[PATCH 4/5] crda: Updated regulatory information for France (FR)

[PATCH 4/5] crda: Updated regulatory information for France (FR)

[Benoit PAPILLAULT]

This patch is for review. It might be tricky to use the INDOOR/OUTDOOR
with those definitions, but it will be another issue.


Signed-off-by: Benoit PAPILLAULT <benoit.papillault@free.fr>
---
 db.txt |   15 ++++++++++-----
 1 files changed, 10 insertions(+), 5 deletions(-)

diff --git a/db.txt b/db.txt
index 6a282b6..bfc91ef 100644
--- a/db.txt
+++ b/db.txt
@@ -199,12 +199,17 @@ country FI:
 	(5170 - 5250 @ 40), (N/A, 20)
 	(5250 - 5330 @ 40), (N/A, 20), DFS
 	(5490 - 5710 @ 40), (N/A, 27), DFS
-
+# Data from http://www.arcep.fr/index.php?id=9272#12931
+# Updated 2008-10-04
+# 1W => 30dBm, 200mW => 23dBm, 100mW => 20dBm, 10mW => 10dBm
+# TPC is needed for 5250-5350 and 5470-5725, which flag to use?
 country FR:
-	(2402 - 2482 @ 40), (N/A, 20)
-	(5170 - 5250 @ 40), (N/A, 20)
-	(5250 - 5330 @ 40), (N/A, 20), DFS
-	(5490 - 5710 @ 40), (N/A, 27), DFS
+	(2400 - 2454 @ 40), (N/A, 20)
+	(2454 - 2483.5 @ 40), (N/A, 20), NO-OUTDOOR
+	(2454 - 2483.5 @ 40), (N/A, 10), NO-INDOOR
+	(5150 - 5250 @ 40), (N/A, 23), NO-OUTDOOR
+	(5250 - 5350 @ 40), (N/A, 23), NO-OUTDOOR, DFS, PASSIVE-SCAN
+	(5470 - 5725 @ 40), (N/A, 30), DFS, PASSIVE-SCAN

 country GE:
 	(2402 - 2482 @ 40), (N/A, 20)
-- 
1.5.6.5

Re: [PATCH 4/5] crda: Updated regulatory information for France (FR)

[Johannes Berg]

[-- Attachment #1: Type: text/plain, Size: 665 bytes --]

On Tue, 2008-10-21 at 22:20 +0200, Benoit PAPILLAULT wrote:

> +# Data from http://www.arcep.fr/index.php?id=9272#12931
> +# Updated 2008-10-04
> +# 1W => 30dBm, 200mW => 23dBm, 100mW => 20dBm, 10mW => 10dBm

You can say "200mW" in the definitions and the tool will automatically
convert it.

> +# TPC is needed for 5250-5350 and 5470-5725, which flag to use?
>  country FR:
> -	(2402 - 2482 @ 40), (N/A, 20)
> -	(5170 - 5250 @ 40), (N/A, 20)
> -	(5250 - 5330 @ 40), (N/A, 20), DFS
> -	(5490 - 5710 @ 40), (N/A, 27), DFS
> +	(2400 - 2454 @ 40), (N/A, 20)

i.e.

(2400 - 2454 @ 40), (N/A, 100mW)

I think I did this for the DE one.

johannes

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [PATCH 4/5] crda: Updated regulatory information for France (FR)

[Luis R. Rodriguez]

CC'ing Jean as he's another French wireless developer.

On Tue, Oct 21, 2008 at 01:20:46PM -0700, Benoit PAPILLAULT wrote:
> This patch is for review. It might be tricky to use the INDOOR/OUTDOOR
> with those definitions, but it will be another issue.

INDOOR/OUTDOOR flag is welcomed when it can be determined. In order
to make use of it we will need to add wireless state in cfg80211, and
let the user be able to update this, say through iw or wpa_supplicant.
But that's for later just as with DFS.

But using it in the db is great, thanks.

