RFR: 8256438: AArch64: Implement match rules with ROR shift register value

[Andrew Haley]

On 1/29/21 8:35 AM, Eric Liu wrote:
> I benchmarked on 7 platforms with jmh test[1], most of them were hard to see any performance changes. On one platform the performance become bad, and seems a little unstable.

Wow. I'll have to look at this.

In theory we prefer combined instructions because it should reduce code
size. However, we can't escape the possibility that on some implementations
combined instructions might cause pipeline stalls.

-- 
Andrew Haley  (he/him)
Java Platform Lead Engineer
Red Hat UK Ltd. <https://www.redhat.com>
https://keybase.io/andrewhaley
EAC8 43EB D3EF DB98 CC77 2FAD A5CD 6035 332F A671