mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Corentin Labbe <clabbe.montjoie@gmail.com>
To: peppe.cavallaro@st.com, alexandre.torgue@st.com, netdev@vger.kernel.org
Cc: davem@davemloft.net, linux-kernel@vger.kernel.org,
	Corentin Labbe <clabbe.montjoie@gmail.com>
Subject: [PATCH 13/17] net: stmmac: Implement NAPI for TX
Date: Tue, 31 Jan 2017 10:11:48 +0100	[thread overview]
Message-ID: <20170131091152.13842-14-clabbe.montjoie@gmail.com> (raw)
In-Reply-To: <20170131091152.13842-1-clabbe.montjoie@gmail.com>

The stmmac driver run TX completion under NAPI but without checking the
work done by the TX completion function.

This patch add work/budget to the TX completion function.

The visible effect is that it keep the driver longer under NAPI and
boost performance.
Under dwmac-sun8i the iperf goes from 140Mbit/s to 500Mbit/s.
Under dwmac-sunxi an iperf run use half less interrupts.

Signed-off-by: Corentin Labbe <clabbe.montjoie@gmail.com>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 18 ++++++++++++++----
 1 file changed, 14 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 2df36bd..e53b727 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -1299,10 +1299,11 @@ static void stmmac_dma_operation_mode(struct stmmac_priv *priv)
  * @priv: driver private structure
  * Description: it reclaims the transmit resources after transmission completes.
  */
-static void stmmac_tx_clean(struct stmmac_priv *priv)
+static int stmmac_tx_clean(struct stmmac_priv *priv, int budget)
 {
 	unsigned int bytes_compl = 0, pkts_compl = 0;
 	unsigned int entry = priv->dirty_tx;
+	int work = 0;
 
 	netif_tx_lock(priv->dev);
 
@@ -1369,6 +1370,9 @@ static void stmmac_tx_clean(struct stmmac_priv *priv)
 		priv->hw->desc->release_tx_desc(p, priv->mode);
 
 		entry = STMMAC_GET_ENTRY(entry, DMA_TX_SIZE);
+		work++;
+		if (work >= budget)
+			break;
 	}
 	priv->dirty_tx = entry;
 
@@ -1386,6 +1390,11 @@ static void stmmac_tx_clean(struct stmmac_priv *priv)
 		mod_timer(&priv->eee_ctrl_timer, STMMAC_LPI_T(eee_timer));
 	}
 	netif_tx_unlock(priv->dev);
+
+	if (work < budget)
+		work = 0;
+
+	return work;
 }
 
 static inline void stmmac_enable_dma_irq(struct stmmac_priv *priv)
@@ -1617,7 +1626,7 @@ static void stmmac_tx_timer(unsigned long data)
 {
 	struct stmmac_priv *priv = (struct stmmac_priv *)data;
 
-	stmmac_tx_clean(priv);
+	stmmac_tx_clean(priv, 256);
 }
 
 /**
@@ -2657,9 +2666,10 @@ static int stmmac_poll(struct napi_struct *napi, int budget)
 	int work_done = 0;
 
 	priv->xstats.napi_poll++;
-	stmmac_tx_clean(priv);
+	work_done += stmmac_tx_clean(priv, budget);
 
-	work_done = stmmac_rx(priv, budget);
+	if (work_done < budget)
+		work_done += stmmac_rx(priv, budget - work_done);
 	if (work_done < budget) {
 		napi_complete(napi);
 		stmmac_enable_dma_irq(priv);
-- 
2.10.2

