[Mac-emacs-users] 'Fatal error (6).Abort'

[Michael P. Wilson]

Heya ppl!


I'm having a bit of trouble and was hoping someone could point me in
the right direction to go track it down meself.

I pulled the cvs current a couple days ago, configured, built &
installed the code, but execution give me the very descriptive error
"Fatal error (6).Abort".

Is there a com-line option to do some kind of full trace on startup
that I can pipe off to a file to figure out what's going on?

I'm pretty used to posix-based systems, but OS-X is different enough
in some strange places that it's throwing me for a loop.

TIA,

Mike

Point me to TFM that I may R it.

[Amos]

On Wed, 7 Aug 2002 13:42:51 -0400,

mpw> I pulled the cvs current a couple days ago, configured, built &
mpw> installed the code, but execution give me the very descriptive error
mpw> "Fatal error (6).Abort".

Did you install the latest Mac OS X security update since the CVS
build? I noticed something quite similar after installing that OS
X update. My guess at the time was that Emacs didn't care for the
change in libraries. So I figured it was time to suck down another
CVS update and redo the build. Things have been peachy since....

[Michael P. Wilson]

I think Steve's response has got it nailed. I wasn't executing with "-nw".

The last update I pulled for OSX was a few days before anyway. Sure
does sound like the type of thing emacs would bellyache over ;)

Post by Amos

On Wed, 7 Aug 2002 13:42:51 -0400,

mpw> I pulled the cvs current a couple days ago, configured, built &
mpw> installed the code, but execution give me the very descriptive error
mpw> "Fatal error (6).Abort".
Did you install the latest Mac OS X security update since the CVS
build? I noticed something quite similar after installing that OS
X update. My guess at the time was that Emacs didn't care for the
change in libraries. So I figured it was time to suck down another
CVS update and redo the build. Things have been peachy since....
--
Amos
-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
Mac-emacs-users mailing list
https://lists.sourceforge.net/lists/listinfo/mac-emacs-users

[Steven Tamm]

I hit the exact same problem. After the security update I had to
recompile from scratch or else I got Fatal Error (10). I assumed -nw
was the problem (probably falsely), but hopefully the patch I posted to
the mac-emacs-devel list will fix that.

WRT, the security update, I think the problem might be in the whole
unexec process. That is where a preliminary emacs (just compiled C
code) reads in a bunch of lisp and then spits itself out as an
executable (so I doesn't take too long to load up). I have found that
when I try to run update_prebinding, it complains that it cannot update
Emacs. Since update_prebinding is what the installer does when it says
"Optimizing" I assumed that unexec might not be spitting out the right
prebinding. However, older versions of emacs I had lying around seemed
to still work after the update so I could be barking up the wrong tree.

Maybe the author of the unexmacosx.c code has an idea?

-Steven

Post by Amos

On Wed, 7 Aug 2002 13:42:51 -0400,

mpw> I pulled the cvs current a couple days ago, configured, built &
mpw> installed the code, but execution give me the very descriptive
error
mpw> "Fatal error (6).Abort".
Did you install the latest Mac OS X security update since the CVS
build? I noticed something quite similar after installing that OS
X update. My guess at the time was that Emacs didn't care for the
change in libraries. So I figured it was time to suck down another
CVS update and redo the build. Things have been peachy since....
--
Amos
-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
Mac-emacs-users mailing list
https://lists.sourceforge.net/lists/listinfo/mac-emacs-users

[Andrew Choi]

[...] WRT, the security update, I think the problem might be in the
whole unexec process. That is where a preliminary emacs (just
compiled C code) reads in a bunch of lisp and then spits itself out
as an executable (so I doesn't take too long to load up). I have
found that when I try to run update_prebinding, it complains that it
cannot update Emacs. Since update_prebinding is what the installer
does when it says "Optimizing" I assumed that unexec might not be
spitting out the right prebinding. However, older versions of emacs
I had lying around seemed to still work after the update so I could
be barking up the wrong tree.
Maybe the author of the unexmacosx.c code has an idea?
-Steven

Hi Steven,

I wrote unexmacosx.c but I don't know if the problem originated there.

When I installed the Security Update, I didn't notice it running
update_prebinding! That would certainly cause the problem, because it
seems to change a few libraries.

If other people experience the same problem, perhaps they can try
running

update_prebinding -root /

manually. Otherwise, rebuilding Emacs will certainly make it work
again.

Andrew.

[Michael P. Wilson]

This is getting interesting :) Glad I said something.

Post by Andrew Choi

[...] WRT, the security update, I think the problem might be in the
whole unexec process. That is where a preliminary emacs (just
compiled C code) reads in a bunch of lisp and then spits itself out
as an executable (so I doesn't take too long to load up). I have
found that when I try to run update_prebinding, it complains that it
cannot update Emacs. Since update_prebinding is what the installer
does when it says "Optimizing" I assumed that unexec might not be
spitting out the right prebinding. However, older versions of emacs
I had lying around seemed to still work after the update so I could
be barking up the wrong tree.
Maybe the author of the unexmacosx.c code has an idea?
-Steven

Hi Steven,
I wrote unexmacosx.c but I don't know if the problem originated there.
When I installed the Security Update, I didn't notice it running
update_prebinding! That would certainly cause the problem, because it
seems to change a few libraries.
If other people experience the same problem, perhaps they can try
running
update_prebinding -root /
manually. Otherwise, rebuilding Emacs will certainly make it work
again.
Andrew.
-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
Mac-emacs-users mailing list
https://lists.sourceforge.net/lists/listinfo/mac-emacs-users

[Jeffrey Poore]

Post by Michael P. Wilson
Heya ppl!
I'm having a bit of trouble and was hoping someone could point me in
the right direction to go track it down meself.
I pulled the cvs current a couple days ago, configured, built &
installed the code, but execution give me the very descriptive error
"Fatal error (6).Abort".

I started having Emacs crash on me on startup (if I run from the command
line, I get a Fatal Error - illegal instruction). I don't think anything in
the application changed, so it might have been some security update from
Apple or something. I re-downloaded the source, built it over again from
scratch, and it seems to work now... You might try building again...

-Jeffrey

[William McCallum]

In addition to the problem every one else describes, which I fixed by
downloading emacs from cvs and recompiling, I have another new problem:
my auctex command to bring up an xdvi window after texing a file no
longer works. It claims to run xdvi, but the window never appears.

My command includes the -sourceposition option (to make xdvi go to the
place in the dvi file corresponding to the current cursor position in
the emacs file). If I eliminate that option, then the problem goes away;
the xdvi window appears as usual. Stranger, if I subsequently, in the
same session, ask emacs to execute the original command with
-sourceposition, it asks me if I want to kill the current xdvi, and then
the original command miraculously starts working again.

Regards,
Bill McCallum