Merge tag 'drm-next-2018-12-14' of git://anongit.freedesktop.org/drm/drm

Pull drm updates from Dave Airlie:
 "Core:
   - shared fencing staging removal
   - drop transactional atomic helpers and move helpers to new location
   - DP/MST atomic cleanup
   - Leasing cleanups and drop EXPORT_SYMBOL
   - Convert drivers to atomic helpers and generic fbdev.
   - removed deprecated obj_ref/unref in favour of get/put
   - Improve dumb callback documentation
   - MODESET_LOCK_BEGIN/END helpers

  panels:
   - CDTech panels, Banana Pi Panel, DLC1010GIG,
   - Olimex LCD-O-LinuXino, Samsung S6D16D0, Truly NT35597 WQXGA,
   - Himax HX8357D, simulated RTSM AEMv8.
   - GPD Win2 panel
   - AUO G101EVN010

  vgem:
   - render node support

  ttm:
   - move global init out of drivers
   - fix LRU handling for ghost objects
   - Support for simultaneous submissions to multiple engines

  scheduler:
   - timeout/fault handling changes to help GPU recovery
   - helpers for hw with preemption support

  i915:
   - Scaler/Watermark fixes
   - DP MST + powerwell fixes
   - PSR fixes
   - Break long get/put shmemfs pages
   - Icelake fixes
   - Icelake DSI video mode enablement
   - Engine workaround improvements

  amdgpu:
   - freesync support
   - GPU reset enabled on CI, VI, SOC15 dGPUs
   - ABM support in DC
   - KFD support for vega12/polaris12
   - SDMA paging queue on vega
   - More amdkfd code sharing
   - DCC scanout on GFX9
   - DC kerneldoc
   - Updated SMU firmware for GFX8 chips
   - XGMI PSP + hive reset support
   - GPU reset
   - DC trace support
   - Powerplay updates for newer Polaris
   - Cursor plane update fast path
   - kfd dma-buf support

  virtio-gpu:
   - add EDID support

  vmwgfx:
   - pageflip with damage support

  nouveau:
   - Initial Turing TU104/TU106 modesetting support

  msm:
   - a2xx gpu support for apq8060 and imx5
   - a2xx gpummu support
   - mdp4 display support for apq8060
   - DPU fixes and cleanups
   - enhanced profiling support
   - debug object naming interface
   - get_iova/page pinning decoupling

  tegra:
   - Tegra194 host1x, VIC and display support enabled
   - Audio over HDMI for Tegra186 and Tegra194

  exynos:
   - DMA/IOMMU refactoring
   - plane alpha + blend mode support
   - Color format fixes for mixer driver

  rcar-du:
   - R8A7744 and R8A77470 support
   - R8A77965 LVDS support

  imx:
   - fbdev emulation fix
   - multi-tiled scalling fixes
   - SPDX identifiers

  rockchip
   - dw_hdmi support
   - dw-mipi-dsi + dual dsi support
   - mailbox read size fix

  qxl:
   - fix cursor pinning

  vc4:
   - YUV support (scaling + cursor)

  v3d:
   - enable TFU (Texture Formatting Unit)

  mali-dp:
   - add support for linear tiled formats

  sun4i:
   - Display Engine 3 support
   - H6 DE3 mixer 0 support
   - H6 display engine support
   - dw-hdmi support
   - H6 HDMI phy support
   - implicit fence waiting
   - BGRX8888 support

  meson:
   - Overlay plane support
   - implicit fence waiting
   - HDMI 1.4 4k modes

  bridge:
   - i2c fixes for sii902x"

* tag 'drm-next-2018-12-14' of git://anongit.freedesktop.org/drm/drm: (1403 commits)
  drm/amd/display: Add fast path for cursor plane updates
  drm/amdgpu: Enable GPU recovery by default for CI
  drm/amd/display: Fix duplicating scaling/underscan connector state
  drm/amd/display: Fix unintialized max_bpc state values
  Revert "drm/amd/display: Set RMX_ASPECT as default"
  drm/amdgpu: Fix stub function name
  drm/msm/dpu: Fix clock issue after bind failure
  drm/msm/dpu: Clean up dpu_media_info.h static inline functions
  drm/msm/dpu: Further cleanups for static inline functions
  drm/msm/dpu: Cleanup the debugfs functions
  drm/msm/dpu: Remove dpu_irq and unused functions
  drm/msm: Make irq_postinstall optional
  drm/msm/dpu: Cleanup callers of dpu_hw_blk_init
  drm/msm/dpu: Remove unused functions
  drm/msm/dpu: Remove dpu_crtc_is_enabled()
  drm/msm/dpu: Remove dpu_crtc_get_mixer_height
  drm/msm/dpu: Remove dpu_dbg
  drm/msm: dpu: Remove crtc_lock
  drm/msm: dpu: Remove vblank_requested flag from dpu_crtc
  drm/msm: dpu: Separate crtc assignment from vblank enable
  ...

