search for: dswp

0 down vote favorite [http://stackoverflow.com/questions/9354312/llvm-2-8-dswp-core-dump-linux#] share [g+] share [fb] share [tw] i new in llvm.i just compile google code and got this error. " Stack dump: 0. Program arguments: opt -load /home/vahid/mywork/llvm-2.8/Release/lib/DSWP.so -dswp obj.o -o out.o 1. Running pass 'Function Pass...

Hi! I'm writing my own project in LLVM and I came across a document describing Decoupled software pipelining algorithm written in LLVM by The Liberty Research Group for their VELOCITY C++ compiler: http://liberty.princeton.edu/Research/DSWP/ http://liberty.princeton.edu/Publications/micro38_dswp.pdf and some of spectacular results: http://liberty.princeton.edu/videos/gimp.php http://liberty.princeton.edu/videos/h263.php I found out, that there was a project to implement this technique as LLVM pass: http://code.google.com/p/15745-pro...

Hello, below is the first draft of my proposal for GSoC. Any comments/advices are apreciated. # Implement Decoupled Software Pipeline (DSWP) ------------------------------------------------------------------------------- # Abstract ------------------------------------------------------------------------------- The goal of this project is to implement the automatic parallelization technique called Decoupled Software Pipeline [1] in L...

Hi Vahid, I suggest you use a more recent version. The latest version is 3.0. Ciao, Duncan. On 20/02/12 09:27, vahid eslami wrote: > > 0 down vote favorite > <http://stackoverflow.com/questions/9354312/llvm-2-8-dswp-core-dump-linux#> > share [g+] share [fb] share [tw] > > i new in llvm.i just compile google code and got this error. " Stack dump: 0. > Program arguments: opt -load /home/vahid/mywork/llvm-2.8/Release/lib/DSWP.so > -dswp obj.o -o out.o 1. Running pass 'Function Pass Ma...

I am MSc student and my thesis subject is "parallelization of sequential program in thread level to run on multi-core processors". I want automatically parallelize sequential program in thread level to run on multi-core processors with software pipelining. I've use LLVM to code analysis and translate code to IR, I could extract PDG graph form IR code and convert it to SCCDAG and

...tep and code generation, I don't have any idea and > information to assign partitions and communications to threads. and I stop > in this position. Unfortunately I don’t have enough time and I am under > force to complete the project. What you want to do is pretty much like what DSWP does, see http://liberty.princeton.edu/Research/DSWP/ What DSWP does is partitioning a loop into producer/consumer thread. I know they had implemented DSWP in LLVM before, but I don't know what the current status is. Maybe you can mail them to ask some idea. Regards, chenwj -- Wei-Ren C...

I've study their work carefully. My problem is the implementation of threads in llvm after partitionning. I want to have information about how to implement producer/consumer thread in llvm. I do not know where I should start in llvm for code generation and create thread and insert produce and consume statement . -------------- next part -------------- An HTML attachment was scrubbed... URL:

Hi, All I want to implement DSWP Which is used for parallelization of loops. For this purpose, the loop was replaced with a new basic block in main function. And new functions were created and basic blocks of Loop assigned to them.I have checked blocks and branches for Succ and Pred relation and I have not found any problems. Howe...

Hello, Erwin > So if you port this SSE code to another platform (Linux, Altivec, > NEON), you could contribute it back to Bullet? I believe this should work as-is on linux. Am I missing something? > optimizations, but haven't contributed this back. > This NEON/VFP, part of the an open source iPhone project, could be a > starting point for this: > http://tinyurl.com/y9gv3e8

Hello, my name is César I'm PhD student at UNICAMP-Brasil. I'm considering submitting a proposal to participate in GSOC, however I would like to hear your comments before writing the proposal: These are the topics I'm considering to write: - Implement DSWP[1]: The proposal will be to implement Decoupled Software Pipeline as a pass in the LLVM-opt. To do this many analysis will be used however many of them are already implemented in opt, so I guess the major part of the work will be generate code to synchronize/communicate between stages. - Implement...

The linux builds are not using SSE right now, but the vector data is 16-byte aligned on all platforms. So if you port this SSE code to another platform (Linux, Altivec, NEON), you could contribute it back to Bullet? The most interesting SSE part is the innerloop of the constraint solver: http://tinyurl.com/ydoapct Some developers replaced some linear algebra functions (in Bullet/LinearMath) with