On Fri, Dec 20, 2013 at 6:14 AM, Ingo Molnar <mingo@xxxxxxxxxx> wrote:
>
>
> So, my thinking was that maybe it's the other way around: the problem
> is with the write not being drained to cache yet.
That is possible. I haven't seen the exact details of the monitor
errata . If it's just about the write buffer and cache interaction,
then I agree that the barrier after the write in
current_set_polling_and_test() is sufficient. But in the absense of
knowledge, I think we should have it after the read too.
Linus
--
To unsubscribe from this list: send the line "unsubscribe linux-tip-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
|