1 files changed, 111 insertions, 49 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index 1580ec60b89f..2dfaf158ef07 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -26,15 +26,26 @@
 #include "amdgpu.h"
 #include "amdgpu_gfx.h"
 #include <linux/module.h>
+#include <linux/dma-buf.h>
 
 const struct kgd2kfd_calls *kgd2kfd;
 
 static const unsigned int compute_vmid_bitmap = 0xFF00;
 
+/* Total memory size in system memory and all GPU VRAM. Used to
+ * estimate worst case amount of memory to reserve for page tables
+ */
+uint64_t amdgpu_amdkfd_total_mem_size;
+
 int amdgpu_amdkfd_init(void)
 {
+	struct sysinfo si;
 	int ret;
 
+	si_meminfo(&si);
+	amdgpu_amdkfd_total_mem_size = si.totalram - si.totalhigh;
+	amdgpu_amdkfd_total_mem_size *= si.mem_unit;
+
 #ifdef CONFIG_HSA_AMD
 	ret = kgd2kfd_init(KFD_INTERFACE_VERSION, &kgd2kfd);
 	if (ret)
@@ -73,9 +84,11 @@ void amdgpu_amdkfd_device_probe(struct amdgpu_device *adev)
 	case CHIP_FIJI:
 	case CHIP_POLARIS10:
 	case CHIP_POLARIS11:
+	case CHIP_POLARIS12:
 		kfd2kgd = amdgpu_amdkfd_gfx_8_0_get_functions();
 		break;
 	case CHIP_VEGA10:
+	case CHIP_VEGA12:
 	case CHIP_VEGA20:
 	case CHIP_RAVEN:
 		kfd2kgd = amdgpu_amdkfd_gfx_9_0_get_functions();
@@ -85,8 +98,11 @@ void amdgpu_amdkfd_device_probe(struct amdgpu_device *adev)
 		return;
 	}
 
-	adev->kfd = kgd2kfd->probe((struct kgd_dev *)adev,
-				   adev->pdev, kfd2kgd);
+	adev->kfd.dev = kgd2kfd->probe((struct kgd_dev *)adev,
+				       adev->pdev, kfd2kgd);
+
+	if (adev->kfd.dev)
+		amdgpu_amdkfd_total_mem_size += adev->gmc.real_vram_size;
 }
 
 /**
@@ -126,7 +142,8 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 {
 	int i, n;
 	int last_valid_bit;
-	if (adev->kfd) {
+
+	if (adev->kfd.dev) {
 		struct kgd2kfd_shared_resources gpu_resources = {
 			.compute_vmid_bitmap = compute_vmid_bitmap,
 			.num_pipe_per_mec = adev->gfx.mec.num_pipe_per_mec,
@@ -144,7 +161,7 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 				  KGD_MAX_QUEUES);
 
 		/* remove the KIQ bit as well */
-		if (adev->gfx.kiq.ring.ready)
+		if (adev->gfx.kiq.ring.sched.ready)
 			clear_bit(amdgpu_gfx_queue_to_bit(adev,
 							  adev->gfx.kiq.ring.me - 1,
 							  adev->gfx.kiq.ring.pipe,
@@ -165,7 +182,7 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 				&gpu_resources.doorbell_start_offset);
 
 		if (adev->asic_type < CHIP_VEGA10) {
-			kgd2kfd->device_init(adev->kfd, &gpu_resources);
+			kgd2kfd->device_init(adev->kfd.dev, &gpu_resources);
 			return;
 		}
 
