Merge tag 'drm-intel-next-2019-07-30' of git://anongit.freedesktop.org/drm/drm-intel into drm-next

- More changes on simplifying locking mechanisms (Chris)
- Selftests fixes and improvements (Chris)
- More work around engine tracking for better handling (Chris, Tvrtko)
- HDCP debug and info improvements (Ram, Ashuman)
- Add DSI properties (Vandita)
- Rework on sdvo support for better debuggability before fixing bugs (Ville)
- Display PLLs fixes and improvements, specially targeting Ice Lake (Imre, Matt, Ville)
- Perf fixes and improvements (Lionel)
- Enumerate scratch buffers (Lionel)
- Add infra to hold off preemption on a request (Lionel)
- Ice Lake color space fixes (Uma)
- Type-C fixes and improvements (Lucas)
- Fix and improvements around workarounds (Chris, John, Tvrtko)
- GuC related fixes and improvements (Chris, Daniele, Michal, Tvrtko)
- Fix on VLV/CHV display power domain (Ville)
- Improvements around Watermark (Ville)
- Favor intel_ types on intel_atomic functions (Ville)
- Don’t pass stack garbage to pcode (Ville)
- Improve display tracepoints (Steven)
- Don’t overestimate 4:2:0 link symbol clock (Ville)
- Add support for 4th pipe and transcoder (Lucas)
- Introduce initial support for Tiger Lake platform (Daniele, Lucas, Mahesh, Jose, Imre, Mika, Vandita, Rodrigo, Michel)
- PPGTT allocation simplification (Chris)
- Standardize function names and suffixes to make clean, symmetric and let checkpatch happy (Janusz)
- Skip SINK_COUNT read on CH7511 (Ville)
- Fix on kernel documentation (Chris, Michal)
- Add modular FIA (Anusha, Lucas)
- Fix EHL display (Matt, Vivek)
- Enable hotplug retry (Imre, Jose)
- Disable preemption under GVT (Chris)
- OA; Reconfigure context on the fly (Chris)
- Fixes and improvements around engine reset. (Chris)
- Small clean up on display pipe fault mask (Ville)
- Make sure cdclk is high enough for DP audio on VLV/CHV (Ville)
- Drop some wmb() and improve pwrite flush (Chris)
- Fix critical PSR regression (DK)
- Remove unused variables (YueHaibing)
- Use dev_get_drvdata for simplification (Chunhong)
- Use upstream version of header tests (Jani)

drm-intel-next-2019-07-08:
- Signal fence completion from i915_request_wait (Chris)
- Fixes and improvements around rings pin/unpin (Chris)
- Display uncore prep patches (Daniele)
- Execlists preemption improvements (Chris)
- Selftests fixes and improvements (Chris)
- More Elkhartlake enabling work (Vandita, Jose, Matt, Vivek)
- Defer address space cleanup to an RCU worker (Chris)
- Implicit dev_priv removal and GT compartmentalization and other related follow-ups (Tvrtko, Chris)
- Prevent dereference of engine before NULL check in error capture (Chris)
- GuC related fixes (Daniele, Robert)
- Many changes on active tracking, timelines and locking mechanisms (Chris)
- Disable SAMPLER_STATE prefetching on Gen11 (HW W/a) (Kenneth)
- I915_perf fixes (Lionel)
- Add Ice Lake PCI ID (Mika)
- eDP backlight fix (Lee)
- Fix various gen2 tracepoints (Ville)
- Some irq vfunc clean-up and improvements (Ville)
- Move OA files to separated folder (Michal)
- Display self contained headers clean-up (Jani)
- Preparation for 4th pile (Lucas)
- Move atomic commit, watermark and other places to use more intel_crtc_state (Maarten)
- Many Ice Lake Type C and Thunderbolt fixes (Imre)
- Fix some Ice Lake hw w/a whitelist regs (Lionel)
- Fix memleak in runtime wakeref tracking (Mika)
- Remove unused Private PPAT manager (Michal)
- Don't check PPGTT presence on PPGTT-only platforms (Michal)
- Fix ICL DSI suspend/resume (Chris)
- Fix ICL Bandwidth issues (Ville)
- Add N & CTS values for 10/12 bit deep color (Aditya)
- Moving more GT related stuff under gt folder (Chris)
- Forcewake related fixes (Chris)
- Show support for accurate sw PMU busyness tracking (Chris)
- Handle gtt double alloc failures (Chris)
- Upgrade to new GuC version (Michal)
- Improve w/a debug dumps and pull engine w/a initialization into a common (Chris)
- Look for instdone on all engines at hangcheck (Tvrtko)
- Engine lookup simplification  (Chris)
- Many plane color formats fixes and improvements (Ville)
- Fix some compilation issues (YueHaibing)
- GTT page directory clean up and improvements (Mika)

Signed-off-by: Dave Airlie <airlied@redhat.com>

From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190801201314.GA23635@intel.com

25 files changed, 300 insertions, 339 deletions

diff --git a/drivers/gpu/drm/i915/gem/Makefile b/drivers/gpu/drm/i915/gem/Makefile
index 07e7b8b840ea..7e73aa587967 100644
--- a/drivers/gpu/drm/i915/gem/Makefile
+++ b/drivers/gpu/drm/i915/gem/Makefile
@@ -1 +1,5 @@
-include $(src)/Makefile.header-test # Extra header tests
+# For building individual subdir files on the command line
+subdir-ccflags-y += -I$(srctree)/$(src)/..
+
+# Extra header tests
+header-test-pattern-$(CONFIG_DRM_I915_WERROR) := *.h
diff --git a/drivers/gpu/drm/i915/gem/Makefile.header-test b/drivers/gpu/drm/i915/gem/Makefile.header-test
deleted file mode 100644
index 61e06cbb4b32..000000000000
--- a/drivers/gpu/drm/i915/gem/Makefile.header-test
+++ /dev/null
@@ -1,16 +0,0 @@
-# SPDX-License-Identifier: MIT
-# Copyright © 2019 Intel Corporation
-
-# Test the headers are compilable as standalone units
-header_test := $(notdir $(wildcard $(src)/*.h))
-
-quiet_cmd_header_test = HDRTEST $@
-      cmd_header_test = echo "\#include \"$(<F)\"" > $@
-
-header_test_%.c: %.h
-	$(call cmd,header_test)
-
-extra-$(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/gem/i915_gem_client_blt.c b/drivers/gpu/drm/i915/gem/i915_gem_client_blt.c
index 1fdab0767a47..2312a0c6af89 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_client_blt.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_client_blt.c
@@ -72,7 +72,6 @@ static struct i915_sleeve *create_sleeve(struct i915_address_space *vm,
 	vma->ops = &proxy_vma_ops;
 
 	sleeve->vma = vma;
-	sleeve->obj = i915_gem_object_get(obj);
 	sleeve->pages = pages;
 	sleeve->page_sizes = *page_sizes;
 
@@ -85,7 +84,6 @@ err_free:
 
 static void destroy_sleeve(struct i915_sleeve *sleeve)
 {
-	i915_gem_object_put(sleeve->obj);
 	kfree(sleeve);
 }
 
@@ -155,7 +153,7 @@ static void clear_pages_worker(struct work_struct *work)
 {
 	struct clear_pages_work *w = container_of(work, typeof(*w), work);
 	struct drm_i915_private *i915 = w->ce->gem_context->i915;
-	struct drm_i915_gem_object *obj = w->sleeve->obj;
+	struct drm_i915_gem_object *obj = w->sleeve->vma->obj;
 	struct i915_vma *vma = w->sleeve->vma;
 	struct i915_request *rq;
 	int err = w->dma.error;
@@ -164,11 +162,12 @@ static void clear_pages_worker(struct work_struct *work)
 		goto out_signal;
 
 	if (obj->cache_dirty) {
-		obj->write_domain = 0;
 		if (i915_gem_object_has_struct_page(obj))
 			drm_clflush_sg(w->sleeve->pages);
 		obj->cache_dirty = false;
 	}
+	obj->read_domains = I915_GEM_GPU_DOMAINS;
+	obj->write_domain = 0;
 
 	/* XXX: we need to kill this */
 	mutex_lock(&i915->drm.struct_mutex);
@@ -193,10 +192,12 @@ static void clear_pages_worker(struct work_struct *work)
 			goto out_request;
 	}
 
-	/* XXX: more feverish nightmares await */
-	i915_vma_lock(vma);
-	err = i915_vma_move_to_active(vma, rq, EXEC_OBJECT_WRITE);
-	i915_vma_unlock(vma);
+	/*
+	 * w->dma is already exported via (vma|obj)->resv we need only
+	 * keep track of the GPU activity within this vma/request, and
+	 * propagate the signal from the request to w->dma.
+	 */
+	err = i915_active_ref(&vma->active, rq->fence.context, rq);
 	if (err)
 		goto out_request;
 
@@ -249,13 +250,11 @@ int i915_gem_schedule_fill_pages_blt(struct drm_i915_gem_object *obj,
 				     u32 value)
 {
 	struct drm_i915_private *i915 = to_i915(obj->base.dev);
-	struct i915_gem_context *ctx = ce->gem_context;
-	struct i915_address_space *vm = ctx->vm ?: &i915->ggtt.vm;
 	struct clear_pages_work *work;
 	struct i915_sleeve *sleeve;
 	int err;
 
