llvm/test/Analysis/CostModel/AMDGPU at small-data - llvm

mirror of https://github.com/AxioDL/llvm.git synced 2026-03-30 11:42:29 -07:00

Files

T

History

Matt Arsenault bb366db643 AMDGPU: Cost model for basic integer operations

This resolves bug 21148 by preventing promotion to
i64 induction variables.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@264376 91177308-0d34-0410-b5e6-96231b3b80d8

2016-03-25 01:16:40 +00:00

add-sub.ll

AMDGPU: Cost model for basic integer operations

2016-03-25 01:16:40 +00:00

addrspacecast.ll

TTI: Report 0 cost for free addrspacecasts

2016-03-25 00:26:29 +00:00

bit-ops.ll

AMDGPU: Cost model for basic integer operations

2016-03-25 01:16:40 +00:00

br.ll

…

extractelement.ll

…

fabs.ll

TTI: Use 0 for cost of fabs if free

2016-03-25 00:26:22 +00:00

fadd.ll

AMDGPU: Partially implement getArithmeticInstrCost for FP ops

2016-03-25 01:00:32 +00:00

fdiv.ll

AMDGPU: Partially implement getArithmeticInstrCost for FP ops

2016-03-25 01:00:32 +00:00

fmul.ll

AMDGPU: Partially implement getArithmeticInstrCost for FP ops

2016-03-25 01:00:32 +00:00

fsub.ll

AMDGPU: Partially implement getArithmeticInstrCost for FP ops

2016-03-25 01:00:32 +00:00

insertelement.ll

AMDGPU: TTI: Make insertelement free.

2016-03-25 00:14:11 +00:00

lit.local.cfg

…

mul.ll

AMDGPU: Cost model for basic integer operations

2016-03-25 01:16:40 +00:00

shifts.ll

AMDGPU: Cost model for basic integer operations

2016-03-25 01:16:40 +00:00