[ARM] MVE VMLAS
authorDavid Green <david.green@arm.com>
Fri, 28 Feb 2020 11:02:20 +0000 (11:02 +0000)
committerDavid Green <david.green@arm.com>
Fri, 28 Feb 2020 14:27:21 +0000 (14:27 +0000)
commite2a2f3f7fcf2175b16a96c47e5d8cb566258d1ae
treec63d8f02bb785ed25969cd57fcd09df2320b559a
parent78e5d1346fb235e8901bb3821e21cd851d04e3a9
[ARM] MVE VMLAS

This addes extra patterns for the VMLAS MVE instruction, which performs
Qda = Qda * Qn + Rm, a similar pattern to the existing VMLA. The sinking
of splat(Rm) into the loop is already performed, meaning we just need
extra Pat's in tablegen.

Differential Revision: https://reviews.llvm.org/D75115
llvm/lib/Target/ARM/ARMInstrMVE.td
llvm/test/CodeGen/Thumb2/LowOverheadLoops/mve-tail-data-types.ll
llvm/test/CodeGen/Thumb2/mve-vmla.ll