Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 29 Jan 2025 02:55:12 +0000
From:      Alexey Dokuchaev <danfe@freebsd.org>
To:        Dimitry Andric <dim@freebsd.org>
Cc:        Jessica Clarke <jrtc27@freebsd.org>, "src-committers@freebsd.org" <src-committers@freebsd.org>, "dev-commits-src-all@freebsd.org" <dev-commits-src-all@freebsd.org>, "dev-commits-src-main@freebsd.org" <dev-commits-src-main@freebsd.org>
Subject:   Re: git: c3450ad127e9 - main - clang: re-downgrade implicit int/function declarations to warning only
Message-ID:  <Z5mYkCwXYsH-CsdQ@FreeBSD.org>
In-Reply-To: <90D8C9F1-0634-44E5-B5B5-20455A91E042@FreeBSD.org>
References:  <202306222111.35MLBAqB066211@gitrepo.freebsd.org> <E1BA0D85-6246-4901-B7F4-B2062A54202A@freebsd.org> <90D8C9F1-0634-44E5-B5B5-20455A91E042@FreeBSD.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Jan 29, 2025 at 12:20:51AM +0100, Dimitry Andric wrote:
> On 28 Jan 2025, at 19:32, Jessica Clarke wrote:
> > On 22 Jun 2023, at 22:11, Dimitry Andric wrote:
> >> 
> >>   clang: re-downgrade implicit int/function declarations to warning only
> >> 
> >>   [...]
> >>   As noted in the upstream discussion, there are many programs that fail
> >>   to configure or build correctly, if these warnings are turned into
> >>   errors by default.
> >> 
> >>   Note that most affected programs in ports are relatively old, and are
> >>   unlikely to be fixed by actually adjusting their declarations, but by
> >>   compiling with -std=gnu89, which downgrades the errors back to warning
> >>   again. Lots of tedious work for very little gain.
> > 
> > This remains as a downstream divergence 1.5 years later, but I do not
> > think we should be carrying this indefinitely. Software needs to build
> > with upstream LLVM, and the fact that many ports do not is now biting
> > us for CheriBSD where we use our own external toolchain that does not
> > have such a diff. In my opinion this should be reverted for FreeBSD 15
> > and it's long time for ports people to step up or have the broken ports
> > removed; we can't keep going on like this with abandonware holding back
> > forward progress.
> 
> While I agree, at the least a full exp-run should be done, and even then
> you can't be entirely sure if ports that build are built the same as
> before, for example due to bad configure tests. So this is potentially a
> massive undertaking.

Can we actually do an exp-run to identify affected ports and at least get
the idea of the fallout size?  I've recently hunted one such subtle bug
exposed by GCC 14 so yeah I fear there'd be a lot, but we need to start with
something.  I presume fixing it won't be particularly hard, just tedious.

./danfe



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Z5mYkCwXYsH-CsdQ>