All of lore.kernel.org
 help / color / mirror / Atom feed
From: Alexander Duyck <alexander.duyck@gmail.com>
To: intel-wired-lan@osuosl.org
Subject: [Intel-wired-lan] [RFC PATCH 20/20] ixgbe: Add support for build_skb
Date: Mon, 19 Dec 2016 12:12:40 -0800	[thread overview]
Message-ID: <20161219201240.27116.6362.stgit@localhost.localdomain> (raw)
In-Reply-To: <20161219200456.27116.28900.stgit@localhost.localdomain>

From: Alexander Duyck <alexander.h.duyck@intel.com>

This patch adds build_skb support to the Rx path.  There are several
advantages to this change.

1.  It avoids the memcpy and skb->head allocation for small packets which
    improves performance by about 5% in my tests.
2.  It avoids the memcpy, skb->head allocation, and eth_get_headlen
    for larger packets improving performance by about 10% in my tests.
3.  For VXLAN packets it allows the full header to be in skb->data which
    improves the performance by as much as 30% in some of my tests.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c |   49 ++++++++++++++++++++++++-
 1 file changed, 48 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index d509b6842b9a..06a97b003639 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -1897,7 +1897,7 @@ static bool ixgbe_cleanup_headers(struct ixgbe_ring *rx_ring,
 	}
 
 	/* place header in linear portion of buffer */
-	if (skb_is_nonlinear(skb))
+	if (!skb_headlen(skb))
 		ixgbe_pull_tail(rx_ring, skb);
 
 #ifdef IXGBE_FCOE
@@ -2120,6 +2120,49 @@ static struct sk_buff *ixgbe_construct_skb(struct ixgbe_ring *rx_ring,
 	return skb;
 }
 
+static struct sk_buff *ixgbe_build_skb(struct ixgbe_ring *rx_ring,
+				       struct ixgbe_rx_buffer *rx_buffer,
+				       union ixgbe_adv_rx_desc *rx_desc,
+				       unsigned int size)
+{
+	void *va = page_address(rx_buffer->page) + rx_buffer->page_offset;
+#if (PAGE_SIZE < 8192)
+	unsigned int truesize = ixgbe_rx_pg_size(rx_ring) / 2;
+#else
+	unsigned int truesize = SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) +
+				SKB_DATA_ALIGN(IXGBE_SKB_PAD + size);
+#endif
+	struct sk_buff *skb;
+
+	/* prefetch first cache line of first page */
+	prefetch(va);
+#if L1_CACHE_BYTES < 128
+	prefetch(va + L1_CACHE_BYTES);
+#endif
+
+	/* build an skb to around the page buffer */
+	skb = build_skb(va - IXGBE_SKB_PAD, truesize);
+	if (unlikely(!skb))
+		return NULL;
+
+	/* update pointers within the skb to store the data */
+	skb_reserve(skb, IXGBE_SKB_PAD);
+	__skb_put(skb, size);
+
+	/* record DMA address if this is the start of a chain of buffers */
+	if (!ixgbe_test_staterr(rx_desc, IXGBE_RXD_STAT_EOP))
+		IXGBE_CB(skb)->dma = rx_buffer->dma;
+
+	/* update buffer offset */
+#if (PAGE_SIZE < 8192)
+	rx_buffer->page_offset ^= truesize;
+#else
+	rx_buffer->page_offset += truesize;
+#endif
+
+	return skb;
+}
+
 /**
  * ixgbe_clean_rx_irq - Clean completed descriptors from Rx ring - bounce buf
  * @q_vector: structure containing interrupt and ring information
@@ -2173,6 +2216,9 @@ static int ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
 		/* retrieve a buffer from the ring */
 		if (skb)
 			ixgbe_add_rx_frag(rx_ring, rx_buffer, skb, size);
+		else if (ring_uses_build_skb(rx_ring))
+			skb = ixgbe_build_skb(rx_ring, rx_buffer,
+					      rx_desc, size);
 		else
 			skb = ixgbe_construct_skb(rx_ring, rx_buffer,
 						  rx_desc, size);
@@ -3924,6 +3970,7 @@ static void ixgbe_set_rx_buffer_len(struct ixgbe_adapter *adapter)
 		if (test_bit(__IXGBE_RX_FCOE, &rx_ring->state))
 			set_bit(__IXGBE_RX_3K_BUFFER, &rx_ring->state);
 
+		clear_bit(__IXGBE_RX_BUILD_SKB_ENABLED, &rx_ring->state);
 		if (adapter->flags2 & IXGBE_FLAG2_RX_LEGACY)
 			continue;
 


      parent reply	other threads:[~2016-12-19 20:12 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-12-19 20:10 [Intel-wired-lan] [RFC PATCH 00/20] Enable the use of build_skb in igb/ixgbe Alexander Duyck
2016-12-19 20:10 ` [Intel-wired-lan] [RFC PATCH 01/20] mm: Rename __alloc_page_frag to page_frag_alloc and __free_page_frag to page_frag_free Alexander Duyck
2016-12-19 20:10 ` [Intel-wired-lan] [RFC PATCH 02/20] mm: Rename __page_frag functions to __page_frag_cache, drop order from drain Alexander Duyck
2016-12-19 20:10 ` [Intel-wired-lan] [RFC PATCH 03/20] mm: Add documentation for page fragment APIs Alexander Duyck
2016-12-19 20:11 ` [Intel-wired-lan] [RFC PATCH 04/20] igb: Add support for DMA_ATTR_WEAK_ORDERING Alexander Duyck
2016-12-19 20:11 ` [Intel-wired-lan] [RFC PATCH 05/20] igb: Use length to determine if descriptor is done Alexander Duyck
2016-12-19 20:11 ` [Intel-wired-lan] [RFC PATCH 06/20] igb: Limit maximum frame Rx based on MTU Alexander Duyck
2016-12-19 20:11 ` [Intel-wired-lan] [RFC PATCH 07/20] igb: Add support for padding packet Alexander Duyck
2016-12-19 20:11 ` [Intel-wired-lan] [RFC PATCH 08/20] igb: Add support for ethtool private flag to allow use of legacy Rx Alexander Duyck
2016-12-19 20:11 ` [Intel-wired-lan] [RFC PATCH 09/20] igb: Break out Rx buffer page management Alexander Duyck
2016-12-19 20:11 ` [Intel-wired-lan] [RFC PATCH 10/20] igb: Revert "igb: Revert support for build_skb in igb" Alexander Duyck
2016-12-19 20:11 ` [Intel-wired-lan] [RFC PATCH 11/20] ixgbe: Add function for checking to see if we can reuse page Alexander Duyck
2016-12-19 20:11 ` [Intel-wired-lan] [RFC PATCH 12/20] ixgbe: Only DMA sync frame length Alexander Duyck
2016-12-19 20:11 ` [Intel-wired-lan] [RFC PATCH 13/20] ixgbe: Update driver to make use of DMA attributes in Rx path Alexander Duyck
2016-12-19 20:12 ` [Intel-wired-lan] [RFC PATCH 14/20] ixgbe: Update code to better handle incrementing page count Alexander Duyck
2016-12-19 20:12 ` [Intel-wired-lan] [RFC PATCH 15/20] ixgbe: Make use of order 1 pages and 3K buffers independent of FCoE Alexander Duyck
2016-12-19 20:12 ` [Intel-wired-lan] [RFC PATCH 16/20] ixgbe: Use length to determine if descriptor is done Alexander Duyck
2016-12-19 20:12 ` [Intel-wired-lan] [RFC PATCH 17/20] ixgbe: Break out Rx buffer page management Alexander Duyck
2016-12-19 20:12 ` [Intel-wired-lan] [RFC PATCH 18/20] ixgbe: Add support for padding packet Alexander Duyck
2016-12-19 20:12 ` [Intel-wired-lan] [RFC PATCH 19/20] ixgbe: Add private flag to control buffer mode Alexander Duyck
2016-12-19 20:12 ` Alexander Duyck [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=20161219201240.27116.6362.stgit@localhost.localdomain \
    --to=alexander.duyck@gmail.com \
    --cc=intel-wired-lan@osuosl.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.