Re: [COMMITTED] 2.22: Bug 18589: Fix strcoll_l incorrect sorting.

[Allan McRae <allan at archlinux dot org>]

On 13/10/15 20:34, Arkadiusz MiÅkiewicz wrote:
> On Monday 12 of October 2015, Florian Weimer wrote:
>> On 10/12/2015 04:23 AM, Allan McRae wrote:
>>> Fun story...
>>>
>>> [2015-10-12 09:47] [ALPM] upgraded glibc (2.22-3 -> 2.22-4)
>>> [2015-10-12 09:47] [ALPM-SCRIPTLET] bash: loadlocale.c:130:
>>> _nl_intern_locale_data: Assertion `cnt < (sizeof
>>> (_nl_value_type_LC_COLLATE) / sizeof (_nl_value_type_LC_COLLATE[0]))'
>>> failed.
>>>
>>> And that was given every time I tried running anything.  It is fun to
>>> watch your system slowly die while trying to figure a way to remove that
>>> update from the repos!
>>>
>>> The only changes between our 2.22-3 and 2.22-4 builds were these patches
>>> you committed.  It seems they introduce incompatibilities to locale
>>> files generated with builds from earlier in the 2.22 release branch?  I
>>> was using a relatively benign locale (en_AU.UTF8).
>>
>> The locale/langinfo.h bits in the reverting commit seem the likely
>> culprit; that would apply to any locale.  It would have been better to
>> drop this from the backport.  Unfortunately, it's too late to fix this
>> now because it would cause similar breakage again for those who suffered
>> through the previous breakage.
> 
> I guess is that my previous problems with 2.21->2.22 (thread "2.22: 
> _nl_intern_locale_data: Assertion `cnt < (sizeof (_nl_value_type_LC_COLLATE) / 
> sizeof (_nl_value_type_LC_COLLATE[0]))' failed." were related to this)
> 
> 
> Do I understand correctly that we have (or will have)
> 
> a) 2.21
> b) 2.22
> c) 2.22+backport
> d) future 2.23
> 
> a->b - breakage
> b->c - breakage
> b->d - breakage
> c->d - no problem
> a->c - no problem

That is my understanding too.   Although c->d is not guaranteed to have
no problem (and there might be changes that already break this...).

Allan