Skip to content

add note about CUDA.jl own use of threads #896

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 2 commits into
base: master
Choose a base branch
from
Open

add note about CUDA.jl own use of threads #896

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
94cadce
add note about CUDA.jl own use of threads
vchuravy Mar 31, 2025
3e92e12
Update docs/src/knownissues.md
giordano Mar 31, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
20 changes: 20 additions & 0 deletions docs/src/knownissues.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -145,6 +145,26 @@ before calling `mpiexec`.

## CUDA-aware MPI

[CUDA.jl](https://github.com/JuliaGPU/CUDA.jl) starts its own internal threads, thus the notes in the section about [Multi-threading and signal handling](@ref) apply.
In particular you may see an error like:

```
Caught signal 11 (Segmentation fault: invalid permissions for mapped object at address 0x151a22497008)
==== backtrace (tid: 226485) ====
0 0x000000000003e730 __GI___sigaction() :0
1 0x0000000001db5d4c julia_multiq_check_empty_75441() ./partr.jl:186
2 0x000000000045dab8 jfptr_multiq_check_empty_75442.1() :0
3 0x000000000004706e _jl_invoke() /cache/build/builder-amdci5-6/julialang/julia-release-1-dot-10/src/gf.c:2895
4 0x000000000009888e check_empty() /cache/build/builder-amdci5-6/julialang/julia-release-1-dot-10/src/partr.c:348
5 0x0000000002205ca8 julia_poptask_75610() ./task.jl:999
6 0x0000000002205ca8 julia_poptask_75610() ./task.jl:1001
7 0x0000000000b3b5c2 julia_wait_74913() ./task.jl:1008
8 0x0000000001be39e3 julia_#wait#645_74932() ./condition.jl:130
9 0x000000000004706e _jl_invoke() /cache/build/builder-amdci5-6/julialang/julia-release-1-dot-10/src/gf.c:2895
10 0x0000000000035ca1 jlcapi_synchronization_worker_11948() text:0
=================================
```

### Memory pool

Using CUDA-aware MPI on multi-GPU nodes with recent CUDA.jl may trigger (see [here](https://github.com/JuliaGPU/CUDA.jl/issues/1053#issue-946826096))
Expand Down