[llvm] r182122 - R600: Factorize Fetch size limit inside AMDGPUSubTarget

[Vincent Lejeune]

Thank for spotting that, I didn't noticed the warning.
I though default value (ie 0) was used for pre HD 4000 gen, ie HD3000
I will provide a patch soon.

Vincent

>________________________________
> De : Renato Golin <renato.golin at linaro.org>
>À : Matt Beaumont-Gay <matthewbg at google.com> 
>Cc : Vincent Lejeune <vljn at ovi.com>; "llvm-commits at cs.uiuc.edu" <llvm-commits at cs.uiuc.edu> 
>Envoyé le : Mercredi 22 mai 2013 16h06
>Objet : Re: [llvm] r182122 - R600: Factorize Fetch size limit inside AMDGPUSubTarget
> 
>
>
>On 17 May 2013 18:49, Matt Beaumont-Gay <matthewbg at google.com> wrote:
>
>> +  TexVTXClauseSize = (Device->getGeneration() >= AMDGPUDeviceInfo::HD4XXX)?16:8;
>>
>>Style nit: Spaces around '?' and ':', please.
>>
>
>I also get a warning on that line:
>
>
>lib/Target/R600/AMDGPUSubtarget.cpp:40:68: warning: comparison of unsigned expression >= 0 is always true [-Wtype-limits]
>
>
>
>cheers,
>--renato
>
>