summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu
AgeCommit message (Collapse)Author
2025-02-27drm/amdgpu/vcn3.0: convert internal functions to use vcn_instAlex Deucher
Pass the vcn instance structure to these functions rather than adev and the instance number. TODO: clean up the function internals to use the vinst state directly rather than accessing it indirectly via adev->vcn.inst[]. Reviewed-by: Boyuan Zhang <Boyuan.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-27drm/amdgpu/vcn: switch vcn helpers to be instance basedAlex Deucher
Pass the instance to the helpers. Reviewed-by: Boyuan Zhang <Boyuan.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-27drm/amdgpu/vcn: move more instanced data to vcn_instanceAlex Deucher
Move more per instance data into the per instance structure. v2: index instances directly on vcn1.0 and 2.0 to make it clear that they only support a single instance (Lijo) v3: fix typo on vcn 2.5 Reviewed-by: Boyuan Zhang <Boyuan.Zhang@amd.com> (v2) Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-27drm/amdgpu/vcn: make powergating status per instanceAlex Deucher
Store it per instance so we can track it per instance. v2: index instances directly on vcn1.0 and 2.0 to make it clear that they only support a single instance (Lijo) Reviewed-by: Boyuan Zhang <Boyuan.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-27drm/amdgpu/vcn: switch work handler to be per instanceAlex Deucher
Have a separate work handler for each VCN instance. This paves the way for per instance VCN power gating at runtime. v2: index instances directly on vcn1.0 and 2.0 to make it clear that they only support a single instance (Lijo) Reviewed-by: Boyuan Zhang <Boyuan.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-27drm/amdgpu/vcn5.0.1: split code along instancesAlex Deucher
Split the code on a per instance basis. This will allow us to use the per instance functions in the future to handle more things per instance. Reviewed-by: Boyuan Zhang <Boyuan.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-27drm/amdgpu/vcn5.0.0: split code along instancesAlex Deucher
Split the code on a per instance basis. This will allow us to use the per instance functions in the future to handle more things per instance. v2: squash in fix for stop() from Boyuan Reviewed-by: Boyuan Zhang <Boyuan.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-27drm/amdgpu/vcn4.0.5: split code along instancesAlex Deucher
Split the code on a per instance basis. This will allow us to use the per instance functions in the future to handle more things per instance. v2: squash in fix for stop() from Boyuan Reviewed-by: Boyuan Zhang <Boyuan.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-27drm/amdgpu/vcn4.0.3: split code along instancesAlex Deucher
Split the code on a per instance basis. This will allow us to use the per instance functions in the future to handle more things per instance. v2: squash in fix for stop() from Boyuan Reviewed-by: Boyuan Zhang <Boyuan.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-27drm/amdgpu/vcn4.0: split code along instancesAlex Deucher
Split the code on a per instance basis. This will allow us to use the per instance functions in the future to handle more things per instance. v2: squash in fix for stop() from Boyuan Reviewed-by: Boyuan Zhang <Boyuan.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-27drm/amdgpu/vcn3.0: split code along instancesAlex Deucher
Split the code on a per instance basis. This will allow us to use the per instance functions in the future to handle more things per instance. v2: squash in fix for stop() from Boyuan Reviewed-by: Boyuan Zhang <Boyuan.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-27drm/amdgpu/vcn2.5: fix VCN stop logicAlex Deucher
Need to make sure we call amdgpu_dpm_enable_vcn() in vcn_v2_5_stop() at the end if there are errors or DPG is enabled. Fixes: ebc25499de12 ("drm/amdgpu/vcn2.5: split code along instances") Reviewed-by: Boyuan Zhang <Boyuan.Zhang@amd.com> Suggested-by: Boyuan Zhang <boyuan.zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-26Merge tag 'amd-drm-next-6.15-2025-02-21' of ↵Dave Airlie
