /
+				(mode->vtotal * drm_mode_vrefresh(mode));
 
 	/* enable external TE after kickoff to avoid premature autorefresh */
 	tc_cfg.hw_vsync_mode = 0;
@@ -424,7 +425,7 @@ static void dpu_encoder_phys_cmd_tearcheck_config(
 	DPU_DEBUG_CMDENC(cmd_enc,
 		"tc %d vsync_clk_speed_hz %u vtotal %u vrefresh %u\n",
 		phys_enc->hw_pp->idx - PINGPONG_0, vsync_hz,
-		mode->vtotal, mode->vrefresh);
+		mode->vtotal, drm_mode_vrefresh(mode));
 	DPU_DEBUG_CMDENC(cmd_enc,
 		"tc %d enable %u start_pos %u rd_ptr_irq %u\n",
 		phys_enc->hw_pp->idx - PINGPONG_0, tc_enable, tc_cfg.start_pos,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c
index 3c4eb470a82c..1b7a335a6140 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c
@@ -18,14 +18,14 @@
 #include "dpu_trace.h"
 
 #define DPU_DEBUG_VIDENC(e, fmt, ...) DPU_DEBUG("enc%d intf%d " fmt, \
-		(e) && (e)->base.parent ? \
-		(e)->base.parent->base.id : -1, \
+		(e) && (e)->parent ? \
+		(e)->parent->base.id : -1, \
 		(e) && (e)->hw_intf ? \
 		(e)->hw_intf->idx - INTF_0 : -1, ##__VA_ARGS__)
 
 #define DPU_ERROR_VIDENC(e, fmt, ...) DPU_ERROR("enc%d intf%d " fmt, \
-		(e) && (e)->base.parent ? \
-		(e)->base.parent->base.id : -1, \
+		(e) && (e)->parent ? \
+		(e)->parent->base.id : -1, \
 		(e) && (e)->hw_intf ? \
 		(e)->hw_intf->idx - INTF_0 : -1, ##__VA_ARGS__)
 
@@ -44,7 +44,7 @@ static bool dpu_encoder_phys_vid_is_master(
 }
 
 static void drm_mode_to_intf_timing_params(
-		const struct dpu_encoder_phys_vid *vid_enc,
+		const struct dpu_encoder_phys *phys_enc,
 		const struct drm_display_mode *mode,
 		struct intf_timing_params *timing)
 {
@@ -92,7 +92,7 @@ static void drm_mode_to_intf_timing_params(
 	timing->hsync_skew = mode->hskew;
 
 	/* DSI controller cannot handle active-low sync signals. */
-	if (vid_enc->hw_intf->cap->type == INTF_DSI) {
+	if (phys_enc->hw_intf->cap->type == INTF_DSI) {
 		timing->hsync_polarity = 0;
 		timing->vsync_polarity = 0;
 	}
@@ -143,11 +143,11 @@ static u32 get_vertical_total(const struct intf_timing_params *timing)
  * lines based on the chip worst case latencies.
  */
 static u32 programmable_fetch_get_num_lines(
-		struct dpu_encoder_phys_vid *vid_enc,
+		struct dpu_encoder_phys *phys_enc,
 		const struct intf_timing_params *timing)
 {
 	u32 worst_case_needed_lines =
-	    vid_enc->hw_intf->cap->prog_fetch_lines_worst_case;
+	    phys_enc->hw_intf->cap->prog_fetch_lines_worst_case;
 	u32 start_of_frame_lines =
 	    timing->v_back_porch + timing->vsync_pulse_width;
 	u32 needed_vfp_lines = worst_case_needed_lines - start_of_frame_lines;
@@ -155,26 +155,26 @@ static u32 programmable_fetch_get_num_lines(
 
 	/* Fetch must be outside active lines, otherwise undefined. */
 	if (start_of_frame_lines >= worst_case_needed_lines) {
-		DPU_DEBUG_VIDENC(vid_enc,
+		DPU_DEBUG_VIDENC(phys_enc,
 				"prog fetch is not needed, large vbp+vsw\n");
 		actual_vfp_lines = 0;
 	} else if (timing->v_front_porch < needed_vfp_lines) {
 		/* Warn fetch needed, but not enough porch in panel config */
 		pr_warn_once
 			("low vbp+vfp may lead to perf issues in some cases\n");
-		DPU_DEBUG_VIDENC(vid_enc,
+		DPU_DEBUG_VIDENC(phys_enc,
 				"less vfp than fetch req, using entire vfp\n");
 		actual_vfp_lines = timing->v_front_porch;
 	} else {
-		DPU_DEBUG_VIDENC(vid_enc, "room in vfp for needed prefetch\n");
+		DPU_DEBUG_VIDENC(phys_enc, "room in vfp for needed prefetch\n");
 		actual_vfp_lines = needed_vfp_lines;
 	}
 
-	DPU_DEBUG_VIDENC(vid_enc,
+	DPU_DEBUG_VIDENC(phys_enc,
 		"v_front_porch %u v_back_porch %u vsync_pulse_width %u\n",
 		timing->v_front_porch, timing->v_back_porch,
 		timing->vsync_pulse_width);
-	DPU_DEBUG_VIDENC(vid_enc,
+	DPU_DEBUG_VIDENC(phys_enc,
 		"wc_lines %u needed_vfp_lines %u actual_vfp_lines %u\n",
 		worst_case_needed_lines, needed_vfp_lines, actual_vfp_lines);
 
@@ -194,8 +194,6 @@ static u32 programmable_fetch_get_num_lines(
 static void programmable_fetch_config(struct dpu_encoder_phys *phys_enc,
 				      const struct intf_timing_params *timing)
 {
-	struct dpu_encoder_phys_vid *vid_enc =
-		to_dpu_encoder_phys_vid(phys_enc);
 	struct intf_prog_fetch f = { 0 };
 	u32 vfp_fetch_lines = 0;
 	u32 horiz_total = 0;
@@ -203,10 +201,10 @@ static void programmable_fetch_config(struct dpu_encoder_phys *phys_enc,
 	u32 vfp_fetch_start_vsync_counter = 0;
 	unsigned long lock_flags;
 
-	if (WARN_ON_ONCE(!vid_enc->hw_intf->ops.setup_prg_fetch))
+	if (WARN_ON_ONCE(!phys_enc->hw_intf->ops.setup_prg_fetch))
 		return;
 
-	vfp_fetch_lines = programmable_fetch_get_num_lines(vid_enc, timing);
+	vfp_fetch_lines = programmable_fetch_get_num_lines(phys_enc, timing);
 	if (vfp_fetch_lines) {
 		vert_total = get_vertical_total(timing);
 		horiz_total = get_horizontal_total(timing);
@@ -216,12 +214,12 @@ static void programmable_fetch_config(struct dpu_encoder_phys *phys_enc,
 		f.fetch_start = vfp_fetch_start_vsync_counter;
 	}
 
-	DPU_DEBUG_VIDENC(vid_enc,
+	DPU_DEBUG_VIDENC(phys_enc,
 		"vfp_fetch_lines %u vfp_fetch_start_vsync_counter %u\n",
 		vfp_fetch_lines, vfp_fetch_start_vsync_counter);
 
 	spin_lock_irqsave(phys_enc->enc_spinlock, lock_flags);
-	vid_enc->hw_intf->ops.setup_prg_fetch(vid_enc->hw_intf, &f);
+	phys_enc->hw_intf->ops.setup_prg_fetch(phys_enc->hw_intf, &f);
 	spin_unlock_irqrestore(phys_enc->enc_spinlock, lock_flags);
 }
 
@@ -231,7 +229,7 @@ static bool dpu_encoder_phys_vid_mode_fixup(
 		struct drm_display_mode *adj_mode)
 {
 	if (phys_enc)
-		DPU_DEBUG_VIDENC(to_dpu_encoder_phys_vid(phys_enc), "\n");
+		DPU_DEBUG_VIDENC(phys_enc, "\n");
 
 	/*
 	 * Modifying mode has consequences when the mode comes back to us
@@ -242,7 +240,6 @@ static bool dpu_encoder_phys_vid_mode_fixup(
 static void dpu_encoder_phys_vid_setup_timing_engine(
 		struct dpu_encoder_phys *phys_enc)
 {
-	struct dpu_encoder_phys_vid *vid_enc;
 	struct drm_display_mode mode;
 	struct intf_timing_params timing_params = { 0 };
 	const struct dpu_format *fmt = NULL;
@@ -256,13 +253,12 @@ static void dpu_encoder_phys_vid_setup_timing_engine(
 	}
 
 	mode = phys_enc->cached_mode;
-	vid_enc = to_dpu_encoder_phys_vid(phys_enc);
-	if (!vid_enc->hw_intf->ops.setup_timing_gen) {
+	if (!phys_enc->hw_intf->ops.setup_timing_gen) {
 		DPU_ERROR("timing engine setup is not supported\n");
 		return;
 	}
 
-	DPU_DEBUG_VIDENC(vid_enc, "enabling mode:\n");
+	DPU_DEBUG_VIDENC(phys_enc, "enabling mode:\n");
 	drm_mode_debug_printmodeline(&mode);
 
 	if (phys_enc->split_role != ENC_ROLE_SOLO) {
@@ -271,32 +267,30 @@ static void dpu_encoder_phys_vid_setup_timing_engine(
 		mode.hsync_start >>= 1;
 		mode.hsync_end >>= 1;
 
-		DPU_DEBUG_VIDENC(vid_enc,
+		DPU_DEBUG_VIDENC(phys_enc,
 			"split_role %d, halve horizontal %d %d %d %d\n",
 			phys_enc->split_role,
 			mode.hdisplay, mode.htotal,
 			mode.hsync_start, mode.hsync_end);
 	}
 
-	drm_mode_to_intf_timing_params(vid_enc, &mode, &timing_params);
+	drm_mode_to_intf_timing_params(phys_enc, &mode, &timing_params);
 
 	fmt = dpu_get_dpu_format(fmt_fourcc);
-	DPU_DEBUG_VIDENC(vid_enc, "fmt_fourcc 0x%X\n", fmt_fourcc);
+	DPU_DEBUG_VIDENC(phys_enc, "fmt_fourcc 0x%X\n", fmt_fourcc);
 
-	intf_cfg.intf = vid_enc->hw_intf->idx;
+	intf_cfg.intf = phys_enc->hw_intf->idx;
 	intf_cfg.intf_mode_sel = DPU_CTL_MODE_SEL_VID;
 	intf_cfg.stream_sel = 0; /* Don't care value for video mode */
 	intf_cfg.mode_3d = dpu_encoder_helper_get_3d_blend_mode(phys_enc);
 
 	spin_lock_irqsave(phys_enc->enc_spinlock, lock_flags);
-	vid_enc->hw_intf->ops.setup_timing_gen(vid_enc->hw_intf,
+	phys_enc->hw_intf->ops.setup_timing_gen(phys_enc->hw_intf,
 			&timing_params, fmt);
 	phys_enc->hw_ctl->ops.setup_intf_cfg(phys_enc->hw_ctl, &intf_cfg);
 	spin_unlock_irqrestore(phys_enc->enc_spinlock, lock_flags);
 
 	programmable_fetch_config(phys_enc, &timing_params);
-
-	vid_enc->timing_params = timing_params;
 }
 
 static void dpu_encoder_phys_vid_vblank_irq(void *arg, int irq_idx)
@@ -353,22 +347,10 @@ static void dpu_encoder_phys_vid_underrun_irq(void *arg, int irq_idx)
 			phys_enc);
 }
 
-static bool _dpu_encoder_phys_is_dual_ctl(struct dpu_encoder_phys *phys_enc)
-{
-	struct dpu_crtc_state *dpu_cstate;
-
-	if (!phys_enc)
-		return false;
-
-	dpu_cstate = to_dpu_crtc_state(phys_enc->parent->crtc->state);
-
-	return dpu_cstate->num_ctls > 1;
-}
-
 static bool dpu_encoder_phys_vid_needs_single_flush(
 		struct dpu_encoder_phys *phys_enc)
 {
-	return (phys_enc && _dpu_encoder_phys_is_dual_ctl(phys_enc));
+	return phys_enc->split_role != ENC_ROLE_SOLO;
 }
 
 static void _dpu_encoder_phys_vid_setup_irq_hw_idx(
@@ -396,19 +378,15 @@ static void dpu_encoder_phys_vid_mode_set(
 		struct drm_display_mode *mode,
 		struct drm_display_mode *adj_mode)
 {
-	struct dpu_encoder_phys_vid *vid_enc;
-
 	if (!phys_enc || !phys_enc->dpu_kms) {
 		DPU_ERROR("invalid encoder/kms\n");
 		return;
 	}
 
-	vid_enc = to_dpu_encoder_phys_vid(phys_enc);
-
 	if (adj_mode) {
 		phys_enc->cached_mode = *adj_mode;
 		drm_mode_debug_printmodeline(adj_mode);
-		DPU_DEBUG_VIDENC(vid_enc, "caching mode:\n");
+		DPU_DEBUG_VIDENC(phys_enc, "caching mode:\n");
 	}
 
 	_dpu_encoder_phys_vid_setup_irq_hw_idx(phys_enc);
@@ -419,7 +397,6 @@ static int dpu_encoder_phys_vid_control_vblank_irq(
 		bool enable)
 {
 	int ret = 0;
-	struct dpu_encoder_phys_vid *vid_enc;
 	int refcount;
 
 	if (!phys_enc) {
@@ -428,7 +405,6 @@ static int dpu_encoder_phys_vid_control_vblank_irq(
 	}
 
 	refcount = atomic_read(&phys_enc->vblank_refcount);
-	vid_enc = to_dpu_encoder_phys_vid(phys_enc);
 
 	/* Slave encoders don't report vblank */
 	if (!dpu_encoder_phys_vid_is_master(phys_enc))
@@ -453,7 +429,7 @@ end:
 	if (ret) {
 		DRM_ERROR("failed: id:%u intf:%d ret:%d enable:%d refcnt:%d\n",
 			  DRMID(phys_enc->parent),
-			  vid_enc->hw_intf->idx - INTF_0, ret, enable,
+			  phys_enc->hw_intf->idx - INTF_0, ret, enable,
 			  refcount);
 	}
 	return ret;
@@ -461,43 +437,17 @@ end:
 
 static void dpu_encoder_phys_vid_enable(struct dpu_encoder_phys *phys_enc)
 {
-	struct msm_drm_private *priv;
-	struct dpu_encoder_phys_vid *vid_enc;
-	struct dpu_rm_hw_iter iter;
 	struct dpu_hw_ctl *ctl;
 	u32 flush_mask = 0;
 
-	if (!phys_enc || !phys_enc->parent || !phys_enc->parent->dev ||
-			!phys_enc->parent->dev->dev_private) {
-		DPU_ERROR("invalid encoder/device\n");
-		return;
-	}
-	priv = phys_enc->parent->dev->dev_private;
-
-	vid_enc = to_dpu_encoder_phys_vid(phys_enc);
 	ctl = phys_enc->hw_ctl;
 
-	dpu_rm_init_hw_iter(&iter, phys_enc->parent->base.id, DPU_HW_BLK_INTF);
-	while (dpu_rm_get_hw(&phys_enc->dpu_kms->rm, &iter)) {
-		struct dpu_hw_intf *hw_intf = (struct dpu_hw_intf *)iter.hw;
-
-		if (hw_intf->idx == phys_enc->intf_idx) {
-			vid_enc->hw_intf = hw_intf;
-			break;
-		}
-	}
-
-	if (!vid_enc->hw_intf) {
-		DPU_ERROR("hw_intf not assigned\n");
-		return;
-	}
-
-	DPU_DEBUG_VIDENC(vid_enc, "\n");
+	DPU_DEBUG_VIDENC(phys_enc, "\n");
 
-	if (WARN_ON(!vid_enc->hw_intf->ops.enable_timing))
+	if (WARN_ON(!phys_enc->hw_intf->ops.enable_timing))
 		return;
 
-	dpu_encoder_helper_split_config(phys_enc, vid_enc->hw_intf->idx);
+	dpu_encoder_helper_split_config(phys_enc, phys_enc->hw_intf->idx);
 
 	dpu_encoder_phys_vid_setup_timing_engine(phys_enc);
 
@@ -510,12 +460,13 @@ static void dpu_encoder_phys_vid_enable(struct dpu_encoder_phys *phys_enc)
 		!dpu_encoder_phys_vid_is_master(phys_enc))
 		goto skip_flush;
 
-	ctl->ops.get_bitmask_intf(ctl, &flush_mask, vid_enc->hw_intf->idx);
+	ctl->ops.get_bitmask_intf(ctl, &flush_mask, phys_enc->hw_intf->idx);
 	ctl->ops.update_pending_flush(ctl, flush_mask);
 
 skip_flush:
-	DPU_DEBUG_VIDENC(vid_enc, "update pending flush ctl %d flush_mask %x\n",
-		ctl->idx - CTL_0, flush_mask);
+	DPU_DEBUG_VIDENC(phys_enc,
+			 "update pending flush ctl %d flush_mask %x\n",
+			 ctl->idx - CTL_0, flush_mask);
 
 	/* ctl_flush & timing engine enable will be triggered by framework */
 	if (phys_enc->enable_state == DPU_ENC_DISABLED)
@@ -524,16 +475,13 @@ skip_flush:
 
 static void dpu_encoder_phys_vid_destroy(struct dpu_encoder_phys *phys_enc)
 {
-	struct dpu_encoder_phys_vid *vid_enc;
-
 	if (!phys_enc) {
 		DPU_ERROR("invalid encoder\n");
 		return;
 	}
 
-	vid_enc = to_dpu_encoder_phys_vid(phys_enc);
-	DPU_DEBUG_VIDENC(vid_enc, "\n");
-	kfree(vid_enc);
+	DPU_DEBUG_VIDENC(phys_enc, "\n");
+	kfree(phys_enc);
 }
 
 static void dpu_encoder_phys_vid_get_hw_resources(
@@ -589,7 +537,6 @@ static int dpu_encoder_phys_vid_wait_for_vblank(
 static void dpu_encoder_phys_vid_prepare_for_kickoff(
 		struct dpu_encoder_phys *phys_enc)
 {
-	struct dpu_encoder_phys_vid *vid_enc;
 	struct dpu_hw_ctl *ctl;
 	int rc;
 
@@ -597,7 +544,6 @@ static void dpu_encoder_phys_vid_prepare_for_kickoff(
 		DPU_ERROR("invalid encoder/parameters\n");
 		return;
 	}
-	vid_enc = to_dpu_encoder_phys_vid(phys_enc);
 
 	ctl = phys_enc->hw_ctl;
 	if (!ctl || !ctl->ops.wait_reset_status)
@@ -609,7 +555,7 @@ static void dpu_encoder_phys_vid_prepare_for_kickoff(
 	 */
 	rc = ctl->ops.wait_reset_status(ctl);
 	if (rc) {
-		DPU_ERROR_VIDENC(vid_enc, "ctl %d reset failure: %d\n",
+		DPU_ERROR_VIDENC(phys_enc, "ctl %d reset failure: %d\n",
 				ctl->idx, rc);
 		dpu_encoder_helper_unregister_irq(phys_enc, INTR_IDX_VSYNC);
 	}
@@ -618,7 +564,6 @@ static void dpu_encoder_phys_vid_prepare_for_kickoff(
 static void dpu_encoder_phys_vid_disable(struct dpu_encoder_phys *phys_enc)
 {
 	struct msm_drm_private *priv;
-	struct dpu_encoder_phys_vid *vid_enc;
 	unsigned long lock_flags;
 	int ret;
 
@@ -629,16 +574,13 @@ static void dpu_encoder_phys_vid_disable(struct dpu_encoder_phys *phys_enc)
 	}
 	priv = phys_enc->parent->dev->dev_private;
 
-	vid_enc = to_dpu_encoder_phys_vid(phys_enc);
-	if (!vid_enc->hw_intf || !phys_enc->hw_ctl) {
+	if (!phys_enc->hw_intf || !phys_enc->hw_ctl) {
 		DPU_ERROR("invalid hw_intf %d hw_ctl %d\n",
-				vid_enc->hw_intf != 0, phys_enc->hw_ctl != 0);
+				phys_enc->hw_intf != 0, phys_enc->hw_ctl != 0);
 		return;
 	}
 
-	DPU_DEBUG_VIDENC(vid_enc, "\n");
-
-	if (WARN_ON(!vid_enc->hw_intf->ops.enable_timing))
+	if (WARN_ON(!phys_enc->hw_intf->ops.enable_timing))
 		return;
 
 	if (phys_enc->enable_state == DPU_ENC_DISABLED) {
@@ -647,7 +589,7 @@ static void dpu_encoder_phys_vid_disable(struct dpu_encoder_phys *phys_enc)
 	}
 
 	spin_lock_irqsave(phys_enc->enc_spinlock, lock_flags);
-	vid_enc->hw_intf->ops.enable_timing(vid_enc->hw_intf, 0);
+	phys_enc->hw_intf->ops.enable_timing(phys_enc->hw_intf, 0);
 	if (dpu_encoder_phys_vid_is_master(phys_enc))
 		dpu_encoder_phys_inc_pending(phys_enc);
 	spin_unlock_irqrestore(phys_enc->enc_spinlock, lock_flags);
@@ -666,7 +608,7 @@ static void dpu_encoder_phys_vid_disable(struct dpu_encoder_phys *phys_enc)
 			atomic_set(&phys_enc->pending_kickoff_cnt, 0);
 			DRM_ERROR("wait disable failed: id:%u intf:%d ret:%d\n",
 				  DRMID(phys_enc->parent),
-				  vid_enc->hw_intf->idx - INTF_0, ret);
+				  phys_enc->hw_intf->idx - INTF_0, ret);
 		}
 	}
 
@@ -677,25 +619,21 @@ static void dpu_encoder_phys_vid_handle_post_kickoff(
 		struct dpu_encoder_phys *phys_enc)
 {
 	unsigned long lock_flags;
-	struct dpu_encoder_phys_vid *vid_enc;
 
 	if (!phys_enc) {
 		DPU_ERROR("invalid encoder\n");
 		return;
 	}
 
-	vid_enc = to_dpu_encoder_phys_vid(phys_enc);
-	DPU_DEBUG_VIDENC(vid_enc, "enable_state %d\n", phys_enc->enable_state);
-
 	/*
 	 * Video mode must flush CTL before enabling timing engine
 	 * Video encoders need to turn on their interfaces now
 	 */
 	if (phys_enc->enable_state == DPU_ENC_ENABLING) {
 		trace_dpu_enc_phys_vid_post_kickoff(DRMID(phys_enc->parent),
-				    vid_enc->hw_intf->idx - INTF_0);
+				    phys_enc->hw_intf->idx - INTF_0);
 		spin_lock_irqsave(phys_enc->enc_spinlock, lock_flags);
-		vid_enc->hw_intf->ops.enable_timing(vid_enc->hw_intf, 1);
+		phys_enc->hw_intf->ops.enable_timing(phys_enc->hw_intf, 1);
 		spin_unlock_irqrestore(phys_enc->enc_spinlock, lock_flags);
 		phys_enc->enable_state = DPU_ENC_ENABLED;
 	}
@@ -704,16 +642,13 @@ static void dpu_encoder_phys_vid_handle_post_kickoff(
 static void dpu_encoder_phys_vid_irq_control(struct dpu_encoder_phys *phys_enc,
 		bool enable)
 {
-	struct dpu_encoder_phys_vid *vid_enc;
 	int ret;
 
 	if (!phys_enc)
 		return;
 
-	vid_enc = to_dpu_encoder_phys_vid(phys_enc);
-
 	trace_dpu_enc_phys_vid_irq_ctrl(DRMID(phys_enc->parent),
-			    vid_enc->hw_intf->idx - INTF_0,
+			    phys_enc->hw_intf->idx - INTF_0,
 			    enable,
 			    atomic_read(&phys_enc->vblank_refcount));
 
@@ -732,19 +667,16 @@ static void dpu_encoder_phys_vid_irq_control(struct dpu_encoder_phys *phys_enc,
 static int dpu_encoder_phys_vid_get_line_count(
 		struct dpu_encoder_phys *phys_enc)
 {
-	struct dpu_encoder_phys_vid *vid_enc;
-
 	if (!phys_enc)
 		return -EINVAL;
 
 	if (!dpu_encoder_phys_vid_is_master(phys_enc))
 		return -EINVAL;
 
-	vid_enc = to_dpu_encoder_phys_vid(phys_enc);
-	if (!vid_enc->hw_intf || !vid_enc->hw_intf->ops.get_line_count)
+	if (!phys_enc->hw_intf || !phys_enc->hw_intf->ops.get_line_count)
 		return -EINVAL;
 
-	return vid_enc->hw_intf->ops.get_line_count(vid_enc->hw_intf);
+	return phys_enc->hw_intf->ops.get_line_count(phys_enc->hw_intf);
 }
 
 static void dpu_encoder_phys_vid_init_ops(struct dpu_encoder_phys_ops *ops)
@@ -771,7 +703,6 @@ struct dpu_encoder_phys *dpu_encoder_phys_vid_init(
 		struct dpu_enc_phys_init_params *p)
 {
 	struct dpu_encoder_phys *phys_enc = NULL;
-	struct dpu_encoder_phys_vid *vid_enc = NULL;
 	struct dpu_encoder_irq *irq;
 	int i, ret = 0;
 
@@ -780,18 +711,16 @@ struct dpu_encoder_phys *dpu_encoder_phys_vid_init(
 		goto fail;
 	}
 
-	vid_enc = kzalloc(sizeof(*vid_enc), GFP_KERNEL);
-	if (!vid_enc) {
+	phys_enc = kzalloc(sizeof(*phys_enc), GFP_KERNEL);
+	if (!phys_enc) {
 		ret = -ENOMEM;
 		goto fail;
 	}
 
-	phys_enc = &vid_enc->base;
-
 	phys_enc->hw_mdptop = p->dpu_kms->hw_mdp;
 	phys_enc->intf_idx = p->intf_idx;
 
-	DPU_DEBUG_VIDENC(vid_enc, "\n");
+	DPU_DEBUG_VIDENC(phys_enc, "\n");
 
 	dpu_encoder_phys_vid_init_ops(&phys_enc->ops);
 	phys_enc->parent = p->parent;
@@ -825,13 +754,13 @@ struct dpu_encoder_phys *dpu_encoder_phys_vid_init(
 	init_waitqueue_head(&phys_enc->pending_kickoff_wq);
 	phys_enc->enable_state = DPU_ENC_DISABLED;
 
-	DPU_DEBUG_VIDENC(vid_enc, "created intf idx:%d\n", p->intf_idx);
+	DPU_DEBUG_VIDENC(phys_enc, "created intf idx:%d\n", p->intf_idx);
 
 	return phys_enc;
 
 fail:
 	DPU_ERROR("failed to create encoder\n");
-	if (vid_enc)
+	if (phys_enc)
 		dpu_encoder_phys_vid_destroy(phys_enc);
 
 	return ERR_PTR(ret);
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h
index ac75cfc267f4..31e9ef96ca5d 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h
@@ -73,9 +73,6 @@
 
 #define DPU_NAME_SIZE  12
 
-/* timeout in frames waiting for frame done */
-#define DPU_FRAME_DONE_TIMEOUT	60
-
 /*
  * struct dpu_irq_callback - IRQ callback handlers
  * @list: list to callback
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
index b01183b309b9..da1f727d7495 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
@@ -387,7 +387,7 @@ static void _dpu_plane_set_ot_limit(struct drm_plane *plane,
 	ot_params.width = drm_rect_width(&pdpu->pipe_cfg.src_rect);
 	ot_params.height = drm_rect_height(&pdpu->pipe_cfg.src_rect);
 	ot_params.is_wfd = !pdpu->is_rt_pipe;
-	ot_params.frame_rate = crtc->mode.vrefresh;
+	ot_params.frame_rate = drm_mode_vrefresh(&crtc->mode);
 	ot_params.vbif_idx = VBIF_RT;
 	ot_params.clk_ctrl = pdpu->pipe_hw->cap->clk_ctrl;
 	ot_params.rd = true;
diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_cmd_encoder.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_cmd_encoder.c
index 9bf9d6065c55..7b9edc21bc2c 100644
--- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_cmd_encoder.c
+++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_cmd_encoder.c
@@ -59,10 +59,10 @@ static int pingpong_tearcheck_setup(struct drm_encoder *encoder,
 		return -EINVAL;
 	}
 
-	total_lines_x100 = mode->vtotal * mode->vrefresh;
+	total_lines_x100 = mode->vtotal * drm_mode_vrefresh(mode);
 	if (!total_lines_x100) {
 		DRM_DEV_ERROR(dev, "%s: vtotal(%d) or vrefresh(%d) is 0\n",
-				__func__, mode->vtotal, mode->vrefresh);
+			      __func__, mode->vtotal, drm_mode_vrefresh(mode));
 		return -EINVAL;
 	}
 
diff --git a/drivers/gpu/drm/msm/msm_debugfs.c b/drivers/gpu/drm/msm/msm_debugfs.c
index fb423d309e91..67ef300559cf 100644
--- a/drivers/gpu/drm/msm/msm_debugfs.c
+++ b/drivers/gpu/drm/msm/msm_debugfs.c
@@ -75,7 +75,7 @@ static int msm_gpu_open(struct inode *inode, struct file *file)
 	struct msm_gpu_show_priv *show_priv;
 	int ret;
 
-	if (!gpu)
+	if (!gpu || !gpu->funcs->gpu_state_get)
 		return -ENODEV;
 
 	show_priv = kmalloc(sizeof(*show_priv), GFP_KERNEL);
diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c
index 0bdd93648761..31deb87abfc6 100644
--- a/drivers/gpu/drm/msm/msm_drv.c
+++ b/drivers/gpu/drm/msm/msm_drv.c
@@ -39,9 +39,10 @@
  *           MSM_GEM_INFO ioctl.
  * - 1.4.0 - softpin, MSM_RELOC_BO_DUMP, and GEM_INFO support to set/get
  *           GEM object's debug name
+ * - 1.5.0 - Add SUBMITQUERY_QUERY ioctl
  */
 #define MSM_VERSION_MAJOR	1
-#define MSM_VERSION_MINOR	4
+#define MSM_VERSION_MINOR	5
 #define MSM_VERSION_PATCHLEVEL	0
 
 static const struct drm_mode_config_funcs mode_config_funcs = {
@@ -457,6 +458,9 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv)
 
 	priv->wq = alloc_ordered_workqueue("msm", 0);
 
+	INIT_WORK(&priv->free_work, msm_gem_free_work);
+	init_llist_head(&priv->free_list);
+
 	INIT_LIST_HEAD(&priv->inactive_list);
 
 	drm_mode_config_init(ddev);
@@ -964,6 +968,11 @@ static int msm_ioctl_submitqueue_new(struct drm_device *dev, void *data,
 		args->flags, &args->id);
 }
 
+static int msm_ioctl_submitqueue_query(struct drm_device *dev, void *data,
+		struct drm_file *file)
+{
+	return msm_submitqueue_query(dev, file->driver_priv, data);
+}
 
 static int msm_ioctl_submitqueue_close(struct drm_device *dev, void *data,
 		struct drm_file *file)
@@ -984,6 +993,7 @@ static const struct drm_ioctl_desc msm_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(MSM_GEM_MADVISE,  msm_ioctl_gem_madvise,  DRM_AUTH|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(MSM_SUBMITQUEUE_NEW,   msm_ioctl_submitqueue_new,   DRM_AUTH|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(MSM_SUBMITQUEUE_CLOSE, msm_ioctl_submitqueue_close, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(MSM_SUBMITQUEUE_QUERY, msm_ioctl_submitqueue_query, DRM_AUTH|DRM_RENDER_ALLOW),
 };
 
 static const struct vm_operations_struct vm_ops = {
@@ -1019,7 +1029,7 @@ static struct drm_driver msm_driver = {
 	.irq_uninstall      = msm_irq_uninstall,
 	.enable_vblank      = msm_enable_vblank,
 	.disable_vblank     = msm_disable_vblank,
-	.gem_free_object    = msm_gem_free_object,
+	.gem_free_object_unlocked = msm_gem_free_object,
 	.gem_vm_ops         = &vm_ops,
 	.dumb_create        = msm_gem_dumb_create,
 	.dumb_map_offset    = msm_gem_dumb_map_offset,
@@ -1027,7 +1037,6 @@ static struct drm_driver msm_driver = {
 	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
 	.gem_prime_export   = drm_gem_prime_export,
 	.gem_prime_import   = drm_gem_prime_import,
-	.gem_prime_res_obj  = msm_gem_prime_res_obj,
 	.gem_prime_pin      = msm_gem_prime_pin,
 	.gem_prime_unpin    = msm_gem_prime_unpin,
 	.gem_prime_get_sg_table = msm_gem_prime_get_sg_table,
diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h
index c56dade2c1dc..eb33d2d00d77 100644
--- a/drivers/gpu/drm/msm/msm_drv.h
+++ b/drivers/gpu/drm/msm/msm_drv.h
@@ -185,6 +185,10 @@ struct msm_drm_private {
 	/* list of GEM objects: */
 	struct list_head inactive_list;
 
+	/* worker for delayed free of objects: */
+	struct work_struct free_work;
+	struct llist_head free_list;
+
 	struct workqueue_struct *wq;
 
 	unsigned int num_planes;
@@ -292,7 +296,6 @@ struct sg_table *msm_gem_prime_get_sg_table(struct drm_gem_object *obj);
 void *msm_gem_prime_vmap(struct drm_gem_object *obj);
 void msm_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr);
 int msm_gem_prime_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma);
-struct reservation_object *msm_gem_prime_res_obj(struct drm_gem_object *obj);
 struct drm_gem_object *msm_gem_prime_import_sg_table(struct drm_device *dev,
 		struct dma_buf_attachment *attach, struct sg_table *sg);
 int msm_gem_prime_pin(struct drm_gem_object *obj);
@@ -325,6 +328,7 @@ void msm_gem_kernel_put(struct drm_gem_object *bo,
 		struct msm_gem_address_space *aspace, bool locked);
 struct drm_gem_object *msm_gem_import(struct drm_device *dev,
 		struct dma_buf *dmabuf, struct sg_table *sgt);
+void msm_gem_free_work(struct work_struct *work);
 
 __printf(2, 3)
 void msm_gem_object_set_name(struct drm_gem_object *bo, const char *fmt, ...);
@@ -420,6 +424,8 @@ struct msm_gpu_submitqueue *msm_submitqueue_get(struct msm_file_private *ctx,
 		u32 id);
 int msm_submitqueue_create(struct drm_device *drm, struct msm_file_private *ctx,
 		u32 prio, u32 flags, u32 *id);
+int msm_submitqueue_query(struct drm_device *drm, struct msm_file_private *ctx,
+		struct drm_msm_submitqueue_query *args);
 int msm_submitqueue_remove(struct msm_file_private *ctx, u32 id);
 void msm_submitqueue_close(struct msm_file_private *ctx);
 
diff --git a/drivers/gpu/drm/msm/msm_fbdev.c b/drivers/gpu/drm/msm/msm_fbdev.c
index c03e860ba737..d088299babf3 100644
--- a/drivers/gpu/drm/msm/msm_fbdev.c
+++ b/drivers/gpu/drm/msm/msm_fbdev.c
@@ -122,13 +122,9 @@ static int msm_fbdev_create(struct drm_fb_helper *helper,
 	fbdev->fb = fb;
 	helper->fb = fb;
 
-	fbi->par = helper;
 	fbi->fbops = &msm_fb_ops;
 
-	strcpy(fbi->fix.id, "msm");
-
-	drm_fb_helper_fill_fix(fbi, fb->pitches[0], fb->format->depth);
-	drm_fb_helper_fill_var(fbi, helper, sizes->fb_width, sizes->fb_height);
+	drm_fb_helper_fill_info(fbi, helper, sizes);
 
 	dev->mode_config.fb_base = paddr;
 
diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c
index 18ca651ab942..31d5a744d84f 100644
--- a/drivers/gpu/drm/msm/msm_gem.c
+++ b/drivers/gpu/drm/msm/msm_gem.c
@@ -672,14 +672,13 @@ void msm_gem_vunmap(struct drm_gem_object *obj, enum msm_gem_lock subclass)
 int msm_gem_sync_object(struct drm_gem_object *obj,
 		struct msm_fence_context *fctx, bool exclusive)
 {
-	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 	struct reservation_object_list *fobj;
 	struct dma_fence *fence;
 	int i, ret;
 
-	fobj = reservation_object_get_list(msm_obj->resv);
+	fobj = reservation_object_get_list(obj->resv);
 	if (!fobj || (fobj->shared_count == 0)) {
-		fence = reservation_object_get_excl(msm_obj->resv);
+		fence = reservation_object_get_excl(obj->resv);
 		/* don't need to wait on our own fences, since ring is fifo */
 		if (fence && (fence->context != fctx->context)) {
 			ret = dma_fence_wait(fence, true);
@@ -693,7 +692,7 @@ int msm_gem_sync_object(struct drm_gem_object *obj,
 
 	for (i = 0; i < fobj->shared_count; i++) {
 		fence = rcu_dereference_protected(fobj->shared[i],
-						reservation_object_held(msm_obj->resv));
+						reservation_object_held(obj->resv));
 		if (fence->context != fctx->context) {
 			ret = dma_fence_wait(fence, true);
 			if (ret)
@@ -711,9 +710,9 @@ void msm_gem_move_to_active(struct drm_gem_object *obj,
 	WARN_ON(msm_obj->madv != MSM_MADV_WILLNEED);
 	msm_obj->gpu = gpu;
 	if (exclusive)
-		reservation_object_add_excl_fence(msm_obj->resv, fence);
+		reservation_object_add_excl_fence(obj->resv, fence);
 	else
-		reservation_object_add_shared_fence(msm_obj->resv, fence);
+		reservation_object_add_shared_fence(obj->resv, fence);
 	list_del_init(&msm_obj->mm_list);
 	list_add_tail(&msm_obj->mm_list, &gpu->active_list);
 }
@@ -733,13 +732,12 @@ void msm_gem_move_to_inactive(struct drm_gem_object *obj)
 
 int msm_gem_cpu_prep(struct drm_gem_object *obj, uint32_t op, ktime_t *timeout)
 {
-	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 	bool write = !!(op & MSM_PREP_WRITE);
 	unsigned long remain =
 		op & MSM_PREP_NOSYNC ? 0 : timeout_to_jiffies(timeout);
 	long ret;
 
-	ret = reservation_object_wait_timeout_rcu(msm_obj->resv, write,
+	ret = reservation_object_wait_timeout_rcu(obj->resv, write,
 						  true,  remain);
 	if (ret == 0)
 		return remain == 0 ? -EBUSY : -ETIMEDOUT;
@@ -771,7 +769,7 @@ static void describe_fence(struct dma_fence *fence, const char *type,
 void msm_gem_describe(struct drm_gem_object *obj, struct seq_file *m)
 {
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
-	struct reservation_object *robj = msm_obj->resv;
+	struct reservation_object *robj = obj->resv;
 	struct reservation_object_list *fobj;
 	struct dma_fence *fence;
 	struct msm_gem_vma *vma;
@@ -853,8 +851,18 @@ void msm_gem_describe_objects(struct list_head *list, struct seq_file *m)
 /* don't call directly!  Use drm_gem_object_put() and friends */
 void msm_gem_free_object(struct drm_gem_object *obj)
 {
-	struct drm_device *dev = obj->dev;
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
+	struct drm_device *dev = obj->dev;
+	struct msm_drm_private *priv = dev->dev_private;
+
+	if (llist_add(&msm_obj->freed, &priv->free_list))
+		queue_work(priv->wq, &priv->free_work);
+}
+
+static void free_object(struct msm_gem_object *msm_obj)
+{
+	struct drm_gem_object *obj = &msm_obj->base;
+	struct drm_device *dev = obj->dev;
 
 	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
 
@@ -883,15 +891,35 @@ void msm_gem_free_object(struct drm_gem_object *obj)
 		put_pages(obj);
 	}
 
-	if (msm_obj->resv == &msm_obj->_resv)
-		reservation_object_fini(msm_obj->resv);
-
 	drm_gem_object_release(obj);
 
 	mutex_unlock(&msm_obj->lock);
 	kfree(msm_obj);
 }
 
+void msm_gem_free_work(struct work_struct *work)
+{
+	struct msm_drm_private *priv =
+		container_of(work, struct msm_drm_private, free_work);
+	struct drm_device *dev = priv->dev;
+	struct llist_node *freed;
+	struct msm_gem_object *msm_obj, *next;
+
+	while ((freed = llist_del_all(&priv->free_list))) {
+
+		mutex_lock(&dev->struct_mutex);
+
+		llist_for_each_entry_safe(msm_obj, next,
+					  freed, freed)
+			free_object(msm_obj);
+
+		mutex_unlock(&dev->struct_mutex);
+
+		if (need_resched())
+			break;
+	}
+}
+
 /* convenience method to construct a GEM buffer object, and userspace handle */
 int msm_gem_new_handle(struct drm_device *dev, struct drm_file *file,
 		uint32_t size, uint32_t flags, uint32_t *handle,
@@ -945,12 +973,8 @@ static int msm_gem_new_impl(struct drm_device *dev,
 	msm_obj->flags = flags;
 	msm_obj->madv = MSM_MADV_WILLNEED;
 
-	if (resv) {
-		msm_obj->resv = resv;
-	} else {
-		msm_obj->resv = &msm_obj->_resv;
-		reservation_object_init(msm_obj->resv);
-	}
+	if (resv)
+		msm_obj->base.resv = resv;
 
 	INIT_LIST_HEAD(&msm_obj->submit_entry);
 	INIT_LIST_HEAD(&msm_obj->vmas);
@@ -1026,6 +1050,13 @@ static struct drm_gem_object *_msm_gem_new(struct drm_device *dev,
 		ret = drm_gem_object_init(dev, obj, size);
 		if (ret)
 			goto fail;
+		/*
+		 * Our buffers are kept pinned, so allocating them from the
+		 * MOVABLE zone is a really bad idea, and conflicts with CMA.
+		 * See comments above new_inode() why this is required _and_
+		 * expected if you're going to pin these pages.
+		 */
+		mapping_set_gfp_mask(obj->filp->f_mapping, GFP_HIGHUSER);
 	}
 
 	return obj;
diff --git a/drivers/gpu/drm/msm/msm_gem.h b/drivers/gpu/drm/msm/msm_gem.h
index 2064fac871b8..c5ac781dffee 100644
--- a/drivers/gpu/drm/msm/msm_gem.h
+++ b/drivers/gpu/drm/msm/msm_gem.h
@@ -84,6 +84,8 @@ struct msm_gem_object {
 
 	struct list_head vmas;    /* list of msm_gem_vma */
 
+	struct llist_node freed;
+
 	/* normally (resv == &_resv) except for imported bo's */
 	struct reservation_object *resv;
 	struct reservation_object _resv;
@@ -133,6 +135,7 @@ enum msm_gem_lock {
 
 void msm_gem_purge(struct drm_gem_object *obj, enum msm_gem_lock subclass);
 void msm_gem_vunmap(struct drm_gem_object *obj, enum msm_gem_lock subclass);
+void msm_gem_free_work(struct work_struct *work);
 
 /* Created per submit-ioctl, to track bo's and cmdstream bufs, etc,
  * associated with the cmdstream submission for synchronization (and
@@ -163,7 +166,10 @@ struct msm_gem_submit {
 	} *cmd;  /* array of size nr_cmds */
 	struct {
 		uint32_t flags;
-		struct msm_gem_object *obj;
+		union {
+			struct msm_gem_object *obj;
+			uint32_t handle;
+		};
 		uint64_t iova;
 	} bos[0];
 };
diff --git a/drivers/gpu/drm/msm/msm_gem_prime.c b/drivers/gpu/drm/msm/msm_gem_prime.c
index 13403c6da6c7..60bb290700ce 100644
--- a/drivers/gpu/drm/msm/msm_gem_prime.c
+++ b/drivers/gpu/drm/msm/msm_gem_prime.c
@@ -70,10 +70,3 @@ void msm_gem_prime_unpin(struct drm_gem_object *obj)
 	if (!obj->import_attach)
 		msm_gem_put_pages(obj);
 }
-
-struct reservation_object *msm_gem_prime_res_obj(struct drm_gem_object *obj)
-{
-	struct msm_gem_object *msm_obj = to_msm_bo(obj);
-
-	return msm_obj->resv;
-}
diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c
index 12b983fc0b56..1b681306aca3 100644
--- a/drivers/gpu/drm/msm/msm_gem_submit.c
+++ b/drivers/gpu/drm/msm/msm_gem_submit.c
@@ -74,27 +74,14 @@ void msm_gem_submit_free(struct msm_gem_submit *submit)
 	kfree(submit);
 }
 
-static inline unsigned long __must_check
-copy_from_user_inatomic(void *to, const void __user *from, unsigned long n)
-{
-	if (access_ok(from, n))
-		return __copy_from_user_inatomic(to, from, n);
-	return -EFAULT;
-}
-
 static int submit_lookup_objects(struct msm_gem_submit *submit,
 		struct drm_msm_gem_submit *args, struct drm_file *file)
 {
 	unsigned i;
 	int ret = 0;
 
-	spin_lock(&file->table_lock);
-	pagefault_disable();
-
 	for (i = 0; i < args->nr_bos; i++) {
 		struct drm_msm_gem_submit_bo submit_bo;
-		struct drm_gem_object *obj;
-		struct msm_gem_object *msm_obj;
 		void __user *userptr =
 			u64_to_user_ptr(args->bos + (i * sizeof(submit_bo)));
 
@@ -103,15 +90,10 @@ static int submit_lookup_objects(struct msm_gem_submit *submit,
 		 */
 		submit->bos[i].flags = 0;
 
-		if (copy_from_user_inatomic(&submit_bo, userptr, sizeof(submit_bo))) {
-			pagefault_enable();
-			spin_unlock(&file->table_lock);
-			if (copy_from_user(&submit_bo, userptr, sizeof(submit_bo))) {
-				ret = -EFAULT;
-				goto out;
-			}
-			spin_lock(&file->table_lock);
-			pagefault_disable();
+		if (copy_from_user(&submit_bo, userptr, sizeof(submit_bo))) {
+			ret = -EFAULT;
+			i = 0;
+			goto out;
 		}
 
 /* at least one of READ and/or WRITE flags should be set: */
@@ -121,19 +103,28 @@ static int submit_lookup_objects(struct msm_gem_submit *submit,
 			!(submit_bo.flags & MANDATORY_FLAGS)) {
 			DRM_ERROR("invalid flags: %x\n", submit_bo.flags);
 			ret = -EINVAL;
-			goto out_unlock;
+			i = 0;
+			goto out;
 		}
 
+		submit->bos[i].handle = submit_bo.handle;
 		submit->bos[i].flags = submit_bo.flags;
 		/* in validate_objects() we figure out if this is true: */
 		submit->bos[i].iova  = submit_bo.presumed;
+	}
+
+	spin_lock(&file->table_lock);
+
+	for (i = 0; i < args->nr_bos; i++) {
+		struct drm_gem_object *obj;
+		struct msm_gem_object *msm_obj;
 
 		/* normally use drm_gem_object_lookup(), but for bulk lookup
 		 * all under single table_lock just hit object_idr directly:
 		 */
-		obj = idr_find(&file->object_idr, submit_bo.handle);
+		obj = idr_find(&file->object_idr, submit->bos[i].handle);
 		if (!obj) {
-			DRM_ERROR("invalid handle %u at index %u\n", submit_bo.handle, i);
+			DRM_ERROR("invalid handle %u at index %u\n", submit->bos[i].handle, i);
 			ret = -EINVAL;
 			goto out_unlock;
 		}
@@ -142,7 +133,7 @@ static int submit_lookup_objects(struct msm_gem_submit *submit,
 
 		if (!list_empty(&msm_obj->submit_entry)) {
 			DRM_ERROR("handle %u at index %u already on submit list\n",
-					submit_bo.handle, i);
+					submit->bos[i].handle, i);
 			ret = -EINVAL;
 			goto out_unlock;
 		}
@@ -155,7 +146,6 @@ static int submit_lookup_objects(struct msm_gem_submit *submit,
 	}
 
 out_unlock:
-	pagefault_enable();
 	spin_unlock(&file->table_lock);
 
 out:
@@ -173,7 +163,7 @@ static void submit_unlock_unpin_bo(struct msm_gem_submit *submit,
 		msm_gem_unpin_iova(&msm_obj->base, submit->gpu->aspace);
 
 	if (submit->bos[i].flags & BO_LOCKED)
-		ww_mutex_unlock(&msm_obj->resv->lock);
+		ww_mutex_unlock(&msm_obj->base.resv->lock);
 
 	if (backoff && !(submit->bos[i].flags & BO_VALID))
 		submit->bos[i].iova = 0;
@@ -196,7 +186,7 @@ retry:
 		contended = i;
 
 		if (!(submit->bos[i].flags & BO_LOCKED)) {
-			ret = ww_mutex_lock_interruptible(&msm_obj->resv->lock,
+			ret = ww_mutex_lock_interruptible(&msm_obj->base.resv->lock,
 					&submit->ticket);
 			if (ret)
 				goto fail;
@@ -218,7 +208,7 @@ fail:
 	if (ret == -EDEADLK) {
 		struct msm_gem_object *msm_obj = submit->bos[contended].obj;
 		/* we lost out in a seqno race, lock and retry.. */
-		ret = ww_mutex_lock_slow_interruptible(&msm_obj->resv->lock,
+		ret = ww_mutex_lock_slow_interruptible(&msm_obj->base.resv->lock,
 				&submit->ticket);
 		if (!ret) {
 			submit->bos[contended].flags |= BO_LOCKED;
@@ -244,7 +234,7 @@ static int submit_fence_sync(struct msm_gem_submit *submit, bool no_implicit)
 			 * strange place to call it.  OTOH this is a
 			 * convenient can-fail point to hook it in.
 			 */
-			ret = reservation_object_reserve_shared(msm_obj->resv,
+			ret = reservation_object_reserve_shared(msm_obj->base.resv,
 								1);
 			if (ret)
 				return ret;
diff --git a/drivers/gpu/drm/msm/msm_gem_vma.c b/drivers/gpu/drm/msm/msm_gem_vma.c
index 49c04829cf34..fcf7a83f0e6f 100644
--- a/drivers/gpu/drm/msm/msm_gem_vma.c
+++ b/drivers/gpu/drm/msm/msm_gem_vma.c
@@ -85,7 +85,7 @@ msm_gem_map_vma(struct msm_gem_address_space *aspace,
 
 	vma->mapped = true;
 
-	if (aspace->mmu)
+	if (aspace && aspace->mmu)
 		ret = aspace->mmu->funcs->map(aspace->mmu, vma->iova, sgt,
 				size, prot);
 
diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c
index 10babd18e286..bf4ee2766431 100644
--- a/drivers/gpu/drm/msm/msm_gpu.c
+++ b/drivers/gpu/drm/msm/msm_gpu.c
@@ -443,24 +443,15 @@ static void recover_worker(struct work_struct *work)
 	if (submit) {
 		struct task_struct *task;
 
+		/* Increment the fault counts */
+		gpu->global_faults++;
+		submit->queue->faults++;
+
 		task = get_pid_task(submit->pid, PIDTYPE_PID);
 		if (task) {
 			comm = kstrdup(task->comm, GFP_KERNEL);
-
-			/*
-			 * So slightly annoying, in other paths like
-			 * mmap'ing gem buffers, mmap_sem is acquired
-			 * before struct_mutex, which means we can't
-			 * hold struct_mutex across the call to
-			 * get_cmdline().  But submits are retired
-			 * from the same in-order workqueue, so we can
-			 * safely drop the lock here without worrying
-			 * about the submit going away.
-			 */
-			mutex_unlock(&dev->struct_mutex);
 			cmd = kstrdup_quotable_cmdline(task, GFP_KERNEL);
 			put_task_struct(task);
-			mutex_lock(&dev->struct_mutex);
 		}
 
 		if (comm && cmd) {
diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h
index 6241986bab51..f2739cd97cea 100644
--- a/drivers/gpu/drm/msm/msm_gpu.h
+++ b/drivers/gpu/drm/msm/msm_gpu.h
@@ -104,6 +104,9 @@ struct msm_gpu {
 	/* does gpu need hw_init? */
 	bool needs_hw_init;
 
+	/* number of GPU hangs (for all contexts) */
+	int global_faults;
+
 	/* worker for handling active-list retiring: */
 	struct work_struct retire_work;
 
diff --git a/drivers/gpu/drm/msm/msm_iommu.c b/drivers/gpu/drm/msm/msm_iommu.c
index 4d62790cd425..12bb54cefd46 100644
--- a/drivers/gpu/drm/msm/msm_iommu.c
+++ b/drivers/gpu/drm/msm/msm_iommu.c
@@ -38,13 +38,8 @@ static int msm_iommu_attach(struct msm_mmu *mmu, const char * const *names,
 			    int cnt)
 {
 	struct msm_iommu *iommu = to_msm_iommu(mmu);
-	int ret;
 
-	pm_runtime_get_sync(mmu->dev);
-	ret = iommu_attach_device(iommu->domain, mmu->dev);
-	pm_runtime_put_sync(mmu->dev);
-
-	return ret;
+	return iommu_attach_device(iommu->domain, mmu->dev);
 }
 
 static void msm_iommu_detach(struct msm_mmu *mmu, const char * const *names,
@@ -52,9 +47,7 @@ static void msm_iommu_detach(struct msm_mmu *mmu, const char * const *names,
 {
 	struct msm_iommu *iommu = to_msm_iommu(mmu);
 
-	pm_runtime_get_sync(mmu->dev);
 	iommu_detach_device(iommu->domain, mmu->dev);
-	pm_runtime_put_sync(mmu->dev);
 }
 
 static int msm_iommu_map(struct msm_mmu *mmu, uint64_t iova,
@@ -63,9 +56,7 @@ static int msm_iommu_map(struct msm_mmu *mmu, uint64_t iova,
 	struct msm_iommu *iommu = to_msm_iommu(mmu);
 	size_t ret;
 
-//	pm_runtime_get_sync(mmu->dev);
 	ret = iommu_map_sg(iommu->domain, iova, sgt->sgl, sgt->nents, prot);
-//	pm_runtime_put_sync(mmu->dev);
 	WARN_ON(!ret);
 
 	return (ret == len) ? 0 : -EINVAL;
@@ -75,9 +66,7 @@ static int msm_iommu_unmap(struct msm_mmu *mmu, uint64_t iova, unsigned len)
 {
 	struct msm_iommu *iommu = to_msm_iommu(mmu);
 
-	pm_runtime_get_sync(mmu->dev);
 	iommu_unmap(iommu->domain, iova, len);
-	pm_runtime_put_sync(mmu->dev);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/msm/msm_submitqueue.c b/drivers/gpu/drm/msm/msm_submitqueue.c
index 5115f75b5b7f..f160ec40a39b 100644
--- a/drivers/gpu/drm/msm/msm_submitqueue.c
+++ b/drivers/gpu/drm/msm/msm_submitqueue.c
@@ -120,6 +120,47 @@ int msm_submitqueue_init(struct drm_device *drm, struct msm_file_private *ctx)
 	return msm_submitqueue_create(drm, ctx, default_prio, 0, NULL);
 }
 
+static int msm_submitqueue_query_faults(struct msm_gpu_submitqueue *queue,
+		struct drm_msm_submitqueue_query *args)
+{
+	size_t size = min_t(size_t, args->len, sizeof(queue->faults));
+	int ret;
+
+	/* If a zero length was passed in, return the data size we expect */
+	if (!args->len) {
+		args->len = sizeof(queue->faults);
+		return 0;
+	}
+
+	/* Set the length to the actual size of the data */
+	args->len = size;
+
+	ret = copy_to_user(u64_to_user_ptr(args->data), &queue->faults, size);
+
+	return ret ? -EFAULT : 0;
+}
+
+int msm_submitqueue_query(struct drm_device *drm, struct msm_file_private *ctx,
+		struct drm_msm_submitqueue_query *args)
+{
+	struct msm_gpu_submitqueue *queue;
+	int ret = -EINVAL;
+
+	if (args->pad)
+		return -EINVAL;
+
+	queue = msm_submitqueue_get(ctx, args->id);
+	if (!queue)
+		return -ENOENT;
+
+	if (args->param == MSM_SUBMITQUEUE_PARAM_FAULTS)
+		ret = msm_submitqueue_query_faults(queue, args);
+
+	msm_submitqueue_put(queue);
+
+	return ret;
+}
+
 int msm_submitqueue_remove(struct msm_file_private *ctx, u32 id)
 {
 	struct msm_gpu_submitqueue *entry;
diff --git a/drivers/gpu/drm/mxsfb/mxsfb_crtc.c b/drivers/gpu/drm/mxsfb/mxsfb_crtc.c
index 0ee1ca8a316a..98e9bda91e80 100644
--- a/drivers/gpu/drm/mxsfb/mxsfb_crtc.c
+++ b/drivers/gpu/drm/mxsfb/mxsfb_crtc.c
@@ -253,12 +253,12 @@ static void mxsfb_crtc_mode_set_nofb(struct mxsfb_drm_private *mxsfb)
 	if (!(bus_flags & DRM_BUS_FLAG_DE_LOW))
 		vdctrl0 |= VDCTRL0_ENABLE_ACT_HIGH;
 	/*
-	 * DRM_BUS_FLAG_PIXDATA_ defines are controller centric,
+	 * DRM_BUS_FLAG_PIXDATA_DRIVE_ defines are controller centric,
 	 * controllers VDCTRL0_DOTCLK is display centric.
 	 * Drive on positive edge       -> display samples on falling edge
-	 * DRM_BUS_FLAG_PIXDATA_POSEDGE -> VDCTRL0_DOTCLK_ACT_FALLING
+	 * DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE -> VDCTRL0_DOTCLK_ACT_FALLING
 	 */
-	if (bus_flags & DRM_BUS_FLAG_PIXDATA_POSEDGE)
+	if (bus_flags & DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE)
 		vdctrl0 |= VDCTRL0_DOTCLK_ACT_FALLING;
 
 	writel(vdctrl0, mxsfb->base + LCDC_VDCTRL0);
diff --git a/drivers/gpu/drm/nouveau/Kbuild b/drivers/gpu/drm/nouveau/Kbuild
index 581404e6544d..378c5dd692b0 100644
--- a/drivers/gpu/drm/nouveau/Kbuild
+++ b/drivers/gpu/drm/nouveau/Kbuild
@@ -1,7 +1,7 @@
-ccflags-y += -I$(src)/include
-ccflags-y += -I$(src)/include/nvkm
-ccflags-y += -I$(src)/nvkm
-ccflags-y += -I$(src)
+ccflags-y += -I $(srctree)/$(src)/include
+ccflags-y += -I $(srctree)/$(src)/include/nvkm
+ccflags-y += -I $(srctree)/$(src)/nvkm
+ccflags-y += -I $(srctree)/$(src)
 
 # NVKM - HW resource manager
 #- code also used by various userspace tools/tests
diff --git a/drivers/gpu/drm/nouveau/Kconfig b/drivers/gpu/drm/nouveau/Kconfig
index 00cd9ab8948d..553c7da5e8e0 100644
--- a/drivers/gpu/drm/nouveau/Kconfig
+++ b/drivers/gpu/drm/nouveau/Kconfig
@@ -17,10 +17,21 @@ config DRM_NOUVEAU
 	select INPUT if ACPI && X86
 	select THERMAL if ACPI && X86
 	select ACPI_VIDEO if ACPI && X86
-	select DRM_VM
 	help
 	  Choose this option for open-source NVIDIA support.
 
+config NOUVEAU_LEGACY_CTX_SUPPORT
+	bool "Nouveau legacy context support"
+	depends on DRM_NOUVEAU
+	select DRM_LEGACY
+	default y
+	help
+	  There was a version of the nouveau DDX that relied on legacy
+	  ctx ioctls not erroring out. But that was back in time a long
+	  ways, so offer a way to disable it now. For uapi compat with
+	  old nouveau ddx this should be on by default, but modern distros
+	  should consider turning it off.
+
 config NOUVEAU_PLATFORM_DRIVER
 	bool "Nouveau (NVIDIA) SoC GPUs"
 	depends on DRM_NOUVEAU && ARCH_TEGRA
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/i2c.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/i2c.h
index eef54e9b5d77..7957eafa5f0e 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/i2c.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/i2c.h
@@ -38,6 +38,7 @@ struct nvkm_i2c_bus {
 	struct mutex mutex;
 	struct list_head head;
 	struct i2c_adapter i2c;
+	u8 enabled;
 };
 
 int nvkm_i2c_bus_acquire(struct nvkm_i2c_bus *);
@@ -57,6 +58,7 @@ struct nvkm_i2c_aux {
 	struct mutex mutex;
 	struct list_head head;
 	struct i2c_adapter i2c;
+	u8 enabled;
 
 	u32 intr;
 };
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c
index 55c0fa451163..832da8e0020d 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.c
+++ b/drivers/gpu/drm/nouveau/nouveau_display.c
@@ -358,15 +358,6 @@ nouveau_display_hpd_work(struct work_struct *work)
 
 #ifdef CONFIG_ACPI
 
-/*
- * Hans de Goede: This define belongs in acpi/video.h, I've submitted a patch
- * to the acpi subsys to move it there from drivers/acpi/acpi_video.c .
- * This should be dropped once that is merged.
- */
-#ifndef ACPI_VIDEO_NOTIFY_PROBE
-#define ACPI_VIDEO_NOTIFY_PROBE			0x81
-#endif
-
 static int
 nouveau_display_acpi_ntfy(struct notifier_block *nb, unsigned long val,
 			  void *data)
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c
index 5020265bfbd9..22cd45845e07 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.c
@@ -802,10 +802,15 @@ fail_display:
 static int
 nouveau_do_resume(struct drm_device *dev, bool runtime)
 {
+	int ret = 0;
 	struct nouveau_drm *drm = nouveau_drm(dev);
 
 	NV_DEBUG(drm, "resuming object tree...\n");
-	nvif_client_resume(&drm->master.base);
+	ret = nvif_client_resume(&drm->master.base);
+	if (ret) {
+		NV_ERROR(drm, "Client resume failed with error: %d\n", ret);
+		return ret;
+	}
 
 	NV_DEBUG(drm, "resuming fence...\n");
 	if (drm->fence && nouveau_fence(drm)->resume)
@@ -925,6 +930,7 @@ nouveau_pmops_runtime_resume(struct device *dev)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
 	struct drm_device *drm_dev = pci_get_drvdata(pdev);
+	struct nouveau_drm *drm = nouveau_drm(drm_dev);
 	struct nvif_device *device = &nouveau_drm(drm_dev)->client.device;
 	int ret;
 
@@ -941,6 +947,10 @@ nouveau_pmops_runtime_resume(struct device *dev)
 	pci_set_master(pdev);
 
 	ret = nouveau_do_resume(drm_dev, true);
+	if (ret) {
+		NV_ERROR(drm, "resume failed with: %d\n", ret);
+		return ret;
+	}
 
 	/* do magic */
 	nvif_mask(&device->object, 0x088488, (1 << 25), (1 << 25));
@@ -1094,8 +1104,11 @@ nouveau_driver_fops = {
 static struct drm_driver
 driver_stub = {
 	.driver_features =
-		DRIVER_GEM | DRIVER_MODESET | DRIVER_PRIME | DRIVER_RENDER |
-		DRIVER_KMS_LEGACY_CONTEXT,
+		DRIVER_GEM | DRIVER_MODESET | DRIVER_PRIME | DRIVER_RENDER
+#if defined(CONFIG_NOUVEAU_LEGACY_CTX_SUPPORT)
+		| DRIVER_KMS_LEGACY_CONTEXT
+#endif
+		,
 
 	.open = nouveau_drm_open,
 	.postclose = nouveau_drm_postclose,
diff --git a/drivers/gpu/drm/nouveau/nouveau_drv.h b/drivers/gpu/drm/nouveau/nouveau_drv.h
index da847244479d..35ff0ca01a3b 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drv.h
+++ b/drivers/gpu/drm/nouveau/nouveau_drv.h
@@ -60,8 +60,6 @@
 struct nouveau_channel;
 struct platform_device;
 
-#define DRM_FILE_PAGE_OFFSET (0x100000000ULL >> PAGE_SHIFT)
-
 #include "nouveau_fence.h"
 #include "nouveau_bios.h"
 #include "nouveau_vmm.h"
diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.c b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
index 0d3cd4e05728..73cc3217068a 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fbcon.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
@@ -365,14 +365,10 @@ nouveau_fbcon_create(struct drm_fb_helper *helper,
 		ret = PTR_ERR(info);
 		goto out_unlock;
 	}
-	info->skip_vt_switch = 1;
-
-	info->par = fbcon;
 
 	/* setup helper */
 	fbcon->helper.fb = &fb->base;
 
-	strcpy(info->fix.id, "nouveaufb");
 	if (!chan)
 		info->flags = FBINFO_HWACCEL_DISABLED;
 	else
@@ -387,9 +383,7 @@ nouveau_fbcon_create(struct drm_fb_helper *helper,
 	info->screen_base = nvbo_kmap_obj_iovirtual(fb->nvbo);
 	info->screen_size = fb->nvbo->bo.mem.num_pages << PAGE_SHIFT;
 
-	drm_fb_helper_fill_fix(info, fb->base.pitches[0],
-			       fb->base.format->depth);
-	drm_fb_helper_fill_var(info, &fbcon->helper, sizes->fb_width, sizes->fb_height);
+	drm_fb_helper_fill_info(info, &fbcon->helper, sizes);
 
 	/* Use default scratch pixmap (info->pixmap.flags = FB_PIXMAP_SYSTEM) */
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.h b/drivers/gpu/drm/nouveau/nouveau_fbcon.h
index db9d52047ef8..73a7eeba3973 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fbcon.h
+++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.h
@@ -32,7 +32,7 @@
 #include "nouveau_display.h"
 
 struct nouveau_fbdev {
-	struct drm_fb_helper helper;
+	struct drm_fb_helper helper; /* must be first */
 	unsigned int saved_flags;
 	struct nvif_object surf2d;
 	struct nvif_object clip;
diff --git a/drivers/gpu/drm/nouveau/nouveau_ttm.c b/drivers/gpu/drm/nouveau/nouveau_ttm.c
index 1543c2f8d3d3..f0daf958e03a 100644
--- a/drivers/gpu/drm/nouveau/nouveau_ttm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_ttm.c
@@ -168,9 +168,6 @@ nouveau_ttm_mmap(struct file *filp, struct vm_area_struct *vma)
 	struct drm_file *file_priv = filp->private_data;
 	struct nouveau_drm *drm = nouveau_drm(file_priv->minor->dev);
 
-	if (unlikely(vma->vm_pgoff < DRM_FILE_PAGE_OFFSET))
-		return drm_legacy_mmap(filp, vma);
-
 	return ttm_bo_mmap(filp, vma, &drm->ttm.bdev);
 }
 
@@ -239,7 +236,6 @@ nouveau_ttm_init(struct nouveau_drm *drm)
 	ret = ttm_bo_device_init(&drm->ttm.bdev,
 				  &nouveau_bo_driver,
 				  dev->anon_inode->i_mapping,
-				  DRM_FILE_PAGE_OFFSET,
 				  drm->client.mmu.dmabits <= 32 ? true : false);
 	if (ret) {
 		NV_ERROR(drm, "error initialising bo driver, %d\n", ret);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/gf100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/gf100.c
index a3ba7f50198b..a3dcb09a40ee 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/gf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/gf100.c
@@ -94,6 +94,8 @@ gf100_bar_oneinit_bar(struct gf100_bar *bar, struct gf100_barN *bar_vm,
 		return ret;
 
 	bar_len = device->func->resource_size(device, bar_nr);
+	if (!bar_len)
+		return -ENOMEM;
 	if (bar_nr == 3 && bar->bar2_halve)
 		bar_len >>= 1;
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/nv50.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/nv50.c
index 157b076a1272..f23a0ccc2bec 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/nv50.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/nv50.c
@@ -109,7 +109,7 @@ nv50_bar_oneinit(struct nvkm_bar *base)
 	struct nvkm_device *device = bar->base.subdev.device;
 	static struct lock_class_key bar1_lock;
 	static struct lock_class_key bar2_lock;
-	u64 start, limit;
+	u64 start, limit, size;
 	int ret;
 
 	ret = nvkm_gpuobj_new(device, 0x20000, 0, false, NULL, &bar->mem);
@@ -127,7 +127,10 @@ nv50_bar_oneinit(struct nvkm_bar *base)
 
 	/* BAR2 */
 	start = 0x0100000000ULL;
-	limit = start + device->func->resource_size(device, 3);
+	size = device->func->resource_size(device, 3);
+	if (!size)
+		return -ENOMEM;
+	limit = start + size;
 
 	ret = nvkm_vmm_new(device, start, limit-- - start, NULL, 0,
 			   &bar2_lock, "bar2", &bar->bar2_vmm);
@@ -164,10 +167,15 @@ nv50_bar_oneinit(struct nvkm_bar *base)
 
 	/* BAR1 */
 	start = 0x0000000000ULL;
-	limit = start + device->func->resource_size(device, 1);
+	size = device->func->resource_size(device, 1);
+	if (!size)
+		return -ENOMEM;
+	limit = start + size;
 
 	ret = nvkm_vmm_new(device, start, limit-- - start, NULL, 0,
 			   &bar1_lock, "bar1", &bar->bar1_vmm);
+	if (ret)
+		return ret;
 
 	atomic_inc(&bar->bar1_vmm->engref[NVKM_SUBDEV_BAR]);
 	bar->bar1_vmm->debug = bar->base.subdev.debug;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgk104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgk104.c
index 8bcb7e79a0cb..456aed1f2a02 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgk104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgk104.c
@@ -1070,7 +1070,7 @@ gk104_ram_calc_xits(struct gk104_ram *ram, struct nvkm_ram_data *next)
 			nvkm_error(subdev, "unable to calc plls\n");
 			return -EINVAL;
 		}
-		nvkm_debug(subdev, "sucessfully calced PLLs for clock %i kHz"
+		nvkm_debug(subdev, "successfully calced PLLs for clock %i kHz"
 				" (refclock: %i kHz)\n", next->freq, ret);
 	} else {
 		/* calculate refpll coefficients */
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/aux.c b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/aux.c
index 4c1f547da463..b4e7404fe660 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/aux.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/aux.c
@@ -105,9 +105,15 @@ nvkm_i2c_aux_acquire(struct nvkm_i2c_aux *aux)
 {
 	struct nvkm_i2c_pad *pad = aux->pad;
 	int ret;
+
 	AUX_TRACE(aux, "acquire");
 	mutex_lock(&aux->mutex);
-	ret = nvkm_i2c_pad_acquire(pad, NVKM_I2C_PAD_AUX);
+
+	if (aux->enabled)
+		ret = nvkm_i2c_pad_acquire(pad, NVKM_I2C_PAD_AUX);
+	else
+		ret = -EIO;
+
 	if (ret)
 		mutex_unlock(&aux->mutex);
 	return ret;
@@ -145,6 +151,24 @@ nvkm_i2c_aux_del(struct nvkm_i2c_aux **paux)
 	}
 }
 
+void
+nvkm_i2c_aux_init(struct nvkm_i2c_aux *aux)
+{
+	AUX_TRACE(aux, "init");
+	mutex_lock(&aux->mutex);
+	aux->enabled = true;
+	mutex_unlock(&aux->mutex);
+}
+
+void
+nvkm_i2c_aux_fini(struct nvkm_i2c_aux *aux)
+{
+	AUX_TRACE(aux, "fini");
+	mutex_lock(&aux->mutex);
+	aux->enabled = false;
+	mutex_unlock(&aux->mutex);
+}
+
 int
 nvkm_i2c_aux_ctor(const struct nvkm_i2c_aux_func *func,
 		  struct nvkm_i2c_pad *pad, int id,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/aux.h b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/aux.h
index 7d56c4ba693c..08f6b2ee64ab 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/aux.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/aux.h
@@ -16,6 +16,8 @@ int nvkm_i2c_aux_ctor(const struct nvkm_i2c_aux_func *, struct nvkm_i2c_pad *,
 int nvkm_i2c_aux_new_(const struct nvkm_i2c_aux_func *, struct nvkm_i2c_pad *,
 		      int id, struct nvkm_i2c_aux **);
 void nvkm_i2c_aux_del(struct nvkm_i2c_aux **);
+void nvkm_i2c_aux_init(struct nvkm_i2c_aux *);
+void nvkm_i2c_aux_fini(struct nvkm_i2c_aux *);
 int nvkm_i2c_aux_xfer(struct nvkm_i2c_aux *, bool retry, u8 type,
 		      u32 addr, u8 *data, u8 *size);
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/base.c
index 4f197b15acf6..ecacb22834d7 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/base.c
@@ -160,8 +160,18 @@ nvkm_i2c_fini(struct nvkm_subdev *subdev, bool suspend)
 {
 	struct nvkm_i2c *i2c = nvkm_i2c(subdev);
 	struct nvkm_i2c_pad *pad;
+	struct nvkm_i2c_bus *bus;
+	struct nvkm_i2c_aux *aux;
 	u32 mask;
 
+	list_for_each_entry(aux, &i2c->aux, head) {
+		nvkm_i2c_aux_fini(aux);
+	}
+
+	list_for_each_entry(bus, &i2c->bus, head) {
+		nvkm_i2c_bus_fini(bus);
+	}
+
 	if ((mask = (1 << i2c->func->aux) - 1), i2c->func->aux_stat) {
 		i2c->func->aux_mask(i2c, NVKM_I2C_ANY, mask, 0);
 		i2c->func->aux_stat(i2c, &mask, &mask, &mask, &mask);
@@ -180,6 +190,7 @@ nvkm_i2c_init(struct nvkm_subdev *subdev)
 	struct nvkm_i2c *i2c = nvkm_i2c(subdev);
 	struct nvkm_i2c_bus *bus;
 	struct nvkm_i2c_pad *pad;
+	struct nvkm_i2c_aux *aux;
 
 	list_for_each_entry(pad, &i2c->pad, head) {
 		nvkm_i2c_pad_init(pad);
@@ -189,6 +200,10 @@ nvkm_i2c_init(struct nvkm_subdev *subdev)
 		nvkm_i2c_bus_init(bus);
 	}
 
+	list_for_each_entry(aux, &i2c->aux, head) {
+		nvkm_i2c_aux_init(aux);
+	}
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/bus.c b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/bus.c
index 807a2b67bd64..ed50cc3736b9 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/bus.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/bus.c
@@ -110,6 +110,19 @@ nvkm_i2c_bus_init(struct nvkm_i2c_bus *bus)
 	BUS_TRACE(bus, "init");
 	if (bus->func->init)
 		bus->func->init(bus);
+
+	mutex_lock(&bus->mutex);
+	bus->enabled = true;
+	mutex_unlock(&bus->mutex);
+}
+
+void
+nvkm_i2c_bus_fini(struct nvkm_i2c_bus *bus)
+{
+	BUS_TRACE(bus, "fini");
+	mutex_lock(&bus->mutex);
+	bus->enabled = false;
+	mutex_unlock(&bus->mutex);
 }
 
 void
@@ -126,9 +139,15 @@ nvkm_i2c_bus_acquire(struct nvkm_i2c_bus *bus)
 {
 	struct nvkm_i2c_pad *pad = bus->pad;
 	int ret;
+
 	BUS_TRACE(bus, "acquire");
 	mutex_lock(&bus->mutex);
-	ret = nvkm_i2c_pad_acquire(pad, NVKM_I2C_PAD_I2C);
+
+	if (bus->enabled)
+		ret = nvkm_i2c_pad_acquire(pad, NVKM_I2C_PAD_I2C);
+	else
+		ret = -EIO;
+
 	if (ret)
 		mutex_unlock(&bus->mutex);
 	return ret;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/bus.h b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/bus.h
index bea0dd33961e..465464bba58b 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/bus.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/bus.h
@@ -18,6 +18,7 @@ int nvkm_i2c_bus_new_(const struct nvkm_i2c_bus_func *, struct nvkm_i2c_pad *,
 		      int id, struct nvkm_i2c_bus **);
 void nvkm_i2c_bus_del(struct nvkm_i2c_bus **);
 void nvkm_i2c_bus_init(struct nvkm_i2c_bus *);
+void nvkm_i2c_bus_fini(struct nvkm_i2c_bus *);
 
 int nvkm_i2c_bit_xfer(struct nvkm_i2c_bus *, struct i2c_msg *, int);
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.c
index fa93f964e6a4..41640e0584ac 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.c
@@ -1783,7 +1783,7 @@ nvkm_vmm_get(struct nvkm_vmm *vmm, u8 page, u64 size, struct nvkm_vma **pvma)
 void
 nvkm_vmm_part(struct nvkm_vmm *vmm, struct nvkm_memory *inst)
 {
-	if (inst && vmm->func->part) {
+	if (inst && vmm && vmm->func->part) {
 		mutex_lock(&vmm->mutex);
 		vmm->func->part(vmm, inst);
 		mutex_unlock(&vmm->mutex);
diff --git a/drivers/gpu/drm/omapdrm/displays/Kconfig b/drivers/gpu/drm/omapdrm/displays/Kconfig
index a349cb61961e..7b0bcb494b5c 100644
--- a/drivers/gpu/drm/omapdrm/displays/Kconfig
+++ b/drivers/gpu/drm/omapdrm/displays/Kconfig
@@ -6,23 +6,12 @@ config DRM_OMAP_ENCODER_OPA362
 	  Driver for OPA362 external analog TV amplifier controlled
 	  through a GPIO.
 
-config DRM_OMAP_ENCODER_TFP410
-        tristate "TFP410 DPI to DVI Encoder"
-	help
-	  Driver for TFP410 DPI to DVI encoder.
-
 config DRM_OMAP_ENCODER_TPD12S015
         tristate "TPD12S015 HDMI ESD protection and level shifter"
 	help
 	  Driver for TPD12S015, which offers HDMI ESD protection and level
 	  shifting.
 
-config DRM_OMAP_CONNECTOR_DVI
-        tristate "DVI Connector"
-	depends on I2C
-	help
-	  Driver for a generic DVI connector.
-
 config DRM_OMAP_CONNECTOR_HDMI
         tristate "HDMI Connector"
 	help
@@ -33,12 +22,6 @@ config DRM_OMAP_CONNECTOR_ANALOG_TV
 	help
 	  Driver for a generic analog TV connector.
 
-config DRM_OMAP_PANEL_DPI
-	tristate "Generic DPI panel"
-	depends on BACKLIGHT_CLASS_DEVICE
-	help
-	  Driver for generic DPI panels.
-
 config DRM_OMAP_PANEL_DSI_CM
 	tristate "Generic DSI Command Mode Panel"
 	depends on BACKLIGHT_CLASS_DEVICE
diff --git a/drivers/gpu/drm/omapdrm/displays/Makefile b/drivers/gpu/drm/omapdrm/displays/Makefile
index d99659e1381b..1db34d4fed64 100644
--- a/drivers/gpu/drm/omapdrm/displays/Makefile
+++ b/drivers/gpu/drm/omapdrm/displays/Makefile
@@ -1,11 +1,8 @@
 # SPDX-License-Identifier: GPL-2.0
 obj-$(CONFIG_DRM_OMAP_ENCODER_OPA362) += encoder-opa362.o
-obj-$(CONFIG_DRM_OMAP_ENCODER_TFP410) += encoder-tfp410.o
 obj-$(CONFIG_DRM_OMAP_ENCODER_TPD12S015) += encoder-tpd12s015.o
-obj-$(CONFIG_DRM_OMAP_CONNECTOR_DVI) += connector-dvi.o
 obj-$(CONFIG_DRM_OMAP_CONNECTOR_HDMI) += connector-hdmi.o
 obj-$(CONFIG_DRM_OMAP_CONNECTOR_ANALOG_TV) += connector-analog-tv.o
-obj-$(CONFIG_DRM_OMAP_PANEL_DPI) += panel-dpi.o
 obj-$(CONFIG_DRM_OMAP_PANEL_DSI_CM) += panel-dsi-cm.o
 obj-$(CONFIG_DRM_OMAP_PANEL_SONY_ACX565AKM) += panel-sony-acx565akm.o
 obj-$(CONFIG_DRM_OMAP_PANEL_LGPHILIPS_LB035Q02) += panel-lgphilips-lb035q02.o
diff --git a/drivers/gpu/drm/omapdrm/displays/connector-analog-tv.c b/drivers/gpu/drm/omapdrm/displays/connector-analog-tv.c
index 28a3ce8f88d2..6c0561101874 100644
--- a/drivers/gpu/drm/omapdrm/displays/connector-analog-tv.c
+++ b/drivers/gpu/drm/omapdrm/displays/connector-analog-tv.c
@@ -35,50 +35,9 @@ static void tvc_disconnect(struct omap_dss_device *src,
 {
 }
 
-static int tvc_enable(struct omap_dss_device *dssdev)
-{
-	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
-	int r;
-
-	dev_dbg(ddata->dev, "enable\n");
-
-	if (!omapdss_device_is_connected(dssdev))
-		return -ENODEV;
-
-	if (omapdss_device_is_enabled(dssdev))
-		return 0;
-
-	r = src->ops->enable(src);
-	if (r)
-		return r;
-
-	dssdev->state = OMAP_DSS_DISPLAY_ACTIVE;
-
-	return r;
-}
-
-static void tvc_disable(struct omap_dss_device *dssdev)
-{
-	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
-
-	dev_dbg(ddata->dev, "disable\n");
-
-	if (!omapdss_device_is_enabled(dssdev))
-		return;
-
-	src->ops->disable(src);
-
-	dssdev->state = OMAP_DSS_DISPLAY_DISABLED;
-}
-
 static const struct omap_dss_device_ops tvc_ops = {
 	.connect		= tvc_connect,
 	.disconnect		= tvc_disconnect,
-
-	.enable			= tvc_enable,
-	.disable		= tvc_disable,
 };
 
 static int tvc_probe(struct platform_device *pdev)
@@ -97,6 +56,7 @@ static int tvc_probe(struct platform_device *pdev)
 	dssdev->ops = &tvc_ops;
 	dssdev->dev = &pdev->dev;
 	dssdev->type = OMAP_DISPLAY_TYPE_VENC;
+	dssdev->display = true;
 	dssdev->owner = THIS_MODULE;
 	dssdev->of_ports = BIT(0);
 
@@ -109,12 +69,9 @@ static int tvc_probe(struct platform_device *pdev)
 static int __exit tvc_remove(struct platform_device *pdev)
 {
 	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
-	struct omap_dss_device *dssdev = &ddata->dssdev;
 
 	omapdss_device_unregister(&ddata->dssdev);
 
-	tvc_disable(dssdev);
-
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/omapdrm/displays/connector-dvi.c b/drivers/gpu/drm/omapdrm/displays/connector-dvi.c
deleted file mode 100644
index 24b14f44248e..000000000000
--- a/drivers/gpu/drm/omapdrm/displays/connector-dvi.c
+++ /dev/null
@@ -1,330 +0,0 @@
-/*
- * Generic DVI Connector driver
- *
- * Copyright (C) 2013 Texas Instruments Incorporated - http://www.ti.com/
- * Author: Tomi Valkeinen <tomi.valkeinen@ti.com>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License version 2 as published by
- * the Free Software Foundation.
- */
-
-#include <linux/gpio/consumer.h>
-#include <linux/i2c.h>
-#include <linux/module.h>
-#include <linux/platform_device.h>
-#include <linux/slab.h>
-
-#include <drm/drm_edid.h>
-
-#include "../dss/omapdss.h"
-
-struct panel_drv_data {
-	struct omap_dss_device dssdev;
-
-	struct i2c_adapter *i2c_adapter;
-
-	struct gpio_desc *hpd_gpio;
-
-	void (*hpd_cb)(void *cb_data, enum drm_connector_status status);
-	void *hpd_cb_data;
-	bool hpd_enabled;
-	/* mutex for hpd fields above */
-	struct mutex hpd_lock;
-};
-
-#define to_panel_data(x) container_of(x, struct panel_drv_data, dssdev)
-
-static int dvic_connect(struct omap_dss_device *src,
-			struct omap_dss_device *dst)
-{
-	return 0;
-}
-
-static void dvic_disconnect(struct omap_dss_device *src,
-			    struct omap_dss_device *dst)
-{
-}
-
-static int dvic_enable(struct omap_dss_device *dssdev)
-{
-	struct omap_dss_device *src = dssdev->src;
-	int r;
-
-	if (!omapdss_device_is_connected(dssdev))
-		return -ENODEV;
-
-	if (omapdss_device_is_enabled(dssdev))
-		return 0;
-
-	r = src->ops->enable(src);
-	if (r)
-		return r;
-
-	dssdev->state = OMAP_DSS_DISPLAY_ACTIVE;
-
-	return 0;
-}
-
-static void dvic_disable(struct omap_dss_device *dssdev)
-{
-	struct omap_dss_device *src = dssdev->src;
-
-	if (!omapdss_device_is_enabled(dssdev))
-		return;
-
-	src->ops->disable(src);
-
-	dssdev->state = OMAP_DSS_DISPLAY_DISABLED;
-}
-
-static int dvic_ddc_read(struct i2c_adapter *adapter,
-		unsigned char *buf, u16 count, u8 offset)
-{
-	int r, retries;
-
-	for (retries = 3; retries > 0; retries--) {
-		struct i2c_msg msgs[] = {
-			{
-				.addr   = DDC_ADDR,
-				.flags  = 0,
-				.len    = 1,
-				.buf    = &offset,
-			}, {
-				.addr   = DDC_ADDR,
-				.flags  = I2C_M_RD,
-				.len    = count,
-				.buf    = buf,
-			}
-		};
-
-		r = i2c_transfer(adapter, msgs, 2);
-		if (r == 2)
-			return 0;
-
-		if (r != -EAGAIN)
-			break;
-	}
-
-	return r < 0 ? r : -EIO;
-}
-
-static int dvic_read_edid(struct omap_dss_device *dssdev,
-		u8 *edid, int len)
-{
-	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	int r, l, bytes_read;
-
-	l = min(EDID_LENGTH, len);
-	r = dvic_ddc_read(ddata->i2c_adapter, edid, l, 0);
-	if (r)
-		return r;
-
-	bytes_read = l;
-
-	/* if there are extensions, read second block */
-	if (len > EDID_LENGTH && edid[0x7e] > 0) {
-		l = min(EDID_LENGTH, len - EDID_LENGTH);
-
-		r = dvic_ddc_read(ddata->i2c_adapter, edid + EDID_LENGTH,
-				l, EDID_LENGTH);
-		if (r)
-			return r;
-
-		bytes_read += l;
-	}
-
-	return bytes_read;
-}
-
-static bool dvic_detect(struct omap_dss_device *dssdev)
-{
-	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	unsigned char out;
-	int r;
-
-	if (ddata->hpd_gpio)
-		return gpiod_get_value_cansleep(ddata->hpd_gpio);
-
-	if (!ddata->i2c_adapter)
-		return true;
-
-	r = dvic_ddc_read(ddata->i2c_adapter, &out, 1, 0);
-
-	return r == 0;
-}
-
-static void dvic_register_hpd_cb(struct omap_dss_device *dssdev,
-				 void (*cb)(void *cb_data,
-					    enum drm_connector_status status),
-				 void *cb_data)
-{
-	struct panel_drv_data *ddata = to_panel_data(dssdev);
-
-	mutex_lock(&ddata->hpd_lock);
-	ddata->hpd_cb = cb;
-	ddata->hpd_cb_data = cb_data;
-	mutex_unlock(&ddata->hpd_lock);
-}
-
-static void dvic_unregister_hpd_cb(struct omap_dss_device *dssdev)
-{
-	struct panel_drv_data *ddata = to_panel_data(dssdev);
-
-	mutex_lock(&ddata->hpd_lock);
-	ddata->hpd_cb = NULL;
-	ddata->hpd_cb_data = NULL;
-	mutex_unlock(&ddata->hpd_lock);
-}
-
-static const struct omap_dss_device_ops dvic_ops = {
-	.connect	= dvic_connect,
-	.disconnect	= dvic_disconnect,
-
-	.enable		= dvic_enable,
-	.disable	= dvic_disable,
-
-	.read_edid	= dvic_read_edid,
-	.detect		= dvic_detect,
-
-	.register_hpd_cb	= dvic_register_hpd_cb,
-	.unregister_hpd_cb	= dvic_unregister_hpd_cb,
-};
-
-static irqreturn_t dvic_hpd_isr(int irq, void *data)
-{
-	struct panel_drv_data *ddata = data;
-
-	mutex_lock(&ddata->hpd_lock);
-	if (ddata->hpd_enabled && ddata->hpd_cb) {
-		enum drm_connector_status status;
-
-		if (dvic_detect(&ddata->dssdev))
-			status = connector_status_connected;
-		else
-			status = connector_status_disconnected;
-
-		ddata->hpd_cb(ddata->hpd_cb_data, status);
-	}
-	mutex_unlock(&ddata->hpd_lock);
-
-	return IRQ_HANDLED;
-}
-
-static int dvic_probe_of(struct platform_device *pdev)
-{
-	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
-	struct device_node *node = pdev->dev.of_node;
-	struct device_node *adapter_node;
-	struct i2c_adapter *adapter;
-	struct gpio_desc *gpio;
-	int r;
-
-	gpio = devm_gpiod_get_optional(&pdev->dev, "hpd", GPIOD_IN);
-	if (IS_ERR(gpio)) {
-		dev_err(&pdev->dev, "failed to parse HPD gpio\n");
-		return PTR_ERR(gpio);
-	}
-
-	ddata->hpd_gpio = gpio;
-
-	mutex_init(&ddata->hpd_lock);
-
-	if (ddata->hpd_gpio) {
-		r = devm_request_threaded_irq(&pdev->dev,
-			gpiod_to_irq(ddata->hpd_gpio), NULL, dvic_hpd_isr,
-			IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING | IRQF_ONESHOT,
-			"DVI HPD", ddata);
-		if (r)
-			return r;
-	}
-
-	adapter_node = of_parse_phandle(node, "ddc-i2c-bus", 0);
-	if (adapter_node) {
-		adapter = of_get_i2c_adapter_by_node(adapter_node);
-		of_node_put(adapter_node);
-		if (adapter == NULL) {
-			dev_err(&pdev->dev, "failed to parse ddc-i2c-bus\n");
-			return -EPROBE_DEFER;
-		}
-
-		ddata->i2c_adapter = adapter;
-	}
-
-	return 0;
-}
-
-static int dvic_probe(struct platform_device *pdev)
-{
-	struct panel_drv_data *ddata;
-	struct omap_dss_device *dssdev;
-	int r;
-
-	ddata = devm_kzalloc(&pdev->dev, sizeof(*ddata), GFP_KERNEL);
-	if (!ddata)
-		return -ENOMEM;
-
-	platform_set_drvdata(pdev, ddata);
-
-	r = dvic_probe_of(pdev);
-	if (r)
-		return r;
-
-	dssdev = &ddata->dssdev;
-	dssdev->ops = &dvic_ops;
-	dssdev->dev = &pdev->dev;
-	dssdev->type = OMAP_DISPLAY_TYPE_DVI;
-	dssdev->owner = THIS_MODULE;
-	dssdev->of_ports = BIT(0);
-
-	if (ddata->hpd_gpio)
-		dssdev->ops_flags |= OMAP_DSS_DEVICE_OP_DETECT
-				  |  OMAP_DSS_DEVICE_OP_HPD;
-	if (ddata->i2c_adapter)
-		dssdev->ops_flags |= OMAP_DSS_DEVICE_OP_DETECT
-				  |  OMAP_DSS_DEVICE_OP_EDID;
-
-	omapdss_display_init(dssdev);
-	omapdss_device_register(dssdev);
-
-	return 0;
-}
-
-static int __exit dvic_remove(struct platform_device *pdev)
-{
-	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
-	struct omap_dss_device *dssdev = &ddata->dssdev;
-
-	omapdss_device_unregister(&ddata->dssdev);
-
-	dvic_disable(dssdev);
-
-	i2c_put_adapter(ddata->i2c_adapter);
-
-	mutex_destroy(&ddata->hpd_lock);
-
-	return 0;
-}
-
-static const struct of_device_id dvic_of_match[] = {
-	{ .compatible = "omapdss,dvi-connector", },
-	{},
-};
-
-MODULE_DEVICE_TABLE(of, dvic_of_match);
-
-static struct platform_driver dvi_connector_driver = {
-	.probe	= dvic_probe,
-	.remove	= __exit_p(dvic_remove),
-	.driver	= {
-		.name	= "connector-dvi",
-		.of_match_table = dvic_of_match,
-		.suppress_bind_attrs = true,
-	},
-};
-
-module_platform_driver(dvi_connector_driver);
-
-MODULE_AUTHOR("Tomi Valkeinen <tomi.valkeinen@ti.com>");
-MODULE_DESCRIPTION("Generic DVI Connector driver");
-MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/omapdrm/displays/connector-hdmi.c b/drivers/gpu/drm/omapdrm/displays/connector-hdmi.c
index e602fa4a50a4..68d6f6e44b03 100644
--- a/drivers/gpu/drm/omapdrm/displays/connector-hdmi.c
+++ b/drivers/gpu/drm/omapdrm/displays/connector-hdmi.c
@@ -41,44 +41,6 @@ static void hdmic_disconnect(struct omap_dss_device *src,
 {
 }
 
-static int hdmic_enable(struct omap_dss_device *dssdev)
-{
-	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
-	int r;
-
-	dev_dbg(ddata->dev, "enable\n");
-
-	if (!omapdss_device_is_connected(dssdev))
-		return -ENODEV;
-
-	if (omapdss_device_is_enabled(dssdev))
-		return 0;
-
-	r = src->ops->enable(src);
-	if (r)
-		return r;
-
-	dssdev->state = OMAP_DSS_DISPLAY_ACTIVE;
-
-	return r;
-}
-
-static void hdmic_disable(struct omap_dss_device *dssdev)
-{
-	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
-
-	dev_dbg(ddata->dev, "disable\n");
-
-	if (!omapdss_device_is_enabled(dssdev))
-		return;
-
-	src->ops->disable(src);
-
-	dssdev->state = OMAP_DSS_DISPLAY_DISABLED;
-}
-
 static bool hdmic_detect(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
@@ -113,9 +75,6 @@ static const struct omap_dss_device_ops hdmic_ops = {
 	.connect		= hdmic_connect,
 	.disconnect		= hdmic_disconnect,
 
-	.enable			= hdmic_enable,
-	.disable		= hdmic_disable,
-
 	.detect			= hdmic_detect,
 	.register_hpd_cb	= hdmic_register_hpd_cb,
 	.unregister_hpd_cb	= hdmic_unregister_hpd_cb,
@@ -181,6 +140,7 @@ static int hdmic_probe(struct platform_device *pdev)
 	dssdev->ops = &hdmic_ops;
 	dssdev->dev = &pdev->dev;
 	dssdev->type = OMAP_DISPLAY_TYPE_HDMI;
+	dssdev->display = true;
 	dssdev->owner = THIS_MODULE;
 	dssdev->of_ports = BIT(0);
 	dssdev->ops_flags = ddata->hpd_gpio
@@ -196,12 +156,9 @@ static int hdmic_probe(struct platform_device *pdev)
 static int __exit hdmic_remove(struct platform_device *pdev)
 {
 	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
-	struct omap_dss_device *dssdev = &ddata->dssdev;
 
 	omapdss_device_unregister(&ddata->dssdev);
 
-	hdmic_disable(dssdev);
-
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/omapdrm/displays/encoder-opa362.c b/drivers/gpu/drm/omapdrm/displays/encoder-opa362.c
index 4fefd80f53bb..29a5a130ebd1 100644
--- a/drivers/gpu/drm/omapdrm/displays/encoder-opa362.c
+++ b/drivers/gpu/drm/omapdrm/displays/encoder-opa362.c
@@ -41,48 +41,20 @@ static void opa362_disconnect(struct omap_dss_device *src,
 	omapdss_device_disconnect(dst, dst->next);
 }
 
-static int opa362_enable(struct omap_dss_device *dssdev)
+static void opa362_enable(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
-	int r;
-
-	dev_dbg(dssdev->dev, "enable\n");
-
-	if (!omapdss_device_is_connected(dssdev))
-		return -ENODEV;
-
-	if (omapdss_device_is_enabled(dssdev))
-		return 0;
-
-	r = src->ops->enable(src);
-	if (r)
-		return r;
 
 	if (ddata->enable_gpio)
 		gpiod_set_value_cansleep(ddata->enable_gpio, 1);
-
-	dssdev->state = OMAP_DSS_DISPLAY_ACTIVE;
-
-	return 0;
 }
 
 static void opa362_disable(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
-
-	dev_dbg(dssdev->dev, "disable\n");
-
-	if (!omapdss_device_is_enabled(dssdev))
-		return;
 
 	if (ddata->enable_gpio)
 		gpiod_set_value_cansleep(ddata->enable_gpio, 0);
-
-	src->ops->disable(src);
-
-	dssdev->state = OMAP_DSS_DISPLAY_DISABLED;
 }
 
 static const struct omap_dss_device_ops opa362_ops = {
@@ -116,7 +88,6 @@ static int opa362_probe(struct platform_device *pdev)
 	dssdev->ops = &opa362_ops;
 	dssdev->dev = &pdev->dev;
 	dssdev->type = OMAP_DISPLAY_TYPE_VENC;
-	dssdev->output_type = OMAP_DISPLAY_TYPE_VENC;
 	dssdev->owner = THIS_MODULE;
 	dssdev->of_ports = BIT(1) | BIT(0);
 
@@ -141,13 +112,7 @@ static int __exit opa362_remove(struct platform_device *pdev)
 		omapdss_device_put(dssdev->next);
 	omapdss_device_unregister(&ddata->dssdev);
 
-	WARN_ON(omapdss_device_is_enabled(dssdev));
-	if (omapdss_device_is_enabled(dssdev))
-		opa362_disable(dssdev);
-
-	WARN_ON(omapdss_device_is_connected(dssdev));
-	if (omapdss_device_is_connected(dssdev))
-		omapdss_device_disconnect(NULL, dssdev);
+	opa362_disable(dssdev);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/omapdrm/displays/encoder-tfp410.c b/drivers/gpu/drm/omapdrm/displays/encoder-tfp410.c
deleted file mode 100644
index f1a748353279..000000000000
--- a/drivers/gpu/drm/omapdrm/displays/encoder-tfp410.c
+++ /dev/null
@@ -1,170 +0,0 @@
-/*
- * TFP410 DPI-to-DVI encoder driver
- *
- * Copyright (C) 2013 Texas Instruments Incorporated - http://www.ti.com/
- * Author: Tomi Valkeinen <tomi.valkeinen@ti.com>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License version 2 as published by
- * the Free Software Foundation.
- */
-
-#include <linux/gpio/consumer.h>
-#include <linux/module.h>
-#include <linux/platform_device.h>
-#include <linux/slab.h>
-
-#include "../dss/omapdss.h"
-
-struct panel_drv_data {
-	struct omap_dss_device dssdev;
-
-	struct gpio_desc *pd_gpio;
-};
-
-#define to_panel_data(x) container_of(x, struct panel_drv_data, dssdev)
-
-static int tfp410_connect(struct omap_dss_device *src,
-			  struct omap_dss_device *dst)
-{
-	return omapdss_device_connect(dst->dss, dst, dst->next);
-}
-
-static void tfp410_disconnect(struct omap_dss_device *src,
-			      struct omap_dss_device *dst)
-{
-	omapdss_device_disconnect(dst, dst->next);
-}
-
-static int tfp410_enable(struct omap_dss_device *dssdev)
-{
-	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
-	int r;
-
-	if (!omapdss_device_is_connected(dssdev))
-		return -ENODEV;
-
-	if (omapdss_device_is_enabled(dssdev))
-		return 0;
-
-	r = src->ops->enable(src);
-	if (r)
-		return r;
-
-	if (ddata->pd_gpio)
-		gpiod_set_value_cansleep(ddata->pd_gpio, 0);
-
-	dssdev->state = OMAP_DSS_DISPLAY_ACTIVE;
-
-	return 0;
-}
-
-static void tfp410_disable(struct omap_dss_device *dssdev)
-{
-	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
-
-	if (!omapdss_device_is_enabled(dssdev))
-		return;
-
-	if (ddata->pd_gpio)
-		gpiod_set_value_cansleep(ddata->pd_gpio, 0);
-
-	src->ops->disable(src);
-
-	dssdev->state = OMAP_DSS_DISPLAY_DISABLED;
-}
-
-static const struct omap_dss_device_ops tfp410_ops = {
-	.connect	= tfp410_connect,
-	.disconnect	= tfp410_disconnect,
-	.enable		= tfp410_enable,
-	.disable	= tfp410_disable,
-};
-
-static int tfp410_probe(struct platform_device *pdev)
-{
-	struct panel_drv_data *ddata;
-	struct omap_dss_device *dssdev;
-	struct gpio_desc *gpio;
-
-	ddata = devm_kzalloc(&pdev->dev, sizeof(*ddata), GFP_KERNEL);
-	if (!ddata)
-		return -ENOMEM;
-
-	platform_set_drvdata(pdev, ddata);
-
-	/* Powerdown GPIO */
-	gpio = devm_gpiod_get_optional(&pdev->dev, "powerdown", GPIOD_OUT_HIGH);
-	if (IS_ERR(gpio)) {
-		dev_err(&pdev->dev, "failed to parse powerdown gpio\n");
-		return PTR_ERR(gpio);
-	}
-
-	ddata->pd_gpio = gpio;
-
-	dssdev = &ddata->dssdev;
-	dssdev->ops = &tfp410_ops;
-	dssdev->dev = &pdev->dev;
-	dssdev->type = OMAP_DISPLAY_TYPE_DPI;
-	dssdev->output_type = OMAP_DISPLAY_TYPE_DVI;
-	dssdev->owner = THIS_MODULE;
-	dssdev->of_ports = BIT(1) | BIT(0);
-	dssdev->bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_SYNC_POSEDGE
-			  | DRM_BUS_FLAG_PIXDATA_POSEDGE;
-
-	dssdev->next = omapdss_of_find_connected_device(pdev->dev.of_node, 1);
-	if (IS_ERR(dssdev->next)) {
-		if (PTR_ERR(dssdev->next) != -EPROBE_DEFER)
-			dev_err(&pdev->dev, "failed to find video sink\n");
-		return PTR_ERR(dssdev->next);
-	}
-
-	omapdss_device_register(dssdev);
-
-	return 0;
-}
-
-static int __exit tfp410_remove(struct platform_device *pdev)
-{
-	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
-	struct omap_dss_device *dssdev = &ddata->dssdev;
-
-	if (dssdev->next)
-		omapdss_device_put(dssdev->next);
-	omapdss_device_unregister(&ddata->dssdev);
-
-	WARN_ON(omapdss_device_is_enabled(dssdev));
-	if (omapdss_device_is_enabled(dssdev))
-		tfp410_disable(dssdev);
-
-	WARN_ON(omapdss_device_is_connected(dssdev));
-	if (omapdss_device_is_connected(dssdev))
-		omapdss_device_disconnect(NULL, dssdev);
-
-	return 0;
-}
-
-static const struct of_device_id tfp410_of_match[] = {
-	{ .compatible = "omapdss,ti,tfp410", },
-	{},
-};
-
-MODULE_DEVICE_TABLE(of, tfp410_of_match);
-
-static struct platform_driver tfp410_driver = {
-	.probe	= tfp410_probe,
-	.remove	= __exit_p(tfp410_remove),
-	.driver	= {
-		.name	= "tfp410",
-		.of_match_table = tfp410_of_match,
-		.suppress_bind_attrs = true,
-	},
-};
-
-module_platform_driver(tfp410_driver);
-
-MODULE_AUTHOR("Tomi Valkeinen <tomi.valkeinen@ti.com>");
-MODULE_DESCRIPTION("TFP410 DPI to DVI encoder driver");
-MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/omapdrm/displays/encoder-tpd12s015.c b/drivers/gpu/drm/omapdrm/displays/encoder-tpd12s015.c
index 94de55fd8884..bc03752d2762 100644
--- a/drivers/gpu/drm/omapdrm/displays/encoder-tpd12s015.c
+++ b/drivers/gpu/drm/omapdrm/displays/encoder-tpd12s015.c
@@ -62,35 +62,6 @@ static void tpd_disconnect(struct omap_dss_device *src,
 	omapdss_device_disconnect(dst, dst->next);
 }
 
-static int tpd_enable(struct omap_dss_device *dssdev)
-{
-	struct omap_dss_device *src = dssdev->src;
-	int r;
-
-	if (dssdev->state == OMAP_DSS_DISPLAY_ACTIVE)
-		return 0;
-
-	r = src->ops->enable(src);
-	if (r)
-		return r;
-
-	dssdev->state = OMAP_DSS_DISPLAY_ACTIVE;
-
-	return r;
-}
-
-static void tpd_disable(struct omap_dss_device *dssdev)
-{
-	struct omap_dss_device *src = dssdev->src;
-
-	if (dssdev->state != OMAP_DSS_DISPLAY_ACTIVE)
-		return;
-
-	src->ops->disable(src);
-
-	dssdev->state = OMAP_DSS_DISPLAY_DISABLED;
-}
-
 static bool tpd_detect(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
@@ -124,8 +95,6 @@ static void tpd_unregister_hpd_cb(struct omap_dss_device *dssdev)
 static const struct omap_dss_device_ops tpd_ops = {
 	.connect		= tpd_connect,
 	.disconnect		= tpd_disconnect,
-	.enable			= tpd_enable,
-	.disable		= tpd_disable,
 	.detect			= tpd_detect,
 	.register_hpd_cb	= tpd_register_hpd_cb,
 	.unregister_hpd_cb	= tpd_unregister_hpd_cb,
@@ -198,7 +167,6 @@ static int tpd_probe(struct platform_device *pdev)
 	dssdev->ops = &tpd_ops;
 	dssdev->dev = &pdev->dev;
 	dssdev->type = OMAP_DISPLAY_TYPE_HDMI;
-	dssdev->output_type = OMAP_DISPLAY_TYPE_HDMI;
 	dssdev->owner = THIS_MODULE;
 	dssdev->of_ports = BIT(1) | BIT(0);
 	dssdev->ops_flags = OMAP_DSS_DEVICE_OP_DETECT
@@ -225,14 +193,6 @@ static int __exit tpd_remove(struct platform_device *pdev)
 		omapdss_device_put(dssdev->next);
 	omapdss_device_unregister(&ddata->dssdev);
 
-	WARN_ON(omapdss_device_is_enabled(dssdev));
-	if (omapdss_device_is_enabled(dssdev))
-		tpd_disable(dssdev);
-
-	WARN_ON(omapdss_device_is_connected(dssdev));
-	if (omapdss_device_is_connected(dssdev))
-		omapdss_device_disconnect(NULL, dssdev);
-
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/omapdrm/displays/panel-dpi.c b/drivers/gpu/drm/omapdrm/displays/panel-dpi.c
deleted file mode 100644
index 465120809eb3..000000000000
--- a/drivers/gpu/drm/omapdrm/displays/panel-dpi.c
+++ /dev/null
@@ -1,221 +0,0 @@
-/*
- * Generic MIPI DPI Panel Driver
- *
- * Copyright (C) 2013 Texas Instruments Incorporated - http://www.ti.com/
- * Author: Tomi Valkeinen <tomi.valkeinen@ti.com>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License version 2 as published by
- * the Free Software Foundation.
- */
-
-#include <linux/gpio/consumer.h>
-#include <linux/module.h>
-#include <linux/platform_device.h>
-#include <linux/slab.h>
-#include <linux/of.h>
-#include <linux/regulator/consumer.h>
-#include <linux/backlight.h>
-
-#include <video/of_display_timing.h>
-
-#include "../dss/omapdss.h"
-
-struct panel_drv_data {
-	struct omap_dss_device dssdev;
-
-	struct videomode vm;
-
-	struct backlight_device *backlight;
-
-	struct gpio_desc *enable_gpio;
-	struct regulator *vcc_supply;
-};
-
-#define to_panel_data(p) container_of(p, struct panel_drv_data, dssdev)
-
-static int panel_dpi_connect(struct omap_dss_device *src,
-			     struct omap_dss_device *dst)
-{
-	return 0;
-}
-
-static void panel_dpi_disconnect(struct omap_dss_device *src,
-				 struct omap_dss_device *dst)
-{
-}
-
-static int panel_dpi_enable(struct omap_dss_device *dssdev)
-{
-	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
-	int r;
-
-	if (!omapdss_device_is_connected(dssdev))
-		return -ENODEV;
-
-	if (omapdss_device_is_enabled(dssdev))
-		return 0;
-
-	r = src->ops->enable(src);
-	if (r)
-		return r;
-
-	r = regulator_enable(ddata->vcc_supply);
-	if (r) {
-		src->ops->disable(src);
-		return r;
-	}
-
-	gpiod_set_value_cansleep(ddata->enable_gpio, 1);
-	backlight_enable(ddata->backlight);
-
-	dssdev->state = OMAP_DSS_DISPLAY_ACTIVE;
-
-	return 0;
-}
-
-static void panel_dpi_disable(struct omap_dss_device *dssdev)
-{
-	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
-
-	if (!omapdss_device_is_enabled(dssdev))
-		return;
-
-	backlight_disable(ddata->backlight);
-
-	gpiod_set_value_cansleep(ddata->enable_gpio, 0);
-	regulator_disable(ddata->vcc_supply);
-
-	src->ops->disable(src);
-
-	dssdev->state = OMAP_DSS_DISPLAY_DISABLED;
-}
-
-static void panel_dpi_get_timings(struct omap_dss_device *dssdev,
-				  struct videomode *vm)
-{
-	struct panel_drv_data *ddata = to_panel_data(dssdev);
-
-	*vm = ddata->vm;
-}
-
-static const struct omap_dss_device_ops panel_dpi_ops = {
-	.connect	= panel_dpi_connect,
-	.disconnect	= panel_dpi_disconnect,
-
-	.enable		= panel_dpi_enable,
-	.disable	= panel_dpi_disable,
-
-	.get_timings	= panel_dpi_get_timings,
-};
-
-static int panel_dpi_probe_of(struct platform_device *pdev)
-{
-	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
-	struct device_node *node = pdev->dev.of_node;
-	int r;
-	struct display_timing timing;
-	struct gpio_desc *gpio;
-
-	gpio = devm_gpiod_get_optional(&pdev->dev, "enable", GPIOD_OUT_LOW);
-	if (IS_ERR(gpio))
-		return PTR_ERR(gpio);
-
-	ddata->enable_gpio = gpio;
-
-	/*
-	 * Many different panels are supported by this driver and there are
-	 * probably very different needs for their reset pins in regards to
-	 * timing and order relative to the enable gpio. So for now it's just
-	 * ensured that the reset line isn't active.
-	 */
-	gpio = devm_gpiod_get_optional(&pdev->dev, "reset", GPIOD_OUT_LOW);
-	if (IS_ERR(gpio))
-		return PTR_ERR(gpio);
-
-	ddata->vcc_supply = devm_regulator_get(&pdev->dev, "vcc");
-	if (IS_ERR(ddata->vcc_supply))
-		return PTR_ERR(ddata->vcc_supply);
-
-	ddata->backlight = devm_of_find_backlight(&pdev->dev);
-
-	if (IS_ERR(ddata->backlight))
-		return PTR_ERR(ddata->backlight);
-
-	r = of_get_display_timing(node, "panel-timing", &timing);
-	if (r) {
-		dev_err(&pdev->dev, "failed to get video timing\n");
-		return r;
-	}
-
-	videomode_from_timing(&timing, &ddata->vm);
-
-	return 0;
-}
-
-static int panel_dpi_probe(struct platform_device *pdev)
-{
-	struct panel_drv_data *ddata;
-	struct omap_dss_device *dssdev;
-	int r;
-
-	ddata = devm_kzalloc(&pdev->dev, sizeof(*ddata), GFP_KERNEL);
-	if (ddata == NULL)
-		return -ENOMEM;
-
-	platform_set_drvdata(pdev, ddata);
-
-	r = panel_dpi_probe_of(pdev);
-	if (r)
-		return r;
-
-	dssdev = &ddata->dssdev;
-	dssdev->dev = &pdev->dev;
-	dssdev->ops = &panel_dpi_ops;
-	dssdev->type = OMAP_DISPLAY_TYPE_DPI;
-	dssdev->owner = THIS_MODULE;
-	dssdev->of_ports = BIT(0);
-	drm_bus_flags_from_videomode(&ddata->vm, &dssdev->bus_flags);
-
-	omapdss_display_init(dssdev);
-	omapdss_device_register(dssdev);
-
-	return 0;
-}
-
-static int __exit panel_dpi_remove(struct platform_device *pdev)
-{
-	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
-	struct omap_dss_device *dssdev = &ddata->dssdev;
-
-	omapdss_device_unregister(dssdev);
-
-	panel_dpi_disable(dssdev);
-
-	return 0;
-}
-
-static const struct of_device_id panel_dpi_of_match[] = {
-	{ .compatible = "omapdss,panel-dpi", },
-	{},
-};
-
-MODULE_DEVICE_TABLE(of, panel_dpi_of_match);
-
-static struct platform_driver panel_dpi_driver = {
-	.probe = panel_dpi_probe,
-	.remove = __exit_p(panel_dpi_remove),
-	.driver = {
-		.name = "panel-dpi",
-		.of_match_table = panel_dpi_of_match,
-		.suppress_bind_attrs = true,
-	},
-};
-
-module_platform_driver(panel_dpi_driver);
-
-MODULE_AUTHOR("Tomi Valkeinen <tomi.valkeinen@ti.com>");
-MODULE_DESCRIPTION("Generic MIPI DPI Panel Driver");
-MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/omapdrm/displays/panel-dsi-cm.c b/drivers/gpu/drm/omapdrm/displays/panel-dsi-cm.c
index 29692a5217c5..741a5e324767 100644
--- a/drivers/gpu/drm/omapdrm/displays/panel-dsi-cm.c
+++ b/drivers/gpu/drm/omapdrm/displays/panel-dsi-cm.c
@@ -24,6 +24,8 @@
 #include <linux/of_device.h>
 #include <linux/regulator/consumer.h>
 
+#include <drm/drm_connector.h>
+
 #include <video/mipi_display.h>
 #include <video/of_display_timing.h>
 
@@ -41,6 +43,7 @@
 
 struct panel_drv_data {
 	struct omap_dss_device dssdev;
+	struct omap_dss_device *src;
 
 	struct videomode vm;
 
@@ -141,7 +144,7 @@ static void hw_guard_wait(struct panel_drv_data *ddata)
 
 static int dsicm_dcs_read_1(struct panel_drv_data *ddata, u8 dcs_cmd, u8 *data)
 {
-	struct omap_dss_device *src = ddata->dssdev.src;
+	struct omap_dss_device *src = ddata->src;
 	int r;
 	u8 buf[1];
 
@@ -157,14 +160,14 @@ static int dsicm_dcs_read_1(struct panel_drv_data *ddata, u8 dcs_cmd, u8 *data)
 
 static int dsicm_dcs_write_0(struct panel_drv_data *ddata, u8 dcs_cmd)
 {
-	struct omap_dss_device *src = ddata->dssdev.src;
+	struct omap_dss_device *src = ddata->src;
 
 	return src->ops->dsi.dcs_write(src, ddata->channel, &dcs_cmd, 1);
 }
 
 static int dsicm_dcs_write_1(struct panel_drv_data *ddata, u8 dcs_cmd, u8 param)
 {
-	struct omap_dss_device *src = ddata->dssdev.src;
+	struct omap_dss_device *src = ddata->src;
 	u8 buf[2] = { dcs_cmd, param };
 
 	return src->ops->dsi.dcs_write(src, ddata->channel, buf, 2);
@@ -173,7 +176,7 @@ static int dsicm_dcs_write_1(struct panel_drv_data *ddata, u8 dcs_cmd, u8 param)
 static int dsicm_sleep_in(struct panel_drv_data *ddata)
 
 {
-	struct omap_dss_device *src = ddata->dssdev.src;
+	struct omap_dss_device *src = ddata->src;
 	u8 cmd;
 	int r;
 
@@ -228,7 +231,7 @@ static int dsicm_get_id(struct panel_drv_data *ddata, u8 *id1, u8 *id2, u8 *id3)
 static int dsicm_set_update_window(struct panel_drv_data *ddata,
 		u16 x, u16 y, u16 w, u16 h)
 {
-	struct omap_dss_device *src = ddata->dssdev.src;
+	struct omap_dss_device *src = ddata->src;
 	int r;
 	u16 x1 = x;
 	u16 x2 = x + w - 1;
@@ -275,7 +278,7 @@ static void dsicm_cancel_ulps_work(struct panel_drv_data *ddata)
 
 static int dsicm_enter_ulps(struct panel_drv_data *ddata)
 {
-	struct omap_dss_device *src = ddata->dssdev.src;
+	struct omap_dss_device *src = ddata->src;
 	int r;
 
 	if (ddata->ulps_enabled)
@@ -309,18 +312,13 @@ err:
 
 static int dsicm_exit_ulps(struct panel_drv_data *ddata)
 {
-	struct omap_dss_device *src = ddata->dssdev.src;
+	struct omap_dss_device *src = ddata->src;
 	int r;
 
 	if (!ddata->ulps_enabled)
 		return 0;
 
-	r = src->ops->enable(src);
-	if (r) {
-		dev_err(&ddata->pdev->dev, "failed to enable DSI\n");
-		goto err1;
-	}
-
+	src->ops->enable(src);
 	src->ops->dsi.enable_hs(src, ddata->channel, true);
 
 	r = _dsicm_enable_te(ddata, true);
@@ -347,7 +345,7 @@ err2:
 			enable_irq(gpiod_to_irq(ddata->ext_te_gpio));
 		ddata->ulps_enabled = false;
 	}
-err1:
+
 	dsicm_queue_ulps_work(ddata);
 
 	return r;
@@ -366,7 +364,7 @@ static int dsicm_wake_up(struct panel_drv_data *ddata)
 static int dsicm_bl_update_status(struct backlight_device *dev)
 {
 	struct panel_drv_data *ddata = dev_get_drvdata(&dev->dev);
-	struct omap_dss_device *src = ddata->dssdev.src;
+	struct omap_dss_device *src = ddata->src;
 	int r = 0;
 	int level;
 
@@ -414,7 +412,7 @@ static ssize_t dsicm_num_errors_show(struct device *dev,
 {
 	struct platform_device *pdev = to_platform_device(dev);
 	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
-	struct omap_dss_device *src = ddata->dssdev.src;
+	struct omap_dss_device *src = ddata->src;
 	u8 errors = 0;
 	int r;
 
@@ -446,7 +444,7 @@ static ssize_t dsicm_hw_revision_show(struct device *dev,
 {
 	struct platform_device *pdev = to_platform_device(dev);
 	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
-	struct omap_dss_device *src = ddata->dssdev.src;
+	struct omap_dss_device *src = ddata->src;
 	u8 id1, id2, id3;
 	int r;
 
@@ -478,7 +476,7 @@ static ssize_t dsicm_store_ulps(struct device *dev,
 {
 	struct platform_device *pdev = to_platform_device(dev);
 	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
-	struct omap_dss_device *src = ddata->dssdev.src;
+	struct omap_dss_device *src = ddata->src;
 	unsigned long t;
 	int r;
 
@@ -528,7 +526,7 @@ static ssize_t dsicm_store_ulps_timeout(struct device *dev,
 {
 	struct platform_device *pdev = to_platform_device(dev);
 	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
-	struct omap_dss_device *src = ddata->dssdev.src;
+	struct omap_dss_device *src = ddata->src;
 	unsigned long t;
 	int r;
 
@@ -603,7 +601,7 @@ static void dsicm_hw_reset(struct panel_drv_data *ddata)
 
 static int dsicm_power_on(struct panel_drv_data *ddata)
 {
-	struct omap_dss_device *src = ddata->dssdev.src;
+	struct omap_dss_device *src = ddata->src;
 	u8 id1, id2, id3;
 	int r;
 	struct omap_dss_dsi_config dsi_config = {
@@ -649,11 +647,7 @@ static int dsicm_power_on(struct panel_drv_data *ddata)
 		goto err_vddi;
 	}
 
-	r = src->ops->enable(src);
-	if (r) {
-		dev_err(&ddata->pdev->dev, "failed to enable DSI\n");
-		goto err_vddi;
-	}
+	src->ops->enable(src);
 
 	dsicm_hw_reset(ddata);
 
@@ -722,7 +716,7 @@ err_vpnl:
 
 static void dsicm_power_off(struct panel_drv_data *ddata)
 {
-	struct omap_dss_device *src = ddata->dssdev.src;
+	struct omap_dss_device *src = ddata->src;
 	int r;
 
 	src->ops->dsi.disable_video_output(src, ddata->channel);
@@ -776,6 +770,7 @@ static int dsicm_connect(struct omap_dss_device *src,
 		return r;
 	}
 
+	ddata->src = src;
 	return 0;
 }
 
@@ -785,28 +780,17 @@ static void dsicm_disconnect(struct omap_dss_device *src,
 	struct panel_drv_data *ddata = to_panel_data(dst);
 
 	src->ops->dsi.release_vc(src, ddata->channel);
+	ddata->src = NULL;
 }
 
-static int dsicm_enable(struct omap_dss_device *dssdev)
+static void dsicm_enable(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
+	struct omap_dss_device *src = ddata->src;
 	int r;
 
-	dev_dbg(&ddata->pdev->dev, "enable\n");
-
 	mutex_lock(&ddata->lock);
 
-	if (!omapdss_device_is_connected(dssdev)) {
-		r = -ENODEV;
-		goto err;
-	}
-
-	if (omapdss_device_is_enabled(dssdev)) {
-		r = 0;
-		goto err;
-	}
-
 	src->ops->dsi.bus_lock(src);
 
 	r = dsicm_power_on(ddata);
@@ -816,27 +800,22 @@ static int dsicm_enable(struct omap_dss_device *dssdev)
 	if (r)
 		goto err;
 
-	dssdev->state = OMAP_DSS_DISPLAY_ACTIVE;
-
 	mutex_unlock(&ddata->lock);
 
 	dsicm_bl_power(ddata, true);
 
-	return 0;
+	return;
 err:
-	dev_dbg(&ddata->pdev->dev, "enable failed\n");
+	dev_dbg(&ddata->pdev->dev, "enable failed (%d)\n", r);
 	mutex_unlock(&ddata->lock);
-	return r;
 }
 
 static void dsicm_disable(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
+	struct omap_dss_device *src = ddata->src;
 	int r;
 
-	dev_dbg(&ddata->pdev->dev, "disable\n");
-
 	dsicm_bl_power(ddata, false);
 
 	mutex_lock(&ddata->lock);
@@ -845,23 +824,19 @@ static void dsicm_disable(struct omap_dss_device *dssdev)
 
 	src->ops->dsi.bus_lock(src);
 
-	if (omapdss_device_is_enabled(dssdev)) {
-		r = dsicm_wake_up(ddata);
-		if (!r)
-			dsicm_power_off(ddata);
-	}
+	r = dsicm_wake_up(ddata);
+	if (!r)
+		dsicm_power_off(ddata);
 
 	src->ops->dsi.bus_unlock(src);
 
-	dssdev->state = OMAP_DSS_DISPLAY_DISABLED;
-
 	mutex_unlock(&ddata->lock);
 }
 
 static void dsicm_framedone_cb(int err, void *data)
 {
 	struct panel_drv_data *ddata = data;
-	struct omap_dss_device *src = ddata->dssdev.src;
+	struct omap_dss_device *src = ddata->src;
 
 	dev_dbg(&ddata->pdev->dev, "framedone, err %d\n", err);
 	src->ops->dsi.bus_unlock(src);
@@ -870,7 +845,7 @@ static void dsicm_framedone_cb(int err, void *data)
 static irqreturn_t dsicm_te_isr(int irq, void *data)
 {
 	struct panel_drv_data *ddata = data;
-	struct omap_dss_device *src = ddata->dssdev.src;
+	struct omap_dss_device *src = ddata->src;
 	int old;
 	int r;
 
@@ -896,7 +871,7 @@ static void dsicm_te_timeout_work_callback(struct work_struct *work)
 {
 	struct panel_drv_data *ddata = container_of(work, struct panel_drv_data,
 					te_timeout_work.work);
-	struct omap_dss_device *src = ddata->dssdev.src;
+	struct omap_dss_device *src = ddata->src;
 
 	dev_err(&ddata->pdev->dev, "TE not received for 250ms!\n");
 
@@ -908,7 +883,7 @@ static int dsicm_update(struct omap_dss_device *dssdev,
 				    u16 x, u16 y, u16 w, u16 h)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
+	struct omap_dss_device *src = ddata->src;
 	int r;
 
 	dev_dbg(&ddata->pdev->dev, "update %d, %d, %d x %d\n", x, y, w, h);
@@ -954,7 +929,7 @@ err:
 static int dsicm_sync(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
+	struct omap_dss_device *src = ddata->src;
 
 	dev_dbg(&ddata->pdev->dev, "sync\n");
 
@@ -970,7 +945,7 @@ static int dsicm_sync(struct omap_dss_device *dssdev)
 
 static int _dsicm_enable_te(struct panel_drv_data *ddata, bool enable)
 {
-	struct omap_dss_device *src = ddata->dssdev.src;
+	struct omap_dss_device *src = ddata->src;
 	int r;
 
 	if (enable)
@@ -990,7 +965,7 @@ static int _dsicm_enable_te(struct panel_drv_data *ddata, bool enable)
 static int dsicm_enable_te(struct omap_dss_device *dssdev, bool enable)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
+	struct omap_dss_device *src = ddata->src;
 	int r;
 
 	mutex_lock(&ddata->lock);
@@ -1041,7 +1016,7 @@ static int dsicm_memory_read(struct omap_dss_device *dssdev,
 		u16 x, u16 y, u16 w, u16 h)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
+	struct omap_dss_device *src = ddata->src;
 	int r;
 	int first = 1;
 	int plen;
@@ -1123,7 +1098,7 @@ static void dsicm_ulps_work(struct work_struct *work)
 	struct panel_drv_data *ddata = container_of(work, struct panel_drv_data,
 			ulps_work.work);
 	struct omap_dss_device *dssdev = &ddata->dssdev;
-	struct omap_dss_device *src = dssdev->src;
+	struct omap_dss_device *src = ddata->src;
 
 	mutex_lock(&ddata->lock);
 
@@ -1140,29 +1115,32 @@ static void dsicm_ulps_work(struct work_struct *work)
 	mutex_unlock(&ddata->lock);
 }
 
-static void dsicm_get_timings(struct omap_dss_device *dssdev,
-			      struct videomode *vm)
+static int dsicm_get_modes(struct omap_dss_device *dssdev,
+			   struct drm_connector *connector)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
 
-	*vm = ddata->vm;
+	connector->display_info.width_mm = ddata->width_mm;
+	connector->display_info.height_mm = ddata->height_mm;
+
+	return omapdss_display_get_modes(connector, &ddata->vm);
 }
 
 static int dsicm_check_timings(struct omap_dss_device *dssdev,
-			       struct videomode *vm)
+			       struct drm_display_mode *mode)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
 	int ret = 0;
 
-	if (vm->hactive != ddata->vm.hactive)
+	if (mode->hdisplay != ddata->vm.hactive)
 		ret = -EINVAL;
 
-	if (vm->vactive != ddata->vm.vactive)
+	if (mode->vdisplay != ddata->vm.vactive)
 		ret = -EINVAL;
 
 	if (ret) {
 		dev_warn(dssdev->dev, "wrong resolution: %d x %d",
-			 vm->hactive, vm->vactive);
+			 mode->hdisplay, mode->vdisplay);
 		dev_warn(dssdev->dev, "panel resolution: %d x %d",
 			 ddata->vm.hactive, ddata->vm.vactive);
 	}
@@ -1170,15 +1148,6 @@ static int dsicm_check_timings(struct omap_dss_device *dssdev,
 	return ret;
 }
 
-static void dsicm_get_size(struct omap_dss_device *dssdev,
-			  unsigned int *width, unsigned int *height)
-{
-	struct panel_drv_data *ddata = to_panel_data(dssdev);
-
-	*width = ddata->width_mm;
-	*height = ddata->height_mm;
-}
-
 static const struct omap_dss_device_ops dsicm_ops = {
 	.connect	= dsicm_connect,
 	.disconnect	= dsicm_disconnect,
@@ -1186,7 +1155,7 @@ static const struct omap_dss_device_ops dsicm_ops = {
 	.enable		= dsicm_enable,
 	.disable	= dsicm_disable,
 
-	.get_timings	= dsicm_get_timings,
+	.get_modes	= dsicm_get_modes,
 	.check_timings	= dsicm_check_timings,
 };
 
@@ -1194,8 +1163,6 @@ static const struct omap_dss_driver dsicm_dss_driver = {
 	.update		= dsicm_update,
 	.sync		= dsicm_sync,
 
-	.get_size	= dsicm_get_size,
-
 	.enable_te	= dsicm_enable_te,
 	.get_te		= dsicm_get_te,
 
@@ -1305,8 +1272,10 @@ static int dsicm_probe(struct platform_device *pdev)
 	dssdev->ops = &dsicm_ops;
 	dssdev->driver = &dsicm_dss_driver;
 	dssdev->type = OMAP_DISPLAY_TYPE_DSI;
+	dssdev->display = true;
 	dssdev->owner = THIS_MODULE;
 	dssdev->of_ports = BIT(0);
+	dssdev->ops_flags = OMAP_DSS_DEVICE_OP_MODES;
 
 	dssdev->caps = OMAP_DSS_DISPLAY_CAP_MANUAL_UPDATE |
 		OMAP_DSS_DISPLAY_CAP_TEAR_ELIM;
@@ -1385,8 +1354,9 @@ static int __exit dsicm_remove(struct platform_device *pdev)
 
 	omapdss_device_unregister(dssdev);
 
-	dsicm_disable(dssdev);
-	omapdss_device_disconnect(dssdev->src, dssdev);
+	if (omapdss_device_is_enabled(dssdev))
+		dsicm_disable(dssdev);
+	omapdss_device_disconnect(ddata->src, dssdev);
 
 	sysfs_remove_group(&pdev->dev.kobj, &dsicm_attr_group);
 
diff --git a/drivers/gpu/drm/omapdrm/displays/panel-lgphilips-lb035q02.c b/drivers/gpu/drm/omapdrm/displays/panel-lgphilips-lb035q02.c
index f6ef8ff964dd..99f2350d462c 100644
--- a/drivers/gpu/drm/omapdrm/displays/panel-lgphilips-lb035q02.c
+++ b/drivers/gpu/drm/omapdrm/displays/panel-lgphilips-lb035q02.c
@@ -123,52 +123,28 @@ static void lb035q02_disconnect(struct omap_dss_device *src,
 {
 }
 
-static int lb035q02_enable(struct omap_dss_device *dssdev)
+static void lb035q02_enable(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
-	int r;
-
-	if (!omapdss_device_is_connected(dssdev))
-		return -ENODEV;
-
-	if (omapdss_device_is_enabled(dssdev))
-		return 0;
-
-	r = src->ops->enable(src);
-	if (r)
-		return r;
 
 	if (ddata->enable_gpio)
 		gpiod_set_value_cansleep(ddata->enable_gpio, 1);
-
-	dssdev->state = OMAP_DSS_DISPLAY_ACTIVE;
-
-	return 0;
 }
 
 static void lb035q02_disable(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
-
-	if (!omapdss_device_is_enabled(dssdev))
-		return;
 
 	if (ddata->enable_gpio)
 		gpiod_set_value_cansleep(ddata->enable_gpio, 0);
-
-	src->ops->disable(src);
-
-	dssdev->state = OMAP_DSS_DISPLAY_DISABLED;
 }
 
-static void lb035q02_get_timings(struct omap_dss_device *dssdev,
-				 struct videomode *vm)
+static int lb035q02_get_modes(struct omap_dss_device *dssdev,
+			      struct drm_connector *connector)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
 
-	*vm = ddata->vm;
+	return omapdss_display_get_modes(connector, &ddata->vm);
 }
 
 static const struct omap_dss_device_ops lb035q02_ops = {
@@ -178,7 +154,7 @@ static const struct omap_dss_device_ops lb035q02_ops = {
 	.enable		= lb035q02_enable,
 	.disable	= lb035q02_disable,
 
-	.get_timings	= lb035q02_get_timings,
+	.get_modes	= lb035q02_get_modes,
 };
 
 static int lb035q02_probe_of(struct spi_device *spi)
@@ -221,16 +197,19 @@ static int lb035q02_panel_spi_probe(struct spi_device *spi)
 	dssdev->dev = &spi->dev;
 	dssdev->ops = &lb035q02_ops;
 	dssdev->type = OMAP_DISPLAY_TYPE_DPI;
+	dssdev->display = true;
 	dssdev->owner = THIS_MODULE;
 	dssdev->of_ports = BIT(0);
+	dssdev->ops_flags = OMAP_DSS_DEVICE_OP_MODES;
 
 	/*
 	 * Note: According to the panel documentation:
 	 * DE is active LOW
 	 * DATA needs to be driven on the FALLING edge
 	 */
-	dssdev->bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_SYNC_NEGEDGE
-			  | DRM_BUS_FLAG_PIXDATA_POSEDGE;
+	dssdev->bus_flags = DRM_BUS_FLAG_DE_HIGH
+			  | DRM_BUS_FLAG_SYNC_DRIVE_NEGEDGE
+			  | DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE;
 
 	omapdss_display_init(dssdev);
 	omapdss_device_register(dssdev);
diff --git a/drivers/gpu/drm/omapdrm/displays/panel-nec-nl8048hl11.c b/drivers/gpu/drm/omapdrm/displays/panel-nec-nl8048hl11.c
index f445de6369f7..c2409815a204 100644
--- a/drivers/gpu/drm/omapdrm/displays/panel-nec-nl8048hl11.c
+++ b/drivers/gpu/drm/omapdrm/displays/panel-nec-nl8048hl11.c
@@ -118,50 +118,26 @@ static void nec_8048_disconnect(struct omap_dss_device *src,
 {
 }
 
-static int nec_8048_enable(struct omap_dss_device *dssdev)
+static void nec_8048_enable(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
-	int r;
-
-	if (!omapdss_device_is_connected(dssdev))
-		return -ENODEV;
-
-	if (omapdss_device_is_enabled(dssdev))
-		return 0;
-
-	r = src->ops->enable(src);
-	if (r)
-		return r;
 
 	gpiod_set_value_cansleep(ddata->res_gpio, 1);
-
-	dssdev->state = OMAP_DSS_DISPLAY_ACTIVE;
-
-	return 0;
 }
 
 static void nec_8048_disable(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
-
-	if (!omapdss_device_is_enabled(dssdev))
-		return;
 
 	gpiod_set_value_cansleep(ddata->res_gpio, 0);
-
-	src->ops->disable(src);
-
-	dssdev->state = OMAP_DSS_DISPLAY_DISABLED;
 }
 
-static void nec_8048_get_timings(struct omap_dss_device *dssdev,
-				 struct videomode *vm)
+static int nec_8048_get_modes(struct omap_dss_device *dssdev,
+			      struct drm_connector *connector)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
 
-	*vm = ddata->vm;
+	return omapdss_display_get_modes(connector, &ddata->vm);
 }
 
 static const struct omap_dss_device_ops nec_8048_ops = {
@@ -171,7 +147,7 @@ static const struct omap_dss_device_ops nec_8048_ops = {
 	.enable		= nec_8048_enable,
 	.disable	= nec_8048_disable,
 
-	.get_timings	= nec_8048_get_timings,
+	.get_modes	= nec_8048_get_modes,
 };
 
 static int nec_8048_probe(struct spi_device *spi)
@@ -216,10 +192,13 @@ static int nec_8048_probe(struct spi_device *spi)
 	dssdev->dev = &spi->dev;
 	dssdev->ops = &nec_8048_ops;
 	dssdev->type = OMAP_DISPLAY_TYPE_DPI;
+	dssdev->display = true;
 	dssdev->owner = THIS_MODULE;
 	dssdev->of_ports = BIT(0);
-	dssdev->bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_SYNC_POSEDGE
-			  | DRM_BUS_FLAG_PIXDATA_POSEDGE;
+	dssdev->ops_flags = OMAP_DSS_DEVICE_OP_MODES;
+	dssdev->bus_flags = DRM_BUS_FLAG_DE_HIGH
+			  | DRM_BUS_FLAG_SYNC_DRIVE_POSEDGE
+			  | DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE;
 
 	omapdss_display_init(dssdev);
 	omapdss_device_register(dssdev);
diff --git a/drivers/gpu/drm/omapdrm/displays/panel-sharp-ls037v7dw01.c b/drivers/gpu/drm/omapdrm/displays/panel-sharp-ls037v7dw01.c
index 64b1369cb274..9c545de430f6 100644
--- a/drivers/gpu/drm/omapdrm/displays/panel-sharp-ls037v7dw01.c
+++ b/drivers/gpu/drm/omapdrm/displays/panel-sharp-ls037v7dw01.c
@@ -62,29 +62,22 @@ static void sharp_ls_disconnect(struct omap_dss_device *src,
 {
 }
 
-static int sharp_ls_enable(struct omap_dss_device *dssdev)
+static void sharp_ls_pre_enable(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
 	int r;
 
-	if (!omapdss_device_is_connected(dssdev))
-		return -ENODEV;
-
-	if (omapdss_device_is_enabled(dssdev))
-		return 0;
-
 	if (ddata->vcc) {
 		r = regulator_enable(ddata->vcc);
-		if (r != 0)
-			return r;
+		if (r)
+			dev_err(dssdev->dev, "%s: failed to enable regulator\n",
+				__func__);
 	}
+}
 
-	r = src->ops->enable(src);
-	if (r) {
-		regulator_disable(ddata->vcc);
-		return r;
-	}
+static void sharp_ls_enable(struct omap_dss_device *dssdev)
+{
+	struct panel_drv_data *ddata = to_panel_data(dssdev);
 
 	/* wait couple of vsyncs until enabling the LCD */
 	msleep(50);
@@ -94,19 +87,11 @@ static int sharp_ls_enable(struct omap_dss_device *dssdev)
 
 	if (ddata->ini_gpio)
 		gpiod_set_value_cansleep(ddata->ini_gpio, 1);
-
-	dssdev->state = OMAP_DSS_DISPLAY_ACTIVE;
-
-	return 0;
 }
 
 static void sharp_ls_disable(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
-
-	if (!omapdss_device_is_enabled(dssdev))
-		return;
 
 	if (ddata->ini_gpio)
 		gpiod_set_value_cansleep(ddata->ini_gpio, 0);
@@ -115,33 +100,35 @@ static void sharp_ls_disable(struct omap_dss_device *dssdev)
 		gpiod_set_value_cansleep(ddata->resb_gpio, 0);
 
 	/* wait at least 5 vsyncs after disabling the LCD */
-
 	msleep(100);
+}
 
-	src->ops->disable(src);
+static void sharp_ls_post_disable(struct omap_dss_device *dssdev)
+{
+	struct panel_drv_data *ddata = to_panel_data(dssdev);
 
 	if (ddata->vcc)
 		regulator_disable(ddata->vcc);
-
-	dssdev->state = OMAP_DSS_DISPLAY_DISABLED;
 }
 
-static void sharp_ls_get_timings(struct omap_dss_device *dssdev,
-				 struct videomode *vm)
+static int sharp_ls_get_modes(struct omap_dss_device *dssdev,
+			      struct drm_connector *connector)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
 
-	*vm = ddata->vm;
+	return omapdss_display_get_modes(connector, &ddata->vm);
 }
 
 static const struct omap_dss_device_ops sharp_ls_ops = {
 	.connect	= sharp_ls_connect,
 	.disconnect	= sharp_ls_disconnect,
 
+	.pre_enable	= sharp_ls_pre_enable,
 	.enable		= sharp_ls_enable,
 	.disable	= sharp_ls_disable,
+	.post_disable	= sharp_ls_post_disable,
 
-	.get_timings	= sharp_ls_get_timings,
+	.get_modes	= sharp_ls_get_modes,
 };
 
 static  int sharp_ls_get_gpio_of(struct device *dev, int index, int val,
@@ -220,15 +207,18 @@ static int sharp_ls_probe(struct platform_device *pdev)
 	dssdev->dev = &pdev->dev;
 	dssdev->ops = &sharp_ls_ops;
 	dssdev->type = OMAP_DISPLAY_TYPE_DPI;
+	dssdev->display = true;
 	dssdev->owner = THIS_MODULE;
 	dssdev->of_ports = BIT(0);
+	dssdev->ops_flags = OMAP_DSS_DEVICE_OP_MODES;
 
 	/*
 	 * Note: According to the panel documentation:
 	 * DATA needs to be driven on the FALLING edge
 	 */
-	dssdev->bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_SYNC_NEGEDGE
-			  | DRM_BUS_FLAG_PIXDATA_POSEDGE;
+	dssdev->bus_flags = DRM_BUS_FLAG_DE_HIGH
+			  | DRM_BUS_FLAG_SYNC_DRIVE_NEGEDGE
+			  | DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE;
 
 	omapdss_display_init(dssdev);
 	omapdss_device_register(dssdev);
@@ -243,7 +233,10 @@ static int __exit sharp_ls_remove(struct platform_device *pdev)
 
 	omapdss_device_unregister(dssdev);
 
-	sharp_ls_disable(dssdev);
+	if (omapdss_device_is_enabled(dssdev)) {
+		sharp_ls_disable(dssdev);
+		sharp_ls_post_disable(dssdev);
+	}
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/omapdrm/displays/panel-sony-acx565akm.c b/drivers/gpu/drm/omapdrm/displays/panel-sony-acx565akm.c
index e04663856b31..2038def14ba1 100644
--- a/drivers/gpu/drm/omapdrm/displays/panel-sony-acx565akm.c
+++ b/drivers/gpu/drm/omapdrm/displays/panel-sony-acx565akm.c
@@ -516,17 +516,9 @@ static void acx565akm_disconnect(struct omap_dss_device *src,
 static int acx565akm_panel_power_on(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
-	int r;
 
 	dev_dbg(&ddata->spi->dev, "%s\n", __func__);
 
-	r = src->ops->enable(src);
-	if (r) {
-		pr_err("%s sdi enable failed\n", __func__);
-		return r;
-	}
-
 	/*FIXME tweak me */
 	msleep(50);
 
@@ -562,7 +554,6 @@ static int acx565akm_panel_power_on(struct omap_dss_device *dssdev)
 static void acx565akm_panel_power_off(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
 
 	dev_dbg(dssdev->dev, "%s\n", __func__);
 
@@ -585,56 +576,32 @@ static void acx565akm_panel_power_off(struct omap_dss_device *dssdev)
 
 	/* FIXME need to tweak this delay */
 	msleep(100);
-
-	src->ops->disable(src);
 }
 
-static int acx565akm_enable(struct omap_dss_device *dssdev)
+static void acx565akm_enable(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	int r;
-
-	dev_dbg(dssdev->dev, "%s\n", __func__);
-
-	if (!omapdss_device_is_connected(dssdev))
-		return -ENODEV;
-
-	if (omapdss_device_is_enabled(dssdev))
-		return 0;
 
 	mutex_lock(&ddata->mutex);
-	r = acx565akm_panel_power_on(dssdev);
+	acx565akm_panel_power_on(dssdev);
 	mutex_unlock(&ddata->mutex);
-	if (r)
-		return r;
-
-	dssdev->state = OMAP_DSS_DISPLAY_ACTIVE;
-
-	return 0;
 }
 
 static void acx565akm_disable(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
 
-	dev_dbg(dssdev->dev, "%s\n", __func__);
-
-	if (!omapdss_device_is_enabled(dssdev))
-		return;
-
 	mutex_lock(&ddata->mutex);
 	acx565akm_panel_power_off(dssdev);
 	mutex_unlock(&ddata->mutex);
-
-	dssdev->state = OMAP_DSS_DISPLAY_DISABLED;
 }
 
-static void acx565akm_get_timings(struct omap_dss_device *dssdev,
-				  struct videomode *vm)
+static int acx565akm_get_modes(struct omap_dss_device *dssdev,
+			       struct drm_connector *connector)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
 
-	*vm = ddata->vm;
+	return omapdss_display_get_modes(connector, &ddata->vm);
 }
 
 static const struct omap_dss_device_ops acx565akm_ops = {
@@ -644,7 +611,7 @@ static const struct omap_dss_device_ops acx565akm_ops = {
 	.enable		= acx565akm_enable,
 	.disable	= acx565akm_disable,
 
-	.get_timings	= acx565akm_get_timings,
+	.get_modes	= acx565akm_get_modes,
 };
 
 static int acx565akm_probe(struct spi_device *spi)
@@ -739,10 +706,13 @@ static int acx565akm_probe(struct spi_device *spi)
 	dssdev->dev = &spi->dev;
 	dssdev->ops = &acx565akm_ops;
 	dssdev->type = OMAP_DISPLAY_TYPE_SDI;
+	dssdev->display = true;
 	dssdev->owner = THIS_MODULE;
 	dssdev->of_ports = BIT(0);
-	dssdev->bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_SYNC_NEGEDGE
-			  | DRM_BUS_FLAG_PIXDATA_POSEDGE;
+	dssdev->ops_flags = OMAP_DSS_DEVICE_OP_MODES;
+	dssdev->bus_flags = DRM_BUS_FLAG_DE_HIGH
+			  | DRM_BUS_FLAG_SYNC_DRIVE_NEGEDGE
+			  | DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE;
 
 	omapdss_display_init(dssdev);
 	omapdss_device_register(dssdev);
@@ -766,7 +736,8 @@ static int acx565akm_remove(struct spi_device *spi)
 
 	omapdss_device_unregister(dssdev);
 
-	acx565akm_disable(dssdev);
+	if (omapdss_device_is_enabled(dssdev))
+		acx565akm_disable(dssdev);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/omapdrm/displays/panel-tpo-td028ttec1.c b/drivers/gpu/drm/omapdrm/displays/panel-tpo-td028ttec1.c
index 7ddc8c574a61..2ad161e33106 100644
--- a/drivers/gpu/drm/omapdrm/displays/panel-tpo-td028ttec1.c
+++ b/drivers/gpu/drm/omapdrm/displays/panel-tpo-td028ttec1.c
@@ -35,6 +35,8 @@ struct panel_drv_data {
 
 	struct videomode vm;
 
+	struct backlight_device *backlight;
+
 	struct spi_device *spi_dev;
 };
 
@@ -169,24 +171,12 @@ static void td028ttec1_panel_disconnect(struct omap_dss_device *src,
 {
 }
 
-static int td028ttec1_panel_enable(struct omap_dss_device *dssdev)
+static void td028ttec1_panel_enable(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
-	int r;
-
-	if (!omapdss_device_is_connected(dssdev))
-		return -ENODEV;
-
-	if (omapdss_device_is_enabled(dssdev))
-		return 0;
-
-	r = src->ops->enable(src);
-	if (r)
-		return r;
+	int r = 0;
 
-	dev_dbg(dssdev->dev, "td028ttec1_panel_enable() - state %d\n",
-		dssdev->state);
+	dev_dbg(dssdev->dev, "%s: state %d\n", __func__, dssdev->state);
 
 	/* three times command zero */
 	r |= jbt_ret_write_0(ddata, 0x00);
@@ -197,8 +187,8 @@ static int td028ttec1_panel_enable(struct omap_dss_device *dssdev)
 	usleep_range(1000, 2000);
 
 	if (r) {
-		dev_warn(dssdev->dev, "transfer error\n");
-		goto transfer_err;
+		dev_warn(dssdev->dev, "%s: transfer error\n", __func__);
+		return;
 	}
 
 	/* deep standby out */
@@ -268,20 +258,17 @@ static int td028ttec1_panel_enable(struct omap_dss_device *dssdev)
 
 	r |= jbt_ret_write_0(ddata, JBT_REG_DISPLAY_ON);
 
-	dssdev->state = OMAP_DSS_DISPLAY_ACTIVE;
-
-transfer_err:
+	if (r)
+		dev_err(dssdev->dev, "%s: write error\n", __func__);
 
-	return r ? -EIO : 0;
+	backlight_enable(ddata->backlight);
 }
 
 static void td028ttec1_panel_disable(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
 
-	if (!omapdss_device_is_enabled(dssdev))
-		return;
+	backlight_disable(ddata->backlight);
 
 	dev_dbg(dssdev->dev, "td028ttec1_panel_disable()\n");
 
@@ -289,18 +276,14 @@ static void td028ttec1_panel_disable(struct omap_dss_device *dssdev)
 	jbt_reg_write_2(ddata, JBT_REG_OUTPUT_CONTROL, 0x8002);
 	jbt_ret_write_0(ddata, JBT_REG_SLEEP_IN);
 	jbt_reg_write_1(ddata, JBT_REG_POWER_ON_OFF, 0x00);
-
-	src->ops->disable(src);
-
-	dssdev->state = OMAP_DSS_DISPLAY_DISABLED;
 }
 
-static void td028ttec1_panel_get_timings(struct omap_dss_device *dssdev,
-					 struct videomode *vm)
+static int td028ttec1_panel_get_modes(struct omap_dss_device *dssdev,
+				      struct drm_connector *connector)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
 
-	*vm = ddata->vm;
+	return omapdss_display_get_modes(connector, &ddata->vm);
 }
 
 static const struct omap_dss_device_ops td028ttec1_ops = {
@@ -310,7 +293,7 @@ static const struct omap_dss_device_ops td028ttec1_ops = {
 	.enable		= td028ttec1_panel_enable,
 	.disable	= td028ttec1_panel_disable,
 
-	.get_timings	= td028ttec1_panel_get_timings,
+	.get_modes	= td028ttec1_panel_get_modes,
 };
 
 static int td028ttec1_panel_probe(struct spi_device *spi)
@@ -334,6 +317,10 @@ static int td028ttec1_panel_probe(struct spi_device *spi)
 	if (ddata == NULL)
 		return -ENOMEM;
 
+	ddata->backlight = devm_of_find_backlight(&spi->dev);
+	if (IS_ERR(ddata->backlight))
+		return PTR_ERR(ddata->backlight);
+
 	dev_set_drvdata(&spi->dev, ddata);
 
 	ddata->spi_dev = spi;
@@ -344,15 +331,18 @@ static int td028ttec1_panel_probe(struct spi_device *spi)
 	dssdev->dev = &spi->dev;
 	dssdev->ops = &td028ttec1_ops;
 	dssdev->type = OMAP_DISPLAY_TYPE_DPI;
+	dssdev->display = true;
 	dssdev->owner = THIS_MODULE;
 	dssdev->of_ports = BIT(0);
+	dssdev->ops_flags = OMAP_DSS_DEVICE_OP_MODES;
 
 	/*
 	 * Note: According to the panel documentation:
 	 * SYNC needs to be driven on the FALLING edge
 	 */
-	dssdev->bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_SYNC_POSEDGE
-			  | DRM_BUS_FLAG_PIXDATA_NEGEDGE;
+	dssdev->bus_flags = DRM_BUS_FLAG_DE_HIGH
+			  | DRM_BUS_FLAG_SYNC_DRIVE_POSEDGE
+			  | DRM_BUS_FLAG_PIXDATA_DRIVE_NEGEDGE;
 
 	omapdss_display_init(dssdev);
 	omapdss_device_register(dssdev);
diff --git a/drivers/gpu/drm/omapdrm/displays/panel-tpo-td043mtea1.c b/drivers/gpu/drm/omapdrm/displays/panel-tpo-td043mtea1.c
index 8440fcb744d9..0b692fc7e5ea 100644
--- a/drivers/gpu/drm/omapdrm/displays/panel-tpo-td043mtea1.c
+++ b/drivers/gpu/drm/omapdrm/displays/panel-tpo-td043mtea1.c
@@ -320,22 +320,11 @@ static void tpo_td043_disconnect(struct omap_dss_device *src,
 {
 }
 
-static int tpo_td043_enable(struct omap_dss_device *dssdev)
+static void tpo_td043_enable(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
 	int r;
 
-	if (!omapdss_device_is_connected(dssdev))
-		return -ENODEV;
-
-	if (omapdss_device_is_enabled(dssdev))
-		return 0;
-
-	r = src->ops->enable(src);
-	if (r)
-		return r;
-
 	/*
 	 * If we are resuming from system suspend, SPI clocks might not be
 	 * enabled yet, so we'll program the LCD from SPI PM resume callback.
@@ -343,38 +332,27 @@ static int tpo_td043_enable(struct omap_dss_device *dssdev)
 	if (!ddata->spi_suspended) {
 		r = tpo_td043_power_on(ddata);
 		if (r) {
-			src->ops->disable(src);
-			return r;
+			dev_err(&ddata->spi->dev, "%s: power on failed (%d)\n",
+				__func__, r);
+			return;
 		}
 	}
-
-	dssdev->state = OMAP_DSS_DISPLAY_ACTIVE;
-
-	return 0;
 }
 
 static void tpo_td043_disable(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
-	struct omap_dss_device *src = dssdev->src;
-
-	if (!omapdss_device_is_enabled(dssdev))
-		return;
-
-	src->ops->disable(src);
 
 	if (!ddata->spi_suspended)
 		tpo_td043_power_off(ddata);
-
-	dssdev->state = OMAP_DSS_DISPLAY_DISABLED;
 }
 
-static void tpo_td043_get_timings(struct omap_dss_device *dssdev,
-				  struct videomode *vm)
+static int tpo_td043_get_modes(struct omap_dss_device *dssdev,
+			       struct drm_connector *connector)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
 
-	*vm = ddata->vm;
+	return omapdss_display_get_modes(connector, &ddata->vm);
 }
 
 static const struct omap_dss_device_ops tpo_td043_ops = {
@@ -384,7 +362,7 @@ static const struct omap_dss_device_ops tpo_td043_ops = {
 	.enable		= tpo_td043_enable,
 	.disable	= tpo_td043_disable,
 
-	.get_timings	= tpo_td043_get_timings,
+	.get_modes	= tpo_td043_get_modes,
 };
 
 static int tpo_td043_probe(struct spi_device *spi)
@@ -442,15 +420,18 @@ static int tpo_td043_probe(struct spi_device *spi)
 	dssdev->dev = &spi->dev;
 	dssdev->ops = &tpo_td043_ops;
 	dssdev->type = OMAP_DISPLAY_TYPE_DPI;
+	dssdev->display = true;
 	dssdev->owner = THIS_MODULE;
 	dssdev->of_ports = BIT(0);
+	dssdev->ops_flags = OMAP_DSS_DEVICE_OP_MODES;
 
 	/*
 	 * Note: According to the panel documentation:
 	 * SYNC needs to be driven on the FALLING edge
 	 */
-	dssdev->bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_SYNC_POSEDGE
-			  | DRM_BUS_FLAG_PIXDATA_NEGEDGE;
+	dssdev->bus_flags = DRM_BUS_FLAG_DE_HIGH
+			  | DRM_BUS_FLAG_SYNC_DRIVE_POSEDGE
+			  | DRM_BUS_FLAG_PIXDATA_DRIVE_NEGEDGE;
 
 	omapdss_display_init(dssdev);
 	omapdss_device_register(dssdev);
@@ -467,7 +448,8 @@ static int tpo_td043_remove(struct spi_device *spi)
 
 	omapdss_device_unregister(dssdev);
 
-	tpo_td043_disable(dssdev);
+	if (omapdss_device_is_enabled(dssdev))
+		tpo_td043_disable(dssdev);
 
 	sysfs_remove_group(&spi->dev.kobj, &tpo_td043_attr_group);
 
diff --git a/drivers/gpu/drm/omapdrm/dss/base.c b/drivers/gpu/drm/omapdrm/dss/base.c
index 472f56e3de70..f8dad99013e8 100644
--- a/drivers/gpu/drm/omapdrm/dss/base.c
+++ b/drivers/gpu/drm/omapdrm/dss/base.c
@@ -19,6 +19,7 @@
 #include <linux/mutex.h>
 #include <linux/of.h>
 #include <linux/of_graph.h>
+#include <linux/platform_device.h>
 
 #include "dss.h"
 #include "omapdss.h"
@@ -112,13 +113,12 @@ void omapdss_device_put(struct omap_dss_device *dssdev)
 }
 EXPORT_SYMBOL(omapdss_device_put);
 
-struct omap_dss_device *omapdss_find_device_by_port(struct device_node *src,
-						    unsigned int port)
+struct omap_dss_device *omapdss_find_device_by_node(struct device_node *node)
 {
 	struct omap_dss_device *dssdev;
 
 	list_for_each_entry(dssdev, &omapdss_devices_list, list) {
-		if (dssdev->dev->of_node == src && dssdev->of_ports & BIT(port))
+		if (dssdev->dev->of_node == node)
 			return omapdss_device_get(dssdev);
 	}
 
@@ -126,13 +126,10 @@ struct omap_dss_device *omapdss_find_device_by_port(struct device_node *src,
 }
 
 /*
- * Search for the next device starting at @from. The type argument specfies
- * which device types to consider when searching. Searching for multiple types
- * is supported by and'ing their type flags. Release the reference to the @from
- * device, and acquire a reference to the returned device if found.
+ * Search for the next output device starting at @from. Release the reference to
+ * the @from device, and acquire a reference to the returned device if found.
  */
-struct omap_dss_device *omapdss_device_get_next(struct omap_dss_device *from,
-						enum omap_dss_device_type type)
+struct omap_dss_device *omapdss_device_next_output(struct omap_dss_device *from)
 {
 	struct omap_dss_device *dssdev;
 	struct list_head *list;
@@ -160,15 +157,8 @@ struct omap_dss_device *omapdss_device_get_next(struct omap_dss_device *from,
 			goto done;
 		}
 
-		/*
-		 * Accept display entities if the display type is requested,
-		 * and output entities if the output type is requested.
-		 */
-		if ((type & OMAP_DSS_DEVICE_TYPE_DISPLAY) &&
-		    !dssdev->output_type)
-			goto done;
-		if ((type & OMAP_DSS_DEVICE_TYPE_OUTPUT) && dssdev->id &&
-		    dssdev->next)
+		if (dssdev->id &&
+		    (dssdev->next || dssdev->bridge || dssdev->panel))
 			goto done;
 	}
 
@@ -183,7 +173,12 @@ done:
 	mutex_unlock(&omapdss_devices_lock);
 	return dssdev;
 }
-EXPORT_SYMBOL(omapdss_device_get_next);
+EXPORT_SYMBOL(omapdss_device_next_output);
+
+static bool omapdss_device_is_connected(struct omap_dss_device *dssdev)
+{
+	return dssdev->dss;
+}
 
 int omapdss_device_connect(struct dss_device *dss,
 			   struct omap_dss_device *src,
@@ -191,7 +186,19 @@ int omapdss_device_connect(struct dss_device *dss,
 {
 	int ret;
 
-	dev_dbg(dst->dev, "connect\n");
+	dev_dbg(&dss->pdev->dev, "connect(%s, %s)\n",
+		src ? dev_name(src->dev) : "NULL",
+		dst ? dev_name(dst->dev) : "NULL");
+
+	if (!dst) {
+		/*
+		 * The destination is NULL when the source is connected to a
+		 * bridge or panel instead of a DSS device. Stop here, we will
+		 * attach the bridge or panel later when we will have a DRM
+		 * encoder.
+		 */
+		return src && (src->bridge || src->panel) ? 0 : -EINVAL;
+	}
 
 	if (omapdss_device_is_connected(dst))
 		return -EBUSY;
@@ -204,12 +211,6 @@ int omapdss_device_connect(struct dss_device *dss,
 		return ret;
 	}
 
-	if (src) {
-		WARN_ON(src->dst);
-		dst->src = src;
-		src->dst = dst;
-	}
-
 	return 0;
 }
 EXPORT_SYMBOL_GPL(omapdss_device_connect);
@@ -217,19 +218,20 @@ EXPORT_SYMBOL_GPL(omapdss_device_connect);
 void omapdss_device_disconnect(struct omap_dss_device *src,
 			       struct omap_dss_device *dst)
 {
-	dev_dbg(dst->dev, "disconnect\n");
+	struct dss_device *dss = src ? src->dss : dst->dss;
 
-	if (!dst->id && !omapdss_device_is_connected(dst)) {
-		WARN_ON(dst->output_type);
+	dev_dbg(&dss->pdev->dev, "disconnect(%s, %s)\n",
+		src ? dev_name(src->dev) : "NULL",
+		dst ? dev_name(dst->dev) : "NULL");
+
+	if (!dst) {
+		WARN_ON(!src->bridge && !src->panel);
 		return;
 	}
 
-	if (src) {
-		if (WARN_ON(dst != src->dst))
-			return;
-
-		dst->src = NULL;
-		src->dst = NULL;
+	if (!dst->id && !omapdss_device_is_connected(dst)) {
+		WARN_ON(!dst->display);
+		return;
 	}
 
 	WARN_ON(dst->state != OMAP_DSS_DISPLAY_DISABLED);
@@ -239,6 +241,58 @@ void omapdss_device_disconnect(struct omap_dss_device *src,
 }
 EXPORT_SYMBOL_GPL(omapdss_device_disconnect);
 
