[Cluster-devel] [PATCH] GFS2: kernel changes to support new gfs2_grow command (Try 3)

[David Teigland]

On Tue, May 08, 2007 at 03:24:41PM +0100, Steven Whitehouse wrote:
> > Well, I retested this code path and found that this code is still 
> > necessary  (unless I should somehow be doing this a different way).
> > Without the code, the gfs2_rindex_hold tries to add a holder,
> > but there's already a holder due to the meta_fs.  Under the right 
> > conditions, you get:
> > 
> > original: gfs2_prepare_write+0x49/0x237 [gfs2]
> > new: gfs2_rindex_hold+0x2b/0x52a [gfs2]
> 
> I agree that this change is needed, its how we've solved the same
> problem elsewhere and it looks ok to me,

OK, I see, it's because we removed the ability for a process to have two
holders on the same glock from gfs2.  I'm assuming that the glock taken
for writing is released after inplace_release().

Dave