-	sleeve = create_sleeve(vm, obj, pages, page_sizes);
+	sleeve = create_sleeve(ce->vm, obj, pages, page_sizes);
 	if (IS_ERR(sleeve))
 		return PTR_ERR(sleeve);
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_context.c b/drivers/gpu/drm/i915/gem/i915_gem_context.c
index 0f2c22a3bcb6..b28c7ca681a8 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_context.c
@@ -316,7 +316,7 @@ static void i915_gem_context_free(struct i915_gem_context *ctx)
 	mutex_destroy(&ctx->engines_mutex);
 
 	if (ctx->timeline)
-		i915_timeline_put(ctx->timeline);
+		intel_timeline_put(ctx->timeline);
 
 	kfree(ctx->name);
 	put_pid(ctx->pid);
@@ -459,8 +459,7 @@ __create_context(struct drm_i915_private *i915)
 	i915_gem_context_set_recoverable(ctx);
 
 	ctx->ring_size = 4 * PAGE_SIZE;
-	ctx->desc_template =
-		default_desc_template(i915, &i915->mm.aliasing_ppgtt->vm);
+	ctx->desc_template = default_desc_template(i915, NULL);
 
 	for (i = 0; i < ARRAY_SIZE(ctx->hang_timestamp); i++)
 		ctx->hang_timestamp[i] = jiffies - CONTEXT_FAST_HANG_JIFFIES;
@@ -476,10 +475,18 @@ static struct i915_address_space *
 __set_ppgtt(struct i915_gem_context *ctx, struct i915_address_space *vm)
 {
 	struct i915_address_space *old = ctx->vm;
+	struct i915_gem_engines_iter it;
+	struct intel_context *ce;
 
 	ctx->vm = i915_vm_get(vm);
 	ctx->desc_template = default_desc_template(ctx->i915, vm);
 
+	for_each_gem_engine(ce, i915_gem_context_lock_engines(ctx), it) {
+		i915_vm_put(ce->vm);
+		ce->vm = i915_vm_get(vm);
+	}
+	i915_gem_context_unlock_engines(ctx);
+
 	return old;
 }
 
@@ -528,9 +535,9 @@ i915_gem_create_context(struct drm_i915_private *dev_priv, unsigned int flags)
 	}
 
 	if (flags & I915_CONTEXT_CREATE_FLAGS_SINGLE_TIMELINE) {
-		struct i915_timeline *timeline;
+		struct intel_timeline *timeline;
 
-		timeline = i915_timeline_create(dev_priv, NULL);
+		timeline = intel_timeline_create(&dev_priv->gt, NULL);
 		if (IS_ERR(timeline)) {
 			context_close(ctx);
 			return ERR_CAST(timeline);
@@ -644,20 +651,13 @@ static void init_contexts(struct drm_i915_private *i915)
 	init_llist_head(&i915->contexts.free_list);
 }
 
-static bool needs_preempt_context(struct drm_i915_private *i915)
-{
-	return HAS_EXECLISTS(i915);
-}
-
 int i915_gem_contexts_init(struct drm_i915_private *dev_priv)
 {
 	struct i915_gem_context *ctx;
 
 	/* Reassure ourselves we are only called once */
 	GEM_BUG_ON(dev_priv->kernel_context);
-	GEM_BUG_ON(dev_priv->preempt_context);
 
-	intel_engine_init_ctx_wa(dev_priv->engine[RCS0]);
 	init_contexts(dev_priv);
 
 	/* lowest priority; idle task */
@@ -677,15 +677,6 @@ int i915_gem_contexts_init(struct drm_i915_private *dev_priv)
 	GEM_BUG_ON(!atomic_read(&ctx->hw_id_pin_count));
 	dev_priv->kernel_context = ctx;
 
-	/* highest priority; preempting task */
-	if (needs_preempt_context(dev_priv)) {
-		ctx = i915_gem_context_create_kernel(dev_priv, INT_MAX);
-		if (!IS_ERR(ctx))
-			dev_priv->preempt_context = ctx;
-		else
-			DRM_ERROR("Failed to create preempt context; disabling preemption\n");
-	}
-
 	DRM_DEBUG_DRIVER("%s context support initialized\n",
 			 DRIVER_CAPS(dev_priv)->has_logical_contexts ?
 			 "logical" : "fake");
@@ -696,8 +687,6 @@ void i915_gem_contexts_fini(struct drm_i915_private *i915)
 {
 	lockdep_assert_held(&i915->drm.struct_mutex);
 
-	if (i915->preempt_context)
-		destroy_kernel_context(&i915->preempt_context);
 	destroy_kernel_context(&i915->kernel_context);
 
 	/* Must free all deferred contexts (via flush_workqueue) first */
@@ -923,8 +912,12 @@ static int context_barrier_task(struct i915_gem_context *ctx,
 	if (!cb)
 		return -ENOMEM;
 
-	i915_active_init(i915, &cb->base, cb_retire);
-	i915_active_acquire(&cb->base);
+	i915_active_init(i915, &cb->base, NULL, cb_retire);
+	err = i915_active_acquire(&cb->base);
+	if (err) {
+		kfree(cb);
+		return err;
+	}
 
 	for_each_gem_engine(ce, i915_gem_context_lock_engines(ctx), it) {
 		struct i915_request *rq;
@@ -1019,7 +1012,7 @@ static void set_ppgtt_barrier(void *data)
 
 static int emit_ppgtt_update(struct i915_request *rq, void *data)
 {
-	struct i915_address_space *vm = rq->gem_context->vm;
+	struct i915_address_space *vm = rq->hw_context->vm;
 	struct intel_engine_cs *engine = rq->engine;
 	u32 base = engine->mmio_base;
 	u32 *cs;
@@ -1128,9 +1121,8 @@ static int set_ppgtt(struct drm_i915_file_private *file_priv,
 				   set_ppgtt_barrier,
 				   old);
 	if (err) {
-		ctx->vm = old;
-		ctx->desc_template = default_desc_template(ctx->i915, old);
-		i915_vm_put(vm);
+		i915_vm_put(__set_ppgtt(ctx, old));
+		i915_vm_put(old);
 	}
 
 unlock:
@@ -1187,26 +1179,11 @@ gen8_modify_rpcs(struct intel_context *ce, struct intel_sseu sseu)
 	if (IS_ERR(rq))
 		return PTR_ERR(rq);
 
-	/* Queue this switch after all other activity by this context. */
-	ret = i915_active_request_set(&ce->ring->timeline->last_request, rq);
-	if (ret)
-		goto out_add;
-
-	/*
-	 * Guarantee context image and the timeline remains pinned until the
-	 * modifying request is retired by setting the ce activity tracker.
-	 *
-	 * But we only need to take one pin on the account of it. Or in other
-	 * words transfer the pinned ce object to tracked active request.
-	 */
-	GEM_BUG_ON(i915_active_is_idle(&ce->active));
-	ret = i915_active_ref(&ce->active, rq->fence.context, rq);
-	if (ret)
-		goto out_add;
-
-	ret = gen8_emit_rpcs_config(rq, ce, sseu);
+	/* Serialise with the remote context */
+	ret = intel_context_prepare_remote_request(ce, rq);
+	if (ret == 0)
+		ret = gen8_emit_rpcs_config(rq, ce, sseu);
 
-out_add:
 	i915_request_add(rq);
 	return ret;
 }
@@ -2015,8 +1992,8 @@ static int clone_timeline(struct i915_gem_context *dst,
 		GEM_BUG_ON(src->timeline == dst->timeline);
 
 		if (dst->timeline)
-			i915_timeline_put(dst->timeline);
-		dst->timeline = i915_timeline_get(src->timeline);
+			intel_timeline_put(dst->timeline);
+		dst->timeline = intel_timeline_get(src->timeline);
 	}
 
 	return 0;
@@ -2141,7 +2118,7 @@ int i915_gem_context_create_ioctl(struct drm_device *dev, void *data,
 	if (args->flags & I915_CONTEXT_CREATE_FLAGS_UNKNOWN)
 		return -EINVAL;
 
-	ret = i915_terminally_wedged(i915);
+	ret = intel_gt_terminally_wedged(&i915->gt);
 	if (ret)
 		return ret;
 
@@ -2287,8 +2264,8 @@ int i915_gem_context_getparam_ioctl(struct drm_device *dev, void *data,
 		args->size = 0;
 		if (ctx->vm)
 			args->value = ctx->vm->total;
-		else if (to_i915(dev)->mm.aliasing_ppgtt)
-			args->value = to_i915(dev)->mm.aliasing_ppgtt->vm.total;
+		else if (to_i915(dev)->ggtt.alias)
+			args->value = to_i915(dev)->ggtt.alias->vm.total;
 		else
 			args->value = to_i915(dev)->ggtt.vm.total;
 		break;
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_context.h b/drivers/gpu/drm/i915/gem/i915_gem_context.h
index 9691dd062f72..106e2ccf7a4c 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_context.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_context.h
@@ -198,12 +198,6 @@ i915_gem_context_unlock_engines(struct i915_gem_context *ctx)
 }
 
 static inline struct intel_context *
-i915_gem_context_lookup_engine(struct i915_gem_context *ctx, unsigned int idx)
-{
-	return i915_gem_context_engines(ctx)->engines[idx];
-}
-
-static inline struct intel_context *
 i915_gem_context_get_engine(struct i915_gem_context *ctx, unsigned int idx)
 {
 	struct intel_context *ce = ERR_PTR(-EINVAL);
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_context_types.h b/drivers/gpu/drm/i915/gem/i915_gem_context_types.h
index cc513410eeef..0ee61482ef94 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_context_types.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_context_types.h
@@ -26,7 +26,7 @@ struct pid;
 struct drm_i915_private;
 struct drm_i915_file_private;
 struct i915_address_space;
-struct i915_timeline;
+struct intel_timeline;
 struct intel_ring;
 
 struct i915_gem_engines {
@@ -77,7 +77,7 @@ struct i915_gem_context {
 	struct i915_gem_engines __rcu *engines;
 	struct mutex engines_mutex; /* guards writes to engines */
 
-	struct i915_timeline *timeline;
+	struct intel_timeline *timeline;
 
 	/**
 	 * @vm: unique address space (GTT)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
index 5fae0e50aad0..cbd7c6e3a1f8 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
@@ -16,6 +16,7 @@
 
 #include "gem/i915_gem_ioctls.h"
 #include "gt/intel_context.h"
+#include "gt/intel_gt.h"
 #include "gt/intel_gt_pm.h"
 
