[llvm-commits] two-byte return optimization for AMD

[Evan Cheng]

On Dec 8, 2008, at 10:04 AM, Chris Lattner wrote:

>
> On Dec 7, 2008, at 11:39 PM, Evan Cheng wrote:
>
>>
>> On Dec 7, 2008, at 11:10 PM, Chris Lattner wrote:
>>
>>>
>>> On Dec 7, 2008, at 11:04 PM, Evan Cheng wrote:
>>>
>>>> I am not crazy about adding a subtarget feature for a micro-
>>>> optimization. If you want to enable a family of optimizations for
>>>> AMD
>>>> chips, please just add something like a FeatureAMD feature.
>>>
>>> Why not?  What if some AMD chips have this "feature/problem" and
>>> others don't?
>>>
>>
>> I don't know. The problem with using SubtargetFeature for these is  
>> you
>> can easily have an explosion of "features". But these are not
>> features, they are micro-optimizations, errata workarounds. I'd  
>> rather
>> break AMD into a several AMD families and teach x86 which families
>> suffer from this.
>
> Okay, it's your call.  Please make this its own method on X86Subtarget
> at least (which can grovel around with the selected cpu), even if
> there is no feature in the .td file.  The problem with not making this
> a feature is that you can't turn it n with -mattr.

Alright, I don't really see a cleaner solution at this point. So I  
guess a SubtargetFeature is fine. We need to do something for the slow  
BR*mr instructions.

Evan

>
>
> -Chris
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits