From: Nick Desaulniers <ndesaulniers@google.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Segher Boessenkool <segher@kernel.crashing.org>,
"Jason A. Donenfeld" <Jason@zx2c4.com>,
linux-kernel@vger.kernel.org, linux-kbuild@vger.kernel.org,
linux-arch@vger.kernel.org, linux-toolchains@vger.kernel.org,
Masahiro Yamada <masahiroy@kernel.org>,
Kees Cook <keescook@chromium.org>,
Andrew Morton <akpm@linux-foundation.org>,
Andy Shevchenko <andriy.shevchenko@linux.intel.com>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Subject: Re: [PATCH] kbuild: treat char as always signed
Date: Wed, 19 Oct 2022 11:10:34 -0700 [thread overview]
Message-ID: <CAKwvOdmDz2VfU1JJkAEnPLTcx4PHH48KfZQfW6gvO6we_QbrRQ@mail.gmail.com> (raw)
In-Reply-To: <CAHk-=whggBoH78ojE0wttyHKwuf48hrSS_X7s3D3Qd_516ayzQ@mail.gmail.com>
On Wed, Oct 19, 2022 at 10:26 AM Linus Torvalds
<torvalds@linux-foundation.org> wrote:
>
> On Wed, Oct 19, 2022 at 10:14 AM Linus Torvalds
> <torvalds@linux-foundation.org> wrote:
> >
> > The pointer-sign thing doesn't actually help (ie it won't find places
> > where you actually compare a char), and it causes untold damage in
> > doing completely insane things.
>
> Side note: several years ago I tried to make up some sane rules to
> have 'sparse' actually be able to warn when a 'char' was used in a
> context where the sign mattered.
Do you have examples? Maybe we could turn this into a compiler feature
request. Having prior art on the problem would be a boon.
>
> I failed miserably.
>
> You actually can see some signs (heh) of that in the sparse sources,
> in that the type system actually has a bit for explicitly signed types
> ("MOD_EXPLICITLY_SIGNED"), but it ends up being almost entirely
> unused.
>
> That bit does still have one particular use: the "bitfield is
> dubiously signed" thing where sparse will complain about bitfields
> that are implicitly (but not explicitly) signed. Because people really
> expect 'int a:1' to have values 0/1, not 0/-1.
Clang's -Wbitfield-constant-conversion can catch that.
commit 5c5c2baad2b5 ("ASoC: mchp-spdiftx: Fix clang
-Wbitfield-constant-conversion")
commit eab9100d9898 ("ASoC: mchp-spdiftx: Fix clang
-Wbitfield-constant-conversion")
commit 37209783c73a ("thunderbolt: Make priority unsigned in struct tb_path")
>
> But the original intent was to find code where people used a 'char'
> that wasn't explicitly signed, and that then had architecture-defined
> behavior.
>
> I just could not come up with any even remotely sane warning
> heuristics that didn't have a metric buttload of false positives.
>
> I still have this feeling that it *should* be possible to warn about
> the situation where you end up doing an implicit type widening (ie the
> normal C "arithmetic is always done in at least 'int'") that then does
> not get narrowed down again without the upper bits ever mattering.
>
> But it needs somebody smarter than me, I'm afraid.
>
> And the fact that I don't think any other compiler has that warning
> either makes me just wonder if my feeling that it should be possible
> is just wrong.
