1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright(c) 2016-2017 Intel Corporation.
4 IPsec Security Gateway Sample Application
5 =========================================
7 The IPsec Security Gateway application is an example of a "real world"
8 application using DPDK cryptodev framework.
13 The application demonstrates the implementation of a Security Gateway
14 (not IPsec compliant, see the Constraints section below) using DPDK based on RFC4301,
15 RFC4303, RFC3602 and RFC2404.
17 Internet Key Exchange (IKE) is not implemented, so only manual setting of
18 Security Policies and Security Associations is supported.
20 The Security Policies (SP) are implemented as ACL rules, the Security
21 Associations (SA) are stored in a table and the routing is implemented
24 The application classifies the ports as *Protected* and *Unprotected*.
25 Thus, traffic received on an Unprotected or Protected port is consider
26 Inbound or Outbound respectively.
28 The application also supports complete IPsec protocol offload to hardware
29 (Look aside crypto accelerator or using ethernet device). It also support
30 inline ipsec processing by the supported ethernet device during transmission.
31 These modes can be selected during the SA creation configuration.
33 In case of complete protocol offload, the processing of headers(ESP and outer
34 IP header) is done by the hardware and the application does not need to
35 add/remove them during outbound/inbound processing.
37 For inline offloaded outbound traffic, the application will not do the LPM
38 lookup for routing, as the port on which the packet has to be forwarded will be
39 part of the SA. Security parameters will be configured on that port only, and
40 sending the packet on other ports could result in unencrypted packets being
43 The Path for IPsec Inbound traffic is:
45 * Read packets from the port.
46 * Classify packets between IPv4 and ESP.
47 * Perform Inbound SA lookup for ESP packets based on their SPI.
48 * Perform Verification/Decryption (Not needed in case of inline ipsec).
49 * Remove ESP and outer IP header (Not needed in case of protocol offload).
50 * Inbound SP check using ACL of decrypted packets and any other IPv4 packets.
52 * Write packet to port.
54 The Path for the IPsec Outbound traffic is:
56 * Read packets from the port.
57 * Perform Outbound SP check using ACL of all IPv4 traffic.
58 * Perform Outbound SA lookup for packets that need IPsec protection.
59 * Add ESP and outer IP header (Not needed in case protocol offload).
60 * Perform Encryption/Digest (Not needed in case of inline ipsec).
62 * Write packet to port.
68 * No IPv6 options headers.
70 * Supported algorithms: AES-CBC, AES-CTR, AES-GCM, 3DES-CBC, HMAC-SHA1 and NULL.
71 * Each SA must be handle by a unique lcore (*1 RX queue per port*).
73 Compiling the Application
74 -------------------------
76 To compile the sample application see :doc:`compiling`.
78 The application is located in the ``ipsec-secgw`` sub-directory.
80 #. [Optional] Build the application for debugging:
81 This option adds some extra flags, disables compiler optimizations and
87 Running the Application
88 -----------------------
90 The application has a number of command line options::
93 ./build/ipsec-secgw [EAL options] --
94 -p PORTMASK -P -u PORTMASK -j FRAMESIZE
95 -l -w REPLAY_WINOW_SIZE -e -a
96 --config (port,queue,lcore)[,(port,queue,lcore]
106 * ``-p PORTMASK``: Hexadecimal bitmask of ports to configure.
108 * ``-P``: *optional*. Sets all ports to promiscuous mode so that packets are
109 accepted regardless of the packet's Ethernet MAC destination address.
110 Without this option, only packets with the Ethernet MAC destination address
111 set to the Ethernet address of the port are accepted (default is enabled).
113 * ``-u PORTMASK``: hexadecimal bitmask of unprotected ports
115 * ``-j FRAMESIZE``: *optional*. data buffer size (in bytes),
116 in other words maximum data size for one segment.
117 Packets with length bigger then FRAMESIZE still can be received,
118 but will be segmented.
119 Default value: RTE_MBUF_DEFAULT_BUF_SIZE (2176)
120 Minimum value: RTE_MBUF_DEFAULT_BUF_SIZE (2176)
121 Maximum value: UINT16_MAX (65535).
123 * ``-l``: enables code-path that uses librte_ipsec.
125 * ``-w REPLAY_WINOW_SIZE``: specifies the IPsec sequence number replay window
126 size for each Security Association (available only with librte_ipsec
129 * ``-e``: enables Security Association extended sequence number processing
130 (available only with librte_ipsec code path).
132 * ``-a``: enables Security Association sequence number atomic behavior
133 (available only with librte_ipsec code path).
135 * ``--config (port,queue,lcore)[,(port,queue,lcore)]``: determines which queues
136 from which ports are mapped to which cores.
138 * ``--single-sa SAIDX``: use a single SA for outbound traffic, bypassing the SP
139 on both Inbound and Outbound. This option is meant for debugging/performance
142 * ``--rxoffload MASK``: RX HW offload capabilities to enable/use on this port
143 (bitmask of DEV_RX_OFFLOAD_* values). It is an optional parameter and
144 allows user to disable some of the RX HW offload capabilities.
145 By default all HW RX offloads are enabled.
147 * ``--txoffload MASK``: TX HW offload capabilities to enable/use on this port
148 (bitmask of DEV_TX_OFFLOAD_* values). It is an optional parameter and
149 allows user to disable some of the TX HW offload capabilities.
150 By default all HW TX offloads are enabled.
152 * ``--mtu MTU``: MTU value (in bytes) on all attached ethernet ports.
153 Outgoing packets with length bigger then MTU will be fragmented.
154 Incoming packets with length bigger then MTU will be discarded.
157 * ``--frag-ttl FRAG_TTL_NS``: fragment lifetime (in nanoseconds).
158 If packet is not reassembled within this time, received fragments
159 will be discarded. Fragment lifetime should be decreased when
160 there is a high fragmented traffic loss in high bandwidth networks.
161 Should be lower for low number of reassembly buckets.
162 Valid values: from 1 ns to 10 s. Default value: 10000000 (10 s).
164 * ``--reassemble NUM``: max number of entries in reassemble fragment table.
165 Zero value disables reassembly functionality.
168 * ``-f CONFIG_FILE_PATH``: the full path of text-based file containing all
169 configuration items for running the application (See Configuration file
170 syntax section below). ``-f CONFIG_FILE_PATH`` **must** be specified.
171 **ONLY** the UNIX format configuration file is accepted.
174 The mapping of lcores to port/queues is similar to other l3fwd applications.
176 For example, given the following command line::
178 ./build/ipsec-secgw -l 20,21 -n 4 --socket-mem 0,2048 \
179 --vdev "crypto_null" -- -p 0xf -P -u 0x3 \
180 --config="(0,0,20),(1,0,20),(2,0,21),(3,0,21)" \
181 -f /path/to/config_file \
183 where each options means:
185 * The ``-l`` option enables cores 20 and 21.
187 * The ``-n`` option sets memory 4 channels.
189 * The ``--socket-mem`` to use 2GB on socket 1.
191 * The ``--vdev "crypto_null"`` option creates virtual NULL cryptodev PMD.
193 * The ``-p`` option enables ports (detected) 0, 1, 2 and 3.
195 * The ``-P`` option enables promiscuous mode.
197 * The ``-u`` option sets ports 1 and 2 as unprotected, leaving 2 and 3 as protected.
199 * The ``--config`` option enables one queue per port with the following mapping:
201 +----------+-----------+-----------+---------------------------------------+
202 | **Port** | **Queue** | **lcore** | **Description** |
204 +----------+-----------+-----------+---------------------------------------+
205 | 0 | 0 | 20 | Map queue 0 from port 0 to lcore 20. |
207 +----------+-----------+-----------+---------------------------------------+
208 | 1 | 0 | 20 | Map queue 0 from port 1 to lcore 20. |
210 +----------+-----------+-----------+---------------------------------------+
211 | 2 | 0 | 21 | Map queue 0 from port 2 to lcore 21. |
213 +----------+-----------+-----------+---------------------------------------+
214 | 3 | 0 | 21 | Map queue 0 from port 3 to lcore 21. |
216 +----------+-----------+-----------+---------------------------------------+
218 * The ``-f /path/to/config_file`` option enables the application read and
219 parse the configuration file specified, and configures the application
220 with a given set of SP, SA and Routing entries accordingly. The syntax of
221 the configuration file will be explained below in more detail. Please
222 **note** the parser only accepts UNIX format text file. Other formats
223 such as DOS/MAC format will cause a parse error.
225 Refer to the *DPDK Getting Started Guide* for general information on running
226 applications and the Environment Abstraction Layer (EAL) options.
228 The application would do a best effort to "map" crypto devices to cores, with
229 hardware devices having priority. Basically, hardware devices if present would
230 be assigned to a core before software ones.
231 This means that if the application is using a single core and both hardware
232 and software crypto devices are detected, hardware devices will be used.
234 A way to achieve the case where you want to force the use of virtual crypto
235 devices is to whitelist the Ethernet devices needed and therefore implicitly
236 blacklisting all hardware crypto devices.
