[Buildroot] [autobuild.buildroot.net] Build results for 2016-05-16

[Buildroot] [autobuild.buildroot.net] Build results for 2016-05-16

[Thomas Petazzoni]

Build statistics for 2016-05-16
===============================

        success : 172
       failures : 60 
       timeouts : 0  
          TOTAL : 232

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

android-tools-4.2.2+git2013... | 5 
                 jack2-v1.9.10 | 4 
                    mpv-0.17.0 | 4 
             stress-ng-0.04.16 | 3 
                    cups-2.1.2 | 3 
                wayland-1.10.0 | 2 
                  ffmpeg-2.8.7 | 2 
             lirc-tools-0.9.3a | 2 
               valgrind-3.11.0 | 2 
flann-04b4a56533faf8c8228d0... | 2 
                  boost-1.60.0 | 2 
                 rsyslog-8.9.0 | 2 
                collectd-5.5.1 | 2 
          libpam-tacplus-1.3.9 | 2 
              gnuradio-3.7.9.1 | 1 
            qt5serialbus-5.6.0 | 1 
                  connman-1.32 | 1 
                     vlc-2.2.3 | 1 
                oprofile-1.1.0 | 1 
      host-gdb-arc-2015.12-gdb | 1 
                  psmisc-22.21 | 1 
                   taskd-1.1.0 | 1 
                    ustr-1.0.4 | 1 
                    php-5.6.21 | 1 
       gadgetfs-test-undefined | 1 
        lttng-babeltrace-1.2.4 | 1 
                 host-go-1.6.2 | 1 
              strongswan-5.4.0 | 1 
                flannel-v0.5.5 | 1 
                    pppd-2.4.7 | 1 
 toolchain-external-legal-info | 1 
liquid-dsp-df5a459fa05dba41... | 1 
          chocolate-doom-2.2.1 | 1 
                 opencv3-3.1.0 | 1 
             ipsec-tools-0.8.2 | 1 
mtr-66de3ecbab28b054b868a73... | 1 
    erlang-p1-stringprep-1.0.0 | 1 

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

       sparc | android-tools-4.2.2+git2013... | NOK | http://autobuild.buildroot.net/results/d0b84c6b512f530c30a917b2f5abd2bc9dadd112/
       sparc | android-tools-4.2.2+git2013... | NOK | http://autobuild.buildroot.net/results/d3099b8fd163772d5601a081c8125a59d186e8cd/
      xtensa | android-tools-4.2.2+git2013... | NOK | http://autobuild.buildroot.net/results/3c2aaab721a29399e9a945c3a7645c54b3e90410/
     powerpc | android-tools-4.2.2+git2013... | NOK | http://autobuild.buildroot.net/results/ce9c0d25448cddc29c398f408a7c5a238451b179/
     powerpc | android-tools-4.2.2+git2013... | NOK | http://autobuild.buildroot.net/results/b17b75bb211d0a0ff2dbc75756809775b17d5637/
microblazeel |                   boost-1.60.0 | NOK | http://autobuild.buildroot.net/results/0b9db4c5cfd80d75eb620dbf7a6201faed7230a1/
       nios2 |                   boost-1.60.0 | NOK | http://autobuild.buildroot.net/results/9b8c5d2cd31d63ae70b41d1434868c83e65b8423/
         arm |           chocolate-doom-2.2.1 | NOK | http://autobuild.buildroot.net/results/31ec8fed2390b8fece70ec7168019c736fe66b01/
         arm |                 collectd-5.5.1 | NOK | http://autobuild.buildroot.net/results/80a75aaebd55fca1220048d1d4dfb434ce4d9804/
      x86_64 |                 collectd-5.5.1 | NOK | http://autobuild.buildroot.net/results/ae1e3a33a1c2b7433ca7a04659467e2017756b00/
      x86_64 |                   connman-1.32 | NOK | http://autobuild.buildroot.net/results/12c91b8ae633112504c5aead84301b3107c23ac6/
         arm |                     cups-2.1.2 | NOK | http://autobuild.buildroot.net/results/a72771300d817ba8d411abd4fa1d232911422f5e/
         arm |                     cups-2.1.2 | NOK | http://autobuild.buildroot.net/results/a8e7bb8f7616aedbf71e7c653004c6ffe00c0790/
         arm |                     cups-2.1.2 | NOK | http://autobuild.buildroot.net/results/fe5e5e4a46007150692cbbc6652a6f2c3a1ee6b7/
         arm |     erlang-p1-stringprep-1.0.0 | NOK | http://autobuild.buildroot.net/results/a8b7208ef9068866365843b11b99d24183d4c7ea/
    mips64el |                   ffmpeg-2.8.7 | NOK | http://autobuild.buildroot.net/results/0f30c6dbf2186af5bfda7fe051db385a389b2251/
    mips64el |                   ffmpeg-2.8.7 | NOK | http://autobuild.buildroot.net/results/ee836d09aadf7fb43f25d59bdcd0658f55844083/
microblazeel | flann-04b4a56533faf8c8228d0... | NOK | http://autobuild.buildroot.net/results/3f44a1f30a88dbe4a3a83055267b472b58769e15/
microblazeel | flann-04b4a56533faf8c8228d0... | NOK | http://autobuild.buildroot.net/results/fc2446484d4167c4f99c4aeccf3532d021d02046/
      x86_64 |                 flannel-v0.5.5 | NOK | http://autobuild.buildroot.net/results/6664189a6f3a815978e8d0a1d7ef408ca47e2874/
     powerpc |        gadgetfs-test-undefined | NOK | http://autobuild.buildroot.net/results/163fbc2ccd88282be2f5d5354b125429bfb4ff4b/
      x86_64 |               gnuradio-3.7.9.1 | NOK | http://autobuild.buildroot.net/results/714019031fb8c8f70c1838aa18e4ec1e4abf165e/
         arc |       host-gdb-arc-2015.12-gdb | NOK | http://autobuild.buildroot.net/results/c6d9ee66842fbb2457277d02d241c439b43bd568/
         arm |                  host-go-1.6.2 | NOK | http://autobuild.buildroot.net/results/42a8d07101d8d954511d1c884ecb66e8d861899e/
         arm |              ipsec-tools-0.8.2 | NOK | http://autobuild.buildroot.net/results/9e68e8bf8b51c7d8fc0a2aff1ac617dfaa36b648/
    mips64el |                  jack2-v1.9.10 | NOK | http://autobuild.buildroot.net/results/43ca1b103434ae582fdf93cb5912b311960f303b/
         arm |                  jack2-v1.9.10 | NOK | http://autobuild.buildroot.net/results/415e2100dc59d35e58646c07f7cdccabecdda966/
         arm |                  jack2-v1.9.10 | NOK | http://autobuild.buildroot.net/results/65edaf15180135b4a30b72ea92c35ae16f001a59/
        i686 |                  jack2-v1.9.10 | NOK | http://autobuild.buildroot.net/results/391e71a988250ea66ec4dbee6f60fdce9eaf2766/
      mipsel |           libpam-tacplus-1.3.9 | NOK | http://autobuild.buildroot.net/results/ac7c9b3ad2e52abfe6b79a80045e4218eeb87175/
      mipsel |           libpam-tacplus-1.3.9 | NOK | http://autobuild.buildroot.net/results/1a778bbcc72a8f0df3b0d25a49c03a25db1b440d/
    mips64el | liquid-dsp-df5a459fa05dba41... | NOK | http://autobuild.buildroot.net/results/1482e2702553f083a500c055c2ef8e6ae2a85182/
        i686 |              lirc-tools-0.9.3a | NOK | http://autobuild.buildroot.net/results/37b5fdc0fa320aad98493f7ae7d4e70a84852266/
        sh4a |              lirc-tools-0.9.3a | NOK | http://autobuild.buildroot.net/results/81e9bfc79d6b4ebba7ead62b7abac68bd9968e88/
         arm |         lttng-babeltrace-1.2.4 | NOK | http://autobuild.buildroot.net/results/0d247c58f788282f92112ac1afb616fa2631c32c/
         sh4 |                     mpv-0.17.0 | NOK | http://autobuild.buildroot.net/results/78e898689c3de1549d92f00b0d39cc0771417027/
         sh4 |                     mpv-0.17.0 | NOK | http://autobuild.buildroot.net/results/7982e81a850f579d51021f620cd3c1e8bd641771/
         sh4 |                     mpv-0.17.0 | NOK | http://autobuild.buildroot.net/results/bf45e609befd6a1c00dc75578f4c363657958380/
         arm |                     mpv-0.17.0 | NOK | http://autobuild.buildroot.net/results/97491961a9a04de03119c08ac21e127971c1202a/
        bfin | mtr-66de3ecbab28b054b868a73... | NOK | http://autobuild.buildroot.net/results/e05632e0b160ee995ce46aaf3ceddf46d3658adf/
         sh4 |                  opencv3-3.1.0 | NOK | http://autobuild.buildroot.net/results/a5a32e8b18583f0c40da22e735ef5e2ed1e415a6/
microblazeel |                 oprofile-1.1.0 | NOK | http://autobuild.buildroot.net/results/1fd1eff35d7945f352561d0eb9bb5d7360942a42/
         arm |                     php-5.6.21 | NOK | http://autobuild.buildroot.net/results/c67d00b19da8bcb2e6070d3f048931b36d5ad467/
      x86_64 |                     pppd-2.4.7 | NOK | http://autobuild.buildroot.net/results/32bab29b2ed19ebf18d34f7dc3324161c89dbda9/
      mipsel |                   psmisc-22.21 | NOK | http://autobuild.buildroot.net/results/382094c3dedfdb3e2489851d8fb1255e440ce39c/
        sh4a |             qt5serialbus-5.6.0 | NOK | http://autobuild.buildroot.net/results/6176a24799b5d80312be954089efbbb498fa1571/
         arc |                  rsyslog-8.9.0 | NOK | http://autobuild.buildroot.net/results/08b5fa07c85e38a771f7d52e2e09fb62901e156c/
      xtensa |                  rsyslog-8.9.0 | NOK | http://autobuild.buildroot.net/results/abb2e06d0b19922627d9a76c75858b86eca1cc2d/
     sparc64 |              stress-ng-0.04.16 | NOK | http://autobuild.buildroot.net/results/99728d80caf490ad2cba78b9c50ec5385c5e6601/
     sparc64 |              stress-ng-0.04.16 | NOK | http://autobuild.buildroot.net/results/9134c246e451061b2ad42d0508d073f18f19961a/
     sparc64 |              stress-ng-0.04.16 | NOK | http://autobuild.buildroot.net/results/c22585c193c48cb1d2ad3bb7ecde73620562d7c8/
       sparc |               strongswan-5.4.0 | NOK | http://autobuild.buildroot.net/results/5859b8e0a1775ecd6afa13132549966080739837/
         arc |                    taskd-1.1.0 | NOK | http://autobuild.buildroot.net/results/67f574af0cc4348ffe2bce026c44766e49c29124/
    mips64el |  toolchain-external-legal-info | NOK | http://autobuild.buildroot.net/results/b047bfa84af0e88f4efa5e1ed7ea5639d120b947/
         arm |                     ustr-1.0.4 | NOK | http://autobuild.buildroot.net/results/cebfd7e29a6521ae7ba4c229586a3ef662aa9de1/
    mips64el |                valgrind-3.11.0 | NOK | http://autobuild.buildroot.net/results/a1168399adb9fd549d610e4d18aa024d9c8bcbff/
    mips64el |                valgrind-3.11.0 | NOK | http://autobuild.buildroot.net/results/61eb30d455cf767f4317b0898c2219b2ab695fc0/
         arm |                      vlc-2.2.3 | NOK | http://autobuild.buildroot.net/results/1a7893c13f6144e0ecd9f37f2f047c596ccedd35/
        bfin |                 wayland-1.10.0 | NOK | http://autobuild.buildroot.net/results/0a5d32f0fa28ecdd540e100951b116708feec6b5/
        bfin |                 wayland-1.10.0 | NOK | http://autobuild.buildroot.net/results/75286a0319094a82934eebeeda6ac4f0b0f36743/


-- 
http://autobuild.buildroot.net

[Buildroot] Analysis of build results for 2016-05-16

[Thomas Petazzoni]

Hello,

Rodrigo, Geoff, Gustavo, Vicente, Gwenhael, Samuel, Yegor, Ben,
Clayton, Bernd, please see below, there are questions/issues for you.

