- Notifications
You must be signed in to change notification settings - Fork759
A Crypto-Secure Reliable-UDP Library for golang with FEC
License
xtaci/kcp-go
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
kcp-go is aReliable-UDP library forgolang.
This library is designed to providesmooth, resilient, ordered, error-checked and anonymous delivery of streams overUDP packets. It has been battle-tested with the open-source projectkcptun. Millions of devices, ranging from low-end MIPS routers to high-end servers, have deployed kcp-go-powered programs in various applications, includingonline games, live broadcasting, file synchronization, and network acceleration.
- Designed forlatency-sensitive scenarios.
- Cache-friendly andmemory-optimized design, offering extremelyhigh performance core.
- Handles>5K concurrent connections on a single commodity server.
- Compatible withnet.Conn andnet.Listener, serving as a drop-in replacement fornet.TCPConn.
- FEC (Forward Error Correction) support withReed-Solomon Codes.
- Packet-level encryption support withAES,TEA,3DES,Blowfish,Cast5,Salsa20, etc., inCFB mode, generating completely anonymous packets.
- Onlya fixed number of goroutines are created for the entire server application, with costs incontext switching between goroutines taken into consideration.
- Compatible withskywind3000's C version with various improvements.
- Platform-dependent optimizations:sendmmsg andrecvmmsg exploited for Linux.
For complete documentation, see the associatedGodoc.
NONCE: 16bytes cryptographically secure random number, nonce changes for every packet. CRC32: CRC-32 checksum of data using the IEEE polynomial FEC TYPE: typeData = 0xF1 typeParity = 0xF2 FEC SEQID: monotonically increasing in range: [0, (0xffffffff/shardSize) * shardSize - 1] SIZE: The size of KCP frame plus 2KCP Header+------------------+| conv uint32 |+------------------+| cmd uint8 |+------------------+| frg uint8 |+------------------+| wnd uint16 |+------------------+| ts uint32 |+------------------+| sn uint32 |+------------------+| una uint32 |+------------------+| rto uint32 |+------------------+| xmit uint32 |+------------------+| resendts uint32 |+------------------+| fastack uint32 |+------------------+| acked uint32 |+------------------+| data []byte |+------------------++-----------------+| SESSION |+-----------------+| KCP(ARQ) |+-----------------+| FEC(OPTIONAL) |+-----------------+| CRYPTO(OPTIONAL)|+-----------------+| UDP(PACKET) |+-----------------+| IP |+-----------------+| LINK |+-----------------+| PHY |+-----------------+- https://github.com/xtaci/libkcp -- FEC enhanced KCP session library for iOS/Android in C++
