Review - 16423 Import fletcher-4 algorithms from OpenZFS

Review - 16423 Import fletcher-4 algorithms from OpenZFS

[Andy Fiddaman]

I've imported changes from OpenZFS that add alternative algorithms for
calculating Fletcher 4 checksums, most of which are significantly faster
than what we have now. Like the raidz maths algorithms, a mini benchmark
is run during boot and the fastest algorithms are independently chosen
for both native and byteswapped data. There's more data in the ticket
including a flamegraph that one of my colleagues generated with the
existing scalar algorithm which shows how much time is spent doing
checksumming in a particular workload, and there are also links to the
OpenZFS commits there.

Please can you review?

    16423 Import fletcher-4 algorithms from OpenZFS
    Portions contributed by: Attila Fülöp <attila@fueloep.org>
    Portions contributed by: Brian Behlendorf <behlendorf1@llnl.gov>
    Portions contributed by: David Quigley <david.quigley@intel.com>
    Portions contributed by: Gvozden Nešković <neskovic@gmail.com>
    Portions contributed by: Jinshan Xiong <jinshan.xiong@intel.com>
    Portions contributed by: Richard Yao <richard.yao@alumni.stonybrook.edu>
    Portions contributed by: Romain Dolbeau <romain.github@dolbeau.name>
    Portions contributed by: Tyler J. Stachecki <stachecki.tyler@gmail.com>
    https://www.illumos.org/issues/16423
    https://code.illumos.org/c/illumos-gate/+/3386

Thanks,

Andy