* Nick Piggin <[EMAIL PROTECTED]> wrote:
> We could just do a set_cpus_allowed, or take the lock,
> set_cpus_allowed, and take the new lock, but that's probably a bit
> heavy if we can avoid it. In the interests of speed in this fast path,
> do you think we can do this in sched_fork, before the task has even
> been put on the tasklist?
yeah, that shouldnt be a problem. Technically we set cpus_allowed up
under the tasklist lock just to be non-preemptible and to copy the
parent's _current_ affinity to the child. But sched_fork() is called
just before and if the parent got its affinity changed between the two
calls, so what? I'd move all of this code into sched_fork().
> That would avoid all locking problems. Passing clone_flags into
> sched_fork would not be a problem if we want to distinguish fork() and
> clone(CLONE_VM).
sure, that was the plan all along with sched_fork() anyway. (i think the
initial versions had the flag)
> Yes? I'll cut a new patch to do just that.
sure, fine by me.
Ingo
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
