Re: [PATCH v2 6/7] MIPS: Fallback CPU -march CFLAG to ISA level if unsupported

From: Nick Desaulniers <ndesaulniers@google.com>
To: Thomas Bogendoerfer <tsbogend@alpha.franken.de>,
	Jiaxun Yang <jiaxun.yang@flygoat.com>
Cc: "linux-mips@vger.kernel.org" <linux-mips@vger.kernel.org>,
	llvm@lists.linux.dev,  Nathan Chancellor <nathan@kernel.org>,
	Huacai Chen <chenhuacai@kernel.org>,
	kernel@xen0n.name,  loongarch@lists.linux.dev
Subject: Re: [PATCH v2 6/7] MIPS: Fallback CPU -march CFLAG to ISA level if unsupported
Date: Thu, 20 Apr 2023 09:32:36 -0700	[thread overview]
Message-ID: <CAKwvOdkZA6DVCbgF=P1xZcHiyGFTnYvGReJZTzVnmeZdsH0L3w@mail.gmail.com> (raw)
In-Reply-To: <20230419221500.GA17797@alpha.franken.de>

On Wed, Apr 19, 2023 at 3:15 PM Thomas Bogendoerfer
<tsbogend@alpha.franken.de> wrote:
>
> On Wed, Apr 19, 2023 at 02:35:02PM -0700, Nick Desaulniers wrote:
> > On Wed, Apr 19, 2023 at 9:50 AM Jiaxun Yang <jiaxun.yang@flygoat.com> wrote:
> > >
> > >
> > >
> > > > 2023年4月18日 21:07，Nick Desaulniers <ndesaulniers@google.com> 写道：
> > > >
> > > > On Fri, Apr 14, 2023 at 1:07 AM Jiaxun Yang <jiaxun.yang@flygoat.com> wrote:
> > > >>
> > > >> LLVM does not implement some of -march options. However those options
> > > >> are not mandatory for kernel to build for those CPUs.
> > > >>
> > > >> Fallback -march CFLAG to ISA level if unsupported by toolchain so
> > > >> we can get those kernel to build with LLVM.
> > > >>
> > > >> Link: https://github.com/ClangBuiltLinux/linux/issues/1544
> > > >> Reported-by: Nathan Chancellor <nathan@kernel.org>
> > > >> Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
> > > >
> > > > Thanks for the patch! Maybe it's more obvious to folks who work on
> > > > mips, but how did you determine that say `p5600` is `mips32r5` or
> > > > `r10000` is `mips4`?
> > >
> > > Wikipedia [1] should fit the purpose.
> > >
> > > [1]: https://en.wikipedia.org/wiki/List_of_MIPS_architecture_processors
> >
> > Mostly! Though I was not able to verify:
> > - p5600
>
> mips32r5
>
> > - r5500
>
> mips4
>
> > - rm5200
>
> mips4
>
> > - sb1
>
> mips64r1

Thomas, thanks for reviewing datasheets to verify this.

Jiaxun, for SB1, you have cc-option fall back to -march=mips4, does
that need to be -march=mips64r1?

(With that last question resolved, I'm happy to sign off on this patch)

>
> all information taken from datasheets.
>
> > - loongson*
>
> probably depends on the exact type, but as I don't have datasheets
> I can't say what it is exactly, probaly mips32r2/mips64r2.

Oh look Jiaxun, you are the maintainer for
"MIPS/LOONGSON64 ARCHITECTURE"
ok then I will trust your knowledge of this architecture!

---
Separate question to Huacai, Wang, and the loongarch list; I know LLVM
landed the Loongarch backend recently; any idea what's the status of
building arch=loongarch with LLVM=1 (linux kernel)?

-- 
Thanks,
~Nick Desaulniers