struct i40e_rx_entry *rxep = &rxq->sw_ring[rxq->rxrearm_start];
struct rte_mbuf *mb0, *mb1;
- vector unsigned long hdr_room = (vector unsigned long){
+ __vector unsigned long hdr_room = (__vector unsigned long){
RTE_PKTMBUF_HEADROOM,
RTE_PKTMBUF_HEADROOM};
- vector unsigned long dma_addr0, dma_addr1;
+ __vector unsigned long dma_addr0, dma_addr1;
rxdp = rxq->rx_ring + rxq->rxrearm_start;
RTE_I40E_RXQ_REARM_THRESH) < 0) {
if (rxq->rxrearm_nb + RTE_I40E_RXQ_REARM_THRESH >=
rxq->nb_rx_desc) {
- dma_addr0 = (vector unsigned long){};
+ dma_addr0 = (__vector unsigned long){};
for (i = 0; i < RTE_I40E_DESCS_PER_LOOP; i++) {
rxep[i].mbuf = &rxq->fake_mbuf;
vec_st(dma_addr0, 0,
- (vector unsigned long *)&rxdp[i].read);
+ (__vector unsigned long *)&rxdp[i].read);
}
}
rte_eth_devices[rxq->port_id].data->rx_mbuf_alloc_failed +=
/* Initialize the mbufs in vector, process 2 mbufs in one loop */
for (i = 0; i < RTE_I40E_RXQ_REARM_THRESH; i += 2, rxep += 2) {
- vector unsigned long vaddr0, vaddr1;
+ __vector unsigned long vaddr0, vaddr1;
uintptr_t p0, p1;
mb0 = rxep[0].mbuf;
*(uint64_t *)p1 = rxq->mbuf_initializer;
/* load buf_addr(lo 64bit) and buf_iova(hi 64bit) */
- vaddr0 = vec_ld(0, (vector unsigned long *)&mb0->buf_addr);
- vaddr1 = vec_ld(0, (vector unsigned long *)&mb1->buf_addr);
+ vaddr0 = vec_ld(0, (__vector unsigned long *)&mb0->buf_addr);
+ vaddr1 = vec_ld(0, (__vector unsigned long *)&mb1->buf_addr);
/* convert pa to dma_addr hdr/data */
dma_addr0 = vec_mergel(vaddr0, vaddr0);
dma_addr1 = vec_add(dma_addr1, hdr_room);
/* flush desc with pa dma_addr */
- vec_st(dma_addr0, 0, (vector unsigned long *)&rxdp++->read);
- vec_st(dma_addr1, 0, (vector unsigned long *)&rxdp++->read);
+ vec_st(dma_addr0, 0, (__vector unsigned long *)&rxdp++->read);
+ vec_st(dma_addr1, 0, (__vector unsigned long *)&rxdp++->read);
}
rxq->rxrearm_start += RTE_I40E_RXQ_REARM_THRESH;
}
static inline void
-desc_to_olflags_v(vector unsigned long descs[4], struct rte_mbuf **rx_pkts)
+desc_to_olflags_v(__vector unsigned long descs[4], struct rte_mbuf **rx_pkts)
{
- vector unsigned int vlan0, vlan1, rss, l3_l4e;
+ __vector unsigned int vlan0, vlan1, rss, l3_l4e;
/* mask everything except RSS, flow director and VLAN flags
* bit2 is for VLAN tag, bit11 for flow director indication
* bit13:12 for RSS indication.
