drm next for 6.12-rc1

string:
 - add mem_is_zero()
 
 core:
 - support more device numbers
 - use XArray for minor ids
 - add backlight constants
 - Split dma fence array creation into alloc and arm
 
 fbdev:
 - remove usage of old fbdev hooks
 
 kms:
 - Add might_fault() to drm_modeset_lock priming
 - Add dynamic per-crtc vblank configuration support
 
 dma-buf:
 - docs cleanup
 
 buddy:
 - Add start address support for trim function
 
 printk:
 - pass description to kmsg_dump
 
 scheduler;
 - Remove full_recover from drm_sched_start
 
 ttm:
 - Make LRU walk restartable after dropping locks
 - Allow direct reclaim to allocate local memory
 
 panic:
 - add display QR code (in rust)
 
 displayport:
 - mst: GUID improvements
 
 bridge:
 - Silence error message on -EPROBE_DEFER
 - analogix: Clean aup
 - bridge-connector: Fix double free
 - lt6505: Disable interrupt when powered off
 - tc358767: Make default DP port preemphasis configurable
 - lt9611uxc: require DRM_BRIDGE_ATTACH_NO_CONNECTOR
 - anx7625: simplify OF array handling
 - dw-hdmi: simplify clock handling
 - lontium-lt8912b: fix mode validation
 - nwl-dsi: fix mode vsync/hsync polarity
 
 xe:
 - Enable LunarLake and Battlemage support
 - Introducing Xe2 ccs modifiers for integrated and discrete graphics
 - rename xe perf to xe observation
 - use wb caching on DGFX for system memory
 - add fence timeouts
 - Lunar Lake graphics/media/display workarounds
 - Battlemage workarounds
 - Battlemage GSC support
 - GSC and HuC fw updates for LL/BM
 - use dma_fence_chain_free
 - refactor hw engine lookup and mmio access
 - enable priority mem read for Xe2
 - Add first GuC BMG fw
 - fix dma-resv lock
 - Fix DGFX display suspend/resume
 - Use xe_managed for kernel BOs
 - Use reserved copy engine for user binds on faulting devices
 - Allow mixing dma-fence jobs and long-running faulting jobs
 - fix media TLB invalidation
 - fix rpm in TTM swapout path
 - track resources and VF state by PF
 
 i915:
 - Type-C programming fix for MTL+
 - FBC cleanup
 - Calc vblank delay more accurately
 - On DP MST, Enable LT fallback for UHBR<->non-UHBR rates
 - Fix DP LTTPR detection
 - limit relocations to INT_MAX
 - fix long hangs in buddy allocator on DG2/A380
 
 amdgpu:
 - Per-queue reset support
 - SDMA devcoredump support
 - DCN 4.0.1 updates
 - GFX12/VCN4/JPEG4 updates
 - Convert vbios embedded EDID to drm_edid
 - GFX9.3/9.4 devcoredump support
 - process isolation framework for GFX 9.4.3/4
 - take IOMMU mappings into account for P2P DMA
 
 amdkfd:
 - CRIU fixes
 - HMM fix
 - Enable process isolation support for GFX 9.4.3/4
 - Allow users to target recommended SDMA engines
 - KFD support for targetting queues on recommended SDMA engines
 
 radeon:
 - remove .load and drm_dev_alloc
 - Fix vbios embedded EDID size handling
 - Convert vbios embedded EDID to drm_edid
 - Use GEM references instead of TTM
 - r100 cp init cleanup
 - Fix potential overflows in evergreen CS offset tracking
 
 msm:
 - DPU:
 - implement DP/PHY mapping on SC8180X
 - Enable writeback on SM8150, SC8180X, SM6125, SM6350
 - DP:
 - Enable widebus on all relevant chipsets
 - MSM8998 HDMI support
 - GPU:
 - A642L speedbin support
 - A615/A306/A621 support
 - A7xx devcoredump support
 
 ast:
 - astdp: Support AST2600 with VGA
 - Clean up HPD
 - Fix timeout loop for DP link training
 - reorganize output code by type (VGA, DP, etc)
 - convert to struct drm_edid
 - fix BMC handling for all outputs
 
 exynos:
 - drop stale MAINTAINERS pattern
 - constify struct
 
 loongson:
 - use GEM refcount over TTM
 
 mgag200:
 - Improve BMC handling
 - Support VBLANK intterupts
 - transparently support BMC outputs
 
 nouveau:
 - Refactor and clean up internals
 - Use GEM refcount over TTM's
 
 gm12u320:
 - convert to struct drm_edid
 
 gma500:
 - update i2c terms
 
 lcdif:
 - pixel clock fix
 
 host1x:
 - fix syncpoint IRQ during resume
 - use iommu_paging_domain_alloc()
 
 imx:
 - ipuv3: convert to struct drm_edid
 
 omapdrm:
 - improve error handling
 - use common helper for_each_endpoint_of_node()
 
 panel:
 - add support for BOE TV101WUM-LL2 plus DT bindings
 - novatek-nt35950: improve error handling
 - nv3051d: improve error handling
 - panel-edp: add support for BOE NE140WUM-N6G; revert support for
   SDC ATNA45AF01
 - visionox-vtdr6130: improve error handling; use
   devm_regulator_bulk_get_const()
 - boe-th101mb31ig002: Support for starry-er88577 MIPI-DSI panel plus
   DT; Fix porch parameter
 - edp: Support AOU B116XTN02.3, AUO B116XAN06.1, AOU B116XAT04.1,
   BOE NV140WUM-N41, BOE NV133WUM-N63, BOE NV116WHM-A4D, CMN N116BCA-EA2,
   CMN N116BCP-EA2, CSW MNB601LS1-4
 - himax-hx8394: Support Microchip AC40T08A MIPI Display panel plus DT
 - ilitek-ili9806e: Support Densitron DMT028VGHMCMI-1D TFT plus DT
 - jd9365da: Support Melfas lmfbx101117480 MIPI-DSI panel plus DT; Refactor
   for code sharing
 - panel-edp: fix name for HKC MB116AN01
 - jd9365da: fix "exit sleep" commands
 - jdi-fhd-r63452: simplify error handling with DSI multi-style
   helpers
 - mantix-mlaf057we51: simplify error handling with DSI multi-style
   helpers
 - simple:
   support Innolux G070ACE-LH3 plus DT bindings
   support On Tat Industrial Company KD50G21-40NT-A1 plus DT bindings
 - st7701:
   decouple DSI and DRM code
   add SPI support
   support Anbernic RG28XX plus DT bindings
 
 mediatek:
 - support alpha blending
 - remove cl in struct cmdq_pkt
 - ovl adaptor fix
 - add power domain binding for mediatek DPI controller
 
 renesas:
 - rz-du: add support for RZ/G2UL plus DT bindings
 
 rockchip:
 - Improve DP sink-capability reporting
 - dw_hdmi: Support 4k@60Hz
 - vop: Support RGB display on Rockchip RK3066; Support 4096px width
 
 sti:
 - convert to struct drm_edid
 
 stm:
 - Avoid UAF wih managed plane and CRTC helpers
 - Fix module owner
 - Fix error handling in probe
 - Depend on COMMON_CLK
 - ltdc: Fix transparency after disabling plane; Remove unused interrupt
 
 tegra:
 - gr3d: improve PM domain handling
 - convert to struct drm_edid
 - Call drm_atomic_helper_shutdown()
 
 vc4:
 - fix PM during detect
 - replace DRM_ERROR() with drm_error()
 - v3d: simplify clock retrieval
 
 v3d:
 - Clean up perfmon
 
 virtio:
 - add DRM capset
 -----BEGIN PGP SIGNATURE-----
 
 iQIzBAABCAAdFiEEEKbZHaGwW9KfbeusDHTzWXnEhr4FAmbq43gACgkQDHTzWXnE
 hr4+lg/+O/r41E7ioitcM0DWeWem0dTlvQr41pJ8jujHvw+bXNdg0BMGWtsTyTLA
 eOft2AwofsFjg+O7l8IFXOT37mQLdIdfjb3+w5brI198InL3OWC3QV8ZSwY9VGET
 n8crO9jFoxNmHZnFniBZbtI6egTyl6H+2ey3E0MTnKiPUKZQvsK/4+x532yVLPob
 UUOze5wcjyGZc7LJEIZPohPVneCb9ki7sabDQqh4cxIQ0Eg+nqPpWjYM4XVd+lTS
 8QmssbR49LrJ7z9m90qVE+8TjYUCn+ChDPMs61KZAAnc8k++nK41btjGZ23mDKPb
 YEguahCYthWJ4U8K18iXBPnLPxZv5+harQ8OIWAUYqdIOWSXHozvuJ2Z84eHV13a
 9mQ5vIymXang8G1nEXwX/vml9uhVhBCeWu3qfdse2jfaTWYUb1YzhqUoFvqI0R0K
 8wT03MyNdx965CSqAhpH5Jd559ueZmpd+jsHOfhAS+1gxfD6NgoPXv7lpnMUmGWX
 SnaeC9RLD4cgy7j2Swo7TEqQHrvK5XhZSwX94kU6RPmFE5RRKqWgFVQmwuikDMId
 UpNqDnPT5NL2UX4TNG4V4coyTXvKgVcSB9TA7j8NSLfwdGHhiz73pkYosaZXKyxe
 u6qKMwMONfZiT20nhD7RhH0AFnnKosAcO14dhn0TKFZPY6Ce9O8=
 =7jR+
 -----END PGP SIGNATURE-----

Merge tag 'drm-next-2024-09-19' of https://gitlab.freedesktop.org/drm/kernel

Pull drm updates from Dave Airlie:
 "This adds a couple of patches outside the drm core, all should be
  acked appropriately, the string and pstore ones are the main ones that
  come to mind.

  Otherwise it's the usual drivers, xe is getting enabled by default on
  some new hardware, we've changed the device number handling to allow
  more devices, and we added some optional rust code to create QR codes
  in the panic handler, an idea first suggested I think 10 years ago :-)

  string:
   - add mem_is_zero()

  core:
   - support more device numbers
   - use XArray for minor ids
   - add backlight constants
   - Split dma fence array creation into alloc and arm

  fbdev:
   - remove usage of old fbdev hooks

  kms:
   - Add might_fault() to drm_modeset_lock priming
   - Add dynamic per-crtc vblank configuration support

  dma-buf:
   - docs cleanup

  buddy:
   - Add start address support for trim function

  printk:
   - pass description to kmsg_dump

  scheduler:
   - Remove full_recover from drm_sched_start

  ttm:
   - Make LRU walk restartable after dropping locks
   - Allow direct reclaim to allocate local memory

  panic:
   - add display QR code (in rust)

  displayport:
   - mst: GUID improvements

  bridge:
   - Silence error message on -EPROBE_DEFER
   - analogix: Clean aup
   - bridge-connector: Fix double free
   - lt6505: Disable interrupt when powered off
   - tc358767: Make default DP port preemphasis configurable
   - lt9611uxc: require DRM_BRIDGE_ATTACH_NO_CONNECTOR
   - anx7625: simplify OF array handling
   - dw-hdmi: simplify clock handling
   - lontium-lt8912b: fix mode validation
   - nwl-dsi: fix mode vsync/hsync polarity

  xe:
   - Enable LunarLake and Battlemage support
   - Introducing Xe2 ccs modifiers for integrated and discrete graphics
   - rename xe perf to xe observation
   - use wb caching on DGFX for system memory
   - add fence timeouts
   - Lunar Lake graphics/media/display workarounds
   - Battlemage workarounds
   - Battlemage GSC support
   - GSC and HuC fw updates for LL/BM
   - use dma_fence_chain_free
   - refactor hw engine lookup and mmio access
   - enable priority mem read for Xe2
   - Add first GuC BMG fw
   - fix dma-resv lock
   - Fix DGFX display suspend/resume
   - Use xe_managed for kernel BOs
   - Use reserved copy engine for user binds on faulting devices
   - Allow mixing dma-fence jobs and long-running faulting jobs
   - fix media TLB invalidation
   - fix rpm in TTM swapout path
   - track resources and VF state by PF

  i915:
   - Type-C programming fix for MTL+
   - FBC cleanup
   - Calc vblank delay more accurately
   - On DP MST, Enable LT fallback for UHBR<->non-UHBR rates
   - Fix DP LTTPR detection
   - limit relocations to INT_MAX
   - fix long hangs in buddy allocator on DG2/A380

  amdgpu:
   - Per-queue reset support
   - SDMA devcoredump support
   - DCN 4.0.1 updates
   - GFX12/VCN4/JPEG4 updates
   - Convert vbios embedded EDID to drm_edid
   - GFX9.3/9.4 devcoredump support
   - process isolation framework for GFX 9.4.3/4
   - take IOMMU mappings into account for P2P DMA

  amdkfd:
   - CRIU fixes
   - HMM fix
   - Enable process isolation support for GFX 9.4.3/4
   - Allow users to target recommended SDMA engines
   - KFD support for targetting queues on recommended SDMA engines

  radeon:
   - remove .load and drm_dev_alloc
   - Fix vbios embedded EDID size handling
   - Convert vbios embedded EDID to drm_edid
   - Use GEM references instead of TTM
   - r100 cp init cleanup
   - Fix potential overflows in evergreen CS offset tracking

  msm:
   - DPU:
      - implement DP/PHY mapping on SC8180X
      - Enable writeback on SM8150, SC8180X, SM6125, SM6350
   - DP:
      - Enable widebus on all relevant chipsets
      - MSM8998 HDMI support
   - GPU:
      - A642L speedbin support
      - A615/A306/A621 support
      - A7xx devcoredump support

  ast:
   - astdp: Support AST2600 with VGA
   - Clean up HPD
   - Fix timeout loop for DP link training
   - reorganize output code by type (VGA, DP, etc)
   - convert to struct drm_edid
   - fix BMC handling for all outputs

  exynos:
   - drop stale MAINTAINERS pattern
   - constify struct

  loongson:
   - use GEM refcount over TTM

  mgag200:
   - Improve BMC handling
   - Support VBLANK intterupts
   - transparently support BMC outputs

  nouveau:
   - Refactor and clean up internals
   - Use GEM refcount over TTM's

  gm12u320:
   - convert to struct drm_edid

  gma500:
   - update i2c terms

  lcdif:
   - pixel clock fix

  host1x:
   - fix syncpoint IRQ during resume
   - use iommu_paging_domain_alloc()

  imx:
   - ipuv3: convert to struct drm_edid

  omapdrm:
   - improve error handling
   - use common helper for_each_endpoint_of_node()

  panel:
   - add support for BOE TV101WUM-LL2 plus DT bindings
   - novatek-nt35950: improve error handling
   - nv3051d: improve error handling
   - panel-edp:
      - add support for BOE NE140WUM-N6G
      - revert support for SDC ATNA45AF01
   - visionox-vtdr6130:
      - improve error handling
      - use devm_regulator_bulk_get_const()
   - boe-th101mb31ig002:
      - Support for starry-er88577 MIPI-DSI panel plus DT
      - Fix porch parameter
   - edp: Support AOU B116XTN02.3, AUO B116XAN06.1, AOU B116XAT04.1, BOE
     NV140WUM-N41, BOE NV133WUM-N63, BOE NV116WHM-A4D, CMN N116BCA-EA2,
     CMN N116BCP-EA2, CSW MNB601LS1-4
   - himax-hx8394: Support Microchip AC40T08A MIPI Display panel plus DT
   - ilitek-ili9806e: Support Densitron DMT028VGHMCMI-1D TFT plus DT
   - jd9365da:
      - Support Melfas lmfbx101117480 MIPI-DSI panel plus DT
      - Refactor for code sharing
   - panel-edp: fix name for HKC MB116AN01
   - jd9365da: fix "exit sleep" commands
   - jdi-fhd-r63452: simplify error handling with DSI multi-style
     helpers
   - mantix-mlaf057we51: simplify error handling with DSI multi-style
     helpers
   - simple:
      - support Innolux G070ACE-LH3 plus DT bindings
      - support On Tat Industrial Company KD50G21-40NT-A1 plus DT
        bindings
   - st7701:
      - decouple DSI and DRM code
      - add SPI support
      - support Anbernic RG28XX plus DT bindings

  mediatek:
   - support alpha blending
   - remove cl in struct cmdq_pkt
   - ovl adaptor fix
   - add power domain binding for mediatek DPI controller

  renesas:
   - rz-du: add support for RZ/G2UL plus DT bindings

  rockchip:
   - Improve DP sink-capability reporting
   - dw_hdmi: Support 4k@60Hz
   - vop:
      - Support RGB display on Rockchip RK3066
      - Support 4096px width

  sti:
   - convert to struct drm_edid

  stm:
   - Avoid UAF wih managed plane and CRTC helpers
   - Fix module owner
   - Fix error handling in probe
   - Depend on COMMON_CLK
   - ltdc:
      - Fix transparency after disabling plane
      - Remove unused interrupt

  tegra:
   - gr3d: improve PM domain handling
   - convert to struct drm_edid
   - Call drm_atomic_helper_shutdown()

  vc4:
   - fix PM during detect
   - replace DRM_ERROR() with drm_error()
   - v3d: simplify clock retrieval

  v3d:
   - Clean up perfmon

  virtio:
   - add DRM capset"

* tag 'drm-next-2024-09-19' of https://gitlab.freedesktop.org/drm/kernel: (1326 commits)
  drm/xe: Fix missing conversion to xe_display_pm_runtime_resume
  drm/xe/xe2hpg: Add Wa_15016589081
  drm/xe: Don't keep stale pointer to bo->ggtt_node
  drm/xe: fix missing 'xe_vm_put'
  drm/xe: fix build warning with CONFIG_PM=n
  drm/xe: Suppress missing outer rpm protection warning
  drm/xe: prevent potential UAF in pf_provision_vf_ggtt()
  drm/amd/display: Add all planes on CRTC to state for overlay cursor
  drm/i915/bios: fix printk format width
  drm/i915/display: Fix BMG CCS modifiers
  drm/amdgpu: get rid of bogus includes of fdtable.h
  drm/amdkfd: CRIU fixes
  drm/amdgpu: fix a race in kfd_mem_export_dmabuf()
  drm: new helper: drm_gem_prime_handle_to_dmabuf()
  drm/amdgpu/atomfirmware: Silence UBSAN warning
  drm/amdgpu: Fix kdoc entry in 'amdgpu_vm_cpu_prepare'
  drm/amd/amdgpu: apply command submission parser for JPEG v1
  drm/amd/amdgpu: apply command submission parser for JPEG v2+
  drm/amd/pm: fix the pp_dpm_pcie issue on smu v14.0.2/3
  drm/amd/pm: update the features set on smu v14.0.2/3
  ...

.mailmap

@@ -613,6 +613,10 @@ Shuah Khan <shuah@kernel.org> <shuah.kh@samsung.com>
 Sibi Sankar <quic_sibis@quicinc.com> <sibis@codeaurora.org>
 Sid Manning <quic_sidneym@quicinc.com> <sidneym@codeaurora.org>
 Simon Arlott <simon@octiron.net> <simon@fire.lp0.eu>
+Simona Vetter <simona.vetter@ffwll.ch> <daniel.vetter@ffwll.ch>
+Simona Vetter <simona.vetter@ffwll.ch> <daniel.vetter@intel.com>
+Simona Vetter <simona.vetter@ffwll.ch> <daniel@ffwll.ch>
+Simona Vetter <simona.vetter@ffwll.ch> <daniel@biene.ffwll.ch>
 Simon Horman <horms@kernel.org> <simon.horman@corigine.com>
 Simon Horman <horms@kernel.org> <simon.horman@netronome.com>
 Simon Kelley <simon@thekelleys.org.uk>

Documentation/ABI/testing/sysfs-driver-intel-i915-hwmon

@@ -75,3 +75,11 @@ Description:	RO. Energy input of device or gt in microjoules.
 		for the gt.
 
 		Only supported for particular Intel i915 graphics platforms.
+
+What:		/sys/bus/pci/drivers/i915/.../hwmon/hwmon<i>/fan1_input
+Date:		November 2024
+KernelVersion:	6.12
+Contact:	intel-gfx@lists.freedesktop.org
+Description:	RO. Fan speed of device in RPM.
+
+		Only supported for particular Intel i915 graphics platforms.

Documentation/accel/qaic/qaic.rst

@@ -147,12 +147,6 @@ DRM_IOCTL_QAIC_PERF_STATS_BO
   recent execution of a BO. This allows userspace to construct an end to end
   timeline of the BO processing for a performance analysis.
 
-DRM_IOCTL_QAIC_PART_DEV
-  This IOCTL allows userspace to request a duplicate "shadow device". This extra
-  accelN device is associated with a specific partition of resources on the
-  AIC100 device and can be used for limiting a process to some subset of
-  resources.
-
 DRM_IOCTL_QAIC_DETACH_SLICE_BO
   This IOCTL allows userspace to remove the slicing information from a BO that
   was originally provided by a call to DRM_IOCTL_QAIC_ATTACH_SLICE_BO. This

Documentation/devicetree/bindings/display/bridge/toshiba,tc358767.yaml

@@ -92,12 +92,31 @@ properties:
             reference to a valid DPI output or input endpoint node.
 
       port@2:
-        $ref: /schemas/graph.yaml#/properties/port
+        $ref: /schemas/graph.yaml#/$defs/port-base
+        unevaluatedProperties: false
         description: |
             eDP/DP output port. The remote endpoint phandle should be a
             reference to a valid eDP panel input endpoint node. This port is
             optional, treated as DP panel if not defined
 
+        properties:
+          endpoint:
+            $ref: /schemas/media/video-interfaces.yaml#
+            unevaluatedProperties: false
+
+            properties:
+              toshiba,pre-emphasis:
+                description:
+                  Display port output Pre-Emphasis settings for both DP lanes.
+                $ref: /schemas/types.yaml#/definitions/uint8-array
+                minItems: 2
+                maxItems: 2
+                items:
+                  enum:
+                    - 0 # No pre-emphasis
+                    - 1 # 3.5dB pre-emphasis
+                    - 2 # 6dB pre-emphasis
+
     oneOf:
       - required:
           - port@0

Documentation/devicetree/bindings/display/mediatek/mediatek,dpi.yaml

@@ -62,6 +62,9 @@ properties:
       - const: default
       - const: sleep
 
+  power-domains:
+    maxItems: 1
+
   port:
     $ref: /schemas/graph.yaml#/properties/port
     description:
@@ -76,6 +79,20 @@ required:
   - clock-names
   - port
 
+allOf:
+  - if:
+      not:
+        properties:
+          compatible:
+            contains:
+              enum:
+                - mediatek,mt6795-dpi
+                - mediatek,mt8173-dpi
+                - mediatek,mt8186-dpi
+    then:
+      properties:
+        power-domains: false
+
 additionalProperties: false
 
 examples:

Documentation/devicetree/bindings/display/msm/hdmi.yaml

@@ -19,14 +19,15 @@ properties:
       - qcom,hdmi-tx-8974
       - qcom,hdmi-tx-8994
       - qcom,hdmi-tx-8996
+      - qcom,hdmi-tx-8998
 
   clocks:
     minItems: 1
-    maxItems: 5
+    maxItems: 8
 
   clock-names:
     minItems: 1
-    maxItems: 5
+    maxItems: 8
 
   reg:
     minItems: 1
@@ -142,6 +143,7 @@ allOf:
       properties:
         clocks:
           minItems: 5
+          maxItems: 5
         clock-names:
           items:
             - const: mdp_core
@@ -151,6 +153,28 @@ allOf:
             - const: extp
         hdmi-mux-supplies: false
 
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - qcom,hdmi-tx-8998
+    then:
+      properties:
+        clocks:
+          minItems: 8
+          maxItems: 8
+        clock-names:
+          items:
+            - const: mdp_core
+            - const: iface
+            - const: core
+            - const: alt_iface
+            - const: extp
+            - const: bus
+            - const: mnoc
+            - const: iface_mmss
+
 additionalProperties: false
 
 examples:

Documentation/devicetree/bindings/display/panel/boe,th101mb31ig002-28a.yaml

@@ -9,20 +9,20 @@ title: BOE TH101MB31IG002-28A WXGA DSI Display Panel
 maintainers:
   - Manuel Traut <manut@mecka.net>
 
-allOf:
-  - $ref: panel-common.yaml#
-
 properties:
   compatible:
     enum:
         # BOE TH101MB31IG002-28A 10.1" WXGA TFT LCD panel
       - boe,th101mb31ig002-28a
+        # The Starry-er88577 is a 10.1" WXGA TFT-LCD panel
+      - starry,er88577
 
   reg:
     maxItems: 1
 
   backlight: true
   enable-gpios: true
+  reset-gpios: true
   power-supply: true
   port: true
   rotation: true
@@ -33,6 +33,20 @@ required:
   - enable-gpios
   - power-supply
 
+allOf:
+  - $ref: panel-common.yaml#
+  - if:
+      properties:
+        compatible:
+          # The Starry-er88577 is a 10.1" WXGA TFT-LCD panel
+          const: starry,er88577
+    then:
+      properties:
+        reset-gpios: false
+    else:
+      required:
+        - reset-gpios
+
 additionalProperties: false
 
 examples:
@@ -47,6 +61,7 @@ examples:
             reg = <0>;
             backlight = <&backlight_lcd0>;
             enable-gpios = <&gpio 45 GPIO_ACTIVE_HIGH>;
+            reset-gpios = <&gpio 55 GPIO_ACTIVE_LOW>;
             rotation = <90>;
             power-supply = <&vcc_3v3>;
             port {

Documentation/devicetree/bindings/display/panel/boe,tv101wum-ll2.yaml

@@ -0,0 +1,63 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/panel/boe,tv101wum-ll2.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: BOE TV101WUM-LL2 DSI Display Panel
+
+maintainers:
+  - Neil Armstrong <neil.armstrong@linaro.org>
+
+allOf:
+  - $ref: panel-common.yaml#
+
+properties:
+  compatible:
+    const: boe,tv101wum-ll2
+
+  reg:
+    maxItems: 1
+    description: DSI virtual channel
+
+  backlight: true
+  reset-gpios: true
+  vsp-supply: true
+  vsn-supply: true
+  port: true
+  rotation: true
+
+required:
+  - compatible
+  - reg
+  - reset-gpios
+  - vsp-supply
+  - vsn-supply
+  - port
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/gpio/gpio.h>
+    dsi {
+        #address-cells = <1>;
+        #size-cells = <0>;
+        panel@0 {
+            compatible = "boe,tv101wum-ll2";
+            reg = <0>;
+
+            vsn-supply = <&vsn_lcd>;
+            vsp-supply = <&vsp_lcd>;
+
+            reset-gpios = <&pio 45 GPIO_ACTIVE_LOW>;
+
+            port {
+                panel_in: endpoint {
+                    remote-endpoint = <&dsi_out>;
+                };
+            };
+        };
+    };
+
+...

Documentation/devicetree/bindings/display/panel/himax,hx8394.yaml

@@ -15,14 +15,12 @@ description:
   such as the HannStar HSD060BHW4 720x1440 TFT LCD panel connected with
   a MIPI-DSI video interface.
 
-allOf:
-  - $ref: panel-common.yaml#
-
 properties:
   compatible:
     items:
       - enum:
           - hannstar,hsd060bhw4
+          - microchip,ac40t08a-mipi-panel
           - powkiddy,x55-panel
       - const: himax,hx8394
 
@@ -46,7 +44,6 @@ properties:
 required:
   - compatible
   - reg
-  - reset-gpios
   - backlight
   - port
   - vcc-supply
@@ -54,6 +51,18 @@ required:
 
 additionalProperties: false
 
+allOf:
+  - $ref: panel-common.yaml#
+  - if:
+      not:
+        properties:
+          compatible:
+            enum:
+              - microchip,ac40t08a-mipi-panel
+    then:
+      required:
+        - reset-gpios
+
 examples:
   - |
     #include <dt-bindings/gpio/gpio.h>

Documentation/devicetree/bindings/display/panel/ilitek,ili9806e.yaml

@@ -16,6 +16,7 @@ properties:
   compatible:
     items:
       - enum:
+          - densitron,dmt028vghmcmi-1d
           - ortustech,com35h3p70ulc
       - const: ilitek,ili9806e
 

Documentation/devicetree/bindings/display/panel/jadard,jd9365da-h3.yaml

@@ -18,6 +18,7 @@ properties:
       - enum:
           - chongzhou,cz101b4001
           - kingdisplay,kd101ne3-40ti
+          - melfas,lmfbx101117480
           - radxa,display-10hd-ad001
           - radxa,display-8hd-ad002
       - const: jadard,jd9365da-h3

Documentation/devicetree/bindings/display/panel/panel-simple.yaml

@@ -158,6 +158,8 @@ properties:
       - innolux,at070tn92
         # Innolux G070ACE-L01 7" WVGA (800x480) TFT LCD panel
       - innolux,g070ace-l01
+        # Innolux G070ACE-LH3 7" WVGA (800x480) TFT LCD panel with WLED backlight
+      - innolux,g070ace-lh3
         # Innolux G070Y2-L01 7" WVGA (800x480) TFT LCD panel
       - innolux,g070y2-l01
         # Innolux G070Y2-T02 7" WVGA (800x480) TFT LCD TTL panel
@@ -222,6 +224,8 @@ properties:
       - okaya,rs800480t-7x0gp
         # Olimex 4.3" TFT LCD panel
       - olimex,lcd-olinuxino-43-ts
+        # On Tat Industrial Company 5" DPI TFT panel.
+      - ontat,kd50g21-40nt-a1
         # On Tat Industrial Company 7" DPI TFT panel.
       - ontat,yx700wv03
         # OrtusTech COM37H3M05DTC Blanview 3.7" VGA portrait TFT-LCD panel

Documentation/devicetree/bindings/display/panel/sitronix,st7701.yaml

@@ -20,21 +20,19 @@ description: |
   Densitron DMT028VGHMCMI-1A is 480x640, 2-lane MIPI DSI LCD panel
   which has built-in ST7701 chip.
 
