Locking fixes for dc(4): please test!

[John Baldwin]

On Monday 15 August 2005 08:42 pm, Marcel Moolenaar wrote:
> On Aug 15, 2005, at 10:22 AM, John Baldwin wrote:
> 
> > On Friday 12 August 2005 06:37 pm, Marcel Moolenaar wrote:
> >
> >> On Aug 12, 2005, at 5:59 AM, John Baldwin wrote:
> >>
> >>
> >>>> With that, if_dc seems to work fine. haven't pushed it though...
> >>>> FYI,
> >>>>
> >>>
> >>> Does it respect any media settings from SRM?
> >>>
> >>
> >> No. The media is always auto-select in FreeBSD, even when I change
> >> it to Twisted-Pair or BNC in SRM. Note that the particular I/F had
> >> no cable attached or was otherwise connected in anyway.
> >>
> >
> > Grr.  Does the stock driver in HEAD respect the settings?  (I.e.,  
> > do my
> > patches contain a regression.)
> 
> No, I get auto-select mode unconditionally with the stock driver as
> well. No regression from where I'm sitting...

Hmm, odd.  Thanks for testing it.  It does seem that the only thing we honor 
is the speed setting (i.e. if you force it to 100 half duplex in SRM we 
should honor that, but it doesn't seem to check for twisted-pair vs BNC).

-- 
John Baldwin <jhb at FreeBSD.org>  <><  http://www.FreeBSD.org/~jhb/
"Power Users Use the Power to Serve"  =  http://www.FreeBSD.org