Note to others: there are also plenty of build failures below with
nobody assigned. Feel free to pick some of them and investigate. Thanks!

On Tue, 17 May 2016 08:30:22 +0200 (CEST), Thomas Petazzoni wrote:

>        sparc | android-tools-4.2.2+git2013... | NOK | http://autobuild.buildroot.net/results/d0b84c6b512f530c30a917b2f5abd2bc9dadd112/
>        sparc | android-tools-4.2.2+git2013... | NOK | http://autobuild.buildroot.net/results/d3099b8fd163772d5601a081c8125a59d186e8cd/
>       xtensa | android-tools-4.2.2+git2013... | NOK | http://autobuild.buildroot.net/results/3c2aaab721a29399e9a945c3a7645c54b3e90410/
>      powerpc | android-tools-4.2.2+git2013... | NOK | http://autobuild.buildroot.net/results/ce9c0d25448cddc29c398f408a7c5a238451b179/
>      powerpc | android-tools-4.2.2+git2013... | NOK | http://autobuild.buildroot.net/results/b17b75bb211d0a0ff2dbc75756809775b17d5637/

Those would all be fixed by applying
http://patchwork.ozlabs.org/patch/622247/.

> microblazeel |                   boost-1.60.0 | NOK | http://autobuild.buildroot.net/results/0b9db4c5cfd80d75eb620dbf7a6201faed7230a1/
>        nios2 |                   boost-1.60.0 | NOK | http://autobuild.buildroot.net/results/9b8c5d2cd31d63ae70b41d1434868c83e65b8423/

Fixed by
https://git.buildroot.net/buildroot/commit/?id=f01f2cde09527319c079d1a076143334411912a3.

>          arm |           chocolate-doom-2.2.1 | NOK | http://autobuild.buildroot.net/results/31ec8fed2390b8fece70ec7168019c736fe66b01/

This is a long-standing issue of static linking with chocolate-doom, it
forgets to link with libstdc++. Rodrigo, you contributed this package,
and already fixed some static linking issues with it. Can you look at
this one, or alternatively decide that we don't want to support static
linking for chocolate-doom?

>          arm |                 collectd-5.5.1 | NOK | http://autobuild.buildroot.net/results/80a75aaebd55fca1220048d1d4dfb434ce4d9804/
>       x86_64 |                 collectd-5.5.1 | NOK | http://autobuild.buildroot.net/results/ae1e3a33a1c2b7433ca7a04659467e2017756b00/

Gustavo, any solution in sight for this collectd problem?

>       x86_64 |                   connman-1.32 | NOK | http://autobuild.buildroot.net/results/12c91b8ae633112504c5aead84301b3107c23ac6/

I guess this is a musl build issue. I don't think we have a patch
pending for this one. Contributions welcome.

>          arm |                     cups-2.1.2 | NOK | http://autobuild.buildroot.net/results/a72771300d817ba8d411abd4fa1d232911422f5e/
>          arm |                     cups-2.1.2 | NOK | http://autobuild.buildroot.net/results/a8e7bb8f7616aedbf71e7c653004c6ffe00c0790/
>          arm |                     cups-2.1.2 | NOK | http://autobuild.buildroot.net/results/fe5e5e4a46007150692cbbc6652a6f2c3a1ee6b7/

Those would be fixed by applying
http://patchwork.ozlabs.org/patch/622248/.

>          arm |     erlang-p1-stringprep-1.0.0 | NOK | http://autobuild.buildroot.net/results/a8b7208ef9068866365843b11b99d24183d4c7ea/

This one would be fixed by applying
http://patchwork.ozlabs.org/patch/622682/.

>     mips64el |                   ffmpeg-2.8.7 | NOK | http://autobuild.buildroot.net/results/0f30c6dbf2186af5bfda7fe051db385a389b2251/
>     mips64el |                   ffmpeg-2.8.7 | NOK | http://autobuild.buildroot.net/results/ee836d09aadf7fb43f25d59bdcd0658f55844083/

Vicente, could you help with this one?

> microblazeel | flann-04b4a56533faf8c8228d0... | NOK | http://autobuild.buildroot.net/results/3f44a1f30a88dbe4a3a83055267b472b58769e15/
> microblazeel | flann-04b4a56533faf8c8228d0... | NOK | http://autobuild.buildroot.net/results/fc2446484d4167c4f99c4aeccf3532d021d02046/

I've started a build with gcc 6.x on Microblaze to see if the problem
still exists. If it does, I will report a bug to the gcc developers and
send a patch to mark this package as not available on Microblaze.

>       x86_64 |                 flannel-v0.5.5 | NOK | http://autobuild.buildroot.net/results/6664189a6f3a815978e8d0a1d7ef408ca47e2874/

Geoff, can you look into this one?

>      powerpc |        gadgetfs-test-undefined | NOK | http://autobuild.buildroot.net/results/163fbc2ccd88282be2f5d5354b125429bfb4ff4b/

Gustavo, another _restgpr_29_x issue, and this is with a Buildroot
toolchain, so we can't blame an old CT-NG or Sourcery toolchain here :)

>       x86_64 |               gnuradio-3.7.9.1 | NOK | http://autobuild.buildroot.net/results/714019031fb8c8f70c1838aa18e4ec1e4abf165e/

internal compiler error: Segmentation fault

Gwenhael, can you look into this problem?

>          arc |       host-gdb-arc-2015.12-gdb | NOK | http://autobuild.buildroot.net/results/c6d9ee66842fbb2457277d02d241c439b43bd568/

Missing dependency on flex it seems. Probably we need to do like we do
for binutils, and depend on host-flex when the gdb version is fetched
from Git.

>          arm |                  host-go-1.6.2 | NOK | http://autobuild.buildroot.net/results/42a8d07101d8d954511d1c884ecb66e8d861899e/

error: #warning requested reentrant code

Geoff, in fact it seems like host-go itself needs thread support in the
target toolchain to build properly, so our plan to have only the Go
packages depend on BR2_TOOLCHAIN_HAS_THREADS does not work.

Can you cook a solution for this?

>          arm |              ipsec-tools-0.8.2 | NOK | http://autobuild.buildroot.net/results/9e68e8bf8b51c7d8fc0a2aff1ac617dfaa36b648/

Musl build issue.

>     mips64el |                  jack2-v1.9.10 | NOK | http://autobuild.buildroot.net/results/43ca1b103434ae582fdf93cb5912b311960f303b/
>          arm |                  jack2-v1.9.10 | NOK | http://autobuild.buildroot.net/results/415e2100dc59d35e58646c07f7cdccabecdda966/
>          arm |                  jack2-v1.9.10 | NOK | http://autobuild.buildroot.net/results/65edaf15180135b4a30b72ea92c35ae16f001a59/
>         i686 |                  jack2-v1.9.10 | NOK | http://autobuild.buildroot.net/results/391e71a988250ea66ec4dbee6f60fdce9eaf2766/

Fixed by
https://git.buildroot.net/buildroot/commit/?id=b7b9426ccc9bc854e337fec8bcfdda150762bcf6.

>       mipsel |           libpam-tacplus-1.3.9 | NOK | http://autobuild.buildroot.net/results/ac7c9b3ad2e52abfe6b79a80045e4218eeb87175/
>       mipsel |           libpam-tacplus-1.3.9 | NOK | http://autobuild.buildroot.net/results/1a778bbcc72a8f0df3b0d25a49c03a25db1b440d/

SSP issue with the MIPS toolchains. Vicente?

>     mips64el | liquid-dsp-df5a459fa05dba41... | NOK | http://autobuild.buildroot.net/results/1482e2702553f083a500c055c2ef8e6ae2a85182/

src/math/src/poly.findroots.c:305:9: error: non-floating-point argument in call to function '__builtin_isnan'

Vicente?

>         i686 |              lirc-tools-0.9.3a | NOK | http://autobuild.buildroot.net/results/37b5fdc0fa320aad98493f7ae7d4e70a84852266/
>         sh4a |              lirc-tools-0.9.3a | NOK | http://autobuild.buildroot.net/results/81e9bfc79d6b4ebba7ead62b7abac68bd9968e88/

ld: cannot find -llirc

Weird, needs investigation. Parallel build issue, maybe?

>          arm |         lttng-babeltrace-1.2.4 | NOK | http://autobuild.buildroot.net/results/0d247c58f788282f92112ac1afb616fa2631c32c/

checking for uuid_create in -lc... no
configure: error: Cannot find libuuid uuid_generate nor libc uuid_create. Use LDFLAGS=-Ldir to specify their location.

No config.log has been saved. Samuel, can you have a look?

>          sh4 |                     mpv-0.17.0 | NOK | http://autobuild.buildroot.net/results/78e898689c3de1549d92f00b0d39cc0771417027/
>          sh4 |                     mpv-0.17.0 | NOK | http://autobuild.buildroot.net/results/7982e81a850f579d51021f620cd3c1e8bd641771/
>          sh4 |                     mpv-0.17.0 | NOK | http://autobuild.buildroot.net/results/bf45e609befd6a1c00dc75578f4c363657958380/
>          arm |                     mpv-0.17.0 | NOK | http://autobuild.buildroot.net/results/97491961a9a04de03119c08ac21e127971c1202a/

Some of these will be fixed by the Buildroot toolchains rebuild that
was deployed today. Let's wait and see if we still have build failures.

>         bfin | mtr-66de3ecbab28b054b868a73... | NOK | http://autobuild.buildroot.net/results/e05632e0b160ee995ce46aaf3ceddf46d3658adf/

Uses fork(), needs a dependency on BR2_USE_MMU.

>          sh4 |                  opencv3-3.1.0 | NOK | http://autobuild.buildroot.net/results/a5a32e8b18583f0c40da22e735ef5e2ed1e415a6/

Toolchain issue:

/tmp/ccF0zot6.s: Assembler messages:
/tmp/ccF0zot6.s:19767: Error: pcrel too far

Samuel, can you retry with the latest gcc/binutils version on SH4 and
see if it fixes the problem?

> microblazeel |                 oprofile-1.1.0 | NOK | http://autobuild.buildroot.net/results/1fd1eff35d7945f352561d0eb9bb5d7360942a42/

operf_utils.cpp: In function 'void OP_perf_utils::op_get_kernel_event_data(mmap_data*, operf_record*)':
operf_utils.cpp:1412:6: error: 'rmb' was not declared in this scope
  rmb();

Architecture not supported it seems.

>          arm |                     php-5.6.21 | NOK | http://autobuild.buildroot.net/results/c67d00b19da8bcb2e6070d3f048931b36d5ad467/

Forgets to link with pthread. Gustavo?

>       x86_64 |                     pppd-2.4.7 | NOK | http://autobuild.buildroot.net/results/32bab29b2ed19ebf18d34f7dc3324161c89dbda9/

Musl build issue. Anyone to look into this?

>       mipsel |                   psmisc-22.21 | NOK | http://autobuild.buildroot.net/results/382094c3dedfdb3e2489851d8fb1255e440ce39c/

SSP MIPS toolchain issue...

>         sh4a |             qt5serialbus-5.6.0 | NOK | http://autobuild.buildroot.net/results/6176a24799b5d80312be954089efbbb498fa1571/

Seems like this package needs a dependency on some recent enough kernel
headers. Yegor, since you added this package, can you have a look?

>          arc |                  rsyslog-8.9.0 | NOK | http://autobuild.buildroot.net/results/08b5fa07c85e38a771f7d52e2e09fb62901e156c/
>       xtensa |                  rsyslog-8.9.0 | NOK | http://autobuild.buildroot.net/results/abb2e06d0b19922627d9a76c75858b86eca1cc2d/

configure: error: in `/home/buildroot/autobuild/run/instance-0/output/build/rsyslog-8.9.0':
configure: error: PgSQL library is missing

Anyone to look into this?

>      sparc64 |              stress-ng-0.04.16 | NOK | http://autobuild.buildroot.net/results/99728d80caf490ad2cba78b9c50ec5385c5e6601/
>      sparc64 |              stress-ng-0.04.16 | NOK | http://autobuild.buildroot.net/results/9134c246e451061b2ad42d0508d073f18f19961a/
>      sparc64 |              stress-ng-0.04.16 | NOK | http://autobuild.buildroot.net/results/c22585c193c48cb1d2ad3bb7ecde73620562d7c8/

This should be fixed by https://git.buildroot.net/buildroot/commit/?id=08fe324b8b0e12a2ada22db8ee268a99cdde4322.

>        sparc |               strongswan-5.4.0 | NOK | http://autobuild.buildroot.net/results/5859b8e0a1775ecd6afa13132549966080739837/

../../src/libstrongswan/.libs/libstrongswan.so: undefined reference to `__atomic_fetch_sub_4'
../../src/libstrongswan/.libs/libstrongswan.so: undefined reference to `__atomic_compare_exchange_4'
../../src/libstrongswan/.libs/libstrongswan.so: undefined reference to `__atomic_fetch_add_4'

Gustavo?

>          arc |                    taskd-1.1.0 | NOK | http://autobuild.buildroot.net/results/67f574af0cc4348ffe2bce026c44766e49c29124/

Forgets to link with pthread. Ben, since you added this package, can
you have a look?

>     mips64el |  toolchain-external-legal-info | NOK | http://autobuild.buildroot.net/results/b047bfa84af0e88f4efa5e1ed7ea5639d120b947/

Network issue, Peter has already added the file to sources.b.o to avoid
the problem in the future.

>          arm |                     ustr-1.0.4 | NOK | http://autobuild.buildroot.net/results/cebfd7e29a6521ae7ba4c229586a3ef662aa9de1/

libustr-debug-1.0.so.1': File exists

Gah, smells like parallel installation issue. Clayton, since you added
this package, can you have a look?

>     mips64el |                valgrind-3.11.0 | NOK | http://autobuild.buildroot.net/results/a1168399adb9fd549d610e4d18aa024d9c8bcbff/
>     mips64el |                valgrind-3.11.0 | NOK | http://autobuild.buildroot.net/results/61eb30d455cf767f4317b0898c2219b2ab695fc0/

Vicente?

>          arm |                      vlc-2.2.3 | NOK | http://autobuild.buildroot.net/results/1a7893c13f6144e0ecd9f37f2f047c596ccedd35/

ld: cannot find -lGL

Bernd?

>         bfin |                 wayland-1.10.0 | NOK | http://autobuild.buildroot.net/results/0a5d32f0fa28ecdd540e100951b116708feec6b5/
>         bfin |                 wayland-1.10.0 | NOK | http://autobuild.buildroot.net/results/75286a0319094a82934eebeeda6ac4f0b0f36743/

I've looked into this, it's the infamous _ prefix issue on Blackfin. It
should be relatively easy to fix, I'll send a patch.

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

[Buildroot] Analysis of build results for 2016-05-16

[Rodrigo Rebello]

Thomas,

2016-05-17 10:36 GMT-03:00 Thomas Petazzoni
<thomas.petazzoni@free-electrons.com>:
[snip]
>>          arm |           chocolate-doom-2.2.1 | NOK | http://autobuild.buildroot.net/results/31ec8fed2390b8fece70ec7168019c736fe66b01/
>
> This is a long-standing issue of static linking with chocolate-doom, it
> forgets to link with libstdc++. Rodrigo, you contributed this package,
> and already fixed some static linking issues with it. Can you look at
> this one, or alternatively decide that we don't want to support static
> linking for chocolate-doom?
>

Sure. I'll have a look into it later when I get back from work.

Regards,
Rodrigo

[Buildroot] Analysis of build results for 2016-05-16

[gwenhael.goavec]

On Tue, 17 May 2016 15:36:11 +0200
Thomas Petazzoni <thomas.petazzoni@free-electrons.com> wrote:

> Hello,
> 
> Rodrigo, Geoff, Gustavo, Vicente, Gwenhael, Samuel, Yegor, Ben,
> Clayton, Bernd, please see below, there are questions/issues for you.
> 
> Note to others: there are also plenty of build failures below with
> nobody assigned. Feel free to pick some of them and investigate.
> Thanks!
> 
[..]
> 
> >       x86_64 |               gnuradio-3.7.9.1 | NOK |
> > http://autobuild.buildroot.net/results/714019031fb8c8f70c1838aa18e4ec1e4abf165e/  
> 
> internal compiler error: Segmentation fault
> 
> Gwenhael, can you look into this problem?
> 
I'm looking into this problem :
- Fails with a binary toolchain (gcc-4.7.2);
- Works with a toolchain generated by buildroot (same options but
  4.8.5 and 4.9.3).
I need to see if it's due to the gcc version or something else.

Gwen

[Buildroot] Analysis of build results for 2016-05-16

[Thomas Petazzoni]

Hello,

On Tue, 17 May 2016 17:13:50 +0200, gwenhael.goavec wrote:

> > >       x86_64 |               gnuradio-3.7.9.1 | NOK |
> > > http://autobuild.buildroot.net/results/714019031fb8c8f70c1838aa18e4ec1e4abf165e/    
> > 
> > internal compiler error: Segmentation fault
> > 
> > Gwenhael, can you look into this problem?
> >   
> I'm looking into this problem :
> - Fails with a binary toolchain (gcc-4.7.2);
> - Works with a toolchain generated by buildroot (same options but
>   4.8.5 and 4.9.3).
> I need to see if it's due to the gcc version or something else.

Potentially seems like a bug in gcc 4.7.x, which has been fixed in gcc
4.8.x, no? We still support gcc 4.7.x in the internal toolchain
backend, if you enable "deprecated options". This way you can check if
a Buildroot toolchain with gcc 4.7 exhibits the same problem.

Note that this kind of issue is often sensitive to optimization flags:
i.e it might fail at -O3 but succeed at -Os.

If we conclude that 4.7 is broken, and >= 4.8 is working, then adding a
dependency on BR2_TOOLCHAIN_GCC_AT_LEAST_4_8 is good enough for me.

Best regards,

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

[Buildroot] Analysis of build results for 2016-05-16

[Bernd Kuhls]

[posted and mailed]

Hi Thomas,

Thomas Petazzoni <thomas.petazzoni@free-electrons.com> wrote 
in news:20160517153611.2a3b4199 at free-electrons.com:
 
>>         bfin | mtr-66de3ecbab28b054b868a73... | NOK | 
http://autobuild.buildroot.net/results/e05632e0b160ee995ce46aaf3ceddf46d3658adf/
> 
> Uses fork(), needs a dependency on BR2_USE_MMU.

fixed by http://patchwork.ozlabs.org/patch/623207/

Regards, Bernd

[Buildroot] Analysis of build results for 2016-05-16

[Bernd Kuhls]

Hi Thomas,

[posted and mailed]

Thomas Petazzoni
<thomas.petazzoni@free-electrons.com> wrote
in news:20160517153611.2a3b4199 at free-electrons.com: 

>>          arm |                      vlc-2.2.3 | NOK |
>>          http://autobuild.buildroot.net/results/1a7893c13f6144e0ecd9f37f
>>          2f047c596ccedd35/ 
> 
> ld: cannot find -lGL
> 
> Bernd?

should be fixed by http://patchwork.ozlabs.org/patch/623236/

VLC has no switch for detecting OpenGL support and searches only for gl.pc

http://git.videolan.org/?p=vlc/vlc-2.2.git;a=blob;f=configure.ac;h=e021546e
284d7daefdb4d94f430eb0bc57722210;hb=HEAD#l3079

which needs to be removed by the mesa3d package when OpenGL is not
available. 

Regards, Bernd

[Buildroot] Analysis of build results for 2016-05-16

[Geoff Levand]

Hi,

I have fixes for these, but have not yet found time to do
some final cleanup before submitting them.  Current patches
are in my master branch.

  https://github.com/glevand/buildroot--buildroot/commits/master

On Tue, 2016-05-17 at 15:36 +0200, Thomas Petazzoni wrote:
>       x86_64 |                 flannel-v0.5.5 | NOK | http://autobuild.buildroot.net/results/6664189a6f3a815978e8d0a1d7ef408ca47e2874/
> 
> Geoff, can you look into this one?

I believe I have the fix for this one in my patch 'package/go: Build
special host binaries'.  This is a reworked version of what I posted
last time.

