From: Andreas Dilger <adilger@clusterfs.com>
To: "Henrik Grubbström" <grubba@grubba.org>
Cc: sct@redhat.com, akpm@osdl.org, neilb@cse.unsw.edu.au,
linux-kernel@vger.kernel.org, ext2-devel@lists.sourceforge.net
Subject: Re: [PATCH] Support for dx directories in ext3_get_parent (NFSD)
Date: Mon, 9 May 2005 03:24:17 -0600 [thread overview]
Message-ID: <20050509092417.GF25935@schnapps.adilger.int> (raw)
In-Reply-To: <Pine.GSO.4.21.0505091032000.22820-100000@jms.roxen.com>
[-- Attachment #1: Type: text/plain, Size: 2079 bytes --]
On May 09, 2005 10:57 +0200, Henrik Grubbström wrote:
> The 2.6.10 ext3_get_parent attempts to use ext3_find_entry to look up the
> entry "..", which fails for dx directories since ".." is not present in
> the directory hash table. The patch below solves this by looking up the
> dotdot entry in the dx_root block.
>
> Typical symptoms of the above bug are intermittent claims by nfsd that
> files or directories are missing on exported ext3 filesystems.
>
> cf https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=150759
> and https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=144556
>
> Signed-off-by: Henrik Grubbström <grubba@grubba.org>
ext3_get_parent() is IMHO the wrong place to fix this bug as it introduces
a lot of internals from htree into that function. Instead, I think this
should be fixed in ext3_find_entry() as in the below patch. This has the
added advantage that it works for any callers of ext3_find_entry() and not
just ext3_lookup_parent().
I thought I submitted this patch to l-k at one point, but now I can't find
any mention of that in the archives...
Signed-off-by: Andreas Diler <adilger@clusterfs.com>
--- linux-2.6.orig/fs/ext3/namei.c 2005-04-04 05:06:46.000000000 -0600
+++ linux-2.6/fs/ext3/namei.c 2005-04-04 05:09:18.000000000 -0600
@@ -926,8 +926,16 @@
struct inode *dir = dentry->d_parent->d_inode;
sb = dir->i_sb;
- if (!(frame = dx_probe(dentry, NULL, &hinfo, frames, err)))
- return NULL;
+ /* NFS may look up ".." - look at dx_root directory block */
+ if (namelen > 2 || name[0] != '.'||(name[1] != '.' && name[1] != '\0')){
+ if (!(frame = dx_probe(dentry, NULL, &hinfo, frames, err)))
+ return NULL;
+ } else {
+ frame = frames;
+ frame->bh = NULL; /* for dx_release() */
+ frame->at = (struct dx_entry *)frames; /* hack for zero entry*/
+ dx_set_block(frame->at, 0); /* dx_root block is 0 */
+ }
hash = hinfo.hash;
do {
block = dx_get_block(frame->at);
Cheers, Andreas
--
Andreas Dilger
Principal Software Engineer
Cluster File Systems, Inc.
[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]
next prev parent reply other threads:[~2005-05-09 9:24 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-05-09 8:57 Henrik Grubbström
2005-05-09 9:24 ` Andreas Dilger [this message]
2005-05-09 9:46 ` Henrik Grubbström
2005-05-09 14:05 ` Stephen C. Tweedie
2005-05-09 14:59 ` Henrik Grubbström
[not found] <42chI-3Rf-49@gated-at.bofh.it>
[not found] ` <42cKy-4jT-9@gated-at.bofh.it>
2005-05-09 14:24 ` Bodo Eggert <harvested.in.lkml@posting.7eggert.dyndns.org>
2005-05-09 15:45 ` Stephen C. Tweedie
2005-05-09 16:55 ` Bodo Eggert
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20050509092417.GF25935@schnapps.adilger.int \
--to=adilger@clusterfs.com \
--cc=akpm@osdl.org \
--cc=ext2-devel@lists.sourceforge.net \
--cc=grubba@grubba.org \
--cc=linux-kernel@vger.kernel.org \
--cc=neilb@cse.unsw.edu.au \
--cc=sct@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®