Received-SPF: pass (google.com: best guess record for domain of linux-wireless-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
DMARC-Filter: OpenDMARC Filter v1.3.2 smtp.codeaurora.org 7EB7F615E8
From:   Kalle Valo <kvalo@codeaurora.org>
To:     Ryder Lee <ryder.lee@mediatek.com>
Cc:     Lorenzo Bianconi <lorenzo.bianconi@redhat.com>,
        Lorenzo Bianconi <lorenzo@kernel.org>,
        Felix Fietkau <nbd@nbd.name>,
        linux-wireless <linux-wireless@vger.kernel.org>,
        Roy Luo <royluo@google.com>, YF Luo <yf.luo@mediatek.com>,
        <linux-mediatek@lists.infradead.org>
Subject: Re: [PATCH 4/6] mt76: mt7615: unlock dfs bands
References: <cover.1561804422.git.lorenzo@kernel.org>
        <33184e0b78983fe7c79fa70c5fbb21042aafa4f5.1561804422.git.lorenzo@kernel.org>
        <87muhzs9qv.fsf@purkki.adurom.net>
        <CAJ0CqmU6TLhFa4ZJxWHBzvpx+5g5E4-WkSPECx47F9d3T=5YjQ@mail.gmail.com>
        <87tvc69odh.fsf@kamboji.qca.qualcomm.com>
        <1561975021.3950.9.camel@mtkswgap22>
Date:   Mon, 01 Jul 2019 13:49:52 +0300
In-Reply-To: <1561975021.3950.9.camel@mtkswgap22> (Ryder Lee's message of
        "Mon, 1 Jul 2019 17:57:01 +0800")
Message-ID: <87pnmu9ghb.fsf@kamboji.qca.qualcomm.com>
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/24.5 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain
Sender: linux-wireless-owner@vger.kernel.org
Precedence: bulk

Ryder Lee <ryder.lee@mediatek.com> writes:

> On Mon, 2019-07-01 at 10:59 +0300, Kalle Valo wrote:
>> Lorenzo Bianconi <lorenzo.bianconi@redhat.com> writes:
>> 
>> >>
>> >> Lorenzo Bianconi <lorenzo@kernel.org> writes:
>> >>
>> >> > Unlock dfs channels since now mt7615 driver supports radar detection
>> >> >
>> >> > Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
>> >> > ---
>> >> >  drivers/net/wireless/mediatek/mt76/mt7615/init.c | 6 ++++++
>> >> >  1 file changed, 6 insertions(+)
>> >> >
>> >> > diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/init.c b/drivers/net/wireless/mediatek/mt76/mt7615/init.c
>> >> > index 5dc4cced5789..6d336d82cafe 100644
>> >> > --- a/drivers/net/wireless/mediatek/mt76/mt7615/init.c
>> >> > +++ b/drivers/net/wireless/mediatek/mt76/mt7615/init.c
>> >> > @@ -152,6 +152,12 @@ static const struct ieee80211_iface_combination if_comb[] = {
>> >> >               .max_interfaces = 4,
>> >> >               .num_different_channels = 1,
>> >> >               .beacon_int_infra_match = true,
>> >> > +             .radar_detect_widths = BIT(NL80211_CHAN_WIDTH_20_NOHT) |
>> >> > +                                    BIT(NL80211_CHAN_WIDTH_20) |
>> >> > +                                    BIT(NL80211_CHAN_WIDTH_40) |
>> >> > +                                    BIT(NL80211_CHAN_WIDTH_80) |
>> >> > +                                    BIT(NL80211_CHAN_WIDTH_160) |
>> >> > +                                    BIT(NL80211_CHAN_WIDTH_80P80),
>> >>
>> >> Isn't it questionable to enable these without any testing on real
>> >> hardware? Getting DFS to work correctly is hard so I'm very suspicious
>> >> about this.
>> >>
>> >> --
>> >> Kalle Valo
>> >
>> > Hi Kalle,
>> >
>> > unfortunately at the moment I am not able to run any tests with a real
>> > signal generator so I just ported the code from vendor sdk.
>> > I am pretty confident it works since the radar pattern detection is
>> > done in fw/hw so I guess it has been already tested in the vendor sdk
>> 
>> DFS is really tricky to get it working right, so I'm not easily
>> convinced :)
>> 
>> > but we can postpone this patch and apply just the rest of the series
>> > until we have some test results.
>> 
>> Yeah, I think it would be best to drop this patch so that DFS is not
>> enabled by default and apply this patch only after positive test
>> results.
>> 
>
> That's why I suggested Lorenzo to add this one - "[6/6] mt76: mt7615:
> add radar pattern test knob to debugfs"
>
> We can feed radar pattern through debugfs to test if a pattern is
> detected as radar pattern or not and verify the fw radar detection
> algorithm.

Sure, that's nice for testing but does not guarantee that it works with
real hardware as well. I take regulatory rules very seriously and that's
why I'm extra careful here.

-- 
Kalle Valo