Merge tag 'drm-fixes-2020-02-21' of git://anongit.freedesktop.org/drm/drm

Pull drm fixes from Dave Airlie:
 "Varied fixes for rc3.

  i915 is the largest, they are seeing some ACPI problems with their CI
  which hopefully get solved soon [1].

  msm has a bunch of fixes for new hw added in the merge, a bunch of
  amdgpu fixes, and nouveau adds support for some new firmwares for
  turing tu11x GPUs that were just released into linux-firmware by
  nvidia, they operate the same as the ones we already have for tu10x so
  should be fine to hook up.

  Otherwise it's just misc fixes for panfrost and sun4i.

  core:
   - Allow only one rotation argument, and allow zero rotation in video
     cmdline.

  i915:
   - Workaround missing Display Stream Compression (DSC) state readout
     by forcing modeset when its enabled at probe
   - Fix EHL port clock voltage level requirements
   - Fix queuing retire workers on the virtual engine
   - Fix use of partially initialized waiters
   - Stop using drm_pci_alloc/drm_pci/free
   - Fix rewind of RING_TAIL by forcing a context reload
   - Fix locking on resetting ring->head
   - Propagate our bug filing URL change to stable kernels

  panfrost:
   - Small compiler warning fix for panfrost.
   - Fix when using performance counters in panfrost when using per fd
     address space.

  sun4xi:
   - Fix dt binding

  nouveau:
   - tu11x modesetting fix
   - ACR/GR firmware support for tu11x (fw is public now)

  msm:
   - fix UBWC on GPU and display side for sc7180
   - fix DSI suspend/resume issue encountered on sc7180
   - fix some breakage on so called "linux-android" devices
      (fallout from sc7180/a618 support, not seen earlier due to
       bootloader/firmware differences)
   - couple other misc fixes

  amdgpu:
   - HDCP fixes
   - xclk fix for raven
   - GFXOFF fixes"

[1] The Intel suspend testing should now be fixed by commit 63fb9623427f
    ("ACPI: PM: s2idle: Check fixed wakeup events in acpi_s2idle_wake()")

* tag 'drm-fixes-2020-02-21' of git://anongit.freedesktop.org/drm/drm: (39 commits)
  drm/amdgpu/display: clean up hdcp workqueue handling
  drm/amdgpu: add is_raven_kicker judgement for raven1
  drm/i915/gt: Avoid resetting ring->head outside of its timeline mutex
  drm/i915/execlists: Always force a context reload when rewinding RING_TAIL
  drm/i915: Wean off drm_pci_alloc/drm_pci_free
  drm/i915/gt: Protect defer_request() from new waiters
  drm/i915/gt: Prevent queuing retire workers on the virtual engine
  drm/i915/dsc: force full modeset whenever DSC is enabled at probe
  drm/i915/ehl: Update port clock voltage level requirements
  drm/i915: Update drm/i915 bug filing URL
  MAINTAINERS: Update drm/i915 bug filing URL
  drm/i915: Initialise basic fence before acquiring seqno
  drm/i915/gem: Require per-engine reset support for non-persistent contexts
  drm/nouveau/kms/gv100-: Re-set LUT after clearing for modesets
  drm/nouveau/gr/tu11x: initial support
  drm/nouveau/acr/tu11x: initial support
  drm/amdgpu/gfx10: disable gfxoff when reading rlc clock
  drm/amdgpu/gfx9: disable gfxoff when reading rlc clock
  drm/amdgpu/soc15: fix xclk for raven
  drm/amd/powerplay: always refetch the enabled features status on dpm enablement
  ...

50 files changed, 486 insertions, 243 deletions

diff --git a/Documentation/devicetree/bindings/display/allwinner,sun4i-a10-tcon.yaml b/Documentation/devicetree/bindings/display/allwinner,sun4i-a10-tcon.yaml
index 86ad617d2327d..5ff9cf26ca380 100644
--- a/Documentation/devicetree/bindings/display/allwinner,sun4i-a10-tcon.yaml
+++ b/Documentation/devicetree/bindings/display/allwinner,sun4i-a10-tcon.yaml
@@ -43,9 +43,13 @@ properties:
         - enum:
           - allwinner,sun8i-h3-tcon-tv
           - allwinner,sun50i-a64-tcon-tv
-          - allwinner,sun50i-h6-tcon-tv
         - const: allwinner,sun8i-a83t-tcon-tv
 
+      - items:
+        - enum:
+          - allwinner,sun50i-h6-tcon-tv
+        - const: allwinner,sun8i-r40-tcon-tv
+
   reg:
     maxItems: 1
 
diff --git a/MAINTAINERS b/MAINTAINERS
index 4e9bdb3c63b25..3ac093da657e4 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -8392,7 +8392,7 @@ M:	Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
 M:	Rodrigo Vivi <rodrigo.vivi@intel.com>
 L:	intel-gfx@lists.freedesktop.org
 W:	https://01.org/linuxgraphics/
-B:	https://01.org/linuxgraphics/documentation/how-report-bugs
+B:	https://gitlab.freedesktop.org/drm/intel/-/wikis/How-to-file-i915-bugs
 C:	irc://chat.freenode.net/intel-gfx
 Q:	http://patchwork.freedesktop.org/project/intel-gfx/
 T:	git git://anongit.freedesktop.org/drm-intel
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 3a1570dafe348..146f96661b6b5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -1013,6 +1013,30 @@ static int psp_dtm_initialize(struct psp_context *psp)
 	return 0;
 }
 
+static int psp_dtm_unload(struct psp_context *psp)
+{
+	int ret;
+	struct psp_gfx_cmd_resp *cmd;
+
+	/*
+	 * TODO: bypass the unloading in sriov for now
+	 */
+	if (amdgpu_sriov_vf(psp->adev))
+		return 0;
+
+	cmd = kzalloc(sizeof(struct psp_gfx_cmd_resp), GFP_KERNEL);
+	if (!cmd)
+		return -ENOMEM;
+
+	psp_prep_ta_unload_cmd_buf(cmd, psp->dtm_context.session_id);
+
+	ret = psp_cmd_submit_buf(psp, NULL, cmd, psp->fence_buf_mc_addr);
+
+	kfree(cmd);
+
+	return ret;
+}
+
 int psp_dtm_invoke(struct psp_context *psp, uint32_t ta_cmd_id)
 {
 	/*
@@ -1037,7 +1061,7 @@ static int psp_dtm_terminate(struct psp_context *psp)
 	if (!psp->dtm_context.dtm_initialized)
 		return 0;
 
-	ret = psp_hdcp_unload(psp);
+	ret = psp_dtm_unload(psp);
 	if (ret)
 		return ret;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 1785fdad6ecba..22bbb36c768e2 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -3923,11 +3923,13 @@ static uint64_t gfx_v10_0_get_gpu_clock_counter(struct amdgpu_device *adev)
 {
 	uint64_t clock;
 
+	amdgpu_gfx_off_ctrl(adev, false);
 	mutex_lock(&adev->gfx.gpu_clock_mutex);
 	WREG32_SOC15(GC, 0, mmRLC_CAPTURE_GPU_CLOCK_COUNT, 1);
 	clock = (uint64_t)RREG32_SOC15(GC, 0, mmRLC_GPU_CLOCK_COUNT_LSB) |
 		((uint64_t)RREG32_SOC15(GC, 0, mmRLC_GPU_CLOCK_COUNT_MSB) << 32ULL);
 	mutex_unlock(&adev->gfx.gpu_clock_mutex);
+	amdgpu_gfx_off_ctrl(adev, true);
 	return clock;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index b33a4eb39193c..3afdbbd6aaad9 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -1193,6 +1193,14 @@ static bool gfx_v9_0_should_disable_gfxoff(struct pci_dev *pdev)
 	return false;
 }
 
+static bool is_raven_kicker(struct amdgpu_device *adev)
+{
+	if (adev->pm.fw_version >= 0x41e2b)
+		return true;
+	else
+		return false;
+}
+
 static void gfx_v9_0_check_if_need_gfxoff(struct amdgpu_device *adev)
 {
 	if (gfx_v9_0_should_disable_gfxoff(adev->pdev))
@@ -1205,9 +1213,8 @@ static void gfx_v9_0_check_if_need_gfxoff(struct amdgpu_device *adev)
 		break;
 	case CHIP_RAVEN:
 		if (!(adev->rev_id >= 0x8 || adev->pdev->device == 0x15d8) &&
-		    ((adev->gfx.rlc_fw_version != 106 &&
+		    ((!is_raven_kicker(adev) &&
 		      adev->gfx.rlc_fw_version < 531) ||
-		     (adev->gfx.rlc_fw_version == 53815) ||
 		     (adev->gfx.rlc_feature_version < 1) ||
 		     !adev->gfx.rlc.is_rlc_v2_1))
 			adev->pm.pp_feature &= ~PP_GFXOFF_MASK;
@@ -3959,6 +3966,7 @@ static uint64_t gfx_v9_0_get_gpu_clock_counter(struct amdgpu_device *adev)
 {
 	uint64_t clock;
 
+	amdgpu_gfx_off_ctrl(adev, false);
 	mutex_lock(&adev->gfx.gpu_clock_mutex);
 	if (adev->asic_type == CHIP_VEGA10 && amdgpu_sriov_runtime(adev)) {
 		uint32_t tmp, lsb, msb, i = 0;
@@ -3977,6 +3985,7 @@ static uint64_t gfx_v9_0_get_gpu_clock_counter(struct amdgpu_device *adev)
 			((uint64_t)RREG32_SOC15(GC, 0, mmRLC_GPU_CLOCK_COUNT_MSB) << 32ULL);
 	}
 	mutex_unlock(&adev->gfx.gpu_clock_mutex);
+	amdgpu_gfx_off_ctrl(adev, true);
 	return clock;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index 15f3424a1ff79..2b488dfb2f21c 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -272,7 +272,12 @@ static u32 soc15_get_config_memsize(struct amdgpu_device *adev)
 
 static u32 soc15_get_xclk(struct amdgpu_device *adev)
 {
-	return adev->clock.spll.reference_freq;
+	u32 reference_clock = adev->clock.spll.reference_freq;
+
+	if (adev->asic_type == CHIP_RAVEN)
+		return reference_clock / 4;
+
+	return reference_clock;
 }
 
 
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 63e8a12a74bcb..e8f66fbf399e5 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -1911,7 +1911,7 @@ static void handle_hpd_irq(void *param)
 	mutex_lock(&aconnector->hpd_lock);
 
