Re: [pnfs] [GIT BISECT] first bad commit: 1f36f774 Switch !O_CREATcase to use of do_last()

[Benny Halevy]

On Mar. 25, 2010, 16:07 +0200, Benny Halevy <bhalevy@xxxxxxxxxxx> wrote:
> On Mar. 25, 2010, 16:06 +0200, Al Viro <viro@xxxxxxxxxxxxxxxxxx> wrote:
>> On Thu, Mar 25, 2010 at 03:52:25PM +0200, Benny Halevy wrote:
>>> On Mar. 25, 2010, 15:37 +0200, Al Viro <viro@xxxxxxxxxxxxxxxxxx> wrote:
>>>> On Thu, Mar 25, 2010 at 03:30:22PM +0200, Boaz Harrosh wrote:
>>>>>> Let's try this: before do_lookup() call there add
>>>>>> 	if (*want_dir)
>>>>>> 		nd->flags |= LOOKUP_DIRECTORY;
>>>>> Yes this fixes it!!
>>>>> 2.6.34-rc2 plus above, now works, horay. (diff attached)
>>>>>
>>>>>> and see how does it behave.
>>>>>>
>>>>>> However, even if it does help, it doesn't explain everything.  Normal
>>>>>> open() on a directory without O_DIRECTORY if flags shouldn't fail with
>>>>>> -EISDIR.  How did that manage to avoid it all along?
>>>> Does open() of directory _without_ O_DIRECTORY work in e.g. vanilla 2.6.33?
>>>> It certainly does for local filesystems and it does for NFSv3; does it work
>>>> for NFSv4?
>>> No, it doesn't.
>>>
>>> # mount localhost:/usr0/nfs4export /mnt/localhost; strace cat /mnt/localhost/server 2>&1 | grep 'open.*server'; umount /mnt/localhost
>>> open("/mnt/localhost/server", O_RDONLY) = 3
>>>
>>> # mount -t nfs4 localhost:/ /mnt/localhost; strace cat /mnt/localhost/server 2>&1 | grep 'open.*server'; umount /mnt/localhost
>>> open("/mnt/localhost/server", O_RDONLY) = -1 EISDIR (Is a directory)
>> Gets better - if you do ls -l /mnt/localhost/server and then repeat that
>> open(), it'll succeed.
> 
> Correct :)

So, previously (pre 1f36f77), do_last called do_path_lookup with
lookup_flags(open_flag)|LOOKUP_OPEN and lookup_flags did
-	if (f & O_DIRECTORY)
-		retval |= LOOKUP_DIRECTORY;

and this was dropped in 1f36f77, right?

Benny

> _______________________________________________
> pNFS mailing list
> pNFS@xxxxxxxxxxxxx
> 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@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/