2011/2/15 Arjan van de Ven <[email protected]>: > On 2/15/2011 8:57 AM, Jeremiah Foster wrote: >> >> None of the replies answers the question, which I'll repeat; is there a >> reason why MeeGo uses one and not the other? > > yes. > > we use the "real" glibc that most folks are using.... the eglibc team is > relative new with little track record > (this is nothing against them, it's just reality) and on the flip side > eglibc doesn't really have real advantages > in the MeeGo context.
Just weighing in here too (and agreeing with Arjan) - there was a quick side discussion between the ARM and IA toolchain guys at some point when we noticed we used glibc and not eglibc (yes, it was in CC emails..), but the conclusion was from both ARM and IA: * MeeGo targets all sorts of targets, not only embedded and we should have similar glibc on all of them to get similar behaviour everywhere * Someone eglibc was diffed with glibc and it doesn't actually bring any ARM specific patches, they're already in upstream glibc-ports * What eglibc brings is mostly #ifdefs and very small changes to help make a smaller library for really-embedded purposes (not too useful in MeeGo) * The ARM patches that are usually seen in Debian's eglibc for instance are actually not part of eglibc upstream and we might as well on a case-to-case/upstream-first basis bring these useful patches (from Linaro) into MeeGo glibc packaging. Based on those findings glibc seemed to be a good choice for MeeGo (and MeeGo ARM too) and I personally don't see any reason why to use eglibc instead of glibc in MeeGo. BR Carsten Munk > > _______________________________________________ > MeeGo-dev mailing list > [email protected] > http://lists.meego.com/listinfo/meego-dev > _______________________________________________ MeeGo-dev mailing list [email protected] http://lists.meego.com/listinfo/meego-dev