 #include "i915_gem_ioctls.h"
@@ -222,7 +223,6 @@ struct i915_execbuffer {
 	struct intel_engine_cs *engine; /** engine to queue the request to */
 	struct intel_context *context; /* logical state for the request */
 	struct i915_gem_context *gem_context; /** caller's context */
-	struct i915_address_space *vm; /** GTT and vma for the request */
 
 	struct i915_request *request; /** our request to build */
 	struct i915_vma *batch; /** identity of the batch obj/vma */
@@ -696,7 +696,7 @@ static int eb_reserve(struct i915_execbuffer *eb)
 
 		case 1:
 			/* Too fragmented, unbind everything and retry */
-			err = i915_gem_evict_vm(eb->vm);
+			err = i915_gem_evict_vm(eb->context->vm);
 			if (err)
 				return err;
 			break;
@@ -724,12 +724,8 @@ static int eb_select_context(struct i915_execbuffer *eb)
 		return -ENOENT;
 
 	eb->gem_context = ctx;
-	if (ctx->vm) {
-		eb->vm = ctx->vm;
+	if (ctx->vm)
 		eb->invalid_flags |= EXEC_OBJECT_NEEDS_GTT;
-	} else {
-		eb->vm = &eb->i915->ggtt.vm;
-	}
 
 	eb->context_flags = 0;
 	if (test_bit(UCONTEXT_NO_ZEROMAP, &ctx->user_flags))
@@ -831,7 +827,7 @@ static int eb_lookup_vmas(struct i915_execbuffer *eb)
 			goto err_vma;
 		}
 
-		vma = i915_vma_instance(obj, eb->vm, NULL);
+		vma = i915_vma_instance(obj, eb->context->vm, NULL);
 		if (IS_ERR(vma)) {
 			err = PTR_ERR(vma);
 			goto err_obj;
@@ -994,7 +990,7 @@ static void reloc_gpu_flush(struct reloc_cache *cache)
 	__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);
+	intel_gt_chipset_flush(cache->rq->engine->gt);
 
 	i915_request_add(cache->rq);
 	cache->rq = NULL;
@@ -1954,7 +1950,7 @@ static int eb_move_to_gpu(struct i915_execbuffer *eb)
 	eb->exec = NULL;
 
 	/* Unconditionally flush any chipset caches (for streaming writes). */
-	i915_gem_chipset_flush(eb->i915);
+	intel_gt_chipset_flush(eb->engine->gt);
 	return 0;
 
 err_skip:
@@ -2129,7 +2125,7 @@ static int eb_pin_context(struct i915_execbuffer *eb, struct intel_context *ce)
 	 * ABI: Before userspace accesses the GPU (e.g. execbuffer), report
 	 * EIO if the GPU is already wedged.
 	 */
-	err = i915_terminally_wedged(eb->i915);
+	err = intel_gt_terminally_wedged(ce->engine->gt);
 	if (err)
 		return err;
 
@@ -2436,7 +2432,7 @@ i915_gem_do_execbuffer(struct drm_device *dev,
 	 * wakeref that we hold until the GPU has been idle for at least
 	 * 100ms.
 	 */
-	intel_gt_pm_get(eb.i915);
+	intel_gt_pm_get(&eb.i915->gt);
 
 	err = i915_mutex_lock_interruptible(dev);
 	if (err)
@@ -2606,7 +2602,7 @@ err_engine:
 err_unlock:
 	mutex_unlock(&dev->struct_mutex);
 err_rpm:
-	intel_gt_pm_put(eb.i915);
+	intel_gt_pm_put(&eb.i915->gt);
 	i915_gem_context_put(eb.gem_context);
 err_destroy:
 	eb_destroy(&eb);
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/i915_gem_mman.c
index 39a661927d8e..dfa525e37eb8 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_mman.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.c
@@ -7,6 +7,8 @@
 #include <linux/mman.h>
 #include <linux/sizes.h>
 
+#include "gt/intel_gt.h"
+
 #include "i915_drv.h"
 #include "i915_gem_gtt.h"
 #include "i915_gem_ioctls.h"
@@ -246,7 +248,7 @@ vm_fault_t i915_gem_fault(struct vm_fault *vmf)
 
 	wakeref = intel_runtime_pm_get(rpm);
 
-	srcu = i915_reset_trylock(i915);
+	srcu = intel_gt_reset_trylock(ggtt->vm.gt);
 	if (srcu < 0) {
 		ret = srcu;
 		goto err_rpm;
@@ -326,7 +328,7 @@ err_unpin:
 err_unlock:
 	mutex_unlock(&dev->struct_mutex);
 err_reset:
-	i915_reset_unlock(i915, srcu);
+	intel_gt_reset_unlock(ggtt->vm.gt, srcu);
 err_rpm:
 	intel_runtime_pm_put(rpm, wakeref);
 	i915_gem_object_unpin_pages(obj);
@@ -339,7 +341,7 @@ err:
 		 * fail). But any other -EIO isn't ours (e.g. swap in failure)
 		 * and so needs to be reported.
 		 */
-		if (!i915_terminally_wedged(i915))
+		if (!intel_gt_is_wedged(ggtt->vm.gt))
 			return VM_FAULT_SIGBUS;
 		/* else, fall through */
 	case -EAGAIN:
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c
index be6caccce0c5..d5197a2a106f 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c
@@ -23,7 +23,7 @@
  */
 
 #include "display/intel_frontbuffer.h"
-
+#include "gt/intel_gt.h"
 #include "i915_drv.h"
 #include "i915_gem_clflush.h"
 #include "i915_gem_context.h"
@@ -146,6 +146,19 @@ void i915_gem_close_object(struct drm_gem_object *gem, struct drm_file *file)
 	}
 }
 
+static void __i915_gem_free_object_rcu(struct rcu_head *head)
+{
+	struct drm_i915_gem_object *obj =
+		container_of(head, typeof(*obj), rcu);
+	struct drm_i915_private *i915 = to_i915(obj->base.dev);
+
+	reservation_object_fini(&obj->base._resv);
+	i915_gem_object_free(obj);
+
+	GEM_BUG_ON(!atomic_read(&i915->mm.free_count));
+	atomic_dec(&i915->mm.free_count);
+}
+
 static void __i915_gem_free_objects(struct drm_i915_private *i915,
 				    struct llist_node *freed)
 {
@@ -160,7 +173,6 @@ static void __i915_gem_free_objects(struct drm_i915_private *i915,
 
 		mutex_lock(&i915->drm.struct_mutex);
 
-		GEM_BUG_ON(i915_gem_object_is_active(obj));
 		list_for_each_entry_safe(vma, vn, &obj->vma.list, obj_link) {
 			GEM_BUG_ON(i915_vma_is_active(vma));
 			vma->flags &= ~I915_VMA_PIN_MASK;
@@ -169,22 +181,6 @@ static void __i915_gem_free_objects(struct drm_i915_private *i915,
 		GEM_BUG_ON(!list_empty(&obj->vma.list));
 		GEM_BUG_ON(!RB_EMPTY_ROOT(&obj->vma.tree));
 
-		/*
-		 * This serializes freeing with the shrinker. Since the free
-		 * is delayed, first by RCU then by the workqueue, we want the
-		 * shrinker to be able to free pages of unreferenced objects,
-		 * or else we may oom whilst there are plenty of deferred
-		 * freed objects.
-		 */
-		if (i915_gem_object_has_pages(obj) &&
-		    i915_gem_object_is_shrinkable(obj)) {
-			unsigned long flags;
-
-			spin_lock_irqsave(&i915->mm.obj_lock, flags);
-			list_del_init(&obj->mm.link);
-			spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
-		}
-
 		mutex_unlock(&i915->drm.struct_mutex);
 
 		GEM_BUG_ON(atomic_read(&obj->bind_count));
@@ -192,25 +188,21 @@ static void __i915_gem_free_objects(struct drm_i915_private *i915,
 		GEM_BUG_ON(atomic_read(&obj->frontbuffer_bits));
 		GEM_BUG_ON(!list_empty(&obj->lut_list));
 
-		if (obj->ops->release)
-			obj->ops->release(obj);
-
 		atomic_set(&obj->mm.pages_pin_count, 0);
 		__i915_gem_object_put_pages(obj, I915_MM_NORMAL);
 		GEM_BUG_ON(i915_gem_object_has_pages(obj));
+		bitmap_free(obj->bit_17);
 
 		if (obj->base.import_attach)
 			drm_prime_gem_destroy(&obj->base, NULL);
 
-		drm_gem_object_release(&obj->base);
+		drm_gem_free_mmap_offset(&obj->base);
 
-		bitmap_free(obj->bit_17);
-		i915_gem_object_free(obj);
-
-		GEM_BUG_ON(!atomic_read(&i915->mm.free_count));
-		atomic_dec(&i915->mm.free_count);
+		if (obj->ops->release)
+			obj->ops->release(obj);
 
-		cond_resched();
+		/* But keep the pointer alive for RCU-protected lookups */
+		call_rcu(&obj->rcu, __i915_gem_free_object_rcu);
 	}
 	intel_runtime_pm_put(&i915->runtime_pm, wakeref);
 }
@@ -261,18 +253,34 @@ static void __i915_gem_free_work(struct work_struct *work)
 	spin_unlock(&i915->mm.free_lock);
 }
 
-static void __i915_gem_free_object_rcu(struct rcu_head *head)
+void i915_gem_free_object(struct drm_gem_object *gem_obj)
 {
-	struct drm_i915_gem_object *obj =
-		container_of(head, typeof(*obj), rcu);
+	struct drm_i915_gem_object *obj = to_intel_bo(gem_obj);
 	struct drm_i915_private *i915 = to_i915(obj->base.dev);
 
 	/*
-	 * We reuse obj->rcu for the freed list, so we had better not treat
-	 * it like a rcu_head from this point forwards. And we expect all
-	 * objects to be freed via this path.
+	 * Before we free the object, make sure any pure RCU-only
+	 * read-side critical sections are complete, e.g.
+	 * i915_gem_busy_ioctl(). For the corresponding synchronized
+	 * lookup see i915_gem_object_lookup_rcu().
 	 */
-	destroy_rcu_head(&obj->rcu);
+	atomic_inc(&i915->mm.free_count);
+
+	/*
+	 * This serializes freeing with the shrinker. Since the free
+	 * is delayed, first by RCU then by the workqueue, we want the
+	 * shrinker to be able to free pages of unreferenced objects,
+	 * or else we may oom whilst there are plenty of deferred
+	 * freed objects.
+	 */
+	if (i915_gem_object_has_pages(obj) &&
+	    i915_gem_object_is_shrinkable(obj)) {
+		unsigned long flags;
+
+		spin_lock_irqsave(&i915->mm.obj_lock, flags);
+		list_del_init(&obj->mm.link);
+		spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
+	}
 
 	/*
 	 * Since we require blocking on struct_mutex to unbind the freed
@@ -288,20 +296,6 @@ static void __i915_gem_free_object_rcu(struct rcu_head *head)
 		queue_work(i915->wq, &i915->mm.free_work);
 }
 
-void i915_gem_free_object(struct drm_gem_object *gem_obj)
-{
-	struct drm_i915_gem_object *obj = to_intel_bo(gem_obj);
-
-	/*
-	 * Before we free the object, make sure any pure RCU-only
-	 * read-side critical sections are complete, e.g.
-	 * i915_gem_busy_ioctl(). For the corresponding synchronized
-	 * lookup see i915_gem_object_lookup_rcu().
-	 */
-	atomic_inc(&to_i915(obj->base.dev)->mm.free_count);
-	call_rcu(&obj->rcu, __i915_gem_free_object_rcu);
-}
-
 static inline enum fb_op_origin
 fb_write_origin(struct drm_i915_gem_object *obj, unsigned int domain)
 {
@@ -319,7 +313,6 @@ void
 i915_gem_object_flush_write_domain(struct drm_i915_gem_object *obj,
 				   unsigned int flush_domains)
 {
-	struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
 	struct i915_vma *vma;
 
 	assert_object_held(obj);
@@ -329,7 +322,8 @@ i915_gem_object_flush_write_domain(struct drm_i915_gem_object *obj,
 
 	switch (obj->write_domain) {
 	case I915_GEM_DOMAIN_GTT:
-		i915_gem_flush_ggtt_writes(dev_priv);
+		for_each_ggtt_vma(vma, obj)
+			intel_gt_flush_ggtt_writes(vma->vm->gt);
 
 		intel_fb_obj_flush(obj,
 				   fb_write_origin(obj, I915_GEM_DOMAIN_GTT));
@@ -340,6 +334,7 @@ i915_gem_object_flush_write_domain(struct drm_i915_gem_object *obj,
 
 			i915_vma_unset_ggtt_write(vma);
 		}
+
 		break;
 
 	case I915_GEM_DOMAIN_WC:
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h
index dfebd5706f16..67aea07ea019 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h
@@ -81,7 +81,7 @@ i915_gem_object_lookup(struct drm_file *file, u32 handle)
 }
 
 __deprecated
-extern struct drm_gem_object *
+struct drm_gem_object *
 drm_gem_object_lookup(struct drm_file *file, u32 handle);
 
 __attribute__((nonnull))
@@ -159,12 +159,6 @@ i915_gem_object_needs_async_cancel(const struct drm_i915_gem_object *obj)
 }
 
 static inline bool
-i915_gem_object_is_active(const struct drm_i915_gem_object *obj)
-{
-	return READ_ONCE(obj->active_count);
-}
-
-static inline bool
 i915_gem_object_is_framebuffer(const struct drm_i915_gem_object *obj)
 {
 	return READ_ONCE(obj->framebuffer_references);
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_blt.c b/drivers/gpu/drm/i915/gem/i915_gem_object_blt.c
index cb42e3a312e2..685064af32d1 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object_blt.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object_blt.c
@@ -47,15 +47,11 @@ int i915_gem_object_fill_blt(struct drm_i915_gem_object *obj,
 			     struct intel_context *ce,
 			     u32 value)
 {
-	struct drm_i915_private *i915 = to_i915(obj->base.dev);
-	struct i915_gem_context *ctx = ce->gem_context;
-	struct i915_address_space *vm = ctx->vm ?: &i915->ggtt.vm;
 	struct i915_request *rq;
 	struct i915_vma *vma;
 	int err;
 
-	/* XXX: ce->vm please */
-	vma = i915_vma_instance(obj, vm, NULL);
+	vma = i915_vma_instance(obj, ce->vm, NULL);
 	if (IS_ERR(vma))
 		return PTR_ERR(vma);
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
index 18bf4f8d6d80..34b51fad02de 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
@@ -154,7 +154,6 @@ struct drm_i915_gem_object {
 
 	/** Count of VMA actually bound by this object */
 	atomic_t bind_count;
-	unsigned int active_count;
 	/** Count of how many global VMA are currently pinned for use by HW */
 	unsigned int pin_global;
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_phys.c b/drivers/gpu/drm/i915/gem/i915_gem_phys.c
index 2deac933cf59..102fd7a23d3d 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_phys.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_phys.c
@@ -13,6 +13,7 @@
 #include <drm/drm_legacy.h> /* for drm_pci.h! */
 #include <drm/drm_pci.h>
 
+#include "gt/intel_gt.h"
 #include "i915_drv.h"
 #include "i915_gem_object.h"
 #include "i915_scatterlist.h"
@@ -60,7 +61,7 @@ static int i915_gem_object_get_pages_phys(struct drm_i915_gem_object *obj)
 		vaddr += PAGE_SIZE;
 	}
 
-	i915_gem_chipset_flush(to_i915(obj->base.dev));
+	intel_gt_chipset_flush(&to_i915(obj->base.dev)->gt);
 
 	st = kmalloc(sizeof(*st), GFP_KERNEL);
 	if (!st) {
@@ -132,16 +133,9 @@ i915_gem_object_put_pages_phys(struct drm_i915_gem_object *obj,
 	drm_pci_free(obj->base.dev, obj->phys_handle);
 }
 
-static void
-i915_gem_object_release_phys(struct drm_i915_gem_object *obj)
-{
-	i915_gem_object_unpin_pages(obj);
-}
-
 static const struct drm_i915_gem_object_ops i915_gem_phys_ops = {
 	.get_pages = i915_gem_object_get_pages_phys,
 	.put_pages = i915_gem_object_put_pages_phys,
-	.release = i915_gem_object_release_phys,
 };
 
 int i915_gem_object_attach_phys(struct drm_i915_gem_object *obj, int align)
@@ -158,7 +152,7 @@ int i915_gem_object_attach_phys(struct drm_i915_gem_object *obj, int align)
 	if (obj->ops != &i915_gem_shmem_ops)
 		return -EINVAL;
 
-	err = i915_gem_object_unbind(obj);
+	err = i915_gem_object_unbind(obj, I915_GEM_OBJECT_UNBIND_ACTIVE);
 	if (err)
 		return err;
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pm.c b/drivers/gpu/drm/i915/gem/i915_gem_pm.c
index 914b5d4112bb..b5561cbdc5ea 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pm.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pm.c
@@ -5,6 +5,7 @@
  */
 
 #include "gem/i915_gem_pm.h"
+#include "gt/intel_gt.h"
 #include "gt/intel_gt_pm.h"
 
 #include "i915_drv.h"
@@ -38,7 +39,7 @@ static void i915_gem_park(struct drm_i915_private *i915)
 		i915_gem_batch_pool_fini(&engine->batch_pool);
 	}
 
-	i915_timelines_park(i915);
+	intel_timelines_park(i915);
 	i915_vma_parked(i915);
 
 	i915_globals_park();
@@ -54,7 +55,8 @@ static void idle_work_handler(struct work_struct *work)
 	mutex_lock(&i915->drm.struct_mutex);
 
 	intel_wakeref_lock(&i915->gt.wakeref);
-	park = !intel_wakeref_active(&i915->gt.wakeref) && !work_pending(work);
+	park = (!intel_wakeref_is_active(&i915->gt.wakeref) &&
+		!work_pending(work));
 	intel_wakeref_unlock(&i915->gt.wakeref);
 	if (park)
 		i915_gem_park(i915);
@@ -105,18 +107,18 @@ static int pm_notifier(struct notifier_block *nb,
 	return NOTIFY_OK;
 }
 
-static bool switch_to_kernel_context_sync(struct drm_i915_private *i915)
+static bool switch_to_kernel_context_sync(struct intel_gt *gt)
 {
-	bool result = !i915_terminally_wedged(i915);
+	bool result = !intel_gt_is_wedged(gt);
 
 	do {
-		if (i915_gem_wait_for_idle(i915,
+		if (i915_gem_wait_for_idle(gt->i915,
 					   I915_WAIT_LOCKED |
 					   I915_WAIT_FOR_IDLE_BOOST,
 					   I915_GEM_IDLE_TIMEOUT) == -ETIME) {
 			/* XXX hide warning from gem_eio */
 			if (i915_modparams.reset) {
-				dev_err(i915->drm.dev,
+				dev_err(gt->i915->drm.dev,
 					"Failed to idle engines, declaring wedged!\n");
 				GEM_TRACE_DUMP();
 			}
@@ -125,18 +127,18 @@ static bool switch_to_kernel_context_sync(struct drm_i915_private *i915)
 			 * Forcibly cancel outstanding work and leave
 			 * the gpu quiet.
 			 */
-			i915_gem_set_wedged(i915);
+			intel_gt_set_wedged(gt);
 			result = false;
 		}
-	} while (i915_retire_requests(i915) && result);
+	} while (i915_retire_requests(gt->i915) && result);
 
