summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
AgeCommit message (Collapse)Author
2025-05-22Revert "drm/amd: Stop evicting resources on APUs in suspend"Alex Deucher
[ Upstream commit d0ce1aaa8531a4a4707711cab5721374751c51b0 ] This reverts commit 3a9626c816db901def438dc2513622e281186d39. This breaks S4 because we end up setting the s3/s0ix flags even when we are entering s4 since prepare is used by both flows. The causes both the S3/s0ix and s4 flags to be set which breaks several checks in the driver which assume they are mutually exclusive. Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3634 Cc: Mario Limonciello <mario.limonciello@amd.com> Reviewed-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit ce8f7d95899c2869b47ea6ce0b3e5bf304b2fff4) Cc: stable@vger.kernel.org Signed-off-by: Sasha Levin <sashal@kernel.org>
2025-05-22drm/amd: Add Suspend/Hibernate notification callback supportMario Limonciello
[ Upstream commit 2965e6355dcdf157b5fafa25a2715f00064da8bf ] As part of the suspend sequence VRAM needs to be evicted on dGPUs. In order to make suspend/resume more reliable we moved this into the pmops prepare() callback so that the suspend sequence would fail but the system could remain operational under high memory usage suspend. Another class of issues exist though where due to memory fragementation there isn't a large enough contiguous space and swap isn't accessible. Add support for a suspend/hibernate notification callback that could evict VRAM before tasks are frozen. This should allow paging out to swap if necessary. Link: https://github.com/ROCm/ROCK-Kernel-Driver/issues/174 Link: https://gitlab.freedesktop.org/drm/amd/-/issues/3476 Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/2362 Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3781 Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Link: https://lore.kernel.org/r/20241128032656.2090059-2-superm1@kernel.org Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Stable-dep-of: d0ce1aaa8531 ("Revert "drm/amd: Stop evicting resources on APUs in suspend"") Signed-off-by: Sasha Levin <sashal@kernel.org>
2025-04-25drm/amdgpu: fix warning of drm_mm_cleanZhenGuo Yin
commit e7afa85a0d0eba5bf2c0a446ff622ebdbc9812d6 upstream. Kernel doorbell BOs needs to be freed before ttm_fini. Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4145 Fixes: 54c30d2a8def ("drm/amdgpu: create kernel doorbell pages") Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: ZhenGuo Yin <zhenguo.yin@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 39938a8ed979e398faa3791a47e282c82bcc6f04) Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2025-04-20drm/amdgpu: grab an additional reference on the gang fence v2Christian König
[ Upstream commit 0d9a95099dcb05b5f4719c830d15bf4fdcad0dc2 ] We keep the gang submission fence around in adev, make sure that it stays alive. v2: fix memory leak on retry Signed-off-by: Christian König <christian.koenig@amd.com> Acked-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
2025-03-07drm/amdgpu: disable BAR resize on Dell G5 SEAlex Deucher
commit 099bffc7cadff40bfab1517c3461c53a7a38a0d7 upstream. 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 Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2025-01-09drm/amdgpu: fix backport of commit 73dae652dcacAlex Deucher
Commit 73dae652dcac ("drm/amdgpu: rework resume handling for display (v2)") missed a small code change when it was backported resulting in an automatic backlight control breakage. Fix the backport. Note that this patch is not in Linus' tree as it is not required there; the bug was introduced in the backport. Fixes: 99a02eab8251 ("drm/amdgpu: rework resume handling for display (v2)") Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3853 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org # 6.11.x Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2024-12-14drm/amdgpu: rework resume handling for display (v2)Alex Deucher
