Skip to content

CI: run the CUDA unit tests on a simulated GPU #817

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鈥檒l occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
saqibkh wants to merge 1 commit into
base: master
Choose a base branch
from
Open

CI: run the CUDA unit tests on a simulated GPU #817

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
CI: run the CUDA unit tests on a simulated GPU 
TF_BUILD_CUDA is off in every job, since the runners have no GPU, so
the cudaGraph tests never run. PantheonSim simulates a GPU on the
runner's CPU; this job builds the CUDA unit tests and runs them there.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
  • Loading branch information
@saqibkh @claude
saqibkh and claude committed Oct 2, 2026
commit 9bcd48e6ab98c3ea330068eb247942b89b1292cc
25 changes: 25 additions & 0 deletions .github/workflows/ubuntu.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,31 @@ on: [push, pull_request]

jobs:

###############################################################################
# CUDA test on a simulated GPU:
###############################################################################

# The runners have no GPU, so PantheonSim simulates one on the CPU and runs
# the CUDA unit tests' kernels there. It checks results, not performance.
cuda-test-simulated-gpu:
# Reads the repository only; the simulator's action uses no token or secrets.
permissions:
contents: read
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3
- name: simulated GPU
uses: pantheongpu/setup-pantheonsim@06648b8c0cb091584196678593215db2e16bd7eb # v0.1.2
with:
gpu: nvidia/t4
cuda-toolkit: '12.8'
- name: cmake
run: cmake -S . -B build -DCMAKE_BUILD_TYPE=Release -DCMAKE_CXX_STANDARD=20 -DTF_BUILD_CUDA=ON -DTF_BUILD_EXAMPLES=OFF -DTF_BUILD_BENCHMARKS=OFF -DCMAKE_CUDA_RUNTIME_LIBRARY=Shared
- name: build
run: cmake --build build --parallel 4 --target test_cuda_objects test_cuda_basics test_cuda_updates test_cuda_matrix test_cuda_kmeans test_cuda_for_each test_cuda_transform
- name: test
run: cd build ; ctest --output-on-failure -E NOT_BUILT

###############################################################################
# C++ 20 standard test:
###############################################################################
Expand Down