RFR: 7903912: JMH: Pad using a mix of bytes and longs [v2]

Tue Jan 7 14:32:58 UTC 2025

On Tue, 7 Jan 2025 14:22:49 GMT, Aleksey Shipilev <shade at openjdk.org> wrote:

> One sec, let's think a bit about the worst case here. If we are accepting that `boolean`-s can be depleted to fill the gaps in superclasses, then only `long`-s are left for padding. That means we actually want 16 booleans and 32 longs? I think it is okay to over-pad here a bit.

I think we better do this, while we are at it ^^^

-------------

PR Comment: https://git.openjdk.org/jmh/pull/147#issuecomment-2575437413