Re: [PATCH][ATM] assorted he driver cleanup

[Alan Cox]

On Sul, 2003-06-01 at 19:57, chas williams wrote:
> In message <[e-mail]>,Arnaldo Carvalho de Melo writes:
> >Sure thing, but hey, spin_lock_irqsave and friends take care of how to behave
> >when in up or smp, i.e. its how all the other drivers use spinlocks 8)
> 
> but on a single processor machine (i.e. #undef CONFIG_SMP) there is no
> chance that there will be reads/writes from other processors so i dont
> need any locking OR protection from interrupts.  so the degenerate case
> of spin_lock_irqsave() isnt quite as dengerate as i would like for this
> particular spin lock.

Then why are you using spin_lock_irqsave ?