-allOf:
-  - $ref: panel-common.yaml#
-
 properties:
   compatible:
     items:
       - enum:
           - anbernic,rg-arc-panel
+          - anbernic,rg28xx-panel
           - densitron,dmt028vghmcmi-1a
           - elida,kd50t048a
           - techstar,ts8550b
       - const: sitronix,st7701
 
   reg:
-    description: DSI virtual channel used by that screen
+    description: DSI / SPI channel used by that screen
     maxItems: 1
 
   VCC-supply:
@@ -43,6 +41,13 @@ properties:
   IOVCC-supply:
     description: I/O system regulator
 
+  dc-gpios:
+    maxItems: 1
+    description:
+      Controller data/command selection (D/CX) in 4-line SPI mode.
+      If not set, the controller is in 3-line SPI mode.
+      Disallowed for DSI.
+
   port: true
   reset-gpios: true
   rotation: true
@@ -57,7 +62,38 @@ required:
   - port
   - reset-gpios
 
-additionalProperties: false
+allOf:
+  - $ref: panel-common.yaml#
+  - if:
+      properties:
+        compatible:
+          contains:
+            # SPI connected panels
+            enum:
+              - anbernic,rg28xx-panel
+    then:
+      $ref: /schemas/spi/spi-peripheral-props.yaml#
+
+  - if:
+      properties:
+        compatible:
+          not:
+            contains:
+              # DSI or SPI without D/CX pin
+              enum:
+                - anbernic,rg-arc-panel
+                - anbernic,rg28xx-panel
+                - densitron,dmt028vghmcmi-1a
+                - elida,kd50t048a
+                - techstar,ts8550b
+    then:
+      required:
+        - dc-gpios
+    else:
+      properties:
+        dc-gpios: false
+
+unevaluatedProperties: false
 
 examples:
   - |
@@ -82,3 +118,26 @@ examples:
             };
         };
     };
+  - |
+    #include <dt-bindings/gpio/gpio.h>
+
+    spi {
+        #address-cells = <1>;
+        #size-cells = <0>;
+
+        panel@0 {
+            compatible = "anbernic,rg28xx-panel", "sitronix,st7701";
+            reg = <0>;
+            spi-max-frequency = <3125000>;
+            VCC-supply = <&reg_lcd>;
+            IOVCC-supply = <&reg_lcd>;
+            reset-gpios = <&pio 8 14 GPIO_ACTIVE_HIGH>; /* LCD-RST: PI14 */
+            backlight = <&backlight>;
+
+            port {
+                panel_in_rgb: endpoint {
+                    remote-endpoint = <&tcon_lcd0_out_lcd>;
+                };
+            };
+        };
+    };

Documentation/devicetree/bindings/display/renesas,rzg2l-du.yaml

@@ -18,6 +18,7 @@ properties:
   compatible:
     oneOf:
       - enum:
+          - renesas,r9a07g043u-du # RZ/G2UL
           - renesas,r9a07g044-du # RZ/G2{L,LC}
       - items:
           - enum:
@@ -60,9 +61,6 @@ properties:
         $ref: /schemas/graph.yaml#/properties/port
         unevaluatedProperties: false
 
-    required:
-      - port@0
-
     unevaluatedProperties: false
 
   renesas,vsps:
@@ -88,6 +86,34 @@ required:
 
 additionalProperties: false
 
+allOf:
+  - if:
+      properties:
+        compatible:
+          contains:
+            const: renesas,r9a07g043u-du
+    then:
+      properties:
+        ports:
+          properties:
+            port@0:
+              description: DPI
+
+          required:
+            - port@0
+    else:
+      properties:
+        ports:
+          properties:
+            port@0:
+              description: DSI
+            port@1:
+              description: DPI
+
+          required:
+            - port@0
+            - port@1
+
 examples:
   # RZ/G2L DU
   - |

Documentation/devicetree/bindings/phy/qcom,hdmi-phy-qmp.yaml

@@ -14,6 +14,7 @@ properties:
   compatible:
     enum:
       - qcom,hdmi-phy-8996
+      - qcom,hdmi-phy-8998
 
   reg:
     maxItems: 6

Documentation/gpu/amdgpu/driver-core.rst

@@ -179,4 +179,4 @@ IP Blocks
    :doc: IP Blocks
 
 .. kernel-doc:: drivers/gpu/drm/amd/include/amd_shared.h
-   :identifiers: amd_ip_block_type amd_ip_funcs
+   :identifiers: amd_ip_block_type amd_ip_funcs DC_DEBUG_MASK

Documentation/gpu/introduction.rst

@@ -154,11 +154,11 @@ Conference talks
 
 * `An Overview of the Linux and Userspace Graphics Stack <https://www.youtube.com/watch?v=wjAJmqwg47k>`_ - Paul Kocialkowski (2020)
 * `Getting pixels on screen on Linux: introduction to Kernel Mode Setting <https://www.youtube.com/watch?v=haes4_Xnc5Q>`_ - Simon Ser (2020)
-* `Everything Great about Upstream Graphics <https://www.youtube.com/watch?v=kVzHOgt6WGE>`_ - Daniel Vetter (2019)
+* `Everything Great about Upstream Graphics <https://www.youtube.com/watch?v=kVzHOgt6WGE>`_ - Simona Vetter (2019)
 * `An introduction to the Linux DRM subsystem <https://www.youtube.com/watch?v=LbDOCJcDRoo>`_ - Maxime Ripard (2017)
-* `Embrace the Atomic (Display) Age <https://www.youtube.com/watch?v=LjiB_JeDn2M>`_ - Daniel Vetter (2016)
+* `Embrace the Atomic (Display) Age <https://www.youtube.com/watch?v=LjiB_JeDn2M>`_ - Simona Vetter (2016)
 * `Anatomy of an Atomic KMS Driver <https://www.youtube.com/watch?v=lihqR9sENpc>`_ - Laurent Pinchart (2015)
-* `Atomic Modesetting for Drivers <https://www.youtube.com/watch?v=kl9suFgbTc8>`_ - Daniel Vetter (2015)
+* `Atomic Modesetting for Drivers <https://www.youtube.com/watch?v=kl9suFgbTc8>`_ - Simona Vetter (2015)
 * `Anatomy of an Embedded KMS Driver <https://www.youtube.com/watch?v=Ja8fM7rTae4>`_ - Laurent Pinchart (2013)
 
 Slides and articles
@@ -169,8 +169,8 @@ Slides and articles
 * `Understanding the Linux Graphics Stack <https://bootlin.com/doc/training/graphics/graphics-slides.pdf>`_ - Bootlin (2022)
 * `DRM KMS overview <https://wiki.st.com/stm32mpu/wiki/DRM_KMS_overview>`_ - STMicroelectronics (2021)
 * `Linux graphic stack <https://studiopixl.com/2017-05-13/linux-graphic-stack-an-overview>`_ - Nathan Gauër (2017)
-* `Atomic mode setting design overview, part 1 <https://lwn.net/Articles/653071/>`_ - Daniel Vetter (2015)
+* `Atomic mode setting design overview, part 1 <https://lwn.net/Articles/653071/>`_ - Simona Vetter (2015)
-* `Atomic mode setting design overview, part 2 <https://lwn.net/Articles/653466/>`_ - Daniel Vetter (2015)
+* `Atomic mode setting design overview, part 2 <https://lwn.net/Articles/653466/>`_ - Simona Vetter (2015)
 * `The DRM/KMS subsystem from a newbie’s point of view <https://bootlin.com/pub/conferences/2014/elce/brezillon-drm-kms/brezillon-drm-kms.pdf>`_ - Boris Brezillon (2014)
 * `A brief introduction to the Linux graphics stack <https://blogs.igalia.com/itoral/2014/07/29/a-brief-introduction-to-the-linux-graphics-stack/>`_ - Iago Toral (2014)
 * `The Linux Graphics Stack <https://blog.mecheye.net/2012/06/the-linux-graphics-stack/>`_ - Jasper St. Pierre (2012)

Documentation/gpu/todo.rst

@@ -37,7 +37,7 @@ Audit each individual driver, make sure it'll work with the generic
 implementation (there's lots of outdated locking leftovers in various
 implementations), and then remove it.
 
-Contact: Daniel Vetter, respective driver maintainers
+Contact: Simona Vetter, respective driver maintainers
 
 Level: Intermediate
 
@@ -61,7 +61,7 @@ do by directly using the new atomic helper driver callbacks.
   .. [2] https://lwn.net/Articles/653071/
   .. [3] https://lwn.net/Articles/653466/
 
-Contact: Daniel Vetter, respective driver maintainers
+Contact: Simona Vetter, respective driver maintainers
 
 Level: Advanced
 
@@ -75,7 +75,7 @@ helper should also be moved from drm_plane_helper.c to the atomic helpers, to
 avoid confusion - the other helpers in that file are all deprecated legacy
 helpers.
 
-Contact: Ville Syrjälä, Daniel Vetter, driver maintainers
+Contact: Ville Syrjälä, Simona Vetter, driver maintainers
 
 Level: Advanced
 
@@ -97,7 +97,7 @@ with the current helpers:
 - Then we could go through all the drivers and remove the more-or-less confused
   checks for plane_state->fb and plane_state->crtc.
 
-Contact: Daniel Vetter
+Contact: Simona Vetter
 
 Level: Advanced
 
@@ -116,7 +116,7 @@ Somewhat related is the legacy_cursor_update hack, which should be replaced with
 the new atomic_async_check/commit functionality in the helpers in drivers that
 still look at that flag.
 
-Contact: Daniel Vetter, respective driver maintainers
+Contact: Simona Vetter, respective driver maintainers
 
 Level: Advanced
 
@@ -169,7 +169,7 @@ interfaces to fix these issues:
   ``_helper_funcs`` since they are not part of the core ABI. There's a
   ``FIXME`` comment in the kerneldoc for each such case in ``drm_crtc.h``.
 
-Contact: Daniel Vetter
+Contact: Simona Vetter
 
 Level: Intermediate
 
@@ -194,7 +194,7 @@ performance-critical drivers it might also be better to go with a more
 fine-grained per-buffer object and per-context lockings scheme. Currently only
 the ``msm`` and `i915` drivers use ``struct_mutex``.
 
-Contact: Daniel Vetter, respective driver maintainers
+Contact: Simona Vetter, respective driver maintainers
 
 Level: Advanced
 
@@ -251,7 +251,7 @@ being rewritten without dependencies on the fbdev module. Some of the
 helpers could further benefit from using struct iosys_map instead of
 raw pointers.
 
-Contact: Thomas Zimmermann <tzimmermann@suse.de>, Daniel Vetter
+Contact: Thomas Zimmermann <tzimmermann@suse.de>, Simona Vetter
 
 Level: Advanced
 
@@ -297,7 +297,7 @@ Various hold-ups:
   version of the varios drm_gem_fb_create functions. Maybe called
   drm_gem_fb_create/_with_dirty/_with_funcs as needed.
 
-Contact: Daniel Vetter
+Contact: Simona Vetter
 
 Level: Intermediate
 
@@ -329,7 +329,7 @@ everything after it has done the write-protect/mkwrite trickery:
 
 Might be good to also have some igt testcases for this.
 
-Contact: Daniel Vetter, Noralf Tronnes
+Contact: Simona Vetter, Noralf Tronnes
 
 Level: Advanced
 
@@ -359,7 +359,7 @@ between setting up the &drm_driver structure and calling drm_dev_register().
 
 - Once all drivers are converted, remove the load/unload callbacks.
 
-Contact: Daniel Vetter
+Contact: Simona Vetter
 
 Level: Intermediate
 
@@ -422,7 +422,7 @@ The task is to use struct iosys_map where it makes sense.
 * TTM might benefit from using struct iosys_map internally.
 * Framebuffer copying and blitting helpers should operate on struct iosys_map.
 
-Contact: Thomas Zimmermann <tzimmermann@suse.de>, Christian König, Daniel Vetter
+Contact: Thomas Zimmermann <tzimmermann@suse.de>, Christian König, Simona Vetter
 
 Level: Intermediate
 
@@ -475,25 +475,22 @@ Remove disable/unprepare in remove/shutdown in panel-simple and panel-edp
 As of commit d2aacaf07395 ("drm/panel: Check for already prepared/enabled in
 drm_panel"), we have a check in the drm_panel core to make sure nobody
 double-calls prepare/enable/disable/unprepare. Eventually that should probably
-be turned into a WARN_ON() or somehow made louder, but right now we actually
+be turned into a WARN_ON() or somehow made louder.
-expect it to trigger and so we don't want it to be too loud.
 
-Specifically, that warning will trigger for panel-edp and panel-simple at
+At the moment, we expect that we may still encounter the warnings in the
-shutdown time because those panels hardcode a call to drm_panel_disable()
+drm_panel core when using panel-simple and panel-edp. Since those panel
-and drm_panel_unprepare() at shutdown and remove time that they call regardless
+drivers are used with a lot of different DRM modeset drivers they still
-of panel state. On systems with a properly coded DRM modeset driver that
+make an extra effort to disable/unprepare the panel themsevles at shutdown
-calls drm_atomic_helper_shutdown() this is pretty much guaranteed to cause
+time. Specifically we could still encounter those warnings if the panel
-the warning to fire.
+driver gets shutdown() _before_ the DRM modeset driver and the DRM modeset
+driver properly calls drm_atomic_helper_shutdown() in its own shutdown()
+callback. Warnings could be avoided in such a case by using something like
+device links to ensure that the panel gets shutdown() after the DRM modeset
+driver.
 
-Unfortunately we can't safely remove the calls in panel-edp and panel-simple
+Once all DRM modeset drivers are known to shutdown properly, the extra
-until we're sure that all DRM modeset drivers that are used with those panels
+calls to disable/unprepare in remove/shutdown in panel-simple and panel-edp
-properly call drm_atomic_helper_shutdown(). This TODO item is to validate
+should be removed and this TODO item marked complete.
-that all DRM modeset drivers used with panel-edp and panel-simple properly
-call drm_atomic_helper_shutdown() and then remove the calls to
-disable/unprepare from those panels. Alternatively, this TODO item could be
-removed by convincing stakeholders that those calls are fine and downgrading
-the error message in drm_panel_disable() / drm_panel_unprepare() to a
-debug-level message.
 
 Contact: Douglas Anderson <dianders@chromium.org>
 
@@ -561,7 +558,7 @@ This is a really varied tasks with lots of little bits and pieces:
   <https://lore.kernel.org/lkml/1446217392-11981-1-git-send-email-alexandru.murtaza@intel.com/>`_
   for some example code that could be reused.
 
-Contact: Daniel Vetter
+Contact: Simona Vetter
 
 Level: Advanced
 
@@ -590,7 +587,7 @@ There's a bunch of issues with it:
   this (together with the drm_minor->drm_device move) would allow us to remove
   debugfs_init.
 
-Contact: Daniel Vetter
+Contact: Simona Vetter
 
 Level: Intermediate
 
@@ -611,7 +608,7 @@ Both these problems can be solved by switching over to drmm_kzalloc(), and the
 various convenience wrappers provided, e.g. drmm_crtc_alloc_with_planes(),
 drmm_universal_plane_alloc(), ... and so on.
 
-Contact: Daniel Vetter
+Contact: Simona Vetter
 
 Level: Intermediate
 
@@ -631,7 +628,7 @@ cache is also tied to &drm_gem_object.import_attach. Meanwhile we paper over
 this problem for USB devices by fishing out the USB host controller device, as
 long as that supports DMA. Otherwise importing can still needlessly fail.
 
-Contact: Thomas Zimmermann <tzimmermann@suse.de>, Daniel Vetter
+Contact: Thomas Zimmermann <tzimmermann@suse.de>, Simona Vetter
 
 Level: Advanced
 
@@ -712,7 +709,7 @@ Plan to fix this:
 2. In all, only look at one of the three status bits set by the above helpers.
 3. Remove the other two status bits.
 
-Contact: Daniel Vetter
+Contact: Simona Vetter
 
 Level: Intermediate
 

Documentation/gpu/xe/xe_mm.rst

@@ -7,6 +7,21 @@ Memory Management
 .. kernel-doc:: drivers/gpu/drm/xe/xe_bo_doc.h
    :doc: Buffer Objects (BO)
 
+GGTT
+====
+
+.. kernel-doc:: drivers/gpu/drm/xe/xe_ggtt.c
+   :doc: Global Graphics Translation Table (GGTT)
+
+GGTT Internal API
+-----------------
+
+.. kernel-doc:: drivers/gpu/drm/xe/xe_ggtt_types.h
+   :internal:
+
+.. kernel-doc:: drivers/gpu/drm/xe/xe_ggtt.c
+   :internal:
+
 Pagetable building
 ==================
 

MAINTAINERS

@@ -1025,6 +1025,13 @@ S:	Supported
 T:	git https://gitlab.freedesktop.org/agd5f/linux.git
 F:	drivers/gpu/drm/amd/display/
 
+AMD DISPLAY CORE - DML
+M:	Chaitanya Dhere <chaitanya.dhere@amd.com>
+M:	Jun Lei <jun.lei@amd.com>
+S:	Supported
+F:	drivers/gpu/drm/amd/display/dc/dml/
+F:	drivers/gpu/drm/amd/display/dc/dml2/
+
 AMD FAM15H PROCESSOR POWER MONITORING DRIVER
 M:	Huang Rui <ray.huang@amd.com>
 L:	linux-hwmon@vger.kernel.org
@@ -1799,6 +1806,7 @@ L:	dri-devel@lists.freedesktop.org
 S:	Supported
 T:	git https://gitlab.freedesktop.org/drm/misc/kernel.git
 F:	Documentation/gpu/panfrost.rst
+F:	drivers/gpu/drm/ci/xfails/panfrost*
 F:	drivers/gpu/drm/panfrost/
 F:	include/uapi/drm/panfrost_drm.h
 
@@ -6724,6 +6732,7 @@ F:	drivers/dma-buf/dma-heap.c
 F:	drivers/dma-buf/heaps/*
 F:	include/linux/dma-heap.h
 F:	include/uapi/linux/dma-heap.h
+F:	tools/testing/selftests/dmabuf-heaps/
 
 DMC FREQUENCY DRIVER FOR SAMSUNG EXYNOS5422
 M:	Lukasz Luba <lukasz.luba@arm.com>
@@ -7398,10 +7407,10 @@ F:	drivers/gpu/drm/udl/
 
 DRM DRIVER FOR VIRTUAL KERNEL MODESETTING (VKMS)
 M:	Rodrigo Siqueira <rodrigosiqueiramelo@gmail.com>
-M:	Melissa Wen <melissa.srw@gmail.com>
 M:	Maíra Canal <mairacanal@riseup.net>
 R:	Haneen Mohammed <hamohammed.sa@gmail.com>
-R:	Daniel Vetter <daniel@ffwll.ch>
+R:	Simona Vetter <simona@ffwll.ch>
+R:	Melissa Wen <melissa.srw@gmail.com>
 L:	dri-devel@lists.freedesktop.org
 S:	Maintained
 T:	git https://gitlab.freedesktop.org/drm/misc/kernel.git
@@ -7434,7 +7443,7 @@ F:	drivers/gpu/drm/panel/panel-widechips-ws2401.c
 
 DRM DRIVERS
 M:	David Airlie <airlied@gmail.com>
-M:	Daniel Vetter <daniel@ffwll.ch>
+M:	Simona Vetter <simona@ffwll.ch>
 L:	dri-devel@lists.freedesktop.org
 S:	Maintained
 B:	https://gitlab.freedesktop.org/drm
@@ -7530,7 +7539,6 @@ M:	Kyungmin Park <kyungmin.park@samsung.com>
 L:	dri-devel@lists.freedesktop.org
 S:	Supported
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/daeinki/drm-exynos.git
-F:	Documentation/devicetree/bindings/display/exynos/
 F:	Documentation/devicetree/bindings/display/samsung/
 F:	drivers/gpu/drm/exynos/
 F:	include/uapi/drm/exynos_drm.h
@@ -8877,7 +8885,7 @@ W:	https://floatingpoint.billm.au/
 F:	arch/x86/math-emu/
 
 FRAMEBUFFER CORE
-M:	Daniel Vetter <daniel@ffwll.ch>
+M:	Simona Vetter <simona@ffwll.ch>
 S:	Odd Fixes
 T:	git https://gitlab.freedesktop.org/drm/misc/kernel.git
 F:	drivers/video/fbdev/core/
@@ -11044,6 +11052,7 @@ T:	git https://gitlab.freedesktop.org/drm/misc/kernel.git
 F:	Documentation/devicetree/bindings/gpu/img,powervr-rogue.yaml
 F:	Documentation/devicetree/bindings/gpu/img,powervr-sgx.yaml
 F:	Documentation/gpu/imagination/
+F:	drivers/gpu/drm/ci/xfails/powervr*
 F:	drivers/gpu/drm/imagination/
 F:	include/uapi/drm/pvr_drm.h
 
@@ -18923,7 +18932,6 @@ F:	include/dt-bindings/clock/qcom,*
 QUALCOMM CLOUD AI (QAIC) DRIVER
 M:	Jeffrey Hugo <quic_jhugo@quicinc.com>
 R:	Carl Vanderlip <quic_carlv@quicinc.com>
-R:	Pranjal Ramajor Asha Kanojiya <quic_pkanojiy@quicinc.com>
 L:	linux-arm-msm@vger.kernel.org
 L:	dri-devel@lists.freedesktop.org
 S:	Supported

arch/powerpc/kernel/nvram_64.c

@@ -73,7 +73,7 @@ static const char *nvram_os_partitions[] = {
 };
 
 static void oops_to_nvram(struct kmsg_dumper *dumper,
-			  enum kmsg_dump_reason reason);
+			  struct kmsg_dump_detail *detail);
 
 static struct kmsg_dumper nvram_kmsg_dumper = {
 	.dump = oops_to_nvram
@@ -643,7 +643,7 @@ void __init nvram_init_oops_partition(int rtas_partition_exists)
  * partition.  If that's too much, go back and capture uncompressed text.
  */
 static void oops_to_nvram(struct kmsg_dumper *dumper,
-			  enum kmsg_dump_reason reason)
+			  struct kmsg_dump_detail *detail)
 {
 	struct oops_log_info *oops_hdr = (struct oops_log_info *)oops_buf;
 	static unsigned int oops_count = 0;
@@ -655,7 +655,7 @@ static void oops_to_nvram(struct kmsg_dumper *dumper,
 	unsigned int err_type = ERR_TYPE_KERNEL_PANIC_GZ;
 	int rc = -1;
 
-	switch (reason) {
+	switch (detail->reason) {
 	case KMSG_DUMP_SHUTDOWN:
 		/* These are almost always orderly shutdowns. */
 		return;
@@ -671,7 +671,7 @@ static void oops_to_nvram(struct kmsg_dumper *dumper,
 		break;
 	default:
 		pr_err("%s: ignoring unrecognized KMSG_DUMP_* reason %d\n",
-		       __func__, (int) reason);
+		       __func__, (int) detail->reason);
 		return;
 	}
 

arch/powerpc/platforms/powernv/opal-kmsg.c

@@ -20,13 +20,13 @@
  * message, it just ensures that OPAL completely flushes the console buffer.
  */
 static void kmsg_dump_opal_console_flush(struct kmsg_dumper *dumper,
-				     enum kmsg_dump_reason reason)
+				     struct kmsg_dump_detail *detail)
 {
 	/*
 	 * Outside of a panic context the pollers will continue to run,
 	 * so we don't need to do any special flushing.
 	 */
-	if (reason != KMSG_DUMP_PANIC)
+	if (detail->reason != KMSG_DUMP_PANIC)
 		return;
 
 	opal_flush_console(0);

arch/um/kernel/kmsg_dump.c

@@ -8,7 +8,7 @@
 #include <os.h>
 
 static void kmsg_dumper_stdout(struct kmsg_dumper *dumper,
-				enum kmsg_dump_reason reason)
+				struct kmsg_dump_detail *detail)
 {
 	static struct kmsg_dump_iter iter;
 	static DEFINE_SPINLOCK(lock);

drivers/accel/drm_accel.c

@@ -8,7 +8,7 @@
 
 #include <linux/debugfs.h>
 #include <linux/device.h>
-#include <linux/idr.h>
+#include <linux/xarray.h>
 
 #include <drm/drm_accel.h>
 #include <drm/drm_auth.h>
@@ -18,8 +18,7 @@
 #include <drm/drm_ioctl.h>
 #include <drm/drm_print.h>
 
-static DEFINE_SPINLOCK(accel_minor_lock);
+DEFINE_XARRAY_ALLOC(accel_minors_xa);
-static struct idr accel_minors_idr;
 
 static struct dentry *accel_debugfs_root;
 
@@ -117,99 +116,6 @@ void accel_set_device_instance_params(struct device *kdev, int index)
 	kdev->type = &accel_sysfs_device_minor;
 }
 
-/**
- * accel_minor_alloc() - Allocates a new accel minor
- *
- * This function access the accel minors idr and allocates from it
- * a new id to represent a new accel minor
- *
- * Return: A new id on success or error code in case idr_alloc failed
- */
-int accel_minor_alloc(void)
-{
-	unsigned long flags;
-	int r;
-
-	spin_lock_irqsave(&accel_minor_lock, flags);
-	r = idr_alloc(&accel_minors_idr, NULL, 0, ACCEL_MAX_MINORS, GFP_NOWAIT);
-	spin_unlock_irqrestore(&accel_minor_lock, flags);
-
-	return r;
-}
-
-/**
- * accel_minor_remove() - Remove an accel minor
- * @index: The minor id to remove.
- *
- * This function access the accel minors idr and removes from
- * it the member with the id that is passed to this function.
- */
-void accel_minor_remove(int index)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&accel_minor_lock, flags);
-	idr_remove(&accel_minors_idr, index);
-	spin_unlock_irqrestore(&accel_minor_lock, flags);
-}
-
-/**
- * accel_minor_replace() - Replace minor pointer in accel minors idr.
- * @minor: Pointer to the new minor.
- * @index: The minor id to replace.
- *
- * This function access the accel minors idr structure and replaces the pointer
- * that is associated with an existing id. Because the minor pointer can be
- * NULL, we need to explicitly pass the index.
- *
- * Return: 0 for success, negative value for error
- */
-void accel_minor_replace(struct drm_minor *minor, int index)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&accel_minor_lock, flags);
-	idr_replace(&accel_minors_idr, minor, index);
-	spin_unlock_irqrestore(&accel_minor_lock, flags);
-}
-
-/*
- * Looks up the given minor-ID and returns the respective DRM-minor object. The
- * refence-count of the underlying device is increased so you must release this
- * object with accel_minor_release().
- *
- * The object can be only a drm_minor that represents an accel device.
- *
- * As long as you hold this minor, it is guaranteed that the object and the
- * minor->dev pointer will stay valid! However, the device may get unplugged and
- * unregistered while you hold the minor.
- */
-static struct drm_minor *accel_minor_acquire(unsigned int minor_id)
-{
-	struct drm_minor *minor;
-	unsigned long flags;
-
-	spin_lock_irqsave(&accel_minor_lock, flags);
-	minor = idr_find(&accel_minors_idr, minor_id);
-	if (minor)
-		drm_dev_get(minor->dev);
-	spin_unlock_irqrestore(&accel_minor_lock, flags);
-
-	if (!minor) {
-		return ERR_PTR(-ENODEV);
-	} else if (drm_dev_is_unplugged(minor->dev)) {
-		drm_dev_put(minor->dev);
-		return ERR_PTR(-ENODEV);
-	}
-
-	return minor;
-}
-
-static void accel_minor_release(struct drm_minor *minor)
-{
-	drm_dev_put(minor->dev);
-}
-
 /**
  * accel_open - open method for ACCEL file
  * @inode: device inode
@@ -227,7 +133,7 @@ int accel_open(struct inode *inode, struct file *filp)
 	struct drm_minor *minor;
 	int retcode;
 
-	minor = accel_minor_acquire(iminor(inode));
+	minor = drm_minor_acquire(&accel_minors_xa, iminor(inode));
 	if (IS_ERR(minor))
 		return PTR_ERR(minor);
 
@@ -246,7 +152,7 @@ int accel_open(struct inode *inode, struct file *filp)
 
 err_undo:
 	atomic_dec(&dev->open_count);
-	accel_minor_release(minor);
+	drm_minor_release(minor);
 	return retcode;
 }
 EXPORT_SYMBOL_GPL(accel_open);
@@ -257,7 +163,7 @@ static int accel_stub_open(struct inode *inode, struct file *filp)
 	struct drm_minor *minor;
 	int err;
 
-	minor = accel_minor_acquire(iminor(inode));
+	minor = drm_minor_acquire(&accel_minors_xa, iminor(inode));
 	if (IS_ERR(minor))
 		return PTR_ERR(minor);
 
@@ -274,7 +180,7 @@ static int accel_stub_open(struct inode *inode, struct file *filp)
 		err = 0;
 
 out:
-	accel_minor_release(minor);
+	drm_minor_release(minor);
 
 	return err;
 }
@@ -290,15 +196,13 @@ void accel_core_exit(void)
 	unregister_chrdev(ACCEL_MAJOR, "accel");
 	debugfs_remove(accel_debugfs_root);
 	accel_sysfs_destroy();
-	idr_destroy(&accel_minors_idr);
+	WARN_ON(!xa_empty(&accel_minors_xa));
 }
 
 int __init accel_core_init(void)
 {
 	int ret;
 
-	idr_init(&accel_minors_idr);
-
 	ret = accel_sysfs_init();
 	if (ret < 0) {
 		DRM_ERROR("Cannot create ACCEL class: %d\n", ret);

drivers/accel/ivpu/ivpu_fw.c

@@ -60,6 +60,10 @@ static struct {
 	{ IVPU_HW_IP_40XX, "intel/vpu/vpu_40xx_v0.0.bin" },
 };
 
+/* Production fw_names from the table above */
+MODULE_FIRMWARE("intel/vpu/vpu_37xx_v0.0.bin");
+MODULE_FIRMWARE("intel/vpu/vpu_40xx_v0.0.bin");
+
 static int ivpu_fw_request(struct ivpu_device *vdev)
 {
 	int ret = -ENOENT;

drivers/dma-buf/dma-fence-array.c

@@ -144,37 +144,38 @@ const struct dma_fence_ops dma_fence_array_ops = {
 EXPORT_SYMBOL(dma_fence_array_ops);
 
 /**
- * dma_fence_array_create - Create a custom fence array
+ * dma_fence_array_alloc - Allocate a custom fence array
+ * @num_fences:		[in]	number of fences to add in the array
+ *
+ * Return dma fence array on success, NULL on failure
+ */
+struct dma_fence_array *dma_fence_array_alloc(int num_fences)
+{
+	struct dma_fence_array *array;
+
+	return kzalloc(struct_size(array, callbacks, num_fences), GFP_KERNEL);
+}
+EXPORT_SYMBOL(dma_fence_array_alloc);
+
+/**
+ * dma_fence_array_init - Init a custom fence array
+ * @array:		[in]	dma fence array to arm
  * @num_fences:		[in]	number of fences to add in the array
  * @fences:		[in]	array containing the fences
  * @context:		[in]	fence context to use
  * @seqno:		[in]	sequence number to use
  * @signal_on_any:	[in]	signal on any fence in the array
  *
- * Allocate a dma_fence_array object and initialize the base fence with
+ * Implementation of @dma_fence_array_create without allocation. Useful to init
- * dma_fence_init().
+ * a preallocated dma fence array in the path of reclaim or dma fence signaling.
- * In case of error it returns NULL.
- *
- * The caller should allocate the fences array with num_fences size
- * and fill it with the fences it wants to add to the object. Ownership of this
- * array is taken and dma_fence_put() is used on each fence on release.
- *
- * If @signal_on_any is true the fence array signals if any fence in the array
- * signals, otherwise it signals when all fences in the array signal.
  */
-struct dma_fence_array *dma_fence_array_create(int num_fences,
+void dma_fence_array_init(struct dma_fence_array *array,
-					       struct dma_fence **fences,
+			  int num_fences, struct dma_fence **fences,
 			  u64 context, unsigned seqno,
 			  bool signal_on_any)
 {
-	struct dma_fence_array *array;
-
 	WARN_ON(!num_fences || !fences);
 
-	array = kzalloc(struct_size(array, callbacks, num_fences), GFP_KERNEL);
-	if (!array)
-		return NULL;
-
 	array->num_fences = num_fences;
 
 	spin_lock_init(&array->lock);
@@ -200,6 +201,41 @@ struct dma_fence_array *dma_fence_array_create(int num_fences,
 	 */
 	while (num_fences--)
 		WARN_ON(dma_fence_is_container(fences[num_fences]));
+}
+EXPORT_SYMBOL(dma_fence_array_init);
+
+/**
+ * dma_fence_array_create - Create a custom fence array
+ * @num_fences:		[in]	number of fences to add in the array
+ * @fences:		[in]	array containing the fences
+ * @context:		[in]	fence context to use
+ * @seqno:		[in]	sequence number to use
+ * @signal_on_any:	[in]	signal on any fence in the array
+ *
+ * Allocate a dma_fence_array object and initialize the base fence with
+ * dma_fence_init().
+ * In case of error it returns NULL.
+ *
+ * The caller should allocate the fences array with num_fences size
+ * and fill it with the fences it wants to add to the object. Ownership of this
+ * array is taken and dma_fence_put() is used on each fence on release.
+ *
+ * If @signal_on_any is true the fence array signals if any fence in the array
+ * signals, otherwise it signals when all fences in the array signal.
+ */
+struct dma_fence_array *dma_fence_array_create(int num_fences,
+					       struct dma_fence **fences,
+					       u64 context, unsigned seqno,
+					       bool signal_on_any)
+{
+	struct dma_fence_array *array;
+
+	array = dma_fence_array_alloc(num_fences);
+	if (!array)
+		return NULL;
+
+	dma_fence_array_init(array, num_fences, fences,
+			     context, seqno, signal_on_any);
 
 	return array;
 }

drivers/dma-buf/dma-heap.c

@@ -7,17 +7,15 @@
  */
 
 #include <linux/cdev.h>
-#include <linux/debugfs.h>
 #include <linux/device.h>
 #include <linux/dma-buf.h>
-#include <linux/err.h>
-#include <linux/xarray.h>
-#include <linux/list.h>
-#include <linux/slab.h>
-#include <linux/nospec.h>
-#include <linux/uaccess.h>
-#include <linux/syscalls.h>
 #include <linux/dma-heap.h>
+#include <linux/err.h>
+#include <linux/list.h>
+#include <linux/nospec.h>
+#include <linux/syscalls.h>
+#include <linux/uaccess.h>
+#include <linux/xarray.h>
 #include <uapi/linux/dma-heap.h>
 
 #define DEVNAME "dma_heap"
@@ -28,9 +26,10 @@
  * struct dma_heap - represents a dmabuf heap in the system
  * @name:		used for debugging/device-node name
  * @ops:		ops struct for this heap
- * @heap_devt		heap device node
+ * @priv:		private data for this heap
- * @list		list head connecting to list of heaps
+ * @heap_devt:		heap device node
- * @heap_cdev		heap char device
+ * @list:		list head connecting to list of heaps
+ * @heap_cdev:		heap char device
  *
  * Represents a heap of memory from which buffers can be made.
  */
@@ -193,11 +192,11 @@ static const struct file_operations dma_heap_fops = {
 };
 
 /**
- * dma_heap_get_drvdata() - get per-subdriver data for the heap
+ * dma_heap_get_drvdata - get per-heap driver data
  * @heap: DMA-Heap to retrieve private data for
  *
  * Returns:
- * The per-subdriver data for the heap.
+ * The per-heap data for the heap.
  */
 void *dma_heap_get_drvdata(struct dma_heap *heap)
 {
@@ -205,8 +204,8 @@ void *dma_heap_get_drvdata(struct dma_heap *heap)
 }
 
 /**
- * dma_heap_get_name() - get heap name
+ * dma_heap_get_name - get heap name
- * @heap: DMA-Heap to retrieve private data for
+ * @heap: DMA-Heap to retrieve the name of
  *
  * Returns:
  * The char* for the heap name.
@@ -216,6 +215,10 @@ const char *dma_heap_get_name(struct dma_heap *heap)
 	return heap->name;
 }
 
+/**
+ * dma_heap_add - adds a heap to dmabuf heaps
+ * @exp_info: information needed to register this heap
+ */
 struct dma_heap *dma_heap_add(const struct dma_heap_export_info *exp_info)
 {
 	struct dma_heap *heap, *h, *err_ret;

drivers/gpu/drm/Kconfig

@@ -107,7 +107,7 @@ config DRM_KMS_HELPER
 
 config DRM_PANIC
 	bool "Display a user-friendly message when a kernel panic occurs"
-	depends on DRM && !(FRAMEBUFFER_CONSOLE && VT_CONSOLE)
+	depends on DRM
 	select FONT_SUPPORT
 	help
 	  Enable a drm panic handler, which will display a user-friendly message
@@ -149,6 +149,37 @@ config DRM_PANIC_SCREEN
 	  or by writing to /sys/module/drm/parameters/panic_screen sysfs entry
 	  Default is "user"
 
+config DRM_PANIC_SCREEN_QR_CODE
+	bool "Add a panic screen with a QR code"
+	depends on DRM_PANIC && RUST
+	help
+	  This option adds a QR code generator, and a panic screen with a QR
+	  code. The QR code will contain the last lines of kmsg and other debug
+	  information. This should be easier for the user to report a kernel
+	  panic, with all debug information available.
+	  To use this panic screen, also set DRM_PANIC_SCREEN to "qr_code"
+
+config DRM_PANIC_SCREEN_QR_CODE_URL
+	string "Base URL of the QR code in the panic screen"
+	depends on DRM_PANIC_SCREEN_QR_CODE
+	help
+	  This option sets the base URL to report the kernel panic. If it's set
+	  the QR code will contain the URL and the kmsg compressed with zlib as
+	  a URL parameter. If it's empty, the QR code will contain the kmsg as
+	  uncompressed text only.
+	  There is a demo code in javascript, to decode and uncompress the kmsg
+	  data from the URL parameter at https://github.com/kdj0c/panic_report
+
+config DRM_PANIC_SCREEN_QR_VERSION
+	int "Maximum version (size) of the QR code."
+	depends on DRM_PANIC_SCREEN_QR_CODE
+	default 40
+	help
+	  This option limits the version (or size) of the QR code. QR code
+	  version ranges from Version 1 (21x21) to Version 40 (177x177).
+	  Smaller QR code are easier to read, but will contain less debugging
+	  data. Default is 40.
+
 config DRM_DEBUG_DP_MST_TOPOLOGY_REFS
         bool "Enable refcount backtrace history in the DP MST helpers"
 	depends on STACKTRACE_SUPPORT

drivers/gpu/drm/Makefile

@@ -89,6 +89,7 @@ drm-$(CONFIG_DRM_PRIVACY_SCREEN) += \
 	drm_privacy_screen_x86.o
 drm-$(CONFIG_DRM_ACCEL) += ../../accel/drm_accel.o
 drm-$(CONFIG_DRM_PANIC) += drm_panic.o
+drm-$(CONFIG_DRM_PANIC_SCREEN_QR_CODE) += drm_panic_qr.o
 obj-$(CONFIG_DRM)	+= drm.o
 
 obj-$(CONFIG_DRM_PANEL_ORIENTATION_QUIRKS) += drm_panel_orientation_quirks.o

drivers/gpu/drm/amd/amdgpu/Makefile

@@ -39,23 +39,7 @@ ccflags-y := -I$(FULL_AMD_PATH)/include/asic_reg \
 	-I$(FULL_AMD_DISPLAY_PATH)/amdgpu_dm \
 	-I$(FULL_AMD_PATH)/amdkfd
 
-subdir-ccflags-y := -Wextra
+# Locally disable W=1 warnings enabled in drm subsystem Makefile
-subdir-ccflags-y += -Wunused
-subdir-ccflags-y += -Wmissing-prototypes
-subdir-ccflags-y += -Wmissing-declarations
-subdir-ccflags-y += -Wmissing-include-dirs
-subdir-ccflags-y += -Wold-style-definition
-subdir-ccflags-y += -Wmissing-format-attribute
-# Need this to avoid recursive variable evaluation issues
-cond-flags := $(call cc-option, -Wunused-but-set-variable) \
-	$(call cc-option, -Wunused-const-variable) \
-	$(call cc-option, -Wstringop-truncation) \
-	$(call cc-option, -Wpacked-not-aligned)
-subdir-ccflags-y += $(cond-flags)
-subdir-ccflags-y += -Wno-unused-parameter
-subdir-ccflags-y += -Wno-type-limits
-subdir-ccflags-y += -Wno-sign-compare
-subdir-ccflags-y += -Wno-missing-field-initializers
 subdir-ccflags-y += -Wno-override-init
 subdir-ccflags-$(CONFIG_DRM_AMDGPU_WERROR) += -Werror
 

drivers/gpu/drm/amd/amdgpu/amdgpu.h

@@ -118,6 +118,8 @@
 
 #define MAX_GPU_INSTANCE		64
 
+#define GFX_SLICE_PERIOD		msecs_to_jiffies(250)
+
 struct amdgpu_gpu_instance {
 	struct amdgpu_device		*adev;
 	int				mgpu_fan_enabled;
@@ -235,6 +237,7 @@ extern int sched_policy;
 extern bool debug_evictions;
 extern bool no_system_mem_limit;
 extern int halt_if_hws_hang;
+extern uint amdgpu_svm_default_granularity;
 #else
 static const int __maybe_unused sched_policy = KFD_SCHED_POLICY_HWS;
 static const bool __maybe_unused debug_evictions; /* = false */
@@ -348,7 +351,7 @@ enum amdgpu_kiq_irq {
 	AMDGPU_CP_KIQ_IRQ_LAST
 };
 #define SRIOV_USEC_TIMEOUT  1200000 /* wait 12 * 100ms for SRIOV */
-#define MAX_KIQ_REG_WAIT (amdgpu_sriov_vf(adev) ? 50000 : 5000) /* in usecs, extend for VF */
+#define MAX_KIQ_REG_WAIT       5000 /* in usecs, 5ms */
 #define MAX_KIQ_REG_BAILOUT_INTERVAL   5 /* in msecs, 5ms */
 #define MAX_KIQ_REG_TRY 1000
 
@@ -823,17 +826,6 @@ struct amdgpu_mqd {
 struct amdgpu_reset_domain;
 struct amdgpu_fru_info;
 
-struct amdgpu_reset_info {
-	/* reset dump register */
-	u32 *reset_dump_reg_list;
-	u32 *reset_dump_reg_value;
-	int num_regs;
-
-#ifdef CONFIG_DEV_COREDUMP
-	struct amdgpu_coredump_info *coredump_info;
-#endif
-};
-
 /*
  * Non-zero (true) if the GPU has VRAM. Zero (false) otherwise.
  */
@@ -1157,8 +1149,6 @@ struct amdgpu_device {
 
 	struct mutex			benchmark_mutex;
 
-	struct amdgpu_reset_info	reset_info;
-
 	bool                            scpm_enabled;
 	uint32_t                        scpm_status;
 
@@ -1175,6 +1165,11 @@ struct amdgpu_device {
 	bool                            debug_disable_soft_recovery;
 	bool                            debug_use_vram_fw_buf;
 	bool                            debug_enable_ras_aca;
+	bool                            debug_exp_resets;
+
+	bool				enforce_isolation[MAX_XCP];
+	/* Added this mutex for cleaner shader isolation between GFX and compute processes */
+	struct mutex                    enforce_isolation_mutex;
 };
 
 static inline uint32_t amdgpu_ip_version(const struct amdgpu_device *adev,
@@ -1484,7 +1479,6 @@ extern const int amdgpu_max_kms_ioctl;
 
 int amdgpu_driver_load_kms(struct amdgpu_device *adev, unsigned long flags);
 void amdgpu_driver_unload_kms(struct drm_device *dev);
-void amdgpu_driver_lastclose_kms(struct drm_device *dev);
 int amdgpu_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv);
 void amdgpu_driver_postclose_kms(struct drm_device *dev,
 				 struct drm_file *file_priv);
@@ -1588,13 +1582,6 @@ static inline bool amdgpu_acpi_is_s3_active(struct amdgpu_device *adev) { return
 static inline void amdgpu_choose_low_power_state(struct amdgpu_device *adev) { }
 #endif
 
-#if defined(CONFIG_DRM_AMD_DC)
-int amdgpu_dm_display_resume(struct amdgpu_device *adev );
-#else
-static inline int amdgpu_dm_display_resume(struct amdgpu_device *adev) { return 0; }
-#endif
-
-
 void amdgpu_register_gpu_instance(struct amdgpu_device *adev);
 void amdgpu_unregister_gpu_instance(struct amdgpu_device *adev);
 

drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c

@@ -80,6 +80,9 @@ static void aca_banks_release(struct aca_banks *banks)
 {
 	struct aca_bank_node *node, *tmp;
 
+	if (list_empty(&banks->list))
+		return;
+
 	list_for_each_entry_safe(node, tmp, &banks->list, node) {
 		list_del(&node->node);
 		kvfree(node);
@@ -453,13 +456,13 @@ static int aca_log_aca_error_data(struct aca_bank_error *bank_error, enum aca_er
 
 	switch (type) {
 	case ACA_ERROR_TYPE_UE:
-		amdgpu_ras_error_statistic_ue_count(err_data, &mcm_info, NULL, count);
+		amdgpu_ras_error_statistic_ue_count(err_data, &mcm_info, count);
 		break;
 	case ACA_ERROR_TYPE_CE:
-		amdgpu_ras_error_statistic_ce_count(err_data, &mcm_info, NULL, count);
+		amdgpu_ras_error_statistic_ce_count(err_data, &mcm_info, count);
 		break;
 	case ACA_ERROR_TYPE_DEFERRED:
-		amdgpu_ras_error_statistic_de_count(err_data, &mcm_info, NULL, count);
+		amdgpu_ras_error_statistic_de_count(err_data, &mcm_info, count);
 		break;
 	default:
 		break;
@@ -562,9 +565,13 @@ static void aca_error_fini(struct aca_error *aerr)
 	struct aca_bank_error *bank_error, *tmp;
 
 	mutex_lock(&aerr->lock);
+	if (list_empty(&aerr->list))
+		goto out_unlock;
+
 	list_for_each_entry_safe(bank_error, tmp, &aerr->list, node)
 		aca_bank_error_remove(aerr, bank_error);
 
+out_unlock:
 	mutex_destroy(&aerr->lock);
 }
 
@@ -680,6 +687,9 @@ static void aca_manager_fini(struct aca_handle_manager *mgr)
 {
 	struct aca_handle *handle, *tmp;
 
+	if (list_empty(&mgr->list))
+		return;
+
 	list_for_each_entry_safe(handle, tmp, &mgr->list, node)
 		amdgpu_aca_remove_handle(handle);
 }

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c

@@ -364,15 +364,15 @@ allocate_mem_reserve_bo_failed:
 	return r;
 }
 
-void amdgpu_amdkfd_free_gtt_mem(struct amdgpu_device *adev, void *mem_obj)
+void amdgpu_amdkfd_free_gtt_mem(struct amdgpu_device *adev, void **mem_obj)
 {
-	struct amdgpu_bo *bo = (struct amdgpu_bo *) mem_obj;
+	struct amdgpu_bo **bo = (struct amdgpu_bo **) mem_obj;
 
-	amdgpu_bo_reserve(bo, true);
+	amdgpu_bo_reserve(*bo, true);
-	amdgpu_bo_kunmap(bo);
+	amdgpu_bo_kunmap(*bo);
-	amdgpu_bo_unpin(bo);
+	amdgpu_bo_unpin(*bo);
-	amdgpu_bo_unreserve(bo);
+	amdgpu_bo_unreserve(*bo);
-	amdgpu_bo_unref(&(bo));
+	amdgpu_bo_unref(bo);
 }
 
 int amdgpu_amdkfd_alloc_gws(struct amdgpu_device *adev, size_t size,
@@ -783,22 +783,6 @@ int amdgpu_amdkfd_send_close_event_drain_irq(struct amdgpu_device *adev,
 	return 0;
 }
 
-bool amdgpu_amdkfd_ras_query_utcl2_poison_status(struct amdgpu_device *adev,
-			int hub_inst, int hub_type)
-{
-	if (!hub_type) {
-		if (adev->gfxhub.funcs->query_utcl2_poison_status)
-			return adev->gfxhub.funcs->query_utcl2_poison_status(adev, hub_inst);
-		else
-			return false;
-	} else {
-		if (adev->mmhub.funcs->query_utcl2_poison_status)
-			return adev->mmhub.funcs->query_utcl2_poison_status(adev, hub_inst);
-		else
-			return false;
-	}
-}
-
 int amdgpu_amdkfd_check_and_lock_kfd(struct amdgpu_device *adev)
 {
 	return kgd2kfd_check_and_lock_kfd();
@@ -887,3 +871,21 @@ free_ring_funcs:
 
 	return r;
 }
+
+/* Stop scheduling on KFD */
+int amdgpu_amdkfd_stop_sched(struct amdgpu_device *adev, uint32_t node_id)
+{
+	if (!adev->kfd.init_complete)
+		return 0;
+
+	return kgd2kfd_stop_sched(adev->kfd.dev, node_id);
+}
+
+/* Start scheduling on KFD */
+int amdgpu_amdkfd_start_sched(struct amdgpu_device *adev, uint32_t node_id)
+{
+	if (!adev->kfd.init_complete)
+		return 0;
+
+	return kgd2kfd_start_sched(adev->kfd.dev, node_id);
+}

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h

@@ -235,7 +235,7 @@ int amdgpu_amdkfd_bo_validate_and_fence(struct amdgpu_bo *bo,
 int amdgpu_amdkfd_alloc_gtt_mem(struct amdgpu_device *adev, size_t size,
 				void **mem_obj, uint64_t *gpu_addr,
 				void **cpu_ptr, bool mqd_gfx9);
-void amdgpu_amdkfd_free_gtt_mem(struct amdgpu_device *adev, void *mem_obj);
+void amdgpu_amdkfd_free_gtt_mem(struct amdgpu_device *adev, void **mem_obj);
 int amdgpu_amdkfd_alloc_gws(struct amdgpu_device *adev, size_t size,
 				void **mem_obj);
 void amdgpu_amdkfd_free_gws(struct amdgpu_device *adev, void *mem_obj);
@@ -264,6 +264,8 @@ int amdgpu_amdkfd_send_close_event_drain_irq(struct amdgpu_device *adev,
 					uint32_t *payload);
 int amdgpu_amdkfd_unmap_hiq(struct amdgpu_device *adev, u32 doorbell_off,
 				u32 inst);
+int amdgpu_amdkfd_start_sched(struct amdgpu_device *adev, uint32_t node_id);
+int amdgpu_amdkfd_stop_sched(struct amdgpu_device *adev, uint32_t node_id);
 
 /* Read user wptr from a specified user address space with page fault
  * disabled. The memory must be pinned and mapped to the hardware when
@@ -322,7 +324,7 @@ int amdgpu_amdkfd_gpuvm_map_gtt_bo_to_kernel(struct kgd_mem *mem,
 					     void **kptr, uint64_t *size);
 void amdgpu_amdkfd_gpuvm_unmap_gtt_bo_from_kernel(struct kgd_mem *mem);
 
-int amdgpu_amdkfd_map_gtt_bo_to_gart(struct amdgpu_bo *bo);
+int amdgpu_amdkfd_map_gtt_bo_to_gart(struct amdgpu_bo *bo, struct amdgpu_bo **bo_gart);
 
 int amdgpu_amdkfd_gpuvm_restore_process_bos(void *process_info,
 					    struct dma_fence __rcu **ef);
@@ -345,11 +347,9 @@ void amdgpu_amdkfd_ras_pasid_poison_consumption_handler(struct amdgpu_device *ad
 			pasid_notify pasid_fn, void *data, uint32_t reset);
 
 bool amdgpu_amdkfd_is_fed(struct amdgpu_device *adev);
-bool amdgpu_amdkfd_bo_mapped_to_dev(struct amdgpu_device *adev, struct kgd_mem *mem);
+bool amdgpu_amdkfd_bo_mapped_to_dev(void *drm_priv, struct kgd_mem *mem);
 void amdgpu_amdkfd_block_mmu_notifications(void *p);
 int amdgpu_amdkfd_criu_resume(void *p);
-bool amdgpu_amdkfd_ras_query_utcl2_poison_status(struct amdgpu_device *adev,
-			int hub_inst, int hub_type);
 int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev,
 		uint64_t size, u32 alloc_flag, int8_t xcp_id);
 void amdgpu_amdkfd_unreserve_mem_limit(struct amdgpu_device *adev,
@@ -426,6 +426,8 @@ void kgd2kfd_set_sram_ecc_flag(struct kfd_dev *kfd);
 void kgd2kfd_smi_event_throttle(struct kfd_dev *kfd, uint64_t throttle_bitmask);
 int kgd2kfd_check_and_lock_kfd(void);
 void kgd2kfd_unlock_kfd(void);
+int kgd2kfd_start_sched(struct kfd_dev *kfd, uint32_t node_id);
+int kgd2kfd_stop_sched(struct kfd_dev *kfd, uint32_t node_id);
 #else
 static inline int kgd2kfd_init(void)
 {
@@ -496,5 +498,15 @@ static inline int kgd2kfd_check_and_lock_kfd(void)
 static inline void kgd2kfd_unlock_kfd(void)
 {
 }
+
+static inline int kgd2kfd_start_sched(struct kfd_dev *kfd, uint32_t node_id)
+{
+	return 0;
+}
+
+static inline int kgd2kfd_stop_sched(struct kfd_dev *kfd, uint32_t node_id)
+{
+	return 0;
+}
 #endif
 #endif /* AMDGPU_AMDKFD_H_INCLUDED */

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c

@@ -191,4 +191,6 @@ const struct kfd2kgd_calls aldebaran_kfd2kgd = {
 	.get_iq_wait_times = kgd_gfx_v9_get_iq_wait_times,
 	.build_grace_period_packet_info = kgd_gfx_v9_build_grace_period_packet_info,
 	.program_trap_handler_settings = kgd_gfx_v9_program_trap_handler_settings,
+	.hqd_get_pq_addr = kgd_gfx_v9_hqd_get_pq_addr,
+	.hqd_reset = kgd_gfx_v9_hqd_reset,
 };

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c

@@ -20,7 +20,6 @@
  * OTHER DEALINGS IN THE SOFTWARE.
  */
 #include <linux/module.h>
-#include <linux/fdtable.h>
 #include <linux/uaccess.h>
 #include <linux/firmware.h>
 #include "amdgpu.h"
@@ -300,7 +299,7 @@ static int suspend_resume_compute_scheduler(struct amdgpu_device *adev, bool sus
 			if (r)
 				goto out;
 		} else {
-			drm_sched_start(&ring->sched, false);
+			drm_sched_start(&ring->sched);
 		}
 	}
 
@@ -418,5 +417,7 @@ const struct kfd2kgd_calls arcturus_kfd2kgd = {
 	.get_iq_wait_times = kgd_gfx_v9_get_iq_wait_times,
 	.build_grace_period_packet_info = kgd_gfx_v9_build_grace_period_packet_info,
 	.get_cu_occupancy = kgd_gfx_v9_get_cu_occupancy,
-	.program_trap_handler_settings = kgd_gfx_v9_program_trap_handler_settings
+	.program_trap_handler_settings = kgd_gfx_v9_program_trap_handler_settings,
+	.hqd_get_pq_addr = kgd_gfx_v9_hqd_get_pq_addr,
+	.hqd_reset = kgd_gfx_v9_hqd_reset
 };

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c

@@ -541,5 +541,7 @@ const struct kfd2kgd_calls gc_9_4_3_kfd2kgd = {
 			kgd_gfx_v9_4_3_set_wave_launch_trap_override,
 	.set_wave_launch_mode = kgd_aldebaran_set_wave_launch_mode,
 	.set_address_watch = kgd_gfx_v9_4_3_set_address_watch,
-	.clear_address_watch = kgd_gfx_v9_4_3_clear_address_watch
+	.clear_address_watch = kgd_gfx_v9_4_3_clear_address_watch,
+	.hqd_get_pq_addr = kgd_gfx_v9_hqd_get_pq_addr,
+	.hqd_reset = kgd_gfx_v9_hqd_reset
 };

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c

@@ -1070,6 +1070,20 @@ static void program_trap_handler_settings(struct amdgpu_device *adev,
 	unlock_srbm(adev);
 }
 
+uint64_t kgd_gfx_v10_hqd_get_pq_addr(struct amdgpu_device *adev,
+				     uint32_t pipe_id, uint32_t queue_id,
+				     uint32_t inst)
+{
+	return 0;
+}
+
+uint64_t kgd_gfx_v10_hqd_reset(struct amdgpu_device *adev,
+			       uint32_t pipe_id, uint32_t queue_id,
+			       uint32_t inst, unsigned int utimeout)
+{
+	return 0;
+}
+
 const struct kfd2kgd_calls gfx_v10_kfd2kgd = {
 	.program_sh_mem_settings = kgd_program_sh_mem_settings,
 	.set_pasid_vmid_mapping = kgd_set_pasid_vmid_mapping,
@@ -1097,4 +1111,6 @@ const struct kfd2kgd_calls gfx_v10_kfd2kgd = {
 	.get_iq_wait_times = kgd_gfx_v10_get_iq_wait_times,
 	.build_grace_period_packet_info = kgd_gfx_v10_build_grace_period_packet_info,
 	.program_trap_handler_settings = program_trap_handler_settings,
+	.hqd_get_pq_addr = kgd_gfx_v10_hqd_get_pq_addr,
+	.hqd_reset = kgd_gfx_v10_hqd_reset
 };

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h

@@ -56,3 +56,12 @@ void kgd_gfx_v10_build_grace_period_packet_info(struct amdgpu_device *adev,
 					       uint32_t grace_period,
 					       uint32_t *reg_offset,
 					       uint32_t *reg_data);
+uint64_t kgd_gfx_v10_hqd_get_pq_addr(struct amdgpu_device *adev,
+				    uint32_t pipe_id,
+				    uint32_t queue_id,
+				    uint32_t inst);
+uint64_t kgd_gfx_v10_hqd_reset(struct amdgpu_device *adev,
+			      uint32_t pipe_id,
+			      uint32_t queue_id,
+			      uint32_t inst,
+			      unsigned int utimeout);

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c

@@ -680,5 +680,7 @@ const struct kfd2kgd_calls gfx_v10_3_kfd2kgd = {
 	.set_wave_launch_trap_override = kgd_gfx_v10_set_wave_launch_trap_override,
 	.set_wave_launch_mode = kgd_gfx_v10_set_wave_launch_mode,
 	.set_address_watch = kgd_gfx_v10_set_address_watch,
-	.clear_address_watch = kgd_gfx_v10_clear_address_watch
+	.clear_address_watch = kgd_gfx_v10_clear_address_watch,
+	.hqd_get_pq_addr = kgd_gfx_v10_hqd_get_pq_addr,
+	.hqd_reset = kgd_gfx_v10_hqd_reset
 };

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c

@@ -786,6 +786,20 @@ static uint32_t kgd_gfx_v11_clear_address_watch(struct amdgpu_device *adev,
 	return 0;
 }
 
+static uint64_t kgd_gfx_v11_hqd_get_pq_addr(struct amdgpu_device *adev,
+					    uint32_t pipe_id, uint32_t queue_id,
+					    uint32_t inst)
+{
+	return 0;
+}
+
+static uint64_t kgd_gfx_v11_hqd_reset(struct amdgpu_device *adev,
+				      uint32_t pipe_id, uint32_t queue_id,
+				      uint32_t inst, unsigned int utimeout)
+{
+	return 0;
+}
+
 const struct kfd2kgd_calls gfx_v11_kfd2kgd = {
 	.program_sh_mem_settings = program_sh_mem_settings_v11,
 	.set_pasid_vmid_mapping = set_pasid_vmid_mapping_v11,
@@ -808,5 +822,7 @@ const struct kfd2kgd_calls gfx_v11_kfd2kgd = {
 	.set_wave_launch_trap_override = kgd_gfx_v11_set_wave_launch_trap_override,
 	.set_wave_launch_mode = kgd_gfx_v11_set_wave_launch_mode,
 	.set_address_watch = kgd_gfx_v11_set_address_watch,
-	.clear_address_watch = kgd_gfx_v11_clear_address_watch
+	.clear_address_watch = kgd_gfx_v11_clear_address_watch,
+	.hqd_get_pq_addr = kgd_gfx_v11_hqd_get_pq_addr,
+	.hqd_reset = kgd_gfx_v11_hqd_reset
 };

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c

@@ -1144,6 +1144,109 @@ void kgd_gfx_v9_program_trap_handler_settings(struct amdgpu_device *adev,
 	kgd_gfx_v9_unlock_srbm(adev, inst);
 }
 
+uint64_t kgd_gfx_v9_hqd_get_pq_addr(struct amdgpu_device *adev,
+				    uint32_t pipe_id, uint32_t queue_id,
+				    uint32_t inst)
+{
+	uint32_t low, high;
+	uint64_t queue_addr = 0;
+
+	if (!adev->debug_exp_resets &&
+	    !adev->gfx.num_gfx_rings)
+		return 0;
+
+	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id, inst);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, inst);
+
+	if (!RREG32_SOC15(GC, GET_INST(GC, inst), mmCP_HQD_ACTIVE))
+		goto unlock_out;
+
+	low = RREG32_SOC15(GC, GET_INST(GC, inst), mmCP_HQD_PQ_BASE);
+	high = RREG32_SOC15(GC, GET_INST(GC, inst), mmCP_HQD_PQ_BASE_HI);
+
+	/* only concerned with user queues. */
+	if (!high)
+		goto unlock_out;
+
+	queue_addr = (((queue_addr | high) << 32) | low) << 8;
+
+unlock_out:
+	amdgpu_gfx_rlc_exit_safe_mode(adev, inst);
+	kgd_gfx_v9_release_queue(adev, inst);
+
+	return queue_addr;
+}
+
+/* assume queue acquired  */
+static int kgd_gfx_v9_hqd_dequeue_wait(struct amdgpu_device *adev, uint32_t inst,
+				       unsigned int utimeout)
+{
+	unsigned long end_jiffies = (utimeout * HZ / 1000) + jiffies;
+
+	while (true) {
+		uint32_t temp = RREG32_SOC15(GC, GET_INST(GC, inst), mmCP_HQD_ACTIVE);
+
+		if (!(temp & CP_HQD_ACTIVE__ACTIVE_MASK))
+			return 0;
+
+		if (time_after(jiffies, end_jiffies))
+			return -ETIME;
+
+		usleep_range(500, 1000);
+	}
+}
+
+uint64_t kgd_gfx_v9_hqd_reset(struct amdgpu_device *adev,
+			      uint32_t pipe_id, uint32_t queue_id,
+			      uint32_t inst, unsigned int utimeout)
+{
+	uint32_t low, high, pipe_reset_data = 0;
+	uint64_t queue_addr = 0;
+
+	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id, inst);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, inst);
+
+	if (!RREG32_SOC15(GC, GET_INST(GC, inst), mmCP_HQD_ACTIVE))
+		goto unlock_out;
+
+	low = RREG32_SOC15(GC, GET_INST(GC, inst), mmCP_HQD_PQ_BASE);
+	high = RREG32_SOC15(GC, GET_INST(GC, inst), mmCP_HQD_PQ_BASE_HI);
+
+	/* only concerned with user queues. */
+	if (!high)
+		goto unlock_out;
+
+	queue_addr = (((queue_addr | high) << 32) | low) << 8;
+
+	pr_debug("Attempting queue reset on XCC %i pipe id %i queue id %i\n",
+		 inst, pipe_id, queue_id);
+
+	/* assume previous dequeue request issued will take affect after reset */
+	WREG32_SOC15(GC, GET_INST(GC, inst), mmSPI_COMPUTE_QUEUE_RESET, 0x1);
+
+	if (!kgd_gfx_v9_hqd_dequeue_wait(adev, inst, utimeout))
+		goto unlock_out;
+
+	pr_debug("Attempting pipe reset on XCC %i pipe id %i\n", inst, pipe_id);
+
+	pipe_reset_data = REG_SET_FIELD(pipe_reset_data, CP_MEC_CNTL, MEC_ME1_PIPE0_RESET, 1);
+	pipe_reset_data = pipe_reset_data << pipe_id;
+
+	WREG32_SOC15(GC, GET_INST(GC, inst), mmCP_MEC_CNTL, pipe_reset_data);
+	WREG32_SOC15(GC, GET_INST(GC, inst), mmCP_MEC_CNTL, 0);
+
+	if (kgd_gfx_v9_hqd_dequeue_wait(adev, inst, utimeout))
+		queue_addr = 0;
+
+unlock_out:
+	pr_debug("queue reset on XCC %i pipe id %i queue id %i %s\n",
+		 inst, pipe_id, queue_id, !!queue_addr ? "succeeded!" : "failed!");
+	amdgpu_gfx_rlc_exit_safe_mode(adev, inst);
+	kgd_gfx_v9_release_queue(adev, inst);
+
+	return queue_addr;
+}
+
 const struct kfd2kgd_calls gfx_v9_kfd2kgd = {
 	.program_sh_mem_settings = kgd_gfx_v9_program_sh_mem_settings,
 	.set_pasid_vmid_mapping = kgd_gfx_v9_set_pasid_vmid_mapping,
@@ -1172,4 +1275,6 @@ const struct kfd2kgd_calls gfx_v9_kfd2kgd = {
 	.build_grace_period_packet_info = kgd_gfx_v9_build_grace_period_packet_info,
 	.get_cu_occupancy = kgd_gfx_v9_get_cu_occupancy,
 	.program_trap_handler_settings = kgd_gfx_v9_program_trap_handler_settings,
+	.hqd_get_pq_addr = kgd_gfx_v9_hqd_get_pq_addr,
+	.hqd_reset = kgd_gfx_v9_hqd_reset
 };

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h

@@ -101,3 +101,12 @@ void kgd_gfx_v9_build_grace_period_packet_info(struct amdgpu_device *adev,
 					       uint32_t grace_period,
 					       uint32_t *reg_offset,
 					       uint32_t *reg_data);
+uint64_t kgd_gfx_v9_hqd_get_pq_addr(struct amdgpu_device *adev,
+				    uint32_t pipe_id,
+				    uint32_t queue_id,
+				    uint32_t inst);
+uint64_t kgd_gfx_v9_hqd_reset(struct amdgpu_device *adev,
+			      uint32_t pipe_id,
+			      uint32_t queue_id,
+			      uint32_t inst,
+			      unsigned int utimeout);

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c

@@ -25,7 +25,6 @@
 #include <linux/pagemap.h>
 #include <linux/sched/mm.h>
 #include <linux/sched/task.h>
-#include <linux/fdtable.h>
 #include <drm/ttm/ttm_tt.h>
 
 #include <drm/drm_exec.h>
@@ -818,18 +817,13 @@ static int kfd_mem_export_dmabuf(struct kgd_mem *mem)
 	if (!mem->dmabuf) {
 		struct amdgpu_device *bo_adev;
 		struct dma_buf *dmabuf;
-		int r, fd;
 
 		bo_adev = amdgpu_ttm_adev(mem->bo->tbo.bdev);
-		r = drm_gem_prime_handle_to_fd(&bo_adev->ddev, bo_adev->kfd.client.file,
+		dmabuf = drm_gem_prime_handle_to_dmabuf(&bo_adev->ddev, bo_adev->kfd.client.file,
 					       mem->gem_handle,
 			mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_WRITABLE ?
-					       DRM_RDWR : 0, &fd);
+					       DRM_RDWR : 0);
-		if (r)
+		if (IS_ERR(dmabuf))
-			return r;
-		dmabuf = dma_buf_get(fd);
-		close_fd(fd);
-		if (WARN_ON_ONCE(IS_ERR(dmabuf)))
 			return PTR_ERR(dmabuf);
 		mem->dmabuf = dmabuf;
 	}
@@ -1252,7 +1246,7 @@ static int unreserve_bo_and_vms(struct bo_vm_reservation_context *ctx,
 	return ret;
 }
 
-static void unmap_bo_from_gpuvm(struct kgd_mem *mem,
+static int unmap_bo_from_gpuvm(struct kgd_mem *mem,
 				struct kfd_mem_attachment *entry,
 				struct amdgpu_sync *sync)
 {
@@ -1260,11 +1254,18 @@ static void unmap_bo_from_gpuvm(struct kgd_mem *mem,
 	struct amdgpu_device *adev = entry->adev;
 	struct amdgpu_vm *vm = bo_va->base.vm;
 
+	if (bo_va->queue_refcount) {
+		pr_debug("bo_va->queue_refcount %d\n", bo_va->queue_refcount);
+		return -EBUSY;
+	}
+
 	amdgpu_vm_bo_unmap(adev, bo_va, entry->va);
 
 	amdgpu_vm_clear_freed(adev, vm, &bo_va->last_pt_update);
 
 	amdgpu_sync_fence(sync, bo_va->last_pt_update);
+
+	return 0;
 }
 
 static int update_gpuvm_pte(struct kgd_mem *mem,
@@ -2191,7 +2192,10 @@ int amdgpu_amdkfd_gpuvm_unmap_memory_from_gpu(
 		pr_debug("\t unmap VA 0x%llx - 0x%llx from entry %p\n",
 			 entry->va, entry->va + bo_size, entry);
 
-		unmap_bo_from_gpuvm(mem, entry, ctx.sync);
+		ret = unmap_bo_from_gpuvm(mem, entry, ctx.sync);
+		if (ret)
+			goto unreserve_out;
+
 		entry->is_mapped = false;
 
 		mem->mapped_to_gpu_memory--;
@@ -2226,11 +2230,12 @@ int amdgpu_amdkfd_gpuvm_sync_memory(
 /**
  * amdgpu_amdkfd_map_gtt_bo_to_gart - Map BO to GART and increment reference count
  * @bo: Buffer object to be mapped
+ * @bo_gart: Return bo reference
  *
  * Before return, bo reference count is incremented. To release the reference and unpin/
  * unmap the BO, call amdgpu_amdkfd_free_gtt_mem.
  */
-int amdgpu_amdkfd_map_gtt_bo_to_gart(struct amdgpu_bo *bo)
+int amdgpu_amdkfd_map_gtt_bo_to_gart(struct amdgpu_bo *bo, struct amdgpu_bo **bo_gart)
 {
 	int ret;
 
@@ -2257,7 +2262,7 @@ int amdgpu_amdkfd_map_gtt_bo_to_gart(struct amdgpu_bo *bo)
 
 	amdgpu_bo_unreserve(bo);
 
-	bo = amdgpu_bo_ref(bo);
+	*bo_gart = amdgpu_bo_ref(bo);
 
 	return 0;
 
@@ -3200,12 +3205,13 @@ int amdgpu_amdkfd_get_tile_config(struct amdgpu_device *adev,
 	return 0;
 }
 
-bool amdgpu_amdkfd_bo_mapped_to_dev(struct amdgpu_device *adev, struct kgd_mem *mem)
+bool amdgpu_amdkfd_bo_mapped_to_dev(void *drm_priv, struct kgd_mem *mem)
 {
+	struct amdgpu_vm *vm = drm_priv_to_vm(drm_priv);
 	struct kfd_mem_attachment *entry;
 
 	list_for_each_entry(entry, &mem->attachments, list) {
-		if (entry->is_mapped && entry->adev == adev)
+		if (entry->is_mapped && entry->bo_va->base.vm == vm)
 			return true;
 	}
 	return false;

drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c

@@ -1018,8 +1018,9 @@ int amdgpu_atombios_get_clock_dividers(struct amdgpu_device *adev,
 		if (clock_type == COMPUTE_ENGINE_PLL_PARAM) {
 			args.v3.ulClockParams = cpu_to_le32((clock_type << 24) | clock);
 
-			amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args,
+			if (amdgpu_atom_execute_table(adev->mode_info.atom_context,
-				sizeof(args));
+			    index, (uint32_t *)&args, sizeof(args)))
+				return -EINVAL;
 
 			dividers->post_div = args.v3.ucPostDiv;
 			dividers->enable_post_div = (args.v3.ucCntlFlag &
@@ -1039,8 +1040,9 @@ int amdgpu_atombios_get_clock_dividers(struct amdgpu_device *adev,
 			if (strobe_mode)
 				args.v5.ucInputFlag = ATOM_PLL_INPUT_FLAG_PLL_STROBE_MODE_EN;
 
-			amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args,
+			if (amdgpu_atom_execute_table(adev->mode_info.atom_context,
-				sizeof(args));
+			    index, (uint32_t *)&args, sizeof(args)))
+				return -EINVAL;
 
 			dividers->post_div = args.v5.ucPostDiv;
 			dividers->enable_post_div = (args.v5.ucCntlFlag &
@@ -1058,8 +1060,9 @@ int amdgpu_atombios_get_clock_dividers(struct amdgpu_device *adev,
 		/* fusion */
 		args.v4.ulClock = cpu_to_le32(clock);	/* 10 khz */
 
-		amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args,
+		if (amdgpu_atom_execute_table(adev->mode_info.atom_context,
-			sizeof(args));
+		    index, (uint32_t *)&args, sizeof(args)))
+			return -EINVAL;
 
 		dividers->post_divider = dividers->post_div = args.v4.ucPostDiv;
 		dividers->real_clock = le32_to_cpu(args.v4.ulClock);
@@ -1070,8 +1073,9 @@ int amdgpu_atombios_get_clock_dividers(struct amdgpu_device *adev,
 		args.v6_in.ulClock.ulComputeClockFlag = clock_type;
 		args.v6_in.ulClock.ulClockFreq = cpu_to_le32(clock);	/* 10 khz */
 
-		amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args,
+		if (amdgpu_atom_execute_table(adev->mode_info.atom_context,
-			sizeof(args));
+		    index, (uint32_t *)&args, sizeof(args)))
+			return -EINVAL;
 
 		dividers->whole_fb_div = le16_to_cpu(args.v6_out.ulFbDiv.usFbDiv);
 		dividers->frac_fb_div = le16_to_cpu(args.v6_out.ulFbDiv.usFbDivFrac);
@@ -1113,8 +1117,9 @@ int amdgpu_atombios_get_memory_pll_dividers(struct amdgpu_device *adev,
 			if (strobe_mode)
 				args.ucInputFlag |= MPLL_INPUT_FLAG_STROBE_MODE_EN;
 
-			amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args,
+			if (amdgpu_atom_execute_table(adev->mode_info.atom_context,
-				sizeof(args));
+			    index, (uint32_t *)&args, sizeof(args)))
+				return -EINVAL;
 
 			mpll_param->clkfrac = le16_to_cpu(args.ulFbDiv.usFbDivFrac);
 			mpll_param->clkf = le16_to_cpu(args.ulFbDiv.usFbDiv);
@@ -1211,8 +1216,9 @@ int amdgpu_atombios_get_max_vddc(struct amdgpu_device *adev, u8 voltage_type,
 		args.v2.ucVoltageMode = 0;
 		args.v2.usVoltageLevel = 0;
 
-		amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args,
+		if (amdgpu_atom_execute_table(adev->mode_info.atom_context,
-			sizeof(args));
+		    index, (uint32_t *)&args, sizeof(args)))
+			return -EINVAL;
 
 		*voltage = le16_to_cpu(args.v2.usVoltageLevel);
 		break;
@@ -1221,8 +1227,9 @@ int amdgpu_atombios_get_max_vddc(struct amdgpu_device *adev, u8 voltage_type,
 		args.v3.ucVoltageMode = ATOM_GET_VOLTAGE_LEVEL;
 		args.v3.usVoltageLevel = cpu_to_le16(voltage_id);
 
-		amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args,
+		if (amdgpu_atom_execute_table(adev->mode_info.atom_context,
-			sizeof(args));
+		    index, (uint32_t *)&args, sizeof(args)))
+			return -EINVAL;
 
 		*voltage = le16_to_cpu(args.v3.usVoltageLevel);
 		break;

drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c

@@ -431,6 +431,11 @@ bool amdgpu_get_bios(struct amdgpu_device *adev)
 		goto success;
 	}
 
+	if (amdgpu_read_platform_bios(adev)) {
+		dev_info(adev->dev, "Fetched VBIOS from platform\n");
+		goto success;
+	}
+
 	if (amdgpu_read_bios(adev)) {
 		dev_info(adev->dev, "Fetched VBIOS from ROM BAR\n");
 		goto success;
@@ -446,11 +451,6 @@ bool amdgpu_get_bios(struct amdgpu_device *adev)
 		goto success;
 	}
 
-	if (amdgpu_read_platform_bios(adev)) {
-		dev_info(adev->dev, "Fetched VBIOS from platform\n");
-		goto success;
-	}
-
 	dev_err(adev->dev, "Unable to locate a BIOS ROM\n");
 	return false;
 

drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c

@@ -414,7 +414,7 @@ static int amdgpu_cgs_get_firmware_info(struct cgs_device *cgs_device,
 				return -EINVAL;
 			}
 
-			err = amdgpu_ucode_request(adev, &adev->pm.fw, fw_name);
+			err = amdgpu_ucode_request(adev, &adev->pm.fw, "%s", fw_name);
 			if (err) {
 				DRM_ERROR("Failed to load firmware \"%s\"", fw_name);
 				amdgpu_ucode_release(&adev->pm.fw);

drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c

@@ -249,11 +249,7 @@ amdgpu_connector_find_encoder(struct drm_connector *connector,
 static struct edid *
 amdgpu_connector_get_hardcoded_edid(struct amdgpu_device *adev)
 {
-	if (adev->mode_info.bios_hardcoded_edid) {
+	return drm_edid_duplicate(drm_edid_raw(adev->mode_info.bios_hardcoded_edid));
-		return kmemdup((unsigned char *)adev->mode_info.bios_hardcoded_edid,
-			       adev->mode_info.bios_hardcoded_edid_size, GFP_KERNEL);
-	}
-	return NULL;
 }
 
 static void amdgpu_connector_get_edid(struct drm_connector *connector)
@@ -442,6 +438,9 @@ static void amdgpu_connector_add_common_modes(struct drm_encoder *encoder,
 			continue;
 
 		mode = drm_cvt_mode(dev, common_modes[i].w, common_modes[i].h, 60, false, false, false);
+		if (!mode)
+			return;
+
 		drm_mode_probed_add(connector, mode);
 	}
 }

drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c

@@ -263,6 +263,10 @@ static int amdgpu_cs_pass1(struct amdgpu_cs_parser *p,
 			if (size < sizeof(struct drm_amdgpu_bo_list_in))
 				goto free_partial_kdata;
 
+			/* Only a single BO list is allowed to simplify handling. */
+			if (p->bo_list)
+				ret = -EINVAL;
+
 			ret = amdgpu_cs_p1_bo_handles(p, p->chunks[i].kdata);
 			if (ret)
 				goto free_partial_kdata;
@@ -292,6 +296,7 @@ static int amdgpu_cs_pass1(struct amdgpu_cs_parser *p,
 				       num_ibs[i], &p->jobs[i]);
 		if (ret)
 			goto free_all_kdata;
+		p->jobs[i]->enforce_isolation = p->adev->enforce_isolation[fpriv->xcp_id];
 	}
 	p->gang_leader = p->jobs[p->gang_leader_idx];
 
@@ -1106,7 +1111,7 @@ static int amdgpu_cs_vm_handling(struct amdgpu_cs_parser *p)
 			struct drm_gpu_scheduler *sched = entity->rq->sched;
 			struct amdgpu_ring *ring = to_amdgpu_ring(sched);
 
-			if (amdgpu_vmid_uses_reserved(vm, ring->vm_hub))
+			if (amdgpu_vmid_uses_reserved(adev, vm, ring->vm_hub))
 				return -EINVAL;
 		}
 	}

drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c

@@ -2026,100 +2026,6 @@ DEFINE_DEBUGFS_ATTRIBUTE(fops_ib_preempt, NULL,
 DEFINE_DEBUGFS_ATTRIBUTE(fops_sclk_set, NULL,
 			amdgpu_debugfs_sclk_set, "%llu\n");
 
-static ssize_t amdgpu_reset_dump_register_list_read(struct file *f,
-				char __user *buf, size_t size, loff_t *pos)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)file_inode(f)->i_private;
-	char reg_offset[12];
-	int i, ret, len = 0;
-
-	if (*pos)
-		return 0;
-
-	memset(reg_offset, 0, 12);
-	ret = down_read_killable(&adev->reset_domain->sem);
-	if (ret)
-		return ret;
-
-	for (i = 0; i < adev->reset_info.num_regs; i++) {
-		sprintf(reg_offset, "0x%x\n", adev->reset_info.reset_dump_reg_list[i]);
-		up_read(&adev->reset_domain->sem);
-		if (copy_to_user(buf + len, reg_offset, strlen(reg_offset)))
-			return -EFAULT;
-
-		len += strlen(reg_offset);
-		ret = down_read_killable(&adev->reset_domain->sem);
-		if (ret)
-			return ret;
-	}
-
-	up_read(&adev->reset_domain->sem);
-	*pos += len;
-
-	return len;
-}
-
-static ssize_t amdgpu_reset_dump_register_list_write(struct file *f,
-			const char __user *buf, size_t size, loff_t *pos)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *)file_inode(f)->i_private;
-	char reg_offset[11];
-	uint32_t *new = NULL, *tmp = NULL;
-	unsigned int len = 0;
-	int ret, i = 0;
-
-	do {
-		memset(reg_offset, 0, 11);
-		if (copy_from_user(reg_offset, buf + len,
-					min(10, (size-len)))) {
-			ret = -EFAULT;
-			goto error_free;
-		}
-
-		new = krealloc_array(tmp, i + 1, sizeof(uint32_t), GFP_KERNEL);
-		if (!new) {
-			ret = -ENOMEM;
-			goto error_free;
-		}
-		tmp = new;
-		if (sscanf(reg_offset, "%X %n", &tmp[i], &ret) != 1) {
-			ret = -EINVAL;
-			goto error_free;
-		}
-
-		len += ret;
-		i++;
-	} while (len < size);
-
-	new = kmalloc_array(i, sizeof(uint32_t), GFP_KERNEL);
-	if (!new) {
-		ret = -ENOMEM;
-		goto error_free;
-	}
-	ret = down_write_killable(&adev->reset_domain->sem);
-	if (ret)
-		goto error_free;
-
-	swap(adev->reset_info.reset_dump_reg_list, tmp);
-	swap(adev->reset_info.reset_dump_reg_value, new);
-	adev->reset_info.num_regs = i;
-	up_write(&adev->reset_domain->sem);
-	ret = size;
-
-error_free:
-	if (tmp != new)
-		kfree(tmp);
-	kfree(new);
-	return ret;
-}
-
-static const struct file_operations amdgpu_reset_dump_register_list = {
-	.owner = THIS_MODULE,
-	.read = amdgpu_reset_dump_register_list_read,
-	.write = amdgpu_reset_dump_register_list_write,
-	.llseek = default_llseek
-};
-
 int amdgpu_debugfs_init(struct amdgpu_device *adev)
 {
 	struct dentry *root = adev_to_drm(adev)->primary->debugfs_root;
@@ -2204,8 +2110,6 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev)
 			    &amdgpu_debugfs_vm_info_fops);
 	debugfs_create_file("amdgpu_benchmark", 0200, root, adev,
 			    &amdgpu_benchmark_fops);
-	debugfs_create_file("amdgpu_reset_dump_register_list", 0644, root, adev,
-			    &amdgpu_reset_dump_register_list);
 
 	adev->debugfs_vbios_blob.data = adev->bios;
 	adev->debugfs_vbios_blob.size = adev->bios_size;

drivers/gpu/drm/amd/amdgpu/amdgpu_dev_coredump.c

@@ -28,8 +28,8 @@
 #include "atom.h"
 
 #ifndef CONFIG_DEV_COREDUMP
-void amdgpu_coredump(struct amdgpu_device *adev, bool vram_lost,
+void amdgpu_coredump(struct amdgpu_device *adev, bool skip_vram_check,
-		     struct amdgpu_reset_context *reset_context)
+		     bool vram_lost, struct amdgpu_job *job)
 {
 }
 #else
@@ -203,7 +203,7 @@ amdgpu_devcoredump_read(char *buffer, loff_t offset, size_t count,
 	struct amdgpu_coredump_info *coredump = data;
 	struct drm_print_iterator iter;
 	struct amdgpu_vm_fault_info *fault_info;
-	int i, ver;
+	int ver;
 
 	iter.data = buffer;
 	iter.offset = 0;
@@ -236,7 +236,7 @@ amdgpu_devcoredump_read(char *buffer, loff_t offset, size_t count,
 	drm_printf(&p, "\nSOC Memory Information\n");
 	drm_printf(&p, "real vram size: %llu\n", coredump->adev->gmc.real_vram_size);
 	drm_printf(&p, "visible vram size: %llu\n", coredump->adev->gmc.visible_vram_size);
-	drm_printf(&p, "visible vram size: %llu\n", coredump->adev->mman.gtt_mgr.manager.size);
+	drm_printf(&p, "gtt size: %llu\n", coredump->adev->mman.gtt_mgr.manager.size);
 
 	/* GDS Config */
 	drm_printf(&p, "\nGDS Config\n");
@@ -315,16 +315,10 @@ amdgpu_devcoredump_read(char *buffer, loff_t offset, size_t count,
 		}
 	}
 
-	if (coredump->reset_vram_lost)
+	if (coredump->skip_vram_check)
+		drm_printf(&p, "VRAM lost check is skipped!\n");
+	else if (coredump->reset_vram_lost)
 		drm_printf(&p, "VRAM is lost due to GPU reset!\n");
-	if (coredump->adev->reset_info.num_regs) {
-		drm_printf(&p, "AMDGPU register dumps:\nOffset:     Value:\n");
-
-		for (i = 0; i < coredump->adev->reset_info.num_regs; i++)
-			drm_printf(&p, "0x%08x: 0x%08x\n",
-				   coredump->adev->reset_info.reset_dump_reg_list[i],
-				   coredump->adev->reset_info.reset_dump_reg_value[i]);
-	}
 
 	return count - iter.remain;
 }
@@ -334,12 +328,11 @@ static void amdgpu_devcoredump_free(void *data)
 	kfree(data);
 }
 
-void amdgpu_coredump(struct amdgpu_device *adev, bool vram_lost,
+void amdgpu_coredump(struct amdgpu_device *adev, bool skip_vram_check,
-		     struct amdgpu_reset_context *reset_context)
+		     bool vram_lost, struct amdgpu_job *job)
 {
-	struct amdgpu_coredump_info *coredump;
 	struct drm_device *dev = adev_to_drm(adev);
-	struct amdgpu_job *job = reset_context->job;
+	struct amdgpu_coredump_info *coredump;
 	struct drm_sched_job *s_job;
 
 	coredump = kzalloc(sizeof(*coredump), GFP_NOWAIT);
@@ -349,11 +342,12 @@ void amdgpu_coredump(struct amdgpu_device *adev, bool vram_lost,
 		return;
 	}
 
+	coredump->skip_vram_check = skip_vram_check;
 	coredump->reset_vram_lost = vram_lost;
 
-	if (reset_context->job && reset_context->job->vm) {
+	if (job && job->vm) {
+		struct amdgpu_vm *vm = job->vm;
 		struct amdgpu_task_info *ti;
-		struct amdgpu_vm *vm = reset_context->job->vm;
 
 		ti = amdgpu_vm_get_task_info_vm(vm);
 		if (ti) {

drivers/gpu/drm/amd/amdgpu/amdgpu_dev_coredump.h

@@ -26,7 +26,6 @@
 #define __AMDGPU_DEV_COREDUMP_H__
 
 #include "amdgpu.h"
-#include "amdgpu_reset.h"
 
 #ifdef CONFIG_DEV_COREDUMP
 
@@ -36,12 +35,12 @@ struct amdgpu_coredump_info {
 	struct amdgpu_device            *adev;
 	struct amdgpu_task_info         reset_task_info;
 	struct timespec64               reset_time;
+	bool                            skip_vram_check;
 	bool                            reset_vram_lost;
 	struct amdgpu_ring              *ring;
 };
 #endif
 
-void amdgpu_coredump(struct amdgpu_device *adev, bool vram_lost,
+void amdgpu_coredump(struct amdgpu_device *adev, bool skip_vram_check,
-		     struct amdgpu_reset_context *reset_context);
+		     bool vram_lost, struct amdgpu_job *job);
-
 #endif

drivers/gpu/drm/amd/amdgpu/amdgpu_device.c

@@ -1916,6 +1916,8 @@ static int amdgpu_device_init_apu_flags(struct amdgpu_device *adev)
  */
 static int amdgpu_device_check_arguments(struct amdgpu_device *adev)
 {
+	int i;
+
 	if (amdgpu_sched_jobs < 4) {
 		dev_warn(adev->dev, "sched jobs (%d) must be at least 4\n",
 			 amdgpu_sched_jobs);
@@ -1970,6 +1972,9 @@ static int amdgpu_device_check_arguments(struct amdgpu_device *adev)
 
 	adev->firmware.load_type = amdgpu_ucode_get_load_type(adev, amdgpu_fw_load_type);
 
+	for (i = 0; i < MAX_XCP; i++)
+		adev->enforce_isolation[i] = !!enforce_isolation;
+
 	return 0;
 }
 
@@ -2471,6 +2476,7 @@ out:
  */
 static int amdgpu_device_ip_early_init(struct amdgpu_device *adev)
 {
+	struct amdgpu_ip_block *ip_block;
 	struct pci_dev *parent;
 	int i, r;
 	bool total;
@@ -2608,7 +2614,10 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev)
 	if (!total)
 		return -ENODEV;
 
+	ip_block = amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_GFX);
+	if (ip_block->status.valid != false)
 		amdgpu_amdkfd_device_probe(adev);
+
 	adev->cg_flags &= amdgpu_cg_mask;
 	adev->pg_flags &= amdgpu_pg_mask;
 
@@ -3948,6 +3957,27 @@ static void amdgpu_device_check_iommu_direct_map(struct amdgpu_device *adev)
 		adev->ram_is_direct_mapped = true;
 }
 
+#if defined(CONFIG_HSA_AMD_P2P)
+/**
+ * amdgpu_device_check_iommu_remap - Check if DMA remapping is enabled.
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * return if IOMMU remapping bar address
+ */
+static bool amdgpu_device_check_iommu_remap(struct amdgpu_device *adev)
+{
+	struct iommu_domain *domain;
+
+	domain = iommu_get_domain_for_dev(adev->dev);
+	if (domain && (domain->type == IOMMU_DOMAIN_DMA ||
+		domain->type ==	IOMMU_DOMAIN_DMA_FQ))
+		return true;
+
+	return false;
+}
+#endif
+
 static const struct attribute *amdgpu_dev_attributes[] = {
 	&dev_attr_pcie_replay_count.attr,
 	NULL
@@ -4055,6 +4085,10 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 	mutex_init(&adev->notifier_lock);
 	mutex_init(&adev->pm.stable_pstate_ctx_lock);
 	mutex_init(&adev->benchmark_mutex);
+	mutex_init(&adev->gfx.reset_sem_mutex);
+	/* Initialize the mutex for cleaner shader isolation between GFX and compute processes */
+	mutex_init(&adev->enforce_isolation_mutex);
+	mutex_init(&adev->gfx.kfd_sch_mutex);
 
 	amdgpu_device_init_apu_flags(adev);
 
@@ -4086,6 +4120,21 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 			  amdgpu_device_delayed_init_work_handler);
 	INIT_DELAYED_WORK(&adev->gfx.gfx_off_delay_work,
 			  amdgpu_device_delay_enable_gfx_off);
+	/*
+	 * Initialize the enforce_isolation work structures for each XCP
+	 * partition.  This work handler is responsible for enforcing shader
+	 * isolation on AMD GPUs.  It counts the number of emitted fences for
+	 * each GFX and compute ring.  If there are any fences, it schedules
+	 * the `enforce_isolation_work` to be run after a delay.  If there are
+	 * no fences, it signals the Kernel Fusion Driver (KFD) to resume the
+	 * runqueue.
+	 */
+	for (i = 0; i < MAX_XCP; i++) {
+		INIT_DELAYED_WORK(&adev->gfx.enforce_isolation[i].work,
+				  amdgpu_gfx_enforce_isolation_handler);
+		adev->gfx.enforce_isolation[i].adev = adev;
+		adev->gfx.enforce_isolation[i].xcp_id = i;
+	}
 
 	INIT_WORK(&adev->xgmi_reset_work, amdgpu_device_xgmi_reset_func);
 
@@ -4482,6 +4531,9 @@ void amdgpu_device_fini_hw(struct amdgpu_device *adev)
 {
 	dev_info(adev->dev, "amdgpu: finishing device.\n");
 	flush_delayed_work(&adev->delayed_init_work);
+
+	if (adev->mman.initialized)
+		drain_workqueue(adev->mman.bdev.wq);
 	adev->shutdown = true;
 
 	/* make sure IB test finished before entering exclusive mode
@@ -4502,9 +4554,6 @@ void amdgpu_device_fini_hw(struct amdgpu_device *adev)
 	}
 	amdgpu_fence_driver_hw_fini(adev);
 
-	if (adev->mman.initialized)
-		drain_workqueue(adev->mman.bdev.wq);
-
 	if (adev->pm.sysfs_initialized)
 		amdgpu_pm_sysfs_fini(adev);
 	if (adev->ucode_sysfs_en)
@@ -5278,16 +5327,15 @@ int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev,
 {
 	int i, r = 0;
 	struct amdgpu_job *job = NULL;
+	struct amdgpu_device *tmp_adev = reset_context->reset_req_dev;
 	bool need_full_reset =
 		test_bit(AMDGPU_NEED_FULL_RESET, &reset_context->flags);
 
 	if (reset_context->reset_req_dev == adev)
 		job = reset_context->job;
 
-	if (amdgpu_sriov_vf(adev)) {
+	if (amdgpu_sriov_vf(adev))
-		/* stop the data exchange thread */
+		amdgpu_virt_pre_reset(adev);
-		amdgpu_virt_fini_data_exchange(adev);
-	}
 
 	amdgpu_fence_driver_isr_toggle(adev, true);
 
@@ -5336,6 +5384,16 @@ int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev,
 			}
 		}
 
+		if (!test_bit(AMDGPU_SKIP_COREDUMP, &reset_context->flags)) {
+			dev_info(tmp_adev->dev, "Dumping IP State\n");
+			/* Trigger ip dump before we reset the asic */
+			for (i = 0; i < tmp_adev->num_ip_blocks; i++)
+				if (tmp_adev->ip_blocks[i].version->funcs->dump_ip_state)
+					tmp_adev->ip_blocks[i].version->funcs
+						->dump_ip_state((void *)tmp_adev);
+			dev_info(tmp_adev->dev, "Dumping IP State Completed\n");
+		}
+
 		if (need_full_reset)
 			r = amdgpu_device_ip_suspend(adev);
 		if (need_full_reset)
@@ -5348,47 +5406,17 @@ int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev,
 	return r;
 }
 
-static int amdgpu_reset_reg_dumps(struct amdgpu_device *adev)
-{
-	int i;
-
-	lockdep_assert_held(&adev->reset_domain->sem);
-
-	for (i = 0; i < adev->reset_info.num_regs; i++) {
-		adev->reset_info.reset_dump_reg_value[i] =
-			RREG32(adev->reset_info.reset_dump_reg_list[i]);
-
-		trace_amdgpu_reset_reg_dumps(adev->reset_info.reset_dump_reg_list[i],
-					     adev->reset_info.reset_dump_reg_value[i]);
-	}
-
-	return 0;
-}
-
 int amdgpu_do_asic_reset(struct list_head *device_list_handle,
 			 struct amdgpu_reset_context *reset_context)
 {
 	struct amdgpu_device *tmp_adev = NULL;
 	bool need_full_reset, skip_hw_reset, vram_lost = false;
 	int r = 0;
-	uint32_t i;
 
 	/* Try reset handler method first */
 	tmp_adev = list_first_entry(device_list_handle, struct amdgpu_device,
 				    reset_list);
 
-	if (!test_bit(AMDGPU_SKIP_COREDUMP, &reset_context->flags)) {
-		amdgpu_reset_reg_dumps(tmp_adev);
-
-		dev_info(tmp_adev->dev, "Dumping IP State\n");
-		/* Trigger ip dump before we reset the asic */
-		for (i = 0; i < tmp_adev->num_ip_blocks; i++)
-			if (tmp_adev->ip_blocks[i].version->funcs->dump_ip_state)
-				tmp_adev->ip_blocks[i].version->funcs
-				->dump_ip_state((void *)tmp_adev);
-		dev_info(tmp_adev->dev, "Dumping IP State Completed\n");
-	}
-
 	reset_context->reset_device_list = device_list_handle;
 	r = amdgpu_reset_perform_reset(tmp_adev, reset_context);
 	/* If reset handler not implemented, continue; otherwise return */
@@ -5461,7 +5489,7 @@ int amdgpu_do_asic_reset(struct list_head *device_list_handle,
 				vram_lost = amdgpu_device_check_vram_lost(tmp_adev);
 
 				if (!test_bit(AMDGPU_SKIP_COREDUMP, &reset_context->flags))
-					amdgpu_coredump(tmp_adev, vram_lost, reset_context);
+					amdgpu_coredump(tmp_adev, false, vram_lost, reset_context->job);
 
 				if (vram_lost) {
 					DRM_INFO("VRAM is lost due to GPU reset!\n");
@@ -5513,7 +5541,7 @@ int amdgpu_do_asic_reset(struct list_head *device_list_handle,
 				 * bad_page_threshold value to fix this once
 				 * probing driver again.
 				 */
-				if (!amdgpu_ras_eeprom_check_err_threshold(tmp_adev)) {
+				if (!amdgpu_ras_is_rma(tmp_adev)) {
 					/* must succeed. */
 					amdgpu_ras_resume(tmp_adev);
 				} else {
@@ -5879,7 +5907,7 @@ skip_hw_reset:
 			if (!amdgpu_ring_sched_ready(ring))
 				continue;
 
-			drm_sched_start(&ring->sched, true);
+			drm_sched_start(&ring->sched);
 		}
 
 		if (!drm_drv_uses_atomic_modeset(adev_to_drm(tmp_adev)) && !job_signaled)
@@ -5891,8 +5919,14 @@ skip_hw_reset:
 		tmp_adev->asic_reset_res = 0;
 
 		if (r) {
-			/* bad news, how to tell it to userspace ? */
+			/* bad news, how to tell it to userspace ?
-			dev_info(tmp_adev->dev, "GPU reset(%d) failed\n", atomic_read(&tmp_adev->gpu_reset_counter));
+			 * for ras error, we should report GPU bad status instead of
+			 * reset failure
+			 */
+			if (reset_context->src != AMDGPU_RESET_SRC_RAS ||
+			    !amdgpu_ras_eeprom_check_err_threshold(tmp_adev))
+				dev_info(tmp_adev->dev, "GPU reset(%d) failed\n",
+					atomic_read(&tmp_adev->gpu_reset_counter));
 			amdgpu_vf_error_put(tmp_adev, AMDGIM_ERROR_VF_GPU_RESET_FAIL, 0, r);
 		} else {
 			dev_info(tmp_adev->dev, "GPU reset(%d) succeeded!\n", atomic_read(&tmp_adev->gpu_reset_counter));
@@ -6138,18 +6172,24 @@ bool amdgpu_device_is_peer_accessible(struct amdgpu_device *adev,
 				      struct amdgpu_device *peer_adev)
 {
 #ifdef CONFIG_HSA_AMD_P2P
-	uint64_t address_mask = peer_adev->dev->dma_mask ?
-		~*peer_adev->dev->dma_mask : ~((1ULL << 32) - 1);
-	resource_size_t aper_limit =
-		adev->gmc.aper_base + adev->gmc.aper_size - 1;
 	bool p2p_access =
 		!adev->gmc.xgmi.connected_to_cpu &&
 		!(pci_p2pdma_distance(adev->pdev, peer_adev->dev, false) < 0);
 
-	return pcie_p2p && p2p_access && (adev->gmc.visible_vram_size &&
+	bool is_large_bar = adev->gmc.visible_vram_size &&
-		adev->gmc.real_vram_size == adev->gmc.visible_vram_size &&
+		adev->gmc.real_vram_size == adev->gmc.visible_vram_size;
-		!(adev->gmc.aper_base & address_mask ||
+	bool p2p_addressable = amdgpu_device_check_iommu_remap(peer_adev);
-		  aper_limit & address_mask));
+
+	if (!p2p_addressable) {
+		uint64_t address_mask = peer_adev->dev->dma_mask ?
+			~*peer_adev->dev->dma_mask : ~((1ULL << 32) - 1);
+		resource_size_t aper_limit =
+			adev->gmc.aper_base + adev->gmc.aper_size - 1;
+
+		p2p_addressable = !(adev->gmc.aper_base & address_mask ||
+				     aper_limit & address_mask);
+	}
+	return is_large_bar && p2p_access && p2p_addressable;
 #else
 	return false;
 #endif
@@ -6374,7 +6414,7 @@ void amdgpu_pci_resume(struct pci_dev *pdev)
 		if (!amdgpu_ring_sched_ready(ring))
 			continue;
 
-		drm_sched_start(&ring->sched, true);
+		drm_sched_start(&ring->sched);
 	}
 
 	amdgpu_device_unset_mp1_state(adev);

drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c

@@ -131,6 +131,7 @@ enum AMDGPU_DEBUG_MASK {
 	AMDGPU_DEBUG_DISABLE_GPU_SOFT_RECOVERY = BIT(2),
 	AMDGPU_DEBUG_USE_VRAM_FW_BUF = BIT(3),
 	AMDGPU_DEBUG_ENABLE_RAS_ACA = BIT(4),
+	AMDGPU_DEBUG_ENABLE_EXP_RESETS = BIT(5),
 };
 
 unsigned int amdgpu_vram_limit = UINT_MAX;
@@ -168,6 +169,16 @@ uint amdgpu_sdma_phase_quantum = 32;
 char *amdgpu_disable_cu;
 char *amdgpu_virtual_display;
 bool enforce_isolation;
+
+/* Specifies the default granularity for SVM, used in buffer
+ * migration and restoration of backing memory when handling
+ * recoverable page faults.
+ *
+ * The value is given as log(numPages(buffer)); for a 2 MiB
+ * buffer it computes to be 9
+ */
+uint amdgpu_svm_default_granularity = 9;
+
 /*
  * OverDrive(bit 14) disabled by default
  * GFX DCS(bit 19) disabled by default
@@ -319,6 +330,13 @@ module_param_named(pcie_gen2, amdgpu_pcie_gen2, int, 0444);
 MODULE_PARM_DESC(msi, "MSI support (1 = enable, 0 = disable, -1 = auto)");
 module_param_named(msi, amdgpu_msi, int, 0444);
 
+/**
+ * DOC: svm_default_granularity (uint)
+ * Used in buffer migration and handling of recoverable page faults
+ */
+MODULE_PARM_DESC(svm_default_granularity, "SVM's default granularity in log(2^Pages), default 9 = 2^9 = 2 MiB");
+module_param_named(svm_default_granularity, amdgpu_svm_default_granularity, uint, 0644);
+
 /**
  * DOC: lockup_timeout (string)
  * Set GPU scheduler timeout value in ms.
@@ -2199,6 +2217,11 @@ static void amdgpu_init_debug_options(struct amdgpu_device *adev)
 		pr_info("debug: enable RAS ACA\n");
 		adev->debug_enable_ras_aca = true;
 	}
+
+	if (amdgpu_debug_mask & AMDGPU_DEBUG_ENABLE_EXP_RESETS) {
+		pr_info("debug: enable experimental reset features\n");
+		adev->debug_exp_resets = true;
+	}
 }
 
 static unsigned long amdgpu_fix_asic_type(struct pci_dev *pdev, unsigned long flags)
@@ -2954,7 +2977,6 @@ static const struct drm_driver amdgpu_kms_driver = {
 	    DRIVER_SYNCOBJ_TIMELINE,
 	.open = amdgpu_driver_open_kms,
 	.postclose = amdgpu_driver_postclose_kms,
-	.lastclose = amdgpu_driver_lastclose_kms,
 	.ioctls = amdgpu_ioctls_kms,
 	.num_ioctls = ARRAY_SIZE(amdgpu_ioctls_kms),
 	.dumb_create = amdgpu_mode_dumb_create,
@@ -2981,7 +3003,6 @@ const struct drm_driver amdgpu_partition_driver = {
 	    DRIVER_SYNCOBJ_TIMELINE,
 	.open = amdgpu_driver_open_kms,
 	.postclose = amdgpu_driver_postclose_kms,
-	.lastclose = amdgpu_driver_lastclose_kms,
 	.ioctls = amdgpu_ioctls_kms,
 	.num_ioctls = ARRAY_SIZE(amdgpu_ioctls_kms),
 	.dumb_create = amdgpu_mode_dumb_create,

drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h

@@ -55,8 +55,6 @@ int amdgpu_gart_table_ram_alloc(struct amdgpu_device *adev);
 void amdgpu_gart_table_ram_free(struct amdgpu_device *adev);
 int amdgpu_gart_table_vram_alloc(struct amdgpu_device *adev);
 void amdgpu_gart_table_vram_free(struct amdgpu_device *adev);
-int amdgpu_gart_table_vram_pin(struct amdgpu_device *adev);
-void amdgpu_gart_table_vram_unpin(struct amdgpu_device *adev);
 int amdgpu_gart_init(struct amdgpu_device *adev);
 void amdgpu_gart_dummy_page_fini(struct amdgpu_device *adev);
 void amdgpu_gart_unbind(struct amdgpu_device *adev, uint64_t offset,

drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c

@@ -24,10 +24,13 @@
  */
 
 #include <linux/firmware.h>
+#include <linux/pm_runtime.h>
+
 #include "amdgpu.h"
 #include "amdgpu_gfx.h"
 #include "amdgpu_rlc.h"
 #include "amdgpu_ras.h"
+#include "amdgpu_reset.h"
 #include "amdgpu_xcp.h"
 #include "amdgpu_xgmi.h"
 
@@ -882,8 +885,11 @@ int amdgpu_gfx_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *r
 	int r;
 
 	if (amdgpu_ras_is_supported(adev, ras_block->block)) {
-		if (!amdgpu_persistent_edc_harvesting_supported(adev))
+		if (!amdgpu_persistent_edc_harvesting_supported(adev)) {
-			amdgpu_ras_reset_error_status(adev, AMDGPU_RAS_BLOCK__GFX);
+			r = amdgpu_ras_reset_error_status(adev, AMDGPU_RAS_BLOCK__GFX);
+			if (r)
+				return r;
+		}
 
 		r = amdgpu_ras_block_late_init(adev, ras_block);
 		if (r)
@@ -1027,7 +1033,10 @@ uint32_t amdgpu_kiq_rreg(struct amdgpu_device *adev, uint32_t reg, uint32_t xcc_
 		pr_err("critical bug! too many kiq readers\n");
 		goto failed_unlock;
 	}
-	amdgpu_ring_alloc(ring, 32);
+	r = amdgpu_ring_alloc(ring, 32);
+	if (r)
+		goto failed_unlock;
+
 	amdgpu_ring_emit_rreg(ring, reg, reg_val_offs);
 	r = amdgpu_fence_emit_polling(ring, &seq, MAX_KIQ_REG_WAIT);
 	if (r)
@@ -1093,7 +1102,10 @@ void amdgpu_kiq_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v, uint3
 	}
 
 	spin_lock_irqsave(&kiq->ring_lock, flags);
-	amdgpu_ring_alloc(ring, 32);
+	r = amdgpu_ring_alloc(ring, 32);
+	if (r)
+		goto failed_unlock;
+
 	amdgpu_ring_emit_wreg(ring, reg, v);
 	r = amdgpu_fence_emit_polling(ring, &seq, MAX_KIQ_REG_WAIT);
 	if (r)
@@ -1129,6 +1141,7 @@ void amdgpu_kiq_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v, uint3
 
 failed_undo:
 	amdgpu_ring_undo(ring);
+failed_unlock:
 	spin_unlock_irqrestore(&kiq->ring_lock, flags);
 failed_kiq_write:
 	dev_err(adev->dev, "failed to write reg:%x\n", reg);
@@ -1381,6 +1394,217 @@ static ssize_t amdgpu_gfx_get_available_compute_partition(struct device *dev,
 	return sysfs_emit(buf, "%s\n", supported_partition);
 }
 
+static int amdgpu_gfx_run_cleaner_shader_job(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+	struct drm_gpu_scheduler *sched = &ring->sched;
+	struct drm_sched_entity entity;
+	struct dma_fence *f;
+	struct amdgpu_job *job;
+	struct amdgpu_ib *ib;
+	int i, r;
+
+	/* Initialize the scheduler entity */
+	r = drm_sched_entity_init(&entity, DRM_SCHED_PRIORITY_NORMAL,
+				  &sched, 1, NULL);
+	if (r) {
+		dev_err(adev->dev, "Failed setting up GFX kernel entity.\n");
+		goto err;
+	}
+
+	r = amdgpu_job_alloc_with_ib(ring->adev, &entity, NULL,
+				     64, 0,
+				     &job);
+	if (r)
+		goto err;
+
+	job->enforce_isolation = true;
+
+	ib = &job->ibs[0];
+	for (i = 0; i <= ring->funcs->align_mask; ++i)
+		ib->ptr[i] = ring->funcs->nop;
+	ib->length_dw = ring->funcs->align_mask + 1;
+
+	f = amdgpu_job_submit(job);
+
+	r = dma_fence_wait(f, false);
+	if (r)
+		goto err;
+
+	dma_fence_put(f);
+
+	/* Clean up the scheduler entity */
+	drm_sched_entity_destroy(&entity);
+	return 0;
+
+err:
+	return r;
+}
+
+static int amdgpu_gfx_run_cleaner_shader(struct amdgpu_device *adev, int xcp_id)
+{
+	int num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	struct amdgpu_ring *ring;
+	int num_xcc_to_clear;
+	int i, r, xcc_id;
+
+	if (adev->gfx.num_xcc_per_xcp)
+		num_xcc_to_clear = adev->gfx.num_xcc_per_xcp;
+	else
+		num_xcc_to_clear = 1;
+
+	for (xcc_id = 0; xcc_id < num_xcc; xcc_id++) {
+		for (i = 0; i < adev->gfx.num_compute_rings; i++) {
+			ring = &adev->gfx.compute_ring[i + xcc_id * adev->gfx.num_compute_rings];
+			if ((ring->xcp_id == xcp_id) && ring->sched.ready) {
+				r = amdgpu_gfx_run_cleaner_shader_job(ring);
+				if (r)
+					return r;
+				num_xcc_to_clear--;
+				break;
+			}
+		}
+	}
+
+	if (num_xcc_to_clear)
+		return -ENOENT;
+
+	return 0;
+}
+
+static ssize_t amdgpu_gfx_set_run_cleaner_shader(struct device *dev,
+						 struct device_attribute *attr,
+						 const char *buf,
+						 size_t count)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	int ret;
+	long value;
+
+	if (amdgpu_in_reset(adev))
+		return -EPERM;
+	if (adev->in_suspend && !adev->in_runpm)
+		return -EPERM;
+
+	ret = kstrtol(buf, 0, &value);
+
+	if (ret)
+		return -EINVAL;
+
+	if (value < 0)
+		return -EINVAL;
+
+	if (adev->xcp_mgr) {
+		if (value >= adev->xcp_mgr->num_xcps)
+			return -EINVAL;
+	} else {
+		if (value > 1)
+			return -EINVAL;
+	}
+
+	ret = pm_runtime_get_sync(ddev->dev);
+	if (ret < 0) {
+		pm_runtime_put_autosuspend(ddev->dev);
+		return ret;
+	}
+
+	ret = amdgpu_gfx_run_cleaner_shader(adev, value);
+
+	pm_runtime_mark_last_busy(ddev->dev);
+	pm_runtime_put_autosuspend(ddev->dev);
+
+	if (ret)
+		return ret;
+
+	return count;
+}
+
+static ssize_t amdgpu_gfx_get_enforce_isolation(struct device *dev,
+						struct device_attribute *attr,
+						char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	int i;
+	ssize_t size = 0;
+
+	if (adev->xcp_mgr) {
+		for (i = 0; i < adev->xcp_mgr->num_xcps; i++) {
+			size += sysfs_emit_at(buf, size, "%u", adev->enforce_isolation[i]);
+			if (i < (adev->xcp_mgr->num_xcps - 1))
+				size += sysfs_emit_at(buf, size, " ");
+		}
+		buf[size++] = '\n';
+	} else {
+		size = sysfs_emit_at(buf, 0, "%u\n", adev->enforce_isolation[0]);
+	}
+
+	return size;
+}
+
+static ssize_t amdgpu_gfx_set_enforce_isolation(struct device *dev,
+						struct device_attribute *attr,
+						const char *buf, size_t count)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	long partition_values[MAX_XCP] = {0};
+	int ret, i, num_partitions;
+	const char *input_buf = buf;
+
+	for (i = 0; i < (adev->xcp_mgr ? adev->xcp_mgr->num_xcps : 1); i++) {
+		ret = sscanf(input_buf, "%ld", &partition_values[i]);
+		if (ret <= 0)
+			break;
+
+		/* Move the pointer to the next value in the string */
+		input_buf = strchr(input_buf, ' ');
+		if (input_buf) {
+			input_buf++;
+		} else {
+			i++;
+			break;
+		}
+	}
+	num_partitions = i;
+
+	if (adev->xcp_mgr && num_partitions != adev->xcp_mgr->num_xcps)
+		return -EINVAL;
+
+	if (!adev->xcp_mgr && num_partitions != 1)
+		return -EINVAL;
+
+	for (i = 0; i < num_partitions; i++) {
+		if (partition_values[i] != 0 && partition_values[i] != 1)
+			return -EINVAL;
+	}
+
+	mutex_lock(&adev->enforce_isolation_mutex);
+
+	for (i = 0; i < num_partitions; i++) {
+		if (adev->enforce_isolation[i] && !partition_values[i]) {
+			/* Going from enabled to disabled */
+			amdgpu_vmid_free_reserved(adev, AMDGPU_GFXHUB(i));
+		} else if (!adev->enforce_isolation[i] && partition_values[i]) {
+			/* Going from disabled to enabled */
+			amdgpu_vmid_alloc_reserved(adev, AMDGPU_GFXHUB(i));
+		}
+		adev->enforce_isolation[i] = partition_values[i];
+	}
+
+	mutex_unlock(&adev->enforce_isolation_mutex);
+
+	return count;
+}
+
+static DEVICE_ATTR(run_cleaner_shader, 0200,
+		   NULL, amdgpu_gfx_set_run_cleaner_shader);
+
+static DEVICE_ATTR(enforce_isolation, 0644,
+		   amdgpu_gfx_get_enforce_isolation,
+		   amdgpu_gfx_set_enforce_isolation);
+
 static DEVICE_ATTR(current_compute_partition, 0644,
 		   amdgpu_gfx_get_current_compute_partition,
 		   amdgpu_gfx_set_compute_partition);
@@ -1406,3 +1630,229 @@ void amdgpu_gfx_sysfs_fini(struct amdgpu_device *adev)
 	device_remove_file(adev->dev, &dev_attr_current_compute_partition);
 	device_remove_file(adev->dev, &dev_attr_available_compute_partition);
 }
+
+int amdgpu_gfx_sysfs_isolation_shader_init(struct amdgpu_device *adev)
+{
+	int r;
+
+	if (!amdgpu_sriov_vf(adev)) {
+		r = device_create_file(adev->dev, &dev_attr_enforce_isolation);
+		if (r)
+			return r;
+	}
+
+	r = device_create_file(adev->dev, &dev_attr_run_cleaner_shader);
+	if (r)
+		return r;
+
+	return 0;
+}
+
+void amdgpu_gfx_sysfs_isolation_shader_fini(struct amdgpu_device *adev)
+{
+	if (!amdgpu_sriov_vf(adev))
+		device_remove_file(adev->dev, &dev_attr_enforce_isolation);
+	device_remove_file(adev->dev, &dev_attr_run_cleaner_shader);
+}
+
+int amdgpu_gfx_cleaner_shader_sw_init(struct amdgpu_device *adev,
+				      unsigned int cleaner_shader_size)
+{
+	if (!adev->gfx.enable_cleaner_shader)
+		return -EOPNOTSUPP;
+
+	return amdgpu_bo_create_kernel(adev, cleaner_shader_size, PAGE_SIZE,
+				       AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT,
+				       &adev->gfx.cleaner_shader_obj,
+				       &adev->gfx.cleaner_shader_gpu_addr,
+				       (void **)&adev->gfx.cleaner_shader_cpu_ptr);
+}
+
+void amdgpu_gfx_cleaner_shader_sw_fini(struct amdgpu_device *adev)
+{
+	if (!adev->gfx.enable_cleaner_shader)
+		return;
+
+	amdgpu_bo_free_kernel(&adev->gfx.cleaner_shader_obj,
+			      &adev->gfx.cleaner_shader_gpu_addr,
+			      (void **)&adev->gfx.cleaner_shader_cpu_ptr);
+}
+
+void amdgpu_gfx_cleaner_shader_init(struct amdgpu_device *adev,
+				    unsigned int cleaner_shader_size,
+				    const void *cleaner_shader_ptr)
+{
+	if (!adev->gfx.enable_cleaner_shader)
+		return;
+
+	if (adev->gfx.cleaner_shader_cpu_ptr && cleaner_shader_ptr)
+		memcpy_toio(adev->gfx.cleaner_shader_cpu_ptr, cleaner_shader_ptr,
+			    cleaner_shader_size);
+}
+
+/**
+ * amdgpu_gfx_kfd_sch_ctrl - Control the KFD scheduler from the KGD (Graphics Driver)
+ * @adev: amdgpu_device pointer
+ * @idx: Index of the scheduler to control
+ * @enable: Whether to enable or disable the KFD scheduler
+ *
+ * This function is used to control the KFD (Kernel Fusion Driver) scheduler
+ * from the KGD. It is part of the cleaner shader feature. This function plays
+ * a key role in enforcing process isolation on the GPU.
+ *
+ * The function uses a reference count mechanism (kfd_sch_req_count) to keep
+ * track of the number of requests to enable the KFD scheduler. When a request
+ * to enable the KFD scheduler is made, the reference count is decremented.
+ * When the reference count reaches zero, a delayed work is scheduled to
+ * enforce isolation after a delay of GFX_SLICE_PERIOD.
+ *
+ * When a request to disable the KFD scheduler is made, the function first
+ * checks if the reference count is zero. If it is, it cancels the delayed work
+ * for enforcing isolation and checks if the KFD scheduler is active. If the
+ * KFD scheduler is active, it sends a request to stop the KFD scheduler and
+ * sets the KFD scheduler state to inactive. Then, it increments the reference
+ * count.
+ *
+ * The function is synchronized using the kfd_sch_mutex to ensure that the KFD
+ * scheduler state and reference count are updated atomically.
+ *
+ * Note: If the reference count is already zero when a request to enable the
+ * KFD scheduler is made, it means there's an imbalance bug somewhere. The
+ * function triggers a warning in this case.
+ */
+static void amdgpu_gfx_kfd_sch_ctrl(struct amdgpu_device *adev, u32 idx,
+				    bool enable)
+{
+	mutex_lock(&adev->gfx.kfd_sch_mutex);
+
+	if (enable) {
+		/* If the count is already 0, it means there's an imbalance bug somewhere.
+		 * Note that the bug may be in a different caller than the one which triggers the
+		 * WARN_ON_ONCE.
+		 */
+		if (WARN_ON_ONCE(adev->gfx.kfd_sch_req_count[idx] == 0)) {
+			dev_err(adev->dev, "Attempted to enable KFD scheduler when reference count is already zero\n");
+			goto unlock;
+		}
+
+		adev->gfx.kfd_sch_req_count[idx]--;
+
+		if (adev->gfx.kfd_sch_req_count[idx] == 0 &&
+		    adev->gfx.kfd_sch_inactive[idx]) {
+			schedule_delayed_work(&adev->gfx.enforce_isolation[idx].work,
+					      GFX_SLICE_PERIOD);
+		}
+	} else {
+		if (adev->gfx.kfd_sch_req_count[idx] == 0) {
+			cancel_delayed_work_sync(&adev->gfx.enforce_isolation[idx].work);
+			if (!adev->gfx.kfd_sch_inactive[idx]) {
+				amdgpu_amdkfd_stop_sched(adev, idx);
+				adev->gfx.kfd_sch_inactive[idx] = true;
+			}
+		}
+
+		adev->gfx.kfd_sch_req_count[idx]++;
+	}
+
+unlock:
+	mutex_unlock(&adev->gfx.kfd_sch_mutex);
+}
+
+/**
+ * amdgpu_gfx_enforce_isolation_handler - work handler for enforcing shader isolation
+ *
+ * @work: work_struct.
+ *
+ * This function is the work handler for enforcing shader isolation on AMD GPUs.
+ * It counts the number of emitted fences for each GFX and compute ring. If there
+ * are any fences, it schedules the `enforce_isolation_work` to be run after a
+ * delay of `GFX_SLICE_PERIOD`. If there are no fences, it signals the Kernel Fusion
+ * Driver (KFD) to resume the runqueue. The function is synchronized using the
+ * `enforce_isolation_mutex`.
+ */
+void amdgpu_gfx_enforce_isolation_handler(struct work_struct *work)
+{
+	struct amdgpu_isolation_work *isolation_work =
+		container_of(work, struct amdgpu_isolation_work, work.work);
+	struct amdgpu_device *adev = isolation_work->adev;
+	u32 i, idx, fences = 0;
+
+	if (isolation_work->xcp_id == AMDGPU_XCP_NO_PARTITION)
+		idx = 0;
+	else
+		idx = isolation_work->xcp_id;
+
+	if (idx >= MAX_XCP)
+		return;
+
+	mutex_lock(&adev->enforce_isolation_mutex);
+	for (i = 0; i < AMDGPU_MAX_GFX_RINGS; ++i) {
+		if (isolation_work->xcp_id == adev->gfx.gfx_ring[i].xcp_id)
+			fences += amdgpu_fence_count_emitted(&adev->gfx.gfx_ring[i]);
+	}
+	for (i = 0; i < (AMDGPU_MAX_COMPUTE_RINGS * AMDGPU_MAX_GC_INSTANCES); ++i) {
+		if (isolation_work->xcp_id == adev->gfx.compute_ring[i].xcp_id)
+			fences += amdgpu_fence_count_emitted(&adev->gfx.compute_ring[i]);
+	}
+	if (fences) {
+		schedule_delayed_work(&adev->gfx.enforce_isolation[idx].work,
+				      GFX_SLICE_PERIOD);
+	} else {
+		/* Tell KFD to resume the runqueue */
+		if (adev->kfd.init_complete) {
+			WARN_ON_ONCE(!adev->gfx.kfd_sch_inactive[idx]);
+			WARN_ON_ONCE(adev->gfx.kfd_sch_req_count[idx]);
+				amdgpu_amdkfd_start_sched(adev, idx);
+				adev->gfx.kfd_sch_inactive[idx] = false;
+		}
+	}
+	mutex_unlock(&adev->enforce_isolation_mutex);
+}
+
+void amdgpu_gfx_enforce_isolation_ring_begin_use(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+	u32 idx;
+
+	if (!adev->gfx.enable_cleaner_shader)
+		return;
+
+	if (ring->xcp_id == AMDGPU_XCP_NO_PARTITION)
+		idx = 0;
+	else
+		idx = ring->xcp_id;
+
+	if (idx >= MAX_XCP)
+		return;
+
+	mutex_lock(&adev->enforce_isolation_mutex);
+	if (adev->enforce_isolation[idx]) {
+		if (adev->kfd.init_complete)
+			amdgpu_gfx_kfd_sch_ctrl(adev, idx, false);
+	}
+	mutex_unlock(&adev->enforce_isolation_mutex);
+}
+
+void amdgpu_gfx_enforce_isolation_ring_end_use(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+	u32 idx;
+
+	if (!adev->gfx.enable_cleaner_shader)
+		return;
+
+	if (ring->xcp_id == AMDGPU_XCP_NO_PARTITION)
+		idx = 0;
+	else
+		idx = ring->xcp_id;
+
+	if (idx >= MAX_XCP)
+		return;
+
+	mutex_lock(&adev->enforce_isolation_mutex);
+	if (adev->enforce_isolation[idx]) {
+		if (adev->kfd.init_complete)
+			amdgpu_gfx_kfd_sch_ctrl(adev, idx, true);
+	}
+	mutex_unlock(&adev->enforce_isolation_mutex);
+}

drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h

@@ -34,6 +34,7 @@
 #include "soc15.h"
 #include "amdgpu_ras.h"
 #include "amdgpu_ring_mux.h"
+#include "amdgpu_xcp.h"
 
 /* GFX current status */
 #define AMDGPU_GFX_NORMAL_MODE			0x00000000L
@@ -138,6 +139,10 @@ struct kiq_pm4_funcs {
 	void (*kiq_invalidate_tlbs)(struct amdgpu_ring *kiq_ring,
 				uint16_t pasid, uint32_t flush_type,
 				bool all_hub);
+	void (*kiq_reset_hw_queue)(struct amdgpu_ring *kiq_ring,
+				   uint32_t queue_type, uint32_t me_id,
+				   uint32_t pipe_id, uint32_t queue_id,
+				   uint32_t xcc_id, uint32_t vmid);
 	/* Packet sizes */
 	int set_resources_size;
 	int map_queues_size;
@@ -345,6 +350,12 @@ struct amdgpu_me {
 	DECLARE_BITMAP(queue_bitmap, AMDGPU_MAX_GFX_QUEUES);
 };
 
+struct amdgpu_isolation_work {
+	struct amdgpu_device		*adev;
+	u32				xcp_id;
+	struct delayed_work		work;
+};
+
 struct amdgpu_gfx {
 	struct mutex			gpu_clock_mutex;
 	struct amdgpu_gfx_config	config;
@@ -397,6 +408,7 @@ struct amdgpu_gfx {
 	struct amdgpu_irq_src		eop_irq;
 	struct amdgpu_irq_src		priv_reg_irq;
 	struct amdgpu_irq_src		priv_inst_irq;
+	struct amdgpu_irq_src		bad_op_irq;
 	struct amdgpu_irq_src		cp_ecc_error_irq;
 	struct amdgpu_irq_src		sq_irq;
 	struct amdgpu_irq_src		rlc_gc_fed_irq;
@@ -445,6 +457,21 @@ struct amdgpu_gfx {
 	uint32_t			*ip_dump_core;
 	uint32_t			*ip_dump_compute_queues;
 	uint32_t			*ip_dump_gfx_queues;
+
+	struct mutex			reset_sem_mutex;
+
+	/* cleaner shader */
+	struct amdgpu_bo		*cleaner_shader_obj;
+	unsigned int                    cleaner_shader_size;
+	u64				cleaner_shader_gpu_addr;
+	void				*cleaner_shader_cpu_ptr;
+	const void			*cleaner_shader_ptr;
+	bool				enable_cleaner_shader;
+	struct amdgpu_isolation_work	enforce_isolation[MAX_XCP];
+	/* Mutex for synchronizing KFD scheduler operations */
+	struct mutex                    kfd_sch_mutex;
+	u64				kfd_sch_req_count[MAX_XCP];
+	bool				kfd_sch_inactive[MAX_XCP];
 };
 
 struct amdgpu_gfx_ras_reg_entry {
@@ -546,6 +573,17 @@ void amdgpu_gfx_ras_error_func(struct amdgpu_device *adev,
 		void *ras_error_status,
 		void (*func)(struct amdgpu_device *adev, void *ras_error_status,
 				int xcc_id));
+int amdgpu_gfx_cleaner_shader_sw_init(struct amdgpu_device *adev,
+				      unsigned int cleaner_shader_size);
+void amdgpu_gfx_cleaner_shader_sw_fini(struct amdgpu_device *adev);
+void amdgpu_gfx_cleaner_shader_init(struct amdgpu_device *adev,
+				    unsigned int cleaner_shader_size,
+				    const void *cleaner_shader_ptr);
+int amdgpu_gfx_sysfs_isolation_shader_init(struct amdgpu_device *adev);
+void amdgpu_gfx_sysfs_isolation_shader_fini(struct amdgpu_device *adev);
+void amdgpu_gfx_enforce_isolation_handler(struct work_struct *work);
+void amdgpu_gfx_enforce_isolation_ring_begin_use(struct amdgpu_ring *ring);
+void amdgpu_gfx_enforce_isolation_ring_end_use(struct amdgpu_ring *ring);
 
 static inline const char *amdgpu_gfx_compute_mode_desc(int mode)
 {

drivers/gpu/drm/amd/amdgpu/amdgpu_gfxhub.h

@@ -38,8 +38,6 @@ struct amdgpu_gfxhub_funcs {
 	void (*mode2_save_regs)(struct amdgpu_device *adev);
 	void (*mode2_restore_regs)(struct amdgpu_device *adev);
 	void (*halt)(struct amdgpu_device *adev);
-	bool (*query_utcl2_poison_status)(struct amdgpu_device *adev,
-			int xcc_id);
 };
 
 struct amdgpu_gfxhub {

drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c

@@ -786,7 +786,8 @@ void amdgpu_gmc_fw_reg_write_reg_wait(struct amdgpu_device *adev,
 		goto failed_kiq;
 
 	might_sleep();
-	while (r < 1 && cnt++ < MAX_KIQ_REG_TRY) {
+	while (r < 1 && cnt++ < MAX_KIQ_REG_TRY &&
+	       !amdgpu_reset_pending(adev->reset_domain)) {
 
 		msleep(MAX_KIQ_REG_BAILOUT_INTERVAL);
 		r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT);

drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c

@@ -424,7 +424,7 @@ int amdgpu_vmid_grab(struct amdgpu_vm *vm, struct amdgpu_ring *ring,
 	if (r || !idle)
 		goto error;
 
-	if (amdgpu_vmid_uses_reserved(vm, vmhub)) {
+	if (amdgpu_vmid_uses_reserved(adev, vm, vmhub)) {
 		r = amdgpu_vmid_grab_reserved(vm, ring, job, &id, fence);
 		if (r || !id)
 			goto error;
@@ -476,15 +476,19 @@ error:
 
 /*
  * amdgpu_vmid_uses_reserved - check if a VM will use a reserved VMID
+ * @adev: amdgpu_device pointer
  * @vm: the VM to check
  * @vmhub: the VMHUB which will be used
  *
  * Returns: True if the VM will use a reserved VMID.
  */
-bool amdgpu_vmid_uses_reserved(struct amdgpu_vm *vm, unsigned int vmhub)
+bool amdgpu_vmid_uses_reserved(struct amdgpu_device *adev,
+			       struct amdgpu_vm *vm, unsigned int vmhub)
 {
 	return vm->reserved_vmid[vmhub] ||
-		(enforce_isolation && (vmhub == AMDGPU_GFXHUB(0)));
+		(adev->enforce_isolation[(vm->root.bo->xcp_id != AMDGPU_XCP_NO_PARTITION) ?
+					 vm->root.bo->xcp_id : 0] &&
+		 AMDGPU_IS_GFXHUB(vmhub));
 }
 
 int amdgpu_vmid_alloc_reserved(struct amdgpu_device *adev,
@@ -600,9 +604,10 @@ void amdgpu_vmid_mgr_init(struct amdgpu_device *adev)
 		}
 	}
 	/* alloc a default reserved vmid to enforce isolation */
-	if (enforce_isolation)
+	for (i = 0; i < (adev->xcp_mgr ? adev->xcp_mgr->num_xcps : 1); i++) {
-		amdgpu_vmid_alloc_reserved(adev, AMDGPU_GFXHUB(0));
+		if (adev->enforce_isolation[i])
-
+			amdgpu_vmid_alloc_reserved(adev, AMDGPU_GFXHUB(i));
+	}
 }
 
 /**

drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h

@@ -78,7 +78,8 @@ void amdgpu_pasid_free_delayed(struct dma_resv *resv,
 
 bool amdgpu_vmid_had_gpu_reset(struct amdgpu_device *adev,
 			       struct amdgpu_vmid *id);
-bool amdgpu_vmid_uses_reserved(struct amdgpu_vm *vm, unsigned int vmhub);
+bool amdgpu_vmid_uses_reserved(struct amdgpu_device *adev,
+			       struct amdgpu_vm *vm, unsigned int vmhub);
 int amdgpu_vmid_alloc_reserved(struct amdgpu_device *adev,
 				unsigned vmhub);
 void amdgpu_vmid_free_reserved(struct amdgpu_device *adev,

drivers/gpu/drm/amd/amdgpu/amdgpu_isp.h

@@ -49,6 +49,7 @@ struct amdgpu_isp {
 	const struct isp_funcs	*funcs;
 	struct mfd_cell *isp_cell;
 	struct resource *isp_res;
+	struct resource *isp_i2c_res;
 	struct isp_platform_data *isp_pdata;
 	unsigned int harvest_config;
 	const struct firmware	*fw;

drivers/gpu/drm/amd/amdgpu/amdgpu_job.c

@@ -30,6 +30,60 @@
 #include "amdgpu.h"
 #include "amdgpu_trace.h"
 #include "amdgpu_reset.h"
+#include "amdgpu_dev_coredump.h"
+#include "amdgpu_xgmi.h"
+
+static void amdgpu_job_do_core_dump(struct amdgpu_device *adev,
+				    struct amdgpu_job *job)
+{
+	int i;
+
+	dev_info(adev->dev, "Dumping IP State\n");
+	for (i = 0; i < adev->num_ip_blocks; i++)
+		if (adev->ip_blocks[i].version->funcs->dump_ip_state)
+			adev->ip_blocks[i].version->funcs
+				->dump_ip_state((void *)adev);
+	dev_info(adev->dev, "Dumping IP State Completed\n");
+
+	amdgpu_coredump(adev, true, false, job);
+}
+
+static void amdgpu_job_core_dump(struct amdgpu_device *adev,
+				 struct amdgpu_job *job)
+{
+	struct list_head device_list, *device_list_handle =  NULL;
+	struct amdgpu_device *tmp_adev = NULL;
+	struct amdgpu_hive_info *hive = NULL;
+
+	if (!amdgpu_sriov_vf(adev))
+		hive = amdgpu_get_xgmi_hive(adev);
+	if (hive)
+		mutex_lock(&hive->hive_lock);
+	/*
+	 * Reuse the logic in amdgpu_device_gpu_recover() to build list of
+	 * devices for code dump
+	 */
+	INIT_LIST_HEAD(&device_list);
+	if (!amdgpu_sriov_vf(adev) && (adev->gmc.xgmi.num_physical_nodes > 1) && hive) {
+		list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head)
+			list_add_tail(&tmp_adev->reset_list, &device_list);
+		if (!list_is_first(&adev->reset_list, &device_list))
+			list_rotate_to_front(&adev->reset_list, &device_list);
+		device_list_handle = &device_list;
+	} else {
+		list_add_tail(&adev->reset_list, &device_list);
+		device_list_handle = &device_list;
+	}
+
+	/* Do the coredump for each device */
+	list_for_each_entry(tmp_adev, device_list_handle, reset_list)
+		amdgpu_job_do_core_dump(tmp_adev, job);
+
+	if (hive) {
+		mutex_unlock(&hive->hive_lock);
+		amdgpu_put_xgmi_hive(hive);
+	}
+}
 
 static enum drm_gpu_sched_stat amdgpu_job_timedout(struct drm_sched_job *s_job)
 {
@@ -48,9 +102,14 @@ static enum drm_gpu_sched_stat amdgpu_job_timedout(struct drm_sched_job *s_job)
 		return DRM_GPU_SCHED_STAT_ENODEV;
 	}
 
-
 	adev->job_hang = true;
 
+	/*
+	 * Do the coredump immediately after a job timeout to get a very
+	 * close dump/snapshot/representation of GPU's current error status
+	 */
+	amdgpu_job_core_dump(adev, job);
+
 	if (amdgpu_gpu_recovery &&
 	    amdgpu_ring_soft_recovery(ring, job->vmid, s_job->s_fence->parent)) {
 		dev_err(adev->dev, "ring %s timeout, but soft recovered\n",
@@ -72,6 +131,26 @@ static enum drm_gpu_sched_stat amdgpu_job_timedout(struct drm_sched_job *s_job)
 
 	dma_fence_set_error(&s_job->s_fence->finished, -ETIME);
 
+	/* attempt a per ring reset */
+	if (amdgpu_gpu_recovery &&
+	    ring->funcs->reset) {
+		/* stop the scheduler, but don't mess with the
+		 * bad job yet because if ring reset fails
+		 * we'll fall back to full GPU reset.
+		 */
+		drm_sched_wqueue_stop(&ring->sched);
+		r = amdgpu_ring_reset(ring, job->vmid);
+		if (!r) {
+			if (amdgpu_ring_sched_ready(ring))
+				drm_sched_stop(&ring->sched, s_job);
+			atomic_inc(&ring->adev->gpu_reset_counter);
+			amdgpu_fence_driver_force_completion(ring);
+			if (amdgpu_ring_sched_ready(ring))
+				drm_sched_start(&ring->sched);
+			goto exit;
+		}
+	}
+
 	if (amdgpu_device_should_recover_gpu(ring->adev)) {
 		struct amdgpu_reset_context reset_context;
 		memset(&reset_context, 0, sizeof(reset_context));
@@ -81,6 +160,12 @@ static enum drm_gpu_sched_stat amdgpu_job_timedout(struct drm_sched_job *s_job)
 		reset_context.src = AMDGPU_RESET_SRC_JOB;
 		clear_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags);
 
+		/*
+		 * To avoid an unnecessary extra coredump, as we have already
+		 * got the very close representation of GPU's error status
+		 */
+		set_bit(AMDGPU_SKIP_COREDUMP, &reset_context.flags);
+
 		r = amdgpu_device_gpu_recover(ring->adev, job, &reset_context);
 		if (r)
 			dev_err(adev->dev, "GPU Recovery Failed: %d\n", r);

drivers/gpu/drm/amd/amdgpu/amdgpu_job.h

@@ -76,6 +76,9 @@ struct amdgpu_job {
 	/* job_run_counter >= 1 means a resubmit job */
 	uint32_t		job_run_counter;
 
+	/* enforce isolation */
+	bool			enforce_isolation;
+
 	uint32_t		num_ibs;
 	struct amdgpu_ib	ibs[];
 };

drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c

@@ -43,6 +43,7 @@
 #include "amdgpu_gem.h"
 #include "amdgpu_display.h"
 #include "amdgpu_ras.h"
+#include "amdgpu_reset.h"
 #include "amd_pcie.h"
 
 void amdgpu_unregister_gpu_instance(struct amdgpu_device *adev)
@@ -778,6 +779,7 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 				    ? -EFAULT : 0;
 	}
 	case AMDGPU_INFO_READ_MMR_REG: {
+		int ret = 0;
 		unsigned int n, alloc_size;
 		uint32_t *regs;
 		unsigned int se_num = (info->read_mmr_reg.instance >>
@@ -787,24 +789,37 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 				   AMDGPU_INFO_MMR_SH_INDEX_SHIFT) &
 				  AMDGPU_INFO_MMR_SH_INDEX_MASK;
 
+		if (!down_read_trylock(&adev->reset_domain->sem))
+			return -ENOENT;
+
 		/* set full masks if the userspace set all bits
 		 * in the bitfields
 		 */
-		if (se_num == AMDGPU_INFO_MMR_SE_INDEX_MASK)
+		if (se_num == AMDGPU_INFO_MMR_SE_INDEX_MASK) {
 			se_num = 0xffffffff;
-		else if (se_num >= AMDGPU_GFX_MAX_SE)
+		} else if (se_num >= AMDGPU_GFX_MAX_SE) {
-			return -EINVAL;
+			ret = -EINVAL;
-		if (sh_num == AMDGPU_INFO_MMR_SH_INDEX_MASK)
+			goto out;
-			sh_num = 0xffffffff;
+		}
-		else if (sh_num >= AMDGPU_GFX_MAX_SH_PER_SE)
-			return -EINVAL;
 
-		if (info->read_mmr_reg.count > 128)
+		if (sh_num == AMDGPU_INFO_MMR_SH_INDEX_MASK) {
-			return -EINVAL;
+			sh_num = 0xffffffff;
+		} else if (sh_num >= AMDGPU_GFX_MAX_SH_PER_SE) {
+			ret = -EINVAL;
+			goto out;
+		}
+
+		if (info->read_mmr_reg.count > 128) {
+			ret = -EINVAL;
+			goto out;
+		}
 
 		regs = kmalloc_array(info->read_mmr_reg.count, sizeof(*regs), GFP_KERNEL);
-		if (!regs)
+		if (!regs) {
-			return -ENOMEM;
+			ret = -ENOMEM;
+			goto out;
+		}
+
 		alloc_size = info->read_mmr_reg.count * sizeof(*regs);
 
 		amdgpu_gfx_off_ctrl(adev, false);
@@ -816,13 +831,17 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 					      info->read_mmr_reg.dword_offset + i);
 				kfree(regs);
 				amdgpu_gfx_off_ctrl(adev, true);
-				return -EFAULT;
+				ret = -EFAULT;
+				goto out;
 			}
 		}
 		amdgpu_gfx_off_ctrl(adev, true);
 		n = copy_to_user(out, regs, min(size, alloc_size));
 		kfree(regs);
-		return n ? -EFAULT : 0;
+		ret = (n ? -EFAULT : 0);
+out:
+		up_read(&adev->reset_domain->sem);
+		return ret;
 	}
 	case AMDGPU_INFO_DEV_INFO: {
 		struct drm_amdgpu_info_device *dev_info;
@@ -1269,23 +1288,6 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 	return 0;
 }
 
-
-/*
- * Outdated mess for old drm with Xorg being in charge (void function now).
- */
-/**
- * amdgpu_driver_lastclose_kms - drm callback for last close
- *
- * @dev: drm dev pointer
- *
- * Switch vga_switcheroo state after last close (all asics).
- */
-void amdgpu_driver_lastclose_kms(struct drm_device *dev)
-{
-	drm_fb_helper_lastclose(dev);
-	vga_switcheroo_process_delayed_switch();
-}
-
 /**
  * amdgpu_driver_open_kms - drm callback for open
  *

drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c

@@ -396,7 +396,6 @@ static int amdgpu_mca_smu_parse_mca_error_count(struct amdgpu_device *adev, enum
 static int amdgpu_mca_dispatch_mca_set(struct amdgpu_device *adev, enum amdgpu_ras_block blk, enum amdgpu_mca_error_type type,
 				       struct mca_bank_set *mca_set, struct ras_err_data *err_data)
 {
-	struct ras_err_addr err_addr;
 	struct amdgpu_smuio_mcm_config_info mcm_info;
 	struct mca_bank_node *node, *tmp;
 	struct mca_bank_entry *entry;
@@ -421,27 +420,20 @@ static int amdgpu_mca_dispatch_mca_set(struct amdgpu_device *adev, enum amdgpu_r
 			continue;
 
 		memset(&mcm_info, 0, sizeof(mcm_info));
-		memset(&err_addr, 0, sizeof(err_addr));
 
 		mcm_info.socket_id = entry->info.socket_id;
 		mcm_info.die_id = entry->info.aid;
 
-		if (blk == AMDGPU_RAS_BLOCK__UMC) {
-			err_addr.err_status = entry->regs[MCA_REG_IDX_STATUS];
-			err_addr.err_ipid = entry->regs[MCA_REG_IDX_IPID];
-			err_addr.err_addr = entry->regs[MCA_REG_IDX_ADDR];
-		}
-
 		if (type == AMDGPU_MCA_ERROR_TYPE_UE) {
 			amdgpu_ras_error_statistic_ue_count(err_data,
-							    &mcm_info, &err_addr, (uint64_t)count);
+							    &mcm_info, (uint64_t)count);
 		} else {
 			if (amdgpu_mca_is_deferred_error(adev, entry->regs[MCA_REG_IDX_STATUS]))
 				amdgpu_ras_error_statistic_de_count(err_data,
-								    &mcm_info, &err_addr, (uint64_t)count);
+								    &mcm_info, (uint64_t)count);
 			else
 				amdgpu_ras_error_statistic_ce_count(err_data,
-								    &mcm_info, &err_addr, (uint64_t)count);
+								    &mcm_info, (uint64_t)count);
 		}
 
 		amdgpu_mca_bank_set_remove_node(mca_set, node);

drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c

@@ -501,60 +501,50 @@ int amdgpu_mes_remove_gang(struct amdgpu_device *adev, int gang_id)
 
 int amdgpu_mes_suspend(struct amdgpu_device *adev)
 {
-	struct idr *idp;
-	struct amdgpu_mes_process *process;
-	struct amdgpu_mes_gang *gang;
 	struct mes_suspend_gang_input input;
-	int r, pasid;
+	int r;
+
+	if (!amdgpu_mes_suspend_resume_all_supported(adev))
+		return 0;
+
+	memset(&input, 0x0, sizeof(struct mes_suspend_gang_input));
+	input.suspend_all_gangs = 1;
 
 	/*
 	 * Avoid taking any other locks under MES lock to avoid circular
 	 * lock dependencies.
 	 */
 	amdgpu_mes_lock(&adev->mes);
-
-	idp = &adev->mes.pasid_idr;
-
-	idr_for_each_entry(idp, process, pasid) {
-		list_for_each_entry(gang, &process->gang_list, list) {
 	r = adev->mes.funcs->suspend_gang(&adev->mes, &input);
-			if (r)
-				DRM_ERROR("failed to suspend pasid %d gangid %d",
-					 pasid, gang->gang_id);
-		}
-	}
-
 	amdgpu_mes_unlock(&adev->mes);
-	return 0;
+	if (r)
+		DRM_ERROR("failed to suspend all gangs");
+
+	return r;
 }
 
 int amdgpu_mes_resume(struct amdgpu_device *adev)
 {
-	struct idr *idp;
-	struct amdgpu_mes_process *process;
-	struct amdgpu_mes_gang *gang;
 	struct mes_resume_gang_input input;
-	int r, pasid;
+	int r;
+
+	if (!amdgpu_mes_suspend_resume_all_supported(adev))
+		return 0;
+
+	memset(&input, 0x0, sizeof(struct mes_resume_gang_input));
+	input.resume_all_gangs = 1;
 
 	/*
 	 * Avoid taking any other locks under MES lock to avoid circular
 	 * lock dependencies.
 	 */
 	amdgpu_mes_lock(&adev->mes);
-
-	idp = &adev->mes.pasid_idr;
-
-	idr_for_each_entry(idp, process, pasid) {
-		list_for_each_entry(gang, &process->gang_list, list) {
 	r = adev->mes.funcs->resume_gang(&adev->mes, &input);
-			if (r)
-				DRM_ERROR("failed to resume pasid %d gangid %d",
-					 pasid, gang->gang_id);
-		}
-	}
-
 	amdgpu_mes_unlock(&adev->mes);
-	return 0;
+	if (r)
+		DRM_ERROR("failed to resume all gangs");
+
+	return r;
 }
 
 static int amdgpu_mes_queue_alloc_mqd(struct amdgpu_device *adev,
@@ -793,6 +783,68 @@ int amdgpu_mes_remove_hw_queue(struct amdgpu_device *adev, int queue_id)
 	return 0;
 }
 
+int amdgpu_mes_reset_hw_queue(struct amdgpu_device *adev, int queue_id)
+{
+	unsigned long flags;
+	struct amdgpu_mes_queue *queue;
+	struct amdgpu_mes_gang *gang;
+	struct mes_reset_queue_input queue_input;
+	int r;
+
+	/*
+	 * Avoid taking any other locks under MES lock to avoid circular
+	 * lock dependencies.
+	 */
+	amdgpu_mes_lock(&adev->mes);
+
+	/* remove the mes gang from idr list */
+	spin_lock_irqsave(&adev->mes.queue_id_lock, flags);
+
+	queue = idr_find(&adev->mes.queue_id_idr, queue_id);
+	if (!queue) {
+		spin_unlock_irqrestore(&adev->mes.queue_id_lock, flags);
+		amdgpu_mes_unlock(&adev->mes);
+		DRM_ERROR("queue id %d doesn't exist\n", queue_id);
+		return -EINVAL;
+	}
+	spin_unlock_irqrestore(&adev->mes.queue_id_lock, flags);
+
+	DRM_DEBUG("try to reset queue, doorbell off = 0x%llx\n",
+		  queue->doorbell_off);
+
+	gang = queue->gang;
+	queue_input.doorbell_offset = queue->doorbell_off;
+	queue_input.gang_context_addr = gang->gang_ctx_gpu_addr;
+
+	r = adev->mes.funcs->reset_hw_queue(&adev->mes, &queue_input);
+	if (r)
+		DRM_ERROR("failed to reset hardware queue, queue id = %d\n",
+			  queue_id);
+
+	amdgpu_mes_unlock(&adev->mes);
+
+	return 0;
+}
+
+int amdgpu_mes_reset_hw_queue_mmio(struct amdgpu_device *adev, int queue_type,
+				   int me_id, int pipe_id, int queue_id, int vmid)
+{
+	struct mes_reset_queue_input queue_input;
+	int r;
+
+	queue_input.queue_type = queue_type;
+	queue_input.use_mmio = true;
+	queue_input.me_id = me_id;
+	queue_input.pipe_id = pipe_id;
+	queue_input.queue_id = queue_id;
+	queue_input.vmid = vmid;
+	r = adev->mes.funcs->reset_hw_queue(&adev->mes, &queue_input);
+	if (r)
+		DRM_ERROR("failed to reset hardware queue by mmio, queue id = %d\n",
+			  queue_id);
+	return r;
+}
+
 int amdgpu_mes_map_legacy_queue(struct amdgpu_device *adev,
 				struct amdgpu_ring *ring)
 {
@@ -838,6 +890,33 @@ int amdgpu_mes_unmap_legacy_queue(struct amdgpu_device *adev,
 	return r;
 }
 
+int amdgpu_mes_reset_legacy_queue(struct amdgpu_device *adev,
+				  struct amdgpu_ring *ring,
+				  unsigned int vmid,
+				  bool use_mmio)
+{
+	struct mes_reset_legacy_queue_input queue_input;
+	int r;
+
+	memset(&queue_input, 0, sizeof(queue_input));
+
+	queue_input.queue_type = ring->funcs->type;
+	queue_input.doorbell_offset = ring->doorbell_index;
+	queue_input.me_id = ring->me;
+	queue_input.pipe_id = ring->pipe;
+	queue_input.queue_id = ring->queue;
+	queue_input.mqd_addr = amdgpu_bo_gpu_offset(ring->mqd_obj);
+	queue_input.wptr_addr = ring->wptr_gpu_addr;
+	queue_input.vmid = vmid;
+	queue_input.use_mmio = use_mmio;
+
+	r = adev->mes.funcs->reset_legacy_queue(&adev->mes, &queue_input);
+	if (r)
+		DRM_ERROR("failed to reset legacy queue\n");
+
+	return r;
+}
+
 uint32_t amdgpu_mes_rreg(struct amdgpu_device *adev, uint32_t reg)
 {
 	struct mes_misc_op_input op_input;
@@ -1533,7 +1612,7 @@ int amdgpu_mes_init_microcode(struct amdgpu_device *adev, int pipe)
 			 pipe == AMDGPU_MES_SCHED_PIPE ? "" : "1");
 	}
 
-	r = amdgpu_ucode_request(adev, &adev->mes.fw[pipe], fw_name);
+	r = amdgpu_ucode_request(adev, &adev->mes.fw[pipe], "%s", fw_name);
 	if (r && need_retry && pipe == AMDGPU_MES_SCHED_PIPE) {
 		dev_info(adev->dev, "try to fall back to %s_mes.bin\n", ucode_prefix);
 		r = amdgpu_ucode_request(adev, &adev->mes.fw[pipe],
@@ -1584,6 +1663,19 @@ out:
 	return r;
 }
 
+bool amdgpu_mes_suspend_resume_all_supported(struct amdgpu_device *adev)
+{
+	uint32_t mes_rev = adev->mes.sched_version & AMDGPU_MES_VERSION_MASK;
+	bool is_supported = false;
+
+	if (amdgpu_ip_version(adev, GC_HWIP, 0) >= IP_VERSION(11, 0, 0) &&
+	    amdgpu_ip_version(adev, GC_HWIP, 0) < IP_VERSION(12, 0, 0) &&
+	    mes_rev >= 0x63)
+		is_supported = true;
+
+	return is_supported;
+}
+
 #if defined(CONFIG_DEBUG_FS)
 
 static int amdgpu_debugfs_mes_event_log_show(struct seq_file *m, void *unused)

drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h

@@ -249,6 +249,18 @@ struct mes_remove_queue_input {
 	uint64_t	gang_context_addr;
 };
 
+struct mes_reset_queue_input {
+	uint32_t	doorbell_offset;
+	uint64_t	gang_context_addr;
+	bool		use_mmio;
+	uint32_t	queue_type;
+	uint32_t	me_id;
+	uint32_t	pipe_id;
+	uint32_t	queue_id;
+	uint32_t	xcc_id;
+	uint32_t	vmid;
+};
+
 struct mes_map_legacy_queue_input {
 	uint32_t                           queue_type;
 	uint32_t                           doorbell_offset;
@@ -280,6 +292,18 @@ struct mes_resume_gang_input {
 	uint64_t	gang_context_addr;
 };
 
+struct mes_reset_legacy_queue_input {
+	uint32_t                           queue_type;
+	uint32_t                           doorbell_offset;
+	bool                               use_mmio;
+	uint32_t                           me_id;
+	uint32_t                           pipe_id;
+	uint32_t                           queue_id;
+	uint64_t                           mqd_addr;
+	uint64_t                           wptr_addr;
+	uint32_t                           vmid;
+};
+
 enum mes_misc_opcode {
 	MES_MISC_OP_WRITE_REG,
 	MES_MISC_OP_READ_REG,
@@ -348,6 +372,12 @@ struct amdgpu_mes_funcs {
 
 	int (*misc_op)(struct amdgpu_mes *mes,
 		       struct mes_misc_op_input *input);
+
+	int (*reset_legacy_queue)(struct amdgpu_mes *mes,
+				  struct mes_reset_legacy_queue_input *input);
+
+	int (*reset_hw_queue)(struct amdgpu_mes *mes,
+			      struct mes_reset_queue_input *input);
 };
 
 #define amdgpu_mes_kiq_hw_init(adev) (adev)->mes.kiq_hw_init((adev))
@@ -375,6 +405,9 @@ int amdgpu_mes_add_hw_queue(struct amdgpu_device *adev, int gang_id,
 			    struct amdgpu_mes_queue_properties *qprops,
 			    int *queue_id);
 int amdgpu_mes_remove_hw_queue(struct amdgpu_device *adev, int queue_id);
+int amdgpu_mes_reset_hw_queue(struct amdgpu_device *adev, int queue_id);
+int amdgpu_mes_reset_hw_queue_mmio(struct amdgpu_device *adev, int queue_type,
+				   int me_id, int pipe_id, int queue_id, int vmid);
 
 int amdgpu_mes_map_legacy_queue(struct amdgpu_device *adev,
 				struct amdgpu_ring *ring);
@@ -382,6 +415,10 @@ int amdgpu_mes_unmap_legacy_queue(struct amdgpu_device *adev,
 				  struct amdgpu_ring *ring,
 				  enum amdgpu_unmap_queues_action action,
 				  u64 gpu_addr, u64 seq);
+int amdgpu_mes_reset_legacy_queue(struct amdgpu_device *adev,
+				  struct amdgpu_ring *ring,
+				  unsigned int vmid,
+				  bool use_mmio);
 
 uint32_t amdgpu_mes_rreg(struct amdgpu_device *adev, uint32_t reg);
 int amdgpu_mes_wreg(struct amdgpu_device *adev,
@@ -479,4 +516,6 @@ static inline void amdgpu_mes_unlock(struct amdgpu_mes *mes)
 	memalloc_noreclaim_restore(mes->saved_flags);
 	mutex_unlock(&mes->mutex_hidden);
 }
+
+bool amdgpu_mes_suspend_resume_all_supported(struct amdgpu_device *adev);
 #endif /* __AMDGPU_MES_H__ */

drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h

@@ -63,8 +63,6 @@ struct amdgpu_mmhub_funcs {
 				uint64_t page_table_base);
 	void (*update_power_gating)(struct amdgpu_device *adev,
                                 bool enable);
-	bool (*query_utcl2_poison_status)(struct amdgpu_device *adev,
-				int hub_inst);
 };
 
 struct amdgpu_mmhub {

drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h

@@ -51,6 +51,7 @@ struct amdgpu_encoder;
 struct amdgpu_router;
 struct amdgpu_hpd;
 struct edid;
+struct drm_edid;
 
 #define to_amdgpu_crtc(x) container_of(x, struct amdgpu_crtc, base)
 #define to_amdgpu_connector(x) container_of(x, struct amdgpu_connector, base)
@@ -326,8 +327,7 @@ struct amdgpu_mode_info {
 	/* FMT dithering */
 	struct drm_property *dither_property;
 	/* hardcoded DFP edid from BIOS */
-	struct edid *bios_hardcoded_edid;
+	const struct drm_edid *bios_hardcoded_edid;
-	int bios_hardcoded_edid_size;
 
 	/* firmware flags */
 	u32 firmware_flags;

drivers/gpu/drm/amd/amdgpu/amdgpu_object.h

@@ -90,6 +90,12 @@ struct amdgpu_bo_va {
 	bool				cleared;
 
 	bool				is_xgmi;
+
+	/*
+	 * protected by vm reservation lock
+	 * if non-zero, cannot unmap from GPU because user queues may still access it
+	 */
+	unsigned int			queue_refcount;
 };
 
 struct amdgpu_bo {

drivers/gpu/drm/amd/amdgpu/amdgpu_pll.c

@@ -94,7 +94,7 @@ static void amdgpu_pll_get_fb_ref_div(struct amdgpu_device *adev, unsigned int n
 		ref_div_max = min(128 / post_div, ref_div_max);
 
 	/* get matching reference and feedback divider */
-	*ref_div = min(max(DIV_ROUND_CLOSEST(den, post_div), 1u), ref_div_max);
+	*ref_div = clamp(DIV_ROUND_CLOSEST(den, post_div), 1u, ref_div_max);
 	*fb_div = DIV_ROUND_CLOSEST(nom * *ref_div * post_div, den);
 
 	/* limit fb divider to its maximum */

drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c

@@ -1223,11 +1223,11 @@ static void amdgpu_rasmgr_error_data_statistic_update(struct ras_manager *obj, s
 		for_each_ras_error(err_node, err_data) {
 			err_info = &err_node->err_info;
 			amdgpu_ras_error_statistic_de_count(&obj->err_data,
-					&err_info->mcm_info, NULL, err_info->de_count);
+					&err_info->mcm_info, err_info->de_count);
 			amdgpu_ras_error_statistic_ce_count(&obj->err_data,
-					&err_info->mcm_info, NULL, err_info->ce_count);
+					&err_info->mcm_info, err_info->ce_count);
 			amdgpu_ras_error_statistic_ue_count(&obj->err_data,
-					&err_info->mcm_info, NULL, err_info->ue_count);
+					&err_info->mcm_info, err_info->ue_count);
 		}
 	} else {
 		/* for legacy asic path which doesn't has error source info */
@@ -2153,7 +2153,7 @@ static void amdgpu_ras_interrupt_poison_consumption_handler(struct ras_manager *
 	/* gpu reset is fallback for failed and default cases.
 	 * For RMA case, amdgpu_umc_poison_handler will handle gpu reset.
 	 */
-	if (poison_stat && !con->is_rma) {
+	if (poison_stat && !amdgpu_ras_is_rma(adev)) {
 		event_id = amdgpu_ras_acquire_event_id(adev, type);
 		RAS_EVENT_LOG(adev, event_id,
 			      "GPU reset for %s RAS poison consumption is issued!\n",
@@ -2881,9 +2881,6 @@ static void amdgpu_ras_ecc_log_init(struct ras_ecc_log_info *ecc_log)
 {
 	mutex_init(&ecc_log->lock);
 
-	/* Set any value as siphash key */
-	memset(&ecc_log->ecc_key, 0xad, sizeof(ecc_log->ecc_key));
-
 	INIT_RADIX_TREE(&ecc_log->de_page_tree, GFP_KERNEL);
 	ecc_log->de_queried_count = 0;
 	ecc_log->prev_de_queried_count = 0;
@@ -2948,7 +2945,7 @@ static void amdgpu_ras_do_page_retirement(struct work_struct *work)
 
 	amdgpu_ras_error_data_fini(&err_data);
 
-	if (err_cnt && con->is_rma)
+	if (err_cnt && amdgpu_ras_is_rma(adev))
 		amdgpu_ras_reset_gpu(adev);
 
 	amdgpu_ras_schedule_retirement_dwork(con,
@@ -3049,7 +3046,7 @@ static int amdgpu_ras_poison_consumption_handler(struct amdgpu_device *adev,
 	}
 
 	/* for RMA, amdgpu_ras_poison_creation_handler will trigger gpu reset */
-	if (reset_flags && !con->is_rma) {
+	if (reset_flags && !amdgpu_ras_is_rma(adev)) {
 		if (reset_flags & AMDGPU_RAS_GPU_RESET_MODE1_RESET)
 			reset = AMDGPU_RAS_GPU_RESET_MODE1_RESET;
 		else if (reset_flags & AMDGPU_RAS_GPU_RESET_MODE2_RESET)
@@ -3195,7 +3192,7 @@ int amdgpu_ras_recovery_init(struct amdgpu_device *adev)
 	 * This calling fails when is_rma is true or
 	 * ret != 0.
 	 */
-	if (con->is_rma || ret)
+	if (amdgpu_ras_is_rma(adev) || ret)
 		goto free;
 
 	if (con->eeprom_control.ras_num_recs) {
@@ -3244,7 +3241,7 @@ out:
 	 * Except error threshold exceeding case, other failure cases in this
 	 * function would not fail amdgpu driver init.
 	 */
-	if (!con->is_rma)
+	if (!amdgpu_ras_is_rma(adev))
 		ret = 0;
 	else
 		ret = -EINVAL;
@@ -4287,7 +4284,7 @@ int amdgpu_ras_reset_gpu(struct amdgpu_device *adev)
 	struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
 
 	/* mode1 is the only selection for RMA status */
-	if (ras->is_rma) {
+	if (amdgpu_ras_is_rma(adev)) {
 		ras->gpu_reset_flags = 0;
 		ras->gpu_reset_flags |= AMDGPU_RAS_GPU_RESET_MODE1_RESET;
 	}
@@ -4611,8 +4608,6 @@ static struct ras_err_info *amdgpu_ras_error_get_info(struct ras_err_data *err_d
 	if (!err_node)
 		return NULL;
 
-	INIT_LIST_HEAD(&err_node->err_info.err_addr_list);
-
 	memcpy(&err_node->err_info.mcm_info, mcm_info, sizeof(*mcm_info));
 
 	err_data->err_list_count++;
@@ -4622,21 +4617,9 @@ static struct ras_err_info *amdgpu_ras_error_get_info(struct ras_err_data *err_d
 	return &err_node->err_info;
 }
 
-void amdgpu_ras_add_mca_err_addr(struct ras_err_info *err_info, struct ras_err_addr *err_addr)
-{
-	/* This function will be retired. */
-	return;
-}
-
-void amdgpu_ras_del_mca_err_addr(struct ras_err_info *err_info, struct ras_err_addr *mca_err_addr)
-{
-	list_del(&mca_err_addr->node);
-	kfree(mca_err_addr);
-}
-
 int amdgpu_ras_error_statistic_ue_count(struct ras_err_data *err_data,
 					struct amdgpu_smuio_mcm_config_info *mcm_info,
-		struct ras_err_addr *err_addr, u64 count)
+					u64 count)
 {
 	struct ras_err_info *err_info;
 
@@ -4650,9 +4633,6 @@ int amdgpu_ras_error_statistic_ue_count(struct ras_err_data *err_data,
 	if (!err_info)
 		return -EINVAL;
 
-	if (err_addr && err_addr->err_status)
-		amdgpu_ras_add_mca_err_addr(err_info, err_addr);
-
 	err_info->ue_count += count;
 	err_data->ue_count += count;
 
@@ -4661,7 +4641,7 @@ int amdgpu_ras_error_statistic_ue_count(struct ras_err_data *err_data,
 
 int amdgpu_ras_error_statistic_ce_count(struct ras_err_data *err_data,
 					struct amdgpu_smuio_mcm_config_info *mcm_info,
-		struct ras_err_addr *err_addr, u64 count)
+					u64 count)
 {
 	struct ras_err_info *err_info;
 
@@ -4683,7 +4663,7 @@ int amdgpu_ras_error_statistic_ce_count(struct ras_err_data *err_data,
 
 int amdgpu_ras_error_statistic_de_count(struct ras_err_data *err_data,
 					struct amdgpu_smuio_mcm_config_info *mcm_info,
-		struct ras_err_addr *err_addr, u64 count)
+					u64 count)
 {
 	struct ras_err_info *err_info;
 
@@ -4697,9 +4677,6 @@ int amdgpu_ras_error_statistic_de_count(struct ras_err_data *err_data,
 	if (!err_info)
 		return -EINVAL;
 
-	if (err_addr && err_addr->err_status)
-		amdgpu_ras_add_mca_err_addr(err_info, err_addr);
-
 	err_info->de_count += count;
 	err_data->de_count += count;
 
@@ -4771,6 +4748,16 @@ static void amdgpu_ras_boot_time_error_reporting(struct amdgpu_device *adev,
 		dev_info(adev->dev,
 			 "socket: %d, aid: %d, hbm: %d, fw_status: 0x%x, hbm bist test failed\n",
 			 socket_id, aid_id, hbm_id, fw_status);
+
+	if (AMDGPU_RAS_GPU_ERR_DATA_ABORT(boot_error))
+		dev_info(adev->dev,
+			 "socket: %d, aid: %d, fw_status: 0x%x, data abort exception\n",
+			 socket_id, aid_id, fw_status);
+
+	if (AMDGPU_RAS_GPU_ERR_UNKNOWN(boot_error))
+		dev_info(adev->dev,
+			 "socket: %d, aid: %d, fw_status: 0x%x, unknown boot time errors\n",
+			 socket_id, aid_id, fw_status);
 }
 
 static bool amdgpu_ras_boot_error_detected(struct amdgpu_device *adev,
@@ -4837,3 +4824,13 @@ void amdgpu_ras_event_log_print(struct amdgpu_device *adev, u64 event_id,
 
 	va_end(args);
 }
+
+bool amdgpu_ras_is_rma(struct amdgpu_device *adev)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+
+	if (!con)
+		return false;
+
+	return con->is_rma;
+}

drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h

@@ -28,7 +28,6 @@
 #include <linux/list.h>
 #include <linux/kfifo.h>
 #include <linux/radix-tree.h>
-#include <linux/siphash.h>
 #include "ta_ras_if.h"
 #include "amdgpu_ras_eeprom.h"
 #include "amdgpu_smuio.h"
@@ -47,6 +46,8 @@ struct amdgpu_iv_entry;
 #define AMDGPU_RAS_GPU_ERR_SOCKET_ID(x)			AMDGPU_GET_REG_FIELD(x, 10, 8)
 #define AMDGPU_RAS_GPU_ERR_AID_ID(x)			AMDGPU_GET_REG_FIELD(x, 12, 11)
 #define AMDGPU_RAS_GPU_ERR_HBM_ID(x)			AMDGPU_GET_REG_FIELD(x, 14, 13)
+#define AMDGPU_RAS_GPU_ERR_DATA_ABORT(x)		AMDGPU_GET_REG_FIELD(x, 29, 29)
+#define AMDGPU_RAS_GPU_ERR_UNKNOWN(x)			AMDGPU_GET_REG_FIELD(x, 30, 30)
 
 #define AMDGPU_RAS_BOOT_STATUS_POLLING_LIMIT	100
 #define AMDGPU_RAS_BOOT_STEADY_STATUS		0xBA
@@ -476,16 +477,15 @@ struct ras_err_pages {
 };
 
 struct ras_ecc_err {
-	u64 hash_index;
 	uint64_t status;
 	uint64_t ipid;
 	uint64_t addr;
+	uint64_t pa_pfn;
 	struct ras_err_pages err_pages;
 };
 
 struct ras_ecc_log_info {
 	struct mutex lock;
-	siphash_key_t ecc_key;
 	struct radix_tree_root de_page_tree;
 	uint64_t	de_queried_count;
 	uint64_t	prev_de_queried_count;
@@ -572,19 +572,11 @@ struct ras_fs_data {
 	char debugfs_name[32];
 };
 
-struct ras_err_addr {
-	struct list_head node;
-	uint64_t err_status;
-	uint64_t err_ipid;
-	uint64_t err_addr;
-};
-
 struct ras_err_info {
 	struct amdgpu_smuio_mcm_config_info mcm_info;
 	u64 ce_count;
 	u64 ue_count;
 	u64 de_count;
-	struct list_head err_addr_list;
 };
 
 struct ras_err_node {
@@ -942,13 +934,13 @@ int amdgpu_ras_error_data_init(struct ras_err_data *err_data);
 void amdgpu_ras_error_data_fini(struct ras_err_data *err_data);
 int amdgpu_ras_error_statistic_ce_count(struct ras_err_data *err_data,
 					struct amdgpu_smuio_mcm_config_info *mcm_info,
-		struct ras_err_addr *err_addr, u64 count);
+					u64 count);
 int amdgpu_ras_error_statistic_ue_count(struct ras_err_data *err_data,
 					struct amdgpu_smuio_mcm_config_info *mcm_info,
-		struct ras_err_addr *err_addr, u64 count);
+					u64 count);
 int amdgpu_ras_error_statistic_de_count(struct ras_err_data *err_data,
 					struct amdgpu_smuio_mcm_config_info *mcm_info,
-		struct ras_err_addr *err_addr, u64 count);
+					u64 count);
 void amdgpu_ras_query_boot_status(struct amdgpu_device *adev, u32 num_instances);
 int amdgpu_ras_bind_aca(struct amdgpu_device *adev, enum amdgpu_ras_block blk,
 			       const struct aca_info *aca_info, void *data);
@@ -957,12 +949,6 @@ int amdgpu_ras_unbind_aca(struct amdgpu_device *adev, enum amdgpu_ras_block blk)
 ssize_t amdgpu_ras_aca_sysfs_read(struct device *dev, struct device_attribute *attr,
 				  struct aca_handle *handle, char *buf, void *data);
 
-void amdgpu_ras_add_mca_err_addr(struct ras_err_info *err_info,
-			struct ras_err_addr *err_addr);
-
-void amdgpu_ras_del_mca_err_addr(struct ras_err_info *err_info,
-		struct ras_err_addr *mca_err_addr);
-
 void amdgpu_ras_set_fed(struct amdgpu_device *adev, bool status);
 bool amdgpu_ras_get_fed_status(struct amdgpu_device *adev);
 
@@ -982,4 +968,5 @@ __printf(3, 4)
 void amdgpu_ras_event_log_print(struct amdgpu_device *adev, u64 event_id,
 				const char *fmt, ...);
 
+bool amdgpu_ras_is_rma(struct amdgpu_device *adev);
 #endif

drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h

@@ -136,6 +136,12 @@ static inline bool amdgpu_reset_domain_schedule(struct amdgpu_reset_domain *doma
 	return queue_work(domain->wq, work);
 }
 
+static inline bool amdgpu_reset_pending(struct amdgpu_reset_domain *domain)
+{
+	lockdep_assert_held(&domain->sem);
+	return rwsem_is_contended(&domain->sem);
+}
+
 void amdgpu_device_lock_reset_domain(struct amdgpu_reset_domain *reset_domain);
 
 void amdgpu_device_unlock_reset_domain(struct amdgpu_reset_domain *reset_domain);

drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c

@@ -144,7 +144,9 @@ void amdgpu_ring_commit(struct amdgpu_ring *ring)
 	/* We pad to match fetch size */
 	count = ring->funcs->align_mask + 1 -
 		(ring->wptr & ring->funcs->align_mask);
-	count %= ring->funcs->align_mask + 1;
+	count &= ring->funcs->align_mask;
+
+	if (count != 0)
 		ring->funcs->insert_nop(ring, count);
 
 	mb();

drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h

@@ -235,6 +235,8 @@ struct amdgpu_ring_funcs {
 	void (*patch_cntl)(struct amdgpu_ring *ring, unsigned offset);
 	void (*patch_ce)(struct amdgpu_ring *ring, unsigned offset);
 	void (*patch_de)(struct amdgpu_ring *ring, unsigned offset);
+	int (*reset)(struct amdgpu_ring *ring, unsigned int vmid);
+	void (*emit_cleaner_shader)(struct amdgpu_ring *ring);
 };
 
 struct amdgpu_ring {
@@ -334,6 +336,7 @@ struct amdgpu_ring {
 #define amdgpu_ring_patch_cntl(r, o) ((r)->funcs->patch_cntl((r), (o)))
 #define amdgpu_ring_patch_ce(r, o) ((r)->funcs->patch_ce((r), (o)))
 #define amdgpu_ring_patch_de(r, o) ((r)->funcs->patch_de((r), (o)))
+#define amdgpu_ring_reset(r, v) (r)->funcs->reset((r), (v))
 
 unsigned int amdgpu_ring_max_ibs(enum amdgpu_ring_type type);
 int amdgpu_ring_alloc(struct amdgpu_ring *ring, unsigned ndw);

drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.c

@@ -410,7 +410,7 @@ void amdgpu_sw_ring_ib_end(struct amdgpu_ring *ring)
 	struct amdgpu_ring_mux *mux = &adev->gfx.muxer;
 
 	WARN_ON(!ring->is_sw_ring);
-	if (ring->hw_prio > AMDGPU_RING_PRIO_DEFAULT)
+	if (adev->gfx.mcbp && ring->hw_prio > AMDGPU_RING_PRIO_DEFAULT)
 		return;
 	amdgpu_ring_mux_end_ib(mux, ring);
 }

drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c

@@ -22,7 +22,6 @@
  * Authors: Andres Rodriguez <andresx7@gmail.com>
  */
 
-#include <linux/fdtable.h>
 #include <linux/file.h>
 #include <linux/pid.h>
 

drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h

@@ -115,6 +115,7 @@ struct amdgpu_sdma {
 	bool			has_page_queue;
 	struct ras_common_if	*ras_if;
 	struct amdgpu_sdma_ras	*ras;
+	uint32_t		*ip_dump;
 };
 
 /*

drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c

@@ -196,7 +196,7 @@ static int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev,
 	amdgpu_umc_handle_bad_pages(adev, ras_error_status);
 
 	if ((err_data->ue_count || err_data->de_count) &&
-	    (reset || (con && con->is_rma))) {
+	    (reset || amdgpu_ras_is_rma(adev))) {
 		con->gpu_reset_flags |= reset;
 		amdgpu_ras_reset_gpu(adev);
 	}
@@ -204,55 +204,6 @@ static int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev,
 	return AMDGPU_RAS_SUCCESS;
 }
 
-int amdgpu_umc_bad_page_polling_timeout(struct amdgpu_device *adev,
-			uint32_t reset, uint32_t timeout_ms)
-{
-	struct ras_err_data err_data;
-	struct ras_common_if head = {
-		.block = AMDGPU_RAS_BLOCK__UMC,
-	};
-	struct ras_manager *obj = amdgpu_ras_find_obj(adev, &head);
-	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
-	uint32_t timeout = timeout_ms;
-
-	memset(&err_data, 0, sizeof(err_data));
-	amdgpu_ras_error_data_init(&err_data);
-
-	do {
-
-		amdgpu_umc_handle_bad_pages(adev, &err_data);
-
-		if (timeout && !err_data.de_count) {
-			msleep(1);
-			timeout--;
-		}
-
-	} while (timeout && !err_data.de_count);
-
-	if (!timeout)
-		dev_warn(adev->dev, "Can't find bad pages\n");
-
-	if (err_data.de_count)
-		dev_info(adev->dev, "%ld new deferred hardware errors detected\n", err_data.de_count);
-
-	if (obj) {
-		obj->err_data.ue_count += err_data.ue_count;
-		obj->err_data.ce_count += err_data.ce_count;
-		obj->err_data.de_count += err_data.de_count;
-	}
-
-	amdgpu_ras_error_data_fini(&err_data);
-
-	kgd2kfd_set_sram_ecc_flag(adev->kfd.dev);
-
-	if (reset || (err_data.err_addr_cnt && con && con->is_rma)) {
-		con->gpu_reset_flags |= reset;
-		amdgpu_ras_reset_gpu(adev);
-	}
-
-	return 0;
-}
-
 int amdgpu_umc_pasid_poison_handler(struct amdgpu_device *adev,
 			enum amdgpu_ras_block block, uint16_t pasid,
 			pasid_notify pasid_fn, void *data, uint32_t reset)
@@ -472,43 +423,6 @@ int amdgpu_umc_update_ecc_status(struct amdgpu_device *adev,
 	return 0;
 }
 
