bug#2956: 23.0.92; anything.el causes BSOD on vista64 dual-processor

bug#2956: 23.0.92; anything.el causes BSOD on vista64 dual-processor

[Erik Charlebois]

[-- Attachment #1: Type: text/plain, Size: 2475 bytes --]

Running "M-x anything" with anything.el, then pressing tab and enter
causes Vista 64 to BSOD with a "Second processor failed to respond to
clock interrupt in time" error. I am running on a dual quad-core system.

I have hit this problem in ERC as well, so I know it is not related
entirely to anything.el. It is simply easier to reproduce consistently
on anything.el. This problem does not occur with 22.3.

(As a software developer myself, I understand how ridiculously esoteric
this bug is and that the OS shouldn't panic from a user-land process. I
understand if this doesn't get fixed. I thought I'd report it anyway.
If I can help provide more information (maybe there is some triage of
version 23 features I could do, given the option names), feel free to
ask (erikcharlebois@gmail.com).)


In GNU Emacs 23.0.92.1 (i386-mingw-nt6.0.6001)
 of 2009-03-30 on SOFT-MJASON
Windowing system distributor `Microsoft Corp.', version 6.0.6001
configured using `configure --with-gcc (3.4)'

Important settings:
  value of $LC_ALL: nil
  value of $LC_COLLATE: nil
  value of $LC_CTYPE: nil
  value of $LC_MESSAGES: nil
  value of $LC_MONETARY: nil
  value of $LC_NUMERIC: nil
  value of $LC_TIME: nil
  value of $LANG: ENU
  value of $XMODIFIERS: nil
  locale-coding-system: cp1252
  default-enable-multibyte-characters: t

Major mode: Lisp Interaction

Minor modes in effect:
  yas/minor-mode: t
  desktop-save-mode: t
  savehist-mode: t
  show-paren-mode: t
  mouse-wheel-mode: t
  file-name-shadow-mode: t
  global-font-lock-mode: t
  font-lock-mode: t
  global-auto-composition-mode: t
  auto-composition-mode: t
  auto-encryption-mode: t
  auto-compression-mode: t
  column-number-mode: t
  line-number-mode: t
  transient-mark-mode: t

Recent input:
M-x g C-g M-x e m a c s - <backspace> <backspace> <backspace>
<backspace> <backspace> <backspace> r e p o r t - e
m <tab> <return>

Recent messages:
Building completion list of all manual topics...
Loading c:/Users/Erik Charlebois/.emacs.d/kuler-colors.el (source)...done
Loading c:/Users/Erik Charlebois/.emacs.d/cedet-1.0pre6/common/cedet.el
(source)...
Outdated speedbar 1.0 shadowed to meet minimum version 1.0.2
Setting up CEDET packages...
Loading `dframe': old-style backquotes detected!
Setting up CEDET packages...done
Loading c:/Users/Erik Charlebois/.emacs.d/cedet-1.0pre6/common/cedet.el
(source)...done
No desktop file.
For information about GNU Emacs and the GNU system, type C-h C-a.
Quit

-- 
Erik Charlebois

[-- Attachment #2: Type: text/html, Size: 3236 bytes --]

bug#2956: 23.0.92; anything.el causes BSOD on vista64 dual-processor

[Stefan Monnier]

> Running "M-x anything" with anything.el, then pressing tab and enter
> causes Vista 64 to BSOD with a "Second processor failed to respond to
> clock interrupt in time" error. I am running on a dual quad-core system.

That's clearly a bug in Vista rather than in Emacs, so there's nothing
for us to fix.
A good fix for you would be to upgrade to a Free operating system.


        Stefan

bug#2956: 23.0.92; anything.el causes BSOD on vista64 dual-processor

[Eli Zaretskii]

> From: Stefan Monnier <monnier@iro.umontreal.ca>
> Date: Sat, 11 Apr 2009 08:55:32 -0400
> Cc: 2956@emacsbugs.donarmstrong.com
> 
> > Running "M-x anything" with anything.el, then pressing tab and enter
> > causes Vista 64 to BSOD with a "Second processor failed to respond to
> > clock interrupt in time" error. I am running on a dual quad-core system.
> 
> That's clearly a bug in Vista rather than in Emacs, so there's nothing
> for us to fix.

The fact (if it is a fact) that a problem manifests itself only on
Vista does not necessarily mean there's no bug in Emacs, especially
since the variety of different systems on which Emacs is routinely
tested is so small.  Some positive evidence is required to deduce that
the bug is in the underlying OS.  Do you have such an evidence?

bug#2956: 23.0.92; anything.el causes BSOD on vista64 dual-processor

[Andreas Schwab]

Eli Zaretskii <eliz@gnu.org> writes:

> The fact (if it is a fact) that a problem manifests itself only on
> Vista does not necessarily mean there's no bug in Emacs, especially
> since the variety of different systems on which Emacs is routinely
> tested is so small.  Some positive evidence is required to deduce that
> the bug is in the underlying OS.  Do you have such an evidence?

