On Mon, Mar 22, 2004 at 10:00:22PM +0000, Angus Leeming wrote:
> > We only need to reap children every 10 minutes or whatever
>
> Then we go back to the timer and the whole thing has been a waste of
> time?
Why do you say that? It's completely normal and expected practice.
> It is trivially easy to execute code from the signal handler. Maybe
> that's OK in this case? The main code is not going to be in a fragile
> state because SIGCHLD is not an error. I guess that we still need
It's still asynchronous. The difference from the current code is that we
know there's likely to be a child waiting to be reaped. We drop out of
the Qt loop to check the sigatomic_t. If it's set, then we go and reap
the child, secure that we're not somewhere in the forked controller
code.
Because everything runs from the Qt event handler, we know we're safe.
Of course, you still have to be careful:
reapTimer()
{
while (sigatomic_value) {
sigatomic_value = 0;
reapAnyAndAllChildren();
}
}
or some such. You probably need a 'volatile' keyword on the sigatomic
too.
I don't know why the current code loops across all children. We should
just repeatedly wait for any child, and reap as needed.
> void child_handler(int)
> {
> while (process_signal) {
> process_signal = false;
> ForkedcallController::get().handleCompletedProcesses();
Racy. If we're in the forked controller code and we take a signal, we
end up calling back into the forked controller code. Suddenly the
controller code needs sigprocmask() all over the place or we run the
risk of racing against ourselves in signal context.
regards,
john
--
"Spammers get STABBED by GOD." - Ron Echeverri
