Re: [PATCH] perf_events: improve Intel event scheduling

[Paul Mackerras]

On Mon, Dec 21, 2009 at 09:59:45PM +0100, Stephane Eranian wrote:

> Ok, so what you are suggesting is that the assignment is actually done
> incrementally in ->enable(). hw_group_sched_in() would simply validate
> that a single group is sane (i.e., can be scheduled if it was alone).

No, hw_group_sched_in needs to validate that this group can go on
along with everything else that has already been enabled.  But as I
have said, if you have the complete list of enabled events to hand,
that's not hard.

On the other hand, hw_perf_event_init does need to validate that a
single group is sane by itself.

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