-	GEM_BUG_ON(i915->gt.awake);
+	GEM_BUG_ON(gt->awake);
 	return result;
 }
 
 bool i915_gem_load_power_context(struct drm_i915_private *i915)
 {
-	return switch_to_kernel_context_sync(i915);
+	return switch_to_kernel_context_sync(&i915->gt);
 }
 
 void i915_gem_suspend(struct drm_i915_private *i915)
@@ -157,7 +159,7 @@ void i915_gem_suspend(struct drm_i915_private *i915)
 	 * state. Fortunately, the kernel_context is disposable and we do
 	 * not rely on its state.
 	 */
-	switch_to_kernel_context_sync(i915);
+	switch_to_kernel_context_sync(&i915->gt);
 
 	mutex_unlock(&i915->drm.struct_mutex);
 
@@ -168,11 +170,11 @@ void i915_gem_suspend(struct drm_i915_private *i915)
 	GEM_BUG_ON(i915->gt.awake);
 	flush_work(&i915->gem.idle_work);
 
-	cancel_delayed_work_sync(&i915->gpu_error.hangcheck_work);
+	cancel_delayed_work_sync(&i915->gt.hangcheck.work);
 
 	i915_gem_drain_freed_objects(i915);
 
-	intel_uc_suspend(i915);
+	intel_uc_suspend(&i915->gt.uc);
 }
 
 static struct drm_i915_gem_object *first_mm_object(struct list_head *list)
@@ -237,7 +239,6 @@ void i915_gem_suspend_late(struct drm_i915_private *i915)
 	}
 	spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
 
-	intel_uc_sanitize(i915);
 	i915_gem_sanitize(i915);
 }
 
@@ -261,10 +262,10 @@ 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.
 	 */
-	if (intel_gt_resume(i915))
+	if (intel_gt_resume(&i915->gt))
 		goto err_wedged;
 
-	intel_uc_resume(i915);
+	intel_uc_resume(&i915->gt.uc);
 
 	/* Always reload a context for powersaving. */
 	if (!i915_gem_load_power_context(i915))
@@ -276,10 +277,10 @@ out_unlock:
 	return;
 
 err_wedged:
-	if (!i915_reset_failed(i915)) {
+	if (!intel_gt_is_wedged(&i915->gt)) {
 		dev_err(i915->drm.dev,
 			"Failed to re-initialize GPU, declaring it wedged!\n");
-		i915_gem_set_wedged(i915);
+		intel_gt_set_wedged(&i915->gt);
 	}
 	goto out_unlock;
 }
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
index 19d9ecdb2894..d2a1158868e7 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
@@ -414,6 +414,11 @@ shmem_pwrite(struct drm_i915_gem_object *obj,
 	return 0;
 }
 
+static void shmem_release(struct drm_i915_gem_object *obj)
+{
+	fput(obj->base.filp);
+}
+
 const struct drm_i915_gem_object_ops i915_gem_shmem_ops = {
 	.flags = I915_GEM_OBJECT_HAS_STRUCT_PAGE |
 		 I915_GEM_OBJECT_IS_SHRINKABLE,
@@ -424,6 +429,8 @@ const struct drm_i915_gem_object_ops i915_gem_shmem_ops = {
 	.writeback = shmem_writeback,
 
 	.pwrite = shmem_pwrite,
+
+	.release = shmem_release,
 };
 
 static int create_shmem(struct drm_i915_private *i915,
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
index 3a926a8755c6..3f4c6bdcc3c3 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
@@ -88,10 +88,18 @@ static bool can_release_pages(struct drm_i915_gem_object *obj)
 	return swap_available() || obj->mm.madv == I915_MADV_DONTNEED;
 }
 
-static bool unsafe_drop_pages(struct drm_i915_gem_object *obj)
+static bool unsafe_drop_pages(struct drm_i915_gem_object *obj,
+			      unsigned long shrink)
 {
-	if (i915_gem_object_unbind(obj) == 0)
+	unsigned long flags;
+
+	flags = 0;
+	if (shrink & I915_SHRINK_ACTIVE)
+		flags = I915_GEM_OBJECT_UNBIND_ACTIVE;
+
+	if (i915_gem_object_unbind(obj, flags) == 0)
 		__i915_gem_object_put_pages(obj, I915_MM_SHRINKER);
+
 	return !i915_gem_object_has_pages(obj);
 }
 
@@ -169,7 +177,6 @@ i915_gem_shrink(struct drm_i915_private *i915,
 	 */
 
 	trace_i915_gem_shrink(i915, target, shrink);
-	i915_retire_requests(i915);
 
 	/*
 	 * Unbinding of objects will require HW access; Let us not wake the
@@ -230,8 +237,7 @@ i915_gem_shrink(struct drm_i915_private *i915,
 				continue;
 
 			if (!(shrink & I915_SHRINK_ACTIVE) &&
-			    (i915_gem_object_is_active(obj) ||
-			     i915_gem_object_is_framebuffer(obj)))
+			    i915_gem_object_is_framebuffer(obj))
 				continue;
 
 			if (!(shrink & I915_SHRINK_BOUND) &&
@@ -246,7 +252,7 @@ i915_gem_shrink(struct drm_i915_private *i915,
 
 			spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
 
-			if (unsafe_drop_pages(obj)) {
+			if (unsafe_drop_pages(obj, shrink)) {
 				/* May arrive from get_pages on another bo */
 				mutex_lock_nested(&obj->mm.lock,
 						  I915_MM_SHRINKER);
@@ -269,8 +275,6 @@ i915_gem_shrink(struct drm_i915_private *i915,
 	if (shrink & I915_SHRINK_BOUND)
 		intel_runtime_pm_put(&i915->runtime_pm, wakeref);
 
-	i915_retire_requests(i915);
-
 	shrinker_unlock(i915, unlock);
 
 	if (nr_scanned)
@@ -427,12 +431,6 @@ i915_gem_shrinker_vmap(struct notifier_block *nb, unsigned long event, void *ptr
 	if (!shrinker_lock(i915, 0, &unlock))
 		return NOTIFY_DONE;
 
-	/* Force everything onto the inactive lists */
-	if (i915_gem_wait_for_idle(i915,
-				   I915_WAIT_LOCKED,
-				   MAX_SCHEDULE_TIMEOUT))
-		goto out;
-
 	with_intel_runtime_pm(&i915->runtime_pm, wakeref)
 		freed_pages += i915_gem_shrink(i915, -1UL, NULL,
 					       I915_SHRINK_BOUND |
@@ -455,7 +453,6 @@ i915_gem_shrinker_vmap(struct notifier_block *nb, unsigned long event, void *ptr
 	}
 	mutex_unlock(&i915->ggtt.vm.mutex);
 
-out:
 	shrinker_unlock(i915, unlock);
 
 	*(unsigned long *)ptr += freed_pages;
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_stolen.c b/drivers/gpu/drm/i915/gem/i915_gem_stolen.c
index de1fab2058ec..639c852bad12 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_stolen.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_stolen.c
@@ -529,8 +529,6 @@ i915_gem_object_release_stolen(struct drm_i915_gem_object *obj)
 
 	GEM_BUG_ON(!stolen);
 
-	__i915_gem_object_unpin_pages(obj);
-
 	i915_gem_stolen_remove_node(dev_priv, stolen);
 	kfree(stolen);
 }
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_throttle.c b/drivers/gpu/drm/i915/gem/i915_gem_throttle.c
index adb3074d9ce2..1e372420771b 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_throttle.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_throttle.c
@@ -41,7 +41,7 @@ i915_gem_throttle_ioctl(struct drm_device *dev, void *data,
 	long ret;
 
 	/* ABI: return -EIO if already wedged */
-	ret = i915_terminally_wedged(to_i915(dev));
+	ret = intel_gt_terminally_wedged(&to_i915(dev)->gt);
 	if (ret)
 		return ret;
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c
index 2caa594322bc..b9d2bb15e4a6 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c
@@ -150,7 +150,8 @@ userptr_mn_invalidate_range_start(struct mmu_notifier *_mn,
 			}
 		}
 
-		ret = i915_gem_object_unbind(obj);
+		ret = i915_gem_object_unbind(obj,
+					     I915_GEM_OBJECT_UNBIND_ACTIVE);
 		if (ret == 0)
 			ret = __i915_gem_object_put_pages(obj, I915_MM_SHRINKER);
 		i915_gem_object_put(obj);
@@ -662,6 +663,14 @@ i915_gem_userptr_put_pages(struct drm_i915_gem_object *obj,
 	__i915_gem_object_release_shmem(obj, pages, true);
 	i915_gem_gtt_finish_pages(obj, pages);
 
+	/*
+	 * We always mark objects as dirty when they are used by the GPU,
+	 * just in case. However, if we set the vma as being read-only we know
+	 * that the object will never have been written to.
+	 */
+	if (i915_gem_object_is_readonly(obj))
+		obj->mm.dirty = false;
+
 	for_each_sgt_page(page, sgt_iter, pages) {
 		if (obj->mm.dirty)
 			/*
diff --git a/drivers/gpu/drm/i915/gem/selftests/huge_pages.c b/drivers/gpu/drm/i915/gem/selftests/huge_pages.c
index b74729b6f353..6cbd4a668c9a 100644
--- a/drivers/gpu/drm/i915/gem/selftests/huge_pages.c
+++ b/drivers/gpu/drm/i915/gem/selftests/huge_pages.c
@@ -10,6 +10,8 @@
 
 #include "gem/i915_gem_pm.h"
 
+#include "gt/intel_gt.h"
+
 #include "igt_gem_utils.h"
 #include "mock_context.h"
 
@@ -926,7 +928,7 @@ gpu_write_dw(struct i915_vma *vma, u64 offset, u32 val)
 	}
 
 	*cmd = MI_BATCH_BUFFER_END;
-	i915_gem_chipset_flush(i915);
+	intel_gt_chipset_flush(vma->vm->gt);
 
 	i915_gem_object_unpin_map(obj);
 
