lkml.org 
[lkml]   [2009]   [Nov]   [20]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
DateFri, 20 Nov 2009 15:24:20 +0900
FromTejun Heo <>
SubjectRe: [PATCH 07/19] acpi: use queue_work_on() instead of binding workqueue worker to cpu0
Hello,

11/20/2009 02:09 PM, Andrew Morton wrote:
>> - kacpid_wq = create_singlethread_workqueue("kacpid");
>> - bind_workqueue(kacpid_wq);
>> - kacpi_notify_wq = create_singlethread_workqueue("kacpi_notify");
>> - bind_workqueue(kacpi_notify_wq);
>> - kacpi_hotplug_wq = create_singlethread_workqueue("kacpi_hotplug");
>> - bind_workqueue(kacpi_hotplug_wq);
>> + kacpid_wq = create_workqueue("kacpid");
>> + kacpi_notify_wq = create_workqueue("kacpi_notify");
>> + kacpi_hotplug_wq = create_workqueue("kacpi_hotplug");
>
> Well that sucks. We create an additional ((num_possible_cpus()-1)*3)
> kernel threads just because the previous code was "unorthodox"?
>
> I guess that problem goes away with concurrency-managed workqueues.
> But please let's not merge this patch without also merging
> concurrency-managed workqueues!

Yeah, it's just prepping the code for the conversion to concurrency
managed workqueues. I'll put this into linux-next for some time
without full conversion but it will never go upstream without it.

Thanks.

--
tejun


\
 
 \ /
  Last update: 2009-11-20 07:27    [from the cache]
©2003-2010