Are you on Linux? Do you want to get good backtraces from an asynchronous signal handler? Do not use abort(3). It generates terrible backtraces, use kill(getpid(), SIGABRT).
-
Yes, it's probably because both raise() and abort() are marked noreturn, and the compiler is smart about it.pic.twitter.com/UcoJKpwQcQ
Hmm.. now I'm confused. raise() should not be a noreturn function. E.g. raise(SIGWINCH) will almost certainly return. So it seems like something else is going on here. Strange.
-
-
I think both raise() and abort() are asynchronous themselves, while kill() isn't? Keep in mind, I'm calling this from an asynchronous signal handler.
-
Hm.. Shouldn't be. raise(x) should really just be a shortcut for kill(getpid(), x). Also, raise() is guaranteed to only return after the signal handler (if any) finished executing. So it mustn't be implemented asynchronously. I get similar "bt" output for all 3 cases in this ex:pic.twitter.com/AXbIWmevDV
- Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.