+void omapdss_device_pre_enable(struct omap_dss_device *dssdev)
+{
+	if (!dssdev)
+		return;
+
+	omapdss_device_pre_enable(dssdev->next);
+
+	if (dssdev->ops->pre_enable)
+		dssdev->ops->pre_enable(dssdev);
+}
+EXPORT_SYMBOL_GPL(omapdss_device_pre_enable);
+
+void omapdss_device_enable(struct omap_dss_device *dssdev)
+{
+	if (!dssdev)
+		return;
+
+	if (dssdev->ops->enable)
+		dssdev->ops->enable(dssdev);
+
+	omapdss_device_enable(dssdev->next);
+
+	dssdev->state = OMAP_DSS_DISPLAY_ACTIVE;
+}
+EXPORT_SYMBOL_GPL(omapdss_device_enable);
+
+void omapdss_device_disable(struct omap_dss_device *dssdev)
+{
+	if (!dssdev)
+		return;
+
+	omapdss_device_disable(dssdev->next);
+
+	if (dssdev->ops->disable)
+		dssdev->ops->disable(dssdev);
+}
+EXPORT_SYMBOL_GPL(omapdss_device_disable);
+
+void omapdss_device_post_disable(struct omap_dss_device *dssdev)
+{
+	if (!dssdev)
+		return;
+
+	if (dssdev->ops->post_disable)
+		dssdev->ops->post_disable(dssdev);
+
+	omapdss_device_post_disable(dssdev->next);
+
+	dssdev->state = OMAP_DSS_DISPLAY_DISABLED;
+}
+EXPORT_SYMBOL_GPL(omapdss_device_post_disable);
+
 /* -----------------------------------------------------------------------------
  * Components Handling
  */
@@ -249,6 +303,7 @@ struct omapdss_comp_node {
 	struct list_head list;
 	struct device_node *node;
 	bool dss_core_component;
+	const char *compat;
 };
 
 static bool omapdss_list_contains(const struct device_node *node)
@@ -266,13 +321,20 @@ static bool omapdss_list_contains(const struct device_node *node)
 static void omapdss_walk_device(struct device *dev, struct device_node *node,
 				bool dss_core)
 {
+	struct omapdss_comp_node *comp;
 	struct device_node *n;
-	struct omapdss_comp_node *comp = devm_kzalloc(dev, sizeof(*comp),
-						      GFP_KERNEL);
+	const char *compat;
+	int ret;
 
+	ret = of_property_read_string(node, "compatible", &compat);
+	if (ret < 0)
+		return;
+
+	comp = devm_kzalloc(dev, sizeof(*comp), GFP_KERNEL);
 	if (comp) {
 		comp->node = node;
 		comp->dss_core_component = dss_core;
+		comp->compat = compat;
 		list_add(&comp->list, &omapdss_comp_list);
 	}
 
@@ -312,12 +374,8 @@ void omapdss_gather_components(struct device *dev)
 
 	omapdss_walk_device(dev, dev->of_node, true);
 
-	for_each_available_child_of_node(dev->of_node, child) {
-		if (!of_find_property(child, "compatible", NULL))
-			continue;
-
+	for_each_available_child_of_node(dev->of_node, child)
 		omapdss_walk_device(dev, child, true);
-	}
 }
 EXPORT_SYMBOL(omapdss_gather_components);
 
@@ -325,6 +383,8 @@ static bool omapdss_component_is_loaded(struct omapdss_comp_node *comp)
 {
 	if (comp->dss_core_component)
 		return true;
+	if (!strstarts(comp->compat, "omapdss,"))
+		return true;
 	if (omapdss_device_is_registered(comp->node))
 		return true;
 
diff --git a/drivers/gpu/drm/omapdrm/dss/display.c b/drivers/gpu/drm/omapdrm/dss/display.c
index 34b2a4ef63a4..e93f61a567a8 100644
--- a/drivers/gpu/drm/omapdrm/dss/display.c
+++ b/drivers/gpu/drm/omapdrm/dss/display.c
@@ -23,6 +23,9 @@
 #include <linux/kernel.h>
 #include <linux/of.h>
 
+#include <drm/drm_connector.h>
+#include <drm/drm_modes.h>
+
 #include "omapdss.h"
 
 static int disp_num_counter;
@@ -39,8 +42,6 @@ void omapdss_display_init(struct omap_dss_device *dssdev)
 	if (id < 0)
 		id = disp_num_counter++;
 
-	dssdev->alias_id = id;
-
 	/* Use 'label' property for name, if it exists */
 	of_property_read_string(dssdev->dev->of_node, "label", &dssdev->name);
 
@@ -58,3 +59,22 @@ struct omap_dss_device *omapdss_display_get(struct omap_dss_device *output)
 	return omapdss_device_get(output);
 }
 EXPORT_SYMBOL_GPL(omapdss_display_get);
+
+int omapdss_display_get_modes(struct drm_connector *connector,
+			      const struct videomode *vm)
+{
+	struct drm_display_mode *mode;
+
+	mode = drm_mode_create(connector->dev);
+	if (!mode)
+		return 0;
+
+	drm_display_mode_from_videomode(vm, mode);
+
+	mode->type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED;
+	drm_mode_set_name(mode);
+	drm_mode_probed_add(connector, mode);
+
+	return 1;
+}
+EXPORT_SYMBOL_GPL(omapdss_display_get_modes);
diff --git a/drivers/gpu/drm/omapdrm/dss/dpi.c b/drivers/gpu/drm/omapdrm/dss/dpi.c
index ca4f3c4c6318..cc78dfa07f04 100644
--- a/drivers/gpu/drm/omapdrm/dss/dpi.c
+++ b/drivers/gpu/drm/omapdrm/dss/dpi.c
@@ -47,8 +47,8 @@ struct dpi_data {
 
 	struct mutex lock;
 
-	struct videomode vm;
 	struct dss_lcd_mgr_config mgr_config;
+	unsigned long pixelclock;
 	int data_lines;
 
 	struct omap_dss_device output;
@@ -347,16 +347,15 @@ static int dpi_set_dispc_clk(struct dpi_data *dpi, unsigned long pck_req,
 
 static int dpi_set_mode(struct dpi_data *dpi)
 {
-	const struct videomode *vm = &dpi->vm;
 	int lck_div = 0, pck_div = 0;
 	unsigned long fck = 0;
 	int r = 0;
 
 	if (dpi->pll)
 		r = dpi_set_pll_clk(dpi, dpi->output.dispc_channel,
-				    vm->pixelclock, &fck, &lck_div, &pck_div);
+				    dpi->pixelclock, &fck, &lck_div, &pck_div);
 	else
-		r = dpi_set_dispc_clk(dpi, vm->pixelclock, &fck,
+		r = dpi_set_dispc_clk(dpi, dpi->pixelclock, &fck,
 				&lck_div, &pck_div);
 	if (r)
 		return r;
@@ -378,7 +377,7 @@ static void dpi_config_lcd_manager(struct dpi_data *dpi)
 	dss_mgr_set_lcd_config(&dpi->output, &dpi->mgr_config);
 }
 
-static int dpi_display_enable(struct omap_dss_device *dssdev)
+static void dpi_display_enable(struct omap_dss_device *dssdev)
 {
 	struct dpi_data *dpi = dpi_get_data_from_dssdev(dssdev);
 	struct omap_dss_device *out = &dpi->output;
@@ -386,12 +385,6 @@ static int dpi_display_enable(struct omap_dss_device *dssdev)
 
 	mutex_lock(&dpi->lock);
 
-	if (!out->dispc_channel_connected) {
-		DSSERR("failed to enable display: no output/manager\n");
-		r = -ENODEV;
-		goto err_no_out_mgr;
-	}
-
 	if (dpi->vdds_dsi_reg) {
 		r = regulator_enable(dpi->vdds_dsi_reg);
 		if (r)
@@ -426,7 +419,7 @@ static int dpi_display_enable(struct omap_dss_device *dssdev)
 
 	mutex_unlock(&dpi->lock);
 
-	return 0;
+	return;
 
 err_mgr_enable:
 err_set_mode:
@@ -439,9 +432,7 @@ err_get_dispc:
 	if (dpi->vdds_dsi_reg)
 		regulator_disable(dpi->vdds_dsi_reg);
 err_reg_enable:
-err_no_out_mgr:
 	mutex_unlock(&dpi->lock);
-	return r;
 }
 
 static void dpi_display_disable(struct omap_dss_device *dssdev)
@@ -467,7 +458,7 @@ static void dpi_display_disable(struct omap_dss_device *dssdev)
 }
 
 static void dpi_set_timings(struct omap_dss_device *dssdev,
-			    const struct videomode *vm)
+			    const struct drm_display_mode *mode)
 {
 	struct dpi_data *dpi = dpi_get_data_from_dssdev(dssdev);
 
@@ -475,13 +466,13 @@ static void dpi_set_timings(struct omap_dss_device *dssdev,
 
 	mutex_lock(&dpi->lock);
 
-	dpi->vm = *vm;
+	dpi->pixelclock = mode->clock * 1000;
 
 	mutex_unlock(&dpi->lock);
 }
 
 static int dpi_check_timings(struct omap_dss_device *dssdev,
-			     struct videomode *vm)
+			     struct drm_display_mode *mode)
 {
 	struct dpi_data *dpi = dpi_get_data_from_dssdev(dssdev);
 	int lck_div, pck_div;
@@ -490,20 +481,20 @@ static int dpi_check_timings(struct omap_dss_device *dssdev,
 	struct dpi_clk_calc_ctx ctx;
 	bool ok;
 
-	if (vm->hactive % 8 != 0)
+	if (mode->hdisplay % 8 != 0)
 		return -EINVAL;
 
-	if (vm->pixelclock == 0)
+	if (mode->clock == 0)
 		return -EINVAL;
 
 	if (dpi->pll) {
-		ok = dpi_pll_clk_calc(dpi, vm->pixelclock, &ctx);
+		ok = dpi_pll_clk_calc(dpi, mode->clock * 1000, &ctx);
 		if (!ok)
 			return -EINVAL;
 
 		fck = ctx.pll_cinfo.clkout[ctx.clkout_idx];
 	} else {
-		ok = dpi_dss_clk_calc(dpi, vm->pixelclock, &ctx);
+		ok = dpi_dss_clk_calc(dpi, mode->clock * 1000, &ctx);
 		if (!ok)
 			return -EINVAL;
 
@@ -515,7 +506,7 @@ static int dpi_check_timings(struct omap_dss_device *dssdev,
 
 	pck = fck / lck_div / pck_div;
 
-	vm->pixelclock = pck;
+	mode->clock = pck / 1000;
 
 	return 0;
 }
@@ -596,23 +587,15 @@ static int dpi_connect(struct omap_dss_device *src,
 		       struct omap_dss_device *dst)
 {
 	struct dpi_data *dpi = dpi_get_data_from_dssdev(dst);
-	int r;
 
 	dpi_init_pll(dpi);
 
-	r = omapdss_device_connect(dst->dss, dst, dst->next);
-	if (r)
-		return r;
-
-	dst->dispc_channel_connected = true;
-	return 0;
+	return omapdss_device_connect(dst->dss, dst, dst->next);
 }
 
 static void dpi_disconnect(struct omap_dss_device *src,
 			   struct omap_dss_device *dst)
 {
-	dst->dispc_channel_connected = false;
-
 	omapdss_device_disconnect(dst, dst->next);
 }
 
@@ -651,25 +634,15 @@ static int dpi_init_output_port(struct dpi_data *dpi, struct device_node *port)
 
 	out->dev = &dpi->pdev->dev;
 	out->id = OMAP_DSS_OUTPUT_DPI;
-	out->output_type = OMAP_DISPLAY_TYPE_DPI;
+	out->type = OMAP_DISPLAY_TYPE_DPI;
 	out->dispc_channel = dpi_get_channel(dpi);
 	out->of_ports = BIT(port_num);
 	out->ops = &dpi_ops;
 	out->owner = THIS_MODULE;
 
-	out->next = omapdss_of_find_connected_device(out->dev->of_node, 0);
-	if (IS_ERR(out->next)) {
-		if (PTR_ERR(out->next) != -EPROBE_DEFER)
-			dev_err(out->dev, "failed to find video sink\n");
-		return PTR_ERR(out->next);
-	}
-
-	r = omapdss_output_validate(out);
-	if (r) {
-		omapdss_device_put(out->next);
-		out->next = NULL;
+	r = omapdss_device_init_output(out);
+	if (r < 0)
 		return r;
-	}
 
 	omapdss_device_register(out);
 
@@ -681,9 +654,8 @@ static void dpi_uninit_output_port(struct device_node *port)
 	struct dpi_data *dpi = port->data;
 	struct omap_dss_device *out = &dpi->output;
 
-	if (out->next)
-		omapdss_device_put(out->next);
 	omapdss_device_unregister(out);
+	omapdss_device_cleanup_output(out);
 }
 
 static const struct soc_device_attribute dpi_soc_devices[] = {
diff --git a/drivers/gpu/drm/omapdrm/dss/dsi.c b/drivers/gpu/drm/omapdrm/dss/dsi.c
index 64fb788b6647..5202862d89b5 100644
--- a/drivers/gpu/drm/omapdrm/dss/dsi.c
+++ b/drivers/gpu/drm/omapdrm/dss/dsi.c
@@ -1342,12 +1342,9 @@ static int dsi_pll_enable(struct dss_pll *pll)
 	 */
 	dsi_enable_scp_clk(dsi);
 
-	if (!dsi->vdds_dsi_enabled) {
-		r = regulator_enable(dsi->vdds_dsi_reg);
-		if (r)
-			goto err0;
-		dsi->vdds_dsi_enabled = true;
-	}
+	r = regulator_enable(dsi->vdds_dsi_reg);
+	if (r)
+		goto err0;
 
 	/* XXX PLL does not come out of reset without this... */
 	dispc_pck_free_enable(dsi->dss->dispc, 1);
@@ -1372,36 +1369,25 @@ static int dsi_pll_enable(struct dss_pll *pll)
 
 	return 0;
 err1:
-	if (dsi->vdds_dsi_enabled) {
-		regulator_disable(dsi->vdds_dsi_reg);
-		dsi->vdds_dsi_enabled = false;
-	}
+	regulator_disable(dsi->vdds_dsi_reg);
 err0:
 	dsi_disable_scp_clk(dsi);
 	dsi_runtime_put(dsi);
 	return r;
 }
 
-static void dsi_pll_uninit(struct dsi_data *dsi, bool disconnect_lanes)
+static void dsi_pll_disable(struct dss_pll *pll)
 {
+	struct dsi_data *dsi = container_of(pll, struct dsi_data, pll);
+
 	dsi_pll_power(dsi, DSI_PLL_POWER_OFF);
-	if (disconnect_lanes) {
-		WARN_ON(!dsi->vdds_dsi_enabled);
-		regulator_disable(dsi->vdds_dsi_reg);
-		dsi->vdds_dsi_enabled = false;
-	}
+
+	regulator_disable(dsi->vdds_dsi_reg);
 
 	dsi_disable_scp_clk(dsi);
 	dsi_runtime_put(dsi);
 
-	DSSDBG("PLL uninit done\n");
-}
-
-static void dsi_pll_disable(struct dss_pll *pll)
-{
-	struct dsi_data *dsi = container_of(pll, struct dsi_data, pll);
-
-	dsi_pll_uninit(dsi, true);
+	DSSDBG("PLL disable done\n");
 }
 
 static int dsi_dump_dsi_clocks(struct seq_file *s, void *p)
@@ -3753,19 +3739,13 @@ static int dsi_enable_video_output(struct omap_dss_device *dssdev, int channel)
 {
 	struct dsi_data *dsi = to_dsi_data(dssdev);
 	int bpp = dsi_get_pixel_size(dsi->pix_fmt);
-	struct omap_dss_device *out = &dsi->output;
 	u8 data_type;
 	u16 word_count;
 	int r;
 
-	if (!out->dispc_channel_connected) {
-		DSSERR("failed to enable display: no output/manager\n");
-		return -ENODEV;
-	}
-
 	r = dsi_display_init_dispc(dsi);
 	if (r)
-		goto err_init_dispc;
+		return r;
 
 	if (dsi->mode == OMAP_DSS_DSI_VIDEO_MODE) {
 		switch (dsi->pix_fmt) {
@@ -3814,7 +3794,6 @@ err_mgr_enable:
 	}
 err_pix_fmt:
 	dsi_display_uninit_dispc(dsi);
-err_init_dispc:
 	return r;
 }
 
@@ -4096,11 +4075,11 @@ static int dsi_display_init_dsi(struct dsi_data *dsi)
 
 	r = dss_pll_enable(&dsi->pll);
 	if (r)
-		goto err0;
+		return r;
 
 	r = dsi_configure_dsi_clocks(dsi);
 	if (r)
-		goto err1;
+		goto err0;
 
 	dss_select_dsi_clk_source(dsi->dss, dsi->module_id,
 				  dsi->module_id == 0 ?
@@ -4108,6 +4087,14 @@ static int dsi_display_init_dsi(struct dsi_data *dsi)
 
 	DSSDBG("PLL OK\n");
 
+	if (!dsi->vdds_dsi_enabled) {
+		r = regulator_enable(dsi->vdds_dsi_reg);
+		if (r)
+			goto err1;
+
+		dsi->vdds_dsi_enabled = true;
+	}
+
 	r = dsi_cio_init(dsi);
 	if (r)
 		goto err2;
@@ -4136,10 +4123,13 @@ static int dsi_display_init_dsi(struct dsi_data *dsi)
 err3:
 	dsi_cio_uninit(dsi);
 err2:
-	dss_select_dsi_clk_source(dsi->dss, dsi->module_id, DSS_CLK_SRC_FCK);
+	regulator_disable(dsi->vdds_dsi_reg);
+	dsi->vdds_dsi_enabled = false;
 err1:
-	dss_pll_disable(&dsi->pll);
+	dss_select_dsi_clk_source(dsi->dss, dsi->module_id, DSS_CLK_SRC_FCK);
 err0:
+	dss_pll_disable(&dsi->pll);
+
 	return r;
 }
 
@@ -4158,13 +4148,18 @@ static void dsi_display_uninit_dsi(struct dsi_data *dsi, bool disconnect_lanes,
 
 	dss_select_dsi_clk_source(dsi->dss, dsi->module_id, DSS_CLK_SRC_FCK);
 	dsi_cio_uninit(dsi);
-	dsi_pll_uninit(dsi, disconnect_lanes);
+	dss_pll_disable(&dsi->pll);
+
+	if (disconnect_lanes) {
+		regulator_disable(dsi->vdds_dsi_reg);
+		dsi->vdds_dsi_enabled = false;
+	}
 }
 
-static int dsi_display_enable(struct omap_dss_device *dssdev)
+static void dsi_display_enable(struct omap_dss_device *dssdev)
 {
 	struct dsi_data *dsi = to_dsi_data(dssdev);
-	int r = 0;
+	int r;
 
 	DSSDBG("dsi_display_enable\n");
 
@@ -4184,14 +4179,13 @@ static int dsi_display_enable(struct omap_dss_device *dssdev)
 
 	mutex_unlock(&dsi->lock);
 
-	return 0;
+	return;
 
 err_init_dsi:
 	dsi_runtime_put(dsi);
 err_get_dsi:
 	mutex_unlock(&dsi->lock);
 	DSSDBG("dsi_display_enable FAILED\n");
-	return r;
 }
 
 static void dsi_display_disable(struct omap_dss_device *dssdev,
@@ -4888,21 +4882,12 @@ static int dsi_get_clocks(struct dsi_data *dsi)
 static int dsi_connect(struct omap_dss_device *src,
 		       struct omap_dss_device *dst)
 {
-	int r;
-
-	r = omapdss_device_connect(dst->dss, dst, dst->next);
-	if (r)
-		return r;
-
-	dst->dispc_channel_connected = true;
-	return 0;
+	return omapdss_device_connect(dst->dss, dst, dst->next);
 }
 
 static void dsi_disconnect(struct omap_dss_device *src,
 			   struct omap_dss_device *dst)
 {
-	dst->dispc_channel_connected = false;
-
 	omapdss_device_disconnect(dst, dst->next);
 }
 
@@ -5138,29 +5123,19 @@ static int dsi_init_output(struct dsi_data *dsi)
 	out->id = dsi->module_id == 0 ?
 			OMAP_DSS_OUTPUT_DSI1 : OMAP_DSS_OUTPUT_DSI2;
 
-	out->output_type = OMAP_DISPLAY_TYPE_DSI;
+	out->type = OMAP_DISPLAY_TYPE_DSI;
 	out->name = dsi->module_id == 0 ? "dsi.0" : "dsi.1";
 	out->dispc_channel = dsi_get_channel(dsi);
 	out->ops = &dsi_ops;
 	out->owner = THIS_MODULE;
 	out->of_ports = BIT(0);
-	out->bus_flags = DRM_BUS_FLAG_PIXDATA_POSEDGE
+	out->bus_flags = DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE
 		       | DRM_BUS_FLAG_DE_HIGH
-		       | DRM_BUS_FLAG_SYNC_NEGEDGE;
-
-	out->next = omapdss_of_find_connected_device(out->dev->of_node, 0);
-	if (IS_ERR(out->next)) {
-		if (PTR_ERR(out->next) != -EPROBE_DEFER)
-			dev_err(out->dev, "failed to find video sink\n");
-		return PTR_ERR(out->next);
-	}
+		       | DRM_BUS_FLAG_SYNC_DRIVE_NEGEDGE;
 
-	r = omapdss_output_validate(out);
-	if (r) {
-		omapdss_device_put(out->next);
-		out->next = NULL;
+	r = omapdss_device_init_output(out);
+	if (r < 0)
 		return r;
-	}
 
 	omapdss_device_register(out);
 
@@ -5171,9 +5146,8 @@ static void dsi_uninit_output(struct dsi_data *dsi)
 {
 	struct omap_dss_device *out = &dsi->output;
 
-	if (out->next)
-		omapdss_device_put(out->next);
 	omapdss_device_unregister(out);
+	omapdss_device_cleanup_output(out);
 }
 
 static int dsi_probe_of(struct dsi_data *dsi)
diff --git a/drivers/gpu/drm/omapdrm/dss/dss-of.c b/drivers/gpu/drm/omapdrm/dss/dss-of.c
index 0422597ac6b0..b2094055c5fc 100644
--- a/drivers/gpu/drm/omapdrm/dss/dss-of.c
+++ b/drivers/gpu/drm/omapdrm/dss/dss-of.c
@@ -12,71 +12,25 @@
  * more details.
  */
 
-#include <linux/device.h>
 #include <linux/err.h>
-#include <linux/module.h>
 #include <linux/of.h>
 #include <linux/of_graph.h>
-#include <linux/seq_file.h>
 
 #include "omapdss.h"
 
-static struct device_node *
-dss_of_port_get_parent_device(struct device_node *port)
-{
-	struct device_node *np;
-	int i;
-
-	if (!port)
-		return NULL;
-
-	np = of_get_parent(port);
-
-	for (i = 0; i < 2 && np; ++i) {
-		struct property *prop;
-
-		prop = of_find_property(np, "compatible", NULL);
-
-		if (prop)
-			return np;
-
-		np = of_get_next_parent(np);
-	}
-
-	return NULL;
-}
-
 struct omap_dss_device *
 omapdss_of_find_connected_device(struct device_node *node, unsigned int port)
 {
-	struct device_node *src_node;
-	struct device_node *src_port;
-	struct device_node *ep;
-	struct omap_dss_device *src;
-	u32 port_number = 0;
+	struct device_node *remote_node;
+	struct omap_dss_device *dssdev;
 
-	/* Get the endpoint... */
-	ep = of_graph_get_endpoint_by_regs(node, port, 0);
-	if (!ep)
+	remote_node = of_graph_get_remote_node(node, port, 0);
+	if (!remote_node)
 		return NULL;
 
-	/* ... and its remote port... */
-	src_port = of_graph_get_remote_port(ep);
-	of_node_put(ep);
-	if (!src_port)
-		return NULL;
-
-	/* ... and the remote port's number and parent... */
-	of_property_read_u32(src_port, "reg", &port_number);
-	src_node = dss_of_port_get_parent_device(src_port);
-	of_node_put(src_port);
-	if (!src_node)
-		return ERR_PTR(-EINVAL);
-
-	/* ... and finally the connected device. */
-	src = omapdss_find_device_by_port(src_node, port_number);
-	of_node_put(src_node);
+	dssdev = omapdss_find_device_by_node(remote_node);
+	of_node_put(remote_node);
 
-	return src ? src : ERR_PTR(-EPROBE_DEFER);
+	return dssdev ? dssdev : ERR_PTR(-EPROBE_DEFER);
 }
 EXPORT_SYMBOL_GPL(omapdss_of_find_connected_device);
diff --git a/drivers/gpu/drm/omapdrm/dss/dss.c b/drivers/gpu/drm/omapdrm/dss/dss.c
index 7553c7fc1c45..55e68863ef15 100644
--- a/drivers/gpu/drm/omapdrm/dss/dss.c
+++ b/drivers/gpu/drm/omapdrm/dss/dss.c
@@ -1560,7 +1560,7 @@ static void dss_shutdown(struct platform_device *pdev)
 
 	DSSDBG("shutdown\n");
 
-	for_each_dss_display(dssdev) {
+	for_each_dss_output(dssdev) {
 		if (dssdev->state == OMAP_DSS_DISPLAY_ACTIVE)
 			dssdev->ops->disable(dssdev);
 	}
diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi4.c b/drivers/gpu/drm/omapdrm/dss/hdmi4.c
index aabdda394c9c..6339e2756b34 100644
--- a/drivers/gpu/drm/omapdrm/dss/hdmi4.c
+++ b/drivers/gpu/drm/omapdrm/dss/hdmi4.c
@@ -249,15 +249,15 @@ static void hdmi_power_off_full(struct omap_hdmi *hdmi)
 }
 
 static void hdmi_display_set_timings(struct omap_dss_device *dssdev,
-				     const struct videomode *vm)
+				     const struct drm_display_mode *mode)
 {
 	struct omap_hdmi *hdmi = dssdev_to_hdmi(dssdev);
 
 	mutex_lock(&hdmi->lock);
 
-	hdmi->cfg.vm = *vm;
+	drm_display_mode_to_videomode(mode, &hdmi->cfg.vm);
 
-	dispc_set_tv_pclk(hdmi->dss->dispc, vm->pixelclock);
+	dispc_set_tv_pclk(hdmi->dss->dispc, mode->clock * 1000);
 
 	mutex_unlock(&hdmi->lock);
 }
@@ -312,26 +312,20 @@ static void hdmi_stop_audio_stream(struct omap_hdmi *hd)
 	hdmi_wp_audio_enable(&hd->wp, false);
 }
 
-static int hdmi_display_enable(struct omap_dss_device *dssdev)
+static void hdmi_display_enable(struct omap_dss_device *dssdev)
 {
 	struct omap_hdmi *hdmi = dssdev_to_hdmi(dssdev);
 	unsigned long flags;
-	int r = 0;
+	int r;
 
 	DSSDBG("ENTER hdmi_display_enable\n");
 
 	mutex_lock(&hdmi->lock);
 
-	if (!dssdev->dispc_channel_connected) {
-		DSSERR("failed to enable display: no output/manager\n");
-		r = -ENODEV;
-		goto err0;
-	}
-
 	r = hdmi_power_on_full(hdmi);
 	if (r) {
 		DSSERR("failed to power on device\n");
-		goto err0;
+		goto done;
 	}
 
 	if (hdmi->audio_configured) {
@@ -351,12 +345,8 @@ static int hdmi_display_enable(struct omap_dss_device *dssdev)
 	hdmi->display_enabled = true;
 	spin_unlock_irqrestore(&hdmi->audio_playing_lock, flags);
 
+done:
 	mutex_unlock(&hdmi->lock);
-	return 0;
-
-err0:
-	mutex_unlock(&hdmi->lock);
-	return r;
 }
 
 static void hdmi_display_disable(struct omap_dss_device *dssdev)
@@ -417,21 +407,12 @@ void hdmi4_core_disable(struct hdmi_core_data *core)
 static int hdmi_connect(struct omap_dss_device *src,
 			struct omap_dss_device *dst)
 {
-	int r;
-
-	r = omapdss_device_connect(dst->dss, dst, dst->next);
-	if (r)
-		return r;
-
-	dst->dispc_channel_connected = true;
-	return 0;
+	return omapdss_device_connect(dst->dss, dst, dst->next);
 }
 
 static void hdmi_disconnect(struct omap_dss_device *src,
 			    struct omap_dss_device *dst)
 {
-	dst->dispc_channel_connected = false;
-
 	omapdss_device_disconnect(dst, dst->next);
 }
 
@@ -698,7 +679,7 @@ static int hdmi4_init_output(struct omap_hdmi *hdmi)
 
 	out->dev = &hdmi->pdev->dev;
 	out->id = OMAP_DSS_OUTPUT_HDMI;
-	out->output_type = OMAP_DISPLAY_TYPE_HDMI;
+	out->type = OMAP_DISPLAY_TYPE_HDMI;
 	out->name = "hdmi.0";
 	out->dispc_channel = OMAP_DSS_CHANNEL_DIGIT;
 	out->ops = &hdmi_ops;
@@ -706,19 +687,9 @@ static int hdmi4_init_output(struct omap_hdmi *hdmi)
 	out->of_ports = BIT(0);
 	out->ops_flags = OMAP_DSS_DEVICE_OP_EDID;
 
-	out->next = omapdss_of_find_connected_device(out->dev->of_node, 0);
-	if (IS_ERR(out->next)) {
-		if (PTR_ERR(out->next) != -EPROBE_DEFER)
-			dev_err(out->dev, "failed to find video sink\n");
-		return PTR_ERR(out->next);
-	}
-
-	r = omapdss_output_validate(out);
-	if (r) {
-		omapdss_device_put(out->next);
-		out->next = NULL;
+	r = omapdss_device_init_output(out);
+	if (r < 0)
 		return r;
-	}
 
 	omapdss_device_register(out);
 
@@ -729,9 +700,8 @@ static void hdmi4_uninit_output(struct omap_hdmi *hdmi)
 {
 	struct omap_dss_device *out = &hdmi->output;
 
-	if (out->next)
-		omapdss_device_put(out->next);
 	omapdss_device_unregister(out);
+	omapdss_device_cleanup_output(out);
 }
 
 static int hdmi4_probe_of(struct omap_hdmi *hdmi)
diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi5.c b/drivers/gpu/drm/omapdrm/dss/hdmi5.c
index 9e8556f67a29..2955bbad13bb 100644
--- a/drivers/gpu/drm/omapdrm/dss/hdmi5.c
+++ b/drivers/gpu/drm/omapdrm/dss/hdmi5.c
@@ -248,15 +248,15 @@ static void hdmi_power_off_full(struct omap_hdmi *hdmi)
 }
 
 static void hdmi_display_set_timings(struct omap_dss_device *dssdev,
-				     const struct videomode *vm)
+				     const struct drm_display_mode *mode)
 {
 	struct omap_hdmi *hdmi = dssdev_to_hdmi(dssdev);
 
 	mutex_lock(&hdmi->lock);
 
-	hdmi->cfg.vm = *vm;
+	drm_display_mode_to_videomode(mode, &hdmi->cfg.vm);
 
-	dispc_set_tv_pclk(hdmi->dss->dispc, vm->pixelclock);
+	dispc_set_tv_pclk(hdmi->dss->dispc, mode->clock * 1000);
 
 	mutex_unlock(&hdmi->lock);
 }
@@ -320,26 +320,20 @@ static void hdmi_stop_audio_stream(struct omap_hdmi *hd)
 	REG_FLD_MOD(hd->wp.base, HDMI_WP_SYSCONFIG, hd->wp_idlemode, 3, 2);
 }
 
-static int hdmi_display_enable(struct omap_dss_device *dssdev)
+static void hdmi_display_enable(struct omap_dss_device *dssdev)
 {
 	struct omap_hdmi *hdmi = dssdev_to_hdmi(dssdev);
 	unsigned long flags;
-	int r = 0;
+	int r;
 
 	DSSDBG("ENTER hdmi_display_enable\n");
 
 	mutex_lock(&hdmi->lock);
 
-	if (!dssdev->dispc_channel_connected) {
-		DSSERR("failed to enable display: no output/manager\n");
-		r = -ENODEV;
-		goto err0;
-	}
-
 	r = hdmi_power_on_full(hdmi);
 	if (r) {
 		DSSERR("failed to power on device\n");
-		goto err0;
+		goto done;
 	}
 
 	if (hdmi->audio_configured) {
@@ -359,12 +353,8 @@ static int hdmi_display_enable(struct omap_dss_device *dssdev)
 	hdmi->display_enabled = true;
 	spin_unlock_irqrestore(&hdmi->audio_playing_lock, flags);
 
+done:
 	mutex_unlock(&hdmi->lock);
-	return 0;
-
-err0:
-	mutex_unlock(&hdmi->lock);
-	return r;
 }
 
 static void hdmi_display_disable(struct omap_dss_device *dssdev)
@@ -422,21 +412,12 @@ static void hdmi_core_disable(struct omap_hdmi *hdmi)
 static int hdmi_connect(struct omap_dss_device *src,
 			struct omap_dss_device *dst)
 {
-	int r;
-
-	r = omapdss_device_connect(dst->dss, dst, dst->next);
-	if (r)
-		return r;
-
-	dst->dispc_channel_connected = true;
-	return 0;
+	return omapdss_device_connect(dst->dss, dst, dst->next);
 }
 
 static void hdmi_disconnect(struct omap_dss_device *src,
 			    struct omap_dss_device *dst)
 {
-	dst->dispc_channel_connected = false;
-
 	omapdss_device_disconnect(dst, dst->next);
 }
 
@@ -682,7 +663,7 @@ static int hdmi5_init_output(struct omap_hdmi *hdmi)
 
 	out->dev = &hdmi->pdev->dev;
 	out->id = OMAP_DSS_OUTPUT_HDMI;
-	out->output_type = OMAP_DISPLAY_TYPE_HDMI;
+	out->type = OMAP_DISPLAY_TYPE_HDMI;
 	out->name = "hdmi.0";
 	out->dispc_channel = OMAP_DSS_CHANNEL_DIGIT;
 	out->ops = &hdmi_ops;
@@ -690,19 +671,9 @@ static int hdmi5_init_output(struct omap_hdmi *hdmi)
 	out->of_ports = BIT(0);
 	out->ops_flags = OMAP_DSS_DEVICE_OP_EDID;
 
-	out->next = omapdss_of_find_connected_device(out->dev->of_node, 0);
-	if (IS_ERR(out->next)) {
-		if (PTR_ERR(out->next) != -EPROBE_DEFER)
-			dev_err(out->dev, "failed to find video sink\n");
-		return PTR_ERR(out->next);
-	}
-
-	r = omapdss_output_validate(out);
-	if (r) {
-		omapdss_device_put(out->next);
-		out->next = NULL;
+	r = omapdss_device_init_output(out);
+	if (r < 0)
 		return r;
-	}
 
 	omapdss_device_register(out);
 
@@ -713,9 +684,8 @@ static void hdmi5_uninit_output(struct omap_hdmi *hdmi)
 {
 	struct omap_dss_device *out = &hdmi->output;
 
-	if (out->next)
-		omapdss_device_put(out->next);
 	omapdss_device_unregister(out);
+	omapdss_device_cleanup_output(out);
 }
 
 static int hdmi5_probe_of(struct omap_hdmi *hdmi)
diff --git a/drivers/gpu/drm/omapdrm/dss/omapdss-boot-init.c b/drivers/gpu/drm/omapdrm/dss/omapdss-boot-init.c
index 3bfb95d230e0..2b41c75ce988 100644
--- a/drivers/gpu/drm/omapdrm/dss/omapdss-boot-init.c
+++ b/drivers/gpu/drm/omapdrm/dss/omapdss-boot-init.c
@@ -184,6 +184,22 @@ static const struct of_device_id omapdss_of_match[] __initconst = {
 	{},
 };
 
+static const struct of_device_id omapdss_of_fixups_whitelist[] __initconst = {
+	{ .compatible = "composite-video-connector" },
+	{ .compatible = "hdmi-connector" },
+	{ .compatible = "lgphilips,lb035q02" },
+	{ .compatible = "nec,nl8048hl11" },
+	{ .compatible = "panel-dsi-cm" },
+	{ .compatible = "sharp,ls037v7dw01" },
+	{ .compatible = "sony,acx565akm" },
+	{ .compatible = "svideo-connector" },
+	{ .compatible = "ti,opa362" },
+	{ .compatible = "ti,tpd12s015" },
+	{ .compatible = "toppoly,td028ttec1" },
+	{ .compatible = "tpo,td028ttec1" },
+	{ .compatible = "tpo,td043mtea1" },
+};
+
 static int __init omapdss_boot_init(void)
 {
 	struct device_node *dss, *child;
@@ -210,7 +226,7 @@ static int __init omapdss_boot_init(void)
 		n = list_first_entry(&dss_conv_list, struct dss_conv_node,
 			list);
 
-		if (!n->root)
+		if (of_match_node(omapdss_of_fixups_whitelist, n->node))
 			omapdss_omapify_node(n->node);
 
 		list_del(&n->list);
diff --git a/drivers/gpu/drm/omapdrm/dss/omapdss.h b/drivers/gpu/drm/omapdrm/dss/omapdss.h
index 33e15cb77efa..0c734d1f89e1 100644
--- a/drivers/gpu/drm/omapdrm/dss/omapdss.h
+++ b/drivers/gpu/drm/omapdrm/dss/omapdss.h
@@ -19,7 +19,6 @@
 #define __OMAP_DRM_DSS_H
 
 #include <linux/list.h>
-#include <linux/kobject.h>
 #include <linux/device.h>
 #include <linux/interrupt.h>
 #include <video/videomode.h>
@@ -68,6 +67,7 @@ struct dss_lcd_mgr_config;
 struct snd_aes_iec958;
 struct snd_cea_861_aud_if;
 struct hdmi_avi_infoframe;
+struct drm_connector;
 
 enum omap_display_type {
 	OMAP_DISPLAY_TYPE_NONE		= 0,
@@ -360,15 +360,15 @@ struct omap_dss_device_ops {
 	void (*disconnect)(struct omap_dss_device *dssdev,
 			struct omap_dss_device *dst);
 
-	int (*enable)(struct omap_dss_device *dssdev);
+	void (*pre_enable)(struct omap_dss_device *dssdev);
+	void (*enable)(struct omap_dss_device *dssdev);
 	void (*disable)(struct omap_dss_device *dssdev);
+	void (*post_disable)(struct omap_dss_device *dssdev);
 
 	int (*check_timings)(struct omap_dss_device *dssdev,
-			     struct videomode *vm);
-	void (*get_timings)(struct omap_dss_device *dssdev,
-			    struct videomode *vm);
+			     struct drm_display_mode *mode);
 	void (*set_timings)(struct omap_dss_device *dssdev,
-			    const struct videomode *vm);
+			    const struct drm_display_mode *mode);
 
 	bool (*detect)(struct omap_dss_device *dssdev);
 
@@ -380,6 +380,9 @@ struct omap_dss_device_ops {
 
 	int (*read_edid)(struct omap_dss_device *dssdev, u8 *buf, int len);
 
+	int (*get_modes)(struct omap_dss_device *dssdev,
+			 struct drm_connector *connector);
+
 	union {
 		const struct omapdss_hdmi_ops hdmi;
 		const struct omapdss_dsi_ops dsi;
@@ -390,42 +393,40 @@ struct omap_dss_device_ops {
  * enum omap_dss_device_ops_flag - Indicates which device ops are supported
  * @OMAP_DSS_DEVICE_OP_DETECT: The device supports output connection detection
  * @OMAP_DSS_DEVICE_OP_HPD: The device supports all hot-plug-related operations
- * @OMAP_DSS_DEVICE_OP_EDID: The device supports readind EDID
+ * @OMAP_DSS_DEVICE_OP_EDID: The device supports reading EDID
+ * @OMAP_DSS_DEVICE_OP_MODES: The device supports reading modes
  */
 enum omap_dss_device_ops_flag {
 	OMAP_DSS_DEVICE_OP_DETECT = BIT(0),
 	OMAP_DSS_DEVICE_OP_HPD = BIT(1),
 	OMAP_DSS_DEVICE_OP_EDID = BIT(2),
-};
-
-enum omap_dss_device_type {
-	OMAP_DSS_DEVICE_TYPE_OUTPUT = (1 << 0),
-	OMAP_DSS_DEVICE_TYPE_DISPLAY = (1 << 1),
+	OMAP_DSS_DEVICE_OP_MODES = BIT(3),
 };
 
 struct omap_dss_device {
-	struct kobject kobj;
 	struct device *dev;
 
 	struct module *owner;
 
 	struct dss_device *dss;
-	struct omap_dss_device *src;
-	struct omap_dss_device *dst;
 	struct omap_dss_device *next;
+	struct drm_bridge *bridge;
+	struct drm_panel *panel;
 
 	struct list_head list;
 
-	unsigned int alias_id;
-
+	/*
+	 * DSS type that this device generates (for DSS internal devices) or
+	 * requires (for external encoders, connectors and panels). Must be a
+	 * non-zero (different than OMAP_DISPLAY_TYPE_NONE) value.
+	 */
 	enum omap_display_type type;
+
 	/*
-	 * DSS output type that this device generates (for DSS internal devices)
-	 * or requires (for external encoders). Must be OMAP_DISPLAY_TYPE_NONE
-	 * for display devices (connectors and panels) and to non-zero value for
-	 * all other devices.
+	 * True if the device is a display (panel or connector) at the end of
+	 * the pipeline, false otherwise.
 	 */
-	enum omap_display_type output_type;
+	bool display;
 
 	const char *name;
 
@@ -434,9 +435,6 @@ struct omap_dss_device {
 	unsigned long ops_flags;
 	u32 bus_flags;
 
-	/* helper variable for driver suspend/resume */
-	bool activate_after_resume;
-
 	enum omap_display_caps caps;
 
 	enum omap_dss_display_state state;
@@ -445,7 +443,6 @@ struct omap_dss_device {
 
 	/* DISPC channel for this output */
 	enum omap_channel dispc_channel;
-	bool dispc_channel_connected;
 
 	/* output instance */
 	enum omap_dss_output_id id;
@@ -465,9 +462,6 @@ struct omap_dss_driver {
 	int (*memory_read)(struct omap_dss_device *dssdev,
 			void *buf, size_t size,
 			u16 x, u16 y, u16 w, u16 h);
-
-	void (*get_size)(struct omap_dss_device *dssdev,
-			 unsigned int *width, unsigned int *height);
 };
 
 struct dss_device *omapdss_get_dss(void);
@@ -477,32 +471,35 @@ static inline bool omapdss_is_initialized(void)
 	return !!omapdss_get_dss();
 }
 
-#define for_each_dss_display(d) \
-	while ((d = omapdss_device_get_next(d, OMAP_DSS_DEVICE_TYPE_DISPLAY)) != NULL)
 void omapdss_display_init(struct omap_dss_device *dssdev);
 struct omap_dss_device *omapdss_display_get(struct omap_dss_device *output);
+int omapdss_display_get_modes(struct drm_connector *connector,
+			      const struct videomode *vm);
 
 void omapdss_device_register(struct omap_dss_device *dssdev);
 void omapdss_device_unregister(struct omap_dss_device *dssdev);
 struct omap_dss_device *omapdss_device_get(struct omap_dss_device *dssdev);
 void omapdss_device_put(struct omap_dss_device *dssdev);
-struct omap_dss_device *omapdss_find_device_by_port(struct device_node *src,
-						    unsigned int port);
-struct omap_dss_device *omapdss_device_get_next(struct omap_dss_device *from,
-						enum omap_dss_device_type type);
+struct omap_dss_device *omapdss_find_device_by_node(struct device_node *node);
 int omapdss_device_connect(struct dss_device *dss,
 			   struct omap_dss_device *src,
 			   struct omap_dss_device *dst);
 void omapdss_device_disconnect(struct omap_dss_device *src,
 			       struct omap_dss_device *dst);
+void omapdss_device_pre_enable(struct omap_dss_device *dssdev);
+void omapdss_device_enable(struct omap_dss_device *dssdev);
+void omapdss_device_disable(struct omap_dss_device *dssdev);
+void omapdss_device_post_disable(struct omap_dss_device *dssdev);
 
 int omap_dss_get_num_overlay_managers(void);
 
 int omap_dss_get_num_overlays(void);
 
 #define for_each_dss_output(d) \
-	while ((d = omapdss_device_get_next(d, OMAP_DSS_DEVICE_TYPE_OUTPUT)) != NULL)
-int omapdss_output_validate(struct omap_dss_device *out);
+	while ((d = omapdss_device_next_output(d)) != NULL)
+struct omap_dss_device *omapdss_device_next_output(struct omap_dss_device *from);
+int omapdss_device_init_output(struct omap_dss_device *out);
+void omapdss_device_cleanup_output(struct omap_dss_device *out);
 
 typedef void (*omap_dispc_isr_t) (void *arg, u32 mask);
 int omap_dispc_register_isr(omap_dispc_isr_t isr, void *arg, u32 mask);
@@ -511,11 +508,6 @@ int omap_dispc_unregister_isr(omap_dispc_isr_t isr, void *arg, u32 mask);
 int omapdss_compat_init(void);
 void omapdss_compat_uninit(void);
 
-static inline bool omapdss_device_is_connected(struct omap_dss_device *dssdev)
-{
-	return dssdev->src;
-}
-
 static inline bool omapdss_device_is_enabled(struct omap_dss_device *dssdev)
 {
 	return dssdev->state == OMAP_DSS_DISPLAY_ACTIVE;
diff --git a/drivers/gpu/drm/omapdrm/dss/output.c b/drivers/gpu/drm/omapdrm/dss/output.c
index 18505bc70f7e..10a9ee5cdc61 100644
--- a/drivers/gpu/drm/omapdrm/dss/output.c
+++ b/drivers/gpu/drm/omapdrm/dss/output.c
@@ -20,20 +20,48 @@
 #include <linux/platform_device.h>
 #include <linux/slab.h>
 #include <linux/of.h>
+#include <linux/of_graph.h>
+
+#include <drm/drm_panel.h>
 
 #include "dss.h"
 #include "omapdss.h"
 
-int omapdss_output_validate(struct omap_dss_device *out)
+int omapdss_device_init_output(struct omap_dss_device *out)
 {
-	if (out->next && out->output_type != out->next->type) {
+	struct device_node *remote_node;
+
+	remote_node = of_graph_get_remote_node(out->dev->of_node, 0, 0);
+	if (!remote_node) {
+		dev_dbg(out->dev, "failed to find video sink\n");
+		return 0;
+	}
+
+	out->next = omapdss_find_device_by_node(remote_node);
+	out->bridge = of_drm_find_bridge(remote_node);
+	out->panel = of_drm_find_panel(remote_node);
+	if (IS_ERR(out->panel))
+		out->panel = NULL;
+
+	of_node_put(remote_node);
+
+	if (out->next && out->type != out->next->type) {
 		dev_err(out->dev, "output type and display type don't match\n");
+		omapdss_device_put(out->next);
+		out->next = NULL;
 		return -EINVAL;
 	}
 
-	return 0;
+	return out->next || out->bridge || out->panel ? 0 : -EPROBE_DEFER;
+}
+EXPORT_SYMBOL(omapdss_device_init_output);
+
+void omapdss_device_cleanup_output(struct omap_dss_device *out)
+{
+	if (out->next)
+		omapdss_device_put(out->next);
 }
-EXPORT_SYMBOL(omapdss_output_validate);
+EXPORT_SYMBOL(omapdss_device_cleanup_output);
 
 int dss_install_mgr_ops(struct dss_device *dss,
 			const struct dss_mgr_ops *mgr_ops,
diff --git a/drivers/gpu/drm/omapdrm/dss/sdi.c b/drivers/gpu/drm/omapdrm/dss/sdi.c
index b2fe2387037a..7aae52984fed 100644
--- a/drivers/gpu/drm/omapdrm/dss/sdi.c
+++ b/drivers/gpu/drm/omapdrm/dss/sdi.c
@@ -37,7 +37,7 @@ struct sdi_device {
 	struct regulator *vdds_sdi_reg;
 
 	struct dss_lcd_mgr_config mgr_config;
-	struct videomode vm;
+	unsigned long pixelclock;
 	int datapairs;
 
 	struct omap_dss_device output;
@@ -129,27 +129,22 @@ static void sdi_config_lcd_manager(struct sdi_device *sdi)
 	dss_mgr_set_lcd_config(&sdi->output, &sdi->mgr_config);
 }
 
-static int sdi_display_enable(struct omap_dss_device *dssdev)
+static void sdi_display_enable(struct omap_dss_device *dssdev)
 {
 	struct sdi_device *sdi = dssdev_to_sdi(dssdev);
 	struct dispc_clock_info dispc_cinfo;
 	unsigned long fck;
 	int r;
 
-	if (!sdi->output.dispc_channel_connected) {
-		DSSERR("failed to enable display: no output/manager\n");
-		return -ENODEV;
-	}
-
 	r = regulator_enable(sdi->vdds_sdi_reg);
 	if (r)
-		goto err_reg_enable;
+		return;
 
 	r = dispc_runtime_get(sdi->dss->dispc);
 	if (r)
 		goto err_get_dispc;
 
-	r = sdi_calc_clock_div(sdi, sdi->vm.pixelclock, &fck, &dispc_cinfo);
+	r = sdi_calc_clock_div(sdi, sdi->pixelclock, &fck, &dispc_cinfo);
 	if (r)
 		goto err_calc_clock_div;
 
@@ -185,7 +180,7 @@ static int sdi_display_enable(struct omap_dss_device *dssdev)
 	if (r)
 		goto err_mgr_enable;
 
-	return 0;
+	return;
 
 err_mgr_enable:
 	dss_sdi_disable(sdi->dss);
@@ -195,8 +190,6 @@ err_calc_clock_div:
 	dispc_runtime_put(sdi->dss->dispc);
 err_get_dispc:
 	regulator_disable(sdi->vdds_sdi_reg);
-err_reg_enable:
-	return r;
 }
 
 static void sdi_display_disable(struct omap_dss_device *dssdev)
@@ -213,36 +206,37 @@ static void sdi_display_disable(struct omap_dss_device *dssdev)
 }
 
 static void sdi_set_timings(struct omap_dss_device *dssdev,
-			    const struct videomode *vm)
+			    const struct drm_display_mode *mode)
 {
 	struct sdi_device *sdi = dssdev_to_sdi(dssdev);
 
-	sdi->vm = *vm;
+	sdi->pixelclock = mode->clock * 1000;
 }
 
 static int sdi_check_timings(struct omap_dss_device *dssdev,
-			     struct videomode *vm)
+			     struct drm_display_mode *mode)
 {
 	struct sdi_device *sdi = dssdev_to_sdi(dssdev);
 	struct dispc_clock_info dispc_cinfo;
+	unsigned long pixelclock = mode->clock * 1000;
 	unsigned long fck;
 	unsigned long pck;
 	int r;
 
-	if (vm->pixelclock == 0)
+	if (pixelclock == 0)
 		return -EINVAL;
 
-	r = sdi_calc_clock_div(sdi, vm->pixelclock, &fck, &dispc_cinfo);
+	r = sdi_calc_clock_div(sdi, pixelclock, &fck, &dispc_cinfo);
 	if (r)
 		return r;
 
 	pck = fck / dispc_cinfo.lck_div / dispc_cinfo.pck_div;
 
-	if (pck != vm->pixelclock) {
+	if (pck != pixelclock) {
 		DSSWARN("Pixel clock adjusted from %lu Hz to %lu Hz\n",
-			vm->pixelclock, pck);
+			pixelclock, pck);
 
-		vm->pixelclock = pck;
+		mode->clock = pck / 1000;
 	}
 
 	return 0;
@@ -251,21 +245,12 @@ static int sdi_check_timings(struct omap_dss_device *dssdev,
 static int sdi_connect(struct omap_dss_device *src,
 		       struct omap_dss_device *dst)
 {
-	int r;
-
-	r = omapdss_device_connect(dst->dss, dst, dst->next);
-	if (r)
-		return r;
-
-	dst->dispc_channel_connected = true;
-	return 0;
+	return omapdss_device_connect(dst->dss, dst, dst->next);
 }
 
 static void sdi_disconnect(struct omap_dss_device *src,
 			   struct omap_dss_device *dst)
 {
-	dst->dispc_channel_connected = false;
-
 	omapdss_device_disconnect(dst, dst->next);
 }
 
@@ -287,29 +272,19 @@ static int sdi_init_output(struct sdi_device *sdi)
 
 	out->dev = &sdi->pdev->dev;
 	out->id = OMAP_DSS_OUTPUT_SDI;
-	out->output_type = OMAP_DISPLAY_TYPE_SDI;
+	out->type = OMAP_DISPLAY_TYPE_SDI;
 	out->name = "sdi.0";
 	out->dispc_channel = OMAP_DSS_CHANNEL_LCD;
 	/* We have SDI only on OMAP3, where it's on port 1 */
 	out->of_ports = BIT(1);
 	out->ops = &sdi_ops;
 	out->owner = THIS_MODULE;
-	out->bus_flags = DRM_BUS_FLAG_PIXDATA_POSEDGE	/* 15.5.9.1.2 */
-		       | DRM_BUS_FLAG_SYNC_POSEDGE;
-
-	out->next = omapdss_of_find_connected_device(out->dev->of_node, 1);
-	if (IS_ERR(out->next)) {
-		if (PTR_ERR(out->next) != -EPROBE_DEFER)
-			dev_err(out->dev, "failed to find video sink\n");
-		return PTR_ERR(out->next);
-	}
+	out->bus_flags = DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE	/* 15.5.9.1.2 */
+		       | DRM_BUS_FLAG_SYNC_DRIVE_POSEDGE;
 
-	r = omapdss_output_validate(out);
-	if (r) {
-		omapdss_device_put(out->next);
-		out->next = NULL;
+	r = omapdss_device_init_output(out);
+	if (r < 0)
 		return r;
-	}
 
 	omapdss_device_register(out);
 
@@ -318,9 +293,8 @@ static int sdi_init_output(struct sdi_device *sdi)
 
 static void sdi_uninit_output(struct sdi_device *sdi)
 {
-	if (sdi->output.next)
-		omapdss_device_put(sdi->output.next);
 	omapdss_device_unregister(&sdi->output);
+	omapdss_device_cleanup_output(&sdi->output);
 }
 
 int sdi_init_port(struct dss_device *dss, struct platform_device *pdev,
diff --git a/drivers/gpu/drm/omapdrm/dss/venc.c b/drivers/gpu/drm/omapdrm/dss/venc.c
index b5f52727f8b1..da43b865d973 100644
--- a/drivers/gpu/drm/omapdrm/dss/venc.c
+++ b/drivers/gpu/drm/omapdrm/dss/venc.c
@@ -267,63 +267,40 @@ enum venc_videomode {
 	VENC_MODE_NTSC,
 };
 
-static const struct videomode omap_dss_pal_vm = {
-	.hactive	= 720,
-	.vactive	= 574,
-	.pixelclock	= 13500000,
-	.hsync_len	= 64,
-	.hfront_porch	= 12,
-	.hback_porch	= 68,
-	.vsync_len	= 5,
-	.vfront_porch	= 5,
-	.vback_porch	= 41,
-
-	.flags		= DISPLAY_FLAGS_INTERLACED | DISPLAY_FLAGS_HSYNC_LOW |
-			  DISPLAY_FLAGS_VSYNC_LOW | DISPLAY_FLAGS_DE_HIGH |
-			  DISPLAY_FLAGS_PIXDATA_POSEDGE |
-			  DISPLAY_FLAGS_SYNC_NEGEDGE,
+static const struct drm_display_mode omap_dss_pal_mode = {
+	.hdisplay	= 720,
+	.hsync_start	= 732,
+	.hsync_end	= 796,
+	.htotal		= 864,
+	.vdisplay	= 574,
+	.vsync_start	= 579,
+	.vsync_end	= 584,
+	.vtotal		= 625,
+	.clock		= 13500,
+
+	.flags		= DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_NHSYNC |
+			  DRM_MODE_FLAG_NVSYNC,
 };
 
-static const struct videomode omap_dss_ntsc_vm = {
-	.hactive	= 720,
-	.vactive	= 482,
-	.pixelclock	= 13500000,
-	.hsync_len	= 64,
-	.hfront_porch	= 16,
-	.hback_porch	= 58,
-	.vsync_len	= 6,
-	.vfront_porch	= 6,
-	.vback_porch	= 31,
-
-	.flags		= DISPLAY_FLAGS_INTERLACED | DISPLAY_FLAGS_HSYNC_LOW |
-			  DISPLAY_FLAGS_VSYNC_LOW | DISPLAY_FLAGS_DE_HIGH |
-			  DISPLAY_FLAGS_PIXDATA_POSEDGE |
-			  DISPLAY_FLAGS_SYNC_NEGEDGE,
+static const struct drm_display_mode omap_dss_ntsc_mode = {
+	.hdisplay	= 720,
+	.hsync_start	= 736,
+	.hsync_end	= 800,
+	.htotal		= 858,
+	.vdisplay	= 482,
+	.vsync_start	= 488,
+	.vsync_end	= 494,
+	.vtotal		= 525,
+	.clock		= 13500,
+
+	.flags		= DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_NHSYNC |
+			  DRM_MODE_FLAG_NVSYNC,
 };
 
-static enum venc_videomode venc_get_videomode(const struct videomode *vm)
-{
-	if (!(vm->flags & DISPLAY_FLAGS_INTERLACED))
-		return VENC_MODE_UNKNOWN;
-
-	if (vm->pixelclock == omap_dss_pal_vm.pixelclock &&
-	    vm->hactive == omap_dss_pal_vm.hactive &&
-	    vm->vactive == omap_dss_pal_vm.vactive)
-		return VENC_MODE_PAL;
-
-	if (vm->pixelclock == omap_dss_ntsc_vm.pixelclock &&
-	    vm->hactive == omap_dss_ntsc_vm.hactive &&
-	    vm->vactive == omap_dss_ntsc_vm.vactive)
-		return VENC_MODE_NTSC;
-
-	return VENC_MODE_UNKNOWN;
-}
-
 struct venc_device {
 	struct platform_device *pdev;
 	void __iomem *base;
 	struct mutex venc_lock;
-	u32 wss_data;
 	struct regulator *vdda_dac_reg;
 	struct dss_device *dss;
 
@@ -331,7 +308,7 @@ struct venc_device {
 
 	struct clk	*tv_dac_clk;
 
-	struct videomode vm;
+	const struct venc_config *config;
 	enum omap_dss_venc_type type;
 	bool invert_polarity;
 	bool requires_tv_dac_clk;
@@ -367,8 +344,7 @@ static void venc_write_config(struct venc_device *venc,
 	venc_write_reg(venc, VENC_BLACK_LEVEL, config->black_level);
 	venc_write_reg(venc, VENC_BLANK_LEVEL, config->blank_level);
 	venc_write_reg(venc, VENC_M_CONTROL, config->m_control);
-	venc_write_reg(venc, VENC_BSTAMP_WSS_DATA, config->bstamp_wss_data |
-		       venc->wss_data);
+	venc_write_reg(venc, VENC_BSTAMP_WSS_DATA, config->bstamp_wss_data);
 	venc_write_reg(venc, VENC_S_CARR, config->s_carr);
 	venc_write_reg(venc, VENC_L21__WC_CTL, config->l21__wc_ctl);
 	venc_write_reg(venc, VENC_SAVID__EAVID, config->savid__eavid);
@@ -452,18 +428,6 @@ static void venc_runtime_put(struct venc_device *venc)
 	WARN_ON(r < 0 && r != -ENOSYS);
 }
 
-static const struct venc_config *venc_timings_to_config(const struct videomode *vm)
-{
-	switch (venc_get_videomode(vm)) {
-	default:
-		WARN_ON_ONCE(1);
-	case VENC_MODE_PAL:
-		return &venc_config_pal_trm;
-	case VENC_MODE_NTSC:
-		return &venc_config_ntsc_trm;
-	}
-}
-
 static int venc_power_on(struct venc_device *venc)
 {
 	u32 l;
@@ -474,7 +438,7 @@ static int venc_power_on(struct venc_device *venc)
 		goto err0;
 
 	venc_reset(venc);
-	venc_write_config(venc, venc_timings_to_config(&venc->vm));
+	venc_write_config(venc, venc->config);
 
 	dss_set_venc_output(venc->dss, venc->type);
 	dss_set_dac_pwrdn_bgz(venc->dss, 1);
@@ -524,33 +488,17 @@ static void venc_power_off(struct venc_device *venc)
 	venc_runtime_put(venc);
 }
 
-static int venc_display_enable(struct omap_dss_device *dssdev)
+static void venc_display_enable(struct omap_dss_device *dssdev)
 {
 	struct venc_device *venc = dssdev_to_venc(dssdev);
-	int r;
 
 	DSSDBG("venc_display_enable\n");
 
 	mutex_lock(&venc->venc_lock);
 
-	if (!dssdev->dispc_channel_connected) {
-		DSSERR("Failed to enable display: no output/manager\n");
-		r = -ENODEV;
-		goto err0;
-	}
-
-	r = venc_power_on(venc);
-	if (r)
-		goto err0;
-
-	venc->wss_data = 0;
+	venc_power_on(venc);
 
 	mutex_unlock(&venc->venc_lock);
-
-	return 0;
-err0:
-	mutex_unlock(&venc->venc_lock);
-	return r;
 }
 
 static void venc_display_disable(struct omap_dss_device *dssdev)
@@ -566,30 +514,70 @@ static void venc_display_disable(struct omap_dss_device *dssdev)
 	mutex_unlock(&venc->venc_lock);
 }
 
-static void venc_get_timings(struct omap_dss_device *dssdev,
-			     struct videomode *vm)
+static int venc_get_modes(struct omap_dss_device *dssdev,
+			  struct drm_connector *connector)
 {
-	struct venc_device *venc = dssdev_to_venc(dssdev);
+	static const struct drm_display_mode *modes[] = {
+		&omap_dss_pal_mode,
+		&omap_dss_ntsc_mode,
+	};
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(modes); ++i) {
+		struct drm_display_mode *mode;
+
+		mode = drm_mode_duplicate(connector->dev, modes[i]);
+		if (!mode)
+			return i;
+
+		mode->type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED;
+		drm_mode_set_name(mode);
+		drm_mode_probed_add(connector, mode);
+	}
 
-	mutex_lock(&venc->venc_lock);
-	*vm = venc->vm;
-	mutex_unlock(&venc->venc_lock);
+	return ARRAY_SIZE(modes);
+}
+
+static enum venc_videomode venc_get_videomode(const struct drm_display_mode *mode)
+{
+	if (!(mode->flags & DRM_MODE_FLAG_INTERLACE))
+		return VENC_MODE_UNKNOWN;
+
+	if (mode->clock == omap_dss_pal_mode.clock &&
+	    mode->hdisplay == omap_dss_pal_mode.hdisplay &&
+	    mode->vdisplay == omap_dss_pal_mode.vdisplay)
+		return VENC_MODE_PAL;
+
+	if (mode->clock == omap_dss_ntsc_mode.clock &&
+	    mode->hdisplay == omap_dss_ntsc_mode.hdisplay &&
+	    mode->vdisplay == omap_dss_ntsc_mode.vdisplay)
+		return VENC_MODE_NTSC;
+
+	return VENC_MODE_UNKNOWN;
 }
 
 static void venc_set_timings(struct omap_dss_device *dssdev,
-			     const struct videomode *vm)
+			     const struct drm_display_mode *mode)
 {
 	struct venc_device *venc = dssdev_to_venc(dssdev);
+	enum venc_videomode venc_mode = venc_get_videomode(mode);
 
 	DSSDBG("venc_set_timings\n");
 
 	mutex_lock(&venc->venc_lock);
 
-	/* Reset WSS data when the TV standard changes. */
-	if (memcmp(&venc->vm, vm, sizeof(*vm)))
-		venc->wss_data = 0;
+	switch (venc_mode) {
+	default:
+		WARN_ON_ONCE(1);
+		/* Fall-through */
+	case VENC_MODE_PAL:
+		venc->config = &venc_config_pal_trm;
+		break;
 
-	venc->vm = *vm;
+	case VENC_MODE_NTSC:
+		venc->config = &venc_config_ntsc_trm;
+		break;
+	}
 
 	dispc_set_tv_pclk(venc->dss->dispc, 13500000);
 
@@ -597,22 +585,26 @@ static void venc_set_timings(struct omap_dss_device *dssdev,
 }
 
 static int venc_check_timings(struct omap_dss_device *dssdev,
-			      struct videomode *vm)
+			      struct drm_display_mode *mode)
 {
 	DSSDBG("venc_check_timings\n");
 
-	switch (venc_get_videomode(vm)) {
+	switch (venc_get_videomode(mode)) {
 	case VENC_MODE_PAL:
-		*vm = omap_dss_pal_vm;
-		return 0;
+		drm_mode_copy(mode, &omap_dss_pal_mode);
+		break;
 
 	case VENC_MODE_NTSC:
-		*vm = omap_dss_ntsc_vm;
-		return 0;
+		drm_mode_copy(mode, &omap_dss_ntsc_mode);
+		break;
 
 	default:
 		return -EINVAL;
 	}
+
+	drm_mode_set_crtcinfo(mode, CRTC_INTERLACE_HALVE_V);
+	drm_mode_set_name(mode);
+	return 0;
 }
 
 static int venc_dump_regs(struct seq_file *s, void *p)
@@ -695,21 +687,12 @@ static int venc_get_clocks(struct venc_device *venc)
 static int venc_connect(struct omap_dss_device *src,
 			struct omap_dss_device *dst)
 {
-	int r;
-
-	r = omapdss_device_connect(dst->dss, dst, dst->next);
-	if (r)
-		return r;
-
-	dst->dispc_channel_connected = true;
-	return 0;
+	return omapdss_device_connect(dst->dss, dst, dst->next);
 }
 
 static void venc_disconnect(struct omap_dss_device *src,
 			    struct omap_dss_device *dst)
 {
-	dst->dispc_channel_connected = false;
-
 	omapdss_device_disconnect(dst, dst->next);
 }
 
@@ -721,8 +704,9 @@ static const struct omap_dss_device_ops venc_ops = {
 	.disable = venc_display_disable,
 
 	.check_timings = venc_check_timings,
-	.get_timings = venc_get_timings,
 	.set_timings = venc_set_timings,
+
+	.get_modes = venc_get_modes,
 };
 
 /* -----------------------------------------------------------------------------
@@ -776,26 +760,17 @@ static int venc_init_output(struct venc_device *venc)
 
 	out->dev = &venc->pdev->dev;
 	out->id = OMAP_DSS_OUTPUT_VENC;
-	out->output_type = OMAP_DISPLAY_TYPE_VENC;
+	out->type = OMAP_DISPLAY_TYPE_VENC;
 	out->name = "venc.0";
 	out->dispc_channel = OMAP_DSS_CHANNEL_DIGIT;
 	out->ops = &venc_ops;
 	out->owner = THIS_MODULE;
 	out->of_ports = BIT(0);
+	out->ops_flags = OMAP_DSS_DEVICE_OP_MODES;
 
-	out->next = omapdss_of_find_connected_device(out->dev->of_node, 0);
-	if (IS_ERR(out->next)) {
-		if (PTR_ERR(out->next) != -EPROBE_DEFER)
-			dev_err(out->dev, "failed to find video sink\n");
-		return PTR_ERR(out->next);
-	}
-
-	r = omapdss_output_validate(out);
-	if (r) {
-		omapdss_device_put(out->next);
-		out->next = NULL;
+	r = omapdss_device_init_output(out);
+	if (r < 0)
 		return r;
-	}
 
 	omapdss_device_register(out);
 
@@ -804,9 +779,8 @@ static int venc_init_output(struct venc_device *venc)
 
 static void venc_uninit_output(struct venc_device *venc)
 {
-	if (venc->output.next)
-		omapdss_device_put(venc->output.next);
 	omapdss_device_unregister(&venc->output);
+	omapdss_device_cleanup_output(&venc->output);
 }
 
 static int venc_probe_of(struct venc_device *venc)
@@ -878,8 +852,7 @@ static int venc_probe(struct platform_device *pdev)
 
 	mutex_init(&venc->venc_lock);
 
-	venc->wss_data = 0;
-	venc->vm = omap_dss_pal_vm;
+	venc->config = &venc_config_pal_trm;
 
 	venc_mem = platform_get_resource(venc->pdev, IORESOURCE_MEM, 0);
 	venc->base = devm_ioremap_resource(&pdev->dev, venc_mem);
diff --git a/drivers/gpu/drm/omapdrm/omap_connector.c b/drivers/gpu/drm/omapdrm/omap_connector.c
index 9da94d10782a..5967283934e1 100644
--- a/drivers/gpu/drm/omapdrm/omap_connector.c
+++ b/drivers/gpu/drm/omapdrm/omap_connector.c
@@ -17,6 +17,7 @@
 
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
+#include <drm/drm_panel.h>
 #include <drm/drm_probe_helper.h>
 
 #include "omap_drv.h"
@@ -30,24 +31,27 @@
 struct omap_connector {
 	struct drm_connector base;
 	struct omap_dss_device *output;
-	struct omap_dss_device *display;
 	struct omap_dss_device *hpd;
 	bool hdmi_mode;
 };
 
 static void omap_connector_hpd_notify(struct drm_connector *connector,
-				      struct omap_dss_device *src,
 				      enum drm_connector_status status)
 {
-	if (status == connector_status_disconnected) {
-		/*
-		 * If the source is an HDMI encoder, notify it of disconnection.
-		 * This is required to let the HDMI encoder reset any internal
-		 * state related to connection status, such as the CEC address.
-		 */
-		if (src && src->type == OMAP_DISPLAY_TYPE_HDMI &&
-		    src->ops->hdmi.lost_hotplug)
-			src->ops->hdmi.lost_hotplug(src);
+	struct omap_connector *omap_connector = to_omap_connector(connector);
+	struct omap_dss_device *dssdev;
+
+	if (status != connector_status_disconnected)
+		return;
+
+	/*
+	 * Notify all devics in the pipeline of disconnection. This is required
+	 * to let the HDMI encoders reset their internal state related to
+	 * connection status, such as the CEC address.
+	 */
+	for (dssdev = omap_connector->output; dssdev; dssdev = dssdev->next) {
+		if (dssdev->ops && dssdev->ops->hdmi.lost_hotplug)
+			dssdev->ops->hdmi.lost_hotplug(dssdev);
 	}
 }
 
@@ -67,7 +71,7 @@ static void omap_connector_hpd_cb(void *cb_data,
 	if (old_status == status)
 		return;
 
-	omap_connector_hpd_notify(connector, omap_connector->hpd, status);
+	omap_connector_hpd_notify(connector, status);
 
 	drm_kms_helper_hotplug_event(dev);
 }
@@ -103,20 +107,20 @@ omap_connector_find_device(struct drm_connector *connector,
 			   enum omap_dss_device_ops_flag op)
 {
 	struct omap_connector *omap_connector = to_omap_connector(connector);
-	struct omap_dss_device *dssdev;
+	struct omap_dss_device *dssdev = NULL;
+	struct omap_dss_device *d;
 
-	for (dssdev = omap_connector->display; dssdev; dssdev = dssdev->src) {
-		if (dssdev->ops_flags & op)
-			return dssdev;
+	for (d = omap_connector->output; d; d = d->next) {
+		if (d->ops_flags & op)
+			dssdev = d;
 	}
 
-	return NULL;
+	return dssdev;
 }
 
 static enum drm_connector_status omap_connector_detect(
 		struct drm_connector *connector, bool force)
 {
-	struct omap_connector *omap_connector = to_omap_connector(connector);
 	struct omap_dss_device *dssdev;
 	enum drm_connector_status status;
 
@@ -128,13 +132,12 @@ static enum drm_connector_status omap_connector_detect(
 		       ? connector_status_connected
 		       : connector_status_disconnected;
 
-		omap_connector_hpd_notify(connector, dssdev->src, status);
+		omap_connector_hpd_notify(connector, status);
 	} else {
-		switch (omap_connector->display->type) {
-		case OMAP_DISPLAY_TYPE_DPI:
-		case OMAP_DISPLAY_TYPE_DBI:
-		case OMAP_DISPLAY_TYPE_SDI:
-		case OMAP_DISPLAY_TYPE_DSI:
+		switch (connector->connector_type) {
+		case DRM_MODE_CONNECTOR_DPI:
+		case DRM_MODE_CONNECTOR_LVDS:
+		case DRM_MODE_CONNECTOR_DSI:
 			status = connector_status_connected;
 			break;
 		default:
@@ -143,7 +146,7 @@ static enum drm_connector_status omap_connector_detect(
 		}
 	}
 
-	VERB("%s: %d (force=%d)", omap_connector->display->name, status, force);
+	VERB("%s: %d (force=%d)", connector->name, status, force);
 
 	return status;
 }
@@ -152,7 +155,7 @@ static void omap_connector_destroy(struct drm_connector *connector)
 {
 	struct omap_connector *omap_connector = to_omap_connector(connector);
 
-	DBG("%s", omap_connector->display->name);
+	DBG("%s", connector->name);
 
 	if (omap_connector->hpd) {
 		struct omap_dss_device *hpd = omap_connector->hpd;
@@ -166,7 +169,6 @@ static void omap_connector_destroy(struct drm_connector *connector)
 	drm_connector_cleanup(connector);
 
 	omapdss_device_put(omap_connector->output);
-	omapdss_device_put(omap_connector->display);
 
 	kfree(omap_connector);
 }
@@ -212,10 +214,8 @@ static int omap_connector_get_modes(struct drm_connector *connector)
 {
 	struct omap_connector *omap_connector = to_omap_connector(connector);
 	struct omap_dss_device *dssdev;
-	struct drm_display_mode *mode;
-	struct videomode vm = {0};
 
-	DBG("%s", omap_connector->display->name);
+	DBG("%s", connector->name);
 
 	/*
 	 * If display exposes EDID, then we parse that in the normal way to
@@ -227,89 +227,71 @@ static int omap_connector_get_modes(struct drm_connector *connector)
 		return omap_connector_get_modes_edid(connector, dssdev);
 
 	/*
-	 * Otherwise we have either a fixed resolution panel or an output that
-	 * doesn't support modes discovery (e.g. DVI or VGA with the DDC bus
-	 * unconnected, or analog TV). Start by querying the size.
+	 * Otherwise if the display pipeline reports modes (e.g. with a fixed
+	 * resolution panel or an analog TV output), query it.
 	 */
-	dssdev = omap_connector->display;
-	if (dssdev->driver && dssdev->driver->get_size)
-		dssdev->driver->get_size(dssdev,
-					 &connector->display_info.width_mm,
-					 &connector->display_info.height_mm);
+	dssdev = omap_connector_find_device(connector,
+					    OMAP_DSS_DEVICE_OP_MODES);
+	if (dssdev)
+		return dssdev->ops->get_modes(dssdev, connector);
 
 	/*
-	 * Iterate over the pipeline to find the first device that can provide
-	 * timing information. If we can't find any, we just let the KMS core
-	 * add the default modes.
+	 * Otherwise if the display pipeline uses a drm_panel, we delegate the
+	 * operation to the panel API.
 	 */
-	for (dssdev = omap_connector->display; dssdev; dssdev = dssdev->src) {
-		if (dssdev->ops->get_timings)
-			break;
-	}
-	if (!dssdev)
-		return 0;
+	if (omap_connector->output->panel)
+		return drm_panel_get_modes(omap_connector->output->panel);
 
-	/* Add a single mode corresponding to the fixed panel timings. */
-	mode = drm_mode_create(connector->dev);
-	if (!mode)
-		return 0;
+	/*
+	 * We can't retrieve modes, which can happen for instance for a DVI or
+	 * VGA output with the DDC bus unconnected. The KMS core will add the
+	 * default modes.
+	 */
+	return 0;
+}
 
-	dssdev->ops->get_timings(dssdev, &vm);
+enum drm_mode_status omap_connector_mode_fixup(struct omap_dss_device *dssdev,
+					const struct drm_display_mode *mode,
+					struct drm_display_mode *adjusted_mode)
+{
+	int ret;
 
-	drm_display_mode_from_videomode(&vm, mode);
+	drm_mode_copy(adjusted_mode, mode);
 
-	mode->type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED;
-	drm_mode_set_name(mode);
-	drm_mode_probed_add(connector, mode);
+	for (; dssdev; dssdev = dssdev->next) {
+		if (!dssdev->ops->check_timings)
+			continue;
+
+		ret = dssdev->ops->check_timings(dssdev, adjusted_mode);
+		if (ret)
+			return MODE_BAD;
+	}
 
-	return 1;
+	return MODE_OK;
 }
 
-static int omap_connector_mode_valid(struct drm_connector *connector,
+static enum drm_mode_status omap_connector_mode_valid(struct drm_connector *connector,
 				 struct drm_display_mode *mode)
 {
 	struct omap_connector *omap_connector = to_omap_connector(connector);
-	enum omap_channel channel = omap_connector->output->dispc_channel;
-	struct omap_drm_private *priv = connector->dev->dev_private;
-	struct omap_dss_device *dssdev;
-	struct videomode vm = {0};
-	struct drm_device *dev = connector->dev;
-	struct drm_display_mode *new_mode;
-	int r, ret = MODE_BAD;
-
-	drm_display_mode_to_videomode(mode, &vm);
-	mode->vrefresh = drm_mode_vrefresh(mode);
+	struct drm_display_mode new_mode = { { 0 } };
+	enum drm_mode_status status;
 
-	r = priv->dispc_ops->mgr_check_timings(priv->dispc, channel, &vm);
-	if (r)
+	status = omap_connector_mode_fixup(omap_connector->output, mode,
+					   &new_mode);
+	if (status != MODE_OK)
 		goto done;
 
-	for (dssdev = omap_connector->output; dssdev; dssdev = dssdev->next) {
-		if (!dssdev->ops->check_timings)
-			continue;
-
-		r = dssdev->ops->check_timings(dssdev, &vm);
-		if (r)
-			goto done;
-	}
-
-	/* check if vrefresh is still valid */
-	new_mode = drm_mode_duplicate(dev, mode);
-	if (!new_mode)
-		return MODE_BAD;
-
-	new_mode->clock = vm.pixelclock / 1000;
-	new_mode->vrefresh = 0;
-	if (mode->vrefresh == drm_mode_vrefresh(new_mode))
-		ret = MODE_OK;
-	drm_mode_destroy(dev, new_mode);
+	/* Check if vrefresh is still valid. */
+	if (drm_mode_vrefresh(mode) != drm_mode_vrefresh(&new_mode))
+		status = MODE_NOCLOCK;
 
 done:
 	DBG("connector: mode %s: " DRM_MODE_FMT,
-			(ret == MODE_OK) ? "valid" : "invalid",
+			(status == MODE_OK) ? "valid" : "invalid",
 			DRM_MODE_ARG(mode));
 
-	return ret;
+	return status;
 }
 
 static const struct drm_connector_funcs omap_connector_funcs = {
@@ -326,9 +308,16 @@ static const struct drm_connector_helper_funcs omap_connector_helper_funcs = {
 	.mode_valid = omap_connector_mode_valid,
 };
 
-static int omap_connector_get_type(struct omap_dss_device *display)
+static int omap_connector_get_type(struct omap_dss_device *output)
 {
-	switch (display->type) {
+	struct omap_dss_device *display;
+	enum omap_display_type type;
+
+	display = omapdss_display_get(output);
+	type = display->type;
+	omapdss_device_put(display);
+
+	switch (type) {
 	case OMAP_DISPLAY_TYPE_HDMI:
 		return DRM_MODE_CONNECTOR_HDMIA;
 	case OMAP_DISPLAY_TYPE_DVI:
@@ -351,28 +340,26 @@ static int omap_connector_get_type(struct omap_dss_device *display)
 /* initialize connector */
 struct drm_connector *omap_connector_init(struct drm_device *dev,
 					  struct omap_dss_device *output,
-					  struct omap_dss_device *display,
 					  struct drm_encoder *encoder)
 {
 	struct drm_connector *connector = NULL;
 	struct omap_connector *omap_connector;
 	struct omap_dss_device *dssdev;
 
-	DBG("%s", display->name);
+	DBG("%s", output->name);
 
 	omap_connector = kzalloc(sizeof(*omap_connector), GFP_KERNEL);
 	if (!omap_connector)
 		goto fail;
 
 	omap_connector->output = omapdss_device_get(output);
-	omap_connector->display = omapdss_device_get(display);
 
 	connector = &omap_connector->base;
 	connector->interlace_allowed = 1;
 	connector->doublescan_allowed = 0;
 
 	drm_connector_init(dev, connector, &omap_connector_funcs,
-			   omap_connector_get_type(display));
+			   omap_connector_get_type(output));
 	drm_connector_helper_add(connector, &omap_connector_helper_funcs);
 
 	/*
diff --git a/drivers/gpu/drm/omapdrm/omap_connector.h b/drivers/gpu/drm/omapdrm/omap_connector.h
index 854099801649..608085219336 100644
--- a/drivers/gpu/drm/omapdrm/omap_connector.h
+++ b/drivers/gpu/drm/omapdrm/omap_connector.h
@@ -22,6 +22,8 @@
 
 #include <linux/types.h>
 
+enum drm_mode_status;
+
 struct drm_connector;
 struct drm_device;
 struct drm_encoder;
@@ -29,12 +31,12 @@ struct omap_dss_device;
 
 struct drm_connector *omap_connector_init(struct drm_device *dev,
 					  struct omap_dss_device *output,
-					  struct omap_dss_device *display,
 					  struct drm_encoder *encoder);
-struct drm_encoder *omap_connector_attached_encoder(
-		struct drm_connector *connector);
 bool omap_connector_get_hdmi_mode(struct drm_connector *connector);
 void omap_connector_enable_hpd(struct drm_connector *connector);
 void omap_connector_disable_hpd(struct drm_connector *connector);
+enum drm_mode_status omap_connector_mode_fixup(struct omap_dss_device *dssdev,
+					const struct drm_display_mode *mode,
+					struct drm_display_mode *adjusted_mode);
 
 #endif /* __OMAPDRM_CONNECTOR_H__ */
diff --git a/drivers/gpu/drm/omapdrm/omap_crtc.c b/drivers/gpu/drm/omapdrm/omap_crtc.c
index d99e24dcc0bf..5a29bf01c0e8 100644
--- a/drivers/gpu/drm/omapdrm/omap_crtc.c
+++ b/drivers/gpu/drm/omapdrm/omap_crtc.c
@@ -128,7 +128,7 @@ static void omap_crtc_set_enabled(struct drm_crtc *crtc, bool enable)
 	if (WARN_ON(omap_crtc->enabled == enable))
 		return;
 
-	if (omap_crtc->pipe->output->output_type == OMAP_DISPLAY_TYPE_HDMI) {
+	if (omap_crtc->pipe->output->type == OMAP_DISPLAY_TYPE_HDMI) {
 		priv->dispc_ops->mgr_enable(priv->dispc, channel, enable);
 		omap_crtc->enabled = enable;
 		return;
@@ -390,6 +390,15 @@ static enum drm_mode_status omap_crtc_mode_valid(struct drm_crtc *crtc,
 					const struct drm_display_mode *mode)
 {
 	struct omap_drm_private *priv = crtc->dev->dev_private;
+	struct omap_crtc *omap_crtc = to_omap_crtc(crtc);
+	struct videomode vm = {0};
+	int r;
+
+	drm_display_mode_to_videomode(mode, &vm);
+	r = priv->dispc_ops->mgr_check_timings(priv->dispc, omap_crtc->channel,
+					       &vm);
+	if (r)
+		return r;
 
 	/* Check for bandwidth limit */
 	if (priv->max_bandwidth) {
@@ -657,7 +666,7 @@ struct drm_crtc *omap_crtc_init(struct drm_device *dev,
 					&omap_crtc_funcs, NULL);
 	if (ret < 0) {
 		dev_err(dev->dev, "%s(): could not init crtc for: %s\n",
-			__func__, pipe->display->name);
+			__func__, pipe->output->name);
 		kfree(omap_crtc);
 		return ERR_PTR(ret);
 	}
diff --git a/drivers/gpu/drm/omapdrm/omap_drv.c b/drivers/gpu/drm/omapdrm/omap_drv.c
index f8292278f57d..1b9b6f5e48e1 100644
--- a/drivers/gpu/drm/omapdrm/omap_drv.c
+++ b/drivers/gpu/drm/omapdrm/omap_drv.c
@@ -23,6 +23,7 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_probe_helper.h>
+#include <drm/drm_panel.h>
 
 #include "omap_dmm_tiler.h"
 #include "omap_drv.h"
@@ -137,12 +138,13 @@ static void omap_disconnect_pipelines(struct drm_device *ddev)
 	for (i = 0; i < priv->num_pipes; i++) {
 		struct omap_drm_pipeline *pipe = &priv->pipes[i];
 
+		if (pipe->output->panel)
+			drm_panel_detach(pipe->output->panel);
+
 		omapdss_device_disconnect(NULL, pipe->output);
 
 		omapdss_device_put(pipe->output);
-		omapdss_device_put(pipe->display);
 		pipe->output = NULL;
-		pipe->display = NULL;
 	}
 
 	memset(&priv->channels, 0, sizeof(priv->channels));
@@ -150,33 +152,17 @@ static void omap_disconnect_pipelines(struct drm_device *ddev)
 	priv->num_pipes = 0;
 }
 
-static int omap_compare_pipes(const void *a, const void *b)
-{
-	const struct omap_drm_pipeline *pipe1 = a;
-	const struct omap_drm_pipeline *pipe2 = b;
-
-	if (pipe1->display->alias_id > pipe2->display->alias_id)
-		return 1;
-	else if (pipe1->display->alias_id < pipe2->display->alias_id)
-		return -1;
-	return 0;
-}
-
 static int omap_connect_pipelines(struct drm_device *ddev)
 {
 	struct omap_drm_private *priv = ddev->dev_private;
 	struct omap_dss_device *output = NULL;
-	unsigned int i;
 	int r;
 
-	if (!omapdss_stack_is_ready())
-		return -EPROBE_DEFER;
-
 	for_each_dss_output(output) {
 		r = omapdss_device_connect(priv->dss, NULL, output);
 		if (r == -EPROBE_DEFER) {
 			omapdss_device_put(output);
-			goto cleanup;
+			return r;
 		} else if (r) {
 			dev_warn(output->dev, "could not connect output %s\n",
 				 output->name);
@@ -185,7 +171,6 @@ static int omap_connect_pipelines(struct drm_device *ddev)
 
 			pipe = &priv->pipes[priv->num_pipes++];
 			pipe->output = omapdss_device_get(output);
-			pipe->display = omapdss_display_get(output);
 
 			if (priv->num_pipes == ARRAY_SIZE(priv->pipes)) {
 				/* To balance the 'for_each_dss_output' loop */
@@ -195,36 +180,19 @@ static int omap_connect_pipelines(struct drm_device *ddev)
 		}
 	}
 
-	/* Sort the list by DT aliases */
-	sort(priv->pipes, priv->num_pipes, sizeof(priv->pipes[0]),
-	     omap_compare_pipes, NULL);
-
-	/*
-	 * Populate the pipeline lookup table by DISPC channel. Only one display
-	 * is allowed per channel.
-	 */
-	for (i = 0; i < priv->num_pipes; ++i) {
-		struct omap_drm_pipeline *pipe = &priv->pipes[i];
-		enum omap_channel channel = pipe->output->dispc_channel;
-
-		if (WARN_ON(priv->channels[channel] != NULL)) {
-			r = -EINVAL;
-			goto cleanup;
-		}
-
-		priv->channels[channel] = pipe;
-	}
-
 	return 0;
+}
 
-cleanup:
-	/*
-	 * if we are deferring probe, we disconnect the devices we previously
-	 * connected
-	 */
-	omap_disconnect_pipelines(ddev);
+static int omap_compare_pipelines(const void *a, const void *b)
+{
+	const struct omap_drm_pipeline *pipe1 = a;
+	const struct omap_drm_pipeline *pipe2 = b;
 
-	return r;
+	if (pipe1->alias_id > pipe2->alias_id)
+		return 1;
+	else if (pipe1->alias_id < pipe2->alias_id)
+		return -1;
+	return 0;
 }
 
 static int omap_modeset_init_properties(struct drm_device *dev)
@@ -240,6 +208,30 @@ static int omap_modeset_init_properties(struct drm_device *dev)
 	return 0;
 }
 
+static int omap_display_id(struct omap_dss_device *output)
+{
+	struct device_node *node = NULL;
+
+	if (output->next) {
+		struct omap_dss_device *display;
+
+		display = omapdss_display_get(output);
+		node = display->dev->of_node;
+		omapdss_device_put(display);
+	} else if (output->bridge) {
+		struct drm_bridge *bridge = output->bridge;
+
+		while (bridge->next)
+			bridge = bridge->next;
+
+		node = bridge->of_node;
+	} else if (output->panel) {
+		node = output->panel->dev->of_node;
+	}
+
+	return node ? of_alias_get_id(node, "display") : -ENODEV;
+}
+
 static int omap_modeset_init(struct drm_device *dev)
 {
 	struct omap_drm_private *priv = dev->dev_private;
@@ -249,6 +241,9 @@ static int omap_modeset_init(struct drm_device *dev)
 	int ret;
 	u32 plane_crtc_mask;
 
+	if (!omapdss_stack_is_ready())
+		return -EPROBE_DEFER;
+
 	drm_mode_config_init(dev);
 
 	ret = omap_modeset_init_properties(dev);
@@ -263,6 +258,10 @@ static int omap_modeset_init(struct drm_device *dev)
 	 * configuration does not match the expectations or exceeds
 	 * the available resources, the configuration is rejected.
 	 */
+	ret = omap_connect_pipelines(dev);
+	if (ret < 0)
+		return ret;
+
 	if (priv->num_pipes > num_mgrs || priv->num_pipes > num_ovls) {
 		dev_err(dev->dev, "%s(): Too many connected displays\n",
 			__func__);
@@ -288,33 +287,75 @@ static int omap_modeset_init(struct drm_device *dev)
 		priv->planes[priv->num_planes++] = plane;
 	}
 
-	/* Create the CRTCs, encoders and connectors. */
+	/*
+	 * Create the encoders, attach the bridges and get the pipeline alias
+	 * IDs.
+	 */
 	for (i = 0; i < priv->num_pipes; i++) {
 		struct omap_drm_pipeline *pipe = &priv->pipes[i];
-		struct omap_dss_device *display = pipe->display;
-		struct drm_connector *connector;
-		struct drm_encoder *encoder;
-		struct drm_crtc *crtc;
+		int id;
 
-		encoder = omap_encoder_init(dev, pipe->output, display);
-		if (!encoder)
+		pipe->encoder = omap_encoder_init(dev, pipe->output);
+		if (!pipe->encoder)
 			return -ENOMEM;
 
-		connector = omap_connector_init(dev, pipe->output, display,
-						encoder);
-		if (!connector)
-			return -ENOMEM;
+		if (pipe->output->bridge) {
+			ret = drm_bridge_attach(pipe->encoder,
+						pipe->output->bridge, NULL);
+			if (ret < 0)
+				return ret;
+		}
+
+		id = omap_display_id(pipe->output);
+		pipe->alias_id = id >= 0 ? id : i;
+	}
+
+	/* Sort the pipelines by DT aliases. */
+	sort(priv->pipes, priv->num_pipes, sizeof(priv->pipes[0]),
+	     omap_compare_pipelines, NULL);
+
+	/*
+	 * Populate the pipeline lookup table by DISPC channel. Only one display
+	 * is allowed per channel.
+	 */
+	for (i = 0; i < priv->num_pipes; ++i) {
+		struct omap_drm_pipeline *pipe = &priv->pipes[i];
+		enum omap_channel channel = pipe->output->dispc_channel;
+
+		if (WARN_ON(priv->channels[channel] != NULL))
+			return -EINVAL;
+
+		priv->channels[channel] = pipe;
+	}
+
+	/* Create the connectors and CRTCs. */
+	for (i = 0; i < priv->num_pipes; i++) {
+		struct omap_drm_pipeline *pipe = &priv->pipes[i];
+		struct drm_encoder *encoder = pipe->encoder;
+		struct drm_crtc *crtc;
+
+		if (!pipe->output->bridge) {
+			pipe->connector = omap_connector_init(dev, pipe->output,
+							      encoder);
+			if (!pipe->connector)
+				return -ENOMEM;
+
+			drm_connector_attach_encoder(pipe->connector, encoder);
+
+			if (pipe->output->panel) {
+				ret = drm_panel_attach(pipe->output->panel,
+						       pipe->connector);
+				if (ret < 0)
+					return ret;
+			}
+		}
 
 		crtc = omap_crtc_init(dev, pipe, priv->planes[i]);
 		if (IS_ERR(crtc))
 			return PTR_ERR(crtc);
 
-		drm_connector_attach_encoder(connector, encoder);
 		encoder->possible_crtcs = 1 << i;
-
 		pipe->crtc = crtc;
-		pipe->encoder = encoder;
-		pipe->connector = connector;
 	}
 
 	DBG("registered %u planes, %u crtcs/encoders/connectors\n",
@@ -351,10 +392,12 @@ static int omap_modeset_init(struct drm_device *dev)
 static void omap_modeset_enable_external_hpd(struct drm_device *ddev)
 {
 	struct omap_drm_private *priv = ddev->dev_private;
-	int i;
+	unsigned int i;
 
-	for (i = 0; i < priv->num_pipes; i++)
-		omap_connector_enable_hpd(priv->pipes[i].connector);
+	for (i = 0; i < priv->num_pipes; i++) {
+		if (priv->pipes[i].connector)
+			omap_connector_enable_hpd(priv->pipes[i].connector);
+	}
 }
 
 /*
@@ -363,10 +406,12 @@ static void omap_modeset_enable_external_hpd(struct drm_device *ddev)
 static void omap_modeset_disable_external_hpd(struct drm_device *ddev)
 {
 	struct omap_drm_private *priv = ddev->dev_private;
-	int i;
+	unsigned int i;
 
-	for (i = 0; i < priv->num_pipes; i++)
-		omap_connector_disable_hpd(priv->pipes[i].connector);
+	for (i = 0; i < priv->num_pipes; i++) {
+		if (priv->pipes[i].connector)
+			omap_connector_disable_hpd(priv->pipes[i].connector);
+	}
 }
 
 /*
@@ -551,10 +596,6 @@ static int omapdrm_init(struct omap_drm_private *priv, struct device *dev)
 
 	omap_crtc_pre_init(priv);
 
-	ret = omap_connect_pipelines(ddev);
-	if (ret)
-		goto err_crtc_uninit;
-
 	soc = soc_device_match(omapdrm_soc_devices);
 	priv->omaprev = soc ? (unsigned int)soc->data : 0;
 	priv->wq = alloc_ordered_workqueue("omapdrm", 0);
@@ -612,7 +653,6 @@ err_gem_deinit:
 	omap_gem_deinit(ddev);
 	destroy_workqueue(priv->wq);
 	omap_disconnect_pipelines(ddev);
-err_crtc_uninit:
 	omap_crtc_pre_uninit(priv);
 	drm_dev_put(ddev);
 	return ret;
@@ -685,54 +725,12 @@ static int pdev_remove(struct platform_device *pdev)
 }
 
 #ifdef CONFIG_PM_SLEEP
-static int omap_drm_suspend_all_displays(struct drm_device *ddev)
-{
-	struct omap_drm_private *priv = ddev->dev_private;
-	int i;
-
-	for (i = 0; i < priv->num_pipes; i++) {
-		struct omap_dss_device *display = priv->pipes[i].display;
-
-		if (display->state == OMAP_DSS_DISPLAY_ACTIVE) {
-			display->ops->disable(display);
-			display->activate_after_resume = true;
-		} else {
-			display->activate_after_resume = false;
-		}
-	}
-
-	return 0;
-}
-
-static int omap_drm_resume_all_displays(struct drm_device *ddev)
-{
-	struct omap_drm_private *priv = ddev->dev_private;
-	int i;
-
-	for (i = 0; i < priv->num_pipes; i++) {
-		struct omap_dss_device *display = priv->pipes[i].display;
-
-		if (display->activate_after_resume) {
-			display->ops->enable(display);
-			display->activate_after_resume = false;
-		}
-	}
-
-	return 0;
-}
-
 static int omap_drm_suspend(struct device *dev)
 {
 	struct omap_drm_private *priv = dev_get_drvdata(dev);
 	struct drm_device *drm_dev = priv->ddev;
 
-	drm_kms_helper_poll_disable(drm_dev);
-
-	drm_modeset_lock_all(drm_dev);
-	omap_drm_suspend_all_displays(drm_dev);
-	drm_modeset_unlock_all(drm_dev);
-
-	return 0;
+	return drm_mode_config_helper_suspend(drm_dev);
 }
 
 static int omap_drm_resume(struct device *dev)
@@ -740,11 +738,7 @@ static int omap_drm_resume(struct device *dev)
 	struct omap_drm_private *priv = dev_get_drvdata(dev);
 	struct drm_device *drm_dev = priv->ddev;
 
-	drm_modeset_lock_all(drm_dev);
-	omap_drm_resume_all_displays(drm_dev);
-	drm_modeset_unlock_all(drm_dev);
-
-	drm_kms_helper_poll_enable(drm_dev);
+	drm_mode_config_helper_resume(drm_dev);
 
 	return omap_gem_resume(drm_dev);
 }
diff --git a/drivers/gpu/drm/omapdrm/omap_drv.h b/drivers/gpu/drm/omapdrm/omap_drv.h
index 0c57d2814c51..3cca45cb25f3 100644
--- a/drivers/gpu/drm/omapdrm/omap_drv.h
+++ b/drivers/gpu/drm/omapdrm/omap_drv.h
@@ -49,7 +49,7 @@ struct omap_drm_pipeline {
 	struct drm_encoder *encoder;
 	struct drm_connector *connector;
 	struct omap_dss_device *output;
-	struct omap_dss_device *display;
+	unsigned int alias_id;
 };
 
 struct omap_drm_private {
diff --git a/drivers/gpu/drm/omapdrm/omap_encoder.c b/drivers/gpu/drm/omapdrm/omap_encoder.c
index 0d85b3a35767..40512419642b 100644
--- a/drivers/gpu/drm/omapdrm/omap_encoder.c
+++ b/drivers/gpu/drm/omapdrm/omap_encoder.c
@@ -20,6 +20,7 @@
 #include <drm/drm_crtc.h>
 #include <drm/drm_modeset_helper_vtables.h>
 #include <drm/drm_edid.h>
+#include <drm/drm_panel.h>
 
 #include "omap_drv.h"
 
@@ -37,7 +38,6 @@
 struct omap_encoder {
 	struct drm_encoder base;
 	struct omap_dss_device *output;
-	struct omap_dss_device *display;
 };
 
 static void omap_encoder_destroy(struct drm_encoder *encoder)
@@ -52,22 +52,43 @@ static const struct drm_encoder_funcs omap_encoder_funcs = {
 	.destroy = omap_encoder_destroy,
 };
 
-static void omap_encoder_hdmi_mode_set(struct drm_encoder *encoder,
+static void omap_encoder_update_videomode_flags(struct videomode *vm,
+						u32 bus_flags)
+{
+	if (!(vm->flags & (DISPLAY_FLAGS_DE_LOW |
+			   DISPLAY_FLAGS_DE_HIGH))) {
+		if (bus_flags & DRM_BUS_FLAG_DE_LOW)
+			vm->flags |= DISPLAY_FLAGS_DE_LOW;
+		else if (bus_flags & DRM_BUS_FLAG_DE_HIGH)
+			vm->flags |= DISPLAY_FLAGS_DE_HIGH;
+	}
+
+	if (!(vm->flags & (DISPLAY_FLAGS_PIXDATA_POSEDGE |
+			   DISPLAY_FLAGS_PIXDATA_NEGEDGE))) {
+		if (bus_flags & DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE)
+			vm->flags |= DISPLAY_FLAGS_PIXDATA_POSEDGE;
+		else if (bus_flags & DRM_BUS_FLAG_PIXDATA_DRIVE_NEGEDGE)
+			vm->flags |= DISPLAY_FLAGS_PIXDATA_NEGEDGE;
+	}
+
+	if (!(vm->flags & (DISPLAY_FLAGS_SYNC_POSEDGE |
+			   DISPLAY_FLAGS_SYNC_NEGEDGE))) {
+		if (bus_flags & DRM_BUS_FLAG_SYNC_DRIVE_POSEDGE)
+			vm->flags |= DISPLAY_FLAGS_SYNC_POSEDGE;
+		else if (bus_flags & DRM_BUS_FLAG_SYNC_DRIVE_NEGEDGE)
+			vm->flags |= DISPLAY_FLAGS_SYNC_NEGEDGE;
+	}
+}
+
+static void omap_encoder_hdmi_mode_set(struct drm_connector *connector,
+				       struct drm_encoder *encoder,
 				       struct drm_display_mode *adjusted_mode)
 {
-	struct drm_device *dev = encoder->dev;
 	struct omap_encoder *omap_encoder = to_omap_encoder(encoder);
 	struct omap_dss_device *dssdev = omap_encoder->output;
-	struct drm_connector *connector;
 	bool hdmi_mode;
 
-	hdmi_mode = false;
-	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
-		if (connector->encoder == encoder) {
-			hdmi_mode = omap_connector_get_hdmi_mode(connector);
-			break;
-		}
-	}
+	hdmi_mode = omap_connector_get_hdmi_mode(connector);
 
 	if (dssdev->ops->hdmi.set_hdmi_mode)
 		dssdev->ops->hdmi.set_hdmi_mode(dssdev, hdmi_mode);
@@ -88,8 +109,18 @@ static void omap_encoder_mode_set(struct drm_encoder *encoder,
 				  struct drm_display_mode *adjusted_mode)
 {
 	struct omap_encoder *omap_encoder = to_omap_encoder(encoder);
+	struct omap_dss_device *output = omap_encoder->output;
 	struct omap_dss_device *dssdev;
+	struct drm_device *dev = encoder->dev;
+	struct drm_connector *connector;
+	struct drm_bridge *bridge;
 	struct videomode vm = { 0 };
+	u32 bus_flags;
+
+	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+		if (connector->encoder == encoder)
+			break;
+	}
 
 	drm_display_mode_to_videomode(adjusted_mode, &vm);
 
@@ -102,66 +133,102 @@ static void omap_encoder_mode_set(struct drm_encoder *encoder,
 	 *
 	 * A better solution is to use DRM's bus-flags through the whole driver.
 	 */
-	for (dssdev = omap_encoder->output; dssdev; dssdev = dssdev->next) {
-		unsigned long bus_flags = dssdev->bus_flags;
-
-		if (!(vm.flags & (DISPLAY_FLAGS_DE_LOW |
-				  DISPLAY_FLAGS_DE_HIGH))) {
-			if (bus_flags & DRM_BUS_FLAG_DE_LOW)
-				vm.flags |= DISPLAY_FLAGS_DE_LOW;
-			else if (bus_flags & DRM_BUS_FLAG_DE_HIGH)
-				vm.flags |= DISPLAY_FLAGS_DE_HIGH;
-		}
-
-		if (!(vm.flags & (DISPLAY_FLAGS_PIXDATA_POSEDGE |
-				  DISPLAY_FLAGS_PIXDATA_NEGEDGE))) {
-			if (bus_flags & DRM_BUS_FLAG_PIXDATA_POSEDGE)
-				vm.flags |= DISPLAY_FLAGS_PIXDATA_POSEDGE;
-			else if (bus_flags & DRM_BUS_FLAG_PIXDATA_NEGEDGE)
-				vm.flags |= DISPLAY_FLAGS_PIXDATA_NEGEDGE;
-		}
-
-		if (!(vm.flags & (DISPLAY_FLAGS_SYNC_POSEDGE |
-				  DISPLAY_FLAGS_SYNC_NEGEDGE))) {
-			if (bus_flags & DRM_BUS_FLAG_SYNC_POSEDGE)
-				vm.flags |= DISPLAY_FLAGS_SYNC_POSEDGE;
-			else if (bus_flags & DRM_BUS_FLAG_SYNC_NEGEDGE)
-				vm.flags |= DISPLAY_FLAGS_SYNC_NEGEDGE;
-		}
+	for (dssdev = output; dssdev; dssdev = dssdev->next)
+		omap_encoder_update_videomode_flags(&vm, dssdev->bus_flags);
+
+	for (bridge = output->bridge; bridge; bridge = bridge->next) {
+		if (!bridge->timings)
+			continue;
+
+		bus_flags = bridge->timings->input_bus_flags;
+		omap_encoder_update_videomode_flags(&vm, bus_flags);
 	}
 
+	bus_flags = connector->display_info.bus_flags;
+	omap_encoder_update_videomode_flags(&vm, bus_flags);
+
 	/* Set timings for all devices in the display pipeline. */
-	dss_mgr_set_timings(omap_encoder->output, &vm);
+	dss_mgr_set_timings(output, &vm);
 
-	for (dssdev = omap_encoder->output; dssdev; dssdev = dssdev->next) {
+	for (dssdev = output; dssdev; dssdev = dssdev->next) {
 		if (dssdev->ops->set_timings)
-			dssdev->ops->set_timings(dssdev, &vm);
+			dssdev->ops->set_timings(dssdev, adjusted_mode);
 	}
 
 	/* Set the HDMI mode and HDMI infoframe if applicable. */
-	if (omap_encoder->output->output_type == OMAP_DISPLAY_TYPE_HDMI)
-		omap_encoder_hdmi_mode_set(encoder, adjusted_mode);
+	if (output->type == OMAP_DISPLAY_TYPE_HDMI)
+		omap_encoder_hdmi_mode_set(connector, encoder, adjusted_mode);
 }
 
 static void omap_encoder_disable(struct drm_encoder *encoder)
 {
 	struct omap_encoder *omap_encoder = to_omap_encoder(encoder);
-	struct omap_dss_device *dssdev = omap_encoder->display;
+	struct omap_dss_device *dssdev = omap_encoder->output;
+	struct drm_device *dev = encoder->dev;
+
+	dev_dbg(dev->dev, "disable(%s)\n", dssdev->name);
+
+	/* Disable the panel if present. */
+	if (dssdev->panel) {
+		drm_panel_disable(dssdev->panel);
+		drm_panel_unprepare(dssdev->panel);
+	}
+
+	/*
+	 * Disable the chain of external devices, starting at the one at the
+	 * internal encoder's output.
+	 */
+	omapdss_device_disable(dssdev->next);
+
+	/*
+	 * Disable the internal encoder. This will disable the DSS output. The
+	 * DSI is treated as an exception as DSI pipelines still use the legacy
+	 * flow where the pipeline output controls the encoder.
+	 */
+	if (dssdev->type != OMAP_DISPLAY_TYPE_DSI) {
+		dssdev->ops->disable(dssdev);
+		dssdev->state = OMAP_DSS_DISPLAY_DISABLED;
+	}
 
-	dssdev->ops->disable(dssdev);
+	/*
+	 * Perform the post-disable operations on the chain of external devices
+	 * to complete the display pipeline disable.
+	 */
+	omapdss_device_post_disable(dssdev->next);
 }
 
 static void omap_encoder_enable(struct drm_encoder *encoder)
 {
 	struct omap_encoder *omap_encoder = to_omap_encoder(encoder);
-	struct omap_dss_device *dssdev = omap_encoder->display;
-	int r;
-
-	r = dssdev->ops->enable(dssdev);
-	if (r)
-		dev_err(encoder->dev->dev,
-			"Failed to enable display '%s': %d\n",
-			dssdev->name, r);
+	struct omap_dss_device *dssdev = omap_encoder->output;
+	struct drm_device *dev = encoder->dev;
+
+	dev_dbg(dev->dev, "enable(%s)\n", dssdev->name);
+
+	/* Prepare the chain of external devices for pipeline enable. */
+	omapdss_device_pre_enable(dssdev->next);
+
+	/*
+	 * Enable the internal encoder. This will enable the DSS output. The
+	 * DSI is treated as an exception as DSI pipelines still use the legacy
+	 * flow where the pipeline output controls the encoder.
+	 */
+	if (dssdev->type != OMAP_DISPLAY_TYPE_DSI) {
+		dssdev->ops->enable(dssdev);
+		dssdev->state = OMAP_DSS_DISPLAY_ACTIVE;
+	}
+
+	/*
+	 * Enable the chain of external devices, starting at the one at the
+	 * internal encoder's output.
+	 */
+	omapdss_device_enable(dssdev->next);
+
+	/* Enable the panel if present. */
+	if (dssdev->panel) {
+		drm_panel_prepare(dssdev->panel);
+		drm_panel_enable(dssdev->panel);
+	}
 }
 
 static int omap_encoder_atomic_check(struct drm_encoder *encoder,
@@ -169,35 +236,17 @@ static int omap_encoder_atomic_check(struct drm_encoder *encoder,
 				     struct drm_connector_state *conn_state)
 {
 	struct omap_encoder *omap_encoder = to_omap_encoder(encoder);
-	enum omap_channel channel = omap_encoder->output->dispc_channel;
-	struct drm_device *dev = encoder->dev;
-	struct omap_drm_private *priv = dev->dev_private;
-	struct omap_dss_device *dssdev;
-	struct videomode vm = { 0 };
-	int ret;
-
-	drm_display_mode_to_videomode(&crtc_state->mode, &vm);
-
-	ret = priv->dispc_ops->mgr_check_timings(priv->dispc, channel, &vm);
-	if (ret)
-		goto done;
-
-	for (dssdev = omap_encoder->output; dssdev; dssdev = dssdev->next) {
-		if (!dssdev->ops->check_timings)
-			continue;
-
-		ret = dssdev->ops->check_timings(dssdev, &vm);
-		if (ret)
-			goto done;
+	enum drm_mode_status status;
+
+	status = omap_connector_mode_fixup(omap_encoder->output,
+					   &crtc_state->mode,
+					   &crtc_state->adjusted_mode);
+	if (status != MODE_OK) {
+		dev_err(encoder->dev->dev, "invalid timings: %d\n", status);
+		return -EINVAL;
 	}
 
-	drm_display_mode_from_videomode(&vm, &crtc_state->adjusted_mode);
-
-done:
-	if (ret)
-		dev_err(dev->dev, "invalid timings: %d\n", ret);
-
-	return ret;
+	return 0;
 }
 
 static const struct drm_encoder_helper_funcs omap_encoder_helper_funcs = {
@@ -209,8 +258,7 @@ static const struct drm_encoder_helper_funcs omap_encoder_helper_funcs = {
 
 /* initialize encoder */
 struct drm_encoder *omap_encoder_init(struct drm_device *dev,
-				      struct omap_dss_device *output,
-				      struct omap_dss_device *display)
+				      struct omap_dss_device *output)
 {
 	struct drm_encoder *encoder = NULL;
 	struct omap_encoder *omap_encoder;
@@ -220,7 +268,6 @@ struct drm_encoder *omap_encoder_init(struct drm_device *dev,
 		goto fail;
 
 	omap_encoder->output = output;
-	omap_encoder->display = display;
 
 	encoder = &omap_encoder->base;
 
diff --git a/drivers/gpu/drm/omapdrm/omap_encoder.h b/drivers/gpu/drm/omapdrm/omap_encoder.h
index a7b5dde63ecb..4aefb3142886 100644
--- a/drivers/gpu/drm/omapdrm/omap_encoder.h
+++ b/drivers/gpu/drm/omapdrm/omap_encoder.h
@@ -25,7 +25,6 @@ struct drm_encoder;
 struct omap_dss_device;
 
 struct drm_encoder *omap_encoder_init(struct drm_device *dev,
-				      struct omap_dss_device *output,
-				      struct omap_dss_device *display);
+				      struct omap_dss_device *output);
 
 #endif /* __OMAPDRM_ENCODER_H__ */
diff --git a/drivers/gpu/drm/omapdrm/omap_fbdev.c b/drivers/gpu/drm/omapdrm/omap_fbdev.c
index 851c59f07eb1..50aabd854f4d 100644
--- a/drivers/gpu/drm/omapdrm/omap_fbdev.c
+++ b/drivers/gpu/drm/omapdrm/omap_fbdev.c
@@ -183,13 +183,9 @@ static int omap_fbdev_create(struct drm_fb_helper *helper,
 	fbdev->fb = fb;
 	helper->fb = fb;
 
-	fbi->par = helper;
 	fbi->fbops = &omap_fb_ops;
 
-	strcpy(fbi->fix.id, MODULE_NAME);
-
-	drm_fb_helper_fill_fix(fbi, fb->pitches[0], fb->format->depth);
-	drm_fb_helper_fill_var(fbi, helper, sizes->fb_width, sizes->fb_height);
+	drm_fb_helper_fill_info(fbi, helper, sizes);
 
 	dev->mode_config.fb_base = dma_addr;
 
diff --git a/drivers/gpu/drm/panel/Kconfig b/drivers/gpu/drm/panel/Kconfig
index 3e070153ef21..e36dbb4df867 100644
--- a/drivers/gpu/drm/panel/Kconfig
+++ b/drivers/gpu/drm/panel/Kconfig
@@ -38,6 +38,15 @@ config DRM_PANEL_SIMPLE
 	  that it can be automatically turned off when the panel goes into a
 	  low power state.
 
+config DRM_PANEL_FEIYANG_FY07024DI26A30D
+	tristate "Feiyang FY07024DI26A30-D MIPI-DSI LCD panel"
+	depends on OF
+	depends on DRM_MIPI_DSI
+	depends on BACKLIGHT_CLASS_DEVICE
+	help
+	  Say Y if you want to enable support for panels based on the
+	  Feiyang FY07024DI26A30-D MIPI-DSI interface.
+
 config DRM_PANEL_ILITEK_IL9322
 	tristate "Ilitek ILI9322 320x240 QVGA panels"
 	depends on OF && SPI
@@ -149,6 +158,28 @@ config DRM_PANEL_RAYDIUM_RM68200
 	  Say Y here if you want to enable support for Raydium RM68200
 	  720x1280 DSI video mode panel.
 
+config DRM_PANEL_ROCKTECH_JH057N00900
+	tristate "Rocktech JH057N00900 MIPI touchscreen panel"
+	depends on OF
+	depends on DRM_MIPI_DSI
+	depends on BACKLIGHT_CLASS_DEVICE
+	help
+	  Say Y here if you want to enable support for Rocktech JH057N00900
+	  MIPI DSI panel as e.g. used in the Librem 5 devkit. It has a
+	  resolution of 720x1440 pixels, a built in backlight and touch
+	  controller.
+	  Touch input support is provided by the goodix driver and needs to be
+	  selected separately.
+
+config DRM_PANEL_RONBO_RB070D30
+	tristate "Ronbo Electronics RB070D30 panel"
+	depends on OF
+	depends on DRM_MIPI_DSI
+	depends on BACKLIGHT_CLASS_DEVICE
+	help
+	  Say Y here if you want to enable support for Ronbo Electronics
+	  RB070D30 1024x600 DSI panel.
+
 config DRM_PANEL_SAMSUNG_S6D16D0
 	tristate "Samsung S6D16D0 DSI video mode panel"
 	depends on OF
diff --git a/drivers/gpu/drm/panel/Makefile b/drivers/gpu/drm/panel/Makefile
index e7ab71968bbf..78e3dc376bdd 100644
--- a/drivers/gpu/drm/panel/Makefile
+++ b/drivers/gpu/drm/panel/Makefile
@@ -2,6 +2,7 @@
 obj-$(CONFIG_DRM_PANEL_ARM_VERSATILE) += panel-arm-versatile.o
 obj-$(CONFIG_DRM_PANEL_LVDS) += panel-lvds.o
 obj-$(CONFIG_DRM_PANEL_SIMPLE) += panel-simple.o
+obj-$(CONFIG_DRM_PANEL_FEIYANG_FY07024DI26A30D) += panel-feiyang-fy07024di26a30d.o
 obj-$(CONFIG_DRM_PANEL_ILITEK_IL9322) += panel-ilitek-ili9322.o
 obj-$(CONFIG_DRM_PANEL_ILITEK_ILI9881C) += panel-ilitek-ili9881c.o
 obj-$(CONFIG_DRM_PANEL_INNOLUX_P079ZCA) += panel-innolux-p079zca.o
@@ -13,6 +14,8 @@ obj-$(CONFIG_DRM_PANEL_ORISETECH_OTM8009A) += panel-orisetech-otm8009a.o
 obj-$(CONFIG_DRM_PANEL_PANASONIC_VVX10F034N00) += panel-panasonic-vvx10f034n00.o
 obj-$(CONFIG_DRM_PANEL_RASPBERRYPI_TOUCHSCREEN) += panel-raspberrypi-touchscreen.o
 obj-$(CONFIG_DRM_PANEL_RAYDIUM_RM68200) += panel-raydium-rm68200.o
+obj-$(CONFIG_DRM_PANEL_ROCKTECH_JH057N00900) += panel-rocktech-jh057n00900.o
+obj-$(CONFIG_DRM_PANEL_RONBO_RB070D30) += panel-ronbo-rb070d30.o
 obj-$(CONFIG_DRM_PANEL_SAMSUNG_LD9040) += panel-samsung-ld9040.o
 obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6D16D0) += panel-samsung-s6d16d0.o
 obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6E3HA2) += panel-samsung-s6e3ha2.o
diff --git a/drivers/gpu/drm/panel/panel-arm-versatile.c b/drivers/gpu/drm/panel/panel-arm-versatile.c
index b428c4678106..a79908dfa3c8 100644
--- a/drivers/gpu/drm/panel/panel-arm-versatile.c
+++ b/drivers/gpu/drm/panel/panel-arm-versatile.c
@@ -191,7 +191,7 @@ static const struct versatile_panel_type versatile_panels[] = {
 			.vrefresh = 390,
 			.flags = DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC,
 		},
-		.bus_flags = DRM_BUS_FLAG_PIXDATA_NEGEDGE,
+		.bus_flags = DRM_BUS_FLAG_PIXDATA_DRIVE_NEGEDGE,
 	},
 	/*
 	 * Sanyo ALR252RGT 240x320 portrait display found on the
@@ -215,7 +215,7 @@ static const struct versatile_panel_type versatile_panels[] = {
 			.vrefresh = 116,
 			.flags = DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC,
 		},
-		.bus_flags = DRM_BUS_FLAG_PIXDATA_NEGEDGE,
+		.bus_flags = DRM_BUS_FLAG_PIXDATA_DRIVE_NEGEDGE,
 		.ib2 = true,
 	},
 };
@@ -264,8 +264,6 @@ static int versatile_panel_get_modes(struct drm_panel *panel)
 	struct versatile_panel *vpanel = to_versatile_panel(panel);
 	struct drm_display_mode *mode;
 
-	strncpy(connector->display_info.name, vpanel->panel_type->name,
-		DRM_DISPLAY_INFO_LEN);
 	connector->display_info.width_mm = vpanel->panel_type->width_mm;
 	connector->display_info.height_mm = vpanel->panel_type->height_mm;
 	connector->display_info.bus_flags = vpanel->panel_type->bus_flags;
diff --git a/drivers/gpu/drm/panel/panel-feiyang-fy07024di26a30d.c b/drivers/gpu/drm/panel/panel-feiyang-fy07024di26a30d.c
new file mode 100644
index 000000000000..dabf59e0f56f
--- /dev/null
+++ b/drivers/gpu/drm/panel/panel-feiyang-fy07024di26a30d.c
@@ -0,0 +1,272 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ * Copyright (C) 2018 Amarula Solutions
+ * Author: Jagan Teki <jagan@amarulasolutions.com>
+ */
+
+#include <drm/drm_mipi_dsi.h>
+#include <drm/drm_modes.h>
+#include <drm/drm_panel.h>
+#include <drm/drm_print.h>
+
+#include <linux/backlight.h>
+#include <linux/gpio/consumer.h>
+#include <linux/delay.h>
+#include <linux/module.h>
+#include <linux/of_device.h>
+#include <linux/regulator/consumer.h>
+
+#define FEIYANG_INIT_CMD_LEN	2
+
+struct feiyang {
+	struct drm_panel	panel;
+	struct mipi_dsi_device	*dsi;
+
+	struct backlight_device	*backlight;
+	struct regulator	*dvdd;
+	struct regulator	*avdd;
+	struct gpio_desc	*reset;
+};
+
+static inline struct feiyang *panel_to_feiyang(struct drm_panel *panel)
+{
+	return container_of(panel, struct feiyang, panel);
+}
+
+struct feiyang_init_cmd {
+	u8 data[FEIYANG_INIT_CMD_LEN];
+};
+
+static const struct feiyang_init_cmd feiyang_init_cmds[] = {
+	{ .data = { 0x80, 0x58 } },
+	{ .data = { 0x81, 0x47 } },
+	{ .data = { 0x82, 0xD4 } },
+	{ .data = { 0x83, 0x88 } },
+	{ .data = { 0x84, 0xA9 } },
+	{ .data = { 0x85, 0xC3 } },
+	{ .data = { 0x86, 0x82 } },
+};
+
+static int feiyang_prepare(struct drm_panel *panel)
+{
+	struct feiyang *ctx = panel_to_feiyang(panel);
+	struct mipi_dsi_device *dsi = ctx->dsi;
+	unsigned int i;
+	int ret;
+
+	ret = regulator_enable(ctx->dvdd);
+	if (ret)
+		return ret;
+
+	/* T1 (dvdd start + dvdd rise) 0 < T1 <= 10ms */
+	msleep(10);
+
+	ret = regulator_enable(ctx->avdd);
+	if (ret)
+		return ret;
+
+	/* T3 (dvdd rise + avdd start + avdd rise) T3 >= 20ms */
+	msleep(20);
+
+	gpiod_set_value(ctx->reset, 0);
+
+	/*
+	 * T5 + T6 (avdd rise + video & logic signal rise)
+	 * T5 >= 10ms, 0 < T6 <= 10ms
+	 */
+	msleep(20);
+
+	gpiod_set_value(ctx->reset, 1);
+
+	/* T12 (video & logic signal rise + backlight rise) T12 >= 200ms */
+	msleep(200);
+
+	for (i = 0; i < ARRAY_SIZE(feiyang_init_cmds); i++) {
+		const struct feiyang_init_cmd *cmd =
+						&feiyang_init_cmds[i];
+
+		ret = mipi_dsi_dcs_write_buffer(dsi, cmd->data,
+						FEIYANG_INIT_CMD_LEN);
+		if (ret < 0)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int feiyang_enable(struct drm_panel *panel)
+{
+	struct feiyang *ctx = panel_to_feiyang(panel);
+
+	/* T12 (video & logic signal rise + backlight rise) T12 >= 200ms */
+	msleep(200);
+
+	mipi_dsi_dcs_set_display_on(ctx->dsi);
+	backlight_enable(ctx->backlight);
+
+	return 0;
+}
+
+static int feiyang_disable(struct drm_panel *panel)
+{
+	struct feiyang *ctx = panel_to_feiyang(panel);
+
+	backlight_disable(ctx->backlight);
+	return mipi_dsi_dcs_set_display_off(ctx->dsi);
+}
+
+static int feiyang_unprepare(struct drm_panel *panel)
+{
+	struct feiyang *ctx = panel_to_feiyang(panel);
+	int ret;
+
+	ret = mipi_dsi_dcs_set_display_off(ctx->dsi);
+	if (ret < 0)
+		DRM_DEV_ERROR(panel->dev, "failed to set display off: %d\n",
+			      ret);
+
+	ret = mipi_dsi_dcs_enter_sleep_mode(ctx->dsi);
+	if (ret < 0)
+		DRM_DEV_ERROR(panel->dev, "failed to enter sleep mode: %d\n",
+			      ret);
+
+	/* T13 (backlight fall + video & logic signal fall) T13 >= 200ms */
+	msleep(200);
+
+	gpiod_set_value(ctx->reset, 0);
+
+	regulator_disable(ctx->avdd);
+
+	/* T11 (dvdd rise to fall) 0 < T11 <= 10ms  */
+	msleep(10);
+
+	regulator_disable(ctx->dvdd);
+
+	return 0;
+}
+
+static const struct drm_display_mode feiyang_default_mode = {
+	.clock		= 55000,
+
+	.hdisplay	= 1024,
+	.hsync_start	= 1024 + 310,
+	.hsync_end	= 1024 + 310 + 20,
+	.htotal		= 1024 + 310 + 20 + 90,
+
+	.vdisplay	= 600,
+	.vsync_start	= 600 + 12,
+	.vsync_end	= 600 + 12 + 2,
+	.vtotal		= 600 + 12 + 2 + 21,
+	.vrefresh	= 60,
+
+	.type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED,
+};
+
+static int feiyang_get_modes(struct drm_panel *panel)
+{
+	struct drm_connector *connector = panel->connector;
+	struct feiyang *ctx = panel_to_feiyang(panel);
+	struct drm_display_mode *mode;
+
+	mode = drm_mode_duplicate(panel->drm, &feiyang_default_mode);
+	if (!mode) {
+		DRM_DEV_ERROR(&ctx->dsi->dev, "failed to add mode %ux%ux@%u\n",
+			      feiyang_default_mode.hdisplay,
+			      feiyang_default_mode.vdisplay,
+			      feiyang_default_mode.vrefresh);
+		return -ENOMEM;
+	}
+
+	drm_mode_set_name(mode);
+
+	drm_mode_probed_add(connector, mode);
+
+	return 1;
+}
+
+static const struct drm_panel_funcs feiyang_funcs = {
+	.disable = feiyang_disable,
+	.unprepare = feiyang_unprepare,
+	.prepare = feiyang_prepare,
+	.enable = feiyang_enable,
+	.get_modes = feiyang_get_modes,
+};
+
+static int feiyang_dsi_probe(struct mipi_dsi_device *dsi)
+{
+	struct feiyang *ctx;
+	int ret;
+
+	ctx = devm_kzalloc(&dsi->dev, sizeof(*ctx), GFP_KERNEL);
+	if (!ctx)
+		return -ENOMEM;
+
+	mipi_dsi_set_drvdata(dsi, ctx);
+	ctx->dsi = dsi;
+
+	drm_panel_init(&ctx->panel);
+	ctx->panel.dev = &dsi->dev;
+	ctx->panel.funcs = &feiyang_funcs;
+
+	ctx->dvdd = devm_regulator_get(&dsi->dev, "dvdd");
+	if (IS_ERR(ctx->dvdd)) {
+		DRM_DEV_ERROR(&dsi->dev, "Couldn't get dvdd regulator\n");
+		return PTR_ERR(ctx->dvdd);
+	}
+
+	ctx->avdd = devm_regulator_get(&dsi->dev, "avdd");
+	if (IS_ERR(ctx->avdd)) {
+		DRM_DEV_ERROR(&dsi->dev, "Couldn't get avdd regulator\n");
+		return PTR_ERR(ctx->avdd);
+	}
+
+	ctx->reset = devm_gpiod_get(&dsi->dev, "reset", GPIOD_OUT_LOW);
+	if (IS_ERR(ctx->reset)) {
+		DRM_DEV_ERROR(&dsi->dev, "Couldn't get our reset GPIO\n");
+		return PTR_ERR(ctx->reset);
+	}
+
+	ctx->backlight = devm_of_find_backlight(&dsi->dev);
+	if (IS_ERR(ctx->backlight))
+		return PTR_ERR(ctx->backlight);
+
+	ret = drm_panel_add(&ctx->panel);
+	if (ret < 0)
+		return ret;
+
+	dsi->mode_flags = MIPI_DSI_MODE_VIDEO_BURST;
+	dsi->format = MIPI_DSI_FMT_RGB888;
+	dsi->lanes = 4;
+
+	return mipi_dsi_attach(dsi);
+}
+
+static int feiyang_dsi_remove(struct mipi_dsi_device *dsi)
+{
+	struct feiyang *ctx = mipi_dsi_get_drvdata(dsi);
+
+	mipi_dsi_detach(dsi);
+	drm_panel_remove(&ctx->panel);
+
+	return 0;
+}
+
+static const struct of_device_id feiyang_of_match[] = {
+	{ .compatible = "feiyang,fy07024di26a30d", },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, feiyang_of_match);
+
+static struct mipi_dsi_driver feiyang_driver = {
+	.probe = feiyang_dsi_probe,
+	.remove = feiyang_dsi_remove,
+	.driver = {
+		.name = "feiyang-fy07024di26a30d",
+		.of_match_table = feiyang_of_match,
+	},
+};
+module_mipi_dsi_driver(feiyang_driver);
+
+MODULE_AUTHOR("Jagan Teki <jagan@amarulasolutions.com>");
+MODULE_DESCRIPTION("Feiyang FY07024DI26A30-D MIPI-DSI LCD panel");
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/panel/panel-ilitek-ili9322.c b/drivers/gpu/drm/panel/panel-ilitek-ili9322.c
index bd38bf4f1ba6..a1c4cd2940fb 100644
--- a/drivers/gpu/drm/panel/panel-ilitek-ili9322.c
+++ b/drivers/gpu/drm/panel/panel-ilitek-ili9322.c
@@ -412,11 +412,11 @@ static int ili9322_init(struct drm_panel *panel, struct ili9322 *ili)
 	if (ili->conf->dclk_active_high) {
 		reg = ILI9322_POL_DCLK;
 		connector->display_info.bus_flags |=
-			DRM_BUS_FLAG_PIXDATA_POSEDGE;
+			DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE;
 	} else {
 		reg = 0;
 		connector->display_info.bus_flags |=
-			DRM_BUS_FLAG_PIXDATA_NEGEDGE;
+			DRM_BUS_FLAG_PIXDATA_DRIVE_NEGEDGE;
 	}
 	if (ili->conf->de_active_high) {
 		reg |= ILI9322_POL_DE;
@@ -662,8 +662,6 @@ static int ili9322_get_modes(struct drm_panel *panel)
 	struct ili9322 *ili = panel_to_ili9322(panel);
 	struct drm_display_mode *mode;
 
-	strncpy(connector->display_info.name, "ILI9322 TFT LCD driver\0",
-		DRM_DISPLAY_INFO_LEN);
 	connector->display_info.width_mm = ili->conf->width_mm;
 	connector->display_info.height_mm = ili->conf->height_mm;
 
diff --git a/drivers/gpu/drm/panel/panel-olimex-lcd-olinuxino.c b/drivers/gpu/drm/panel/panel-olimex-lcd-olinuxino.c
index 5e8d4523e9ed..a1d8d92fac2b 100644
--- a/drivers/gpu/drm/panel/panel-olimex-lcd-olinuxino.c
+++ b/drivers/gpu/drm/panel/panel-olimex-lcd-olinuxino.c
@@ -190,7 +190,6 @@ static int lcd_olinuxino_get_modes(struct drm_panel *panel)
 		num++;
 	}
 
-	memcpy(connector->display_info.name, lcd_info->name, 32);
 	connector->display_info.width_mm = lcd_info->width_mm;
 	connector->display_info.height_mm = lcd_info->height_mm;
 	connector->display_info.bpc = lcd_info->bpc;
diff --git a/drivers/gpu/drm/panel/panel-orisetech-otm8009a.c b/drivers/gpu/drm/panel/panel-orisetech-otm8009a.c
index 87fa316e1d7b..f27a7e426574 100644
--- a/drivers/gpu/drm/panel/panel-orisetech-otm8009a.c
+++ b/drivers/gpu/drm/panel/panel-orisetech-otm8009a.c
@@ -67,15 +67,15 @@ struct otm8009a {
 };
 
 static const struct drm_display_mode default_mode = {
-	.clock = 32729,
+	.clock = 29700,
 	.hdisplay = 480,
-	.hsync_start = 480 + 120,
-	.hsync_end = 480 + 120 + 63,
-	.htotal = 480 + 120 + 63 + 120,
+	.hsync_start = 480 + 98,
+	.hsync_end = 480 + 98 + 32,
+	.htotal = 480 + 98 + 32 + 98,
 	.vdisplay = 800,
-	.vsync_start = 800 + 12,
-	.vsync_end = 800 + 12 + 12,
-	.vtotal = 800 + 12 + 12 + 12,
+	.vsync_start = 800 + 15,
+	.vsync_end = 800 + 15 + 10,
+	.vtotal = 800 + 15 + 10 + 14,
 	.vrefresh = 50,
 	.flags = 0,
 	.width_mm = 52,
@@ -248,6 +248,9 @@ static int otm8009a_init_sequence(struct otm8009a *ctx)
 	/* Send Command GRAM memory write (no parameters) */
 	dcs_write_seq(ctx, MIPI_DCS_WRITE_MEMORY_START);
 
+	/* Wait a short while to let the panel be ready before the 1st frame */
+	mdelay(10);
+
 	return 0;
 }
 
@@ -433,7 +436,8 @@ static int otm8009a_probe(struct mipi_dsi_device *dsi)
 	ctx->supply = devm_regulator_get(dev, "power");
 	if (IS_ERR(ctx->supply)) {
 		ret = PTR_ERR(ctx->supply);
-		dev_err(dev, "failed to request regulator: %d\n", ret);
+		if (ret != -EPROBE_DEFER)
+			dev_err(dev, "failed to request regulator: %d\n", ret);
 		return ret;
 	}
 
diff --git a/drivers/gpu/drm/panel/panel-raydium-rm68200.c b/drivers/gpu/drm/panel/panel-raydium-rm68200.c
index 77593533abcd..14186827e591 100644
--- a/drivers/gpu/drm/panel/panel-raydium-rm68200.c
+++ b/drivers/gpu/drm/panel/panel-raydium-rm68200.c
@@ -383,7 +383,8 @@ static int rm68200_probe(struct mipi_dsi_device *dsi)
 	ctx->supply = devm_regulator_get(dev, "power");
 	if (IS_ERR(ctx->supply)) {
 		ret = PTR_ERR(ctx->supply);
-		dev_err(dev, "cannot get regulator: %d\n", ret);
+		if (ret != -EPROBE_DEFER)
+			dev_err(dev, "cannot get regulator: %d\n", ret);
 		return ret;
 	}
 
diff --git a/drivers/gpu/drm/panel/panel-rocktech-jh057n00900.c b/drivers/gpu/drm/panel/panel-rocktech-jh057n00900.c
new file mode 100644
index 000000000000..d88ea8da2ec2
--- /dev/null
+++ b/drivers/gpu/drm/panel/panel-rocktech-jh057n00900.c
@@ -0,0 +1,387 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Rockteck jh057n00900 5.5" MIPI-DSI panel driver
+ *
+ * Copyright (C) Purism SPC 2019
+ */
+
+#include <drm/drm_mipi_dsi.h>
+#include <drm/drm_modes.h>
+#include <drm/drm_panel.h>
+#include <drm/drm_print.h>
+#include <linux/backlight.h>
+#include <linux/debugfs.h>
+#include <linux/delay.h>
+#include <linux/gpio/consumer.h>
+#include <linux/media-bus-format.h>
+#include <linux/module.h>
+#include <video/display_timing.h>
+#include <video/mipi_display.h>
+
+#define DRV_NAME "panel-rocktech-jh057n00900"
+
+/* Manufacturer specific Commands send via DSI */
+#define ST7703_CMD_ALL_PIXEL_OFF 0x22
+#define ST7703_CMD_ALL_PIXEL_ON	 0x23
+#define ST7703_CMD_SETDISP	 0xB2
+#define ST7703_CMD_SETRGBIF	 0xB3
+#define ST7703_CMD_SETCYC	 0xB4
+#define ST7703_CMD_SETBGP	 0xB5
+#define ST7703_CMD_SETVCOM	 0xB6
+#define ST7703_CMD_SETOTP	 0xB7
+#define ST7703_CMD_SETPOWER_EXT	 0xB8
+#define ST7703_CMD_SETEXTC	 0xB9
+#define ST7703_CMD_SETMIPI	 0xBA
+#define ST7703_CMD_SETVDC	 0xBC
+#define ST7703_CMD_SETSCR	 0xC0
+#define ST7703_CMD_SETPOWER	 0xC1
+#define ST7703_CMD_SETPANEL	 0xCC
+#define ST7703_CMD_SETGAMMA	 0xE0
+#define ST7703_CMD_SETEQ	 0xE3
+#define ST7703_CMD_SETGIP1	 0xE9
+#define ST7703_CMD_SETGIP2	 0xEA
+
+struct jh057n {
+	struct device *dev;
+	struct drm_panel panel;
+	struct gpio_desc *reset_gpio;
+	struct backlight_device *backlight;
+	bool prepared;
+
+	struct dentry *debugfs;
+};
+
+static inline struct jh057n *panel_to_jh057n(struct drm_panel *panel)
+{
+	return container_of(panel, struct jh057n, panel);
+}
+
+#define dsi_generic_write_seq(dsi, seq...) do {				\
+		static const u8 d[] = { seq };				\
+		int ret;						\
+		ret = mipi_dsi_generic_write(dsi, d, ARRAY_SIZE(d));	\
+		if (ret < 0)						\
+			return ret;					\
+	} while (0)
+
+static int jh057n_init_sequence(struct jh057n *ctx)
+{
+	struct mipi_dsi_device *dsi = to_mipi_dsi_device(ctx->dev);
+	struct device *dev = ctx->dev;
+	int ret;
+
+	/*
+	 * Init sequence was supplied by the panel vendor. Most of the commands
+	 * resemble the ST7703 but the number of parameters often don't match
+	 * so it's likely a clone.
+	 */
+	dsi_generic_write_seq(dsi, ST7703_CMD_SETEXTC,
+			      0xF1, 0x12, 0x83);
+	dsi_generic_write_seq(dsi, ST7703_CMD_SETRGBIF,
+			      0x10, 0x10, 0x05, 0x05, 0x03, 0xFF, 0x00, 0x00,
+			      0x00, 0x00);
+	dsi_generic_write_seq(dsi, ST7703_CMD_SETSCR,
+			      0x73, 0x73, 0x50, 0x50, 0x00, 0x00, 0x08, 0x70,
+			      0x00);
+	dsi_generic_write_seq(dsi, ST7703_CMD_SETVDC, 0x4E);
+	dsi_generic_write_seq(dsi, ST7703_CMD_SETPANEL, 0x0B);
+	dsi_generic_write_seq(dsi, ST7703_CMD_SETCYC, 0x80);
+	dsi_generic_write_seq(dsi, ST7703_CMD_SETDISP, 0xF0, 0x12, 0x30);
+	dsi_generic_write_seq(dsi, ST7703_CMD_SETEQ,
+			      0x07, 0x07, 0x0B, 0x0B, 0x03, 0x0B, 0x00, 0x00,
+			      0x00, 0x00, 0xFF, 0x00, 0xC0, 0x10);
+	dsi_generic_write_seq(dsi, ST7703_CMD_SETBGP, 0x08, 0x08);
+	msleep(20);
+
+	dsi_generic_write_seq(dsi, ST7703_CMD_SETVCOM, 0x3F, 0x3F);
+	dsi_generic_write_seq(dsi, 0xBF, 0x02, 0x11, 0x00);
+	dsi_generic_write_seq(dsi, ST7703_CMD_SETGIP1,
+			      0x82, 0x10, 0x06, 0x05, 0x9E, 0x0A, 0xA5, 0x12,
+			      0x31, 0x23, 0x37, 0x83, 0x04, 0xBC, 0x27, 0x38,
+			      0x0C, 0x00, 0x03, 0x00, 0x00, 0x00, 0x0C, 0x00,
+			      0x03, 0x00, 0x00, 0x00, 0x75, 0x75, 0x31, 0x88,
+			      0x88, 0x88, 0x88, 0x88, 0x88, 0x13, 0x88, 0x64,
+			      0x64, 0x20, 0x88, 0x88, 0x88, 0x88, 0x88, 0x88,
+			      0x02, 0x88, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+			      0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00);
+	dsi_generic_write_seq(dsi, ST7703_CMD_SETGIP2,
+			      0x02, 0x21, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+			      0x00, 0x00, 0x00, 0x00, 0x02, 0x46, 0x02, 0x88,
+			      0x88, 0x88, 0x88, 0x88, 0x88, 0x64, 0x88, 0x13,
+			      0x57, 0x13, 0x88, 0x88, 0x88, 0x88, 0x88, 0x88,
+			      0x75, 0x88, 0x23, 0x14, 0x00, 0x00, 0x02, 0x00,
+			      0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+			      0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x30, 0x0A,
+			      0xA5, 0x00, 0x00, 0x00, 0x00);
+	dsi_generic_write_seq(dsi, ST7703_CMD_SETGAMMA,
+			      0x00, 0x09, 0x0E, 0x29, 0x2D, 0x3C, 0x41, 0x37,
+			      0x07, 0x0B, 0x0D, 0x10, 0x11, 0x0F, 0x10, 0x11,
+			      0x18, 0x00, 0x09, 0x0E, 0x29, 0x2D, 0x3C, 0x41,
+			      0x37, 0x07, 0x0B, 0x0D, 0x10, 0x11, 0x0F, 0x10,
+			      0x11, 0x18);
+	msleep(20);
+
+	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
+	if (ret < 0) {
+		DRM_DEV_ERROR(dev, "Failed to exit sleep mode\n");
+		return ret;
+	}
+	/* Panel is operational 120 msec after reset */
+	msleep(60);
+	ret = mipi_dsi_dcs_set_display_on(dsi);
+	if (ret)
+		return ret;
+
+	DRM_DEV_DEBUG_DRIVER(dev, "Panel init sequence done\n");
+	return 0;
+}
+
+static int jh057n_enable(struct drm_panel *panel)
+{
+	struct jh057n *ctx = panel_to_jh057n(panel);
+
+	return backlight_enable(ctx->backlight);
+}
+
+static int jh057n_disable(struct drm_panel *panel)
+{
+	struct jh057n *ctx = panel_to_jh057n(panel);
+
+	return backlight_disable(ctx->backlight);
+}
+
+static int jh057n_unprepare(struct drm_panel *panel)
+{
+	struct jh057n *ctx = panel_to_jh057n(panel);
+	struct mipi_dsi_device *dsi = to_mipi_dsi_device(ctx->dev);
+
+	if (!ctx->prepared)
+		return 0;
+
+	mipi_dsi_dcs_set_display_off(dsi);
+	ctx->prepared = false;
+
+	return 0;
+}
+
+static int jh057n_prepare(struct drm_panel *panel)
+{
+	struct jh057n *ctx = panel_to_jh057n(panel);
+	int ret;
+
+	if (ctx->prepared)
+		return 0;
+
+	DRM_DEV_DEBUG_DRIVER(ctx->dev, "Resetting the panel\n");
+	gpiod_set_value_cansleep(ctx->reset_gpio, 1);
+	usleep_range(20, 40);
+	gpiod_set_value_cansleep(ctx->reset_gpio, 0);
+	msleep(20);
+
+	ret = jh057n_init_sequence(ctx);
+	if (ret < 0) {
+		DRM_DEV_ERROR(ctx->dev, "Panel init sequence failed: %d\n",
+			      ret);
+		return ret;
+	}
+
+	ctx->prepared = true;
+
+	return 0;
+}
+
+static const struct drm_display_mode default_mode = {
+	.hdisplay    = 720,
+	.hsync_start = 720 + 90,
+	.hsync_end   = 720 + 90 + 20,
+	.htotal	     = 720 + 90 + 20 + 20,
+	.vdisplay    = 1440,
+	.vsync_start = 1440 + 20,
+	.vsync_end   = 1440 + 20 + 4,
+	.vtotal	     = 1440 + 20 + 4 + 12,
+	.vrefresh    = 60,
+	.clock	     = 75276,
+	.flags	     = DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC,
+	.width_mm    = 65,
+	.height_mm   = 130,
+};
+
+static int jh057n_get_modes(struct drm_panel *panel)
+{
+	struct jh057n *ctx = panel_to_jh057n(panel);
+	struct drm_display_mode *mode;
+
+	mode = drm_mode_duplicate(panel->drm, &default_mode);
+	if (!mode) {
+		DRM_DEV_ERROR(ctx->dev, "Failed to add mode %ux%u@%u\n",
+			      default_mode.hdisplay, default_mode.vdisplay,
+			      default_mode.vrefresh);
+		return -ENOMEM;
+	}
+
+	drm_mode_set_name(mode);
+
+	mode->type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED;
+	panel->connector->display_info.width_mm = mode->width_mm;
+	panel->connector->display_info.height_mm = mode->height_mm;
+	drm_mode_probed_add(panel->connector, mode);
+
+	return 1;
+}
+
+static const struct drm_panel_funcs jh057n_drm_funcs = {
+	.disable   = jh057n_disable,
+	.unprepare = jh057n_unprepare,
+	.prepare   = jh057n_prepare,
+	.enable	   = jh057n_enable,
+	.get_modes = jh057n_get_modes,
+};
+
+static int allpixelson_set(void *data, u64 val)
+{
+	struct jh057n *ctx = data;
+	struct mipi_dsi_device *dsi = to_mipi_dsi_device(ctx->dev);
+
+	DRM_DEV_DEBUG_DRIVER(ctx->dev, "Setting all pixels on\n");
+	dsi_generic_write_seq(dsi, ST7703_CMD_ALL_PIXEL_ON);
+	msleep(val * 1000);
+	/* Reset the panel to get video back */
+	drm_panel_disable(&ctx->panel);
+	drm_panel_unprepare(&ctx->panel);
+	drm_panel_prepare(&ctx->panel);
+	drm_panel_enable(&ctx->panel);
+
+	return 0;
+}
+
+DEFINE_SIMPLE_ATTRIBUTE(allpixelson_fops, NULL,
+			allpixelson_set, "%llu\n");
+
+static int jh057n_debugfs_init(struct jh057n *ctx)
+{
+	struct dentry *f;
+
+	ctx->debugfs = debugfs_create_dir(DRV_NAME, NULL);
+	if (!ctx->debugfs)
+		return -ENOMEM;
+
+	f = debugfs_create_file("allpixelson", 0600,
+				ctx->debugfs, ctx, &allpixelson_fops);
+	if (!f)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static void jh057n_debugfs_remove(struct jh057n *ctx)
+{
+	debugfs_remove_recursive(ctx->debugfs);
+	ctx->debugfs = NULL;
+}
+
+static int jh057n_probe(struct mipi_dsi_device *dsi)
+{
+	struct device *dev = &dsi->dev;
+	struct jh057n *ctx;
+	int ret;
+
+	ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL);
+	if (!ctx)
+		return -ENOMEM;
+
+	ctx->reset_gpio = devm_gpiod_get(dev, "reset", GPIOD_OUT_LOW);
+	if (IS_ERR(ctx->reset_gpio)) {
+		DRM_DEV_ERROR(dev, "cannot get reset gpio\n");
+		return PTR_ERR(ctx->reset_gpio);
+	}
+
+	mipi_dsi_set_drvdata(dsi, ctx);
+
+	ctx->dev = dev;
+
+	dsi->lanes = 4;
+	dsi->format = MIPI_DSI_FMT_RGB888;
+	dsi->mode_flags = MIPI_DSI_MODE_VIDEO |
+		MIPI_DSI_MODE_VIDEO_BURST | MIPI_DSI_MODE_VIDEO_SYNC_PULSE;
+
+	ctx->backlight = devm_of_find_backlight(dev);
+	if (IS_ERR(ctx->backlight))
+		return PTR_ERR(ctx->backlight);
+
+	drm_panel_init(&ctx->panel);
+	ctx->panel.dev = dev;
+	ctx->panel.funcs = &jh057n_drm_funcs;
+
+	drm_panel_add(&ctx->panel);
+
+	ret = mipi_dsi_attach(dsi);
+	if (ret < 0) {
+		DRM_DEV_ERROR(dev, "mipi_dsi_attach failed. Is host ready?\n");
+		drm_panel_remove(&ctx->panel);
+		return ret;
+	}
+
+	DRM_DEV_INFO(dev, "%ux%u@%u %ubpp dsi %udl - ready\n",
+		     default_mode.hdisplay, default_mode.vdisplay,
+		     default_mode.vrefresh,
+		     mipi_dsi_pixel_format_to_bpp(dsi->format), dsi->lanes);
+
+	jh057n_debugfs_init(ctx);
+	return 0;
+}
+
+static void jh057n_shutdown(struct mipi_dsi_device *dsi)
+{
+	struct jh057n *ctx = mipi_dsi_get_drvdata(dsi);
+	int ret;
+
+	ret = jh057n_unprepare(&ctx->panel);
+	if (ret < 0)
+		DRM_DEV_ERROR(&dsi->dev, "Failed to unprepare panel: %d\n",
+			      ret);
+
+	ret = jh057n_disable(&ctx->panel);
+	if (ret < 0)
+		DRM_DEV_ERROR(&dsi->dev, "Failed to disable panel: %d\n",
+			      ret);
+}
+
+static int jh057n_remove(struct mipi_dsi_device *dsi)
+{
+	struct jh057n *ctx = mipi_dsi_get_drvdata(dsi);
+	int ret;
+
+	jh057n_shutdown(dsi);
+
+	ret = mipi_dsi_detach(dsi);
+	if (ret < 0)
+		DRM_DEV_ERROR(&dsi->dev, "Failed to detach from DSI host: %d\n",
+			      ret);
+
+	drm_panel_remove(&ctx->panel);
+
+	jh057n_debugfs_remove(ctx);
+
+	return 0;
+}
+
+static const struct of_device_id jh057n_of_match[] = {
+	{ .compatible = "rocktech,jh057n00900" },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, jh057n_of_match);
+
+static struct mipi_dsi_driver jh057n_driver = {
+	.probe	= jh057n_probe,
+	.remove = jh057n_remove,
+	.shutdown = jh057n_shutdown,
+	.driver = {
+		.name = DRV_NAME,
+		.of_match_table = jh057n_of_match,
+	},
+};
+module_mipi_dsi_driver(jh057n_driver);
+
+MODULE_AUTHOR("Guido Günther <agx@sigxcpu.org>");
+MODULE_DESCRIPTION("DRM driver for Rocktech JH057N00900 MIPI DSI panel");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/gpu/drm/panel/panel-ronbo-rb070d30.c b/drivers/gpu/drm/panel/panel-ronbo-rb070d30.c
new file mode 100644
index 000000000000..3c15764f0c03
--- /dev/null
+++ b/drivers/gpu/drm/panel/panel-ronbo-rb070d30.c
@@ -0,0 +1,258 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ * Copyright (C) 2018-2019, Bridge Systems BV
+ * Copyright (C) 2018-2019, Bootlin
+ * Copyright (C) 2017, Free Electrons
+ *
+ * This file based on panel-ilitek-ili9881c.c
+ */
+
+#include <linux/backlight.h>
+#include <linux/delay.h>
+#include <linux/device.h>
+#include <linux/err.h>
+#include <linux/errno.h>
+#include <linux/fb.h>
+#include <linux/kernel.h>
+#include <linux/media-bus-format.h>
+#include <linux/module.h>
+
+#include <linux/gpio/consumer.h>
+#include <linux/regulator/consumer.h>
+
+#include <drm/drm_connector.h>
+#include <drm/drm_mipi_dsi.h>
+#include <drm/drm_modes.h>
+#include <drm/drm_panel.h>
+#include <drm/drm_print.h>
+
+struct rb070d30_panel {
+	struct drm_panel panel;
+	struct mipi_dsi_device *dsi;
+	struct backlight_device *backlight;
+	struct regulator *supply;
+
+	struct {
+		struct gpio_desc *power;
+		struct gpio_desc *reset;
+		struct gpio_desc *updn;
+		struct gpio_desc *shlr;
+	} gpios;
+};
+
+static inline struct rb070d30_panel *panel_to_rb070d30_panel(struct drm_panel *panel)
+{
+	return container_of(panel, struct rb070d30_panel, panel);
+}
+
+static int rb070d30_panel_prepare(struct drm_panel *panel)
+{
+	struct rb070d30_panel *ctx = panel_to_rb070d30_panel(panel);
+	int ret;
+
+	ret = regulator_enable(ctx->supply);
+	if (ret < 0) {
+		DRM_DEV_ERROR(&ctx->dsi->dev, "Failed to enable supply: %d\n", ret);
+		return ret;
+	}
+
+	msleep(20);
+	gpiod_set_value(ctx->gpios.power, 1);
+	msleep(20);
+	gpiod_set_value(ctx->gpios.reset, 1);
+	msleep(20);
+	return 0;
+}
+
+static int rb070d30_panel_unprepare(struct drm_panel *panel)
+{
+	struct rb070d30_panel *ctx = panel_to_rb070d30_panel(panel);
+
+	gpiod_set_value(ctx->gpios.reset, 0);
+	gpiod_set_value(ctx->gpios.power, 0);
+	regulator_disable(ctx->supply);
+
+	return 0;
+}
+
+static int rb070d30_panel_enable(struct drm_panel *panel)
+{
+	struct rb070d30_panel *ctx = panel_to_rb070d30_panel(panel);
+	int ret;
+
+	ret = mipi_dsi_dcs_exit_sleep_mode(ctx->dsi);
+	if (ret)
+		return ret;
+
+	ret = backlight_enable(ctx->backlight);
+	if (ret)
+		goto out;
+
+	return 0;
+
+out:
+	mipi_dsi_dcs_enter_sleep_mode(ctx->dsi);
+	return ret;
+}
+
+static int rb070d30_panel_disable(struct drm_panel *panel)
+{
+	struct rb070d30_panel *ctx = panel_to_rb070d30_panel(panel);
+
+	backlight_disable(ctx->backlight);
+	return mipi_dsi_dcs_enter_sleep_mode(ctx->dsi);
+}
+
+/* Default timings */
+static const struct drm_display_mode default_mode = {
+	.clock		= 51206,
+	.hdisplay	= 1024,
+	.hsync_start	= 1024 + 160,
+	.hsync_end	= 1024 + 160 + 80,
+	.htotal		= 1024 + 160 + 80 + 80,
+	.vdisplay	= 600,
+	.vsync_start	= 600 + 12,
+	.vsync_end	= 600 + 12 + 10,
+	.vtotal		= 600 + 12 + 10 + 13,
+	.vrefresh	= 60,
+
+	.width_mm	= 154,
+	.height_mm	= 85,
+};
+
+static int rb070d30_panel_get_modes(struct drm_panel *panel)
+{
+	struct drm_connector *connector = panel->connector;
+	struct rb070d30_panel *ctx = panel_to_rb070d30_panel(panel);
+	struct drm_display_mode *mode;
+	static const u32 bus_format = MEDIA_BUS_FMT_RGB888_1X24;
+
+	mode = drm_mode_duplicate(panel->drm, &default_mode);
+	if (!mode) {
+		DRM_DEV_ERROR(&ctx->dsi->dev,
+			      "Failed to add mode " DRM_MODE_FMT "\n",
+			      DRM_MODE_ARG(&default_mode));
+		return -EINVAL;
+	}
+
+	drm_mode_set_name(mode);
+
+	mode->type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED;
+	drm_mode_probed_add(connector, mode);
+
+	panel->connector->display_info.bpc = 8;
+	panel->connector->display_info.width_mm = mode->width_mm;
+	panel->connector->display_info.height_mm = mode->height_mm;
+	drm_display_info_set_bus_formats(&connector->display_info,
+					 &bus_format, 1);
+
+	return 1;
+}
+
+static const struct drm_panel_funcs rb070d30_panel_funcs = {
+	.get_modes	= rb070d30_panel_get_modes,
+	.prepare	= rb070d30_panel_prepare,
+	.enable		= rb070d30_panel_enable,
+	.disable	= rb070d30_panel_disable,
+	.unprepare	= rb070d30_panel_unprepare,
+};
+
+static int rb070d30_panel_dsi_probe(struct mipi_dsi_device *dsi)
+{
+	struct rb070d30_panel *ctx;
+	int ret;
+
+	ctx = devm_kzalloc(&dsi->dev, sizeof(*ctx), GFP_KERNEL);
+	if (!ctx)
+		return -ENOMEM;
+
+	ctx->supply = devm_regulator_get(&dsi->dev, "vcc-lcd");
+	if (IS_ERR(ctx->supply))
+		return PTR_ERR(ctx->supply);
+
+	mipi_dsi_set_drvdata(dsi, ctx);
+	ctx->dsi = dsi;
+
+	drm_panel_init(&ctx->panel);
+	ctx->panel.dev = &dsi->dev;
+	ctx->panel.funcs = &rb070d30_panel_funcs;
+
+	ctx->gpios.reset = devm_gpiod_get(&dsi->dev, "reset", GPIOD_OUT_LOW);
+	if (IS_ERR(ctx->gpios.reset)) {
+		DRM_DEV_ERROR(&dsi->dev, "Couldn't get our reset GPIO\n");
+		return PTR_ERR(ctx->gpios.reset);
+	}
+
+	ctx->gpios.power = devm_gpiod_get(&dsi->dev, "power", GPIOD_OUT_LOW);
+	if (IS_ERR(ctx->gpios.power)) {
+		DRM_DEV_ERROR(&dsi->dev, "Couldn't get our power GPIO\n");
+		return PTR_ERR(ctx->gpios.power);
+	}
+
+	/*
+	 * We don't change the state of that GPIO later on but we need
+	 * to force it into a low state.
+	 */
+	ctx->gpios.updn = devm_gpiod_get(&dsi->dev, "updn", GPIOD_OUT_LOW);
+	if (IS_ERR(ctx->gpios.updn)) {
+		DRM_DEV_ERROR(&dsi->dev, "Couldn't get our updn GPIO\n");
+		return PTR_ERR(ctx->gpios.updn);
+	}
+
+	/*
+	 * We don't change the state of that GPIO later on but we need
+	 * to force it into a low state.
+	 */
+	ctx->gpios.shlr = devm_gpiod_get(&dsi->dev, "shlr", GPIOD_OUT_LOW);
+	if (IS_ERR(ctx->gpios.shlr)) {
+		DRM_DEV_ERROR(&dsi->dev, "Couldn't get our shlr GPIO\n");
+		return PTR_ERR(ctx->gpios.shlr);
+	}
+
+	ctx->backlight = devm_of_find_backlight(&dsi->dev);
+	if (IS_ERR(ctx->backlight)) {
+		DRM_DEV_ERROR(&dsi->dev, "Couldn't get our backlight\n");
+		return PTR_ERR(ctx->backlight);
+	}
+
+	ret = drm_panel_add(&ctx->panel);
+	if (ret < 0)
+		return ret;
+
+	dsi->mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_BURST | MIPI_DSI_MODE_LPM;
+	dsi->format = MIPI_DSI_FMT_RGB888;
+	dsi->lanes = 4;
+
+	return mipi_dsi_attach(dsi);
+}
+
+static int rb070d30_panel_dsi_remove(struct mipi_dsi_device *dsi)
+{
+	struct rb070d30_panel *ctx = mipi_dsi_get_drvdata(dsi);
+
+	mipi_dsi_detach(dsi);
+	drm_panel_remove(&ctx->panel);
+
+	return 0;
+}
+
+static const struct of_device_id rb070d30_panel_of_match[] = {
+	{ .compatible = "ronbo,rb070d30" },
+	{ /* sentinel */ },
+};
+MODULE_DEVICE_TABLE(of, rb070d30_panel_of_match);
+
+static struct mipi_dsi_driver rb070d30_panel_driver = {
+	.probe = rb070d30_panel_dsi_probe,
+	.remove = rb070d30_panel_dsi_remove,
+	.driver = {
+		.name = "panel-ronbo-rb070d30",
+		.of_match_table	= rb070d30_panel_of_match,
+	},
+};
+module_mipi_dsi_driver(rb070d30_panel_driver);
+
+MODULE_AUTHOR("Boris Brezillon <boris.brezillon@bootlin.com>");
+MODULE_AUTHOR("Konstantin Sudakov <k.sudakov@integrasources.com>");
+MODULE_DESCRIPTION("Ronbo RB070D30 Panel Driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/panel/panel-samsung-s6d16d0.c b/drivers/gpu/drm/panel/panel-samsung-s6d16d0.c
index 33c22ee036f8..f75bef24e050 100644
--- a/drivers/gpu/drm/panel/panel-samsung-s6d16d0.c
+++ b/drivers/gpu/drm/panel/panel-samsung-s6d16d0.c
@@ -148,9 +148,6 @@ static int s6d16d0_get_modes(struct drm_panel *panel)
 	struct drm_connector *connector = panel->connector;
 	struct drm_display_mode *mode;
 
