diff --git a/drivers/gpu/drm/i915/gvt/cmd_parser.c b/drivers/gpu/drm/i915/gvt/cmd_parser.c
index db6b94dda5df..d85939bd7b47 100644
--- a/drivers/gpu/drm/i915/gvt/cmd_parser.c
+++ b/drivers/gpu/drm/i915/gvt/cmd_parser.c
@@ -1080,6 +1080,7 @@ static int cmd_handler_mi_user_interrupt(struct parser_exec_state *s)
 {
 	set_bit(cmd_interrupt_events[s->ring_id].mi_user_interrupt,
 			s->workload->pending_events);
+	patch_value(s, cmd_ptr(s, 0), MI_NOOP);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/i915/gvt/display.c b/drivers/gpu/drm/i915/gvt/display.c
index dd96ffc878ac..6d8180e8d1e2 100644
--- a/drivers/gpu/drm/i915/gvt/display.c
+++ b/drivers/gpu/drm/i915/gvt/display.c
@@ -169,6 +169,8 @@ static u8 dpcd_fix_data[DPCD_HEADER_SIZE] = {
 static void emulate_monitor_status_change(struct intel_vgpu *vgpu)
 {
 	struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
+	int pipe;
+
 	vgpu_vreg_t(vgpu, SDEISR) &= ~(SDE_PORTB_HOTPLUG_CPT |
 			SDE_PORTC_HOTPLUG_CPT |
 			SDE_PORTD_HOTPLUG_CPT);
@@ -267,6 +269,14 @@ static void emulate_monitor_status_change(struct intel_vgpu *vgpu)
 	if (IS_BROADWELL(dev_priv))
 		vgpu_vreg_t(vgpu, PCH_ADPA) &= ~ADPA_CRT_HOTPLUG_MONITOR_MASK;
 
+	/* Disable Primary/Sprite/Cursor plane */
+	for_each_pipe(dev_priv, pipe) {
+		vgpu_vreg_t(vgpu, DSPCNTR(pipe)) &= ~DISPLAY_PLANE_ENABLE;
+		vgpu_vreg_t(vgpu, SPRCTL(pipe)) &= ~SPRITE_ENABLE;
+		vgpu_vreg_t(vgpu, CURCNTR(pipe)) &= ~CURSOR_MODE;
+		vgpu_vreg_t(vgpu, CURCNTR(pipe)) |= CURSOR_MODE_DISABLE;
+	}
+
 	vgpu_vreg_t(vgpu, PIPECONF(PIPE_A)) |= PIPECONF_ENABLE;
 }
 
diff --git a/drivers/gpu/drm/i915/gvt/dmabuf.c b/drivers/gpu/drm/i915/gvt/dmabuf.c
index b555eb26f9ce..6f4f8e941fc2 100644
--- a/drivers/gpu/drm/i915/gvt/dmabuf.c
+++ b/drivers/gpu/drm/i915/gvt/dmabuf.c
@@ -323,6 +323,7 @@ static void update_fb_info(struct vfio_device_gfx_plane_info *gvt_dmabuf,
 		      struct intel_vgpu_fb_info *fb_info)
 {
 	gvt_dmabuf->drm_format = fb_info->drm_format;
+	gvt_dmabuf->drm_format_mod = fb_info->drm_format_mod;
 	gvt_dmabuf->width = fb_info->width;
 	gvt_dmabuf->height = fb_info->height;
 	gvt_dmabuf->stride = fb_info->stride;
diff --git a/drivers/gpu/drm/i915/gvt/fb_decoder.c b/drivers/gpu/drm/i915/gvt/fb_decoder.c
index 6b50fe78dc1b..1c120683e958 100644
--- a/drivers/gpu/drm/i915/gvt/fb_decoder.c
+++ b/drivers/gpu/drm/i915/gvt/fb_decoder.c
@@ -245,16 +245,13 @@ int intel_vgpu_decode_primary_plane(struct intel_vgpu *vgpu,
 	plane->hw_format = fmt;
 
