DPDK patches and discussions
 help / color / mirror / Atom feed
* Re: [dpdk-dev] [PATCH 1/1] net/i40e: fix compilation failure on core-avx-i
@ 2021-07-14  1:06 Lu, Wenzhuo
  2021-08-02  0:17 ` Zhang, Qi Z
  0 siblings, 1 reply; 3+ messages in thread
From: Lu, Wenzhuo @ 2021-07-14  1:06 UTC (permalink / raw)
  To: shaikh; +Cc: dev, stable, Xing,  Beilei

Hi Shahed,



> -----Original Message-----

> From: Shahed Shaikh <shaikh@niometrics.com<mailto:shaikh@niometrics.com>>

> Sent: Tuesday, June 22, 2021 5:37 PM

> To: Xing, Beilei <beilei.xing@intel.com<mailto:beilei.xing@intel.com>>

> Cc: dev@dpdk.org<mailto:dev@dpdk.org>; Shahed Shaikh <shaikh@niometrics.com<mailto:shaikh@niometrics.com>>;

> stable@dpdk.org<mailto:stable@dpdk.org>

> Subject: [PATCH 1/1] net/i40e: fix compilation failure on core-avx-i

>

> i40e_rxtx_vec_sse.c fails to compile with below configuration:

> - #define RTE_LIBRTE_I40E_16BYTE_RX_DESC 1 in config/rte_config.h

> - cpu=core-axv-i

> - gcc which supports -mavx2 (e.g. gcc 4.8.5)

>

> This is because commit 0604b1f2208f ("net/i40e: fix crash in AVX512")

> added

> i40e_rxq_rearm_common() to i40e_rxtx_vec_common.h which is included by

> i40e_rxtx_vec_sse.c.

>

> This function is enabled for compilation if CC_AVX2_SUPPORT is defined.

> As per drivers/net/i40e/meson.build, CC_AVX2_SUPPORT is defined when

> either CPU supports __AVX2__ or compiler supports -mavx2 option.

>

> So for given configuration, CC_AVX2_SUPPORT gets defined but we don't

> pass -mavx2 explicitly to gcc while compiling i40e_rxtx_vec_sse.c.

Looks like it's better and more clear if a new specific avx head file is added.

Just want to understand more about the problem. I don't see the same problem when using 'gcc version 4.8.5 20150623 (Red Hat 4.8.5-44) (GCC)', don't understand why -mavx2 is missing.

And more, if so, why the avx2 code doesn't meet the same problem?

Thanks.



> Hence it fails due to avx2 specific code from i40e_rxq_rearm_command().

>

> This patch tries to fix the compilation by moving

> i40e_rxq_rearm_common() to a new header file which will only be

> included by i40e_rxtx_vec_avx2.c and i40e_rxtx_vec_avx512.c.

>

> Fixes: 0604b1f2208f ("net/i40e: fix crash in AVX512")

> Cc: stable@dpdk.org<mailto:stable@dpdk.org>

>

> Signed-off-by: Shahed Shaikh <shaikh@niometrics.com<mailto:shaikh@niometrics.com>>


^ permalink raw reply	[flat|nested] 3+ messages in thread
* [dpdk-dev] [PATCH 1/1] net/i40e: fix compilation failure on core-avx-i
@ 2021-06-22  9:37 Shahed Shaikh
  0 siblings, 0 replies; 3+ messages in thread
From: Shahed Shaikh @ 2021-06-22  9:37 UTC (permalink / raw)
  To: beilei.xing; +Cc: dev, Shahed Shaikh, stable

i40e_rxtx_vec_sse.c fails to compile with below configuration:
- #define RTE_LIBRTE_I40E_16BYTE_RX_DESC 1 in config/rte_config.h
- cpu=core-axv-i
- gcc which supports -mavx2 (e.g. gcc 4.8.5)

This is because commit 0604b1f2208f ("net/i40e: fix crash in AVX512")
added i40e_rxq_rearm_common() to i40e_rxtx_vec_common.h which is
included by i40e_rxtx_vec_sse.c.

This function is enabled for compilation if CC_AVX2_SUPPORT is defined.
As per drivers/net/i40e/meson.build, CC_AVX2_SUPPORT is defined when
either CPU supports __AVX2__ or compiler supports -mavx2 option.

