Paul Mundt wrote:
On Wed, Jun 29, 2005 at 07:09:38PM +1000, Nick Piggin wrote:
Now this might introduce some power and heat penalty. What's more,
your race isn't a fatal one: in the worst case, it should just
stall until the next timer interrupt (aside, that might be fatal
with a tickless kernel).
After incorporating your changes, how about this?
So, just ignore the race, remove the irq disabling completely?
(with the hlt_counter fallback for busy waiting). If you're
happy with that then I think it looks good. Thanks Paul.
I'll also update the Documentation/ file to point out that such
a race isn't fatal.
--
SUSE Labs, Novell Inc.
Send instant messages to your online friends http://au.messenger.yahoo.com