> >          arm |                  host-go-1.6.2 | NOK | http://autobuild.buildroot.net/results/42a8d07101d8d954511d1c884ecb66e8d861899e/
> 
> error: #warning requested reentrant code
> 
> Geoff, in fact it seems like host-go itself needs thread support in the
> target toolchain to build properly, so our plan to have only the Go
> packages depend on BR2_TOOLCHAIN_HAS_THREADS does not work.

Yes, the solution I came up with is that the go target package must
specify BR2_TOOLCHAIN_HAS_THREADS if it uses cgo support.  If
BR2_TOOLCHAIN_HAS_THREADS is set, then host-go sets CGO_ENABLED=1 and
builds a compiler with cgo support.  If BR2_TOOLCHAIN_HAS_THREADS is
not set, then host-go does not build in cgo support.  The two cases:

A go package which does not use cgo sets BR2_TOOLCHAIN_HAS_THREADS=n,
the host-go is not build with cgo support.  Both the target package
and host-go should build OK since no thread support is needed.

A package that uses cgo sets BR2_TOOLCHAIN_HAS_THREADS=y, host-go
is build with cgo support.  Both build OK since the toolchain has
thread support.

The patches for this are in (package/go: Add HOST_GO_CGO_ENABLED) and
(package/flannel: Add BR2_TOOLCHAIN_HAS_THREADS), and a non-cgo test
program is in (go-sample: new package).

-Geoff

[Buildroot] Analysis of build results for 2016-05-16

[Thomas Petazzoni]

Hello,

On Tue, 17 May 2016 11:56:07 -0700, Geoff Levand wrote:

> On Tue, 2016-05-17 at 15:36 +0200, Thomas Petazzoni wrote:
> >       x86_64 |                 flannel-v0.5.5 | NOK | http://autobuild.buildroot.net/results/6664189a6f3a815978e8d0a1d7ef408ca47e2874/
> > 
> > Geoff, can you look into this one?  
> 
> I believe I have the fix for this one in my patch 'package/go: Build
> special host binaries'.  This is a reworked version of what I posted
> last time.

OK.

> > >          arm |                  host-go-1.6.2 | NOK | http://autobuild.buildroot.net/results/42a8d07101d8d954511d1c884ecb66e8d861899e/  
> > 
> > error: #warning requested reentrant code
> > 
> > Geoff, in fact it seems like host-go itself needs thread support in the
> > target toolchain to build properly, so our plan to have only the Go
> > packages depend on BR2_TOOLCHAIN_HAS_THREADS does not work.  
> 
> Yes, the solution I came up with is that the go target package must

By "go target package", you mean a target package that is written in
Go, and not specifically the package/go/ package (which doesn't have a
target variant), right?

> specify BR2_TOOLCHAIN_HAS_THREADS if it uses cgo support.  If
> BR2_TOOLCHAIN_HAS_THREADS is set, then host-go sets CGO_ENABLED=1 and
> builds a compiler with cgo support.  If BR2_TOOLCHAIN_HAS_THREADS is
> not set, then host-go does not build in cgo support.  The two cases:
> 
> A go package which does not use cgo sets BR2_TOOLCHAIN_HAS_THREADS=n,

A package cannot *set* BR2_TOOLCHAIN_HAS_THREADS.
BR2_TOOLCHAIN_HAS_THREADS is a property of the toolchain.

> the host-go is not build with cgo support.  Both the target package
> and host-go should build OK since no thread support is needed.
> 
> A package that uses cgo sets BR2_TOOLCHAIN_HAS_THREADS=y, host-go
> is build with cgo support.  Both build OK since the toolchain has
> thread support.

What needs to happen instead is the following (and is quite similar to
what you said, except for the interaction with
BR2_TOOLCHAIN_HAS_THREADS=y) :

 *) The host-go package will build cgo support when
    BR2_TOOLCHAIN_HAS_THREADS=y and otherwise not build cgo support.

 *) Any target package that is written in Go and that needs cgo support
    should have a "depends on BR2_TOOLCHAIN_HAS_THREADS" in its
    dependencies.

This needs to be explained by a comment in package/go/go.mk, as it is
not trivial.

Thanks!

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

[Buildroot] Analysis of build results for 2016-05-16

[Geoff Levand]

On Tue, 2016-05-17 at 21:34 +0200, Thomas Petazzoni wrote:
>          arm |                  host-go-1.6.2 | NOK | http://au
> > > > tobuild.buildroot.net/results/42a8d07101d8d954511d1c884ecb66e8d
> > > > 861899e/  ;
> > > 
> > > error: #warning requested reentrant code
> > > 
> > > Geoff, in fact it seems like host-go itself needs thread support
> > > in the
> > > target toolchain to build properly, so our plan to have only the
> > > Go
> > > packages depend on BR2_TOOLCHAIN_HAS_THREADS does not work.  
> > 
> > Yes, the solution I came up with is that the go target package must
> 
> By "go target package", you mean a target package that is written in
> Go, and not specifically the package/go/ package (which doesn't have
> a
> target variant), right?

Yes, a target package that has go source code.

> > specify BR2_TOOLCHAIN_HAS_THREADS if it uses cgo support.  If
> > BR2_TOOLCHAIN_HAS_THREADS is set, then host-go sets CGO_ENABLED=1
> > and
> > builds a compiler with cgo support.  If BR2_TOOLCHAIN_HAS_THREADS
> > is
> > not set, then host-go does not build in cgo support.  The two
> > cases:
> > 
> > A go package which does not use cgo sets
> > BR2_TOOLCHAIN_HAS_THREADS=n,
> 
> A package cannot *set* BR2_TOOLCHAIN_HAS_THREADS.
> BR2_TOOLCHAIN_HAS_THREADS is a property of the toolchain.

> > the host-go is not build with cgo support.  Both the target package
> > and host-go should build OK since no thread support is needed.
> > 
> > A package that uses cgo sets BR2_TOOLCHAIN_HAS_THREADS=y, host-go
> > is build with cgo support.  Both build OK since the toolchain has
> > thread support.
> 
> What needs to happen instead is the following (and is quite similar
> to
> what you said, except for the interaction with
> BR2_TOOLCHAIN_HAS_THREADS=y) :
> 
>  *) The host-go package will build cgo support when
>     BR2_TOOLCHAIN_HAS_THREADS=y and otherwise not build cgo support.
> 
>  *) Any target package that is written in Go and that needs cgo
> support
>     should have a "depends on BR2_TOOLCHAIN_HAS_THREADS" in its
>     dependencies.

Yes, sorry, that is correct.

> This needs to be explained by a comment in package/go/go.mk, as it is
> not trivial.

OK.

[Buildroot] Analysis of build results for 2016-05-16

[Yegor Yefremov]