 	plane->base = vgpu_vreg_t(vgpu, DSPSURF(pipe)) & I915_GTT_PAGE_MASK;
-	if (!intel_gvt_ggtt_validate_range(vgpu, plane->base, 0)) {
-		gvt_vgpu_err("invalid gma address: %lx\n",
-			     (unsigned long)plane->base);
+	if (!intel_gvt_ggtt_validate_range(vgpu, plane->base, 0))
 		return  -EINVAL;
-	}
 
 	plane->base_gpa = intel_vgpu_gma_to_gpa(vgpu->gtt.ggtt_mm, plane->base);
 	if (plane->base_gpa == INTEL_GVT_INVALID_ADDR) {
-		gvt_vgpu_err("invalid gma address: %lx\n",
-				(unsigned long)plane->base);
+		gvt_vgpu_err("Translate primary plane gma 0x%x to gpa fail\n",
+				plane->base);
 		return  -EINVAL;
 	}
 
@@ -371,16 +368,13 @@ int intel_vgpu_decode_cursor_plane(struct intel_vgpu *vgpu,
 			alpha_plane, alpha_force);
 
 	plane->base = vgpu_vreg_t(vgpu, CURBASE(pipe)) & I915_GTT_PAGE_MASK;
-	if (!intel_gvt_ggtt_validate_range(vgpu, plane->base, 0)) {
-		gvt_vgpu_err("invalid gma address: %lx\n",
-			     (unsigned long)plane->base);
+	if (!intel_gvt_ggtt_validate_range(vgpu, plane->base, 0))
 		return  -EINVAL;
-	}
 
 	plane->base_gpa = intel_vgpu_gma_to_gpa(vgpu->gtt.ggtt_mm, plane->base);
 	if (plane->base_gpa == INTEL_GVT_INVALID_ADDR) {
-		gvt_vgpu_err("invalid gma address: %lx\n",
-				(unsigned long)plane->base);
+		gvt_vgpu_err("Translate cursor plane gma 0x%x to gpa fail\n",
+				plane->base);
 		return  -EINVAL;
 	}
 
@@ -476,16 +470,13 @@ int intel_vgpu_decode_sprite_plane(struct intel_vgpu *vgpu,
 	plane->drm_format = drm_format;
 
 	plane->base = vgpu_vreg_t(vgpu, SPRSURF(pipe)) & I915_GTT_PAGE_MASK;
-	if (!intel_gvt_ggtt_validate_range(vgpu, plane->base, 0)) {
-		gvt_vgpu_err("invalid gma address: %lx\n",
-			     (unsigned long)plane->base);
+	if (!intel_gvt_ggtt_validate_range(vgpu, plane->base, 0))
 		return  -EINVAL;
-	}
 
 	plane->base_gpa = intel_vgpu_gma_to_gpa(vgpu->gtt.ggtt_mm, plane->base);
 	if (plane->base_gpa == INTEL_GVT_INVALID_ADDR) {
-		gvt_vgpu_err("invalid gma address: %lx\n",
-				(unsigned long)plane->base);
+		gvt_vgpu_err("Translate sprite plane gma 0x%x to gpa fail\n",
+				plane->base);
 		return  -EINVAL;
 	}
 
diff --git a/drivers/gpu/drm/i915/gvt/gtt.c b/drivers/gpu/drm/i915/gvt/gtt.c
index d29281231507..78e55aafc8bc 100644
--- a/drivers/gpu/drm/i915/gvt/gtt.c
+++ b/drivers/gpu/drm/i915/gvt/gtt.c
@@ -530,6 +530,16 @@ static void ggtt_set_guest_entry(struct intel_vgpu_mm *mm,
 			   false, 0, mm->vgpu);
 }
 
+static void ggtt_get_host_entry(struct intel_vgpu_mm *mm,
+		struct intel_gvt_gtt_entry *entry, unsigned long index)
+{
+	struct intel_gvt_gtt_pte_ops *pte_ops = mm->vgpu->gvt->gtt.pte_ops;
+
+	GEM_BUG_ON(mm->type != INTEL_GVT_MM_GGTT);
+
+	pte_ops->get_entry(NULL, entry, index, false, 0, mm->vgpu);
+}
+
 static void ggtt_set_host_entry(struct intel_vgpu_mm *mm,
 		struct intel_gvt_gtt_entry *entry, unsigned long index)
 {
@@ -1818,6 +1828,18 @@ int intel_vgpu_emulate_ggtt_mmio_read(struct intel_vgpu *vgpu, unsigned int off,
 	return ret;
 }
 
+static void ggtt_invalidate_pte(struct intel_vgpu *vgpu,
+		struct intel_gvt_gtt_entry *entry)
+{
+	struct intel_gvt_gtt_pte_ops *pte_ops = vgpu->gvt->gtt.pte_ops;
+	unsigned long pfn;
+
+	pfn = pte_ops->get_pfn(entry);
+	if (pfn != vgpu->gvt->gtt.scratch_mfn)
+		intel_gvt_hypervisor_dma_unmap_guest_page(vgpu,
+						pfn << PAGE_SHIFT);
+}
+
 static int emulate_ggtt_mmio_write(struct intel_vgpu *vgpu, unsigned int off,
 	void *p_data, unsigned int bytes)
 {
@@ -1844,10 +1866,10 @@ static int emulate_ggtt_mmio_write(struct intel_vgpu *vgpu, unsigned int off,
 
 	memcpy((void *)&e.val64 + (off & (info->gtt_entry_size - 1)), p_data,
 			bytes);
-	m = e;
 
 	if (ops->test_present(&e)) {
 		gfn = ops->get_pfn(&e);
+		m = e;
 
 		/* one PTE update may be issued in multiple writes and the
 		 * first write may not construct a valid gfn
@@ -1868,8 +1890,12 @@ static int emulate_ggtt_mmio_write(struct intel_vgpu *vgpu, unsigned int off,
 			ops->set_pfn(&m, gvt->gtt.scratch_mfn);
 		} else
 			ops->set_pfn(&m, dma_addr >> PAGE_SHIFT);
-	} else
+	} else {
+		ggtt_get_host_entry(ggtt_mm, &m, g_gtt_index);
+		ggtt_invalidate_pte(vgpu, &m);
 		ops->set_pfn(&m, gvt->gtt.scratch_mfn);
+		ops->clear_present(&m);
+	}
 
 out:
 	ggtt_set_host_entry(ggtt_mm, &m, g_gtt_index);
@@ -2030,7 +2056,7 @@ int intel_vgpu_init_gtt(struct intel_vgpu *vgpu)
 		return PTR_ERR(gtt->ggtt_mm);
 	}
 
-	intel_vgpu_reset_ggtt(vgpu);
+	intel_vgpu_reset_ggtt(vgpu, false);
 
 	return create_scratch_page_tree(vgpu);
 }
@@ -2315,17 +2341,19 @@ void intel_vgpu_invalidate_ppgtt(struct intel_vgpu *vgpu)
 /**
  * intel_vgpu_reset_ggtt - reset the GGTT entry
  * @vgpu: a vGPU
+ * @invalidate_old: invalidate old entries
  *
  * This function is called at the vGPU create stage
  * to reset all the GGTT entries.
  *
  */
-void intel_vgpu_reset_ggtt(struct intel_vgpu *vgpu)
+void intel_vgpu_reset_ggtt(struct intel_vgpu *vgpu, bool invalidate_old)
 {
 	struct intel_gvt *gvt = vgpu->gvt;
 	struct drm_i915_private *dev_priv = gvt->dev_priv;
 	struct intel_gvt_gtt_pte_ops *pte_ops = vgpu->gvt->gtt.pte_ops;
 	struct intel_gvt_gtt_entry entry = {.type = GTT_TYPE_GGTT_PTE};
+	struct intel_gvt_gtt_entry old_entry;
 	u32 index;
 	u32 num_entries;
 
@@ -2334,13 +2362,23 @@ void intel_vgpu_reset_ggtt(struct intel_vgpu *vgpu)
 
 	index = vgpu_aperture_gmadr_base(vgpu) >> PAGE_SHIFT;
 	num_entries = vgpu_aperture_sz(vgpu) >> PAGE_SHIFT;
-	while (num_entries--)
+	while (num_entries--) {
+		if (invalidate_old) {
+			ggtt_get_host_entry(vgpu->gtt.ggtt_mm, &old_entry, index);
+			ggtt_invalidate_pte(vgpu, &old_entry);
+		}
 		ggtt_set_host_entry(vgpu->gtt.ggtt_mm, &entry, index++);
+	}
 
