Skip to content

GitLab

Commit 50b6b104 authored by Chao Liu's avatar Chao Liu
Browse files

Merge remote-tracking branch 'origin/develop' into groupnorm_check

parents d7bb21c2 e9d4e893
Expand all Hide whitespace changes
Inline Side-by-side
Showing with 160 additions and 31 deletions
Jenkinsfile
View file @ 50b6b104
......@@ -663,8 +663,8 @@ pipeline {
{
agent{ label rocmnode("gfx908 || gfx90a") }
environment{
setup_args = "${params.COMPILER_VERSION == "ck-9110" ? """ -DBUILD_DEV=Off -DCMAKE_INSTALL_PREFIX=../install -D CMAKE_CXX_FLAGS="--offload-arch=gfx908 --offload-arch=gfx90a -O3 -Xclang -mlink-builtin-bitcode -Xclang /opt/rocm/amdgcn/bitcode/oclc_abi_version_400.bc" """ : """ -DBUILD_DEV=Off -DCMAKE_INSTALL_PREFIX=../install -D CMAKE_CXX_FLAGS="--offload-arch=gfx908 --offload-arch=gfx90a -O3 " """ }"
execute_args = "${params.COMPILER_VERSION == "ck-9110" ? """ cd ../client_example && rm -rf build && mkdir build && cd build && cmake -D CMAKE_PREFIX_PATH="${env.WORKSPACE}/install;/opt/rocm" -D CMAKE_CXX_FLAGS=" --offload-arch=gfx908 --offload-arch=gfx90a -O3 -Xclang -mlink-builtin-bitcode -Xclang /opt/rocm/amdgcn/bitcode/oclc_abi_version_400.bc" -D CMAKE_CXX_COMPILER="${build_compiler()}" .. && make -j """ : """ cd ../client_example && rm -rf build && mkdir build && cd build && cmake -D CMAKE_PREFIX_PATH="${env.WORKSPACE}/install;/opt/rocm" -D CMAKE_CXX_FLAGS=" --offload-arch=gfx908 --offload-arch=gfx90a -O3" -D CMAKE_CXX_COMPILER="${build_compiler()}" .. && make -j """ }"
setup_args = "${params.COMPILER_VERSION == "ck-9110" ? """ -DBUILD_DEV=Off -DCMAKE_INSTALL_PREFIX=../install -DGPU_TARGETS="gfx908;gfx90a" -DCMAKE_CXX_FLAGS="-O3 -Xclang -mlink-builtin-bitcode -Xclang /opt/rocm/amdgcn/bitcode/oclc_abi_version_400.bc" """ : """ -DBUILD_DEV=Off -DCMAKE_INSTALL_PREFIX=../install -DGPU_TARGETS="gfx908;gfx90a" -DCMAKE_CXX_FLAGS="-O3 " """ }"
execute_args = "${params.COMPILER_VERSION == "ck-9110" ? """ cd ../client_example && rm -rf build && mkdir build && cd build && cmake -D CMAKE_PREFIX_PATH="${env.WORKSPACE}/install;/opt/rocm" -DGPU_TARGETS="gfx908;gfx90a" -DCMAKE_CXX_FLAGS="-O3 -Xclang -mlink-builtin-bitcode -Xclang /opt/rocm/amdgcn/bitcode/oclc_abi_version_400.bc" -D CMAKE_CXX_COMPILER="${build_compiler()}" .. && make -j """ : """ cd ../client_example && rm -rf build && mkdir build && cd build && cmake -D CMAKE_PREFIX_PATH="${env.WORKSPACE}/install;/opt/rocm" -DGPU_TARGETS="gfx908,gfx90a" -DCMAKE_CXX_FLAGS="-O3" -D CMAKE_CXX_COMPILER="${build_compiler()}" .. && make -j """ }"
}
steps{
Build_CK_and_Reboot(setup_args: setup_args, config_targets: "install", no_reboot:true, build_type: 'Release', execute_cmd: execute_args, prefixpath: '/usr/local')
......@@ -689,8 +689,8 @@ pipeline {
{
agent{ label rocmnode("gfx908")}
environment{
