minutes of meeting

[Jim Gettys]

Alan, Jim, Egbert, Stuart, Leon, Kevin 

Mike Harris committed stuff he shouldn't have, that were new features.

Process problems: we need to deal with the process, particularly on
release branches.

We need to be more careful right now, than say, in a year when we
expect many more developers to be running the code than has been the
case recently on XF86.

What happens on release branches?  What happens on different parts of
the system?

It is OK for anyone to put bugs and proposed patches on the release
blocker's list, to ensure they get discussed and decisions taken.

The issue is we've branched for release: no new features. So the
PPC-64 is probably a bad idea to bring in at this date.  Egbert's
decisions rule.

http://freedesktop.org/cgi-bin/bugzilla/show_bug.cgi?id=297
The issue with the sleep problem is mostly: do we really understand the
issue?  Is there a lurking bug that really needs to get fixed?
The sleep is *clearly* bogus, but just removing it without understanding
is probably a mistake; there may be practical security issues to
deal with, according to the comments.

We need to develop significant policies for different parts of the
system.  For example, fully new drivers that won't effect other parts of
the system can be acceptable late in a release cycle, but things that
might make other parts of the system unstable should not.

We reviewed bugs in:
http://freedesktop.org/cgi-bin/bugzilla/show_bug.cgi?id=269
These need to be broken out into separate bugs, so that we
can track what goes on the release branch versus current.

Jim reported on meeting with lawyers.  We'll get list of what to do into
bugzilla, so we can get them worked on.
                             - Jim


-- 
Jim Gettys <Jim.Gettys@hp.com>
HP Labs, Cambridge Research Laboratory