From mboxrd@z Thu Jan 1 00:00:00 1970 Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.remote-ip=100.103.45.18 ARC-Seal: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; cv=none; t=1794453470; b=LyEmf6Zmn+blR7uCTSWH3ZSRjfX4kLGRtIqFPW93kLLo5n7+GKp+CLatMryLF94lWtok j+omNGhBaRxbgcbGp+YQV8szSdb+m73Q+SvGluTUZ5airQkh1Qdj18WEbQzY99KpVruhF +dAM2q9vqsnXVSogKL0HJR0KPl8RU/sp+mV/nUXClzihZSH/aRxNAfzrWC+vEpZHUaGvn tJhg0f9Y5+qhzRca3hG0hw2zHlg5a1IqlY4c0wSSIr9xNkbmmannSWokb4xp23uGPyJX9 Vd86P0nsFY5AqW0RH6N6djAcAOQ+g5vySDVUmY0VTh7BWSjBd/jafh3WjxFnV4cmLhw== ARC-Message-Signature: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; c=relaxed/relaxed; t=1794453470; h=Received:DKIM-Signature:From:To:Cc:Subject:Date:Message-ID:X-Mailer: In-Reply-To:References:MIME-Version:Content-Transfer-Encoding; bh=On2c35btmNhmiH0etPL6d6aMVFYONbECDtu+DNw/Lvc=; b=CLS5PXpfkRqHXmXfSIEtuFcJRhDeR05CohnMjGYVGWeR1xdmtSjY/RomSKUTlBkRJ4kI krVBZvnzkqdz4evO6Lb3qK5z906H5FJkpQrXr6NYJEXz1WAX5axCMkywyavdTOpiG4Kbl ZFtqfZmfOYBez6a/vQNTyrr6qvbTdlCqYKfdmd/RJCNsUM+1AAS4bHKL3D+RMkDfPu/RU 2oj73ZLoDQtYUzUeETkmG8wpkLuE5eN0T22EiicuKaNxSURcVSJFVUTRPW7CiqkPEGFLv 7sVItfwYDJSHKLs2oS/t9ZVvEfjxJcZnwJxpbyOQRC8Mvu9D/GGmRY5zPxEts8r6/Yw== ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass header.d=kernel.org header.i=@kernel.org header.a=rsa-sha256 header.s=k20260515 header.b=FkgRUAFf; arc=none smtp.remote-ip=100.103.45.18 Received: from smtp.kernel.org (quasi.space.kernel.org [100.103.45.18]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (prime256v1)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id EBCFC606A321 for ; Thu, 12 Nov 2026 03:17:49 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.a=rsa-sha256 header.s=k20260515 header.b=FkgRUAFf Received: by smtp.kernel.org (Postfix) with ESMTPSA id 06E7C1F00A06; Thu, 12 Nov 2026 03:17:48 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1794453469; bh=On2c35btmNhmiH0etPL6d6aMVFYONbECDtu+DNw/Lvc=; h=From:To:Cc:Subject:Date:In-Reply-To:References; b=FkgRUAFfPHyluaymmxs84LgvWdc31X7cbCikooexE6Wzum3CgExpfr6ZXGsy09Xij y6yKWG/3DlxNpRzQ3CXiMZNJlbwdCiJ+uOQQun8hBKOFusXkcwAA/1oRwyZ8whl1x9 jVocjgfDaS+hCY398oyljzB5WhF6xGbrL47iXyBo9tA9tuwaTJasZK6OaG2tQPnmMd sy76iBIEwDxU0Slqaodb2cXbJBdh1/YpZNhNfxKddgIguVkA74MEhTBNpH1TwUvQZl vi8ztax2tY4OvAT4ZmAopstY/hb9eY2s5IldCLb/uvweIHj+WBFYoNmUH5uxzmbSY/ 1hDEoulQSXCOA== From: Sasha Levin To: stable@vger.kernel.org Cc: Prike Liang , Lijo Lazar , Alex Deucher , Sasha Levin Subject: [PATCH 6.6.y] drm/amdgpu: finish GPU recovery callbacks before restarting schedulers Date: Wed, 11 Nov 2026 22:17:47 -0500 Message-ID: <20261112031732.1416232-1-sashal@kernel.org> X-Mailer: git-send-email 2.53.0 In-Reply-To: <20261109201417-failed-6.6-733389cb31e2@kernel.org> References: <20261109201417-failed-6.6-733389cb31e2@kernel.org> X-Mailing-List: stable@vger.kernel.org List-Id: List-Post: List-Help: List-Subscribe: List-Unsubscribe: List-Owner: Precedence: list MIME-Version: 1.0 Content-Transfer-Encoding: 8bit From: Prike Liang [ Upstream commit 733389cb31e2c55e355410759fdc68afe74a4582 ] Run GPU post-reset callbacks, including KFD and display-audio resume, before restarting DRM scheduler workqueues. IP hardware initialization already happens inside ASIC recovery; this change orders the remaining callbacks before new scheduler submissions. Preserve the emergency-stop and guity job completition path's callbacks without starting schedulers, Signed-off-by: Prike Liang Acked-by: Lijo Lazar Signed-off-by: Alex Deucher (cherry picked from commit 1f6433e2b264e75c56ae95f187788806201bdbc3) Cc: stable@vger.kernel.org [ sashal: Adapted helper reordering to inline code in amdgpu_device_gpu_recover() because the upstream recovery helpers are absent. ] Signed-off-by: Sasha Levin --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 47 +++++++++++----------- 1 file changed, 24 insertions(+), 23 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 2cdbee6700ecd..41fac9d273622 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -5511,7 +5511,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, } if (need_emergency_restart) - goto skip_sched_resume; + goto post_resume; /* * Must check guilty signal here since after this point all old @@ -5522,7 +5522,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, if (job && dma_fence_is_signaled(&job->hw_fence.base)) { job_signaled = true; dev_info(adev->dev, "Guilty job already signaled, skipping HW reset"); - goto skip_hw_reset; + goto post_resume; } retry: /* Rest of adevs pre asic reset from XGMI hive. */ @@ -5567,7 +5567,28 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, goto recover_end; } -skip_hw_reset: +post_resume: + list_for_each_entry(tmp_adev, device_list_handle, reset_list) { + /* unlock kfd: SRIOV would do it separately */ + if (!need_emergency_restart && !amdgpu_sriov_vf(tmp_adev)) + amdgpu_amdkfd_post_reset(tmp_adev); + + /* kfd_post_reset will do nothing if kfd device is not initialized, + * need to bring up kfd here if it's not be initialized before + */ + if (!adev->kfd.init_complete) + amdgpu_amdkfd_device_init(adev); + + if (audio_suspended) + amdgpu_device_resume_display_audio(tmp_adev); + + amdgpu_device_unset_mp1_state(tmp_adev); + + amdgpu_ras_set_error_query_ready(tmp_adev, true); + } + + if (need_emergency_restart) + goto recover_end; /* Post ASIC reset for all devs .*/ list_for_each_entry(tmp_adev, device_list_handle, reset_list) { @@ -5603,26 +5624,6 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, } } -skip_sched_resume: - list_for_each_entry(tmp_adev, device_list_handle, reset_list) { - /* unlock kfd: SRIOV would do it separately */ - if (!need_emergency_restart && !amdgpu_sriov_vf(tmp_adev)) - amdgpu_amdkfd_post_reset(tmp_adev); - - /* kfd_post_reset will do nothing if kfd device is not initialized, - * need to bring up kfd here if it's not be initialized before - */ - if (!adev->kfd.init_complete) - amdgpu_amdkfd_device_init(adev); - - if (audio_suspended) - amdgpu_device_resume_display_audio(tmp_adev); - - amdgpu_device_unset_mp1_state(tmp_adev); - - amdgpu_ras_set_error_query_ready(tmp_adev, true); - } - recover_end: tmp_adev = list_first_entry(device_list_handle, struct amdgpu_device, reset_list); -- 2.53.0