update: building emacs-21.2

update: building emacs-21.2

[Jochen Küpper]

This is a copy of an article posted on gnu.emacs.help. I get the same
error on multiple systems with RedHat-7.0 and RedHat-7.1 updated to
gcc-3.2.x.

On 24 Aug 2002 12:50:37 -0400 Jochen Küpper wrote:

Jochen> temacs is succesfully build:
Jochen> ,----[ldd src/temacs]
Jochen> |         libXaw3d.so.7 => /usr/X11R6/lib/libXaw3d.so.7 (0x40018000)
Jochen> |         libXmu.so.6 => /usr/X11R6/lib/libXmu.so.6 (0x40069000)
Jochen> |         libXt.so.6 => /usr/X11R6/lib/libXt.so.6 (0x4007f000)
Jochen> |         libSM.so.6 => /usr/X11R6/lib/libSM.so.6 (0x400cc000)
Jochen> |         libICE.so.6 => /usr/X11R6/lib/libICE.so.6 (0x400d5000)
Jochen> |         libXext.so.6 => /usr/X11R6/lib/libXext.so.6 (0x400ed000)
Jochen> |         libtiff.so.3 => /usr/lib/libtiff.so.3 (0x400fb000)
Jochen> |         libjpeg.so.62 => /usr/lib/libjpeg.so.62 (0x4013e000)
Jochen> |         libpng.so.2 => /usr/lib/libpng.so.2 (0x4015d000)
Jochen> |         libz.so.1 => /usr/lib/libz.so.1 (0x4017e000)
Jochen> |         libm.so.6 => /lib/libm.so.6 (0x4018c000)
Jochen> |         libungif.so.4 => /usr/lib/libungif.so.4 (0x401b0000)
Jochen> |         libXpm.so.4 => /usr/X11R6/lib/libXpm.so.4 (0x401b8000)
Jochen> |         libX11.so.6 => /usr/X11R6/lib/libX11.so.6 (0x401c7000)
Jochen> |         libncurses.so.5 => /usr/lib/libncurses.so.5 (0x40295000)
Jochen> |         libc.so.6 => /lib/libc.so.6 (0x402d7000)
Jochen> |         /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)
Jochen> `----
Jochen> but doesn't do it's job:
Jochen> ,----[end of build-process]
Jochen> | LC_ALL=C ./temacs -batch -l loadup dump
Jochen> | Loading loadup (source)...
Jochen> | Using load-path (/tmp/emacs-21.2/lisp)
Jochen> | Loading byte-run...
Jochen> | Loading subr...
Jochen> | Loading version.el (source)...
Jochen> | Loading widget...
Jochen> | Loading custom...
Jochen> | Loading emacs-lisp/backquote...
Jochen> | Loading map-ynp...
Jochen> | Loading env...
Jochen> | Loading cus-start (source)...
Jochen> | Loading international/mule...
Jochen> | Loading international/mule-conf.el (source)...
Jochen> | Loading format...
Jochen> | Loading bindings...
Jochen> | Loading simple...
Jochen> | Loading files...
Jochen> | Lists of integers (garbage collection statistics) are normal output
Jochen> | while building Emacs; they do not indicate a problem.
Jochen> | ((7024 . 6016) (3780 . 0) (485 . 3) 11478 21356 (3 . 0) (16 . 0) (617 . 1714))
Jochen> | Loading loaddefs.el (source)...
Jochen> | Maximum buffer size exceeded
Jochen> | make[1]: *** [emacs] Error 255
Jochen> | make[1]: Leaving directory `/tmp/emacs-21.2/src'
Jochen> | make: *** [src] Error 2
Jochen> `----

Jochen> Searching google I found recent messages about this problem only where
Jochen> people actually tried to load *large* files.

Jochen> Here lisp/loaddefs.h is about 800kB (as for everybody else, I guess:))

Greetings,
Jochen
-- 
University of North Carolina                       phone: +1-919-962-4403
Department of Chemistry                            phone: +1-919-962-1579
Venable Hall CB#3290 (Kenan C148)                    fax: +1-919-843-6041
Chapel Hill, NC 27599, USA                            GnuPG key: 44BCCD8E

Re: update: building emacs-21.2

[Richard Stallman]

Since this does not fail for most everyone else, it probably depends
on some peculiar local situation.  Unfortunately that means
only you can debug it (or someone else who happens to observe
the same problem).  I suggest you try running the command
   temacs -batch -l loadup
under GDB with a breakpoint at Fsignal and try to debug it.

The workaround you found might be a clue, but unless someone gets a
brilliant inspiration from it and guesses the answer, the problem
still needs to be debugged.

Re: update: building emacs-21.2

[Jochen Küpper]

On Sun, 25 Aug 2002 18:36:49 -0600 Richard Stallman wrote:

Yep, to busy... I'll keep working on my schedule. Here's a quick
update.

Richard> Since this does not fail for most everyone else, it probably
Richard> depends on some peculiar local situation. Unfortunately that
Richard> means only you can debug it (or someone else who happens to
Richard> observe the same problem). I suggest you try running the
Richard> command temacs -batch -l loadup under GDB with a breakpoint
Richard> at Fsignal and try to debug it.

The problem seems to be related to gcc's -malign-double switch,
independent of optimization level. 

I had fired up a debugger and seen where we are coming from, but I
need some time to dig into emacs' source. Hopefully I'll have some
passion later this month to work on it (I should have, and I am
planning to do it.).

Greetings,
Jochen
-- 
Einigkeit und Recht und Freiheit                http://www.Jochen-Kuepper.de
    Liberté, Égalité, Fraternité                GnuPG key: 44BCCD8E
        Sex, drugs and rock-n-roll

Re: update: building emacs-21.2

[Andreas Schwab]

"Jochen Küpper" <jochen@unc.edu> writes:

|> On Sun, 25 Aug 2002 18:36:49 -0600 Richard Stallman wrote:
|> 
|> Yep, to busy... I'll keep working on my schedule. Here's a quick
|> update.
|> 
|> Richard> Since this does not fail for most everyone else, it probably
|> Richard> depends on some peculiar local situation. Unfortunately that
|> Richard> means only you can debug it (or someone else who happens to
|> Richard> observe the same problem). I suggest you try running the
|> Richard> command temacs -batch -l loadup under GDB with a breakpoint
|> Richard> at Fsignal and try to debug it.
|> 
|> The problem seems to be related to gcc's -malign-double switch,
|> independent of optimization level. 

Don't use -malign-double, it changes the ABI.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux AG, Deutschherrnstr. 15-19, D-90429 Nürnberg
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."