Subject: Re: [PATCH] mt76: mt7915: add mu-mimo and ofdma debugfs knobs
To:     meichia chiu <meichia.chiu@mediatek.com>,
        Felix Fietkau <nbd@nbd.name>
Cc:     Lorenzo Bianconi <lorenzo.bianconi@redhat.com>,
        Shayne Chen <shayne.chen@mediatek.com>,
        Evelyn Tsai <evelyn.tsai@mediatek.com>,
        Ryder Lee <ryder.lee@mediatek.com>,
        Money Wang <Money.Wang@mediatek.com>,
        linux-wireless@vger.kernel.org, linux-mediatek@lists.infradead.org
References: <20211206141006.29632-1-MeiChia.Chiu@mediatek.com>
 <7158a937-57f3-2bd8-591e-124a18d2d8e8@candelatech.com>
 <72ca4f5fe3c351ec56dee4935b28c7b7f26188c2.camel@mediatek.com>
From:   Ben Greear <greearb@candelatech.com>
Organization: Candela Technologies
Message-ID: <e4a8301c-53a1-9ca0-7e1d-fd235a22ccd6@candelatech.com>
Date:   Tue, 7 Dec 2021 07:33:25 -0800
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.10.1
MIME-Version: 1.0
In-Reply-To: <72ca4f5fe3c351ec56dee4935b28c7b7f26188c2.camel@mediatek.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-MW
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 12/6/21 11:25 PM, meichia chiu wrote:
> On Mon, 2021-12-06 at 08:43 -0800, Ben Greear wrote:
>> On 12/6/21 6:10 AM, MeiChia Chiu wrote:
>>> From: MeiChia Chiu <meichia.chiu@mediatek.com>
>>>
>>> Add mu-mimo and ofdma packet counters statistics.
>>
>> Hello,
>>
>> Are these stats clear-on-read in the firmware?  If so, then we need
>> to accumulate
>> them in the driver.
>>
>> Is there any significant downside to enabling the stats?  For
>> instance, a performance
>> degradation in the firmware/hardware?
>>
>> Thanks,
>> Ben
> 
> Hi Ben,
> 
> Yes, the current design is clear-on-read and its sole purpose is for
> debugging only. If you need statistics, this may have to be done in
> another way.

So we could poll these stats in the other logic that polls firmware stats and accumulate
them in the driver and display with debugfs and/or ethtool.  If you don't want
to do the accumulation, and lease add a comment to your patch about the
values being clear-on-read and someone else can add the accumulation logic
later.

> Yes, a performance degradation is expected since enabling this debug
> function would burden MCU with more calculation.

Have you done any testing to verify this?  If it is just bumping a few
counters, it may not matter.  But I don't know how the firmware is actually
implementing this, so maybe it is a lot more expensive than expected?

Thanks,
Ben

> 
> Thanks,
> Meichia
> 


-- 
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc  http://www.candelatech.com