X86_64 Ubuntu18.04编译教程

[LEMON-LM]

大佬们有没有详细一点的编译教程？
我的环境是Ubutun18.04 + Nvidia T4 ,
我看CMakeLits 依赖xnnpack/oneapi/popart 而且一些路径都是写死的都是“/opt/*”这种（ODLA/platforms/CMakeLists.txt文件中），对于eigen,以及popart 都有变量设置编译的时候不编译，但是对于intel gpu的没有找到编译开关，编译的的时候一直找不到DNNL_COMPILER

TensorRT/POPart/oneAPI 是对接NV GPU 和Graphcore IPU以及Intel GPU的，XNNPACK和eigen 是针对CPU的吗？

ODLA_BUILD_DNNL_GPU 这个变量是针对那个设备的呢？

Cmake 报错信息：
root@3e04a12126ca:~/HALO/heterogeneity-aware-lowering-and-optimization/build# /root/cmake-3.15.5-Linux-x86_64/bin/cmake -DCMAKE_BUILD_TYPE=Release -G Ninja -DHALO_USE_GLOG=[ON] -DHALO_CCACHE_BUILD=[ON] ..
-- clang-tidy found: /usr/bin/clang-tidy
-- Found Protobuf 3.0.0: /usr/lib/x86_64-linux-gnu/libprotobuf.so
-- Found Flatbuffers 1.12.0:
-- clang project is enabled
-- clang-tools-extra project is disabled
-- compiler-rt project is disabled
-- cross-project-tests project is disabled
-- libc project is disabled
-- libclc project is disabled
-- libcxx project is disabled
-- libcxxabi project is disabled
-- libunwind project is disabled
-- lld project is enabled
-- lldb project is disabled
-- mlir project is disabled
-- openmp project is disabled
-- polly project is disabled
-- pstl project is disabled
-- flang project is disabled
-- Could NOT find LibXml2 (missing: LIBXML2_LIBRARY LIBXML2_INCLUDE_DIR)
-- Native target architecture is X86
-- Threads enabled.
-- Doxygen disabled.
-- Go bindings disabled.
-- Ninja version: 1.7.2
-- Could NOT find OCaml (missing: OCAMLFIND OCAML_VERSION OCAML_STDLIB_PATH)
-- Could NOT find OCaml (missing: OCAMLFIND OCAML_VERSION OCAML_STDLIB_PATH)
-- OCaml bindings disabled.
-- LLVM host triple: x86_64-unknown-linux-gnu
-- LLVM default target triple: x86_64-unknown-linux-gnu
-- Building with -fPIC
-- Targeting AArch64
-- Targeting AMDGPU
-- Targeting ARM
-- Targeting AVR
-- Targeting BPF
-- Targeting Hexagon
-- Targeting Lanai
-- Targeting Mips
-- Targeting MSP430
-- Targeting NVPTX
-- Targeting PowerPC
-- Targeting RISCV
-- Targeting Sparc
-- Targeting SystemZ
-- Targeting VE
-- Targeting WebAssembly
-- Targeting X86
-- Targeting XCore
-- Clang version: 14.0.0
-- Not building amdgpu-arch: hsa-runtime64 not found
-- LLD version: 14.0.0
-- Registering Bye as a pass plugin (static build: OFF)
-- Failed to find LLVM FileCheck
-- git version: v0.0.0 normalized to 0.0.0
-- Version: 1.6.0
-- Performing Test HAVE_GNU_POSIX_REGEX -- failed to compile
-- Performing Test HAVE_POSIX_REGEX -- success
-- Performing Test HAVE_STEADY_CLOCK -- success
-- Eigen not found
-- GCC10 path not found, skip enable bf16 on odla_dnnl
-- GPU_ARCHS is not defined. Generating CUDA code for default SMs: 35;53;61;70;75
CMake Error: The following variables are used in this project, but they are set to NOTFOUND.
Please set them or make sure they are set and tested correctly in the CMake files:
dnnl
linked by target "odla_dnnl_interpreter" in directory /root/HALO/heterogeneity-aware-lowering-and-optimization/ODLA/platforms/dnnl
linked by target "odla_dnnl" in directory /root/HALO/heterogeneity-aware-lowering-and-optimization/ODLA/platforms/dnnl

-- Configuring incomplete, errors occurred!
See also "/root/HALO/heterogeneity-aware-lowering-and-optimization/build/CMakeFiles/CMakeOutput.log".
See also "/root/HALO/heterogeneity-aware-lowering-and-optimization/build/CMakeFiles/CMakeError.log".

[LEMON-LM]

搞定了~~对应平台的编译开关在ODLA/platforms/CMakeLists.txt文件中
ODLA_BUILD_EIGEN 对应eigen
ODLA_BUILD_DNNL 对应intel oneAPI
ODLA_BUILD_POPART 对应Graphcore IPU
ODLA_BUILD_TRT 对应 NV TensorRT
ODLA_BUILD_VODLA 对应NV DLA??(这个不认识)
ODLA_BUILD_XNNPACK 对应Google XNNPACK

[weimingzha0]

开发/编译的环境 docker pull registry-intl.us-west-1.aliyuncs.com/computation/halo:0.7.6-devel-cuda11.4.2-cudnn8-ubuntu18.04

可以参考 https://github.com/alibaba/heterogeneity-aware-lowering-and-optimization/blob/master/utils/docker/Dockerfile
自己构建image.