| Message ID | 20231222101442.626837-14-martin.kaistra@linutronix.de |
|---|---|
| State | New |
| Headers | show |
| Series | wifi: rtl8xxxu: Add concurrent mode for 8188f | expand |
On Fri, Dec 22, 2023 at 6:16 PM Martin Kaistra <martin.kaistra@linutronix.de> wrote: > > Check first whether priv->vifs[0] exists and is of type STATION, then go > to priv->vifs[1]. Make sure to call refresh_rate_mask for both > interfaces. > > Signed-off-by: Martin Kaistra <martin.kaistra@linutronix.de> Reviewed-by: Ping-Ke Shih <pkshih@realtek.com> Logically, this patch is fine to me, but I would like to know if you have tested two STATION interfaces connect two different AP? Especially, one AP is 802.11g mode and the other is 802.11n mode, so we can clearly know this patch is expected. Ping-Ke
Am 25.12.23 um 06:22 schrieb Ping-Ke Shih: > > On Fri, Dec 22, 2023 at 6:16 PM Martin Kaistra > <martin.kaistra@linutronix.de> wrote: >> >> Check first whether priv->vifs[0] exists and is of type STATION, then go >> to priv->vifs[1]. Make sure to call refresh_rate_mask for both >> interfaces. >> >> Signed-off-by: Martin Kaistra <martin.kaistra@linutronix.de> > > Reviewed-by: Ping-Ke Shih <pkshih@realtek.com> > > Logically, this patch is fine to me, but I would like to know if you have > tested two STATION interfaces connect two different AP? Especially, one AP > is 802.11g mode and the other is 802.11n mode, so we can clearly know this > patch is expected. Yes, I did try to connect the two interfaces to two different AP (on the same channel) at the same time, and it seems to behave fine.
> -----Original Message----- > From: Martin Kaistra <martin.kaistra@linutronix.de> > Sent: Monday, January 8, 2024 5:43 PM > To: Ping-Ke Shih <pkshih@realtek.com>; linux-wireless@vger.kernel.org > Cc: Jes Sorensen <Jes.Sorensen@gmail.com>; Kalle Valo <kvalo@kernel.org>; Bitterblue Smith > <rtl8821cerfe2@gmail.com>; Sebastian Andrzej Siewior <bigeasy@linutronix.de> > Subject: Re: [PATCH v3 13/21] wifi: rtl8xxxu: support multiple interfaces in watchdog_callback() > > Yes, I did try to connect the two interfaces to two different AP (on the same > channel) at the same time, and it seems to behave fine. Nice! Thanks for your work. :-) Ping-Ke
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c index c5b71892369c9..fd0108668bcda 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c @@ -7200,11 +7200,15 @@ static void rtl8xxxu_watchdog_callback(struct work_struct *work) { struct ieee80211_vif *vif; struct rtl8xxxu_priv *priv; + int i; priv = container_of(work, struct rtl8xxxu_priv, ra_watchdog.work); - vif = priv->vif; + for (i = 0; i < ARRAY_SIZE(priv->vifs); i++) { + vif = priv->vifs[i]; + + if (!vif || vif->type != NL80211_IFTYPE_STATION) + continue; - if (vif && vif->type == NL80211_IFTYPE_STATION) { int signal; struct ieee80211_sta *sta; @@ -7215,22 +7219,21 @@ static void rtl8xxxu_watchdog_callback(struct work_struct *work) dev_dbg(dev, "%s: no sta found\n", __func__); rcu_read_unlock(); - goto out; + continue; } rcu_read_unlock(); signal = ieee80211_ave_rssi(vif); - priv->fops->report_rssi(priv, 0, + priv->fops->report_rssi(priv, rtl8xxxu_get_macid(priv, sta), rtl8xxxu_signal_to_snr(signal)); - if (priv->fops->set_crystal_cap) - rtl8xxxu_track_cfo(priv); - rtl8xxxu_refresh_rate_mask(priv, signal, sta, false); } -out: + if (priv->fops->set_crystal_cap) + rtl8xxxu_track_cfo(priv); + schedule_delayed_work(&priv->ra_watchdog, 2 * HZ); }
Check first whether priv->vifs[0] exists and is of type STATION, then go to priv->vifs[1]. Make sure to call refresh_rate_mask for both interfaces. Signed-off-by: Martin Kaistra <martin.kaistra@linutronix.de> --- .../wireless/realtek/rtl8xxxu/rtl8xxxu_core.c | 19 +++++++++++-------- 1 file changed, 11 insertions(+), 8 deletions(-)