[ARM] Parallel DSP Pass
Armv6 introduced instructions to perform 32-bit SIMD operations. The purpose of
this pass is to do some straightforward IR pattern matching to create ACLE DSP
intrinsics, which map on these 32-bit SIMD operations.
Currently, only the SMLAD instruction gets recognised. This instruction
performs two multiplications with 16-bit operands, and stores the result in an
accumulator. We will follow this up with patches to recognise SMLAD in more
cases, and also to generate other DSP instructions (like e.g. SADD16).
Patch by: Sam Parker and Sjoerd Meijer
Differential Revision: https://reviews.llvm.org/D48128
llvm-svn: 335850
diff --git a/llvm/lib/Target/ARM/ARM.h b/llvm/lib/Target/ARM/ARM.h
index 9ffb4c2..644dcbe 100644
--- a/llvm/lib/Target/ARM/ARM.h
+++ b/llvm/lib/Target/ARM/ARM.h
@@ -15,6 +15,7 @@
#ifndef LLVM_LIB_TARGET_ARM_ARM_H
#define LLVM_LIB_TARGET_ARM_ARM_H
+#include "llvm/IR/LegacyPassManager.h"
#include "llvm/Support/CodeGen.h"
#include <functional>
#include <vector>
@@ -35,6 +36,8 @@
class MCInst;
class PassRegistry;
+
+Pass *createARMParallelDSPPass();
FunctionPass *createARMISelDag(ARMBaseTargetMachine &TM,
CodeGenOpt::Level OptLevel);
FunctionPass *createA15SDOptimizerPass();
@@ -57,6 +60,8 @@
BasicBlockInfo &BBI);
std::vector<BasicBlockInfo> computeAllBlockSizes(MachineFunction *MF);
+
+void initializeARMParallelDSPPass(PassRegistry &);
void initializeARMLoadStoreOptPass(PassRegistry &);
void initializeARMPreAllocLoadStoreOptPass(PassRegistry &);
void initializeARMConstantIslandsPass(PassRegistry &);