PreviousNext

Pipes

Pipes are a mechanism for efficiently handling large quantities of data by overlapping the transfer and processing of data. Input data is transferred in chunks to the server for processing, and output data is processed by the server in chunks and transferred to the client. A pipe is declared in a type definition of an interface definition, and the data type is used as parameters in the operations of the interface. The server manager calls stub pipe support routines in a loop, and the client stub calls pipe support routines that the client application must provide.

One of the pipe support routines that the client must provide is an alloc routine, which allocates a buffer for each chunk of pipe data. Given that pipes are intended to process data asynchronously, consuming it as it arrives, the alloc routine should not just blindly allocate a new buffer each time it is called, since the net effect would be to allocate space for the whole stream. A reasonable approach is either to declare a buffer statically or allocate it on the first call (per thread), and thereafter simply return the same buffer. The following code example shows the form an alloc routine takes in client application code.

#define CLIENT_BUFFER_SIZE 2048

idl_byte client_buffer[CLIENT_BUFFER_SIZE];

void client_alloc (state, bsize, buf, bcount)

rpc_ss_pipe_state_t state;

unsigned int bsize;

byte **buf;

unsigned int *bcount;

{

*buf = client_buffer;

*bcount = CLIENT_BUFFER_SIZE;

}

More:

Input Pipes

Output Pipes

Pipe Summary