Remove the spinlock protecting the pages allocation
On Mon, 23 May 2011, Joe Thornber wrote:
> On Thu, 2011-05-19 at 12:58 -0400, Mikulas Patocka wrote:
> > The spinlock is only taken on initialization or from single-threaded
> > workqueue.
> > Therefore, the spinlock is useless.
> The spinlock has memory barrier semantics as well. How are you
> guaranteeing the changes from the initialisation thread are visible in
> the workqueue thread?
There is another spinlock, kc->job_lock, that is taken each time someone
pushes or pops some work for the worker thread.
Once we take kc->job_lock, we guarantee that any written memory is
visible to the other CPUs.
> - Joe
> dm-devel mailing list
dm-devel mailing list