DPDK patches and discussions
 help / color / mirror / Atom feed
From: Bruce Richardson <bruce.richardson@intel.com>
To: dev@dpdk.org
Subject: [dpdk-dev] [PATCH v2 2/5] ixgbe: add prefetch to improve slow-path tx perf
Date: Tue, 23 Sep 2014 12:08:14 +0100
Message-ID: <1411470497-10209-3-git-send-email-bruce.richardson@intel.com> (raw)
In-Reply-To: <1411470497-10209-1-git-send-email-bruce.richardson@intel.com>

Make a small improvement to slow path TX performance by adding in a
prefetch for the second mbuf cache line.
Also move assignment of l2/l3 length values only when needed.

What I've done with the prefetches is two-fold:
1) changed it from prefetching the mbuf (first cache line) to prefetching
the mbuf pool pointer (second cache line) so that when we go to access
the pool pointer to free transmitted mbufs we don't get a cache miss. When
clearing the ring and freeing mbufs, the pool pointer is the only mbuf
field used, so we don't need that first cache line.
2) changed the code to prefetch earlier - in effect to prefetch one mbuf
ahead. The original code prefetched the mbuf to be freed as soon as it
started processing the mbuf to replace it. Instead now, every time we
calculate what the next mbuf position is going to be we prefetch the mbuf
in that position (i.e. the mbuf pool pointer we are going to free the mbuf
to), even while we are still updating the previous mbuf slot on the ring.
This gives the prefetch much more time to resolve and get the data we need
in the cache before we need it.

In terms of performance difference, a quick sanity test using testpmd
on a Xeon (Sandy Bridge uarch) platform showed performance increases
between approx 8-18%, depending on the particular RX path used in
conjuntion with this TX path code.

Changes in V2:
* Expanded commit message with extra details of change.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
---
 lib/librte_pmd_ixgbe/ixgbe_rxtx.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/lib/librte_pmd_ixgbe/ixgbe_rxtx.c b/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
index 6f702b3..c0bb49f 100644
--- a/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
+++ b/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
@@ -565,25 +565,26 @@ ixgbe_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts,
 		ixgbe_xmit_cleanup(txq);
 	}
 
+	rte_prefetch0(&txe->mbuf->pool);
+
 	/* TX loop */
 	for (nb_tx = 0; nb_tx < nb_pkts; nb_tx++) {
 		new_ctx = 0;
 		tx_pkt = *tx_pkts++;
 		pkt_len = tx_pkt->pkt_len;
 
-		RTE_MBUF_PREFETCH_TO_FREE(txe->mbuf);
-
 		/*
 		 * Determine how many (if any) context descriptors
 		 * are needed for offload functionality.
 		 */
 		ol_flags = tx_pkt->ol_flags;
-		vlan_macip_lens.f.vlan_tci = tx_pkt->vlan_tci;
-		vlan_macip_lens.f.l2_l3_len = tx_pkt->l2_l3_len;
 
 		/* If hardware offload required */
 		tx_ol_req = ol_flags & PKT_TX_OFFLOAD_MASK;
 		if (tx_ol_req) {
+			vlan_macip_lens.f.vlan_tci = tx_pkt->vlan_tci;
+			vlan_macip_lens.f.l2_l3_len = tx_pkt->l2_l3_len;
+
 			/* If new context need be built or reuse the exist ctx. */
 			ctx = what_advctx_update(txq, tx_ol_req,
 				vlan_macip_lens.data);
@@ -720,7 +721,7 @@ ixgbe_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts,
 				    &txr[tx_id];
 
 				txn = &sw_ring[txe->next_id];
-				RTE_MBUF_PREFETCH_TO_FREE(txn->mbuf);
+				rte_prefetch0(&txn->mbuf->pool);
 
 				if (txe->mbuf != NULL) {
 					rte_pktmbuf_free_seg(txe->mbuf);
@@ -749,6 +750,7 @@ ixgbe_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts,
 		do {
 			txd = &txr[tx_id];
 			txn = &sw_ring[txe->next_id];
+			rte_prefetch0(&txn->mbuf->pool);
 
