[ceph.git] / ceph / src / spdk / dpdk / drivers / net / virtio / virtio_rxtx_simple_altivec.c

/* SPDX-License-Identifier: BSD-3-Clause
 * Copyright(c) 2010-2015 Intel Corporation
 * Copyright(C) 2019 IBM Corporation
 */

#include <stdint.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <errno.h>

#include <rte_altivec.h>
#include <rte_byteorder.h>
#include <rte_branch_prediction.h>
#include <rte_cycles.h>
#include <rte_ether.h>
#include <rte_ethdev_driver.h>
#include <rte_errno.h>
#include <rte_memory.h>
#include <rte_mempool.h>
#include <rte_malloc.h>
#include <rte_mbuf.h>
#include <rte_prefetch.h>
#include <rte_string_fns.h>

#include "virtio_rxtx_simple.h"

#define RTE_VIRTIO_DESC_PER_LOOP 8

/* virtio vPMD receive routine, only accept(nb_pkts >= RTE_VIRTIO_DESC_PER_LOOP)
 *
 * This routine is for non-mergeable RX, one desc for each guest buffer.
 * This routine is based on the RX ring layout optimization. Each entry in the
 * avail ring points to the desc with the same index in the desc ring and this
 * will never be changed in the driver.
 *
 * - nb_pkts < RTE_VIRTIO_DESC_PER_LOOP, just return no packet
 */
uint16_t
virtio_recv_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts,
	uint16_t nb_pkts)
{
	struct virtnet_rx *rxvq = rx_queue;
	struct virtqueue *vq = rxvq->vq;
	struct virtio_hw *hw = vq->hw;
	uint16_t nb_used, nb_total;
	uint16_t desc_idx;
	struct vring_used_elem *rused;
	struct rte_mbuf **sw_ring;
	struct rte_mbuf **sw_ring_end;
	struct rte_mbuf **ref_rx_pkts;
	uint16_t nb_pkts_received = 0;
	const vector unsigned char zero = {0};

	const vector unsigned char shuf_msk1 = {
		0xFF, 0xFF, 0xFF, 0xFF,	/* packet type */
		4, 5, 0xFF, 0xFF, /* vlan tci */
		4, 5,			/* dat len */
		0xFF, 0xFF,		/* vlan tci */
		0xFF, 0xFF, 0xFF, 0xFF
	};

	const vector unsigned char shuf_msk2 = {
		0xFF, 0xFF, 0xFF, 0xFF,	/* packet type */
		12, 13, 0xFF, 0xFF,	/* pkt len */
		12, 13,			/* dat len */
		0xFF, 0xFF,		/* vlan tci */
		0xFF, 0xFF, 0xFF, 0xFF
	};

	/*
	 * Subtract the header length.
	 *  In which case do we need the header length in used->len ?
	 */
	const vector unsigned short len_adjust = {
		0, 0,
		(uint16_t)-vq->hw->vtnet_hdr_size, 0,
		(uint16_t)-vq->hw->vtnet_hdr_size, 0,
		0, 0
	};

	if (unlikely(hw->started == 0))
		return nb_pkts_received;

	if (unlikely(nb_pkts < RTE_VIRTIO_DESC_PER_LOOP))
		return 0;

	nb_used = virtqueue_nused(vq);

	rte_compiler_barrier();

	if (unlikely(nb_used == 0))
		return 0;

	nb_pkts = RTE_ALIGN_FLOOR(nb_pkts, RTE_VIRTIO_DESC_PER_LOOP);
	nb_used = RTE_MIN(nb_used, nb_pkts);

	desc_idx = (uint16_t)(vq->vq_used_cons_idx & (vq->vq_nentries - 1));
	rused = &vq->vq_split.ring.used->ring[desc_idx];
	sw_ring  = &vq->sw_ring[desc_idx];
	sw_ring_end = &vq->sw_ring[vq->vq_nentries];

	rte_prefetch0(rused);

