Received-SPF: pass (google.com: domain of linux-kernel+bounces-48803-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Precedence: bulk
MIME-Version: 1.0
References: <CAK7LNASGqfMkTuzP28qydpYCC0ct3cAgMpbPpmgHuQHZbtLhbA@mail.gmail.com>
In-Reply-To: <CAK7LNASGqfMkTuzP28qydpYCC0ct3cAgMpbPpmgHuQHZbtLhbA@mail.gmail.com>
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Thu, 1 Feb 2024 12:06:19 -0800
Message-ID: <CAHk-=whJKZNZWsa-VNDKafS_VfY4a5dAjG-r8BZgWk_a-xSepw@mail.gmail.com>
Subject: Re: [GIT PULL] Kbuild fixes for v6.8-rc3
To: Masahiro Yamada <masahiroy@kernel.org>
Cc: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>, 
	Linux Kbuild mailing list <linux-kbuild@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"

On Thu, 1 Feb 2024 at 05:40, Masahiro Yamada <masahiroy@kernel.org> wrote:
>
>  - Replace tabs with spaces when followed by conditionals for
>    future GNU Make versions

This is horrid.

Now, the whole "whitespace type matters" is broken in Make anyway, so
clearly this is a fundamental make problem, but this commit makes
things worse by making the tab replacement use eight spaces, so it
really visually is entirely indistinguishable.

Don't make a 'make' problem worse by not visually distinguishing tabs
from spaces.

IOW, those "that can't be a tab" cases should have used pretty much
_anything_ but 8 spaces. Yes on indentation of nested 'if' statements,
but no on then using something that visually makes no sense.

IOW, those nested if-statements should use perhaps just 2-4 spaces
instead. That tends to match what we sometimes see in C files too, and
it is visually very clearly not a tab with the kernel coding rules
(yes, yes, some people set tabstops to smaller values, that's _their_
problem).

I've pulled this, but please fix it, and don't make an insane Makefile
whitespace situation worse.

                Linus