[llvm] [NVPTX] fold movs into loads and stores (PR #144581)

[Princeton Ferro via llvm-commits]

================
@@ -138,9 +138,9 @@ define ptx_kernel void @foo13(ptr noalias readonly %from, ptr %to) {
 }
 
 ; SM20-LABEL: .visible .entry foo14(
-; SM20: ld.global.v4.b16
+; SM20: ld.global.v2.b32
 ; SM35-LABEL: .visible .entry foo14(
-; SM35: ld.global.nc.v4.b16
+; SM35: ld.global.nc.v2.b32
----------------
Prince781 wrote:

This is happening because of the changes in `getVectorLoweringShape()` to prefer packing all types maximally first, then allowing these new combiner rules to split them later as needed.

https://github.com/llvm/llvm-project/pull/144581