On Tue, May 17, 2016 at 3:36 PM, Thomas Petazzoni
<thomas.petazzoni@free-electrons.com> wrote:
> Hello,
>
> Rodrigo, Geoff, Gustavo, Vicente, Gwenhael, Samuel, Yegor, Ben,
> Clayton, Bernd, please see below, there are questions/issues for you.
>
> Note to others: there are also plenty of build failures below with
> nobody assigned. Feel free to pick some of them and investigate. Thanks!
>
> On Tue, 17 May 2016 08:30:22 +0200 (CEST), Thomas Petazzoni wrote:
>
>>        sparc | android-tools-4.2.2+git2013... | NOK | http://autobuild.buildroot.net/results/d0b84c6b512f530c30a917b2f5abd2bc9dadd112/
>>        sparc | android-tools-4.2.2+git2013... | NOK | http://autobuild.buildroot.net/results/d3099b8fd163772d5601a081c8125a59d186e8cd/
>>       xtensa | android-tools-4.2.2+git2013... | NOK | http://autobuild.buildroot.net/results/3c2aaab721a29399e9a945c3a7645c54b3e90410/
>>      powerpc | android-tools-4.2.2+git2013... | NOK | http://autobuild.buildroot.net/results/ce9c0d25448cddc29c398f408a7c5a238451b179/
>>      powerpc | android-tools-4.2.2+git2013... | NOK | http://autobuild.buildroot.net/results/b17b75bb211d0a0ff2dbc75756809775b17d5637/
>
> Those would all be fixed by applying
> http://patchwork.ozlabs.org/patch/622247/.
>
>> microblazeel |                   boost-1.60.0 | NOK | http://autobuild.buildroot.net/results/0b9db4c5cfd80d75eb620dbf7a6201faed7230a1/
>>        nios2 |                   boost-1.60.0 | NOK | http://autobuild.buildroot.net/results/9b8c5d2cd31d63ae70b41d1434868c83e65b8423/
>
> Fixed by
> https://git.buildroot.net/buildroot/commit/?id=f01f2cde09527319c079d1a076143334411912a3.
>
>>          arm |           chocolate-doom-2.2.1 | NOK | http://autobuild.buildroot.net/results/31ec8fed2390b8fece70ec7168019c736fe66b01/
>
> This is a long-standing issue of static linking with chocolate-doom, it
> forgets to link with libstdc++. Rodrigo, you contributed this package,
> and already fixed some static linking issues with it. Can you look at
> this one, or alternatively decide that we don't want to support static
> linking for chocolate-doom?
>
>>          arm |                 collectd-5.5.1 | NOK | http://autobuild.buildroot.net/results/80a75aaebd55fca1220048d1d4dfb434ce4d9804/
>>       x86_64 |                 collectd-5.5.1 | NOK | http://autobuild.buildroot.net/results/ae1e3a33a1c2b7433ca7a04659467e2017756b00/
>
> Gustavo, any solution in sight for this collectd problem?
>
>>       x86_64 |                   connman-1.32 | NOK | http://autobuild.buildroot.net/results/12c91b8ae633112504c5aead84301b3107c23ac6/
>
> I guess this is a musl build issue. I don't think we have a patch
> pending for this one. Contributions welcome.
>
>>          arm |                     cups-2.1.2 | NOK | http://autobuild.buildroot.net/results/a72771300d817ba8d411abd4fa1d232911422f5e/
>>          arm |                     cups-2.1.2 | NOK | http://autobuild.buildroot.net/results/a8e7bb8f7616aedbf71e7c653004c6ffe00c0790/
>>          arm |                     cups-2.1.2 | NOK | http://autobuild.buildroot.net/results/fe5e5e4a46007150692cbbc6652a6f2c3a1ee6b7/
>
> Those would be fixed by applying
> http://patchwork.ozlabs.org/patch/622248/.
>
>>          arm |     erlang-p1-stringprep-1.0.0 | NOK | http://autobuild.buildroot.net/results/a8b7208ef9068866365843b11b99d24183d4c7ea/
>
> This one would be fixed by applying
> http://patchwork.ozlabs.org/patch/622682/.
>
>>     mips64el |                   ffmpeg-2.8.7 | NOK | http://autobuild.buildroot.net/results/0f30c6dbf2186af5bfda7fe051db385a389b2251/
>>     mips64el |                   ffmpeg-2.8.7 | NOK | http://autobuild.buildroot.net/results/ee836d09aadf7fb43f25d59bdcd0658f55844083/
>
> Vicente, could you help with this one?
>
>> microblazeel | flann-04b4a56533faf8c8228d0... | NOK | http://autobuild.buildroot.net/results/3f44a1f30a88dbe4a3a83055267b472b58769e15/
>> microblazeel | flann-04b4a56533faf8c8228d0... | NOK | http://autobuild.buildroot.net/results/fc2446484d4167c4f99c4aeccf3532d021d02046/
>
> I've started a build with gcc 6.x on Microblaze to see if the problem
> still exists. If it does, I will report a bug to the gcc developers and
> send a patch to mark this package as not available on Microblaze.
>
>>       x86_64 |                 flannel-v0.5.5 | NOK | http://autobuild.buildroot.net/results/6664189a6f3a815978e8d0a1d7ef408ca47e2874/
>
> Geoff, can you look into this one?
>
>>      powerpc |        gadgetfs-test-undefined | NOK | http://autobuild.buildroot.net/results/163fbc2ccd88282be2f5d5354b125429bfb4ff4b/
>
> Gustavo, another _restgpr_29_x issue, and this is with a Buildroot
> toolchain, so we can't blame an old CT-NG or Sourcery toolchain here :)
>
>>       x86_64 |               gnuradio-3.7.9.1 | NOK | http://autobuild.buildroot.net/results/714019031fb8c8f70c1838aa18e4ec1e4abf165e/
>
> internal compiler error: Segmentation fault
>
> Gwenhael, can you look into this problem?
>
>>          arc |       host-gdb-arc-2015.12-gdb | NOK | http://autobuild.buildroot.net/results/c6d9ee66842fbb2457277d02d241c439b43bd568/
>
> Missing dependency on flex it seems. Probably we need to do like we do
> for binutils, and depend on host-flex when the gdb version is fetched
> from Git.
>
>>          arm |                  host-go-1.6.2 | NOK | http://autobuild.buildroot.net/results/42a8d07101d8d954511d1c884ecb66e8d861899e/
>
> error: #warning requested reentrant code
>
> Geoff, in fact it seems like host-go itself needs thread support in the
> target toolchain to build properly, so our plan to have only the Go
> packages depend on BR2_TOOLCHAIN_HAS_THREADS does not work.
>
> Can you cook a solution for this?
>
>>          arm |              ipsec-tools-0.8.2 | NOK | http://autobuild.buildroot.net/results/9e68e8bf8b51c7d8fc0a2aff1ac617dfaa36b648/
>
> Musl build issue.
>
>>     mips64el |                  jack2-v1.9.10 | NOK | http://autobuild.buildroot.net/results/43ca1b103434ae582fdf93cb5912b311960f303b/
>>          arm |                  jack2-v1.9.10 | NOK | http://autobuild.buildroot.net/results/415e2100dc59d35e58646c07f7cdccabecdda966/
>>          arm |                  jack2-v1.9.10 | NOK | http://autobuild.buildroot.net/results/65edaf15180135b4a30b72ea92c35ae16f001a59/
>>         i686 |                  jack2-v1.9.10 | NOK | http://autobuild.buildroot.net/results/391e71a988250ea66ec4dbee6f60fdce9eaf2766/
>
> Fixed by
> https://git.buildroot.net/buildroot/commit/?id=b7b9426ccc9bc854e337fec8bcfdda150762bcf6.
>
>>       mipsel |           libpam-tacplus-1.3.9 | NOK | http://autobuild.buildroot.net/results/ac7c9b3ad2e52abfe6b79a80045e4218eeb87175/
>>       mipsel |           libpam-tacplus-1.3.9 | NOK | http://autobuild.buildroot.net/results/1a778bbcc72a8f0df3b0d25a49c03a25db1b440d/
>
> SSP issue with the MIPS toolchains. Vicente?
>
>>     mips64el | liquid-dsp-df5a459fa05dba41... | NOK | http://autobuild.buildroot.net/results/1482e2702553f083a500c055c2ef8e6ae2a85182/
>
> src/math/src/poly.findroots.c:305:9: error: non-floating-point argument in call to function '__builtin_isnan'
>
> Vicente?
>
>>         i686 |              lirc-tools-0.9.3a | NOK | http://autobuild.buildroot.net/results/37b5fdc0fa320aad98493f7ae7d4e70a84852266/
>>         sh4a |              lirc-tools-0.9.3a | NOK | http://autobuild.buildroot.net/results/81e9bfc79d6b4ebba7ead62b7abac68bd9968e88/
>
> ld: cannot find -llirc
>
> Weird, needs investigation. Parallel build issue, maybe?
>
>>          arm |         lttng-babeltrace-1.2.4 | NOK | http://autobuild.buildroot.net/results/0d247c58f788282f92112ac1afb616fa2631c32c/
>
> checking for uuid_create in -lc... no
> configure: error: Cannot find libuuid uuid_generate nor libc uuid_create. Use LDFLAGS=-Ldir to specify their location.
>
> No config.log has been saved. Samuel, can you have a look?
>
>>          sh4 |                     mpv-0.17.0 | NOK | http://autobuild.buildroot.net/results/78e898689c3de1549d92f00b0d39cc0771417027/
>>          sh4 |                     mpv-0.17.0 | NOK | http://autobuild.buildroot.net/results/7982e81a850f579d51021f620cd3c1e8bd641771/
>>          sh4 |                     mpv-0.17.0 | NOK | http://autobuild.buildroot.net/results/bf45e609befd6a1c00dc75578f4c363657958380/
>>          arm |                     mpv-0.17.0 | NOK | http://autobuild.buildroot.net/results/97491961a9a04de03119c08ac21e127971c1202a/
>
> Some of these will be fixed by the Buildroot toolchains rebuild that
> was deployed today. Let's wait and see if we still have build failures.
>
>>         bfin | mtr-66de3ecbab28b054b868a73... | NOK | http://autobuild.buildroot.net/results/e05632e0b160ee995ce46aaf3ceddf46d3658adf/
>
> Uses fork(), needs a dependency on BR2_USE_MMU.
>
>>          sh4 |                  opencv3-3.1.0 | NOK | http://autobuild.buildroot.net/results/a5a32e8b18583f0c40da22e735ef5e2ed1e415a6/
>
> Toolchain issue:
>
> /tmp/ccF0zot6.s: Assembler messages:
> /tmp/ccF0zot6.s:19767: Error: pcrel too far
>
> Samuel, can you retry with the latest gcc/binutils version on SH4 and
> see if it fixes the problem?
>
>> microblazeel |                 oprofile-1.1.0 | NOK | http://autobuild.buildroot.net/results/1fd1eff35d7945f352561d0eb9bb5d7360942a42/
>
> operf_utils.cpp: In function 'void OP_perf_utils::op_get_kernel_event_data(mmap_data*, operf_record*)':
> operf_utils.cpp:1412:6: error: 'rmb' was not declared in this scope
>   rmb();
>
> Architecture not supported it seems.
>
>>          arm |                     php-5.6.21 | NOK | http://autobuild.buildroot.net/results/c67d00b19da8bcb2e6070d3f048931b36d5ad467/
>
> Forgets to link with pthread. Gustavo?
>
>>       x86_64 |                     pppd-2.4.7 | NOK | http://autobuild.buildroot.net/results/32bab29b2ed19ebf18d34f7dc3324161c89dbda9/
>
> Musl build issue. Anyone to look into this?
>
>>       mipsel |                   psmisc-22.21 | NOK | http://autobuild.buildroot.net/results/382094c3dedfdb3e2489851d8fb1255e440ce39c/
>
> SSP MIPS toolchain issue...
>
>>         sh4a |             qt5serialbus-5.6.0 | NOK | http://autobuild.buildroot.net/results/6176a24799b5d80312be954089efbbb498fa1571/
>
> Seems like this package needs a dependency on some recent enough kernel
> headers. Yegor, since you added this package, can you have a look?

CANFD_MTU stuff seems to be added in kernel 3.5. How can I express
this in Buildroot dependency terms?

Yegor

>
>>          arc |                  rsyslog-8.9.0 | NOK | http://autobuild.buildroot.net/results/08b5fa07c85e38a771f7d52e2e09fb62901e156c/
>>       xtensa |                  rsyslog-8.9.0 | NOK | http://autobuild.buildroot.net/results/abb2e06d0b19922627d9a76c75858b86eca1cc2d/
>
> configure: error: in `/home/buildroot/autobuild/run/instance-0/output/build/rsyslog-8.9.0':
> configure: error: PgSQL library is missing
>
> Anyone to look into this?
>
>>      sparc64 |              stress-ng-0.04.16 | NOK | http://autobuild.buildroot.net/results/99728d80caf490ad2cba78b9c50ec5385c5e6601/
>>      sparc64 |              stress-ng-0.04.16 | NOK | http://autobuild.buildroot.net/results/9134c246e451061b2ad42d0508d073f18f19961a/
>>      sparc64 |              stress-ng-0.04.16 | NOK | http://autobuild.buildroot.net/results/c22585c193c48cb1d2ad3bb7ecde73620562d7c8/
>
> This should be fixed by https://git.buildroot.net/buildroot/commit/?id=08fe324b8b0e12a2ada22db8ee268a99cdde4322.
>
>>        sparc |               strongswan-5.4.0 | NOK | http://autobuild.buildroot.net/results/5859b8e0a1775ecd6afa13132549966080739837/
>
> ../../src/libstrongswan/.libs/libstrongswan.so: undefined reference to `__atomic_fetch_sub_4'
> ../../src/libstrongswan/.libs/libstrongswan.so: undefined reference to `__atomic_compare_exchange_4'
> ../../src/libstrongswan/.libs/libstrongswan.so: undefined reference to `__atomic_fetch_add_4'
>
> Gustavo?
>
>>          arc |                    taskd-1.1.0 | NOK | http://autobuild.buildroot.net/results/67f574af0cc4348ffe2bce026c44766e49c29124/
>
> Forgets to link with pthread. Ben, since you added this package, can
> you have a look?
>
>>     mips64el |  toolchain-external-legal-info | NOK | http://autobuild.buildroot.net/results/b047bfa84af0e88f4efa5e1ed7ea5639d120b947/
>
> Network issue, Peter has already added the file to sources.b.o to avoid
> the problem in the future.
>
>>          arm |                     ustr-1.0.4 | NOK | http://autobuild.buildroot.net/results/cebfd7e29a6521ae7ba4c229586a3ef662aa9de1/
>
> libustr-debug-1.0.so.1': File exists
>
> Gah, smells like parallel installation issue. Clayton, since you added
> this package, can you have a look?
>
>>     mips64el |                valgrind-3.11.0 | NOK | http://autobuild.buildroot.net/results/a1168399adb9fd549d610e4d18aa024d9c8bcbff/
>>     mips64el |                valgrind-3.11.0 | NOK | http://autobuild.buildroot.net/results/61eb30d455cf767f4317b0898c2219b2ab695fc0/
>
> Vicente?
>
>>          arm |                      vlc-2.2.3 | NOK | http://autobuild.buildroot.net/results/1a7893c13f6144e0ecd9f37f2f047c596ccedd35/
>
> ld: cannot find -lGL
>
> Bernd?
>
>>         bfin |                 wayland-1.10.0 | NOK | http://autobuild.buildroot.net/results/0a5d32f0fa28ecdd540e100951b116708feec6b5/
>>         bfin |                 wayland-1.10.0 | NOK | http://autobuild.buildroot.net/results/75286a0319094a82934eebeeda6ac4f0b0f36743/
>
> I've looked into this, it's the infamous _ prefix issue on Blackfin. It
> should be relatively easy to fix, I'll send a patch.
>
> Thomas
> --
> Thomas Petazzoni, CTO, Free Electrons
> Embedded Linux, Kernel and Android engineering
> http://free-electrons.com