https://gitlab.freedesktop.org/agd5f/linux into drm-next amd-drm-next-6.15-2025-02-21: amdgpu: - Add OEM i2c support for RGB lights, etc. - Add support for GC 11.5.3 - Add support for GC 11.5.2 - Add support for SDMA 6.1.3 - Add support for NBIO 7.11.2 - Add support for NBIO 7.9.1 - Add support for MMHUB 3.3.2 - Add support for MMHUB 1.8.1 - Add support for SMU 14.0.5 - Add support for SMUIO 13.0.11 - Add support for PSP 14.0.5 - Add support for UMC 12.5.0 - Add support for DCN 3.6.0 - JPEG 4.0.3 updates - Add dynamic workload profile switching for GC 10-12 - support larger vbios sizes - GC 9.5.0 updates - SMU 13.0.12 updates - SMU 13.0.6 updates - IP discovery updates - GC 10 queue reset updates - DCN 4.0.1 updates - UHBR link rate fixes - Aborted suspend fix - Mark gttsize parameter as deprecated - GC 10 cleaner shader updates - PSR-SU fixes - Clean up PM4 headers - Cursor fixes - Enable devcoredump for JPEG - Misc cleanups - Runpm cleanups - MES updates - GC 9 gfxoff fixes - Vbios fetching cleanups - Documentation updates - Update secondary plane handling - DML2 updates - SDMA fixes for MI - Cleaner shader fixes for GC 11/12 - ACA updates - Initial JPEG queue reset support - RAS updates - Initial RAS CPER support - DCN/DCE panic screen handling cleanup - BT2020 fixes - SR-IOV fixes amdkfd: - synchronize pasid values between KGD and KFD - Misc cleanups - Improve GTT/VRAM handling for APUs - Topology updates - Fix user queue validation on GC 7/8 UAPI: - Enable "Broadcast RGB" drm property - Add INFO IOCTL query for virtualization mode Proposed userspace: https://github.com/ROCm/amdsmi/commit/e663bed7d6b3df79f5959e73981749b1f22ec698 From: Alex Deucher <alexander.deucher@amd.com> Link: https://patchwork.freedesktop.org/patch/msgid/20250221213651.4176031-1-alexander.deucher@amd.com Signed-off-by: Dave Airlie <airlied@redhat.com>
2025-02-25drm/amdgpu: init return value in amdgpu_ttm_clear_bufferPierre-Eric Pelloux-Prayer
Otherwise an uninitialized value can be returned if amdgpu_res_cleared returns true for all regions. Possibly closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3812 Fixes: a68c7eaa7a8f ("drm/amdgpu: Enable clear page functionality") Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 7c62aacc3b452f73a1284198c81551035fac6d71) Cc: stable@vger.kernel.org
2025-02-25drm/amdgpu/mes: keep enforce isolation up to dateAlex Deucher
Re-send the mes message on resume to make sure the mes state is up to date. Fixes: 8521e3c5f058 ("drm/amd/amdgpu: limit single process inside MES") Acked-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: Shaoyun Liu <shaoyun.liu@amd.com> Cc: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 27b791514789844e80da990c456c2465325e0851)
2025-02-25drm/amdgpu/gfx: only call mes for enforce isolation if supportedAlex Deucher
This should not be called on chips without MES so check if MES is enabled and if the cleaner shader is supported. Fixes: 8521e3c5f058 ("drm/amd/amdgpu: limit single process inside MES") Reviewed-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: Shaoyun Liu <shaoyun.liu@amd.com> Cc: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> (cherry picked from commit 80513e389765c8f9543b26d8fa4bbdf0e59ff8bc)
2025-02-25drm/amdgpu: disable BAR resize on Dell G5 SEAlex Deucher
There was a quirk added to add a workaround for a Sapphire RX 5600 XT Pulse that didn't allow BAR resizing. However, the quirk caused a regression with runtime pm on Dell laptops using those chips, rather than narrowing the scope of the resizing quirk, add a quirk to prevent amdgpu from resizing the BAR on those Dell platforms unless runtime pm is disabled. v2: update commit message, add runpm check Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/1707 Fixes: 907830b0fc9e ("PCI: Add a REBAR size quirk for Sapphire RX 5600 XT Pulse") Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 5235053f443cef4210606e5fb71f99b915a9723d) Cc: stable@vger.kernel.org
2025-02-25drm/amdgpu: increase AMDGPU_MAX_RINGSTao Zhou
Increase it since a cper ring is introduced. Signed-off-by: Tao Zhou <tao.zhou1@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: Fix correct parameter desc for VCN idle check functionsSrinivasan Shanmugam
Fixes the kdoc for the following VCN idle check functions by updating the parameter description from 'handle' to 'ip_block': - vcn_v4_0_is_idle - vcn_v4_0_3_is_idle - vcn_v4_0_5_is_idle - vcn_v5_0_1_is_idle Fixes the below with gcc W=1: drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c:935: warning: Function parameter or struct member 'ip_block' not described in 'vcn_v5_0_1_is_idle' drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c:935: warning: Excess function parameter 'handle' description in 'vcn_v5_0_1_is_idle' drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c:1972: warning: Function parameter or struct member 'ip_block' not described in 'vcn_v4_0_is_idle' drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c:1972: warning: Excess function parameter 'handle' description in 'vcn_v4_0_is_idle' drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c:1583: warning: Function parameter or struct member 'ip_block' not described in 'vcn_v4_0_3_is_idle' drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c:1583: warning: Excess function parameter 'handle' description in 'vcn_v4_0_3_is_idle' drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c:1200: warning: Function parameter or struct member 'ip_block' not described in 'vcn_v5_0_0_is_idle' drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c:1200: warning: Excess function parameter 'handle' description in 'vcn_v5_0_0_is_idle' drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c:1460: warning: Function parameter or struct member 'ip_block' not described in 'vcn_v4_0_5_is_idle' drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c:1460: warning: Excess function parameter 'handle' description in 'vcn_v4_0_5_is_idle' Cc: Christian König <christian.koenig@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: init return value in amdgpu_ttm_clear_bufferPierre-Eric Pelloux-Prayer
Otherwise an uninitialized value can be returned if amdgpu_res_cleared returns true for all regions. Possibly closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3812 Fixes: a68c7eaa7a8f ("drm/amdgpu: Enable clear page functionality") Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: Change page/record number calculation based on npsganglxie
save only one record to save eeprom space,and bad_page_num = pa_rec_num + mca_rec_num*16 Signed-off-by: ganglxie <ganglxie@amd.com> Reviewed-by: Tao Zhou <tao.zhou1@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: Refine bad page addingganglxie
bad page adding can be simpler with nps info Signed-off-by: ganglxie <ganglxie@amd.com> Reviewed-by: Tao Zhou <tao.zhou1@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: update SDMA sysfs reset mask in late_initJesse.zhang@amd.com
- Added `sdma_v4_4_2_update_reset_mask` function to update the reset mask. - update the sysfs reset mask to the `late_init` stage to ensure that the SMU initialization and capability setup are completed before checking the SDMA reset capability. - For IP versions 9.4.3 and 9.4.4, enable per-queue reset if the MEC firmware version is at least 0xb0 and PMFW supports queue reset. - Add a TODO comment for future support of per-queue reset for IP version 9.5.0. This change ensures that per-queue reset is only enabled when the MEC and PMFW support it. v2: fix ip version (9.5.4 -> 9.5.0)(Lijo) Suggested-by: Jonathan Kim <Jonathan.Kim@amd.com> Signed-off-by: Vitaly Prosyak <vitaly.prosyak@amd.com> Signed-off-by: Jesse Zhang <jesse.zhang@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: Set CPER enabled flag after ring initiailizedXiang Liu
Setting cper.enabled to be true only after cper ring is successfully created. Signed-off-by: Xiang Liu <xiang.liu@amd.com> Reviewed-by: Tao Zhou <tao.zhou1@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: Save nps to eepromganglxie
nps info saved together with bad page makes bad page parsing more efficient Signed-off-by: ganglxie <ganglxie@amd.com> Reviewed-by: Tao Zhou <tao.zhou1@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: Check if CPER enabled when generating CPERXiang Liu
In the case of CPER disabled, generating CPER will cause kernel NULL pointer dereference without checking. Signed-off-by: Xiang Liu <xiang.liu@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: simplify xgmi peer info callsJonathan Kim
