Do not CSE constant operands for GT_SIMD vector constructors

[TIHan]

Description
If a GT_SIMD op contains all constants and is an Init/InitN, then mark each constant as DoNotCSE. This logic already exists for GT_HWINTRINSIC nodes when creating vectors.

Acceptance Criteria

• CI passes

Diffs

[ghost]

Tagging subscribers to this area: @JulieLeeMSFT, @jakobbotsch
See info in area-owners.md if you want to be subscribed.

Issue Details

---

Description
If a GT_SIMD op contains all constants, then mark each constant as DoNotCSE.

Acceptance Criteria

• CI passes

Author:	TIHan
Assignees:	TIHan
Labels:	area-CodeGen-coreclr
Milestone:	-

[TIHan]

@dotnet/jit-contrib This is ready. Just will wait for CI to pass.

The diffs that I got were some improvements, but this is going to help us tweak the double constant heuristic in #65176 since we do not want to hoist or CSE these kinds of constants.

[EgorBo]

There is already a similar logic in that method that skips CSE for IMM-able arguments and also for Vector128/256.Create() (because if we do CSE for any argument of constant Vector128 we likely won't be able to replace it with a data section load)

Can you show some examples? I assume you target new Vector(...) - in this case you can just extend the existing logic

runtime/src/coreclr/jit/morph.cpp

Lines 13923 to 13938 in df3bd55

	case NI_Vector128_Create:
	#if defined(TARGET_XARCH)
	case NI_Vector256_Create:
	#elif defined(TARGET_ARMARCH)
	case NI_Vector64_Create:
	#endif
	{
	bool hwAllArgsAreConst = true;
	for (GenTree** use : multiOp->UseEdges())
	{
	if (!(*use)->OperIsConst())
	{
	hwAllArgsAreConst = false;
	break;
	}
	}

[TIHan]

An example of this is:

new Vector3(1, 0, 0)

I did see the existing logic there. Wanted to keep it separate but makes sense to combine them since they are so similar.

[TIHan]

Diffs

[TIHan]

@dotnet/jit-contrib This is ready again, diffs look pretty good. x64 has more improvements because we always CSE double constants in x64, except for positive zero.

[kunalspathak]

LGTM