From 9fd00ee006ccd4996bbc756397b039343d2fde94 Mon Sep 17 00:00:00 2001 From: Andreas Karatzas Date: Sat, 27 Jun 2026 04:08:54 -0500 Subject: [PATCH] [ROCm][CI] Move remaining mi250_2 tests out of the MI250 queue (#46905) Signed-off-by: Codex Co-authored-by: Codex --- .buildkite/test-amd.yaml | 214 ++++++++++++--------------------------- 1 file changed, 64 insertions(+), 150 deletions(-) diff --git a/.buildkite/test-amd.yaml b/.buildkite/test-amd.yaml index 083aa024b2..a9608cc332 100644 --- a/.buildkite/test-amd.yaml +++ b/.buildkite/test-amd.yaml @@ -112,33 +112,6 @@ steps: # # ######################################################################################################################################### -#----------------------------------------------------- mi250 · basic_correctness -----------------------------------------------------# - -- label: Distributed Model Tests (2 GPUs) # TBD - timeout_in_minutes: 180 - mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250] - agent_pool: mi250_2 - num_gpus: 2 - working_dir: "/vllm-workspace/tests" - source_file_dependencies: - - vllm/model_executor/model_loader/sharded_state_loader.py - - vllm/model_executor/models/ - - vllm/model_executor/layers/ - - vllm/v1/attention/backends/ - - vllm/v1/attention/selector.py - - tests/basic_correctness/ - - tests/model_executor/model_loader/test_sharded_state_loader.py - - tests/models/ - - vllm/_aiter_ops.py - - vllm/platforms/rocm.py - commands: - - TARGET_TEST_SUITE=MI250 pytest basic_correctness/ -v -s -m 'distributed(num_gpus=2)' - - CUDA_VISIBLE_DEVICES=0,1 pytest -v -s model_executor/model_loader/test_sharded_state_loader.py -m '(not slow_test)' - - pytest models/test_transformers.py -v -s -m 'distributed(num_gpus=2)' - - pytest models/language -v -s -m 'distributed(num_gpus=2)' - - pytest models/multimodal -v -s -m 'distributed(num_gpus=2)' --ignore models/multimodal/generation/test_whisper.py - - VLLM_WORKER_MULTIPROC_METHOD=spawn pytest models/multimodal/generation/test_whisper.py -v -s -m 'distributed(num_gpus=2)' - #---------------------------------------------------------- mi250 · compile ----------------------------------------------------------# - label: PyTorch Compilation Unit Tests # TBD @@ -179,48 +152,8 @@ steps: commands: - "find compile/fullgraph/ -name 'test_*.py' -not -name 'test_full_graph.py' -exec pytest -s -v {} \\\\;" -- label: Distributed Compile + RPC Tests (2 GPUs) # TBD - timeout_in_minutes: 180 - mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250] - agent_pool: mi250_2 - num_gpus: 2 - optional: true - working_dir: "/vllm-workspace/tests" - source_file_dependencies: - - vllm/compilation/ - - vllm/distributed/ - - vllm/engine/ - - vllm/executor/ - - vllm/worker/worker_base.py - - vllm/v1/engine/ - - vllm/v1/worker/ - - tests/compile/fullgraph/test_basic_correctness.py - - tests/compile/test_wrapper.py - - tests/entrypoints/llm/test_collective_rpc.py - - vllm/platforms/rocm.py - commands: - - pytest -v -s entrypoints/llm/test_collective_rpc.py - - pytest -v -s ./compile/fullgraph/test_basic_correctness.py - - pytest -v -s ./compile/test_wrapper.py - #-------------------------------------------------------- mi250 · distributed --------------------------------------------------------# -- label: Distributed Comm Ops # TBD - timeout_in_minutes: 180 - mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250] - agent_pool: mi250_2 - num_gpus: 2 - working_dir: "/vllm-workspace/tests" - source_file_dependencies: - - vllm/distributed - - tests/distributed - - vllm/platforms/rocm.py - commands: - - pytest -v -s distributed/test_comm_ops.py - - pytest -v -s distributed/test_shm_broadcast.py - - pytest -v -s distributed/test_shm_buffer.py - - pytest -v -s distributed/test_shm_storage.py - - label: Pipeline + Context Parallelism (4 GPUs) # TBD timeout_in_minutes: 180 mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250] @@ -330,54 +263,6 @@ steps: - pytest -v -s models/multimodal/generation/test_common.py -m core_model -k "not qwen2 and not qwen3 and not gemma" - pytest -v -s models/multimodal/generation/test_qwen2_vl.py -m core_model -#---------------------------------------------------------- mi250 · plugins ----------------------------------------------------------# - -- label: Plugin Tests (2 GPUs) # TBD - timeout_in_minutes: 180 - mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250] - agent_pool: mi250_2 - num_gpus: 2 - optional: true - working_dir: "/vllm-workspace/tests" - source_file_dependencies: - - vllm/plugins/ - - tests/plugins/ - - vllm/platforms/rocm.py - commands: - # BEGIN: platform plugin and general plugin tests, all the code in-between runs on dummy platform - - pip install -e ./plugins/vllm_add_dummy_platform - - pytest -v -s plugins_tests/test_platform_plugins.py - - pip uninstall vllm_add_dummy_platform -y - # END: platform plugin tests - # BEGIN: `io_processor` plugins test, all the code in between uses the `prithvi_io_processor` plugin - - pip install -e ./plugins/prithvi_io_processor_plugin - - pytest -v -s plugins_tests/test_io_processor_plugins.py - - pytest -v -s plugins_tests/test_terratorch_io_processor_plugins.py - - pip uninstall prithvi_io_processor_plugin -y - # END: `io_processor` plugins test - # BEGIN: `bge_m3_sparse io_processor` test - - pip install -e ./plugins/bge_m3_sparse_plugin - - pytest -v -s plugins_tests/test_bge_m3_sparse_io_processor_plugins.py - - pip uninstall bge_m3_sparse_plugin -y - # END: `bge_m3_sparse io_processor` test - # BEGIN: `colbert_query io_processor` test - - pip install -e ./plugins/colbert_query_plugin - - pytest -v -s plugins_tests/test_colbert_query_io_processor_plugins.py - - pip uninstall colbert_query_plugin -y - # END: `colbert_query io_processor` test - # BEGIN: `stat_logger` plugins test - - pip install -e ./plugins/vllm_add_dummy_stat_logger - - pytest -v -s plugins_tests/test_stats_logger_plugins.py - - pip uninstall dummy_stat_logger -y - # END: `stat_logger` plugins test - # BEGIN: other tests - - pytest -v -s plugins_tests/test_scheduler_plugins.py - - pip install -e ./plugins/vllm_add_dummy_model - - pytest -v -s distributed/test_distributed_oot.py - - pytest -v -s plugins_tests/test_oot_registration_online.py # it needs a clean process - - pytest -v -s plugins_tests/test_oot_registration_offline.py # it needs a clean process - - pytest -v -s plugins_tests/lora_resolvers # unit tests for in-tree lora resolver plugins - #------------------------------------------------------------ mi250 · v1 -------------------------------------------------------------# - label: Batch Invariance (H100-MI250) # TBD @@ -505,41 +390,6 @@ steps: commands: - pytest -v -s v1/attention -- label: Distributed DP Tests (2 GPUs) # TBD - timeout_in_minutes: 180 - mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250] - agent_pool: mi250_2 - num_gpus: 2 - optional: true - working_dir: "/vllm-workspace/tests" - source_file_dependencies: - - vllm/distributed/ - - vllm/engine/ - - vllm/executor/ - - vllm/worker/worker_base.py - - vllm/v1/engine/ - - vllm/v1/worker/ - - tests/v1/distributed - - tests/entrypoints/openai/test_multi_api_servers.py - - vllm/platforms/rocm.py - commands: - - TP_SIZE=1 DP_SIZE=2 pytest -v -s v1/distributed/test_async_llm_dp.py - - TP_SIZE=1 DP_SIZE=2 pytest -v -s v1/distributed/test_eagle_dp.py - - TP_SIZE=1 DP_SIZE=2 pytest -v -s v1/distributed/test_external_lb_dp.py - - DP_SIZE=2 pytest -v -s entrypoints/openai/test_multi_api_servers.py - -- label: V1 e2e (2 GPUs) # TBD - timeout_in_minutes: 180 - mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250] - agent_pool: mi250_2 - optional: true - working_dir: "/vllm-workspace/tests" - source_file_dependencies: - - vllm/ - - tests/v1/e2e - commands: - - pytest -v -s v1/e2e/spec_decode/test_spec_decode.py -k "tensor_parallelism" - #------------------------------------------------------------- mi250 · misc ------------------------------------------------------------# - label: Async Engine, Inputs, Utils, Worker, Config (CPU) # TBD @@ -782,6 +632,22 @@ steps: #-------------------------------------------------------- mi300 · distributed --------------------------------------------------------# +- label: Distributed Comm Ops # TBD + timeout_in_minutes: 180 + mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300] + agent_pool: mi300_2 + num_gpus: 2 + working_dir: "/vllm-workspace/tests" + source_file_dependencies: + - vllm/distributed + - tests/distributed + - vllm/platforms/rocm.py + commands: + - pytest -v -s distributed/test_comm_ops.py + - pytest -v -s distributed/test_shm_broadcast.py + - pytest -v -s distributed/test_shm_buffer.py + - pytest -v -s distributed/test_shm_storage.py + - label: EPLB Algorithm # TBD timeout_in_minutes: 180 mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300] @@ -1806,6 +1672,54 @@ steps: - PYTORCH_NVML_BASED_CUDA_CHECK=1 python3 examples/generate/multimodal/vision_language_offline.py --model-type qwen2_5_vl - VLLM_WORKER_MULTIPROC_METHOD=spawn python3 examples/generate/multimodal/audio_language_offline.py --model-type whisper +#---------------------------------------------------------- mi300 · plugins ----------------------------------------------------------# + +- label: Plugin Tests (2 GPUs) # TBD + timeout_in_minutes: 180 + mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300] + agent_pool: mi300_2 + num_gpus: 2 + optional: true + working_dir: "/vllm-workspace/tests" + source_file_dependencies: + - vllm/plugins/ + - tests/plugins/ + - vllm/platforms/rocm.py + commands: + # BEGIN: platform plugin and general plugin tests, all the code in-between runs on dummy platform + - pip install -e ./plugins/vllm_add_dummy_platform + - pytest -v -s plugins_tests/test_platform_plugins.py + - pip uninstall vllm_add_dummy_platform -y + # END: platform plugin tests + # BEGIN: `io_processor` plugins test, all the code in between uses the `prithvi_io_processor` plugin + - pip install -e ./plugins/prithvi_io_processor_plugin + - pytest -v -s plugins_tests/test_io_processor_plugins.py + - pytest -v -s plugins_tests/test_terratorch_io_processor_plugins.py + - pip uninstall prithvi_io_processor_plugin -y + # END: `io_processor` plugins test + # BEGIN: `bge_m3_sparse io_processor` test + - pip install -e ./plugins/bge_m3_sparse_plugin + - pytest -v -s plugins_tests/test_bge_m3_sparse_io_processor_plugins.py + - pip uninstall bge_m3_sparse_plugin -y + # END: `bge_m3_sparse io_processor` test + # BEGIN: `colbert_query io_processor` test + - pip install -e ./plugins/colbert_query_plugin + - pytest -v -s plugins_tests/test_colbert_query_io_processor_plugins.py + - pip uninstall colbert_query_plugin -y + # END: `colbert_query io_processor` test + # BEGIN: `stat_logger` plugins test + - pip install -e ./plugins/vllm_add_dummy_stat_logger + - pytest -v -s plugins_tests/test_stats_logger_plugins.py + - pip uninstall dummy_stat_logger -y + # END: `stat_logger` plugins test + # BEGIN: other tests + - pytest -v -s plugins_tests/test_scheduler_plugins.py + - pip install -e ./plugins/vllm_add_dummy_model + - pytest -v -s distributed/test_distributed_oot.py + - pytest -v -s plugins_tests/test_oot_registration_online.py # it needs a clean process + - pytest -v -s plugins_tests/test_oot_registration_offline.py # it needs a clean process + - pytest -v -s plugins_tests/lora_resolvers # unit tests for in-tree lora resolver plugins + #------------------------------------------------------- mi300 · quantization --------------------------------------------------------# - label: Quantization # TBD