MIME-Version: 1.0
References: <20230206201455.1790329-1-evan@rivosinc.com> <20230206201455.1790329-7-evan@rivosinc.com>
 <Y+FwvI+xRg7zMhSx@sirena.org.uk>
In-Reply-To: <Y+FwvI+xRg7zMhSx@sirena.org.uk>
From:   Evan Green <evan@rivosinc.com>
Date:   Thu, 9 Feb 2023 10:44:36 -0800
Message-ID: <CALs-HsvQiRm-guFMy-E4J+CTLoOMoxb=JiedbtFrdXwzz-oSoA@mail.gmail.com>
Subject: Re: [PATCH v2 6/6] selftests: Test the new RISC-V hwprobe interface
To:     Mark Brown <broonie@kernel.org>
Cc:     Palmer Dabbelt <palmer@rivosinc.com>,
        Conor Dooley <conor@kernel.org>, vineetg@rivosinc.com,
        heiko@sntech.de, slewis@rivosinc.com,
        Albert Ou <aou@eecs.berkeley.edu>,
        Catalin Marinas <catalin.marinas@arm.com>,
        Palmer Dabbelt <palmer@dabbelt.com>,
        Paul Walmsley <paul.walmsley@sifive.com>,
        Shuah Khan <shuah@kernel.org>, linux-kernel@vger.kernel.org,
        linux-kselftest@vger.kernel.org, linux-riscv@lists.infradead.org
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Mon, Feb 6, 2023 at 1:27 PM Mark Brown <broonie@kernel.org> wrote:
>
> On Mon, Feb 06, 2023 at 12:14:55PM -0800, Evan Green wrote:
>
> > +int main(int argc, char **argv)
> > +{
>
> > --- /dev/null
> > +++ b/tools/testing/selftests/riscv/libc.S
>
> > +.global _start
> > +_start:
> > +.option push
> > +.option norelax
> > +     la gp, __global_pointer$
> > +.option pop
> > +
> > +     la sp, stack
> > +
> > +     la t0, heap
> > +     la t1, brk
> > +     REG_S t0, 0(t1)
> > +
> > +     li a0, 0
> > +     li a1, 0
> > +
> > +     call main
>
> This looks like it's just a standard program entry but I don't speak
> RISC-V asm so I might be missing something.  If that's the case might it
> make sense to use nolibc here?

I think I can just remove this file entirely along with -nostdlib, and
just let the compiler add in this glue.
-Evan