[cryptography] [zfs] [Review] 4185 New hash algorithm support

CodesInChaos codesinchaos at gmail.com
Mon Oct 21 13:45:55 EDT 2013


> If that is all you want, have you considered SipHash? It is much faster
> than the other algorithms, yet more secure than CityHash, Murmurhash and
> friends. And it provides an IV/salt to make it per instance unique.

Is SipHash really that fast in this context? AFAIK it's only much
faster for short strings, since its block size is so small.

The downsides of SipHash are:

* lack of collision resistance when the key is known
* small 64 bit output, which means that collisions will happen
frequently and need to be handled

SipHash is great for use in hash tables, where it's fine if a bucket
contains two or three items, but I don't think it's a good match for ZFS.


More information about the cryptography mailing list