-	strncpy(connector->display_info.name, "Samsung S6D16D0\0",
-		DRM_DISPLAY_INFO_LEN);
-
 	mode = drm_mode_duplicate(panel->drm, &samsung_s6d16d0_mode);
 	if (!mode) {
 		DRM_ERROR("bad mode or failed to add mode\n");
diff --git a/drivers/gpu/drm/panel/panel-seiko-43wvf1g.c b/drivers/gpu/drm/panel/panel-seiko-43wvf1g.c
index 2d99e28ff117..bdcc5d80823d 100644
--- a/drivers/gpu/drm/panel/panel-seiko-43wvf1g.c
+++ b/drivers/gpu/drm/panel/panel-seiko-43wvf1g.c
@@ -328,7 +328,7 @@ static const struct seiko_panel_desc seiko_43wvf1g = {
 		.height = 57,
 	},
 	.bus_format = MEDIA_BUS_FMT_RGB888_1X24,
-	.bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_NEGEDGE,
+	.bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_DRIVE_NEGEDGE,
 };
 
 static const struct of_device_id platform_of_match[] = {
diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c
index 9e8218f6a3f2..569be4efd8d1 100644
--- a/drivers/gpu/drm/panel/panel-simple.c
+++ b/drivers/gpu/drm/panel/panel-simple.c
@@ -914,7 +914,7 @@ static const struct panel_desc cdtech_s043wq26h_ct7 = {
 		.width = 95,
 		.height = 54,
 	},
-	.bus_flags = DRM_BUS_FLAG_PIXDATA_POSEDGE,
+	.bus_flags = DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE,
 };
 
 static const struct drm_display_mode cdtech_s070wv95_ct16_mode = {
@@ -1034,7 +1034,7 @@ static const struct panel_desc dataimage_scf0700c48ggu18 = {
 		.height = 91,
 	},
 	.bus_format = MEDIA_BUS_FMT_RGB888_1X24,
-	.bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_POSEDGE,
+	.bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE,
 };
 
 static const struct display_timing dlc_dlc0700yzg_1_timing = {
@@ -1119,7 +1119,7 @@ static const struct panel_desc edt_et057090dhu = {
 		.height = 86,
 	},
 	.bus_format = MEDIA_BUS_FMT_RGB666_1X18,
-	.bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_NEGEDGE,
+	.bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_DRIVE_NEGEDGE,
 };
 
 static const struct drm_display_mode edt_etm0700g0dh6_mode = {
@@ -1145,7 +1145,7 @@ static const struct panel_desc edt_etm0700g0dh6 = {
 		.height = 91,
 	},
 	.bus_format = MEDIA_BUS_FMT_RGB666_1X18,
-	.bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_NEGEDGE,
+	.bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_DRIVE_NEGEDGE,
 };
 
 static const struct panel_desc edt_etm0700g0bdh6 = {
@@ -1157,7 +1157,7 @@ static const struct panel_desc edt_etm0700g0bdh6 = {
 		.height = 91,
 	},
 	.bus_format = MEDIA_BUS_FMT_RGB666_1X18,
-	.bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_POSEDGE,
+	.bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE,
 };
 
 static const struct drm_display_mode foxlink_fl500wvr00_a0t_mode = {
@@ -1311,7 +1311,7 @@ static const struct panel_desc innolux_at043tn24 = {
 		.height = 54,
 	},
 	.bus_format = MEDIA_BUS_FMT_RGB888_1X24,
-	.bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_POSEDGE,
+	.bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE,
 };
 
 static const struct drm_display_mode innolux_at070tn92_mode = {
@@ -1818,7 +1818,7 @@ static const struct panel_desc nec_nl4827hc19_05b = {
 		.height = 54,
 	},
 	.bus_format = MEDIA_BUS_FMT_RGB888_1X24,
-	.bus_flags = DRM_BUS_FLAG_PIXDATA_POSEDGE,
+	.bus_flags = DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE,
 };
 
 static const struct drm_display_mode netron_dy_e231732_mode = {
@@ -1867,8 +1867,8 @@ static const struct panel_desc newhaven_nhd_43_480272ef_atxl = {
 		.height = 54,
 	},
 	.bus_format = MEDIA_BUS_FMT_RGB888_1X24,
-	.bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_POSEDGE |
-		     DRM_BUS_FLAG_SYNC_POSEDGE,
+	.bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE |
+		     DRM_BUS_FLAG_SYNC_DRIVE_POSEDGE,
 };
 
 static const struct display_timing nlt_nl192108ac18_02d_timing = {
@@ -2029,7 +2029,33 @@ static const struct panel_desc ortustech_com43h4m85ulc = {
 		.height = 93,
 	},
 	.bus_format = MEDIA_BUS_FMT_RGB888_1X24,
-	.bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_POSEDGE,
+	.bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE,
+};
+
+static const struct drm_display_mode osddisplays_osd070t1718_19ts_mode  = {
+	.clock = 33000,
+	.hdisplay = 800,
+	.hsync_start = 800 + 210,
+	.hsync_end = 800 + 210 + 30,
+	.htotal = 800 + 210 + 30 + 16,
+	.vdisplay = 480,
+	.vsync_start = 480 + 22,
+	.vsync_end = 480 + 22 + 13,
+	.vtotal = 480 + 22 + 13 + 10,
+	.vrefresh = 60,
+	.flags = DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_NHSYNC,
+};
+
+static const struct panel_desc osddisplays_osd070t1718_19ts = {
+	.modes = &osddisplays_osd070t1718_19ts_mode,
+	.num_modes = 1,
+	.bpc = 8,
+	.size = {
+		.width = 152,
+		.height = 91,
+	},
+	.bus_format = MEDIA_BUS_FMT_RGB888_1X24,
+	.bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE,
 };
 
 static const struct drm_display_mode pda_91_00156_a0_mode = {
@@ -2398,7 +2424,7 @@ static const struct panel_desc toshiba_lt089ac29000 = {
 		.height = 116,
 	},
 	.bus_format = MEDIA_BUS_FMT_RGB888_1X24,
-	.bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_POSEDGE,
+	.bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE,
 };
 
 static const struct drm_display_mode tpk_f07a_0102_mode = {
@@ -2421,7 +2447,7 @@ static const struct panel_desc tpk_f07a_0102 = {
 		.width = 152,
 		.height = 91,
 	},
-	.bus_flags = DRM_BUS_FLAG_PIXDATA_POSEDGE,
+	.bus_flags = DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE,
 };
 
 static const struct drm_display_mode tpk_f10a_0102_mode = {
@@ -2737,6 +2763,9 @@ static const struct of_device_id platform_of_match[] = {
 		.compatible = "ortustech,com43h4m85ulc",
 		.data = &ortustech_com43h4m85ulc,
 	}, {
+		.compatible = "osddisplays,osd070t1718-19ts",
+		.data = &osddisplays_osd070t1718_19ts,
+	}, {
 		.compatible = "pda,91-00156-a0",
 		.data = &pda_91_00156_a0,
 	}, {
@@ -2996,6 +3025,34 @@ static const struct panel_desc_dsi panasonic_vvx10f004b00 = {
 	.lanes = 4,
 };
 
+static const struct drm_display_mode lg_acx467akm_7_mode = {
+	.clock = 150000,
+	.hdisplay = 1080,
+	.hsync_start = 1080 + 2,
+	.hsync_end = 1080 + 2 + 2,
+	.htotal = 1080 + 2 + 2 + 2,
+	.vdisplay = 1920,
+	.vsync_start = 1920 + 2,
+	.vsync_end = 1920 + 2 + 2,
+	.vtotal = 1920 + 2 + 2 + 2,
+	.vrefresh = 60,
+};
+
+static const struct panel_desc_dsi lg_acx467akm_7 = {
+	.desc = {
+		.modes = &lg_acx467akm_7_mode,
+		.num_modes = 1,
+		.bpc = 8,
+		.size = {
+			.width = 62,
+			.height = 110,
+		},
+	},
+	.flags = 0,
+	.format = MIPI_DSI_FMT_RGB888,
+	.lanes = 4,
+};
+
 static const struct of_device_id dsi_of_match[] = {
 	{
 		.compatible = "auo,b080uan01",
@@ -3013,6 +3070,9 @@ static const struct of_device_id dsi_of_match[] = {
 		.compatible = "panasonic,vvx10f004b00",
 		.data = &panasonic_vvx10f004b00
 	}, {
+		.compatible = "lg,acx467akm-7",
+		.data = &lg_acx467akm_7
+	}, {
 		/* sentinel */
 	}
 };
diff --git a/drivers/gpu/drm/panel/panel-tpo-tpg110.c b/drivers/gpu/drm/panel/panel-tpo-tpg110.c
index 5a9f8f4d5d24..71591e5f5938 100644
--- a/drivers/gpu/drm/panel/panel-tpo-tpg110.c
+++ b/drivers/gpu/drm/panel/panel-tpo-tpg110.c
@@ -118,7 +118,7 @@ static const struct tpg110_panel_mode tpg110_modes[] = {
 			.vtotal = 480 + 10 + 1 + 35,
 			.vrefresh = 60,
 		},
-		.bus_flags = DRM_BUS_FLAG_PIXDATA_POSEDGE,
+		.bus_flags = DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE,
 	},
 	{
 		.name = "640x480 RGB",
@@ -135,7 +135,7 @@ static const struct tpg110_panel_mode tpg110_modes[] = {
 			.vtotal = 480 + 18 + 1 + 27,
 			.vrefresh = 60,
 		},
-		.bus_flags = DRM_BUS_FLAG_PIXDATA_POSEDGE,
+		.bus_flags = DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE,
 	},
 	{
 		.name = "480x272 RGB",
@@ -152,7 +152,7 @@ static const struct tpg110_panel_mode tpg110_modes[] = {
 			.vtotal = 272 + 2 + 1 + 12,
 			.vrefresh = 60,
 		},
-		.bus_flags = DRM_BUS_FLAG_PIXDATA_POSEDGE,
+		.bus_flags = DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE,
 	},
 	{
 		.name = "480x640 RGB",
@@ -169,7 +169,7 @@ static const struct tpg110_panel_mode tpg110_modes[] = {
 			.vtotal = 640 + 4 + 1 + 8,
 			.vrefresh = 60,
 		},
-		.bus_flags = DRM_BUS_FLAG_PIXDATA_POSEDGE,
+		.bus_flags = DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE,
 	},
 	{
 		.name = "400x240 RGB",
@@ -186,7 +186,7 @@ static const struct tpg110_panel_mode tpg110_modes[] = {
 			.vtotal = 240 + 2 + 1 + 20,
 			.vrefresh = 60,
 		},
-		.bus_flags = DRM_BUS_FLAG_PIXDATA_POSEDGE,
+		.bus_flags = DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE,
 	},
 };
 
@@ -390,8 +390,6 @@ static int tpg110_get_modes(struct drm_panel *panel)
 	struct tpg110 *tpg = to_tpg110(panel);
 	struct drm_display_mode *mode;
 
-	strncpy(connector->display_info.name, tpg->panel_mode->name,
-		DRM_DISPLAY_INFO_LEN);
 	connector->display_info.width_mm = tpg->width;
 	connector->display_info.height_mm = tpg->height;
 	connector->display_info.bus_flags = tpg->panel_mode->bus_flags;
diff --git a/drivers/gpu/drm/panfrost/Kconfig b/drivers/gpu/drm/panfrost/Kconfig
new file mode 100644
index 000000000000..591611dc4e34
--- /dev/null
+++ b/drivers/gpu/drm/panfrost/Kconfig
@@ -0,0 +1,14 @@
+# SPDX-License-Identifier: GPL-2.0
+
+config DRM_PANFROST
+	tristate "Panfrost (DRM support for ARM Mali Midgard/Bifrost GPUs)"
+	depends on DRM
+	depends on ARM || ARM64 || (COMPILE_TEST && !GENERIC_ATOMIC64)
+	depends on MMU
+	select DRM_SCHED
+	select IOMMU_SUPPORT
+	select IOMMU_IO_PGTABLE_LPAE
+	select DRM_GEM_SHMEM_HELPER
+	help
+	  DRM driver for ARM Mali Midgard (T6xx, T7xx, T8xx) and
+	  Bifrost (G3x, G5x, G7x) GPUs.
diff --git a/drivers/gpu/drm/panfrost/Makefile b/drivers/gpu/drm/panfrost/Makefile
new file mode 100644
index 000000000000..6de72d13c58f
--- /dev/null
+++ b/drivers/gpu/drm/panfrost/Makefile
@@ -0,0 +1,12 @@
+# SPDX-License-Identifier: GPL-2.0
+
+panfrost-y := \
+	panfrost_drv.o \
+	panfrost_device.o \
+	panfrost_devfreq.o \
+	panfrost_gem.o \
+	panfrost_gpu.o \
+	panfrost_job.o \
+	panfrost_mmu.o
+
+obj-$(CONFIG_DRM_PANFROST) += panfrost.o
diff --git a/drivers/gpu/drm/panfrost/TODO b/drivers/gpu/drm/panfrost/TODO
new file mode 100644
index 000000000000..c2e44add37d8
--- /dev/null
+++ b/drivers/gpu/drm/panfrost/TODO
@@ -0,0 +1,27 @@
+- Thermal support.
+
+- Bifrost support:
+  - DT bindings (Neil, WIP)
+  - MMU page table format and address space setup
+  - Bifrost specific feature and issue handling
+  - Coherent DMA support
+
+- Support for 2MB pages. The io-pgtable code already supports this. Finishing
+  support involves either copying or adapting the iommu API to handle passing
+  aligned addresses and sizes to the io-pgtable code.
+
+- Per FD address space support. The h/w supports multiple addresses spaces.
+  The hard part is handling when more address spaces are needed than what
+  the h/w provides.
+
+- Support pinning pages on demand (GPU page faults).
+
+- Support userspace controlled GPU virtual addresses. Needed for Vulkan. (Tomeu)
+
+- Support for madvise and a shrinker.
+
+- Compute job support. So called 'compute only' jobs need to be plumbed up to
+  userspace.
+
+- Performance counter support. (Boris)
+
diff --git a/drivers/gpu/drm/panfrost/panfrost_devfreq.c b/drivers/gpu/drm/panfrost/panfrost_devfreq.c
new file mode 100644
index 000000000000..238bd1d89d43
--- /dev/null
+++ b/drivers/gpu/drm/panfrost/panfrost_devfreq.c
@@ -0,0 +1,219 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright 2019 Collabora ltd. */
+#include <linux/devfreq.h>
+#include <linux/platform_device.h>
+#include <linux/pm_opp.h>
+#include <linux/clk.h>
+#include <linux/regulator/consumer.h>
+
+#include "panfrost_device.h"
+#include "panfrost_devfreq.h"
+#include "panfrost_features.h"
+#include "panfrost_issues.h"
+#include "panfrost_gpu.h"
+#include "panfrost_regs.h"
+
+static void panfrost_devfreq_update_utilization(struct panfrost_device *pfdev, int slot);
+
+static int panfrost_devfreq_target(struct device *dev, unsigned long *freq,
+				   u32 flags)
+{
+	struct panfrost_device *pfdev = platform_get_drvdata(to_platform_device(dev));
+	struct dev_pm_opp *opp;
+	unsigned long old_clk_rate = pfdev->devfreq.cur_freq;
+	unsigned long target_volt, target_rate;
+	int err;
+
+	opp = devfreq_recommended_opp(dev, freq, flags);
+	if (IS_ERR(opp))
+		return PTR_ERR(opp);
+
+	target_rate = dev_pm_opp_get_freq(opp);
+	target_volt = dev_pm_opp_get_voltage(opp);
+	dev_pm_opp_put(opp);
+
+	if (old_clk_rate == target_rate)
+		return 0;
+
+	/*
+	 * If frequency scaling from low to high, adjust voltage first.
+	 * If frequency scaling from high to low, adjust frequency first.
+	 */
+	if (old_clk_rate < target_rate) {
+		err = regulator_set_voltage(pfdev->regulator, target_volt,
+					    target_volt);
+		if (err) {
+			dev_err(dev, "Cannot set voltage %lu uV\n",
+				target_volt);
+			return err;
+		}
+	}
+
+	err = clk_set_rate(pfdev->clock, target_rate);
+	if (err) {
+		dev_err(dev, "Cannot set frequency %lu (%d)\n", target_rate,
+			err);
+		regulator_set_voltage(pfdev->regulator, pfdev->devfreq.cur_volt,
+				      pfdev->devfreq.cur_volt);
+		return err;
+	}
+
+	if (old_clk_rate > target_rate) {
+		err = regulator_set_voltage(pfdev->regulator, target_volt,
+					    target_volt);
+		if (err)
+			dev_err(dev, "Cannot set voltage %lu uV\n", target_volt);
+	}
+
+	pfdev->devfreq.cur_freq = target_rate;
+	pfdev->devfreq.cur_volt = target_volt;
+
+	return 0;
+}
+
+static void panfrost_devfreq_reset(struct panfrost_device *pfdev)
+{
+	ktime_t now = ktime_get();
+	int i;
+
+	for (i = 0; i < NUM_JOB_SLOTS; i++) {
+		pfdev->devfreq.slot[i].busy_time = 0;
+		pfdev->devfreq.slot[i].idle_time = 0;
+		pfdev->devfreq.slot[i].time_last_update = now;
+	}
+}
+
+static int panfrost_devfreq_get_dev_status(struct device *dev,
+					   struct devfreq_dev_status *status)
+{
+	struct panfrost_device *pfdev = platform_get_drvdata(to_platform_device(dev));
+	int i;
+
+	for (i = 0; i < NUM_JOB_SLOTS; i++) {
+		panfrost_devfreq_update_utilization(pfdev, i);
+	}
+
+	status->current_frequency = clk_get_rate(pfdev->clock);
+	status->total_time = ktime_to_ns(ktime_add(pfdev->devfreq.slot[0].busy_time,
+						   pfdev->devfreq.slot[0].idle_time));
+
+	status->busy_time = 0;
+	for (i = 0; i < NUM_JOB_SLOTS; i++) {
+		status->busy_time += ktime_to_ns(pfdev->devfreq.slot[i].busy_time);
+	}
+
+	/* We're scheduling only to one core atm, so don't divide for now */
+	/* status->busy_time /= NUM_JOB_SLOTS; */
+
+	panfrost_devfreq_reset(pfdev);
+
+	dev_dbg(pfdev->dev, "busy %lu total %lu %lu %% freq %lu MHz\n", status->busy_time,
+		status->total_time,
+		status->busy_time / (status->total_time / 100),
+		status->current_frequency / 1000 / 1000);
+
+	return 0;
+}
+
+static int panfrost_devfreq_get_cur_freq(struct device *dev, unsigned long *freq)
+{
+	struct panfrost_device *pfdev = platform_get_drvdata(to_platform_device(dev));
+
+	*freq = pfdev->devfreq.cur_freq;
+
+	return 0;
+}
+
+static struct devfreq_dev_profile panfrost_devfreq_profile = {
+	.polling_ms = 50, /* ~3 frames */
+	.target = panfrost_devfreq_target,
+	.get_dev_status = panfrost_devfreq_get_dev_status,
+	.get_cur_freq = panfrost_devfreq_get_cur_freq,
+};
+
+int panfrost_devfreq_init(struct panfrost_device *pfdev)
+{
+	int ret;
+	struct dev_pm_opp *opp;
+
+	if (!pfdev->regulator)
+		return 0;
+
+	ret = dev_pm_opp_of_add_table(&pfdev->pdev->dev);
+	if (ret == -ENODEV) /* Optional, continue without devfreq */
+		return 0;
+
+	panfrost_devfreq_reset(pfdev);
+
+	pfdev->devfreq.cur_freq = clk_get_rate(pfdev->clock);
+
+	opp = devfreq_recommended_opp(&pfdev->pdev->dev, &pfdev->devfreq.cur_freq, 0);
+	if (IS_ERR(opp))
+		return PTR_ERR(opp);
+
+	panfrost_devfreq_profile.initial_freq = pfdev->devfreq.cur_freq;
+	dev_pm_opp_put(opp);
+
+	pfdev->devfreq.devfreq = devm_devfreq_add_device(&pfdev->pdev->dev,
+			&panfrost_devfreq_profile, "simple_ondemand", NULL);
+	if (IS_ERR(pfdev->devfreq.devfreq)) {
+		DRM_DEV_ERROR(&pfdev->pdev->dev, "Couldn't initialize GPU devfreq\n");
+		ret = PTR_ERR(pfdev->devfreq.devfreq);
+		pfdev->devfreq.devfreq = NULL;
+		return ret;
+	}
+
+	return 0;
+}
+
+void panfrost_devfreq_resume(struct panfrost_device *pfdev)
+{
+	int i;
+
+	if (!pfdev->devfreq.devfreq)
+		return;
+
+	panfrost_devfreq_reset(pfdev);
+	for (i = 0; i < NUM_JOB_SLOTS; i++)
+		pfdev->devfreq.slot[i].busy = false;
+
+	devfreq_resume_device(pfdev->devfreq.devfreq);
+}
+
+void panfrost_devfreq_suspend(struct panfrost_device *pfdev)
+{
+	if (!pfdev->devfreq.devfreq)
+		return;
+
+	devfreq_suspend_device(pfdev->devfreq.devfreq);
+}
+
+static void panfrost_devfreq_update_utilization(struct panfrost_device *pfdev, int slot)
+{
+	struct panfrost_devfreq_slot *devfreq_slot = &pfdev->devfreq.slot[slot];
+	ktime_t now;
+	ktime_t last;
+
+	if (!pfdev->devfreq.devfreq)
+		return;
+
+	now = ktime_get();
+	last = pfdev->devfreq.slot[slot].time_last_update;
+
+	/* If we last recorded a transition to busy, we have been idle since */
+	if (devfreq_slot->busy)
+		pfdev->devfreq.slot[slot].busy_time += ktime_sub(now, last);
+	else
+		pfdev->devfreq.slot[slot].idle_time += ktime_sub(now, last);
+
+	pfdev->devfreq.slot[slot].time_last_update = now;
+}
+
+/* The job scheduler is expected to call this at every transition busy <-> idle */
+void panfrost_devfreq_record_transition(struct panfrost_device *pfdev, int slot)
+{
+	struct panfrost_devfreq_slot *devfreq_slot = &pfdev->devfreq.slot[slot];
+
+	panfrost_devfreq_update_utilization(pfdev, slot);
+	devfreq_slot->busy = !devfreq_slot->busy;
+}
diff --git a/drivers/gpu/drm/panfrost/panfrost_devfreq.h b/drivers/gpu/drm/panfrost/panfrost_devfreq.h
new file mode 100644
index 000000000000..eb999531ed90
--- /dev/null
+++ b/drivers/gpu/drm/panfrost/panfrost_devfreq.h
@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright 2019 Collabora ltd. */
+
+#ifndef __PANFROST_DEVFREQ_H__
+#define __PANFROST_DEVFREQ_H__
+
+int panfrost_devfreq_init(struct panfrost_device *pfdev);
+
+void panfrost_devfreq_resume(struct panfrost_device *pfdev);
+void panfrost_devfreq_suspend(struct panfrost_device *pfdev);
+
+void panfrost_devfreq_record_transition(struct panfrost_device *pfdev, int slot);
+
+#endif /* __PANFROST_DEVFREQ_H__ */
diff --git a/drivers/gpu/drm/panfrost/panfrost_device.c b/drivers/gpu/drm/panfrost/panfrost_device.c
new file mode 100644
index 000000000000..970f669c6d29
--- /dev/null
+++ b/drivers/gpu/drm/panfrost/panfrost_device.c
@@ -0,0 +1,253 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright 2018 Marty E. Plummer <hanetzer@startmail.com> */
+/* Copyright 2019 Linaro, Ltd, Rob Herring <robh@kernel.org> */
+
+#include <linux/clk.h>
+#include <linux/reset.h>
+#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
+#include <linux/regulator/consumer.h>
+
+#include "panfrost_device.h"
+#include "panfrost_devfreq.h"
+#include "panfrost_features.h"
+#include "panfrost_gpu.h"
+#include "panfrost_job.h"
+#include "panfrost_mmu.h"
+
+static int panfrost_reset_init(struct panfrost_device *pfdev)
+{
+	int err;
+
+	pfdev->rstc = devm_reset_control_array_get(pfdev->dev, false, true);
+	if (IS_ERR(pfdev->rstc)) {
+		dev_err(pfdev->dev, "get reset failed %ld\n", PTR_ERR(pfdev->rstc));
+		return PTR_ERR(pfdev->rstc);
+	}
+
+	err = reset_control_deassert(pfdev->rstc);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+static void panfrost_reset_fini(struct panfrost_device *pfdev)
+{
+	reset_control_assert(pfdev->rstc);
+}
+
+static int panfrost_clk_init(struct panfrost_device *pfdev)
+{
+	int err;
+	unsigned long rate;
+
+	pfdev->clock = devm_clk_get(pfdev->dev, NULL);
+	if (IS_ERR(pfdev->clock)) {
+		dev_err(pfdev->dev, "get clock failed %ld\n", PTR_ERR(pfdev->clock));
+		return PTR_ERR(pfdev->clock);
+	}
+
+	rate = clk_get_rate(pfdev->clock);
+	dev_info(pfdev->dev, "clock rate = %lu\n", rate);
+
+	err = clk_prepare_enable(pfdev->clock);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+static void panfrost_clk_fini(struct panfrost_device *pfdev)
+{
+	clk_disable_unprepare(pfdev->clock);
+}
+
+static int panfrost_regulator_init(struct panfrost_device *pfdev)
+{
+	int ret;
+
+	pfdev->regulator = devm_regulator_get_optional(pfdev->dev, "mali");
+	if (IS_ERR(pfdev->regulator)) {
+		ret = PTR_ERR(pfdev->regulator);
+		pfdev->regulator = NULL;
+		if (ret == -ENODEV)
+			return 0;
+		dev_err(pfdev->dev, "failed to get regulator: %d\n", ret);
+		return ret;
+	}
+
+	ret = regulator_enable(pfdev->regulator);
+	if (ret < 0) {
+		dev_err(pfdev->dev, "failed to enable regulator: %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static void panfrost_regulator_fini(struct panfrost_device *pfdev)
+{
+	if (pfdev->regulator)
+		regulator_disable(pfdev->regulator);
+}
+
+int panfrost_device_init(struct panfrost_device *pfdev)
+{
+	int err;
+	struct resource *res;
+
+	mutex_init(&pfdev->sched_lock);
+	mutex_init(&pfdev->reset_lock);
+	INIT_LIST_HEAD(&pfdev->scheduled_jobs);
+
+	spin_lock_init(&pfdev->hwaccess_lock);
+
+	err = panfrost_clk_init(pfdev);
+	if (err) {
+		dev_err(pfdev->dev, "clk init failed %d\n", err);
+		return err;
+	}
+
+	err = panfrost_regulator_init(pfdev);
+	if (err) {
+		dev_err(pfdev->dev, "regulator init failed %d\n", err);
+		goto err_out0;
+	}
+
+	err = panfrost_reset_init(pfdev);
+	if (err) {
+		dev_err(pfdev->dev, "reset init failed %d\n", err);
+		goto err_out1;
+	}
+
+	res = platform_get_resource(pfdev->pdev, IORESOURCE_MEM, 0);
+	pfdev->iomem = devm_ioremap_resource(pfdev->dev, res);
+	if (IS_ERR(pfdev->iomem)) {
+		dev_err(pfdev->dev, "failed to ioremap iomem\n");
+		err = PTR_ERR(pfdev->iomem);
+		goto err_out2;
+	}
+
+	err = panfrost_gpu_init(pfdev);
+	if (err)
+		goto err_out2;
+
+	err = panfrost_mmu_init(pfdev);
+	if (err)
+		goto err_out3;
+
+	err = panfrost_job_init(pfdev);
+	if (err)
+		goto err_out4;
+
+	/* runtime PM will wake us up later */
+	panfrost_gpu_power_off(pfdev);
+
+	pm_runtime_set_active(pfdev->dev);
+	pm_runtime_get_sync(pfdev->dev);
+	pm_runtime_mark_last_busy(pfdev->dev);
+	pm_runtime_put_autosuspend(pfdev->dev);
+
+	return 0;
+err_out4:
+	panfrost_mmu_fini(pfdev);
+err_out3:
+	panfrost_gpu_fini(pfdev);
+err_out2:
+	panfrost_reset_fini(pfdev);
+err_out1:
+	panfrost_regulator_fini(pfdev);
+err_out0:
+	panfrost_clk_fini(pfdev);
+	return err;
+}
+
+void panfrost_device_fini(struct panfrost_device *pfdev)
+{
+	panfrost_regulator_fini(pfdev);
+	panfrost_clk_fini(pfdev);
+}
+
+const char *panfrost_exception_name(struct panfrost_device *pfdev, u32 exception_code)
+{
+	switch (exception_code) {
+		/* Non-Fault Status code */
+	case 0x00: return "NOT_STARTED/IDLE/OK";
+	case 0x01: return "DONE";
+	case 0x02: return "INTERRUPTED";
+	case 0x03: return "STOPPED";
+	case 0x04: return "TERMINATED";
+	case 0x08: return "ACTIVE";
+		/* Job exceptions */
+	case 0x40: return "JOB_CONFIG_FAULT";
+	case 0x41: return "JOB_POWER_FAULT";
+	case 0x42: return "JOB_READ_FAULT";
+	case 0x43: return "JOB_WRITE_FAULT";
+	case 0x44: return "JOB_AFFINITY_FAULT";
+	case 0x48: return "JOB_BUS_FAULT";
+	case 0x50: return "INSTR_INVALID_PC";
+	case 0x51: return "INSTR_INVALID_ENC";
+	case 0x52: return "INSTR_TYPE_MISMATCH";
+	case 0x53: return "INSTR_OPERAND_FAULT";
+	case 0x54: return "INSTR_TLS_FAULT";
+	case 0x55: return "INSTR_BARRIER_FAULT";
+	case 0x56: return "INSTR_ALIGN_FAULT";
+	case 0x58: return "DATA_INVALID_FAULT";
+	case 0x59: return "TILE_RANGE_FAULT";
+	case 0x5A: return "ADDR_RANGE_FAULT";
+	case 0x60: return "OUT_OF_MEMORY";
+		/* GPU exceptions */
+	case 0x80: return "DELAYED_BUS_FAULT";
+	case 0x88: return "SHAREABILITY_FAULT";
+		/* MMU exceptions */
+	case 0xC1: return "TRANSLATION_FAULT_LEVEL1";
+	case 0xC2: return "TRANSLATION_FAULT_LEVEL2";
+	case 0xC3: return "TRANSLATION_FAULT_LEVEL3";
+	case 0xC4: return "TRANSLATION_FAULT_LEVEL4";
+	case 0xC8: return "PERMISSION_FAULT";
+	case 0xC9 ... 0xCF: return "PERMISSION_FAULT";
+	case 0xD1: return "TRANSTAB_BUS_FAULT_LEVEL1";
+	case 0xD2: return "TRANSTAB_BUS_FAULT_LEVEL2";
+	case 0xD3: return "TRANSTAB_BUS_FAULT_LEVEL3";
+	case 0xD4: return "TRANSTAB_BUS_FAULT_LEVEL4";
+	case 0xD8: return "ACCESS_FLAG";
+	case 0xD9 ... 0xDF: return "ACCESS_FLAG";
+	case 0xE0 ... 0xE7: return "ADDRESS_SIZE_FAULT";
+	case 0xE8 ... 0xEF: return "MEMORY_ATTRIBUTES_FAULT";
+	}
+
+	return "UNKNOWN";
+}
+
+#ifdef CONFIG_PM
+int panfrost_device_resume(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct panfrost_device *pfdev = platform_get_drvdata(pdev);
+
+	panfrost_gpu_soft_reset(pfdev);
+
+	/* TODO: Re-enable all other address spaces */
+	panfrost_gpu_power_on(pfdev);
+	panfrost_mmu_enable(pfdev, 0);
+	panfrost_job_enable_interrupts(pfdev);
+	panfrost_devfreq_resume(pfdev);
+
+	return 0;
+}
+
+int panfrost_device_suspend(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct panfrost_device *pfdev = platform_get_drvdata(pdev);
+
+	if (!panfrost_job_is_idle(pfdev))
+		return -EBUSY;
+
+	panfrost_devfreq_suspend(pfdev);
+	panfrost_gpu_power_off(pfdev);
+
+	return 0;
+}
+#endif
diff --git a/drivers/gpu/drm/panfrost/panfrost_device.h b/drivers/gpu/drm/panfrost/panfrost_device.h
new file mode 100644
index 000000000000..56f452dfb490
--- /dev/null
+++ b/drivers/gpu/drm/panfrost/panfrost_device.h
@@ -0,0 +1,125 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright 2018 Marty E. Plummer <hanetzer@startmail.com> */
+/* Copyright 2019 Linaro, Ltd, Rob Herring <robh@kernel.org> */
+
+#ifndef __PANFROST_DEVICE_H__
+#define __PANFROST_DEVICE_H__
+
+#include <linux/spinlock.h>
+#include <drm/drm_device.h>
+#include <drm/drm_mm.h>
+#include <drm/gpu_scheduler.h>
+
+struct panfrost_device;
+struct panfrost_mmu;
+struct panfrost_job_slot;
+struct panfrost_job;
+
+#define NUM_JOB_SLOTS 3
+
+struct panfrost_features {
+	u16 id;
+	u16 revision;
+
+	u64 shader_present;
+	u64 tiler_present;
+	u64 l2_present;
+	u64 stack_present;
+	u32 as_present;
+	u32 js_present;
+
+	u32 l2_features;
+	u32 core_features;
+	u32 tiler_features;
+	u32 mem_features;
+	u32 mmu_features;
+	u32 thread_features;
+	u32 max_threads;
+	u32 thread_max_workgroup_sz;
+	u32 thread_max_barrier_sz;
+	u32 coherency_features;
+	u32 texture_features[4];
+	u32 js_features[16];
+
+	u32 nr_core_groups;
+
+	unsigned long hw_features[64 / BITS_PER_LONG];
+	unsigned long hw_issues[64 / BITS_PER_LONG];
+};
+
+struct panfrost_devfreq_slot {
+	ktime_t busy_time;
+	ktime_t idle_time;
+	ktime_t time_last_update;
+	bool busy;
+};
+
+struct panfrost_device {
+	struct device *dev;
+	struct drm_device *ddev;
+	struct platform_device *pdev;
+
+	spinlock_t hwaccess_lock;
+
+	struct drm_mm mm;
+	spinlock_t mm_lock;
+
+	void __iomem *iomem;
+	struct clk *clock;
+	struct regulator *regulator;
+	struct reset_control *rstc;
+
+	struct panfrost_features features;
+
+	struct panfrost_mmu *mmu;
+	struct panfrost_job_slot *js;
+
+	struct panfrost_job *jobs[NUM_JOB_SLOTS];
+	struct list_head scheduled_jobs;
+
+	struct mutex sched_lock;
+	struct mutex reset_lock;
+
+	struct {
+		struct devfreq *devfreq;
+		struct thermal_cooling_device *cooling;
+		unsigned long cur_freq;
+		unsigned long cur_volt;
+		struct panfrost_devfreq_slot slot[NUM_JOB_SLOTS];
+	} devfreq;
+};
+
+struct panfrost_file_priv {
+	struct panfrost_device *pfdev;
+
+	struct drm_sched_entity sched_entity[NUM_JOB_SLOTS];
+};
+
+static inline struct panfrost_device *to_panfrost_device(struct drm_device *ddev)
+{
+	return ddev->dev_private;
+}
+
+static inline int panfrost_model_cmp(struct panfrost_device *pfdev, s32 id)
+{
+	s32 match_id = pfdev->features.id;
+
+	if (match_id & 0xf000)
+		match_id &= 0xf00f;
+	return match_id - id;
+}
+
+static inline bool panfrost_model_eq(struct panfrost_device *pfdev, s32 id)
+{
+	return !panfrost_model_cmp(pfdev, id);
+}
+
+int panfrost_device_init(struct panfrost_device *pfdev);
+void panfrost_device_fini(struct panfrost_device *pfdev);
+
+int panfrost_device_resume(struct device *dev);
+int panfrost_device_suspend(struct device *dev);
+
+const char *panfrost_exception_name(struct panfrost_device *pfdev, u32 exception_code);
+
+#endif
diff --git a/drivers/gpu/drm/panfrost/panfrost_drv.c b/drivers/gpu/drm/panfrost/panfrost_drv.c
new file mode 100644
index 000000000000..94b0819ad50b
--- /dev/null
+++ b/drivers/gpu/drm/panfrost/panfrost_drv.c
@@ -0,0 +1,474 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright 2018 Marty E. Plummer <hanetzer@startmail.com> */
+/* Copyright 2019 Linaro, Ltd., Rob Herring <robh@kernel.org> */
+/* Copyright 2019 Collabora ltd. */
+
+#include <linux/module.h>
+#include <linux/of_platform.h>
+#include <linux/pagemap.h>
+#include <linux/pm_runtime.h>
+#include <drm/panfrost_drm.h>
+#include <drm/drm_drv.h>
+#include <drm/drm_ioctl.h>
+#include <drm/drm_syncobj.h>
+#include <drm/drm_utils.h>
+
+#include "panfrost_device.h"
+#include "panfrost_devfreq.h"
+#include "panfrost_gem.h"
+#include "panfrost_mmu.h"
+#include "panfrost_job.h"
+#include "panfrost_gpu.h"
+
+static int panfrost_ioctl_get_param(struct drm_device *ddev, void *data, struct drm_file *file)
+{
+	struct drm_panfrost_get_param *param = data;
+	struct panfrost_device *pfdev = ddev->dev_private;
+
+	if (param->pad != 0)
+		return -EINVAL;
+
+	switch (param->param) {
+	case DRM_PANFROST_PARAM_GPU_PROD_ID:
+		param->value = pfdev->features.id;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int panfrost_ioctl_create_bo(struct drm_device *dev, void *data,
+		struct drm_file *file)
+{
+	int ret;
+	struct drm_gem_shmem_object *shmem;
+	struct drm_panfrost_create_bo *args = data;
+
+	if (!args->size || args->flags || args->pad)
+		return -EINVAL;
+
+	shmem = drm_gem_shmem_create_with_handle(file, dev, args->size,
+						 &args->handle);
+	if (IS_ERR(shmem))
+		return PTR_ERR(shmem);
+
+	ret = panfrost_mmu_map(to_panfrost_bo(&shmem->base));
+	if (ret)
+		goto err_free;
+
+	args->offset = to_panfrost_bo(&shmem->base)->node.start << PAGE_SHIFT;
+
+	return 0;
+
+err_free:
+	drm_gem_object_put_unlocked(&shmem->base);
+	return ret;
+}
+
+/**
+ * panfrost_lookup_bos() - Sets up job->bo[] with the GEM objects
+ * referenced by the job.
+ * @dev: DRM device
+ * @file_priv: DRM file for this fd
+ * @args: IOCTL args
+ * @job: job being set up
+ *
+ * Resolve handles from userspace to BOs and attach them to job.
+ *
+ * Note that this function doesn't need to unreference the BOs on
+ * failure, because that will happen at panfrost_job_cleanup() time.
+ */
+static int
+panfrost_lookup_bos(struct drm_device *dev,
+		  struct drm_file *file_priv,
+		  struct drm_panfrost_submit *args,
+		  struct panfrost_job *job)
+{
+	job->bo_count = args->bo_handle_count;
+
+	if (!job->bo_count)
+		return 0;
+
+	job->implicit_fences = kvmalloc_array(job->bo_count,
+				  sizeof(struct dma_fence *),
+				  GFP_KERNEL | __GFP_ZERO);
+	if (!job->implicit_fences)
+		return -ENOMEM;
+
+	return drm_gem_objects_lookup(file_priv,
+				      (void __user *)(uintptr_t)args->bo_handles,
+				      job->bo_count, &job->bos);
+}
+
+/**
+ * panfrost_copy_in_sync() - Sets up job->in_fences[] with the sync objects
+ * referenced by the job.
+ * @dev: DRM device
+ * @file_priv: DRM file for this fd
+ * @args: IOCTL args
+ * @job: job being set up
+ *
+ * Resolve syncobjs from userspace to fences and attach them to job.
+ *
+ * Note that this function doesn't need to unreference the fences on
+ * failure, because that will happen at panfrost_job_cleanup() time.
+ */
+static int
+panfrost_copy_in_sync(struct drm_device *dev,
+		  struct drm_file *file_priv,
+		  struct drm_panfrost_submit *args,
+		  struct panfrost_job *job)
+{
+	u32 *handles;
+	int ret = 0;
+	int i;
+
+	job->in_fence_count = args->in_sync_count;
+
+	if (!job->in_fence_count)
+		return 0;
+
+	job->in_fences = kvmalloc_array(job->in_fence_count,
+					sizeof(struct dma_fence *),
+					GFP_KERNEL | __GFP_ZERO);
+	if (!job->in_fences) {
+		DRM_DEBUG("Failed to allocate job in fences\n");
+		return -ENOMEM;
+	}
+
+	handles = kvmalloc_array(job->in_fence_count, sizeof(u32), GFP_KERNEL);
+	if (!handles) {
+		ret = -ENOMEM;
+		DRM_DEBUG("Failed to allocate incoming syncobj handles\n");
+		goto fail;
+	}
+
+	if (copy_from_user(handles,
+			   (void __user *)(uintptr_t)args->in_syncs,
+			   job->in_fence_count * sizeof(u32))) {
+		ret = -EFAULT;
+		DRM_DEBUG("Failed to copy in syncobj handles\n");
+		goto fail;
+	}
+
+	for (i = 0; i < job->in_fence_count; i++) {
+		ret = drm_syncobj_find_fence(file_priv, handles[i], 0, 0,
+					     &job->in_fences[i]);
+		if (ret == -EINVAL)
+			goto fail;
+	}
+
+fail:
+	kvfree(handles);
+	return ret;
+}
+
+static int panfrost_ioctl_submit(struct drm_device *dev, void *data,
+		struct drm_file *file)
+{
+	struct panfrost_device *pfdev = dev->dev_private;
+	struct drm_panfrost_submit *args = data;
+	struct drm_syncobj *sync_out = NULL;
+	struct panfrost_job *job;
+	int ret = 0;
+
+	if (!args->jc)
+		return -EINVAL;
+
+	if (args->requirements && args->requirements != PANFROST_JD_REQ_FS)
+		return -EINVAL;
+
+	if (args->out_sync > 0) {
+		sync_out = drm_syncobj_find(file, args->out_sync);
+		if (!sync_out)
+			return -ENODEV;
+	}
+
+	job = kzalloc(sizeof(*job), GFP_KERNEL);
+	if (!job) {
+		ret = -ENOMEM;
+		goto fail_out_sync;
+	}
+
+	kref_init(&job->refcount);
+
+	job->pfdev = pfdev;
+	job->jc = args->jc;
+	job->requirements = args->requirements;
+	job->flush_id = panfrost_gpu_get_latest_flush_id(pfdev);
+	job->file_priv = file->driver_priv;
+
+	ret = panfrost_copy_in_sync(dev, file, args, job);
+	if (ret)
+		goto fail_job;
+
+	ret = panfrost_lookup_bos(dev, file, args, job);
+	if (ret)
+		goto fail_job;
+
+	ret = panfrost_job_push(job);
+	if (ret)
+		goto fail_job;
+
+	/* Update the return sync object for the job */
+	if (sync_out)
+		drm_syncobj_replace_fence(sync_out, job->render_done_fence);
+
+fail_job:
+	panfrost_job_put(job);
+fail_out_sync:
+	drm_syncobj_put(sync_out);
+
+	return ret;
+}
+
+static int
+panfrost_ioctl_wait_bo(struct drm_device *dev, void *data,
+		       struct drm_file *file_priv)
+{
+	long ret;
+	struct drm_panfrost_wait_bo *args = data;
+	struct drm_gem_object *gem_obj;
+	unsigned long timeout = drm_timeout_abs_to_jiffies(args->timeout_ns);
+
+	if (args->pad)
+		return -EINVAL;
+
+	gem_obj = drm_gem_object_lookup(file_priv, args->handle);
+	if (!gem_obj)
+		return -ENOENT;
+
+	ret = reservation_object_wait_timeout_rcu(gem_obj->resv, true,
+						  true, timeout);
+	if (!ret)
+		ret = timeout ? -ETIMEDOUT : -EBUSY;
+
+	drm_gem_object_put_unlocked(gem_obj);
+
+	return ret;
+}
+
+static int panfrost_ioctl_mmap_bo(struct drm_device *dev, void *data,
+		      struct drm_file *file_priv)
+{
+	struct drm_panfrost_mmap_bo *args = data;
+	struct drm_gem_object *gem_obj;
+	int ret;
+
+	if (args->flags != 0) {
+		DRM_INFO("unknown mmap_bo flags: %d\n", args->flags);
+		return -EINVAL;
+	}
+
+	gem_obj = drm_gem_object_lookup(file_priv, args->handle);
+	if (!gem_obj) {
+		DRM_DEBUG("Failed to look up GEM BO %d\n", args->handle);
+		return -ENOENT;
+	}
+
+	ret = drm_gem_create_mmap_offset(gem_obj);
+	if (ret == 0)
+		args->offset = drm_vma_node_offset_addr(&gem_obj->vma_node);
+	drm_gem_object_put_unlocked(gem_obj);
+
+	return ret;
+}
+
+static int panfrost_ioctl_get_bo_offset(struct drm_device *dev, void *data,
+			    struct drm_file *file_priv)
+{
+	struct drm_panfrost_get_bo_offset *args = data;
+	struct drm_gem_object *gem_obj;
+	struct panfrost_gem_object *bo;
+
+	gem_obj = drm_gem_object_lookup(file_priv, args->handle);
+	if (!gem_obj) {
+		DRM_DEBUG("Failed to look up GEM BO %d\n", args->handle);
+		return -ENOENT;
+	}
+	bo = to_panfrost_bo(gem_obj);
+
+	args->offset = bo->node.start << PAGE_SHIFT;
+
+	drm_gem_object_put_unlocked(gem_obj);
+	return 0;
+}
+
+static int
+panfrost_open(struct drm_device *dev, struct drm_file *file)
+{
+	struct panfrost_device *pfdev = dev->dev_private;
+	struct panfrost_file_priv *panfrost_priv;
+
+	panfrost_priv = kzalloc(sizeof(*panfrost_priv), GFP_KERNEL);
+	if (!panfrost_priv)
+		return -ENOMEM;
+
+	panfrost_priv->pfdev = pfdev;
+	file->driver_priv = panfrost_priv;
+
+	return panfrost_job_open(panfrost_priv);
+}
+
+static void
+panfrost_postclose(struct drm_device *dev, struct drm_file *file)
+{
+	struct panfrost_file_priv *panfrost_priv = file->driver_priv;
+
+	panfrost_job_close(panfrost_priv);
+
+	kfree(panfrost_priv);
+}
+
+/* DRM_AUTH is required on SUBMIT for now, while all clients share a single
+ * address space.  Note that render nodes would be able to submit jobs that
+ * could access BOs from clients authenticated with the master node.
+ */
+static const struct drm_ioctl_desc panfrost_drm_driver_ioctls[] = {
+#define PANFROST_IOCTL(n, func, flags) \
+	DRM_IOCTL_DEF_DRV(PANFROST_##n, panfrost_ioctl_##func, flags)
+
+	PANFROST_IOCTL(SUBMIT,		submit,		DRM_RENDER_ALLOW | DRM_AUTH),
+	PANFROST_IOCTL(WAIT_BO,		wait_bo,	DRM_RENDER_ALLOW),
+	PANFROST_IOCTL(CREATE_BO,	create_bo,	DRM_RENDER_ALLOW),
+	PANFROST_IOCTL(MMAP_BO,		mmap_bo,	DRM_RENDER_ALLOW),
+	PANFROST_IOCTL(GET_PARAM,	get_param,	DRM_RENDER_ALLOW),
+	PANFROST_IOCTL(GET_BO_OFFSET,	get_bo_offset,	DRM_RENDER_ALLOW),
+};
+
+DEFINE_DRM_GEM_SHMEM_FOPS(panfrost_drm_driver_fops);
+
+static struct drm_driver panfrost_drm_driver = {
+	.driver_features	= DRIVER_RENDER | DRIVER_GEM | DRIVER_PRIME |
+				  DRIVER_SYNCOBJ,
+	.open			= panfrost_open,
+	.postclose		= panfrost_postclose,
+	.ioctls			= panfrost_drm_driver_ioctls,
+	.num_ioctls		= ARRAY_SIZE(panfrost_drm_driver_ioctls),
+	.fops			= &panfrost_drm_driver_fops,
+	.name			= "panfrost",
+	.desc			= "panfrost DRM",
+	.date			= "20180908",
+	.major			= 1,
+	.minor			= 0,
+
+	.gem_create_object	= panfrost_gem_create_object,
+	.prime_handle_to_fd	= drm_gem_prime_handle_to_fd,
+	.prime_fd_to_handle	= drm_gem_prime_fd_to_handle,
+	.gem_prime_import_sg_table = panfrost_gem_prime_import_sg_table,
+	.gem_prime_mmap		= drm_gem_prime_mmap,
+};
+
+static int panfrost_probe(struct platform_device *pdev)
+{
+	struct panfrost_device *pfdev;
+	struct drm_device *ddev;
+	int err;
+
+	pfdev = devm_kzalloc(&pdev->dev, sizeof(*pfdev), GFP_KERNEL);
+	if (!pfdev)
+		return -ENOMEM;
+
+	pfdev->pdev = pdev;
+	pfdev->dev = &pdev->dev;
+
+	platform_set_drvdata(pdev, pfdev);
+
+	/* Allocate and initialze the DRM device. */
+	ddev = drm_dev_alloc(&panfrost_drm_driver, &pdev->dev);
+	if (IS_ERR(ddev))
+		return PTR_ERR(ddev);
+
+	ddev->dev_private = pfdev;
+	pfdev->ddev = ddev;
+
+	spin_lock_init(&pfdev->mm_lock);
+
+	/* 4G enough for now. can be 48-bit */
+	drm_mm_init(&pfdev->mm, SZ_32M >> PAGE_SHIFT, (SZ_4G - SZ_32M) >> PAGE_SHIFT);
+
+	pm_runtime_use_autosuspend(pfdev->dev);
+	pm_runtime_set_autosuspend_delay(pfdev->dev, 50); /* ~3 frames */
+	pm_runtime_enable(pfdev->dev);
+
+	err = panfrost_device_init(pfdev);
+	if (err) {
+		if (err != -EPROBE_DEFER)
+			dev_err(&pdev->dev, "Fatal error during GPU init\n");
+		goto err_out0;
+	}
+
+	err = panfrost_devfreq_init(pfdev);
+	if (err) {
+		if (err != -EPROBE_DEFER)
+			dev_err(&pdev->dev, "Fatal error during devfreq init\n");
+		goto err_out1;
+	}
+
+	/*
+	 * Register the DRM device with the core and the connectors with
+	 * sysfs
+	 */
+	err = drm_dev_register(ddev, 0);
+	if (err < 0)
+		goto err_out1;
+
+	return 0;
+
+err_out1:
+	panfrost_device_fini(pfdev);
+err_out0:
+	pm_runtime_disable(pfdev->dev);
+	drm_dev_put(ddev);
+	return err;
+}
+
+static int panfrost_remove(struct platform_device *pdev)
+{
+	struct panfrost_device *pfdev = platform_get_drvdata(pdev);
+	struct drm_device *ddev = pfdev->ddev;
+
+	drm_dev_unregister(ddev);
+	pm_runtime_get_sync(pfdev->dev);
+	pm_runtime_put_sync_autosuspend(pfdev->dev);
+	pm_runtime_disable(pfdev->dev);
+	panfrost_device_fini(pfdev);
+	drm_dev_put(ddev);
+	return 0;
+}
+
+static const struct of_device_id dt_match[] = {
+	{ .compatible = "arm,mali-t604" },
+	{ .compatible = "arm,mali-t624" },
+	{ .compatible = "arm,mali-t628" },
+	{ .compatible = "arm,mali-t720" },
+	{ .compatible = "arm,mali-t760" },
+	{ .compatible = "arm,mali-t820" },
+	{ .compatible = "arm,mali-t830" },
+	{ .compatible = "arm,mali-t860" },
+	{ .compatible = "arm,mali-t880" },
+	{}
+};
+MODULE_DEVICE_TABLE(of, dt_match);
+
+static const struct dev_pm_ops panfrost_pm_ops = {
+	SET_SYSTEM_SLEEP_PM_OPS(pm_runtime_force_suspend, pm_runtime_force_resume)
+	SET_RUNTIME_PM_OPS(panfrost_device_suspend, panfrost_device_resume, NULL)
+};
+
+static struct platform_driver panfrost_driver = {
+	.probe		= panfrost_probe,
+	.remove		= panfrost_remove,
+	.driver		= {
+		.name	= "panfrost",
+		.pm	= &panfrost_pm_ops,
+		.of_match_table = dt_match,
+	},
+};
+module_platform_driver(panfrost_driver);
+
+MODULE_AUTHOR("Panfrost Project Developers");
+MODULE_DESCRIPTION("Panfrost DRM Driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/gpu/drm/panfrost/panfrost_features.h b/drivers/gpu/drm/panfrost/panfrost_features.h
new file mode 100644
index 000000000000..5056777c7744
--- /dev/null
+++ b/drivers/gpu/drm/panfrost/panfrost_features.h
@@ -0,0 +1,309 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* (C) COPYRIGHT 2014-2018 ARM Limited. All rights reserved. */
+/* Copyright 2019 Linaro, Ltd., Rob Herring <robh@kernel.org> */
+#ifndef __PANFROST_FEATURES_H__
+#define __PANFROST_FEATURES_H__
+
+#include <linux/bitops.h>
+
+#include "panfrost_device.h"
+
+enum panfrost_hw_feature {
+	HW_FEATURE_JOBCHAIN_DISAMBIGUATION,
+	HW_FEATURE_PWRON_DURING_PWROFF_TRANS,
+	HW_FEATURE_XAFFINITY,
+	HW_FEATURE_OUT_OF_ORDER_EXEC,
+	HW_FEATURE_MRT,
+	HW_FEATURE_BRNDOUT_CC,
+	HW_FEATURE_INTERPIPE_REG_ALIASING,
+	HW_FEATURE_LD_ST_TILEBUFFER,
+	HW_FEATURE_MSAA_16X,
+	HW_FEATURE_32_BIT_UNIFORM_ADDRESS,
+	HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL,
+	HW_FEATURE_OPTIMIZED_COVERAGE_MASK,
+	HW_FEATURE_T7XX_PAIRING_RULES,
+	HW_FEATURE_LD_ST_LEA_TEX,
+	HW_FEATURE_LINEAR_FILTER_FLOAT,
+	HW_FEATURE_WORKGROUP_ROUND_MULTIPLE_OF_4,
+	HW_FEATURE_IMAGES_IN_FRAGMENT_SHADERS,
+	HW_FEATURE_TEST4_DATUM_MODE,
+	HW_FEATURE_NEXT_INSTRUCTION_TYPE,
+	HW_FEATURE_BRNDOUT_KILL,
+	HW_FEATURE_WARPING,
+	HW_FEATURE_V4,
+	HW_FEATURE_FLUSH_REDUCTION,
+	HW_FEATURE_PROTECTED_MODE,
+	HW_FEATURE_COHERENCY_REG,
+	HW_FEATURE_PROTECTED_DEBUG_MODE,
+	HW_FEATURE_AARCH64_MMU,
+	HW_FEATURE_TLS_HASHING,
+	HW_FEATURE_THREAD_GROUP_SPLIT,
+	HW_FEATURE_3BIT_EXT_RW_L2_MMU_CONFIG,
+};
+
+#define hw_features_t600 (\
+	BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \
+	BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \
+	BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \
+	BIT_ULL(HW_FEATURE_V4))
+
+#define hw_features_t620 (\
+	BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \
+	BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \
+	BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \
+	BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \
+	BIT_ULL(HW_FEATURE_V4))
+
+#define hw_features_t720 (\
+	BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \
+	BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \
+	BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \
+	BIT_ULL(HW_FEATURE_OPTIMIZED_COVERAGE_MASK) | \
+	BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \
+	BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \
+	BIT_ULL(HW_FEATURE_WORKGROUP_ROUND_MULTIPLE_OF_4) | \
+	BIT_ULL(HW_FEATURE_WARPING) | \
+	BIT_ULL(HW_FEATURE_V4))
+
+
+#define hw_features_t760 (\
+	BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \
+	BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \
+	BIT_ULL(HW_FEATURE_XAFFINITY) | \
+	BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \
+	BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \
+	BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \
+	BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \
+	BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \
+	BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \
+	BIT_ULL(HW_FEATURE_MRT) | \
+	BIT_ULL(HW_FEATURE_MSAA_16X) | \
+	BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \
+	BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \
+	BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \
+	BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT))
+
+// T860
+#define hw_features_t860 (\
+	BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \
+	BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \
+	BIT_ULL(HW_FEATURE_XAFFINITY) | \
+	BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \
+	BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \
+	BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \
+	BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \
+	BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \
+	BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \
+	BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \
+	BIT_ULL(HW_FEATURE_MRT) | \
+	BIT_ULL(HW_FEATURE_MSAA_16X) | \
+	BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \
+	BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \
+	BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \
+	BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \
+	BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT))
+
+#define hw_features_t880 hw_features_t860
+
+#define hw_features_t830 (\
+	BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \
+	BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \
+	BIT_ULL(HW_FEATURE_XAFFINITY) | \
+	BIT_ULL(HW_FEATURE_WARPING) | \
+	BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \
+	BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \
+	BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \
+	BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \
+	BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \
+	BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \
+	BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \
+	BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \
+	BIT_ULL(HW_FEATURE_MRT) | \
+	BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \
+	BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \
+	BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \
+	BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \
+	BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT))
+
+#define hw_features_t820 (\
+	BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \
+	BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \
+	BIT_ULL(HW_FEATURE_XAFFINITY) | \
+	BIT_ULL(HW_FEATURE_WARPING) | \
+	BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \
+	BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \
+	BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \
+	BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \
+	BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \
+	BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \
+	BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \
+	BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \
+	BIT_ULL(HW_FEATURE_MRT) | \
+	BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \
+	BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \
+	BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \
+	BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \
+	BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT))
+
+#define hw_features_g71 (\
+	BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \
+	BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \
+	BIT_ULL(HW_FEATURE_XAFFINITY) | \
+	BIT_ULL(HW_FEATURE_WARPING) | \
+	BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \
+	BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \
+	BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \
+	BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \
+	BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \
+	BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \
+	BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \
+	BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \
+	BIT_ULL(HW_FEATURE_MRT) | \
+	BIT_ULL(HW_FEATURE_MSAA_16X) | \
+	BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \
+	BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \
+	BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \
+	BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \
+	BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \
+	BIT_ULL(HW_FEATURE_FLUSH_REDUCTION) | \
+	BIT_ULL(HW_FEATURE_PROTECTED_MODE) | \
+	BIT_ULL(HW_FEATURE_COHERENCY_REG))
+
+#define hw_features_g72 (\
+	BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \
+	BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \
+	BIT_ULL(HW_FEATURE_XAFFINITY) | \
+	BIT_ULL(HW_FEATURE_WARPING) | \
+	BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \
+	BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \
+	BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \
+	BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \
+	BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \
+	BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \
+	BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \
+	BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \
+	BIT_ULL(HW_FEATURE_MRT) | \
+	BIT_ULL(HW_FEATURE_MSAA_16X) | \
+	BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \
+	BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \
+	BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \
+	BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \
+	BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \
+	BIT_ULL(HW_FEATURE_FLUSH_REDUCTION) | \
+	BIT_ULL(HW_FEATURE_PROTECTED_MODE) | \
+	BIT_ULL(HW_FEATURE_PROTECTED_DEBUG_MODE) | \
+	BIT_ULL(HW_FEATURE_COHERENCY_REG))
+
+#define hw_features_g51 (\
+	BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \
+	BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \
+	BIT_ULL(HW_FEATURE_XAFFINITY) | \
+	BIT_ULL(HW_FEATURE_WARPING) | \
+	BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \
+	BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \
+	BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \
+	BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \
+	BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \
+	BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \
+	BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \
+	BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \
+	BIT_ULL(HW_FEATURE_MRT) | \
+	BIT_ULL(HW_FEATURE_MSAA_16X) | \
+	BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \
+	BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \
+	BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \
+	BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \
+	BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \
+	BIT_ULL(HW_FEATURE_FLUSH_REDUCTION) | \
+	BIT_ULL(HW_FEATURE_PROTECTED_MODE) | \
+	BIT_ULL(HW_FEATURE_PROTECTED_DEBUG_MODE) | \
+	BIT_ULL(HW_FEATURE_COHERENCY_REG))
+
+#define hw_features_g52 (\
+	BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \
+	BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \
+	BIT_ULL(HW_FEATURE_XAFFINITY) | \
+	BIT_ULL(HW_FEATURE_WARPING) | \
+	BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \
+	BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \
+	BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \
+	BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \
+	BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \
+	BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \
+	BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \
+	BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \
+	BIT_ULL(HW_FEATURE_MRT) | \
+	BIT_ULL(HW_FEATURE_MSAA_16X) | \
+	BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \
+	BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \
+	BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \
+	BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \
+	BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \
+	BIT_ULL(HW_FEATURE_FLUSH_REDUCTION) | \
+	BIT_ULL(HW_FEATURE_PROTECTED_MODE) | \
+	BIT_ULL(HW_FEATURE_PROTECTED_DEBUG_MODE) | \
+	BIT_ULL(HW_FEATURE_COHERENCY_REG))
+
+#define hw_features_g76 (\
+	BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \
+	BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \
+	BIT_ULL(HW_FEATURE_XAFFINITY) | \
+	BIT_ULL(HW_FEATURE_WARPING) | \
+	BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \
+	BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \
+	BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \
+	BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \
+	BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \
+	BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \
+	BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \
+	BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \
+	BIT_ULL(HW_FEATURE_MRT) | \
+	BIT_ULL(HW_FEATURE_MSAA_16X) | \
+	BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \
+	BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \
+	BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \
+	BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \
+	BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \
+	BIT_ULL(HW_FEATURE_FLUSH_REDUCTION) | \
+	BIT_ULL(HW_FEATURE_PROTECTED_MODE) | \
+	BIT_ULL(HW_FEATURE_PROTECTED_DEBUG_MODE) | \
+	BIT_ULL(HW_FEATURE_COHERENCY_REG) | \
+	BIT_ULL(HW_FEATURE_AARCH64_MMU) | \
+	BIT_ULL(HW_FEATURE_TLS_HASHING) | \
+	BIT_ULL(HW_FEATURE_3BIT_EXT_RW_L2_MMU_CONFIG))
+
+#define hw_features_g31 (\
+	BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \
+	BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \
+	BIT_ULL(HW_FEATURE_XAFFINITY) | \
+	BIT_ULL(HW_FEATURE_WARPING) | \
+	BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \
+	BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \
+	BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \
+	BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \
+	BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \
+	BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \
+	BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \
+	BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \
+	BIT_ULL(HW_FEATURE_MRT) | \
+	BIT_ULL(HW_FEATURE_MSAA_16X) | \
+	BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \
+	BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \
+	BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \
+	BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \
+	BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \
+	BIT_ULL(HW_FEATURE_FLUSH_REDUCTION) | \
+	BIT_ULL(HW_FEATURE_PROTECTED_MODE) | \
+	BIT_ULL(HW_FEATURE_PROTECTED_DEBUG_MODE) | \
+	BIT_ULL(HW_FEATURE_COHERENCY_REG) | \
+	BIT_ULL(HW_FEATURE_AARCH64_MMU) | \
+	BIT_ULL(HW_FEATURE_TLS_HASHING) | \
+	BIT_ULL(HW_FEATURE_3BIT_EXT_RW_L2_MMU_CONFIG))
+
+static inline bool panfrost_has_hw_feature(struct panfrost_device *pfdev,
+					   enum panfrost_hw_feature feat)
+{
+	return test_bit(feat, pfdev->features.hw_features);
+}
+
+#endif
diff --git a/drivers/gpu/drm/panfrost/panfrost_gem.c b/drivers/gpu/drm/panfrost/panfrost_gem.c
new file mode 100644
index 000000000000..a5528a360ef4
--- /dev/null
+++ b/drivers/gpu/drm/panfrost/panfrost_gem.c
@@ -0,0 +1,95 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright 2019 Linaro, Ltd, Rob Herring <robh@kernel.org> */
+
+#include <linux/err.h>
+#include <linux/slab.h>
+#include <linux/dma-buf.h>
+#include <linux/dma-mapping.h>
+
+#include <drm/panfrost_drm.h>
+#include "panfrost_device.h"
+#include "panfrost_gem.h"
+#include "panfrost_mmu.h"
+
+/* Called DRM core on the last userspace/kernel unreference of the
+ * BO.
+ */
+static void panfrost_gem_free_object(struct drm_gem_object *obj)
+{
+	struct panfrost_gem_object *bo = to_panfrost_bo(obj);
+	struct panfrost_device *pfdev = obj->dev->dev_private;
+
+	panfrost_mmu_unmap(bo);
+
+	spin_lock(&pfdev->mm_lock);
+	drm_mm_remove_node(&bo->node);
+	spin_unlock(&pfdev->mm_lock);
+
+	drm_gem_shmem_free_object(obj);
+}
+
+static const struct drm_gem_object_funcs panfrost_gem_funcs = {
+	.free = panfrost_gem_free_object,
+	.print_info = drm_gem_shmem_print_info,
+	.pin = drm_gem_shmem_pin,
+	.unpin = drm_gem_shmem_unpin,
+	.get_sg_table = drm_gem_shmem_get_sg_table,
+	.vmap = drm_gem_shmem_vmap,
+	.vunmap = drm_gem_shmem_vunmap,
+	.vm_ops = &drm_gem_shmem_vm_ops,
+};
+
+/**
+ * panfrost_gem_create_object - Implementation of driver->gem_create_object.
+ * @dev: DRM device
+ * @size: Size in bytes of the memory the object will reference
+ *
+ * This lets the GEM helpers allocate object structs for us, and keep
+ * our BO stats correct.
+ */
+struct drm_gem_object *panfrost_gem_create_object(struct drm_device *dev, size_t size)
+{
+	int ret;
+	struct panfrost_device *pfdev = dev->dev_private;
+	struct panfrost_gem_object *obj;
+
+	obj = kzalloc(sizeof(*obj), GFP_KERNEL);
+	if (!obj)
+		return NULL;
+
+	obj->base.base.funcs = &panfrost_gem_funcs;
+
+	spin_lock(&pfdev->mm_lock);
+	ret = drm_mm_insert_node(&pfdev->mm, &obj->node,
+				 roundup(size, PAGE_SIZE) >> PAGE_SHIFT);
+	spin_unlock(&pfdev->mm_lock);
+	if (ret)
+		goto free_obj;
+
+	return &obj->base.base;
+
+free_obj:
+	kfree(obj);
+	return ERR_PTR(ret);
+}
+
+struct drm_gem_object *
+panfrost_gem_prime_import_sg_table(struct drm_device *dev,
+				   struct dma_buf_attachment *attach,
+				   struct sg_table *sgt)
+{
+	struct drm_gem_object *obj;
+	struct panfrost_gem_object *pobj;
+
+	obj = drm_gem_shmem_prime_import_sg_table(dev, attach, sgt);
+	if (IS_ERR(obj))
+		return ERR_CAST(obj);
+
+	pobj = to_panfrost_bo(obj);
+
+	obj->resv = attach->dmabuf->resv;
+
+	panfrost_mmu_map(pobj);
+
+	return obj;
+}
diff --git a/drivers/gpu/drm/panfrost/panfrost_gem.h b/drivers/gpu/drm/panfrost/panfrost_gem.h
new file mode 100644
index 000000000000..045000eb5fcf
--- /dev/null
+++ b/drivers/gpu/drm/panfrost/panfrost_gem.h
@@ -0,0 +1,29 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright 2019 Linaro, Ltd, Rob Herring <robh@kernel.org> */
+
+#ifndef __PANFROST_GEM_H__
+#define __PANFROST_GEM_H__
+
+#include <drm/drm_gem_shmem_helper.h>
+#include <drm/drm_mm.h>
+
+struct panfrost_gem_object {
+	struct drm_gem_shmem_object base;
+
+	struct drm_mm_node node;
+};
+
+static inline
+struct  panfrost_gem_object *to_panfrost_bo(struct drm_gem_object *obj)
+{
+	return container_of(to_drm_gem_shmem_obj(obj), struct panfrost_gem_object, base);
+}
+
+struct drm_gem_object *panfrost_gem_create_object(struct drm_device *dev, size_t size);
+
+struct drm_gem_object *
+panfrost_gem_prime_import_sg_table(struct drm_device *dev,
+				   struct dma_buf_attachment *attach,
+				   struct sg_table *sgt);
+
+#endif /* __PANFROST_GEM_H__ */
diff --git a/drivers/gpu/drm/panfrost/panfrost_gpu.c b/drivers/gpu/drm/panfrost/panfrost_gpu.c
new file mode 100644
index 000000000000..58ef25573cda
--- /dev/null
+++ b/drivers/gpu/drm/panfrost/panfrost_gpu.c
@@ -0,0 +1,367 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright 2018 Marty E. Plummer <hanetzer@startmail.com> */
+/* Copyright 2019 Linaro, Ltd., Rob Herring <robh@kernel.org> */
+/* Copyright 2019 Collabora ltd. */
+#include <linux/bitfield.h>
+#include <linux/bitmap.h>
+#include <linux/delay.h>
+#include <linux/dma-mapping.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/iopoll.h>
+#include <linux/platform_device.h>
+
+#include "panfrost_device.h"
+#include "panfrost_features.h"
+#include "panfrost_issues.h"
+#include "panfrost_gpu.h"
+#include "panfrost_regs.h"
+
+#define gpu_write(dev, reg, data) writel(data, dev->iomem + reg)
+#define gpu_read(dev, reg) readl(dev->iomem + reg)
+
+static irqreturn_t panfrost_gpu_irq_handler(int irq, void *data)
+{
+	struct panfrost_device *pfdev = data;
+	u32 state = gpu_read(pfdev, GPU_INT_STAT);
+	u32 fault_status = gpu_read(pfdev, GPU_FAULT_STATUS);
+
+	if (!state)
+		return IRQ_NONE;
+
+	if (state & GPU_IRQ_MASK_ERROR) {
+		u64 address = (u64) gpu_read(pfdev, GPU_FAULT_ADDRESS_HI) << 32;
+		address |= gpu_read(pfdev, GPU_FAULT_ADDRESS_LO);
+
+		dev_warn(pfdev->dev, "GPU Fault 0x%08x (%s) at 0x%016llx\n",
+			 fault_status & 0xFF, panfrost_exception_name(pfdev, fault_status),
+			 address);
+
+		if (state & GPU_IRQ_MULTIPLE_FAULT)
+			dev_warn(pfdev->dev, "There were multiple GPU faults - some have not been reported\n");
+
+		gpu_write(pfdev, GPU_INT_MASK, 0);
+	}
+
+	gpu_write(pfdev, GPU_INT_CLEAR, state);
+
+	return IRQ_HANDLED;
+}
+
+int panfrost_gpu_soft_reset(struct panfrost_device *pfdev)
+{
+	int ret;
+	u32 val;
+
+	gpu_write(pfdev, GPU_INT_MASK, 0);
+	gpu_write(pfdev, GPU_INT_CLEAR, GPU_IRQ_RESET_COMPLETED);
+	gpu_write(pfdev, GPU_CMD, GPU_CMD_SOFT_RESET);
+
+	ret = readl_relaxed_poll_timeout(pfdev->iomem + GPU_INT_RAWSTAT,
+		val, val & GPU_IRQ_RESET_COMPLETED, 100, 10000);
+
+	if (ret) {
+		dev_err(pfdev->dev, "gpu soft reset timed out\n");
+		return ret;
+	}
+
+	gpu_write(pfdev, GPU_INT_CLEAR, GPU_IRQ_MASK_ALL);
+	gpu_write(pfdev, GPU_INT_MASK, GPU_IRQ_MASK_ALL);
+
+	return 0;
+}
+
+static void panfrost_gpu_init_quirks(struct panfrost_device *pfdev)
+{
+	u32 quirks = 0;
+
+	if (panfrost_has_hw_issue(pfdev, HW_ISSUE_8443) ||
+	    panfrost_has_hw_issue(pfdev, HW_ISSUE_11035))
+		quirks |= SC_LS_PAUSEBUFFER_DISABLE;
+
+	if (panfrost_has_hw_issue(pfdev, HW_ISSUE_10327))
+		quirks |= SC_SDC_DISABLE_OQ_DISCARD;
+
+	if (panfrost_has_hw_issue(pfdev, HW_ISSUE_10797))
+		quirks |= SC_ENABLE_TEXGRD_FLAGS;
+
+	if (!panfrost_has_hw_issue(pfdev, GPUCORE_1619)) {
+		if (panfrost_model_cmp(pfdev, 0x750) < 0) /* T60x, T62x, T72x */
+			quirks |= SC_LS_ATTR_CHECK_DISABLE;
+		else if (panfrost_model_cmp(pfdev, 0x880) <= 0) /* T76x, T8xx */
+			quirks |= SC_LS_ALLOW_ATTR_TYPES;
+	}
+
+	if (panfrost_has_hw_feature(pfdev, HW_FEATURE_TLS_HASHING))
+		quirks |= SC_TLS_HASH_ENABLE;
+
+	if (quirks)
+		gpu_write(pfdev, GPU_SHADER_CONFIG, quirks);
+
+
+	quirks = gpu_read(pfdev, GPU_TILER_CONFIG);
+
+	/* Set tiler clock gate override if required */
+	if (panfrost_has_hw_issue(pfdev, HW_ISSUE_T76X_3953))
+		quirks |= TC_CLOCK_GATE_OVERRIDE;
+
+	gpu_write(pfdev, GPU_TILER_CONFIG, quirks);
+
+
+	quirks = gpu_read(pfdev, GPU_L2_MMU_CONFIG);
+
+	/* Limit read & write ID width for AXI */
+	if (panfrost_has_hw_feature(pfdev, HW_FEATURE_3BIT_EXT_RW_L2_MMU_CONFIG))
+		quirks &= ~(L2_MMU_CONFIG_3BIT_LIMIT_EXTERNAL_READS |
+			    L2_MMU_CONFIG_3BIT_LIMIT_EXTERNAL_WRITES);
+	else
+		quirks &= ~(L2_MMU_CONFIG_LIMIT_EXTERNAL_READS |
+			    L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES);
+
+	gpu_write(pfdev, GPU_L2_MMU_CONFIG, quirks);
+
+	quirks = 0;
+	if ((panfrost_model_eq(pfdev, 0x860) || panfrost_model_eq(pfdev, 0x880)) &&
+	    pfdev->features.revision >= 0x2000)
+		quirks |= JM_MAX_JOB_THROTTLE_LIMIT << JM_JOB_THROTTLE_LIMIT_SHIFT;
+	else if (panfrost_model_eq(pfdev, 0x6000) &&
+		 pfdev->features.coherency_features == COHERENCY_ACE)
+		quirks |= (COHERENCY_ACE_LITE | COHERENCY_ACE) <<
+			   JM_FORCE_COHERENCY_FEATURES_SHIFT;
+
+	if (quirks)
+		gpu_write(pfdev, GPU_JM_CONFIG, quirks);
+}
+
+#define MAX_HW_REVS 6
+
+struct panfrost_model {
+	const char *name;
+	u32 id;
+	u32 id_mask;
+	u64 features;
+	u64 issues;
+	struct {
+		u32 revision;
+		u64 issues;
+	} revs[MAX_HW_REVS];
+};
+
+#define GPU_MODEL(_name, _id, ...) \
+{\
+	.name = __stringify(_name),				\
+	.id = _id,						\
+	.features = hw_features_##_name,			\
+	.issues = hw_issues_##_name,				\
+	.revs = { __VA_ARGS__ },				\
+}
+
+#define GPU_REV_EXT(name, _rev, _p, _s, stat) \
+{\
+	.revision = (_rev) << 12 | (_p) << 4 | (_s),		\
+	.issues = hw_issues_##name##_r##_rev##p##_p##stat,	\
+}
+#define GPU_REV(name, r, p) GPU_REV_EXT(name, r, p, 0, )
+
+static const struct panfrost_model gpu_models[] = {
+	/* T60x has an oddball version */
+	GPU_MODEL(t600, 0x600,
+		GPU_REV_EXT(t600, 0, 0, 1, _15dev0)),
+	GPU_MODEL(t620, 0x620,
+		GPU_REV(t620, 0, 1), GPU_REV(t620, 1, 0)),
+	GPU_MODEL(t720, 0x720),
+	GPU_MODEL(t760, 0x750,
+		GPU_REV(t760, 0, 0), GPU_REV(t760, 0, 1),
+		GPU_REV_EXT(t760, 0, 1, 0, _50rel0),
+		GPU_REV(t760, 0, 2), GPU_REV(t760, 0, 3)),
+	GPU_MODEL(t820, 0x820),
+	GPU_MODEL(t830, 0x830),
+	GPU_MODEL(t860, 0x860),
+	GPU_MODEL(t880, 0x880),
+
+	GPU_MODEL(g71, 0x6000,
+		GPU_REV_EXT(g71, 0, 0, 1, _05dev0)),
+	GPU_MODEL(g72, 0x6001),
+	GPU_MODEL(g51, 0x7000),
+	GPU_MODEL(g76, 0x7001),
+	GPU_MODEL(g52, 0x7002),
+	GPU_MODEL(g31, 0x7003,
+		GPU_REV(g31, 1, 0)),
+};
+
+static void panfrost_gpu_init_features(struct panfrost_device *pfdev)
+{
+	u32 gpu_id, num_js, major, minor, status, rev;
+	const char *name = "unknown";
+	u64 hw_feat = 0;
+	u64 hw_issues = hw_issues_all;
+	const struct panfrost_model *model;
+	int i;
+
+	pfdev->features.l2_features = gpu_read(pfdev, GPU_L2_FEATURES);
+	pfdev->features.core_features = gpu_read(pfdev, GPU_CORE_FEATURES);
+	pfdev->features.tiler_features = gpu_read(pfdev, GPU_TILER_FEATURES);
+	pfdev->features.mem_features = gpu_read(pfdev, GPU_MEM_FEATURES);
+	pfdev->features.mmu_features = gpu_read(pfdev, GPU_MMU_FEATURES);
+	pfdev->features.thread_features = gpu_read(pfdev, GPU_THREAD_FEATURES);
+	pfdev->features.coherency_features = gpu_read(pfdev, GPU_COHERENCY_FEATURES);
+	for (i = 0; i < 4; i++)
+		pfdev->features.texture_features[i] = gpu_read(pfdev, GPU_TEXTURE_FEATURES(i));
+
+	pfdev->features.as_present = gpu_read(pfdev, GPU_AS_PRESENT);
+
+	pfdev->features.js_present = gpu_read(pfdev, GPU_JS_PRESENT);
+	num_js = hweight32(pfdev->features.js_present);
+	for (i = 0; i < num_js; i++)
+		pfdev->features.js_features[i] = gpu_read(pfdev, GPU_JS_FEATURES(i));
+
+	pfdev->features.shader_present = gpu_read(pfdev, GPU_SHADER_PRESENT_LO);
+	pfdev->features.shader_present |= (u64)gpu_read(pfdev, GPU_SHADER_PRESENT_HI) << 32;
+
+	pfdev->features.tiler_present = gpu_read(pfdev, GPU_TILER_PRESENT_LO);
+	pfdev->features.tiler_present |= (u64)gpu_read(pfdev, GPU_TILER_PRESENT_HI) << 32;
+
+	pfdev->features.l2_present = gpu_read(pfdev, GPU_L2_PRESENT_LO);
+	pfdev->features.l2_present |= (u64)gpu_read(pfdev, GPU_L2_PRESENT_HI) << 32;
+	pfdev->features.nr_core_groups = hweight64(pfdev->features.l2_present);
+
+	pfdev->features.stack_present = gpu_read(pfdev, GPU_STACK_PRESENT_LO);
+	pfdev->features.stack_present |= (u64)gpu_read(pfdev, GPU_STACK_PRESENT_HI) << 32;
+
+	gpu_id = gpu_read(pfdev, GPU_ID);
+	pfdev->features.revision = gpu_id & 0xffff;
+	pfdev->features.id = gpu_id >> 16;
+
+	/* The T60x has an oddball ID value. Fix it up to the standard Midgard
+	 * format so we (and userspace) don't have to special case it.
+	 */
+	if (pfdev->features.id == 0x6956)
+		pfdev->features.id = 0x0600;
+
+	major = (pfdev->features.revision >> 12) & 0xf;
+	minor = (pfdev->features.revision >> 4) & 0xff;
+	status = pfdev->features.revision & 0xf;
+	rev = pfdev->features.revision;
+
+	gpu_id = pfdev->features.id;
+
+	for (model = gpu_models; model->name; model++) {
+		int best = -1;
+
+		if (!panfrost_model_eq(pfdev, model->id))
+			continue;
+
+		name = model->name;
+		hw_feat = model->features;
+		hw_issues |= model->issues;
+		for (i = 0; i < MAX_HW_REVS; i++) {
+			if (model->revs[i].revision == rev) {
+				best = i;
+				break;
+			} else if (model->revs[i].revision == (rev & ~0xf))
+				best = i;
+		}
+
+		if (best >= 0)
+			hw_issues |= model->revs[best].issues;
+
+		break;
+	}
+
+	bitmap_from_u64(pfdev->features.hw_features, hw_feat);
+	bitmap_from_u64(pfdev->features.hw_issues, hw_issues);
+
+	dev_info(pfdev->dev, "mali-%s id 0x%x major 0x%x minor 0x%x status 0x%x",
+		 name, gpu_id, major, minor, status);
+	dev_info(pfdev->dev, "features: %64pb, issues: %64pb",
+		 pfdev->features.hw_features,
+		 pfdev->features.hw_issues);
+
+	dev_info(pfdev->dev, "Features: L2:0x%08x Shader:0x%08x Tiler:0x%08x Mem:0x%0x MMU:0x%08x AS:0x%x JS:0x%x",
+		 pfdev->features.l2_features,
+		 pfdev->features.core_features,
+		 pfdev->features.tiler_features,
+		 pfdev->features.mem_features,
+		 pfdev->features.mmu_features,
+		 pfdev->features.as_present,
+		 pfdev->features.js_present);
+
+	dev_info(pfdev->dev, "shader_present=0x%0llx l2_present=0x%0llx",
+		 pfdev->features.shader_present, pfdev->features.l2_present);
+}
+
+void panfrost_gpu_power_on(struct panfrost_device *pfdev)
+{
+	int ret;
+	u32 val;
+
+	/* Just turn on everything for now */
+	gpu_write(pfdev, L2_PWRON_LO, pfdev->features.l2_present);
+	ret = readl_relaxed_poll_timeout(pfdev->iomem + L2_READY_LO,
+		val, val == pfdev->features.l2_present, 100, 1000);
+
+	gpu_write(pfdev, STACK_PWRON_LO, pfdev->features.stack_present);
+	ret |= readl_relaxed_poll_timeout(pfdev->iomem + STACK_READY_LO,
+		val, val == pfdev->features.stack_present, 100, 1000);
+
+	gpu_write(pfdev, SHADER_PWRON_LO, pfdev->features.shader_present);
+	ret |= readl_relaxed_poll_timeout(pfdev->iomem + SHADER_READY_LO,
+		val, val == pfdev->features.shader_present, 100, 1000);
+
+	gpu_write(pfdev, TILER_PWRON_LO, pfdev->features.tiler_present);
+	ret |= readl_relaxed_poll_timeout(pfdev->iomem + TILER_READY_LO,
+		val, val == pfdev->features.tiler_present, 100, 1000);
+
+	if (ret)
+		dev_err(pfdev->dev, "error powering up gpu");
+}
+
+void panfrost_gpu_power_off(struct panfrost_device *pfdev)
+{
+	gpu_write(pfdev, TILER_PWROFF_LO, 0);
+	gpu_write(pfdev, SHADER_PWROFF_LO, 0);
+	gpu_write(pfdev, STACK_PWROFF_LO, 0);
+	gpu_write(pfdev, L2_PWROFF_LO, 0);
+}
+
+int panfrost_gpu_init(struct panfrost_device *pfdev)
+{
+	int err, irq;
+
+	err = panfrost_gpu_soft_reset(pfdev);
+	if (err)
+		return err;
+
+	panfrost_gpu_init_features(pfdev);
+
+	dma_set_mask_and_coherent(pfdev->dev,
+		DMA_BIT_MASK(FIELD_GET(0xff00, pfdev->features.mmu_features)));
+
+	irq = platform_get_irq_byname(to_platform_device(pfdev->dev), "gpu");
+	if (irq <= 0)
+		return -ENODEV;
+
+	err = devm_request_irq(pfdev->dev, irq, panfrost_gpu_irq_handler,
+			       IRQF_SHARED, "gpu", pfdev);
+	if (err) {
+		dev_err(pfdev->dev, "failed to request gpu irq");
+		return err;
+	}
+
+	panfrost_gpu_init_quirks(pfdev);
+	panfrost_gpu_power_on(pfdev);
+
+	return 0;
+}
+
+void panfrost_gpu_fini(struct panfrost_device *pfdev)
+{
+	panfrost_gpu_power_off(pfdev);
+}
+
+u32 panfrost_gpu_get_latest_flush_id(struct panfrost_device *pfdev)
+{
+	if (panfrost_has_hw_feature(pfdev, HW_FEATURE_FLUSH_REDUCTION))
+		return gpu_read(pfdev, GPU_LATEST_FLUSH_ID);
+	return 0;
+}
diff --git a/drivers/gpu/drm/panfrost/panfrost_gpu.h b/drivers/gpu/drm/panfrost/panfrost_gpu.h
new file mode 100644
index 000000000000..4112412087b2
--- /dev/null
+++ b/drivers/gpu/drm/panfrost/panfrost_gpu.h
@@ -0,0 +1,19 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright 2018 Marty E. Plummer <hanetzer@startmail.com> */
+/* Copyright 2019 Collabora ltd. */
+
+#ifndef __PANFROST_GPU_H__
+#define __PANFROST_GPU_H__
+
+struct panfrost_device;
+
+int panfrost_gpu_init(struct panfrost_device *pfdev);
+void panfrost_gpu_fini(struct panfrost_device *pfdev);
+
+u32 panfrost_gpu_get_latest_flush_id(struct panfrost_device *pfdev);
+
+int panfrost_gpu_soft_reset(struct panfrost_device *pfdev);
+void panfrost_gpu_power_on(struct panfrost_device *pfdev);
+void panfrost_gpu_power_off(struct panfrost_device *pfdev);
+
+#endif
diff --git a/drivers/gpu/drm/panfrost/panfrost_issues.h b/drivers/gpu/drm/panfrost/panfrost_issues.h
new file mode 100644
index 000000000000..cec6dcdadb5c
--- /dev/null
+++ b/drivers/gpu/drm/panfrost/panfrost_issues.h
@@ -0,0 +1,176 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* (C) COPYRIGHT 2014-2018 ARM Limited. All rights reserved. */
+/* Copyright 2019 Linaro, Ltd., Rob Herring <robh@kernel.org> */
+#ifndef __PANFROST_ISSUES_H__
+#define __PANFROST_ISSUES_H__
+
+#include <linux/bitops.h>
+
+#include "panfrost_device.h"
+
+/*
+ * This is not a complete list of issues, but only the ones the driver needs
+ * to care about.
+ */
+enum panfrost_hw_issue {
+	HW_ISSUE_6367,
+	HW_ISSUE_6787,
+	HW_ISSUE_8186,
+	HW_ISSUE_8245,
+	HW_ISSUE_8316,
+	HW_ISSUE_8394,
+	HW_ISSUE_8401,
+	HW_ISSUE_8408,
+	HW_ISSUE_8443,
+	HW_ISSUE_8987,
+	HW_ISSUE_9435,
+	HW_ISSUE_9510,
+	HW_ISSUE_9630,
+	HW_ISSUE_10327,
+	HW_ISSUE_10649,
+	HW_ISSUE_10676,
+	HW_ISSUE_10797,
+	HW_ISSUE_10817,
+	HW_ISSUE_10883,
+	HW_ISSUE_10959,
+	HW_ISSUE_10969,
+	HW_ISSUE_11020,
+	HW_ISSUE_11024,
+	HW_ISSUE_11035,
+	HW_ISSUE_11056,
+	HW_ISSUE_T76X_3542,
+	HW_ISSUE_T76X_3953,
+	HW_ISSUE_TMIX_8463,
+	GPUCORE_1619,
+	HW_ISSUE_TMIX_8438,
+	HW_ISSUE_TGOX_R1_1234,
+	HW_ISSUE_END
+};
+
+#define hw_issues_all (\
+	BIT_ULL(HW_ISSUE_9435))
+
+#define hw_issues_t600 (\
+	BIT_ULL(HW_ISSUE_6367) | \
+	BIT_ULL(HW_ISSUE_6787) | \
+	BIT_ULL(HW_ISSUE_8408) | \
+	BIT_ULL(HW_ISSUE_9510) | \
+	BIT_ULL(HW_ISSUE_10649) | \
+	BIT_ULL(HW_ISSUE_10676) | \
+	BIT_ULL(HW_ISSUE_10883) | \
+	BIT_ULL(HW_ISSUE_11020) | \
+	BIT_ULL(HW_ISSUE_11035) | \
+	BIT_ULL(HW_ISSUE_11056) | \
+	BIT_ULL(HW_ISSUE_TMIX_8438))
+
+#define hw_issues_t600_r0p0_15dev0 (\
+	BIT_ULL(HW_ISSUE_8186) | \
+	BIT_ULL(HW_ISSUE_8245) | \
+	BIT_ULL(HW_ISSUE_8316) | \
+	BIT_ULL(HW_ISSUE_8394) | \
+	BIT_ULL(HW_ISSUE_8401) | \
+	BIT_ULL(HW_ISSUE_8443) | \
+	BIT_ULL(HW_ISSUE_8987) | \
+	BIT_ULL(HW_ISSUE_9630) | \
+	BIT_ULL(HW_ISSUE_10969) | \
+	BIT_ULL(GPUCORE_1619))
+
+#define hw_issues_t620 (\
+	BIT_ULL(HW_ISSUE_10649) | \
+	BIT_ULL(HW_ISSUE_10883) | \
+	BIT_ULL(HW_ISSUE_10959) | \
+	BIT_ULL(HW_ISSUE_11056) | \
+	BIT_ULL(HW_ISSUE_TMIX_8438))
+
+#define hw_issues_t620_r0p1 (\
+	BIT_ULL(HW_ISSUE_10327) | \
+	BIT_ULL(HW_ISSUE_10676) | \
+	BIT_ULL(HW_ISSUE_10817) | \
+	BIT_ULL(HW_ISSUE_11020) | \
+	BIT_ULL(HW_ISSUE_11024) | \
+	BIT_ULL(HW_ISSUE_11035))
+
+#define hw_issues_t620_r1p0 (\
+	BIT_ULL(HW_ISSUE_11020) | \
+	BIT_ULL(HW_ISSUE_11024))
+
+#define hw_issues_t720 (\
+	BIT_ULL(HW_ISSUE_10649) | \
+	BIT_ULL(HW_ISSUE_10797) | \
+	BIT_ULL(HW_ISSUE_10883) | \
+	BIT_ULL(HW_ISSUE_11056) | \
+	BIT_ULL(HW_ISSUE_TMIX_8438))
+
+#define hw_issues_t760 (\
+	BIT_ULL(HW_ISSUE_10883) | \
+	BIT_ULL(HW_ISSUE_T76X_3953) | \
+	BIT_ULL(HW_ISSUE_TMIX_8438))
+
+#define hw_issues_t760_r0p0 (\
+	BIT_ULL(HW_ISSUE_11020) | \
+	BIT_ULL(HW_ISSUE_11024) | \
+	BIT_ULL(HW_ISSUE_T76X_3542))
+
+#define hw_issues_t760_r0p1 (\
+	BIT_ULL(HW_ISSUE_11020) | \
+	BIT_ULL(HW_ISSUE_11024) | \
+	BIT_ULL(HW_ISSUE_T76X_3542))
+
+#define hw_issues_t760_r0p1_50rel0 (\
+	BIT_ULL(HW_ISSUE_T76X_3542))
+
+#define hw_issues_t760_r0p2 (\
+	BIT_ULL(HW_ISSUE_11020) | \
+	BIT_ULL(HW_ISSUE_11024) | \
+	BIT_ULL(HW_ISSUE_T76X_3542))
+
+#define hw_issues_t760_r0p3 (\
+	BIT_ULL(HW_ISSUE_T76X_3542))
+
+#define hw_issues_t820 (\
+	BIT_ULL(HW_ISSUE_10883) | \
+	BIT_ULL(HW_ISSUE_T76X_3953) | \
+	BIT_ULL(HW_ISSUE_TMIX_8438))
+
+#define hw_issues_t830 (\
+	BIT_ULL(HW_ISSUE_10883) | \
+	BIT_ULL(HW_ISSUE_T76X_3953) | \
+	BIT_ULL(HW_ISSUE_TMIX_8438))
+
+#define hw_issues_t860 (\
+	BIT_ULL(HW_ISSUE_10883) | \
+	BIT_ULL(HW_ISSUE_T76X_3953) | \
+	BIT_ULL(HW_ISSUE_TMIX_8438))
+
+#define hw_issues_t880 (\
+	BIT_ULL(HW_ISSUE_10883) | \
+	BIT_ULL(HW_ISSUE_T76X_3953) | \
+	BIT_ULL(HW_ISSUE_TMIX_8438))
+
+#define hw_issues_g31 0
+
+#define hw_issues_g31_r1p0 (\
+	BIT_ULL(HW_ISSUE_TGOX_R1_1234))
+
+#define hw_issues_g51 0
+
+#define hw_issues_g52 0
+
+#define hw_issues_g71 (\
+	BIT_ULL(HW_ISSUE_TMIX_8463) | \
+	BIT_ULL(HW_ISSUE_TMIX_8438))
+
+#define hw_issues_g71_r0p0_05dev0 (\
+	BIT_ULL(HW_ISSUE_T76X_3953))
+
+#define hw_issues_g72 0
+
+#define hw_issues_g76 0
+
+static inline bool panfrost_has_hw_issue(struct panfrost_device *pfdev,
+					 enum panfrost_hw_issue issue)
+{
+	return test_bit(issue, pfdev->features.hw_issues);
+}
+
+#endif /* __PANFROST_ISSUES_H__ */
diff --git a/drivers/gpu/drm/panfrost/panfrost_job.c b/drivers/gpu/drm/panfrost/panfrost_job.c
new file mode 100644
index 000000000000..a5716c8fe8b3
--- /dev/null
+++ b/drivers/gpu/drm/panfrost/panfrost_job.c
@@ -0,0 +1,564 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright 2019 Linaro, Ltd, Rob Herring <robh@kernel.org> */
+/* Copyright 2019 Collabora ltd. */
+#include <linux/delay.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
+#include <linux/reservation.h>
+#include <drm/gpu_scheduler.h>
+#include <drm/panfrost_drm.h>
+
+#include "panfrost_device.h"
+#include "panfrost_devfreq.h"
+#include "panfrost_job.h"
+#include "panfrost_features.h"
+#include "panfrost_issues.h"
+#include "panfrost_gem.h"
+#include "panfrost_regs.h"
+#include "panfrost_gpu.h"
+#include "panfrost_mmu.h"
+
+#define job_write(dev, reg, data) writel(data, dev->iomem + (reg))
+#define job_read(dev, reg) readl(dev->iomem + (reg))
+
+struct panfrost_queue_state {
+	struct drm_gpu_scheduler sched;
+
+	u64 fence_context;
+	u64 emit_seqno;
+};
+
+struct panfrost_job_slot {
+	struct panfrost_queue_state queue[NUM_JOB_SLOTS];
+	spinlock_t job_lock;
+};
+
+static struct panfrost_job *
+to_panfrost_job(struct drm_sched_job *sched_job)
+{
+	return container_of(sched_job, struct panfrost_job, base);
+}
+
+struct panfrost_fence {
+	struct dma_fence base;
+	struct drm_device *dev;
+	/* panfrost seqno for signaled() test */
+	u64 seqno;
+	int queue;
+};
+
+static inline struct panfrost_fence *
+to_panfrost_fence(struct dma_fence *fence)
+{
+	return (struct panfrost_fence *)fence;
+}
+
+static const char *panfrost_fence_get_driver_name(struct dma_fence *fence)
+{
+	return "panfrost";
+}
+
+static const char *panfrost_fence_get_timeline_name(struct dma_fence *fence)
+{
+	struct panfrost_fence *f = to_panfrost_fence(fence);
+
+	switch (f->queue) {
+	case 0:
+		return "panfrost-js-0";
+	case 1:
+		return "panfrost-js-1";
+	case 2:
+		return "panfrost-js-2";
+	default:
+		return NULL;
+	}
+}
+
+static const struct dma_fence_ops panfrost_fence_ops = {
+	.get_driver_name = panfrost_fence_get_driver_name,
+	.get_timeline_name = panfrost_fence_get_timeline_name,
+};
+
+static struct dma_fence *panfrost_fence_create(struct panfrost_device *pfdev, int js_num)
+{
+	struct panfrost_fence *fence;
+	struct panfrost_job_slot *js = pfdev->js;
+
+	fence = kzalloc(sizeof(*fence), GFP_KERNEL);
+	if (!fence)
+		return ERR_PTR(-ENOMEM);
+
+	fence->dev = pfdev->ddev;
+	fence->queue = js_num;
+	fence->seqno = ++js->queue[js_num].emit_seqno;
+	dma_fence_init(&fence->base, &panfrost_fence_ops, &js->job_lock,
+		       js->queue[js_num].fence_context, fence->seqno);
+
+	return &fence->base;
+}
+
+static int panfrost_job_get_slot(struct panfrost_job *job)
+{
+	/* JS0: fragment jobs.
+	 * JS1: vertex/tiler jobs
+	 * JS2: compute jobs
+	 */
+	if (job->requirements & PANFROST_JD_REQ_FS)
+		return 0;
+
+/* Not exposed to userspace yet */
+#if 0
+	if (job->requirements & PANFROST_JD_REQ_ONLY_COMPUTE) {
+		if ((job->requirements & PANFROST_JD_REQ_CORE_GRP_MASK) &&
+		    (job->pfdev->features.nr_core_groups == 2))
+			return 2;
+		if (panfrost_has_hw_issue(job->pfdev, HW_ISSUE_8987))
+			return 2;
+	}
+#endif
+	return 1;
+}
+
+static void panfrost_job_write_affinity(struct panfrost_device *pfdev,
+					u32 requirements,
+					int js)
+{
+	u64 affinity;
+
+	/*
+	 * Use all cores for now.
+	 * Eventually we may need to support tiler only jobs and h/w with
+	 * multiple (2) coherent core groups
+	 */
+	affinity = pfdev->features.shader_present;
+
+	job_write(pfdev, JS_AFFINITY_NEXT_LO(js), affinity & 0xFFFFFFFF);
+	job_write(pfdev, JS_AFFINITY_NEXT_HI(js), affinity >> 32);
+}
+
+static void panfrost_job_hw_submit(struct panfrost_job *job, int js)
+{
+	struct panfrost_device *pfdev = job->pfdev;
+	unsigned long flags;
+	u32 cfg;
+	u64 jc_head = job->jc;
+	int ret;
+
+	ret = pm_runtime_get_sync(pfdev->dev);
+	if (ret < 0)
+		return;
+
+	if (WARN_ON(job_read(pfdev, JS_COMMAND_NEXT(js))))
+		goto end;
+
+	panfrost_devfreq_record_transition(pfdev, js);
+	spin_lock_irqsave(&pfdev->hwaccess_lock, flags);
+
+	job_write(pfdev, JS_HEAD_NEXT_LO(js), jc_head & 0xFFFFFFFF);
+	job_write(pfdev, JS_HEAD_NEXT_HI(js), jc_head >> 32);
+
+	panfrost_job_write_affinity(pfdev, job->requirements, js);
+
+	/* start MMU, medium priority, cache clean/flush on end, clean/flush on
+	 * start */
+	/* TODO: different address spaces */
+	cfg = JS_CONFIG_THREAD_PRI(8) |
+		JS_CONFIG_START_FLUSH_CLEAN_INVALIDATE |
+		JS_CONFIG_END_FLUSH_CLEAN_INVALIDATE;
+
+	if (panfrost_has_hw_feature(pfdev, HW_FEATURE_FLUSH_REDUCTION))
+		cfg |= JS_CONFIG_ENABLE_FLUSH_REDUCTION;
+
+	if (panfrost_has_hw_issue(pfdev, HW_ISSUE_10649))
+		cfg |= JS_CONFIG_START_MMU;
+
+	job_write(pfdev, JS_CONFIG_NEXT(js), cfg);
+
+	if (panfrost_has_hw_feature(pfdev, HW_FEATURE_FLUSH_REDUCTION))
+		job_write(pfdev, JS_FLUSH_ID_NEXT(js), job->flush_id);
+
+	/* GO ! */
+	dev_dbg(pfdev->dev, "JS: Submitting atom %p to js[%d] with head=0x%llx",
+				job, js, jc_head);
+
+	job_write(pfdev, JS_COMMAND_NEXT(js), JS_COMMAND_START);
+
+	spin_unlock_irqrestore(&pfdev->hwaccess_lock, flags);
+
+end:
+	pm_runtime_mark_last_busy(pfdev->dev);
+	pm_runtime_put_autosuspend(pfdev->dev);
+}
+
+static void panfrost_acquire_object_fences(struct drm_gem_object **bos,
+					   int bo_count,
+					   struct dma_fence **implicit_fences)
+{
+	int i;
+
+	for (i = 0; i < bo_count; i++)
+		implicit_fences[i] = reservation_object_get_excl_rcu(bos[i]->resv);
+}
+
+static void panfrost_attach_object_fences(struct drm_gem_object **bos,
+					  int bo_count,
+					  struct dma_fence *fence)
+{
+	int i;
+
+	for (i = 0; i < bo_count; i++)
+		reservation_object_add_excl_fence(bos[i]->resv, fence);
+}
+
+int panfrost_job_push(struct panfrost_job *job)
+{
+	struct panfrost_device *pfdev = job->pfdev;
+	int slot = panfrost_job_get_slot(job);
+	struct drm_sched_entity *entity = &job->file_priv->sched_entity[slot];
+	struct ww_acquire_ctx acquire_ctx;
+	int ret = 0;
+
+	mutex_lock(&pfdev->sched_lock);
+
+	ret = drm_gem_lock_reservations(job->bos, job->bo_count,
+					    &acquire_ctx);
+	if (ret) {
+		mutex_unlock(&pfdev->sched_lock);
+		return ret;
+	}
+
+	ret = drm_sched_job_init(&job->base, entity, NULL);
+	if (ret) {
+		mutex_unlock(&pfdev->sched_lock);
+		goto unlock;
+	}
+
+	job->render_done_fence = dma_fence_get(&job->base.s_fence->finished);
+
+	kref_get(&job->refcount); /* put by scheduler job completion */
+
+	panfrost_acquire_object_fences(job->bos, job->bo_count,
+				       job->implicit_fences);
+
+	drm_sched_entity_push_job(&job->base, entity);
+
+	mutex_unlock(&pfdev->sched_lock);
+
+	panfrost_attach_object_fences(job->bos, job->bo_count,
+				      job->render_done_fence);
+
+unlock:
+	drm_gem_unlock_reservations(job->bos, job->bo_count, &acquire_ctx);
+
+	return ret;
+}
+
+static void panfrost_job_cleanup(struct kref *ref)
+{
+	struct panfrost_job *job = container_of(ref, struct panfrost_job,
+						refcount);
+	unsigned int i;
+
+	if (job->in_fences) {
+		for (i = 0; i < job->in_fence_count; i++)
+			dma_fence_put(job->in_fences[i]);
+		kvfree(job->in_fences);
+	}
+	if (job->implicit_fences) {
+		for (i = 0; i < job->bo_count; i++)
+			dma_fence_put(job->implicit_fences[i]);
+		kvfree(job->implicit_fences);
+	}
+	dma_fence_put(job->done_fence);
+	dma_fence_put(job->render_done_fence);
+
+	if (job->bos) {
+		for (i = 0; i < job->bo_count; i++)
+			drm_gem_object_put_unlocked(job->bos[i]);
+		kvfree(job->bos);
+	}
+
+	kfree(job);
+}
+
+void panfrost_job_put(struct panfrost_job *job)
+{
+	kref_put(&job->refcount, panfrost_job_cleanup);
+}
+
+static void panfrost_job_free(struct drm_sched_job *sched_job)
+{
+	struct panfrost_job *job = to_panfrost_job(sched_job);
+
+	drm_sched_job_cleanup(sched_job);
+
+	panfrost_job_put(job);
+}
+
+static struct dma_fence *panfrost_job_dependency(struct drm_sched_job *sched_job,
+						 struct drm_sched_entity *s_entity)
+{
+	struct panfrost_job *job = to_panfrost_job(sched_job);
+	struct dma_fence *fence;
+	unsigned int i;
+
+	/* Explicit fences */
+	for (i = 0; i < job->in_fence_count; i++) {
+		if (job->in_fences[i]) {
+			fence = job->in_fences[i];
+			job->in_fences[i] = NULL;
+			return fence;
+		}
+	}
+
+	/* Implicit fences, max. one per BO */
+	for (i = 0; i < job->bo_count; i++) {
+		if (job->implicit_fences[i]) {
+			fence = job->implicit_fences[i];
+			job->implicit_fences[i] = NULL;
+			return fence;
+		}
+	}
+
+	return NULL;
+}
+
+static struct dma_fence *panfrost_job_run(struct drm_sched_job *sched_job)
+{
+	struct panfrost_job *job = to_panfrost_job(sched_job);
+	struct panfrost_device *pfdev = job->pfdev;
+	int slot = panfrost_job_get_slot(job);
+	struct dma_fence *fence = NULL;
+
+	if (unlikely(job->base.s_fence->finished.error))
+		return NULL;
+
+	pfdev->jobs[slot] = job;
+
+	fence = panfrost_fence_create(pfdev, slot);
+	if (IS_ERR(fence))
+		return NULL;
+
+	if (job->done_fence)
+		dma_fence_put(job->done_fence);
+	job->done_fence = dma_fence_get(fence);
+
+	panfrost_job_hw_submit(job, slot);
+
+	return fence;
+}
+
+void panfrost_job_enable_interrupts(struct panfrost_device *pfdev)
+{
+	int j;
+	u32 irq_mask = 0;
+
+	for (j = 0; j < NUM_JOB_SLOTS; j++) {
+		irq_mask |= MK_JS_MASK(j);
+	}
+
+	job_write(pfdev, JOB_INT_CLEAR, irq_mask);
+	job_write(pfdev, JOB_INT_MASK, irq_mask);
+}
+
+static void panfrost_job_timedout(struct drm_sched_job *sched_job)
+{
+	struct panfrost_job *job = to_panfrost_job(sched_job);
+	struct panfrost_device *pfdev = job->pfdev;
+	int js = panfrost_job_get_slot(job);
+	int i;
+
+	/*
+	 * If the GPU managed to complete this jobs fence, the timeout is
+	 * spurious. Bail out.
+	 */
+	if (dma_fence_is_signaled(job->done_fence))
+		return;
+
+	dev_err(pfdev->dev, "gpu sched timeout, js=%d, status=0x%x, head=0x%x, tail=0x%x, sched_job=%p",
+		js,
+		job_read(pfdev, JS_STATUS(js)),
+		job_read(pfdev, JS_HEAD_LO(js)),
+		job_read(pfdev, JS_TAIL_LO(js)),
+		sched_job);
+
+	mutex_lock(&pfdev->reset_lock);
+
+	for (i = 0; i < NUM_JOB_SLOTS; i++)
+		drm_sched_stop(&pfdev->js->queue[i].sched);
+
+	if (sched_job)
+		drm_sched_increase_karma(sched_job);
+
+	/* panfrost_core_dump(pfdev); */
+
+	panfrost_devfreq_record_transition(pfdev, js);
+	panfrost_gpu_soft_reset(pfdev);
+
+	/* TODO: Re-enable all other address spaces */
+	panfrost_mmu_enable(pfdev, 0);
+	panfrost_gpu_power_on(pfdev);
+	panfrost_job_enable_interrupts(pfdev);
+
+	for (i = 0; i < NUM_JOB_SLOTS; i++)
+		drm_sched_resubmit_jobs(&pfdev->js->queue[i].sched);
+
+	/* restart scheduler after GPU is usable again */
+	for (i = 0; i < NUM_JOB_SLOTS; i++)
+		drm_sched_start(&pfdev->js->queue[i].sched, true);
+
+	mutex_unlock(&pfdev->reset_lock);
+}
+
+static const struct drm_sched_backend_ops panfrost_sched_ops = {
+	.dependency = panfrost_job_dependency,
+	.run_job = panfrost_job_run,
+	.timedout_job = panfrost_job_timedout,
+	.free_job = panfrost_job_free
+};
+
+static irqreturn_t panfrost_job_irq_handler(int irq, void *data)
+{
+	struct panfrost_device *pfdev = data;
+	u32 status = job_read(pfdev, JOB_INT_STAT);
+	int j;
+
+	dev_dbg(pfdev->dev, "jobslot irq status=%x\n", status);
+
+	if (!status)
+		return IRQ_NONE;
+
+	pm_runtime_mark_last_busy(pfdev->dev);
+
+	for (j = 0; status; j++) {
+		u32 mask = MK_JS_MASK(j);
+
+		if (!(status & mask))
+			continue;
+
+		job_write(pfdev, JOB_INT_CLEAR, mask);
+
+		if (status & JOB_INT_MASK_ERR(j)) {
+			job_write(pfdev, JS_COMMAND_NEXT(j), JS_COMMAND_NOP);
+
+			dev_err(pfdev->dev, "js fault, js=%d, status=%s, head=0x%x, tail=0x%x",
+				j,
+				panfrost_exception_name(pfdev, job_read(pfdev, JS_STATUS(j))),
+				job_read(pfdev, JS_HEAD_LO(j)),
+				job_read(pfdev, JS_TAIL_LO(j)));
+
+			drm_sched_fault(&pfdev->js->queue[j].sched);
+		}
+
+		if (status & JOB_INT_MASK_DONE(j)) {
+			panfrost_devfreq_record_transition(pfdev, j);
+			dma_fence_signal(pfdev->jobs[j]->done_fence);
+		}
+
+		status &= ~mask;
+	}
+
+	return IRQ_HANDLED;
+}
+
+int panfrost_job_init(struct panfrost_device *pfdev)
+{
+	struct panfrost_job_slot *js;
+	int ret, j, irq;
+
+	pfdev->js = js = devm_kzalloc(pfdev->dev, sizeof(*js), GFP_KERNEL);
+	if (!js)
+		return -ENOMEM;
+
+	spin_lock_init(&js->job_lock);
+
+	irq = platform_get_irq_byname(to_platform_device(pfdev->dev), "job");
+	if (irq <= 0)
+		return -ENODEV;
+
+	ret = devm_request_irq(pfdev->dev, irq, panfrost_job_irq_handler,
+			       IRQF_SHARED, "job", pfdev);
+	if (ret) {
+		dev_err(pfdev->dev, "failed to request job irq");
+		return ret;
+	}
+
+	for (j = 0; j < NUM_JOB_SLOTS; j++) {
+		js->queue[j].fence_context = dma_fence_context_alloc(1);
+
+		ret = drm_sched_init(&js->queue[j].sched,
+				     &panfrost_sched_ops,
+				     1, 0, msecs_to_jiffies(500),
+				     "pan_js");
+		if (ret) {
+			dev_err(pfdev->dev, "Failed to create scheduler: %d.", ret);
+			goto err_sched;
+		}
+	}
+
+	panfrost_job_enable_interrupts(pfdev);
+
+	return 0;
+
+err_sched:
+	for (j--; j >= 0; j--)
+		drm_sched_fini(&js->queue[j].sched);
+
+	return ret;
+}
+
+void panfrost_job_fini(struct panfrost_device *pfdev)
+{
+	struct panfrost_job_slot *js = pfdev->js;
+	int j;
+
+	job_write(pfdev, JOB_INT_MASK, 0);
+
+	for (j = 0; j < NUM_JOB_SLOTS; j++)
+		drm_sched_fini(&js->queue[j].sched);
+
+}
+
+int panfrost_job_open(struct panfrost_file_priv *panfrost_priv)
+{
+	struct panfrost_device *pfdev = panfrost_priv->pfdev;
+	struct panfrost_job_slot *js = pfdev->js;
+	struct drm_sched_rq *rq;
+	int ret, i;
+
+	for (i = 0; i < NUM_JOB_SLOTS; i++) {
+		rq = &js->queue[i].sched.sched_rq[DRM_SCHED_PRIORITY_NORMAL];
+		ret = drm_sched_entity_init(&panfrost_priv->sched_entity[i], &rq, 1, NULL);
+		if (WARN_ON(ret))
+			return ret;
+	}
+	return 0;
+}
+
+void panfrost_job_close(struct panfrost_file_priv *panfrost_priv)
+{
+	int i;
+
+	for (i = 0; i < NUM_JOB_SLOTS; i++)
+		drm_sched_entity_destroy(&panfrost_priv->sched_entity[i]);
+}
+
+int panfrost_job_is_idle(struct panfrost_device *pfdev)
+{
+	struct panfrost_job_slot *js = pfdev->js;
+	int i;
+
+	for (i = 0; i < NUM_JOB_SLOTS; i++) {
+		/* If there are any jobs in the HW queue, we're not idle */
+		if (atomic_read(&js->queue[i].sched.hw_rq_count))
+			return false;
+
+		/* Check whether the hardware is idle */
+		if (pfdev->devfreq.slot[i].busy)
+			return false;
+	}
+
+	return true;
+}
diff --git a/drivers/gpu/drm/panfrost/panfrost_job.h b/drivers/gpu/drm/panfrost/panfrost_job.h
new file mode 100644
index 000000000000..62454128a792
--- /dev/null
+++ b/drivers/gpu/drm/panfrost/panfrost_job.h
@@ -0,0 +1,51 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright 2019 Collabora ltd. */
+
+#ifndef __PANFROST_JOB_H__
+#define __PANFROST_JOB_H__
+
+#include <uapi/drm/panfrost_drm.h>
+#include <drm/gpu_scheduler.h>
+
+struct panfrost_device;
+struct panfrost_gem_object;
+struct panfrost_file_priv;
+
+struct panfrost_job {
+	struct drm_sched_job base;
+
+	struct kref refcount;
+
+	struct panfrost_device *pfdev;
+	struct panfrost_file_priv *file_priv;
+
+	/* Optional fences userspace can pass in for the job to depend on. */
+	struct dma_fence **in_fences;
+	u32 in_fence_count;
+
+	/* Fence to be signaled by IRQ handler when the job is complete. */
+	struct dma_fence *done_fence;
+
+	__u64 jc;
+	__u32 requirements;
+	__u32 flush_id;
+
+	/* Exclusive fences we have taken from the BOs to wait for */
+	struct dma_fence **implicit_fences;
+	struct drm_gem_object **bos;
+	u32 bo_count;
+
+	/* Fence to be signaled by drm-sched once its done with the job */
+	struct dma_fence *render_done_fence;
+};
+
+int panfrost_job_init(struct panfrost_device *pfdev);
+void panfrost_job_fini(struct panfrost_device *pfdev);
+int panfrost_job_open(struct panfrost_file_priv *panfrost_priv);
+void panfrost_job_close(struct panfrost_file_priv *panfrost_priv);
+int panfrost_job_push(struct panfrost_job *job);
+void panfrost_job_put(struct panfrost_job *job);
+void panfrost_job_enable_interrupts(struct panfrost_device *pfdev);
+int panfrost_job_is_idle(struct panfrost_device *pfdev);
+
+#endif
diff --git a/drivers/gpu/drm/panfrost/panfrost_mmu.c b/drivers/gpu/drm/panfrost/panfrost_mmu.c
new file mode 100644
index 000000000000..762b1bd2a8c2
--- /dev/null
+++ b/drivers/gpu/drm/panfrost/panfrost_mmu.c
@@ -0,0 +1,386 @@
+// SPDX-License-Identifier:	GPL-2.0
+/* Copyright 2019 Linaro, Ltd, Rob Herring <robh@kernel.org> */
+#include <linux/bitfield.h>
+#include <linux/delay.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/iopoll.h>
+#include <linux/io-pgtable.h>
+#include <linux/iommu.h>
+#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
+#include <linux/sizes.h>
+
+#include "panfrost_device.h"
+#include "panfrost_mmu.h"
+#include "panfrost_gem.h"
+#include "panfrost_features.h"
+#include "panfrost_regs.h"
+
+#define mmu_write(dev, reg, data) writel(data, dev->iomem + reg)
+#define mmu_read(dev, reg) readl(dev->iomem + reg)
+
+struct panfrost_mmu {
+	struct io_pgtable_cfg pgtbl_cfg;
+	struct io_pgtable_ops *pgtbl_ops;
+	struct mutex lock;
+};
+
+static int wait_ready(struct panfrost_device *pfdev, u32 as_nr)
+{
+	int ret;
+	u32 val;
+
+	/* Wait for the MMU status to indicate there is no active command, in
+	 * case one is pending. */
+	ret = readl_relaxed_poll_timeout_atomic(pfdev->iomem + AS_STATUS(as_nr),
+		val, !(val & AS_STATUS_AS_ACTIVE), 10, 1000);
+
+	if (ret)
+		dev_err(pfdev->dev, "AS_ACTIVE bit stuck\n");
+
+	return ret;
+}
+
+static int write_cmd(struct panfrost_device *pfdev, u32 as_nr, u32 cmd)
+{
+	int status;
+
+	/* write AS_COMMAND when MMU is ready to accept another command */
+	status = wait_ready(pfdev, as_nr);
+	if (!status)
+		mmu_write(pfdev, AS_COMMAND(as_nr), cmd);
+
+	return status;
+}
+
+static void lock_region(struct panfrost_device *pfdev, u32 as_nr,
+			u64 iova, size_t size)
+{
+	u8 region_width;
+	u64 region = iova & PAGE_MASK;
+	/*
+	 * fls returns:
+	 * 1 .. 32
+	 *
+	 * 10 + fls(num_pages)
+	 * results in the range (11 .. 42)
+	 */
+
+	size = round_up(size, PAGE_SIZE);
+
+	region_width = 10 + fls(size >> PAGE_SHIFT);
+	if ((size >> PAGE_SHIFT) != (1ul << (region_width - 11))) {
+		/* not pow2, so must go up to the next pow2 */
+		region_width += 1;
+	}
+	region |= region_width;
+
+	/* Lock the region that needs to be updated */
+	mmu_write(pfdev, AS_LOCKADDR_LO(as_nr), region & 0xFFFFFFFFUL);
+	mmu_write(pfdev, AS_LOCKADDR_HI(as_nr), (region >> 32) & 0xFFFFFFFFUL);
+	write_cmd(pfdev, as_nr, AS_COMMAND_LOCK);
+}
+
+
+static int mmu_hw_do_operation(struct panfrost_device *pfdev, u32 as_nr,
+		u64 iova, size_t size, u32 op)
+{
+	unsigned long flags;
+	int ret;
+
+	spin_lock_irqsave(&pfdev->hwaccess_lock, flags);
+
+	if (op != AS_COMMAND_UNLOCK)
+		lock_region(pfdev, as_nr, iova, size);
+
+	/* Run the MMU operation */
+	write_cmd(pfdev, as_nr, op);
+
+	/* Wait for the flush to complete */
+	ret = wait_ready(pfdev, as_nr);
+
+	spin_unlock_irqrestore(&pfdev->hwaccess_lock, flags);
+
+	return ret;
+}
+
+void panfrost_mmu_enable(struct panfrost_device *pfdev, u32 as_nr)
+{
+	struct io_pgtable_cfg *cfg = &pfdev->mmu->pgtbl_cfg;
+	u64 transtab = cfg->arm_mali_lpae_cfg.transtab;
+	u64 memattr = cfg->arm_mali_lpae_cfg.memattr;
+
+	mmu_write(pfdev, MMU_INT_CLEAR, ~0);
+	mmu_write(pfdev, MMU_INT_MASK, ~0);
+
+	mmu_write(pfdev, AS_TRANSTAB_LO(as_nr), transtab & 0xffffffffUL);
+	mmu_write(pfdev, AS_TRANSTAB_HI(as_nr), transtab >> 32);
+
+	/* Need to revisit mem attrs.
+	 * NC is the default, Mali driver is inner WT.
+	 */
+	mmu_write(pfdev, AS_MEMATTR_LO(as_nr), memattr & 0xffffffffUL);
+	mmu_write(pfdev, AS_MEMATTR_HI(as_nr), memattr >> 32);
+
+	write_cmd(pfdev, as_nr, AS_COMMAND_UPDATE);
+}
+
+static void mmu_disable(struct panfrost_device *pfdev, u32 as_nr)
+{
+	mmu_write(pfdev, AS_TRANSTAB_LO(as_nr), 0);
+	mmu_write(pfdev, AS_TRANSTAB_HI(as_nr), 0);
+
+	mmu_write(pfdev, AS_MEMATTR_LO(as_nr), 0);
+	mmu_write(pfdev, AS_MEMATTR_HI(as_nr), 0);
+
+	write_cmd(pfdev, as_nr, AS_COMMAND_UPDATE);
+}
+
+static size_t get_pgsize(u64 addr, size_t size)
+{
+	if (addr & (SZ_2M - 1) || size < SZ_2M)
+		return SZ_4K;
+
+	return SZ_2M;
+}
+
+int panfrost_mmu_map(struct panfrost_gem_object *bo)
+{
+	struct drm_gem_object *obj = &bo->base.base;
+	struct panfrost_device *pfdev = to_panfrost_device(obj->dev);
+	struct io_pgtable_ops *ops = pfdev->mmu->pgtbl_ops;
+	u64 iova = bo->node.start << PAGE_SHIFT;
+	unsigned int count;
+	struct scatterlist *sgl;
+	struct sg_table *sgt;
+	int ret;
+
+	sgt = drm_gem_shmem_get_pages_sgt(obj);
+	if (WARN_ON(IS_ERR(sgt)))
+		return PTR_ERR(sgt);
+
+	ret = pm_runtime_get_sync(pfdev->dev);
+	if (ret < 0)
+		return ret;
+
+	mutex_lock(&pfdev->mmu->lock);
+
+	for_each_sg(sgt->sgl, sgl, sgt->nents, count) {
+		unsigned long paddr = sg_dma_address(sgl);
+		size_t len = sg_dma_len(sgl);
+
+		dev_dbg(pfdev->dev, "map: iova=%llx, paddr=%lx, len=%zx", iova, paddr, len);
+
+		while (len) {
+			size_t pgsize = get_pgsize(iova | paddr, len);
+
+			ops->map(ops, iova, paddr, pgsize, IOMMU_WRITE | IOMMU_READ);
+			iova += pgsize;
+			paddr += pgsize;
+			len -= pgsize;
+		}
+	}
+
+	mmu_hw_do_operation(pfdev, 0, bo->node.start << PAGE_SHIFT,
+			    bo->node.size << PAGE_SHIFT, AS_COMMAND_FLUSH_PT);
+
+	mutex_unlock(&pfdev->mmu->lock);
+
+	pm_runtime_mark_last_busy(pfdev->dev);
+	pm_runtime_put_autosuspend(pfdev->dev);
+
+	return 0;
+}
+
+void panfrost_mmu_unmap(struct panfrost_gem_object *bo)
+{
+	struct drm_gem_object *obj = &bo->base.base;
+	struct panfrost_device *pfdev = to_panfrost_device(obj->dev);
+	struct io_pgtable_ops *ops = pfdev->mmu->pgtbl_ops;
+	u64 iova = bo->node.start << PAGE_SHIFT;
+	size_t len = bo->node.size << PAGE_SHIFT;
+	size_t unmapped_len = 0;
+	int ret;
+
+	dev_dbg(pfdev->dev, "unmap: iova=%llx, len=%zx", iova, len);
+
+	ret = pm_runtime_get_sync(pfdev->dev);
+	if (ret < 0)
+		return;
+
+	mutex_lock(&pfdev->mmu->lock);
+
+	while (unmapped_len < len) {
+		size_t unmapped_page;
+		size_t pgsize = get_pgsize(iova, len - unmapped_len);
+
+		unmapped_page = ops->unmap(ops, iova, pgsize);
+		if (!unmapped_page)
+			break;
+
+		iova += unmapped_page;
+		unmapped_len += unmapped_page;
+	}
+
+	mmu_hw_do_operation(pfdev, 0, bo->node.start << PAGE_SHIFT,
+			    bo->node.size << PAGE_SHIFT, AS_COMMAND_FLUSH_PT);
+
+	mutex_unlock(&pfdev->mmu->lock);
+
+	pm_runtime_mark_last_busy(pfdev->dev);
+	pm_runtime_put_autosuspend(pfdev->dev);
+}
+
+static void mmu_tlb_inv_context_s1(void *cookie)
+{
+	struct panfrost_device *pfdev = cookie;
+
+	mmu_hw_do_operation(pfdev, 0, 0, ~0UL, AS_COMMAND_FLUSH_MEM);
+}
+
+static void mmu_tlb_inv_range_nosync(unsigned long iova, size_t size,
+				     size_t granule, bool leaf, void *cookie)
+{}
+
+static void mmu_tlb_sync_context(void *cookie)
+{
+	//struct panfrost_device *pfdev = cookie;
+	// TODO: Wait 1000 GPU cycles for HW_ISSUE_6367/T60X
+}
+
+static const struct iommu_gather_ops mmu_tlb_ops = {
+	.tlb_flush_all	= mmu_tlb_inv_context_s1,
+	.tlb_add_flush	= mmu_tlb_inv_range_nosync,
+	.tlb_sync	= mmu_tlb_sync_context,
+};
+
+static const char *access_type_name(struct panfrost_device *pfdev,
+		u32 fault_status)
+{
+	switch (fault_status & AS_FAULTSTATUS_ACCESS_TYPE_MASK) {
+	case AS_FAULTSTATUS_ACCESS_TYPE_ATOMIC:
+		if (panfrost_has_hw_feature(pfdev, HW_FEATURE_AARCH64_MMU))
+			return "ATOMIC";
+		else
+			return "UNKNOWN";
+	case AS_FAULTSTATUS_ACCESS_TYPE_READ:
+		return "READ";
+	case AS_FAULTSTATUS_ACCESS_TYPE_WRITE:
+		return "WRITE";
+	case AS_FAULTSTATUS_ACCESS_TYPE_EX:
+		return "EXECUTE";
+	default:
+		WARN_ON(1);
+		return NULL;
+	}
+}
+
+static irqreturn_t panfrost_mmu_irq_handler(int irq, void *data)
+{
+	struct panfrost_device *pfdev = data;
+	u32 status = mmu_read(pfdev, MMU_INT_STAT);
+	int i;
+
+	if (!status)
+		return IRQ_NONE;
+
+	dev_err(pfdev->dev, "mmu irq status=%x\n", status);
+
+	for (i = 0; status; i++) {
+		u32 mask = BIT(i) | BIT(i + 16);
+		u64 addr;
+		u32 fault_status;
+		u32 exception_type;
+		u32 access_type;
+		u32 source_id;
+
+		if (!(status & mask))
+			continue;
+
+		fault_status = mmu_read(pfdev, AS_FAULTSTATUS(i));
+		addr = mmu_read(pfdev, AS_FAULTADDRESS_LO(i));
+		addr |= (u64)mmu_read(pfdev, AS_FAULTADDRESS_HI(i)) << 32;
+
+		/* decode the fault status */
+		exception_type = fault_status & 0xFF;
+		access_type = (fault_status >> 8) & 0x3;
+		source_id = (fault_status >> 16);
+
+		/* terminal fault, print info about the fault */
+		dev_err(pfdev->dev,
+			"Unhandled Page fault in AS%d at VA 0x%016llX\n"
+			"Reason: %s\n"
+			"raw fault status: 0x%X\n"
+			"decoded fault status: %s\n"
+			"exception type 0x%X: %s\n"
+			"access type 0x%X: %s\n"
+			"source id 0x%X\n",
+			i, addr,
+			"TODO",
+			fault_status,
+			(fault_status & (1 << 10) ? "DECODER FAULT" : "SLAVE FAULT"),
+			exception_type, panfrost_exception_name(pfdev, exception_type),
+			access_type, access_type_name(pfdev, fault_status),
+			source_id);
+
+		mmu_write(pfdev, MMU_INT_CLEAR, mask);
+
+		status &= ~mask;
+	}
+
+	return IRQ_HANDLED;
+};
+
+int panfrost_mmu_init(struct panfrost_device *pfdev)
+{
+	struct io_pgtable_ops *pgtbl_ops;
+	int err, irq;
+
+	pfdev->mmu = devm_kzalloc(pfdev->dev, sizeof(*pfdev->mmu), GFP_KERNEL);
+	if (!pfdev->mmu)
+		return -ENOMEM;
+
+	mutex_init(&pfdev->mmu->lock);
+
+	irq = platform_get_irq_byname(to_platform_device(pfdev->dev), "mmu");
+	if (irq <= 0)
+		return -ENODEV;
+
+	err = devm_request_irq(pfdev->dev, irq, panfrost_mmu_irq_handler,
+			       IRQF_SHARED, "mmu", pfdev);
+
+	if (err) {
+		dev_err(pfdev->dev, "failed to request mmu irq");
+		return err;
+	}
+	mmu_write(pfdev, MMU_INT_CLEAR, ~0);
+	mmu_write(pfdev, MMU_INT_MASK, ~0);
+
+	pfdev->mmu->pgtbl_cfg = (struct io_pgtable_cfg) {
+		.pgsize_bitmap	= SZ_4K | SZ_2M,
+		.ias		= FIELD_GET(0xff, pfdev->features.mmu_features),
+		.oas		= FIELD_GET(0xff00, pfdev->features.mmu_features),
+		.tlb		= &mmu_tlb_ops,
+		.iommu_dev	= pfdev->dev,
+	};
+
+	pgtbl_ops = alloc_io_pgtable_ops(ARM_MALI_LPAE, &pfdev->mmu->pgtbl_cfg,
+					 pfdev);
+	if (!pgtbl_ops)
+		return -ENOMEM;
+
+	pfdev->mmu->pgtbl_ops = pgtbl_ops;
+
+	panfrost_mmu_enable(pfdev, 0);
+
+	return 0;
+}
+
+void panfrost_mmu_fini(struct panfrost_device *pfdev)
+{
+	mmu_write(pfdev, MMU_INT_MASK, 0);
+	mmu_disable(pfdev, 0);
+
+	free_io_pgtable_ops(pfdev->mmu->pgtbl_ops);
+}
diff --git a/drivers/gpu/drm/panfrost/panfrost_mmu.h b/drivers/gpu/drm/panfrost/panfrost_mmu.h
new file mode 100644
index 000000000000..f5878d86a5ce
--- /dev/null
+++ b/drivers/gpu/drm/panfrost/panfrost_mmu.h
@@ -0,0 +1,17 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright 2019 Linaro, Ltd, Rob Herring <robh@kernel.org> */
+
+#ifndef __PANFROST_MMU_H__
+#define __PANFROST_MMU_H__
+
+struct panfrost_gem_object;
+
+int panfrost_mmu_map(struct panfrost_gem_object *bo);
+void panfrost_mmu_unmap(struct panfrost_gem_object *bo);
+
+int panfrost_mmu_init(struct panfrost_device *pfdev);
+void panfrost_mmu_fini(struct panfrost_device *pfdev);
+
+void panfrost_mmu_enable(struct panfrost_device *pfdev, u32 as_nr);
+
+#endif
diff --git a/drivers/gpu/drm/panfrost/panfrost_regs.h b/drivers/gpu/drm/panfrost/panfrost_regs.h
new file mode 100644
index 000000000000..578c5fc2188b
--- /dev/null
+++ b/drivers/gpu/drm/panfrost/panfrost_regs.h
@@ -0,0 +1,298 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright 2018 Marty E. Plummer <hanetzer@startmail.com> */
+/* Copyright 2019 Linaro, Ltd, Rob Herring <robh@kernel.org> */
+/*
+ * Register definitions based on mali_midg_regmap.h
+ * (C) COPYRIGHT 2010-2018 ARM Limited. All rights reserved.
+ */
+#ifndef __PANFROST_REGS_H__
+#define __PANFROST_REGS_H__
+
+#define GPU_ID				0x00
+#define GPU_L2_FEATURES			0x004	/* (RO) Level 2 cache features */
+#define GPU_CORE_FEATURES		0x008	/* (RO) Shader Core Features */
+#define GPU_TILER_FEATURES		0x00C	/* (RO) Tiler Features */
+#define GPU_MEM_FEATURES		0x010	/* (RO) Memory system features */
+#define   GROUPS_L2_COHERENT		BIT(0)	/* Cores groups are l2 coherent */
+
+#define GPU_MMU_FEATURES		0x014	/* (RO) MMU features */
+#define GPU_AS_PRESENT			0x018	/* (RO) Address space slots present */
+#define GPU_JS_PRESENT			0x01C	/* (RO) Job slots present */
+
+#define GPU_INT_RAWSTAT			0x20
+#define GPU_INT_CLEAR			0x24
+#define GPU_INT_MASK			0x28
+#define GPU_INT_STAT			0x2c
+#define   GPU_IRQ_FAULT			BIT(0)
+#define   GPU_IRQ_MULTIPLE_FAULT	BIT(7)
+#define   GPU_IRQ_RESET_COMPLETED	BIT(8)
+#define   GPU_IRQ_POWER_CHANGED		BIT(9)
+#define   GPU_IRQ_POWER_CHANGED_ALL	BIT(10)
+#define   GPU_IRQ_PERFCNT_SAMPLE_COMPLETED BIT(16)
+#define   GPU_IRQ_CLEAN_CACHES_COMPLETED BIT(17)
+#define   GPU_IRQ_MASK_ALL			 \
+	  (GPU_IRQ_FAULT			|\
+	   GPU_IRQ_MULTIPLE_FAULT		|\
+	   GPU_IRQ_RESET_COMPLETED		|\
+	   GPU_IRQ_POWER_CHANGED		|\
+	   GPU_IRQ_POWER_CHANGED_ALL		|\
+	   GPU_IRQ_PERFCNT_SAMPLE_COMPLETED	|\
+	   GPU_IRQ_CLEAN_CACHES_COMPLETED)
+#define GPU_IRQ_MASK_ERROR	   		\
+	(					\
+	 GPU_IRQ_FAULT				|\
+	 GPU_IRQ_MULTIPLE_FAULT)
+#define GPU_CMD				0x30
+#define   GPU_CMD_SOFT_RESET		0x01
+#define GPU_STATUS			0x34
+#define GPU_LATEST_FLUSH_ID		0x38
+#define GPU_FAULT_STATUS		0x3C
+#define GPU_FAULT_ADDRESS_LO		0x40
+#define GPU_FAULT_ADDRESS_HI		0x44
+
+#define GPU_THREAD_MAX_THREADS		0x0A0	/* (RO) Maximum number of threads per core */
+#define GPU_THREAD_MAX_WORKGROUP_SIZE	0x0A4	/* (RO) Maximum workgroup size */
+#define GPU_THREAD_MAX_BARRIER_SIZE	0x0A8	/* (RO) Maximum threads waiting at a barrier */
+#define GPU_THREAD_FEATURES		0x0AC	/* (RO) Thread features */
+#define GPU_THREAD_TLS_ALLOC		0x310   /* (RO) Number of threads per core that
+						 * TLS must be allocated for */
+
+#define GPU_TEXTURE_FEATURES(n)		(0x0B0 + ((n) * 4))
+#define GPU_JS_FEATURES(n)		(0x0C0 + ((n) * 4))
+
+#define GPU_SHADER_PRESENT_LO		0x100	/* (RO) Shader core present bitmap, low word */
+#define GPU_SHADER_PRESENT_HI		0x104	/* (RO) Shader core present bitmap, high word */
+#define GPU_TILER_PRESENT_LO		0x110	/* (RO) Tiler core present bitmap, low word */
+#define GPU_TILER_PRESENT_HI		0x114	/* (RO) Tiler core present bitmap, high word */
+
+#define GPU_L2_PRESENT_LO		0x120	/* (RO) Level 2 cache present bitmap, low word */
+#define GPU_L2_PRESENT_HI		0x124	/* (RO) Level 2 cache present bitmap, high word */
+
+#define GPU_COHERENCY_FEATURES		0x300	/* (RO) Coherency features present */
+#define   COHERENCY_ACE_LITE		BIT(0)
+#define   COHERENCY_ACE			BIT(1)
+
+#define GPU_STACK_PRESENT_LO		0xE00   /* (RO) Core stack present bitmap, low word */
+#define GPU_STACK_PRESENT_HI		0xE04   /* (RO) Core stack present bitmap, high word */
+
+#define SHADER_READY_LO			0x140	/* (RO) Shader core ready bitmap, low word */
+#define SHADER_READY_HI			0x144	/* (RO) Shader core ready bitmap, high word */
+
+#define TILER_READY_LO			0x150	/* (RO) Tiler core ready bitmap, low word */
+#define TILER_READY_HI			0x154	/* (RO) Tiler core ready bitmap, high word */
+
+#define L2_READY_LO			0x160	/* (RO) Level 2 cache ready bitmap, low word */
+#define L2_READY_HI			0x164	/* (RO) Level 2 cache ready bitmap, high word */
+
+#define STACK_READY_LO			0xE10   /* (RO) Core stack ready bitmap, low word */
+#define STACK_READY_HI			0xE14   /* (RO) Core stack ready bitmap, high word */
+
+
+#define SHADER_PWRON_LO			0x180	/* (WO) Shader core power on bitmap, low word */
+#define SHADER_PWRON_HI			0x184	/* (WO) Shader core power on bitmap, high word */
+
+#define TILER_PWRON_LO			0x190	/* (WO) Tiler core power on bitmap, low word */
+#define TILER_PWRON_HI			0x194	/* (WO) Tiler core power on bitmap, high word */
+
+#define L2_PWRON_LO			0x1A0	/* (WO) Level 2 cache power on bitmap, low word */
+#define L2_PWRON_HI			0x1A4	/* (WO) Level 2 cache power on bitmap, high word */
+
+#define STACK_PWRON_LO			0xE20   /* (RO) Core stack power on bitmap, low word */
+#define STACK_PWRON_HI			0xE24   /* (RO) Core stack power on bitmap, high word */
+
+
+#define SHADER_PWROFF_LO		0x1C0	/* (WO) Shader core power off bitmap, low word */
+#define SHADER_PWROFF_HI		0x1C4	/* (WO) Shader core power off bitmap, high word */
+
+#define TILER_PWROFF_LO			0x1D0	/* (WO) Tiler core power off bitmap, low word */
+#define TILER_PWROFF_HI			0x1D4	/* (WO) Tiler core power off bitmap, high word */
+
+#define L2_PWROFF_LO			0x1E0	/* (WO) Level 2 cache power off bitmap, low word */
+#define L2_PWROFF_HI			0x1E4	/* (WO) Level 2 cache power off bitmap, high word */
+
+#define STACK_PWROFF_LO			0xE30   /* (RO) Core stack power off bitmap, low word */
+#define STACK_PWROFF_HI			0xE34   /* (RO) Core stack power off bitmap, high word */
+
+
+#define SHADER_PWRTRANS_LO		0x200	/* (RO) Shader core power transition bitmap, low word */
+#define SHADER_PWRTRANS_HI		0x204	/* (RO) Shader core power transition bitmap, high word */
+
+#define TILER_PWRTRANS_LO		0x210	/* (RO) Tiler core power transition bitmap, low word */
+#define TILER_PWRTRANS_HI		0x214	/* (RO) Tiler core power transition bitmap, high word */
+
+#define L2_PWRTRANS_LO			0x220	/* (RO) Level 2 cache power transition bitmap, low word */
+#define L2_PWRTRANS_HI			0x224	/* (RO) Level 2 cache power transition bitmap, high word */
+
+#define STACK_PWRTRANS_LO		0xE40   /* (RO) Core stack power transition bitmap, low word */
+#define STACK_PWRTRANS_HI		0xE44   /* (RO) Core stack power transition bitmap, high word */
+
+
+#define SHADER_PWRACTIVE_LO		0x240	/* (RO) Shader core active bitmap, low word */
+#define SHADER_PWRACTIVE_HI		0x244	/* (RO) Shader core active bitmap, high word */
+
+#define TILER_PWRACTIVE_LO		0x250	/* (RO) Tiler core active bitmap, low word */
+#define TILER_PWRACTIVE_HI		0x254	/* (RO) Tiler core active bitmap, high word */
+
+#define L2_PWRACTIVE_LO			0x260	/* (RO) Level 2 cache active bitmap, low word */
+#define L2_PWRACTIVE_HI			0x264	/* (RO) Level 2 cache active bitmap, high word */
+
+#define GPU_JM_CONFIG			0xF00   /* (RW) Job Manager configuration register (Implementation specific register) */
+#define GPU_SHADER_CONFIG		0xF04	/* (RW) Shader core configuration settings (Implementation specific register) */
+#define GPU_TILER_CONFIG		0xF08   /* (RW) Tiler core configuration settings (Implementation specific register) */
+#define GPU_L2_MMU_CONFIG		0xF0C	/* (RW) Configuration of the L2 cache and MMU (Implementation specific register) */
+
+/* L2_MMU_CONFIG register */
+#define L2_MMU_CONFIG_ALLOW_SNOOP_DISPARITY_SHIFT	23
+#define L2_MMU_CONFIG_ALLOW_SNOOP_DISPARITY		(0x1 << L2_MMU_CONFIG_ALLOW_SNOOP_DISPARITY_SHIFT)
+#define L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_SHIFT	24
+#define L2_MMU_CONFIG_LIMIT_EXTERNAL_READS		(0x3 << L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_SHIFT)
+#define L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_OCTANT	(0x1 << L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_SHIFT)
+#define L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_QUARTER	(0x2 << L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_SHIFT)
+#define L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_HALF		(0x3 << L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_SHIFT)
+
+#define L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_SHIFT	26
+#define L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES		(0x3 << L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_SHIFT)
+#define L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_OCTANT	(0x1 << L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_SHIFT)
+#define L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_QUARTER	(0x2 << L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_SHIFT)
+#define L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_HALF	(0x3 << L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_SHIFT)
+
+#define L2_MMU_CONFIG_3BIT_LIMIT_EXTERNAL_READS_SHIFT	12
+#define L2_MMU_CONFIG_3BIT_LIMIT_EXTERNAL_READS		(0x7 << L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_SHIFT)
+
+#define L2_MMU_CONFIG_3BIT_LIMIT_EXTERNAL_WRITES_SHIFT	15
+#define L2_MMU_CONFIG_3BIT_LIMIT_EXTERNAL_WRITES	(0x7 << L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_SHIFT)
+
+/* SHADER_CONFIG register */
+#define SC_ALT_COUNTERS			BIT(3)
+#define SC_OVERRIDE_FWD_PIXEL_KILL	BIT(4)
+#define SC_SDC_DISABLE_OQ_DISCARD	BIT(6)
+#define SC_LS_ALLOW_ATTR_TYPES		BIT(16)
+#define SC_LS_PAUSEBUFFER_DISABLE	BIT(16)
+#define SC_TLS_HASH_ENABLE		BIT(17)
+#define SC_LS_ATTR_CHECK_DISABLE	BIT(18)
+#define SC_ENABLE_TEXGRD_FLAGS		BIT(25)
+/* End SHADER_CONFIG register */
+
+/* TILER_CONFIG register */
+#define TC_CLOCK_GATE_OVERRIDE		BIT(0)
+
+/* JM_CONFIG register */
+#define JM_TIMESTAMP_OVERRIDE		BIT(0)
+#define JM_CLOCK_GATE_OVERRIDE		BIT(1)
+#define JM_JOB_THROTTLE_ENABLE		BIT(2)
+#define JM_JOB_THROTTLE_LIMIT_SHIFT	3
+#define JM_MAX_JOB_THROTTLE_LIMIT	0x3F
+#define JM_FORCE_COHERENCY_FEATURES_SHIFT 2
+#define JM_IDVS_GROUP_SIZE_SHIFT	16
+#define JM_MAX_IDVS_GROUP_SIZE		0x3F
+
+
+/* Job Control regs */
+#define JOB_INT_RAWSTAT			0x1000
+#define JOB_INT_CLEAR			0x1004
+#define JOB_INT_MASK			0x1008
+#define JOB_INT_STAT			0x100c
+#define JOB_INT_JS_STATE		0x1010
+#define JOB_INT_THROTTLE		0x1014
+
+#define MK_JS_MASK(j)			(0x10001 << (j))
+#define JOB_INT_MASK_ERR(j)		BIT((j) + 16)
+#define JOB_INT_MASK_DONE(j)		BIT(j)
+
+#define JS_BASE				0x1800
+#define JS_HEAD_LO(n)			(JS_BASE + ((n) * 0x80) + 0x00)
+#define JS_HEAD_HI(n)			(JS_BASE + ((n) * 0x80) + 0x04)
+#define JS_TAIL_LO(n)			(JS_BASE + ((n) * 0x80) + 0x08)
+#define JS_TAIL_HI(n)			(JS_BASE + ((n) * 0x80) + 0x0c)
+#define JS_AFFINITY_LO(n)		(JS_BASE + ((n) * 0x80) + 0x10)
+#define JS_AFFINITY_HI(n)		(JS_BASE + ((n) * 0x80) + 0x14)
+#define JS_CONFIG(n)			(JS_BASE + ((n) * 0x80) + 0x18)
+#define JS_XAFFINITY(n)			(JS_BASE + ((n) * 0x80) + 0x1c)
+#define JS_COMMAND(n)			(JS_BASE + ((n) * 0x80) + 0x20)
+#define JS_STATUS(n)			(JS_BASE + ((n) * 0x80) + 0x24)
+#define JS_HEAD_NEXT_LO(n)		(JS_BASE + ((n) * 0x80) + 0x40)
+#define JS_HEAD_NEXT_HI(n)		(JS_BASE + ((n) * 0x80) + 0x44)
+#define JS_AFFINITY_NEXT_LO(n)		(JS_BASE + ((n) * 0x80) + 0x50)
+#define JS_AFFINITY_NEXT_HI(n)		(JS_BASE + ((n) * 0x80) + 0x54)
+#define JS_CONFIG_NEXT(n)		(JS_BASE + ((n) * 0x80) + 0x58)
+#define JS_COMMAND_NEXT(n)		(JS_BASE + ((n) * 0x80) + 0x60)
+#define JS_FLUSH_ID_NEXT(n)		(JS_BASE + ((n) * 0x80) + 0x70)
+
+/* Possible values of JS_CONFIG and JS_CONFIG_NEXT registers */
+#define JS_CONFIG_START_FLUSH_CLEAN		BIT(8)
+#define JS_CONFIG_START_FLUSH_CLEAN_INVALIDATE	(3u << 8)
+#define JS_CONFIG_START_MMU			BIT(10)
+#define JS_CONFIG_JOB_CHAIN_FLAG		BIT(11)
+#define JS_CONFIG_END_FLUSH_CLEAN		BIT(12)
+#define JS_CONFIG_END_FLUSH_CLEAN_INVALIDATE	(3u << 12)
+#define JS_CONFIG_ENABLE_FLUSH_REDUCTION	BIT(14)
+#define JS_CONFIG_DISABLE_DESCRIPTOR_WR_BK	BIT(15)
+#define JS_CONFIG_THREAD_PRI(n)			((n) << 16)
+
+#define JS_COMMAND_NOP			0x00
+#define JS_COMMAND_START		0x01
+#define JS_COMMAND_SOFT_STOP		0x02	/* Gently stop processing a job chain */
+#define JS_COMMAND_HARD_STOP		0x03	/* Rudely stop processing a job chain */
+#define JS_COMMAND_SOFT_STOP_0		0x04	/* Execute SOFT_STOP if JOB_CHAIN_FLAG is 0 */
+#define JS_COMMAND_HARD_STOP_0		0x05	/* Execute HARD_STOP if JOB_CHAIN_FLAG is 0 */
+#define JS_COMMAND_SOFT_STOP_1		0x06	/* Execute SOFT_STOP if JOB_CHAIN_FLAG is 1 */
+#define JS_COMMAND_HARD_STOP_1		0x07	/* Execute HARD_STOP if JOB_CHAIN_FLAG is 1 */
+
+#define JS_STATUS_EVENT_ACTIVE		0x08
+
+
+/* MMU regs */
+#define MMU_INT_RAWSTAT			0x2000
+#define MMU_INT_CLEAR			0x2004
+#define MMU_INT_MASK			0x2008
+#define MMU_INT_STAT			0x200c
+
+/* AS_COMMAND register commands */
+#define AS_COMMAND_NOP			0x00	/* NOP Operation */
+#define AS_COMMAND_UPDATE		0x01	/* Broadcasts the values in AS_TRANSTAB and ASn_MEMATTR to all MMUs */
+#define AS_COMMAND_LOCK			0x02	/* Issue a lock region command to all MMUs */
+#define AS_COMMAND_UNLOCK		0x03	/* Issue a flush region command to all MMUs */
+#define AS_COMMAND_FLUSH		0x04	/* Flush all L2 caches then issue a flush region command to all MMUs
+						   (deprecated - only for use with T60x) */
+#define AS_COMMAND_FLUSH_PT		0x04	/* Flush all L2 caches then issue a flush region command to all MMUs */
+#define AS_COMMAND_FLUSH_MEM		0x05	/* Wait for memory accesses to complete, flush all the L1s cache then
+						   flush all L2 caches then issue a flush region command to all MMUs */
+
+#define MMU_AS(as)			(0x2400 + ((as) << 6))
+
+#define AS_TRANSTAB_LO(as)		(MMU_AS(as) + 0x00) /* (RW) Translation Table Base Address for address space n, low word */
+#define AS_TRANSTAB_HI(as)		(MMU_AS(as) + 0x04) /* (RW) Translation Table Base Address for address space n, high word */
+#define AS_MEMATTR_LO(as)		(MMU_AS(as) + 0x08) /* (RW) Memory attributes for address space n, low word. */
+#define AS_MEMATTR_HI(as)		(MMU_AS(as) + 0x0C) /* (RW) Memory attributes for address space n, high word. */
+#define AS_LOCKADDR_LO(as)		(MMU_AS(as) + 0x10) /* (RW) Lock region address for address space n, low word */
+#define AS_LOCKADDR_HI(as)		(MMU_AS(as) + 0x14) /* (RW) Lock region address for address space n, high word */
+#define AS_COMMAND(as)			(MMU_AS(as) + 0x18) /* (WO) MMU command register for address space n */
+#define AS_FAULTSTATUS(as)		(MMU_AS(as) + 0x1C) /* (RO) MMU fault status register for address space n */
+#define AS_FAULTADDRESS_LO(as)		(MMU_AS(as) + 0x20) /* (RO) Fault Address for address space n, low word */
+#define AS_FAULTADDRESS_HI(as)		(MMU_AS(as) + 0x24) /* (RO) Fault Address for address space n, high word */
+#define AS_STATUS(as)			(MMU_AS(as) + 0x28) /* (RO) Status flags for address space n */
+/* Additional Bifrost AS regsiters */
+#define AS_TRANSCFG_LO(as)		(MMU_AS(as) + 0x30) /* (RW) Translation table configuration for address space n, low word */
+#define AS_TRANSCFG_HI(as)		(MMU_AS(as) + 0x34) /* (RW) Translation table configuration for address space n, high word */
+#define AS_FAULTEXTRA_LO(as)		(MMU_AS(as) + 0x38) /* (RO) Secondary fault address for address space n, low word */
+#define AS_FAULTEXTRA_HI(as)		(MMU_AS(as) + 0x3C) /* (RO) Secondary fault address for address space n, high word */
+
+/*
+ * Begin LPAE MMU TRANSTAB register values
+ */
+#define AS_TRANSTAB_LPAE_ADDR_SPACE_MASK	0xfffffffffffff000
+#define AS_TRANSTAB_LPAE_ADRMODE_IDENTITY	0x2
+#define AS_TRANSTAB_LPAE_ADRMODE_TABLE		0x3
+#define AS_TRANSTAB_LPAE_ADRMODE_MASK		0x3
+#define AS_TRANSTAB_LPAE_READ_INNER		BIT(2)
+#define AS_TRANSTAB_LPAE_SHARE_OUTER		BIT(4)
+
+#define AS_STATUS_AS_ACTIVE			0x01
+
+#define AS_FAULTSTATUS_ACCESS_TYPE_MASK		(0x3 << 8)
+#define AS_FAULTSTATUS_ACCESS_TYPE_ATOMIC	(0x0 << 8)
+#define AS_FAULTSTATUS_ACCESS_TYPE_EX		(0x1 << 8)
+#define AS_FAULTSTATUS_ACCESS_TYPE_READ		(0x2 << 8)
+#define AS_FAULTSTATUS_ACCESS_TYPE_WRITE	(0x3 << 8)
+
+#endif
diff --git a/drivers/gpu/drm/pl111/pl111_display.c b/drivers/gpu/drm/pl111/pl111_display.c
index 754f6b25f265..0c5d391f0a8f 100644
--- a/drivers/gpu/drm/pl111/pl111_display.c
+++ b/drivers/gpu/drm/pl111/pl111_display.c
@@ -188,7 +188,7 @@ static void pl111_display_enable(struct drm_simple_display_pipe *pipe,
 			tim2 |= TIM2_IOE;
 
 		if (connector->display_info.bus_flags &
-		    DRM_BUS_FLAG_PIXDATA_NEGEDGE)
+		    DRM_BUS_FLAG_PIXDATA_DRIVE_NEGEDGE)
 			tim2 |= TIM2_IPC;
 	}
 
diff --git a/drivers/gpu/drm/pl111/pl111_versatile.c b/drivers/gpu/drm/pl111/pl111_versatile.c
index b9baefdba38a..1c318ad32a8c 100644
--- a/drivers/gpu/drm/pl111/pl111_versatile.c
+++ b/drivers/gpu/drm/pl111/pl111_versatile.c
@@ -330,6 +330,7 @@ int pl111_versatile_init(struct device *dev, struct pl111_drm_dev_private *priv)
 		ret = vexpress_muxfpga_init();
 		if (ret) {
 			dev_err(dev, "unable to initialize muxfpga driver\n");
+			of_node_put(np);
 			return ret;
 		}
 
@@ -337,17 +338,20 @@ int pl111_versatile_init(struct device *dev, struct pl111_drm_dev_private *priv)
 		pdev = of_find_device_by_node(np);
 		if (!pdev) {
 			dev_err(dev, "can't find the sysreg device, deferring\n");
+			of_node_put(np);
 			return -EPROBE_DEFER;
 		}
 		map = dev_get_drvdata(&pdev->dev);
 		if (!map) {
 			dev_err(dev, "sysreg has not yet probed\n");
 			platform_device_put(pdev);
+			of_node_put(np);
 			return -EPROBE_DEFER;
 		}
 	} else {
 		map = syscon_node_to_regmap(np);
 	}
+	of_node_put(np);
 
 	if (IS_ERR(map)) {
 		dev_err(dev, "no Versatile syscon regmap\n");
diff --git a/drivers/gpu/drm/qxl/qxl_display.c b/drivers/gpu/drm/qxl/qxl_display.c
index 08c725544a2f..8b319ebbb0fb 100644
--- a/drivers/gpu/drm/qxl/qxl_display.c
+++ b/drivers/gpu/drm/qxl/qxl_display.c
@@ -535,7 +535,7 @@ static void qxl_primary_atomic_update(struct drm_plane *plane,
 {
 	struct qxl_device *qdev = plane->dev->dev_private;
 	struct qxl_bo *bo = gem_to_qxl_bo(plane->state->fb->obj[0]);
-	struct qxl_bo *bo_old, *primary;
+	struct qxl_bo *primary;
 	struct drm_clip_rect norect = {
 	    .x1 = 0,
 	    .y1 = 0,
@@ -544,12 +544,6 @@ static void qxl_primary_atomic_update(struct drm_plane *plane,
 	};
 	uint32_t dumb_shadow_offset = 0;
 
-	if (old_state->fb) {
-		bo_old = gem_to_qxl_bo(old_state->fb->obj[0]);
-	} else {
-		bo_old = NULL;
-	}
-
 	primary = bo->shadow ? bo->shadow : bo;
 
 	if (!primary->is_primary) {
diff --git a/drivers/gpu/drm/qxl/qxl_drv.h b/drivers/gpu/drm/qxl/qxl_drv.h
index 4a0331b3ff7d..2896bb6fdbf4 100644
--- a/drivers/gpu/drm/qxl/qxl_drv.h
+++ b/drivers/gpu/drm/qxl/qxl_drv.h
@@ -65,9 +65,6 @@
 extern int qxl_num_crtc;
 extern int qxl_max_ioctls;
 
-#define DRM_FILE_OFFSET 0x100000000ULL
-#define DRM_FILE_PAGE_OFFSET (DRM_FILE_OFFSET >> PAGE_SHIFT)
-
 #define QXL_INTERRUPT_MASK (\
 	QXL_INTERRUPT_DISPLAY |\
 	QXL_INTERRUPT_CURSOR |\
diff --git a/drivers/gpu/drm/qxl/qxl_ttm.c b/drivers/gpu/drm/qxl/qxl_ttm.c
index 92f5db5b296f..0234f8556ada 100644
--- a/drivers/gpu/drm/qxl/qxl_ttm.c
+++ b/drivers/gpu/drm/qxl/qxl_ttm.c
@@ -63,15 +63,10 @@ static vm_fault_t qxl_ttm_fault(struct vm_fault *vmf)
 
 int qxl_mmap(struct file *filp, struct vm_area_struct *vma)
 {
-	struct drm_file *file_priv;
-	struct qxl_device *qdev;
 	int r;
+	struct drm_file *file_priv = filp->private_data;
+	struct qxl_device *qdev = file_priv->minor->dev->dev_private;
 
-	if (unlikely(vma->vm_pgoff < DRM_FILE_PAGE_OFFSET))
-		return -EINVAL;
-
-	file_priv = filp->private_data;
-	qdev = file_priv->minor->dev->dev_private;
 	if (qdev == NULL) {
 		DRM_ERROR(
 		 "filp->private_data->minor->dev->dev_private == NULL\n");
@@ -328,7 +323,7 @@ int qxl_ttm_init(struct qxl_device *qdev)
 	r = ttm_bo_device_init(&qdev->mman.bdev,
 			       &qxl_bo_driver,
 			       qdev->ddev.anon_inode->i_mapping,
-			       DRM_FILE_PAGE_OFFSET, 0);
+			       false);
 	if (r) {
 		DRM_ERROR("failed initializing buffer object driver(%d).\n", r);
 		return r;
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index 53f29a115104..0a9312ea250a 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -1388,7 +1388,7 @@ int radeon_device_init(struct radeon_device *rdev,
 		pci_set_consistent_dma_mask(rdev->pdev, DMA_BIT_MASK(32));
 		pr_warn("radeon: No coherent DMA available\n");
 	}
-	rdev->need_swiotlb = drm_get_max_iomem() > ((u64)1 << dma_bits);
+	rdev->need_swiotlb = drm_need_swiotlb(dma_bits);
 
 	/* Registers mapping */
 	/* TODO: block userspace mapping of io register */
diff --git a/drivers/gpu/drm/radeon/radeon_drv.h b/drivers/gpu/drm/radeon/radeon_drv.h
index afef2d9fccd8..173deb463414 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.h
+++ b/drivers/gpu/drm/radeon/radeon_drv.h
@@ -35,7 +35,6 @@
 #include <linux/platform_device.h>
 #include <drm/drm_legacy.h>
 
-#include <drm/ati_pcigart.h>
 #include "radeon_family.h"
 
 /* General customization:
diff --git a/drivers/gpu/drm/radeon/radeon_fb.c b/drivers/gpu/drm/radeon/radeon_fb.c
index 1179034024ae..1298b84cb1c7 100644
--- a/drivers/gpu/drm/radeon/radeon_fb.c
+++ b/drivers/gpu/drm/radeon/radeon_fb.c
@@ -42,7 +42,7 @@
  * the helper contains a pointer to radeon framebuffer baseclass.
  */
 struct radeon_fbdev {
-	struct drm_fb_helper helper;
+	struct drm_fb_helper helper; /* must be first */
 	struct drm_framebuffer fb;
 	struct radeon_device *rdev;
 };
@@ -244,7 +244,8 @@ static int radeonfb_create(struct drm_fb_helper *helper,
 		goto out;
 	}
 
-	info->par = rfbdev;
+	/* radeon resume is fragile and needs a vt switch to help it along */
+	info->skip_vt_switch = false;
 
 	ret = radeon_framebuffer_init(rdev->ddev, &rfbdev->fb, &mode_cmd, gobj);
 	if (ret) {
@@ -259,10 +260,6 @@ static int radeonfb_create(struct drm_fb_helper *helper,
 
 	memset_io(rbo->kptr, 0x0, radeon_bo_size(rbo));
 
-	strcpy(info->fix.id, "radeondrmfb");
-
-	drm_fb_helper_fill_fix(info, fb->pitches[0], fb->format->depth);
-
 	info->fbops = &radeonfb_ops;
 
 	tmp = radeon_bo_gpu_offset(rbo) - rdev->mc.vram_start;
@@ -271,7 +268,7 @@ static int radeonfb_create(struct drm_fb_helper *helper,
 	info->screen_base = rbo->kptr;
 	info->screen_size = radeon_bo_size(rbo);
 
-	drm_fb_helper_fill_var(info, &rfbdev->helper, sizes->fb_width, sizes->fb_height);
+	drm_fb_helper_fill_info(info, &rfbdev->helper, sizes);
 
 	/* setup aperture base/size for vesafb takeover */
 	info->apertures->ranges[0].base = rdev->ddev->mode_config.fb_base;
diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c
index 9920a6fc11bf..5d42f8d8e68d 100644
--- a/drivers/gpu/drm/radeon/radeon_ttm.c
+++ b/drivers/gpu/drm/radeon/radeon_ttm.c
@@ -45,8 +45,6 @@
 #include "radeon_reg.h"
 #include "radeon.h"
 
