From: Jisheng Zhang <jszhang@marvell.com>
To: <sebastian.hesselbarth@gmail.com>, <davem@davemloft.net>
Cc: <netdev@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
<linux-arm-kernel@lists.infradead.org>,
Jisheng Zhang <jszhang@marvell.com>
Subject: [PATCH 2/2] net: mv643xx_eth: use dma_wmb/rmb where appropriate
Date: Fri, 13 May 2016 19:59:20 +0800 [thread overview]
Message-ID: <1463140760-7128-3-git-send-email-jszhang@marvell.com> (raw)
In-Reply-To: <1463140760-7128-1-git-send-email-jszhang@marvell.com>
Update the mv643xx_eth driver to use the dma_rmb/wmb calls instead of
the full barriers in order to improve performance.
Signed-off-by: Jisheng Zhang <jszhang@marvell.com>
---
drivers/net/ethernet/marvell/mv643xx_eth.c | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mv643xx_eth.c b/drivers/net/ethernet/marvell/mv643xx_eth.c
index c6d8124..13b71e3 100644
--- a/drivers/net/ethernet/marvell/mv643xx_eth.c
+++ b/drivers/net/ethernet/marvell/mv643xx_eth.c
@@ -536,7 +536,7 @@ static int rxq_process(struct rx_queue *rxq, int budget)
cmd_sts = rx_desc->cmd_sts;
if (cmd_sts & BUFFER_OWNED_BY_DMA)
break;
- rmb();
+ dma_rmb();
skb = rxq->rx_skb[rxq->rx_curr_desc];
rxq->rx_skb[rxq->rx_curr_desc] = NULL;
@@ -647,9 +647,9 @@ static int rxq_refill(struct rx_queue *rxq, int budget)
DMA_FROM_DEVICE);
rx_desc->buf_size = size;
rxq->rx_skb[rx] = skb;
- wmb();
+ dma_wmb();
rx_desc->cmd_sts = BUFFER_OWNED_BY_DMA | RX_ENABLE_INTERRUPT;
- wmb();
+ dma_wmb();
/*
* The hardware automatically prepends 2 bytes of
@@ -889,7 +889,7 @@ static int txq_submit_tso(struct tx_queue *txq, struct sk_buff *skb,
skb_tx_timestamp(skb);
/* ensure all other descriptors are written before first cmd_sts */
- wmb();
+ dma_wmb();
first_tx_desc->cmd_sts = first_cmd_sts;
/* clear TX_END status */
@@ -994,7 +994,7 @@ static int txq_submit_skb(struct tx_queue *txq, struct sk_buff *skb,
skb_tx_timestamp(skb);
/* ensure all other descriptors are written before first cmd_sts */
- wmb();
+ dma_wmb();
desc->cmd_sts = cmd_sts;
/* clear TX_END status */
--
2.8.1
prev parent reply other threads:[~2016-05-13 12:03 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-05-13 11:59 [PATCH 0/2] net: mv643xx_eth: improve performance Jisheng Zhang
2016-05-13 11:59 ` [PATCH 1/2] net: mv643xx_eth: use {readl|writel}_relaxed instead of readl/writel Jisheng Zhang
2016-05-13 12:09 ` Arnd Bergmann
2016-05-13 12:19 ` Jisheng Zhang
2016-05-13 15:24 ` Andrew Lunn
2016-05-13 11:59 ` Jisheng Zhang [this message]
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=1463140760-7128-3-git-send-email-jszhang@marvell.com \
--to=jszhang@marvell.com \
--cc=davem@davemloft.net \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=sebastian.hesselbarth@gmail.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;
as well as URLs for NNTP newsgroup(s).