RFR(S): 8224580: Matcher can cause oop field/array element to be reloaded

[Roland Westrelin]

> Yes, sure. My question was, if the code-quality-problem is a general
> problem, or would only affect offset==0 case, because as I understood,
> what you changed is set_shared() for offset==0 case too, and other cases
> have already been covered?

To be clear, there's no code quality or performance problem unless
there's a bug in the change I propose. What it does as far I understand
is simply extend the current behavior for LoadN to LoadN at offset
0. Now if I'm wrong it could introduce a code quality/performance
regression that could affect all LoadN, not only those at offset 0.

Roland.