*/
- const vector unsigned int rss_vlan_msk = (vector unsigned int){
+ const __vector unsigned int rss_vlan_msk = (__vector unsigned int){
(int32_t)0x1c03804, (int32_t)0x1c03804,
(int32_t)0x1c03804, (int32_t)0x1c03804};
/* map rss and vlan type to rss hash and vlan flag */
- const vector unsigned char vlan_flags = (vector unsigned char){
+ const __vector unsigned char vlan_flags = (__vector unsigned char){
0, 0, 0, 0,
- PKT_RX_VLAN | PKT_RX_VLAN_STRIPPED, 0, 0, 0,
+ RTE_MBUF_F_RX_VLAN | RTE_MBUF_F_RX_VLAN_STRIPPED, 0, 0, 0,
0, 0, 0, 0,
0, 0, 0, 0};
- const vector unsigned char rss_flags = (vector unsigned char){
- 0, PKT_RX_FDIR, 0, 0,
- 0, 0, PKT_RX_RSS_HASH, PKT_RX_RSS_HASH | PKT_RX_FDIR,
+ const __vector unsigned char rss_flags = (__vector unsigned char){
+ 0, RTE_MBUF_F_RX_FDIR, 0, 0,
+ 0, 0, RTE_MBUF_F_RX_RSS_HASH, RTE_MBUF_F_RX_RSS_HASH | RTE_MBUF_F_RX_FDIR,
0, 0, 0, 0,
0, 0, 0, 0};
- const vector unsigned char l3_l4e_flags = (vector unsigned char){
+ const __vector unsigned char l3_l4e_flags = (__vector unsigned char){
0,
- PKT_RX_IP_CKSUM_BAD,
- PKT_RX_L4_CKSUM_BAD,
- PKT_RX_L4_CKSUM_BAD | PKT_RX_IP_CKSUM_BAD,
- PKT_RX_OUTER_IP_CKSUM_BAD,
- PKT_RX_OUTER_IP_CKSUM_BAD | PKT_RX_IP_CKSUM_BAD,
- PKT_RX_OUTER_IP_CKSUM_BAD | PKT_RX_L4_CKSUM_BAD,
- PKT_RX_OUTER_IP_CKSUM_BAD | PKT_RX_L4_CKSUM_BAD
- | PKT_RX_IP_CKSUM_BAD,
+ RTE_MBUF_F_RX_IP_CKSUM_BAD,
+ RTE_MBUF_F_RX_L4_CKSUM_BAD,
+ RTE_MBUF_F_RX_L4_CKSUM_BAD | RTE_MBUF_F_RX_IP_CKSUM_BAD,
+ RTE_MBUF_F_RX_OUTER_IP_CKSUM_BAD,
+ RTE_MBUF_F_RX_OUTER_IP_CKSUM_BAD | RTE_MBUF_F_RX_IP_CKSUM_BAD,
+ RTE_MBUF_F_RX_OUTER_IP_CKSUM_BAD | RTE_MBUF_F_RX_L4_CKSUM_BAD,
+ RTE_MBUF_F_RX_OUTER_IP_CKSUM_BAD | RTE_MBUF_F_RX_L4_CKSUM_BAD
+ | RTE_MBUF_F_RX_IP_CKSUM_BAD,
0, 0, 0, 0, 0, 0, 0, 0};
- vlan0 = (vector unsigned int)vec_mergel(descs[0], descs[1]);
- vlan1 = (vector unsigned int)vec_mergel(descs[2], descs[3]);
- vlan0 = (vector unsigned int)vec_mergeh(vlan0, vlan1);
+ vlan0 = (__vector unsigned int)vec_mergel(descs[0], descs[1]);
+ vlan1 = (__vector unsigned int)vec_mergel(descs[2], descs[3]);
+ vlan0 = (__vector unsigned int)vec_mergeh(vlan0, vlan1);
vlan1 = vec_and(vlan0, rss_vlan_msk);
- vlan0 = (vector unsigned int)vec_perm(vlan_flags,
- (vector unsigned char){},
- *(vector unsigned char *)&vlan1);
+ vlan0 = (__vector unsigned int)vec_perm(vlan_flags,
+ (__vector unsigned char){},
+ *(__vector unsigned char *)&vlan1);
- rss = vec_sr(vlan1, (vector unsigned int){11, 11, 11, 11});
- rss = (vector unsigned int)vec_perm(rss_flags, (vector unsigned char){},
- *(vector unsigned char *)&rss);
+ rss = vec_sr(vlan1, (__vector unsigned int){11, 11, 11, 11});
+ rss = (__vector unsigned int)vec_perm(rss_flags, (__vector unsigned char){},
+ *(__vector unsigned char *)&rss);
- l3_l4e = vec_sr(vlan1, (vector unsigned int){22, 22, 22, 22});
- l3_l4e = (vector unsigned int)vec_perm(l3_l4e_flags,
- (vector unsigned char){},
- *(vector unsigned char *)&l3_l4e);
+ l3_l4e = vec_sr(vlan1, (__vector unsigned int){22, 22, 22, 22});
+ l3_l4e = (__vector unsigned int)vec_perm(l3_l4e_flags,
+ (__vector unsigned char){},
+ *(__vector unsigned char *)&l3_l4e);
vlan0 = vec_or(vlan0, rss);
