Re: [PATCH] ide: replace the global ide_lock spinlock by per-hwgroup spinlocks

[Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>]

On Sunday 12 October 2008, Elias Oltmanns wrote:
> Bartlomiej Zolnierkiewicz <bzolnier@gmail.com> wrote:
> > From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
> > Subject: [PATCH] ide: replace the global ide_lock spinlock by per-hwgroup spinlocks
> >
> > Now that (almost) all host drivers have been fixed not to abuse ide_lock
> > and core code usage of ide_lock has been sanitized we may safely replace
> > ide_lock by per-hwgroup locks.
> >
> > This patch is partially based on earlier patch from Ravikiran G Thirumalai.
> >
> > While at it:
> > - don't use deprecated HWIF() and HWGROUP() macros
> > - update locking documentation in ide.h
> >
> > Cc: Vaibhav V. Nivargi <vaibhav.nivargi@gmail.com>
> > Cc: Alok N. Kataria <alokk@calsoftinc.com>
> > Cc: Ravikiran Thirumalai <kiran@scalex86.org>
> > Cc: Shai Fultheim <shai@scalex86.org>
> > Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
> > ---
> > this is against 2.6.27 + pata tree + pre-patchset posted on Wednesday
> > (http://lkml.org/lkml/2008/10/8/221)
> 
> I've only had a casual look at this patch, but there is one thing:
> 
> [...]
> > Index: b/drivers/ide/ide-probe.c
> > ===================================================================
> > --- a/drivers/ide/ide-probe.c
> > +++ b/drivers/ide/ide-probe.c
> [...]
> > @@ -1091,11 +1092,11 @@ static int init_irq (ide_hwif_t *hwif)
> >  		 * linked list, the first entry is the hwif that owns
> >  		 * hwgroup->handler - do not change that.
> >  		 */
> > -		spin_lock_irq(&ide_lock);
> > +		spin_lock_irq(&hwgroup->lock);
> >  		hwif->next = hwgroup->hwif->next;
> >  		hwgroup->hwif->next = hwif;
> >  		BUG_ON(hwif->next == hwif);
> > -		spin_unlock_irq(&ide_lock);
> > +		spin_unlock_irq(&hwgroup->lock);
> >  	} else {
> >  		hwgroup = kmalloc_node(sizeof(*hwgroup), GFP_KERNEL|__GFP_ZERO,
> >  				       hwif_to_node(hwif));
> 
> Something like
> 
> 	spin_lock_init(&hwgroup->lock);
> 
> should go into this else clause too.

Thanks!

I fixed this in v2, obvious interdiff:

...
v2:
Add missing spin_lock_init(&hwgroup->lock).  (Noticed by Elias Oltmanns)
...

diff -u b/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
--- b/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1103,6 +1103,8 @@
 		if (hwgroup == NULL)
 			goto out_up;
 
+		spin_lock_init(&hwgroup->lock);
+
 		hwif->hwgroup = hwgroup;
 		hwgroup->hwif = hwif->next = hwif;