drm/amdgpu: correct the gpu reset handling for job != NULL case

commit 207ac68 upstream.

Current code wrongly treat all cases as job == NULL.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-and-tested-by: Jane Jian <Jane.Jian@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/gpu/drm/amd/amdgpu/amdgpu_device.c

@@ -3890,7 +3890,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
 
 		amdgpu_device_lock_adev(tmp_adev, false);
 		r = amdgpu_device_pre_asic_reset(tmp_adev,
-						 NULL,
+						 (tmp_adev == adev) ? job : NULL,
 						 &need_full_reset);
 		/*TODO Should we stop ?*/
 		if (r) {