minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

[CI] update estimated elapsed time of some unittests (#14347)

Browse Source
This commit is contained in:
Cheng Wan
2025-12-03 01:21:40 -08:00
committed by GitHub
parent 77512ae0d7
commit 96cc10834a
1 changed files with 80 additions and 81 deletions
Download Patch File Download Diff File Expand all files Collapse all files
test/srt/run_suite.py
+80 -81
View File
@@ -8,130 +8,129 @@ from sglang.test.ci.ci_utils import TestFile, run_unittest_files
suites = { suites = {
"per-commit-1-gpu": [ "per-commit-1-gpu": [
TestFile("test_eagle_constrained_decoding.py", 100), TestFile("test_eagle_constrained_decoding.py", 100),
TestFile("debug_utils/test_tensor_dump_forward_hook.py", 15), TestFile("debug_utils/test_tensor_dump_forward_hook.py", 9),
TestFile("hicache/test_hicache_storage.py", 127), TestFile("hicache/test_hicache_storage.py", 96),
TestFile("hicache/test_hicache_variants.py", 393), TestFile("hicache/test_hicache_variants.py", 368),
TestFile("layers/attention/mamba/test_causal_conv1d.py", 25), TestFile("layers/attention/mamba/test_causal_conv1d.py", 25),
TestFile("layers/attention/mamba/test_mamba_ssm.py", 50), TestFile("layers/attention/mamba/test_mamba_ssm.py", 7),
TestFile("layers/attention/mamba/test_mamba_ssm_ssd.py", 20), TestFile("layers/attention/mamba/test_mamba_ssm_ssd.py", 13),
TestFile("lora/test_lora.py", 150), TestFile("lora/test_lora.py", 82),
TestFile("lora/test_lora_eviction.py", 240), TestFile("lora/test_lora_eviction.py", 224),
TestFile("lora/test_lora_update.py", 600), TestFile("lora/test_lora_update.py", 451),
TestFile("lora/test_lora_backend.py", 99), TestFile("lora/test_lora_backend.py", 200),
TestFile("lora/test_multi_lora_backend.py", 60), TestFile("lora/test_multi_lora_backend.py", 60),
TestFile("models/test_compressed_tensors_models.py", 42), TestFile("models/test_compressed_tensors_models.py", 42),
TestFile("models/test_cross_encoder_models.py", 100), TestFile("models/test_cross_encoder_models.py", 100),
TestFile("models/test_embedding_models.py", 73), TestFile("models/test_embedding_models.py", 73),
TestFile("models/test_encoder_embedding_models.py", 460), TestFile("models/test_encoder_embedding_models.py", 221),
TestFile("models/test_generation_models.py", 103), TestFile("models/test_generation_models.py", 103),
TestFile("models/test_nvidia_nemotron_nano_v2.py", 160), TestFile("models/test_nvidia_nemotron_nano_v2.py", 132),
TestFile("models/test_nvidia_nemotron_nano_v2_vl.py", 350), # GSM8k + MMMU TestFile("models/test_nvidia_nemotron_nano_v2_vl.py", 214), # GSM8k + MMMU
TestFile("models/test_qwen_models.py", 150), TestFile("models/test_qwen_models.py", 90),
TestFile("models/test_reward_models.py", 132), TestFile("models/test_reward_models.py", 103),
TestFile("models/test_transformers_models.py", 320), TestFile("models/test_transformers_models.py", 245),
TestFile("models/test_vlm_models.py", 741), TestFile("models/test_vlm_models.py", 282),
TestFile("openai_server/basic/test_openai_embedding.py", 79), TestFile("openai_server/basic/test_openai_embedding.py", 70),
TestFile("openai_server/basic/test_openai_server.py", 270), TestFile("openai_server/basic/test_openai_server.py", 184),
TestFile("openai_server/basic/test_protocol.py", 10), TestFile("openai_server/basic/test_protocol.py", 3),
TestFile("openai_server/basic/test_serving_chat.py", 10), TestFile("openai_server/basic/test_serving_chat.py", 10),
TestFile("openai_server/basic/test_serving_completions.py", 10), TestFile("openai_server/basic/test_serving_completions.py", 10),
