linux-kernel - Re: [GIT PULL] regmap updates for v6.1

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: <CAHk-=wiVAhwNSNtDB7dXhetBoHHDL70FF=Zx4T72yfT7kdYJ+Q@mail.gmail.com>
Date:   Tue, 4 Oct 2022 19:27:20 -0700
From:   Linus Torvalds <torvalds@...ux-foundation.org>
To:     Mark Brown <broonie@...nel.org>
Cc:     linux-kernel@...r.kernel.org
Subject: Re: [GIT PULL] regmap updates for v6.1

I've pulled this, but...

On Mon, Oct 3, 2022 at 5:46 AM Mark Brown <broonie@...nel.org> wrote:
>
> Andy Shevchenko (11):
>       swab: Add array operations

I wish people didn't add pointless inline functions to core headers
for things that just aren't that important or common.

Partly because it just makes compiles slower. But partly because I
think the code generation is questionable too. Does inlining really
help or matter? No.

And are these actually ever likely to have more than one or two
odd-ball users in the regmap code? No.

IOW, this would have been better off just being kept in regmap code,
and never hit a generic header file that is included by absolutely
*EVERYTHING*.

Now, if would be one thing if this was some complicated
super-optimized code that might actually matter elsewhere too. But it
*really* isn't. It's a completely trivial (and fairly bad)
implementation that is "good enough" for the one single use it has.

IOW, if we actually cared about something like swab16_array() enough
that we'd want to seriously export it, you'd do it a word at a time
(two 8-bit shifts, and a masked combining bitop).

But we clearly don't care.  And for the same reason we don't care
enough to make a really fancy version of it, I don't think it should
be in a generic header file and make every build slower.

                  Linus