LLVM BoF at SC13

All,

I have arranged for an LLVM BoF at SC13 in Denver:

http://sc13.supercomputing.org/schedule/event_detail.php?evid=bof143

If you will be at SC13, have an interesting project based on or within LLVM, and would be willing to come to the BoF and spend 10 minutes or so giving an overview of it, please email me. I am trying to get a collection of people together to talk about vectorization and parallelization in particular. DSLs targeting HPC would also be interesting. Of course you can also attend without presenting, and please let people interested in LLVM know about this event!

Thanks!

Michael McCool (Intel)

Dear Michael & Hal,

We had a submission to SC13 on beating OpenACC compilers with LLVM+Polly+NVPTX [1]. Unfortunately, reviewers somehow expected we want to go into polyhedral tools intercomparison, which was absolutely not our goal. The goal was exactly to bring a competitive HPC/GPU toolchain for end users. Only NV works for now, MIC is in progress. If you think this contribution is worth a short presentation at the LLVM BoF, please let me know!

[1] http://kernelgen.org/tech_report2013/

Thanks,

  • D.

I’ll be there and will send some ideas in the near future.



From: Dmitry Mikushin
Sent: Thursday, October 31, 2013 9:49 AM
To: Mccool, Michael
Cc: llvmdev@cs.uiuc.edu
Subject: Re: [LLVMdev] LLVM BoF at SC13

Dear Michael & Hal,

We had a submission to SC13 on beating OpenACC compilers with LLVM+Polly+NVPTX [1]. Unfortunately, reviewers somehow expected we want to go into polyhedral tools intercomparison, which was absolutely not our goal. The goal was exactly to bring a competitive HPC/GPU toolchain for end users. Only NV works for now, MIC is in progress. If you think this contribution is worth a short presentation at the LLVM BoF, please let me know!

[1] http://kernelgen.org/tech_report2013/

Thanks,

  • D.

ATT00001…txt (183 Bytes)

Hi Michael,

you may want to add this event to the LLVM homepage. We already use it to keep track of other meetings such as the FOSDEM room. The relevant
page is.

http://llvm.org/devmtg/

Cheers,
Tobias

Hi Michael,

We’re working on a program slicing project based on LLVMSlicer to generate I/O benchmarks from HPC applications. I wonder whether others are interesting in this idea. I can send you several pages of slides to introduce our work. Any comment is welcome.

I’ll present a technical paper at 11:00 am ~ 11:30 am that day. Perhaps I will be late for a while if I have to discuss with others about that work.

Thanks.