4 * Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
7 * Redistribution and use in source and binary forms, with or without
8 * modification, are permitted provided that the following conditions
11 * * Redistributions of source code must retain the above copyright
12 * notice, this list of conditions and the following disclaimer.
13 * * Redistributions in binary form must reproduce the above copyright
14 * notice, this list of conditions and the following disclaimer in
15 * the documentation and/or other materials provided with the
17 * * Neither the name of Intel Corporation nor the names of its
18 * contributors may be used to endorse or promote products derived
19 * from this software without specific prior written permission.
21 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
22 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
23 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
24 * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
25 * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
26 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
27 * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
28 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
29 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
30 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
31 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
34 #ifndef _RTE_LPM_SSE_H_
35 #define _RTE_LPM_SSE_H_
37 #include <rte_branch_prediction.h>
38 #include <rte_byteorder.h>
39 #include <rte_common.h>
48 rte_lpm_lookupx4(const struct rte_lpm *lpm, xmm_t ip, uint32_t hop[4],
54 uint64_t idx, pt, pt2;
58 _mm_set_epi32(UINT8_MAX, UINT8_MAX, UINT8_MAX, UINT8_MAX);
61 * RTE_LPM_VALID_EXT_ENTRY_BITMASK for 2 LPM entries
62 * as one 64-bit value (0x0300000003000000).
64 const uint64_t mask_xv =
65 ((uint64_t)RTE_LPM_VALID_EXT_ENTRY_BITMASK |
66 (uint64_t)RTE_LPM_VALID_EXT_ENTRY_BITMASK << 32);
69 * RTE_LPM_LOOKUP_SUCCESS for 2 LPM entries
70 * as one 64-bit value (0x0100000001000000).
72 const uint64_t mask_v =
73 ((uint64_t)RTE_LPM_LOOKUP_SUCCESS |
74 (uint64_t)RTE_LPM_LOOKUP_SUCCESS << 32);
76 /* get 4 indexes for tbl24[]. */
77 i24 = _mm_srli_epi32(ip, CHAR_BIT);
79 /* extract values from tbl24[] */
80 idx = _mm_cvtsi128_si64(i24);
81 /* With -O0 option, gcc 4.8 - 5.4 fails to fold sizeof() into a constant */
82 i24 = _mm_srli_si128(i24, /* sizeof(uint64_t) */ 8);
84 ptbl = (const uint32_t *)&lpm->tbl24[(uint32_t)idx];
86 ptbl = (const uint32_t *)&lpm->tbl24[idx >> 32];
89 idx = _mm_cvtsi128_si64(i24);
91 ptbl = (const uint32_t *)&lpm->tbl24[(uint32_t)idx];
93 ptbl = (const uint32_t *)&lpm->tbl24[idx >> 32];
96 /* get 4 indexes for tbl8[]. */
97 i8.x = _mm_and_si128(ip, mask8);
99 pt = (uint64_t)tbl[0] |
100 (uint64_t)tbl[1] << 32;
101 pt2 = (uint64_t)tbl[2] |
102 (uint64_t)tbl[3] << 32;
104 /* search successfully finished for all 4 IP addresses. */
105 if (likely((pt & mask_xv) == mask_v) &&
106 likely((pt2 & mask_xv) == mask_v)) {
107 *(uint64_t *)hop = pt & RTE_LPM_MASKX4_RES;
108 *(uint64_t *)(hop + 2) = pt2 & RTE_LPM_MASKX4_RES;
112 if (unlikely((pt & RTE_LPM_VALID_EXT_ENTRY_BITMASK) ==
113 RTE_LPM_VALID_EXT_ENTRY_BITMASK)) {
114 i8.u32[0] = i8.u32[0] +
115 (uint8_t)tbl[0] * RTE_LPM_TBL8_GROUP_NUM_ENTRIES;
116 ptbl = (const uint32_t *)&lpm->tbl8[i8.u32[0]];
119 if (unlikely((pt >> 32 & RTE_LPM_VALID_EXT_ENTRY_BITMASK) ==
120 RTE_LPM_VALID_EXT_ENTRY_BITMASK)) {
121 i8.u32[1] = i8.u32[1] +
122 (uint8_t)tbl[1] * RTE_LPM_TBL8_GROUP_NUM_ENTRIES;
123 ptbl = (const uint32_t *)&lpm->tbl8[i8.u32[1]];
126 if (unlikely((pt2 & RTE_LPM_VALID_EXT_ENTRY_BITMASK) ==
127 RTE_LPM_VALID_EXT_ENTRY_BITMASK)) {
128 i8.u32[2] = i8.u32[2] +
129 (uint8_t)tbl[2] * RTE_LPM_TBL8_GROUP_NUM_ENTRIES;
130 ptbl = (const uint32_t *)&lpm->tbl8[i8.u32[2]];
133 if (unlikely((pt2 >> 32 & RTE_LPM_VALID_EXT_ENTRY_BITMASK) ==
134 RTE_LPM_VALID_EXT_ENTRY_BITMASK)) {
135 i8.u32[3] = i8.u32[3] +
136 (uint8_t)tbl[3] * RTE_LPM_TBL8_GROUP_NUM_ENTRIES;
137 ptbl = (const uint32_t *)&lpm->tbl8[i8.u32[3]];
141 hop[0] = (tbl[0] & RTE_LPM_LOOKUP_SUCCESS) ? tbl[0] & 0x00FFFFFF : defv;
142 hop[1] = (tbl[1] & RTE_LPM_LOOKUP_SUCCESS) ? tbl[1] & 0x00FFFFFF : defv;
143 hop[2] = (tbl[2] & RTE_LPM_LOOKUP_SUCCESS) ? tbl[2] & 0x00FFFFFF : defv;
144 hop[3] = (tbl[3] & RTE_LPM_LOOKUP_SUCCESS) ? tbl[3] & 0x00FFFFFF : defv;
151 #endif /* _RTE_LPM_SSE_H_ */