[llvm] [X86] Don't rely on global -fp-contract=fast on X86 CodeGen tests (PR #158026)

[Evgenii Kudriashov via llvm-commits]

================
@@ -1007,14 +1007,6 @@ define <4 x float> @fmul_pow_shl_cnt_vec_preserve_fma(<4 x i32> %cnt, <4 x float
 ; CHECK-AVX2-NEXT:    vaddps %xmm1, %xmm0, %xmm0
 ; CHECK-AVX2-NEXT:    retq
 ;
-; CHECK-NO-FASTFMA-LABEL: fmul_pow_shl_cnt_vec_preserve_fma:
-; CHECK-NO-FASTFMA:       # %bb.0:
-; CHECK-NO-FASTFMA-NEXT:    vpslld $23, %xmm0, %xmm0
-; CHECK-NO-FASTFMA-NEXT:    vpbroadcastd {{.*#+}} xmm2 = [1092616192,1092616192,1092616192,1092616192]
-; CHECK-NO-FASTFMA-NEXT:    vpaddd %xmm2, %xmm0, %xmm0
-; CHECK-NO-FASTFMA-NEXT:    vaddps %xmm1, %xmm0, %xmm0
-; CHECK-NO-FASTFMA-NEXT:    retq
-;
----------------
e-kud wrote:

I'm not sure but we may want to duplicate this test w/wo contract. FMA is generated only here and the scenario of  `<no contract-fma>` is lost here.

https://github.com/llvm/llvm-project/pull/158026