Re: [PATCH v7 1/4] spinlock: A new lockref structure for locklessupdate of refcount

[Waiman Long]

On 08/30/2013 04:26 PM, Al Viro wrote:
> On Fri, Aug 30, 2013 at 03:52:49PM -0400, Waiman Long wrote:
>
> > So it is the mnput_no_expire() function that is doing all the
> > lg_global_lock() calls.
> Interesting...  So you are getting a lot of mntput() with ->mnt_ns being
> NULL?  I wonder which type it is...  Note that anything mounted will
> have non-NULL ->mnt_ns until umount and anything obtained via
> kern_mount/kern_mount_data will also have a non-NULL ->mnt_ns - until
> kern_unmount().
>
> Could you try to gather stats of that sort?  Normally that path should
> be only hit by fput() when we have a lazy-unmounted fs and close an opened
> file on it...
>
> I see one potential stupidity in that area (simple_pin_fs() ought to set
> ->mnt_ns, with simple_release_fs() clearing it), but there's not a lot
> of fs types that would use the damn thing...

The AIM7 test was run on a set of 16 ramdisk formated with ext3 
filesystem with the following mount options: 
barrier=0,async,noatime,nodiratime. Maybe that is a factor.

Regards,
Longman
--
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/