Hi everyone,
I’m exploring an experimental approach to formally verifying CUDA kernels that rely on cuda::barrier/mbarrier and async copy (agentically). Is anyone interested on working on this with me? Would people find it useful if it worked?
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Agents in GPU Seeking Agents or implementations of Finite State Machine in CUDA. | 0 | 845 | November 27, 2009 | |
| CUDA AI Code Editor for faster CUDA kernals | 2 | 108 | November 7, 2025 | |
| I made 64 swarm agents compete to write gpu kernels | 3 | 75 | January 17, 2026 | |
| CUDA Kernel self-suspension ? Can a CUDA Kernel conditionally suspend its execution ? | 46 | 45546 | April 17, 2011 | |
| test test123 | 0 | 1916 | December 4, 2009 | |
| test | 0 | 3650 | July 24, 2008 | |
| The performance result of CUDA sample globalToShmemAsyncCopy is puzzled | 1 | 630 | December 9, 2021 | |
| Block sync | 0 | 373 | November 25, 2020 | |
| [closed] | 0 | 612 | December 8, 2017 | |
| Please delete this | 0 | 363 | December 14, 2018 |