Re: [gritsch@iue.tuwien.ac.at: Severe hscroll-margin Bug]

Re: [gritsch@iue.tuwien.ac.at: Severe hscroll-margin Bug]

[Kim F. Storm]

Both incidents of the bug described below were observed on Windoze
(compiling with mingw).  I tried the recipe on X+GNU/Linux, but cannot
reproduce it.

Can one of the W32 maintainers pls try to reproduce it on their
system -- and possibly debug it to see why it loops.

Markus Gritsch <gritsch@iue.tuwien.ac.at> wrote:

I can now build Emacs from CVS and have done so on 2006-02-21 four days 
ago. I still get a frozen Emacs wich takes 100% CPU when I use the 
following .emacs file

(setq-default truncate-lines t)
(custom-set-variables
 '(hscroll-margin 20)
 '(hscroll-step 1)
)

and resize the Emacs frame to something quite narrow. If you would have 
another fix, I could probably try it out.

GNU Emacs 22.0.50.1 (i386-mingw-nt5.2.3790) of 2006-02-21 on MGROOVY



Markus also wrote:

When I reported the bug the first time some months ago, David Hunter was 
also able to reproduce the problem:
  http://lists.gnu.org/archive/html/emacs-devel/2005-07/msg00077.html

.. which says: ... on today's CVS, my mingw build, Windows XP SP2.

This does probably not help very much, but at least I am not the only 
one, who can reproduce this :)

> Does it also loop if you start emacs -Q -D and execute the above
> lines directly?

Yes.

> Can you pls. run emacs under gdb and try to interrupt it to see where
> it is looping.	 You can find instructions in etc/DEBUG.

MGroovy:/cygdrive/c/emacs_CVS/emacs/bin$ gdb emacs.exe
GNU gdb 6.3.50_2004-12-28-cvs (cygwin-special)
Copyright 2004 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain 
conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for details.
This GDB was configured as "i686-pc-cygwin"...
(gdb) run -Q -D
Starting program: /cygdrive/c/emacs_CVS/emacs/bin/emacs.exe -Q -D
Error: dll starting at 0x78d41000 not found.

Error: dll starting at 0x78d41000 not found.

Error: dll starting at 0x78c31000 not found.


Program received signal SIGINT, Interrupt.
[Switching to thread 4164.0x5b8]
0x7d542a49 in SetConsoleNlsMode ()
(gdb) where
#0  0x7d542a49 in SetConsoleNlsMode ()
#1  0x00000000 in ?? () from
(gdb) list
828	#ifdef VMS
829	, envp
830	#endif
831	)
832	     int argc;
833	     char **argv;
834	#ifdef VMS
835	     char **envp;
836	#endif
837	{
(gdb)



-- 
Kim F. Storm <storm@cua.dk> http://www.cua.dk

Re: [gritsch@iue.tuwien.ac.at: Severe hscroll-margin Bug]

[Markus Gritsch]

The Problem described below is still present in current CVS Emacs.

Kim F. Storm wrote:
> Both incidents of the bug described below were observed on Windoze
> (compiling with mingw).  I tried the recipe on X+GNU/Linux, but cannot
> reproduce it.
> 
> Can one of the W32 maintainers pls try to reproduce it on their
> system -- and possibly debug it to see why it loops.
> 
> Markus Gritsch <gritsch@iue.tuwien.ac.at> wrote:
> 
> I can now build Emacs from CVS and have done so on 2006-02-21 four days 
> ago. I still get a frozen Emacs wich takes 100% CPU when I use the 
> following .emacs file
> 
> (setq-default truncate-lines t)
> (custom-set-variables
>  '(hscroll-margin 20)
>  '(hscroll-step 1)
> )
> 
> and resize the Emacs frame to something quite narrow. If you would have 
> another fix, I could probably try it out.
> 
> GNU Emacs 22.0.50.1 (i386-mingw-nt5.2.3790) of 2006-02-21 on MGROOVY
> 
> 
> 
> Markus also wrote:
> 
> When I reported the bug the first time some months ago, David Hunter was 
> also able to reproduce the problem:
>   http://lists.gnu.org/archive/html/emacs-devel/2005-07/msg00077.html
> 
> .. which says: ... on today's CVS, my mingw build, Windows XP SP2.
> 
> This does probably not help very much, but at least I am not the only 
> one, who can reproduce this :)
> 
>> Does it also loop if you start emacs -Q -D and execute the above
>> lines directly?
> 
> Yes.
> 
>> Can you pls. run emacs under gdb and try to interrupt it to see where
>> it is looping.	 You can find instructions in etc/DEBUG.
> 
> MGroovy:/cygdrive/c/emacs_CVS/emacs/bin$ gdb emacs.exe
> GNU gdb 6.3.50_2004-12-28-cvs (cygwin-special)
> Copyright 2004 Free Software Foundation, Inc.
> GDB is free software, covered by the GNU General Public License, and you are
> welcome to change it and/or distribute copies of it under certain 
> conditions.
> Type "show copying" to see the conditions.
> There is absolutely no warranty for GDB.  Type "show warranty" for details.
> This GDB was configured as "i686-pc-cygwin"...
> (gdb) run -Q -D
> Starting program: /cygdrive/c/emacs_CVS/emacs/bin/emacs.exe -Q -D
> Error: dll starting at 0x78d41000 not found.
> 
> Error: dll starting at 0x78d41000 not found.
> 
> Error: dll starting at 0x78c31000 not found.
> 
> 
> Program received signal SIGINT, Interrupt.
> [Switching to thread 4164.0x5b8]
> 0x7d542a49 in SetConsoleNlsMode ()
> (gdb) where
> #0  0x7d542a49 in SetConsoleNlsMode ()
> #1  0x00000000 in ?? () from
> (gdb) list
> 828	#ifdef VMS
> 829	, envp
> 830	#endif
> 831	)
> 832	     int argc;
> 833	     char **argv;
> 834	#ifdef VMS
> 835	     char **envp;
> 836	#endif
> 837	{
> (gdb)

Re: [gritsch@iue.tuwien.ac.at: Severe hscroll-margin Bug]

[Eli Zaretskii]

> Date: Thu, 13 Apr 2006 07:27:11 +0200
> From: Markus Gritsch <gritsch@iue.tuwien.ac.at>
> Cc: emacs-devel@gnu.org
> 
> The Problem described below is still present in current CVS Emacs.
> 
> Kim F. Storm wrote:
> > Both incidents of the bug described below were observed on Windoze
> > (compiling with mingw).  I tried the recipe on X+GNU/Linux, but cannot
> > reproduce it.
> > 
> > Can one of the W32 maintainers pls try to reproduce it on their
> > system -- and possibly debug it to see why it loops.

I can reproduce this on my system.  I will try to see where it loops,
but it isn't easy on Windows (since interrupting a running program in
GDB does not work reliably).  Ideas and tricks are welcome.

> > Program received signal SIGINT, Interrupt.
> > [Switching to thread 4164.0x5b8]
> > 0x7d542a49 in SetConsoleNlsMode ()
> > (gdb) where
> > #0  0x7d542a49 in SetConsoleNlsMode ()
> > #1  0x00000000 in ?? () from

In particular, this backtrace is obviously useless.