From: Konstantin Khlebnikov <khlebnikov@openvz.org>
To: Hugh Dickins <hughd@google.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
Ying Han <yinghan@google.com>,
"linux-mm@kvack.org" <linux-mm@kvack.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH RFC 00/15] mm: memory book keeping and lru_lock splitting
Date: Sat, 18 Feb 2012 10:35:35 +0400 [thread overview]
Message-ID: <4F3F46B7.40100@openvz.org> (raw)
In-Reply-To: <alpine.LSU.2.00.1202171803380.25191@eggly.anvils>
Hugh Dickins wrote:
> On Thu, 16 Feb 2012, Hugh Dickins wrote:
>>
>> Yours are not the only patches I was testing in that tree, I tried to
>> gather several other series which I should be reviewing if I ever have
>> time: Kamezawa-san's page cgroup diet 6, Xiao Guangrong's 4 prio_tree
>> cleanups, your 3 radix_tree changes, your 6 shmem changes, your 4 memcg
>> miscellaneous, and then your 15 books.
>>
>> The tree before your final 15 did well under pressure, until I tried to
>> rmdir one of the cgroups afterwards: then it crashed nastily, I'll have
>> to bisect into that, probably either Kamezawa's or your memcg changes.
>
> So far I haven't succeeded in reproducing that at all: it was real,
> but obviously harder to get than I assumed - indeed, no good reason
> to associate it with any of those patches, might even be in 3.3-rc.
>
> It did involve a NULL pointer dereference in mem_cgroup_page_lruvec(),
> somewhere below compact_zone() - but repercussions were causing the
> stacktrace to scroll offscreen, so I didn't get good details.
There some stupid bugs in my v1 patchset, it shouldn't works at all.
I did not expect that someone will try to use it. I sent it just to discuss.
Most destructive bug is this PageCgroupUsed() below:
+struct book *page_book(struct page *page)
+{
+ struct mem_cgroup_per_zone *mz;
+ struct page_cgroup *pc;
+
+ if (mem_cgroup_disabled())
+ return &page_zone(page)->book;
+
+ pc = lookup_page_cgroup(page);
+ if (!PageCgroupUsed(pc))
+ return &page_zone(page)->book;
+ /* Ensure pc->mem_cgroup is visible after reading PCG_USED. */
+ smp_rmb();
+ mz = mem_cgroup_zoneinfo(pc->mem_cgroup,
+ page_to_nid(page), page_zonenum(page));
+ return &mz->book;
+}
Thus after page uncharge I remove page from wrong book, under wrong lock =)
[ as I wrote, updated patchset there: https://github.com/koct9i/linux ]
>
> Hugh
--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
WARNING: multiple messages have this Message-ID (diff)
From: Konstantin Khlebnikov <khlebnikov@openvz.org>
To: Hugh Dickins <hughd@google.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
Ying Han <yinghan@google.com>,
"linux-mm@kvack.org" <linux-mm@kvack.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH RFC 00/15] mm: memory book keeping and lru_lock splitting
Date: Sat, 18 Feb 2012 10:35:35 +0400 [thread overview]
Message-ID: <4F3F46B7.40100@openvz.org> (raw)
In-Reply-To: <alpine.LSU.2.00.1202171803380.25191@eggly.anvils>
Hugh Dickins wrote:
> On Thu, 16 Feb 2012, Hugh Dickins wrote:
>>
>> Yours are not the only patches I was testing in that tree, I tried to
>> gather several other series which I should be reviewing if I ever have
>> time: Kamezawa-san's page cgroup diet 6, Xiao Guangrong's 4 prio_tree
>> cleanups, your 3 radix_tree changes, your 6 shmem changes, your 4 memcg
>> miscellaneous, and then your 15 books.
>>
>> The tree before your final 15 did well under pressure, until I tried to
>> rmdir one of the cgroups afterwards: then it crashed nastily, I'll have
>> to bisect into that, probably either Kamezawa's or your memcg changes.
>
> So far I haven't succeeded in reproducing that at all: it was real,
> but obviously harder to get than I assumed - indeed, no good reason
> to associate it with any of those patches, might even be in 3.3-rc.
>
> It did involve a NULL pointer dereference in mem_cgroup_page_lruvec(),
> somewhere below compact_zone() - but repercussions were causing the
> stacktrace to scroll offscreen, so I didn't get good details.
There some stupid bugs in my v1 patchset, it shouldn't works at all.
I did not expect that someone will try to use it. I sent it just to discuss.
Most destructive bug is this PageCgroupUsed() below:
+struct book *page_book(struct page *page)
+{
+ struct mem_cgroup_per_zone *mz;
+ struct page_cgroup *pc;
+
+ if (mem_cgroup_disabled())
+ return &page_zone(page)->book;
+
+ pc = lookup_page_cgroup(page);
+ if (!PageCgroupUsed(pc))
+ return &page_zone(page)->book;
+ /* Ensure pc->mem_cgroup is visible after reading PCG_USED. */
+ smp_rmb();
+ mz = mem_cgroup_zoneinfo(pc->mem_cgroup,
+ page_to_nid(page), page_zonenum(page));
+ return &mz->book;
+}
Thus after page uncharge I remove page from wrong book, under wrong lock =)
[ as I wrote, updated patchset there: https://github.com/koct9i/linux ]
>
> Hugh
next prev parent reply other threads:[~2012-02-18 6:35 UTC|newest]
Thread overview: 62+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-02-15 22:57 [PATCH RFC 00/15] mm: memory book keeping and lru_lock splitting Konstantin Khlebnikov
2012-02-15 22:57 ` Konstantin Khlebnikov
2012-02-15 22:57 ` [PATCH RFC 01/15] mm: rename struct lruvec into struct book Konstantin Khlebnikov
2012-02-15 22:57 ` Konstantin Khlebnikov
2012-02-15 22:57 ` [PATCH RFC 02/15] mm: memory bookkeeping core Konstantin Khlebnikov
2012-02-15 22:57 ` Konstantin Khlebnikov
2012-02-15 22:57 ` [PATCH RFC 03/15] mm: add book->pages_count Konstantin Khlebnikov
2012-02-15 22:57 ` Konstantin Khlebnikov
2012-02-15 22:57 ` [PATCH RFC 04/15] mm: unify inactive_list_is_low() Konstantin Khlebnikov
2012-02-15 22:57 ` Konstantin Khlebnikov
2012-02-15 22:57 ` [PATCH RFC 05/15] mm: add book->reclaim_stat Konstantin Khlebnikov
2012-02-15 22:57 ` Konstantin Khlebnikov
2012-02-15 22:57 ` [PATCH RFC 06/15] mm: kill struct mem_cgroup_zone Konstantin Khlebnikov
2012-02-15 22:57 ` Konstantin Khlebnikov
2012-02-15 22:57 ` [PATCH RFC 07/15] mm: move page-to-book translation upper Konstantin Khlebnikov
2012-02-15 22:57 ` Konstantin Khlebnikov
2012-02-15 22:57 ` [PATCH RFC 08/15] mm: introduce book locking primitives Konstantin Khlebnikov
2012-02-15 22:57 ` Konstantin Khlebnikov
2012-02-15 22:57 ` [PATCH RFC 09/15] mm: handle book relocks on lumpy reclaim Konstantin Khlebnikov
2012-02-15 22:57 ` Konstantin Khlebnikov
2012-02-15 22:57 ` [PATCH RFC 10/15] mm: handle book relocks in compaction Konstantin Khlebnikov
2012-02-15 22:57 ` Konstantin Khlebnikov
2012-02-15 22:57 ` [PATCH RFC 11/15] mm: handle book relock in memory controller Konstantin Khlebnikov
2012-02-15 22:57 ` Konstantin Khlebnikov
2012-02-15 22:57 ` [PATCH RFC 12/15] mm: optimize books in update_page_reclaim_stat() Konstantin Khlebnikov
2012-02-15 22:57 ` Konstantin Khlebnikov
2012-02-15 22:57 ` [PATCH RFC 13/15] mm: optimize books in pagevec_lru_move_fn() Konstantin Khlebnikov
2012-02-15 22:57 ` Konstantin Khlebnikov
2012-02-15 22:57 ` [PATCH RFC 14/15] mm: optimize putback for 0-order reclaim Konstantin Khlebnikov
2012-02-15 22:57 ` Konstantin Khlebnikov
2012-02-15 22:58 ` [PATCH RFC 15/15] mm: split zone->lru_lock Konstantin Khlebnikov
2012-02-15 22:58 ` Konstantin Khlebnikov
2012-02-16 2:04 ` [PATCH RFC 00/15] mm: memory book keeping and lru_lock splitting KAMEZAWA Hiroyuki
2012-02-16 2:04 ` KAMEZAWA Hiroyuki
2012-02-16 5:43 ` Konstantin Khlebnikov
2012-02-16 5:43 ` Konstantin Khlebnikov
2012-02-16 8:24 ` KAMEZAWA Hiroyuki
2012-02-16 8:24 ` KAMEZAWA Hiroyuki
2012-02-16 11:02 ` Konstantin Khlebnikov
2012-02-16 11:02 ` Konstantin Khlebnikov
2012-02-16 15:54 ` Konstantin Khlebnikov
2012-02-16 15:54 ` Konstantin Khlebnikov
2012-02-16 23:54 ` KAMEZAWA Hiroyuki
2012-02-16 23:54 ` KAMEZAWA Hiroyuki
2012-02-18 9:09 ` Konstantin Khlebnikov
2012-02-18 9:09 ` Konstantin Khlebnikov
2012-02-16 2:37 ` Hugh Dickins
2012-02-16 2:37 ` Hugh Dickins
2012-02-16 4:51 ` Konstantin Khlebnikov
2012-02-16 4:51 ` Konstantin Khlebnikov
2012-02-16 21:37 ` Hugh Dickins
2012-02-16 21:37 ` Hugh Dickins
2012-02-17 19:56 ` Konstantin Khlebnikov
2012-02-17 19:56 ` Konstantin Khlebnikov
2012-02-18 2:13 ` Hugh Dickins
2012-02-18 2:13 ` Hugh Dickins
2012-02-18 6:35 ` Konstantin Khlebnikov [this message]
2012-02-18 6:35 ` Konstantin Khlebnikov
2012-02-18 7:14 ` Hugh Dickins
2012-02-18 7:14 ` Hugh Dickins
2012-02-20 0:32 ` KAMEZAWA Hiroyuki
2012-02-20 0:32 ` KAMEZAWA Hiroyuki
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=4F3F46B7.40100@openvz.org \
--to=khlebnikov@openvz.org \
--cc=hughd@google.com \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=yinghan@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: link
Be 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.