On Mon, Apr 03, 2017 at 03:11:11PM -0700, Andrew Morton wrote: > On Fri, 24 Mar 2017 20:48:50 +0300 Alexander Polakov <apolyakov@xxxxxxxx> wrote: > > > Better seen in context: https://github.com/torvalds/linux/blob/master/m > > m/page_alloc.c#L4500 > > > > Signed-off-by: Alexander Polyakov <apolyakov@xxxxxxxxx> > > --- a/mm/page_alloc.c~fix-print-order-in-show_free_areas > +++ a/mm/page_alloc.c > @@ -4519,13 +4519,13 @@ void show_free_areas(unsigned int filter > K(node_page_state(pgdat, NR_FILE_MAPPED)), > K(node_page_state(pgdat, NR_FILE_DIRTY)), > K(node_page_state(pgdat, NR_WRITEBACK)), > + K(node_page_state(pgdat, NR_SHMEM)), > #ifdef CONFIG_TRANSPARENT_HUGEPAGE > K(node_page_state(pgdat, NR_SHMEM_THPS) * HPAGE_PMD_NR), > K(node_page_state(pgdat, NR_SHMEM_PMDMAPPED) > * HPAGE_PMD_NR), > K(node_page_state(pgdat, NR_ANON_THPS) * HPAGE_PMD_NR), > #endif > - K(node_page_state(pgdat, NR_SHMEM)), > K(node_page_state(pgdat, NR_WRITEBACK_TEMP)), > K(node_page_state(pgdat, NR_UNSTABLE_NFS)), > node_page_state(pgdat, NR_PAGES_SCANNED), > _ > > huh. It looks like this has been broken for nearly a year, by > > : commit 11fb998986a72aa7e997d96d63d52582a01228c5 > : Author: Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx> > : AuthorDate: Thu Jul 28 15:46:20 2016 -0700 > : Commit: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> > : CommitDate: Thu Jul 28 16:07:41 2016 -0700 > : > : mm: move most file-based accounting to the node > Yes, this was careless. Thanks for catching it Alexander. > I'm surprised nobody noticed until now. > Probably because vmstat was not affected which is consumed more often than the output from sysrq or an oom kill message. -- Mel Gorman SUSE Labs -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>
