On Mon, 2021-05-10 09:08:47 +0200, Jan-Benedict Glaw <jbglaw%lug-owl.de@localhost> wrote: > On Mon, 2021-05-10 07:18:02 +1000, matthew green <mrg%eterna.com.au@localhost> wrote: > > > Oh and now it's verified: I don't export HAVE_GCC=3D10 anywhere. > > > (Though the initial GCC was a "gcc (Debian 10.2.1-6) 10.2.1 20210110", > > > which is now the system GCC and not the more recent one from the > > > "gcc-snapshot" DEB (which'd identify as "gcc (Debian 20210320-1) > > > 11.0.1 20210320 (experimental) [master revision > > > 3279a9a5a9a:6526c452d22:5f256a70a05fcfc5a1caf56678ceb12b4f87f781]").) > > > > as you previously said, the host compiler _should not_ affect > > what is being built for the target sun2. > > > > just to confirm that this isn't happening, running this will > > tell you what target GCC is being used: > > > > /var/lib/laminar/run/netbsd-sun2-m68000/nn/obj/tooldir.Linux-5.9.0-2-amd64-x86_64/bin/m68010--netbsdelf-gcc -v > > > > (at least, i think that is your path.) [...] > Will now start a "bullseye"-based run. Just as a note of amazement, > I'd like to notice how self-contained the NetBSD build is and how > minimal (gcc, g++, zlib1g-dev and possibly make) the external > dependencies are. Testing in Docker so far: Debian stable "Buster": successful build for tools+release Debian testing "Bullseye": successful build for tools+release Debian unstable "Sid": Next to go! MfG, JBG --
Attachment:
signature.asc
Description: PGP signature