This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: recognize x86 CPU variants and default SSE/MMX/3dNOW support


> On Tue, Dec 11, 2001 at 06:26:54PM +0100, Jan Hubicka wrote:
> > What is wrong with defaulting to SSE math on these CPUs?
> 
> Take up this discussion with Uli.
> 
> > I can prepare the patch today or tomorrow. The obstacke is that we do use
> > constraint letter for SSE that disappears if SSE is not present, so for SSE
> > match we will need to invent yet another letter.
> 
> Err.. why?
OK, to totally disable any fp support w/o duplicating insn patterns I need that,
but I see I can just disable the arithmetic pattern that is quite easy and keep
the comparisons/moves/similar stuff that can not cause precision to be lost.
That should be OK I guess.  I am just testing the patch.

Honza
> 
> 
> r~


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]