So for given configuration, CC_AVX2_SUPPORT gets defined but we
don't pass -mavx2 explicitly to gcc while compiling i40e_rxtx_vec_sse.c.
Hence it fails due to avx2 specific code from i40e_rxq_rearm_command().

This patch tries to fix the compilation by moving
i40e_rxq_rearm_common() to a new header file which will only be
included by i40e_rxtx_vec_avx2.c and i40e_rxtx_vec_avx512.c.

Fixes: 0604b1f2208f ("net/i40e: fix crash in AVX512")
Cc: stable@dpdk.org

Signed-off-by: Shahed Shaikh <shaikh@niometrics.com>
---
 drivers/net/i40e/i40e_rxtx_vec_avx2.c       |   2 +-
 drivers/net/i40e/i40e_rxtx_vec_avx512.c     |   2 +-
 drivers/net/i40e/i40e_rxtx_vec_avx_common.h | 210 ++++++++++++++++++++
 drivers/net/i40e/i40e_rxtx_vec_common.h     | 201 -------------------
 4 files changed, 212 insertions(+), 203 deletions(-)
 create mode 100644 drivers/net/i40e/i40e_rxtx_vec_avx_common.h

diff --git a/drivers/net/i40e/i40e_rxtx_vec_avx2.c b/drivers/net/i40e/i40e_rxtx_vec_avx2.c
index 3b9eef91a9..2afbb71b75 100644
--- a/drivers/net/i40e/i40e_rxtx_vec_avx2.c
+++ b/drivers/net/i40e/i40e_rxtx_vec_avx2.c
@@ -10,7 +10,7 @@
 #include "base/i40e_type.h"
 #include "i40e_ethdev.h"
 #include "i40e_rxtx.h"
-#include "i40e_rxtx_vec_common.h"
+#include "i40e_rxtx_vec_avx_common.h"

 #include <rte_vect.h>

diff --git a/drivers/net/i40e/i40e_rxtx_vec_avx512.c b/drivers/net/i40e/i40e_rxtx_vec_avx512.c
index bd21d64223..ad225b0e54 100644
--- a/drivers/net/i40e/i40e_rxtx_vec_avx512.c
+++ b/drivers/net/i40e/i40e_rxtx_vec_avx512.c
@@ -10,7 +10,7 @@
 #include "base/i40e_type.h"
 #include "i40e_ethdev.h"
 #include "i40e_rxtx.h"
-#include "i40e_rxtx_vec_common.h"
+#include "i40e_rxtx_vec_avx_common.h"

 #include <rte_vect.h>

diff --git a/drivers/net/i40e/i40e_rxtx_vec_avx_common.h b/drivers/net/i40e/i40e_rxtx_vec_avx_common.h
new file mode 100644
index 0000000000..9f34e52efb
--- /dev/null
+++ b/drivers/net/i40e/i40e_rxtx_vec_avx_common.h
@@ -0,0 +1,210 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2021 Intel Corporation
+ */
+
+#ifndef _I40E_RXTX_VEC_AVX_COMMON_H_
+#define _I40E_RXTX_VEC_AVX_COMMON_H_
+
+#include "i40e_rxtx_vec_common.h"
+
+#ifndef __INTEL_COMPILER
+#pragma GCC diagnostic ignored "-Wcast-qual"
+#endif
+
+#ifdef CC_AVX2_SUPPORT
+static __rte_always_inline void
+i40e_rxq_rearm_common(struct i40e_rx_queue *rxq, __rte_unused bool avx512)
+{
+	int i;
+	uint16_t rx_id;
+	volatile union i40e_rx_desc *rxdp;
+	struct i40e_rx_entry *rxep = &rxq->sw_ring[rxq->rxrearm_start];
+
+	rxdp = rxq->rx_ring + rxq->rxrearm_start;
+
+	/* Pull 'n' more MBUFs into the software ring */
+	if (rte_mempool_get_bulk(rxq->mp,
+				 (void *)rxep,
+				 RTE_I40E_RXQ_REARM_THRESH) < 0) {
+		if (rxq->rxrearm_nb + RTE_I40E_RXQ_REARM_THRESH >=
+		    rxq->nb_rx_desc) {
+			__m128i dma_addr0;
+			dma_addr0 = _mm_setzero_si128();
+			for (i = 0; i < RTE_I40E_DESCS_PER_LOOP; i++) {
+				rxep[i].mbuf = &rxq->fake_mbuf;
+				_mm_store_si128((__m128i *)&rxdp[i].read,
+						dma_addr0);
+			}
+		}
+		rte_eth_devices[rxq->port_id].data->rx_mbuf_alloc_failed +=
+			RTE_I40E_RXQ_REARM_THRESH;
+		return;
+	}
+
+#ifndef RTE_LIBRTE_I40E_16BYTE_RX_DESC
+	struct rte_mbuf *mb0, *mb1;
+	__m128i dma_addr0, dma_addr1;
+	__m128i hdr_room = _mm_set_epi64x(RTE_PKTMBUF_HEADROOM,
+			RTE_PKTMBUF_HEADROOM);
+	/* Initialize the mbufs in vector, process 2 mbufs in one loop */
+	for (i = 0; i < RTE_I40E_RXQ_REARM_THRESH; i += 2, rxep += 2) {
+		__m128i vaddr0, vaddr1;
+
+		mb0 = rxep[0].mbuf;
+		mb1 = rxep[1].mbuf;
+
+		/* load buf_addr(lo 64bit) and buf_iova(hi 64bit) */
+		RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, buf_iova) !=
+				offsetof(struct rte_mbuf, buf_addr) + 8);
+		vaddr0 = _mm_loadu_si128((__m128i *)&mb0->buf_addr);
+		vaddr1 = _mm_loadu_si128((__m128i *)&mb1->buf_addr);
+
+		/* convert pa to dma_addr hdr/data */
+		dma_addr0 = _mm_unpackhi_epi64(vaddr0, vaddr0);
+		dma_addr1 = _mm_unpackhi_epi64(vaddr1, vaddr1);
+
+		/* add headroom to pa values */
+		dma_addr0 = _mm_add_epi64(dma_addr0, hdr_room);
+		dma_addr1 = _mm_add_epi64(dma_addr1, hdr_room);
+
+		/* flush desc with pa dma_addr */
+		_mm_store_si128((__m128i *)&rxdp++->read, dma_addr0);
+		_mm_store_si128((__m128i *)&rxdp++->read, dma_addr1);
+	}
+#else
+#ifdef CC_AVX512_SUPPORT
+	if (avx512) {
+		struct rte_mbuf *mb0, *mb1, *mb2, *mb3;
+		struct rte_mbuf *mb4, *mb5, *mb6, *mb7;
+		__m512i dma_addr0_3, dma_addr4_7;
+		__m512i hdr_room = _mm512_set1_epi64(RTE_PKTMBUF_HEADROOM);
+		/* Initialize the mbufs in vector, process 8 mbufs in one loop */
+		for (i = 0; i < RTE_I40E_RXQ_REARM_THRESH;
+				i += 8, rxep += 8, rxdp += 8) {
+			__m128i vaddr0, vaddr1, vaddr2, vaddr3;
+			__m128i vaddr4, vaddr5, vaddr6, vaddr7;
+			__m256i vaddr0_1, vaddr2_3;
+			__m256i vaddr4_5, vaddr6_7;
+			__m512i vaddr0_3, vaddr4_7;
+
+			mb0 = rxep[0].mbuf;
+			mb1 = rxep[1].mbuf;
+			mb2 = rxep[2].mbuf;
+			mb3 = rxep[3].mbuf;
+			mb4 = rxep[4].mbuf;
+			mb5 = rxep[5].mbuf;
+			mb6 = rxep[6].mbuf;
+			mb7 = rxep[7].mbuf;
+
+			/* load buf_addr(lo 64bit) and buf_iova(hi 64bit) */
+			RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, buf_iova) !=
+					offsetof(struct rte_mbuf, buf_addr) + 8);
+			vaddr0 = _mm_loadu_si128((__m128i *)&mb0->buf_addr);
+			vaddr1 = _mm_loadu_si128((__m128i *)&mb1->buf_addr);
+			vaddr2 = _mm_loadu_si128((__m128i *)&mb2->buf_addr);
+			vaddr3 = _mm_loadu_si128((__m128i *)&mb3->buf_addr);
+			vaddr4 = _mm_loadu_si128((__m128i *)&mb4->buf_addr);
+			vaddr5 = _mm_loadu_si128((__m128i *)&mb5->buf_addr);
+			vaddr6 = _mm_loadu_si128((__m128i *)&mb6->buf_addr);
+			vaddr7 = _mm_loadu_si128((__m128i *)&mb7->buf_addr);
+
+			/**
+			 * merge 0 & 1, by casting 0 to 256-bit and inserting 1
+			 * into the high lanes. Similarly for 2 & 3, and so on.
+			 */
+			vaddr0_1 =
+				_mm256_inserti128_si256(_mm256_castsi128_si256(vaddr0),
+							vaddr1, 1);
+			vaddr2_3 =
+				_mm256_inserti128_si256(_mm256_castsi128_si256(vaddr2),
+							vaddr3, 1);
+			vaddr4_5 =
+				_mm256_inserti128_si256(_mm256_castsi128_si256(vaddr4),
+							vaddr5, 1);
+			vaddr6_7 =
+				_mm256_inserti128_si256(_mm256_castsi128_si256(vaddr6),
+							vaddr7, 1);
+			vaddr0_3 =
+				_mm512_inserti64x4(_mm512_castsi256_si512(vaddr0_1),
+							vaddr2_3, 1);
+			vaddr4_7 =
+				_mm512_inserti64x4(_mm512_castsi256_si512(vaddr4_5),
+							vaddr6_7, 1);
+
+			/* convert pa to dma_addr hdr/data */
+			dma_addr0_3 = _mm512_unpackhi_epi64(vaddr0_3, vaddr0_3);
+			dma_addr4_7 = _mm512_unpackhi_epi64(vaddr4_7, vaddr4_7);
+
+			/* add headroom to pa values */
+			dma_addr0_3 = _mm512_add_epi64(dma_addr0_3, hdr_room);
+			dma_addr4_7 = _mm512_add_epi64(dma_addr4_7, hdr_room);
+
+			/* flush desc with pa dma_addr */
