1.. BSD LICENSE 2 Copyright(c) 2010-2014 Intel Corporation. All rights reserved. 3 All rights reserved. 4 5 Redistribution and use in source and binary forms, with or without 6 modification, are permitted provided that the following conditions 7 are met: 8 9 * Redistributions of source code must retain the above copyright 10 notice, this list of conditions and the following disclaimer. 11 * Redistributions in binary form must reproduce the above copyright 12 notice, this list of conditions and the following disclaimer in 13 the documentation and/or other materials provided with the 14 distribution. 15 * Neither the name of Intel Corporation nor the names of its 16 contributors may be used to endorse or promote products derived 17 from this software without specific prior written permission. 18 19 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS 20 "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT 21 LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR 22 A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT 23 OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, 24 SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT 25 LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, 26 DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY 27 THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT 28 (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE 29 OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 30 31LPM6 Library 32============ 33 34The LPM6 (LPM for IPv6) library component implements the Longest Prefix Match (LPM) table search method for 128-bit keys 35that is typically used to find the best match route in IPv6 forwarding applications. 36 37LPM6 API Overview 38----------------- 39 40The main configuration parameters for the LPM6 library are: 41 42* Maximum number of rules: This defines the size of the table that holds the rules, 43 and therefore the maximum number of rules that can be added. 44 45* Number of tbl8s: A tbl8 is a node of the trie that the LPM6 algorithm is based on. 46 47This parameter is related to the number of rules you can have, 48but there is no way to accurately predict the number needed to hold a specific number of rules, 49since it strongly depends on the depth and IP address of every rule. 50One tbl8 consumes 1 kb of memory. As a recommendation, 65536 tbl8s should be sufficient to store 51several thousand IPv6 rules, but the number can vary depending on the case. 52 53An LPM prefix is represented by a pair of parameters (128-bit key, depth), with depth in the range of 1 to 128. 54An LPM rule is represented by an LPM prefix and some user data associated with the prefix. 55The prefix serves as the unique identifier for the LPM rule. 56In this implementation, the user data is 1-byte long and is called "next hop", 57which corresponds to its main use of storing the ID of the next hop in a routing table entry. 58 59The main methods exported for the LPM component are: 60 61* Add LPM rule: The LPM rule is provided as input. 62 If there is no rule with the same prefix present in the table, then the new rule is added to the LPM table. 63 If a rule with the same prefix is already present in the table, the next hop of the rule is updated. 64 An error is returned when there is no available space left. 65 66* Delete LPM rule: The prefix of the LPM rule is provided as input. 67 If a rule with the specified prefix is present in the LPM table, then it is removed. 68 69* Lookup LPM key: The 128-bit key is provided as input. 70 The algorithm selects the rule that represents the best match for the given key and returns the next hop of that rule. 71 In the case that there are multiple rules present in the LPM table that have the same 128-bit value, 72 the algorithm picks the rule with the highest depth as the best match rule, 73 which means the rule has the highest number of most significant bits matching between the input key and the rule key. 74 75Implementation Details 76~~~~~~~~~~~~~~~~~~~~~~ 77 78This is a modification of the algorithm used for IPv4 (see Section 19.2 "Implementation Details"). 79In this case, instead of using two levels, one with a tbl24 and a second with a tbl8, 14 levels are used. 80 81The implementation can be seen as a multi-bit trie where the *stride* 82or number of bits inspected on each level varies from level to level. 83Specifically, 24 bits are inspected on the root node, and the remaining 104 bits are inspected in groups of 8 bits. 84This effectively means that the trie has 14 levels at the most, depending on the rules that are added to the table. 85 86The algorithm allows the lookup operation to be performed with a number of memory accesses 87that directly depends on the length of the rule and 88whether there are other rules with bigger depths and the same key in the data structure. 89It can vary from 1 to 14 memory accesses, with 5 being the average value for the lengths 90that are most commonly used in IPv6. 91 92The main data structure is built using the following elements: 93 94* A table with 224 entries 95 96* A number of tables, configurable by the user through the API, with 28 entries 97 98The first table, called tbl24, is indexed using the first 24 bits of the IP address be looked up, 99while the rest of the tables, called tbl8s, 100are indexed using the rest of the bytes of the IP address, in chunks of 8 bits. 101This means that depending on the outcome of trying to match the IP address of an incoming packet to the rule stored in the tbl24 102or the subsequent tbl8s we might need to continue the lookup process in deeper levels of the tree. 103 104Similar to the limitation presented in the algorithm for IPv4, 105to store every possible IPv6 rule, we would need a table with 2^128 entries. 106This is not feasible due to resource restrictions. 107 108By splitting the process in different tables/levels and limiting the number of tbl8s, 109we can greatly reduce memory consumption while maintaining a very good lookup speed (one memory access per level). 