cvs commit: src/sys/dev/ath if_ath.c if_athvar.h 1.153

[Daniel Dvořák]

Okay about vendors,

a) and b)
compex and its WPE54AG supports "smartselect" funcion aka DFS and country
select as we talked about it recently.
http://www.i4shop.net/cz/iObchod/Catalog.asp?ca=1620&it=9259
http://www.inwifi.cz/stahnete-si/soubor.php?f=4f1d47475a466cb321f82dea200f92
d1

I tested these 2 boxes, they worked in smartselection mode, yes true, I am
not very sure about that because I did not scan it by kismet, so perhaps
they worked seeming.

About complaining,

I did not want to complain you at all !!! Read my e-mail clearly I said I
disagree with you only one thing, it is how you wrote "slipped requirement".

And directly about Atheros,

they do not answer repeatedly on my question about win32 drivers oid´s so
now I do not count that even they answer if I complain them now. You know.
:)

Daniel

> -----Original Message-----
> From: Sam Leffler [mailto:sam at errno.com] 
> Sent: Wednesday, September 20, 2006 2:33 AM
> To: dandee at volny.cz
> Cc: 'Henrik Brix Andersen'; cvs-src at FreeBSD.org; 
> src-committers at FreeBSD.org; cvs-all at FreeBSD.org
> Subject: Re: cvs commit: src/sys/dev/ath if_ath.c if_athvar.h 1.153
> 
> Daniel Dvor(ák wrote:
> > Hi all,
> > 
> > I disagree with that the requirement had been slipped. 
> Where is it written ?
> > 
> > In our country this requirement still exists and the same 
> is with ETSI 
> > the whole Europe countries. It was required one year ago 
> and the same is now.
> > 
> > Our regulator have not said something similar to "Okay, 
> this DFS was 
> > mistake and after 1 year, now, we do not require it. We wanted that 
> > simple because of funny."
> > 
> > I am surprised, how hardware vendors with whole source of hal still 
> > can respect licence to transmit in 5G band legally with 
> working DFS, 
> > but open source systems did not offer this requirement after many 
> > months and even DFS is abolished.
> > 
> > I am sorry if I was rude, but I get upset that there was 
> not legally 
> > atheros drivers for OSS and is not and it seems it will not.
> > 
> > I still beliefed DFS will be in ath drivers, but not. Do 
> you remember 
> > my questions and interest about DFS and TPC and your answers ?
> > 
> > I understand I am not a developer so I am not entitled to 
> be irritated 
> > with this, but I have a law to tell you what I think about 
> whole it. 
> > (IMHO)
> > 
> > It is a big pity!
> > 
> > Sam, you know that I admire and respect your hard work for 
> our beloved 
> > fbsd, I am sorry if my unexpected words make you angry with 
> me, but I 
> > deeply, very deeply disagree with you about slipped requirement.
> > 
> > There still exists.
> > 
> > Thanks for your attention.
> 
> TPC has been supported for a long time.  DFS can be 
> implemented w/o the hal blindly interfering.  Talk to the 
> madwifi folks about why having the hal try to do radar 
> processing is bad; they suffered through the one hal release 
> Atheros made (note Atheros, not me) where there was an 
> attempt at detecting radar for the purpose of implementing DFS.
> 
> If you read my original commit msg it clearly states I am 
> removing stub code.  If you want to complain that Atheros 
> isn't giving away a radar/DFS implementation then talk to 
> them.  But be sure to show up with a list of vendors that: a) 
> currently do so, or b) provide sufficient information about 
> their hardware that you can implement it yourself.  I figure 
> if you can supply either they will immediately provide 
> code--if it's even possible.
> 
> 	Sam
>