 #ifdef CONFIG_DRM_AMD_DC_HDCP
-	if (adev->asic_type >= CHIP_RAVEN)
+	if (adev->dm.hdcp_workqueue)
 		hdcp_reset_display(adev->dm.hdcp_workqueue, aconnector->dc_link->link_index);
 #endif
 	if (aconnector->fake_enable)
@@ -2088,8 +2088,10 @@ static void handle_hpd_rx_irq(void *param)
 		}
 	}
 #ifdef CONFIG_DRM_AMD_DC_HDCP
-	if (hpd_irq_data.bytes.device_service_irq.bits.CP_IRQ)
-		hdcp_handle_cpirq(adev->dm.hdcp_workqueue,  aconnector->base.index);
+	    if (hpd_irq_data.bytes.device_service_irq.bits.CP_IRQ) {
+		    if (adev->dm.hdcp_workqueue)
+			    hdcp_handle_cpirq(adev->dm.hdcp_workqueue,  aconnector->base.index);
+	    }
 #endif
 	if ((dc_link->cur_link_settings.lane_count != LANE_COUNT_UNKNOWN) ||
 	    (dc_link->type == dc_connection_mst_branch))
@@ -5702,7 +5704,7 @@ void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm,
 		drm_connector_attach_vrr_capable_property(
 			&aconnector->base);
 #ifdef CONFIG_DRM_AMD_DC_HDCP
-		if (adev->asic_type >= CHIP_RAVEN)
+		if (adev->dm.hdcp_workqueue)
 			drm_connector_attach_content_protection_property(&aconnector->base, true);
 #endif
 	}
diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c
index f730b94ac3c06..55246711700ba 100644
--- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c
+++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c
@@ -46,8 +46,8 @@ static inline enum mod_hdcp_status check_hdcp2_capable(struct mod_hdcp *hdcp)
 	enum mod_hdcp_status status;
 
 	if (is_dp_hdcp(hdcp))
-		status = (hdcp->auth.msg.hdcp2.rxcaps_dp[2] & HDCP_2_2_RX_CAPS_VERSION_VAL) &&
-				HDCP_2_2_DP_HDCP_CAPABLE(hdcp->auth.msg.hdcp2.rxcaps_dp[0]) ?
+		status = (hdcp->auth.msg.hdcp2.rxcaps_dp[0] == HDCP_2_2_RX_CAPS_VERSION_VAL) &&
+				HDCP_2_2_DP_HDCP_CAPABLE(hdcp->auth.msg.hdcp2.rxcaps_dp[2]) ?
 				MOD_HDCP_STATUS_SUCCESS :
 				MOD_HDCP_STATUS_HDCP2_NOT_CAPABLE;
 	else
diff --git a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c
index 0dc49479a7ebd..b06c057a90024 100644
--- a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c
+++ b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c
@@ -898,6 +898,9 @@ int smu_v11_0_system_features_control(struct smu_context *smu,
 	if (ret)
 		return ret;
 
+	bitmap_zero(feature->enabled, feature->feature_num);
+	bitmap_zero(feature->supported, feature->feature_num);
+
 	if (en) {
 		ret = smu_feature_get_enabled_mask(smu, feature_mask, 2);
 		if (ret)
@@ -907,9 +910,6 @@ int smu_v11_0_system_features_control(struct smu_context *smu,
 			    feature->feature_num);
 		bitmap_copy(feature->supported, (unsigned long *)&feature_mask,
 			    feature->feature_num);
-	} else {
-		bitmap_zero(feature->enabled, feature->feature_num);
-		bitmap_zero(feature->supported, feature->feature_num);
 	}
 
 	return ret;
diff --git a/drivers/gpu/drm/bridge/tc358767.c b/drivers/gpu/drm/bridge/tc358767.c
index 3709e5ace7246..fbdb42d4e772e 100644
--- a/drivers/gpu/drm/bridge/tc358767.c
+++ b/drivers/gpu/drm/bridge/tc358767.c
@@ -297,7 +297,7 @@ static inline int tc_poll_timeout(struct tc_data *tc, unsigned int addr,
 
 static int tc_aux_wait_busy(struct tc_data *tc)
 {
-	return tc_poll_timeout(tc, DP0_AUXSTATUS, AUX_BUSY, 0, 1000, 100000);
+	return tc_poll_timeout(tc, DP0_AUXSTATUS, AUX_BUSY, 0, 100, 100000);
 }
 
 static int tc_aux_write_data(struct tc_data *tc, const void *data,
@@ -640,7 +640,7 @@ static int tc_aux_link_setup(struct tc_data *tc)
 	if (ret)
 		goto err;
 
-	ret = tc_poll_timeout(tc, DP_PHY_CTRL, PHY_RDY, PHY_RDY, 1, 1000);
+	ret = tc_poll_timeout(tc, DP_PHY_CTRL, PHY_RDY, PHY_RDY, 100, 100000);
 	if (ret == -ETIMEDOUT) {
 		dev_err(tc->dev, "Timeout waiting for PHY to become ready");
 		return ret;
@@ -876,7 +876,7 @@ static int tc_wait_link_training(struct tc_data *tc)
 	int ret;
 
 	ret = tc_poll_timeout(tc, DP0_LTSTAT, LT_LOOPDONE,
-			      LT_LOOPDONE, 1, 1000);
+			      LT_LOOPDONE, 500, 100000);
 	if (ret) {
 		dev_err(tc->dev, "Link training timeout waiting for LT_LOOPDONE!\n");
 		return ret;
@@ -949,7 +949,7 @@ static int tc_main_link_enable(struct tc_data *tc)
 	dp_phy_ctrl &= ~(DP_PHY_RST | PHY_M1_RST | PHY_M0_RST);
 	ret = regmap_write(tc->regmap, DP_PHY_CTRL, dp_phy_ctrl);
 
-	ret = tc_poll_timeout(tc, DP_PHY_CTRL, PHY_RDY, PHY_RDY, 1, 1000);
+	ret = tc_poll_timeout(tc, DP_PHY_CTRL, PHY_RDY, PHY_RDY, 500, 100000);
 	if (ret) {
 		dev_err(dev, "timeout waiting for phy become ready");
 		return ret;
diff --git a/drivers/gpu/drm/bridge/ti-tfp410.c b/drivers/gpu/drm/bridge/ti-tfp410.c
index 6f6d6d1e60ae9..f195a4732e0ba 100644
--- a/drivers/gpu/drm/bridge/ti-tfp410.c
+++ b/drivers/gpu/drm/bridge/ti-tfp410.c
@@ -140,7 +140,8 @@ static int tfp410_attach(struct drm_bridge *bridge)
 					  dvi->connector_type,
 					  dvi->ddc);
 	if (ret) {
-		dev_err(dvi->dev, "drm_connector_init() failed: %d\n", ret);
+		dev_err(dvi->dev, "drm_connector_init_with_ddc() failed: %d\n",
+			ret);
 		return ret;
 	}
 
diff --git a/drivers/gpu/drm/drm_client_modeset.c b/drivers/gpu/drm/drm_client_modeset.c
index 6d4a29e99ae26..3035584f6dc72 100644
--- a/drivers/gpu/drm/drm_client_modeset.c
+++ b/drivers/gpu/drm/drm_client_modeset.c
@@ -951,7 +951,8 @@ bool drm_client_rotation(struct drm_mode_set *modeset, unsigned int *rotation)
 	 * depending on the hardware this may require the framebuffer
 	 * to be in a specific tiling format.
 	 */
-	if ((*rotation & DRM_MODE_ROTATE_MASK) != DRM_MODE_ROTATE_180 ||
+	if (((*rotation & DRM_MODE_ROTATE_MASK) != DRM_MODE_ROTATE_0 &&
+	     (*rotation & DRM_MODE_ROTATE_MASK) != DRM_MODE_ROTATE_180) ||
 	    !plane->rotation_property)
 		return false;
 
diff --git a/drivers/gpu/drm/drm_modes.c b/drivers/gpu/drm/drm_modes.c
index 10336b144c722..d4d64518e11b8 100644
--- a/drivers/gpu/drm/drm_modes.c
+++ b/drivers/gpu/drm/drm_modes.c
@@ -1698,6 +1698,13 @@ static int drm_mode_parse_cmdline_options(const char *str,
 	if (rotation && freestanding)
 		return -EINVAL;
 
+	if (!(rotation & DRM_MODE_ROTATE_MASK))
+		rotation |= DRM_MODE_ROTATE_0;
+
+	/* Make sure there is exactly one rotation defined */
+	if (!is_power_of_2(rotation & DRM_MODE_ROTATE_MASK))
+		return -EINVAL;
+
 	mode->rotation_reflection = rotation;
 
 	return 0;
diff --git a/drivers/gpu/drm/i915/Kconfig b/drivers/gpu/drm/i915/Kconfig
index ba9595960bbeb..907c4471f5916 100644
--- a/drivers/gpu/drm/i915/Kconfig
+++ b/drivers/gpu/drm/i915/Kconfig
@@ -75,9 +75,8 @@ config DRM_I915_CAPTURE_ERROR
 	help
 	  This option enables capturing the GPU state when a hang is detected.
 	  This information is vital for triaging hangs and assists in debugging.
-	  Please report any hang to
-	    https://bugs.freedesktop.org/enter_bug.cgi?product=DRI
-	  for triaging.
+	  Please report any hang for triaging according to:
+	    https://gitlab.freedesktop.org/drm/intel/-/wikis/How-to-file-i915-bugs
 
 	  If in doubt, say "Y".
 
diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index 33f1dc3d7c1a6..d9a61f341070b 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -4251,7 +4251,9 @@ static bool intel_ddi_is_audio_enabled(struct drm_i915_private *dev_priv,
 void intel_ddi_compute_min_voltage_level(struct drm_i915_private *dev_priv,
 					 struct intel_crtc_state *crtc_state)
 {
-	if (INTEL_GEN(dev_priv) >= 11 && crtc_state->port_clock > 594000)
+	if (IS_ELKHARTLAKE(dev_priv) && crtc_state->port_clock > 594000)
+		crtc_state->min_voltage_level = 3;
+	else if (INTEL_GEN(dev_priv) >= 11 && crtc_state->port_clock > 594000)
 		crtc_state->min_voltage_level = 1;
 	else if (IS_CANNONLAKE(dev_priv) && crtc_state->port_clock > 594000)
 		crtc_state->min_voltage_level = 2;
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 064dd99bbc499..aa453953908b5 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -11087,7 +11087,7 @@ static u32 intel_cursor_base(const struct intel_plane_state *plane_state)
 	u32 base;
 
 	if (INTEL_INFO(dev_priv)->display.cursor_needs_physical)
-		base = obj->phys_handle->busaddr;
+		base = sg_dma_address(obj->mm.pages->sgl);
 	else
 		base = intel_plane_ggtt_offset(plane_state);
 
