bug#20134: Acknowledgement (24.4.90; Emacs slowdown to eventual hang)

[Phillip Lord]

Eli Zaretskii <eliz@gnu.org> writes:

>> From: phillip.lord@newcastle.ac.uk (Phillip Lord)
>> Cc: <20134@debbugs.gnu.org>
>> Date: Tue, 07 Apr 2015 16:28:23 +0100
>> 
>> > When Emacs becomes sluggish, does pressing and holding an arrow key
>> > produces a slow response time for the first keypress, followed by more
>> > or less normal response times for the next keypresses?
>> 
>> Nope. The lag is general. If I carry one for long enough, emacs hangs
>> totally, but Emacs because unusable long before that point.
>
> Then I guess the primary suspects are non-idle timers and
> pre/post-command hooks.
>
>> timer-list is a variable defined in `C source code'.
>> Its value is
>> ([nil 21795 62426 202965 60 ac-clear-variables-every-minute nil nil 1000])
>
> Did you try running without this timer?
>
> What about the pre/post-command hooks?


Okay, I think this is likely to be correct, and also why the backtraces
are so useless. Emacs is making extensive use of CPU (running at 100%)
but only when I am typing. When I stop (which I obviously have to do to
check gdb, then the CPU drops again.

I've tried removing all post/pre command hooks with no success so far.