===Model Name:MacBook ProModel Identifier:MacBookPro14,1Processor Name:Intel Core i5Processor Speed:3.1 GHzNumber of Processors:1Total Number of Cores:2L2 Cache (per Core):256 KBL3 Cache:4 MBMemory:8 GB===$ go test -v -run=^$ -bench .beginning tests, encryption:salsa20, fec:10/3goos: darwingoarch: amd64pkg: github.com/xtaci/kcp-goBenchmarkSM4-4 50000 32180 ns/op 93.23 MB/s 0 B/op 0 allocs/opBenchmarkAES128-4 500000 3285 ns/op 913.21 MB/s 0 B/op 0 allocs/opBenchmarkAES192-4 300000 3623 ns/op 827.85 MB/s 0 B/op 0 allocs/opBenchmarkAES256-4 300000 3874 ns/op 774.20 MB/s 0 B/op 0 allocs/opBenchmarkTEA-4 100000 15384 ns/op 195.00 MB/s 0 B/op 0 allocs/opBenchmarkXOR-4 20000000 89.9 ns/op33372.00 MB/s 0 B/op 0 allocs/opBenchmarkBlowfish-4 50000 26927 ns/op 111.41 MB/s 0 B/op 0 allocs/opBenchmarkNone-4 30000000 45.7 ns/op65597.94 MB/s 0 B/op 0 allocs/opBenchmarkCast5-4 50000 34258 ns/op 87.57 MB/s 0 B/op 0 allocs/opBenchmark3DES-4 10000 117149 ns/op 25.61 MB/s 0 B/op 0 allocs/opBenchmarkTwofish-4 50000 33538 ns/op 89.45 MB/s 0 B/op 0 allocs/opBenchmarkXTEA-4 30000 45666 ns/op 65.69 MB/s 0 B/op 0 allocs/opBenchmarkSalsa20-4 500000 3308 ns/op 906.76 MB/s 0 B/op 0 allocs/opBenchmarkCRC32-4 20000000 65.2 ns/op15712.43 MB/sBenchmarkCsprngSystem-4 1000000 1150 ns/op 13.91 MB/sBenchmarkCsprngMD5-4 10000000 145 ns/op 110.26 MB/sBenchmarkCsprngSHA1-4 10000000 158 ns/op 126.54 MB/sBenchmarkCsprngNonceMD5-4 10000000 153 ns/op 104.22 MB/sBenchmarkCsprngNonceAES128-4 100000000 19.1 ns/op 837.81 MB/sBenchmarkFECDecode-4 1000000 1119 ns/op1339.61 MB/s 1606 B/op 2 allocs/opBenchmarkFECEncode-4 2000000 832 ns/op1801.83 MB/s 17 B/op 0 allocs/opBenchmarkFlush-4 5000000 272 ns/op 0 B/op 0 allocs/opBenchmarkEchoSpeed4K-4 5000 259617 ns/op 15.78 MB/s 5451 B/op 149 allocs/opBenchmarkEchoSpeed64K-4 1000 1706084 ns/op 38.41 MB/s 56002 B/op 1604 allocs/opBenchmarkEchoSpeed512K-4 100 14345505 ns/op 36.55 MB/s 482597 B/op 13045 allocs/opBenchmarkEchoSpeed1M-4 30 34859104 ns/op 30.08 MB/s 1143773 B/op 27186 allocs/opBenchmarkSinkSpeed4K-4 50000 31369 ns/op 130.57 MB/s 1566 B/op 30 allocs/opBenchmarkSinkSpeed64K-4 5000 329065 ns/op 199.16 MB/s 21529 B/op 453 allocs/opBenchmarkSinkSpeed256K-4 500 2373354 ns/op 220.91 MB/s 166332 B/op 3554 allocs/opBenchmarkSinkSpeed1M-4 300 5117927 ns/op 204.88 MB/s 310378 B/op 6988 allocs/opPASSok github.com/xtaci/kcp-go50.349s=== Raspberry Pi 4 ===➜ kcp-go git:(master) cat /proc/cpuinfoprocessor: 0model name: ARMv7 Processor rev 3 (v7l)BogoMIPS: 108.00Features: half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae evtstrm crc32CPU implementer: 0x41CPU architecture: 7CPU variant: 0x0CPU part: 0xd08CPU revision: 3➜ kcp-go git:(master) go test -run=^$ -bench .2020/01/05 19:25:13 beginning tests, encryption:salsa20, fec:10/3goos: linuxgoarch: armpkg: github.com/xtaci/kcp-go/v5BenchmarkSM4-4 20000 86475 ns/op 34.69 MB/s 0 B/op 0 allocs/opBenchmarkAES128-4 20000 62254 ns/op 48.19 MB/s 0 B/op 0 allocs/opBenchmarkAES192-4 20000 71802 ns/op 41.78 MB/s 0 B/op 0 allocs/opBenchmarkAES256-4 20000 80570 ns/op 37.23 MB/s 0 B/op 0 allocs/opBenchmarkTEA-4 50000 37343 ns/op 80.34 MB/s 0 B/op 0 allocs/opBenchmarkXOR-4 100000 22266 ns/op 134.73 MB/s 0 B/op 0 allocs/opBenchmarkBlowfish-4 20000 66123 ns/op 45.37 MB/s 0 B/op 0 allocs/opBenchmarkNone-4 3000000 518 ns/op 5786.77 MB/s 0 B/op 0 allocs/opBenchmarkCast5-4 20000 76705 ns/op 39.11 MB/s 0 B/op 0 allocs/opBenchmark3DES-4 5000 418868 ns/op 7.16 MB/s 0 B/op 0 allocs/opBenchmarkTwofish-4 5000 326896 ns/op 9.18 MB/s 0 B/op 0 allocs/opBenchmarkXTEA-4 10000 114418 ns/op 26.22 MB/s 0 B/op 0 allocs/opBenchmarkSalsa20-4 50000 36736 ns/op 81.66 MB/s 0 B/op 0 allocs/opBenchmarkCRC32-4 1000000 1735 ns/op 589.98 MB/sBenchmarkCsprngSystem-4 1000000 2179 ns/op 7.34 MB/sBenchmarkCsprngMD5-4 2000000 811 ns/op 19.71 MB/sBenchmarkCsprngSHA1-4 2000000 862 ns/op 23.19 MB/sBenchmarkCsprngNonceMD5-4 2000000 878 ns/op 18.22 MB/sBenchmarkCsprngNonceAES128-4 5000000 326 ns/op 48.97 MB/sBenchmarkFECDecode-4 200000 9081 ns/op 165.16 MB/s 140 B/op 1 allocs/opBenchmarkFECEncode-4 100000 12039 ns/op 124.59 MB/s 11 B/op 0 allocs/opBenchmarkFlush-4 100000 21704 ns/op 0 B/op 0 allocs/opBenchmarkEchoSpeed4K-4 2000 981182 ns/op 4.17 MB/s 12384 B/op 424 allocs/opBenchmarkEchoSpeed64K-4 100 10503324 ns/op 6.24 MB/s 123616 B/op 3779 allocs/opBenchmarkEchoSpeed512K-4 20 138633802 ns/op 3.78 MB/s 1606584 B/op 29233 allocs/opBenchmarkEchoSpeed1M-4 5 372903568 ns/op 2.81 MB/s 4080504 B/op 63600 allocs/opBenchmarkSinkSpeed4K-4 10000 121239 ns/op 33.78 MB/s 4647 B/op 104 allocs/opBenchmarkSinkSpeed64K-4 1000 1587906 ns/op 41.27 MB/s 50914 B/op 1115 allocs/opBenchmarkSinkSpeed256K-4 100 16277830 ns/op 32.21 MB/s 453027 B/op 9296 allocs/opBenchmarkSinkSpeed1M-4 100 31040703 ns/op 33.78 MB/s 898097 B/op 18932 allocs/opPASSok github.com/xtaci/kcp-go/v5 64.151s
kcp.flush() loops through the send queue for retransmission checking every 20 ms.
I wrote a benchmark comparing sequential loops through aslice and acontainer/listhere:
BenchmarkLoopSlice-4 2000000000 0.39 ns/opBenchmarkLoopList-4 100000000 54.6 ns/opThe list structure introducesheavy cache misses compared to the slice, which has betterlocality. For 5,000 connections with a 32-window size and a 20 ms interval, using a slice costs 6 μs (0.03% CPU) perkcp.flush(), while using a list costs 8.7 ms (43.5% CPU).
Timing iscritical to theRTT estimator. Inaccurate timing leads to false retransmissions in KCP, but callingtime.Now() costs 42 cycles (10.5 ns on a 4 GHz CPU, 15.6 ns on my MacBook Pro 2.7 GHz).
The benchmark fortime.Now() ishere:
BenchmarkNow-4 100000000 15.6 ns/opIn kcp-go, after eachkcp.output() function call, the current clock time is updated upon return. For a singlekcp.flush() operation, the current time is queried from the system once. For 5,000 connections, this costs 5000 * 15.6 ns = 78 μs (a fixed cost when no packet needs to be sent). For 10 MB/s data transfer with a 1400 MTU,kcp.output() is called around 7500 times, costing 117 μs fortime.Now() every second.