-#define DRM_FILE_PAGE_OFFSET (0x100000000ULL >> PAGE_SHIFT)
-
 static int radeon_ttm_debugfs_init(struct radeon_device *rdev);
 static void radeon_ttm_debugfs_fini(struct radeon_device *rdev);
 
@@ -253,14 +251,12 @@ static int radeon_move_vram_ram(struct ttm_buffer_object *bo,
 				struct ttm_mem_reg *new_mem)
 {
 	struct ttm_operation_ctx ctx = { interruptible, no_wait_gpu };
-	struct radeon_device *rdev;
 	struct ttm_mem_reg *old_mem = &bo->mem;
 	struct ttm_mem_reg tmp_mem;
 	struct ttm_place placements;
 	struct ttm_placement placement;
 	int r;
 
-	rdev = radeon_get_rdev(bo->bdev);
 	tmp_mem = *new_mem;
 	tmp_mem.mm_node = NULL;
 	placement.num_placement = 1;
@@ -300,14 +296,12 @@ static int radeon_move_ram_vram(struct ttm_buffer_object *bo,
 				struct ttm_mem_reg *new_mem)
 {
 	struct ttm_operation_ctx ctx = { interruptible, no_wait_gpu };
-	struct radeon_device *rdev;
 	struct ttm_mem_reg *old_mem = &bo->mem;
 	struct ttm_mem_reg tmp_mem;
 	struct ttm_placement placement;
 	struct ttm_place placements;
 	int r;
 
-	rdev = radeon_get_rdev(bo->bdev);
 	tmp_mem = *new_mem;
 	tmp_mem.mm_node = NULL;
 	placement.num_placement = 1;
@@ -792,7 +786,6 @@ int radeon_ttm_init(struct radeon_device *rdev)
 	r = ttm_bo_device_init(&rdev->mman.bdev,
 			       &radeon_bo_driver,
 			       rdev->ddev->anon_inode->i_mapping,
-			       DRM_FILE_PAGE_OFFSET,
 			       rdev->need_dma32);
 	if (r) {
 		DRM_ERROR("failed initializing buffer object driver(%d).\n", r);
@@ -901,16 +894,10 @@ static vm_fault_t radeon_ttm_fault(struct vm_fault *vmf)
 
 int radeon_mmap(struct file *filp, struct vm_area_struct *vma)
 {
-	struct drm_file *file_priv;
-	struct radeon_device *rdev;
 	int r;
+	struct drm_file *file_priv = filp->private_data;
+	struct radeon_device *rdev = file_priv->minor->dev->dev_private;
 
-	if (unlikely(vma->vm_pgoff < DRM_FILE_PAGE_OFFSET)) {
-		return -EINVAL;
-	}
-
-	file_priv = filp->private_data;
-	rdev = file_priv->minor->dev->dev_private;
 	if (rdev == NULL) {
 		return -EINVAL;
 	}
diff --git a/drivers/gpu/drm/rcar-du/Kconfig b/drivers/gpu/drm/rcar-du/Kconfig
index 7c36e2777a15..1529849e217e 100644
--- a/drivers/gpu/drm/rcar-du/Kconfig
+++ b/drivers/gpu/drm/rcar-du/Kconfig
@@ -36,3 +36,7 @@ config DRM_RCAR_VSP
 	depends on VIDEO_RENESAS_VSP1=y || (VIDEO_RENESAS_VSP1 && DRM_RCAR_DU=m)
 	help
 	  Enable support to expose the R-Car VSP Compositor as KMS planes.
+
+config DRM_RCAR_WRITEBACK
+	bool
+	default y if ARM64
diff --git a/drivers/gpu/drm/rcar-du/Makefile b/drivers/gpu/drm/rcar-du/Makefile
index 2a3b8d7972b5..6c2ed9c46467 100644
--- a/drivers/gpu/drm/rcar-du/Makefile
+++ b/drivers/gpu/drm/rcar-du/Makefile
@@ -4,7 +4,7 @@ rcar-du-drm-y := rcar_du_crtc.o \
 		 rcar_du_encoder.o \
 		 rcar_du_group.o \
 		 rcar_du_kms.o \
-		 rcar_du_plane.o
+		 rcar_du_plane.o \
 
 rcar-du-drm-$(CONFIG_DRM_RCAR_LVDS)	+= rcar_du_of.o \
 					   rcar_du_of_lvds_r8a7790.dtb.o \
@@ -13,6 +13,7 @@ rcar-du-drm-$(CONFIG_DRM_RCAR_LVDS)	+= rcar_du_of.o \
 					   rcar_du_of_lvds_r8a7795.dtb.o \
 					   rcar_du_of_lvds_r8a7796.dtb.o
 rcar-du-drm-$(CONFIG_DRM_RCAR_VSP)	+= rcar_du_vsp.o
+rcar-du-drm-$(CONFIG_DRM_RCAR_WRITEBACK) += rcar_du_writeback.o
 
 obj-$(CONFIG_DRM_RCAR_DU)		+= rcar-du-drm.o
 obj-$(CONFIG_DRM_RCAR_DW_HDMI)		+= rcar_dw_hdmi.o
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_crtc.c b/drivers/gpu/drm/rcar-du/rcar_du_crtc.c
index 4cdea14d552f..2da46e3dc4ae 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_crtc.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_crtc.c
@@ -32,21 +32,21 @@
 
 static u32 rcar_du_crtc_read(struct rcar_du_crtc *rcrtc, u32 reg)
 {
-	struct rcar_du_device *rcdu = rcrtc->group->dev;
+	struct rcar_du_device *rcdu = rcrtc->dev;
 
 	return rcar_du_read(rcdu, rcrtc->mmio_offset + reg);
 }
 
 static void rcar_du_crtc_write(struct rcar_du_crtc *rcrtc, u32 reg, u32 data)
 {
-	struct rcar_du_device *rcdu = rcrtc->group->dev;
+	struct rcar_du_device *rcdu = rcrtc->dev;
 
 	rcar_du_write(rcdu, rcrtc->mmio_offset + reg, data);
 }
 
 static void rcar_du_crtc_clr(struct rcar_du_crtc *rcrtc, u32 reg, u32 clr)
 {
-	struct rcar_du_device *rcdu = rcrtc->group->dev;
+	struct rcar_du_device *rcdu = rcrtc->dev;
 
 	rcar_du_write(rcdu, rcrtc->mmio_offset + reg,
 		      rcar_du_read(rcdu, rcrtc->mmio_offset + reg) & ~clr);
@@ -54,7 +54,7 @@ static void rcar_du_crtc_clr(struct rcar_du_crtc *rcrtc, u32 reg, u32 clr)
 
 static void rcar_du_crtc_set(struct rcar_du_crtc *rcrtc, u32 reg, u32 set)
 {
-	struct rcar_du_device *rcdu = rcrtc->group->dev;
+	struct rcar_du_device *rcdu = rcrtc->dev;
 
 	rcar_du_write(rcdu, rcrtc->mmio_offset + reg,
 		      rcar_du_read(rcdu, rcrtc->mmio_offset + reg) | set);
@@ -62,7 +62,7 @@ static void rcar_du_crtc_set(struct rcar_du_crtc *rcrtc, u32 reg, u32 set)
 
 void rcar_du_crtc_dsysr_clr_set(struct rcar_du_crtc *rcrtc, u32 clr, u32 set)
 {
-	struct rcar_du_device *rcdu = rcrtc->group->dev;
+	struct rcar_du_device *rcdu = rcrtc->dev;
 
 	rcrtc->dsysr = (rcrtc->dsysr & ~clr) | set;
 	rcar_du_write(rcdu, rcrtc->mmio_offset + DSYSR, rcrtc->dsysr);
@@ -157,10 +157,9 @@ static void rcar_du_dpll_divider(struct rcar_du_crtc *rcrtc,
 	}
 
 done:
-	dev_dbg(rcrtc->group->dev->dev,
+	dev_dbg(rcrtc->dev->dev,
 		"output:%u, fdpll:%u, n:%u, m:%u, diff:%lu\n",
-		 dpll->output, dpll->fdpll, dpll->n, dpll->m,
-		 best_diff);
+		 dpll->output, dpll->fdpll, dpll->n, dpll->m, best_diff);
 }
 
 struct du_clk_params {
@@ -212,7 +211,7 @@ static const struct soc_device_attribute rcar_du_r8a7795_es1[] = {
 static void rcar_du_crtc_set_display_timing(struct rcar_du_crtc *rcrtc)
 {
 	const struct drm_display_mode *mode = &rcrtc->crtc.state->adjusted_mode;
-	struct rcar_du_device *rcdu = rcrtc->group->dev;
+	struct rcar_du_device *rcdu = rcrtc->dev;
 	unsigned long mode_clock = mode->clock * 1000;
 	u32 dsmr;
 	u32 escr;
@@ -277,7 +276,7 @@ static void rcar_du_crtc_set_display_timing(struct rcar_du_crtc *rcrtc)
 			rcar_du_escr_divider(rcrtc->extclock, mode_clock,
 					     ESCR_DCLKSEL_DCLKIN, &params);
 
-		dev_dbg(rcrtc->group->dev->dev,	"mode clock %lu %s rate %lu\n",
+		dev_dbg(rcrtc->dev->dev, "mode clock %lu %s rate %lu\n",
 			mode_clock, params.clk == rcrtc->clock ? "cpg" : "ext",
 			params.rate);
 
@@ -285,7 +284,7 @@ static void rcar_du_crtc_set_display_timing(struct rcar_du_crtc *rcrtc)
 		escr = params.escr;
 	}
 
-	dev_dbg(rcrtc->group->dev->dev, "%s: ESCR 0x%08x\n", __func__, escr);
+	dev_dbg(rcrtc->dev->dev, "%s: ESCR 0x%08x\n", __func__, escr);
 
 	rcar_du_crtc_write(rcrtc, rcrtc->index % 2 ? ESCR13 : ESCR02, escr);
 	rcar_du_crtc_write(rcrtc, rcrtc->index % 2 ? OTAR13 : OTAR02, 0);
@@ -333,7 +332,7 @@ plane_format(struct rcar_du_plane *plane)
 static void rcar_du_crtc_update_planes(struct rcar_du_crtc *rcrtc)
 {
 	struct rcar_du_plane *planes[RCAR_DU_NUM_HW_PLANES];
-	struct rcar_du_device *rcdu = rcrtc->group->dev;
+	struct rcar_du_device *rcdu = rcrtc->dev;
 	unsigned int num_planes = 0;
 	unsigned int dptsr_planes;
 	unsigned int hwplanes = 0;
@@ -463,7 +462,7 @@ static bool rcar_du_crtc_page_flip_pending(struct rcar_du_crtc *rcrtc)
 
 static void rcar_du_crtc_wait_page_flip(struct rcar_du_crtc *rcrtc)
 {
-	struct rcar_du_device *rcdu = rcrtc->group->dev;
+	struct rcar_du_device *rcdu = rcrtc->dev;
 
 	if (wait_event_timeout(rcrtc->flip_wait,
 			       !rcar_du_crtc_page_flip_pending(rcrtc),
@@ -493,7 +492,7 @@ static void rcar_du_crtc_setup(struct rcar_du_crtc *rcrtc)
 	rcar_du_group_write(rcrtc->group, rcrtc->index % 2 ? DS2PR : DS1PR, 0);
 
 	/* Enable the VSP compositor. */
-	if (rcar_du_has(rcrtc->group->dev, RCAR_DU_FEATURE_VSP1_SOURCE))
+	if (rcar_du_has(rcrtc->dev, RCAR_DU_FEATURE_VSP1_SOURCE))
 		rcar_du_vsp_enable(rcrtc);
 
 	/* Turn vertical blanking interrupt reporting on. */
@@ -564,7 +563,7 @@ static void rcar_du_crtc_start(struct rcar_du_crtc *rcrtc)
 
 static void rcar_du_crtc_disable_planes(struct rcar_du_crtc *rcrtc)
 {
-	struct rcar_du_device *rcdu = rcrtc->group->dev;
+	struct rcar_du_device *rcdu = rcrtc->dev;
 	struct drm_crtc *crtc = &rcrtc->crtc;
 	u32 status;
 
@@ -617,7 +616,7 @@ static void rcar_du_crtc_stop(struct rcar_du_crtc *rcrtc)
 	drm_crtc_vblank_off(crtc);
 
 	/* Disable the VSP compositor. */
-	if (rcar_du_has(rcrtc->group->dev, RCAR_DU_FEATURE_VSP1_SOURCE))
+	if (rcar_du_has(rcrtc->dev, RCAR_DU_FEATURE_VSP1_SOURCE))
 		rcar_du_vsp_disable(rcrtc);
 
 	/*
@@ -627,7 +626,7 @@ static void rcar_du_crtc_stop(struct rcar_du_crtc *rcrtc)
 	 * TODO: Find another way to stop the display for DUs that don't support
 	 * TVM sync.
 	 */
-	if (rcar_du_has(rcrtc->group->dev, RCAR_DU_FEATURE_TVM_SYNC))
+	if (rcar_du_has(rcrtc->dev, RCAR_DU_FEATURE_TVM_SYNC))
 		rcar_du_crtc_dsysr_clr_set(rcrtc, DSYSR_TVM_MASK,
 					   DSYSR_TVM_SWITCH);
 
@@ -648,8 +647,13 @@ static int rcar_du_crtc_atomic_check(struct drm_crtc *crtc,
 	rstate->outputs = 0;
 
 	drm_for_each_encoder_mask(encoder, crtc->dev, state->encoder_mask) {
-		struct rcar_du_encoder *renc = to_rcar_encoder(encoder);
+		struct rcar_du_encoder *renc;
 
+		/* Skip the writeback encoder. */
+		if (encoder->encoder_type == DRM_MODE_ENCODER_VIRTUAL)
+			continue;
+
+		renc = to_rcar_encoder(encoder);
 		rstate->outputs |= BIT(renc->output);
 	}
 
@@ -661,7 +665,7 @@ static void rcar_du_crtc_atomic_enable(struct drm_crtc *crtc,
 {
 	struct rcar_du_crtc *rcrtc = to_rcar_crtc(crtc);
 	struct rcar_du_crtc_state *rstate = to_rcar_crtc_state(crtc->state);
-	struct rcar_du_device *rcdu = rcrtc->group->dev;
+	struct rcar_du_device *rcdu = rcrtc->dev;
 
 	rcar_du_crtc_get(rcrtc);
 
@@ -689,7 +693,7 @@ static void rcar_du_crtc_atomic_disable(struct drm_crtc *crtc,
 {
 	struct rcar_du_crtc *rcrtc = to_rcar_crtc(crtc);
 	struct rcar_du_crtc_state *rstate = to_rcar_crtc_state(old_state);
-	struct rcar_du_device *rcdu = rcrtc->group->dev;
+	struct rcar_du_device *rcdu = rcrtc->dev;
 
 	rcar_du_crtc_stop(rcrtc);
 	rcar_du_crtc_put(rcrtc);
@@ -735,7 +739,7 @@ static void rcar_du_crtc_atomic_begin(struct drm_crtc *crtc,
 	 */
 	rcar_du_crtc_get(rcrtc);
 
-	if (rcar_du_has(rcrtc->group->dev, RCAR_DU_FEATURE_VSP1_SOURCE))
+	if (rcar_du_has(rcrtc->dev, RCAR_DU_FEATURE_VSP1_SOURCE))
 		rcar_du_vsp_atomic_begin(rcrtc);
 }
 
@@ -757,15 +761,16 @@ static void rcar_du_crtc_atomic_flush(struct drm_crtc *crtc,
 		spin_unlock_irqrestore(&dev->event_lock, flags);
 	}
 
-	if (rcar_du_has(rcrtc->group->dev, RCAR_DU_FEATURE_VSP1_SOURCE))
+	if (rcar_du_has(rcrtc->dev, RCAR_DU_FEATURE_VSP1_SOURCE))
 		rcar_du_vsp_atomic_flush(rcrtc);
 }
 
-enum drm_mode_status rcar_du_crtc_mode_valid(struct drm_crtc *crtc,
-				   const struct drm_display_mode *mode)
+static enum drm_mode_status
+rcar_du_crtc_mode_valid(struct drm_crtc *crtc,
+			const struct drm_display_mode *mode)
 {
 	struct rcar_du_crtc *rcrtc = to_rcar_crtc(crtc);
-	struct rcar_du_device *rcdu = rcrtc->group->dev;
+	struct rcar_du_device *rcdu = rcrtc->dev;
 	bool interlaced = mode->flags & DRM_MODE_FLAG_INTERLACE;
 	unsigned int vbp;
 
@@ -797,7 +802,7 @@ static const struct drm_crtc_helper_funcs crtc_helper_funcs = {
 
 static void rcar_du_crtc_crc_init(struct rcar_du_crtc *rcrtc)
 {
-	struct rcar_du_device *rcdu = rcrtc->group->dev;
+	struct rcar_du_device *rcdu = rcrtc->dev;
 	const char **sources;
 	unsigned int count;
 	int i = -1;
@@ -981,8 +986,8 @@ static int rcar_du_crtc_verify_crc_source(struct drm_crtc *crtc,
 	return 0;
 }
 
-const char *const *rcar_du_crtc_get_crc_sources(struct drm_crtc *crtc,
-						size_t *count)
+static const char *const *
+rcar_du_crtc_get_crc_sources(struct drm_crtc *crtc, size_t *count)
 {
 	struct rcar_du_crtc *rcrtc = to_rcar_crtc(crtc);
 
@@ -1079,7 +1084,7 @@ static const struct drm_crtc_funcs crtc_funcs_gen3 = {
 static irqreturn_t rcar_du_crtc_irq(int irq, void *arg)
 {
 	struct rcar_du_crtc *rcrtc = arg;
-	struct rcar_du_device *rcdu = rcrtc->group->dev;
+	struct rcar_du_device *rcdu = rcrtc->dev;
 	irqreturn_t ret = IRQ_NONE;
 	u32 status;
 
@@ -1171,6 +1176,7 @@ int rcar_du_crtc_create(struct rcar_du_group *rgrp, unsigned int swindex,
 	init_waitqueue_head(&rcrtc->vblank_wait);
 	spin_lock_init(&rcrtc->vblank_lock);
 
+	rcrtc->dev = rcdu;
 	rcrtc->group = rgrp;
 	rcrtc->mmio_offset = mmio_offsets[hwindex];
 	rcrtc->index = hwindex;
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_crtc.h b/drivers/gpu/drm/rcar-du/rcar_du_crtc.h
index bcb35b0b7612..3b7fc668996f 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_crtc.h
+++ b/drivers/gpu/drm/rcar-du/rcar_du_crtc.h
@@ -15,6 +15,7 @@
 #include <linux/wait.h>
 
 #include <drm/drm_crtc.h>
+#include <drm/drm_writeback.h>
 
 #include <media/vsp1.h>
 
@@ -24,10 +25,11 @@ struct rcar_du_vsp;
 /**
  * struct rcar_du_crtc - the CRTC, representing a DU superposition processor
  * @crtc: base DRM CRTC
+ * @dev: the DU device
  * @clock: the CRTC functional clock
  * @extclock: external pixel dot clock (optional)
  * @mmio_offset: offset of the CRTC registers in the DU MMIO block
- * @index: CRTC software and hardware index
+ * @index: CRTC hardware index
  * @initialized: whether the CRTC has been initialized and clocks enabled
  * @dsysr: cached value of the DSYSR register
  * @vblank_enable: whether vblank events are enabled on this CRTC
@@ -39,10 +41,12 @@ struct rcar_du_vsp;
  * @group: CRTC group this CRTC belongs to
  * @vsp: VSP feeding video to this CRTC
  * @vsp_pipe: index of the VSP pipeline feeding video to this CRTC
+ * @writeback: the writeback connector
  */
 struct rcar_du_crtc {
 	struct drm_crtc crtc;
 
+	struct rcar_du_device *dev;
 	struct clk *clock;
 	struct clk *extclock;
 	unsigned int mmio_offset;
@@ -65,9 +69,12 @@ struct rcar_du_crtc {
 
 	const char *const *sources;
 	unsigned int sources_count;
+
+	struct drm_writeback_connector writeback;
 };
 
-#define to_rcar_crtc(c)	container_of(c, struct rcar_du_crtc, crtc)
+#define to_rcar_crtc(c)		container_of(c, struct rcar_du_crtc, crtc)
+#define wb_to_rcar_crtc(c)	container_of(c, struct rcar_du_crtc, writeback)
 
 /**
  * struct rcar_du_crtc_state - Driver-specific CRTC state
@@ -97,8 +104,6 @@ enum rcar_du_output {
 
 int rcar_du_crtc_create(struct rcar_du_group *rgrp, unsigned int swindex,
 			unsigned int hwindex);
-void rcar_du_crtc_suspend(struct rcar_du_crtc *rcrtc);
-void rcar_du_crtc_resume(struct rcar_du_crtc *rcrtc);
 
 void rcar_du_crtc_finish_page_flip(struct rcar_du_crtc *rcrtc);
 
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_encoder.c b/drivers/gpu/drm/rcar-du/rcar_du_encoder.c
index 8ee4e762f4e5..6c91753af7bc 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_encoder.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_encoder.c
@@ -28,13 +28,33 @@ static const struct drm_encoder_funcs encoder_funcs = {
 	.destroy = drm_encoder_cleanup,
 };
 
+static unsigned int rcar_du_encoder_count_ports(struct device_node *node)
+{
+	struct device_node *ports;
+	struct device_node *port;
+	unsigned int num_ports = 0;
+
+	ports = of_get_child_by_name(node, "ports");
+	if (!ports)
+		ports = of_node_get(node);
+
+	for_each_child_of_node(ports, port) {
+		if (of_node_name_eq(port, "port"))
+			num_ports++;
+	}
+
+	of_node_put(ports);
+
+	return num_ports;
+}
+
 int rcar_du_encoder_init(struct rcar_du_device *rcdu,
 			 enum rcar_du_output output,
 			 struct device_node *enc_node)
 {
 	struct rcar_du_encoder *renc;
 	struct drm_encoder *encoder;
-	struct drm_bridge *bridge = NULL;
+	struct drm_bridge *bridge;
 	int ret;
 
 	renc = devm_kzalloc(rcdu->dev, sizeof(*renc), GFP_KERNEL);
@@ -48,11 +68,33 @@ int rcar_du_encoder_init(struct rcar_du_device *rcdu,
 	dev_dbg(rcdu->dev, "initializing encoder %pOF for output %u\n",
 		enc_node, output);
 
-	/* Locate the DRM bridge from the encoder DT node. */
-	bridge = of_drm_find_bridge(enc_node);
-	if (!bridge) {
-		ret = -EPROBE_DEFER;
-		goto done;
+	/*
+	 * Locate the DRM bridge from the DT node. For the DPAD outputs, if the
+	 * DT node has a single port, assume that it describes a panel and
+	 * create a panel bridge.
+	 */
+	if ((output == RCAR_DU_OUTPUT_DPAD0 ||
+	     output == RCAR_DU_OUTPUT_DPAD1) &&
+	    rcar_du_encoder_count_ports(enc_node) == 1) {
+		struct drm_panel *panel = of_drm_find_panel(enc_node);
+
+		if (IS_ERR(panel)) {
+			ret = PTR_ERR(panel);
+			goto done;
+		}
+
+		bridge = devm_drm_panel_bridge_add(rcdu->dev, panel,
+						   DRM_MODE_CONNECTOR_DPI);
+		if (IS_ERR(bridge)) {
+			ret = PTR_ERR(bridge);
+			goto done;
+		}
+	} else {
+		bridge = of_drm_find_bridge(enc_node);
+		if (!bridge) {
+			ret = -EPROBE_DEFER;
+			goto done;
+		}
 	}
 
 	ret = drm_encoder_init(rcdu->ddev, encoder, &encoder_funcs,
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_kms.c b/drivers/gpu/drm/rcar-du/rcar_du_kms.c
index 3b7d50a8fb9b..f8f7fff34dff 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_kms.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_kms.c
@@ -26,6 +26,7 @@
 #include "rcar_du_kms.h"
 #include "rcar_du_regs.h"
 #include "rcar_du_vsp.h"
+#include "rcar_du_writeback.h"
 
 /* -----------------------------------------------------------------------------
  * Format helpers
@@ -34,60 +35,70 @@
 static const struct rcar_du_format_info rcar_du_format_infos[] = {
 	{
 		.fourcc = DRM_FORMAT_RGB565,
+		.v4l2 = V4L2_PIX_FMT_RGB565,
 		.bpp = 16,
 		.planes = 1,
 		.pnmr = PnMR_SPIM_TP | PnMR_DDDF_16BPP,
 		.edf = PnDDCR4_EDF_NONE,
 	}, {
 		.fourcc = DRM_FORMAT_ARGB1555,
+		.v4l2 = V4L2_PIX_FMT_ARGB555,
 		.bpp = 16,
 		.planes = 1,
 		.pnmr = PnMR_SPIM_ALP | PnMR_DDDF_ARGB,
 		.edf = PnDDCR4_EDF_NONE,
 	}, {
 		.fourcc = DRM_FORMAT_XRGB1555,
+		.v4l2 = V4L2_PIX_FMT_XRGB555,
 		.bpp = 16,
 		.planes = 1,
 		.pnmr = PnMR_SPIM_ALP | PnMR_DDDF_ARGB,
 		.edf = PnDDCR4_EDF_NONE,
 	}, {
 		.fourcc = DRM_FORMAT_XRGB8888,
+		.v4l2 = V4L2_PIX_FMT_XBGR32,
 		.bpp = 32,
 		.planes = 1,
 		.pnmr = PnMR_SPIM_TP | PnMR_DDDF_16BPP,
 		.edf = PnDDCR4_EDF_RGB888,
 	}, {
 		.fourcc = DRM_FORMAT_ARGB8888,
+		.v4l2 = V4L2_PIX_FMT_ABGR32,
 		.bpp = 32,
 		.planes = 1,
 		.pnmr = PnMR_SPIM_ALP | PnMR_DDDF_16BPP,
 		.edf = PnDDCR4_EDF_ARGB8888,
 	}, {
 		.fourcc = DRM_FORMAT_UYVY,
+		.v4l2 = V4L2_PIX_FMT_UYVY,
 		.bpp = 16,
 		.planes = 1,
 		.pnmr = PnMR_SPIM_TP_OFF | PnMR_DDDF_YC,
 		.edf = PnDDCR4_EDF_NONE,
 	}, {
 		.fourcc = DRM_FORMAT_YUYV,
+		.v4l2 = V4L2_PIX_FMT_YUYV,
 		.bpp = 16,
 		.planes = 1,
 		.pnmr = PnMR_SPIM_TP_OFF | PnMR_DDDF_YC,
 		.edf = PnDDCR4_EDF_NONE,
 	}, {
 		.fourcc = DRM_FORMAT_NV12,
+		.v4l2 = V4L2_PIX_FMT_NV12M,
 		.bpp = 12,
 		.planes = 2,
 		.pnmr = PnMR_SPIM_TP_OFF | PnMR_DDDF_YC,
 		.edf = PnDDCR4_EDF_NONE,
 	}, {
 		.fourcc = DRM_FORMAT_NV21,
+		.v4l2 = V4L2_PIX_FMT_NV21M,
 		.bpp = 12,
 		.planes = 2,
 		.pnmr = PnMR_SPIM_TP_OFF | PnMR_DDDF_YC,
 		.edf = PnDDCR4_EDF_NONE,
 	}, {
 		.fourcc = DRM_FORMAT_NV16,
+		.v4l2 = V4L2_PIX_FMT_NV16M,
 		.bpp = 16,
 		.planes = 2,
 		.pnmr = PnMR_SPIM_TP_OFF | PnMR_DDDF_YC,
@@ -99,62 +110,77 @@ static const struct rcar_du_format_info rcar_du_format_infos[] = {
 	 */
 	{
 		.fourcc = DRM_FORMAT_RGB332,
+		.v4l2 = V4L2_PIX_FMT_RGB332,
 		.bpp = 8,
 		.planes = 1,
 	}, {
 		.fourcc = DRM_FORMAT_ARGB4444,
+		.v4l2 = V4L2_PIX_FMT_ARGB444,
 		.bpp = 16,
 		.planes = 1,
 	}, {
 		.fourcc = DRM_FORMAT_XRGB4444,
+		.v4l2 = V4L2_PIX_FMT_XRGB444,
 		.bpp = 16,
 		.planes = 1,
 	}, {
 		.fourcc = DRM_FORMAT_BGR888,
+		.v4l2 = V4L2_PIX_FMT_RGB24,
 		.bpp = 24,
 		.planes = 1,
 	}, {
 		.fourcc = DRM_FORMAT_RGB888,
+		.v4l2 = V4L2_PIX_FMT_BGR24,
 		.bpp = 24,
 		.planes = 1,
 	}, {
 		.fourcc = DRM_FORMAT_BGRA8888,
+		.v4l2 = V4L2_PIX_FMT_ARGB32,
 		.bpp = 32,
 		.planes = 1,
 	}, {
 		.fourcc = DRM_FORMAT_BGRX8888,
+		.v4l2 = V4L2_PIX_FMT_XRGB32,
 		.bpp = 32,
 		.planes = 1,
 	}, {
 		.fourcc = DRM_FORMAT_YVYU,
+		.v4l2 = V4L2_PIX_FMT_YVYU,
 		.bpp = 16,
 		.planes = 1,
 	}, {
 		.fourcc = DRM_FORMAT_NV61,
+		.v4l2 = V4L2_PIX_FMT_NV61M,
 		.bpp = 16,
 		.planes = 2,
 	}, {
 		.fourcc = DRM_FORMAT_YUV420,
+		.v4l2 = V4L2_PIX_FMT_YUV420M,
 		.bpp = 12,
 		.planes = 3,
 	}, {
 		.fourcc = DRM_FORMAT_YVU420,
+		.v4l2 = V4L2_PIX_FMT_YVU420M,
 		.bpp = 12,
 		.planes = 3,
 	}, {
 		.fourcc = DRM_FORMAT_YUV422,
+		.v4l2 = V4L2_PIX_FMT_YUV422M,
 		.bpp = 16,
 		.planes = 3,
 	}, {
 		.fourcc = DRM_FORMAT_YVU422,
+		.v4l2 = V4L2_PIX_FMT_YVU422M,
 		.bpp = 16,
 		.planes = 3,
 	}, {
 		.fourcc = DRM_FORMAT_YUV444,
+		.v4l2 = V4L2_PIX_FMT_YUV444M,
 		.bpp = 24,
 		.planes = 3,
 	}, {
 		.fourcc = DRM_FORMAT_YVU444,
+		.v4l2 = V4L2_PIX_FMT_YVU444M,
 		.bpp = 24,
 		.planes = 3,
 	},
@@ -639,6 +665,17 @@ int rcar_du_modeset_init(struct rcar_du_device *rcdu)
 		encoder->possible_clones = (1 << num_encoders) - 1;
 	}
 
+	/* Create the writeback connectors. */
+	if (rcdu->info->gen >= 3) {
+		for (i = 0; i < rcdu->num_crtcs; ++i) {
+			struct rcar_du_crtc *rcrtc = &rcdu->crtcs[i];
+
+			ret = rcar_du_writeback_init(rcdu, rcrtc);
+			if (ret < 0)
+				return ret;
+		}
+	}
+
 	/*
 	 * Initialize the default DPAD0 source to the index of the first DU
 	 * channel that can be connected to DPAD0. The exact value doesn't
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_kms.h b/drivers/gpu/drm/rcar-du/rcar_du_kms.h
index e171527abdaa..0346504d8c59 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_kms.h
+++ b/drivers/gpu/drm/rcar-du/rcar_du_kms.h
@@ -19,6 +19,7 @@ struct rcar_du_device;
 
 struct rcar_du_format_info {
 	u32 fourcc;
+	u32 v4l2;
 	unsigned int bpp;
 	unsigned int planes;
 	unsigned int pnmr;
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_vsp.c b/drivers/gpu/drm/rcar-du/rcar_du_vsp.c
index 0878accbd134..5e4faf258c31 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_vsp.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_vsp.c
@@ -10,6 +10,7 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_fb_cma_helper.h>
+#include <drm/drm_fourcc.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_plane_helper.h>
@@ -26,16 +27,19 @@
 #include "rcar_du_drv.h"
 #include "rcar_du_kms.h"
 #include "rcar_du_vsp.h"
+#include "rcar_du_writeback.h"
 
-static void rcar_du_vsp_complete(void *private, bool completed, u32 crc)
+static void rcar_du_vsp_complete(void *private, unsigned int status, u32 crc)
 {
 	struct rcar_du_crtc *crtc = private;
 
 	if (crtc->vblank_enable)
 		drm_crtc_handle_vblank(&crtc->crtc);
 
-	if (completed)
+	if (status & VSP1_DU_STATUS_COMPLETE)
 		rcar_du_crtc_finish_page_flip(crtc);
+	if (status & VSP1_DU_STATUS_WRITEBACK)
+		rcar_du_writeback_complete(crtc);
 
 	drm_crtc_add_crc_entry(&crtc->crtc, false, 0, &crc);
 }
@@ -43,7 +47,7 @@ static void rcar_du_vsp_complete(void *private, bool completed, u32 crc)
 void rcar_du_vsp_enable(struct rcar_du_crtc *crtc)
 {
 	const struct drm_display_mode *mode = &crtc->crtc.state->adjusted_mode;
-	struct rcar_du_device *rcdu = crtc->group->dev;
+	struct rcar_du_device *rcdu = crtc->dev;
 	struct vsp1_du_lif_config cfg = {
 		.width = mode->hdisplay,
 		.height = mode->vdisplay,
@@ -107,11 +111,12 @@ void rcar_du_vsp_atomic_flush(struct rcar_du_crtc *crtc)
 	state = to_rcar_crtc_state(crtc->crtc.state);
 	cfg.crc = state->crc;
 
+	rcar_du_writeback_setup(crtc, &cfg.writeback);
+
 	vsp1_du_atomic_flush(crtc->vsp->vsp, crtc->vsp_pipe, &cfg);
 }
 
-/* Keep the two tables in sync. */
-static const u32 formats_kms[] = {
+static const u32 rcar_du_vsp_formats[] = {
 	DRM_FORMAT_RGB332,
 	DRM_FORMAT_ARGB4444,
 	DRM_FORMAT_XRGB4444,
@@ -139,40 +144,13 @@ static const u32 formats_kms[] = {
 	DRM_FORMAT_YVU444,
 };
 
-static const u32 formats_v4l2[] = {
-	V4L2_PIX_FMT_RGB332,
-	V4L2_PIX_FMT_ARGB444,
-	V4L2_PIX_FMT_XRGB444,
-	V4L2_PIX_FMT_ARGB555,
-	V4L2_PIX_FMT_XRGB555,
-	V4L2_PIX_FMT_RGB565,
-	V4L2_PIX_FMT_RGB24,
-	V4L2_PIX_FMT_BGR24,
-	V4L2_PIX_FMT_ARGB32,
-	V4L2_PIX_FMT_XRGB32,
-	V4L2_PIX_FMT_ABGR32,
-	V4L2_PIX_FMT_XBGR32,
-	V4L2_PIX_FMT_UYVY,
-	V4L2_PIX_FMT_YUYV,
-	V4L2_PIX_FMT_YVYU,
-	V4L2_PIX_FMT_NV12M,
-	V4L2_PIX_FMT_NV21M,
-	V4L2_PIX_FMT_NV16M,
-	V4L2_PIX_FMT_NV61M,
-	V4L2_PIX_FMT_YUV420M,
-	V4L2_PIX_FMT_YVU420M,
-	V4L2_PIX_FMT_YUV422M,
-	V4L2_PIX_FMT_YVU422M,
-	V4L2_PIX_FMT_YUV444M,
-	V4L2_PIX_FMT_YVU444M,
-};
-
 static void rcar_du_vsp_plane_setup(struct rcar_du_vsp_plane *plane)
 {
 	struct rcar_du_vsp_plane_state *state =
 		to_rcar_vsp_plane_state(plane->plane.state);
 	struct rcar_du_crtc *crtc = to_rcar_crtc(state->state.crtc);
 	struct drm_framebuffer *fb = plane->plane.state->fb;
+	const struct rcar_du_format_info *format;
 	struct vsp1_du_atomic_config cfg = {
 		.pixelformat = 0,
 		.pitch = fb->pitches[0],
@@ -195,37 +173,23 @@ static void rcar_du_vsp_plane_setup(struct rcar_du_vsp_plane *plane)
 		cfg.mem[i] = sg_dma_address(state->sg_tables[i].sgl)
 			   + fb->offsets[i];
 
-	for (i = 0; i < ARRAY_SIZE(formats_kms); ++i) {
-		if (formats_kms[i] == state->format->fourcc) {
-			cfg.pixelformat = formats_v4l2[i];
-			break;
-		}
-	}
+	format = rcar_du_format_info(state->format->fourcc);
+	cfg.pixelformat = format->v4l2;
 
 	vsp1_du_atomic_update(plane->vsp->vsp, crtc->vsp_pipe,
 			      plane->index, &cfg);
 }
 
-static int rcar_du_vsp_plane_prepare_fb(struct drm_plane *plane,
-					struct drm_plane_state *state)
+int rcar_du_vsp_map_fb(struct rcar_du_vsp *vsp, struct drm_framebuffer *fb,
+		       struct sg_table sg_tables[3])
 {
-	struct rcar_du_vsp_plane_state *rstate = to_rcar_vsp_plane_state(state);
-	struct rcar_du_vsp *vsp = to_rcar_vsp_plane(plane)->vsp;
 	struct rcar_du_device *rcdu = vsp->dev;
 	unsigned int i;
 	int ret;
 
-	/*
-	 * There's no need to prepare (and unprepare) the framebuffer when the
-	 * plane is not visible, as it will not be displayed.
-	 */
-	if (!state->visible)
-		return 0;
-
-	for (i = 0; i < rstate->format->planes; ++i) {
-		struct drm_gem_cma_object *gem =
-			drm_fb_cma_get_gem_obj(state->fb, i);
-		struct sg_table *sgt = &rstate->sg_tables[i];
+	for (i = 0; i < fb->format->num_planes; ++i) {
+		struct drm_gem_cma_object *gem = drm_fb_cma_get_gem_obj(fb, i);
+		struct sg_table *sgt = &sg_tables[i];
 
 		ret = dma_get_sgtable(rcdu->dev, sgt, gem->vaddr, gem->paddr,
 				      gem->base.size);
@@ -240,15 +204,11 @@ static int rcar_du_vsp_plane_prepare_fb(struct drm_plane *plane,
 		}
 	}
 
-	ret = drm_gem_fb_prepare_fb(plane, state);
-	if (ret)
-		goto fail;
-
 	return 0;
 
 fail:
 	while (i--) {
-		struct sg_table *sgt = &rstate->sg_tables[i];
+		struct sg_table *sgt = &sg_tables[i];
 
 		vsp1_du_unmap_sg(vsp->vsp, sgt);
 		sg_free_table(sgt);
@@ -257,24 +217,52 @@ fail:
 	return ret;
 }
 
-static void rcar_du_vsp_plane_cleanup_fb(struct drm_plane *plane,
-					 struct drm_plane_state *state)
+static int rcar_du_vsp_plane_prepare_fb(struct drm_plane *plane,
+					struct drm_plane_state *state)
 {
 	struct rcar_du_vsp_plane_state *rstate = to_rcar_vsp_plane_state(state);
 	struct rcar_du_vsp *vsp = to_rcar_vsp_plane(plane)->vsp;
-	unsigned int i;
+	int ret;
 
+	/*
+	 * There's no need to prepare (and unprepare) the framebuffer when the
+	 * plane is not visible, as it will not be displayed.
+	 */
 	if (!state->visible)
-		return;
+		return 0;
+
+	ret = rcar_du_vsp_map_fb(vsp, state->fb, rstate->sg_tables);
+	if (ret < 0)
+		return ret;
+
+	return drm_gem_fb_prepare_fb(plane, state);
+}
+
+void rcar_du_vsp_unmap_fb(struct rcar_du_vsp *vsp, struct drm_framebuffer *fb,
+			  struct sg_table sg_tables[3])
+{
+	unsigned int i;
 
-	for (i = 0; i < rstate->format->planes; ++i) {
-		struct sg_table *sgt = &rstate->sg_tables[i];
+	for (i = 0; i < fb->format->num_planes; ++i) {
+		struct sg_table *sgt = &sg_tables[i];
 
 		vsp1_du_unmap_sg(vsp->vsp, sgt);
 		sg_free_table(sgt);
 	}
 }
 
+static void rcar_du_vsp_plane_cleanup_fb(struct drm_plane *plane,
+					 struct drm_plane_state *state)
+{
+	struct rcar_du_vsp_plane_state *rstate = to_rcar_vsp_plane_state(state);
+	struct rcar_du_vsp *vsp = to_rcar_vsp_plane(plane)->vsp;
+
+	if (!state->visible)
+		return;
+
+	rcar_du_vsp_unmap_fb(vsp, state->fb, rstate->sg_tables);
+}
+
 static int rcar_du_vsp_plane_atomic_check(struct drm_plane *plane,
 					  struct drm_plane_state *state)
 {
@@ -395,8 +383,8 @@ int rcar_du_vsp_init(struct rcar_du_vsp *vsp, struct device_node *np,
 
 		ret = drm_universal_plane_init(rcdu->ddev, &plane->plane, crtcs,
 					       &rcar_du_vsp_plane_funcs,
-					       formats_kms,
-					       ARRAY_SIZE(formats_kms),
+					       rcar_du_vsp_formats,
+					       ARRAY_SIZE(rcar_du_vsp_formats),
 					       NULL, type, NULL);
 		if (ret < 0)
 			return ret;
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_vsp.h b/drivers/gpu/drm/rcar-du/rcar_du_vsp.h
index db232037f24a..9b4724159378 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_vsp.h
+++ b/drivers/gpu/drm/rcar-du/rcar_du_vsp.h
@@ -12,8 +12,10 @@
 
 #include <drm/drm_plane.h>
 
+struct drm_framebuffer;
 struct rcar_du_format_info;
 struct rcar_du_vsp;
+struct sg_table;
 
 struct rcar_du_vsp_plane {
 	struct drm_plane plane;
@@ -60,6 +62,10 @@ void rcar_du_vsp_enable(struct rcar_du_crtc *crtc);
 void rcar_du_vsp_disable(struct rcar_du_crtc *crtc);
 void rcar_du_vsp_atomic_begin(struct rcar_du_crtc *crtc);
 void rcar_du_vsp_atomic_flush(struct rcar_du_crtc *crtc);
+int rcar_du_vsp_map_fb(struct rcar_du_vsp *vsp, struct drm_framebuffer *fb,
+		       struct sg_table sg_tables[3]);
+void rcar_du_vsp_unmap_fb(struct rcar_du_vsp *vsp, struct drm_framebuffer *fb,
+			  struct sg_table sg_tables[3]);
 #else
 static inline int rcar_du_vsp_init(struct rcar_du_vsp *vsp,
 				   struct device_node *np,
@@ -71,6 +77,17 @@ static inline void rcar_du_vsp_enable(struct rcar_du_crtc *crtc) { };
 static inline void rcar_du_vsp_disable(struct rcar_du_crtc *crtc) { };
 static inline void rcar_du_vsp_atomic_begin(struct rcar_du_crtc *crtc) { };
 static inline void rcar_du_vsp_atomic_flush(struct rcar_du_crtc *crtc) { };
+static inline int rcar_du_vsp_map_fb(struct rcar_du_vsp *vsp,
+				     struct drm_framebuffer *fb,
+				     struct sg_table sg_tables[3])
+{
+	return -ENXIO;
+}
+static inline void rcar_du_vsp_unmap_fb(struct rcar_du_vsp *vsp,
+					struct drm_framebuffer *fb,
+					struct sg_table sg_tables[3])
+{
+}
 #endif
 
 #endif /* __RCAR_DU_VSP_H__ */
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_writeback.c b/drivers/gpu/drm/rcar-du/rcar_du_writeback.c
new file mode 100644
index 000000000000..989a0be94131
--- /dev/null
+++ b/drivers/gpu/drm/rcar-du/rcar_du_writeback.c
@@ -0,0 +1,243 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * rcar_du_writeback.c  --  R-Car Display Unit Writeback Support
+ *
+ * Copyright (C) 2019 Laurent Pinchart <laurent.pinchart@ideasonboard.com>
+ */
+
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_device.h>
+#include <drm/drm_fourcc.h>
+#include <drm/drm_probe_helper.h>
+#include <drm/drm_writeback.h>
+
+#include "rcar_du_crtc.h"
+#include "rcar_du_drv.h"
+#include "rcar_du_kms.h"
+
+/**
+ * struct rcar_du_wb_conn_state - Driver-specific writeback connector state
+ * @state: base DRM connector state
+ * @format: format of the writeback framebuffer
+ */
+struct rcar_du_wb_conn_state {
+	struct drm_connector_state state;
+	const struct rcar_du_format_info *format;
+};
+
+#define to_rcar_wb_conn_state(s) \
+	container_of(s, struct rcar_du_wb_conn_state, state)
+
+/**
+ * struct rcar_du_wb_job - Driver-private data for writeback jobs
+ * @sg_tables: scatter-gather tables for the framebuffer memory
+ */
+struct rcar_du_wb_job {
+	struct sg_table sg_tables[3];
+};
+
+static int rcar_du_wb_conn_get_modes(struct drm_connector *connector)
+{
+	struct drm_device *dev = connector->dev;
+
+	return drm_add_modes_noedid(connector, dev->mode_config.max_width,
+				    dev->mode_config.max_height);
+}
+
+static int rcar_du_wb_prepare_job(struct drm_writeback_connector *connector,
+				  struct drm_writeback_job *job)
+{
+	struct rcar_du_crtc *rcrtc = wb_to_rcar_crtc(connector);
+	struct rcar_du_wb_job *rjob;
+	int ret;
+
+	if (!job->fb)
+		return 0;
+
+	rjob = kzalloc(sizeof(*rjob), GFP_KERNEL);
+	if (!rjob)
+		return -ENOMEM;
+
+	/* Map the framebuffer to the VSP. */
+	ret = rcar_du_vsp_map_fb(rcrtc->vsp, job->fb, rjob->sg_tables);
+	if (ret < 0) {
+		kfree(rjob);
+		return ret;
+	}
+
+	job->priv = rjob;
+	return 0;
+}
+
+static void rcar_du_wb_cleanup_job(struct drm_writeback_connector *connector,
+				   struct drm_writeback_job *job)
+{
+	struct rcar_du_crtc *rcrtc = wb_to_rcar_crtc(connector);
+	struct rcar_du_wb_job *rjob = job->priv;
+
+	if (!job->fb)
+		return;
+
+	rcar_du_vsp_unmap_fb(rcrtc->vsp, job->fb, rjob->sg_tables);
+	kfree(rjob);
+}
+
+static const struct drm_connector_helper_funcs rcar_du_wb_conn_helper_funcs = {
+	.get_modes = rcar_du_wb_conn_get_modes,
+	.prepare_writeback_job = rcar_du_wb_prepare_job,
+	.cleanup_writeback_job = rcar_du_wb_cleanup_job,
+};
+
+static struct drm_connector_state *
+rcar_du_wb_conn_duplicate_state(struct drm_connector *connector)
+{
+	struct rcar_du_wb_conn_state *copy;
+
+	if (WARN_ON(!connector->state))
+		return NULL;
+
+	copy = kzalloc(sizeof(*copy), GFP_KERNEL);
+	if (!copy)
+		return NULL;
+
+	__drm_atomic_helper_connector_duplicate_state(connector, &copy->state);
+
+	return &copy->state;
+}
+
+static void rcar_du_wb_conn_destroy_state(struct drm_connector *connector,
+					  struct drm_connector_state *state)
+{
+	__drm_atomic_helper_connector_destroy_state(state);
+	kfree(to_rcar_wb_conn_state(state));
+}
+
+static void rcar_du_wb_conn_reset(struct drm_connector *connector)
+{
+	struct rcar_du_wb_conn_state *state;
+
+	if (connector->state) {
+		rcar_du_wb_conn_destroy_state(connector, connector->state);
+		connector->state = NULL;
+	}
+
+	state = kzalloc(sizeof(*state), GFP_KERNEL);
+	if (state == NULL)
+		return;
+
+	__drm_atomic_helper_connector_reset(connector, &state->state);
+}
+
+static const struct drm_connector_funcs rcar_du_wb_conn_funcs = {
+	.reset = rcar_du_wb_conn_reset,
+	.fill_modes = drm_helper_probe_single_connector_modes,
+	.destroy = drm_connector_cleanup,
+	.atomic_duplicate_state = rcar_du_wb_conn_duplicate_state,
+	.atomic_destroy_state = rcar_du_wb_conn_destroy_state,
+};
+
+static int rcar_du_wb_enc_atomic_check(struct drm_encoder *encoder,
+				       struct drm_crtc_state *crtc_state,
+				       struct drm_connector_state *conn_state)
+{
+	struct rcar_du_wb_conn_state *wb_state =
+		to_rcar_wb_conn_state(conn_state);
+	const struct drm_display_mode *mode = &crtc_state->mode;
+	struct drm_device *dev = encoder->dev;
+	struct drm_framebuffer *fb;
+
+	if (!conn_state->writeback_job || !conn_state->writeback_job->fb)
+		return 0;
+
+	fb = conn_state->writeback_job->fb;
+
+	/*
+	 * Verify that the framebuffer format is supported and that its size
+	 * matches the current mode.
+	 */
+	if (fb->width != mode->hdisplay || fb->height != mode->vdisplay) {
+		dev_dbg(dev->dev, "%s: invalid framebuffer size %ux%u\n",
+			__func__, fb->width, fb->height);
+		return -EINVAL;
+	}
+
+	wb_state->format = rcar_du_format_info(fb->format->format);
+	if (wb_state->format == NULL) {
+		dev_dbg(dev->dev, "%s: unsupported format %08x\n", __func__,
+			fb->format->format);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static const struct drm_encoder_helper_funcs rcar_du_wb_enc_helper_funcs = {
+	.atomic_check = rcar_du_wb_enc_atomic_check,
+};
+
+/*
+ * Only RGB formats are currently supported as the VSP outputs RGB to the DU
+ * and can't convert to YUV separately for writeback.
+ */
+static const u32 writeback_formats[] = {
+	DRM_FORMAT_RGB332,
+	DRM_FORMAT_ARGB4444,
+	DRM_FORMAT_XRGB4444,
+	DRM_FORMAT_ARGB1555,
+	DRM_FORMAT_XRGB1555,
+	DRM_FORMAT_RGB565,
+	DRM_FORMAT_BGR888,
+	DRM_FORMAT_RGB888,
+	DRM_FORMAT_BGRA8888,
+	DRM_FORMAT_BGRX8888,
+	DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_XRGB8888,
+};
+
+int rcar_du_writeback_init(struct rcar_du_device *rcdu,
+			   struct rcar_du_crtc *rcrtc)
+{
+	struct drm_writeback_connector *wb_conn = &rcrtc->writeback;
+
+	wb_conn->encoder.possible_crtcs = 1 << drm_crtc_index(&rcrtc->crtc);
+	drm_connector_helper_add(&wb_conn->base,
+				 &rcar_du_wb_conn_helper_funcs);
+
+	return drm_writeback_connector_init(rcdu->ddev, wb_conn,
+					    &rcar_du_wb_conn_funcs,
+					    &rcar_du_wb_enc_helper_funcs,
+					    writeback_formats,
+					    ARRAY_SIZE(writeback_formats));
+}
+
+void rcar_du_writeback_setup(struct rcar_du_crtc *rcrtc,
+			     struct vsp1_du_writeback_config *cfg)
+{
+	struct rcar_du_wb_conn_state *wb_state;
+	struct drm_connector_state *state;
+	struct rcar_du_wb_job *rjob;
+	struct drm_framebuffer *fb;
+	unsigned int i;
+
+	state = rcrtc->writeback.base.state;
+	if (!state || !state->writeback_job || !state->writeback_job->fb)
+		return;
+
+	fb = state->writeback_job->fb;
+	rjob = state->writeback_job->priv;
+	wb_state = to_rcar_wb_conn_state(state);
+
+	cfg->pixelformat = wb_state->format->v4l2;
+	cfg->pitch = fb->pitches[0];
+
+	for (i = 0; i < wb_state->format->planes; ++i)
+		cfg->mem[i] = sg_dma_address(rjob->sg_tables[i].sgl)
+			    + fb->offsets[i];
+
+	drm_writeback_queue_job(&rcrtc->writeback, state);
+}
+
+void rcar_du_writeback_complete(struct rcar_du_crtc *rcrtc)
+{
+	drm_writeback_signal_completion(&rcrtc->writeback, 0);
+}
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_writeback.h b/drivers/gpu/drm/rcar-du/rcar_du_writeback.h
new file mode 100644
index 000000000000..fa87ebf8d21f
--- /dev/null
+++ b/drivers/gpu/drm/rcar-du/rcar_du_writeback.h
@@ -0,0 +1,39 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/*
+ * rcar_du_writeback.h  --  R-Car Display Unit Writeback Support
+ *
+ * Copyright (C) 2019 Laurent Pinchart <laurent.pinchart@ideasonboard.com>
+ */
+
+#ifndef __RCAR_DU_WRITEBACK_H__
+#define __RCAR_DU_WRITEBACK_H__
+
+#include <drm/drm_plane.h>
+
+struct rcar_du_crtc;
+struct rcar_du_device;
+struct vsp1_du_atomic_pipe_config;
+
+#ifdef CONFIG_DRM_RCAR_WRITEBACK
+int rcar_du_writeback_init(struct rcar_du_device *rcdu,
+			   struct rcar_du_crtc *rcrtc);
+void rcar_du_writeback_setup(struct rcar_du_crtc *rcrtc,
+			     struct vsp1_du_writeback_config *cfg);
+void rcar_du_writeback_complete(struct rcar_du_crtc *rcrtc);
+#else
+static inline int rcar_du_writeback_init(struct rcar_du_device *rcdu,
+					 struct rcar_du_crtc *rcrtc)
+{
+	return -ENXIO;
+}
+static inline void
+rcar_du_writeback_setup(struct rcar_du_crtc *rcrtc,
+			struct vsp1_du_writeback_config *cfg)
+{
+}
+static inline void rcar_du_writeback_complete(struct rcar_du_crtc *rcrtc)
+{
+}
+#endif
+
+#endif /* __RCAR_DU_WRITEBACK_H__ */
diff --git a/drivers/gpu/drm/rcar-du/rcar_lvds.c b/drivers/gpu/drm/rcar-du/rcar_lvds.c
index 7ef97b2a6eda..620b51aab291 100644
--- a/drivers/gpu/drm/rcar-du/rcar_lvds.c
+++ b/drivers/gpu/drm/rcar-du/rcar_lvds.c
@@ -283,7 +283,7 @@ static void rcar_lvds_d3_e3_pll_calc(struct rcar_lvds *lvds, struct clk *clk,
 				 * divider.
 				 */
 				fout = fvco / (1 << e) / div7;
-				div = DIV_ROUND_CLOSEST(fout, target);
+				div = max(1UL, DIV_ROUND_CLOSEST(fout, target));
 				diff = abs(fout / div - target);
 
 				if (diff < pll->diff) {
@@ -485,9 +485,13 @@ static void rcar_lvds_enable(struct drm_bridge *bridge)
 	}
 
 	if (lvds->info->quirks & RCAR_LVDS_QUIRK_GEN3_LVEN) {
-		/* Turn on the LVDS PHY. */
+		/*
+		 * Turn on the LVDS PHY. On D3, the LVEN and LVRES bit must be
+		 * set at the same time, so don't write the register yet.
+		 */
 		lvdcr0 |= LVDCR0_LVEN;
-		rcar_lvds_write(lvds, LVDCR0, lvdcr0);
+		if (!(lvds->info->quirks & RCAR_LVDS_QUIRK_PWD))
+			rcar_lvds_write(lvds, LVDCR0, lvdcr0);
 	}
 
 	if (!(lvds->info->quirks & RCAR_LVDS_QUIRK_EXT_PLL)) {
@@ -531,11 +535,16 @@ static bool rcar_lvds_mode_fixup(struct drm_bridge *bridge,
 				 const struct drm_display_mode *mode,
 				 struct drm_display_mode *adjusted_mode)
 {
+	struct rcar_lvds *lvds = bridge_to_rcar_lvds(bridge);
+	int min_freq;
+
 	/*
 	 * The internal LVDS encoder has a restricted clock frequency operating
-	 * range (31MHz to 148.5MHz). Clamp the clock accordingly.
+	 * range, from 5MHz to 148.5MHz on D3 and E3, and from 31MHz to
+	 * 148.5MHz on all other platforms. Clamp the clock accordingly.
 	 */
-	adjusted_mode->clock = clamp(adjusted_mode->clock, 31000, 148500);
+	min_freq = lvds->info->quirks & RCAR_LVDS_QUIRK_EXT_PLL ? 5000 : 31000;
+	adjusted_mode->clock = clamp(adjusted_mode->clock, min_freq, 148500);
 
 	return true;
 }
diff --git a/drivers/gpu/drm/rockchip/Kconfig b/drivers/gpu/drm/rockchip/Kconfig
index 1e75196f9659..2cdf3b62d559 100644
--- a/drivers/gpu/drm/rockchip/Kconfig
+++ b/drivers/gpu/drm/rockchip/Kconfig
@@ -77,4 +77,12 @@ config ROCKCHIP_RGB
 	  Some Rockchip CRTCs, like rv1108, can directly output parallel
 	  and serial RGB format to panel or connect to a conversion chip.
 	  say Y to enable its driver.
+
+config ROCKCHIP_RK3066_HDMI
+	bool "Rockchip specific extensions for RK3066 HDMI"
+	depends on DRM_ROCKCHIP
+	help
+	  This selects support for Rockchip SoC specific extensions
+	  for the RK3066 HDMI driver. If you want to enable
+	  HDMI on RK3066 based SoC, you should select this option.
 endif
diff --git a/drivers/gpu/drm/rockchip/Makefile b/drivers/gpu/drm/rockchip/Makefile
index f6fc9d5dd0ad..524684ba7f6a 100644
--- a/drivers/gpu/drm/rockchip/Makefile
+++ b/drivers/gpu/drm/rockchip/Makefile
@@ -15,5 +15,6 @@ rockchipdrm-$(CONFIG_ROCKCHIP_DW_MIPI_DSI) += dw-mipi-dsi-rockchip.o
 rockchipdrm-$(CONFIG_ROCKCHIP_INNO_HDMI) += inno_hdmi.o
 rockchipdrm-$(CONFIG_ROCKCHIP_LVDS) += rockchip_lvds.o
 rockchipdrm-$(CONFIG_ROCKCHIP_RGB) += rockchip_rgb.o
+rockchipdrm-$(CONFIG_ROCKCHIP_RK3066_HDMI) += rk3066_hdmi.o
 
 obj-$(CONFIG_DRM_ROCKCHIP) += rockchipdrm.o
diff --git a/drivers/gpu/drm/rockchip/rk3066_hdmi.c b/drivers/gpu/drm/rockchip/rk3066_hdmi.c
new file mode 100644
index 000000000000..85fc5f01f761
--- /dev/null
+++ b/drivers/gpu/drm/rockchip/rk3066_hdmi.c
@@ -0,0 +1,876 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) Fuzhou Rockchip Electronics Co.Ltd
+ *    Zheng Yang <zhengyang@rock-chips.com>
+ */
+
+#include <drm/drm_of.h>
+#include <drm/drm_probe_helper.h>
+
+#include <linux/clk.h>
+#include <linux/mfd/syscon.h>
+#include <linux/platform_device.h>
+#include <linux/regmap.h>
+
+#include "rk3066_hdmi.h"
+
+#include "rockchip_drm_drv.h"
+#include "rockchip_drm_vop.h"
+
+#define DEFAULT_PLLA_RATE 30000000
+
+struct hdmi_data_info {
+	int vic; /* The CEA Video ID (VIC) of the current drm display mode. */
+	bool sink_is_hdmi;
+	unsigned int enc_out_format;
+	unsigned int colorimetry;
+};
+
+struct rk3066_hdmi_i2c {
+	struct i2c_adapter adap;
+
+	u8 ddc_addr;
+	u8 segment_addr;
+	u8 stat;
+
+	struct mutex i2c_lock; /* For i2c operation. */
+	struct completion cmpltn;
+};
+
+struct rk3066_hdmi {
+	struct device *dev;
+	struct drm_device *drm_dev;
+	struct regmap *grf_regmap;
+	int irq;
+	struct clk *hclk;
+	void __iomem *regs;
+
+	struct drm_connector connector;
+	struct drm_encoder encoder;
+
+	struct rk3066_hdmi_i2c *i2c;
+	struct i2c_adapter *ddc;
+
+	unsigned int tmdsclk;
+
+	struct hdmi_data_info hdmi_data;
+	struct drm_display_mode previous_mode;
+};
+
+#define to_rk3066_hdmi(x) container_of(x, struct rk3066_hdmi, x)
+
+static inline u8 hdmi_readb(struct rk3066_hdmi *hdmi, u16 offset)
+{
+	return readl_relaxed(hdmi->regs + offset);
+}
+
+static inline void hdmi_writeb(struct rk3066_hdmi *hdmi, u16 offset, u32 val)
+{
+	writel_relaxed(val, hdmi->regs + offset);
+}
+
+static inline void hdmi_modb(struct rk3066_hdmi *hdmi, u16 offset,
+			     u32 msk, u32 val)
+{
+	u8 temp = hdmi_readb(hdmi, offset) & ~msk;
+
+	temp |= val & msk;
+	hdmi_writeb(hdmi, offset, temp);
+}
+
+static void rk3066_hdmi_i2c_init(struct rk3066_hdmi *hdmi)
+{
+	int ddc_bus_freq;
+
+	ddc_bus_freq = (hdmi->tmdsclk >> 2) / HDMI_SCL_RATE;
+
+	hdmi_writeb(hdmi, HDMI_DDC_BUS_FREQ_L, ddc_bus_freq & 0xFF);
+	hdmi_writeb(hdmi, HDMI_DDC_BUS_FREQ_H, (ddc_bus_freq >> 8) & 0xFF);
+
+	/* Clear the EDID interrupt flag and mute the interrupt. */
+	hdmi_modb(hdmi, HDMI_INTR_MASK1, HDMI_INTR_EDID_MASK, 0);
+	hdmi_writeb(hdmi, HDMI_INTR_STATUS1, HDMI_INTR_EDID_MASK);
+}
+
+static inline u8 rk3066_hdmi_get_power_mode(struct rk3066_hdmi *hdmi)
+{
+	return hdmi_readb(hdmi, HDMI_SYS_CTRL) & HDMI_SYS_POWER_MODE_MASK;
+}
+
+static void rk3066_hdmi_set_power_mode(struct rk3066_hdmi *hdmi, int mode)
+{
+	u8 current_mode, next_mode;
+	u8 i = 0;
+
+	current_mode = rk3066_hdmi_get_power_mode(hdmi);
+
+	DRM_DEV_DEBUG(hdmi->dev, "mode         :%d\n", mode);
+	DRM_DEV_DEBUG(hdmi->dev, "current_mode :%d\n", current_mode);
+
+	if (current_mode == mode)
+		return;
+
+	do {
+		if (current_mode > mode) {
+			next_mode = current_mode / 2;
+		} else {
+			if (current_mode < HDMI_SYS_POWER_MODE_A)
+				next_mode = HDMI_SYS_POWER_MODE_A;
+			else
+				next_mode = current_mode * 2;
+		}
+
+		DRM_DEV_DEBUG(hdmi->dev, "%d: next_mode :%d\n", i, next_mode);
+
+		if (next_mode != HDMI_SYS_POWER_MODE_D) {
+			hdmi_modb(hdmi, HDMI_SYS_CTRL,
+				  HDMI_SYS_POWER_MODE_MASK, next_mode);
+		} else {
+			hdmi_writeb(hdmi, HDMI_SYS_CTRL,
+				    HDMI_SYS_POWER_MODE_D |
+				    HDMI_SYS_PLL_RESET_MASK);
+			usleep_range(90, 100);
+			hdmi_writeb(hdmi, HDMI_SYS_CTRL,
+				    HDMI_SYS_POWER_MODE_D |
+				    HDMI_SYS_PLLB_RESET);
+			usleep_range(90, 100);
+			hdmi_writeb(hdmi, HDMI_SYS_CTRL,
+				    HDMI_SYS_POWER_MODE_D);
+		}
+		current_mode = next_mode;
+		i = i + 1;
+	} while ((next_mode != mode) && (i < 5));
+
+	/*
+	 * When the IP controller isn't configured with accurate video timing,
+	 * DDC_CLK should be equal to the PLLA frequency, which is 30MHz,
+	 * so we need to init the TMDS rate to the PCLK rate and reconfigure
+	 * the DDC clock.
+	 */
+	if (mode < HDMI_SYS_POWER_MODE_D)
+		hdmi->tmdsclk = DEFAULT_PLLA_RATE;
+}
+
+static int
+rk3066_hdmi_upload_frame(struct rk3066_hdmi *hdmi, int setup_rc,
+			 union hdmi_infoframe *frame, u32 frame_index,
+			 u32 mask, u32 disable, u32 enable)
+{
+	if (mask)
+		hdmi_modb(hdmi, HDMI_CP_AUTO_SEND_CTRL, mask, disable);
+
+	hdmi_writeb(hdmi, HDMI_CP_BUF_INDEX, frame_index);
+
+	if (setup_rc >= 0) {
+		u8 packed_frame[HDMI_MAXIMUM_INFO_FRAME_SIZE];
+		ssize_t rc, i;
+
+		rc = hdmi_infoframe_pack(frame, packed_frame,
+					 sizeof(packed_frame));
+		if (rc < 0)
+			return rc;
+
+		for (i = 0; i < rc; i++)
+			hdmi_writeb(hdmi, HDMI_CP_BUF_ACC_HB0 + i * 4,
+				    packed_frame[i]);
+
+		if (mask)
+			hdmi_modb(hdmi, HDMI_CP_AUTO_SEND_CTRL, mask, enable);
+	}
+
+	return setup_rc;
+}
+
+static int rk3066_hdmi_config_avi(struct rk3066_hdmi *hdmi,
+				  struct drm_display_mode *mode)
+{
+	union hdmi_infoframe frame;
+	int rc;
+
+	rc = drm_hdmi_avi_infoframe_from_display_mode(&frame.avi,
+						      &hdmi->connector, mode);
+
+	if (hdmi->hdmi_data.enc_out_format == HDMI_COLORSPACE_YUV444)
+		frame.avi.colorspace = HDMI_COLORSPACE_YUV444;
+	else if (hdmi->hdmi_data.enc_out_format == HDMI_COLORSPACE_YUV422)
+		frame.avi.colorspace = HDMI_COLORSPACE_YUV422;
+	else
+		frame.avi.colorspace = HDMI_COLORSPACE_RGB;
+
+	frame.avi.colorimetry = hdmi->hdmi_data.colorimetry;
+	frame.avi.scan_mode = HDMI_SCAN_MODE_NONE;
+
+	return rk3066_hdmi_upload_frame(hdmi, rc, &frame,
+					HDMI_INFOFRAME_AVI, 0, 0, 0);
+}
+
+static int rk3066_hdmi_config_video_timing(struct rk3066_hdmi *hdmi,
+					   struct drm_display_mode *mode)
+{
+	int value, vsync_offset;
+
+	/* Set the details for the external polarity and interlace mode. */
+	value = HDMI_EXT_VIDEO_SET_EN;
+	value |= mode->flags & DRM_MODE_FLAG_PHSYNC ?
+		 HDMI_VIDEO_HSYNC_ACTIVE_HIGH : HDMI_VIDEO_HSYNC_ACTIVE_LOW;
+	value |= mode->flags & DRM_MODE_FLAG_PVSYNC ?
+		 HDMI_VIDEO_VSYNC_ACTIVE_HIGH : HDMI_VIDEO_VSYNC_ACTIVE_LOW;
+	value |= mode->flags & DRM_MODE_FLAG_INTERLACE ?
+		 HDMI_VIDEO_MODE_INTERLACE : HDMI_VIDEO_MODE_PROGRESSIVE;
+
+	if (hdmi->hdmi_data.vic == 2 || hdmi->hdmi_data.vic == 3)
+		vsync_offset = 6;
+	else
+		vsync_offset = 0;
+
+	value |= vsync_offset << HDMI_VIDEO_VSYNC_OFFSET_SHIFT;
+	hdmi_writeb(hdmi, HDMI_EXT_VIDEO_PARA, value);
+
+	/* Set the details for the external video timing. */
+	value = mode->htotal;
+	hdmi_writeb(hdmi, HDMI_EXT_HTOTAL_L, value & 0xFF);
+	hdmi_writeb(hdmi, HDMI_EXT_HTOTAL_H, (value >> 8) & 0xFF);
+
+	value = mode->htotal - mode->hdisplay;
+	hdmi_writeb(hdmi, HDMI_EXT_HBLANK_L, value & 0xFF);
+	hdmi_writeb(hdmi, HDMI_EXT_HBLANK_H, (value >> 8) & 0xFF);
+
+	value = mode->htotal - mode->hsync_start;
+	hdmi_writeb(hdmi, HDMI_EXT_HDELAY_L, value & 0xFF);
+	hdmi_writeb(hdmi, HDMI_EXT_HDELAY_H, (value >> 8) & 0xFF);
+
+	value = mode->hsync_end - mode->hsync_start;
+	hdmi_writeb(hdmi, HDMI_EXT_HDURATION_L, value & 0xFF);
+	hdmi_writeb(hdmi, HDMI_EXT_HDURATION_H, (value >> 8) & 0xFF);
+
+	value = mode->vtotal;
+	hdmi_writeb(hdmi, HDMI_EXT_VTOTAL_L, value & 0xFF);
+	hdmi_writeb(hdmi, HDMI_EXT_VTOTAL_H, (value >> 8) & 0xFF);
+
+	value = mode->vtotal - mode->vdisplay;
+	hdmi_writeb(hdmi, HDMI_EXT_VBLANK_L, value & 0xFF);
+
+	value = mode->vtotal - mode->vsync_start + vsync_offset;
+	hdmi_writeb(hdmi, HDMI_EXT_VDELAY, value & 0xFF);
+
+	value = mode->vsync_end - mode->vsync_start;
+	hdmi_writeb(hdmi, HDMI_EXT_VDURATION, value & 0xFF);
+
+	return 0;
+}
+
+static void
+rk3066_hdmi_phy_write(struct rk3066_hdmi *hdmi, u16 offset, u8 value)
+{
+	hdmi_writeb(hdmi, offset, value);
+	hdmi_modb(hdmi, HDMI_SYS_CTRL,
+		  HDMI_SYS_PLL_RESET_MASK, HDMI_SYS_PLL_RESET);
+	usleep_range(90, 100);
+	hdmi_modb(hdmi, HDMI_SYS_CTRL, HDMI_SYS_PLL_RESET_MASK, 0);
+	usleep_range(900, 1000);
+}
+
+static void rk3066_hdmi_config_phy(struct rk3066_hdmi *hdmi)
+{
+	/* TMDS uses the same frequency as dclk. */
+	hdmi_writeb(hdmi, HDMI_DEEP_COLOR_MODE, 0x22);
+
+	/*
+	 * The semi-public documentation does not describe the hdmi registers
+	 * used by the function rk3066_hdmi_phy_write(), so we keep using
+	 * these magic values for now.
+	 */
+	if (hdmi->tmdsclk > 100000000) {
+		rk3066_hdmi_phy_write(hdmi, 0x158, 0x0E);
+		rk3066_hdmi_phy_write(hdmi, 0x15c, 0x00);
+		rk3066_hdmi_phy_write(hdmi, 0x160, 0x60);
+		rk3066_hdmi_phy_write(hdmi, 0x164, 0x00);
+		rk3066_hdmi_phy_write(hdmi, 0x168, 0xDA);
+		rk3066_hdmi_phy_write(hdmi, 0x16c, 0xA1);
+		rk3066_hdmi_phy_write(hdmi, 0x170, 0x0e);
+		rk3066_hdmi_phy_write(hdmi, 0x174, 0x22);
+		rk3066_hdmi_phy_write(hdmi, 0x178, 0x00);
+	} else if (hdmi->tmdsclk > 50000000) {
+		rk3066_hdmi_phy_write(hdmi, 0x158, 0x06);
+		rk3066_hdmi_phy_write(hdmi, 0x15c, 0x00);
+		rk3066_hdmi_phy_write(hdmi, 0x160, 0x60);
+		rk3066_hdmi_phy_write(hdmi, 0x164, 0x00);
+		rk3066_hdmi_phy_write(hdmi, 0x168, 0xCA);
+		rk3066_hdmi_phy_write(hdmi, 0x16c, 0xA3);
+		rk3066_hdmi_phy_write(hdmi, 0x170, 0x0e);
+		rk3066_hdmi_phy_write(hdmi, 0x174, 0x20);
+		rk3066_hdmi_phy_write(hdmi, 0x178, 0x00);
+	} else {
+		rk3066_hdmi_phy_write(hdmi, 0x158, 0x02);
+		rk3066_hdmi_phy_write(hdmi, 0x15c, 0x00);
+		rk3066_hdmi_phy_write(hdmi, 0x160, 0x60);
+		rk3066_hdmi_phy_write(hdmi, 0x164, 0x00);
+		rk3066_hdmi_phy_write(hdmi, 0x168, 0xC2);
+		rk3066_hdmi_phy_write(hdmi, 0x16c, 0xA2);
+		rk3066_hdmi_phy_write(hdmi, 0x170, 0x0e);
+		rk3066_hdmi_phy_write(hdmi, 0x174, 0x20);
+		rk3066_hdmi_phy_write(hdmi, 0x178, 0x00);
+	}
+}
+
+static int rk3066_hdmi_setup(struct rk3066_hdmi *hdmi,
+			     struct drm_display_mode *mode)
+{
+	hdmi->hdmi_data.vic = drm_match_cea_mode(mode);
+	hdmi->hdmi_data.enc_out_format = HDMI_COLORSPACE_RGB;
+
+	if (hdmi->hdmi_data.vic == 6 || hdmi->hdmi_data.vic == 7 ||
+	    hdmi->hdmi_data.vic == 21 || hdmi->hdmi_data.vic == 22 ||
+	    hdmi->hdmi_data.vic == 2 || hdmi->hdmi_data.vic == 3 ||
+	    hdmi->hdmi_data.vic == 17 || hdmi->hdmi_data.vic == 18)
+		hdmi->hdmi_data.colorimetry = HDMI_COLORIMETRY_ITU_601;
+	else
+		hdmi->hdmi_data.colorimetry = HDMI_COLORIMETRY_ITU_709;
+
+	hdmi->tmdsclk = mode->clock * 1000;
+
+	/* Mute video and audio output. */
+	hdmi_modb(hdmi, HDMI_VIDEO_CTRL2, HDMI_VIDEO_AUDIO_DISABLE_MASK,
+		  HDMI_AUDIO_DISABLE | HDMI_VIDEO_DISABLE);
+
+	/* Set power state to mode B. */
+	if (rk3066_hdmi_get_power_mode(hdmi) != HDMI_SYS_POWER_MODE_B)
+		rk3066_hdmi_set_power_mode(hdmi, HDMI_SYS_POWER_MODE_B);
+
+	/* Input video mode is RGB 24 bit. Use external data enable signal. */
+	hdmi_modb(hdmi, HDMI_AV_CTRL1,
+		  HDMI_VIDEO_DE_MASK, HDMI_VIDEO_EXTERNAL_DE);
+	hdmi_writeb(hdmi, HDMI_VIDEO_CTRL1,
+		    HDMI_VIDEO_OUTPUT_RGB444 |
+		    HDMI_VIDEO_INPUT_DATA_DEPTH_8BIT |
+		    HDMI_VIDEO_INPUT_COLOR_RGB);
+	hdmi_writeb(hdmi, HDMI_DEEP_COLOR_MODE, 0x20);
+
+	rk3066_hdmi_config_video_timing(hdmi, mode);
+
+	if (hdmi->hdmi_data.sink_is_hdmi) {
+		hdmi_modb(hdmi, HDMI_HDCP_CTRL, HDMI_VIDEO_MODE_MASK,
+			  HDMI_VIDEO_MODE_HDMI);
+		rk3066_hdmi_config_avi(hdmi, mode);
+	} else {
+		hdmi_modb(hdmi, HDMI_HDCP_CTRL, HDMI_VIDEO_MODE_MASK, 0);
+	}
+
+	rk3066_hdmi_config_phy(hdmi);
+
+	rk3066_hdmi_set_power_mode(hdmi, HDMI_SYS_POWER_MODE_E);
+
+	/*
+	 * When the IP controller is configured with accurate video
+	 * timing, the TMDS clock source should be switched to
+	 * DCLK_LCDC, so we need to init the TMDS rate to the pixel mode
+	 * clock rate and reconfigure the DDC clock.
+	 */
+	rk3066_hdmi_i2c_init(hdmi);
+
+	/* Unmute video output. */
+	hdmi_modb(hdmi, HDMI_VIDEO_CTRL2,
+		  HDMI_VIDEO_AUDIO_DISABLE_MASK, HDMI_AUDIO_DISABLE);
+	return 0;
+}
+
+static void
+rk3066_hdmi_encoder_mode_set(struct drm_encoder *encoder,
+			     struct drm_display_mode *mode,
+			     struct drm_display_mode *adj_mode)
+{
+	struct rk3066_hdmi *hdmi = to_rk3066_hdmi(encoder);
+
+	/* Store the display mode for plugin/DPMS poweron events. */
+	memcpy(&hdmi->previous_mode, adj_mode, sizeof(hdmi->previous_mode));
+}
+
+static void rk3066_hdmi_encoder_enable(struct drm_encoder *encoder)
+{
+	struct rk3066_hdmi *hdmi = to_rk3066_hdmi(encoder);
+	int mux, val;
+
+	mux = drm_of_encoder_active_endpoint_id(hdmi->dev->of_node, encoder);
+	if (mux)
+		val = (HDMI_VIDEO_SEL << 16) | HDMI_VIDEO_SEL;
+	else
+		val = HDMI_VIDEO_SEL << 16;
+
+	regmap_write(hdmi->grf_regmap, GRF_SOC_CON0, val);
+
+	DRM_DEV_DEBUG(hdmi->dev, "hdmi encoder enable select: vop%s\n",
+		      (mux) ? "1" : "0");
+
+	rk3066_hdmi_setup(hdmi, &hdmi->previous_mode);
+}
+
+static void rk3066_hdmi_encoder_disable(struct drm_encoder *encoder)
+{
+	struct rk3066_hdmi *hdmi = to_rk3066_hdmi(encoder);
+
+	DRM_DEV_DEBUG(hdmi->dev, "hdmi encoder disable\n");
+
+	if (rk3066_hdmi_get_power_mode(hdmi) == HDMI_SYS_POWER_MODE_E) {
+		hdmi_writeb(hdmi, HDMI_VIDEO_CTRL2,
+			    HDMI_VIDEO_AUDIO_DISABLE_MASK);
+		hdmi_modb(hdmi, HDMI_VIDEO_CTRL2,
+			  HDMI_AUDIO_CP_LOGIC_RESET_MASK,
+			  HDMI_AUDIO_CP_LOGIC_RESET);
+		usleep_range(500, 510);
+	}
+	rk3066_hdmi_set_power_mode(hdmi, HDMI_SYS_POWER_MODE_A);
+}
+
+static bool
+rk3066_hdmi_encoder_mode_fixup(struct drm_encoder *encoder,
+			       const struct drm_display_mode *mode,
+			       struct drm_display_mode *adj_mode)
+{
+	return true;
+}
+
+static int
+rk3066_hdmi_encoder_atomic_check(struct drm_encoder *encoder,
+				 struct drm_crtc_state *crtc_state,
+				 struct drm_connector_state *conn_state)
+{
+	struct rockchip_crtc_state *s = to_rockchip_crtc_state(crtc_state);
+
+	s->output_mode = ROCKCHIP_OUT_MODE_P888;
+	s->output_type = DRM_MODE_CONNECTOR_HDMIA;
+
+	return 0;
+}
+
+static const
+struct drm_encoder_helper_funcs rk3066_hdmi_encoder_helper_funcs = {
+	.enable       = rk3066_hdmi_encoder_enable,
+	.disable      = rk3066_hdmi_encoder_disable,
+	.mode_fixup   = rk3066_hdmi_encoder_mode_fixup,
+	.mode_set     = rk3066_hdmi_encoder_mode_set,
+	.atomic_check = rk3066_hdmi_encoder_atomic_check,
+};
+
+static const struct drm_encoder_funcs rk3066_hdmi_encoder_funcs = {
+	.destroy = drm_encoder_cleanup,
+};
+
+static enum drm_connector_status
+rk3066_hdmi_connector_detect(struct drm_connector *connector, bool force)
+{
+	struct rk3066_hdmi *hdmi = to_rk3066_hdmi(connector);
+
+	return (hdmi_readb(hdmi, HDMI_HPG_MENS_STA) & HDMI_HPG_IN_STATUS_HIGH) ?
+		connector_status_connected : connector_status_disconnected;
+}
+
+static int rk3066_hdmi_connector_get_modes(struct drm_connector *connector)
+{
+	struct rk3066_hdmi *hdmi = to_rk3066_hdmi(connector);
+	struct edid *edid;
+	int ret = 0;
+
+	if (!hdmi->ddc)
+		return 0;
+
+	edid = drm_get_edid(connector, hdmi->ddc);
+	if (edid) {
+		hdmi->hdmi_data.sink_is_hdmi = drm_detect_hdmi_monitor(edid);
+		drm_connector_update_edid_property(connector, edid);
+		ret = drm_add_edid_modes(connector, edid);
+		kfree(edid);
+	}
+
+	return ret;
+}
+
+static enum drm_mode_status
+rk3066_hdmi_connector_mode_valid(struct drm_connector *connector,
+				 struct drm_display_mode *mode)
+{
+	u32 vic = drm_match_cea_mode(mode);
+
+	if (vic > 1)
+		return MODE_OK;
+	else
+		return MODE_BAD;
+}
+
+static struct drm_encoder *
+rk3066_hdmi_connector_best_encoder(struct drm_connector *connector)
+{
+	struct rk3066_hdmi *hdmi = to_rk3066_hdmi(connector);
+
+	return &hdmi->encoder;
+}
+
+static int
+rk3066_hdmi_probe_single_connector_modes(struct drm_connector *connector,
+					 uint32_t maxX, uint32_t maxY)
+{
+	if (maxX > 1920)
+		maxX = 1920;
+	if (maxY > 1080)
+		maxY = 1080;
+
+	return drm_helper_probe_single_connector_modes(connector, maxX, maxY);
+}
+
+static void rk3066_hdmi_connector_destroy(struct drm_connector *connector)
+{
+	drm_connector_unregister(connector);
+	drm_connector_cleanup(connector);
+}
+
+static const struct drm_connector_funcs rk3066_hdmi_connector_funcs = {
+	.fill_modes = rk3066_hdmi_probe_single_connector_modes,
+	.detect = rk3066_hdmi_connector_detect,
+	.destroy = rk3066_hdmi_connector_destroy,
+	.reset = drm_atomic_helper_connector_reset,
+	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
+};
+
+static const
+struct drm_connector_helper_funcs rk3066_hdmi_connector_helper_funcs = {
+	.get_modes = rk3066_hdmi_connector_get_modes,
+	.mode_valid = rk3066_hdmi_connector_mode_valid,
+	.best_encoder = rk3066_hdmi_connector_best_encoder,
+};
+
+static int
+rk3066_hdmi_register(struct drm_device *drm, struct rk3066_hdmi *hdmi)
+{
+	struct drm_encoder *encoder = &hdmi->encoder;
+	struct device *dev = hdmi->dev;
+
+	encoder->possible_crtcs =
+		drm_of_find_possible_crtcs(drm, dev->of_node);
+
+	/*
+	 * If we failed to find the CRTC(s) which this encoder is
+	 * supposed to be connected to, it's because the CRTC has
+	 * not been registered yet.  Defer probing, and hope that
+	 * the required CRTC is added later.
+	 */
+	if (encoder->possible_crtcs == 0)
+		return -EPROBE_DEFER;
+
+	drm_encoder_helper_add(encoder, &rk3066_hdmi_encoder_helper_funcs);
+	drm_encoder_init(drm, encoder, &rk3066_hdmi_encoder_funcs,
+			 DRM_MODE_ENCODER_TMDS, NULL);
+
+	hdmi->connector.polled = DRM_CONNECTOR_POLL_HPD;
+
+	drm_connector_helper_add(&hdmi->connector,
+				 &rk3066_hdmi_connector_helper_funcs);
+	drm_connector_init(drm, &hdmi->connector,
+			   &rk3066_hdmi_connector_funcs,
+			   DRM_MODE_CONNECTOR_HDMIA);
+
+	drm_connector_attach_encoder(&hdmi->connector, encoder);
+
+	return 0;
+}
+
+static irqreturn_t rk3066_hdmi_hardirq(int irq, void *dev_id)
+{
+	struct rk3066_hdmi *hdmi = dev_id;
+	irqreturn_t ret = IRQ_NONE;
+	u8 interrupt;
+
+	if (rk3066_hdmi_get_power_mode(hdmi) == HDMI_SYS_POWER_MODE_A)
+		hdmi_writeb(hdmi, HDMI_SYS_CTRL, HDMI_SYS_POWER_MODE_B);
+
+	interrupt = hdmi_readb(hdmi, HDMI_INTR_STATUS1);
+	if (interrupt)
+		hdmi_writeb(hdmi, HDMI_INTR_STATUS1, interrupt);
+
+	if (interrupt & HDMI_INTR_EDID_MASK) {
+		hdmi->i2c->stat = interrupt;
+		complete(&hdmi->i2c->cmpltn);
+	}
+
+	if (interrupt & (HDMI_INTR_HOTPLUG | HDMI_INTR_MSENS))
+		ret = IRQ_WAKE_THREAD;
+
+	return ret;
+}
+
+static irqreturn_t rk3066_hdmi_irq(int irq, void *dev_id)
+{
+	struct rk3066_hdmi *hdmi = dev_id;
+
+	drm_helper_hpd_irq_event(hdmi->connector.dev);
+
+	return IRQ_HANDLED;
+}
+
+static int rk3066_hdmi_i2c_read(struct rk3066_hdmi *hdmi, struct i2c_msg *msgs)
+{
+	int length = msgs->len;
+	u8 *buf = msgs->buf;
+	int ret;
+
+	ret = wait_for_completion_timeout(&hdmi->i2c->cmpltn, HZ / 10);
+	if (!ret || hdmi->i2c->stat & HDMI_INTR_EDID_ERR)
+		return -EAGAIN;
+
+	while (length--)
+		*buf++ = hdmi_readb(hdmi, HDMI_DDC_READ_FIFO_ADDR);
+
+	return 0;
+}
+
+static int rk3066_hdmi_i2c_write(struct rk3066_hdmi *hdmi, struct i2c_msg *msgs)
+{
+	/*
+	 * The DDC module only supports read EDID message, so
+	 * we assume that each word write to this i2c adapter
+	 * should be the offset of the EDID word address.
+	 */
+	if (msgs->len != 1 ||
+	    (msgs->addr != DDC_ADDR && msgs->addr != DDC_SEGMENT_ADDR))
+		return -EINVAL;
+
+	reinit_completion(&hdmi->i2c->cmpltn);
+
+	if (msgs->addr == DDC_SEGMENT_ADDR)
+		hdmi->i2c->segment_addr = msgs->buf[0];
+	if (msgs->addr == DDC_ADDR)
+		hdmi->i2c->ddc_addr = msgs->buf[0];
+
+	/* Set edid word address 0x00/0x80. */
+	hdmi_writeb(hdmi, HDMI_EDID_WORD_ADDR, hdmi->i2c->ddc_addr);
+
+	/* Set edid segment pointer. */
+	hdmi_writeb(hdmi, HDMI_EDID_SEGMENT_POINTER, hdmi->i2c->segment_addr);
+
+	return 0;
+}
+
+static int rk3066_hdmi_i2c_xfer(struct i2c_adapter *adap,
+				struct i2c_msg *msgs, int num)
+{
+	struct rk3066_hdmi *hdmi = i2c_get_adapdata(adap);
+	struct rk3066_hdmi_i2c *i2c = hdmi->i2c;
+	int i, ret = 0;
+
+	mutex_lock(&i2c->i2c_lock);
+
+	rk3066_hdmi_i2c_init(hdmi);
+
+	/* Unmute HDMI EDID interrupt. */
+	hdmi_modb(hdmi, HDMI_INTR_MASK1,
+		  HDMI_INTR_EDID_MASK, HDMI_INTR_EDID_MASK);
+	i2c->stat = 0;
+
+	for (i = 0; i < num; i++) {
+		DRM_DEV_DEBUG(hdmi->dev,
+			      "xfer: num: %d/%d, len: %d, flags: %#x\n",
+			      i + 1, num, msgs[i].len, msgs[i].flags);
+
+		if (msgs[i].flags & I2C_M_RD)
+			ret = rk3066_hdmi_i2c_read(hdmi, &msgs[i]);
+		else
+			ret = rk3066_hdmi_i2c_write(hdmi, &msgs[i]);
+
+		if (ret < 0)
+			break;
+	}
+
+	if (!ret)
+		ret = num;
+
+	/* Mute HDMI EDID interrupt. */
+	hdmi_modb(hdmi, HDMI_INTR_MASK1, HDMI_INTR_EDID_MASK, 0);
+
+	mutex_unlock(&i2c->i2c_lock);
+
+	return ret;
+}
+
+static u32 rk3066_hdmi_i2c_func(struct i2c_adapter *adapter)
+{
+	return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL;
+}
+
+static const struct i2c_algorithm rk3066_hdmi_algorithm = {
+	.master_xfer   = rk3066_hdmi_i2c_xfer,
+	.functionality = rk3066_hdmi_i2c_func,
+};
+
+static struct i2c_adapter *rk3066_hdmi_i2c_adapter(struct rk3066_hdmi *hdmi)
+{
+	struct i2c_adapter *adap;
+	struct rk3066_hdmi_i2c *i2c;
+	int ret;
+
+	i2c = devm_kzalloc(hdmi->dev, sizeof(*i2c), GFP_KERNEL);
+	if (!i2c)
+		return ERR_PTR(-ENOMEM);
+
+	mutex_init(&i2c->i2c_lock);
+	init_completion(&i2c->cmpltn);
+
+	adap = &i2c->adap;
+	adap->class = I2C_CLASS_DDC;
+	adap->owner = THIS_MODULE;
+	adap->dev.parent = hdmi->dev;
+	adap->dev.of_node = hdmi->dev->of_node;
+	adap->algo = &rk3066_hdmi_algorithm;
+	strlcpy(adap->name, "RK3066 HDMI", sizeof(adap->name));
+	i2c_set_adapdata(adap, hdmi);
+
+	ret = i2c_add_adapter(adap);
+	if (ret) {
+		DRM_DEV_ERROR(hdmi->dev, "cannot add %s I2C adapter\n",
+			      adap->name);
+		devm_kfree(hdmi->dev, i2c);
+		return ERR_PTR(ret);
+	}
+
+	hdmi->i2c = i2c;
+
+	DRM_DEV_DEBUG(hdmi->dev, "registered %s I2C bus driver\n", adap->name);
+
+	return adap;
+}
+
+static int rk3066_hdmi_bind(struct device *dev, struct device *master,
+			    void *data)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct drm_device *drm = data;
+	struct rk3066_hdmi *hdmi;
+	struct resource *iores;
+	int irq;
+	int ret;
+
+	hdmi = devm_kzalloc(dev, sizeof(*hdmi), GFP_KERNEL);
+	if (!hdmi)
+		return -ENOMEM;
+
+	hdmi->dev = dev;
+	hdmi->drm_dev = drm;
+
+	iores = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!iores)
+		return -ENXIO;
+
+	hdmi->regs = devm_ioremap_resource(dev, iores);
+	if (IS_ERR(hdmi->regs))
+		return PTR_ERR(hdmi->regs);
+
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0)
+		return irq;
+
+	hdmi->hclk = devm_clk_get(dev, "hclk");
+	if (IS_ERR(hdmi->hclk)) {
+		DRM_DEV_ERROR(dev, "unable to get HDMI hclk clock\n");
+		return PTR_ERR(hdmi->hclk);
+	}
+
+	ret = clk_prepare_enable(hdmi->hclk);
+	if (ret) {
+		DRM_DEV_ERROR(dev, "cannot enable HDMI hclk clock: %d\n", ret);
+		return ret;
+	}
+
+	hdmi->grf_regmap = syscon_regmap_lookup_by_phandle(dev->of_node,
+							   "rockchip,grf");
+	if (IS_ERR(hdmi->grf_regmap)) {
+		DRM_DEV_ERROR(dev, "unable to get rockchip,grf\n");
+		ret = PTR_ERR(hdmi->grf_regmap);
+		goto err_disable_hclk;
+	}
+
+	/* internal hclk = hdmi_hclk / 25 */
+	hdmi_writeb(hdmi, HDMI_INTERNAL_CLK_DIVIDER, 25);
+
+	hdmi->ddc = rk3066_hdmi_i2c_adapter(hdmi);
+	if (IS_ERR(hdmi->ddc)) {
+		ret = PTR_ERR(hdmi->ddc);
+		hdmi->ddc = NULL;
+		goto err_disable_hclk;
+	}
+
+	rk3066_hdmi_set_power_mode(hdmi, HDMI_SYS_POWER_MODE_B);
+	usleep_range(999, 1000);
+	hdmi_writeb(hdmi, HDMI_INTR_MASK1, HDMI_INTR_HOTPLUG);
+	hdmi_writeb(hdmi, HDMI_INTR_MASK2, 0);
+	hdmi_writeb(hdmi, HDMI_INTR_MASK3, 0);
+	hdmi_writeb(hdmi, HDMI_INTR_MASK4, 0);
+	rk3066_hdmi_set_power_mode(hdmi, HDMI_SYS_POWER_MODE_A);
+
+	ret = rk3066_hdmi_register(drm, hdmi);
+	if (ret)
+		goto err_disable_i2c;
+
+	dev_set_drvdata(dev, hdmi);
+
+	ret = devm_request_threaded_irq(dev, irq, rk3066_hdmi_hardirq,
+					rk3066_hdmi_irq, IRQF_SHARED,
+					dev_name(dev), hdmi);
+	if (ret) {
+		DRM_DEV_ERROR(dev, "failed to request hdmi irq: %d\n", ret);
+		goto err_cleanup_hdmi;
+	}
+
+	return 0;
+
+err_cleanup_hdmi:
+	hdmi->connector.funcs->destroy(&hdmi->connector);
+	hdmi->encoder.funcs->destroy(&hdmi->encoder);
+err_disable_i2c:
+	i2c_put_adapter(hdmi->ddc);
+err_disable_hclk:
+	clk_disable_unprepare(hdmi->hclk);
+
+	return ret;
+}
+
+static void rk3066_hdmi_unbind(struct device *dev, struct device *master,
+			       void *data)
+{
+	struct rk3066_hdmi *hdmi = dev_get_drvdata(dev);
+
+	hdmi->connector.funcs->destroy(&hdmi->connector);
+	hdmi->encoder.funcs->destroy(&hdmi->encoder);
+
+	i2c_put_adapter(hdmi->ddc);
+	clk_disable_unprepare(hdmi->hclk);
+}
+
+static const struct component_ops rk3066_hdmi_ops = {
+	.bind   = rk3066_hdmi_bind,
+	.unbind = rk3066_hdmi_unbind,
+};
+
+static int rk3066_hdmi_probe(struct platform_device *pdev)
+{
+	return component_add(&pdev->dev, &rk3066_hdmi_ops);
+}
+
+static int rk3066_hdmi_remove(struct platform_device *pdev)
+{
+	component_del(&pdev->dev, &rk3066_hdmi_ops);
+
+	return 0;
+}
+
+static const struct of_device_id rk3066_hdmi_dt_ids[] = {
+	{ .compatible = "rockchip,rk3066-hdmi" },
+	{ /* sentinel */ },
+};
+MODULE_DEVICE_TABLE(of, rk3066_hdmi_dt_ids);
+
+struct platform_driver rk3066_hdmi_driver = {
+	.probe  = rk3066_hdmi_probe,
+	.remove = rk3066_hdmi_remove,
+	.driver = {
+		.name = "rockchip-rk3066-hdmi",
+		.of_match_table = rk3066_hdmi_dt_ids,
+	},
+};
diff --git a/drivers/gpu/drm/rockchip/rk3066_hdmi.h b/drivers/gpu/drm/rockchip/rk3066_hdmi.h
new file mode 100644
index 000000000000..39a31c62a428
--- /dev/null
+++ b/drivers/gpu/drm/rockchip/rk3066_hdmi.h
@@ -0,0 +1,229 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) Fuzhou Rockchip Electronics Co.Ltd
+ *    Zheng Yang <zhengyang@rock-chips.com>
+ */
+
+#ifndef __RK3066_HDMI_H__
+#define __RK3066_HDMI_H__
+
+#define GRF_SOC_CON0				0x150
+#define HDMI_VIDEO_SEL				BIT(14)
+
+#define DDC_SEGMENT_ADDR			0x30
+#define HDMI_SCL_RATE				(50 * 1000)
+#define HDMI_MAXIMUM_INFO_FRAME_SIZE		0x11
+
+#define N_32K					0x1000
+#define N_441K					0x1880
+#define N_882K					0x3100
+#define N_1764K					0x6200
+#define N_48K					0x1800
+#define N_96K					0x3000
+#define N_192K					0x6000
+
+#define HDMI_SYS_CTRL				0x000
+#define HDMI_LR_SWAP_N3				0x004
+#define HDMI_N2					0x008
+#define HDMI_N1					0x00c
+#define HDMI_SPDIF_FS_CTS_INT3			0x010
+#define HDMI_CTS_INT2				0x014
+#define HDMI_CTS_INT1				0x018
+#define HDMI_CTS_EXT3				0x01c
+#define HDMI_CTS_EXT2				0x020
+#define HDMI_CTS_EXT1				0x024
+#define HDMI_AUDIO_CTRL1			0x028
+#define HDMI_AUDIO_CTRL2			0x02c
+#define HDMI_I2S_AUDIO_CTRL			0x030
+#define HDMI_I2S_SWAP				0x040
+#define HDMI_AUDIO_STA_BIT_CTRL1		0x044
+#define HDMI_AUDIO_STA_BIT_CTRL2		0x048
+#define HDMI_AUDIO_SRC_NUM_AND_LENGTH		0x050
+#define HDMI_AV_CTRL1				0x054
+#define HDMI_VIDEO_CTRL1			0x058
+#define HDMI_DEEP_COLOR_MODE			0x05c
+
+#define HDMI_EXT_VIDEO_PARA			0x0c0
+#define HDMI_EXT_HTOTAL_L			0x0c4
+#define HDMI_EXT_HTOTAL_H			0x0c8
+#define HDMI_EXT_HBLANK_L			0x0cc
+#define HDMI_EXT_HBLANK_H			0x0d0
+#define HDMI_EXT_HDELAY_L			0x0d4
+#define HDMI_EXT_HDELAY_H			0x0d8
+#define HDMI_EXT_HDURATION_L			0x0dc
+#define HDMI_EXT_HDURATION_H			0x0e0
+#define HDMI_EXT_VTOTAL_L			0x0e4
+#define HDMI_EXT_VTOTAL_H			0x0e8
+#define HDMI_AV_CTRL2				0x0ec
+#define HDMI_EXT_VBLANK_L			0x0f4
+#define HDMI_EXT_VBLANK_H			0x10c
+#define HDMI_EXT_VDELAY				0x0f8
+#define HDMI_EXT_VDURATION			0x0fc
+
+#define HDMI_CP_MANU_SEND_CTRL			0x100
+#define HDMI_CP_AUTO_SEND_CTRL			0x104
+#define HDMI_AUTO_CHECKSUM_OPT			0x108
+
+#define HDMI_VIDEO_CTRL2			0x114
+
+#define HDMI_PHY_OPTION				0x144
+
+#define HDMI_CP_BUF_INDEX			0x17c
+#define HDMI_CP_BUF_ACC_HB0			0x180
+#define HDMI_CP_BUF_ACC_HB1			0x184
+#define HDMI_CP_BUF_ACC_HB2			0x188
+#define HDMI_CP_BUF_ACC_PB0			0x18c
+
+#define HDMI_DDC_READ_FIFO_ADDR			0x200
+#define HDMI_DDC_BUS_FREQ_L			0x204
+#define HDMI_DDC_BUS_FREQ_H			0x208
+#define HDMI_DDC_BUS_CTRL			0x2dc
+#define HDMI_DDC_I2C_LEN			0x278
+#define HDMI_DDC_I2C_OFFSET			0x280
+#define HDMI_DDC_I2C_CTRL			0x284
+#define HDMI_DDC_I2C_READ_BUF0			0x288
+#define HDMI_DDC_I2C_READ_BUF1			0x28c
+#define HDMI_DDC_I2C_READ_BUF2			0x290
+#define HDMI_DDC_I2C_READ_BUF3			0x294
+#define HDMI_DDC_I2C_WRITE_BUF0			0x298
+#define HDMI_DDC_I2C_WRITE_BUF1			0x29c
+#define HDMI_DDC_I2C_WRITE_BUF2			0x2a0
+#define HDMI_DDC_I2C_WRITE_BUF3			0x2a4
+#define HDMI_DDC_I2C_WRITE_BUF4			0x2ac
+#define HDMI_DDC_I2C_WRITE_BUF5			0x2b0
+#define HDMI_DDC_I2C_WRITE_BUF6			0x2b4
+
+#define HDMI_INTR_MASK1				0x248
+#define HDMI_INTR_MASK2				0x24c
+#define HDMI_INTR_STATUS1			0x250
+#define HDMI_INTR_STATUS2			0x254
+#define HDMI_INTR_MASK3				0x258
+#define HDMI_INTR_MASK4				0x25c
+#define HDMI_INTR_STATUS3			0x260
+#define HDMI_INTR_STATUS4			0x264
+
+#define HDMI_HDCP_CTRL				0x2bc
+
+#define HDMI_EDID_SEGMENT_POINTER		0x310
+#define HDMI_EDID_WORD_ADDR			0x314
+#define HDMI_EDID_FIFO_ADDR			0x318
+
+#define HDMI_HPG_MENS_STA			0x37c
+
+#define HDMI_INTERNAL_CLK_DIVIDER		0x800
+
+enum {
+	/* HDMI_SYS_CTRL */
+	HDMI_SYS_POWER_MODE_MASK = 0xf0,
+	HDMI_SYS_POWER_MODE_A = 0x10,
+	HDMI_SYS_POWER_MODE_B = 0x20,
+	HDMI_SYS_POWER_MODE_D = 0x40,
+	HDMI_SYS_POWER_MODE_E = 0x80,
+	HDMI_SYS_PLL_RESET_MASK = 0x0c,
+	HDMI_SYS_PLL_RESET = 0x0c,
+	HDMI_SYS_PLLB_RESET = 0x08,
+
+	/* HDMI_LR_SWAP_N3 */
+	HDMI_AUDIO_LR_SWAP_MASK = 0xf0,
+	HDMI_AUDIO_LR_SWAP_SUBPACKET0 = 0x10,
+	HDMI_AUDIO_LR_SWAP_SUBPACKET1 = 0x20,
+	HDMI_AUDIO_LR_SWAP_SUBPACKET2 = 0x40,
+	HDMI_AUDIO_LR_SWAP_SUBPACKET3 = 0x80,
+	HDMI_AUDIO_N_19_16_MASK = 0x0f,
+
+	/* HDMI_AUDIO_CTRL1 */
+	HDMI_AUDIO_EXTERNAL_CTS = BIT(7),
+	HDMI_AUDIO_INPUT_IIS = 0,
+	HDMI_AUDIO_INPUT_SPDIF = 0x08,
+	HDMI_AUDIO_INPUT_MCLK_ACTIVE = 0x04,
+	HDMI_AUDIO_INPUT_MCLK_DEACTIVE = 0,
+	HDMI_AUDIO_INPUT_MCLK_RATE_128X = 0,
+	HDMI_AUDIO_INPUT_MCLK_RATE_256X = 1,
+	HDMI_AUDIO_INPUT_MCLK_RATE_384X = 2,
+	HDMI_AUDIO_INPUT_MCLK_RATE_512X = 3,
+
+	/* HDMI_I2S_AUDIO_CTRL */
+	HDMI_AUDIO_I2S_FORMAT_STANDARD = 0,
+	HDMI_AUDIO_I2S_CHANNEL_1_2 = 0x04,
+	HDMI_AUDIO_I2S_CHANNEL_3_4 = 0x0c,
+	HDMI_AUDIO_I2S_CHANNEL_5_6 = 0x1c,
+	HDMI_AUDIO_I2S_CHANNEL_7_8 = 0x3c,
+
+	/* HDMI_AV_CTRL1 */
+	HDMI_AUDIO_SAMPLE_FRE_MASK = 0xf0,
+	HDMI_AUDIO_SAMPLE_FRE_32000 = 0x30,
+	HDMI_AUDIO_SAMPLE_FRE_44100 = 0,
+	HDMI_AUDIO_SAMPLE_FRE_48000 = 0x20,
+	HDMI_AUDIO_SAMPLE_FRE_88200 = 0x80,
+	HDMI_AUDIO_SAMPLE_FRE_96000 = 0xa0,
+	HDMI_AUDIO_SAMPLE_FRE_176400 = 0xc0,
+	HDMI_AUDIO_SAMPLE_FRE_192000 = 0xe0,
+	HDMI_AUDIO_SAMPLE_FRE_768000 = 0x90,
+
+	HDMI_VIDEO_INPUT_FORMAT_MASK = 0x0e,
+	HDMI_VIDEO_INPUT_RGB_YCBCR444 = 0,
+	HDMI_VIDEO_INPUT_YCBCR422 = 0x02,
+	HDMI_VIDEO_DE_MASK = 0x1,
+	HDMI_VIDEO_INTERNAL_DE = 0,
+	HDMI_VIDEO_EXTERNAL_DE = 0x01,
+
+	/* HDMI_VIDEO_CTRL1 */
+	HDMI_VIDEO_OUTPUT_FORMAT_MASK = 0xc0,
+	HDMI_VIDEO_OUTPUT_RGB444 = 0,
+	HDMI_VIDEO_OUTPUT_YCBCR444 = 0x40,
+	HDMI_VIDEO_OUTPUT_YCBCR422 = 0x80,
+	HDMI_VIDEO_INPUT_DATA_DEPTH_MASK = 0x30,
+	HDMI_VIDEO_INPUT_DATA_DEPTH_12BIT = 0,
+	HDMI_VIDEO_INPUT_DATA_DEPTH_10BIT = 0x10,
+	HDMI_VIDEO_INPUT_DATA_DEPTH_8BIT = 0x30,
+	HDMI_VIDEO_INPUT_COLOR_MASK = 1,
+	HDMI_VIDEO_INPUT_COLOR_RGB = 0,
+	HDMI_VIDEO_INPUT_COLOR_YCBCR = 1,
+
+	/* HDMI_EXT_VIDEO_PARA */
+	HDMI_VIDEO_VSYNC_OFFSET_SHIFT = 4,
+	HDMI_VIDEO_VSYNC_ACTIVE_HIGH = BIT(3),
+	HDMI_VIDEO_VSYNC_ACTIVE_LOW = 0,
+	HDMI_VIDEO_HSYNC_ACTIVE_HIGH = BIT(2),
+	HDMI_VIDEO_HSYNC_ACTIVE_LOW = 0,
+	HDMI_VIDEO_MODE_INTERLACE = BIT(1),
+	HDMI_VIDEO_MODE_PROGRESSIVE = 0,
+	HDMI_EXT_VIDEO_SET_EN = BIT(0),
+
+	/* HDMI_CP_AUTO_SEND_CTRL */
+
+	/* HDMI_VIDEO_CTRL2 */
+	HDMI_VIDEO_AV_MUTE_MASK = 0xc0,
+	HDMI_VIDEO_CLR_AV_MUTE = BIT(7),
+	HDMI_VIDEO_SET_AV_MUTE = BIT(6),
+	HDMI_AUDIO_CP_LOGIC_RESET_MASK = BIT(2),
+	HDMI_AUDIO_CP_LOGIC_RESET = BIT(2),
+	HDMI_VIDEO_AUDIO_DISABLE_MASK = 0x3,
+	HDMI_AUDIO_DISABLE = BIT(1),
+	HDMI_VIDEO_DISABLE = BIT(0),
+
+	/* HDMI_CP_BUF_INDEX */
+	HDMI_INFOFRAME_VSI = 0x05,
+	HDMI_INFOFRAME_AVI = 0x06,
+	HDMI_INFOFRAME_AAI = 0x08,
+
+	/* HDMI_INTR_MASK1 */
+	/* HDMI_INTR_STATUS1 */
+	HDMI_INTR_HOTPLUG = BIT(7),
+	HDMI_INTR_MSENS = BIT(6),
+	HDMI_INTR_VSYNC = BIT(5),
+	HDMI_INTR_AUDIO_FIFO_FULL = BIT(4),
+	HDMI_INTR_EDID_MASK = 0x6,
+	HDMI_INTR_EDID_READY = BIT(2),
+	HDMI_INTR_EDID_ERR = BIT(1),
+
+	/* HDMI_HDCP_CTRL */
+	HDMI_VIDEO_MODE_MASK = BIT(1),
+	HDMI_VIDEO_MODE_HDMI = BIT(1),
+
+	/* HDMI_HPG_MENS_STA */
+	HDMI_HPG_IN_STATUS_HIGH = BIT(7),
+	HDMI_MSENS_IN_STATUS_HIGH = BIT(6),
+};
+
+#endif /* __RK3066_HDMI_H__ */
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
index d7fa17f12769..cb938d3cd3c2 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
@@ -448,6 +448,14 @@ static int rockchip_drm_platform_remove(struct platform_device *pdev)
 	return 0;
 }
 
+static void rockchip_drm_platform_shutdown(struct platform_device *pdev)
+{
+	struct drm_device *drm = platform_get_drvdata(pdev);
+
+	if (drm)
+		drm_atomic_helper_shutdown(drm);
+}
+
 static const struct of_device_id rockchip_drm_dt_ids[] = {
 	{ .compatible = "rockchip,display-subsystem", },
 	{ /* sentinel */ },
@@ -457,6 +465,7 @@ MODULE_DEVICE_TABLE(of, rockchip_drm_dt_ids);
 static struct platform_driver rockchip_drm_platform_driver = {
 	.probe = rockchip_drm_platform_probe,
 	.remove = rockchip_drm_platform_remove,
+	.shutdown = rockchip_drm_platform_shutdown,
 	.driver = {
 		.name = "rockchip-drm",
 		.of_match_table = rockchip_drm_dt_ids,
@@ -486,6 +495,8 @@ static int __init rockchip_drm_init(void)
 	ADD_ROCKCHIP_SUB_DRIVER(dw_mipi_dsi_rockchip_driver,
 				CONFIG_ROCKCHIP_DW_MIPI_DSI);
 	ADD_ROCKCHIP_SUB_DRIVER(inno_hdmi_driver, CONFIG_ROCKCHIP_INNO_HDMI);
+	ADD_ROCKCHIP_SUB_DRIVER(rk3066_hdmi_driver,
+				CONFIG_ROCKCHIP_RK3066_HDMI);
 
 	ret = platform_register_drivers(rockchip_sub_drivers,
 					num_rockchip_sub_drivers);
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_drv.h b/drivers/gpu/drm/rockchip/rockchip_drm_drv.h
index ce48568ec8a0..e4bc4322bc3f 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_drv.h
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_drv.h
@@ -73,4 +73,5 @@ extern struct platform_driver inno_hdmi_driver;
 extern struct platform_driver rockchip_dp_driver;
 extern struct platform_driver rockchip_lvds_driver;
 extern struct platform_driver vop_platform_driver;
+extern struct platform_driver rk3066_hdmi_driver;
 #endif /* _ROCKCHIP_DRM_DRV_H_ */
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.c b/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.c
index 8ce68bd508be..30459de66b67 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.c
@@ -90,12 +90,10 @@ static int rockchip_drm_fbdev_create(struct drm_fb_helper *helper,
 		goto out;
 	}
 
-	fbi->par = helper;
 	fbi->fbops = &rockchip_drm_fbdev_ops;
 
 	fb = helper->fb;
-	drm_fb_helper_fill_fix(fbi, fb->pitches[0], fb->format->depth);
-	drm_fb_helper_fill_var(fbi, helper, sizes->fb_width, sizes->fb_height);
+	drm_fb_helper_fill_info(fbi, helper, sizes);
 
 	offset = fbi->var.xoffset * bytes_per_pixel;
 	offset += fbi->var.yoffset * fb->pitches[0];
@@ -110,8 +108,6 @@ static int rockchip_drm_fbdev_create(struct drm_fb_helper *helper,
 		      rk_obj->kvaddr,
 		      offset, size);
 
-	fbi->skip_vt_switch = true;
-
 	return 0;
 
 out:
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
index 0d4ade9d4722..20a9c296d027 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
@@ -1041,6 +1041,7 @@ static void vop_crtc_atomic_enable(struct drm_crtc *crtc,
 	u16 vact_st = adjusted_mode->vtotal - adjusted_mode->vsync_start;
 	u16 vact_end = vact_st + vdisplay;
 	uint32_t pin_pol, val;
+	int dither_bpc = s->output_bpc ? s->output_bpc : 10;
 	int ret;
 
 	mutex_lock(&vop->vop_lock);
@@ -1098,11 +1099,19 @@ static void vop_crtc_atomic_enable(struct drm_crtc *crtc,
 	    !(vop_data->feature & VOP_FEATURE_OUTPUT_RGB10))
 		s->output_mode = ROCKCHIP_OUT_MODE_P888;
 
-	if (s->output_mode == ROCKCHIP_OUT_MODE_AAAA && s->output_bpc == 8)
+	if (s->output_mode == ROCKCHIP_OUT_MODE_AAAA && dither_bpc <= 8)
 		VOP_REG_SET(vop, common, pre_dither_down, 1);
 	else
 		VOP_REG_SET(vop, common, pre_dither_down, 0);
 
+	if (dither_bpc == 6) {
+		VOP_REG_SET(vop, common, dither_down_sel, DITHER_DOWN_ALLEGRO);
+		VOP_REG_SET(vop, common, dither_down_mode, RGB888_TO_RGB666);
+		VOP_REG_SET(vop, common, dither_down_en, 1);
+	} else {
+		VOP_REG_SET(vop, common, dither_down_en, 0);
+	}
+
 	VOP_REG_SET(vop, common, out_mode, s->output_mode);
 
 	VOP_REG_SET(vop, modeset, htotal_pw, (htotal << 16) | hsync_len);
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop.h b/drivers/gpu/drm/rockchip/rockchip_drm_vop.h
index 04ed401d2325..e64351dab610 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_vop.h
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop.h
@@ -71,7 +71,9 @@ struct vop_common {
 	struct vop_reg dsp_blank;
 	struct vop_reg data_blank;
 	struct vop_reg pre_dither_down;
-	struct vop_reg dither_down;
+	struct vop_reg dither_down_sel;
+	struct vop_reg dither_down_mode;
+	struct vop_reg dither_down_en;
 	struct vop_reg dither_up;
 	struct vop_reg gate_en;
 	struct vop_reg mmu_en;
@@ -287,6 +289,16 @@ enum scale_down_mode {
 	SCALE_DOWN_AVG = 0x1
 };
 
+enum dither_down_mode {
+	RGB888_TO_RGB565 = 0x0,
+	RGB888_TO_RGB666 = 0x1
+};
+
+enum dither_down_mode_sel {
+	DITHER_DOWN_ALLEGRO = 0x0,
+	DITHER_DOWN_FRC = 0x1
+};
+
 enum vop_pol {
 	HSYNC_POSITIVE = 0,
 	VSYNC_POSITIVE = 1,
diff --git a/drivers/gpu/drm/rockchip/rockchip_vop_reg.c b/drivers/gpu/drm/rockchip/rockchip_vop_reg.c
index bd76328c0fdb..e732b73033c8 100644
--- a/drivers/gpu/drm/rockchip/rockchip_vop_reg.c
+++ b/drivers/gpu/drm/rockchip/rockchip_vop_reg.c
@@ -137,6 +137,9 @@ static const struct vop_common rk3036_common = {
 	.standby = VOP_REG_SYNC(RK3036_SYS_CTRL, 0x1, 30),
 	.out_mode = VOP_REG(RK3036_DSP_CTRL0, 0xf, 0),
 	.dsp_blank = VOP_REG(RK3036_DSP_CTRL1, 0x1, 24),
+	.dither_down_sel = VOP_REG(RK3036_DSP_CTRL0, 0x1, 27),
+	.dither_down_en = VOP_REG(RK3036_DSP_CTRL0, 0x1, 11),
+	.dither_down_mode = VOP_REG(RK3036_DSP_CTRL0, 0x1, 10),
 	.cfg_done = VOP_REG_SYNC(RK3036_REG_CFG_DONE, 0x1, 0),
 };
 
@@ -200,6 +203,9 @@ static const struct vop_common px30_common = {
 	.standby = VOP_REG_SYNC(PX30_SYS_CTRL2, 0x1, 1),
 	.out_mode = VOP_REG(PX30_DSP_CTRL2, 0xf, 16),
 	.dsp_blank = VOP_REG(PX30_DSP_CTRL2, 0x1, 14),
+	.dither_down_en = VOP_REG(PX30_DSP_CTRL2, 0x1, 8),
+	.dither_down_sel = VOP_REG(PX30_DSP_CTRL2, 0x1, 7),
+	.dither_down_mode = VOP_REG(PX30_DSP_CTRL2, 0x1, 6),
 	.cfg_done = VOP_REG_SYNC(PX30_REG_CFG_DONE, 0x1, 0),
 };
 
@@ -365,6 +371,8 @@ static const struct vop_common rk3066_common = {
 	.standby = VOP_REG(RK3066_SYS_CTRL0, 0x1, 1),
 	.out_mode = VOP_REG(RK3066_DSP_CTRL0, 0xf, 0),
 	.cfg_done = VOP_REG(RK3066_REG_CFG_DONE, 0x1, 0),
+	.dither_down_en = VOP_REG(RK3066_DSP_CTRL0, 0x1, 11),
+	.dither_down_mode = VOP_REG(RK3066_DSP_CTRL0, 0x1, 10),
 	.dsp_blank = VOP_REG(RK3066_DSP_CTRL1, 0x1, 24),
 };
 
@@ -458,6 +466,9 @@ static const struct vop_common rk3188_common = {
 	.standby = VOP_REG(RK3188_SYS_CTRL, 0x1, 30),
 	.out_mode = VOP_REG(RK3188_DSP_CTRL0, 0xf, 0),
 	.cfg_done = VOP_REG(RK3188_REG_CFG_DONE, 0x1, 0),
+	.dither_down_sel = VOP_REG(RK3188_DSP_CTRL0, 0x1, 27),
+	.dither_down_en = VOP_REG(RK3188_DSP_CTRL0, 0x1, 11),
+	.dither_down_mode = VOP_REG(RK3188_DSP_CTRL0, 0x1, 10),
 	.dsp_blank = VOP_REG(RK3188_DSP_CTRL1, 0x3, 24),
 };
 
@@ -585,8 +596,10 @@ static const struct vop_common rk3288_common = {
 	.standby = VOP_REG_SYNC(RK3288_SYS_CTRL, 0x1, 22),
 	.gate_en = VOP_REG(RK3288_SYS_CTRL, 0x1, 23),
 	.mmu_en = VOP_REG(RK3288_SYS_CTRL, 0x1, 20),
+	.dither_down_sel = VOP_REG(RK3288_DSP_CTRL1, 0x1, 4),
+	.dither_down_mode = VOP_REG(RK3288_DSP_CTRL1, 0x1, 3),
+	.dither_down_en = VOP_REG(RK3288_DSP_CTRL1, 0x1, 2),
 	.pre_dither_down = VOP_REG(RK3288_DSP_CTRL1, 0x1, 1),
-	.dither_down = VOP_REG(RK3288_DSP_CTRL1, 0xf, 1),
 	.dither_up = VOP_REG(RK3288_DSP_CTRL1, 0x1, 6),
 	.data_blank = VOP_REG(RK3288_DSP_CTRL0, 0x1, 19),
 	.dsp_blank = VOP_REG(RK3288_DSP_CTRL0, 0x3, 18),
@@ -878,7 +891,10 @@ static const struct vop_misc rk3328_misc = {
 
 static const struct vop_common rk3328_common = {
 	.standby = VOP_REG_SYNC(RK3328_SYS_CTRL, 0x1, 22),
-	.dither_down = VOP_REG(RK3328_DSP_CTRL1, 0xf, 1),
+	.dither_down_sel = VOP_REG(RK3328_DSP_CTRL1, 0x1, 4),
+	.dither_down_mode = VOP_REG(RK3328_DSP_CTRL1, 0x1, 3),
+	.dither_down_en = VOP_REG(RK3328_DSP_CTRL1, 0x1, 2),
+	.pre_dither_down = VOP_REG(RK3328_DSP_CTRL1, 0x1, 1),
 	.dither_up = VOP_REG(RK3328_DSP_CTRL1, 0x1, 6),
 	.dsp_blank = VOP_REG(RK3328_DSP_CTRL0, 0x3, 18),
 	.out_mode = VOP_REG(RK3328_DSP_CTRL0, 0xf, 0),
diff --git a/drivers/gpu/drm/selftests/test-drm_mm.c b/drivers/gpu/drm/selftests/test-drm_mm.c
index fbed2c90fd51..286a0eeefcb6 100644
--- a/drivers/gpu/drm/selftests/test-drm_mm.c
+++ b/drivers/gpu/drm/selftests/test-drm_mm.c
@@ -1615,7 +1615,7 @@ static int igt_topdown(void *ignored)
 	DRM_RND_STATE(prng, random_seed);
 	const unsigned int count = 8192;
 	unsigned int size;
-	unsigned long *bitmap = NULL;
+	unsigned long *bitmap;
 	struct drm_mm mm;
 	struct drm_mm_node *nodes, *node, *next;
 	unsigned int *order, n, m, o = 0;
@@ -1631,8 +1631,7 @@ static int igt_topdown(void *ignored)
 	if (!nodes)
 		goto err;
 
-	bitmap = kcalloc(count / BITS_PER_LONG, sizeof(unsigned long),
-			 GFP_KERNEL);
+	bitmap = bitmap_zalloc(count, GFP_KERNEL);
 	if (!bitmap)
 		goto err_nodes;
 
@@ -1717,7 +1716,7 @@ out:
 	drm_mm_takedown(&mm);
 	kfree(order);
 err_bitmap:
-	kfree(bitmap);
+	bitmap_free(bitmap);
 err_nodes:
 	vfree(nodes);
 err:
@@ -1745,8 +1744,7 @@ static int igt_bottomup(void *ignored)
 	if (!nodes)
 		goto err;
 
-	bitmap = kcalloc(count / BITS_PER_LONG, sizeof(unsigned long),
-			 GFP_KERNEL);
+	bitmap = bitmap_zalloc(count, GFP_KERNEL);
 	if (!bitmap)
 		goto err_nodes;
 
@@ -1818,7 +1816,7 @@ out:
 	drm_mm_takedown(&mm);
 	kfree(order);
 err_bitmap:
-	kfree(bitmap);
+	bitmap_free(bitmap);
 err_nodes:
 	vfree(nodes);
 err:
diff --git a/drivers/gpu/drm/stm/Kconfig b/drivers/gpu/drm/stm/Kconfig
index 35367ada3bc1..d15b10de1da6 100644
--- a/drivers/gpu/drm/stm/Kconfig
+++ b/drivers/gpu/drm/stm/Kconfig
@@ -6,7 +6,7 @@ config DRM_STM
 	select DRM_KMS_CMA_HELPER
 	select DRM_PANEL_BRIDGE
 	select VIDEOMODE_HELPERS
-	select FB_PROVIDE_GET_FB_UNMAPPED_AREA
+	select FB_PROVIDE_GET_FB_UNMAPPED_AREA if FB
 
 	help
 	  Enable support for the on-chip display controller on
diff --git a/drivers/gpu/drm/stm/drv.c b/drivers/gpu/drm/stm/drv.c
index 0a7f933ab007..5834ef56fbaa 100644
--- a/drivers/gpu/drm/stm/drv.c
+++ b/drivers/gpu/drm/stm/drv.c
@@ -129,6 +129,40 @@ static void drv_unload(struct drm_device *ddev)
 	drm_mode_config_cleanup(ddev);
 }
 
+static __maybe_unused int drv_suspend(struct device *dev)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct ltdc_device *ldev = ddev->dev_private;
+	struct drm_atomic_state *state;
+
+	drm_kms_helper_poll_disable(ddev);
+	state = drm_atomic_helper_suspend(ddev);
+	if (IS_ERR(state)) {
+		drm_kms_helper_poll_enable(ddev);
+		return PTR_ERR(state);
+	}
+	ldev->suspend_state = state;
+	ltdc_suspend(ddev);
+
+	return 0;
+}
+
+static __maybe_unused int drv_resume(struct device *dev)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct ltdc_device *ldev = ddev->dev_private;
+
+	ltdc_resume(ddev);
+	drm_atomic_helper_resume(ddev, ldev->suspend_state);
+	drm_kms_helper_poll_enable(ddev);
+
+	return 0;
+}
+
+static const struct dev_pm_ops drv_pm_ops = {
+	SET_SYSTEM_SLEEP_PM_OPS(drv_suspend, drv_resume)
+};
+
 static int stm_drm_platform_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
@@ -186,6 +220,7 @@ static struct platform_driver stm_drm_platform_driver = {
 	.driver = {
 		.name = "stm32-display",
 		.of_match_table = drv_dt_ids,
+		.pm = &drv_pm_ops,
 	},
 };
 
diff --git a/drivers/gpu/drm/stm/dw_mipi_dsi-stm.c b/drivers/gpu/drm/stm/dw_mipi_dsi-stm.c
index a672b59a2226..1bef73e8c8fe 100644
--- a/drivers/gpu/drm/stm/dw_mipi_dsi-stm.c
+++ b/drivers/gpu/drm/stm/dw_mipi_dsi-stm.c
@@ -356,12 +356,40 @@ static int dw_mipi_dsi_stm_remove(struct platform_device *pdev)
 	return 0;
 }
 
+static int __maybe_unused dw_mipi_dsi_stm_suspend(struct device *dev)
+{
+	struct dw_mipi_dsi_stm *dsi = dw_mipi_dsi_stm_plat_data.priv_data;
+
+	DRM_DEBUG_DRIVER("\n");
+
+	clk_disable_unprepare(dsi->pllref_clk);
+
+	return 0;
+}
+
+static int __maybe_unused dw_mipi_dsi_stm_resume(struct device *dev)
+{
+	struct dw_mipi_dsi_stm *dsi = dw_mipi_dsi_stm_plat_data.priv_data;
+
+	DRM_DEBUG_DRIVER("\n");
+
+	clk_prepare_enable(dsi->pllref_clk);
+
+	return 0;
+}
+
+static const struct dev_pm_ops dw_mipi_dsi_stm_pm_ops = {
+	SET_SYSTEM_SLEEP_PM_OPS(dw_mipi_dsi_stm_suspend,
+				dw_mipi_dsi_stm_resume)
+};
+
 static struct platform_driver dw_mipi_dsi_stm_driver = {
 	.probe		= dw_mipi_dsi_stm_probe,
 	.remove		= dw_mipi_dsi_stm_remove,
 	.driver		= {
 		.of_match_table = dw_mipi_dsi_stm_dt_ids,
 		.name	= "stm32-display-dsi",
+		.pm = &dw_mipi_dsi_stm_pm_ops,
 	},
 };
 
diff --git a/drivers/gpu/drm/stm/ltdc.c b/drivers/gpu/drm/stm/ltdc.c
index b1741a9d5be2..32fd6a3b37fb 100644
--- a/drivers/gpu/drm/stm/ltdc.c
+++ b/drivers/gpu/drm/stm/ltdc.c
@@ -1062,6 +1062,30 @@ static int ltdc_get_caps(struct drm_device *ddev)
 	return 0;
 }
 
+void ltdc_suspend(struct drm_device *ddev)
+{
+	struct ltdc_device *ldev = ddev->dev_private;
+
+	DRM_DEBUG_DRIVER("\n");
+	clk_disable_unprepare(ldev->pixel_clk);
+}
+
+int ltdc_resume(struct drm_device *ddev)
+{
+	struct ltdc_device *ldev = ddev->dev_private;
+	int ret;
+
+	DRM_DEBUG_DRIVER("\n");
+
+	ret = clk_prepare_enable(ldev->pixel_clk);
+	if (ret) {
+		DRM_ERROR("failed to enable pixel clock (%d)\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
 int ltdc_load(struct drm_device *ddev)
 {
 	struct platform_device *pdev = to_platform_device(ddev->dev);
diff --git a/drivers/gpu/drm/stm/ltdc.h b/drivers/gpu/drm/stm/ltdc.h
index e46f477a8494..a1ad0ae3b006 100644
--- a/drivers/gpu/drm/stm/ltdc.h
+++ b/drivers/gpu/drm/stm/ltdc.h
@@ -36,6 +36,7 @@ struct ltdc_device {
 	u32 error_status;
 	u32 irq_status;
 	struct fps_info plane_fpsi[LTDC_MAX_LAYER];
+	struct drm_atomic_state *suspend_state;
 };
 
 bool ltdc_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe,
@@ -45,5 +46,7 @@ bool ltdc_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe,
 
 int ltdc_load(struct drm_device *ddev);
 void ltdc_unload(struct drm_device *ddev);
+void ltdc_suspend(struct drm_device *ddev);
+int ltdc_resume(struct drm_device *ddev);
 
 #endif
diff --git a/drivers/gpu/drm/sun4i/sun4i_backend.c b/drivers/gpu/drm/sun4i/sun4i_backend.c
index 4c0d51f73237..4e5922c89d7b 100644
--- a/drivers/gpu/drm/sun4i/sun4i_backend.c
+++ b/drivers/gpu/drm/sun4i/sun4i_backend.c
@@ -361,13 +361,6 @@ int sun4i_backend_update_layer_buffer(struct sun4i_backend *backend,
 	paddr = drm_fb_cma_get_gem_addr(fb, state, 0);
 	DRM_DEBUG_DRIVER("Setting buffer address to %pad\n", &paddr);
 
-	/*
-	 * backend DMA accesses DRAM directly, bypassing the system
-	 * bus. As such, the address range is different and the buffer
-	 * address needs to be corrected.
-	 */
-	paddr -= PHYS_OFFSET;
-
 	if (fb->format->is_yuv)
 		return sun4i_backend_update_yuv_buffer(backend, fb, paddr);
 
@@ -720,33 +713,22 @@ static int sun4i_backend_free_sat(struct device *dev) {
  */
 static int sun4i_backend_of_get_id(struct device_node *node)
 {
-	struct device_node *port, *ep;
-	int ret = -EINVAL;
+	struct device_node *ep, *remote;
+	struct of_endpoint of_ep;
 
-	/* input is port 0 */
-	port = of_graph_get_port_by_id(node, 0);
-	if (!port)
+	/* Input port is 0, and we want the first endpoint. */
+	ep = of_graph_get_endpoint_by_regs(node, 0, -1);
+	if (!ep)
 		return -EINVAL;
 
-	/* try finding an upstream endpoint */
-	for_each_available_child_of_node(port, ep) {
-		struct device_node *remote;
-		u32 reg;
-
-		remote = of_graph_get_remote_endpoint(ep);
-		if (!remote)
-			continue;
-
-		ret = of_property_read_u32(remote, "reg", &reg);
-		if (ret)
-			continue;
-
-		ret = reg;
-	}
-
-	of_node_put(port);
+	remote = of_graph_get_remote_endpoint(ep);
+	of_node_put(ep);
+	if (!remote)
+		return -EINVAL;
 
-	return ret;
+	of_graph_parse_endpoint(remote, &of_ep);
+	of_node_put(remote);
+	return of_ep.id;
 }
 
 /* TODO: This needs to take multiple pipelines into account */
@@ -814,6 +796,27 @@ static int sun4i_backend_bind(struct device *dev, struct device *master,
 	dev_set_drvdata(dev, backend);
 	spin_lock_init(&backend->frontend_lock);
 
+	if (of_find_property(dev->of_node, "interconnects", NULL)) {
+		/*
+		 * This assume we have the same DMA constraints for all our the
+		 * devices in our pipeline (all the backends, but also the
+		 * frontends). This sounds bad, but it has always been the case
+		 * for us, and DRM doesn't do per-device allocation either, so
+		 * we would need to fix DRM first...
+		 */
+		ret = of_dma_configure(drm->dev, dev->of_node, true);
+		if (ret)
+			return ret;
+	} else {
+		/*
+		 * If we don't have the interconnect property, most likely
+		 * because of an old DT, we need to set the DMA offset by hand
+		 * on our device since the RAM mapping is at 0 for the DMA bus,
+		 * unlike the CPU.
+		 */
+		drm->dev->dma_pfn_offset = PHYS_PFN_OFFSET;
+	}
+
 	backend->engine.node = dev->of_node;
 	backend->engine.ops = &sun4i_backend_engine_ops;
 	backend->engine.id = sun4i_backend_of_get_id(dev->of_node);
diff --git a/drivers/gpu/drm/sun4i/sun4i_hdmi.h b/drivers/gpu/drm/sun4i/sun4i_hdmi.h
index b685ee11623d..b08c4453d47c 100644
--- a/drivers/gpu/drm/sun4i/sun4i_hdmi.h
+++ b/drivers/gpu/drm/sun4i/sun4i_hdmi.h
@@ -269,6 +269,7 @@ struct sun4i_hdmi {
 	struct clk		*tmds_clk;
 
 	struct i2c_adapter	*i2c;
+	struct i2c_adapter	*ddc_i2c;
 
 	/* Regmap fields for I2C adapter */
 	struct regmap_field	*field_ddc_en;
diff --git a/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c b/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c
index d18862629301..8c122e637697 100644
--- a/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c
+++ b/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c
@@ -217,7 +217,7 @@ static int sun4i_hdmi_get_modes(struct drm_connector *connector)
 	struct edid *edid;
 	int ret;
 
-	edid = drm_get_edid(connector, hdmi->i2c);
+	edid = drm_get_edid(connector, hdmi->ddc_i2c ?: hdmi->i2c);
 	if (!edid)
 		return 0;
 
@@ -233,6 +233,28 @@ static int sun4i_hdmi_get_modes(struct drm_connector *connector)
 	return ret;
 }
 
+static struct i2c_adapter *sun4i_hdmi_get_ddc(struct device *dev)
+{
+	struct device_node *phandle, *remote;
+	struct i2c_adapter *ddc;
+
+	remote = of_graph_get_remote_node(dev->of_node, 1, -1);
+	if (!remote)
+		return ERR_PTR(-EINVAL);
+
+	phandle = of_parse_phandle(remote, "ddc-i2c-bus", 0);
+	of_node_put(remote);
+	if (!phandle)
+		return ERR_PTR(-ENODEV);
+
+	ddc = of_get_i2c_adapter_by_node(phandle);
+	of_node_put(phandle);
+	if (!ddc)
+		return ERR_PTR(-EPROBE_DEFER);
+
+	return ddc;
+}
+
 static const struct drm_connector_helper_funcs sun4i_hdmi_connector_helper_funcs = {
 	.get_modes	= sun4i_hdmi_get_modes,
 };
@@ -580,6 +602,15 @@ static int sun4i_hdmi_bind(struct device *dev, struct device *master,
 		goto err_disable_mod_clk;
 	}
 
+	hdmi->ddc_i2c = sun4i_hdmi_get_ddc(dev);
+	if (IS_ERR(hdmi->ddc_i2c)) {
+		ret = PTR_ERR(hdmi->ddc_i2c);
+		if (ret == -ENODEV)
+			hdmi->ddc_i2c = NULL;
+		else
+			goto err_del_i2c_adapter;
+	}
+
 	drm_encoder_helper_add(&hdmi->encoder,
 			       &sun4i_hdmi_helper_funcs);
 	ret = drm_encoder_init(drm,
@@ -589,14 +620,14 @@ static int sun4i_hdmi_bind(struct device *dev, struct device *master,
 			       NULL);
 	if (ret) {
 		dev_err(dev, "Couldn't initialise the HDMI encoder\n");
-		goto err_del_i2c_adapter;
+		goto err_put_ddc_i2c;
 	}
 
 	hdmi->encoder.possible_crtcs = drm_of_find_possible_crtcs(drm,
 								  dev->of_node);
 	if (!hdmi->encoder.possible_crtcs) {
 		ret = -EPROBE_DEFER;
-		goto err_del_i2c_adapter;
+		goto err_put_ddc_i2c;
 	}
 
 #ifdef CONFIG_DRM_SUN4I_HDMI_CEC
@@ -635,6 +666,8 @@ static int sun4i_hdmi_bind(struct device *dev, struct device *master,
 err_cleanup_connector:
 	cec_delete_adapter(hdmi->cec_adap);
 	drm_encoder_cleanup(&hdmi->encoder);
+err_put_ddc_i2c:
+	i2c_put_adapter(hdmi->ddc_i2c);
 err_del_i2c_adapter:
 	i2c_del_adapter(hdmi->i2c);
 err_disable_mod_clk:
@@ -655,6 +688,7 @@ static void sun4i_hdmi_unbind(struct device *dev, struct device *master,
 	drm_connector_cleanup(&hdmi->connector);
 	drm_encoder_cleanup(&hdmi->encoder);
 	i2c_del_adapter(hdmi->i2c);
+	i2c_put_adapter(hdmi->ddc_i2c);
 	clk_disable_unprepare(hdmi->mod_clk);
 	clk_disable_unprepare(hdmi->bus_clk);
 }
diff --git a/drivers/gpu/drm/sun4i/sun4i_lvds.c b/drivers/gpu/drm/sun4i/sun4i_lvds.c
index 147b97ed1a09..3a3ba99fed22 100644
--- a/drivers/gpu/drm/sun4i/sun4i_lvds.c
+++ b/drivers/gpu/drm/sun4i/sun4i_lvds.c
@@ -20,7 +20,7 @@ struct sun4i_lvds {
 	struct drm_connector	connector;
 	struct drm_encoder	encoder;
 
-	struct sun4i_tcon	*tcon;
+	struct drm_panel	*panel;
 };
 
 static inline struct sun4i_lvds *
@@ -41,9 +41,8 @@ static int sun4i_lvds_get_modes(struct drm_connector *connector)
 {
 	struct sun4i_lvds *lvds =
 		drm_connector_to_sun4i_lvds(connector);
-	struct sun4i_tcon *tcon = lvds->tcon;
 
-	return drm_panel_get_modes(tcon->panel);
+	return drm_panel_get_modes(lvds->panel);
 }
 
 static struct drm_connector_helper_funcs sun4i_lvds_con_helper_funcs = {
@@ -54,9 +53,8 @@ static void
 sun4i_lvds_connector_destroy(struct drm_connector *connector)
 {
 	struct sun4i_lvds *lvds = drm_connector_to_sun4i_lvds(connector);
-	struct sun4i_tcon *tcon = lvds->tcon;
 
-	drm_panel_detach(tcon->panel);
+	drm_panel_detach(lvds->panel);
 	drm_connector_cleanup(connector);
 }
 
@@ -71,26 +69,24 @@ static const struct drm_connector_funcs sun4i_lvds_con_funcs = {
 static void sun4i_lvds_encoder_enable(struct drm_encoder *encoder)
 {
 	struct sun4i_lvds *lvds = drm_encoder_to_sun4i_lvds(encoder);
-	struct sun4i_tcon *tcon = lvds->tcon;
 
 	DRM_DEBUG_DRIVER("Enabling LVDS output\n");
 
-	if (tcon->panel) {
-		drm_panel_prepare(tcon->panel);
-		drm_panel_enable(tcon->panel);
+	if (lvds->panel) {
+		drm_panel_prepare(lvds->panel);
+		drm_panel_enable(lvds->panel);
 	}
 }
 
 static void sun4i_lvds_encoder_disable(struct drm_encoder *encoder)
 {
 	struct sun4i_lvds *lvds = drm_encoder_to_sun4i_lvds(encoder);
-	struct sun4i_tcon *tcon = lvds->tcon;
 
 	DRM_DEBUG_DRIVER("Disabling LVDS output\n");
 
-	if (tcon->panel) {
-		drm_panel_disable(tcon->panel);
-		drm_panel_unprepare(tcon->panel);
+	if (lvds->panel) {
+		drm_panel_disable(lvds->panel);
+		drm_panel_unprepare(lvds->panel);
 	}
 }
 
@@ -113,11 +109,10 @@ int sun4i_lvds_init(struct drm_device *drm, struct sun4i_tcon *tcon)
 	lvds = devm_kzalloc(drm->dev, sizeof(*lvds), GFP_KERNEL);
 	if (!lvds)
 		return -ENOMEM;
-	lvds->tcon = tcon;
 	encoder = &lvds->encoder;
 
 	ret = drm_of_find_panel_or_bridge(tcon->dev->of_node, 1, 0,
-					  &tcon->panel, &bridge);
+					  &lvds->panel, &bridge);
 	if (ret) {
 		dev_info(drm->dev, "No panel or bridge found... LVDS output disabled\n");
 		return 0;
@@ -138,7 +133,7 @@ int sun4i_lvds_init(struct drm_device *drm, struct sun4i_tcon *tcon)
 	/* The LVDS encoder can only work with the TCON channel 0 */
 	lvds->encoder.possible_crtcs = drm_crtc_mask(&tcon->crtc->crtc);
 
-	if (tcon->panel) {
+	if (lvds->panel) {
 		drm_connector_helper_add(&lvds->connector,
 					 &sun4i_lvds_con_helper_funcs);
 		ret = drm_connector_init(drm, &lvds->connector,
@@ -152,7 +147,7 @@ int sun4i_lvds_init(struct drm_device *drm, struct sun4i_tcon *tcon)
 		drm_connector_attach_encoder(&lvds->connector,
 						  &lvds->encoder);
 
-		ret = drm_panel_attach(tcon->panel, &lvds->connector);
+		ret = drm_panel_attach(lvds->panel, &lvds->connector);
 		if (ret) {
 			dev_err(drm->dev, "Couldn't attach our panel\n");
 			goto err_cleanup_connector;
diff --git a/drivers/gpu/drm/sun4i/sun4i_rgb.c b/drivers/gpu/drm/sun4i/sun4i_rgb.c
index cae19e7bbeaa..d9e2502b49fa 100644
--- a/drivers/gpu/drm/sun4i/sun4i_rgb.c
+++ b/drivers/gpu/drm/sun4i/sun4i_rgb.c
@@ -27,6 +27,8 @@ struct sun4i_rgb {
 	struct drm_encoder	encoder;
 
 	struct sun4i_tcon	*tcon;
+	struct drm_panel	*panel;
+	struct drm_bridge	*bridge;
 };
 
 static inline struct sun4i_rgb *
@@ -47,11 +49,18 @@ static int sun4i_rgb_get_modes(struct drm_connector *connector)
 {
 	struct sun4i_rgb *rgb =
 		drm_connector_to_sun4i_rgb(connector);
-	struct sun4i_tcon *tcon = rgb->tcon;
 
-	return drm_panel_get_modes(tcon->panel);
+	return drm_panel_get_modes(rgb->panel);
 }
 
+/*
+ * VESA DMT defines a tolerance of 0.5% on the pixel clock, while the
+ * CVT spec reuses that tolerance in its examples, so it looks to be a
+ * good default tolerance for the EDID-based modes. Define it to 5 per
+ * mille to avoid floating point operations.
+ */
+#define SUN4I_RGB_DOTCLOCK_TOLERANCE_PER_MILLE	5
+
 static enum drm_mode_status sun4i_rgb_mode_valid(struct drm_encoder *crtc,
 						 const struct drm_display_mode *mode)
 {
@@ -59,8 +68,9 @@ static enum drm_mode_status sun4i_rgb_mode_valid(struct drm_encoder *crtc,
 	struct sun4i_tcon *tcon = rgb->tcon;
 	u32 hsync = mode->hsync_end - mode->hsync_start;
 	u32 vsync = mode->vsync_end - mode->vsync_start;
-	unsigned long rate = mode->clock * 1000;
-	long rounded_rate;
+	unsigned long long rate = mode->clock * 1000;
+	unsigned long long lowest, highest;
+	unsigned long long rounded_rate;
 
 	DRM_DEBUG_DRIVER("Validating modes...\n");
 
@@ -92,15 +102,39 @@ static enum drm_mode_status sun4i_rgb_mode_valid(struct drm_encoder *crtc,
 
 	DRM_DEBUG_DRIVER("Vertical parameters OK\n");
 
+	/*
+	 * TODO: We should use the struct display_timing if available
+	 * and / or trying to stretch the timings within that
+	 * tolerancy to take care of panels that we wouldn't be able
+	 * to have a exact match for.
+	 */
+	if (rgb->panel) {
+		DRM_DEBUG_DRIVER("RGB panel used, skipping clock rate checks");
+		goto out;
+	}
+
+	/*
+	 * That shouldn't ever happen unless something is really wrong, but it
+	 * doesn't harm to check.
+	 */
+	if (!rgb->bridge)
+		goto out;
+
 	tcon->dclk_min_div = 6;
 	tcon->dclk_max_div = 127;
 	rounded_rate = clk_round_rate(tcon->dclk, rate);
-	if (rounded_rate < rate)
+
+	lowest = rate * (1000 - SUN4I_RGB_DOTCLOCK_TOLERANCE_PER_MILLE);
+	do_div(lowest, 1000);
+	if (rounded_rate < lowest)
 		return MODE_CLOCK_LOW;
 
-	if (rounded_rate > rate)
+	highest = rate * (1000 + SUN4I_RGB_DOTCLOCK_TOLERANCE_PER_MILLE);
+	do_div(highest, 1000);
+	if (rounded_rate > highest)
 		return MODE_CLOCK_HIGH;
 
+out:
 	DRM_DEBUG_DRIVER("Clock rate OK\n");
 
 	return MODE_OK;
@@ -114,9 +148,8 @@ static void
 sun4i_rgb_connector_destroy(struct drm_connector *connector)
 {
 	struct sun4i_rgb *rgb = drm_connector_to_sun4i_rgb(connector);
-	struct sun4i_tcon *tcon = rgb->tcon;
 
-	drm_panel_detach(tcon->panel);
+	drm_panel_detach(rgb->panel);
 	drm_connector_cleanup(connector);
 }
 
@@ -131,26 +164,24 @@ static const struct drm_connector_funcs sun4i_rgb_con_funcs = {
 static void sun4i_rgb_encoder_enable(struct drm_encoder *encoder)
 {
 	struct sun4i_rgb *rgb = drm_encoder_to_sun4i_rgb(encoder);
-	struct sun4i_tcon *tcon = rgb->tcon;
 
 	DRM_DEBUG_DRIVER("Enabling RGB output\n");
 
-	if (tcon->panel) {
-		drm_panel_prepare(tcon->panel);
-		drm_panel_enable(tcon->panel);
+	if (rgb->panel) {
+		drm_panel_prepare(rgb->panel);
+		drm_panel_enable(rgb->panel);
 	}
 }
 
 static void sun4i_rgb_encoder_disable(struct drm_encoder *encoder)
 {
 	struct sun4i_rgb *rgb = drm_encoder_to_sun4i_rgb(encoder);
-	struct sun4i_tcon *tcon = rgb->tcon;
 
 	DRM_DEBUG_DRIVER("Disabling RGB output\n");
 
-	if (tcon->panel) {
-		drm_panel_disable(tcon->panel);
-		drm_panel_unprepare(tcon->panel);
+	if (rgb->panel) {
+		drm_panel_disable(rgb->panel);
+		drm_panel_unprepare(rgb->panel);
 	}
 }
 
@@ -172,7 +203,6 @@ static struct drm_encoder_funcs sun4i_rgb_enc_funcs = {
 int sun4i_rgb_init(struct drm_device *drm, struct sun4i_tcon *tcon)
 {
 	struct drm_encoder *encoder;
-	struct drm_bridge *bridge;
 	struct sun4i_rgb *rgb;
 	int ret;
 
@@ -183,7 +213,7 @@ int sun4i_rgb_init(struct drm_device *drm, struct sun4i_tcon *tcon)
 	encoder = &rgb->encoder;
 
 	ret = drm_of_find_panel_or_bridge(tcon->dev->of_node, 1, 0,
-					  &tcon->panel, &bridge);
+					  &rgb->panel, &rgb->bridge);
 	if (ret) {
 		dev_info(drm->dev, "No panel or bridge found... RGB output disabled\n");
 		return 0;
@@ -204,7 +234,7 @@ int sun4i_rgb_init(struct drm_device *drm, struct sun4i_tcon *tcon)
 	/* The RGB encoder can only work with the TCON channel 0 */
 	rgb->encoder.possible_crtcs = drm_crtc_mask(&tcon->crtc->crtc);
 
-	if (tcon->panel) {
+	if (rgb->panel) {
 		drm_connector_helper_add(&rgb->connector,
 					 &sun4i_rgb_con_helper_funcs);
 		ret = drm_connector_init(drm, &rgb->connector,
@@ -218,15 +248,15 @@ int sun4i_rgb_init(struct drm_device *drm, struct sun4i_tcon *tcon)
 		drm_connector_attach_encoder(&rgb->connector,
 						  &rgb->encoder);
 
-		ret = drm_panel_attach(tcon->panel, &rgb->connector);
+		ret = drm_panel_attach(rgb->panel, &rgb->connector);
 		if (ret) {
 			dev_err(drm->dev, "Couldn't attach our panel\n");
 			goto err_cleanup_connector;
 		}
 	}
 
-	if (bridge) {
-		ret = drm_bridge_attach(encoder, bridge, NULL);
+	if (rgb->bridge) {
+		ret = drm_bridge_attach(encoder, rgb->bridge, NULL);
 		if (ret) {
 			dev_err(drm->dev, "Couldn't attach our bridge\n");
 			goto err_cleanup_connector;
diff --git a/drivers/gpu/drm/sun4i/sun4i_tcon.c b/drivers/gpu/drm/sun4i/sun4i_tcon.c
index 7136fc91c603..9d8d8124b1f6 100644
--- a/drivers/gpu/drm/sun4i/sun4i_tcon.c
+++ b/drivers/gpu/drm/sun4i/sun4i_tcon.c
@@ -236,8 +236,8 @@ static struct sun4i_tcon *sun4i_get_tcon0(struct drm_device *drm)
 	return NULL;
 }
 
-void sun4i_tcon_set_mux(struct sun4i_tcon *tcon, int channel,
-			const struct drm_encoder *encoder)
+static void sun4i_tcon_set_mux(struct sun4i_tcon *tcon, int channel,
+			       const struct drm_encoder *encoder)
 {
 	int ret = -ENOTSUPP;
 
@@ -341,8 +341,8 @@ static void sun4i_tcon0_mode_set_cpu(struct sun4i_tcon *tcon,
 	u32 block_space, start_delay;
 	u32 tcon_div;
 
-	tcon->dclk_min_div = 4;
-	tcon->dclk_max_div = 127;
+	tcon->dclk_min_div = SUN6I_DSI_TCON_DIV;
+	tcon->dclk_max_div = SUN6I_DSI_TCON_DIV;
 
 	sun4i_tcon0_mode_set_common(tcon, mode);
 
@@ -561,10 +561,10 @@ static void sun4i_tcon0_mode_set_rgb(struct sun4i_tcon *tcon,
 	 * Following code is a way to avoid quirks all around TCON
 	 * and DOTCLOCK drivers.
 	 */
-	if (display_info.bus_flags & DRM_BUS_FLAG_PIXDATA_POSEDGE)
+	if (display_info.bus_flags & DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE)
 		clk_set_phase(tcon->dclk, 240);
 
-	if (display_info.bus_flags & DRM_BUS_FLAG_PIXDATA_NEGEDGE)
+	if (display_info.bus_flags & DRM_BUS_FLAG_PIXDATA_DRIVE_NEGEDGE)
 		clk_set_phase(tcon->dclk, 0);
 
 	regmap_update_bits(tcon->regs, SUN4I_TCON0_IO_POL_REG,
diff --git a/drivers/gpu/drm/sun4i/sun4i_tcon.h b/drivers/gpu/drm/sun4i/sun4i_tcon.h
index b5214d71610f..84cfb1952ff7 100644
--- a/drivers/gpu/drm/sun4i/sun4i_tcon.h
+++ b/drivers/gpu/drm/sun4i/sun4i_tcon.h
@@ -257,8 +257,6 @@ struct sun4i_tcon {
 	struct reset_control		*lcd_rst;
 	struct reset_control		*lvds_rst;
 
-	struct drm_panel		*panel;
-
 	/* Platform adjustments */
 	const struct sun4i_tcon_quirks	*quirks;
 
diff --git a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c
index 318994cd1b85..6ff585055a07 100644
--- a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c
+++ b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c
@@ -24,7 +24,9 @@
 #include <drm/drm_panel.h>
 #include <drm/drm_probe_helper.h>
 
+#include "sun4i_crtc.h"
 #include "sun4i_drv.h"
+#include "sun4i_tcon.h"
 #include "sun6i_mipi_dsi.h"
 
 #include <video/mipi_display.h>
@@ -33,6 +35,8 @@
 #define SUN6I_DSI_CTL_EN			BIT(0)
 
 #define SUN6I_DSI_BASIC_CTL_REG		0x00c
+#define SUN6I_DSI_BASIC_CTL_TRAIL_INV(n)		(((n) & 0xf) << 4)
+#define SUN6I_DSI_BASIC_CTL_TRAIL_FILL		BIT(3)
 #define SUN6I_DSI_BASIC_CTL_HBP_DIS		BIT(2)
 #define SUN6I_DSI_BASIC_CTL_HSA_HSE_DIS		BIT(1)
 #define SUN6I_DSI_BASIC_CTL_VIDEO_BURST		BIT(0)
@@ -153,6 +157,8 @@
 
 #define SUN6I_DSI_CMD_TX_REG(n)		(0x300 + (n) * 0x04)
 
+#define SUN6I_DSI_SYNC_POINT		40
+
 enum sun6i_dsi_start_inst {
 	DSI_START_LPRX,
 	DSI_START_LPTX,
@@ -358,7 +364,54 @@ static void sun6i_dsi_inst_init(struct sun6i_dsi *dsi,
 static u16 sun6i_dsi_get_video_start_delay(struct sun6i_dsi *dsi,
 					   struct drm_display_mode *mode)
 {
-	return mode->vtotal - (mode->vsync_end - mode->vdisplay) + 1;
+	u16 start = clamp(mode->vtotal - mode->vdisplay - 10, 8, 100);
+	u16 delay = mode->vtotal - (mode->vsync_end - mode->vdisplay) + start;
+
+	if (delay > mode->vtotal)
+		delay = delay % mode->vtotal;
+
+	return max_t(u16, delay, 1);
+}
+
+static u16 sun6i_dsi_get_line_num(struct sun6i_dsi *dsi,
+				  struct drm_display_mode *mode)
+{
+	struct mipi_dsi_device *device = dsi->device;
+	unsigned int Bpp = mipi_dsi_pixel_format_to_bpp(device->format) / 8;
+
+	return mode->htotal * Bpp / device->lanes;
+}
+
+static u16 sun6i_dsi_get_drq_edge0(struct sun6i_dsi *dsi,
+				   struct drm_display_mode *mode,
+				   u16 line_num, u16 edge1)
+{
+	u16 edge0 = edge1;
+
+	edge0 += (mode->hdisplay + 40) * SUN6I_DSI_TCON_DIV / 8;
+
+	if (edge0 > line_num)
+		return edge0 - line_num;
+
+	return 1;
+}
+
+static u16 sun6i_dsi_get_drq_edge1(struct sun6i_dsi *dsi,
+				   struct drm_display_mode *mode,
+				   u16 line_num)
+{
+	struct mipi_dsi_device *device = dsi->device;
+	unsigned int Bpp = mipi_dsi_pixel_format_to_bpp(device->format) / 8;
+	unsigned int hbp = mode->htotal - mode->hsync_end;
+	u16 edge1;
+
+	edge1 = SUN6I_DSI_SYNC_POINT;
+	edge1 += (mode->hdisplay + hbp + 20) * Bpp / device->lanes;
+
+	if (edge1 > line_num)
+		return line_num;
+
+	return edge1;
 }
 
 static void sun6i_dsi_setup_burst(struct sun6i_dsi *dsi,
@@ -367,7 +420,23 @@ static void sun6i_dsi_setup_burst(struct sun6i_dsi *dsi,
 	struct mipi_dsi_device *device = dsi->device;
 	u32 val = 0;
 
-	if ((mode->hsync_end - mode->hdisplay) > 20) {
+	if (device->mode_flags & MIPI_DSI_MODE_VIDEO_BURST) {
+		u16 line_num = sun6i_dsi_get_line_num(dsi, mode);
+		u16 edge0, edge1;
+
+		edge1 = sun6i_dsi_get_drq_edge1(dsi, mode, line_num);
+		edge0 = sun6i_dsi_get_drq_edge0(dsi, mode, line_num, edge1);
+
+		regmap_write(dsi->regs, SUN6I_DSI_BURST_DRQ_REG,
+			     SUN6I_DSI_BURST_DRQ_EDGE0(edge0) |
+			     SUN6I_DSI_BURST_DRQ_EDGE1(edge1));
+
+		regmap_write(dsi->regs, SUN6I_DSI_BURST_LINE_REG,
+			     SUN6I_DSI_BURST_LINE_NUM(line_num) |
+			     SUN6I_DSI_BURST_LINE_SYNC_POINT(SUN6I_DSI_SYNC_POINT));
+
+		val = SUN6I_DSI_TCON_DRQ_ENABLE_MODE;
+	} else if ((mode->hsync_end - mode->hdisplay) > 20) {
 		/* Maaaaaagic */
 		u16 drq = (mode->hsync_end - mode->hdisplay) - 20;
 
@@ -384,8 +453,19 @@ static void sun6i_dsi_setup_burst(struct sun6i_dsi *dsi,
 static void sun6i_dsi_setup_inst_loop(struct sun6i_dsi *dsi,
 				      struct drm_display_mode *mode)
 {
+	struct mipi_dsi_device *device = dsi->device;
 	u16 delay = 50 - 1;
 
+	if (device->mode_flags & MIPI_DSI_MODE_VIDEO_BURST) {
+		delay = (mode->htotal - mode->hdisplay) * 150;
+		delay /= (mode->clock / 1000) * 8;
+		delay -= 50;
+	}
+
+	regmap_write(dsi->regs, SUN6I_DSI_INST_LOOP_SEL_REG,
+		     2 << (4 * DSI_INST_ID_LP11) |
+		     3 << (4 * DSI_INST_ID_DLY));
+
 	regmap_write(dsi->regs, SUN6I_DSI_INST_LOOP_NUM_REG(0),
 		     SUN6I_DSI_INST_LOOP_NUM_N0(50 - 1) |
 		     SUN6I_DSI_INST_LOOP_NUM_N1(delay));
@@ -451,49 +531,68 @@ static void sun6i_dsi_setup_timings(struct sun6i_dsi *dsi,
 {
 	struct mipi_dsi_device *device = dsi->device;
 	unsigned int Bpp = mipi_dsi_pixel_format_to_bpp(device->format) / 8;
-	u16 hbp, hfp, hsa, hblk, vblk;
+	u16 hbp = 0, hfp = 0, hsa = 0, hblk = 0, vblk = 0;
+	u32 basic_ctl = 0;
 	size_t bytes;
 	u8 *buffer;
 
 	/* Do all timing calculations up front to allocate buffer space */
 
-	/*
-	 * A sync period is composed of a blanking packet (4 bytes +
-	 * payload + 2 bytes) and a sync event packet (4 bytes). Its
-	 * minimal size is therefore 10 bytes
-	 */
-#define HSA_PACKET_OVERHEAD	10
-	hsa = max((unsigned int)HSA_PACKET_OVERHEAD,
-		  (mode->hsync_end - mode->hsync_start) * Bpp - HSA_PACKET_OVERHEAD);
+	if (device->mode_flags & MIPI_DSI_MODE_VIDEO_BURST) {
+		hblk = mode->hdisplay * Bpp;
+		basic_ctl = SUN6I_DSI_BASIC_CTL_VIDEO_BURST |
+			    SUN6I_DSI_BASIC_CTL_HSA_HSE_DIS |
+			    SUN6I_DSI_BASIC_CTL_HBP_DIS;
 