commit 73dae652dcac776296890da215ee7dec357a1032 upstream. Split resume into a 3rd step to handle displays when DCC is enabled on DCN 4.0.1. Move display after the buffer funcs have been re-enabled so that the GPU will do the move and properly set the DCC metadata for DCN. v2: fix fence irq resume ordering Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org # 6.11.x Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2024-12-14drm/amdgpu: skip amdgpu_device_cache_pci_state under sriovVictor Zhao
[ Upstream commit afe260df55ac280cd56306248cb6d8a6b0db095c ] Under sriov, host driver will save and restore vf pci cfg space during reset. And during device init, under sriov, pci_restore_state happens after fullaccess released, and it can have race condition with mmio protection enable from host side leading to missing interrupts. So skip amdgpu_device_cache_pci_state for sriov. Signed-off-by: Victor Zhao <Victor.Zhao@amd.com> Acked-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
2024-12-09drm/amdgpu: fix usage slab after freeVitaly Prosyak
commit b61badd20b443eabe132314669bb51a263982e5c upstream. [ +0.000021] BUG: KASAN: slab-use-after-free in drm_sched_entity_flush+0x6cb/0x7a0 [gpu_sched] [ +0.000027] Read of size 8 at addr ffff8881b8605f88 by task amd_pci_unplug/2147 [ +0.000023] CPU: 6 PID: 2147 Comm: amd_pci_unplug Not tainted 6.10.0+ #1 [ +0.000016] Hardware name: ASUS System Product Name/ROG STRIX B550-F GAMING (WI-FI), BIOS 1401 12/03/2020 [ +0.000016] Call Trace: [ +0.000008] <TASK> [ +0.000009] dump_stack_lvl+0x76/0xa0 [ +0.000017] print_report+0xce/0x5f0 [ +0.000017] ? drm_sched_entity_flush+0x6cb/0x7a0 [gpu_sched] [ +0.000019] ? srso_return_thunk+0x5/0x5f [ +0.000015] ? kasan_complete_mode_report_info+0x72/0x200 [ +0.000016] ? drm_sched_entity_flush+0x6cb/0x7a0 [gpu_sched] [ +0.000019] kasan_report+0xbe/0x110 [ +0.000015] ? drm_sched_entity_flush+0x6cb/0x7a0 [gpu_sched] [ +0.000023] __asan_report_load8_noabort+0x14/0x30 [ +0.000014] drm_sched_entity_flush+0x6cb/0x7a0 [gpu_sched] [ +0.000020] ? srso_return_thunk+0x5/0x5f [ +0.000013] ? __kasan_check_write+0x14/0x30 [ +0.000016] ? __pfx_drm_sched_entity_flush+0x10/0x10 [gpu_sched] [ +0.000020] ? srso_return_thunk+0x5/0x5f [ +0.000013] ? __kasan_check_write+0x14/0x30 [ +0.000013] ? srso_return_thunk+0x5/0x5f [ +0.000013] ? enable_work+0x124/0x220 [ +0.000015] ? __pfx_enable_work+0x10/0x10 [ +0.000013] ? srso_return_thunk+0x5/0x5f [ +0.000014] ? free_large_kmalloc+0x85/0xf0 [ +0.000016] drm_sched_entity_destroy+0x18/0x30 [gpu_sched] [ +0.000020] amdgpu_vce_sw_fini+0x55/0x170 [amdgpu] [ +0.000735] ? __kasan_check_read+0x11/0x20 [ +0.000016] vce_v4_0_sw_fini+0x80/0x110 [amdgpu] [ +0.000726] amdgpu_device_fini_sw+0x331/0xfc0 [amdgpu] [ +0.000679] ? mutex_unlock+0x80/0xe0 [ +0.000017] ? __pfx_amdgpu_device_fini_sw+0x10/0x10 [amdgpu] [ +0.000662] ? srso_return_thunk+0x5/0x5f [ +0.000014] ? __kasan_check_write+0x14/0x30 [ +0.000013] ? srso_return_thunk+0x5/0x5f [ +0.000013] ? mutex_unlock+0x80/0xe0 [ +0.000016] amdgpu_driver_release_kms+0x16/0x80 [amdgpu] [ +0.000663] drm_minor_release+0xc9/0x140 [drm] [ +0.000081] drm_release+0x1fd/0x390 [drm] [ +0.000082] __fput+0x36c/0xad0 [ +0.000018] __fput_sync+0x3c/0x50 [ +0.000014] __x64_sys_close+0x7d/0xe0 [ +0.000014] x64_sys_call+0x1bc6/0x2680 [ +0.000014] do_syscall_64+0x70/0x130 [ +0.000014] ? srso_return_thunk+0x5/0x5f [ +0.000014] ? irqentry_exit_to_user_mode+0x60/0x190 [ +0.000015] ? srso_return_thunk+0x5/0x5f [ +0.000014] ? irqentry_exit+0x43/0x50 [ +0.000012] ? srso_return_thunk+0x5/0x5f [ +0.000013] ? exc_page_fault+0x7c/0x110 [ +0.000015] entry_SYSCALL_64_after_hwframe+0x76/0x7e [ +0.000014] RIP: 0033:0x7ffff7b14f67 [ +0.000013] Code: ff e8 0d 16 02 00 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 03 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 41 c3 48 83 ec 18 89 7c 24 0c e8 73 ba f7 ff [ +0.000026] RSP: 002b:00007fffffffe378 EFLAGS: 00000246 ORIG_RAX: 0000000000000003 [ +0.000019] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007ffff7b14f67 [ +0.000014] RDX: 0000000000000000 RSI: 00007ffff7f6f47a