Primary memory allocation is done from a global buffer pool,xmit.Buf. In kcp-go, when we need to allocate some bytes, we get them from that pool, which returns a fixed-capacity 1500 bytes (mtuLimit). The rx queue, tx queue, and fec queue all receive bytes from this pool and return them after use to prevent unnecessary zeroing of bytes. The pool mechanism maintains a high watermark for slice objects, allowing these in-flight objects to survive periodic garbage collection, while also being able to return memory to the runtime when idle.
kcp-go is shipped with built-in packet encryption powered by various block encryption algorithms and works inCipher Feedback Mode. For each packet to be sent, the encryption process starts by encrypting anonce from thesystem entropy, ensuring that encryption of the same plaintext never results in the same ciphertext.
The contents of the packets are completely anonymous with encryption, including the headers (FEC, KCP), checksums, and contents. Note that no matter which encryption method you choose at the upper layer, if you disable encryption, the transmission will be insecure, as the header isplaintext and susceptible to tampering, such as jamming thesliding window size,round-trip time,FEC properties, andchecksums.AES-128 is suggested for minimal encryption, as modern CPUs come withAES-NI instructions and perform better thansalsa20 (check the table above).
Other possible attacks on kcp-go include:
- Traffic analysis: Data flow on specific websites may have patterns while exchanging data. This type of eavesdropping has been mitigated by adoptingsmux to mix data streams and introduce noise. A perfect solution has not yet appeared, but theoretically, shuffling/mixing messages on a larger scale network may mitigate this problem.
- Replay attack: Since asymmetrical encryption has not been introduced into kcp-go, capturing packets and replaying them on a different machine is possible. Note that hijacking the session and decrypting the contents is stillimpossible. Upper layers should use an asymmetrical encryption system to guarantee the authenticity of each message (to process each message exactly once), such as HTTPS/OpenSSL/LibreSSL. Signing requests with private keys can eliminate this type of attack.
Control messages likeSYN/FIN/RST in TCPare not defined in KCP. You need akeepalive/heartbeat mechanism at the application level. A real-world example is to use amultiplexing protocol over the session, such assmux (which has an embedded keepalive mechanism). Seekcptun for an example.
Q: I'm handling >5K connections on my server, and the CPU utilization is so high.
A: A standaloneagent orgate server for running kcp-go is suggested, not only to reduce CPU utilization but also to improve theprecision of RTT measurements (timing), which indirectly affects retransmission. Increasing the updateinterval withSetNoDelay, such asconn.SetNoDelay(1, 40, 1, 1), will dramatically reduce system load but may lower performance.
Q: When should I enable FEC?
A: Forward error correction is critical for long-distance transmission because packet loss incurs a huge time penalty. In the complex packet routing networks of the modern world, round-trip time-based loss checks are not always efficient. The significant deviation of RTT samples over long distances usually leads to a larger RTO value in typical RTT estimators, which slows down the transmission.
Q: Should I enable encryption?
A: Yes, for the security of the protocol, even if the upper layer has encryption.
- https://github.com/xtaci/kcptun -- A Secure Tunnel Based On KCP over UDP.
- https://github.com/getlantern/lantern -- Lantern delivers fast access to the open Internet.
- https://github.com/smallnest/rpcx -- A RPC service framework based on net/rpc like alibaba Dubbo and weibo Motan.
- https://github.com/gonet2/agent -- A gateway for games with stream multiplexing.
- https://github.com/syncthing/syncthing -- Open Source Continuous File Synchronization.
- https://github.com/xtaci/smux/ -- A Stream Multiplexing Library for golang with least memory
- https://github.com/xtaci/libkcp -- FEC enhanced KCP session library for iOS/Android in C++
- https://github.com/skywind3000/kcp -- A Fast and Reliable ARQ Protocol
- https://github.com/klauspost/reedsolomon -- Reed-Solomon Erasure Coding in Go
About
A Crypto-Secure Reliable-UDP Library for golang with FEC
Topics
Resources
License
Uh oh!
There was an error while loading.Please reload this page.
Stars
Watchers
Forks
Packages0
Uh oh!
There was an error while loading.Please reload this page.