linux-kernel - Re: [GIT PULL] MM updates for 5.20-rc1

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <Yu2wDNcOYE0RdkA3@P9FQF9L96D.corp.robot.car>
Date:   Fri, 5 Aug 2022 17:04:28 -0700
From:   Roman Gushchin <roman.gushchin@...ux.dev>
To:     Linus Torvalds <torvalds@...ux-foundation.org>
Cc:     Andrew Morton <akpm@...ux-foundation.org>,
        Kent Overstreet <kent.overstreet@...il.com>,
        mm-commits@...r.kernel.org, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org, Dave Chinner <dchinner@...hat.com>
Subject: Re: [GIT PULL] MM updates for 5.20-rc1

On Fri, Aug 05, 2022 at 04:32:34PM -0700, Linus Torvalds wrote:
> On Wed, Aug 3, 2022 at 1:35 PM Andrew Morton <akpm@...ux-foundation.org> wrote:
> >
> > - memcg debug/visibility work from Roman Gushchin
> 
> Well, not just memcg.
> 
> There's that new CONFIG_SHRINKER_DEBUG too.
> 
> Which looks like yet another "people add Kconfig options incorrectly".
> 
> We don't make new features "default y" unless there's some truly
> pressing reason for it (ie "99.9% of all people will want this" or "we
> used to always compile this in, now it's optional").

Hi Linus!

It actually was 'default n' in one of the earlier version of the patchset
and has been switched to 'default y' based on the following feedback from
Dave Chinner (https://lore.kernel.org/lkml/YmiWK56bOHyrr64u@rh/):

    No. The argument that "if we turn it off there's no overhead" means
    one of two things:

    1. nobody turns it on and it never gets tested and so bitrots and is
    useless, or
    2. distro's all turn it on because some tool they ship or customer
    they ship to wants it.

    Either way, hiding it behind a config option is not an acceptible
    solution for mering poorly thought out infrastructure.

Personally I think that the feature is not that useful for the majority
of users (this is why default was n), but it's not adding much of the
overhead, so I had no strong reasons to oppose Dave.
Cc'ing him just in case.

Thanks!