Commit cf8b92a7 authored by Roman Li's avatar Roman Li Committed by Alex Deucher
Browse files

drm/amd/display: fix potential gpu reset deadlock 



[Why]
In gpu reset dc_lock acquired in dm_suspend().
Asynchronously handle_hpd_rx_irq can also be called
through amdgpu_dm_irq_suspend->flush_work, which also
tries to acquire dc_lock. That causes a deadlock.

[How]
Check if amdgpu executing reset before acquiring dc_lock.

Signed-off-by: default avatarLang Yu <Lang.Yu@amd.com>
Signed-off-by: default avatarRoman Li <Roman.Li@amd.com>
Reviewed-by: default avatarQingqing Zhuo <Qingqing.Zhuo@amd.com>
Acked-by: default avatarWayne Lin <Wayne.Lin@amd.com>
Tested-by: default avatarDaniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: default avatarAlex Deucher <alexander.deucher@amd.com>
parent 2e4ec251

drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c

+4 −2
Original line number Diff line number Diff line
@@ -2726,12 +2726,14 @@ static void handle_hpd_rx_irq(void *param) 
		}

	}



	if (!amdgpu_in_reset(adev))

		mutex_lock(&adev->dm.dc_lock);

#ifdef CONFIG_DRM_AMD_DC_HDCP

	result = dc_link_handle_hpd_rx_irq(dc_link, &hpd_irq_data, NULL);

#else

	result = dc_link_handle_hpd_rx_irq(dc_link, NULL, NULL);

#endif

	if (!amdgpu_in_reset(adev))

		mutex_unlock(&adev->dm.dc_lock);



out: