Re: linux interrupt handling problem

Linus Torvalds (torvalds@transmeta.com)
Tue, 9 Nov 1999 10:59:49 -0800 (PST)

On Tue, 9 Nov 1999, Roman Zippel wrote:
>
> 1. I had such a patch already, but it was rejected by Jes for some reason.
> 2. It adds an extra check to an often used macro for spinlocks for only a
> few cases that could be avoided completly.

Hmm..

I don't think it is used in spinlocks ever. At least it SHOUDLN'T be used.

The spinlock code uses "local_irq_save()" and "local_irq_restore()", both
of which should be fine in all the contexts you're talking about (they'll
never end up re-enabling interrupts).

The "unlock_irq()" version is basically almost never used, so yes, while
it woul dget slightly longer it shouldn't much matter.

Linus

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