linux-kernel - Re: Subject: [RFC] clang tooling cleanups

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-=wi0urcZx_TU=U5_hWEFbuwd67QV=7k7H5q7uWtvnV72kg@mail.gmail.com>
Date:   Tue, 27 Oct 2020 12:39:18 -0700
From:   Linus Torvalds <torvalds@...ux-foundation.org>
To:     Nick Desaulniers <ndesaulniers@...gle.com>
Cc:     Tom Rix <trix@...hat.com>, LKML <linux-kernel@...r.kernel.org>,
        clang-built-linux <clang-built-linux@...glegroups.com>,
        linux-toolchains@...r.kernel.org, Joe Perches <joe@...ches.com>,
        Julia Lawall <Julia.Lawall@...6.fr>,
        Masahiro Yamada <masahiroy@...nel.org>,
        Nathan Huckleberry <nhuck15@...il.com>
Subject: Re: Subject: [RFC] clang tooling cleanups

On Tue, Oct 27, 2020 at 11:42 AM Nick Desaulniers
<ndesaulniers@...gle.com> wrote:
>
> Most recently Joe sent a treewide fix for section attributes that
> Linux pulled just after the merge window closed, IIUC.  Maybe that
> would be the best time, since automation makes it trivial for anyone
> to run the treewide fixit whenever.

Well, it worked fine timing-wise, but the fact that it wasn't in
linux-next did mean that it then hit a few small snags once merged.

It's not a big deal - I never got the feeling that that patch was
rushed or that Joe hadn't vetted it enough, and it was well worth it,
but I'm pointing that out simply as an example of the model having a
few gotchas.

So avoiding linux-next (in order to avoid merge pain) does have
downsides. And even obvious and 100% automated fixes can cause issues
if there are #ifdef's or other architecture-specific things that then
mean that the extra semicolon might matter after all. Usually for
horribly bad reasons, but still..

So it would be best if this got a lot of multi-architecture (and
multi-config) coverage if it avoids linux-next.

                Linus