[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAHk-=wi6U90ddnUvh5A0AxBecSPczswRj4hsd2rvKeMv4pUNgQ@mail.gmail.com>
Date: Tue, 27 May 2025 11:37:12 -0700
From: Linus Torvalds <torvalds@...uxfoundation.org>
To: Jonathan Corbet <corbet@....net>
Cc: linux-kernel@...r.kernel.org, linux-doc@...r.kernel.org
Subject: Re: [GIT PULL] Documentation for 6.16
On Tue, 27 May 2025 at 09:01, Jonathan Corbet <corbet@....net> wrote:
>
> scripts/kernel-doc | 2440 +-------------------
> scripts/kernel-doc.pl | 2439 +++++++++++++++++++
> [...]
> 54 files changed, 7336 insertions(+), 2815 deletions(-)
> [...]
> mode change 100755 => 120000 scripts/kernel-doc
> create mode 100755 scripts/kernel-doc.pl
> create mode 100755 scripts/kernel-doc.py
Bah, I was unhappy about git creating this horribly diffstat, but it
turns out that it's because the pull-request scripts by default don't
use the fancier 'git diff' options.
In particular, with the '-B' option, git will give much better results:
scripts/kernel-doc | 2440 +-------------------
scripts/{kernel-doc => kernel-doc.pl} | 2 +-
[...]
54 files changed, 4903 insertions(+), 2821 deletions(-)
[...]
rewrite scripts/kernel-doc (100%)
mode change 100755 => 120000
copy scripts/{kernel-doc => kernel-doc.pl} (99%)
but I think we never enabled the 'break rewrites' option by default
because it generates diffs that I am not convinced non-git sources
understand.
Oh well. I just wanted to point out that git *can* do better than what
the default settings make it do.
Linus
Powered by blists - more mailing lists