Messages in this thread | | | Date | Tue, 16 Jul 2013 12:39:39 +0200 | From | Niels de Vos <> | Subject | Re: [fuse-devel] [PATCH] fuse: fix occasional dentry leak when readdirplus is used |
| |
On Mon, Jul 15, 2013 at 04:08:22PM -0400, Brian Foster wrote: > On 07/15/2013 08:59 AM, Niels de Vos wrote: > > In case d_lookup() returns a dentry with d_inode == NULL, the dentry is > > not returned with dput(). This results in triggering a BUG() in > > shrink_dcache_for_umount_subtree(): > > > > BUG: Dentry ...{i=0,n=...} still in use (1) [unmount of fuse fuse] > > > > Reported-by: Justin Clift <jclift@redhat.com> > > Signed-off-by: Niels de Vos <ndevos@redhat.com> > > > > -- > > Reproducing the BUG() on kernels with fuse that support READDIRPLUS can > > be done with the GlusterFS tests: > > - http://www.gluster.org/community/documentation/index.php/Using_the_Gluster_Test_Framework > > > > After some stressing of the VFS and fuse mountpoints, bug-860663.t will > > hit the BUG(). It does not happen on running this test stand-alone. > > Hi Neils, > > FYI, this is fairly easy to reproduce on-demand with gluster: > > - mount a volume to two local mountpoints (i.e., I used a single > storage/posix translator volume): > glusterfs --volfile=./test.vol /mnt/{1,2} --use-readdirp=1 > - create a negative dentry in one mountpoint: > ls /mnt/1/file (results in ENOENT) > - create the file via the second mountpoint: > touch /mnt/2/file > - run a readdirp on the first mountpoint: > ls /mnt/1/ > - umount /mnt/2 /mnt/1
Thanks, that definitely makes it easier to verify the fix.
> > --- > > fs/fuse/dir.c | 4 +++- > > 1 files changed, 3 insertions(+), 1 deletions(-) > > > > diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c > > index 0eda527..da67a15 100644 > > --- a/fs/fuse/dir.c > > +++ b/fs/fuse/dir.c > > @@ -1246,7 +1246,9 @@ static int fuse_direntplus_link(struct file *file, > > if (err) > > goto out; > > dput(dentry); > > - dentry = NULL; > > + } else if (dentry) { > > + /* this dentry does not have a d_inode, just drop it */ > > + dput(dentry); > > } > > I'm not really familiar with the dcache code, but is it appropriate to > also d_invalidate() the dentry in this case (as the previous code block > does)? Perhaps Miklos or somebody more familiar with dcache can confirm...
I do not *think* d_invalidate() is needed. The vmcores I have seem where this BUG() happened, only have dentry->d_flags = 0x18 which translates to (DCACHE_OP_DELETE | DCACHE_OP_PRUNE) and d_subdirs as an empty list. d_invalidate() only calls __d_drop(), which only does something when the dentry is hashed.
I am not sure if a dentry can be hashed, but still does not have a valid non-NULL d_inode. If that is the case, d_invalidate() should indeed be called.
Thanks, Niels
> Brian > > > > > dentry = d_alloc(parent, &name); > > >
-- Niels de Vos Sr. Software Maintenance Engineer Support Engineering Group Red Hat Global Support Services
| |