An unprivileged process able to crash the OS is always a bug in the OS.
This is independent of whether there is a workaround.

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."

bug#2956: 23.0.92; anything.el causes BSOD on vista64 dual-processor

[Eli Zaretskii]

> From: Andreas Schwab <schwab@linux-m68k.org>
> Cc: 2956@emacsbugs.donarmstrong.com,  Stefan Monnier <monnier@iro.umontreal.ca>,  erikcharlebois@gmail.com
> Date: Sat, 11 Apr 2009 16:39:00 +0200
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> > The fact (if it is a fact) that a problem manifests itself only on
> > Vista does not necessarily mean there's no bug in Emacs, especially
> > since the variety of different systems on which Emacs is routinely
> > tested is so small.  Some positive evidence is required to deduce that
> > the bug is in the underlying OS.  Do you have such an evidence?
> 
> An unprivileged process able to crash the OS is always a bug in the OS.

Please re-read what I wrote: I also said that this doesn't necessarily
mean there's no bug in Emacs.  And when I wrote about some positive
evidence that the bug is in the OS, I really meant to say ``the bug is
ONLY in the OS''.

We don't care about Vista bugs here, but I think we do care about
Emacs.

bug#2956: 23.0.92; anything.el causes BSOD on vista64 dual-processor

[Andreas Schwab]

Eli Zaretskii <eliz@gnu.org> writes:

>> From: Andreas Schwab <schwab@linux-m68k.org>
>> Cc: 2956@emacsbugs.donarmstrong.com,  Stefan Monnier <monnier@iro.umontreal.ca>,  erikcharlebois@gmail.com
>> Date: Sat, 11 Apr 2009 16:39:00 +0200
>> 
>> Eli Zaretskii <eliz@gnu.org> writes:
>> 
>> > The fact (if it is a fact) that a problem manifests itself only on
>> > Vista does not necessarily mean there's no bug in Emacs, especially
>> > since the variety of different systems on which Emacs is routinely
>> > tested is so small.  Some positive evidence is required to deduce that
>> > the bug is in the underlying OS.  Do you have such an evidence?
>> 
>> An unprivileged process able to crash the OS is always a bug in the OS.
>
> Please re-read what I wrote: I also said that this doesn't necessarily
> mean there's no bug in Emacs.

Please re-read what I wrote: I also said that this is independent of
whether there is a workaround.  An unprivileged process must _never_ be
able to crash the OS.

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."

bug#2956: 23.0.92; anything.el causes BSOD on vista64 dual-processor

[Lennart Borgman]

On Sat, Apr 11, 2009 at 6:28 PM, Andreas Schwab <schwab@linux-m68k.org> wrote:
> Please re-read what I wrote: I also said that this is independent of
> whether there is a workaround.  An unprivileged process must _never_ be
> able to crash the OS.

Excuse me, I did not follow, but was the Emacs process unprivileged?

bug#2956: 23.0.92; anything.el causes BSOD on vista64 dual-processor

[Eli Zaretskii]

> From: Andreas Schwab <schwab@linux-m68k.org>
> Cc: 2956@emacsbugs.donarmstrong.com,  monnier@iro.umontreal.ca,  erikcharlebois@gmail.com
> Date: Sat, 11 Apr 2009 18:28:32 +0200
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> An unprivileged process must _never_ be able to crash the OS.

Right, but that's irrelevant to the Emacs bug tracker.

bug#2956: marked as done (23.0.92; anything.el causes BSOD on vista64 dual-processor)

[Emacs bug Tracking System]

[-- Attachment #1: Type: text/plain, Size: 933 bytes --]


Your message dated Sat, 11 Apr 2009 14:31:26 -0400
with message-id <jwvprfjyrlk.fsf-monnier+emacsbugreports@gnu.org>
and subject line Re: bug#2956: 23.0.92; anything.el causes BSOD on vista64 dual-processor
has caused the Emacs bug report #2956,
regarding 23.0.92; anything.el causes BSOD on vista64 dual-processor
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact owner@emacsbugs.donarmstrong.com
immediately.)


-- 
2956: http://emacsbugs.donarmstrong.com/cgi-bin/bugreport.cgi?bug=2956
Emacs Bug Tracking System
Contact owner@emacsbugs.donarmstrong.com with problems

[-- Attachment #2: Type: message/rfc822, Size: 8963 bytes --]

[-- Attachment #2.1.1: Type: text/plain, Size: 2475 bytes --]

Running "M-x anything" with anything.el, then pressing tab and enter
causes Vista 64 to BSOD with a "Second processor failed to respond to
clock interrupt in time" error. I am running on a dual quad-core system.

