svn commit: r189594 - head

[Max Laier]

On Monday 09 March 2009 23:27:05 Steve Kargl wrote:
> On Mon, Mar 09, 2009 at 12:58:06PM -0700, Andrew Thompson wrote:
> > Not sure if the problems you are referring to extend beyond your
> > previous mail about it being rushed, if so please advise.
> >
> > While the merge hasnt been perfect it couldnt have really been done
> > differently due to finite resources and I dont believe it was
> > unreasonable for a change in HEAD.
>
> I have no problems with a transition for old USB to USB2 in HEAD.
> The manner of execution of the transition leaves much to desire.
> Don't the 20090215 and 20090216 entries in src/UPDATING send up
> a red flag that perhaps the people rushing USB2 into the tree
> might want to ask portmngr to build the port collection on pointyhat

There is a clear chicken-egg problem here.  This is current (aka the bleeding 
edge) for a reason.  This is the place where painful operations like this 
happen from time to time and users can either live with the pain that 
sometimes comes with running current or hold off on updating while the 
problems are being worked out.  Better yet, pitch in with constructive reports 
on specific problems and maybe possible solutions.  Complaining, esp. without 
specific problem reports, is - IMHO - not an option.  Remember that all this 
is done so that things will be in order when 8.0 comes.  Judging from my 
experience with USB2 - things will not only be in order, but also in far 
better shape then they have been for a long time now.

So please, get off the back of those that stepped up to take care of this 
vital piece of software and let them do their work - better yet, help with 
this truly thankless job with specific problem reports and maybe investigating 
solutions.  All this nay-saying isn't helping at all.

> to gauge the damager?  When Mark Linimon, a member of portmngr, posts
> on Feb 26th
>
> (http://lists.freebsd.org/pipermail/freebsd-ports/2009-February/053282.html
>
> There appears to be a disconnect with USB2 development and the
> rest of FreeBSD.
>
> Don't the 20090223, 20090227, and 20090309 entries suggest to you
> that USB2 is going to get very limited testing by the actual user
> community?  Asking users to rebuild world/kernel 2 or 3 times in
> a span of 20 days, and all the ports that use USB (with the hope
> that the ports actuall build) seems destiny to limit testing.
>
>
> After a complete build{world,kernel}/install{world,kernel} dance, including
> a 'make delete-old-libs' and a reboot.
>
> REMOVE:root[214] ll /usr/lib/libusb*
> -r--r--r--  1 root  wheel  - 32136 Mar  9 14:33 /usr/lib/libusb.so.1
> -r--r--r--  1 root  wheel  - 37580 Mar  5 05:01 /usr/lib/libusb20.a
> lrwxr-xr-x  1 root  wheel  -    13 Mar  5 05:01 /usr/lib/libusb20.so@ ->
> libusb20.so.1 -r--r--r--  1 root  wheel  - 39960 Mar  5 05:01
> /usr/lib/libusb20_p.a -r--r--r--  1 root  wheel  - 11874 Mar  9 14:33
> /usr/lib/libusbhid.a lrwxr-xr-x  1 root  wheel  -    14 Mar  9 14:33
> /usr/lib/libusbhid.so@ -> libusbhid.so.3 -r--r--r--  1 root  wheel  - 11284
> Mar  9 14:33 /usr/lib/libusbhid.so.3 -r--r--r--  1 root  wheel  - 12240 Mar
>  9 14:33 /usr/lib/libusbhid_p.a
>
> It seems that libusb20 lives.

-- 
/"\  Best regards,                      | mlaier at freebsd.org
\ /  Max Laier                          | ICQ #67774661
 X   http://pf4freebsd.love2party.net/  | mlaier at EFnet
/ \  ASCII Ribbon Campaign              | Against HTML Mail and News