svn commit: r189594 - head

M. Warner Losh imp at bsdimp.com
Thu Mar 12 00:01:34 PDT 2009


In message: <20090310021926.GA51405 at troutmask.apl.washington.edu>
            Steve Kargl <sgk at troutmask.apl.washington.edu> writes:
: On Mon, Mar 09, 2009 at 08:38:10PM -0500, Mark Linimon wrote:
: > 
: > As mlaier has pointed out, -current has sharp edges.  It's one of
: > 3 choices open to you, the other two being -stable (which will still
: > have ports regressions from time to time -- see xorg -- and sometimes
: > even src regressions), and a release, which is the best we can do with
: > respect to QA.  If you can't deal with having your system out of
: > commission on occasion, then -current isn't for you.
: > 
: 
: Oh Please!
: 
: I've run -current since it was called 386bsd+patchkit.  I've
: lived through the gcc 2.6.3 to gcc 3.x transition, the replacement
: of devfs by phk with a new improved devfs, the problems with
: libm and the changes to stdio.h among many others.  The facts
: remain that the USB2 transistion was poorly executed.  Contrast USB2
: with Ed's new TTY layer.  Ed gave a month or more headsup that a
: new TTY layer was coming.  He enumerated the drivers that were broken
: and actively solicited people with the affected hardware for help.
: He furthermore helped those people fix as many driver as possible
: before committing the new TTY layer.  As part of portmngr, you
: know Ed also actively fixed many ports broken by the new TTY layer
: and/or helped others fix the ports before the new layer became
: standard.  The fact that USB2 broke such a fundamentally important
: port as Xorg suggests a lack of testing and planning by those who
: rushed the USB2 transition.
: 
: If you and others take off your rose colored glasses, you'll see
: that the USB2 transition could have been handled better.  Hopefully,
: you're willing to learn from your mistakes.

I don't think anybody thinks the usb2 transition was a paragon of
virtue.  However, it was handled better than many others.  The usb2
stuff has been in transition into the tree for much longer than the
tty later, and was, frankly, a much bigger change.  I think that it
was handled well enough, all things considered.

The biggest problem with the usb2 transition is that nobody had the
time to drive it into the tree.  There was no one person that
volunteered to do all the steps.  We had several different volunteers
do good work to make it happen, but there wasn't a single point of
contact for it.  Rather than complain about it, you should view future
transitions as an opportunity to help out more...

Warner


More information about the svn-src-all mailing list