On Wed, Nov 21, 2007 at 07:11:02PM +0100, Pieter Palmers wrote:
> It is worthwhile to specify whether the CPU is a pentium or not:
> i.e.: -march=pentium
> since this enables more efficient float <=> int conversion. The 
> performance penalty of the stock GCC conversion is rather large.
> 
> if you can figure out during configuration whether the architecture is 
> i686 and supports mmx/sse, the following will gain you even more float 
> performance:
> -mmmx -msse -mfpmath=sse

This is the purpose of my 'tuned' optimisation. As Albert says it
only works with gcc >= 4.2.0 but it saves having to work out ourselves
which CPU features are supported. It's a shame it took so long for
gcc to get this feature, but now that it has it, it doesn't seem worth
us trying to duplicate it for older versions.

Ben


-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Mixxx-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/mixxx-devel

Reply via email to