[llvm] [AMDGPU] Add getCmpSelInstrCost override to re-enable SimplifyCFG speculation for vector types (PR #208043)

[Matt Arsenault via llvm-commits]

================
@@ -0,0 +1,193 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 6
+
+; RUN: opt -S -passes=simplifycfg -mtriple=amdgpu9.0a-amd-amdhsa < %s | FileCheck %s --check-prefix=IR
+; RUN: opt -S -passes=simplifycfg -mtriple=amdgpu9.0a-amd-amdhsa < %s \
----------------
arsenm wrote:

Avoid using -O3, and you don't need the triple on llc after opt sets it 

https://github.com/llvm/llvm-project/pull/208043