Java Vectorization

[GoogleCodeExporter]

What version of the product are you using? On what operating system?
Aparapi on Ubuntu 12.04

Please provide any additional information below.
I am looking into the possibility of using the Aparapi framework to add 
vectorization to java. The idea is to take advantage of the kernel interface 
and the fact that anything in separate kernels is independent and combine 
consecutive loop iterations inside the kernel interface into vector 
instructions. 

The project requires two steps. The first would be to modify Aparapi to have it 
output code that is partially loop unrolled and annotated in a way that a 
modified JVM could identify the vectorization section. I was wondering if 
anyone could point me in the right direction as to which part of the Aparapi 
code I need to look at to get this working.

The second step would be to implement this modified JVM.

Thanks

Original issue reported on code.google.com by falbe...@vt.edu on 23 Oct 2013 at 7:00

[GoogleCodeExporter]

I know this isn't a defect, but I can't figure out how to change it

Original comment by falbe...@vt.edu on 23 Oct 2013 at 7:01

[GoogleCodeExporter]

Thanks for raising this. I changed it to a an enhancement request, rather than 
a bug. 

Actually this will be hard to fully implement in Aparapi.  As you indicate, 
this would need a modified JVM, and of course Aparapi tries to be JVM agnostic 
and does not have access to JVM internals (except as exposed via 
JVMTI/JNI/Unsafe).

I think this is a great idea, but one that I suspect may need to be looked at 
in the context of 'Project Sumatra', which is aiming at bringing GPU 
compute/offload to the JVM itself - and as such is indeed a modified JVM. 

http://openjdk.java.net/projects/sumatra/
http://www.extremetech.com/computing/137628-project-sumatra-improves-java-perfor
mance-with-opencl-graphics-card-acceleration

I would be interested in a use case (no matter how trivial) and en example of 
how you would like to see (for example) data auto-vettorized. 

Gary

Original comment by frost.g...@gmail.com on 24 Oct 2013 at 1:46

• Added labels: Type-Enhancement
• Removed labels: Type-Defect

[GoogleCodeExporter]

I was looking at using Oracle's Maxine JVM to modify to actually produce the 
vectorized assembly code but was going to use the Aparapi interface to justify 
that the variables are independent and vectorizable since they must be to be 
split up across a gpu. The part of Aparapi that I am looking to modify would be 
to preform the loop unroll and mark the code to tell the modified JVM to 
produce vectorized instructions for those variables if possible. Which part of 
the Aparapi code should I look into to preform this or should I just implement 
this separately and run it on the code before Aparapi?

Original comment by falbe...@vt.edu on 24 Oct 2013 at 7:33

[GoogleCodeExporter]

For a basic example say there is a for loop

for(int i=0; i < 64; i++)
{
   a[i] = b[i] * c[i];
}

could be vectorized to 
for(int i=0; i < 32; i++)
{
   a[i&i+1] = b[i&i+1] * c[i&i+1];
}

There are some benchmarks where the GPU is faster than the JTP and this 
obviously wouldn't apply to them but when JTP is faster due to the overhead of 
copying data to and form the GPU my thought is that this would speed up the JTP 
execution just like vectorization does in C.

Original comment by falbe...@vt.edu on 24 Oct 2013 at 7:38

[GoogleCodeExporter]

Ah I see.

So you might want to look at using Graal for this. 
https://wikis.oracle.com/display/MaxineVM/MaxineGraal

Graal (derived from Maxine) as far as I can tell is also the engine we are 
using in Sumatra to generate HSAIL code for GPU offload.  The idea of 
vectorizing via Graal is not new apparently (although was new to me :) ) 

http://www.boston-technology.com/blog/what-does-graal-mean-for-java/

BTW having played with Graal for a while now, I am considering using it for GPU 
code generation for Aparapi.  It supports interesting optimizations and is 
great at inlining.

I think this is a great idea for a project. 

Gary

Original comment by frost.g...@gmail.com on 24 Oct 2013 at 10:02