[petsc-dev] First call to cudaMalloc or cudaFree is very slow on summit

Zhang, Hong hongzhang at anl.gov
Sat Feb 8 17:21:57 CST 2020



On Feb 8, 2020, at 5:03 PM, Matthew Knepley <knepley at gmail.com<mailto:knepley at gmail.com>> wrote:

On Sat, Feb 8, 2020 at 4:34 PM Zhang, Hong via petsc-dev <petsc-dev at mcs.anl.gov<mailto:petsc-dev at mcs.anl.gov>> wrote:
I did some further investigation. The overhead persists for both the PETSc shared library and the static library. In the previous example, it does not call any PETSc function, the first CUDA function becomes very slow when it is linked to the petsc so. This indicates that the slowdown occurs if the symbol (cudafree)is searched through the petsc so, but does not occur if the symbol is found directly in the cuda runtime lib.

So the issue has nothing to do with the dynamic linker. The following example can be used to easily reproduce the problem (cudaFree(0) always takes ~7.5 seconds).

1) This should go to OLCF admin as Jeff suggests

I had sent this to OLCF admin before the discussion was started here. Thomas Papatheodore has followed up. I am trying to help him reproduce the problem on summit.


2) Just to make sure I understand, a static executable with this code is still slow on the cudaFree(), since CUDA is a shared library by default.

I prepared the code as a minimal example to reproduce the problem. It would be fair to say any code using PETSc (with CUDA enabled, built statically or dynamically) on summit suffers a 7.5-second overhead on the first CUDA function call (either in the user code or inside PETSc).

Thanks,
Hong


I think we should try:

  a) Forcing a full static link, if possible

  b) Asking OLCF about link resolution order

It sounds like a similar thing I have seen in the past where link resolution order can exponentially increase load time.

  Thanks,

     Matt

bash-4.2$ cat ex_simple_petsc.c
#include <time.h>
#include <cuda_runtime.h>
#include <stdio.h>
#include <petscmat.h>

int main(int argc,char **args)
{
  clock_t start,s1,s2,s3;
  double  cputime;
  double  *init,tmp[100] = {0};
  PetscErrorCode ierr=0;

  ierr = PetscInitialize(&argc,&args,(char*)0,NULL);if (ierr) return ierr;
  start = clock();
  cudaFree(0);
  s1 = clock();
  cudaMalloc((void **)&init,100*sizeof(double));
  s2 = clock();
  cudaMemcpy(init,tmp,100*sizeof(double),cudaMemcpyHostToDevice);
  s3 = clock();
  printf("free time =%lf malloc time =%lf copy time =%lf\n",((double) (s1 - start)) / CLOCKS_PER_SEC,((double) (s2 - s1)) / CLOCKS_PER_SEC,((double) (s3 - s2)) / CLOCKS_PER_SEC);
  ierr = PetscFinalize();
  return ierr;
}

Hong

On Feb 7, 2020, at 3:09 PM, Zhang, Hong <hongzhang at anl.gov<mailto:hongzhang at anl.gov>> wrote:

Note that the overhead was triggered by the first call to a CUDA function. So it seems that the first CUDA function triggered loading petsc so (if petsc so is linked), which is slow on the summit file system.

Hong

On Feb 7, 2020, at 2:54 PM, Zhang, Hong via petsc-dev <petsc-dev at mcs.anl.gov<mailto:petsc-dev at mcs.anl.gov>> wrote:

Linking any other shared library does not slow down the execution. The PETSc shared library is the only one causing trouble.

Here are the ldd output for two different versions. For the first version, I removed -lpetsc and it ran very fast. The second (slow) version was linked to petsc so.

