[PATCH 2/5] kernel.h: Add non_block_start/end()

[Daniel Vetter]

On Wed, Aug 14, 2019 at 01:45:58PM -0700, Andrew Morton wrote:
> On Wed, 14 Aug 2019 22:20:24 +0200 Daniel Vetter <daniel.vetter at ffwll.ch> wrote:
> 
> > In some special cases we must not block, but there's not a
> > spinlock, preempt-off, irqs-off or similar critical section already
> > that arms the might_sleep() debug checks. Add a non_block_start/end()
> > pair to annotate these.
> > 
> > This will be used in the oom paths of mmu-notifiers, where blocking is
> > not allowed to make sure there's forward progress. Quoting Michal:
> > 
> > "The notifier is called from quite a restricted context - oom_reaper -
> > which shouldn't depend on any locks or sleepable conditionals. The code
> > should be swift as well but we mostly do care about it to make a forward
> > progress. Checking for sleepable context is the best thing we could come
> > up with that would describe these demands at least partially."
> > 
> > Peter also asked whether we want to catch spinlocks on top, but Michal
> > said those are less of a problem because spinlocks can't have an
> > indirect dependency upon the page allocator and hence close the loop
> > with the oom reaper.
> 
> I continue to struggle with this.  It introduces a new kernel state
> "running preemptibly but must not call schedule()".  How does this make
> any sense?
> 
> Perhaps a much, much more detailed description of the oom_reaper
> situation would help out.

I agree on both points, but I guess I'm not the expert to explain why we
have this. All I'm trying to do is that drivers hold up their side. If you
want to have better documentation for why the oom case needs this special
new mode, you're looking at the wrong guy for that :-)

Of course if you folks all decide that you just don't want to be
remembered about that I guess we can drop this one here, but you're just
shooting the messenger I think.
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch