[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <23267.1316131006@turing-police.cc.vt.edu>
Date: Thu, 15 Sep 2011 19:56:46 -0400
From: Valdis.Kletnieks@...edu
To: Seth Jennings <sjenning@...ux.vnet.ibm.com>
Cc: gregkh@...e.de, devel@...verdev.osuosl.org, linux-mm@...ck.org,
ngupta@...are.org, linux-kernel@...r.kernel.org,
francis.moro@...il.com, dan.magenheimer@...cle.com
Subject: Re: [PATCH] staging: zcache: fix cleancache crash
On Tue, 13 Sep 2011 14:19:22 CDT, Seth Jennings said:
> After commit, c5f5c4db, cleancache crashes on the first
> successful get. This was caused by a remaining virt_to_page()
> call in zcache_pampd_get_data_and_free() that only gets
> run in the cleancache path.
>
> The patch converts the virt_to_page() to struct page
> casting like was done for other instances in c5f5c4db.
>
> Based on 3.1-rc4
>
> Signed-off-by: Seth Jennings <sjenning@...ux.vnet.ibm.com>
I was seeing all sorts of bizzare memory corruptions and panics
and crashes while testing zcache - average uptime was only 2-3 hours.
With this patch applied, now have close to 24 hours of crash-free operation.
Feel free to add this if it isn't in somebody's tree already:
Tested-By: Valdis Kletnieks <valdis.kletnieks@...edu>
Content of type "application/pgp-signature" skipped
Powered by blists - more mailing lists