portelli/Grid
1
0
Fork 0
mirror of https://github.com/paboyle/Grid.git synced 2025-04-09 05:30:46 +01:00
Code Releases Activity

Peer to peer on GPU's setup

Browse Source
This commit is contained in:
Peter Boyle 2018-09-10 11:26:20 +01:00
parent bc503b60e6
commit f02c7ea534
2 changed files with 101 additions and 29 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
lib/communicator/SharedMemory.h
View File

@ -96,6 +96,8 @@ public:
/////////////////////////////////////////////////// ///////////////////////////////////////////////////
static void SharedMemoryAllocate(uint64_t bytes, int flags); static void SharedMemoryAllocate(uint64_t bytes, int flags);
static void SharedMemoryFree(void); static void SharedMemoryFree(void);
static void SharedMemoryCopy(void *dest,const void *src,size_t bytes);
static void SharedMemoryZero(void *dest,size_t bytes);
}; };
@ -135,6 +137,7 @@ public:
// Call on any instance // Call on any instance
/////////////////////////////////////////////////// ///////////////////////////////////////////////////
void SharedMemoryTest(void); void SharedMemoryTest(void);
void *ShmBufferSelf(void); void *ShmBufferSelf(void);
void *ShmBuffer (int rank); void *ShmBuffer (int rank);
void *ShmBufferTranslate(int rank,void * local_p); void *ShmBufferTranslate(int rank,void * local_p);

127
lib/communicator/SharedMemoryMPI.cc
View File

@ -28,6 +28,10 @@ Author: Peter Boyle <paboyle@ph.ed.ac.uk>
#include <Grid/GridCore.h> #include <Grid/GridCore.h>
#ifdef GRID_NVCC
#include <cuda_runtime_api.h>
#endif
NAMESPACE_BEGIN(Grid); NAMESPACE_BEGIN(Grid);
/*Construct from an MPI communicator*/ /*Construct from an MPI communicator*/
@ -42,15 +46,12 @@ void GlobalSharedMemory::Init(Grid_MPI_Comm comm)
///////////////////////////////////////////////////////////////////// /////////////////////////////////////////////////////////////////////
// Split into groups that can share memory // Split into groups that can share memory
///////////////////////////////////////////////////////////////////// /////////////////////////////////////////////////////////////////////
#ifdef GRID_NVCC
MPI_Comm_split(comm, WorldRank , 0,&WorldShmComm);
#else
MPI_Comm_split_type(comm, MPI_COMM_TYPE_SHARED, 0, MPI_INFO_NULL,&WorldShmComm); MPI_Comm_split_type(comm, MPI_COMM_TYPE_SHARED, 0, MPI_INFO_NULL,&WorldShmComm);
#endif
MPI_Comm_rank(WorldShmComm ,&WorldShmRank); MPI_Comm_rank(WorldShmComm ,&WorldShmRank);
MPI_Comm_size(WorldShmComm ,&WorldShmSize); MPI_Comm_size(WorldShmComm ,&WorldShmSize);
std::cout << " World communicator of size " <<WorldSize << std::endl;
std::cout << " Shared communicator of size " <<WorldShmSize << std::endl; std::cout << " GRID initialisation: World communicator of size " <<WorldSize << std::endl;
std::cout << " GRID initialisation: Node communicator of size " <<WorldShmSize << std::endl;
// WorldShmComm, WorldShmSize, WorldShmRank // WorldShmComm, WorldShmSize, WorldShmRank
// WorldNodes // WorldNodes
@ -192,26 +193,74 @@ void GlobalSharedMemory::SharedMemoryAllocate(uint64_t bytes, int flags)
assert(_ShmSetup==1); assert(_ShmSetup==1);
assert(_ShmAlloc==0); assert(_ShmAlloc==0);
//////////////////////////////////////////////////////////////////////////////////////////////////////////
// allocate the pointer array for shared windows for our group
//////////////////////////////////////////////////////////////////////////////////////////////////////////
MPI_Barrier(WorldShmComm);
WorldShmCommBufs.resize(WorldShmSize);
//std::cerr << " allocating "<<bytes <<" bytes "<< std::endl; ///////////////////////////////////////////////////////////////////////////////////////////////////////////
auto err = cudaMallocManaged(&ShmCommBuf, bytes); // Each MPI rank should allocate our own buffer
///////////////////////////////////////////////////////////////////////////////////////////////////////////
auto err = cudaMalloc(&ShmCommBuf, bytes);
if ( err != cudaSuccess) { if ( err != cudaSuccess) {
std::cerr << " SharedMemoryMPI.cc cudaMallocManaged failed for " << bytes<<" bytes " <<cudaGetErrorString(err)<< std::endl; std::cerr << " SharedMemoryMPI.cc cudaMallocManaged failed for " << bytes<<" bytes " <<cudaGetErrorString(err)<< std::endl;
exit(EXIT_FAILURE); exit(EXIT_FAILURE);
} }
if (ShmCommBuf == (void *)NULL ) { if (ShmCommBuf == (void *)NULL ) {
std::cerr << " SharedMemoryMPI.cc cudaMallocManaged failed NULL pointer for " << bytes<<" bytes " << std::endl; std::cerr << " SharedMemoryMPI.cc cudaMallocManaged failed NULL pointer for " << bytes<<" bytes " << std::endl;
exit(EXIT_FAILURE); exit(EXIT_FAILURE);
} }
std::cerr << " Cuda allocated managed memory at "<<std::hex<<ShmCommBuf <<" - " << ((uint64_t)ShmCommBuf + bytes)<<std::dec<< std::endl; SharedMemoryZero(ShmCommBuf,bytes);
bzero(ShmCommBuf,bytes);
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// Loop over ranks/gpu's on our node
///////////////////////////////////////////////////////////////////////////////////////////////////////////
for(int r=0;r<WorldShmSize;r++){
//////////////////////////////////////////////////
// If it is me, pass around the IPC access key
//////////////////////////////////////////////////
cudaIpcMemHandle_t handle;
if ( r==WorldShmRank ) {
err = cudaIpcGetMemHandle(&handle,ShmCommBuf);
if ( err != cudaSuccess) {
std::cerr << " SharedMemoryMPI.cc cudaIpcGetMemHandle failed for rank" << r <<" "<<cudaGetErrorString(err)<< std::endl;
exit(EXIT_FAILURE);
}
}
//////////////////////////////////////////////////
// Share this IPC handle across the Shm Comm
//////////////////////////////////////////////////
{
int ierr=MPI_Bcast(&handle,
sizeof(handle),
MPI_BYTE,
r,
WorldShmComm);
assert(ierr==0);
}
///////////////////////////////////////////////////////////////
// If I am not the source, overwrite thisBuf with remote buffer
///////////////////////////////////////////////////////////////
void * thisBuf = ShmCommBuf;
if ( r!=WorldShmRank ) {
err = cudaIpcOpenMemHandle(&thisBuf,handle,cudaIpcMemLazyEnablePeerAccess);
if ( err != cudaSuccess) {
std::cerr << " SharedMemoryMPI.cc cudaIpcOpenMemHandle failed for rank" << r <<" "<<cudaGetErrorString(err)<< std::endl;
exit(EXIT_FAILURE);
}
}
///////////////////////////////////////////////////////////////
// Save a copy of the device buffers
///////////////////////////////////////////////////////////////
WorldShmCommBufs[r] = thisBuf;
}
WorldShmCommBufs.resize(1);
WorldShmCommBufs[0] = ShmCommBuf;
_ShmAllocBytes=bytes; _ShmAllocBytes=bytes;
_ShmAlloc=1; _ShmAlloc=1;
//std::cerr << " cudaMallocManaged Returning ; "<< bytes <<" bytes allocated at "<<std::hex<<ShmCommBuf <<std::dec<< std::endl;
} }
#else #else
#ifdef GRID_MPI3_SHMMMAP #ifdef GRID_MPI3_SHMMMAP
@ -326,7 +375,27 @@ void GlobalSharedMemory::SharedMemoryAllocate(uint64_t bytes, int flags)
_ShmAllocBytes = bytes; _ShmAllocBytes = bytes;
} }
#endif #endif
#endif // End NVCC case for GPU device buffers
/////////////////////////////////////////////////////////////////////////
// Routines accessing shared memory should route through for GPU safety
/////////////////////////////////////////////////////////////////////////
void GlobalSharedMemory::SharedMemoryZero(void *dest,size_t bytes)
{
#ifdef GRID_NVCC
cudaMemset(dest,0,bytes);
#else
bzero(dest,bytes);
#endif #endif
}
void GlobalSharedMemory::SharedMemoryCopy(void *dest,const void *src,size_t bytes)
{
#ifdef GRID_NVCC
cudaMemcpy(dest,src,bytes,cudaMemcpyDefault);
#else
bcopy(src,dest,bytes);
#endif
}
//////////////////////////////////////////////////////// ////////////////////////////////////////////////////////
// Global shared functionality finished // Global shared functionality finished
// Now move to per communicator functionality // Now move to per communicator functionality
@ -341,11 +410,7 @@ void SharedMemory::SetCommunicator(Grid_MPI_Comm comm)
///////////////////////////////////////////////////////////////////// /////////////////////////////////////////////////////////////////////
// Split into groups that can share memory // Split into groups that can share memory
///////////////////////////////////////////////////////////////////// /////////////////////////////////////////////////////////////////////
#ifdef GRID_NVCC
MPI_Comm_split(comm, rank, 0,&ShmComm);
#else
MPI_Comm_split_type(comm, MPI_COMM_TYPE_SHARED, 0, MPI_INFO_NULL,&ShmComm); MPI_Comm_split_type(comm, MPI_COMM_TYPE_SHARED, 0, MPI_INFO_NULL,&ShmComm);
#endif
MPI_Comm_rank(ShmComm ,&ShmRank); MPI_Comm_rank(ShmComm ,&ShmRank);
MPI_Comm_size(ShmComm ,&ShmSize); MPI_Comm_size(ShmComm ,&ShmSize);
ShmCommBufs.resize(ShmSize); ShmCommBufs.resize(ShmSize);
@ -374,6 +439,8 @@ void SharedMemory::SetCommunicator(Grid_MPI_Comm comm)
std::vector<int> ranks(size); for(int r=0;r<size;r++) ranks[r]=r; std::vector<int> ranks(size); for(int r=0;r<size;r++) ranks[r]=r;
MPI_Group_translate_ranks (FullGroup,size,&ranks[0],ShmGroup, &ShmRanks[0]); MPI_Group_translate_ranks (FullGroup,size,&ranks[0],ShmGroup, &ShmRanks[0]);
SharedMemoryTest();
} }
////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////
// On node barrier // On node barrier
@ -388,24 +455,26 @@ void SharedMemory::ShmBarrier(void)
void SharedMemory::SharedMemoryTest(void) void SharedMemory::SharedMemoryTest(void)
{ {
ShmBarrier(); ShmBarrier();
uint64_t check[3];
uint64_t magic = 0x5A5A5A;
if ( ShmRank == 0 ) { if ( ShmRank == 0 ) {
for(int r=0;r<ShmSize;r++){ for(uint64_t r=0;r<ShmSize;r++){
uint64_t * check = (uint64_t *) ShmCommBufs[r]; check[0]=GlobalSharedMemory::WorldNode;
check[0] = GlobalSharedMemory::WorldNode; check[1]=r;
check[1] = r; check[2]=magic;
check[2] = 0x5A5A5A; GlobalSharedMemory::SharedMemoryCopy( ShmCommBufs[r], check, 3*sizeof(uint64_t));
} }
} }
ShmBarrier(); ShmBarrier();
for(int r=0;r<ShmSize;r++){ for(uint64_t r=0;r<ShmSize;r++){
uint64_t * check = (uint64_t *) ShmCommBufs[r]; ShmBarrier();
GlobalSharedMemory::SharedMemoryCopy(check,ShmCommBufs[r], 3*sizeof(uint64_t));
ShmBarrier();
assert(check[0]==GlobalSharedMemory::WorldNode); assert(check[0]==GlobalSharedMemory::WorldNode);
assert(check[1]==r); assert(check[1]==r);
assert(check[2]==0x5A5A5A); assert(check[2]==magic);
ShmBarrier();
} }
ShmBarrier();
} }
void *SharedMemory::ShmBuffer(int rank) void *SharedMemory::ShmBuffer(int rank)