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;
From:   Toke =?utf-8?Q?H=C3=B8iland-J=C3=B8rgensen?= <toke@redhat.com>
To:     Felix Fietkau <nbd@nbd.name>, linux-wireless@vger.kernel.org
Subject: Re: [PATCH 05/15] mt76: track rx airtime for airtime fairness and survey
In-Reply-To: <6af9b976-48e8-64c2-dffe-b3eaa03b9c73@nbd.name>
References: <20190926174732.42375-1-nbd@nbd.name> <20190926174732.42375-5-nbd@nbd.name> <87v9teyw40.fsf@toke.dk> <6af9b976-48e8-64c2-dffe-b3eaa03b9c73@nbd.name>
Date:   Fri, 27 Sep 2019 11:04:21 +0200
Message-ID: <87pnjmyusq.fsf@toke.dk>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Sender: linux-wireless-owner@vger.kernel.org
Precedence: bulk

Felix Fietkau <nbd@nbd.name> writes:

> On 2019-09-27 10:35, Toke H=C3=B8iland-J=C3=B8rgensen wrote:
>> Felix Fietkau <nbd@nbd.name> writes:
>>> +=09switch (status->encoding) {
>>> +=09case RX_ENC_LEGACY:
>>> +=09=09if (WARN_ON_ONCE(status->band > NL80211_BAND_5GHZ))
>>> +=09=09=09return 0;
>>> +
>>> +=09=09sband =3D dev->hw->wiphy->bands[status->band];
>>> +=09=09if (!sband || status->rate_idx > sband->n_bitrates)
>>> +=09=09=09return 0;
>>> +
>>> +=09=09rate =3D &sband->bitrates[status->rate_idx];
>>> +
>>> +=09=09return mt76_calc_legacy_rate_duration(rate, sp, len);
>>> +=09case RX_ENC_VHT:
>>> +=09=09streams =3D status->nss;
>>> +=09=09idx =3D status->rate_idx;
>>> +=09=09group =3D VHT_GROUP_IDX(streams, sgi, bw);
>>> +=09=09break;
>>> +=09case RX_ENC_HT:
>>> +=09=09streams =3D ((status->rate_idx >> 3) & 3) + 1;
>>> +=09=09idx =3D status->rate_idx & 7;
>>> +=09=09group =3D HT_GROUP_IDX(streams, sgi, bw);
>>> +=09=09break;
>>> +=09default:
>>> +=09=09WARN_ON_ONCE(1);
>>> +=09=09return 0;
>>> +=09}
>>> +
>>> +=09if (WARN_ON_ONCE(streams > 4))
>>> +=09=09return 0;
>>> +
>>> +=09duration =3D airtime_mcs_groups[group].duration[idx];
>>> +=09duration <<=3D airtime_mcs_groups[group].shift;
>>> +=09duration *=3D len;
>>> +=09duration /=3D AVG_PKT_SIZE;
>>> +=09duration /=3D 1024;
>>=20
>> On an earlier patch of mine you expressed concern over divisions in the
>> fast path. Does this mean this is no longer a concern? Or is the
>> compiler doing fancy things with the constant division here? :)
> I made the constants power-of-2, so that the compiler can turn them into
> bit shifts. I verified the disassembly: no divisions left in the code.

I see. Neat :)

There's still a regular division in the legacy rate calc further up,
though, right? But I guess that doesn't happen so often?

-Toke