* [Buildroot] [autobuild.buildroot.net] Daily results for 2019-09-17
@ 2019-09-18 6:16 Thomas Petazzoni
2019-09-18 20:56 ` Peter Seiderer
` (2 more replies)
0 siblings, 3 replies; 7+ messages in thread
From: Thomas Petazzoni @ 2019-09-18 6:16 UTC (permalink / raw)
To: buildroot
Hello,
Autobuild statistics for 2019-09-17
===============================
branch | OK | NOK | TIM | TOT |
2019.02.x | 19 | 0 | 0 | 19 |
2019.05.x | 15 | 5 | 0 | 20 |
2019.08.x | 16 | 4 | 0 | 20 |
master | 116 | 38 | 0 | 154 |
Results for branch 'master'
===========================
Classification of failures by reason
------------------------------------
wpa_supplicant-2.9 | 6
exempi-2.5.1 | 3
flac-1.3.3 | 2
host-libglib2-2.62.0 | 2
host-mfgtools-0.02 | 2
icu-64-2 | 2
luksmeta-9 | 2
strace-5.2 | 2
ytree-1.99pl1 | 2
apr-1.7.0 | 1
cmocka-1.1.5 | 1
ffmpeg-4.2 | 1
fio-fio-3.9 | 1
gst1-plugins-base-1.16.0 | 1
haproxy-2.0.5 | 1
iputils-20190709 | 1
libnice-0.1.14 | 1
mimic-1.1.0 | 1
mpv-0.29.1 | 1
python-sip-4.19.13 | 1
spandsp-20180108 | 1
unknown | 1
upmpdcli-1.4.2 | 1
vdr-plugin-vnsiserver-1.8.0 | 1
Detail of failures
------------------
arch | reason | OK? | url | orph?
-------------+--------------------------------+-----+---------------------------------------------------------------------------------+-------
arm | apr-1.7.0 | NOK | http://autobuild.buildroot.net/results/4da73982f160305ac2d5d2b4f615fa98eec5a7a3 |
riscv64 | cmocka-1.1.5 | NOK | http://autobuild.buildroot.net/results/5e526515157df607737331cd3b8e2ec8dc2fe425 |
powerpc | exempi-2.5.1 | NOK | http://autobuild.buildroot.net/results/8860cd17bbdfd14117369486fc37a2b881b5e8aa |
powerpc | exempi-2.5.1 | NOK | http://autobuild.buildroot.net/results/7ee587945f4d20c7d8a56846366e014e3e210e3e |
powerpc | exempi-2.5.1 | NOK | http://autobuild.buildroot.net/results/eab0a04b6d9cb384489f7f1c20ce4b25b142e0f3 |
x86_64 | ffmpeg-4.2 | NOK | http://autobuild.buildroot.net/results/061fd3cd6e29c86f49ebfe47b4abb7497ee17fa5 |
arm | fio-fio-3.9 | NOK | http://autobuild.buildroot.net/results/b0a5a9f61cada527e8d480d77fc810489a92fa2b |
powerpc | flac-1.3.3 | NOK | http://autobuild.buildroot.net/results/eb3a59a78dbb0c83cff78aac2384edea890af830 |
powerpc | flac-1.3.3 | NOK | http://autobuild.buildroot.net/results/f1360435f0a198ac35bcfd965650da1e204576bf |
riscv32 | gst1-plugins-base-1.16.0 | NOK | http://autobuild.buildroot.net/results/2c5bc3fa3784e7916156c52b2155502b349acaa3 |
microblazeel | haproxy-2.0.5 | NOK | http://autobuild.buildroot.net/results/4426a5ce9e1b1772f8e70cdf802d19a1d6751dd0 |
arm | host-libglib2-2.62.0 | NOK | http://autobuild.buildroot.net/results/2b730a6bc0096b15feaea076df094c19362e1ac8 |
aarch64 | host-libglib2-2.62.0 | NOK | http://autobuild.buildroot.net/results/702880f74b5517e3c73c80379748eae5893e2c05 |
arm | host-mfgtools-0.02 | NOK | http://autobuild.buildroot.net/results/6ae9a69b717bf6ed0898779f8b413e6b73870dd5 |
arm | host-mfgtools-0.02 | NOK | http://autobuild.buildroot.net/results/47a10f29f9085c1bdd4e31397357aa36b07f87b9 |
