make bootstrap on CVS head fails

make bootstrap on CVS head fails

[Frank Schmitt]

Hello

I just said "make distclean && cvs -z3 update -Pd && make
bootstrap". Make bootstrap run into the following error:

gnus/gnus-cite.el:32:1:Error: Symbol's function definition is void:
gnus-image-type-available-p

Before this error, I got dozens of warnings like this one: "Source file
`/home/fschmitt/cvs/emacs/lisp/gnus/mm-util.el' newer than byte-compiled
file"

I said "find . -name '*.elc' -exec rm {} \; && make bootstrap" and this
time it worked. Is there an error somewhere in the Makefiles or did I
something wrong?

-- 
Did you ever realize how much text fits in eighty columns? If you now consider
that a signature usually consists of up to four lines, this gives you enough
space to spread a tremendous amount of information with your messages. So seize
this opportunity and don't waste your signature with bullshit nobody will read.

Re: make bootstrap on CVS head fails

[Sébastien Kirche]

Hi,

I have also a problem with make bootstrap today : it stops with the 
following (last command):

> gcc -I/sw/include -L/sw/lib -fpascal-strings -fno-common -DMAC_OSX 
> -I../mac/src -DHAVE_CONFIG_H -I. -I../src 
> -I/Users/seki/Projets/emacs/mac/../lib-src 
> -I/Users/seki/Projets/emacs/mac/../lib-src/../src  -fpascal-strings 
> -fno-common -DMAC_OSX -I../mac/src   -g -O2 -DVERSION="\"21.3.50\"" 
> /Users/seki/Projets/emacs/mac/../lib-src/ebrowse.c getopt.o getopt1.o  
> -o ebrowse
> make[2]: *** No rule to make target 
> `/Users/seki/Projets/emacs/mac/../lib-src/rcs2log', needed by 
> `rcs2log'.  Stop.
> make[1]: *** [lib-src] Error 2
> make: *** [bootstrap] Error 2
> Make bootstrap failed...  Aborting make-package.

I try to build on OSX with make-package, but i've also get same error 
«no rule to build... rcs2log» with Linux IA32.

CVS checkout today around 1:00 am GMT+2

Regards,
Sébastien Kirche

Re: make bootstrap on CVS head fails

[Luc Teirlinck]

Frank Schmitt wrote:

   I just said "make distclean && cvs -z3 update -Pd && make
   bootstrap". Make bootstrap run into the following error:

Use `make maintainer-clean' instead of `make distclean'.

Sincerely,

Luc.

Re: make bootstrap on CVS head fails

[Frank Schmitt]

Luc Teirlinck <teirllm@dms.auburn.edu> writes:

> Frank Schmitt wrote:
>
>    I just said "make distclean && cvs -z3 update -Pd && make
>    bootstrap". Make bootstrap run into the following error:
>
> Use `make maintainer-clean' instead of `make distclean'.

If this is necessary before a bootstrap, why doesn't bootstrap depend on
maintainer-clean? And IMO, a distclean should get rid of all compiled
files including elc-Files. 

-- 
Did you ever realize how much text fits in eighty columns? If you now consider
that a signature usually consists of up to four lines, this gives you enough
space to spread a tremendous amount of information with your messages. So seize
this opportunity and don't waste your signature with bullshit nobody will read.

Re: make bootstrap on CVS head fails

[Luc Teirlinck]

Frank-Schmitt wrote:

   If this is necessary before a bootstrap, why doesn't bootstrap depend on
   maintainer-clean?

It used to, but that was changed to make bootstrapping run faster at
the expense of an occasional failure.  (Personally, I believe that
change was a mistake.)

   And IMO, a distclean should get rid of all compiled files including
   elc-Files.

I believe that here the argument is that .elc files are included with
distributions of released versions (although not with CVS).

Sincerely,

Luc.

Re: make bootstrap on CVS head fails

[Kim F. Storm]

Luc Teirlinck <teirllm@dms.auburn.edu> writes:

> Frank-Schmitt wrote:
>
>    If this is necessary before a bootstrap, why doesn't bootstrap depend on
>    maintainer-clean?
>
> It used to, but that was changed to make bootstrapping run faster at
> the expense of an occasional failure.  (Personally, I believe that
> change was a mistake.)

I agree that bootstrap should do maintainer-clean.

For experts, there could be a bootstrap-no-clean target (with alias bootnc if
too long).

>
>    And IMO, a distclean should get rid of all compiled files including
>    elc-Files.
>
> I believe that here the argument is that .elc files are included with
> distributions of released versions (although not with CVS).


-- 
Kim F. Storm <storm@cua.dk> http://www.cua.dk

Re: make bootstrap on CVS head fails

[Jason Rumney]

[-- Attachment #1.1: Type: text/plain, Size: 764 bytes --]

Kim F. Storm wrote:

>Luc Teirlinck <teirllm@dms.auburn.edu> writes:
>  
>
>>Frank-Schmitt wrote:
>>
>>   If this is necessary before a bootstrap, why doesn't bootstrap depend on
>>   maintainer-clean?
>>
>>It used to, but that was changed to make bootstrapping run faster at
>>the expense of an occasional failure.  (Personally, I believe that
>>change was a mistake.)
>>    
>>
>
>I agree that bootstrap should do maintainer-clean.
>  
>
I wasn't aware of the change so I was a bit confused by the original 
report. If you want to compile faster, use "make recompile" to recompile 
the .elc files followed by a normal "make".

The cases where recompile is not enough and bootstrap is really 
necessary are probably the same cases of "occasional failure" above.


[-- Attachment #1.2: Type: text/html, Size: 1261 bytes --]

[-- Attachment #2: Type: text/plain, Size: 142 bytes --]

_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/emacs-devel

Re: make bootstrap on CVS head fails

[Richard Stallman]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 364 bytes --]

    I try to build on OSX with make-package, but i've also get same error 
    «no rule to build... rcs2log» with Linux IA32.

You mean GNU/Linux IS32, right?

    > make[2]: *** No rule to make target 
    > `/Users/seki/Projets/emacs/mac/../lib-src/rcs2log', needed by 
    > `rcs2log'.  Stop.

lib-src/rcs2log is a source file.  Does it exist in your checkout?

Re: make bootstrap on CVS head fails

[Sébastien Kirche]

[sorry forget to CC the list :( ]

Le 7 sept. 2004, à 5:43, Richard Stallman a écrit :

> You mean GNU/Linux IS32, right?

Err, *GNU*/Linux of course. Debian GNU/Linux to be precise.
But what about S in IS32 ?

>
>> make[2]: *** No rule to make target
>> `/Users/seki/Projets/emacs/mac/../lib-src/rcs2log', needed by
>> `rcs2log'.  Stop.
>
> lib-src/rcs2log is a source file.  Does it exist in your checkout?

Well, i made this morning another checkout and it seems that rcs2log 
was flagged "updated" (??).
Now it compiles nice as it used to.

Weird indeed. I made no changes on it.
Don't know at all from where the problem came from, but it is ok now.

Thanks for attention.
Regards.
Sébastien Kirche

Re: make bootstrap on CVS head fails

[Piet van Oostrum]

On MacOS X with the current CVS the following sequence fails:

make maintainer-clean
./configure  --enable-carbon-app --without-x 
make bootstrap

The error is:

Loading loaddefs.el (source)...
Symbol's function definition is void: define-compilation-mode

loaddefs requires define-compilation-mode, which is defined in compile.el,
but compile hasn't been loaded yet.
-- 
Piet van Oostrum <piet@cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum@hccnet.nl