Re: aio stress panic on 2.6.11-mm1

[Andrew Morton]

Arjan van de Ven <arjan@xxxxxxxxxxxxx> wrote:
>
> On Wed, 2005-03-09 at 16:34 +0530, Suparna Bhattacharya wrote:
> > Any sense of how costly it is to use spin_lock_irq's vs spin_lock
> > (across different architectures)
> 
> on x86 it makes a difference of maybe a few cycles. At most.
> However please consider using spin_lock_irqsave(); the _irq() variant,
> while it can be used correctly, is a major source of bugs since it
> unconditionally enables interrupts on unlock.
> 

spin_lock_irq() is OK for down_*(), since down() can call schedule() anyway.

spin_lock_irqsave() should be used in up_*() and I guess down_*_trylock(),
although the latter shouldn't need to go into the slowpath anyway.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/