Re: [PATCH] mm: reduce spinlock contention in release_pages()

[Michal Hocko <mhocko@suse.com>]

On Mon 29-11-21 13:23:19, Matthew Wilcox wrote:
> On Mon, Nov 29, 2021 at 09:39:16AM +0100, Michal Hocko wrote:
> > On Fri 26-11-21 16:26:23, Hao Lee wrote:
> > [...]
> > > I will try Matthew's idea to use semaphore or mutex to limit the number of BE
> > > jobs that are in the exiting path. This sounds like a feasible approach for
> > > our scenario...
> > 
> > I am not really sure this is something that would be acceptable. Your
> > problem is resource partitioning. Papering that over by a lock is not
> > the right way to go. Besides that you will likely hit a hard question on
> > how many tasks to allow to run concurrently. Whatever the value some
> > workload will very likely going to suffer. We cannot assume admin to
> > chose the right value because there is no clear answer for that. Not to
> > mention other potential problems - e.g. even more priority inversions
> > etc.
> 
> I don't see how we get priority inversions.  These tasks are exiting; at
> the point they take the semaphore, they should not be holding any locks.
> They're holding a resource (memory) that needs to be released, but a
> task wanting to acquire memory must already be prepared to sleep.

At least these scenarios come to mind
- a task being blocked by other lower priority tasks slowly tearing down
  their address space - essentially a different incarnation of the same
  problem this is trying to handle
- a huge memory backed task waiting many for smaller ones to finish
- waste of resources on properly partitioned systems. Why should
  somebody block tasks when they are acting on different lruvecs and
  cpus?

-- 
Michal Hocko
SUSE Labs