Re: Allegro assembly code (was: Re: [AD] proposing a new official .lib name for VC static CRT version)

[Chris <chris.kcat@xxxxxxxxxx>]

On Monday 16 January 2006 03:17, Andrei Ellman wrote:
> Just out of interest, does anyone know which flags are needed
> to build executables with GCC that will work on a 386SX?

IIRC, there's a package that's distributed with DJGPP for doing floating point 
stuff in software (and thus using it on 386SX and 486SX). As far as I 
remember, it was used automagically.

> Anyway, now that compilers have improved, is there still an advantage to
> using ASM even on 386s and 486s, or are the compilers good enough at
> creating C code just as fast or even faster than what we already have -
> even on the 386s and 486s?

It's certainly possible. Feel free to do benchmarks. :)

> Also, are the compilers smart enough to be 
> able to guess when they should make use of things like MMX, SSE etc
> instructions and try to vectorise similar operations so they can be
> speeded up on such proccessors?

If you use the -march= option, it should turn on mmx, 3dnow, sse and stuff if 
the given CPU is capable of it. There are also -mmmx/-m3dnow/etc switches to 
do it manually. The -mtune= (formerly -mcpu=) option, which is implied when 
using -march, is supposed to generate code that runs best on the given CPU, 
though still retain compatibility for processors back to the one specified by 
-march (386 by default, I think).