Matt Brown 751ba79cc5 lib/raid6/altivec: Add vpermxor implementation for raid6 Q syndrome
This patch uses the vpermxor instruction to optimise the raid6 Q
syndrome. This instruction was made available with POWER8, ISA version
2.07. It allows for both vperm and vxor instructions to be done in a
single instruction. This has been tested for correctness on a ppc64le
vm with a basic RAID6 setup containing 5 drives.

The performance benchmarks are from the raid6test in the
/lib/raid6/test directory. These results are from an IBM Firestone
machine with ppc64le architecture. The benchmark results show a 35%
speed increase over the best existing algorithm for powerpc (altivec).
The raid6test has also been run on a big-endian ppc64 vm to ensure it
also works for big-endian architectures.

Performance benchmarks:
  raid6: altivecx4 gen() 18773 MB/s
  raid6: altivecx8 gen() 19438 MB/s

  raid6: vpermxor4 gen() 25112 MB/s
  raid6: vpermxor8 gen() 26279 MB/s

Signed-off-by: Matt Brown <matthew.brown.dev@gmail.com>
Reviewed-by: Daniel Axtens <dja@axtens.net>
[mpe: Add VPERMXOR macro so we can build with old binutils]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2018-03-20 16:47:25 +11:00
..
2018-01-21 01:06:10 +11:00
2015-05-28 15:31:53 +02:00
2017-08-15 22:55:58 +10:00
2015-01-12 05:04:13 -08:00
2015-02-03 18:47:18 +01:00
2016-03-04 23:50:27 -06:00
2016-03-01 19:27:20 +11:00
2017-04-21 21:08:47 +10:00
2017-06-28 06:54:33 -07:00
2016-08-05 09:00:54 -04:00
2018-02-10 13:16:35 -08:00
2018-01-21 23:21:14 +11:00
2018-02-02 10:01:04 -08:00
2017-11-16 12:47:46 -08:00
2017-11-16 12:47:46 -08:00
2018-01-21 01:06:10 +11:00
2016-06-14 13:58:26 +10:00
2018-01-24 11:42:58 +11:00
2016-06-14 13:58:26 +10:00
2018-01-21 22:43:43 +11:00
2017-08-31 14:26:56 +10:00
2017-11-16 12:47:46 -08:00
2018-01-15 09:35:26 +01:00
2017-11-16 12:47:46 -08:00
2016-06-14 13:58:26 +10:00
2017-11-12 09:03:10 +11:00
2015-05-05 13:35:39 -06:00
2018-02-10 13:16:35 -08:00