Deprecate KFD XGMI peer info calls in favour of calling directly from simplified XGMI peer info functions. Signed-off-by: Jonathan Kim <jonathan.kim@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: Remove unused nbif_v6_3_1_sriov_funcsDr. David Alan Gilbert
The nbif_v6_3_1_sriov_funcs instance of amdgpu_nbio_funcs was added in commit 894c6d3522d1 ("drm/amdgpu: Add nbif v6_3_1 ip block support") but has remained unused. Alex has confirmed it wasn't needed. Remove it, together with the four unused stub functions: nbif_v6_3_1_sriov_ih_doorbell_range nbif_v6_3_1_sriov_gc_doorbell_init nbif_v6_3_1_sriov_vcn_doorbell_range nbif_v6_3_1_sriov_sdma_doorbell_range Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: Add ring reset callback for JPEG5_0_1Sathishkumar S
Add ring reset function callback for JPEG5_0_1 to recover from job timeouts without a full gpu reset. Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com> Reviewed-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: Log after a successful ring resetAndré Almeida
When a ring reset happens, the kernel log shows only "amdgpu: Starting <ring name> ring reset", but when it finishes nothing appears in the log. Explicitly write in the log that the reset has finished correctly. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: André Almeida <andrealmeid@igalia.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: Log the creation of a coredump fileAndré Almeida
After a GPU reset happens, the driver creates a coredump file. However, the user might not be aware of it. Log the file creation the user can find more information about the device and add the file to bug reports. This is similar to what the xe driver does. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: André Almeida <andrealmeid@igalia.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu/mes: keep enforce isolation up to dateAlex Deucher
Re-send the mes message on resume to make sure the mes state is up to date. Fixes: 8521e3c5f058 ("drm/amd/amdgpu: limit single process inside MES") Acked-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: Shaoyun Liu <shaoyun.liu@amd.com> Cc: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: Add core reset registers for JPEG5_0_1Sathishkumar S
Add core reset control register definitions and align all prior register definitions to end at 100 column length for uniformity. Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com> Reviewed-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: Per-instance init func for JPEG5_0_1Sathishkumar S
Add helper functions to handle per-instance and per-core initialization and deinitialization in JPEG5_0_1. Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com> Reviewed-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: disable BAR resize on Dell G5 SEAlex Deucher
There was a quirk added to add a workaround for a Sapphire RX 5600 XT Pulse that didn't allow BAR resizing. However, the quirk caused a regression with runtime pm on Dell laptops using those chips, rather than narrowing the scope of the resizing quirk, add a quirk to prevent amdgpu from resizing the BAR on those Dell platforms unless runtime pm is disabled. v2: update commit message, add runpm check Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/1707 Fixes: 907830b0fc9e ("PCI: Add a REBAR size quirk for Sapphire RX 5600 XT Pulse") Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amd/pm: Fetch fru product info for smu_v13_0_12Asad Kamal
Fetch fru product info for smu_v13_0_12 from static metrics table v2: Field by field copy for fru info(Lijo) Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: Update amdgpu_job_timedout to check if the ring is guiltyJesse.zhang@amd.com
This patch updates the `amdgpu_job_timedout` function to check if the ring is actually guilty of causing the timeout. If not, it skips error handling and fence completion. v2: move the is_guilty check down into the queue reset area (Alex) v3: need to call is_guilty before reset (Alex) v4: squash in is_guilty logic fixes (Alex) Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Jesse Zhang <jesse.zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: Add reset function pointer for SDMA v4.4.2 page ringJesse.zhang@amd.com
