[llvm] r340016 - [PowerPC] Generate Power9 extswsli extend sign and shift immediate instruction

Eric Christopher via llvm-commits llvm-commits at lists.llvm.org
Wed Aug 22 09:32:48 PDT 2018


Ali is working on it and we'll see if we can get it to you today.

-eric

On Wed, Aug 22, 2018 at 5:28 AM Nemanja Ivanovic <nemanja.i.ibm at gmail.com>
wrote:

> Thanks for reverting Eric,
>
> I look forward to a test case so we can get this sorted out.
> Nemanja
>
> On Tue, Aug 21, 2018 at 7:03 PM Eric Christopher <echristo at gmail.com>
> wrote:
>
>> I went ahead and reverted this here:
>>
>> Author: Eric Christopher <echristo at gmail.com>
>> Date:   Tue Aug 21 18:35:08 2018 +0000
>>
>>     Temporarily Revert "[PowerPC] Generate Power9 extswsli extend sign
>> and shift immediate instruction" due to it causing a compiler crash on
>> valid.
>>
>>     This reverts commit r340016, testcase forthcoming.
>>
>>     git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@340315
>> 91177308-0d34-0410-b5e6-96231b3b80d8
>>
>> after talking with Nemanja.
>>
>> -eric
>>
>> On Tue, Aug 21, 2018 at 11:11 AM Eric Christopher <echristo at gmail.com>
>> wrote:
>>
>>> Hi Nemanja,
>>>
>>> We're seeing a compiler crash on valid with this patch - while we're
>>> working up a testcase would you mind reverting?
>>>
>>> -eric
>>>
>>> On Fri, Aug 17, 2018 at 5:36 AM Nemanja Ivanovic via llvm-commits <
>>> llvm-commits at lists.llvm.org> wrote:
>>>
>>>> Author: nemanjai
>>>> Date: Fri Aug 17 05:35:44 2018
>>>> New Revision: 340016
>>>>
>>>> URL: http://llvm.org/viewvc/llvm-project?rev=340016&view=rev
>>>> Log:
>>>> [PowerPC] Generate Power9 extswsli extend sign and shift immediate
>>>> instruction
>>>>
>>>> Add a DAG combine for the PowerPC code generator to generate the Power9
>>>> extswsli
>>>> extend sign and shift immediate instruction.
>>>>
>>>> Patch by RolandF.
>>>>
>>>> Differential revision: https://reviews.llvm.org/D49879
>>>>
>>>> Added:
>>>>     llvm/trunk/llvm/
>>>>     llvm/trunk/llvm/test/
>>>>     llvm/trunk/llvm/test/CodeGen/
>>>>     llvm/trunk/llvm/test/CodeGen/PowerPC/
>>>>     llvm/trunk/llvm/test/CodeGen/PowerPC/extswsli.ll
>>>>     llvm/trunk/test/CodeGen/PowerPC/extswsli.ll
>>>> Modified:
>>>>     llvm/trunk/lib/Target/PowerPC/PPCISelLowering.cpp
>>>>     llvm/trunk/lib/Target/PowerPC/PPCISelLowering.h
>>>>     llvm/trunk/lib/Target/PowerPC/PPCInstr64Bit.td
>>>>     llvm/trunk/lib/Target/PowerPC/PPCInstrInfo.td
>>>>
>>>> Modified: llvm/trunk/lib/Target/PowerPC/PPCISelLowering.cpp
>>>> URL:
>>>> http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/PowerPC/PPCISelLowering.cpp?rev=340016&r1=340015&r2=340016&view=diff
>>>>
>>>> ==============================================================================
>>>> --- llvm/trunk/lib/Target/PowerPC/PPCISelLowering.cpp (original)
>>>> +++ llvm/trunk/lib/Target/PowerPC/PPCISelLowering.cpp Fri Aug 17
>>>> 05:35:44 2018
>>>> @@ -1351,6 +1351,7 @@ const char *PPCTargetLowering::getTarget
>>>>    case PPCISD::QBFLT:           return "PPCISD::QBFLT";
>>>>    case PPCISD::QVLFSb:          return "PPCISD::QVLFSb";
>>>>    case PPCISD::BUILD_FP128:     return "PPCISD::BUILD_FP128";
>>>> +  case PPCISD::EXTSWSLI:        return "PPCISD::EXTSWSLI";
>>>>    }
>>>>    return nullptr;
>>>>  }
>>>> @@ -14102,7 +14103,30 @@ SDValue PPCTargetLowering::combineSHL(SD
>>>>    if (auto Value = stripModuloOnShift(*this, N, DCI.DAG))
>>>>      return Value;
>>>>
>>>> -  return SDValue();
>>>> +  SDValue N0 = N->getOperand(0);
>>>> +  ConstantSDNode *CN1 = dyn_cast<ConstantSDNode>(N->getOperand(1));
>>>> +  if (!Subtarget.isISA3_0() ||
>>>> +      N0.getOpcode() != ISD::SIGN_EXTEND ||
>>>> +      N0.getOperand(0).getValueType() != MVT::i32 ||
>>>> +      CN1 == nullptr)
>>>> +    return SDValue();
>>>> +
>>>> +  // We can't save an operation here if the value is already extended,
>>>> and
>>>> +  // the existing shift is easier to combine.
>>>> +  SDValue ExtsSrc = N0.getOperand(0);
>>>> +  if (ExtsSrc.getOpcode() == ISD::TRUNCATE &&
>>>> +      ExtsSrc.getOperand(0).getOpcode() == ISD::AssertSext)
>>>> +    return SDValue();
>>>> +
>>>> +  SDLoc DL(N0);
>>>> +  SDValue ShiftBy = SDValue(CN1, 0);
>>>> +  // We want the shift amount to be i32 on the extswli, but the shift
>>>> could
>>>> +  // have an i64.
>>>> +  if (ShiftBy.getValueType() == MVT::i64)
>>>> +    ShiftBy = DCI.DAG.getConstant(CN1->getZExtValue(), DL, MVT::i32);
>>>> +
>>>> +  return DCI.DAG.getNode(PPCISD::EXTSWSLI, DL, MVT::i64,
>>>> N0->getOperand(0),
>>>> +                         ShiftBy);
>>>>  }
>>>>
>>>>  SDValue PPCTargetLowering::combineSRA(SDNode *N, DAGCombinerInfo &DCI)
>>>> const {
>>>>
>>>> Modified: llvm/trunk/lib/Target/PowerPC/PPCISelLowering.h
>>>> URL:
>>>> http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/PowerPC/PPCISelLowering.h?rev=340016&r1=340015&r2=340016&view=diff
>>>>
>>>> ==============================================================================
>>>> --- llvm/trunk/lib/Target/PowerPC/PPCISelLowering.h (original)
>>>> +++ llvm/trunk/lib/Target/PowerPC/PPCISelLowering.h Fri Aug 17 05:35:44
>>>> 2018
>>>> @@ -149,6 +149,10 @@ namespace llvm {
>>>>        /// For vector types, only the last n bits are used. See vsld.
>>>>        SRL, SRA, SHL,
>>>>
>>>> +      /// EXTSWSLI = The PPC extswsli instruction, which does an
>>>> extend-sign
>>>> +      /// word and shift left immediate.
>>>> +      EXTSWSLI,
>>>> +
>>>>        /// The combination of sra[wd]i and addze used to implemented
>>>> signed
>>>>        /// integer division by a power of 2. The first operand is the
>>>> dividend,
>>>>        /// and the second is the constant shift amount (representing the
>>>>
>>>> Modified: llvm/trunk/lib/Target/PowerPC/PPCInstr64Bit.td
>>>> URL:
>>>> http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/PowerPC/PPCInstr64Bit.td?rev=340016&r1=340015&r2=340016&view=diff
>>>>
>>>> ==============================================================================
>>>> --- llvm/trunk/lib/Target/PowerPC/PPCInstr64Bit.td (original)
>>>> +++ llvm/trunk/lib/Target/PowerPC/PPCInstr64Bit.td Fri Aug 17 05:35:44
>>>> 2018
>>>> @@ -717,9 +717,10 @@ defm SRADI  : XSForm_1rc<31, 413, (outs
>>>>                           "sradi", "$rA, $rS, $SH", IIC_IntRotateDI,
>>>>                           [(set i64:$rA, (sra i64:$rS, (i32
>>>> imm:$SH)))]>, isPPC64;
>>>>
>>>> -defm EXTSWSLI : XSForm_1r<31, 445, (outs g8rc:$rA), (ins g8rc:$rS,
>>>> u6imm:$SH),
>>>> +defm EXTSWSLI : XSForm_1r<31, 445, (outs g8rc:$rA), (ins gprc:$rS,
>>>> u6imm:$SH),
>>>>                            "extswsli", "$rA, $rS, $SH", IIC_IntRotateDI,
>>>> -                          []>, isPPC64;
>>>> +                          [(set i64:$rA, (PPCextswsli i32:$rS, (i32
>>>> imm:$SH)))]>,
>>>> +                          isPPC64, Requires<[IsISA3_0]>;
>>>>
>>>>  // For fast-isel:
>>>>  let isCodeGenOnly = 1, Defs = [CARRY] in
>>>>
>>>> Modified: llvm/trunk/lib/Target/PowerPC/PPCInstrInfo.td
>>>> URL:
>>>> http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/PowerPC/PPCInstrInfo.td?rev=340016&r1=340015&r2=340016&view=diff
>>>>
>>>> ==============================================================================
>>>> --- llvm/trunk/lib/Target/PowerPC/PPCInstrInfo.td (original)
>>>> +++ llvm/trunk/lib/Target/PowerPC/PPCInstrInfo.td Fri Aug 17 05:35:44
>>>> 2018
>>>> @@ -114,6 +114,10 @@ def SDT_PPCqvlfsb : SDTypeProfile<1, 1,
>>>>    SDTCisVec<0>, SDTCisPtrTy<1>
>>>>  ]>;
>>>>
>>>> +def SDT_PPCextswsli : SDTypeProfile<1, 2, [  // extswsli
>>>> +  SDTCisInt<0>, SDTCisInt<1>, SDTCisOpSmallerThanOp<1, 0>, SDTCisInt<2>
>>>> +]>;
>>>> +
>>>>
>>>>  //===----------------------------------------------------------------------===//
>>>>  // PowerPC specific DAG Nodes.
>>>>  //
>>>> @@ -218,6 +222,8 @@ def PPCsrl        : SDNode<"PPCISD::SRL"
>>>>  def PPCsra        : SDNode<"PPCISD::SRA"       , SDTIntShiftOp>;
>>>>  def PPCshl        : SDNode<"PPCISD::SHL"       , SDTIntShiftOp>;
>>>>
>>>> +def PPCextswsli : SDNode<"PPCISD::EXTSWSLI" , SDT_PPCextswsli>;
>>>> +
>>>>  // Move 2 i64 values into a VSX register
>>>>  def PPCbuild_fp128: SDNode<"PPCISD::BUILD_FP128",
>>>>                             SDTypeProfile<1, 2,
>>>>
>>>> Added: llvm/trunk/llvm/test/CodeGen/PowerPC/extswsli.ll
>>>> URL:
>>>> http://llvm.org/viewvc/llvm-project/llvm/trunk/llvm/test/CodeGen/PowerPC/extswsli.ll?rev=340016&view=auto
>>>>
>>>> ==============================================================================
>>>> --- llvm/trunk/llvm/test/CodeGen/PowerPC/extswsli.ll (added)
>>>> +++ llvm/trunk/llvm/test/CodeGen/PowerPC/extswsli.ll Fri Aug 17
>>>> 05:35:44 2018
>>>> @@ -0,0 +1,17 @@
>>>> +; RUN: llc -verify-machineinstrs
>>>> -mtriple=powerpc64le-unknown-linux-gnu \
>>>> +; RUN:     -mcpu=pwr9 -ppc-asm-full-reg-names < %s | FileCheck %s
>>>> +
>>>> + at z = external local_unnamed_addr global i32*, align 8
>>>> +
>>>> +; Function Attrs: norecurse nounwind readonly
>>>> +define signext i32 @_Z2tcii(i32 signext %x, i32 signext %y)
>>>> local_unnamed_addr #0 {
>>>> +entry:
>>>> +  %0 = load i32*, i32** @z, align 8
>>>> +  %add = add nsw i32 %y, %x
>>>> +  %idxprom = sext i32 %add to i64
>>>> +  %arrayidx = getelementptr inbounds i32, i32* %0, i64 %idxprom
>>>> +  %1 = load i32, i32* %arrayidx, align 4
>>>> +  ret i32 %1
>>>> +; CHECK-LABEL: @_Z2tcii
>>>> +; CHECK: extswsli {{r[0-9]+}}, {{r[0-9]+}}, 2
>>>> +}
>>>>
>>>> Added: llvm/trunk/test/CodeGen/PowerPC/extswsli.ll
>>>> URL:
>>>> http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/PowerPC/extswsli.ll?rev=340016&view=auto
>>>>
>>>> ==============================================================================
>>>> --- llvm/trunk/test/CodeGen/PowerPC/extswsli.ll (added)
>>>> +++ llvm/trunk/test/CodeGen/PowerPC/extswsli.ll Fri Aug 17 05:35:44 2018
>>>> @@ -0,0 +1,17 @@
>>>> +; RUN: llc -verify-machineinstrs
>>>> -mtriple=powerpc64le-unknown-linux-gnu \
>>>> +; RUN:     -mcpu=pwr9 -ppc-asm-full-reg-names < %s | FileCheck %s
>>>> +
>>>> + at z = external local_unnamed_addr global i32*, align 8
>>>> +
>>>> +; Function Attrs: norecurse nounwind readonly
>>>> +define signext i32 @_Z2tcii(i32 signext %x, i32 signext %y)
>>>> local_unnamed_addr #0 {
>>>> +entry:
>>>> +  %0 = load i32*, i32** @z, align 8
>>>> +  %add = add nsw i32 %y, %x
>>>> +  %idxprom = sext i32 %add to i64
>>>> +  %arrayidx = getelementptr inbounds i32, i32* %0, i64 %idxprom
>>>> +  %1 = load i32, i32* %arrayidx, align 4
>>>> +  ret i32 %1
>>>> +; CHECK-LABEL: @_Z2tcii
>>>> +; CHECK: extswsli {{r[0-9]+}}, {{r[0-9]+}}, 2
>>>> +}
>>>>
>>>>
>>>> _______________________________________________
>>>> llvm-commits mailing list
>>>> llvm-commits at lists.llvm.org
>>>> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits
>>>>
>>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20180822/f90d0235/attachment.html>


More information about the llvm-commits mailing list