@@ -17433,6 +17433,24 @@ retry:
 			 * have readout for pipe gamma enable.
 			 */
 			crtc_state->uapi.color_mgmt_changed = true;
+
+			/*
+			 * FIXME hack to force full modeset when DSC is being
+			 * used.
+			 *
+			 * As long as we do not have full state readout and
+			 * config comparison of crtc_state->dsc, we have no way
+			 * to ensure reliable fastset. Remove once we have
+			 * readout for DSC.
+			 */
+			if (crtc_state->dsc.compression_enable) {
+				ret = drm_atomic_add_affected_connectors(state,
+									 &crtc->base);
+				if (ret)
+					goto out;
+				crtc_state->uapi.mode_changed = true;
+				drm_dbg_kms(dev, "Force full modeset for DSC\n");
+			}
 		}
 	}
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_context.c b/drivers/gpu/drm/i915/gem/i915_gem_context.c
index a2e57e62af30a..151a1e8ae36ab 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_context.c
@@ -565,6 +565,22 @@ static int __context_set_persistence(struct i915_gem_context *ctx, bool state)
 		if (!(ctx->i915->caps.scheduler & I915_SCHEDULER_CAP_PREEMPTION))
 			return -ENODEV;
 
+		/*
+		 * If the cancel fails, we then need to reset, cleanly!
+		 *
+		 * If the per-engine reset fails, all hope is lost! We resort
+		 * to a full GPU reset in that unlikely case, but realistically
+		 * if the engine could not reset, the full reset does not fare
+		 * much better. The damage has been done.
+		 *
+		 * However, if we cannot reset an engine by itself, we cannot
+		 * cleanup a hanging persistent context without causing
+		 * colateral damage, and we should not pretend we can by
+		 * exposing the interface.
+		 */
+		if (!intel_has_reset_engine(&ctx->i915->gt))
+			return -ENODEV;
+
 		i915_gem_context_clear_persistence(ctx);
 	}
 
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 f64ad77e6b1ee..c2174da35bb0f 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
@@ -285,9 +285,6 @@ struct drm_i915_gem_object {
 
 		void *gvt_info;
 	};
-
-	/** for phys allocated objects */
-	struct drm_dma_handle *phys_handle;
 };
 
 static inline struct drm_i915_gem_object *
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_phys.c b/drivers/gpu/drm/i915/gem/i915_gem_phys.c
index b1b7c1b3038aa..b07bb40edd5a3 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_phys.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_phys.c
@@ -22,88 +22,87 @@
 static int i915_gem_object_get_pages_phys(struct drm_i915_gem_object *obj)
 {
 	struct address_space *mapping = obj->base.filp->f_mapping;
-	struct drm_dma_handle *phys;
-	struct sg_table *st;
 	struct scatterlist *sg;
-	char *vaddr;
+	struct sg_table *st;
+	dma_addr_t dma;
+	void *vaddr;
+	void *dst;
 	int i;
-	int err;
 
 	if (WARN_ON(i915_gem_object_needs_bit17_swizzle(obj)))
 		return -EINVAL;
 
-	/* Always aligning to the object size, allows a single allocation
+	/*
+	 * Always aligning to the object size, allows a single allocation
 	 * to handle all possible callers, and given typical object sizes,
 	 * the alignment of the buddy allocation will naturally match.
 	 */
-	phys = drm_pci_alloc(obj->base.dev,
-			     roundup_pow_of_two(obj->base.size),
-			     roundup_pow_of_two(obj->base.size));
-	if (!phys)
+	vaddr = dma_alloc_coherent(&obj->base.dev->pdev->dev,
+				   roundup_pow_of_two(obj->base.size),
+				   &dma, GFP_KERNEL);
+	if (!vaddr)
 		return -ENOMEM;
 
-	vaddr = phys->vaddr;
+	st = kmalloc(sizeof(*st), GFP_KERNEL);
+	if (!st)
+		goto err_pci;
+
+	if (sg_alloc_table(st, 1, GFP_KERNEL))
+		goto err_st;
+
+	sg = st->sgl;
+	sg->offset = 0;
+	sg->length = obj->base.size;
+
+	sg_assign_page(sg, (struct page *)vaddr);
+	sg_dma_address(sg) = dma;
+	sg_dma_len(sg) = obj->base.size;
+
+	dst = vaddr;
 	for (i = 0; i < obj->base.size / PAGE_SIZE; i++) {
 		struct page *page;
-		char *src;
+		void *src;
 
 		page = shmem_read_mapping_page(mapping, i);
-		if (IS_ERR(page)) {
-			err = PTR_ERR(page);
-			goto err_phys;
-		}
+		if (IS_ERR(page))
+			goto err_st;
 
 		src = kmap_atomic(page);
-		memcpy(vaddr, src, PAGE_SIZE);
-		drm_clflush_virt_range(vaddr, PAGE_SIZE);
+		memcpy(dst, src, PAGE_SIZE);
+		drm_clflush_virt_range(dst, PAGE_SIZE);
 		kunmap_atomic(src);
 
 		put_page(page);
-		vaddr += PAGE_SIZE;
+		dst += PAGE_SIZE;
 	}
 
 	intel_gt_chipset_flush(&to_i915(obj->base.dev)->gt);
 
-	st = kmalloc(sizeof(*st), GFP_KERNEL);
-	if (!st) {
-		err = -ENOMEM;
-		goto err_phys;
-	}
-
-	if (sg_alloc_table(st, 1, GFP_KERNEL)) {
-		kfree(st);
-		err = -ENOMEM;
-		goto err_phys;
-	}
-
-	sg = st->sgl;
-	sg->offset = 0;
-	sg->length = obj->base.size;
-
-	sg_dma_address(sg) = phys->busaddr;
-	sg_dma_len(sg) = obj->base.size;
-
-	obj->phys_handle = phys;
-
 	__i915_gem_object_set_pages(obj, st, sg->length);
 
 	return 0;
 
-err_phys:
-	drm_pci_free(obj->base.dev, phys);
-
-	return err;
+err_st:
+	kfree(st);
+err_pci:
+	dma_free_coherent(&obj->base.dev->pdev->dev,
+			  roundup_pow_of_two(obj->base.size),
+			  vaddr, dma);
+	return -ENOMEM;
 }
 
 static void
 i915_gem_object_put_pages_phys(struct drm_i915_gem_object *obj,
 			       struct sg_table *pages)
 {
+	dma_addr_t dma = sg_dma_address(pages->sgl);
+	void *vaddr = sg_page(pages->sgl);
+
 	__i915_gem_object_release_shmem(obj, pages, false);
 
 	if (obj->mm.dirty) {
 		struct address_space *mapping = obj->base.filp->f_mapping;
-		char *vaddr = obj->phys_handle->vaddr;
+		void *src = vaddr;
 		int i;
 
 		for (i = 0; i < obj->base.size / PAGE_SIZE; i++) {
@@ -115,15 +114,16 @@ i915_gem_object_put_pages_phys(struct drm_i915_gem_object *obj,
 				continue;
 
 			dst = kmap_atomic(page);
-			drm_clflush_virt_range(vaddr, PAGE_SIZE);
-			memcpy(dst, vaddr, PAGE_SIZE);
+			drm_clflush_virt_range(src, PAGE_SIZE);
+			memcpy(dst, src, PAGE_SIZE);
 			kunmap_atomic(dst);
 
 			set_page_dirty(page);
 			if (obj->mm.madv == I915_MADV_WILLNEED)
 				mark_page_accessed(page);
 			put_page(page);
-			vaddr += PAGE_SIZE;
+
+			src += PAGE_SIZE;
 		}
 		obj->mm.dirty = false;
 	}
@@ -131,7 +131,9 @@ i915_gem_object_put_pages_phys(struct drm_i915_gem_object *obj,
 	sg_free_table(pages);
 	kfree(pages);
 
-	drm_pci_free(obj->base.dev, obj->phys_handle);
+	dma_free_coherent(&obj->base.dev->pdev->dev,
+			  roundup_pow_of_two(obj->base.size),
+			  vaddr, dma);
 }
 
 static void phys_release(struct drm_i915_gem_object *obj)
diff --git a/drivers/gpu/drm/i915/gt/intel_breadcrumbs.c b/drivers/gpu/drm/i915/gt/intel_breadcrumbs.c
index 0ba524a414c68..cbad7fe722ceb 100644
--- a/drivers/gpu/drm/i915/gt/intel_breadcrumbs.c
+++ b/drivers/gpu/drm/i915/gt/intel_breadcrumbs.c
@@ -136,6 +136,9 @@ static void add_retire(struct intel_breadcrumbs *b, struct intel_timeline *tl)
 	struct intel_engine_cs *engine =
 		container_of(b, struct intel_engine_cs, breadcrumbs);
 
+	if (unlikely(intel_engine_is_virtual(engine)))
+		engine = intel_virtual_engine_get_sibling(engine, 0);
+
 	intel_engine_add_retire(engine, tl);
 }
 
diff --git a/drivers/gpu/drm/i915/gt/intel_gt_requests.c b/drivers/gpu/drm/i915/gt/intel_gt_requests.c
index 7ef1d37970f6d..8a5054f21bf88 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_requests.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt_requests.c
@@ -99,6 +99,9 @@ static bool add_retire(struct intel_engine_cs *engine,
 void intel_engine_add_retire(struct intel_engine_cs *engine,
 			     struct intel_timeline *tl)
 {
+	/* We don't deal well with the engine disappearing beneath us */
+	GEM_BUG_ON(intel_engine_is_virtual(engine));
+
 	if (add_retire(engine, tl))
 		schedule_work(&engine->retire_work);
 }
diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c
index a13a8c4b65ab9..fe8a59aaa629a 100644
--- a/drivers/gpu/drm/i915/gt/intel_lrc.c
+++ b/drivers/gpu/drm/i915/gt/intel_lrc.c
@@ -237,7 +237,8 @@ static void execlists_init_reg_state(u32 *reg_state,
 				     bool close);
 static void
 __execlists_update_reg_state(const struct intel_context *ce,
-			     const struct intel_engine_cs *engine);
+			     const struct intel_engine_cs *engine,
+			     u32 head);
 
 static void mark_eio(struct i915_request *rq)
 {
@@ -1186,12 +1187,11 @@ static void reset_active(struct i915_request *rq,
 		head = rq->tail;
 	else
 		head = active_request(ce->timeline, rq)->head;
-	ce->ring->head = intel_ring_wrap(ce->ring, head);
-	intel_ring_update_space(ce->ring);
+	head = intel_ring_wrap(ce->ring, head);
 
