Messages in this thread |  | | | Date | Tue, 9 Jan 2007 02:15:51 -0800 | | From | Andrew Morton <> | | Subject | Re: [PATCH] flush_cpu_workqueue: don't flush an empty ->worklist |
| |
On Tue, 9 Jan 2007 15:39:26 +0530 Srivatsa Vaddagiri <vatsa@in.ibm.com> wrote:
> I just hope the latency of freeze_processes() is tolerable ..
It'll be roughly proportional to the number of processes, I guess: if we have 100,000 processes (or threads) doing sleep(1000000) then yeah, it'll take some time to wake them all up and capture them in refrigerator(). But I suspect a suitable fix for any problems which arise there is to implement gang-offlining and onlining, rather than the present one-cpu-at-a-time. That'd be pretty simple to do: we already have sysfs interfaces which take a cpumask. - 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/
|  |