A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 4, 03:45 UTC
[CI][Arm] skip e2e model tests if HF_TOKEN is not set (#41919)
Signed-off-by: Fadi Arafeh <fadi.arafeh@arm.com>
Fadi Arafeh committed
May 7, 2026 at 12:31 UTC
b20731d0aea48c426441880b0ebe28c5d0cefece
1 file changed
+15
-7
.buildkite/scripts/hardware_ci/run-cpu-test-arm.sh
+15
-7
@@ -31,6 +31,21 @@ function cpu_tests() {
31
set -e
32
pip list"
33
34
+ # Run kernel tests
35
+ docker exec cpu-test bash -c "
36
+ set -e
37
+ pytest -x -v -s tests/kernels/test_onednn.py
38
+ pytest -x -v -s tests/kernels/attention/test_cpu_attn.py
39
+ pytest -x -v -s tests/kernels/core/test_cpu_activation.py
40
+ pytest -x -v -s tests/kernels/moe/test_moe.py -k test_cpu_fused_moe_basic"
41
+
42
+ # skip tests requiring model downloads if HF_TOKEN is not set
43
+ # due to rate-limits
44
+ if [ -z "$HF_TOKEN" ]; then
45
+ echo "Warning: HF_TOKEN is not set. Skipping tests that require model downloads."
46
+ return
47
+ fi
48
+
49
# offline inference
50
docker exec cpu-test bash -c "
51
set -e
@@ -46,13 +61,6 @@ function cpu_tests() {
61
set -e
62
pytest -x -v -s tests/quantization/test_compressed_tensors.py::test_compressed_tensors_w8a8_logprobs"
63
49
- # Run kernel tests
50
- docker exec cpu-test bash -c "
51
- set -e
52
- pytest -x -v -s tests/kernels/test_onednn.py
53
- pytest -x -v -s tests/kernels/attention/test_cpu_attn.py
54
- pytest -x -v -s tests/kernels/core/test_cpu_activation.py
55
- pytest -x -v -s tests/kernels/moe/test_moe.py -k test_cpu_fused_moe_basic"
64
65
# basic online serving
66
docker exec cpu-test bash -c '