aclnnBitwiseXorTensor&aclnnInplaceBitwiseXorTensor
支持的产品型号
- Atlas 推理系列产品(Ascend 310P处理器)。
- Atlas 训练系列产品。
- Atlas A2训练系列产品/Atlas 800I A2推理产品。
接口原型
aclnnBitwiseXorTensor和aclnnInplaceBitwiseXorTensor实现相同的功能,使用区别如下,请根据自身实际场景选择合适的算子。
- aclnnBitwiseXorTensor:需新建一个输出张量对象存储计算结果。
- aclnnInplaceBitwiseXorTensor:无需新建输出张量对象,直接在输入张量的内存中存储计算结果。
每个算子分为两段式接口,必须先调用“aclnnBitwiseXorTensorGetWorkspaceSize”或者”aclnnInplaceBitwiseXorTensorGetWorkspaceSize“接口获取计算所需workspace大小以及包含了算子计算流程的执行器,再调用“aclnnBitwiseXorTensor”或者”aclnnInplaceBitwiseXorTensor“接口执行计算。
aclnnStatus aclnnBitwiseXorTensorGetWorkspaceSize(const aclTensor *self, const aclTensor *other, aclTensor *out, uint64_t *workspaceSize, aclOpExecutor **executor)
aclnnStatus aclnnBitwiseXorTensor(void *workspace, uint64_t workspaceSize, aclOpExecutor *executor, aclrtStream stream)
aclnnStatus aclnnInplaceBitwiseXorTensorGetWorkspaceSize(aclTensor *selfRef, const aclTensor *other, uint64_t *workspaceSize, aclOpExecutor **executor)
aclnnStatus aclnnInplaceBitwiseXorTensor(void *workspace, uint64_t workspaceSize, aclOpExecutor *executor, aclrtStream stream)
功能描述
- 算子功能:计算输入张量self中每个元素与输入张量other中对应位置元素的按位异或,输入self和other必须是整数或布尔类型,对于布尔类型,计算逻辑异或。
- 计算公式:
aclnnBitwiseXorTensorGetWorkspaceSize
参数说明:
- self(const aclTensor *, 计算输入):Device侧的aclTensor,数据类型支持BOOL、INT8、INT16、INT32、INT64、UINT8,且数据类型与other的数据类型需满足数据类型推导规则(参见互推导关系),shape需要与other满足broadcast关系,支持非连续的Tensor,数据格式支持ND,数据维度不支持8维以上。
- other(const aclTensor *, 计算输入):Device侧的aclTensor,数据类型支持BOOL、INT8、INT16、INT32、INT64、UINT8,且数据类型与self的数据类型需满足数据类型推导规则(参见互推导关系),shape需要与self满足broadcast关系,支持非连续的Tensor,数据格式支持ND,数据维度不支持8维以上。
- out(aclTensor *, 计算输出):Device侧的aclTensor,数据类型支持BOOL、INT8、INT16、INT32、INT64、UINT8、FLOAT、FLOAT16、DOUBLE、BFLOAT16(仅Atlas A2训练系列产品/Atlas 800I A2推理产品支持)、COMPLEX64、COMPLEX128,且数据类型需要是self与other推导之后可转换的数据类型,shape需要是self与other broadcast之后的shape,支持非连续的Tensor,数据格式支持ND,数据维度不支持8维以上。
- workspaceSize(uint64_t *, 出参):返回需要在Device侧申请的workspace大小。
- executor(aclOpExecutor **, 出参):返回op执行器,包含了算子计算流程。
返回值:
aclnnStatus:返回状态码,具体参见aclnn返回码。
161001(ACLNN_ERR_PARAM_NULLPTR): 1. 参数self、other、out是空指针。 161002(ACLNN_ERR_PARAM_INVALID): 1. 参数self、other的数据类型不在支持范围内。 2. 参数self和other不满足数据类型推导规则。 3. 参数self和other推导出的数据类型不能转换为out的数据类型。 4. 参数self和other的shape无法做broadcast。 5. 参数out的shape不是self和other经过broadcast之后的shape。 6. 参数self、other,out的维度大于8。
aclnnBitwiseXorTensor
参数说明:
- workspace(void *, 入参):在Device侧申请的workspace内存地址。
- workspaceSize(uint64_t, 入参):在Device侧申请的workspace大小,由第一段接口aclnnBitwiseXorTensorGetWorkspaceSize获取。
- executor(aclOpExecutor *, 入参):op执行器,包含了算子计算流程。
- stream(aclrtStream, 入参):指定执行任务的AscendCL Stream流。
返回值:
aclnnStatus:返回状态码,具体参见aclnn返回码。
aclnnInplaceBitwiseXorTensorGetWorkspaceSize
参数说明:
