On Wed, Mar 3, 2010 at 10:01 AM, Bernhard Reutner-Fischer <[email protected]> wrote: > On Tue, Mar 02, 2010 at 03:22:44PM -0500, Lennart Sorensen wrote: > >>That sure needs a lot of other patches pulled in to apply. Doesn't seem >>like 0.9.30.3 material unfortunately. I guess the new busybox 1.16.0 >>simply won't be working with uClibc 0.9.30.x. >> >>At least there is hope that someday the master tree will get turned into >>a new release given it works fine. > > During the weekend I've been pondering to release master as 0.9.31 now > (without NPTL), branch it so we can cleanup master after the NPTL merge. > In my mind this would be the best approach: > > .plan is now: > tag 0.9.30.3 off the 0_9_30 branch, push a tarball.
OK > declare master frozen, push a 0.9.31-rc1 tarball (resp point people to > e.g. > http://git.uclibc.org/uClibc/snapshot/uClibc-63859f89f327e48037a4cdba982cd6afa3007da7.tar.bz2 > to be called 0.9.31-rc1; That's current master ATM). > wait and fixup; branch 0.9.31, release it. > merge NPTL into master (later on to be known as 0.9.32).] I would like to propose that we have a longer RC period and stabilize master after merging nptl. 1. There is enough interest base who want it and they will still have to wait for next release which if we go by averages is about a year and chances are nptl will rot once more. It has taken significant effort to get it where it is. 2. other threading implementations remain as it is so all existing users are not affected. > > FYI.. > _______________________________________________ > uClibc mailing list > [email protected] > http://lists.busybox.net/mailman/listinfo/uclibc > _______________________________________________ uClibc mailing list [email protected] http://lists.busybox.net/mailman/listinfo/uclibc
