On Mon, Oct 24, 2011 at 04:22:45PM +0200, Roman Divacky wrote: > On Mon, Oct 24, 2011 at 01:53:32PM +0000, Konstantin Belousov wrote: > > Author: kib > > Date: Mon Oct 24 13:53:32 2011 > > New Revision: 226687 > > URL: http://svn.freebsd.org/changeset/base/226687 > > > > Log: > > The only possible error return from null_nodeget() is due to insmntque1 > > failure (the getnewvnode cannot return an error). In this case, the > > null_insmntque_dtr() already unlocked the reclaimed vnode, so VOP_UNLOCK() > > in the nullfs_mount() after null_nodeget() failure is wrong. > > > > Tested by: pho > > MFC after: 1 week > > > > Modified: > > head/sys/fs/nullfs/null_vfsops.c > > > > Modified: head/sys/fs/nullfs/null_vfsops.c > > ============================================================================== > > --- head/sys/fs/nullfs/null_vfsops.c Mon Oct 24 13:48:13 2011 > > (r226686) > > +++ head/sys/fs/nullfs/null_vfsops.c Mon Oct 24 13:53:32 2011 > > (r226687) > > @@ -157,7 +157,6 @@ nullfs_mount(struct mount *mp) > > * Make sure the node alias worked > > */ > > if (error) { > > - VOP_UNLOCK(vp, 0); > > Maybe you want to assert that it's indeed unlocked at this point?
No, I don't, since the vnode pointer is already invalid at this point. The vput() call in dtr released the vnode reference owned by current thread.
pgpbc9pqDE9NH.pgp
Description: PGP signature
