Re: CVS commit: pkgsrc/lang/gcc12

[Jason Bacon <jtocino%gmx.com@localhost>]

On 12/19/24 15:05, Jason Bacon wrote:
> On 12/19/24 14:01, maya%NetBSD.org@localhost wrote:
> > On Thu, Dec 19, 2024 at 10:51:47AM -0700, Brook Milligan wrote:
> > > > On Dec 19, 2024, at 02:41, Jonathan Perkin <jperkin%pkgsrc.org@localhost> wrote:
> > > >
> > > > * On 2024-12-15 at 03:05 GMT, Brook Milligan wrote:
> > > >
> > > > > Module Name:   pkgsrc
> > > > > Committed By:  brook
> > > > > Date:          Sun Dec 15 03:05:12 UTC 2024
> > > > >
> > > > > Modified Files:
> > > > >       pkgsrc/lang/gcc12: distinfo
> > > > >       pkgsrc/lang/gcc12/patches: patch-libgcc_config.host
> > > > > Added Files:
> > > > >       pkgsrc/lang/gcc12/patches: patch-libgcc_config_i386_t-darwin
> > > > >           patch-libgcc_config_rs6000_t-darwin
> > > > >           patch-libgcc_config_t-darwin-libgccs1
> > > > >           patch-libgcc_config_t-darwin-min-11
> > > > >
> > > > > Log Message:
> > > > > The lang/gcc12, lang/gcc13, and lang/gcc14 fail to build on Darwin 24
> > > > > (MacOS 15) + Xcode 16.  This is a known problem that has been fixed
> > > > > upstream; see
> > > >
> > > > This broke the build:
> > >
> > > I have been looking into this.  My cvs is up-to-date, i.e., cvs 
> > > status is all up-to-date, cvs update -PAd changes nothing, cvs diff 
> > > shows nothing, and all the patches work fine for me (just reverified 
> > > on both NetBSD and Darwin).
> > >
> > > Is there any indication of what is broken?  For example, a rejected 
> > > part of the patch or something?
> > >
> > > Sorry about this, but I’m not sure where things went wrong as I’m not 
> > > seeing this.
> > >
> > > Cheers,
> > > Brook
> >
> > There are two things changing libgcc/config.host in lang/gcc12:
> > patches/patch-libgcc_config.host, but also gcc-12.4.0.diff, a
> > downloaded patch file.
> >
> > I don't know what about it triggers the issue, but knowing this helped
> > clarify some of my confusion about why the original file doesn't have
> > some of the context mentionedin the patch.
>
> Commenting out
>
>      tmake_file="${tmake_file} ${cpu_type}/t-lse t-slibgcc-libgcc"
>
> eliminates the error for me.  I don't understand the script, but maybe 
> this info will help someone who does.

Scratch that, there was another part missing that I overlooked.

--
Life is a game.  Play hard.  Play fair.  Have fun.