vlan0 = vec_or(vlan0, l3_l4e);
#define PKTLEN_SHIFT 10
static inline void
-desc_to_ptype_v(vector unsigned long descs[4], struct rte_mbuf **rx_pkts,
+desc_to_ptype_v(__vector unsigned long descs[4], struct rte_mbuf **rx_pkts,
uint32_t *ptype_tbl)
{
- vector unsigned long ptype0 = vec_mergel(descs[0], descs[1]);
- vector unsigned long ptype1 = vec_mergel(descs[2], descs[3]);
+ __vector unsigned long ptype0 = vec_mergel(descs[0], descs[1]);
+ __vector unsigned long ptype1 = vec_mergel(descs[2], descs[3]);
- ptype0 = vec_sr(ptype0, (vector unsigned long){30, 30});
- ptype1 = vec_sr(ptype1, (vector unsigned long){30, 30});
+ ptype0 = vec_sr(ptype0, (__vector unsigned long){30, 30});
+ ptype1 = vec_sr(ptype1, (__vector unsigned long){30, 30});
rx_pkts[0]->packet_type =
- ptype_tbl[(*(vector unsigned char *)&ptype0)[0]];
+ ptype_tbl[(*(__vector unsigned char *)&ptype0)[0]];
rx_pkts[1]->packet_type =
- ptype_tbl[(*(vector unsigned char *)&ptype0)[8]];
+ ptype_tbl[(*(__vector unsigned char *)&ptype0)[8]];
rx_pkts[2]->packet_type =
- ptype_tbl[(*(vector unsigned char *)&ptype1)[0]];
+ ptype_tbl[(*(__vector unsigned char *)&ptype1)[0]];
rx_pkts[3]->packet_type =
- ptype_tbl[(*(vector unsigned char *)&ptype1)[8]];
+ ptype_tbl[(*(__vector unsigned char *)&ptype1)[8]];
}
/**
uint16_t nb_pkts_recd;
int pos;
uint64_t var;
- vector unsigned char shuf_msk;
+ __vector unsigned char shuf_msk;
uint32_t *ptype_tbl = rxq->vsi->adapter->ptype_tbl;
- vector unsigned short crc_adjust = (vector unsigned short){
+ __vector unsigned short crc_adjust = (__vector unsigned short){
0, 0, /* ignore pkt_type field */
rxq->crc_len, /* sub crc on pkt_len */
0, /* ignore high-16bits of pkt_len */
rxq->crc_len, /* sub crc on data_len */
0, 0, 0 /* ignore non-length fields */
};
- vector unsigned long dd_check, eop_check;
+ __vector unsigned long dd_check, eop_check;
/* nb_pkts has to be floor-aligned to RTE_I40E_DESCS_PER_LOOP */
nb_pkts = RTE_ALIGN_FLOOR(nb_pkts, RTE_I40E_DESCS_PER_LOOP);
return 0;
/* 4 packets DD mask */
- dd_check = (vector unsigned long){0x0000000100000001ULL,
+ dd_check = (__vector unsigned long){0x0000000100000001ULL,
0x0000000100000001ULL};
/* 4 packets EOP mask */
- eop_check = (vector unsigned long){0x0000000200000002ULL,
+ eop_check = (__vector unsigned long){0x0000000200000002ULL,
0x0000000200000002ULL};
/* mask to shuffle from desc. to mbuf */
- shuf_msk = (vector unsigned char){
+ shuf_msk = (__vector unsigned char){
0xFF, 0xFF, /* pkt_type set as unknown */
0xFF, 0xFF, /* pkt_type set as unknown */
14, 15, /* octet 15~14, low 16 bits pkt_len */
for (pos = 0, nb_pkts_recd = 0; pos < nb_pkts;
pos += RTE_I40E_DESCS_PER_LOOP,
rxdp += RTE_I40E_DESCS_PER_LOOP) {
- vector unsigned long descs[RTE_I40E_DESCS_PER_LOOP];
- vector unsigned char pkt_mb1, pkt_mb2, pkt_mb3, pkt_mb4;
- vector unsigned short staterr, sterr_tmp1, sterr_tmp2;
- vector unsigned long mbp1, mbp2; /* two mbuf pointer
+ __vector unsigned long descs[RTE_I40E_DESCS_PER_LOOP];
+ __vector unsigned char pkt_mb1, pkt_mb2, pkt_mb3, pkt_mb4;
+ __vector unsigned short staterr, sterr_tmp1, sterr_tmp2;
+ __vector unsigned long mbp1, mbp2; /* two mbuf pointer
* in one XMM reg.
