build breakage: python-native-2.6.1

build breakage: python-native-2.6.1

[Robert P. J. Day]

  building beagleboard-demo-image:

NOTE: make -j4 BUILD_SYS= HOST_SYS= LIBC= STAGING_LIBDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/lib STAGING_INCDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/include
*** WARNING: renaming "dbm" since importing it failed: build/lib.linux-x86_64-2.6/dbm.so: undefined symbol: dbm_firstkey
/bin/sh: line 1:   515 Segmentation fault      LD_LIBRARY_PATH=/home/rpjday/oe/angstrom-dev/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1: CC='ccache gcc -pthread' LDSHARED='ccache gcc -pthread -shared' OPT='-DNDEBUG -g  -O3 -Wall -Wstrict-prototypes' ./python -E ./setup.py -q build
make: *** [sharedmods] Error 139
FATAL: oe_runmake failed

  that's it, that's the entire log file.  and it's entirely
reproducible.

rday
--

========================================================================
Robert P. J. Day                               Waterloo, Ontario, CANADA

            Linux Consulting, Training and Kernel Pedantry.

Web page:                                          http://crashcourse.ca
Twitter:                                       http://twitter.com/rpjday
========================================================================

Re: build breakage: python-native-2.6.1

[Philip Balister]

On 11/13/2009 04:45 AM, Robert P. J. Day wrote:
>
>    building beagleboard-demo-image:
>
> NOTE: make -j4 BUILD_SYS= HOST_SYS= LIBC= STAGING_LIBDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/lib STAGING_INCDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/include
> *** WARNING: renaming "dbm" since importing it failed: build/lib.linux-x86_64-2.6/dbm.so: undefined symbol: dbm_firstkey
> /bin/sh: line 1:   515 Segmentation fault      LD_LIBRARY_PATH=/home/rpjday/oe/angstrom-dev/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1: CC='ccache gcc -pthread' LDSHARED='ccache gcc -pthread -shared' OPT='-DNDEBUG -g  -O3 -Wall -Wstrict-prototypes' ./python -E ./setup.py -q build
> make: *** [sharedmods] Error 139
> FATAL: oe_runmake failed
>
>    that's it, that's the entire log file.  and it's entirely
> reproducible.

Well, your and my setups are the same :) Graeme is the the other F11 
user and it is working for him :(

Philip

>
> rday
> --
>
> ========================================================================
> Robert P. J. Day                               Waterloo, Ontario, CANADA
>
>              Linux Consulting, Training and Kernel Pedantry.
>
> Web page:                                          http://crashcourse.ca
> Twitter:                                       http://twitter.com/rpjday
> ========================================================================
>
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
>

Re: build breakage: python-native-2.6.1

[Robert P. J. Day]

On Fri, 13 Nov 2009, Philip Balister wrote:

> On 11/13/2009 04:45 AM, Robert P. J. Day wrote:
> >
> >    building beagleboard-demo-image:
> >
> > NOTE: make -j4 BUILD_SYS= HOST_SYS= LIBC=
> > STAGING_LIBDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/lib
> > STAGING_INCDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/include
> > *** WARNING: renaming "dbm" since importing it failed:
> > build/lib.linux-x86_64-2.6/dbm.so: undefined symbol: dbm_firstkey
> > /bin/sh: line 1:   515 Segmentation fault
> > LD_LIBRARY_PATH=/home/rpjday/oe/angstrom-dev/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1:
> > CC='ccache gcc -pthread' LDSHARED='ccache gcc -pthread -shared'
> > OPT='-DNDEBUG -g  -O3 -Wall -Wstrict-prototypes' ./python -E ./setup.py -q
> > build
> > make: *** [sharedmods] Error 139
> > FATAL: oe_runmake failed
> >
> >    that's it, that's the entire log file.  and it's entirely
> > reproducible.
>
> Well, your and my setups are the same :) Graeme is the the other F11 user and
> it is working for him :(

  i'm not on F11 anymore, i'm effectively on F12 (fully-updated f12
beta).  i made the upgrade a couple days ago and that's when those
other packages broke.  hmmmm ... i really should have made that
chronological connection.

  but after the upgrade, i did a full rebuild of everything and those
were the only packages to fail.  so, percentage-wise, that's not bad.

rday
--


========================================================================
Robert P. J. Day                               Waterloo, Ontario, CANADA

            Linux Consulting, Training and Kernel Pedantry.

Web page:                                          http://crashcourse.ca
Twitter:                                       http://twitter.com/rpjday
========================================================================

Re: build breakage: python-native-2.6.1

[Philip Balister]

On 11/13/2009 07:34 AM, Robert P. J. Day wrote:
> On Fri, 13 Nov 2009, Philip Balister wrote:
>
>> On 11/13/2009 04:45 AM, Robert P. J. Day wrote:
>>>
>>>     building beagleboard-demo-image:
>>>
>>> NOTE: make -j4 BUILD_SYS= HOST_SYS= LIBC=
>>> STAGING_LIBDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/lib
>>> STAGING_INCDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/include
>>> *** WARNING: renaming "dbm" since importing it failed:
>>> build/lib.linux-x86_64-2.6/dbm.so: undefined symbol: dbm_firstkey
>>> /bin/sh: line 1:   515 Segmentation fault
>>> LD_LIBRARY_PATH=/home/rpjday/oe/angstrom-dev/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1:
>>> CC='ccache gcc -pthread' LDSHARED='ccache gcc -pthread -shared'
>>> OPT='-DNDEBUG -g  -O3 -Wall -Wstrict-prototypes' ./python -E ./setup.py -q
>>> build
>>> make: *** [sharedmods] Error 139
>>> FATAL: oe_runmake failed
>>>
>>>     that's it, that's the entire log file.  and it's entirely
>>> reproducible.
>>
>> Well, your and my setups are the same :) Graeme is the the other F11 user and
>> it is working for him :(
>
>    i'm not on F11 anymore, i'm effectively on F12 (fully-updated f12
> beta).  i made the upgrade a couple days ago and that's when those
> other packages broke.  hmmmm ... i really should have made that
> chronological connection.
>
>    but after the upgrade, i did a full rebuild of everything and those
> were the only packages to fail.  so, percentage-wise, that's not bad.

I'm on F11 fwiw.

