真机验证发现的问题: 1. qwen3_5.py 做 'from ex_engine.python.ix_bridge' 但包结构是 ex_engine.ix_bridge → 创建 python/ 子目录 + symlinks 2. flash_qla_sm70 只部署到 /workspace 没有到 vllm models 目录 → 显式 cp -r 到 VLLM/model_executor/models/ 3. 从 vllm v0.5.5 搬 MoE CUDA kernels (torch::Tensor API): - topk_softmax_kernels.cu (506行, CUB BlockReduce) - moe_align_block_size_kernels.cu (134行) - moe_pybind.cpp (pybind11 入口) 真机验证结果: ✓ ix_bridge import OK, available=True ✓ topk_softmax (64 experts, top8) OK — CUDA kernel 命中 ✓ ix_full_bridge silu_and_mul OK ✓ qwen3_5.py import OK ✓ ex_engine build 2/2 factors ✓ moe_topk_softmax_v3.so 编译成功 ✓ flash_qla_sm70_gdn_strided.so 编译成功 ✗ 单卡 32GB OOM (正常, 竞赛 4卡 tp=4)
36 lines
1.5 KiB
C
36 lines
1.5 KiB
C
/*
|
|
* Adapted from
|
|
* https://github.com/pytorch/pytorch/blob/v2.0.1/aten/src/ATen/Dispatch.h
|
|
*/
|
|
#pragma once
|
|
|
|
#include <torch/all.h>
|
|
|
|
#define VLLM_DISPATCH_CASE_FLOATING_TYPES(...) \
|
|
AT_DISPATCH_CASE(at::ScalarType::Float, __VA_ARGS__) \
|
|
AT_DISPATCH_CASE(at::ScalarType::Half, __VA_ARGS__) \
|
|
AT_DISPATCH_CASE(at::ScalarType::BFloat16, __VA_ARGS__)
|
|
|
|
#define VLLM_DISPATCH_FLOATING_TYPES(TYPE, NAME, ...) \
|
|
AT_DISPATCH_SWITCH(TYPE, NAME, VLLM_DISPATCH_CASE_FLOATING_TYPES(__VA_ARGS__))
|
|
|
|
#define VLLM_DISPATCH_CASE_FLOATING_AND_BYTE_TYPES(...) \
|
|
AT_DISPATCH_CASE(at::ScalarType::Float, __VA_ARGS__) \
|
|
AT_DISPATCH_CASE(at::ScalarType::Half, __VA_ARGS__) \
|
|
AT_DISPATCH_CASE(at::ScalarType::BFloat16, __VA_ARGS__) \
|
|
AT_DISPATCH_CASE(at::ScalarType::Byte, __VA_ARGS__)
|
|
|
|
#define VLLM_DISPATCH_FLOATING_AND_BYTE_TYPES(TYPE, NAME, ...) \
|
|
AT_DISPATCH_SWITCH(TYPE, NAME, \
|
|
VLLM_DISPATCH_CASE_FLOATING_AND_BYTE_TYPES(__VA_ARGS__))
|
|
|
|
#define VLLM_DISPATCH_CASE_INTEGRAL_TYPES(...) \
|
|
AT_DISPATCH_CASE(at::ScalarType::Byte, __VA_ARGS__) \
|
|
AT_DISPATCH_CASE(at::ScalarType::Char, __VA_ARGS__) \
|
|
AT_DISPATCH_CASE(at::ScalarType::Short, __VA_ARGS__) \
|
|
AT_DISPATCH_CASE(at::ScalarType::Int, __VA_ARGS__) \
|
|
AT_DISPATCH_CASE(at::ScalarType::Long, __VA_ARGS__)
|
|
|
|
#define VLLM_DISPATCH_INTEGRAL_TYPES(TYPE, NAME, ...) \
|
|
AT_DISPATCH_SWITCH(TYPE, NAME, VLLM_DISPATCH_CASE_INTEGRAL_TYPES(__VA_ARGS__))
|