All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ron Angeles <ronangeles@gmail.com>
To: netdev@vger.kernel.org
Subject: [PATCH net-next] net: atl1c: add BQL support
Date: Tue, 11 Aug 2015 23:01:20 -0700	[thread overview]
Message-ID: <55CAE130.8040406@gmail.com> (raw)

This BQL implementation is mostly derived from its related driver, alx.
Tested on AR8131 (rev c0) [1969:1063]. Saturated a 100mbps link with 5
concurrent runs of netperf. Ping latency dropped from 14ms to 3ms.

Signed-off-by: Ron Angeles <ronangeles@gmail.com>
---
 drivers/net/ethernet/atheros/atl1c/atl1c_main.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
index 932bd18..2795d6d 100644
--- a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
+++ b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
@@ -874,6 +874,8 @@ static void atl1c_clean_tx_ring(struct atl1c_adapter *adapter,
 		atl1c_clean_buffer(pdev, buffer_info);
 	}
 
+	netdev_reset_queue(adapter->netdev);
+
 	/* Zero out Tx-buffers */
 	memset(tpd_ring->desc, 0, sizeof(struct atl1c_tpd_desc) *
 		ring_count);
@@ -1551,6 +1553,7 @@ static bool atl1c_clean_tx_irq(struct atl1c_adapter *adapter,
 	u16 next_to_clean = atomic_read(&tpd_ring->next_to_clean);
 	u16 hw_next_to_clean;
 	u16 reg;
+	unsigned int total_bytes = 0, total_packets = 0;
 
 	reg = type == atl1c_trans_high ? REG_TPD_PRI1_CIDX : REG_TPD_PRI0_CIDX;
 
@@ -1558,12 +1561,18 @@ static bool atl1c_clean_tx_irq(struct atl1c_adapter *adapter,
 
 	while (next_to_clean != hw_next_to_clean) {
 		buffer_info = &tpd_ring->buffer_info[next_to_clean];
+		if (buffer_info->skb) {
+			total_bytes += buffer_info->skb->len;
+			total_packets++;
+		}
 		atl1c_clean_buffer(pdev, buffer_info);
 		if (++next_to_clean == tpd_ring->count)
 			next_to_clean = 0;
 		atomic_set(&tpd_ring->next_to_clean, next_to_clean);
 	}
 
+	netdev_completed_queue(adapter->netdev, total_packets, total_bytes);
+
 	if (netif_queue_stopped(adapter->netdev) &&
 			netif_carrier_ok(adapter->netdev)) {
 		netif_wake_queue(adapter->netdev);
@@ -2256,6 +2265,7 @@ static netdev_tx_t atl1c_xmit_frame(struct sk_buff *skb,
 		spin_unlock_irqrestore(&adapter->tx_lock, flags);
 		dev_kfree_skb_any(skb);
 	} else {
+		netdev_sent_queue(adapter->netdev, skb->len);
 		atl1c_tx_queue(adapter, skb, tpd, type);
 		spin_unlock_irqrestore(&adapter->tx_lock, flags);
 	}
-- 
2.4.6

             reply	other threads:[~2015-08-12  6:01 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-08-12  6:01 Ron Angeles [this message]
2015-08-12 22:40 ` [PATCH net-next] net: atl1c: add BQL support Francois Romieu
2015-08-13  5:12   ` Ron Angeles
2015-08-13 22:16     ` Francois Romieu
2015-08-12 23:27 ` David Miller

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=55CAE130.8040406@gmail.com \
    --to=ronangeles@gmail.com \
    --cc=netdev@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.