bug#24201: 25.1.50; TLS connections sometimes hang

[Andreas Schwab]

On Feb 19 2018, Lars Ingebrigtsen <larsi@gnus.org> wrote:

> Paul Eggert <eggert@cs.ucla.edu> writes:
>
>> Lars Ingebrigtsen wrote:
>>> #0  pthread_sigmask (how=0, newmask=<optimized out>, oldmask=0x7fffffff8a60)
>>>      at ../sysdeps/unix/sysv/linux/pthread_sigmask.c:50
>>
>> When you say that Emacs "hangs" surely you don't mean that the call to
>> pthread_sigmask is hanging. So, can we assume that Emacs is
>> busy-waiting for something? Is the CPU pegged? What does the busy-wait
>> loop look like?
>
> It's using 100% CPU, so it's busy-waiting, but every time I stop it,
> it's in that pthread_sigmask, as far as I'm able to tell...

Since it's making a syscall the loop is likely spending a significant
part of the time there.  That doesn't mean it's reponsible for the busy
wait.

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."