RDI: 0000000000000003 [ +0.000014] RBP: 00007fffffffe3a0 R08: 0000555555569890 R09: 0000000000000000 [ +0.000014] R10: 0000000000000000 R11: 0000000000000246 R12: 00007fffffffe5c8 [ +0.000013] R13: 00005555555552a9 R14: 0000555555557d48 R15: 00007ffff7ffd040 [ +0.000020] </TASK> [ +0.000016] Allocated by task 383 on cpu 7 at 26.880319s: [ +0.000014] kasan_save_stack+0x28/0x60 [ +0.000008] kasan_save_track+0x18/0x70 [ +0.000007] kasan_save_alloc_info+0x38/0x60 [ +0.000007] __kasan_kmalloc+0xc1/0xd0 [ +0.000007] kmalloc_trace_noprof+0x180/0x380 [ +0.000007] drm_sched_init+0x411/0xec0 [gpu_sched] [ +0.000012] amdgpu_device_init+0x695f/0xa610 [amdgpu] [ +0.000658] amdgpu_driver_load_kms+0x1a/0x120 [amdgpu] [ +0.000662] amdgpu_pci_probe+0x361/0xf30 [amdgpu] [ +0.000651] local_pci_probe+0xe7/0x1b0 [ +0.000009] pci_device_probe+0x248/0x890 [ +0.000008] really_probe+0x1fd/0x950 [ +0.000008] __driver_probe_device+0x307/0x410 [ +0.000007] driver_probe_device+0x4e/0x150 [ +0.000007] __driver_attach+0x223/0x510 [ +0.000006] bus_for_each_dev+0x102/0x1a0 [ +0.000007] driver_attach+0x3d/0x60 [ +0.000006] bus_add_driver+0x2ac/0x5f0 [ +0.000006] driver_register+0x13d/0x490 [ +0.000008] __pci_register_driver+0x1ee/0x2b0 [ +0.000007] llc_sap_close+0xb0/0x160 [llc] [ +0.000009] do_one_initcall+0x9c/0x3e0 [ +0.000008] do_init_module+0x241/0x760 [ +0.000008] load_module+0x51ac/0x6c30 [ +0.000006] __do_sys_init_module+0x234/0x270 [ +0.000007] __x64_sys_init_module+0x73/0xc0 [ +0.000006] x64_sys_call+0xe3/0x2680 [ +0.000006] do_syscall_64+0x70/0x130 [ +0.000007] entry_SYSCALL_64_after_hwframe+0x76/0x7e [ +0.000015] Freed by task 2147 on cpu 6 at 160.507651s: [ +0.000013] kasan_save_stack+0x28/0x60 [ +0.000007] kasan_save_track+0x18/0x70 [ +0.000007] kasan_save_free_info+0x3b/0x60 [ +0.000007] poison_slab_object+0x115/0x1c0 [ +0.000007] __kasan_slab_free+0x34/0x60 [ +0.000007] kfree+0xfa/0x2f0 [ +0.000007] drm_sched_fini+0x19d/0x410 [gpu_sched] [ +0.000012] amdgpu_fence_driver_sw_fini+0xc4/0x2f0 [amdgpu] [ +0.000662] amdgpu_device_fini_sw+0x77/0xfc0 [amdgpu] [ +0.000653] amdgpu_driver_release_kms+0x16/0x80 [amdgpu] [ +0.000655] drm_minor_release+0xc9/0x140 [drm] [ +0.000071] drm_release+0x1fd/0x390 [drm] [ +0.000071] __fput+0x36c/0xad0 [ +0.000008] __fput_sync+0x3c/0x50 [ +0.000007] __x64_sys_close+0x7d/0xe0 [ +0.000007] x64_sys_call+0x1bc6/0x2680 [ +0.000007] do_syscall_64+0x70/0x130 [ +0.000007] entry_SYSCALL_64_after_hwframe+0x76/0x7e [ +0.000014] The buggy address belongs to the object at ffff8881b8605f80 which belongs to the cache kmalloc-64 of size 64 [ +0.000020] The buggy address is located 8 bytes inside of freed 64-byte region [ffff8881b8605f80, ffff8881b8605fc0) [ +0.000028] The buggy address belongs to the physical page: [ +0.000011] page: refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x1b8605 [ +0.000008] anon flags: 0x17ffffc0000000(node=0|zone=2|lastcpupid=0x1fffff) [ +0.000007] page_type: 0xffffefff(slab) [ +0.000009] raw: 0017ffffc0000000 ffff8881000428c0 0000000000000000 dead000000000001 [ +0.000006] raw: 0000000000000000 0000000000200020 00000001ffffefff 0000000000000000 [ +0.000006] page dumped because: kasan: bad access detected [ +0.000012] Memory state around the buggy address: [ +0.000011] ffff8881b8605e80: fa fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc [ +0.000015] ffff8881b8605f00: 00 00 00 00 00 00 00 00 fc fc fc fc fc fc fc fc [ +0.000015] >ffff8881b8605f80: fa fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc [ +0.000013] ^ [ +0.000011] ffff8881b8606000: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fc [ +0.000014] ffff8881b8606080: fc fc fc fc fc fc fc fa fb fb fb fb fb fb fb fb [ +0.000013] ================================================================== The issue reproduced on VG20 during the IGT pci_unplug test. The root cause of the issue is that the function drm_sched_fini is called before drm_sched_entity_kill. In drm_sched_fini, the drm_sched_rq structure is freed, but this structure is later accessed by each entity within the run queue, leading to invalid memory access. To resolve this, the order of cleanup calls is updated: Before: amdgpu_fence_driver_sw_fini amdgpu_device_ip_fini After: amdgpu_device_ip_fini amdgpu_fence_driver_sw_fini This updated order ensures that all entities in the IPs are cleaned up first, followed by proper cleanup of the schedulers. Additional Investigation: During debugging, another issue was identified in the amdgpu_vce_sw_fini function. The vce.vcpu_bo buffer must be freed only as the final step in the cleanup process to prevent any premature access during earlier cleanup stages. v2: Using Christian suggestion call drm_sched_entity_destroy before drm_sched_fini. Cc: Christian König <christian.koenig@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Vitaly Prosyak <vitaly.prosyak@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2024-09-18drm/amdgpu: nuke the VM PD/PT shadow handlingChristian König
This was only used as workaround for recovering the page tables after VRAM was lost and is no longer necessary after the function amdgpu_vm_bo_reset_state_machine() started to do the same. Compute never used shadows either, so the only proplematic case left is SVM and that is most likely not recoverable in any way when VRAM is lost. Signed-off-by: Christian König <christian.koenig@amd.com> Acked-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-17drm/amdgpu: Fix missing check pcie_p2p module paramBob Zhou
The module param pcie_p2p should be checked for kfd p2p feature, so add it. Fixes: 75f0efbc4b3b ("drm/amdgpu: Take IOMMU remapping into account for p2p checks") Signed-off-by: Bob Zhou <bob.zhou@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-08-29drm/amd/amdgpu: move drain_workqueue before shutdown is setVictor Zhao
[background] when unloading amdgpu driver right after running a workload, drain_workqueue is causing "Fence fallback timer expired on ring sdma0.0". Under sriov, this issue will cause sriov full access timeout and a reset happening. move drain_workqueue before shutdown is set to allow ih process and before enter full access under sriov to avoid full access time cost. Signed-off-by: Victor Zhao <Victor.Zhao@amd.com> Reviewed-by: Feifei Xu <Feifei.Xu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-08-29drm/amdgpu: skip printing vram_lost if neededTrigger Huang
The vm lost status can only be obtained after a GPU reset occurs, but sometimes a dev core dump can be happened before GPU reset. So a new argument is added to tell the dev core dump implementation whether to skip printing the vram_lost status in the dump. And this patch is also trying to decouple the core dump function from the GPU reset function, by replacing the argument amdgpu_reset_context with amdgpu_job to specify the context for core dump. V2: Inform user if VRAM lost check is skipped so users don't assume VRAM wasn't lost (Alex) Signed-off-by: Trigger Huang <Trigger.Huang@amd.com> Suggested-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-08-27Merge tag 'amd-drm-next-6.12-2024-08-26' of ↵Daniel Vetter
https://gitlab.freedesktop.org/agd5f/linux into drm-next amd-drm-next-6.12-2024-08-26: amdgpu: - SDMA devcoredump support - DCN 4.0.1 updates - DC SUBVP fixes - Refactor OPP in DC - Refactor MMHUBBUB in DC - DC DML 2.1 updates - DC FAMS2 updates - RAS updates - GFX12 updates - VCN 4.0.3 updates - JPEG 4.0.3 updates - Enable wave kill (soft recovery) for compute queues - Clean up CP error interrupt handling - Enable CP bad opcode interrupts - VCN 4.x fixes - VCN 5.x fixes - GPU reset fixes - Fix vbios embedded EDID size handling - SMU 14.x updates - Misc code cleanups and spelling fixes - VCN devcoredump support - ISP MFD i2c support - DC vblank fixes - GFX 12 fixes - PSR fixes - Convert vbios embedded EDID to drm_edid - DCN 3.5 updates - DMCUB updates - Cursor fixes - Overdrive support for SMU 14.x - GFX CP padding optimizations - DCC fixes - DSC fixes - Preliminary per queue reset infrastructure - Initial per queue reset support for GFX 9 - Initial per queue reset support for GFX 7, 8 - DCN 3.2 fixes - DP MST fixes - SR-IOV fixes - GFX 9.4.3/4 devcoredump support - Add process isolation framework - Enable process isolation support for GFX 9.4.3/4 - Take IOMMU remapping into account for P2P DMA checks amdkfd: - CRIU fixes - Improved input validation for user queues - HMM fix - Enable process isolation support for GFX 9.4.3/4 - Initial per queue reset support for GFX 9 - Allow users to target 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 UAPI: - KFD support for targetting queues on recommended SDMA engines Proposed userspace: https://github.com/ROCm/ROCR-Runtime/commit/2f588a24065f41c208c3701945e20be746d8faf7 https://github.com/ROCm/ROCR-Runtime/commit/eb30a5bbc7719c6ffcf2d2dd2878bc53a47b3f30 drm/buddy: - Add start address support for trim function From: Alex Deucher <alexander.deucher@amd.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240826201528.55307-1-alexander.deucher@amd.com
2024-08-23drm/amdgpu: Take IOMMU remapping into account for p2p checksRahul Jain
when trying to enable p2p the amdgpu_device_is_peer_accessible() checks the condition where address_mask overlaps the aper_base and hence returns 0, due to which the p2p disables for this platform IOMMU should remap the BAR addresses so the device can access them. Hence check if peer_adev is remapping DMA v5: (Felix, Alex) - fixing comment as per Alex feedback - refactor code as per Felix v4: (Alex) - fix the comment and description v3: - remove iommu_remap variable v2: (Alex) - Fix as per review comments - add new function amdgpu_device_check_iommu_remap to check if iommu remap Signed-off-by: Rahul Jain <Rahul.Jain@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-08-20drm/amdgpu: Implement Enforce Isolation Handler for KGD/KFD serializationSrinivasan Shanmugam
This commit introduces the Enforce Isolation Handler designed to enforce shader isolation on AMD GPUs, which helps to prevent data leakage between different processes. The handler 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`. This commit also introduces 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. Cc: Christian König <christian.koenig@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Suggested-by: Christian König <christian.koenig@amd.com> Suggested-by: Alex Deucher <alexander.deucher@amd.com>
2024-08-20drm/amdgpu: Add enforce_isolation sysfs attributeSrinivasan Shanmugam
This commit adds a new sysfs attribute 'enforce_isolation' to control the 'enforce_isolation' setting per GPU. The attribute can be read and written, and accepts values 0 (disabled) and 1 (enabled). When 'enforce_isolation' is enabled, reserved VMIDs are allocated for each ring. When it's disabled, the reserved VMIDs are freed. The set function locks a mutex before changing the 'enforce_isolation' flag and the VMIDs, and unlocks it afterwards. This ensures that these operations are atomic and prevents race conditions and other concurrency issues. Cc: Christian König <christian.koenig@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Suggested-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-08-16drm/amdgpu: Make enforce_isolation setting per GPUSrinivasan Shanmugam
This commit makes enforce_isolation setting to be per GPU and per partition by adding the enforce_isolation array to the adev structure. The adev variable is set based on the global enforce_isolation module parameter during device initialization. In amdgpu_ids.c, the adev->enforce_isolation value for the current GPU is used to determine whether to enforce isolation between graphics and compute processes on that GPU. In amdgpu_ids.c, the adev->enforce_isolation value for the current GPU and partition is used to determine whether to enforce isolation between graphics and compute processes on that GPU and partition. This allows the enforce_isolation setting to be controlled individually for each GPU and each partition, which is useful in a system with multiple GPUs and partitions where different isolation settings might be desired for different GPUs and partitions. v2: fix loop in amdgpu_vmid_mgr_init() (Alex) Cc: Christian König <christian.koenig@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Suggested-by: Christian König <christian.koenig@amd.com>