I have hit this problem in ERC as well, so I know it is not related
entirely to anything.el. It is simply easier to reproduce consistently
on anything.el. This problem does not occur with 22.3.

(As a software developer myself, I understand how ridiculously esoteric
this bug is and that the OS shouldn't panic from a user-land process. I
understand if this doesn't get fixed. I thought I'd report it anyway.
If I can help provide more information (maybe there is some triage of
version 23 features I could do, given the option names), feel free to
ask (erikcharlebois@gmail.com).)


In GNU Emacs 23.0.92.1 (i386-mingw-nt6.0.6001)
 of 2009-03-30 on SOFT-MJASON
Windowing system distributor `Microsoft Corp.', version 6.0.6001
configured using `configure --with-gcc (3.4)'

Important settings:
  value of $LC_ALL: nil
  value of $LC_COLLATE: nil
  value of $LC_CTYPE: nil
  value of $LC_MESSAGES: nil
  value of $LC_MONETARY: nil
  value of $LC_NUMERIC: nil
  value of $LC_TIME: nil
  value of $LANG: ENU
  value of $XMODIFIERS: nil
  locale-coding-system: cp1252
  default-enable-multibyte-characters: t

Major mode: Lisp Interaction

Minor modes in effect:
  yas/minor-mode: t
  desktop-save-mode: t
  savehist-mode: t
  show-paren-mode: t
  mouse-wheel-mode: t
  file-name-shadow-mode: t
  global-font-lock-mode: t
  font-lock-mode: t
  global-auto-composition-mode: t
  auto-composition-mode: t
  auto-encryption-mode: t
  auto-compression-mode: t
  column-number-mode: t
  line-number-mode: t
  transient-mark-mode: t

Recent input:
M-x g C-g M-x e m a c s - <backspace> <backspace> <backspace>
<backspace> <backspace> <backspace> r e p o r t - e
m <tab> <return>

Recent messages:
Building completion list of all manual topics...
Loading c:/Users/Erik Charlebois/.emacs.d/kuler-colors.el (source)...done
Loading c:/Users/Erik Charlebois/.emacs.d/cedet-1.0pre6/common/cedet.el
(source)...
Outdated speedbar 1.0 shadowed to meet minimum version 1.0.2
Setting up CEDET packages...
Loading `dframe': old-style backquotes detected!
Setting up CEDET packages...done
Loading c:/Users/Erik Charlebois/.emacs.d/cedet-1.0pre6/common/cedet.el
(source)...done
No desktop file.
For information about GNU Emacs and the GNU system, type C-h C-a.
Quit

-- 
Erik Charlebois

[-- Attachment #2.1.2: Type: text/html, Size: 3236 bytes --]

[-- Attachment #3: Type: message/rfc822, Size: 2982 bytes --]

From: Stefan Monnier <monnier@iro.umontreal.ca>
To: Eli Zaretskii <eliz@gnu.org>
Cc: 2956-done@emacsbugs.donarmstrong.com
Subject: Re: bug#2956: 23.0.92; anything.el causes BSOD on vista64 dual-processor
Date: Sat, 11 Apr 2009 14:31:26 -0400
Message-ID: <jwvprfjyrlk.fsf-monnier+emacsbugreports@gnu.org>

>> > The fact (if it is a fact) that a problem manifests itself only on
>> > Vista does not necessarily mean there's no bug in Emacs, especially
>> > since the variety of different systems on which Emacs is routinely
>> > tested is so small.  Some positive evidence is required to deduce that
>> > the bug is in the underlying OS.  Do you have such an evidence?
>> An unprivileged process able to crash the OS is always a bug in the OS.
> Please re-read what I wrote: I also said that this doesn't necessarily
> mean there's no bug in Emacs.

We know for a fact that there are many bugs in Emacs, so there's no risk
of me thinking that "there's no bug in Emacs" ;-)

> We don't care about Vista bugs here, but I think we do care about Emacs.

Indeed.  But he described a bug in Vista, which happens to appear to be
triggered by something in Emacs.  Since we don't know what is that thing
in Emacs that triggers the bug, we can't even decide if it's itself
a bug or not.  We have plenty of real confirmed bugs not to worry about
this one until Vista gets fixed.  And I already mentioned a good "fix".


        Stefan

bug#2956: 23.0.92; anything.el causes BSOD on vista64 dual-processor

[Eli Zaretskii]

> From: Stefan Monnier <monnier@iro.umontreal.ca>
> Cc: 2956-done@emacsbugs.donarmstrong.com
> Date: Sat, 11 Apr 2009 14:31:26 -0400
> 
> > We don't care about Vista bugs here, but I think we do care about Emacs.
> 
> Indeed.  But he described a bug in Vista, which happens to appear to be
> triggered by something in Emacs.

That something could very well be a bug.  So I think marking this bug
"done" is waaaay too far-fetched.