From 21e621f84c295c668240041ae6e7b8f5079e5b11 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Zeyu=20Huang=20=7C=20=E9=BB=83=E6=BE=A4=E5=AE=87?= <11222265+fhfuih@users.noreply.github.com> Date: Thu, 12 Mar 2026 21:32:43 +0800 Subject: [PATCH] [Test] Reduce SP & Offloading test cases for L2 (#1839) Signed-off-by: Huang, Zeyu <11222265+fhfuih@users.noreply.github.com> Co-authored-by: Hongsheng Liu --- .buildkite/test-ready.yml | 10 +--------- tests/e2e/offline_inference/test_sequence_parallel.py | 3 --- 2 files changed, 1 insertion(+), 12 deletions(-) diff --git a/.buildkite/test-ready.yml b/.buildkite/test-ready.yml index 092d6f11a..66598f3ea 100644 --- a/.buildkite/test-ready.yml +++ b/.buildkite/test-ready.yml @@ -34,15 +34,7 @@ steps: - label: "Diffusion Model CPU offloading Test" depends_on: upload-ready-pipeline commands: - - | - timeout 20m bash -c ' - set +e - pytest -s -v tests/e2e/offline_inference/test_diffusion_cpu_offload.py - EXIT1=$$? - pytest -s -v tests/e2e/offline_inference/test_diffusion_layerwise_offload.py - EXIT2=$$? - exit $$((EXIT1 | EXIT2)) - ' + - timeout 10m pytest -s -v tests/e2e/offline_inference/test_diffusion_cpu_offload.py agents: queue: "gpu_1_queue" # g6.4xlarge instance on AWS, has 1 L4 GPU plugins: diff --git a/tests/e2e/offline_inference/test_sequence_parallel.py b/tests/e2e/offline_inference/test_sequence_parallel.py index d66104aa1..0f0255b44 100644 --- a/tests/e2e/offline_inference/test_sequence_parallel.py +++ b/tests/e2e/offline_inference/test_sequence_parallel.py @@ -146,9 +146,6 @@ def _run_inference( # - warmup: whether to run warmup for this SP config # - is_perf_test: whether this is a performance test (show speedup metrics) SP_CONFIGS_L2 = [ - # Ulysses-2 - performance test - (2, 1, DEFAULT_HEIGHT, DEFAULT_WIDTH, True, True), - (1, 2, DEFAULT_HEIGHT, DEFAULT_WIDTH, True, True), # Ring-2 - performance test # Hybrid - correctness only (2, 2, DEFAULT_HEIGHT, DEFAULT_WIDTH, False, False), ]