|
|
* Bill Davidsen <davidsen@xxxxxxx> wrote:
> >What are you suggesting here, that the implementation of
> >FUTEX_WAKE_OP is "ignoring inefficiencies"? Please explain why and
> >what you would do differently to solve that inefficiency.
>
> I am suggesting that "There are no such situations anymore in an
> optimal userlevel implementation" is not the right approach to the
> original post. What I would do differently is to evaluate the original
> suggestion to see if it would in fact be more efficient. [...]
that's what we did a year ago for a similar futex rescheduling
inefficiency, and came up with the FUTEX_WAKE_OP solution, which is now
part of the kernel and is used by glibc. But we are all ears for
alternative suggestions.
Ingo
-
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 www.tux.org/lkml/">http://www.tux.org/lkml/
|
|