TestFile("openai_server/basic/test_serving_embedding.py", 10), TestFile("openai_server/basic/test_serving_embedding.py", 10),
TestFile("openai_server/features/test_enable_thinking.py", 70), TestFile("openai_server/features/test_enable_thinking.py", 70),
TestFile("openai_server/features/test_json_mode.py", 120), TestFile("openai_server/features/test_json_mode.py", 109),
TestFile("openai_server/features/test_openai_server_ebnf.py", 20), TestFile("openai_server/features/test_openai_server_ebnf.py", 7),
TestFile("openai_server/features/test_openai_server_hidden_states.py", 240), TestFile("openai_server/features/test_openai_server_hidden_states.py", 186),
TestFile("openai_server/features/test_reasoning_content.py", 89), TestFile("openai_server/features/test_reasoning_content.py", 89),
TestFile("openai_server/function_call/test_openai_function_calling.py", 60), TestFile("openai_server/function_call/test_openai_function_calling.py", 60),
TestFile("openai_server/function_call/test_tool_choice.py", 120), TestFile("openai_server/function_call/test_tool_choice.py", 120),
TestFile("openai_server/validation/test_large_max_new_tokens.py", 41), TestFile("openai_server/validation/test_large_max_new_tokens.py", 41),
TestFile("openai_server/validation/test_matched_stop.py", 60), TestFile("openai_server/validation/test_matched_stop.py", 40),
TestFile("openai_server/validation/test_openai_server_ignore_eos.py", 85), TestFile("openai_server/validation/test_openai_server_ignore_eos.py", 6),
TestFile("openai_server/validation/test_request_length_validation.py", 31), TestFile("openai_server/validation/test_request_length_validation.py", 38),
TestFile("ops/test_repeat_interleave.py", 60), TestFile("ops/test_repeat_interleave.py", 60),
TestFile("quant/test_block_int8.py", 22), TestFile("quant/test_block_int8.py", 44),
TestFile("quant/test_fp8_kernel.py", 8), TestFile("quant/test_fp8_kernel.py", 10),
TestFile("quant/test_int8_kernel.py", 8), TestFile("quant/test_int8_kernel.py", 8),
TestFile("quant/test_triton_scaled_mm.py", 8), TestFile("quant/test_triton_scaled_mm.py", 8),
TestFile("quant/test_w8a8_quantization.py", 160), TestFile("quant/test_w8a8_quantization.py", 160),
TestFile("quant/test_autoround.py", 60), TestFile("quant/test_autoround.py", 77),
TestFile("rl/test_fp32_lm_head.py", 30), TestFile("rl/test_fp32_lm_head.py", 9),
# TestFile("rl/test_update_weights_from_disk.py", 210), # Temporarily disabled, see https://github.com/sgl-project/sglang/pull/13998 # TestFile("rl/test_update_weights_from_disk.py", 210), # Temporarily disabled, see https://github.com/sgl-project/sglang/pull/13998
TestFile("rl/test_update_weights_from_tensor.py", 80), TestFile("rl/test_update_weights_from_tensor.py", 195),
TestFile("test_abort.py", 190), TestFile("test_abort.py", 131),
TestFile("test_build_eagle_tree.py", 8), TestFile("test_build_eagle_tree.py", 3),
TestFile("test_chunked_prefill.py", 410), TestFile("test_chunked_prefill.py", 312),
TestFile("test_create_kvindices.py", 2), TestFile("test_create_kvindices.py", 7),
TestFile("test_deterministic.py", 400), TestFile("test_deterministic.py", 228),
TestFile("test_eagle_infer_a.py", 750), TestFile("test_eagle_infer_a.py", 470),
TestFile("test_eagle_infer_b.py", 750), TestFile("test_eagle_infer_b.py", 473),
TestFile("test_eagle_infer_beta.py", 90), TestFile("test_eagle_infer_beta.py", 194),
TestFile("test_constrained_decoding.py", 150), TestFile("test_constrained_decoding.py", 111),
TestFile("test_eval_fp8_accuracy.py", 303), TestFile("test_eval_fp8_accuracy.py", 250),
TestFile("test_external_models.py", 155), TestFile("test_external_models.py", 30),
