From: Klemens Nanni Subject: Re: binutils: enable ld.bfd ? To: Frederic Cambus , ports Date: Mon, 01 Dec 2025 22:12:36 +0000 30.11.2025 19:17, Klemens Nanni пишет: > 30.11.2025 18:58, Jeremie Courreges-Anglas пишет: >> Nothing depends on binutils at the moment. Indeed this port will need >> some tweaks on some architectures (backports from base probably). >> Sigh. Maybe restricting this to amd64 and arm64 is more reasonable >> indeed... > > Yes, hence my uncertainty as to what's the best approach here. I'm starting to prefer a small subset of archs; the Makefile bits are not much and so far only two would use ld.bfd, so looking into build failures on archs like arm is eating time that I'd like to spend on productive stuff instead. > >> I wonder if it would be less effort to provide a stripped down >> binutils port for bare x86_64, like arm-none-eabi/binutils. Care to >> show your ipxe port and how it is currently using gld from binutils? > > - "new net/ipxe" https://marc.info/?l=openbsd-ports&m=176381141712670&w=2 > - https://github.com/jasperla/openbsd-wip/tree/master/net/ipxe > (since got updated and gained .iso images, rest is the same) > > tl;dr: Usual COMPILER* for gcc15 and AS=gas, etc. via MAKE_FLAGS. > >>> Could someone please give this a try on anything other than: >>> arm64 amd64 i386 octeon? >> >> Builds and packages on sparc64 > > Thanks! > >> The list of "safe" architectures where we could enable gld without >> patches can probably be determined, but gotta head out now. > > I'll try to cover more architectures. riscv64 builds fine as-is, arm needs the equivalent to arm64's configure.tgt patch, but then further fails with earmelf_obsd.c:56:5: error: non-void function 'gldarmelf_obsd_get_script' should return a value [-Wreturn-mismatch] 56 | return | ^ Oddly enough, that file does not exist under WRKDIR.