Re: [PATCH 2/2] workqueue: Keep activate-order equals to queue_work()-order

[Tejun Heo <tj@kernel.org>]

Hello, Lai.

On Wed, Sep 19, 2012 at 05:57:12PM +0800, Lai Jiangshan wrote:
> (disorder of execution is OK for current WQ. because we can launch new worker
> to execute the next work if the previous one is waiting something)
> 
> But I concern activate-order, not execution order. A non-delayed work
> may delay a delayed work for ever, and if a non-delayed work needs something
> which will be produced by delayed one, the two work may wait each other.
> 
> {
> a subsystem queues a work to release resource.
> and them
> a subsystem queues a work to use the resource.
> }
> Is this behavior acceptable?

Even on workqueues with a rescuer, forward progress is not guaranteed
if there are more than one co-dependent work items.  workqueue doesn't
guarantee anything regarding activation or execution order and any
user which depends on that is broken.

In general, I think it's a bad idea to give that kind of guarantee and
encourage such usages which can lead to extremely subtle breakages
which cannot be detected in any automated way - we don't have any way
to mark dependencies among work items.

Thanks.

-- 
tejun