[Top] [All Lists]

Re: Kernel New RAID 5 Bug (oops when writing Samba -> RAID5)

To: Chuck Ebbert <cebbert@xxxxxxxxxx>
Subject: Re: Kernel New RAID 5 Bug (oops when writing Samba -> RAID5)
From: Neil Brown <neilb@xxxxxxx>
Date: Tue, 23 Jan 2007 08:59:36 +1100
Cc: Justin Piszcz <jpiszcz@xxxxxxxxxxxxxxx>, linux-kernel@xxxxxxxxxxxxxxx, linux-raid@xxxxxxxxxxxxxxx, xfs@xxxxxxxxxxx
In-reply-to: message from Chuck Ebbert on Monday January 22
References: <Pine.LNX.4.64.0701200718290.29223@xxxxxxxxxxxxxxxx> <45B5261B.1050104@xxxxxxxxxx>
Sender: xfs-bounce@xxxxxxxxxxx
On Monday January 22, cebbert@xxxxxxxxxx wrote:
> Justin Piszcz wrote:
> > My .config is attached, please let me know if any other information is 
> > needed and please CC (lkml) as I am not on the list, thanks!
> >
> > Running Kernel on a MD RAID5 volume.  Copying files over Samba to 
> > the RAID5 running XFS.
> >
> > Any idea what happened here?
> >   
> Without digging too deeply, I'd say you've hit the same bug Sami Farin 
> and others
> have reported starting with 2.6.19: pages mapped with kmap_atomic() 
> become unmapped
> during memcpy() or similar operations.  Try disabling preempt -- that 
> seems to be the
> common factor.

That is exactly the conclusion I had just come to (a kmap_atomic page
must be being unmapped during memcpy).  I wasn't aware that others had
reported it - thanks for that.

Turning off CONFIG_PREEMPT certainly seems like a good idea.


<Prev in Thread] Current Thread [Next in Thread>