HomePhabricator

[ARM][DAGCOMBINE][FIX] PerformVMOVRRDCombine

Description

[ARM][DAGCOMBINE][FIX] PerformVMOVRRDCombine

Summary:
PerformVMOVRRDCombine ommits adding a offset
of 4 to the PointerInfo, when converting a
f64 = load[M]
to
{i32, i32} = {load[M], load[M + 4]}

Which would allow the machine scheduller
to break dependencies with the second load.

  • pr42638

Reviewers: eli.friedman, dmgreen, ostannard

Reviewed By: ostannard

Subscribers: ostannard, javed.absar, kristof.beyls, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D64870

Details

Committed
dnsampaioJul 18 2019, 3:05 AM
Reviewer
ostannard
Differential Revision
D64870: [ARM][DAGCOMBINE][FIX] PerformVMOVRRDCombine
Parents
rL366422: [OpenCL] Improve destructor support in C++ for OpenCL
Branches
Unknown
Tags
Unknown