Re: [PATCH] iget-locked [2/6]

Jan Harkes (jaharkes@cs.cmu.edu)
Fri, 10 May 2002 23:04:37 -0400


On Sat, May 11, 2002 at 12:48:46PM +1000, Keith Owens wrote:
> On Fri, 10 May 2002 21:21:16 -0500 (CDT),
> Kai Germaschewski <kai@tp1.ruhr-uni-bochum.de> wrote:
> >This is not true anymore in 2.5, this limitation was removed when ALSA
> >went in.
>
> True, but if the iget change goes into 2.5 it will probably be
> backported to 2.4 later, 2.4 still has the restriction.
>
> As for modversions on 2.5, well you know my opinion ;).

A backport is not that likely. The patch removes iget4 and as a result
breaks compatibility for binary-only kernel modules that use iget and/or
iget4. So, I don't believe this patch is appropriate for a stable series.

I'm going to fix the iget4 race in 2.4 by adding a per-superblock
semaphore around the call to iget4 in Coda. My guess is that NFS and
ReiserFS will have to do something similar. Filesystems that do not
use a special 'find_actor' (i.e. iget) don't have a problem in 2.4.

Jan

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/