setup_args = "${params.COMPILER_VERSION == "ck-9110" ? """ -DBUILD_DEV=Off -DCMAKE_INSTALL_PREFIX=../install -D CMAKE_CXX_FLAGS="--offload-arch=gfx908 -O3 -Xclang -mlink-builtin-bitcode -Xclang /opt/rocm/amdgcn/bitcode/oclc_abi_version_400.bc" """ : """ -DBUILD_DEV=Off -DCMAKE_INSTALL_PREFIX=../install -D CMAKE_CXX_FLAGS="--offload-arch=gfx908 -O3 " """ }"
execute_args = "${params.COMPILER_VERSION == "ck-9110" ? """ cd ../client_example && rm -rf build && mkdir build && cd build && cmake -D CMAKE_PREFIX_PATH="${env.WORKSPACE}/install;/opt/rocm" -D CMAKE_CXX_FLAGS=" --offload-arch=gfx908 -O3 -Xclang -mlink-builtin-bitcode -Xclang /opt/rocm/amdgcn/bitcode/oclc_abi_version_400.bc" -D CMAKE_CXX_COMPILER="${build_compiler()}" .. && make -j """ : """ cd ../client_example && rm -rf build && mkdir build && cd build && cmake -D CMAKE_PREFIX_PATH="${env.WORKSPACE}/install;/opt/rocm" -D CMAKE_CXX_FLAGS=" --offload-arch=gfx908 -O3" -D CMAKE_CXX_COMPILER="${build_compiler()}" .. && make -j """ }"
setup_args = "${params.COMPILER_VERSION == "ck-9110" ? """ -DBUILD_DEV=Off -DCMAKE_INSTALL_PREFIX=../install -DGPU_TARGETS="gfx908;gfx90a" -DCMAKE_CXX_FLAGS="-O3 -Xclang -mlink-builtin-bitcode -Xclang /opt/rocm/amdgcn/bitcode/oclc_abi_version_400.bc" """ : """ -DBUILD_DEV=Off -DCMAKE_INSTALL_PREFIX=../install -DGPU_TARGETS="gfx908;gfx90a" -DCMAKE_CXX_FLAGS="-O3 " """ }"
execute_args = "${params.COMPILER_VERSION == "ck-9110" ? """ cd ../client_example && rm -rf build && mkdir build && cd build && cmake -D CMAKE_PREFIX_PATH="${env.WORKSPACE}/install;/opt/rocm" -DGPU_TARGETS="gfx908;gfx90a" -DCMAKE_CXX_FLAGS="-O3 -Xclang -mlink-builtin-bitcode -Xclang /opt/rocm/amdgcn/bitcode/oclc_abi_version_400.bc" -D CMAKE_CXX_COMPILER="${build_compiler()}" .. && make -j """ : """ cd ../client_example && rm -rf build && mkdir build && cd build && cmake -D CMAKE_PREFIX_PATH="${env.WORKSPACE}/install;/opt/rocm" -DGPU_TARGETS="gfx908;gfx90a" -DCMAKE_CXX_FLAGS="-O3" -D CMAKE_CXX_COMPILER="${build_compiler()}" .. && make -j """ }"
}
steps{
buildHipClangJobAndReboot(setup_args: setup_args, config_targets: "install", no_reboot:true, build_type: 'Release', execute_cmd: execute_args, prefixpath: '/usr/local')
......@@ -712,7 +712,7 @@ pipeline {
options { retry(2) }
agent{ label rocmnode("gfx908 || gfx90a")}
environment{
setup_args = "${params.COMPILER_VERSION == "ck-9110" ? """ -D CMAKE_CXX_FLAGS=" --offload-arch=gfx908 -O3 -Xclang -mlink-builtin-bitcode -Xclang /opt/rocm/amdgcn/bitcode/oclc_abi_version_400.bc" -DBUILD_DEV=On """ : """ -D CMAKE_CXX_FLAGS=" --offload-arch=gfx908 -O3 " -DBUILD_DEV=On """}"
setup_args = "${params.COMPILER_VERSION == "ck-9110" ? """ -DGPU_TARGETS="gfx908;gfx90a" -DCMAKE_CXX_FLAGS=" -O3 -Xclang -mlink-builtin-bitcode -Xclang /opt/rocm/amdgcn/bitcode/oclc_abi_version_400.bc" -DBUILD_DEV=On """ : """ -DGPU_TARGETS="gfx908;gfx90a" -DCMAKE_CXX_FLAGS=" -O3 " -DBUILD_DEV=On """}"
}
steps{
runPerfTest(setup_args:setup_args, config_targets: "ckProfiler", no_reboot:true, build_type: 'Release')
......@@ -727,7 +727,7 @@ pipeline {
