bug#24201: 25.1.50; TLS connections sometimes hang

[Lars Ingebrigtsen <larsi@gnus.org>]

Lars Ingebrigtsen <larsi@gnus.org> writes:

>>   Ffuncall (nargs=4, args=args@entry=0x7fffffffb420) at eval.c:2826
>>   2826	  lisp_eval_depth--;
>>   Value returned is $12 = (union Lisp_X *) 0x0
>>
>> don't type "finish", but instead do this:
>>
>>   (gdb) p args[0]
>>   (gdb) xsymbol

It's frustrating -- when running under gdb, this hang seems to happen a
lot less often, and it hasn't happen against since last time.  But I was
just looking at the code to try to understand the previous backtrace.

So, the "finish" stuff seemed to show that we weren't hanging in
accept-process-output at all, but were instead just calling functions as
normal, and looping on the Lisp level?  And, for instance, Ffuncall has
a maybe_quit, so C-g should work in these instances?  But they don't,
which is confusing.

Is it even possible for the Lisp-ier functions in the backtrace to
inhibit C-g?

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no