orbslam移植ios------cholmod一系列问题解决11

• 错误1
  Undefined symbols for architecture x86_64:
  “_jpeg_free_large”, referenced from:
  _free_pool in opencv2(jmemmgr.o)
  “_jpeg_free_small”, referenced from:
  _free_pool in opencv2(jmemmgr.o)
  _self_destruct in opencv2(jmemmgr.o)
  “_jpeg_get_large”, referenced from:
  _alloc_large in opencv2(jmemmgr.o)
  _alloc_barray in opencv2(jmemmgr.o)
  “_jpeg_get_small”, referenced from:
  _jinit_memory_mgr in opencv2(jmemmgr.o)
  _alloc_small in opencv2(jmemmgr.o)
  “_jpeg_mem_available”, referenced from:
  _realize_virt_arrays in opencv2(jmemmgr.o)
  “_jpeg_mem_init”, referenced from:
  _jinit_memory_mgr in opencv2(jmemmgr.o)
  “_jpeg_mem_term”, referenced from:
  _jinit_memory_mgr in opencv2(jmemmgr.o)
  _self_destruct in opencv2(jmemmgr.o)
  “_jpeg_open_backing_store”, referenced from:
  _realize_virt_arrays in opencv2(jmemmgr.o)
  ld: symbol(s) not found for architecture x86_64
  clang: error: linker command failed with exit code 1 (use -v to see invocation)

  解决办法：
  安装：libjpeg-turbo.dmg ： https://sourceforge.net/projects/libjpeg-turbo/files/1.4.0/
  在终端运行：lipo -info /opt/libjpeg-turbo/lib/libjpeg.a 将会出现以下内容：
  Architectures in the fat file: /opt/libjpeg-turbo/lib/libjpeg.a are: i386 x86_64 armv6 armv7 armv7s arm64
  然后将/opt/libjpeg-turbo/lib/libjpeg.a 加到你的工程里面

• 错误2
  Undefined symbols for architecture arm64:
  “zherk”, referenced from:
  _cholmod_super_numeric in libcholmod.a(cholmod_super_numeric.o)
  “zpotrf”, referenced from:
  _cholmod_super_numeric in libcholmod.a(cholmod_super_numeric.o)
  “dsyrk”, referenced from:
  _cholmod_super_numeric in libcholmod.a(cholmod_super_numeric.o)
  “dpotrf”, referenced from:
  _cholmod_super_numeric in libcholmod.a(cholmod_super_numeric.o)
  “ztrsv”, referenced from:
  _cholmod_super_lsolve in libcholmod.a(cholmod_super_solve.o)
  _cholmod_super_ltsolve in libcholmod.a(cholmod_super_solve.o)
  “zgemv”, referenced from:
  _cholmod_super_lsolve in libcholmod.a(cholmod_super_solve.o)
  _cholmod_super_ltsolve in libcholmod.a(cholmod_super_solve.o)
  “dtrsv”, referenced from:
  _cholmod_super_lsolve in libcholmod.a(cholmod_super_solve.o)
  _cholmod_super_ltsolve in libcholmod.a(cholmod_super_solve.o)
  “dgemv”, referenced from:
  _cholmod_super_lsolve in libcholmod.a(cholmod_super_solve.o)
  _cholmod_super_ltsolve in libcholmod.a(cholmod_super_solve.o)
  “VIO::VIO()”, referenced from:
  ___cxx_global_var_init.10 in ViewController.o
  “ztrsm”, referenced from:
  _cholmod_super_lsolve in libcholmod.a(cholmod_super_solve.o)
  _cholmod_super_ltsolve in libcholmod.a(cholmod_super_solve.o)
  _cholmod_super_numeric in libcholmod.a(cholmod_super_numeric.o)
  “zgemm”, referenced from:
  _cholmod_super_lsolve in libcholmod.a(cholmod_super_solve.o)
  _cholmod_super_ltsolve in libcholmod.a(cholmod_super_solve.o)
  _cholmod_super_numeric in libcholmod.a(cholmod_super_numeric.o)
  “dtrsm”, referenced from:
  _cholmod_super_lsolve in libcholmod.a(cholmod_super_solve.o)
  _cholmod_super_ltsolve in libcholmod.a(cholmod_super_solve.o)
  _cholmod_super_numeric in libcholmod.a(cholmod_super_numeric.o)
  “dgemm”, referenced from:
  _cholmod_super_lsolve in libcholmod.a(cholmod_super_solve.o)
  _cholmod_super_ltsolve in libcholmod.a(cholmod_super_solve.o)
  _cholmod_super_numeric in libcholmod.a(cholmod_super_numeric.o)
  ld: symbol(s) not found for architecture arm64