+			_mm512_store_si512((__m512i *)&rxdp->read, dma_addr0_3);
+			_mm512_store_si512((__m512i *)&(rxdp + 4)->read, dma_addr4_7);
+		}
+	} else
+#endif
+	{
+		struct rte_mbuf *mb0, *mb1, *mb2, *mb3;
+		__m256i dma_addr0_1, dma_addr2_3;
+		__m256i hdr_room = _mm256_set1_epi64x(RTE_PKTMBUF_HEADROOM);
+		/* Initialize the mbufs in vector, process 4 mbufs in one loop */
+		for (i = 0; i < RTE_I40E_RXQ_REARM_THRESH;
+				i += 4, rxep += 4, rxdp += 4) {
+			__m128i vaddr0, vaddr1, vaddr2, vaddr3;
+			__m256i vaddr0_1, vaddr2_3;
+
+			mb0 = rxep[0].mbuf;
+			mb1 = rxep[1].mbuf;
+			mb2 = rxep[2].mbuf;
+			mb3 = rxep[3].mbuf;
+
+			/* load buf_addr(lo 64bit) and buf_iova(hi 64bit) */
+			RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, buf_iova) !=
+					offsetof(struct rte_mbuf, buf_addr) + 8);
+			vaddr0 = _mm_loadu_si128((__m128i *)&mb0->buf_addr);
+			vaddr1 = _mm_loadu_si128((__m128i *)&mb1->buf_addr);
+			vaddr2 = _mm_loadu_si128((__m128i *)&mb2->buf_addr);
+			vaddr3 = _mm_loadu_si128((__m128i *)&mb3->buf_addr);
+
+			/*
+			 * merge 0 & 1, by casting 0 to 256-bit and inserting 1
+			 * into the high lanes. Similarly for 2 & 3
+			 */
+			vaddr0_1 = _mm256_inserti128_si256(
+					_mm256_castsi128_si256(vaddr0), vaddr1, 1);
+			vaddr2_3 = _mm256_inserti128_si256(
+					_mm256_castsi128_si256(vaddr2), vaddr3, 1);
+
+			/* convert pa to dma_addr hdr/data */
+			dma_addr0_1 = _mm256_unpackhi_epi64(vaddr0_1, vaddr0_1);
+			dma_addr2_3 = _mm256_unpackhi_epi64(vaddr2_3, vaddr2_3);
+
+			/* add headroom to pa values */
+			dma_addr0_1 = _mm256_add_epi64(dma_addr0_1, hdr_room);
+			dma_addr2_3 = _mm256_add_epi64(dma_addr2_3, hdr_room);
+
+			/* flush desc with pa dma_addr */
+			_mm256_store_si256((__m256i *)&rxdp->read, dma_addr0_1);
+			_mm256_store_si256((__m256i *)&(rxdp + 2)->read, dma_addr2_3);
+		}
+	}
+
+#endif
+
+	rxq->rxrearm_start += RTE_I40E_RXQ_REARM_THRESH;
+	if (rxq->rxrearm_start >= rxq->nb_rx_desc)
+		rxq->rxrearm_start = 0;
+
+	rxq->rxrearm_nb -= RTE_I40E_RXQ_REARM_THRESH;
+
+	rx_id = (uint16_t)((rxq->rxrearm_start == 0) ?
+			     (rxq->nb_rx_desc - 1) : (rxq->rxrearm_start - 1));
+
+	/* Update the tail pointer on the NIC */
+	I40E_PCI_REG_WC_WRITE(rxq->qrx_tail, rx_id);
+}
+#endif
+
+#endif
diff --git a/drivers/net/i40e/i40e_rxtx_vec_common.h b/drivers/net/i40e/i40e_rxtx_vec_common.h
index 16fcf0aec6..33cebbe88b 100644
--- a/drivers/net/i40e/i40e_rxtx_vec_common.h
+++ b/drivers/net/i40e/i40e_rxtx_vec_common.h
@@ -11,10 +11,6 @@
 #include "i40e_ethdev.h"
 #include "i40e_rxtx.h"

