diff options
author | Kerry McLaughlin <kerry.mclaughlin@arm.com> | 2021-04-27 14:50:27 +0100 |
---|---|---|
committer | Kerry McLaughlin <kerry.mclaughlin@arm.com> | 2021-04-28 11:29:01 +0100 |
commit | 9cc217ab36eb003dee8ace8359e6f535ac059cfd (patch) | |
tree | 6e52932c2b38eda261a8821107de296f91459d14 /llvm/utils/UpdateTestChecks/common.py | |
parent | 858a9583e1fed24aa57e5f2769f4117883264ceb (diff) | |
download | llvm-9cc217ab36eb003dee8ace8359e6f535ac059cfd.zip llvm-9cc217ab36eb003dee8ace8359e6f535ac059cfd.tar.gz llvm-9cc217ab36eb003dee8ace8359e6f535ac059cfd.tar.bz2 |
[LoopVectorize] Prevent multiple Phis being generated with in-order reductions
When using the -enable-strict-reductions flag where UF>1 we generate multiple
Phi nodes, though only one of these is used as an input to the vector.reduce.fadd
intrinsics. The unused Phi nodes are removed later by instcombine.
This patch changes widenPHIInstruction/fixReduction to only generate
one Phi, and adds an additional test for unrolling to strict-fadd.ll
Reviewed By: david-arm
Differential Revision: https://reviews.llvm.org/D100570
Diffstat (limited to 'llvm/utils/UpdateTestChecks/common.py')
0 files changed, 0 insertions, 0 deletions