xtensa | icu-64-2 | NOK | http://autobuild.buildroot.net/results/b8841f5642ce9d12928f133c92a5839c6e45fe09 | ORPH
xtensa | icu-64-2 | NOK | http://autobuild.buildroot.net/results/8db12ea86d769cba5c40e6b1e76096f68bfae791 | ORPH
x86_64 | iputils-20190709 | NOK | http://autobuild.buildroot.net/results/c00deb41e465a049dc8dcc9f9ece3ffbf4292c4c |
x86_64 | libnice-0.1.14 | NOK | http://autobuild.buildroot.net/results/4e7e441fecceaec06b452058434be459068d5902 |
i686 | luksmeta-9 | NOK | http://autobuild.buildroot.net/results/c009d5e137318bf6cad7380e445cd8def99e19ca |
nios2 | luksmeta-9 | NOK | http://autobuild.buildroot.net/results/003311c3dd7d0076ef7298cea8bed1c1823dae7e |
powerpc | mimic-1.1.0 | NOK | http://autobuild.buildroot.net/results/474b99a6ae402bdc750e3336cee9e01b2e76a809 |
riscv64 | mpv-0.29.1 | NOK | http://autobuild.buildroot.net/results/2aa60497e38edde022461c23671a7aa771af7554 |
riscv32 | python-sip-4.19.13 | NOK | http://autobuild.buildroot.net/results/2c5d763e8fa632912b68e042d79ea2b168ce4c11 |
i686 | spandsp-20180108 | NOK | http://autobuild.buildroot.net/results/8427e399c27a15858bd144df72ec850cc4185611 |
m68k | strace-5.2 | NOK | http://autobuild.buildroot.net/results/fb2715ba38c6f9a57033c20acbbd739a13af5c2d |
m68k | strace-5.2 | NOK | http://autobuild.buildroot.net/results/1a9af7e54715533c85fc3010fc3cf8c9ff0fe093 |
sparc | unknown | NOK | http://autobuild.buildroot.net/results/b26b9e6bb1bb26632cf8681539618de6ea04b8a0 |
arc | upmpdcli-1.4.2 | NOK | http://autobuild.buildroot.net/results/604753a82da6c90dcceaeea1fbdc9f13ff4850ff |
arm | vdr-plugin-vnsiserver-1.8.0 | NOK | http://autobuild.buildroot.net/results/2ecf4fa96149ddb47f15f22e05decf563b04fefe |
sparc | wpa_supplicant-2.9 | NOK | http://autobuild.buildroot.net/results/fe56301266a5198550264ccdee7988fb4c8d4c1c |
arm | wpa_supplicant-2.9 | NOK | http://autobuild.buildroot.net/results/0d39920cfa45fce092cac9b24c4a9033a53c433a |
powerpc64le | wpa_supplicant-2.9 | NOK | http://autobuild.buildroot.net/results/76ca8cd3feca37fc5eaa02fd29bb2964308958e8 |
powerpc64 | wpa_supplicant-2.9 | NOK | http://autobuild.buildroot.net/results/b5f059176a75207c5912b38856d915f118fcd485 |
arm | wpa_supplicant-2.9 | NOK | http://autobuild.buildroot.net/results/0d3f4f2a6bd19691894def7fe642040ed30c1c07 |
sh4 | wpa_supplicant-2.9 | NOK | http://autobuild.buildroot.net/results/3c7d494221c71092841da2b3a6c1376e484a61c2 |
m68k | ytree-1.99pl1 | NOK | http://autobuild.buildroot.net/results/14ee31c5535acae61a2ceb3e2dcd31e4280d97d7 |
arm | ytree-1.99pl1 | NOK | http://autobuild.buildroot.net/results/f394983f7ca8a34018c3061d53bf63104a5e10b4 |
Results for branch '2019.05.x'
==============================
Classification of failures by reason
------------------------------------
host-qemu-3.1.1 | 2
host-mfgtools-v0.02 | 1
libgpg-error-1.35 | 1
xen-4.11.2 | 1
Detail of failures
------------------
arch | reason | OK? | url | orph?
-------------+--------------------------------+-----+---------------------------------------------------------------------------------+-------
arm | host-mfgtools-v0.02 | NOK | http://autobuild.buildroot.net/results/df9b1e988218265770ca14eb9da2a95a7724b42c |
m68k | host-qemu-3.1.1 | NOK | http://autobuild.buildroot.net/results/15a75c794f98884a40d1904b90d9647d778b8f11 | ORPH
powerpc | host-qemu-3.1.1 | NOK | http://autobuild.buildroot.net/results/8b1bcda19ab16bda40d757300350588d3211b19b | ORPH
riscv64 | libgpg-error-1.35 | NOK | http://autobuild.buildroot.net/results/b8065cf04b52caac4c6369e9b473f8e264d9baa7 | ORPH
aarch64 | xen-4.11.2 | NOK | http://autobuild.buildroot.net/results/2168bf338e8dcb46d39322b03c7ee1c29ccd8129 |
Results for branch '2019.08.x'
==============================
Classification of failures by reason
------------------------------------
libgpg-error-1.36 | 2
piglit-ef761094be04d14e258a... | 1
vdr-plugin-vnsiserver-1.8.0 | 1
Detail of failures
------------------
arch | reason | OK? | url | orph?
-------------+--------------------------------+-----+---------------------------------------------------------------------------------+-------
mipsel | libgpg-error-1.36 | NOK | http://autobuild.buildroot.net/results/0f8af2135b96283a6310402c7cf597f9854c90c2 | ORPH
arm | libgpg-error-1.36 | NOK | http://autobuild.buildroot.net/results/c61eaa5da9496e22120acd5b5d51f4db8dfdea5d | ORPH
mips | piglit-ef761094be04d14e258a... | NOK | http://autobuild.buildroot.net/results/50146e20c58b9b88f286695d1182b3ac2d9d9f27 |
powerpc64le | vdr-plugin-vnsiserver-1.8.0 | NOK | http://autobuild.buildroot.net/results/727108af1f73d91b2a956d959029f984df9936d5 |
--
http://autobuild.buildroot.net
^ permalink raw reply [flat|nested] 7+ messages in thread* [Buildroot] [autobuild.buildroot.net] Daily results for 2019-09-17 2019-09-18 6:16 [Buildroot] [autobuild.buildroot.net] Daily results for 2019-09-17 Thomas Petazzoni @ 2019-09-18 20:56 ` Peter Seiderer 2019-09-21 11:26 ` Arnout Vandecappelle 2019-09-19 18:35 ` Peter Seiderer 2019-09-19 19:35 ` Peter Seiderer 2 siblings, 1 reply; 7+ messages in thread From: Peter Seiderer @ 2019-09-18 20:56 UTC (permalink / raw) To: buildroot Hello, On Wed, 18 Sep 2019 06:16:20 -0000, Thomas Petazzoni <thomas.petazzoni@bootlin.com> wrote: [...] > > Results for branch 'master' > =========================== [...] > x86_64 | ffmpeg-4.2 | NOK | http://autobuild.buildroot.net/results/061fd3cd6e29c86f49ebfe47b4abb7497ee17fa5 | The ffmpeg configure step fails with: ERROR: gnutls not found using pkg-config From build/ffmpeg-4.2/ffbuild/config.log: .../host/bin/../x86_64-buildroot-linux-gnu/sysroot/usr/lib/libgnutls.so: undefined reference to `__get_cpuid_count' From build/gnutls-3.6.9/lib/accelerated/x86/x86-common.c: 41 #ifdef HAVE_CPUID_H 42 # include <cpuid.h> 43 #else 44 # define __get_cpuid(...) 0 45 # define __get_cpuid_count(...) 0 46 #endif The given toolchain cpuid.h file is missing the get_cpuid_count definition (for an cpuid.h file with it see e.g. [1])... The gnutls search for cpuid.h can be disabled by the '--disable-hardware-acceleration' configure option, so the build could be fixed by the following patch: --- a/package/gnutls/gnutls.mk +++ b/package/gnutls/gnutls.mk @@ -27,7 +27,8 @@ GNUTLS_CONF_OPTS = \ --with-librt-prefix=$(STAGING_DIR) \ --without-tpm \ $(if $(BR2_PACKAGE_GNUTLS_OPENSSL),--enable,--disable)-openssl-compatibility \ - $(if $(BR2_PACKAGE_GNUTLS_TOOLS),--enable-tools,--disable-tools) + $(if $(BR2_PACKAGE_GNUTLS_TOOLS),--enable-tools,--disable-tools) \ + --disable-hardware-acceleration GNUTLS_CONF_ENV = gl_cv_socket_ipv6=yes \ ac_cv_header_wchar_h=$(if $(BR2_USE_WCHAR),yes,no) \ gt_cv_c_wchar_t=$(if $(BR2_USE_WCHAR),yes,no) \ Unconditionally disabling hardware acceleration seems to be overkill, but did not spot the right condition (yet)... Or alternatively patch configure to check not only for cpuid.h header, but although check for __get_cpuid/__get_cpuid_count presence... Regards, Peter [1] https://github.com/gcc-mirror/gcc/blob/master/gcc/config/i386/cpuid.h ^ permalink raw reply [flat|nested] 7+ messages in thread
* [Buildroot] [autobuild.buildroot.net] Daily results for 2019-09-17 2019-09-18 20:56 ` Peter Seiderer @ 2019-09-21 11:26 ` Arnout Vandecappelle 2019-09-23 19:02 ` Peter Seiderer 0 siblings, 1 reply; 7+ messages in thread From: Arnout Vandecappelle @ 2019-09-21 11:26 UTC (permalink / raw) To: buildroot On 18/09/2019 22:56, Peter Seiderer wrote: > Hello, > > On Wed, 18 Sep 2019 06:16:20 -0000, Thomas Petazzoni <thomas.petazzoni@bootlin.com> wrote: > > [...] >> >> Results for branch 'master' >> =========================== > [...] > >> x86_64 | ffmpeg-4.2 | NOK | http://autobuild.buildroot.net/results/061fd3cd6e29c86f49ebfe47b4abb7497ee17fa5 | > > The ffmpeg configure step fails with: > > ERROR: gnutls not found using pkg-config > > From build/ffmpeg-4.2/ffbuild/config.log: > > .../host/bin/../x86_64-buildroot-linux-gnu/sysroot/usr/lib/libgnutls.so: undefined reference to `__get_cpuid_count' > > From build/gnutls-3.6.9/lib/accelerated/x86/x86-common.c: > > 41 #ifdef HAVE_CPUID_H > 42 # include <cpuid.h> > 43 #else > 44 # define __get_cpuid(...) 0 > 45 # define __get_cpuid_count(...) 0 > 46 #endif > > The given toolchain cpuid.h file is missing the get_cpuid_count definition (for > an cpuid.h file with it see e.g. [1])... > > The gnutls search for cpuid.h can be disabled by the > '--disable-hardware-acceleration' configure option, so the build could > be fixed by the following patch: > > --- a/package/gnutls/gnutls.mk > +++ b/package/gnutls/gnutls.mk > @@ -27,7 +27,8 @@ GNUTLS_CONF_OPTS = \ > --with-librt-prefix=$(STAGING_DIR) \ > --without-tpm \ > $(if $(BR2_PACKAGE_GNUTLS_OPENSSL),--enable,--disable)-openssl-compatibility \ > - $(if $(BR2_PACKAGE_GNUTLS_TOOLS),--enable-tools,--disable-tools) > + $(if $(BR2_PACKAGE_GNUTLS_TOOLS),--enable-tools,--disable-tools) \ > + --disable-hardware-acceleration > GNUTLS_CONF_ENV = gl_cv_socket_ipv6=yes \ > ac_cv_header_wchar_h=$(if $(BR2_USE_WCHAR),yes,no) \ > gt_cv_c_wchar_t=$(if $(BR2_USE_WCHAR),yes,no) \ > > > Unconditionally disabling hardware acceleration seems to be overkill, but > did not spot the right condition (yet)... > > Or alternatively patch configure to check not only for cpuid.h header, > but although check for __get_cpuid/__get_cpuid_count presence... Fixed by http://patchwork.ozlabs.org/patch/1157883/ I guess? Regards, Arnout > > Regards, > Peter > > [1] https://github.com/gcc-mirror/gcc/blob/master/gcc/config/i386/cpuid.h > _______________________________________________ > buildroot mailing list > buildroot at busybox.net > http://lists.busybox.net/mailman/listinfo/buildroot > ^ permalink raw reply [flat|nested] 7+ messages in thread
* [Buildroot] [autobuild.buildroot.net] Daily results for 2019-09-17 2019-09-21 11:26 ` Arnout Vandecappelle @ 2019-09-23 19:02 ` Peter Seiderer 0 siblings, 0 replies; 7+ messages in thread From: Peter Seiderer @ 2019-09-23 19:02 UTC (permalink / raw) To: buildroot Hello Arnout, On Sat, 21 Sep 2019 13:26:00 +0200, Arnout Vandecappelle <arnout@mind.be> wrote: > On 18/09/2019 22:56, Peter Seiderer wrote: > > Hello, > > > > On Wed, 18 Sep 2019 06:16:20 -0000, Thomas Petazzoni <thomas.petazzoni@bootlin.com> wrote: > > > > [...] > >> > >> Results for branch 'master' > >> =========================== > > [...] > > > >> x86_64 | ffmpeg-4.2 | NOK | http://autobuild.buildroot.net/results/061fd3cd6e29c86f49ebfe47b4abb7497ee17fa5 | > > > > The ffmpeg configure step fails with: > > > > ERROR: gnutls not found using pkg-config > > > > From build/ffmpeg-4.2/ffbuild/config.log: > > > > .../host/bin/../x86_64-buildroot-linux-gnu/sysroot/usr/lib/libgnutls.so: undefined reference to `__get_cpuid_count' > > > > From build/gnutls-3.6.9/lib/accelerated/x86/x86-common.c: > > > > 41 #ifdef HAVE_CPUID_H > > 42 # include <cpuid.h> > > 43 #else > > 44 # define __get_cpuid(...) 0 > > 45 # define __get_cpuid_count(...) 0 > > 46 #endif > > > > The given toolchain cpuid.h file is missing the get_cpuid_count definition (for > > an cpuid.h file with it see e.g. [1])... > > > > The gnutls search for cpuid.h can be disabled by the > > '--disable-hardware-acceleration' configure option, so the build could > > be fixed by the following patch: > > > > --- a/package/gnutls/gnutls.mk > > +++ b/package/gnutls/gnutls.mk > > @@ -27,7 +27,8 @@ GNUTLS_CONF_OPTS = \ > > --with-librt-prefix=$(STAGING_DIR) \ > > --without-tpm \ > > $(if $(BR2_PACKAGE_GNUTLS_OPENSSL),--enable,--disable)-openssl-compatibility \ > > - $(if $(BR2_PACKAGE_GNUTLS_TOOLS),--enable-tools,--disable-tools) > > + $(if $(BR2_PACKAGE_GNUTLS_TOOLS),--enable-tools,--disable-tools) \ > > + --disable-hardware-acceleration > > GNUTLS_CONF_ENV = gl_cv_socket_ipv6=yes \ > > ac_cv_header_wchar_h=$(if $(BR2_USE_WCHAR),yes,no) \ > > gt_cv_c_wchar_t=$(if $(BR2_USE_WCHAR),yes,no) \ > > > > > > Unconditionally disabling hardware acceleration seems to be overkill, but > > did not spot the right condition (yet)... > > > > Or alternatively patch configure to check not only for cpuid.h header, > > but although check for __get_cpuid/__get_cpuid_count