Hamish Marson wrote:
>
>I've take a quick look, and small nit, but you aren't checking the condition after
>the condition wait... Sometimes it's possible
>for the condition wait to return for a reason other than another thread broadcasting
>to it... Stevens (Volume 2) has a good
>chapter on pthreads & using condition variables & mutexes if you're interested...
>
Thanks... But what do you mean exactly with "checking the condition
after the condition wait" ?
Hmmm i worked on ppoa3 this night I give you some results from my
experiments:
The problem is that pppoa3 establish a connection only when using -v 2
switch. And when I use it with -v 0 or -v 1, it never establishes the
connection (pppd sends data but no answer from the modem ie the pppoa3 pty).
So i decided to merge pppoa3 and pppoa2 to skip all related thread
problems away. (Back to the fork method :-))
1 - I've merged the new report function.
It still ran fine so it is not a mutex related problem ( i've
changed the pthread mutex into IPC semaphore using the existing semaphore)
2 - I've merged the new ATM-AAL5/PPP stuff...
I've discovered something strange: the pppoa2 child was eating 96.8
CPU time. The child is responsible of the usb->pppd data stream. So we
are currently working on that part of the code but we (me ,bozo_fr and
stsquad) doesn't find nothing suspicious... :-(
My plans are now to audit the ATM code... I will keep you informed
through this ML or my website
Edouard Gomez
Liste de diffusion modem ALCATEL SpeedTouch USB
Pour se désinscrire : mailto:[EMAIL PROTECTED]?subject=unsubscribe
