From: Ralph Campbell <rcampb...@nvidia.com>
[ Upstream commit ed710a6ed797430026aa5116dd0ab22378798b69 ]
If system memory is migrated to device private memory and no GPU MMU
page table entry exists, the GPU will fault and call hmm_range_fault()
to get the PFN for the page. Since the .dev_private_owner pointer in
struct hmm_range is not set, hmm_range_fault returns an error which
results in the GPU program stopping with a fatal fault.
Fix this by setting .dev_private_owner appropriately.
Fixes: 08ddddda667b ("mm/hmm: check the device private page owner in
hmm_range_fault()")
Cc: sta...@vger.kernel.org
Signed-off-by: Ralph Campbell <rcampb...@nvidia.com>
Reviewed-by: Jason Gunthorpe <j...@mellanox.com>
Signed-off-by: Ben Skeggs <bske...@redhat.com>
Signed-off-by: Sasha Levin <sas...@kernel.org>
---
drivers/gpu/drm/nouveau/nouveau_svm.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/gpu/drm/nouveau/nouveau_svm.c
b/drivers/gpu/drm/nouveau/nouveau_svm.c
index 645fedd77e21b..a9ce86740799f 100644
--- a/drivers/gpu/drm/nouveau/nouveau_svm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_svm.c
@@ -534,6 +534,7 @@ static int nouveau_range_fault(struct nouveau_svmm *svmm,
.flags = nouveau_svm_pfn_flags,
.values = nouveau_svm_pfn_values,
.pfn_shift = NVIF_VMM_PFNMAP_V0_ADDR_SHIFT,
+ .dev_private_owner = drm->dev,
};
struct mm_struct *mm = notifier->notifier.mm;
long ret;
--
2.25.1
