Am 26.01.24 um 03:41 schrieb Ping-Ke Shih:
-----Original Message-----
From: Martin Kaistra <martin.kaistra@xxxxxxxxxxxxx>
Sent: Wednesday, January 24, 2024 4:27 PM
To: linux-wireless@xxxxxxxxxxxxxxx
Cc: Jes Sorensen <Jes.Sorensen@xxxxxxxxx>; Kalle Valo <kvalo@xxxxxxxxxx>; Ping-Ke Shih
<pkshih@xxxxxxxxxxx>; Bitterblue Smith <rtl8821cerfe2@xxxxxxxxx>; Sebastian Andrzej Siewior
<bigeasy@xxxxxxxxxxxxx>
Subject: [PATCH v2] wifi: rtl8xxxu: update rate mask per sta
Until now, rtl8xxxu_watchdog_callback() only fetches RSSI and updates
the rate mask in station mode. This means, in AP mode only the default
rate mask is used.
In order to have the rate mask reflect the actual connection quality,
extend rtl8xxxu_watchdog_callback() to iterate over every sta. Like in
the rtw88 driver, add a function to collect all currently present stas
and then iterate over a list of copies to ensure no RCU lock problems
for register access via USB. Remove the existing RCU lock in
rtl8xxxu_refresh_rate_mask().
Since the currently used ieee80211_ave_rssi() is only for 'vif', add
driver-level tracking of RSSI per sta.
Signed-off-by: Martin Kaistra <martin.kaistra@xxxxxxxxxxxxx>
---
changes v1->v2: move 'rssi_level' into struct rtl8xxxu_sta_info
v1: https://lore.kernel.org/linux-wireless/20240117145516.497966-1-martin.kaistra@xxxxxxxxxxxxx/
[...]
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
index 3b954c2fe448f..3820d3c308759 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c
@@ -4993,8 +4993,8 @@ rtl8xxxu_bss_info_changed(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
struct device *dev = &priv->udev->dev;
struct ieee80211_sta *sta;
struct rtl8xxxu_ra_report *rarpt;
+ u8 val8, macid;
u32 val32;
- u8 val8;
rarpt = &priv->ra_report;
@@ -5004,6 +5004,7 @@ rtl8xxxu_bss_info_changed(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
rtl8xxxu_set_linktype(priv, vif->type, rtlvif->port_num);
if (vif->cfg.assoc) {
+ struct rtl8xxxu_sta_info *sta_info;
nit: I remember that declaration at beginning of function is preferred.
I will move it to the beginning of the function.
u32 ramask;
int sgi = 0;
u8 highest_rate;
@@ -5017,6 +5018,7 @@ rtl8xxxu_bss_info_changed(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
rcu_read_unlock();
goto error;
}
+ macid = rtl8xxxu_get_macid(priv, sta);
if (sta->deflink.ht_cap.ht_supported)
dev_info(dev, "%s: HT supported\n", __func__);
@@ -5037,14 +5039,15 @@ rtl8xxxu_bss_info_changed(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
bw = RATE_INFO_BW_40;
else
bw = RATE_INFO_BW_20;
+
+ sta_info = (struct rtl8xxxu_sta_info *)sta->drv_priv;
+ sta_info->rssi_level = RTL8XXXU_RATR_STA_INIT;
For AP mode, we should do this as well at rtl8xxxu_sta_add() before calling
rtl8xxxu_refresh_rate_mask()?
refresh_rate_mask is called from sta_add with force == true, so a new value is
set there to sta_info->rssi_level regardless of the previous value, but it can't
hurt to have a proper initialisation.
rcu_read_unlock();
rtl8xxxu_update_ra_report(rarpt, highest_rate, sgi, bw);
- priv->rssi_level = RTL8XXXU_RATR_STA_INIT;
-
priv->fops->update_rate_mask(priv, ramask, 0, sgi,
- bw == RATE_INFO_BW_40, 0);
+ bw == RATE_INFO_BW_40, macid);
rtl8xxxu_write8(priv, REG_BCN_MAX_ERR, 0xff);
@@ -6317,6 +6320,76 @@ static void rtl8188e_c2hcmd_callback(struct work_struct *work)
}
}
+#define rtl8xxxu_iterate_vifs_atomic(priv, iterator, data) \
+ ieee80211_iterate_active_interfaces_atomic((priv)->hw, \
+ IEEE80211_IFACE_ITER_NORMAL, iterator, data)
+
+struct rtl8xxxu_rx_addr_match_data {
+ struct rtl8xxxu_priv *priv;
+ struct ieee80211_hdr *hdr;
+ struct ieee80211_rx_status *rx_status;
+ u8 *bssid;
+};
+
+static void rtl8xxxu_rx_addr_match_iter(void *data, u8 *mac,
+ struct ieee80211_vif *vif)
+{
+ struct rtl8xxxu_rx_addr_match_data *iter_data = data;
+ struct ieee80211_sta *sta;
+ struct ieee80211_hdr *hdr = iter_data->hdr;
+ struct rtl8xxxu_priv *priv = iter_data->priv;
+ struct rtl8xxxu_sta_info *sta_info;
+ struct ieee80211_rx_status *rx_status = iter_data->rx_status;
+ u8 *bssid = iter_data->bssid;
+
+ if (!ether_addr_equal(vif->bss_conf.bssid, bssid))
+ return;
+
+ if (!(ether_addr_equal(vif->addr, hdr->addr1) ||
+ ieee80211_is_beacon(hdr->frame_control)))
+ return;
nit: Here checks bssid, addr1 and beacon frame. For me, if you want to combine
some of them, would it be reasonable to combine bssid and add1?
Ok, I will do that for v3.
+
+ sta = ieee80211_find_sta_by_ifaddr(priv->hw, hdr->addr2,
+ vif->addr);
+ if (!sta)
+ return;
+
+ sta_info = (struct rtl8xxxu_sta_info *)sta->drv_priv;
+ ewma_rssi_add(&sta_info->avg_rssi, -rx_status->signal);
+}
+
+static inline u8 *get_hdr_bssid(struct ieee80211_hdr *hdr)
Would you like to add a ieee80211_get_BSSID() to ieee80211.h in separated
patch? But I wonder if it is enough to check addr1 only?
I just saw that there already is a ieee80211_get_bssid() in net/mac80211/util.c.
Only problem is that it needs enum nl80211_iftype type as parameter and I
currently do not have that available in rtl8xxxu_parse_rxdesc16()..
+{
+ __le16 fc = hdr->frame_control;
+ u8 *bssid;
+
+ if (ieee80211_has_tods(fc))
+ bssid = hdr->addr1;
+ else if (ieee80211_has_fromds(fc))
+ bssid = hdr->addr2;
+ else
+ bssid = hdr->addr3;
+
+ return bssid;
+}
+
+static void rtl8xxxu_rx_addr_match(struct rtl8xxxu_priv *priv,
+ struct ieee80211_rx_status *rx_status,
+ struct ieee80211_hdr *hdr)
+{
+ struct rtl8xxxu_rx_addr_match_data data = {};
+
+ if (ieee80211_is_ctl(hdr->frame_control))
+ return;
+
+ data.priv = priv;
+ data.hdr = hdr;
+ data.rx_status = rx_status;
+ data.bssid = get_hdr_bssid(hdr);
+
+ rtl8xxxu_iterate_vifs_atomic(priv, rtl8xxxu_rx_addr_match_iter, &data);
+}
+
int rtl8xxxu_parse_rxdesc16(struct rtl8xxxu_priv *priv, struct sk_buff *skb)
{
struct ieee80211_hw *hw = priv->hw;
@@ -6376,18 +6449,26 @@ int rtl8xxxu_parse_rxdesc16(struct rtl8xxxu_priv *priv, struct sk_buff *skb)
skb_queue_tail(&priv->c2hcmd_queue, skb);
schedule_work(&priv->c2hcmd_work);
} else {
+ struct ieee80211_hdr *hdr;
+
phy_stats = (struct rtl8723au_phy_stats *)skb->data;
skb_pull(skb, drvinfo_sz + desc_shift);
skb_trim(skb, pkt_len);
- if (rx_desc->phy_stats)
+ hdr = (struct ieee80211_hdr *)skb->data;
+ if (rx_desc->phy_stats) {
priv->fops->parse_phystats(
priv, rx_status, phy_stats,
rx_desc->rxmcs,
- (struct ieee80211_hdr *)skb->data,
+ hdr,
rx_desc->crc32 || rx_desc->icverr);
+ if (!rx_desc->crc32 && !rx_desc->icverr)
+ rtl8xxxu_rx_addr_match(priv,
+ rx_status,
+ hdr);
This function name isn't clear, because it doesn't just match rx addr,
but it is to update RSSI by rx_status for corresponding station.
Does rtl8xxxu_rx_update_rssi() sound better to you?
+ }
rx_status->mactime = rx_desc->tsfl;
rx_status->flag |= RX_FLAG_MACTIME_START;
[...]
