^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) .. SPDX-License-Identifier: GPL-2.0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) ====================================
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) HOWTO for the linux packet generator
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) ====================================
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) Enable CONFIG_NET_PKTGEN to compile and build pktgen either in-kernel
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8) or as a module. A module is preferred; modprobe pktgen if needed. Once
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9) running, pktgen creates a thread for each CPU with affinity to that CPU.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) Monitoring and controlling is done via /proc. It is easiest to select a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) suitable sample script and configure that.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) On a dual CPU::
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) ps aux | grep pkt
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) root 129 0.3 0.0 0 0 ? SW 2003 523:20 [kpktgend_0]
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) root 130 0.3 0.0 0 0 ? SW 2003 509:50 [kpktgend_1]
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) For monitoring and control pktgen creates::
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) /proc/net/pktgen/pgctrl
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) /proc/net/pktgen/kpktgend_X
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) /proc/net/pktgen/ethX
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) Tuning NIC for max performance
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28) ==============================
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) The default NIC settings are (likely) not tuned for pktgen's artificial
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) overload type of benchmarking, as this could hurt the normal use-case.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33) Specifically increasing the TX ring buffer in the NIC::
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) # ethtool -G ethX tx 1024
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) A larger TX ring can improve pktgen's performance, while it can hurt
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) in the general case, 1) because the TX ring buffer might get larger
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39) than the CPU's L1/L2 cache, 2) because it allows more queueing in the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) NIC HW layer (which is bad for bufferbloat).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42) One should hesitate to conclude that packets/descriptors in the HW
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43) TX ring cause delay. Drivers usually delay cleaning up the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) ring-buffers for various performance reasons, and packets stalling
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) the TX ring might just be waiting for cleanup.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) This cleanup issue is specifically the case for the driver ixgbe
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48) (Intel 82599 chip). This driver (ixgbe) combines TX+RX ring cleanups,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49) and the cleanup interval is affected by the ethtool --coalesce setting
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) of parameter "rx-usecs".
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52) For ixgbe use e.g. "30" resulting in approx 33K interrupts/sec (1/30*10^6)::
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) # ethtool -C ethX rx-usecs 30
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57) Kernel threads
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58) ==============
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59) Pktgen creates a thread for each CPU with affinity to that CPU.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60) Which is controlled through procfile /proc/net/pktgen/kpktgend_X.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62) Example: /proc/net/pktgen/kpktgend_0::
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) Running:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65) Stopped: eth4@0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66) Result: OK: add_device=eth4@0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68) Most important are the devices assigned to the thread.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) The two basic thread commands are:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) * add_device DEVICE@NAME -- adds a single device
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73) * rem_device_all -- remove all associated devices
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) When adding a device to a thread, a corresponding procfile is created
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76) which is used for configuring this device. Thus, device names need to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77) be unique.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79) To support adding the same device to multiple threads, which is useful
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) with multi queue NICs, the device naming scheme is extended with "@":
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81) device@something
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83) The part after "@" can be anything, but it is custom to use the thread
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84) number.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86) Viewing devices
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87) ===============
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) The Params section holds configured information. The Current section
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) holds running statistics. The Result is printed after a run or after
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91) interruption. Example::
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) /proc/net/pktgen/eth4@0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) Params: count 100000 min_pkt_size: 60 max_pkt_size: 60
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96) frags: 0 delay: 0 clone_skb: 64 ifname: eth4@0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97) flows: 0 flowlen: 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98) queue_map_min: 0 queue_map_max: 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99) dst_min: 192.168.81.2 dst_max:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100) src_min: src_max:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) src_mac: 90:e2:ba:0a:56:b4 dst_mac: 00:1b:21:3c:9d:f8
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) udp_src_min: 9 udp_src_max: 109 udp_dst_min: 9 udp_dst_max: 9
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) src_mac_count: 0 dst_mac_count: 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) Flags: UDPSRC_RND NO_TIMESTAMP QUEUE_MAP_CPU
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105) Current:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) pkts-sofar: 100000 errors: 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107) started: 623913381008us stopped: 623913396439us idle: 25us
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108) seq_num: 100001 cur_dst_mac_offset: 0 cur_src_mac_offset: 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) cur_saddr: 192.168.8.3 cur_daddr: 192.168.81.2
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) cur_udp_dst: 9 cur_udp_src: 42
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111) cur_queue_map: 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) flows: 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) Result: OK: 15430(c15405+d25) usec, 100000 (60byte,0frags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) 6480562pps 3110Mb/sec (3110669760bps) errors: 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117) Configuring devices
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118) ===================
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) This is done via the /proc interface, and most easily done via pgset
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) as defined in the sample scripts.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) You need to specify PGDEV environment variable to use functions from sample
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) scripts, i.e.::
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124) export PGDEV=/proc/net/pktgen/eth4@0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125) source samples/pktgen/functions.sh
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127) Examples::
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) pg_ctrl start starts injection.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130) pg_ctrl stop aborts injection. Also, ^C aborts generator.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) pgset "clone_skb 1" sets the number of copies of the same packet
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) pgset "clone_skb 0" use single SKB for all transmits
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134) pgset "burst 8" uses xmit_more API to queue 8 copies of the same
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135) packet and update HW tx queue tail pointer once.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136) "burst 1" is the default
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137) pgset "pkt_size 9014" sets packet size to 9014
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138) pgset "frags 5" packet will consist of 5 fragments
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) pgset "count 200000" sets number of packets to send, set to zero
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140) for continuous sends until explicitly stopped.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142) pgset "delay 5000" adds delay to hard_start_xmit(). nanoseconds
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144) pgset "dst 10.0.0.1" sets IP destination address
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) (BEWARE! This generator is very aggressive!)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147) pgset "dst_min 10.0.0.1" Same as dst
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148) pgset "dst_max 10.0.0.254" Set the maximum destination IP.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149) pgset "src_min 10.0.0.1" Set the minimum (or only) source IP.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150) pgset "src_max 10.0.0.254" Set the maximum source IP.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151) pgset "dst6 fec0::1" IPV6 destination address
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152) pgset "src6 fec0::2" IPV6 source address
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153) pgset "dstmac 00:00:00:00:00:00" sets MAC destination address
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154) pgset "srcmac 00:00:00:00:00:00" sets MAC source address
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156) pgset "queue_map_min 0" Sets the min value of tx queue interval
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157) pgset "queue_map_max 7" Sets the max value of tx queue interval, for multiqueue devices
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158) To select queue 1 of a given device,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) use queue_map_min=1 and queue_map_max=1
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) pgset "src_mac_count 1" Sets the number of MACs we'll range through.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162) The 'minimum' MAC is what you set with srcmac.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164) pgset "dst_mac_count 1" Sets the number of MACs we'll range through.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165) The 'minimum' MAC is what you set with dstmac.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167) pgset "flag [name]" Set a flag to determine behaviour. Current flags
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) are: IPSRC_RND # IP source is random (between min/max)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) IPDST_RND # IP destination is random
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170) UDPSRC_RND, UDPDST_RND,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171) MACSRC_RND, MACDST_RND
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172) TXSIZE_RND, IPV6,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173) MPLS_RND, VID_RND, SVID_RND
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) FLOW_SEQ,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175) QUEUE_MAP_RND # queue map random
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) QUEUE_MAP_CPU # queue map mirrors smp_processor_id()
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177) UDPCSUM,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178) IPSEC # IPsec encapsulation (needs CONFIG_XFRM)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) NODE_ALLOC # node specific memory allocation
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) NO_TIMESTAMP # disable timestamping
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181) pgset 'flag ![name]' Clear a flag to determine behaviour.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182) Note that you might need to use single quote in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) interactive mode, so that your shell wouldn't expand
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184) the specified flag as a history command.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186) pgset "spi [SPI_VALUE]" Set specific SA used to transform packet.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) pgset "udp_src_min 9" set UDP source port min, If < udp_src_max, then
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189) cycle through the port range.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191) pgset "udp_src_max 9" set UDP source port max.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192) pgset "udp_dst_min 9" set UDP destination port min, If < udp_dst_max, then
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) cycle through the port range.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194) pgset "udp_dst_max 9" set UDP destination port max.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196) pgset "mpls 0001000a,0002000a,0000000a" set MPLS labels (in this example
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197) outer label=16,middle label=32,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198) inner label=0 (IPv4 NULL)) Note that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199) there must be no spaces between the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) arguments. Leading zeros are required.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201) Do not set the bottom of stack bit,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) that's done automatically. If you do
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203) set the bottom of stack bit, that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204) indicates that you want to randomly
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205) generate that address and the flag
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) MPLS_RND will be turned on. You
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207) can have any mix of random and fixed
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208) labels in the label stack.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) pgset "mpls 0" turn off mpls (or any invalid argument works too!)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212) pgset "vlan_id 77" set VLAN ID 0-4095
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213) pgset "vlan_p 3" set priority bit 0-7 (default 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) pgset "vlan_cfi 0" set canonical format identifier 0-1 (default 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216) pgset "svlan_id 22" set SVLAN ID 0-4095
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217) pgset "svlan_p 3" set priority bit 0-7 (default 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218) pgset "svlan_cfi 0" set canonical format identifier 0-1 (default 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) pgset "vlan_id 9999" > 4095 remove vlan and svlan tags
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221) pgset "svlan 9999" > 4095 remove svlan tag
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) pgset "tos XX" set former IPv4 TOS field (e.g. "tos 28" for AF11 no ECN, default 00)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225) pgset "traffic_class XX" set former IPv6 TRAFFIC CLASS (e.g. "traffic_class B8" for EF no ECN, default 00)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227) pgset "rate 300M" set rate to 300 Mb/s
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) pgset "ratep 1000000" set rate to 1Mpps
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230) pgset "xmit_mode netif_receive" RX inject into stack netif_receive_skb()
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231) Works with "burst" but not with "clone_skb".
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232) Default xmit_mode is "start_xmit".
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234) Sample scripts
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) ==============
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237) A collection of tutorial scripts and helpers for pktgen is in the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238) samples/pktgen directory. The helper parameters.sh file support easy
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239) and consistent parameter parsing across the sample scripts.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) Usage example and help::
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) ./pktgen_sample01_simple.sh -i eth4 -m 00:1B:21:3C:9D:F8 -d 192.168.8.2
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) Usage:::
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247) ./pktgen_sample01_simple.sh [-vx] -i ethX
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249) -i : ($DEV) output interface/device (required)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250) -s : ($PKT_SIZE) packet size
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251) -d : ($DEST_IP) destination IP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252) -m : ($DST_MAC) destination MAC-addr
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) -t : ($THREADS) threads to start
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254) -c : ($SKB_CLONE) SKB clones send before alloc new SKB
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255) -b : ($BURST) HW level bursting of SKBs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) -v : ($VERBOSE) verbose
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257) -x : ($DEBUG) debug
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259) The global variables being set are also listed. E.g. the required
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260) interface/device parameter "-i" sets variable $DEV. Copy the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261) pktgen_sampleXX scripts and modify them to fit your own needs.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263) The old scripts::
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265) pktgen.conf-1-2 # 1 CPU 2 dev
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266) pktgen.conf-1-1-rdos # 1 CPU 1 dev w. route DoS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267) pktgen.conf-1-1-ip6 # 1 CPU 1 dev ipv6
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268) pktgen.conf-1-1-ip6-rdos # 1 CPU 1 dev ipv6 w. route DoS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269) pktgen.conf-1-1-flows # 1 CPU 1 dev multiple flows.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272) Interrupt affinity
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273) ===================
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) Note that when adding devices to a specific CPU it is a good idea to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275) also assign /proc/irq/XX/smp_affinity so that the TX interrupts are bound
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276) to the same CPU. This reduces cache bouncing when freeing skbs.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278) Plus using the device flag QUEUE_MAP_CPU, which maps the SKBs TX queue
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279) to the running threads CPU (directly from smp_processor_id()).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281) Enable IPsec
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282) ============
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283) Default IPsec transformation with ESP encapsulation plus transport mode
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284) can be enabled by simply setting::
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286) pgset "flag IPSEC"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287) pgset "flows 1"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289) To avoid breaking existing testbed scripts for using AH type and tunnel mode,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290) you can use "pgset spi SPI_VALUE" to specify which transformation mode
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) to employ.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294) Current commands and configuration options
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295) ==========================================
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297) **Pgcontrol commands**::
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299) start
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300) stop
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301) reset
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303) **Thread commands**::
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305) add_device
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306) rem_device_all
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) **Device commands**::
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311) count
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) clone_skb
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313) burst
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) debug
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316) frags
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317) delay
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319) src_mac_count
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320) dst_mac_count
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322) pkt_size
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323) min_pkt_size
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324) max_pkt_size
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326) queue_map_min
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327) queue_map_max
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328) skb_priority
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330) tos (ipv4)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331) traffic_class (ipv6)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) mpls
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335) udp_src_min
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336) udp_src_max
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338) udp_dst_min
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) udp_dst_max
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341) node
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343) flag
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344) IPSRC_RND
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345) IPDST_RND
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346) UDPSRC_RND
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347) UDPDST_RND
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348) MACSRC_RND
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349) MACDST_RND
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350) TXSIZE_RND
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351) IPV6
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352) MPLS_RND
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353) VID_RND
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354) SVID_RND
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355) FLOW_SEQ
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356) QUEUE_MAP_RND
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357) QUEUE_MAP_CPU
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358) UDPCSUM
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359) IPSEC
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) NODE_ALLOC
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361) NO_TIMESTAMP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) spi (ipsec)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) dst_min
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366) dst_max
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368) src_min
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369) src_max
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371) dst_mac
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372) src_mac
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374) clear_counters
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376) src6
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377) dst6
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378) dst6_max
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379) dst6_min
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381) flows
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382) flowlen
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384) rate
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385) ratep
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387) xmit_mode <start_xmit|netif_receive>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389) vlan_cfi
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390) vlan_id
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391) vlan_p
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393) svlan_cfi
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 394) svlan_id
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 395) svlan_p
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 396)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398) References:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400) - ftp://robur.slu.se/pub/Linux/net-development/pktgen-testing/
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401) - tp://robur.slu.se/pub/Linux/net-development/pktgen-testing/examples/
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 402)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 403) Paper from Linux-Kongress in Erlangen 2004.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 404) - ftp://robur.slu.se/pub/Linux/net-development/pktgen-testing/pktgen_paper.pdf
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 405)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 406) Thanks to:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 407)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 408) Grant Grundler for testing on IA-64 and parisc, Harald Welte, Lennert Buytenhek
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 409) Stephen Hemminger, Andi Kleen, Dave Miller and many others.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 410)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 411)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 412) Good luck with the linux net-development.