bash-4.2$ ldd ex_simple
        linux-vdso64.so.1 =>  (0x0000200000050000)
        liblapack.so.0 => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/liblapack.so.0 (0x0000200000070000)
        libblas.so.0 => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libblas.so.0 (0x00002000009b0000)
        libhdf5hl_fortran.so.100 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/hdf5-1.10.3-pgiul2yf4auv7krecd72t6vupd7e3qgn/lib/libhdf5hl_fortran.so.100 (0x0000200000e80000)
        libhdf5_fortran.so.100 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/hdf5-1.10.3-pgiul2yf4auv7krecd72t6vupd7e3qgn/lib/libhdf5_fortran.so.100 (0x0000200000ed0000)
        libhdf5_hl.so.100 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/hdf5-1.10.3-pgiul2yf4auv7krecd72t6vupd7e3qgn/lib/libhdf5_hl.so.100 (0x0000200000f50000)
        libhdf5.so.103 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/hdf5-1.10.3-pgiul2yf4auv7krecd72t6vupd7e3qgn/lib/libhdf5.so.103 (0x0000200000fb0000)
        libX11.so.6 => /usr/lib64/libX11.so.6 (0x00002000015e0000)
        libcufft.so.10 => /sw/summit/cuda/10.1.168/lib64/libcufft.so.10 (0x0000200001770000)
        libcublas.so.10 => /sw/summit/cuda/10.1.168/lib64/libcublas.so.10 (0x0000200009b00000)
        libcudart.so.10.1 => /sw/summit/cuda/10.1.168/lib64/libcudart.so.10.1 (0x000020000d950000)
        libcusparse.so.10 => /sw/summit/cuda/10.1.168/lib64/libcusparse.so.10 (0x000020000d9f0000)
        libcusolver.so.10 => /sw/summit/cuda/10.1.168/lib64/libcusolver.so.10 (0x0000200012f50000)
        libstdc++.so.6 => /usr/lib64/libstdc++.so.6 (0x000020001dc40000)
        libdl.so.2 => /usr/lib64/libdl.so.2 (0x000020001ddd0000)
        libpthread.so.0 => /usr/lib64/libpthread.so.0 (0x000020001de00000)
        libmpiprofilesupport.so.3 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/spectrum-mpi-10.3.0.1-20190611-4ymaahbai7ehhw4rves5jjiwon2laz3a/lib/libmpiprofilesupport.so.3 (0x000020001de40000)
        libmpi_ibm_usempi.so => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/spectrum-mpi-10.3.0.1-20190611-4ymaahbai7ehhw4rves5jjiwon2laz3a/lib/libmpi_ibm_usempi.so (0x000020001de70000)
        libmpi_ibm_mpifh.so.3 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/spectrum-mpi-10.3.0.1-20190611-4ymaahbai7ehhw4rves5jjiwon2laz3a/lib/libmpi_ibm_mpifh.so.3 (0x000020001dea0000)
        libmpi_ibm.so.3 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/spectrum-mpi-10.3.0.1-20190611-4ymaahbai7ehhw4rves5jjiwon2laz3a/lib/libmpi_ibm.so.3 (0x000020001df40000)
        libpgf90rtl.so => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libpgf90rtl.so (0x000020001e0b0000)
        libpgf90.so => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libpgf90.so (0x000020001e0f0000)
        libpgf90_rpm1.so => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libpgf90_rpm1.so (0x000020001e6a0000)
        libpgf902.so => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libpgf902.so (0x000020001e6d0000)
        libpgftnrtl.so => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libpgftnrtl.so (0x000020001e700000)
        libatomic.so.1 => /usr/lib64/libatomic.so.1 (0x000020001e730000)
        libpgkomp.so => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libpgkomp.so (0x000020001e760000)
        libomp.so => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libomp.so (0x000020001e790000)
        libomptarget.so => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libomptarget.so (0x000020001e880000)
        libpgmath.so => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libpgmath.so (0x000020001e8b0000)
        libpgc.so => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libpgc.so (0x000020001e9d0000)
        librt.so.1 => /usr/lib64/librt.so.1 (0x000020001eb40000)
        libm.so.6 => /usr/lib64/libm.so.6 (0x000020001eb70000)
        libgcc_s.so.1 => /usr/lib64/libgcc_s.so.1 (0x000020001ec60000)
        libc.so.6 => /usr/lib64/libc.so.6 (0x000020001eca0000)
        libz.so.1 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/zlib-1.2.11-2htm7ws4hgrthi5tyjnqxtjxgpfklxsc/lib/libz.so.1 (0x000020001ee90000)
        libxcb.so.1 => /usr/lib64/libxcb.so.1 (0x000020001eef0000)
        /lib64/ld64.so.2 (0x0000200000000000)
        libcublasLt.so.10 => /sw/summit/cuda/10.1.168/lib64/libcublasLt.so.10 (0x000020001ef40000)
        libutil.so.1 => /usr/lib64/libutil.so.1 (0x0000200020e50000)
        libhwloc_ompi.so.15 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/spectrum-mpi-10.3.0.1-20190611-4ymaahbai7ehhw4rves5jjiwon2laz3a/lib/libhwloc_ompi.so.15 (0x0000200020e80000)
        libevent-2.1.so.6 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/spectrum-mpi-10.3.0.1-20190611-4ymaahbai7ehhw4rves5jjiwon2laz3a/lib/libevent-2.1.so.6 (0x0000200020ef0000)
        libevent_pthreads-2.1.so.6 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/spectrum-mpi-10.3.0.1-20190611-4ymaahbai7ehhw4rves5jjiwon2laz3a/lib/libevent_pthreads-2.1.so.6 (0x0000200020f70000)
        libopen-rte.so.3 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/spectrum-mpi-10.3.0.1-20190611-4ymaahbai7ehhw4rves5jjiwon2laz3a/lib/libopen-rte.so.3 (0x0000200020fa0000)
        libopen-pal.so.3 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/spectrum-mpi-10.3.0.1-20190611-4ymaahbai7ehhw4rves5jjiwon2laz3a/lib/libopen-pal.so.3 (0x00002000210b0000)
        libXau.so.6 => /usr/lib64/libXau.so.6 (0x00002000211a0000)


bash-4.2$ ldd ex_simple_slow
        linux-vdso64.so.1 =>  (0x0000200000050000)
        libpetsc.so.3.012 => /autofs/nccs-svm1_home1/hongzh/Projects/petsc/arch-olcf-summit-sell-opt/lib/libpetsc.so.3.012 (0x0000200000070000)
        liblapack.so.0 => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/liblapack.so.0 (0x0000200002be0000)
        libblas.so.0 => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libblas.so.0 (0x0000200003520000)
        libhdf5hl_fortran.so.100 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/hdf5-1.10.3-pgiul2yf4auv7krecd72t6vupd7e3qgn/lib/libhdf5hl_fortran.so.100 (0x00002000039f0000)
        libhdf5_fortran.so.100 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/hdf5-1.10.3-pgiul2yf4auv7krecd72t6vupd7e3qgn/lib/libhdf5_fortran.so.100 (0x0000200003a40000)
        libhdf5_hl.so.100 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/hdf5-1.10.3-pgiul2yf4auv7krecd72t6vupd7e3qgn/lib/libhdf5_hl.so.100 (0x0000200003ac0000)
        libhdf5.so.103 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/hdf5-1.10.3-pgiul2yf4auv7krecd72t6vupd7e3qgn/lib/libhdf5.so.103 (0x0000200003b20000)
        libX11.so.6 => /usr/lib64/libX11.so.6 (0x0000200004150000)
        libcufft.so.10 => /sw/summit/cuda/10.1.168/lib64/libcufft.so.10 (0x00002000042e0000)
        libcublas.so.10 => /sw/summit/cuda/10.1.168/lib64/libcublas.so.10 (0x000020000c670000)
        libcudart.so.10.1 => /sw/summit/cuda/10.1.168/lib64/libcudart.so.10.1 (0x00002000104c0000)
        libcusparse.so.10 => /sw/summit/cuda/10.1.168/lib64/libcusparse.so.10 (0x0000200010560000)
        libcusolver.so.10 => /sw/summit/cuda/10.1.168/lib64/libcusolver.so.10 (0x0000200015ac0000)
        libstdc++.so.6 => /usr/lib64/libstdc++.so.6 (0x00002000207b0000)
        libdl.so.2 => /usr/lib64/libdl.so.2 (0x0000200020940000)
        libpthread.so.0 => /usr/lib64/libpthread.so.0 (0x0000200020970000)
        libmpiprofilesupport.so.3 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/spectrum-mpi-10.3.0.1-20190611-4ymaahbai7ehhw4rves5jjiwon2laz3a/lib/libmpiprofilesupport.so.3 (0x00002000209b0000)
        libmpi_ibm_usempi.so => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/spectrum-mpi-10.3.0.1-20190611-4ymaahbai7ehhw4rves5jjiwon2laz3a/lib/libmpi_ibm_usempi.so (0x00002000209e0000)
        libmpi_ibm_mpifh.so.3 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/spectrum-mpi-10.3.0.1-20190611-4ymaahbai7ehhw4rves5jjiwon2laz3a/lib/libmpi_ibm_mpifh.so.3 (0x0000200020a10000)
        libmpi_ibm.so.3 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/spectrum-mpi-10.3.0.1-20190611-4ymaahbai7ehhw4rves5jjiwon2laz3a/lib/libmpi_ibm.so.3 (0x0000200020ab0000)
        libpgf90rtl.so => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libpgf90rtl.so (0x0000200020c20000)
        libpgf90.so => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libpgf90.so (0x0000200020c60000)
        libpgf90_rpm1.so => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libpgf90_rpm1.so (0x0000200021210000)
        libpgf902.so => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libpgf902.so (0x0000200021240000)
        libpgftnrtl.so => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libpgftnrtl.so (0x0000200021270000)
        libatomic.so.1 => /usr/lib64/libatomic.so.1 (0x00002000212a0000)
        libpgkomp.so => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libpgkomp.so (0x00002000212d0000)
        libomp.so => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libomp.so (0x0000200021300000)
        libomptarget.so => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libomptarget.so (0x00002000213f0000)
        libpgmath.so => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libpgmath.so (0x0000200021420000)
        libpgc.so => /autofs/nccs-svm1_sw/summit/.swci/0-core/opt/spack/20180914/linux-rhel7-ppc64le/gcc-4.8.5/pgi-19.4-6acz4xyqjlpoaonjiiqjme2aknrfnzoy/linuxpower/19.4/lib/libpgc.so (0x0000200021540000)
        librt.so.1 => /usr/lib64/librt.so.1 (0x00002000216b0000)
        libm.so.6 => /usr/lib64/libm.so.6 (0x00002000216e0000)
        libgcc_s.so.1 => /usr/lib64/libgcc_s.so.1 (0x00002000217d0000)
        libc.so.6 => /usr/lib64/libc.so.6 (0x0000200021810000)
        libz.so.1 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/zlib-1.2.11-2htm7ws4hgrthi5tyjnqxtjxgpfklxsc/lib/libz.so.1 (0x0000200021a10000)
        libxcb.so.1 => /usr/lib64/libxcb.so.1 (0x0000200021a60000)
        /lib64/ld64.so.2 (0x0000200000000000)
        libcublasLt.so.10 => /sw/summit/cuda/10.1.168/lib64/libcublasLt.so.10 (0x0000200021ab0000)
        libutil.so.1 => /usr/lib64/libutil.so.1 (0x00002000239c0000)
        libhwloc_ompi.so.15 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/spectrum-mpi-10.3.0.1-20190611-4ymaahbai7ehhw4rves5jjiwon2laz3a/lib/libhwloc_ompi.so.15 (0x00002000239f0000)
        libevent-2.1.so.6 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/spectrum-mpi-10.3.0.1-20190611-4ymaahbai7ehhw4rves5jjiwon2laz3a/lib/libevent-2.1.so.6 (0x0000200023a60000)
        libevent_pthreads-2.1.so.6 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/spectrum-mpi-10.3.0.1-20190611-4ymaahbai7ehhw4rves5jjiwon2laz3a/lib/libevent_pthreads-2.1.so.6 (0x0000200023ae0000)
        libopen-rte.so.3 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/spectrum-mpi-10.3.0.1-20190611-4ymaahbai7ehhw4rves5jjiwon2laz3a/lib/libopen-rte.so.3 (0x0000200023b10000)
        libopen-pal.so.3 => /autofs/nccs-svm1_sw/summit/.swci/1-compute/opt/spack/20180914/linux-rhel7-ppc64le/pgi-19.4/spectrum-mpi-10.3.0.1-20190611-4ymaahbai7ehhw4rves5jjiwon2laz3a/lib/libopen-pal.so.3 (0x0000200023c20000)
        libXau.so.6 => /usr/lib64/libXau.so.6 (0x0000200023d10000)


On Feb 7, 2020, at 2:31 PM, Smith, Barry F. <bsmith at mcs.anl.gov<mailto:bsmith at mcs.anl.gov>> wrote:


 ldd -o on the executable of both linkings of your code.

 My guess is that without PETSc it is linking the static version of the needed libraries and with PETSc the shared. And, in typical fashion, the shared libraries are off on some super slow file system so take a long time to be loaded and linked in on demand.

  Still a performance bug in Summit.

  Barry


On Feb 7, 2020, at 12:23 PM, Zhang, Hong via petsc-dev <petsc-dev at mcs.anl.gov<mailto:petsc-dev at mcs.anl.gov>> wrote:

Hi all,

Previously I have noticed that the first call to a CUDA function such as cudaMalloc and cudaFree in PETSc takes a long time (7.5 seconds) on summit. Then I prepared a simple example as attached to help OCLF reproduce the problem. It turned out that the problem was  caused by PETSc. The 7.5-second overhead can be observed only when the PETSc lib is linked. If I do not link PETSc, it runs normally. Does anyone have any idea why this happens and how to fix it?

Hong (Mr.)

bash-4.2$ cat ex_simple.c
#include <time.h>
#include <cuda_runtime.h>
#include <stdio.h>

int main(int argc,char **args)
{
clock_t start,s1,s2,s3;
double  cputime;
double   *init,tmp[100] = {0};

start = clock();
cudaFree(0);
s1 = clock();
cudaMalloc((void **)&init,100*sizeof(double));
s2 = clock();
cudaMemcpy(init,tmp,100*sizeof(double),cudaMemcpyHostToDevice);
s3 = clock();
printf("free time =%lf malloc time =%lf copy time =%lf\n",((double) (s1 - start)) / CLOCKS_PER_SEC,((double) (s2 - s1)) / CLOCKS_PER_SEC,((double) (s3 - s2)) / CLOCKS_PER_SEC);

return 0;
}








--
What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.
-- Norbert Wiener

https://www.cse.buffalo.edu/~knepley/<http://www.cse.buffalo.edu/~knepley/>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20200208/e91ee103/attachment-0001.html>


More information about the petsc-dev mailing list