ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-11 00:40:09 +00:00
Code Issues Packages Projects Releases Wiki Activity

Compile for gfx908 and gfx90a (#130)

Browse Source 
* adding compilation for multiple targets

* fix build

* clean

* update Jekinsfile

* update readme

* update Jenkins

* use ck::half_t instead of ushort for bf16

* rename enum classes

* clean

* rename

* clean
This commit is contained in:
Chao Liu
2022-03-31 12:33:34 -05:00
committed by GitHub
parent ecf337bab5
commit cd167e492a
227 changed files with 1398 additions and 2944 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
include/ck/utility/amd_buffer_addressing.hpp
View File

@@ -1,6 +1,4 @@
#ifndef CK_AMD_BUFFER_ADDRESSING_HPP
#define CK_AMD_BUFFER_ADDRESSING_HPP
#pragma once
#include "data_type.hpp"
namespace ck {
@@ -87,6 +85,7 @@ llvm_amdgcn_raw_buffer_load_i32x4(int32x4_t srsrc,
index_t voffset,
index_t soffset,
index_t glc_slc) __asm("llvm.amdgcn.raw.buffer.load.v4i32");
// buffer load fp16
__device__ half_t
llvm_amdgcn_raw_buffer_load_fp16(int32x4_t srsrc,
@@ -212,6 +211,7 @@ llvm_amdgcn_raw_buffer_store_fp16x4(half4_t vdata,
index_t voffset,
index_t soffset,
index_t glc_slc) __asm("llvm.amdgcn.raw.buffer.store.v4f16");
// buffer store fp32
__device__ void
llvm_amdgcn_raw_buffer_store_fp32(float vdata,
@@ -233,6 +233,7 @@ llvm_amdgcn_raw_buffer_store_fp32x4(float4_t vdata,
index_t voffset,
index_t soffset,
index_t glc_slc) __asm("llvm.amdgcn.raw.buffer.store.v4f32");
// buffer atomic-add fp16
__device__ half2_t llvm_amdgcn_raw_buffer_atomic_add_fp16x2(
half2_t vdata,
@@ -1046,4 +1047,3 @@ amd_buffer_atomic_add(const typename vector_type_maker<T, N>::type::type src_thr
}
} // namespace ck
#endif