Displaying 3 results from an estimated 3 matches for "polybench_scops".
2013 Jun 30
4
[LLVMdev] [Polly][GSOC2013] FastPolly -- SCOP Detection Pass
Hi all,
I have investigated the compile-time overhead of "Polly Scop Detection" pass based on LNT testing results.
This mail is to share some results I have found.
(1) Analysis of "SCOP Detection Pass" for PolyBench (Attached file PolyBench_SCoPs.log)
Experimental results show that the "SCOP Detection pass" does not lead to significant extra compile-time overhead for compiling PolyBench. The percent of compile-time overhead caused by "SCOP Detection Pass" is usually less than 4% of total compile-time. Details for each be...
2013 Jun 30
0
[LLVMdev] [Polly][GSOC2013] FastPolly -- SCOP Detection Pass
...te:
> Hi all,
>
>
>
> I have investigated the compile-time overhead of "Polly Scop Detection" pass based on LNT testing results.
> This mail is to share some results I have found.
>
>
> (1) Analysis of "SCOP Detection Pass" for PolyBench (Attached file PolyBench_SCoPs.log)
> Experimental results show that the "SCOP Detection pass" does not lead to significant extra compile-time overhead for compiling PolyBench. The percent of compile-time overhead caused by "SCOP Detection Pass" is usually less than 4% of total compile-time. Details for ea...
2013 Jul 01
1
[LLVMdev] [Polly][GSOC2013] FastPolly -- SCOP Detection Pass
...t;
>>
>> I have investigated the compile-time overhead of "Polly Scop Detection" pass based on LNT testing results.
>> This mail is to share some results I have found.
>>
>>
>> (1) Analysis of "SCOP Detection Pass" for PolyBench (Attached file PolyBench_SCoPs.log)
>> Experimental results show that the "SCOP Detection pass" does not lead to significant extra compile-time overhead for compiling PolyBench. The percent of compile-time overhead caused by "SCOP Detection Pass" is usually less than 4% of total compile-time. Details fo...