Pending GCC-3.1 upgrade - do or don't

Sam Barnett-Cormack sambc at nights.force9.co.uk
Thu May 9 04:57:14 PDT 2002


> -----Original Message-----
> From: lfs-dev-bounce at linuxfromscratch.org
> [mailto:lfs-dev-bounce at linuxfromscratch.org]On Behalf Of Ian Molton
>
> On Wed, 08 May 2002 10:35:26 -0400
> Michael W Daniels <daniels at ling.ohio-state.edu> wrote:
>
<SNIP>
> > http://gcc.gnu.org/gcc-3.1/changes.html
>
> Er. Im seeing code generation speedups to put it on a par with 2.95.3
> and ABI incompatability. It doesnt mention the generated code running
> faster...
>
> Whats the big advantage?

According to the SPECInt2000 results on an AMD Athlon CPU, the code
generated by GCC 3.1 is 6% faster on the average (8.2% faster with profile
feedback) compared to GCC 3.0. The code produced by *GCC 3.0 is about 2.1%
faster compared to 2.95.3*. Tests were done using the -O2 -march=athlon
command-line options.

Is that not faster then?

Sam

-- 
Unsubscribe: send email to listar at linuxfromscratch.org
and put 'unsubscribe lfs-dev' in the subject header of the message



More information about the lfs-dev mailing list