bug#7907: 24.0.50; trunk build failure

bug#7907: 24.0.50; trunk build failure

[Sean Sieger]

On Windows 7:

gcc -I. -c -gdwarf-2 -g3  -DEMACSDEBUG -DENABLE_CHECKING -DXASSERTS -fno-crossju
mping  -Ic:/imagesupport/include -Demacs=1 -DHAVE_CONFIG_H -I../nt/inc -DHAVE_NT
GUI=1 -DUSE_CRT_DLL=1 -DPURESIZE=5000000 -o oo/i386/print.o print.c
print.c:53:21: error: ftoastr.h: No such file or directory
make[2]: *** [oo/i386/print.o] Error 1
make[2]: Leaving directory `c:/emacs-24.0.50/src'
make[1]: *** [bootstrap-temacs] Error 2
make[1]: Leaving directory `c:/emacs-24.0.50/src'
make: *** [bootstrap-gmake] Error 2

bug#7907: 24.0.50; trunk build failure

[Eli Zaretskii]

> Date: Mon, 24 Jan 2011 09:10:59 -0500
> From: Sean Sieger <sean.sieger@gmail.com>
> Cc: 
> 
> On Windows 7:
> 
> gcc -I. -c -gdwarf-2 -g3  -DEMACSDEBUG -DENABLE_CHECKING -DXASSERTS -fno-crossju
> mping  -Ic:/imagesupport/include -Demacs=1 -DHAVE_CONFIG_H -I../nt/inc -DHAVE_NT
> GUI=1 -DUSE_CRT_DLL=1 -DPURESIZE=5000000 -o oo/i386/print.o print.c
> print.c:53:21: error: ftoastr.h: No such file or directory
> make[2]: *** [oo/i386/print.o] Error 1
> make[2]: Leaving directory `c:/emacs-24.0.50/src'
> make[1]: *** [bootstrap-temacs] Error 2
> make[1]: Leaving directory `c:/emacs-24.0.50/src'
> make: *** [bootstrap-gmake] Error 2

The Windows build is broken by importing a few files from gnulib.  I
hope to have it fixed in a few days, but feel free to beat me to it.

bug#7907: 24.0.50; trunk build failure

[Eli Zaretskii]

> Date: Mon, 24 Jan 2011 09:10:59 -0500
> From: Sean Sieger <sean.sieger@gmail.com>
> Cc: 
> 
> On Windows 7:
> 
> gcc -I. -c -gdwarf-2 -g3  -DEMACSDEBUG -DENABLE_CHECKING -DXASSERTS -fno-crossju
> mping  -Ic:/imagesupport/include -Demacs=1 -DHAVE_CONFIG_H -I../nt/inc -DHAVE_NT
> GUI=1 -DUSE_CRT_DLL=1 -DPURESIZE=5000000 -o oo/i386/print.o print.c
> print.c:53:21: error: ftoastr.h: No such file or directory
> make[2]: *** [oo/i386/print.o] Error 1
> make[2]: Leaving directory `c:/emacs-24.0.50/src'
> make[1]: *** [bootstrap-temacs] Error 2
> make[1]: Leaving directory `c:/emacs-24.0.50/src'
> make: *** [bootstrap-gmake] Error 2

Should be fixed now (as of revno 103022).

bug#7907: 24.0.50; trunk build failure

[Sean Sieger]

Eli Zaretskii <eliz@gnu.org> writes:

    Should be fixed now (as of revno 103022).

Wonderful.  I'll give it a try.

bug#7907: 24.0.50; trunk build failure

[Sean Sieger]

Sean Sieger <sean.sieger@gmail.com> writes:

    Eli Zaretskii <eliz@gnu.org> writes:

        Should be fixed now (as of revno 103022).

    Wonderful.  I'll give it a try.

Nope.

