[Fwd: gdm bug fix]



Damn isp's mail server didn't deliver this.  Try again:


Hi,

I found a bug in gdm.  Can someone fix this in the CVS.  The bug was a
race-condition in function gdm_slave_init (gdmslave.c:390)

The signal handler for SIG_USR1 should be set before the fork, otherwise
X might send the signal before the proper handler is in place.  If this
happens, gdmgreeter won't be started...

Here's the diff, to apply to gdmslave.c:

394,400c394
<
< /*
<     the SIGUSR1 handler must go before the fork, to ensure the proper
handler
<     is in place when X sends its signal
< */
<  signal(SIGUSR1, gdm_slave_usr1_handler);
<
---
>
402c396
<
---
>
427a422
>  signal(SIGUSR1, gdm_slave_usr1_handler);



BTW, should I be sending these to the list, or somewhere else?

Andrew Clausen





[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]