From aca48733fc3ffb9c186fe8012f09c20ee43ce9c4 Mon Sep 17 00:00:00 2001 From: Daniel Vega-Myhre Date: Mon, 14 Jul 2025 11:46:04 -0700 Subject: [PATCH 1/2] run test_everything.sh in float8 test CI using linux.aws.h100.4 --- .github/workflows/float8_test.yml | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/.github/workflows/float8_test.yml b/.github/workflows/float8_test.yml index 91083df0bf..2e891d5fef 100644 --- a/.github/workflows/float8_test.yml +++ b/.github/workflows/float8_test.yml @@ -29,7 +29,7 @@ jobs: gpu-arch-type: "cuda" gpu-arch-version: "12.6" - name: H100 - runs-on: linux.aws.h100 + runs-on: linux.aws.h100.4 torch-spec: '--pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/cu126' gpu-arch-type: "cuda" gpu-arch-version: "12.4" @@ -53,6 +53,11 @@ jobs: uv pip install -r dev-requirements.txt uv pip install vllm pip install . - pytest test/float8 --verbose -s - pytest test/integration --verbose -s - pytest test/dtypes/test_affine_quantized_float.py --verbose -s + GPU_COUNT=$(nvidia-smi -L 2>/dev/null | wc -l) + if [ "$GPU_COUNT" -ge 4 ]; then + echo "Found $GPU_COUNT GPUs - running test_everything.sh" + ./test/float8/test_everything.sh + else + echo "Only $GPU_COUNT GPUs available. Need at least 4 GPUs to run test_everything.sh" + exit 0 + fi From 2530c2ded285b1b85b3aaf996d60efd53497c4f1 Mon Sep 17 00:00:00 2001 From: Daniel Vega-Myhre Date: Mon, 14 Jul 2025 17:20:58 -0700 Subject: [PATCH 2/2] add back tests --- .github/workflows/float8_test.yml | 3 +++ 1 file changed, 3 insertions(+) diff --git a/.github/workflows/float8_test.yml b/.github/workflows/float8_test.yml index 2e891d5fef..bf58f520c6 100644 --- a/.github/workflows/float8_test.yml +++ b/.github/workflows/float8_test.yml @@ -53,6 +53,9 @@ jobs: uv pip install -r dev-requirements.txt uv pip install vllm pip install . + pytest test/float8 --verbose -s + pytest test/integration --verbose -s + pytest test/dtypes/test_affine_quantized_float.py --verbose -s GPU_COUNT=$(nvidia-smi -L 2>/dev/null | wc -l) if [ "$GPU_COUNT" -ge 4 ]; then echo "Found $GPU_COUNT GPUs - running test_everything.sh"