>
> Linus
--
Thanks,
~Nick Desaulniers
next prev parent reply other threads:[~2022-10-19 18:11 UTC|newest]
Thread overview: 70+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-10-19 16:26 [PATCH] kbuild: treat char as always signed Jason A. Donenfeld
2022-10-19 16:54 ` Segher Boessenkool
2022-10-19 17:14 ` Linus Torvalds
2022-10-19 17:26 ` Linus Torvalds
2022-10-19 18:10 ` Nick Desaulniers [this message]
2022-10-19 18:35 ` Linus Torvalds
2022-10-19 19:23 ` Andy Shevchenko
2022-10-19 19:36 ` Linus Torvalds
2022-10-19 17:43 ` Segher Boessenkool
2022-10-19 18:11 ` Linus Torvalds
2022-10-19 18:20 ` Nick Desaulniers
2022-10-19 18:56 ` Linus Torvalds
2022-10-19 19:11 ` Kees Cook
2022-10-19 19:30 ` Linus Torvalds
2022-10-19 20:35 ` Jason A. Donenfeld
2022-10-20 0:10 ` Linus Torvalds
2022-10-20 3:11 ` Jason A. Donenfeld
2022-10-19 20:15 ` Segher Boessenkool
2022-10-19 21:07 ` David Laight
2022-10-19 21:26 ` Segher Boessenkool
2022-10-20 10:41 ` Gabriel Paubert
2022-10-21 22:46 ` Linus Torvalds
2022-10-22 6:06 ` Gabriel Paubert
2022-10-22 18:16 ` Linus Torvalds
2022-10-23 20:23 ` Gabriel Paubert
2022-10-25 23:00 ` Kees Cook
2022-10-26 0:04 ` Jason A. Donenfeld
2022-10-26 15:41 ` Kees Cook
2022-10-19 19:54 ` Linus Torvalds
2022-10-19 20:23 ` Jason A. Donenfeld
2022-10-19 20:30 ` [PATCH v2] kbuild: treat char as always unsigned Jason A. Donenfeld
2022-10-19 23:56 ` Linus Torvalds
2022-10-20 0:02 ` Jason A. Donenfeld
2022-10-20 0:38 ` Linus Torvalds
2022-10-20 2:59 ` Jason A. Donenfeld
2022-10-20 18:41 ` Kees Cook
2022-10-21 1:01 ` Jason A. Donenfeld
2022-10-20 20:24 ` Segher Boessenkool
2022-10-24 9:24 ` Dan Carpenter
2022-10-24 9:30 ` Dan Carpenter
2022-10-24 16:33 ` Jason A. Donenfeld
2022-10-24 17:10 ` Linus Torvalds
2022-10-24 17:17 ` Jason A. Donenfeld
2022-10-25 19:22 ` Kalle Valo
2022-10-25 10:16 ` David Laight
2022-10-24 15:17 ` Jason A. Donenfeld
2022-12-21 14:53 ` Guenter Roeck
2022-12-21 15:05 ` Geert Uytterhoeven
2022-12-21 15:23 ` Guenter Roeck
2022-12-21 15:29 ` Rasmus Villemoes
2022-12-21 15:56 ` Guenter Roeck
2022-12-21 17:06 ` Linus Torvalds
2022-12-21 17:19 ` Guenter Roeck
2022-12-21 18:46 ` Linus Torvalds
2022-12-21 19:08 ` Linus Torvalds
2022-12-21 21:01 ` Guenter Roeck
2022-12-22 13:05 ` Geert Uytterhoeven
2022-12-22 10:41 ` David Laight
[not found] ` <f02e0ac7f2d805020a7ba66803aaff3e31b5eeff.camel@t-online.de>
2022-12-24 9:47 ` Geert Uytterhoeven
2022-12-30 11:39 ` David Laight
2022-12-30 13:13 ` David Laight
2023-01-02 8:29 ` Geert Uytterhoeven
2022-12-21 17:49 ` Andreas Schwab
2022-12-21 16:57 ` Geert Uytterhoeven
2022-10-19 20:58 ` [PATCH] kbuild: treat char as always signed David Laight
2022-10-26 0:10 ` make ctype ascii only? (was [PATCH] kbuild: treat char as always signed) Rasmus Villemoes
2022-10-26 18:10 ` Linus Torvalds
2022-10-27 7:59 ` Rasmus Villemoes
2022-10-27 18:28 ` Linus Torvalds
[not found] ` <202210201618.8XhEGsLd-lkp@intel.com>
2022-10-20 16:33 ` [PATCH] kbuild: treat char as always signed Jason A. Donenfeld
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CAKwvOdmDz2VfU1JJkAEnPLTcx4PHH48KfZQfW6gvO6we_QbrRQ@mail.gmail.com \
--to=ndesaulniers@google.com \
--cc=Jason@zx2c4.com \
--cc=akpm@linux-foundation.org \
--cc=andriy.shevchenko@linux.intel.com \
--cc=gregkh@linuxfoundation.org \
--cc=keescook@chromium.org \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kbuild@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-toolchains@vger.kernel.org \
--cc=masahiroy@kernel.org \
--cc=segher@kernel.crashing.org \
--cc=torvalds@linux-foundation.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).