  解决办法：由报错第一行可以判定该报错与Cholmod库有关。无法正确链接Cholmod库。而Cholmod库最终是用的suitesparse库里的
  brew install suite-sparse
  然后在Header Search Paths 和Library Search Paths里添加suitesparse include和lib路径
  后面还是会有这个错，是因为一开始编译g2o的时候，就没真正编译好，重新把suitesparse也导入到g2o工程里，编译生成g2o.a就好了

• 错误3
  Undefined symbols for architecture arm64:
  “_cholmod_analyze”, referenced from:
  g2o::LinearSolverCholmod<Eigen::Matrix<double, -1, -1, 0, -1, -1> >::computeSymbolicDecomposition(g2o::SparseBlockMatrix<Eigen::Matrix<double, -1, -1, 0, -1, -1> > const&) in Optimizer.o
  g2o::LinearSolverCholmod<Eigen::Matrix<double, 6, 6, 0, 6, 6> >::computeSymbolicDecomposition(g2o::SparseBlockMatrix<Eigen::Matrix<double, 6, 6, 0, 6, 6> > const&) in Optimizer.o
  “_cholmod_amd”, referenced from:
  g2o::LinearSolverCholmod<Eigen::Matrix<double, -1, -1, 0, -1, -1> >::computeSymbolicDecomposition(g2o::SparseBlockMatrix<Eigen::Matrix<double, -1, -1, 0, -1, -1> > const&) in Optimizer.o
  g2o::LinearSolverCholmod<Eigen::Matrix<double, 6, 6, 0, 6, 6> >::computeSymbolicDecomposition(g2o::SparseBlockMatrix<Eigen::Matrix<double, 6, 6, 0, 6, 6> > const&) in Optimizer.o
  “_cholmod_factorize”, referenced from:
  g2o::LinearSolverCholmod<Eigen::Matrix<double, -1, -1, 0, -1, -1> >::solve(g2o::SparseBlockMatrix<Eigen::Matrix<double, -1, -1, 0, -1, -1> > const&, double*, double*) in Optimizer.o
  g2o::LinearSolverCholmod<Eigen::Matrix<double, -1, -1, 0, -1, -1> >::solveBlocks(double**&, g2o::SparseBlockMatrix<Eigen::Matrix<double, -1, -1, 0, -1, -1> > const&) in Optimizer.o
  g2o::LinearSolverCholmod<Eigen::Matrix<double, -1, -1, 0, -1, -1> >::solvePattern(g2o::SparseBlockMatrix<Eigen::Matrix<double, -1, -1, 0, -1, -1> >&, std::__1::vector<std::__1::pair<int, int>, std::__1::allocator<std::__1::pair<int, int> > > const&, g2o::SparseBlockMatrix<Eigen::Matrix<double, -1, -1, 0, -1, -1> > const&) in Optimizer.o
  g2o::LinearSolverCholmod<Eigen::Matrix<double, 6, 6, 0, 6, 6> >::solve(g2o::SparseBlockMatrix<Eigen::Matrix<double, 6, 6, 0, 6, 6> > const&, double*, double*) in Optimizer.o
  g2o::LinearSolverCholmod<Eigen::Matrix<double, 6, 6, 0, 6, 6> >::solveBlocks(double**&, g2o::SparseBlockMatrix<Eigen::Matrix<double, 6, 6, 0, 6, 6> > const&) in Optimizer.o
  g2o::LinearSolverCholmod<Eigen::Matrix<double, 6, 6, 0, 6, 6> >::solvePattern(g2o::SparseBlockMatrix<Eigen::Matrix<double, -1, -1, 0, -1, -1> >&, std::__1::vector<std::__1::pair<int, int>, std::__1::allocator<std::__1::pair<int, int> > > const&, g2o::SparseBlockMatrix<Eigen::Matrix<double, 6, 6, 0, 6, 6> > const&) in Optimizer.o
  “_cholmod_change_factor”, referenced from:
  g2o::LinearSolverCholmod<Eigen::Matrix<double, -1, -1, 0, -1, -1> >::solveBlocks(double**&, g2o::SparseBlockMatrix<Eigen::Matrix<double, -1, -1, 0, -1, -1> > const&) in Optimizer.o
  g2o::LinearSolverCholmod<Eigen::Matrix<double, -1, -1, 0, -1, -1> >::solvePattern(g2o::SparseBlockMatrix<Eigen::Matrix<double, -1, -1, 0, -1, -1> >&, std::__1::vector<std::__1::pair<int, int>, std::__1::allocator<std::__1::pair<int, int> > > const&, g2o::SparseBlockMatrix<Eigen::Matrix<double, -1, -1, 0, -1, -1> > const&) in Optimizer.o