 	/* Scrub the context image to prevent replaying the previous batch */
 	restore_default_state(ce, engine);
-	__execlists_update_reg_state(ce, engine);
+	__execlists_update_reg_state(ce, engine, head);
 
 	/* We've switched away, so this should be a no-op, but intent matters */
 	ce->lrc_desc |= CTX_DESC_FORCE_RESTORE;
@@ -1321,7 +1321,7 @@ static u64 execlists_update_context(struct i915_request *rq)
 {
 	struct intel_context *ce = rq->context;
 	u64 desc = ce->lrc_desc;
-	u32 tail;
+	u32 tail, prev;
 
 	/*
 	 * WaIdleLiteRestore:bdw,skl
@@ -1334,9 +1334,15 @@ static u64 execlists_update_context(struct i915_request *rq)
 	 * subsequent resubmissions (for lite restore). Should that fail us,
 	 * and we try and submit the same tail again, force the context
 	 * reload.
+	 *
+	 * If we need to return to a preempted context, we need to skip the
+	 * lite-restore and force it to reload the RING_TAIL. Otherwise, the
+	 * HW has a tendency to ignore us rewinding the TAIL to the end of
+	 * an earlier request.
 	 */
 	tail = intel_ring_set_tail(rq->ring, rq->tail);
-	if (unlikely(ce->lrc_reg_state[CTX_RING_TAIL] == tail))
+	prev = ce->lrc_reg_state[CTX_RING_TAIL];
+	if (unlikely(intel_ring_direction(rq->ring, tail, prev) <= 0))
 		desc |= CTX_DESC_FORCE_RESTORE;
 	ce->lrc_reg_state[CTX_RING_TAIL] = tail;
 	rq->tail = rq->wa_tail;
@@ -1605,6 +1611,11 @@ last_active(const struct intel_engine_execlists *execlists)
 	return *last;
 }
 
+#define for_each_waiter(p__, rq__) \
+	list_for_each_entry_lockless(p__, \
+				     &(rq__)->sched.waiters_list, \
+				     wait_link)
+
 static void defer_request(struct i915_request *rq, struct list_head * const pl)
 {
 	LIST_HEAD(list);
@@ -1622,7 +1633,7 @@ static void defer_request(struct i915_request *rq, struct list_head * const pl)
 		GEM_BUG_ON(i915_request_is_active(rq));
 		list_move_tail(&rq->sched.link, pl);
 
-		list_for_each_entry(p, &rq->sched.waiters_list, wait_link) {
+		for_each_waiter(p, rq) {
 			struct i915_request *w =
 				container_of(p->waiter, typeof(*w), sched);
 
@@ -1834,14 +1845,6 @@ static void execlists_dequeue(struct intel_engine_cs *engine)
 			 */
 			__unwind_incomplete_requests(engine);
 
-			/*
-			 * If we need to return to the preempted context, we
-			 * need to skip the lite-restore and force it to
-			 * reload the RING_TAIL. Otherwise, the HW has a
-			 * tendency to ignore us rewinding the TAIL to the
-			 * end of an earlier request.
-			 */
-			last->context->lrc_desc |= CTX_DESC_FORCE_RESTORE;
 			last = NULL;
 		} else if (need_timeslice(engine, last) &&
 			   timer_expired(&engine->execlists.timer)) {
@@ -2860,16 +2863,17 @@ static void execlists_context_unpin(struct intel_context *ce)
 
 static void
 __execlists_update_reg_state(const struct intel_context *ce,
-			     const struct intel_engine_cs *engine)
+			     const struct intel_engine_cs *engine,
+			     u32 head)
 {
 	struct intel_ring *ring = ce->ring;
 	u32 *regs = ce->lrc_reg_state;
 
-	GEM_BUG_ON(!intel_ring_offset_valid(ring, ring->head));
+	GEM_BUG_ON(!intel_ring_offset_valid(ring, head));
 	GEM_BUG_ON(!intel_ring_offset_valid(ring, ring->tail));
 
 	regs[CTX_RING_START] = i915_ggtt_offset(ring->vma);
-	regs[CTX_RING_HEAD] = ring->head;
+	regs[CTX_RING_HEAD] = head;
 	regs[CTX_RING_TAIL] = ring->tail;
 
 	/* RPCS */
@@ -2898,7 +2902,7 @@ __execlists_context_pin(struct intel_context *ce,
 
 	ce->lrc_desc = lrc_descriptor(ce, engine) | CTX_DESC_FORCE_RESTORE;
 	ce->lrc_reg_state = vaddr + LRC_STATE_PN * PAGE_SIZE;
-	__execlists_update_reg_state(ce, engine);
+	__execlists_update_reg_state(ce, engine, ce->ring->tail);
 
 	return 0;
 }
@@ -2939,7 +2943,7 @@ static void execlists_context_reset(struct intel_context *ce)
 	/* Scrub away the garbage */
 	execlists_init_reg_state(ce->lrc_reg_state,
 				 ce, ce->engine, ce->ring, true);
-	__execlists_update_reg_state(ce, ce->engine);
+	__execlists_update_reg_state(ce, ce->engine, ce->ring->tail);
 
 	ce->lrc_desc |= CTX_DESC_FORCE_RESTORE;
 }
@@ -3494,6 +3498,7 @@ static void __execlists_reset(struct intel_engine_cs *engine, bool stalled)
 	struct intel_engine_execlists * const execlists = &engine->execlists;
 	struct intel_context *ce;
 	struct i915_request *rq;
+	u32 head;
 
 	mb(); /* paranoia: read the CSB pointers from after the reset */
 	clflush(execlists->csb_write);
@@ -3521,15 +3526,15 @@ static void __execlists_reset(struct intel_engine_cs *engine, bool stalled)
 
 	if (i915_request_completed(rq)) {
 		/* Idle context; tidy up the ring so we can restart afresh */
-		ce->ring->head = intel_ring_wrap(ce->ring, rq->tail);
+		head = intel_ring_wrap(ce->ring, rq->tail);
 		goto out_replay;
 	}
 
 	/* Context has requests still in-flight; it should not be idle! */
 	GEM_BUG_ON(i915_active_is_idle(&ce->active));
 	rq = active_request(ce->timeline, rq);
-	ce->ring->head = intel_ring_wrap(ce->ring, rq->head);
-	GEM_BUG_ON(ce->ring->head == ce->ring->tail);
+	head = intel_ring_wrap(ce->ring, rq->head);
+	GEM_BUG_ON(head == ce->ring->tail);
 
 	/*
 	 * If this request hasn't started yet, e.g. it is waiting on a
@@ -3574,10 +3579,9 @@ static void __execlists_reset(struct intel_engine_cs *engine, bool stalled)
 
 out_replay:
 	ENGINE_TRACE(engine, "replay {head:%04x, tail:%04x}\n",
-		     ce->ring->head, ce->ring->tail);
-	intel_ring_update_space(ce->ring);
+		     head, ce->ring->tail);
 	__execlists_reset_reg_state(ce, engine);
-	__execlists_update_reg_state(ce, engine);
+	__execlists_update_reg_state(ce, engine, head);
 	ce->lrc_desc |= CTX_DESC_FORCE_RESTORE; /* paranoid: GPU was reset! */
 
 unwind:
@@ -5220,10 +5224,7 @@ void intel_lr_context_reset(struct intel_engine_cs *engine,
 		restore_default_state(ce, engine);
 
 	/* Rerun the request; its payload has been neutered (if guilty). */
-	ce->ring->head = head;
-	intel_ring_update_space(ce->ring);
-
-	__execlists_update_reg_state(ce, engine);
+	__execlists_update_reg_state(ce, engine, head);
 }
 
 bool
diff --git a/drivers/gpu/drm/i915/gt/intel_ring.c b/drivers/gpu/drm/i915/gt/intel_ring.c
index 374b28f13ca0b..6ff803f397c4d 100644
--- a/drivers/gpu/drm/i915/gt/intel_ring.c
+++ b/drivers/gpu/drm/i915/gt/intel_ring.c
@@ -145,6 +145,7 @@ intel_engine_create_ring(struct intel_engine_cs *engine, int size)
 
 	kref_init(&ring->ref);
 	ring->size = size;
+	ring->wrap = BITS_PER_TYPE(ring->size) - ilog2(size);
 
 	/*
 	 * Workaround an erratum on the i830 which causes a hang if
diff --git a/drivers/gpu/drm/i915/gt/intel_ring.h b/drivers/gpu/drm/i915/gt/intel_ring.h
index ea2839d9e0445..5bdce24994aa0 100644
--- a/drivers/gpu/drm/i915/gt/intel_ring.h
+++ b/drivers/gpu/drm/i915/gt/intel_ring.h
@@ -56,6 +56,14 @@ static inline u32 intel_ring_wrap(const struct intel_ring *ring, u32 pos)
 	return pos & (ring->size - 1);
 }
 
+static inline int intel_ring_direction(const struct intel_ring *ring,
+				       u32 next, u32 prev)
+{
+	typecheck(typeof(ring->size), next);
+	typecheck(typeof(ring->size), prev);
+	return (next - prev) << ring->wrap;
+}
+
 static inline bool
 intel_ring_offset_valid(const struct intel_ring *ring,
 			unsigned int pos)
diff --git a/drivers/gpu/drm/i915/gt/intel_ring_types.h b/drivers/gpu/drm/i915/gt/intel_ring_types.h
index d9f17f38e0cce..1a189ea00fd82 100644
--- a/drivers/gpu/drm/i915/gt/intel_ring_types.h
+++ b/drivers/gpu/drm/i915/gt/intel_ring_types.h
@@ -39,12 +39,13 @@ struct intel_ring {
 	 */
 	atomic_t pin_count;
 
-	u32 head;
-	u32 tail;
-	u32 emit;
+	u32 head; /* updated during retire, loosely tracks RING_HEAD */
+	u32 tail; /* updated on submission, used for RING_TAIL */
+	u32 emit; /* updated during request construction */
 
 	u32 space;
 	u32 size;
+	u32 wrap;
 	u32 effective_size;
 };
 
diff --git a/drivers/gpu/drm/i915/gt/selftest_lrc.c b/drivers/gpu/drm/i915/gt/selftest_lrc.c
index 65718ca2326e8..b292f8cbd0bf1 100644
--- a/drivers/gpu/drm/i915/gt/selftest_lrc.c
+++ b/drivers/gpu/drm/i915/gt/selftest_lrc.c
@@ -186,7 +186,7 @@ static int live_unlite_restore(struct intel_gt *gt, int prio)
 		}
 		GEM_BUG_ON(!ce[1]->ring->size);
 		intel_ring_reset(ce[1]->ring, ce[1]->ring->size / 2);
-		__execlists_update_reg_state(ce[1], engine);
+		__execlists_update_reg_state(ce[1], engine, ce[1]->ring->head);
 
 		rq[0] = igt_spinner_create_request(&spin, ce[0], MI_ARB_CHECK);
 		if (IS_ERR(rq[0])) {
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index c2de2f45b459b..5f6e639528219 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -180,7 +180,7 @@ i915_gem_phys_pwrite(struct drm_i915_gem_object *obj,
 		     struct drm_i915_gem_pwrite *args,
 		     struct drm_file *file)
 {
-	void *vaddr = obj->phys_handle->vaddr + args->offset;
+	void *vaddr = sg_page(obj->mm.pages->sgl) + args->offset;
 	char __user *user_data = u64_to_user_ptr(args->data_ptr);
 
