NAME
sc::RDMAMemoryGrp -
The RDMAMemoryGrp abstract class specializes the MsgMemoryGrp class.
SYNOPSIS
#include <memrdma.h>
Inherits sc::MsgMemoryGrp.
Inherited by sc::ARMCIMemoryGrp.
Public Member Functions
RDMAMemoryGrp (const Ref< MessageGrp > &msg)
RDMAMemoryGrp (const Ref< KeyVal > &)
void * localdata ()
Returns a pointer to the local data.
void set_localsize (size_t localsize)
Set the size of locally held memory.
void * obtain_writeonly (distsize_t offset, int size)
This gives write access to the memory location. No locking is done.
void * obtain_readwrite (distsize_t offset, int size)
Only one thread can have an unreleased obtain_readwrite at a time.
void * obtain_readonly (distsize_t offset, int size)
This gives read access to the memory location. No locking is done.
void release_readonly (void *data, distsize_t offset, int size)
This is called when read access is no longer needed.
void release_writeonly (void *data, distsize_t offset, int size)
This is called when write access is no longer needed.
void release_readwrite (void *data, distsize_t offset, int size)
This is called when read/write access is no longer needed.
void sum_reduction (double *data, distsize_t doffset, int dsize)
void sum_reduction_on_node (double *data, size_t doffset, int dsize,
int node=-1)
void print (std::ostream &o=ExEnv::out0()) const
Prints out information about the object.
Protected Member Functions
virtual void retrieve_data (void *, int node, int offset, int size, int
lock)=0
virtual void replace_data (void *, int node, int offset, int size, int
unlock)=0
virtual void sum_data (double *data, int node, int doffset, int
dsize)=0
void * malloc_region (size_t nbyte)
void free_region (void *)
Protected Attributes
char * data_
std::vector< Pool * > pools_
size_t default_pool_size_
Detailed Description
The RDMAMemoryGrp abstract class specializes the MsgMemoryGrp class.
It uses RDMA to implement global shared memory.
Member Function Documentation
void* sc::RDMAMemoryGrp::obtain_readwrite (distsize_t offset, int size)
[virtual]
Only one thread can have an unreleased obtain_readwrite at a time. The
actual memory region locked can be larger than that requested. If the
memory region is already locked this will block. For this reason, data
should be held as read/write for as short a time as possible.
Implements sc::MemoryGrp.
void sc::RDMAMemoryGrp::release_readwrite (void * data, distsize_t offset,
int size) [virtual]
This is called when read/write access is no longer needed. The memory
will be unlocked.
Implements sc::MemoryGrp.
void sc::RDMAMemoryGrp::set_localsize (size_t) [virtual]
Set the size of locally held memory. When memory is accessed using a
global offset counting starts at node 0 and proceeds up to node n() -
1.
Reimplemented from sc::MsgMemoryGrp.
Reimplemented in sc::ARMCIMemoryGrp.
Author
Generated automatically by Doxygen for MPQC from the source code.