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

[Phoebe Wang via llvm-commits]

================
@@ -208,12 +198,12 @@ define <32 x half> @test_fminimum_v32f16_szero(<32 x half> %x, <32 x half> %y) "
 define <32 x half> @test_fmaximum_v32f16_nans_szero(<32 x half> %x, <32 x half> %y) {
 ; CHECK-LABEL: test_fmaximum_v32f16_nans_szero:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    vpmovw2m %zmm0, %k1
-; CHECK-NEXT:    vpblendmw %zmm1, %zmm0, %zmm2 {%k1}
+; CHECK-NEXT:    vmaxph %zmm1, %zmm0, %zmm2
+; CHECK-NEXT:    vpbroadcastw {{.*#+}} zmm1 = [NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN]
+; CHECK-NEXT:    vpternlogq {{.*#+}} zmm1 = zmm2 & (zmm1 | zmm0)
----------------
phoebewang wrote:

The old code seems better since no memory load there.

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