[foreign-memaccess+abi] RFR: Add benchmarks to MemorySegmentVsBits

[Maurizio Cimadamore]

On Tue, 3 Jan 2023 19:03:59 GMT, Paul Sandoz <psandoz at openjdk.org> wrote:

> > All checks are hoisted out of the hot loop - so in principle the segment code should vectorize as well?
> 
> The addressing modes of loads from the long array and stores to the underlying memory are slightly different, but connected via the loop variable. I wonder if additional layers involved with memory segments means C2 is loosing that information, which makes the auto-vectorizer bail out.

I thought of that - but not sure why the buffer benchmark is also slow - heap buffers are implemented using array access aren't they?

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

PR: https://git.openjdk.org/panama-foreign/pull/762