I got this from gsb and a core file:

http://pastebin.ca/1669812

Philip

>
> rday
> --
>
>
> ========================================================================
> Robert P. J. Day                               Waterloo, Ontario, CANADA
>
>              Linux Consulting, Training and Kernel Pedantry.
>
> Web page:                                          http://crashcourse.ca
> Twitter:                                       http://twitter.com/rpjday
> ========================================================================
>
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
>

Re: build breakage: python-native-2.6.1

[Petr Štetiar]

Philip Balister <philip@balister.org> [2009-11-13 07:42:40]:

> I'm on F11 fwiw.
>
> I got this from gsb and a core file:
>
> http://pastebin.ca/1669812
>
> Philip

I had problems building lot of native packages on Ubuntu 8.10, so I've tried
to update to 9.04 and voalá, beagleboard-demo-image builds now without any
problem.

What I'm trying to say here, is that there should be some prebuilt chroot
environment with sane host tools everyone can use. Or we can build everything
from scratch using OE, so we don't need to use anything from the host system.

It's simply impossible to have same result on different systems,
distributions.

-- ynezz

Re: build breakage: python-native-2.6.1

[Adrian Alonso]

I workaround python-native-2.6.1 segfault by un-comment

readline readline.c -lreadline -ltermcap

in the file

work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/Setup

I'm also under Fedora 12 beta, hope this help;



2009/11/13 Petr Štetiar <ynezzor@gmail.com>

> Philip Balister <philip@balister.org> [2009-11-13 07:42:40]:
>
> > I'm on F11 fwiw.
> >
> > I got this from gsb and a core file:
> >
> > http://pastebin.ca/1669812
> >
> > Philip
>
> I had problems building lot of native packages on Ubuntu 8.10, so I've
> tried
> to update to 9.04 and voalá, beagleboard-demo-image builds now without any
> problem.
>
> What I'm trying to say here, is that there should be some prebuilt chroot
> environment with sane host tools everyone can use. Or we can build
> everything
> from scratch using OE, so we don't need to use anything from the host
> system.
>
> It's simply impossible to have same result on different systems,
> distributions.
>
> -- ynezz
>
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
>



-- 
Saludos
Adrian Alonso
http://aalonso.wordpress.com
Fedora Project Contributor
https://fedoraproject.org/wiki/User:Aalonso

Re: build breakage: python-native-2.6.1

[Robert P. J. Day]

On Fri, 13 Nov 2009, Adrian Alonso wrote:

> I workaround python-native-2.6.1 segfault by un-comment
>
> readline readline.c -lreadline -ltermcap
>
> in the file
>
> work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/Setup
>
> I'm also under Fedora 12 beta, hope this help;

  huh..  that worked for me, too.  thanks.  now the only breakage left
for me is guile-native.

rday
--

========================================================================
Robert P. J. Day                               Waterloo, Ontario, CANADA

            Linux Consulting, Training and Kernel Pedantry.

Web page:                                          http://crashcourse.ca
Twitter:                                       http://twitter.com/rpjday
========================================================================

Re: build breakage: python-native-2.6.1

[Richard Purdie]

On Fri, 2009-11-13 at 07:42 -0500, Philip Balister wrote:
> On 11/13/2009 07:34 AM, Robert P. J. Day wrote:
> > On Fri, 13 Nov 2009, Philip Balister wrote:
> >
> >> On 11/13/2009 04:45 AM, Robert P. J. Day wrote:
> >>>
> >>>     building beagleboard-demo-image:
> >>>
> >>> NOTE: make -j4 BUILD_SYS= HOST_SYS= LIBC=
> >>> STAGING_LIBDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/lib
> >>> STAGING_INCDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/include
> >>> *** WARNING: renaming "dbm" since importing it failed:
> >>> build/lib.linux-x86_64-2.6/dbm.so: undefined symbol: dbm_firstkey
> >>> /bin/sh: line 1:   515 Segmentation fault
> >>> LD_LIBRARY_PATH=/home/rpjday/oe/angstrom-dev/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1:
> >>> CC='ccache gcc -pthread' LDSHARED='ccache gcc -pthread -shared'
> >>> OPT='-DNDEBUG -g  -O3 -Wall -Wstrict-prototypes' ./python -E ./setup.py -q
> >>> build
> >>> make: *** [sharedmods] Error 139
> >>> FATAL: oe_runmake failed
> >>>
> >>>     that's it, that's the entire log file.  and it's entirely
> >>> reproducible.
> >>
> >> Well, your and my setups are the same :) Graeme is the the other F11 user and
> >> it is working for him :(
> >
> >    i'm not on F11 anymore, i'm effectively on F12 (fully-updated f12
> > beta).  i made the upgrade a couple days ago and that's when those
> > other packages broke.  hmmmm ... i really should have made that
> > chronological connection.
> >
> >    but after the upgrade, i did a full rebuild of everything and those
> > were the only packages to fail.  so, percentage-wise, that's not bad.
> 
> I'm on F11 fwiw.
> 
> I got this from gsb and a core file:
> 
> http://pastebin.ca/1669812
> 
> Philip

Can someone with this problem please run a python-native build with
http://tim.rpsys.net/python-debug.patch applied?

There should be some debug at the end of the do_compile log afterwards.
I see:

Looking in ncursesw for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in ncurses for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libncurses.so

and I'm guessing it finds ncursesw in /lib/lib64 for some people.

I also see some other worrying entries on my system :(

Cheers,

Richard

Re: build breakage: python-native-2.6.1

[Robert P. J. Day]

On Fri, 13 Nov 2009, Richard Purdie wrote:

> Can someone with this problem please run a python-native build with
> http://tim.rpsys.net/python-debug.patch applied?
>
> There should be some debug at the end of the do_compile log afterwards.
> I see:
>
> Looking in ncursesw for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
> Looking in ncurses for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
> Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libncurses.so
>
> and I'm guessing it finds ncursesw in /lib/lib64 for some people.
>
> I also see some other worrying entries on my system :(

  do you still want this tested given the earlier posting about how
this was fixed with that readline tweak?

rday
--

========================================================================
Robert P. J. Day                               Waterloo, Ontario, CANADA

            Linux Consulting, Training and Kernel Pedantry.

Web page:                                          http://crashcourse.ca
Twitter:                                       http://twitter.com/rpjday
========================================================================

Re: build breakage: python-native-2.6.1

[Richard Purdie]

On Fri, 2009-11-13 at 11:29 -0500, Robert P. J. Day wrote:
> On Fri, 13 Nov 2009, Richard Purdie wrote:
> 
> > Can someone with this problem please run a python-native build with
> > http://tim.rpsys.net/python-debug.patch applied?
> >
> > There should be some debug at the end of the do_compile log afterwards.
> > I see:
> >
> > Looking in ncursesw for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
> > Looking in ncurses for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
> > Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libncurses.so
> >
> > and I'm guessing it finds ncursesw in /lib/lib64 for some people.
> >
> > I also see some other worrying entries on my system :(
> 
>   do you still want this tested given the earlier posting about how
> this was fixed with that readline tweak?

That readline tweak is a hack and there is a bigger problem here. I'd
like to solve it completely so yes, I'd like to confirm that was the
problem please. Once someone has done that I also have a proposed fix:

http://tim.rpsys.net/python-hostfix.patch

For completeness, I've included some before/after logs below for what
we're now potentially avoiding. I'm cc'ing the python maintainer to make
sure he sees it, its his call whether to include this patch. I'd also
argue it could be worth removing more "/usr" references from setup.py
but that is his call and my patch solves 90% of the problem (but not
sqlite3 notice). Also some modules now failed to build - DEPENDS would
need to be fixed to get them back if we want them.

Cheers,

Richard

BEFORE
------

Looking in readline for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libreadline.so
Looking in ncursesw for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in ncurses for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libncurses.so
Looking in crypt for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Found /usr/lib64/libcrypt.so
Looking in ssl for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib', '/usr/local/ssl/lib', '/usr/contrib/ssl/lib/']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libssl.so
Looking in db-4.3 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libdb-4.3.so
Looking in sqlite3 for ['/usr/include/../lib64', '/usr/include/../lib', '/usr/include/../../lib64', '/usr/include/../../lib', '/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Found /usr/include/../lib64/libsqlite3.so
Looking in gdbm for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libgdbm.so
Looking in gdbm for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libgdbm.so
Looking in nsl for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Found /usr/lib64/libnsl.so
Looking in ncursesw for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in ncurses for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libncurses.so
Looking in panel for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libpanel.so
Looking in z for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libz.so
Looking in bz2 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libbz2.so
Looking in tk8.5 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tcl8.5 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tk85 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tcl85 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/u/media/build1/builds/oe/angstrom/tmp/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/_testcapimodule.c: In function 'test_u_code':
/media/build1/builds/oe/angstrom/tmp/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/_testcapimodule.c:489: warning: unused variable 'x'
/media/build1/builds/oe/angstrom/tmp/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/cPickle.c: In function 'Unpickler_noload':
/media/build1/builds/oe/angstrom/tmp/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/cPickle.c:4317: warning: assuming signed overflow does not occur when assuming that (X - c) > X is always false
/media/build1/builds/oe/angstrom/tmp/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/cPickle.c:194: warning: assuming signed overflow does not occur when assuming that (X - c) > X is always false
/media/build1/builds/oe/angstrom/tmp/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/cPickle.c: In function 'load':
/media/build1/builds/oe/angstrom/tmp/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/cPickle.c:4317: warning: assuming signed overflow does not occur when assuming that (X - c) > X is always false
In file included from /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/include/readline/readline.h:38,
                 from /media/build1/builds/oe/angstrom/tmp/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/readline.c:32:
/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/include/readline/rltypedefs.h:35: warning: function declaration isn't a prototype
/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/include/readline/rltypedefs.h:36: warning: function declaration isn't a prototype
/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/include/readline/rltypedefs.h:37: warning: function declaration isn't a prototype
/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/include/readline/rltypedefs.h:38: warning: function declaration isn't a prototype
In file included from /media/build1/builds/oe/angstrom/tmp/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/readline.c:32:
/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/include/readline/readline.h:378: warning: function declaration isn't a prototype
gcc: unrecognized option '-R/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib'
gcc: unrecognized option '-R/usr/lib64'
gcc: unrecognized option '-R/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib'
In file included from /media/build1/builds/oe/angstrom/tmp/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/gdbmmodule.c:13:
/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/include/gdbm.h:85: warning: function declaration isn't a prototype
aclocal.m4:14: error: this file was generated for autoconf 2.61.
You have another version of autoconf.  If you want to use that,
you should regenerate the build system entirely.
aclocal.m4:14: the top level
autom4te: /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/bin/m4 failed with exit status: 63
sr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tk8.4 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tcl8.4 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libtcl8.4.so
Looking in tk84 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tcl84 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tk8.3 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tcl8.3 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tk83 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tcl83 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tk8.2 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tcl8.2 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tk82 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tcl82 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tk8.1 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tcl8.1 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tk81 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tcl81 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tk8.0 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tcl8.0 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tk80 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Looking in tcl80 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/usr/local/lib', '.', '/lib64', '/usr/lib64', '/lib', '/usr/lib']
Failed to configure _ctypes module (res=16128) or missing conffile=build/temp.linux-x86_64-2.6/libffi/fficonfig.py

Failed to find the necessary bits to build these modules:
_tkinter           bsddb185           dl              
imageop            sunaudiodev                        
To find the necessary bits, look in setup.py in detect_modules() for the module's name.


