[llvm] [X86] Respect denormal mode in f32-to-bf16 conversions (PR #221052)

[Simon Pilgrim via llvm-commits]

================
@@ -1,7 +1,7 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 6
-; RUN: llc < %s -verify-machineinstrs -mtriple=x86_64-unknown-unknown -mattr=+avx10.2 | FileCheck %s --check-prefixes=AVX10_2
-; RUN: llc < %s -verify-machineinstrs -mtriple=x86_64-unknown-unknown -mattr=+avx512bf16,+avx512vl | FileCheck %s --check-prefixes=AVX512BF16
-; RUN: llc < %s -verify-machineinstrs -mtriple=x86_64-unknown-unknown -mattr=+avxneconvert | FileCheck %s --check-prefixes=AVXNECONVERT
+; RUN: llc < %s -verify-machineinstrs -denormal-fp-math=preserve-sign -mtriple=x86_64-unknown-unknown -mattr=+avx10.2 | FileCheck %s --check-prefixes=AVX10_2
+; RUN: llc < %s -verify-machineinstrs -denormal-fp-math=preserve-sign -mtriple=x86_64-unknown-unknown -mattr=+avx512bf16,+avx512vl | FileCheck %s --check-prefixes=AVX512BF16
+; RUN: llc < %s -verify-machineinstrs -denormal-fp-math=preserve-sign -mtriple=x86_64-unknown-unknown -mattr=+avxneconvert | FileCheck %s --check-prefixes=AVXNECONVERT
----------------
RKSimon wrote:

retain test coverage without the tag as well

https://github.com/llvm/llvm-project/pull/221052