Received-SPF: pass (google.com: domain of linux-wireless-owner@vger.kernel.org designates 23.128.96.37 as permitted sender) client-ip=23.128.96.37;
Message-ID: <75b4b34f9e49e1af2fac5fdf8a1d022f8b0b2d5f.camel@sipsolutions.net>
Subject: Re: [PATCH] wifi: ath11k: mac: fix struct
 ieee80211_sband_iftype_data handling
From:   Johannes Berg <johannes@sipsolutions.net>
To:     Kalle Valo <kvalo@kernel.org>, ath11k@lists.infradead.org
Cc:     linux-wireless@vger.kernel.org
Date:   Sat, 23 Sep 2023 22:08:56 +0200
In-Reply-To: <87ediqa674.fsf@kernel.org>
References: <20230921075440.1539515-1-kvalo@kernel.org>
         <87ediqa674.fsf@kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
User-Agent: Evolution 3.48.4 (3.48.4-1.fc38) 
MIME-Version: 1.0
Precedence: bulk

On Fri, 2023-09-22 at 12:49 +0300, Kalle Valo wrote:
>=20
> > +		if (iftd->types_mask & BIT(arvif->vif->type)) {
>=20
> Johannes pointed out that I should use ieee80211_get_he_iftype_cap_vif()
> instead. I'll submit v2.

And I'll note that it's not just because it saves open-coding it, also
the BIT(vif->type) in mac80211 will be wrong because it's always STATION
even for P2P-Client, so you need to add the vif->p2p bit in.

We maybe shouldn't have done that ... but it seemed kinda nicer at the
time to treat them both the same except in special cases.

johannes