tech-pkg archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: rust problems on the branch: NetBSD 9/i386 is troubled

Tobias Nygren <> writes:

>> It seems that perhaps having a different bootstrap for 9/current is in
>> order.  Can someone who deals with rust fix things on 9/i386, or suggest
>> what else wes hould do?
> We could just stop using the upstream binaries which evidently give
> us more headaches than benefits on all platforms.
> Yes, it's nice and all that they support NetBSD, but pkgsrc is not
> NetBSD. It is not worth bending over backwards to fix issues like
> this when we could just rebuild the bootstrap either statically linked
> or separate tgz's for -8, -9 and current.
> I could set up the automation for this but I'm not really interested as
> long as there is pushback from the camp who prefer we use the upstream
> binaries.

I thought we were building our own bootstrap?  I am talking about i386,
and there is no rust-bin for that, so I assumed upstream did not build

I am fine with the notion of building a bootstrap that is statically
linked for NetBSD-8/i386 and having that used on all NetBSD i386
versions.  Does anybody object to that, and if so, what do they think we
should do instead?  (All with the assumption that a get-well plan needs
to be figured out in a day or maybe two, and executed within maybe 5.)

I haven't tried to build for 9/amd64 yet.  Despite rust-bin, rust(src)
also needs to work, I think.  As someone eloquently said before, this is
not pkgbin :-)    (I am not agitating to back off RUST_TYPE=bin; we are
still having too much pain from rust.)

Attachment: signature.asc
Description: PGP signature

Home | Main Index | Thread Index | Old Index