AFTER
-----

Looking in readline for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libreadline.so
Looking in ncursesw for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in ncurses for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libncurses.so
Looking in crypt for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in ssl for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.', '/usr/local/ssl/lib', '/usr/contrib/ssl/lib/']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libssl.so
Looking in db-4.3 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libdb-4.3.so
Looking in sqlite3 for ['/usr/include/../lib64', '/usr/include/../lib', '/usr/include/../../lib64', '/usr/include/../../lib', '/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Found /usr/include/../lib64/libsqlite3.so
Looking in gdbm for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libgdbm.so
Looking in gdbm for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libgdbm.so
Looking in nsl for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in ncursesw for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in ncurses for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libncurses.so
Looking in panel for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libpanel.so
Looking in z for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libz.so
Looking in bz2 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libbz2.so
Looking in tk8.5 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tcl8.5 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tk85 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tcl85 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tk8.4 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tcl8.4 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Found /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib/libtcl8.4.so
Looking in tk84 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tcl84 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tk8.3 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tcl8.3 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tk83 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tcl83 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tk8.2 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tcl8.2 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tk82 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tcl82 for ['/media/media/build1/builds/oe/angstrom/tmp/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/_testcapimodule.c: In function 'test_u_code':
/media/build1/builds/oe/angstrom/tmp/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/_testcapimodule.c:489: warning: unused variable 'x'
/media/build1/builds/oe/angstrom/tmp/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/cPickle.c: In function 'Unpickler_noload':
/media/build1/builds/oe/angstrom/tmp/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/cPickle.c:4317: warning: assuming signed overflow does not occur when assuming that (X - c) > X is always false
/media/build1/builds/oe/angstrom/tmp/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/cPickle.c:194: warning: assuming signed overflow does not occur when assuming that (X - c) > X is always false
/media/build1/builds/oe/angstrom/tmp/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/cPickle.c: In function 'load':
/media/build1/builds/oe/angstrom/tmp/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/cPickle.c:4317: warning: assuming signed overflow does not occur when assuming that (X - c) > X is always false
In file included from /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/include/readline/readline.h:38,
                 from /media/build1/builds/oe/angstrom/tmp/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/readline.c:32:
/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/include/readline/rltypedefs.h:35: warning: function declaration isn't a prototype
/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/include/readline/rltypedefs.h:36: warning: function declaration isn't a prototype
/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/include/readline/rltypedefs.h:37: warning: function declaration isn't a prototype
/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/include/readline/rltypedefs.h:38: warning: function declaration isn't a prototype
In file included from /media/build1/builds/oe/angstrom/tmp/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/readline.c:32:
/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/include/readline/readline.h:378: warning: function declaration isn't a prototype
/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tk8.1 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tcl8.1 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tk81 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tcl81 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tk8.0 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tcl8.0 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tk80 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
Looking in tcl80 for ['/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib', '.']
*** WARNING: renaming "crypt" since importing it failed: build/lib.linux-x86_64-2.6/crypt.so: undefined symbol: crypt
gcc: unrecognized option '-R/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib'
gcc: unrecognized option '-R/usr/lib64'
gcc: unrecognized option '-R/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/lib'
In file included from /media/build1/builds/oe/angstrom/tmp/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1/Modules/gdbmmodule.c:13:
/media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/include/gdbm.h:85: warning: function declaration isn't a prototype
*** WARNING: renaming "nis" since importing it failed: build/lib.linux-x86_64-2.6/nis.so: undefined symbol: yp_get_default_domain
aclocal.m4:14: error: this file was generated for autoconf 2.61.
You have another version of autoconf.  If you want to use that,
you should regenerate the build system entirely.
aclocal.m4:14: the top level
autom4te: /media/build1/builds/oe/angstrom/tmp/staging/x86_64-linux/usr/bin/m4 failed with exit status: 63
Failed to configure _ctypes module (res=16128) or missing conffile=build/temp.linux-x86_64-2.6/libffi/fficonfig.py

Failed to find the necessary bits to build these modules:
_tkinter           bsddb185           dl              
imageop            sunaudiodev                        
To find the necessary bits, look in setup.py in detect_modules() for the module's name.


Failed to build these modules:
crypt              nis                                

Re: build breakage: python-native-2.6.1

[Mike Westerhof]

Philip Balister wrote:
> On 11/13/2009 07:34 AM, Robert P. J. Day wrote:
>>    i'm not on F11 anymore, i'm effectively on F12 (fully-updated f12
>> beta).  i made the upgrade a couple days ago and that's when those
>> other packages broke.  hmmmm ... i really should have made that
>> chronological connection.
>>
>>    but after the upgrade, i did a full rebuild of everything and those
>> were the only packages to fail.  so, percentage-wise, that's not bad.
> 
> I'm on F11 fwiw.
> 
> I got this from gsb and a core file:
> 
> http://pastebin.ca/1669812
> 
> Philip
> 
>>
>> rday

Same python-native failure.

Fedora 10, the host hasn't been updated in many many weeks (so host
changes are ruled out).  CCACHE is disabled.  xterm is not installed on
the host.  It is a 64-bit install.

-Mike (mwester)

Re: build breakage: python-native-2.6.1

[Mike Westerhof]

Richard Purdie wrote:

> That readline tweak is a hack and there is a bigger problem here. I'd
> like to solve it completely so yes, I'd like to confirm that was the
> problem please. Once someone has done that I also have a proposed fix:
> 
> http://tim.rpsys.net/python-hostfix.patch
> 
> For completeness, I've included some before/after logs below for what
> we're now potentially avoiding. I'm cc'ing the python maintainer to make
> sure he sees it, its his call whether to include this patch. I'd also
> argue it could be worth removing more "/usr" references from setup.py
> but that is his call and my patch solves 90% of the problem (but not
> sqlite3 notice). Also some modules now failed to build - DEPENDS would
> need to be fixed to get them back if we want them.
> 
> Cheers,
> 
> Richard

The python-hostfix.patch resolves the issue for me.  The logs (before
and after) show that with the patch, we no longer have references to
/usr/lib/ and /usr/lib64/ -- thanks!

-Mike (mwester)

Re: build breakage: python-native-2.6.1

[Adrian Alonso]

After applying the patch http://tim.rpsys.net/python-hostfix.patch
and rebuild python-native_2.6.1.bb

