[PATCH v2 2/3] mutex: add support for reservation style locks, v2

Peter Zijlstra peterz at infradead.org
Thu Apr 4 09:54:36 PDT 2013


On Thu, 2013-04-04 at 15:31 +0200, Daniel Vetter wrote:
> Well, it was a good read and I'm rather happy that we agree on the
> ww_ctx
> thing (whatever it's called in the end), even though we have slightly
> different reasons for it.

Yeah, I tried various weirdness to get out from under it, but the whole
progress/fairness thing made it rather hard. Ideally you'd be able to
use some existing scheduler state since its the same goal, but the
whole wakeup-retry muck makes that hard.

> I don't really have a useful idea to make the retry handling for users
> more rusty-compliant though, and I'm still unhappy with all current
> naming
> proposals ;-)

Ah, naming,.. yeah I'm not too terribly attached to most of them. I
just want to avoid something that's reasonably well known to mean
something different.

Furthermore, since we use the wound/wait symmetry breaking it would
make sense for that to appear somewhere in the name.





More information about the dri-devel mailing list