	if (vq->vq_free_cnt >= RTE_VIRTIO_VPMD_RX_REARM_THRESH) {
		virtio_rxq_rearm_vec(rxvq);
		if (unlikely(virtqueue_kick_prepare(vq)))
			virtqueue_notify(vq);
	}

	nb_total = nb_used;
	ref_rx_pkts = rx_pkts;
	for (nb_pkts_received = 0;
		nb_pkts_received < nb_total;) {
		vector unsigned char desc[RTE_VIRTIO_DESC_PER_LOOP / 2];
		vector unsigned char mbp[RTE_VIRTIO_DESC_PER_LOOP / 2];
		vector unsigned char pkt_mb[RTE_VIRTIO_DESC_PER_LOOP];

		mbp[0] = vec_vsx_ld(0, (unsigned char const *)(sw_ring + 0));
		desc[0] = vec_vsx_ld(0, (unsigned char const *)(rused + 0));
		*(vector unsigned char *)&rx_pkts[0] = mbp[0];

		mbp[1] = vec_vsx_ld(0, (unsigned char const *)(sw_ring + 2));
		desc[1] = vec_vsx_ld(0, (unsigned char const *)(rused + 2));
		*(vector unsigned char *)&rx_pkts[2] = mbp[1];

		mbp[2] = vec_vsx_ld(0, (unsigned char const *)(sw_ring + 4));
		desc[2] = vec_vsx_ld(0, (unsigned char const *)(rused + 4));
		*(vector unsigned char *)&rx_pkts[4] = mbp[2];

		mbp[3] = vec_vsx_ld(0, (unsigned char const *)(sw_ring + 6));
		desc[3] = vec_vsx_ld(0, (unsigned char const *)(rused + 6));
		*(vector unsigned char *)&rx_pkts[6] = mbp[3];

		pkt_mb[0] = vec_perm(desc[0], zero, shuf_msk1);
		pkt_mb[1] = vec_perm(desc[0], zero, shuf_msk2);
		pkt_mb[0] = (vector unsigned char)
			((vector unsigned short)pkt_mb[0] + len_adjust);
		pkt_mb[1] = (vector unsigned char)
			((vector unsigned short)pkt_mb[1] + len_adjust);
		*(vector unsigned char *)&rx_pkts[0]->rx_descriptor_fields1 =
			pkt_mb[0];
		*(vector unsigned char *)&rx_pkts[1]->rx_descriptor_fields1 =
			pkt_mb[1];

		pkt_mb[2] = vec_perm(desc[1], zero, shuf_msk1);
		pkt_mb[3] = vec_perm(desc[1], zero, shuf_msk2);
		pkt_mb[2] = (vector unsigned char)
			((vector unsigned short)pkt_mb[2] + len_adjust);
		pkt_mb[3] = (vector unsigned char)
			((vector unsigned short)pkt_mb[3] + len_adjust);
		*(vector unsigned char *)&rx_pkts[2]->rx_descriptor_fields1 =
			pkt_mb[2];
		*(vector unsigned char *)&rx_pkts[3]->rx_descriptor_fields1 =
			pkt_mb[3];

		pkt_mb[4] = vec_perm(desc[2], zero, shuf_msk1);
		pkt_mb[5] = vec_perm(desc[2], zero, shuf_msk2);
		pkt_mb[4] = (vector unsigned char)
			((vector unsigned short)pkt_mb[4] + len_adjust);
		pkt_mb[5] = (vector unsigned char)
			((vector unsigned short)pkt_mb[5] + len_adjust);
		*(vector unsigned char *)&rx_pkts[4]->rx_descriptor_fields1 =
			pkt_mb[4];
		*(vector unsigned char *)&rx_pkts[5]->rx_descriptor_fields1 =
			pkt_mb[5];

		pkt_mb[6] = vec_perm(desc[3], zero, shuf_msk1);
		pkt_mb[7] = vec_perm(desc[3], zero, shuf_msk2);
		pkt_mb[6] = (vector unsigned char)
			((vector unsigned short)pkt_mb[6] + len_adjust);
		pkt_mb[7] = (vector unsigned char)
			((vector unsigned short)pkt_mb[7] + len_adjust);
		*(vector unsigned char *)&rx_pkts[6]->rx_descriptor_fields1 =
			pkt_mb[6];
		*(vector unsigned char *)&rx_pkts[7]->rx_descriptor_fields1 =
			pkt_mb[7];

