Re: poudriere loop: llvm19-19.1.7: missed shlib PORTREVISION chase

On Feb 1, 2025, at 03:04, Nuno Teixeira <eduardo@freebsd.org> wrote:

> Hello Mark!
> 
> Note: I do not know why, but aarch64 does not get MULTILIB to
> also span armv7 (aarch32). So aarch64 might not have this
> problem being visible as stands.
> 
> I confirm that aarch64 doesn't have this problem.

This suggests the test of trying to use poudriere-devel with
PKG_NO_VERSION_FOR_DEPS=yes to build, say, gcc13 with MULTILIB
disabled for the target architecture(s) that would otherwise
have it enabled --in order to see if things then work overall,
absent the 32-bit support being built for the 64-bit target
archtecture.

If there are more lang/gcc*'s with MULTILIB normally enabled
that are normally being built, you might want each such have
MULTILIB disabled so that nothing attempts to build 32-bit
support for a 64-bit target and then see if everything works
overall for such a context.

If it worked, then the thing to be fixed would be handling of
contexts were 64-bit targets also build some 32-bit materials.
I wonder if MULTILIB like contexts are the only examples.

I'll note that the existence of MULTILIB as an OPTION for
amd64 and powerpc64 is based on the test structure:

.if exists(/usr/lib32/libc.so)
. . .
.else
. . .
.endif

So, if the poudriere(-devel) jail does not have that file,
then MULTILIB will not be involved in the build.

> main-n275011-dbedcc169f70
> poudriere-devel
> PKG_NO_VERSION_FOR_DEPS=yes
> 
> `pkg check -dn` shows no conflicts
> 


===
Mark Millard
marklmi at yahoo.com