From: Michal Hocko <mhocko@kernel.org> To: Minchan Kim <minchan@kernel.org> Cc: Andrew Morton <akpm@linux-foundation.org>, "Kirill A. Shutemov" <kirill@shutemov.name>, Johannes Weiner <hannes@cmpxchg.org>, linux-mm@kvack.org, linux-kernel@vger.kernel.org Subject: Re: memcg uncharge page counter mismatch Date: Thu, 3 Dec 2015 09:54:52 +0100 [thread overview] Message-ID: <20151203085451.GC9264@dhcp22.suse.cz> (raw) In-Reply-To: <20151203021006.GA31041@bbox> On Thu 03-12-15 11:10:06, Minchan Kim wrote: > On Thu, Dec 03, 2015 at 10:34:04AM +0900, Minchan Kim wrote: > > On Wed, Dec 02, 2015 at 11:16:43AM +0100, Michal Hocko wrote: > > > On Tue 01-12-15 22:34:55, Minchan Kim wrote: > > > > With new test on mmotm-2015-11-25-17-08, I saw below WARNING message > > > > several times. I couldn't see it with reverting new THP refcount > > > > redesign. > > > > > > Just a wild guess. What prevents migration/compaction from calling > > > split_huge_page on thp zero page? There is VM_BUG_ON but it is not clear > > > > I guess migration should work with LRU pages now but zero page couldn't > > stay there. Ahh, you are right. I have missed PageLRU check in isolate_migratepages_block pfn walker. > > > whether you run with CONFIG_DEBUG_VM enabled. > > > > I enabled VM_DEBUG_VM. > > > > > > > > Also, how big is the underflow? [...] > > nr_pages 293 new -324 > > nr_pages 16 new -340 > > nr_pages 342 new -91 > > nr_pages 246 new -337 > > nr_pages 15 new -352 > > nr_pages 15 new -367 They are quite large but that is not that surprising if we consider that we are batching many uncharges at once. > My guess is that it's related to new feature of Kirill's THP 'PageDoubleMap' > so a THP page could be mapped a pte but !pmd_trans_huge(*pmd) so memcg > precharge in move_charge should handle it? I am not familiar with the current state of THP after the rework unfortunately. So if I got you right then you are saying that pmd_trans_huge_lock fails to notice a THP so we will not charge it as THP and only charge one head page and then the tear down path will correctly recognize it as a THP and uncharge the full size, right? I have to admit I have no idea how PageDoubleMap works and how can pmd_trans_huge fail. -- Michal Hocko SUSE Labs
WARNING: multiple messages have this Message-ID (diff)
From: Michal Hocko <mhocko@kernel.org> To: Minchan Kim <minchan@kernel.org> Cc: Andrew Morton <akpm@linux-foundation.org>, "Kirill A. Shutemov" <kirill@shutemov.name>, Johannes Weiner <hannes@cmpxchg.org>, linux-mm@kvack.org, linux-kernel@vger.kernel.org Subject: Re: memcg uncharge page counter mismatch Date: Thu, 3 Dec 2015 09:54:52 +0100 [thread overview] Message-ID: <20151203085451.GC9264@dhcp22.suse.cz> (raw) In-Reply-To: <20151203021006.GA31041@bbox> On Thu 03-12-15 11:10:06, Minchan Kim wrote: > On Thu, Dec 03, 2015 at 10:34:04AM +0900, Minchan Kim wrote: > > On Wed, Dec 02, 2015 at 11:16:43AM +0100, Michal Hocko wrote: > > > On Tue 01-12-15 22:34:55, Minchan Kim wrote: > > > > With new test on mmotm-2015-11-25-17-08, I saw below WARNING message > > > > several times. I couldn't see it with reverting new THP refcount > > > > redesign. > > > > > > Just a wild guess. What prevents migration/compaction from calling > > > split_huge_page on thp zero page? There is VM_BUG_ON but it is not clear > > > > I guess migration should work with LRU pages now but zero page couldn't > > stay there. Ahh, you are right. I have missed PageLRU check in isolate_migratepages_block pfn walker. > > > whether you run with CONFIG_DEBUG_VM enabled. > > > > I enabled VM_DEBUG_VM. > > > > > > > > Also, how big is the underflow? [...] > > nr_pages 293 new -324 > > nr_pages 16 new -340 > > nr_pages 342 new -91 > > nr_pages 246 new -337 > > nr_pages 15 new -352 > > nr_pages 15 new -367 They are quite large but that is not that surprising if we consider that we are batching many uncharges at once. > My guess is that it's related to new feature of Kirill's THP 'PageDoubleMap' > so a THP page could be mapped a pte but !pmd_trans_huge(*pmd) so memcg > precharge in move_charge should handle it? I am not familiar with the current state of THP after the rework unfortunately. So if I got you right then you are saying that pmd_trans_huge_lock fails to notice a THP so we will not charge it as THP and only charge one head page and then the tear down path will correctly recognize it as a THP and uncharge the full size, right? I have to admit I have no idea how PageDoubleMap works and how can pmd_trans_huge fail. -- Michal Hocko SUSE Labs -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2015-12-03 8:54 UTC|newest] Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top 2015-12-01 13:34 memcg uncharge page counter mismatch Minchan Kim 2015-12-01 13:34 ` Minchan Kim 2015-12-02 10:16 ` Michal Hocko 2015-12-02 10:16 ` Michal Hocko 2015-12-03 1:34 ` Minchan Kim 2015-12-03 1:34 ` Minchan Kim 2015-12-03 2:10 ` Minchan Kim 2015-12-03 2:10 ` Minchan Kim 2015-12-03 8:54 ` Michal Hocko [this message] 2015-12-03 8:54 ` Michal Hocko 2015-12-03 12:59 ` Minchan Kim 2015-12-03 12:59 ` Minchan Kim 2015-12-03 13:37 ` Michal Hocko 2015-12-03 13:37 ` Michal Hocko 2015-12-03 13:43 ` Michal Hocko 2015-12-03 13:43 ` Michal Hocko 2015-12-03 14:58 ` Michal Hocko 2015-12-03 14:58 ` Michal Hocko 2015-12-03 15:47 ` Michal Hocko 2015-12-03 15:47 ` Michal Hocko 2015-12-04 5:35 ` Minchan Kim 2015-12-04 5:35 ` Minchan Kim 2015-12-04 8:52 ` Michal Hocko 2015-12-04 8:52 ` Michal Hocko 2015-12-04 9:16 ` Minchan Kim 2015-12-04 9:16 ` Minchan Kim 2015-12-04 9:58 ` Michal Hocko 2015-12-04 9:58 ` Michal Hocko 2015-12-04 13:35 ` Minchan Kim 2015-12-04 13:35 ` Minchan Kim 2015-12-04 16:01 ` Johannes Weiner 2015-12-04 16:01 ` Johannes Weiner
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=20151203085451.GC9264@dhcp22.suse.cz \ --to=mhocko@kernel.org \ --cc=akpm@linux-foundation.org \ --cc=hannes@cmpxchg.org \ --cc=kirill@shutemov.name \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-mm@kvack.org \ --cc=minchan@kernel.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.