X-Git-Url: http://git.droids-corp.org/?a=blobdiff_plain;f=drivers%2Fnet%2Fixgbe%2Fixgbe_rxtx_vec_sse.c;h=e704a7f355e764642661a87e9e9d3681d4fc9b31;hb=3b212853e478e45490668ea054ec351da1fbda43;hp=203f092dfae834f4c2d2df07a7073be53caf98f3;hpb=631a2f530b6c1ea17e1c504f6170e6127439509a;p=dpdk.git diff --git a/drivers/net/ixgbe/ixgbe_rxtx_vec_sse.c b/drivers/net/ixgbe/ixgbe_rxtx_vec_sse.c index 203f092dfa..e704a7f355 100644 --- a/drivers/net/ixgbe/ixgbe_rxtx_vec_sse.c +++ b/drivers/net/ixgbe/ixgbe_rxtx_vec_sse.c @@ -216,26 +216,11 @@ desc_to_olflags_v(__m128i descs[4], __m128i mbuf_init, uint8_t vlan_flags, * appropriate flags means that we have to do a shift and blend for * each mbuf before we do the write. */ -#ifdef RTE_MACHINE_CPUFLAG_SSE4_2 - rearm0 = _mm_blend_epi16(mbuf_init, _mm_slli_si128(vtag1, 8), 0x10); rearm1 = _mm_blend_epi16(mbuf_init, _mm_slli_si128(vtag1, 6), 0x10); rearm2 = _mm_blend_epi16(mbuf_init, _mm_slli_si128(vtag1, 4), 0x10); rearm3 = _mm_blend_epi16(mbuf_init, _mm_slli_si128(vtag1, 2), 0x10); -#else - rearm0 = _mm_slli_si128(vtag1, 14); - rearm1 = _mm_slli_si128(vtag1, 12); - rearm2 = _mm_slli_si128(vtag1, 10); - rearm3 = _mm_slli_si128(vtag1, 8); - - rearm0 = _mm_or_si128(mbuf_init, _mm_srli_epi64(rearm0, 48)); - rearm1 = _mm_or_si128(mbuf_init, _mm_srli_epi64(rearm1, 48)); - rearm2 = _mm_or_si128(mbuf_init, _mm_srli_epi64(rearm2, 48)); - rearm3 = _mm_or_si128(mbuf_init, _mm_srli_epi64(rearm3, 48)); - -#endif /* RTE_MACHINE_CPUFLAG_SSE4_2 */ - /* write the rearm data and the olflags in one write */ RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, ol_flags) != offsetof(struct rte_mbuf, rearm_data) + 8); @@ -247,6 +232,68 @@ desc_to_olflags_v(__m128i descs[4], __m128i mbuf_init, uint8_t vlan_flags, _mm_store_si128((__m128i *)&rx_pkts[3]->rearm_data, rearm3); } +static inline uint32_t get_packet_type(int index, + uint32_t pkt_info, + uint32_t etqf_check, + uint32_t tunnel_check) +{ + if (etqf_check & (0x02 << (index * RTE_IXGBE_DESCS_PER_LOOP))) + return RTE_PTYPE_UNKNOWN; + + if (tunnel_check & (0x02 << (index * RTE_IXGBE_DESCS_PER_LOOP))) { + pkt_info &= IXGBE_PACKET_TYPE_MASK_TUNNEL; + return ptype_table_tn[pkt_info]; + } + + pkt_info &= IXGBE_PACKET_TYPE_MASK_82599; + return ptype_table[pkt_info]; +} + +static inline void +desc_to_ptype_v(__m128i descs[4], uint16_t pkt_type_mask, + struct rte_mbuf **rx_pkts) +{ + __m128i etqf_mask = _mm_set_epi64x(0x800000008000LL, 0x800000008000LL); + __m128i ptype_mask = _mm_set_epi32( + pkt_type_mask, pkt_type_mask, pkt_type_mask, pkt_type_mask); + __m128i tunnel_mask = + _mm_set_epi64x(0x100000001000LL, 0x100000001000LL); + + uint32_t etqf_check, tunnel_check, pkt_info; + + __m128i ptype0 = _mm_unpacklo_epi32(descs[0], descs[2]); + __m128i ptype1 = _mm_unpacklo_epi32(descs[1], descs[3]); + + /* interleave low 32 bits, + * now we have 4 ptypes in a XMM register + */ + ptype0 = _mm_unpacklo_epi32(ptype0, ptype1); + + /* create a etqf bitmask based on the etqf bit. */ + etqf_check = _mm_movemask_epi8(_mm_and_si128(ptype0, etqf_mask)); + + /* shift left by IXGBE_PACKET_TYPE_SHIFT, and apply ptype mask */ + ptype0 = _mm_and_si128(_mm_srli_epi32(ptype0, IXGBE_PACKET_TYPE_SHIFT), + ptype_mask); + + /* create a tunnel bitmask based on the tunnel bit */ + tunnel_check = _mm_movemask_epi8( + _mm_slli_epi32(_mm_and_si128(ptype0, tunnel_mask), 0x3)); + + pkt_info = _mm_extract_epi32(ptype0, 0); + rx_pkts[0]->packet_type = + get_packet_type(0, pkt_info, etqf_check, tunnel_check); + pkt_info = _mm_extract_epi32(ptype0, 1); + rx_pkts[1]->packet_type = + get_packet_type(1, pkt_info, etqf_check, tunnel_check); + pkt_info = _mm_extract_epi32(ptype0, 2); + rx_pkts[2]->packet_type = + get_packet_type(2, pkt_info, etqf_check, tunnel_check); + pkt_info = _mm_extract_epi32(ptype0, 3); + rx_pkts[3]->packet_type = + get_packet_type(3, pkt_info, etqf_check, tunnel_check); +} + /* * vPMD raw receive routine, only accept(nb_pkts >= RTE_IXGBE_DESCS_PER_LOOP) * @@ -476,6 +523,8 @@ _recv_raw_pkts_vec(struct ixgbe_rx_queue *rxq, struct rte_mbuf **rx_pkts, _mm_storeu_si128((void *)&rx_pkts[pos]->rx_descriptor_fields1, pkt_mb1); + desc_to_ptype_v(descs, rxq->pkt_type_mask, &rx_pkts[pos]); + /* C.4 calc avaialbe number of desc */ var = __builtin_popcountll(_mm_cvtsi128_si64(staterr)); nb_pkts_recd += var;