-#ifndef __INTEL_COMPILER
-#pragma GCC diagnostic ignored "-Wcast-qual"
-#endif
-
 static inline uint16_t
 reassemble_packets(struct i40e_rx_queue *rxq, struct rte_mbuf **rx_bufs,
 		   uint16_t nb_bufs, uint8_t *split_flags)
@@ -256,201 +252,4 @@ i40e_rx_vec_dev_conf_condition_check_default(struct rte_eth_dev *dev)
 	return -1;
 #endif
 }
-
-#ifdef CC_AVX2_SUPPORT
-static __rte_always_inline void
-i40e_rxq_rearm_common(struct i40e_rx_queue *rxq, __rte_unused bool avx512)
-{
-	int i;
-	uint16_t rx_id;
-	volatile union i40e_rx_desc *rxdp;
-	struct i40e_rx_entry *rxep = &rxq->sw_ring[rxq->rxrearm_start];
-
-	rxdp = rxq->rx_ring + rxq->rxrearm_start;
-
-	/* Pull 'n' more MBUFs into the software ring */
-	if (rte_mempool_get_bulk(rxq->mp,
-				 (void *)rxep,
-				 RTE_I40E_RXQ_REARM_THRESH) < 0) {
-		if (rxq->rxrearm_nb + RTE_I40E_RXQ_REARM_THRESH >=
-		    rxq->nb_rx_desc) {
-			__m128i dma_addr0;
-			dma_addr0 = _mm_setzero_si128();
-			for (i = 0; i < RTE_I40E_DESCS_PER_LOOP; i++) {
-				rxep[i].mbuf = &rxq->fake_mbuf;
-				_mm_store_si128((__m128i *)&rxdp[i].read,
-						dma_addr0);
-			}
-		}
-		rte_eth_devices[rxq->port_id].data->rx_mbuf_alloc_failed +=
-			RTE_I40E_RXQ_REARM_THRESH;
-		return;
-	}
-
-#ifndef RTE_LIBRTE_I40E_16BYTE_RX_DESC
-	struct rte_mbuf *mb0, *mb1;
-	__m128i dma_addr0, dma_addr1;
-	__m128i hdr_room = _mm_set_epi64x(RTE_PKTMBUF_HEADROOM,
-			RTE_PKTMBUF_HEADROOM);
-	/* Initialize the mbufs in vector, process 2 mbufs in one loop */
-	for (i = 0; i < RTE_I40E_RXQ_REARM_THRESH; i += 2, rxep += 2) {
-		__m128i vaddr0, vaddr1;
-
-		mb0 = rxep[0].mbuf;
-		mb1 = rxep[1].mbuf;
-
-		/* load buf_addr(lo 64bit) and buf_iova(hi 64bit) */
-		RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, buf_iova) !=
-				offsetof(struct rte_mbuf, buf_addr) + 8);
-		vaddr0 = _mm_loadu_si128((__m128i *)&mb0->buf_addr);
-		vaddr1 = _mm_loadu_si128((__m128i *)&mb1->buf_addr);
-
-		/* convert pa to dma_addr hdr/data */
-		dma_addr0 = _mm_unpackhi_epi64(vaddr0, vaddr0);
-		dma_addr1 = _mm_unpackhi_epi64(vaddr1, vaddr1);
-
-		/* add headroom to pa values */
-		dma_addr0 = _mm_add_epi64(dma_addr0, hdr_room);
-		dma_addr1 = _mm_add_epi64(dma_addr1, hdr_room);
-
-		/* flush desc with pa dma_addr */
-		_mm_store_si128((__m128i *)&rxdp++->read, dma_addr0);
-		_mm_store_si128((__m128i *)&rxdp++->read, dma_addr1);
-	}
-#else
-#ifdef CC_AVX512_SUPPORT
-	if (avx512) {
-		struct rte_mbuf *mb0, *mb1, *mb2, *mb3;
-		struct rte_mbuf *mb4, *mb5, *mb6, *mb7;
-		__m512i dma_addr0_3, dma_addr4_7;
-		__m512i hdr_room = _mm512_set1_epi64(RTE_PKTMBUF_HEADROOM);
-		/* Initialize the mbufs in vector, process 8 mbufs in one loop */
-		for (i = 0; i < RTE_I40E_RXQ_REARM_THRESH;
-				i += 8, rxep += 8, rxdp += 8) {
-			__m128i vaddr0, vaddr1, vaddr2, vaddr3;
-			__m128i vaddr4, vaddr5, vaddr6, vaddr7;
-			__m256i vaddr0_1, vaddr2_3;
-			__m256i vaddr4_5, vaddr6_7;
-			__m512i vaddr0_3, vaddr4_7;
-
-			mb0 = rxep[0].mbuf;
-			mb1 = rxep[1].mbuf;
-			mb2 = rxep[2].mbuf;
-			mb3 = rxep[3].mbuf;
-			mb4 = rxep[4].mbuf;
-			mb5 = rxep[5].mbuf;
-			mb6 = rxep[6].mbuf;
-			mb7 = rxep[7].mbuf;
-
-			/* load buf_addr(lo 64bit) and buf_iova(hi 64bit) */
-			RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, buf_iova) !=
-					offsetof(struct rte_mbuf, buf_addr) + 8);
-			vaddr0 = _mm_loadu_si128((__m128i *)&mb0->buf_addr);
-			vaddr1 = _mm_loadu_si128((__m128i *)&mb1->buf_addr);
-			vaddr2 = _mm_loadu_si128((__m128i *)&mb2->buf_addr);
-			vaddr3 = _mm_loadu_si128((__m128i *)&mb3->buf_addr);
-			vaddr4 = _mm_loadu_si128((__m128i *)&mb4->buf_addr);
-			vaddr5 = _mm_loadu_si128((__m128i *)&mb5->buf_addr);
-			vaddr6 = _mm_loadu_si128((__m128i *)&mb6->buf_addr);
-			vaddr7 = _mm_loadu_si128((__m128i *)&mb7->buf_addr);
-
-			/**