[Buildroot] Analysis of build results for 2016-05-16

[Yegor Yefremov]

On Tue, May 17, 2016 at 11:03 PM, Yegor Yefremov
<yegorslists@googlemail.com> wrote:
> On Tue, May 17, 2016 at 3:36 PM, Thomas Petazzoni
> <thomas.petazzoni@free-electrons.com> wrote:
>> Hello,
>>
>> Rodrigo, Geoff, Gustavo, Vicente, Gwenhael, Samuel, Yegor, Ben,
>> Clayton, Bernd, please see below, there are questions/issues for you.
>>
>> Note to others: there are also plenty of build failures below with
>> nobody assigned. Feel free to pick some of them and investigate. Thanks!
>>
>> On Tue, 17 May 2016 08:30:22 +0200 (CEST), Thomas Petazzoni wrote:
>>
>>>        sparc | android-tools-4.2.2+git2013... | NOK | http://autobuild.buildroot.net/results/d0b84c6b512f530c30a917b2f5abd2bc9dadd112/
>>>        sparc | android-tools-4.2.2+git2013... | NOK | http://autobuild.buildroot.net/results/d3099b8fd163772d5601a081c8125a59d186e8cd/
>>>       xtensa | android-tools-4.2.2+git2013... | NOK | http://autobuild.buildroot.net/results/3c2aaab721a29399e9a945c3a7645c54b3e90410/
>>>      powerpc | android-tools-4.2.2+git2013... | NOK | http://autobuild.buildroot.net/results/ce9c0d25448cddc29c398f408a7c5a238451b179/
>>>      powerpc | android-tools-4.2.2+git2013... | NOK | http://autobuild.buildroot.net/results/b17b75bb211d0a0ff2dbc75756809775b17d5637/
>>
>> Those would all be fixed by applying
>> http://patchwork.ozlabs.org/patch/622247/.
>>
>>> microblazeel |                   boost-1.60.0 | NOK | http://autobuild.buildroot.net/results/0b9db4c5cfd80d75eb620dbf7a6201faed7230a1/
>>>        nios2 |                   boost-1.60.0 | NOK | http://autobuild.buildroot.net/results/9b8c5d2cd31d63ae70b41d1434868c83e65b8423/
>>
>> Fixed by
>> https://git.buildroot.net/buildroot/commit/?id=f01f2cde09527319c079d1a076143334411912a3.
>>
>>>          arm |           chocolate-doom-2.2.1 | NOK | http://autobuild.buildroot.net/results/31ec8fed2390b8fece70ec7168019c736fe66b01/
>>
>> This is a long-standing issue of static linking with chocolate-doom, it
>> forgets to link with libstdc++. Rodrigo, you contributed this package,
>> and already fixed some static linking issues with it. Can you look at
>> this one, or alternatively decide that we don't want to support static
>> linking for chocolate-doom?
>>
>>>          arm |                 collectd-5.5.1 | NOK | http://autobuild.buildroot.net/results/80a75aaebd55fca1220048d1d4dfb434ce4d9804/
>>>       x86_64 |                 collectd-5.5.1 | NOK | http://autobuild.buildroot.net/results/ae1e3a33a1c2b7433ca7a04659467e2017756b00/
>>
>> Gustavo, any solution in sight for this collectd problem?
>>
>>>       x86_64 |                   connman-1.32 | NOK | http://autobuild.buildroot.net/results/12c91b8ae633112504c5aead84301b3107c23ac6/
>>
>> I guess this is a musl build issue. I don't think we have a patch
>> pending for this one. Contributions welcome.
>>
>>>          arm |                     cups-2.1.2 | NOK | http://autobuild.buildroot.net/results/a72771300d817ba8d411abd4fa1d232911422f5e/
>>>          arm |                     cups-2.1.2 | NOK | http://autobuild.buildroot.net/results/a8e7bb8f7616aedbf71e7c653004c6ffe00c0790/
>>>          arm |                     cups-2.1.2 | NOK | http://autobuild.buildroot.net/results/fe5e5e4a46007150692cbbc6652a6f2c3a1ee6b7/
>>
>> Those would be fixed by applying
>> http://patchwork.ozlabs.org/patch/622248/.
>>
>>>          arm |     erlang-p1-stringprep-1.0.0 | NOK | http://autobuild.buildroot.net/results/a8b7208ef9068866365843b11b99d24183d4c7ea/
>>
>> This one would be fixed by applying
>> http://patchwork.ozlabs.org/patch/622682/.
>>
>>>     mips64el |                   ffmpeg-2.8.7 | NOK | http://autobuild.buildroot.net/results/0f30c6dbf2186af5bfda7fe051db385a389b2251/
>>>     mips64el |                   ffmpeg-2.8.7 | NOK | http://autobuild.buildroot.net/results/ee836d09aadf7fb43f25d59bdcd0658f55844083/
>>
>> Vicente, could you help with this one?
>>
>>> microblazeel | flann-04b4a56533faf8c8228d0... | NOK | http://autobuild.buildroot.net/results/3f44a1f30a88dbe4a3a83055267b472b58769e15/
>>> microblazeel | flann-04b4a56533faf8c8228d0... | NOK | http://autobuild.buildroot.net/results/fc2446484d4167c4f99c4aeccf3532d021d02046/
>>
>> I've started a build with gcc 6.x on Microblaze to see if the problem
>> still exists. If it does, I will report a bug to the gcc developers and
>> send a patch to mark this package as not available on Microblaze.
>>
>>>       x86_64 |                 flannel-v0.5.5 | NOK | http://autobuild.buildroot.net/results/6664189a6f3a815978e8d0a1d7ef408ca47e2874/
>>
>> Geoff, can you look into this one?
>>
>>>      powerpc |        gadgetfs-test-undefined | NOK | http://autobuild.buildroot.net/results/163fbc2ccd88282be2f5d5354b125429bfb4ff4b/
>>
>> Gustavo, another _restgpr_29_x issue, and this is with a Buildroot
>> toolchain, so we can't blame an old CT-NG or Sourcery toolchain here :)
>>
>>>       x86_64 |               gnuradio-3.7.9.1 | NOK | http://autobuild.buildroot.net/results/714019031fb8c8f70c1838aa18e4ec1e4abf165e/
>>
>> internal compiler error: Segmentation fault
>>
>> Gwenhael, can you look into this problem?
>>
>>>          arc |       host-gdb-arc-2015.12-gdb | NOK | http://autobuild.buildroot.net/results/c6d9ee66842fbb2457277d02d241c439b43bd568/
>>
>> Missing dependency on flex it seems. Probably we need to do like we do
>> for binutils, and depend on host-flex when the gdb version is fetched
>> from Git.
>>
>>>          arm |                  host-go-1.6.2 | NOK | http://autobuild.buildroot.net/results/42a8d07101d8d954511d1c884ecb66e8d861899e/
>>
>> error: #warning requested reentrant code
>>
>> Geoff, in fact it seems like host-go itself needs thread support in the
>> target toolchain to build properly, so our plan to have only the Go
>> packages depend on BR2_TOOLCHAIN_HAS_THREADS does not work.
>>
>> Can you cook a solution for this?
>>
>>>          arm |              ipsec-tools-0.8.2 | NOK | http://autobuild.buildroot.net/results/9e68e8bf8b51c7d8fc0a2aff1ac617dfaa36b648/
>>
>> Musl build issue.
>>
>>>     mips64el |                  jack2-v1.9.10 | NOK | http://autobuild.buildroot.net/results/43ca1b103434ae582fdf93cb5912b311960f303b/
>>>          arm |                  jack2-v1.9.10 | NOK | http://autobuild.buildroot.net/results/415e2100dc59d35e58646c07f7cdccabecdda966/
>>>          arm |                  jack2-v1.9.10 | NOK | http://autobuild.buildroot.net/results/65edaf15180135b4a30b72ea92c35ae16f001a59/
>>>         i686 |                  jack2-v1.9.10 | NOK | http://autobuild.buildroot.net/results/391e71a988250ea66ec4dbee6f60fdce9eaf2766/
>>
>> Fixed by
>> https://git.buildroot.net/buildroot/commit/?id=b7b9426ccc9bc854e337fec8bcfdda150762bcf6.
>>
>>>       mipsel |           libpam-tacplus-1.3.9 | NOK | http://autobuild.buildroot.net/results/ac7c9b3ad2e52abfe6b79a80045e4218eeb87175/
>>>       mipsel |           libpam-tacplus-1.3.9 | NOK | http://autobuild.buildroot.net/results/1a778bbcc72a8f0df3b0d25a49c03a25db1b440d/
>>
>> SSP issue with the MIPS toolchains. Vicente?
>>
>>>     mips64el | liquid-dsp-df5a459fa05dba41... | NOK | http://autobuild.buildroot.net/results/1482e2702553f083a500c055c2ef8e6ae2a85182/
>>
>> src/math/src/poly.findroots.c:305:9: error: non-floating-point argument in call to function '__builtin_isnan'
>>
>> Vicente?
>>
>>>         i686 |              lirc-tools-0.9.3a | NOK | http://autobuild.buildroot.net/results/37b5fdc0fa320aad98493f7ae7d4e70a84852266/
>>>         sh4a |              lirc-tools-0.9.3a | NOK | http://autobuild.buildroot.net/results/81e9bfc79d6b4ebba7ead62b7abac68bd9968e88/
>>
>> ld: cannot find -llirc
>>
>> Weird, needs investigation. Parallel build issue, maybe?
>>
>>>          arm |         lttng-babeltrace-1.2.4 | NOK | http://autobuild.buildroot.net/results/0d247c58f788282f92112ac1afb616fa2631c32c/
>>
>> checking for uuid_create in -lc... no
>> configure: error: Cannot find libuuid uuid_generate nor libc uuid_create. Use LDFLAGS=-Ldir to specify their location.
>>
>> No config.log has been saved. Samuel, can you have a look?
>>
>>>          sh4 |                     mpv-0.17.0 | NOK | http://autobuild.buildroot.net/results/78e898689c3de1549d92f00b0d39cc0771417027/
>>>          sh4 |                     mpv-0.17.0 | NOK | http://autobuild.buildroot.net/results/7982e81a850f579d51021f620cd3c1e8bd641771/
>>>          sh4 |                     mpv-0.17.0 | NOK | http://autobuild.buildroot.net/results/bf45e609befd6a1c00dc75578f4c363657958380/
>>>          arm |                     mpv-0.17.0 | NOK | http://autobuild.buildroot.net/results/97491961a9a04de03119c08ac21e127971c1202a/
>>
>> Some of these will be fixed by the Buildroot toolchains rebuild that
>> was deployed today. Let's wait and see if we still have build failures.
>>
>>>         bfin | mtr-66de3ecbab28b054b868a73... | NOK | http://autobuild.buildroot.net/results/e05632e0b160ee995ce46aaf3ceddf46d3658adf/
>>
>> Uses fork(), needs a dependency on BR2_USE_MMU.
>>
>>>          sh4 |                  opencv3-3.1.0 | NOK | http://autobuild.buildroot.net/results/a5a32e8b18583f0c40da22e735ef5e2ed1e415a6/
>>
>> Toolchain issue:
>>
>> /tmp/ccF0zot6.s: Assembler messages:
>> /tmp/ccF0zot6.s:19767: Error: pcrel too far
>>
>> Samuel, can you retry with the latest gcc/binutils version on SH4 and
>> see if it fixes the problem?
>>
>>> microblazeel |                 oprofile-1.1.0 | NOK | http://autobuild.buildroot.net/results/1fd1eff35d7945f352561d0eb9bb5d7360942a42/
>>
>> operf_utils.cpp: In function 'void OP_perf_utils::op_get_kernel_event_data(mmap_data*, operf_record*)':
>> operf_utils.cpp:1412:6: error: 'rmb' was not declared in this scope
>>   rmb();
>>
>> Architecture not supported it seems.
>>
>>>          arm |                     php-5.6.21 | NOK | http://autobuild.buildroot.net/results/c67d00b19da8bcb2e6070d3f048931b36d5ad467/
>>
>> Forgets to link with pthread. Gustavo?
>>
>>>       x86_64 |                     pppd-2.4.7 | NOK | http://autobuild.buildroot.net/results/32bab29b2ed19ebf18d34f7dc3324161c89dbda9/
>>
>> Musl build issue. Anyone to look into this?
>>
>>>       mipsel |                   psmisc-22.21 | NOK | http://autobuild.buildroot.net/results/382094c3dedfdb3e2489851d8fb1255e440ce39c/
>>
>> SSP MIPS toolchain issue...
>>
>>>         sh4a |             qt5serialbus-5.6.0 | NOK | http://autobuild.buildroot.net/results/6176a24799b5d80312be954089efbbb498fa1571/
>>
>> Seems like this package needs a dependency on some recent enough kernel
>> headers. Yegor, since you added this package, can you have a look?
>
> CANFD_MTU stuff seems to be added in kernel 3.5. How can I express
> this in Buildroot dependency terms?