110 111.. image40_png has been renamed 112 113|tbl24_tbl8_tbl8| 114 115An entry in a table contains the following fields: 116 117* next hop / index to the tbl8 118 119* depth of the rule (length) 120 121* valid flag 122 123* valid group flag 124 125* external entry flag 126 127The first field can either contain a number indicating the tbl8 in which the lookup process should continue 128or the next hop itself if the longest prefix match has already been found. 129The depth or length of the rule is the number of bits of the rule that is stored in a specific entry. 130The flags are used to determine whether the entry/table is valid or not 131and whether the search process have finished or not respectively. 132 133Both types of tables share the same structure. 134 135The other main data structure is a table containing the main information about the rules (IP, next hop and depth). 136This is a higher level table, used for different things: 137 138* Check whether a rule already exists or not, prior to addition or deletion, 139 without having to actually perform a lookup. 140 141When deleting, to check whether there is a rule containing the one that is to be deleted. 142This is important, since the main data structure will have to be updated accordingly. 143 144Addition 145~~~~~~~~ 146 147When adding a rule, there are different possibilities. 148If the rule's depth is exactly 24 bits, then: 149 150* Use the rule (IP address) as an index to the tbl24. 151 152* If the entry is invalid (i.e. it doesn't already contain a rule) then set its next hop to its value, 153 the valid flag to 1 (meaning this entry is in use), 154 and the external entry flag to 0 (meaning the lookup process ends at this point, 155 since this is the longest prefix that matches). 156 157If the rule's depth is bigger than 24 bits but a multiple of 8, then: 158 159* Use the first 24 bits of the rule as an index to the tbl24. 160 161* If the entry is invalid (i.e. it doesn't already contain a rule) then look for a free tbl8, 162 set the index to the tbl8 to this value, 163 the valid flag to 1 (meaning this entry is in use), 164 and the external entry flag to 1 165 (meaning the lookup process must continue since the rule hasn't been explored completely). 166 167* Use the following 8 bits of the rule as an index to the next tbl8. 168 169* Repeat the process until the tbl8 at the right level (depending on the depth) has been reached 170 and fill it with the next hop, setting the next entry flag to 0. 171 172If the rule's depth is any other value, prefix expansion must be performed. 173This means the rule is copied to all the entries (as long as they are not in use) which would also cause a match. 174 175As a simple example, let's assume the depth is 20 bits. 176This means that there are 2^(24-20) = 16 different combinations of the first 24 bits of an IP address that would cause a match. 177Hence, in this case, we copy the exact same entry to every position indexed by one of these combinations. 178 179By doing this we ensure that during the lookup process, if a rule matching the IP address exists, 180it is found in, at the most, 14 memory accesses, 181depending on how many times we need to move to the next table. 182Prefix expansion is one of the keys of this algorithm, since it improves the speed dramatically by adding redundancy. 183 184Prefix expansion can be performed at any level. 185So, for example, is the depth is 34 bits, it will be performed in the third level (second tbl8-based level). 186 187Lookup 188~~~~~~ 189 190The lookup process is much simpler and quicker. In this case: 191 192* Use the first 24 bits of the IP address as an index to the tbl24. 193 If the entry is not in use, then it means we don't have a rule matching this IP. 194 If it is valid and the external entry flag is set to 0, then the next hop is returned. 195 196* If it is valid and the external entry flag is set to 1, then we use the tbl8 index to find out the tbl8 to be checked, 197 and the next 8 bits of the IP address as an index to this table. 198 Similarly, if the entry is not in use, then we don't have a rule matching this IP address. 199 If it is valid then check the external entry flag for a new tbl8 to be inspected. 200 201* Repeat the process until either we find an invalid entry (lookup miss) or a valid entry with the external entry flag set to 0. 202 Return the next hop in the latter case. 203 204Limitations in the Number of Rules 205~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 206 207There are different things that limit the number of rules that can be added. 208The first one is the maximum number of rules, which is a parameter passed through the API. 209Once this number is reached, it is not possible to add any more rules to the routing table unless one or more are removed. 210 211The second limitation is in the number of tbl8s available. 212If we exhaust tbl8s, we won't be able to add any more rules. 213How to know how many of them are necessary for a specific routing table is hard to determine in advance. 214 215In this algorithm, the maximum number of tbl8s a single rule can consume is 13, 216which is the number of levels minus one, since the first three bytes are resolved in the tbl24. However: 217 218* Typically, on IPv6, routes are not longer than 48 bits, which means rules usually take up to 3 tbl8s. 219 220As explained in the LPM for IPv4 algorithm, it is possible and very likely that several rules will share one or more tbl8s, 221depending on what their first bytes are. 222If they share the same first 24 bits, for instance, the tbl8 at the second level will be shared. 223This might happen again in deeper levels, so, effectively, 224two 48 bit-long rules may use the same three tbl8s if the only difference is in their last byte. 225 226The number of tbl8s is a parameter exposed to the user through the API in this version of the algorithm, 227due to its impact in memory consumption and the number or rules that can be added to the LPM table. 228One tbl8 consumes 1 kilobyte of memory. 229 230Use Case: IPv6 Forwarding 231------------------------- 232 233The LPM algorithm is used to implement the Classless Inter-Domain Routing (CIDR) strategy used by routers implementing IP forwarding. 234 235.. |tbl24_tbl8_tbl8| image:: img/tbl24_tbl8_tbl8.* 236