p71924506
/
linux_old1
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

mm: filemap_unaccount_folio() large skip mapcount fixup 

The page_mapcount_reset() when folio_mapped() while mapping_exiting() was
devised long before there were huge or compound pages in the cache.  It is
still valid for small pages, but not at all clear what's right to check
and reset on large pages.  Just don't try when folio_test_large().

Link: https://lkml.kernel.org/r/879c4426-4122-da9c-1a86-697f2c9a083@google.com
Signed-off-by: Hugh Dickins <hughd@google.com>
Cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This commit is contained in:
Hugh Dickins 2022-03-24 18:09:52 -07:00 committed by Linus Torvalds
parent bb43b14b57
commit 85207ad8ea
1 changed files with 13 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
mm/filemap.c
View File

@ -152,27 +152,27 @@ static void filemap_unaccount_folio(struct address_space *mapping,
VM_BUG_ON_FOLIO(folio_mapped(folio), folio); VM_BUG_ON_FOLIO(folio_mapped(folio), folio);
if (!IS_ENABLED(CONFIG_DEBUG_VM) && unlikely(folio_mapped(folio))) { if (!IS_ENABLED(CONFIG_DEBUG_VM) && unlikely(folio_mapped(folio))) {
int mapcount;
pr_alert("BUG: Bad page cache in process %s pfn:%05lx\n", pr_alert("BUG: Bad page cache in process %s pfn:%05lx\n",
current->comm, folio_pfn(folio)); current->comm, folio_pfn(folio));
dump_page(&folio->page, "still mapped when deleted"); dump_page(&folio->page, "still mapped when deleted");
dump_stack(); dump_stack();
add_taint(TAINT_BAD_PAGE, LOCKDEP_NOW_UNRELIABLE); add_taint(TAINT_BAD_PAGE, LOCKDEP_NOW_UNRELIABLE);
mapcount = page_mapcount(&folio->page); if (mapping_exiting(mapping) && !folio_test_large(folio)) {
if (mapping_exiting(mapping) && int mapcount = page_mapcount(&folio->page);
folio_ref_count(folio) >= mapcount + 2) {
if (folio_ref_count(folio) >= mapcount + 2) {
/* /*
* All vmas have already been torn down, so it's * All vmas have already been torn down, so it's
* a good bet that actually the folio is unmapped, * a good bet that actually the page is unmapped
* and we'd prefer not to leak it: if we're wrong, * and we'd rather not leak it: if we're wrong,
* some other bad page check should catch it later. * another bad page check should catch it later.
*/ */
page_mapcount_reset(&folio->page); page_mapcount_reset(&folio->page);
folio_ref_sub(folio, mapcount); folio_ref_sub(folio, mapcount);
} }
} }
}
/* hugetlb folios do not participate in page cache accounting. */ /* hugetlb folios do not participate in page cache accounting. */
if (folio_test_hugetlb(folio)) if (folio_test_hugetlb(folio))