Re-entrancy ? problem

["daniel sheltraw" <l5gibson@hotmail.com>]

Hello RTAIers

Hello everyone. I have a sort of reentrant problem. I know this
isn't a terribly specific question but rather than work on it
until it is more specific I thought I needed to catch you before
the weekend.

This is what my code does correctly:

  A kernel-space module contains a one-shot mode RT task.
  I load the module and after the task is started I immediately
  suspend it. I start the task with a write from user-space to a
  START_FIFO and a corresponding fifo handler resumes the task.
  The task again suspends after a certain number of one shot
  events and the user-space program returns. Everything works fine
  the first time I run the user-space code.

But this what happens next:

  The task is now suspended. When I try to resume the RT task
  a second time (the RT kernel module is still loaded), by
  writing to the START_FIFO from user-space again, the task
  resumes for a few events and then just hangs there (does
  not lock up the system).

I use shared memory as well as fifos to communicate between user
and kernel-spaces.

Question:
  Are there any reentrancy issues that I should be aware of with
  fifos and shared memory? I am using RTAI 0.7 (will upgrade next
  week when my new machine arrives).

Daniel S.
________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com