Re: [PATCH 1/4] ext2: remove inode constructor

["Satyam Sharma" <satyam.sharma@gmail.com>]

On 5/10/07, Andrew Morton <akpm@linux-foundation.org> wrote:
> On Sat, 5 May 2007 12:57:46 +0300 (EEST)
> Pekka J Enberg <penberg@cs.helsinki.fi> wrote:
>
> > From: Pekka Enberg <penberg@cs.helsinki.fi>
> >
> > As alloc_inode() touches the same cache line as init_once(), we gain
> > nothing from using slab constructors.
> >
> > Cc: Stephen C. Tweedie <sct@redhat.com>
> > Cc: Andreas Dilger <adilger@clusterfs.com>
> > Cc: Christoph Lameter <clameter@sgi.com>
> > Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
> > ---
> >  fs/ext2/super.c |   32 +++++++++++++-------------------
> >  1 file changed, 13 insertions(+), 19 deletions(-)
> >
> > Index: 26-mm/fs/ext2/super.c
> > ===================================================================
> > --- 26-mm.orig/fs/ext2/super.c        2007-05-05 12:26:15.000000000 +0300
> > +++ 26-mm/fs/ext2/super.c     2007-05-05 12:30:50.000000000 +0300
> > @@ -140,16 +140,24 @@ static struct kmem_cache * ext2_inode_ca
> >  static struct inode *ext2_alloc_inode(struct super_block *sb)
> >  {
> >       struct ext2_inode_info *ei;
> > -     ei = (struct ext2_inode_info *)kmem_cache_alloc(ext2_inode_cachep,
> > -                                             GFP_KERNEL|__GFP_RECLAIMABLE);
> > +     struct inode *inode;
> > +
> > +     ei = kmem_cache_alloc(ext2_inode_cachep, GFP_KERNEL|__GFP_RECLAIMABLE);
> >       if (!ei)
> >               return NULL;
> > +     rwlock_init(&ei->i_meta_lock);
> > +#ifdef CONFIG_EXT2_FS_XATTR
> > +     init_rwsem(&ei->xattr_sem);
> > +#endif
> > +     mutex_init(&ei->truncate_mutex);
>
> ext2 has no truncate_mutex.

2.6.21-mm2/broken-out/ext2-reservations.patch seems to have added one.

> These patches are rather tangled up with the unmerged __GFP_RECLAIMABLE
> stuff.

So _and_ the unmerged ext2-reservations stuff, you mean :-)

> I'll duck them for now.