[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH rust] Add new target for GNU/Hurd
From: |
Samuel Thibault |
Subject: |
Re: [PATCH rust] Add new target for GNU/Hurd |
Date: |
Wed, 12 Jul 2023 00:28:17 +0200 |
User-agent: |
NeoMutt/20170609 (1.8.3) |
Sergey Bugaev, le mar. 11 juil. 2023 17:52:41 +0300, a ecrit:
> On Mon, Jul 10, 2023 at 10:53 PM Samuel Thibault
> <samuel.thibault@gnu.org> wrote:
> > llvm calls it i686-unknown-hurd-gnu (that was an intended behavior, even
> > when knowing that gnu tools call it i686-unknown-gnu), so we should
> > probably stick to that.
> >
> > I'd tend to keep it "hurd", because "gnu" is confusing for people.
> > Notably in llvm there are various parts called "gnu" which are the parts
> > shared between GNU/Linux, GNU/kFreeBSD, and GNU/Hurd.
>
> That explains it, thanks. Damien, Vedant, please mention this in the
> commit message and in comments.
>
> Still, OS = "hurd" sounds very wrong to me. It's probably too late to
> change the LLVM triplet (quadruplet?),
It's still a triplet, the os part just happens to contain a dash.
> but perhaps OS = "gnu-hurd"
> (read: GNU/Hurd, not GNU Hurd) would be a good compromise for Rust?
It would look quite incoherent when the LLVM os part is hurd-gnu. Note
that the ordering in the triplet is indeed reversed compared to the
ordering in "/" notation: i686-unknown-gnu does mean "the gnu userspace
running on a machine from an unknown manufacturer that uses an i686
cpu", while GNU/Hurd means "the GNU user space running on top of the
Hurd.
So it does make sense to me to use "hurd-gnu" as OS.
Samuel
[PATCH rust-libc] i386-gnu: Add GNU/Hurd os, Damien Zammit, 2023/07/09