[llvm] [AMDGPU] Hints for prefering allocating AGPR registers for buffer loads (PR #215697)

[Matt Arsenault via llvm-commits]

================
@@ -0,0 +1,64 @@
+# RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx942 \
----------------
arsenm wrote:

Use new subarch triples 

https://github.com/llvm/llvm-project/pull/215697