for all the processors where I have tried it, and even when it might not help
performance, the cost is quite low. The opportunities for duplicating
indirect branches are limited by other factors so code size does not change
much due to tail duplicating indirect branches aggressively.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@90144
91177308-0d34-0410-b5e6-
96231b3b80d8
/// length.
virtual unsigned getInlineAsmLength(const char *Str,
const MCAsmInfo &MAI) const;
-
- /// isProfitableToDuplicateIndirectBranch - Returns true if tail duplication
- /// is especially profitable for indirect branches.
- virtual bool isProfitableToDuplicateIndirectBranch() const { return false; }
};
/// TargetInstrInfoImpl - This is the default implementation of
unsigned MaxDuplicateCount;
if (MF.getFunction()->hasFnAttr(Attribute::OptimizeForSize))
MaxDuplicateCount = 1;
- else if (TII->isProfitableToDuplicateIndirectBranch() &&
- !TailBB->empty() && TailBB->back().getDesc().isIndirectBranch())
+ else if (!TailBB->empty() && TailBB->back().getDesc().isIndirectBranch())
// If the target has hardware branch prediction that can handle indirect
// branches, duplicating them can often make them predictable when there
// are common paths through the code. The limit needs to be high enough
return TargetInstrInfoImpl::isIdentical(MI0, MI1, MRI);
}
-bool ARMBaseInstrInfo::isProfitableToDuplicateIndirectBranch() const {
- // If the target processor can predict indirect branches, it is highly
- // desirable to duplicate them, since it can often make them predictable.
- return getSubtarget().hasBranchTargetBuffer();
-}
-
/// getInstrPredicate - If instruction is predicated, returns its predicate
/// condition, otherwise returns AL. It also returns the condition code
/// register by reference.
virtual bool isIdentical(const MachineInstr *MI, const MachineInstr *Other,
const MachineRegisterInfo *MRI) const;
-
- virtual bool isProfitableToDuplicateIndirectBranch() const;
};
static inline
if (UseNEONFP.getPosition() == 0)
UseNEONForSinglePrecisionFP = true;
}
- HasBranchTargetBuffer = (CPUString == "cortex-a8" ||
- CPUString == "cortex-a9");
}
/// GVIsIndirectSymbol - true if the GV will be accessed via an indirect symbol.
/// determine if NEON should actually be used.
bool UseNEONForSinglePrecisionFP;
- /// HasBranchTargetBuffer - True if processor can predict indirect branches.
- bool HasBranchTargetBuffer;
-
/// IsThumb - True if we are in thumb mode, false if in ARM mode.
bool IsThumb;
bool isThumb2() const { return IsThumb && (ThumbMode == Thumb2); }
bool hasThumb2() const { return ThumbMode >= Thumb2; }
- bool hasBranchTargetBuffer() const { return HasBranchTargetBuffer; }
-
bool isR9Reserved() const { return IsR9Reserved; }
bool useMovt() const { return UseMovt && hasV6T2Ops(); }
/// instruction may be. This returns the maximum number of bytes.
///
virtual unsigned GetInstSizeInBytes(const MachineInstr *MI) const;
-
- virtual bool isProfitableToDuplicateIndirectBranch() const { return true; }
};
}
///
unsigned getGlobalBaseReg(MachineFunction *MF) const;
- virtual bool isProfitableToDuplicateIndirectBranch() const { return true; }
-
private:
MachineInstr* foldMemoryOperandImpl(MachineFunction &MF,
MachineInstr* MI,