From: Qi Zhang <qi.z.zhang@intel.com>
To: ferruh.yigit@intel.com, bruce.richardson@intel.com,
keith.wiles@intel.com, konstantin.ananyev@intel.com
Cc: dev@dpdk.org, wenzhuo.lu@intel.com, bernard.iremonger@intel.com,
Qi Zhang <qi.z.zhang@intel.com>
Subject: [PATCH v2 3/3] app/testpmd: further improve MAC swap performance for x86
Date: Tue, 11 Dec 2018 13:55:11 +0800 [thread overview]
Message-ID: <20181211055511.32284-4-qi.z.zhang@intel.com> (raw)
In-Reply-To: <20181211055511.32284-1-qi.z.zhang@intel.com>
Do four packets macswap in same loop iterate to squeeze more
CPU cycles.
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
---
app/test-pmd/macswap_sse.h | 62 +++++++++++++++++++++++++++++++++++++---------
1 file changed, 50 insertions(+), 12 deletions(-)
diff --git a/app/test-pmd/macswap_sse.h b/app/test-pmd/macswap_sse.h
index 79f4f9a7c..df2875ace 100644
--- a/app/test-pmd/macswap_sse.h
+++ b/app/test-pmd/macswap_sse.h
@@ -11,11 +11,12 @@ static inline void
do_macswap(struct rte_mbuf *pkts[], uint16_t nb,
struct rte_port *txp)
{
- struct ether_hdr *eth_hdr;
- struct rte_mbuf *mb;
+ struct ether_hdr *eth_hdr[4];
+ struct rte_mbuf *mb[4];
uint64_t ol_flags;
int i;
- __m128i addr;
+ int r;
+ __m128i addr0, addr1, addr2, addr3;
__m128i shfl_msk = _mm_set_epi8(15, 14, 13, 12,
5, 4, 3, 2,
1, 0, 11, 10,
@@ -25,19 +26,56 @@ do_macswap(struct rte_mbuf *pkts[], uint16_t nb,
vlan_qinq_set(pkts, nb, ol_flags,
txp->tx_vlan_id, txp->tx_vlan_id_outer);
- for (i = 0; i < nb; i++) {
- if (likely(i < nb - 1))
- rte_prefetch0(rte_pktmbuf_mtod(pkts[i+1], void *));
- mb = pkts[i];
+ i = 0;
+ r = nb;
+
+ while (r >= 4) {
+ mb[0] = pkts[i++];
+ eth_hdr[0] = rte_pktmbuf_mtod(mb[0], struct ether_hdr *);
+ addr0 = _mm_loadu_si128((__m128i *)eth_hdr[0]);
+
+ mb[1] = pkts[i++];
+ eth_hdr[1] = rte_pktmbuf_mtod(mb[1], struct ether_hdr *);
+ addr1 = _mm_loadu_si128((__m128i *)eth_hdr[1]);
+
+
+ mb[2] = pkts[i++];
+ eth_hdr[2] = rte_pktmbuf_mtod(mb[2], struct ether_hdr *);
+ addr2 = _mm_loadu_si128((__m128i *)eth_hdr[2]);
+
+ mb[3] = pkts[i++];
+ eth_hdr[3] = rte_pktmbuf_mtod(mb[3], struct ether_hdr *);
+ addr3 = _mm_loadu_si128((__m128i *)eth_hdr[3]);
- eth_hdr = rte_pktmbuf_mtod(mb, struct ether_hdr *);
+ addr0 = _mm_shuffle_epi8(addr0, shfl_msk);
+ addr1 = _mm_shuffle_epi8(addr1, shfl_msk);
+ addr2 = _mm_shuffle_epi8(addr2, shfl_msk);
+ addr3 = _mm_shuffle_epi8(addr3, shfl_msk);
+
+ _mm_storeu_si128((__m128i *)eth_hdr[0], addr0);
+ _mm_storeu_si128((__m128i *)eth_hdr[1], addr1);
+ _mm_storeu_si128((__m128i *)eth_hdr[2], addr2);
+ _mm_storeu_si128((__m128i *)eth_hdr[3], addr3);
+
+ mbuf_field_set(mb[0], ol_flags);
+ mbuf_field_set(mb[1], ol_flags);
+ mbuf_field_set(mb[2], ol_flags);
+ mbuf_field_set(mb[3], ol_flags);
+ r -= 4;
+ }
+
+ for ( ; i < nb; i++) {
+ if (i < nb - 1)
