public inbox for linux-kbuild@vger.kernel.org
 help / color / mirror / Atom feed
From: Nathan Chancellor <nathan@kernel.org>
To: Fangrui Song <maskray@google.com>,
	Masahiro Yamada <masahiroy@kernel.org>
Cc: Nicolas Schier <nicolas@fjasle.eu>,
	Bill Wendling <morbo@google.com>,
	Justin Stitt <justinstitt@google.com>,
	linux-kbuild@vger.kernel.org, llvm@lists.linux.dev,
	patches@lists.linux.dev, stable@vger.kernel.org
Subject: Re: [PATCH] kbuild: Update ld-version.sh for change in LLD version output
Date: Fri, 5 Jul 2024 09:00:07 -0700	[thread overview]
Message-ID: <20240705160007.GA875035@thelio-3990X> (raw)
In-Reply-To: <CAFP8O3JUgH-tBJtqO-QS0HmO4mrFBE6Dz+tnrBcse=gw_Q_4vQ@mail.gmail.com>

On Thu, Jul 04, 2024 at 02:23:46PM -0700, Fangrui Song wrote:
> On Thu, Jul 4, 2024 at 9:19 AM Nathan Chancellor <nathan@kernel.org> wrote:
> >
> > After [1] in upstream LLVM, ld.lld's version output is slightly
> > different when the cmake configuration option LLVM_APPEND_VC_REV is
> > disabled.
> >
> > Before:
> >
> >   Debian LLD 19.0.0 (compatible with GNU linkers)
> >
> > After:
> >
> >   Debian LLD 19.0.0, compatible with GNU linkers
> >
> > This results in ld-version.sh failing with
> >
> >   scripts/ld-version.sh: 19: arithmetic expression: expecting EOF: "10000 * 19 + 100 * 0 + 0,"
> >
> > because the trailing comma is included in the patch level part of the
> > expression. Remove the trailing comma when assigning the version
> > variable in the LLD block to resolve the error, resulting in the proper
> > output:
> >
> >   LLD 190000
> >
> > With LLVM_APPEND_VC_REV enabled, there is no issue with the new output
> > because it is treated the same as the prior LLVM_APPEND_VC_REV=OFF
> > version string was.
> >
> >   ClangBuiltLinux LLD 19.0.0 (https://github.com/llvm/llvm-project a3c5c83273358a85a4e02f5f76379b1a276e7714), compatible with GNU linkers
> >
> > Cc: stable@vger.kernel.org
> > Fixes: 02aff8592204 ("kbuild: check the minimum linker version in Kconfig")
> > Link: https://github.com/llvm/llvm-project/commit/0f9fbbb63cfcd2069441aa2ebef622c9716f8dbb [1]
> > Signed-off-by: Nathan Chancellor <nathan@kernel.org>
> > ---
> >  scripts/ld-version.sh | 4 +++-
> >  1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/scripts/ld-version.sh b/scripts/ld-version.sh
> > index a78b804b680c..f2f425322524 100755
> > --- a/scripts/ld-version.sh
> > +++ b/scripts/ld-version.sh
> > @@ -47,7 +47,9 @@ else
> >         done
> >
> >         if [ "$1" = LLD ]; then
> > -               version=$2
> > +               # LLD after https://github.com/llvm/llvm-project/commit/0f9fbbb63cfcd2069441aa2ebef622c9716f8dbb
> > +               # may have a trailing comma on the patch version with LLVM_APPEND_VC_REV=off.
> > +               version=${2%,}
> >                 min_version=$($min_tool_version llvm)
> >                 name=LLD
> >                 disp_name=LLD
> >
> > ---
> > base-commit: 22a40d14b572deb80c0648557f4bd502d7e83826
> > change-id: 20240704-update-ld-version-for-new-lld-ver-str-b7a4afbbd5f1
> >
> > Best regards,
> > --
> > Nathan Chancellor <nathan@kernel.org>
> >
> 
> Thanks for catching the issue.
> If we want to minimize the number of special cases, perhaps we can
> adjust `version=${version%-*}` below to
> 
> version=${version%%[^0-9.]*}

Thanks for the suggestion! I think this wants to be

  version=${version%%[!0-9.]*}

because of "If an open bracket introduces a bracket expression as in XBD
RE Bracket Expression, except that the <exclamation-mark> character
('!') shall replace the <circumflex> character ('^') in its role in a
non-matching list in the regular expression notation, it shall introduce
a pattern bracket expression." from the link that you have below.

That does work for me with all the different linker versions that I can
easily access (Arch, Debian, Fedora) along with my own self built
toolchains, so it seems like it should be pretty robust.

Masahiro, would you be okay with me sending a v2 with that change or do
you foresee any issues where it would not be sufficient? I would
probably change the comment to:

  # There may be something after the version, such as a distribution's
  # package release number (2.34-4.fc32) or a comma (like LLD adds
  # before the "compatible with GNU linkers" string), so remove anything
  # that is not a number or a period.

> (POSIX shell doc:
> https://pubs.opengroup.org/onlinepubs/9699919799/utilities/V3_chap02.html#:~:text=Remove%20Largest)
> 
> ${version%%[^0-9.]*} is a simpler form than what glibc uses:
> 
>   "LLD"*)
>   # Accept LLD 13.0.0 or higher
>     AC_CHECK_PROG_VER(LD, $LD, --version,
>                     [LLD.* \([0-9][0-9]*\.[0-9.]*\)],
>                     [1[3-9].*|[2-9][0-9].*],

Cheers,
Nathan

  reply	other threads:[~2024-07-05 16:00 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-07-04 16:18 [PATCH] kbuild: Update ld-version.sh for change in LLD version output Nathan Chancellor
2024-07-04 21:23 ` Fangrui Song
2024-07-05 16:00   ` Nathan Chancellor [this message]
2024-07-06  3:23     ` Fangrui Song
2024-07-06  5:08     ` Masahiro Yamada

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=20240705160007.GA875035@thelio-3990X \
    --to=nathan@kernel.org \
    --cc=justinstitt@google.com \
    --cc=linux-kbuild@vger.kernel.org \
    --cc=llvm@lists.linux.dev \
    --cc=masahiroy@kernel.org \
    --cc=maskray@google.com \
    --cc=morbo@google.com \
    --cc=nicolas@fjasle.eu \
    --cc=patches@lists.linux.dev \
    --cc=stable@vger.kernel.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