DPDK patches and discussions
 help / color / mirror / Atom feed
From: Vipin Varghese <vipin.varghese@amd.com>
To: <ferruh.yigit@amd.com>, <bruce.richardson@intel.com>,
	<konstantin.v.ananyev@yandex.ru>, <aman.deep.singh@intel.com>,
	<dev@dpdk.org>
Subject: [PATCH v2 2/3] app/testpmd: move offload update
Date: Wed, 21 Aug 2024 20:08:56 +0530	[thread overview]
Message-ID: <20240821143857.1972-3-vipin.varghese@amd.com> (raw)
In-Reply-To: <20240821143857.1972-1-vipin.varghese@amd.com>

Moving the offload flag update from end to start of the loop,
helps to reduce L1 or L2 cache evictions and amortize shuffle.
This helps to improve RX packet in mac-swap processing.

Test Result:
 * Platform: AMD EPYC 9554 @3.1GHz, no boost
 * Test scenarios: TEST-PMD 64B IO vs MAC-SWAP
 * NIC: broadcom P2100: loopback 2*100Gbps

 <mode : Mpps Ingress: Mpps Egress>
 ------------------------------------------------
  - MAC-SWAP original: 45.75 : 43.8
  - MAC-SWAP register mod: 45.73 : 44.83
  - MAC-SWAP register+ofl modified: 46.36 : 44.79

Signed-off-by: Vipin Varghese <vipin.varghese@amd.com>
---
 app/test-pmd/macswap_sse.h | 13 ++++++-------
 1 file changed, 6 insertions(+), 7 deletions(-)

diff --git a/app/test-pmd/macswap_sse.h b/app/test-pmd/macswap_sse.h
index 29088843b7..67ff7fdfbb 100644
--- a/app/test-pmd/macswap_sse.h
+++ b/app/test-pmd/macswap_sse.h
@@ -45,19 +45,22 @@ do_macswap(struct rte_mbuf *pkts[], uint16_t nb,
 		mb[0] = pkts[i++];
 		eth_hdr[0] = rte_pktmbuf_mtod(mb[0], struct rte_ether_hdr *);
 		addr0 = _mm_loadu_si128((__m128i *)eth_hdr[0]);
+		mbuf_field_set(mb[0], ol_flags);
 
 		mb[1] = pkts[i++];
 		eth_hdr[1] = rte_pktmbuf_mtod(mb[1], struct rte_ether_hdr *);
 		addr1 = _mm_loadu_si128((__m128i *)eth_hdr[1]);
-
+		mbuf_field_set(mb[1], ol_flags);
 
 		mb[2] = pkts[i++];
 		eth_hdr[2] = rte_pktmbuf_mtod(mb[2], struct rte_ether_hdr *);
 		addr2 = _mm_loadu_si128((__m128i *)eth_hdr[2]);
+		mbuf_field_set(mb[2], ol_flags);
 
 		mb[3] = pkts[i++];
 		eth_hdr[3] = rte_pktmbuf_mtod(mb[3], struct rte_ether_hdr *);
 		addr3 = _mm_loadu_si128((__m128i *)eth_hdr[3]);
+		mbuf_field_set(mb[3], ol_flags);
 
 		addr0 = _mm_shuffle_epi8(addr0, shfl_msk);
 		addr1 = _mm_shuffle_epi8(addr1, shfl_msk);
@@ -69,10 +72,6 @@ do_macswap(struct rte_mbuf *pkts[], uint16_t nb,
 		_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;
 	}
 
@@ -84,10 +83,10 @@ do_macswap(struct rte_mbuf *pkts[], uint16_t nb,
 
 		/* Swap dest and src mac addresses. */
 		addr0 = _mm_loadu_si128((__m128i *)eth_hdr[0]);
+		mbuf_field_set(mb[0], ol_flags);
+
 		addr0 = _mm_shuffle_epi8(addr0, shfl_msk);
 		_mm_storeu_si128((__m128i *)eth_hdr[0], addr0);
-
-		mbuf_field_set(mb[0], ol_flags);
 	}
 }
 
-- 
2.34.1


  parent reply	other threads:[~2024-08-21 14:39 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-07-16  6:37 [PATCH] app/testpmd: improve sse based macswap Vipin Varghese
2024-07-23 16:45 ` Ferruh Yigit
2024-07-23 17:12   ` Bruce Richardson
2024-07-25 12:47     ` Varghese, Vipin
2024-07-25 12:52       ` Bruce Richardson
2024-08-21 14:38 ` [PATCH v2 0/3] " Vipin Varghese
2024-08-21 14:38   ` [PATCH v2 1/3] app/testpmd: add register keyword Vipin Varghese
2024-08-21 14:55     ` Stephen Hemminger
2024-08-27 15:32       ` Varghese, Vipin
2024-08-27 17:39         ` Stephen Hemminger
2024-08-29  8:14           ` Varghese, Vipin
2024-09-03 11:52           ` Konstantin Ananyev
2024-09-06 13:02             ` Varghese, Vipin
2024-08-21 14:38   ` Vipin Varghese [this message]
2024-08-21 14:38   ` [PATCH v2 3/3] app/testpmd: interleave SSE SIMD Vipin Varghese

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=20240821143857.1972-3-vipin.varghese@amd.com \
    --to=vipin.varghese@amd.com \
    --cc=aman.deep.singh@intel.com \
    --cc=bruce.richardson@intel.com \
    --cc=dev@dpdk.org \
    --cc=ferruh.yigit@amd.com \
    --cc=konstantin.v.ananyev@yandex.ru \
    /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).