[PATCH] D90554: [CostModel] remove cost-kind predicate for intrinsics in basic TTI implementation

[Sanjay Patel via Phabricator via llvm-commits]

spatel added inline comments.


================
Comment at: llvm/test/Analysis/CostModel/X86/fmaxnum-size-latency.ll:18
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4F32 = call <4 x float> @llvm.maxnum.v4f32(<4 x float> undef, <4 x float> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V8F32 = call <8 x float> @llvm.maxnum.v8f32(<8 x float> undef, <8 x float> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16F32 = call <16 x float> @llvm.maxnum.v16f32(<16 x float> undef, <16 x float> undef)
----------------
RKSimon wrote:
> Not sure why ymm cost is higher than scalar/xmm - on AVX2 they should be the same (same for f64)
Looks like the entries for FMAXNUM are missing in the `AVX2CostTbl`.
I'll draft a fix.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D90554/new/

https://reviews.llvm.org/D90554