#53161 [ROCm][Perf][DeepSeek V4] Fuse native FP8 shared expert with MXFP4 routed experts
实现拆解
-
核心源码路径
-vllm/models/deepseek_v4/amd/model.py(data-contract):源码主路径;涉及符号 _use_heterogeneous_fhmoe, _validate_heterogeneous_routes, _heterogeneous_shared_expert_enabled, _prepare_native_fp8_shared_expert;包含 导入关系调整、控制流调整、配置键调整;+377/-5;关键符号:_use_heterogeneous_fhmoe, _validate_heterogeneous_routes, _heterogeneous_shared_expert_enabled, _prepare_native_fp8_shared_expert
-vllm/_aiter_ops.py(dependency-wiring):源码主路径;涉及符号 _validate_rocm_aiter_fused_moe_shared_expert_args, _probe_dsv4_i384_fhmoe_capability, fused_moe_supports_heterogeneous_shared_expert, shuffle_scale;包含 导入关系调整、控制流调整、配置键调整;+107/-0;关键符号:_validate_rocm_aiter_fused_moe_shared_expert_args, _probe_dsv4_i384_fhmoe_capability, fused_moe_supports_heterogeneous_shared_expert, shuffle_scale
-vllm/model_executor/layers/fused_moe/experts/rocm_aiter_moe.py(data-contract):源码主路径;包含 控制流调整、配置键调整;+15/-0 -
测试与验证
-tests/model_executor/layers/test_fused_shared_expert.py(test-coverage):测试配套;涉及符号 test_deepseek_v4_heterogeneous_fhmoe_keeps_native_intermediate_width, test_deepseek_v4_heterogeneous_fhmoe_token_policy, supports, _supported_fhmoe_signature;包含 测试覆盖调整、导入关系调整、控制流调整;+340/-1;关键符号:test_deepseek_v4_heterogeneous_fhmoe_keeps_native_intermediate_width, test_deepseek_v4_heterogeneous_fhmoe_token_policy, supports, _supported_fhmoe_signature
| 文件 | 模块 | 状态 | 重要度 |
|---|---|---|---|
vllm/models/deepseek_v4/amd/model.py |
vllm/models | modified | 9.21 |
vllm/_aiter_ops.py |
vllm | modified | 8.55 |
tests/model_executor/layers/test_fused_shared_expert.py |
fused/shared/expert | modified | 8.1 |
vllm/model_executor/layers/fused_moe/experts/rocm_aiter_moe.py |
vllm/model_executor | modified | 6.37 |
分析完成后,这里会展示 LLM 生成的相对完整源码片段和详细注释。
评论区精华
没有提炼出高价值讨论线程
当前评论区没有形成足够清晰的争议点或结论,后续有更多讨论时会体现在这里。
参与讨论