Something like this I suppose :-)

depends on BR2_TOOLCHAIN_HEADERS_AT_LEAST_3_5

Will send a patch.

>>
>>>          arc |                  rsyslog-8.9.0 | NOK | http://autobuild.buildroot.net/results/08b5fa07c85e38a771f7d52e2e09fb62901e156c/
>>>       xtensa |                  rsyslog-8.9.0 | NOK | http://autobuild.buildroot.net/results/abb2e06d0b19922627d9a76c75858b86eca1cc2d/
>>
>> configure: error: in `/home/buildroot/autobuild/run/instance-0/output/build/rsyslog-8.9.0':
>> configure: error: PgSQL library is missing
>>
>> Anyone to look into this?
>>
>>>      sparc64 |              stress-ng-0.04.16 | NOK | http://autobuild.buildroot.net/results/99728d80caf490ad2cba78b9c50ec5385c5e6601/
>>>      sparc64 |              stress-ng-0.04.16 | NOK | http://autobuild.buildroot.net/results/9134c246e451061b2ad42d0508d073f18f19961a/
>>>      sparc64 |              stress-ng-0.04.16 | NOK | http://autobuild.buildroot.net/results/c22585c193c48cb1d2ad3bb7ecde73620562d7c8/
>>
>> This should be fixed by https://git.buildroot.net/buildroot/commit/?id=08fe324b8b0e12a2ada22db8ee268a99cdde4322.
>>
>>>        sparc |               strongswan-5.4.0 | NOK | http://autobuild.buildroot.net/results/5859b8e0a1775ecd6afa13132549966080739837/
>>
>> ../../src/libstrongswan/.libs/libstrongswan.so: undefined reference to `__atomic_fetch_sub_4'
>> ../../src/libstrongswan/.libs/libstrongswan.so: undefined reference to `__atomic_compare_exchange_4'
>> ../../src/libstrongswan/.libs/libstrongswan.so: undefined reference to `__atomic_fetch_add_4'
>>
>> Gustavo?
>>
>>>          arc |                    taskd-1.1.0 | NOK | http://autobuild.buildroot.net/results/67f574af0cc4348ffe2bce026c44766e49c29124/
>>
>> Forgets to link with pthread. Ben, since you added this package, can
>> you have a look?
>>
>>>     mips64el |  toolchain-external-legal-info | NOK | http://autobuild.buildroot.net/results/b047bfa84af0e88f4efa5e1ed7ea5639d120b947/
>>
>> Network issue, Peter has already added the file to sources.b.o to avoid
>> the problem in the future.
>>
>>>          arm |                     ustr-1.0.4 | NOK | http://autobuild.buildroot.net/results/cebfd7e29a6521ae7ba4c229586a3ef662aa9de1/
>>
>> libustr-debug-1.0.so.1': File exists
>>
>> Gah, smells like parallel installation issue. Clayton, since you added
>> this package, can you have a look?
>>
>>>     mips64el |                valgrind-3.11.0 | NOK | http://autobuild.buildroot.net/results/a1168399adb9fd549d610e4d18aa024d9c8bcbff/
>>>     mips64el |                valgrind-3.11.0 | NOK | http://autobuild.buildroot.net/results/61eb30d455cf767f4317b0898c2219b2ab695fc0/
>>
>> Vicente?
>>
>>>          arm |                      vlc-2.2.3 | NOK | http://autobuild.buildroot.net/results/1a7893c13f6144e0ecd9f37f2f047c596ccedd35/
>>
>> ld: cannot find -lGL
>>
>> Bernd?
>>
>>>         bfin |                 wayland-1.10.0 | NOK | http://autobuild.buildroot.net/results/0a5d32f0fa28ecdd540e100951b116708feec6b5/
>>>         bfin |                 wayland-1.10.0 | NOK | http://autobuild.buildroot.net/results/75286a0319094a82934eebeeda6ac4f0b0f36743/
>>
>> I've looked into this, it's the infamous _ prefix issue on Blackfin. It
>> should be relatively easy to fix, I'll send a patch.
>>
>> Thomas
>> --
>> Thomas Petazzoni, CTO, Free Electrons
>> Embedded Linux, Kernel and Android engineering
>> http://free-electrons.com

[Buildroot] Analysis of build results for 2016-05-16

[gwenhael.goavec]

Hi,

On Tue, 17 May 2016 17:20:40 +0200
Thomas Petazzoni <thomas.petazzoni@free-electrons.com> wrote:

> Hello,
> 
> On Tue, 17 May 2016 17:13:50 +0200, gwenhael.goavec wrote:
> 
> > > >       x86_64 |               gnuradio-3.7.9.1 | NOK |
> > > > http://autobuild.buildroot.net/results/714019031fb8c8f70c1838aa18e4ec1e4abf165e/      
> > > 
> > > internal compiler error: Segmentation fault
> > > 
> > > Gwenhael, can you look into this problem?
> > >     
> > I'm looking into this problem :
> > - Fails with a binary toolchain (gcc-4.7.2);
> > - Works with a toolchain generated by buildroot (same options but
> >   4.8.5 and 4.9.3).
> > I need to see if it's due to the gcc version or something else.  
> 
> Potentially seems like a bug in gcc 4.7.x, which has been fixed in gcc
> 4.8.x, no? We still support gcc 4.7.x in the internal toolchain
> backend, if you enable "deprecated options". This way you can check if
> a Buildroot toolchain with gcc 4.7 exhibits the same problem.
> 
> Note that this kind of issue is often sensitive to optimization flags:
> i.e it might fail at -O3 but succeed at -Os.
> 
> If we conclude that 4.7 is broken, and >= 4.8 is working, then adding
> a dependency on BR2_TOOLCHAIN_GCC_AT_LEAST_4_8 is good enough for me.
> 
With a toolchain based on gcc-4.7.4 the compilation is working... Maybe
some options enabled/disabled with the binary toolchain.

I will search...
Gwen

[Buildroot] Analysis of build results for 2016-05-16

[Samuel Martin]

Hi Thomas, all

On Tue, May 17, 2016 at 3:36 PM, Thomas Petazzoni
<thomas.petazzoni@free-electrons.com> wrote:
> Hello,
>
> Rodrigo, Geoff, Gustavo, Vicente, Gwenhael, Samuel, Yegor, Ben,
> Clayton, Bernd, please see below, there are questions/issues for you.
>
> Note to others: there are also plenty of build failures below with
> nobody assigned. Feel free to pick some of them and investigate. Thanks!
[...]
>>          sh4 |                  opencv3-3.1.0 | NOK | http://autobuild.buildroot.net/results/a5a32e8b18583f0c40da22e735ef5e2ed1e415a6/
>
> Toolchain issue:
>
> /tmp/ccF0zot6.s: Assembler messages:
> /tmp/ccF0zot6.s:19767: Error: pcrel too far
>
> Samuel, can you retry with the latest gcc/binutils version on SH4 and
> see if it fixes the problem?

I've not finished yet to check all configurations, but from the
preliminary results I got so far, it seems that opencv3 for sh4 builds
correctly with gcc-5.3 and gcc-6.1.

This may give some hints for the next sh4 toolchain configuration for the a.b.o.

I let you know when I get the final results.

