Receive a pulse on a channel
#include <sys/neutrino.h> int MsgReceivePulsev( int chid, const iov_t * piov, int parts, struct _msg_info * info ); int MsgReceivePulsev_r( int chid, const iov_t * piov, int parts, struct _msg_info * info );
|The first buffer of the IOV (input/output vector) must be big enough to contain a struct _pulse structure. If it isn't, you'll get an EFAULT.|
Use the -l c option to qcc to link against this library. This library is usually included automatically.
The MsgReceivePulsev() and MsgReceivePulsev_r() kernel calls wait for a pulse to arrive on the channel identified by chid and place the received data in the first entry of the array of buffers pointed to by piov. If the buffer is large enough, the number of bytes written to it is the size of a struct _pulse structure.
These functions are identical, except in the way they indicate errors; see the Returns section for details. These functions aren't very useful because the “array of buffers” is really a single element that has to be the size of a struct _pulse structure. MsgReceivePulse() has a cleaner set of arguments.
If a pulse is waiting on the channel when you call MsgReceivePulsev(), the calling thread doesn't block, and the pulse is immediately copied. If a pulse isn't waiting, the calling thread enters the RECEIVE-blocked state until a pulse arrives.
If multiple pulses are sent to a channel without a thread waiting to receive them, the pulses are queued in priority order.
When a thread receives a pulse:
|Don't reply to a pulse.|
The only difference between MsgReceivePulsev() and MsgReceivePulsev_r() is the way they indicate errors. On success, they both return 0.
If an error occurred:
MsgDeliverEvent(), MsgReceive(), MsgReceivePulse(), MsgReceivev(), MsgSendPulse(), _pulse, TimerTimeout()
Message Passing chapter of Getting Started with QNX Neutrino