1ccc49924a
D79164/2596da31740f changed getCFInstrCost to return 1 per default.
AArch64 did not have its own implementation, hence the throughput cost
of CFI instructions is overestimated. On most cores, most branches should
be predicated and essentially free throughput wise.
This restores a 9% performance regression on a SPEC2006 benchmark on
AArch64 with -O3 LTO & PGO.
This patch effectively restores pre
|
||
|---|---|---|
| .. | ||
| aarch64-predication.ll | ||
| aarch64-unroll.ll | ||
| arbitrary-induction-step.ll | ||
| arm64-unroll.ll | ||
| backedge-overflow.ll | ||
| deterministic-type-shrinkage.ll | ||
| extractvalue-no-scalarization-required.ll | ||
| gather-cost.ll | ||
| induction-trunc.ll | ||
| interleaved-vs-scalar.ll | ||
| interleaved_cost.ll | ||
| lit.local.cfg | ||
| loop-vectorization-factors.ll | ||
| max-vf-for-interleaved.ll | ||
| no_vector_instructions.ll | ||
| nontemporal-load-store.ll | ||
| outer_loop_test1_no_explicit_vect_width.ll | ||
| pr31900.ll | ||
| pr33053.ll | ||
| pr36032.ll | ||
| predication_costs.ll | ||
| reduction-small-size.ll | ||
| sdiv-pow2.ll | ||
| smallest-and-widest-types.ll | ||
| type-shrinkage-insertelt.ll | ||