Re: [PATCH 00/22][RFC] Unionfs: Stackable Namespace Unification Filesystem

[Josef Sipek <jsipek@fsl.cs.sunysb.edu>]

On Sun, Sep 03, 2006 at 07:42:53PM +0200, Jan Engelhardt wrote:
> >> > This set of patches constitutes Unionfs version 2.0. We are presenting it to
> >> > be reviewed and considered for inclusion into the kernel.
> >> 
> >> Small nit: is it possible to order these patches so that the kernel
> >> builds at each intermediate point (so we can use git bisect).  The
> >> easiest way to achieve this would be to do the Kconfig and Makefile
> >> updates last.
> >
> >Ideally, when Unionfs is commited into git, the whole thing would be one
> >commit - what's the point of having half of a filesystem?
> 
> So that you can eliminate e.g. locking bugs by searching in less code.

I think you misunderstood my comment. What I meant to say was that there is
_no way_ you can compile a filesystem that has only dentry ops but not
superblock ops - this would happen if you tried to bisect and you landed
half way in the series of commits for the filesystem. For the _initial_
commit one cset makes sense. For subsequent fixes one commit per fix is the
only logical thing to do.

Josef "Jeff" Sipek.

-- 
Bad pun of the week: The formula 1 control computer suffered from a race
condition

-- 
VGER BF report: H 7.95683e-07