Messages in this thread | | | From | David Howells <> | Subject | Re: [PATCHSET] workqueue: concurrency managed workqueue, take#4 | Date | Fri, 12 Mar 2010 11:23:55 +0000 |
| |
Tejun Heo <tj@kernel.org> wrote:
> > So when a work item is running on a CPU, any work items it queues > > (including requeueing itself) will be queued upon that CPU for attention > > only by that CPU (assuming that CPU doesn't get pulled out)? > > Yes, that's the one of the characteristics of workqueue. For IO bound > stuff, it usually is a plus.
Hmmm. So when, say, an FS-Cache index object finishes creating itself on disk and then releases all the several thousand data objects waiting on that event so that they can then create themselves, it will do this by calling queue_work() on each of them in turn. I take it this will stack them all up on that one CPU's work queue.
David
| |