Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 13 Mar 2009 13:29:37 -0400
From:      John Baldwin <jhb@FreeBSD.org>
To:        Attilio Rao <attilio@FreeBSD.org>
Cc:        Pawel Jakub Dawidek <pjd@FreeBSD.org>, Mark Powell <M.S.Powell@salford.ac.uk>, John Baldwin <jhb@FreeBSD.org>, "freebsd-current@freebsd.org" <freebsd-current@FreeBSD.org>, Tim Kientzle <kientzle@FreeBSD.org>, Anonymous <swell.k@gmail.com>, Peter Schuller <peter.schuller@infidyne.com>
Subject:   Re: repeatable ZFS panic: share->excl
Message-ID:  <49BA9801.5080505@FreeBSD.org>
In-Reply-To: <3bbf2fe10903122156u650417f0s5c49b68bdf4ffa07@mail.gmail.com>
References:  <20090312175345.Y80227@rust.salford.ac.uk>	 <20090312191333.GA97342@hyperion.scode.org>	 <49B97617.8010709@freebsd.org> <86r6124f2v.fsf@gmail.com>	 <3bbf2fe10903122035i20b2767cod2322c39c6f850ee@mail.gmail.com>	 <29C8FA04-D5B1-49B7-ACF0-4185537367B0@baldwin.cx> <3bbf2fe10903122156u650417f0s5c49b68bdf4ffa07@mail.gmail.com>

next in thread | previous in thread | raw e-mail | index | archive | help
Attilio Rao wrote:
> 2009/3/13, John Baldwin <john@baldwin.cx>:
>> This is similar to the patch I've asked lulf@ to test except that 
>> it is longer and I fix a bug where zfs_lookup() can leak a vnode 
>> lock if the access check fails. :-)  The last one I sent to lulf@ 
>> is at www.FreeBSD.org/~jhb/patches/zfs_ea.patch.
> 
> I really thought zfs_lookup() was returning the lock held as a 
> feature assming no LOCKLEAF. Is that a bug instead?

Yes, I think that is the real bug.  Looking at this further I think
zfs_get_xattrdir() will return the vnode locked if it has to create a
new node via zfs_make_attrdir() but only returns it held and unlocked if
it finds an existing one.  So my new patch is to just fix
zfs_get_xattrdir() to unlock the vnode if it creates a new one like so:

(Sorry, TBird is probably going to butcher all the whitespace):

---
//depot/user/jhb/lock/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_dir.c
+++
/Users/jhb/work/p4/lock/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_dir.c
@@ -940,6 +940,7 @@
  		/* NB: we already did dmu_tx_wait() if necessary */
  		goto top;
  	}
+	VOP_UNLOCK(*xvpp, 0);

  	return (error);
  }

A non-butchered version is at www.FreeBSD.org/~jhb/patches/zfs_ea.patch.

-- 
John Baldwin



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?49BA9801.5080505>