[HPGMG Forum] Do we want the benchmark to go into intrinsics?

Sam Williams swwilliams at lbl.gov
Tue Apr 29 16:14:40 UTC 2014


I would tend to agree.  

The only requirement I have for the FV code is really just a C compiler.
Of course if you want distributed memory, then the only path I include in the reference version is MPI.
If you want threading, the only path in the reference version is OpenMP 3.1.  
I didn't want to put ISA specific-optimizations (or intrinsics or other threading models ...) into a *reference* implementation. 
As such, everything is pure C.  In fact, all the ISA-specific timers were replaced with omp/mpi versions.


On Apr 29, 2014, at 9:03 AM, Brian Van Straalen <bvstraalen at lbl.gov> wrote:

> 
> 
> Jed Brown	7b171e1       	 fe: loop optimizations to TensorContract_QPX	
> 28 Apr 2014
> Jed Brown	339691b       	 fe: initial QPX version of tensor contraction	
> 28 Apr 2014
> Jed Brown	b1189a3       	 make: remove redundant link flags	
> 28 Apr 2014
> 
> This seems like a pretty unportable benchmark idea.  Does HPL do this for the download version?  Or are these commits to the research branch?
> 
> Brian
> 
> Brian Van Straalen         Lawrence Berkeley Lab
> BVStraalen at lbl.gov         Computational Research
> (510) 486-4976             Division (crd.lbl.gov)
> 
> 
> 
> 
> _______________________________________________
> HPGMG-Forum mailing list
> HPGMG-Forum at hpgmg.org
> https://hpgmg.org/lists/listinfo/hpgmg-forum



More information about the HPGMG-Forum mailing list