options { retry(2) }
agent{ label rocmnode("gfx90a")}
environment{
setup_args = "${params.COMPILER_VERSION == "ck-9110" ? """ -D CMAKE_CXX_FLAGS=" --offload-arch=gfx90a -O3 -Xclang -mlink-builtin-bitcode -Xclang /opt/rocm/amdgcn/bitcode/oclc_abi_version_400.bc" -DBUILD_DEV=On """ : """ -D CMAKE_CXX_FLAGS=" --offload-arch=gfx90a -O3 " -DBUILD_DEV=On """}"
setup_args = "${params.COMPILER_VERSION == "ck-9110" ? """ -DGPU_TARGETS="gfx90a" -DCMAKE_CXX_FLAGS=" -O3 -Xclang -mlink-builtin-bitcode -Xclang /opt/rocm/amdgcn/bitcode/oclc_abi_version_400.bc" -DBUILD_DEV=On """ : """ -DGPU_TARGETS="gfx90a" -DCMAKE_CXX_FLAGS=" -O3 " -DBUILD_DEV=On """}"
}
steps{
runPerfTest(setup_args:setup_args, config_targets: "ckProfiler", no_reboot:true, build_type: 'Release')
......
client_example/CMakeLists.txt
View file @ 50b6b104
......@@ -9,7 +9,7 @@ message(STATUS "Build with HIP ${hip_VERSION}")
# add all example subdir
file(GLOB dir_list LIST_DIRECTORIES true *)
FOREACH(subdir ${dir_list})
IF(IS_DIRECTORY "${subdir}")
IF(IS_DIRECTORY "${subdir}" AND (NOT "${subdir}" MATCHES "build"))
add_subdirectory(${subdir})
ENDIF()
ENDFOREACH()
include/ck/tensor_operation/gpu/device/device_gemm_multiple_d_xdl_cshuffle.hpp
View file @ 50b6b104
......@@ -332,7 +332,10 @@ struct DeviceGemmMultipleD_Xdl_CShuffle : public DeviceGemmMultipleD<ALayout,
block_2_etile_map_{GridwiseGemm::MakeDefaultBlock2ETileMap(e_grid_desc_m_n_)},
a_element_op_{a_element_op},
b_element_op_{b_element_op},
cde_element_op_{cde_element_op}
cde_element_op_{cde_element_op},
MRaw_{MRaw},
NRaw_{NRaw},
KRaw_{KRaw}
{
// populate pointer, desc for Ds
static_for<0, NumDTensor, 1>{}([&](auto i) {
......@@ -400,6 +403,11 @@ struct DeviceGemmMultipleD_Xdl_CShuffle : public DeviceGemmMultipleD<ALayout,
AElementwiseOperation a_element_op_;
BElementwiseOperation b_element_op_;
CDEElementwiseOperation cde_element_op_;
// for checking vector load/store
index_t MRaw_;
index_t NRaw_;
index_t KRaw_;
};
// Invoker
......@@ -486,6 +494,86 @@ struct DeviceGemmMultipleD_Xdl_CShuffle : public DeviceGemmMultipleD<ALayout,
return false;
}
// check vector load/store
{
using Row = ck::tensor_layout::gemm::RowMajor;
using Col = ck::tensor_layout::gemm::ColumnMajor;
// check vector load of A
if constexpr(is_same_v<ALayout, Row> && ABlockTransferSrcVectorDim == 2)
{
if(arg.KRaw_ % ABlockTransferSrcScalarPerVector != 0)
{
return false;
}
}
else if constexpr(is_same_v<ALayout, Col> && ABlockTransferSrcVectorDim == 1)
{
// FIXME: not rigorous
if(arg.MRaw_ % ABlockTransferSrcScalarPerVector != 0)
{
return false;
}
}
else
{
return false;
}
// check vector laod of B
if constexpr(is_same_v<BLayout, Col> && BBlockTransferSrcVectorDim == 2)
{
if(arg.KRaw_ % BBlockTransferSrcScalarPerVector != 0)
{
return false;
}
}
else if constexpr(is_same_v<BLayout, Row> && BBlockTransferSrcVectorDim == 1)
{
// FIXME: not rigorous
if(arg.NRaw_ % BBlockTransferSrcScalarPerVector != 0)
{
return false;