 	/*
@@ -844,10 +844,10 @@ i915_gem_pwrite_ioctl(struct drm_device *dev, void *data,
 		ret = i915_gem_gtt_pwrite_fast(obj, args);
 
 	if (ret == -EFAULT || ret == -ENOSPC) {
-		if (obj->phys_handle)
-			ret = i915_gem_phys_pwrite(obj, args, file);
-		else
+		if (i915_gem_object_has_struct_page(obj))
 			ret = i915_gem_shmem_pwrite(obj, args);
+		else
+			ret = i915_gem_phys_pwrite(obj, args, file);
 	}
 
 	i915_gem_object_unpin_pages(obj);
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 594341e27a47d..9e401a5fcae8c 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -1852,7 +1852,8 @@ void i915_error_state_store(struct i915_gpu_coredump *error)
 	if (!xchg(&warned, true) &&
 	    ktime_get_real_seconds() - DRIVER_TIMESTAMP < DAY_AS_SECONDS(180)) {
 		pr_info("GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.\n");
-		pr_info("Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel\n");
+		pr_info("Please file a _new_ bug report at https://gitlab.freedesktop.org/drm/intel/issues/new.\n");
+		pr_info("Please see https://gitlab.freedesktop.org/drm/intel/-/wikis/How-to-file-i915-bugs for details.\n");
 		pr_info("drm/i915 developers can then reassign to the right component if it's not a kernel issue.\n");
 		pr_info("The GPU crash dump is required to analyze GPU hangs, so please always attach it.\n");
 		pr_info("GPU crash dump saved to /sys/class/drm/card%d/error\n",
diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c
index 78a5f5d3c0700..f56b046a32de1 100644
--- a/drivers/gpu/drm/i915/i915_request.c
+++ b/drivers/gpu/drm/i915/i915_request.c
@@ -595,6 +595,8 @@ static void __i915_request_ctor(void *arg)
 	i915_sw_fence_init(&rq->submit, submit_notify);
 	i915_sw_fence_init(&rq->semaphore, semaphore_notify);
 
+	dma_fence_init(&rq->fence, &i915_fence_ops, &rq->lock, 0, 0);
+
 	rq->file_priv = NULL;
 	rq->capture_list = NULL;
 
@@ -653,25 +655,30 @@ __i915_request_create(struct intel_context *ce, gfp_t gfp)
 		}
 	}
 
-	ret = intel_timeline_get_seqno(tl, rq, &seqno);
-	if (ret)
-		goto err_free;
-
 	rq->i915 = ce->engine->i915;
 	rq->context = ce;
 	rq->engine = ce->engine;
 	rq->ring = ce->ring;
 	rq->execution_mask = ce->engine->mask;
 
+	kref_init(&rq->fence.refcount);
+	rq->fence.flags = 0;
+	rq->fence.error = 0;
+	INIT_LIST_HEAD(&rq->fence.cb_list);
+
+	ret = intel_timeline_get_seqno(tl, rq, &seqno);
+	if (ret)
+		goto err_free;
+
+	rq->fence.context = tl->fence_context;
+	rq->fence.seqno = seqno;
+
 	RCU_INIT_POINTER(rq->timeline, tl);
 	RCU_INIT_POINTER(rq->hwsp_cacheline, tl->hwsp_cacheline);
 	rq->hwsp_seqno = tl->hwsp_seqno;
 
 	rq->rcustate = get_state_synchronize_rcu(); /* acts as smp_mb() */
 
-	dma_fence_init(&rq->fence, &i915_fence_ops, &rq->lock,
-		       tl->fence_context, seqno);
-
 	/* We bump the ref for the fence chain */
 	i915_sw_fence_reinit(&i915_request_get(rq)->submit);
 	i915_sw_fence_reinit(&i915_request_get(rq)->semaphore);
diff --git a/drivers/gpu/drm/i915/i915_scheduler.c b/drivers/gpu/drm/i915/i915_scheduler.c
index 5d96cfba40f88..34b654b4e58af 100644
--- a/drivers/gpu/drm/i915/i915_scheduler.c
+++ b/drivers/gpu/drm/i915/i915_scheduler.c
@@ -423,8 +423,6 @@ bool __i915_sched_node_add_dependency(struct i915_sched_node *node,
 
 	if (!node_signaled(signal)) {
 		INIT_LIST_HEAD(&dep->dfs_link);
-		list_add(&dep->wait_link, &signal->waiters_list);
-		list_add(&dep->signal_link, &node->signalers_list);
 		dep->signaler = signal;
 		dep->waiter = node;
 		dep->flags = flags;
@@ -434,6 +432,10 @@ bool __i915_sched_node_add_dependency(struct i915_sched_node *node,
 		    !node_started(signal))
 			node->flags |= I915_SCHED_HAS_SEMAPHORE_CHAIN;
 
+		/* All set, now publish. Beware the lockless walkers. */
+		list_add(&dep->signal_link, &node->signalers_list);
+		list_add_rcu(&dep->wait_link, &signal->waiters_list);
+
 		/*
 		 * As we do not allow WAIT to preempt inflight requests,
 		 * once we have executed a request, along with triggering
diff --git a/drivers/gpu/drm/i915/i915_utils.c b/drivers/gpu/drm/i915/i915_utils.c
index c47261ae86eab..632d6953c78da 100644
--- a/drivers/gpu/drm/i915/i915_utils.c
+++ b/drivers/gpu/drm/i915/i915_utils.c
@@ -8,9 +8,8 @@
 #include "i915_drv.h"
 #include "i915_utils.h"
 
-#define FDO_BUG_URL "https://bugs.freedesktop.org/enter_bug.cgi?product=DRI"
-#define FDO_BUG_MSG "Please file a bug at " FDO_BUG_URL " against DRM/Intel " \
-		    "providing the dmesg log by booting with drm.debug=0xf"
+#define FDO_BUG_URL "https://gitlab.freedesktop.org/drm/intel/-/wikis/How-to-file-i915-bugs"
+#define FDO_BUG_MSG "Please file a bug on drm/i915; see " FDO_BUG_URL " for details."
 
 void
 __i915_printk(struct drm_i915_private *dev_priv, const char *level,
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
index 983afeaee737e..748cd379065f1 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
@@ -796,12 +796,41 @@ bool a6xx_gmu_isidle(struct a6xx_gmu *gmu)
 	return true;
 }
 
+#define GBIF_CLIENT_HALT_MASK             BIT(0)
+#define GBIF_ARB_HALT_MASK                BIT(1)
+
+static void a6xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu)
+{
+	struct msm_gpu *gpu = &adreno_gpu->base;
+
+	if (!a6xx_has_gbif(adreno_gpu)) {
+		gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0xf);
+		spin_until((gpu_read(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL1) &
+								0xf) == 0xf);
+		gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0);
+
+		return;
+	}
+
+	/* Halt new client requests on GBIF */
+	gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_CLIENT_HALT_MASK);
+	spin_until((gpu_read(gpu, REG_A6XX_GBIF_HALT_ACK) &
+			(GBIF_CLIENT_HALT_MASK)) == GBIF_CLIENT_HALT_MASK);
+
+	/* Halt all AXI requests on GBIF */
+	gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_ARB_HALT_MASK);
+	spin_until((gpu_read(gpu,  REG_A6XX_GBIF_HALT_ACK) &
+			(GBIF_ARB_HALT_MASK)) == GBIF_ARB_HALT_MASK);
+
+	/* The GBIF halt needs to be explicitly cleared */
+	gpu_write(gpu, REG_A6XX_GBIF_HALT, 0x0);
+}
+
 /* Gracefully try to shut down the GMU and by extension the GPU */
 static void a6xx_gmu_shutdown(struct a6xx_gmu *gmu)
 {
 	struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu);
 	struct adreno_gpu *adreno_gpu = &a6xx_gpu->base;
-	struct msm_gpu *gpu = &adreno_gpu->base;
 	u32 val;
 
 	/*
@@ -819,11 +848,7 @@ static void a6xx_gmu_shutdown(struct a6xx_gmu *gmu)
 			return;
 		}
 
-		/* Clear the VBIF pipe before shutting down */
-		gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0xf);
-		spin_until((gpu_read(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL1) & 0xf)
-			== 0xf);
-		gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0);
+		a6xx_bus_clear_pending_transactions(adreno_gpu);
 
 		/* tell the GMU we want to slumber */
 		a6xx_gmu_notify_slumber(gmu);
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index daf07800cde02..68af24150de57 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -378,18 +378,6 @@ static int a6xx_hw_init(struct msm_gpu *gpu)
 	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
 	int ret;
 
-	/*
-	 * During a previous slumber, GBIF halt is asserted to ensure
-	 * no further transaction can go through GPU before GPU
-	 * headswitch is turned off.
-	 *
-	 * This halt is deasserted once headswitch goes off but
-	 * incase headswitch doesn't goes off clear GBIF halt
-	 * here to ensure GPU wake-up doesn't fail because of
-	 * halted GPU transactions.
-	 */
-	gpu_write(gpu, REG_A6XX_GBIF_HALT, 0x0);
-
 	/* Make sure the GMU keeps the GPU on while we set it up */
 	a6xx_gmu_set_oob(&a6xx_gpu->gmu, GMU_OOB_GPU_SET);
 
@@ -470,10 +458,12 @@ static int a6xx_hw_init(struct msm_gpu *gpu)
 	/* Select CP0 to always count cycles */
 	gpu_write(gpu, REG_A6XX_CP_PERFCTR_CP_SEL_0, PERF_CP_ALWAYS_COUNT);
 
