Messages in this thread | | | Date | Thu, 22 Sep 2005 14:59:50 +0200 | From | Ingo Molnar <> | Subject | Re: [ANNOUNCE] ktimers subsystem |
| |
* Thomas Gleixner <tglx@linutronix.de> wrote:
> > > This revealed a reasonable explanation for this behaviour. Both > > > networking and disk I/O arm a lot of timeout timers (the maximum number > > > of armed timers during the tests observed was ~400000). > > > > This triggers the obvious question: where are these timers coming from? > > You don't think that having that much timers in first place is little > > insane (especially if these are kernel timers)? > > Quick answer: Networking and disk I/O. Insane load on a 4 way SMP > machine. Check yourself. :)
a busy network server can easily have millions of timers pending. I once had to increase a server's 16 million tw timer sysctl limit ...
Ingo - 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/
| |