}
}
else
{
return false;
}
// check vector load of Ds
// only support RowMajor for now
bool all_valid = true;
static_for<0, NumDTensor, 1>{}([&](auto i) {
using DLayout = remove_cvref_t<tuple_element_t<i.value, DsLayout>>;
if constexpr(!is_same_v<DLayout, Row>)
{
all_valid = false;
}
});
if(!all_valid)
{
return false;
}
// check vector store of E
// only support RowMajor for now
if constexpr(is_same_v<ELayout, Row>)
{
if(arg.NRaw_ % CDEBlockTransferScalarPerVector_NPerBlock != 0)
{
return false;
}
}
else
{
return false;
}
}
return GridwiseGemm::CheckValidity(arg.a_grid_desc_m_k_,
arg.b_grid_desc_n_k_,
arg.ds_grid_desc_m_n_,
......
profiler/src/profiler.cpp
View file @ 50b6b104
......@@ -3,27 +3,27 @@
#include <cstring>
// int profile_gemm(int, char*[]);
// int profile_gemm_splitk(int, char*[]);
// int profile_gemm_bilinear(int, char*[]);
// int profile_gemm_add_add_fastgelu(int, char*[]);
// int profile_gemm_reduce(int, char*[]);
// int profile_gemm_bias_add_reduce(int, char*[]);
// int profile_batched_gemm(int, char*[]);
// int profile_batched_gemm_gemm(int, char*[]);
// int profile_batched_gemm_add_relu_gemm_add(int, char*[]);
// int profile_batched_gemm_reduce(int, char*[]);
// int profile_grouped_gemm(int, char*[]);
// int profile_conv_fwd(int, char*[]);
// int profile_conv_fwd_bias_relu(int, char*[]);
// int profile_conv_fwd_bias_relu_add(int, char*[]);
// int profile_conv_bwd_data(int, char*[]);
// int profile_conv_bwd_weight(int, char*[]);
// int profile_grouped_conv_fwd(int, char*[]);
// int profile_normalization(int, char*[]);
int profile_gemm(int, char*[]);
int profile_gemm_splitk(int, char*[]);
int profile_gemm_bilinear(int, char*[]);
int profile_gemm_add_add_fastgelu(int, char*[]);
int profile_gemm_reduce(int, char*[]);
int profile_gemm_bias_add_reduce(int, char*[]);
int profile_batched_gemm(int, char*[]);
int profile_batched_gemm_gemm(int, char*[]);
int profile_batched_gemm_add_relu_gemm_add(int, char*[]);
int profile_batched_gemm_reduce(int, char*[]);
int profile_grouped_gemm(int, char*[]);
int profile_conv_fwd(int, char*[]);
int profile_conv_fwd_bias_relu(int, char*[]);
int profile_conv_fwd_bias_relu_add(int, char*[]);
int profile_conv_bwd_data(int, char*[]);
int profile_conv_bwd_weight(int, char*[]);
int profile_grouped_conv_fwd(int, char*[]);
int profile_normalization(int, char*[]);
int profile_layernorm(int, char*[]);
int profile_groupnorm(int, char*[]);
// int profile_reduce(int, char*[]);
int profile_reduce(int, char*[]);
static void print_helper_message()
{
......@@ -57,7 +57,6 @@ int main(int argc, char* argv[])
return 0;
}
#if 0
else if(strcmp(argv[1], "gemm") == 0)
{
return profile_gemm(argc, argv);
......@@ -134,7 +133,6 @@ int main(int argc, char* argv[])
{
return profile_normalization(argc, argv);
}
#endif
else if(strcmp(argv[1], "layernorm") == 0)
{
return profile_layernorm(argc, argv);
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment