Hangs on redisplay_internal

Hangs on redisplay_internal

[Eric Abrahamsen]

I've been updating master every three or four days, and over the past
week (more?) feel like I've been getting a lot of inexplicable hangs.
They last as long as five or six seconds, but I can usually quit out of
them. I started running with debug on quit as a matter of course, and
just managed to kill one of them, and got:

Debugger entered--Lisp error: (quit)
  redisplay_internal\ \(C\ function\)()

That's all that was in *Backtrace*.

I'm on Arch Linux, and haven't done anything special to the build
process. I know this isn't very helpful, but maybe I can track down
something more.

Thanks,
Eric

Re: Hangs on redisplay_internal

[Eli Zaretskii]

On June 3, 2019 8:56:33 AM GMT+03:00, Eric Abrahamsen <eric@ericabrahamsen.net> wrote:
> I've been updating master every three or four days, and over the past
> week (more?) feel like I've been getting a lot of inexplicable hangs.
> They last as long as five or six seconds, but I can usually quit out
> of
> them. I started running with debug on quit as a matter of course, and
> just managed to kill one of them, and got:
> 
> Debugger entered--Lisp error: (quit)
>   redisplay_internal\ \(C\ function\)()
> 
> That's all that was in *Backtrace*.
> 
> I'm on Arch Linux, and haven't done anything special to the build
> process. I know this isn't very helpful, but maybe I can track down
> something more.

Run Emacs under GDB, and when Emacs habgs, stop it with SIGTSTP, as explained in etc/DEBUG.  Then get a more informative backtrace with the GDB's "bt" command.  Be sure to invoke GDB from the src directory, so that it will also show the Lisp backtrace.

Re: Hangs on redisplay_internal

[Eric Abrahamsen]

On 06/03/19 09:43 AM, Eli Zaretskii wrote:
> On June 3, 2019 8:56:33 AM GMT+03:00, Eric Abrahamsen <eric@ericabrahamsen.net> wrote:
>> I've been updating master every three or four days, and over the past
>> week (more?) feel like I've been getting a lot of inexplicable hangs.
>> They last as long as five or six seconds, but I can usually quit out
>> of
>> them. I started running with debug on quit as a matter of course, and
>> just managed to kill one of them, and got:
>> 
>> Debugger entered--Lisp error: (quit)
>>   redisplay_internal\ \(C\ function\)()
>> 
>> That's all that was in *Backtrace*.
>> 
>> I'm on Arch Linux, and haven't done anything special to the build
>> process. I know this isn't very helpful, but maybe I can track down
>> something more.
>
> Run Emacs under GDB, and when Emacs habgs, stop it with SIGTSTP, as
> explained in etc/DEBUG. Then get a more informative backtrace with the
> GDB's "bt" command. Be sure to invoke GDB from the src directory, so
> that it will also show the Lisp backtrace.

Will do. Hopefully I'll report back in a day or so with something
useful.

Thanks,
Eric

Re: Hangs on redisplay_internal

[Eric Abrahamsen]

Eli Zaretskii <eliz@gnu.org> writes:

> On June 3, 2019 8:56:33 AM GMT+03:00, Eric Abrahamsen <eric@ericabrahamsen.net> wrote:
>> I've been updating master every three or four days, and over the past
>> week (more?) feel like I've been getting a lot of inexplicable hangs.
>> They last as long as five or six seconds, but I can usually quit out
>> of
>> them. I started running with debug on quit as a matter of course, and
>> just managed to kill one of them, and got:
>> 
>> Debugger entered--Lisp error: (quit)
>>   redisplay_internal\ \(C\ function\)()
>> 
>> That's all that was in *Backtrace*.
>> 
>> I'm on Arch Linux, and haven't done anything special to the build
>> process. I know this isn't very helpful, but maybe I can track down
>> something more.
>
> Run Emacs under GDB, and when Emacs habgs, stop it with SIGTSTP, as
> explained in etc/DEBUG. Then get a more informative backtrace with the
> GDB's "bt" command. Be sure to invoke GDB from the src directory, so
> that it will also show the Lisp backtrace.

And, of course, it never happened again. At least I know what to do now
if it does.

Re: Hangs on redisplay_internal

[Kaushal Modi]

[-- Attachment #1: Type: text/plain, Size: 555 bytes --]

On Mon, Jun 3, 2019, 1:56 AM Eric Abrahamsen <eric@ericabrahamsen.net>
wrote:

> I've been updating master every three or four days, and over the past
> week (more?) feel like I've been getting a lot of inexplicable hangs.
> They last as long as five or six seconds, but I can usually quit out of
> them.
>

I've experienced those too! I am on RHEL 6.8.

I think it happened 3-4 times in the last two days. But it would happen
when I'm deep in some work, and I'd just frantically C-g out of it.

Let me see if I can get a useful gdb backtrace as well.

>

[-- Attachment #2: Type: text/html, Size: 1170 bytes --]