> Signed-off-by: Benoit PAPILLAULT <benoit.papillault@free.fr>
> ---
>  db.txt |   15 ++++++++++-----
>  1 files changed, 10 insertions(+), 5 deletions(-)
> 
> diff --git a/db.txt b/db.txt
> index 6a282b6..bfc91ef 100644
> --- a/db.txt
> +++ b/db.txt
> @@ -199,12 +199,17 @@ country FI:
>         (5170 - 5250 @ 40), (N/A, 20)
>         (5250 - 5330 @ 40), (N/A, 20), DFS
>         (5490 - 5710 @ 40), (N/A, 27), DFS
> -
> +# Data from http://www.arcep.fr/index.php?id=9272#12931
> +# Updated 2008-10-04
> +# 1W => 30dBm, 200mW => 23dBm, 100mW => 20dBm, 10mW => 10dBm
> +# TPC is needed for 5250-5350 and 5470-5725, which flag to use?
>  country FR:
> -       (2402 - 2482 @ 40), (N/A, 20)
> -       (5170 - 5250 @ 40), (N/A, 20)
> -       (5250 - 5330 @ 40), (N/A, 20), DFS
> -       (5490 - 5710 @ 40), (N/A, 27), DFS
> +       (2400 - 2454 @ 40), (N/A, 20)
> +       (2454 - 2483.5 @ 40), (N/A, 20), NO-OUTDOOR

2483.5 - 2454 = 29.5 so 29.5 should be the max bandwidth.

> +       (2454 - 2483.5 @ 40), (N/A, 10), NO-INDOOR

Same here.

> +       (5150 - 5250 @ 40), (N/A, 23), NO-OUTDOOR
> +       (5250 - 5350 @ 40), (N/A, 23), NO-OUTDOOR, DFS, PASSIVE-SCAN
> +       (5470 - 5725 @ 40), (N/A, 30), DFS, PASSIVE-SCAN

Hm, actually so I had removed PASSIVE-SCAN from all entries
as I had determined that this was only used for DFS purposes.
As you can see in your case both have DFS so I'd leave only DFS.
So essentially we can get rid of these:

        RRF_PASSIVE_SCAN        = 1<<7, /* passive scan is required */
        RRF_NO_IBSS             = 1<<8, /* IBSS is not allowed */

What I noted was that PASSIVE-SCAN for example was only used for when
we don't have DFS in STA and NO_IBSS when we don't have DFS in IBSS.

So we can just stick to DFS flag, unless you are aware of other
considerations for them. Thoughts?

  Luis

Re: [PATCH 4/5] crda: Updated regulatory information for France (FR)

[Jean Tourrilhes]

On Wed, Oct 22, 2008 at 08:03:23AM -0700, Luis R. Rodriguez wrote:
> CC'ing Jean as he's another French wireless developer.
> 

	I have not been living in France for a while, like more than
10 years ;-)
	I think Benoit is on top of things, and as I'm not familiar
with it, I'll defer to him.

	Jean

Re: [PATCH 4/5] crda: Updated regulatory information for France (FR)

[Benoit PAPILLAULT]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Jean Tourrilhes a =E9crit :
> On Wed, Oct 22, 2008 at 08:03:23AM -0700, Luis R. Rodriguez wrote:
>> CC'ing Jean as he's another French wireless developer.
>>
>=20
> 	I have not been living in France for a while, like more than
> 10 years ;-)
> 	I think Benoit is on top of things, and as I'm not familiar
> with it, I'll defer to him.
>=20
> 	Jean

Indeed, I read lots of paper, website, source code and I know a lot for
sure. But at the end, that's my own interpretation and I might still be
wrong, so having peer review is still needed.

