I'm still pursuing my timeout problem. The documentation for
DeviceRequestTO does not indicate that it returns an error if it the
operation times out. Is there a specific error message returned in this
case?
Also, if you use DeviceRequestAsyncTO, then how do you find if an error
occurred on the operation?
Currently, my code is calling DeviceRequest from a pthread that is
spawned each time it is called.
After DeviceRequest completes, the calling thread clears a busy a flag
so the main process knows it is done. For some reason this flag is
getting set or not getting cleared properly. The main thread is
responsible for setting the busy flag and starting a new thread that
calls DeviceRequest and the thread that calls DeviceRequest only clears
this flag just before is completes and the thread expires. This should
be safe. The flag is always checked to make sure it is cleared before a
new thread which calls DeviceRequest is issued by the main thread, so I
don't believe my code is nesting DeviceRequest calls.
But for some strange reason after about 12 feet of film is output by
the hardware, the busy flag doesn't get cleared or it gets set by a
gremlin. Console syslog messages appear to indicate that the call to
DeviceRequest in the thread completed and the flag was cleared.
Is there anything wrong with the concept of repeatedly calling
DeviceRequest from a newly spawned pthread, letting that thread expire
after the call and repeating it continuously?
Also, I am seeing some strange things when I use GDB within
ProjectBuilder to place a hardware watchpoint on the busy flag. It
indicates the busy flag is changing value in the plugin main routine in
a place where the code has nothing to do with it. How robust is the GDB
watch command.
Thanks for the earlier tip on installing the proper debug libraries.
Paul
_______________________________________________
usb mailing list | email@hidden
Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/usb
Do not post admin requests to the list. They will be ignored.
