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

[Matt Arsenault via llvm-commits]

================
@@ -0,0 +1,110 @@
+; RUN: llc < %s -mtriple=x86_64-linux-gnu -mattr=+avx512bf16,+avx512vl -verify-machineinstrs | FileCheck %s --check-prefixes=CHECK,AVX512BF16
+; RUN: llc < %s -mtriple=x86_64-linux-gnu -mattr=+avxneconvert -verify-machineinstrs | FileCheck %s --check-prefixes=CHECK,AVXNECONVERT
----------------
arsenm wrote:

```suggestion
; RUN: llc < %s -mtriple=x86_64-linux-gnu -mattr=+avx512bf16,+avx512vl | FileCheck %s --check-prefixes=CHECK,AVX512BF16
; RUN: llc < %s -mtriple=x86_64-linux-gnu -mattr=+avxneconvert  | FileCheck %s --check-prefixes=CHECK,AVXNECONVERT
```

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