Messages in this thread |  | | | From | Roland McGrath <> | | Subject | Re: regression introduced by - timers: fix itimer/many thread hang | | Date | Mon, 17 Nov 2008 13:49:46 -0800 (PST) |
| |
> But please note that currently this already happens for sub-threads (and > if we protect ->signal with rcu too), the exiting sub-thread does not > contribute to accounting after release_task(self). Also, when the last > thread exits the process can be reaped by its parent, but after that > the threads can still use CPU.
Understood (I think I mentioned this earlier). All this is what makes it seem potentially attractive in the long run to reorganize this more thoroughly.
> > Yes, I think you're right. The best solution that comes to mind off hand > > is to protect the update/read of that u64 with a seqcount_t on 32-bit. > > Oh, but we need them to be per-cpu, and both read and write need memory > barriers... Not that I argue, this will fix the problem of course, just > I don't know how this impacts the perfomance.
I agree it's a sticky question. Just the only thing I've thought of so far.
Thanks, Roland
|  |