- selfRef(aclTensor *,计算输入|计算输出):输入输出tensor,即公式中的input与out。Device侧的aclTensor,数据类型支持BOOL、INT8、INT16、INT32、INT64、UINT8,且数据类型与other的数据类型需满足数据类型推导规则(参见互推导关系),且推导后的数据类型需要能转换成selfRef自身的数据类型,shape需要与other满足broadcast关系,且broadcast之后的shape需要与selfRef自身的shape相同,支持非连续的Tensor,数据格式支持ND,数据维度不支持8维以上。
- other(const aclTensor *,计算输入):计算输入,Device侧的aclTensor,数据类型支持BOOL、INT8、INT16、INT32、INT64、UINT8,且数据类型与selfRef的数据类型需满足数据类型推导规则(参见互推导关系),shape需要与selfRef满足broadcast关系,支持非连续的Tensor,数据格式支持ND,数据维度不支持8维以上。
- workspaceSize(uint64_t *,出参):返回需要在Device侧申请的workspace大小。
- executor(aclOpExecutor **,出参):返回op执行器,包含了算子计算流程。
返回值:
aclnnStatus:返回状态码,具体参见aclnn返回码。
161001(ACLNN_ERR_PARAM_NULLPTR): 1. 参数selfRef、other是空指针。 161002(ACLNN_ERR_PARAM_INVALID): 1. 参数selfRef、other的数据类型不在支持范围内。 2. 参数selfRef和other不满足数据类型推导规则。 3. 参数selfRef和other推导出的数据类型不能转换为selfRef的数据类型。 4. 参数selfRef和other的shape无法做broadcast。 5. 参数selfRef的shape不是selfRef和other经过broadcast之后的shape。 6. 参数selfRef、other的维度大于8。
aclnnInplaceBitwiseXorTensor
参数说明:
- workspace(void *, 入参):在Device侧申请的workspace内存地址。
- workspaceSize(uint64_t, 入参):在Device侧申请的workspace大小,由第一段接口aclnnInplaceBitwiseXorTensorGetWorkspaceSize获取。
- executor(aclOpExecutor *, 入参):op执行器,包含了算子计算流程。
- stream(aclrtStream, 入参):指定执行任务的AscendCL Stream流。
返回值:
aclnnStatus:返回状态码,具体参见aclnn返回码。
约束与限制
无
调用示例
示例代码如下,仅供参考,具体编译和执行过程请参考编译与运行样例。
aclnnBitwiseXorTensor示例代码:
#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_bitwise_xor_tensor.h"
#define CHECK_RET(cond, return_expr) \
do { \
if (!(cond)) { \
return_expr; \
} \
} while (0)
#define LOG_PRINT(message, ...) \
do { \
printf(message, ##__VA_ARGS__); \
} while (0)
int64_t GetShapeSize(const std::vector<int64_t>& shape) {
int64_t shape_size = 1;
for (auto i : shape) {
shape_size *= i;
}
return shape_size;
}
int Init(int32_t deviceId, aclrtStream* stream) {
// 固定写法,AscendCL初始化
auto ret = aclInit(nullptr);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclInit failed. ERROR: %d\n", ret); return ret);
ret = aclrtSetDevice(deviceId);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSetDevice failed. ERROR: %d\n", ret); return ret);
ret = aclrtCreateStream(stream);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtCreateStream failed. ERROR: %d\n", ret); return ret);
return 0;
}
template <typename T>
int CreateAclTensor(const std::vector<T>& hostData, const std::vector<int64_t>& shape, void** deviceAddr,
aclDataType dataType, aclTensor** tensor) {
auto size = GetShapeSize(shape) * sizeof(T);
// 调用aclrtMalloc申请device侧内存
auto ret = aclrtMalloc(deviceAddr, size, ACL_MEM_MALLOC_HUGE_FIRST);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtMalloc failed. ERROR: %d\n", ret); return ret);
// 调用aclrtMemcpy将host侧数据拷贝到device侧内存上
ret = aclrtMemcpy(*deviceAddr, size, hostData.data(), size, ACL_MEMCPY_HOST_TO_DEVICE);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtMemcpy failed. ERROR: %d\n", ret); return ret);
// 计算连续tensor的strides
std::vector<int64_t> strides(shape.size(), 1);
for (int64_t i = shape.size() - 2; i >= 0; i--) {
strides[i] = shape[i + 1] * strides[i + 1];
}
// 调用aclCreateTensor接口创建aclTensor
*tensor = aclCreateTensor(shape.data(), shape.size(), dataType, strides.data(), 0, aclFormat::ACL_FORMAT_ND,
shape.data(), shape.size(), *deviceAddr);
return 0;
}
int main() {
// 1. (固定写法)device/stream初始化, 参考AscendCL对外接口列表
// 根据自己的实际device填写deviceId
int32_t deviceId = 0;
aclrtStream stream;
auto ret = Init(deviceId, &stream);
// check根据自己的需要处理
CHECK_RET(ret == 0, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);
// 2. 构造输入与输出,需要根据API的接口自定义构造
std::vector<int64_t> selfShape = {4, 2};
std::vector<int64_t> otherShape = {4, 2};
std::vector<int64_t> outShape = {4, 2};
void* selfDeviceAddr = nullptr;
void* otherDeviceAddr = nullptr;
void* outDeviceAddr = nullptr;
aclTensor* self = nullptr;
aclTensor* other = nullptr;
aclTensor* out = nullptr;
std::vector<int64_t> selfHostData = {0, 1, 2, 3, 4, 5, 6, 7};
std::vector<int64_t> otherHostData = {0, 1, 1, 9, 3, 4, 5, 6};
std::vector<int64_t> outHostData = {0, 0, 0, 0, 0, 0, 0, 0};
// 创建self aclTensor
ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_INT64, &self);
CHECK_RET(ret == ACL_SUCCESS, return ret);
// 创建other aclTensor
ret = CreateAclTensor(otherHostData, otherShape, &otherDeviceAddr, aclDataType::ACL_INT64, &other);
CHECK_RET(ret == ACL_SUCCESS, return ret);
// 创建out aclTensor
ret = CreateAclTensor(outHostData, outShape, &outDeviceAddr, aclDataType::ACL_INT64, &out);
CHECK_RET(ret == ACL_SUCCESS, return ret);
// 3. 调用CANN算子库API,需要修改为具体的API
uint64_t workspaceSize = 0;
aclOpExecutor* executor;
// 调用aclnnBitwiseXorTensor第一段接口
ret = aclnnBitwiseXorTensorGetWorkspaceSize(self, other, out, &workspaceSize, &executor);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnBitwiseXorTensorGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
// 根据第一段接口计算出的workspaceSize申请device内存
void* workspaceAddr = nullptr;
if (workspaceSize > 0) {
ret = aclrtMalloc(&workspaceAddr, workspaceSize, ACL_MEM_MALLOC_HUGE_FIRST);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("allocate workspace failed. ERROR: %d\n", ret); return ret;);
}
// 调用aclnnBitwiseXorTensor第二段接口
ret = aclnnBitwiseXorTensor(workspaceAddr, workspaceSize, executor, stream);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnBitwiseXorTensor failed. ERROR: %d\n", ret); return ret);
// 4. (固定写法)同步等待任务执行结束
ret = aclrtSynchronizeStream(stream);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSynchronizeStream failed. ERROR: %d\n", ret); return ret);
// 5. 获取输出的值,将device侧内存上的结果拷贝至host侧,需要根据具体API的接口定义修改
auto size = GetShapeSize(outShape);
std::vector<int64_t> resultData(size, 0);
ret = aclrtMemcpy(resultData.data(), resultData.size() * sizeof(resultData[0]), outDeviceAddr,
size * sizeof(resultData[0]), ACL_MEMCPY_DEVICE_TO_HOST);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("copy result from device to host failed. ERROR: %d\n", ret); return ret);
for (int64_t i = 0; i < size; i++) {
LOG_PRINT("result[%ld] is: %ld\n", i, resultData[i]);
}
// 6. 释放aclTensor,需要根据具体API的接口定义修改
aclDestroyTensor(self);
aclDestroyTensor(other);
aclDestroyTensor(out);
// 7. 释放device资源,需要根据具体API的接口定义修改
aclrtFree(selfDeviceAddr);
aclrtFree(otherDeviceAddr);
aclrtFree(outDeviceAddr);
if (workspaceSize > 0) {
aclrtFree(workspaceAddr);
}
aclrtDestroyStream(stream);
aclrtResetDevice(deviceId);
aclFinalize();
return 0;
}
aclnnInplaceBitwiseXorTensor示例代码:
#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_bitwise_xor_tensor.h"
#define CHECK_RET(cond, return_expr) \
do { \
if (!(cond)) { \
return_expr; \
} \
} while (0)
#define LOG_PRINT(message, ...) \
do { \
printf(message, ##__VA_ARGS__); \
} while (0)
int64_t GetShapeSize(const std::vector<int64_t>& shape) {
int64_t shape_size = 1;
for (auto i : shape) {
shape_size *= i;
}
return shape_size;
}
int Init(int32_t deviceId, aclrtStream* stream) {
// 固定写法,AscendCL初始化
auto ret = aclInit(nullptr);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclInit failed. ERROR: %d\n", ret); return ret);
ret = aclrtSetDevice(deviceId);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSetDevice failed. ERROR: %d\n", ret); return ret);
ret = aclrtCreateStream(stream);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtCreateStream failed. ERROR: %d\n", ret); return ret);
return 0;
}
template <typename T>
int CreateAclTensor(const std::vector<T>& hostData, const std::vector<int64_t>& shape, void** deviceAddr,
aclDataType dataType, aclTensor** tensor) {
auto size = GetShapeSize(shape) * sizeof(T);
// 调用aclrtMalloc申请device侧内存
auto ret = aclrtMalloc(deviceAddr, size, ACL_MEM_MALLOC_HUGE_FIRST);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtMalloc failed. ERROR: %d\n", ret); return ret);
// 调用aclrtMemcpy将host侧数据拷贝到device侧内存上
ret = aclrtMemcpy(*deviceAddr, size, hostData.data(), size, ACL_MEMCPY_HOST_TO_DEVICE);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtMemcpy failed. ERROR: %d\n", ret); return ret);
// 计算连续tensor的strides
std::vector<int64_t> strides(shape.size(), 1);
for (int64_t i = shape.size() - 2; i >= 0; i--) {
strides[i] = shape[i + 1] * strides[i + 1];
}
// 调用aclCreateTensor接口创建aclTensor
*tensor = aclCreateTensor(shape.data(), shape.size(), dataType, strides.data(), 0, aclFormat::ACL_FORMAT_ND,
shape.data(), shape.size(), *deviceAddr);
return 0;
}
int main() {
// 1. (固定写法)device/stream初始化, 参考AscendCL对外接口列表
// 根据自己的实际device填写deviceId
int32_t deviceId = 0;
aclrtStream stream;
auto ret = Init(deviceId, &stream);
// check根据自己的需要处理
CHECK_RET(ret == 0, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);
// 2. 构造输入与输出,需要根据API的接口自定义构造
std::vector<int64_t> selfRefShape = {4, 2};
std::vector<int64_t> otherShape = {4, 2};
void* selfRefDeviceAddr = nullptr;
void* otherDeviceAddr = nullptr;
aclTensor* selfRef = nullptr;
aclTensor* other = nullptr;
std::vector<int64_t> selfRefHostData = {0, 1, 2, 3, 4, 5, 6, 7};
std::vector<int64_t> otherHostData = {0, 1, 1, 9, 3, 4, 5, 6};
// 创建selfRef aclTensor
ret = CreateAclTensor(selfRefHostData, selfRefShape, &selfRefDeviceAddr, aclDataType::ACL_INT64, &selfRef);
CHECK_RET(ret == ACL_SUCCESS, return ret);
// 创建other aclTensor
ret = CreateAclTensor(otherHostData, otherShape, &otherDeviceAddr, aclDataType::ACL_INT64, &other);
CHECK_RET(ret == ACL_SUCCESS, return ret);
// 3. 调用CANN算子库API,需要修改为具体的API
uint64_t workspaceSize = 0;
aclOpExecutor* executor;
// 调用aclnnInplaceBitwiseXorTensor第一段接口
ret = aclnnInplaceBitwiseXorTensorGetWorkspaceSize(selfRef, other, &workspaceSize, &executor);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnInplaceBitwiseXorTensorGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
// 根据第一段接口计算出的workspaceSize申请device内存
void* workspaceAddr = nullptr;
if (workspaceSize > 0) {
ret = aclrtMalloc(&workspaceAddr, workspaceSize, ACL_MEM_MALLOC_HUGE_FIRST);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("allocate workspace failed. ERROR: %d\n", ret); return ret;);
}
// 调用aclnnInplaceBitwiseXorTensor第二段接口
ret = aclnnInplaceBitwiseXorTensor(workspaceAddr, workspaceSize, executor, stream);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnInplaceBitwiseXorTensor failed. ERROR: %d\n", ret); return ret);
// 4. (固定写法)同步等待任务执行结束
ret = aclrtSynchronizeStream(stream);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSynchronizeStream failed. ERROR: %d\n", ret); return ret);
// 5. 获取输出的值,将device侧内存上的结果拷贝至host侧,需要根据具体API的接口定义修改
auto size = GetShapeSize(selfRefShape);
std::vector<int64_t> resultData(size, 0);
ret = aclrtMemcpy(resultData.data(), resultData.size() * sizeof(resultData[0]), selfRefDeviceAddr,
size * sizeof(resultData[0]), ACL_MEMCPY_DEVICE_TO_HOST);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("copy result from device to host failed. ERROR: %d\n", ret); return ret);
for (int64_t i = 0; i < size; i++) {
LOG_PRINT("result[%ld] is: %ld\n", i, resultData[i]);
}
// 6. 释放aclTensor,需要根据具体API的接口定义修改
aclDestroyTensor(selfRef);
aclDestroyTensor(other);
// 7. 释放device资源,需要根据具体API的接口定义修改
aclrtFree(selfRefDeviceAddr);
aclrtFree(otherDeviceAddr);
if (workspaceSize > 0) {
aclrtFree(workspaceAddr);
}
aclrtDestroyStream(stream);
aclrtResetDevice(deviceId);
aclFinalize();
return 0;
}