Works with out `readline tweak`

Here is the build log

[aalonso@localhost conf]$ bitbake -b
~/workspace/oe/openembedded/recipes/python/python-native_2.6.1.bb
NOTE: Preparing runqueue
NOTE: Executing runqueue
NOTE: Running task 1 of 17 (ID: 1,
/home/aalonso/workspace/oe/openembedded/recipes/python/
python-native_2.6.1.bb, do_setscene)
NOTE: Running task 2 of 17 (ID: 2,
/home/aalonso/workspace/oe/openembedded/recipes/python/
python-native_2.6.1.bb, do_fetch)
NOTE: Running task 3 of 17 (ID: 3,
/home/aalonso/workspace/oe/openembedded/recipes/python/
python-native_2.6.1.bb, do_unpack)
NOTE: Running task 4 of 17 (ID: 13,
/home/aalonso/workspace/oe/openembedded/recipes/python/
python-native_2.6.1.bb, do_distribute_sources)
NOTE: Unpacking ../../sources/Python-2.6.1.tar.bz2 to
tmp/work/x86_64-linux/python-native-2.6.1-ml8.2/
NOTE: Running task 5 of 17 (ID: 4,
/home/aalonso/workspace/oe/openembedded/recipes/python/
python-native_2.6.1.bb, do_patch)
NOTE: Applying patch '00-fix-bindir-libdir-for-cross.patch'
(../../openembedded/recipes/python/python-native-2.6.1/00-fix-bindir-libdir-for-cross.patch)
NOTE: Applying patch '04-default-is-optimized.patch'
(../../openembedded/recipes/python/python-native-2.6.1/04-default-is-optimized.patch)
NOTE: Applying patch '05-enable-ctypes-cross-build.patch'
(../../openembedded/recipes/python/python-2.6.1/05-enable-ctypes-cross-build.patch)
NOTE: Applying patch '10-distutils-fix-swig-parameter.patch'
(../../openembedded/recipes/python/python-native-2.6.1/10-distutils-fix-swig-parameter.patch)
NOTE: Applying patch '11-distutils-never-modify-shebang-line.patch'
(../../openembedded/recipes/python/python-native-2.6.1/11-distutils-never-modify-shebang-line.patch)
NOTE: Applying patch '12-distutils-prefix-is-inside-staging-area.patch'
(../../openembedded/recipes/python/python-native-2.6.1/12-distutils-prefix-is-inside-staging-area.patch)
NOTE: Applying patch 'debug.patch'
(../../openembedded/recipes/python/python-native-2.6.1/debug.patch)
NOTE: Applying patch 'nohostlibs.patch'
(../../openembedded/recipes/python/python-native-2.6.1/nohostlibs.patch)
NOTE: Running task 6 of 17 (ID: 5,
/home/aalonso/workspace/oe/openembedded/recipes/python/
python-native_2.6.1.bb, do_configure)
NOTE: Running task 7 of 17 (ID: 6,
/home/aalonso/workspace/oe/openembedded/recipes/python/
python-native_2.6.1.bb, do_qa_configure)
NOTE: Running task 8 of 17 (ID: 7,
/home/aalonso/workspace/oe/openembedded/recipes/python/
python-native_2.6.1.bb, do_compile)
NOTE: Running task 9 of 17 (ID: 8,
/home/aalonso/workspace/oe/openembedded/recipes/python/
python-native_2.6.1.bb, do_install)
NOTE: Running task 10 of 17 (ID: 12,
/home/aalonso/workspace/oe/openembedded/recipes/python/
python-native_2.6.1.bb, do_package)
NOTE: Running task 11 of 17 (ID: 9,
/home/aalonso/workspace/oe/openembedded/recipes/python/
python-native_2.6.1.bb, do_populate_staging)
NOTE: Legacy staging mode for
/home/aalonso/workspace/oe/openembedded/recipes/python/
python-native_2.6.1.bb
NOTE: Running task 12 of 17 (ID: 16,
/home/aalonso/workspace/oe/openembedded/recipes/python/
python-native_2.6.1.bb, do_package_write_ipk)
NOTE: Running task 13 of 17 (ID: 11,
/home/aalonso/workspace/oe/openembedded/recipes/python/
python-native_2.6.1.bb, do_package_write)
NOTE: Running task 14 of 17 (ID: 15,
/home/aalonso/workspace/oe/openembedded/recipes/python/
python-native_2.6.1.bb, do_package_stage)
NOTE: Running task 15 of 17 (ID: 10,
/home/aalonso/workspace/oe/openembedded/recipes/python/
python-native_2.6.1.bb, do_qa_staging)
NOTE: Running task 16 of 17 (ID: 14,
/home/aalonso/workspace/oe/openembedded/recipes/python/
python-native_2.6.1.bb, do_package_stage_all)
NOTE: Running task 17 of 17 (ID: 0,
/home/aalonso/workspace/oe/openembedded/recipes/python/
python-native_2.6.1.bb, do_build)
NOTE: Tasks Summary: Attempted 17 tasks of which 0 didn't need to be rerun
and 0 failed.


On Fri, Nov 13, 2009 at 10:42 AM, Mike Westerhof <mike@mwester.net> wrote:

> Philip Balister wrote:
> > On 11/13/2009 07:34 AM, Robert P. J. Day wrote:
> >>    i'm not on F11 anymore, i'm effectively on F12 (fully-updated f12
> >> beta).  i made the upgrade a couple days ago and that's when those
> >> other packages broke.  hmmmm ... i really should have made that
> >> chronological connection.
> >>
> >>    but after the upgrade, i did a full rebuild of everything and those
> >> were the only packages to fail.  so, percentage-wise, that's not bad.
> >
> > I'm on F11 fwiw.
> >
> > I got this from gsb and a core file:
> >
> > http://pastebin.ca/1669812
> >
> > Philip
> >
> >>
> >> rday
>
> Same python-native failure.
>
> Fedora 10, the host hasn't been updated in many many weeks (so host
> changes are ruled out).  CCACHE is disabled.  xterm is not installed on
> the host.  It is a 64-bit install.
>
> -Mike (mwester)
>
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
>



-- 
Saludos
Adrian Alonso
http://aalonso.wordpress.com
Fedora Project Contributor
https://fedoraproject.org/wiki/User:Aalonso

Re: build breakage: python-native-2.6.1

[Robert P. J. Day]

On Fri, 13 Nov 2009, Adrian Alonso wrote:

> After applying the patch http://tim.rpsys.net/python-hostfix.patch
> and rebuild python-native_2.6.1.bb
>
> Works with out `readline tweak`

  ok, since those are two data points both confirming, i'm going to
sit this one out for a change.  :-)  i'll just wait for the patch to
hit git and take it from there.

  only one to go.  yee ha.

rday
--

========================================================================
Robert P. J. Day                               Waterloo, Ontario, CANADA

            Linux Consulting, Training and Kernel Pedantry.

Web page:                                          http://crashcourse.ca
Twitter:                                       http://twitter.com/rpjday
========================================================================

Re: build breakage: python-native-2.6.1

[Michael 'Mickey' Lauer]

Hi folks,

> That readline tweak is a hack and there is a bigger problem here. I'd
> like to solve it completely so yes, I'd like to confirm that was the
> problem please. Once someone has done that I also have a proposed fix:
> 
> http://tim.rpsys.net/python-hostfix.patch

Acked-by: Michael 'Mickey' Lauer <mickey@vanille-media.de>

:M:

Re: build breakage: python-native-2.6.1

[Philip Balister]

On 11/13/2009 05:41 PM, Michael 'Mickey' Lauer wrote:
> Hi folks,
>
>> That readline tweak is a hack and there is a bigger problem here. I'd
>> like to solve it completely so yes, I'd like to confirm that was the
>> problem please. Once someone has done that I also have a proposed fix:
>>
>> http://tim.rpsys.net/python-hostfix.patch
>
> Acked-by: Michael 'Mickey' Lauer<mickey@vanille-media.de>

Tested-by: Philip Balister <philip@balister.org>

Fixed on F11.

Philip

PS: Yes, I know RP just pushed the patch.

