Re: [patch] Re: setitimer lowlatency-2.2.13-A1 questions

Manfred Spraul (manfreds@colorfullife.com)
Mon, 06 Dec 1999 21:51:30 +0100


Andrea Arcangeli wrote:
>
> >
> >Is there a special reason why do_bottom_half() calls __sti() and __cli()
> >directly? This seems wrong, what about using the normal
> >__save_flags()/__sti()/__restore_flags()?
>
> The irq handler case by design doesn't need it. You must enter there with
> irq disabled and you must exit with irq disabled (so you must enabled
> and disable by hand before and after running the bhs).
>
> For the ret to userspace case it could decrease a bit the bh latency but
> it seems not a big issue, as right now a ret from syscall will act like
> while returning from an irq after the bottom half path.
>
I know that it's safe:
* "real interrupts" need it.
* schedule() calls "spin_lock_irq()" immediately after returning from
do_bottom_half().
* interrupts are cleared during IRET.

But nevertheless I think that it's ugly that a function sometimes (even
worse: rarely) disables interrupts: What if a cpu doesn't reenable
interrupts during return to user space? do_bottom_half() is not
architecture specific.

--
	Manfred

- 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/