-static int amdgpu_umc_uint64_cmp(const void *a, const void *b)
-{
-	uint64_t *addr_a = (uint64_t *)a;
-	uint64_t *addr_b = (uint64_t *)b;
-
-	if (*addr_a > *addr_b)
-		return 1;
-	else if (*addr_a < *addr_b)
-		return -1;
-	else
-		return 0;
-}
-
-/* Use string hash to avoid logging the same bad pages repeatedly */
-int amdgpu_umc_build_pages_hash(struct amdgpu_device *adev,
-		uint64_t *pfns, int len, uint64_t *val)
-{
-	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
-	char buf[MAX_UMC_HASH_STRING_SIZE] = {0};
-	int offset = 0, i = 0;
-	uint64_t hash_val;
-
-	if (!pfns || !len)
-		return -EINVAL;
-
-	sort(pfns, len, sizeof(uint64_t), amdgpu_umc_uint64_cmp, NULL);
-
-	for (i = 0; i < len; i++)
-		offset += snprintf(&buf[offset], sizeof(buf) - offset, "%llx", pfns[i]);
-
-	hash_val = siphash(buf, offset, &con->umc_ecc_log.ecc_key);
-
-	*val = hash_val;
-
-	return 0;
-}
-
 int amdgpu_umc_logs_ecc_err(struct amdgpu_device *adev,
 		struct radix_tree_root *ecc_tree, struct ras_ecc_err *ecc_err)
 {
@@ -519,18 +433,10 @@ int amdgpu_umc_logs_ecc_err(struct amdgpu_device *adev,
 	ecc_log = &con->umc_ecc_log;
 
 	mutex_lock(&ecc_log->lock);
-	ret = radix_tree_insert(ecc_tree, ecc_err->hash_index, ecc_err);
+	ret = radix_tree_insert(ecc_tree, ecc_err->pa_pfn, ecc_err);
-	if (!ret) {
+	if (!ret)
-		struct ras_err_pages *err_pages = &ecc_err->err_pages;
-		int i;
-
-		/* Reserve memory */
-		for (i = 0; i < err_pages->count; i++)
-			amdgpu_ras_reserve_page(adev, err_pages->pfn[i]);
-
 		radix_tree_tag_set(ecc_tree,
-			ecc_err->hash_index, UMC_ECC_NEW_DETECTED_TAG);
+			ecc_err->pa_pfn, UMC_ECC_NEW_DETECTED_TAG);
-	}
 	mutex_unlock(&ecc_log->lock);
 
 	return ret;

drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h

@@ -127,13 +127,8 @@ int amdgpu_umc_page_retirement_mca(struct amdgpu_device *adev,
 int amdgpu_umc_loop_channels(struct amdgpu_device *adev,
 			umc_func func, void *data);
 
-int amdgpu_umc_bad_page_polling_timeout(struct amdgpu_device *adev,
-			uint32_t reset, uint32_t timeout_ms);
-
 int amdgpu_umc_update_ecc_status(struct amdgpu_device *adev,
 				uint64_t status, uint64_t ipid, uint64_t addr);
-int amdgpu_umc_build_pages_hash(struct amdgpu_device *adev,
-		uint64_t *pfns, int len, uint64_t *val);
 int amdgpu_umc_logs_ecc_err(struct amdgpu_device *adev,
 		struct radix_tree_root *ecc_tree, struct ras_ecc_err *ecc_err);
 

drivers/gpu/drm/amd/amdgpu/amdgpu_umsch_mm.c

@@ -587,7 +587,7 @@ int amdgpu_umsch_mm_init_microcode(struct amdgpu_umsch_mm *umsch)
 		break;
 	}
 
-	r = amdgpu_ucode_request(adev, &adev->umsch_mm.fw, fw_name);
+	r = amdgpu_ucode_request(adev, &adev->umsch_mm.fw, "%s", fw_name);
 	if (r) {
 		release_firmware(adev->umsch_mm.fw);
 		adev->umsch_mm.fw = NULL;

drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c

@@ -260,7 +260,7 @@ int amdgpu_uvd_sw_init(struct amdgpu_device *adev)
 		return -EINVAL;
 	}
 
-	r = amdgpu_ucode_request(adev, &adev->uvd.fw, fw_name);
+	r = amdgpu_ucode_request(adev, &adev->uvd.fw, "%s", fw_name);
 	if (r) {
 		dev_err(adev->dev, "amdgpu_uvd: Can't validate firmware \"%s\"\n",
 			fw_name);
@@ -1088,7 +1088,6 @@ int amdgpu_uvd_ring_parse_cs(struct amdgpu_cs_parser *parser,
 	int r;
 
 	job->vm = NULL;
-	ib->gpu_addr = amdgpu_sa_bo_gpu_addr(ib->sa_bo);
 
 	if (ib->length_dw % 16) {
 		DRM_ERROR("UVD IB length (%d) not 16 dwords aligned!\n",

drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c

@@ -158,7 +158,7 @@ int amdgpu_vce_sw_init(struct amdgpu_device *adev, unsigned long size)
 		return -EINVAL;
 	}
 
-	r = amdgpu_ucode_request(adev, &adev->vce.fw, fw_name);
+	r = amdgpu_ucode_request(adev, &adev->vce.fw, "%s", fw_name);
 	if (r) {
 		dev_err(adev->dev, "amdgpu_vce: Can't validate firmware \"%s\"\n",
 			fw_name);
@@ -749,7 +749,6 @@ int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p,
 	int i, r = 0;
 
 	job->vm = NULL;
-	ib->gpu_addr = amdgpu_sa_bo_gpu_addr(ib->sa_bo);
 
 	for (idx = 0; idx < ib->length_dw;) {
 		uint32_t len = amdgpu_ib_get_value(ib, idx);
@@ -1044,7 +1043,6 @@ out:
 	if (!r) {
 		/* No error, free all destroyed handle slots */
 		tmp = destroyed;
-		amdgpu_ib_free(p->adev, ib, NULL);
 	} else {
 		/* Error during parsing, free all allocated handle slots */
 		tmp = allocated;

drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h

@@ -330,6 +330,9 @@ struct amdgpu_vcn {
 	uint16_t inst_mask;
 	uint8_t	num_inst_per_aid;
 	bool using_unified_queue;
+
+	/* IP reg dump */
+	uint32_t		*ip_dump;
 };
 
 struct amdgpu_fw_shared_rb_ptrs_struct {

drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c

@@ -33,6 +33,7 @@
 #include "amdgpu.h"
 #include "amdgpu_ras.h"
 #include "amdgpu_reset.h"
+#include "amdgpu_dpm.h"
 #include "vi.h"
 #include "soc15.h"
 #include "nv.h"
@@ -849,6 +850,13 @@ enum amdgpu_sriov_vf_mode amdgpu_virt_get_sriov_vf_mode(struct amdgpu_device *ad
 	return mode;
 }
 
+void amdgpu_virt_pre_reset(struct amdgpu_device *adev)
+{
+	/* stop the data exchange thread */
+	amdgpu_virt_fini_data_exchange(adev);
+	amdgpu_dpm_set_mp1_state(adev, PP_MP1_STATE_FLR);
+}
+
 void amdgpu_virt_post_reset(struct amdgpu_device *adev)
 {
 	if (amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(11, 0, 3)) {

drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h

@@ -376,6 +376,7 @@ u32 amdgpu_sriov_rreg(struct amdgpu_device *adev,
 		      u32 offset, u32 acc_flags, u32 hwip, u32 xcc_id);
 bool amdgpu_virt_fw_load_skip_check(struct amdgpu_device *adev,
 			uint32_t ucode_id);
+void amdgpu_virt_pre_reset(struct amdgpu_device *adev);
 void amdgpu_virt_post_reset(struct amdgpu_device *adev);
 bool amdgpu_sriov_xnack_support(struct amdgpu_device *adev);
 bool amdgpu_virt_get_rlcg_reg_access_flag(struct amdgpu_device *adev,

drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c

@@ -549,7 +549,7 @@ static int amdgpu_vkms_sw_fini(void *handle)
 
 	adev->mode_info.mode_config_initialized = false;
 
-	kfree(adev->mode_info.bios_hardcoded_edid);
+	drm_edid_free(adev->mode_info.bios_hardcoded_edid);
 	kfree(adev->amdgpu_vkms_output);
 	return 0;
 }

drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c

@@ -681,6 +681,11 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring, struct amdgpu_job *job,
 	pasid_mapping_needed &= adev->gmc.gmc_funcs->emit_pasid_mapping &&
 		ring->funcs->emit_wreg;
 
+	if (adev->gfx.enable_cleaner_shader &&
+	    ring->funcs->emit_cleaner_shader &&
+	    job->enforce_isolation)
+		ring->funcs->emit_cleaner_shader(ring);
+
 	if (!vm_flush_needed && !gds_switch_needed && !need_pipe_sync)
 		return 0;
 
@@ -742,6 +747,7 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring, struct amdgpu_job *job,
 		amdgpu_ring_emit_switch_buffer(ring);
 		amdgpu_ring_emit_switch_buffer(ring);
 	}
+
 	amdgpu_ring_ib_end(ring);
 	return 0;
 }
@@ -838,7 +844,7 @@ int amdgpu_vm_update_pdes(struct amdgpu_device *adev,
 	params.vm = vm;
 	params.immediate = immediate;
 
-	r = vm->update_funcs->prepare(&params, NULL, AMDGPU_SYNC_EXPLICIT);
+	r = vm->update_funcs->prepare(&params, NULL);
 	if (r)
 		goto error;
 
@@ -902,10 +908,12 @@ amdgpu_vm_tlb_flush(struct amdgpu_vm_update_params *params,
 {
 	struct amdgpu_vm *vm = params->vm;
 
-	if (!fence || !*fence)
-		return;
-
 	tlb_cb->vm = vm;
+	if (!fence || !*fence) {
+		amdgpu_vm_tlb_seq_cb(NULL, &tlb_cb->cb);
+		return;
+	}
+
 	if (!dma_fence_add_callback(*fence, &tlb_cb->cb,
 				    amdgpu_vm_tlb_seq_cb)) {
 		dma_fence_put(vm->last_tlb_flush);
@@ -933,7 +941,7 @@ amdgpu_vm_tlb_flush(struct amdgpu_vm_update_params *params,
  * @unlocked: unlocked invalidation during MM callback
  * @flush_tlb: trigger tlb invalidation after update completed
  * @allow_override: change MTYPE for local NUMA nodes
- * @resv: fences we need to sync to
+ * @sync: fences we need to sync to
  * @start: start of mapped range
  * @last: last mapped entry
  * @flags: flags for the entries
@@ -949,16 +957,16 @@ amdgpu_vm_tlb_flush(struct amdgpu_vm_update_params *params,
  * 0 for success, negative erro code for failure.
  */
 int amdgpu_vm_update_range(struct amdgpu_device *adev, struct amdgpu_vm *vm,
-			   bool immediate, bool unlocked, bool flush_tlb, bool allow_override,
+			   bool immediate, bool unlocked, bool flush_tlb,
-			   struct dma_resv *resv, uint64_t start, uint64_t last,
+			   bool allow_override, struct amdgpu_sync *sync,
-			   uint64_t flags, uint64_t offset, uint64_t vram_base,
+			   uint64_t start, uint64_t last, uint64_t flags,
+			   uint64_t offset, uint64_t vram_base,
 			   struct ttm_resource *res, dma_addr_t *pages_addr,
 			   struct dma_fence **fence)
 {
 	struct amdgpu_vm_tlb_seq_struct *tlb_cb;
 	struct amdgpu_vm_update_params params;
 	struct amdgpu_res_cursor cursor;
-	enum amdgpu_sync_mode sync_mode;
 	int r, idx;
 
 	if (!drm_dev_enter(adev_to_drm(adev), &idx))
@@ -991,14 +999,6 @@ int amdgpu_vm_update_range(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	params.allow_override = allow_override;
 	INIT_LIST_HEAD(&params.tlb_flush_waitlist);
 
-	/* Implicitly sync to command submissions in the same VM before
-	 * unmapping. Sync to moving fences before mapping.
-	 */
-	if (!(flags & AMDGPU_PTE_VALID))
-		sync_mode = AMDGPU_SYNC_EQ_OWNER;
-	else
-		sync_mode = AMDGPU_SYNC_EXPLICIT;
-
 	amdgpu_vm_eviction_lock(vm);
 	if (vm->evicting) {
 		r = -EBUSY;
@@ -1013,7 +1013,7 @@ int amdgpu_vm_update_range(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 		dma_fence_put(tmp);
 	}
 
-	r = vm->update_funcs->prepare(&params, resv, sync_mode);
+	r = vm->update_funcs->prepare(&params, sync);
 	if (r)
 		goto error_free;
 
@@ -1155,23 +1155,30 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
 	struct amdgpu_bo *bo = bo_va->base.bo;
 	struct amdgpu_vm *vm = bo_va->base.vm;
 	struct amdgpu_bo_va_mapping *mapping;
+	struct dma_fence **last_update;
 	dma_addr_t *pages_addr = NULL;
 	struct ttm_resource *mem;
-	struct dma_fence **last_update;
+	struct amdgpu_sync sync;
 	bool flush_tlb = clear;
-	bool uncached;
-	struct dma_resv *resv;
 	uint64_t vram_base;
 	uint64_t flags;
+	bool uncached;
 	int r;
 
+	amdgpu_sync_create(&sync);
 	if (clear || !bo) {
 		mem = NULL;
-		resv = vm->root.bo->tbo.base.resv;
+
+		/* Implicitly sync to command submissions in the same VM before
+		 * unmapping.
+		 */
+		r = amdgpu_sync_resv(adev, &sync, vm->root.bo->tbo.base.resv,
+				     AMDGPU_SYNC_EQ_OWNER, vm);
+		if (r)
+			goto error_free;
 	} else {
 		struct drm_gem_object *obj = &bo->tbo.base;
 
-		resv = bo->tbo.base.resv;
 		if (obj->import_attach && bo_va->is_xgmi) {
 			struct dma_buf *dma_buf = obj->import_attach->dmabuf;
 			struct drm_gem_object *gobj = dma_buf->priv;
@@ -1185,6 +1192,12 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
 		if (mem && (mem->mem_type == TTM_PL_TT ||
 			    mem->mem_type == AMDGPU_PL_PREEMPT))
 			pages_addr = bo->tbo.ttm->dma_address;
+
+		/* Implicitly sync to moving fences before mapping anything */
+		r = amdgpu_sync_resv(adev, &sync, bo->tbo.base.resv,
+				     AMDGPU_SYNC_EXPLICIT, vm);
+		if (r)
+			goto error_free;
 	}
 
 	if (bo) {
@@ -1234,12 +1247,12 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
 		trace_amdgpu_vm_bo_update(mapping);
 
 		r = amdgpu_vm_update_range(adev, vm, false, false, flush_tlb,
-					   !uncached, resv, mapping->start, mapping->last,
+					   !uncached, &sync, mapping->start,
-					   update_flags, mapping->offset,
+					   mapping->last, update_flags,
-					   vram_base, mem, pages_addr,
+					   mapping->offset, vram_base, mem,
-					   last_update);
+					   pages_addr, last_update);
 		if (r)
-			return r;
+			goto error_free;
 	}
 
 	/* If the BO is not in its preferred location add it back to
@@ -1267,7 +1280,9 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
 			trace_amdgpu_vm_bo_mapping(mapping);
 	}
 
-	return 0;
+error_free:
+	amdgpu_sync_free(&sync);
+	return r;
 }
 
 /**
@@ -1414,25 +1429,34 @@ int amdgpu_vm_clear_freed(struct amdgpu_device *adev,
 			  struct amdgpu_vm *vm,
 			  struct dma_fence **fence)
 {
-	struct dma_resv *resv = vm->root.bo->tbo.base.resv;
 	struct amdgpu_bo_va_mapping *mapping;
-	uint64_t init_pte_value = 0;
 	struct dma_fence *f = NULL;
+	struct amdgpu_sync sync;
 	int r;
 
+
+	/*
+	 * Implicitly sync to command submissions in the same VM before
+	 * unmapping.
+	 */
+	amdgpu_sync_create(&sync);
+	r = amdgpu_sync_resv(adev, &sync, vm->root.bo->tbo.base.resv,
+			     AMDGPU_SYNC_EQ_OWNER, vm);
+	if (r)
+		goto error_free;
+
 	while (!list_empty(&vm->freed)) {
 		mapping = list_first_entry(&vm->freed,
 			struct amdgpu_bo_va_mapping, list);
 		list_del(&mapping->list);
 
 		r = amdgpu_vm_update_range(adev, vm, false, false, true, false,
-					   resv, mapping->start, mapping->last,
+					   &sync, mapping->start, mapping->last,
-					   init_pte_value, 0, 0, NULL, NULL,
+					   0, 0, 0, NULL, NULL, &f);
-					   &f);
 		amdgpu_vm_free_mapping(adev, vm, mapping, f);
 		if (r) {
 			dma_fence_put(f);
-			return r;
+			goto error_free;
 		}
 	}
 
@@ -1443,7 +1467,9 @@ int amdgpu_vm_clear_freed(struct amdgpu_device *adev,
 		dma_fence_put(f);
 	}
 
-	return 0;
+error_free:
+	amdgpu_sync_free(&sync);
+	return r;
 
 }
 
@@ -2218,7 +2244,7 @@ void amdgpu_vm_adjust_size(struct amdgpu_device *adev, uint32_t min_vm_size,
 		phys_ram_gb = ((uint64_t)si.totalram * si.mem_unit +
 			       (1 << 30) - 1) >> 30;
 		vm_size = roundup_pow_of_two(
-			min(max(phys_ram_gb * 3, min_vm_size), max_size));
+			clamp(phys_ram_gb * 3, min_vm_size, max_size));
 	}
 
 	adev->vm_manager.max_pfn = (uint64_t)vm_size << 18;
@@ -2421,6 +2447,8 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	if (r)
 		return r;
 
+	ttm_lru_bulk_move_init(&vm->lru_bulk_move);
+
 	vm->is_compute_context = false;
 
 	vm->use_cpu_for_update = !!(adev->vm_manager.vm_update_mode &
@@ -2485,6 +2513,7 @@ error_free_root:
 error_free_delayed:
 	dma_fence_put(vm->last_tlb_flush);
 	dma_fence_put(vm->last_unlocked);
+	ttm_lru_bulk_move_fini(&adev->mman.bdev, &vm->lru_bulk_move);
 	amdgpu_vm_fini_entities(vm);
 
 	return r;
@@ -2641,6 +2670,7 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 		}
 	}
 
+	ttm_lru_bulk_move_fini(&adev->mman.bdev, &vm->lru_bulk_move);
 }
 
 /**
@@ -2754,6 +2784,7 @@ int amdgpu_vm_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
  * amdgpu_vm_handle_fault - graceful handling of VM faults.
  * @adev: amdgpu device pointer
  * @pasid: PASID of the VM
+ * @ts: Timestamp of the fault
  * @vmid: VMID, only used for GFX 9.4.3.
  * @node_id: Node_id received in IH cookie. Only applicable for
  *           GFX 9.4.3.
@@ -2764,7 +2795,7 @@ int amdgpu_vm_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
  * shouldn't be reported any more.
  */
 bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid,
-			    u32 vmid, u32 node_id, uint64_t addr,
+			    u32 vmid, u32 node_id, uint64_t addr, uint64_t ts,
 			    bool write_fault)
 {
 	bool is_compute_context = false;
@@ -2790,7 +2821,7 @@ bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid,
 	addr /= AMDGPU_GPU_PAGE_SIZE;
 
 	if (is_compute_context && !svm_range_restore_pages(adev, pasid, vmid,
-	    node_id, addr, write_fault)) {
+	    node_id, addr, ts, write_fault)) {
 		amdgpu_bo_unref(&root);
 		return true;
 	}

drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h

@@ -304,8 +304,8 @@ struct amdgpu_vm_update_params {
 
 struct amdgpu_vm_update_funcs {
 	int (*map_table)(struct amdgpu_bo_vm *bo);
-	int (*prepare)(struct amdgpu_vm_update_params *p, struct dma_resv *resv,
+	int (*prepare)(struct amdgpu_vm_update_params *p,
-		       enum amdgpu_sync_mode sync_mode);
+		       struct amdgpu_sync *sync);
 	int (*update)(struct amdgpu_vm_update_params *p,
 		      struct amdgpu_bo_vm *bo, uint64_t pe, uint64_t addr,
 		      unsigned count, uint32_t incr, uint64_t flags);
@@ -505,9 +505,10 @@ int amdgpu_vm_flush_compute_tlb(struct amdgpu_device *adev,
 void amdgpu_vm_bo_base_init(struct amdgpu_vm_bo_base *base,
 			    struct amdgpu_vm *vm, struct amdgpu_bo *bo);
 int amdgpu_vm_update_range(struct amdgpu_device *adev, struct amdgpu_vm *vm,
-			   bool immediate, bool unlocked, bool flush_tlb, bool allow_override,
+			   bool immediate, bool unlocked, bool flush_tlb,
-			   struct dma_resv *resv, uint64_t start, uint64_t last,
+			   bool allow_override, struct amdgpu_sync *sync,
-			   uint64_t flags, uint64_t offset, uint64_t vram_base,
+			   uint64_t start, uint64_t last, uint64_t flags,
+			   uint64_t offset, uint64_t vram_base,
 			   struct ttm_resource *res, dma_addr_t *pages_addr,
 			   struct dma_fence **fence);
 int amdgpu_vm_bo_update(struct amdgpu_device *adev,
@@ -558,7 +559,7 @@ amdgpu_vm_get_task_info_vm(struct amdgpu_vm *vm);
 void amdgpu_vm_put_task_info(struct amdgpu_task_info *task_info);
 
 bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid,
-			    u32 vmid, u32 node_id, uint64_t addr,
+			    u32 vmid, u32 node_id, uint64_t addr, uint64_t ts,
 			    bool write_fault);
 
 void amdgpu_vm_set_task_info(struct amdgpu_vm *vm);

drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c

@@ -39,20 +39,18 @@ static int amdgpu_vm_cpu_map_table(struct amdgpu_bo_vm *table)
  * amdgpu_vm_cpu_prepare - prepare page table update with the CPU
  *
  * @p: see amdgpu_vm_update_params definition
- * @resv: reservation object with embedded fence
+ * @sync: sync obj with fences to wait on
- * @sync_mode: synchronization mode
  *
  * Returns:
  * Negativ errno, 0 for success.
  */
 static int amdgpu_vm_cpu_prepare(struct amdgpu_vm_update_params *p,
-				 struct dma_resv *resv,
+				 struct amdgpu_sync *sync)
-				 enum amdgpu_sync_mode sync_mode)
 {
-	if (!resv)
+	if (!sync)
 		return 0;
 
-	return amdgpu_bo_sync_wait_resv(p->adev, resv, sync_mode, p->vm, true);
+	return amdgpu_sync_wait(sync, true);
 }
 
 /**

drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c

@@ -403,7 +403,7 @@ int amdgpu_vm_pt_clear(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	params.vm = vm;
 	params.immediate = immediate;
 
-	r = vm->update_funcs->prepare(&params, NULL, AMDGPU_SYNC_EXPLICIT);
+	r = vm->update_funcs->prepare(&params, NULL);
 	if (r)
 		goto exit;
 

drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c

@@ -77,32 +77,24 @@ static int amdgpu_vm_sdma_alloc_job(struct amdgpu_vm_update_params *p,
  * amdgpu_vm_sdma_prepare - prepare SDMA command submission
  *
  * @p: see amdgpu_vm_update_params definition
- * @resv: reservation object with embedded fence
+ * @sync: amdgpu_sync object with fences to wait for
- * @sync_mode: synchronization mode
  *
  * Returns:
  * Negativ errno, 0 for success.
  */
 static int amdgpu_vm_sdma_prepare(struct amdgpu_vm_update_params *p,
-				  struct dma_resv *resv,
+				  struct amdgpu_sync *sync)
-				  enum amdgpu_sync_mode sync_mode)
 {
-	struct amdgpu_sync sync;
 	int r;
 
 	r = amdgpu_vm_sdma_alloc_job(p, 0);
 	if (r)
 		return r;
 
-	if (!resv)
+	if (!sync)
 		return 0;
 
-	amdgpu_sync_create(&sync);
+	r = amdgpu_sync_push_to_job(sync, p->job);
-	r = amdgpu_sync_resv(p->adev, &sync, resv, sync_mode, p->vm);
-	if (!r)
-		r = amdgpu_sync_push_to_job(&sync, p->job);
-	amdgpu_sync_free(&sync);
-
 	if (r) {
 		p->num_dw_left = 0;
 		amdgpu_job_free(p->job);

drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c

@@ -1389,10 +1389,10 @@ static void __xgmi_v6_4_0_query_error_count(struct amdgpu_device *adev, struct a
 
 	switch (xgmi_v6_4_0_pcs_mca_get_error_type(adev, status)) {
 	case ACA_ERROR_TYPE_UE:
-		amdgpu_ras_error_statistic_ue_count(err_data, mcm_info, NULL, 1ULL);
+		amdgpu_ras_error_statistic_ue_count(err_data, mcm_info, 1ULL);
 		break;
 	case ACA_ERROR_TYPE_CE:
-		amdgpu_ras_error_statistic_ce_count(err_data, mcm_info, NULL, 1ULL);
+		amdgpu_ras_error_statistic_ce_count(err_data, mcm_info, 1ULL);
 		break;
 	default:
 		break;

drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h

@@ -213,7 +213,7 @@ struct amd_sriov_msg_pf2vf_info {
 	uint32_t gpu_capacity;
 	/* reserved */
 	uint32_t reserved[256 - AMD_SRIOV_MSG_PF2VF_INFO_FILLED_SIZE];
-};
+} __packed;
 
 struct amd_sriov_msg_vf2pf_info_header {
 	/* the total structure size in byte */
@@ -273,7 +273,7 @@ struct amd_sriov_msg_vf2pf_info {
 	uint32_t mes_info_size;
 	/* reserved */
 	uint32_t reserved[256 - AMD_SRIOV_MSG_VF2PF_INFO_FILLED_SIZE];
-};
+} __packed;
 
 /* mailbox message send from guest to host  */
 enum amd_sriov_mailbox_request_message {

drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c

@@ -75,6 +75,8 @@ static void aqua_vanjaram_set_xcp_id(struct amdgpu_device *adev,
 	uint32_t inst_mask;
 
 	ring->xcp_id = AMDGPU_XCP_NO_PARTITION;
+	if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE)
+		adev->gfx.enforce_isolation[0].xcp_id = ring->xcp_id;
 	if (adev->xcp_mgr->mode == AMDGPU_XCP_MODE_NONE)
 		return;
 
@@ -103,6 +105,8 @@ static void aqua_vanjaram_set_xcp_id(struct amdgpu_device *adev,
 	for (xcp_id = 0; xcp_id < adev->xcp_mgr->num_xcps; xcp_id++) {
 		if (adev->xcp_mgr->xcp[xcp_id].ip[ip_blk].inst_mask & inst_mask) {
 			ring->xcp_id = xcp_id;
+			if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE)
+				adev->gfx.enforce_isolation[xcp_id].xcp_id = xcp_id;
 			break;
 		}
 	}

drivers/gpu/drm/amd/amdgpu/atombios_encoders.c

@@ -215,7 +215,7 @@ void amdgpu_atombios_encoder_init_backlight(struct amdgpu_encoder *amdgpu_encode
 	dig->bl_dev = bd;
 
 	bd->props.brightness = amdgpu_atombios_encoder_get_backlight_brightness(bd);
-	bd->props.power = FB_BLANK_UNBLANK;
+	bd->props.power = BACKLIGHT_POWER_ON;
 	backlight_update_status(bd);
 
 	DRM_INFO("amdgpu atom DIG backlight initialized\n");
@@ -2064,27 +2064,25 @@ amdgpu_atombios_encoder_get_lcd_info(struct amdgpu_encoder *encoder)
 				case LCD_FAKE_EDID_PATCH_RECORD_TYPE:
 					fake_edid_record = (ATOM_FAKE_EDID_PATCH_RECORD *)record;
 					if (fake_edid_record->ucFakeEDIDLength) {
-						struct edid *edid;
+						const struct drm_edid *edid;
-						int edid_size =
+						int edid_size;
-							max((int)EDID_LENGTH, (int)fake_edid_record->ucFakeEDIDLength);
-						edid = kmalloc(edid_size, GFP_KERNEL);
-						if (edid) {
-							memcpy((u8 *)edid, (u8 *)&fake_edid_record->ucFakeEDIDString[0],
-							       fake_edid_record->ucFakeEDIDLength);
 
-							if (drm_edid_is_valid(edid)) {
+						if (fake_edid_record->ucFakeEDIDLength == 128)
+							edid_size = fake_edid_record->ucFakeEDIDLength;
+						else
+							edid_size = fake_edid_record->ucFakeEDIDLength * 128;
+						edid = drm_edid_alloc(fake_edid_record->ucFakeEDIDString, edid_size);
+						if (drm_edid_valid(edid))
 							adev->mode_info.bios_hardcoded_edid = edid;
-								adev->mode_info.bios_hardcoded_edid_size = edid_size;
+						else
-							} else
+							drm_edid_free(edid);
-								kfree(edid);
+						record += struct_size(fake_edid_record,
-						}
-					}
-					record += fake_edid_record->ucFakeEDIDLength ?
-						  struct_size(fake_edid_record,
 								      ucFakeEDIDString,
-							      fake_edid_record->ucFakeEDIDLength) :
+								      edid_size);
+					} else {
 						/* empty fake edid record must be 3 bytes long */
-						  sizeof(ATOM_FAKE_EDID_PATCH_RECORD) + 1;
+						record += sizeof(ATOM_FAKE_EDID_PATCH_RECORD) + 1;
+					}
 					break;
 				case LCD_PANEL_RESOLUTION_RECORD_TYPE:
 					panel_res_record = (ATOM_PANEL_RESOLUTION_PATCH_RECORD *)record;

drivers/gpu/drm/amd/amdgpu/cikd.h

@@ -364,6 +364,7 @@
 		 * 1 - Stream
 		 * 2 - Bypass
 		 */
+#define		EOP_EXEC				(1 << 28) /* For Trailing Fence */
 #define		DATA_SEL(x)                             ((x) << 29)
 		/* 0 - discard
 		 * 1 - send low 32bit data

drivers/gpu/drm/amd/amdgpu/dce_v10_0.c

@@ -2846,7 +2846,7 @@ static int dce_v10_0_sw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	kfree(adev->mode_info.bios_hardcoded_edid);
+	drm_edid_free(adev->mode_info.bios_hardcoded_edid);
 
 	drm_kms_helper_poll_fini(adev_to_drm(adev));