tu-c0r1n00 - 0 device=0 binding=--interleave=0,1 tu-c0r1n00 - 1 device=1 binding=--interleave=2,3 tu-c0r1n00 - 2 device=2 binding=--interleave=4,5 tu-c0r1n00 - 3 device=3 binding=--interleave=6,7 tu-c0r2n06 - 0 device=0 binding=--interleave=0,1 tu-c0r2n06 - 2 device=2 binding=--interleave=4,5 tu-c0r2n06 - 3 device=3 binding=--interleave=6,7 tu-c0r2n06 - 1 device=1 binding=--interleave=2,3 tu-c0r2n09 - 1 device=1 binding=--interleave=2,3 tu-c0r2n09 - 0 device=0 binding=--interleave=0,1 tu-c0r2n09 - 3 device=3 binding=--interleave=6,7 tu-c0r2n09 - 2 device=2 binding=--interleave=4,5 tu-c0r1n21 - 0 device=0 binding=--interleave=0,1 tu-c0r1n21 - 1 device=1 binding=--interleave=2,3 tu-c0r1n21 - 2 device=2 binding=--interleave=4,5 tu-c0r1n21 - 3 device=3 binding=--interleave=6,7 tu-c0r2n15 - 1 device=1 binding=--interleave=2,3 tu-c0r2n15 - 0 device=0 binding=--interleave=0,1 tu-c0r2n15 - 3 device=3 binding=--interleave=6,7 tu-c0r2n15 - 2 device=2 binding=--interleave=4,5 tu-c0r1n09 - 2 device=2 binding=--interleave=4,5 tu-c0r1n09 - 1 device=1 binding=--interleave=2,3 tu-c0r1n09 - 3 device=3 binding=--interleave=6,7 tu-c0r1n09 - 0 device=0 binding=--interleave=0,1 tu-c0r2n21 - 2 device=2 binding=--interleave=4,5 tu-c0r2n21 - 1 device=1 binding=--interleave=2,3 tu-c0r2n21 - 0 device=0 binding=--interleave=0,1 tu-c0r2n21 - 3 device=3 binding=--interleave=6,7 tu-c0r1n12 - 0 device=0 binding=--interleave=0,1 tu-c0r1n12 - 3 device=3 binding=--interleave=6,7 tu-c0r1n12 - 1 device=1 binding=--interleave=2,3 tu-c0r1n12 - 2 device=2 binding=--interleave=4,5 tu-c0r2n12 - 0 device=0 binding=--interleave=0,1 tu-c0r2n12 - 1 device=1 binding=--interleave=2,3 tu-c0r2n12 - 3 device=3 binding=--interleave=6,7 tu-c0r2n12 - 2 device=2 binding=--interleave=4,5 tu-c0r1n18 - 3 device=3 binding=--interleave=6,7 tu-c0r1n18 - 1 device=1 binding=--interleave=2,3 tu-c0r1n06 - 1 device=1 binding=--interleave=2,3 tu-c0r1n06 - 0 device=0 binding=--interleave=0,1 tu-c0r1n18 - 0 device=0 binding=--interleave=0,1 tu-c0r1n18 - 2 device=2 binding=--interleave=4,5 tu-c0r1n06 - 2 device=2 binding=--interleave=4,5 tu-c0r1n06 - 3 device=3 binding=--interleave=6,7 tu-c0r2n00 - 2 device=2 binding=--interleave=4,5 tu-c0r2n00 - 1 device=1 binding=--interleave=2,3 tu-c0r2n00 - 3 device=3 binding=--interleave=6,7 tu-c0r2n03 - 3 device=3 binding=--interleave=6,7 tu-c0r2n03 - 0 device=0 binding=--interleave=0,1 tu-c0r2n00 - 0 device=0 binding=--interleave=0,1 tu-c0r2n03 - 2 device=2 binding=--interleave=4,5 tu-c0r1n03 - 1 device=1 binding=--interleave=2,3 tu-c0r1n03 - 3 device=3 binding=--interleave=6,7 tu-c0r1n03 - 0 device=0 binding=--interleave=0,1 tu-c0r1n03 - 2 device=2 binding=--interleave=4,5 tu-c0r2n03 - 1 device=1 binding=--interleave=2,3 tu-c0r1n15 - 2 device=2 binding=--interleave=4,5 tu-c0r1n15 - 1 device=1 binding=--interleave=2,3 tu-c0r1n15 - 3 device=3 binding=--interleave=6,7 tu-c0r1n15 - 0 device=0 binding=--interleave=0,1 tu-c0r2n18 - 2 device=2 binding=--interleave=4,5 tu-c0r2n18 - 0 device=0 binding=--interleave=0,1 tu-c0r2n18 - 1 device=1 binding=--interleave=2,3 tu-c0r2n18 - 3 device=3 