TestFile("test_fa3.py", 420), TestFile("test_fa3.py", 420),
TestFile("test_flashmla.py", 230), TestFile("test_flashmla.py", 230),
TestFile("test_fp8_utils.py", 5), TestFile("test_fp8_utils.py", 9),
TestFile("rotary_embedding/test_mrope.py", 10), TestFile("rotary_embedding/test_mrope.py", 10),
TestFile("test_fused_moe.py", 80), TestFile("test_fused_moe.py", 80),
TestFile("test_gpt_oss_1gpu.py", 750), TestFile("test_gpt_oss_1gpu.py", 402),
TestFile("test_harmony_parser.py", 20), TestFile("test_harmony_parser.py", 6),
TestFile("test_hidden_states.py", 55), TestFile("test_hidden_states.py", 55),
TestFile("test_hybrid_attn_backend.py", 379), TestFile("test_hybrid_attn_backend.py", 379),
TestFile("test_input_embeddings.py", 38), TestFile("test_input_embeddings.py", 38),
TestFile("test_io_struct.py", 8), TestFile("test_io_struct.py", 8),
TestFile("test_jinja_template_utils.py", 1), TestFile("test_jinja_template_utils.py", 7),
TestFile("test_mamba_unittest.py", 4), TestFile("test_mamba_unittest.py", 9),
TestFile("test_metrics.py", 32), TestFile("test_metrics.py", 32),
TestFile("test_metrics_utils.py", 1), TestFile("test_metrics_utils.py", 1),
TestFile("test_mla.py", 180), TestFile("test_mla.py", 194),
TestFile("test_mla_deepseek_v3.py", 500), TestFile("test_mla_deepseek_v3.py", 442),
TestFile("test_mla_flashinfer.py", 302), TestFile("test_mla_flashinfer.py", 302),
TestFile("test_mla_fp8.py", 93), TestFile("test_mla_fp8.py", 77),
TestFile("test_mla_int8_deepseek_v3.py", 300), TestFile("test_mla_int8_deepseek_v3.py", 300),
TestFile("test_model_hooks.py", 1), TestFile("test_model_hooks.py", 6),
TestFile("test_modelopt_loader.py", 30), TestFile("test_modelopt_loader.py", 11),
TestFile("test_multi_tokenizer.py", 230), TestFile("test_multi_tokenizer.py", 230),
TestFile("test_ngram_speculative_decoding.py", 290), TestFile("test_ngram_speculative_decoding.py", 177),
TestFile("test_no_chunked_prefill.py", 108), TestFile("test_no_chunked_prefill.py", 108),
TestFile("test_no_overlap_scheduler.py", 234), TestFile("test_no_overlap_scheduler.py", 217),
TestFile("test_original_logprobs.py", 41), TestFile("test_original_logprobs.py", 41),
TestFile("test_page_size.py", 60), TestFile("test_page_size.py", 60),
TestFile("test_penalty.py", 82), TestFile("test_penalty.py", 82),
TestFile("test_priority_scheduling.py", 130), TestFile("test_priority_scheduling.py", 130),
TestFile("test_pytorch_sampling_backend.py", 66), TestFile("test_pytorch_sampling_backend.py", 66),
TestFile("test_radix_attention.py", 105), TestFile("test_radix_attention.py", 105),
TestFile("test_radix_cache_unit.py", 5), TestFile("test_radix_cache_unit.py", 8),
TestFile("test_reasoning_parser.py", 5), TestFile("test_reasoning_parser.py", 5),
TestFile("test_request_queue_validation.py", 30), TestFile("test_request_queue_validation.py", 47),
TestFile("test_retract_decode.py", 450), TestFile("test_retract_decode.py", 259),
TestFile("test_score_api.py", 310), TestFile("test_score_api.py", 260),
TestFile("test_server_args.py", 1), TestFile("test_server_args.py", 9),
TestFile("test_speculative_registry.py", 1), TestFile("test_speculative_registry.py", 8),
TestFile("test_skip_tokenizer_init.py", 117), TestFile("test_skip_tokenizer_init.py", 77),
TestFile("test_srt_endpoint.py", 130), TestFile("test_srt_endpoint.py", 127),
TestFile("test_srt_engine.py", 450), TestFile("test_srt_engine.py", 252),
TestFile("test_standalone_speculative_decoding.py", 150), TestFile("test_standalone_speculative_decoding.py", 150),
TestFile("test_start_profile.py", 180), TestFile("test_start_profile.py", 41),
