Re: igc, macOS avoiding signals

[Helmut Eller]

On Tue, Dec 31 2024, Eli Zaretskii wrote:

>> Except the POSIX police: it says that pthread_mutex_trylock isn't async
>> signal safe.  I suppose this also makes it's unsafe to use MPS's fault
>> handler in an async signal handler.  Bummer.  (Does the police take
>> bribes?)
>
> Doesn't MPS itself call it from a SIGSEGV handler?

Yes.  But I'm not sure that SIGSEGV is considered an async signal.  I
supposed there is something like synchronous signals.

>> I wonder if the backtrace that we see in the signal handler is any
>> different from the backrace that we would see at the next safe point
>> (i.e. the next time maybe_quit is called).
>
> I think we cannot rely on that, because maybe_quit must be called by
> hand, it isn't magic.  We call it from various places in the
> interpreter, which could well be in some other place of a Lisp
> program.

Maybe somebody could do an experiment: record both backtraces for a
while and see if they differ.

> Once again, why not ask the MPS folks to give us a callback?  Or maybe
> we could try hacking MPS ourselves first, to see if that does the job,
> and ask them then?

Nobody stops you from doing this. :-)

Helmut