Saving file c:/emacs-24.0.50/lisp/net/tramp-loaddefs.el...
Loading c:/emacs-24.0.50/lisp/vc/vc-bzr.el (source)...
Wrote c:/emacs-24.0.50/lisp/net/tramp-loaddefs.el
Loading c:/emacs-24.0.50/lisp/vc/vc-bzr.el (source)...
Symbol's value as variable is void: <<<<<<<
make[1]: *** [c:/emacs-24.0.50/lisp/net/tramp-loaddefs.el] Error -1
make[1]: Leaving directory `c:/emacs-24.0.50/lisp'
make: *** [bootstrap-gmake] Error 2

bug#7907: 24.0.50; trunk build failure

[Stefan Monnier]

> Symbol's value as variable is void: <<<<<<<

You have conflict markers in the file.  While they may have been
committed to the trunk, it's at least as likely that they are only in
your checkout tree because of other changes you've performed that
collide with recent changes.


        Stefan

bug#7907: 24.0.50; trunk build failure

[Eli Zaretskii]

> From: Sean Sieger <sean.sieger@gmail.com>
> Date: Sat, 29 Jan 2011 10:00:07 -0500
> Cc: 
> 
> Saving file c:/emacs-24.0.50/lisp/net/tramp-loaddefs.el...
> Loading c:/emacs-24.0.50/lisp/vc/vc-bzr.el (source)...
> Wrote c:/emacs-24.0.50/lisp/net/tramp-loaddefs.el
> Loading c:/emacs-24.0.50/lisp/vc/vc-bzr.el (source)...
> Symbol's value as variable is void: <<<<<<<
> make[1]: *** [c:/emacs-24.0.50/lisp/net/tramp-loaddefs.el] Error -1
> make[1]: Leaving directory `c:/emacs-24.0.50/lisp'
> make: *** [bootstrap-gmake] Error 2

I don't think it's related.  Could be something related to these
errors I see when byte-compiling many Lisp files:

  Compiling D:/gnu/bzr/emacs/test2/lisp/expand.el

  In toplevel form:
  D:/gnu/bzr/emacs/test2/lisp/expand.el:97:1:Error: Symbol's chain of function indirections contains a loop: vc-update

  Compiling D:/gnu/bzr/emacs/test2/lisp/generic-x.el

  In toplevel form:
  D:/gnu/bzr/emacs/test2/lisp/generic-x.el:1547:7:Error: Symbol's chain of function indirections contains a loop: vc-update

Btw, are you compiling with "-j N" switch?  Because otherwise I don't
understand how come Emacs loads vc-bzr _after_ displaying the "Wrote
c:/emacs-24.0.50/lisp/net/tramp-loaddefs.el" message.

bug#7907: 24.0.50; trunk build failure

[Harald Maier]

Sean Sieger <sean.sieger@gmail.com> writes:

> Saving file c:/emacs-24.0.50/lisp/net/tramp-loaddefs.el...
> Loading c:/emacs-24.0.50/lisp/vc/vc-bzr.el (source)...
> Wrote c:/emacs-24.0.50/lisp/net/tramp-loaddefs.el
> Loading c:/emacs-24.0.50/lisp/vc/vc-bzr.el (source)...
> Symbol's value as variable is void: <<<<<<<
> make[1]: *** [c:/emacs-24.0.50/lisp/net/tramp-loaddefs.el] Error -1
> make[1]: Leaving directory `c:/emacs-24.0.50/lisp'
> make: *** [bootstrap-gmake] Error 2

Hello Sean,

I saw this error some days before on OS X in a "nextstep" build. With a
new complete checkout the error doesn't happen any more. So I don't
think this is related to Eli's gnulib work.

The strange thing about this error was, that it didn't occur on a "x11"
build with the same checkout.

