p83571920
/
mindspore
forked from mindspore-Ecosystem/mindspore
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

!48518 upgrade_ascend_20270207_master 

Merge pull request !48518 from 胡彬/upgrade_ascend_20230207_master
This commit is contained in:
i-robot 2023-02-13 10:07:39 +00:00 committed by Gitee
parent 2a40438f81 2151611a32
commit 486cb563a0
No known key found for this signature in database
GPG Key ID: 173E9B9CA92EEF8F
67 changed files with 163 additions and 129 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
CMakeLists.txt
View File

@ -110,17 +110,17 @@ include_directories(${PYTHON_INCLUDE_DIRS})
set(MS_CCSRC_PATH ${CMAKE_SOURCE_DIR}/mindspore/ccsrc)
set(MS_CCSRC_BUILD_PATH ${BUILD_PATH}/mindspore/mindspore/ccsrc)
if(ENABLE_D OR ENABLE_ACL OR ENABLE_TESTCASES)
include(${CMAKE_SOURCE_DIR}/cmake/dependency_graphengine.cmake)
endif()
if(NOT MSVC)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fvisibility=hidden")
endif()
include(${CMAKE_SOURCE_DIR}/cmake/init.cmake)
add_subdirectory_with_faster_option(mindspore/ccsrc)
add_subdirectory(mindspore/core)
if(ENABLE_D OR ENABLE_ACL OR ENABLE_TESTCASES)
include(${CMAKE_SOURCE_DIR}/cmake/dependency_graphengine.cmake)
endif()
add_subdirectory_with_faster_option(mindspore/ccsrc)
if(ENABLE_TESTCASES OR ENABLE_CPP_ST)
add_subdirectory(tests)

10
cmake/ascend_variables.cmake
View File

@ -9,14 +9,16 @@ set(ASCEND_DRIVER_HAL_PATH ${ASCEND_PATH}/driver/lib64/driver)
# CANN packages
set(ASCEND_CANN_RUNTIME_PATH ${ASCEND_PATH}/latest/lib64)
set(ASCEND_CANN_OPP_PATH ${ASCEND_PATH}/latest/opp/op_impl/built-in/ai_core/tbe/op_tiling)
set(ASCEND_CANN_OPP_PATH_TEMP ${ASCEND_PATH}/latest/opp/built-in/op_impl/ai_core/tbe/op_tiling)
set(ASCEND_CANN_OPP_PATH ${ASCEND_PATH}/latest/opp/built-in/op_impl/ai_core/tbe/op_tiling/lib/linux)
set(ASCEND_CANN_OPP_AARCH64_PATH ${ASCEND_CANN_OPP_PATH}/aarch64)
set(ASCEND_CANN_OPP_X86_64_PATH ${ASCEND_CANN_OPP_PATH}/x86_64)
set(ASCEND_CANN_PLUGIN_PATH ${ASCEND_CANN_RUNTIME_PATH}/plugin/opskernel)
# Ascend-toolkit packages
set(ASCEND_TOOLKIT_RUNTIME_PATH ${ASCEND_PATH}/ascend-toolkit/latest/lib64)
set(ASCEND_TOOLKIT_OPP_PATH ${ASCEND_PATH}/ascend-toolkit/latest/opp/op_impl/built-in/ai_core/tbe/op_tiling)
set(ASCEND_TOOLKIT_OPP_PATH_TEMP ${ASCEND_PATH}/ascend-toolkit/latest/opp/built-in/op_impl/ai_core/tbe/op_tiling)
set(ASCEND_TOOLKIT_OPP_PATH ${ASCEND_PATH}/ascend-toolkit/latest/opp/built-in/op_impl/ai_core/tbe/op_tiling/lib/linux)
set(ASCEND_TOOLKIT_OPP_AARCH64_PATH ${ASCEND_TOOLKIT_OPP_PATH}/aarch64)
set(ASCEND_TOOLKIT_OPP_X86_64_PATH ${ASCEND_TOOLKIT_OPP_PATH}/x86_64)
set(ASCEND_TOOLKIT_PLUGIN_PATH ${ASCEND_TOOLKIT_RUNTIME_PATH}/plugin/opskernel)
# nnae packages (for rpath only)

14
cmake/graphengine_variables.cmake Normal file
View File

@ -0,0 +1,14 @@
# path variables for graphengine submodule, it has to be included after mindspore/core
# and minspore/ccsrc to prevent conflict of op headers
if(ENABLE_D OR ENABLE_ACL OR ENABLE_TESTCASES)
include_directories(${CMAKE_SOURCE_DIR}/graphengine/inc)
include_directories(${CMAKE_SOURCE_DIR}/graphengine/inc/external)
include_directories(${CMAKE_SOURCE_DIR}/graphengine/inc/framework)
include_directories(${CMAKE_SOURCE_DIR}/graphengine/base)
include_directories(${CMAKE_SOURCE_DIR}/graphengine/third_party/fwkacllib/inc)
include_directories(${CMAKE_SOURCE_DIR}/graphengine/third_party/fwkacllib/inc/aicpu)
include_directories(${CMAKE_SOURCE_DIR}/graphengine/third_party/fwkacllib/inc/toolchain)
include_directories(${CMAKE_SOURCE_DIR}/graphengine/metadef/inc)
include_directories(${CMAKE_SOURCE_DIR}/graphengine/metadef/inc/external)
include_directories(${CMAKE_SOURCE_DIR}/graphengine/metadef/inc/external/graph)
endif()