@@ -179,25 +196,14 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 			 * process in case of 64-bit doorbells so we
 			 * can use each doorbell assignment twice.
 			 */
-			if (adev->asic_type == CHIP_VEGA10) {
-				gpu_resources.sdma_doorbell[0][i] =
-					AMDGPU_VEGA10_DOORBELL64_sDMA_ENGINE0 + (i >> 1);
-				gpu_resources.sdma_doorbell[0][i+1] =
-					AMDGPU_VEGA10_DOORBELL64_sDMA_ENGINE0 + 0x200 + (i >> 1);
-				gpu_resources.sdma_doorbell[1][i] =
-					AMDGPU_VEGA10_DOORBELL64_sDMA_ENGINE1 + (i >> 1);
-				gpu_resources.sdma_doorbell[1][i+1] =
-					AMDGPU_VEGA10_DOORBELL64_sDMA_ENGINE1 + 0x200 + (i >> 1);
-			} else {
-				gpu_resources.sdma_doorbell[0][i] =
-					AMDGPU_DOORBELL64_sDMA_ENGINE0 + (i >> 1);
-				gpu_resources.sdma_doorbell[0][i+1] =
-					AMDGPU_DOORBELL64_sDMA_ENGINE0 + 0x200 + (i >> 1);
-				gpu_resources.sdma_doorbell[1][i] =
-					AMDGPU_DOORBELL64_sDMA_ENGINE1 + (i >> 1);
-				gpu_resources.sdma_doorbell[1][i+1] =
-					AMDGPU_DOORBELL64_sDMA_ENGINE1 + 0x200 + (i >> 1);
-			}
+			gpu_resources.sdma_doorbell[0][i] =
+				adev->doorbell_index.sdma_engine0 + (i >> 1);
+			gpu_resources.sdma_doorbell[0][i+1] =
+				adev->doorbell_index.sdma_engine0 + 0x200 + (i >> 1);
+			gpu_resources.sdma_doorbell[1][i] =
+				adev->doorbell_index.sdma_engine1 + (i >> 1);
+			gpu_resources.sdma_doorbell[1][i+1] =
+				adev->doorbell_index.sdma_engine1 + 0x200 + (i >> 1);
 		}
 		/* Doorbells 0x0e0-0ff and 0x2e0-2ff are reserved for
 		 * SDMA, IH and VCN. So don't use them for the CP.
@@ -205,37 +211,37 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 		gpu_resources.reserved_doorbell_mask = 0x1e0;
 		gpu_resources.reserved_doorbell_val  = 0x0e0;
 
-		kgd2kfd->device_init(adev->kfd, &gpu_resources);
+		kgd2kfd->device_init(adev->kfd.dev, &gpu_resources);
 	}
 }
 
 void amdgpu_amdkfd_device_fini(struct amdgpu_device *adev)
 {
-	if (adev->kfd) {
-		kgd2kfd->device_exit(adev->kfd);
-		adev->kfd = NULL;
+	if (adev->kfd.dev) {
+		kgd2kfd->device_exit(adev->kfd.dev);
+		adev->kfd.dev = NULL;
 	}
 }
 
 void amdgpu_amdkfd_interrupt(struct amdgpu_device *adev,
 		const void *ih_ring_entry)
 {
-	if (adev->kfd)
-		kgd2kfd->interrupt(adev->kfd, ih_ring_entry);
+	if (adev->kfd.dev)
+		kgd2kfd->interrupt(adev->kfd.dev, ih_ring_entry);
 }
 
 void amdgpu_amdkfd_suspend(struct amdgpu_device *adev)
 {
-	if (adev->kfd)
-		kgd2kfd->suspend(adev->kfd);
+	if (adev->kfd.dev)
+		kgd2kfd->suspend(adev->kfd.dev);
 }
 
 int amdgpu_amdkfd_resume(struct amdgpu_device *adev)
 {
 	int r = 0;
 
-	if (adev->kfd)
-		r = kgd2kfd->resume(adev->kfd);
+	if (adev->kfd.dev)
+		r = kgd2kfd->resume(adev->kfd.dev);
 
 	return r;
 }
@@ -244,8 +250,8 @@ int amdgpu_amdkfd_pre_reset(struct amdgpu_device *adev)
 {
 	int r = 0;
 
-	if (adev->kfd)
-		r = kgd2kfd->pre_reset(adev->kfd);
+	if (adev->kfd.dev)
+		r = kgd2kfd->pre_reset(adev->kfd.dev);
 
 	return r;
 }
@@ -254,8 +260,8 @@ int amdgpu_amdkfd_post_reset(struct amdgpu_device *adev)
 {
 	int r = 0;
 
-	if (adev->kfd)
-		r = kgd2kfd->post_reset(adev->kfd);
+	if (adev->kfd.dev)
+		r = kgd2kfd->post_reset(adev->kfd.dev);
 
 	return r;
 }
@@ -268,9 +274,9 @@ void amdgpu_amdkfd_gpu_reset(struct kgd_dev *kgd)
 		amdgpu_device_gpu_recover(adev, NULL);
 }
 
-int alloc_gtt_mem(struct kgd_dev *kgd, size_t size,
-			void **mem_obj, uint64_t *gpu_addr,
-			void **cpu_ptr, bool mqd_gfx9)
+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)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
 	struct amdgpu_bo *bo = NULL;
@@ -340,7 +346,7 @@ allocate_mem_reserve_bo_failed:
 	return r;
 }
 
-void free_gtt_mem(struct kgd_dev *kgd, void *mem_obj)
+void amdgpu_amdkfd_free_gtt_mem(struct kgd_dev *kgd, void *mem_obj)
 {
 	struct amdgpu_bo *bo = (struct amdgpu_bo *) mem_obj;
 
@@ -351,8 +357,8 @@ void free_gtt_mem(struct kgd_dev *kgd, void *mem_obj)
 	amdgpu_bo_unref(&(bo));
 }
 
