- 22 Dec, 2022 1 commit
-
-
Dave Airlie authored
Merge tag 'amd-drm-fixes-6.2-2022-12-21' of https://gitlab.freedesktop.org/agd5f/linux into drm-next amd-drm-fixes-6.2-2022-12-21: amdgpu: - Avoid large variable on the stack - S0ix fixes - SMU 13.x fixes - VCN fix - Add missing fence reference amdkfd: - Fix init vm error handling - Fix double release of compute pasid Signed-off-by:
Dave Airlie <airlied@redhat.com> From: Alex Deucher <alexander.deucher@amd.com> Link: https://patchwork.freedesktop.org/patch/msgid/20221221205828.6093-1-alexander.deucher@amd.com
-
- 21 Dec, 2022 2 commits
-
-
Christian König authored
That function consumes the reference. Reviewed-by:
Luben Tuikov <luben.tuikov@amd.com> Reported-by:
Borislav Petkov (AMD) <bp@alien8.de> Tested-by:
Christian König <christian.koenig@amd.com> Fixes: aab9cf7b ("drm/amdgpu: use scheduler dependencies for VM updates") Signed-off-by:
Alex Deucher <alexander.deucher@amd.com>
-
Saleemkhan Jamadar authored
Enable VCN Dynamic Power Gating control for GC IP v11.0.4. Signed-off-by:
Saleemkhan Jamadar <saleemkhan.jamadar@amd.com> Reviewed-by:
Veerabadhran Gopalakrishnan <veerabadhran.gopalakrishnan@amd.com> Signed-off-by:
-
- 20 Dec, 2022 9 commits
-
-
Tim Huang authored
MES is part of gfxoff and MES suspend and resume are skipped for S0i3. But the mes_self_test call path is still in the amdgpu_device_ip_late_init. it's should also be skipped for s0ix as no hardware re-initialization happened. Besides, mes_self_test will free the BO that triggers a lot of warning messages while in the suspend state. [ 81.656085] WARNING: CPU: 2 PID: 1550 at drivers/gpu/drm/amd/amdgpu/amdgpu_object.c:425 amdgpu_bo_free_kernel+0xfc/0x110 [amdgpu] [ 81.679435] Call Trace: [ 81.679726] <TASK> [ 81.679981] amdgpu_mes_remove_hw_queue+0x17a/0x230 [amdgpu] [ 81.680857] amdgpu_mes_self_test+0x390/0x430 [amdgpu] [ 81.681665] mes_v11_0_late_init+0x37/0x50 [amdgpu] [ 81.682423] amdgpu_device_ip_late_init+0x53/0x280 [amdgpu] [ 81.683257] amdgpu_device_resume+0xae/0x2a0 [amdgpu] [ 81.684043] amdgpu_pmops_resume+0x37/0x70 [amdgpu] [ 81.684818] pci_pm_resume+0x5c/0xa0 [ 81.685247] ? pci_pm_thaw+0x90/0x90 [ 81.685658] dpm_run_callback+0x4e/0x160 [ 81.686110] device_resume+0xad/0x210 [ 81.686529] async_resume+0x1e/0x40 [ 81.686931] async_run_entry_fn+0x33/0x120 [ 81.687405] process_one_work+0x21d/0x3f0 [ 81.687869] worker_thread+0x4a/0x3c0 [ 81.688293] ? process_one_work+0x3f0/0x3f0 [ 81.688777] kthread+0xff/0x130 [ 81.689157] ? kthread_complete_and_exit+0x20/0x20 [ 81.689707] ret_from_fork+0x22/0x30 [ 81.690118] </TASK> [ 81.690380] ---[ end trace 0000000000000000 ]--- v2: make the comment clean and use adev->in_s0ix instead of adev->suspend Signed-off-by:
Tim Huang <tim.huang@amd.com> Reviewed-by:
Mario Limonciello <mario.limonciello@amd.com> Signed-off-by:
-
Evan Quan authored
For SMU 13.0.0 and 13.0.7, the output from PMFW is in percent. Driver need to convert that into correct PMW(255) based. Signed-off-by:
Evan Quan <evan.quan@amd.com> Acked-by:
-
Evan Quan authored
To fit the latest PMFW and suppress the warning emerged on driver loading. Signed-off-by:
-
Alex Deucher authored
It's also part of gfxoff. Cc: stable@vger.kernel.org # 6.0, 6.1 Reviewed-by:
-
Arnd Bergmann authored
