Prev: [PATCH 0/5] blkdev: discard optimizations v2 RESEND
Next: [PATCH 4/5] powernow-k8: Fix frequency reporting
From: Al Viro on 25 Mar 2010 07:10
On Thu, Mar 25, 2010 at 12:56:40PM +0200, Benny Halevy wrote:
> - if (*want_dir && !path->dentry->d_inode->i_op->lookup)
> + if (*want_dir && !path->dentry->d_inode->i_op->lookup) {
> + WARN_ON(1);
> goto exit_dput;
> + }
> path_to_nameidata(path, nd);
> audit_inode(pathname, nd->path.dentry);
> goto ok;
>
> but it is NOT tripping for this scenario.
>
> (for some reason I saw it tripping when building the kernel over nfs
> but it's benign)

It's ENOTDIR, not EISDIR, anyway. Happens if you ask to open foo/ or
foo with O_DIRECTORY when foo is not a directory.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Benny Halevy on 25 Mar 2010 07:20
On Mar. 25, 2010, 13:12 +0200, Benny Halevy <bhalevy(a)panasas.com> wrote:
> On Mar. 25, 2010, 13:00 +0200, Al Viro <viro(a)ZenIV.linux.org.uk> wrote:
>> On Thu, Mar 25, 2010 at 12:56:40PM +0200, Benny Halevy wrote:
>>> - if (*want_dir && !path->dentry->d_inode->i_op->lookup)
>>> + if (*want_dir && !path->dentry->d_inode->i_op->lookup) {
>>> + WARN_ON(1);
>>> goto exit_dput;
>>> + }
>>> path_to_nameidata(path, nd);
>>> audit_inode(pathname, nd->path.dentry);
>>> goto ok;
>>>
>>> but it is NOT tripping for this scenario.
>>>
>>> (for some reason I saw it tripping when building the kernel over nfs
>>> but it's benign)
>> It's ENOTDIR, not EISDIR, anyway. Happens if you ask to open foo/ or
>> foo with O_DIRECTORY when foo is not a directory.
>
> Hmm, not according to my strace (or am I blind today? :)

Duh, you were referring to the hunk I added a WARN_ON to.
Sure. you're right!

>
>>> open(".git/objects/pack", O_RDONLY|O_NONBLOCK|O_DIRECTORY|0x80000) = -1 EISDIR (Is a directory)
>
> The odd thing is the app incorrectly gets EISDIR despite
> the fact that it is opening a directory with O_DIRECTORY.
> _______________________________________________
> pNFS mailing list
> pNFS(a)linux-nfs.org
> http://linux-nfs.org/cgi-bin/mailman/listinfo/pnfs
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Benny Halevy on 25 Mar 2010 07:20
On Mar. 25, 2010, 12:54 +0200, Al Viro <viro(a)ZenIV.linux.org.uk> wrote:
> On Thu, Mar 25, 2010 at 10:12:31AM +0000, Al Viro wrote:
>> On Thu, Mar 25, 2010 at 11:39:38AM +0200, Boaz Harrosh wrote:
>>
>>> It makes no difference, fails just the same. Would an "strace" help?
>> It might, especially if you ran it for identical repositories on local
>> fs and on NFS; at least that way it would be possible to see where do
>> they diverge...
>
> I wonder... What happens if you add
> if (error == -EISDIR)
> printk("blah: %s", pathname);

I'll add that check too.

Benny

> right after do_lookup() call in do_last()? That would separate -EISDIR
> coming from NFS from the same thing coming from fs/namei.c...
> _______________________________________________
> pNFS mailing list
> pNFS(a)linux-nfs.org
> http://linux-nfs.org/cgi-bin/mailman/listinfo/pnfs
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Al Viro on 25 Mar 2010 08:00
On Thu, Mar 25, 2010 at 10:49:24AM +0000, Al Viro wrote:

> Does that happen with nfsv3 or is that v4-only? I'm going to set up v4
> server and client and see what happens, but that information could be
> useful...

*grumble*
I'd managed to forget that rpc.imapd in its infinite wisdom relies on
the FPOS I normally disable. Oh, well... rebuilding testbox kernel
with dnotify enabled ;-/
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Benny Halevy on 25 Mar 2010 08:10
On Mar. 25, 2010, 12:54 +0200, Al Viro <viro(a)ZenIV.linux.org.uk> wrote:
> On Thu, Mar 25, 2010 at 10:12:31AM +0000, Al Viro wrote:
>> On Thu, Mar 25, 2010 at 11:39:38AM +0200, Boaz Harrosh wrote:
>>
>>> It makes no difference, fails just the same. Would an "strace" help?
>> It might, especially if you ran it for identical repositories on local
>> fs and on NFS; at least that way it would be possible to see where do
>> they diverge...
>
> I wonder... What happens if you add
> if (error == -EISDIR)
> printk("blah: %s", pathname);
> right after do_lookup() call in do_last()? That would separate -EISDIR
> coming from NFS from the same thing coming from fs/namei.c...

Bingo.
It is returned from do_lookup.

@@ -1648,6 +1654,8 @@ static struct file *do_last(struct nameidata *nd, struct path *path,
/* just plain open? */
if (!(open_flag & O_CREAT)) {
error = do_lookup(nd, &nd->last, path);
+ if (error == -EISDIR)
+ printk("%s: do_lookup returned -EISDIR %s\n", __func__, pathname);
if (error)
goto exit;
error = -ENOENT;


> _______________________________________________
> pNFS mailing list
> pNFS(a)linux-nfs.org
> http://linux-nfs.org/cgi-bin/mailman/listinfo/pnfs
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
First  |  Prev  |  Next  |  Last
Pages: 1 2 3 4 5 6 7
Prev: [PATCH 0/5] blkdev: discard optimizations v2 RESEND
Next: [PATCH 4/5] powernow-k8: Fix frequency reporting