presence... > > Fixed by http://patchwork.ozlabs.org/patch/1157883/ I guess? Yes, I can confirm the ffmpeg 'gnutls not found using pkg-config' failure is fixed by the above patch (which is already applied on master)... Regards, Peter > > Regards, > Arnout > > > > > Regards, > > Peter > > > > [1] https://github.com/gcc-mirror/gcc/blob/master/gcc/config/i386/cpuid.h > > _______________________________________________ > > buildroot mailing list > > buildroot at busybox.net > > http://lists.busybox.net/mailman/listinfo/buildroot > > ^ permalink raw reply [flat|nested] 7+ messages in thread
* [Buildroot] [autobuild.buildroot.net] Daily results for 2019-09-17 2019-09-18 6:16 [Buildroot] [autobuild.buildroot.net] Daily results for 2019-09-17 Thomas Petazzoni 2019-09-18 20:56 ` Peter Seiderer @ 2019-09-19 18:35 ` Peter Seiderer 2019-09-21 9:33 ` Arnout Vandecappelle 2019-09-19 19:35 ` Peter Seiderer 2 siblings, 1 reply; 7+ messages in thread From: Peter Seiderer @ 2019-09-19 18:35 UTC (permalink / raw) To: buildroot Hello, On Wed, 18 Sep 2019 06:16:20 -0000, Thomas Petazzoni <thomas.petazzoni@bootlin.com> wrote: > Hello, > > Autobuild statistics for 2019-09-17 > =============================== [...] > > arch | reason | OK? | url | orph? > -------------+--------------------------------+-----+---------------------------------------------------------------------------------+------- > arm | apr-1.7.0 | NOK | http://autobuild.buildroot.net/results/4da73982f160305ac2d5d2b4f615fa98eec5a7a3 | From build/apr-1.7.0/config.log: configure:24288: checking whether int64_t and long long use fmt %lld configure:24320: /home/seiderer/Work/Buildroot/build_arm_apr_001/host/bin/arm-linux-gcc -c -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -Os -Werror -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -DLINUX -D_REENTRANT -D_GNU_SOURCE conftest.c >&5 In file included from /home/seiderer/Work/Buildroot/build_arm_apr_001/host/arm-buildroot-linux-uclibcgnueabihf/sysroot/usr/include/sys/types.h:26:0, from conftest.c:147: /home/seiderer/Work/Buildroot/build_arm_apr_001/host/arm-buildroot-linux-uclibcgnueabihf/sysroot/usr/include/features.h:218:5: error: #warning requested reentrant code, but thread support was disabled [-Werror=cpp] # warning requested reentrant code, but thread support was disabled ^~~~~~~ cc1: all warnings being treated as errors configure:24320: $? = 1 configure: failed program was: [...] | | #include <sys/types.h> | #include <stdio.h> | #ifdef HAVE_STDINT_H | #include <stdint.h> | #endif | | int main(int argc, const char *const *argv) { | | int64_t chk1, *ptr1; | long long chk2, *ptr2 = &chk1; | ptr1 = &chk2; | *ptr1 = *ptr2 = 0; | printf("%lld %lld", chk1, chk2); | | return 0; } | configure:24330: result: no The test failure is a result of a toolchain without threads '# BR2_TOOLCHAIN_EXTERNAL_HAS_THREADS is not set' and the given '-D_REENTRANT' compile flag (the same test without '-D_REENTRANT' compiles fine) plus '-Werror'... Disable apr for toolchains without thread support? Regards, Peter ^ permalink raw reply [flat|nested] 7+ messages in thread
* [Buildroot] [autobuild.buildroot.net] Daily results for 2019-09-17 2019-09-19 18:35 ` Peter Seiderer @ 2019-09-21 9:33 ` Arnout Vandecappelle 0 siblings, 0 replies; 7+ messages in thread From: Arnout Vandecappelle @ 2019-09-21 9:33 UTC (permalink / raw) To: buildroot On 19/09/2019 20:35, Peter Seiderer wrote: > Hello, > > On Wed, 18 Sep 2019 06:16:20 -0000, Thomas Petazzoni <thomas.petazzoni@bootlin.com> wrote: > >> Hello, >> >> Autobuild statistics for 2019-09-17 >> =============================== > [...] >> >> arch | reason | OK? | url | orph? >> -------------+--------------------------------+-----+---------------------------------------------------------------------------------+------- >> arm | apr-1.7.0 | NOK | http://autobuild.buildroot.net/results/4da73982f160305ac2d5d2b4f615fa98eec5a7a3 | > > From build/apr-1.7.0/config.log: > > configure:24288: checking whether int64_t and long long use fmt %lld > configure:24320: /home/seiderer/Work/Buildroot/build_arm_apr_001/host/bin/arm-linux-gcc -c -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -Os -Werror -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -DLINUX -D_REENTRANT -D_GNU_SOURCE conftest.c >&5 > In file included from /home/seiderer/Work/Buildroot/build_arm_apr_001/host/arm-buildroot-linux-uclibcgnueabihf/sysroot/usr/include/sys/types.h:26:0, > from conftest.c:147: > /home/seiderer/Work/Buildroot/build_arm_apr_001/host/arm-buildroot-linux-uclibcgnueabihf/sysroot/usr/include/features.h:218:5: error: #warning requested reentrant code, but thread support was disabled [-Werror=cpp] > # warning requested reentrant code, but thread support was disabled > ^~~~~~~ > cc1: all warnings being treated as errors > configure:24320: $? = 1 > configure: failed program was: > [...] > | > | #include <sys/types.h> > | #include <stdio.h> > | #ifdef HAVE_STDINT_H > | #include <stdint.h> > | #endif > | > | int main(int argc, const char *const *argv) { > | > | int64_t chk1, *ptr1; > | long long chk2, *ptr2 = &chk1; > | ptr1 = &chk2; > | *ptr1 = *ptr2 = 0; > | printf("%lld %lld", chk1, chk2); > | > | return 0; } > | > configure:24330: result: no > > The test failure is a result of a toolchain without threads '# BR2_TOOLCHAIN_EXTERNAL_HAS_THREADS is not set' > and the given '-D_REENTRANT' compile flag (the same test without '-D_REENTRANT' compiles fine) plus '-Werror'... So the problem is that apr adds this -D_REENTRANT unconditionally on linux (cfr. build/apr_hints.m4). That suggests two possible solutions: - patch apr_hints.m4 to not add -D_REENTRANT, and add it when we have threads. - bypass the hints entirely by setting apr_preload_done=yes in the environment, and manually passing -DLINUX -D_GNU_SOURCE and a conditional -DREENTRANT in CFLAGS. Regards, Arnout > > Disable apr for toolchains without thread support? > > Regards, > Peter > _______________________________________________ > buildroot mailing list > buildroot at busybox.net > http://lists.busybox.net/mailman/listinfo/buildroot > ^ permalink raw reply [flat|nested] 7+ messages in thread
* [Buildroot] [autobuild.buildroot.net] Daily results for 2019-09-17 2019-09-18 6:16 [Buildroot] [autobuild.buildroot.net] Daily results for 2019-09-17 Thomas Petazzoni 2019-09-18 20:56 ` Peter Seiderer 2019-09-19 18:35 ` Peter Seiderer @ 2019-09-19 19:35 ` Peter Seiderer 2 siblings, 0 replies; 7+ messages in thread From: Peter Seiderer @ 2019-09-19 19:35 UTC (permalink / raw) To: buildroot Hello, On Wed, 18 Sep 2019 06:16:20 -0000, Thomas Petazzoni <thomas.petazzoni@bootlin.com> wrote: > Hello, > > Autobuild statistics for 2019-09-17 > =============================== [...] > powerpc | flac-1.3.3 | NOK | http://autobuild.buildroot.net/results/eb3a59a78dbb0c83cff78aac2384edea890af830 | > powerpc | flac-1.3.3 | NOK | http://autobuild.buildroot.net/results/f1360435f0a198ac35bcfd965650da1e204576bf | Fails with: cpu.c:57:10: fatal error: sys/auxv.h: No such file or directory #include <sys/auxv.h> ^~~~~~~~~~~~ compilation terminated. From build/flac-1.3.3/src/libFLAC/cpu.c: 56 #if defined FLAC__CPU_PPC 57 #include <sys/auxv.h> 58 #endif 236 static void 237 ppc_cpu_info (FLAC__CPUInfo *info) 238 { 239 #if defined FLAC__CPU_PPC 240 #ifndef PPC_FEATURE2_ARCH_3_00 241 #define PPC_FEATURE2_ARCH_3_00 0x00800000 242 #endif 243 244 #ifndef PPC_FEATURE2_ARCH_2_07 245 #define PPC_FEATURE2_ARCH_2_07 0x80000000 246 #endif 247 248 if (getauxval(AT_HWCAP2) & PPC_FEATURE2_ARCH_3_00) { 249 info->ppc.arch_3_00 = true; 250 } else if (getauxval(AT_HWCAP2) & PPC_FEATURE2_ARCH_2_07) { 251 info->ppc.arch_2_07 = true; 252 } 253 #else 254 info->ppc.arch_2_07 = false; 255 info->ppc.arch_3_00 = false; 256 #endif 257 } Could be fixed by the following flac configure.ac patch (check for sys/auxv.h before defining FLAC__CPU_PPC): diff --git a/configure.ac b/configure.ac index 0228a12..64cb3f2 100644 --- a/configure.ac +++ b/configure.ac @@ -144,7 +144,7 @@ case "$host_cpu" in powerpc64|powerpc64le) cpu_ppc64=true cpu_ppc=true - AC_DEFINE(FLAC__CPU_PPC) + AC_CHECK_HEADER(sys/auxv.h, AC_DEFINE(FLAC__CPU_PPC)) AH_TEMPLATE(FLAC__CPU_PPC, [define if building for PowerPC]) AC_DEFINE(FLAC__CPU_PPC64) AH_TEMPLATE(FLAC__CPU_PPC64, [define if building for PowerPC64]) @@ -152,7 +152,7 @@ case "$host_cpu" in ;; powerpc|powerpcle) cpu_ppc=true - AC_DEFINE(FLAC__CPU_PPC) + AC_CHECK_HEADER(sys/auxv.h, AC_DEFINE(FLAC__CPU_PPC)) AH_TEMPLATE(FLAC__CPU_PPC, [define if building for PowerPC]) asm_optimisation=$asm_opt ;; Regards, Peter ^ permalink raw reply related [flat|nested] 7+ messages in thread
end of thread, other threads:[~2019-09-23 19:02 UTC | newest] Thread overview: 7+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2019-09-18 6:16 [Buildroot] [autobuild.buildroot.net] Daily results for 2019-09-17 Thomas Petazzoni 2019-09-18 20:56 ` Peter Seiderer 2019-09-21 11:26 ` Arnout Vandecappelle 2019-09-23 19:02 ` Peter Seiderer 2019-09-19 18:35 ` Peter Seiderer 2019-09-21 9:33 ` Arnout Vandecappelle 2019-09-19 19:35 ` Peter Seiderer
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox