[PATCH] D53257: [ARM][NEON] Improve vector popcnt lowering with PADDL (PR39281)

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon created this revision.
RKSimon added reviewers: efriedma, t.p.northover, samparker, spatel.
Herald added a reviewer: javed.absar.
Herald added subscribers: chrib, kristof.beyls.

As I suggested on PR39281, this patch uses PADDL pairwise addition to widen from the vXi8 CTPOP result to the target vector type.

This is a blocker for generic vector CTPOP expansion (P32655) - ARM's vXi64 CTPOP currently expands, which would generate a vXi64 MUL but ARM's lowering expands the general MUL case and vectors aren't well handled in LegalizeDAG - improving the CTPOP lowering was a lot easier than fixing the MUL lowering......


Repository:
  rL LLVM

https://reviews.llvm.org/D53257

Files:
  lib/Target/ARM/ARMISelLowering.cpp
  test/CodeGen/ARM/popcnt.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D53257.169603.patch
Type: text/x-patch
Size: 14238 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20181014/4aa5a374/attachment.bin>