On Mon, Sep 20, 2021 at 09:13:44AM +0100, Peter Maydell wrote:
> On Mon, 20 Sept 2021 at 06:07, David Gibson <david@gibson.dropbear.id.au> wrote:
> > On Sat, Sep 18, 2021 at 01:01:35PM -0700, Richard Henderson wrote:
> > > We dropped host support for sparcv8 (true 32-bit) a long time ago.
> > > We only support sparcv9 in ilp32 (sparcv8plus) and lp64 (sparc64).
> >
> > We really need to update
> > https://qemu-project.gitlab.io/qemu/about/build-platforms.html
> > to clarify this then.  I don't really know what the procedures are for
> > updating the website.
> 
> It's automatically updated by building the documentation from current
> head-of-git, so the answer is "submit a patch to change
> docs/about/build-platforms.rst". (That's a pretty new file, and
> the stuff about CPU architectures has gone in only very recently,
> so it's not unsurprising if Marc-André and I got some things wrong:
> we were just looking through tcg/ to see what it seemed to have
> support for.)

Oh.. yeah.. that's pretty obvious now I look at it </sheepish>

> 
> The structure of the build-platforms page currently assumes that
> "supported host CPU architectures" and "supported host OSes" are
> basically orthogonal, because historically the nature of QEMU has
> been that this is more-or-less true. If we want to try to be more
> specific about that then we'd need to re-jig things.

Yeah, I think we kind of need to do that.  Even apart from the Rust
support thing, I think it would help to clarify which build platforms
are merely kind of unusual versus which are very unusual (obscure OS *
obscure ISA).

-- 
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson