[Buildroot] [autobuild.buildroot.net] Build results for 2014-05-19

[Buildroot] [autobuild.buildroot.net] Build results for 2014-05-19

[Thomas Petazzoni]

Build statistics for 2014-05-19
===============================

        success : 79 
       failures : 7  
       timeouts : 0  
          TOTAL : 86 

Classification of failures by reason
====================================

                  samba-3.6.23 | 1 
         matchbox-keyboard-0.1 | 1 
                busybox-1.22.1 | 1 
                 host-gdb-6.6a | 1 
               alsa-lib-1.0.26 | 1 
                 cairo-1.12.10 | 1 
             lttng-tools-2.4.1 | 1 

Detail of failures
===================

    x86_64 |                alsa-lib-1.0.26 | NOK | http://autobuild.buildroot.net/results/3ecb780be9e7912dac9670d9e7ad9a810c6243af/
      i686 |                 busybox-1.22.1 | NOK | http://autobuild.buildroot.net/results/725962e174198f0b26b7cbd335691d9a23be73f5/
    mipsel |                  cairo-1.12.10 | NOK | http://autobuild.buildroot.net/results/cde0a6eda1db441cd16ef933fa27cd37411b837e/
      bfin |                  host-gdb-6.6a | NOK | http://autobuild.buildroot.net/results/b47f85553336b7f63ee4ecdf8598374ce4a225a3/
       arm |              lttng-tools-2.4.1 | NOK | http://autobuild.buildroot.net/results/5cde56a0bf8dfac22d01752a98d399f71f664e07/
       arm |          matchbox-keyboard-0.1 | NOK | http://autobuild.buildroot.net/results/47380d57e507ec395a1132dfd7b32d35459c1c7a/
     avr32 |                   samba-3.6.23 | NOK | http://autobuild.buildroot.net/results/0142b1d5b8c3f20f1bff76ece646d901407f7490/


-- 
http://autobuild.buildroot.net

[Buildroot] [autobuild.buildroot.net] Build results for 2014-05-19

[Thomas Petazzoni]

Hello,

On Tue, 20 May 2014 08:30:07 +0200 (CEST), Thomas Petazzoni wrote:

>     x86_64 |                alsa-lib-1.0.26 | NOK | http://autobuild.buildroot.net/results/3ecb780be9e7912dac9670d9e7ad9a810c6243af/

Multiple definitions of vfork in libc.a and libpthread.a. To me, this
looks like a uClibc bug. Do other agree about this? Should we mark
alsa-lib as not available on pure static builds with uClibc? Is there a
uClibc patch that fixes that?

>       i686 |                 busybox-1.22.1 | NOK | http://autobuild.buildroot.net/results/725962e174198f0b26b7cbd335691d9a23be73f5/

A build error in Busybox? Really weird:

libbb/appletlib.c:164:13: error: ???NUM_APPLETS??? undeclared (first use in this function)
libbb/appletlib.c:164:13: note: each undeclared identifier is reported only once for each function it appears in
libbb/appletlib.c: In function ???install_links???:
libbb/appletlib.c:595:18: error: invalid application of ???sizeof??? to incomplete type ???int (* const[])(int,  char **)???

Peter, any idea?

>     mipsel |                  cairo-1.12.10 | NOK | http://autobuild.buildroot.net/results/cde0a6eda1db441cd16ef933fa27cd37411b837e/

The infamous static linking problem with libstdc++.so.

>       bfin |                  host-gdb-6.6a | NOK | http://autobuild.buildroot.net/results/b47f85553336b7f63ee4ecdf8598374ce4a225a3/

Fixed by http://git.buildroot.net/buildroot/commit/?id=dfc3cc23af3ceb715f322dcb63d82d9d9abfe3ac.

>        arm |              lttng-tools-2.4.1 | NOK | http://autobuild.buildroot.net/results/5cde56a0bf8dfac22d01752a98d399f71f664e07/

The usual gcc issue. Last night, I've restarted the build of the
Buildroot external toolchains used in the autobuilders with Buildroot
2014.05-rc1. When I'll deploy them, it should hopefully fix the issue
since we now have the gcc patch that fixes this problem.