		if (unlikely(nb_used <= RTE_VIRTIO_DESC_PER_LOOP)) {
			if (sw_ring + nb_used <= sw_ring_end)
				nb_pkts_received += nb_used;
			else
				nb_pkts_received += sw_ring_end - sw_ring;
			break;
		} else {
			if (unlikely(sw_ring + RTE_VIRTIO_DESC_PER_LOOP >=
				sw_ring_end)) {
				nb_pkts_received += sw_ring_end - sw_ring;
				break;
			} else {
				nb_pkts_received += RTE_VIRTIO_DESC_PER_LOOP;

				rx_pkts += RTE_VIRTIO_DESC_PER_LOOP;
				sw_ring += RTE_VIRTIO_DESC_PER_LOOP;
				rused   += RTE_VIRTIO_DESC_PER_LOOP;
				nb_used -= RTE_VIRTIO_DESC_PER_LOOP;
			}
		}
	}

	vq->vq_used_cons_idx += nb_pkts_received;
	vq->vq_free_cnt += nb_pkts_received;
	rxvq->stats.packets += nb_pkts_received;
	for (nb_used = 0; nb_used < nb_pkts_received; nb_used++)
		virtio_update_packet_stats(&rxvq->stats, ref_rx_pkts[nb_used]);

	return nb_pkts_received;
}
Commit	Line	Data
f67539c2 TL	1	/* SPDX-License-Identifier: BSD-3-Clause
	2	* Copyright(c) 2010-2015 Intel Corporation
	3	* Copyright(C) 2019 IBM Corporation
	4	*/
	5
	6	#include <stdint.h>
	7	#include <stdio.h>
	8	#include <stdlib.h>
	9	#include <string.h>
	10	#include <errno.h>
	11
	12	#include <rte_altivec.h>
	13	#include <rte_byteorder.h>
	14	#include <rte_branch_prediction.h>
	15	#include <rte_cycles.h>
	16	#include <rte_ether.h>
	17	#include <rte_ethdev_driver.h>
	18	#include <rte_errno.h>
	19	#include <rte_memory.h>
	20	#include <rte_mempool.h>
	21	#include <rte_malloc.h>
	22	#include <rte_mbuf.h>
	23	#include <rte_prefetch.h>
	24	#include <rte_string_fns.h>
	25
	26	#include "virtio_rxtx_simple.h"
	27
	28	#define RTE_VIRTIO_DESC_PER_LOOP 8
	29
	30	/* virtio vPMD receive routine, only accept(nb_pkts >= RTE_VIRTIO_DESC_PER_LOOP)
	31	*
	32	* This routine is for non-mergeable RX, one desc for each guest buffer.
	33	* This routine is based on the RX ring layout optimization. Each entry in the
	34	* avail ring points to the desc with the same index in the desc ring and this
	35	* will never be changed in the driver.
	36	*
	37	* - nb_pkts < RTE_VIRTIO_DESC_PER_LOOP, just return no packet
	38	*/
	39	uint16_t
	40	virtio_recv_pkts_vec(void rx_queue, struct rte_mbuf *rx_pkts,
	41	uint16_t nb_pkts)
	42	{
	43	struct virtnet_rx *rxvq = rx_queue;
	44	struct virtqueue *vq = rxvq->vq;
	45	struct virtio_hw *hw = vq->hw;
	46	uint16_t nb_used, nb_total;
	47	uint16_t desc_idx;
	48	struct vring_used_elem *rused;
	49	struct rte_mbuf **sw_ring;
	50	struct rte_mbuf **sw_ring_end;
	51	struct rte_mbuf **ref_rx_pkts;
	52	uint16_t nb_pkts_received = 0;
	53	const vector unsigned char zero = {0};
	54
	55	const vector unsigned char shuf_msk1 = {
	56	0xFF, 0xFF, 0xFF, 0xFF, /* packet type */
	57	4, 5, 0xFF, 0xFF, /* vlan tci */
	58	4, 5, /* dat len */
	59	0xFF, 0xFF, /* vlan tci */
	60	0xFF, 0xFF, 0xFF, 0xFF
	61	};
	62
	63	const vector unsigned char shuf_msk2 = {
	64	0xFF, 0xFF, 0xFF, 0xFF, /* packet type */
65	12, 13, 0xFF, 0xFF, /* pkt len */
66	12, 13, /* dat len */
67	0xFF, 0xFF, /* vlan tci */
68	0xFF, 0xFF, 0xFF, 0xFF
69	};
70
71	/*
72	* Subtract the header length.
73	* In which case do we need the header length in used->len ?
74	*/
75	const vector unsigned short len_adjust = {
76	0, 0,
77	(uint16_t)-vq->hw->vtnet_hdr_size, 0,
78	(uint16_t)-vq->hw->vtnet_hdr_size, 0,
79	0, 0
80	};
81
82	if (unlikely(hw->started == 0))
83	return nb_pkts_received;
84
85	if (unlikely(nb_pkts < RTE_VIRTIO_DESC_PER_LOOP))
86	return 0;
87
88	nb_used = virtqueue_nused(vq);
89
90	rte_compiler_barrier();
91
92	if (unlikely(nb_used == 0))
93	return 0;
94
95	nb_pkts = RTE_ALIGN_FLOOR(nb_pkts, RTE_VIRTIO_DESC_PER_LOOP);
96	nb_used = RTE_MIN(nb_used, nb_pkts);
97
98	desc_idx = (uint16_t)(vq->vq_used_cons_idx & (vq->vq_nentries - 1));
99	rused = &vq->vq_split.ring.used->ring[desc_idx];
100	sw_ring = &vq->sw_ring[desc_idx];
101	sw_ring_end = &vq->sw_ring[vq->vq_nentries];
102
103	rte_prefetch0(rused);
104
105	if (vq->vq_free_cnt >= RTE_VIRTIO_VPMD_RX_REARM_THRESH) {
106	virtio_rxq_rearm_vec(rxvq);
107	if (unlikely(virtqueue_kick_prepare(vq)))
108	virtqueue_notify(vq);
109	}
110
111	nb_total = nb_used;
112	ref_rx_pkts = rx_pkts;
113	for (nb_pkts_received = 0;
114	nb_pkts_received < nb_total;) {
115	vector unsigned char desc[RTE_VIRTIO_DESC_PER_LOOP / 2];
116	vector unsigned char mbp[RTE_VIRTIO_DESC_PER_LOOP / 2];
117	vector unsigned char pkt_mb[RTE_VIRTIO_DESC_PER_LOOP];
118
119	mbp[0] = vec_vsx_ld(0, (unsigned char const *)(sw_ring + 0));
120	desc[0] = vec_vsx_ld(0, (unsigned char const *)(rused + 0));
121	(vector unsigned char )&rx_pkts[0] = mbp[0];
122
123	mbp[1] = vec_vsx_ld(0, (unsigned char const *)(sw_ring + 2));
124	desc[1] = vec_vsx_ld(0, (unsigned char const *)(rused + 2));
125	(vector unsigned char )&rx_pkts[2] = mbp[1];
126
127	mbp[2] = vec_vsx_ld(0, (unsigned char const *)(sw_ring + 4));
128	desc[2] = vec_vsx_ld(0, (unsigned char const *)(rused + 4));
129	(vector unsigned char )&rx_pkts[4] = mbp[2];
130
131	mbp[3] = vec_vsx_ld(0, (unsigned char const *)(sw_ring + 6));
132	desc[3] = vec_vsx_ld(0, (unsigned char const *)(rused + 6));
133	(vector unsigned char )&rx_pkts[6] = mbp[3];
134
135	pkt_mb[0] = vec_perm(desc[0], zero, shuf_msk1);
136	pkt_mb[1] = vec_perm(desc[0], zero, shuf_msk2);
137	pkt_mb[0] = (vector unsigned char)
138	((vector unsigned short)pkt_mb[0] + len_adjust);
139	pkt_mb[1] = (vector unsigned char)
140	((vector unsigned short)pkt_mb[1] + len_adjust);
141	(vector unsigned char )&rx_pkts[0]->rx_descriptor_fields1 =
142	pkt_mb[0];
143	(vector unsigned char )&rx_pkts[1]->rx_descriptor_fields1 =
144	pkt_mb[1];
145
146	pkt_mb[2] = vec_perm(desc[1], zero, shuf_msk1);
147	pkt_mb[3] = vec_perm(desc[1], zero, shuf_msk2);
148	pkt_mb[2] = (vector unsigned char)
149	((vector unsigned short)pkt_mb[2] + len_adjust);
150	pkt_mb[3] = (vector unsigned char)
151	((vector unsigned short)pkt_mb[3] + len_adjust);
152	(vector unsigned char )&rx_pkts[2]->rx_descriptor_fields1 =
153	pkt_mb[2];
154	(vector unsigned char )&rx_pkts[3]->rx_descriptor_fields1 =
155	pkt_mb[3];
156
157	pkt_mb[4] = vec_perm(desc[2], zero, shuf_msk1);
158	pkt_mb[5] = vec_perm(desc[2], zero, shuf_msk2);
159	pkt_mb[4] = (vector unsigned char)
160	((vector unsigned short)pkt_mb[4] + len_adjust);
161	pkt_mb[5] = (vector unsigned char)
162	((vector unsigned short)pkt_mb[5] + len_adjust);
163	(vector unsigned char )&rx_pkts[4]->rx_descriptor_fields1 =
164	pkt_mb[4];
165	(vector unsigned char )&rx_pkts[5]->rx_descriptor_fields1 =
166	pkt_mb[5];
167
168	pkt_mb[6] = vec_perm(desc[3], zero, shuf_msk1);
169	pkt_mb[7] = vec_perm(desc[3], zero, shuf_msk2);
170	pkt_mb[6] = (vector unsigned char)
171	((vector unsigned short)pkt_mb[6] + len_adjust);
172	pkt_mb[7] = (vector unsigned char)
173	((vector unsigned short)pkt_mb[7] + len_adjust);
174	(vector unsigned char )&rx_pkts[6]->rx_descriptor_fields1 =
175	pkt_mb[6];
176	(vector unsigned char )&rx_pkts[7]->rx_descriptor_fields1 =
177	pkt_mb[7];
178
179	if (unlikely(nb_used <= RTE_VIRTIO_DESC_PER_LOOP)) {
180	if (sw_ring + nb_used <= sw_ring_end)
181	nb_pkts_received += nb_used;
182	else
183	nb_pkts_received += sw_ring_end - sw_ring;
184	break;
185	} else {
186	if (unlikely(sw_ring + RTE_VIRTIO_DESC_PER_LOOP >=
187	sw_ring_end)) {
188	nb_pkts_received += sw_ring_end - sw_ring;
189	break;
190	} else {
191	nb_pkts_received += RTE_VIRTIO_DESC_PER_LOOP;
192
193	rx_pkts += RTE_VIRTIO_DESC_PER_LOOP;
194	sw_ring += RTE_VIRTIO_DESC_PER_LOOP;
195	rused += RTE_VIRTIO_DESC_PER_LOOP;
196	nb_used -= RTE_VIRTIO_DESC_PER_LOOP;
197	}
198	}
199	}
200
201	vq->vq_used_cons_idx += nb_pkts_received;
202	vq->vq_free_cnt += nb_pkts_received;
203	rxvq->stats.packets += nb_pkts_received;
204	for (nb_used = 0; nb_used < nb_pkts_received; nb_used++)
205	virtio_update_packet_stats(&rxvq->stats, ref_rx_pkts[nb_used]);
206
207	return nb_pkts_received;
208	}