>
> :M:
>
>
>
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
>

Re: build breakage: python-native-2.6.1

[Robert P. J. Day]

On Sat, 14 Nov 2009, Philip Balister wrote:

>
>
> On 11/13/2009 05:41 PM, Michael 'Mickey' Lauer wrote:
> > Hi folks,
> >
> > > That readline tweak is a hack and there is a bigger problem here. I'd
> > > like to solve it completely so yes, I'd like to confirm that was the
> > > problem please. Once someone has done that I also have a proposed fix:
> > >
> > > http://tim.rpsys.net/python-hostfix.patch
> >
> > Acked-by: Michael 'Mickey' Lauer<mickey@vanille-media.de>
>
> Tested-by: Philip Balister <philip@balister.org>
>
> Fixed on F11.
>
> Philip
>
> PS: Yes, I know RP just pushed the patch.

  so there's only one package between me and a build of the
beagleboard demo image -- guile-native -- and adrian alonso(?)
allegedly posted a usable fix for that yesterday.  looking good.

rday
--

========================================================================
Robert P. J. Day                               Waterloo, Ontario, CANADA

            Linux Consulting, Training and Kernel Pedantry.

Web page:                                          http://crashcourse.ca
Twitter:                                       http://twitter.com/rpjday
========================================================================

Re: build breakage: python-native-2.6.1

[Philip Balister]

On 11/14/2009 07:52 AM, Robert P. J. Day wrote:
> On Sat, 14 Nov 2009, Philip Balister wrote:
>
>>
>>
>> On 11/13/2009 05:41 PM, Michael 'Mickey' Lauer wrote:
>>> Hi folks,
>>>
>>>> That readline tweak is a hack and there is a bigger problem here. I'd
>>>> like to solve it completely so yes, I'd like to confirm that was the
>>>> problem please. Once someone has done that I also have a proposed fix:
>>>>
>>>> http://tim.rpsys.net/python-hostfix.patch
>>>
>>> Acked-by: Michael 'Mickey' Lauer<mickey@vanille-media.de>
>>
>> Tested-by: Philip Balister<philip@balister.org>
>>
>> Fixed on F11.
>>
>> Philip
>>
>> PS: Yes, I know RP just pushed the patch.
>
>    so there's only one package between me and a build of the
> beagleboard demo image -- guile-native -- and adrian alonso(?)
> allegedly posted a usable fix for that yesterday.  looking good.

I just started a beagleboard-demo-image build. Too bad it is a nice day 
here :)

Philip


>
> rday
> --
>
> ========================================================================
> Robert P. J. Day                               Waterloo, Ontario, CANADA
>
>              Linux Consulting, Training and Kernel Pedantry.
>
> Web page:                                          http://crashcourse.ca
> Twitter:                                       http://twitter.com/rpjday
> ========================================================================
>
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
>

Re: build breakage: python-native-2.6.1

[Robert P. J. Day]

On Sat, 14 Nov 2009, Robert P. J. Day wrote:

> On Sat, 14 Nov 2009, Philip Balister wrote:
>
> >
> >
> > On 11/13/2009 05:41 PM, Michael 'Mickey' Lauer wrote:
> > > Hi folks,
> > >
> > > > That readline tweak is a hack and there is a bigger problem here. I'd
> > > > like to solve it completely so yes, I'd like to confirm that was the
> > > > problem please. Once someone has done that I also have a proposed fix:
> > > >
> > > > http://tim.rpsys.net/python-hostfix.patch
> > >
> > > Acked-by: Michael 'Mickey' Lauer<mickey@vanille-media.de>
> >
> > Tested-by: Philip Balister <philip@balister.org>
> >
> > Fixed on F11.
> >
> > Philip
> >
> > PS: Yes, I know RP just pushed the patch.
>
>   so there's only one package between me and a build of the
> beagleboard demo image -- guile-native -- and adrian alonso(?)
> allegedly posted a usable fix for that yesterday.  looking good.

  i just pulled that python patch and i'm still getting the same build
error as yesterday -- segmentation fault:

NOTE: make -j4 BUILD_SYS= HOST_SYS= LIBC=
STAGING_LIBDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/lib
STAGING_INCDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/include
*** WARNING: renaming "dbm" since importing it failed:
build/lib.linux-x86_64-2.6/dbm.so: undefined symbol: dbm_firstkey
/bin/sh: line 1: 12842 Segmentation fault      (core dumped)
LD_LIBRARY_PATH=/home/rpjday/oe/angstrom-dev/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1:
CC='ccache gcc -pthread' LDSHARED='ccache gcc -pthread -shared'
OPT='-DNDEBUG -g  -O3 -Wall -Wstrict-prototypes' ./python -E
./setup.py -q build
make: *** [sharedmods] Error 139
FATAL: oe_runmake failed

  this is after "bitbake -c distclean python-native".

rday
--

========================================================================
Robert P. J. Day                               Waterloo, Ontario, CANADA

            Linux Consulting, Training and Kernel Pedantry.

Web page:                                          http://crashcourse.ca
Twitter:                                       http://twitter.com/rpjday
========================================================================

Re: build breakage: python-native-2.6.1

[Philip Balister]

On 11/14/2009 07:59 AM, Robert P. J. Day wrote:
> On Sat, 14 Nov 2009, Robert P. J. Day wrote:
>
>> On Sat, 14 Nov 2009, Philip Balister wrote:
>>
>>>
>>>
>>> On 11/13/2009 05:41 PM, Michael 'Mickey' Lauer wrote:
>>>> Hi folks,
>>>>
>>>>> That readline tweak is a hack and there is a bigger problem here. I'd
>>>>> like to solve it completely so yes, I'd like to confirm that was the
>>>>> problem please. Once someone has done that I also have a proposed fix:
>>>>>
>>>>> http://tim.rpsys.net/python-hostfix.patch
>>>>
>>>> Acked-by: Michael 'Mickey' Lauer<mickey@vanille-media.de>
>>>
>>> Tested-by: Philip Balister<philip@balister.org>
>>>
>>> Fixed on F11.
>>>
>>> Philip
>>>
>>> PS: Yes, I know RP just pushed the patch.
>>
>>    so there's only one package between me and a build of the
>> beagleboard demo image -- guile-native -- and adrian alonso(?)
>> allegedly posted a usable fix for that yesterday.  looking good.
>
>    i just pulled that python patch and i'm still getting the same build
> error as yesterday -- segmentation fault:
>
> NOTE: make -j4 BUILD_SYS= HOST_SYS= LIBC=
> STAGING_LIBDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/lib
> STAGING_INCDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/include
> *** WARNING: renaming "dbm" since importing it failed:
> build/lib.linux-x86_64-2.6/dbm.so: undefined symbol: dbm_firstkey
> /bin/sh: line 1: 12842 Segmentation fault      (core dumped)
> LD_LIBRARY_PATH=/home/rpjday/oe/angstrom-dev/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1:
> CC='ccache gcc -pthread' LDSHARED='ccache gcc -pthread -shared'
> OPT='-DNDEBUG -g  -O3 -Wall -Wstrict-prototypes' ./python -E
> ./setup.py -q build
> make: *** [sharedmods] Error 139
> FATAL: oe_runmake failed
>
>    this is after "bitbake -c distclean python-native".

