/* SPDX-License-Identifier: BSD-3-Clause */
-/* Copyright(c) 2019-2020 Broadcom All rights reserved. */
+/* Copyright(c) 2019-2021 Broadcom All rights reserved. */
#include <inttypes.h>
#include <stdbool.h>
uint32_t tmp, of; \
\
of = _mm_extract_epi32((rss_flags), (pi)) | \
- bnxt_ol_flags_table[_mm_extract_epi32((ol_index), (pi))]; \
+ rxr->ol_flags_table[_mm_extract_epi32((ol_index), (pi))]; \
\
tmp = _mm_extract_epi32((errors), (pi)); \
if (tmp) \
- of |= bnxt_ol_flags_err_table[tmp]; \
+ of |= rxr->ol_flags_err_table[tmp]; \
(ol_flags) = of; \
}
static inline void
descs_to_mbufs(__m128i mm_rxcmp[4], __m128i mm_rxcmp1[4],
- __m128i mbuf_init, struct rte_mbuf **mbuf)
+ __m128i mbuf_init, struct rte_mbuf **mbuf,
+ struct bnxt_rx_ring_info *rxr)
{
const __m128i shuf_msk =
_mm_set_epi8(15, 14, 13, 12, /* rss */
const __m128i flags_type_mask =
_mm_set1_epi32(RX_PKT_CMPL_FLAGS_ITYPE_MASK);
const __m128i flags2_mask1 =
- _mm_set1_epi32(RX_PKT_CMPL_FLAGS2_META_FORMAT_VLAN |
- RX_PKT_CMPL_FLAGS2_T_IP_CS_CALC);
+ _mm_set1_epi32(CMPL_FLAGS2_VLAN_TUN_MSK);
const __m128i flags2_mask2 =
_mm_set1_epi32(RX_PKT_CMPL_FLAGS2_IP_TYPE);
const __m128i rss_mask =
_mm_set1_epi32(RX_PKT_CMPL_FLAGS_RSS_VALID);
__m128i t0, t1, flags_type, flags2, index, errors, rss_flags;
- __m128i ptype_idx;
+ __m128i ptype_idx, is_tunnel;
uint32_t ol_flags;
+ /* Validate ptype table indexing at build time. */
+ bnxt_check_ptype_constants();
+
/* Compute packet type table indexes for four packets */
t0 = _mm_unpacklo_epi32(mm_rxcmp[0], mm_rxcmp[1]);
t1 = _mm_unpacklo_epi32(mm_rxcmp[2], mm_rxcmp[3]);
flags_type = _mm_unpacklo_epi64(t0, t1);
- ptype_idx =
- _mm_srli_epi32(_mm_and_si128(flags_type, flags_type_mask), 9);
+ ptype_idx = _mm_srli_epi32(_mm_and_si128(flags_type, flags_type_mask),
+ RX_PKT_CMPL_FLAGS_ITYPE_SFT - BNXT_PTYPE_TBL_TYPE_SFT);
t0 = _mm_unpacklo_epi32(mm_rxcmp1[0], mm_rxcmp1[1]);
t1 = _mm_unpacklo_epi32(mm_rxcmp1[2], mm_rxcmp1[3]);
flags2 = _mm_unpacklo_epi64(t0, t1);
ptype_idx = _mm_or_si128(ptype_idx,
- _mm_srli_epi32(_mm_and_si128(flags2, flags2_mask1), 2));
+ _mm_srli_epi32(_mm_and_si128(flags2, flags2_mask1),
+ RX_PKT_CMPL_FLAGS2_META_FORMAT_SFT -
+ BNXT_PTYPE_TBL_VLAN_SFT));
ptype_idx = _mm_or_si128(ptype_idx,
- _mm_srli_epi32(_mm_and_si128(flags2, flags2_mask2), 7));
+ _mm_srli_epi32(_mm_and_si128(flags2, flags2_mask2),
+ RX_PKT_CMPL_FLAGS2_IP_TYPE_SFT -
+ BNXT_PTYPE_TBL_IP_VER_SFT));
/* Extract RSS valid flags for four packets. */
rss_flags = _mm_srli_epi32(_mm_and_si128(flags_type, rss_mask), 9);
t1 = _mm_unpackhi_epi32(mm_rxcmp1[2], mm_rxcmp1[3]);
/* Compute ol_flags and checksum error indexes for four packets. */
+ is_tunnel = _mm_and_si128(flags2, _mm_set1_epi32(4));
+ is_tunnel = _mm_slli_epi32(is_tunnel, 3);
flags2 = _mm_and_si128(flags2, _mm_set1_epi32(0x1F));
errors = _mm_srli_epi32(_mm_unpacklo_epi64(t0, t1), 4);
errors = _mm_and_si128(errors, flags2);
index = _mm_andnot_si128(errors, flags2);
+ errors = _mm_or_si128(errors, _mm_srli_epi32(is_tunnel, 1));
+ index = _mm_or_si128(index, is_tunnel);
/* Update mbuf rearm_data for four packets. */
GET_OL_FLAGS(rss_flags, index, errors, 0, ol_flags);
_mm_store_si128((void *)&mbuf[3]->rx_descriptor_fields1, t0);
}
-uint16_t
-bnxt_recv_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts,
- uint16_t nb_pkts)
+static uint16_t
+recv_burst_vec_sse(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
{
struct bnxt_rx_queue *rxq = rx_queue;
const __m128i mbuf_init = _mm_set_epi64x(0, rxq->mbuf_initializer);
if (rxq->rxrearm_nb >= rxq->rx_free_thresh)
bnxt_rxq_rearm(rxq, rxr);
- /* Return no more than RTE_BNXT_MAX_RX_BURST per call. */
- nb_pkts = RTE_MIN(nb_pkts, RTE_BNXT_MAX_RX_BURST);
-
cons = raw_cons & (cp_ring_size - 1);
mbcons = (raw_cons / 2) & (rx_ring_size - 1);
* maximum number of packets to receive to be a multiple of the per-
* loop count.
*/
- if (nb_pkts < RTE_BNXT_DESCS_PER_LOOP)
- desc_valid_mask >>= 16 * (RTE_BNXT_DESCS_PER_LOOP - nb_pkts);
- else
- nb_pkts = RTE_ALIGN_FLOOR(nb_pkts, RTE_BNXT_DESCS_PER_LOOP);
+ if (nb_pkts < BNXT_RX_DESCS_PER_LOOP_VEC128) {
+ desc_valid_mask >>=
+ 16 * (BNXT_RX_DESCS_PER_LOOP_VEC128 - nb_pkts);
+ } else {
+ nb_pkts =
+ RTE_ALIGN_FLOOR(nb_pkts, BNXT_RX_DESCS_PER_LOOP_VEC128);
+ }
/* Handle RX burst request */
- for (i = 0; i < nb_pkts; i += RTE_BNXT_DESCS_PER_LOOP,
- cons += RTE_BNXT_DESCS_PER_LOOP * 2,
- mbcons += RTE_BNXT_DESCS_PER_LOOP) {
- __m128i rxcmp1[RTE_BNXT_DESCS_PER_LOOP];
- __m128i rxcmp[RTE_BNXT_DESCS_PER_LOOP];
+ for (i = 0; i < nb_pkts; i += BNXT_RX_DESCS_PER_LOOP_VEC128,
+ cons += BNXT_RX_DESCS_PER_LOOP_VEC128 * 2,
+ mbcons += BNXT_RX_DESCS_PER_LOOP_VEC128) {
+ __m128i rxcmp1[BNXT_RX_DESCS_PER_LOOP_VEC128];
+ __m128i rxcmp[BNXT_RX_DESCS_PER_LOOP_VEC128];
__m128i tmp0, tmp1, info3_v;
uint32_t num_valid;
#endif
/* Prefetch four descriptor pairs for next iteration. */
- if (i + RTE_BNXT_DESCS_PER_LOOP < nb_pkts) {
+ if (i + BNXT_RX_DESCS_PER_LOOP_VEC128 < nb_pkts) {
rte_prefetch0(&cp_desc_ring[cons + 8]);
rte_prefetch0(&cp_desc_ring[cons + 12]);
}
valid = _mm_cvtsi128_si64(_mm_packs_epi32(info3_v, info3_v));
num_valid = __builtin_popcountll(valid & desc_valid_mask);
- switch (num_valid) {
- case 4:
- rxr->rx_buf_ring[mbcons + 3] = NULL;
- /* FALLTHROUGH */
- case 3:
- rxr->rx_buf_ring[mbcons + 2] = NULL;
- /* FALLTHROUGH */
- case 2:
- rxr->rx_buf_ring[mbcons + 1] = NULL;
- /* FALLTHROUGH */
- case 1:
- rxr->rx_buf_ring[mbcons + 0] = NULL;
+ if (num_valid == 0)
break;
- case 0:
- goto out;
- }
- descs_to_mbufs(rxcmp, rxcmp1, mbuf_init, &rx_pkts[nb_rx_pkts]);
+ descs_to_mbufs(rxcmp, rxcmp1, mbuf_init, &rx_pkts[nb_rx_pkts],
+ rxr);
nb_rx_pkts += num_valid;
- if (num_valid < RTE_BNXT_DESCS_PER_LOOP)
+ if (num_valid < BNXT_RX_DESCS_PER_LOOP_VEC128)
break;
}
-out:
if (nb_rx_pkts) {
rxr->rx_raw_prod = RING_ADV(rxr->rx_raw_prod, nb_rx_pkts);
return nb_rx_pkts;
}
+uint16_t
+bnxt_recv_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
+{
+ uint16_t cnt = 0;
+
+ while (nb_pkts > RTE_BNXT_MAX_RX_BURST) {
+ uint16_t burst;
+
+ burst = recv_burst_vec_sse(rx_queue, rx_pkts + cnt,
+ RTE_BNXT_MAX_RX_BURST);
+
+ cnt += burst;
+ nb_pkts -= burst;
+
+ if (burst < RTE_BNXT_MAX_RX_BURST)
+ return cnt;
+ }
+
+ return cnt + recv_burst_vec_sse(rx_queue, rx_pkts + cnt, nb_pkts);
+}
+
static void
bnxt_handle_tx_cp_vec(struct bnxt_tx_queue *txq)
{
static inline void
bnxt_xmit_one(struct rte_mbuf *mbuf, struct tx_bd_long *txbd,
- struct bnxt_sw_tx_bd *tx_buf)
+ struct rte_mbuf **tx_buf)
{
__m128i desc;
- tx_buf->mbuf = mbuf;
- tx_buf->nr_bds = 1;
+ *tx_buf = mbuf;
desc = _mm_set_epi64x(mbuf->buf_iova + mbuf->data_off,
bnxt_xmit_flags_len(mbuf->data_len,
struct bnxt_tx_ring_info *txr = txq->tx_ring;
uint16_t tx_prod, tx_raw_prod = txr->tx_raw_prod;
struct tx_bd_long *txbd;
- struct bnxt_sw_tx_bd *tx_buf;
+ struct rte_mbuf **tx_buf;
uint16_t to_send;
tx_prod = RING_IDX(txr->tx_ring_struct, tx_raw_prod);
/* Handle TX burst request */
to_send = nb_pkts;
- while (to_send >= RTE_BNXT_DESCS_PER_LOOP) {
+ while (to_send >= BNXT_TX_DESCS_PER_LOOP) {
/* Prefetch next transmit buffer descriptors. */
rte_prefetch0(txbd + 4);
rte_prefetch0(txbd + 7);
bnxt_xmit_one(tx_pkts[2], txbd++, tx_buf++);
bnxt_xmit_one(tx_pkts[3], txbd++, tx_buf++);
- to_send -= RTE_BNXT_DESCS_PER_LOOP;
- tx_pkts += RTE_BNXT_DESCS_PER_LOOP;
+ to_send -= BNXT_TX_DESCS_PER_LOOP;
+ tx_pkts += BNXT_TX_DESCS_PER_LOOP;
}
while (to_send) {
git.droids-corp.org / dpdk.git / blobdiff