On 3/1/06, Chris Buechler <cbuechler at gmail dot com> wrote:
> (moved to the -dev list, so Scott, primary pfsense developer, will see
> this as he's not on the main list.  he may want to fill in some clue,
> since i'm pretty deficient on this matter)  :)
>
 >
> On 3/1/06, Manuel Kasper <mk at neon1 dot net> wrote:
> >
> > Yep, that's a bug in FreeBSD 6.0's wicontrol. I think FreeBSD PR
> > bin/78570 (http://www.freebsd.org/cgi/query-pr.cgi?pr=bin/78570)
> > describes it, and the patch that it mentions works. Why this hasn't
> > been fixed in 6.0-RELEASE is beyond me.

Sam considers wicontrol dead now that ifconfig does everything.  See:
http://unix.derkeiler.com/Mailing-Lists/FreeBSD/current/2005-05/0611.html
 We worked around this by getting the information from ifconfig
instead to match the NWO.

> They seem to have pushed a ton of wireless changes/fixes out to 6.1.
> I believe for some reason many (most/all?) of the changes couldn't be
> merged into RELENG_6_0 (ABI changes, or whatever their typical
> reasoning is that escapes me at the moment)

We stuck with RELENG_6 mainly for wireless fixes and countless other
small perks that RELENG_6 brings to the table.  In addition I tried to
time our release as close to the current release that we could since
we are so "bleeding edge".

>
> > > ath0: device timeout
> >
> > Hmm, I've experienced that with a somewhat stripped (but otherwisee
> > standard) FreeBSD 6.0-RELEASE installation on a WRAP as well, but
> > didn't investigate. Maybe Jim Thompson can comment?
> >
>
> From some off list emails from Jim related to pfsense development, I'm
> guessing this is probably from the outdated HAL in RELENG_6_0.  I'm
> not really invovled in pfsense development (though I get copied on a
> lot of the emails), but there was a bumpy road with 6_0 for quite a
> while, and pfsense went for quite some time with a HAL from SamL that
> wasn't yet committed into FreeBSD.  This is all in RELENG_6, soon to
> be RELENG_6_1, though.  There are a few other reasons that I don't
> recall that pfsense is tracking RELENG_6 rather than 6_0.

We have used the newer HAL for atleast 6-8 months.  Now its merged
into RELENG_6.

> This wasn't very helpful, but my suggestion is to make the next alpha
> image based on RELENG_6_1, which should be tagged within the next
> week.

If you have any other questions just let me know, getting to where
pfSense is now was definitely not a walk in the park.

Scott