-			 * merge 0 & 1, by casting 0 to 256-bit and inserting 1
-			 * into the high lanes. Similarly for 2 & 3, and so on.
-			 */
-			vaddr0_1 =
-				_mm256_inserti128_si256(_mm256_castsi128_si256(vaddr0),
-							vaddr1, 1);
-			vaddr2_3 =
-				_mm256_inserti128_si256(_mm256_castsi128_si256(vaddr2),
-							vaddr3, 1);
-			vaddr4_5 =
-				_mm256_inserti128_si256(_mm256_castsi128_si256(vaddr4),
-							vaddr5, 1);
-			vaddr6_7 =
-				_mm256_inserti128_si256(_mm256_castsi128_si256(vaddr6),
-							vaddr7, 1);
-			vaddr0_3 =
-				_mm512_inserti64x4(_mm512_castsi256_si512(vaddr0_1),
-							vaddr2_3, 1);
-			vaddr4_7 =
-				_mm512_inserti64x4(_mm512_castsi256_si512(vaddr4_5),
-							vaddr6_7, 1);
-
-			/* convert pa to dma_addr hdr/data */
-			dma_addr0_3 = _mm512_unpackhi_epi64(vaddr0_3, vaddr0_3);
-			dma_addr4_7 = _mm512_unpackhi_epi64(vaddr4_7, vaddr4_7);
-
-			/* add headroom to pa values */
-			dma_addr0_3 = _mm512_add_epi64(dma_addr0_3, hdr_room);
-			dma_addr4_7 = _mm512_add_epi64(dma_addr4_7, hdr_room);
-
-			/* flush desc with pa dma_addr */
-			_mm512_store_si512((__m512i *)&rxdp->read, dma_addr0_3);
-			_mm512_store_si512((__m512i *)&(rxdp + 4)->read, dma_addr4_7);
-		}
-	} else
-#endif
-	{
-		struct rte_mbuf *mb0, *mb1, *mb2, *mb3;
-		__m256i dma_addr0_1, dma_addr2_3;
-		__m256i hdr_room = _mm256_set1_epi64x(RTE_PKTMBUF_HEADROOM);
-		/* Initialize the mbufs in vector, process 4 mbufs in one loop */
-		for (i = 0; i < RTE_I40E_RXQ_REARM_THRESH;
-				i += 4, rxep += 4, rxdp += 4) {
-			__m128i vaddr0, vaddr1, vaddr2, vaddr3;
-			__m256i vaddr0_1, vaddr2_3;
-
-			mb0 = rxep[0].mbuf;
-			mb1 = rxep[1].mbuf;
-			mb2 = rxep[2].mbuf;
-			mb3 = rxep[3].mbuf;
-
-			/* load buf_addr(lo 64bit) and buf_iova(hi 64bit) */
-			RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, buf_iova) !=
-					offsetof(struct rte_mbuf, buf_addr) + 8);
-			vaddr0 = _mm_loadu_si128((__m128i *)&mb0->buf_addr);
-			vaddr1 = _mm_loadu_si128((__m128i *)&mb1->buf_addr);
-			vaddr2 = _mm_loadu_si128((__m128i *)&mb2->buf_addr);
-			vaddr3 = _mm_loadu_si128((__m128i *)&mb3->buf_addr);
-
-			/*
-			 * merge 0 & 1, by casting 0 to 256-bit and inserting 1
-			 * into the high lanes. Similarly for 2 & 3
-			 */
-			vaddr0_1 = _mm256_inserti128_si256(
-					_mm256_castsi128_si256(vaddr0), vaddr1, 1);
-			vaddr2_3 = _mm256_inserti128_si256(
-					_mm256_castsi128_si256(vaddr2), vaddr3, 1);
-
-			/* convert pa to dma_addr hdr/data */
-			dma_addr0_1 = _mm256_unpackhi_epi64(vaddr0_1, vaddr0_1);
-			dma_addr2_3 = _mm256_unpackhi_epi64(vaddr2_3, vaddr2_3);
-
-			/* add headroom to pa values */
-			dma_addr0_1 = _mm256_add_epi64(dma_addr0_1, hdr_room);
-			dma_addr2_3 = _mm256_add_epi64(dma_addr2_3, hdr_room);
-
-			/* flush desc with pa dma_addr */
-			_mm256_store_si256((__m256i *)&rxdp->read, dma_addr0_1);
-			_mm256_store_si256((__m256i *)&(rxdp + 2)->read, dma_addr2_3);
-		}
-	}
-
-#endif
-
-	rxq->rxrearm_start += RTE_I40E_RXQ_REARM_THRESH;
-	if (rxq->rxrearm_start >= rxq->nb_rx_desc)
-		rxq->rxrearm_start = 0;
-
-	rxq->rxrearm_nb -= RTE_I40E_RXQ_REARM_THRESH;
-
-	rx_id = (uint16_t)((rxq->rxrearm_start == 0) ?
-			     (rxq->nb_rx_desc - 1) : (rxq->rxrearm_start - 1));
-
-	/* Update the tail pointer on the NIC */
-	I40E_PCI_REG_WC_WRITE(rxq->qrx_tail, rx_id);
-}
-#endif
-
 #endif
--
2.29.2


^ permalink raw reply	[flat|nested] 3+ messages in thread
end of thread, other threads:[~2021-08-02  0:17 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-07-14  1:06 [dpdk-dev] [PATCH 1/1] net/i40e: fix compilation failure on core-avx-i Lu, Wenzhuo
2021-08-02  0:17 ` Zhang, Qi Z
  -- strict thread matches above, loose matches on Subject: below --
2021-06-22  9:37 Shahed Shaikh

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).