+ rte_prefetch0(rte_pktmbuf_mtod(pkts[i+1], void *));
+ mb[0] = pkts[i];
+ eth_hdr[0] = rte_pktmbuf_mtod(mb[0], struct ether_hdr *);
/* Swap dest and src mac addresses. */
- addr = _mm_loadu_si128((__m128i *)eth_hdr);
- addr = _mm_shuffle_epi8(addr, shfl_msk);
- _mm_storeu_si128((__m128i *)eth_hdr, addr);
+ addr0 = _mm_loadu_si128((__m128i *)eth_hdr);
+ addr0 = _mm_shuffle_epi8(addr0, shfl_msk);
+ _mm_storeu_si128((__m128i *)eth_hdr[0], addr0);
- mbuf_field_set(mb, ol_flags);
+ mbuf_field_set(mb[0], ol_flags);
}
}
--
2.13.6
next prev parent reply other threads:[~2018-12-11 5:53 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-11-22 17:26 [PATCH 0/3] improve MAC swap performance Qi Zhang
2018-11-22 17:26 ` [PATCH 1/3] app/testpmd: code refactory for macswap Qi Zhang
2018-11-22 17:26 ` [PATCH 2/3] app/testpmd: improve MAC swap performance for x86 Qi Zhang
2018-11-22 17:26 ` [PATCH 3/3] app/testpmd: further " Qi Zhang
2018-11-22 17:38 ` [PATCH v2 0/3] improve MAC swap performance Qi Zhang
2018-11-22 17:38 ` [PATCH v2 1/3] app/testpmd: code refactory for macswap Qi Zhang
2018-12-10 17:44 ` Ferruh Yigit
2018-12-11 4:02 ` Zhang, Qi Z
2018-11-22 17:38 ` [PATCH v2 2/3] app/testpmd: improve MAC swap performance for x86 Qi Zhang
2018-12-10 17:44 ` Ferruh Yigit
2018-11-22 17:38 ` [PATCH v2 3/3] app/testpmd: further " Qi Zhang
2018-12-11 5:55 ` [PATCH v2 0/3] improve MAC swap performance Qi Zhang
2018-12-11 5:55 ` [PATCH v2 1/3] app/testpmd: code refactory for macswap Qi Zhang
2018-12-11 15:48 ` Ferruh Yigit
2018-12-14 11:14 ` Iremonger, Bernard
2018-12-14 11:21 ` Zhang, Qi Z
2018-12-11 5:55 ` [PATCH v2 2/3] app/testpmd: improve MAC swap performance for x86 Qi Zhang
2018-12-13 19:50 ` [EXT] " Jerin Jacob Kollanukkaran
2018-12-14 11:21 ` Iremonger, Bernard
2018-12-14 11:36 ` Zhang, Qi Z
2018-12-11 5:55 ` Qi Zhang [this message]
2018-12-16 0:58 ` [PATCH v3 0/3] improve MAC swap performance Qi Zhang
2018-12-16 0:58 ` [PATCH v3 1/3] app/testpmd: code refactory for macswap Qi Zhang
2018-12-18 11:06 ` Iremonger, Bernard
2018-12-16 0:58 ` [PATCH v3 2/3] app/testpmd: improve MAC swap performance for x86 Qi Zhang
2018-12-18 11:07 ` Iremonger, Bernard
2018-12-16 0:58 ` [PATCH v3 3/3] app/testpmd: further " Qi Zhang
2018-12-18 11:09 ` Iremonger, Bernard
2018-12-18 0:15 ` [PATCH v3 0/3] improve MAC swap performance Ferruh Yigit
2018-12-18 16:29 ` Ferruh Yigit
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=20181211055511.32284-4-qi.z.zhang@intel.com \
--to=qi.z.zhang@intel.com \
--cc=bernard.iremonger@intel.com \
--cc=bruce.richardson@intel.com \
--cc=dev@dpdk.org \
--cc=ferruh.yigit@intel.com \
--cc=keith.wiles@intel.com \
--cc=konstantin.ananyev@intel.com \
--cc=wenzhuo.lu@intel.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 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.