This patch adds a reset function pointer to the SDMA v4.4.2 page ring functionality. The new function pointer `reset` is set to `sdma_v4_4_2_reset_queue`, which is responsible for resetting the SDMA queue. Changes: - Add `reset` function pointer to `sdma_v4_4_2_page_ring_funcs`. Signed-off-by: Jesse Zhang <jesse.zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: Improve SDMA reset logic with guilty queue trackingJesse.zhang@amd.com
This patch includes the remaining improvements to the SDMA reset logic: - Added `gfx_guilty` and `page_guilty` flags to track guilty queues. - Updated the reset and resume functions to handle the guilty state. - Cached the `rptr` before reset. v2: 1.replace the caller with a guilty bool. If the queue is the guilty one, set the rptr and wptr to the saved wptr value, else, set the rptr and wptr to the saved rptr value. (Alex) 2. cache the rptr before the reset. (Alex) v3: Keeping intermediate variables like u64 rwptr simplifies resotre rptr/wptr.(Lijo) Suggested-by: Alex Deucher <alexander.deucher@amd.com> Suggested-by: Jiadong Zhu <Jiadong.Zhu@amd.com> Signed-off-by: Jesse Zhang <jesse.zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu/sdma: Introduce is_guilty callbacks for sdma GFX and PAGE ringsJesse.zhang@amd.com
This patch introduces the `is_guilty` callbacks for the GFX and PAGE rings. These callbacks check if a ring is guilty of causing a timeout or error. Suggested-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Jesse Zhang <jesse.zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: Introduce cached_rptr and is_guilty callback in amdgpu_ringJesse.zhang@amd.com
This patch introduces the following changes: - Add `cached_rptr` to the `amdgpu_ring` structure to store the read pointer before a reset. - Add `is_guilty` callback to the `amdgpu_ring_funcs` structure to check if a ring is guilty of causing a timeout. Suggested-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Jesse Zhang <jesse.zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: Introduce conditional user queue suspension for SDMA resetsJesse.zhang@amd.com
- Modify the `amdgpu_sdma_reset_engine` function to accept a `suspend_user_queues` parameter. - This parameter allows the function to conditionally suspend and resume user queues during SDMA resets. - Ensure that user queues are suspended only when necessary to avoid unnecessary overhead and potential deadlocks. - Restart the scheduler's work queue for the GFX and page rings after the reset to allow new tasks to be submitted. This change improves synchronization between the KGD and the KFD during SDMA resets, ensuring proper handling of user queues and avoiding race conditions. V2: replace the ring_lock with the existed the scheduler locks for the queues (ring->sched) on the sdma engine.(Alex) v3: call drm_sched_wqueue_stop() rather than job_list_lock. If a GPU ring reset was already initiated for one ring at amdgpu_job_timedout, skip resetting that ring and call drm_sched_wqueue_stop() for the other rings (Alex) replace the common lock (sdma_reset_lock) with DQM lock to to resolve reset races between the two driver sections during KFD eviction.(Jon) Rename the caller to Reset_src and Change AMDGPU_RESET_SRC_SDMA_KGD/KFD to AMDGPU_RESET_SRC_SDMA_HWS/RING (Jon) v4: restart the wqueue if the reset was successful, or fall back to a full adapter reset. (Alex) move definition of reset source to enumeration AMDGPU_RESET_SRCS, and check reset src in amdgpu_sdma_reset_instance (Jon) v5: Call amdgpu_amdkfd_suspend/resume at the start/end of reset function respectively under !SRC_HWS conditions only (Jon) v6: replace the paramter src with a bool suspend_user_queues, remove the paramter src in pre/post func. (Jon) Suggested-by: Alex Deucher <alexander.deucher@amd.com> Suggested-by: Jiadong Zhu <Jiadong.Zhu@amd.com> Suggested-by: Jonathan Kim <Jonathan.Kim@amd.com> Signed-off-by: Jesse Zhang <jesse.zhang@amd.com> Acked-by: Jonathan Kim <jonathan.kim@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: Remove redundant logic in GC v9.4.3Lijo Lazar
