4 * Copyright(c) 2010-2012 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.
35 #ifndef __INCLUDE_RTE_IPV4_FRAG_H__
36 #define __INCLUDE_RTE_IPV4_FRAG_H__
41 * RTE IPv4 Fragmentation
43 * Implementation of IPv4 fragmentation.
48 * Default byte size for the IPv4 Maximum Transfer Unit (MTU).
49 * This value includes the size of IPv4 header.
51 #define IPV4_MTU_DEFAULT ETHER_MTU
54 * Default payload in bytes for the IPv4 packet.
56 #define IPV4_DEFAULT_PAYLOAD (IPV4_MTU_DEFAULT - sizeof(struct ipv4_hdr))
59 * MAX number of fragments per packet allowed.
61 #define IPV4_MAX_FRAGS_PER_PACKET 0x80
65 #ifdef RTE_IPV4_FRAG_DEBUG
67 #define RTE_IPV4_FRAG_ASSERT(exp) \
69 rte_panic("function %s, line%d\tassert \"" #exp "\" failed\n", \
70 __func__, __LINE__); \
73 #else /*RTE_IPV4_FRAG_DEBUG*/
75 #define RTE_IPV4_FRAG_ASSERT(exp) do { } while(0)
77 #endif /*RTE_IPV4_FRAG_DEBUG*/
80 #define IPV4_HDR_DF_SHIFT 14
81 #define IPV4_HDR_MF_SHIFT 13
82 #define IPV4_HDR_FO_SHIFT 3
84 #define IPV4_HDR_DF_MASK (1 << IPV4_HDR_DF_SHIFT)
85 #define IPV4_HDR_MF_MASK (1 << IPV4_HDR_MF_SHIFT)
87 #define IPV4_HDR_FO_MASK ((1 << IPV4_HDR_FO_SHIFT) - 1)
89 static inline void __fill_ipv4hdr_frag(struct ipv4_hdr *dst,
90 const struct ipv4_hdr *src, uint16_t len, uint16_t fofs,
91 uint16_t dofs, uint32_t mf)
93 rte_memcpy(dst, src, sizeof(*dst));
94 fofs = (uint16_t)(fofs + (dofs >> IPV4_HDR_FO_SHIFT));
95 fofs = (uint16_t)(fofs | mf << IPV4_HDR_MF_SHIFT);
96 dst->fragment_offset = rte_cpu_to_be_16(fofs);
97 dst->total_length = rte_cpu_to_be_16(len);
98 dst->hdr_checksum = 0;
101 static inline void __free_fragments(struct rte_mbuf *mb[], uint32_t num)
104 for (i = 0; i != num; i++)
105 rte_pktmbuf_free(mb[i]);
109 * IPv4 fragmentation.
111 * This function implements the fragmentation of IPv4 packets.
116 * Array storing the output fragments.
118 * Size in bytes of the Maximum Transfer Unit (MTU) for the outgoing IPv4
119 * datagrams. This value includes the size of the IPv4 header.
121 * MBUF pool used for allocating direct buffers for the output fragments.
122 * @param pool_indirect
123 * MBUF pool used for allocating indirect buffers for the output fragments.
125 * Upon successful completion - number of output fragments placed
126 * in the pkts_out array.
127 * Otherwise - (-1) * <errno>.
129 static inline int32_t rte_ipv4_fragmentation(struct rte_mbuf *pkt_in,
130 struct rte_mbuf **pkts_out,
131 uint16_t nb_pkts_out,
133 struct rte_mempool *pool_direct,
134 struct rte_mempool *pool_indirect)
136 struct rte_mbuf *in_seg = NULL;
137 struct ipv4_hdr *in_hdr;
138 uint32_t out_pkt_pos, in_seg_data_pos;
139 uint32_t more_in_segs;
140 uint16_t fragment_offset, flag_offset, frag_size;
142 frag_size = (uint16_t)(mtu_size - sizeof(struct ipv4_hdr));
144 /* Fragment size should be a multiply of 8. */
145 RTE_IPV4_FRAG_ASSERT((frag_size & IPV4_HDR_FO_MASK) == 0);
147 /* Fragment size should be a multiply of 8. */
148 RTE_IPV4_FRAG_ASSERT(IPV4_MAX_FRAGS_PER_PACKET * frag_size >=
149 (uint16_t)(pkt_in->pkt.pkt_len - sizeof (struct ipv4_hdr)));
151 in_hdr = (struct ipv4_hdr*) pkt_in->pkt.data;
152 flag_offset = rte_cpu_to_be_16(in_hdr->fragment_offset);
154 /* If Don't Fragment flag is set */
155 if (unlikely ((flag_offset & IPV4_HDR_DF_MASK) != 0))
158 /* Check that pkts_out is big enough to hold all fragments */
159 if (unlikely (frag_size * nb_pkts_out <
160 (uint16_t)(pkt_in->pkt.pkt_len - sizeof (struct ipv4_hdr))))
164 in_seg_data_pos = sizeof(struct ipv4_hdr);
169 while (likely(more_in_segs)) {
170 struct rte_mbuf *out_pkt = NULL, *out_seg_prev = NULL;
171 uint32_t more_out_segs;
172 struct ipv4_hdr *out_hdr;
174 /* Allocate direct buffer */
175 out_pkt = rte_pktmbuf_alloc(pool_direct);
176 if (unlikely(out_pkt == NULL)) {
177 __free_fragments(pkts_out, out_pkt_pos);
181 /* Reserve space for the IP header that will be built later */
182 out_pkt->pkt.data_len = sizeof(struct ipv4_hdr);
183 out_pkt->pkt.pkt_len = sizeof(struct ipv4_hdr);
185 out_seg_prev = out_pkt;
187 while (likely(more_out_segs && more_in_segs)) {
188 struct rte_mbuf *out_seg = NULL;
191 /* Allocate indirect buffer */
192 out_seg = rte_pktmbuf_alloc(pool_indirect);
193 if (unlikely(out_seg == NULL)) {
194 rte_pktmbuf_free(out_pkt);
195 __free_fragments(pkts_out, out_pkt_pos);
198 out_seg_prev->pkt.next = out_seg;
199 out_seg_prev = out_seg;
201 /* Prepare indirect buffer */
202 rte_pktmbuf_attach(out_seg, in_seg);
203 len = mtu_size - out_pkt->pkt.pkt_len;
204 if (len > (in_seg->pkt.data_len - in_seg_data_pos)) {
205 len = in_seg->pkt.data_len - in_seg_data_pos;
207 out_seg->pkt.data = (char*) in_seg->pkt.data + (uint16_t)in_seg_data_pos;
208 out_seg->pkt.data_len = (uint16_t)len;
209 out_pkt->pkt.pkt_len = (uint16_t)(len +
210 out_pkt->pkt.pkt_len);
211 out_pkt->pkt.nb_segs += 1;
212 in_seg_data_pos += len;
214 /* Current output packet (i.e. fragment) done ? */
215 if (unlikely(out_pkt->pkt.pkt_len >= mtu_size)) {
219 /* Current input segment done ? */
220 if (unlikely(in_seg_data_pos == in_seg->pkt.data_len)) {
221 in_seg = in_seg->pkt.next;
224 if (unlikely(in_seg == NULL)) {
230 /* Build the IP header */
232 out_hdr = (struct ipv4_hdr*) out_pkt->pkt.data;
234 __fill_ipv4hdr_frag(out_hdr, in_hdr,
235 (uint16_t)out_pkt->pkt.pkt_len,
236 flag_offset, fragment_offset, more_in_segs);
238 fragment_offset = (uint16_t)(fragment_offset +
239 out_pkt->pkt.pkt_len - sizeof(struct ipv4_hdr));
241 out_pkt->ol_flags |= PKT_TX_IP_CKSUM;
242 out_pkt->pkt.l3_len = sizeof(struct ipv4_hdr);
244 /* Write the fragment to the output list */
245 pkts_out[out_pkt_pos] = out_pkt;
249 return (out_pkt_pos);