binding=--interleave=6,7 OPENMPI detected AcceleratorCudaInit: using default device AcceleratorCudaInit: assume user either uses AcceleratorCudaInit: a) IBM jsrun, or AcceleratorCudaInit: b) invokes through a wrapping script to set CUDA_VISIBLE_DEVICES, UCX_NET_DEVICES, and numa binding AcceleratorCudaInit: Configure options --enable-setdevice=no OPENMPI detected AcceleratorCudaInit[0]: ======================== AcceleratorCudaInit[0]: Device Number : 0 AcceleratorCudaInit[0]: ======================== AcceleratorCudaInit[0]: Device identifier: NVIDIA A100-SXM4-40GB AcceleratorCudaInit[0]: totalGlobalMem: 42505273344 AcceleratorCudaInit[0]: managedMemory: 1 AcceleratorCudaInit[0]: isMultiGpuBoard: 0 AcceleratorCudaInit[0]: warpSize: 32 AcceleratorCudaInit[0]: pciBusID: 3 AcceleratorCudaInit[0]: pciDeviceID: 0 AcceleratorCudaInit[0]: maxGridSize (2147483647,65535,65535) AcceleratorCudaInit: using default device AcceleratorCudaInit: assume user either uses AcceleratorCudaInit: a) IBM jsrun, or AcceleratorCudaInit: b) invokes through a wrapping script to set CUDA_VISIBLE_DEVICES, UCX_NET_DEVICES, and numa binding AcceleratorCudaInit: Configure options --enable-setdevice=no OPENMPI detected AcceleratorCudaInit: using default device AcceleratorCudaInit: assume user either uses AcceleratorCudaInit: a) IBM jsrun, or AcceleratorCudaInit: b) invokes through a wrapping script to set CUDA_VISIBLE_DEVICES, UCX_NET_DEVICES, and numa binding AcceleratorCudaInit: Configure options --enable-setdevice=no OPENMPI detected AcceleratorCudaInit: using default device AcceleratorCudaInit: assume user either uses AcceleratorCudaInit: a) IBM jsrun, or AcceleratorCudaInit: b) invokes through a wrapping script to set CUDA_VISIBLE_DEVICES, UCX_NET_DEVICES, and numa binding AcceleratorCudaInit: Configure options --enable-setdevice=no OPENMPI detected AcceleratorCudaInit[0]: ======================== AcceleratorCudaInit[0]: Device Number : 0 AcceleratorCudaInit[0]: ======================== AcceleratorCudaInit[0]: Device identifier: NVIDIA A100-SXM4-40GB AcceleratorCudaInit[0]: totalGlobalMem: 42505273344 AcceleratorCudaInit[0]: managedMemory: 1 AcceleratorCudaInit[0]: isMultiGpuBoard: 0 AcceleratorCudaInit[0]: warpSize: 32 AcceleratorCudaInit[0]: pciBusID: 3 AcceleratorCudaInit[0]: pciDeviceID: 0 AcceleratorCudaInit[0]: maxGridSize (2147483647,65535,65535) AcceleratorCudaInit: using default device AcceleratorCudaInit: assume user either uses AcceleratorCudaInit: a) IBM jsrun, or AcceleratorCudaInit: b) invokes through a wrapping script to set CUDA_VISIBLE_DEVICES, UCX_NET_DEVICES, and numa binding AcceleratorCudaInit: Configure options --enable-setdevice=no OPENMPI detected AcceleratorCudaInit: using default device AcceleratorCudaInit: assume user either uses AcceleratorCudaInit: a) IBM jsrun, or AcceleratorCudaInit: b) invokes through a wrapping script to set CUDA_VISIBLE_DEVICES, UCX_NET_DEVICES, and numa binding AcceleratorCudaInit: Configure options --enable-setdevice=no OPENMPI detected AcceleratorCudaInit: using default device AcceleratorCudaInit: assume user either uses AcceleratorCudaInit: a) IBM jsrun, or AcceleratorCudaInit: b) invokes through a wrapping script to set CUDA_VISIBLE_DEVICES, UCX_NET_DEVICES, and numa binding AcceleratorCudaInit: Configure options --enable-setdevice=no local rank 0 device 0 bus id: 0000:03:00.0 AcceleratorCudaInit: ================================================ AcceleratorCudaInit: ================================================ AcceleratorCudaInit: ================================================ AcceleratorCudaInit: ================================================ OPENMPI detected AcceleratorCudaInit: using default device AcceleratorCudaInit: assume user either uses AcceleratorCudaInit: a) IBM jsrun, or AcceleratorCudaInit: b) invokes through a wrapping script to set CUDA_VISIBLE_DEVICES, UCX_NET_DEVICES, and numa binding AcceleratorCudaInit: Configure options --enable-setdevice=no local rank 2 device 0 bus id: 0000:84:00.0 local rank 0 device 0 bus id: 0000:03:00.0 AcceleratorCudaInit: ================================================ AcceleratorCudaInit: ================================================ AcceleratorCudaInit: ================================================ AcceleratorCudaInit: ================================================ local rank 1 device 0 bus id: 0000:44:00.0 local rank 3 device 0 bus id: 0000:C4:00.0 SharedMemoryMpi: World communicator of size 64 SharedMemoryMpi: Node communicator of size 4 0SharedMemoryMpi: SharedMemoryMPI.cc acceleratorAllocDevice 2147483648bytes at 0x14c200000000 for comms buffers Setting up IPC __|__|__|__|__|__|__|__|__|__|__|__|__|__|__ __|__|__|__|__|__|__|__|__|__|__|__|__|__|__ __|_ | | | | | | | | | | | | _|__ __|_ _|__ __|_ GGGG RRRR III DDDD _|__ __|_ G R R I D D _|__ __|_ G R R I D D _|__ __|_ G GG RRRR I D D _|__ __|_ G G R R I D D _|__ __|_ GGGG R R III DDDD _|__ __|_ _|__ __|__|__|__|__|__|__|__|__|__|__|__|__|__|__ __|__|__|__|__|__|__|__|__|__|__|__|__|__|__ | | | | | | | | | | | | | | Copyright (C) 2015 Peter Boyle, Azusa Yamaguchi, Guido Cossu, Antonin Portelli and other authors This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. Current Grid git commit hash=188d2c7a4dc77807b545f5f2813cdb589b9e44ca: (HEAD -> develop, gh/develop, gh/HEAD) uncommited changes Grid : Message : ================================================ Grid : Message : MPI is initialised and logging filters activated Grid : Message : ================================================ Grid : Message : Requested 2147483648 byte stencil comms buffers Grid : Message : MemoryManager Cache 34004218675 bytes Grid : Message : MemoryManager::Init() setting up Grid : Message : MemoryManager::Init() cache pool for recent allocations: SMALL 8 LARGE 2 Grid : Message : MemoryManager::Init() Non unified: Caching accelerator data in dedicated memory Grid : Message : MemoryManager::Init() Using cudaMalloc Grid : Message : 1.644920 s : Grid Layout Grid : Message : 1.644924 s : Global lattice size : 64 64 64 256 Grid : Message : 1.644931 s : OpenMP threads : 4 Grid : Message : 1.644934 s : MPI tasks : 2 2 2 8 Grid : Message : 1.686165 s : Making s innermost grids Grid : Message : 1.749166 s : Initialising 4d RNG Grid : Message : 1.841034 s : Intialising parallel RNG with unique string 'The 4D RNG' Grid : Message : 1.841066 s : Seed SHA256: 49db4542db694e3b1a74bf2592a8c1b83bfebbe18401693c2609a4c3af1 Grid : Message : 3.424395 s : Initialising 5d RNG Grid : Message : 4.892259 s : Intialising parallel RNG with unique string 'The 5D RNG' Grid : Message : 4.892305 s : Seed SHA256: b6316f2fac44ce14111f93e0296389330b077bfd0a7b359f781c58589f8a Grid : Message : 38.140451 s : Initialised RNGs Grid : Message : 43.525986 s : Drawing gauge field Grid : Message : 44.448636 s : Random gauge initialised Grid : Message : 44.460922 s : Setting up Cshift based reference Grid : Message : 73.407183 s : ***************************************************************** Grid : Message : 73.407213 s : * Kernel options --dslash-generic, --dslash-unroll, --dslash-asm Grid : Message : 73.407215 s : ***************************************************************** Grid : Message : 73.407216 s : ***************************************************************** Grid : Message : 73.407217 s : * Benchmarking DomainWallFermionR::Dhop Grid : Message : 73.407218 s : * Vectorising space-time by 8 Grid : Message : 73.407219 s : * VComplexF size is 64 B Grid : Message : 73.407220 s : * SINGLE precision Grid : Message : 73.407223 s : * Using Overlapped Comms/Compute Grid : Message : 73.407224 s : * Using GENERIC Nc WilsonKernels Grid : Message : 73.407225 s : ***************************************************************** Grid : Message : 75.563449 s : Called warmup Grid : Message : 379.575070 s : Called Dw 30000 times in 3.04011e+08 us Grid : Message : 379.575131 s : mflop/s = 1.39864e+08 Grid : Message : 379.575133 s : mflop/s per rank = 2.18537e+06 Grid : Message : 379.575135 s : mflop/s per node = 8.74148e+06 Grid : Message : 379.575137 s : RF GiB/s (base 2) = 284200 Grid : Message : 379.575139 s : mem GiB/s (base 2) = 177625 Grid : Message : 379.578632 s : norm diff 1.06407e-13 Grid : Message : 379.631872 s : #### Dhop calls report Grid : Message : 379.631879 s : WilsonFermion5D Number of DhopEO Calls : 60002 Grid : Message : 379.631882 s : WilsonFermion5D TotalTime /Calls : 5070.92 us Grid : Message : 379.631884 s : WilsonFermion5D CommTime /Calls : 3546.45 us Grid : Message : 379.631886 s : WilsonFermion5D FaceTime /Calls : 535.07 us Grid : Message : 379.631888 s : WilsonFermion5D ComputeTime1/Calls : 4.5944 us Grid : Message : 379.631890 s : WilsonFermion5D ComputeTime2/Calls : 1003.42 us Grid : Message : 379.631978 s : Average mflops/s per call : 1.24725e+11 Grid : Message : 379.631982 s : Average mflops/s per call per rank : 1.94882e+09 Grid : Message : 379.631984 s : Average mflops/s per call per node : 7.7953e+09 Grid : Message : 379.631986 s : Average mflops/s per call (full) : 1.42293e+08 Grid : Message : 379.631988 s : Average mflops/s per call per rank (full): 2.22332e+06 Grid : Message : 379.631990 s : Average mflops/s per call per node (full): 8.89328e+06 Grid : Message : 379.631992 s : WilsonFermion5D Stencil Grid : Message : 379.631993 s : WilsonFermion5D