002f9879b2
Replaces cherry-picked upstream_ref with complete source trees. xllm/ — Iluvatar official C++ inference engine (15MB, 1470 files) Complete: kernels → layers → models → runtime → scheduler → api Excluded: .git, binary images, third_party submodule checkouts ds_vllm/ — Iluvatar official vllm fork (8MB, 703 files) Included: csrc/ (ALL CUDA kernels), fused_moe/, qwen3_5 model, _custom_ops Excluded: tests, benchmarks, docs, examples (not needed for reference) Critical call chains now fully traceable: MoE: moe_topk_softmax_kernels.cuh → ixformer.h → fused_moe.cpp → layer GDN: qwen3_gated_delta_net_base.cpp → qwen3_5_gated_delta_net.cpp Attention: ixformer.h → xllm_paged_attention → attention.cpp
19 lines
446 B
C++
19 lines
446 B
C++
#pragma once
|
|
|
|
#ifndef USE_ROCM
|
|
#include <cub/cub.cuh>
|
|
#if CUB_VERSION >= 200800
|
|
#include <cuda/std/functional>
|
|
using CubAddOp = cuda::std::plus<>;
|
|
using CubMaxOp = cuda::maximum<>;
|
|
#else // if CUB_VERSION < 200800
|
|
using CubAddOp = cub::Sum;
|
|
using CubMaxOp = cub::Max;
|
|
#endif // CUB_VERSION
|
|
#else
|
|
#include <hipcub/hipcub.hpp>
|
|
namespace cub = hipcub;
|
|
using CubAddOp = hipcub::Sum;
|
|
using CubMaxOp = hipcub::Max;
|
|
#endif // USE_ROCM
|