On 2025-10-24 13:26, Sunil Khatri wrote:
we need to unreserve the bo's too during clean up along
with freeing the memory of context.
Fixes: c549912cd8 (drm/amdkfd: add missing return value check for range)
Signed-off-by: Sunil Khatri<[email protected]>
---
drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 7f0ab73e2396..76e36201742a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1740,7 +1740,7 @@ static int svm_range_validate_and_map(struct mm_struct
*mm,
range = amdgpu_hmm_range_alloc(NULL);
if (unlikely(!range)) {
r = -ENOMEM;
- goto free_ctx;
+ break;
Sorry, looking at the latest code, it's even more complicated because of
p->svms.faulting_task, which also needs to be reset.
WRITE_ONCE(p->svms.faulting_task, current);
range = amdgpu_hmm_range_alloc(NULL);
if (likely(range))
r =
amdgpu_hmm_range_get_pages(&prange->notifier, addr,
npages,
readonly, owner,
range);
else
r = -ENOMEM;
WRITE_ONCE(p->svms.faulting_task, NULL);
if (r) {
amdgpu_hmm_range_free(range);
pr_debug("failed %d to get svm range pages\n",
r);
}
Regards,
Felix
}
r = amdgpu_hmm_range_get_pages(&prange->notifier, addr, npages,
