Re: More tests failing under highly parallel builds.

[Adhemerval Zanella <adhemerval dot zanella at linaro dot org>]

On 19/07/2016 06:26, Florian Weimer wrote:
> On 07/16/2016 02:27 AM, Carlos O'Donell wrote:
>> During glibc-2.24 release testing for AArch64, ppc64, ppc64le,
>> and x86_64 I would, depending on the system load see:
>>
>> FAIL: localedata/bug-iconv-trans
>> FAIL: localedata/tst-mbswcs6
>> FAIL: localedata/tst-trans
>> FAIL: localedata/tst-wctype
>> FAIL: localedata/tst_mblen
>> FAIL: localedata/tst_mbrlen
>> FAIL: localedata/tst_mbsrtowcs
>> FAIL: localedata/tst_mbstowcs
>> FAIL: localedata/tst_mbtowc
>> FAIL: localedata/tst_swscanf
>> FAIL: localedata/tst_wcrtomb
>> FAIL: localedata/tst_wcsrtombs
>> FAIL: localedata/tst_wcstombs
>> FAIL: localedata/tst_wctob
>> FAIL: localedata/tst_wctomb
>>
>> They fail with SIGSEGV.
> 
> I have never seen these.

I have seen these even in low parallel builds (armhf with -j2) and it
shows the same behaviour).  I think we have a couple of registered bugs
for this with one begin BZ#10596 [1].

[1] https://sourceware.org/bugzilla/show_bug.cgi?id=10596

> 
>> cat glibc/localedata/bug-iconv-trans.test-result
>> FAIL: localedata/bug-iconv-trans
>> original exit status 139
>>
>> Run manually the test passes just fine.
> 
> Can you patch the test skeleton not to suppress core dumps and try again to reproduce?  Maybe we'll get a usable core file this way.
> 
> Thanks,
> Florian