| Message ID | 20240301214712.2853147-4-willy@infradead.org (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | Remove some races around folio_test_hugetlb | expand |
diff --git a/mm/memory-failure.c b/mm/memory-failure.c index 9349948f1abf..e42d0cd7888d 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -1987,12 +1987,12 @@ void folio_clear_hugetlb_hwpoison(struct folio *folio) int __get_huge_page_for_hwpoison(unsigned long pfn, int flags, bool *migratable_cleared) { - struct page *page = pfn_to_page(pfn); - struct folio *folio = page_folio(page); + struct folio *folio; int ret = 2; /* fallback to normal page handling */ bool count_increased = false; - if (!folio_test_hugetlb(folio)) + folio = hwpoison_pfn_folio(pfn); + if (!folio) goto out; if (flags & MF_COUNT_INCREASED) {
Calling folio_test_hugetlb() is inefficient; hugetlb_pfn_folio() does less work. This is a safe use of hugetlb_pfn_folio() as __get_huge_page_for_hwpoison() holds the hugetlb_lock. Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org> --- mm/memory-failure.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)