Harald

bug#7907: 24.0.50; trunk build failure

[Harald Maier]

Eli Zaretskii <eliz@gnu.org> writes:

>> gcc -I. -c -gdwarf-2 -g3 -DEMACSDEBUG -DENABLE_CHECKING -DXASSERTS
>> -fno-crossju
>> mping -Ic:/imagesupport/include -Demacs=1 -DHAVE_CONFIG_H
>> -I../nt/inc -DHAVE_NT
>> GUI=1 -DUSE_CRT_DLL=1 -DPURESIZE=5000000 -o oo/i386/print.o print.c
>> print.c:53:21: error: ftoastr.h: No such file or directory
>> make[2]: *** [oo/i386/print.o] Error 1
>> make[2]: Leaving directory `c:/emacs-24.0.50/src'
>> make[1]: *** [bootstrap-temacs] Error 2
>> make[1]: Leaving directory `c:/emacs-24.0.50/src'
>> make: *** [bootstrap-gmake] Error 2
>
> Should be fixed now (as of revno 103022).

Compilation works here. Thanks Eli.

But I am facing some issues:

1. During bootrstraping I got tons of "Symbol's chain of function
   indirections contains a loop". For example vc-pull, vc-update.

2. Compiling of vc.el during bootstraping does not finish. I have to
   kill the emacs process.

3. After bootstraping the lisp directory contains only 990 .elc files.

4. But an "recompile" in the lisp directory after bootstraping
   works. Then the count of .elc files is 1287.

Harald

--- configure options ---
In GNU Emacs 24.0.50.1 (i386-mingw-nt5.1.2600)
 of 2011-01-30 on ATE-XP-32
Windowing system distributor `Microsoft Corp.', version 5.1.2600
configured using `configure --with-gcc (3.4) --cflags -Ic:/Programme/GnuWin32/include'

bug#7907: 24.0.50; trunk build failure

[Eli Zaretskii]

> From: Harald Maier <harald@maierh.de>
> CC: bug-gnu-emacs@gnu.org
> Date: Sun, 30 Jan 2011 10:49:51 +0100
> 
> 1. During bootrstraping I got tons of "Symbol's chain of function
>    indirections contains a loop". For example vc-pull, vc-update.
> 
> 2. Compiling of vc.el during bootstraping does not finish. I have to
>    kill the emacs process.
> 
> 3. After bootstraping the lisp directory contains only 990 .elc files.
> 
> 4. But an "recompile" in the lisp directory after bootstraping
>    works. Then the count of .elc files is 1287.

Exactly my observations.  It appears that emacs.exe produced by the
1st stage of "make bootstrap", the one with 5000000 as pure size, has
some problems, because an emacs.exe compiled normally doesn't have
this problem.

If someone could debug this or suggest ideas why this would happen, it
will be greatly appreciated.  Failing that, I will try to debug this
myself, time permitting.

bug#7907: 24.0.50; trunk build failure

[Sean Sieger]

Harald Maier <harald@maierh.de> writes:

    Hello Sean,

    I saw this error some days before on OS X in a "nextstep" build. With a
    new complete checkout the error doesn't happen any more. So I don't
    think this is related to Eli's gnulib work.

    The strange thing about this error was, that it didn't occur on a "x11"
    build with the same checkout.

    Harald

Thank you, Harald.  Pulled a fresh copy and the trunk built fine.

bug#7907: 24.0.50; trunk build failure

[Jason Rumney]

On 29/01/2011 23:56, Stefan Monnier wrote:
>> Symbol's value as variable is void:<<<<<<<
>>      
> You have conflict markers in the file.  While they may have been
> committed to the trunk, it's at least as likely that they are only in
> your checkout tree because of other changes you've performed that
> collide with recent changes.
>    

The same happened to me recently with dired.el. The conflict was caused 
by the hash value at the end of the generated autoloads block 
(everything else was the same). Has something changed recently that 
affects the generation of that hash?

Should the build process be changing files like dired.el, that are not 
dedicated autoload files?  It is very confusing to users to get a 
conflict in a file they have not changed which is not obviously an 
autogenerated file.