Giuseppe Rossini e94a0b300a

[AMDGPU] Fix vector legalization for bf16 valu ops (#158439) ...

Add v4,v8,v16,v32 legalizations for the following operations:
- `FADD`
- `FMUL`
- `FMA`
- `FCANONICALIZE`

2025-09-25 10:07:11 +01:00

..

abs.ll

…

add-sub.ll

…

addrspacecast.ll

…

aggregates.ll

…

arith-sminmax.ll

…

arith-ssat.ll

…

arith-uminmax.ll

…

arith-usat.ll

…

arithmetic_fence.ll

…

bit-ops.ll

…

canonicalize.ll

[AMDGPU] Fix vector legalization for bf16 valu ops (#158439)

2025-09-25 10:07:11 +01:00

cast.ll

…

control-flow.ll

…

copysign.ll

AMDGPU: Reduce cost of f64 copysign (#141944)

2025-06-18 08:10:53 +09:00

div.ll

…

exp2.ll

…

exp10.ll

…

exp.ll

…

extractelement.ll

…

fabs.ll

…

fadd.ll

[AMDGPU] Fix vector legalization for bf16 valu ops (#158439)

2025-09-25 10:07:11 +01:00

fdiv.ll

…

fma.ll

[AMDGPU] Fix vector legalization for bf16 valu ops (#158439)

2025-09-25 10:07:11 +01:00

fmul.ll

[AMDGPU] Fix vector legalization for bf16 valu ops (#158439)

2025-09-25 10:07:11 +01:00

fmuladd.ll

…

fneg.ll

…

fptosi.ll

…

fptoui.ll

…

frexp.ll

…

fround.ll

…

fsub.ll

[AMDGPU] Fix vector legalization for bf16 valu ops (#158439)

2025-09-25 10:07:11 +01:00

fused_costs.ll

…

gep.ll

…

insertelement.ll

…

is_fpclass.ll

…

ldexp.ll

[BasicTTIImpl] Add cost entries for ldexp, [l]lround (#146373)

2025-07-07 09:26:30 +01:00

lit.local.cfg

…

load-to-trunc.ll

…

load.ll

[AMDGPU] Enable vectorization of i8 values. (#134934)

2025-06-26 19:15:31 -04:00

log2.ll

…

log10.ll

…

log.ll

…

logicalop.ll

…

maximum.ll

…

maximumnum.ll

AMDGPU: Cost model for minimumnum/maximumnum (#141946)

2025-06-18 08:19:06 +09:00

maxnum.ll

…

minimum.ll

…

minimumnum.ll

AMDGPU: Cost model for minimumnum/maximumnum (#141946)

2025-06-18 08:19:06 +09:00

minnum.ll

…

mul.ll

…

ptrmask.ll

…

reduce-and.ll

[CostModel] Handle all cost kinds in getCmpSelInstrCost (#148233)

2025-07-15 18:08:52 +01:00

reduce-or.ll

[CostModel] Handle all cost kinds in getCmpSelInstrCost (#148233)

2025-07-15 18:08:52 +01:00

rem.ll

…

shifts.ll

…

shufflevector.ll

[AMDGPU] Enable vectorization of i8 values. (#134934)

2025-06-26 19:15:31 -04:00

special-argument-intrinsics.ll

AMDGPU: Report special input intrinsics as free (#141948)

2025-06-18 08:24:58 +09:00

sqrt.ll

…

store.ll

[AMDGPU] Enable vectorization of i8 values. (#134934)

2025-06-26 19:15:31 -04:00