*/
/* B.1 load 2 mbuf point */
- mbp1 = *(vector unsigned long *)&sw_ring[pos];
+ mbp1 = *(__vector unsigned long *)&sw_ring[pos];
/* Read desc statuses backwards to avoid race condition */
/* A.1 load desc[3] */
- descs[3] = *(vector unsigned long *)(rxdp + 3);
+ descs[3] = *(__vector unsigned long *)(rxdp + 3);
rte_compiler_barrier();
/* B.2 copy 2 mbuf point into rx_pkts */
- *(vector unsigned long *)&rx_pkts[pos] = mbp1;
+ *(__vector unsigned long *)&rx_pkts[pos] = mbp1;
/* B.1 load 2 mbuf point */
- mbp2 = *(vector unsigned long *)&sw_ring[pos + 2];
+ mbp2 = *(__vector unsigned long *)&sw_ring[pos + 2];
/* A.1 load desc[2-0] */
- descs[2] = *(vector unsigned long *)(rxdp + 2);
+ descs[2] = *(__vector unsigned long *)(rxdp + 2);
rte_compiler_barrier();
- descs[1] = *(vector unsigned long *)(rxdp + 1);
+ descs[1] = *(__vector unsigned long *)(rxdp + 1);
rte_compiler_barrier();
- descs[0] = *(vector unsigned long *)(rxdp);
+ descs[0] = *(__vector unsigned long *)(rxdp);
/* B.2 copy 2 mbuf point into rx_pkts */
- *(vector unsigned long *)&rx_pkts[pos + 2] = mbp2;
+ *(__vector unsigned long *)&rx_pkts[pos + 2] = mbp2;
if (split_packet) {
rte_mbuf_prefetch_part2(rx_pkts[pos]);
rte_compiler_barrier();
/* pkt 3,4 shift the pktlen field to be 16-bit aligned*/
- const vector unsigned int len3 = vec_sl(
- vec_ld(0, (vector unsigned int *)&descs[3]),
- (vector unsigned int){0, 0, 0, PKTLEN_SHIFT});
+ const __vector unsigned int len3 = vec_sl(
+ vec_ld(0, (__vector unsigned int *)&descs[3]),
+ (__vector unsigned int){0, 0, 0, PKTLEN_SHIFT});
- const vector unsigned int len2 = vec_sl(
- vec_ld(0, (vector unsigned int *)&descs[2]),
- (vector unsigned int){0, 0, 0, PKTLEN_SHIFT});
+ const __vector unsigned int len2 = vec_sl(
+ vec_ld(0, (__vector unsigned int *)&descs[2]),
+ (__vector unsigned int){0, 0, 0, PKTLEN_SHIFT});
/* merge the now-aligned packet length fields back in */
- descs[3] = (vector unsigned long)len3;
- descs[2] = (vector unsigned long)len2;
+ descs[3] = (__vector unsigned long)len3;
+ descs[2] = (__vector unsigned long)len2;
/* D.1 pkt 3,4 convert format from desc to pktmbuf */
- pkt_mb4 = vec_perm((vector unsigned char)descs[3],
- (vector unsigned char){}, shuf_msk);
- pkt_mb3 = vec_perm((vector unsigned char)descs[2],
- (vector unsigned char){}, shuf_msk);
+ pkt_mb4 = vec_perm((__vector unsigned char)descs[3],
+ (__vector unsigned char){}, shuf_msk);
+ pkt_mb3 = vec_perm((__vector unsigned char)descs[2],
+ (__vector unsigned char){}, shuf_msk);
/* C.1 4=>2 filter staterr info only */
- sterr_tmp2 = vec_mergel((vector unsigned short)descs[3],
- (vector unsigned short)descs[2]);
+ sterr_tmp2 = vec_mergel((__vector unsigned short)descs[3],
+ (__vector unsigned short)descs[2]);
