diff --git a/.buildkite/test-amd.yaml b/.buildkite/test-amd.yaml index c23f114a41..c74f516c08 100644 --- a/.buildkite/test-amd.yaml +++ b/.buildkite/test-amd.yaml @@ -1025,6 +1025,23 @@ steps: commands: - pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-small.txt +- label: MRCR Eval Small Models # TBD + timeout_in_minutes: 180 + mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300] + agent_pool: mi300_1 + optional: true + working_dir: "/vllm-workspace/tests" + source_file_dependencies: + - vllm/model_executor/models/ + - vllm/model_executor/model_loader/ + - vllm/v1/attention/backends/ + - vllm/v1/attention/selector.py + - vllm/_aiter_ops.py + - vllm/platforms/rocm.py + - tests/evals/mrcr/ + commands: + - pytest -s -v evals/mrcr/test_mrcr_correctness.py --config-list-file=evals/mrcr/configs/models-small.txt + - label: LM Eval Small Models (MI300) # TBD timeout_in_minutes: 180 mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300] @@ -1283,6 +1300,21 @@ steps: #---------------------------------------------------------- mi300 · kernels ----------------------------------------------------------# +- label: vLLM IR Tests # TBD + timeout_in_minutes: 30 + mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300] + agent_pool: mi300_1 + optional: true + working_dir: "/vllm-workspace/" + source_file_dependencies: + - vllm/ir + - vllm/kernels + - vllm/_aiter_ops.py + - vllm/platforms/rocm.py + commands: + - pytest -v -s tests/ir + - pytest -v -s tests/kernels/ir + - label: Kernels Attention Test %N # TBD timeout_in_minutes: 100 mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300] @@ -1317,6 +1349,21 @@ steps: commands: - pytest -v -s kernels/core --ignore=kernels/core/test_minimax_reduce_rms.py kernels/test_concat_mla_q.py kernels/test_top_k_per_row.py +- label: Kernels KDA Test # TBD + timeout_in_minutes: 30 + mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300] + agent_pool: mi300_1 + optional: true + working_dir: "/vllm-workspace/tests" + source_file_dependencies: + - vllm/model_executor/layers/fla/ops/kda.py + - vllm/model_executor/layers/fla/ops/chunk_delta_h.py + - vllm/model_executor/layers/fla/ops/l2norm.py + - tests/kernels/test_kda.py + - vllm/platforms/rocm.py + commands: + - pytest -v -s kernels/test_kda.py + - label: Kernels MoE Test %N # TBD timeout_in_minutes: 95 mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300] @@ -1429,6 +1476,128 @@ steps: - pytest -v -s model_executor -m '(not slow_test)' - pytest -v -s entrypoints/openai/completion/test_tensorizer_entrypoint.py +#---------------------------------------------------- mi300 · model_runner_v2 -------------------------------------------------------# + +- label: Model Runner V2 Core Tests # TBD + timeout_in_minutes: 60 + mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300] + agent_pool: mi300_1 + optional: true + working_dir: "/vllm-workspace/tests" + source_file_dependencies: + - vllm/v1/worker/gpu/ + - vllm/v1/worker/gpu_worker.py + - vllm/v1/core/sched/ + - vllm/v1/attention/ + - tests/v1/engine/test_llm_engine.py + - tests/v1/e2e/ + - tests/entrypoints/llm/test_struct_output_generate.py + - vllm/platforms/rocm.py + commands: + - set -x + - export VLLM_USE_V2_MODEL_RUNNER=1 + - pytest -v -s v1/engine/test_llm_engine.py -k "not test_engine_metrics" + - ENFORCE_EAGER=1 pytest -v -s v1/e2e/general/test_async_scheduling.py -k "not ngram" + - pytest -v -s