Linux wireless drivers development
 help / color / mirror / Atom feed
From: Felix Fietkau <nbd@nbd.name>
To: linux-wireless@vger.kernel.org
Subject: [PATCH 2/4] mt76: mt76x02: remove irqsave/restore in locking for tx status fifo
Date: Mon, 29 Apr 2019 19:17:10 +0200	[thread overview]
Message-ID: <20190429171712.52983-2-nbd@nbd.name> (raw)
In-Reply-To: <20190429171712.52983-1-nbd@nbd.name>

Use a separate lock and spin_trylock to avoid disabling interrupts.
Should improve performance and latency

Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt76x02.h      | 1 +
 drivers/net/wireless/mediatek/mt76/mt76x02_mac.c  | 7 ++++---
 drivers/net/wireless/mediatek/mt76/mt76x02_mmio.c | 1 +
 3 files changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02.h b/drivers/net/wireless/mediatek/mt76/mt76x02.h
index cd37f44580ba..9f103c2506db 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76x02.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76x02.h
@@ -86,6 +86,7 @@ struct mt76x02_dev {
 
 	u8 txdone_seq;
 	DECLARE_KFIFO_PTR(txstatus_fifo, struct mt76x02_tx_status);
+	spinlock_t txstatus_fifo_lock;
 
 	struct sk_buff *rx_head;
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_mac.c b/drivers/net/wireless/mediatek/mt76/mt76x02_mac.c
index 28851060aa0f..c1f041e1a279 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76x02_mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76x02_mac.c
@@ -739,7 +739,6 @@ int mt76x02_mac_process_rx(struct mt76x02_dev *dev, struct sk_buff *skb,
 void mt76x02_mac_poll_tx_status(struct mt76x02_dev *dev, bool irq)
 {
 	struct mt76x02_tx_status stat = {};
-	unsigned long flags;
 	u8 update = 1;
 	bool ret;
 
@@ -749,9 +748,11 @@ void mt76x02_mac_poll_tx_status(struct mt76x02_dev *dev, bool irq)
 	trace_mac_txstat_poll(dev);
 
 	while (!irq || !kfifo_is_full(&dev->txstatus_fifo)) {
-		spin_lock_irqsave(&dev->mt76.mmio.irq_lock, flags);
+		if (!spin_trylock(&dev->txstatus_fifo_lock))
+			break;
+
 		ret = mt76x02_mac_load_tx_status(dev, &stat);
-		spin_unlock_irqrestore(&dev->mt76.mmio.irq_lock, flags);
+		spin_unlock(&dev->txstatus_fifo_lock);
 
 		if (!ret)
 			break;
diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_mmio.c b/drivers/net/wireless/mediatek/mt76/mt76x02_mmio.c
index 87e14af7a93b..5bc1b901f897 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76x02_mmio.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76x02_mmio.c
@@ -201,6 +201,7 @@ int mt76x02_dma_init(struct mt76x02_dev *dev)
 	tasklet_init(&dev->pre_tbtt_tasklet, mt76x02_pre_tbtt_tasklet,
 		     (unsigned long)dev);
 
+	spin_lock_init(&dev->txstatus_fifo_lock);
 	kfifo_init(&dev->txstatus_fifo, status_fifo, fifo_size);
 
 	mt76_dma_attach(&dev->mt76);
-- 
2.17.0


  reply	other threads:[~2019-04-29 17:17 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-04-29 17:17 [PATCH 1/4] mt76: mt76x02: use napi polling for tx cleanup Felix Fietkau
2019-04-29 17:17 ` Felix Fietkau [this message]
2019-04-29 17:17 ` [PATCH 3/4] mt76: mt7603: fix initialization of max rx length Felix Fietkau
2019-04-29 17:17 ` [PATCH 4/4] mt76: mt7615: use sizeof instead of sizeof_field Felix Fietkau

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=20190429171712.52983-2-nbd@nbd.name \
    --to=nbd@nbd.name \
    --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