Received-SPF: pass (google.com: domain of linux-wireless-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <9145cfd7-5c7c-da58-7453-3b8fd47bf76f@linaro.org>
Date:   Wed, 14 Sep 2022 21:38:20 +0100
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101
 Thunderbird/91.11.0
Subject: Re: [PATCH] wcn36xx: Add RX frame SNR as a source of system entropy
Content-Language: en-US
To:     Loic Poulain <loic.poulain@linaro.org>
Cc:     kvalo@kernel.org, davem@davemloft.net, edumazet@google.com,
        kuba@kernel.org, pabeni@redhat.com, wcn36xx@lists.infradead.org,
        linux-wireless@vger.kernel.org, netdev@vger.kernel.org,
        "Jason A . Donenfeld" <Jason@zx2c4.com>
References: <20220913174224.1399480-1-bryan.odonoghue@linaro.org>
 <CAMZdPi_jZEHSz9+WaR3N-MANU7YdYK0zqO7VXNwAcES=YkNaxg@mail.gmail.com>
From:   Bryan O'Donoghue <bryan.odonoghue@linaro.org>
In-Reply-To: <CAMZdPi_jZEHSz9+WaR3N-MANU7YdYK0zqO7VXNwAcES=YkNaxg@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 14/09/2022 13:53, Loic Poulain wrote:
> Hi Bryan,
> 
> On Tue, 13 Sept 2022 at 19:42, Bryan O'Donoghue
> <bryan.odonoghue@linaro.org> wrote:
>>
>> The signal-to-noise-ratio of a received frame is a representation of noise
>> in a given received frame.
>>
>> RSSI - received signal strength indication can appear pretty static
>> frame-to-frame but noise will "bounce around" more depending on the EM
>> environment, temperature or placement of obstacles between the transmitter
>> and receiver.
>>
>> Other WiFi drivers offer up the noise component of the FFT as an entropy
>> source for the random pool i.e.
>>
>> Commit: 2aa56cca3571 ("ath9k: Mix the received FFT bins to the random pool")
>>
>> I attended Jason's talk on sources of randomness at Plumbers and it occured
>> to me that SNR is a reasonable candidate to add.
>>
>> Cc: Jason A. Donenfeld <Jason@zx2c4.com>
>> Signed-off-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
>> ---
>>   drivers/net/wireless/ath/wcn36xx/txrx.c | 3 +++
>>   1 file changed, 3 insertions(+)
>>
>> diff --git a/drivers/net/wireless/ath/wcn36xx/txrx.c b/drivers/net/wireless/ath/wcn36xx/txrx.c
>> index 8da3955995b6e..f3b77d7ffebe4 100644
>> --- a/drivers/net/wireless/ath/wcn36xx/txrx.c
>> +++ b/drivers/net/wireless/ath/wcn36xx/txrx.c
>> @@ -16,6 +16,7 @@
>>
>>   #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
>>
>> +#include <linux/random.h>
>>   #include "txrx.h"
>>
>>   static inline int get_rssi0(struct wcn36xx_rx_bd *bd)
>> @@ -297,6 +298,8 @@ static void wcn36xx_update_survey(struct wcn36xx *wcn, int rssi, int snr,
>>          wcn->chan_survey[idx].rssi = rssi;
>>          wcn->chan_survey[idx].snr = snr;
>>          spin_unlock(&wcn->survey_lock);
>> +
>> +       add_device_randomness(&snr, sizeof(int));
> 
> We store the SNR in an integer, but isn't it reported as u8 (or s8) by
> the firmware? So maybe we should just inject the LSByte since the upper
> ones will always be 0?

sizeof(s8) makes sense to me, also I can massage the commit text a bit, 
it doesn't scan well on a second reading

---
bod