[prev in list] [next in list] [prev in thread] [next in thread] 

List:       gcc
Subject:    Re: GCC8
From:       Janne Blomqvist <blomqvist.janne () gmail ! com>
Date:       2018-06-24 18:29:16
Message-ID: CAO9iq9HJVg_JYiu+GYfLVHYFM7mYgOyx_Ue+c_kc5WVxfS54aQ () mail ! gmail ! com
[Download RAW message or body]

On Sun, Jun 24, 2018 at 5:45 PM, Jerome Huck <jeromehuck@orange.fr> wrote:

> Private communication. If someone has clues/hints...Thanks in advance !!!
>
> Hi.
>
> I just have a new small laptop with N4200 proc. In the past I have one
> N3470.
>
> With GCC , a couple of years ago,  I used the Linpack(Fortran) in single
> precision to have an idea of the proc capabilities.
>
> I attached the Fortran source code. I downloaded GCC from this web page
> see http://www.equation.com/servlet/equation.cmd?fa=fortran
>
> With the N3470, I was able to have more than 1 GigaFlops. (-O2, -O4...)
> and there was a difference when you changed the level of optimisation.
>
> I just tried it  with the N4200, same result with different levels of
> optimisation, benchmark around 300 Megaflops !!!
>
> I was just expecting something a little above the N3470 !
>
> Clues/Hints?
>
> Thanks in advance.
>
> Best regards.
>

Last I heard, equation.com have (substantially) modified the GCC version
they are distributing. I recommend that you either try to recreate your
problem with a build of the upstream sources (e.g. from the mingw-w64
project), or that you contact equation.com for help with your issue with
their product.

(Added the fortran list to CC, in case other gfortran developers have
better up to date information about the equation.com builds)

-- 
Janne Blomqvist
[prev in list] [next in list] [prev in thread] [next in thread]