Linux wireless drivers development
 help / color / mirror / Atom feed
* [PATCH 01/11] mt76: mt76x02: reduce false positives in ED/CCA tx blocking
@ 2019-03-04  8:02 Felix Fietkau
  2019-03-04  8:02 ` [PATCH 02/11] mt76: mt7603: fix tx status HT rate validation Felix Fietkau
                   ` (9 more replies)
  0 siblings, 10 replies; 11+ messages in thread
From: Felix Fietkau @ 2019-03-04  8:02 UTC (permalink / raw)
  To: linux-wireless

Full tx blocking (as opposed to CCA blocking) should only happen if there
is a continuous non-802.11 signal above the energy detect threshold.
Unfortunately the ED/CCA counter can't detect that, as it also counts 802.11
signals as busy.

Similar to the vendor code, implement a learning mode that waits until the AGC
gain has already been adjusted to the lowest value (due to false CCA events),
and the number of false CCA events still remains high, and the blocking
threshold is exceeded for more than 5 seconds.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt76x02.h  |  3 +++
 .../net/wireless/mediatek/mt76/mt76x02_mac.c  | 25 ++++++++++++++++---
 .../net/wireless/mediatek/mt76/mt76x02_phy.c  |  2 ++
 3 files changed, 26 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02.h b/drivers/net/wireless/mediatek/mt76/mt76x02.h
index 3d93017c3227..0f0fab54e4f1 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76x02.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76x02.h
@@ -51,6 +51,7 @@ struct mt76x02_calibration {
 	u16 false_cca;
 	s8 avg_rssi_all;
 	s8 agc_gain_adjust;
+	s8 agc_lowest_gain;
 	s8 low_gain;
 
 	s8 temp_vco;
@@ -114,8 +115,10 @@ struct mt76x02_dev {
 	struct mt76x02_dfs_pattern_detector dfs_pd;
 
 	/* edcca monitor */
+	unsigned long ed_trigger_timeout;
 	bool ed_tx_blocked;
 	bool ed_monitor;
+	u8 ed_monitor_learning;
 	u8 ed_trigger;
 	u8 ed_silent;
 	ktime_t ed_time;
diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_mac.c b/drivers/net/wireless/mediatek/mt76/mt76x02_mac.c
index e1e0c8da5a8c..9ed231abe916 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76x02_mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76x02_mac.c
@@ -960,6 +960,7 @@ void mt76x02_edcca_init(struct mt76x02_dev *dev, bool enable)
 		}
 	}
 	mt76x02_edcca_tx_enable(dev, true);
+	dev->ed_monitor_learning = true;
 
 	/* clear previous CCA timer value */
 	mt76_rr(dev, MT_ED_CCA_TIMER);
@@ -969,6 +970,10 @@ EXPORT_SYMBOL_GPL(mt76x02_edcca_init);
 
 #define MT_EDCCA_TH		92
 #define MT_EDCCA_BLOCK_TH	2
+#define MT_EDCCA_LEARN_TH	50
+#define MT_EDCCA_LEARN_CCA	180
+#define MT_EDCCA_LEARN_TIMEOUT	(20 * HZ)
+
 static void mt76x02_edcca_check(struct mt76x02_dev *dev)
 {
 	ktime_t cur_time;
@@ -991,11 +996,23 @@ static void mt76x02_edcca_check(struct mt76x02_dev *dev)
 		dev->ed_trigger = 0;
 	}
 
-	if (dev->ed_trigger > MT_EDCCA_BLOCK_TH &&
-	    !dev->ed_tx_blocked)
+	if (dev->cal.agc_lowest_gain &&
+	    dev->cal.false_cca > MT_EDCCA_LEARN_CCA &&
+	    dev->ed_trigger > MT_EDCCA_LEARN_TH) {
+		dev->ed_monitor_learning = false;
+		dev->ed_trigger_timeout = jiffies + 20 * HZ;
+	} else if (!dev->ed_monitor_learning &&
+		   time_is_after_jiffies(dev->ed_trigger_timeout)) {
+		dev->ed_monitor_learning = true;
+		mt76x02_edcca_tx_enable(dev, true);
+	}
+
+	if (dev->ed_monitor_learning)
+		return;
+
+	if (dev->ed_trigger > MT_EDCCA_BLOCK_TH && !dev->ed_tx_blocked)
 		mt76x02_edcca_tx_enable(dev, false);
-	else if (dev->ed_silent > MT_EDCCA_BLOCK_TH &&
-		 dev->ed_tx_blocked)
+	else if (dev->ed_silent > MT_EDCCA_BLOCK_TH && dev->ed_tx_blocked)
 		mt76x02_edcca_tx_enable(dev, true);
 }
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_phy.c b/drivers/net/wireless/mediatek/mt76/mt76x02_phy.c
index a020c757ba5c..a54b63a96eae 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76x02_phy.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76x02_phy.c
@@ -194,6 +194,8 @@ bool mt76x02_phy_adjust_vga_gain(struct mt76x02_dev *dev)
 		ret = true;
 	}
 
+	dev->cal.agc_lowest_gain = dev->cal.agc_gain_adjust >= limit;
+
 	return ret;
 }
 EXPORT_SYMBOL_GPL(mt76x02_phy_adjust_vga_gain);
-- 
2.17.0


^ permalink raw reply related	[flat|nested] 11+ messages in thread
end of thread, other threads:[~2019-03-04  8:02 UTC | newest]

Thread overview: 11+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2019-03-04  8:02 [PATCH 01/11] mt76: mt76x02: reduce false positives in ED/CCA tx blocking Felix Fietkau
2019-03-04  8:02 ` [PATCH 02/11] mt76: mt7603: fix tx status HT rate validation Felix Fietkau
2019-03-04  8:02 ` [PATCH 03/11] mt76: mt76x2: fix external LNA gain settings Felix Fietkau
2019-03-04  8:02 ` [PATCH 04/11] mt76: mt76x2: fix 2.4 GHz channel " Felix Fietkau
2019-03-04  8:02 ` [PATCH 05/11] mt76: mt7603: clear ps filtering mode before releasing buffered frames Felix Fietkau
2019-03-04  8:02 ` [PATCH 06/11] mt76: mt7603: fix up hardware queue index for PS filtered packets Felix Fietkau
2019-03-04  8:02 ` [PATCH 07/11] mt76: mt7603: notify mac80211 about buffered frames in ps queue Felix Fietkau
2019-03-04  8:02 ` [PATCH 08/11] mt76: mt7603: clear the service period on releasing PS filtered packets Felix Fietkau
2019-03-04  8:02 ` [PATCH 09/11] mt76: when releasing PS frames, end the service period if no frame was found Felix Fietkau
2019-03-04  8:02 ` [PATCH 10/11] mt76: mt76x02: disable ED/CCA by default Felix Fietkau
2019-03-04  8:02 ` [PATCH 11/11] mt76: mt7603: set moredata flag when queueing ps-filtered packets Felix Fietkau

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox