Re: Faster timers for Linux 2.1.22
Keith Owens (kaos@ocs.com.au)
Tue, 28 Jan 1997 09:46:47 +1100
On Mon, 27 Jan 1997 09:54:26 +0000 (GMT),
Alan Cox <alan@cymru.net> wrote:
> Keith Owens wrote
>> As for 'buffering', after disable_bh is called, the next interrupt of
>> any kind including (AFAIK) the timer tick will run any outstanding
>> bottom half code. This takes precedence over user driven code. It's a
>> tradeoff, a more local lock and a slightly delayed bh against a global
>> lock that hits every piece of code.
>
>Thats too expensive for networking and for other things. However what
>Ingo suggested which is having an enable_bh() or variant that does run
>pending bh handlers solves it
So to get back to the original suggestion. Is it worth associating
timers with a specific bh instead of having all timers run under a
global bh?