Re: NetBSD/vax vm for github testing
Jan-Benedict Glaw <[email protected]>
| Newsgroups | gmane.os.netbsd.ports.vax |
|---|---|
| Message-ID | <[email protected]> |
On Mon, 2025-01-06 19:29:32 +0000, John Klos <[email protected]> wrote: > > However it takes 24 minutes even to build ancient small emacs-18.59. > > simh emulating microvax3900 on my i5-4570 NetBSD/amd64 took ~70 hours > > to build perl 5.38.2 (from pkgsrc-2023Q4) so it won't finish in > > 6 hour github CI limit. > > I suppose because additional cores can't be thrown at simh, this > is a difficult limit. > > What do others do when something comes close to that limit? Do > they just add cores and increase -j? Just wait for it. I personally prefer to not do parallel builds as I quite often compare build logs. That just won't fly... I haven't tried that, but maybe it would be worth to give it a shot: Can we use the Tools cross-gcc as a backend to distcc on a fast amd64 host and use that either on a VAX to compile stuff, or even for cross-building pkgsrc packages on an amd64 host? I've never been in that route and don't know anything about whether that's already working / supported. MfG, JBG --
signature.asc
(application/pgp-signature, 195 B)
-----BEGIN PGP SIGNATURE----- iF0EABECAB0WIQQlDTvPcScNjKREqWEdvV51g5nhuwUCZ3xDyAAKCRAdvV51g5nh u8CxAJ9bhVisLVUVzqJBJaIZ76i+51dtgACeN1EkJx3llScl9/eFMaOI2820C54= =A7bv -----END PGP SIGNATURE-----