v1/e2e/general/test_context_length.py + - pytest -v -s v1/e2e/general/test_min_tokens.py + - pytest -v -s entrypoints/llm/test_struct_output_generate.py -k "xgrammar and not speculative_config6 and not speculative_config7 and not speculative_config8 and not speculative_config0" + +- label: Model Runner V2 Examples # TBD + timeout_in_minutes: 60 + mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300] + agent_pool: mi300_1 + optional: true + working_dir: "/vllm-workspace/examples" + source_file_dependencies: + - vllm/v1/worker/gpu/ + - vllm/v1/core/sched/ + - vllm/v1/worker/gpu_worker.py + - examples/basic/offline_inference/ + - examples/generate/multimodal/ + - examples/features/ + - examples/pooling/embed/vision_embedding_offline.py + - examples/features/tensorize_vllm_model.py + - examples/deployment/llm_engine_example.py + - vllm/platforms/rocm.py + commands: + - set -x + - export VLLM_USE_V2_MODEL_RUNNER=1 + - pip install tensorizer + - python3 basic/offline_inference/chat.py + - python3 basic/offline_inference/generate.py --model facebook/opt-125m + - python3 generate/multimodal/audio_language_offline.py --seed 0 + - python3 generate/multimodal/vision_language_offline.py --seed 0 + - python3 generate/multimodal/vision_language_multi_image_offline.py --seed 0 + - python3 generate/multimodal/encoder_decoder_multimodal_offline.py --model-type whisper --seed 0 + - python3 pooling/embed/vision_embedding_offline.py --seed 0 + - python3 features/automatic_prefix_caching/prefix_caching_offline.py + - python3 deployment/llm_engine_example.py + - python3 features/tensorize_vllm_model.py --model facebook/opt-125m serialize --serialized-directory /tmp/ --suffix v1 && python3 features/tensorize_vllm_model.py --model facebook/opt-125m deserialize --path-to-tensors /tmp/vllm/facebook/opt-125m/v1/model.tensors + - python3 features/speculative_decoding/spec_decode_offline.py --test --method eagle --num_spec_tokens 3 --dataset-name hf --dataset-path philschmid/mt-bench --num-prompts 80 --temp 0 --top-p 1.0 --top-k -1 --tp 1 --enable-chunked-prefill --max-model-len 2048 + - python3 features/speculative_decoding/spec_decode_offline.py --test --method eagle3 --num_spec_tokens 3 --dataset-name hf --dataset-path philschmid/mt-bench --num-prompts 80 --temp 0 --top-p 1.0 --top-k -1 --tp 1 --enable-chunked-prefill --max-model-len 1536 + +- label: Model Runner V2 Distributed (2 GPUs) # TBD + timeout_in_minutes: 60 + mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300] + agent_pool: mi300_2 + num_gpus: 2 + optional: true + working_dir: "/vllm-workspace/tests" + source_file_dependencies: + - vllm/v1/worker/gpu/ + - vllm/v1/worker/gpu_worker.py + - tests/basic_correctness/test_basic_correctness.py + - tests/v1/distributed/test_async_llm_dp.py + - tests/v1/distributed/test_eagle_dp.py + - vllm/platforms/rocm.py + commands: + - set -x + - export VLLM_USE_V2_MODEL_RUNNER=1 + - TARGET_TEST_SUITE=MI300 pytest -v -s basic_correctness/test_basic_correctness.py -m 'distributed(num_gpus=2)' -k "not ray and not True" + - TP_SIZE=1 DP_SIZE=2 pytest -v -s v1/distributed/test_async_llm_dp.py -k "not ray" + - TP_SIZE=1 DP_SIZE=2 pytest -v -s v1/distributed/test_eagle_dp.py + +- label: Model Runner V2 Pipeline Parallelism (4 GPUs) # TBD + timeout_in_minutes: 60 + mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300] + agent_pool: mi300_4 + num_gpus: 4 + optional: true + working_dir: "/vllm-workspace/tests" + source_file_dependencies: + - vllm/v1/worker/gpu/ + - vllm/v1/worker/gpu_worker.py + - tests/distributed/test_pipeline_parallel.py + - tests/distributed/test_pp_cudagraph.py + - vllm/platforms/rocm.py + commands: + - set -x + - export VLLM_USE_V2_MODEL_RUNNER=1 + - pytest -v -s distributed/test_pipeline_parallel.py -k "not ray and not Jamba" + - pytest -v -s distributed/test_pp_cudagraph.py -k "not ray" + +- label: Model Runner V2 Spec Decode # TBD + timeout_in_minutes: 45 + mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300] + agent_pool: mi300_1 + optional: true + working_dir: "/vllm-workspace/tests" + source_file_dependencies: + - vllm/v1/worker/gpu/ + - vllm/v1/worker/gpu_worker.py + - tests/v1/spec_decode/test_max_len.py + - tests/v1/spec_decode/test_rejection_sampler_utils.py + - tests/v1/spec_decode/test_synthetic_rejection_sampler_utils.py + - tests/v1/e2e/spec_decode/test_spec_decode.py + - vllm/platforms/rocm.py + commands: + - set -x + - export VLLM_USE_V2_MODEL_RUNNER=1 + - pytest -v -s v1/spec_decode/test_max_len.py -k "eagle or mtp" + - pytest -v -s v1/spec_decode/test_rejection_sampler_utils.py + - pytest -v -s v1/spec_decode/test_synthetic_rejection_sampler_utils.py + - pytest -v -s v1/e2e/spec_decode/test_spec_decode.py -k "eagle or mtp" + #------------------------------------------------------ mi300 · models / basic -------------------------------------------------------# - label: Basic Models Tests (Extra Initialization) %N # TBD @@ -1745,6 +1914,22 @@ steps: - 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 +- label: GGUF Plugin # TBD + timeout_in_minutes: 30 + mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300] + agent_pool: mi300_1 + optional: true + soft_fail: true + working_dir: "/vllm-workspace/tests" + source_file_dependencies: + - vllm/model_executor/layers/quantization + - tests/plugins_tests/test_gguf_plugin.py + - tests/plugins_tests/gguf + - vllm/platforms/rocm.py + commands: + - pip install "vllm-gguf-plugin >= 0.0.2" + - pytest -v -s plugins_tests/gguf + #------------------------------------------------------- mi300 · quantization --------------------------------------------------------# - label: Quantization # TBD @@ -2770,7 +2955,7 @@ steps: #--------------------------------------------------------- mi355 · examples ----------------------------------------------------------# - label: Examples # TBD - timeout_in_minutes: 45 + timeout_in_minutes: 100 mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355] agent_pool: mi355_1 working_dir: "/vllm-workspace/examples" diff --git a/vllm/model_executor/layers/fla/ops/chunk_delta_h.py b/vllm/model_executor/layers/fla/ops/chunk_delta_h.py index 89a8c4d909..eb1c3af152 100644 --- a/vllm/model_executor/layers/fla/ops/chunk_delta_h.py +++ b/vllm/model_executor/layers/fla/ops/chunk_delta_h.py @@ -17,6 +17,8 @@ from .op import exp, exp2 from .utils import FLA_CHUNK_SIZE, use_cuda_graph NUM_WARPS = [2, 4, 8, 16] +# Triton's AMD backend fails to lower this kernel with num_stages=4. +_CHUNK_DELTA_H_NUM_STAGES = [2, 3] if torch.version.hip else [2, 3, 4] @triton.heuristics( @@ -33,7 +35,7 @@ NUM_WARPS = [2, 4, 8, 16] configs=[ triton.Config({"BV": BV}, num_warps=num_warps, num_stages=num_stages) for num_warps in [2, 4] - for num_stages in [2, 3, 4] + for num_stages in _CHUNK_DELTA_H_NUM_STAGES for BV in [32, 64] ], key=["H", "K", "V", "BT"],