-	/*
-	 * The backporch is set using a blanking packet (4 bytes +
-	 * payload + 2 bytes). Its minimal size is therefore 6 bytes
-	 */
+		if (device->lanes == 4)
+			basic_ctl |= SUN6I_DSI_BASIC_CTL_TRAIL_FILL |
+				     SUN6I_DSI_BASIC_CTL_TRAIL_INV(0xc);
+	} else {
+		/*
+		 * A sync period is composed of a blanking packet (4
+		 * bytes + payload + 2 bytes) and a sync event packet
+		 * (4 bytes). Its minimal size is therefore 10 bytes
+		 */
+#define HSA_PACKET_OVERHEAD	10
+		hsa = max((unsigned int)HSA_PACKET_OVERHEAD,
+			  (mode->hsync_end - mode->hsync_start) * Bpp - HSA_PACKET_OVERHEAD);
+
+		/*
+		 * The backporch is set using a blanking packet (4
+		 * bytes + payload + 2 bytes). Its minimal size is
+		 * therefore 6 bytes
+		 */
 #define HBP_PACKET_OVERHEAD	6
-	hbp = max((unsigned int)HBP_PACKET_OVERHEAD,
-		  (mode->hsync_start - mode->hdisplay) * Bpp - HBP_PACKET_OVERHEAD);
-
-	/*
-	 * The frontporch is set using a blanking packet (4 bytes +
-	 * payload + 2 bytes). Its minimal size is therefore 6 bytes
-	 */
+		hbp = max((unsigned int)HBP_PACKET_OVERHEAD,
+			  (mode->htotal - mode->hsync_end) * Bpp - HBP_PACKET_OVERHEAD);
+
+		/*
+		 * The frontporch is set using a blanking packet (4
+		 * bytes + payload + 2 bytes). Its minimal size is
+		 * therefore 6 bytes
+		 */
 #define HFP_PACKET_OVERHEAD	6
-	hfp = max((unsigned int)HFP_PACKET_OVERHEAD,
-		  (mode->htotal - mode->hsync_end) * Bpp - HFP_PACKET_OVERHEAD);
-
-	/*
-	 * hblk seems to be the line + porches length.
-	 */
-	hblk = mode->htotal * Bpp - hsa;
-
-	/*
-	 * And I'm not entirely sure what vblk is about. The driver in
-	 * Allwinner BSP is using a rather convoluted calculation
-	 * there only for 4 lanes. However, using 0 (the !4 lanes
-	 * case) even with a 4 lanes screen seems to work...
-	 */
-	vblk = 0;
+		hfp = max((unsigned int)HFP_PACKET_OVERHEAD,
+			  (mode->hsync_start - mode->hdisplay) * Bpp - HFP_PACKET_OVERHEAD);
+
+		/*
+		 * The blanking is set using a sync event (4 bytes)
+		 * and a blanking packet (4 bytes + payload + 2
+		 * bytes). Its minimal size is therefore 10 bytes.
+		 */
+#define HBLK_PACKET_OVERHEAD	10
+		hblk = max((unsigned int)HBLK_PACKET_OVERHEAD,
+			   (mode->htotal - (mode->hsync_end - mode->hsync_start)) * Bpp -
+			   HBLK_PACKET_OVERHEAD);
+
+		/*
+		 * And I'm not entirely sure what vblk is about. The driver in
+		 * Allwinner BSP is using a rather convoluted calculation
+		 * there only for 4 lanes. However, using 0 (the !4 lanes
+		 * case) even with a 4 lanes screen seems to work...
+		 */
+		vblk = 0;
+	}
 
 	/* How many bytes do we need to send all payloads? */
 	bytes = max_t(size_t, max(max(hfp, hblk), max(hsa, hbp)), vblk);
@@ -501,7 +600,7 @@ static void sun6i_dsi_setup_timings(struct sun6i_dsi *dsi,
 	if (WARN_ON(!buffer))
 		return;
 
-	regmap_write(dsi->regs, SUN6I_DSI_BASIC_CTL_REG, 0);
+	regmap_write(dsi->regs, SUN6I_DSI_BASIC_CTL_REG, basic_ctl);
 
 	regmap_write(dsi->regs, SUN6I_DSI_SYNC_HSS_REG,
 		     sun6i_dsi_build_sync_pkt(MIPI_DSI_H_SYNC_START,
@@ -526,8 +625,8 @@ static void sun6i_dsi_setup_timings(struct sun6i_dsi *dsi,
 	regmap_write(dsi->regs, SUN6I_DSI_BASIC_SIZE0_REG,
 		     SUN6I_DSI_BASIC_SIZE0_VSA(mode->vsync_end -
 					       mode->vsync_start) |
-		     SUN6I_DSI_BASIC_SIZE0_VBP(mode->vsync_start -
-					       mode->vdisplay));
+		     SUN6I_DSI_BASIC_SIZE0_VBP(mode->vtotal -
+					       mode->vsync_end));
 
 	regmap_write(dsi->regs, SUN6I_DSI_BASIC_SIZE1_REG,
 		     SUN6I_DSI_BASIC_SIZE1_VACT(mode->vdisplay) |
diff --git a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.h b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.h
index a07090579f84..5c3ad5be0690 100644
--- a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.h
+++ b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.h
@@ -13,6 +13,8 @@
 #include <drm/drm_encoder.h>
 #include <drm/drm_mipi_dsi.h>
 
+#define SUN6I_DSI_TCON_DIV	4
+
 struct sun6i_dsi {
 	struct drm_connector	connector;
 	struct drm_encoder	encoder;
diff --git a/drivers/gpu/drm/sun4i/sun8i_mixer.c b/drivers/gpu/drm/sun4i/sun8i_mixer.c
index 30a2eff55687..fd20a928cf4d 100644
--- a/drivers/gpu/drm/sun4i/sun8i_mixer.c
+++ b/drivers/gpu/drm/sun4i/sun8i_mixer.c
@@ -325,38 +325,22 @@ static struct regmap_config sun8i_mixer_regmap_config = {
 
 static int sun8i_mixer_of_get_id(struct device_node *node)
 {
-	struct device_node *port, *ep;
-	int ret = -EINVAL;
+	struct device_node *ep, *remote;
+	struct of_endpoint of_ep;
 
-	/* output is port 1 */
-	port = of_graph_get_port_by_id(node, 1);
-	if (!port)
+	/* Output port is 1, and we want the first endpoint. */
+	ep = of_graph_get_endpoint_by_regs(node, 1, -1);
+	if (!ep)
 		return -EINVAL;
 
-	/* try to find downstream endpoint */
-	for_each_available_child_of_node(port, ep) {
-		struct device_node *remote;
-		u32 reg;
-
-		remote = of_graph_get_remote_endpoint(ep);
-		if (!remote)
-			continue;
-
-		ret = of_property_read_u32(remote, "reg", &reg);
-		if (!ret) {
-			of_node_put(remote);
-			of_node_put(ep);
-			of_node_put(port);
-
-			return reg;
-		}
-
-		of_node_put(remote);
-	}
-
-	of_node_put(port);
+	remote = of_graph_get_remote_endpoint(ep);
+	of_node_put(ep);
+	if (!remote)
+		return -EINVAL;
 
-	return ret;
+	of_graph_parse_endpoint(remote, &of_ep);
+	of_node_put(remote);
+	return of_ep.id;
 }
 
 static int sun8i_mixer_bind(struct device *dev, struct device *master,
@@ -554,6 +538,7 @@ static int sun8i_mixer_remove(struct platform_device *pdev)
 static const struct sun8i_mixer_cfg sun8i_a83t_mixer0_cfg = {
 	.ccsc		= 0,
 	.scaler_mask	= 0xf,
+	.scanline_yuv	= 2048,
 	.ui_num		= 3,
 	.vi_num		= 1,
 };
@@ -561,6 +546,7 @@ static const struct sun8i_mixer_cfg sun8i_a83t_mixer0_cfg = {
 static const struct sun8i_mixer_cfg sun8i_a83t_mixer1_cfg = {
 	.ccsc		= 1,
 	.scaler_mask	= 0x3,
+	.scanline_yuv	= 2048,
 	.ui_num		= 1,
 	.vi_num		= 1,
 };
@@ -569,6 +555,7 @@ static const struct sun8i_mixer_cfg sun8i_h3_mixer0_cfg = {
 	.ccsc		= 0,
 	.mod_rate	= 432000000,
 	.scaler_mask	= 0xf,
+	.scanline_yuv	= 2048,
 	.ui_num		= 3,
 	.vi_num		= 1,
 };
@@ -577,6 +564,7 @@ static const struct sun8i_mixer_cfg sun8i_r40_mixer0_cfg = {
 	.ccsc		= 0,
 	.mod_rate	= 297000000,
 	.scaler_mask	= 0xf,
+	.scanline_yuv	= 2048,
 	.ui_num		= 3,
 	.vi_num		= 1,
 };
@@ -585,6 +573,7 @@ static const struct sun8i_mixer_cfg sun8i_r40_mixer1_cfg = {
 	.ccsc		= 1,
 	.mod_rate	= 297000000,
 	.scaler_mask	= 0x3,
+	.scanline_yuv	= 2048,
 	.ui_num		= 1,
 	.vi_num		= 1,
 };
@@ -593,6 +582,7 @@ static const struct sun8i_mixer_cfg sun8i_v3s_mixer_cfg = {
 	.vi_num = 2,
 	.ui_num = 1,
 	.scaler_mask = 0x3,
+	.scanline_yuv = 2048,
 	.ccsc = 0,
 	.mod_rate = 150000000,
 };
@@ -601,6 +591,7 @@ static const struct sun8i_mixer_cfg sun50i_a64_mixer0_cfg = {
 	.ccsc		= 0,
 	.mod_rate	= 297000000,
 	.scaler_mask	= 0xf,
+	.scanline_yuv	= 4096,
 	.ui_num		= 3,
 	.vi_num		= 1,
 };
@@ -609,6 +600,7 @@ static const struct sun8i_mixer_cfg sun50i_a64_mixer1_cfg = {
 	.ccsc		= 1,
 	.mod_rate	= 297000000,
 	.scaler_mask	= 0x3,
+	.scanline_yuv	= 2048,
 	.ui_num		= 1,
 	.vi_num		= 1,
 };
@@ -618,6 +610,7 @@ static const struct sun8i_mixer_cfg sun50i_h6_mixer0_cfg = {
 	.is_de3		= true,
 	.mod_rate	= 600000000,
 	.scaler_mask	= 0xf,
+	.scanline_yuv	= 4096,
 	.ui_num		= 3,
 	.vi_num		= 1,
 };
diff --git a/drivers/gpu/drm/sun4i/sun8i_mixer.h b/drivers/gpu/drm/sun4i/sun8i_mixer.h
index 913d14ce68b0..80e084caa084 100644
--- a/drivers/gpu/drm/sun4i/sun8i_mixer.h
+++ b/drivers/gpu/drm/sun4i/sun8i_mixer.h
@@ -159,6 +159,7 @@ struct de2_fmt_info {
  * @mod_rate: module clock rate that needs to be set in order to have
  *	a functional block.
  * @is_de3: true, if this is next gen display engine 3.0, false otherwise.
+ * @scaline_yuv: size of a scanline for VI scaler for YUV formats.
  */
 struct sun8i_mixer_cfg {
 	int		vi_num;
@@ -167,6 +168,7 @@ struct sun8i_mixer_cfg {
 	int		ccsc;
 	unsigned long	mod_rate;
 	unsigned int	is_de3 : 1;
+	unsigned int	scanline_yuv;
 };
 
 struct sun8i_mixer {
diff --git a/drivers/gpu/drm/sun4i/sun8i_tcon_top.c b/drivers/gpu/drm/sun4i/sun8i_tcon_top.c
index b1e7c76e9c17..3267d0f9b9b2 100644
--- a/drivers/gpu/drm/sun4i/sun8i_tcon_top.c
+++ b/drivers/gpu/drm/sun4i/sun8i_tcon_top.c
@@ -269,12 +269,12 @@ static int sun8i_tcon_top_remove(struct platform_device *pdev)
 	return 0;
 }
 
-const struct sun8i_tcon_top_quirks sun8i_r40_tcon_top_quirks = {
+static const struct sun8i_tcon_top_quirks sun8i_r40_tcon_top_quirks = {
 	.has_tcon_tv1	= true,
 	.has_dsi	= true,
 };
 
-const struct sun8i_tcon_top_quirks sun50i_h6_tcon_top_quirks = {
+static const struct sun8i_tcon_top_quirks sun50i_h6_tcon_top_quirks = {
 	/* Nothing special */
 };
 
diff --git a/drivers/gpu/drm/sun4i/sun8i_vi_layer.c b/drivers/gpu/drm/sun4i/sun8i_vi_layer.c
index 8a0616238467..bb8e026d6405 100644
--- a/drivers/gpu/drm/sun4i/sun8i_vi_layer.c
+++ b/drivers/gpu/drm/sun4i/sun8i_vi_layer.c
@@ -80,6 +80,8 @@ static int sun8i_vi_layer_update_coord(struct sun8i_mixer *mixer, int channel,
 	u32 bld_base, ch_base;
 	u32 outsize, insize;
 	u32 hphase, vphase;
+	u32 hn = 0, hm = 0;
+	u32 vn = 0, vm = 0;
 	bool subsampled;
 
 	DRM_DEBUG_DRIVER("Updating VI channel %d overlay %d\n",
@@ -137,12 +139,41 @@ static int sun8i_vi_layer_update_coord(struct sun8i_mixer *mixer, int channel,
 	subsampled = format->hsub > 1 || format->vsub > 1;
 
 	if (insize != outsize || subsampled || hphase || vphase) {
-		u32 hscale, vscale;
+		unsigned int scanline, required;
+		struct drm_display_mode *mode;
+		u32 hscale, vscale, fps;
+		u64 ability;
 
 		DRM_DEBUG_DRIVER("HW scaling is enabled\n");
 
-		hscale = state->src_w / state->crtc_w;
-		vscale = state->src_h / state->crtc_h;
+		mode = &plane->state->crtc->state->mode;
+		fps = (mode->clock * 1000) / (mode->vtotal * mode->htotal);
+		ability = clk_get_rate(mixer->mod_clk);
+		/* BSP algorithm assumes 80% efficiency of VI scaler unit */
+		ability *= 80;
+		do_div(ability, mode->vdisplay * fps * max(src_w, dst_w));
+
+		required = src_h * 100 / dst_h;
+
+		if (ability < required) {
+			DRM_DEBUG_DRIVER("Using vertical coarse scaling\n");
+			vm = src_h;
+			vn = (u32)ability * dst_h / 100;
+			src_h = vn;
+		}
+
+		/* it seems that every RGB scaler has buffer for 2048 pixels */
+		scanline = subsampled ? mixer->cfg->scanline_yuv : 2048;
+
+		if (src_w > scanline) {
+			DRM_DEBUG_DRIVER("Using horizontal coarse scaling\n");
+			hm = src_w;
+			hn = scanline;
+			src_w = hn;
+		}
+
+		hscale = (src_w << 16) / dst_w;
+		vscale = (src_h << 16) / dst_h;
 
 		sun8i_vi_scaler_setup(mixer, channel, src_w, src_h, dst_w,
 				      dst_h, hscale, vscale, hphase, vphase,
@@ -153,6 +184,23 @@ static int sun8i_vi_layer_update_coord(struct sun8i_mixer *mixer, int channel,
 		sun8i_vi_scaler_enable(mixer, channel, false);
 	}
 
+	regmap_write(mixer->engine.regs,
+		     SUN8I_MIXER_CHAN_VI_HDS_Y(ch_base),
+		     SUN8I_MIXER_CHAN_VI_DS_N(hn) |
+		     SUN8I_MIXER_CHAN_VI_DS_M(hm));
+	regmap_write(mixer->engine.regs,
+		     SUN8I_MIXER_CHAN_VI_HDS_UV(ch_base),
+		     SUN8I_MIXER_CHAN_VI_DS_N(hn) |
+		     SUN8I_MIXER_CHAN_VI_DS_M(hm));
+	regmap_write(mixer->engine.regs,
+		     SUN8I_MIXER_CHAN_VI_VDS_Y(ch_base),
+		     SUN8I_MIXER_CHAN_VI_DS_N(vn) |
+		     SUN8I_MIXER_CHAN_VI_DS_M(vm));
+	regmap_write(mixer->engine.regs,
+		     SUN8I_MIXER_CHAN_VI_VDS_UV(ch_base),
+		     SUN8I_MIXER_CHAN_VI_DS_N(vn) |
+		     SUN8I_MIXER_CHAN_VI_DS_M(vm));
+
 	/* Set base coordinates */
 	DRM_DEBUG_DRIVER("Layer destination coordinates X: %d Y: %d\n",
 			 state->dst.x1, state->dst.y1);
diff --git a/drivers/gpu/drm/sun4i/sun8i_vi_layer.h b/drivers/gpu/drm/sun4i/sun8i_vi_layer.h
index 8a5e6d01c85d..a223a4839f45 100644
--- a/drivers/gpu/drm/sun4i/sun8i_vi_layer.h
+++ b/drivers/gpu/drm/sun4i/sun8i_vi_layer.h
@@ -24,6 +24,14 @@
 		((base) + 0x30 * (layer) + 0x18 + 4 * (plane))
 #define SUN8I_MIXER_CHAN_VI_OVL_SIZE(base) \
 		((base) + 0xe8)
+#define SUN8I_MIXER_CHAN_VI_HDS_Y(base) \
+		((base) + 0xf0)
+#define SUN8I_MIXER_CHAN_VI_HDS_UV(base) \
+		((base) + 0xf4)
+#define SUN8I_MIXER_CHAN_VI_VDS_Y(base) \
+		((base) + 0xf8)
+#define SUN8I_MIXER_CHAN_VI_VDS_UV(base) \
+		((base) + 0xfc)
 
 #define SUN8I_MIXER_CHAN_VI_LAYER_ATTR_EN		BIT(0)
 /* RGB mode should be set for RGB formats and cleared for YCbCr */
@@ -33,6 +41,9 @@
 #define SUN50I_MIXER_CHAN_VI_LAYER_ATTR_ALPHA_MASK	GENMASK(31, 24)
 #define SUN50I_MIXER_CHAN_VI_LAYER_ATTR_ALPHA(x)	((x) << 24)
 
+#define SUN8I_MIXER_CHAN_VI_DS_N(x)			((x) << 16)
+#define SUN8I_MIXER_CHAN_VI_DS_M(x)			((x) << 0)
+
 struct sun8i_mixer;
 
 struct sun8i_vi_layer {
diff --git a/drivers/gpu/drm/tegra/fb.c b/drivers/gpu/drm/tegra/fb.c
index 0a4ce05e00ab..1dd83a757dba 100644
--- a/drivers/gpu/drm/tegra/fb.c
+++ b/drivers/gpu/drm/tegra/fb.c
@@ -255,11 +255,9 @@ static int tegra_fbdev_probe(struct drm_fb_helper *helper,
 	helper->fb = fb;
 	helper->fbdev = info;
 
-	info->par = helper;
 	info->fbops = &tegra_fb_ops;
 
-	drm_fb_helper_fill_fix(info, fb->pitches[0], fb->format->depth);
-	drm_fb_helper_fill_var(info, helper, fb->width, fb->height);
+	drm_fb_helper_fill_info(info, helper, sizes);
 
 	offset = info->var.xoffset * bytes_per_pixel +
 		 info->var.yoffset * fb->pitches[0];
diff --git a/drivers/gpu/drm/tegra/gem.c b/drivers/gpu/drm/tegra/gem.c
index 4f80100ff5f3..4cce11fd8836 100644
--- a/drivers/gpu/drm/tegra/gem.c
+++ b/drivers/gpu/drm/tegra/gem.c
@@ -204,7 +204,7 @@ static void tegra_bo_free(struct drm_device *drm, struct tegra_bo *bo)
 {
 	if (bo->pages) {
 		dma_unmap_sg(drm->dev, bo->sgt->sgl, bo->sgt->nents,
-			     DMA_BIDIRECTIONAL);
+			     DMA_FROM_DEVICE);
 		drm_gem_put_pages(&bo->gem, bo->pages, true, true);
 		sg_free_table(bo->sgt);
 		kfree(bo->sgt);
@@ -230,7 +230,7 @@ static int tegra_bo_get_pages(struct drm_device *drm, struct tegra_bo *bo)
 	}
 
 	err = dma_map_sg(drm->dev, bo->sgt->sgl, bo->sgt->nents,
-			 DMA_BIDIRECTIONAL);
+			 DMA_FROM_DEVICE);
 	if (err == 0) {
 		err = -EFAULT;
 		goto free_sgt;
diff --git a/drivers/gpu/drm/tegra/sor.c b/drivers/gpu/drm/tegra/sor.c
index 40057106f5f3..5be5a0817dfe 100644
--- a/drivers/gpu/drm/tegra/sor.c
+++ b/drivers/gpu/drm/tegra/sor.c
@@ -2871,6 +2871,13 @@ static int tegra_sor_init(struct host1x_client *client)
 	 * kernel is possible.
 	 */
 	if (sor->rst) {
+		err = reset_control_acquire(sor->rst);
+		if (err < 0) {
+			dev_err(sor->dev, "failed to acquire SOR reset: %d\n",
+				err);
+			return err;
+		}
+
 		err = reset_control_assert(sor->rst);
 		if (err < 0) {
 			dev_err(sor->dev, "failed to assert SOR reset: %d\n",
@@ -2894,6 +2901,8 @@ static int tegra_sor_init(struct host1x_client *client)
 				err);
 			return err;
 		}
+
+		reset_control_release(sor->rst);
 	}
 
 	err = clk_prepare_enable(sor->clk_safe);
@@ -3331,7 +3340,7 @@ static int tegra_sor_probe(struct platform_device *pdev)
 		goto remove;
 	}
 
-	sor->rst = devm_reset_control_get(&pdev->dev, "sor");
+	sor->rst = devm_reset_control_get_exclusive_released(&pdev->dev, "sor");
 	if (IS_ERR(sor->rst)) {
 		err = PTR_ERR(sor->rst);
 
@@ -3519,6 +3528,8 @@ static int tegra_sor_suspend(struct device *dev)
 			dev_err(dev, "failed to assert reset: %d\n", err);
 			return err;
 		}
+
+		reset_control_release(sor->rst);
 	}
 
 	usleep_range(1000, 2000);
@@ -3542,9 +3553,17 @@ static int tegra_sor_resume(struct device *dev)
 	usleep_range(1000, 2000);
 
 	if (sor->rst) {
+		err = reset_control_acquire(sor->rst);
+		if (err < 0) {
+			dev_err(dev, "failed to acquire reset: %d\n", err);
+			clk_disable_unprepare(sor->clk);
+			return err;
+		}
+
 		err = reset_control_deassert(sor->rst);
 		if (err < 0) {
 			dev_err(dev, "failed to deassert reset: %d\n", err);
+			reset_control_release(sor->rst);
 			clk_disable_unprepare(sor->clk);
 			return err;
 		}
diff --git a/drivers/gpu/drm/tinydrm/core/Makefile b/drivers/gpu/drm/tinydrm/core/Makefile
index fb221e6f8885..6f8f764560e0 100644
--- a/drivers/gpu/drm/tinydrm/core/Makefile
+++ b/drivers/gpu/drm/tinydrm/core/Makefile
@@ -1,3 +1,3 @@
-tinydrm-y := tinydrm-core.o tinydrm-pipe.o tinydrm-helpers.o
+tinydrm-y := tinydrm-pipe.o tinydrm-helpers.o
 
 obj-$(CONFIG_DRM_TINYDRM) += tinydrm.o
diff --git a/drivers/gpu/drm/tinydrm/core/tinydrm-core.c b/drivers/gpu/drm/tinydrm/core/tinydrm-core.c
deleted file mode 100644
index 554abd5d3b53..000000000000
--- a/drivers/gpu/drm/tinydrm/core/tinydrm-core.c
+++ /dev/null
@@ -1,183 +0,0 @@
-/*
- * Copyright (C) 2016 Noralf Trønnes
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- */
-
-#include <drm/drm_atomic.h>
-#include <drm/drm_atomic_helper.h>
-#include <drm/drm_drv.h>
-#include <drm/drm_fb_helper.h>
-#include <drm/drm_gem_framebuffer_helper.h>
-#include <drm/drm_probe_helper.h>
-#include <drm/drm_print.h>
-#include <drm/tinydrm/tinydrm.h>
-#include <linux/device.h>
-#include <linux/dma-buf.h>
-#include <linux/module.h>
-
-/**
- * DOC: overview
- *
- * This library provides driver helpers for very simple display hardware.
- *
- * It is based on &drm_simple_display_pipe coupled with a &drm_connector which
- * has only one fixed &drm_display_mode. The framebuffers are backed by the
- * cma helper and have support for framebuffer flushing (dirty).
- * fbdev support is also included.
- *
- */
-
-/**
- * DOC: core
- *
- * The driver allocates &tinydrm_device, initializes it using
- * devm_tinydrm_init(), sets up the pipeline using tinydrm_display_pipe_init()
- * and registers the DRM device using devm_tinydrm_register().
- */
-
-static const struct drm_mode_config_funcs tinydrm_mode_config_funcs = {
-	.fb_create = drm_gem_fb_create_with_dirty,
-	.atomic_check = drm_atomic_helper_check,
-	.atomic_commit = drm_atomic_helper_commit,
-};
-
-static int tinydrm_init(struct device *parent, struct tinydrm_device *tdev,
-			struct drm_driver *driver)
-{
-	struct drm_device *drm;
-
-	/*
-	 * We don't embed drm_device, because that prevent us from using
-	 * devm_kzalloc() to allocate tinydrm_device in the driver since
-	 * drm_dev_put() frees the structure. The devm_ functions provide
-	 * for easy error handling.
-	 */
-	drm = drm_dev_alloc(driver, parent);
-	if (IS_ERR(drm))
-		return PTR_ERR(drm);
-
-	tdev->drm = drm;
-	drm->dev_private = tdev;
-	drm_mode_config_init(drm);
-	drm->mode_config.funcs = &tinydrm_mode_config_funcs;
-	drm->mode_config.allow_fb_modifiers = true;
-
-	return 0;
-}
-
-static void tinydrm_fini(struct tinydrm_device *tdev)
-{
-	drm_mode_config_cleanup(tdev->drm);
-	tdev->drm->dev_private = NULL;
-	drm_dev_put(tdev->drm);
-}
-
-static void devm_tinydrm_release(void *data)
-{
-	tinydrm_fini(data);
-}
-
-/**
- * devm_tinydrm_init - Initialize tinydrm device
- * @parent: Parent device object
- * @tdev: tinydrm device
- * @driver: DRM driver
- *
- * This function initializes @tdev, the underlying DRM device and it's
- * mode_config. Resources will be automatically freed on driver detach (devres)
- * using drm_mode_config_cleanup() and drm_dev_put().
- *
- * Returns:
- * Zero on success, negative error code on failure.
- */
-int devm_tinydrm_init(struct device *parent, struct tinydrm_device *tdev,
-		      struct drm_driver *driver)
-{
-	int ret;
-
-	ret = tinydrm_init(parent, tdev, driver);
-	if (ret)
-		return ret;
-
-	ret = devm_add_action(parent, devm_tinydrm_release, tdev);
-	if (ret)
-		tinydrm_fini(tdev);
-
-	return ret;
-}
-EXPORT_SYMBOL(devm_tinydrm_init);
-
-static int tinydrm_register(struct tinydrm_device *tdev)
-{
-	struct drm_device *drm = tdev->drm;
-	int ret;
-
-	ret = drm_dev_register(tdev->drm, 0);
-	if (ret)
-		return ret;
-
-	ret = drm_fbdev_generic_setup(drm, 0);
-	if (ret)
-		DRM_ERROR("Failed to initialize fbdev: %d\n", ret);
-
-	return 0;
-}
-
-static void tinydrm_unregister(struct tinydrm_device *tdev)
-{
-	drm_atomic_helper_shutdown(tdev->drm);
-	drm_dev_unregister(tdev->drm);
-}
-
-static void devm_tinydrm_register_release(void *data)
-{
-	tinydrm_unregister(data);
-}
-
-/**
- * devm_tinydrm_register - Register tinydrm device
- * @tdev: tinydrm device
- *
- * This function registers the underlying DRM device and fbdev.
- * These resources will be automatically unregistered on driver detach (devres)
- * and the display pipeline will be disabled.
- *
- * Returns:
- * Zero on success, negative error code on failure.
- */
-int devm_tinydrm_register(struct tinydrm_device *tdev)
-{
-	struct device *dev = tdev->drm->dev;
-	int ret;
-
-	ret = tinydrm_register(tdev);
-	if (ret)
-		return ret;
-
-	ret = devm_add_action(dev, devm_tinydrm_register_release, tdev);
-	if (ret)
-		tinydrm_unregister(tdev);
-
-	return ret;
-}
-EXPORT_SYMBOL(devm_tinydrm_register);
-
-/**
- * tinydrm_shutdown - Shutdown tinydrm
- * @tdev: tinydrm device
- *
- * This function makes sure that the display pipeline is disabled.
- * Used by drivers in their shutdown callback to turn off the display
- * on machine shutdown and reboot.
- */
-void tinydrm_shutdown(struct tinydrm_device *tdev)
-{
-	drm_atomic_helper_shutdown(tdev->drm);
-}
-EXPORT_SYMBOL(tinydrm_shutdown);
-
-MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/tinydrm/core/tinydrm-helpers.c b/drivers/gpu/drm/tinydrm/core/tinydrm-helpers.c
index 2737b6fdadc8..6d540d93758f 100644
--- a/drivers/gpu/drm/tinydrm/core/tinydrm-helpers.c
+++ b/drivers/gpu/drm/tinydrm/core/tinydrm-helpers.c
@@ -26,164 +26,6 @@ static unsigned int spi_max;
 module_param(spi_max, uint, 0400);
 MODULE_PARM_DESC(spi_max, "Set a lower SPI max transfer size");
 
-/**
- * tinydrm_memcpy - Copy clip buffer
- * @dst: Destination buffer
- * @vaddr: Source buffer
- * @fb: DRM framebuffer
- * @clip: Clip rectangle area to copy
- */
-void tinydrm_memcpy(void *dst, void *vaddr, struct drm_framebuffer *fb,
-		    struct drm_rect *clip)
-{
-	unsigned int cpp = drm_format_plane_cpp(fb->format->format, 0);
-	unsigned int pitch = fb->pitches[0];
-	void *src = vaddr + (clip->y1 * pitch) + (clip->x1 * cpp);
-	size_t len = (clip->x2 - clip->x1) * cpp;
-	unsigned int y;
-
-	for (y = clip->y1; y < clip->y2; y++) {
-		memcpy(dst, src, len);
-		src += pitch;
-		dst += len;
-	}
-}
-EXPORT_SYMBOL(tinydrm_memcpy);
-
-/**
- * tinydrm_swab16 - Swap bytes into clip buffer
- * @dst: RGB565 destination buffer
- * @vaddr: RGB565 source buffer
- * @fb: DRM framebuffer
- * @clip: Clip rectangle area to copy
- */
-void tinydrm_swab16(u16 *dst, void *vaddr, struct drm_framebuffer *fb,
-		    struct drm_rect *clip)
-{
-	size_t len = (clip->x2 - clip->x1) * sizeof(u16);
-	unsigned int x, y;
-	u16 *src, *buf;
-
-	/*
-	 * The cma memory is write-combined so reads are uncached.
-	 * Speed up by fetching one line at a time.
-	 */
-	buf = kmalloc(len, GFP_KERNEL);
-	if (!buf)
-		return;
-
-	for (y = clip->y1; y < clip->y2; y++) {
-		src = vaddr + (y * fb->pitches[0]);
-		src += clip->x1;
-		memcpy(buf, src, len);
-		src = buf;
-		for (x = clip->x1; x < clip->x2; x++)
-			*dst++ = swab16(*src++);
-	}
-
-	kfree(buf);
-}
-EXPORT_SYMBOL(tinydrm_swab16);
-
-/**
- * tinydrm_xrgb8888_to_rgb565 - Convert XRGB8888 to RGB565 clip buffer
- * @dst: RGB565 destination buffer
- * @vaddr: XRGB8888 source buffer
- * @fb: DRM framebuffer
- * @clip: Clip rectangle area to copy
- * @swap: Swap bytes
- *
- * Drivers can use this function for RGB565 devices that don't natively
- * support XRGB8888.
- */
-void tinydrm_xrgb8888_to_rgb565(u16 *dst, void *vaddr,
-				struct drm_framebuffer *fb,
-				struct drm_rect *clip, bool swap)
-{
-	size_t len = (clip->x2 - clip->x1) * sizeof(u32);
-	unsigned int x, y;
-	u32 *src, *buf;
-	u16 val16;
-
-	buf = kmalloc(len, GFP_KERNEL);
-	if (!buf)
-		return;
-
-	for (y = clip->y1; y < clip->y2; y++) {
-		src = vaddr + (y * fb->pitches[0]);
-		src += clip->x1;
-		memcpy(buf, src, len);
-		src = buf;
-		for (x = clip->x1; x < clip->x2; x++) {
-			val16 = ((*src & 0x00F80000) >> 8) |
-				((*src & 0x0000FC00) >> 5) |
-				((*src & 0x000000F8) >> 3);
-			src++;
-			if (swap)
-				*dst++ = swab16(val16);
-			else
-				*dst++ = val16;
-		}
-	}
-
-	kfree(buf);
-}
-EXPORT_SYMBOL(tinydrm_xrgb8888_to_rgb565);
-
-/**
- * tinydrm_xrgb8888_to_gray8 - Convert XRGB8888 to grayscale
- * @dst: 8-bit grayscale destination buffer
- * @vaddr: XRGB8888 source buffer
- * @fb: DRM framebuffer
- * @clip: Clip rectangle area to copy
- *
- * Drm doesn't have native monochrome or grayscale support.
- * Such drivers can announce the commonly supported XR24 format to userspace
- * and use this function to convert to the native format.
- *
- * Monochrome drivers will use the most significant bit,
- * where 1 means foreground color and 0 background color.
- *
- * ITU BT.601 is used for the RGB -> luma (brightness) conversion.
- */
-void tinydrm_xrgb8888_to_gray8(u8 *dst, void *vaddr, struct drm_framebuffer *fb,
-			       struct drm_rect *clip)
-{
-	unsigned int len = (clip->x2 - clip->x1) * sizeof(u32);
-	unsigned int x, y;
-	void *buf;
-	u32 *src;
-
-	if (WARN_ON(fb->format->format != DRM_FORMAT_XRGB8888))
-		return;
-	/*
-	 * The cma memory is write-combined so reads are uncached.
-	 * Speed up by fetching one line at a time.
-	 */
-	buf = kmalloc(len, GFP_KERNEL);
-	if (!buf)
-		return;
-
-	for (y = clip->y1; y < clip->y2; y++) {
-		src = vaddr + (y * fb->pitches[0]);
-		src += clip->x1;
-		memcpy(buf, src, len);
-		src = buf;
-		for (x = clip->x1; x < clip->x2; x++) {
-			u8 r = (*src & 0x00ff0000) >> 16;
-			u8 g = (*src & 0x0000ff00) >> 8;
-			u8 b =  *src & 0x000000ff;
-
-			/* ITU BT.601: Y = 0.299 R + 0.587 G + 0.114 B */
-			*dst++ = (3 * r + 6 * g + b) / 10;
-			src++;
-		}
-	}
-
-	kfree(buf);
-}
-EXPORT_SYMBOL(tinydrm_xrgb8888_to_gray8);
-
 #if IS_ENABLED(CONFIG_SPI)
 
 /**
@@ -365,3 +207,5 @@ int tinydrm_spi_transfer(struct spi_device *spi, u32 speed_hz,
 EXPORT_SYMBOL(tinydrm_spi_transfer);
 
 #endif /* CONFIG_SPI */
+
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/tinydrm/core/tinydrm-pipe.c b/drivers/gpu/drm/tinydrm/core/tinydrm-pipe.c
index bb5b1c1e21ba..bb8a7ed8ddf6 100644
--- a/drivers/gpu/drm/tinydrm/core/tinydrm-pipe.c
+++ b/drivers/gpu/drm/tinydrm/core/tinydrm-pipe.c
@@ -13,7 +13,7 @@
 #include <drm/drm_modes.h>
 #include <drm/drm_probe_helper.h>
 #include <drm/drm_print.h>
-#include <drm/tinydrm/tinydrm.h>
+#include <drm/drm_simple_kms_helper.h>
 
 struct tinydrm_connector {
 	struct drm_connector base;
@@ -129,7 +129,8 @@ static int tinydrm_rotate_mode(struct drm_display_mode *mode,
 
 /**
  * tinydrm_display_pipe_init - Initialize display pipe
- * @tdev: tinydrm device
+ * @drm: DRM device
+ * @pipe: Display pipe
  * @funcs: Display pipe functions
  * @connector_type: Connector type
  * @formats: Array of supported formats (DRM_FORMAT\_\*)
@@ -143,16 +144,15 @@ static int tinydrm_rotate_mode(struct drm_display_mode *mode,
  * Returns:
  * Zero on success, negative error code on failure.
  */
-int
-tinydrm_display_pipe_init(struct tinydrm_device *tdev,
-			  const struct drm_simple_display_pipe_funcs *funcs,
-			  int connector_type,
-			  const uint32_t *formats,
-			  unsigned int format_count,
-			  const struct drm_display_mode *mode,
-			  unsigned int rotation)
+int tinydrm_display_pipe_init(struct drm_device *drm,
+			      struct drm_simple_display_pipe *pipe,
+			      const struct drm_simple_display_pipe_funcs *funcs,
+			      int connector_type,
+			      const uint32_t *formats,
+			      unsigned int format_count,
+			      const struct drm_display_mode *mode,
+			      unsigned int rotation)
 {
-	struct drm_device *drm = tdev->drm;
 	struct drm_display_mode mode_copy;
 	struct drm_connector *connector;
 	int ret;
@@ -177,7 +177,7 @@ tinydrm_display_pipe_init(struct tinydrm_device *tdev,
 	if (IS_ERR(connector))
 		return PTR_ERR(connector);
 
-	return drm_simple_display_pipe_init(drm, &tdev->pipe, funcs, formats,
+	return drm_simple_display_pipe_init(drm, pipe, funcs, formats,
 					    format_count, modifiers, connector);
 }
 EXPORT_SYMBOL(tinydrm_display_pipe_init);
diff --git a/drivers/gpu/drm/tinydrm/hx8357d.c b/drivers/gpu/drm/tinydrm/hx8357d.c
index 8bbd0beafc6a..5773d0fb6ca1 100644
--- a/drivers/gpu/drm/tinydrm/hx8357d.c
+++ b/drivers/gpu/drm/tinydrm/hx8357d.c
@@ -16,7 +16,9 @@
 #include <linux/property.h>
 #include <linux/spi/spi.h>
 
+#include <drm/drm_atomic_helper.h>
 #include <drm/drm_drv.h>
+#include <drm/drm_fb_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_modeset_helper.h>
@@ -46,16 +48,18 @@ static void yx240qv29_enable(struct drm_simple_display_pipe *pipe,
 			     struct drm_crtc_state *crtc_state,
 			     struct drm_plane_state *plane_state)
 {
-	struct tinydrm_device *tdev = pipe_to_tinydrm(pipe);
-	struct mipi_dbi *mipi = mipi_dbi_from_tinydrm(tdev);
+	struct mipi_dbi *mipi = drm_to_mipi_dbi(pipe->crtc.dev);
 	u8 addr_mode;
-	int ret;
+	int ret, idx;
+
+	if (!drm_dev_enter(pipe->crtc.dev, &idx))
+		return;
 
 	DRM_DEBUG_KMS("\n");
 
 	ret = mipi_dbi_poweron_conditional_reset(mipi);
 	if (ret < 0)
-		return;
+		goto out_exit;
 	if (ret == 1)
 		goto out_enable;
 
@@ -171,6 +175,8 @@ out_enable:
 	}
 	mipi_dbi_command(mipi, MIPI_DCS_SET_ADDRESS_MODE, addr_mode);
 	mipi_dbi_enable_flush(mipi, crtc_state, plane_state);
+out_exit:
+	drm_dev_exit(idx);
 }
 
 static const struct drm_simple_display_pipe_funcs hx8357d_pipe_funcs = {
@@ -181,7 +187,7 @@ static const struct drm_simple_display_pipe_funcs hx8357d_pipe_funcs = {
 };
 
 static const struct drm_display_mode yx350hv15_mode = {
-	TINYDRM_MODE(320, 480, 60, 75),
+	DRM_SIMPLE_MODE(320, 480, 60, 75),
 };
 
 DEFINE_DRM_GEM_CMA_FOPS(hx8357d_fops);
@@ -189,6 +195,7 @@ DEFINE_DRM_GEM_CMA_FOPS(hx8357d_fops);
 static struct drm_driver hx8357d_driver = {
 	.driver_features	= DRIVER_GEM | DRIVER_MODESET | DRIVER_PRIME | DRIVER_ATOMIC,
 	.fops			= &hx8357d_fops,
+	.release		= mipi_dbi_release,
 	DRM_GEM_CMA_VMAP_DRIVER_OPS,
 	.debugfs_init		= mipi_dbi_debugfs_init,
 	.name			= "hx8357d",
@@ -213,15 +220,25 @@ MODULE_DEVICE_TABLE(spi, hx8357d_id);
 static int hx8357d_probe(struct spi_device *spi)
 {
 	struct device *dev = &spi->dev;
+	struct drm_device *drm;
 	struct mipi_dbi *mipi;
 	struct gpio_desc *dc;
 	u32 rotation = 0;
 	int ret;
 
-	mipi = devm_kzalloc(dev, sizeof(*mipi), GFP_KERNEL);
+	mipi = kzalloc(sizeof(*mipi), GFP_KERNEL);
 	if (!mipi)
 		return -ENOMEM;
 
+	drm = &mipi->drm;
+	ret = devm_drm_dev_init(dev, drm, &hx8357d_driver);
+	if (ret) {
+		kfree(mipi);
+		return ret;
+	}
+
+	drm_mode_config_init(drm);
+
 	dc = devm_gpiod_get(dev, "dc", GPIOD_OUT_LOW);
 	if (IS_ERR(dc)) {
 		DRM_DEV_ERROR(dev, "Failed to get gpio 'dc'\n");
@@ -238,21 +255,36 @@ static int hx8357d_probe(struct spi_device *spi)
 	if (ret)
 		return ret;
 
-	ret = mipi_dbi_init(&spi->dev, mipi, &hx8357d_pipe_funcs,
-			    &hx8357d_driver, &yx350hv15_mode, rotation);
+	ret = mipi_dbi_init(mipi, &hx8357d_pipe_funcs, &yx350hv15_mode, rotation);
 	if (ret)
 		return ret;
 
-	spi_set_drvdata(spi, mipi);
+	drm_mode_config_reset(drm);
 
-	return devm_tinydrm_register(&mipi->tinydrm);
+	ret = drm_dev_register(drm, 0);
+	if (ret)
+		return ret;
+
+	spi_set_drvdata(spi, drm);
+
+	drm_fbdev_generic_setup(drm, 0);
+
+	return 0;
 }
 
-static void hx8357d_shutdown(struct spi_device *spi)
+static int hx8357d_remove(struct spi_device *spi)
 {
-	struct mipi_dbi *mipi = spi_get_drvdata(spi);
+	struct drm_device *drm = spi_get_drvdata(spi);
+
+	drm_dev_unplug(drm);
+	drm_atomic_helper_shutdown(drm);
 
-	tinydrm_shutdown(&mipi->tinydrm);
+	return 0;
+}
+
+static void hx8357d_shutdown(struct spi_device *spi)
+{
+	drm_atomic_helper_shutdown(spi_get_drvdata(spi));
 }
 
 static struct spi_driver hx8357d_spi_driver = {
@@ -262,6 +294,7 @@ static struct spi_driver hx8357d_spi_driver = {
 	},
 	.id_table = hx8357d_id,
 	.probe = hx8357d_probe,
+	.remove = hx8357d_remove,
 	.shutdown = hx8357d_shutdown,
 };
 module_spi_driver(hx8357d_spi_driver);
diff --git a/drivers/gpu/drm/tinydrm/ili9225.c b/drivers/gpu/drm/tinydrm/ili9225.c
index 43a3b68d90a2..4b1a587c0134 100644
--- a/drivers/gpu/drm/tinydrm/ili9225.c
+++ b/drivers/gpu/drm/tinydrm/ili9225.c
@@ -20,9 +20,11 @@
 #include <linux/spi/spi.h>
 #include <video/mipi_display.h>
 
+#include <drm/drm_atomic_helper.h>
 #include <drm/drm_damage_helper.h>
 #include <drm/drm_drv.h>
 #include <drm/drm_fb_cma_helper.h>
+#include <drm/drm_fb_helper.h>
 #include <drm/drm_fourcc.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
@@ -81,20 +83,22 @@ static inline int ili9225_command(struct mipi_dbi *mipi, u8 cmd, u16 data)
 static void ili9225_fb_dirty(struct drm_framebuffer *fb, struct drm_rect *rect)
 {
 	struct drm_gem_cma_object *cma_obj = drm_fb_cma_get_gem_obj(fb, 0);
-	struct tinydrm_device *tdev = fb->dev->dev_private;
-	struct mipi_dbi *mipi = mipi_dbi_from_tinydrm(tdev);
+	struct mipi_dbi *mipi = drm_to_mipi_dbi(fb->dev);
 	unsigned int height = rect->y2 - rect->y1;
 	unsigned int width = rect->x2 - rect->x1;
 	bool swap = mipi->swap_bytes;
 	u16 x_start, y_start;
 	u16 x1, x2, y1, y2;
-	int ret = 0;
+	int idx, ret = 0;
 	bool full;
 	void *tr;
 
 	if (!mipi->enabled)
 		return;
 
+	if (!drm_dev_enter(fb->dev, &idx))
+		return;
+
 	full = width == fb->width && height == fb->height;
 
 	DRM_DEBUG_KMS("Flushing [FB:%d] " DRM_RECT_FMT "\n", fb->base.id, DRM_RECT_ARG(rect));
@@ -157,6 +161,8 @@ static void ili9225_fb_dirty(struct drm_framebuffer *fb, struct drm_rect *rect)
 err_msg:
 	if (ret)
 		dev_err_once(fb->dev->dev, "Failed to update display %d\n", ret);
+
+	drm_dev_exit(idx);
 }
 
 static void ili9225_pipe_update(struct drm_simple_display_pipe *pipe,
@@ -181,19 +187,21 @@ static void ili9225_pipe_enable(struct drm_simple_display_pipe *pipe,
 				struct drm_crtc_state *crtc_state,
 				struct drm_plane_state *plane_state)
 {
-	struct tinydrm_device *tdev = pipe_to_tinydrm(pipe);
-	struct mipi_dbi *mipi = mipi_dbi_from_tinydrm(tdev);
+	struct mipi_dbi *mipi = drm_to_mipi_dbi(pipe->crtc.dev);
 	struct drm_framebuffer *fb = plane_state->fb;
-	struct device *dev = tdev->drm->dev;
+	struct device *dev = pipe->crtc.dev->dev;
 	struct drm_rect rect = {
 		.x1 = 0,
 		.x2 = fb->width,
 		.y1 = 0,
 		.y2 = fb->height,
 	};
-	int ret;
+	int ret, idx;
 	u8 am_id;
 
+	if (!drm_dev_enter(pipe->crtc.dev, &idx))
+		return;
+
 	DRM_DEBUG_KMS("\n");
 
 	mipi_dbi_hw_reset(mipi);
@@ -207,7 +215,7 @@ static void ili9225_pipe_enable(struct drm_simple_display_pipe *pipe,
 	ret = ili9225_command(mipi, ILI9225_POWER_CONTROL_1, 0x0000);
 	if (ret) {
 		DRM_DEV_ERROR(dev, "Error sending command %d\n", ret);
-		return;
+		goto out_exit;
 	}
 	ili9225_command(mipi, ILI9225_POWER_CONTROL_2, 0x0000);
 	ili9225_command(mipi, ILI9225_POWER_CONTROL_3, 0x0000);
@@ -280,15 +288,23 @@ static void ili9225_pipe_enable(struct drm_simple_display_pipe *pipe,
 
 	mipi->enabled = true;
 	ili9225_fb_dirty(fb, &rect);
+out_exit:
+	drm_dev_exit(idx);
 }
 
 static void ili9225_pipe_disable(struct drm_simple_display_pipe *pipe)
 {
-	struct tinydrm_device *tdev = pipe_to_tinydrm(pipe);
-	struct mipi_dbi *mipi = mipi_dbi_from_tinydrm(tdev);
+	struct mipi_dbi *mipi = drm_to_mipi_dbi(pipe->crtc.dev);
 
 	DRM_DEBUG_KMS("\n");
 
+	/*
+	 * This callback is not protected by drm_dev_enter/exit since we want to
+	 * turn off the display on regular driver unload. It's highly unlikely
+	 * that the underlying SPI controller is gone should this be called after
+	 * unplug.
+	 */
+
 	if (!mipi->enabled)
 		return;
 
@@ -301,7 +317,7 @@ static void ili9225_pipe_disable(struct drm_simple_display_pipe *pipe)
 	mipi->enabled = false;
 }
 
-static int ili9225_dbi_command(struct mipi_dbi *mipi, u8 cmd, u8 *par,
+static int ili9225_dbi_command(struct mipi_dbi *mipi, u8 *cmd, u8 *par,
 			       size_t num)
 {
 	struct spi_device *spi = mipi->spi;
@@ -311,11 +327,11 @@ static int ili9225_dbi_command(struct mipi_dbi *mipi, u8 cmd, u8 *par,
 
 	gpiod_set_value_cansleep(mipi->dc, 0);
 	speed_hz = mipi_dbi_spi_cmd_max_speed(spi, 1);
-	ret = tinydrm_spi_transfer(spi, speed_hz, NULL, 8, &cmd, 1);
+	ret = tinydrm_spi_transfer(spi, speed_hz, NULL, 8, cmd, 1);
 	if (ret || !num)
 		return ret;
 
-	if (cmd == ILI9225_WRITE_DATA_TO_GRAM && !mipi->swap_bytes)
+	if (*cmd == ILI9225_WRITE_DATA_TO_GRAM && !mipi->swap_bytes)
 		bpw = 16;
 
 	gpiod_set_value_cansleep(mipi->dc, 1);
@@ -332,7 +348,7 @@ static const struct drm_simple_display_pipe_funcs ili9225_pipe_funcs = {
 };
 
 static const struct drm_display_mode ili9225_mode = {
-	TINYDRM_MODE(176, 220, 35, 44),
+	DRM_SIMPLE_MODE(176, 220, 35, 44),
 };
 
 DEFINE_DRM_GEM_CMA_FOPS(ili9225_fops);
@@ -341,6 +357,7 @@ static struct drm_driver ili9225_driver = {
 	.driver_features	= DRIVER_GEM | DRIVER_MODESET | DRIVER_PRIME |
 				  DRIVER_ATOMIC,
 	.fops			= &ili9225_fops,
+	.release		= mipi_dbi_release,
 	DRM_GEM_CMA_VMAP_DRIVER_OPS,
 	.name			= "ili9225",
 	.desc			= "Ilitek ILI9225",
@@ -364,15 +381,25 @@ MODULE_DEVICE_TABLE(spi, ili9225_id);
 static int ili9225_probe(struct spi_device *spi)
 {
 	struct device *dev = &spi->dev;
+	struct drm_device *drm;
 	struct mipi_dbi *mipi;
 	struct gpio_desc *rs;
 	u32 rotation = 0;
 	int ret;
 
-	mipi = devm_kzalloc(dev, sizeof(*mipi), GFP_KERNEL);
+	mipi = kzalloc(sizeof(*mipi), GFP_KERNEL);
 	if (!mipi)
 		return -ENOMEM;
 
+	drm = &mipi->drm;
+	ret = devm_drm_dev_init(dev, drm, &ili9225_driver);
+	if (ret) {
+		kfree(mipi);
+		return ret;
+	}
+
+	drm_mode_config_init(drm);
+
 	mipi->reset = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH);
 	if (IS_ERR(mipi->reset)) {
 		DRM_DEV_ERROR(dev, "Failed to get gpio 'reset'\n");
@@ -394,21 +421,36 @@ static int ili9225_probe(struct spi_device *spi)
 	/* override the command function set in  mipi_dbi_spi_init() */
 	mipi->command = ili9225_dbi_command;
 
-	ret = mipi_dbi_init(&spi->dev, mipi, &ili9225_pipe_funcs,
-			    &ili9225_driver, &ili9225_mode, rotation);
+	ret = mipi_dbi_init(mipi, &ili9225_pipe_funcs, &ili9225_mode, rotation);
 	if (ret)
 		return ret;
 
-	spi_set_drvdata(spi, mipi);
+	drm_mode_config_reset(drm);
+
+	ret = drm_dev_register(drm, 0);
+	if (ret)
+		return ret;
+
+	spi_set_drvdata(spi, drm);
+
+	drm_fbdev_generic_setup(drm, 0);
 
-	return devm_tinydrm_register(&mipi->tinydrm);
+	return 0;
 }
 
-static void ili9225_shutdown(struct spi_device *spi)
+static int ili9225_remove(struct spi_device *spi)
 {
-	struct mipi_dbi *mipi = spi_get_drvdata(spi);
+	struct drm_device *drm = spi_get_drvdata(spi);
+
+	drm_dev_unplug(drm);
+	drm_atomic_helper_shutdown(drm);
+
+	return 0;
+}
 
-	tinydrm_shutdown(&mipi->tinydrm);
+static void ili9225_shutdown(struct spi_device *spi)
+{
+	drm_atomic_helper_shutdown(spi_get_drvdata(spi));
 }
 
 static struct spi_driver ili9225_spi_driver = {
@@ -419,6 +461,7 @@ static struct spi_driver ili9225_spi_driver = {
 	},
 	.id_table = ili9225_id,
 	.probe = ili9225_probe,
+	.remove = ili9225_remove,
 	.shutdown = ili9225_shutdown,
 };
 module_spi_driver(ili9225_spi_driver);
diff --git a/drivers/gpu/drm/tinydrm/ili9341.c b/drivers/gpu/drm/tinydrm/ili9341.c
index 713bb2dd7e04..4ade9e4b924f 100644
--- a/drivers/gpu/drm/tinydrm/ili9341.c
+++ b/drivers/gpu/drm/tinydrm/ili9341.c
@@ -15,7 +15,9 @@
 #include <linux/property.h>
 #include <linux/spi/spi.h>
 
+#include <drm/drm_atomic_helper.h>
 #include <drm/drm_drv.h>
+#include <drm/drm_fb_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_modeset_helper.h>
@@ -52,16 +54,18 @@ static void yx240qv29_enable(struct drm_simple_display_pipe *pipe,
 			     struct drm_crtc_state *crtc_state,
 			     struct drm_plane_state *plane_state)
 {
-	struct tinydrm_device *tdev = pipe_to_tinydrm(pipe);
-	struct mipi_dbi *mipi = mipi_dbi_from_tinydrm(tdev);
+	struct mipi_dbi *mipi = drm_to_mipi_dbi(pipe->crtc.dev);
 	u8 addr_mode;
-	int ret;
+	int ret, idx;
+
+	if (!drm_dev_enter(pipe->crtc.dev, &idx))
+		return;
 
 	DRM_DEBUG_KMS("\n");
 
 	ret = mipi_dbi_poweron_conditional_reset(mipi);
 	if (ret < 0)
-		return;
+		goto out_exit;
 	if (ret == 1)
 		goto out_enable;
 
@@ -127,6 +131,8 @@ out_enable:
 	addr_mode |= ILI9341_MADCTL_BGR;
 	mipi_dbi_command(mipi, MIPI_DCS_SET_ADDRESS_MODE, addr_mode);
 	mipi_dbi_enable_flush(mipi, crtc_state, plane_state);
+out_exit:
+	drm_dev_exit(idx);
 }
 
 static const struct drm_simple_display_pipe_funcs ili9341_pipe_funcs = {
@@ -137,7 +143,7 @@ static const struct drm_simple_display_pipe_funcs ili9341_pipe_funcs = {
 };
 
 static const struct drm_display_mode yx240qv29_mode = {
-	TINYDRM_MODE(240, 320, 37, 49),
+	DRM_SIMPLE_MODE(240, 320, 37, 49),
 };
 
 DEFINE_DRM_GEM_CMA_FOPS(ili9341_fops);
@@ -145,6 +151,7 @@ DEFINE_DRM_GEM_CMA_FOPS(ili9341_fops);
 static struct drm_driver ili9341_driver = {
 	.driver_features	= DRIVER_GEM | DRIVER_MODESET | DRIVER_PRIME | DRIVER_ATOMIC,
 	.fops			= &ili9341_fops,
+	.release		= mipi_dbi_release,
 	DRM_GEM_CMA_VMAP_DRIVER_OPS,
 	.debugfs_init		= mipi_dbi_debugfs_init,
 	.name			= "ili9341",
@@ -169,15 +176,25 @@ MODULE_DEVICE_TABLE(spi, ili9341_id);
 static int ili9341_probe(struct spi_device *spi)
 {
 	struct device *dev = &spi->dev;
+	struct drm_device *drm;
 	struct mipi_dbi *mipi;
 	struct gpio_desc *dc;
 	u32 rotation = 0;
 	int ret;
 
-	mipi = devm_kzalloc(dev, sizeof(*mipi), GFP_KERNEL);
+	mipi = kzalloc(sizeof(*mipi), GFP_KERNEL);
 	if (!mipi)
 		return -ENOMEM;
 
+	drm = &mipi->drm;
+	ret = devm_drm_dev_init(dev, drm, &ili9341_driver);
+	if (ret) {
+		kfree(mipi);
+		return ret;
+	}
+
+	drm_mode_config_init(drm);
+
 	mipi->reset = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_HIGH);
 	if (IS_ERR(mipi->reset)) {
 		DRM_DEV_ERROR(dev, "Failed to get gpio 'reset'\n");
@@ -200,21 +217,36 @@ static int ili9341_probe(struct spi_device *spi)
 	if (ret)
 		return ret;
 
-	ret = mipi_dbi_init(&spi->dev, mipi, &ili9341_pipe_funcs,
-			    &ili9341_driver, &yx240qv29_mode, rotation);
+	ret = mipi_dbi_init(mipi, &ili9341_pipe_funcs, &yx240qv29_mode, rotation);
 	if (ret)
 		return ret;
 
-	spi_set_drvdata(spi, mipi);
+	drm_mode_config_reset(drm);
 
-	return devm_tinydrm_register(&mipi->tinydrm);
+	ret = drm_dev_register(drm, 0);
+	if (ret)
+		return ret;
+
+	spi_set_drvdata(spi, drm);
+
+	drm_fbdev_generic_setup(drm, 0);
+
+	return 0;
 }
 
-static void ili9341_shutdown(struct spi_device *spi)
+static int ili9341_remove(struct spi_device *spi)
 {
-	struct mipi_dbi *mipi = spi_get_drvdata(spi);
+	struct drm_device *drm = spi_get_drvdata(spi);
+
+	drm_dev_unplug(drm);
+	drm_atomic_helper_shutdown(drm);
 
-	tinydrm_shutdown(&mipi->tinydrm);
+	return 0;
+}
+
+static void ili9341_shutdown(struct spi_device *spi)
+{
+	drm_atomic_helper_shutdown(spi_get_drvdata(spi));
 }
 
 static struct spi_driver ili9341_spi_driver = {
@@ -224,6 +256,7 @@ static struct spi_driver ili9341_spi_driver = {
 	},
 	.id_table = ili9341_id,
 	.probe = ili9341_probe,
+	.remove = ili9341_remove,
 	.shutdown = ili9341_shutdown,
 };
 module_spi_driver(ili9341_spi_driver);
diff --git a/drivers/gpu/drm/tinydrm/mi0283qt.c b/drivers/gpu/drm/tinydrm/mi0283qt.c
index 82a92ec9ae3c..8e169846fbd8 100644
--- a/drivers/gpu/drm/tinydrm/mi0283qt.c
+++ b/drivers/gpu/drm/tinydrm/mi0283qt.c
@@ -17,7 +17,9 @@
 #include <linux/regulator/consumer.h>
 #include <linux/spi/spi.h>
 
+#include <drm/drm_atomic_helper.h>
 #include <drm/drm_drv.h>
+#include <drm/drm_fb_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_modeset_helper.h>
@@ -54,16 +56,18 @@ static void mi0283qt_enable(struct drm_simple_display_pipe *pipe,
 			    struct drm_crtc_state *crtc_state,
 			    struct drm_plane_state *plane_state)
 {
-	struct tinydrm_device *tdev = pipe_to_tinydrm(pipe);
-	struct mipi_dbi *mipi = mipi_dbi_from_tinydrm(tdev);
+	struct mipi_dbi *mipi = drm_to_mipi_dbi(pipe->crtc.dev);
 	u8 addr_mode;
-	int ret;
+	int ret, idx;
+
+	if (!drm_dev_enter(pipe->crtc.dev, &idx))
+		return;
 
 	DRM_DEBUG_KMS("\n");
 
 	ret = mipi_dbi_poweron_conditional_reset(mipi);
 	if (ret < 0)
-		return;
+		goto out_exit;
 	if (ret == 1)
 		goto out_enable;
 
@@ -135,6 +139,8 @@ out_enable:
 	addr_mode |= ILI9341_MADCTL_BGR;
 	mipi_dbi_command(mipi, MIPI_DCS_SET_ADDRESS_MODE, addr_mode);
 	mipi_dbi_enable_flush(mipi, crtc_state, plane_state);
+out_exit:
+	drm_dev_exit(idx);
 }
 
 static const struct drm_simple_display_pipe_funcs mi0283qt_pipe_funcs = {
@@ -145,7 +151,7 @@ static const struct drm_simple_display_pipe_funcs mi0283qt_pipe_funcs = {
 };
 
 static const struct drm_display_mode mi0283qt_mode = {
-	TINYDRM_MODE(320, 240, 58, 43),
+	DRM_SIMPLE_MODE(320, 240, 58, 43),
 };
 
 DEFINE_DRM_GEM_CMA_FOPS(mi0283qt_fops);
@@ -154,6 +160,7 @@ static struct drm_driver mi0283qt_driver = {
 	.driver_features	= DRIVER_GEM | DRIVER_MODESET | DRIVER_PRIME |
 				  DRIVER_ATOMIC,
 	.fops			= &mi0283qt_fops,
+	.release		= mipi_dbi_release,
 	DRM_GEM_CMA_VMAP_DRIVER_OPS,
 	.debugfs_init		= mipi_dbi_debugfs_init,
 	.name			= "mi0283qt",
@@ -178,15 +185,25 @@ MODULE_DEVICE_TABLE(spi, mi0283qt_id);
 static int mi0283qt_probe(struct spi_device *spi)
 {
 	struct device *dev = &spi->dev;
+	struct drm_device *drm;
 	struct mipi_dbi *mipi;
 	struct gpio_desc *dc;
 	u32 rotation = 0;
 	int ret;
 
-	mipi = devm_kzalloc(dev, sizeof(*mipi), GFP_KERNEL);
+	mipi = kzalloc(sizeof(*mipi), GFP_KERNEL);
 	if (!mipi)
 		return -ENOMEM;
 
+	drm = &mipi->drm;
+	ret = devm_drm_dev_init(dev, drm, &mi0283qt_driver);
+	if (ret) {
+		kfree(mipi);
+		return ret;
+	}
+
+	drm_mode_config_init(drm);
+
 	mipi->reset = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_HIGH);
 	if (IS_ERR(mipi->reset)) {
 		DRM_DEV_ERROR(dev, "Failed to get gpio 'reset'\n");
@@ -213,35 +230,46 @@ static int mi0283qt_probe(struct spi_device *spi)
 	if (ret)
 		return ret;
 
-	ret = mipi_dbi_init(&spi->dev, mipi, &mi0283qt_pipe_funcs,
-			    &mi0283qt_driver, &mi0283qt_mode, rotation);
+	ret = mipi_dbi_init(mipi, &mi0283qt_pipe_funcs, &mi0283qt_mode, rotation);
+	if (ret)
+		return ret;
+
+	drm_mode_config_reset(drm);
+
+	ret = drm_dev_register(drm, 0);
 	if (ret)
 		return ret;
 
-	spi_set_drvdata(spi, mipi);
+	spi_set_drvdata(spi, drm);
 
-	return devm_tinydrm_register(&mipi->tinydrm);
+	drm_fbdev_generic_setup(drm, 0);
+
+	return 0;
 }
 
-static void mi0283qt_shutdown(struct spi_device *spi)
+static int mi0283qt_remove(struct spi_device *spi)
 {
-	struct mipi_dbi *mipi = spi_get_drvdata(spi);
+	struct drm_device *drm = spi_get_drvdata(spi);
 
-	tinydrm_shutdown(&mipi->tinydrm);
+	drm_dev_unplug(drm);
+	drm_atomic_helper_shutdown(drm);
+
+	return 0;
 }
 
-static int __maybe_unused mi0283qt_pm_suspend(struct device *dev)
+static void mi0283qt_shutdown(struct spi_device *spi)
 {
-	struct mipi_dbi *mipi = dev_get_drvdata(dev);
+	drm_atomic_helper_shutdown(spi_get_drvdata(spi));
+}
 
-	return drm_mode_config_helper_suspend(mipi->tinydrm.drm);
+static int __maybe_unused mi0283qt_pm_suspend(struct device *dev)
+{
+	return drm_mode_config_helper_suspend(dev_get_drvdata(dev));
 }
 
 static int __maybe_unused mi0283qt_pm_resume(struct device *dev)
 {
-	struct mipi_dbi *mipi = dev_get_drvdata(dev);
-
-	drm_mode_config_helper_resume(mipi->tinydrm.drm);
+	drm_mode_config_helper_resume(dev_get_drvdata(dev));
 
 	return 0;
 }
@@ -259,6 +287,7 @@ static struct spi_driver mi0283qt_spi_driver = {
 	},
 	.id_table = mi0283qt_id,
 	.probe = mi0283qt_probe,
+	.remove = mi0283qt_remove,
 	.shutdown = mi0283qt_shutdown,
 };
 module_spi_driver(mi0283qt_spi_driver);
diff --git a/drivers/gpu/drm/tinydrm/mipi-dbi.c b/drivers/gpu/drm/tinydrm/mipi-dbi.c
index 918f77c7de34..85761b4abb83 100644
--- a/drivers/gpu/drm/tinydrm/mipi-dbi.c
+++ b/drivers/gpu/drm/tinydrm/mipi-dbi.c
@@ -21,6 +21,7 @@
 #include <drm/drm_drv.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_cma_helper.h>
+#include <drm/drm_format_helper.h>
 #include <drm/drm_fourcc.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_vblank.h>
@@ -153,16 +154,42 @@ EXPORT_SYMBOL(mipi_dbi_command_read);
  */
 int mipi_dbi_command_buf(struct mipi_dbi *mipi, u8 cmd, u8 *data, size_t len)
 {
+	u8 *cmdbuf;
 	int ret;
 
+	/* SPI requires dma-safe buffers */
+	cmdbuf = kmemdup(&cmd, 1, GFP_KERNEL);
+	if (!cmdbuf)
+		return -ENOMEM;
+
 	mutex_lock(&mipi->cmdlock);
-	ret = mipi->command(mipi, cmd, data, len);
+	ret = mipi->command(mipi, cmdbuf, data, len);
 	mutex_unlock(&mipi->cmdlock);
 
+	kfree(cmdbuf);
+
 	return ret;
 }
 EXPORT_SYMBOL(mipi_dbi_command_buf);
 
+/* This should only be used by mipi_dbi_command() */
+int mipi_dbi_command_stackbuf(struct mipi_dbi *mipi, u8 cmd, u8 *data, size_t len)
+{
+	u8 *buf;
+	int ret;
+
+	buf = kmemdup(data, len, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	ret = mipi_dbi_command_buf(mipi, cmd, buf, len);
+
+	kfree(buf);
+
+	return ret;
+}
+EXPORT_SYMBOL(mipi_dbi_command_stackbuf);
+
 /**
  * mipi_dbi_buf_copy - Copy a framebuffer, transforming it if necessary
  * @dst: The destination buffer
@@ -192,12 +219,12 @@ int mipi_dbi_buf_copy(void *dst, struct drm_framebuffer *fb,
 	switch (fb->format->format) {
 	case DRM_FORMAT_RGB565:
 		if (swap)
-			tinydrm_swab16(dst, src, fb, clip);
+			drm_fb_swab16(dst, src, fb, clip);
 		else
-			tinydrm_memcpy(dst, src, fb, clip);
+			drm_fb_memcpy(dst, src, fb, clip);
 		break;
 	case DRM_FORMAT_XRGB8888:
-		tinydrm_xrgb8888_to_rgb565(dst, src, fb, clip, swap);
+		drm_fb_xrgb8888_to_rgb565(dst, src, fb, clip, swap);
 		break;
 	default:
 		dev_err_once(fb->dev->dev, "Format is not supported: %s\n",
@@ -216,18 +243,20 @@ EXPORT_SYMBOL(mipi_dbi_buf_copy);
 static void mipi_dbi_fb_dirty(struct drm_framebuffer *fb, struct drm_rect *rect)
 {
 	struct drm_gem_cma_object *cma_obj = drm_fb_cma_get_gem_obj(fb, 0);
-	struct tinydrm_device *tdev = fb->dev->dev_private;
-	struct mipi_dbi *mipi = mipi_dbi_from_tinydrm(tdev);
+	struct mipi_dbi *mipi = drm_to_mipi_dbi(fb->dev);
 	unsigned int height = rect->y2 - rect->y1;
 	unsigned int width = rect->x2 - rect->x1;
 	bool swap = mipi->swap_bytes;
-	int ret = 0;
+	int idx, ret = 0;
 	bool full;
 	void *tr;
 
 	if (!mipi->enabled)
 		return;
 
+	if (!drm_dev_enter(fb->dev, &idx))
+		return;
+
 	full = width == fb->width && height == fb->height;
 
 	DRM_DEBUG_KMS("Flushing [FB:%d] " DRM_RECT_FMT "\n", fb->base.id, DRM_RECT_ARG(rect));
@@ -254,6 +283,8 @@ static void mipi_dbi_fb_dirty(struct drm_framebuffer *fb, struct drm_rect *rect)
 err_msg:
 	if (ret)
 		dev_err_once(fb->dev->dev, "Failed to update display %d\n", ret);
+
+	drm_dev_exit(idx);
 }
 
 /**
@@ -308,19 +339,29 @@ void mipi_dbi_enable_flush(struct mipi_dbi *mipi,
 		.y1 = 0,
 		.y2 = fb->height,
 	};
+	int idx;
+
+	if (!drm_dev_enter(&mipi->drm, &idx))
+		return;
 
 	mipi->enabled = true;
 	mipi_dbi_fb_dirty(fb, &rect);
 	backlight_enable(mipi->backlight);
+
+	drm_dev_exit(idx);
 }
 EXPORT_SYMBOL(mipi_dbi_enable_flush);
 
 static void mipi_dbi_blank(struct mipi_dbi *mipi)
 {
-	struct drm_device *drm = mipi->tinydrm.drm;
+	struct drm_device *drm = &mipi->drm;
 	u16 height = drm->mode_config.min_height;
 	u16 width = drm->mode_config.min_width;
 	size_t len = width * height * 2;
+	int idx;
+
+	if (!drm_dev_enter(drm, &idx))
+		return;
 
 	memset(mipi->tx_buf, 0, len);
 
@@ -330,6 +371,8 @@ static void mipi_dbi_blank(struct mipi_dbi *mipi)
 			 (height >> 8) & 0xFF, (height - 1) & 0xFF);
 	mipi_dbi_command_buf(mipi, MIPI_DCS_WRITE_MEMORY_START,
 			     (u8 *)mipi->tx_buf, len);
+
+	drm_dev_exit(idx);
 }
 
 /**
@@ -342,8 +385,10 @@ static void mipi_dbi_blank(struct mipi_dbi *mipi)
  */
 void mipi_dbi_pipe_disable(struct drm_simple_display_pipe *pipe)
 {
-	struct tinydrm_device *tdev = pipe_to_tinydrm(pipe);
-	struct mipi_dbi *mipi = mipi_dbi_from_tinydrm(tdev);
+	struct mipi_dbi *mipi = drm_to_mipi_dbi(pipe->crtc.dev);
+
+	if (!mipi->enabled)
+		return;
 
 	DRM_DEBUG_KMS("\n");
 
@@ -359,6 +404,12 @@ void mipi_dbi_pipe_disable(struct drm_simple_display_pipe *pipe)
 }
 EXPORT_SYMBOL(mipi_dbi_pipe_disable);
 
+static const struct drm_mode_config_funcs mipi_dbi_mode_config_funcs = {
+	.fb_create = drm_gem_fb_create_with_dirty,
+	.atomic_check = drm_atomic_helper_check,
+	.atomic_commit = drm_atomic_helper_commit,
+};
+
 static const uint32_t mipi_dbi_formats[] = {
 	DRM_FORMAT_RGB565,
 	DRM_FORMAT_XRGB8888,
@@ -366,31 +417,27 @@ static const uint32_t mipi_dbi_formats[] = {
 
 /**
  * mipi_dbi_init - MIPI DBI initialization
- * @dev: Parent device
  * @mipi: &mipi_dbi structure to initialize
- * @pipe_funcs: Display pipe functions
- * @driver: DRM driver
+ * @funcs: Display pipe functions
  * @mode: Display mode
  * @rotation: Initial rotation in degrees Counter Clock Wise
  *
- * This function initializes a &mipi_dbi structure and it's underlying
- * @tinydrm_device. It also sets up the display pipeline.
+ * This function sets up a &drm_simple_display_pipe with a &drm_connector that
+ * has one fixed &drm_display_mode which is rotated according to @rotation.
+ * This mode is used to set the mode config min/max width/height properties.
+ * Additionally &mipi_dbi.tx_buf is allocated.
  *
  * Supported formats: Native RGB565 and emulated XRGB8888.
  *
- * Objects created by this function will be automatically freed on driver
- * detach (devres).
- *
  * Returns:
  * Zero on success, negative error code on failure.
  */
-int mipi_dbi_init(struct device *dev, struct mipi_dbi *mipi,
-		  const struct drm_simple_display_pipe_funcs *pipe_funcs,
-		  struct drm_driver *driver,
+int mipi_dbi_init(struct mipi_dbi *mipi,
+		  const struct drm_simple_display_pipe_funcs *funcs,
 		  const struct drm_display_mode *mode, unsigned int rotation)
 {
 	size_t bufsize = mode->vdisplay * mode->hdisplay * sizeof(u16);
-	struct tinydrm_device *tdev = &mipi->tinydrm;
+	struct drm_device *drm = &mipi->drm;
 	int ret;
 
 	if (!mipi->command)
@@ -398,16 +445,12 @@ int mipi_dbi_init(struct device *dev, struct mipi_dbi *mipi,
 
 	mutex_init(&mipi->cmdlock);
 
-	mipi->tx_buf = devm_kmalloc(dev, bufsize, GFP_KERNEL);
+	mipi->tx_buf = devm_kmalloc(drm->dev, bufsize, GFP_KERNEL);
 	if (!mipi->tx_buf)
 		return -ENOMEM;
 
-	ret = devm_tinydrm_init(dev, tdev, driver);
-	if (ret)
-		return ret;
-
 	/* TODO: Maybe add DRM_MODE_CONNECTOR_SPI */
-	ret = tinydrm_display_pipe_init(tdev, pipe_funcs,
+	ret = tinydrm_display_pipe_init(drm, &mipi->pipe, funcs,
 					DRM_MODE_CONNECTOR_VIRTUAL,
 					mipi_dbi_formats,
 					ARRAY_SIZE(mipi_dbi_formats), mode,
@@ -415,21 +458,40 @@ int mipi_dbi_init(struct device *dev, struct mipi_dbi *mipi,
 	if (ret)
 		return ret;
 
-	drm_plane_enable_fb_damage_clips(&tdev->pipe.plane);
+	drm_plane_enable_fb_damage_clips(&mipi->pipe.plane);
 
-	tdev->drm->mode_config.preferred_depth = 16;
+	drm->mode_config.funcs = &mipi_dbi_mode_config_funcs;
+	drm->mode_config.preferred_depth = 16;
 	mipi->rotation = rotation;
 
-	drm_mode_config_reset(tdev->drm);
-
 	DRM_DEBUG_KMS("preferred_depth=%u, rotation = %u\n",
-		      tdev->drm->mode_config.preferred_depth, rotation);
+		      drm->mode_config.preferred_depth, rotation);
 
 	return 0;
 }
 EXPORT_SYMBOL(mipi_dbi_init);
 
 /**
+ * mipi_dbi_release - DRM driver release helper
+ * @drm: DRM device
+ *
+ * This function finalizes and frees &mipi_dbi.
+ *
+ * Drivers can use this as their &drm_driver->release callback.
+ */
+void mipi_dbi_release(struct drm_device *drm)
+{
+	struct mipi_dbi *dbi = drm_to_mipi_dbi(drm);
+
+	DRM_DEBUG_DRIVER("\n");
+
+	drm_mode_config_cleanup(drm);
+	drm_dev_fini(drm);
+	kfree(dbi);
+}
+EXPORT_SYMBOL(mipi_dbi_release);
+
+/**
  * mipi_dbi_hw_reset - Hardware reset of controller
  * @mipi: MIPI DBI structure
  *
@@ -481,7 +543,7 @@ EXPORT_SYMBOL(mipi_dbi_display_is_on);
 
 static int mipi_dbi_poweron_reset_conditional(struct mipi_dbi *mipi, bool cond)
 {
-	struct device *dev = mipi->tinydrm.drm->dev;
+	struct device *dev = mipi->drm.dev;
 	int ret;
 
 	if (mipi->regulator) {
@@ -774,18 +836,18 @@ static int mipi_dbi_spi1_transfer(struct mipi_dbi *mipi, int dc,
 	return 0;
 }
 
-static int mipi_dbi_typec1_command(struct mipi_dbi *mipi, u8 cmd,
+static int mipi_dbi_typec1_command(struct mipi_dbi *mipi, u8 *cmd,
 				   u8 *parameters, size_t num)
 {
-	unsigned int bpw = (cmd == MIPI_DCS_WRITE_MEMORY_START) ? 16 : 8;
+	unsigned int bpw = (*cmd == MIPI_DCS_WRITE_MEMORY_START) ? 16 : 8;
 	int ret;
 
-	if (mipi_dbi_command_is_read(mipi, cmd))
+	if (mipi_dbi_command_is_read(mipi, *cmd))
 		return -ENOTSUPP;
 
-	MIPI_DBI_DEBUG_COMMAND(cmd, parameters, num);
+	MIPI_DBI_DEBUG_COMMAND(*cmd, parameters, num);
 
-	ret = mipi_dbi_spi1_transfer(mipi, 0, &cmd, 1, 8);
+	ret = mipi_dbi_spi1_transfer(mipi, 0, cmd, 1, 8);
 	if (ret || !num)
 		return ret;
 
@@ -794,7 +856,7 @@ static int mipi_dbi_typec1_command(struct mipi_dbi *mipi, u8 cmd,
 
 /* MIPI DBI Type C Option 3 */
 
-static int mipi_dbi_typec3_command_read(struct mipi_dbi *mipi, u8 cmd,
+static int mipi_dbi_typec3_command_read(struct mipi_dbi *mipi, u8 *cmd,
 					u8 *data, size_t len)
 {
 	struct spi_device *spi = mipi->spi;
@@ -803,7 +865,7 @@ static int mipi_dbi_typec3_command_read(struct mipi_dbi *mipi, u8 cmd,
 	struct spi_transfer tr[2] = {
 		{
 			.speed_hz = speed_hz,
-			.tx_buf = &cmd,
+			.tx_buf = cmd,
 			.len = 1,
 		}, {
 			.speed_hz = speed_hz,
@@ -821,8 +883,8 @@ static int mipi_dbi_typec3_command_read(struct mipi_dbi *mipi, u8 cmd,
 	 * Support non-standard 24-bit and 32-bit Nokia read commands which
 	 * start with a dummy clock, so we need to read an extra byte.
 	 */
-	if (cmd == MIPI_DCS_GET_DISPLAY_ID ||
-	    cmd == MIPI_DCS_GET_DISPLAY_STATUS) {
+	if (*cmd == MIPI_DCS_GET_DISPLAY_ID ||
+	    *cmd == MIPI_DCS_GET_DISPLAY_STATUS) {
 		if (!(len == 3 || len == 4))
 			return -EINVAL;
 
@@ -852,7 +914,7 @@ static int mipi_dbi_typec3_command_read(struct mipi_dbi *mipi, u8 cmd,
 			data[i] = (buf[i] << 1) | !!(buf[i + 1] & BIT(7));
 	}
 
-	MIPI_DBI_DEBUG_COMMAND(cmd, data, len);
+	MIPI_DBI_DEBUG_COMMAND(*cmd, data, len);
 
 err_free:
 	kfree(buf);
@@ -860,7 +922,7 @@ err_free:
 	return ret;
 }
 
-static int mipi_dbi_typec3_command(struct mipi_dbi *mipi, u8 cmd,
+static int mipi_dbi_typec3_command(struct mipi_dbi *mipi, u8 *cmd,
 				   u8 *par, size_t num)
 {
 	struct spi_device *spi = mipi->spi;
@@ -868,18 +930,18 @@ static int mipi_dbi_typec3_command(struct mipi_dbi *mipi, u8 cmd,
 	u32 speed_hz;
 	int ret;
 
-	if (mipi_dbi_command_is_read(mipi, cmd))
+	if (mipi_dbi_command_is_read(mipi, *cmd))
 		return mipi_dbi_typec3_command_read(mipi, cmd, par, num);
 
-	MIPI_DBI_DEBUG_COMMAND(cmd, par, num);
+	MIPI_DBI_DEBUG_COMMAND(*cmd, par, num);
 
 	gpiod_set_value_cansleep(mipi->dc, 0);
 	speed_hz = mipi_dbi_spi_cmd_max_speed(spi, 1);
-	ret = tinydrm_spi_transfer(spi, speed_hz, NULL, 8, &cmd, 1);
+	ret = tinydrm_spi_transfer(spi, speed_hz, NULL, 8, cmd, 1);
 	if (ret || !num)
 		return ret;
 
-	if (cmd == MIPI_DCS_WRITE_MEMORY_START && !mipi->swap_bytes)
+	if (*cmd == MIPI_DCS_WRITE_MEMORY_START && !mipi->swap_bytes)
 		bpw = 16;
 
 	gpiod_set_value_cansleep(mipi->dc, 1);
@@ -926,7 +988,7 @@ int mipi_dbi_spi_init(struct spi_device *spi, struct mipi_dbi *mipi,
 	 * Even though it's not the SPI device that does DMA (the master does),
 	 * the dma mask is necessary for the dma_alloc_wc() in
 	 * drm_gem_cma_create(). The dma_addr returned will be a physical
-	 * adddress which might be different from the bus address, but this is
+	 * address which might be different from the bus address, but this is
 	 * not a problem since the address will not be used.
 	 * The virtual address is used in the transfer and the SPI core
 	 * re-maps it on the SPI master device using the DMA streaming API
@@ -976,11 +1038,16 @@ static ssize_t mipi_dbi_debugfs_command_write(struct file *file,
 	u8 val, cmd = 0, parameters[64];
 	char *buf, *pos, *token;
 	unsigned int i;
-	int ret;
+	int ret, idx;
+
+	if (!drm_dev_enter(&mipi->drm, &idx))
+		return -ENODEV;
 
 	buf = memdup_user_nul(ubuf, count);
-	if (IS_ERR(buf))
-		return PTR_ERR(buf);
+	if (IS_ERR(buf)) {
+		ret = PTR_ERR(buf);
+		goto err_exit;
+	}
 
 	/* strip trailing whitespace */
 	for (i = count - 1; i > 0; i--)
@@ -1016,6 +1083,8 @@ static ssize_t mipi_dbi_debugfs_command_write(struct file *file,
 
 err_free:
 	kfree(buf);
+err_exit:
+	drm_dev_exit(idx);
 
 	return ret < 0 ? ret : count;
 }
@@ -1024,8 +1093,11 @@ static int mipi_dbi_debugfs_command_show(struct seq_file *m, void *unused)
 {
 	struct mipi_dbi *mipi = m->private;
 	u8 cmd, val[4];
+	int ret, idx;
 	size_t len;
-	int ret;
+
+	if (!drm_dev_enter(&mipi->drm, &idx))
+		return -ENODEV;
 
 	for (cmd = 0; cmd < 255; cmd++) {
 		if (!mipi_dbi_command_is_read(mipi, cmd))
@@ -1056,6 +1128,8 @@ static int mipi_dbi_debugfs_command_show(struct seq_file *m, void *unused)
 		seq_printf(m, "%*phN\n", (int)len, val);
 	}
 
+	drm_dev_exit(idx);
+
 	return 0;
 }
 
@@ -1088,8 +1162,7 @@ static const struct file_operations mipi_dbi_debugfs_command_fops = {
  */
 int mipi_dbi_debugfs_init(struct drm_minor *minor)
 {
-	struct tinydrm_device *tdev = minor->dev->dev_private;
-	struct mipi_dbi *mipi = mipi_dbi_from_tinydrm(tdev);
+	struct mipi_dbi *mipi = drm_to_mipi_dbi(minor->dev);
 	umode_t mode = S_IFREG | S_IWUSR;
 
 	if (mipi->read_commands)
diff --git a/drivers/gpu/drm/tinydrm/repaper.c b/drivers/gpu/drm/tinydrm/repaper.c
index b037c6540cf3..370629e2de94 100644
--- a/drivers/gpu/drm/tinydrm/repaper.c
+++ b/drivers/gpu/drm/tinydrm/repaper.c
@@ -26,14 +26,17 @@
 #include <linux/spi/spi.h>
 #include <linux/thermal.h>
 
+#include <drm/drm_atomic_helper.h>
 #include <drm/drm_damage_helper.h>
 #include <drm/drm_drv.h>
 #include <drm/drm_fb_cma_helper.h>
+#include <drm/drm_fb_helper.h>
+#include <drm/drm_format_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_rect.h>
 #include <drm/drm_vblank.h>
-#include <drm/tinydrm/tinydrm.h>
+#include <drm/drm_simple_kms_helper.h>
 #include <drm/tinydrm/tinydrm-helpers.h>
 
 #define REPAPER_RID_G2_COG_ID	0x12
@@ -59,7 +62,8 @@ enum repaper_epd_border_byte {
 };
 
 struct repaper_epd {
-	struct tinydrm_device tinydrm;
+	struct drm_device drm;
+	struct drm_simple_display_pipe pipe;
 	struct spi_device *spi;
 
 	struct gpio_desc *panel_on;
@@ -88,10 +92,9 @@ struct repaper_epd {
 	bool partial;
 };
 
-static inline struct repaper_epd *
-epd_from_tinydrm(struct tinydrm_device *tdev)
+static inline struct repaper_epd *drm_to_epd(struct drm_device *drm)
 {
-	return container_of(tdev, struct repaper_epd, tinydrm);
+	return container_of(drm, struct repaper_epd, drm);
 }
 
 static int repaper_spi_transfer(struct spi_device *spi, u8 header,
@@ -529,11 +532,16 @@ static int repaper_fb_dirty(struct drm_framebuffer *fb)
 {
 	struct drm_gem_cma_object *cma_obj = drm_fb_cma_get_gem_obj(fb, 0);
 	struct dma_buf_attachment *import_attach = cma_obj->base.import_attach;
-	struct tinydrm_device *tdev = fb->dev->dev_private;
-	struct repaper_epd *epd = epd_from_tinydrm(tdev);
+	struct repaper_epd *epd = drm_to_epd(fb->dev);
 	struct drm_rect clip;
+	int idx, ret = 0;
 	u8 *buf = NULL;
-	int ret = 0;
+
+	if (!epd->enabled)
+		return 0;
+
+	if (!drm_dev_enter(fb->dev, &idx))
+		return -ENODEV;
 
 	/* repaper can't do partial updates */
 	clip.x1 = 0;
@@ -541,17 +549,16 @@ static int repaper_fb_dirty(struct drm_framebuffer *fb)
 	clip.y1 = 0;
 	clip.y2 = fb->height;
 
-	if (!epd->enabled)
-		return 0;
-
 	repaper_get_temperature(epd);
 
 	DRM_DEBUG("Flushing [FB:%d] st=%ums\n", fb->base.id,
 		  epd->factored_stage_time);
 
 	buf = kmalloc_array(fb->width, fb->height, GFP_KERNEL);
-	if (!buf)
-		return -ENOMEM;
+	if (!buf) {
+		ret = -ENOMEM;
+		goto out_exit;
+	}
 
 	if (import_attach) {
 		ret = dma_buf_begin_cpu_access(import_attach->dmabuf,
@@ -560,7 +567,7 @@ static int repaper_fb_dirty(struct drm_framebuffer *fb)
 			goto out_free;
 	}
 
-	tinydrm_xrgb8888_to_gray8(buf, cma_obj->vaddr, fb, &clip);
+	drm_fb_xrgb8888_to_gray8(buf, cma_obj->vaddr, fb, &clip);
 
 	if (import_attach) {
 		ret = dma_buf_end_cpu_access(import_attach->dmabuf,
@@ -620,6 +627,8 @@ static int repaper_fb_dirty(struct drm_framebuffer *fb)
 
 out_free:
 	kfree(buf);
+out_exit:
+	drm_dev_exit(idx);
 
 	return ret;
 }
@@ -645,12 +654,14 @@ static void repaper_pipe_enable(struct drm_simple_display_pipe *pipe,
 				struct drm_crtc_state *crtc_state,
 				struct drm_plane_state *plane_state)
 {
-	struct tinydrm_device *tdev = pipe_to_tinydrm(pipe);
-	struct repaper_epd *epd = epd_from_tinydrm(tdev);
+	struct repaper_epd *epd = drm_to_epd(pipe->crtc.dev);
 	struct spi_device *spi = epd->spi;
 	struct device *dev = &spi->dev;
 	bool dc_ok = false;
-	int i, ret;
+	int i, ret, idx;
+
+	if (!drm_dev_enter(pipe->crtc.dev, &idx))
+		return;
 
 	DRM_DEBUG_DRIVER("\n");
 
@@ -689,7 +700,7 @@ static void repaper_pipe_enable(struct drm_simple_display_pipe *pipe,
 	if (!i) {
 		DRM_DEV_ERROR(dev, "timeout waiting for panel to become ready.\n");
 		power_off(epd);
-		return;
+		goto out_exit;
 	}
 
 	repaper_read_id(spi);
@@ -700,7 +711,7 @@ static void repaper_pipe_enable(struct drm_simple_display_pipe *pipe,
 		else
 			dev_err(dev, "wrong COG ID 0x%02x\n", ret);
 		power_off(epd);
-		return;
+		goto out_exit;
 	}
 
 	/* Disable OE */
@@ -713,7 +724,7 @@ static void repaper_pipe_enable(struct drm_simple_display_pipe *pipe,
 		else
 			DRM_DEV_ERROR(dev, "panel is reported broken\n");
 		power_off(epd);
-		return;
+		goto out_exit;
 	}
 
 	/* Power saving mode */
@@ -753,7 +764,7 @@ static void repaper_pipe_enable(struct drm_simple_display_pipe *pipe,
 		if (ret < 0) {
 			DRM_DEV_ERROR(dev, "failed to read chip (%d)\n", ret);
 			power_off(epd);
-			return;
+			goto out_exit;
 		}
 
 		if (ret & 0x40) {
@@ -765,7 +776,7 @@ static void repaper_pipe_enable(struct drm_simple_display_pipe *pipe,
 	if (!dc_ok) {
 		DRM_DEV_ERROR(dev, "dc/dc failed\n");
 		power_off(epd);
-		return;
+		goto out_exit;
 	}
 
 	/*
@@ -776,15 +787,26 @@ static void repaper_pipe_enable(struct drm_simple_display_pipe *pipe,
 
 	epd->enabled = true;
 	epd->partial = false;
+out_exit:
+	drm_dev_exit(idx);
 }
 
 static void repaper_pipe_disable(struct drm_simple_display_pipe *pipe)
 {
-	struct tinydrm_device *tdev = pipe_to_tinydrm(pipe);
-	struct repaper_epd *epd = epd_from_tinydrm(tdev);
+	struct repaper_epd *epd = drm_to_epd(pipe->crtc.dev);
 	struct spi_device *spi = epd->spi;
 	unsigned int line;
 
+	/*
+	 * This callback is not protected by drm_dev_enter/exit since we want to
+	 * turn off the display on regular driver unload. It's highly unlikely
+	 * that the underlying SPI controller is gone should this be called after
+	 * unplug.
+	 */
+
+	if (!epd->enabled)
+		return;
+
 	DRM_DEBUG_DRIVER("\n");
 
 	epd->enabled = false;
@@ -855,33 +877,50 @@ static const struct drm_simple_display_pipe_funcs repaper_pipe_funcs = {
 	.prepare_fb = drm_gem_fb_simple_display_pipe_prepare_fb,
 };
 
+static const struct drm_mode_config_funcs repaper_mode_config_funcs = {
+	.fb_create = drm_gem_fb_create_with_dirty,
+	.atomic_check = drm_atomic_helper_check,
+	.atomic_commit = drm_atomic_helper_commit,
+};
+
+static void repaper_release(struct drm_device *drm)
+{
+	struct repaper_epd *epd = drm_to_epd(drm);
+
+	DRM_DEBUG_DRIVER("\n");
+
+	drm_mode_config_cleanup(drm);
+	drm_dev_fini(drm);
+	kfree(epd);
+}
+
 static const uint32_t repaper_formats[] = {
 	DRM_FORMAT_XRGB8888,
 };
 
 static const struct drm_display_mode repaper_e1144cs021_mode = {
-	TINYDRM_MODE(128, 96, 29, 22),
+	DRM_SIMPLE_MODE(128, 96, 29, 22),
 };
 
 static const u8 repaper_e1144cs021_cs[] = { 0x00, 0x00, 0x00, 0x00,
 					    0x00, 0x0f, 0xff, 0x00 };
 
 static const struct drm_display_mode repaper_e1190cs021_mode = {
-	TINYDRM_MODE(144, 128, 36, 32),
+	DRM_SIMPLE_MODE(144, 128, 36, 32),
 };
 
 static const u8 repaper_e1190cs021_cs[] = { 0x00, 0x00, 0x00, 0x03,
 					    0xfc, 0x00, 0x00, 0xff };
 
 static const struct drm_display_mode repaper_e2200cs021_mode = {
-	TINYDRM_MODE(200, 96, 46, 22),
+	DRM_SIMPLE_MODE(200, 96, 46, 22),
 };
 
 static const u8 repaper_e2200cs021_cs[] = { 0x00, 0x00, 0x00, 0x00,
 					    0x01, 0xff, 0xe0, 0x00 };
 
 static const struct drm_display_mode repaper_e2271cs021_mode = {
-	TINYDRM_MODE(264, 176, 57, 38),
+	DRM_SIMPLE_MODE(264, 176, 57, 38),
 };
 
 static const u8 repaper_e2271cs021_cs[] = { 0x00, 0x00, 0x00, 0x7f,
@@ -893,6 +932,7 @@ static struct drm_driver repaper_driver = {
 	.driver_features	= DRIVER_GEM | DRIVER_MODESET | DRIVER_PRIME |
 				  DRIVER_ATOMIC,
 	.fops			= &repaper_fops,
+	.release		= repaper_release,
 	DRM_GEM_CMA_VMAP_DRIVER_OPS,
 	.name			= "repaper",
 	.desc			= "Pervasive Displays RePaper e-ink panels",
@@ -925,11 +965,11 @@ static int repaper_probe(struct spi_device *spi)
 	const struct spi_device_id *spi_id;
 	const struct of_device_id *match;
 	struct device *dev = &spi->dev;
-	struct tinydrm_device *tdev;
 	enum repaper_model model;
 	const char *thermal_zone;
 	struct repaper_epd *epd;
 	size_t line_buffer_size;
+	struct drm_device *drm;
 	int ret;
 
 	match = of_match_device(repaper_of_match, dev);
@@ -949,10 +989,21 @@ static int repaper_probe(struct spi_device *spi)
 		}
 	}
 
-	epd = devm_kzalloc(dev, sizeof(*epd), GFP_KERNEL);
+	epd = kzalloc(sizeof(*epd), GFP_KERNEL);
 	if (!epd)
 		return -ENOMEM;
 
+	drm = &epd->drm;
+
+	ret = devm_drm_dev_init(dev, drm, &repaper_driver);
+	if (ret) {
+		kfree(epd);
+		return ret;
+	}
+
+	drm_mode_config_init(drm);
+	drm->mode_config.funcs = &repaper_mode_config_funcs;
+
 	epd->spi = spi;
 
 	epd->panel_on = devm_gpiod_get(dev, "panel-on", GPIOD_OUT_LOW);
@@ -1063,32 +1114,41 @@ static int repaper_probe(struct spi_device *spi)
 	if (!epd->current_frame)
 		return -ENOMEM;
 
-	tdev = &epd->tinydrm;
-
-	ret = devm_tinydrm_init(dev, tdev, &repaper_driver);
-	if (ret)
-		return ret;
-
-	ret = tinydrm_display_pipe_init(tdev, &repaper_pipe_funcs,
+	ret = tinydrm_display_pipe_init(drm, &epd->pipe, &repaper_pipe_funcs,
 					DRM_MODE_CONNECTOR_VIRTUAL,
 					repaper_formats,
 					ARRAY_SIZE(repaper_formats), mode, 0);
 	if (ret)
 		return ret;
 
-	drm_mode_config_reset(tdev->drm);
-	spi_set_drvdata(spi, tdev);
+	drm_mode_config_reset(drm);
+
+	ret = drm_dev_register(drm, 0);
+	if (ret)
+		return ret;
+
+	spi_set_drvdata(spi, drm);
 
 	DRM_DEBUG_DRIVER("SPI speed: %uMHz\n", spi->max_speed_hz / 1000000);
 
-	return devm_tinydrm_register(tdev);
+	drm_fbdev_generic_setup(drm, 0);
+
+	return 0;
 }
 
-static void repaper_shutdown(struct spi_device *spi)
+static int repaper_remove(struct spi_device *spi)
 {
-	struct tinydrm_device *tdev = spi_get_drvdata(spi);
+	struct drm_device *drm = spi_get_drvdata(spi);
+
+	drm_dev_unplug(drm);
+	drm_atomic_helper_shutdown(drm);
+
+	return 0;
+}
 
-	tinydrm_shutdown(tdev);
+static void repaper_shutdown(struct spi_device *spi)
+{
+	drm_atomic_helper_shutdown(spi_get_drvdata(spi));
 }
 
 static struct spi_driver repaper_spi_driver = {
@@ -1099,6 +1159,7 @@ static struct spi_driver repaper_spi_driver = {
 	},
 	.id_table = repaper_id,
 	.probe = repaper_probe,
+	.remove = repaper_remove,
 	.shutdown = repaper_shutdown,
 };
 module_spi_driver(repaper_spi_driver);
diff --git a/drivers/gpu/drm/tinydrm/st7586.c b/drivers/gpu/drm/tinydrm/st7586.c
index 01a8077954b3..36bb16a15f7e 100644
--- a/drivers/gpu/drm/tinydrm/st7586.c
+++ b/drivers/gpu/drm/tinydrm/st7586.c
@@ -17,9 +17,12 @@
 #include <linux/spi/spi.h>
 #include <video/mipi_display.h>
 
+#include <drm/drm_atomic_helper.h>
 #include <drm/drm_damage_helper.h>
 #include <drm/drm_drv.h>
 #include <drm/drm_fb_cma_helper.h>
+#include <drm/drm_fb_helper.h>
+#include <drm/drm_format_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_rect.h>
@@ -75,7 +78,7 @@ static void st7586_xrgb8888_to_gray332(u8 *dst, void *vaddr,
 	if (!buf)
 		return;
 
-	tinydrm_xrgb8888_to_gray8(buf, vaddr, fb, clip);
+	drm_fb_xrgb8888_to_gray8(buf, vaddr, fb, clip);
 	src = buf;
 
 	for (y = clip->y1; y < clip->y2; y++) {
@@ -116,14 +119,15 @@ static int st7586_buf_copy(void *dst, struct drm_framebuffer *fb,
 
 static void st7586_fb_dirty(struct drm_framebuffer *fb, struct drm_rect *rect)
 {
-	struct tinydrm_device *tdev = fb->dev->dev_private;
-	struct mipi_dbi *mipi = mipi_dbi_from_tinydrm(tdev);
-	int start, end;
-	int ret = 0;
+	struct mipi_dbi *mipi = drm_to_mipi_dbi(fb->dev);
+	int start, end, idx, ret = 0;
 
 	if (!mipi->enabled)
 		return;
 
+	if (!drm_dev_enter(fb->dev, &idx))
+		return;
+
 	/* 3 pixels per byte, so grow clip to nearest multiple of 3 */
 	rect->x1 = rounddown(rect->x1, 3);
 	rect->x2 = roundup(rect->x2, 3);
@@ -151,6 +155,8 @@ static void st7586_fb_dirty(struct drm_framebuffer *fb, struct drm_rect *rect)
 err_msg:
 	if (ret)
 		dev_err_once(fb->dev->dev, "Failed to update display %d\n", ret);
+
+	drm_dev_exit(idx);
 }
 
 static void st7586_pipe_update(struct drm_simple_display_pipe *pipe,
@@ -175,8 +181,7 @@ static void st7586_pipe_enable(struct drm_simple_display_pipe *pipe,
 			       struct drm_crtc_state *crtc_state,
 			       struct drm_plane_state *plane_state)
 {
-	struct tinydrm_device *tdev = pipe_to_tinydrm(pipe);
-	struct mipi_dbi *mipi = mipi_dbi_from_tinydrm(tdev);
+	struct mipi_dbi *mipi = drm_to_mipi_dbi(pipe->crtc.dev);
 	struct drm_framebuffer *fb = plane_state->fb;
 	struct drm_rect rect = {
 		.x1 = 0,
@@ -184,14 +189,17 @@ static void st7586_pipe_enable(struct drm_simple_display_pipe *pipe,
 		.y1 = 0,
 		.y2 = fb->height,
 	};
-	int ret;
+	int idx, ret;
 	u8 addr_mode;
 
+	if (!drm_dev_enter(pipe->crtc.dev, &idx))
+		return;
+
 	DRM_DEBUG_KMS("\n");
 
 	ret = mipi_dbi_poweron_reset(mipi);
 	if (ret)
-		return;
+		goto out_exit;
 
 	mipi_dbi_command(mipi, ST7586_AUTO_READ_CTRL, 0x9f);
 	mipi_dbi_command(mipi, ST7586_OTP_RW_CTRL, 0x00);
@@ -244,12 +252,20 @@ static void st7586_pipe_enable(struct drm_simple_display_pipe *pipe,
 	st7586_fb_dirty(fb, &rect);
 
 	mipi_dbi_command(mipi, MIPI_DCS_SET_DISPLAY_ON);
+out_exit:
+	drm_dev_exit(idx);
 }
 
 static void st7586_pipe_disable(struct drm_simple_display_pipe *pipe)
 {
-	struct tinydrm_device *tdev = pipe_to_tinydrm(pipe);
-	struct mipi_dbi *mipi = mipi_dbi_from_tinydrm(tdev);
+	struct mipi_dbi *mipi = drm_to_mipi_dbi(pipe->crtc.dev);
+
+	/*
+	 * This callback is not protected by drm_dev_enter/exit since we want to
+	 * turn off the display on regular driver unload. It's highly unlikely
+	 * that the underlying SPI controller is gone should this be called after
+	 * unplug.
+	 */
 
 	DRM_DEBUG_KMS("\n");
 
@@ -264,46 +280,6 @@ static const u32 st7586_formats[] = {
 	DRM_FORMAT_XRGB8888,
 };
 
-static int st7586_init(struct device *dev, struct mipi_dbi *mipi,
-		const struct drm_simple_display_pipe_funcs *pipe_funcs,
-		struct drm_driver *driver, const struct drm_display_mode *mode,
-		unsigned int rotation)
-{
-	size_t bufsize = (mode->vdisplay + 2) / 3 * mode->hdisplay;
-	struct tinydrm_device *tdev = &mipi->tinydrm;
-	int ret;
-
-	mutex_init(&mipi->cmdlock);
-
-	mipi->tx_buf = devm_kmalloc(dev, bufsize, GFP_KERNEL);
-	if (!mipi->tx_buf)
-		return -ENOMEM;
-
-	ret = devm_tinydrm_init(dev, tdev, driver);
-	if (ret)
-		return ret;
-
-	ret = tinydrm_display_pipe_init(tdev, pipe_funcs,
-					DRM_MODE_CONNECTOR_VIRTUAL,
-					st7586_formats,
-					ARRAY_SIZE(st7586_formats),
-					mode, rotation);
-	if (ret)
-		return ret;
-
-	drm_plane_enable_fb_damage_clips(&tdev->pipe.plane);
-
-	tdev->drm->mode_config.preferred_depth = 32;
-	mipi->rotation = rotation;
-
-	drm_mode_config_reset(tdev->drm);
-
-	DRM_DEBUG_KMS("preferred_depth=%u, rotation = %u\n",
-		      tdev->drm->mode_config.preferred_depth, rotation);
-
-	return 0;
-}
-
 static const struct drm_simple_display_pipe_funcs st7586_pipe_funcs = {
 	.enable		= st7586_pipe_enable,
 	.disable	= st7586_pipe_disable,
@@ -311,8 +287,14 @@ static const struct drm_simple_display_pipe_funcs st7586_pipe_funcs = {
 	.prepare_fb	= drm_gem_fb_simple_display_pipe_prepare_fb,
 };
 
+static const struct drm_mode_config_funcs st7586_mode_config_funcs = {
+	.fb_create = drm_gem_fb_create_with_dirty,
+	.atomic_check = drm_atomic_helper_check,
+	.atomic_commit = drm_atomic_helper_commit,
+};
+
 static const struct drm_display_mode st7586_mode = {
-	TINYDRM_MODE(178, 128, 37, 27),
+	DRM_SIMPLE_MODE(178, 128, 37, 27),
 };
 
 DEFINE_DRM_GEM_CMA_FOPS(st7586_fops);
@@ -321,6 +303,7 @@ static struct drm_driver st7586_driver = {
 	.driver_features	= DRIVER_GEM | DRIVER_MODESET | DRIVER_PRIME |
 				  DRIVER_ATOMIC,
 	.fops			= &st7586_fops,
+	.release		= mipi_dbi_release,
 	DRM_GEM_CMA_VMAP_DRIVER_OPS,
 	.debugfs_init		= mipi_dbi_debugfs_init,
 	.name			= "st7586",
@@ -345,15 +328,35 @@ MODULE_DEVICE_TABLE(spi, st7586_id);
 static int st7586_probe(struct spi_device *spi)
 {
 	struct device *dev = &spi->dev;
+	struct drm_device *drm;
 	struct mipi_dbi *mipi;
 	struct gpio_desc *a0;
 	u32 rotation = 0;
+	size_t bufsize;
 	int ret;
 
-	mipi = devm_kzalloc(dev, sizeof(*mipi), GFP_KERNEL);
+	mipi = kzalloc(sizeof(*mipi), GFP_KERNEL);
 	if (!mipi)
 		return -ENOMEM;
 
+	drm = &mipi->drm;
+	ret = devm_drm_dev_init(dev, drm, &st7586_driver);
+	if (ret) {
+		kfree(mipi);
+		return ret;
+	}
+
+	drm_mode_config_init(drm);
+	drm->mode_config.preferred_depth = 32;
+	drm->mode_config.funcs = &st7586_mode_config_funcs;
+
+	mutex_init(&mipi->cmdlock);
+
+	bufsize = (st7586_mode.vdisplay + 2) / 3 * st7586_mode.hdisplay;
+	mipi->tx_buf = devm_kmalloc(dev, bufsize, GFP_KERNEL);
+	if (!mipi->tx_buf)
+		return -ENOMEM;
+
 	mipi->reset = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH);
 	if (IS_ERR(mipi->reset)) {
 		DRM_DEV_ERROR(dev, "Failed to get gpio 'reset'\n");
@@ -367,6 +370,7 @@ static int st7586_probe(struct spi_device *spi)
 	}
 
 	device_property_read_u32(dev, "rotation", &rotation);
+	mipi->rotation = rotation;
 
 	ret = mipi_dbi_spi_init(spi, mipi, a0);
 	if (ret)
@@ -384,21 +388,44 @@ static int st7586_probe(struct spi_device *spi)
 	 */
 	mipi->swap_bytes = true;
 
-	ret = st7586_init(&spi->dev, mipi, &st7586_pipe_funcs, &st7586_driver,
-			  &st7586_mode, rotation);
+	ret = tinydrm_display_pipe_init(drm, &mipi->pipe, &st7586_pipe_funcs,
+					DRM_MODE_CONNECTOR_VIRTUAL,
+					st7586_formats, ARRAY_SIZE(st7586_formats),
+					&st7586_mode, rotation);
 	if (ret)
 		return ret;
 
-	spi_set_drvdata(spi, mipi);
+	drm_plane_enable_fb_damage_clips(&mipi->pipe.plane);
 
-	return devm_tinydrm_register(&mipi->tinydrm);
+	drm_mode_config_reset(drm);
+
+	ret = drm_dev_register(drm, 0);
+	if (ret)
+		return ret;
+
+	spi_set_drvdata(spi, drm);
+
+	DRM_DEBUG_KMS("preferred_depth=%u, rotation = %u\n",
+		      drm->mode_config.preferred_depth, rotation);
+
+	drm_fbdev_generic_setup(drm, 0);
+
+	return 0;
 }
 
-static void st7586_shutdown(struct spi_device *spi)
+static int st7586_remove(struct spi_device *spi)
 {
-	struct mipi_dbi *mipi = spi_get_drvdata(spi);
+	struct drm_device *drm = spi_get_drvdata(spi);
+
+	drm_dev_unplug(drm);
+	drm_atomic_helper_shutdown(drm);
+
+	return 0;
+}
 
-	tinydrm_shutdown(&mipi->tinydrm);
+static void st7586_shutdown(struct spi_device *spi)
+{
+	drm_atomic_helper_shutdown(spi_get_drvdata(spi));
 }
 
 static struct spi_driver st7586_spi_driver = {
@@ -409,6 +436,7 @@ static struct spi_driver st7586_spi_driver = {
 	},
 	.id_table = st7586_id,
 	.probe = st7586_probe,
+	.remove = st7586_remove,
 	.shutdown = st7586_shutdown,
 };
 module_spi_driver(st7586_spi_driver);
diff --git a/drivers/gpu/drm/tinydrm/st7735r.c b/drivers/gpu/drm/tinydrm/st7735r.c
index 3bab9a9569a6..ce9109e613e0 100644
--- a/drivers/gpu/drm/tinydrm/st7735r.c
+++ b/drivers/gpu/drm/tinydrm/st7735r.c
@@ -14,7 +14,9 @@
 #include <linux/spi/spi.h>
 #include <video/mipi_display.h>
 
+#include <drm/drm_atomic_helper.h>
 #include <drm/drm_drv.h>
+#include <drm/drm_fb_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/tinydrm/mipi-dbi.h>
@@ -41,16 +43,18 @@ static void jd_t18003_t01_pipe_enable(struct drm_simple_display_pipe *pipe,
 				      struct drm_crtc_state *crtc_state,
 				      struct drm_plane_state *plane_state)
 {
-	struct tinydrm_device *tdev = pipe_to_tinydrm(pipe);
-	struct mipi_dbi *mipi = mipi_dbi_from_tinydrm(tdev);
-	int ret;
+	struct mipi_dbi *mipi = drm_to_mipi_dbi(pipe->crtc.dev);
+	int ret, idx;
 	u8 addr_mode;
 
+	if (!drm_dev_enter(pipe->crtc.dev, &idx))
+		return;
+
 	DRM_DEBUG_KMS("\n");
 
 	ret = mipi_dbi_poweron_reset(mipi);
 	if (ret)
-		return;
+		goto out_exit;
 
 	msleep(150);
 
@@ -101,6 +105,8 @@ static void jd_t18003_t01_pipe_enable(struct drm_simple_display_pipe *pipe,
 	msleep(20);
 
 	mipi_dbi_enable_flush(mipi, crtc_state, plane_state);
+out_exit:
+	drm_dev_exit(idx);
 }
 
 static const struct drm_simple_display_pipe_funcs jd_t18003_t01_pipe_funcs = {
@@ -111,7 +117,7 @@ static const struct drm_simple_display_pipe_funcs jd_t18003_t01_pipe_funcs = {
 };
 
 static const struct drm_display_mode jd_t18003_t01_mode = {
-	TINYDRM_MODE(128, 160, 28, 35),
+	DRM_SIMPLE_MODE(128, 160, 28, 35),
 };
 
 DEFINE_DRM_GEM_CMA_FOPS(st7735r_fops);
@@ -120,6 +126,7 @@ static struct drm_driver st7735r_driver = {
 	.driver_features	= DRIVER_GEM | DRIVER_MODESET | DRIVER_PRIME |
 				  DRIVER_ATOMIC,
 	.fops			= &st7735r_fops,
+	.release		= mipi_dbi_release,
 	DRM_GEM_CMA_VMAP_DRIVER_OPS,
 	.debugfs_init		= mipi_dbi_debugfs_init,
 	.name			= "st7735r",
@@ -144,15 +151,25 @@ MODULE_DEVICE_TABLE(spi, st7735r_id);
 static int st7735r_probe(struct spi_device *spi)
 {
 	struct device *dev = &spi->dev;
+	struct drm_device *drm;
 	struct mipi_dbi *mipi;
 	struct gpio_desc *dc;
 	u32 rotation = 0;
 	int ret;
 
-	mipi = devm_kzalloc(dev, sizeof(*mipi), GFP_KERNEL);
+	mipi = kzalloc(sizeof(*mipi), GFP_KERNEL);
 	if (!mipi)
 		return -ENOMEM;
 
+	drm = &mipi->drm;
+	ret = devm_drm_dev_init(dev, drm, &st7735r_driver);
+	if (ret) {
+		kfree(mipi);
+		return ret;
+	}
+
+	drm_mode_config_init(drm);
+
 	mipi->reset = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH);
 	if (IS_ERR(mipi->reset)) {
 		DRM_DEV_ERROR(dev, "Failed to get gpio 'reset'\n");
@@ -178,21 +195,36 @@ static int st7735r_probe(struct spi_device *spi)
 	/* Cannot read from Adafruit 1.8" display via SPI */
 	mipi->read_commands = NULL;
 
-	ret = mipi_dbi_init(&spi->dev, mipi, &jd_t18003_t01_pipe_funcs,
-			    &st7735r_driver, &jd_t18003_t01_mode, rotation);
+	ret = mipi_dbi_init(mipi, &jd_t18003_t01_pipe_funcs, &jd_t18003_t01_mode, rotation);
 	if (ret)
 		return ret;
 
-	spi_set_drvdata(spi, mipi);
+	drm_mode_config_reset(drm);
 
-	return devm_tinydrm_register(&mipi->tinydrm);
+	ret = drm_dev_register(drm, 0);
+	if (ret)
+		return ret;
+
+	spi_set_drvdata(spi, drm);
+
+	drm_fbdev_generic_setup(drm, 0);
+
+	return 0;
 }
 
-static void st7735r_shutdown(struct spi_device *spi)
+static int st7735r_remove(struct spi_device *spi)
 {
-	struct mipi_dbi *mipi = spi_get_drvdata(spi);
+	struct drm_device *drm = spi_get_drvdata(spi);
+
+	drm_dev_unplug(drm);
+	drm_atomic_helper_shutdown(drm);
 
-	tinydrm_shutdown(&mipi->tinydrm);
+	return 0;
+}
+
+static void st7735r_shutdown(struct spi_device *spi)
+{
+	drm_atomic_helper_shutdown(spi_get_drvdata(spi));
 }
 
 static struct spi_driver st7735r_spi_driver = {
@@ -203,6 +235,7 @@ static struct spi_driver st7735r_spi_driver = {
 	},
 	.id_table = st7735r_id,
 	.probe = st7735r_probe,
+	.remove = st7735r_remove,
 	.shutdown = st7735r_shutdown,
 };
 module_spi_driver(st7735r_spi_driver);
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index 1a01669b159a..2845fceb2fbd 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -1626,7 +1626,6 @@ EXPORT_SYMBOL(ttm_bo_device_release);
 int ttm_bo_device_init(struct ttm_bo_device *bdev,
 		       struct ttm_bo_driver *driver,
 		       struct address_space *mapping,
-		       uint64_t file_page_offset,
 		       bool need_dma32)
 {
 	struct ttm_bo_global *glob = &ttm_bo_glob;
@@ -1648,8 +1647,9 @@ int ttm_bo_device_init(struct ttm_bo_device *bdev,
 	if (unlikely(ret != 0))
 		goto out_no_sys;
 
-	drm_vma_offset_manager_init(&bdev->vma_manager, file_page_offset,
-				    0x10000000);
+	drm_vma_offset_manager_init(&bdev->vma_manager,
+				    DRM_FILE_PAGE_OFFSET_START,
+				    DRM_FILE_PAGE_OFFSET_SIZE);
 	INIT_DELAYED_WORK(&bdev->wq, ttm_bo_delayed_workqueue);
 	INIT_LIST_HEAD(&bdev->ddestroy);
 	bdev->dev_mapping = mapping;
diff --git a/drivers/gpu/drm/ttm/ttm_bo_vm.c b/drivers/gpu/drm/ttm/ttm_bo_vm.c
index e86a29a1e51f..6dacff49c1cc 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_vm.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_vm.c
@@ -432,6 +432,9 @@ int ttm_bo_mmap(struct file *filp, struct vm_area_struct *vma,
 	struct ttm_buffer_object *bo;
 	int ret;
 
+	if (unlikely(vma->vm_pgoff < DRM_FILE_PAGE_OFFSET_START))
+		return -EINVAL;
+
 	bo = ttm_bo_vm_lookup(bdev, vma->vm_pgoff, vma_pages(vma));
 	if (unlikely(!bo))
 		return -EINVAL;
diff --git a/drivers/gpu/drm/ttm/ttm_execbuf_util.c b/drivers/gpu/drm/ttm/ttm_execbuf_util.c
index 93860346c426..0075eb9a0b52 100644
--- a/drivers/gpu/drm/ttm/ttm_execbuf_util.c
+++ b/drivers/gpu/drm/ttm/ttm_execbuf_util.c
@@ -188,13 +188,11 @@ void ttm_eu_fence_buffer_objects(struct ww_acquire_ctx *ticket,
 	struct ttm_validate_buffer *entry;
 	struct ttm_buffer_object *bo;
 	struct ttm_bo_global *glob;
-	struct ttm_bo_device *bdev;
 
 	if (list_empty(list))
 		return;
 
 	bo = list_first_entry(list, struct ttm_validate_buffer, head)->bo;
-	bdev = bo->bdev;
 	glob = bo->bdev->glob;
 
 	spin_lock(&glob->lru_lock);
diff --git a/drivers/gpu/drm/ttm/ttm_memory.c b/drivers/gpu/drm/ttm/ttm_memory.c
index 9a0909decb36..8617958b7ae6 100644
--- a/drivers/gpu/drm/ttm/ttm_memory.c
+++ b/drivers/gpu/drm/ttm/ttm_memory.c
@@ -81,7 +81,7 @@ static void ttm_mem_zone_kobj_release(struct kobject *kobj)
 	struct ttm_mem_zone *zone =
 		container_of(kobj, struct ttm_mem_zone, kobj);
 
-	pr_info("Zone %7s: Used memory at exit: %llu kiB\n",
+	pr_info("Zone %7s: Used memory at exit: %llu KiB\n",
 		zone->name, (unsigned long long)zone->used_mem >> 10);
 	kfree(zone);
 }
@@ -448,7 +448,7 @@ int ttm_mem_global_init(struct ttm_mem_global *glob)
 #endif
 	for (i = 0; i < glob->num_zones; ++i) {
 		zone = glob->zones[i];
-		pr_info("Zone %7s: Available graphics memory: %llu kiB\n",
+		pr_info("Zone %7s: Available graphics memory: %llu KiB\n",
 			zone->name, (unsigned long long)zone->max_mem >> 10);
 	}
 	ttm_page_alloc_init(glob, glob->zone_kernel->max_mem/(2*PAGE_SIZE));
@@ -523,7 +523,7 @@ static void ttm_mem_global_free_zone(struct ttm_mem_global *glob,
 void ttm_mem_global_free(struct ttm_mem_global *glob,
 			 uint64_t amount)
 {
-	return ttm_mem_global_free_zone(glob, NULL, amount);
+	return ttm_mem_global_free_zone(glob, glob->zone_kernel, amount);
 }
 EXPORT_SYMBOL(ttm_mem_global_free);
 
@@ -622,10 +622,10 @@ int ttm_mem_global_alloc(struct ttm_mem_global *glob, uint64_t memory,
 {
 	/**
 	 * Normal allocations of kernel memory are registered in
-	 * all zones.
+	 * the kernel zone.
 	 */
 
-	return ttm_mem_global_alloc_zone(glob, NULL, memory, ctx);
+	return ttm_mem_global_alloc_zone(glob, glob->zone_kernel, memory, ctx);
 }
 EXPORT_SYMBOL(ttm_mem_global_alloc);
 
diff --git a/drivers/gpu/drm/tve200/tve200_display.c b/drivers/gpu/drm/tve200/tve200_display.c
index e8723a2412a6..d775d10dbe6a 100644
--- a/drivers/gpu/drm/tve200/tve200_display.c
+++ b/drivers/gpu/drm/tve200/tve200_display.c
@@ -149,7 +149,8 @@ static void tve200_display_enable(struct drm_simple_display_pipe *pipe,
 	/* Vsync IRQ at start of Vsync at first */
 	ctrl1 |= TVE200_VSTSTYPE_VSYNC;
 
-	if (connector->display_info.bus_flags & DRM_BUS_FLAG_PIXDATA_NEGEDGE)
+	if (connector->display_info.bus_flags &
+	    DRM_BUS_FLAG_PIXDATA_DRIVE_NEGEDGE)
 		ctrl1 |= TVE200_CTRL_TVCLKP;
 
 	if ((mode->hdisplay == 352 && mode->vdisplay == 240) || /* SIF(525) */
diff --git a/drivers/gpu/drm/udl/udl_drv.c b/drivers/gpu/drm/udl/udl_drv.c
index ff47f890e6ad..312bf324841a 100644
--- a/drivers/gpu/drm/udl/udl_drv.c
+++ b/drivers/gpu/drm/udl/udl_drv.c
@@ -48,10 +48,16 @@ static const struct file_operations udl_driver_fops = {
 	.llseek = noop_llseek,
 };
 
+static void udl_driver_release(struct drm_device *dev)
+{
+	udl_fini(dev);
+	udl_modeset_cleanup(dev);
+	drm_dev_fini(dev);
+	kfree(dev);
+}
+
 static struct drm_driver driver = {
 	.driver_features = DRIVER_MODESET | DRIVER_GEM | DRIVER_PRIME,
-	.load = udl_driver_load,
-	.unload = udl_driver_unload,
 	.release = udl_driver_release,
 
 	/* gem hooks */
@@ -75,28 +81,56 @@ static struct drm_driver driver = {
 	.patchlevel = DRIVER_PATCHLEVEL,
 };
 
+static struct udl_device *udl_driver_create(struct usb_interface *interface)
+{
+	struct usb_device *udev = interface_to_usbdev(interface);
+	struct udl_device *udl;
+	int r;
+
+	udl = kzalloc(sizeof(*udl), GFP_KERNEL);
+	if (!udl)
+		return ERR_PTR(-ENOMEM);
+
+	r = drm_dev_init(&udl->drm, &driver, &interface->dev);
+	if (r) {
+		kfree(udl);
+		return ERR_PTR(r);
+	}
+
+	udl->udev = udev;
+	udl->drm.dev_private = udl;
+
+	r = udl_init(udl);
+	if (r) {
+		drm_dev_fini(&udl->drm);
+		kfree(udl);
+		return ERR_PTR(r);
+	}
+
+	usb_set_intfdata(interface, udl);
+	return udl;
+}
+
 static int udl_usb_probe(struct usb_interface *interface,
 			 const struct usb_device_id *id)
 {
-	struct usb_device *udev = interface_to_usbdev(interface);
-	struct drm_device *dev;
 	int r;
+	struct udl_device *udl;
 
-	dev = drm_dev_alloc(&driver, &interface->dev);
-	if (IS_ERR(dev))
-		return PTR_ERR(dev);
+	udl = udl_driver_create(interface);
+	if (IS_ERR(udl))
+		return PTR_ERR(udl);
 
-	r = drm_dev_register(dev, (unsigned long)udev);
+	r = drm_dev_register(&udl->drm, 0);
 	if (r)
 		goto err_free;
 
-	usb_set_intfdata(interface, dev);
-	DRM_INFO("Initialized udl on minor %d\n", dev->primary->index);
+	DRM_INFO("Initialized udl on minor %d\n", udl->drm.primary->index);
 
 	return 0;
 
 err_free:
-	drm_dev_put(dev);
+	drm_dev_put(&udl->drm);
 	return r;
 }
 
@@ -108,6 +142,7 @@ static void udl_usb_disconnect(struct usb_interface *interface)
 	udl_fbdev_unplug(dev);
 	udl_drop_usb(dev);
 	drm_dev_unplug(dev);
+	drm_dev_put(dev);
 }
 
 /*
diff --git a/drivers/gpu/drm/udl/udl_drv.h b/drivers/gpu/drm/udl/udl_drv.h
index 4ae67d882eae..35c1f33fbc1a 100644
--- a/drivers/gpu/drm/udl/udl_drv.h
+++ b/drivers/gpu/drm/udl/udl_drv.h
@@ -50,8 +50,8 @@ struct urb_list {
 struct udl_fbdev;
 
 struct udl_device {
+	struct drm_device drm;
 	struct device *dev;
-	struct drm_device *ddev;
 	struct usb_device *udev;
 	struct drm_crtc *crtc;
 
@@ -71,6 +71,8 @@ struct udl_device {
 	atomic_t cpu_kcycles_used; /* transpired during pixel processing */
 };
 
+#define to_udl(x) container_of(x, struct udl_device, drm)
+
 struct udl_gem_object {
 	struct drm_gem_object base;
 	struct page **pages;
@@ -102,9 +104,8 @@ struct urb *udl_get_urb(struct drm_device *dev);
 int udl_submit_urb(struct drm_device *dev, struct urb *urb, size_t len);
 void udl_urb_completion(struct urb *urb);
 
-int udl_driver_load(struct drm_device *dev, unsigned long flags);
-void udl_driver_unload(struct drm_device *dev);
-void udl_driver_release(struct drm_device *dev);
+int udl_init(struct udl_device *udl);
+void udl_fini(struct drm_device *dev);
 
 int udl_fbdev_init(struct drm_device *dev);
 void udl_fbdev_cleanup(struct drm_device *dev);
diff --git a/drivers/gpu/drm/udl/udl_fb.c b/drivers/gpu/drm/udl/udl_fb.c
index dd9ffded223b..b9b67a546d4c 100644
--- a/drivers/gpu/drm/udl/udl_fb.c
+++ b/drivers/gpu/drm/udl/udl_fb.c
@@ -32,7 +32,7 @@ module_param(fb_bpp, int, S_IWUSR | S_IRUSR | S_IWGRP | S_IRGRP);
 module_param(fb_defio, int, S_IWUSR | S_IRUSR | S_IWGRP | S_IRGRP);
 
 struct udl_fbdev {
-	struct drm_fb_helper helper;
+	struct drm_fb_helper helper; /* must be first */
 	struct udl_framebuffer ufb;
 	int fb_count;
 };
@@ -82,7 +82,7 @@ int udl_handle_damage(struct udl_framebuffer *fb, int x, int y,
 		      int width, int height)
 {
 	struct drm_device *dev = fb->base.dev;
-	struct udl_device *udl = dev->dev_private;
+	struct udl_device *udl = to_udl(dev);
 	int i, ret;
 	char *cmd;
 	cycles_t start_cycles, end_cycles;
@@ -210,10 +210,10 @@ static int udl_fb_open(struct fb_info *info, int user)
 {
 	struct udl_fbdev *ufbdev = info->par;
 	struct drm_device *dev = ufbdev->ufb.base.dev;
-	struct udl_device *udl = dev->dev_private;
+	struct udl_device *udl = to_udl(dev);
 
 	/* If the USB device is gone, we don't accept new opens */
-	if (drm_dev_is_unplugged(udl->ddev))
+	if (drm_dev_is_unplugged(&udl->drm))
 		return -ENODEV;
 
 	ufbdev->fb_count++;
@@ -392,7 +392,6 @@ static int udlfb_create(struct drm_fb_helper *helper,
 		ret = PTR_ERR(info);
 		goto out_gfree;
 	}
-	info->par = ufbdev;
 
 	ret = udl_framebuffer_init(dev, &ufbdev->ufb, &mode_cmd, obj);
 	if (ret)
@@ -402,15 +401,12 @@ static int udlfb_create(struct drm_fb_helper *helper,
 
 	ufbdev->helper.fb = fb;
 
-	strcpy(info->fix.id, "udldrmfb");
-
 	info->screen_base = ufbdev->ufb.obj->vmapping;
 	info->fix.smem_len = size;
 	info->fix.smem_start = (unsigned long)ufbdev->ufb.obj->vmapping;
 
 	info->fbops = &udlfb_ops;
-	drm_fb_helper_fill_fix(info, fb->pitches[0], fb->format->depth);
-	drm_fb_helper_fill_var(info, &ufbdev->helper, sizes->fb_width, sizes->fb_height);
+	drm_fb_helper_fill_info(info, &ufbdev->helper, sizes);
 
 	DRM_DEBUG_KMS("allocated %dx%d vmal %p\n",
 		      fb->width, fb->height,
@@ -441,7 +437,7 @@ static void udl_fbdev_destroy(struct drm_device *dev,
 
 int udl_fbdev_init(struct drm_device *dev)
 {
-	struct udl_device *udl = dev->dev_private;
+	struct udl_device *udl = to_udl(dev);
 	int bpp_sel = fb_bpp;
 	struct udl_fbdev *ufbdev;
 	int ret;
@@ -480,7 +476,7 @@ free:
 
 void udl_fbdev_cleanup(struct drm_device *dev)
 {
-	struct udl_device *udl = dev->dev_private;
+	struct udl_device *udl = to_udl(dev);
 	if (!udl->fbdev)
 		return;
 
@@ -491,7 +487,7 @@ void udl_fbdev_cleanup(struct drm_device *dev)
 
 void udl_fbdev_unplug(struct drm_device *dev)
 {
-	struct udl_device *udl = dev->dev_private;
+	struct udl_device *udl = to_udl(dev);
 	struct udl_fbdev *ufbdev;
 	if (!udl->fbdev)
 		return;
diff --git a/drivers/gpu/drm/udl/udl_gem.c b/drivers/gpu/drm/udl/udl_gem.c
index bb7b58407039..3b3e17652bb2 100644
--- a/drivers/gpu/drm/udl/udl_gem.c
+++ b/drivers/gpu/drm/udl/udl_gem.c
@@ -203,7 +203,7 @@ int udl_gem_mmap(struct drm_file *file, struct drm_device *dev,
 {
 	struct udl_gem_object *gobj;
 	struct drm_gem_object *obj;
-	struct udl_device *udl = dev->dev_private;
+	struct udl_device *udl = to_udl(dev);
 	int ret = 0;
 
 	mutex_lock(&udl->gem_lock);
diff --git a/drivers/gpu/drm/udl/udl_main.c b/drivers/gpu/drm/udl/udl_main.c
index 1f8ef34ade24..6743eaef4594 100644
--- a/drivers/gpu/drm/udl/udl_main.c
+++ b/drivers/gpu/drm/udl/udl_main.c
@@ -30,7 +30,7 @@
 static int udl_parse_vendor_descriptor(struct drm_device *dev,
 				       struct usb_device *usbdev)
 {
-	struct udl_device *udl = dev->dev_private;
+	struct udl_device *udl = to_udl(dev);
 	char *desc;
 	char *buf;
 	char *desc_end;
@@ -166,7 +166,7 @@ void udl_urb_completion(struct urb *urb)
 
 static void udl_free_urb_list(struct drm_device *dev)
 {
-	struct udl_device *udl = dev->dev_private;
+	struct udl_device *udl = to_udl(dev);
 	int count = udl->urbs.count;
 	struct list_head *node;
 	struct urb_node *unode;
@@ -199,7 +199,7 @@ static void udl_free_urb_list(struct drm_device *dev)
 
 static int udl_alloc_urb_list(struct drm_device *dev, int count, size_t size)
 {
-	struct udl_device *udl = dev->dev_private;
+	struct udl_device *udl = to_udl(dev);
 	struct urb *urb;
 	struct urb_node *unode;
 	char *buf;
@@ -263,7 +263,7 @@ retry:
 
 struct urb *udl_get_urb(struct drm_device *dev)
 {
-	struct udl_device *udl = dev->dev_private;
+	struct udl_device *udl = to_udl(dev);
 	int ret = 0;
 	struct list_head *entry;
 	struct urb_node *unode;
@@ -296,7 +296,7 @@ error:
 
 int udl_submit_urb(struct drm_device *dev, struct urb *urb, size_t len)
 {
-	struct udl_device *udl = dev->dev_private;
+	struct udl_device *udl = to_udl(dev);
 	int ret;
 
 	BUG_ON(len > udl->urbs.size);
@@ -311,20 +311,12 @@ int udl_submit_urb(struct drm_device *dev, struct urb *urb, size_t len)
 	return ret;
 }
 
-int udl_driver_load(struct drm_device *dev, unsigned long flags)
+int udl_init(struct udl_device *udl)
 {
-	struct usb_device *udev = (void*)flags;
-	struct udl_device *udl;
+	struct drm_device *dev = &udl->drm;
 	int ret = -ENOMEM;
 
 	DRM_DEBUG("\n");
-	udl = kzalloc(sizeof(struct udl_device), GFP_KERNEL);
-	if (!udl)
-		return -ENOMEM;
-
-	udl->udev = udev;
-	udl->ddev = dev;
-	dev->dev_private = udl;
 
 	mutex_init(&udl->gem_lock);
 
@@ -358,7 +350,6 @@ int udl_driver_load(struct drm_device *dev, unsigned long flags)
 err:
 	if (udl->urbs.count)
 		udl_free_urb_list(dev);
-	kfree(udl);
 	DRM_ERROR("%d\n", ret);
 	return ret;
 }
@@ -369,9 +360,9 @@ int udl_drop_usb(struct drm_device *dev)
 	return 0;
 }
 
-void udl_driver_unload(struct drm_device *dev)
+void udl_fini(struct drm_device *dev)
 {
-	struct udl_device *udl = dev->dev_private;
+	struct udl_device *udl = to_udl(dev);
 
 	drm_kms_helper_poll_fini(dev);
 
@@ -379,12 +370,4 @@ void udl_driver_unload(struct drm_device *dev)
 		udl_free_urb_list(dev);
 
 	udl_fbdev_cleanup(dev);
-	kfree(udl);
-}
-
-void udl_driver_release(struct drm_device *dev)
-{
-	udl_modeset_cleanup(dev);
-	drm_dev_fini(dev);
-	kfree(dev);
 }
diff --git a/drivers/gpu/drm/v3d/Kconfig b/drivers/gpu/drm/v3d/Kconfig
index 1552bf552c94..75a74c45f109 100644
--- a/drivers/gpu/drm/v3d/Kconfig
+++ b/drivers/gpu/drm/v3d/Kconfig
@@ -5,6 +5,7 @@ config DRM_V3D
 	depends on COMMON_CLK
 	depends on MMU
 	select DRM_SCHED
+	select DRM_GEM_SHMEM_HELPER
 	help
 	  Choose this option if you have a system that has a Broadcom
 	  V3D 3.x or newer GPU, such as BCM7268.
diff --git a/drivers/gpu/drm/v3d/v3d_bo.c b/drivers/gpu/drm/v3d/v3d_bo.c
index a08766d39eab..a22b75a3a533 100644
--- a/drivers/gpu/drm/v3d/v3d_bo.c
+++ b/drivers/gpu/drm/v3d/v3d_bo.c
@@ -25,162 +25,6 @@
 #include "v3d_drv.h"
 #include "uapi/drm/v3d_drm.h"
 
-/* Pins the shmem pages, fills in the .pages and .sgt fields of the BO, and maps
- * it for DMA.
- */
-static int
-v3d_bo_get_pages(struct v3d_bo *bo)
-{
-	struct drm_gem_object *obj = &bo->base;
-	struct drm_device *dev = obj->dev;
-	int npages = obj->size >> PAGE_SHIFT;
-	int ret = 0;
-
-	mutex_lock(&bo->lock);
-	if (bo->pages_refcount++ != 0)
-		goto unlock;
-
-	if (!obj->import_attach) {
-		bo->pages = drm_gem_get_pages(obj);
-		if (IS_ERR(bo->pages)) {
-			ret = PTR_ERR(bo->pages);
-			goto unlock;
-		}
-
-		bo->sgt = drm_prime_pages_to_sg(bo->pages, npages);
-		if (IS_ERR(bo->sgt)) {
-			ret = PTR_ERR(bo->sgt);
-			goto put_pages;
-		}
-
-		/* Map the pages for use by the GPU. */
-		dma_map_sg(dev->dev, bo->sgt->sgl,
-			   bo->sgt->nents, DMA_BIDIRECTIONAL);
-	} else {
-		bo->pages = kcalloc(npages, sizeof(*bo->pages), GFP_KERNEL);
-		if (!bo->pages)
-			goto put_pages;
-
-		drm_prime_sg_to_page_addr_arrays(bo->sgt, bo->pages,
-						 NULL, npages);
-
-		/* Note that dma-bufs come in mapped. */
-	}
-
-	mutex_unlock(&bo->lock);
-
-	return 0;
-
-put_pages:
-	drm_gem_put_pages(obj, bo->pages, true, true);
-	bo->pages = NULL;
-unlock:
-	bo->pages_refcount--;
-	mutex_unlock(&bo->lock);
-	return ret;
-}
-
-static void
-v3d_bo_put_pages(struct v3d_bo *bo)
-{
-	struct drm_gem_object *obj = &bo->base;
-
-	mutex_lock(&bo->lock);
-	if (--bo->pages_refcount == 0) {
-		if (!obj->import_attach) {
-			dma_unmap_sg(obj->dev->dev, bo->sgt->sgl,
-				     bo->sgt->nents, DMA_BIDIRECTIONAL);
-			sg_free_table(bo->sgt);
-			kfree(bo->sgt);
-			drm_gem_put_pages(obj, bo->pages, true, true);
-		} else {
-			kfree(bo->pages);
-		}
-	}
-	mutex_unlock(&bo->lock);
-}
-
-static struct v3d_bo *v3d_bo_create_struct(struct drm_device *dev,
-					   size_t unaligned_size)
-{
-	struct v3d_dev *v3d = to_v3d_dev(dev);
-	struct drm_gem_object *obj;
-	struct v3d_bo *bo;
-	size_t size = roundup(unaligned_size, PAGE_SIZE);
-	int ret;
-
-	if (size == 0)
-		return ERR_PTR(-EINVAL);
-
-	bo = kzalloc(sizeof(*bo), GFP_KERNEL);
-	if (!bo)
-		return ERR_PTR(-ENOMEM);
-	obj = &bo->base;
-
-	INIT_LIST_HEAD(&bo->vmas);
-	INIT_LIST_HEAD(&bo->unref_head);
-	mutex_init(&bo->lock);
-
-	ret = drm_gem_object_init(dev, obj, size);
-	if (ret)
-		goto free_bo;
-
-	spin_lock(&v3d->mm_lock);
-	ret = drm_mm_insert_node_generic(&v3d->mm, &bo->node,
-					 obj->size >> PAGE_SHIFT,
-					 GMP_GRANULARITY >> PAGE_SHIFT, 0, 0);
-	spin_unlock(&v3d->mm_lock);
-	if (ret)
-		goto free_obj;
-
-	return bo;
-
-free_obj:
-	drm_gem_object_release(obj);
-free_bo:
-	kfree(bo);
-	return ERR_PTR(ret);
-}
-
-struct v3d_bo *v3d_bo_create(struct drm_device *dev, struct drm_file *file_priv,
-			     size_t unaligned_size)
-{
-	struct v3d_dev *v3d = to_v3d_dev(dev);
-	struct drm_gem_object *obj;
-	struct v3d_bo *bo;
-	int ret;
-
-	bo = v3d_bo_create_struct(dev, unaligned_size);
-	if (IS_ERR(bo))
-		return bo;
-	obj = &bo->base;
-
-	bo->resv = &bo->_resv;
-	reservation_object_init(bo->resv);
-
-	ret = v3d_bo_get_pages(bo);
-	if (ret)
-		goto free_mm;
-
-	v3d_mmu_insert_ptes(bo);
-
-	mutex_lock(&v3d->bo_lock);
-	v3d->bo_stats.num_allocated++;
-	v3d->bo_stats.pages_allocated += obj->size >> PAGE_SHIFT;
-	mutex_unlock(&v3d->bo_lock);
-
-	return bo;
-
-free_mm:
-	spin_lock(&v3d->mm_lock);
-	drm_mm_remove_node(&bo->node);
-	spin_unlock(&v3d->mm_lock);
-
-	drm_gem_object_release(obj);
-	kfree(bo);
-	return ERR_PTR(ret);
-}
-
 /* Called DRM core on the last userspace/kernel unreference of the
  * BO.
  */
@@ -189,92 +33,116 @@ void v3d_free_object(struct drm_gem_object *obj)
 	struct v3d_dev *v3d = to_v3d_dev(obj->dev);
 	struct v3d_bo *bo = to_v3d_bo(obj);
 
+	v3d_mmu_remove_ptes(bo);
+
 	mutex_lock(&v3d->bo_lock);
 	v3d->bo_stats.num_allocated--;
 	v3d->bo_stats.pages_allocated -= obj->size >> PAGE_SHIFT;
 	mutex_unlock(&v3d->bo_lock);
 
-	reservation_object_fini(&bo->_resv);
-
-	v3d_bo_put_pages(bo);
-
-	if (obj->import_attach)
-		drm_prime_gem_destroy(obj, bo->sgt);
-
-	v3d_mmu_remove_ptes(bo);
 	spin_lock(&v3d->mm_lock);
 	drm_mm_remove_node(&bo->node);
 	spin_unlock(&v3d->mm_lock);
 
-	mutex_destroy(&bo->lock);
+	/* GPU execution may have dirtied any pages in the BO. */
+	bo->base.pages_mark_dirty_on_put = true;
 
-	drm_gem_object_release(obj);
-	kfree(bo);
+	drm_gem_shmem_free_object(obj);
 }
 
-struct reservation_object *v3d_prime_res_obj(struct drm_gem_object *obj)
+static const struct drm_gem_object_funcs v3d_gem_funcs = {
+	.free = v3d_free_object,
+	.print_info = drm_gem_shmem_print_info,
+	.pin = drm_gem_shmem_pin,
+	.unpin = drm_gem_shmem_unpin,
+	.get_sg_table = drm_gem_shmem_get_sg_table,
+	.vmap = drm_gem_shmem_vmap,
+	.vunmap = drm_gem_shmem_vunmap,
+	.vm_ops = &drm_gem_shmem_vm_ops,
+};
+
+/* gem_create_object function for allocating a BO struct and doing
+ * early setup.
+ */
+struct drm_gem_object *v3d_create_object(struct drm_device *dev, size_t size)
 {
-	struct v3d_bo *bo = to_v3d_bo(obj);
+	struct v3d_bo *bo;
+	struct drm_gem_object *obj;
 
-	return bo->resv;
-}
+	if (size == 0)
+		return NULL;
 
-static void
-v3d_set_mmap_vma_flags(struct vm_area_struct *vma)
-{
-	vma->vm_flags &= ~VM_PFNMAP;
-	vma->vm_flags |= VM_MIXEDMAP;
-	vma->vm_page_prot = pgprot_writecombine(vm_get_page_prot(vma->vm_flags));
-}
+	bo = kzalloc(sizeof(*bo), GFP_KERNEL);
+	if (!bo)
+		return NULL;
+	obj = &bo->base.base;
 
-vm_fault_t v3d_gem_fault(struct vm_fault *vmf)
-{
-	struct vm_area_struct *vma = vmf->vma;
-	struct drm_gem_object *obj = vma->vm_private_data;
-	struct v3d_bo *bo = to_v3d_bo(obj);
-	pfn_t pfn;
-	pgoff_t pgoff;
+	obj->funcs = &v3d_gem_funcs;
 
-	/* We don't use vmf->pgoff since that has the fake offset: */
-	pgoff = (vmf->address - vma->vm_start) >> PAGE_SHIFT;
-	pfn = __pfn_to_pfn_t(page_to_pfn(bo->pages[pgoff]), PFN_DEV);
+	INIT_LIST_HEAD(&bo->unref_head);
 
-	return vmf_insert_mixed(vma, vmf->address, pfn);
+	return &bo->base.base;
 }
 
-int v3d_mmap(struct file *filp, struct vm_area_struct *vma)
+static int
+v3d_bo_create_finish(struct drm_gem_object *obj)
 {
+	struct v3d_dev *v3d = to_v3d_dev(obj->dev);
+	struct v3d_bo *bo = to_v3d_bo(obj);
+	struct sg_table *sgt;
 	int ret;
 
-	ret = drm_gem_mmap(filp, vma);
+	/* So far we pin the BO in the MMU for its lifetime, so use
+	 * shmem's helper for getting a lifetime sgt.
+	 */
+	sgt = drm_gem_shmem_get_pages_sgt(&bo->base.base);
+	if (IS_ERR(sgt))
+		return PTR_ERR(sgt);
+
+	spin_lock(&v3d->mm_lock);
+	/* Allocate the object's space in the GPU's page tables.
+	 * Inserting PTEs will happen later, but the offset is for the
+	 * lifetime of the BO.
+	 */
+	ret = drm_mm_insert_node_generic(&v3d->mm, &bo->node,
+					 obj->size >> PAGE_SHIFT,
+					 GMP_GRANULARITY >> PAGE_SHIFT, 0, 0);
+	spin_unlock(&v3d->mm_lock);
 	if (ret)
 		return ret;
 
-	v3d_set_mmap_vma_flags(vma);
+	/* Track stats for /debug/dri/n/bo_stats. */
+	mutex_lock(&v3d->bo_lock);
+	v3d->bo_stats.num_allocated++;
+	v3d->bo_stats.pages_allocated += obj->size >> PAGE_SHIFT;
+	mutex_unlock(&v3d->bo_lock);
 
-	return ret;
+	v3d_mmu_insert_ptes(bo);
+
+	return 0;
 }
 
-int v3d_prime_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma)
+struct v3d_bo *v3d_bo_create(struct drm_device *dev, struct drm_file *file_priv,
+			     size_t unaligned_size)
 {
+	struct drm_gem_shmem_object *shmem_obj;
+	struct v3d_bo *bo;
 	int ret;
 
-	ret = drm_gem_mmap_obj(obj, obj->size, vma);
-	if (ret < 0)
-		return ret;
-
-	v3d_set_mmap_vma_flags(vma);
+	shmem_obj = drm_gem_shmem_create(dev, unaligned_size);
+	if (IS_ERR(shmem_obj))
+		return ERR_CAST(shmem_obj);
+	bo = to_v3d_bo(&shmem_obj->base);
 
-	return 0;
-}
+	ret = v3d_bo_create_finish(&shmem_obj->base);
+	if (ret)
+		goto free_obj;
 
-struct sg_table *
-v3d_prime_get_sg_table(struct drm_gem_object *obj)
-{
-	struct v3d_bo *bo = to_v3d_bo(obj);
-	int npages = obj->size >> PAGE_SHIFT;
+	return bo;
 
-	return drm_prime_pages_to_sg(bo->pages, npages);
+free_obj:
+	drm_gem_shmem_free_object(&shmem_obj->base);
+	return ERR_PTR(ret);
 }
 
 struct drm_gem_object *
@@ -283,20 +151,17 @@ v3d_prime_import_sg_table(struct drm_device *dev,
 			  struct sg_table *sgt)
 {
 	struct drm_gem_object *obj;
-	struct v3d_bo *bo;
-
-	bo = v3d_bo_create_struct(dev, attach->dmabuf->size);
-	if (IS_ERR(bo))
-		return ERR_CAST(bo);
-	obj = &bo->base;
-
-	bo->resv = attach->dmabuf->resv;
+	int ret;
 
-	bo->sgt = sgt;
-	obj->import_attach = attach;
-	v3d_bo_get_pages(bo);
+	obj = drm_gem_shmem_prime_import_sg_table(dev, attach, sgt);
+	if (IS_ERR(obj))
+		return obj;
 
-	v3d_mmu_insert_ptes(bo);
+	ret = v3d_bo_create_finish(obj);
+	if (ret) {
+		drm_gem_shmem_free_object(obj);
+		return ERR_PTR(ret);
+	}
 
 	return obj;
 }
@@ -319,8 +184,8 @@ int v3d_create_bo_ioctl(struct drm_device *dev, void *data,
 
 	args->offset = bo->node.start << PAGE_SHIFT;
 
-	ret = drm_gem_handle_create(file_priv, &bo->base, &args->handle);
-	drm_gem_object_put_unlocked(&bo->base);
+	ret = drm_gem_handle_create(file_priv, &bo->base.base, &args->handle);
+	drm_gem_object_put_unlocked(&bo->base.base);
 
 	return ret;
 }
@@ -330,7 +195,6 @@ int v3d_mmap_bo_ioctl(struct drm_device *dev, void *data,
 {
 	struct drm_v3d_mmap_bo *args = data;
 	struct drm_gem_object *gem_obj;
-	int ret;
 
 	if (args->flags != 0) {
 		DRM_INFO("unknown mmap_bo flags: %d\n", args->flags);
@@ -343,12 +207,10 @@ int v3d_mmap_bo_ioctl(struct drm_device *dev, void *data,
 		return -ENOENT;
 	}
 
-	ret = drm_gem_create_mmap_offset(gem_obj);
-	if (ret == 0)
-		args->offset = drm_vma_node_offset_addr(&gem_obj->vma_node);
+	args->offset = drm_vma_node_offset_addr(&gem_obj->vma_node);
 	drm_gem_object_put_unlocked(gem_obj);
 
-	return ret;
+	return 0;
 }
 
 int v3d_get_bo_offset_ioctl(struct drm_device *dev, void *data,
diff --git a/drivers/gpu/drm/v3d/v3d_debugfs.c b/drivers/gpu/drm/v3d/v3d_debugfs.c
index eb2b2d2f8553..a24af2d2f574 100644
--- a/drivers/gpu/drm/v3d/v3d_debugfs.c
+++ b/drivers/gpu/drm/v3d/v3d_debugfs.c
@@ -187,6 +187,11 @@ static int v3d_measure_clock(struct seq_file *m, void *unused)
 	uint32_t cycles;
 	int core = 0;
 	int measure_ms = 1000;
+	int ret;
+
+	ret = pm_runtime_get_sync(v3d->dev);
+	if (ret < 0)
+		return ret;
 
 	if (v3d->ver >= 40) {
 		V3D_CORE_WRITE(core, V3D_V4_PCTR_0_SRC_0_3,
@@ -210,6 +215,9 @@ static int v3d_measure_clock(struct seq_file *m, void *unused)
 		   cycles / (measure_ms * 1000),
 		   (cycles / (measure_ms * 100)) % 10);
 
+	pm_runtime_mark_last_busy(v3d->dev);
+	pm_runtime_put_autosuspend(v3d->dev);
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/v3d/v3d_drv.c b/drivers/gpu/drm/v3d/v3d_drv.c
index f0afcec72c34..a06b05f714a5 100644
--- a/drivers/gpu/drm/v3d/v3d_drv.c
+++ b/drivers/gpu/drm/v3d/v3d_drv.c
@@ -7,9 +7,9 @@
  * This driver supports the Broadcom V3D 3.3 and 4.1 OpenGL ES GPUs.
  * For V3D 2.x support, see the VC4 driver.
  *
- * Currently only single-core rendering using the binner and renderer
- * is supported.  The TFU (texture formatting unit) and V3D 4.x's CSD
- * (compute shader dispatch) are not yet supported.
+ * Currently only single-core rendering using the binner and renderer,
+ * along with TFU (texture formatting unit) rendering is supported.
+ * V3D 4.x's CSD (compute shader dispatch) is not yet supported.
  */
 
 #include <linux/clk.h>
@@ -19,6 +19,7 @@
 #include <linux/of_platform.h>
 #include <linux/platform_device.h>
 #include <linux/pm_runtime.h>
+#include <linux/reset.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_fb_helper.h>
 
@@ -101,6 +102,8 @@ static int v3d_get_param_ioctl(struct drm_device *dev, void *data,
 			return -EINVAL;
 
 		ret = pm_runtime_get_sync(v3d->dev);
+		if (ret < 0)
+			return ret;
 		if (args->param >= DRM_V3D_PARAM_V3D_CORE0_IDENT0 &&
 		    args->param <= DRM_V3D_PARAM_V3D_CORE0_IDENT2) {
 			args->value = V3D_CORE_READ(0, offset);
@@ -160,17 +163,7 @@ v3d_postclose(struct drm_device *dev, struct drm_file *file)
 	kfree(v3d_priv);
 }
 
-static const struct file_operations v3d_drm_fops = {
-	.owner = THIS_MODULE,
-	.open = drm_open,
-	.release = drm_release,
-	.unlocked_ioctl = drm_ioctl,
-	.mmap = v3d_mmap,
-	.poll = drm_poll,
-	.read = drm_read,
-	.compat_ioctl = drm_compat_ioctl,
-	.llseek = noop_llseek,
-};
+DEFINE_DRM_GEM_SHMEM_FOPS(v3d_drm_fops);
 
 /* DRM_AUTH is required on SUBMIT_CL for now, while we don't have GMP
  * protection between clients.  Note that render nodes would be be
@@ -188,12 +181,6 @@ static const struct drm_ioctl_desc v3d_drm_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(V3D_SUBMIT_TFU, v3d_submit_tfu_ioctl, DRM_RENDER_ALLOW | DRM_AUTH),
 };
 
-static const struct vm_operations_struct v3d_vm_ops = {
-	.fault = v3d_gem_fault,
-	.open = drm_gem_vm_open,
-	.close = drm_gem_vm_close,
-};
-
 static struct drm_driver v3d_drm_driver = {
 	.driver_features = (DRIVER_GEM |
 			    DRIVER_RENDER |
@@ -207,17 +194,11 @@ static struct drm_driver v3d_drm_driver = {
 	.debugfs_init = v3d_debugfs_init,
 #endif
 
-	.gem_free_object_unlocked = v3d_free_object,
-	.gem_vm_ops = &v3d_vm_ops,
-
+	.gem_create_object = v3d_create_object,
 	.prime_handle_to_fd = drm_gem_prime_handle_to_fd,
 	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
-	.gem_prime_import = drm_gem_prime_import,
-	.gem_prime_export = drm_gem_prime_export,
-	.gem_prime_res_obj = v3d_prime_res_obj,
-	.gem_prime_get_sg_table	= v3d_prime_get_sg_table,
 	.gem_prime_import_sg_table = v3d_prime_import_sg_table,
-	.gem_prime_mmap = v3d_prime_mmap,
+	.gem_prime_mmap = drm_gem_prime_mmap,
 
 	.ioctls = v3d_drm_ioctls,
 	.num_ioctls = ARRAY_SIZE(v3d_drm_ioctls),
@@ -265,10 +246,6 @@ static int v3d_platform_drm_probe(struct platform_device *pdev)
 	v3d->pdev = pdev;
 	drm = &v3d->drm;
 
-	ret = map_regs(v3d, &v3d->bridge_regs, "bridge");
-	if (ret)
-		goto dev_free;
-
 	ret = map_regs(v3d, &v3d->hub_regs, "hub");
 	if (ret)
 		goto dev_free;
@@ -283,6 +260,22 @@ static int v3d_platform_drm_probe(struct platform_device *pdev)
 	v3d->cores = V3D_GET_FIELD(ident1, V3D_HUB_IDENT1_NCORES);
 	WARN_ON(v3d->cores > 1); /* multicore not yet implemented */
 
+	v3d->reset = devm_reset_control_get_exclusive(dev, NULL);
+	if (IS_ERR(v3d->reset)) {
+		ret = PTR_ERR(v3d->reset);
+
+		if (ret == -EPROBE_DEFER)
+			goto dev_free;
+
+		v3d->reset = NULL;
+		ret = map_regs(v3d, &v3d->bridge_regs, "bridge");
+		if (ret) {
+			dev_err(dev,
+				"Failed to get reset control or bridge regs\n");
+			goto dev_free;
+		}
+	}
+
 	if (v3d->ver < 41) {
 		ret = map_regs(v3d, &v3d->gca_regs, "gca");
 		if (ret)
@@ -312,14 +305,18 @@ static int v3d_platform_drm_probe(struct platform_device *pdev)
 	if (ret)
 		goto dev_destroy;
 
-	v3d_irq_init(v3d);
+	ret = v3d_irq_init(v3d);
+	if (ret)
+		goto gem_destroy;
 
 	ret = drm_dev_register(drm, 0);
 	if (ret)
-		goto gem_destroy;
+		goto irq_disable;
 
 	return 0;
 
+irq_disable:
+	v3d_irq_disable(v3d);
 gem_destroy:
 	v3d_gem_destroy(drm);
 dev_destroy:
diff --git a/drivers/gpu/drm/v3d/v3d_drv.h b/drivers/gpu/drm/v3d/v3d_drv.h
index fdda3037f7af..e9d4a2fdcf44 100644
--- a/drivers/gpu/drm/v3d/v3d_drv.h
+++ b/drivers/gpu/drm/v3d/v3d_drv.h
@@ -1,11 +1,11 @@
 // SPDX-License-Identifier: GPL-2.0+
 /* Copyright (C) 2015-2018 Broadcom */
 
-#include <linux/reservation.h>
 #include <linux/mm_types.h>
 #include <drm/drmP.h>
 #include <drm/drm_encoder.h>
 #include <drm/drm_gem.h>
+#include <drm/drm_gem_shmem_helper.h>
 #include <drm/gpu_scheduler.h>
 #include "uapi/drm/v3d_drm.h"
 
@@ -34,6 +34,7 @@ struct v3d_dev {
 	 * and revision.
 	 */
 	int ver;
+	bool single_irq_line;
 
 	struct device *dev;
 	struct platform_device *pdev;
@@ -42,6 +43,7 @@ struct v3d_dev {
 	void __iomem *bridge_regs;
 	void __iomem *gca_regs;
 	struct clk *clk;
+	struct reset_control *reset;
 
 	/* Virtual and DMA addresses of the single shared page table. */
 	volatile u32 *pt;
@@ -109,34 +111,15 @@ struct v3d_file_priv {
 	struct drm_sched_entity sched_entity[V3D_MAX_QUEUES];
 };
 
-/* Tracks a mapping of a BO into a per-fd address space */
-struct v3d_vma {
-	struct v3d_page_table *pt;
-	struct list_head list; /* entry in v3d_bo.vmas */
-};
-
 struct v3d_bo {
-	struct drm_gem_object base;
-
-	struct mutex lock;
+	struct drm_gem_shmem_object base;
 
 	struct drm_mm_node node;
 
-	u32 pages_refcount;
-	struct page **pages;
-	struct sg_table *sgt;
-	void *vaddr;
-
-	struct list_head vmas;    /* list of v3d_vma */
-
 	/* List entry for the BO's position in
 	 * v3d_exec_info->unref_list
 	 */
 	struct list_head unref_head;
-
-	/* normally (resv == &_resv) except for imported bo's */
-	struct reservation_object *resv;
-	struct reservation_object _resv;
 };
 
 static inline struct v3d_bo *
@@ -180,7 +163,7 @@ struct v3d_job {
 	struct dma_fence *in_fence;
 
 	/* v3d fence to be signaled by IRQ handler when the job is complete. */
-	struct dma_fence *done_fence;
+	struct dma_fence *irq_fence;
 
 	/* GPU virtual addresses of the start/end of the CL job. */
 	u32 start, end;
@@ -227,7 +210,7 @@ struct v3d_tfu_job {
 	struct dma_fence *in_fence;
 
 	/* v3d fence to be signaled by IRQ handler when the job is complete. */
-	struct dma_fence *done_fence;
+	struct dma_fence *irq_fence;
 
 	struct v3d_dev *v3d;
 
@@ -270,6 +253,7 @@ static inline unsigned long nsecs_to_jiffies_timeout(const u64 n)
 }
 
 /* v3d_bo.c */
+struct drm_gem_object *v3d_create_object(struct drm_device *dev, size_t size);
 void v3d_free_object(struct drm_gem_object *gem_obj);
 struct v3d_bo *v3d_bo_create(struct drm_device *dev, struct drm_file *file_priv,
 			     size_t size);
@@ -279,11 +263,6 @@ int v3d_mmap_bo_ioctl(struct drm_device *dev, void *data,
 		      struct drm_file *file_priv);
 int v3d_get_bo_offset_ioctl(struct drm_device *dev, void *data,
 			    struct drm_file *file_priv);
-vm_fault_t v3d_gem_fault(struct vm_fault *vmf);
-int v3d_mmap(struct file *filp, struct vm_area_struct *vma);
-struct reservation_object *v3d_prime_res_obj(struct drm_gem_object *obj);
-int v3d_prime_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma);
-struct sg_table *v3d_prime_get_sg_table(struct drm_gem_object *obj);
 struct drm_gem_object *v3d_prime_import_sg_table(struct drm_device *dev,
 						 struct dma_buf_attachment *attach,
 						 struct sg_table *sgt);
@@ -310,7 +289,7 @@ void v3d_reset(struct v3d_dev *v3d);
 void v3d_invalidate_caches(struct v3d_dev *v3d);
 
