llvm-project

History

[AArch64] Enable RT and partial unrolling with reductions for Apple CPUs. (#149699 )

Update unrolling preferences for Apple Silicon CPUs to enable partial
unrolling and runtime unrolling for small loops with reductions.

This builds on top of unroller changes to introduce parallel reduction
phis, if possible: https://github.com/llvm/llvm-project/pull/149470.

PR: https://github.com/llvm/llvm-project/pull/149699

2025-09-09 13:23:30 +00:00

apple-unrolling-multi-exit.ll

[AArch64] Enable unrolling for small multi-exit loops (#131998 )

2025-04-09 10:34:27 +01:00

apple-unrolling.ll

[AArch64] Enable RT and partial unrolling with reductions for Apple CPUs. (#149699 )

2025-09-09 13:23:30 +00:00

falkor-prefetch.ll

…

full-unroll-trip-count-upper-bound.ll

…

large-vector-ops.ll

…

lit.local.cfg

…

partial.ll

…

runtime-loop.ll

…

runtime-unroll-generic.ll

…

scalable-vec-ins-ext.ll

…

thresholdO3-cost-model.ll

…

unroll-optsize.ll

…

unroll-upperbound.ll

…

unrolling-multi-exit.ll

[AArch64] Enable unrolling for small multi-exit loops (#131998 )

2025-04-09 10:34:27 +01:00

vector.ll

[AArch64] Allow unrolling of scalar epilogue loops (#151164 )

2025-07-31 11:03:41 +01:00