[llvm] [X86] Lower `minimum`/`maximum`/`minimumnum`/`maximumnum` using bitwise operations (PR #170069)

[via llvm-commits]

================
@@ -3,13 +3,16 @@
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx      | FileCheck %s --check-prefixes=AVX,AVX1
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f  | FileCheck %s --check-prefixes=AVX,AVX512,AVX512F
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512dq | FileCheck %s --check-prefixes=AVX,AVX512,AVX512DQ
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512bf16,+avx512vl | FileCheck %s --check-prefixes=AVX,AVX512,AVX512BF16
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx10.2  | FileCheck %s --check-prefixes=AVX10_2
 ; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+avx        | FileCheck %s --check-prefixes=X86
 
 declare float @llvm.maximumnum.f32(float, float)
 declare double @llvm.maximumnum.f64(double, double)
+declare bfloat @llvm.maximumnum.bf16(bfloat, bfloat)
 declare float @llvm.minimumnum.f32(float, float)
 declare double @llvm.minimumnum.f64(double, double)
+declare bfloat @llvm.minimumnum.bf16(bfloat, bfloat)
----------------
valadaptive wrote:

I never know what to do with these. I know declarations aren't required in regression tests anymore, but it feels wrong to leave them out for only new functions, or to remove them all entirely.

https://github.com/llvm/llvm-project/pull/170069