Regards,
Benoit
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFJAAisOR6EySwP7oIRAhW9AJ9vUcgHiPLQGcUwgyHRBsKdMLZgcwCfQEyn
bwGnbRXBEARYX6TCy3eqVns=3D
=3DRr7k
-----END PGP SIGNATURE-----
--
To unsubscribe from this list: send the line "unsubscribe linux-wireles=
s" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 4/5] crda: Updated regulatory information for France (FR)

[Benoit PAPILLAULT]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Luis R. Rodriguez a =E9crit :
> CC'ing Jean as he's another French wireless developer.
>=20
> On Tue, Oct 21, 2008 at 01:20:46PM -0700, Benoit PAPILLAULT wrote:
>> This patch is for review. It might be tricky to use the INDOOR/OUTDO=
OR
>> with those definitions, but it will be another issue.
>=20
> INDOOR/OUTDOOR flag is welcomed when it can be determined. In order
> to make use of it we will need to add wireless state in cfg80211, and
> let the user be able to update this, say through iw or wpa_supplicant=
=2E
> But that's for later just as with DFS.

=46ine with me here.

>=20
> But using it in the db is great, thanks.
>=20
>> Signed-off-by: Benoit PAPILLAULT <benoit.papillault@free.fr>
>> ---
>>  db.txt |   15 ++++++++++-----
>>  1 files changed, 10 insertions(+), 5 deletions(-)
>>
>> diff --git a/db.txt b/db.txt
>> index 6a282b6..bfc91ef 100644
>> --- a/db.txt
>> +++ b/db.txt
>> @@ -199,12 +199,17 @@ country FI:
>>         (5170 - 5250 @ 40), (N/A, 20)
>>         (5250 - 5330 @ 40), (N/A, 20), DFS
>>         (5490 - 5710 @ 40), (N/A, 27), DFS
>> -
>> +# Data from http://www.arcep.fr/index.php?id=3D9272#12931
>> +# Updated 2008-10-04
>> +# 1W =3D> 30dBm, 200mW =3D> 23dBm, 100mW =3D> 20dBm, 10mW =3D> 10dB=
m
>> +# TPC is needed for 5250-5350 and 5470-5725, which flag to use?
>>  country FR:
>> -       (2402 - 2482 @ 40), (N/A, 20)
>> -       (5170 - 5250 @ 40), (N/A, 20)
>> -       (5250 - 5330 @ 40), (N/A, 20), DFS
>> -       (5490 - 5710 @ 40), (N/A, 27), DFS
>> +       (2400 - 2454 @ 40), (N/A, 20)
>> +       (2454 - 2483.5 @ 40), (N/A, 20), NO-OUTDOOR
>=20
> 2483.5 - 2454 =3D 29.5 so 29.5 should be the max bandwidth.

Not exactly. The maximum power in this band depends on whether we are
indoor or outdoor. If we are indoor, we are allowed to have 20dBm in th=
e
whole 2400 - 2454 MHz band and can still use 40MHz bandwidth. Maybe I
should describe that different?

	(2412 - 2483.5 @ 40), (N/A, 20), NO-OUTDOOR
        (2412 - 2454 @ 40), (N/A, 20), NO-INDOOR
        (2454 - 2483.5 @ 40) (N/A, 10), NO-INDOOR

In fact, this depends on the indoor/outdoor flags issue which can be
resolved later.

>=20
>> +       (2454 - 2483.5 @ 40), (N/A, 10), NO-INDOOR
>=20
> Same here.
>=20
>> +       (5150 - 5250 @ 40), (N/A, 23), NO-OUTDOOR
>> +       (5250 - 5350 @ 40), (N/A, 23), NO-OUTDOOR, DFS, PASSIVE-SCAN
>> +       (5470 - 5725 @ 40), (N/A, 30), DFS, PASSIVE-SCAN
>=20
> Hm, actually so I had removed PASSIVE-SCAN from all entries
> as I had determined that this was only used for DFS purposes.
> As you can see in your case both have DFS so I'd leave only DFS.
> So essentially we can get rid of these:
>=20
>         RRF_PASSIVE_SCAN        =3D 1<<7, /* passive scan is required=
 */
>         RRF_NO_IBSS             =3D 1<<8, /* IBSS is not allowed */
>=20
> What I noted was that PASSIVE-SCAN for example was only used for when
> we don't have DFS in STA and NO_IBSS when we don't have DFS in IBSS.
>=20
> So we can just stick to DFS flag, unless you are aware of other
> considerations for them. Thoughts?

If it's documented that the DFS flag implies PASSIVE_SCAN, that's fine
with me.

