[LLVMdev] changing -mattr behavior with mmx and sse
clattner at apple.com
Thu Nov 20 10:41:23 PST 2008
On Nov 20, 2008, at 8:31 AM, Preston Gurd wrote:
> Might you instead consider just adding a -disable-mmx option?
I agree, this is a better approach. This distinguishes between
capabilities of the chip and the desire to codegen specific vectors
one way or another.
> On Thu, 2008-20-11 at 02:57 -0500, Mon Ping Wang wrote:
>> When setting -mattr option on X86, I would like to treat MMX
>> separately from SSE levels. This would allow a client who sets the
>> attributes directly to set the SSE level independent of MMX, e.g.,
>> -march=x86 -mattr=sse41, one would get sse4.1 with mmx disabled while
>> llc -march=x86 -mattr=mmx -mattr=sse42 will get mmx and sse42. If
>> anyone objects to this change, please let me know.
>> -- Mon Ping
>> LLVM Developers mailing list
>> LLVMdev at cs.uiuc.edu http://llvm.cs.uiuc.edu
> LLVM Developers mailing list
> LLVMdev at cs.uiuc.edu http://llvm.cs.uiuc.edu
More information about the llvm-dev