[Buildroot] [autobuild.buildroot.net] Build results for 2015-03-15

Max Filippov jcmvbkbc at gmail.com
Mon Mar 16 22:56:38 UTC 2015

On Tue, Mar 17, 2015 at 1:29 AM, Max Filippov <jcmvbkbc at gmail.com> wrote:
> On Tue, Mar 17, 2015 at 1:17 AM, Thomas Petazzoni
> <thomas.petazzoni at free-electrons.com> wrote:
>> Dear Max Filippov,
>> On Tue, 17 Mar 2015 01:02:28 +0300, Max Filippov wrote:
>>> On Mon, Mar 16, 2015 at 10:30 AM, Thomas Petazzoni
>>> <thomas.petazzoni at free-electrons.com> wrote:
>>> >       xtensa |                   libnl-3.2.25 | NOK | http://autobuild.buildroot.net/results/90c318e3a9ef5e626d0a80e36695d34eda21f97e/
>>> >       xtensa |               memcached-1.4.22 | NOK | http://autobuild.buildroot.net/results/cd8e9d9a5f454c321071c21192958aeee9c39e73/
>>> Caused by disabled UCLIBC_HAS_GNU_GETSUBOPT option in the uclibc config.
>>> Also reproducible with other architectures that use uclibc.
>>> I guess we can just enable it in the config, additional code is really small.
>> On libnl, the error only happened twice, and only on xtensa:
>> http://autobuild.buildroot.org/?reason=libnl-3.2.25. What's weird is
>> that in uClibc-0.9.33.config, which we use on other architectures, we
>> also have UCLIBC_HAS_GNU_GETSUBOPT disabled. But still there is a
>> getsubopt symbol in the generated uClibc .so file.
>> Needs more investigation, it seems.
> Ok, will look at it some more...

It's a bug in uclibc- fixed in commits
b222a04ae768080121458d28bff743b4afade9ba and

Doesn't look like an easy backport, as these commits depend on
major Makefile style change. Probably easier to reimplement.
The patch enabling UCLIBC_HAS_GNU_GETSUBOPT in uclibc
config is still valid after fixing that.

-- Max

More information about the buildroot mailing list