all messages for Emacs-related lists mirrored at yhetil.org
 help / color / mirror / code / Atom feed
* Lots of hangs Windows 10 26.0.91
@ 2018-03-28 14:18 Adam Taylor
  2018-03-29 12:08 ` Eli Zaretskii
                   ` (2 more replies)
  0 siblings, 3 replies; 6+ messages in thread
From: Adam Taylor @ 2018-03-28 14:18 UTC (permalink / raw
  To: help-gnu-emacs

Hi,
I'm looking for ideas on how to debug a fairly common hang I'm getting
while running emacs 26.0.91 under Windows 10 (I know, that's my first
problem).

I've been using mostly magit and elpy (python-mode) stuff when it happens.
I use desktop-save-mode which I suspected was the culprit until I turned it
off and the problem still happened.

What happens is that emacs starts to consume 30-40% of the cpu and starts
to eat memory at a good clip eventually causing my machine to start to page.

C-g does not work. I have to kill emacs from the task manager.

It does not seem to be linked to any particular action I make - sometimes
I'm just editing the python, sometimes I'm updating the magit status
window. No rhyme or reason that I can divine.

My guess is some issue in communicating with a subprocess (magit and elpy
both work with subprocesses).

I seem to remember some chatter about subprocess problems in the emacs
developer mailing list and wasn't sure if this was the same issue.

I would like to debug this, but since I didn't build the windows emacs, it
might be really hard to connect it to gdb and see where it has gone south.

Any help/pointers is greatly appreciated.
Thanks,
-Adam


^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: Lots of hangs Windows 10 26.0.91
  2018-03-28 14:18 Lots of hangs Windows 10 26.0.91 Adam Taylor
@ 2018-03-29 12:08 ` Eli Zaretskii
  2018-03-29 20:27 ` Phillip Lord
  2018-03-29 21:51 ` Stefan Monnier
  2 siblings, 0 replies; 6+ messages in thread
From: Eli Zaretskii @ 2018-03-29 12:08 UTC (permalink / raw
  To: help-gnu-emacs

> From: Adam Taylor <mr.adtaylor@gmail.com>
> Date: Wed, 28 Mar 2018 10:18:13 -0400
> 
> I'm looking for ideas on how to debug a fairly common hang I'm getting
> while running emacs 26.0.91 under Windows 10 (I know, that's my first
> problem).
> 
> I've been using mostly magit and elpy (python-mode) stuff when it happens.
> I use desktop-save-mode which I suspected was the culprit until I turned it
> off and the problem still happened.
> 
> What happens is that emacs starts to consume 30-40% of the cpu and starts
> to eat memory at a good clip eventually causing my machine to start to page.
> 
> C-g does not work. I have to kill emacs from the task manager.
> 
> It does not seem to be linked to any particular action I make - sometimes
> I'm just editing the python, sometimes I'm updating the magit status
> window. No rhyme or reason that I can divine.
> 
> My guess is some issue in communicating with a subprocess (magit and elpy
> both work with subprocesses).
> 
> I seem to remember some chatter about subprocess problems in the emacs
> developer mailing list and wasn't sure if this was the same issue.
> 
> I would like to debug this, but since I didn't build the windows emacs, it
> might be really hard to connect it to gdb and see where it has gone south.
> 
> Any help/pointers is greatly appreciated.

Please start by reporting this is a bug, using "M-x report-emacs-bug".
This forum is not the right place to discuss debugging of Emacs
problems.

Given the description of the issue, I'd start by asking whether you
are sure this is a "hang".  Did you leave Emacs alone long enough to
maybe finish whatever it's doing and become responsive again?  How
long did you wait before giving up and killing it?

Next, can you try disabling all or most of your customizations?  E.g.,
how about disabling all the non-default values of variables, and only
leaving the add-on packages that you load?  If that fixes the problem,
then you need to look at the customizations for the possible cause.

Did you try disabling the add-on packages one by one, perhaps some of
them has a bug.

If none of the above gives a clue, start Emacs under GDB, and when
that happens interrupt Emacs with Ctrl-C and show the C-level
backtrace.  The file etc/DEBUG in the distribution sho0uld explain how
to set things up so that Emacs will be interruptible in such
situations when run under GDB.  You will need to install a Windows
port of GDB, of course, which is available from the MSYS2 project.



^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: Lots of hangs Windows 10 26.0.91
  2018-03-28 14:18 Lots of hangs Windows 10 26.0.91 Adam Taylor
  2018-03-29 12:08 ` Eli Zaretskii
@ 2018-03-29 20:27 ` Phillip Lord
  2018-03-30  7:42   ` Eli Zaretskii
  2018-03-29 21:51 ` Stefan Monnier
  2 siblings, 1 reply; 6+ messages in thread
From: Phillip Lord @ 2018-03-29 20:27 UTC (permalink / raw
  To: Adam Taylor; +Cc: help-gnu-emacs

Adam Taylor <mr.adtaylor@gmail.com> writes:

> I'm looking for ideas on how to debug a fairly common hang I'm getting
> while running emacs 26.0.91 under Windows 10 (I know, that's my first
> problem).

