[vectorIntrinsics+mask] RFR: 8266287: Basic mask IR implementation for the Vector API masking feature support [v2]

[Xiaohong Gong]

On Wed, 30 Jun 2021 16:49:10 GMT, Jatin Bhateja <jbhateja at openjdk.org> wrote:

>>> Hi @XiaohongGong , I just missed noticing this query earlier, apologies!, I think VectorLoadMask and VectorStoreMask serves the purpose to load/store a raw mask (boolean array) from/to a vector/predicated register. An explicit mask casting IR (VectorMaskCastNode) has been introduced recently which can be plugged after loading mask value if casting is needed.
>> 
>> It's ok. I think it's ok to use `VectorMaskCastNode` (actually we think this node is more reasonable instead of `VectorLoadMask/VectorStoreMask`). It needs a further consideration for it since we'd better to revisit its usage for all platforms while not only SVE.
>
> Hi @XiaohongGong ,  I think there are still few opens like StoreVectorMask/LoadVectorMask which may be specific for ARM or creation of different nodes for logic mask operation (OrVMask/AndVMask/XorVMask) etc, since we are anyways working on a forked branch from vectorIntrinsics any further tuning can be done later and it will give us an opportunity to start the backend work using common IR.
> 
> Hi @PaulSandoz , @sviswa7 , what are your thoughts.

Thanks for the review @jatin-bhateja @sviswa7 @iwanowww @PaulSandoz !

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

PR: https://git.openjdk.java.net/panama-vector/pull/78