portelli/Grid
1
0
Fork 0
mirror of https://github.com/paboyle/Grid.git synced 2025-06-18 15:57:05 +01:00
Code Releases Activity

Merge branch 'develop' into feature/hadrons

Browse Source
This commit is contained in:
Antonin Portelli
2017-08-29 11:03:53 -06:00
parent 5846566728 4b4c2a715b
commit a56e3b40c4
20 changed files with 1174 additions and 184 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
lib/Makefile.am
View File

@ -10,8 +10,8 @@ if BUILD_COMMS_MPI3
extra_sources+=communicator/Communicator_base.cc
endif
if BUILD_COMMS_MPI3L
extra_sources+=communicator/Communicator_mpi3_leader.cc
if BUILD_COMMS_MPIT
extra_sources+=communicator/Communicator_mpit.cc
extra_sources+=communicator/Communicator_base.cc
endif

37
lib/allocator/AlignedAllocator.cc
View File

@ -1,7 +1,5 @@
#include <Grid/GridCore.h>
#include <fcntl.h>
namespace Grid {
@ -63,4 +61,37 @@ void *PointerCache::Lookup(size_t bytes) {
return NULL;
}
void check_huge_pages(void *Buf,uint64_t BYTES)
{
#ifdef __linux__
int fd = open("/proc/self/pagemap", O_RDONLY);
assert(fd >= 0);
const int page_size = 4096;
uint64_t virt_pfn = (uint64_t)Buf / page_size;
off_t offset = sizeof(uint64_t) * virt_pfn;
uint64_t npages = (BYTES + page_size-1) / page_size;
uint64_t pagedata[npages];
uint64_t ret = lseek(fd, offset, SEEK_SET);
assert(ret == offset);
ret = ::read(fd, pagedata, sizeof(uint64_t)*npages);
assert(ret == sizeof(uint64_t) * npages);
int nhugepages = npages / 512;
int n4ktotal, nnothuge;
n4ktotal = 0;
nnothuge = 0;
for (int i = 0; i < nhugepages; ++i) {
uint64_t baseaddr = (pagedata[i*512] & 0x7fffffffffffffULL) * page_size;
for (int j = 0; j < 512; ++j) {
uint64_t pageaddr = (pagedata[i*512+j] & 0x7fffffffffffffULL) * page_size;
++n4ktotal;
if (pageaddr != baseaddr + j * page_size)
++nnothuge;
}
}
int rank = CartesianCommunicator::RankWorld();
printf("rank %d Allocated %d 4k pages, %d not in huge pages\n", rank, n4ktotal, nnothuge);
#endif
}
}

21
lib/allocator/AlignedAllocator.h
View File

@ -64,6 +64,8 @@ namespace Grid {
};
void check_huge_pages(void *Buf,uint64_t BYTES);
////////////////////////////////////////////////////////////////////
// A lattice of something, but assume the something is SIMDized.
////////////////////////////////////////////////////////////////////
@ -92,12 +94,20 @@ public:
size_type bytes = __n*sizeof(_Tp);
_Tp *ptr = (_Tp *) PointerCache::Lookup(bytes);
// if ( ptr != NULL )
// std::cout << "alignedAllocator "<<__n << " cache hit "<< std::hex << ptr <<std::dec <<std::endl;
//////////////////
// Hack 2MB align; could make option probably doesn't need configurability
//////////////////
//define GRID_ALLOC_ALIGN (128)
#define GRID_ALLOC_ALIGN (2*1024*1024)
#ifdef HAVE_MM_MALLOC_H
if ( ptr == (_Tp *) NULL ) ptr = (_Tp *) _mm_malloc(bytes,128);
if ( ptr == (_Tp *) NULL ) ptr = (_Tp *) _mm_malloc(bytes,GRID_ALLOC_ALIGN);
#else
if ( ptr == (_Tp *) NULL ) ptr = (_Tp *) memalign(128,bytes);
if ( ptr == (_Tp *) NULL ) ptr = (_Tp *) memalign(GRID_ALLOC_ALIGN,bytes);
#endif
// std::cout << "alignedAllocator " << std::hex << ptr <<std::dec <<std::endl;
// First touch optimise in threaded loop
uint8_t *cp = (uint8_t *)ptr;
#ifdef GRID_OMP
@ -111,6 +121,7 @@ public:
void deallocate(pointer __p, size_type __n) {
size_type bytes = __n * sizeof(_Tp);
pointer __freeme = (pointer)PointerCache::Insert((void *)__p,bytes);
#ifdef HAVE_MM_MALLOC_H
@ -189,9 +200,9 @@ public:
pointer allocate(size_type __n, const void* _p= 0)
{
#ifdef HAVE_MM_MALLOC_H
_Tp * ptr = (_Tp *) _mm_malloc(__n*sizeof(_Tp),128);
_Tp * ptr = (_Tp *) _mm_malloc(__n*sizeof(_Tp),GRID_ALLOC_ALIGN);
#else
_Tp * ptr = (_Tp *) memalign(128,__n*sizeof(_Tp));
_Tp * ptr = (_Tp *) memalign(GRID_ALLOC_ALIGN,__n*sizeof(_Tp));
#endif
size_type bytes = __n*sizeof(_Tp);
uint8_t *cp = (uint8_t *)ptr;

59
lib/communicator/Communicator_base.cc
View File

@ -26,6 +26,10 @@ Author: Peter Boyle <paboyle@ph.ed.ac.uk>
*************************************************************************************/
/* END LEGAL */
#include <Grid/GridCore.h>
#include <fcntl.h>
#include <unistd.h>
#include <limits.h>
#include <sys/mman.h>
namespace Grid {
@ -34,7 +38,10 @@ namespace Grid {
///////////////////////////////////////////////////////////////
void * CartesianCommunicator::ShmCommBuf;
uint64_t CartesianCommunicator::MAX_MPI_SHM_BYTES = 128*1024*1024;
CartesianCommunicator::CommunicatorPolicy_t CartesianCommunicator::CommunicatorPolicy= CartesianCommunicator::CommunicatorPolicyConcurrent;
CartesianCommunicator::CommunicatorPolicy_t
CartesianCommunicator::CommunicatorPolicy= CartesianCommunicator::CommunicatorPolicyConcurrent;
int CartesianCommunicator::nCommThreads = -1;
int CartesianCommunicator::Hugepages = 0;
/////////////////////////////////
// Alloc, free shmem region
@ -89,25 +96,43 @@ void CartesianCommunicator::GlobalSumVector(ComplexD *c,int N)
GlobalSumVector((double *)c,2*N);
}
#if !defined( GRID_COMMS_MPI3) && !defined (GRID_COMMS_MPI3L)
#if !defined( GRID_COMMS_MPI3)
int CartesianCommunicator::NodeCount(void) { return ProcessorCount();};
int CartesianCommunicator::RankCount(void) { return ProcessorCount();};
double CartesianCommunicator::StencilSendToRecvFromBegin(std::vector<CommsRequest_t> &list,
void *xmit,
int xmit_to_rank,
void *recv,
int recv_from_rank,
int bytes)
#endif
#if !defined( GRID_COMMS_MPI3) && !defined (GRID_COMMS_MPIT)
double CartesianCommunicator::StencilSendToRecvFrom( void *xmit,
int xmit_to_rank,
void *recv,
int recv_from_rank,
int bytes, int dir)
{
std::vector<CommsRequest_t> list;
// Discard the "dir"
SendToRecvFromBegin (list,xmit,xmit_to_rank,recv,recv_from_rank,bytes);
SendToRecvFromComplete(list);
return 2.0*bytes;
}
double CartesianCommunicator::StencilSendToRecvFromBegin(std::vector<CommsRequest_t> &list,
void *xmit,
int xmit_to_rank,
void *recv,
int recv_from_rank,
int bytes, int dir)
{
// Discard the "dir"
SendToRecvFromBegin(list,xmit,xmit_to_rank,recv,recv_from_rank,bytes);
return 2.0*bytes;
}
void CartesianCommunicator::StencilSendToRecvFromComplete(std::vector<CommsRequest_t> &waitall)
void CartesianCommunicator::StencilSendToRecvFromComplete(std::vector<CommsRequest_t> &waitall,int dir)
{
SendToRecvFromComplete(waitall);
}
#endif
#if !defined( GRID_COMMS_MPI3)
void CartesianCommunicator::StencilBarrier(void){};
commVector<uint8_t> CartesianCommunicator::ShmBufStorageVector;
@ -121,8 +146,22 @@ void *CartesianCommunicator::ShmBufferTranslate(int rank,void * local_p) {
return NULL;
}
void CartesianCommunicator::ShmInitGeneric(void){
#if 1
int mmap_flag = MAP_SHARED | MAP_ANONYMOUS;
#ifdef MAP_HUGETLB
if ( Hugepages ) mmap_flag |= MAP_HUGETLB;
#endif
ShmCommBuf =(void *) mmap(NULL, MAX_MPI_SHM_BYTES, PROT_READ | PROT_WRITE, mmap_flag, -1, 0);
if (ShmCommBuf == (void *)MAP_FAILED) {
perror("mmap failed ");
exit(EXIT_FAILURE);
}
#else
ShmBufStorageVector.resize(MAX_MPI_SHM_BYTES);
ShmCommBuf=(void *)&ShmBufStorageVector[0];
#endif
bzero(ShmCommBuf,MAX_MPI_SHM_BYTES);
}
#endif

55
lib/communicator/Communicator_base.h
View File

@ -38,7 +38,7 @@ Author: Peter Boyle <paboyle@ph.ed.ac.uk>
#ifdef GRID_COMMS_MPI3
#include <mpi.h>
#endif
#ifdef GRID_COMMS_MPI3L
#ifdef GRID_COMMS_MPIT
#include <mpi.h>
#endif
#ifdef GRID_COMMS_SHMEM
@ -50,12 +50,24 @@ namespace Grid {
class CartesianCommunicator {
public:
// 65536 ranks per node adequate for now
////////////////////////////////////////////
// Isend/Irecv/Wait, or Sendrecv blocking
////////////////////////////////////////////
enum CommunicatorPolicy_t { CommunicatorPolicyConcurrent, CommunicatorPolicySequential };
static CommunicatorPolicy_t CommunicatorPolicy;
static void SetCommunicatorPolicy(CommunicatorPolicy_t policy ) { CommunicatorPolicy = policy; }
///////////////////////////////////////////
// Up to 65536 ranks per node adequate for now
// 128MB shared memory for comms enought for 48^4 local vol comms
// Give external control (command line override?) of this
static const int MAXLOG2RANKSPERNODE = 16;
static uint64_t MAX_MPI_SHM_BYTES;
///////////////////////////////////////////
static const int MAXLOG2RANKSPERNODE = 16;
static uint64_t MAX_MPI_SHM_BYTES;
static int nCommThreads;
// use explicit huge pages
static int Hugepages;
// Communicator should know nothing of the physics grid, only processor grid.
int _Nprocessors; // How many in all
@ -64,14 +76,18 @@ class CartesianCommunicator {
std::vector<int> _processor_coor; // linear processor coordinate
unsigned long _ndimension;
#if defined (GRID_COMMS_MPI) || defined (GRID_COMMS_MPI3) || defined (GRID_COMMS_MPI3L)
#if defined (GRID_COMMS_MPI) || defined (GRID_COMMS_MPI3) || defined (GRID_COMMS_MPIT)
static MPI_Comm communicator_world;
MPI_Comm communicator;
MPI_Comm communicator;
std::vector<MPI_Comm> communicator_halo;
typedef MPI_Request CommsRequest_t;
#else
typedef int CommsRequest_t;
#endif
////////////////////////////////////////////////////////////////////
// Helper functionality for SHM Windows common to all other impls
////////////////////////////////////////////////////////////////////
@ -117,11 +133,7 @@ class CartesianCommunicator {
/////////////////////////////////
static void * ShmCommBuf;
// Isend/Irecv/Wait, or Sendrecv blocking
enum CommunicatorPolicy_t { CommunicatorPolicyConcurrent, CommunicatorPolicySequential };
static CommunicatorPolicy_t CommunicatorPolicy;
static void SetCommunicatorPolicy(CommunicatorPolicy_t policy ) { CommunicatorPolicy = policy; }
size_t heap_top;
size_t heap_bytes;
@ -211,14 +223,21 @@ class CartesianCommunicator {
void SendToRecvFromComplete(std::vector<CommsRequest_t> &waitall);
double StencilSendToRecvFrom(void *xmit,
int xmit_to_rank,
void *recv,
int recv_from_rank,
int bytes,int dir);
double StencilSendToRecvFromBegin(std::vector<CommsRequest_t> &list,
void *xmit,
int xmit_to_rank,
void *recv,
int recv_from_rank,
int bytes);
void *xmit,
int xmit_to_rank,
void *recv,
int recv_from_rank,
int bytes,int dir);
void StencilSendToRecvFromComplete(std::vector<CommsRequest_t> &waitall);
void StencilSendToRecvFromComplete(std::vector<CommsRequest_t> &waitall,int i);
void StencilBarrier(void);
////////////////////////////////////////////////////////////

62
lib/communicator/Communicator_mpi3.cc
View File

@ -41,9 +41,7 @@ Author: Peter Boyle <paboyle@ph.ed.ac.uk>
#ifdef HAVE_NUMAIF_H
#include <numaif.h>
#endif
#ifndef SHM_HUGETLB
#define SHM_HUGETLB 04000
#endif
namespace Grid {
@ -213,13 +211,19 @@ void CartesianCommunicator::Init(int *argc, char ***argv) {
int fd=shm_open(shm_name,O_RDWR|O_CREAT,0666);
if ( fd < 0 ) { perror("failed shm_open"); assert(0); }
ftruncate(fd, size);
int mmap_flag = MAP_SHARED;
#ifdef MAP_HUGETLB
if (Hugepages) mmap_flag |= MAP_HUGETLB;
#endif
void * ptr = mmap(NULL,size, PROT_READ | PROT_WRITE, mmap_flag, fd, 0);
void * ptr = mmap(NULL,size, PROT_READ | PROT_WRITE, MAP_SHARED, fd, 0);
if ( ptr == MAP_FAILED ) { perror("failed mmap"); assert(0); }
assert(((uint64_t)ptr&0x3F)==0);
// Try to force numa domain on the shm segment if we have numaif.h
#ifdef HAVE_NUMAIF_H
// Experiments; Experiments; Try to force numa domain on the shm segment if we have numaif.h
#if 0
//#ifdef HAVE_NUMAIF_H
int status;
int flags=MPOL_MF_MOVE;
#ifdef KNL
@ -266,7 +270,11 @@ void CartesianCommunicator::Init(int *argc, char ***argv) {
for(int r=0;r<ShmSize;r++){
size_t size = CartesianCommunicator::MAX_MPI_SHM_BYTES;
key_t key = 0x4545 + r;
if ((shmids[r]= shmget(key,size, SHM_HUGETLB | IPC_CREAT | SHM_R | SHM_W)) < 0) {
int flags = IPC_CREAT | SHM_R | SHM_W;
#ifdef SHM_HUGETLB
flags|=SHM_HUGETLB;
#endif
if ((shmids[r]= shmget(key,size, flags)) < 0) {
int errsv = errno;
printf("Errno %d\n",errsv);
perror("shmget");
@ -397,8 +405,14 @@ CartesianCommunicator::CartesianCommunicator(const std::vector<int> &processors)
{
int ierr;
communicator=communicator_world;
_ndimension = processors.size();
communicator_halo.resize (2*_ndimension);
for(int i=0;i<_ndimension*2;i++){
MPI_Comm_dup(communicator,&communicator_halo[i]);
}
////////////////////////////////////////////////////////////////
// Assert power of two shm_size.
////////////////////////////////////////////////////////////////
@ -621,13 +635,27 @@ void CartesianCommunicator::SendToRecvFromBegin(std::vector<CommsRequest_t> &lis
}
}
double CartesianCommunicator::StencilSendToRecvFromBegin(std::vector<CommsRequest_t> &list,
void *xmit,
int dest,
void *recv,
int from,
int bytes)
double CartesianCommunicator::StencilSendToRecvFrom( void *xmit,
int dest,
void *recv,
int from,
int bytes,int dir)
{
std::vector<CommsRequest_t> list;
double offbytes = StencilSendToRecvFromBegin(list,xmit,dest,recv,from,bytes,dir);
StencilSendToRecvFromComplete(list,dir);
return offbytes;
}
double CartesianCommunicator::StencilSendToRecvFromBegin(std::vector<CommsRequest_t> &list,
void *xmit,
int dest,
void *recv,
int from,
int bytes,int dir)
{
assert(dir < communicator_halo.size());
MPI_Request xrq;
MPI_Request rrq;
@ -646,26 +674,26 @@ double CartesianCommunicator::StencilSendToRecvFromBegin(std::vector<CommsReques
gfrom = MPI_UNDEFINED;
#endif
if ( gfrom ==MPI_UNDEFINED) {
ierr=MPI_Irecv(recv, bytes, MPI_CHAR,from,from,communicator,&rrq);
ierr=MPI_Irecv(recv, bytes, MPI_CHAR,from,from,communicator_halo[dir],&rrq);
assert(ierr==0);
list.push_back(rrq);
off_node_bytes+=bytes;
}
if ( gdest == MPI_UNDEFINED ) {
ierr =MPI_Isend(xmit, bytes, MPI_CHAR,dest,_processor,communicator,&xrq);
ierr =MPI_Isend(xmit, bytes, MPI_CHAR,dest,_processor,communicator_halo[dir],&xrq);
assert(ierr==0);
list.push_back(xrq);
off_node_bytes+=bytes;
}
if ( CommunicatorPolicy == CommunicatorPolicySequential ) {
this->StencilSendToRecvFromComplete(list);
this->StencilSendToRecvFromComplete(list,dir);
}
return off_node_bytes;
}
void CartesianCommunicator::StencilSendToRecvFromComplete(std::vector<CommsRequest_t> &waitall)
void CartesianCommunicator::StencilSendToRecvFromComplete(std::vector<CommsRequest_t> &waitall,int dir)
{
SendToRecvFromComplete(waitall);
}

286
lib/communicator/Communicator_mpit.cc Normal file
View File

@ -0,0 +1,286 @@
/*************************************************************************************
Grid physics library, www.github.com/paboyle/Grid
Source file: ./lib/communicator/Communicator_mpi.cc
Copyright (C) 2015
Author: Peter Boyle <paboyle@ph.ed.ac.uk>
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.
You should have received a copy of the GNU General Public License along
with this program; if not, write to the Free Software Foundation, Inc.,
51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
See the full license in the file "LICENSE" in the top level distribution directory
*************************************************************************************/
/* END LEGAL */
#include <Grid/GridCore.h>
#include <Grid/GridQCDcore.h>
#include <Grid/qcd/action/ActionCore.h>
#include <mpi.h>
namespace Grid {
///////////////////////////////////////////////////////////////////////////////////////////////////
// Info that is setup once and indept of cartesian layout
///////////////////////////////////////////////////////////////////////////////////////////////////
MPI_Comm CartesianCommunicator::communicator_world;
// Should error check all MPI calls.
void CartesianCommunicator::Init(int *argc, char ***argv) {
int flag;
int provided;
MPI_Initialized(&flag); // needed to coexist with other libs apparently
if ( !flag ) {
MPI_Init_thread(argc,argv,MPI_THREAD_MULTIPLE,&provided);
if ( provided != MPI_THREAD_MULTIPLE ) {
QCD::WilsonKernelsStatic::Comms = QCD::WilsonKernelsStatic::CommsThenCompute;
}
}
MPI_Comm_dup (MPI_COMM_WORLD,&communicator_world);
ShmInitGeneric();
}
CartesianCommunicator::CartesianCommunicator(const std::vector<int> &processors)
{
_ndimension = processors.size();
std::vector<int> periodic(_ndimension,1);
_Nprocessors=1;
_processors = processors;
_processor_coor.resize(_ndimension);
MPI_Cart_create(communicator_world, _ndimension,&_processors[0],&periodic[0],1,&communicator);
MPI_Comm_rank(communicator,&_processor);
MPI_Cart_coords(communicator,_processor,_ndimension,&_processor_coor[0]);
for(int i=0;i<_ndimension;i++){
_Nprocessors*=_processors[i];
}
communicator_halo.resize (2*_ndimension);
for(int i=0;i<_ndimension*2;i++){
MPI_Comm_dup(communicator,&communicator_halo[i]);
}
int Size;
MPI_Comm_size(communicator,&Size);
assert(Size==_Nprocessors);
}
void CartesianCommunicator::GlobalSum(uint32_t &u){
int ierr=MPI_Allreduce(MPI_IN_PLACE,&u,1,MPI_UINT32_T,MPI_SUM,communicator);
assert(ierr==0);
}
void CartesianCommunicator::GlobalSum(uint64_t &u){
int ierr=MPI_Allreduce(MPI_IN_PLACE,&u,1,MPI_UINT64_T,MPI_SUM,communicator);
assert(ierr==0);
}
void CartesianCommunicator::GlobalXOR(uint32_t &u){
int ierr=MPI_Allreduce(MPI_IN_PLACE,&u,1,MPI_UINT32_T,MPI_BXOR,communicator);
assert(ierr==0);
}
void CartesianCommunicator::GlobalXOR(uint64_t &u){
int ierr=MPI_Allreduce(MPI_IN_PLACE,&u,1,MPI_UINT64_T,MPI_BXOR,communicator);
assert(ierr==0);
}
void CartesianCommunicator::GlobalSum(float &f){
int ierr=MPI_Allreduce(MPI_IN_PLACE,&f,1,MPI_FLOAT,MPI_SUM,communicator);
assert(ierr==0);
}
void CartesianCommunicator::GlobalSumVector(float *f,int N)
{
int ierr=MPI_Allreduce(MPI_IN_PLACE,f,N,MPI_FLOAT,MPI_SUM,communicator);
assert(ierr==0);
}
void CartesianCommunicator::GlobalSum(double &d)
{
int ierr = MPI_Allreduce(MPI_IN_PLACE,&d,1,MPI_DOUBLE,MPI_SUM,communicator);
assert(ierr==0);
}
void CartesianCommunicator::GlobalSumVector(double *d,int N)
{
int ierr = MPI_Allreduce(MPI_IN_PLACE,d,N,MPI_DOUBLE,MPI_SUM,communicator);
assert(ierr==0);
}
void CartesianCommunicator::ShiftedRanks(int dim,int shift,int &source,int &dest)
{
int ierr=MPI_Cart_shift(communicator,dim,shift,&source,&dest);
assert(ierr==0);
}
int CartesianCommunicator::RankFromProcessorCoor(std::vector<int> &coor)
{
int rank;
int ierr=MPI_Cart_rank (communicator, &coor[0], &rank);
assert(ierr==0);
return rank;
}
void CartesianCommunicator::ProcessorCoorFromRank(int rank, std::vector<int> &coor)
{
coor.resize(_ndimension);
int ierr=MPI_Cart_coords (communicator, rank, _ndimension,&coor[0]);
assert(ierr==0);
}
// Basic Halo comms primitive
void CartesianCommunicator::SendToRecvFrom(void *xmit,
int dest,
void *recv,
int from,
int bytes)
{
std::vector<CommsRequest_t> reqs(0);
SendToRecvFromBegin(reqs,xmit,dest,recv,from,bytes);
SendToRecvFromComplete(reqs);
}
void CartesianCommunicator::SendRecvPacket(void *xmit,
void *recv,
int sender,
int receiver,
int bytes)
{
MPI_Status stat;
assert(sender != receiver);
int tag = sender;
if ( _processor == sender ) {
MPI_Send(xmit, bytes, MPI_CHAR,receiver,tag,communicator);
}
if ( _processor == receiver ) {
MPI_Recv(recv, bytes, MPI_CHAR,sender,tag,communicator,&stat);
}
}
// Basic Halo comms primitive
void CartesianCommunicator::SendToRecvFromBegin(std::vector<CommsRequest_t> &list,
void *xmit,
int dest,
void *recv,
int from,
int bytes)
{
int myrank = _processor;
int ierr;
if ( CommunicatorPolicy == CommunicatorPolicyConcurrent ) {
MPI_Request xrq;
MPI_Request rrq;
ierr =MPI_Irecv(recv, bytes, MPI_CHAR,from,from,communicator,&rrq);
ierr|=MPI_Isend(xmit, bytes, MPI_CHAR,dest,_processor,communicator,&xrq);
assert(ierr==0);
list.push_back(xrq);
list.push_back(rrq);
} else {
// Give the CPU to MPI immediately; can use threads to overlap optionally
ierr=MPI_Sendrecv(xmit,bytes,MPI_CHAR,dest,myrank,
recv,bytes,MPI_CHAR,from, from,
communicator,MPI_STATUS_IGNORE);
assert(ierr==0);
}
}
void CartesianCommunicator::SendToRecvFromComplete(std::vector<CommsRequest_t> &list)
{
if ( CommunicatorPolicy == CommunicatorPolicyConcurrent ) {
int nreq=list.size();
std::vector<MPI_Status> status(nreq);
int ierr = MPI_Waitall(nreq,&list[0],&status[0]);
assert(ierr==0);
}
}
void CartesianCommunicator::Barrier(void)
{
int ierr = MPI_Barrier(communicator);
assert(ierr==0);
}
void CartesianCommunicator::Broadcast(int root,void* data, int bytes)
{
int ierr=MPI_Bcast(data,
bytes,
MPI_BYTE,
root,
communicator);
assert(ierr==0);
}
///////////////////////////////////////////////////////
// Should only be used prior to Grid Init finished.
// Check for this?
///////////////////////////////////////////////////////
int CartesianCommunicator::RankWorld(void){
int r;
MPI_Comm_rank(communicator_world,&r);
return r;
}
void CartesianCommunicator::BroadcastWorld(int root,void* data, int bytes)
{
int ierr= MPI_Bcast(data,
bytes,
MPI_BYTE,
root,
communicator_world);
assert(ierr==0);
}
double CartesianCommunicator::StencilSendToRecvFromBegin(std::vector<CommsRequest_t> &list,
void *xmit,
int xmit_to_rank,
void *recv,
int recv_from_rank,
int bytes,int dir)
{
int myrank = _processor;
int ierr;
assert(dir < communicator_halo.size());
// std::cout << " sending on communicator "<<dir<<" " <<communicator_halo[dir]<<std::endl;
// Give the CPU to MPI immediately; can use threads to overlap optionally
MPI_Request req[2];
MPI_Irecv(recv,bytes,MPI_CHAR,recv_from_rank,recv_from_rank, communicator_halo[dir],&req[1]);
MPI_Isend(xmit,bytes,MPI_CHAR,xmit_to_rank ,myrank , communicator_halo[dir],&req[0]);
list.push_back(req[0]);
list.push_back(req[1]);
return 2.0*bytes;
}
void CartesianCommunicator::StencilSendToRecvFromComplete(std::vector<CommsRequest_t> &waitall,int dir)
{
int nreq=waitall.size();
MPI_Waitall(nreq, &waitall[0], MPI_STATUSES_IGNORE);
};
double CartesianCommunicator::StencilSendToRecvFrom(void *xmit,
int xmit_to_rank,
void *recv,
int recv_from_rank,
int bytes,int dir)
{
int myrank = _processor;
int ierr;
assert(dir < communicator_halo.size());
// std::cout << " sending on communicator "<<dir<<" " <<communicator_halo[dir]<<std::endl;
// Give the CPU to MPI immediately; can use threads to overlap optionally
MPI_Request req[2];
MPI_Irecv(recv,bytes,MPI_CHAR,recv_from_rank,recv_from_rank, communicator_halo[dir],&req[1]);
MPI_Isend(xmit,bytes,MPI_CHAR,xmit_to_rank ,myrank , communicator_halo[dir],&req[0]);
MPI_Waitall(2, req, MPI_STATUSES_IGNORE);
return 2.0*bytes;
}
}

2
lib/cshift/Cshift.h
View File

@ -42,7 +42,7 @@ Author: Peter Boyle <paboyle@ph.ed.ac.uk>
#include <Grid/cshift/Cshift_mpi.h>
#endif
#ifdef GRID_COMMS_MPI3L
#ifdef GRID_COMMS_MPIT
#include <Grid/cshift/Cshift_mpi.h>
#endif

2
lib/log/Log.cc
View File

@ -95,7 +95,7 @@ void GridLogConfigure(std::vector<std::string> &logstreams) {
////////////////////////////////////////////////////////////
void Grid_quiesce_nodes(void) {
int me = 0;
#if defined(GRID_COMMS_MPI) || defined(GRID_COMMS_MPI3) || defined(GRID_COMMS_MPI3L)
#if defined(GRID_COMMS_MPI) || defined(GRID_COMMS_MPI3) || defined(GRID_COMMS_MPIT)
MPI_Comm_rank(MPI_COMM_WORLD, &me);
#endif
#ifdef GRID_COMMS_SHMEM

2
lib/parallelIO/BinaryIO.h
View File

@ -29,7 +29,7 @@
#ifndef GRID_BINARY_IO_H
#define GRID_BINARY_IO_H
#if defined(GRID_COMMS_MPI) || defined(GRID_COMMS_MPI3)
#if defined(GRID_COMMS_MPI) || defined(GRID_COMMS_MPI3) || defined(GRID_COMMS_MPIT)
#define USE_MPI_IO
#else
#undef USE_MPI_IO

14
lib/qcd/action/fermion/CayleyFermion5D.cc
View File

@ -414,7 +414,7 @@ void CayleyFermion5D<Impl>::SetCoefficientsInternal(RealD zolo_hi,std::vector<Co
for(int i=0; i < Ls; i++){
as[i] = 1.0;
omega[i] = gamma[i]*zolo_hi; //NB reciprocal relative to Chroma NEF code
// assert(fabs(omega[i])>0.0);
assert(omega[i]!=Coeff_t(0.0));
bs[i] = 0.5*(bpc/omega[i] + bmc);
cs[i] = 0.5*(bpc/omega[i] - bmc);
}
@ -429,7 +429,7 @@ void CayleyFermion5D<Impl>::SetCoefficientsInternal(RealD zolo_hi,std::vector<Co
for(int i=0;i<Ls;i++){
bee[i]=as[i]*(bs[i]*(4.0-this->M5) +1.0);
// assert(fabs(bee[i])>0.0);
assert(bee[i]!=Coeff_t(0.0));
cee[i]=as[i]*(1.0-cs[i]*(4.0-this->M5));
beo[i]=as[i]*bs[i];
ceo[i]=-as[i]*cs[i];
@ -455,11 +455,17 @@ void CayleyFermion5D<Impl>::SetCoefficientsInternal(RealD zolo_hi,std::vector<Co
dee[i] = bee[i];
if ( i < Ls-1 ) {
assert(bee[i]!=Coeff_t(0.0));
assert(bee[0]!=Coeff_t(0.0));
lee[i] =-cee[i+1]/bee[i]; // sub-diag entry on the ith column
leem[i]=mass*cee[Ls-1]/bee[0];
for(int j=0;j<i;j++) leem[i]*= aee[j]/bee[j+1];
for(int j=0;j<i;j++) {
assert(bee[j+1]!=Coeff_t(0.0));
leem[i]*= aee[j]/bee[j+1];
}
uee[i] =-aee[i]/bee[i]; // up-diag entry on the ith row
@ -478,7 +484,7 @@ void CayleyFermion5D<Impl>::SetCoefficientsInternal(RealD zolo_hi,std::vector<Co
{
Coeff_t delta_d=mass*cee[Ls-1];
for(int j=0;j<Ls-1;j++) {
// assert(fabs(bee[j])>0.0);
assert(bee[j] != Coeff_t(0.0));
delta_d *= cee[j]/bee[j];
}
dee[Ls-1] += delta_d;

48
lib/qcd/action/fermion/WilsonCompressor.h
View File

@ -238,7 +238,33 @@ template<typename HCS,typename HS,typename S> using WilsonCompressor = WilsonCom
template<class vobj,class cobj>
class WilsonStencil : public CartesianStencil<vobj,cobj> {
public:
double timer0;
double timer1;
double timer2;
double timer3;
double timer4;
double timer5;
double timer6;
uint64_t callsi;
void ZeroCountersi(void)
{
timer0=0;
timer1=0;
timer2=0;
timer3=0;
timer4=0;
timer5=0;
timer6=0;
callsi=0;
}
void Reporti(int calls)
{
if ( timer0 ) std::cout << GridLogMessage << " timer0 (HaloGatherOpt) " <<timer0/calls <<std::endl;
if ( timer1 ) std::cout << GridLogMessage << " timer1 (Communicate) " <<timer1/calls <<std::endl;
if ( timer2 ) std::cout << GridLogMessage << " timer2 (CommsMerge ) " <<timer2/calls <<std::endl;
if ( timer3 ) std::cout << GridLogMessage << " timer3 (commsMergeShm) " <<timer3/calls <<std::endl;
if ( timer4 ) std::cout << GridLogMessage << " timer4 " <<timer4 <<std::endl;
}
typedef CartesianCommunicator::CommsRequest_t CommsRequest_t;
std::vector<int> same_node;
@ -252,6 +278,7 @@ public:
: CartesianStencil<vobj,cobj> (grid,npoints,checkerboard,directions,distances) ,
same_node(npoints)
{
ZeroCountersi();
surface_list.resize(0);
};
@ -261,7 +288,6 @@ public:
// Here we know the distance is 1 for WilsonStencil
for(int point=0;point<this->_npoints;point++){
same_node[point] = this->SameNode(point);
// std::cout << " dir " <<point<<" same_node " <<same_node[point]<<std::endl;
}
for(int site = 0 ;site< vol4;site++){
@ -282,17 +308,28 @@ public:
{
std::vector<std::vector<CommsRequest_t> > reqs;
this->HaloExchangeOptGather(source,compress);
this->CommunicateBegin(reqs);
this->CommunicateComplete(reqs);
double t1=usecond();
// Asynchronous MPI calls multidirectional, Isend etc...
// this->CommunicateBegin(reqs);
// this->CommunicateComplete(reqs);
// Non-overlapped directions within a thread. Asynchronous calls except MPI3, threaded up to comm threads ways.
this->Communicate();
double t2=usecond(); timer1 += t2-t1;
this->CommsMerge(compress);
double t3=usecond(); timer2 += t3-t2;
this->CommsMergeSHM(compress);
double t4=usecond(); timer3 += t4-t3;
}
template <class compressor>
void HaloExchangeOptGather(const Lattice<vobj> &source,compressor &compress)
{
this->Prepare();
double t0=usecond();
this->HaloGatherOpt(source,compress);
double t1=usecond();
timer0 += t1-t0;
callsi++;
}
template <class compressor>
@ -304,7 +341,9 @@ public:
typedef typename compressor::SiteHalfSpinor SiteHalfSpinor;
typedef typename compressor::SiteHalfCommSpinor SiteHalfCommSpinor;
this->mpi3synctime_g-=usecond();
this->_grid->StencilBarrier();
this->mpi3synctime_g+=usecond();
assert(source._grid==this->_grid);
this->halogtime-=usecond();
@ -323,7 +362,6 @@ public:
int dag = compress.dag;
int face_idx=0;
if ( dag ) {
// std::cout << " Optimised Dagger compress " <<std::endl;
assert(same_node[Xp]==this->HaloGatherDir(source,XpCompress,Xp,face_idx));
assert(same_node[Yp]==this->HaloGatherDir(source,YpCompress,Yp,face_idx));
assert(same_node[Zp]==this->HaloGatherDir(source,ZpCompress,Zp,face_idx));

107
lib/qcd/action/fermion/WilsonFermion5D.cc
View File

@ -124,22 +124,24 @@ WilsonFermion5D<Impl>::WilsonFermion5D(GaugeField &_Umu,
int vol4;
vol4=FourDimGrid.oSites();
Stencil.BuildSurfaceList(LLs,vol4);
vol4=FourDimRedBlackGrid.oSites();
StencilEven.BuildSurfaceList(LLs,vol4);
StencilOdd.BuildSurfaceList(LLs,vol4);
std::cout << GridLogMessage << " SurfaceLists "<< Stencil.surface_list.size()
<<" " << StencilEven.surface_list.size()<<std::endl;
// std::cout << GridLogMessage << " SurfaceLists "<< Stencil.surface_list.size()
// <<" " << StencilEven.surface_list.size()<<std::endl;
}
template<class Impl>
void WilsonFermion5D<Impl>::Report(void)
{
std::vector<int> latt = GridDefaultLatt();
RealD volume = Ls; for(int mu=0;mu<Nd;mu++) volume=volume*latt[mu];
RealD NP = _FourDimGrid->_Nprocessors;
RealD NN = _FourDimGrid->NodeCount();
RealD NP = _FourDimGrid->_Nprocessors;
RealD NN = _FourDimGrid->NodeCount();
RealD volume = Ls;
std::vector<int> latt = _FourDimGrid->GlobalDimensions();
for(int mu=0;mu<Nd;mu++) volume=volume*latt[mu];
if ( DhopCalls > 0 ) {
std::cout << GridLogMessage << "#### Dhop calls report " << std::endl;
@ -185,6 +187,11 @@ void WilsonFermion5D<Impl>::Report(void)
std::cout << GridLogMessage << "WilsonFermion5D StencilEven"<<std::endl; StencilEven.Report();
std::cout << GridLogMessage << "WilsonFermion5D StencilOdd" <<std::endl; StencilOdd.Report();
}
if ( DhopCalls > 0){
std::cout << GridLogMessage << "WilsonFermion5D Stencil Reporti()" <<std::endl; Stencil.Reporti(DhopCalls);
std::cout << GridLogMessage << "WilsonFermion5D StencilEven Reporti()"<<std::endl; StencilEven.Reporti(DhopCalls);
std::cout << GridLogMessage << "WilsonFermion5D StencilOdd Reporti()" <<std::endl; StencilOdd.Reporti(DhopCalls);
}
}
template<class Impl>
@ -204,6 +211,9 @@ void WilsonFermion5D<Impl>::ZeroCounters(void) {
Stencil.ZeroCounters();
StencilEven.ZeroCounters();
StencilOdd.ZeroCounters();
Stencil.ZeroCountersi();
StencilEven.ZeroCountersi();
StencilOdd.ZeroCountersi();
}
@ -380,7 +390,6 @@ void WilsonFermion5D<Impl>::DhopInternalOverlappedComms(StencilImpl & st, Lebesg
{
#ifdef GRID_OMP
// assert((dag==DaggerNo) ||(dag==DaggerYes));
typedef CartesianCommunicator::CommsRequest_t CommsRequest_t;
Compressor compressor(dag);
@ -389,46 +398,70 @@ void WilsonFermion5D<Impl>::DhopInternalOverlappedComms(StencilImpl & st, Lebesg
DhopFaceTime-=usecond();
st.HaloExchangeOptGather(in,compressor);
DhopFaceTime+=usecond();
std::vector<std::vector<CommsRequest_t> > reqs;
// Rely on async comms; start comms before merge of local data
DhopCommTime-=usecond();
st.CommunicateBegin(reqs);
DhopFaceTime-=usecond();
st.CommsMergeSHM(compressor);
st.CommsMergeSHM(compressor);// Could do this inside parallel region overlapped with comms
DhopFaceTime+=usecond();
// Perhaps use omp task and region
#pragma omp parallel
double ctime=0;
double ptime=0;
//////////////////////////////////////////////////////////////////////////////////////////////////////
// Ugly explicit thread mapping introduced for OPA reasons.
//////////////////////////////////////////////////////////////////////////////////////////////////////
#pragma omp parallel reduction(max:ctime) reduction(max:ptime)
{
int tid = omp_get_thread_num();
int nthreads = omp_get_num_threads();
int me = omp_get_thread_num();
int myoff, mywork;
GridThread::GetWork(len,me-1,mywork,myoff,nthreads-1);
int sF = LLs * myoff;
if ( me == 0 ) {
st.CommunicateComplete(reqs);
DhopCommTime+=usecond();
} else {
// Interior links in stencil
if ( me==1 ) DhopComputeTime-=usecond();
if (dag == DaggerYes) Kernels::DhopSiteDag(st,lo,U,st.CommBuf(),sF,myoff,LLs,mywork,in,out,1,0);
else Kernels::DhopSite(st,lo,U,st.CommBuf(),sF,myoff,LLs,mywork,in,out,1,0);
if ( me==1 ) DhopComputeTime+=usecond();
int ncomms = CartesianCommunicator::nCommThreads;
if (ncomms == -1) ncomms = 1;
assert(nthreads > ncomms);
if (tid >= ncomms) {
double start = usecond();
nthreads -= ncomms;
int ttid = tid - ncomms;
int n = U._grid->oSites();
int chunk = n / nthreads;
int rem = n % nthreads;
int myblock, myn;
if (ttid < rem) {
myblock = ttid * chunk + ttid;
myn = chunk+1;
} else {
myblock = ttid*chunk + rem;
myn = chunk;
}
// do the compute
if (dag == DaggerYes) {
for (int ss = myblock; ss < myblock+myn; ++ss) {
int sU = ss;
int sF = LLs * sU;
Kernels::DhopSiteDag(st,lo,U,st.CommBuf(),sF,sU,LLs,1,in,out,1,0);
}
} else {
for (int ss = myblock; ss < myblock+myn; ++ss) {
int sU = ss;
int sF = LLs * sU;
Kernels::DhopSite(st,lo,U,st.CommBuf(),sF,sU,LLs,1,in,out,1,0);
}
}
ptime = usecond() - start;
}
{
double start = usecond();
st.CommunicateThreaded();
ctime = usecond() - start;
}
}
DhopCommTime += ctime;
DhopComputeTime+=ptime;
// First to enter, last to leave timing
st.CollateThreads();
DhopFaceTime-=usecond();
st.CommsMerge(compressor);
DhopFaceTime+=usecond();
// Load imbalance alert. Should use dynamic schedule OMP for loop
// Perhaps create a list of only those sites with face work, and
// load balance process the list.
DhopComputeTime2-=usecond();
if (dag == DaggerYes) {
int sz=st.surface_list.size();
@ -449,11 +482,9 @@ void WilsonFermion5D<Impl>::DhopInternalOverlappedComms(StencilImpl & st, Lebesg
#else
assert(0);
#endif
}
template<class Impl>
void WilsonFermion5D<Impl>::DhopInternalSerialComms(StencilImpl & st, LebesgueOrder &lo,
DoubledGaugeField & U,

145
lib/stencil/Stencil.h
View File

@ -176,6 +176,9 @@ class CartesianStencil { // Stencil runs along coordinate axes only; NO diagonal
// Timing info; ugly; possibly temporary
/////////////////////////////////////////
double commtime;
double mpi3synctime;
double mpi3synctime_g;
double shmmergetime;
double gathertime;
double gathermtime;
double halogtime;
@ -185,6 +188,10 @@ class CartesianStencil { // Stencil runs along coordinate axes only; NO diagonal
double splicetime;
double nosplicetime;
double calls;
std::vector<double> comm_bytes_thr;
std::vector<double> comm_time_thr;
std::vector<double> comm_enter_thr;
std::vector<double> comm_leave_thr;
////////////////////////////////////////
// Stencil query
@ -248,35 +255,120 @@ class CartesianStencil { // Stencil runs along coordinate axes only; NO diagonal
//////////////////////////////////////////
// Comms packet queue for asynch thread
//////////////////////////////////////////
void CommunicateThreaded()
{
#ifdef GRID_OMP
// must be called in parallel region
int mythread = omp_get_thread_num();
int nthreads = CartesianCommunicator::nCommThreads;
#else
int mythread = 0;
int nthreads = 1;
#endif
if (nthreads == -1) nthreads = 1;
if (mythread < nthreads) {
comm_enter_thr[mythread] = usecond();
for (int i = mythread; i < Packets.size(); i += nthreads) {
uint64_t bytes = _grid->StencilSendToRecvFrom(Packets[i].send_buf,
Packets[i].to_rank,
Packets[i].recv_buf,
Packets[i].from_rank,
Packets[i].bytes,i);
comm_bytes_thr[mythread] += bytes;
}
comm_leave_thr[mythread]= usecond();
comm_time_thr[mythread] += comm_leave_thr[mythread] - comm_enter_thr[mythread];
}
}
void CollateThreads(void)
{
int nthreads = CartesianCommunicator::nCommThreads;
double first=0.0;
double last =0.0;
for(int t=0;t<nthreads;t++) {
double t0 = comm_enter_thr[t];
double t1 = comm_leave_thr[t];
comms_bytes+=comm_bytes_thr[t];
comm_enter_thr[t] = 0.0;
comm_leave_thr[t] = 0.0;
comm_time_thr[t] = 0.0;
comm_bytes_thr[t]=0;
if ( first == 0.0 ) first = t0; // first is t0
if ( (t0 > 0.0) && ( t0 < first ) ) first = t0; // min time seen
if ( t1 > last ) last = t1; // max time seen
}
commtime+= last-first;
}
void CommunicateBegin(std::vector<std::vector<CommsRequest_t> > &reqs)
{
reqs.resize(Packets.size());
commtime-=usecond();
for(int i=0;i<Packets.size();i++){
comms_bytes+=_grid->StencilSendToRecvFromBegin(reqs[i],
Packets[i].send_buf,
Packets[i].to_rank,
Packets[i].recv_buf,
Packets[i].from_rank,
Packets[i].bytes);
Packets[i].send_buf,
Packets[i].to_rank,
Packets[i].recv_buf,
Packets[i].from_rank,
Packets[i].bytes,i);
}
}
void CommunicateComplete(std::vector<std::vector<CommsRequest_t> > &reqs)
{
for(int i=0;i<Packets.size();i++){
_grid->StencilSendToRecvFromComplete(reqs[i]);
_grid->StencilSendToRecvFromComplete(reqs[i],i);
}
commtime+=usecond();
}
void Communicate(void)
{
#ifdef GRID_OMP
#pragma omp parallel
{
// must be called in parallel region
int mythread = omp_get_thread_num();
int maxthreads= omp_get_max_threads();
int nthreads = CartesianCommunicator::nCommThreads;
assert(nthreads <= maxthreads);
if (nthreads == -1) nthreads = 1;
#else
int mythread = 0;
int nthreads = 1;
#endif
if (mythread < nthreads) {
for (int i = mythread; i < Packets.size(); i += nthreads) {
double start = usecond();
comm_bytes_thr[mythread] += _grid->StencilSendToRecvFrom(Packets[i].send_buf,
Packets[i].to_rank,
Packets[i].recv_buf,
Packets[i].from_rank,
Packets[i].bytes,i);
comm_time_thr[mythread] += usecond() - start;
}
}
#ifdef GRID_OMP
}
#endif
}
template<class compressor> void HaloExchange(const Lattice<vobj> &source,compressor &compress)
{
std::vector<std::vector<CommsRequest_t> > reqs;
Prepare();
HaloGather(source,compress);
CommunicateBegin(reqs);
CommunicateComplete(reqs);
// Concurrent
//CommunicateBegin(reqs);
//CommunicateComplete(reqs);
// Sequential, possibly threaded
Communicate();
CommsMergeSHM(compress);
CommsMerge(compress);
}
@ -337,7 +429,9 @@ class CartesianStencil { // Stencil runs along coordinate axes only; NO diagonal
template<class compressor>
void HaloGather(const Lattice<vobj> &source,compressor &compress)
{
mpi3synctime_g-=usecond();
_grid->StencilBarrier();// Synch shared memory on a single nodes
mpi3synctime_g+=usecond();
// conformable(source._grid,_grid);
assert(source._grid==_grid);
@ -397,8 +491,12 @@ class CartesianStencil { // Stencil runs along coordinate axes only; NO diagonal
CommsMerge(decompress,Mergers,Decompressions);
}
template<class decompressor> void CommsMergeSHM(decompressor decompress) {
mpi3synctime-=usecond();
_grid->StencilBarrier();// Synch shared memory on a single nodes
mpi3synctime+=usecond();
shmmergetime-=usecond();
CommsMerge(decompress,MergersSHM,DecompressionsSHM);
shmmergetime+=usecond();
}
template<class decompressor>
@ -442,7 +540,12 @@ class CartesianStencil { // Stencil runs along coordinate axes only; NO diagonal
int checkerboard,
const std::vector<int> &directions,
const std::vector<int> &distances)
: _permute_type(npoints), _comm_buf_size(npoints)
: _permute_type(npoints),
_comm_buf_size(npoints),
comm_bytes_thr(npoints),
comm_enter_thr(npoints),
comm_leave_thr(npoints),
comm_time_thr(npoints)
{
face_table_computed=0;
_npoints = npoints;
@ -996,6 +1099,15 @@ class CartesianStencil { // Stencil runs along coordinate axes only; NO diagonal
void ZeroCounters(void) {
gathertime = 0.;
commtime = 0.;
mpi3synctime=0.;
mpi3synctime_g=0.;
shmmergetime=0.;
for(int i=0;i<_npoints;i++){
comm_time_thr[i]=0;
comm_bytes_thr[i]=0;
comm_enter_thr[i]=0;
comm_leave_thr[i]=0;
}
halogtime = 0.;
mergetime = 0.;
decompresstime = 0.;
@ -1011,6 +1123,18 @@ class CartesianStencil { // Stencil runs along coordinate axes only; NO diagonal
#define PRINTIT(A) AVERAGE(A); std::cout << GridLogMessage << " Stencil " << #A << " "<< A/calls<<std::endl;
RealD NP = _grid->_Nprocessors;
RealD NN = _grid->NodeCount();
double t = 0;
// if comm_time_thr is set they were all done in parallel so take the max
// but add up the bytes
int threaded = 0 ;
for (int i = 0; i < 8; ++i) {
if ( comm_time_thr[i]>0.0 ) {
threaded = 1;
comms_bytes += comm_bytes_thr[i];
if (t < comm_time_thr[i]) t = comm_time_thr[i];
}
}
if (threaded) commtime += t;
_grid->GlobalSum(commtime); commtime/=NP;
if ( calls > 0. ) {
@ -1026,6 +1150,9 @@ class CartesianStencil { // Stencil runs along coordinate axes only; NO diagonal
std::cout << GridLogMessage << " Stencil " << comms_bytes/commtime/1000. << " GB/s per rank"<<std::endl;
std::cout << GridLogMessage << " Stencil " << comms_bytes/commtime/1000.*NP/NN << " GB/s per node"<<std::endl;
}
PRINTIT(mpi3synctime);
PRINTIT(mpi3synctime_g);
PRINTIT(shmmergetime);
PRINTIT(splicetime);
PRINTIT(nosplicetime);
}

22
lib/util/Init.cc
View File

@ -222,6 +222,11 @@ void Grid_init(int *argc,char ***argv)
CartesianCommunicator::MAX_MPI_SHM_BYTES = MB*1024*1024;
}
if( GridCmdOptionExists(*argv,*argv+*argc,"--shm-hugepages") ){
CartesianCommunicator::Hugepages = 1;
}
if( GridCmdOptionExists(*argv,*argv+*argc,"--debug-signals") ){
Grid_debug_handler_init();
}
@ -304,6 +309,7 @@ void Grid_init(int *argc,char ***argv)
std::cout<<GridLogMessage<<" --threads n : default number of OMP threads"<<std::endl;
std::cout<<GridLogMessage<<" --grid n.n.n.n : default Grid size"<<std::endl;
std::cout<<GridLogMessage<<" --shm M : allocate M megabytes of shared memory for comms"<<std::endl;
std::cout<<GridLogMessage<<" --shm-hugepages : use explicit huge pages in mmap call "<<std::endl;
std::cout<<GridLogMessage<<std::endl;
std::cout<<GridLogMessage<<"Verbose and debug:"<<std::endl;
std::cout<<GridLogMessage<<std::endl;
@ -317,7 +323,7 @@ void Grid_init(int *argc,char ***argv)
std::cout<<GridLogMessage<<std::endl;
std::cout<<GridLogMessage<<" --comms-concurrent : Asynchronous MPI calls; several dirs at a time "<<std::endl;
std::cout<<GridLogMessage<<" --comms-sequential : Synchronous MPI calls; one dirs at a time "<<std::endl;
std::cout<<GridLogMessage<<" --comms-overlap : Overlap comms with compute "<<std::endl;
std::cout<<GridLogMessage<<" --comms-overlap : Overlap comms with compute "<<std::endl;
std::cout<<GridLogMessage<<std::endl;
std::cout<<GridLogMessage<<" --dslash-generic: Wilson kernel for generic Nc"<<std::endl;
std::cout<<GridLogMessage<<" --dslash-unroll : Wilson kernel for Nc=3"<<std::endl;
@ -356,10 +362,15 @@ void Grid_init(int *argc,char ***argv)
if( GridCmdOptionExists(*argv,*argv+*argc,"--comms-sequential") ){
CartesianCommunicator::SetCommunicatorPolicy(CartesianCommunicator::CommunicatorPolicySequential);
}
if( GridCmdOptionExists(*argv,*argv+*argc,"--lebesgue") ){
LebesgueOrder::UseLebesgueOrder=1;
}
CartesianCommunicator::nCommThreads = -1;
if( GridCmdOptionExists(*argv,*argv+*argc,"--comms-threads") ){
arg= GridCmdOptionPayload(*argv,*argv+*argc,"--comms-threads");
GridCmdOptionInt(arg,CartesianCommunicator::nCommThreads);
}
if( GridCmdOptionExists(*argv,*argv+*argc,"--cacheblocking") ){
arg= GridCmdOptionPayload(*argv,*argv+*argc,"--cacheblocking");
GridCmdOptionIntVector(arg,LebesgueOrder::Block);
@ -374,7 +385,10 @@ void Grid_init(int *argc,char ***argv)
Grid_default_latt,
Grid_default_mpi);
std::cout << GridLogDebug << "Requesting "<< CartesianCommunicator::MAX_MPI_SHM_BYTES <<" byte stencil comms buffers "<<std::endl;
std::cout << GridLogMessage << "Requesting "<< CartesianCommunicator::MAX_MPI_SHM_BYTES <<" byte stencil comms buffers "<<std::endl;
if ( CartesianCommunicator::Hugepages) {
std::cout << GridLogMessage << "Mapped stencil comms buffers as MAP_HUGETLB "<<std::endl;
}
if( GridCmdOptionExists(*argv,*argv+*argc,"--decomposition") ){
std::cout<<GridLogMessage<<"Grid Default Decomposition patterns\n";
@ -393,7 +407,7 @@ void Grid_init(int *argc,char ***argv)
void Grid_finalize(void)
{
#if defined (GRID_COMMS_MPI) || defined (GRID_COMMS_MPI3)
#if defined (GRID_COMMS_MPI) || defined (GRID_COMMS_MPI3) || defined (GRID_COMMS_MPIT)
MPI_Finalize();
Grid_unquiesce_nodes();
#endif