llnl / llnl/blt

blt_hip_smoke false positive

Open
#586 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

bug
Dominant language
C++
Stars
296
Forks
66
Avg merge
1h 3m
Merged PRs (30d)
1

Description

I tried running blt_hip_runtime_smoke in a rocm container, but with no rocm devices.

here is the tail of the output:

...
Device Number: 25952
  Device name: 
  Memory Clock Rate (KHz): 0
  Memory Bus Width (bits): 0
  Peak Memory Bandwidth (GB/s): 0.000000

Device Number: 25953
  Device name: 
  Memory Clock Rate (KHz): 0
  Memory Bus Width (bits): 0
  Peak Memory Bandwidth (GB/s): 0.000000

Device Number: 25954
  Device name: 
  Memory Clock Rate (KHz): 0
  Memory Bus Width (bits): 0
  Peak Memory Bandwidth (GB/s): 0.000000
(return code 0)

I expected this to fail.

Obviously the clock rates etc hint that this is bogus, but if it was just run with ctest w/o verbose output, it would have looked a-ok.

Contributor guide

Open the contributing guide

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

Start with the blt_hip_runtime_smoke entry point and run it in a ROCm container without ROCm devices, then compare its behavior when invoked directly and through ctest. Trace how the smoke test determines success from the reported device data. Done means the no-device case returns failure so non-verbose ctest does not report a false positive.

Written by the indexing model from the issue text.

Assessment

Tech stack
cmake, cpp
Domain
build-system, testing
Issue type
Bug
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
45/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.