[llvm] [WIP][X86] lowerBuildVectorAsBroadcast - don't convert constant vectors to broadcasts on AVX512VL targets (PR #73509)

[Phoebe Wang via llvm-commits]

================
@@ -8,7 +8,7 @@ define <4 x float> @test_int_x86_avx512fp8_mask_cfmadd_ph_bst(<4 x float> %x0, <
 ; CHECK-LABEL: test_int_x86_avx512fp8_mask_cfmadd_ph_bst:
 ; CHECK:       ## %bb.0:
 ; CHECK-NEXT:    kmovd %edi, %k1
-; CHECK-NEXT:    vfmaddcph {{\.?LCPI[0-9]+_[0-9]+}}(%rip){1to4}, %xmm0, %xmm1 {%k1}
+; CHECK-NEXT:    vfmaddcph {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm1 {%k1}
----------------
phoebewang wrote:

Regression? Seem all FP16 broadcast affected?

https://github.com/llvm/llvm-project/pull/73509