 /* v3d_irq.c */
-void v3d_irq_init(struct v3d_dev *v3d);
+int v3d_irq_init(struct v3d_dev *v3d);
 void v3d_irq_enable(struct v3d_dev *v3d);
 void v3d_irq_disable(struct v3d_dev *v3d);
 void v3d_irq_reset(struct v3d_dev *v3d);
diff --git a/drivers/gpu/drm/v3d/v3d_gem.c b/drivers/gpu/drm/v3d/v3d_gem.c
index 803f31467ec1..93ff8fcbe475 100644
--- a/drivers/gpu/drm/v3d/v3d_gem.c
+++ b/drivers/gpu/drm/v3d/v3d_gem.c
@@ -6,6 +6,7 @@
 #include <linux/module.h>
 #include <linux/platform_device.h>
 #include <linux/pm_runtime.h>
+#include <linux/reset.h>
 #include <linux/device.h>
 #include <linux/io.h>
 #include <linux/sched/signal.h>
@@ -24,7 +25,8 @@ v3d_init_core(struct v3d_dev *v3d, int core)
 	 * type.  If you want the default behavior, you can still put
 	 * "2" in the indirect texture state's output_type field.
 	 */
-	V3D_CORE_WRITE(core, V3D_CTL_MISCCFG, V3D_MISCCFG_OVRTMUOUT);
+	if (v3d->ver < 40)
+		V3D_CORE_WRITE(core, V3D_CTL_MISCCFG, V3D_MISCCFG_OVRTMUOUT);
 
 	/* Whenever we flush the L2T cache, we always want to flush
 	 * the whole thing.
@@ -69,7 +71,7 @@ v3d_idle_gca(struct v3d_dev *v3d)
 }
 
 static void
-v3d_reset_v3d(struct v3d_dev *v3d)
+v3d_reset_by_bridge(struct v3d_dev *v3d)
 {
 	int version = V3D_BRIDGE_READ(V3D_TOP_GR_BRIDGE_REVISION);
 
@@ -89,6 +91,15 @@ v3d_reset_v3d(struct v3d_dev *v3d)
 				 V3D_TOP_GR_BRIDGE_SW_INIT_1_V3D_CLK_108_SW_INIT);
 		V3D_BRIDGE_WRITE(V3D_TOP_GR_BRIDGE_SW_INIT_1, 0);
 	}
+}
+
+static void
+v3d_reset_v3d(struct v3d_dev *v3d)
+{
+	if (v3d->reset)
+		reset_control_reset(v3d->reset);
+	else
+		v3d_reset_by_bridge(v3d);
 
 	v3d_init_hw_state(v3d);
 }
@@ -190,7 +201,8 @@ v3d_attach_object_fences(struct v3d_bo **bos, int bo_count,
 
 	for (i = 0; i < bo_count; i++) {
 		/* XXX: Use shared fences for read-only objects. */
-		reservation_object_add_excl_fence(bos[i]->resv, fence);
+		reservation_object_add_excl_fence(bos[i]->base.base.resv,
+						  fence);
 	}
 }
 
@@ -199,12 +211,8 @@ v3d_unlock_bo_reservations(struct v3d_bo **bos,
 			   int bo_count,
 			   struct ww_acquire_ctx *acquire_ctx)
 {
-	int i;
-
-	for (i = 0; i < bo_count; i++)
-		ww_mutex_unlock(&bos[i]->resv->lock);
-
-	ww_acquire_fini(acquire_ctx);
+	drm_gem_unlock_reservations((struct drm_gem_object **)bos, bo_count,
+				    acquire_ctx);
 }
 
 /* Takes the reservation lock on all the BOs being referenced, so that
@@ -219,58 +227,19 @@ v3d_lock_bo_reservations(struct v3d_bo **bos,
 			 int bo_count,
 			 struct ww_acquire_ctx *acquire_ctx)
 {
-	int contended_lock = -1;
 	int i, ret;
 
-	ww_acquire_init(acquire_ctx, &reservation_ww_class);
-
-retry:
-	if (contended_lock != -1) {
-		struct v3d_bo *bo = bos[contended_lock];
-
-		ret = ww_mutex_lock_slow_interruptible(&bo->resv->lock,
-						       acquire_ctx);
-		if (ret) {
-			ww_acquire_done(acquire_ctx);
-			return ret;
-		}
-	}
-
-	for (i = 0; i < bo_count; i++) {
-		if (i == contended_lock)
-			continue;
-
-		ret = ww_mutex_lock_interruptible(&bos[i]->resv->lock,
-						  acquire_ctx);
-		if (ret) {
-			int j;
-
-			for (j = 0; j < i; j++)
-				ww_mutex_unlock(&bos[j]->resv->lock);
-
-			if (contended_lock != -1 && contended_lock >= i) {
-				struct v3d_bo *bo = bos[contended_lock];
-
-				ww_mutex_unlock(&bo->resv->lock);
-			}
-
-			if (ret == -EDEADLK) {
-				contended_lock = i;
-				goto retry;
-			}
-
-			ww_acquire_done(acquire_ctx);
-			return ret;
-		}
-	}
-
-	ww_acquire_done(acquire_ctx);
+	ret = drm_gem_lock_reservations((struct drm_gem_object **)bos,
+					bo_count, acquire_ctx);
+	if (ret)
+		return ret;
 
 	/* Reserve space for our shared (read-only) fence references,
 	 * before we commit the CL to the hardware.
 	 */
 	for (i = 0; i < bo_count; i++) {
-		ret = reservation_object_reserve_shared(bos[i]->resv, 1);
+		ret = reservation_object_reserve_shared(bos[i]->base.base.resv,
+							1);
 		if (ret) {
 			v3d_unlock_bo_reservations(bos, bo_count,
 						   acquire_ctx);
@@ -371,18 +340,18 @@ v3d_exec_cleanup(struct kref *ref)
 	dma_fence_put(exec->bin.in_fence);
 	dma_fence_put(exec->render.in_fence);
 
-	dma_fence_put(exec->bin.done_fence);
-	dma_fence_put(exec->render.done_fence);
+	dma_fence_put(exec->bin.irq_fence);
+	dma_fence_put(exec->render.irq_fence);
 
 	dma_fence_put(exec->bin_done_fence);
 	dma_fence_put(exec->render_done_fence);
 
 	for (i = 0; i < exec->bo_count; i++)
-		drm_gem_object_put_unlocked(&exec->bo[i]->base);
+		drm_gem_object_put_unlocked(&exec->bo[i]->base.base);
 	kvfree(exec->bo);
 
 	list_for_each_entry_safe(bo, save, &exec->unref_list, unref_head) {
-		drm_gem_object_put_unlocked(&bo->base);
+		drm_gem_object_put_unlocked(&bo->base.base);
 	}
 
 	pm_runtime_mark_last_busy(v3d->dev);
@@ -405,11 +374,11 @@ v3d_tfu_job_cleanup(struct kref *ref)
 	unsigned int i;
 
 	dma_fence_put(job->in_fence);
-	dma_fence_put(job->done_fence);
+	dma_fence_put(job->irq_fence);
 
 	for (i = 0; i < ARRAY_SIZE(job->bo); i++) {
 		if (job->bo[i])
-			drm_gem_object_put_unlocked(&job->bo[i]->base);
+			drm_gem_object_put_unlocked(&job->bo[i]->base.base);
 	}
 
 	pm_runtime_mark_last_busy(v3d->dev);
@@ -429,8 +398,6 @@ v3d_wait_bo_ioctl(struct drm_device *dev, void *data,
 {
 	int ret;
 	struct drm_v3d_wait_bo *args = data;
-	struct drm_gem_object *gem_obj;
-	struct v3d_bo *bo;
 	ktime_t start = ktime_get();
 	u64 delta_ns;
 	unsigned long timeout_jiffies =
@@ -439,21 +406,8 @@ v3d_wait_bo_ioctl(struct drm_device *dev, void *data,
 	if (args->pad != 0)
 		return -EINVAL;
 
-	gem_obj = drm_gem_object_lookup(file_priv, args->handle);
-	if (!gem_obj) {
-		DRM_DEBUG("Failed to look up GEM BO %d\n", args->handle);
-		return -EINVAL;
-	}
-	bo = to_v3d_bo(gem_obj);
-
-	ret = reservation_object_wait_timeout_rcu(bo->resv,
-						  true, true,
-						  timeout_jiffies);
-
-	if (ret == 0)
-		ret = -ETIME;
-	else if (ret > 0)
-		ret = 0;
+	ret = drm_gem_reservation_object_wait(file_priv, args->handle,
+					      true, timeout_jiffies);
 
 	/* Decrement the user's timeout, in case we got interrupted
 	 * such that the ioctl will be restarted.
@@ -468,8 +422,6 @@ v3d_wait_bo_ioctl(struct drm_device *dev, void *data,
 	if (ret == -ETIME && args->timeout_ns)
 		ret = -EAGAIN;
 
-	drm_gem_object_put_unlocked(gem_obj);
-
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/v3d/v3d_irq.c b/drivers/gpu/drm/v3d/v3d_irq.c
index 69338da70ddc..aa0a180ae700 100644
--- a/drivers/gpu/drm/v3d/v3d_irq.c
+++ b/drivers/gpu/drm/v3d/v3d_irq.c
@@ -27,6 +27,9 @@
 			    V3D_HUB_INT_MMU_CAP |	\
 			    V3D_HUB_INT_TFUC))
 
+static irqreturn_t
+v3d_hub_irq(int irq, void *arg);
+
 static void
 v3d_overflow_mem_work(struct work_struct *work)
 {
@@ -34,12 +37,14 @@ v3d_overflow_mem_work(struct work_struct *work)
 		container_of(work, struct v3d_dev, overflow_mem_work);
 	struct drm_device *dev = &v3d->drm;
 	struct v3d_bo *bo = v3d_bo_create(dev, NULL /* XXX: GMP */, 256 * 1024);
+	struct drm_gem_object *obj;
 	unsigned long irqflags;
 
 	if (IS_ERR(bo)) {
 		DRM_ERROR("Couldn't allocate binner overflow mem\n");
 		return;
 	}
+	obj = &bo->base.base;
 
 	/* We lost a race, and our work task came in after the bin job
 	 * completed and exited.  This can happen because the HW
@@ -56,15 +61,15 @@ v3d_overflow_mem_work(struct work_struct *work)
 		goto out;
 	}
 
-	drm_gem_object_get(&bo->base);
+	drm_gem_object_get(obj);
 	list_add_tail(&bo->unref_head, &v3d->bin_job->unref_list);
 	spin_unlock_irqrestore(&v3d->job_lock, irqflags);
 
 	V3D_CORE_WRITE(0, V3D_PTB_BPOA, bo->node.start << PAGE_SHIFT);
-	V3D_CORE_WRITE(0, V3D_PTB_BPOS, bo->base.size);
+	V3D_CORE_WRITE(0, V3D_PTB_BPOS, obj->size);
 
 out:
-	drm_gem_object_put_unlocked(&bo->base);
+	drm_gem_object_put_unlocked(obj);
 }
 
 static irqreturn_t
@@ -82,7 +87,8 @@ v3d_irq(int irq, void *arg)
 	if (intsts & V3D_INT_OUTOMEM) {
 		/* Note that the OOM status is edge signaled, so the
 		 * interrupt won't happen again until the we actually
-		 * add more memory.
+		 * add more memory.  Also, as of V3D 4.1, FLDONE won't
+		 * be reported until any OOM state has been cleared.
 		 */
 		schedule_work(&v3d->overflow_mem_work);
 		status = IRQ_HANDLED;
@@ -90,7 +96,7 @@ v3d_irq(int irq, void *arg)
 
 	if (intsts & V3D_INT_FLDONE) {
 		struct v3d_fence *fence =
-			to_v3d_fence(v3d->bin_job->bin.done_fence);
+			to_v3d_fence(v3d->bin_job->bin.irq_fence);
 
 		trace_v3d_bcl_irq(&v3d->drm, fence->seqno);
 		dma_fence_signal(&fence->base);
@@ -99,7 +105,7 @@ v3d_irq(int irq, void *arg)
 
 	if (intsts & V3D_INT_FRDONE) {
 		struct v3d_fence *fence =
-			to_v3d_fence(v3d->render_job->render.done_fence);
+			to_v3d_fence(v3d->render_job->render.irq_fence);
 
 		trace_v3d_rcl_irq(&v3d->drm, fence->seqno);
 		dma_fence_signal(&fence->base);
@@ -112,6 +118,12 @@ v3d_irq(int irq, void *arg)
 	if (intsts & V3D_INT_GMPV)
 		dev_err(v3d->dev, "GMP violation\n");
 
+	/* V3D 4.2 wires the hub and core IRQs together, so if we &
+	 * didn't see the common one then check hub for MMU IRQs.
+	 */
+	if (v3d->single_irq_line && status == IRQ_NONE)
+		return v3d_hub_irq(irq, arg);
+
 	return status;
 }
 
@@ -129,7 +141,7 @@ v3d_hub_irq(int irq, void *arg)
 
 	if (intsts & V3D_HUB_INT_TFUC) {
 		struct v3d_fence *fence =
-			to_v3d_fence(v3d->tfu_job->done_fence);
+			to_v3d_fence(v3d->tfu_job->irq_fence);
 
 		trace_v3d_tfu_irq(&v3d->drm, fence->seqno);
 		dma_fence_signal(&fence->base);
@@ -156,10 +168,10 @@ v3d_hub_irq(int irq, void *arg)
 	return status;
 }
 
-void
+int
 v3d_irq_init(struct v3d_dev *v3d)
 {
-	int ret, core;
+	int irq1, ret, core;
 
 	INIT_WORK(&v3d->overflow_mem_work, v3d_overflow_mem_work);
 
@@ -170,16 +182,37 @@ v3d_irq_init(struct v3d_dev *v3d)
 		V3D_CORE_WRITE(core, V3D_CTL_INT_CLR, V3D_CORE_IRQS);
 	V3D_WRITE(V3D_HUB_INT_CLR, V3D_HUB_IRQS);
 
-	ret = devm_request_irq(v3d->dev, platform_get_irq(v3d->pdev, 0),
-			       v3d_hub_irq, IRQF_SHARED,
-			       "v3d_hub", v3d);
-	ret = devm_request_irq(v3d->dev, platform_get_irq(v3d->pdev, 1),
-			       v3d_irq, IRQF_SHARED,
-			       "v3d_core0", v3d);
-	if (ret)
-		dev_err(v3d->dev, "IRQ setup failed: %d\n", ret);
+	irq1 = platform_get_irq(v3d->pdev, 1);
+	if (irq1 == -EPROBE_DEFER)
+		return irq1;
+	if (irq1 > 0) {
+		ret = devm_request_irq(v3d->dev, irq1,
+				       v3d_irq, IRQF_SHARED,
+				       "v3d_core0", v3d);
+		if (ret)
+			goto fail;
+		ret = devm_request_irq(v3d->dev, platform_get_irq(v3d->pdev, 0),
+				       v3d_hub_irq, IRQF_SHARED,
+				       "v3d_hub", v3d);
+		if (ret)
+			goto fail;
+	} else {
+		v3d->single_irq_line = true;
+
+		ret = devm_request_irq(v3d->dev, platform_get_irq(v3d->pdev, 0),
+				       v3d_irq, IRQF_SHARED,
+				       "v3d", v3d);
+		if (ret)
+			goto fail;
+	}
 
 	v3d_irq_enable(v3d);
+	return 0;
+
+fail:
+	if (ret != -EPROBE_DEFER)
+		dev_err(v3d->dev, "IRQ setup failed: %d\n", ret);
+	return ret;
 }
 
 void
diff --git a/drivers/gpu/drm/v3d/v3d_mmu.c b/drivers/gpu/drm/v3d/v3d_mmu.c
index b00f97c31b70..7a21f1787ab1 100644
--- a/drivers/gpu/drm/v3d/v3d_mmu.c
+++ b/drivers/gpu/drm/v3d/v3d_mmu.c
@@ -83,13 +83,14 @@ int v3d_mmu_set_page_table(struct v3d_dev *v3d)
 
 void v3d_mmu_insert_ptes(struct v3d_bo *bo)
 {
-	struct v3d_dev *v3d = to_v3d_dev(bo->base.dev);
+	struct drm_gem_shmem_object *shmem_obj = &bo->base;
+	struct v3d_dev *v3d = to_v3d_dev(shmem_obj->base.dev);
 	u32 page = bo->node.start;
 	u32 page_prot = V3D_PTE_WRITEABLE | V3D_PTE_VALID;
 	unsigned int count;
 	struct scatterlist *sgl;
 
-	for_each_sg(bo->sgt->sgl, sgl, bo->sgt->nents, count) {
+	for_each_sg(shmem_obj->sgt->sgl, sgl, shmem_obj->sgt->nents, count) {
 		u32 page_address = sg_dma_address(sgl) >> V3D_MMU_PAGE_SHIFT;
 		u32 pte = page_prot | page_address;
 		u32 i;
@@ -102,7 +103,7 @@ void v3d_mmu_insert_ptes(struct v3d_bo *bo)
 	}
 
 	WARN_ON_ONCE(page - bo->node.start !=
-		     bo->base.size >> V3D_MMU_PAGE_SHIFT);
+		     shmem_obj->base.size >> V3D_MMU_PAGE_SHIFT);
 
 	if (v3d_mmu_flush_all(v3d))
 		dev_err(v3d->dev, "MMU flush timeout\n");
@@ -110,8 +111,8 @@ void v3d_mmu_insert_ptes(struct v3d_bo *bo)
 
 void v3d_mmu_remove_ptes(struct v3d_bo *bo)
 {
-	struct v3d_dev *v3d = to_v3d_dev(bo->base.dev);
-	u32 npages = bo->base.size >> V3D_MMU_PAGE_SHIFT;
+	struct v3d_dev *v3d = to_v3d_dev(bo->base.base.dev);
+	u32 npages = bo->base.base.size >> V3D_MMU_PAGE_SHIFT;
 	u32 page;
 
 	for (page = bo->node.start; page < bo->node.start + npages; page++)
diff --git a/drivers/gpu/drm/v3d/v3d_regs.h b/drivers/gpu/drm/v3d/v3d_regs.h
index 6ccdee9d47bd..8e88af237610 100644
--- a/drivers/gpu/drm/v3d/v3d_regs.h
+++ b/drivers/gpu/drm/v3d/v3d_regs.h
@@ -216,6 +216,8 @@
 # define V3D_IDENT2_BCG_INT                            BIT(28)
 
 #define V3D_CTL_MISCCFG                                0x00018
+# define V3D_CTL_MISCCFG_QRMAXCNT_MASK                 V3D_MASK(3, 1)
+# define V3D_CTL_MISCCFG_QRMAXCNT_SHIFT                1
 # define V3D_MISCCFG_OVRTMUOUT                         BIT(0)
 
 #define V3D_CTL_L2CACTL                                0x00020
diff --git a/drivers/gpu/drm/v3d/v3d_sched.c b/drivers/gpu/drm/v3d/v3d_sched.c
index 4704b2df3688..e740f3b99aa5 100644
--- a/drivers/gpu/drm/v3d/v3d_sched.c
+++ b/drivers/gpu/drm/v3d/v3d_sched.c
@@ -156,9 +156,9 @@ static struct dma_fence *v3d_job_run(struct drm_sched_job *sched_job)
 	if (IS_ERR(fence))
 		return NULL;
 
-	if (job->done_fence)
-		dma_fence_put(job->done_fence);
-	job->done_fence = dma_fence_get(fence);
+	if (job->irq_fence)
+		dma_fence_put(job->irq_fence);
+	job->irq_fence = dma_fence_get(fence);
 
 	trace_v3d_submit_cl(dev, q == V3D_RENDER, to_v3d_fence(fence)->seqno,
 			    job->start, job->end);
@@ -199,9 +199,9 @@ v3d_tfu_job_run(struct drm_sched_job *sched_job)
 		return NULL;
 
 	v3d->tfu_job = job;
-	if (job->done_fence)
-		dma_fence_put(job->done_fence);
-	job->done_fence = dma_fence_get(fence);
+	if (job->irq_fence)
+		dma_fence_put(job->irq_fence);
+	job->irq_fence = dma_fence_get(fence);
 
 	trace_v3d_submit_tfu(dev, to_v3d_fence(fence)->seqno);
 
@@ -231,20 +231,17 @@ v3d_gpu_reset_for_timeout(struct v3d_dev *v3d, struct drm_sched_job *sched_job)
 	mutex_lock(&v3d->reset_lock);
 
 	/* block scheduler */
-	for (q = 0; q < V3D_MAX_QUEUES; q++) {
-		struct drm_gpu_scheduler *sched = &v3d->queue[q].sched;
-
-		drm_sched_stop(sched);
+	for (q = 0; q < V3D_MAX_QUEUES; q++)
+		drm_sched_stop(&v3d->queue[q].sched);
 
-		if(sched_job)
-			drm_sched_increase_karma(sched_job);
-	}
+	if (sched_job)
+		drm_sched_increase_karma(sched_job);
 
 	/* get the GPU back into the init state */
 	v3d_reset(v3d);
 
 	for (q = 0; q < V3D_MAX_QUEUES; q++)
-		drm_sched_resubmit_jobs(sched_job->sched);
+		drm_sched_resubmit_jobs(&v3d->queue[q].sched);
 
 	/* Unblock schedulers and restart their jobs. */
 	for (q = 0; q < V3D_MAX_QUEUES; q++) {
diff --git a/drivers/staging/vboxvideo/Kconfig b/drivers/gpu/drm/vboxvideo/Kconfig
index d6ab955c0768..d6ab955c0768 100644
--- a/drivers/staging/vboxvideo/Kconfig
+++ b/drivers/gpu/drm/vboxvideo/Kconfig
diff --git a/drivers/staging/vboxvideo/Makefile b/drivers/gpu/drm/vboxvideo/Makefile
index 1224f313af0c..1224f313af0c 100644
--- a/drivers/staging/vboxvideo/Makefile
+++ b/drivers/gpu/drm/vboxvideo/Makefile
diff --git a/drivers/staging/vboxvideo/hgsmi_base.c b/drivers/gpu/drm/vboxvideo/hgsmi_base.c
index 361d3193258e..361d3193258e 100644
--- a/drivers/staging/vboxvideo/hgsmi_base.c
+++ b/drivers/gpu/drm/vboxvideo/hgsmi_base.c
diff --git a/drivers/staging/vboxvideo/hgsmi_ch_setup.h b/drivers/gpu/drm/vboxvideo/hgsmi_ch_setup.h
index 4e93418d6a13..4e93418d6a13 100644
--- a/drivers/staging/vboxvideo/hgsmi_ch_setup.h
+++ b/drivers/gpu/drm/vboxvideo/hgsmi_ch_setup.h
diff --git a/drivers/staging/vboxvideo/hgsmi_channels.h b/drivers/gpu/drm/vboxvideo/hgsmi_channels.h
index 9b83f4ff3faf..9b83f4ff3faf 100644
--- a/drivers/staging/vboxvideo/hgsmi_channels.h
+++ b/drivers/gpu/drm/vboxvideo/hgsmi_channels.h
diff --git a/drivers/staging/vboxvideo/hgsmi_defs.h b/drivers/gpu/drm/vboxvideo/hgsmi_defs.h
index 6c8df1cdb087..6c8df1cdb087 100644
--- a/drivers/staging/vboxvideo/hgsmi_defs.h
+++ b/drivers/gpu/drm/vboxvideo/hgsmi_defs.h
diff --git a/drivers/staging/vboxvideo/modesetting.c b/drivers/gpu/drm/vboxvideo/modesetting.c
index 7580b9002379..7580b9002379 100644
--- a/drivers/staging/vboxvideo/modesetting.c
+++ b/drivers/gpu/drm/vboxvideo/modesetting.c
diff --git a/drivers/staging/vboxvideo/vbox_drv.c b/drivers/gpu/drm/vboxvideo/vbox_drv.c
index e7755a179850..fb6a0f0b8167 100644
--- a/drivers/staging/vboxvideo/vbox_drv.c
+++ b/drivers/gpu/drm/vboxvideo/vbox_drv.c
@@ -200,36 +200,11 @@ static const struct file_operations vbox_fops = {
 	.read = drm_read,
 };
 
-static int vbox_master_set(struct drm_device *dev,
-			   struct drm_file *file_priv, bool from_open)
-{
-	struct vbox_private *vbox = dev->dev_private;
-
-	/*
-	 * We do not yet know whether the new owner can handle hotplug, so we
-	 * do not advertise dynamic modes on the first query and send a
-	 * tentative hotplug notification after that to see if they query again.
-	 */
-	vbox->initial_mode_queried = false;
-
-	return 0;
-}
-
-static void vbox_master_drop(struct drm_device *dev, struct drm_file *file_priv)
-{
-	struct vbox_private *vbox = dev->dev_private;
-
-	/* See vbox_master_set() */
-	vbox->initial_mode_queried = false;
-}
-
 static struct drm_driver driver = {
 	.driver_features =
 	    DRIVER_MODESET | DRIVER_GEM | DRIVER_PRIME | DRIVER_ATOMIC,
 
 	.lastclose = drm_fb_helper_lastclose,
-	.master_set = vbox_master_set,
-	.master_drop = vbox_master_drop,
 
 	.fops = &vbox_fops,
 	.irq_handler = vbox_irq_handler,
diff --git a/drivers/staging/vboxvideo/vbox_drv.h b/drivers/gpu/drm/vboxvideo/vbox_drv.h
index aa40e5cc2861..ece31f395540 100644
--- a/drivers/staging/vboxvideo/vbox_drv.h
+++ b/drivers/gpu/drm/vboxvideo/vbox_drv.h
@@ -14,7 +14,6 @@
 #include <linux/io.h>
 #include <linux/irqreturn.h>
 #include <linux/string.h>
-#include <linux/version.h>
 
 #include <drm/drm_encoder.h>
 #include <drm/drm_fb_helper.h>
@@ -83,12 +82,6 @@ struct vbox_private {
 	} ttm;
 
 	struct mutex hw_mutex; /* protects modeset and accel/vbva accesses */
-	/*
-	 * We decide whether or not user-space supports display hot-plug
-	 * depending on whether they react to a hot-plug event after the initial
-	 * mode query.
-	 */
-	bool initial_mode_queried;
 	struct work_struct hotplug_work;
 	u32 input_mapping_width;
 	u32 input_mapping_height;
@@ -209,8 +202,6 @@ int vbox_dumb_mmap_offset(struct drm_file *file,
 			  struct drm_device *dev,
 			  u32 handle, u64 *offset);
 
-#define DRM_FILE_PAGE_OFFSET (0x10000000ULL >> PAGE_SHIFT)
-
 int vbox_mm_init(struct vbox_private *vbox);
 void vbox_mm_fini(struct vbox_private *vbox);
 
diff --git a/drivers/staging/vboxvideo/vbox_fb.c b/drivers/gpu/drm/vboxvideo/vbox_fb.c
index 83a04afd1766..b724fe7c0c30 100644
--- a/drivers/staging/vboxvideo/vbox_fb.c
+++ b/drivers/gpu/drm/vboxvideo/vbox_fb.c
@@ -90,13 +90,9 @@ int vboxfb_create(struct drm_fb_helper *helper,
 	if (IS_ERR(info->screen_base))
 		return PTR_ERR(info->screen_base);
 
-	info->par = helper;
-
 	fb = &vbox->afb.base;
 	helper->fb = fb;
 
-	strcpy(info->fix.id, "vboxdrmfb");
-
 	info->fbops = &vboxfb_ops;
 
 	/*
@@ -106,9 +102,7 @@ int vboxfb_create(struct drm_fb_helper *helper,
 	info->apertures->ranges[0].base = pci_resource_start(pdev, 0);
 	info->apertures->ranges[0].size = pci_resource_len(pdev, 0);
 
-	drm_fb_helper_fill_fix(info, fb->pitches[0], fb->format->depth);
-	drm_fb_helper_fill_var(info, helper, sizes->fb_width,
-			       sizes->fb_height);
+	drm_fb_helper_fill_info(info, helper, sizes);
 
 	gpu_addr = vbox_bo_gpu_offset(bo);
 	info->fix.smem_start = info->apertures->ranges[0].base + gpu_addr;
diff --git a/drivers/staging/vboxvideo/vbox_hgsmi.c b/drivers/gpu/drm/vboxvideo/vbox_hgsmi.c
index 94b60654a012..94b60654a012 100644
--- a/drivers/staging/vboxvideo/vbox_hgsmi.c
+++ b/drivers/gpu/drm/vboxvideo/vbox_hgsmi.c
diff --git a/drivers/staging/vboxvideo/vbox_irq.c b/drivers/gpu/drm/vboxvideo/vbox_irq.c
index 195484713365..16a1e29f5292 100644
--- a/drivers/staging/vboxvideo/vbox_irq.c
+++ b/drivers/gpu/drm/vboxvideo/vbox_irq.c
@@ -105,6 +105,7 @@ static void validate_or_set_position_hints(struct vbox_private *vbox)
 /* Query the host for the most recent video mode hints. */
 static void vbox_update_mode_hints(struct vbox_private *vbox)
 {
+	struct drm_connector_list_iter conn_iter;
 	struct drm_device *dev = &vbox->ddev;
 	struct drm_connector *connector;
 	struct vbox_connector *vbox_conn;
@@ -122,8 +123,10 @@ static void vbox_update_mode_hints(struct vbox_private *vbox)
 	}
 
 	validate_or_set_position_hints(vbox);
-	drm_modeset_lock_all(dev);
-	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+
+	drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
+	drm_connector_list_iter_begin(dev, &conn_iter);
+	drm_for_each_connector_iter(connector, &conn_iter) {
 		vbox_conn = to_vbox_connector(connector);
 
 		hints = &vbox->last_mode_hints[vbox_conn->vbox_crtc->crtc_id];
@@ -152,7 +155,8 @@ static void vbox_update_mode_hints(struct vbox_private *vbox)
 
 		vbox_conn->vbox_crtc->disconnected = disconnected;
 	}
-	drm_modeset_unlock_all(dev);
+	drm_connector_list_iter_end(&conn_iter);
+	drm_modeset_unlock(&dev->mode_config.connection_mutex);
 }
 
 static void vbox_hotplug_worker(struct work_struct *work)
diff --git a/drivers/staging/vboxvideo/vbox_main.c b/drivers/gpu/drm/vboxvideo/vbox_main.c
index e1fb70a42d32..f4d02de5518a 100644
--- a/drivers/staging/vboxvideo/vbox_main.c
+++ b/drivers/gpu/drm/vboxvideo/vbox_main.c
@@ -32,9 +32,9 @@ void vbox_report_caps(struct vbox_private *vbox)
 	u32 caps = VBVACAPS_DISABLE_CURSOR_INTEGRATION |
 		   VBVACAPS_IRQ | VBVACAPS_USE_VBVA_ONLY;
 
-	if (vbox->initial_mode_queried)
-		caps |= VBVACAPS_VIDEO_MODE_HINTS;
-
+	/* The host only accepts VIDEO_MODE_HINTS if it is send separately. */
+	hgsmi_send_caps_info(vbox->guest_pool, caps);
+	caps |= VBVACAPS_VIDEO_MODE_HINTS;
 	hgsmi_send_caps_info(vbox->guest_pool, caps);
 }
 
diff --git a/drivers/staging/vboxvideo/vbox_mode.c b/drivers/gpu/drm/vboxvideo/vbox_mode.c
index f105360a574e..58cea131470e 100644
--- a/drivers/staging/vboxvideo/vbox_mode.c
+++ b/drivers/gpu/drm/vboxvideo/vbox_mode.c
@@ -735,29 +735,12 @@ static int vbox_get_modes(struct drm_connector *connector)
 
 	vbox_connector = to_vbox_connector(connector);
 	vbox = connector->dev->dev_private;
-	/*
-	 * Heuristic: we do not want to tell the host that we support dynamic
-	 * resizing unless we feel confident that the user space client using
-	 * the video driver can handle hot-plug events.  So the first time modes
-	 * are queried after a "master" switch we tell the host that we do not,
-	 * and immediately after we send the client a hot-plug notification as
-	 * a test to see if they will respond and query again.
-	 * That is also the reason why capabilities are reported to the host at
-	 * this place in the code rather than elsewhere.
-	 * We need to report the flags location before reporting the IRQ
-	 * capability.
-	 */
+
 	hgsmi_report_flags_location(vbox->guest_pool, GUEST_HEAP_OFFSET(vbox) +
 				    HOST_FLAGS_OFFSET);
 	if (vbox_connector->vbox_crtc->crtc_id == 0)
 		vbox_report_caps(vbox);
-	if (!vbox->initial_mode_queried) {
-		if (vbox_connector->vbox_crtc->crtc_id == 0) {
-			vbox->initial_mode_queried = true;
-			vbox_report_hotplug(vbox);
-		}
-		return drm_add_modes_noedid(connector, 800, 600);
-	}
+
 	num_modes = drm_add_modes_noedid(connector, 2560, 1600);
 	preferred_width = vbox_connector->mode_hint.width ?
 			  vbox_connector->mode_hint.width : 1024;
diff --git a/drivers/staging/vboxvideo/vbox_prime.c b/drivers/gpu/drm/vboxvideo/vbox_prime.c
index 702b1aa53494..702b1aa53494 100644
--- a/drivers/staging/vboxvideo/vbox_prime.c
+++ b/drivers/gpu/drm/vboxvideo/vbox_prime.c
diff --git a/drivers/staging/vboxvideo/vbox_ttm.c b/drivers/gpu/drm/vboxvideo/vbox_ttm.c
index 30f270027acf..9d78438c2877 100644
--- a/drivers/staging/vboxvideo/vbox_ttm.c
+++ b/drivers/gpu/drm/vboxvideo/vbox_ttm.c
@@ -156,7 +156,7 @@ int vbox_mm_init(struct vbox_private *vbox)
 	ret = ttm_bo_device_init(&vbox->ttm.bdev,
 				 &vbox_bo_driver,
 				 dev->anon_inode->i_mapping,
-				 DRM_FILE_PAGE_OFFSET, true);
+				 true);
 	if (ret) {
 		DRM_ERROR("Error initialising bo driver; %d\n", ret);
 		return ret;
@@ -357,14 +357,8 @@ int vbox_bo_push_sysram(struct vbox_bo *bo)
 
 int vbox_mmap(struct file *filp, struct vm_area_struct *vma)
 {
-	struct drm_file *file_priv;
-	struct vbox_private *vbox;
-
-	if (unlikely(vma->vm_pgoff < DRM_FILE_PAGE_OFFSET))
-		return -EINVAL;
-
-	file_priv = filp->private_data;
-	vbox = file_priv->minor->dev->dev_private;
+	struct drm_file *file_priv = filp->private_data;
+	struct vbox_private *vbox = file_priv->minor->dev->dev_private;
 
 	return ttm_bo_mmap(filp, vma, &vbox->ttm.bdev);
 }
diff --git a/drivers/staging/vboxvideo/vboxvideo.h b/drivers/gpu/drm/vboxvideo/vboxvideo.h
index 0592004f71aa..0592004f71aa 100644
--- a/drivers/staging/vboxvideo/vboxvideo.h
+++ b/drivers/gpu/drm/vboxvideo/vboxvideo.h
diff --git a/drivers/staging/vboxvideo/vboxvideo_guest.h b/drivers/gpu/drm/vboxvideo/vboxvideo_guest.h
index 55fcee3a6470..55fcee3a6470 100644
--- a/drivers/staging/vboxvideo/vboxvideo_guest.h
+++ b/drivers/gpu/drm/vboxvideo/vboxvideo_guest.h
diff --git a/drivers/staging/vboxvideo/vboxvideo_vbe.h b/drivers/gpu/drm/vboxvideo/vboxvideo_vbe.h
index 427235869297..427235869297 100644
--- a/drivers/staging/vboxvideo/vboxvideo_vbe.h
+++ b/drivers/gpu/drm/vboxvideo/vboxvideo_vbe.h
diff --git a/drivers/staging/vboxvideo/vbva_base.c b/drivers/gpu/drm/vboxvideo/vbva_base.c
index 36bc9824ec3f..36bc9824ec3f 100644
--- a/drivers/staging/vboxvideo/vbva_base.c
+++ b/drivers/gpu/drm/vboxvideo/vbva_base.c
diff --git a/drivers/gpu/drm/vc4/vc4_bo.c b/drivers/gpu/drm/vc4/vc4_bo.c
index 8dcce7182bb7..88ebd681d7eb 100644
--- a/drivers/gpu/drm/vc4/vc4_bo.c
+++ b/drivers/gpu/drm/vc4/vc4_bo.c
@@ -40,7 +40,7 @@ static bool is_user_label(int label)
 	return label >= VC4_BO_TYPE_COUNT;
 }
 
-static void vc4_bo_stats_dump(struct vc4_dev *vc4)
+static void vc4_bo_stats_print(struct drm_printer *p, struct vc4_dev *vc4)
 {
 	int i;
 
@@ -48,58 +48,35 @@ static void vc4_bo_stats_dump(struct vc4_dev *vc4)
 		if (!vc4->bo_labels[i].num_allocated)
 			continue;
 
-		DRM_INFO("%30s: %6dkb BOs (%d)\n",
-			 vc4->bo_labels[i].name,
-			 vc4->bo_labels[i].size_allocated / 1024,
-			 vc4->bo_labels[i].num_allocated);
+		drm_printf(p, "%30s: %6dkb BOs (%d)\n",
+			   vc4->bo_labels[i].name,
+			   vc4->bo_labels[i].size_allocated / 1024,
+			   vc4->bo_labels[i].num_allocated);
 	}
 
 	mutex_lock(&vc4->purgeable.lock);
 	if (vc4->purgeable.num)
-		DRM_INFO("%30s: %6zdkb BOs (%d)\n", "userspace BO cache",
-			 vc4->purgeable.size / 1024, vc4->purgeable.num);
+		drm_printf(p, "%30s: %6zdkb BOs (%d)\n", "userspace BO cache",
+			   vc4->purgeable.size / 1024, vc4->purgeable.num);
 
 	if (vc4->purgeable.purged_num)
-		DRM_INFO("%30s: %6zdkb BOs (%d)\n", "total purged BO",
-			 vc4->purgeable.purged_size / 1024,
-			 vc4->purgeable.purged_num);
+		drm_printf(p, "%30s: %6zdkb BOs (%d)\n", "total purged BO",
+			   vc4->purgeable.purged_size / 1024,
+			   vc4->purgeable.purged_num);
 	mutex_unlock(&vc4->purgeable.lock);
 }
 
-#ifdef CONFIG_DEBUG_FS
-int vc4_bo_stats_debugfs(struct seq_file *m, void *unused)
+static int vc4_bo_stats_debugfs(struct seq_file *m, void *unused)
 {
 	struct drm_info_node *node = (struct drm_info_node *)m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	int i;
-
-	mutex_lock(&vc4->bo_lock);
-	for (i = 0; i < vc4->num_labels; i++) {
-		if (!vc4->bo_labels[i].num_allocated)
-			continue;
-
-		seq_printf(m, "%30s: %6dkb BOs (%d)\n",
-			   vc4->bo_labels[i].name,
-			   vc4->bo_labels[i].size_allocated / 1024,
-			   vc4->bo_labels[i].num_allocated);
-	}
-	mutex_unlock(&vc4->bo_lock);
-
-	mutex_lock(&vc4->purgeable.lock);
-	if (vc4->purgeable.num)
-		seq_printf(m, "%30s: %6zdkb BOs (%d)\n", "userspace BO cache",
-			   vc4->purgeable.size / 1024, vc4->purgeable.num);
+	struct drm_printer p = drm_seq_file_printer(m);
 
-	if (vc4->purgeable.purged_num)
-		seq_printf(m, "%30s: %6zdkb BOs (%d)\n", "total purged BO",
-			   vc4->purgeable.purged_size / 1024,
-			   vc4->purgeable.purged_num);
-	mutex_unlock(&vc4->purgeable.lock);
+	vc4_bo_stats_print(&p, vc4);
 
 	return 0;
 }
-#endif
 
 /* Takes ownership of *name and returns the appropriate slot for it in
  * the bo_labels[] array, extending it as necessary.
@@ -201,8 +178,6 @@ static void vc4_bo_destroy(struct vc4_bo *bo)
 		bo->validated_shader = NULL;
 	}
 
-	reservation_object_fini(&bo->_resv);
-
 	drm_gem_cma_free_object(obj);
 }
 
@@ -427,8 +402,6 @@ struct drm_gem_object *vc4_create_object(struct drm_device *dev, size_t size)
 	vc4->bo_labels[VC4_BO_TYPE_KERNEL].num_allocated++;
 	vc4->bo_labels[VC4_BO_TYPE_KERNEL].size_allocated += size;
 	mutex_unlock(&vc4->bo_lock);
-	bo->resv = &bo->_resv;
-	reservation_object_init(bo->resv);
 
 	return &bo->base.base;
 }
@@ -479,8 +452,9 @@ struct vc4_bo *vc4_bo_create(struct drm_device *dev, size_t unaligned_size,
 	}
 
 	if (IS_ERR(cma_obj)) {
+		struct drm_printer p = drm_info_printer(vc4->dev->dev);
 		DRM_ERROR("Failed to allocate from CMA:\n");
-		vc4_bo_stats_dump(vc4);
+		vc4_bo_stats_print(&p, vc4);
 		return ERR_PTR(-ENOMEM);
 	}
 	bo = to_vc4_bo(&cma_obj->base);
@@ -684,13 +658,6 @@ static void vc4_bo_cache_time_timer(struct timer_list *t)
 	schedule_work(&vc4->bo_cache.time_work);
 }
 
-struct reservation_object *vc4_prime_res_obj(struct drm_gem_object *obj)
-{
-	struct vc4_bo *bo = to_vc4_bo(obj);
-
-	return bo->resv;
-}
-
 struct dma_buf *
 vc4_prime_export(struct drm_device *dev, struct drm_gem_object *obj, int flags)
 {
@@ -822,14 +789,12 @@ vc4_prime_import_sg_table(struct drm_device *dev,
 			  struct sg_table *sgt)
 {
 	struct drm_gem_object *obj;
-	struct vc4_bo *bo;
 
 	obj = drm_gem_cma_prime_import_sg_table(dev, attach, sgt);
 	if (IS_ERR(obj))
 		return obj;
 
-	bo = to_vc4_bo(obj);
-	bo->resv = attach->dmabuf->resv;
+	obj->resv = attach->dmabuf->resv;
 
 	return obj;
 }
@@ -1038,6 +1003,8 @@ int vc4_bo_cache_init(struct drm_device *dev)
 
 	mutex_init(&vc4->bo_lock);
 
+	vc4_debugfs_add_file(dev, "bo_stats", vc4_bo_stats_debugfs, NULL);
+
 	INIT_LIST_HEAD(&vc4->bo_cache.time_list);
 
 	INIT_WORK(&vc4->bo_cache.time_work, vc4_bo_cache_time_work);
diff --git a/drivers/gpu/drm/vc4/vc4_crtc.c b/drivers/gpu/drm/vc4/vc4_crtc.c
index 1baa10e94484..5e09389e1514 100644
--- a/drivers/gpu/drm/vc4/vc4_crtc.c
+++ b/drivers/gpu/drm/vc4/vc4_crtc.c
@@ -35,6 +35,7 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_atomic_uapi.h>
+#include <drm/drm_print.h>
 #include <drm/drm_probe_helper.h>
 #include <linux/clk.h>
 #include <drm/drm_fb_cma_helper.h>
@@ -67,67 +68,22 @@ to_vc4_crtc_state(struct drm_crtc_state *crtc_state)
 #define CRTC_WRITE(offset, val) writel(val, vc4_crtc->regs + (offset))
 #define CRTC_READ(offset) readl(vc4_crtc->regs + (offset))
 
-#define CRTC_REG(reg) { reg, #reg }
-static const struct {
-	u32 reg;
-	const char *name;
-} crtc_regs[] = {
-	CRTC_REG(PV_CONTROL),
-	CRTC_REG(PV_V_CONTROL),
-	CRTC_REG(PV_VSYNCD_EVEN),
-	CRTC_REG(PV_HORZA),
-	CRTC_REG(PV_HORZB),
-	CRTC_REG(PV_VERTA),
-	CRTC_REG(PV_VERTB),
-	CRTC_REG(PV_VERTA_EVEN),
-	CRTC_REG(PV_VERTB_EVEN),
-	CRTC_REG(PV_INTEN),
-	CRTC_REG(PV_INTSTAT),
-	CRTC_REG(PV_STAT),
-	CRTC_REG(PV_HACT_ACT),
+static const struct debugfs_reg32 crtc_regs[] = {
+	VC4_REG32(PV_CONTROL),
+	VC4_REG32(PV_V_CONTROL),
+	VC4_REG32(PV_VSYNCD_EVEN),
+	VC4_REG32(PV_HORZA),
+	VC4_REG32(PV_HORZB),
+	VC4_REG32(PV_VERTA),
+	VC4_REG32(PV_VERTB),
+	VC4_REG32(PV_VERTA_EVEN),
+	VC4_REG32(PV_VERTB_EVEN),
+	VC4_REG32(PV_INTEN),
+	VC4_REG32(PV_INTSTAT),
+	VC4_REG32(PV_STAT),
+	VC4_REG32(PV_HACT_ACT),
 };
 
-static void vc4_crtc_dump_regs(struct vc4_crtc *vc4_crtc)
-{
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(crtc_regs); i++) {
-		DRM_INFO("0x%04x (%s): 0x%08x\n",
-			 crtc_regs[i].reg, crtc_regs[i].name,
-			 CRTC_READ(crtc_regs[i].reg));
-	}
-}
-
-#ifdef CONFIG_DEBUG_FS
-int vc4_crtc_debugfs_regs(struct seq_file *m, void *unused)
-{
-	struct drm_info_node *node = (struct drm_info_node *)m->private;
-	struct drm_device *dev = node->minor->dev;
-	int crtc_index = (uintptr_t)node->info_ent->data;
-	struct drm_crtc *crtc;
-	struct vc4_crtc *vc4_crtc;
-	int i;
-
-	i = 0;
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
-		if (i == crtc_index)
-			break;
-		i++;
-	}
-	if (!crtc)
-		return 0;
-	vc4_crtc = to_vc4_crtc(crtc);
-
-	for (i = 0; i < ARRAY_SIZE(crtc_regs); i++) {
-		seq_printf(m, "%s (0x%04x): 0x%08x\n",
-			   crtc_regs[i].name, crtc_regs[i].reg,
-			   CRTC_READ(crtc_regs[i].reg));
-	}
-
-	return 0;
-}
-#endif
-
 bool vc4_crtc_get_scanoutpos(struct drm_device *dev, unsigned int crtc_id,
 			     bool in_vblank_irq, int *vpos, int *hpos,
 			     ktime_t *stime, ktime_t *etime,
@@ -434,8 +390,10 @@ static void vc4_crtc_mode_set_nofb(struct drm_crtc *crtc)
 	bool debug_dump_regs = false;
 
 	if (debug_dump_regs) {
-		DRM_INFO("CRTC %d regs before:\n", drm_crtc_index(crtc));
-		vc4_crtc_dump_regs(vc4_crtc);
+		struct drm_printer p = drm_info_printer(&vc4_crtc->pdev->dev);
+		dev_info(&vc4_crtc->pdev->dev, "CRTC %d regs before:\n",
+			 drm_crtc_index(crtc));
+		drm_print_regset32(&p, &vc4_crtc->regset);
 	}
 
 	if (vc4_crtc->channel == 2) {
@@ -476,8 +434,10 @@ static void vc4_crtc_mode_set_nofb(struct drm_crtc *crtc)
 	vc4_crtc_lut_load(crtc);
 
 	if (debug_dump_regs) {
-		DRM_INFO("CRTC %d regs after:\n", drm_crtc_index(crtc));
-		vc4_crtc_dump_regs(vc4_crtc);
+		struct drm_printer p = drm_info_printer(&vc4_crtc->pdev->dev);
+		dev_info(&vc4_crtc->pdev->dev, "CRTC %d regs after:\n",
+			 drm_crtc_index(crtc));
+		drm_print_regset32(&p, &vc4_crtc->regset);
 	}
 }
 
@@ -834,6 +794,14 @@ static void vc4_crtc_handle_page_flip(struct vc4_crtc *vc4_crtc)
 		drm_crtc_send_vblank_event(crtc, vc4_crtc->event);
 		vc4_crtc->event = NULL;
 		drm_crtc_vblank_put(crtc);
+
+		/* Wait for the page flip to unmask the underrun to ensure that
+		 * the display list was updated by the hardware. Before that
+		 * happens, the HVS will be using the previous display list with
+		 * the CRTC and encoder already reconfigured, leading to
+		 * underruns. This can be seen when reconfiguring the CRTC.
+		 */
+		vc4_hvs_unmask_underrun(dev, vc4_crtc->channel);
 	}
 	spin_unlock_irqrestore(&dev->event_lock, flags);
 }
@@ -1075,6 +1043,7 @@ static const struct drm_crtc_helper_funcs vc4_crtc_helper_funcs = {
 
 static const struct vc4_crtc_data pv0_data = {
 	.hvs_channel = 0,
+	.debugfs_name = "crtc0_regs",
 	.encoder_types = {
 		[PV_CONTROL_CLK_SELECT_DSI] = VC4_ENCODER_TYPE_DSI0,
 		[PV_CONTROL_CLK_SELECT_DPI_SMI_HDMI] = VC4_ENCODER_TYPE_DPI,
@@ -1083,6 +1052,7 @@ static const struct vc4_crtc_data pv0_data = {
 
 static const struct vc4_crtc_data pv1_data = {
 	.hvs_channel = 2,
+	.debugfs_name = "crtc1_regs",
 	.encoder_types = {
 		[PV_CONTROL_CLK_SELECT_DSI] = VC4_ENCODER_TYPE_DSI1,
 		[PV_CONTROL_CLK_SELECT_DPI_SMI_HDMI] = VC4_ENCODER_TYPE_SMI,
@@ -1091,6 +1061,7 @@ static const struct vc4_crtc_data pv1_data = {
 
 static const struct vc4_crtc_data pv2_data = {
 	.hvs_channel = 1,
+	.debugfs_name = "crtc2_regs",
 	.encoder_types = {
 		[PV_CONTROL_CLK_SELECT_DPI_SMI_HDMI] = VC4_ENCODER_TYPE_HDMI,
 		[PV_CONTROL_CLK_SELECT_VEC] = VC4_ENCODER_TYPE_VEC,
@@ -1169,11 +1140,16 @@ static int vc4_crtc_bind(struct device *dev, struct device *master, void *data)
 	if (!match)
 		return -ENODEV;
 	vc4_crtc->data = match->data;
+	vc4_crtc->pdev = pdev;
 
 	vc4_crtc->regs = vc4_ioremap_regs(pdev, 0);
 	if (IS_ERR(vc4_crtc->regs))
 		return PTR_ERR(vc4_crtc->regs);
 
+	vc4_crtc->regset.base = vc4_crtc->regs;
+	vc4_crtc->regset.regs = crtc_regs;
+	vc4_crtc->regset.nregs = ARRAY_SIZE(crtc_regs);
+
 	/* For now, we create just the primary and the legacy cursor
 	 * planes.  We should be able to stack more planes on easily,
 	 * but to do that we would need to compute the bandwidth
@@ -1247,6 +1223,9 @@ static int vc4_crtc_bind(struct device *dev, struct device *master, void *data)
 
 	platform_set_drvdata(pdev, vc4_crtc);
 
+	vc4_debugfs_add_regset32(drm, vc4_crtc->data->debugfs_name,
+				 &vc4_crtc->regset);
+
 	return 0;
 
 err_destroy_planes:
diff --git a/drivers/gpu/drm/vc4/vc4_debugfs.c b/drivers/gpu/drm/vc4/vc4_debugfs.c
index 7a0003de71ab..f9dec08267dc 100644
--- a/drivers/gpu/drm/vc4/vc4_debugfs.c
+++ b/drivers/gpu/drm/vc4/vc4_debugfs.c
@@ -15,26 +15,82 @@
 #include "vc4_drv.h"
 #include "vc4_regs.h"
 
-static const struct drm_info_list vc4_debugfs_list[] = {
-	{"bo_stats", vc4_bo_stats_debugfs, 0},
-	{"dpi_regs", vc4_dpi_debugfs_regs, 0},
-	{"dsi1_regs", vc4_dsi_debugfs_regs, 0, (void *)(uintptr_t)1},
-	{"hdmi_regs", vc4_hdmi_debugfs_regs, 0},
-	{"vec_regs", vc4_vec_debugfs_regs, 0},
-	{"txp_regs", vc4_txp_debugfs_regs, 0},
-	{"hvs_regs", vc4_hvs_debugfs_regs, 0},
-	{"crtc0_regs", vc4_crtc_debugfs_regs, 0, (void *)(uintptr_t)0},
-	{"crtc1_regs", vc4_crtc_debugfs_regs, 0, (void *)(uintptr_t)1},
-	{"crtc2_regs", vc4_crtc_debugfs_regs, 0, (void *)(uintptr_t)2},
-	{"v3d_ident", vc4_v3d_debugfs_ident, 0},
-	{"v3d_regs", vc4_v3d_debugfs_regs, 0},
+struct vc4_debugfs_info_entry {
+	struct list_head link;
+	struct drm_info_list info;
 };
 
-#define VC4_DEBUGFS_ENTRIES ARRAY_SIZE(vc4_debugfs_list)
-
+/**
+ * Called at drm_dev_register() time on each of the minors registered
+ * by the DRM device, to attach the debugfs files.
+ */
 int
 vc4_debugfs_init(struct drm_minor *minor)
 {
-	return drm_debugfs_create_files(vc4_debugfs_list, VC4_DEBUGFS_ENTRIES,
-					minor->debugfs_root, minor);
+	struct vc4_dev *vc4 = to_vc4_dev(minor->dev);
+	struct vc4_debugfs_info_entry *entry;
+	struct dentry *dentry;
+
+	dentry = debugfs_create_bool("hvs_load_tracker", S_IRUGO | S_IWUSR,
+				     minor->debugfs_root,
+				     &vc4->load_tracker_enabled);
+	if (!dentry)
+		return -ENOMEM;
+
+	list_for_each_entry(entry, &vc4->debugfs_list, link) {
+		int ret = drm_debugfs_create_files(&entry->info, 1,
+						   minor->debugfs_root, minor);
+
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int vc4_debugfs_regset32(struct seq_file *m, void *unused)
+{
+	struct drm_info_node *node = (struct drm_info_node *)m->private;
+	struct debugfs_regset32 *regset = node->info_ent->data;
+	struct drm_printer p = drm_seq_file_printer(m);
+
+	drm_print_regset32(&p, regset);
+
+	return 0;
+}
+
+/**
+ * Registers a debugfs file with a callback function for a vc4 component.
+ *
+ * This is like drm_debugfs_create_files(), but that can only be
+ * called a given DRM minor, while the various VC4 components want to
+ * register their debugfs files during the component bind process.  We
+ * track the request and delay it to be called on each minor during
+ * vc4_debugfs_init().
+ */
+void vc4_debugfs_add_file(struct drm_device *dev,
+			  const char *name,
+			  int (*show)(struct seq_file*, void*),
+			  void *data)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+
+	struct vc4_debugfs_info_entry *entry =
+		devm_kzalloc(dev->dev, sizeof(*entry), GFP_KERNEL);
+
+	if (!entry)
+		return;
+
+	entry->info.name = name;
+	entry->info.show = show;
+	entry->info.data = data;
+
+	list_add(&entry->link, &vc4->debugfs_list);
+}
+
+void vc4_debugfs_add_regset32(struct drm_device *drm,
+			      const char *name,
+			      struct debugfs_regset32 *regset)
+{
+	vc4_debugfs_add_file(drm, name, vc4_debugfs_regset32, regset);
 }
diff --git a/drivers/gpu/drm/vc4/vc4_dpi.c b/drivers/gpu/drm/vc4/vc4_dpi.c
index 169521e547ba..34f90ca8f479 100644
--- a/drivers/gpu/drm/vc4/vc4_dpi.c
+++ b/drivers/gpu/drm/vc4/vc4_dpi.c
@@ -101,6 +101,8 @@ struct vc4_dpi {
 
 	struct clk *pixel_clock;
 	struct clk *core_clock;
+
+	struct debugfs_regset32 regset;
 };
 
 #define DPI_READ(offset) readl(dpi->regs + (offset))
@@ -118,37 +120,11 @@ to_vc4_dpi_encoder(struct drm_encoder *encoder)
 	return container_of(encoder, struct vc4_dpi_encoder, base.base);
 }
 
-#define DPI_REG(reg) { reg, #reg }
-static const struct {
-	u32 reg;
-	const char *name;
-} dpi_regs[] = {
-	DPI_REG(DPI_C),
-	DPI_REG(DPI_ID),
+static const struct debugfs_reg32 dpi_regs[] = {
+	VC4_REG32(DPI_C),
+	VC4_REG32(DPI_ID),
 };
 
-#ifdef CONFIG_DEBUG_FS
-int vc4_dpi_debugfs_regs(struct seq_file *m, void *unused)
-{
-	struct drm_info_node *node = (struct drm_info_node *)m->private;
-	struct drm_device *dev = node->minor->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	struct vc4_dpi *dpi = vc4->dpi;
-	int i;
-
-	if (!dpi)
-		return 0;
-
-	for (i = 0; i < ARRAY_SIZE(dpi_regs); i++) {
-		seq_printf(m, "%s (0x%04x): 0x%08x\n",
-			   dpi_regs[i].name, dpi_regs[i].reg,
-			   DPI_READ(dpi_regs[i].reg));
-	}
-
-	return 0;
-}
-#endif
-
 static const struct drm_encoder_funcs vc4_dpi_encoder_funcs = {
 	.destroy = drm_encoder_cleanup,
 };
@@ -314,6 +290,9 @@ static int vc4_dpi_bind(struct device *dev, struct device *master, void *data)
 	dpi->regs = vc4_ioremap_regs(pdev, 0);
 	if (IS_ERR(dpi->regs))
 		return PTR_ERR(dpi->regs);
+	dpi->regset.base = dpi->regs;
+	dpi->regset.regs = dpi_regs;
+	dpi->regset.nregs = ARRAY_SIZE(dpi_regs);
 
 	if (DPI_READ(DPI_ID) != DPI_ID_VALUE) {
 		dev_err(dev, "Port returned 0x%08x for ID instead of 0x%08x\n",
@@ -352,6 +331,8 @@ static int vc4_dpi_bind(struct device *dev, struct device *master, void *data)
 
 	vc4->dpi = dpi;
 
+	vc4_debugfs_add_regset32(drm, "dpi_regs", &dpi->regset);
+
 	return 0;
 
 err_destroy_encoder:
diff --git a/drivers/gpu/drm/vc4/vc4_drv.c b/drivers/gpu/drm/vc4/vc4_drv.c
index 5fcd2f0da7f7..6d9be20a32be 100644
--- a/drivers/gpu/drm/vc4/vc4_drv.c
+++ b/drivers/gpu/drm/vc4/vc4_drv.c
@@ -72,30 +72,30 @@ static int vc4_get_param_ioctl(struct drm_device *dev, void *data,
 	if (args->pad != 0)
 		return -EINVAL;
 
+	if (!vc4->v3d)
+		return -ENODEV;
+
 	switch (args->param) {
 	case DRM_VC4_PARAM_V3D_IDENT0:
-		ret = pm_runtime_get_sync(&vc4->v3d->pdev->dev);
-		if (ret < 0)
+		ret = vc4_v3d_pm_get(vc4);
+		if (ret)
 			return ret;
 		args->value = V3D_READ(V3D_IDENT0);
-		pm_runtime_mark_last_busy(&vc4->v3d->pdev->dev);
-		pm_runtime_put_autosuspend(&vc4->v3d->pdev->dev);
+		vc4_v3d_pm_put(vc4);
 		break;
 	case DRM_VC4_PARAM_V3D_IDENT1:
-		ret = pm_runtime_get_sync(&vc4->v3d->pdev->dev);
-		if (ret < 0)
+		ret = vc4_v3d_pm_get(vc4);
+		if (ret)
 			return ret;
 		args->value = V3D_READ(V3D_IDENT1);
-		pm_runtime_mark_last_busy(&vc4->v3d->pdev->dev);
-		pm_runtime_put_autosuspend(&vc4->v3d->pdev->dev);
+		vc4_v3d_pm_put(vc4);
 		break;
 	case DRM_VC4_PARAM_V3D_IDENT2:
-		ret = pm_runtime_get_sync(&vc4->v3d->pdev->dev);
-		if (ret < 0)
+		ret = vc4_v3d_pm_get(vc4);
+		if (ret)
 			return ret;
 		args->value = V3D_READ(V3D_IDENT2);
-		pm_runtime_mark_last_busy(&vc4->v3d->pdev->dev);
-		pm_runtime_put_autosuspend(&vc4->v3d->pdev->dev);
+		vc4_v3d_pm_put(vc4);
 		break;
 	case DRM_VC4_PARAM_SUPPORTS_BRANCHES:
 	case DRM_VC4_PARAM_SUPPORTS_ETC1:
@@ -200,7 +200,6 @@ static struct drm_driver vc4_drm_driver = {
 	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
 	.gem_prime_import = drm_gem_prime_import,
 	.gem_prime_export = vc4_prime_export,
-	.gem_prime_res_obj = vc4_prime_res_obj,
 	.gem_prime_get_sg_table	= drm_gem_cma_prime_get_sg_table,
 	.gem_prime_import_sg_table = vc4_prime_import_sg_table,
 	.gem_prime_vmap = vc4_prime_vmap,
@@ -252,6 +251,7 @@ static int vc4_drm_bind(struct device *dev)
 	struct platform_device *pdev = to_platform_device(dev);
 	struct drm_device *drm;
 	struct vc4_dev *vc4;
+	struct device_node *node;
 	int ret = 0;
 
 	dev->coherent_dma_mask = DMA_BIT_MASK(32);
@@ -260,12 +260,19 @@ static int vc4_drm_bind(struct device *dev)
 	if (!vc4)
 		return -ENOMEM;
 
+	/* If VC4 V3D is missing, don't advertise render nodes. */
+	node = of_find_matching_node_and_match(NULL, vc4_v3d_dt_match, NULL);
+	if (!node || !of_device_is_available(node))
+		vc4_drm_driver.driver_features &= ~DRIVER_RENDER;
+	of_node_put(node);
+
 	drm = drm_dev_alloc(&vc4_drm_driver, dev);
 	if (IS_ERR(drm))
 		return PTR_ERR(drm);
 	platform_set_drvdata(pdev, drm);
 	vc4->dev = drm;
 	drm->dev_private = vc4;
+	INIT_LIST_HEAD(&vc4->debugfs_list);
 
 	ret = vc4_bo_cache_init(drm);
 	if (ret)
@@ -281,13 +288,15 @@ static int vc4_drm_bind(struct device *dev)
 
 	drm_fb_helper_remove_conflicting_framebuffers(NULL, "vc4drmfb", false);
 
-	ret = drm_dev_register(drm, 0);
+	ret = vc4_kms_load(drm);
 	if (ret < 0)
 		goto unbind_all;
 
-	vc4_kms_load(drm);
+	ret = drm_dev_register(drm, 0);
+	if (ret < 0)
+		goto unbind_all;
 
-	drm_fbdev_generic_setup(drm, 32);
+	drm_fbdev_generic_setup(drm, 16);
 
 	return 0;
 
@@ -312,6 +321,7 @@ static void vc4_drm_unbind(struct device *dev)
 
 	drm_mode_config_cleanup(drm);
 
+	drm_atomic_private_obj_fini(&vc4->load_tracker);
 	drm_atomic_private_obj_fini(&vc4->ctm_manager);
 
 	drm_dev_put(drm);
diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h
index 2c635f001c71..4f13f6262491 100644
--- a/drivers/gpu/drm/vc4/vc4_drv.h
+++ b/drivers/gpu/drm/vc4/vc4_drv.h
@@ -7,7 +7,6 @@
  */
 
 #include <linux/mm_types.h>
-#include <linux/reservation.h>
 #include <drm/drmP.h>
 #include <drm/drm_util.h>
 #include <drm/drm_encoder.h>
@@ -185,10 +184,20 @@ struct vc4_dev {
 	/* Bitmask of the current bin_alloc used for overflow memory. */
 	uint32_t bin_alloc_overflow;
 
+	/* Incremented when an underrun error happened after an atomic commit.
+	 * This is particularly useful to detect when a specific modeset is too
+	 * demanding in term of memory or HVS bandwidth which is hard to guess
+	 * at atomic check time.
+	 */
+	atomic_t underrun;
+
 	struct work_struct overflow_mem_work;
 
 	int power_refcount;
 
+	/* Set to true when the load tracker is active. */
+	bool load_tracker_enabled;
+
 	/* Mutex controlling the power refcount. */
 	struct mutex power_lock;
 
@@ -201,6 +210,12 @@ struct vc4_dev {
 
 	struct drm_modeset_lock ctm_state_lock;
 	struct drm_private_obj ctm_manager;
+	struct drm_private_obj load_tracker;
+
+	/* List of vc4_debugfs_info_entry for adding to debugfs once
+	 * the minor is available (after drm_dev_register()).
+	 */
+	struct list_head debugfs_list;
 };
 
 static inline struct vc4_dev *
@@ -240,10 +255,6 @@ struct vc4_bo {
 	 */
 	struct vc4_validated_shader_info *validated_shader;
 
-	/* normally (resv == &_resv) except for imported bo's */
-	struct reservation_object *resv;
-	struct reservation_object _resv;
-
 	/* One of enum vc4_kernel_bo_type, or VC4_BO_TYPE_COUNT + i
 	 * for user-allocated labels.
 	 */
@@ -290,6 +301,7 @@ struct vc4_v3d {
 	struct platform_device *pdev;
 	void __iomem *regs;
 	struct clk *clk;
+	struct debugfs_regset32 regset;
 };
 
 struct vc4_hvs {
@@ -306,6 +318,7 @@ struct vc4_hvs {
 	spinlock_t mm_lock;
 
 	struct drm_mm_node mitchell_netravali_filter;
+	struct debugfs_regset32 regset;
 };
 
 struct vc4_plane {
@@ -376,6 +389,16 @@ struct vc4_plane_state {
 	 * when async update is not possible.
 	 */
 	bool dlist_initialized;
+
+	/* Load of this plane on the HVS block. The load is expressed in HVS
+	 * cycles/sec.
+	 */
+	u64 hvs_load;
+
+	/* Memory bandwidth needed for this plane. This is expressed in
+	 * bytes/sec.
+	 */
+	u64 membus_load;
 };
 
 static inline struct vc4_plane_state *
@@ -411,10 +434,12 @@ struct vc4_crtc_data {
 	int hvs_channel;
 
 	enum vc4_encoder_type encoder_types[4];
+	const char *debugfs_name;
 };
 
 struct vc4_crtc {
 	struct drm_crtc base;
+	struct platform_device *pdev;
 	const struct vc4_crtc_data *data;
 	void __iomem *regs;
 
@@ -431,6 +456,8 @@ struct vc4_crtc {
 	u32 cob_size;
 
 	struct drm_pending_vblank_event *event;
+
+	struct debugfs_regset32 regset;
 };
 
 static inline struct vc4_crtc *
@@ -444,6 +471,8 @@ to_vc4_crtc(struct drm_crtc *crtc)
 #define HVS_READ(offset) readl(vc4->hvs->regs + offset)
 #define HVS_WRITE(offset, val) writel(val, vc4->hvs->regs + offset)
 
+#define VC4_REG32(reg) { .name = #reg, .offset = reg }
+
 struct vc4_exec_info {
 	/* Sequence number for this bin/render job. */
 	uint64_t seqno;
@@ -685,7 +714,6 @@ int vc4_label_bo_ioctl(struct drm_device *dev, void *data,
 		       struct drm_file *file_priv);
 vm_fault_t vc4_fault(struct vm_fault *vmf);
 int vc4_mmap(struct file *filp, struct vm_area_struct *vma);
-struct reservation_object *vc4_prime_res_obj(struct drm_gem_object *obj);
 int vc4_prime_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma);
 struct drm_gem_object *vc4_prime_import_sg_table(struct drm_device *dev,
 						 struct dma_buf_attachment *attach,
@@ -693,7 +721,6 @@ struct drm_gem_object *vc4_prime_import_sg_table(struct drm_device *dev,
 void *vc4_prime_vmap(struct drm_gem_object *obj);
 int vc4_bo_cache_init(struct drm_device *dev);
 void vc4_bo_cache_destroy(struct drm_device *dev);
-int vc4_bo_stats_debugfs(struct seq_file *m, void *arg);
 int vc4_bo_inc_usecnt(struct vc4_bo *bo);
 void vc4_bo_dec_usecnt(struct vc4_bo *bo);
 void vc4_bo_add_to_purgeable_pool(struct vc4_bo *bo);
@@ -701,7 +728,6 @@ void vc4_bo_remove_from_purgeable_pool(struct vc4_bo *bo);
 
 /* vc4_crtc.c */
 extern struct platform_driver vc4_crtc_driver;
-int vc4_crtc_debugfs_regs(struct seq_file *m, void *arg);
 bool vc4_crtc_get_scanoutpos(struct drm_device *dev, unsigned int crtc_id,
 			     bool in_vblank_irq, int *vpos, int *hpos,
 			     ktime_t *stime, ktime_t *etime,
@@ -714,17 +740,37 @@ void vc4_crtc_get_margins(struct drm_crtc_state *state,
 
 /* vc4_debugfs.c */
 int vc4_debugfs_init(struct drm_minor *minor);
+#ifdef CONFIG_DEBUG_FS
+void vc4_debugfs_add_file(struct drm_device *drm,
+			  const char *filename,
+			  int (*show)(struct seq_file*, void*),
+			  void *data);
+void vc4_debugfs_add_regset32(struct drm_device *drm,
+			      const char *filename,
+			      struct debugfs_regset32 *regset);
+#else
+static inline void vc4_debugfs_add_file(struct drm_device *drm,
+					const char *filename,
+					int (*show)(struct seq_file*, void*),
+					void *data)
+{
+}
+
+static inline void vc4_debugfs_add_regset32(struct drm_device *drm,
+					    const char *filename,
+					    struct debugfs_regset32 *regset)
+{
+}
+#endif
 
 /* vc4_drv.c */
 void __iomem *vc4_ioremap_regs(struct platform_device *dev, int index);
 
 /* vc4_dpi.c */
 extern struct platform_driver vc4_dpi_driver;
-int vc4_dpi_debugfs_regs(struct seq_file *m, void *unused);
 
 /* vc4_dsi.c */
 extern struct platform_driver vc4_dsi_driver;
-int vc4_dsi_debugfs_regs(struct seq_file *m, void *unused);
 
 /* vc4_fence.c */
 extern const struct dma_fence_ops vc4_fence_ops;
@@ -752,15 +798,12 @@ int vc4_gem_madvise_ioctl(struct drm_device *dev, void *data,
 
 /* vc4_hdmi.c */
 extern struct platform_driver vc4_hdmi_driver;
-int vc4_hdmi_debugfs_regs(struct seq_file *m, void *unused);
 
 /* vc4_vec.c */
 extern struct platform_driver vc4_vec_driver;
-int vc4_vec_debugfs_regs(struct seq_file *m, void *unused);
 
 /* vc4_txp.c */
 extern struct platform_driver vc4_txp_driver;
-int vc4_txp_debugfs_regs(struct seq_file *m, void *unused);
 
 /* vc4_irq.c */
 irqreturn_t vc4_irq(int irq, void *arg);
@@ -772,7 +815,8 @@ void vc4_irq_reset(struct drm_device *dev);
 /* vc4_hvs.c */
 extern struct platform_driver vc4_hvs_driver;
 void vc4_hvs_dump_state(struct drm_device *dev);
-int vc4_hvs_debugfs_regs(struct seq_file *m, void *unused);
+void vc4_hvs_unmask_underrun(struct drm_device *dev, int channel);
+void vc4_hvs_mask_underrun(struct drm_device *dev, int channel);
 
 /* vc4_kms.c */
 int vc4_kms_load(struct drm_device *dev);
@@ -787,9 +831,10 @@ void vc4_plane_async_set_fb(struct drm_plane *plane,
 
 /* vc4_v3d.c */
 extern struct platform_driver vc4_v3d_driver;
-int vc4_v3d_debugfs_ident(struct seq_file *m, void *unused);
-int vc4_v3d_debugfs_regs(struct seq_file *m, void *unused);
+extern const struct of_device_id vc4_v3d_dt_match[];
 int vc4_v3d_get_bin_slot(struct vc4_dev *vc4);
+int vc4_v3d_pm_get(struct vc4_dev *vc4);
+void vc4_v3d_pm_put(struct vc4_dev *vc4);
 
 /* vc4_validate.c */
 int
diff --git a/drivers/gpu/drm/vc4/vc4_dsi.c b/drivers/gpu/drm/vc4/vc4_dsi.c
index 11702e1d9011..9412709067f5 100644
--- a/drivers/gpu/drm/vc4/vc4_dsi.c
+++ b/drivers/gpu/drm/vc4/vc4_dsi.c
@@ -545,6 +545,8 @@ struct vc4_dsi {
 
 	struct completion xfer_completion;
 	int xfer_result;
+
+	struct debugfs_regset32 regset;
 };
 
 #define host_to_dsi(host) container_of(host, struct vc4_dsi, dsi_host)
@@ -605,113 +607,56 @@ to_vc4_dsi_encoder(struct drm_encoder *encoder)
 	return container_of(encoder, struct vc4_dsi_encoder, base.base);
 }
 
-#define DSI_REG(reg) { reg, #reg }
-static const struct {
-	u32 reg;
-	const char *name;
-} dsi0_regs[] = {
-	DSI_REG(DSI0_CTRL),
-	DSI_REG(DSI0_STAT),
-	DSI_REG(DSI0_HSTX_TO_CNT),
-	DSI_REG(DSI0_LPRX_TO_CNT),
-	DSI_REG(DSI0_TA_TO_CNT),
-	DSI_REG(DSI0_PR_TO_CNT),
-	DSI_REG(DSI0_DISP0_CTRL),
-	DSI_REG(DSI0_DISP1_CTRL),
-	DSI_REG(DSI0_INT_STAT),
-	DSI_REG(DSI0_INT_EN),
-	DSI_REG(DSI0_PHYC),
-	DSI_REG(DSI0_HS_CLT0),
-	DSI_REG(DSI0_HS_CLT1),
-	DSI_REG(DSI0_HS_CLT2),
-	DSI_REG(DSI0_HS_DLT3),
-	DSI_REG(DSI0_HS_DLT4),
-	DSI_REG(DSI0_HS_DLT5),
-	DSI_REG(DSI0_HS_DLT6),
-	DSI_REG(DSI0_HS_DLT7),
-	DSI_REG(DSI0_PHY_AFEC0),
-	DSI_REG(DSI0_PHY_AFEC1),
-	DSI_REG(DSI0_ID),
+static const struct debugfs_reg32 dsi0_regs[] = {
+	VC4_REG32(DSI0_CTRL),
+	VC4_REG32(DSI0_STAT),
+	VC4_REG32(DSI0_HSTX_TO_CNT),
+	VC4_REG32(DSI0_LPRX_TO_CNT),
+	VC4_REG32(DSI0_TA_TO_CNT),
+	VC4_REG32(DSI0_PR_TO_CNT),
+	VC4_REG32(DSI0_DISP0_CTRL),
+	VC4_REG32(DSI0_DISP1_CTRL),
+	VC4_REG32(DSI0_INT_STAT),
+	VC4_REG32(DSI0_INT_EN),
+	VC4_REG32(DSI0_PHYC),
+	VC4_REG32(DSI0_HS_CLT0),
+	VC4_REG32(DSI0_HS_CLT1),
+	VC4_REG32(DSI0_HS_CLT2),
+	VC4_REG32(DSI0_HS_DLT3),
+	VC4_REG32(DSI0_HS_DLT4),
+	VC4_REG32(DSI0_HS_DLT5),
+	VC4_REG32(DSI0_HS_DLT6),
+	VC4_REG32(DSI0_HS_DLT7),
+	VC4_REG32(DSI0_PHY_AFEC0),
+	VC4_REG32(DSI0_PHY_AFEC1),
+	VC4_REG32(DSI0_ID),
 };
 
-static const struct {
-	u32 reg;
-	const char *name;
-} dsi1_regs[] = {
-	DSI_REG(DSI1_CTRL),
-	DSI_REG(DSI1_STAT),
-	DSI_REG(DSI1_HSTX_TO_CNT),
-	DSI_REG(DSI1_LPRX_TO_CNT),
-	DSI_REG(DSI1_TA_TO_CNT),
-	DSI_REG(DSI1_PR_TO_CNT),
-	DSI_REG(DSI1_DISP0_CTRL),
-	DSI_REG(DSI1_DISP1_CTRL),
-	DSI_REG(DSI1_INT_STAT),
-	DSI_REG(DSI1_INT_EN),
-	DSI_REG(DSI1_PHYC),
-	DSI_REG(DSI1_HS_CLT0),
-	DSI_REG(DSI1_HS_CLT1),
-	DSI_REG(DSI1_HS_CLT2),
-	DSI_REG(DSI1_HS_DLT3),
-	DSI_REG(DSI1_HS_DLT4),
-	DSI_REG(DSI1_HS_DLT5),
-	DSI_REG(DSI1_HS_DLT6),
-	DSI_REG(DSI1_HS_DLT7),
-	DSI_REG(DSI1_PHY_AFEC0),
-	DSI_REG(DSI1_PHY_AFEC1),
-	DSI_REG(DSI1_ID),
+static const struct debugfs_reg32 dsi1_regs[] = {
+	VC4_REG32(DSI1_CTRL),
+	VC4_REG32(DSI1_STAT),
+	VC4_REG32(DSI1_HSTX_TO_CNT),
+	VC4_REG32(DSI1_LPRX_TO_CNT),
+	VC4_REG32(DSI1_TA_TO_CNT),
+	VC4_REG32(DSI1_PR_TO_CNT),
+	VC4_REG32(DSI1_DISP0_CTRL),
+	VC4_REG32(DSI1_DISP1_CTRL),
+	VC4_REG32(DSI1_INT_STAT),
+	VC4_REG32(DSI1_INT_EN),
+	VC4_REG32(DSI1_PHYC),
+	VC4_REG32(DSI1_HS_CLT0),
+	VC4_REG32(DSI1_HS_CLT1),
+	VC4_REG32(DSI1_HS_CLT2),
+	VC4_REG32(DSI1_HS_DLT3),
+	VC4_REG32(DSI1_HS_DLT4),
+	VC4_REG32(DSI1_HS_DLT5),
+	VC4_REG32(DSI1_HS_DLT6),
+	VC4_REG32(DSI1_HS_DLT7),
+	VC4_REG32(DSI1_PHY_AFEC0),
+	VC4_REG32(DSI1_PHY_AFEC1),
+	VC4_REG32(DSI1_ID),
 };
 
-static void vc4_dsi_dump_regs(struct vc4_dsi *dsi)
-{
-	int i;
-
-	if (dsi->port == 0) {
-		for (i = 0; i < ARRAY_SIZE(dsi0_regs); i++) {
-			DRM_INFO("0x%04x (%s): 0x%08x\n",
-				 dsi0_regs[i].reg, dsi0_regs[i].name,
-				 DSI_READ(dsi0_regs[i].reg));
-		}
-	} else {
-		for (i = 0; i < ARRAY_SIZE(dsi1_regs); i++) {
-			DRM_INFO("0x%04x (%s): 0x%08x\n",
-				 dsi1_regs[i].reg, dsi1_regs[i].name,
-				 DSI_READ(dsi1_regs[i].reg));
-		}
-	}
-}
-
-#ifdef CONFIG_DEBUG_FS
-int vc4_dsi_debugfs_regs(struct seq_file *m, void *unused)
-{
-	struct drm_info_node *node = (struct drm_info_node *)m->private;
-	struct drm_device *drm = node->minor->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(drm);
-	int dsi_index = (uintptr_t)node->info_ent->data;
-	struct vc4_dsi *dsi = (dsi_index == 1 ? vc4->dsi1 : NULL);
-	int i;
-
-	if (!dsi)
-		return 0;
-
-	if (dsi->port == 0) {
-		for (i = 0; i < ARRAY_SIZE(dsi0_regs); i++) {
-			seq_printf(m, "0x%04x (%s): 0x%08x\n",
-				   dsi0_regs[i].reg, dsi0_regs[i].name,
-				   DSI_READ(dsi0_regs[i].reg));
-		}
-	} else {
-		for (i = 0; i < ARRAY_SIZE(dsi1_regs); i++) {
-			seq_printf(m, "0x%04x (%s): 0x%08x\n",
-				   dsi1_regs[i].reg, dsi1_regs[i].name,
-				   DSI_READ(dsi1_regs[i].reg));
-		}
-	}
-
-	return 0;
-}
-#endif
-
 static void vc4_dsi_encoder_destroy(struct drm_encoder *encoder)
 {
 	drm_encoder_cleanup(encoder);
@@ -900,8 +845,9 @@ static void vc4_dsi_encoder_enable(struct drm_encoder *encoder)
 	}
 
 	if (debug_dump_regs) {
-		DRM_INFO("DSI regs before:\n");
-		vc4_dsi_dump_regs(dsi);
+		struct drm_printer p = drm_info_printer(&dsi->pdev->dev);
+		dev_info(&dsi->pdev->dev, "DSI regs before:\n");
+		drm_print_regset32(&p, &dsi->regset);
 	}
 
 	/* Round up the clk_set_rate() request slightly, since
@@ -1135,8 +1081,9 @@ static void vc4_dsi_encoder_enable(struct drm_encoder *encoder)
 	drm_bridge_enable(dsi->bridge);
 
 	if (debug_dump_regs) {
-		DRM_INFO("DSI regs after:\n");
-		vc4_dsi_dump_regs(dsi);
+		struct drm_printer p = drm_info_printer(&dsi->pdev->dev);
+		dev_info(&dsi->pdev->dev, "DSI regs after:\n");
+		drm_print_regset32(&p, &dsi->regset);
 	}
 }
 
@@ -1527,6 +1474,15 @@ static int vc4_dsi_bind(struct device *dev, struct device *master, void *data)
 	if (IS_ERR(dsi->regs))
 		return PTR_ERR(dsi->regs);
 
+	dsi->regset.base = dsi->regs;
+	if (dsi->port == 0) {
+		dsi->regset.regs = dsi0_regs;
+		dsi->regset.nregs = ARRAY_SIZE(dsi0_regs);
+	} else {
+		dsi->regset.regs = dsi1_regs;
+		dsi->regset.nregs = ARRAY_SIZE(dsi1_regs);
+	}
+
 	if (DSI_PORT_READ(ID) != DSI_ID_VALUE) {
 		dev_err(dev, "Port returned 0x%08x for ID instead of 0x%08x\n",
 			DSI_PORT_READ(ID), DSI_ID_VALUE);
@@ -1662,6 +1618,11 @@ static int vc4_dsi_bind(struct device *dev, struct device *master, void *data)
 	 */
 	dsi->encoder->bridge = NULL;
 
+	if (dsi->port == 0)
+		vc4_debugfs_add_regset32(drm, "dsi0_regs", &dsi->regset);
+	else
+		vc4_debugfs_add_regset32(drm, "dsi1_regs", &dsi->regset);
+
 	pm_runtime_enable(dev);
 
 	return 0;
diff --git a/drivers/gpu/drm/vc4/vc4_gem.c b/drivers/gpu/drm/vc4/vc4_gem.c
index aea2b8dfec17..d9311be32a4f 100644
--- a/drivers/gpu/drm/vc4/vc4_gem.c
+++ b/drivers/gpu/drm/vc4/vc4_gem.c
@@ -74,6 +74,11 @@ vc4_get_hang_state_ioctl(struct drm_device *dev, void *data,
 	u32 i;
 	int ret = 0;
 
+	if (!vc4->v3d) {
+		DRM_DEBUG("VC4_GET_HANG_STATE with no VC4 V3D probed\n");
+		return -ENODEV;
+	}
+
 	spin_lock_irqsave(&vc4->job_lock, irqflags);
 	kernel_state = vc4->hang_state;
 	if (!kernel_state) {
@@ -536,7 +541,7 @@ vc4_update_bo_seqnos(struct vc4_exec_info *exec, uint64_t seqno)
 		bo = to_vc4_bo(&exec->bo[i]->base);
 		bo->seqno = seqno;
 
-		reservation_object_add_shared_fence(bo->resv, exec->fence);
+		reservation_object_add_shared_fence(bo->base.base.resv, exec->fence);
 	}
 
 	list_for_each_entry(bo, &exec->unref_list, unref_head) {
@@ -547,7 +552,7 @@ vc4_update_bo_seqnos(struct vc4_exec_info *exec, uint64_t seqno)
 		bo = to_vc4_bo(&exec->rcl_write_bo[i]->base);
 		bo->write_seqno = seqno;
 
-		reservation_object_add_excl_fence(bo->resv, exec->fence);
+		reservation_object_add_excl_fence(bo->base.base.resv, exec->fence);
 	}
 }
 
@@ -559,7 +564,7 @@ vc4_unlock_bo_reservations(struct drm_device *dev,
 	int i;
 
 	for (i = 0; i < exec->bo_count; i++) {
-		struct vc4_bo *bo = to_vc4_bo(&exec->bo[i]->base);
+		struct drm_gem_object *bo = &exec->bo[i]->base;
 
 		ww_mutex_unlock(&bo->resv->lock);
 	}
@@ -581,13 +586,13 @@ vc4_lock_bo_reservations(struct drm_device *dev,
 {
 	int contended_lock = -1;
 	int i, ret;
-	struct vc4_bo *bo;
+	struct drm_gem_object *bo;
 
 	ww_acquire_init(acquire_ctx, &reservation_ww_class);
 
 retry:
 	if (contended_lock != -1) {
-		bo = to_vc4_bo(&exec->bo[contended_lock]->base);
+		bo = &exec->bo[contended_lock]->base;
 		ret = ww_mutex_lock_slow_interruptible(&bo->resv->lock,
 						       acquire_ctx);
 		if (ret) {
@@ -600,19 +605,19 @@ retry:
 		if (i == contended_lock)
 			continue;
 
-		bo = to_vc4_bo(&exec->bo[i]->base);
+		bo = &exec->bo[i]->base;
 
 		ret = ww_mutex_lock_interruptible(&bo->resv->lock, acquire_ctx);
 		if (ret) {
 			int j;
 
 			for (j = 0; j < i; j++) {
-				bo = to_vc4_bo(&exec->bo[j]->base);
+				bo = &exec->bo[j]->base;
 				ww_mutex_unlock(&bo->resv->lock);
 			}
 
 			if (contended_lock != -1 && contended_lock >= i) {
-				bo = to_vc4_bo(&exec->bo[contended_lock]->base);
+				bo = &exec->bo[contended_lock]->base;
 
 				ww_mutex_unlock(&bo->resv->lock);
 			}
@@ -633,7 +638,7 @@ retry:
 	 * before we commit the CL to the hardware.
 	 */
 	for (i = 0; i < exec->bo_count; i++) {
-		bo = to_vc4_bo(&exec->bo[i]->base);
+		bo = &exec->bo[i]->base;
 
 		ret = reservation_object_reserve_shared(bo->resv, 1);
 		if (ret) {
@@ -964,12 +969,7 @@ vc4_complete_exec(struct drm_device *dev, struct vc4_exec_info *exec)
 	/* Release the reference we had on the perf monitor. */
 	vc4_perfmon_put(exec->perfmon);
 
-	mutex_lock(&vc4->power_lock);
-	if (--vc4->power_refcount == 0) {
-		pm_runtime_mark_last_busy(&vc4->v3d->pdev->dev);
-		pm_runtime_put_autosuspend(&vc4->v3d->pdev->dev);
-	}
-	mutex_unlock(&vc4->power_lock);
+	vc4_v3d_pm_put(vc4);
 
 	kfree(exec);
 }
@@ -1124,6 +1124,11 @@ vc4_submit_cl_ioctl(struct drm_device *dev, void *data,
 	struct dma_fence *in_fence;
 	int ret = 0;
 
+	if (!vc4->v3d) {
+		DRM_DEBUG("VC4_SUBMIT_CL with no VC4 V3D probed\n");
+		return -ENODEV;
+	}
+
 	if ((args->flags & ~(VC4_SUBMIT_CL_USE_CLEAR_COLOR |
 			     VC4_SUBMIT_CL_FIXED_RCL_ORDER |
 			     VC4_SUBMIT_CL_RCL_ORDER_INCREASING_X |
@@ -1143,17 +1148,11 @@ vc4_submit_cl_ioctl(struct drm_device *dev, void *data,
 		return -ENOMEM;
 	}
 
-	mutex_lock(&vc4->power_lock);
-	if (vc4->power_refcount++ == 0) {
-		ret = pm_runtime_get_sync(&vc4->v3d->pdev->dev);
-		if (ret < 0) {
-			mutex_unlock(&vc4->power_lock);
-			vc4->power_refcount--;
-			kfree(exec);
-			return ret;
-		}
+	ret = vc4_v3d_pm_get(vc4);
+	if (ret) {
+		kfree(exec);
+		return ret;
 	}
-	mutex_unlock(&vc4->power_lock);
 
 	exec->args = args;
 	INIT_LIST_HEAD(&exec->unref_list);
diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
index 88fd5df7e7dc..99fc8569e0f5 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.c
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
@@ -97,6 +97,9 @@ struct vc4_hdmi {
 
 	struct clk *pixel_clock;
 	struct clk *hsm_clock;
+
+	struct debugfs_regset32 hdmi_regset;
+	struct debugfs_regset32 hd_regset;
 };
 
 #define HDMI_READ(offset) readl(vc4->hdmi->hdmicore_regs + offset)
@@ -134,103 +137,69 @@ to_vc4_hdmi_connector(struct drm_connector *connector)
 	return container_of(connector, struct vc4_hdmi_connector, base);
 }
 
-#define HDMI_REG(reg) { reg, #reg }
-static const struct {
-	u32 reg;
-	const char *name;
-} hdmi_regs[] = {
-	HDMI_REG(VC4_HDMI_CORE_REV),
-	HDMI_REG(VC4_HDMI_SW_RESET_CONTROL),
-	HDMI_REG(VC4_HDMI_HOTPLUG_INT),
-	HDMI_REG(VC4_HDMI_HOTPLUG),
-	HDMI_REG(VC4_HDMI_MAI_CHANNEL_MAP),
-	HDMI_REG(VC4_HDMI_MAI_CONFIG),
-	HDMI_REG(VC4_HDMI_MAI_FORMAT),
-	HDMI_REG(VC4_HDMI_AUDIO_PACKET_CONFIG),
-	HDMI_REG(VC4_HDMI_RAM_PACKET_CONFIG),
-	HDMI_REG(VC4_HDMI_HORZA),
-	HDMI_REG(VC4_HDMI_HORZB),
-	HDMI_REG(VC4_HDMI_FIFO_CTL),
-	HDMI_REG(VC4_HDMI_SCHEDULER_CONTROL),
-	HDMI_REG(VC4_HDMI_VERTA0),
-	HDMI_REG(VC4_HDMI_VERTA1),
-	HDMI_REG(VC4_HDMI_VERTB0),
-	HDMI_REG(VC4_HDMI_VERTB1),
-	HDMI_REG(VC4_HDMI_TX_PHY_RESET_CTL),
-	HDMI_REG(VC4_HDMI_TX_PHY_CTL0),
-
-	HDMI_REG(VC4_HDMI_CEC_CNTRL_1),
-	HDMI_REG(VC4_HDMI_CEC_CNTRL_2),
-	HDMI_REG(VC4_HDMI_CEC_CNTRL_3),
-	HDMI_REG(VC4_HDMI_CEC_CNTRL_4),
-	HDMI_REG(VC4_HDMI_CEC_CNTRL_5),
-	HDMI_REG(VC4_HDMI_CPU_STATUS),
-	HDMI_REG(VC4_HDMI_CPU_MASK_STATUS),
-
-	HDMI_REG(VC4_HDMI_CEC_RX_DATA_1),
-	HDMI_REG(VC4_HDMI_CEC_RX_DATA_2),
-	HDMI_REG(VC4_HDMI_CEC_RX_DATA_3),
-	HDMI_REG(VC4_HDMI_CEC_RX_DATA_4),
-	HDMI_REG(VC4_HDMI_CEC_TX_DATA_1),
-	HDMI_REG(VC4_HDMI_CEC_TX_DATA_2),
-	HDMI_REG(VC4_HDMI_CEC_TX_DATA_3),
-	HDMI_REG(VC4_HDMI_CEC_TX_DATA_4),
+static const struct debugfs_reg32 hdmi_regs[] = {
+	VC4_REG32(VC4_HDMI_CORE_REV),
+	VC4_REG32(VC4_HDMI_SW_RESET_CONTROL),
+	VC4_REG32(VC4_HDMI_HOTPLUG_INT),
+	VC4_REG32(VC4_HDMI_HOTPLUG),
+	VC4_REG32(VC4_HDMI_MAI_CHANNEL_MAP),
+	VC4_REG32(VC4_HDMI_MAI_CONFIG),
+	VC4_REG32(VC4_HDMI_MAI_FORMAT),
+	VC4_REG32(VC4_HDMI_AUDIO_PACKET_CONFIG),
+	VC4_REG32(VC4_HDMI_RAM_PACKET_CONFIG),
+	VC4_REG32(VC4_HDMI_HORZA),
+	VC4_REG32(VC4_HDMI_HORZB),
+	VC4_REG32(VC4_HDMI_FIFO_CTL),
+	VC4_REG32(VC4_HDMI_SCHEDULER_CONTROL),
+	VC4_REG32(VC4_HDMI_VERTA0),
+	VC4_REG32(VC4_HDMI_VERTA1),
+	VC4_REG32(VC4_HDMI_VERTB0),
+	VC4_REG32(VC4_HDMI_VERTB1),
+	VC4_REG32(VC4_HDMI_TX_PHY_RESET_CTL),
+	VC4_REG32(VC4_HDMI_TX_PHY_CTL0),
+
+	VC4_REG32(VC4_HDMI_CEC_CNTRL_1),
+	VC4_REG32(VC4_HDMI_CEC_CNTRL_2),
+	VC4_REG32(VC4_HDMI_CEC_CNTRL_3),
+	VC4_REG32(VC4_HDMI_CEC_CNTRL_4),
+	VC4_REG32(VC4_HDMI_CEC_CNTRL_5),
+	VC4_REG32(VC4_HDMI_CPU_STATUS),
+	VC4_REG32(VC4_HDMI_CPU_MASK_STATUS),
+
+	VC4_REG32(VC4_HDMI_CEC_RX_DATA_1),
+	VC4_REG32(VC4_HDMI_CEC_RX_DATA_2),
+	VC4_REG32(VC4_HDMI_CEC_RX_DATA_3),
+	VC4_REG32(VC4_HDMI_CEC_RX_DATA_4),
+	VC4_REG32(VC4_HDMI_CEC_TX_DATA_1),
+	VC4_REG32(VC4_HDMI_CEC_TX_DATA_2),
+	VC4_REG32(VC4_HDMI_CEC_TX_DATA_3),
+	VC4_REG32(VC4_HDMI_CEC_TX_DATA_4),
 };
 
-static const struct {
-	u32 reg;
-	const char *name;
-} hd_regs[] = {
-	HDMI_REG(VC4_HD_M_CTL),
-	HDMI_REG(VC4_HD_MAI_CTL),
-	HDMI_REG(VC4_HD_MAI_THR),
-	HDMI_REG(VC4_HD_MAI_FMT),
-	HDMI_REG(VC4_HD_MAI_SMP),
-	HDMI_REG(VC4_HD_VID_CTL),
-	HDMI_REG(VC4_HD_CSC_CTL),
-	HDMI_REG(VC4_HD_FRAME_COUNT),
+static const struct debugfs_reg32 hd_regs[] = {
+	VC4_REG32(VC4_HD_M_CTL),
+	VC4_REG32(VC4_HD_MAI_CTL),
+	VC4_REG32(VC4_HD_MAI_THR),
+	VC4_REG32(VC4_HD_MAI_FMT),
+	VC4_REG32(VC4_HD_MAI_SMP),
+	VC4_REG32(VC4_HD_VID_CTL),
+	VC4_REG32(VC4_HD_CSC_CTL),
+	VC4_REG32(VC4_HD_FRAME_COUNT),
 };
 
-#ifdef CONFIG_DEBUG_FS
-int vc4_hdmi_debugfs_regs(struct seq_file *m, void *unused)
+static int vc4_hdmi_debugfs_regs(struct seq_file *m, void *unused)
 {
 	struct drm_info_node *node = (struct drm_info_node *)m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(hdmi_regs); i++) {
-		seq_printf(m, "%s (0x%04x): 0x%08x\n",
-			   hdmi_regs[i].name, hdmi_regs[i].reg,
-			   HDMI_READ(hdmi_regs[i].reg));
-	}
+	struct vc4_hdmi *hdmi = vc4->hdmi;
+	struct drm_printer p = drm_seq_file_printer(m);
 
-	for (i = 0; i < ARRAY_SIZE(hd_regs); i++) {
-		seq_printf(m, "%s (0x%04x): 0x%08x\n",
-			   hd_regs[i].name, hd_regs[i].reg,
-			   HD_READ(hd_regs[i].reg));
-	}
+	drm_print_regset32(&p, &hdmi->hdmi_regset);
+	drm_print_regset32(&p, &hdmi->hd_regset);
 
 	return 0;
 }
-#endif /* CONFIG_DEBUG_FS */
-
-static void vc4_hdmi_dump_regs(struct drm_device *dev)
-{
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(hdmi_regs); i++) {
-		DRM_INFO("0x%04x (%s): 0x%08x\n",
-			 hdmi_regs[i].reg, hdmi_regs[i].name,
-			 HDMI_READ(hdmi_regs[i].reg));
-	}
-	for (i = 0; i < ARRAY_SIZE(hd_regs); i++) {
-		DRM_INFO("0x%04x (%s): 0x%08x\n",
-			 hd_regs[i].reg, hd_regs[i].name,
-			 HD_READ(hd_regs[i].reg));
-	}
-}
 
 static enum drm_connector_status
 vc4_hdmi_connector_detect(struct drm_connector *connector, bool force)
@@ -561,8 +530,11 @@ static void vc4_hdmi_encoder_enable(struct drm_encoder *encoder)
 	HDMI_WRITE(VC4_HDMI_TX_PHY_RESET_CTL, 0);
 
 	if (debug_dump_regs) {
-		DRM_INFO("HDMI regs before:\n");
-		vc4_hdmi_dump_regs(dev);
+		struct drm_printer p = drm_info_printer(&hdmi->pdev->dev);
+
+		dev_info(&hdmi->pdev->dev, "HDMI regs before:\n");
+		drm_print_regset32(&p, &hdmi->hdmi_regset);
+		drm_print_regset32(&p, &hdmi->hd_regset);
 	}
 
 	HD_WRITE(VC4_HD_VID_CTL, 0);
@@ -637,8 +609,11 @@ static void vc4_hdmi_encoder_enable(struct drm_encoder *encoder)
 	HDMI_WRITE(VC4_HDMI_FIFO_CTL, VC4_HDMI_FIFO_CTL_MASTER_SLAVE_N);
 
 	if (debug_dump_regs) {
-		DRM_INFO("HDMI regs after:\n");
-		vc4_hdmi_dump_regs(dev);
+		struct drm_printer p = drm_info_printer(&hdmi->pdev->dev);
+
+		dev_info(&hdmi->pdev->dev, "HDMI regs after:\n");
+		drm_print_regset32(&p, &hdmi->hdmi_regset);
+		drm_print_regset32(&p, &hdmi->hd_regset);
 	}
 
 	HD_WRITE(VC4_HD_VID_CTL,
@@ -1333,6 +1308,13 @@ static int vc4_hdmi_bind(struct device *dev, struct device *master, void *data)
 	if (IS_ERR(hdmi->hd_regs))
 		return PTR_ERR(hdmi->hd_regs);
 
+	hdmi->hdmi_regset.base = hdmi->hdmicore_regs;
+	hdmi->hdmi_regset.regs = hdmi_regs;
+	hdmi->hdmi_regset.nregs = ARRAY_SIZE(hdmi_regs);
+	hdmi->hd_regset.base = hdmi->hd_regs;
+	hdmi->hd_regset.regs = hd_regs;
+	hdmi->hd_regset.nregs = ARRAY_SIZE(hd_regs);
+
 	hdmi->pixel_clock = devm_clk_get(dev, "pixel");
 	if (IS_ERR(hdmi->pixel_clock)) {
 		DRM_ERROR("Failed to get pixel clock\n");
@@ -1448,6 +1430,8 @@ static int vc4_hdmi_bind(struct device *dev, struct device *master, void *data)
 	if (ret)
 		goto err_destroy_encoder;
 
+	vc4_debugfs_add_file(drm, "hdmi_regs", vc4_hdmi_debugfs_regs, hdmi);
+
 	return 0;
 
 #ifdef CONFIG_DRM_VC4_HDMI_CEC
diff --git a/drivers/gpu/drm/vc4/vc4_hvs.c b/drivers/gpu/drm/vc4/vc4_hvs.c
index 5d8c749c9749..f746e9a7a88c 100644
--- a/drivers/gpu/drm/vc4/vc4_hvs.c
+++ b/drivers/gpu/drm/vc4/vc4_hvs.c
@@ -22,58 +22,52 @@
  * each CRTC.
  */
 
+#include <drm/drm_atomic_helper.h>
 #include <linux/component.h>
 #include "vc4_drv.h"
 #include "vc4_regs.h"
 
-#define HVS_REG(reg) { reg, #reg }
-static const struct {
-	u32 reg;
-	const char *name;
-} hvs_regs[] = {
-	HVS_REG(SCALER_DISPCTRL),
-	HVS_REG(SCALER_DISPSTAT),
-	HVS_REG(SCALER_DISPID),
-	HVS_REG(SCALER_DISPECTRL),
-	HVS_REG(SCALER_DISPPROF),
-	HVS_REG(SCALER_DISPDITHER),
-	HVS_REG(SCALER_DISPEOLN),
-	HVS_REG(SCALER_DISPLIST0),
-	HVS_REG(SCALER_DISPLIST1),
-	HVS_REG(SCALER_DISPLIST2),
-	HVS_REG(SCALER_DISPLSTAT),
-	HVS_REG(SCALER_DISPLACT0),
-	HVS_REG(SCALER_DISPLACT1),
-	HVS_REG(SCALER_DISPLACT2),
-	HVS_REG(SCALER_DISPCTRL0),
-	HVS_REG(SCALER_DISPBKGND0),
-	HVS_REG(SCALER_DISPSTAT0),
-	HVS_REG(SCALER_DISPBASE0),
-	HVS_REG(SCALER_DISPCTRL1),
-	HVS_REG(SCALER_DISPBKGND1),
-	HVS_REG(SCALER_DISPSTAT1),
-	HVS_REG(SCALER_DISPBASE1),
-	HVS_REG(SCALER_DISPCTRL2),
-	HVS_REG(SCALER_DISPBKGND2),
-	HVS_REG(SCALER_DISPSTAT2),
-	HVS_REG(SCALER_DISPBASE2),
-	HVS_REG(SCALER_DISPALPHA2),
-	HVS_REG(SCALER_OLEDOFFS),
-	HVS_REG(SCALER_OLEDCOEF0),
-	HVS_REG(SCALER_OLEDCOEF1),
-	HVS_REG(SCALER_OLEDCOEF2),
+static const struct debugfs_reg32 hvs_regs[] = {
+	VC4_REG32(SCALER_DISPCTRL),
+	VC4_REG32(SCALER_DISPSTAT),
+	VC4_REG32(SCALER_DISPID),
+	VC4_REG32(SCALER_DISPECTRL),
+	VC4_REG32(SCALER_DISPPROF),
+	VC4_REG32(SCALER_DISPDITHER),
+	VC4_REG32(SCALER_DISPEOLN),
+	VC4_REG32(SCALER_DISPLIST0),
+	VC4_REG32(SCALER_DISPLIST1),
+	VC4_REG32(SCALER_DISPLIST2),
+	VC4_REG32(SCALER_DISPLSTAT),
+	VC4_REG32(SCALER_DISPLACT0),
+	VC4_REG32(SCALER_DISPLACT1),
+	VC4_REG32(SCALER_DISPLACT2),
+	VC4_REG32(SCALER_DISPCTRL0),
+	VC4_REG32(SCALER_DISPBKGND0),
+	VC4_REG32(SCALER_DISPSTAT0),
+	VC4_REG32(SCALER_DISPBASE0),
+	VC4_REG32(SCALER_DISPCTRL1),
+	VC4_REG32(SCALER_DISPBKGND1),
+	VC4_REG32(SCALER_DISPSTAT1),
+	VC4_REG32(SCALER_DISPBASE1),
+	VC4_REG32(SCALER_DISPCTRL2),
+	VC4_REG32(SCALER_DISPBKGND2),
+	VC4_REG32(SCALER_DISPSTAT2),
+	VC4_REG32(SCALER_DISPBASE2),
+	VC4_REG32(SCALER_DISPALPHA2),
+	VC4_REG32(SCALER_OLEDOFFS),
+	VC4_REG32(SCALER_OLEDCOEF0),
+	VC4_REG32(SCALER_OLEDCOEF1),
+	VC4_REG32(SCALER_OLEDCOEF2),
 };
 
 void vc4_hvs_dump_state(struct drm_device *dev)
 {
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	struct drm_printer p = drm_info_printer(&vc4->hvs->pdev->dev);
 	int i;
 
-	for (i = 0; i < ARRAY_SIZE(hvs_regs); i++) {
-		DRM_INFO("0x%04x (%s): 0x%08x\n",
-			 hvs_regs[i].reg, hvs_regs[i].name,
-			 HVS_READ(hvs_regs[i].reg));
-	}
+	drm_print_regset32(&p, &vc4->hvs->regset);
 
 	DRM_INFO("HVS ctx:\n");
 	for (i = 0; i < 64; i += 4) {
@@ -86,23 +80,17 @@ void vc4_hvs_dump_state(struct drm_device *dev)
 	}
 }
 
-#ifdef CONFIG_DEBUG_FS
-int vc4_hvs_debugfs_regs(struct seq_file *m, void *unused)
+static int vc4_hvs_debugfs_underrun(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *)m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	int i;
+	struct drm_printer p = drm_seq_file_printer(m);
 
-	for (i = 0; i < ARRAY_SIZE(hvs_regs); i++) {
-		seq_printf(m, "%s (0x%04x): 0x%08x\n",
-			   hvs_regs[i].name, hvs_regs[i].reg,
-			   HVS_READ(hvs_regs[i].reg));
-	}
+	drm_printf(&p, "%d\n", atomic_read(&vc4->underrun));
 
 	return 0;
 }
-#endif
 
 /* The filter kernel is composed of dwords each containing 3 9-bit
  * signed integers packed next to each other.
@@ -166,6 +154,67 @@ static int vc4_hvs_upload_linear_kernel(struct vc4_hvs *hvs,
 	return 0;
 }
 
+void vc4_hvs_mask_underrun(struct drm_device *dev, int channel)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	u32 dispctrl = HVS_READ(SCALER_DISPCTRL);
+
+	dispctrl &= ~SCALER_DISPCTRL_DSPEISLUR(channel);
+
+	HVS_WRITE(SCALER_DISPCTRL, dispctrl);
+}
+
+void vc4_hvs_unmask_underrun(struct drm_device *dev, int channel)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	u32 dispctrl = HVS_READ(SCALER_DISPCTRL);
+
+	dispctrl |= SCALER_DISPCTRL_DSPEISLUR(channel);
+
+	HVS_WRITE(SCALER_DISPSTAT,
+		  SCALER_DISPSTAT_EUFLOW(channel));
+	HVS_WRITE(SCALER_DISPCTRL, dispctrl);
+}
+
+static void vc4_hvs_report_underrun(struct drm_device *dev)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+
+	atomic_inc(&vc4->underrun);
+	DRM_DEV_ERROR(dev->dev, "HVS underrun\n");
+}
+
+static irqreturn_t vc4_hvs_irq_handler(int irq, void *data)
+{
+	struct drm_device *dev = data;
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	irqreturn_t irqret = IRQ_NONE;
+	int channel;
+	u32 control;
+	u32 status;
+
+	status = HVS_READ(SCALER_DISPSTAT);
+	control = HVS_READ(SCALER_DISPCTRL);
+
+	for (channel = 0; channel < SCALER_CHANNELS_COUNT; channel++) {
+		/* Interrupt masking is not always honored, so check it here. */
+		if (status & SCALER_DISPSTAT_EUFLOW(channel) &&
+		    control & SCALER_DISPCTRL_DSPEISLUR(channel)) {
+			vc4_hvs_mask_underrun(dev, channel);
+			vc4_hvs_report_underrun(dev);
+
+			irqret = IRQ_HANDLED;
+		}
+	}
+
+	/* Clear every per-channel interrupt flag. */
+	HVS_WRITE(SCALER_DISPSTAT, SCALER_DISPSTAT_IRQMASK(0) |
+				   SCALER_DISPSTAT_IRQMASK(1) |
+				   SCALER_DISPSTAT_IRQMASK(2));
+
+	return irqret;
+}
+
 static int vc4_hvs_bind(struct device *dev, struct device *master, void *data)
 {
 	struct platform_device *pdev = to_platform_device(dev);
@@ -185,6 +234,10 @@ static int vc4_hvs_bind(struct device *dev, struct device *master, void *data)
 	if (IS_ERR(hvs->regs))
 		return PTR_ERR(hvs->regs);
 
+	hvs->regset.base = hvs->regs;
+	hvs->regset.regs = hvs_regs;
+	hvs->regset.nregs = ARRAY_SIZE(hvs_regs);
+
 	hvs->dlist = hvs->regs + SCALER_DLIST_START;
 
 	spin_lock_init(&hvs->mm_lock);
@@ -219,15 +272,40 @@ static int vc4_hvs_bind(struct device *dev, struct device *master, void *data)
 	dispctrl = HVS_READ(SCALER_DISPCTRL);
 
 	dispctrl |= SCALER_DISPCTRL_ENABLE;
+	dispctrl |= SCALER_DISPCTRL_DISPEIRQ(0) |
+		    SCALER_DISPCTRL_DISPEIRQ(1) |
+		    SCALER_DISPCTRL_DISPEIRQ(2);
 
 	/* Set DSP3 (PV1) to use HVS channel 2, which would otherwise
 	 * be unused.
 	 */
 	dispctrl &= ~SCALER_DISPCTRL_DSP3_MUX_MASK;
+	dispctrl &= ~(SCALER_DISPCTRL_DMAEIRQ |
+		      SCALER_DISPCTRL_SLVWREIRQ |
+		      SCALER_DISPCTRL_SLVRDEIRQ |
+		      SCALER_DISPCTRL_DSPEIEOF(0) |
+		      SCALER_DISPCTRL_DSPEIEOF(1) |
+		      SCALER_DISPCTRL_DSPEIEOF(2) |
+		      SCALER_DISPCTRL_DSPEIEOLN(0) |
+		      SCALER_DISPCTRL_DSPEIEOLN(1) |
+		      SCALER_DISPCTRL_DSPEIEOLN(2) |
+		      SCALER_DISPCTRL_DSPEISLUR(0) |
+		      SCALER_DISPCTRL_DSPEISLUR(1) |
+		      SCALER_DISPCTRL_DSPEISLUR(2) |
+		      SCALER_DISPCTRL_SCLEIRQ);
 	dispctrl |= VC4_SET_FIELD(2, SCALER_DISPCTRL_DSP3_MUX);
 
 	HVS_WRITE(SCALER_DISPCTRL, dispctrl);
 
+	ret = devm_request_irq(dev, platform_get_irq(pdev, 0),
+			       vc4_hvs_irq_handler, 0, "vc4 hvs", drm);
+	if (ret)
+		return ret;
+
+	vc4_debugfs_add_regset32(drm, "hvs_regs", &hvs->regset);
+	vc4_debugfs_add_file(drm, "hvs_underrun", vc4_hvs_debugfs_underrun,
+			     NULL);
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/vc4/vc4_irq.c b/drivers/gpu/drm/vc4/vc4_irq.c
index 4cd2ccfe15f4..ffd0a4388752 100644
--- a/drivers/gpu/drm/vc4/vc4_irq.c
+++ b/drivers/gpu/drm/vc4/vc4_irq.c
@@ -229,6 +229,9 @@ vc4_irq_preinstall(struct drm_device *dev)
 {
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
 
+	if (!vc4->v3d)
+		return;
+
 	init_waitqueue_head(&vc4->job_wait_queue);
 	INIT_WORK(&vc4->overflow_mem_work, vc4_overflow_mem_work);
 
@@ -243,6 +246,9 @@ vc4_irq_postinstall(struct drm_device *dev)
 {
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
 
+	if (!vc4->v3d)
+		return 0;
+
 	/* Enable both the render done and out of memory interrupts. */
 	V3D_WRITE(V3D_INTENA, V3D_DRIVER_IRQS);
 
@@ -254,6 +260,9 @@ vc4_irq_uninstall(struct drm_device *dev)
 {
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
 
+	if (!vc4->v3d)
+		return;
+
 	/* Disable sending interrupts for our driver's IRQs. */
 	V3D_WRITE(V3D_INTDIS, V3D_DRIVER_IRQS);
 
diff --git a/drivers/gpu/drm/vc4/vc4_kms.c b/drivers/gpu/drm/vc4/vc4_kms.c
index 91b8c72ff361..295dacc8bcb9 100644
--- a/drivers/gpu/drm/vc4/vc4_kms.c
+++ b/drivers/gpu/drm/vc4/vc4_kms.c
@@ -34,6 +34,18 @@ static struct vc4_ctm_state *to_vc4_ctm_state(struct drm_private_state *priv)
 	return container_of(priv, struct vc4_ctm_state, base);
 }
 
+struct vc4_load_tracker_state {
+	struct drm_private_state base;
+	u64 hvs_load;
+	u64 membus_load;
+};
+
+static struct vc4_load_tracker_state *
+to_vc4_load_tracker_state(struct drm_private_state *priv)
+{
+	return container_of(priv, struct vc4_load_tracker_state, base);
+}
+
 static struct vc4_ctm_state *vc4_get_ctm_state(struct drm_atomic_state *state,
 					       struct drm_private_obj *manager)
 {
@@ -138,6 +150,16 @@ vc4_atomic_complete_commit(struct drm_atomic_state *state)
 {
 	struct drm_device *dev = state->dev;
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	struct vc4_crtc *vc4_crtc;
+	int i;
+
+	for (i = 0; i < dev->mode_config.num_crtc; i++) {
+		if (!state->crtcs[i].ptr || !state->crtcs[i].commit)
+			continue;
+
+		vc4_crtc = to_vc4_crtc(state->crtcs[i].ptr);
+		vc4_hvs_mask_underrun(dev, vc4_crtc->channel);
+	}
 
 	drm_atomic_helper_wait_for_fences(dev, state, false);
 
@@ -385,6 +407,85 @@ vc4_ctm_atomic_check(struct drm_device *dev, struct drm_atomic_state *state)
 	return 0;
 }
 
+static int vc4_load_tracker_atomic_check(struct drm_atomic_state *state)
+{
+	struct drm_plane_state *old_plane_state, *new_plane_state;
+	struct vc4_dev *vc4 = to_vc4_dev(state->dev);
+	struct vc4_load_tracker_state *load_state;
+	struct drm_private_state *priv_state;
+	struct drm_plane *plane;
+	int i;
+
+	priv_state = drm_atomic_get_private_obj_state(state,
+						      &vc4->load_tracker);
+	if (IS_ERR(priv_state))
+		return PTR_ERR(priv_state);
+
+	load_state = to_vc4_load_tracker_state(priv_state);
+	for_each_oldnew_plane_in_state(state, plane, old_plane_state,
+				       new_plane_state, i) {
+		struct vc4_plane_state *vc4_plane_state;
+
+		if (old_plane_state->fb && old_plane_state->crtc) {
+			vc4_plane_state = to_vc4_plane_state(old_plane_state);
+			load_state->membus_load -= vc4_plane_state->membus_load;
+			load_state->hvs_load -= vc4_plane_state->hvs_load;
+		}
+
+		if (new_plane_state->fb && new_plane_state->crtc) {
+			vc4_plane_state = to_vc4_plane_state(new_plane_state);
+			load_state->membus_load += vc4_plane_state->membus_load;
+			load_state->hvs_load += vc4_plane_state->hvs_load;
+		}
+	}
+
+	/* Don't check the load when the tracker is disabled. */
+	if (!vc4->load_tracker_enabled)
+		return 0;
+
+	/* The absolute limit is 2Gbyte/sec, but let's take a margin to let
+	 * the system work when other blocks are accessing the memory.
+	 */
+	if (load_state->membus_load > SZ_1G + SZ_512M)
+		return -ENOSPC;
+
+	/* HVS clock is supposed to run @ 250Mhz, let's take a margin and
+	 * consider the maximum number of cycles is 240M.
+	 */
+	if (load_state->hvs_load > 240000000ULL)
+		return -ENOSPC;
+
+	return 0;
+}
+
+static struct drm_private_state *
+vc4_load_tracker_duplicate_state(struct drm_private_obj *obj)
+{
+	struct vc4_load_tracker_state *state;
+
+	state = kmemdup(obj->state, sizeof(*state), GFP_KERNEL);
+	if (!state)
+		return NULL;
+
+	__drm_atomic_helper_private_obj_duplicate_state(obj, &state->base);
+
+	return &state->base;
+}
+
+static void vc4_load_tracker_destroy_state(struct drm_private_obj *obj,
+					   struct drm_private_state *state)
+{
+	struct vc4_load_tracker_state *load_state;
+
+	load_state = to_vc4_load_tracker_state(state);
+	kfree(load_state);
+}
+
+static const struct drm_private_state_funcs vc4_load_tracker_state_funcs = {
+	.atomic_duplicate_state = vc4_load_tracker_duplicate_state,
+	.atomic_destroy_state = vc4_load_tracker_destroy_state,
+};
+
 static int
 vc4_atomic_check(struct drm_device *dev, struct drm_atomic_state *state)
 {
@@ -394,7 +495,11 @@ vc4_atomic_check(struct drm_device *dev, struct drm_atomic_state *state)
 	if (ret < 0)
 		return ret;
 
-	return drm_atomic_helper_check(dev, state);
+	ret = drm_atomic_helper_check(dev, state);
+	if (ret)
+		return ret;
+
+	return vc4_load_tracker_atomic_check(state);
 }
 
 static const struct drm_mode_config_funcs vc4_mode_funcs = {
@@ -407,13 +512,20 @@ int vc4_kms_load(struct drm_device *dev)
 {
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	struct vc4_ctm_state *ctm_state;
+	struct vc4_load_tracker_state *load_state;
 	int ret;
 
+	/* Start with the load tracker enabled. Can be disabled through the
+	 * debugfs load_tracker file.
+	 */
+	vc4->load_tracker_enabled = true;
+
 	sema_init(&vc4->async_modeset, 1);
 
 	/* Set support for vblank irq fast disable, before drm_vblank_init() */
 	dev->vblank_disable_immediate = true;
 
+	dev->irq_enabled = true;
 	ret = drm_vblank_init(dev, dev->mode_config.num_crtc);
 	if (ret < 0) {
 		dev_err(dev->dev, "failed to initialize vblank\n");
@@ -436,6 +548,15 @@ int vc4_kms_load(struct drm_device *dev)
 	drm_atomic_private_obj_init(dev, &vc4->ctm_manager, &ctm_state->base,
 				    &vc4_ctm_state_funcs);
 
+	load_state = kzalloc(sizeof(*load_state), GFP_KERNEL);
+	if (!load_state) {
+		drm_atomic_private_obj_fini(&vc4->ctm_manager);
+		return -ENOMEM;
+	}
+
+	drm_atomic_private_obj_init(dev, &vc4->load_tracker, &load_state->base,
+				    &vc4_load_tracker_state_funcs);
+
 	drm_mode_config_reset(dev);
 
 	drm_kms_helper_poll_init(dev);
diff --git a/drivers/gpu/drm/vc4/vc4_perfmon.c b/drivers/gpu/drm/vc4/vc4_perfmon.c
index 495150415020..f4aa75efd16b 100644
--- a/drivers/gpu/drm/vc4/vc4_perfmon.c
+++ b/drivers/gpu/drm/vc4/vc4_perfmon.c
@@ -100,12 +100,18 @@ void vc4_perfmon_close_file(struct vc4_file *vc4file)
 int vc4_perfmon_create_ioctl(struct drm_device *dev, void *data,
 			     struct drm_file *file_priv)
 {
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	struct vc4_file *vc4file = file_priv->driver_priv;
 	struct drm_vc4_perfmon_create *req = data;
 	struct vc4_perfmon *perfmon;
 	unsigned int i;
 	int ret;
 
+	if (!vc4->v3d) {
+		DRM_DEBUG("Creating perfmon no VC4 V3D probed\n");
+		return -ENODEV;
+	}
+
 	/* Number of monitored counters cannot exceed HW limits. */
 	if (req->ncounters > DRM_VC4_MAX_PERF_COUNTERS ||
 	    !req->ncounters)
@@ -146,10 +152,16 @@ int vc4_perfmon_create_ioctl(struct drm_device *dev, void *data,
 int vc4_perfmon_destroy_ioctl(struct drm_device *dev, void *data,
 			      struct drm_file *file_priv)
 {
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	struct vc4_file *vc4file = file_priv->driver_priv;
 	struct drm_vc4_perfmon_destroy *req = data;
 	struct vc4_perfmon *perfmon;
 
+	if (!vc4->v3d) {
+		DRM_DEBUG("Destroying perfmon no VC4 V3D probed\n");
+		return -ENODEV;
+	}
+
 	mutex_lock(&vc4file->perfmon.lock);
 	perfmon = idr_remove(&vc4file->perfmon.idr, req->id);
 	mutex_unlock(&vc4file->perfmon.lock);
@@ -164,11 +176,17 @@ int vc4_perfmon_destroy_ioctl(struct drm_device *dev, void *data,
 int vc4_perfmon_get_values_ioctl(struct drm_device *dev, void *data,
 				 struct drm_file *file_priv)
 {
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	struct vc4_file *vc4file = file_priv->driver_priv;
 	struct drm_vc4_perfmon_get_values *req = data;
 	struct vc4_perfmon *perfmon;
 	int ret;
 
+	if (!vc4->v3d) {
+		DRM_DEBUG("Getting perfmon no VC4 V3D probed\n");
+		return -ENODEV;
+	}
+
 	mutex_lock(&vc4file->perfmon.lock);
 	perfmon = idr_find(&vc4file->perfmon.idr, req->id);
 	vc4_perfmon_get(perfmon);
diff --git a/drivers/gpu/drm/vc4/vc4_plane.c b/drivers/gpu/drm/vc4/vc4_plane.c
index d098337c10e9..4d918d3e4858 100644
--- a/drivers/gpu/drm/vc4/vc4_plane.c
+++ b/drivers/gpu/drm/vc4/vc4_plane.c
@@ -488,6 +488,61 @@ static void vc4_write_scaling_parameters(struct drm_plane_state *state,
 	}
 }
 
+static void vc4_plane_calc_load(struct drm_plane_state *state)
+{
+	unsigned int hvs_load_shift, vrefresh, i;
+	struct drm_framebuffer *fb = state->fb;
+	struct vc4_plane_state *vc4_state;
+	struct drm_crtc_state *crtc_state;
+	unsigned int vscale_factor;
+
+	vc4_state = to_vc4_plane_state(state);
+	crtc_state = drm_atomic_get_existing_crtc_state(state->state,
+							state->crtc);
+	vrefresh = drm_mode_vrefresh(&crtc_state->adjusted_mode);
+
+	/* The HVS is able to process 2 pixels/cycle when scaling the source,
+	 * 4 pixels/cycle otherwise.
+	 * Alpha blending step seems to be pipelined and it's always operating
+	 * at 4 pixels/cycle, so the limiting aspect here seems to be the
+	 * scaler block.
+	 * HVS load is expressed in clk-cycles/sec (AKA Hz).
+	 */
+	if (vc4_state->x_scaling[0] != VC4_SCALING_NONE ||
+	    vc4_state->x_scaling[1] != VC4_SCALING_NONE ||
+	    vc4_state->y_scaling[0] != VC4_SCALING_NONE ||
+	    vc4_state->y_scaling[1] != VC4_SCALING_NONE)
+		hvs_load_shift = 1;
+	else
+		hvs_load_shift = 2;
+
+	vc4_state->membus_load = 0;
+	vc4_state->hvs_load = 0;
+	for (i = 0; i < fb->format->num_planes; i++) {
+		/* Even if the bandwidth/plane required for a single frame is
+		 *
+		 * vc4_state->src_w[i] * vc4_state->src_h[i] * cpp * vrefresh
+		 *
+		 * when downscaling, we have to read more pixels per line in
+		 * the time frame reserved for a single line, so the bandwidth
+		 * demand can be punctually higher. To account for that, we
+		 * calculate the down-scaling factor and multiply the plane
+		 * load by this number. We're likely over-estimating the read
+		 * demand, but that's better than under-estimating it.
+		 */
+		vscale_factor = DIV_ROUND_UP(vc4_state->src_h[i],
+					     vc4_state->crtc_h);
+		vc4_state->membus_load += vc4_state->src_w[i] *
+					  vc4_state->src_h[i] * vscale_factor *
+					  fb->format->cpp[i];
+		vc4_state->hvs_load += vc4_state->crtc_h * vc4_state->crtc_w;
+	}
+
+	vc4_state->hvs_load *= vrefresh;
+	vc4_state->hvs_load >>= hvs_load_shift;
+	vc4_state->membus_load *= vrefresh;
+}
+
 static int vc4_plane_allocate_lbm(struct drm_plane_state *state)
 {
 	struct vc4_dev *vc4 = to_vc4_dev(state->plane->dev);
@@ -875,6 +930,8 @@ static int vc4_plane_mode_set(struct drm_plane *plane,
 	 */
 	vc4_state->dlist_initialized = 1;
 
+	vc4_plane_calc_load(state);
+
 	return 0;
 }
 
@@ -1082,7 +1139,7 @@ static int vc4_prepare_fb(struct drm_plane *plane,
 
 	bo = to_vc4_bo(&drm_fb_cma_get_gem_obj(state->fb, 0)->base);
 
-	fence = reservation_object_get_excl_rcu(bo->resv);
+	fence = reservation_object_get_excl_rcu(bo->base.base.resv);
 	drm_atomic_set_fence_for_plane(state, fence);
 
 	if (plane->state->fb == state->fb)
diff --git a/drivers/gpu/drm/vc4/vc4_regs.h b/drivers/gpu/drm/vc4/vc4_regs.h
index 931088014272..c0c5fadaf7e3 100644
--- a/drivers/gpu/drm/vc4/vc4_regs.h
+++ b/drivers/gpu/drm/vc4/vc4_regs.h
@@ -212,11 +212,11 @@
 
 #define PV_HACT_ACT				0x30
 
+#define SCALER_CHANNELS_COUNT			3
+
 #define SCALER_DISPCTRL                         0x00000000
 /* Global register for clock gating the HVS */
 # define SCALER_DISPCTRL_ENABLE			BIT(31)
-# define SCALER_DISPCTRL_DSP2EISLUR		BIT(15)
-# define SCALER_DISPCTRL_DSP1EISLUR		BIT(14)
 # define SCALER_DISPCTRL_DSP3_MUX_MASK		VC4_MASK(19, 18)
 # define SCALER_DISPCTRL_DSP3_MUX_SHIFT		18
 
@@ -224,45 +224,25 @@
  * SCALER_DISPSTAT_IRQDISP0.  Note that short frame contributions are
  * always enabled.
  */
-# define SCALER_DISPCTRL_DSP0EISLUR		BIT(13)
-# define SCALER_DISPCTRL_DSP2EIEOLN		BIT(12)
-# define SCALER_DISPCTRL_DSP2EIEOF		BIT(11)
-# define SCALER_DISPCTRL_DSP1EIEOLN		BIT(10)
-# define SCALER_DISPCTRL_DSP1EIEOF		BIT(9)
+# define SCALER_DISPCTRL_DSPEISLUR(x)		BIT(13 + (x))
 /* Enables Display 0 end-of-line-N contribution to
  * SCALER_DISPSTAT_IRQDISP0
  */
-# define SCALER_DISPCTRL_DSP0EIEOLN		BIT(8)
+# define SCALER_DISPCTRL_DSPEIEOLN(x)		BIT(8 + ((x) * 2))
 /* Enables Display 0 EOF contribution to SCALER_DISPSTAT_IRQDISP0 */
-# define SCALER_DISPCTRL_DSP0EIEOF		BIT(7)
+# define SCALER_DISPCTRL_DSPEIEOF(x)		BIT(7 + ((x) * 2))
 
 # define SCALER_DISPCTRL_SLVRDEIRQ		BIT(6)
 # define SCALER_DISPCTRL_SLVWREIRQ		BIT(5)
 # define SCALER_DISPCTRL_DMAEIRQ		BIT(4)
-# define SCALER_DISPCTRL_DISP2EIRQ		BIT(3)
-# define SCALER_DISPCTRL_DISP1EIRQ		BIT(2)
 /* Enables interrupt generation on the enabled EOF/EOLN/EISLUR
  * bits and short frames..
  */
-# define SCALER_DISPCTRL_DISP0EIRQ		BIT(1)
+# define SCALER_DISPCTRL_DISPEIRQ(x)		BIT(1 + (x))
 /* Enables interrupt generation on scaler profiler interrupt. */
 # define SCALER_DISPCTRL_SCLEIRQ		BIT(0)
 
 #define SCALER_DISPSTAT                         0x00000004
-# define SCALER_DISPSTAT_COBLOW2		BIT(29)
-# define SCALER_DISPSTAT_EOLN2			BIT(28)
-# define SCALER_DISPSTAT_ESFRAME2		BIT(27)
-# define SCALER_DISPSTAT_ESLINE2		BIT(26)
-# define SCALER_DISPSTAT_EUFLOW2		BIT(25)
-# define SCALER_DISPSTAT_EOF2			BIT(24)
-
-# define SCALER_DISPSTAT_COBLOW1		BIT(21)
-# define SCALER_DISPSTAT_EOLN1			BIT(20)
-# define SCALER_DISPSTAT_ESFRAME1		BIT(19)
-# define SCALER_DISPSTAT_ESLINE1		BIT(18)
-# define SCALER_DISPSTAT_EUFLOW1		BIT(17)
-# define SCALER_DISPSTAT_EOF1			BIT(16)
-
 # define SCALER_DISPSTAT_RESP_MASK		VC4_MASK(15, 14)
 # define SCALER_DISPSTAT_RESP_SHIFT		14
 # define SCALER_DISPSTAT_RESP_OKAY		0
@@ -270,23 +250,26 @@
 # define SCALER_DISPSTAT_RESP_SLVERR		2
 # define SCALER_DISPSTAT_RESP_DECERR		3
 
-# define SCALER_DISPSTAT_COBLOW0		BIT(13)
+# define SCALER_DISPSTAT_COBLOW(x)		BIT(13 + ((x) * 8))
 /* Set when the DISPEOLN line is done compositing. */
-# define SCALER_DISPSTAT_EOLN0			BIT(12)
+# define SCALER_DISPSTAT_EOLN(x)		BIT(12 + ((x) * 8))
 /* Set when VSTART is seen but there are still pixels in the current
  * output line.
  */
-# define SCALER_DISPSTAT_ESFRAME0		BIT(11)
+# define SCALER_DISPSTAT_ESFRAME(x)		BIT(11 + ((x) * 8))
 /* Set when HSTART is seen but there are still pixels in the current
  * output line.
  */
-# define SCALER_DISPSTAT_ESLINE0		BIT(10)
+# define SCALER_DISPSTAT_ESLINE(x)		BIT(10 + ((x) * 8))
 /* Set when the the downstream tries to read from the display FIFO
  * while it's empty.
  */
-# define SCALER_DISPSTAT_EUFLOW0		BIT(9)
+# define SCALER_DISPSTAT_EUFLOW(x)		BIT(9 + ((x) * 8))
 /* Set when the display mode changes from RUN to EOF */
-# define SCALER_DISPSTAT_EOF0			BIT(8)
+# define SCALER_DISPSTAT_EOF(x)			BIT(8 + ((x) * 8))
+
+# define SCALER_DISPSTAT_IRQMASK(x)		VC4_MASK(13 + ((x) * 8), \
+							 8 + ((x) * 8))
 
 /* Set on AXI invalid DMA ID error. */
 # define SCALER_DISPSTAT_DMA_ERROR		BIT(7)
@@ -298,12 +281,10 @@
  * SCALER_DISPSTAT_RESP_ERROR is not SCALER_DISPSTAT_RESP_OKAY.
  */
 # define SCALER_DISPSTAT_IRQDMA			BIT(4)
-# define SCALER_DISPSTAT_IRQDISP2		BIT(3)
-# define SCALER_DISPSTAT_IRQDISP1		BIT(2)
 /* Set when any of the EOF/EOLN/ESFRAME/ESLINE bits are set and their
  * corresponding interrupt bit is enabled in DISPCTRL.
  */
-# define SCALER_DISPSTAT_IRQDISP0		BIT(1)
+# define SCALER_DISPSTAT_IRQDISP(x)		BIT(1 + (x))
 /* On read, the profiler interrupt.  On write, clear *all* interrupt bits. */
 # define SCALER_DISPSTAT_IRQSCL			BIT(0)
 
diff --git a/drivers/gpu/drm/vc4/vc4_render_cl.c b/drivers/gpu/drm/vc4/vc4_render_cl.c
index 273984f71ae2..3c918eeaf56e 100644
--- a/drivers/gpu/drm/vc4/vc4_render_cl.c
+++ b/drivers/gpu/drm/vc4/vc4_render_cl.c
@@ -148,6 +148,12 @@ static void emit_tile(struct vc4_exec_info *exec,
 	}
 
 	if (setup->zs_read) {
+		if (setup->color_read) {
+			/* Exec previous load. */
+			vc4_tile_coordinates(setup, x, y);
+			vc4_store_before_load(setup);
+		}
+
 		if (args->zs_read.flags &
 		    VC4_SUBMIT_RCL_SURFACE_READ_IS_FULL_RES) {
 			rcl_u8(setup, VC4_PACKET_LOAD_FULL_RES_TILE_BUFFER);
@@ -156,12 +162,6 @@ static void emit_tile(struct vc4_exec_info *exec,
 						    &args->zs_read, x, y) |
 				VC4_LOADSTORE_FULL_RES_DISABLE_COLOR);
 		} else {
-			if (setup->color_read) {
-				/* Exec previous load. */
-				vc4_tile_coordinates(setup, x, y);
-				vc4_store_before_load(setup);
-			}
-
 			rcl_u8(setup, VC4_PACKET_LOAD_TILE_BUFFER_GENERAL);
 			rcl_u16(setup, args->zs_read.bits);
 			rcl_u32(setup, setup->zs_read->paddr +
@@ -291,16 +291,15 @@ static int vc4_create_rcl_bo(struct drm_device *dev, struct vc4_exec_info *exec,
 		}
 	}
 	if (setup->zs_read) {
+		if (setup->color_read) {
+			loop_body_size += VC4_PACKET_TILE_COORDINATES_SIZE;
+			loop_body_size += VC4_PACKET_STORE_TILE_BUFFER_GENERAL_SIZE;
+		}
+
 		if (args->zs_read.flags &
 		    VC4_SUBMIT_RCL_SURFACE_READ_IS_FULL_RES) {
 			loop_body_size += VC4_PACKET_LOAD_FULL_RES_TILE_BUFFER_SIZE;
 		} else {
-			if (setup->color_read &&
-			    !(args->color_read.flags &
-			      VC4_SUBMIT_RCL_SURFACE_READ_IS_FULL_RES)) {
-				loop_body_size += VC4_PACKET_TILE_COORDINATES_SIZE;
-				loop_body_size += VC4_PACKET_STORE_TILE_BUFFER_GENERAL_SIZE;
-			}
 			loop_body_size += VC4_PACKET_LOAD_TILE_BUFFER_GENERAL_SIZE;
 		}
 	}
diff --git a/drivers/gpu/drm/vc4/vc4_txp.c b/drivers/gpu/drm/vc4/vc4_txp.c
index aa279b5b0de7..c8b89a78f9f4 100644
--- a/drivers/gpu/drm/vc4/vc4_txp.c
+++ b/drivers/gpu/drm/vc4/vc4_txp.c
@@ -148,6 +148,7 @@ struct vc4_txp {
 	struct drm_writeback_connector connector;
 
 	void __iomem *regs;
+	struct debugfs_regset32 regset;
 };
 
 static inline struct vc4_txp *encoder_to_vc4_txp(struct drm_encoder *encoder)
@@ -160,40 +161,14 @@ static inline struct vc4_txp *connector_to_vc4_txp(struct drm_connector *conn)
 	return container_of(conn, struct vc4_txp, connector.base);
 }
 
-#define TXP_REG(reg) { reg, #reg }
-static const struct {
-	u32 reg;
-	const char *name;
-} txp_regs[] = {
-	TXP_REG(TXP_DST_PTR),
-	TXP_REG(TXP_DST_PITCH),
-	TXP_REG(TXP_DIM),
-	TXP_REG(TXP_DST_CTRL),
-	TXP_REG(TXP_PROGRESS),
+static const struct debugfs_reg32 txp_regs[] = {
+	VC4_REG32(TXP_DST_PTR),
+	VC4_REG32(TXP_DST_PITCH),
+	VC4_REG32(TXP_DIM),
+	VC4_REG32(TXP_DST_CTRL),
+	VC4_REG32(TXP_PROGRESS),
 };
 
-#ifdef CONFIG_DEBUG_FS
-int vc4_txp_debugfs_regs(struct seq_file *m, void *unused)
-{
-	struct drm_info_node *node = (struct drm_info_node *)m->private;
-	struct drm_device *dev = node->minor->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	struct vc4_txp *txp = vc4->txp;
-	int i;
-
-	if (!txp)
-		return 0;
-
-	for (i = 0; i < ARRAY_SIZE(txp_regs); i++) {
-		seq_printf(m, "%s (0x%04x): 0x%08x\n",
-			   txp_regs[i].name, txp_regs[i].reg,
-			   TXP_READ(txp_regs[i].reg));
-	}
-
-	return 0;
-}
-#endif
-
 static int vc4_txp_connector_get_modes(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
@@ -249,7 +224,6 @@ static int vc4_txp_connector_atomic_check(struct drm_connector *conn,
 					struct drm_connector_state *conn_state)
 {
 	struct drm_crtc_state *crtc_state;
-	struct drm_gem_cma_object *gem;
 	struct drm_framebuffer *fb;
 	int i;
 
@@ -275,8 +249,6 @@ static int vc4_txp_connector_atomic_check(struct drm_connector *conn,
 	if (i == ARRAY_SIZE(drm_fmts))
 		return -EINVAL;
 
-	gem = drm_fb_cma_get_gem_obj(fb, 0);
-
 	/* Pitch must be aligned on 16 bytes. */
 	if (fb->pitches[0] & GENMASK(3, 0))
 		return -EINVAL;
@@ -327,7 +299,7 @@ static void vc4_txp_connector_atomic_commit(struct drm_connector *conn,
 
 	TXP_WRITE(TXP_DST_CTRL, ctrl);
 
-	drm_writeback_queue_job(&txp->connector, conn_state->writeback_job);
+	drm_writeback_queue_job(&txp->connector, conn_state);
 }
 
 static const struct drm_connector_helper_funcs vc4_txp_connector_helper_funcs = {
@@ -413,6 +385,9 @@ static int vc4_txp_bind(struct device *dev, struct device *master, void *data)
 	txp->regs = vc4_ioremap_regs(pdev, 0);
 	if (IS_ERR(txp->regs))
 		return PTR_ERR(txp->regs);
+	txp->regset.base = txp->regs;
+	txp->regset.regs = txp_regs;
+	txp->regset.nregs = ARRAY_SIZE(txp_regs);
 
 	drm_connector_helper_add(&txp->connector.base,
 				 &vc4_txp_connector_helper_funcs);
@@ -431,6 +406,8 @@ static int vc4_txp_bind(struct device *dev, struct device *master, void *data)
 	dev_set_drvdata(dev, txp);
 	vc4->txp = txp;
 
+	vc4_debugfs_add_regset32(drm, "txp_regs", &txp->regset);
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/vc4/vc4_v3d.c b/drivers/gpu/drm/vc4/vc4_v3d.c
index e47e29426078..a4b6859e3af6 100644
--- a/drivers/gpu/drm/vc4/vc4_v3d.c
+++ b/drivers/gpu/drm/vc4/vc4_v3d.c
@@ -22,129 +22,145 @@
 #include "vc4_drv.h"
 #include "vc4_regs.h"
 
-#ifdef CONFIG_DEBUG_FS
-#define REGDEF(reg) { reg, #reg }
-static const struct {
-	uint32_t reg;
-	const char *name;
-} vc4_reg_defs[] = {
-	REGDEF(V3D_IDENT0),
-	REGDEF(V3D_IDENT1),
-	REGDEF(V3D_IDENT2),
-	REGDEF(V3D_SCRATCH),
-	REGDEF(V3D_L2CACTL),
-	REGDEF(V3D_SLCACTL),
-	REGDEF(V3D_INTCTL),
-	REGDEF(V3D_INTENA),
-	REGDEF(V3D_INTDIS),
-	REGDEF(V3D_CT0CS),
-	REGDEF(V3D_CT1CS),
-	REGDEF(V3D_CT0EA),
-	REGDEF(V3D_CT1EA),
-	REGDEF(V3D_CT0CA),
-	REGDEF(V3D_CT1CA),
-	REGDEF(V3D_CT00RA0),
-	REGDEF(V3D_CT01RA0),
-	REGDEF(V3D_CT0LC),
-	REGDEF(V3D_CT1LC),
-	REGDEF(V3D_CT0PC),
-	REGDEF(V3D_CT1PC),
-	REGDEF(V3D_PCS),
-	REGDEF(V3D_BFC),
-	REGDEF(V3D_RFC),
-	REGDEF(V3D_BPCA),
-	REGDEF(V3D_BPCS),
-	REGDEF(V3D_BPOA),
-	REGDEF(V3D_BPOS),
-	REGDEF(V3D_BXCF),
-	REGDEF(V3D_SQRSV0),
-	REGDEF(V3D_SQRSV1),
-	REGDEF(V3D_SQCNTL),
-	REGDEF(V3D_SRQPC),
-	REGDEF(V3D_SRQUA),
-	REGDEF(V3D_SRQUL),
-	REGDEF(V3D_SRQCS),
-	REGDEF(V3D_VPACNTL),
-	REGDEF(V3D_VPMBASE),
-	REGDEF(V3D_PCTRC),
-	REGDEF(V3D_PCTRE),
-	REGDEF(V3D_PCTR(0)),
-	REGDEF(V3D_PCTRS(0)),
-	REGDEF(V3D_PCTR(1)),
-	REGDEF(V3D_PCTRS(1)),
-	REGDEF(V3D_PCTR(2)),
-	REGDEF(V3D_PCTRS(2)),
-	REGDEF(V3D_PCTR(3)),
-	REGDEF(V3D_PCTRS(3)),
-	REGDEF(V3D_PCTR(4)),
-	REGDEF(V3D_PCTRS(4)),
-	REGDEF(V3D_PCTR(5)),
-	REGDEF(V3D_PCTRS(5)),
-	REGDEF(V3D_PCTR(6)),
-	REGDEF(V3D_PCTRS(6)),
-	REGDEF(V3D_PCTR(7)),
-	REGDEF(V3D_PCTRS(7)),
-	REGDEF(V3D_PCTR(8)),
-	REGDEF(V3D_PCTRS(8)),
-	REGDEF(V3D_PCTR(9)),
-	REGDEF(V3D_PCTRS(9)),
-	REGDEF(V3D_PCTR(10)),
-	REGDEF(V3D_PCTRS(10)),
-	REGDEF(V3D_PCTR(11)),
-	REGDEF(V3D_PCTRS(11)),
-	REGDEF(V3D_PCTR(12)),
-	REGDEF(V3D_PCTRS(12)),
-	REGDEF(V3D_PCTR(13)),
-	REGDEF(V3D_PCTRS(13)),
-	REGDEF(V3D_PCTR(14)),
-	REGDEF(V3D_PCTRS(14)),
-	REGDEF(V3D_PCTR(15)),
-	REGDEF(V3D_PCTRS(15)),
-	REGDEF(V3D_DBGE),
-	REGDEF(V3D_FDBGO),
-	REGDEF(V3D_FDBGB),
-	REGDEF(V3D_FDBGR),
-	REGDEF(V3D_FDBGS),
-	REGDEF(V3D_ERRSTAT),
+static const struct debugfs_reg32 v3d_regs[] = {
+	VC4_REG32(V3D_IDENT0),
+	VC4_REG32(V3D_IDENT1),
+	VC4_REG32(V3D_IDENT2),
+	VC4_REG32(V3D_SCRATCH),
+	VC4_REG32(V3D_L2CACTL),
+	VC4_REG32(V3D_SLCACTL),
+	VC4_REG32(V3D_INTCTL),
+	VC4_REG32(V3D_INTENA),
+	VC4_REG32(V3D_INTDIS),
+	VC4_REG32(V3D_CT0CS),
+	VC4_REG32(V3D_CT1CS),
+	VC4_REG32(V3D_CT0EA),
+	VC4_REG32(V3D_CT1EA),
+	VC4_REG32(V3D_CT0CA),
+	VC4_REG32(V3D_CT1CA),
+	VC4_REG32(V3D_CT00RA0),
+	VC4_REG32(V3D_CT01RA0),
+	VC4_REG32(V3D_CT0LC),
+	VC4_REG32(V3D_CT1LC),
+	VC4_REG32(V3D_CT0PC),
+	VC4_REG32(V3D_CT1PC),
+	VC4_REG32(V3D_PCS),
+	VC4_REG32(V3D_BFC),
+	VC4_REG32(V3D_RFC),
+	VC4_REG32(V3D_BPCA),
+	VC4_REG32(V3D_BPCS),
+	VC4_REG32(V3D_BPOA),
+	VC4_REG32(V3D_BPOS),
+	VC4_REG32(V3D_BXCF),
+	VC4_REG32(V3D_SQRSV0),
+	VC4_REG32(V3D_SQRSV1),
+	VC4_REG32(V3D_SQCNTL),
+	VC4_REG32(V3D_SRQPC),
+	VC4_REG32(V3D_SRQUA),
+	VC4_REG32(V3D_SRQUL),
+	VC4_REG32(V3D_SRQCS),
+	VC4_REG32(V3D_VPACNTL),
+	VC4_REG32(V3D_VPMBASE),
+	VC4_REG32(V3D_PCTRC),
+	VC4_REG32(V3D_PCTRE),
+	VC4_REG32(V3D_PCTR(0)),
+	VC4_REG32(V3D_PCTRS(0)),
+	VC4_REG32(V3D_PCTR(1)),
+	VC4_REG32(V3D_PCTRS(1)),
+	VC4_REG32(V3D_PCTR(2)),
+	VC4_REG32(V3D_PCTRS(2)),
+	VC4_REG32(V3D_PCTR(3)),
+	VC4_REG32(V3D_PCTRS(3)),
+	VC4_REG32(V3D_PCTR(4)),
+	VC4_REG32(V3D_PCTRS(4)),
+	VC4_REG32(V3D_PCTR(5)),
+	VC4_REG32(V3D_PCTRS(5)),
+	VC4_REG32(V3D_PCTR(6)),
+	VC4_REG32(V3D_PCTRS(6)),
+	VC4_REG32(V3D_PCTR(7)),
+	VC4_REG32(V3D_PCTRS(7)),
+	VC4_REG32(V3D_PCTR(8)),
+	VC4_REG32(V3D_PCTRS(8)),
+	VC4_REG32(V3D_PCTR(9)),
+	VC4_REG32(V3D_PCTRS(9)),
+	VC4_REG32(V3D_PCTR(10)),
+	VC4_REG32(V3D_PCTRS(10)),
+	VC4_REG32(V3D_PCTR(11)),
+	VC4_REG32(V3D_PCTRS(11)),
+	VC4_REG32(V3D_PCTR(12)),
+	VC4_REG32(V3D_PCTRS(12)),
+	VC4_REG32(V3D_PCTR(13)),
+	VC4_REG32(V3D_PCTRS(13)),
+	VC4_REG32(V3D_PCTR(14)),
+	VC4_REG32(V3D_PCTRS(14)),
+	VC4_REG32(V3D_PCTR(15)),
+	VC4_REG32(V3D_PCTRS(15)),
+	VC4_REG32(V3D_DBGE),
+	VC4_REG32(V3D_FDBGO),
+	VC4_REG32(V3D_FDBGB),
+	VC4_REG32(V3D_FDBGR),
+	VC4_REG32(V3D_FDBGS),
+	VC4_REG32(V3D_ERRSTAT),
 };
 
-int vc4_v3d_debugfs_regs(struct seq_file *m, void *unused)
+static int vc4_v3d_debugfs_ident(struct seq_file *m, void *unused)
 {
 	struct drm_info_node *node = (struct drm_info_node *)m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(vc4_reg_defs); i++) {
-		seq_printf(m, "%s (0x%04x): 0x%08x\n",
-			   vc4_reg_defs[i].name, vc4_reg_defs[i].reg,
-			   V3D_READ(vc4_reg_defs[i].reg));
+	int ret = vc4_v3d_pm_get(vc4);
+
+	if (ret == 0) {
+		uint32_t ident1 = V3D_READ(V3D_IDENT1);
+		uint32_t nslc = VC4_GET_FIELD(ident1, V3D_IDENT1_NSLC);
+		uint32_t tups = VC4_GET_FIELD(ident1, V3D_IDENT1_TUPS);
+		uint32_t qups = VC4_GET_FIELD(ident1, V3D_IDENT1_QUPS);
+
+		seq_printf(m, "Revision:   %d\n",
+			   VC4_GET_FIELD(ident1, V3D_IDENT1_REV));
+		seq_printf(m, "Slices:     %d\n", nslc);
+		seq_printf(m, "TMUs:       %d\n", nslc * tups);
+		seq_printf(m, "QPUs:       %d\n", nslc * qups);
+		seq_printf(m, "Semaphores: %d\n",
+			   VC4_GET_FIELD(ident1, V3D_IDENT1_NSEM));
+		vc4_v3d_pm_put(vc4);
 	}
 
 	return 0;
 }
 
-int vc4_v3d_debugfs_ident(struct seq_file *m, void *unused)
+/**
+ * Wraps pm_runtime_get_sync() in a refcount, so that we can reliably
+ * get the pm_runtime refcount to 0 in vc4_reset().
+ */
+int
+vc4_v3d_pm_get(struct vc4_dev *vc4)
 {
-	struct drm_info_node *node = (struct drm_info_node *)m->private;
-	struct drm_device *dev = node->minor->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	uint32_t ident1 = V3D_READ(V3D_IDENT1);
-	uint32_t nslc = VC4_GET_FIELD(ident1, V3D_IDENT1_NSLC);
-	uint32_t tups = VC4_GET_FIELD(ident1, V3D_IDENT1_TUPS);
-	uint32_t qups = VC4_GET_FIELD(ident1, V3D_IDENT1_QUPS);
-
-	seq_printf(m, "Revision:   %d\n",
-		   VC4_GET_FIELD(ident1, V3D_IDENT1_REV));
-	seq_printf(m, "Slices:     %d\n", nslc);
-	seq_printf(m, "TMUs:       %d\n", nslc * tups);
-	seq_printf(m, "QPUs:       %d\n", nslc * qups);
-	seq_printf(m, "Semaphores: %d\n",
-		   VC4_GET_FIELD(ident1, V3D_IDENT1_NSEM));
+	mutex_lock(&vc4->power_lock);
+	if (vc4->power_refcount++ == 0) {
+		int ret = pm_runtime_get_sync(&vc4->v3d->pdev->dev);
+
+		if (ret < 0) {
+			vc4->power_refcount--;
+			mutex_unlock(&vc4->power_lock);
+			return ret;
+		}
+	}
+	mutex_unlock(&vc4->power_lock);
 
 	return 0;
 }
-#endif /* CONFIG_DEBUG_FS */
+
+void
+vc4_v3d_pm_put(struct vc4_dev *vc4)
+{
+	mutex_lock(&vc4->power_lock);
+	if (--vc4->power_refcount == 0) {
+		pm_runtime_mark_last_busy(&vc4->v3d->pdev->dev);
+		pm_runtime_put_autosuspend(&vc4->v3d->pdev->dev);
+	}
+	mutex_unlock(&vc4->power_lock);
+}
 
 static void vc4_v3d_init_hw(struct drm_device *dev)
 {
@@ -354,6 +370,9 @@ static int vc4_v3d_bind(struct device *dev, struct device *master, void *data)
 	v3d->regs = vc4_ioremap_regs(pdev, 0);
 	if (IS_ERR(v3d->regs))
 		return PTR_ERR(v3d->regs);
+	v3d->regset.base = v3d->regs;
+	v3d->regset.regs = v3d_regs;
+	v3d->regset.nregs = ARRAY_SIZE(v3d_regs);
 
 	vc4->v3d = v3d;
 	v3d->vc4 = vc4;
@@ -409,6 +428,9 @@ static int vc4_v3d_bind(struct device *dev, struct device *master, void *data)
 	pm_runtime_set_autosuspend_delay(dev, 40); /* a little over 2 frames. */
 	pm_runtime_enable(dev);
 
+	vc4_debugfs_add_file(drm, "v3d_ident", vc4_v3d_debugfs_ident, NULL);
+	vc4_debugfs_add_regset32(drm, "v3d_regs", &v3d->regset);
+
 	return 0;
 }
 
@@ -452,7 +474,7 @@ static int vc4_v3d_dev_remove(struct platform_device *pdev)
 	return 0;
 }
 
-static const struct of_device_id vc4_v3d_dt_match[] = {
+const struct of_device_id vc4_v3d_dt_match[] = {
 	{ .compatible = "brcm,bcm2835-v3d" },
 	{ .compatible = "brcm,cygnus-v3d" },
 	{ .compatible = "brcm,vc4-v3d" },
diff --git a/drivers/gpu/drm/vc4/vc4_vec.c b/drivers/gpu/drm/vc4/vc4_vec.c
index 858c3a483229..0a27e48fab31 100644
--- a/drivers/gpu/drm/vc4/vc4_vec.c
+++ b/drivers/gpu/drm/vc4/vc4_vec.c
@@ -176,6 +176,8 @@ struct vc4_vec {
 	struct clk *clock;
 
 	const struct vc4_vec_tv_mode *tv_mode;
+
+	struct debugfs_regset32 regset;
 };
 
 #define VEC_READ(offset) readl(vec->regs + (offset))
@@ -223,59 +225,33 @@ struct vc4_vec_tv_mode {
 	void (*mode_set)(struct vc4_vec *vec);
 };
 
-#define VEC_REG(reg) { reg, #reg }
-static const struct {
-	u32 reg;
-	const char *name;
-} vec_regs[] = {
-	VEC_REG(VEC_WSE_CONTROL),
-	VEC_REG(VEC_WSE_WSS_DATA),
-	VEC_REG(VEC_WSE_VPS_DATA1),
-	VEC_REG(VEC_WSE_VPS_CONTROL),
-	VEC_REG(VEC_REVID),
-	VEC_REG(VEC_CONFIG0),
-	VEC_REG(VEC_SCHPH),
-	VEC_REG(VEC_CLMP0_START),
-	VEC_REG(VEC_CLMP0_END),
-	VEC_REG(VEC_FREQ3_2),
-	VEC_REG(VEC_FREQ1_0),
-	VEC_REG(VEC_CONFIG1),
-	VEC_REG(VEC_CONFIG2),
-	VEC_REG(VEC_INTERRUPT_CONTROL),
-	VEC_REG(VEC_INTERRUPT_STATUS),
-	VEC_REG(VEC_FCW_SECAM_B),
-	VEC_REG(VEC_SECAM_GAIN_VAL),
-	VEC_REG(VEC_CONFIG3),
-	VEC_REG(VEC_STATUS0),
-	VEC_REG(VEC_MASK0),
-	VEC_REG(VEC_CFG),
-	VEC_REG(VEC_DAC_TEST),
-	VEC_REG(VEC_DAC_CONFIG),
-	VEC_REG(VEC_DAC_MISC),
+static const struct debugfs_reg32 vec_regs[] = {
+	VC4_REG32(VEC_WSE_CONTROL),
+	VC4_REG32(VEC_WSE_WSS_DATA),
+	VC4_REG32(VEC_WSE_VPS_DATA1),
+	VC4_REG32(VEC_WSE_VPS_CONTROL),
+	VC4_REG32(VEC_REVID),
+	VC4_REG32(VEC_CONFIG0),
+	VC4_REG32(VEC_SCHPH),
+	VC4_REG32(VEC_CLMP0_START),
+	VC4_REG32(VEC_CLMP0_END),
+	VC4_REG32(VEC_FREQ3_2),
+	VC4_REG32(VEC_FREQ1_0),
+	VC4_REG32(VEC_CONFIG1),
+	VC4_REG32(VEC_CONFIG2),
+	VC4_REG32(VEC_INTERRUPT_CONTROL),
+	VC4_REG32(VEC_INTERRUPT_STATUS),
+	VC4_REG32(VEC_FCW_SECAM_B),
+	VC4_REG32(VEC_SECAM_GAIN_VAL),
+	VC4_REG32(VEC_CONFIG3),
+	VC4_REG32(VEC_STATUS0),
+	VC4_REG32(VEC_MASK0),
+	VC4_REG32(VEC_CFG),
+	VC4_REG32(VEC_DAC_TEST),
+	VC4_REG32(VEC_DAC_CONFIG),
+	VC4_REG32(VEC_DAC_MISC),
 };
 
-#ifdef CONFIG_DEBUG_FS
-int vc4_vec_debugfs_regs(struct seq_file *m, void *unused)
-{
-	struct drm_info_node *node = (struct drm_info_node *)m->private;
-	struct drm_device *dev = node->minor->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	struct vc4_vec *vec = vc4->vec;
-	int i;
-
-	if (!vec)
-		return 0;
-
-	for (i = 0; i < ARRAY_SIZE(vec_regs); i++) {
-		seq_printf(m, "%s (0x%04x): 0x%08x\n",
-			   vec_regs[i].name, vec_regs[i].reg,
-			   VEC_READ(vec_regs[i].reg));
-	}
-
-	return 0;
-}
-#endif
-
 static void vc4_vec_ntsc_mode_set(struct vc4_vec *vec)
 {
 	VEC_WRITE(VEC_CONFIG0, VEC_CONFIG0_NTSC_STD | VEC_CONFIG0_PDEN);
@@ -587,6 +563,9 @@ static int vc4_vec_bind(struct device *dev, struct device *master, void *data)
 	vec->regs = vc4_ioremap_regs(pdev, 0);
 	if (IS_ERR(vec->regs))
 		return PTR_ERR(vec->regs);
+	vec->regset.base = vec->regs;
+	vec->regset.regs = vec_regs;
+	vec->regset.nregs = ARRAY_SIZE(vec_regs);
 
 	vec->clock = devm_clk_get(dev, NULL);
 	if (IS_ERR(vec->clock)) {
@@ -612,6 +591,8 @@ static int vc4_vec_bind(struct device *dev, struct device *master, void *data)
 
 	vc4->vec = vec;
 
+	vc4_debugfs_add_regset32(drm, "vec_regs", &vec->regset);
+
 	return 0;
 
 err_destroy_encoder:
diff --git a/drivers/gpu/drm/virtio/virtgpu_debugfs.c b/drivers/gpu/drm/virtio/virtgpu_debugfs.c
index 73dc99046c43..ed0fcda713c3 100644
--- a/drivers/gpu/drm/virtio/virtgpu_debugfs.c
+++ b/drivers/gpu/drm/virtio/virtgpu_debugfs.c
@@ -28,6 +28,30 @@
 
 #include "virtgpu_drv.h"
 
+static void virtio_add_bool(struct seq_file *m, const char *name,
+				    bool value)
+{
+	seq_printf(m, "%-16s : %s\n", name, value ? "yes" : "no");
+}
+
+static void virtio_add_int(struct seq_file *m, const char *name,
+				   int value)
+{
+	seq_printf(m, "%-16s : %d\n", name, value);
+}
+
+static int virtio_gpu_features(struct seq_file *m, void *data)
+{
+	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct virtio_gpu_device *vgdev = node->minor->dev->dev_private;
+
+	virtio_add_bool(m, "virgl", vgdev->has_virgl_3d);
+	virtio_add_bool(m, "edid", vgdev->has_edid);
+	virtio_add_int(m, "cap sets", vgdev->num_capsets);
+	virtio_add_int(m, "scanouts", vgdev->num_scanouts);
+	return 0;
+}
+
 static int
 virtio_gpu_debugfs_irq_info(struct seq_file *m, void *data)
 {
@@ -41,7 +65,8 @@ virtio_gpu_debugfs_irq_info(struct seq_file *m, void *data)
 }
 
 static struct drm_info_list virtio_gpu_debugfs_list[] = {
-	{ "irq_fence", virtio_gpu_debugfs_irq_info, 0, NULL },
+	{ "virtio-gpu-features", virtio_gpu_features },
+	{ "virtio-gpu-irq-fence", virtio_gpu_debugfs_irq_info, 0, NULL },
 };
 
