wireless-regdb: Allow 6ghz in the US

[Chen-Yu Tsai]

(When replying, please reply to all)

On Wed, Feb 26, 2025 at 5:05 PM <rmandrad at gmail.com> wrote:
>
> Thank you. Apologies this is the first time submitting a kernel patch. The original submission set NO-IR where clearly the document notes it is allowed as section 59 states hence the patch - with increase usage of 6Ghz we have now Openwrt users complaining they can't the band in the US due to NO-IR
>
> Shall I update the patch to only remove NO-IR and increase power to 30 dBm ?

As explained, the limit is lowered from 30 dBm to 12 dBm to comply with
the PSD requirements when using a 20 MHz channel. I suspect no one would
use such a narrow band, but the requirements exist and the database rule
should not be written in a way that allows the user to exceed the limit,
i.e. using 30 dBm on a 20 MHz channel.

So please update the patch to only remove NO-IR. You can look at previous
commits to see how we put URLs in the commit message.

Thanks
ChenYu

> Thank you again
>
> -----Original Message-----
> From: Chen-Yu Tsai <wens at kernel.org>
> Sent: 26 February 2025 08:58
> To: rmandrad at gmail.com
> Cc: linux-wireless at vger.kernel.org; wireless-regdb at lists.infradead.org; Dennis Bland <dennis at dbperformance.com>; Ping-Ke Shih <pkshih at realtek.com>
> Subject: Re: wireless-regdb: Allow 6ghz in the US
>
> CC-ing Dennis, the original submitter, and also Ping-Ke, who has done a lot of 6 GHz updates, for more information.
>
> On Wed, Feb 26, 2025 at 4:03 PM <rmandrad at gmail.com> wrote:
> >
> > Allow 6ghz in the US
> >
> > https://www.federalregister.gov/documents/2020/05/26/2020-11236/unlice
> > nsed-u se-of-the-6-ghz-band allows the use of 6ghz in the US namely
> > section
> > 59 https://www.federalregister.gov/d/2020-11236/p-66 with absolute
> > radiated power of 30 dBm for the 320 megahertz channel
>
> Please don't wrap URLs.
>
> Please see the original submission [1], which explains why the power limit is so low. Basically, neither the database nor the kernel supports specifying power spectral density limits, so we can only take the narrowest bandwidth to calculate the applicable power limit.
>
> [1] https://lore.kernel.org/wireless-regdb/CAPRryQp6j4UKvLZCkMAuQdaxepMBETQUJ1eNULJSh3ZWXC0f5Q@mail.gmail.com/
>
> > based on this remove NO-IR flag and allow 30 dBm max power
>
> The original submission mentioned NO-IR requirements, though I did not find such wording. Dennis, do you have any ideas?
>
> > Signed-off-by: Rudy Andram <rmandrad at gmail.com>
> >
> > diff --git a/db.txt b/db.txt
> > index 803f1bc..bc2b4fe 100644
> > --- a/db.txt
> > +++ b/db.txt
> > @@ -1953,7 +1953,8 @@ country US: DFS-FCC
> >         (5850 - 5895 @ 40), (27), NO-OUTDOOR, AUTO-BW, NO-IR
> >         # 6g band
> >         #
> > https://www.federalregister.gov/documents/2020/05/26/2020-11236/unlice
> > nsed-u
> > se-of-the-6ghz-band
> > -       (5925 - 7125 @ 320), (12), NO-OUTDOOR, NO-IR
> > +       (5925 - 6425 @ 320), (30), NO-OUTDOOR
> > +       (6525 - 6875 @ 320), (30), NO-OUTDOOR
>
> The database entry targets LPI usage instead of standard usage, which requires the presence of AFC, which is also a requirement that the database is unable to represent. And under LPI usage, the full 6GHz band (U-NII-5, 6, 7) is available.
>
>
> Thanks
> ChenYu
>
> >         # 60g band
> >         # reference: section IV-D
> > https://docs.fcc.gov/public/attachments/FCC-16-89A1.pdf
> >         # channels 1-6 EIRP=40dBm(43dBm peak)
> >
> >
> >
>