Awesome

SMhasher

The sortable table variants:

• Default AMD Ryzen 5 3350G 3.6GHz
• Intel i5-2300 2.8GHz
• Intel i5-2300 2.8GHz 32bit
• AMD Ryzen 3 3200U 3.5GHz
• Mac Air i7-4650
• Cortex-A53 2GHz (Sony XPeria L4)

Summary

I added some SSE assisted hashes and fast intel/arm CRC32-C, AES and SHA HW variants. See also the old https://github.com/aappleby/smhasher/wiki, the improved, but unmaintained fork https://github.com/demerphq/smhasher, and the new improved version SMHasher3 https://gitlab.com/fwojcik/smhasher3.

So the fastest hash functions on x86_64 without quality problems are:

• rapidhash (an improved wyhash)
• xxh3low
• wyhash
• umash (even universal!)
• ahash64
• t1ha2_atonce
• komihash
• FarmHash (not portable, too machine specific: 64 vs 32bit, old gcc, ...)
• halftime_hash128
• Spooky32
• pengyhash
• nmhash32
• mx3
• MUM/mir (different results on 32/64-bit archs, lots of bad seeds to filter out)
• fasthash32

Hash functions for symbol tables or hash tables typically use 32 bit hashes, for databases, file systems and file checksums typically 64 or 128bit, for crypto now starting with 256 bit.

Typical median key size in perl5 is 20, the most common 4. Similar for all other dynamic languages. See github.com/rurban/perl-hash-stats

When used in a hash table the instruction cache will usually beat the CPU and throughput measured here. In my tests the smallest FNV1A beats the fastest crc32_hw1 with Perl 5 hash tables. Even if those worse hash functions will lead to more collisions, the overall speed advantage and inline-ability beats the slightly worse quality. See e.g. A Seven-Dimensional Analysis of Hashing Methods and its Implications on Query Processing for a concise overview of the best hash table strategies, confirming that the simplest Mult hashing (bernstein, FNV*, x17, sdbm) always beat "better" hash functions (Tabulation, Murmur, Farm, ...) when used in a hash table.

The fast hash functions tested here are recommendable as fast for file digests and maybe bigger databases, but not for 32bit hash tables. The "Quality problems" lead to less uniform distribution, i.e. more collisions and worse performance, but are rarely related to real security attacks, just the 2nd sanity AppendZeroes test against \0 invariance is security relevant.

Columns

MiB/sec: The average of the Bulk key speed test for alignments 0-7 with 262144-byte keys. The higher the better.

cycl./hash: The average of the Small key speed test for 1-31 byte keys. The smaller the better.

cycl./map: The result of the Hashmap test for /usr/dict/words with fast C++ hashmap get queries, with the standard deviation in brackets. This tests the inlinability of the hash function in practise (see size). The smaller the better.

size: The object size in byte on AMD64. This affects the inlinability in e.g. hash tables. The smaller the better.

Quality problems: See the failures in the linked doc. The less the better.

Other

• https://github.com/martinus/better-faster-stronger-mixer Mixers. And in his blog the best C++ hashmap benchmarks.
• http://nohatcoder.dk/2019-05-19-1.html gives a new, useful hash level classification 1-5.
• http://www.strchr.com/hash_functions lists other benchmarks and quality of most simple and fast hash functions.
• http://bench.cr.yp.to/primitives-hash.html lists the benchmarks of all currently tested secure hashes.
• http://valerieaurora.org/hash.html Lifetimes of cryptographic hash functions

SECURITY

The hash table attacks described in SipHash against City, Murmur or Perl JenkinsOAAT or at Hash Function Lounge are not included here. We list some known attacks at GH #186.

Such an attack avoidance cannot be the problem of the hash function, but only the hash table collision resolution scheme. You can attack every single hash function, even the best and most secure if you detect the seed, e.g. from language (mis-)features, side-channel attacks, collision timings and independly the sort-order, so you need to protect your collision handling scheme from the worst-case O(n), i.e. separate chaining with linked lists. Linked lists chaining allows high load factors, but is very cache-unfriendly. The only recommendable linked list scheme is inlining the key or hash into the array. Nowadays everybody uses fast open addressing, even if the load factor needs to be ~50%, unless you use Cuckoo Hashing.

I.e. the usage of SipHash for their hash table in Python 3.4, ruby, rust, systemd, OpenDNS, Haskell and OpenBSD is pure security theatre. SipHash is not secure enough for security purposes and not fast enough for general usage. Brute-force generation of ~32k collisions need 2-4m for all these hashes. siphash being the slowest needs max 4m, other typically max 2m30s, with <10s for practical 16k collision attacks with all hash functions. Using Murmur is usually slower than a simple Mult, even in the worst case. Provable secure is only uniform hashing, i.e. 2-5 independent Mult or Tabulation, or using a guaranteed logarithmic collision scheme (a tree) or a linear collision scheme, such as swisstable/folly-F14, Robin Hood or Cuckoo hashing with collision counting.

