Re: Fwd: [PATCH] [RFC] ataflop: remove buggy IRQ disable from do_fd_request()

[Jiri Kosina <jkosina@suse.cz>]

On Sun, 18 Oct 2009, Michael Schmitz wrote:

> > The patch is not removing any locking. It only
> > 
> > 1) removes the local_irq_disable() that has been commented out for many 
> >    years already anyway
> > 2) removes the saving and restoring of CPU flags around do_fd_request(), 
> >    which is rather clearly a nop than any kind of "locking"
> > 
> > > > [1] http://lkml.org/lkml/2002/12/27/58
> > > > 
> > > > Signed-off-by: Jiri Kosina <jkosina@suse.cz>
> > > 
> > > NAck for my part. 
> > 
> > Please elaborate a little bit more which of the two points above you base 
> > your NACK on.
> 
> The removal of local_irq_disable() (which should have been local_irq_enable()) 
> just raised a flag, and I didn't immediately see why the interrupt enable had 
> been commented out. 

Yes, it has been commented out in a very non-intuitive way.

> With a bit of further thought on the matter I am satisfied that this patch will 
> not impact on driver function at all, and do not wish to sustain my objection.
> 
> IOW: Ack, and my sincere apologies for wasting your time. 

Thanks, I have added

	Acked-by: Michael Schmitz <schmitz@biophys.uni-duesseldorf.de>

to the patch changelog in my tree.

-- 
Jiri Kosina
SUSE Labs, Novell Inc.