2024-08-16drm/amdgpu/gfx11: add a mutex for the gfx semaphoreAlex Deucher
This will be used in more places in the future so add a mutex. Acked-by: Vitaly Prosyak <vitaly.prosyak@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-08-13drm/amdgpu: Disable dpm_enabled flag while VF is in resetVictor Skvortsov
VFs do not perform HW fini/suspend in FLR, so the dpm_enabled is incorrectly kept enabled. Add interface to disable it in virt_pre_reset call. v2: Made implementation generic for all asics v3: Re-order conditionals so PP_MP1_STATE_FLR is only evaluated on VF Signed-off-by: Victor Skvortsov <victor.skvortsov@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-08-06drm/amdgpu: report bad status in GPU recoveryTao Zhou
Instead of printing GPU reset failed. v2: add check for reset_context->src. 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>
2024-08-06drm/amdgpu: update bad state check in GPU recoveryTao Zhou
Return RMA status without message print. 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>
2024-08-06drm/amdgpu: Clean up the register dump via debugfs listSunil Khatri
debugfs register list for dump is cleaned as it have some issues related to proper power state of the IP before register read. Since the above mentioned is removed we no longer want this to be dumped part of the devcoredump and hence removed. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Sunil Khatri <sunil.khatri@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-07-29Merge drm/drm-next into drm-misc-nextThomas Zimmermann
Backmerging to get a late RC of v6.10 before moving into v6.11. Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
2024-07-27drm/amdgpu: trigger ip dump before suspend of IP'sSunil Khatri
Problem: IP dump right now is done post suspend of all IP's which for some IP's could change power state and software state too which we do not want to reflect in the dump as it might not be same at the time of hang. Solution: IP should be dumped as close to the HW state when the GPU was in hung state without trying to reinitialize any resource. Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sunil Khatri <sunil.khatri@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-07-25drm/scheduler: remove full_recover from drm_sched_startChristian König
This was basically just another one of amdgpus hacks. The parameter allowed to restart the scheduler without turning fence signaling on again. That this is absolutely not a good idea should be obvious by now since the fences will then just sit there and never signal. While at it cleanup the code a bit. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Matthew Brost <matthew.brost@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240722083816.99685-1-christian.koenig@amd.com
2024-07-24drm/amdgpu: skip kfd init if GFX is not ready.Yifan Zhang
avoid kfd init crash in that case. Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Tested-by: Jesse Zhang <Jesse.Zhang@amd.com> Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-07-08drm/amdgpu: sysfs node disable query error count during gpu resetYiPeng Chai
Sysfs node disable query error count during gpu reset. Signed-off-by: YiPeng Chai <YiPeng.Chai@amd.com> Reviewed-by: Stanley.Yang <Stanley.Yang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-06-27drm/amdgpu: process RAS fatal error MB notificationVignesh Chander
For RAS error scenario, VF guest driver will check mailbox and set fed flag to avoid unnecessary HW accesses. additionally, poll for reset completion message first to avoid accidentally spamming multiple reset requests to host. v2: add another mailbox check for handling case where kfd detects timeout first v3: set host_flr bit and use wait_for_reset Signed-off-by: Vignesh Chander <Vignesh.Chander@amd.com> Reviewed-by: Zhigang Luo <Zhigang.Luo@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-06-27drm/amdgpu: Fix pci state save during mode-1 resetLijo Lazar
