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: [patch] require sse3 for fisttp


Andrew Pinski wrote:
On Sat, 2006-09-16 at 21:41 -0700, Eric Christopher wrote:
Not if he wanted all of the processor features as well.

Like SSE3? Or fisttp? Well make a no-fisttp instead. fisttp should not be linked to SSE3, just because Intel decided to rename PNI to SEE3 without taking into account that fisttp does not touch any SSE register.

A -mno-fisttp would be fine, however, it doesn't address the issue I'm trying to solve in the patch. Your perception of Intel's incompetence is also irrelevant.

You still haven't addressed the issue of "it is considered an sse3 instruction in that the processor will fault". Until you do, please refrain from arguing this more.

-eric


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