StencilEven Grid : Message : 379.631994 s : WilsonFermion5D StencilOdd Grid : Message : 379.631995 s : WilsonFermion5D Stencil Reporti() Grid : Message : 379.631996 s : WilsonFermion5D StencilEven Reporti() Grid : Message : 379.631997 s : WilsonFermion5D StencilOdd Reporti() Grid : Message : 435.228602 s : Compare to naive wilson implementation Dag to verify correctness Grid : Message : 435.228625 s : Called DwDag Grid : Message : 435.228626 s : norm dag result 12.0421 Grid : Message : 435.243914 s : norm dag ref 12.0421 Grid : Message : 435.261408 s : norm dag diff 7.21924e-14 Grid : Message : 435.313437 s : Calling Deo and Doe and //assert Deo+Doe == Dunprec Grid : Message : 435.660767 s : src_e0.499998 Grid : Message : 436.112111 s : src_o0.500002 Grid : Message : 436.258880 s : ********************************************************* Grid : Message : 436.258883 s : * Benchmarking DomainWallFermionF::DhopEO Grid : Message : 436.258884 s : * Vectorising space-time by 8 Grid : Message : 436.258885 s : * SINGLE precision Grid : Message : 436.258886 s : * Using Overlapped Comms/Compute Grid : Message : 436.258887 s : * Using GENERIC Nc WilsonKernels Grid : Message : 436.258888 s : ********************************************************* Grid : Message : 592.381302 s : Deo mflop/s = 1.36248e+08 Grid : Message : 592.381328 s : Deo mflop/s per rank 2.12887e+06 Grid : Message : 592.381330 s : Deo mflop/s per node 8.51549e+06 Grid : Message : 592.381333 s : #### Dhop calls report Grid : Message : 592.381335 s : WilsonFermion5D Number of DhopEO Calls : 30001 Grid : Message : 592.381337 s : WilsonFermion5D TotalTime /Calls : 5203.74 us Grid : Message : 592.381339 s : WilsonFermion5D CommTime /Calls : 3583.01 us Grid : Message : 592.381341 s : WilsonFermion5D FaceTime /Calls : 696.099 us Grid : Message : 592.381343 s : WilsonFermion5D ComputeTime1/Calls : 5.72691 us Grid : Message : 592.381345 s : WilsonFermion5D ComputeTime2/Calls : 950.432 us Grid : Message : 592.381374 s : Average mflops/s per call : 1.03144e+11 Grid : Message : 592.381379 s : Average mflops/s per call per rank : 1.61163e+09 Grid : Message : 592.381381 s : Average mflops/s per call per node : 6.44653e+09 Grid : Message : 592.381384 s : Average mflops/s per call (full) : 1.38661e+08 Grid : Message : 592.381386 s : Average mflops/s per call per rank (full): 2.16657e+06 Grid : Message : 592.381389 s : Average mflops/s per call per node (full): 8.66629e+06 Grid : Message : 592.381391 s : WilsonFermion5D Stencil Grid : Message : 592.381393 s : WilsonFermion5D StencilEven Grid : Message : 592.381395 s : WilsonFermion5D StencilOdd Grid : Message : 592.381396 s : WilsonFermion5D Stencil Reporti() Grid : Message : 592.381398 s : WilsonFermion5D StencilEven Reporti() Grid : Message : 592.381400 s : WilsonFermion5D StencilOdd Reporti() Grid : Message : 592.456431 s : r_e6.02108 Grid : Message : 592.464000 s : r_o6.02101 Grid : Message : 592.471236 s : res12.0421 Grid : Message : 593.988570 s : norm diff 0 Grid : Message : 593.966561 s : norm diff even 0 Grid : Message : 594.396991 s : norm diff odd 0