@@ -1037,8 +1039,7 @@ static int __igt_write_huge(struct i915_gem_context *ctx,
 			    u64 size, u64 offset,
 			    u32 dword, u32 val)
 {
-	struct drm_i915_private *i915 = to_i915(obj->base.dev);
-	struct i915_address_space *vm = ctx->vm ?: &i915->ggtt.vm;
+	struct i915_address_space *vm = ctx->vm ?: &engine->gt->ggtt->vm;
 	unsigned int flags = PIN_USER | PIN_OFFSET_FIXED;
 	struct i915_vma *vma;
 	int err;
@@ -1421,6 +1422,9 @@ static int igt_ppgtt_pin_update(void *arg)
 	struct drm_i915_gem_object *obj;
 	struct i915_vma *vma;
 	unsigned int flags = PIN_USER | PIN_OFFSET_FIXED;
+	struct intel_engine_cs *engine;
+	enum intel_engine_id id;
+	unsigned int n;
 	int first, last;
 	int err;
 
@@ -1518,11 +1522,20 @@ static int igt_ppgtt_pin_update(void *arg)
 	 * land in the now stale 2M page.
 	 */
 
-	err = gpu_write(vma, ctx, dev_priv->engine[RCS0], 0, 0xdeadbeaf);
-	if (err)
-		goto out_unpin;
+	n = 0;
+	for_each_engine(engine, dev_priv, id) {
+		if (!intel_engine_can_store_dword(engine))
+			continue;
 
-	err = cpu_check(obj, 0, 0xdeadbeaf);
+		err = gpu_write(vma, ctx, engine, n++, 0xdeadbeaf);
+		if (err)
+			goto out_unpin;
+	}
+	while (n--) {
+		err = cpu_check(obj, n, 0xdeadbeaf);
+		if (err)
+			goto out_unpin;
+	}
 
 out_unpin:
 	i915_vma_unpin(vma);
@@ -1598,8 +1611,11 @@ static int igt_shrink_thp(void *arg)
 	struct drm_i915_private *i915 = ctx->i915;
 	struct i915_address_space *vm = ctx->vm ?: &i915->ggtt.vm;
 	struct drm_i915_gem_object *obj;
+	struct intel_engine_cs *engine;
+	enum intel_engine_id id;
 	struct i915_vma *vma;
 	unsigned int flags = PIN_USER;
+	unsigned int n;
 	int err;
 
 	/*
@@ -1635,9 +1651,15 @@ static int igt_shrink_thp(void *arg)
 	if (err)
 		goto out_unpin;
 
-	err = gpu_write(vma, ctx, i915->engine[RCS0], 0, 0xdeadbeaf);
-	if (err)
-		goto out_unpin;
+	n = 0;
+	for_each_engine(engine, i915, id) {
+		if (!intel_engine_can_store_dword(engine))
+			continue;
+
+		err = gpu_write(vma, ctx, engine, n++, 0xdeadbeaf);
+		if (err)
+			goto out_unpin;
+	}
 
 	i915_vma_unpin(vma);
 
@@ -1662,7 +1684,12 @@ static int igt_shrink_thp(void *arg)
 	if (err)
 		goto out_close;
 
-	err = cpu_check(obj, 0, 0xdeadbeaf);
+	while (n--) {
+		err = cpu_check(obj, n, 0xdeadbeaf);
+		if (err)
+			goto out_unpin;
+	}
+
 
 out_unpin:
 	i915_vma_unpin(vma);
@@ -1726,7 +1753,7 @@ out_unlock:
 	return err;
 }
 
-int i915_gem_huge_page_live_selftests(struct drm_i915_private *dev_priv)
+int i915_gem_huge_page_live_selftests(struct drm_i915_private *i915)
 {
 	static const struct i915_subtest tests[] = {
 		SUBTEST(igt_shrink_thp),
@@ -1741,22 +1768,22 @@ int i915_gem_huge_page_live_selftests(struct drm_i915_private *dev_priv)
 	intel_wakeref_t wakeref;
 	int err;
 
-	if (!HAS_PPGTT(dev_priv)) {
+	if (!HAS_PPGTT(i915)) {
 		pr_info("PPGTT not supported, skipping live-selftests\n");
 		return 0;
 	}
 
-	if (i915_terminally_wedged(dev_priv))
+	if (intel_gt_is_wedged(&i915->gt))
 		return 0;
 
-	file = mock_file(dev_priv);
+	file = mock_file(i915);
 	if (IS_ERR(file))
 		return PTR_ERR(file);
 
-	mutex_lock(&dev_priv->drm.struct_mutex);
-	wakeref = intel_runtime_pm_get(&dev_priv->runtime_pm);
+	mutex_lock(&i915->drm.struct_mutex);
+	wakeref = intel_runtime_pm_get(&i915->runtime_pm);
 
-	ctx = live_context(dev_priv, file);
+	ctx = live_context(i915, file);
 	if (IS_ERR(ctx)) {
 		err = PTR_ERR(ctx);
 		goto out_unlock;
@@ -1768,10 +1795,10 @@ int i915_gem_huge_page_live_selftests(struct drm_i915_private *dev_priv)
 	err = i915_subtests(tests, ctx);
 
 out_unlock:
-	intel_runtime_pm_put(&dev_priv->runtime_pm, wakeref);
-	mutex_unlock(&dev_priv->drm.struct_mutex);
+	intel_runtime_pm_put(&i915->runtime_pm, wakeref);
+	mutex_unlock(&i915->drm.struct_mutex);
 
-	mock_file_free(dev_priv, file);
+	mock_file_free(i915, file);
 
 	return err;
 }
diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c
index f3a5eb807c1c..275c28926067 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c
@@ -5,14 +5,16 @@
 
 #include "i915_selftest.h"
 
+#include "gt/intel_gt.h"
+
 #include "selftests/igt_flush_test.h"
 #include "selftests/mock_drm.h"
 #include "mock_context.h"
 
 static int igt_client_fill(void *arg)
 {
-	struct intel_context *ce = arg;
-	struct drm_i915_private *i915 = ce->gem_context->i915;
+	struct drm_i915_private *i915 = arg;
+	struct intel_context *ce = i915->engine[BCS0]->kernel_context;
 	struct drm_i915_gem_object *obj;
 	struct rnd_state prng;
 	IGT_TIMEOUT(end);
@@ -63,17 +65,6 @@ static int igt_client_fill(void *arg)
 		if (err)
 			goto err_unpin;
 
-		/*
-		 * XXX: For now do the wait without the object resv lock to
-		 * ensure we don't deadlock.
-		 */
-		err = i915_gem_object_wait(obj,
-					   I915_WAIT_INTERRUPTIBLE |
-					   I915_WAIT_ALL,
-					   MAX_SCHEDULE_TIMEOUT);
-		if (err)
-			goto err_unpin;
-
 		i915_gem_object_lock(obj);
 		err = i915_gem_object_set_to_cpu_domain(obj, false);
 		i915_gem_object_unlock(obj);
@@ -100,11 +91,6 @@ err_unpin:
 err_put:
 	i915_gem_object_put(obj);
 err_flush:
-	mutex_lock(&i915->drm.struct_mutex);
-	if (igt_flush_test(i915, I915_WAIT_LOCKED))
-		err = -EIO;
-	mutex_unlock(&i915->drm.struct_mutex);
-
 	if (err == -ENOMEM)
 		err = 0;
 
@@ -117,11 +103,11 @@ int i915_gem_client_blt_live_selftests(struct drm_i915_private *i915)
 		SUBTEST(igt_client_fill),
 	};
 
-	if (i915_terminally_wedged(i915))
+	if (intel_gt_is_wedged(&i915->gt))
 		return 0;
 
 	if (!HAS_ENGINE(i915, BCS0))
 		return 0;
 
-	return i915_subtests(tests, i915->engine[BCS0]->kernel_context);
+	return i915_live_subtests(tests, i915);
 }
diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_coherency.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_coherency.c
index 8f22d3f18422..a1a4b53cdc4a 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_coherency.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_coherency.c
@@ -6,6 +6,8 @@
 
 #include <linux/prime_numbers.h>
 
+#include "gt/intel_gt.h"
+
 #include "i915_selftest.h"
 #include "selftests/i915_random.h"
 
@@ -242,12 +244,15 @@ static bool always_valid(struct drm_i915_private *i915)
 
 static bool needs_fence_registers(struct drm_i915_private *i915)
 {
-	return !i915_terminally_wedged(i915);
+	return !intel_gt_is_wedged(&i915->gt);
 }
 
 static bool needs_mi_store_dword(struct drm_i915_private *i915)
 {
-	if (i915_terminally_wedged(i915))
+	if (intel_gt_is_wedged(&i915->gt))
+		return false;
+
+	if (!HAS_ENGINE(i915, RCS0))
 		return false;
 
 	return intel_engine_can_store_dword(i915->engine[RCS0]);
diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c
index eaa2b16574c7..7f9f6701b32c 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c
@@ -7,6 +7,7 @@
 #include <linux/prime_numbers.h>
 
 #include "gem/i915_gem_pm.h"
+#include "gt/intel_gt.h"
 #include "gt/intel_reset.h"
 #include "i915_selftest.h"
 
