opsis

Hi there!

I have a program waiting for messages from another process and it must have an infinite loop like for(; but I wouldn't like to have it using 100% CPU while it doesn't have messages to read.

How can I solve this problem? The messages are sent via shared memory.

I've been trying to use pthread_cond (since the reading part is in another thread). But the thread with an infinite loop is the one that can see when there is or not a new message.

Thanks in advance!

Salem

Some reading - http://www.llnl.gov/computing/tutorials/pthreads/
The sending thread writes to shared memory (why not use thread messages?), and sets the condition variable.
The receiving thread waits for the condition variable, then reads shared memory.

__________________
If you dance barefoot on the broken glass of undefined behaviour, you've got to expect the occasional cut.

opsis

No, you got me wrong...

I have a sending *process* and a receiver process with a receiving thread. That's why I can't use it (I guess).

CornedBee

You could use a SysV message queue simply for the purpose of making one process wait for another.

__________________
All the buzzt!
CornedBee

"There is not now, nor has there ever been, nor will there ever be, any programming language in which it is the least bit difficult to write bad code."
- Flon's Law

Salem

Are you doing anything else if a message cannot be read?

Consider using say usleep() to allow you to poll for a message, then wait for a short while.

Or maybe use a pipe between the two processes which can be used to signal that a message is available.

__________________
If you dance barefoot on the broken glass of undefined behaviour, you've got to expect the occasional cut.