bug#23577: 24.5; emacs_backtrace.txt

bug#23577: 24.5; emacs_backtrace.txt

[Drew Adams]

Exception 0xc0000005 at this address:
75217786

Backtrace:
011d988d
011d9902
010d965c
010f8726
010f86fe
010f875b
01242f31
011d97fe
771703c7
77865b63
778298d1

Exception 0xc0000005 at this address:
0118fe3e

Backtrace:
01151828




In GNU Emacs 24.5.1 (i686-pc-mingw32)
 of 2015-04-11 on LEG570
Windowing system distributor `Microsoft Corp.', version 6.1.7601
Configured using:
 `configure --prefix=/c/usr --host=i686-pc-mingw32'

bug#23577: 24.5; emacs_backtrace.txt

[Eli Zaretskii]

> Date: Wed, 18 May 2016 12:38:32 -0700 (PDT)
> From: Drew Adams <drew.adams@oracle.com>
> 
> 
> 
> Exception 0xc0000005 at this address:
> 75217786
> 
> Backtrace:
> 011d988d
> 011d9902
> 010d965c
> 010f8726
> 010f86fe
> 010f875b
> 01242f31
> 011d97fe
> 771703c7
> 77865b63
> 778298d1
> 
> Exception 0xc0000005 at this address:
> 0118fe3e
> 
> Backtrace:
> 01151828

Can someone who has this binary:

> In GNU Emacs 24.5.1 (i686-pc-mingw32)
>  of 2015-04-11 on LEG570
> Windowing system distributor `Microsoft Corp.', version 6.1.7601
> Configured using:
>  `configure --prefix=/c/usr --host=i686-pc-mingw32'

and also addr2line please produce a human-readable backtrace from
these addresses?  Otherwise, they are useless.

bug#23577: 24.5; emacs_backtrace.txt

[John Wiegley]

[-- Attachment #1: Type: text/plain, Size: 698 bytes --]

>>>>> Drew Adams <drew.adams@oracle.com> writes:

> Exception 0xc0000005 at this address:
> 75217786

> Backtrace:
> 011d988d
> 011d9902
> 010d965c
> 010f8726
> 010f86fe
> 010f875b
> 01242f31
> 011d97fe
> 771703c7
> 77865b63
> 778298d1

> Exception 0xc0000005 at this address:
> 0118fe3e

> Backtrace:
> 01151828

Hi Drew,

It would actually be better to not create bug reports that look like this. I'm
not sure any of them have ever resulted in positive action, since no one can
make heads or tails of the trace.

-- 
John Wiegley                  GPG fingerprint = 4710 CF98 AF9B 327B B80F
http://newartisans.com                          60E1 46C4 BD1A 7AC1 4BA2

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 629 bytes --]

bug#23577: 24.5; emacs_backtrace.txt

[Drew Adams]

> It would actually be better to not create bug reports that look like this.
> I'm
> not sure any of them have ever resulted in positive action, since no one can
> make heads or tails of the trace.

Fine.  I'll be glad to oblige.  Can you define "look like this"?

You're not sure that any of them have ever resulted in
positive action, but it is definitely the case that several
of the emacs_backtrace.txt reports I've filed have been resolved,
thanks to the info in the backtrace files.  (None recently, it
is true.)  All of those that were resolved also looked, to me,
"like this".  But perhaps not to the informed eye.

I guess I'll just not send any emacs_backtrace.txt reports in
the future.  Is there a way for me to prevent Emacs from
writing such useless files, and return to the way things were
before Emacs started cleverly writing them?  I certainly have
no need for them, if they are not helpful to report.  How to
turn off this "feature"?

Better yet would be for Emacs to write only useful ones, if
possible.  If it cannot write something useful, why write
anything?

Eli calls out to whoever built the Emacs binary to provide
extra information, to help decipher the backtrace.  Juanma
did that in the past, for his binaries.  But in this case
the binary was built and distributed by GNU Emacs, I believe
- it is the 24.5.1 release, AFAIK.

