InfiniTensor/include/kunlun/kunlun_runtime.h

74 lines
2.4 KiB
C
Raw Normal View History

Xpu (#82) * support kunlun xpu and add an operator named Add * add sub, mul, div, pow, maximum, minimum * add code * add xpu code * add code * add matmul * add transpose * add unary operator * add unary operator * add some operator * add code * support run resnet18 on xpu * add code * add max pool2d * fix xpu code, let it can run. * 添加XPU算子 (#120) * add floordiv for xpu * add batchnorm for xpu * add more cast types for xpu * add conv_trans for xpu * add pad for xpu * add logical ops for xpu * fix format for xpu src and include * fix format for xpu test * fix format for xpu src --------- Co-authored-by: Bolun <bolunz@u.nus.edu> * Xpu abs (#121) * add: unary kernel for xpu * formatting * format * format * format * fix: pointer jump * fix optype comments * fix bug introduced while resolving conflict * change cmake option for kunlunxin xpu from 'xpu' to 'kunlun'; fix bug after merging distributed infrastructure * Add doc support for xpu (#141) * fix * fix * fix pooling test * format * format * fix * fix * set cmake version requirement * fix cmakelists * rename xpu to kunlun * fix * fix format * fix format * fix format * fix change name to kunlun * format * fix format * clang format * fix format --------- Co-authored-by: root <root@localhost.localdomain> Co-authored-by: wanghailu <wanghailu@qiyuanlab.com> Co-authored-by: wanghailu <wanghailu0717@163.com> Co-authored-by: Bolun Zhang <48948016+Chamberlain0w0@users.noreply.github.com> Co-authored-by: Bolun <bolunz@u.nus.edu> Co-authored-by: zhangyue207 <138768300+zhangyue207@users.noreply.github.com> Co-authored-by: Haojie Wang <haojie0429@gmail.com> Co-authored-by: baominghelly <41820386+baominghelly@users.noreply.github.com> Co-authored-by: Bolun <chamberlain0w0@gmail.com>
2023-10-16 10:57:08 +08:00
#pragma once
#include "core/runtime.h"
#include "kunlun/kunlun_common.h"
namespace infini {
class KUNLUNRuntimeObj : public RuntimeObj {
private:
baidu::xpu::api::Context *xdnn;
KUNLUNPtr workspace;
size_t workspaceSize;
public:
KUNLUNRuntimeObj() : RuntimeObj(Device::KUNLUN) {
xdnn = baidu::xpu::api::create_context();
// 10GB for Longformer
// size_t longformerNum = 3lu * (1 << 30);
workspaceSize = 3ll << 30; // 3 GB
// std::cout<<workspaceSize/1024/1024/1024<< std::endl;
// std::cout<<std::bitset<64>(workspaceSize)<< std::endl;
workspace = alloc(workspaceSize);
}
virtual ~KUNLUNRuntimeObj() {
dealloc(workspace);
baidu::xpu::api::destroy_context(xdnn);
}
string toString() const override;
void run(const Graph &graph, bool tune = false,
bool profiling = false) const;
// double runEvaluation(const Graph &graph, int nWarmups,
// int nEvaluations) const;
void sync() const;
KUNLUNPtr alloc(size_t size) override {
void *ptr;
checkKUNLUNError(
xpu_malloc_ex((void **)&ptr, size, XPUMemoryKind::XPU_MEM_MAIN));
return ptr;
}
void dealloc(void *ptr) override { xpu_free(ptr); }
baidu::xpu::api::Context *KUNLUNHandle() const { return xdnn; }
KUNLUNPtr getWorkspace(size_t size) const {
IT_ASSERT(size <= workspaceSize);
return workspace;
}
void copyBlobFromCPU(void *dst, const void *src,
size_t bytes) const override {
xpu_memcpy(dst, const_cast<void *>(src), bytes,
XPUMemcpyKind::XPU_HOST_TO_DEVICE);
}
void copyBlobToCPU(void *dst, const void *src,
size_t bytes) const override {
xpu_memcpy(dst, const_cast<void *>(src), bytes,
XPUMemcpyKind::XPU_DEVICE_TO_HOST);
}
void copyBlobInsideRuntime(void *dst, const void *src,
size_t bytes) const override {
xpu_memcpy(dst, const_cast<void *>(src), bytes,
XPUMemcpyKind::XPU_DEVICE_TO_DEVICE);
}
void initComm(const string &, int, int) override { IT_TODO_HALT(); }
CommunicatorObj &getCommunicator() const override { IT_TODO_HALT(); }
private:
void runWithoutSync(const Graph &graph, bool tune, bool profiling) const;
};
} // namespace infini