/* C.1 4=>2 filter staterr info only */
- sterr_tmp1 = vec_mergel((vector unsigned short)descs[1],
- (vector unsigned short)descs[0]);
+ sterr_tmp1 = vec_mergel((__vector unsigned short)descs[1],
+ (__vector unsigned short)descs[0]);
/* D.2 pkt 3,4 set in_port/nb_seg and remove crc */
- pkt_mb4 = (vector unsigned char)vec_sub(
- (vector unsigned short)pkt_mb4, crc_adjust);
- pkt_mb3 = (vector unsigned char)vec_sub(
- (vector unsigned short)pkt_mb3, crc_adjust);
+ pkt_mb4 = (__vector unsigned char)vec_sub(
+ (__vector unsigned short)pkt_mb4, crc_adjust);
+ pkt_mb3 = (__vector unsigned char)vec_sub(
+ (__vector unsigned short)pkt_mb3, crc_adjust);
/* pkt 1,2 shift the pktlen field to be 16-bit aligned*/
- const vector unsigned int len1 = vec_sl(
- vec_ld(0, (vector unsigned int *)&descs[1]),
- (vector unsigned int){0, 0, 0, PKTLEN_SHIFT});
- const vector unsigned int len0 = vec_sl(
- vec_ld(0, (vector unsigned int *)&descs[0]),
- (vector unsigned int){0, 0, 0, PKTLEN_SHIFT});
+ const __vector unsigned int len1 = vec_sl(
+ vec_ld(0, (__vector unsigned int *)&descs[1]),
+ (__vector unsigned int){0, 0, 0, PKTLEN_SHIFT});
+ const __vector unsigned int len0 = vec_sl(
+ vec_ld(0, (__vector unsigned int *)&descs[0]),
+ (__vector unsigned int){0, 0, 0, PKTLEN_SHIFT});
/* merge the now-aligned packet length fields back in */
- descs[1] = (vector unsigned long)len1;
- descs[0] = (vector unsigned long)len0;
+ descs[1] = (__vector unsigned long)len1;
+ descs[0] = (__vector unsigned long)len0;
/* D.1 pkt 1,2 convert format from desc to pktmbuf */
- pkt_mb2 = vec_perm((vector unsigned char)descs[1],
- (vector unsigned char){}, shuf_msk);
- pkt_mb1 = vec_perm((vector unsigned char)descs[0],
- (vector unsigned char){}, shuf_msk);
+ pkt_mb2 = vec_perm((__vector unsigned char)descs[1],
+ (__vector unsigned char){}, shuf_msk);
+ pkt_mb1 = vec_perm((__vector unsigned char)descs[0],
+ (__vector unsigned char){}, shuf_msk);
/* C.2 get 4 pkts staterr value */
- staterr = (vector unsigned short)vec_mergeh(
+ staterr = (__vector unsigned short)vec_mergeh(
sterr_tmp1, sterr_tmp2);
/* D.3 copy final 3,4 data to rx_pkts */
vec_st(pkt_mb4, 0,
- (vector unsigned char *)&rx_pkts[pos + 3]
+ (__vector unsigned char *)&rx_pkts[pos + 3]
->rx_descriptor_fields1
);
vec_st(pkt_mb3, 0,
- (vector unsigned char *)&rx_pkts[pos + 2]
+ (__vector unsigned char *)&rx_pkts[pos + 2]
->rx_descriptor_fields1
);
/* D.2 pkt 1,2 set in_port/nb_seg and remove crc */
- pkt_mb2 = (vector unsigned char)vec_sub(
- (vector unsigned short)pkt_mb2, crc_adjust);
- pkt_mb1 = (vector unsigned char)vec_sub(
- (vector unsigned short)pkt_mb1, crc_adjust);
+ pkt_mb2 = (__vector unsigned char)vec_sub(
+ (__vector unsigned short)pkt_mb2, crc_adjust);
+ pkt_mb1 = (__vector unsigned char)vec_sub(
+ (__vector unsigned short)pkt_mb1, crc_adjust);
/* C* extract and record EOP bit */
if (split_packet) {
- vector unsigned char eop_shuf_mask =
- (vector unsigned char){
+ __vector unsigned char eop_shuf_mask =
+ (__vector unsigned char){
0xFF, 0xFF, 0xFF, 0xFF,
0xFF, 0xFF, 0xFF, 0xFF,
0xFF, 0xFF, 0xFF, 0xFF,
};
/* and with mask to extract bits, flipping 1-0 */
- vector unsigned char eop_bits = vec_and(
- (vector unsigned char)vec_nor(staterr, staterr),
- (vector unsigned char)eop_check);
+ __vector unsigned char eop_bits = vec_and(
+ (__vector unsigned char)vec_nor(staterr, staterr),
+ (__vector unsigned char)eop_check);
/* the staterr values are not in order, as the count
* of dd bits doesn't care. However, for end of
* packet tracking, we do care, so shuffle. This also
* compresses the 32-bit values to 8-bit
*/
- eop_bits = vec_perm(eop_bits, (vector unsigned char){},
+ eop_bits = vec_perm(eop_bits, (__vector unsigned char){},
eop_shuf_mask);
/* store the resulting 32-bit value */
*split_packet = (vec_ld(0,
- (vector unsigned int *)&eop_bits))[0];
+ (__vector unsigned int *)&eop_bits))[0];
split_packet += RTE_I40E_DESCS_PER_LOOP;
/* zero-out next pointers */
}
/* C.3 calc available number of desc */
- staterr = vec_and(staterr, (vector unsigned short)dd_check);
+ staterr = vec_and(staterr, (__vector unsigned short)dd_check);
/* D.3 copy final 1,2 data to rx_pkts */
vec_st(pkt_mb2, 0,
- (vector unsigned char *)&rx_pkts[pos + 1]
+ (__vector unsigned char *)&rx_pkts[pos + 1]
->rx_descriptor_fields1
);
vec_st(pkt_mb1, 0,
- (vector unsigned char *)&rx_pkts[pos]->rx_descriptor_fields1
+ (__vector unsigned char *)&rx_pkts[pos]->rx_descriptor_fields1
);
desc_to_ptype_v(descs, &rx_pkts[pos], ptype_tbl);
desc_to_olflags_v(descs, &rx_pkts[pos]);
- /* C.4 calc avaialbe number of desc */
+ /* C.4 calc available number of desc */
var = __builtin_popcountll((vec_ld(0,
- (vector unsigned long *)&staterr)[0]));
+ (__vector unsigned long *)&staterr)[0]));
nb_pkts_recd += var;
if (likely(var != RTE_I40E_DESCS_PER_LOOP))
break;
((uint64_t)flags << I40E_TXD_QW1_CMD_SHIFT) |
((uint64_t)pkt->data_len << I40E_TXD_QW1_TX_BUF_SZ_SHIFT));
- vector unsigned long descriptor = (vector unsigned long){
+ __vector unsigned long descriptor = (__vector unsigned long){
pkt->buf_iova + pkt->data_off, high_qw};
- *(vector unsigned long *)txdp = descriptor;
+ *(__vector unsigned long *)txdp = descriptor;
}
static inline void
uint64_t rs = I40E_TX_DESC_CMD_RS | I40E_TD_CMD;
int i;
- /* cross rx_thresh boundary is not allowed */
- nb_pkts = RTE_MIN(nb_pkts, txq->tx_rs_thresh);
-
if (txq->nb_tx_free < txq->tx_free_thresh)
i40e_tx_free_bufs(txq);
git.droids-corp.org / dpdk.git / blobdiff