>        arm |          matchbox-keyboard-0.1 | NOK | http://autobuild.buildroot.net/results/47380d57e507ec395a1132dfd7b32d35459c1c7a/

checking for FAKEKEY... yes
checking for XFT... configure: error: *** Required xft Librarys not found ***
make: *** [/scratch/peko/build/matchbox-keyboard-0.1/.stamp_configured] Error 1

>      avr32 |                   samba-3.6.23 | NOK | http://autobuild.buildroot.net/results/0142b1d5b8c3f20f1bff76ece646d901407f7490/

Could be fixed by applying http://patchwork.ozlabs.org/patch/350543/.

Best regards,

Thomas
-- 
Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

[Buildroot] [autobuild.buildroot.net] Build results for 2014-05-19

[Peter Korsgaard]

>>>>> "Thomas" == Thomas Petazzoni <thomas.petazzoni@free-electrons.com> writes:

 > Hello,
 > On Tue, 20 May 2014 08:30:07 +0200 (CEST), Thomas Petazzoni wrote:

 >> x86_64 |                alsa-lib-1.0.26 | NOK | http://autobuild.buildroot.net/results/3ecb780be9e7912dac9670d9e7ad9a810c6243af/

 > Multiple definitions of vfork in libc.a and libpthread.a. To me, this
 > looks like a uClibc bug. Do other agree about this? Should we mark
 > alsa-lib as not available on pure static builds with uClibc? Is there a
 > uClibc patch that fixes that?

It looks like it. I am no aware of any patches (but I haven't really
looked either).


 >> i686 |                 busybox-1.22.1 | NOK | http://autobuild.buildroot.net/results/725962e174198f0b26b7cbd335691d9a23be73f5/

 > A build error in Busybox? Really weird:

 > libbb/appletlib.c:164:13: error: ???NUM_APPLETS??? undeclared (first use in this function)
 > libbb/appletlib.c:164:13: note: each undeclared identifier is reported only once for each function it appears in
 > libbb/appletlib.c: In function ???install_links???:
 > libbb/appletlib.c:595:18: error: invalid application of ???sizeof??? to incomplete type ???int (* const[])(int,  char **)???

 > Peter, any idea?

Very odd. Glitch in the matrix? I would say just ignore it unless it
happens again.


 >> arm |              lttng-tools-2.4.1 | NOK | http://autobuild.buildroot.net/results/5cde56a0bf8dfac22d01752a98d399f71f664e07/

 > The usual gcc issue. Last night, I've restarted the build of the
 > Buildroot external toolchains used in the autobuilders with Buildroot
 > 2014.05-rc1. When I'll deploy them, it should hopefully fix the issue
 > since we now have the gcc patch that fixes this problem.


Thanks!


 >> avr32 |                   samba-3.6.23 | NOK | http://autobuild.buildroot.net/results/0142b1d5b8c3f20f1bff76ece646d901407f7490/

 > Could be fixed by applying http://patchwork.ozlabs.org/patch/350543/.

Will apply now.

-- 
Bye, Peter Korsgaard

[Buildroot] [autobuild.buildroot.net] Build results for 2014-05-19

[Arnout Vandecappelle]

On 20/05/14 09:25, Thomas Petazzoni wrote:
> On Tue, 20 May 2014 08:30:07 +0200 (CEST), Thomas Petazzoni wrote:
> 
>> >     x86_64 |                alsa-lib-1.0.26 | NOK | http://autobuild.buildroot.net/results/3ecb780be9e7912dac9670d9e7ad9a810c6243af/
> Multiple definitions of vfork in libc.a and libpthread.a. To me, this
> looks like a uClibc bug. Do other agree about this? Should we mark
> alsa-lib as not available on pure static builds with uClibc? Is there a
> uClibc patch that fixes that?

 When I did that allpackageyes of a static arm configuration, there were dozens
of packages that show the same or a similar problem (duplicates in libpthread
and libc). I looked at one of them in more detail, and I could make it link by
playing around with the link options, but I haven't been able to find the real
cause.

 Regards,
 Arnout


-- 
Arnout Vandecappelle                          arnout at mind be
Senior Embedded Software Architect            +32-16-286500
Essensium/Mind                                http://www.mind.be
G.Geenslaan 9, 3001 Leuven, Belgium           BE 872 984 063 RPR Leuven
LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle
GPG fingerprint:  7CB5 E4CC 6C2E EFD4 6E3D A754 F963 ECAB 2450 2F1F