drm/xe/pf: Check for fence error on VRAM save/restore

The code incorrectly assumes that the VRAM save/restore fence is valid.
Fix it by checking for error.

Fixes: 49cf1b9 ("drm/xe/pf: Handle VRAM migration data as part of PF control")
Suggested-by: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://patch.msgid.link/20251114122339.1791026-1-michal.winiarski@intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
(cherry picked from commit 78ff838)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>

Author: mwiniars

drivers/gpu/drm/xe/xe_gt_sriov_pf_migration.c

@@ -571,6 +571,10 @@ static int pf_save_vram_chunk(struct xe_gt *gt, unsigned int vfid,
 	fence = __pf_save_restore_vram(gt, vfid,
 				       src_vram, src_vram_offset,
 				       data->bo, 0, size, true);
+	if (IS_ERR(fence)) {
+		ret = PTR_ERR(fence);
+		goto fail;
+	}
 
 	ret = dma_fence_wait_timeout(fence, false, PF_VRAM_SAVE_RESTORE_TIMEOUT);
 	dma_fence_put(fence);
@@ -654,6 +658,11 @@ static int pf_restore_vf_vram_mig_data(struct xe_gt *gt, unsigned int vfid,
 
 	fence = __pf_save_restore_vram(gt, vfid, vram, data->hdr.offset,
 				       data->bo, 0, data->hdr.size, false);
+	if (IS_ERR(fence)) {
+		ret = PTR_ERR(fence);
+		goto err;
+	}
+
 	ret = dma_fence_wait_timeout(fence, false, PF_VRAM_SAVE_RESTORE_TIMEOUT);
 	dma_fence_put(fence);
 	if (!ret) {