I no longer see releases in directory /gnu/emacs/windows/,
however.  All that is there now are Emacs 25 pretests.
(And do these pretest binaries have the magic sauce that
would make their emacs_backtrace files useful?  I don't
think so.)

(And BTW, why is there a windows/ directory of pretests at
the top level there, in addition to there being a windows/
directory under each of the other directories, pretest/ and
pretests/?)

bug#23577: 24.5; emacs_backtrace.txt

[Eli Zaretskii]

> Date: Wed, 18 May 2016 22:47:53 -0700 (PDT)
> From: Drew Adams <drew.adams@oracle.com>
> Cc: 23577@debbugs.gnu.org
> 
> Is there a way for me to prevent Emacs from writing such useless
> files, and return to the way things were before Emacs started
> cleverly writing them?

No.

bug#23577:

[Glenn Morris]

Nothing can be done with this information.

bug#23577: 24.5; emacs_backtrace.txt

[Drew Adams]

> Can someone who has this binary:
> 
> > In GNU Emacs 24.5.1 (i686-pc-mingw32)
> >  of 2015-04-11 on LEG570
> > Windowing system distributor `Microsoft Corp.', version 6.1.7601
> > Configured using:
> >  `configure --prefix=/c/usr --host=i686-pc-mingw32'
> 
> and also addr2line please produce a human-readable backtrace from
> these addresses?

FWIW, it's the Emacs 24.5 release.

bug#23577: 24.5; emacs_backtrace.txt

[Eli Zaretskii]

> Date: Wed, 18 May 2016 14:08:26 -0700 (PDT)
> From: Drew Adams <drew.adams@oracle.com>
> Cc: 23577@debbugs.gnu.org
> 
> > Can someone who has this binary:
> > 
> > > In GNU Emacs 24.5.1 (i686-pc-mingw32)
> > >  of 2015-04-11 on LEG570
> > > Windowing system distributor `Microsoft Corp.', version 6.1.7601
> > > Configured using:
> > >  `configure --prefix=/c/usr --host=i686-pc-mingw32'
> > 
> > and also addr2line please produce a human-readable backtrace from
> > these addresses?
> 
> FWIW, it's the Emacs 24.5 release.

Where can I find it?

bug#23577: 24.5; emacs_backtrace.txt

[Nicolas Richard]

>> and also addr2line please produce a human-readable backtrace from
>> these addresses?
>
> FWIW, it's the Emacs 24.5 release.

I grabbed
http://ftp.gnu.org/gnu/emacs/windows/emacs-24.5-bin-i686-mingw32.zip
unzipped it, and tried various invocations of addr2line, but I only get
"??" in return.

Sample output:
$ addr2line -a -p -e emacs.exe < ~/tmp/bt
0x011d988d: ??:0
0x011d9902: ??:0
0x010d965c: ??:0
0x010f8726: ??:0
0x010f86fe: ??:0
0x010f875b: ??:0
0x01242f31: ??:0
0x011d97fe: ??:0
0x771703c7: ??:0
0x77865b63: ??:0
0x778298d1: ??:0

I'm on GNU/Linux -- does that explain it ?

-- 
Nicolas

bug#23577: 24.5; emacs_backtrace.txt

[Eli Zaretskii]

> From: Nicolas Richard <youngfrog@members.fsf.org>
> Cc: Eli Zaretskii <eliz@gnu.org>,  23577@debbugs.gnu.org
> Date: Thu, 19 May 2016 12:16:26 +0200
> 
> I grabbed
> http://ftp.gnu.org/gnu/emacs/windows/emacs-24.5-bin-i686-mingw32.zip
> unzipped it, and tried various invocations of addr2line, but I only get
> "??" in return.
> 
> Sample output:
> $ addr2line -a -p -e emacs.exe < ~/tmp/bt
> 0x011d988d: ??:0
> 0x011d9902: ??:0
> 0x010d965c: ??:0
> 0x010f8726: ??:0
> 0x010f86fe: ??:0
> 0x010f875b: ??:0
> 0x01242f31: ??:0
> 0x011d97fe: ??:0
> 0x771703c7: ??:0
> 0x77865b63: ??:0
> 0x778298d1: ??:0
> 
> I'm on GNU/Linux -- does that explain it ?

It could be stripped.

bug#23577: 24.5; emacs_backtrace.txt

[Drew Adams]

> > FWIW, it's the Emacs 24.5 release.
> 
> Where can I find it?

Sorry, I don't know, at this point.  I thought that I
picked it up from GNU - e.g. here:
https://ftp.gnu.org/gnu/emacs/windows/

Or perhaps here (but I guess not, as this one is I guess
only for pretests): http://alpha.gnu.org/gnu/emacs/windows/).

If not, sorry, I don't know.  If the info in the bug
report (e.g. "2015-04-11 on LEG570") doesn't identify it
sufficiently then too bad, I guess.

(You can see by the version number that it is a release and
not a dev version, at least.)

---

For comparison, here's the info for release 23.4, which I'm
pretty sure I got from GNU Emacs:

In GNU Emacs 23.4.1 (i386-mingw-nt6.1.7601)
 of 2012-02-04 on MARVIN
Windowing system distributor `Microsoft Corp.', version 6.1.7601
configured using `configure --with-gcc (4.4) --cflags -ID:/devel/emacs/libs/libXpm-3.5.8/include -ID:/devel/emacs/libs/libXpm-3.5.8/src -ID:/devel/emacs/libs/libpng-dev_1.4.3-1/include -ID:/devel/emacs/libs/zlib-dev_1.2.5-2/include -ID:/devel/emacs/libs/giflib-4.1.4-1/include -ID:/devel/emacs/libs/jpeg-6b-4/include -ID:/devel/emacs/libs/tiff-3.8.2-1/include'

bug#23577: 24.5; emacs_backtrace.txt

[Nicolas Richard]

Drew Adams <drew.adams@oracle.com> writes:

>> > FWIW, it's the Emacs 24.5 release.
>> 
>> Where can I find it?
>
> Sorry, I don't know, at this point.  I thought that I
> picked it up from GNU - e.g. here:
> https://ftp.gnu.org/gnu/emacs/windows/

I mentionned it elsewhere, but it is probably this one:
http://ftp.gnu.org/gnu/emacs/windows/emacs-24.5-bin-i686-mingw32.zip

After unzipping it, I see:
    $ strings bin/emacs.exe | grep LEG
    $Id: GNU Emacs 24.5 (i686-pc-mingw32)  of 2015-04-11 on LEG570 $
which matches your version string.

Unfortunately it seems to have been stripped (which might be considered
normal for a release ?), so I guess the backtrace is useless.

    $ file bin/emacs.exe 
    bin/emacs.exe: PE32 executable (console) Intel 80386 (stripped to external PDB), for MS Windows

-- 
Nicolas

bug#23577: 24.5; emacs_backtrace.txt

[Drew Adams]

> >> > FWIW, it's the Emacs 24.5 release.
> >>
> >> Where can I find it?
> >
> > Sorry, I don't know, at this point.  I thought that I
> > picked it up from GNU - e.g. here:
> > https://ftp.gnu.org/gnu/emacs/windows/
> 
> I mentionned it elsewhere, but it is probably this one:
> http://ftp.gnu.org/gnu/emacs/windows/emacs-24.5-bin-i686-mingw32.zip
> 
> After unzipping it, I see:
>     $ strings bin/emacs.exe | grep LEG
>     $Id: GNU Emacs 24.5 (i686-pc-mingw32)  of 2015-04-11 on LEG570 $
> which matches your version string.
> 
> Unfortunately it seems to have been stripped (which might be considered
> normal for a release ?), so I guess the backtrace is useless.
> 
>     $ file bin/emacs.exe
>     bin/emacs.exe: PE32 executable (console) Intel 80386 (stripped to
>     external PDB), for MS Windows

Thanks for tracking it down, Nicholas.