linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Arnd Bergmann <arnd@kernel.org>
To: Nick Desaulniers <ndesaulniers@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Rasmus Villemoes <linux@rasmusvillemoes.dk>,
	Nathan Chancellor <nathan@kernel.org>,
	Masahiro Yamada <masahiroy@kernel.org>,
	Joe Perches <joe@perches.com>,
	Stephen Rothwell <sfr@canb.auug.org.au>,
	llvm@lists.linux.dev,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 00/10] raise minimum GCC version to 5.1
Date: Mon, 13 Sep 2021 13:27:03 +0200	[thread overview]
Message-ID: <CAK8P3a3kD4vd6Xiq_L1YKQPd41=pNbXoA59fuZK3dTtShezoKg@mail.gmail.com> (raw)
In-Reply-To: <20210910234047.1019925-1-ndesaulniers@google.com>

On Sat, Sep 11, 2021 at 1:41 AM Nick Desaulniers
<ndesaulniers@google.com> wrote:
>
> commit fad7cd3310db ("nbd: add the check to prevent overflow in
> __nbd_ioctl()")
>
> raised an issue from the fallback helpers added in
>
> commit f0907827a8a9 ("compiler.h: enable builtin overflow checkers and add fallback code")
>
> Specifically, the helpers for checking whether the results of a
> multiplication overflowed (__unsigned_mul_overflow,
> __signed_add_overflow) use the division operator when
> !COMPILER_HAS_GENERIC_BUILTIN_OVERFLOW. This is problematic for 64b
> operands on 32b hosts.
>
> Also, because the macro is type agnostic, it is very difficult to write
> a similarly type generic macro that dispatches to one of:
> * div64_s64
> * div64_u64
> * div_s64
> * div_u64
>
> Raising the minimum supported versions allows us to remove all of the
> fallback helpers for !COMPILER_HAS_GENERIC_BUILTIN_OVERFLOW, instead
> dispatching the compiler builtins.
>
> arm64 has already raised the minimum supported GCC version to 5.1, do
> this for all targets now. See the link below for the previous
> discussion.
>
> Additional patches cleaning up a few obsolete version checks.

Agreed, I think this is a good time for removing gcc-4.9 support, with
the tradeoff between the number of remaining users of that compiler and
the number of problems that are solved in gcc-5.1.

After this, I think we can also change the --std=gnu89 to --std=gnu11,
as an additional benefit.

According to distrowatch, this will lose support for the distro gcc version
of Debian 8 (Jessie), Ubuntu 15.04, and older Android AOSP (new versions
use clang to build kernels, old versions also need older kernels).
I think that is acceptable.

For reference, the following distro releases use

Centos 8: gcc-8.3
Debian 9: gcc-6.3
RHEL 8: gcc-8.4
Slackware-14.2: gcc-5.3
SLES15: gcc-10.2
Ubuntu 16.04: gcc-5.3

Most older releases of these already don't support building current kernels.
I expect the number of users of Ubuntu 16.04 and Slackware-14.2 that
want to build their own mainline kernels instead of running what was
provided by the distro to be really low.

The other case that always gets brought up are embedded users that
want to use an ancient user space that is only validated with an old gcc,
but need a new kernel without revalidating user space. The only answer
for those is to use two different compilers.

       Arnd

      parent reply	other threads:[~2021-09-13 11:27 UTC|newest]

Thread overview: 45+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-10 23:40 [PATCH 00/10] raise minimum GCC version to 5.1 Nick Desaulniers
2021-09-10 23:40 ` [PATCH 01/10] Documentation: raise minimum supported version of GCC " Nick Desaulniers
2021-09-10 23:55   ` Nathan Chancellor
2021-09-11  0:16     ` Nick Desaulniers
2021-09-13 16:22   ` Kees Cook
2021-09-10 23:40 ` [PATCH 02/10] compiler.h: drop fallback overflow checkers Nick Desaulniers
2021-09-11  0:04   ` Nathan Chancellor
2021-09-14 15:33     ` Nick Desaulniers
2021-09-14 16:04       ` Nathan Chancellor
2021-09-13 16:21   ` Kees Cook
2021-09-10 23:40 ` [PATCH 03/10] mm/ksm: remove old GCC 4.9+ check Nick Desaulniers
2021-09-11  0:07   ` Nathan Chancellor
2021-09-13 16:22   ` Kees Cook
2021-09-10 23:40 ` [PATCH 04/10] Kconfig.debug: drop GCC 5+ version check for DWARF5 Nick Desaulniers
2021-09-11  0:08   ` Nathan Chancellor
2021-09-13 16:23   ` Kees Cook
2021-09-10 23:40 ` [PATCH 05/10] riscv: remove Kconfig check for GCC version for ARCH_RV64I Nick Desaulniers
2021-09-11  0:09   ` Nathan Chancellor
2021-09-13 16:23   ` Kees Cook
2021-10-05  0:40     ` Palmer Dabbelt
2021-10-05  0:50       ` Kees Cook
2021-10-05  0:57         ` Palmer Dabbelt
2021-09-10 23:40 ` [PATCH 06/10] powerpc: remove GCC version check for UPD_CONSTR Nick Desaulniers
2021-09-10 23:48   ` Nathan Chancellor
2021-09-11 10:43     ` Michael Ellerman
2021-09-11 15:34   ` Christophe Leroy
2021-09-10 23:40 ` [PATCH 07/10] arm64: remove GCC version check for ARCH_SUPPORTS_INT128 Nick Desaulniers
2021-09-11  0:10   ` Nathan Chancellor
2021-09-13 16:25   ` Kees Cook
2021-09-16 13:23   ` Catalin Marinas
2021-09-10 23:40 ` [PATCH 08/10] Makefile: drop GCC < 5 -fno-var-tracking-assignments workaround Nick Desaulniers
2021-09-11  0:11   ` Nathan Chancellor
2021-09-13 16:25   ` Kees Cook
2021-09-10 23:40 ` [PATCH 09/10] compiler-gcc.h: drop checks for older GCC versions Nick Desaulniers
2021-09-11  0:12   ` Nathan Chancellor
2021-09-13 16:26   ` Kees Cook
2021-09-10 23:40 ` [PATCH 10/10] vmlinux.lds.h: remove old check for GCC 4.9 Nick Desaulniers
2021-09-10 23:50   ` Nathan Chancellor
2021-09-14 15:21     ` Nick Desaulniers
2021-09-13 16:28   ` Kees Cook
2021-09-11  2:19 ` [PATCH 00/10] raise minimum GCC version to 5.1 Kees Cook
2021-09-11 10:42   ` Michael Ellerman
2021-09-13  9:49 ` Pavel Machek
2021-09-13 16:20   ` Kees Cook
2021-09-13 11:27 ` Arnd Bergmann [this message]

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='CAK8P3a3kD4vd6Xiq_L1YKQPd41=pNbXoA59fuZK3dTtShezoKg@mail.gmail.com' \
    --to=arnd@kernel.org \
    --cc=akpm@linux-foundation.org \
    --cc=joe@perches.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux@rasmusvillemoes.dk \
    --cc=llvm@lists.linux.dev \
    --cc=masahiroy@kernel.org \
    --cc=nathan@kernel.org \
    --cc=ndesaulniers@google.com \
    --cc=sfr@canb.auug.org.au \
    --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).