[llvm] [AMDGPU][SDAG] Legalise v2i32 or/xor/and instructions to make use of 64-bit wide instructions (PR #140694)

[Chris Jackson via llvm-commits]

================
@@ -1,8 +1,13 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 3
 ; RUN:  llc -amdgpu-scalarize-global-loads=false  -mtriple=amdgcn -mcpu=verde -verify-machineinstrs < %s | FileCheck -check-prefix=GFX6 %s
 ; RUN:  llc -amdgpu-scalarize-global-loads=false  -mtriple=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=GFX8 %s
+; RUN:  llc -amdgpu-scalarize-global-loads=true  -mtriple=amdgcn -mcpu=verde -verify-machineinstrs < %s | FileCheck -check-prefix=GFX6S %s
+; RUN:  llc -amdgpu-scalarize-global-loads=true  -mtriple=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=GFX8S %s
 ; RUN:  llc -amdgpu-scalarize-global-loads=false  -mtriple=r600 -mcpu=redwood < %s | FileCheck -check-prefix=EG %s
 
+;; Added tests with "-amdgpu-scalarize-global-loads=true" to allow the generation of s_or_b64, particularly in the v2i32 case. See SWDEV-517886.
----------------
chrisjbris wrote:

All tests added. r600 separated out to different files as seems to be convention / there is strong precedence looking at the test/CodeGen/AMDGPU/ folder.

https://github.com/llvm/llvm-project/pull/140694