[llvm] [X86] Truncates i64 add arithmetic to i32 when known zeroes in upper 33 bits (PR #191619)

[Takashi Idobe via llvm-commits]

================
@@ -0,0 +1,178 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 6
+; RUN: llc < %s -mtriple=i686-unknown-unknown | FileCheck %s --check-prefix=X86
+; RUN: llc < %s -mtriple=x86_64-linux | FileCheck %s --check-prefix=X64-LINUX
+; RUN: llc < %s -mtriple=x86_64-win32 | FileCheck %s --check-prefix=X64-WIN32
+
----------------
Takashiidobe wrote:

Since this code doesn't explicitly handle linux/windows separately, these two lines can be:
; RUN: llc < %s -mtriple=x86_64-unknown-unknown | FileCheck %s --check-prefix=X64

Taking a look at the check labels, all that changes are the registers between the windows and linux checks which aren't very notable. 

https://github.com/llvm/llvm-project/pull/191619