Messages in this thread | | | Date | Sat, 05 Feb 2005 10:15:27 +1100 | From | Nick Piggin <> | Subject | Re: Race condition in ptrace |
| |
Andrew Morton wrote: > Nick Piggin <nickpiggin@yahoo.com.au> wrote: > >>Bodo Stroesser wrote: >> >>>Nick Piggin wrote: >>> >>> >>>>Bodo Stroesser wrote: >> >>>>I don't see how this could help because AFAIKS, child->saving is only >>>>set and cleared while the runqueue is locked. And the same runqueue lock >>>>is taken by wait_task_inactive. >>>> >>> >>>Sorry, that not right. There are some routines called by sched(), that >>>release >>>and reacquire the runqueue lock. >>> >> >>Oh yeah, it is the wake_sleeping_dependent / dependent_sleeper crap. >>Sorry, you are right. And that's definitely a bug in sched.c, because >>it breaks wait_task_inactive, as you've rightly observed. >> >>Andrew, IMO this is another bug to hold 2.6.11 for. > > > Sure. I wouldn't consider Bodo's patch to be the one to use though..
No. Something similar could be done that works on all architectures and all wait_task_inactive callers (and is confined to sched.c). That would still be more or less a hack to work around smtnice's unfortunate locking though.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |