ff Namespace Reference

The principal namespace for the FastFlow library. More...

Classes
class	Barrier
	Blocking barrier - Used only to start all nodes synchronously. More...
class	ff_allocator
	The ff_allocator, based on the idea of the Slab allocator More...
class	ff_dinout
	A ff::ff_dnode serving both Input and Output to the network. More...
class	ff_dnode
	Defines ff::ff_dnode. More...
class	ff_farm
	The Farm skeleton, with Emitter (lb_t) and Collector (gt_t). More...
class	ff_gatherer
	A class representing the Collector node in a Farm skeleton. More...
class	ff_loadbalancer
	A class representing the Emitter node in a typical Farm skeleton. More...
class	ff_mdf
	Macro Data Flow executor. More...
class	ff_minode
	Multiple input ff_node (the MPSC mediator) More...
class	ff_monode
	Multiple output ff_node (the SPMC mediator) More...
class	ff_node
	The FastFlow abstract contanier for a parallel activity (actor). More...
class	ff_oclNode
	OpenCL implementation of FastFlow node. More...
class	ff_ofarm
	The ordered Farm skeleton. More...
class	ff_pipe
	Pipeline pattern (high-level pattern syntax) More...
class	ff_pipeline
	The Pipeline skeleton (low-level syntax) More...
class	ff_thread
	thread container for (leaves) ff_node More...
class	FFAllocator
	A user-space parallel allocator (process-wide) More...
class	Lamport_Buffer
	Implementation of the well-known Lamport's wait-free circular buffer. Not currently used. More...
class	MPMC_Ptr_Queue
	An implementation of the bounded Multi-Producer/Multi-Consumer queue. Not currently used. More...
class	MSqueue
	Michael and Scott MPMC. Not currently used. More...
class	ofarm_gt
	Ordered farm Collector. More...
class	ofarm_lb
	Ordered farm emitter. More...
class	ParallelFor
	Parallel for loop. Run automatically. More...
class	ParallelForPipeReduce
	ParallelForPipeReduce class. More...
class	ParallelForReduce
	Parallel for and reduce. Run automatically. More...
class	poolEvolution
	The pool evolution parallel pattern. More...
class	spinBarrier
	Nonblocking barrier - Used only to start all nodes synchronously. More...
class	squeue
	Unbound sequential FIFO queue. More...
class	svector
	Simple yet efficient dynamic vector. More...
class	SWSR_Ptr_Buffer
	SPSC bound channel (Single-Writer/Single-Reader) More...
class	threadMapper
	The thread mapper allows to map threads to specific core using a predefined mapping policy. More...
class	uSWSR_Ptr_Buffer
	Unbounded Single-Writer/Single-Reader buffer (FastFlow unbound channel) More...

Functions
static unsigned int	address_hash_function (void *address)
template<typename Function >
static void	parallel_for (long first, long last, const Function &body, const long nw=FF_AUTO)
	Parallel loop over a range of indexes (step=1)
template<typename Function >
static void	parallel_for (long first, long last, long step, const Function &body, const long nw=FF_AUTO)
	Parallel loop over a range of indexes using a given step.
template<typename Function >
static void	parallel_for (long first, long last, long step, long grain, const Function &body, const long nw=FF_AUTO)
	Parallel loop over a range of indexes using a given step and granularity.

Variables
static const size_t	FF_EOS_NOFREEZE = (FF_EOS-0x1)
	automatically propagated
static const size_t	FF_GO_ON = (FF_EOS-0x2)
	non automatically propagated
static const size_t	FF_GO_OUT = (FF_EOS-0x3)
	non automatically propagated

Detailed Description

The principal namespace for the FastFlow library.

All classes and functions in the FastFlow library are contained within this namespace.

Function Documentation

static unsigned int ff::address_hash_function ( void *  address )

inlinestatic

Hash function to map addresses, cut into "long" size chunks, then XOR. The result will be matched to hash table size using mod in the hash table implementation