The unlikely predates an era where we were checking for
hwpoisoned/migration entries prior to checking whether the pte was
present.
Currently, we check for the pte to be a migration/hwpoison entry after we
have checked that is not present, so it must be either one or the other.
Link: https://lkml.kernel.org/r/20250627102904.107202-6-osalvador@suse.de
Signed-off-by: Oscar Salvador <osalvador@suse.de>
Cc: David Hildenbrand <david@redhat.com>
Cc: Gavin Guo <gavinguo@igalia.com>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Peter Xu <peterx@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
mm/hugetlb.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index ba078aa1cb96..fda6b748e985 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -6747,7 +6747,7 @@ vm_fault_t hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma,
/* Not present, either a migration or a hwpoisoned entry */
if (!pte_present(vmf.orig_pte)) {
- if (unlikely(is_hugetlb_entry_migration(vmf.orig_pte))) {
+ if (is_hugetlb_entry_migration(vmf.orig_pte)) {
/*
* Release the hugetlb fault lock now, but retain
* the vma lock, because it is needed to guard the
@@ -6758,7 +6758,7 @@ vm_fault_t hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma,
mutex_unlock(&hugetlb_fault_mutex_table[hash]);
migration_entry_wait_huge(vma, vmf.address, vmf.pte);
return 0;
- } else if (unlikely(is_hugetlb_entry_hwpoisoned(vmf.orig_pte)))
+ } else if (is_hugetlb_entry_hwpoisoned(vmf.orig_pte))
ret = VM_FAULT_HWPOISON_LARGE |
VM_FAULT_SET_HINDEX(hstate_index(h));
goto out_mutex;
--
2.50.0