Cache the PCI state before bus master is disabled. The saved state is later used for other cases like restoring config space after mode-2 reset. Fixes: 5c03e5843e6b ("drm/amdgpu:add smu mode1/2 support for aldebaran") Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Feifei Xu <Feifei.Xu@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-06-19drm/amdgpu: fix using the reserved VMID with gang submitChristian König
We need to ensure that even when using a reserved VMID that the gang members can still run in parallel. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-06-14drm/amdgpu: create amdgpu_ras_in_recovery to simplify codeTao Zhou
Reduce redundant code and user doesn't need to pay attention to RAS details. 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>
2024-06-14drm/amdgpu: refine gpu_info firmware loadingYang Wang
refine gpu_info firmware loading Signed-off-by: Yang Wang <kevinyang.wang@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-06-14drm/amdgpu: fix sriov host flr handlerYunxiang Li
We send back the ready to reset message before we stop anything. This is wrong. Move it to when we are actually ready for the FLR to happen. In the current state since we take tens of seconds to stop everything, it is very likely that host would give up waiting and reset the GPU before we send ready, so it would be the same as before. But this gets rid of the hack with reset_domain locking and also let us tell how slow ready to reset actually is from the host. The ready to reset speed can be improved later. Signed-off-by: Yunxiang Li <Yunxiang.Li@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Reviewed-by: Emily Deng <Emily.Deng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-06-05drm/amdkfd: add reset cause in gpu pre-reset smi eventEric Huang
reset cause is requested by customer as additional info for gpu reset smi event. v2: integerate reset sources suggested by Lijo Lazar Signed-off-by: Eric Huang <jinhuieric.huang@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-05-29drm/amdgpu: Add lock around VF RLCG interfaceVictor Skvortsov
flush_gpu_tlb may be called from another thread while device_gpu_recover is running. Both of these threads access registers through the VF RLCG interface during VF Full Access. Add a lock around this interface to prevent race conditions between these threads. Signed-off-by: Victor Skvortsov <victor.skvortsov@amd.com> Reviewed-by: Zhigang Luo <zhigang.luo@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-05-29drm/amdgpu: Adjust logic in amdgpu_device_partner_bandwidth()Alex Deucher
Use current speed/width on devices which don't support dynamic PCIe switching. Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3289 Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-05-23drm/amdgpu: add prints in IP State dumpSunil Khatri
add prints before and after ip state is dumped. It avoids user to think of system being stuck/hung as dump could take some time after a gpu hang. Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sunil Khatri <sunil.khatri@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-05-23drm/amd/amdgpu: fix the inst passed to amdgpu_virt_rlcg_reg_rwVictor Zhao
the inst passed to amdgpu_virt_rlcg_reg_rw should be physical instance. Fix the miss matched code. Signed-off-by: Victor Zhao <Victor.Zhao@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-05-17drm/amdgpu: Check if NBIO funcs are NULL in amdgpu_device_baco_exitFriedrich Vock
The special case for VM passthrough doesn't check adev->nbio.funcs before dereferencing it. If GPUs that don't have an NBIO block are passed through, this leads to a NULL pointer dereference on startup. Signed-off-by: Friedrich Vock <friedrich.vock@gmx.de> Fixes: 1bece222eabe ("drm/amdgpu: Clear doorbell interrupt status for Sienna Cichlid") Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Christian König <christian.koenig@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-05-13drm/amdgpu: fix dereference after null checkJesse Zhang
