NVIDIA / NVIDIA/nvbandwidth

multinode benchmarks fail to start with error - [CUDA_ERROR_INVALID_VALUE] invalid argument in expression cuMemSetAccess

Open
#44 19 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
C++
Stars
778
Forks
93
PR merge metrics
No merged PRs in 30d

Description

Running multinode benchmarks works on single node, but fails with the following error when running on multiple nodes:

mpirun --map-by ppr:8:node --bind-to core -np 16 --report-bindings --hostfile hostfile  /usr/local/bin/nvbandwidth -p multinode
...
Running multinode_device_to_device_memcpy_read_ce.
[CUDA_ERROR_INVALID_VALUE] invalid argument in expression cuMemSetAccess((CUdeviceptr) buffer, roundedUpAllocationSize, &desc, 1 ) on gpub200-nom6ae0518, rank = 8 in MultinodeMemoryAllocationUnicast::MultinodeMemoryAllocationUnicast(size_t, int)() : /home/code/external/nvbandwidth/multinode_memcpy.cpp:76

Binary was build with Cuda 12.8

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Reproduce the multinode benchmark with the shown mpirun command and CUDA 12.8, then inspect external/nvbandwidth/multinode_memcpy.cpp at line 76 where cuMemSetAccess fails. Compare the single-node and multinode paths and verify that the multinode benchmark starts and completes without CUDA_ERROR_INVALID_VALUE.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp
Domain
distributed-systems, performance
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
50/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.