From: Andrew Morton <akpm@linux-foundation.org>
To: Minchan Kim <minchan@kernel.org>
Cc: Richard Davies <richard@arachsys.com>, KVM <kvm@vger.kernel.org>,
QEMU-devel <qemu-devel@nongnu.org>,
LKML <linux-kernel@vger.kernel.org>,
Linux-MM <linux-mm@kvack.org>, Mel Gorman <mgorman@suse.de>,
Shaohua Li <shli@kernel.org>, Avi Kivity <avi@redhat.com>
Subject: Re: [Qemu-devel] [PATCH 5/9] mm: compaction: Acquire the zone->lru_lock as late as possible
Date: Tue, 25 Sep 2012 14:39:31 -0700 [thread overview]
Message-ID: <20120925143931.f404ca22.akpm@linux-foundation.org> (raw)
In-Reply-To: <20120925081327.GA7759@bbox>
On Tue, 25 Sep 2012 17:13:27 +0900
Minchan Kim <minchan@kernel.org> wrote:
> I see. To me, your saying is better than current comment.
> I hope comment could be more explicit.
>
> diff --git a/mm/compaction.c b/mm/compaction.c
> index df01b4e..f1d2cc7 100644
> --- a/mm/compaction.c
> +++ b/mm/compaction.c
> @@ -542,8 +542,9 @@ isolate_migratepages_range(struct zone *zone, struct compact_control *cc,
> * splitting and collapsing (collapsing has already happened
> * if PageLRU is set) but the lock is not necessarily taken
> * here and it is wasteful to take it just to check transhuge.
> - * Check transhuge without lock and skip if it's either a
> - * transhuge or hugetlbfs page.
> + * Check transhuge without lock and *skip* if it's either a
> + * transhuge or hugetlbfs page because it's not safe to call
> + * compound_order.
> */
> if (PageTransHuge(page)) {
> if (!locked)
Going a bit further:
--- a/mm/compaction.c~mm-compaction-acquire-the-zone-lru_lock-as-late-as-possible-fix
+++ a/mm/compaction.c
@@ -415,7 +415,8 @@ isolate_migratepages_range(struct zone *
* if PageLRU is set) but the lock is not necessarily taken
* here and it is wasteful to take it just to check transhuge.
* Check transhuge without lock and skip if it's either a
- * transhuge or hugetlbfs page.
+ * transhuge or hugetlbfs page because calling compound_order()
+ * requires lru_lock to exclude isolation and splitting.
*/
if (PageTransHuge(page)) {
if (!locked)
_
but... the requirement to hold lru_lock for compound_order() is news
to me. It doesn't seem to be written down or explained anywhere, and
one wonders why the cheerily undocumented compound_lock() doesn't have
this effect. What's going on here??
next prev parent reply other threads:[~2012-09-25 21:39 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-09-21 10:46 [Qemu-devel] [PATCH 0/9] Reduce compaction scanning and lock contention Mel Gorman
2012-09-21 10:46 ` [Qemu-devel] [PATCH 1/9] Revert "mm: compaction: check lock contention first before taking lock" Mel Gorman
2012-09-21 17:46 ` Rafael Aquini
2012-09-21 10:46 ` [Qemu-devel] [PATCH 2/9] Revert "mm-compaction-abort-compaction-loop-if-lock-is-contended-or-run-too-long-fix" Mel Gorman
2012-09-21 17:47 ` Rafael Aquini
2012-09-21 10:46 ` [Qemu-devel] [PATCH 3/9] Revert "mm: compaction: abort compaction loop if lock is contended or run too long" Mel Gorman
2012-09-21 17:48 ` Rafael Aquini
2012-09-21 10:46 ` [Qemu-devel] [PATCH 4/9] mm: compaction: Abort compaction loop if lock is contended or run too long Mel Gorman
2012-09-21 17:50 ` Rafael Aquini
2012-09-21 21:31 ` Andrew Morton
2012-09-25 7:34 ` Minchan Kim
2012-09-21 10:46 ` [Qemu-devel] [PATCH 5/9] mm: compaction: Acquire the zone->lru_lock as late as possible Mel Gorman
2012-09-21 17:51 ` Rafael Aquini
2012-09-25 7:05 ` Minchan Kim
2012-09-25 7:51 ` Mel Gorman
2012-09-25 8:13 ` Minchan Kim
2012-09-25 21:39 ` Andrew Morton [this message]
2012-09-26 0:23 ` Minchan Kim
2012-09-26 10:17 ` Mel Gorman
2012-09-21 10:46 ` [Qemu-devel] [PATCH 6/9] mm: compaction: Acquire the zone->lock " Mel Gorman
2012-09-21 17:52 ` Rafael Aquini
2012-09-21 21:35 ` Andrew Morton
2012-09-24 8:52 ` Mel Gorman
2012-09-25 7:36 ` Minchan Kim
2012-09-25 7:35 ` Minchan Kim
2012-09-21 10:46 ` [Qemu-devel] [PATCH 7/9] Revert "mm: have order > 0 compaction start off where it left" Mel Gorman
2012-09-21 17:52 ` Rafael Aquini
2012-09-25 7:37 ` Minchan Kim
2012-09-21 10:46 ` [Qemu-devel] [PATCH 8/9] mm: compaction: Cache if a pageblock was scanned and no pages were isolated Mel Gorman
2012-09-21 17:53 ` Rafael Aquini
2012-09-21 21:36 ` Andrew Morton
2012-09-24 9:39 ` Mel Gorman
2012-09-24 21:26 ` Andrew Morton
2012-09-25 9:12 ` Mel Gorman
2012-09-25 20:03 ` Andrew Morton
2012-09-27 12:06 ` [Qemu-devel] [PATCH] mm: compaction: cache if a pageblock was scanned and no pages were isolated -fix2 Mel Gorman
2012-09-27 13:12 ` [Qemu-devel] [PATCH 8/9] mm: compaction: Cache if a pageblock was scanned and no pages were isolated Mel Gorman
2012-09-26 0:49 ` Minchan Kim
2012-09-27 12:14 ` Mel Gorman
2012-09-21 10:46 ` [Qemu-devel] [PATCH 9/9] mm: compaction: Restart compaction from near where it left off Mel Gorman
2012-09-21 17:54 ` Rafael Aquini
2012-09-21 13:51 ` [Qemu-devel] [PATCH 0/9] Reduce compaction scanning and lock contention Rik van Riel
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=20120925143931.f404ca22.akpm@linux-foundation.org \
--to=akpm@linux-foundation.org \
--cc=avi@redhat.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@suse.de \
--cc=minchan@kernel.org \
--cc=qemu-devel@nongnu.org \
--cc=richard@arachsys.com \
--cc=shli@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: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).