 			if (txe->mbuf != NULL)
 				rte_pktmbuf_free_seg(txe->mbuf);
-- 
1.9.3

  parent reply	other threads:[~2014-09-23 11:02 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-09-17 10:01 [dpdk-dev] [PATCH 0/5] Mbuf Structure Rework, part 3 Bruce Richardson
2014-09-17 10:01 ` [dpdk-dev] [PATCH 1/5] mbuf: ensure next pointer is set to null on free Bruce Richardson
2014-09-17 10:01 ` [dpdk-dev] [PATCH 2/5] ixgbe: add prefetch to improve slow-path tx perf Bruce Richardson
2014-09-17 15:21   ` Neil Horman
2014-09-17 15:35     ` Richardson, Bruce
2014-09-17 17:59       ` Neil Horman
2014-09-18 13:36         ` Bruce Richardson
2014-09-18 15:29           ` Neil Horman
2014-09-18 15:42             ` Bruce Richardson
2014-09-18 17:56               ` Neil Horman
2014-09-17 10:01 ` [dpdk-dev] [PATCH 3/5] testpmd: Change rxfreet default to 32 Bruce Richardson
2014-09-17 15:29   ` Neil Horman
2014-09-18 15:53     ` Richardson, Bruce
2014-09-18 17:13       ` Thomas Monjalon
2014-09-18 18:08         ` Neil Horman
2014-09-19  9:18           ` Richardson, Bruce
2014-09-19 10:24             ` Neil Horman
2014-09-19 10:28               ` Richardson, Bruce
2014-09-19 15:18                 ` Neil Horman
2014-09-18 18:03       ` Neil Horman
2014-09-17 10:01 ` [dpdk-dev] [PATCH 4/5] mbuf: add userdata pointer field Bruce Richardson
2014-09-17 15:35   ` Neil Horman
2014-09-17 16:02     ` Richardson, Bruce
2014-09-17 18:29       ` Neil Horman
2014-09-17 10:01 ` [dpdk-dev] [PATCH 5/5] mbuf: Add in second vlan tag field to mbuf Bruce Richardson
2014-09-17 20:46   ` Stephen Hemminger
2014-09-23 11:08 ` [dpdk-dev] [PATCH v2 0/5] Mbuf Structure Rework, part 3 Bruce Richardson
2014-09-23 11:08   ` [dpdk-dev] [PATCH v2 1/5] mbuf: ensure next pointer is set to null on free Bruce Richardson
2014-09-23 11:08   ` Bruce Richardson [this message]
2014-09-23 11:08   ` [dpdk-dev] [PATCH v2 3/5] testpmd: Change rxfreet default to 32 Bruce Richardson
2014-09-23 17:02     ` Neil Horman
2014-09-24  9:03       ` Richardson, Bruce
2014-09-24 10:05         ` Neil Horman
2014-11-07 12:30         ` Thomas Monjalon
2014-11-07 13:49           ` Bruce Richardson
2014-09-23 11:08   ` [dpdk-dev] [PATCH v2 4/5] mbuf: add userdata pointer field Bruce Richardson
2014-09-23 11:08   ` [dpdk-dev] [PATCH v2 5/5] mbuf: switch vlan_tci and reserved2 fields Bruce Richardson
2014-09-29 15:58   ` [dpdk-dev] [PATCH v2 0/5] Mbuf Structure Rework, part 3 De Lara Guarch, Pablo
2014-10-08 12:31     ` Thomas Monjalon

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=1411470497-10209-3-git-send-email-bruce.richardson@intel.com \
    --to=bruce.richardson@intel.com \
    --cc=dev@dpdk.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

DPDK patches and discussions

This inbox may be cloned and mirrored by anyone:

	git clone --mirror https://inbox.dpdk.org/dev/0 dev/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 dev dev/ https://inbox.dpdk.org/dev \
		dev@dpdk.org
	public-inbox-index dev

Example config snippet for mirrors.
Newsgroup available over NNTP:
	nntp://inbox.dpdk.org/inbox.dpdk.dev


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git