Commit Graph

2 Commits

Author SHA1 Message Date
Binyang Li
610db6f023 Fix test script (#655)
Fix: #654. Address correctness_test.py crash issue

Co-authored-by: Changho Hwang <changhohwang@microsoft.com>
2025-10-21 19:57:17 +00:00
Binyang Li
ba4c4aaeb8 Integrate MSCCL++ with torch workload (#626)
Integrate MSCCL++ with torch
Introduce `NCCL audit shim library`, use can use following commands to
launch torch library. Also avoid break build pipeline in the CPU machine
```bash
export LD_AUDIT=$MSCCLPP_INSTALL_DIR/libmscclpp_audit_nccl.so
export LD_LIBRARY_PATH=$MSCCLPP_INSTALL_DIR:$LD_LIBRARY_PATH
torchrun --nnodes=1 --nproc_per_node=8 your_script.py
```
2025-09-09 13:28:32 -07:00