GFXOFF check is not needed for GC v9.4.3. Also, save/restore list is available by default. Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: Do not poweroff UVDJ in JPEG4_0_3Sathishkumar S
Update power gate setting to not poweroff UVDJ in JPEG4_0_3. Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com> Reviewed-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu/sdma: Refactor SDMA reset functionality and add callback supportJesse.zhang@amd.com
This patch refactors the SDMA reset functionality in the `sdma_v4_4_2` driver to improve modularity and support shared usage between AMDGPU and KFD. The changes include: 1. **Refactored SDMA Reset Logic**: - Split the `sdma_v4_4_2_reset_queue` function into two separate functions: - `sdma_v4_4_2_stop_queue`: Stops the SDMA queue before reset. - `sdma_v4_4_2_restore_queue`: Restores the SDMA queue after reset. - These functions are now used as callbacks for the shared reset mechanism. 2. **Added Callback Support**: - Introduced a new structure `sdma_v4_4_2_reset_funcs` to hold the stop and restore callbacks. - Added `sdma_v4_4_2_set_reset_funcs` to register these callbacks with the shared reset mechanism using `amdgpu_set_on_reset_callbacks`. 3. **Fixed Reset Queue Function**: - Modified `sdma_v4_4_2_reset_queue` to use the shared `amdgpu_sdma_reset_queue` function, ensuring consistency across the driver. This patch ensures that SDMA reset functionality is more modular, reusable, and aligned with the shared reset mechanism between AMDGPU and KFD. v2: Renamed sdma_v4_4_2_set_reset_funcs to sdma_v4_4_2_set_engine_reset_funcs. Renamed sdma_v4_4_2_reset_funcs to sdma_v4_4_2_engine_reset_funcs.(Alex) Suggested-by: Jiadong Zhu <Jiadong.Zhu@amd.com> Suggested-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Jesse Zhang <jesse.zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu/kfd: Add shared SDMA reset functionality with callback supportJesse.zhang@amd.com
This patch introduces shared SDMA reset functionality between AMDGPU and KFD. The implementation includes the following key changes: 1. Added `amdgpu_sdma_reset_queue`: - Resets a specific SDMA queue by instance ID. - Invokes registered pre-reset and post-reset callbacks to allow KFD and AMDGPU to save/restore their state during the reset process. 2. Added `amdgpu_set_on_reset_callbacks`: - Allows KFD and AMDGPU to register callback functions for pre-reset and post-reset operations. - Callbacks are stored in a global linked list and invoked in the correct order during SDMA reset. This patch ensures that both AMDGPU and KFD can handle SDMA reset events gracefully, with proper state saving and restoration. It also provides a flexible callback mechanism for future extensions. v2: fix CamelCase and put the SDMA helper into amdgpu_sdma.c (Alex) v3: rename the `amdgpu_register_on_reset_callbacks` function to `amdgpu_sdma_register_on_reset_callbacks` move global reset_callback_list to struct amdgpu_sdma (Alex) v4: Update the reset callback function description and rename the reset function to amdgpu_sdma_reset_engine (Alex) Suggested-by: Alex Deucher <alexander.deucher@amd.com> Suggested-by: Jiadong Zhu <Jiadong.Zhu@amd.com> Signed-off-by: Jesse Zhang <jesse.zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: correct the name of mes_pipe structureLikun Gao
Correct the structure name admgpu_mes_pipe to amdgpu_mes_pipe. Signed-off-by: Likun Gao <Likun.Gao@amd.com> Reviewed-by: Mukul Joshi <mukul.joshi@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25drm/amdgpu: update the handle ptr in is_idleSunil Khatri
Update the *handle to amdgpu_ip_block ptr for all functions pointers of is_idle. Signed-off-by: Sunil Khatri <sunil.khatri@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-02-25Merge drm/drm-next into drm-misc-nextThomas Zimmermann
Backmerging to get fixes from v6.14-rc4. Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
2025-02-25Merge tag 'v6.14-rc4' into drm-nextDave Airlie
Backmerge Linux 6.14-rc4 at the request of tzimmermann so misc-next can base on rc4. Signed-off-by: Dave Airlie <airlied@redhat.com>
generated by cgit v1.2.3 (git 2.39.1) at 2025-06-21 00:02:13 +0000