238 For example, something like the following command line:
240 .. code-block:: console
242 ./build/ipsec-secgw -l 20,21 -n 4 --socket-mem 0,2048 \
243 -w 81:00.0 -w 81:00.1 -w 81:00.2 -w 81:00.3 \
244 --vdev "crypto_aesni_mb" --vdev "crypto_null" \
246 -p 0xf -P -u 0x3 --config="(0,0,20),(1,0,20),(2,0,21),(3,0,21)" \
253 The following sections provide the syntax of configurations to initialize
254 your SP, SA, Routing and Neighbour tables.
255 Configurations shall be specified in the configuration file to be passed to
256 the application. The file is then parsed by the application. The successful
257 parsing will result in the appropriate rules being applied to the tables
261 Configuration File Syntax
262 ~~~~~~~~~~~~~~~~~~~~~~~~~
264 As mention in the overview, the Security Policies are ACL rules.
265 The application parsers the rules specified in the configuration file and
266 passes them to the ACL table, and replicates them per socket in use.
268 Following are the configuration file syntax.
273 The parse treats one line in the configuration file as one configuration
274 item (unless the line concatenation symbol exists). Every configuration
275 item shall follow the syntax of either SP, SA, Routing or Neighbour
276 rules specified below.
278 The configuration parser supports the following special symbols:
280 * Comment symbol **#**. Any character from this symbol to the end of
281 line is treated as comment and will not be parsed.
283 * Line concatenation symbol **\\**. This symbol shall be placed in the end
284 of the line to be concatenated to the line below. Multiple lines'
285 concatenation is supported.
291 The SP rule syntax is shown as follows:
293 .. code-block:: console
295 sp <ip_ver> <dir> esp <action> <priority> <src_ip> <dst_ip>
296 <proto> <sport> <dport>
299 where each options means:
303 * IP protocol version
309 * *ipv4*: IP protocol version 4
310 * *ipv6*: IP protocol version 6
314 * The traffic direction
320 * *in*: inbound traffic
321 * *out*: outbound traffic
331 * *protect <SA_idx>*: the specified traffic is protected by SA rule
333 * *bypass*: the specified traffic traffic is bypassed
334 * *discard*: the specified traffic is discarded
340 * Optional: Yes, default priority 0 will be used
346 * The source IP address and mask
348 * Optional: Yes, default address 0.0.0.0 and mask of 0 will be used
352 * *src X.X.X.X/Y* for IPv4
353 * *src XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX/Y* for IPv6
357 * The destination IP address and mask
359 * Optional: Yes, default address 0.0.0.0 and mask of 0 will be used
363 * *dst X.X.X.X/Y* for IPv4
364 * *dst XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX/Y* for IPv6
368 * The protocol start and end range
370 * Optional: yes, default range of 0 to 0 will be used
372 * Syntax: *proto X:Y*
376 * The source port start and end range
378 * Optional: yes, default range of 0 to 0 will be used
380 * Syntax: *sport X:Y*
384 * The destination port start and end range
386 * Optional: yes, default range of 0 to 0 will be used
388 * Syntax: *dport X:Y*
392 .. code-block:: console
394 sp ipv4 out esp protect 105 pri 1 dst 192.168.115.0/24 sport 0:65535 \
397 sp ipv6 in esp bypass pri 1 dst 0000:0000:0000:0000:5555:5555:\
398 0000:0000/96 sport 0:65535 dport 0:65535
404 The successfully parsed SA rules will be stored in an array table.
406 The SA rule syntax is shown as follows:
408 .. code-block:: console
410 sa <dir> <spi> <cipher_algo> <cipher_key> <auth_algo> <auth_key>
411 <mode> <src_ip> <dst_ip> <action_type> <port_id> <fallback>
413 where each options means:
417 * The traffic direction
423 * *in*: inbound traffic
424 * *out*: outbound traffic
432 * Syntax: unsigned integer number
438 * Optional: Yes, unless <aead_algo> is not used
442 * *null*: NULL algorithm
443 * *aes-128-cbc*: AES-CBC 128-bit algorithm
444 * *aes-256-cbc*: AES-CBC 256-bit algorithm
445 * *aes-128-ctr*: AES-CTR 128-bit algorithm
446 * *3des-cbc*: 3DES-CBC 192-bit algorithm
448 * Syntax: *cipher_algo <your algorithm>*
452 * Cipher key, NOT available when 'null' algorithm is used
454 * Optional: Yes, unless <aead_algo> is not used.
455 Must be followed by <cipher_algo> option
457 * Syntax: Hexadecimal bytes (0x0-0xFF) concatenate by colon symbol ':'.
458 The number of bytes should be as same as the specified cipher algorithm
461 For example: *cipher_key A1:B2:C3:D4:A1:B2:C3:D4:A1:B2:C3:D4:
466 * Authentication algorithm
468 * Optional: Yes, unless <aead_algo> is not used
472 * *null*: NULL algorithm
473 * *sha1-hmac*: HMAC SHA1 algorithm
477 * Authentication key, NOT available when 'null' or 'aes-128-gcm' algorithm
480 * Optional: Yes, unless <aead_algo> is not used.
481 Must be followed by <auth_algo> option
483 * Syntax: Hexadecimal bytes (0x0-0xFF) concatenate by colon symbol ':'.
484 The number of bytes should be as same as the specified authentication
487 For example: *auth_key A1:B2:C3:D4:A1:B2:C3:D4:A1:B2:C3:D4:A1:B2:C3:D4:
494 * Optional: Yes, unless <cipher_algo> and <auth_algo> are not used
498 * *aes-128-gcm*: AES-GCM 128-bit algorithm
500 * Syntax: *cipher_algo <your algorithm>*
504 * Cipher key, NOT available when 'null' algorithm is used
506 * Optional: Yes, unless <cipher_algo> and <auth_algo> are not used.
507 Must be followed by <aead_algo> option
509 * Syntax: Hexadecimal bytes (0x0-0xFF) concatenate by colon symbol ':'.
510 The number of bytes should be as same as the specified AEAD algorithm
513 For example: *aead_key A1:B2:C3:D4:A1:B2:C3:D4:A1:B2:C3:D4:
524 * *ipv4-tunnel*: Tunnel mode for IPv4 packets
525 * *ipv6-tunnel*: Tunnel mode for IPv6 packets
526 * *transport*: transport mode
532 * The source IP address. This option is not available when
533 transport mode is used
535 * Optional: Yes, default address 0.0.0.0 will be used
539 * *src X.X.X.X* for IPv4
540 * *src XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX* for IPv6
544 * The destination IP address. This option is not available when
545 transport mode is used
547 * Optional: Yes, default address 0.0.0.0 will be used
551 * *dst X.X.X.X* for IPv4
552 * *dst XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX* for IPv6
556 * Action type to specify the security action. This option specify
557 the SA to be performed with look aside protocol offload to HW
558 accelerator or protocol offload on ethernet device or inline
559 crypto processing on the ethernet device during transmission.
561 * Optional: Yes, default type *no-offload*
565 * *lookaside-protocol-offload*: look aside protocol offload to HW accelerator
566 * *inline-protocol-offload*: inline protocol offload on ethernet device
567 * *inline-crypto-offload*: inline crypto processing on ethernet device
568 * *no-offload*: no offloading to hardware
572 * Port/device ID of the ethernet/crypto accelerator for which the SA is
573 configured. For *inline-crypto-offload* and *inline-protocol-offload*, this
574 port will be used for routing. The routing table will not be referred in
577 * Optional: No, if *type* is not *no-offload*
581 * *port_id X* X is a valid device number in decimal
585 * Action type for ingress IPsec packets that inline processor failed to
586 process. Only a combination of *inline-crypto-offload* as a primary
587 session and *lookaside-none* as a fall-back session is supported at the
590 If used in conjunction with IPsec window, its width needs be increased
591 due to different processing times of inline and lookaside modes which
592 results in packet reordering.
598 * *lookaside-none*: use automatically chosen cryptodev to process packets
602 * *fallback lookaside-none*
606 .. code-block:: console
608 sa out 5 cipher_algo null auth_algo null mode ipv4-tunnel \
609 src 172.16.1.5 dst 172.16.2.5
611 sa out 25 cipher_algo aes-128-cbc \
612 cipher_key c3:c3:c3:c3:c3:c3:c3:c3:c3:c3:c3:c3:c3:c3:c3:c3 \
613 auth_algo sha1-hmac \
614 auth_key c3:c3:c3:c3:c3:c3:c3:c3:c3:c3:c3:c3:c3:c3:c3:c3:c3:c3:c3:c3 \
616 src 1111:1111:1111:1111:1111:1111:1111:5555 \
617 dst 2222:2222:2222:2222:2222:2222:2222:5555
619 sa in 105 aead_algo aes-128-gcm \
620 aead_key de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef \
621 mode ipv4-tunnel src 172.16.2.5 dst 172.16.1.5
623 sa out 5 cipher_algo aes-128-cbc cipher_key 0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0 \
624 auth_algo sha1-hmac auth_key 0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0 \
625 mode ipv4-tunnel src 172.16.1.5 dst 172.16.2.5 \
626 type lookaside-protocol-offload port_id 4
628 sa in 35 aead_algo aes-128-gcm \
629 aead_key de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef \
630 mode ipv4-tunnel src 172.16.2.5 dst 172.16.1.5 \
631 type inline-crypto-offload port_id 0
636 The Routing rule syntax is shown as follows:
638 .. code-block:: console
640 rt <ip_ver> <src_ip> <dst_ip> <port>
643 where each options means:
647 * IP protocol version
653 * *ipv4*: IP protocol version 4
654 * *ipv6*: IP protocol version 6
658 * The source IP address and mask
660 * Optional: Yes, default address 0.0.0.0 and mask of 0 will be used
664 * *src X.X.X.X/Y* for IPv4
665 * *src XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX/Y* for IPv6
669 * The destination IP address and mask
671 * Optional: Yes, default address 0.0.0.0 and mask of 0 will be used
675 * *dst X.X.X.X/Y* for IPv4
676 * *dst XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX/Y* for IPv6
680 * The traffic output port id
682 * Optional: yes, default output port 0 will be used
688 .. code-block:: console
690 rt ipv4 dst 172.16.1.5/32 port 0
692 rt ipv6 dst 1111:1111:1111:1111:1111:1111:1111:5555/116 port 0
694 Neighbour rule syntax
695 ^^^^^^^^^^^^^^^^^^^^^
697 The Neighbour rule syntax is shown as follows:
699 .. code-block:: console
701 neigh <port> <dst_mac>
704 where each options means:
716 * The destination ethernet address to use for that port
724 Example Neighbour rules:
726 .. code-block:: console
728 neigh port 0 DE:AD:BE:EF:01:02
733 The test directory contains scripts for testing the various encryption
736 The purpose of the scripts is to automate ipsec-secgw testing
737 using another system running linux as a DUT.
739 The user must setup the following environment variables:
741 * ``SGW_PATH``: path to the ipsec-secgw binary to test.
743 * ``REMOTE_HOST``: IP address/hostname of the DUT.
745 * ``REMOTE_IFACE``: interface name for the test-port on the DUT.
747 * ``ETH_DEV``: ethernet device to be used on the SUT by DPDK ('-w <pci-id>')
749 Also the user can optionally setup:
751 * ``SGW_LCORE``: lcore to run ipsec-secgw on (default value is 0)
753 * ``CRYPTO_DEV``: crypto device to be used ('-w <pci-id>'). If none specified
754 appropriate vdevs will be created by the script
756 * ``MULTI_SEG_TEST``: ipsec-secgw option to enable reassembly support and
757 specify size of reassembly table (e.g.
758 ``MULTI_SEG_TEST='--reassemble 128'``). This option must be set for
759 fallback session tests.
761 Note that most of the tests require the appropriate crypto PMD/device to be
767 Two servers are required for the tests, SUT and DUT.
769 Make sure the user from the SUT can ssh to the DUT without entering the password.
770 To enable this feature keys must be setup on the DUT.
772 ``ssh-keygen`` will make a private & public key pair on the SUT.
774 ``ssh-copy-id`` <user name>@<target host name> on the SUT will copy the public
775 key to the DUT. It will ask for credentials so that it can upload the public key.
777 The SUT and DUT are connected through at least 2 NIC ports.
779 One NIC port is expected to be managed by linux on both machines and will be
780 used as a control path.
782 The second NIC port (test-port) should be bound to DPDK on the SUT, and should
783 be managed by linux on the DUT.
785 The script starts ``ipsec-secgw`` with 2 NIC devices: ``test-port`` and
788 It then configures the local tap interface and the remote interface and IPsec
789 policies in the following way:
791 Traffic going over the test-port in both directions has to be protected by IPsec.
793 Traffic going over the TAP port in both directions does not have to be protected.
797 DUT OS(NIC1)--(IPsec)-->(NIC1)ipsec-secgw(TAP)--(plain)-->(TAP)SUT OS
799 SUT OS(TAP)--(plain)-->(TAP)psec-secgw(NIC1)--(IPsec)-->(NIC1)DUT OS
801 It then tries to perform some data transfer using the scheme described above.
806 In the ipsec-secgw/test directory
808 to run one test for IPv4 or IPv6
810 /bin/bash linux_test(4|6).sh <ipsec_mode>
812 to run all tests for IPv4 or IPv6
814 /bin/bash run_test.sh -4|-6
816 For the list of available modes please refer to run_test.sh.