[llvm][polly][patch]Remove use of llvm.codegen intrinsic

Yabin Hu yabin.hwu at gmail.com
Tue Jun 24 00:36:51 PDT 2014


Hi Tobi,

Add a 1d loop test case.
;int A[1024];

;int gpu() {
;  int i;
;
;  for(i = 0; i < 1024; i++)
;    A[i] = i*128 + 508;
;
;  return 0;
;}
;
;int main() {
;  int b = gpu();
;  return 0;
;}

checked with make polly-test.
Is this OK to commit?

Thanks,
Yabin


2014-06-24 14:32 GMT+08:00 Tobias Grosser <tobias at grosser.es>:

> On 24/06/2014 08:27, Yabin Hu wrote:
>
>> Hi Tobias,
>>
>> The attached patch removes use of llvm.codegen intrinsic for GPGPU codegen
>> in polly.
>> Do you think it is good enough to commit?
>>
>
> If you add a trivial test case this looks good to me.
>
> Tobias
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140624/a365faeb/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0002-Remove-use-of-llvm.codegen-intrinsic-for-GPGPU-codeg.patch
Type: text/x-patch
Size: 32342 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140624/a365faeb/attachment.bin>


More information about the llvm-commits mailing list