Received-SPF: pass (google.com: domain of linux-wireless+bounces-3409-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Message-ID: <d3f6cdef5b05fba8eadedf67129c9c49471a08f0.camel@sipsolutions.net>
Subject: Re: [RFC] mac80211: add AQL support for broadcast packets
From: Johannes Berg <johannes@sipsolutions.net>
To: Felix Fietkau <nbd@nbd.name>, linux-wireless@vger.kernel.org
Date: Fri, 09 Feb 2024 20:32:42 +0100
In-Reply-To: <20240209184730.69589-1-nbd@nbd.name>
References: <20240209184730.69589-1-nbd@nbd.name>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
User-Agent: Evolution 3.50.3 (3.50.3-1.fc39) 
Precedence: bulk
MIME-Version: 1.0


> -	if (sscanf(buf, "%u %u %u", &ac, &q_limit_low, &q_limit_high) !=3D 3)
> +	n =3D sscanf(buf, "%u %u %u", &ac, &q_limit_low, &q_limit_high);
> +	if (n < 2)
>  		return -EINVAL;
> =20
> -	if (ac >=3D IEEE80211_NUM_ACS)
> +	if (ac =3D=3D IEEE80211_NUM_ACS) {
> +		local->aql_txq_limit_bc =3D q_limit_low;
> +		return count;
> +	}

If we keep this, it should probably check n =3D=3D 2?

But not sure I like it - in other places ac =3D=3D NUM_ACS means management
rather than not multicast; could we just check something like

	strncmp(buf, "mcast ", 6)

instead? Or "BC/MC" matching the output? Though we don't match VO on
input either.

johannes