check the pointer hive before use. Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com> Reviewed-by: Tim Huang <Tim.Huang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-05-02drm/amd/amdgpu: Check tbo resource pointerAsad Kamal
Validate tbo resource pointer, skip if NULL Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-05-02drm/amdgpu: Add gfx v9_4_4 ip blockHawking Zhang
Add gfx v9_4_4 ip block support Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Le Ma <le.ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-05-02drm/amdgpu: Move ras resume into SRIOV functionYunxiang Li
This is part of the reset, move it into the reset function. Signed-off-by: Yunxiang Li <Yunxiang.Li@amd.com> Reviewed-by: Emily Deng <Emily.Deng@amd.com> Reviewed-by: Zhigang Luo <zhigang.luo@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-05-02drm/amdgpu: Fix amdgpu_device_reset_sriov retry logicYunxiang Li
The retry loop for SRIOV reset have refcount and memory leak issue. Depending on which function call fails it can potentially call amdgpu_amdkfd_pre/post_reset different number of times and causes kfd_locked count to be wrong. This will block all future attempts at opening /dev/kfd. The retry loop also leakes resources by calling amdgpu_virt_init_data_exchange multiple times without calling the corresponding fini function. Align with the bare-metal reset path which doesn't have these issues. This means taking the amdgpu_amdkfd_pre/post_reset functions out of the reset loop and calling amdgpu_device_pre_asic_reset each retry which properly free the resources from previous try by calling amdgpu_virt_fini_data_exchange. Signed-off-by: Yunxiang Li <Yunxiang.Li@amd.com> Reviewed-by: Emily Deng <Emily.Deng@amd.com> Reviewed-by: Zhigang Luo <zhigang.luo@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-05-02drm/amdgpu: Add reset_context flag for host FLRYunxiang Li
There are other reset sources that pass NULL as the job pointer, such as amdgpu_amdkfd_reset_work. Therefore, using the job pointer to check if the FLR comes from the host does not work. Add a flag in reset_context to explicitly mark host triggered reset, and set this flag when we receive host reset notification. Signed-off-by: Yunxiang Li <Yunxiang.Li@amd.com> Reviewed-by: Emily Deng <Emily.Deng@amd.com> Reviewed-by: Zhigang Luo <zhigang.luo@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-05-02drm/amdgpu: Fix two reset triggered in a rowYunxiang Li
Some times a hang GPU causes multiple reset sources to schedule resets. The second source will be able to trigger an unnecessary reset if they schedule after we call amdgpu_device_stop_pending_resets. Move amdgpu_device_stop_pending_resets to after the reset is done. Since at this point the GPU is supposedly in a good state, any reset scheduled after this point would be a legitimate reset. Remove unnecessary and incorrect checks for amdgpu_in_reset that was kinda serving this purpose. Signed-off-by: Yunxiang Li <Yunxiang.Li@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-04-26drm/amdgpu: skip ip dump if devcoredump flag is setSunil Khatri
Do not dump the ip registers during driver reload in passthrough environment. Signed-off-by: Sunil Khatri <sunil.khatri@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-04-26drm/amdgpu: add a spinlock to wb allocationAlex Deucher
As we use wb slots more dynamically, we need to lock access to avoid racing on allocation or free. Reviewed-by: Shaoyun.liu <shaoyunl@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-04-26drm/amdgpu: dump ip state before reset for each ipSunil Khatri
Invoke the dump_ip_state function for each ip before the asic resets and save the register values for debugging via devcoredump. Signed-off-by: Sunil Khatri <sunil.khatri@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
generated by cgit v1.2.3 (git 2.39.1) at 2025-08-01 19:50:24 +0000