 #define VIRTIO_GPU_DEBUGFS_ENTRIES ARRAY_SIZE(virtio_gpu_debugfs_list)
diff --git a/drivers/gpu/drm/virtio/virtgpu_display.c b/drivers/gpu/drm/virtio/virtgpu_display.c
index 653ec7d0bf4d..86843a4d6102 100644
--- a/drivers/gpu/drm/virtio/virtgpu_display.c
+++ b/drivers/gpu/drm/virtio/virtgpu_display.c
@@ -385,5 +385,6 @@ void virtio_gpu_modeset_fini(struct virtio_gpu_device *vgdev)
 
 	for (i = 0 ; i < vgdev->num_scanouts; ++i)
 		kfree(vgdev->outputs[i].edid);
+	drm_atomic_helper_shutdown(vgdev->ddev);
 	drm_mode_config_cleanup(vgdev->ddev);
 }
diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.c b/drivers/gpu/drm/virtio/virtgpu_drv.c
index af92964b6889..c50868753132 100644
--- a/drivers/gpu/drm/virtio/virtgpu_drv.c
+++ b/drivers/gpu/drm/virtio/virtgpu_drv.c
@@ -209,8 +209,6 @@ static struct drm_driver driver = {
 	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
 	.gem_prime_export = drm_gem_prime_export,
 	.gem_prime_import = drm_gem_prime_import,
-	.gem_prime_pin = virtgpu_gem_prime_pin,
-	.gem_prime_unpin = virtgpu_gem_prime_unpin,
 	.gem_prime_get_sg_table = virtgpu_gem_prime_get_sg_table,
 	.gem_prime_import_sg_table = virtgpu_gem_prime_import_sg_table,
 	.gem_prime_vmap = virtgpu_gem_prime_vmap,
diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.h b/drivers/gpu/drm/virtio/virtgpu_drv.h
index d577cb76f5ad..b69ae10ca238 100644
--- a/drivers/gpu/drm/virtio/virtgpu_drv.h
+++ b/drivers/gpu/drm/virtio/virtgpu_drv.h
@@ -50,6 +50,23 @@
 #define DRIVER_MINOR 1
 #define DRIVER_PATCHLEVEL 0
 
+struct virtio_gpu_object_params {
+	uint32_t format;
+	uint32_t width;
+	uint32_t height;
+	unsigned long size;
+	bool dumb;
+	/* 3d */
+	bool virgl;
+	uint32_t target;
+	uint32_t bind;
+	uint32_t depth;
+	uint32_t array_size;
+	uint32_t last_level;
+	uint32_t nr_samples;
+	uint32_t flags;
+};
+
 struct virtio_gpu_object {
 	struct drm_gem_object gem_base;
 	uint32_t hw_res_handle;
@@ -204,6 +221,9 @@ struct virtio_gpu_fpriv {
 /* virtio_ioctl.c */
 #define DRM_VIRTIO_NUM_IOCTLS 10
 extern struct drm_ioctl_desc virtio_gpu_ioctls[DRM_VIRTIO_NUM_IOCTLS];
+int virtio_gpu_object_list_validate(struct ww_acquire_ctx *ticket,
+				    struct list_head *head);
+void virtio_gpu_unref_list(struct list_head *head);
 
 /* virtio_kms.c */
 int virtio_gpu_init(struct drm_device *dev);
@@ -217,16 +237,17 @@ int virtio_gpu_gem_init(struct virtio_gpu_device *vgdev);
 void virtio_gpu_gem_fini(struct virtio_gpu_device *vgdev);
 int virtio_gpu_gem_create(struct drm_file *file,
 			  struct drm_device *dev,
-			  uint64_t size,
+			  struct virtio_gpu_object_params *params,
 			  struct drm_gem_object **obj_p,
 			  uint32_t *handle_p);
 int virtio_gpu_gem_object_open(struct drm_gem_object *obj,
 			       struct drm_file *file);
 void virtio_gpu_gem_object_close(struct drm_gem_object *obj,
 				 struct drm_file *file);
-struct virtio_gpu_object *virtio_gpu_alloc_object(struct drm_device *dev,
-						  size_t size, bool kernel,
-						  bool pinned);
+struct virtio_gpu_object*
+virtio_gpu_alloc_object(struct drm_device *dev,
+			struct virtio_gpu_object_params *params,
+			struct virtio_gpu_fence *fence);
 int virtio_gpu_mode_dumb_create(struct drm_file *file_priv,
 				struct drm_device *dev,
 				struct drm_mode_create_dumb *args);
@@ -243,9 +264,8 @@ int virtio_gpu_alloc_vbufs(struct virtio_gpu_device *vgdev);
 void virtio_gpu_free_vbufs(struct virtio_gpu_device *vgdev);
 void virtio_gpu_cmd_create_resource(struct virtio_gpu_device *vgdev,
 				    struct virtio_gpu_object *bo,
-				    uint32_t format,
-				    uint32_t width,
-				    uint32_t height);
+				    struct virtio_gpu_object_params *params,
+				    struct virtio_gpu_fence *fence);
 void virtio_gpu_cmd_unref_resource(struct virtio_gpu_device *vgdev,
 				   uint32_t resource_id);
 void virtio_gpu_cmd_transfer_to_host_2d(struct virtio_gpu_device *vgdev,
@@ -304,7 +324,8 @@ void virtio_gpu_cmd_transfer_to_host_3d(struct virtio_gpu_device *vgdev,
 void
 virtio_gpu_cmd_resource_create_3d(struct virtio_gpu_device *vgdev,
 				  struct virtio_gpu_object *bo,
-				  struct virtio_gpu_resource_create_3d *rc_3d);
+				  struct virtio_gpu_object_params *params,
+				  struct virtio_gpu_fence *fence);
 void virtio_gpu_ctrl_ack(struct virtqueue *vq);
 void virtio_gpu_cursor_ack(struct virtqueue *vq);
 void virtio_gpu_fence_ack(struct virtqueue *vq);
@@ -332,6 +353,7 @@ void virtio_gpu_ttm_fini(struct virtio_gpu_device *vgdev);
 int virtio_gpu_mmap(struct file *filp, struct vm_area_struct *vma);
 
 /* virtio_gpu_fence.c */
+bool virtio_fence_signaled(struct dma_fence *f);
 struct virtio_gpu_fence *virtio_gpu_fence_alloc(
 	struct virtio_gpu_device *vgdev);
 int virtio_gpu_fence_emit(struct virtio_gpu_device *vgdev,
@@ -342,8 +364,9 @@ void virtio_gpu_fence_event_process(struct virtio_gpu_device *vdev,
 
 /* virtio_gpu_object */
 int virtio_gpu_object_create(struct virtio_gpu_device *vgdev,
-			     unsigned long size, bool kernel, bool pinned,
-			     struct virtio_gpu_object **bo_ptr);
+			     struct virtio_gpu_object_params *params,
+			     struct virtio_gpu_object **bo_ptr,
+			     struct virtio_gpu_fence *fence);
 void virtio_gpu_object_kunmap(struct virtio_gpu_object *bo);
 int virtio_gpu_object_kmap(struct virtio_gpu_object *bo);
 int virtio_gpu_object_get_sg_table(struct virtio_gpu_device *qdev,
@@ -352,8 +375,6 @@ void virtio_gpu_object_free_sg_table(struct virtio_gpu_object *bo);
 int virtio_gpu_object_wait(struct virtio_gpu_object *bo, bool no_wait);
 
 /* virtgpu_prime.c */
-int virtgpu_gem_prime_pin(struct drm_gem_object *obj);
-void virtgpu_gem_prime_unpin(struct drm_gem_object *obj);
 struct sg_table *virtgpu_gem_prime_get_sg_table(struct drm_gem_object *obj);
 struct drm_gem_object *virtgpu_gem_prime_import_sg_table(
 	struct drm_device *dev, struct dma_buf_attachment *attach,
diff --git a/drivers/gpu/drm/virtio/virtgpu_fence.c b/drivers/gpu/drm/virtio/virtgpu_fence.c
index 21bd4c4a32d1..87d1966192f4 100644
--- a/drivers/gpu/drm/virtio/virtgpu_fence.c
+++ b/drivers/gpu/drm/virtio/virtgpu_fence.c
@@ -36,7 +36,7 @@ static const char *virtio_get_timeline_name(struct dma_fence *f)
 	return "controlq";
 }
 
-static bool virtio_signaled(struct dma_fence *f)
+bool virtio_fence_signaled(struct dma_fence *f)
 {
 	struct virtio_gpu_fence *fence = to_virtio_fence(f);
 
@@ -62,7 +62,7 @@ static void virtio_timeline_value_str(struct dma_fence *f, char *str, int size)
 static const struct dma_fence_ops virtio_fence_ops = {
 	.get_driver_name     = virtio_get_driver_name,
 	.get_timeline_name   = virtio_get_timeline_name,
-	.signaled            = virtio_signaled,
+	.signaled            = virtio_fence_signaled,
 	.fence_value_str     = virtio_fence_value_str,
 	.timeline_value_str  = virtio_timeline_value_str,
 };
diff --git a/drivers/gpu/drm/virtio/virtgpu_gem.c b/drivers/gpu/drm/virtio/virtgpu_gem.c
index f06586393974..1e49e08dd545 100644
--- a/drivers/gpu/drm/virtio/virtgpu_gem.c
+++ b/drivers/gpu/drm/virtio/virtgpu_gem.c
@@ -34,15 +34,16 @@ void virtio_gpu_gem_free_object(struct drm_gem_object *gem_obj)
 		virtio_gpu_object_unref(&obj);
 }
 
-struct virtio_gpu_object *virtio_gpu_alloc_object(struct drm_device *dev,
-						  size_t size, bool kernel,
-						  bool pinned)
+struct virtio_gpu_object*
+virtio_gpu_alloc_object(struct drm_device *dev,
+			struct virtio_gpu_object_params *params,
+			struct virtio_gpu_fence *fence)
 {
 	struct virtio_gpu_device *vgdev = dev->dev_private;
 	struct virtio_gpu_object *obj;
 	int ret;
 
-	ret = virtio_gpu_object_create(vgdev, size, kernel, pinned, &obj);
+	ret = virtio_gpu_object_create(vgdev, params, &obj, fence);
 	if (ret)
 		return ERR_PTR(ret);
 
@@ -51,7 +52,7 @@ struct virtio_gpu_object *virtio_gpu_alloc_object(struct drm_device *dev,
 
 int virtio_gpu_gem_create(struct drm_file *file,
 			  struct drm_device *dev,
-			  uint64_t size,
+			  struct virtio_gpu_object_params *params,
 			  struct drm_gem_object **obj_p,
 			  uint32_t *handle_p)
 {
@@ -59,7 +60,7 @@ int virtio_gpu_gem_create(struct drm_file *file,
 	int ret;
 	u32 handle;
 
-	obj = virtio_gpu_alloc_object(dev, size, false, false);
+	obj = virtio_gpu_alloc_object(dev, params, NULL);
 	if (IS_ERR(obj))
 		return PTR_ERR(obj);
 
@@ -82,12 +83,10 @@ int virtio_gpu_mode_dumb_create(struct drm_file *file_priv,
 				struct drm_device *dev,
 				struct drm_mode_create_dumb *args)
 {
-	struct virtio_gpu_device *vgdev = dev->dev_private;
 	struct drm_gem_object *gobj;
-	struct virtio_gpu_object *obj;
+	struct virtio_gpu_object_params params = { 0 };
 	int ret;
 	uint32_t pitch;
-	uint32_t format;
 
 	if (args->bpp != 32)
 		return -EINVAL;
@@ -96,22 +95,16 @@ int virtio_gpu_mode_dumb_create(struct drm_file *file_priv,
 	args->size = pitch * args->height;
 	args->size = ALIGN(args->size, PAGE_SIZE);
 
-	ret = virtio_gpu_gem_create(file_priv, dev, args->size, &gobj,
+	params.format = virtio_gpu_translate_format(DRM_FORMAT_HOST_XRGB8888);
+	params.width = args->width;
+	params.height = args->height;
+	params.size = args->size;
+	params.dumb = true;
+	ret = virtio_gpu_gem_create(file_priv, dev, &params, &gobj,
 				    &args->handle);
 	if (ret)
 		goto fail;
 
-	format = virtio_gpu_translate_format(DRM_FORMAT_HOST_XRGB8888);
-	obj = gem_to_virtio_gpu_obj(gobj);
-	virtio_gpu_cmd_create_resource(vgdev, obj, format,
-				       args->width, args->height);
-
-	/* attach the object to the resource */
-	ret = virtio_gpu_object_attach(vgdev, obj, NULL);
-	if (ret)
-		goto fail;
-
-	obj->dumb = true;
 	args->pitch = pitch;
 	return ret;
 
diff --git a/drivers/gpu/drm/virtio/virtgpu_ioctl.c b/drivers/gpu/drm/virtio/virtgpu_ioctl.c
index 14ce8188c052..949a264985fc 100644
--- a/drivers/gpu/drm/virtio/virtgpu_ioctl.c
+++ b/drivers/gpu/drm/virtio/virtgpu_ioctl.c
@@ -54,8 +54,8 @@ static int virtio_gpu_map_ioctl(struct drm_device *dev, void *data,
 					 &virtio_gpu_map->offset);
 }
 
-static int virtio_gpu_object_list_validate(struct ww_acquire_ctx *ticket,
-					   struct list_head *head)
+int virtio_gpu_object_list_validate(struct ww_acquire_ctx *ticket,
+				    struct list_head *head)
 {
 	struct ttm_operation_ctx ctx = { false, false };
 	struct ttm_validate_buffer *buf;
@@ -79,7 +79,7 @@ static int virtio_gpu_object_list_validate(struct ww_acquire_ctx *ticket,
 	return 0;
 }
 
-static void virtio_gpu_unref_list(struct list_head *head)
+void virtio_gpu_unref_list(struct list_head *head)
 {
 	struct ttm_validate_buffer *buf;
 	struct ttm_buffer_object *bo;
@@ -275,16 +275,12 @@ static int virtio_gpu_resource_create_ioctl(struct drm_device *dev, void *data,
 {
 	struct virtio_gpu_device *vgdev = dev->dev_private;
 	struct drm_virtgpu_resource_create *rc = data;
+	struct virtio_gpu_fence *fence;
 	int ret;
 	struct virtio_gpu_object *qobj;
 	struct drm_gem_object *obj;
 	uint32_t handle = 0;
-	uint32_t size;
-	struct list_head validate_list;
-	struct ttm_validate_buffer mainbuf;
-	struct virtio_gpu_fence *fence = NULL;
-	struct ww_acquire_ctx ticket;
-	struct virtio_gpu_resource_create_3d rc_3d;
+	struct virtio_gpu_object_params params = { 0 };
 
 	if (vgdev->has_virgl_3d == false) {
 		if (rc->depth > 1)
@@ -299,94 +295,43 @@ static int virtio_gpu_resource_create_ioctl(struct drm_device *dev, void *data,
 			return -EINVAL;
 	}
 
-	INIT_LIST_HEAD(&validate_list);
-	memset(&mainbuf, 0, sizeof(struct ttm_validate_buffer));
-
-	size = rc->size;
-
+	params.format = rc->format;
+	params.width = rc->width;
+	params.height = rc->height;
+	params.size = rc->size;
+	if (vgdev->has_virgl_3d) {
+		params.virgl = true;
+		params.target = rc->target;
+		params.bind = rc->bind;
+		params.depth = rc->depth;
+		params.array_size = rc->array_size;
+		params.last_level = rc->last_level;
+		params.nr_samples = rc->nr_samples;
+		params.flags = rc->flags;
+	}
 	/* allocate a single page size object */
-	if (size == 0)
-		size = PAGE_SIZE;
+	if (params.size == 0)
+		params.size = PAGE_SIZE;
 
-	qobj = virtio_gpu_alloc_object(dev, size, false, false);
+	fence = virtio_gpu_fence_alloc(vgdev);
+	if (!fence)
+		return -ENOMEM;
+	qobj = virtio_gpu_alloc_object(dev, &params, fence);
+	dma_fence_put(&fence->f);
 	if (IS_ERR(qobj))
 		return PTR_ERR(qobj);
 	obj = &qobj->gem_base;
 
-	if (!vgdev->has_virgl_3d) {
-		virtio_gpu_cmd_create_resource(vgdev, qobj, rc->format,
-					       rc->width, rc->height);
-
-		ret = virtio_gpu_object_attach(vgdev, qobj, NULL);
-	} else {
-		/* use a gem reference since unref list undoes them */
-		drm_gem_object_get(&qobj->gem_base);
-		mainbuf.bo = &qobj->tbo;
-		list_add(&mainbuf.head, &validate_list);
-
-		ret = virtio_gpu_object_list_validate(&ticket, &validate_list);
-		if (ret) {
-			DRM_DEBUG("failed to validate\n");
-			goto fail_unref;
-		}
-
-		rc_3d.resource_id = cpu_to_le32(qobj->hw_res_handle);
-		rc_3d.target = cpu_to_le32(rc->target);
-		rc_3d.format = cpu_to_le32(rc->format);
-		rc_3d.bind = cpu_to_le32(rc->bind);
-		rc_3d.width = cpu_to_le32(rc->width);
-		rc_3d.height = cpu_to_le32(rc->height);
-		rc_3d.depth = cpu_to_le32(rc->depth);
-		rc_3d.array_size = cpu_to_le32(rc->array_size);
-		rc_3d.last_level = cpu_to_le32(rc->last_level);
-		rc_3d.nr_samples = cpu_to_le32(rc->nr_samples);
-		rc_3d.flags = cpu_to_le32(rc->flags);
-
-		fence = virtio_gpu_fence_alloc(vgdev);
-		if (!fence) {
-			ret = -ENOMEM;
-			goto fail_backoff;
-		}
-
-		virtio_gpu_cmd_resource_create_3d(vgdev, qobj, &rc_3d);
-		ret = virtio_gpu_object_attach(vgdev, qobj, fence);
-		if (ret) {
-			dma_fence_put(&fence->f);
-			goto fail_backoff;
-		}
-		ttm_eu_fence_buffer_objects(&ticket, &validate_list, &fence->f);
-	}
-
 	ret = drm_gem_handle_create(file_priv, obj, &handle);
 	if (ret) {
-
 		drm_gem_object_release(obj);
-		if (vgdev->has_virgl_3d) {
-			virtio_gpu_unref_list(&validate_list);
-			dma_fence_put(&fence->f);
-		}
 		return ret;
 	}
 	drm_gem_object_put_unlocked(obj);
 
 	rc->res_handle = qobj->hw_res_handle; /* similiar to a VM address */
 	rc->bo_handle = handle;
-
-	if (vgdev->has_virgl_3d) {
-		virtio_gpu_unref_list(&validate_list);
-		dma_fence_put(&fence->f);
-	}
 	return 0;
-fail_backoff:
-	ttm_eu_backoff_reservation(&ticket, &validate_list);
-fail_unref:
-	if (vgdev->has_virgl_3d) {
-		virtio_gpu_unref_list(&validate_list);
-		dma_fence_put(&fence->f);
-	}
-//fail_obj:
-//	drm_gem_object_handle_unreference_unlocked(obj);
-	return ret;
 }
 
 static int virtio_gpu_resource_info_ioctl(struct drm_device *dev, void *data,
diff --git a/drivers/gpu/drm/virtio/virtgpu_object.c b/drivers/gpu/drm/virtio/virtgpu_object.c
index e7e946035027..b2da31310d24 100644
--- a/drivers/gpu/drm/virtio/virtgpu_object.c
+++ b/drivers/gpu/drm/virtio/virtgpu_object.c
@@ -23,6 +23,8 @@
  * WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
  */
 
+#include <drm/ttm/ttm_execbuf_util.h>
+
 #include "virtgpu_drv.h"
 
 static int virtio_gpu_resource_id_get(struct virtio_gpu_device *vgdev,
@@ -74,39 +76,34 @@ static void virtio_gpu_ttm_bo_destroy(struct ttm_buffer_object *tbo)
 	kfree(bo);
 }
 
-static void virtio_gpu_init_ttm_placement(struct virtio_gpu_object *vgbo,
-					  bool pinned)
+static void virtio_gpu_init_ttm_placement(struct virtio_gpu_object *vgbo)
 {
 	u32 c = 1;
-	u32 pflag = pinned ? TTM_PL_FLAG_NO_EVICT : 0;
 
 	vgbo->placement.placement = &vgbo->placement_code;
 	vgbo->placement.busy_placement = &vgbo->placement_code;
 	vgbo->placement_code.fpfn = 0;
 	vgbo->placement_code.lpfn = 0;
 	vgbo->placement_code.flags =
-		TTM_PL_MASK_CACHING | TTM_PL_FLAG_TT | pflag;
+		TTM_PL_MASK_CACHING | TTM_PL_FLAG_TT |
+		TTM_PL_FLAG_NO_EVICT;
 	vgbo->placement.num_placement = c;
 	vgbo->placement.num_busy_placement = c;
 
 }
 
 int virtio_gpu_object_create(struct virtio_gpu_device *vgdev,
-			     unsigned long size, bool kernel, bool pinned,
-			     struct virtio_gpu_object **bo_ptr)
+			     struct virtio_gpu_object_params *params,
+			     struct virtio_gpu_object **bo_ptr,
+			     struct virtio_gpu_fence *fence)
 {
 	struct virtio_gpu_object *bo;
-	enum ttm_bo_type type;
 	size_t acc_size;
 	int ret;
 
-	if (kernel)
-		type = ttm_bo_type_kernel;
-	else
-		type = ttm_bo_type_device;
 	*bo_ptr = NULL;
 
-	acc_size = ttm_bo_dma_acc_size(&vgdev->mman.bdev, size,
+	acc_size = ttm_bo_dma_acc_size(&vgdev->mman.bdev, params->size,
 				       sizeof(struct virtio_gpu_object));
 
 	bo = kzalloc(sizeof(struct virtio_gpu_object), GFP_KERNEL);
@@ -117,23 +114,62 @@ int virtio_gpu_object_create(struct virtio_gpu_device *vgdev,
 		kfree(bo);
 		return ret;
 	}
-	size = roundup(size, PAGE_SIZE);
-	ret = drm_gem_object_init(vgdev->ddev, &bo->gem_base, size);
+	params->size = roundup(params->size, PAGE_SIZE);
+	ret = drm_gem_object_init(vgdev->ddev, &bo->gem_base, params->size);
 	if (ret != 0) {
 		virtio_gpu_resource_id_put(vgdev, bo->hw_res_handle);
 		kfree(bo);
 		return ret;
 	}
-	bo->dumb = false;
-	virtio_gpu_init_ttm_placement(bo, pinned);
+	bo->dumb = params->dumb;
+
+	if (params->virgl) {
+		virtio_gpu_cmd_resource_create_3d(vgdev, bo, params, fence);
+	} else {
+		virtio_gpu_cmd_create_resource(vgdev, bo, params, fence);
+	}
 
-	ret = ttm_bo_init(&vgdev->mman.bdev, &bo->tbo, size, type,
-			  &bo->placement, 0, !kernel, acc_size,
-			  NULL, NULL, &virtio_gpu_ttm_bo_destroy);
+	virtio_gpu_init_ttm_placement(bo);
+	ret = ttm_bo_init(&vgdev->mman.bdev, &bo->tbo, params->size,
+			  ttm_bo_type_device, &bo->placement, 0,
+			  true, acc_size, NULL, NULL,
+			  &virtio_gpu_ttm_bo_destroy);
 	/* ttm_bo_init failure will call the destroy */
 	if (ret != 0)
 		return ret;
 
+	if (fence) {
+		struct virtio_gpu_fence_driver *drv = &vgdev->fence_drv;
+		struct list_head validate_list;
+		struct ttm_validate_buffer mainbuf;
+		struct ww_acquire_ctx ticket;
+		unsigned long irq_flags;
+		bool signaled;
+
+		INIT_LIST_HEAD(&validate_list);
+		memset(&mainbuf, 0, sizeof(struct ttm_validate_buffer));
+
+		/* use a gem reference since unref list undoes them */
+		drm_gem_object_get(&bo->gem_base);
+		mainbuf.bo = &bo->tbo;
+		list_add(&mainbuf.head, &validate_list);
+
+		ret = virtio_gpu_object_list_validate(&ticket, &validate_list);
+		if (ret == 0) {
+			spin_lock_irqsave(&drv->lock, irq_flags);
+			signaled = virtio_fence_signaled(&fence->f);
+			if (!signaled)
+				/* virtio create command still in flight */
+				ttm_eu_fence_buffer_objects(&ticket, &validate_list,
+							    &fence->f);
+			spin_unlock_irqrestore(&drv->lock, irq_flags);
+			if (signaled)
+				/* virtio create command finished */
+				ttm_eu_backoff_reservation(&ticket, &validate_list);
+		}
+		virtio_gpu_unref_list(&validate_list);
+	}
+
 	*bo_ptr = bo;
 	return 0;
 }
diff --git a/drivers/gpu/drm/virtio/virtgpu_prime.c b/drivers/gpu/drm/virtio/virtgpu_prime.c
index eb51a78e1199..8fbf71bd0c5e 100644
--- a/drivers/gpu/drm/virtio/virtgpu_prime.c
+++ b/drivers/gpu/drm/virtio/virtgpu_prime.c
@@ -28,20 +28,16 @@
  * device that might share buffers with virtgpu
  */
 
-int virtgpu_gem_prime_pin(struct drm_gem_object *obj)
+struct sg_table *virtgpu_gem_prime_get_sg_table(struct drm_gem_object *obj)
 {
-	WARN_ONCE(1, "not implemented");
-	return -ENODEV;
-}
+	struct virtio_gpu_object *bo = gem_to_virtio_gpu_obj(obj);
 
-void virtgpu_gem_prime_unpin(struct drm_gem_object *obj)
-{
-	WARN_ONCE(1, "not implemented");
-}
+	if (!bo->tbo.ttm->pages || !bo->tbo.ttm->num_pages)
+		/* should not happen */
+		return ERR_PTR(-EINVAL);
 
-struct sg_table *virtgpu_gem_prime_get_sg_table(struct drm_gem_object *obj)
-{
-	return ERR_PTR(-ENODEV);
+	return drm_prime_pages_to_sg(bo->tbo.ttm->pages,
+				     bo->tbo.ttm->num_pages);
 }
 
 struct drm_gem_object *virtgpu_gem_prime_import_sg_table(
@@ -68,7 +64,10 @@ void virtgpu_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr)
 }
 
 int virtgpu_gem_prime_mmap(struct drm_gem_object *obj,
-		       struct vm_area_struct *area)
+			   struct vm_area_struct *vma)
 {
-	return -ENODEV;
+	struct virtio_gpu_object *bo = gem_to_virtio_gpu_obj(obj);
+
+	bo->gem_base.vma_node.vm_node.start = bo->tbo.vma_node.vm_node.start;
+	return drm_gem_prime_mmap(obj, vma);
 }
diff --git a/drivers/gpu/drm/virtio/virtgpu_ttm.c b/drivers/gpu/drm/virtio/virtgpu_ttm.c
index 4bfbf25fabff..300ef3a83538 100644
--- a/drivers/gpu/drm/virtio/virtgpu_ttm.c
+++ b/drivers/gpu/drm/virtio/virtgpu_ttm.c
@@ -37,8 +37,6 @@
 
 #include <linux/delay.h>
 
-#define DRM_FILE_PAGE_OFFSET (0x100000000ULL >> PAGE_SHIFT)
-
 static struct
 virtio_gpu_device *virtio_gpu_get_vgdev(struct ttm_bo_device *bdev)
 {
@@ -116,10 +114,6 @@ static const struct ttm_mem_type_manager_func virtio_gpu_bo_manager_func = {
 static int virtio_gpu_init_mem_type(struct ttm_bo_device *bdev, uint32_t type,
 				    struct ttm_mem_type_manager *man)
 {
-	struct virtio_gpu_device *vgdev;
-
-	vgdev = virtio_gpu_get_vgdev(bdev);
-
 	switch (type) {
 	case TTM_PL_SYSTEM:
 		/* System memory */
@@ -194,42 +188,45 @@ static void virtio_gpu_ttm_io_mem_free(struct ttm_bo_device *bdev,
  */
 struct virtio_gpu_ttm_tt {
 	struct ttm_dma_tt		ttm;
-	struct virtio_gpu_device	*vgdev;
-	u64				offset;
+	struct virtio_gpu_object        *obj;
 };
 
-static int virtio_gpu_ttm_backend_bind(struct ttm_tt *ttm,
-				       struct ttm_mem_reg *bo_mem)
+static int virtio_gpu_ttm_tt_bind(struct ttm_tt *ttm,
+				  struct ttm_mem_reg *bo_mem)
 {
-	struct virtio_gpu_ttm_tt *gtt = (void *)ttm;
+	struct virtio_gpu_ttm_tt *gtt =
+		container_of(ttm, struct virtio_gpu_ttm_tt, ttm.ttm);
+	struct virtio_gpu_device *vgdev =
+		virtio_gpu_get_vgdev(gtt->obj->tbo.bdev);
 
-	gtt->offset = (unsigned long)(bo_mem->start << PAGE_SHIFT);
-	if (!ttm->num_pages)
-		WARN(1, "nothing to bind %lu pages for mreg %p back %p!\n",
-		     ttm->num_pages, bo_mem, ttm);
-
-	/* Not implemented */
+	virtio_gpu_object_attach(vgdev, gtt->obj, NULL);
 	return 0;
 }
 
-static int virtio_gpu_ttm_backend_unbind(struct ttm_tt *ttm)
+static int virtio_gpu_ttm_tt_unbind(struct ttm_tt *ttm)
 {
-	/* Not implemented */
+	struct virtio_gpu_ttm_tt *gtt =
+		container_of(ttm, struct virtio_gpu_ttm_tt, ttm.ttm);
+	struct virtio_gpu_device *vgdev =
+		virtio_gpu_get_vgdev(gtt->obj->tbo.bdev);
+
+	virtio_gpu_object_detach(vgdev, gtt->obj);
 	return 0;
 }
 
-static void virtio_gpu_ttm_backend_destroy(struct ttm_tt *ttm)
+static void virtio_gpu_ttm_tt_destroy(struct ttm_tt *ttm)
 {
-	struct virtio_gpu_ttm_tt *gtt = (void *)ttm;
+	struct virtio_gpu_ttm_tt *gtt =
+		container_of(ttm, struct virtio_gpu_ttm_tt, ttm.ttm);
 
 	ttm_dma_tt_fini(&gtt->ttm);
 	kfree(gtt);
 }
 
-static struct ttm_backend_func virtio_gpu_backend_func = {
-	.bind = &virtio_gpu_ttm_backend_bind,
-	.unbind = &virtio_gpu_ttm_backend_unbind,
-	.destroy = &virtio_gpu_ttm_backend_destroy,
+static struct ttm_backend_func virtio_gpu_tt_func = {
+	.bind = &virtio_gpu_ttm_tt_bind,
+	.unbind = &virtio_gpu_ttm_tt_unbind,
+	.destroy = &virtio_gpu_ttm_tt_destroy,
 };
 
 static struct ttm_tt *virtio_gpu_ttm_tt_create(struct ttm_buffer_object *bo,
@@ -242,8 +239,8 @@ static struct ttm_tt *virtio_gpu_ttm_tt_create(struct ttm_buffer_object *bo,
 	gtt = kzalloc(sizeof(struct virtio_gpu_ttm_tt), GFP_KERNEL);
 	if (gtt == NULL)
 		return NULL;
-	gtt->ttm.ttm.func = &virtio_gpu_backend_func;
-	gtt->vgdev = vgdev;
+	gtt->ttm.ttm.func = &virtio_gpu_tt_func;
+	gtt->obj = container_of(bo, struct virtio_gpu_object, tbo);
 	if (ttm_dma_tt_init(&gtt->ttm, bo, page_flags)) {
 		kfree(gtt);
 		return NULL;
@@ -251,58 +248,11 @@ static struct ttm_tt *virtio_gpu_ttm_tt_create(struct ttm_buffer_object *bo,
 	return &gtt->ttm.ttm;
 }
 
-static void virtio_gpu_move_null(struct ttm_buffer_object *bo,
-				 struct ttm_mem_reg *new_mem)
-{
-	struct ttm_mem_reg *old_mem = &bo->mem;
-
-	BUG_ON(old_mem->mm_node != NULL);
-	*old_mem = *new_mem;
-	new_mem->mm_node = NULL;
-}
-
-static int virtio_gpu_bo_move(struct ttm_buffer_object *bo, bool evict,
-			      struct ttm_operation_ctx *ctx,
-			      struct ttm_mem_reg *new_mem)
-{
-	int ret;
-
-	ret = ttm_bo_wait(bo, ctx->interruptible, ctx->no_wait_gpu);
-	if (ret)
-		return ret;
-
-	virtio_gpu_move_null(bo, new_mem);
-	return 0;
-}
-
-static void virtio_gpu_bo_move_notify(struct ttm_buffer_object *tbo,
-				      bool evict,
-				      struct ttm_mem_reg *new_mem)
-{
-	struct virtio_gpu_object *bo;
-	struct virtio_gpu_device *vgdev;
-
-	bo = container_of(tbo, struct virtio_gpu_object, tbo);
-	vgdev = (struct virtio_gpu_device *)bo->gem_base.dev->dev_private;
-
-	if (!new_mem || (new_mem->placement & TTM_PL_FLAG_SYSTEM)) {
-		if (bo->hw_res_handle)
-			virtio_gpu_object_detach(vgdev, bo);
-
-	} else if (new_mem->placement & TTM_PL_FLAG_TT) {
-		if (bo->hw_res_handle) {
-			virtio_gpu_object_attach(vgdev, bo, NULL);
-		}
-	}
-}
-
 static void virtio_gpu_bo_swap_notify(struct ttm_buffer_object *tbo)
 {
 	struct virtio_gpu_object *bo;
-	struct virtio_gpu_device *vgdev;
 
 	bo = container_of(tbo, struct virtio_gpu_object, tbo);
-	vgdev = (struct virtio_gpu_device *)bo->gem_base.dev->dev_private;
 
 	if (bo->pages)
 		virtio_gpu_object_free_sg_table(bo);
@@ -314,11 +264,9 @@ static struct ttm_bo_driver virtio_gpu_bo_driver = {
 	.init_mem_type = &virtio_gpu_init_mem_type,
 	.eviction_valuable = ttm_bo_eviction_valuable,
 	.evict_flags = &virtio_gpu_evict_flags,
-	.move = &virtio_gpu_bo_move,
 	.verify_access = &virtio_gpu_verify_access,
 	.io_mem_reserve = &virtio_gpu_ttm_io_mem_reserve,
 	.io_mem_free = &virtio_gpu_ttm_io_mem_free,
-	.move_notify = &virtio_gpu_bo_move_notify,
 	.swap_notify = &virtio_gpu_bo_swap_notify,
 };
 
@@ -330,7 +278,7 @@ int virtio_gpu_ttm_init(struct virtio_gpu_device *vgdev)
 	r = ttm_bo_device_init(&vgdev->mman.bdev,
 			       &virtio_gpu_bo_driver,
 			       vgdev->ddev->anon_inode->i_mapping,
-			       DRM_FILE_PAGE_OFFSET, 0);
+			       false);
 	if (r) {
 		DRM_ERROR("failed initializing buffer object driver(%d).\n", r);
 		goto err_dev_init;
diff --git a/drivers/gpu/drm/virtio/virtgpu_vq.c b/drivers/gpu/drm/virtio/virtgpu_vq.c
index 6bc2008b0d0d..e62fe24b1a2e 100644
--- a/drivers/gpu/drm/virtio/virtgpu_vq.c
+++ b/drivers/gpu/drm/virtio/virtgpu_vq.c
@@ -376,9 +376,8 @@ retry:
 /* create a basic resource */
 void virtio_gpu_cmd_create_resource(struct virtio_gpu_device *vgdev,
 				    struct virtio_gpu_object *bo,
-				    uint32_t format,
-				    uint32_t width,
-				    uint32_t height)
+				    struct virtio_gpu_object_params *params,
+				    struct virtio_gpu_fence *fence)
 {
 	struct virtio_gpu_resource_create_2d *cmd_p;
 	struct virtio_gpu_vbuffer *vbuf;
@@ -388,11 +387,11 @@ void virtio_gpu_cmd_create_resource(struct virtio_gpu_device *vgdev,
 
 	cmd_p->hdr.type = cpu_to_le32(VIRTIO_GPU_CMD_RESOURCE_CREATE_2D);
 	cmd_p->resource_id = cpu_to_le32(bo->hw_res_handle);
-	cmd_p->format = cpu_to_le32(format);
-	cmd_p->width = cpu_to_le32(width);
-	cmd_p->height = cpu_to_le32(height);
+	cmd_p->format = cpu_to_le32(params->format);
+	cmd_p->width = cpu_to_le32(params->width);
+	cmd_p->height = cpu_to_le32(params->height);
 
-	virtio_gpu_queue_ctrl_buffer(vgdev, vbuf);
+	virtio_gpu_queue_fenced_ctrl_buffer(vgdev, vbuf, &cmd_p->hdr, fence);
 	bo->created = true;
 }
 
@@ -828,7 +827,8 @@ void virtio_gpu_cmd_context_detach_resource(struct virtio_gpu_device *vgdev,
 void
 virtio_gpu_cmd_resource_create_3d(struct virtio_gpu_device *vgdev,
 				  struct virtio_gpu_object *bo,
-				  struct virtio_gpu_resource_create_3d *rc_3d)
+				  struct virtio_gpu_object_params *params,
+				  struct virtio_gpu_fence *fence)
 {
 	struct virtio_gpu_resource_create_3d *cmd_p;
 	struct virtio_gpu_vbuffer *vbuf;
@@ -836,11 +836,21 @@ virtio_gpu_cmd_resource_create_3d(struct virtio_gpu_device *vgdev,
 	cmd_p = virtio_gpu_alloc_cmd(vgdev, &vbuf, sizeof(*cmd_p));
 	memset(cmd_p, 0, sizeof(*cmd_p));
 
-	*cmd_p = *rc_3d;
 	cmd_p->hdr.type = cpu_to_le32(VIRTIO_GPU_CMD_RESOURCE_CREATE_3D);
-	cmd_p->hdr.flags = 0;
+	cmd_p->resource_id = cpu_to_le32(bo->hw_res_handle);
+	cmd_p->format = cpu_to_le32(params->format);
+	cmd_p->width = cpu_to_le32(params->width);
+	cmd_p->height = cpu_to_le32(params->height);
+
+	cmd_p->target = cpu_to_le32(params->target);
+	cmd_p->bind = cpu_to_le32(params->bind);
+	cmd_p->depth = cpu_to_le32(params->depth);
+	cmd_p->array_size = cpu_to_le32(params->array_size);
+	cmd_p->last_level = cpu_to_le32(params->last_level);
+	cmd_p->nr_samples = cpu_to_le32(params->nr_samples);
+	cmd_p->flags = cpu_to_le32(params->flags);
 
-	virtio_gpu_queue_ctrl_buffer(vgdev, vbuf);
+	virtio_gpu_queue_fenced_ctrl_buffer(vgdev, vbuf, &cmd_p->hdr, fence);
 	bo->created = true;
 }
 
@@ -924,8 +934,8 @@ int virtio_gpu_object_attach(struct virtio_gpu_device *vgdev,
 	struct scatterlist *sg;
 	int si, nents;
 
-	if (!obj->created)
-		return 0;
+	if (WARN_ON_ONCE(!obj->created))
+		return -EINVAL;
 
 	if (!obj->pages) {
 		int ret;
diff --git a/drivers/gpu/drm/vkms/vkms_crtc.c b/drivers/gpu/drm/vkms/vkms_crtc.c
index 8a9aeb0a9ea8..bb66dbcd5e3f 100644
--- a/drivers/gpu/drm/vkms/vkms_crtc.c
+++ b/drivers/gpu/drm/vkms/vkms_crtc.c
@@ -219,6 +219,8 @@ int vkms_crtc_init(struct drm_device *dev, struct drm_crtc *crtc,
 	spin_lock_init(&vkms_out->state_lock);
 
 	vkms_out->crc_workq = alloc_ordered_workqueue("vkms_crc_workq", 0);
+	if (!vkms_out->crc_workq)
+		return -ENOMEM;
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_binding.c b/drivers/gpu/drm/vmwgfx/vmwgfx_binding.c
index 0b9ee7fb45d6..66e14e38d5e8 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_binding.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_binding.c
@@ -499,12 +499,9 @@ static int vmw_binding_scrub_shader(struct vmw_ctx_bindinfo *bi, bool rebind)
 		SVGA3dCmdSetShader body;
 	} *cmd;
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for shader "
-			  "unbinding.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	cmd->header.id = SVGA_3D_CMD_SET_SHADER;
 	cmd->header.size = sizeof(cmd->body);
@@ -534,12 +531,9 @@ static int vmw_binding_scrub_render_target(struct vmw_ctx_bindinfo *bi,
 		SVGA3dCmdSetRenderTarget body;
 	} *cmd;
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for render target "
-			  "unbinding.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	cmd->header.id = SVGA_3D_CMD_SETRENDERTARGET;
 	cmd->header.size = sizeof(cmd->body);
@@ -576,12 +570,9 @@ static int vmw_binding_scrub_texture(struct vmw_ctx_bindinfo *bi,
 		} body;
 	} *cmd;
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for texture "
-			  "unbinding.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	cmd->header.id = SVGA_3D_CMD_SETTEXTURESTATE;
 	cmd->header.size = sizeof(cmd->body);
@@ -610,12 +601,10 @@ static int vmw_binding_scrub_dx_shader(struct vmw_ctx_bindinfo *bi, bool rebind)
 		SVGA3dCmdDXSetShader body;
 	} *cmd;
 
-	cmd = vmw_fifo_reserve_dx(dev_priv, sizeof(*cmd), bi->ctx->id);
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for DX shader "
-			  "unbinding.\n");
+	cmd = VMW_FIFO_RESERVE_DX(dev_priv, sizeof(*cmd), bi->ctx->id);
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
+
 	cmd->header.id = SVGA_3D_CMD_DX_SET_SHADER;
 	cmd->header.size = sizeof(cmd->body);
 	cmd->body.type = binding->shader_slot + SVGA3D_SHADERTYPE_MIN;
@@ -641,12 +630,9 @@ static int vmw_binding_scrub_cb(struct vmw_ctx_bindinfo *bi, bool rebind)
 		SVGA3dCmdDXSetSingleConstantBuffer body;
 	} *cmd;
 
-	cmd = vmw_fifo_reserve_dx(dev_priv, sizeof(*cmd), bi->ctx->id);
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for DX shader "
-			  "unbinding.\n");
+	cmd = VMW_FIFO_RESERVE_DX(dev_priv, sizeof(*cmd), bi->ctx->id);
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	cmd->header.id = SVGA_3D_CMD_DX_SET_SINGLE_CONSTANT_BUFFER;
 	cmd->header.size = sizeof(cmd->body);
@@ -768,12 +754,9 @@ static int vmw_emit_set_sr(struct vmw_ctx_binding_state *cbs,
 
 	view_id_size = cbs->bind_cmd_count*sizeof(uint32);
 	cmd_size = sizeof(*cmd) + view_id_size;
-	cmd = vmw_fifo_reserve_dx(ctx->dev_priv, cmd_size, ctx->id);
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for DX shader"
-			  " resource binding.\n");
+	cmd = VMW_FIFO_RESERVE_DX(ctx->dev_priv, cmd_size, ctx->id);
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	cmd->header.id = SVGA_3D_CMD_DX_SET_SHADER_RESOURCES;
 	cmd->header.size = sizeof(cmd->body) + view_id_size;
@@ -807,12 +790,9 @@ static int vmw_emit_set_rt(struct vmw_ctx_binding_state *cbs)
 	vmw_collect_view_ids(cbs, loc, SVGA3D_MAX_SIMULTANEOUS_RENDER_TARGETS);
 	view_id_size = cbs->bind_cmd_count*sizeof(uint32);
 	cmd_size = sizeof(*cmd) + view_id_size;
-	cmd = vmw_fifo_reserve_dx(ctx->dev_priv, cmd_size, ctx->id);
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for DX render-target"
-			  " binding.\n");
+	cmd = VMW_FIFO_RESERVE_DX(ctx->dev_priv, cmd_size, ctx->id);
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	cmd->header.id = SVGA_3D_CMD_DX_SET_RENDERTARGETS;
 	cmd->header.size = sizeof(cmd->body) + view_id_size;
@@ -894,12 +874,9 @@ static int vmw_emit_set_so(struct vmw_ctx_binding_state *cbs)
 
 	so_target_size = cbs->bind_cmd_count*sizeof(SVGA3dSoTarget);
 	cmd_size = sizeof(*cmd) + so_target_size;
-	cmd = vmw_fifo_reserve_dx(ctx->dev_priv, cmd_size, ctx->id);
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for DX SO target"
-			  " binding.\n");
+	cmd = VMW_FIFO_RESERVE_DX(ctx->dev_priv, cmd_size, ctx->id);
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	cmd->header.id = SVGA_3D_CMD_DX_SET_SOTARGETS;
 	cmd->header.size = sizeof(cmd->body) + so_target_size;
@@ -1011,12 +988,9 @@ static int vmw_emit_set_vb(struct vmw_ctx_binding_state *cbs)
 
 	set_vb_size = cbs->bind_cmd_count*sizeof(SVGA3dVertexBuffer);
 	cmd_size = sizeof(*cmd) + set_vb_size;
-	cmd = vmw_fifo_reserve_dx(ctx->dev_priv, cmd_size, ctx->id);
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for DX vertex buffer"
-			  " binding.\n");
+	cmd = VMW_FIFO_RESERVE_DX(ctx->dev_priv, cmd_size, ctx->id);
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	cmd->header.id = SVGA_3D_CMD_DX_SET_VERTEX_BUFFERS;
 	cmd->header.size = sizeof(cmd->body) + set_vb_size;
@@ -1167,12 +1141,10 @@ static int vmw_binding_scrub_ib(struct vmw_ctx_bindinfo *bi, bool rebind)
 		SVGA3dCmdDXSetIndexBuffer body;
 	} *cmd;
 
-	cmd = vmw_fifo_reserve_dx(dev_priv, sizeof(*cmd), bi->ctx->id);
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for DX index buffer "
-			  "binding.\n");
+	cmd = VMW_FIFO_RESERVE_DX(dev_priv, sizeof(*cmd), bi->ctx->id);
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
+
 	cmd->header.id = SVGA_3D_CMD_DX_SET_INDEX_BUFFER;
 	cmd->header.size = sizeof(cmd->body);
 	if (rebind) {
@@ -1269,6 +1241,32 @@ void vmw_binding_state_reset(struct vmw_ctx_binding_state *cbs)
 		vmw_binding_drop(entry);
 }
 
+/**
+ * vmw_binding_dirtying - Return whether a binding type is dirtying its resource
+ * @binding_type: The binding type
+ *
+ * Each time a resource is put on the validation list as the result of a
+ * context binding referencing it, we need to determine whether that resource
+ * will be dirtied (written to by the GPU) as a result of the corresponding
+ * GPU operation. Currently rendertarget-, depth-stencil-, and
+ * stream-output-target bindings are capable of dirtying its resource.
+ *
+ * Return: Whether the binding type dirties the resource its binding points to.
+ */
+u32 vmw_binding_dirtying(enum vmw_ctx_binding_type binding_type)
+{
+	static u32 is_binding_dirtying[vmw_ctx_binding_max] = {
+		[vmw_ctx_binding_rt] = VMW_RES_DIRTY_SET,
+		[vmw_ctx_binding_dx_rt] = VMW_RES_DIRTY_SET,
+		[vmw_ctx_binding_ds] = VMW_RES_DIRTY_SET,
+		[vmw_ctx_binding_so] = VMW_RES_DIRTY_SET,
+	};
+
+	/* Review this function as new bindings are added. */
+	BUILD_BUG_ON(vmw_ctx_binding_max != 11);
+	return is_binding_dirtying[binding_type];
+}
+
 /*
  * This function is unused at run-time, and only used to hold various build
  * asserts important for code optimization assumptions.
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_binding.h b/drivers/gpu/drm/vmwgfx/vmwgfx_binding.h
index 6a2a9d69043b..f6ab79d23923 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_binding.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_binding.h
@@ -205,5 +205,7 @@ extern void vmw_binding_state_free(struct vmw_ctx_binding_state *cbs);
 extern struct list_head *
 vmw_binding_state_list(struct vmw_ctx_binding_state *cbs);
 extern void vmw_binding_state_reset(struct vmw_ctx_binding_state *cbs);
+extern u32 vmw_binding_dirtying(enum vmw_ctx_binding_type binding_type);
+
 
 #endif
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c
index 70dab55e7888..56979e412ca8 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c
@@ -393,6 +393,7 @@ static void vmw_cmdbuf_ctx_process(struct vmw_cmdbuf_man *man,
 			__vmw_cmdbuf_header_free(entry);
 			break;
 		case SVGA_CB_STATUS_COMMAND_ERROR:
+			WARN_ONCE(true, "Command buffer error.\n");
 			entry->cb_header->status = SVGA_CB_STATUS_NONE;
 			list_add_tail(&entry->list, &man->error);
 			schedule_work(&man->work);
@@ -511,17 +512,14 @@ static void vmw_cmdbuf_work_func(struct work_struct *work)
 		container_of(work, struct vmw_cmdbuf_man, work);
 	struct vmw_cmdbuf_header *entry, *next;
 	uint32_t dummy;
-	bool restart[SVGA_CB_CONTEXT_MAX];
 	bool send_fence = false;
 	struct list_head restart_head[SVGA_CB_CONTEXT_MAX];
 	int i;
 	struct vmw_cmdbuf_context *ctx;
 	bool global_block = false;
 
-	for_each_cmdbuf_ctx(man, i, ctx) {
+	for_each_cmdbuf_ctx(man, i, ctx)
 		INIT_LIST_HEAD(&restart_head[i]);
-		restart[i] = false;
-	}
 
 	mutex_lock(&man->error_mutex);
 	spin_lock(&man->lock);
@@ -533,23 +531,23 @@ static void vmw_cmdbuf_work_func(struct work_struct *work)
 		const char *cmd_name;
 
 		list_del_init(&entry->list);
-		restart[entry->cb_context] = true;
 		global_block = true;
 
 		if (!vmw_cmd_describe(header, &error_cmd_size, &cmd_name)) {
-			DRM_ERROR("Unknown command causing device error.\n");
-			DRM_ERROR("Command buffer offset is %lu\n",
-				  (unsigned long) cb_hdr->errorOffset);
+			VMW_DEBUG_USER("Unknown command causing device error.\n");
+			VMW_DEBUG_USER("Command buffer offset is %lu\n",
+				       (unsigned long) cb_hdr->errorOffset);
 			__vmw_cmdbuf_header_free(entry);
 			send_fence = true;
 			continue;
 		}
 
-		DRM_ERROR("Command \"%s\" causing device error.\n", cmd_name);
-		DRM_ERROR("Command buffer offset is %lu\n",
-			  (unsigned long) cb_hdr->errorOffset);
-		DRM_ERROR("Command size is %lu\n",
-			  (unsigned long) error_cmd_size);
+		VMW_DEBUG_USER("Command \"%s\" causing device error.\n",
+			       cmd_name);
+		VMW_DEBUG_USER("Command buffer offset is %lu\n",
+			       (unsigned long) cb_hdr->errorOffset);
+		VMW_DEBUG_USER("Command size is %lu\n",
+			       (unsigned long) error_cmd_size);
 
 		new_start_offset = cb_hdr->errorOffset + error_cmd_size;
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_context.c b/drivers/gpu/drm/vmwgfx/vmwgfx_context.c
index 14bd760a62fd..63f111068a44 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_context.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_context.c
@@ -156,12 +156,9 @@ static void vmw_hw_context_destroy(struct vmw_resource *res)
 	}
 
 	vmw_execbuf_release_pinned_bo(dev_priv);
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for surface "
-			  "destruction.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (unlikely(cmd == NULL))
 		return;
-	}
 
 	cmd->header.id = SVGA_3D_CMD_CONTEXT_DESTROY;
 	cmd->header.size = sizeof(cmd->body);
@@ -210,7 +207,7 @@ static int vmw_gb_context_init(struct vmw_private *dev_priv,
 		for (i = 0; i < SVGA_COTABLE_DX10_MAX; ++i) {
 			uctx->cotables[i] = vmw_cotable_alloc(dev_priv,
 							      &uctx->res, i);
-			if (unlikely(IS_ERR(uctx->cotables[i]))) {
+			if (IS_ERR(uctx->cotables[i])) {
 				ret = PTR_ERR(uctx->cotables[i]);
 				goto out_cotables;
 			}
@@ -259,9 +256,8 @@ static int vmw_context_init(struct vmw_private *dev_priv,
 		return -ENOMEM;
 	}
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
 	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Fifo reserve failed.\n");
 		vmw_resource_unreference(&res);
 		return -ENOMEM;
 	}
@@ -311,10 +307,8 @@ static int vmw_gb_context_create(struct vmw_resource *res)
 		goto out_no_fifo;
 	}
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
 	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for context "
-			  "creation.\n");
 		ret = -ENOMEM;
 		goto out_no_fifo;
 	}
@@ -345,12 +339,10 @@ static int vmw_gb_context_bind(struct vmw_resource *res,
 
 	BUG_ON(bo->mem.mem_type != VMW_PL_MOB);
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for context "
-			  "binding.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
+
 	cmd->header.id = SVGA_3D_CMD_BIND_GB_CONTEXT;
 	cmd->header.size = sizeof(cmd->body);
 	cmd->body.cid = res->id;
@@ -391,10 +383,8 @@ static int vmw_gb_context_unbind(struct vmw_resource *res,
 
 	submit_size = sizeof(*cmd2) + (readback ? sizeof(*cmd1) : 0);
 
-	cmd = vmw_fifo_reserve(dev_priv, submit_size);
+	cmd = VMW_FIFO_RESERVE(dev_priv, submit_size);
 	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for context "
-			  "unbinding.\n");
 		mutex_unlock(&dev_priv->binding_mutex);
 		return -ENOMEM;
 	}
@@ -441,12 +431,9 @@ static int vmw_gb_context_destroy(struct vmw_resource *res)
 	if (likely(res->id == -1))
 		return 0;
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for context "
-			  "destruction.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	cmd->header.id = SVGA_3D_CMD_DESTROY_GB_CONTEXT;
 	cmd->header.size = sizeof(cmd->body);
@@ -487,10 +474,8 @@ static int vmw_dx_context_create(struct vmw_resource *res)
 		goto out_no_fifo;
 	}
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
 	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for context "
-			  "creation.\n");
 		ret = -ENOMEM;
 		goto out_no_fifo;
 	}
@@ -521,12 +506,9 @@ static int vmw_dx_context_bind(struct vmw_resource *res,
 
 	BUG_ON(bo->mem.mem_type != VMW_PL_MOB);
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for context "
-			  "binding.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	cmd->header.id = SVGA_3D_CMD_DX_BIND_CONTEXT;
 	cmd->header.size = sizeof(cmd->body);
@@ -615,10 +597,8 @@ static int vmw_dx_context_unbind(struct vmw_resource *res,
 
 	submit_size = sizeof(*cmd2) + (readback ? sizeof(*cmd1) : 0);
 
-	cmd = vmw_fifo_reserve(dev_priv, submit_size);
+	cmd = VMW_FIFO_RESERVE(dev_priv, submit_size);
 	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for context "
-			  "unbinding.\n");
 		mutex_unlock(&dev_priv->binding_mutex);
 		return -ENOMEM;
 	}
@@ -665,12 +645,9 @@ static int vmw_dx_context_destroy(struct vmw_resource *res)
 	if (likely(res->id == -1))
 		return 0;
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for context "
-			  "destruction.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	cmd->header.id = SVGA_3D_CMD_DX_DESTROY_CONTEXT;
 	cmd->header.size = sizeof(cmd->body);
@@ -751,7 +728,7 @@ static int vmw_context_define(struct drm_device *dev, void *data,
 	int ret;
 
 	if (!dev_priv->has_dx && dx) {
-		DRM_ERROR("DX contexts not supported by device.\n");
+		VMW_DEBUG_USER("DX contexts not supported by device.\n");
 		return -EINVAL;
 	}
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c
index 44f3f6f107d3..b4f6e1217c9d 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c
@@ -171,12 +171,9 @@ static int vmw_cotable_unscrub(struct vmw_resource *res)
 	WARN_ON_ONCE(bo->mem.mem_type != VMW_PL_MOB);
 	lockdep_assert_held(&bo->resv->lock.base);
 
-	cmd = vmw_fifo_reserve_dx(dev_priv, sizeof(*cmd), SVGA3D_INVALID_ID);
-	if (!cmd) {
-		DRM_ERROR("Failed reserving FIFO space for cotable "
-			  "binding.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (!cmd)
 		return -ENOMEM;
-	}
 
 	WARN_ON(vcotbl->ctx->id == SVGA3D_INVALID_ID);
 	WARN_ON(bo->mem.mem_type != VMW_PL_MOB);
@@ -262,12 +259,9 @@ int vmw_cotable_scrub(struct vmw_resource *res, bool readback)
 	if (readback)
 		submit_size += sizeof(*cmd0);
 
-	cmd1 = vmw_fifo_reserve_dx(dev_priv, submit_size, SVGA3D_INVALID_ID);
-	if (!cmd1) {
-		DRM_ERROR("Failed reserving FIFO space for cotable "
-			  "unbinding.\n");
+	cmd1 = VMW_FIFO_RESERVE(dev_priv, submit_size);
+	if (!cmd1)
 		return -ENOMEM;
-	}
 
 	vcotbl->size_read_back = 0;
 	if (readback) {
@@ -351,13 +345,10 @@ static int vmw_cotable_readback(struct vmw_resource *res)
 	struct vmw_fence_obj *fence;
 
 	if (!vcotbl->scrubbed) {
-		cmd = vmw_fifo_reserve_dx(dev_priv, sizeof(*cmd),
-					  SVGA3D_INVALID_ID);
-		if (!cmd) {
-			DRM_ERROR("Failed reserving FIFO space for cotable "
-				  "readback.\n");
+		cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+		if (!cmd)
 			return -ENOMEM;
-		}
+
 		cmd->header.id = SVGA_3D_CMD_DX_READBACK_COTABLE;
 		cmd->header.size = sizeof(cmd->body);
 		cmd->body.cid = vcotbl->ctx->id;
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index 1bfa353d995c..bf6c3500d363 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -828,7 +828,6 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
 	ret = ttm_bo_device_init(&dev_priv->bdev,
 				 &vmw_bo_driver,
 				 dev->anon_inode->i_mapping,
-				 VMWGFX_FILE_PAGE_OFFSET,
 				 false);
 	if (unlikely(ret != 0)) {
 		DRM_ERROR("Failed initializing TTM buffer object driver.\n");
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
index accb2fafe2f1..96983c47fb40 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
@@ -48,7 +48,6 @@
 #define VMWGFX_DRIVER_MAJOR 2
 #define VMWGFX_DRIVER_MINOR 15
 #define VMWGFX_DRIVER_PATCHLEVEL 0
-#define VMWGFX_FILE_PAGE_OFFSET 0x00100000
 #define VMWGFX_FIFO_STATIC_SIZE (1024*1024)
 #define VMWGFX_MAX_RELOCATIONS 2048
 #define VMWGFX_MAX_VALIDATIONS 2048
@@ -700,6 +699,8 @@ extern int vmw_user_stream_lookup(struct vmw_private *dev_priv,
 				  uint32_t *inout_id,
 				  struct vmw_resource **out);
 extern void vmw_resource_unreserve(struct vmw_resource *res,
+				   bool dirty_set,
+				   bool dirty,
 				   bool switch_backup,
 				   struct vmw_buffer_object *new_backup,
 				   unsigned long new_backup_offset);
@@ -812,7 +813,6 @@ extern int vmw_fifo_init(struct vmw_private *dev_priv,
 			 struct vmw_fifo_state *fifo);
 extern void vmw_fifo_release(struct vmw_private *dev_priv,
 			     struct vmw_fifo_state *fifo);
-extern void *vmw_fifo_reserve(struct vmw_private *dev_priv, uint32_t bytes);
 extern void *
 vmw_fifo_reserve_dx(struct vmw_private *dev_priv, uint32_t bytes, int ctx_id);
 extern void vmw_fifo_commit(struct vmw_private *dev_priv, uint32_t bytes);
@@ -828,6 +828,18 @@ extern int vmw_fifo_emit_dummy_query(struct vmw_private *dev_priv,
 extern int vmw_fifo_flush(struct vmw_private *dev_priv,
 			  bool interruptible);
 
+#define VMW_FIFO_RESERVE_DX(__priv, __bytes, __ctx_id)                        \
+({                                                                            \
+	vmw_fifo_reserve_dx(__priv, __bytes, __ctx_id) ? : ({                 \
+		DRM_ERROR("FIFO reserve failed at %s for %u bytes\n",         \
+			  __func__, (unsigned int) __bytes);                  \
+		NULL;                                                         \
+	});                                                                   \
+})
+
+#define VMW_FIFO_RESERVE(__priv, __bytes)                                     \
+	VMW_FIFO_RESERVE_DX(__priv, __bytes, SVGA3D_INVALID_ID)
+
 /**
  * TTM glue - vmwgfx_ttm_glue.c
  */
@@ -1312,6 +1324,20 @@ int vmw_host_get_guestinfo(const char *guest_info_param,
 			   char *buffer, size_t *length);
 int vmw_host_log(const char *log);
 
+/* VMW logging */
+
+/**
+ * VMW_DEBUG_USER - Debug output for user-space debugging.
+ *
+ * @fmt: printf() like format string.
+ *
+ * This macro is for logging user-space error and debugging messages for e.g.
+ * command buffer execution errors due to malformed commands, invalid context,
+ * etc.
+ */
+#define VMW_DEBUG_USER(fmt, ...)                                              \
+	DRM_DEBUG_DRIVER(fmt, ##__VA_ARGS__)
+
 /**
  * Inline helper functions
  */
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
index 88b8178d4687..2ff7ba04d8c8 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
@@ -36,6 +36,25 @@
 #define VMW_RES_HT_ORDER 12
 
 /*
+ * Helper macro to get dx_ctx_node if available otherwise print an error
+ * message. This is for use in command verifier function where if dx_ctx_node
+ * is not set then command is invalid.
+ */
+#define VMW_GET_CTX_NODE(__sw_context)                                        \
+({                                                                            \
+	__sw_context->dx_ctx_node ? __sw_context->dx_ctx_node : ({            \
+		VMW_DEBUG_USER("SM context is not set at %s\n", __func__);    \
+		__sw_context->dx_ctx_node;                                    \
+	});                                                                   \
+})
+
+#define VMW_DECLARE_CMD_VAR(__var, __type)                                    \
+	struct {                                                              \
+		SVGA3dCmdHeader header;                                       \
+		__type body;                                                  \
+	} __var
+
+/**
  * struct vmw_relocation - Buffer object relocation
  *
  * @head: List head for the command submission context's relocation list
@@ -59,9 +78,8 @@ struct vmw_relocation {
  * command stream is replaced with the actual id after validation.
  * @vmw_res_rel_nop: NOP relocation. The command is unconditionally replaced
  * with a NOP.
- * @vmw_res_rel_cond_nop: Conditional NOP relocation. If the resource id
- * after validation is -1, the command is replaced with a NOP. Otherwise no
- * action.
+ * @vmw_res_rel_cond_nop: Conditional NOP relocation. If the resource id after
+ * validation is -1, the command is replaced with a NOP. Otherwise no action.
  */
 enum vmw_resource_relocation_type {
 	vmw_res_rel_normal,
@@ -75,8 +93,8 @@ enum vmw_resource_relocation_type {
  *
  * @head: List head for the software context's relocation list.
  * @res: Non-ref-counted pointer to the resource.
- * @offset: Offset of single byte entries into the command buffer where the
- * id that needs fixup is located.
+ * @offset: Offset of single byte entries into the command buffer where the id
+ * that needs fixup is located.
  * @rel_type: Type of relocation.
  */
 struct vmw_resource_relocation {
@@ -86,8 +104,9 @@ struct vmw_resource_relocation {
 	enum vmw_resource_relocation_type rel_type:3;
 };
 
-/*
+/**
  * struct vmw_ctx_validation_info - Extra validation metadata for contexts
+ *
  * @head: List head of context list
  * @ctx: The context resource
  * @cur: The context's persistent binding state
@@ -142,9 +161,10 @@ static size_t vmw_ptr_diff(void *a, void *b)
 
 /**
  * vmw_execbuf_bindings_commit - Commit modified binding state
+ *
  * @sw_context: The command submission context
- * @backoff: Whether this is part of the error path and binding state
- * changes should be ignored
+ * @backoff: Whether this is part of the error path and binding state changes
+ * should be ignored
  */
 static void vmw_execbuf_bindings_commit(struct vmw_sw_context *sw_context,
 					bool backoff)
@@ -154,6 +174,7 @@ static void vmw_execbuf_bindings_commit(struct vmw_sw_context *sw_context,
 	list_for_each_entry(entry, &sw_context->ctx_list, head) {
 		if (!backoff)
 			vmw_binding_state_commit(entry->cur, entry->staged);
+
 		if (entry->staged != sw_context->staged_bindings)
 			vmw_binding_state_free(entry->staged);
 		else
@@ -166,6 +187,7 @@ static void vmw_execbuf_bindings_commit(struct vmw_sw_context *sw_context,
 
 /**
  * vmw_bind_dx_query_mob - Bind the DX query MOB if referenced
+ *
  * @sw_context: The command submission context
  */
 static void vmw_bind_dx_query_mob(struct vmw_sw_context *sw_context)
@@ -176,8 +198,8 @@ static void vmw_bind_dx_query_mob(struct vmw_sw_context *sw_context)
 }
 
 /**
- * vmw_cmd_ctx_first_setup - Perform the setup needed when a context is
- * added to the validate list.
+ * vmw_cmd_ctx_first_setup - Perform the setup needed when a context is added to
+ * the validate list.
  *
  * @dev_priv: Pointer to the device private:
  * @sw_context: The command submission context
@@ -195,11 +217,8 @@ static int vmw_cmd_ctx_first_setup(struct vmw_private *dev_priv,
 		goto out_err;
 
 	if (!sw_context->staged_bindings) {
-		sw_context->staged_bindings =
-			vmw_binding_state_alloc(dev_priv);
+		sw_context->staged_bindings = vmw_binding_state_alloc(dev_priv);
 		if (IS_ERR(sw_context->staged_bindings)) {
-			DRM_ERROR("Failed to allocate context binding "
-				  "information.\n");
 			ret = PTR_ERR(sw_context->staged_bindings);
 			sw_context->staged_bindings = NULL;
 			goto out_err;
@@ -209,8 +228,6 @@ static int vmw_cmd_ctx_first_setup(struct vmw_private *dev_priv,
 	if (sw_context->staged_bindings_inuse) {
 		node->staged = vmw_binding_state_alloc(dev_priv);
 		if (IS_ERR(node->staged)) {
-			DRM_ERROR("Failed to allocate context binding "
-				  "information.\n");
 			ret = PTR_ERR(node->staged);
 			node->staged = NULL;
 			goto out_err;
@@ -225,19 +242,20 @@ static int vmw_cmd_ctx_first_setup(struct vmw_private *dev_priv,
 	list_add_tail(&node->head, &sw_context->ctx_list);
 
 	return 0;
+
 out_err:
 	return ret;
 }
 
 /**
- * vmw_execbuf_res_size - calculate extra size fore the resource validation
- * node
+ * vmw_execbuf_res_size - calculate extra size fore the resource validation node
+ *
  * @dev_priv: Pointer to the device private struct.
  * @res_type: The resource type.
  *
- * Guest-backed contexts and DX contexts require extra size to store
- * execbuf private information in the validation node. Typically the
- * binding manager associated data structures.
+ * Guest-backed contexts and DX contexts require extra size to store execbuf
+ * private information in the validation node. Typically the binding manager
+ * associated data structures.
  *
  * Returns: The extra size requirement based on resource type.
  */
@@ -254,8 +272,8 @@ static unsigned int vmw_execbuf_res_size(struct vmw_private *dev_priv,
  *
  * @rcache: Pointer to the entry to update.
  * @res: Pointer to the resource.
- * @private: Pointer to the execbuf-private space in the resource
- * validation node.
+ * @private: Pointer to the execbuf-private space in the resource validation
+ * node.
  */
 static void vmw_execbuf_rcache_update(struct vmw_res_cache_entry *rcache,
 				      struct vmw_resource *res,
@@ -268,17 +286,19 @@ static void vmw_execbuf_rcache_update(struct vmw_res_cache_entry *rcache,
 }
 
 /**
- * vmw_execbuf_res_noref_val_add - Add a resource described by an
- * unreferenced rcu-protected pointer to the validation list.
+ * vmw_execbuf_res_noref_val_add - Add a resource described by an unreferenced
+ * rcu-protected pointer to the validation list.
+ *
  * @sw_context: Pointer to the software context.
  * @res: Unreferenced rcu-protected pointer to the resource.
+ * @dirty: Whether to change dirty status.
  *
- * Returns: 0 on success. Negative error code on failure. Typical error
- * codes are %-EINVAL on inconsistency and %-ESRCH if the resource was
- * doomed.
+ * Returns: 0 on success. Negative error code on failure. Typical error codes
+ * are %-EINVAL on inconsistency and %-ESRCH if the resource was doomed.
  */
 static int vmw_execbuf_res_noref_val_add(struct vmw_sw_context *sw_context,
-					 struct vmw_resource *res)
+					 struct vmw_resource *res,
+					 u32 dirty)
 {
 	struct vmw_private *dev_priv = res->dev_priv;
 	int ret;
@@ -290,13 +310,17 @@ static int vmw_execbuf_res_noref_val_add(struct vmw_sw_context *sw_context,
 
 	rcache = &sw_context->res_cache[res_type];
 	if (likely(rcache->valid && rcache->res == res)) {
+		if (dirty)
+			vmw_validation_res_set_dirty(sw_context->ctx,
+						     rcache->private, dirty);
 		vmw_user_resource_noref_release();
 		return 0;
 	}
 
 	priv_size = vmw_execbuf_res_size(dev_priv, res_type);
 	ret = vmw_validation_add_resource(sw_context->ctx, res, priv_size,
-					  (void **)&ctx_info, &first_usage);
+					  dirty, (void **)&ctx_info,
+					  &first_usage);
 	vmw_user_resource_noref_release();
 	if (ret)
 		return ret;
@@ -304,8 +328,10 @@ static int vmw_execbuf_res_noref_val_add(struct vmw_sw_context *sw_context,
 	if (priv_size && first_usage) {
 		ret = vmw_cmd_ctx_first_setup(dev_priv, sw_context, res,
 					      ctx_info);
-		if (ret)
+		if (ret) {
+			VMW_DEBUG_USER("Failed first usage context setup.\n");
 			return ret;
+		}
 	}
 
 	vmw_execbuf_rcache_update(rcache, res, ctx_info);
@@ -315,13 +341,16 @@ static int vmw_execbuf_res_noref_val_add(struct vmw_sw_context *sw_context,
 /**
  * vmw_execbuf_res_noctx_val_add - Add a non-context resource to the resource
  * validation list if it's not already on it
+ *
  * @sw_context: Pointer to the software context.
  * @res: Pointer to the resource.
+ * @dirty: Whether to change dirty status.
  *
  * Returns: Zero on success. Negative error code on failure.
  */
 static int vmw_execbuf_res_noctx_val_add(struct vmw_sw_context *sw_context,
-					 struct vmw_resource *res)
+					 struct vmw_resource *res,
+					 u32 dirty)
 {
 	struct vmw_res_cache_entry *rcache;
 	enum vmw_res_type res_type = vmw_res_type(res);
@@ -329,10 +358,15 @@ static int vmw_execbuf_res_noctx_val_add(struct vmw_sw_context *sw_context,
 	int ret;
 
 	rcache = &sw_context->res_cache[res_type];
-	if (likely(rcache->valid && rcache->res == res))
+	if (likely(rcache->valid && rcache->res == res)) {
+		if (dirty)
+			vmw_validation_res_set_dirty(sw_context->ctx,
+						     rcache->private, dirty);
 		return 0;
+	}
 
-	ret = vmw_validation_add_resource(sw_context->ctx, res, 0, &ptr, NULL);
+	ret = vmw_validation_add_resource(sw_context->ctx, res, 0, dirty,
+					  &ptr, NULL);
 	if (ret)
 		return ret;
 
@@ -342,8 +376,8 @@ static int vmw_execbuf_res_noctx_val_add(struct vmw_sw_context *sw_context,
 }
 
 /**
- * vmw_view_res_val_add - Add a view and the surface it's pointing to
- * to the validation list
+ * vmw_view_res_val_add - Add a view and the surface it's pointing to to the
+ * validation list
  *
  * @sw_context: The software context holding the validation list.
  * @view: Pointer to the view resource.
@@ -356,27 +390,29 @@ static int vmw_view_res_val_add(struct vmw_sw_context *sw_context,
 	int ret;
 
 	/*
-	 * First add the resource the view is pointing to, otherwise
-	 * it may be swapped out when the view is validated.
+	 * First add the resource the view is pointing to, otherwise it may be
+	 * swapped out when the view is validated.
 	 */
-	ret = vmw_execbuf_res_noctx_val_add(sw_context, vmw_view_srf(view));
+	ret = vmw_execbuf_res_noctx_val_add(sw_context, vmw_view_srf(view),
+					    vmw_view_dirtying(view));
 	if (ret)
 		return ret;
 
-	return vmw_execbuf_res_noctx_val_add(sw_context, view);
+	return vmw_execbuf_res_noctx_val_add(sw_context, view,
+					     VMW_RES_DIRTY_NONE);
 }
 
 /**
- * vmw_view_id_val_add - Look up a view and add it and the surface it's
- * pointing to to the validation list.
+ * vmw_view_id_val_add - Look up a view and add it and the surface it's pointing
+ * to to the validation list.
  *
  * @sw_context: The software context holding the validation list.
  * @view_type: The view type to look up.
  * @id: view id of the view.
  *
- * The view is represented by a view id and the DX context it's created on,
- * or scheduled for creation on. If there is no DX context set, the function
- * will return an -EINVAL error pointer.
+ * The view is represented by a view id and the DX context it's created on, or
+ * scheduled for creation on. If there is no DX context set, the function will
+ * return an -EINVAL error pointer.
  *
  * Returns: Unreferenced pointer to the resource on success, negative error
  * pointer on failure.
@@ -389,10 +425,8 @@ vmw_view_id_val_add(struct vmw_sw_context *sw_context,
 	struct vmw_resource *view;
 	int ret;
 
-	if (!ctx_node) {
-		DRM_ERROR("DX Context not set.\n");
+	if (!ctx_node)
 		return ERR_PTR(-EINVAL);
-	}
 
 	view = vmw_view_lookup(sw_context->man, view_type, id);
 	if (IS_ERR(view))
@@ -413,8 +447,8 @@ vmw_view_id_val_add(struct vmw_sw_context *sw_context,
  * @sw_context: Pointer to a software context used for this command submission
  * @ctx: Pointer to the context resource
  *
- * This function puts all resources that were previously bound to @ctx on
- * the resource validation list. This is part of the context state reemission
+ * This function puts all resources that were previously bound to @ctx on the
+ * resource validation list. This is part of the context state reemission
  */
 static int vmw_resource_context_res_add(struct vmw_private *dev_priv,
 					struct vmw_sw_context *sw_context,
@@ -433,13 +467,13 @@ static int vmw_resource_context_res_add(struct vmw_private *dev_priv,
 			if (IS_ERR(res))
 				continue;
 
-			ret = vmw_execbuf_res_noctx_val_add(sw_context, res);
+			ret = vmw_execbuf_res_noctx_val_add(sw_context, res,
+							    VMW_RES_DIRTY_SET);
 			if (unlikely(ret != 0))
 				return ret;
 		}
 	}
 
-
 	/* Add all resources bound to the context to the validation list */
 	mutex_lock(&dev_priv->binding_mutex);
 	binding_list = vmw_context_binding_list(ctx);
@@ -448,8 +482,9 @@ static int vmw_resource_context_res_add(struct vmw_private *dev_priv,
 		if (vmw_res_type(entry->res) == vmw_res_view)
 			ret = vmw_view_res_val_add(sw_context, entry->res);
 		else
-			ret = vmw_execbuf_res_noctx_val_add(sw_context,
-							    entry->res);
+			ret = vmw_execbuf_res_noctx_val_add
+				(sw_context, entry->res,
+				 vmw_binding_dirtying(entry->bt));
 		if (unlikely(ret != 0))
 			break;
 	}
@@ -472,8 +507,8 @@ static int vmw_resource_context_res_add(struct vmw_private *dev_priv,
  *
  * @list: Pointer to head of relocation list.
  * @res: The resource.
- * @offset: Offset into the command buffer currently being parsed where the
- * id that needs fixup is located. Granularity is one byte.
+ * @offset: Offset into the command buffer currently being parsed where the id
+ * that needs fixup is located. Granularity is one byte.
  * @rel_type: Relocation type.
  */
 static int vmw_resource_relocation_add(struct vmw_sw_context *sw_context,
@@ -486,7 +521,7 @@ static int vmw_resource_relocation_add(struct vmw_sw_context *sw_context,
 
 	rel = vmw_validation_mem_alloc(sw_context->ctx, sizeof(*rel));
 	if (unlikely(!rel)) {
-		DRM_ERROR("Failed to allocate a resource relocation.\n");
+		VMW_DEBUG_USER("Failed to allocate a resource relocation.\n");
 		return -ENOMEM;
 	}
 
@@ -506,17 +541,15 @@ static int vmw_resource_relocation_add(struct vmw_sw_context *sw_context,
 static void vmw_resource_relocations_free(struct list_head *list)
 {
 	/* Memory is validation context memory, so no need to free it */
-
 	INIT_LIST_HEAD(list);
 }
 
 /**
  * vmw_resource_relocations_apply - Apply all relocations on a list
  *
- * @cb: Pointer to the start of the command buffer bein patch. This need
- * not be the same buffer as the one being parsed when the relocation
- * list was built, but the contents must be the same modulo the
- * resource ids.
+ * @cb: Pointer to the start of the command buffer bein patch. This need not be
+ * the same buffer as the one being parsed when the relocation list was built,
+ * but the contents must be the same modulo the resource ids.
  * @list: Pointer to the head of the relocation list.
  */
 static void vmw_resource_relocations_apply(uint32_t *cb,
@@ -560,14 +593,14 @@ static int vmw_cmd_ok(struct vmw_private *dev_priv,
 }
 
 /**
- * vmw_resources_reserve - Reserve all resources on the sw_context's
- * resource list.
+ * vmw_resources_reserve - Reserve all resources on the sw_context's resource
+ * list.
  *
  * @sw_context: Pointer to the software context.
  *
- * Note that since vmware's command submission currently is protected by
- * the cmdbuf mutex, no fancy deadlock avoidance is required for resources,
- * since only a single thread at once will attempt this.
+ * Note that since vmware's command submission currently is protected by the
+ * cmdbuf mutex, no fancy deadlock avoidance is required for resources, since
+ * only a single thread at once will attempt this.
  */
 static int vmw_resources_reserve(struct vmw_sw_context *sw_context)
 {
@@ -592,22 +625,24 @@ static int vmw_resources_reserve(struct vmw_sw_context *sw_context)
 }
 
 /**
- * vmw_cmd_res_check - Check that a resource is present and if so, put it
- * on the resource validate list unless it's already there.
+ * vmw_cmd_res_check - Check that a resource is present and if so, put it on the
+ * resource validate list unless it's already there.
  *
  * @dev_priv: Pointer to a device private structure.
  * @sw_context: Pointer to the software context.
  * @res_type: Resource type.
+ * @dirty: Whether to change dirty status.
  * @converter: User-space visisble type specific information.
- * @id_loc: Pointer to the location in the command buffer currently being
- * parsed from where the user-space resource id handle is located.
- * @p_val: Pointer to pointer to resource validalidation node. Populated
- * on exit.
+ * @id_loc: Pointer to the location in the command buffer currently being parsed
+ * from where the user-space resource id handle is located.
+ * @p_val: Pointer to pointer to resource validalidation node. Populated on
+ * exit.
  */
 static int
 vmw_cmd_res_check(struct vmw_private *dev_priv,
 		  struct vmw_sw_context *sw_context,
 		  enum vmw_res_type res_type,
+		  u32 dirty,
 		  const struct vmw_user_resource_conv *converter,
 		  uint32_t *id_loc,
 		  struct vmw_resource **p_res)
@@ -621,7 +656,7 @@ vmw_cmd_res_check(struct vmw_private *dev_priv,
 
 	if (*id_loc == SVGA3D_INVALID_ID) {
 		if (res_type == vmw_res_context) {
-			DRM_ERROR("Illegal context invalid id.\n");
+			VMW_DEBUG_USER("Illegal context invalid id.\n");
 			return -EINVAL;
 		}
 		return 0;
@@ -629,6 +664,9 @@ vmw_cmd_res_check(struct vmw_private *dev_priv,
 
 	if (likely(rcache->valid_handle && *id_loc == rcache->handle)) {
 		res = rcache->res;
+		if (dirty)
+			vmw_validation_res_set_dirty(sw_context->ctx,
+						     rcache->private, dirty);
 	} else {
 		unsigned int size = vmw_execbuf_res_size(dev_priv, res_type);
 
@@ -638,13 +676,13 @@ vmw_cmd_res_check(struct vmw_private *dev_priv,
 
 		res = vmw_user_resource_noref_lookup_handle
 			(dev_priv, sw_context->fp->tfile, *id_loc, converter);
-		if (unlikely(IS_ERR(res))) {
-			DRM_ERROR("Could not find or use resource 0x%08x.\n",
-				  (unsigned int) *id_loc);
+		if (IS_ERR(res)) {
+			VMW_DEBUG_USER("Could not find/use resource 0x%08x.\n",
+				       (unsigned int) *id_loc);
 			return PTR_ERR(res);
 		}
 
-		ret = vmw_execbuf_res_noref_val_add(sw_context, res);
+		ret = vmw_execbuf_res_noref_val_add(sw_context, res, dirty);
 		if (unlikely(ret != 0))
 			return ret;
 
@@ -675,23 +713,16 @@ static int vmw_rebind_all_dx_query(struct vmw_resource *ctx_res)
 {
 	struct vmw_private *dev_priv = ctx_res->dev_priv;
 	struct vmw_buffer_object *dx_query_mob;
-	struct {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdDXBindAllQuery body;
-	} *cmd;
-
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXBindAllQuery);
 
 	dx_query_mob = vmw_context_get_dx_query_mob(ctx_res);
 
 	if (!dx_query_mob || dx_query_mob->dx_query_ctx)
 		return 0;
 
-	cmd = vmw_fifo_reserve_dx(dev_priv, sizeof(*cmd), ctx_res->id);
-
-	if (cmd == NULL) {
-		DRM_ERROR("Failed to rebind queries.\n");
+	cmd = VMW_FIFO_RESERVE_DX(dev_priv, sizeof(*cmd), ctx_res->id);
+	if (cmd == NULL)
 		return -ENOMEM;
-	}
 
 	cmd->header.id = SVGA_3D_CMD_DX_BIND_ALL_QUERY;
 	cmd->header.size = sizeof(cmd->body);
@@ -705,8 +736,8 @@ static int vmw_rebind_all_dx_query(struct vmw_resource *ctx_res)
 }
 
 /**
- * vmw_rebind_contexts - Rebind all resources previously bound to
- * referenced contexts.
+ * vmw_rebind_contexts - Rebind all resources previously bound to referenced
+ * contexts.
  *
  * @sw_context: Pointer to the software context.
  *
@@ -721,21 +752,23 @@ static int vmw_rebind_contexts(struct vmw_sw_context *sw_context)
 		ret = vmw_binding_rebind_all(val->cur);
 		if (unlikely(ret != 0)) {
 			if (ret != -ERESTARTSYS)
-				DRM_ERROR("Failed to rebind context.\n");
+				VMW_DEBUG_USER("Failed to rebind context.\n");
 			return ret;
 		}
 
 		ret = vmw_rebind_all_dx_query(val->ctx);
-		if (ret != 0)
+		if (ret != 0) {
+			VMW_DEBUG_USER("Failed to rebind queries.\n");
 			return ret;
+		}
 	}
 
 	return 0;
 }
 
 /**
- * vmw_view_bindings_add - Add an array of view bindings to a context
- * binding state tracker.
+ * vmw_view_bindings_add - Add an array of view bindings to a context binding
+ * state tracker.
  *
  * @sw_context: The execbuf state used for this command.
  * @view_type: View type for the bindings.
@@ -752,13 +785,11 @@ static int vmw_view_bindings_add(struct vmw_sw_context *sw_context,
 				 uint32 view_ids[], u32 num_views,
 				 u32 first_slot)
 {
-	struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node;
+	struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context);
 	u32 i;
 
-	if (!ctx_node) {
-		DRM_ERROR("DX Context not set.\n");
+	if (!ctx_node)
 		return -EINVAL;
-	}
 
 	for (i = 0; i < num_views; ++i) {
 		struct vmw_ctx_bindinfo_view binding;
@@ -768,7 +799,7 @@ static int vmw_view_bindings_add(struct vmw_sw_context *sw_context,
 			view = vmw_view_id_val_add(sw_context, view_type,
 						   view_ids[i]);
 			if (IS_ERR(view)) {
-				DRM_ERROR("View not found.\n");
+				VMW_DEBUG_USER("View not found.\n");
 				return PTR_ERR(view);
 			}
 		}
@@ -798,19 +829,18 @@ static int vmw_cmd_cid_check(struct vmw_private *dev_priv,
 			     struct vmw_sw_context *sw_context,
 			     SVGA3dCmdHeader *header)
 {
-	struct vmw_cid_cmd {
-		SVGA3dCmdHeader header;
-		uint32_t cid;
-	} *cmd;
+	VMW_DECLARE_CMD_VAR(*cmd, uint32_t) =
+		container_of(header, typeof(*cmd), header);
 
-	cmd = container_of(header, struct vmw_cid_cmd, header);
 	return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context,
-				 user_context_converter, &cmd->cid, NULL);
+				 VMW_RES_DIRTY_SET, user_context_converter,
+				 &cmd->body, NULL);
 }
 
 /**
  * vmw_execbuf_info_from_res - Get the private validation metadata for a
  * recently validated resource
+ *
  * @sw_context: Pointer to the command submission context
  * @res: The resource
  *
@@ -818,8 +848,8 @@ static int vmw_cmd_cid_check(struct vmw_private *dev_priv,
  * context's resource cache and hence the last resource of that type to be
  * processed by the validation code.
  *
- * Return: a pointer to the private metadata of the resource, or NULL
- * if it wasn't found
+ * Return: a pointer to the private metadata of the resource, or NULL if it
+ * wasn't found
  */
 static struct vmw_ctx_validation_info *
 vmw_execbuf_info_from_res(struct vmw_sw_context *sw_context,
@@ -835,36 +865,32 @@ vmw_execbuf_info_from_res(struct vmw_sw_context *sw_context,
 	return NULL;
 }
 
-
 static int vmw_cmd_set_render_target_check(struct vmw_private *dev_priv,
 					   struct vmw_sw_context *sw_context,
 					   SVGA3dCmdHeader *header)
 {
-	struct vmw_sid_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdSetRenderTarget body;
-	} *cmd;
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdSetRenderTarget);
 	struct vmw_resource *ctx;
 	struct vmw_resource *res;
 	int ret;
 
-	cmd = container_of(header, struct vmw_sid_cmd, header);
+	cmd = container_of(header, typeof(*cmd), header);
 
 	if (cmd->body.type >= SVGA3D_RT_MAX) {
-		DRM_ERROR("Illegal render target type %u.\n",
-			  (unsigned) cmd->body.type);
+		VMW_DEBUG_USER("Illegal render target type %u.\n",
+			       (unsigned int) cmd->body.type);
 		return -EINVAL;
 	}
 
 	ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context,
-				user_context_converter, &cmd->body.cid,
-				&ctx);
+				VMW_RES_DIRTY_SET, user_context_converter,
+				&cmd->body.cid, &ctx);
 	if (unlikely(ret != 0))
 		return ret;
 
 	ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				user_surface_converter, &cmd->body.target.sid,
-				&res);
+				VMW_RES_DIRTY_SET, user_surface_converter,
+				&cmd->body.target.sid, &res);
 	if (unlikely(ret))
 		return ret;
 
@@ -890,44 +916,38 @@ static int vmw_cmd_surface_copy_check(struct vmw_private *dev_priv,
 				      struct vmw_sw_context *sw_context,
 				      SVGA3dCmdHeader *header)
 {
-	struct vmw_sid_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdSurfaceCopy body;
-	} *cmd;
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdSurfaceCopy);
 	int ret;
 
-	cmd = container_of(header, struct vmw_sid_cmd, header);
+	cmd = container_of(header, typeof(*cmd), header);
 
 	ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-					  user_surface_converter,
-					  &cmd->body.src.sid, NULL);
+				VMW_RES_DIRTY_NONE, user_surface_converter,
+				&cmd->body.src.sid, NULL);
 	if (ret)
 		return ret;
 
 	return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				 user_surface_converter,
+				 VMW_RES_DIRTY_SET, user_surface_converter,
 				 &cmd->body.dest.sid, NULL);
 }
 
 static int vmw_cmd_buffer_copy_check(struct vmw_private *dev_priv,
-				      struct vmw_sw_context *sw_context,
-				      SVGA3dCmdHeader *header)
+				     struct vmw_sw_context *sw_context,
+				     SVGA3dCmdHeader *header)
 {
-	struct {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdDXBufferCopy body;
-	} *cmd;
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXBufferCopy);
 	int ret;
 
 	cmd = container_of(header, typeof(*cmd), header);
 	ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				user_surface_converter,
+				VMW_RES_DIRTY_NONE, user_surface_converter,
 				&cmd->body.src, NULL);
 	if (ret != 0)
 		return ret;
 
 	return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				 user_surface_converter,
+				 VMW_RES_DIRTY_SET, user_surface_converter,
 				 &cmd->body.dest, NULL);
 }
 
@@ -935,21 +955,18 @@ static int vmw_cmd_pred_copy_check(struct vmw_private *dev_priv,
 				   struct vmw_sw_context *sw_context,
 				   SVGA3dCmdHeader *header)
 {
-	struct {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdDXPredCopyRegion body;
-	} *cmd;
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXPredCopyRegion);
 	int ret;
 
 	cmd = container_of(header, typeof(*cmd), header);
 	ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				user_surface_converter,
+				VMW_RES_DIRTY_NONE, user_surface_converter,
 				&cmd->body.srcSid, NULL);
 	if (ret != 0)
 		return ret;
 
 	return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				 user_surface_converter,
+				 VMW_RES_DIRTY_SET, user_surface_converter,
 				 &cmd->body.dstSid, NULL);
 }
 
@@ -957,20 +974,18 @@ static int vmw_cmd_stretch_blt_check(struct vmw_private *dev_priv,
 				     struct vmw_sw_context *sw_context,
 				     SVGA3dCmdHeader *header)
 {
-	struct vmw_sid_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdSurfaceStretchBlt body;
-	} *cmd;
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdSurfaceStretchBlt);
 	int ret;
 
-	cmd = container_of(header, struct vmw_sid_cmd, header);
+	cmd = container_of(header, typeof(*cmd), header);
 	ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				user_surface_converter,
+				VMW_RES_DIRTY_NONE, user_surface_converter,
 				&cmd->body.src.sid, NULL);
 	if (unlikely(ret != 0))
 		return ret;
+
 	return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				 user_surface_converter,
+				 VMW_RES_DIRTY_SET, user_surface_converter,
 				 &cmd->body.dest.sid, NULL);
 }
 
@@ -978,15 +993,11 @@ static int vmw_cmd_blt_surf_screen_check(struct vmw_private *dev_priv,
 					 struct vmw_sw_context *sw_context,
 					 SVGA3dCmdHeader *header)
 {
-	struct vmw_sid_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdBlitSurfaceToScreen body;
-	} *cmd;
-
-	cmd = container_of(header, struct vmw_sid_cmd, header);
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdBlitSurfaceToScreen) =
+		container_of(header, typeof(*cmd), header);
 
 	return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				 user_surface_converter,
+				 VMW_RES_DIRTY_NONE, user_surface_converter,
 				 &cmd->body.srcImage.sid, NULL);
 }
 
@@ -994,17 +1005,12 @@ static int vmw_cmd_present_check(struct vmw_private *dev_priv,
 				 struct vmw_sw_context *sw_context,
 				 SVGA3dCmdHeader *header)
 {
-	struct vmw_sid_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdPresent body;
-	} *cmd;
-
-
-	cmd = container_of(header, struct vmw_sid_cmd, header);
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdPresent) =
+		container_of(header, typeof(*cmd), header);
 
 	return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				 user_surface_converter, &cmd->body.sid,
-				 NULL);
+				 VMW_RES_DIRTY_NONE, user_surface_converter,
+				 &cmd->body.sid, NULL);
 }
 
 /**
@@ -1014,11 +1020,10 @@ static int vmw_cmd_present_check(struct vmw_private *dev_priv,
  * @new_query_bo: The new buffer holding query results.
  * @sw_context: The software context used for this command submission.
  *
- * This function checks whether @new_query_bo is suitable for holding
- * query results, and if another buffer currently is pinned for query
- * results. If so, the function prepares the state of @sw_context for
- * switching pinned buffers after successful submission of the current
- * command batch.
+ * This function checks whether @new_query_bo is suitable for holding query
+ * results, and if another buffer currently is pinned for query results. If so,
+ * the function prepares the state of @sw_context for switching pinned buffers
+ * after successful submission of the current command batch.
  */
 static int vmw_query_bo_switch_prepare(struct vmw_private *dev_priv,
 				       struct vmw_buffer_object *new_query_bo,
@@ -1034,7 +1039,7 @@ static int vmw_query_bo_switch_prepare(struct vmw_private *dev_priv,
 	if (unlikely(new_query_bo != sw_context->cur_query_bo)) {
 
 		if (unlikely(new_query_bo->base.num_pages > 4)) {
-			DRM_ERROR("Query buffer too large.\n");
+			VMW_DEBUG_USER("Query buffer too large.\n");
 			return -EINVAL;
 		}
 
@@ -1053,13 +1058,11 @@ static int vmw_query_bo_switch_prepare(struct vmw_private *dev_priv,
 					    dev_priv->has_mob, false);
 		if (unlikely(ret != 0))
 			return ret;
-
 	}
 
 	return 0;
 }
 
-
 /**
  * vmw_query_bo_switch_commit - Finalize switching pinned query buffer
  *
@@ -1068,11 +1071,11 @@ static int vmw_query_bo_switch_prepare(struct vmw_private *dev_priv,
  *
  * This function will check if we're switching query buffers, and will then,
  * issue a dummy occlusion query wait used as a query barrier. When the fence
- * object following that query wait has signaled, we are sure that all
- * preceding queries have finished, and the old query buffer can be unpinned.
- * However, since both the new query buffer and the old one are fenced with
- * that fence, we can do an asynchronus unpin now, and be sure that the
- * old query buffer won't be moved until the fence has signaled.
+ * object following that query wait has signaled, we are sure that all preceding
+ * queries have finished, and the old query buffer can be unpinned. However,
+ * since both the new query buffer and the old one are fenced with that fence,
+ * we can do an asynchronus unpin now, and be sure that the old query buffer
+ * won't be moved until the fence has signaled.
  *
  * As mentioned above, both the new - and old query buffers need to be fenced
  * using a sequence emitted *after* calling this function.
@@ -1084,7 +1087,6 @@ static void vmw_query_bo_switch_commit(struct vmw_private *dev_priv,
 	 * The validate list should still hold references to all
 	 * contexts here.
 	 */
-
 	if (sw_context->needs_post_query_barrier) {
 		struct vmw_res_cache_entry *ctx_entry =
 			&sw_context->res_cache[vmw_res_context];
@@ -1097,7 +1099,7 @@ static void vmw_query_bo_switch_commit(struct vmw_private *dev_priv,
 		ret = vmw_fifo_emit_dummy_query(dev_priv, ctx->id);
 
 		if (unlikely(ret != 0))
-			DRM_ERROR("Out of fifo space for dummy query.\n");
+			VMW_DEBUG_USER("Out of fifo space for dummy query.\n");
 	}
 
 	if (dev_priv->pinned_bo != sw_context->cur_query_bo) {
@@ -1111,10 +1113,9 @@ static void vmw_query_bo_switch_commit(struct vmw_private *dev_priv,
 
 			/*
 			 * We pin also the dummy_query_bo buffer so that we
-			 * don't need to validate it when emitting
-			 * dummy queries in context destroy paths.
+			 * don't need to validate it when emitting dummy queries
+			 * in context destroy paths.
 			 */
-
 			if (!dev_priv->dummy_query_bo_pinned) {
 				vmw_bo_pin_reserved(dev_priv->dummy_query_bo,
 						    true);
@@ -1131,22 +1132,24 @@ static void vmw_query_bo_switch_commit(struct vmw_private *dev_priv,
 }
 
 /**
- * vmw_translate_mob_pointer - Prepare to translate a user-space buffer
- * handle to a MOB id.
+ * vmw_translate_mob_pointer - Prepare to translate a user-space buffer handle
+ * to a MOB id.
  *
  * @dev_priv: Pointer to a device private structure.
  * @sw_context: The software context used for this command batch validation.
  * @id: Pointer to the user-space handle to be translated.
- * @vmw_bo_p: Points to a location that, on successful return will carry
- * a non-reference-counted pointer to the buffer object identified by the
+ * @vmw_bo_p: Points to a location that, on successful return will carry a
+ * non-reference-counted pointer to the buffer object identified by the
  * user-space handle in @id.
  *
  * This function saves information needed to translate a user-space buffer
  * handle to a MOB id. The translation does not take place immediately, but
- * during a call to vmw_apply_relocations(). This function builds a relocation
- * list and a list of buffers to validate. The former needs to be freed using
- * either vmw_apply_relocations() or vmw_free_relocations(). The latter
- * needs to be freed using vmw_clear_validations.
+ * during a call to vmw_apply_relocations().
+ *
+ * This function builds a relocation list and a list of buffers to validate. The
+ * former needs to be freed using either vmw_apply_relocations() or
+ * vmw_free_relocations(). The latter needs to be freed using
+ * vmw_clear_validations.
  */
 static int vmw_translate_mob_ptr(struct vmw_private *dev_priv,
 				 struct vmw_sw_context *sw_context,
@@ -1161,7 +1164,7 @@ static int vmw_translate_mob_ptr(struct vmw_private *dev_priv,
 	vmw_validation_preload_bo(sw_context->ctx);
 	vmw_bo = vmw_user_bo_noref_lookup(sw_context->fp->tfile, handle);
 	if (IS_ERR(vmw_bo)) {
-		DRM_ERROR("Could not find or use MOB buffer.\n");
+		VMW_DEBUG_USER("Could not find or use MOB buffer.\n");
 		return PTR_ERR(vmw_bo);
 	}
 
@@ -1184,19 +1187,20 @@ static int vmw_translate_mob_ptr(struct vmw_private *dev_priv,
 }
 
 /**
- * vmw_translate_guest_pointer - Prepare to translate a user-space buffer
- * handle to a valid SVGAGuestPtr
+ * vmw_translate_guest_pointer - Prepare to translate a user-space buffer handle
+ * to a valid SVGAGuestPtr
  *
  * @dev_priv: Pointer to a device private structure.
  * @sw_context: The software context used for this command batch validation.
  * @ptr: Pointer to the user-space handle to be translated.
- * @vmw_bo_p: Points to a location that, on successful return will carry
- * a non-reference-counted pointer to the DMA buffer identified by the
- * user-space handle in @id.
+ * @vmw_bo_p: Points to a location that, on successful return will carry a
+ * non-reference-counted pointer to the DMA buffer identified by the user-space
+ * handle in @id.
  *
  * This function saves information needed to translate a user-space buffer
  * handle to a valid SVGAGuestPtr. The translation does not take place
  * immediately, but during a call to vmw_apply_relocations().
+ *
  * This function builds a relocation list and a list of buffers to validate.
  * The former needs to be freed using either vmw_apply_relocations() or
  * vmw_free_relocations(). The latter needs to be freed using
@@ -1215,7 +1219,7 @@ static int vmw_translate_guest_ptr(struct vmw_private *dev_priv,
 	vmw_validation_preload_bo(sw_context->ctx);
 	vmw_bo = vmw_user_bo_noref_lookup(sw_context->fp->tfile, handle);
 	if (IS_ERR(vmw_bo)) {
-		DRM_ERROR("Could not find or use GMR region.\n");
+		VMW_DEBUG_USER("Could not find or use GMR region.\n");
 		return PTR_ERR(vmw_bo);
 	}
 
@@ -1236,10 +1240,8 @@ static int vmw_translate_guest_ptr(struct vmw_private *dev_priv,
 	return 0;
 }
 
-
-
 /**
- * vmw_cmd_dx_define_query - validate a SVGA_3D_CMD_DX_DEFINE_QUERY command.
+ * vmw_cmd_dx_define_query - validate SVGA_3D_CMD_DX_DEFINE_QUERY command.
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context used for this command submission.
@@ -1251,67 +1253,52 @@ static int vmw_cmd_dx_define_query(struct vmw_private *dev_priv,
 				   struct vmw_sw_context *sw_context,
 				   SVGA3dCmdHeader *header)
 {
-	struct vmw_dx_define_query_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdDXDefineQuery q;
-	} *cmd;
-
-	int    ret;
-	struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node;
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXDefineQuery);
+	struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context);
 	struct vmw_resource *cotable_res;
+	int ret;
 
-
-	if (ctx_node == NULL) {
-		DRM_ERROR("DX Context not set for query.\n");
+	if (!ctx_node)
 		return -EINVAL;
-	}
 
-	cmd = container_of(header, struct vmw_dx_define_query_cmd, header);
+	cmd = container_of(header, typeof(*cmd), header);
 
-	if (cmd->q.type <  SVGA3D_QUERYTYPE_MIN ||
-	    cmd->q.type >= SVGA3D_QUERYTYPE_MAX)
+	if (cmd->body.type <  SVGA3D_QUERYTYPE_MIN ||
+	    cmd->body.type >= SVGA3D_QUERYTYPE_MAX)
 		return -EINVAL;
 
 	cotable_res = vmw_context_cotable(ctx_node->ctx, SVGA_COTABLE_DXQUERY);
-	ret = vmw_cotable_notify(cotable_res, cmd->q.queryId);
+	ret = vmw_cotable_notify(cotable_res, cmd->body.queryId);
 
 	return ret;
 }
 
-
-
 /**
- * vmw_cmd_dx_bind_query - validate a SVGA_3D_CMD_DX_BIND_QUERY command.
+ * vmw_cmd_dx_bind_query - validate SVGA_3D_CMD_DX_BIND_QUERY command.
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context used for this command submission.
  * @header: Pointer to the command header in the command stream.
  *
- * The query bind operation will eventually associate the query ID
- * with its backing MOB.  In this function, we take the user mode
- * MOB ID and use vmw_translate_mob_ptr() to translate it to its
- * kernel mode equivalent.
+ * The query bind operation will eventually associate the query ID with its
+ * backing MOB.  In this function, we take the user mode MOB ID and use
+ * vmw_translate_mob_ptr() to translate it to its kernel mode equivalent.
  */
 static int vmw_cmd_dx_bind_query(struct vmw_private *dev_priv,
 				 struct vmw_sw_context *sw_context,
 				 SVGA3dCmdHeader *header)
 {
-	struct vmw_dx_bind_query_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdDXBindQuery q;
-	} *cmd;
-
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXBindQuery);
 	struct vmw_buffer_object *vmw_bo;
-	int    ret;
-
+	int ret;
 
-	cmd = container_of(header, struct vmw_dx_bind_query_cmd, header);
+	cmd = container_of(header, typeof(*cmd), header);
 
 	/*
 	 * Look up the buffer pointed to by q.mobid, put it on the relocation
 	 * list so its kernel mode MOB ID can be filled in later
 	 */
-	ret = vmw_translate_mob_ptr(dev_priv, sw_context, &cmd->q.mobid,
+	ret = vmw_translate_mob_ptr(dev_priv, sw_context, &cmd->body.mobid,
 				    &vmw_bo);
 
 	if (ret != 0)
@@ -1322,10 +1309,8 @@ static int vmw_cmd_dx_bind_query(struct vmw_private *dev_priv,
 	return 0;
 }
 
-
-
 /**
- * vmw_cmd_begin_gb_query - validate a  SVGA_3D_CMD_BEGIN_GB_QUERY command.
+ * vmw_cmd_begin_gb_query - validate SVGA_3D_CMD_BEGIN_GB_QUERY command.
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context used for this command submission.
@@ -1335,21 +1320,16 @@ static int vmw_cmd_begin_gb_query(struct vmw_private *dev_priv,
 				  struct vmw_sw_context *sw_context,
 				  SVGA3dCmdHeader *header)
 {
-	struct vmw_begin_gb_query_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdBeginGBQuery q;
-	} *cmd;
-
-	cmd = container_of(header, struct vmw_begin_gb_query_cmd,
-			   header);
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdBeginGBQuery) =
+		container_of(header, typeof(*cmd), header);
 
 	return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context,
-				 user_context_converter, &cmd->q.cid,
-				 NULL);
+				 VMW_RES_DIRTY_SET, user_context_converter,
+				 &cmd->body.cid, NULL);
 }
 
 /**
- * vmw_cmd_begin_query - validate a  SVGA_3D_CMD_BEGIN_QUERY command.
+ * vmw_cmd_begin_query - validate SVGA_3D_CMD_BEGIN_QUERY command.
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context used for this command submission.
@@ -1359,38 +1339,30 @@ static int vmw_cmd_begin_query(struct vmw_private *dev_priv,
 			       struct vmw_sw_context *sw_context,
 			       SVGA3dCmdHeader *header)
 {
-	struct vmw_begin_query_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdBeginQuery q;
-	} *cmd;
-
-	cmd = container_of(header, struct vmw_begin_query_cmd,
-			   header);
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdBeginQuery) =
+		container_of(header, typeof(*cmd), header);
 
 	if (unlikely(dev_priv->has_mob)) {
-		struct {
-			SVGA3dCmdHeader header;
-			SVGA3dCmdBeginGBQuery q;
-		} gb_cmd;
+		VMW_DECLARE_CMD_VAR(gb_cmd, SVGA3dCmdBeginGBQuery);
 
 		BUG_ON(sizeof(gb_cmd) != sizeof(*cmd));
 
 		gb_cmd.header.id = SVGA_3D_CMD_BEGIN_GB_QUERY;
 		gb_cmd.header.size = cmd->header.size;
-		gb_cmd.q.cid = cmd->q.cid;
-		gb_cmd.q.type = cmd->q.type;
+		gb_cmd.body.cid = cmd->body.cid;
+		gb_cmd.body.type = cmd->body.type;
 
 		memcpy(cmd, &gb_cmd, sizeof(*cmd));
 		return vmw_cmd_begin_gb_query(dev_priv, sw_context, header);
 	}
 
 	return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context,
-				 user_context_converter, &cmd->q.cid,
-				 NULL);
+				 VMW_RES_DIRTY_SET, user_context_converter,
+				 &cmd->body.cid, NULL);
 }
 
 /**
- * vmw_cmd_end_gb_query - validate a  SVGA_3D_CMD_END_GB_QUERY command.
+ * vmw_cmd_end_gb_query - validate SVGA_3D_CMD_END_GB_QUERY command.
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context used for this command submission.
@@ -1401,19 +1373,15 @@ static int vmw_cmd_end_gb_query(struct vmw_private *dev_priv,
 				SVGA3dCmdHeader *header)
 {
 	struct vmw_buffer_object *vmw_bo;
-	struct vmw_query_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdEndGBQuery q;
-	} *cmd;
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdEndGBQuery);
 	int ret;
 
-	cmd = container_of(header, struct vmw_query_cmd, header);
+	cmd = container_of(header, typeof(*cmd), header);
 	ret = vmw_cmd_cid_check(dev_priv, sw_context, header);
 	if (unlikely(ret != 0))
 		return ret;
 
-	ret = vmw_translate_mob_ptr(dev_priv, sw_context,
-				    &cmd->q.mobid,
+	ret = vmw_translate_mob_ptr(dev_priv, sw_context, &cmd->body.mobid,
 				    &vmw_bo);
 	if (unlikely(ret != 0))
 		return ret;
@@ -1424,7 +1392,7 @@ static int vmw_cmd_end_gb_query(struct vmw_private *dev_priv,
 }
 
 /**
- * vmw_cmd_end_query - validate a  SVGA_3D_CMD_END_QUERY command.
+ * vmw_cmd_end_query - validate SVGA_3D_CMD_END_QUERY command.
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context used for this command submission.
@@ -1435,27 +1403,21 @@ static int vmw_cmd_end_query(struct vmw_private *dev_priv,
 			     SVGA3dCmdHeader *header)
 {
 	struct vmw_buffer_object *vmw_bo;
-	struct vmw_query_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdEndQuery q;
-	} *cmd;
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdEndQuery);
 	int ret;
 
-	cmd = container_of(header, struct vmw_query_cmd, header);
+	cmd = container_of(header, typeof(*cmd), header);
 	if (dev_priv->has_mob) {
-		struct {
-			SVGA3dCmdHeader header;
-			SVGA3dCmdEndGBQuery q;
-		} gb_cmd;
+		VMW_DECLARE_CMD_VAR(gb_cmd, SVGA3dCmdEndGBQuery);
 
 		BUG_ON(sizeof(gb_cmd) != sizeof(*cmd));
 
 		gb_cmd.header.id = SVGA_3D_CMD_END_GB_QUERY;
 		gb_cmd.header.size = cmd->header.size;
-		gb_cmd.q.cid = cmd->q.cid;
-		gb_cmd.q.type = cmd->q.type;
-		gb_cmd.q.mobid = cmd->q.guestResult.gmrId;
-		gb_cmd.q.offset = cmd->q.guestResult.offset;
+		gb_cmd.body.cid = cmd->body.cid;
+		gb_cmd.body.type = cmd->body.type;
+		gb_cmd.body.mobid = cmd->body.guestResult.gmrId;
+		gb_cmd.body.offset = cmd->body.guestResult.offset;
 
 		memcpy(cmd, &gb_cmd, sizeof(*cmd));
 		return vmw_cmd_end_gb_query(dev_priv, sw_context, header);
@@ -1466,8 +1428,7 @@ static int vmw_cmd_end_query(struct vmw_private *dev_priv,
 		return ret;
 
 	ret = vmw_translate_guest_ptr(dev_priv, sw_context,
-				      &cmd->q.guestResult,
-				      &vmw_bo);
+				      &cmd->body.guestResult, &vmw_bo);
 	if (unlikely(ret != 0))
 		return ret;
 
@@ -1477,7 +1438,7 @@ static int vmw_cmd_end_query(struct vmw_private *dev_priv,
 }
 
 /**
- * vmw_cmd_wait_gb_query - validate a  SVGA_3D_CMD_WAIT_GB_QUERY command.
+ * vmw_cmd_wait_gb_query - validate SVGA_3D_CMD_WAIT_GB_QUERY command.
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context used for this command submission.
@@ -1488,19 +1449,15 @@ static int vmw_cmd_wait_gb_query(struct vmw_private *dev_priv,
 				 SVGA3dCmdHeader *header)
 {
 	struct vmw_buffer_object *vmw_bo;
-	struct vmw_query_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdWaitForGBQuery q;
-	} *cmd;
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdWaitForGBQuery);
 	int ret;
 
-	cmd = container_of(header, struct vmw_query_cmd, header);
+	cmd = container_of(header, typeof(*cmd), header);
 	ret = vmw_cmd_cid_check(dev_priv, sw_context, header);
 	if (unlikely(ret != 0))
 		return ret;
 
-	ret = vmw_translate_mob_ptr(dev_priv, sw_context,
-				    &cmd->q.mobid,
+	ret = vmw_translate_mob_ptr(dev_priv, sw_context, &cmd->body.mobid,
 				    &vmw_bo);
 	if (unlikely(ret != 0))
 		return ret;
@@ -1509,7 +1466,7 @@ static int vmw_cmd_wait_gb_query(struct vmw_private *dev_priv,
 }
 
 /**
- * vmw_cmd_wait_query - validate a  SVGA_3D_CMD_WAIT_QUERY command.
+ * vmw_cmd_wait_query - validate SVGA_3D_CMD_WAIT_QUERY command.
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context used for this command submission.
@@ -1520,27 +1477,21 @@ static int vmw_cmd_wait_query(struct vmw_private *dev_priv,
 			      SVGA3dCmdHeader *header)
 {
 	struct vmw_buffer_object *vmw_bo;
-	struct vmw_query_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdWaitForQuery q;
-	} *cmd;
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdWaitForQuery);
 	int ret;
 
-	cmd = container_of(header, struct vmw_query_cmd, header);
+	cmd = container_of(header, typeof(*cmd), header);
 	if (dev_priv->has_mob) {
-		struct {
-			SVGA3dCmdHeader header;
-			SVGA3dCmdWaitForGBQuery q;
-		} gb_cmd;
+		VMW_DECLARE_CMD_VAR(gb_cmd, SVGA3dCmdWaitForGBQuery);
 
 		BUG_ON(sizeof(gb_cmd) != sizeof(*cmd));
 
 		gb_cmd.header.id = SVGA_3D_CMD_WAIT_FOR_GB_QUERY;
 		gb_cmd.header.size = cmd->header.size;
-		gb_cmd.q.cid = cmd->q.cid;
-		gb_cmd.q.type = cmd->q.type;
-		gb_cmd.q.mobid = cmd->q.guestResult.gmrId;
-		gb_cmd.q.offset = cmd->q.guestResult.offset;
+		gb_cmd.body.cid = cmd->body.cid;
+		gb_cmd.body.type = cmd->body.type;
+		gb_cmd.body.mobid = cmd->body.guestResult.gmrId;
+		gb_cmd.body.offset = cmd->body.guestResult.offset;
 
 		memcpy(cmd, &gb_cmd, sizeof(*cmd));
 		return vmw_cmd_wait_gb_query(dev_priv, sw_context, header);
@@ -1551,8 +1502,7 @@ static int vmw_cmd_wait_query(struct vmw_private *dev_priv,
 		return ret;
 
 	ret = vmw_translate_guest_ptr(dev_priv, sw_context,
-				      &cmd->q.guestResult,
-				      &vmw_bo);
+				      &cmd->body.guestResult, &vmw_bo);
 	if (unlikely(ret != 0))
 		return ret;
 
@@ -1565,54 +1515,52 @@ static int vmw_cmd_dma(struct vmw_private *dev_priv,
 {
 	struct vmw_buffer_object *vmw_bo = NULL;
 	struct vmw_surface *srf = NULL;
-	struct vmw_dma_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdSurfaceDMA dma;
-	} *cmd;
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdSurfaceDMA);
 	int ret;
 	SVGA3dCmdSurfaceDMASuffix *suffix;
 	uint32_t bo_size;
+	bool dirty;
 
-	cmd = container_of(header, struct vmw_dma_cmd, header);
-	suffix = (SVGA3dCmdSurfaceDMASuffix *)((unsigned long) &cmd->dma +
+	cmd = container_of(header, typeof(*cmd), header);
+	suffix = (SVGA3dCmdSurfaceDMASuffix *)((unsigned long) &cmd->body +
 					       header->size - sizeof(*suffix));
 
 	/* Make sure device and verifier stays in sync. */
 	if (unlikely(suffix->suffixSize != sizeof(*suffix))) {
-		DRM_ERROR("Invalid DMA suffix size.\n");
+		VMW_DEBUG_USER("Invalid DMA suffix size.\n");
 		return -EINVAL;
 	}
 
 	ret = vmw_translate_guest_ptr(dev_priv, sw_context,
-				      &cmd->dma.guest.ptr,
-				      &vmw_bo);
+				      &cmd->body.guest.ptr, &vmw_bo);
 	if (unlikely(ret != 0))
 		return ret;
 
 	/* Make sure DMA doesn't cross BO boundaries. */
 	bo_size = vmw_bo->base.num_pages * PAGE_SIZE;
-	if (unlikely(cmd->dma.guest.ptr.offset > bo_size)) {
-		DRM_ERROR("Invalid DMA offset.\n");
+	if (unlikely(cmd->body.guest.ptr.offset > bo_size)) {
+		VMW_DEBUG_USER("Invalid DMA offset.\n");
 		return -EINVAL;
 	}
 
-	bo_size -= cmd->dma.guest.ptr.offset;
+	bo_size -= cmd->body.guest.ptr.offset;
 	if (unlikely(suffix->maximumOffset > bo_size))
 		suffix->maximumOffset = bo_size;
 
+	dirty = (cmd->body.transfer == SVGA3D_WRITE_HOST_VRAM) ?
+		VMW_RES_DIRTY_SET : 0;
 	ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				user_surface_converter, &cmd->dma.host.sid,
-				NULL);
+				dirty, user_surface_converter,
+				&cmd->body.host.sid, NULL);
 	if (unlikely(ret != 0)) {
 		if (unlikely(ret != -ERESTARTSYS))
-			DRM_ERROR("could not find surface for DMA.\n");
+			VMW_DEBUG_USER("could not find surface for DMA.\n");
 		return ret;
 	}
 
 	srf = vmw_res_to_srf(sw_context->res_cache[vmw_res_surface].res);
 
-	vmw_kms_cursor_snoop(srf, sw_context->fp->tfile, &vmw_bo->base,
-			     header);
+	vmw_kms_cursor_snoop(srf, sw_context->fp->tfile, &vmw_bo->base, header);
 
 	return 0;
 }
@@ -1621,10 +1569,7 @@ static int vmw_cmd_draw(struct vmw_private *dev_priv,
 			struct vmw_sw_context *sw_context,
 			SVGA3dCmdHeader *header)
 {
-	struct vmw_draw_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdDrawPrimitives body;
-	} *cmd;
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDrawPrimitives);
 	SVGA3dVertexDecl *decl = (SVGA3dVertexDecl *)(
 		(unsigned long)header + sizeof(*cmd));
 	SVGA3dPrimitiveRange *range;
@@ -1636,16 +1581,17 @@ static int vmw_cmd_draw(struct vmw_private *dev_priv,
 	if (unlikely(ret != 0))
 		return ret;
 
-	cmd = container_of(header, struct vmw_draw_cmd, header);
+	cmd = container_of(header, typeof(*cmd), header);
 	maxnum = (header->size - sizeof(cmd->body)) / sizeof(*decl);
 
 	if (unlikely(cmd->body.numVertexDecls > maxnum)) {
-		DRM_ERROR("Illegal number of vertex declarations.\n");
+		VMW_DEBUG_USER("Illegal number of vertex declarations.\n");
 		return -EINVAL;
 	}
 
 	for (i = 0; i < cmd->body.numVertexDecls; ++i, ++decl) {
 		ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
+					VMW_RES_DIRTY_NONE,
 					user_surface_converter,
 					&decl->array.surfaceId, NULL);
 		if (unlikely(ret != 0))
@@ -1655,13 +1601,14 @@ static int vmw_cmd_draw(struct vmw_private *dev_priv,
 	maxnum = (header->size - sizeof(cmd->body) -
 		  cmd->body.numVertexDecls * sizeof(*decl)) / sizeof(*range);
 	if (unlikely(cmd->body.numRanges > maxnum)) {
-		DRM_ERROR("Illegal number of index ranges.\n");
+		VMW_DEBUG_USER("Illegal number of index ranges.\n");
 		return -EINVAL;
 	}
 
 	range = (SVGA3dPrimitiveRange *) decl;
 	for (i = 0; i < cmd->body.numRanges; ++i, ++range) {
 		ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
+					VMW_RES_DIRTY_NONE,
 					user_surface_converter,
 					&range->indexArray.surfaceId, NULL);
 		if (unlikely(ret != 0))
@@ -1670,30 +1617,24 @@ static int vmw_cmd_draw(struct vmw_private *dev_priv,
 	return 0;
 }
 
-
 static int vmw_cmd_tex_state(struct vmw_private *dev_priv,
 			     struct vmw_sw_context *sw_context,
 			     SVGA3dCmdHeader *header)
 {
-	struct vmw_tex_state_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdSetTextureState state;
-	} *cmd;
-
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdSetTextureState);
 	SVGA3dTextureState *last_state = (SVGA3dTextureState *)
 	  ((unsigned long) header + header->size + sizeof(header));
 	SVGA3dTextureState *cur_state = (SVGA3dTextureState *)
-		((unsigned long) header + sizeof(struct vmw_tex_state_cmd));
+		((unsigned long) header + sizeof(*cmd));
 	struct vmw_resource *ctx;
 	struct vmw_resource *res;
 	int ret;
 
-	cmd = container_of(header, struct vmw_tex_state_cmd,
-			   header);
+	cmd = container_of(header, typeof(*cmd), header);
 
 	ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context,
-				user_context_converter, &cmd->state.cid,
-				&ctx);
+				VMW_RES_DIRTY_SET, user_context_converter,
+				&cmd->body.cid, &ctx);
 	if (unlikely(ret != 0))
 		return ret;
 
@@ -1702,12 +1643,13 @@ static int vmw_cmd_tex_state(struct vmw_private *dev_priv,
 			continue;
 
 		if (cur_state->stage >= SVGA3D_NUM_TEXTURE_UNITS) {
-			DRM_ERROR("Illegal texture/sampler unit %u.\n",
-				  (unsigned) cur_state->stage);
+			VMW_DEBUG_USER("Illegal texture/sampler unit %u.\n",
+				       (unsigned int) cur_state->stage);
 			return -EINVAL;
 		}
 
 		ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
+					VMW_RES_DIRTY_NONE,
 					user_surface_converter,
 					&cur_state->value, &res);
 		if (unlikely(ret != 0))
@@ -1744,12 +1686,10 @@ static int vmw_cmd_check_define_gmrfb(struct vmw_private *dev_priv,
 		SVGAFifoCmdDefineGMRFB body;
 	} *cmd = buf;
 
-	return vmw_translate_guest_ptr(dev_priv, sw_context,
-				       &cmd->body.ptr,
+	return vmw_translate_guest_ptr(dev_priv, sw_context, &cmd->body.ptr,
 				       &vmw_bo);
 }
 
-
 /**
  * vmw_cmd_res_switch_backup - Utility function to handle backup buffer
  * switching
@@ -1761,14 +1701,13 @@ static int vmw_cmd_check_define_gmrfb(struct vmw_private *dev_priv,
  * stream.
  * @backup_offset: Offset of backup into MOB.
  *
- * This function prepares for registering a switch of backup buffers
- * in the resource metadata just prior to unreserving. It's basically a wrapper
- * around vmw_cmd_res_switch_backup with a different interface.
+ * This function prepares for registering a switch of backup buffers in the
+ * resource metadata just prior to unreserving. It's basically a wrapper around
+ * vmw_cmd_res_switch_backup with a different interface.
  */
 static int vmw_cmd_res_switch_backup(struct vmw_private *dev_priv,
 				     struct vmw_sw_context *sw_context,
-				     struct vmw_resource *res,
-				     uint32_t *buf_id,
+				     struct vmw_resource *res, uint32_t *buf_id,
 				     unsigned long backup_offset)
 {
 	struct vmw_buffer_object *vbo;
@@ -1788,7 +1727,6 @@ static int vmw_cmd_res_switch_backup(struct vmw_private *dev_priv,
 	return 0;
 }
 
-
 /**
  * vmw_cmd_switch_backup - Utility function to handle backup buffer switching
  *
@@ -1801,34 +1739,31 @@ static int vmw_cmd_res_switch_backup(struct vmw_private *dev_priv,
  * stream.
  * @backup_offset: Offset of backup into MOB.
  *
- * This function prepares for registering a switch of backup buffers
- * in the resource metadata just prior to unreserving. It's basically a wrapper
- * around vmw_cmd_res_switch_backup with a different interface.
+ * This function prepares for registering a switch of backup buffers in the
+ * resource metadata just prior to unreserving. It's basically a wrapper around
+ * vmw_cmd_res_switch_backup with a different interface.
  */
 static int vmw_cmd_switch_backup(struct vmw_private *dev_priv,
 				 struct vmw_sw_context *sw_context,
 				 enum vmw_res_type res_type,
 				 const struct vmw_user_resource_conv
-				 *converter,
-				 uint32_t *res_id,
-				 uint32_t *buf_id,
+				 *converter, uint32_t *res_id, uint32_t *buf_id,
 				 unsigned long backup_offset)
 {
 	struct vmw_resource *res;
 	int ret;
 
 	ret = vmw_cmd_res_check(dev_priv, sw_context, res_type,
-				converter, res_id, &res);
+				VMW_RES_DIRTY_NONE, converter, res_id, &res);
 	if (ret)
 		return ret;
 
-	return vmw_cmd_res_switch_backup(dev_priv, sw_context, res,
-					 buf_id, backup_offset);
+	return vmw_cmd_res_switch_backup(dev_priv, sw_context, res, buf_id,
+					 backup_offset);
 }
 
 /**
- * vmw_cmd_bind_gb_surface - Validate an SVGA_3D_CMD_BIND_GB_SURFACE
- * command
+ * vmw_cmd_bind_gb_surface - Validate SVGA_3D_CMD_BIND_GB_SURFACE command
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -1838,22 +1773,16 @@ static int vmw_cmd_bind_gb_surface(struct vmw_private *dev_priv,
 				   struct vmw_sw_context *sw_context,
 				   SVGA3dCmdHeader *header)
 {
-	struct vmw_bind_gb_surface_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdBindGBSurface body;
-	} *cmd;
-
-	cmd = container_of(header, struct vmw_bind_gb_surface_cmd, header);
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdBindGBSurface) =
+		container_of(header, typeof(*cmd), header);
 
 	return vmw_cmd_switch_backup(dev_priv, sw_context, vmw_res_surface,
-				     user_surface_converter,
-				     &cmd->body.sid, &cmd->body.mobid,
-				     0);
+				     user_surface_converter, &cmd->body.sid,
+				     &cmd->body.mobid, 0);
 }
 
 /**
- * vmw_cmd_update_gb_image - Validate an SVGA_3D_CMD_UPDATE_GB_IMAGE
- * command
+ * vmw_cmd_update_gb_image - Validate SVGA_3D_CMD_UPDATE_GB_IMAGE command
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -1863,21 +1792,16 @@ static int vmw_cmd_update_gb_image(struct vmw_private *dev_priv,
 				   struct vmw_sw_context *sw_context,
 				   SVGA3dCmdHeader *header)
 {
-	struct vmw_gb_surface_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdUpdateGBImage body;
-	} *cmd;
-
-	cmd = container_of(header, struct vmw_gb_surface_cmd, header);
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdUpdateGBImage) =
+		container_of(header, typeof(*cmd), header);
 
 	return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				 user_surface_converter,
+				 VMW_RES_DIRTY_NONE, user_surface_converter,
 				 &cmd->body.image.sid, NULL);
 }
 
 /**
- * vmw_cmd_update_gb_surface - Validate an SVGA_3D_CMD_UPDATE_GB_SURFACE
- * command
+ * vmw_cmd_update_gb_surface - Validate SVGA_3D_CMD_UPDATE_GB_SURFACE command
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -1887,21 +1811,16 @@ static int vmw_cmd_update_gb_surface(struct vmw_private *dev_priv,
 				     struct vmw_sw_context *sw_context,
 				     SVGA3dCmdHeader *header)
 {
-	struct vmw_gb_surface_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdUpdateGBSurface body;
-	} *cmd;
-
-	cmd = container_of(header, struct vmw_gb_surface_cmd, header);
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdUpdateGBSurface) =
+		container_of(header, typeof(*cmd), header);
 
 	return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				 user_surface_converter,
+				 VMW_RES_DIRTY_CLEAR, user_surface_converter,
 				 &cmd->body.sid, NULL);
 }
 
 /**
- * vmw_cmd_readback_gb_image - Validate an SVGA_3D_CMD_READBACK_GB_IMAGE
- * command
+ * vmw_cmd_readback_gb_image - Validate SVGA_3D_CMD_READBACK_GB_IMAGE command
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -1911,20 +1830,16 @@ static int vmw_cmd_readback_gb_image(struct vmw_private *dev_priv,
 				     struct vmw_sw_context *sw_context,
 				     SVGA3dCmdHeader *header)
 {
-	struct vmw_gb_surface_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdReadbackGBImage body;
-	} *cmd;
-
-	cmd = container_of(header, struct vmw_gb_surface_cmd, header);
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdReadbackGBImage) =
+		container_of(header, typeof(*cmd), header);
 
 	return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				 user_surface_converter,
+				 VMW_RES_DIRTY_NONE, user_surface_converter,
 				 &cmd->body.image.sid, NULL);
 }
 
 /**
- * vmw_cmd_readback_gb_surface - Validate an SVGA_3D_CMD_READBACK_GB_SURFACE
+ * vmw_cmd_readback_gb_surface - Validate SVGA_3D_CMD_READBACK_GB_SURFACE
  * command
  *
  * @dev_priv: Pointer to a device private struct.
@@ -1935,20 +1850,16 @@ static int vmw_cmd_readback_gb_surface(struct vmw_private *dev_priv,
 				       struct vmw_sw_context *sw_context,
 				       SVGA3dCmdHeader *header)
 {
-	struct vmw_gb_surface_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdReadbackGBSurface body;
-	} *cmd;
-
-	cmd = container_of(header, struct vmw_gb_surface_cmd, header);
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdReadbackGBSurface) =
+		container_of(header, typeof(*cmd), header);
 
 	return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				 user_surface_converter,
+				 VMW_RES_DIRTY_CLEAR, user_surface_converter,
 				 &cmd->body.sid, NULL);
 }
 
 /**
- * vmw_cmd_invalidate_gb_image - Validate an SVGA_3D_CMD_INVALIDATE_GB_IMAGE
+ * vmw_cmd_invalidate_gb_image - Validate SVGA_3D_CMD_INVALIDATE_GB_IMAGE
  * command
  *
  * @dev_priv: Pointer to a device private struct.
@@ -1959,21 +1870,17 @@ static int vmw_cmd_invalidate_gb_image(struct vmw_private *dev_priv,
 				       struct vmw_sw_context *sw_context,
 				       SVGA3dCmdHeader *header)
 {
-	struct vmw_gb_surface_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdInvalidateGBImage body;
-	} *cmd;
-
-	cmd = container_of(header, struct vmw_gb_surface_cmd, header);
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdInvalidateGBImage) =
+		container_of(header, typeof(*cmd), header);
 
 	return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				 user_surface_converter,
+				 VMW_RES_DIRTY_NONE, user_surface_converter,
 				 &cmd->body.image.sid, NULL);
 }
 
 /**
- * vmw_cmd_invalidate_gb_surface - Validate an
- * SVGA_3D_CMD_INVALIDATE_GB_SURFACE command
+ * vmw_cmd_invalidate_gb_surface - Validate SVGA_3D_CMD_INVALIDATE_GB_SURFACE
+ * command
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -1983,22 +1890,16 @@ static int vmw_cmd_invalidate_gb_surface(struct vmw_private *dev_priv,
 					 struct vmw_sw_context *sw_context,
 					 SVGA3dCmdHeader *header)
 {
-	struct vmw_gb_surface_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdInvalidateGBSurface body;
-	} *cmd;
-
-	cmd = container_of(header, struct vmw_gb_surface_cmd, header);
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdInvalidateGBSurface) =
+		container_of(header, typeof(*cmd), header);
 
 	return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				 user_surface_converter,
+				 VMW_RES_DIRTY_CLEAR, user_surface_converter,
 				 &cmd->body.sid, NULL);
 }
 
-
 /**
- * vmw_cmd_shader_define - Validate an SVGA_3D_CMD_SHADER_DEFINE
- * command
+ * vmw_cmd_shader_define - Validate SVGA_3D_CMD_SHADER_DEFINE command
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -2008,20 +1909,16 @@ static int vmw_cmd_shader_define(struct vmw_private *dev_priv,
 				 struct vmw_sw_context *sw_context,
 				 SVGA3dCmdHeader *header)
 {
-	struct vmw_shader_define_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdDefineShader body;
-	} *cmd;
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDefineShader);
 	int ret;
 	size_t size;
 	struct vmw_resource *ctx;
 
-	cmd = container_of(header, struct vmw_shader_define_cmd,
-			   header);
+	cmd = container_of(header, typeof(*cmd), header);
 
 	ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context,
-				user_context_converter, &cmd->body.cid,
-				&ctx);
+				VMW_RES_DIRTY_SET, user_context_converter,
+				&cmd->body.cid, &ctx);
 	if (unlikely(ret != 0))
 		return ret;
 
@@ -2029,24 +1926,20 @@ static int vmw_cmd_shader_define(struct vmw_private *dev_priv,
 		return 0;
 
 	size = cmd->header.size - sizeof(cmd->body);
-	ret = vmw_compat_shader_add(dev_priv,
-				    vmw_context_res_man(ctx),
-				    cmd->body.shid, cmd + 1,
-				    cmd->body.type, size,
-				    &sw_context->staged_cmd_res);
+	ret = vmw_compat_shader_add(dev_priv, vmw_context_res_man(ctx),
+				    cmd->body.shid, cmd + 1, cmd->body.type,
+				    size, &sw_context->staged_cmd_res);
 	if (unlikely(ret != 0))
 		return ret;
 
-	return vmw_resource_relocation_add(sw_context,
-					   NULL,
+	return vmw_resource_relocation_add(sw_context, NULL,
 					   vmw_ptr_diff(sw_context->buf_start,
 							&cmd->header.id),
 					   vmw_res_rel_nop);
 }
 
 /**
- * vmw_cmd_shader_destroy - Validate an SVGA_3D_CMD_SHADER_DESTROY
- * command
+ * vmw_cmd_shader_destroy - Validate SVGA_3D_CMD_SHADER_DESTROY command
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -2056,42 +1949,34 @@ static int vmw_cmd_shader_destroy(struct vmw_private *dev_priv,
 				  struct vmw_sw_context *sw_context,
 				  SVGA3dCmdHeader *header)
 {
-	struct vmw_shader_destroy_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdDestroyShader body;
-	} *cmd;
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDestroyShader);
 	int ret;
 	struct vmw_resource *ctx;
 
-	cmd = container_of(header, struct vmw_shader_destroy_cmd,
-			   header);
+	cmd = container_of(header, typeof(*cmd), header);
 
 	ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context,
-				user_context_converter, &cmd->body.cid,
-				&ctx);
+				VMW_RES_DIRTY_SET, user_context_converter,
+				&cmd->body.cid, &ctx);
 	if (unlikely(ret != 0))
 		return ret;
 
 	if (unlikely(!dev_priv->has_mob))
 		return 0;
 
-	ret = vmw_shader_remove(vmw_context_res_man(ctx),
-				cmd->body.shid,
-				cmd->body.type,
-				&sw_context->staged_cmd_res);
+	ret = vmw_shader_remove(vmw_context_res_man(ctx), cmd->body.shid,
+				cmd->body.type, &sw_context->staged_cmd_res);
 	if (unlikely(ret != 0))
 		return ret;
 
-	return vmw_resource_relocation_add(sw_context,
-					   NULL,
+	return vmw_resource_relocation_add(sw_context, NULL,
 					   vmw_ptr_diff(sw_context->buf_start,
 							&cmd->header.id),
 					   vmw_res_rel_nop);
 }
 
 /**
- * vmw_cmd_set_shader - Validate an SVGA_3D_CMD_SET_SHADER
- * command
+ * vmw_cmd_set_shader - Validate SVGA_3D_CMD_SET_SHADER command
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -2101,27 +1986,23 @@ static int vmw_cmd_set_shader(struct vmw_private *dev_priv,
 			      struct vmw_sw_context *sw_context,
 			      SVGA3dCmdHeader *header)
 {
-	struct vmw_set_shader_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdSetShader body;
-	} *cmd;
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdSetShader);
 	struct vmw_ctx_bindinfo_shader binding;
 	struct vmw_resource *ctx, *res = NULL;
 	struct vmw_ctx_validation_info *ctx_info;
 	int ret;
 
-	cmd = container_of(header, struct vmw_set_shader_cmd,
-			   header);
+	cmd = container_of(header, typeof(*cmd), header);
 
 	if (cmd->body.type >= SVGA3D_SHADERTYPE_PREDX_MAX) {
-		DRM_ERROR("Illegal shader type %u.\n",
-			  (unsigned) cmd->body.type);
+		VMW_DEBUG_USER("Illegal shader type %u.\n",
+			       (unsigned int) cmd->body.type);
 		return -EINVAL;
 	}
 
 	ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context,
-				user_context_converter, &cmd->body.cid,
-				&ctx);
+				VMW_RES_DIRTY_SET, user_context_converter,
+				&cmd->body.cid, &ctx);
 	if (unlikely(ret != 0))
 		return ret;
 
@@ -2130,21 +2011,20 @@ static int vmw_cmd_set_shader(struct vmw_private *dev_priv,
 
 	if (cmd->body.shid != SVGA3D_INVALID_ID) {
 		res = vmw_shader_lookup(vmw_context_res_man(ctx),
-					cmd->body.shid,
-					cmd->body.type);
-
+					cmd->body.shid, cmd->body.type);
 		if (!IS_ERR(res)) {
-			ret = vmw_execbuf_res_noctx_val_add(sw_context, res);
+			ret = vmw_execbuf_res_noctx_val_add(sw_context, res,
+							    VMW_RES_DIRTY_NONE);
 			if (unlikely(ret != 0))
 				return ret;
 		}
 	}
 
 	if (IS_ERR_OR_NULL(res)) {
-		ret = vmw_cmd_res_check(dev_priv, sw_context,
-					vmw_res_shader,
-					user_shader_converter,
-					&cmd->body.shid, &res);
+		ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_shader,
+					VMW_RES_DIRTY_NONE,
+					user_shader_converter, &cmd->body.shid,
+					&res);
 		if (unlikely(ret != 0))
 			return ret;
 	}
@@ -2157,14 +2037,13 @@ static int vmw_cmd_set_shader(struct vmw_private *dev_priv,
 	binding.bi.res = res;
 	binding.bi.bt = vmw_ctx_binding_shader;
 	binding.shader_slot = cmd->body.type - SVGA3D_SHADERTYPE_MIN;
-	vmw_binding_add(ctx_info->staged, &binding.bi,
-			binding.shader_slot, 0);
+	vmw_binding_add(ctx_info->staged, &binding.bi, binding.shader_slot, 0);
+
 	return 0;
 }
 
 /**
- * vmw_cmd_set_shader_const - Validate an SVGA_3D_CMD_SET_SHADER_CONST
- * command
+ * vmw_cmd_set_shader_const - Validate SVGA_3D_CMD_SET_SHADER_CONST command
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -2174,18 +2053,14 @@ static int vmw_cmd_set_shader_const(struct vmw_private *dev_priv,
 				    struct vmw_sw_context *sw_context,
 				    SVGA3dCmdHeader *header)
 {
-	struct vmw_set_shader_const_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdSetShaderConst body;
-	} *cmd;
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdSetShaderConst);
 	int ret;
 
-	cmd = container_of(header, struct vmw_set_shader_const_cmd,
-			   header);
+	cmd = container_of(header, typeof(*cmd), header);
 
 	ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context,
-				user_context_converter, &cmd->body.cid,
-				NULL);
+				VMW_RES_DIRTY_SET, user_context_converter,
+				&cmd->body.cid, NULL);
 	if (unlikely(ret != 0))
 		return ret;
 
@@ -2196,8 +2071,7 @@ static int vmw_cmd_set_shader_const(struct vmw_private *dev_priv,
 }
 
 /**
- * vmw_cmd_bind_gb_shader - Validate an SVGA_3D_CMD_BIND_GB_SHADER
- * command
+ * vmw_cmd_bind_gb_shader - Validate SVGA_3D_CMD_BIND_GB_SHADER command
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -2207,22 +2081,16 @@ static int vmw_cmd_bind_gb_shader(struct vmw_private *dev_priv,
 				  struct vmw_sw_context *sw_context,
 				  SVGA3dCmdHeader *header)
 {
-	struct vmw_bind_gb_shader_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdBindGBShader body;
-	} *cmd;
-
-	cmd = container_of(header, struct vmw_bind_gb_shader_cmd,
-			   header);
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdBindGBShader) =
+		container_of(header, typeof(*cmd), header);
 
 	return vmw_cmd_switch_backup(dev_priv, sw_context, vmw_res_shader,
-				     user_shader_converter,
-				     &cmd->body.shid, &cmd->body.mobid,
-				     cmd->body.offsetInBytes);
+				     user_shader_converter, &cmd->body.shid,
+				     &cmd->body.mobid, cmd->body.offsetInBytes);
 }
 
 /**
- * vmw_cmd_dx_set_single_constant_buffer - Validate an
+ * vmw_cmd_dx_set_single_constant_buffer - Validate
  * SVGA_3D_CMD_DX_SET_SINGLE_CONSTANT_BUFFER command.
  *
  * @dev_priv: Pointer to a device private struct.
@@ -2234,23 +2102,18 @@ vmw_cmd_dx_set_single_constant_buffer(struct vmw_private *dev_priv,
 				      struct vmw_sw_context *sw_context,
 				      SVGA3dCmdHeader *header)
 {
-	struct {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdDXSetSingleConstantBuffer body;
-	} *cmd;
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXSetSingleConstantBuffer);
 	struct vmw_resource *res = NULL;
-	struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node;
+	struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context);
 	struct vmw_ctx_bindinfo_cb binding;
 	int ret;
 
-	if (unlikely(ctx_node == NULL)) {
-		DRM_ERROR("DX Context not set.\n");
+	if (!ctx_node)
 		return -EINVAL;
-	}
 
 	cmd = container_of(header, typeof(*cmd), header);
 	ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				user_surface_converter,
+				VMW_RES_DIRTY_NONE, user_surface_converter,
 				&cmd->body.sid, &res);
 	if (unlikely(ret != 0))
 		return ret;
@@ -2265,21 +2128,21 @@ vmw_cmd_dx_set_single_constant_buffer(struct vmw_private *dev_priv,
 
 	if (binding.shader_slot >= SVGA3D_NUM_SHADERTYPE_DX10 ||
 	    binding.slot >= SVGA3D_DX_MAX_CONSTBUFFERS) {
-		DRM_ERROR("Illegal const buffer shader %u slot %u.\n",
-			  (unsigned) cmd->body.type,
-			  (unsigned) binding.slot);
+		VMW_DEBUG_USER("Illegal const buffer shader %u slot %u.\n",
+			       (unsigned int) cmd->body.type,
+			       (unsigned int) binding.slot);
 		return -EINVAL;
 	}
 
-	vmw_binding_add(ctx_node->staged, &binding.bi,
-			binding.shader_slot, binding.slot);
+	vmw_binding_add(ctx_node->staged, &binding.bi, binding.shader_slot,
+			binding.slot);
 
 	return 0;
 }
 
 /**
- * vmw_cmd_dx_set_shader_res - Validate an
- * SVGA_3D_CMD_DX_SET_SHADER_RESOURCES command
+ * vmw_cmd_dx_set_shader_res - Validate SVGA_3D_CMD_DX_SET_SHADER_RESOURCES
+ * command
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -2289,17 +2152,15 @@ static int vmw_cmd_dx_set_shader_res(struct vmw_private *dev_priv,
 				     struct vmw_sw_context *sw_context,
 				     SVGA3dCmdHeader *header)
 {
-	struct {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdDXSetShaderResources body;
-	} *cmd = container_of(header, typeof(*cmd), header);
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXSetShaderResources) =
+		container_of(header, typeof(*cmd), header);
 	u32 num_sr_view = (cmd->header.size - sizeof(cmd->body)) /
 		sizeof(SVGA3dShaderResourceViewId);
 
 	if ((u64) cmd->body.startView + (u64) num_sr_view >
 	    (u64) SVGA3D_DX_MAX_SRVIEWS ||
 	    cmd->body.type >= SVGA3D_SHADERTYPE_DX10_MAX) {
-		DRM_ERROR("Invalid shader binding.\n");
+		VMW_DEBUG_USER("Invalid shader binding.\n");
 		return -EINVAL;
 	}
 
@@ -2311,8 +2172,7 @@ static int vmw_cmd_dx_set_shader_res(struct vmw_private *dev_priv,
 }
 
 /**
- * vmw_cmd_dx_set_shader - Validate an SVGA_3D_CMD_DX_SET_SHADER
- * command
+ * vmw_cmd_dx_set_shader - Validate SVGA_3D_CMD_DX_SET_SHADER command
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -2322,36 +2182,32 @@ static int vmw_cmd_dx_set_shader(struct vmw_private *dev_priv,
 				 struct vmw_sw_context *sw_context,
 				 SVGA3dCmdHeader *header)
 {
-	struct {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdDXSetShader body;
-	} *cmd;
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXSetShader);
 	struct vmw_resource *res = NULL;
-	struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node;
+	struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context);
 	struct vmw_ctx_bindinfo_shader binding;
 	int ret = 0;
 
-	if (unlikely(ctx_node == NULL)) {
-		DRM_ERROR("DX Context not set.\n");
+	if (!ctx_node)
 		return -EINVAL;
-	}
 
 	cmd = container_of(header, typeof(*cmd), header);
 
 	if (cmd->body.type >= SVGA3D_SHADERTYPE_DX10_MAX) {
-		DRM_ERROR("Illegal shader type %u.\n",
-			  (unsigned) cmd->body.type);
+		VMW_DEBUG_USER("Illegal shader type %u.\n",
+			       (unsigned int) cmd->body.type);
 		return -EINVAL;
 	}
 
 	if (cmd->body.shaderId != SVGA3D_INVALID_ID) {
 		res = vmw_shader_lookup(sw_context->man, cmd->body.shaderId, 0);
 		if (IS_ERR(res)) {
-			DRM_ERROR("Could not find shader for binding.\n");
+			VMW_DEBUG_USER("Could not find shader for binding.\n");
 			return PTR_ERR(res);
 		}
 
-		ret = vmw_execbuf_res_noctx_val_add(sw_context, res);
+		ret = vmw_execbuf_res_noctx_val_add(sw_context, res,
+						    VMW_RES_DIRTY_NONE);
 		if (ret)
 			return ret;
 	}
@@ -2361,15 +2217,14 @@ static int vmw_cmd_dx_set_shader(struct vmw_private *dev_priv,
 	binding.bi.bt = vmw_ctx_binding_dx_shader;
 	binding.shader_slot = cmd->body.type - SVGA3D_SHADERTYPE_MIN;
 
-	vmw_binding_add(ctx_node->staged, &binding.bi,
-			binding.shader_slot, 0);
+	vmw_binding_add(ctx_node->staged, &binding.bi, binding.shader_slot, 0);
 
 	return 0;
 }
 
 /**
- * vmw_cmd_dx_set_vertex_buffers - Validates an
- * SVGA_3D_CMD_DX_SET_VERTEX_BUFFERS command
+ * vmw_cmd_dx_set_vertex_buffers - Validates SVGA_3D_CMD_DX_SET_VERTEX_BUFFERS
+ * command
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -2379,7 +2234,7 @@ static int vmw_cmd_dx_set_vertex_buffers(struct vmw_private *dev_priv,
 					 struct vmw_sw_context *sw_context,
 					 SVGA3dCmdHeader *header)
 {
-	struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node;
+	struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context);
 	struct vmw_ctx_bindinfo_vb binding;
 	struct vmw_resource *res;
 	struct {
@@ -2389,22 +2244,21 @@ static int vmw_cmd_dx_set_vertex_buffers(struct vmw_private *dev_priv,
 	} *cmd;
 	int i, ret, num;
 
-	if (unlikely(ctx_node == NULL)) {
-		DRM_ERROR("DX Context not set.\n");
+	if (!ctx_node)
 		return -EINVAL;
-	}
 
 	cmd = container_of(header, typeof(*cmd), header);
 	num = (cmd->header.size - sizeof(cmd->body)) /
 		sizeof(SVGA3dVertexBuffer);
 	if ((u64)num + (u64)cmd->body.startBuffer >
 	    (u64)SVGA3D_DX_MAX_VERTEXBUFFERS) {
-		DRM_ERROR("Invalid number of vertex buffers.\n");
+		VMW_DEBUG_USER("Invalid number of vertex buffers.\n");
 		return -EINVAL;
 	}
 
 	for (i = 0; i < num; i++) {
 		ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
+					VMW_RES_DIRTY_NONE,
 					user_surface_converter,
 					&cmd->buf[i].sid, &res);
 		if (unlikely(ret != 0))
@@ -2417,15 +2271,14 @@ static int vmw_cmd_dx_set_vertex_buffers(struct vmw_private *dev_priv,
 		binding.stride = cmd->buf[i].stride;
 		binding.slot = i + cmd->body.startBuffer;
 
-		vmw_binding_add(ctx_node->staged, &binding.bi,
-				0, binding.slot);
+		vmw_binding_add(ctx_node->staged, &binding.bi, 0, binding.slot);
 	}
 
 	return 0;
 }
 
 /**
- * vmw_cmd_dx_ia_set_vertex_buffers - Validate an
+ * vmw_cmd_dx_ia_set_vertex_buffers - Validate
  * SVGA_3D_CMD_DX_IA_SET_INDEX_BUFFER command.
  *
  * @dev_priv: Pointer to a device private struct.
@@ -2436,23 +2289,18 @@ static int vmw_cmd_dx_set_index_buffer(struct vmw_private *dev_priv,
 				       struct vmw_sw_context *sw_context,
 				       SVGA3dCmdHeader *header)
 {
-	struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node;
+	struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context);
 	struct vmw_ctx_bindinfo_ib binding;
 	struct vmw_resource *res;
-	struct {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdDXSetIndexBuffer body;
-	} *cmd;
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXSetIndexBuffer);
 	int ret;
 
-	if (unlikely(ctx_node == NULL)) {
-		DRM_ERROR("DX Context not set.\n");
+	if (!ctx_node)
 		return -EINVAL;
-	}
 
 	cmd = container_of(header, typeof(*cmd), header);
 	ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				user_surface_converter,
+				VMW_RES_DIRTY_NONE, user_surface_converter,
 				&cmd->body.sid, &res);
 	if (unlikely(ret != 0))
 		return ret;
@@ -2469,8 +2317,8 @@ static int vmw_cmd_dx_set_index_buffer(struct vmw_private *dev_priv,
 }
 
 /**
- * vmw_cmd_dx_set_rendertarget - Validate an
- * SVGA_3D_CMD_DX_SET_RENDERTARGETS command
+ * vmw_cmd_dx_set_rendertarget - Validate SVGA_3D_CMD_DX_SET_RENDERTARGETS
+ * command
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -2480,32 +2328,29 @@ static int vmw_cmd_dx_set_rendertargets(struct vmw_private *dev_priv,
 					struct vmw_sw_context *sw_context,
 					SVGA3dCmdHeader *header)
 {
-	struct {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdDXSetRenderTargets body;
-	} *cmd = container_of(header, typeof(*cmd), header);
-	int ret;
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXSetRenderTargets) =
+		container_of(header, typeof(*cmd), header);
 	u32 num_rt_view = (cmd->header.size - sizeof(cmd->body)) /
 		sizeof(SVGA3dRenderTargetViewId);
+	int ret;
 
 	if (num_rt_view > SVGA3D_MAX_SIMULTANEOUS_RENDER_TARGETS) {
-		DRM_ERROR("Invalid DX Rendertarget binding.\n");
+		VMW_DEBUG_USER("Invalid DX Rendertarget binding.\n");
 		return -EINVAL;
 	}
 
-	ret = vmw_view_bindings_add(sw_context, vmw_view_ds,
-				    vmw_ctx_binding_ds, 0,
-				    &cmd->body.depthStencilViewId, 1, 0);
+	ret = vmw_view_bindings_add(sw_context, vmw_view_ds, vmw_ctx_binding_ds,
+				    0, &cmd->body.depthStencilViewId, 1, 0);
 	if (ret)
 		return ret;
 
 	return vmw_view_bindings_add(sw_context, vmw_view_rt,
-				     vmw_ctx_binding_dx_rt, 0,
-				     (void *)&cmd[1], num_rt_view, 0);
+				     vmw_ctx_binding_dx_rt, 0, (void *)&cmd[1],
+				     num_rt_view, 0);
 }
 
 /**
- * vmw_cmd_dx_clear_rendertarget_view - Validate an
+ * vmw_cmd_dx_clear_rendertarget_view - Validate
  * SVGA_3D_CMD_DX_CLEAR_RENDERTARGET_VIEW command
  *
  * @dev_priv: Pointer to a device private struct.
@@ -2516,17 +2361,15 @@ static int vmw_cmd_dx_clear_rendertarget_view(struct vmw_private *dev_priv,
 					      struct vmw_sw_context *sw_context,
 					      SVGA3dCmdHeader *header)
 {
-	struct {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdDXClearRenderTargetView body;
-	} *cmd = container_of(header, typeof(*cmd), header);
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXClearRenderTargetView) =
+		container_of(header, typeof(*cmd), header);
 
 	return PTR_RET(vmw_view_id_val_add(sw_context, vmw_view_rt,
 					   cmd->body.renderTargetViewId));
 }
 
 /**
- * vmw_cmd_dx_clear_rendertarget_view - Validate an
+ * vmw_cmd_dx_clear_rendertarget_view - Validate
  * SVGA_3D_CMD_DX_CLEAR_DEPTHSTENCIL_VIEW command
  *
  * @dev_priv: Pointer to a device private struct.
@@ -2537,10 +2380,8 @@ static int vmw_cmd_dx_clear_depthstencil_view(struct vmw_private *dev_priv,
 					      struct vmw_sw_context *sw_context,
 					      SVGA3dCmdHeader *header)
 {
-	struct {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdDXClearDepthStencilView body;
-	} *cmd = container_of(header, typeof(*cmd), header);
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXClearDepthStencilView) =
+		container_of(header, typeof(*cmd), header);
 
 	return PTR_RET(vmw_view_id_val_add(sw_context, vmw_view_ds,
 					   cmd->body.depthStencilViewId));
@@ -2550,14 +2391,14 @@ static int vmw_cmd_dx_view_define(struct vmw_private *dev_priv,
 				  struct vmw_sw_context *sw_context,
 				  SVGA3dCmdHeader *header)
 {
-	struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node;
+	struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context);
 	struct vmw_resource *srf;
 	struct vmw_resource *res;
 	enum vmw_view_type view_type;
 	int ret;
 	/*
-	 * This is based on the fact that all affected define commands have
-	 * the same initial command body layout.
+	 * This is based on the fact that all affected define commands have the
+	 * same initial command body layout.
 	 */
 	struct {
 		SVGA3dCmdHeader header;
@@ -2565,17 +2406,16 @@ static int vmw_cmd_dx_view_define(struct vmw_private *dev_priv,
 		uint32 sid;
 	} *cmd;
 
-	if (unlikely(ctx_node == NULL)) {
-		DRM_ERROR("DX Context not set.\n");
+	if (!ctx_node)
 		return -EINVAL;
-	}
 
 	view_type = vmw_view_cmd_to_type(header->id);
 	if (view_type == vmw_view_max)
 		return -EINVAL;
+
 	cmd = container_of(header, typeof(*cmd), header);
 	ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				user_surface_converter,
+				VMW_RES_DIRTY_NONE, user_surface_converter,
 				&cmd->sid, &srf);
 	if (unlikely(ret != 0))
 		return ret;
@@ -2585,19 +2425,14 @@ static int vmw_cmd_dx_view_define(struct vmw_private *dev_priv,
 	if (unlikely(ret != 0))
 		return ret;
 
-	return vmw_view_add(sw_context->man,
-			    ctx_node->ctx,
-			    srf,
-			    view_type,
-			    cmd->defined_id,
-			    header,
+	return vmw_view_add(sw_context->man, ctx_node->ctx, srf, view_type,
+			    cmd->defined_id, header,
 			    header->size + sizeof(*header),
 			    &sw_context->staged_cmd_res);
 }
 
 /**
- * vmw_cmd_dx_set_so_targets - Validate an
- * SVGA_3D_CMD_DX_SET_SOTARGETS command.
+ * vmw_cmd_dx_set_so_targets - Validate SVGA_3D_CMD_DX_SET_SOTARGETS command.
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -2607,7 +2442,7 @@ static int vmw_cmd_dx_set_so_targets(struct vmw_private *dev_priv,
 				     struct vmw_sw_context *sw_context,
 				     SVGA3dCmdHeader *header)
 {
-	struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node;
+	struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context);
 	struct vmw_ctx_bindinfo_so binding;
 	struct vmw_resource *res;
 	struct {
@@ -2617,22 +2452,20 @@ static int vmw_cmd_dx_set_so_targets(struct vmw_private *dev_priv,
 	} *cmd;
 	int i, ret, num;
 
-	if (unlikely(ctx_node == NULL)) {
-		DRM_ERROR("DX Context not set.\n");
+	if (!ctx_node)
 		return -EINVAL;
-	}
 
 	cmd = container_of(header, typeof(*cmd), header);
-	num = (cmd->header.size - sizeof(cmd->body)) /
-		sizeof(SVGA3dSoTarget);
+	num = (cmd->header.size - sizeof(cmd->body)) / sizeof(SVGA3dSoTarget);
 
 	if (num > SVGA3D_DX_MAX_SOTARGETS) {
-		DRM_ERROR("Invalid DX SO binding.\n");
+		VMW_DEBUG_USER("Invalid DX SO binding.\n");
 		return -EINVAL;
 	}
 
 	for (i = 0; i < num; i++) {
 		ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
+					VMW_RES_DIRTY_SET,
 					user_surface_converter,
 					&cmd->targets[i].sid, &res);
 		if (unlikely(ret != 0))
@@ -2645,8 +2478,7 @@ static int vmw_cmd_dx_set_so_targets(struct vmw_private *dev_priv,
 		binding.size = cmd->targets[i].sizeInBytes;
 		binding.slot = i;
 
-		vmw_binding_add(ctx_node->staged, &binding.bi,
-				0, binding.slot);
+		vmw_binding_add(ctx_node->staged, &binding.bi, 0, binding.slot);
 	}
 
 	return 0;
@@ -2656,7 +2488,7 @@ static int vmw_cmd_dx_so_define(struct vmw_private *dev_priv,
 				struct vmw_sw_context *sw_context,
 				SVGA3dCmdHeader *header)
 {
-	struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node;
+	struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context);
 	struct vmw_resource *res;
 	/*
 	 * This is based on the fact that all affected define commands have
@@ -2669,10 +2501,8 @@ static int vmw_cmd_dx_so_define(struct vmw_private *dev_priv,
 	enum vmw_so_type so_type;
 	int ret;
 
-	if (unlikely(ctx_node == NULL)) {
-		DRM_ERROR("DX Context not set.\n");
+	if (!ctx_node)
 		return -EINVAL;
-	}
 
 	so_type = vmw_so_cmd_to_type(header->id);
 	res = vmw_context_cotable(ctx_node->ctx, vmw_so_cotables[so_type]);
@@ -2683,8 +2513,8 @@ static int vmw_cmd_dx_so_define(struct vmw_private *dev_priv,
 }
 
 /**
- * vmw_cmd_dx_check_subresource - Validate an
- * SVGA_3D_CMD_DX_[X]_SUBRESOURCE command
+ * vmw_cmd_dx_check_subresource - Validate SVGA_3D_CMD_DX_[X]_SUBRESOURCE
+ * command
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -2714,7 +2544,7 @@ static int vmw_cmd_dx_check_subresource(struct vmw_private *dev_priv,
 	cmd = container_of(header, typeof(*cmd), header);
 
 	return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				 user_surface_converter,
+				 VMW_RES_DIRTY_NONE, user_surface_converter,
 				 &cmd->sid, NULL);
 }
 
@@ -2722,32 +2552,30 @@ static int vmw_cmd_dx_cid_check(struct vmw_private *dev_priv,
 				struct vmw_sw_context *sw_context,
 				SVGA3dCmdHeader *header)
 {
-	struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node;
+	struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context);
 