It is working for me. Check and see if the patch is applied.

Philip


>
> rday
> --
>
> ========================================================================
> Robert P. J. Day                               Waterloo, Ontario, CANADA
>
>              Linux Consulting, Training and Kernel Pedantry.
>
> Web page:                                          http://crashcourse.ca
> Twitter:                                       http://twitter.com/rpjday
> ========================================================================
>
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
>

Re: build breakage: python-native-2.6.1

[Michael 'Mickey' Lauer]

Am Samstag, den 14.11.2009, 07:50 -0500 schrieb Philip Balister:
> 
> On 11/13/2009 05:41 PM, Michael 'Mickey' Lauer wrote:
> > Hi folks,
> >
> >> That readline tweak is a hack and there is a bigger problem here. I'd
> >> like to solve it completely so yes, I'd like to confirm that was the
> >> problem please. Once someone has done that I also have a proposed fix:
> >>
> >> http://tim.rpsys.net/python-hostfix.patch
> >
> > Acked-by: Michael 'Mickey' Lauer<mickey@vanille-media.de>
> 
> Tested-by: Philip Balister <philip@balister.org>
> 
> Fixed on F11.
> 
> Philip
> 
> PS: Yes, I know RP just pushed the patch.

For the records... I pushed it w/ --author. I consider this being
equivalent to SOB by $AUTHOR and ACK by $COMMITER.

-- 
:M:

Re: build breakage: python-native-2.6.1

[Robert P. J. Day]

On Sat, 14 Nov 2009, Philip Balister wrote:

> On 11/14/2009 07:59 AM, Robert P. J. Day wrote:
> > On Sat, 14 Nov 2009, Robert P. J. Day wrote:
> >
> > > On Sat, 14 Nov 2009, Philip Balister wrote:
> > >
> > > >
> > > >
> > > > On 11/13/2009 05:41 PM, Michael 'Mickey' Lauer wrote:
> > > > > Hi folks,
> > > > >
> > > > > > That readline tweak is a hack and there is a bigger problem here.
> > > > > > I'd
> > > > > > like to solve it completely so yes, I'd like to confirm that was the
> > > > > > problem please. Once someone has done that I also have a proposed
> > > > > > fix:
> > > > > >
> > > > > > http://tim.rpsys.net/python-hostfix.patch
> > > > >
> > > > > Acked-by: Michael 'Mickey' Lauer<mickey@vanille-media.de>
> > > >
> > > > Tested-by: Philip Balister<philip@balister.org>
> > > >
> > > > Fixed on F11.
> > > >
> > > > Philip
> > > >
> > > > PS: Yes, I know RP just pushed the patch.
> > >
> > >    so there's only one package between me and a build of the
> > > beagleboard demo image -- guile-native -- and adrian alonso(?)
> > > allegedly posted a usable fix for that yesterday.  looking good.
> >
> >    i just pulled that python patch and i'm still getting the same build
> > error as yesterday -- segmentation fault:
> >
> > NOTE: make -j4 BUILD_SYS= HOST_SYS= LIBC=
> > STAGING_LIBDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/lib
> > STAGING_INCDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/include
> > *** WARNING: renaming "dbm" since importing it failed:
> > build/lib.linux-x86_64-2.6/dbm.so: undefined symbol: dbm_firstkey
> > /bin/sh: line 1: 12842 Segmentation fault      (core dumped)
> > LD_LIBRARY_PATH=/home/rpjday/oe/angstrom-dev/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1:
> > CC='ccache gcc -pthread' LDSHARED='ccache gcc -pthread -shared'
> > OPT='-DNDEBUG -g  -O3 -Wall -Wstrict-prototypes' ./python -E
> > ./setup.py -q build
> > make: *** [sharedmods] Error 139
> > FATAL: oe_runmake failed
> >
> >    this is after "bitbake -c distclean python-native".
>
> It is working for me. Check and see if the patch is applied.

$ git log
commit a4bddfe9e6a513a8ad1881ed2ffd419390cf20e7
Author: Richard Purdie <rpurdie@linux.intel.com>
Date:   Sat Nov 14 13:25:03 2009 +0100

    python-native: do not look in host paths for shared objects

yes?

rday
--

========================================================================
Robert P. J. Day                               Waterloo, Ontario, CANADA

            Linux Consulting, Training and Kernel Pedantry.

Web page:                                          http://crashcourse.ca
Twitter:                                       http://twitter.com/rpjday
========================================================================

Re: build breakage: python-native-2.6.1

[Robert P. J. Day]

On Sat, 14 Nov 2009, Philip Balister wrote:

> On 11/14/2009 07:59 AM, Robert P. J. Day wrote:
> > On Sat, 14 Nov 2009, Robert P. J. Day wrote:
> >
> > > On Sat, 14 Nov 2009, Philip Balister wrote:
> > >
> > > >
> > > >
> > > > On 11/13/2009 05:41 PM, Michael 'Mickey' Lauer wrote:
> > > > > Hi folks,
> > > > >
> > > > > > That readline tweak is a hack and there is a bigger problem here.
> > > > > > I'd
> > > > > > like to solve it completely so yes, I'd like to confirm that was the
> > > > > > problem please. Once someone has done that I also have a proposed
> > > > > > fix:
> > > > > >
> > > > > > http://tim.rpsys.net/python-hostfix.patch
> > > > >
> > > > > Acked-by: Michael 'Mickey' Lauer<mickey@vanille-media.de>
> > > >
> > > > Tested-by: Philip Balister<philip@balister.org>
> > > >
> > > > Fixed on F11.
> > > >
> > > > Philip
> > > >
> > > > PS: Yes, I know RP just pushed the patch.
> > >
> > >    so there's only one package between me and a build of the
> > > beagleboard demo image -- guile-native -- and adrian alonso(?)
> > > allegedly posted a usable fix for that yesterday.  looking good.
> >
> >    i just pulled that python patch and i'm still getting the same build
> > error as yesterday -- segmentation fault:
> >
> > NOTE: make -j4 BUILD_SYS= HOST_SYS= LIBC=
> > STAGING_LIBDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/lib
> > STAGING_INCDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/include
> > *** WARNING: renaming "dbm" since importing it failed:
> > build/lib.linux-x86_64-2.6/dbm.so: undefined symbol: dbm_firstkey
> > /bin/sh: line 1: 12842 Segmentation fault      (core dumped)
> > LD_LIBRARY_PATH=/home/rpjday/oe/angstrom-dev/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1:
> > CC='ccache gcc -pthread' LDSHARED='ccache gcc -pthread -shared'
> > OPT='-DNDEBUG -g  -O3 -Wall -Wstrict-prototypes' ./python -E
> > ./setup.py -q build
> > make: *** [sharedmods] Error 139
> > FATAL: oe_runmake failed
> >
> >    this is after "bitbake -c distclean python-native".
>
> It is working for me. Check and see if the patch is applied.

