a few questions on regdomain.xml

From: Andriy Gapon <avg_at_FreeBSD.org>
Date: Thu, 25 Sep 2025 15:24:16 UTC
My first question is where does the PASSIVE flag come from?
My impression from skimming through some web pages is that DFS implies passive scan.
But I see that we have some channels with DFS but without PASSIVE.
Can that be correct?

There are also channels with PASSIVE but without DFS.
E.g., this one (channel 165 covering 5815..5835 MHz band) under FCC:
     <band>
       <freqband ref="F1_5825_5825"/>
       <maxpower>23</maxpower>
       <flags>IEEE80211_CHAN_PASSIVE</flags>
     </band>
But I do not see anything special about it in Linux regdb [1] or in the 
Wikipedia page about WLAN channels [2].

Also, comparing our regdomain.xml and Linux regdb I fail to see what their 
equivalent of our PASSIVE flag is and what our equivalent of their AUTO-BW flag is.
That would be my another question (or a couple).

Then, looking at ETSI domain I see a couple of strange VHT160 channels.
First:
     <!-- 5150-5250/80, 200 mW, indoor -->
...
     <band>
       <freqband ref="AC2_5180_5320_160"/>
       <maxpower>22</maxpower>
       <flags>IEEE80211_CHAN_HT40</flags>
       <flags>IEEE80211_CHAN_VHT160</flags>
       <flags>IEEE80211_CHAN_DFS</flags>
     </band>
As I understand, 160 MHz channels are not allowed in this frequency range at 
all.  Besides, the channel extends beyond the range into a range with different 
requirements.  Also, it's missing INDOOR.  But I think that the channel just 
shouldn't be there.

Second such channel.  Ah, sorry, bz@ recently (relatively) removed it.  It was 
AC2_5745_5885_160 and extended beyond the allowed frequency range.
FWIW, freqband AC2_5745_5885_160 still remains in the file.  Maybe it could be 
useful in some other domain...

My last question is about some "borderline" ETSI channels.
There is an odd 5725 MHz line that separates the short range device frequency 
range.  It's odd because it is not on a channel boundary.  It cuts through 
channel 144 (5710..5730 MHz, center frequency 5720).

Our solution to that oddity seems to be to disallow channel 144 (20 MHz band) 
and also potential 40 MHz and 80 MHz channels that would have that band at the end.

 From a survey of what others do in that area it seems that they actually allow 
those channels these days.  Since the channels step a little bit into the SRD 
territory, they seems to have to obey the limits from both sides of the border.
That is, both DFS and 25 mW power limit.
But if the Wikipedia article to be trusted, that applies only to the 40 and 80 
MHz channels.  The 20 MHz channel (channel 144) has only the DFS restriction.
Not sure what that claim is based and, so, if it is true.

Links:
[1] - 
https://git.kernel.org/pub/scm/linux/kernel/git/wens/wireless-regdb.git/tree/db.txt
[2] - 
https://en.wikipedia.org/wiki/List_of_WLAN_channels#5_GHz_(802.11a/h/n/ac/ax/be)

-- 
Andriy Gapon