-	if (unlikely(ctx_node == NULL)) {
-		DRM_ERROR("DX Context not set.\n");
+	if (!ctx_node)
 		return -EINVAL;
-	}
 
 	return 0;
 }
 
 /**
- * vmw_cmd_dx_view_remove - validate a view remove command and
- * schedule the view resource for removal.
+ * vmw_cmd_dx_view_remove - validate a view remove command and schedule the view
+ * resource for removal.
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
  * @header: Pointer to the command header in the command stream.
  *
- * Check that the view exists, and if it was not created using this
- * command batch, conditionally make this command a NOP.
+ * Check that the view exists, and if it was not created using this command
+ * batch, conditionally make this command a NOP.
  */
 static int vmw_cmd_dx_view_remove(struct vmw_private *dev_priv,
 				  struct vmw_sw_context *sw_context,
 				  SVGA3dCmdHeader *header)
 {
-	struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node;
+	struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context);
 	struct {
 		SVGA3dCmdHeader header;
 		union vmw_view_destroy body;
@@ -2756,15 +2584,11 @@ static int vmw_cmd_dx_view_remove(struct vmw_private *dev_priv,
 	struct vmw_resource *view;
 	int ret;
 
-	if (!ctx_node) {
-		DRM_ERROR("DX Context not set.\n");
+	if (!ctx_node)
 		return -EINVAL;
-	}
 
-	ret = vmw_view_remove(sw_context->man,
-			      cmd->body.view_id, view_type,
-			      &sw_context->staged_cmd_res,
-			      &view);
+	ret = vmw_view_remove(sw_context->man, cmd->body.view_id, view_type,
+			      &sw_context->staged_cmd_res, &view);
 	if (ret || !view)
 		return ret;
 
@@ -2774,16 +2598,14 @@ static int vmw_cmd_dx_view_remove(struct vmw_private *dev_priv,
 	 * relocation to conditionally make this command a NOP to avoid
 	 * device errors.
 	 */
-	return vmw_resource_relocation_add(sw_context,
-					   view,
+	return vmw_resource_relocation_add(sw_context, view,
 					   vmw_ptr_diff(sw_context->buf_start,
 							&cmd->header.id),
 					   vmw_res_rel_cond_nop);
 }
 
 /**
- * vmw_cmd_dx_define_shader - Validate an SVGA_3D_CMD_DX_DEFINE_SHADER
- * command
+ * vmw_cmd_dx_define_shader - Validate SVGA_3D_CMD_DX_DEFINE_SHADER command
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -2793,18 +2615,14 @@ static int vmw_cmd_dx_define_shader(struct vmw_private *dev_priv,
 				    struct vmw_sw_context *sw_context,
 				    SVGA3dCmdHeader *header)
 {
-	struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node;
+	struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context);
 	struct vmw_resource *res;
-	struct {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdDXDefineShader body;
-	} *cmd = container_of(header, typeof(*cmd), header);
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXDefineShader) =
+		container_of(header, typeof(*cmd), header);
 	int ret;
 
-	if (!ctx_node) {
-		DRM_ERROR("DX Context not set.\n");
+	if (!ctx_node)
 		return -EINVAL;
-	}
 
 	res = vmw_context_cotable(ctx_node->ctx, SVGA_COTABLE_DXSHADER);
 	ret = vmw_cotable_notify(res, cmd->body.shaderId);
@@ -2817,8 +2635,7 @@ static int vmw_cmd_dx_define_shader(struct vmw_private *dev_priv,
 }
 
 /**
- * vmw_cmd_dx_destroy_shader - Validate an SVGA_3D_CMD_DX_DESTROY_SHADER
- * command
+ * vmw_cmd_dx_destroy_shader - Validate SVGA_3D_CMD_DX_DESTROY_SHADER command
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -2828,29 +2645,22 @@ static int vmw_cmd_dx_destroy_shader(struct vmw_private *dev_priv,
 				     struct vmw_sw_context *sw_context,
 				     SVGA3dCmdHeader *header)
 {
-	struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node;
-	struct {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdDXDestroyShader body;
-	} *cmd = container_of(header, typeof(*cmd), header);
+	struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context);
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXDestroyShader) =
+		container_of(header, typeof(*cmd), header);
 	int ret;
 
-	if (!ctx_node) {
-		DRM_ERROR("DX Context not set.\n");
+	if (!ctx_node)
 		return -EINVAL;
-	}
 
 	ret = vmw_shader_remove(sw_context->man, cmd->body.shaderId, 0,
 				&sw_context->staged_cmd_res);
-	if (ret)
-		DRM_ERROR("Could not find shader to remove.\n");
 
 	return ret;
 }
 
 /**
- * vmw_cmd_dx_bind_shader - Validate an SVGA_3D_CMD_DX_BIND_SHADER
- * command
+ * vmw_cmd_dx_bind_shader - Validate SVGA_3D_CMD_DX_BIND_SHADER command
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -2862,36 +2672,37 @@ static int vmw_cmd_dx_bind_shader(struct vmw_private *dev_priv,
 {
 	struct vmw_resource *ctx;
 	struct vmw_resource *res;
-	struct {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdDXBindShader body;
-	} *cmd = container_of(header, typeof(*cmd), header);
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXBindShader) =
+		container_of(header, typeof(*cmd), header);
 	int ret;
 
 	if (cmd->body.cid != SVGA3D_INVALID_ID) {
 		ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context,
-					user_context_converter,
-					&cmd->body.cid, &ctx);
+					VMW_RES_DIRTY_SET,
+					user_context_converter, &cmd->body.cid,
+					&ctx);
 		if (ret)
 			return ret;
 	} else {
-		if (!sw_context->dx_ctx_node) {
-			DRM_ERROR("DX Context not set.\n");
+		struct vmw_ctx_validation_info *ctx_node =
+			VMW_GET_CTX_NODE(sw_context);
+
+		if (!ctx_node)
 			return -EINVAL;
-		}
-		ctx = sw_context->dx_ctx_node->ctx;
+
+		ctx = ctx_node->ctx;
 	}
 
-	res = vmw_shader_lookup(vmw_context_res_man(ctx),
-				cmd->body.shid, 0);
+	res = vmw_shader_lookup(vmw_context_res_man(ctx), cmd->body.shid, 0);
 	if (IS_ERR(res)) {
-		DRM_ERROR("Could not find shader to bind.\n");
+		VMW_DEBUG_USER("Could not find shader to bind.\n");
 		return PTR_ERR(res);
 	}
 
-	ret = vmw_execbuf_res_noctx_val_add(sw_context, res);
+	ret = vmw_execbuf_res_noctx_val_add(sw_context, res,
+					    VMW_RES_DIRTY_NONE);
 	if (ret) {
-		DRM_ERROR("Error creating resource validation node.\n");
+		VMW_DEBUG_USER("Error creating resource validation node.\n");
 		return ret;
 	}
 
@@ -2901,7 +2712,7 @@ static int vmw_cmd_dx_bind_shader(struct vmw_private *dev_priv,
 }
 
 /**
- * vmw_cmd_dx_genmips - Validate an SVGA_3D_CMD_DX_GENMIPS command
+ * vmw_cmd_dx_genmips - Validate SVGA_3D_CMD_DX_GENMIPS command
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -2911,18 +2722,16 @@ static int vmw_cmd_dx_genmips(struct vmw_private *dev_priv,
 			      struct vmw_sw_context *sw_context,
 			      SVGA3dCmdHeader *header)
 {
-	struct {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdDXGenMips body;
-	} *cmd = container_of(header, typeof(*cmd), header);
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXGenMips) =
+		container_of(header, typeof(*cmd), header);
 
 	return PTR_RET(vmw_view_id_val_add(sw_context, vmw_view_sr,
 					   cmd->body.shaderResourceViewId));
 }
 
 /**
- * vmw_cmd_dx_transfer_from_buffer -
- * Validate an SVGA_3D_CMD_DX_TRANSFER_FROM_BUFFER command
+ * vmw_cmd_dx_transfer_from_buffer - Validate
+ * SVGA_3D_CMD_DX_TRANSFER_FROM_BUFFER command
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -2932,26 +2741,23 @@ static int vmw_cmd_dx_transfer_from_buffer(struct vmw_private *dev_priv,
 					   struct vmw_sw_context *sw_context,
 					   SVGA3dCmdHeader *header)
 {
-	struct {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdDXTransferFromBuffer body;
-	} *cmd = container_of(header, typeof(*cmd), header);
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXTransferFromBuffer) =
+		container_of(header, typeof(*cmd), header);
 	int ret;
 
 	ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				user_surface_converter,
+				VMW_RES_DIRTY_NONE, user_surface_converter,
 				&cmd->body.srcSid, NULL);
 	if (ret != 0)
 		return ret;
 
 	return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				 user_surface_converter,
+				 VMW_RES_DIRTY_SET, user_surface_converter,
 				 &cmd->body.destSid, NULL);
 }
 
 /**
- * vmw_cmd_intra_surface_copy -
- * Validate an SVGA_3D_CMD_INTRA_SURFACE_COPY command
+ * vmw_cmd_intra_surface_copy - Validate SVGA_3D_CMD_INTRA_SURFACE_COPY command
  *
  * @dev_priv: Pointer to a device private struct.
  * @sw_context: The software context being used for this batch.
@@ -2961,20 +2767,17 @@ static int vmw_cmd_intra_surface_copy(struct vmw_private *dev_priv,
 					   struct vmw_sw_context *sw_context,
 					   SVGA3dCmdHeader *header)
 {
-	struct {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdIntraSurfaceCopy body;
-	} *cmd = container_of(header, typeof(*cmd), header);
+	VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdIntraSurfaceCopy) =
+		container_of(header, typeof(*cmd), header);
 
 	if (!(dev_priv->capabilities2 & SVGA_CAP2_INTRA_SURFACE_COPY))
 		return -EINVAL;
 
 	return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
-				user_surface_converter,
-				&cmd->body.surface.sid, NULL);
+				 VMW_RES_DIRTY_SET, user_surface_converter,
+				 &cmd->body.surface.sid, NULL);
 }
 
-
 static int vmw_cmd_check_not_3d(struct vmw_private *dev_priv,
 				struct vmw_sw_context *sw_context,
 				void *buf, uint32_t *size)
@@ -2997,18 +2800,18 @@ static int vmw_cmd_check_not_3d(struct vmw_private *dev_priv,
 		*size = sizeof(uint32_t) + sizeof(SVGAFifoCmdBlitGMRFBToScreen);
 		break;
 	default:
-		DRM_ERROR("Unsupported SVGA command: %u.\n", cmd_id);
+		VMW_DEBUG_USER("Unsupported SVGA command: %u.\n", cmd_id);
 		return -EINVAL;
 	}
 
 	if (*size > size_remaining) {
-		DRM_ERROR("Invalid SVGA command (size mismatch):"
-			  " %u.\n", cmd_id);
+		VMW_DEBUG_USER("Invalid SVGA command (size mismatch): %u.\n",
+			       cmd_id);
 		return -EINVAL;
 	}
 
 	if (unlikely(!sw_context->kernel)) {
-		DRM_ERROR("Kernel only SVGA command: %u.\n", cmd_id);
+		VMW_DEBUG_USER("Kernel only SVGA command: %u.\n", cmd_id);
 		return -EPERM;
 	}
 
@@ -3196,9 +2999,7 @@ static const struct vmw_cmd_entry vmw_cmd_entries[SVGA_3D_CMD_MAX] = {
 	VMW_CMD_DEF(SVGA_3D_CMD_DEFINE_GB_SURFACE_V2, &vmw_cmd_invalid,
 		    false, false, true),
 
-	/*
-	 * DX commands
-	 */
+	/* SM commands */
 	VMW_CMD_DEF(SVGA_3D_CMD_DX_DEFINE_CONTEXT, &vmw_cmd_invalid,
 		    false, false, true),
 	VMW_CMD_DEF(SVGA_3D_CMD_DX_DESTROY_CONTEXT, &vmw_cmd_invalid,
@@ -3380,8 +3181,8 @@ bool vmw_cmd_describe(const void *buf, u32 *size, char const **cmd)
 }
 
 static int vmw_cmd_check(struct vmw_private *dev_priv,
-			 struct vmw_sw_context *sw_context,
-			 void *buf, uint32_t *size)
+			 struct vmw_sw_context *sw_context, void *buf,
+			 uint32_t *size)
 {
 	uint32_t cmd_id;
 	uint32_t size_remaining = *size;
@@ -3420,31 +3221,33 @@ static int vmw_cmd_check(struct vmw_private *dev_priv,
 		goto out_new;
 
 	ret = entry->func(dev_priv, sw_context, header);
-	if (unlikely(ret != 0))
-		goto out_invalid;
+	if (unlikely(ret != 0)) {
+		VMW_DEBUG_USER("SVGA3D command: %d failed with error %d\n",
+			       cmd_id + SVGA_3D_CMD_BASE, ret);
+		return ret;
+	}
 
 	return 0;
 out_invalid:
-	DRM_ERROR("Invalid SVGA3D command: %d\n",
-		  cmd_id + SVGA_3D_CMD_BASE);
+	VMW_DEBUG_USER("Invalid SVGA3D command: %d\n",
+		       cmd_id + SVGA_3D_CMD_BASE);
 	return -EINVAL;
 out_privileged:
-	DRM_ERROR("Privileged SVGA3D command: %d\n",
-		  cmd_id + SVGA_3D_CMD_BASE);
+	VMW_DEBUG_USER("Privileged SVGA3D command: %d\n",
+		       cmd_id + SVGA_3D_CMD_BASE);
 	return -EPERM;
 out_old:
-	DRM_ERROR("Deprecated (disallowed) SVGA3D command: %d\n",
-		  cmd_id + SVGA_3D_CMD_BASE);
+	VMW_DEBUG_USER("Deprecated (disallowed) SVGA3D command: %d\n",
+		       cmd_id + SVGA_3D_CMD_BASE);
 	return -EINVAL;
 out_new:
-	DRM_ERROR("SVGA3D command: %d not supported by virtual hardware.\n",
-		  cmd_id + SVGA_3D_CMD_BASE);
+	VMW_DEBUG_USER("SVGA3D command: %d not supported by virtual device.\n",
+		       cmd_id + SVGA_3D_CMD_BASE);
 	return -EINVAL;
 }
 
 static int vmw_cmd_check_all(struct vmw_private *dev_priv,
-			     struct vmw_sw_context *sw_context,
-			     void *buf,
+			     struct vmw_sw_context *sw_context, void *buf,
 			     uint32_t size)
 {
 	int32_t cur_size = size;
@@ -3462,7 +3265,7 @@ static int vmw_cmd_check_all(struct vmw_private *dev_priv,
 	}
 
 	if (unlikely(cur_size != 0)) {
-		DRM_ERROR("Command verifier out of sync.\n");
+		VMW_DEBUG_USER("Command verifier out of sync.\n");
 		return -EINVAL;
 	}
 
@@ -3472,7 +3275,6 @@ static int vmw_cmd_check_all(struct vmw_private *dev_priv,
 static void vmw_free_relocations(struct vmw_sw_context *sw_context)
 {
 	/* Memory is validation context memory, so no need to free it */
-
 	INIT_LIST_HEAD(&sw_context->bo_relocations);
 }
 
@@ -3520,7 +3322,7 @@ static int vmw_resize_cmd_bounce(struct vmw_sw_context *sw_context,
 	sw_context->cmd_bounce = vmalloc(sw_context->cmd_bounce_size);
 
 	if (sw_context->cmd_bounce == NULL) {
-		DRM_ERROR("Failed to allocate command bounce buffer.\n");
+		VMW_DEBUG_USER("Failed to allocate command bounce buffer.\n");
 		sw_context->cmd_bounce_size = 0;
 		return -ENOMEM;
 	}
@@ -3535,8 +3337,8 @@ static int vmw_resize_cmd_bounce(struct vmw_sw_context *sw_context,
  * If this fails for some reason, We sync the fifo and return NULL.
  * It is then safe to fence buffers with a NULL pointer.
  *
- * If @p_handle is not NULL @file_priv must also not be NULL. Creates
- * a userspace handle if @p_handle is not NULL, otherwise not.
+ * If @p_handle is not NULL @file_priv must also not be NULL. Creates a
+ * userspace handle if @p_handle is not NULL, otherwise not.
  */
 
 int vmw_execbuf_fence_commands(struct drm_file *file_priv,
@@ -3553,7 +3355,7 @@ int vmw_execbuf_fence_commands(struct drm_file *file_priv,
 
 	ret = vmw_fifo_send_fence(dev_priv, &sequence);
 	if (unlikely(ret != 0)) {
-		DRM_ERROR("Fence submission error. Syncing.\n");
+		VMW_DEBUG_USER("Fence submission error. Syncing.\n");
 		synced = true;
 	}
 
@@ -3564,9 +3366,8 @@ int vmw_execbuf_fence_commands(struct drm_file *file_priv,
 		ret = vmw_fence_create(dev_priv->fman, sequence, p_fence);
 
 	if (unlikely(ret != 0 && !synced)) {
-		(void) vmw_fallback_wait(dev_priv, false, false,
-					 sequence, false,
-					 VMW_FENCE_WAIT_TIMEOUT);
+		(void) vmw_fallback_wait(dev_priv, false, false, sequence,
+					 false, VMW_FENCE_WAIT_TIMEOUT);
 		*p_fence = NULL;
 	}
 
@@ -3574,36 +3375,32 @@ int vmw_execbuf_fence_commands(struct drm_file *file_priv,
 }
 
 /**
- * vmw_execbuf_copy_fence_user - copy fence object information to
- * user-space.
+ * vmw_execbuf_copy_fence_user - copy fence object information to user-space.
  *
  * @dev_priv: Pointer to a vmw_private struct.
  * @vmw_fp: Pointer to the struct vmw_fpriv representing the calling file.
  * @ret: Return value from fence object creation.
- * @user_fence_rep: User space address of a struct drm_vmw_fence_rep to
- * which the information should be copied.
+ * @user_fence_rep: User space address of a struct drm_vmw_fence_rep to which
+ * the information should be copied.
  * @fence: Pointer to the fenc object.
  * @fence_handle: User-space fence handle.
  * @out_fence_fd: exported file descriptor for the fence.  -1 if not used
  * @sync_file:  Only used to clean up in case of an error in this function.
  *
- * This function copies fence information to user-space. If copying fails,
- * The user-space struct drm_vmw_fence_rep::error member is hopefully
- * left untouched, and if it's preloaded with an -EFAULT by user-space,
- * the error will hopefully be detected.
- * Also if copying fails, user-space will be unable to signal the fence
- * object so we wait for it immediately, and then unreference the
- * user-space reference.
+ * This function copies fence information to user-space. If copying fails, the
+ * user-space struct drm_vmw_fence_rep::error member is hopefully left
+ * untouched, and if it's preloaded with an -EFAULT by user-space, the error
+ * will hopefully be detected.
+ *
+ * Also if copying fails, user-space will be unable to signal the fence object
+ * so we wait for it immediately, and then unreference the user-space reference.
  */
 void
 vmw_execbuf_copy_fence_user(struct vmw_private *dev_priv,
-			    struct vmw_fpriv *vmw_fp,
-			    int ret,
+			    struct vmw_fpriv *vmw_fp, int ret,
 			    struct drm_vmw_fence_rep __user *user_fence_rep,
-			    struct vmw_fence_obj *fence,
-			    uint32_t fence_handle,
-			    int32_t out_fence_fd,
-			    struct sync_file *sync_file)
+			    struct vmw_fence_obj *fence, uint32_t fence_handle,
+			    int32_t out_fence_fd, struct sync_file *sync_file)
 {
 	struct drm_vmw_fence_rep fence_rep;
 
@@ -3624,16 +3421,16 @@ vmw_execbuf_copy_fence_user(struct vmw_private *dev_priv,
 	}
 
 	/*
-	 * copy_to_user errors will be detected by user space not
-	 * seeing fence_rep::error filled in. Typically
-	 * user-space would have pre-set that member to -EFAULT.
+	 * copy_to_user errors will be detected by user space not seeing
+	 * fence_rep::error filled in. Typically user-space would have pre-set
+	 * that member to -EFAULT.
 	 */
 	ret = copy_to_user(user_fence_rep, &fence_rep,
 			   sizeof(fence_rep));
 
 	/*
-	 * User-space lost the fence object. We need to sync
-	 * and unreference the handle.
+	 * User-space lost the fence object. We need to sync and unreference the
+	 * handle.
 	 */
 	if (unlikely(ret != 0) && (fence_rep.error == 0)) {
 		if (sync_file)
@@ -3644,42 +3441,39 @@ vmw_execbuf_copy_fence_user(struct vmw_private *dev_priv,
 			fence_rep.fd = -1;
 		}
 
-		ttm_ref_object_base_unref(vmw_fp->tfile,
-					  fence_handle, TTM_REF_USAGE);
-		DRM_ERROR("Fence copy error. Syncing.\n");
+		ttm_ref_object_base_unref(vmw_fp->tfile, fence_handle,
+					  TTM_REF_USAGE);
+		VMW_DEBUG_USER("Fence copy error. Syncing.\n");
 		(void) vmw_fence_obj_wait(fence, false, false,
 					  VMW_FENCE_WAIT_TIMEOUT);
 	}
 }
 
 /**
- * vmw_execbuf_submit_fifo - Patch a command batch and submit it using
- * the fifo.
+ * vmw_execbuf_submit_fifo - Patch a command batch and submit it using the fifo.
  *
  * @dev_priv: Pointer to a device private structure.
  * @kernel_commands: Pointer to the unpatched command batch.
  * @command_size: Size of the unpatched command batch.
  * @sw_context: Structure holding the relocation lists.
  *
- * Side effects: If this function returns 0, then the command batch
- * pointed to by @kernel_commands will have been modified.
+ * Side effects: If this function returns 0, then the command batch pointed to
+ * by @kernel_commands will have been modified.
  */
 static int vmw_execbuf_submit_fifo(struct vmw_private *dev_priv,
-				   void *kernel_commands,
-				   u32 command_size,
+				   void *kernel_commands, u32 command_size,
 				   struct vmw_sw_context *sw_context)
 {
 	void *cmd;
 
 	if (sw_context->dx_ctx_node)
-		cmd = vmw_fifo_reserve_dx(dev_priv, command_size,
+		cmd = VMW_FIFO_RESERVE_DX(dev_priv, command_size,
 					  sw_context->dx_ctx_node->ctx->id);
 	else
-		cmd = vmw_fifo_reserve(dev_priv, command_size);
-	if (!cmd) {
-		DRM_ERROR("Failed reserving fifo space for commands.\n");
+		cmd = VMW_FIFO_RESERVE(dev_priv, command_size);
+
+	if (!cmd)
 		return -ENOMEM;
-	}
 
 	vmw_apply_relocations(sw_context);
 	memcpy(cmd, kernel_commands, command_size);
@@ -3691,16 +3485,16 @@ static int vmw_execbuf_submit_fifo(struct vmw_private *dev_priv,
 }
 
 /**
- * vmw_execbuf_submit_cmdbuf - Patch a command batch and submit it using
- * the command buffer manager.
+ * vmw_execbuf_submit_cmdbuf - Patch a command batch and submit it using the
+ * command buffer manager.
  *
  * @dev_priv: Pointer to a device private structure.
  * @header: Opaque handle to the command buffer allocation.
  * @command_size: Size of the unpatched command batch.
  * @sw_context: Structure holding the relocation lists.
  *
- * Side effects: If this function returns 0, then the command buffer
- * represented by @header will have been modified.
+ * Side effects: If this function returns 0, then the command buffer represented
+ * by @header will have been modified.
  */
 static int vmw_execbuf_submit_cmdbuf(struct vmw_private *dev_priv,
 				     struct vmw_cmdbuf_header *header,
@@ -3709,8 +3503,8 @@ static int vmw_execbuf_submit_cmdbuf(struct vmw_private *dev_priv,
 {
 	u32 id = ((sw_context->dx_ctx_node) ? sw_context->dx_ctx_node->ctx->id :
 		  SVGA3D_INVALID_ID);
-	void *cmd = vmw_cmdbuf_reserve(dev_priv->cman, command_size,
-				       id, false, header);
+	void *cmd = vmw_cmdbuf_reserve(dev_priv->cman, command_size, id, false,
+				       header);
 
 	vmw_apply_relocations(sw_context);
 	vmw_resource_relocations_apply(cmd, &sw_context->res_relocations);
@@ -3730,22 +3524,23 @@ static int vmw_execbuf_submit_cmdbuf(struct vmw_private *dev_priv,
  * @header: Out parameter returning the opaque pointer to the command buffer.
  *
  * This function checks whether we can use the command buffer manager for
- * submission and if so, creates a command buffer of suitable size and
- * copies the user data into that buffer.
+ * submission and if so, creates a command buffer of suitable size and copies
+ * the user data into that buffer.
  *
  * On successful return, the function returns a pointer to the data in the
  * command buffer and *@header is set to non-NULL.
- * If command buffers could not be used, the function will return the value
- * of @kernel_commands on function call. That value may be NULL. In that case,
- * the value of *@header will be set to NULL.
+ *
+ * If command buffers could not be used, the function will return the value of
+ * @kernel_commands on function call. That value may be NULL. In that case, the
+ * value of *@header will be set to NULL.
+ *
  * If an error is encountered, the function will return a pointer error value.
  * If the function is interrupted by a signal while sleeping, it will return
  * -ERESTARTSYS casted to a pointer error value.
  */
 static void *vmw_execbuf_cmdbuf(struct vmw_private *dev_priv,
 				void __user *user_commands,
-				void *kernel_commands,
-				u32 command_size,
+				void *kernel_commands, u32 command_size,
 				struct vmw_cmdbuf_header **header)
 {
 	size_t cmdbuf_size;
@@ -3753,7 +3548,7 @@ static void *vmw_execbuf_cmdbuf(struct vmw_private *dev_priv,
 
 	*header = NULL;
 	if (command_size > SVGA_CB_MAX_SIZE) {
-		DRM_ERROR("Command buffer is too large.\n");
+		VMW_DEBUG_USER("Command buffer is too large.\n");
 		return ERR_PTR(-EINVAL);
 	}
 
@@ -3763,15 +3558,14 @@ static void *vmw_execbuf_cmdbuf(struct vmw_private *dev_priv,
 	/* If possible, add a little space for fencing. */
 	cmdbuf_size = command_size + 512;
 	cmdbuf_size = min_t(size_t, cmdbuf_size, SVGA_CB_MAX_SIZE);
-	kernel_commands = vmw_cmdbuf_alloc(dev_priv->cman, cmdbuf_size,
-					   true, header);
+	kernel_commands = vmw_cmdbuf_alloc(dev_priv->cman, cmdbuf_size, true,
+					   header);
 	if (IS_ERR(kernel_commands))
 		return kernel_commands;
 
-	ret = copy_from_user(kernel_commands, user_commands,
-			     command_size);
+	ret = copy_from_user(kernel_commands, user_commands, command_size);
 	if (ret) {
-		DRM_ERROR("Failed copying commands.\n");
+		VMW_DEBUG_USER("Failed copying commands.\n");
 		vmw_cmdbuf_header_free(*header);
 		*header = NULL;
 		return ERR_PTR(-EFAULT);
@@ -3799,13 +3593,13 @@ static int vmw_execbuf_tie_context(struct vmw_private *dev_priv,
 	res = vmw_user_resource_noref_lookup_handle
 		(dev_priv, sw_context->fp->tfile, handle,
 		 user_context_converter);
-	if (unlikely(IS_ERR(res))) {
-		DRM_ERROR("Could not find or user DX context 0x%08x.\n",
-			  (unsigned) handle);
+	if (IS_ERR(res)) {
+		VMW_DEBUG_USER("Could not find or user DX context 0x%08x.\n",
+			       (unsigned int) handle);
 		return PTR_ERR(res);
 	}
 
-	ret = vmw_execbuf_res_noref_val_add(sw_context, res);
+	ret = vmw_execbuf_res_noref_val_add(sw_context, res, VMW_RES_DIRTY_SET);
 	if (unlikely(ret != 0))
 		return ret;
 
@@ -3817,19 +3611,16 @@ static int vmw_execbuf_tie_context(struct vmw_private *dev_priv,
 
 int vmw_execbuf_process(struct drm_file *file_priv,
 			struct vmw_private *dev_priv,
-			void __user *user_commands,
-			void *kernel_commands,
-			uint32_t command_size,
-			uint64_t throttle_us,
+			void __user *user_commands, void *kernel_commands,
+			uint32_t command_size, uint64_t throttle_us,
 			uint32_t dx_context_handle,
 			struct drm_vmw_fence_rep __user *user_fence_rep,
-			struct vmw_fence_obj **out_fence,
-			uint32_t flags)
+			struct vmw_fence_obj **out_fence, uint32_t flags)
 {
 	struct vmw_sw_context *sw_context = &dev_priv->ctx;
 	struct vmw_fence_obj *fence = NULL;
 	struct vmw_cmdbuf_header *header;
-	uint32_t handle;
+	uint32_t handle = 0;
 	int ret;
 	int32_t out_fence_fd = -1;
 	struct sync_file *sync_file = NULL;
@@ -3840,7 +3631,7 @@ int vmw_execbuf_process(struct drm_file *file_priv,
 	if (flags & DRM_VMW_EXECBUF_FLAG_EXPORT_FENCE_FD) {
 		out_fence_fd = get_unused_fd_flags(O_CLOEXEC);
 		if (out_fence_fd < 0) {
-			DRM_ERROR("Failed to get a fence file descriptor.\n");
+			VMW_DEBUG_USER("Failed to get a fence fd.\n");
 			return out_fence_fd;
 		}
 	}
@@ -3873,18 +3664,18 @@ int vmw_execbuf_process(struct drm_file *file_priv,
 		if (unlikely(ret != 0))
 			goto out_unlock;
 
-
-		ret = copy_from_user(sw_context->cmd_bounce,
-				     user_commands, command_size);
-
+		ret = copy_from_user(sw_context->cmd_bounce, user_commands,
+				     command_size);
 		if (unlikely(ret != 0)) {
 			ret = -EFAULT;
-			DRM_ERROR("Failed copying commands.\n");
+			VMW_DEBUG_USER("Failed copying commands.\n");
 			goto out_unlock;
 		}
+
 		kernel_commands = sw_context->cmd_bounce;
-	} else if (!header)
+	} else if (!header) {
 		sw_context->kernel = true;
+	}
 
 	sw_context->fp = vmw_fpriv(file_priv);
 	INIT_LIST_HEAD(&sw_context->ctx_list);
@@ -3897,6 +3688,7 @@ int vmw_execbuf_process(struct drm_file *file_priv,
 	memset(sw_context->res_cache, 0, sizeof(sw_context->res_cache));
 	INIT_LIST_HEAD(&sw_context->res_relocations);
 	INIT_LIST_HEAD(&sw_context->bo_relocations);
+
 	if (sw_context->staged_bindings)
 		vmw_binding_state_reset(sw_context->staged_bindings);
 
@@ -3904,8 +3696,10 @@ int vmw_execbuf_process(struct drm_file *file_priv,
 		ret = drm_ht_create(&sw_context->res_ht, VMW_RES_HT_ORDER);
 		if (unlikely(ret != 0))
 			goto out_unlock;
+
 		sw_context->res_ht_initialized = true;
 	}
+
 	INIT_LIST_HEAD(&sw_context->staged_cmd_res);
 	sw_context->ctx = &val_ctx;
 	ret = vmw_execbuf_tie_context(dev_priv, sw_context, dx_context_handle);
@@ -3932,6 +3726,7 @@ int vmw_execbuf_process(struct drm_file *file_priv,
 	ret = vmw_validation_res_validate(&val_ctx, true);
 	if (unlikely(ret != 0))
 		goto out_err;
+
 	vmw_validation_drop_ht(&val_ctx);
 
 	ret = mutex_lock_interruptible(&dev_priv->binding_mutex);
@@ -3959,17 +3754,15 @@ int vmw_execbuf_process(struct drm_file *file_priv,
 		goto out_err;
 
 	vmw_query_bo_switch_commit(dev_priv, sw_context);
-	ret = vmw_execbuf_fence_commands(file_priv, dev_priv,
-					 &fence,
+	ret = vmw_execbuf_fence_commands(file_priv, dev_priv, &fence,
 					 (user_fence_rep) ? &handle : NULL);
 	/*
 	 * This error is harmless, because if fence submission fails,
 	 * vmw_fifo_send_fence will sync. The error will be propagated to
 	 * user-space in @fence_rep
 	 */
-
 	if (ret != 0)
-		DRM_ERROR("Fence submission error. Syncing.\n");
+		VMW_DEBUG_USER("Fence submission error. Syncing.\n");
 
 	vmw_execbuf_bindings_commit(sw_context, false);
 	vmw_bind_dx_query_mob(sw_context);
@@ -3977,21 +3770,19 @@ int vmw_execbuf_process(struct drm_file *file_priv,
 
 	vmw_validation_bo_fence(sw_context->ctx, fence);
 
-	if (unlikely(dev_priv->pinned_bo != NULL &&
-		     !dev_priv->query_cid_valid))
+	if (unlikely(dev_priv->pinned_bo != NULL && !dev_priv->query_cid_valid))
 		__vmw_execbuf_release_pinned_bo(dev_priv, fence);
 
 	/*
-	 * If anything fails here, give up trying to export the fence
-	 * and do a sync since the user mode will not be able to sync
-	 * the fence itself.  This ensures we are still functionally
-	 * correct.
+	 * If anything fails here, give up trying to export the fence and do a
+	 * sync since the user mode will not be able to sync the fence itself.
+	 * This ensures we are still functionally correct.
 	 */
 	if (flags & DRM_VMW_EXECBUF_FLAG_EXPORT_FENCE_FD) {
 
 		sync_file = sync_file_create(&fence->base);
 		if (!sync_file) {
-			DRM_ERROR("Unable to create sync file for fence\n");
+			VMW_DEBUG_USER("Sync file create failed for fence\n");
 			put_unused_fd(out_fence_fd);
 			out_fence_fd = -1;
 
@@ -4004,8 +3795,8 @@ int vmw_execbuf_process(struct drm_file *file_priv,
 	}
 
 	vmw_execbuf_copy_fence_user(dev_priv, vmw_fpriv(file_priv), ret,
-				    user_fence_rep, fence, handle,
-				    out_fence_fd, sync_file);
+				    user_fence_rep, fence, handle, out_fence_fd,
+				    sync_file);
 
 	/* Don't unreference when handing fence out */
 	if (unlikely(out_fence != NULL)) {
@@ -4019,8 +3810,8 @@ int vmw_execbuf_process(struct drm_file *file_priv,
 	mutex_unlock(&dev_priv->cmdbuf_mutex);
 
 	/*
-	 * Unreference resources outside of the cmdbuf_mutex to
-	 * avoid deadlocks in resource destruction paths.
+	 * Unreference resources outside of the cmdbuf_mutex to avoid deadlocks
+	 * in resource destruction paths.
 	 */
 	vmw_validation_unref_lists(&val_ctx);
 
@@ -4035,8 +3826,7 @@ out_err_nores:
 	vmw_validation_res_unreserve(&val_ctx, true);
 	vmw_resource_relocations_free(&sw_context->res_relocations);
 	vmw_free_relocations(sw_context);
-	if (unlikely(dev_priv->pinned_bo != NULL &&
-		     !dev_priv->query_cid_valid))
+	if (unlikely(dev_priv->pinned_bo != NULL && !dev_priv->query_cid_valid))
 		__vmw_execbuf_release_pinned_bo(dev_priv, NULL);
 out_unlock:
 	vmw_cmdbuf_res_revert(&sw_context->staged_cmd_res);
@@ -4045,8 +3835,8 @@ out_unlock:
 	mutex_unlock(&dev_priv->cmdbuf_mutex);
 
 	/*
-	 * Unreference resources outside of the cmdbuf_mutex to
-	 * avoid deadlocks in resource destruction paths.
+	 * Unreference resources outside of the cmdbuf_mutex to avoid deadlocks
+	 * in resource destruction paths.
 	 */
 	vmw_validation_unref_lists(&val_ctx);
 out_free_header:
@@ -4064,13 +3854,13 @@ out_free_fence_fd:
  *
  * @dev_priv: The device private structure.
  *
- * This function is called to idle the fifo and unpin the query buffer
- * if the normal way to do this hits an error, which should typically be
- * extremely rare.
+ * This function is called to idle the fifo and unpin the query buffer if the
+ * normal way to do this hits an error, which should typically be extremely
+ * rare.
  */
 static void vmw_execbuf_unpin_panic(struct vmw_private *dev_priv)
 {
-	DRM_ERROR("Can't unpin query buffer. Trying to recover.\n");
+	VMW_DEBUG_USER("Can't unpin query buffer. Trying to recover.\n");
 
 	(void) vmw_fallback_wait(dev_priv, false, true, 0, false, 10*HZ);
 	vmw_bo_pin_reserved(dev_priv->pinned_bo, false);
@@ -4082,28 +3872,27 @@ static void vmw_execbuf_unpin_panic(struct vmw_private *dev_priv)
 
 
 /**
- * __vmw_execbuf_release_pinned_bo - Flush queries and unpin the pinned
- * query bo.
+ * __vmw_execbuf_release_pinned_bo - Flush queries and unpin the pinned query
+ * bo.
  *
  * @dev_priv: The device private structure.
- * @fence: If non-NULL should point to a struct vmw_fence_obj issued
- * _after_ a query barrier that flushes all queries touching the current
- * buffer pointed to by @dev_priv->pinned_bo
+ * @fence: If non-NULL should point to a struct vmw_fence_obj issued _after_ a
+ * query barrier that flushes all queries touching the current buffer pointed to
+ * by @dev_priv->pinned_bo
  *
- * This function should be used to unpin the pinned query bo, or
- * as a query barrier when we need to make sure that all queries have
- * finished before the next fifo command. (For example on hardware
- * context destructions where the hardware may otherwise leak unfinished
- * queries).
+ * This function should be used to unpin the pinned query bo, or as a query
+ * barrier when we need to make sure that all queries have finished before the
+ * next fifo command. (For example on hardware context destructions where the
+ * hardware may otherwise leak unfinished queries).
  *
- * This function does not return any failure codes, but make attempts
- * to do safe unpinning in case of errors.
+ * This function does not return any failure codes, but make attempts to do safe
+ * unpinning in case of errors.
  *
- * The function will synchronize on the previous query barrier, and will
- * thus not finish until that barrier has executed.
+ * The function will synchronize on the previous query barrier, and will thus
+ * not finish until that barrier has executed.
  *
- * the @dev_priv->cmdbuf_mutex needs to be held by the current thread
- * before calling this function.
+ * the @dev_priv->cmdbuf_mutex needs to be held by the current thread before
+ * calling this function.
  */
 void __vmw_execbuf_release_pinned_bo(struct vmw_private *dev_priv,
 				     struct vmw_fence_obj *fence)
@@ -4153,35 +3942,32 @@ void __vmw_execbuf_release_pinned_bo(struct vmw_private *dev_priv,
 
 	vmw_validation_unref_lists(&val_ctx);
 	vmw_bo_unreference(&dev_priv->pinned_bo);
+
 out_unlock:
 	return;
-
 out_no_emit:
 	vmw_validation_bo_backoff(&val_ctx);
 out_no_reserve:
 	vmw_validation_unref_lists(&val_ctx);
 	vmw_execbuf_unpin_panic(dev_priv);
 	vmw_bo_unreference(&dev_priv->pinned_bo);
-
 }
 
 /**
- * vmw_execbuf_release_pinned_bo - Flush queries and unpin the pinned
- * query bo.
+ * vmw_execbuf_release_pinned_bo - Flush queries and unpin the pinned query bo.
  *
  * @dev_priv: The device private structure.
  *
- * This function should be used to unpin the pinned query bo, or
- * as a query barrier when we need to make sure that all queries have
- * finished before the next fifo command. (For example on hardware
- * context destructions where the hardware may otherwise leak unfinished
- * queries).
+ * This function should be used to unpin the pinned query bo, or as a query
+ * barrier when we need to make sure that all queries have finished before the
+ * next fifo command. (For example on hardware context destructions where the
+ * hardware may otherwise leak unfinished queries).
  *
- * This function does not return any failure codes, but make attempts
- * to do safe unpinning in case of errors.
+ * This function does not return any failure codes, but make attempts to do safe
+ * unpinning in case of errors.
  *
- * The function will synchronize on the previous query barrier, and will
- * thus not finish until that barrier has executed.
+ * The function will synchronize on the previous query barrier, and will thus
+ * not finish until that barrier has executed.
  */
 void vmw_execbuf_release_pinned_bo(struct vmw_private *dev_priv)
 {
@@ -4203,8 +3989,8 @@ int vmw_execbuf_ioctl(struct drm_device *dev, unsigned long data,
 	struct dma_fence *in_fence = NULL;
 
 	if (unlikely(size < copy_offset[0])) {
-		DRM_ERROR("Invalid command size, ioctl %d\n",
-			  DRM_VMW_EXECBUF);
+		VMW_DEBUG_USER("Invalid command size, ioctl %d\n",
+			       DRM_VMW_EXECBUF);
 		return -EINVAL;
 	}
 
@@ -4212,23 +3998,19 @@ int vmw_execbuf_ioctl(struct drm_device *dev, unsigned long data,
 		return -EFAULT;
 
 	/*
-	 * Extend the ioctl argument while
-	 * maintaining backwards compatibility:
-	 * We take different code paths depending on the value of
-	 * arg.version.
+	 * Extend the ioctl argument while maintaining backwards compatibility:
+	 * We take different code paths depending on the value of arg.version.
 	 */
-
 	if (unlikely(arg.version > DRM_VMW_EXECBUF_VERSION ||
 		     arg.version == 0)) {
-		DRM_ERROR("Incorrect execbuf version.\n");
+		VMW_DEBUG_USER("Incorrect execbuf version.\n");
 		return -EINVAL;
 	}
 
 	if (arg.version > 1 &&
 	    copy_from_user(&arg.context_handle,
 			   (void __user *) (data + copy_offset[0]),
-			   copy_offset[arg.version - 1] -
-			   copy_offset[0]) != 0)
+			   copy_offset[arg.version - 1] - copy_offset[0]) != 0)
 		return -EFAULT;
 
 	switch (arg.version) {
@@ -4240,13 +4022,12 @@ int vmw_execbuf_ioctl(struct drm_device *dev, unsigned long data,
 		break;
 	}
 
-
 	/* If imported a fence FD from elsewhere, then wait on it */
 	if (arg.flags & DRM_VMW_EXECBUF_FLAG_IMPORT_FENCE_FD) {
 		in_fence = sync_file_get_fence(arg.imported_fence_fd);
 
 		if (!in_fence) {
-			DRM_ERROR("Cannot get imported fence\n");
+			VMW_DEBUG_USER("Cannot get imported fence\n");
 			return -EINVAL;
 		}
 
@@ -4264,8 +4045,8 @@ int vmw_execbuf_ioctl(struct drm_device *dev, unsigned long data,
 				  NULL, arg.command_size, arg.throttle_us,
 				  arg.context_handle,
 				  (void __user *)(unsigned long)arg.fence_rep,
-				  NULL,
-				  arg.flags);
+				  NULL, arg.flags);
+
 	ttm_read_unlock(&dev_priv->reservation_sem);
 	if (unlikely(ret != 0))
 		goto out;
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c b/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c
index 2a9112515f46..972e8fda6d35 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c
@@ -642,12 +642,11 @@ int vmw_fb_init(struct vmw_private *vmw_priv)
 	struct vmw_fb_par *par;
 	struct fb_info *info;
 	unsigned fb_width, fb_height;
-	unsigned fb_bpp, fb_depth, fb_offset, fb_pitch, fb_size;
+	unsigned int fb_bpp, fb_pitch, fb_size;
 	struct drm_display_mode *init_mode;
 	int ret;
 
 	fb_bpp = 32;
-	fb_depth = 24;
 
 	/* XXX As shouldn't these be as well. */
 	fb_width = min(vmw_priv->fb_max_width, (unsigned)2048);
@@ -655,7 +654,6 @@ int vmw_fb_init(struct vmw_private *vmw_priv)
 
 	fb_pitch = fb_width * fb_bpp / 8;
 	fb_size = fb_pitch * fb_height;
-	fb_offset = vmw_read(vmw_priv, SVGA_REG_FB_OFFSET);
 
 	info = framebuffer_alloc(sizeof(*par), device);
 	if (!info)
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c b/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c
index d0fd147ef75f..ff3586cb6851 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c
@@ -395,12 +395,8 @@ void *vmw_fifo_reserve_dx(struct vmw_private *dev_priv, uint32_t bytes,
 		WARN(1, "Command buffer has not been allocated.\n");
 		ret = NULL;
 	}
-	if (IS_ERR_OR_NULL(ret)) {
-		DRM_ERROR("Fifo reserve failure of %u bytes.\n",
-			  (unsigned) bytes);
-		dump_stack();
+	if (IS_ERR_OR_NULL(ret))
 		return NULL;
-	}
 
 	return ret;
 }
@@ -544,7 +540,7 @@ int vmw_fifo_send_fence(struct vmw_private *dev_priv, uint32_t *seqno)
 	int ret = 0;
 	uint32_t bytes = sizeof(u32) + sizeof(*cmd_fence);
 
-	fm = vmw_fifo_reserve(dev_priv, bytes);
+	fm = VMW_FIFO_RESERVE(dev_priv, bytes);
 	if (unlikely(fm == NULL)) {
 		*seqno = atomic_read(&dev_priv->marker_seq);
 		ret = -ENOMEM;
@@ -603,12 +599,9 @@ static int vmw_fifo_emit_dummy_legacy_query(struct vmw_private *dev_priv,
 		SVGA3dCmdWaitForQuery body;
 	} *cmd;
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Out of fifo space for dummy query.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	cmd->header.id = SVGA_3D_CMD_WAIT_FOR_QUERY;
 	cmd->header.size = sizeof(cmd->body);
@@ -652,12 +645,9 @@ static int vmw_fifo_emit_dummy_gb_query(struct vmw_private *dev_priv,
 		SVGA3dCmdWaitForGBQuery body;
 	} *cmd;
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Out of fifo space for dummy query.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	cmd->header.id = SVGA_3D_CMD_WAIT_FOR_GB_QUERY;
 	cmd->header.size = sizeof(cmd->body);
@@ -699,8 +689,3 @@ int vmw_fifo_emit_dummy_query(struct vmw_private *dev_priv,
 
 	return vmw_fifo_emit_dummy_legacy_query(dev_priv, cid);
 }
-
-void *vmw_fifo_reserve(struct vmw_private *dev_priv, uint32_t bytes)
-{
-	return vmw_fifo_reserve_dx(dev_priv, bytes, SVGA3D_INVALID_ID);
-}
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_gmr.c b/drivers/gpu/drm/vmwgfx/vmwgfx_gmr.c
index 007a0cc7f232..ae7acc6f3dda 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_gmr.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_gmr.c
@@ -51,7 +51,7 @@ static int vmw_gmr2_bind(struct vmw_private *dev_priv,
 	uint32_t cmd_size = define_size + remap_size;
 	uint32_t i;
 
-	cmd_orig = cmd = vmw_fifo_reserve(dev_priv, cmd_size);
+	cmd_orig = cmd = VMW_FIFO_RESERVE(dev_priv, cmd_size);
 	if (unlikely(cmd == NULL))
 		return -ENOMEM;
 
@@ -110,11 +110,10 @@ static void vmw_gmr2_unbind(struct vmw_private *dev_priv,
 	uint32_t define_size = sizeof(define_cmd) + 4;
 	uint32_t *cmd;
 
-	cmd = vmw_fifo_reserve(dev_priv, define_size);
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("GMR2 unbind failed.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, define_size);
+	if (unlikely(cmd == NULL))
 		return;
-	}
+
 	define_cmd.gmrId = gmr_id;
 	define_cmd.numPages = 0;
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
index 172a6ba6539c..a15375eb476e 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
@@ -188,7 +188,7 @@ int vmw_get_cap_3d_ioctl(struct drm_device *dev, void *data,
 	struct vmw_fpriv *vmw_fp = vmw_fpriv(file_priv);
 
 	if (unlikely(arg->pad64 != 0 || arg->max_size == 0)) {
-		DRM_ERROR("Illegal GET_3D_CAP argument.\n");
+		VMW_DEBUG_USER("Illegal GET_3D_CAP argument.\n");
 		return -EINVAL;
 	}
 
@@ -268,7 +268,7 @@ int vmw_present_ioctl(struct drm_device *dev, void *data,
 		return 0;
 
 	if (clips_ptr == NULL) {
-		DRM_ERROR("Variable clips_ptr must be specified.\n");
+		VMW_DEBUG_USER("Variable clips_ptr must be specified.\n");
 		ret = -EINVAL;
 		goto out_clips;
 	}
@@ -291,7 +291,7 @@ int vmw_present_ioctl(struct drm_device *dev, void *data,
 
 	fb = drm_framebuffer_lookup(dev, file_priv, arg->fb_id);
 	if (!fb) {
-		DRM_ERROR("Invalid framebuffer id.\n");
+		VMW_DEBUG_USER("Invalid framebuffer id.\n");
 		ret = -ENOENT;
 		goto out_no_fb;
 	}
@@ -351,7 +351,7 @@ int vmw_present_readback_ioctl(struct drm_device *dev, void *data,
 		return 0;
 
 	if (clips_ptr == NULL) {
-		DRM_ERROR("Argument clips_ptr must be specified.\n");
+		VMW_DEBUG_USER("Argument clips_ptr must be specified.\n");
 		ret = -EINVAL;
 		goto out_clips;
 	}
@@ -374,14 +374,14 @@ int vmw_present_readback_ioctl(struct drm_device *dev, void *data,
 
 	fb = drm_framebuffer_lookup(dev, file_priv, arg->fb_id);
 	if (!fb) {
-		DRM_ERROR("Invalid framebuffer id.\n");
+		VMW_DEBUG_USER("Invalid framebuffer id.\n");
 		ret = -ENOENT;
 		goto out_no_fb;
 	}
 
 	vfb = vmw_framebuffer_to_vfb(fb);
 	if (!vfb->bo) {
-		DRM_ERROR("Framebuffer not buffer backed.\n");
+		VMW_DEBUG_USER("Framebuffer not buffer backed.\n");
 		ret = -EINVAL;
 		goto out_no_ttm_lock;
 	}
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
index ed2f67822f45..b97bc8e5944b 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
@@ -64,11 +64,9 @@ static int vmw_cursor_update_image(struct vmw_private *dev_priv,
 	if (!image)
 		return -EINVAL;
 
-	cmd = vmw_fifo_reserve(dev_priv, cmd_size);
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Fifo reserve failed.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, cmd_size);
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	memset(cmd, 0, sizeof(*cmd));
 
@@ -1202,7 +1200,7 @@ static int vmw_create_bo_proxy(struct drm_device *dev,
 	vmw_bo_unreference(&res->backup);
 	res->backup = vmw_bo_reference(bo_mob);
 	res->backup_offset = 0;
-	vmw_resource_unreserve(res, false, NULL, 0);
+	vmw_resource_unreserve(res, false, false, false, NULL, 0);
 	mutex_unlock(&res->dev_priv->cmdbuf_mutex);
 
 	return 0;
@@ -2468,13 +2466,11 @@ int vmw_kms_helper_dirty(struct vmw_private *dev_priv,
 
 		dirty->unit = unit;
 		if (dirty->fifo_reserve_size > 0) {
-			dirty->cmd = vmw_fifo_reserve(dev_priv,
+			dirty->cmd = VMW_FIFO_RESERVE(dev_priv,
 						      dirty->fifo_reserve_size);
-			if (!dirty->cmd) {
-				DRM_ERROR("Couldn't reserve fifo space "
-					  "for dirty blits.\n");
+			if (!dirty->cmd)
 				return -ENOMEM;
-			}
+
 			memset(dirty->cmd, 0, dirty->fifo_reserve_size);
 		}
 		dirty->num_hits = 0;
@@ -2604,12 +2600,9 @@ int vmw_kms_update_proxy(struct vmw_resource *res,
 	if (!clips)
 		return 0;
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd) * num_clips);
-	if (!cmd) {
-		DRM_ERROR("Couldn't reserve fifo space for proxy surface "
-			  "update.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd) * num_clips);
+	if (!cmd)
 		return -ENOMEM;
-	}
 
 	for (i = 0; i < num_clips; ++i, clips += increment, ++cmd) {
 		box = &cmd->body.box;
@@ -2827,7 +2820,8 @@ int vmw_du_helper_plane_update(struct vmw_du_update_plane *update)
 			container_of(update->vfb, typeof(*vfbs), base);
 
 		ret = vmw_validation_add_resource(&val_ctx, &vfbs->surface->res,
-						  0, NULL, NULL);
+						  0, VMW_RES_DIRTY_NONE, NULL,
+						  NULL);
 	}
 
 	if (ret)
@@ -2838,7 +2832,7 @@ int vmw_du_helper_plane_update(struct vmw_du_update_plane *update)
 		goto out_unref;
 
 	reserved_size = update->calc_fifo_size(update, num_hits);
-	cmd_start = vmw_fifo_reserve(update->dev_priv, reserved_size);
+	cmd_start = VMW_FIFO_RESERVE(update->dev_priv, reserved_size);
 	if (!cmd_start) {
 		ret = -ENOMEM;
 		goto out_revert;
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c
index 16be515c4c0f..25e6343bcf21 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c
@@ -554,11 +554,9 @@ int vmw_kms_ldu_do_bo_dirty(struct vmw_private *dev_priv,
 	} *cmd;
 
 	fifo_size = sizeof(*cmd) * num_clips;
-	cmd = vmw_fifo_reserve(dev_priv, fifo_size);
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Fifo reserve failed.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, fifo_size);
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	memset(cmd, 0, fifo_size);
 	for (i = 0; i < num_clips; i++, clips += increment) {
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c b/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c
index d83cc66e1210..406edc8cef35 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c
@@ -146,9 +146,8 @@ static int vmw_setup_otable_base(struct vmw_private *dev_priv,
 		mob->pt_level += VMW_MOBFMT_PTDEPTH_1 - SVGA3D_MOBFMT_PTDEPTH_1;
 	}
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
 	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for OTable setup.\n");
 		ret = -ENOMEM;
 		goto out_no_fifo;
 	}
@@ -202,12 +201,9 @@ static void vmw_takedown_otable_base(struct vmw_private *dev_priv,
 		return;
 
 	bo = otable->page_table->pt_bo;
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for OTable "
-			  "takedown.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (unlikely(cmd == NULL))
 		return;
-	}
 
 	memset(cmd, 0, sizeof(*cmd));
 	cmd->header.id = SVGA_3D_CMD_SET_OTABLE_BASE;
@@ -614,16 +610,14 @@ void vmw_mob_unbind(struct vmw_private *dev_priv,
 		BUG_ON(ret != 0);
 	}
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for Memory "
-			  "Object unbinding.\n");
-	} else {
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (cmd) {
 		cmd->header.id = SVGA_3D_CMD_DESTROY_GB_MOB;
 		cmd->header.size = sizeof(cmd->body);
 		cmd->body.mobid = mob->id;
 		vmw_fifo_commit(dev_priv, sizeof(*cmd));
 	}
+
 	if (bo) {
 		vmw_bo_fence_single(bo, NULL);
 		ttm_bo_unreserve(bo);
@@ -683,12 +677,9 @@ int vmw_mob_bind(struct vmw_private *dev_priv,
 
 	vmw_fifo_resource_inc(dev_priv);
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for Memory "
-			  "Object binding.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (unlikely(cmd == NULL))
 		goto out_no_cmd_space;
-	}
 
 	cmd->header.id = SVGA_3D_CMD_DEFINE_GB_MOB64;
 	cmd->header.size = sizeof(cmd->body);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c b/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c
index 9f1b9d289bec..d5ef8cf802de 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c
@@ -124,7 +124,7 @@ static int vmw_overlay_send_put(struct vmw_private *dev_priv,
 
 	fifo_size = sizeof(*cmds) + sizeof(*flush) + sizeof(*items) * num_items;
 
-	cmds = vmw_fifo_reserve(dev_priv, fifo_size);
+	cmds = VMW_FIFO_RESERVE(dev_priv, fifo_size);
 	/* hardware has hung, can't do anything here */
 	if (!cmds)
 		return -ENOMEM;
@@ -194,7 +194,7 @@ static int vmw_overlay_send_stop(struct vmw_private *dev_priv,
 	int ret;
 
 	for (;;) {
-		cmds = vmw_fifo_reserve(dev_priv, sizeof(*cmds));
+		cmds = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmds));
 		if (cmds)
 			break;
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
index a7c30e567f09..711f8fd0dd45 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
@@ -365,14 +365,6 @@ static int vmw_resource_do_validate(struct vmw_resource *res,
 			list_add_tail(&res->mob_head, &res->backup->res_list);
 	}
 
-	/*
-	 * Only do this on write operations, and move to
-	 * vmw_resource_unreserve if it can be called after
-	 * backup buffers have been unreserved. Otherwise
-	 * sort out locking.
-	 */
-	res->res_dirty = true;
-
 	return 0;
 
 out_bind_failed:
@@ -386,6 +378,8 @@ out_bind_failed:
  * command submission.
  *
  * @res:               Pointer to the struct vmw_resource to unreserve.
+ * @dirty_set:         Change dirty status of the resource.
+ * @dirty:             When changing dirty status indicates the new status.
  * @switch_backup:     Backup buffer has been switched.
  * @new_backup:        Pointer to new backup buffer if command submission
  *                     switched. May be NULL.
@@ -395,6 +389,8 @@ out_bind_failed:
  * resource lru list, so that it can be evicted if necessary.
  */
 void vmw_resource_unreserve(struct vmw_resource *res,
+			    bool dirty_set,
+			    bool dirty,
 			    bool switch_backup,
 			    struct vmw_buffer_object *new_backup,
 			    unsigned long new_backup_offset)
@@ -422,6 +418,9 @@ void vmw_resource_unreserve(struct vmw_resource *res,
 	if (switch_backup)
 		res->backup_offset = new_backup_offset;
 
+	if (dirty_set)
+		res->res_dirty = dirty;
+
 	if (!res->func->may_evict || res->id == -1 || res->pin_count)
 		return;
 
@@ -696,7 +695,7 @@ void vmw_resource_unbind_list(struct vmw_buffer_object *vbo)
 		if (!res->func->unbind)
 			continue;
 
-		(void) res->func->unbind(res, true, &val_buf);
+		(void) res->func->unbind(res, res->res_dirty, &val_buf);
 		res->backup_dirty = true;
 		res->res_dirty = false;
 		list_del_init(&res->mob_head);
@@ -731,12 +730,9 @@ int vmw_query_readback_all(struct vmw_buffer_object *dx_query_mob)
 	dx_query_ctx = dx_query_mob->dx_query_ctx;
 	dev_priv     = dx_query_ctx->dev_priv;
 
-	cmd = vmw_fifo_reserve_dx(dev_priv, sizeof(*cmd), dx_query_ctx->id);
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for "
-			  "query MOB read back.\n");
+	cmd = VMW_FIFO_RESERVE_DX(dev_priv, sizeof(*cmd), dx_query_ctx->id);
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	cmd->header.id   = SVGA_3D_CMD_DX_READBACK_ALL_QUERY;
 	cmd->header.size = sizeof(cmd->body);
@@ -932,7 +928,7 @@ int vmw_resource_pin(struct vmw_resource *res, bool interruptible)
 	res->pin_count++;
 
 out_no_validate:
-	vmw_resource_unreserve(res, false, NULL, 0UL);
+	vmw_resource_unreserve(res, false, false, false, NULL, 0UL);
 out_no_reserve:
 	mutex_unlock(&dev_priv->cmdbuf_mutex);
 	ttm_write_unlock(&dev_priv->reservation_sem);
@@ -968,7 +964,7 @@ void vmw_resource_unpin(struct vmw_resource *res)
 		ttm_bo_unreserve(&vbo->base);
 	}
 
-	vmw_resource_unreserve(res, false, NULL, 0UL);
+	vmw_resource_unreserve(res, false, false, false, NULL, 0UL);
 
 	mutex_unlock(&dev_priv->cmdbuf_mutex);
 	ttm_read_unlock(&dev_priv->reservation_sem);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c
index cd586c52af7e..9a2a3836d89a 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c
@@ -130,12 +130,9 @@ static int vmw_sou_fifo_create(struct vmw_private *dev_priv,
 	BUG_ON(!sou->buffer);
 
 	fifo_size = sizeof(*cmd);
-	cmd = vmw_fifo_reserve(dev_priv, fifo_size);
-	/* The hardware has hung, nothing we can do about it here. */
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Fifo reserve failed.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, fifo_size);
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	memset(cmd, 0, fifo_size);
 	cmd->header.cmdType = SVGA_CMD_DEFINE_SCREEN;
@@ -182,12 +179,9 @@ static int vmw_sou_fifo_destroy(struct vmw_private *dev_priv,
 		return 0;
 
 	fifo_size = sizeof(*cmd);
-	cmd = vmw_fifo_reserve(dev_priv, fifo_size);
-	/* the hardware has hung, nothing we can do about it here */
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Fifo reserve failed.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, fifo_size);
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	memset(cmd, 0, fifo_size);
 	cmd->header.cmdType = SVGA_CMD_DESTROY_SCREEN;
@@ -998,11 +992,9 @@ static int do_bo_define_gmrfb(struct vmw_private *dev_priv,
 	if (depth == 32)
 		depth = 24;
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-	if (!cmd) {
-		DRM_ERROR("Out of fifo space for dirty framebuffer command.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (!cmd)
 		return -ENOMEM;
-	}
 
 	cmd->header = SVGA_CMD_DEFINE_GMRFB;
 	cmd->body.format.bitsPerPixel = framebuffer->base.format->cpp[0] * 8;
@@ -1148,7 +1140,8 @@ int vmw_kms_sou_do_surface_dirty(struct vmw_private *dev_priv,
 	if (!srf)
 		srf = &vfbs->surface->res;
 
-	ret = vmw_validation_add_resource(&val_ctx, srf, 0, NULL, NULL);
+	ret = vmw_validation_add_resource(&val_ctx, srf, 0, VMW_RES_DIRTY_NONE,
+					  NULL, NULL);
 	if (ret)
 		return ret;
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c
index bf32fe446219..d310d21f0d54 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c
@@ -218,10 +218,8 @@ static int vmw_gb_shader_create(struct vmw_resource *res)
 		goto out_no_fifo;
 	}
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
 	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for shader "
-			  "creation.\n");
 		ret = -ENOMEM;
 		goto out_no_fifo;
 	}
@@ -254,12 +252,9 @@ static int vmw_gb_shader_bind(struct vmw_resource *res,
 
 	BUG_ON(bo->mem.mem_type != VMW_PL_MOB);
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for shader "
-			  "binding.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	cmd->header.id = SVGA_3D_CMD_BIND_GB_SHADER;
 	cmd->header.size = sizeof(cmd->body);
@@ -285,12 +280,9 @@ static int vmw_gb_shader_unbind(struct vmw_resource *res,
 
 	BUG_ON(res->backup->base.mem.mem_type != VMW_PL_MOB);
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for shader "
-			  "unbinding.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	cmd->header.id = SVGA_3D_CMD_BIND_GB_SHADER;
 	cmd->header.size = sizeof(cmd->body);
@@ -328,10 +320,8 @@ static int vmw_gb_shader_destroy(struct vmw_resource *res)
 	mutex_lock(&dev_priv->binding_mutex);
 	vmw_binding_res_list_scrub(&res->binding_head);
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
 	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for shader "
-			  "destruction.\n");
 		mutex_unlock(&dev_priv->binding_mutex);
 		return -ENOMEM;
 	}
@@ -400,13 +390,9 @@ static int vmw_dx_shader_unscrub(struct vmw_resource *res)
 	if (!list_empty(&shader->cotable_head) || !shader->committed)
 		return 0;
 
-	cmd = vmw_fifo_reserve_dx(dev_priv, sizeof(*cmd),
-				  shader->ctx->id);
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for shader "
-			  "scrubbing.\n");
+	cmd = VMW_FIFO_RESERVE_DX(dev_priv, sizeof(*cmd), shader->ctx->id);
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	cmd->header.id = SVGA_3D_CMD_DX_BIND_SHADER;
 	cmd->header.size = sizeof(cmd->body);
@@ -491,12 +477,9 @@ static int vmw_dx_shader_scrub(struct vmw_resource *res)
 		return 0;
 
 	WARN_ON_ONCE(!shader->committed);
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Failed reserving FIFO space for shader "
-			  "scrubbing.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	cmd->header.id = SVGA_3D_CMD_DX_BIND_SHADER;
 	cmd->header.size = sizeof(cmd->body);
@@ -865,14 +848,13 @@ static int vmw_shader_define(struct drm_device *dev, struct drm_file *file_priv,
 		ret = vmw_user_bo_lookup(tfile, buffer_handle,
 					     &buffer, NULL);
 		if (unlikely(ret != 0)) {
-			DRM_ERROR("Could not find buffer for shader "
-				  "creation.\n");
+			VMW_DEBUG_USER("Couldn't find buffer for shader creation.\n");
 			return ret;
 		}
 
 		if ((u64)buffer->base.num_pages * PAGE_SIZE <
 		    (u64)size + (u64)offset) {
-			DRM_ERROR("Illegal buffer- or shader size.\n");
+			VMW_DEBUG_USER("Illegal buffer- or shader size.\n");
 			ret = -EINVAL;
 			goto out_bad_arg;
 		}
@@ -886,7 +868,7 @@ static int vmw_shader_define(struct drm_device *dev, struct drm_file *file_priv,
 		shader_type = SVGA3D_SHADERTYPE_PS;
 		break;
 	default:
-		DRM_ERROR("Illegal shader type.\n");
+		VMW_DEBUG_USER("Illegal shader type.\n");
 		ret = -EINVAL;
 		goto out_bad_arg;
 	}
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_simple_resource.c b/drivers/gpu/drm/vmwgfx/vmwgfx_simple_resource.c
index 6a6865384e91..73e9a487e659 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_simple_resource.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_simple_resource.c
@@ -239,17 +239,17 @@ vmw_simple_resource_lookup(struct ttm_object_file *tfile,
 
 	base = ttm_base_object_lookup(tfile, handle);
 	if (!base) {
-		DRM_ERROR("Invalid %s handle 0x%08lx.\n",
-			  func->res_func.type_name,
-			  (unsigned long) handle);
+		VMW_DEBUG_USER("Invalid %s handle 0x%08lx.\n",
+			       func->res_func.type_name,
+			       (unsigned long) handle);
 		return ERR_PTR(-ESRCH);
 	}
 
 	if (ttm_base_object_type(base) != func->ttm_res_type) {
 		ttm_base_object_unref(&base);
-		DRM_ERROR("Invalid type of %s handle 0x%08lx.\n",
-			  func->res_func.type_name,
-			  (unsigned long) handle);
+		VMW_DEBUG_USER("Invalid type of %s handle 0x%08lx.\n",
+			       func->res_func.type_name,
+			       (unsigned long) handle);
 		return ERR_PTR(-EINVAL);
 	}
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_so.c b/drivers/gpu/drm/vmwgfx/vmwgfx_so.c
index bc8bb690f1ea..63807361e16f 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_so.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_so.c
@@ -170,13 +170,12 @@ static int vmw_view_create(struct vmw_resource *res)
 		return 0;
 	}
 
-	cmd = vmw_fifo_reserve_dx(res->dev_priv, view->cmd_size,
-				  view->ctx->id);
+	cmd = VMW_FIFO_RESERVE_DX(res->dev_priv, view->cmd_size, view->ctx->id);
 	if (!cmd) {
-		DRM_ERROR("Failed reserving FIFO space for view creation.\n");
 		mutex_unlock(&dev_priv->binding_mutex);
 		return -ENOMEM;
 	}
+
 	memcpy(cmd, &view->cmd, view->cmd_size);
 	WARN_ON(cmd->body.view_id != view->view_id);
 	/* Sid may have changed due to surface eviction. */
@@ -214,12 +213,9 @@ static int vmw_view_destroy(struct vmw_resource *res)
 	if (!view->committed || res->id == -1)
 		return 0;
 
-	cmd = vmw_fifo_reserve_dx(dev_priv, sizeof(*cmd), view->ctx->id);
-	if (!cmd) {
-		DRM_ERROR("Failed reserving FIFO space for view "
-			  "destruction.\n");
+	cmd = VMW_FIFO_RESERVE_DX(dev_priv, sizeof(*cmd), view->ctx->id);
+	if (!cmd)
 		return -ENOMEM;
-	}
 
 	cmd->header.id = vmw_view_destroy_cmds[view->view_type];
 	cmd->header.size = sizeof(cmd->body);
@@ -338,12 +334,12 @@ int vmw_view_add(struct vmw_cmdbuf_res_manager *man,
 
 	if (cmd_size != vmw_view_define_sizes[view_type] +
 	    sizeof(SVGA3dCmdHeader)) {
-		DRM_ERROR("Illegal view create command size.\n");
+		VMW_DEBUG_USER("Illegal view create command size.\n");
 		return -EINVAL;
 	}
 
 	if (!vmw_view_id_ok(user_key, view_type)) {
-		DRM_ERROR("Illegal view add view id.\n");
+		VMW_DEBUG_USER("Illegal view add view id.\n");
 		return -EINVAL;
 	}
 
@@ -352,8 +348,7 @@ int vmw_view_add(struct vmw_cmdbuf_res_manager *man,
 	ret = ttm_mem_global_alloc(vmw_mem_glob(dev_priv), size, &ttm_opt_ctx);
 	if (ret) {
 		if (ret != -ERESTARTSYS)
-			DRM_ERROR("Out of graphics memory for view"
-				  " creation.\n");
+			DRM_ERROR("Out of graphics memory for view creation\n");
 		return ret;
 	}
 
@@ -413,7 +408,7 @@ int vmw_view_remove(struct vmw_cmdbuf_res_manager *man,
 		    struct vmw_resource **res_p)
 {
 	if (!vmw_view_id_ok(user_key, view_type)) {
-		DRM_ERROR("Illegal view remove view id.\n");
+		VMW_DEBUG_USER("Illegal view remove view id.\n");
 		return -EINVAL;
 	}
 
@@ -497,6 +492,30 @@ struct vmw_resource *vmw_view_lookup(struct vmw_cmdbuf_res_manager *man,
 				     vmw_view_key(user_key, view_type));
 }
 
+/**
+ * vmw_view_dirtying - Return whether a view type is dirtying its resource
+ * @res: Pointer to the view
+ *
+ * Each time a resource is put on the validation list as the result of a
+ * view pointing to it, we need to determine whether that resource will
+ * be dirtied (written to by the GPU) as a result of the corresponding
+ * GPU operation. Currently only rendertarget- and depth-stencil views are
+ * capable of dirtying its resource.
+ *
+ * Return: Whether the view type of @res dirties the resource it points to.
+ */
+u32 vmw_view_dirtying(struct vmw_resource *res)
+{
+	static u32 view_is_dirtying[vmw_view_max] = {
+		[vmw_view_rt] = VMW_RES_DIRTY_SET,
+		[vmw_view_ds] = VMW_RES_DIRTY_SET,
+	};
+
+	/* Update this function as we add more view types */
+	BUILD_BUG_ON(vmw_view_max != 3);
+	return view_is_dirtying[vmw_view(res)->view_type];
+}
+
 const u32 vmw_view_destroy_cmds[] = {
 	[vmw_view_sr] = SVGA_3D_CMD_DX_DESTROY_SHADERRESOURCE_VIEW,
 	[vmw_view_rt] = SVGA_3D_CMD_DX_DESTROY_RENDERTARGET_VIEW,
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_so.h b/drivers/gpu/drm/vmwgfx/vmwgfx_so.h
index b80c7252f2fd..12565047bc55 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_so.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_so.h
@@ -157,4 +157,5 @@ extern struct vmw_resource *vmw_view_srf(struct vmw_resource *res);
 extern struct vmw_resource *vmw_view_lookup(struct vmw_cmdbuf_res_manager *man,
 					    enum vmw_view_type view_type,
 					    u32 user_key);
+extern u32 vmw_view_dirtying(struct vmw_resource *res);
 #endif
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c
index 096c2941a8e4..f803bb5e782b 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c
@@ -111,7 +111,7 @@ struct vmw_stdu_update_gb_image {
  */
 struct vmw_screen_target_display_unit {
 	struct vmw_display_unit base;
-	const struct vmw_surface *display_srf;
+	struct vmw_surface *display_srf;
 	enum stdu_content_type content_fb_type;
 	s32 display_width, display_height;
 
@@ -167,12 +167,9 @@ static int vmw_stdu_define_st(struct vmw_private *dev_priv,
 		SVGA3dCmdDefineGBScreenTarget body;
 	} *cmd;
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Out of FIFO space defining Screen Target\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	cmd->header.id   = SVGA_3D_CMD_DEFINE_GB_SCREENTARGET;
 	cmd->header.size = sizeof(cmd->body);
@@ -229,12 +226,9 @@ static int vmw_stdu_bind_st(struct vmw_private *dev_priv,
 	memset(&image, 0, sizeof(image));
 	image.sid = res ? res->id : SVGA3D_INVALID_ID;
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Out of FIFO space binding a screen target\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	cmd->header.id   = SVGA_3D_CMD_BIND_GB_SCREENTARGET;
 	cmd->header.size = sizeof(cmd->body);
@@ -296,12 +290,9 @@ static int vmw_stdu_update_st(struct vmw_private *dev_priv,
 		return -EINVAL;
 	}
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Out of FIFO space updating a Screen Target\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	vmw_stdu_populate_update(cmd, stdu->base.unit,
 				 0, stdu->display_width,
@@ -335,12 +326,9 @@ static int vmw_stdu_destroy_st(struct vmw_private *dev_priv,
 	if (unlikely(!stdu->defined))
 		return 0;
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-
-	if (unlikely(cmd == NULL)) {
-		DRM_ERROR("Out of FIFO space, screen target not destroyed\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+	if (unlikely(cmd == NULL))
 		return -ENOMEM;
-	}
 
 	cmd->header.id   = SVGA_3D_CMD_DESTROY_GB_SCREENTARGET;
 	cmd->header.size = sizeof(cmd->body);
@@ -533,6 +521,7 @@ static void vmw_stdu_bo_fifo_commit(struct vmw_kms_dirty *dirty)
 
 	vmw_fifo_commit(dirty->dev_priv, sizeof(*cmd) + blit_size);
 
+	stdu->display_srf->res.res_dirty = true;
 	ddirty->left = ddirty->top = S32_MAX;
 	ddirty->right = ddirty->bottom = S32_MIN;
 }
@@ -629,20 +618,16 @@ static void vmw_stdu_bo_cpu_commit(struct vmw_kms_dirty *dirty)
 		region.x2 = diff.rect.x2;
 		region.y1 = diff.rect.y1;
 		region.y2 = diff.rect.y2;
-		ret = vmw_kms_update_proxy(
-			(struct vmw_resource *) &stdu->display_srf->res,
-			(const struct drm_clip_rect *) &region, 1, 1);
+		ret = vmw_kms_update_proxy(&stdu->display_srf->res, &region,
+					   1, 1);
 		if (ret)
 			goto out_cleanup;
 
 
 		dev_priv = vmw_priv(stdu->base.crtc.dev);
-		cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
-
-		if (!cmd) {
-			DRM_ERROR("Cannot reserve FIFO space to update STDU");
+		cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
+		if (!cmd)
 			goto out_cleanup;
-		}
 
 		vmw_stdu_populate_update(cmd, stdu->base.unit,
 					 region.x1, region.x2,
@@ -820,6 +805,7 @@ static void vmw_kms_stdu_surface_fifo_commit(struct vmw_kms_dirty *dirty)
 		cmd->body.dest.sid = stdu->display_srf->res.id;
 		update = (struct vmw_stdu_update *) &blit[dirty->num_hits];
 		commit_size = sizeof(*cmd) + blit_size + sizeof(*update);
+		stdu->display_srf->res.res_dirty = true;
 	} else {
 		update = dirty->cmd;
 		commit_size = sizeof(*update);
@@ -876,7 +862,8 @@ int vmw_kms_stdu_surface_dirty(struct vmw_private *dev_priv,
 	if (!srf)
 		srf = &vfbs->surface->res;
 
-	ret = vmw_validation_add_resource(&val_ctx, srf, 0, NULL, NULL);
+	ret = vmw_validation_add_resource(&val_ctx, srf, 0, VMW_RES_DIRTY_NONE,
+					  NULL, NULL);
 	if (ret)
 		return ret;
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c
index ef09f7edf931..219471903bc1 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c
@@ -342,12 +342,9 @@ static void vmw_hw_surface_destroy(struct vmw_resource *res)
 
 	if (res->id != -1) {
 
-		cmd = vmw_fifo_reserve(dev_priv, vmw_surface_destroy_size());
-		if (unlikely(!cmd)) {
-			DRM_ERROR("Failed reserving FIFO space for surface "
-				  "destruction.\n");
+		cmd = VMW_FIFO_RESERVE(dev_priv, vmw_surface_destroy_size());
+		if (unlikely(!cmd))
 			return;
-		}
 
 		vmw_surface_destroy_encode(res->id, cmd);
 		vmw_fifo_commit(dev_priv, vmw_surface_destroy_size());
@@ -414,10 +411,8 @@ static int vmw_legacy_srf_create(struct vmw_resource *res)
 	 */
 
 	submit_size = vmw_surface_define_size(srf);
-	cmd = vmw_fifo_reserve(dev_priv, submit_size);
+	cmd = VMW_FIFO_RESERVE(dev_priv, submit_size);
 	if (unlikely(!cmd)) {
-		DRM_ERROR("Failed reserving FIFO space for surface "
-			  "creation.\n");
 		ret = -ENOMEM;
 		goto out_no_fifo;
 	}
@@ -468,12 +463,10 @@ static int vmw_legacy_srf_dma(struct vmw_resource *res,
 
 	BUG_ON(!val_buf->bo);
 	submit_size = vmw_surface_dma_size(srf);
-	cmd = vmw_fifo_reserve(dev_priv, submit_size);
-	if (unlikely(!cmd)) {
-		DRM_ERROR("Failed reserving FIFO space for surface "
-			  "DMA.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, submit_size);
+	if (unlikely(!cmd))
 		return -ENOMEM;
-	}
+
 	vmw_bo_get_guest_ptr(val_buf->bo, &ptr);
 	vmw_surface_dma_encode(srf, cmd, &ptr, bind);
 
@@ -556,12 +549,9 @@ static int vmw_legacy_srf_destroy(struct vmw_resource *res)
 	 */
 
 	submit_size = vmw_surface_destroy_size();
-	cmd = vmw_fifo_reserve(dev_priv, submit_size);
-	if (unlikely(!cmd)) {
-		DRM_ERROR("Failed reserving FIFO space for surface "
-			  "eviction.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, submit_size);
+	if (unlikely(!cmd))
 		return -ENOMEM;
-	}
 
 	vmw_surface_destroy_encode(res->id, cmd);
 	vmw_fifo_commit(dev_priv, submit_size);
@@ -748,11 +738,10 @@ int vmw_surface_define_ioctl(struct drm_device *dev, void *data,
 		ttm_round_pot(num_sizes * sizeof(struct drm_vmw_size)) +
 		ttm_round_pot(num_sizes * sizeof(struct vmw_surface_offset));
 
-
 	desc = svga3dsurface_get_desc(req->format);
 	if (unlikely(desc->block_desc == SVGA3DBLOCKDESC_NONE)) {
-		DRM_ERROR("Invalid surface format for surface creation.\n");
-		DRM_ERROR("Format requested is: %d\n", req->format);
+		VMW_DEBUG_USER("Invalid format %d for surface creation.\n",
+			       req->format);
 		return -EINVAL;
 	}
 
@@ -764,8 +753,7 @@ int vmw_surface_define_ioctl(struct drm_device *dev, void *data,
 				   size, &ctx);
 	if (unlikely(ret != 0)) {
 		if (ret != -ERESTARTSYS)
-			DRM_ERROR("Out of graphics memory for surface"
-				  " creation.\n");
+			DRM_ERROR("Out of graphics memory for surface.\n");
 		goto out_unlock;
 	}
 
@@ -939,12 +927,12 @@ vmw_surface_handle_reference(struct vmw_private *dev_priv,
 	ret = -EINVAL;
 	base = ttm_base_object_lookup_for_ref(dev_priv->tdev, handle);
 	if (unlikely(!base)) {
-		DRM_ERROR("Could not find surface to reference.\n");
+		VMW_DEBUG_USER("Could not find surface to reference.\n");
 		goto out_no_lookup;
 	}
 
 	if (unlikely(ttm_base_object_type(base) != VMW_RES_SURFACE)) {
-		DRM_ERROR("Referenced object is not a surface.\n");
+		VMW_DEBUG_USER("Referenced object is not a surface.\n");
 		goto out_bad_resource;
 	}
 
@@ -1022,8 +1010,8 @@ int vmw_surface_reference_ioctl(struct drm_device *dev, void *data,
 		ret = copy_to_user(user_sizes, &srf->base_size,
 				   sizeof(srf->base_size));
 	if (unlikely(ret != 0)) {
-		DRM_ERROR("copy_to_user failed %p %u\n",
-			  user_sizes, srf->num_sizes);
+		VMW_DEBUG_USER("copy_to_user failed %p %u\n", user_sizes,
+			       srf->num_sizes);
 		ttm_ref_object_base_unref(tfile, base->handle, TTM_REF_USAGE);
 		ret = -EFAULT;
 	}
@@ -1088,12 +1076,10 @@ static int vmw_gb_surface_create(struct vmw_resource *res)
 		submit_len = sizeof(*cmd);
 	}
 
-	cmd = vmw_fifo_reserve(dev_priv, submit_len);
+	cmd = VMW_FIFO_RESERVE(dev_priv, submit_len);
 	cmd2 = (typeof(cmd2))cmd;
 	cmd3 = (typeof(cmd3))cmd;
 	if (unlikely(!cmd)) {
-		DRM_ERROR("Failed reserving FIFO space for surface "
-			  "creation.\n");
 		ret = -ENOMEM;
 		goto out_no_fifo;
 	}
@@ -1171,12 +1157,9 @@ static int vmw_gb_surface_bind(struct vmw_resource *res,
 
 	submit_size = sizeof(*cmd1) + (res->backup_dirty ? sizeof(*cmd2) : 0);
 
-	cmd1 = vmw_fifo_reserve(dev_priv, submit_size);
-	if (unlikely(!cmd1)) {
-		DRM_ERROR("Failed reserving FIFO space for surface "
-			  "binding.\n");
+	cmd1 = VMW_FIFO_RESERVE(dev_priv, submit_size);
+	if (unlikely(!cmd1))
 		return -ENOMEM;
-	}
 
 	cmd1->header.id = SVGA_3D_CMD_BIND_GB_SURFACE;
 	cmd1->header.size = sizeof(cmd1->body);
@@ -1221,12 +1204,9 @@ static int vmw_gb_surface_unbind(struct vmw_resource *res,
 	BUG_ON(bo->mem.mem_type != VMW_PL_MOB);
 
 	submit_size = sizeof(*cmd3) + (readback ? sizeof(*cmd1) : sizeof(*cmd2));
-	cmd = vmw_fifo_reserve(dev_priv, submit_size);
-	if (unlikely(!cmd)) {
-		DRM_ERROR("Failed reserving FIFO space for surface "
-			  "unbinding.\n");
+	cmd = VMW_FIFO_RESERVE(dev_priv, submit_size);
+	if (unlikely(!cmd))
 		return -ENOMEM;
-	}
 
 	if (readback) {
 		cmd1 = (void *) cmd;
@@ -1280,10 +1260,8 @@ static int vmw_gb_surface_destroy(struct vmw_resource *res)
 	vmw_view_surface_list_destroy(dev_priv, &srf->view_list);
 	vmw_binding_res_list_scrub(&res->binding_head);
 
-	cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd));
+	cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd));
 	if (unlikely(!cmd)) {
-		DRM_ERROR("Failed reserving FIFO space for surface "
-			  "destruction.\n");
 		mutex_unlock(&dev_priv->binding_mutex);
 		return -ENOMEM;
 	}
@@ -1405,16 +1383,16 @@ int vmw_surface_gb_priv_define(struct drm_device *dev,
 
 	if (for_scanout) {
 		if (!svga3dsurface_is_screen_target_format(format)) {
-			DRM_ERROR("Invalid Screen Target surface format.");
+			VMW_DEBUG_USER("Invalid Screen Target surface format.");
 			return -EINVAL;
 		}
 
 		if (size.width > dev_priv->texture_max_width ||
 		    size.height > dev_priv->texture_max_height) {
-			DRM_ERROR("%ux%u\n, exceeds max surface size %ux%u",
-				  size.width, size.height,
-				  dev_priv->texture_max_width,
-				  dev_priv->texture_max_height);
+			VMW_DEBUG_USER("%ux%u\n, exceeds max surface size %ux%u",
+				       size.width, size.height,
+				       dev_priv->texture_max_width,
+				       dev_priv->texture_max_height);
 			return -EINVAL;
 		}
 	} else {
@@ -1422,14 +1400,14 @@ int vmw_surface_gb_priv_define(struct drm_device *dev,
 
 		desc = svga3dsurface_get_desc(format);
 		if (unlikely(desc->block_desc == SVGA3DBLOCKDESC_NONE)) {
-			DRM_ERROR("Invalid surface format.\n");
+			VMW_DEBUG_USER("Invalid surface format.\n");
 			return -EINVAL;
 		}
 	}
 
 	/* array_size must be null for non-GL3 host. */
 	if (array_size > 0 && !dev_priv->has_dx) {
-		DRM_ERROR("Tried to create DX surface on non-DX host.\n");
+		VMW_DEBUG_USER("Tried to create DX surface on non-DX host.\n");
 		return -EINVAL;
 	}
 
@@ -1651,7 +1629,7 @@ vmw_gb_surface_define_internal(struct drm_device *dev,
 		if (ret == 0) {
 			if (res->backup->base.num_pages * PAGE_SIZE <
 			    res->backup_size) {
-				DRM_ERROR("Surface backup buffer too small.\n");
+				VMW_DEBUG_USER("Surface backup buffer too small.\n");
 				vmw_bo_unreference(&res->backup);
 				ret = -EINVAL;
 				goto out_unlock;
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c
index e6d75e377dd8..8bafa6eac5a8 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c
@@ -30,16 +30,9 @@
 
 int vmw_mmap(struct file *filp, struct vm_area_struct *vma)
 {
-	struct drm_file *file_priv;
-	struct vmw_private *dev_priv;
+	struct drm_file *file_priv = filp->private_data;
+	struct vmw_private *dev_priv = vmw_priv(file_priv->minor->dev);
 
-	if (unlikely(vma->vm_pgoff < VMWGFX_FILE_PAGE_OFFSET)) {
-		DRM_ERROR("Illegal attempt to mmap old fifo space.\n");
-		return -EINVAL;
-	}
-
-	file_priv = filp->private_data;
-	dev_priv = vmw_priv(file_priv->minor->dev);
 	return ttm_bo_mmap(filp, vma, &dev_priv->bdev);
 }
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c b/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c
index e9944ac2e057..f611b2290a1b 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c
@@ -76,6 +76,8 @@ struct vmw_validation_res_node {
 	u32 switching_backup : 1;
 	u32 first_usage : 1;
 	u32 reserved : 1;
+	u32 dirty : 1;
+	u32 dirty_set : 1;
 	unsigned long private[0];
 };
 
@@ -299,6 +301,7 @@ int vmw_validation_add_bo(struct vmw_validation_context *ctx,
  * @ctx: The validation context.
  * @res: The resource.
  * @priv_size: Size of private, additional metadata.
+ * @dirty: Whether to change dirty status.
  * @p_node: Output pointer of additional metadata address.
  * @first_usage: Whether this was the first time this resource was seen.
  *
@@ -307,6 +310,7 @@ int vmw_validation_add_bo(struct vmw_validation_context *ctx,
 int vmw_validation_add_resource(struct vmw_validation_context *ctx,
 				struct vmw_resource *res,
 				size_t priv_size,
+				u32 dirty,
 				void **p_node,
 				bool *first_usage)
 {
@@ -321,8 +325,7 @@ int vmw_validation_add_resource(struct vmw_validation_context *ctx,
 
 	node = vmw_validation_mem_alloc(ctx, sizeof(*node) + priv_size);
 	if (!node) {
-		DRM_ERROR("Failed to allocate a resource validation "
-			  "entry.\n");
+		VMW_DEBUG_USER("Failed to allocate a resource validation entry.\n");
 		return -ENOMEM;
 	}
 
@@ -358,6 +361,11 @@ int vmw_validation_add_resource(struct vmw_validation_context *ctx,
 	}
 
 out_fill:
+	if (dirty) {
+		node->dirty_set = 1;
+		/* Overwriting previous information here is intentional! */
+		node->dirty = (dirty & VMW_RES_DIRTY_SET) ? 1 : 0;
+	}
 	if (first_usage)
 		*first_usage = node->first_usage;
 	if (p_node)
@@ -367,6 +375,29 @@ out_fill:
 }
 
 /**
+ * vmw_validation_res_set_dirty - Register a resource dirty set or clear during
+ * validation.
+ * @ctx: The validation context.
+ * @val_private: The additional meta-data pointer returned when the
+ * resource was registered with the validation context. Used to identify
+ * the resource.
+ * @dirty: Dirty information VMW_RES_DIRTY_XX
+ */
+void vmw_validation_res_set_dirty(struct vmw_validation_context *ctx,
+				  void *val_private, u32 dirty)
+{
+	struct vmw_validation_res_node *val;
+
+	if (!dirty)
+		return;
+
+	val = container_of(val_private, typeof(*val), private);
+	val->dirty_set = 1;
+	/* Overwriting previous information here is intentional! */
+	val->dirty = (dirty & VMW_RES_DIRTY_SET) ? 1 : 0;
+}
+
+/**
  * vmw_validation_res_switch_backup - Register a backup MOB switch during
  * validation.
  * @ctx: The validation context.
@@ -450,15 +481,23 @@ void vmw_validation_res_unreserve(struct vmw_validation_context *ctx,
 	struct vmw_validation_res_node *val;
 
 	list_splice_init(&ctx->resource_ctx_list, &ctx->resource_list);
-
-	list_for_each_entry(val, &ctx->resource_list, head) {
-		if (val->reserved)
-			vmw_resource_unreserve(val->res,
-					       !backoff &&
-					       val->switching_backup,
-					       val->new_backup,
-					       val->new_backup_offset);
-	}
+	if (backoff)
+		list_for_each_entry(val, &ctx->resource_list, head) {
+			if (val->reserved)
+				vmw_resource_unreserve(val->res,
+						       false, false, false,
+						       NULL, 0);
+		}
+	else
+		list_for_each_entry(val, &ctx->resource_list, head) {
+			if (val->reserved)
+				vmw_resource_unreserve(val->res,
+						       val->dirty_set,
+						       val->dirty,
+						       val->switching_backup,
+						       val->new_backup,
+						       val->new_backup_offset);
+		}
 }
 
 /**
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_validation.h b/drivers/gpu/drm/vmwgfx/vmwgfx_validation.h
index 3b396fea40d7..523f6ac5c335 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_validation.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_validation.h
@@ -33,6 +33,10 @@
 #include <linux/ww_mutex.h>
 #include <drm/ttm/ttm_execbuf_util.h>
 
+#define VMW_RES_DIRTY_NONE 0
+#define VMW_RES_DIRTY_SET BIT(0)
+#define VMW_RES_DIRTY_CLEAR BIT(1)
+
 /**
  * struct vmw_validation_mem - Custom interface to provide memory reservations
  * for the validation code.
@@ -237,6 +241,7 @@ void vmw_validation_unref_lists(struct vmw_validation_context *ctx);
 int vmw_validation_add_resource(struct vmw_validation_context *ctx,
 				struct vmw_resource *res,
 				size_t priv_size,
+				u32 dirty,
 				void **p_node,
 				bool *first_usage);
 void vmw_validation_drop_ht(struct vmw_validation_context *ctx);
@@ -261,4 +266,6 @@ void *vmw_validation_mem_alloc(struct vmw_validation_context *ctx,
 int vmw_validation_preload_bo(struct vmw_validation_context *ctx);
 int vmw_validation_preload_res(struct vmw_validation_context *ctx,
 			       unsigned int size);
+void vmw_validation_res_set_dirty(struct vmw_validation_context *ctx,
+				  void *val_private, u32 dirty);
 #endif
diff --git a/drivers/gpu/drm/xen/xen_drm_front.c b/drivers/gpu/drm/xen/xen_drm_front.c
index 3e78a832d7f9..84aa4d61dc42 100644
--- a/drivers/gpu/drm/xen/xen_drm_front.c
+++ b/drivers/gpu/drm/xen/xen_drm_front.c
@@ -582,6 +582,7 @@ static void xen_drm_drv_fini(struct xen_drm_front_info *front_info)
 
 	drm_kms_helper_poll_fini(dev);
 	drm_dev_unplug(dev);
+	drm_dev_put(dev);
 
 	front_info->drm_info = NULL;
 
diff --git a/drivers/iommu/io-pgtable-arm.c b/drivers/iommu/io-pgtable-arm.c
index d3700ec15cbd..4e21efbc4459 100644
--- a/drivers/iommu/io-pgtable-arm.c
+++ b/drivers/iommu/io-pgtable-arm.c
@@ -172,6 +172,10 @@
 #define ARM_LPAE_MAIR_ATTR_IDX_CACHE	1
 #define ARM_LPAE_MAIR_ATTR_IDX_DEV	2
 
+#define ARM_MALI_LPAE_TTBR_ADRMODE_TABLE (3u << 0)
+#define ARM_MALI_LPAE_TTBR_READ_INNER	BIT(2)
+#define ARM_MALI_LPAE_TTBR_SHARE_OUTER	BIT(4)
+
 /* IOPTE accessors */
 #define iopte_deref(pte,d) __va(iopte_to_paddr(pte, d))
 
@@ -180,11 +184,6 @@
 
 #define iopte_prot(pte)	((pte) & ARM_LPAE_PTE_ATTR_MASK)
 
-#define iopte_leaf(pte,l)					\
-	(l == (ARM_LPAE_MAX_LEVELS - 1) ?			\
-		(iopte_type(pte,l) == ARM_LPAE_PTE_TYPE_PAGE) :	\
-		(iopte_type(pte,l) == ARM_LPAE_PTE_TYPE_BLOCK))
-
 struct arm_lpae_io_pgtable {
 	struct io_pgtable	iop;
 
@@ -198,6 +197,15 @@ struct arm_lpae_io_pgtable {
 
 typedef u64 arm_lpae_iopte;
 
+static inline bool iopte_leaf(arm_lpae_iopte pte, int lvl,
+			      enum io_pgtable_fmt fmt)
+{
+	if (lvl == (ARM_LPAE_MAX_LEVELS - 1) && fmt != ARM_MALI_LPAE)
+		return iopte_type(pte, lvl) == ARM_LPAE_PTE_TYPE_PAGE;
+
+	return iopte_type(pte, lvl) == ARM_LPAE_PTE_TYPE_BLOCK;
+}
+
 static arm_lpae_iopte paddr_to_iopte(phys_addr_t paddr,
 				     struct arm_lpae_io_pgtable *data)
 {
@@ -303,12 +311,14 @@ static void __arm_lpae_init_pte(struct arm_lpae_io_pgtable *data,
 	if (data->iop.cfg.quirks & IO_PGTABLE_QUIRK_ARM_NS)
 		pte |= ARM_LPAE_PTE_NS;
 
-	if (lvl == ARM_LPAE_MAX_LEVELS - 1)
+	if (data->iop.fmt != ARM_MALI_LPAE && lvl == ARM_LPAE_MAX_LEVELS - 1)
 		pte |= ARM_LPAE_PTE_TYPE_PAGE;
 	else
 		pte |= ARM_LPAE_PTE_TYPE_BLOCK;
 
-	pte |= ARM_LPAE_PTE_AF | ARM_LPAE_PTE_SH_IS;
+	if (data->iop.fmt != ARM_MALI_LPAE)
+		pte |= ARM_LPAE_PTE_AF;
+	pte |= ARM_LPAE_PTE_SH_IS;
 	pte |= paddr_to_iopte(paddr, data);
 
 	__arm_lpae_set_pte(ptep, pte, &data->iop.cfg);
@@ -321,7 +331,7 @@ static int arm_lpae_init_pte(struct arm_lpae_io_pgtable *data,
 {
 	arm_lpae_iopte pte = *ptep;
 
-	if (iopte_leaf(pte, lvl)) {
+	if (iopte_leaf(pte, lvl, data->iop.fmt)) {
 		/* We require an unmap first */
 		WARN_ON(!selftest_running);
 		return -EEXIST;
@@ -409,7 +419,7 @@ static int __arm_lpae_map(struct arm_lpae_io_pgtable *data, unsigned long iova,
 		__arm_lpae_sync_pte(ptep, cfg);
 	}
 
-	if (pte && !iopte_leaf(pte, lvl)) {
+	if (pte && !iopte_leaf(pte, lvl, data->iop.fmt)) {
 		cptep = iopte_deref(pte, data);
 	} else if (pte) {
 		/* We require an unmap first */
@@ -429,31 +439,37 @@ static arm_lpae_iopte arm_lpae_prot_to_pte(struct arm_lpae_io_pgtable *data,
 	if (data->iop.fmt == ARM_64_LPAE_S1 ||
 	    data->iop.fmt == ARM_32_LPAE_S1) {
 		pte = ARM_LPAE_PTE_nG;
-
 		if (!(prot & IOMMU_WRITE) && (prot & IOMMU_READ))
 			pte |= ARM_LPAE_PTE_AP_RDONLY;
-
 		if (!(prot & IOMMU_PRIV))
 			pte |= ARM_LPAE_PTE_AP_UNPRIV;
-
-		if (prot & IOMMU_MMIO)
-			pte |= (ARM_LPAE_MAIR_ATTR_IDX_DEV
-				<< ARM_LPAE_PTE_ATTRINDX_SHIFT);
-		else if (prot & IOMMU_CACHE)
-			pte |= (ARM_LPAE_MAIR_ATTR_IDX_CACHE
-				<< ARM_LPAE_PTE_ATTRINDX_SHIFT);
 	} else {
 		pte = ARM_LPAE_PTE_HAP_FAULT;
 		if (prot & IOMMU_READ)
 			pte |= ARM_LPAE_PTE_HAP_READ;
 		if (prot & IOMMU_WRITE)
 			pte |= ARM_LPAE_PTE_HAP_WRITE;
+	}
+
+	/*
+	 * Note that this logic is structured to accommodate Mali LPAE
+	 * having stage-1-like attributes but stage-2-like permissions.
+	 */
+	if (data->iop.fmt == ARM_64_LPAE_S2 ||
+	    data->iop.fmt == ARM_32_LPAE_S2) {
 		if (prot & IOMMU_MMIO)
 			pte |= ARM_LPAE_PTE_MEMATTR_DEV;
 		else if (prot & IOMMU_CACHE)
 			pte |= ARM_LPAE_PTE_MEMATTR_OIWB;
 		else
 			pte |= ARM_LPAE_PTE_MEMATTR_NC;
+	} else {
+		if (prot & IOMMU_MMIO)
+			pte |= (ARM_LPAE_MAIR_ATTR_IDX_DEV
+				<< ARM_LPAE_PTE_ATTRINDX_SHIFT);
+		else if (prot & IOMMU_CACHE)
+			pte |= (ARM_LPAE_MAIR_ATTR_IDX_CACHE
+				<< ARM_LPAE_PTE_ATTRINDX_SHIFT);
 	}
 
 	if (prot & IOMMU_NOEXEC)
@@ -511,7 +527,7 @@ static void __arm_lpae_free_pgtable(struct arm_lpae_io_pgtable *data, int lvl,
 	while (ptep != end) {
 		arm_lpae_iopte pte = *ptep++;
 
-		if (!pte || iopte_leaf(pte, lvl))
+		if (!pte || iopte_leaf(pte, lvl, data->iop.fmt))
 			continue;
 
 		__arm_lpae_free_pgtable(data, lvl + 1, iopte_deref(pte, data));
@@ -602,7 +618,7 @@ static size_t __arm_lpae_unmap(struct arm_lpae_io_pgtable *data,
 	if (size == ARM_LPAE_BLOCK_SIZE(lvl, data)) {
 		__arm_lpae_set_pte(ptep, 0, &iop->cfg);
 
-		if (!iopte_leaf(pte, lvl)) {
+		if (!iopte_leaf(pte, lvl, iop->fmt)) {
 			/* Also flush any partial walks */
 			io_pgtable_tlb_add_flush(iop, iova, size,
 						ARM_LPAE_GRANULE(data), false);
@@ -621,7 +637,7 @@ static size_t __arm_lpae_unmap(struct arm_lpae_io_pgtable *data,
 		}
 
 		return size;
-	} else if (iopte_leaf(pte, lvl)) {
+	} else if (iopte_leaf(pte, lvl, iop->fmt)) {
 		/*
 		 * Insert a table at the next level to map the old region,
 		 * minus the part we want to unmap
@@ -669,7 +685,7 @@ static phys_addr_t arm_lpae_iova_to_phys(struct io_pgtable_ops *ops,
 			return 0;
 
 		/* Leaf entry? */
-		if (iopte_leaf(pte,lvl))
+		if (iopte_leaf(pte, lvl, data->iop.fmt))
 			goto found_translation;
 
 		/* Take it to the next level */
@@ -995,6 +1011,32 @@ arm_32_lpae_alloc_pgtable_s2(struct io_pgtable_cfg *cfg, void *cookie)
 	return iop;
 }
 
+static struct io_pgtable *
+arm_mali_lpae_alloc_pgtable(struct io_pgtable_cfg *cfg, void *cookie)
+{
+	struct io_pgtable *iop;
+
+	if (cfg->ias != 48 || cfg->oas > 40)
+		return NULL;
+
+	cfg->pgsize_bitmap &= (SZ_4K | SZ_2M | SZ_1G);
+	iop = arm_64_lpae_alloc_pgtable_s1(cfg, cookie);
+	if (iop) {
+		u64 mair, ttbr;
+
+		/* Copy values as union fields overlap */
+		mair = cfg->arm_lpae_s1_cfg.mair[0];
+		ttbr = cfg->arm_lpae_s1_cfg.ttbr[0];
+
+		cfg->arm_mali_lpae_cfg.memattr = mair;
+		cfg->arm_mali_lpae_cfg.transtab = ttbr |
+			ARM_MALI_LPAE_TTBR_READ_INNER |
+			ARM_MALI_LPAE_TTBR_ADRMODE_TABLE;
+	}
+
+	return iop;
+}
+
 struct io_pgtable_init_fns io_pgtable_arm_64_lpae_s1_init_fns = {
 	.alloc	= arm_64_lpae_alloc_pgtable_s1,
 	.free	= arm_lpae_free_pgtable,
@@ -1015,6 +1057,11 @@ struct io_pgtable_init_fns io_pgtable_arm_32_lpae_s2_init_fns = {
 	.free	= arm_lpae_free_pgtable,
 };
 
+struct io_pgtable_init_fns io_pgtable_arm_mali_lpae_init_fns = {
+	.alloc	= arm_mali_lpae_alloc_pgtable,
+	.free	= arm_lpae_free_pgtable,
+};
+
 #ifdef CONFIG_IOMMU_IO_PGTABLE_LPAE_SELFTEST
 
 static struct io_pgtable_cfg *cfg_cookie;
diff --git a/drivers/iommu/io-pgtable.c b/drivers/iommu/io-pgtable.c
index 93f2880be6c6..5227cfdbb65b 100644
--- a/drivers/iommu/io-pgtable.c
+++ b/drivers/iommu/io-pgtable.c
@@ -30,6 +30,7 @@ io_pgtable_init_table[IO_PGTABLE_NUM_FMTS] = {
 	[ARM_32_LPAE_S2] = &io_pgtable_arm_32_lpae_s2_init_fns,
 	[ARM_64_LPAE_S1] = &io_pgtable_arm_64_lpae_s1_init_fns,
 	[ARM_64_LPAE_S2] = &io_pgtable_arm_64_lpae_s2_init_fns,
+	[ARM_MALI_LPAE] = &io_pgtable_arm_mali_lpae_init_fns,
 #endif
 #ifdef CONFIG_IOMMU_IO_PGTABLE_ARMV7S
 	[ARM_V7S] = &io_pgtable_arm_v7s_init_fns,
diff --git a/drivers/media/platform/vsp1/vsp1_brx.c b/drivers/media/platform/vsp1/vsp1_brx.c
index 58ad248cd7f7..2d86c718a5cf 100644
--- a/drivers/media/platform/vsp1/vsp1_brx.c
+++ b/drivers/media/platform/vsp1/vsp1_brx.c
@@ -283,6 +283,7 @@ static const struct v4l2_subdev_ops brx_ops = {
 
 static void brx_configure_stream(struct vsp1_entity *entity,
 				 struct vsp1_pipeline *pipe,
+				 struct vsp1_dl_list *dl,
 				 struct vsp1_dl_body *dlb)
 {
 	struct vsp1_brx *brx = to_brx(&entity->subdev);
diff --git a/drivers/media/platform/vsp1/vsp1_clu.c b/drivers/media/platform/vsp1/vsp1_clu.c
index 942fc14c19d1..a47b23bf5abf 100644
--- a/drivers/media/platform/vsp1/vsp1_clu.c
+++ b/drivers/media/platform/vsp1/vsp1_clu.c
@@ -171,6 +171,7 @@ static const struct v4l2_subdev_ops clu_ops = {
 
 static void clu_configure_stream(struct vsp1_entity *entity,
 				 struct vsp1_pipeline *pipe,
+				 struct vsp1_dl_list *dl,
 				 struct vsp1_dl_body *dlb)
 {
 	struct vsp1_clu *clu = to_clu(&entity->subdev);
diff --git a/drivers/media/platform/vsp1/vsp1_dl.c b/drivers/media/platform/vsp1/vsp1_dl.c
index 26289adaf658..104b6f514536 100644
--- a/drivers/media/platform/vsp1/vsp1_dl.c
+++ b/drivers/media/platform/vsp1/vsp1_dl.c
@@ -178,7 +178,7 @@ struct vsp1_dl_cmd_pool {
  * @post_cmd: post command to be issued through extended dl header
  * @has_chain: if true, indicates that there's a partition chain
  * @chain: entry in the display list partition chain
- * @internal: whether the display list is used for internal purpose
+ * @flags: display list flags, a combination of VSP1_DL_FRAME_END_*
  */
 struct vsp1_dl_list {
 	struct list_head list;
@@ -197,7 +197,7 @@ struct vsp1_dl_list {
 	bool has_chain;
 	struct list_head chain;
 
-	bool internal;
+	unsigned int flags;
 };
 
 /**
@@ -699,8 +699,8 @@ struct vsp1_dl_body *vsp1_dl_list_get_body0(struct vsp1_dl_list *dl)
  * which bodies are added.
  *
  * Adding a body to a display list passes ownership of the body to the list. The
- * caller retains its reference to the fragment when adding it to the display
- * list, but is not allowed to add new entries to the body.
+ * caller retains its reference to the body when adding it to the display list,
+ * but is not allowed to add new entries to the body.
  *
  * The reference must be explicitly released by a call to vsp1_dl_body_put()
  * when the body isn't needed anymore.
@@ -770,17 +770,35 @@ static void vsp1_dl_list_fill_header(struct vsp1_dl_list *dl, bool is_last)
 	}
 
 	dl->header->num_lists = num_lists;
+	dl->header->flags = 0;
 
-	if (!list_empty(&dl->chain) && !is_last) {
+	/*
+	 * Enable the interrupt for the end of each frame. In continuous mode
+	 * chained lists are used with one list per frame, so enable the
+	 * interrupt for each list. In singleshot mode chained lists are used
+	 * to partition a single frame, so enable the interrupt for the last
+	 * list only.
+	 */
+	if (!dlm->singleshot || is_last)
+		dl->header->flags |= VSP1_DLH_INT_ENABLE;
+
+	/*
+	 * In continuous mode enable auto-start for all lists, as the VSP must
+	 * loop on the same list until a new one is queued. In singleshot mode
+	 * enable auto-start for all lists but the last to chain processing of
+	 * partitions without software intervention.
+	 */
+	if (!dlm->singleshot || !is_last)
+		dl->header->flags |= VSP1_DLH_AUTO_START;
+
+	if (!is_last) {
 		/*
-		 * If this display list's chain is not empty, we are on a list,
-		 * and the next item is the display list that we must queue for
-		 * automatic processing by the hardware.
+		 * If this is not the last display list in the chain, queue the
+		 * next item for automatic processing by the hardware.
 		 */
 		struct vsp1_dl_list *next = list_next_entry(dl, chain);
 
 		dl->header->next_header = next->dma;
-		dl->header->flags = VSP1_DLH_AUTO_START;
 	} else if (!dlm->singleshot) {
 		/*
 		 * if the display list manager works in continuous mode, the VSP
@@ -788,13 +806,6 @@ static void vsp1_dl_list_fill_header(struct vsp1_dl_list *dl, bool is_last)
 		 * instructed to do otherwise.
 		 */
 		dl->header->next_header = dl->dma;
-		dl->header->flags = VSP1_DLH_INT_ENABLE | VSP1_DLH_AUTO_START;
-	} else {
-		/*
-		 * Otherwise, in mem-to-mem mode, we work in single-shot mode
-		 * and the next display list must not be started automatically.
-		 */
-		dl->header->flags = VSP1_DLH_INT_ENABLE;
 	}
 
 	if (!dl->extension)
@@ -861,13 +872,15 @@ static void vsp1_dl_list_commit_continuous(struct vsp1_dl_list *dl)
 	 *
 	 * If a display list is already pending we simply drop it as the new
 	 * display list is assumed to contain a more recent configuration. It is
-	 * an error if the already pending list has the internal flag set, as
-	 * there is then a process waiting for that list to complete. This
-	 * shouldn't happen as the waiting process should perform proper
-	 * locking, but warn just in case.
+	 * an error if the already pending list has the
+	 * VSP1_DL_FRAME_END_INTERNAL flag set, as there is then a process
+	 * waiting for that list to complete. This shouldn't happen as the
+	 * waiting process should perform proper locking, but warn just in
+	 * case.
 	 */
 	if (vsp1_dl_list_hw_update_pending(dlm)) {
-		WARN_ON(dlm->pending && dlm->pending->internal);
+		WARN_ON(dlm->pending &&
+			(dlm->pending->flags & VSP1_DL_FRAME_END_INTERNAL));
 		__vsp1_dl_list_put(dlm->pending);
 		dlm->pending = dl;
 		return;
@@ -897,7 +910,7 @@ static void vsp1_dl_list_commit_singleshot(struct vsp1_dl_list *dl)
 	dlm->active = dl;
 }
 
-void vsp1_dl_list_commit(struct vsp1_dl_list *dl, bool internal)
+void vsp1_dl_list_commit(struct vsp1_dl_list *dl, unsigned int dl_flags)
 {
 	struct vsp1_dl_manager *dlm = dl->dlm;
 	struct vsp1_dl_list *dl_next;
@@ -912,7 +925,7 @@ void vsp1_dl_list_commit(struct vsp1_dl_list *dl, bool internal)
 		vsp1_dl_list_fill_header(dl_next, last);
 	}
 
-	dl->internal = internal;
+	dl->flags = dl_flags & ~VSP1_DL_FRAME_END_COMPLETED;
 
 	spin_lock_irqsave(&dlm->lock, flags);
 
@@ -941,9 +954,13 @@ void vsp1_dl_list_commit(struct vsp1_dl_list *dl, bool internal)
  * set in single-shot mode as display list processing is then not continuous and
  * races never occur.
  *
- * The VSP1_DL_FRAME_END_INTERNAL flag indicates that the previous display list
- * has completed and had been queued with the internal notification flag.
- * Internal notification is only supported for continuous mode.
+ * The following flags are only supported for continuous mode.
+ *
+ * The VSP1_DL_FRAME_END_INTERNAL flag indicates that the display list that just
+ * became active had been queued with the internal notification flag.
+ *
+ * The VSP1_DL_FRAME_END_WRITEBACK flag indicates that the previously active
+ * display list had been queued with the writeback flag.
  */
 unsigned int vsp1_dlm_irq_frame_end(struct vsp1_dl_manager *dlm)
 {
@@ -982,13 +999,24 @@ unsigned int vsp1_dlm_irq_frame_end(struct vsp1_dl_manager *dlm)
 		goto done;
 
 	/*
+	 * If the active display list has the writeback flag set, the frame
+	 * completion marks the end of the writeback capture. Return the
+	 * VSP1_DL_FRAME_END_WRITEBACK flag and reset the display list's
+	 * writeback flag.
+	 */
+	if (dlm->active && (dlm->active->flags & VSP1_DL_FRAME_END_WRITEBACK)) {
+		flags |= VSP1_DL_FRAME_END_WRITEBACK;
+		dlm->active->flags &= ~VSP1_DL_FRAME_END_WRITEBACK;
+	}
+
+	/*
 	 * The device starts processing the queued display list right after the
 	 * frame end interrupt. The display list thus becomes active.
 	 */
 	if (dlm->queued) {
-		if (dlm->queued->internal)
+		if (dlm->queued->flags & VSP1_DL_FRAME_END_INTERNAL)
 			flags |= VSP1_DL_FRAME_END_INTERNAL;
-		dlm->queued->internal = false;
+		dlm->queued->flags &= ~VSP1_DL_FRAME_END_INTERNAL;
 
 		__vsp1_dl_list_put(dlm->active);
 		dlm->active = dlm->queued;
diff --git a/drivers/media/platform/vsp1/vsp1_dl.h b/drivers/media/platform/vsp1/vsp1_dl.h
index 125750dc8b5c..bebe16483ca5 100644
--- a/drivers/media/platform/vsp1/vsp1_dl.h
+++ b/drivers/media/platform/vsp1/vsp1_dl.h
@@ -17,8 +17,10 @@ struct vsp1_dl_body_pool;
 struct vsp1_dl_list;
 struct vsp1_dl_manager;
 
+/* Keep these flags in sync with VSP1_DU_STATUS_* in include/media/vsp1.h. */
 #define VSP1_DL_FRAME_END_COMPLETED		BIT(0)
-#define VSP1_DL_FRAME_END_INTERNAL		BIT(1)
+#define VSP1_DL_FRAME_END_WRITEBACK		BIT(1)
+#define VSP1_DL_FRAME_END_INTERNAL		BIT(2)
 
 /**
  * struct vsp1_dl_ext_cmd - Extended Display command
@@ -61,7 +63,7 @@ struct vsp1_dl_list *vsp1_dl_list_get(struct vsp1_dl_manager *dlm);
 void vsp1_dl_list_put(struct vsp1_dl_list *dl);
 struct vsp1_dl_body *vsp1_dl_list_get_body0(struct vsp1_dl_list *dl);
 struct vsp1_dl_ext_cmd *vsp1_dl_get_pre_cmd(struct vsp1_dl_list *dl);
-void vsp1_dl_list_commit(struct vsp1_dl_list *dl, bool internal);
+void vsp1_dl_list_commit(struct vsp1_dl_list *dl, unsigned int dl_flags);
 
 struct vsp1_dl_body_pool *
 vsp1_dl_body_pool_create(struct vsp1_device *vsp1, unsigned int num_bodies,
diff --git a/drivers/media/platform/vsp1/vsp1_drm.c b/drivers/media/platform/vsp1/vsp1_drm.c
index 84895385d2e5..a4a45d68a6ef 100644
--- a/drivers/media/platform/vsp1/vsp1_drm.c
+++ b/drivers/media/platform/vsp1/vsp1_drm.c
@@ -34,14 +34,16 @@ static void vsp1_du_pipeline_frame_end(struct vsp1_pipeline *pipe,
 				       unsigned int completion)
 {
 	struct vsp1_drm_pipeline *drm_pipe = to_vsp1_drm_pipeline(pipe);
-	bool complete = completion == VSP1_DL_FRAME_END_COMPLETED;
 
 	if (drm_pipe->du_complete) {
 		struct vsp1_entity *uif = drm_pipe->uif;
+		unsigned int status = completion
+				    & (VSP1_DU_STATUS_COMPLETE |
+				       VSP1_DU_STATUS_WRITEBACK);
 		u32 crc;
 
 		crc = uif ? vsp1_uif_get_crc(to_uif(&uif->subdev)) : 0;
-		drm_pipe->du_complete(drm_pipe->du_private, complete, crc);
+		drm_pipe->du_complete(drm_pipe->du_private, status, crc);
 	}
 
 	if (completion & VSP1_DL_FRAME_END_INTERNAL) {
@@ -537,6 +539,12 @@ static void vsp1_du_pipeline_configure(struct vsp1_pipeline *pipe)
 	struct vsp1_entity *next;
 	struct vsp1_dl_list *dl;
 	struct vsp1_dl_body *dlb;
+	unsigned int dl_flags = 0;
+
+	if (drm_pipe->force_brx_release)
+		dl_flags |= VSP1_DL_FRAME_END_INTERNAL;
+	if (pipe->output->writeback)
+		dl_flags |= VSP1_DL_FRAME_END_WRITEBACK;
 
 	dl = vsp1_dl_list_get(pipe->output->dlm);
 	dlb = vsp1_dl_list_get_body0(dl);
@@ -554,12 +562,42 @@ static void vsp1_du_pipeline_configure(struct vsp1_pipeline *pipe)
 		}
 
 		vsp1_entity_route_setup(entity, pipe, dlb);
-		vsp1_entity_configure_stream(entity, pipe, dlb);
+		vsp1_entity_configure_stream(entity, pipe, dl, dlb);
 		vsp1_entity_configure_frame(entity, pipe, dl, dlb);
 		vsp1_entity_configure_partition(entity, pipe, dl, dlb);
 	}
 
-	vsp1_dl_list_commit(dl, drm_pipe->force_brx_release);
+	vsp1_dl_list_commit(dl, dl_flags);
+}
+
+static int vsp1_du_pipeline_set_rwpf_format(struct vsp1_device *vsp1,
+					    struct vsp1_rwpf *rwpf,
+					    u32 pixelformat, unsigned int pitch)
+{
+	const struct vsp1_format_info *fmtinfo;
+	unsigned int chroma_hsub;
+
+	fmtinfo = vsp1_get_format_info(vsp1, pixelformat);
+	if (!fmtinfo) {
+		dev_dbg(vsp1->dev, "Unsupported pixel format %08x\n",
+			pixelformat);
+		return -EINVAL;
+	}
+
+	/*
+	 * Only formats with three planes can affect the chroma planes pitch.
+	 * All formats with two planes have a horizontal subsampling value of 2,
+	 * but combine U and V in a single chroma plane, which thus results in
+	 * the luma plane and chroma plane having the same pitch.
+	 */
+	chroma_hsub = (fmtinfo->planes == 3) ? fmtinfo->hsub : 1;
+
+	rwpf->fmtinfo = fmtinfo;
+	rwpf->format.num_planes = fmtinfo->planes;
+	rwpf->format.plane_fmt[0].bytesperline = pitch;
+	rwpf->format.plane_fmt[1].bytesperline = pitch / chroma_hsub;
+
+	return 0;
 }
 
 /* -----------------------------------------------------------------------------
@@ -700,8 +738,8 @@ int vsp1_du_setup_lif(struct device *dev, unsigned int pipe_index,
 	drm_pipe->du_private = cfg->callback_data;
 
 	/* Disable the display interrupts. */
-	vsp1_write(vsp1, VI6_DISP_IRQ_STA, 0);
-	vsp1_write(vsp1, VI6_DISP_IRQ_ENB, 0);
+	vsp1_write(vsp1, VI6_DISP_IRQ_STA(pipe_index), 0);
+	vsp1_write(vsp1, VI6_DISP_IRQ_ENB(pipe_index), 0);
 
 	/* Configure all entities in the pipeline. */
 	vsp1_du_pipeline_configure(pipe);
@@ -769,9 +807,8 @@ int vsp1_du_atomic_update(struct device *dev, unsigned int pipe_index,
 {
 	struct vsp1_device *vsp1 = dev_get_drvdata(dev);
 	struct vsp1_drm_pipeline *drm_pipe = &vsp1->drm->pipe[pipe_index];
-	const struct vsp1_format_info *fmtinfo;
-	unsigned int chroma_hsub;
 	struct vsp1_rwpf *rpf;
+	int ret;
 
 	if (rpf_index >= vsp1->info->rpf_count)
 		return -EINVAL;
@@ -804,25 +841,11 @@ int vsp1_du_atomic_update(struct device *dev, unsigned int pipe_index,
 	 * Store the format, stride, memory buffer address, crop and compose
 	 * rectangles and Z-order position and for the input.
 	 */
-	fmtinfo = vsp1_get_format_info(vsp1, cfg->pixelformat);
-	if (!fmtinfo) {
-		dev_dbg(vsp1->dev, "Unsupported pixel format %08x for RPF\n",
-			cfg->pixelformat);
-		return -EINVAL;
-	}
-
-	/*
-	 * Only formats with three planes can affect the chroma planes pitch.
-	 * All formats with two planes have a horizontal subsampling value of 2,
-	 * but combine U and V in a single chroma plane, which thus results in
-	 * the luma plane and chroma plane having the same pitch.
-	 */
-	chroma_hsub = (fmtinfo->planes == 3) ? fmtinfo->hsub : 1;
+	ret = vsp1_du_pipeline_set_rwpf_format(vsp1, rpf, cfg->pixelformat,
+					       cfg->pitch);
+	if (ret < 0)
+		return ret;
 
-	rpf->fmtinfo = fmtinfo;
-	rpf->format.num_planes = fmtinfo->planes;
-	rpf->format.plane_fmt[0].bytesperline = cfg->pitch;
-	rpf->format.plane_fmt[1].bytesperline = cfg->pitch / chroma_hsub;
 	rpf->alpha = cfg->alpha;
 
 	rpf->mem.addr[0] = cfg->mem[0];
@@ -851,12 +874,31 @@ void vsp1_du_atomic_flush(struct device *dev, unsigned int pipe_index,
 	struct vsp1_device *vsp1 = dev_get_drvdata(dev);
 	struct vsp1_drm_pipeline *drm_pipe = &vsp1->drm->pipe[pipe_index];
 	struct vsp1_pipeline *pipe = &drm_pipe->pipe;
+	int ret;
 
 	drm_pipe->crc = cfg->crc;
 
 	mutex_lock(&vsp1->drm->lock);
+
+	if (cfg->writeback.pixelformat) {
+		const struct vsp1_du_writeback_config *wb_cfg = &cfg->writeback;
+
+		ret = vsp1_du_pipeline_set_rwpf_format(vsp1, pipe->output,
+						       wb_cfg->pixelformat,
+						       wb_cfg->pitch);
+		if (WARN_ON(ret < 0))
+			goto done;
+
+		pipe->output->mem.addr[0] = wb_cfg->mem[0];
+		pipe->output->mem.addr[1] = wb_cfg->mem[1];
+		pipe->output->mem.addr[2] = wb_cfg->mem[2];
+		pipe->output->writeback = true;
+	}
+
 	vsp1_du_pipeline_setup_inputs(vsp1, pipe);
 	vsp1_du_pipeline_configure(pipe);
+
+done:
 	mutex_unlock(&vsp1->drm->lock);
 }
 EXPORT_SYMBOL_GPL(vsp1_du_atomic_flush);
diff --git a/drivers/media/platform/vsp1/vsp1_drm.h b/drivers/media/platform/vsp1/vsp1_drm.h
index 8dfd274a59e2..e85ad4366fbb 100644
--- a/drivers/media/platform/vsp1/vsp1_drm.h
+++ b/drivers/media/platform/vsp1/vsp1_drm.h
@@ -42,7 +42,7 @@ struct vsp1_drm_pipeline {
 	struct vsp1_du_crc_config crc;
 
 	/* Frame synchronisation */
-	void (*du_complete)(void *data, bool completed, u32 crc);
+	void (*du_complete)(void *data, unsigned int status, u32 crc);
 	void *du_private;
 };
 
diff --git a/drivers/media/platform/vsp1/vsp1_entity.c b/drivers/media/platform/vsp1/vsp1_entity.c
index a54ab528b060..aa9d2286056e 100644
--- a/drivers/media/platform/vsp1/vsp1_entity.c
+++ b/drivers/media/platform/vsp1/vsp1_entity.c
@@ -71,10 +71,11 @@ void vsp1_entity_route_setup(struct vsp1_entity *entity,
 
 void vsp1_entity_configure_stream(struct vsp1_entity *entity,
 				  struct vsp1_pipeline *pipe,
+				  struct vsp1_dl_list *dl,
 				  struct vsp1_dl_body *dlb)
 {
 	if (entity->ops->configure_stream)
-		entity->ops->configure_stream(entity, pipe, dlb);
+		entity->ops->configure_stream(entity, pipe, dl, dlb);
 }
 
 void vsp1_entity_configure_frame(struct vsp1_entity *entity,
diff --git a/drivers/media/platform/vsp1/vsp1_entity.h b/drivers/media/platform/vsp1/vsp1_entity.h
index 97acb7795cf1..a1ceb37bb837 100644
--- a/drivers/media/platform/vsp1/vsp1_entity.h
+++ b/drivers/media/platform/vsp1/vsp1_entity.h
@@ -67,7 +67,9 @@ struct vsp1_route {
  * struct vsp1_entity_operations - Entity operations
  * @destroy:	Destroy the entity.
  * @configure_stream:	Setup the hardware parameters for the stream which do
- *			not vary between frames (pipeline, formats).
+ *			not vary between frames (pipeline, formats). Note that
+ *			the vsp1_dl_list argument is only valid for display
+ *			pipeline and will be NULL for mem-to-mem pipelines.
  * @configure_frame:	Configure the runtime parameters for each frame.
  * @configure_partition: Configure partition specific parameters.
  * @max_width:	Return the max supported width of data that the entity can
@@ -78,7 +80,7 @@ struct vsp1_route {
 struct vsp1_entity_operations {
 	void (*destroy)(struct vsp1_entity *);
 	void (*configure_stream)(struct vsp1_entity *, struct vsp1_pipeline *,
-				 struct vsp1_dl_body *);
+				 struct vsp1_dl_list *, struct vsp1_dl_body *);
 	void (*configure_frame)(struct vsp1_entity *, struct vsp1_pipeline *,
 				struct vsp1_dl_list *, struct vsp1_dl_body *);
 	void (*configure_partition)(struct vsp1_entity *,
@@ -155,6 +157,7 @@ void vsp1_entity_route_setup(struct vsp1_entity *entity,
 
 void vsp1_entity_configure_stream(struct vsp1_entity *entity,
 				  struct vsp1_pipeline *pipe,
+				  struct vsp1_dl_list *dl,
 				  struct vsp1_dl_body *dlb);
 
 void vsp1_entity_configure_frame(struct vsp1_entity *entity,
diff --git a/drivers/media/platform/vsp1/vsp1_hgo.c b/drivers/media/platform/vsp1/vsp1_hgo.c
index 827373c25351..bf3f981f93a1 100644
--- a/drivers/media/platform/vsp1/vsp1_hgo.c
+++ b/drivers/media/platform/vsp1/vsp1_hgo.c
@@ -131,6 +131,7 @@ static const struct v4l2_ctrl_config hgo_num_bins_control = {
 
 static void hgo_configure_stream(struct vsp1_entity *entity,
 				 struct vsp1_pipeline *pipe,
+				 struct vsp1_dl_list *dl,
 				 struct vsp1_dl_body *dlb)
 {
 	struct vsp1_hgo *hgo = to_hgo(&entity->subdev);
diff --git a/drivers/media/platform/vsp1/vsp1_hgt.c b/drivers/media/platform/vsp1/vsp1_hgt.c
index bb6ce6fdd5f4..aa1c718e0453 100644
--- a/drivers/media/platform/vsp1/vsp1_hgt.c
+++ b/drivers/media/platform/vsp1/vsp1_hgt.c
@@ -127,6 +127,7 @@ static const struct v4l2_ctrl_config hgt_hue_areas = {
 
 static void hgt_configure_stream(struct vsp1_entity *entity,
 				 struct vsp1_pipeline *pipe,
+				 struct vsp1_dl_list *dl,
 				 struct vsp1_dl_body *dlb)
 {
 	struct vsp1_hgt *hgt = to_hgt(&entity->subdev);
diff --git a/drivers/media/platform/vsp1/vsp1_hsit.c b/drivers/media/platform/vsp1/vsp1_hsit.c
index 39ab2e0c7c18..d5ebd9d08c8a 100644
--- a/drivers/media/platform/vsp1/vsp1_hsit.c
+++ b/drivers/media/platform/vsp1/vsp1_hsit.c
@@ -129,6 +129,7 @@ static const struct v4l2_subdev_ops hsit_ops = {
 
 static void hsit_configure_stream(struct vsp1_entity *entity,
 				  struct vsp1_pipeline *pipe,
+				  struct vsp1_dl_list *dl,
 				  struct vsp1_dl_body *dlb)
 {
 	struct vsp1_hsit *hsit = to_hsit(&entity->subdev);
diff --git a/drivers/media/platform/vsp1/vsp1_lif.c b/drivers/media/platform/vsp1/vsp1_lif.c
index 8b0a26335d70..14ed5d7bd061 100644
--- a/drivers/media/platform/vsp1/vsp1_lif.c
+++ b/drivers/media/platform/vsp1/vsp1_lif.c
@@ -84,6 +84,7 @@ static const struct v4l2_subdev_ops lif_ops = {
 
 static void lif_configure_stream(struct vsp1_entity *entity,
 				 struct vsp1_pipeline *pipe,
+				 struct vsp1_dl_list *dl,
 				 struct vsp1_dl_body *dlb)
 {
 	const struct v4l2_mbus_framefmt *format;
diff --git a/drivers/media/platform/vsp1/vsp1_lut.c b/drivers/media/platform/vsp1/vsp1_lut.c
index 64c48d9459b0..9f88842d7048 100644
--- a/drivers/media/platform/vsp1/vsp1_lut.c
+++ b/drivers/media/platform/vsp1/vsp1_lut.c
@@ -147,6 +147,7 @@ static const struct v4l2_subdev_ops lut_ops = {
 
 static void lut_configure_stream(struct vsp1_entity *entity,
 				 struct vsp1_pipeline *pipe,
+				 struct vsp1_dl_list *dl,
 				 struct vsp1_dl_body *dlb)
 {
 	struct vsp1_lut *lut = to_lut(&entity->subdev);
diff --git a/drivers/media/platform/vsp1/vsp1_regs.h b/drivers/media/platform/vsp1/vsp1_regs.h
index f6e4157095cc..1bb1d39c60d9 100644
--- a/drivers/media/platform/vsp1/vsp1_regs.h
+++ b/drivers/media/platform/vsp1/vsp1_regs.h
@@ -39,12 +39,12 @@
 #define VI6_WFP_IRQ_STA_DFE		(1 << 1)
 #define VI6_WFP_IRQ_STA_FRE		(1 << 0)
 
-#define VI6_DISP_IRQ_ENB		0x0078
+#define VI6_DISP_IRQ_ENB(n)		(0x0078 + (n) * 60)
 #define VI6_DISP_IRQ_ENB_DSTE		(1 << 8)
 #define VI6_DISP_IRQ_ENB_MAEE		(1 << 5)
 #define VI6_DISP_IRQ_ENB_LNEE(n)	(1 << (n))
 
-#define VI6_DISP_IRQ_STA		0x007c
+#define VI6_DISP_IRQ_STA(n)		(0x007c + (n) * 60)
 #define VI6_DISP_IRQ_STA_DST		(1 << 8)
 #define VI6_DISP_IRQ_STA_MAE		(1 << 5)
 #define VI6_DISP_IRQ_STA_LNE(n)		(1 << (n))
@@ -307,7 +307,7 @@
 #define VI6_WPF_DSTM_ADDR_C0		0x1028
 #define VI6_WPF_DSTM_ADDR_C1		0x102c
 
-#define VI6_WPF_WRBCK_CTRL		0x1034
+#define VI6_WPF_WRBCK_CTRL(n)		(0x1034 + (n) * 0x100)
 #define VI6_WPF_WRBCK_CTRL_WBMD		(1 << 0)
 
 /* -----------------------------------------------------------------------------
diff --git a/drivers/media/platform/vsp1/vsp1_rpf.c b/drivers/media/platform/vsp1/vsp1_rpf.c
index 616afa7e165f..85587c1b6a37 100644
--- a/drivers/media/platform/vsp1/vsp1_rpf.c
+++ b/drivers/media/platform/vsp1/vsp1_rpf.c
@@ -57,6 +57,7 @@ static const struct v4l2_subdev_ops rpf_ops = {
 
 static void rpf_configure_stream(struct vsp1_entity *entity,
 				 struct vsp1_pipeline *pipe,
+				 struct vsp1_dl_list *dl,
 				 struct vsp1_dl_body *dlb)
 {
 	struct vsp1_rwpf *rpf = to_rwpf(&entity->subdev);
diff --git a/drivers/media/platform/vsp1/vsp1_rwpf.h b/drivers/media/platform/vsp1/vsp1_rwpf.h
index 70742ecf766f..2f3582590618 100644
--- a/drivers/media/platform/vsp1/vsp1_rwpf.h
+++ b/drivers/media/platform/vsp1/vsp1_rwpf.h
@@ -61,6 +61,7 @@ struct vsp1_rwpf {
 	} flip;
 
 	struct vsp1_rwpf_memory mem;
+	bool writeback;
 
 	struct vsp1_dl_manager *dlm;
 };
diff --git a/drivers/media/platform/vsp1/vsp1_sru.c b/drivers/media/platform/vsp1/vsp1_sru.c
index b1617cb1f2b9..2b65457ee12f 100644
--- a/drivers/media/platform/vsp1/vsp1_sru.c
+++ b/drivers/media/platform/vsp1/vsp1_sru.c
@@ -269,6 +269,7 @@ static const struct v4l2_subdev_ops sru_ops = {
 
 static void sru_configure_stream(struct vsp1_entity *entity,
 				 struct vsp1_pipeline *pipe,
+				 struct vsp1_dl_list *dl,
 				 struct vsp1_dl_body *dlb)
 {
 	const struct vsp1_sru_param *param;
diff --git a/drivers/media/platform/vsp1/vsp1_uds.c b/drivers/media/platform/vsp1/vsp1_uds.c
index 27012af973b2..5fc04c082d1a 100644
--- a/drivers/media/platform/vsp1/vsp1_uds.c
+++ b/drivers/media/platform/vsp1/vsp1_uds.c
@@ -257,6 +257,7 @@ static const struct v4l2_subdev_ops uds_ops = {
 
 static void uds_configure_stream(struct vsp1_entity *entity,
 				 struct vsp1_pipeline *pipe,
+				 struct vsp1_dl_list *dl,
 				 struct vsp1_dl_body *dlb)
 {
 	struct vsp1_uds *uds = to_uds(&entity->subdev);
diff --git a/drivers/media/platform/vsp1/vsp1_uif.c b/drivers/media/platform/vsp1/vsp1_uif.c
index 4b58d51df231..467d1072577b 100644
--- a/drivers/media/platform/vsp1/vsp1_uif.c
+++ b/drivers/media/platform/vsp1/vsp1_uif.c
@@ -192,6 +192,7 @@ static const struct v4l2_subdev_ops uif_ops = {
 
 static void uif_configure_stream(struct vsp1_entity *entity,
 				 struct vsp1_pipeline *pipe,
+				 struct vsp1_dl_list *dl,
 				 struct vsp1_dl_body *dlb)
 {
 	struct vsp1_uif *uif = to_uif(&entity->subdev);
diff --git a/drivers/media/platform/vsp1/vsp1_video.c b/drivers/media/platform/vsp1/vsp1_video.c
index 7ceaf3222145..fd98e483b2f4 100644
--- a/drivers/media/platform/vsp1/vsp1_video.c
+++ b/drivers/media/platform/vsp1/vsp1_video.c
@@ -307,11 +307,6 @@ static int vsp1_video_pipeline_setup_partitions(struct vsp1_pipeline *pipe)
  * This function completes the current buffer by filling its sequence number,
  * time stamp and payload size, and hands it back to the videobuf core.
  *
- * When operating in DU output mode (deep pipeline to the DU through the LIF),
- * the VSP1 needs to constantly supply frames to the display. In that case, if
- * no other buffer is queued, reuse the one that has just been processed instead
- * of handing it back to the videobuf core.
- *
  * Return the next queued buffer or NULL if the queue is empty.
  */
 static struct vsp1_vb2_buffer *
@@ -333,12 +328,6 @@ vsp1_video_complete_buffer(struct vsp1_video *video)
 	done = list_first_entry(&video->irqqueue,
 				struct vsp1_vb2_buffer, queue);
 
-	/* In DU output mode reuse the buffer if the list is singular. */
-	if (pipe->lif && list_is_singular(&video->irqqueue)) {
-		spin_unlock_irqrestore(&video->irqlock, flags);
-		return done;
-	}
-
 	list_del(&done->queue);
 
 	if (!list_empty(&video->irqqueue))
@@ -432,7 +421,7 @@ static void vsp1_video_pipeline_run(struct vsp1_pipeline *pipe)
 	}
 
 	/* Complete, and commit the head display list. */
-	vsp1_dl_list_commit(dl, false);
+	vsp1_dl_list_commit(dl, 0);
 	pipe->configured = true;
 
 	vsp1_pipeline_run(pipe);
@@ -836,7 +825,8 @@ static int vsp1_video_setup_pipeline(struct vsp1_pipeline *pipe)
 
 	list_for_each_entry(entity, &pipe->entities, list_pipe) {
 		vsp1_entity_route_setup(entity, pipe, pipe->stream_config);
-		vsp1_entity_configure_stream(entity, pipe, pipe->stream_config);
+		vsp1_entity_configure_stream(entity, pipe, NULL,
+					     pipe->stream_config);
 	}
 
 	return 0;
diff --git a/drivers/media/platform/vsp1/vsp1_wpf.c b/drivers/media/platform/vsp1/vsp1_wpf.c
index 32bb207b2007..208498fa6ed7 100644
--- a/drivers/media/platform/vsp1/vsp1_wpf.c
+++ b/drivers/media/platform/vsp1/vsp1_wpf.c
@@ -232,17 +232,41 @@ static void vsp1_wpf_destroy(struct vsp1_entity *entity)
 	vsp1_dlm_destroy(wpf->dlm);
 }
 
+static int wpf_configure_writeback_chain(struct vsp1_rwpf *wpf,
+					 struct vsp1_dl_list *dl)
+{
+	unsigned int index = wpf->entity.index;
+	struct vsp1_dl_list *dl_next;
+	struct vsp1_dl_body *dlb;
+
+	dl_next = vsp1_dl_list_get(wpf->dlm);
+	if (!dl_next) {
+		dev_err(wpf->entity.vsp1->dev,
+			"Failed to obtain a dl list, disabling writeback\n");
+		return -ENOMEM;
+	}
+
+	dlb = vsp1_dl_list_get_body0(dl_next);
+	vsp1_dl_body_write(dlb, VI6_WPF_WRBCK_CTRL(index), 0);
+	vsp1_dl_list_add_chain(dl, dl_next);
+
+	return 0;
+}
+
 static void wpf_configure_stream(struct vsp1_entity *entity,
 				 struct vsp1_pipeline *pipe,
+				 struct vsp1_dl_list *dl,
 				 struct vsp1_dl_body *dlb)
 {
 	struct vsp1_rwpf *wpf = to_rwpf(&entity->subdev);
 	struct vsp1_device *vsp1 = wpf->entity.vsp1;
 	const struct v4l2_mbus_framefmt *source_format;
 	const struct v4l2_mbus_framefmt *sink_format;
+	unsigned int index = wpf->entity.index;
 	unsigned int i;
 	u32 outfmt = 0;
 	u32 srcrpf = 0;
+	int ret;
 
 	sink_format = vsp1_entity_get_pad_format(&wpf->entity,
 						 wpf->entity.config,
@@ -250,8 +274,9 @@ static void wpf_configure_stream(struct vsp1_entity *entity,
 	source_format = vsp1_entity_get_pad_format(&wpf->entity,
 						   wpf->entity.config,
 						   RWPF_PAD_SOURCE);
+
 	/* Format */
-	if (!pipe->lif) {
+	if (!pipe->lif || wpf->writeback) {
 		const struct v4l2_pix_format_mplane *format = &wpf->format;
 		const struct vsp1_format_info *fmtinfo = wpf->fmtinfo;
 
@@ -276,8 +301,7 @@ static void wpf_configure_stream(struct vsp1_entity *entity,
 
 		vsp1_wpf_write(wpf, dlb, VI6_WPF_DSWAP, fmtinfo->swap);
 
-		if (vsp1_feature(vsp1, VSP1_HAS_WPF_HFLIP) &&
-		    wpf->entity.index == 0)
+		if (vsp1_feature(vsp1, VSP1_HAS_WPF_HFLIP) && index == 0)
 			vsp1_wpf_write(wpf, dlb, VI6_WPF_ROT_CTRL,
 				       VI6_WPF_ROT_CTRL_LN16 |
 				       (256 << VI6_WPF_ROT_CTRL_LMEM_WD_SHIFT));
@@ -288,11 +312,9 @@ static void wpf_configure_stream(struct vsp1_entity *entity,
 
 	wpf->outfmt = outfmt;
 
-	vsp1_dl_body_write(dlb, VI6_DPR_WPF_FPORCH(wpf->entity.index),
+	vsp1_dl_body_write(dlb, VI6_DPR_WPF_FPORCH(index),
 			   VI6_DPR_WPF_FPORCH_FP_WPFN);
 
-	vsp1_dl_body_write(dlb, VI6_WPF_WRBCK_CTRL, 0);
-
 	/*
 	 * Sources. If the pipeline has a single input and BRx is not used,
 	 * configure it as the master layer. Otherwise configure all
@@ -318,9 +340,26 @@ static void wpf_configure_stream(struct vsp1_entity *entity,
 	vsp1_wpf_write(wpf, dlb, VI6_WPF_SRCRPF, srcrpf);
 
 	/* Enable interrupts. */
-	vsp1_dl_body_write(dlb, VI6_WPF_IRQ_STA(wpf->entity.index), 0);
-	vsp1_dl_body_write(dlb, VI6_WPF_IRQ_ENB(wpf->entity.index),
+	vsp1_dl_body_write(dlb, VI6_WPF_IRQ_STA(index), 0);
+	vsp1_dl_body_write(dlb, VI6_WPF_IRQ_ENB(index),
 			   VI6_WFP_IRQ_ENB_DFEE);
+
+	/*
+	 * Configure writeback for display pipelines (the wpf writeback flag is
+	 * never set for memory-to-memory pipelines). Start by adding a chained
+	 * display list to disable writeback after a single frame, and process
+	 * to enable writeback. If the display list allocation fails don't
+	 * enable writeback as we wouldn't be able to safely disable it,
+	 * resulting in possible memory corruption.
+	 */
+	if (wpf->writeback) {
+		ret = wpf_configure_writeback_chain(wpf, dl);
+		if (ret < 0)
+			wpf->writeback = false;
+	}
+
+	vsp1_dl_body_write(dlb, VI6_WPF_WRBCK_CTRL(index),
+			   wpf->writeback ? VI6_WPF_WRBCK_CTRL_WBMD : 0);
 }
 
 static void wpf_configure_frame(struct vsp1_entity *entity,
@@ -362,6 +401,7 @@ static void wpf_configure_partition(struct vsp1_entity *entity,
 	const struct vsp1_format_info *fmtinfo = wpf->fmtinfo;
 	unsigned int width;
 	unsigned int height;
+	unsigned int left;
 	unsigned int offset;
 	unsigned int flip;
 	unsigned int i;
@@ -371,13 +411,16 @@ static void wpf_configure_partition(struct vsp1_entity *entity,
 						 RWPF_PAD_SINK);
 	width = sink_format->width;
 	height = sink_format->height;
+	left = 0;
 
 	/*
 	 * Cropping. The partition algorithm can split the image into
 	 * multiple slices.
 	 */
-	if (pipe->partitions > 1)
+	if (pipe->partitions > 1) {
 		width = pipe->partition->wpf.width;
+		left = pipe->partition->wpf.left;
+	}
 
 	vsp1_wpf_write(wpf, dlb, VI6_WPF_HSZCLIP, VI6_WPF_SZCLIP_EN |
 		       (0 << VI6_WPF_SZCLIP_OFST_SHIFT) |
@@ -386,7 +429,11 @@ static void wpf_configure_partition(struct vsp1_entity *entity,
 		       (0 << VI6_WPF_SZCLIP_OFST_SHIFT) |
 		       (height << VI6_WPF_SZCLIP_SIZE_SHIFT));
 
-	if (pipe->lif)
+	/*
+	 * For display pipelines without writeback enabled there's no memory
+	 * address to configure, return now.
+	 */
+	if (pipe->lif && !wpf->writeback)
 		return;
 
 	/*
@@ -408,13 +455,11 @@ static void wpf_configure_partition(struct vsp1_entity *entity,
 	flip = wpf->flip.active;
 
 	if (flip & BIT(WPF_CTRL_HFLIP) && !wpf->flip.rotate)
-		offset = format->width - pipe->partition->wpf.left
-			- pipe->partition->wpf.width;
+		offset = format->width - left - width;
 	else if (flip & BIT(WPF_CTRL_VFLIP) && wpf->flip.rotate)
-		offset = format->height - pipe->partition->wpf.left
-			- pipe->partition->wpf.width;
+		offset = format->height - left - width;
 	else
-		offset = pipe->partition->wpf.left;
+		offset = left;
 
 	for (i = 0; i < format->num_planes; ++i) {
 		unsigned int hsub = i > 0 ? fmtinfo->hsub : 1;
@@ -436,7 +481,7 @@ static void wpf_configure_partition(struct vsp1_entity *entity,
 		 * image height.
 		 */
 		if (wpf->flip.rotate)
-			height = pipe->partition->wpf.width;
+			height = width;
 		else
 			height = format->height;
 
@@ -477,6 +522,12 @@ static void wpf_configure_partition(struct vsp1_entity *entity,
 	vsp1_wpf_write(wpf, dlb, VI6_WPF_DSTM_ADDR_Y, mem.addr[0]);
 	vsp1_wpf_write(wpf, dlb, VI6_WPF_DSTM_ADDR_C0, mem.addr[1]);
 	vsp1_wpf_write(wpf, dlb, VI6_WPF_DSTM_ADDR_C1, mem.addr[2]);
+
+	/*
+	 * Writeback operates in single-shot mode and lasts for a single frame,
+	 * reset the writeback flag to false for the next frame.
+	 */
+	wpf->writeback = false;
 }
 
 static unsigned int wpf_max_width(struct vsp1_entity *entity,
diff --git a/drivers/reset/core.c b/drivers/reset/core.c
index 9582efb70025..81ea77cba123 100644
--- a/drivers/reset/core.c
+++ b/drivers/reset/core.c
@@ -34,6 +34,7 @@ static LIST_HEAD(reset_lookup_list);
  * @id: ID of the reset controller in the reset
  *      controller device
  * @refcnt: Number of gets of this reset_control
+ * @acquired: Only one reset_control may be acquired for a given rcdev and id.
  * @shared: Is this a shared (1), or an exclusive (0) reset_control?
  * @deassert_cnt: Number of times this reset line has been deasserted
  * @triggered_count: Number of times this reset line has been reset. Currently
@@ -45,6 +46,7 @@ struct reset_control {
 	struct list_head list;
 	unsigned int id;
 	struct kref refcnt;
+	bool acquired;
 	bool shared;
 	bool array;
 	atomic_t deassert_count;
@@ -63,6 +65,17 @@ struct reset_control_array {
 	struct reset_control *rstc[];
 };
 
+static const char *rcdev_name(struct reset_controller_dev *rcdev)
+{
+	if (rcdev->dev)
+		return dev_name(rcdev->dev);
+
+	if (rcdev->of_node)
+		return rcdev->of_node->full_name;
+
+	return NULL;
+}
+
 /**
  * of_reset_simple_xlate - translate reset_spec to the reset line number
  * @rcdev: a pointer to the reset controller device
@@ -232,6 +245,34 @@ err:
 	return ret;
 }
 
+static int reset_control_array_acquire(struct reset_control_array *resets)
+{
+	unsigned int i;
+	int err;
+
+	for (i = 0; i < resets->num_rstcs; i++) {
+		err = reset_control_acquire(resets->rstc[i]);
+		if (err < 0)
+			goto release;
+	}
+
+	return 0;
+
+release:
+	while (i--)
+		reset_control_release(resets->rstc[i]);
+
+	return err;
+}
+
+static void reset_control_array_release(struct reset_control_array *resets)
+{
+	unsigned int i;
+
+	for (i = 0; i < resets->num_rstcs; i++)
+		reset_control_release(resets->rstc[i]);
+}
+
 static inline bool reset_control_is_array(struct reset_control *rstc)
 {
 	return rstc->array;
@@ -272,6 +313,9 @@ int reset_control_reset(struct reset_control *rstc)
 
 		if (atomic_inc_return(&rstc->triggered_count) != 1)
 			return 0;
+	} else {
+		if (!rstc->acquired)
+			return -EPERM;
 	}
 
 	ret = rstc->rcdev->ops->reset(rstc->rcdev, rstc->id);
@@ -334,6 +378,12 @@ int reset_control_assert(struct reset_control *rstc)
 		 */
 		if (!rstc->rcdev->ops->assert)
 			return -ENOTSUPP;
+
+		if (!rstc->acquired) {
+			WARN(1, "reset %s (ID: %u) is not acquired\n",
+			     rcdev_name(rstc->rcdev), rstc->id);
+			return -EPERM;
+		}
 	}
 
 	return rstc->rcdev->ops->assert(rstc->rcdev, rstc->id);
@@ -369,6 +419,12 @@ int reset_control_deassert(struct reset_control *rstc)
 
 		if (atomic_inc_return(&rstc->deassert_count) != 1)
 			return 0;
+	} else {
+		if (!rstc->acquired) {
+			WARN(1, "reset %s (ID: %u) is not acquired\n",
+			     rcdev_name(rstc->rcdev), rstc->id);
+			return -EPERM;
+		}
 	}
 
 	/*
@@ -406,9 +462,87 @@ int reset_control_status(struct reset_control *rstc)
 }
 EXPORT_SYMBOL_GPL(reset_control_status);
 
+/**
+ * reset_control_acquire() - acquires a reset control for exclusive use
+ * @rstc: reset control
+ *
+ * This is used to explicitly acquire a reset control for exclusive use. Note
+ * that exclusive resets are requested as acquired by default. In order for a
+ * second consumer to be able to control the reset, the first consumer has to
+ * release it first. Typically the easiest way to achieve this is to call the
+ * reset_control_get_exclusive_released() to obtain an instance of the reset
+ * control. Such reset controls are not acquired by default.
+ *
+ * Consumers implementing shared access to an exclusive reset need to follow
+ * a specific protocol in order to work together. Before consumers can change
+ * a reset they must acquire exclusive access using reset_control_acquire().
+ * After they are done operating the reset, they must release exclusive access
+ * with a call to reset_control_release(). Consumers are not granted exclusive
+ * access to the reset as long as another consumer hasn't released a reset.
+ *
+ * See also: reset_control_release()
+ */
+int reset_control_acquire(struct reset_control *rstc)
+{
+	struct reset_control *rc;
+
+	if (!rstc)
+		return 0;
+
+	if (WARN_ON(IS_ERR(rstc)))
+		return -EINVAL;
+
+	if (reset_control_is_array(rstc))
+		return reset_control_array_acquire(rstc_to_array(rstc));
+
+	mutex_lock(&reset_list_mutex);
+
+	if (rstc->acquired) {
+		mutex_unlock(&reset_list_mutex);
+		return 0;
+	}
+
+	list_for_each_entry(rc, &rstc->rcdev->reset_control_head, list) {
+		if (rstc != rc && rstc->id == rc->id) {
+			if (rc->acquired) {
+				mutex_unlock(&reset_list_mutex);
+				return -EBUSY;
+			}
+		}
+	}
+
+	rstc->acquired = true;
+
+	mutex_unlock(&reset_list_mutex);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(reset_control_acquire);
+
+/**
+ * reset_control_release() - releases exclusive access to a reset control
+ * @rstc: reset control
+ *
+ * Releases exclusive access right to a reset control previously obtained by a
+ * call to reset_control_acquire(). Until a consumer calls this function, no
+ * other consumers will be granted exclusive access.
+ *
+ * See also: reset_control_acquire()
+ */
+void reset_control_release(struct reset_control *rstc)
+{
+	if (!rstc || WARN_ON(IS_ERR(rstc)))
+		return;
+
+	if (reset_control_is_array(rstc))
+		reset_control_array_release(rstc_to_array(rstc));
+	else
+		rstc->acquired = false;
+}
+EXPORT_SYMBOL_GPL(reset_control_release);
+
 static struct reset_control *__reset_control_get_internal(
 				struct reset_controller_dev *rcdev,
-				unsigned int index, bool shared)
+				unsigned int index, bool shared, bool acquired)
 {
 	struct reset_control *rstc;
 
@@ -416,6 +550,14 @@ static struct reset_control *__reset_control_get_internal(
 
 	list_for_each_entry(rstc, &rcdev->reset_control_head, list) {
 		if (rstc->id == index) {
+			/*
+			 * Allow creating a secondary exclusive reset_control
+			 * that is initially not acquired for an already
+			 * controlled reset line.
+			 */
+			if (!rstc->shared && !shared && !acquired)
+				break;
+
 			if (WARN_ON(!rstc->shared || !shared))
 				return ERR_PTR(-EBUSY);
 
@@ -434,6 +576,7 @@ static struct reset_control *__reset_control_get_internal(
 	list_add(&rstc->list, &rcdev->reset_control_head);
 	rstc->id = index;
 	kref_init(&rstc->refcnt);
+	rstc->acquired = acquired;
 	rstc->shared = shared;
 
 	return rstc;
@@ -461,7 +604,7 @@ static void __reset_control_put_internal(struct reset_control *rstc)
 
 struct reset_control *__of_reset_control_get(struct device_node *node,
 				     const char *id, int index, bool shared,
-				     bool optional)
+				     bool optional, bool acquired)
 {
 	struct reset_control *rstc;
 	struct reset_controller_dev *r, *rcdev;
@@ -514,7 +657,7 @@ struct reset_control *__of_reset_control_get(struct device_node *node,
 	}
 
 	/* reset_list_mutex also protects the rcdev's reset_control list */
-	rstc = __reset_control_get_internal(rcdev, rstc_id, shared);
+	rstc = __reset_control_get_internal(rcdev, rstc_id, shared, acquired);
 
 out:
 	mutex_unlock(&reset_list_mutex);
@@ -544,7 +687,7 @@ __reset_controller_by_name(const char *name)
 
 static struct reset_control *
 __reset_control_get_from_lookup(struct device *dev, const char *con_id,
-				bool shared, bool optional)
+				bool shared, bool optional, bool acquired)
 {
 	const struct reset_control_lookup *lookup;
 	struct reset_controller_dev *rcdev;
@@ -574,7 +717,7 @@ __reset_control_get_from_lookup(struct device *dev, const char *con_id,
 
 			rstc = __reset_control_get_internal(rcdev,
 							    lookup->index,
-							    shared);
+							    shared, acquired);
 			mutex_unlock(&reset_list_mutex);
 			break;
 		}
@@ -589,13 +732,18 @@ __reset_control_get_from_lookup(struct device *dev, const char *con_id,
 }
 
 struct reset_control *__reset_control_get(struct device *dev, const char *id,
-					  int index, bool shared, bool optional)
+					  int index, bool shared, bool optional,
+					  bool acquired)
 {
+	if (WARN_ON(shared && acquired))
+		return ERR_PTR(-EINVAL);
+
 	if (dev->of_node)
 		return __of_reset_control_get(dev->of_node, id, index, shared,
-					      optional);
+					      optional, acquired);
 
-	return __reset_control_get_from_lookup(dev, id, shared, optional);
+	return __reset_control_get_from_lookup(dev, id, shared, optional,
+					       acquired);
 }
 EXPORT_SYMBOL_GPL(__reset_control_get);
 
@@ -636,7 +784,7 @@ static void devm_reset_control_release(struct device *dev, void *res)
 
 struct reset_control *__devm_reset_control_get(struct device *dev,
 				     const char *id, int index, bool shared,
-				     bool optional)
+				     bool optional, bool acquired)
 {
 	struct reset_control **ptr, *rstc;
 
@@ -645,7 +793,7 @@ struct reset_control *__devm_reset_control_get(struct device *dev,
 	if (!ptr)
 		return ERR_PTR(-ENOMEM);
 
-	rstc = __reset_control_get(dev, id, index, shared, optional);
+	rstc = __reset_control_get(dev, id, index, shared, optional, acquired);
 	if (!IS_ERR(rstc)) {
 		*ptr = rstc;
 		devres_add(dev, ptr);
@@ -672,7 +820,7 @@ int __device_reset(struct device *dev, bool optional)
 	struct reset_control *rstc;
 	int ret;
 
-	rstc = __reset_control_get(dev, NULL, 0, 0, optional);
+	rstc = __reset_control_get(dev, NULL, 0, 0, optional, true);
 	if (IS_ERR(rstc))
 		return PTR_ERR(rstc);
 
@@ -716,12 +864,15 @@ static int of_reset_control_get_count(struct device_node *node)
  * @np: device node for the device that requests the reset controls array
  * @shared: whether reset controls are shared or not
  * @optional: whether it is optional to get the reset controls
+ * @acquired: only one reset control may be acquired for a given controller
+ *            and ID
  *
  * Returns pointer to allocated reset_control_array on success or
  * error on failure
  */
 struct reset_control *
-of_reset_control_array_get(struct device_node *np, bool shared, bool optional)
+of_reset_control_array_get(struct device_node *np, bool shared, bool optional,
+			   bool acquired)
 {
 	struct reset_control_array *resets;
 	struct reset_control *rstc;
@@ -736,7 +887,8 @@ of_reset_control_array_get(struct device_node *np, bool shared, bool optional)
 		return ERR_PTR(-ENOMEM);
 
 	for (i = 0; i < num; i++) {
-		rstc = __of_reset_control_get(np, NULL, i, shared, optional);
+		rstc = __of_reset_control_get(np, NULL, i, shared, optional,
+					      acquired);
 		if (IS_ERR(rstc))
 			goto err_rst;
 		resets->rstc[i] = rstc;
@@ -783,7 +935,7 @@ devm_reset_control_array_get(struct device *dev, bool shared, bool optional)
 	if (!devres)
 		return ERR_PTR(-ENOMEM);
 
-	rstc = of_reset_control_array_get(dev->of_node, shared, optional);
+	rstc = of_reset_control_array_get(dev->of_node, shared, optional, true);
 	if (IS_ERR(rstc)) {
 		devres_free(devres);
 		return rstc;
diff --git a/drivers/staging/Kconfig b/drivers/staging/Kconfig
index 86001a9f67e0..d5f771fafc21 100644
--- a/drivers/staging/Kconfig
+++ b/drivers/staging/Kconfig
@@ -94,8 +94,6 @@ source "drivers/staging/greybus/Kconfig"
 
 source "drivers/staging/vc04_services/Kconfig"
 
-source "drivers/staging/vboxvideo/Kconfig"
-
 source "drivers/staging/pi433/Kconfig"
 
 source "drivers/staging/mt7621-pci/Kconfig"
diff --git a/drivers/staging/Makefile b/drivers/staging/Makefile
index dc3da72b3ff9..0da0d3f0b5e4 100644
--- a/drivers/staging/Makefile
+++ b/drivers/staging/Makefile
@@ -37,7 +37,6 @@ obj-$(CONFIG_MOST)		+= most/
 obj-$(CONFIG_KS7010)		+= ks7010/
 obj-$(CONFIG_GREYBUS)		+= greybus/
 obj-$(CONFIG_BCM2835_VCHIQ)	+= vc04_services/
-obj-$(CONFIG_DRM_VBOXVIDEO)	+= vboxvideo/
 obj-$(CONFIG_PI433)		+= pi433/
 obj-$(CONFIG_PCI_MT7621)	+= mt7621-pci/
 obj-$(CONFIG_PCI_MT7621_PHY)	+= mt7621-pci-phy/
diff --git a/drivers/staging/vboxvideo/TODO b/drivers/staging/vboxvideo/TODO
deleted file mode 100644
index 7f97c47a4042..000000000000
--- a/drivers/staging/vboxvideo/TODO
+++ /dev/null
@@ -1,10 +0,0 @@
-TODO:
--Get a full review from the drm-maintainers on dri-devel done on this driver
--Drop all the logic around initial_mode_queried, the master_set and
- master_drop callbacks and everything related to this. kms clients can handle
- hotplugs.
--Extend this TODO with the results of that review
-
-Please send any patches to Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
-Hans de Goede <hdegoede@redhat.com>, Michael Thayer <michael.thayer@oracle.com>
-and dri-devel@lists.freedesktop.org .
diff --git a/drivers/usb/dwc3/dwc3-of-simple.c b/drivers/usb/dwc3/dwc3-of-simple.c
index c4da82dd15c7..bdac3e7d7b18 100644
--- a/drivers/usb/dwc3/dwc3-of-simple.c
+++ b/drivers/usb/dwc3/dwc3-of-simple.c
@@ -60,7 +60,8 @@ static int dwc3_of_simple_probe(struct platform_device *pdev)
 		simple->pulse_resets = true;
 	}
 
-	simple->resets = of_reset_control_array_get(np, shared_resets, true);
+	simple->resets = of_reset_control_array_get(np, shared_resets, true,
+						    true);
 	if (IS_ERR(simple->resets)) {
 		ret = PTR_ERR(simple->resets);
 		dev_err(dev, "failed to get device resets, err=%d\n", ret);