The activity_monitor_external[] array is too big to fit on the kernel stack, resulting in this warning with clang: drivers/gpu/drm/amd/amdgpu/../pm/swsmu/smu13/smu_v13_0_7_ppt.c:1438:12: error: stack frame size (1040) exceeds limit (1024) in 'smu_v13_0_7_get_power_profile_mode' [-Werror,-Wframe-larger-than] Use dynamic allocation instead. It should also be possible to have single element here instead of the array, but this seems easier. v2: fix up argument to sizeof() (Alex) Fixes: 334682ae ("drm/amd/pm: enable workload type change on smu_v13_0_7") Signed-off-by:
Arnd Bergmann <arnd@arndb.de> Signed-off-by:
-
Philip Yang authored
If kfd_process_device_init_vm returns failure after vm is converted to compute vm and vm->pasid set to compute pasid, KFD will not take pdd->drm_file reference. As a result, drm close file handler maybe called to release the compute pasid before KFD process destroy worker to release the same pasid and set vm->pasid to zero, this generates below WARNING backtrace and NULL pointer access. Add helper amdgpu_amdkfd_gpuvm_set_vm_pasid and call it at the last step of kfd_process_device_init_vm, to ensure vm pasid is the original pasid if acquiring vm failed or is the compute pasid with pdd->drm_file reference taken to avoid double release same pasid. amdgpu: Failed to create process VM object ida_free called for id=32770 which is not allocated. WARNING: CPU: 57 PID: 72542 at ../lib/idr.c:522 ida_free+0x96/0x140 RIP: 0010:ida_free+0x96/0x140 Call Trace: amdgpu_pasid_free_delayed+0xe1/0x2a0 [amdgpu] amdgpu_driver_postclose_kms+0x2d8/0x340 [amdgpu] drm_file_free.part.13+0x216/0x270 [drm] drm_close_helper.isra.14+0x60/0x70 [drm] drm_release+0x6e/0xf0 [drm] __fput+0xcc/0x280 ____fput+0xe/0x20 task_work_run+0x96/0xc0 do_exit+0x3d0/0xc10 BUG: kernel NULL pointer dereference, address: 0000000000000000 RIP: 0010:ida_free+0x76/0x140 Call Trace: amdgpu_pasid_free_delayed+0xe1/0x2a0 [amdgpu] amdgpu_driver_postclose_kms+0x2d8/0x340 [amdgpu] drm_file_free.part.13+0x216/0x270 [drm] drm_close_helper.isra.14+0x60/0x70 [drm] drm_release+0x6e/0xf0 [drm] __fput+0xcc/0x280 ____fput+0xe/0x20 task_work_run+0x96/0xc0 do_exit+0x3d0/0xc10 Signed-off-by:
Philip Yang <Philip.Yang@amd.com> Reviewed-by:
Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by:
-
Philip Yang authored
Should only destroy the ib_mem and let process cleanup worker to free the outstanding BOs. Reset the pointer in pdd->qpd structure, to avoid NULL pointer access in process destroy worker. BUG: kernel NULL pointer dereference, address: 0000000000000010 Call Trace: amdgpu_amdkfd_gpuvm_unmap_gtt_bo_from_kernel+0x46/0xb0 [amdgpu] kfd_process_device_destroy_cwsr_dgpu+0x40/0x70 [amdgpu] kfd_process_destroy_pdds+0x71/0x190 [amdgpu] kfd_process_wq_release+0x2a2/0x3b0 [amdgpu] process_one_work+0x2a1/0x600 worker_thread+0x39/0x3d0 Signed-off-by:
-
Dave Airlie authored
Merge tag 'drm-intel-next-fixes-2022-12-15' of git://anongit.freedesktop.org/drm/drm-intel into drm-next - Documentation fixe (Matt, Miaoqian) - OA-perf related fix (Umesh) - VLV/CHV HDMI/DP audio fix (Ville) - Display DDI/Transcoder fix (Khaled) - Migrate fixes (Chris, Matt) Signed-off-by:
-
Dave Airlie authored
Merge tag 'amd-drm-fixes-6.2-2022-12-15' of https://gitlab.freedesktop.org/agd5f/linux into drm-next amd-drm-fixes-6.2-2022-12-15: amdgpu: - Spelling fix - BO pin fix - Properly handle polaris 10/11 overlap asics - GMC9 fix - SR-IOV suspend fix - DCN 3.1.4 fix - KFD userptr locking fix - SMU13.x fixes - GDS/GWS/OA handling fix - Reserved VMID handling fixes - FRU EEPROM fix - BO validation fixes Signed-off-by:
-
