[llvm] [AMDGPU][DAG] Enable ganging up of memcpy loads/stores for AMDGPU (PR #96185)

[Matt Arsenault via llvm-commits]

================
@@ -0,0 +1,138 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
+
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1030 %s -o - | FileCheck %s
+
+; Testing codegen for memcpy and memmove with scalar reads.
+
+
+define void @memcpy_p1_p4_sz16_align_4_4(ptr addrspace(1) align 4  %dst, ptr addrspace(4) align 4 readonly inreg %src) {
----------------
arsenm wrote:

Extra space between align 4 and %st throughout 

https://github.com/llvm/llvm-project/pull/96185