From: Michal Hocko <mhocko@kernel.org> To: <linux-mm@kvack.org> Cc: Mikulas Patocka <mpatocka@redhat.com>, Ondrej Kozina <okozina@redhat.com>, David Rientjes <rientjes@google.com>, Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>, Mel Gorman <mgorman@suse.de>, Neil Brown <neilb@suse.de>, Andrew Morton <akpm@linux-foundation.org>, LKML <linux-kernel@vger.kernel.org>, dm-devel@redhat.com Subject: [RFC PATCH 0/2] mempool vs. page allocator interaction Date: Mon, 18 Jul 2016 10:39:22 +0200 [thread overview] Message-ID: <1468831164-26621-1-git-send-email-mhocko@kernel.org> (raw) Hi, there have been two issues identified when investigating dm-crypt backed swap recently [1]. The first one looks like a regression from f9054c70d28b ("mm, mempool: only set __GFP_NOMEMALLOC if there are free elements") because swapout path can now deplete all the available memory reserves. The first patch tries to address that issue by dropping __GFP_NOMEMALLOC only to TIF_MEMDIE tasks. The second issue is that dm writeout path which relies on mempool allocator gets throttled by the direct reclaim in throttle_vm_writeout which just makes the whole memory pressure problem even worse. The patch2 just makes sure that we annotate mempool users to be throttled less by PF_LESS_THROTTLE flag and prevent from throttle_vm_writeout for that path. mempool users are usually the IO path and throttle them less sounds like a reasonable way to go. I do not have any more complicated dm setup available so I would appreciate if dm people (CCed) could give these two a try. Also it would be great to iron out concerns from David. He has posted a deadlock stack trace [2] which has led to f9054c70d28b which is bio allocation lockup because the TIF_MEMDIE process cannot make a forward progress without access to memory reserve. This case should be fixed by patch 1 AFAICS. There are other potential cases when the stuck mempool is called from PF_MEMALLOC context and blocks the oom victim indirectly (over a lock) but I believe those are much less likely and we have the oom reaper to make a forward progress. Sorry of pulling the discussion outside of the original email thread but there were more lines of dicussion there and I felt discussing particualr solution with its justification has a greater chance of moving towards a solution. I am sending this as an RFC because this needs a deep review as there might be other side effects I do not see (especially about patch 2). Any comments, suggestions are welcome. --- [1] http://lkml.kernel.org/r/alpine.LRH.2.02.1607111027080.14327@file01.intranet.prod.int.rdu2.redhat.com [2] http://lkml.kernel.org/r/alpine.DEB.2.10.1607131644590.92037@chino.kir.corp.google.com
WARNING: multiple messages have this Message-ID (diff)
From: Michal Hocko <mhocko@kernel.org> To: linux-mm@kvack.org Cc: Mikulas Patocka <mpatocka@redhat.com>, Ondrej Kozina <okozina@redhat.com>, David Rientjes <rientjes@google.com>, Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>, Mel Gorman <mgorman@suse.de>, Neil Brown <neilb@suse.de>, Andrew Morton <akpm@linux-foundation.org>, LKML <linux-kernel@vger.kernel.org>, dm-devel@redhat.com Subject: [RFC PATCH 0/2] mempool vs. page allocator interaction Date: Mon, 18 Jul 2016 10:39:22 +0200 [thread overview] Message-ID: <1468831164-26621-1-git-send-email-mhocko@kernel.org> (raw) Hi, there have been two issues identified when investigating dm-crypt backed swap recently [1]. The first one looks like a regression from f9054c70d28b ("mm, mempool: only set __GFP_NOMEMALLOC if there are free elements") because swapout path can now deplete all the available memory reserves. The first patch tries to address that issue by dropping __GFP_NOMEMALLOC only to TIF_MEMDIE tasks. The second issue is that dm writeout path which relies on mempool allocator gets throttled by the direct reclaim in throttle_vm_writeout which just makes the whole memory pressure problem even worse. The patch2 just makes sure that we annotate mempool users to be throttled less by PF_LESS_THROTTLE flag and prevent from throttle_vm_writeout for that path. mempool users are usually the IO path and throttle them less sounds like a reasonable way to go. I do not have any more complicated dm setup available so I would appreciate if dm people (CCed) could give these two a try. Also it would be great to iron out concerns from David. He has posted a deadlock stack trace [2] which has led to f9054c70d28b which is bio allocation lockup because the TIF_MEMDIE process cannot make a forward progress without access to memory reserve. This case should be fixed by patch 1 AFAICS. There are other potential cases when the stuck mempool is called from PF_MEMALLOC context and blocks the oom victim indirectly (over a lock) but I believe those are much less likely and we have the oom reaper to make a forward progress. Sorry of pulling the discussion outside of the original email thread but there were more lines of dicussion there and I felt discussing particualr solution with its justification has a greater chance of moving towards a solution. I am sending this as an RFC because this needs a deep review as there might be other side effects I do not see (especially about patch 2). Any comments, suggestions are welcome. --- [1] http://lkml.kernel.org/r/alpine.LRH.2.02.1607111027080.14327@file01.intranet.prod.int.rdu2.redhat.com [2] http://lkml.kernel.org/r/alpine.DEB.2.10.1607131644590.92037@chino.kir.corp.google.com -- 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 reply other threads:[~2016-07-18 8:39 UTC|newest] Thread overview: 102+ messages / expand[flat|nested] mbox.gz Atom feed top 2016-07-18 8:39 Michal Hocko [this message] 2016-07-18 8:39 ` [RFC PATCH 0/2] mempool vs. page allocator interaction Michal Hocko 2016-07-18 8:41 ` [RFC PATCH 1/2] mempool: do not consume memory reserves from the reclaim path Michal Hocko 2016-07-18 8:41 ` Michal Hocko 2016-07-18 8:41 ` [RFC PATCH 2/2] mm, mempool: do not throttle PF_LESS_THROTTLE tasks Michal Hocko 2016-07-18 8:41 ` Michal Hocko 2016-07-19 21:50 ` Mikulas Patocka 2016-07-19 21:50 ` Mikulas Patocka 2016-07-22 8:46 ` NeilBrown 2016-07-22 9:04 ` NeilBrown 2016-07-22 9:15 ` Michal Hocko 2016-07-22 9:15 ` Michal Hocko 2016-07-23 0:12 ` NeilBrown 2016-07-25 8:32 ` Michal Hocko 2016-07-25 8:32 ` Michal Hocko 2016-07-25 19:23 ` Michal Hocko 2016-07-25 19:23 ` Michal Hocko 2016-07-25 19:23 ` Michal Hocko 2016-07-26 7:07 ` Michal Hocko 2016-07-26 7:07 ` Michal Hocko 2016-07-27 3:43 ` [dm-devel] " NeilBrown 2016-07-27 18:24 ` Michal Hocko 2016-07-27 18:24 ` Michal Hocko 2016-07-27 21:33 ` NeilBrown 2016-07-28 7:17 ` Michal Hocko 2016-07-28 7:17 ` Michal Hocko 2016-08-03 12:53 ` Mikulas Patocka 2016-08-03 12:53 ` Mikulas Patocka 2016-08-03 14:34 ` Michal Hocko 2016-08-03 14:34 ` Michal Hocko 2016-08-04 18:49 ` Mikulas Patocka 2016-08-04 18:49 ` Mikulas Patocka 2016-08-12 12:32 ` Michal Hocko 2016-08-12 12:32 ` Michal Hocko 2016-08-13 17:34 ` Mikulas Patocka 2016-08-13 17:34 ` Mikulas Patocka 2016-08-14 10:34 ` Michal Hocko 2016-08-14 10:34 ` Michal Hocko 2016-08-15 16:15 ` Mikulas Patocka 2016-08-15 16:15 ` Mikulas Patocka 2016-11-23 21:11 ` Mikulas Patocka 2016-11-23 21:11 ` Mikulas Patocka 2016-11-24 13:29 ` Michal Hocko 2016-11-24 13:29 ` Michal Hocko 2016-11-24 17:10 ` Mikulas Patocka 2016-11-24 17:10 ` Mikulas Patocka 2016-11-28 14:06 ` Michal Hocko 2016-11-28 14:06 ` Michal Hocko 2016-07-25 21:52 ` Mikulas Patocka 2016-07-25 21:52 ` Mikulas Patocka 2016-07-26 7:25 ` Michal Hocko 2016-07-26 7:25 ` Michal Hocko 2016-07-27 4:02 ` [dm-devel] " NeilBrown 2016-07-27 14:28 ` Mikulas Patocka 2016-07-27 14:28 ` Mikulas Patocka 2016-07-27 18:40 ` Michal Hocko 2016-07-27 18:40 ` Michal Hocko 2016-08-03 13:59 ` Mikulas Patocka 2016-08-03 13:59 ` Mikulas Patocka 2016-08-03 14:42 ` Michal Hocko 2016-08-03 14:42 ` Michal Hocko 2016-08-04 18:46 ` Mikulas Patocka 2016-08-04 18:46 ` Mikulas Patocka 2016-07-27 21:36 ` NeilBrown 2016-07-19 2:00 ` [RFC PATCH 1/2] mempool: do not consume memory reserves from the reclaim path David Rientjes 2016-07-19 2:00 ` David Rientjes 2016-07-19 7:49 ` Michal Hocko 2016-07-19 7:49 ` Michal Hocko 2016-07-19 13:54 ` Johannes Weiner 2016-07-19 13:54 ` Johannes Weiner 2016-07-19 14:19 ` Michal Hocko 2016-07-19 14:19 ` Michal Hocko 2016-07-19 22:01 ` Mikulas Patocka 2016-07-19 22:01 ` Mikulas Patocka 2016-07-19 20:45 ` David Rientjes 2016-07-19 20:45 ` David Rientjes 2016-07-20 8:15 ` Michal Hocko 2016-07-20 8:15 ` Michal Hocko 2016-07-20 21:06 ` David Rientjes 2016-07-20 21:06 ` David Rientjes 2016-07-21 8:52 ` Michal Hocko 2016-07-21 8:52 ` Michal Hocko 2016-07-21 12:13 ` Johannes Weiner 2016-07-21 12:13 ` Johannes Weiner 2016-07-21 14:53 ` Michal Hocko 2016-07-21 14:53 ` Michal Hocko 2016-07-21 14:53 ` Michal Hocko 2016-07-21 15:26 ` Johannes Weiner 2016-07-21 15:26 ` Johannes Weiner 2016-07-22 1:41 ` NeilBrown 2016-07-22 6:37 ` Michal Hocko 2016-07-22 6:37 ` Michal Hocko 2016-07-22 12:26 ` Vlastimil Babka 2016-07-22 12:26 ` Vlastimil Babka 2016-07-22 19:44 ` Andrew Morton 2016-07-22 19:44 ` Andrew Morton 2016-07-23 18:52 ` Vlastimil Babka 2016-07-23 18:52 ` Vlastimil Babka 2016-07-19 21:50 ` Mikulas Patocka 2016-07-19 21:50 ` Mikulas Patocka 2016-07-20 6:44 ` Michal Hocko 2016-07-20 6:44 ` Michal Hocko
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=1468831164-26621-1-git-send-email-mhocko@kernel.org \ --to=mhocko@kernel.org \ --cc=akpm@linux-foundation.org \ --cc=dm-devel@redhat.com \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-mm@kvack.org \ --cc=mgorman@suse.de \ --cc=mpatocka@redhat.com \ --cc=neilb@suse.de \ --cc=okozina@redhat.com \ --cc=penguin-kernel@I-love.SAKURA.ne.jp \ --cc=rientjes@google.com \ /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.