It's a supported platform, so it's reasonable to complain!

> I've been using mostly magit and elpy (python-mode) stuff when it happens.
> I use desktop-save-mode which I suspected was the culprit until I turned it
> off and the problem still happened.
>
> What happens is that emacs starts to consume 30-40% of the cpu and starts
> to eat memory at a good clip eventually causing my machine to start to page.
>
> C-g does not work. I have to kill emacs from the task manager.

This is clearly a bug -- C-g should always work.

>
> It does not seem to be linked to any particular action I make - sometimes
> I'm just editing the python, sometimes I'm updating the magit status
> window. No rhyme or reason that I can divine.
>
> My guess is some issue in communicating with a subprocess (magit and elpy
> both work with subprocesses).
>
> I seem to remember some chatter about subprocess problems in the emacs
> developer mailing list and wasn't sure if this was the same issue.
>
> I would like to debug this, but since I didn't build the windows emacs, it
> might be really hard to connect it to gdb and see where it has gone south.
>
> Any help/pointers is greatly appreciated.

I'd ask on the devel mailing list, especially as the release candidate
is around the corner.

Phil



^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: Lots of hangs Windows 10 26.0.91
  2018-03-28 14:18 Lots of hangs Windows 10 26.0.91 Adam Taylor
  2018-03-29 12:08 ` Eli Zaretskii
  2018-03-29 20:27 ` Phillip Lord
@ 2018-03-29 21:51 ` Stefan Monnier
  2 siblings, 0 replies; 6+ messages in thread
From: Stefan Monnier @ 2018-03-29 21:51 UTC (permalink / raw
  To: help-gnu-emacs

> C-g does not work. I have to kill emacs from the task manager.

These are most commonly due to either a "pathological regexp" or some
broken code run with inhibit-quit (e.g. from a timer or process filter).

The best way to track it down (and fix it) starts by running Emacs
under GDB and to look at the backtrace (including and especially the
Lisp backtrace) during those hangs.


        Stefan




^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: Lots of hangs Windows 10 26.0.91
  2018-03-29 20:27 ` Phillip Lord
@ 2018-03-30  7:42   ` Eli Zaretskii
  2018-03-30 12:39     ` Kevin Buchs
  0 siblings, 1 reply; 6+ messages in thread
From: Eli Zaretskii @ 2018-03-30  7:42 UTC (permalink / raw
  To: help-gnu-emacs

> From: phillip.lord@russet.org.uk (Phillip Lord)
> Date: Thu, 29 Mar 2018 21:27:01 +0100
> Cc: help-gnu-emacs@gnu.org
> 
> > C-g does not work. I have to kill emacs from the task manager.
> 
> This is clearly a bug -- C-g should always work.

Unfortunately, due to how Emacs input works on Windows, there are
legitimate situations where C-g doesn't (and cannot) work.  These
periods should be short, of course.  But saying that C-g must always
work is factually in correct.



^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: Lots of hangs Windows 10 26.0.91
  2018-03-30  7:42   ` Eli Zaretskii
@ 2018-03-30 12:39     ` Kevin Buchs
  0 siblings, 0 replies; 6+ messages in thread
From: Kevin Buchs @ 2018-03-30 12:39 UTC (permalink / raw
  To: Eli Zaretskii; +Cc: help-gnu-emacs

I have certainly experienced similar problems on Windows. It was always
accompanied by a massive CPU consumption by Emacs. So, something got into a
tight loop. It has maybe happened 3-4 times, since Win 10 came out, so not
the frequency the OP suggested. I'm sure it has already been suggested that
you start up with --no-init-file, and if the problem vanishes, start
digging into what your init file is loading/executing to determine the
issue.

On Fri, Mar 30, 2018 at 2:42 AM, Eli Zaretskii <eliz@gnu.org> wrote:

> > From: phillip.lord@russet.org.uk (Phillip Lord)
> > Date: Thu, 29 Mar 2018 21:27:01 +0100
> > Cc: help-gnu-emacs@gnu.org
> >
> > > C-g does not work. I have to kill emacs from the task manager.
> >
> > This is clearly a bug -- C-g should always work.
>
> Unfortunately, due to how Emacs input works on Windows, there are
> legitimate situations where C-g doesn't (and cannot) work.  These
> periods should be short, of course.  But saying that C-g must always
> work is factually in correct.
>
>


^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2018-03-30 12:39 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2018-03-28 14:18 Lots of hangs Windows 10 26.0.91 Adam Taylor
2018-03-29 12:08 ` Eli Zaretskii
2018-03-29 20:27 ` Phillip Lord
2018-03-30  7:42   ` Eli Zaretskii
2018-03-30 12:39     ` Kevin Buchs
2018-03-29 21:51 ` Stefan Monnier

Code repositories for project(s) associated with this external index

	https://git.savannah.gnu.org/cgit/emacs.git
	https://git.savannah.gnu.org/cgit/emacs/org-mode.git

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.