One more note regarding security: Nowadays even SHA1 can be solved in a solver, like Z3 (or faster ones) for practical hash table collision attacks (i.e. 14-20 bits). All hash functions with less than 160 bits tested here cannot be considered "secure" at all.

The '\0' vulnerability attack with binary keys is tested in the 2nd Sanity Zero test.

CRYPTO

Our crypto hashes are hardened with an added size_t seed, mixed into the initial state, and the versions which require zero-padding are hardened by adding the len also, to prevent from collisions with AppendedZeroes for the padding. The libtomcrypt implementations already provide for that, but others might not. Without, such crypto hash functions are unsuitable for normal tasks, as it's trivial to create collisions by padding or bad seeds.

The official NIST hash function testsuite does not do such extensive statistical tests, to search for weak ranges in the bits. Also crypto does not change the initial state, which we do here for our random 32bit seed. Crypto mostly cares about unreversable key -> hash functions without changing the initial fixed state and timings/sidechannel attacks.

The NIST "Cryptographic Algorithm Validation Program" (CAVP) involves the testing of the implementations of FIPS-approved and NIST-recommended cryptographic algorithms. During the NIST SHA-3 competition, the testing methodology was borrowed from the "CAVP", as the KATs and MCTs of the SHA-3 Competition Test Suite were based on the CAVP tests for SHA-2. In addition to this, the “Extremely Long Message Test,” not present in the CAVP for SHA-2, required the submitters to generate the hash value corresponding to a message with a length of 1 GiB. “NIST - Cryptographic Algorithm Validation Program (CAVP),” June 2017. Available: http://csrc.nist.gov/groups/STM/cavp (No testing source code provided, just high-level descriptions)

Two other independent third party testsuites found an extensive number of bugs and weaknesses in the SHA3 candidates. "Finding Bugs in Cryptographic Hash Function Implementations", Nicky Mouha, Mohammad S Raunak, D. Richard Kuhn, and Raghu Kacker, 2017. https://eprint.iacr.org/2017/891.pdf

Maybe independent researchers should come together to do a better public SHA-4 round, based on better and more testing methods, open source code for the tests, and using standard industry practices, such as valgrind, address-sanitizer and ubsan to detect obvious bugs.

PROBLEMS

• Bad Seeds

  Hash functions are typically initialized with a random seed. But some seed values may lead to bad hash functions, regardless of the key. In the regular case with random seeds the probablity of such bad seeds is very low, like 2^32 or 2^64. A practical application needs to know if bad seeds exist and choose another one. See e.g. mirhash_seed_init() and mirhash_bad_seeds() in Hashes.h. Note that a bad seed is not really a problem when you skip this seed during initialization. It can still be a GOOD or recommended hash function. But a bad seed of 0 leading to collisions is considered a bug, a bad hash function.

  We test for internal secrets, if they will be multiplied with 0. This is also called "blinding multiplication". main.cpp lists some secrets for each hash function we test against. The function <hash>_bad_seeds() lists the confirmed bad seeds.

  Special care needs to be taken for crc, most FNV1 variants, fletcher, Jenkins. And with GOOD hashes most MUM variants, like mirhash, MUM, wyhash.

  Independently from this, when the attacker knows the seed it will lead to DDOS attacks. Even with crypto hashes in power2 hashtables.

Typical undefined behaviour (UB) problems:

• Misaligned

  Many word-wise hashes (in opposite to safe byte-wise processing) don't check the input buffer for proper word alignment, which will fail with ubsan or Sparc. word being int32_t or int64_t or even more. On some old RISC hardware this will be a BUS error, you can even let Intel HW generate such a bus error by setting some CPU flag. But generally using misaligned accesses is fine.

  These are: mx3, Spooky, mirhash (but not strict), MUM, fasthash, Murmur3*, Murmur2*, metrohash* (all but cmetro*), Crap8, beamsplitter, lookup3, fletcher4, fletcher2, all sanmayce FNV1a_ variants (FNV1a_YT, FNV1A_Pippip_Yurii, FNV1A_Totenschiff, ...), fibonacci.

  The usual mitigation is to check the buffer alignment either in the caller, provide a pre-processing loop for the misaligned prefix, or copy the whole buffer into a fresh aligned area. Put that extra code inside #ifdef HAVE_ALIGNED_ACCESS_REQUIRED.

• oob - Out of bounds

  Some hash function assume a padded input buffer which can be accessed past its length up to the word size. This allows for faster loop processing, as no 2nd loop or switch table for the rest is needed, but it requires a cooperative calling enviroment and is as such considered cheating.

• Signed integer overflow

  A simple type error, this hash needs to use unsigned integer types internally, to avoid undefined and inconsistent behaviour. i.e. SuperFastHash: signed integer overflow: -2147483641 + -113 cannot be represented in type 'int'

• shift exponent overflow

  With: FNV1A_Pippip_Yurii, FNV1A_Totenschiff, pair_multiply_shift, sumhash32 shift exponent 64 is too large for 64-bit type 'long unsigned int'