From: Pavel Tatashin <pasha.tatashin@xxxxxxxxxx> [ Upstream commit 83c02c23d0747a7bdcd71f99a538aacec94b146c ] When pages are isolated in check_and_migrate_movable_pages() we skip compound number of pages at a time. However, as Jason noted, it is not necessary correct that pages[i] corresponds to the pages that we skipped. This is because it is possible that the addresses in this range had split_huge_pmd()/split_huge_pud(), and these functions do not update the compound page metadata. The problem can be reproduced if something like this occurs: 1. User faulted huge pages. 2. split_huge_pmd() was called for some reason 3. User has unmapped some sub-pages in the range 4. User tries to longterm pin the addresses. The resulting pages[i] might end-up having pages which are not compound size page aligned. Link: https://lkml.kernel.org/r/20210215161349.246722-3-pasha.tatashin@xxxxxxxxxx Fixes: aa712399c1e8 ("mm/gup: speed up check_and_migrate_cma_pages() on huge page") Signed-off-by: Pavel Tatashin <pasha.tatashin@xxxxxxxxxx> Reported-by: Jason Gunthorpe <jgg@xxxxxxxxxx> Reviewed-by: Jason Gunthorpe <jgg@xxxxxxxxxx> Cc: Dan Williams <dan.j.williams@xxxxxxxxx> Cc: David Hildenbrand <david@xxxxxxxxxx> Cc: David Rientjes <rientjes@xxxxxxxxxx> Cc: Ingo Molnar <mingo@xxxxxxxxxx> Cc: Ira Weiny <ira.weiny@xxxxxxxxx> Cc: James Morris <jmorris@xxxxxxxxx> Cc: Jason Gunthorpe <jgg@xxxxxxxx> Cc: John Hubbard <jhubbard@xxxxxxxxxx> Cc: Joonsoo Kim <iamjoonsoo.kim@xxxxxxx> Cc: Matthew Wilcox <willy@xxxxxxxxxxxxx> Cc: Mel Gorman <mgorman@xxxxxxx> Cc: Michal Hocko <mhocko@xxxxxxxxxx> Cc: Michal Hocko <mhocko@xxxxxxxx> Cc: Mike Kravetz <mike.kravetz@xxxxxxxxxx> Cc: Oscar Salvador <osalvador@xxxxxxx> Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx> Cc: Sasha Levin <sashal@xxxxxxxxxx> Cc: Steven Rostedt (VMware) <rostedt@xxxxxxxxxxx> Cc: Tyler Hicks <tyhicks@xxxxxxxxxxxxxxxxxxx> Cc: Vlastimil Babka <vbabka@xxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> Signed-off-by: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx> --- mm/gup.c | 19 +++++++------------ 1 file changed, 7 insertions(+), 12 deletions(-) diff --git a/mm/gup.c b/mm/gup.c index ef7d2da9f03f..6530672131db 100644 --- a/mm/gup.c +++ b/mm/gup.c @@ -1552,26 +1552,23 @@ static long check_and_migrate_cma_pages(struct mm_struct *mm, unsigned int gup_flags) { unsigned long i; - unsigned long step; bool drain_allow = true; bool migrate_allow = true; LIST_HEAD(cma_page_list); long ret = nr_pages; + struct page *prev_head, *head; struct migration_target_control mtc = { .nid = NUMA_NO_NODE, .gfp_mask = GFP_USER | __GFP_MOVABLE | __GFP_NOWARN, }; check_again: - for (i = 0; i < nr_pages;) { - - struct page *head = compound_head(pages[i]); - - /* - * gup may start from a tail page. Advance step by the left - * part. - */ - step = compound_nr(head) - (pages[i] - head); + prev_head = NULL; + for (i = 0; i < nr_pages; i++) { + head = compound_head(pages[i]); + if (head == prev_head) + continue; + prev_head = head; /* * If we get a page from the CMA zone, since we are going to * be pinning these entries, we might as well move them out @@ -1595,8 +1592,6 @@ check_again: } } } - - i += step; } if (!list_empty(&cma_page_list)) { -- 2.30.2