[dpdk-dev] [PATCH v2 2/3] app/testpmd: improve MAC swap performance for x86

[Qi Zhang <qi.z.zhang@intel.com>]

The patch optimizes the mac swap operation by taking advantage
of SSE instructions, it only impacts x86 platform.

Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
---
 app/test-pmd/macswap.c     |  4 ++++
 app/test-pmd/macswap_sse.h | 43 +++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 47 insertions(+)
 create mode 100644 app/test-pmd/macswap_sse.h

diff --git a/app/test-pmd/macswap.c b/app/test-pmd/macswap.c
index 849194fe2..cbb41b728 100644
--- a/app/test-pmd/macswap.c
+++ b/app/test-pmd/macswap.c
@@ -66,7 +66,11 @@
 #include <rte_flow.h>
 
 #include "testpmd.h"
+#ifdef RTE_ARCH_X86
+#include "macswap_sse.h"
+#else
 #include "macswap.h"
+#endif
 
 /*
  * MAC swap forwarding mode: Swap the source and the destination Ethernet
diff --git a/app/test-pmd/macswap_sse.h b/app/test-pmd/macswap_sse.h
new file mode 100644
index 000000000..d5b0f6a21
--- /dev/null
+++ b/app/test-pmd/macswap_sse.h
@@ -0,0 +1,43 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2018 Intel Corporation
+ */
+
+#ifndef _L2FWD_SSE_H_
+#define _L2FWD_SSE_H_
+
+#include "macswap_common.h"
+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;
+	uint64_t ol_flags;
+	int i;
+	__m128i addr;
+	__m128i shfl_msk = _mm_set_epi8(15, 14, 13, 12,
+					5, 4, 3, 2,
+					1, 0, 11, 10,
+					9, 8, 7, 6);
+
+	ol_flags = ol_flags_init(txp->dev_conf.txmode.offloads);
+
+	for (i = 0; i < nb; i++) {
+		if (likely(i < nb - 1))
+			rte_prefetch0(rte_pktmbuf_mtod(pkts[i+1], void *));
+		mb = pkts[i];
+
+		eth_hdr = rte_pktmbuf_mtod(mb, 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);
+
+		mbuf_field_set(mb, ol_flags, txp->tx_vlan_id,
+				txp->tx_vlan_id_outer);
+	}
+}
+
+#endif /* _BPF_CMD_H_ */
+
-- 
2.13.6