 	index = vgpu_hidden_gmadr_base(vgpu) >> PAGE_SHIFT;
 	num_entries = vgpu_hidden_sz(vgpu) >> PAGE_SHIFT;
-	while (num_entries--)
+	while (num_entries--) {
+		if (invalidate_old) {
+			ggtt_get_host_entry(vgpu->gtt.ggtt_mm, &old_entry, index);
+			ggtt_invalidate_pte(vgpu, &old_entry);
+		}
 		ggtt_set_host_entry(vgpu->gtt.ggtt_mm, &entry, index++);
+	}
 
 	ggtt_invalidate(dev_priv);
 }
@@ -2360,5 +2398,5 @@ void intel_vgpu_reset_gtt(struct intel_vgpu *vgpu)
 	 * removing the shadow pages.
 	 */
 	intel_vgpu_destroy_all_ppgtt_mm(vgpu);
-	intel_vgpu_reset_ggtt(vgpu);
+	intel_vgpu_reset_ggtt(vgpu, true);
 }
diff --git a/drivers/gpu/drm/i915/gvt/gtt.h b/drivers/gpu/drm/i915/gvt/gtt.h
index a8b369cd352b..3792f2b7f4ff 100644
--- a/drivers/gpu/drm/i915/gvt/gtt.h
+++ b/drivers/gpu/drm/i915/gvt/gtt.h
@@ -193,7 +193,7 @@ struct intel_vgpu_gtt {
 
 extern int intel_vgpu_init_gtt(struct intel_vgpu *vgpu);
 extern void intel_vgpu_clean_gtt(struct intel_vgpu *vgpu);
-void intel_vgpu_reset_ggtt(struct intel_vgpu *vgpu);
+void intel_vgpu_reset_ggtt(struct intel_vgpu *vgpu, bool invalidate_old);
 void intel_vgpu_invalidate_ppgtt(struct intel_vgpu *vgpu);
 
 extern int intel_gvt_init_gtt(struct intel_gvt *gvt);
diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c
index 8c5d5d005854..a33c1c3e4a21 100644
--- a/drivers/gpu/drm/i915/gvt/handlers.c
+++ b/drivers/gpu/drm/i915/gvt/handlers.c
@@ -1150,6 +1150,7 @@ static int handle_g2v_notification(struct intel_vgpu *vgpu, int notification)
 	switch (notification) {
 	case VGT_G2V_PPGTT_L3_PAGE_TABLE_CREATE:
 		root_entry_type = GTT_TYPE_PPGTT_ROOT_L3_ENTRY;
+		/* fall through */
 	case VGT_G2V_PPGTT_L4_PAGE_TABLE_CREATE:
 		mm = intel_vgpu_get_ppgtt_mm(vgpu, root_entry_type, pdps);
 		return PTR_ERR_OR_ZERO(mm);
diff --git a/drivers/gpu/drm/i915/gvt/kvmgt.c b/drivers/gpu/drm/i915/gvt/kvmgt.c
index c16a492449d7..1466d8769ec9 100644
--- a/drivers/gpu/drm/i915/gvt/kvmgt.c
+++ b/drivers/gpu/drm/i915/gvt/kvmgt.c
@@ -1301,7 +1301,7 @@ static long intel_vgpu_ioctl(struct mdev_device *mdev, unsigned int cmd,
 
 	}
 
-	return 0;
+	return -ENOTTY;
 }
 
 static ssize_t
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 07c07d55398b..be8555049c93 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -1102,30 +1102,32 @@ static int i915_driver_init_hw(struct drm_i915_private *dev_priv)
 
 	ret = i915_ggtt_probe_hw(dev_priv);
 	if (ret)
-		return ret;
+		goto err_perf;
 
-	/* WARNING: Apparently we must kick fbdev drivers before vgacon,
-	 * otherwise the vga fbdev driver falls over. */
+	/*
+	 * WARNING: Apparently we must kick fbdev drivers before vgacon,
+	 * otherwise the vga fbdev driver falls over.
+	 */
 	ret = i915_kick_out_firmware_fb(dev_priv);
 	if (ret) {
 		DRM_ERROR("failed to remove conflicting framebuffer drivers\n");
-		goto out_ggtt;
+		goto err_ggtt;
 	}
 
 	ret = i915_kick_out_vgacon(dev_priv);
 	if (ret) {
 		DRM_ERROR("failed to remove conflicting VGA console\n");
-		goto out_ggtt;
+		goto err_ggtt;
 	}
 
 	ret = i915_ggtt_init_hw(dev_priv);
 	if (ret)
-		return ret;
+		goto err_ggtt;
 
 	ret = i915_ggtt_enable_hw(dev_priv);
 	if (ret) {
 		DRM_ERROR("failed to enable GGTT\n");
-		goto out_ggtt;
+		goto err_ggtt;
 	}
 
 	pci_set_master(pdev);
@@ -1136,7 +1138,7 @@ static int i915_driver_init_hw(struct drm_i915_private *dev_priv)
 		if (ret) {
 			DRM_ERROR("failed to set DMA mask\n");
 
-			goto out_ggtt;
+			goto err_ggtt;
 		}
 	}
 
@@ -1154,7 +1156,7 @@ static int i915_driver_init_hw(struct drm_i915_private *dev_priv)
 		if (ret) {
 			DRM_ERROR("failed to set DMA mask\n");
 
-			goto out_ggtt;
+			goto err_ggtt;
 		}
 	}
 
@@ -1187,13 +1189,14 @@ static int i915_driver_init_hw(struct drm_i915_private *dev_priv)
 
 	ret = intel_gvt_init(dev_priv);
 	if (ret)
-		goto out_ggtt;
+		goto err_ggtt;
 
 	return 0;
 
-out_ggtt:
+err_ggtt:
 	i915_ggtt_cleanup_hw(dev_priv);
-
+err_perf:
+	i915_perf_fini(dev_priv);
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 8c170db8495d..0414228cd2b5 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -728,7 +728,7 @@ static int eb_lookup_vmas(struct i915_execbuffer *eb)
 
 		err = radix_tree_insert(handles_vma, handle, vma);
 		if (unlikely(err)) {
-			kfree(lut);
+			kmem_cache_free(eb->i915->luts, lut);
 			goto err_obj;
 		}
 
diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c
index d8feb9053e0c..f0519e31543a 100644
--- a/drivers/gpu/drm/i915/i915_pmu.c
+++ b/drivers/gpu/drm/i915/i915_pmu.c
@@ -473,20 +473,37 @@ static u64 get_rc6(struct drm_i915_private *i915)
 		spin_lock_irqsave(&i915->pmu.lock, flags);
 		spin_lock(&kdev->power.lock);
 
-		if (!i915->pmu.sample[__I915_SAMPLE_RC6_ESTIMATED].cur)
-			i915->pmu.suspended_jiffies_last =
-						kdev->power.suspended_jiffies;
+		/*
+		 * After the above branch intel_runtime_pm_get_if_in_use failed
+		 * to get the runtime PM reference we cannot assume we are in
+		 * runtime suspend since we can either: a) race with coming out
+		 * of it before we took the power.lock, or b) there are other
+		 * states than suspended which can bring us here.
+		 *
+		 * We need to double-check that we are indeed currently runtime
+		 * suspended and if not we cannot do better than report the last
+		 * known RC6 value.
+		 */
+		if (kdev->power.runtime_status == RPM_SUSPENDED) {
+			if (!i915->pmu.sample[__I915_SAMPLE_RC6_ESTIMATED].cur)
+				i915->pmu.suspended_jiffies_last =
+						  kdev->power.suspended_jiffies;
 
-		val = kdev->power.suspended_jiffies -
-		      i915->pmu.suspended_jiffies_last;
-		val += jiffies - kdev->power.accounting_timestamp;
+			val = kdev->power.suspended_jiffies -
+			      i915->pmu.suspended_jiffies_last;
+			val += jiffies - kdev->power.accounting_timestamp;
+
+			val = jiffies_to_nsecs(val);
+			val += i915->pmu.sample[__I915_SAMPLE_RC6].cur;
+
+			i915->pmu.sample[__I915_SAMPLE_RC6_ESTIMATED].cur = val;
+		} else if (i915->pmu.sample[__I915_SAMPLE_RC6_ESTIMATED].cur) {
+			val = i915->pmu.sample[__I915_SAMPLE_RC6_ESTIMATED].cur;
+		} else {
+			val = i915->pmu.sample[__I915_SAMPLE_RC6].cur;
+		}
 
 		spin_unlock(&kdev->power.lock);
-
-		val = jiffies_to_nsecs(val);
-		val += i915->pmu.sample[__I915_SAMPLE_RC6].cur;
-		i915->pmu.sample[__I915_SAMPLE_RC6_ESTIMATED].cur = val;
-
 		spin_unlock_irqrestore(&i915->pmu.lock, flags);
 	}
 
diff --git a/drivers/gpu/drm/i915/intel_audio.c b/drivers/gpu/drm/i915/intel_audio.c
index 709d6ca68074..3ea566f99450 100644
--- a/drivers/gpu/drm/i915/intel_audio.c
+++ b/drivers/gpu/drm/i915/intel_audio.c
@@ -729,7 +729,7 @@ static void i915_audio_component_codec_wake_override(struct device *kdev,
 	struct drm_i915_private *dev_priv = kdev_to_i915(kdev);
 	u32 tmp;
 
-	if (!IS_GEN9_BC(dev_priv))
+	if (!IS_GEN9(dev_priv))
 		return;
 
 	i915_audio_component_get_power(kdev);
diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c
index c5c7530ba157..447b721c3be9 100644
--- a/drivers/gpu/drm/i915/intel_bios.c
+++ b/drivers/gpu/drm/i915/intel_bios.c
@@ -1256,7 +1256,6 @@ static void parse_ddi_port(struct drm_i915_private *dev_priv, enum port port,
 		return;
 
 	aux_channel = child->aux_channel;
-	ddc_pin = child->ddc_pin;
 
 	is_dvi = child->device_type & DEVICE_TYPE_TMDS_DVI_SIGNALING;
 	is_dp = child->device_type & DEVICE_TYPE_DISPLAYPORT_OUTPUT;
@@ -1303,9 +1302,15 @@ static void parse_ddi_port(struct drm_i915_private *dev_priv, enum port port,
 		DRM_DEBUG_KMS("Port %c is internal DP\n", port_name(port));
 
 	if (is_dvi) {
-		info->alternate_ddc_pin = map_ddc_pin(dev_priv, ddc_pin);
-
-		sanitize_ddc_pin(dev_priv, port);
+		ddc_pin = map_ddc_pin(dev_priv, child->ddc_pin);
+		if (intel_gmbus_is_valid_pin(dev_priv, ddc_pin)) {
+			info->alternate_ddc_pin = ddc_pin;
+			sanitize_ddc_pin(dev_priv, port);
+		} else {
+			DRM_DEBUG_KMS("Port %c has invalid DDC pin %d, "
+				      "sticking to defaults\n",
+				      port_name(port), ddc_pin);
+		}
 	}
 
 	if (is_dp) {
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 697af5add78b..e3a5f673ff67 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -577,6 +577,8 @@ static void execlists_dequeue(struct intel_engine_cs *engine)
 		 * know the next preemption status we see corresponds
 		 * to this ELSP update.
 		 */
+		GEM_BUG_ON(!execlists_is_active(execlists,
+						EXECLISTS_ACTIVE_USER));
 		GEM_BUG_ON(!port_count(&port[0]));
 		if (port_count(&port[0]) > 1)
 			goto unlock;
@@ -738,6 +740,8 @@ execlists_cancel_port_requests(struct intel_engine_execlists * const execlists)
 		memset(port, 0, sizeof(*port));
 		port++;
 	}
+
+	execlists_clear_active(execlists, EXECLISTS_ACTIVE_USER);
 }
 
 static void execlists_cancel_requests(struct intel_engine_cs *engine)
@@ -1001,6 +1005,11 @@ static void execlists_submission_tasklet(unsigned long data)
 
 	if (fw)
 		intel_uncore_forcewake_put(dev_priv, execlists->fw_domains);
+
+	/* If the engine is now idle, so should be the flag; and vice versa. */
+	GEM_BUG_ON(execlists_is_active(&engine->execlists,
+				       EXECLISTS_ACTIVE_USER) ==
+		   !port_isset(engine->execlists.port));
 }
 
 static void queue_request(struct intel_engine_cs *engine,