llvm-project

History

dfukalov 76a0c0ee6f [AMDGPU][CostModel] Improve cost estimation for fused {fadd|fsub}(a,fmul(b,c))

Summary:
If result of fmul(b,c) has one use, in almost all cases (except denormals are
IEEE) the pair of operations will be fused in one fma/mad/mac/etc.

Reviewers: rampitec

Reviewed By: rampitec

Subscribers: arsenm, kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, hiraditya, llvm-commits, kerbowa

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D83919

2020-07-16 03:06:38 +03:00

add-sub.ll

[AMDGPU] Tweak getTypeLegalizationCost()

2020-07-06 14:07:48 -07:00

addrspacecast.ll

[AMDGPU] Improve code size cost model (part 2)

2019-11-06 13:55:48 +03:00

bit-ops.ll

[AMDGPU] Fixed cost model for packed 16 bit ops

2019-12-17 15:14:17 -08:00

br.ll

…