Re: [patch] 2.6.5 Improve interrupt handling on contended spinlocks

[Keith Owens <kaos@sgi.com>]

On Thu, 8 Apr 2004 17:07:44 -0700, 
David Mosberger <davidm@napali.hpl.hp.com> wrote:
>>>>>> On Thu, 08 Apr 2004 13:19:37 +1000, Keith Owens <kaos@sgi.com> said:
>
>  Keith> spin_lock_irqsave() on a contended lock leaves interrupts disabled
>  Keith> while waiting for the lock.  If the interrupts were enabled before
>  Keith> spin_lock_irqsave then this has the nasty side effect of preventing
>  Keith> interrupt handling while waiting for the lock.  Change the contention
>  Keith> path to enable interrupts if it is safe to do so, allowing the cpu to
>  Keith> process interrupts while waiting for the lock.
>
>Seems like a reasonable thing to me.  I'm fine with the ia64-specific
>bits, so I guess the only question is whether the linux/spinlock.h
>bits will be accepted.

Catch 22.  The ia64 code cannot be changed without the change to
linux/spinlock.h, that change cannot be justified without at least one
architecture using it.  David, could you send the change to Linus as an
ia64 patch, other architectures can define _raw_spin_lock_flags if they
want it.