I_MPI_PIN_PROCESSOR_LIST=[<procset>][:[grain=<grain>][,shift=<shift>][,preoffset=<preoffset>][,postoffset=<postoffset>] If grain is not being specified, then the syntax should be, I_MPI_PIN_PROCESSOR_LIST=allcores:,shift=3,preoffset=2 However, with both variants, ...
On Lomonosov-2 supercomputer (http://hpc.msu.ru/node/159, partition "Pascal"), I observe strange slowdowns in processing with I_MPI_ASYNC_PROGRESS=1/I_MPI_PIN_PROCESSOR_LIST=... parameters combination. The reproducer code is like: ---int main(int argc, char...
I_MPI_PIN_PROCESSOR_LIST=[<procset>][:[grain=<grain>][,shift=<shift>][,preoffset=<preoffset>][,postoffset=<postoffset>] If grain is not being specified, then the syntax should be, I_MPI_PIN_PROCESSOR_LIST=allcores:,shift=3,preoffset=2 However, with both variants, ...
I_MPI_PIN_PROCESSOR_LIST=[<procset>][:[grain=<grain>][,shift=<shift>][,preoffset=<preoffset>][,postoffset=<postoffset>] If grain is not being specified, then the syntax should be, I_MPI_PIN_PROCESSOR_LIST=allcores:,shift=3,preoffset=2 However, with both variants, t...
On Lomonosov-2 supercomputer (http://hpc.msu.ru/node/159, partition "Pascal"), I observe strange slowdowns in processing with I_MPI_ASYNC_PROGRESS=1/I_MPI_PIN_PROCESSOR_LIST=... parameters combination. The reproducer code is like: ---int main(int argc, char **argv){...
On Lomonosov-2 supercomputer (http://hpc.msu.ru/node/159, partition "Pascal"), I observe strange slowdowns in processing with I_MPI_ASYNC_PROGRESS=1/I_MPI_PIN_PROCESSOR_LIST=... parameters combination. The reproducer code is like: ---int main(int argc, char **argv){...
I've got timing results on stdout from 3 consecutive executions like: 10.2775; 10.1705; 10.243 If I add explicit pinning like this: # export I_MPI_ASYNC_PROGRESS=1# export I_MPI_PIN_PROCESSOR_LIST=0,1,2,3# mpiexec.hydra -f hosts -np 4 -ppn 2 --errfile-pattern=er...