understanding why cuda workflows are so tricky

.github/workflows/parallel.yml

@@ -0,0 +1,23 @@
+name: NVCC
+
+on:
+  push:
+    branches: [ "main" ]
+  pull_request:
+    branches: [ "main" ]
+
+jobs:
+  build-parallel:
+
+    runs-on: ubuntu-latest
+
+    steps:
+      - uses: actions/checkout@v3
+      - uses: Jimver/cuda-toolkit@v0.2.11
+        id: cuda-toolkit
+        with:
+          cuda: '12.1.0'  
+      - name: cuda-build 
+        shell: |
+          bash  
+      - run: nvcc -V

.github/workflows/sequential.yml

@@ -0,0 +1,20 @@
+name: C Binary build
+
+on:
+  push:
+    branches: [ "main" ]
+  pull_request:
+    branches: [ "main" ]
+
+jobs:
+  build-sequential:
+
+    runs-on: ubuntu-latest
+
+    steps:
+    - uses: actions/checkout@v3
+    - name: make clean 
+      run: make clean
+
+    - name: make sequential
+      run: make sequential

README.md

@@ -20,6 +20,6 @@ make parallel
 ```
 
 ### TODO LIST
-- [ ] ⚠️ **IMPORTANT** more tests on my GPU (working on my Mac M1 rn🥲)
+- [X] ⚠️ **IMPORTANT** more tests on my GPU (working on my Mac M1 rn🥲)
 - [ ] more performances evaluation 
 - [ ] CUDA workflow