Subject: Re: Oops in bdflush with 2.4.1[4|7]-xfs
From: Eric Sandeen <sandeen@sgi.com>
To: Eric Sandeen <sandeen@sgi.com>
Cc: Nathan Poznick <poznick@conwaycorp.net>, linux-xfs@oss.sgi.com
In-Reply-To: <1012429546.4291.19.camel@stout.americas.sgi.com>
References: <20020130214108.GA25792@conwaycorp.net>
	<1012428591.4291.15.camel@stout.americas.sgi.com> 
	<20020130222207.GA26516@conwaycorp.net> 
	<1012429546.4291.19.camel@stout.americas.sgi.com>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Date: 30 Jan 2002 16:52:08 -0600
Message-Id: <1012431129.4291.23.camel@stout.americas.sgi.com>
Mime-Version: 1.0
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Status: O
Content-Length: 964
Lines: 37

The other thing you can do is disassemble the code it oopsed in, so we
can see where it was when it went belly up...

I'm not sure where pagebuf lives in your current kernel tree...

Can you try commenting out the last line of page_buf.c*, then 
chmod +x /localhome/eric/2.4.x-xfs/workarea-reallyclean/linux/scripts/makelst

and

make fs/xfs/pagebuf/page_buf.lst

or

make fs/pagebuf/page_buf.lst

then send the resulting page_buf.lst to me, off-list?

(don't forget to un-comment the last line when you're done).

That way we can see where things go wrong.

-Eric

*(it doesn't deal with the EXPORT_SYMBOL() for some reason)

On Wed, 2002-01-30 at 16:25, Eric Sandeen wrote:
> If you're not using it, try turning it off.  DMAPI-enabled kernels don't
> get as much testing as non-DMAPI, it could be causing the problem.  Just
> a guess at this point.
> 
> -Eric

-- 
Eric Sandeen      XFS for Linux     http://oss.sgi.com/projects/xfs
sandeen@sgi.com   SGI, Inc.