[Mageia-dev] [changelog] [RPM] cauldron core/release libbsd-0.3.0-4.mga3

[Simple w]

2012/6/17 Anssi Hannula <anssi at mageia.org>:
> (@shlomif, sorry for double post, I sent the previous one from an
> unsubscribed address)
>
> 06.06.2012 21:50, shlomif kirjoitti:
>> Name        : libbsd                       Relocations: (not relocatable)
>> Version     : 0.3.0                             Vendor: Mageia.Org
>> Release     : 4.mga3                        Build Date: Wed Jun  6 20:50:30 2012
>> Install Date: (not installed)               Build Host: ecosse.mageia.org
>> Group       : Development/C                 Source RPM: (none)
>> Size        : 105088                           License: BSD and ISC and Copyright only and Public Domain
>> Signature   : (none)
>> Packager    : shlomif <shlomif>
>> URL         : http://libbsd.freedesktop.org/
>> Summary     : Library providing BSD-compatible functions for portability
>> Description :
>> libbsd provides useful functions commonly found on BSD systems, and
>> lacking on others like GNU systems, thus making it easier to port
>> projects with strong BSD origins, without needing to embed the same
>> code over and over again on each project.
>>
>> shlomif <shlomif> 0.3.0-4.mga3:
>> + Revision: 256602
>> - Correct the devel package Group
>> - Correct the group
>> - Bump the release
>> - Fix the man pages globs
>>
>>   + kamil <kamil>
>>     - imported package libbsd
>
> We already have a libbsd.a from glibc. This package will cause software
> being linked with "-lbsd" to be linked against libbsd.so of libbsd
> package instead of libbsd.a of glibc.
>
> Why do we need this?

Its needed for bumblebee.
I remember you saying this when i asked you to review bumblebee
packages and i explained you why was needed.

> We need to either rename or drop one of them (libbsd or libbsd.a from
> glibc), otherwise this will cause bad linking behavior.
>
> I raised this originally in my libbsd.spec review at
> https://www.mageia.org/pipermail/mageia-dev/2012-April/014102.html
>
> I hope the other points in the above review have been addressed (haven't
> checked yet).
>
> --
> Anssi Hannula