forked from OSchip/llvm-project
b7a11274f9
In InnerLoopVectorizer::widenPHIInstruction there are cases where we have to scalarise a pointer induction variable after vectorisation. For scalable vectors we already deal with the case where the pointer induction variable is uniform, but we currently crash if not uniform. For fixed width vectors we calculate every lane of the scalarised pointer induction variable for a given VF, however this cannot work for scalable vectors. In this case I have added support for caching the whole vector value for each unrolled part so that we can always extract an arbitrary element. Additionally, we still continue to cache the known minimum number of lanes too in order to improve code quality by avoiding an extractelement operation. I have adapted an existing test `pointer_iv_mixed` from the file: Transforms/LoopVectorize/consecutive-ptr-uniforms.ll and added it here for scalable vectors instead: Transforms/LoopVectorize/AArch64/sve-widen-phi.ll Differential Revision: https://reviews.llvm.org/D101294 |
||
|---|---|---|
| .. | ||
| Oz-and-forced-vectorize.ll | ||
| aarch64-predication.ll | ||
| aarch64-unroll.ll | ||
| arbitrary-induction-step.ll | ||
| arm64-unroll.ll | ||
| backedge-overflow.ll | ||
| deterministic-type-shrinkage.ll | ||
| extend-vectorization-factor-for-unprofitable-memops.ll | ||
| extractvalue-no-scalarization-required.ll | ||
| first-order-recurrence.ll | ||
| gather-cost.ll | ||
| induction-trunc.ll | ||
| interleaved-vs-scalar.ll | ||
| interleaved_cost.ll | ||
| intrinsiccost.ll | ||
| lit.local.cfg | ||
| loop-vectorization-factors.ll | ||
| loopvectorize_pr33804_double.ll | ||
| masked-op-cost.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 | ||
| pr46950-load-cast-context-crash.ll | ||
| predication_costs.ll | ||
| reduction-small-size.ll | ||
| runtime-check-size-based-threshold.ll | ||
| scalable-call.ll | ||
| scalable-reductions.ll | ||
| scalable-strict-fadd.ll | ||
| scalable-vf-analysis.ll | ||
| scalable-vf-hint.ll | ||
| scalarize-store-with-predication.ll | ||
| sdiv-pow2.ll | ||
| select-costs.ll | ||
| smallest-and-widest-types.ll | ||
| strict-fadd.ll | ||
| sve-basic-vec.ll | ||
| sve-cond-inv-loads.ll | ||
| sve-extract-last-veclane.ll | ||
| sve-gather-scatter.ll | ||
| sve-inductions.ll | ||
| sve-inv-loads.ll | ||
| sve-large-strides.ll | ||
| sve-masked-loadstore.ll | ||
| sve-scalable-load-in-loop.ll | ||
| sve-type-conv.ll | ||
| sve-vector-reverse-mask4.ll | ||
| sve-vector-reverse.ll | ||
| sve-widen-phi.ll | ||
| type-shrinkage-insertelt.ll | ||
| unsafe-vf-hint-remark.ll | ||
| veclib-calls-libsystem-darwin.ll | ||
| vector-reverse-mask4.ll | ||
| vector-reverse.ll | ||