This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: Release branch maintenance
- From: Siddhesh Poyarekar <siddhesh at redhat dot com>
- To: Adam Conrad <adconrad at 0c3 dot net>
- Cc: Allan McRae <allan at archlinux dot org>, "Joseph S. Myers" <joseph at codesourcery dot com>, Alexandre Oliva <aoliva at redhat dot com>, "Carlos O'Donell" <carlos at redhat dot com>, Roland McGrath <roland at hack dot frob dot com>, libc-alpha <libc-alpha at sourceware dot org>
- Date: Mon, 3 Feb 2014 09:40:20 +0530
- Subject: Re: Release branch maintenance
- Authentication-results: sourceware.org; auth=none
- References: <52E9DEB7 dot 4000709 at redhat dot com> <52E9E84F dot 50907 at redhat dot com> <52EA682D dot 90900 at archlinux dot org> <ormwid428y dot fsf at livre dot home> <Pine dot LNX dot 4 dot 64 dot 1401302131080 dot 12540 at digraph dot polyomino dot org dot uk> <20140131032418 dot GK2149 at spoyarek dot pnq dot redhat dot com> <Pine dot LNX dot 4 dot 64 dot 1401311709310 dot 26476 at digraph dot polyomino dot org dot uk> <CAAHN_R2r6jiHWwj2HGDS-sruGewKEMmeQkAS9PXqZtOS95gKAw at mail dot gmail dot com> <52EC58E6 dot 6060702 at archlinux dot org> <20140201111221 dot GS15976 at 0c3 dot net>
On Sat, Feb 01, 2014 at 04:12:21AM -0700, Adam Conrad wrote:
> We (both Debian and Ubuntu) pull a *lot* of patches from master back
> into stables and I'll admit that I've been monumentally bad about
> committing those on the upstream branch when we do. I suppose the
You're not alone; I believe most major distributions (including Fedora
and RHEL) are guilty of this.
> thought has been that I'm often not even sure if anyone other than
> us is using the branch in question, so there may not be much shared
> benefit to committing back, but if we all got into the habit, it'd
> certainly make stable maintenance easier, even if we didn't update
> from the stable branches directly. Getting an overview of what fixes
> others deemed important would be helpful.
While I agree that it is the Right Thing to do, it is an overhead to
clone an upstream bugzilla for the branch, post the patch for the
backport, get it reviewed (and maybe even get an ack from the branch
maintainer if (s)he is not the reviewer) and then pull the patch in.
The benefits are even less (zero?) if they don't intend to ever rebase
their distribution tarballs to use the point releases.
Maybe everyone (me and Carlos included) needs to go back into their
distro huddles and decide if it makes sense for their distribution to
do point release rebases at regular intervals. That should be a good
incentive to maintain the branch releases and even monitor what goes
in there.
Siddhesh