kfree(adev->irq.client[i].sources); +adev->irq.client[i].sources = NULL; Set adev->irq.client[i].sources to NULL in amdgpu_irq_fini also can fix NULL ptr in amdgpu_irq_disable_all.. But I didn't check why amdgpu_device_fini was called twice. This patch looks good. Best Regards Rex -----Original Message----- From: amd-gfx [mailto:amd-gfx-bounces@xxxxxxxxxxxxxxxxxxxxx] On Behalf Of Andrey Grodzovsky Sent: Thursday, March 22, 2018 2:23 AM To: amd-gfx at lists.freedesktop.org Cc: Grodzovsky, Andrey Subject: [PATCH] drm/amdgpu: Fix NULL ptr on driver unload due to init failure. Problem: When unloading due to failure amdgpu_device_fini was called twice which was leading to NULL ptr in amdgpu_irq_disable_all. Fix: Call amdgpu_device_fini only once from amdgpu_driver_unload_kms. Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky at amd.com> --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 60e577c..c51be05 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -2023,7 +2023,6 @@ int amdgpu_device_init(struct amdgpu_device *adev, } dev_err(adev->dev, "amdgpu_device_ip_init failed\n"); amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_AMDGPU_INIT_FAIL, 0, 0); - amdgpu_device_ip_fini(adev); goto failed; } -- 2.7.4 _______________________________________________ amd-gfx mailing list amd-gfx at lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/amd-gfx
