[OpenIndiana-discuss] RAIDZ performance

[Sašo Kiselkov]

On 02/22/2013 02:16 AM, Timothy Coalson wrote:
> The first parity uses straight XOR on uint64_t, while the second parity
> performs the LFSR on all bytes in a uint64_t with some bitwise math (search
> for VDEV_RAIDZ_64MUL_2) that adds up to 8 operators by my count, followed
> by xor - using the LFSR lookup table on each byte might have a chance at
> being faster, but its hard to know without testing.  I am not at all
> surprised it takes significantly more cpu time, and I see no way to make it
> significantly closer in speed to a simple XOR.

I just tested by replacing some copy-loops with memcpy and restructuring
some conditionals and got some appreciable speed gains. I wouldn't want
to jump the gun yet, though.

> However, the test indicated more than 1GB/s, is the N40L's processor really
> 10x slower per core than the machine you ran the test on?  I mean, sure,
> its from 2010 and running at 1.5GHz, but still...

I probably forgot to mention that these tests were conducted on an
Athlon II Neo N36L @ 1.3GHz - pretty close to the N40L.

Cheers,
--
Saso