- 15 Dec, 2022 4 commits
-
-
Evan Quan authored
Update the reported maximum shader clock to the value which can be guarded to be achieved on all cards. This is to align with Window setting. Signed-off-by:
-
Evan Quan authored
Correct the pstate standard/peak profiling mode clock settings for SMU13.0.0. Signed-off-by:
-
Evan Quan authored
To better support UMD pstate profilings, the GPO feature needs to be switched on/off accordingly. Signed-off-by:
-
Evan Quan authored
To better support UMD pstate profilings, the GPO feature needs to be switched on/off accordingly. Signed-off-by:
-
- 14 Dec, 2022 16 commits
-
-
Christian König authored
This reverts commit f9d00a4a. This causes problem for KFD because when we overcommit we accidentially bind the BO to GTT for moving it into VRAM. We also need to make sure that this is done only as fallback after trying to evict first. Signed-off-by:
-
Luben Tuikov authored
Remove the "domain" argument to amdgpu_bo_create_kernel_at() since this function takes an "offset" argument which is the offset off of VRAM, and as such allocation always takes place in VRAM. Thus, the "domain" argument is unnecessary. Cc: Alex Deucher <Alexander.Deucher@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: AMD Graphics <amd-gfx@lists.freedesktop.org> Signed-off-by:
-
Luben Tuikov authored
Fix amdgpu_bo_validate_size() to check whether the TTM domain manager for the requested memory exists, else we get a kernel oops when dereferencing "man". v2: Make the patch standalone, i.e. not dependent on local patches. v3: Preserve old behaviour and just check that the manager pointer is not NULL. v4: Complain if GTT domain requested and it is uninitialized--most likely a bug. Cc: Alex Deucher <Alexander.Deucher@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: AMD Graphics <amd-gfx@lists.freedesktop.org> Signed-off-by: -
Luben Tuikov authored
Always check if fru_addr is not NULL. This commit also fixes a "smatch" warning. v2: Add a Fixes tag. Cc: Alex Deucher <Alexander.Deucher@amd.com> Cc: Dan Carpenter <error27@gmail.com> Cc: kernel test robot <lkp@intel.com> Cc: AMD Graphics <amd-gfx@lists.freedesktop.org> Fixes: afbe5d1e ("drm/amdgpu: Bug-fix: Reading I2C FRU data on newer ASICs") Signed-off-by:
Kent Russell <kent.russell@amd.com> Signed-off-by:
-
Matthew Auld authored
It seems we can have one or more framebuffers that are still pinned when suspending lmem, in such a case we end up creating a shmem backup object, instead of evicting the object directly, but this will skip copying the CCS aux state, since we don't allocate the extra storage for the CCS pages as part of the ttm_tt construction. Since we can already deal with pinned objects just fine, it doesn't seem too nasty to just extend to support dealing with the CCS aux state, if the object is a pinned framebuffer. This fixes display corruption (like in gnome-shell) seen on DG2 when returning from suspend. Fixes: da0595ae ("drm/i915/migrate: Evict and restore the flatccs capable lmem obj") Signed-off-by:
Matthew Auld <matthew.auld@intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Nirmoy Das <nirmoy.das@intel.com> Cc: Andrzej Hajda <andrzej.hajda@intel.com> Cc: Shuicheng Lin <shuicheng.lin@intel.com> Cc: <stable@vger.kernel.org> # v5.19+ Tested-by:
Nirmoy Das <nirmoy.das@intel.com> Reviewed-by:
Rodrigo Vivi <rodrigo.vivi@intel.com>
-
Matthew Auld authored
In the case of lmem -> lmem transfers, which is currently only possible with small-bar systems, we need to ensure we copy the CCS aux state as-is, rather than nuke it. This should fix some nasty display corruption sometimes seen on DG2 small-bar systems, when also using DG2_RC_CCS_CC for the surface. Fixes: e3afc690 ("drm/i915/display: consider DG2_RC_CCS_CC when migrating buffers") Signed-off-by:
-
Christian König authored
Instead of reserving a VMID for a single process allow that many processes use the reserved ID. This allows for proper isolation between the processes. Signed-off-by:
-
Christian König authored
This is completely pointless since the VMID always stays allocated until the VM is idle. Signed-off-by:
-
Christian König authored
This should probably not access job->vm and also emit the SPM switch under the conditional execute. Signed-off-by:
-
Christian König authored
Bas pointed out that this isn't working as expected and could cause crashes. Fix the handling by storing the marker that a switch is needed inside the job instead. Reported-by:
Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl> Signed-off-by:
-
Evan Quan authored
Without this, the pp_dpm_vclk and pp_dpm_dclk outputs are not with correct data. Signed-off-by:
-
Evan Quan authored
Without this, the pp_dpm_vclk and pp_dpm_dclk outputs are not with correct data. Signed-off-by:
-
Felix Kuehling authored
Add a per-process MMU notifier lock for processing notifiers from userptrs. Use that lock to properly synchronize page table updates with MMU notifiers. Signed-off-by:
-
Yifan Zhang authored
Add display SG support for DCN 3.1.4. Signed-off-by:
Yifan Zhang <yifan1.zhang@amd.com> Reviewed-by:
-
Christian König authored
When buffers are freed during suspend there is no guarantee that they can be re-allocated during resume. The PSP subsystem seems to be quite buggy regarding this, so add a WARN_ON() to point out those bugs. Signed-off-by:
Alex Deucher <alexdeucher@amd.com> Tested-by:
Guilherme G. Piccoli <gpiccoli@igalia.com> Signed-off-by:
-
Christian König authored
We not only need to make sure that we have a BO, but also that the BO has some backing store. Fixes: d1a372af ("drm/amdgpu: Set MTYPE in PTE based on BO flags") Signed-off-by:
-
- 13 Dec, 2022 4 commits
-
-
Shikang Fan authored
- evict_resource is taking too long causing sriov full access mode timeout. So, add an extra evict_resource in the beginning as an early evict. Signed-off-by:
Shikang Fan <shikang.fan@amd.com> Reviewed-by:
-
Chris Wilson authored
If the ring is nearly full when calling into emit_pte(), we might incorrectly trample the reserved_space when constructing the packet to emit the PTEs. This then triggers the GEM_BUG_ON(rq->reserved_space > ring->space) when later submitting the request, since the request itself doesn't have enough space left in the ring to emit things like workarounds, breadcrumbs etc. v2: Fix the whitespace errors Testcase: igt@i915_selftests@live_emit_pte_full_ring Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/7535 Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/6889 Fixes: cf586021 ("drm/i915/gt: Pipelined page migration") Signed-off-by:
Chris Wilson <chris.p.wilson@intel.com> Signed-off-by:
Andrzej Hajda <andrzej.hajda@intel.com> Reviewed-by:
Andi Shyti <andi.shyti@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20221202122844.428006-1-matthew.auld@intel.com (cherry picked from commit 35168a6c) Signed-off-by:
-
Khaled Almahallawy authored
Bspecs has updated recently to remove the restriction to disable DDI/Transcoder before setting PHY test pattern. This update is to address PHY compliance test failures observed on a port with LTTPR. The issue is that when Transc. is disabled, the main link signals fed to LTTPR will be dropped invalidating link training, which will affect the quality of the phy test pattern when the transcoder is enabled again. v2: Update commit message (Clint) v3: Add missing Signed-off in v2 v4: Update Bspec and commit message for pre-gen12 (Jani) Bspec: 50482, 7555 Fixes: 8cdf7271 ("drm/i915/dp: Program vswing, pre-emphasis, test-pattern") Cc: Imre Deak <imre.deak@intel.com> Cc: Clint Taylor <clinton.a.taylor@intel.com> CC: Jani Nikula <jani.nikula@intel.com> Tested-by:
Khaled Almahallawy <khaled.almahallawy@intel.com> Reviewed-by:
Clint Taylor <clinton.a.taylor@intel.com> Signed-off-by:
Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20221123220926.170034-1-khaled.almahallawy@intel.com (cherry picked from commit be4a8476) Signed-off-by:
-
Ville Syrjälä authored
Despite what I claimed in commit c3c5dc1d ("drm/i915/audio: Do the vblank waits") the vblank interrupts are in fact not enabled yet when we do the audio enable sequence on VLV/CHV (all other platforms are fine). Reorder the enable sequence on VLV/CHV to match that of the other platforms so that the audio enable happens after the pipe has been enabled. Fixes: c3c5dc1d ("drm/i915/audio: Do the vblank waits") Signed-off-by:
Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20221207225219.29060-1-ville.syrjala@linux.intel.comReviewed-by:
-
- 09 Dec, 2022 4 commits
-
-
Alex Deucher authored
Some special polaris 10 chips overlap with the polaris11 DID range. Handle this properly in the driver. v2: use local flags for other function calls. Acked-by:
-
Alex Deucher authored
Only apply the static threshold for Stoney and Carrizo. This hardware has certain requirements that don't allow mixing of GTT and VRAM. Newer asics do not have these requirements so we should be able to be more flexible with where buffers end up. Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/2270 Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/2291 Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/2255Acked-by:
-
Colin Ian King authored
There is a spelling mistake in the struct field dram_clk_chanage. Fix it. Signed-off-by:
Colin Ian King <colin.i.king@gmail.com> Signed-off-by:
Hamza Mahfooz <hamza.mahfooz@amd.com> Signed-off-by:
-
Umesh Nerlige Ramappa authored
An earlier commit introduced a mechanism to parse the context image to find the OA context control offset. This resulted in an NPD on haswell when gem_context was passed into i915_perf_open_ioctl params. Haswell does not support logical ring contexts, so ensure that the context image is parsed only for platforms with logical ring contexts and also validate lrc_reg_state. v2: Fix build failure v3: Fix checkpatch error Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/7432 Fixes: a5c3a3cb ("drm/i915/perf: Determine gen12 oa ctx offset at runtime") Signed-off-by:
Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com> Reviewed-by:
Ashutosh Dixit <ashutosh.dixit@intel.com> Signed-off-by:
John Harrison <John.C.Harrison@Intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20221123235342.713068-1-umesh.nerlige.ramappa@intel.com (cherry picked from commit 95c713d7) Signed-off-by:
-
