RFR: 8292898: [vectorapi] Unify vector mask cast operation [v4]

[Jatin Bhateja]

On Fri, 23 Sep 2022 10:15:55 GMT, Quan Anh Mai <duke at openjdk.org> wrote:

>> src/hotspot/share/opto/vectorIntrinsics.cpp line 2559:
>> 
>>> 2557:           op = gvn().transform(new VectorMaskCastNode(op, dst_type));
>>> 2558:         } else {
>>> 2559:           op = VectorMaskCastNode::makeCastNode(&gvn(), op, dst_type);
>> 
>> Masks are either predicated registers or vectors, in former case cast is a no-op and in later case mask casting is similar to vector casting where each vector lane is holding either a 0 or -1 value and we leverage backend implementation for existing VectorCastX2Y IR nodes.  Unification here in a way duplicating some backed implementation at least for X86.
>
> I believe the intention is to simplify the mid-end as well as to achieve better code gen on x86

Hi @merykitty , Can you please post the benchmark results with the new changes.

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

PR: https://git.openjdk.org/jdk/pull/10192