-void get_local_mem_info(struct kgd_dev *kgd,
-			struct kfd_local_mem_info *mem_info)
+void amdgpu_amdkfd_get_local_mem_info(struct kgd_dev *kgd,
+				      struct kfd_local_mem_info *mem_info)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
 	uint64_t address_mask = adev->dev->dma_mask ? ~*adev->dev->dma_mask :
@@ -383,7 +389,7 @@ void get_local_mem_info(struct kgd_dev *kgd,
 		mem_info->mem_clk_max = 100;
 }
 
-uint64_t get_gpu_clock_counter(struct kgd_dev *kgd)
+uint64_t amdgpu_amdkfd_get_gpu_clock_counter(struct kgd_dev *kgd)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
 
@@ -392,7 +398,7 @@ uint64_t get_gpu_clock_counter(struct kgd_dev *kgd)
 	return 0;
 }
 
-uint32_t get_max_engine_clock_in_mhz(struct kgd_dev *kgd)
+uint32_t amdgpu_amdkfd_get_max_engine_clock_in_mhz(struct kgd_dev *kgd)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
 
@@ -405,7 +411,7 @@ uint32_t get_max_engine_clock_in_mhz(struct kgd_dev *kgd)
 		return 100;
 }
 
-void get_cu_info(struct kgd_dev *kgd, struct kfd_cu_info *cu_info)
+void amdgpu_amdkfd_get_cu_info(struct kgd_dev *kgd, struct kfd_cu_info *cu_info)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
 	struct amdgpu_cu_info acu_info = adev->gfx.cu_info;
@@ -428,6 +434,62 @@ void get_cu_info(struct kgd_dev *kgd, struct kfd_cu_info *cu_info)
 	cu_info->lds_size = acu_info.lds_size;
 }
 
+int amdgpu_amdkfd_get_dmabuf_info(struct kgd_dev *kgd, int dma_buf_fd,
+				  struct kgd_dev **dma_buf_kgd,
+				  uint64_t *bo_size, void *metadata_buffer,
+				  size_t buffer_size, uint32_t *metadata_size,
+				  uint32_t *flags)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
+	struct dma_buf *dma_buf;
+	struct drm_gem_object *obj;
+	struct amdgpu_bo *bo;
+	uint64_t metadata_flags;
+	int r = -EINVAL;
+
+	dma_buf = dma_buf_get(dma_buf_fd);
+	if (IS_ERR(dma_buf))
+		return PTR_ERR(dma_buf);
+
+	if (dma_buf->ops != &amdgpu_dmabuf_ops)
+		/* Can't handle non-graphics buffers */
+		goto out_put;
+
+	obj = dma_buf->priv;
+	if (obj->dev->driver != adev->ddev->driver)
+		/* Can't handle buffers from different drivers */
+		goto out_put;
+
+	adev = obj->dev->dev_private;
+	bo = gem_to_amdgpu_bo(obj);
+	if (!(bo->preferred_domains & (AMDGPU_GEM_DOMAIN_VRAM |
+				    AMDGPU_GEM_DOMAIN_GTT)))
+		/* Only VRAM and GTT BOs are supported */
+		goto out_put;
+
+	r = 0;
+	if (dma_buf_kgd)
+		*dma_buf_kgd = (struct kgd_dev *)adev;
+	if (bo_size)
+		*bo_size = amdgpu_bo_size(bo);
+	if (metadata_size)
+		*metadata_size = bo->metadata_size;
+	if (metadata_buffer)
+		r = amdgpu_bo_get_metadata(bo, metadata_buffer, buffer_size,
+					   metadata_size, &metadata_flags);
+	if (flags) {
+		*flags = (bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM) ?
+			ALLOC_MEM_FLAGS_VRAM : ALLOC_MEM_FLAGS_GTT;
+
+		if (bo->flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED)
+			*flags |= ALLOC_MEM_FLAGS_PUBLIC;
+	}
+
+out_put:
+	dma_buf_put(dma_buf);
+	return r;
+}
+
 uint64_t amdgpu_amdkfd_get_vram_usage(struct kgd_dev *kgd)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
@@ -510,7 +572,7 @@ void amdgpu_amdkfd_set_compute_idle(struct kgd_dev *kgd, bool idle)
 
 bool amdgpu_amdkfd_is_kfd_vmid(struct amdgpu_device *adev, u32 vmid)
 {
-	if (adev->kfd) {
+	if (adev->kfd.dev) {
 		if ((1 << vmid) & compute_vmid_bitmap)
 			return true;
 	}
@@ -524,7 +586,7 @@ bool amdkfd_fence_check_mm(struct dma_fence *f, struct mm_struct *mm)
 	return false;
 }
 
-void amdgpu_amdkfd_unreserve_system_memory_limit(struct amdgpu_bo *bo)
+void amdgpu_amdkfd_unreserve_memory_limit(struct amdgpu_bo *bo)
 {
 }