On Wed 12-05-21 00:10:15, Naoya Horiguchi wrote:
> From: Naoya Horiguchi <naoya.horiguchi@xxxxxxx>
>
> When hugetlb page fault (under overcommiting situation) and
> memory_failure() race, VM_BUG_ON_PAGE() is triggered by the following race:
>
> CPU0: CPU1:
>
> gather_surplus_pages()
> page = alloc_surplus_huge_page()
> memory_failure_hugetlb()
> get_hwpoison_page(page)
> __get_hwpoison_page(page)
> get_page_unless_zero(page)
> zero = put_page_testzero(page)
> VM_BUG_ON_PAGE(!zero, page)
> enqueue_huge_page(h, page)
> put_page(page)
>
> __get_hwpoison_page() only checks page refcount before taking additional
> one for memory error handling, which is wrong because there's time
> windows where compound pages have non-zero refcount during initialization.
>
> So makes __get_hwpoison_page() check page status a bit more for a few
> types of compound pages. PageSlab() check is added because otherwise
> "non anonymous thp" path is wrongly chosen.
This should really describe the fix in more details. E.g.
[...]
> @@ -1095,30 +1095,43 @@ static int __get_hwpoison_page(struct page *page)
> {
> struct page *head = compound_head(page);
>
> - if (!PageHuge(head) && PageTransHuge(head)) {
> - /*
> - * Non anonymous thp exists only in allocation/free time. We
> - * can't handle such a case correctly, so let's give it up.
> - * This should be better than triggering BUG_ON when kernel
> - * tries to touch the "partially handled" page.
> - */
> - if (!PageAnon(head)) {
> - pr_err("Memory failure: %#lx: non anonymous thp\n",
> - page_to_pfn(page));
> - return 0;
> + if (PageCompound(page)) {
So you do rely on PageCompound to be true. Which is prone to races as
well. All you need is to check before prep_compound_page and run into
get_page_unless_zero (down in this function) before hugetlb reaches
put_page_testzero. Or do I miss something that would prevent from that?
--
Michal Hocko
SUSE Labs
