On Fri, Aug 26, 2022 at 9:56 AM Minchan Kim <minchan@xxxxxxxxxx> wrote: > > On Fri, Aug 26, 2022 at 11:20:58AM +0800, Yin, Fengwei wrote: > > > > > > On 8/26/2022 2:46 AM, Matthew Wilcox wrote: > > >>> Looks like my analysis from yesterday was dropped: > > >>> > > >>> : This all seems quite plausible. The reproducer seems to (correct me > > >>> : if I'm wrong) create an AF_PACKET socket and mmap it. af_packet.c > > >>> : seems to create compound pages and mmap them. This isn't folio-related > > >>> : at all; I just moved the code that warns about it from mm/vmscan.c to > > >>> : folio-compat.c. > > >>> : > > >>> : Looks like a long-standing bug in MADV_PAGEOUT to me. > > >> Such page should never be on lru, right? We could test lru before > > >> calling isolate_lru_page() for this case? I know isolate_lru_page() > > >> does the check, but the tail page warning is raised before the check. > > >> > > >> Could the tail page warning be moved under the lru flag test? Seems > > >> possible, but it should need extra handling (re-set lru flag). Seems a > > >> little bit overkilling. > > > There's a number of ways of solving this. I'm interested in seeing > > > which one Minchan thinks is best. > > > > > > > My understanding is: > > PageTransCompound() return false for compound page if THP is disabled > > in kernel config. Replacing PageTransCompound() with PageCompound() > > could work here. But for the long term, folio should be the answer. :). > > Thanks for reporting and analysis, folks, > > I agree with Yang since the MADV_PAGEOUT should work with only > LRU pages. > > From 0a43ac31c903bc23299a868a6d6724ff5b807e3d Mon Sep 17 00:00:00 2001 > From: Minchan Kim <minchan@xxxxxxxxxx> > Date: Fri, 26 Aug 2022 09:37:34 -0700 > Subject: [PATCH] mm: fix madivse_pageout mishandling on non-LRU page > MIME-Version: 1.0 > Content-Type: text/plain; charset=UTF-8 > Content-Transfer-Encoding: 8bit > > MADV_PAGEOUT tries to isolate non-LRU pages and get the warning > from isolate_lru_page below. > Fix it with checking PageLRU in advance. > > ------------[ cut here ]------------ > trying to isolate tail page > WARNING: CPU: 0 PID: 6175 at mm/folio-compat.c:158 isolate_lru_page+0x130/0x140 > Modules linked in: > CPU: 0 PID: 6175 Comm: syz-executor.0 Not tainted 5.18.12 #1 > Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.13.0-1ubuntu1.1 04/01/2014 > RIP: 0010:isolate_lru_page+0x130/0x140 > > Link: https://lore.kernel.org/linux-mm/485f8c33.2471b.182d5726afb.Coremail.hantianshuo@xxxxxxxxx/ > Reported-by: 韩天硕 <hantianshuo@xxxxxxxxx> > Suggested-by: Yang Shi <shy828301@xxxxxxxxx> > Fixes: 1a4e58cce84e ("mm: introduce MADV_PAGEOUT") > Cc: stable@xxxxxxxxxxxxxxx > Signed-off-by: Minchan Kim <minchan@xxxxxxxxxx> Thanks for the patch, looks good to me. Will you post it to the mailing list? Anyway you could have my ack. > --- > mm/madvise.c | 7 +++++-- > 1 file changed, 5 insertions(+), 2 deletions(-) > > diff --git a/mm/madvise.c b/mm/madvise.c > index 682e1d161aef..a3fc4cd32ed3 100644 > --- a/mm/madvise.c > +++ b/mm/madvise.c > @@ -452,8 +452,11 @@ static int madvise_cold_or_pageout_pte_range(pmd_t *pmd, > continue; > } > > - /* Do not interfere with other mappings of this page */ > - if (page_mapcount(page) != 1) > + /* > + * Do not interfere with other mappings of this page and > + * non-LRU page. > + */ > + if (!PageLRU(page) || page_mapcount(page) != 1) > continue; > > VM_BUG_ON_PAGE(PageTransCompound(page), page); > -- > 2.37.2.672.g94769d06f0-goog >