@@ -31,7 +32,6 @@ static int live_nop_switch(void *arg)
 	struct intel_engine_cs *engine;
 	struct i915_gem_context **ctx;
 	enum intel_engine_id id;
-	intel_wakeref_t wakeref;
 	struct igt_live_test t;
 	struct drm_file *file;
 	unsigned long n;
@@ -53,7 +53,6 @@ static int live_nop_switch(void *arg)
 		return PTR_ERR(file);
 
 	mutex_lock(&i915->drm.struct_mutex);
-	wakeref = intel_runtime_pm_get(&i915->runtime_pm);
 
 	ctx = kcalloc(nctx, sizeof(*ctx), GFP_KERNEL);
 	if (!ctx) {
@@ -85,7 +84,7 @@ static int live_nop_switch(void *arg)
 		}
 		if (i915_request_wait(rq, 0, HZ / 5) < 0) {
 			pr_err("Failed to populated %d contexts\n", nctx);
-			i915_gem_set_wedged(i915);
+			intel_gt_set_wedged(&i915->gt);
 			err = -EIO;
 			goto out_unlock;
 		}
@@ -129,7 +128,7 @@ static int live_nop_switch(void *arg)
 			if (i915_request_wait(rq, 0, HZ / 5) < 0) {
 				pr_err("Switching between %ld contexts timed out\n",
 				       prime);
-				i915_gem_set_wedged(i915);
+				intel_gt_set_wedged(&i915->gt);
 				break;
 			}
 
@@ -152,7 +151,6 @@ static int live_nop_switch(void *arg)
 	}
 
 out_unlock:
-	intel_runtime_pm_put(&i915->runtime_pm, wakeref);
 	mutex_unlock(&i915->drm.struct_mutex);
 	mock_file_free(i915, file);
 	return err;
@@ -237,8 +235,7 @@ static int gpu_fill(struct drm_i915_gem_object *obj,
 		    struct intel_engine_cs *engine,
 		    unsigned int dw)
 {
-	struct drm_i915_private *i915 = to_i915(obj->base.dev);
-	struct i915_address_space *vm = ctx->vm ?: &i915->ggtt.vm;
+	struct i915_address_space *vm = ctx->vm ?: &engine->gt->ggtt->vm;
 	struct i915_request *rq;
 	struct i915_vma *vma;
 	struct i915_vma *batch;
@@ -431,6 +428,9 @@ create_test_object(struct i915_gem_context *ctx,
 	u64 size;
 	int err;
 
+	/* Keep in GEM's good graces */
+	i915_retire_requests(ctx->i915);
+
 	size = min(vm->total / 2, 1024ull * DW_PER_PAGE * PAGE_SIZE);
 	size = round_down(size, DW_PER_PAGE * PAGE_SIZE);
 
@@ -507,7 +507,6 @@ static int igt_ctx_exec(void *arg)
 		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)) {
@@ -523,8 +522,7 @@ static int igt_ctx_exec(void *arg)
 				}
 			}
 
-			with_intel_runtime_pm(&i915->runtime_pm, wakeref)
-				err = gpu_fill(obj, ctx, engine, dw);
+			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),
@@ -565,6 +563,8 @@ out_unlock:
 		mock_file_free(i915, file);
 		if (err)
 			return err;
+
+		i915_gem_drain_freed_objects(i915);
 	}
 
 	return 0;
@@ -623,7 +623,6 @@ static int igt_shared_ctx_exec(void *arg)
 		ncontexts = 0;
 		while (!time_after(jiffies, end_time)) {
 			struct i915_gem_context *ctx;
-			intel_wakeref_t wakeref;
 
 			ctx = kernel_context(i915);
 			if (IS_ERR(ctx)) {
@@ -642,9 +641,7 @@ static int igt_shared_ctx_exec(void *arg)
 				}
 			}
 
-			err = 0;
-			with_intel_runtime_pm(&i915->runtime_pm, wakeref)
-				err = gpu_fill(obj, ctx, engine, dw);
+			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),
@@ -678,6 +675,10 @@ static int igt_shared_ctx_exec(void *arg)
 
 			dw += rem;
 		}
+
+		mutex_unlock(&i915->drm.struct_mutex);
+		i915_gem_drain_freed_objects(i915);
+		mutex_lock(&i915->drm.struct_mutex);
 	}
 out_test:
 	if (igt_live_test_end(&t))
@@ -746,7 +747,7 @@ emit_rpcs_query(struct drm_i915_gem_object *obj,
 
 	GEM_BUG_ON(!intel_engine_can_store_dword(ce->engine));
 
-	vma = i915_vma_instance(obj, ce->gem_context->vm, NULL);
+	vma = i915_vma_instance(obj, ce->vm, NULL);
 	if (IS_ERR(vma))
 		return PTR_ERR(vma);
 
@@ -956,7 +957,7 @@ __sseu_finish(struct drm_i915_private *i915,
 	int ret = 0;
 
 	if (flags & TEST_RESET) {
-		ret = i915_reset_engine(ce->engine, "sseu");
+		ret = intel_engine_reset(ce->engine, "sseu");
 		if (ret)
 			goto out;
 	}
@@ -1025,35 +1026,33 @@ __igt_ctx_sseu(struct drm_i915_private *i915,
 	       unsigned int flags)
 {
 	struct intel_engine_cs *engine = i915->engine[RCS0];
-	struct intel_sseu default_sseu = engine->sseu;
 	struct drm_i915_gem_object *obj;
 	struct i915_gem_context *ctx;
 	struct intel_context *ce;
 	struct intel_sseu pg_sseu;
-	intel_wakeref_t wakeref;
 	struct drm_file *file;
 	int ret;
 
-	if (INTEL_GEN(i915) < 9)
+	if (INTEL_GEN(i915) < 9 || !engine)
 		return 0;
 
 	if (!RUNTIME_INFO(i915)->sseu.has_slice_pg)
 		return 0;
 
-	if (hweight32(default_sseu.slice_mask) < 2)
+	if (hweight32(engine->sseu.slice_mask) < 2)
 		return 0;
 
 	/*
 	 * Gen11 VME friendly power-gated configuration with half enabled
 	 * sub-slices.
 	 */
-	pg_sseu = default_sseu;
+	pg_sseu = engine->sseu;
 	pg_sseu.slice_mask = 1;
 	pg_sseu.subslice_mask =
-		~(~0 << (hweight32(default_sseu.subslice_mask) / 2));
+		~(~0 << (hweight32(engine->sseu.subslice_mask) / 2));
 
 	pr_info("SSEU subtest '%s', flags=%x, def_slices=%u, pg_slices=%u\n",
-		name, flags, hweight32(default_sseu.slice_mask),
+		name, flags, hweight32(engine->sseu.slice_mask),
 		hweight32(pg_sseu.slice_mask));
 
 	file = mock_file(i915);
@@ -1061,7 +1060,7 @@ __igt_ctx_sseu(struct drm_i915_private *i915,
 		return PTR_ERR(file);
 
 	if (flags & TEST_RESET)
-		igt_global_reset_lock(i915);
+		igt_global_reset_lock(&i915->gt);
 
 	mutex_lock(&i915->drm.struct_mutex);
 
@@ -1078,12 +1077,10 @@ __igt_ctx_sseu(struct drm_i915_private *i915,
 		goto out_unlock;
 	}
 
-	wakeref = intel_runtime_pm_get(&i915->runtime_pm);
-
 	ce = i915_gem_context_get_engine(ctx, RCS0);
 	if (IS_ERR(ce)) {
 		ret = PTR_ERR(ce);
-		goto out_rpm;
+		goto out_put;
 	}
 
 	ret = intel_context_pin(ce);
@@ -1091,7 +1088,7 @@ __igt_ctx_sseu(struct drm_i915_private *i915,
 		goto out_context;
 
 	/* First set the default mask. */
-	ret = __sseu_test(i915, name, flags, ce, obj, default_sseu);
+	ret = __sseu_test(i915, name, flags, ce, obj, engine->sseu);
 	if (ret)
 		goto out_fail;
 
@@ -1101,7 +1098,7 @@ __igt_ctx_sseu(struct drm_i915_private *i915,
 		goto out_fail;
 
 	/* Back to defaults. */
-	ret = __sseu_test(i915, name, flags, ce, obj, default_sseu);
+	ret = __sseu_test(i915, name, flags, ce, obj, engine->sseu);
 	if (ret)
 		goto out_fail;
 
@@ -1117,15 +1114,14 @@ out_fail:
 	intel_context_unpin(ce);
 out_context:
 	intel_context_put(ce);
-out_rpm:
-	intel_runtime_pm_put(&i915->runtime_pm, wakeref);
+out_put:
 	i915_gem_object_put(obj);
 
 out_unlock:
 	mutex_unlock(&i915->drm.struct_mutex);
 
 	if (flags & TEST_RESET)
-		igt_global_reset_unlock(i915);
+		igt_global_reset_unlock(&i915->gt);
 
 	mock_file_free(i915, file);
 
@@ -1194,7 +1190,7 @@ static int igt_ctx_readonly(void *arg)
 		goto out_unlock;
 	}
 
-	vm = ctx->vm ?: &i915->mm.aliasing_ppgtt->vm;
+	vm = ctx->vm ?: &i915->ggtt.alias->vm;
 	if (!vm || !vm->has_read_only) {
 		err = 0;
 		goto out_unlock;
@@ -1207,8 +1203,6 @@ static int igt_ctx_readonly(void *arg)
 		unsigned int id;
 
 		for_each_engine(engine, i915, id) {
-			intel_wakeref_t wakeref;
-
 			if (!intel_engine_can_store_dword(engine))
 				continue;
 
@@ -1223,9 +1217,7 @@ static int igt_ctx_readonly(void *arg)
 					i915_gem_object_set_readonly(obj);
 			}
 
-			err = 0;
-			with_intel_runtime_pm(&i915->runtime_pm, wakeref)
-				err = gpu_fill(obj, ctx, engine, dw);
+			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),
@@ -1488,7 +1480,6 @@ static int igt_vm_isolation(void *arg)
 	struct drm_i915_private *i915 = arg;
 	struct i915_gem_context *ctx_a, *ctx_b;
 	struct intel_engine_cs *engine;
-	intel_wakeref_t wakeref;
 	struct igt_live_test t;
 	struct drm_file *file;
 	I915_RND_STATE(prng);
@@ -1535,8 +1526,6 @@ static int igt_vm_isolation(void *arg)
 	GEM_BUG_ON(ctx_b->vm->total != vm_total);
 	vm_total -= I915_GTT_PAGE_SIZE;
 
-	wakeref = intel_runtime_pm_get(&i915->runtime_pm);
-
 	count = 0;
 	for_each_engine(engine, i915, id) {
 		IGT_TIMEOUT(end_time);
@@ -1551,7 +1540,7 @@ static int igt_vm_isolation(void *arg)
 
 			div64_u64_rem(i915_prandom_u64_state(&prng),
 				      vm_total, &offset);
-			offset &= -sizeof(u32);
+			offset = round_down(offset, alignof_dword);
 			offset += I915_GTT_PAGE_SIZE;
 
 			err = write_to_scratch(ctx_a, engine,
@@ -1560,7 +1549,7 @@ static int igt_vm_isolation(void *arg)
 				err = read_from_scratch(ctx_b, engine,
 							offset, &value);
 			if (err)
-				goto out_rpm;
+				goto out_unlock;
 
 			if (value) {
 				pr_err("%s: Read %08x from scratch (offset 0x%08x_%08x), after %lu reads!\n",
@@ -1569,7 +1558,7 @@ static int igt_vm_isolation(void *arg)
 				       lower_32_bits(offset),
 				       this);
 				err = -EINVAL;
-				goto out_rpm;
+				goto out_unlock;
 			}
 
 			this++;
@@ -1579,8 +1568,6 @@ static int igt_vm_isolation(void *arg)
 	pr_info("Checked %lu scratch offsets across %d engines\n",
 		count, RUNTIME_INFO(i915)->num_engines);
 
-out_rpm:
-	intel_runtime_pm_put(&i915->runtime_pm, wakeref);
 out_unlock:
 	if (igt_live_test_end(&t))
 		err = -EIO;
@@ -1736,7 +1723,7 @@ int i915_gem_context_mock_selftests(void)
 	return err;
 }
 
-int i915_gem_context_live_selftests(struct drm_i915_private *dev_priv)
+int i915_gem_context_live_selftests(struct drm_i915_private *i915)
 {
 	static const struct i915_subtest tests[] = {
 		SUBTEST(live_nop_switch),
@@ -1747,8 +1734,8 @@ int i915_gem_context_live_selftests(struct drm_i915_private *dev_priv)
 		SUBTEST(igt_vm_isolation),
 	};
 
-	if (i915_terminally_wedged(dev_priv))
+	if (intel_gt_is_wedged(&i915->gt))
 		return 0;
 
-	return i915_subtests(tests, dev_priv);
+	return i915_live_subtests(tests, i915);
 }
diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c
index 5c81f4b4813a..01857c12f12f 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c
@@ -6,6 +6,7 @@
 
 #include <linux/prime_numbers.h>
 
+#include "gt/intel_gt.h"
 #include "gt/intel_gt_pm.h"
 #include "huge_gem_object.h"
 #include "i915_selftest.h"
@@ -143,7 +144,7 @@ static int check_partial_mapping(struct drm_i915_gem_object *obj,
 		if (offset >= obj->base.size)
 			continue;
 
-		i915_gem_flush_ggtt_writes(to_i915(obj->base.dev));
+		intel_gt_flush_ggtt_writes(&to_i915(obj->base.dev)->gt);
 
 		p = i915_gem_object_get_page(obj, offset >> PAGE_SHIFT);
 		cpu = kmap(p) + offset_in_page(offset);
@@ -327,7 +328,8 @@ out:
 static int make_obj_busy(struct drm_i915_gem_object *obj)
 {
 	struct drm_i915_private *i915 = to_i915(obj->base.dev);
-	struct i915_request *rq;
+	struct intel_engine_cs *engine;
+	enum intel_engine_id id;
 	struct i915_vma *vma;
 	int err;
 
@@ -339,17 +341,21 @@ static int make_obj_busy(struct drm_i915_gem_object *obj)
 	if (err)
 		return err;
 
-	rq = i915_request_create(i915->engine[RCS0]->kernel_context);
-	if (IS_ERR(rq)) {
-		i915_vma_unpin(vma);
-		return PTR_ERR(rq);
-	}
+	for_each_engine(engine, i915, id) {
+		struct i915_request *rq;
+
+		rq = i915_request_create(engine->kernel_context);
+		if (IS_ERR(rq)) {
+			i915_vma_unpin(vma);
+			return PTR_ERR(rq);
+		}
 
-	i915_vma_lock(vma);
-	err = i915_vma_move_to_active(vma, rq, EXEC_OBJECT_WRITE);
-	i915_vma_unlock(vma);
+		i915_vma_lock(vma);
+		err = i915_vma_move_to_active(vma, rq, EXEC_OBJECT_WRITE);
+		i915_vma_unlock(vma);
 
-	i915_request_add(rq);
+		i915_request_add(rq);
+	}
 
 	i915_vma_unpin(vma);
 	i915_gem_object_put(obj); /* leave it only alive via its active ref */
@@ -378,7 +384,7 @@ static void disable_retire_worker(struct drm_i915_private *i915)
 {
 	i915_gem_shrinker_unregister(i915);
 
-	intel_gt_pm_get(i915);
+	intel_gt_pm_get(&i915->gt);
 
 	cancel_delayed_work_sync(&i915->gem.retire_work);
 	flush_work(&i915->gem.idle_work);
@@ -386,7 +392,7 @@ static void disable_retire_worker(struct drm_i915_private *i915)
 
 static void restore_retire_worker(struct drm_i915_private *i915)
 {
-	intel_gt_pm_put(i915);
+	intel_gt_pm_put(&i915->gt);
 
 	mutex_lock(&i915->drm.struct_mutex);
 	igt_flush_test(i915, I915_WAIT_LOCKED);
@@ -395,6 +401,18 @@ static void restore_retire_worker(struct drm_i915_private *i915)
 	i915_gem_shrinker_register(i915);
 }
 
+static void mmap_offset_lock(struct drm_i915_private *i915)
+	__acquires(&i915->drm.vma_offset_manager->vm_lock)
+{
+	write_lock(&i915->drm.vma_offset_manager->vm_lock);
+}
+
+static void mmap_offset_unlock(struct drm_i915_private *i915)
+	__releases(&i915->drm.vma_offset_manager->vm_lock)
+{
+	write_unlock(&i915->drm.vma_offset_manager->vm_lock);
+}
+
 static int igt_mmap_offset_exhaustion(void *arg)
 {
 	struct drm_i915_private *i915 = arg;
@@ -413,7 +431,9 @@ static int igt_mmap_offset_exhaustion(void *arg)
 	drm_mm_for_each_hole(hole, mm, hole_start, hole_end) {
 		resv.start = hole_start;
 		resv.size = hole_end - hole_start - 1; /* PAGE_SIZE units */
+		mmap_offset_lock(i915);
 		err = drm_mm_reserve_node(mm, &resv);
+		mmap_offset_unlock(i915);
 		if (err) {
 			pr_err("Failed to trim VMA manager, err=%d\n", err);
 			goto out_park;
@@ -458,7 +478,7 @@ static int igt_mmap_offset_exhaustion(void *arg)
 
 	/* Now fill with busy dead objects that we expect to reap */
 	for (loop = 0; loop < 3; loop++) {
-		if (i915_terminally_wedged(i915))
+		if (intel_gt_is_wedged(&i915->gt))
 			break;
 
 		obj = i915_gem_object_create_internal(i915, PAGE_SIZE);
@@ -474,19 +494,12 @@ static int igt_mmap_offset_exhaustion(void *arg)
 			pr_err("[loop %d] Failed to busy the object\n", loop);
 			goto err_obj;
 		}
-
-		/* NB we rely on the _active_ reference to access obj now */
-		GEM_BUG_ON(!i915_gem_object_is_active(obj));
-		err = create_mmap_offset(obj);
-		if (err) {
-			pr_err("[loop %d] create_mmap_offset failed with err=%d\n",
-			       loop, err);
-			goto out;
-		}
 	}
 
 out:
+	mmap_offset_lock(i915);
 	drm_mm_remove_node(&resv);
+	mmap_offset_unlock(i915);
 out_park:
 	restore_retire_worker(i915);
 	return err;
diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_object_blt.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_object_blt.c
index e23d8c9e9298..19843acc84d3 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_object_blt.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_object_blt.c
@@ -3,6 +3,8 @@
  * Copyright © 2019 Intel Corporation
  */
 
+#include "gt/intel_gt.h"
+
 #include "i915_selftest.h"
 
 #include "selftests/igt_flush_test.h"
@@ -11,8 +13,8 @@
 
 static int igt_fill_blt(void *arg)
 {
-	struct intel_context *ce = arg;
-	struct drm_i915_private *i915 = ce->gem_context->i915;
+	struct drm_i915_private *i915 = arg;
+	struct intel_context *ce = i915->engine[BCS0]->kernel_context;
 	struct drm_i915_gem_object *obj;
 	struct rnd_state prng;
 	IGT_TIMEOUT(end);
@@ -83,11 +85,6 @@ err_unpin:
 err_put:
 	i915_gem_object_put(obj);
 err_flush:
-	mutex_lock(&i915->drm.struct_mutex);
-	if (igt_flush_test(i915, I915_WAIT_LOCKED))
-		err = -EIO;
-	mutex_unlock(&i915->drm.struct_mutex);
-
 	if (err == -ENOMEM)
 		err = 0;
 
@@ -100,11 +97,11 @@ int i915_gem_object_blt_live_selftests(struct drm_i915_private *i915)
 		SUBTEST(igt_fill_blt),
 	};
 
-	if (i915_terminally_wedged(i915))
+	if (intel_gt_is_wedged(&i915->gt))
 		return 0;
 
 	if (!HAS_ENGINE(i915, BCS0))
 		return 0;
 
-	return i915_subtests(tests, i915->engine[BCS0]->kernel_context);
+	return i915_live_subtests(tests, i915);
 }