From f4322b9f8ad5f9f62add288c785d2e10bb6a5efe Mon Sep 17 00:00:00 2001 From: Yunxiang Li Date: Mon, 22 Apr 2024 14:59:02 -0400 Subject: drm/amdgpu: Fix two reset triggered in a row 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 Reviewed-by: Lijo Lazar Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 861ccff78af9..8befd10bf007 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -5070,8 +5070,6 @@ static int amdgpu_device_reset_sriov(struct amdgpu_device *adev, retry: amdgpu_amdkfd_pre_reset(adev); - amdgpu_device_stop_pending_resets(adev); - if (from_hypervisor) r = amdgpu_virt_request_full_gpu(adev, true); else @@ -5823,13 +5821,6 @@ retry: /* Rest of adevs pre asic reset from XGMI hive. */ r, adev_to_drm(tmp_adev)->unique); tmp_adev->asic_reset_res = r; } - - if (!amdgpu_sriov_vf(tmp_adev)) - /* - * Drop all pending non scheduler resets. Scheduler resets - * were already dropped during drm_sched_stop - */ - amdgpu_device_stop_pending_resets(tmp_adev); } /* Actual ASIC resets if needed.*/ @@ -5851,6 +5842,16 @@ retry: /* Rest of adevs pre asic reset from XGMI hive. */ goto retry; } + list_for_each_entry(tmp_adev, device_list_handle, reset_list) { + /* + * Drop any pending non scheduler resets queued before reset is done. + * Any reset scheduled after this point would be valid. Scheduler resets + * were already dropped during drm_sched_stop and no new ones can come + * in before drm_sched_start. + */ + amdgpu_device_stop_pending_resets(tmp_adev); + } + skip_hw_reset: /* Post ASIC reset for all devs .*/ -- cgit