linux-kernel - Re: [PATCH v2] mux: remove the Kconfig question for the subsystem

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-=whWKBKmoq-BOCgFGfZQ45vPP2ujb-wAG8q2PCvjngz3yg@mail.gmail.com>
Date:   Wed, 7 Dec 2022 11:16:52 -0800
From:   Linus Torvalds <torvalds@...ux-foundation.org>
To:     Randy Dunlap <rdunlap@...radead.org>
Cc:     Arnd Bergmann <arnd@...db.de>, Peter Rosin <peda@...ntia.se>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v2] mux: remove the Kconfig question for the subsystem

On Wed, Dec 7, 2022 at 11:03 AM Randy Dunlap <rdunlap@...radead.org> wrote:
> On 12/7/22 10:57, Arnd Bergmann wrote:
> >
> > They cannot, that is the entire point of hiding the subsystem
> > when it is not used.
>
> OK, if you say so. That doesn't make any sense to me, but whatever,
> I'll drop it.

The config phase of the kernel is one of the more annoying parts (the
rest is pretty much "make" and "make install"), and we should strive
to make it as simple and obvious as possible.

One big thing for that is to *not* ask questions that make no sense.

We already hide a number of them, either by disallowing them ("this
driver only makes sense on this architecture" kinds of things) or by
always forcing them on  ("if you're building for a PC, you're getting
the keyboard driver whether you want it or not").

And in many cases, we don't have questions at all, but enable code
based on automatic selection (ie "you wanted ACPI support, so we'll
enable CRC32 support because the ACPI code needs it").

I don't understand *why* you'd ever want more nonsensical questions?

Why would you ever want to ask a human "do you want this code that is
never used"?

That, to me, is actively a horrible thing to do. It only makes the
Kconfig worse. Don't do it.

         Linus