>=20
>   Luis

Regards,
Benoit
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFJAAwhOR6EySwP7oIRApcqAKC/+B/fPUu6+8caHOqnA/GwsnGCdwCfcuSR
YX46su0CSrQCynR/RQ1CLnM=3D
=3D9NM9
-----END PGP SIGNATURE-----
--
To unsubscribe from this list: send the line "unsubscribe linux-wireles=
s" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 4/5] crda: Updated regulatory information for France (FR)

[Luis R. Rodriguez]

On Wed, Oct 22, 2008 at 10:31:13PM -0700, Benoit PAPILLAULT wrote:
> >> +# Data from http://www.arcep.fr/index.php?id=9272#12931
> >> +# Updated 2008-10-04
> >> +# 1W => 30dBm, 200mW => 23dBm, 100mW => 20dBm, 10mW => 10dBm
> >> +# TPC is needed for 5250-5350 and 5470-5725, which flag to use?
> >>  country FR:
> >> -       (2402 - 2482 @ 40), (N/A, 20)
> >> -       (5170 - 5250 @ 40), (N/A, 20)
> >> -       (5250 - 5330 @ 40), (N/A, 20), DFS
> >> -       (5490 - 5710 @ 40), (N/A, 27), DFS
> >> +       (2400 - 2454 @ 40), (N/A, 20)
> >> +       (2454 - 2483.5 @ 40), (N/A, 20), NO-OUTDOOR
> >
> > 2483.5 - 2454 = 29.5 so 29.5 should be the max bandwidth.
> 
> Not exactly. The maximum power in this band depends on whether we are
> indoor or outdoor. If we are indoor, we are allowed to have 20dBm in the
> whole 2400 - 2454 MHz band and can still use 40MHz bandwidth. Maybe I
> should describe that different?
> 
>         (2412 - 2483.5 @ 40), (N/A, 20), NO-OUTDOOR
>         (2412 - 2454 @ 40), (N/A, 20), NO-INDOOR
>         (2454 - 2483.5 @ 40) (N/A, 10), NO-INDOOR
> 
> In fact, this depends on the indoor/outdoor flags issue which can be
> resolved later.

But I wasn't speaking about power, I was speaking about channel
bandwidth. You cannot fit a 40 MHz channel into the freq range
2483.5 - 2454.

> >> +       (2454 - 2483.5 @ 40), (N/A, 10), NO-INDOOR
> >
> > Same here.
> >
> >> +       (5150 - 5250 @ 40), (N/A, 23), NO-OUTDOOR
> >> +       (5250 - 5350 @ 40), (N/A, 23), NO-OUTDOOR, DFS, PASSIVE-SCAN
> >> +       (5470 - 5725 @ 40), (N/A, 30), DFS, PASSIVE-SCAN
> >
> > Hm, actually so I had removed PASSIVE-SCAN from all entries
> > as I had determined that this was only used for DFS purposes.
> > As you can see in your case both have DFS so I'd leave only DFS.
> > So essentially we can get rid of these:
> >
> >         RRF_PASSIVE_SCAN        = 1<<7, /* passive scan is required */
> >         RRF_NO_IBSS             = 1<<8, /* IBSS is not allowed */
> >
> > What I noted was that PASSIVE-SCAN for example was only used for when
> > we don't have DFS in STA and NO_IBSS when we don't have DFS in IBSS.
> >
> > So we can just stick to DFS flag, unless you are aware of other
> > considerations for them. Thoughts?
> 
> If it's documented that the DFS flag implies PASSIVE_SCAN, that's fine
> with me.

We'll have to add the docs, this is just what what I determined from my
own review on regulatory, with regulatory folks. I asked a "why" for
each flag.

