cvs commit: src/sys/conf files kern.pre.mk src/sys/dev/em LICENSE e1000_80003es2lan.c e1000_80003es2lan.h e1000_82540.c e1000_82541.c e1000_82541.h e1000_82542.c e1000_82543.c e1000_82543.h e1000_82571.c e1000_82571.h e1000_82575.c ...

[Erik Trulsson]

On Sat, Oct 06, 2007 at 09:53:06PM +0200, Dag-Erling Smørgrav wrote:
> Scott Long <scottl at samsco.org> writes:
> > Dag-Erling Smørgrav <des at des.no> writes:
> > > We used to have a principle that commit bits were granted to individuals
> > > on their individual merit, not simply because they represented a vendor
> > > and were paid to work on drivers for that vendor's hardware.
> > Are you implying that Jack has no merit?  That's an unfortunate
> > assertion.  What evidence do you have to support that?
> 
> It is you who implied rather strongly that Jack is neither more nor less
> than an Intel representative.  You spoke of "lecturing a vendor" when
> all Erik did was point out or own rules to a *committer*.
> 
> > > We also used to have a principle that changes should be tested
> > > before being committed, especially to -STABLE.
> > I guess you missed that part where Jack said that the changes had
> > undergone extensive testing.
> 
> I guess you missed the part where his commit broke the tinderbox,
> because he clearly did not test the DEVICE_POLLING case.

In Jack's defense that particular problem was apparently beacuse a 7.x
interface was left in the code  (bus_setup_intr(9) takes one more argument
in 7.x than it does in 6.x.)  That would not have been caught by having
the code go into -CURRENT first.


-- 
<Insert your favourite quote here.>
Erik Trulsson
ertr1013 at student.uu.se