  g2o::LinearSolverCholmod<Eigen::Matrix<double, 6, 6, 0, 6, 6> >::solveBlocks(double**&, g2o::SparseBlockMatrix<Eigen::Matrix<double, 6, 6, 0, 6, 6> > const&) in Optimizer.o
  g2o::LinearSolverCholmod<Eigen::Matrix<double, 6, 6, 0, 6, 6> >::solvePattern(g2o::SparseBlockMatrix<Eigen::Matrix<double, -1, -1, 0, -1, -1> >&, std::__1::vector<std::__1::pair<int, int>, std::__1::allocator<std::__1::pair<int, int> > > const&, g2o::SparseBlockMatrix<Eigen::Matrix<double, 6, 6, 0, 6, 6> > const&) in Optimizer.o
  “_cholmod_free_factor”, referenced from:
  g2o::LinearSolverCholmod<Eigen::Matrix<double, -1, -1, 0, -1, -1> >::init() in Optimizer.o
  g2o::LinearSolverCholmod<Eigen::Matrix<double, -1, -1, 0, -1, -1> >::~LinearSolverCholmod() in Optimizer.o
  g2o::LinearSolverCholmod<Eigen::Matrix<double, 6, 6, 0, 6, 6> >::init() in Optimizer.o
  g2o::LinearSolverCholmod<Eigen::Matrix<double, 6, 6, 0, 6, 6> >::~LinearSolverCholmod() in Optimizer.o
  “_cholmod_solve”, referenced from:
  g2o::LinearSolverCholmod<Eigen::Matrix<double, -1, -1, 0, -1, -1> >::solve(g2o::SparseBlockMatrix<Eigen::Matrix<double, -1, -1, 0, -1, -1> > const&, double*, double*) in Optimizer.o
  g2o::LinearSolverCholmod<Eigen::Matrix<double, 6, 6, 0, 6, 6> >::solve(g2o::SparseBlockMatrix<Eigen::Matrix<double, 6, 6, 0, 6, 6> > const&, double*, double*) in Optimizer.o
  “_cholmod_free_dense”, referenced from:
  g2o::LinearSolverCholmod<Eigen::Matrix<double, -1, -1, 0, -1, -1> >::solve(g2o::SparseBlockMatrix<Eigen::Matrix<double, -1, -1, 0, -1, -1> > const&, double*, double*) in Optimizer.o
  g2o::LinearSolverCholmod<Eigen::Matrix<double, 6, 6, 0, 6, 6> >::solve(g2o::SparseBlockMatrix<Eigen::Matrix<double, 6, 6, 0, 6, 6> > const&, double*, double*) in Optimizer.o
  “_cholmod_start”, referenced from:
  g2o::LinearSolverCholmod<Eigen::Matrix<double, -1, -1, 0, -1, -1> >::LinearSolverCholmod() in Optimizer.o
  g2o::LinearSolverCholmod<Eigen::Matrix<double, 6, 6, 0, 6, 6> >::LinearSolverCholmod() in Optimizer.o
  “_cholmod_analyze_p”, referenced from:
  g2o::LinearSolverCholmod<Eigen::Matrix<double, -1, -1, 0, -1, -1> >::computeSymbolicDecomposition(g2o::SparseBlockMatrix<Eigen::Matrix<double, -1, -1, 0, -1, -1> > const&) in Optimizer.o
  g2o::LinearSolverCholmod<Eigen::Matrix<double, 6, 6, 0, 6, 6> >::computeSymbolicDecomposition(g2o::SparseBlockMatrix<Eigen::Matrix<double, 6, 6, 0, 6, 6> > const&) in Optimizer.o
  “_cholmod_finish”, referenced from:
  g2o::LinearSolverCholmod<Eigen::Matrix<double, -1, -1, 0, -1, -1> >::~LinearSolverCholmod() in Optimizer.o
  g2o::LinearSolverCholmod<Eigen::Matrix<double, 6, 6, 0, 6, 6> >::~LinearSolverCholmod() in Optimizer.o
  “VIO::VIO()”, referenced from:
  ___cxx_global_var_init.10 in ViewController.o
  ld: symbol(s) not found for architecture arm64

后来发现，链接的库还是有问题
最终是自己去网上找了suitesparse源码，然后编译，这样模拟器就可以编译通过了
不要直接在别人的代码里直接找cholmod的头文件和库文件，这样很容易出问题（当然可以直接先这样尝试，如果报某个包相关的错误，那么就应该去网上找源码，然后自己按照官网编译即可，此时可能会出现模拟器编译通过，但ios真机编译失败，而且生成的总是dylib，没有.a静态库文件，此时就是编译的方式换一换，一般是修改CMakeLists.txt，同时生成dylib和.a静态库文件）

前面这么说的原因，自然是因为别人导了这些头文件和库文件，但很有可能它还要依赖某些头文件和库文件，但是你有时候会找不到它到底在哪里导了，或者说有没有导这些依赖库

通过CMakeLists.txt生成.a静态库文件