diff options
| author | Weiming Zhao <weimingz@codeaurora.org> | 2014-04-30 21:07:24 +0000 |
|---|---|---|
| committer | Weiming Zhao <weimingz@codeaurora.org> | 2014-04-30 21:07:24 +0000 |
| commit | 7f6daf1799a0dc93e37edc4d83ba4ef7c184e3a6 (patch) | |
| tree | db7cc9066440bf1a9e4106da6c979e0ecdae8272 /llvm/lib/Target/ARM64 | |
| parent | dd2647edcf99a9e7b1af9b407ff1489b4ee45739 (diff) | |
| download | bcm5719-llvm-7f6daf1799a0dc93e37edc4d83ba4ef7c184e3a6.tar.gz bcm5719-llvm-7f6daf1799a0dc93e37edc4d83ba4ef7c184e3a6.zip | |
[ARM64] Prevent bit extraction to be adjusted by following shift
For pattern like ((x >> C1) & Mask) << C2, DAG combiner may convert it
into (x >> (C1-C2)) & (Mask << C2), which makes pattern matching of ubfx
more difficult.
For example:
Given
%shr = lshr i64 %x, 4
%and = and i64 %shr, 15
%arrayidx = getelementptr inbounds [8 x [64 x i64]]* @arr, i64 0, %i64 2, i64 %and
%0 = load i64* %arrayidx
With current shift folding, it takes 3 instrs to compute base address:
lsr x8, x0, #1
and x8, x8, #0x78
add x8, x9, x8
If using ubfx, it only needs 2 instrs:
ubfx x8, x0, #4, #4
add x8, x9, x8, lsl #3
This fixes bug 19589
llvm-svn: 207702
Diffstat (limited to 'llvm/lib/Target/ARM64')
| -rw-r--r-- | llvm/lib/Target/ARM64/ARM64ISelLowering.cpp | 15 | ||||
| -rw-r--r-- | llvm/lib/Target/ARM64/ARM64ISelLowering.h | 3 |
2 files changed, 18 insertions, 0 deletions
diff --git a/llvm/lib/Target/ARM64/ARM64ISelLowering.cpp b/llvm/lib/Target/ARM64/ARM64ISelLowering.cpp index ccd80175faf..6dd588c3705 100644 --- a/llvm/lib/Target/ARM64/ARM64ISelLowering.cpp +++ b/llvm/lib/Target/ARM64/ARM64ISelLowering.cpp @@ -5931,6 +5931,21 @@ ARM64TargetLowering::getScratchRegisters(CallingConv::ID) const { return ScratchRegs; } +bool ARM64TargetLowering::isDesirableToCommuteWithShift(const SDNode *N) const { + EVT VT = N->getValueType(0); + // If N is unsigned bit extraction: ((x >> C) & mask), then do not combine + // it with shift to let it be lowered to UBFX. + if (N->getOpcode() == ISD::AND && (VT == MVT::i32 || VT == MVT::i64) && + isa<ConstantSDNode>(N->getOperand(1))) { + uint64_t TruncMask = N->getConstantOperandVal(1); + if (isMask_64(TruncMask) && + N->getOperand(0).getOpcode() == ISD::SRL && + isa<ConstantSDNode>(N->getOperand(0)->getOperand(1))) + return false; + } + return true; +} + bool ARM64TargetLowering::shouldConvertConstantLoadToIntImm(const APInt &Imm, Type *Ty) const { assert(Ty->isIntegerTy()); diff --git a/llvm/lib/Target/ARM64/ARM64ISelLowering.h b/llvm/lib/Target/ARM64/ARM64ISelLowering.h index ddaf347bf1a..96a32ea85c0 100644 --- a/llvm/lib/Target/ARM64/ARM64ISelLowering.h +++ b/llvm/lib/Target/ARM64/ARM64ISelLowering.h @@ -284,6 +284,9 @@ public: const MCPhysReg *getScratchRegisters(CallingConv::ID CC) const override; + /// \brief Returns false if N is a bit extraction pattern of (X >> C) & Mask. + bool isDesirableToCommuteWithShift(const SDNode *N) const override; + /// \brief Returns true if it is beneficial to convert a load of a constant /// to just the constant itself. bool shouldConvertConstantLoadToIntImm(const APInt &Imm, |

