Linux wireless drivers development
 help / color / mirror / Atom feed
From: Felix Fietkau <nbd@nbd.name>
To: linux-wireless@vger.kernel.org
Cc: kvalo@codeaurora.org
Subject: [PATCH 4/5] mt76: wait for pending tx to complete before switching channel
Date: Sun, 20 May 2018 07:43:48 +0200	[thread overview]
Message-ID: <20180520054349.49422-4-nbd@nbd.name> (raw)
In-Reply-To: <20180520054349.49422-1-nbd@nbd.name>

Reduces interruption caused by scanning

Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/dma.c      |  4 ++++
 drivers/net/wireless/mediatek/mt76/mac80211.c | 16 ++++++++++++++++
 drivers/net/wireless/mediatek/mt76/mt76.h     |  2 ++
 3 files changed, 22 insertions(+)

diff --git a/drivers/net/wireless/mediatek/mt76/dma.c b/drivers/net/wireless/mediatek/mt76/dma.c
index 3518703524e7..3dbedcedc2c4 100644
--- a/drivers/net/wireless/mediatek/mt76/dma.c
+++ b/drivers/net/wireless/mediatek/mt76/dma.c
@@ -178,6 +178,10 @@ mt76_dma_tx_cleanup(struct mt76_dev *dev, enum mt76_txq_id qid, bool flush)
 		mt76_dma_sync_idx(dev, q);
 
 	wake = wake && qid < IEEE80211_NUM_ACS && q->queued < q->ndesc - 8;
+
+	if (!q->queued)
+		wake_up(&dev->tx_wait);
+
 	spin_unlock_bh(&q->lock);
 
 	if (wake)
diff --git a/drivers/net/wireless/mediatek/mt76/mac80211.c b/drivers/net/wireless/mediatek/mt76/mac80211.c
index d1044e5b25db..fcd079a96782 100644
--- a/drivers/net/wireless/mediatek/mt76/mac80211.c
+++ b/drivers/net/wireless/mediatek/mt76/mac80211.c
@@ -283,6 +283,7 @@ mt76_alloc_device(unsigned int size, const struct ieee80211_ops *ops)
 	spin_lock_init(&dev->rx_lock);
 	spin_lock_init(&dev->lock);
 	spin_lock_init(&dev->cc_lock);
+	init_waitqueue_head(&dev->tx_wait);
 
 	return dev;
 }
@@ -377,18 +378,33 @@ void mt76_rx(struct mt76_dev *dev, enum mt76_rxq_id q, struct sk_buff *skb)
 }
 EXPORT_SYMBOL_GPL(mt76_rx);
 
+static bool mt76_has_tx_pending(struct mt76_dev *dev)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(dev->q_tx); i++) {
+		if (dev->q_tx[i].queued)
+			return true;
+	}
+
+	return false;
+}
+
 void mt76_set_channel(struct mt76_dev *dev)
 {
 	struct ieee80211_hw *hw = dev->hw;
 	struct cfg80211_chan_def *chandef = &hw->conf.chandef;
 	struct mt76_channel_state *state;
 	bool offchannel = hw->conf.flags & IEEE80211_CONF_OFFCHANNEL;
+	int timeout = HZ / 5;
 
 	if (offchannel)
 		set_bit(MT76_OFFCHANNEL, &dev->state);
 	else
 		clear_bit(MT76_OFFCHANNEL, &dev->state);
 
+	wait_event_timeout(dev->tx_wait, !mt76_has_tx_pending(dev), timeout);
+
 	if (dev->drv->update_survey)
 		dev->drv->update_survey(dev);
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt76.h b/drivers/net/wireless/mediatek/mt76/mt76.h
index bb158f867d7c..d2166fbf50ff 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76.h
@@ -251,6 +251,8 @@ struct mt76_dev {
 	struct mt76_queue q_rx[__MT_RXQ_MAX];
 	const struct mt76_queue_ops *queue_ops;
 
+	wait_queue_head_t tx_wait;
+
 	u8 macaddr[ETH_ALEN];
 	u32 rev;
 	unsigned long state;
-- 
2.17.0

  parent reply	other threads:[~2018-05-20  5:43 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-05-20  5:43 [PATCH 1/5] mt76: only stop tx queues on offchannel, not during the entire scan Felix Fietkau
2018-05-20  5:43 ` [PATCH 2/5] mt76: prevent tx scheduling during channel change Felix Fietkau
2018-05-20  5:43 ` [PATCH 3/5] mt76: move ieee80211_hw allocation to common core Felix Fietkau
2018-05-20  5:43 ` Felix Fietkau [this message]
2018-05-20  5:43 ` [PATCH 5/5] mt76: use udelay instead of usleep_range in mt76x2_mac_stop Felix Fietkau
2018-05-23  7:59 ` [1/5] mt76: only stop tx queues on offchannel, not during the entire scan Kalle Valo

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20180520054349.49422-4-nbd@nbd.name \
    --to=nbd@nbd.name \
    --cc=kvalo@codeaurora.org \
    --cc=linux-wireless@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox