Re: [PATCH] - Fix potential race condition in wake_up_forked_process()

[Ingo Molnar]

* Jack Steiner <steiner@xxxxxxx> wrote:

> There appears to be a tiny (microscopic) timing hole in
> wake_up_forked_process(). It is possible that a load_balancing
> application could change cpus_allowed on a task as it is being forked.
> Unlikely, but it can happen, especially if preemption is enabled.

there has been alot of work in this area and sched.c in -BK is much
different from 2.6.8.1's sched.c. In particular there's no
wake_up_forked_process() but wake_up_new_task().

more importantly, the whole bootstrapping of a new task has been
reworked and the set_cpus_allowed() race should not exist in -BK
anymore. Could you double-check to make sure?

	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  http://www.tux.org/lkml/