12
cmake/mind_expression.cmake
View File

@ -64,18 +64,6 @@ if(ENABLE_GPU AND GPU_BACKEND_CUDA)
endif()
endif()
if(ENABLE_D OR ENABLE_ACL OR ENABLE_TESTCASES)
include_directories(${CMAKE_SOURCE_DIR}/graphengine/inc)
include_directories(${CMAKE_SOURCE_DIR}/graphengine/inc/external)
include_directories(${CMAKE_SOURCE_DIR}/graphengine/inc/framework)
include_directories(${CMAKE_SOURCE_DIR}/graphengine/third_party/fwkacllib/inc)
include_directories(${CMAKE_SOURCE_DIR}/graphengine/third_party/fwkacllib/inc/aicpu)
include_directories(${CMAKE_SOURCE_DIR}/graphengine/third_party/fwkacllib/inc/toolchain)
include_directories(${CMAKE_SOURCE_DIR}/graphengine/metadef/inc)
include_directories(${CMAKE_SOURCE_DIR}/graphengine/metadef/inc/external)
include_directories(${CMAKE_SOURCE_DIR}/graphengine/metadef/inc/external/graph)
endif()
if(ENABLE_MINDDATA)
include(${CMAKE_SOURCE_DIR}/cmake/external_libs/icu4c.cmake)
include(${CMAKE_SOURCE_DIR}/cmake/external_libs/opencv.cmake)

6
config/super_bar_config.json
View File

@ -415,7 +415,8 @@
"TransData ": "support boll",
"ScatterNdD ": "Accuracy issues",
"Trace": "Hadn't adapted tbe implementation",
"AssignAdd": "Frac_nz in pangu not support"
"AssignAdd": "Frac_nz in pangu not support",
"Range": "not support dynamic shape with tiling failed"
},
"SkipNodes": [
"BroadcastTo",
@ -444,7 +445,8 @@
"ACos",
"TransData",
"ScatterNdD",
"AssignAdd"
"AssignAdd",
"Range"
],
"FallbackOps": {
"DeformableOffsets": [

2
graphengine

@ -1 +1 @@
Subproject commit 236001806129e36c0f48b240c4f61b2e1d92c470
Subproject commit f5f74bb7a124f99bbaeaec17b55aa466fdd34285

4
mindspore/ccsrc/CMakeLists.txt
View File

@ -5,6 +5,10 @@ include_directories(${CMAKE_BINARY_DIR})
include_directories(${CMAKE_SOURCE_DIR}/mindspore/core/mindrt/include)
include_directories(${CMAKE_SOURCE_DIR}/mindspore/core/mindrt/src)
# graphengine include directories
if(ENABLE_D OR ENABLE_ACL OR ENABLE_TESTCASES)
include(${CMAKE_SOURCE_DIR}/cmake/graphengine_variables.cmake)
endif()
set(SERVER_FLATBUFFER_OUTPUT "${CMAKE_BINARY_DIR}/schema")

6
mindspore/ccsrc/cxx_api/model/acl/model_converter.cc
View File

@ -19,6 +19,7 @@
#include "include/transform/graph_ir/utils.h"
#include "cxx_api/model/model_converter_utils/multi_process.h"
#include "graph/model.h"
#include "graph/utils/graph_utils_ex.h"
#include "acl/acl_rt.h"
#include "cxx_api/model/aoe/auto_tune_process.h"
#include "plugin/device/ascend/optimizer/ge_optimization.h"
@ -343,7 +344,7 @@ Buffer ModelConverter::LoadMindIR(const FuncGraphPtr &func_graph) {
}
ge::Model model;
ge::Buffer model_data;
model.SetGraph(*df_graph);
model.SetGraph(::ge::GraphUtilsEx::GetComputeGraph(*df_graph));
auto ge_ret = model.Save(model_data);
if (ge_ret != ge::SUCCESS) {
MS_LOG(ERROR) << "Save ge model to buffer failed.";
@ -413,7 +414,8 @@ Buffer ModelConverter::LoadAscendIRInner(const Buffer &model_data) {
return Buffer();
}
transform::DfGraphPtr df_graph = std::make_shared<transform::DfGraph>(load_model.GetGraph());
transform::DfGraphPtr df_graph =
std::make_shared<transform::DfGraph>(::ge::GraphUtilsEx::CreateGraphFromComputeGraph(load_model.GetGraph()));
if (df_graph == nullptr) {
MS_LOG(ERROR) << "Convert FuncGraph to AscendIR failed.";
return Buffer();

28
mindspore/ccsrc/plugin/device/ascend/CMakeLists.txt
View File

@ -6,11 +6,22 @@ include_directories(${CMAKE_SOURCE_DIR}/mindspore/ccsrc/minddata/dataset)
set(ASCEND_RPATH ${ASCEND_RPATH}:/usr/local/Ascend/nnae/latest/lib64)
set(ASCEND_RPATH ${ASCEND_RPATH}:/usr/local/Ascend/ascend-toolkit/latest/lib64)
set(ASCEND_RPATH ${ASCEND_RPATH}:/usr/local/Ascend/latest/lib64)
set(ASCEND_RPATH ${ASCEND_RPATH}:/usr/local/Ascend/opp/built-in/op_impl/ai_core/tbe/op_tiling)
set(ASCEND_RPATH ${ASCEND_RPATH}:/usr/local/Ascend/nnae/latest/opp/built-in/op_impl/ai_core/tbe/op_tiling)
set(ASCEND_RPATH
${ASCEND_RPATH}:/usr/local/Ascend/ascend-toolkit/latest/opp/built-in/op_impl/ai_core/tbe/op_tiling)
set(ASCEND_RPATH ${ASCEND_RPATH}:/usr/local/Ascend/latest/opp/built-in/op_impl/ai_core/tbe/op_tiling)
if(CMAKE_HOST_SYSTEM_PROCESSOR MATCHES "x86_64")
set(ASCEND_RPATH
${ASCEND_RPATH}:/usr/local/Ascend/nnae/latest/opp/built-in/op_impl/ai_core/tbe/op_tiling/lib/linux/x86_64)
set(ASCEND_RPATH
${ASCEND_RPATH}:/usr/local/Ascend/ascend-toolkit/latest/opp/built-in/op_impl/ai_core/tbe/op_tiling/lib/linux/x86_64)
set(ASCEND_RPATH
${ASCEND_RPATH}:/usr/local/Ascend/latest/opp/built-in/op_impl/ai_core/tbe/op_tiling/lib/linux/x86_64)
else()
set(ASCEND_RPATH
${ASCEND_RPATH}:/usr/local/Ascend/nnae/latest/opp/built-in/op_impl/ai_core/tbe/op_tiling/lib/linux/aarch64)
set(ASCEND_RPATH
${ASCEND_RPATH}:/usr/local/Ascend/ascend-toolkit/latest/opp/built-in/op_impl/ai_core/tbe/op_tiling/lib/linux/aarch64)
set(ASCEND_RPATH
${ASCEND_RPATH}:/usr/local/Ascend/latest/opp/built-in/op_impl/ai_core/tbe/op_tiling/lib/linux/aarch64)
endif()
### cxx api need file ###
if(ENABLE_ACL)
@ -87,8 +98,11 @@ if(MODE_ASCEND_ALL)
find_library(PLATFORM platform ${ASCEND_CANN_RUNTIME_PATH} ${ASCEND_TOOLKIT_RUNTIME_PATH})
find_library(OPT_FEATURE opt_feature ${ASCEND_CANN_RUNTIME_PATH} ${ASCEND_TOOLKIT_RUNTIME_PATH})
find_library(adump_server libadump_server.a ${ASCEND_CANN_RUNTIME_PATH} ${ASCEND_TOOLKIT_RUNTIME_PATH})
find_library(OPTILING optiling ${ASCEND_CANN_OPP_PATH} ${ASCEND_TOOLKIT_OPP_PATH}
${ASCEND_CANN_OPP_PATH_TEMP} ${ASCEND_TOOLKIT_OPP_PATH_TEMP})
if(CMAKE_HOST_SYSTEM_PROCESSOR MATCHES "x86_64")
find_library(OPTILING optiling ${ASCEND_CANN_OPP_X86_64_PATH} ${ASCEND_TOOLKIT_OPP_X86_64_PATH})
else()
find_library(OPTILING optiling ${ASCEND_CANN_OPP_AARCH64_PATH} ${ASCEND_TOOLKIT_OPP_AARCH64_PATH})
endif()
find_library(ACL_OP_COMPILER acl_op_compiler ${ASCEND_CANN_RUNTIME_PATH} ${ASCEND_TOOLKIT_RUNTIME_PATH})
target_link_libraries(mindspore_ascend PRIVATE ${RUNTIME_LIB} ${TSDCLIENT} ${DATATRANSFER} ${ERROR_MANAGER}

2
mindspore/ccsrc/plugin/device/ascend/hal/device/ascend_memory_adapter.cc
View File

@ -254,7 +254,7 @@ size_t AscendMemAdapter::GetDeviceMemSizeFromContext() const {
uint8_t *AscendMemAdapter::MallocFromRts(size_t size) const {
uint8_t *ptr = nullptr;
auto ret = rtMalloc(reinterpret_cast<void **>(&ptr), size, RT_MEMORY_HBM);
auto ret = rtMalloc(reinterpret_cast<void **>(&ptr), size, RT_MEMORY_HBM, 0);
if (ret != ACL_RT_SUCCESS) {
if (ret == ACL_ERROR_RT_MEMORY_ALLOCATION) {
auto context_ptr = MsContext::GetInstance();

4
mindspore/ccsrc/plugin/device/ascend/hal/device/dump/data_dumper.cc
View File

@ -277,12 +277,12 @@ void DataDumper::OpDebugRegister() {
MS_LOG(EXCEPTION) << "[DataDump] Call rt api rtGetRtCapability failed, ret = " << rt_ret;
}
auto memory_type = (value == static_cast<int64_t>(RT_CAPABILITY_SUPPORT)) ? RT_MEMORY_TS : RT_MEMORY_HBM;
rt_ret = rtMalloc(&op_debug_buffer_addr_, kOpDebugHostMemSize, memory_type);
rt_ret = rtMalloc(&op_debug_buffer_addr_, kOpDebugHostMemSize, memory_type, 0);
if (rt_ret != RT_ERROR_NONE) {
MS_LOG(EXCEPTION) << "[DataDump] Call rt api rtMalloc failed, ret = " << rt_ret;
}
rt_ret = rtMalloc(&op_debug_dump_args_, kOpDebugDevMemSize, RT_MEMORY_HBM);
rt_ret = rtMalloc(&op_debug_dump_args_, kOpDebugDevMemSize, RT_MEMORY_HBM, 0);
if (rt_ret != RT_ERROR_NONE) {
MS_LOG(EXCEPTION) << "[DataDump] Call rtMalloc failed, ret = " << rt_ret;
}

2
mindspore/ccsrc/plugin/device/ascend/hal/device/dump/dumper_base.cc
View File

@ -101,7 +101,7 @@ void RtLoadDumpData(const aicpu::dump::OpMappingInfo &dump_info, void **ptr) {
return;
}
rtError_t rt_ret = rtMalloc(ptr, proto_size, RT_MEMORY_HBM);
rtError_t rt_ret = rtMalloc(ptr, proto_size, RT_MEMORY_HBM, 0);
if (rt_ret != RT_ERROR_NONE) {
MS_LOG(EXCEPTION) << "[DumperBase] Call rtMalloc failed";
}

8
mindspore/ccsrc/plugin/device/ascend/hal/device/dump/kernel_dumper.cc
View File

