[PATCH] mac80211_hwsim: set the maximum EIRP output power for 5GHz

[PATCH] mac80211_hwsim: set the maximum EIRP output power for 5GHz

[Ramon Fontes]

ETSI has been set the maximum EIRP output power to 36 dBm (4000 mW)
Source: https://www.etsi.org/deliver/etsi_en/302500_302599/302502/01.02.01_60/en_302502v010201p.pdf

Signed-off-by: Ramon Fontes <ramonreisfontes@gmail.com>
---
 drivers/net/wireless/mac80211_hwsim.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c
index 14f562cd7..af83791df 100644
--- a/drivers/net/wireless/mac80211_hwsim.c
+++ b/drivers/net/wireless/mac80211_hwsim.c
@@ -305,7 +305,7 @@ static struct net_device *hwsim_mon; /* global monitor netdev */
 	.band = NL80211_BAND_5GHZ, \
 	.center_freq = (_freq), \
 	.hw_value = (_freq), \
-	.max_power = 20, \
+	.max_power = 36, \
 }
 
 static const struct ieee80211_channel hwsim_channels_2ghz[] = {
-- 
2.17.1

Re: [PATCH] mac80211_hwsim: set the maximum EIRP output power for 5GHz

[Johannes Berg]

On Fri, 2019-11-08 at 12:20 -0300, Ramon Fontes wrote:
> ETSI has been set the maximum EIRP output power to 36 dBm (4000 mW)
> Source: https://www.etsi.org/deliver/etsi_en/302500_302599/302502/01.02.01_60/en_302502v010201p.pdf

How is hwsim related to ETSI? What does it matter?

johannes

Re: [PATCH] mac80211_hwsim: set the maximum EIRP output power for 5GHz

[Ramon Fontes]

> How is hwsim related to ETSI? What does it matter?

It's well known that the frequency bands 2,4 GHz and 5 GHz are mainly
used by Radio LANs and in many cases, the deployed technology is based
on the IEEE 802.11 standards family. However, other technologies such
as LTE-LAA are deployed in those frequency bands as well. That said,
considering that hwsim is an excellent module that can be used in
different network simulation scenarios; that it is not only used in
North America; and also considering that some regulatory power limits
are taken from the ETSI standards, why not set a maximum value
supported by a renowned Institute? Without this new value, regdomain
will not work as expected for some countries.

--
Ramon Fontes

Re: [PATCH] mac80211_hwsim: set the maximum EIRP output power for 5GHz

[Johannes Berg]

On Fri, 2019-11-22 at 09:52 -0300, Ramon Fontes wrote:
> > How is hwsim related to ETSI? What does it matter?
> 
> It's well known that the frequency bands 2,4 GHz and 5 GHz are mainly
> used by Radio LANs and in many cases, the deployed technology is based
> on the IEEE 802.11 standards family. However, other technologies such
> as LTE-LAA are deployed in those frequency bands as well. That said,
> considering that hwsim is an excellent module that can be used in
> different network simulation scenarios; that it is not only used in
> North America; and also considering that some regulatory power limits
> are taken from the ETSI standards, why not set a maximum value
> supported by a renowned Institute? Without this new value, regdomain
> will not work as expected for some countries.

Right, so the commit log should say that it should be incremented to
allow regdb to work, rather than worry about ETSI specifics?

Or maybe this limit should just be removed entirely?

johannes

Re: [PATCH] mac80211_hwsim: set the maximum EIRP output power for 5GHz

[Ramon Fontes]

> Right, so the commit log should say that it should be incremented to
> allow regdb to work, rather than worry about ETSI specifics?
>
> Or maybe this limit should just be removed entirely?

Hmm.. not sure. Perhaps we should add only one more information:

ETSI has been set the maximum EIRP output power to 36 dBm (4000 mW)
Source: https://www.etsi.org/deliver/etsi_en/302500_302599/302502/01.02.01_60/en_302502v010201p.pdf

+ The new maximum EIRP output power also allows regdb to work
correctly when txpower is greater than 20 dBm.

Since there is no standard defining greater txpower, in my opinion we
should keep the maximum value. What do you think?

Do I need to submit a new patch?

Re: [PATCH] mac80211_hwsim: set the maximum EIRP output power for 5GHz

[Johannes Berg]

On Fri, 2019-11-22 at 11:19 -0300, Ramon Fontes wrote:
> > Right, so the commit log should say that it should be incremented to
> > allow regdb to work, rather than worry about ETSI specifics?
> > 
> > Or maybe this limit should just be removed entirely?
> 
> Hmm.. not sure. Perhaps we should add only one more information:
> 
> ETSI has been set the maximum EIRP output power to 36 dBm (4000 mW)
> Source: https://www.etsi.org/deliver/etsi_en/302500_302599/302502/01.02.01_60/en_302502v010201p.pdf
> 
> + The new maximum EIRP output power also allows regdb to work
> correctly when txpower is greater than 20 dBm.
> 
> Since there is no standard defining greater txpower, in my opinion we
> should keep the maximum value. What do you think?

It just feels to me like if the only restriction in the driver is
regulatory, we shouldn't have it in the driver. That's what we have the
regulatory database for.

If there's some other (physical?) restriction in the driver, sure, maybe
it should have one there, but for pure regulatory I'm not sure I see it.

That's why the pointer here to ETSI feels so strange to me.

> Do I need to submit a new patch?

I'll need to see if we can remove it, but if we can I'll do that, and
otherwise I can just commit your patch but with a changed commit
message.

Note that I just sent my final pull request for the current kernel, so
this'll probably have to wait some time.

johannes

Re: [PATCH] mac80211_hwsim: set the maximum EIRP output power for 5GHz

[Ramon Fontes]

> If there's some other (physical?) restriction in the driver, sure, maybe
> it should have one there, but for pure regulatory I'm not sure I see it.
> That's why the pointer here to ETSI feels so strange to me.

Ok. I see. You can change the commit msg then.

> Note that I just sent my final pull request for the current kernel, so
> this'll probably have to wait some time.

Ok. No problem.

--
Ramon