TestFile("test_profile_merger.py", 60), TestFile("test_profile_merger.py", 8),
TestFile("test_profile_merger_http_api.py", 15), TestFile("test_profile_merger_http_api.py", 9),
TestFile("test_swa_unittest.py", 1), TestFile("test_swa_unittest.py", 8),
TestFile("test_torch_compile.py", 76), TestFile("test_torch_compile.py", 190),
TestFile("test_torch_compile_moe.py", 210), TestFile("test_torch_compile_moe.py", 210),
TestFile("test_triton_fused_moe.py", 80), TestFile("test_triton_fused_moe.py", 12),
TestFile("test_torch_native_attention_backend.py", 123), TestFile("test_torch_native_attention_backend.py", 221),
TestFile("test_torchao.py", 70), TestFile("test_torchao.py", 103),
TestFile("test_triton_attention_kernels.py", 4), TestFile("test_triton_attention_kernels.py", 4),
TestFile("test_triton_attention_backend.py", 150), TestFile("test_triton_attention_backend.py", 203),
TestFile("test_triton_attention_kernels.py", 4), TestFile("test_triton_attention_kernels.py", 4),
TestFile("test_triton_moe_channel_fp8_kernel.py", 25), TestFile("test_triton_moe_channel_fp8_kernel.py", 16),
TestFile("test_triton_sliding_window.py", 100), TestFile("test_triton_sliding_window.py", 84),
TestFile("test_utils_update_weights.py", 48), TestFile("test_utils_update_weights.py", 29),
TestFile("test_video_utils.py", 5), TestFile("test_video_utils.py", 5),
TestFile("test_vision_chunked_prefill.py", 170), TestFile("test_vision_chunked_prefill.py", 117),
TestFile("test_vision_openai_server_a.py", 900), TestFile("test_vision_openai_server_a.py", 778),
TestFile("test_vlm_input_format.py", 300), TestFile("test_vlm_input_format.py", 166),
TestFile("test_modelopt_loader.py", 30), TestFile("test_modelopt_export.py", 9),
TestFile("test_modelopt_export.py", 30),
], ],
"per-commit-2-gpu": [ "per-commit-2-gpu": [
TestFile("ep/test_moe_ep.py", 140), TestFile("ep/test_moe_ep.py", 140),
@@ -246,7 +245,7 @@ suite_amd = {
TestFile("openai_server/features/test_json_mode.py", 120), TestFile("openai_server/features/test_json_mode.py", 120),
TestFile("openai_server/features/test_openai_server_ebnf.py", 20), TestFile("openai_server/features/test_openai_server_ebnf.py", 20),
TestFile("openai_server/features/test_reasoning_content.py", 89), TestFile("openai_server/features/test_reasoning_content.py", 89),
TestFile("openai_server/function_call/test_openai_function_calling.py", 60), TestFile("openai_server/function_call/test_openai_function_calling.py", 73),
TestFile("openai_server/function_call/test_tool_choice.py", 120), TestFile("openai_server/function_call/test_tool_choice.py", 120),
TestFile("openai_server/validation/test_large_max_new_tokens.py", 41), TestFile("openai_server/validation/test_large_max_new_tokens.py", 41),
TestFile("openai_server/validation/test_matched_stop.py", 60), TestFile("openai_server/validation/test_matched_stop.py", 60),
@@ -258,7 +257,7 @@ suite_amd = {
# TestFile("rl/test_update_weights_from_disk.py", 210), # Temporarily disabled, see https://github.com/sgl-project/sglang/pull/13998 # TestFile("rl/test_update_weights_from_disk.py", 210), # Temporarily disabled, see https://github.com/sgl-project/sglang/pull/13998
TestFile("test_abort.py", 51), TestFile("test_abort.py", 51),
TestFile("test_bench_typebaseddispatcher.py", 10), TestFile("test_bench_typebaseddispatcher.py", 10),
TestFile("test_chunked_prefill.py", 410), TestFile("test_chunked_prefill.py", 312),
TestFile("test_create_kvindices.py", 2), TestFile("test_create_kvindices.py", 2),
TestFile("test_eval_fp8_accuracy.py", 303), TestFile("test_eval_fp8_accuracy.py", 303),
TestFile("test_fused_moe.py", 30), TestFile("test_fused_moe.py", 30),