Re: git: b9cb80883bce - main - Don't omit bsd.compiler.mk for the nested delete-old in buildworld.
Date: Mon, 21 Nov 2022 08:20:58 UTC
Quoting John Baldwin <jhb@freebsd.org> (from Sat, 19 Nov 2022 04:13:10 GMT): > The branch main has been updated by jhb: > > URL: > https://cgit.FreeBSD.org/src/commit/?id=b9cb80883bce6dc992cf05ae2e59089a60d311ec > > commit b9cb80883bce6dc992cf05ae2e59089a60d311ec > Author: John Baldwin <jhb@FreeBSD.org> > AuthorDate: 2022-11-19 04:11:07 +0000 > Commit: John Baldwin <jhb@FreeBSD.org> > CommitDate: 2022-11-19 04:11:07 +0000 > > Don't omit bsd.compiler.mk for the nested delete-old in buildworld. > > src.opts.mk will need bsd.compiler.mk to determine if CXX is > supported. Without this, src.opts.mk always marks CXX as broken and > attempts to delete all dependencies of MK_CXX from WORLDTMP. Is this the reason why a buildworld directly after a successful buildworld was (re-)building parts of llvm (without doing _any_ change in between)? I noticed this just last week but didn't took the time to have a look or discuss somewhere (well, I thought "maybe it's a feature of the bootstrap, it would surely get noticed if not"). Bye, Alexander. -- http://www.Leidinger.net Alexander@Leidinger.net: PGP 0x8F31830F9F2772BF http://www.FreeBSD.org netchild@FreeBSD.org : PGP 0x8F31830F9F2772BF