Regards,

-- 
Samuel

[Buildroot] Analysis of build results for 2016-05-16

[Arnout Vandecappelle]

On 05/18/16 09:09, gwenhael.goavec wrote:
> Hi,
>
> On Tue, 17 May 2016 17:20:40 +0200
> Thomas Petazzoni <thomas.petazzoni@free-electrons.com> wrote:
>
>> Hello,
>>
>> On Tue, 17 May 2016 17:13:50 +0200, gwenhael.goavec wrote:
>>
>>>>>       x86_64 |               gnuradio-3.7.9.1 | NOK |
>>>>> http://autobuild.buildroot.net/results/714019031fb8c8f70c1838aa18e4ec1e4abf165e/
>>>>
>>>> internal compiler error: Segmentation fault
>>>>
>>>> Gwenhael, can you look into this problem?
>>>>
>>> I'm looking into this problem :
>>> - Fails with a binary toolchain (gcc-4.7.2);
>>> - Works with a toolchain generated by buildroot (same options but
>>>   4.8.5 and 4.9.3).
>>> I need to see if it's due to the gcc version or something else.
>>
>> Potentially seems like a bug in gcc 4.7.x, which has been fixed in gcc
>> 4.8.x, no? We still support gcc 4.7.x in the internal toolchain
>> backend, if you enable "deprecated options". This way you can check if
>> a Buildroot toolchain with gcc 4.7 exhibits the same problem.
>>
>> Note that this kind of issue is often sensitive to optimization flags:
>> i.e it might fail at -O3 but succeed at -Os.
>>
>> If we conclude that 4.7 is broken, and >= 4.8 is working, then adding
>> a dependency on BR2_TOOLCHAIN_GCC_AT_LEAST_4_8 is good enough for me.
>>
> With a toolchain based on gcc-4.7.4 the compilation is working... Maybe
> some options enabled/disabled with the binary toolchain.


  In my opinion, ICE issues are not really our problem. Since the user can 
configure any kind of toolchain combination, I don't think it makes a whole lot 
of sense even to add dependencies in the Config.in to avoid broken toolchains 
(like we used to have for BR2_TOOLCHAIN_EXTERNAL_CODESOURCERY_NIOSII).

  So I'd just add an autobuilder exception.

  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:  7493 020B C7E3 8618 8DEC 222C 82EB F404 F9AC 0DDF

[Buildroot] Analysis of build results for 2016-05-16

[Thomas Petazzoni]

Hello,

On Wed, 18 May 2016 21:40:38 +0200, Arnout Vandecappelle wrote:

>   In my opinion, ICE issues are not really our problem. Since the user can 
> configure any kind of toolchain combination, I don't think it makes a whole lot 
> of sense even to add dependencies in the Config.in to avoid broken toolchains 
> (like we used to have for BR2_TOOLCHAIN_EXTERNAL_CODESOURCERY_NIOSII).
> 
>   So I'd just add an autobuilder exception.

I'm hesitant to do this, because it means that normal users can also
fall into the problem.

On the other hand, it's annoying to encode all those exceptions in
Buildroot itself.

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

[Buildroot] Analysis of build results for 2016-05-16

[Arnout Vandecappelle]

On 05/18/16 23:54, Thomas Petazzoni wrote:
> Hello,
>
> On Wed, 18 May 2016 21:40:38 +0200, Arnout Vandecappelle wrote:
>
>>   In my opinion, ICE issues are not really our problem. Since the user can
>> configure any kind of toolchain combination, I don't think it makes a whole lot
>> of sense even to add dependencies in the Config.in to avoid broken toolchains
>> (like we used to have for BR2_TOOLCHAIN_EXTERNAL_CODESOURCERY_NIOSII).
>>
>>   So I'd just add an autobuilder exception.
>
> I'm hesitant to do this, because it means that normal users can also
> fall into the problem.

  My point is: excluding one specific (external) toolchain doesn't really stop 
normal users from encountering the problem, because they are just as likely to 
use a completely different (buildroot-generated, crosstool-NG, or 
vendor-provided) toolchain.

  My personal experience with customer projects is that maybe we start with a 
predefined external toolchain, but in the end we either build our own, or we 
bump buildroot but want to keep the existing toolchain which already has been 
replaced. So we always end up with a custom external toolchain.


> On the other hand, it's annoying to encode all those exceptions in
> Buildroot itself.

  My point is not so much that it's annoying, my point is that it's giving a 
false sense of security. We're adding exceptions for things we happen to 
encounter, but normal users are just as likely to still run into them.


  Note that I'm really talking about toolchain bugs here (i.e. ICE and the 
like), because these are particularly difficult to identify reliably. The 
situation is entirely different for things that can be handled by 
BR2_TOOLCHAIN_GCC_AT_LEAST_ and the like.


  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:  7493 020B C7E3 8618 8DEC 222C 82EB F404 F9AC 0DDF

[Buildroot] Analysis of build results for 2016-05-16

[Gustavo Zacarias]

On 17/05/16 10:36, Thomas Petazzoni wrote:

>>           arm |                 collectd-5.5.1 | NOK | http://autobuild.buildroot.net/results/80a75aaebd55fca1220048d1d4dfb434ce4d9804/
>>        x86_64 |                 collectd-5.5.1 | NOK | http://autobuild.buildroot.net/results/ae1e3a33a1c2b7433ca7a04659467e2017756b00/
>
> Gustavo, any solution in sight for this collectd problem?

Hi.
This is the same problem that was fixed for connman in 0dece985, fixed 
in linux via 4a91cb61.
The problem is that the connman fix breaks the proper fix in the kernel, 
and now that 4.5.5 and 4.6 are out i think the best approach is to bump 
the 4.5.x series for the release and drop the connman patch.
This has the side-effect of making connman/collectd with iptables and 
possibly some other netfilter-wielding code to be effectively broken for 
4.5.0 -> 4.5.4 headers though, but as long as we don't ship these in a 
release i think we should genrally be fine.
Regarding external toolchains time will tell, but any package patches we 
make about this will likely fall on silent ears upstream.
http://patchwork.ozlabs.org/patch/624067/ and 
http://patchwork.ozlabs.org/patch/624068/ address this.
Regards.

[Buildroot] Analysis of build results for 2016-05-16

[Gustavo Zacarias]

On 17/05/16 10:36, Thomas Petazzoni wrote:

>>       powerpc |        gadgetfs-test-undefined | NOK | http://autobuild.buildroot.net/results/163fbc2ccd88282be2f5d5354b125429bfb4ff4b/
>
> Gustavo, another _restgpr_29_x issue, and this is with a Buildroot
> toolchain, so we can't blame an old CT-NG or Sourcery toolchain here :)

Hi.
No no no, the other was a _restgpr_31_x issue ;)
The recurring problem is that gcc targetting powerpc has bugs when using 
-Os.
We could override per-package given this condition, or simply switch to 
-O2 as default for powerpc. I'm more inclined for the second option.
I believe there are patches floating around the internets to fix this 
but they were never committed upstream.
Regards.

[Buildroot] Analysis of build results for 2016-05-16

[Arnout Vandecappelle]

On 05/19/16 20:52, Gustavo Zacarias wrote:
> On 17/05/16 10:36, Thomas Petazzoni wrote:
> 
>>>       powerpc |        gadgetfs-test-undefined | NOK |
>>> http://autobuild.buildroot.net/results/163fbc2ccd88282be2f5d5354b125429bfb4ff4b/
>>
>> Gustavo, another _restgpr_29_x issue, and this is with a Buildroot
>> toolchain, so we can't blame an old CT-NG or Sourcery toolchain here :)
> 
> Hi.
> No no no, the other was a _restgpr_31_x issue ;)
> The recurring problem is that gcc targetting powerpc has bugs when using -Os.
> We could override per-package given this condition, or simply switch to -O2 as
> default for powerpc. I'm more inclined for the second option.

 Even better: disallow -Os completely for powerpc. If it's fixed in recent gcc
we can add a gcc version dependency as well.

 Regards,
 Arnout

> I believe there are patches floating around the internets to fix this but they
> were never committed upstream.
> Regards.
> _______________________________________________
> buildroot mailing list
> buildroot at busybox.net
> http://lists.busybox.net/mailman/listinfo/buildroot
> 


-- 
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:  7493 020B C7E3 8618 8DEC 222C 82EB F404 F9AC 0DDF

[Buildroot] Analysis of build results for 2016-05-16

[Thomas Petazzoni]

Hello,

On Thu, 19 May 2016 23:31:32 +0200, Arnout Vandecappelle wrote:

> > No no no, the other was a _restgpr_31_x issue ;)
> > The recurring problem is that gcc targetting powerpc has bugs when using -Os.
> > We could override per-package given this condition, or simply switch to -O2 as
> > default for powerpc. I'm more inclined for the second option.  
> 
>  Even better: disallow -Os completely for powerpc. If it's fixed in recent gcc
> we can add a gcc version dependency as well.

+1 on this suggestion.

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

[Buildroot] Analysis of build results for 2016-05-16

[Clayton Shotwell]

Thomas,

On Tue, May 17, 2016 at 8:36 AM, Thomas Petazzoni
<thomas.petazzoni@free-electrons.com> wrote:
>
> Hello,
>
> Rodrigo, Geoff, Gustavo, Vicente, Gwenhael, Samuel, Yegor, Ben,
> Clayton, Bernd, please see below, there are questions/issues for you.
>
> Note to others: there are also plenty of build failures below with
> nobody assigned. Feel free to pick some of them and investigate. Thanks!
>
>
> >          arm |                     ustr-1.0.4 | NOK | http://autobuild.buildroot.net/results/cebfd7e29a6521ae7ba4c229586a3ef662aa9de1/
>
> libustr-debug-1.0.so.1': File exists
>
> Gah, smells like parallel installation issue. Clayton, since you added
> this package, can you have a look?

I took a look at the errors but I can't seem to reproduce it on my
local setup. I can submit a patch that tries to fix it but I'm not
sure what the root cause of the error is. The commands make is running
should not allow for the error to happen (removes the link before it
is created). I'll submit the patch for review and then we can go from
there.

Thanks,
Clayton

Clayton Shotwell
Senior Software Engineer, Rockwell Collins
clayton.shotwell at rockwellcollins.com

[Buildroot] Analysis of build results for 2016-05-16

[Ben Boeckel]

Sorry for the delay; had been on vacation and am catching up on things.

On Tue, 17 May, 2016 at 13:36:11 GMT, Thomas Petazzoni wrote:
>>          arc |                    taskd-1.1.0 | NOK | http://autobuild.buildroot.net/results/67f574af0cc4348ffe2bce026c44766e49c29124/
>
> Forgets to link with pthread. Ben, since you added this package, can
> you have a look?

So I took the config file pointed there and built taskd with it using
git master as of yesterday. It looks like a problem with how taskd is
linking to gnutls:

    gnutls_compress.c:(.text+0x480): undefined reference to `inflate'

and lots of other missing symbols from libgnutls.a (looks to be similar
for libuuid.a).

Since everything is static, taskd needs to link to gnutls' transitive
dependencies, but CMake's FindGnuTLS module only finds the library
itself, so the final executable ends up with missing symbols.

I'll look into improving CMake's FindGnuTLS for this. The code for
linking to libuuid in taskd seems to have similar problems.

--Ben

[Buildroot] Analysis of build results for 2016-05-16

[Ben Boeckel]

On Fri, 17 Jun, 2016 at 01:21:23 GMT, Ben Boeckel wrote:
> Since everything is static, taskd needs to link to gnutls' transitive
> dependencies, but CMake's FindGnuTLS module only finds the library
> itself, so the final executable ends up with missing symbols.

Fixed it with a hacky patch. Proper fixes will need to wait for a
revamped FindGnuTLS module (earliest would be CMake 3.7.0; features
required will be in CMake 3.6.0).

--Ben