  Luis

Re: [PATCH 4/5] crda: Updated regulatory information for France (FR)

[Benoit PAPILLAULT]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Luis R. Rodriguez a =E9crit :
> On Wed, Oct 22, 2008 at 10:31:13PM -0700, Benoit PAPILLAULT wrote:
>>> 2483.5 - 2454 =3D 29.5 so 29.5 should be the max bandwidth.
>> Not exactly. The maximum power in this band depends on whether we ar=
e
>> indoor or outdoor. If we are indoor, we are allowed to have 20dBm in=
 the
>> whole 2400 - 2454 MHz band and can still use 40MHz bandwidth. Maybe =
I
>> should describe that different?
>>
>>         (2412 - 2483.5 @ 40), (N/A, 20), NO-OUTDOOR
>>         (2412 - 2454 @ 40), (N/A, 20), NO-INDOOR
>>         (2454 - 2483.5 @ 40) (N/A, 10), NO-INDOOR
>>
>> In fact, this depends on the indoor/outdoor flags issue which can be
>> resolved later.
>=20
> But I wasn't speaking about power, I was speaking about channel
> bandwidth. You cannot fit a 40 MHz channel into the freq range
> 2483.5 - 2454.

The whole band is still 2412-2483.5MHz in both cases (indoor/outdoor),
this is just that different TX power rules apply for frequencies below
2454 MHz and above when used outdoor.

To me, it's valid to have a 40MHz channel centered at 2462 MHz covering
the range [2442, 2482] MHz. When used outdoor, it will be limited to 10
dBm on the whole range.

Regards,
Benoit
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFJAYuHOR6EySwP7oIRAseyAKCjiDkOs39MIBnhjA3OQDsgfCJzYQCgr1aA
R9UNky9HVhPQwtl9pwInL88=3D
=3D7Rzj
-----END PGP SIGNATURE-----
--
To unsubscribe from this list: send the line "unsubscribe linux-wireles=
s" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 4/5] crda: Updated regulatory information for France (FR)

[Luis R. Rodriguez]

On Fri, Oct 24, 2008 at 01:47:03AM -0700, Benoit PAPILLAULT wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>=20
> Luis R. Rodriguez a =E9crit :
> > On Wed, Oct 22, 2008 at 10:31:13PM -0700, Benoit PAPILLAULT wrote:
> >>> 2483.5 - 2454 =3D 29.5 so 29.5 should be the max bandwidth.
> >> Not exactly. The maximum power in this band depends on whether we =
are
> >> indoor or outdoor. If we are indoor, we are allowed to have 20dBm =
in the
> >> whole 2400 - 2454 MHz band and can still use 40MHz bandwidth. Mayb=
e I
> >> should describe that different?
> >>
> >>         (2412 - 2483.5 @ 40), (N/A, 20), NO-OUTDOOR
> >>         (2412 - 2454 @ 40), (N/A, 20), NO-INDOOR
> >>         (2454 - 2483.5 @ 40) (N/A, 10), NO-INDOOR
> >>
> >> In fact, this depends on the indoor/outdoor flags issue which can =
be
> >> resolved later.
> >
> > But I wasn't speaking about power, I was speaking about channel
> > bandwidth. You cannot fit a 40 MHz channel into the freq range
> > 2483.5 - 2454.
>=20
> The whole band is still 2412-2483.5MHz in both cases (indoor/outdoor)=
,
> this is just that different TX power rules apply for frequencies belo=
w
> 2454 MHz and above when used outdoor.
>=20
> To me, it's valid to have a 40MHz channel centered at 2462 MHz coveri=
ng
> the range [2442, 2482] MHz. When used outdoor, it will be limited to =
10
> dBm on the whole range.

This is an interesting case... so right now we apply a reg rule to
a channel upon register or when a regulatory domain changes. We do
this in handle_channel(). This calls freq_reg_info() and one parameter
we can pass there is the max_bandwidth we know we want or 0 if we have
no requirement. We use 0 on max_bandwidth right now as we haven't added
code for checking max bandwidth yet in mac80211, I think we can add
desired tx power here as well and it'll then be able to pick up the
correct rule. However this does require adding code to reg.c. and
maybe mac80211. We'll have to keep this in mind and add it, but
the reg rule for 2454 - 2483.5 should still not have 40 MHz for max
bandwidth as only 29.5 MHz fits in it.

  Luis
--
To unsubscribe from this list: send the line "unsubscribe linux-wireles=
s" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html