Oh, I think I got it. childcb() schedules EV_SIGNAL, which triggers
another call to childcb() from eventloop.
Am I right?
2012/5/10 Brandon Black :
> On Thu, May 10, 2012 at 6:28 AM, Vetoshkin Nikita
> wrote:
>> So correct SIGCHLD handler should loop calling waitpid() until no zombi
Hi!
Not sure if this issue was on the list, google suggests nothing.
If I'm not mistaken, upon receiving SIGHCLD libev calls waitpid(-1,
...) only once, which is wrong because multiple SIGCHLDs can be merged
into one if many child processes died while we were in sighandler
code.
So correct SIGCHLD