[llvm] [AArch64][SVE] Enable factor-3 scalable interleave operations (PR #200424)

[David Sherwood via llvm-commits]

================
@@ -45,6 +45,17 @@ define { <vscale x 2 x double>, <vscale x 2 x double> } @foo_ld2_nxv2f64(<vscale
   ret { <vscale x 2 x double>, <vscale x 2 x double> } %deinterleaved.vec
 }
 
+define { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @foo_ld3_nxv16i8(<vscale x 16 x i1> %mask, ptr %p) {
----------------
david-arm wrote:

It would be good to have variants for other element sizes too, similar to the ld4 case below.

https://github.com/llvm/llvm-project/pull/200424