On Wed, 17 Nov 2021 16:17:26 -0700 Jens Axboe [off-list ref] wrote:

On 11/17/21 3:26 PM, Johannes Weiner wrote:

quoted

quoted

Link: https://lkml.kernel.org/r/20211028080813.15966-1-sir@cmpwn.com
Signed-off-by: Drew DeVault <redacted>
Acked-by: Jens Axboe <axboe@kernel.dk>
Acked-by: Cyril Hrubis <chrubis@suse.cz>
Cc: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

Acked-by: Johannes Weiner <hannes@cmpxchg.org>

As per above, I think basing it off of RAM size would be better, but
this increase is overdue given all the new users beyond mlock(), and
8M is much better than the current value.

That's basically my reasoning too. Let's just get something going that
will at least unblock some valid use cases, and not get bogged down with
aiming for perfection. The latter can happen in parallel, but it should
not hold it up imho.

Nobody's aiming for perfection.  We're discussing aiming for "better".

What we should have done on day one was to set the default MLOCK_LIMIT
to zero bytes.  Then everyone would have infrastructure to tune it from
userspace and we wouldn't ever have this discussion.