  parent reply	other threads:[~2017-01-31  9:16 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-01-31  9:11 [PATCH 00/17] net: stmmac: misc fix Corentin Labbe
2017-01-31  9:11 ` [PATCH 01/17] net: stmmac: fix the typo on MAC_RNABLE_RX Corentin Labbe
2017-01-31 10:03   ` Giuseppe CAVALLARO
2017-01-31  9:11 ` [PATCH 02/17] net: stmmac: Remove the bus_setup function pointer Corentin Labbe
2017-01-31 10:02   ` Giuseppe CAVALLARO
2017-01-31  9:11 ` [PATCH 03/17] net: stmmac: fix some typos in comments Corentin Labbe
2017-01-31 10:03   ` Giuseppe CAVALLARO
2017-01-31  9:11 ` [PATCH 04/17] net: stmmac: remove freesoftware address Corentin Labbe
2017-01-31  9:11 ` [PATCH 05/17] net: stmmac: remplace asm/io.h by linux/io.h Corentin Labbe
2017-01-31  9:11 ` [PATCH 06/17] net: stmmac: fix some code style problem Corentin Labbe
2017-01-31 10:05   ` Giuseppe CAVALLARO
2017-01-31  9:11 ` [PATCH 07/17] net: stmmac: replace stmmac_mdio_busy_wait by readl_poll_timeout Corentin Labbe
2017-01-31 10:13   ` Giuseppe CAVALLARO
2017-01-31 10:39     ` Corentin Labbe
2017-01-31 10:44       ` Giuseppe CAVALLARO
2017-01-31  9:11 ` [PATCH 08/17] net: stmmac: Use readl_poll_timeout Corentin Labbe
2017-01-31  9:11 ` [PATCH 09/17] net: stmmac: replace ENOSYS by EINVAL Corentin Labbe
2017-01-31 10:06   ` Giuseppe CAVALLARO
2017-01-31  9:11 ` [PATCH 10/17] net: stmmac: Correct the error message about invalid speed Corentin Labbe
2017-01-31 10:07   ` Giuseppe CAVALLARO
2017-01-31  9:11 ` [PATCH 11/17] net: stmmac: Rewrite two test against NULL value Corentin Labbe
2017-01-31 10:07   ` Giuseppe CAVALLARO
2017-01-31  9:11 ` [PATCH 12/17] net: stmmac: rename rx_crc to rx_crc_errors Corentin Labbe
2017-01-31 10:08   ` Giuseppe CAVALLARO
2017-01-31  9:11 ` Corentin Labbe [this message]
2017-01-31 10:28   ` [PATCH 13/17] net: stmmac: Implement NAPI for TX Giuseppe CAVALLARO
2017-01-31 13:38     ` Corentin Labbe
2017-02-01  4:12   ` David Miller
2017-02-03 13:41     ` Corentin Labbe
2017-02-03 15:15       ` David Miller
2017-02-03 15:58         ` Corentin Labbe
2017-01-31  9:11 ` [PATCH 14/17] net: stmmac: print phy information Corentin Labbe
2017-01-31 10:10   ` Giuseppe CAVALLARO
2017-02-03 13:16     ` Corentin Labbe
2017-01-31  9:11 ` [PATCH 15/17] net: stmmac: remove dead code in stmmac_tx_clean Corentin Labbe
2017-01-31 10:21   ` Giuseppe CAVALLARO
2017-01-31  9:11 ` [PATCH 16/17] net: stmmac: remove unused variable in sysfs_display_ring Corentin Labbe
2017-01-31 10:11   ` Giuseppe CAVALLARO
2017-01-31  9:11 ` [PATCH 17/17] net: stmmac: replace unsigned by u32 Corentin Labbe
2017-01-31 10:12   ` Giuseppe CAVALLARO
2017-01-31 10:00 ` [PATCH 00/17] net: stmmac: misc fix Giuseppe CAVALLARO
2017-01-31 10:23   ` Joao Pinto
2017-01-31 10:33     ` Giuseppe CAVALLARO
2017-01-31 10:37       ` Joao Pinto

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=20170131091152.13842-14-clabbe.montjoie@gmail.com \
    --to=clabbe.montjoie@gmail.com \
    --cc=alexandre.torgue@st.com \
    --cc=davem@davemloft.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=peppe.cavallaro@st.com \
    /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

all inboxes | Powered by JetHome®