@ -184,7 +184,7 @@ void KernelDumper::ExecutorDumpOp(const aicpu::dump::OpMappingInfo &op_mapping_i
}
std::string proto_json;
(void)google::protobuf::util::MessageToJsonString(op_mapping_info, &proto_json);
rtError_t rt_ret = rtMalloc(&proto_dev_mem_, proto_size, RT_MEMORY_HBM);
rtError_t rt_ret = rtMalloc(&proto_dev_mem_, proto_size, RT_MEMORY_HBM, 0);
if (rt_ret != RT_ERROR_NONE) {
MS_LOG(ERROR) << "[KernelDumper] Call rt api rtMalloc failed, ret = " << rt_ret;
return;
@ -196,7 +196,7 @@ void KernelDumper::ExecutorDumpOp(const aicpu::dump::OpMappingInfo &op_mapping_i
return;
}
rt_ret = rtMalloc(&proto_size_dev_mem_, sizeof(size_t), RT_MEMORY_HBM);
rt_ret = rtMalloc(&proto_size_dev_mem_, sizeof(size_t), RT_MEMORY_HBM, 0);
if (rt_ret != RT_ERROR_NONE) {
MS_LOG(ERROR) << "[KernelDumper] Call rt api rtMalloc failed, ret = " << rt_ret;
return;
@ -372,7 +372,7 @@ void KernelDumper::MallocP2PDebugMem(const void *const op_debug_addr) {
MS_LOG(EXCEPTION) << "[KernelDumper] Call rt api rtGetRtCapability failed, ret = " << rt_ret;
}
auto memory_type = (value == static_cast<int64_t>(RT_CAPABILITY_SUPPORT)) ? RT_MEMORY_TS : RT_MEMORY_HBM;
rtMalloc(&p2p_debug_addr_, kDebugP2pSize, memory_type);
rtMalloc(&p2p_debug_addr_, kDebugP2pSize, memory_type, 0);
rtMemcpy(p2p_debug_addr_, sizeof(uint64_t), &debug_addrs_tmp, sizeof(uint64_t), RT_MEMCPY_HOST_TO_DEVICE);
}
@ -416,7 +416,7 @@ void KernelDumper::OpDebugRegisterForStream(const CNodePtr &kernel) {
MS_LOG(EXCEPTION) << "[KernelDumper] Call rt api rtGetRtCapability failed, ret = " << rt_ret;
}
auto memory_type = (value == static_cast<int64_t>(RT_CAPABILITY_SUPPORT)) ? RT_MEMORY_TS : RT_MEMORY_HBM;
rt_ret = rtMalloc(&op_debug_task->op_debug_addr, kOpDebugMemorySize, memory_type);
rt_ret = rtMalloc(&op_debug_task->op_debug_addr, kOpDebugMemorySize, memory_type, 0);
if (rt_ret != RT_ERROR_NONE) {
MS_LOG(EXCEPTION) << "[KernelDumper] Call rt api rtMalloc failed, ret = " << rt_ret;
}

4
mindspore/ccsrc/plugin/device/ascend/hal/device/ge_runtime/task/aicpu_task.cc
View File

@ -91,7 +91,7 @@ void AicpuTask::Distribute() {
sizeof(uint32_t);
// Malloc device memory for args
rtError_t rt_ret = rtMalloc(&args_, args_size_, RT_MEMORY_HBM);
rtError_t rt_ret = rtMalloc(&args_, args_size_, RT_MEMORY_HBM, 0);
if (rt_ret != RT_ERROR_NONE) {
MS_LOG(EXCEPTION) << "Call rt api rtMalloc failed, ret: " << rt_ret;
}
@ -165,7 +165,7 @@ void AicpuTask::SetAicpuParamHead(uint32_t args_size, uint32_t io_addrs_num) {
}
}
// alloc extinfo address
rtError_t flag = rtMalloc(&ext_info_addr_, ext_info_handler->GetExtInfoLen(), RT_MEMORY_HBM);
rtError_t flag = rtMalloc(&ext_info_addr_, ext_info_handler->GetExtInfoLen(), RT_MEMORY_HBM, 0);
if (flag != RT_ERROR_NONE) {
MS_LOG(EXCEPTION) << "Call rt api rtMalloc failed, ret: " << flag;
}

2
mindspore/ccsrc/plugin/device/ascend/hal/device/ge_runtime/task/label_goto_task.cc
View File

@ -61,7 +61,7 @@ void LabelGotoTask::Distribute() {
MS_EXCEPTION_IF_NULL(label_info_);
if (index_value_ == nullptr) {
rtError_t rt_ret = rtMalloc(&index_value_, sizeof(uint64_t), RT_MEMORY_HBM);
rtError_t rt_ret = rtMalloc(&index_value_, sizeof(uint64_t), RT_MEMORY_HBM, 0);
if (rt_ret != RT_ERROR_NONE) {
MS_LOG(EXCEPTION) << "Call rt api rtMalloc failed, ret: " << rt_ret;
}

2
mindspore/ccsrc/plugin/device/ascend/hal/device/ge_runtime/task/label_manager.cc
View File

@ -106,7 +106,7 @@ std::shared_ptr<LabelGuard> LabelManager::GetLabelInfo(rtModel_t model, const st
return nullptr;
}
rt_ret = rtMalloc(&label_info, label_info_size, (value == RT_CAPABILITY_SUPPORT) ? RT_MEMORY_TS : RT_MEMORY_HBM);
rt_ret = rtMalloc(&label_info, label_info_size, (value == RT_CAPABILITY_SUPPORT) ? RT_MEMORY_TS : RT_MEMORY_HBM, 0);
if (rt_ret != RT_ERROR_NONE) {
MS_LOG(ERROR) << "Call rt api rtMalloc failed, ret: " << rt_ret;
return nullptr;

2
mindspore/ccsrc/plugin/device/ascend/hal/device/ge_runtime/task/tbe_task.cc
View File

@ -78,7 +78,7 @@ void TbeTask::Distribute() {
task_info_->workspace_addrs().cend());
args_size_ = static_cast<uint32_t>(tensor_device_addrs.size() * sizeof(void *));
rt_ret = rtMalloc(&args_, args_size_, RT_MEMORY_HBM);
rt_ret = rtMalloc(&args_, args_size_, RT_MEMORY_HBM, 0);
if (rt_ret != RT_ERROR_NONE) {
MS_LOG(EXCEPTION) << "Call rt api rtMalloc failed, ret: " << rt_ret << " mem size " << args_size_;
}

3
mindspore/ccsrc/plugin/device/ascend/hal/hardware/ascend_deprecated_interface.cc
View File

@ -31,6 +31,7 @@
#include "plugin/device/ascend/hal/profiler/parallel_strategy_profiling.h"
#include "plugin/device/ascend/optimizer/enhancer/add_placeholder_for_dynamic_rnn.h"
#include "cxx_api/graph/acl/acl_env_guard.h"
#include "graph/utils/graph_utils_ex.h"
using mindspore::abstract::AbstractScalar;
using mindspore::abstract::AbstractTensor;
@ -166,7 +167,7 @@ void AscendDeprecatedInterface::ExportDFGraph(const std::string &file_name, cons
}
// get model stream
::ge::Model model("", "");
model.SetGraph(*ge_graph);
model.SetGraph(::ge::GraphUtilsEx::GetComputeGraph(*ge_graph));
::ge::Buffer model_data;
auto ge_ret = model.Save(model_data);
if (ge_ret != ::ge::SUCCESS) {

6
mindspore/ccsrc/plugin/device/ascend/kernel/aicpu/aicpu_kernel_load.cc
View File

@ -224,14 +224,14 @@ bool AicpuOpKernelLoad::CacheBinaryFileToDevice(const uintptr_t &resource_id, st
void *d_aicpu_data = nullptr;
void *d_so_name = nullptr;
status = rtMalloc(&d_aicpu_data, aicpu_data_length, RT_MEMORY_HBM);
status = rtMalloc(&d_aicpu_data, aicpu_data_length, RT_MEMORY_HBM, 0);
if (status != RT_ERROR_NONE) {
MS_LOG(ERROR) << "Call rtMalloc failed, size:" << aicpu_data_length << ", ret = 0x" << status;
return false;
}
allocated_mem->emplace_back(d_aicpu_data);
status = rtMalloc(&d_so_name, so_name.size(), RT_MEMORY_HBM);
status = rtMalloc(&d_so_name, so_name.size(), RT_MEMORY_HBM, 0);
if (status != RT_ERROR_NONE) {
MS_LOG(ERROR) << "Call rtMalloc failed, size:" << so_name.size() << ", ret = 0x" << status;
return false;
@ -267,7 +267,7 @@ bool AicpuOpKernelLoad::CacheBinaryFileToDevice(const uintptr_t &resource_id, st
void *args = nullptr;
uint32_t args_size = sizeof(CustAicpuSoBuf) * v_cust_so.size();
status = rtMalloc(&args, args_size, RT_MEMORY_HBM);
status = rtMalloc(&args, args_size, RT_MEMORY_HBM, 0);
if (status != RT_ERROR_NONE) {
MS_LOG(ERROR) << "Call rtMalloc failed, size:" << args_size << ", ret = 0x" << status;
return false;

2
mindspore/ccsrc/plugin/device/ascend/kernel/tbe/tbe_convert_utils.cc
View File

@ -78,7 +78,7 @@ const std::map<TypeId, std::string> type_id_str_maps = {
const std::unordered_map<std::string, size_t> type_nbyte_maps = {
{"float16", sizeof(float) / 2}, {"float32", sizeof(float)},
{"float64", sizeof(float) * 2}, {"int8", sizeof(int) / 4},
{"uint1", sizeof(int) / 4}, {"int16", sizeof(int) / 2},
{"int1", sizeof(int) / 8}, {"int16", sizeof(int) / 2},
{"int32", sizeof(int)}, {"int64", sizeof(int) * 2},
{"uint8", sizeof(int) / 4}, {"uint16", sizeof(int) / 2},
{"uint32", sizeof(int)}, {"uint64", sizeof(int) * 2},

3
mindspore/ccsrc/utils/python_fallback_running.cc
View File

@ -14,6 +14,9 @@
* limitations under the License.
*/
#include "include/common/utils/python_fallback_running.h"
#include "ops/tuple_get_item.h"
#include "ops/primitive_c.h"
#include "mindapi/src/helper.h"
namespace mindspore {
ScopedFallbackRunning::ScopedFallbackRunning() { on_ = true; }

1
mindspore/lite/tools/converter/adapter/acl/CMakeLists.txt
View File

@ -2,6 +2,7 @@ include_directories(${TOP_DIR}/graphengine/metadef/inc/external)
include_directories(${TOP_DIR}/graphengine/inc)
include_directories(${TOP_DIR}/graphengine/inc/external)
include_directories(${TOP_DIR}/graphengine/ge)
include_directories(${TOP_DIR}/graphengine/base)
include_directories(${TOP_DIR}/graphengine/metadef/inc)
include_directories(${TOP_DIR}/graphengine/inc/framework)
include_directories(${TOP_DIR}/graphengine/third_party/fwkacllib/inc)

2
mindspore/python/mindspore/ops/_op_impl/_custom_op/batch_matmul_impl.py
View File

@ -16,7 +16,7 @@
from __future__ import absolute_import
from te import tik
from topi.cce import util
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType

4
mindspore/python/mindspore/ops/_op_impl/_custom_op/batchnorm_fold.py
View File

@ -18,8 +18,8 @@ from __future__ import absolute_import
import te
from te import tvm
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
batch_norm_op_info = TBERegOp("BatchNormFoldD") \

4
mindspore/python/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2.py
View File

@ -17,8 +17,8 @@
import te.lang.cce
from te import tvm
from te.platform.fusion_manager import fusion_manager
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
SHAPE_SIZE_LIMIT = 2147483648

4
mindspore/python/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad.py
View File

@ -17,8 +17,8 @@
import te.lang.cce
from te import tvm
from te.platform.fusion_manager import fusion_manager
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
SHAPE_SIZE_LIMIT = 2147483648

8
mindspore/python/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad_reduce.py
View File

@ -17,9 +17,9 @@
import te.lang.cce
from te import tvm
from te.platform.fusion_manager import fusion_manager
from te.platform.cce_build import build_config
from topi import generic
from topi.cce import util
from tbe.common.buildcfg import build_config
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
SHAPE_SIZE_LIMIT = 2147483648
@ -101,5 +101,5 @@ def batchnorm_fold2_grad_reduce(dout, x, dout_reduce, dout_x_reduce, kernel_name
return
from impl.bn_training_reduce import bn_training_reduce_schedule_nd
sch, tensor_list = bn_training_reduce_schedule_nd(res_list)
with build_config:
with build_config():
tvm.build(sch, tensor_list, "cce", name=kernel_name)

4
mindspore/python/mindspore/ops/_op_impl/_custom_op/batchnorm_fold_grad.py
View File

@ -18,8 +18,8 @@
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
import te.lang.cce
from te import tvm
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
batch_norm_op_info = TBERegOp("BatchNormFoldGradD") \
.fusion_type("OPAQUE") \

2
mindspore/python/mindspore/ops/_op_impl/_custom_op/cholesky_trsm_impl.py
View File

@ -17,7 +17,7 @@ from __future__ import absolute_import
import logging
from te import tik
from topi.cce import util
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
logging.basicConfig(level=logging.DEBUG, format="%(asctime)s %(levelname)s %(message)s")

4
mindspore/python/mindspore/ops/_op_impl/_custom_op/correction_mul.py
View File

@ -17,8 +17,8 @@
import te.lang.cce
from te import tvm
from te.platform.fusion_manager import fusion_manager
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
SHAPE_SIZE_LIMIT = 2147483648

4
mindspore/python/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py
View File

@ -17,8 +17,8 @@
import te.lang.cce
from te import tvm
from te.platform.fusion_manager import fusion_manager
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
SHAPE_SIZE_LIMIT = 2147483648

2
mindspore/python/mindspore/ops/_op_impl/_custom_op/dsd_back_impl.py
View File

@ -15,7 +15,7 @@
"""dsd back impl"""
from __future__ import absolute_import
from te import tik
from topi.cce import util
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import DataType, TBERegOp, op_info_register
dsd_grad_info = TBERegOp('DSDGrad') \

2
mindspore/python/mindspore/ops/_op_impl/_custom_op/dsd_impl.py
View File

@ -15,7 +15,7 @@
""" dense sparse to densne matmul"""
from __future__ import absolute_import
from te import tik
from topi.cce import util
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import DataType, TBERegOp, op_info_register
dsd_matmul_info = TBERegOp('DSDMatmul') \

4
mindspore/python/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel.py
View File

@ -17,8 +17,8 @@
import te.lang.cce
from te import tvm
from te.platform.fusion_manager import fusion_manager
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
fake_learned_scale_quant_perchannel_op_info = TBERegOp("FakeLearnedScaleQuantPerChannel") \

4
mindspore/python/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad.py
View File

@ -17,8 +17,8 @@
import te.lang.cce
from te import tvm
from te.platform.fusion_manager import fusion_manager
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
NEG_SCALAR_MIN_FP16 = -(2 ** (-24))

4
mindspore/python/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad_reduce.py
View File

@ -17,8 +17,8 @@
import te.lang.cce
from te import tvm
from te.platform.fusion_manager import fusion_manager
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType

4
mindspore/python/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer.py
View File

@ -18,8 +18,8 @@ from functools import reduce as functools_reduce
import te.lang.cce
from te import tvm
from te.platform.fusion_manager import fusion_manager
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
fake_learned_scale_quant_perlayer_op_info = TBERegOp("FakeLearnedScaleQuantPerLayer") \

4
mindspore/python/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad.py
View File

@ -19,8 +19,8 @@ from functools import reduce as functools_reduce
import te.lang.cce
from te import tvm
from te.platform.fusion_manager import fusion_manager
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
NEG_SCALAR_MIN_FP16 = -(2 ** (-24))

4
mindspore/python/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad_reduce.py
View File

@ -19,8 +19,8 @@ from functools import reduce as functools_reduce
import te.lang.cce
from te import tvm
from te.platform.fusion_manager import fusion_manager
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType

4
mindspore/python/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel.py
View File

@ -19,8 +19,8 @@ from __future__ import absolute_import
import te.lang.cce
from te import tvm
from te.platform.fusion_manager import fusion_manager
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
fake_quant_perchannel_op_info = TBERegOp("FakeQuantPerChannel") \

4
mindspore/python/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel_grad.py
View File

@ -19,8 +19,8 @@ from __future__ import absolute_import
import te.lang.cce
from te import tvm
from te.platform.fusion_manager import fusion_manager
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
SHAPE_SIZE_LIMIT = 2147483648

4
mindspore/python/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer.py
View File

@ -20,8 +20,8 @@ from functools import reduce as functools_reduce
import te.lang.cce
from te import tvm
from te.platform.fusion_manager import fusion_manager
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
fake_quant_per_layer_op_info = TBERegOp("FakeQuantPerLayer") \

4
mindspore/python/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer_grad.py
View File

@ -20,8 +20,8 @@ from functools import reduce as functools_reduce
import te.lang.cce
from te import tvm
from te.platform.fusion_manager import fusion_manager
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
SHAPE_SIZE_LIMIT = 2147483648

2
mindspore/python/mindspore/ops/_op_impl/_custom_op/fused_abs_max1_impl.py
View File

@ -16,7 +16,7 @@
from __future__ import absolute_import
from te import tik
from topi.cce import util
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType

2
mindspore/python/mindspore/ops/_op_impl/_custom_op/img2col_impl.py
View File

@ -16,7 +16,7 @@
from __future__ import absolute_import
from te import tik
from topi.cce import util
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
cus_img2col_info = TBERegOp("CusImg2Col") \

4
mindspore/python/mindspore/ops/_op_impl/_custom_op/matmul_cube_dense_left_impl.py
View File

@ -21,8 +21,8 @@ import te.lang.cce
import te.platform.cce_params as cce
from te import tik
from te import tvm
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
from mindspore.ops._op_impl._custom_op._basic import _shape_check, _get_bias, _get_input_shape

2
mindspore/python/mindspore/ops/_op_impl/_custom_op/matmul_cube_dense_right_impl.py
View File

@ -20,7 +20,7 @@ matmul
from __future__ import absolute_import
from te import tik
from topi.cce import util
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
matmul_cube_dense_right_op_info = TBERegOp("CusMatMulCubeDenseRight") \

2
mindspore/python/mindspore/ops/_op_impl/_custom_op/matmul_cube_fracz_left_cast_impl.py
View File

@ -22,7 +22,7 @@ import collections
import te.platform.cce_params as cce
from te import tik
from topi.cce import util
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
from mindspore.ops._op_impl._custom_op._basic import _shape_check, _get_bias, _get_input_shape

2
mindspore/python/mindspore/ops/_op_impl/_custom_op/matmul_cube_fracz_right_mul_impl.py
View File

@ -23,7 +23,7 @@ from collections import namedtuple
import logging
from te import tik
from topi.cce import util
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
# General limitation of the size for input shape: 2**31

4
mindspore/python/mindspore/ops/_op_impl/_custom_op/matmul_cube_impl.py
View File

@ -22,8 +22,8 @@ from impl.matmul_vector import matmul_vector_cce
import te.platform.cce_params as cce
import te.lang.cce
from te import tvm
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
from mindspore.ops._op_impl._custom_op._basic import _shape_check, _get_bias, _get_input_shape

2
mindspore/python/mindspore/ops/_op_impl/_custom_op/matrix_combine_impl.py
View File

@ -16,7 +16,7 @@
from __future__ import absolute_import
from te import tik
from topi.cce import util
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
cus_matrix_combine_op_info = TBERegOp("CusMatrixCombine") \

4
mindspore/python/mindspore/ops/_op_impl/_custom_op/minmax_update_perchannel.py
View File

@ -17,8 +17,8 @@
import te.lang.cce
from te import tvm
from te.platform.fusion_manager import fusion_manager
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
minmax_update_perchannel_op_info = TBERegOp("MinMaxUpdatePerChannel") \

4
mindspore/python/mindspore/ops/_op_impl/_custom_op/minmax_update_perlayer.py
View File

@ -18,8 +18,8 @@ from functools import reduce as functools_reduce
import te.lang.cce
from te import tvm
from te.platform.fusion_manager import fusion_manager
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
minmax_update_perlayer_op_info = TBERegOp("MinMaxUpdatePerLayer") \

2
mindspore/python/mindspore/ops/_op_impl/_custom_op/transpose02314_impl.py
View File

@ -16,7 +16,7 @@
from __future__ import absolute_import
from te import tik
from topi.cce import util
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
cus_transpose02314_op_info = TBERegOp("CusTranspose02314") \

12
mindspore/python/mindspore/run_check/_check_version.py
View File

@ -255,7 +255,7 @@ class AscendEnvChecker(EnvChecker):
def __init__(self, library_path):
self.library_path = library_path
self.version = ["1.84"]
self.version = ["6.3"]
atlas_nnae_version = "/usr/local/Ascend/nnae/latest/compiler/version.info"
atlas_toolkit_version = "/usr/local/Ascend/ascend-toolkit/latest/compiler/version.info"
hisi_fwk_version = "/usr/local/Ascend/latest/compiler/version.info"
@ -331,7 +331,7 @@ class AscendEnvChecker(EnvChecker):
def check_deps_version(self):
"""
te, topi, hccl wheel package version check
te and hccl wheel package version check
in order to update the change of 'LD_LIBRARY_PATH' env, run a sub process
"""
@ -345,12 +345,6 @@ class AscendEnvChecker(EnvChecker):
attention_warning = True
logger.warning(f"MindSpore version {mindspore_version} and \"te\" wheel package version {v} does not "
"match, reference to the match info on: https://www.mindspore.cn/install")
from topi import version as topiver
v = '.'.join(topiver.version.split('.')[0:2])
if v not in supported_version:
attention_warning = True
logger.warning(f"MindSpore version {mindspore_version} and \"topi\" wheel package version {v} does not "
"match, reference to the match info on: https://www.mindspore.cn/install")
from hccl import sys_version as hccl_version
v = '.'.join(hccl_version.__sys_version__.split('.')[0:2])
if v not in supported_version:
@ -361,7 +355,7 @@ class AscendEnvChecker(EnvChecker):
# pylint: disable=broad-except
except Exception as e:
logger.error("CheckFailed:", e.args)
logger.error("MindSpore relies on the 3 whl packages of \"te\", \"topi\" and \"hccl\" in the \"latest\" "
logger.error("MindSpore relies on whl packages of \"te\" and \"hccl\" in the \"latest\" "
"folder of the Ascend AI software package (Ascend Data Center Solution), please check whether"
" they are installed correctly or not, reference to the match info on: "
"https://www.mindspore.cn/install")

5
tests/st/cpp/CMakeLists.txt
View File

@ -8,6 +8,11 @@ include_directories(${CMAKE_SOURCE_DIR}/mindspore/core)
include_directories(${CMAKE_BINARY_DIR})
include_directories(${CUDA_INCLUDE_DIRS})
# graphengine include directories
if(ENABLE_D OR ENABLE_ACL OR ENABLE_TESTCASES)
include(${CMAKE_SOURCE_DIR}/cmake/graphengine_variables.cmake)
endif()
if(ENABLE_ACL)
add_definitions(-D ENABLE_ACL)
endif()

2
tests/st/networks/models/bert/bert_performance/test_bert_tdt_lossscale.py
View File

@ -174,7 +174,7 @@ class TimeMonitor(Callback):
self.per_step_mseconds_list.append(epoch_mseconds / self.data_size)
@pytest.mark.level0
@pytest.mark.level1
@pytest.mark.platform_arm_ascend_training
@pytest.mark.platform_x86_ascend_training
@pytest.mark.env_onecard

2
tests/st/ops/ascend/test_deformable_conv.py
View File

@ -23,7 +23,7 @@ from mindspore import Tensor
context.set_context(device_target="Ascend")
@pytest.mark.level0
@pytest.mark.level1
@pytest.mark.platform_x86_ascend_training
@pytest.mark.platform_arm_ascend_training
@pytest.mark.env_onecard

2
tests/st/ops/ascend/test_tbe_ops/test_deformable_offsets_grad.py
View File

@ -59,7 +59,7 @@ class Grad(nn.Cell):
return self.grad(self.network)(x, w, offset, output_grad)
@pytest.mark.level0
@pytest.mark.level1
@pytest.mark.platform_x86_ascend_training
@pytest.mark.platform_arm_ascend_training
@pytest.mark.env_onecard

2
tests/st/ops/ascend/test_unsorted_segment_op.py
View File

@ -126,7 +126,7 @@ class TestUnsortedSegmentArithmeticNet(nn.Cell):
return self.func(x, segment_ids, self.num_segments)
@pytest.mark.level0
@pytest.mark.level1
@pytest.mark.platform_arm_ascend_training
@pytest.mark.platform_x86_ascend_training
@pytest.mark.env_onecard

4
tests/st/ops/custom_ops_tbe/add3_impl.py
View File

@ -16,8 +16,8 @@ from __future__ import absolute_import
import te.lang.cce
from te import tvm
from te.platform.fusion_manager import fusion_manager
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
@fusion_manager.register("add3")

6
tests/st/ops/custom_ops_tbe/conv_layer.py
View File

@ -15,9 +15,9 @@
import te.lang.cce
from te import tvm
from te.platform import CUBE_MKN
from topi import generic
from topi.cce import util
from topi.cce.util import is_v200_version
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from tbe.tvm.topi.cce.util import is_v200_version
# pylint: disable=R0912,R0913,R0914,R0915,E1101
# the dim of shape in conv must be 4

4
tests/st/ops/custom_ops_tbe/square_impl.py
View File

@ -17,8 +17,8 @@ from __future__ import absolute_import
import te.lang.cce
from te import tvm
from te.platform.fusion_manager import fusion_manager
from topi import generic
from topi.cce import util
from tbe.tvm.topi import generic
from tbe.tvm.topi.cce import util
from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType

2
tests/st/ops/dynamic_shape/test_dynamic_setitem.py
View File

@ -114,7 +114,7 @@ class TensorSetItem(nn.Cell):
return tensor1, tensor2
@pytest.mark.level0
@pytest.mark.level1
@pytest.mark.platform_x86_cpu
@pytest.mark.platform_x86_gpu_training
@pytest.mark.platform_arm_ascend_training

2
tests/st/ops/graph_kernel/custom/test_custom_tbe.py
View File

@ -33,7 +33,7 @@ from mindspore.ops.composite.multitype_ops.zeros_like_impl import zeros_like
def square_with_bias(input_x, output_y, bias=0.0, kernel_name="square_with_bias"):
import te.lang.cce
from te import tvm
from topi.cce import util
from tbe.tvm.topi.cce import util
shape = input_x.get("shape")
dtype = input_x.get("dtype").lower()

4
tests/ut/cpp/CMakeLists.txt
View File

@ -34,6 +34,10 @@ include_directories(${CMAKE_BINARY_DIR}/proto/metadef_protos)
include_directories(${CMAKE_BINARY_DIR}/proto/ge)
include_directories(${CUDA_INCLUDE_DIRS})
include_directories(${CMAKE_SOURCE_DIR}/mindspore/ccsrc/plugin/device/cpu/kernel)
# graphengine include directories
if(ENABLE_D OR ENABLE_ACL OR ENABLE_TESTCASES)
include(${CMAKE_SOURCE_DIR}/cmake/graphengine_variables.cmake)
endif()
MESSAGE("check ut_test ${CMAKE_BINARY_DIR}")
link_directories(${MS_CCSRC_BUILD_PATH})

2
tests/ut/cpp/stub/runtime/runtime_stub.cc
View File

@ -30,7 +30,7 @@ rtError_t rtEventCreateWithFlag(rtEvent_t *event, uint32_t flag) { return RT_ERR
rtError_t rtEventElapsedTime(float *time, rtEvent_t start, rtEvent_t end) { return RT_ERROR_NONE; }
rtError_t rtMalloc(void **devPtr, uint64_t size, rtMemType_t type) { return RT_ERROR_NONE; }
rtError_t rtMalloc(void **devPtr, uint64_t size, rtMemType_t type, const uint16_t moduleId) { return RT_ERROR_NONE; }
rtError_t rtMemcpy(void *dst, uint64_t destMax, const void *src, uint64_t count, rtMemcpyKind_t kind) {
return RT_ERROR_NONE;