Re: NULL poniter dereference in isolate_lru_pages 2.6.39.1

On Tue, 19 Jul 2011 07:48:11 +0900 Minchan Kim [off-list ref] wrote:

Thanks for the test, Chris.
Andrew.
We should push this into -stable.

That's

commit d179e84ba5da1d0024087d1759a2938817a00f3f
Author:     Andrea Arcangeli [off-list ref]
AuthorDate: Wed Jun 15 15:08:51 2011 -0700
Commit:     Linus Torvalds [off-list ref]
CommitDate: Wed Jun 15 20:04:02 2011 -0700

    mm: vmscan: do not use page_count without a page pin

On Fri, Jul 15, 2011 at 10:48 AM, Chris Pearson
[off-list ref] wrote:

quoted

That definately fixed it. __MTBF was about 20 days on those systems, since that patch we've had 200 server days with no problems.

Thanks!

On Tue, 12 Jul 2011, Minchan Kim wrote:

quoted

Date: Tue, 12 Jul 2011 09:16:09 +0900
From: Minchan Kim <redacted>
To: Chris Pearson <redacted>
Cc: linux-mm@kvack.org, Andrea Arcangeli <redacted>,
__ __Andrew Morton [off-list ref], stable [off-list ref]
Subject: Re: NULL poniter dereference in isolate_lru_pages 2.6.39.1

On Tue, Jul 12, 2011 at 5:52 AM, Chris Pearson
[off-list ref] wrote:

quoted

We applied the patch to many servers. __No problems so far.

The .config is attached.

Thanks. I verified. The point where isolate_lru_pages + 0x225 is
page_count exactly. So Andrea patch solves this problem apparently.
Couldn't we throw this patch to stable tree?

https://patchwork.kernel.org/patch/857442/

quoted

What's the config option to get that debugging info in the future?

CONFIG_DEBUG_INFO helps you. :)

--
Kind regards,
Minchan Kim

-- 
Kind regards,
Minchan Kim

--
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>