First add

d3dd8642 · Rayyyyy · d3dd8642 · d3dd8642 · d3dd8642 · d3dd8642
Commit d3dd8642 authored Jun 26, 2024 by Rayyyyy
20 changed files
--- a/vllm/csrc/attention/dtype_float32.cuh
+++ b/vllm/csrc/attention/dtype_float32.cuh
--- a/vllm/csrc/attention/dtype_fp8.cuh
+++ b/vllm/csrc/attention/dtype_fp8.cuh
--- a/vllm/csrc/cache.h
+++ b/vllm/csrc/cache.h
--- a/vllm/csrc/cache_kernels.cu
+++ b/vllm/csrc/cache_kernels.cu
--- a/vllm/csrc/cpu/activation.cpp
+++ b/vllm/csrc/cpu/activation.cpp
--- a/vllm/csrc/cpu/attention.cpp
+++ b/vllm/csrc/cpu/attention.cpp
--- a/vllm/csrc/cpu/cache.cpp
+++ b/vllm/csrc/cpu/cache.cpp
--- a/vllm/csrc/cpu/cpu_types.hpp
+++ b/vllm/csrc/cpu/cpu_types.hpp
--- a/vllm/csrc/cpu/layernorm.cpp
+++ b/vllm/csrc/cpu/layernorm.cpp
--- a/vllm/csrc/cpu/pos_encoding.cpp
+++ b/vllm/csrc/cpu/pos_encoding.cpp
--- a/vllm/csrc/cpu/pybind.cpp
+++ b/vllm/csrc/cpu/pybind.cpp
--- a/vllm/csrc/cuda_compat.h
+++ b/vllm/csrc/cuda_compat.h
--- a/vllm/csrc/cuda_utils.h
+++ b/vllm/csrc/cuda_utils.h
--- a/vllm/csrc/cuda_utils_kernels.cu
+++ b/vllm/csrc/cuda_utils_kernels.cu
--- a/vllm/csrc/custom_all_reduce.cu
+++ b/vllm/csrc/custom_all_reduce.cu
--- a/vllm/csrc/custom_all_reduce.cuh
+++ b/vllm/csrc/custom_all_reduce.cuh
--- a/vllm/csrc/custom_all_reduce_test.cu
+++ b/vllm/csrc/custom_all_reduce_test.cu
--- a/vllm/csrc/dispatch_utils.h
+++ b/vllm/csrc/dispatch_utils.h
--- a/vllm/csrc/layernorm_kernels.cu
+++ b/vllm/csrc/layernorm_kernels.cu
--- a/vllm/csrc/moe/moe_ops.cpp
+++ b/vllm/csrc/moe/moe_ops.cpp