Re: [PATCH 1/2] bswap.h: drop unaligned loads

From: Jeff King <peff@peff.net>
Date: Sep 25, 2020, 09:09 UTC
Message-ID: <20200925090906.GA66146@coredump.intra.peff.net>
In-Reply-To: <CAPUEspjpUeaqCCuBJpQLSTv=C_P4f5h22HoTPm9+9rB7k0NkaQ@mail.gmail.com>

On Fri, Sep 25, 2020 at 02:05:09AM -0700, Carlo Arenas wrote:

Show 6 quoted lines

> > >   [stock]
> > >   Benchmark #1: t/helper/test-tool sha1 <foo.rand
> > >     Time (mean ± σ):      6.638 s ±  0.081 s    [User: 6.269 s, System: 0.368 s]
> > >     Range (min … max):    6.550 s …  6.841 s    10 runs
> 
> slightly offtopic but what generates this nicely formatted output?

It's this:

  https://github.com/sharkdp/hyperfine

It will actually run both versions and compare them, but it's a little more involved to set up (since you have to do a build step in between).

Show 7 quoted lines

> > I cannot speak for s390, since I have never owned one
> 
> I happen to be lucky enough to have access to one (RHEL 8.2/z15, gcc
> 8.3.1) and seems (third consecutive run):
> 
> stock: user: 7.555s, system: 1.191s
> -DNO_UNALIGNED_LOADS: user: 7.561s, system: 1.189s

Thanks. That's not too surprising. gcc 8 seems to be able to optimize both versions to the same thing (though I have no idea if s390 has a bswap instruction).

-Peff