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=1794455786; b=NRqJGOnVF3hsQBkmci/nbv5qiYDsUUqobxrgv0Az/VgxbDWGkzmtFhkXCCKGZFf0aqcy kt3ueFPo/7MU9LFts2gdzmpQlGyW1coKcrf6cJ9HDFfpVhEgP8tVdFHoiHsPqyX5lI7Rk HwAK7Qi05bnieeBYN12fa+weLFyvJfd5KTOC0aBCvv0BTpLBfB71OlzJHh7up0RcfcdZv 4B/1I45fZGbNpreWZ3UYjKp1EXvaiAEuENBKWhC2jTHeyjTE1Jh+cCO1bz1yK/VoDWZ5r ezlS0GHst6E+Lf76c+ZKUdSvn3SoClHqH+IYvHgCjSzuIh9Upd26I2oYG6na76AEjyA== ARC-Message-Signature: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; c=relaxed/relaxed; t=1794455786; h=Received:DKIM-Signature:From:To:Cc:Subject:Date:Message-ID:X-Mailer: In-Reply-To:References:MIME-Version:Content-Transfer-Encoding; bh=EAhXseB6H0YcI2lUHSQWeRc1GnlOIDH/HLV7tFJX42k=; b=qadYePuZoU/QYhFA+rBd9bOvKxhGMEeyU8kdA3sMEi3gYIjbpZjVUBJtJYnJ+AOXL+Ix U0gaq643w0kMg6YDsVHW54lBKZm78DK/q+vrjh9kimjKcz/YaxLZ1WJst0feiZolljnd2 NwZjD9gruubkRXzmuLYAvZUWaTikXLpZ9N9Ol/zeS+YxiAu2JuGzHCM9wg/O7ypH+Ofbf P4RXK/Cb2g+YVYVm0WdJNOUoqs1CSIgPPBIKd3cWeda4Jau5hDBRGu62RUvB2Rd6Qr+mK 6qHCMKdNUo+E+x4EwHvk/UFIaTuY4+pXXWhEQLd3W8l+qtMxBVxCqnZ9drS+8iuGZ3A== 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=IEF6wWxb; 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 33126606A31C for ; Thu, 12 Nov 2026 03:56:25 +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=IEF6wWxb Received: by smtp.kernel.org (Postfix) with ESMTPSA id 33C831F00A05; Thu, 12 Nov 2026 03:56:25 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1794455785; bh=EAhXseB6H0YcI2lUHSQWeRc1GnlOIDH/HLV7tFJX42k=; h=From:To:Cc:Subject:Date:In-Reply-To:References; b=IEF6wWxblxzVRmyFplXnQMX53TIG9lbpRQiEFE88U9ckEErpp8VIQwExXdRPZhPto 2XkM/LdPS/vdHPg+f/ak+UqBjOJV279MaYiGefDWz101rVJ0h5sJcWNOGYSQeRM+3W MZQkBUa8EBrda5XeaG8fEmTZWwmYH8mVGYr/r5UVK1gOEEnYql5XMb+2frAJcPVD1Z FgUWGpfbkAKV9uYqWR6c3L68kwlRJ3q351LEDwroDzGd2rqyZicuEJs1SM8y4IYl27 evMmYvgpi+v7+JDwk4Fb5/i5f+mLVzRZAcpsku0Yc4s3aKx8od7dceFdUw2P24Geru Ho6BeuaA5XdOg== From: Sasha Levin To: stable@vger.kernel.org Cc: Prike Liang , Lijo Lazar , Alex Deucher , Sasha Levin Subject: [PATCH 6.1.y] drm/amdgpu: finish GPU recovery callbacks before restarting schedulers Date: Wed, 11 Nov 2026 22:56:23 -0500 Message-ID: <20261112035607.1542041-1-sashal@kernel.org> X-Mailer: git-send-email 2.53.0 In-Reply-To: <20261109201906-failed-6.1-733389cb31e2@kernel.org> References: <20261109201906-failed-6.1-733389cb31e2@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 upstream recovery helper reordering to the existing inline recovery code. ] Signed-off-by: Sasha Levin --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 43 ++++++++++++---------- 1 file changed, 24 insertions(+), 19 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 8bd9ecb88c94a..3cd3fb1a5eb4b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -5621,7 +5621,7 @@ 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 @@ -5693,7 +5693,30 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, !(tmp_vram_lost_counter < atomic_read(&adev->vram_lost_counter))) amdgpu_device_recheck_guilty_jobs( tmp_adev, device_list_handle, reset_context); + } + +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); + } + + if (need_emergency_restart) + goto recover_end; + list_for_each_entry(tmp_adev, device_list_handle, reset_list) { for (i = 0; i < AMDGPU_MAX_RINGS; ++i) { struct amdgpu_ring *ring = tmp_adev->rings[i]; @@ -5730,24 +5753,6 @@ 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); - } - recover_end: tmp_adev = list_first_entry(device_list_handle, struct amdgpu_device, reset_list); -- 2.53.0