On Sat, Dec 18, 2010 at 01:01:18PM +1100, Nick Piggin wrote:
> On Fri, Dec 17, 2010 at 01:00:23PM -0500, J. Bruce Fields wrote:
> > From: J. Bruce Fields <bfields@xxxxxxxxxx>
> >
> > Without this patch
> >
> > Â Â Â Âclient$ mount -tnfs4 server:/export/ /mnt/
> > Â Â Â Âclient$ tail -f /mnt/FOO
> > Â Â Â Â...
> > Â Â Â Âserver$ df -i /export
> > Â Â Â Âserver$ rm /export/FOO
> > Â Â Â Â(^C the tail -f)
> > Â Â Â Âserver$ df -i /export
> > Â Â Â Âserver$ echo 2 >/proc/sys/vm/drop_caches
> > Â Â Â Âserver$ df -i /export
> >
> > the df's will show that the inode is not freed on the filesystem until
> > the last step, when it could have been freed after killing the client's
> > tail -f. ÂOn-disk data won't be deallocated either, leading to possible
> > spurious ENOSPC.
> >
> > This occurs because when the client does the close, it arrives in a
> > compound with a putfh and a close, processed like:
> >
> > Â Â Â Â- putfh: look up the filehandle. ÂThe only alias found for the
> > Â Â Â Â Âinode will be DCACHE_UNHASHED alias referenced by the filp
> > Â Â Â Â Âassociated with the nfsd open. Âd_obtain_alias() doesn't like
> > Â Â Â Â Âthis, so it creates a new DCACHE_DISCONECTED dentry and
> > Â Â Â Â Âreturns that instead.
> >
> > Nick Piggin suggested fixing this by allowing d_obtain_alias to return
> > the unhashed dentry that is referenced by the filp, instead of making it
> > create a new dentry.
> >
> > Leave __d_find_alias() alone to avoid changing behavior of other
> > callers.
> >
> > Also nfsd doesn't need all the checks of __d_find_alias(); any dentry,
> > hashed or unhashed, disconnected or not, should work.
> >
> > Signed-off-by: J. Bruce Fields <bfields@xxxxxxxxxx>
> > ---
> > fs/dcache.c | 26 ++++++++++++++++++++++++--
> > 1 files changed, 24 insertions(+), 2 deletions(-)
> >
> > On Tue, Dec 14, 2010 at 05:01:02PM -0500, J. Bruce Fields wrote:
> > > On Mon, Dec 13, 2010 at 04:19:44PM +1100, Nick Piggin wrote:
> > > > Not sure where Al's hiding...
> > > >
> > > > But I would like to update the comments, and perhaps even a new
> > > > add a new function here (or new flag to __d_find_alias).
> > > >
> > > > AFAIKS, the callers are OK, however I suppose d_splice_alias and
> > > > d_materialise_unique should not have unlinked inodes at this point,
> > > > so at least a BUG_ON for them might be a good idea?
> > >
> > > That does sound safer. I'm pretty confused by the various
> > > __di_splice_alias callers. I'll go search through them and see if I can
> > > understand better....
> >
> > I think a new __d_find_alias flag would just make it more confusing.
> > And it looks like all d_obtain_alias needs is something much simpler.
> > This works for me.
>
> Well I don't really know if this is clearer. By convention,
> __ prefix version should be same as non prefix version but just
> take fewer locks. Why do you have the new d_find_any_alias()?
Argh, apologies; with typo fixed.
--b.
commit 5a911af645aae9992d465d57c397237fb35d1f93
Author: J. Bruce Fields <bfields@xxxxxxxxxx>
Date: Fri Dec 17 09:05:04 2010 -0500
fs/dcache: allow __d_obtain_alias() to return unhashed dentries
Without this patch
   Âclient$ mount -tnfs4 server:/export/ /mnt/
   Âclient$ tail -f /mnt/FOO
   Â...
   Âserver$ df -i /export
   Âserver$ rm /export/FOO
   Â(^C the tail -f)
   Âserver$ df -i /export
   Âserver$ echo 2 >/proc/sys/vm/drop_caches
   Âserver$ df -i /export
the df's will show that the inode is not freed on the filesystem until
the last step, when it could have been freed after killing the client's
tail -f. ÂOn-disk data won't be deallocated either, leading to possible
spurious ENOSPC.
This occurs because when the client does the close, it arrives in a
compound with a putfh and a close, processed like:
   Â- putfh: look up the filehandle. ÂThe only alias found for the
    Âinode will be DCACHE_UNHASHED alias referenced by the filp
    Âassociated with the nfsd open. Âd_obtain_alias() doesn't like
    Âthis, so it creates a new DCACHE_DISCONECTED dentry and
    Âreturns that instead.
Nick Piggin suggested fixing this by allowing d_obtain_alias to return
the unhashed dentry that is referenced by the filp, instead of making it
create a new dentry.
Leave __d_find_alias() alone to avoid changing behavior of other
callers.
Also nfsd doesn't need all the checks of __d_find_alias(); any dentry,
hashed or unhashed, disconnected or not, should work.
Signed-off-by: J. Bruce Fields <bfields@xxxxxxxxxx>
diff --git a/fs/dcache.c b/fs/dcache.c
index 5ed93cd..5c014a5 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -1135,6 +1135,28 @@ static inline struct hlist_head *d_hash(struct dentry *parent,
return dentry_hashtable + (hash & D_HASHMASK);
}
+static struct dentry * __d_find_any_alias(struct inode *inode)
+{
+ struct dentry *alias;
+
+ if (list_empty(&inode->i_dentry))
+ return NULL;
+ alias = list_first_entry(&inode->i_dentry, struct dentry, d_alias);
+ __dget_locked(alias);
+ return alias;
+}
+
+static struct dentry * d_find_any_alias(struct inode *inode)
+{
+ struct dentry *de;
+
+ spin_lock(&dcache_lock);
+ de = __d_find_any_alias(inode);
+ spin_unlock(&dcache_lock);
+ return de;
+}
+
+
/**
* d_obtain_alias - find or allocate a dentry for a given inode
* @inode: inode to allocate the dentry for
@@ -1164,7 +1186,7 @@ struct dentry *d_obtain_alias(struct inode *inode)
if (IS_ERR(inode))
return ERR_CAST(inode);
- res = d_find_alias(inode);
+ res = d_find_any_alias(inode);
if (res)
goto out_iput;
@@ -1176,7 +1198,7 @@ struct dentry *d_obtain_alias(struct inode *inode)
tmp->d_parent = tmp; /* make sure dput doesn't croak */
spin_lock(&dcache_lock);
- res = __d_find_alias(inode, 0);
+ res = __d_find_any_alias(inode);
if (res) {
spin_unlock(&dcache_lock);
dput(tmp);
--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
