Re: [PATCH 1/2 (repost)] mm: serialize OOM kill operations

[Paul Jackson]

Andrew wrote:
> Note that these will occupy the same machine word.

That's why I did it.  Yup.

> So they'll need
> locking.  (Good luck trying to demonstrate the race though!)

Oops.  Good catch.  Thanks, Andrew.

Probably solvable (lockable) but this line of thought is
getting to be more trouble than I suspect it's worth.

I'm still a little surprised that this per-mm 'oom_notify' bit
was needed to implement what I thought was a single, global
system wide oom killer serializer.

But I'm too ignorant and lazy, and too distracted by other
tasks, to actually think that surprise through.

Good luck with it, Dave.

-- 
                  I won't rest till it's the best ...
                  Programmer, Linux Scalability
                  Paul Jackson <pj@xxxxxxx> 1.925.600.0401
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/