 weird.  i retried the distclean on that package, tried to build again
and now it's fine.  how odd.

rday
--

========================================================================
Robert P. J. Day                               Waterloo, Ontario, CANADA

            Linux Consulting, Training and Kernel Pedantry.

Web page:                                          http://crashcourse.ca
Twitter:                                       http://twitter.com/rpjday
========================================================================

Re: build breakage: python-native-2.6.1

[Philip Balister]

On 11/14/2009 08:27 AM, Robert P. J. Day wrote:
> On Sat, 14 Nov 2009, Philip Balister wrote:
>
>> On 11/14/2009 07:59 AM, Robert P. J. Day wrote:
>>> On Sat, 14 Nov 2009, Robert P. J. Day wrote:
>>>
>>>> On Sat, 14 Nov 2009, Philip Balister wrote:
>>>>
>>>>>
>>>>>
>>>>> On 11/13/2009 05:41 PM, Michael 'Mickey' Lauer wrote:
>>>>>> Hi folks,
>>>>>>
>>>>>>> That readline tweak is a hack and there is a bigger problem here.
>>>>>>> I'd
>>>>>>> like to solve it completely so yes, I'd like to confirm that was the
>>>>>>> problem please. Once someone has done that I also have a proposed
>>>>>>> fix:
>>>>>>>
>>>>>>> http://tim.rpsys.net/python-hostfix.patch
>>>>>>
>>>>>> Acked-by: Michael 'Mickey' Lauer<mickey@vanille-media.de>
>>>>>
>>>>> Tested-by: Philip Balister<philip@balister.org>
>>>>>
>>>>> Fixed on F11.
>>>>>
>>>>> Philip
>>>>>
>>>>> PS: Yes, I know RP just pushed the patch.
>>>>
>>>>     so there's only one package between me and a build of the
>>>> beagleboard demo image -- guile-native -- and adrian alonso(?)
>>>> allegedly posted a usable fix for that yesterday.  looking good.
>>>
>>>     i just pulled that python patch and i'm still getting the same build
>>> error as yesterday -- segmentation fault:
>>>
>>> NOTE: make -j4 BUILD_SYS= HOST_SYS= LIBC=
>>> STAGING_LIBDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/lib
>>> STAGING_INCDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/include
>>> *** WARNING: renaming "dbm" since importing it failed:
>>> build/lib.linux-x86_64-2.6/dbm.so: undefined symbol: dbm_firstkey
>>> /bin/sh: line 1: 12842 Segmentation fault      (core dumped)
>>> LD_LIBRARY_PATH=/home/rpjday/oe/angstrom-dev/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1:
>>> CC='ccache gcc -pthread' LDSHARED='ccache gcc -pthread -shared'
>>> OPT='-DNDEBUG -g  -O3 -Wall -Wstrict-prototypes' ./python -E
>>> ./setup.py -q build
>>> make: *** [sharedmods] Error 139
>>> FATAL: oe_runmake failed
>>>
>>>     this is after "bitbake -c distclean python-native".
>>
>> It is working for me. Check and see if the patch is applied.
>
>   weird.  i retried the distclean on that package, tried to build again
> and now it's fine.  how odd.

Obviously not enough coffee :)

Philip

>
> rday
> --
>
> ========================================================================
> Robert P. J. Day                               Waterloo, Ontario, CANADA
>
>              Linux Consulting, Training and Kernel Pedantry.
>
> Web page:                                          http://crashcourse.ca
> Twitter:                                       http://twitter.com/rpjday
> ========================================================================
>
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
>

build breakage: python-native-2.6.4 (was build breakage: python-native-2.6.1)

[Steffen Sledz]

Robert P. J. Day wrote:
> 
>   building beagleboard-demo-image:
> 
> NOTE: make -j4 BUILD_SYS= HOST_SYS= LIBC=
> STAGING_LIBDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/lib
> STAGING_INCDIR=/home/rpjday/oe/angstrom-dev/staging/x86_64-linux/usr/include
> *** WARNING: renaming "dbm" since importing it failed:
> build/lib.linux-x86_64-2.6/dbm.so: undefined symbol: dbm_firstkey
> /bin/sh: line 1:   515 Segmentation fault     
> LD_LIBRARY_PATH=/home/rpjday/oe/angstrom-dev/work/x86_64-linux/python-native-2.6.1-ml8.2/Python-2.6.1:
> CC='ccache gcc -pthread' LDSHARED='ccache gcc -pthread -shared'
> OPT='-DNDEBUG -g  -O3 -Wall -Wstrict-prototypes' ./python -E ./setup.py
> -q build
> make: *** [sharedmods] Error 139
> FATAL: oe_runmake failed
> 
>   that's it, that's the entire log file.  and it's entirely
> reproducible.

The problem is back with python-native-2.6.4 and openSUSE 11.2. :(

It was solved for python-native-2.6.1 by this patch:

commit a4bddfe9e6a513a8ad1881ed2ffd419390cf20e7
Author: Richard Purdie <rpurdie <at> linux.intel.com>
Date:   Sat Nov 14 13:25:03 2009 +0100

    python-native: do not look in host paths for shared objects

Needs this patch to be adapted for all other python-native versions?

Steffen

-- 
Steffen Sledz
DResearch Digital Media Systems GmbH
Otto-Schmirgal-Str.3, D-10319 Berlin, Germany
Tel: +49 (30) 515932237 mailto:sledz@DResearch.DE
Fax: +49 (30) 515932299 http://www.DResearch.DE
Geschäftsführer: Dr. Michael Weber, Werner Mögle;
Amtsgericht Berlin Charlottenburg; HRB 54412;
Ust.-IDNr. DE169013825; WEEE Reg.-Nr. DE 85995642