'ComPile: A Large IR Dataset from Production Sources'
by Aiden Grossman, Ludger Paehler, Konstantinos Parasyris, Tal Ben-Nun, Jacob Hegna, William S. Moses, Jose M Monsalve Diaz, Mircea Trofin, Johannes Doerfert
Action Editor: Sebastian Schelter
https://t.co/JWPwwyrGsb
#Code #Multilingual #Compiler #LLVM #IntermediateRepresentation
👉 If you are at #ISC24 in Hamburg, consider joining the #OpenMP BoF on Tuesday from 16h45 to 17h45.
👉 The theme this year is "What to Expect from OpenMP API Version 6.0"!
👉 Speakers include @cterboven, @simonmcs and @llvm_dev#HPC#Embedded
A quick update on our large dataset of LLVM intermediate representation - we have published the full dataset on HuggingFace datasets by now (https://t.co/FxCW7yXtoG) with more work to come in the near future :)
We've also revised the preprint: https://t.co/i9v7NFxg2x
For portability, it’s crucial to analyze the reliability of various compilers’ #OpenACC implementations across system architectures. Join our webinar 12/7 to explore the OpenACC Validation and Verification Testsuite and its underlying infrastructure. https://t.co/fqhBfuG5gu
ICYMI @Livermore_Lab received a few awards at #SC23:
✔️ @llvm_dev: IEEE-CS Technical Community on #HPC Early Career Researchers Award
✔️ SCREAM team: Gordon Bell #climate modeling
✔️ @spackpm: HPCwire’s Editor's Choice for Best HPC Programming Tool or Technology
Congratulations to Dr. Johannes Doerfert! He is one of the three awardees who has received the #IEEECS Technical Community on High Performance Computing Early Career Researchers Award for Excellence in HPC! Read more @ https://t.co/TtUx9yDjqx
#IEEE
Super happy to be able to finally announce the first iteration or our new code data "ComPile", a 182B tokens/2.4TB textual IR dataset of compiled production sources. In collaboration with @aidengrossman1@__koparasy__@tbennun@wsmoses@josemonsalve2 Mircea Trofin & @llvm_dev
We are looking for people to work with us on #compilers for #hpc: https://t.co/4PV2UhhefT
There are also other opportunities, reach out (via email) and spread the word!
The @llvmorg Developer Meeting videos are up! Check out our panel discussion about moving Clang C++ compiler to MLIR + opportunities and tradeoffs that entails. Thx to Ivan for initiating, Johannes as protagonist🧐 and Bruno/Alex/Mehdi for a great chat. https://t.co/DJf6ERaOQ0
The first dedicated work on Enzyme (https://t.co/6iSzAtQ3oi) in Boulder from February 23-25 is finally shaping up. No matter if you are already using Enzyme, or are only thinking of using Enzyme in the future, we want to hear from you!
https://t.co/mBX99VHXQH
@argonne_lcf@argonne@MIT@TU_Muenchen Congrats to all! A special shout out to one of our newest members of technical staff, @llvm_dev You have truly hit the ground running here at @Livermore_Lab and it's no surprise to us that this paper won. https://t.co/XBOZwcXT2N
Honored to win Best Student Paper @Supercomputing for our paper "Scalable Automatic Differentiation of Multiple Parallel Paradigms through Compiler Augmentation" with a great team from @MIT_CSAIL@argonne and @TU_Muenchen#SC22
We spy 👀 one of our booth speakers in this ASCR Discovery Newsletter: https://t.co/7Cgizb92bv
Check out @llvm_dev 's booth talk with Sunita Chandrasekaran 1pm CST Tuesday, 11/15 at the DOE Booth (1600) at @Supercomputing