-	gpu_write(gpu, REG_A6XX_RB_NC_MODE_CNTL, 2 << 1);
-	gpu_write(gpu, REG_A6XX_TPL1_NC_MODE_CNTL, 2 << 1);
-	gpu_write(gpu, REG_A6XX_SP_NC_MODE_CNTL, 2 << 1);
-	gpu_write(gpu, REG_A6XX_UCHE_MODE_CNTL, 2 << 21);
+	if (adreno_is_a630(adreno_gpu)) {
+		gpu_write(gpu, REG_A6XX_RB_NC_MODE_CNTL, 2 << 1);
+		gpu_write(gpu, REG_A6XX_TPL1_NC_MODE_CNTL, 2 << 1);
+		gpu_write(gpu, REG_A6XX_SP_NC_MODE_CNTL, 2 << 1);
+		gpu_write(gpu, REG_A6XX_UCHE_MODE_CNTL, 2 << 21);
+	}
 
 	/* Enable fault detection */
 	gpu_write(gpu, REG_A6XX_RBBM_INTERFACE_HANG_INT_CNTL,
@@ -748,39 +738,6 @@ static const u32 a6xx_register_offsets[REG_ADRENO_REGISTER_MAX] = {
 	REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_CNTL, REG_A6XX_CP_RB_CNTL),
 };
 
-#define GBIF_CLIENT_HALT_MASK             BIT(0)
-#define GBIF_ARB_HALT_MASK                BIT(1)
-
-static void a6xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu)
-{
-	struct msm_gpu *gpu = &adreno_gpu->base;
-
-	if(!a6xx_has_gbif(adreno_gpu)){
-		gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0xf);
-		spin_until((gpu_read(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL1) &
-								0xf) == 0xf);
-		gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0);
-
-		return;
-	}
-
-	/* Halt new client requests on GBIF */
-	gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_CLIENT_HALT_MASK);
-	spin_until((gpu_read(gpu, REG_A6XX_GBIF_HALT_ACK) &
-			(GBIF_CLIENT_HALT_MASK)) == GBIF_CLIENT_HALT_MASK);
-
-	/* Halt all AXI requests on GBIF */
-	gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_ARB_HALT_MASK);
-	spin_until((gpu_read(gpu,  REG_A6XX_GBIF_HALT_ACK) &
-			(GBIF_ARB_HALT_MASK)) == GBIF_ARB_HALT_MASK);
-
-	/*
-	 * GMU needs DDR access in slumber path. Deassert GBIF halt now
-	 * to allow for GMU to access system memory.
-	 */
-	gpu_write(gpu, REG_A6XX_GBIF_HALT, 0x0);
-}
-
 static int a6xx_pm_resume(struct msm_gpu *gpu)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
@@ -805,16 +762,6 @@ static int a6xx_pm_suspend(struct msm_gpu *gpu)
 
 	devfreq_suspend_device(gpu->devfreq.devfreq);
 
-	/*
-	 * Make sure the GMU is idle before continuing (because some transitions
-	 * may use VBIF
-	 */
-	a6xx_gmu_wait_for_idle(&a6xx_gpu->gmu);
-
-	/* Clear the VBIF pipe before shutting down */
-	/* FIXME: This accesses the GPU - do we need to make sure it is on? */
-	a6xx_bus_clear_pending_transactions(adreno_gpu);
-
 	return a6xx_gmu_stop(a6xx_gpu);
 }
 
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c
index eda11abc5f011..e450e0b972115 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c
@@ -7,6 +7,7 @@
 
 #include "a6xx_gmu.h"
 #include "a6xx_gmu.xml.h"
+#include "a6xx_gpu.h"
 
 #define HFI_MSG_ID(val) [val] = #val
 
@@ -216,48 +217,82 @@ static int a6xx_hfi_send_perf_table(struct a6xx_gmu *gmu)
 		NULL, 0);
 }
 
-static int a6xx_hfi_send_bw_table(struct a6xx_gmu *gmu)
+static void a618_build_bw_table(struct a6xx_hfi_msg_bw_table *msg)
 {
-	struct a6xx_hfi_msg_bw_table msg = { 0 };
+	/* Send a single "off" entry since the 618 GMU doesn't do bus scaling */
+	msg->bw_level_num = 1;
+
+	msg->ddr_cmds_num = 3;
+	msg->ddr_wait_bitmask = 0x01;
+
+	msg->ddr_cmds_addrs[0] = 0x50000;
+	msg->ddr_cmds_addrs[1] = 0x5003c;
+	msg->ddr_cmds_addrs[2] = 0x5000c;
+
+	msg->ddr_cmds_data[0][0] =  0x40000000;
+	msg->ddr_cmds_data[0][1] =  0x40000000;
+	msg->ddr_cmds_data[0][2] =  0x40000000;
 
 	/*
-	 * The sdm845 GMU doesn't do bus frequency scaling on its own but it
-	 * does need at least one entry in the list because it might be accessed
-	 * when the GMU is shutting down. Send a single "off" entry.
+	 * These are the CX (CNOC) votes - these are used by the GMU but the
+	 * votes are known and fixed for the target
 	 */
+	msg->cnoc_cmds_num = 1;
+	msg->cnoc_wait_bitmask = 0x01;
+
+	msg->cnoc_cmds_addrs[0] = 0x5007c;
+	msg->cnoc_cmds_data[0][0] =  0x40000000;
+	msg->cnoc_cmds_data[1][0] =  0x60000001;
+}
 
-	msg.bw_level_num = 1;
+static void a6xx_build_bw_table(struct a6xx_hfi_msg_bw_table *msg)
+{
+	/* Send a single "off" entry since the 630 GMU doesn't do bus scaling */
+	msg->bw_level_num = 1;
 
-	msg.ddr_cmds_num = 3;
-	msg.ddr_wait_bitmask = 0x07;
+	msg->ddr_cmds_num = 3;
+	msg->ddr_wait_bitmask = 0x07;
 
-	msg.ddr_cmds_addrs[0] = 0x50000;
-	msg.ddr_cmds_addrs[1] = 0x5005c;
-	msg.ddr_cmds_addrs[2] = 0x5000c;
+	msg->ddr_cmds_addrs[0] = 0x50000;
+	msg->ddr_cmds_addrs[1] = 0x5005c;
+	msg->ddr_cmds_addrs[2] = 0x5000c;
 
-	msg.ddr_cmds_data[0][0] =  0x40000000;
-	msg.ddr_cmds_data[0][1] =  0x40000000;
-	msg.ddr_cmds_data[0][2] =  0x40000000;
+	msg->ddr_cmds_data[0][0] =  0x40000000;
+	msg->ddr_cmds_data[0][1] =  0x40000000;
+	msg->ddr_cmds_data[0][2] =  0x40000000;
 
 	/*
 	 * These are the CX (CNOC) votes.  This is used but the values for the
 	 * sdm845 GMU are known and fixed so we can hard code them.
 	 */
 
-	msg.cnoc_cmds_num = 3;
-	msg.cnoc_wait_bitmask = 0x05;
+	msg->cnoc_cmds_num = 3;
+	msg->cnoc_wait_bitmask = 0x05;
 
-	msg.cnoc_cmds_addrs[0] = 0x50034;
-	msg.cnoc_cmds_addrs[1] = 0x5007c;
-	msg.cnoc_cmds_addrs[2] = 0x5004c;
+	msg->cnoc_cmds_addrs[0] = 0x50034;
+	msg->cnoc_cmds_addrs[1] = 0x5007c;
+	msg->cnoc_cmds_addrs[2] = 0x5004c;
 
-	msg.cnoc_cmds_data[0][0] =  0x40000000;
-	msg.cnoc_cmds_data[0][1] =  0x00000000;
-	msg.cnoc_cmds_data[0][2] =  0x40000000;
+	msg->cnoc_cmds_data[0][0] =  0x40000000;
+	msg->cnoc_cmds_data[0][1] =  0x00000000;
+	msg->cnoc_cmds_data[0][2] =  0x40000000;
+
+	msg->cnoc_cmds_data[1][0] =  0x60000001;
+	msg->cnoc_cmds_data[1][1] =  0x20000001;
+	msg->cnoc_cmds_data[1][2] =  0x60000001;
+}
+
+
+static int a6xx_hfi_send_bw_table(struct a6xx_gmu *gmu)
+{
+	struct a6xx_hfi_msg_bw_table msg = { 0 };
+	struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu);
+	struct adreno_gpu *adreno_gpu = &a6xx_gpu->base;
 
-	msg.cnoc_cmds_data[1][0] =  0x60000001;
-	msg.cnoc_cmds_data[1][1] =  0x20000001;
-	msg.cnoc_cmds_data[1][2] =  0x60000001;
+	if (adreno_is_a618(adreno_gpu))
+		a618_build_bw_table(&msg);
+	else
+		a6xx_build_bw_table(&msg);
 
 	return a6xx_hfi_send_msg(gmu, HFI_H2F_MSG_BW_TABLE, &msg, sizeof(msg),
 		NULL, 0);
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c
index 528632690f1ef..a05282dede91b 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c
@@ -255,13 +255,13 @@ static const struct dpu_format dpu_format_map[] = {
 
 	INTERLEAVED_RGB_FMT(RGB565,
 		0, COLOR_5BIT, COLOR_6BIT, COLOR_5BIT,
-		C2_R_Cr, C0_G_Y, C1_B_Cb, 0, 3,
+		C1_B_Cb, C0_G_Y, C2_R_Cr, 0, 3,
 		false, 2, 0,
 		DPU_FETCH_LINEAR, 1),
 
 	INTERLEAVED_RGB_FMT(BGR565,
 		0, COLOR_5BIT, COLOR_6BIT, COLOR_5BIT,
-		C1_B_Cb, C0_G_Y, C2_R_Cr, 0, 3,
+		C2_R_Cr, C0_G_Y, C1_B_Cb, 0, 3,
 		false, 2, 0,
 		DPU_FETCH_LINEAR, 1),
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c
index 29705e773a4b7..80d3cfc140070 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c
@@ -12,6 +12,7 @@
 
 #define to_dpu_mdss(x) container_of(x, struct dpu_mdss, base)
 
+#define HW_REV				0x0
 #define HW_INTR_STATUS			0x0010
 
 /* Max BW defined in KBps */
@@ -22,6 +23,17 @@ struct dpu_irq_controller {
 	struct irq_domain *domain;
 };
 
+struct dpu_hw_cfg {
+	u32 val;
+	u32 offset;
+};
+
+struct dpu_mdss_hw_init_handler {
+	u32 hw_rev;
+	u32 hw_reg_count;
+	struct dpu_hw_cfg* hw_cfg;
+};
+
 struct dpu_mdss {
 	struct msm_mdss base;
 	void __iomem *mmio;
@@ -32,6 +44,44 @@ struct dpu_mdss {
 	u32 num_paths;
 };
 
+static struct dpu_hw_cfg hw_cfg[] = {
+    {
+	/* UBWC global settings */
+	.val = 0x1E,
+	.offset = 0x144,
+    }
+};
+
+static struct dpu_mdss_hw_init_handler cfg_handler[] = {
+    { .hw_rev = DPU_HW_VER_620,
+      .hw_reg_count = ARRAY_SIZE(hw_cfg),
+      .hw_cfg = hw_cfg
+    },
+};
+
+static void dpu_mdss_hw_init(struct dpu_mdss *dpu_mdss, u32 hw_rev)
+{
+	int i;
+	u32 count = 0;
+	struct dpu_hw_cfg *hw_cfg = NULL;
+
+	for (i = 0; i < ARRAY_SIZE(cfg_handler); i++) {
+		if (cfg_handler[i].hw_rev == hw_rev) {
+			hw_cfg = cfg_handler[i].hw_cfg;
+			count = cfg_handler[i].hw_reg_count;
+			break;
+	    }
+	}
+
+	for (i = 0; i < count; i++ ) {
+		writel_relaxed(hw_cfg->val,
+			dpu_mdss->mmio + hw_cfg->offset);
+		hw_cfg++;
+	}
+
+    return;
+}
+
 static int dpu_mdss_parse_data_bus_icc_path(struct drm_device *dev,
 						struct dpu_mdss *dpu_mdss)
 {
@@ -174,12 +224,18 @@ static int dpu_mdss_enable(struct msm_mdss *mdss)
 	struct dpu_mdss *dpu_mdss = to_dpu_mdss(mdss);
 	struct dss_module_power *mp = &dpu_mdss->mp;
 	int ret;
+	u32 mdss_rev;
 
 	dpu_mdss_icc_request_bw(mdss);
 
 	ret = msm_dss_enable_clk(mp->clk_config, mp->num_clk, true);
-	if (ret)
+	if (ret) {
 		DPU_ERROR("clock enable failed, ret:%d\n", ret);
+		return ret;
+	}
+
+	mdss_rev = readl_relaxed(dpu_mdss->mmio + HW_REV);
+	dpu_mdss_hw_init(dpu_mdss, mdss_rev);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c
index 05cc04f729d63..e1cc541e0ef2e 100644
--- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c
+++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c
@@ -1109,8 +1109,8 @@ static void mdp5_crtc_wait_for_pp_done(struct drm_crtc *crtc)
 	ret = wait_for_completion_timeout(&mdp5_crtc->pp_completion,
 						msecs_to_jiffies(50));
 	if (ret == 0)
-		dev_warn(dev->dev, "pp done time out, lm=%d\n",
-			 mdp5_cstate->pipeline.mixer->lm);
+		dev_warn_ratelimited(dev->dev, "pp done time out, lm=%d\n",
+				     mdp5_cstate->pipeline.mixer->lm);
 }
 
 static void mdp5_crtc_wait_for_flush_done(struct drm_crtc *crtc)
diff --git a/drivers/gpu/drm/msm/dsi/dsi_manager.c b/drivers/gpu/drm/msm/dsi/dsi_manager.c
index 104115d112eba..4864b9558f65a 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_manager.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_manager.c
@@ -336,7 +336,7 @@ static int dsi_mgr_connector_get_modes(struct drm_connector *connector)
 	return num;
 }
 
-static int dsi_mgr_connector_mode_valid(struct drm_connector *connector,
+static enum drm_mode_status dsi_mgr_connector_mode_valid(struct drm_connector *connector,
 				struct drm_display_mode *mode)
 {
 	int id = dsi_mgr_connector_get_id(connector);
@@ -506,6 +506,7 @@ static void dsi_mgr_bridge_post_disable(struct drm_bridge *bridge)
 	struct msm_dsi *msm_dsi1 = dsi_mgr_get_dsi(DSI_1);
 	struct mipi_dsi_host *host = msm_dsi->host;
 	struct drm_panel *panel = msm_dsi->panel;
+	struct msm_dsi_pll *src_pll;
 	bool is_dual_dsi = IS_DUAL_DSI();
 	int ret;
 
@@ -539,6 +540,10 @@ static void dsi_mgr_bridge_post_disable(struct drm_bridge *bridge)
 								id, ret);
 	}
 
+	/* Save PLL status if it is a clock source */
+	src_pll = msm_dsi_phy_get_pll(msm_dsi->phy);
+	msm_dsi_pll_save_state(src_pll);
+
 	ret = msm_dsi_host_power_off(host);
 	if (ret)
 		pr_err("%s: host %d power off failed,%d\n", __func__, id, ret);
diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c
index b0cfa67d2a578..f509ebd77500f 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c
@@ -724,10 +724,6 @@ void msm_dsi_phy_disable(struct msm_dsi_phy *phy)
 	if (!phy || !phy->cfg->ops.disable)
 		return;
 
-	/* Save PLL status if it is a clock source */
-	if (phy->usecase != MSM_DSI_PHY_SLAVE)
-		msm_dsi_pll_save_state(phy->pll);
-
 	phy->cfg->ops.disable(phy);
 
 	dsi_phy_regulator_disable(phy);
diff --git a/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c
index 1c894548dd725..6ac04fc303f56 100644
--- a/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c
+++ b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c
@@ -411,6 +411,12 @@ static int dsi_pll_10nm_vco_prepare(struct clk_hw *hw)
 	if (pll_10nm->slave)
 		dsi_pll_enable_pll_bias(pll_10nm->slave);
 
+	rc = dsi_pll_10nm_vco_set_rate(hw,pll_10nm->vco_current_rate, 0);
+	if (rc) {
+		pr_err("vco_set_rate failed, rc=%d\n", rc);
+		return rc;
+	}
+
 	/* Start PLL */
 	pll_write(pll_10nm->phy_cmn_mmio + REG_DSI_10nm_PHY_CMN_PLL_CNTRL,
 		  0x01);
diff --git a/drivers/gpu/drm/nouveau/dispnv50/wndw.c b/drivers/gpu/drm/nouveau/dispnv50/wndw.c
index 890315291b01e..bb737f9281e69 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/wndw.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/wndw.c
@@ -458,6 +458,8 @@ nv50_wndw_atomic_check(struct drm_plane *plane, struct drm_plane_state *state)
 		asyw->clr.ntfy = armw->ntfy.handle != 0;
 		asyw->clr.sema = armw->sema.handle != 0;
 		asyw->clr.xlut = armw->xlut.handle != 0;
+		if (asyw->clr.xlut && asyw->visible)
+			asyw->set.xlut = asyw->xlut.handle != 0;
 		asyw->clr.csc  = armw->csc.valid;
 		if (wndw->func->image_clr)
 			asyw->clr.image = armw->image.handle[0] != 0;
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
index c7d700916eae7..8ebbe16560083 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
@@ -2579,6 +2579,7 @@ nv166_chipset = {
 static const struct nvkm_device_chip
 nv167_chipset = {
 	.name = "TU117",
+	.acr = tu102_acr_new,
 	.bar = tu102_bar_new,
 	.bios = nvkm_bios_new,
 	.bus = gf100_bus_new,
@@ -2607,6 +2608,7 @@ nv167_chipset = {
 	.disp = tu102_disp_new,
 	.dma = gv100_dma_new,
 	.fifo = tu102_fifo_new,
+	.gr = tu102_gr_new,
 	.nvdec[0] = gm107_nvdec_new,
 	.nvenc[0] = gm107_nvenc_new,
 	.sec2 = tu102_sec2_new,
@@ -2615,6 +2617,7 @@ nv167_chipset = {
 static const struct nvkm_device_chip
 nv168_chipset = {
 	.name = "TU116",
+	.acr = tu102_acr_new,
 	.bar = tu102_bar_new,
 	.bios = nvkm_bios_new,
 	.bus = gf100_bus_new,
@@ -2643,6 +2646,7 @@ nv168_chipset = {
 	.disp = tu102_disp_new,
 	.dma = gv100_dma_new,
 	.fifo = tu102_fifo_new,
+	.gr = tu102_gr_new,
 	.nvdec[0] = gm107_nvdec_new,
 	.nvenc[0] = gm107_nvenc_new,
 	.sec2 = tu102_sec2_new,
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/tu102.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/tu102.c
index 454668b1cf54d..a9efa4d78be92 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/tu102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/tu102.c
@@ -164,6 +164,32 @@ MODULE_FIRMWARE("nvidia/tu106/gr/sw_nonctx.bin");
 MODULE_FIRMWARE("nvidia/tu106/gr/sw_bundle_init.bin");
 MODULE_FIRMWARE("nvidia/tu106/gr/sw_method_init.bin");
 
+MODULE_FIRMWARE("nvidia/tu117/gr/fecs_bl.bin");
+MODULE_FIRMWARE("nvidia/tu117/gr/fecs_inst.bin");
+MODULE_FIRMWARE("nvidia/tu117/gr/fecs_data.bin");
+MODULE_FIRMWARE("nvidia/tu117/gr/fecs_sig.bin");
+MODULE_FIRMWARE("nvidia/tu117/gr/gpccs_bl.bin");
+MODULE_FIRMWARE("nvidia/tu117/gr/gpccs_inst.bin");
+MODULE_FIRMWARE("nvidia/tu117/gr/gpccs_data.bin");
+MODULE_FIRMWARE("nvidia/tu117/gr/gpccs_sig.bin");
+MODULE_FIRMWARE("nvidia/tu117/gr/sw_ctx.bin");
+MODULE_FIRMWARE("nvidia/tu117/gr/sw_nonctx.bin");
+MODULE_FIRMWARE("nvidia/tu117/gr/sw_bundle_init.bin");
+MODULE_FIRMWARE("nvidia/tu117/gr/sw_method_init.bin");
+
+MODULE_FIRMWARE("nvidia/tu116/gr/fecs_bl.bin");
+MODULE_FIRMWARE("nvidia/tu116/gr/fecs_inst.bin");
+MODULE_FIRMWARE("nvidia/tu116/gr/fecs_data.bin");
+MODULE_FIRMWARE("nvidia/tu116/gr/fecs_sig.bin");
+MODULE_FIRMWARE("nvidia/tu116/gr/gpccs_bl.bin");
+MODULE_FIRMWARE("nvidia/tu116/gr/gpccs_inst.bin");
+MODULE_FIRMWARE("nvidia/tu116/gr/gpccs_data.bin");
+MODULE_FIRMWARE("nvidia/tu116/gr/gpccs_sig.bin");
+MODULE_FIRMWARE("nvidia/tu116/gr/sw_ctx.bin");
+MODULE_FIRMWARE("nvidia/tu116/gr/sw_nonctx.bin");
+MODULE_FIRMWARE("nvidia/tu116/gr/sw_bundle_init.bin");
+MODULE_FIRMWARE("nvidia/tu116/gr/sw_method_init.bin");
+
 static const struct gf100_gr_fwif
 tu102_gr_fwif[] = {
 	{ 0, gm200_gr_load, &tu102_gr, &gp108_gr_fecs_acr, &gp108_gr_gpccs_acr },
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/acr/tu102.c b/drivers/gpu/drm/nouveau/nvkm/subdev/acr/tu102.c
index 7f4b89d82d320..d28d8f36ae248 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/acr/tu102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/acr/tu102.c
@@ -107,6 +107,12 @@ MODULE_FIRMWARE("nvidia/tu104/acr/ucode_unload.bin");
 MODULE_FIRMWARE("nvidia/tu106/acr/unload_bl.bin");
 MODULE_FIRMWARE("nvidia/tu106/acr/ucode_unload.bin");
 
+MODULE_FIRMWARE("nvidia/tu116/acr/unload_bl.bin");
+MODULE_FIRMWARE("nvidia/tu116/acr/ucode_unload.bin");
+
+MODULE_FIRMWARE("nvidia/tu117/acr/unload_bl.bin");
+MODULE_FIRMWARE("nvidia/tu117/acr/ucode_unload.bin");
+
 static const struct nvkm_acr_hsf_fwif
 tu102_acr_unload_fwif[] = {
 	{  0, nvkm_acr_hsfw_load, &gp108_acr_unload_0 },
@@ -130,6 +136,8 @@ tu102_acr_asb_0 = {
 MODULE_FIRMWARE("nvidia/tu102/acr/ucode_asb.bin");
 MODULE_FIRMWARE("nvidia/tu104/acr/ucode_asb.bin");
 MODULE_FIRMWARE("nvidia/tu106/acr/ucode_asb.bin");
+MODULE_FIRMWARE("nvidia/tu116/acr/ucode_asb.bin");
+MODULE_FIRMWARE("nvidia/tu117/acr/ucode_asb.bin");
 
 static const struct nvkm_acr_hsf_fwif
 tu102_acr_asb_fwif[] = {
@@ -154,6 +162,12 @@ MODULE_FIRMWARE("nvidia/tu104/acr/ucode_ahesasc.bin");
 MODULE_FIRMWARE("nvidia/tu106/acr/bl.bin");
 MODULE_FIRMWARE("nvidia/tu106/acr/ucode_ahesasc.bin");
 
+MODULE_FIRMWARE("nvidia/tu116/acr/bl.bin");
+MODULE_FIRMWARE("nvidia/tu116/acr/ucode_ahesasc.bin");
+
+MODULE_FIRMWARE("nvidia/tu117/acr/bl.bin");
+MODULE_FIRMWARE("nvidia/tu117/acr/ucode_ahesasc.bin");
+
 static const struct nvkm_acr_hsf_fwif
 tu102_acr_ahesasc_fwif[] = {
 	{  0, nvkm_acr_hsfw_load, &tu102_acr_ahesasc_0 },
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gv100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gv100.c
index 389bad312bf27..10ff5d053f7ea 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gv100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gv100.c
@@ -51,3 +51,5 @@ MODULE_FIRMWARE("nvidia/gv100/nvdec/scrubber.bin");
 MODULE_FIRMWARE("nvidia/tu102/nvdec/scrubber.bin");
 MODULE_FIRMWARE("nvidia/tu104/nvdec/scrubber.bin");
 MODULE_FIRMWARE("nvidia/tu106/nvdec/scrubber.bin");
+MODULE_FIRMWARE("nvidia/tu116/nvdec/scrubber.bin");
+MODULE_FIRMWARE("nvidia/tu117/nvdec/scrubber.bin");
diff --git a/drivers/gpu/drm/panfrost/panfrost_job.c b/drivers/gpu/drm/panfrost/panfrost_job.c
index 7157dfd7dea3c..9a1a72a748e72 100644
--- a/drivers/gpu/drm/panfrost/panfrost_job.c
+++ b/drivers/gpu/drm/panfrost/panfrost_job.c
@@ -280,12 +280,8 @@ static void panfrost_job_cleanup(struct kref *ref)
 	}
 
 	if (job->bos) {
-		struct panfrost_gem_object *bo;
-
-		for (i = 0; i < job->bo_count; i++) {
-			bo = to_panfrost_bo(job->bos[i]);
+		for (i = 0; i < job->bo_count; i++)
 			drm_gem_object_put_unlocked(job->bos[i]);
-		}
 
 		kvfree(job->bos);
 	}
diff --git a/drivers/gpu/drm/panfrost/panfrost_mmu.c b/drivers/gpu/drm/panfrost/panfrost_mmu.c
index 763cfca886a73..3107b0738e401 100644
--- a/drivers/gpu/drm/panfrost/panfrost_mmu.c
+++ b/drivers/gpu/drm/panfrost/panfrost_mmu.c
@@ -151,7 +151,12 @@ u32 panfrost_mmu_as_get(struct panfrost_device *pfdev, struct panfrost_mmu *mmu)
 	as = mmu->as;
 	if (as >= 0) {
 		int en = atomic_inc_return(&mmu->as_count);
-		WARN_ON(en >= NUM_JOB_SLOTS);
+
+		/*
+		 * AS can be retained by active jobs or a perfcnt context,
+		 * hence the '+ 1' here.
+		 */
+		WARN_ON(en >= (NUM_JOB_SLOTS + 1));
 
 		list_move(&mmu->list, &pfdev->as_lru_list);
 		goto out;
diff --git a/drivers/gpu/drm/panfrost/panfrost_perfcnt.c b/drivers/gpu/drm/panfrost/panfrost_perfcnt.c
index 684820448be31..6913578d5aa72 100644
--- a/drivers/gpu/drm/panfrost/panfrost_perfcnt.c
+++ b/drivers/gpu/drm/panfrost/panfrost_perfcnt.c
@@ -73,7 +73,7 @@ static int panfrost_perfcnt_enable_locked(struct panfrost_device *pfdev,
 	struct panfrost_file_priv *user = file_priv->driver_priv;
 	struct panfrost_perfcnt *perfcnt = pfdev->perfcnt;
 	struct drm_gem_shmem_object *bo;
-	u32 cfg;
+	u32 cfg, as;
 	int ret;
 
 	if (user == perfcnt->user)
@@ -126,12 +126,8 @@ static int panfrost_perfcnt_enable_locked(struct panfrost_device *pfdev,
 
 	perfcnt->user = user;
 
-	/*
-	 * Always use address space 0 for now.
-	 * FIXME: this needs to be updated when we start using different
-	 * address space.
-	 */
-	cfg = GPU_PERFCNT_CFG_AS(0) |
+	as = panfrost_mmu_as_get(pfdev, perfcnt->mapping->mmu);
+	cfg = GPU_PERFCNT_CFG_AS(as) |
 	      GPU_PERFCNT_CFG_MODE(GPU_PERFCNT_CFG_MODE_MANUAL);
 
 	/*
@@ -195,6 +191,7 @@ static int panfrost_perfcnt_disable_locked(struct panfrost_device *pfdev,
 	drm_gem_shmem_vunmap(&perfcnt->mapping->obj->base.base, perfcnt->buf);
 	perfcnt->buf = NULL;
 	panfrost_gem_close(&perfcnt->mapping->obj->base.base, file_priv);
+	panfrost_mmu_as_put(pfdev, perfcnt->mapping->mmu);
 	panfrost_gem_mapping_put(perfcnt->mapping);
 	perfcnt->mapping = NULL;
 	pm_runtime_mark_last_busy(pfdev->dev);
diff --git a/drivers/gpu/drm/selftests/drm_cmdline_selftests.h b/drivers/gpu/drm/selftests/drm_cmdline_selftests.h
index ceac7af9a172d..29e367db6118b 100644
--- a/drivers/gpu/drm/selftests/drm_cmdline_selftests.h
+++ b/drivers/gpu/drm/selftests/drm_cmdline_selftests.h
@@ -53,6 +53,7 @@ cmdline_test(drm_cmdline_test_rotate_0)
 cmdline_test(drm_cmdline_test_rotate_90)
 cmdline_test(drm_cmdline_test_rotate_180)
 cmdline_test(drm_cmdline_test_rotate_270)
+cmdline_test(drm_cmdline_test_rotate_multiple)
 cmdline_test(drm_cmdline_test_rotate_invalid_val)
 cmdline_test(drm_cmdline_test_rotate_truncated)
 cmdline_test(drm_cmdline_test_hmirror)
diff --git a/drivers/gpu/drm/selftests/test-drm_cmdline_parser.c b/drivers/gpu/drm/selftests/test-drm_cmdline_parser.c
index 520f3e66a384a..d96cd890def6e 100644
--- a/drivers/gpu/drm/selftests/test-drm_cmdline_parser.c
+++ b/drivers/gpu/drm/selftests/test-drm_cmdline_parser.c
@@ -856,6 +856,17 @@ static int drm_cmdline_test_rotate_270(void *ignored)
 	return 0;
 }
 
+static int drm_cmdline_test_rotate_multiple(void *ignored)
+{
+	struct drm_cmdline_mode mode = { };
+
+	FAIL_ON(drm_mode_parse_command_line_for_connector("720x480,rotate=0,rotate=90",
+							  &no_connector,
+							  &mode));
+
+	return 0;
+}
+
 static int drm_cmdline_test_rotate_invalid_val(void *ignored)
 {
 	struct drm_cmdline_mode mode = { };
@@ -888,7 +899,7 @@ static int drm_cmdline_test_hmirror(void *ignored)
 	FAIL_ON(!mode.specified);
 	FAIL_ON(mode.xres != 720);
 	FAIL_ON(mode.yres != 480);
-	FAIL_ON(mode.rotation_reflection != DRM_MODE_REFLECT_X);
+	FAIL_ON(mode.rotation_reflection != (DRM_MODE_ROTATE_0 | DRM_MODE_REFLECT_X));
 
 	FAIL_ON(mode.refresh_specified);
 
@@ -913,7 +924,7 @@ static int drm_cmdline_test_vmirror(void *ignored)
 	FAIL_ON(!mode.specified);
 	FAIL_ON(mode.xres != 720);
 	FAIL_ON(mode.yres != 480);
-	FAIL_ON(mode.rotation_reflection != DRM_MODE_REFLECT_Y);
+	FAIL_ON(mode.rotation_reflection != (DRM_MODE_ROTATE_0 | DRM_MODE_REFLECT_Y));
 
 	FAIL_ON(mode.refresh_specified);