re: HEADS UP: plan to switch many ports over to GCC 12 soon

[matthew green <mrg%eterna23.net@localhost>]

------- Blind-Carbon-Copy

to: Jan-Benedict Glaw <jbglaw%lug-owl.de@localhost>
cc: port-vax%netbsd.org@localhost
subject: re: HEADS UP: plan to switch many ports over to GCC 12 soon
in-reply-to: <20240610040023.GD16219%lug-owl.de@localhost>
from: matthew green <mrg%eterna23.net@localhost>
organisation: people's front against (bozotic) www (softwar foundation)
x-other-organisation: The NetBSD Foundation.
x-message: step back from the shadows of war!  seek out the ways of peace
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <28441.1717994017.1%splode.eterna23.net@localhost>
Date: Mon, 10 Jun 2024 14:33:37 +1000
Message-ID: <12502.1717994017%splode.eterna23.net@localhost>

[ bcc: tech-toolchain%netbsd.org@localhost, current-users%netbsd.org@localhost ]

Jan-Benedict Glaw writes:
> On Mon, 2024-06-10 05:39:27 +1000, matthew green <mrg%eterna23.net@localhost> wrote:
> > at least, arm64, x86, sparc*, ia64, and riscv targets very soon,
> > with arm32 and alpha probably soon after.   more testing needed
> > for m68k, ppc, mips, and hppa.  vax needs the gcc12 version of
> > the Kalvis fixes, sh3 seems pretty bad off.
>
> For VAX, it would be nice to track which patches were necessary to
> make it work, as GCC 12 contains a lot of new VAX patches. So Kalvis's
> patches should also go upstream.

yes - that's the plan.  as i recall, Kalvis published a separate
set of diffs for gcc12, though many of the fixes remain common
between them IIRC.  (it's been a while :).

some of the changes are not in vax-specific code (though it is in
code not used by more than half the GCC ports we use.)

i've been trying to find time to test those changes enabled for
non-vax, to see if any other bugs we've seen there are fixed.
(this is not working so well.)


.mrg.

------- End of Blind-Carbon-Copy