Re: [**BULK SPAM**] Re: bd_mount_mutex -> bd_mount_sem (was Re: xfs_file_ioctl / xfs_freeze: BUG: warning at kernel/mutex-debug.c:80/debug_mutex_unlock())

[Nathan Scott <nscott@aconex.com>]

On Tue, 2007-01-09 at 15:49 +1100, David Chinner wrote:
> On Tue, Jan 09, 2007 at 03:17:03PM +1100, Nathan Scott wrote:
> > On Mon, 2007-01-08 at 19:51 -0800, Andrew Morton wrote:
> > > If that's not true, then what _is_ happening in there?
> > 
> > This particular case was a device mapper stack trace, hence the
> > confusion, I think.  Both XFS and DM are making the same generic
> > block layer call here though (freeze_bdev).
> 
> Yup. it's the freeze_bdev/thaw_bdev use of the bd_mount_mutex()
> that's the problem. I fail to see _why_ we need to hold a lock
> across the freeze/thaw - the only reason i can think of is to
> hold out new calls to sget() (via get_sb_bdev()) while the
> filesystem is frozen though I'm not sure why you'd need to
> do that. Can someone explain why we are holding the lock from
> freeze to thaw?

Not me.  If it's really not needed, then...

> > > If that _is_ true then, well, that sucks a bit.
> > 
> > Indeed, its a fairly ordinary interface, but thats too late to go
> > fix now I guess (since its exposed to userspace already). 
> 
> Userspace knows nothing about that lock, so we can change that without
> changing the the userspace API.

...that would be true, AFAICS.

cheers.

-- 
Nathan