This patch adds a new NVPTX back-end to LLVM which supports code generation for NVIDIA PTX 3.0. This back-end will (eventually) replace the current PTX back-end, while maintaining compatibility with it.

The new target machines are:

nvptx (old ptx32) => 32-bit PTX
nvptx64 (old ptx64) => 64-bit PTX

The sources are based on the internal NVIDIA NVPTX back-end, and
contain more functionality than the current PTX back-end currently
provides.

NV_CONTRIB

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@156196 91177308-0d34-0410-b5e6-96231b3b80d8

1 files changed, 326 insertions, 0 deletions

diff --git a/lib/Target/NVPTX/NVPTXInstrInfo.cpp b/lib/Target/NVPTX/NVPTXInstrInfo.cpp
new file mode 100644
index 0000000000..b57592dea6
--- /dev/null
+++ b/lib/Target/NVPTX/NVPTXInstrInfo.cpp
@@ -0,0 +1,326 @@
+//===- NVPTXInstrInfo.cpp - NVPTX Instruction Information -----------------===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+//
+// This file contains the NVPTX implementation of the TargetInstrInfo class.
+//
+//===----------------------------------------------------------------------===//
+
+#include "NVPTX.h"
+#include "NVPTXInstrInfo.h"
+#include "NVPTXTargetMachine.h"
+#define GET_INSTRINFO_CTOR
+#include "NVPTXGenInstrInfo.inc"
+#include "llvm/Function.h"
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/CodeGen/MachineFunction.h"
+#include "llvm/CodeGen/MachineInstrBuilder.h"
+#include "llvm/CodeGen/MachineRegisterInfo.h"
+#include <cstdio>
+
+
+using namespace llvm;
+
+// FIXME: Add the subtarget support on this constructor.
+NVPTXInstrInfo::NVPTXInstrInfo(NVPTXTargetMachine &tm)
+: NVPTXGenInstrInfo(),
+  TM(tm),
+  RegInfo(*this, *TM.getSubtargetImpl()) {}
+
+
+void NVPTXInstrInfo::copyPhysReg (MachineBasicBlock &MBB,
+                                  MachineBasicBlock::iterator I, DebugLoc DL,
+                                  unsigned DestReg, unsigned SrcReg,
+                                  bool KillSrc) const {
+  if (NVPTX::Int32RegsRegClass.contains(DestReg) &&
+      NVPTX::Int32RegsRegClass.contains(SrcReg))
+    BuildMI(MBB, I, DL, get(NVPTX::IMOV32rr), DestReg)
+    .addReg(SrcReg, getKillRegState(KillSrc));
+  else if (NVPTX::Int8RegsRegClass.contains(DestReg) &&
+      NVPTX::Int8RegsRegClass.contains(SrcReg))
+    BuildMI(MBB, I, DL, get(NVPTX::IMOV8rr), DestReg)
+    .addReg(SrcReg, getKillRegState(KillSrc));
+  else if (NVPTX::Int1RegsRegClass.contains(DestReg) &&
+      NVPTX::Int1RegsRegClass.contains(SrcReg))
+    BuildMI(MBB, I, DL, get(NVPTX::IMOV1rr), DestReg)
+    .addReg(SrcReg, getKillRegState(KillSrc));
+  else if (NVPTX::Float32RegsRegClass.contains(DestReg) &&
+      NVPTX::Float32RegsRegClass.contains(SrcReg))
+    BuildMI(MBB, I, DL, get(NVPTX::FMOV32rr), DestReg)
+    .addReg(SrcReg, getKillRegState(KillSrc));
+  else if (NVPTX::Int16RegsRegClass.contains(DestReg) &&
+      NVPTX::Int16RegsRegClass.contains(SrcReg))
+    BuildMI(MBB, I, DL, get(NVPTX::IMOV16rr), DestReg)
+    .addReg(SrcReg, getKillRegState(KillSrc));
+  else if (NVPTX::Int64RegsRegClass.contains(DestReg) &&
+      NVPTX::Int64RegsRegClass.contains(SrcReg))
+    BuildMI(MBB, I, DL, get(NVPTX::IMOV64rr), DestReg)
+    .addReg(SrcReg, getKillRegState(KillSrc));
+  else if (NVPTX::Float64RegsRegClass.contains(DestReg) &&
+      NVPTX::Float64RegsRegClass.contains(SrcReg))
+    BuildMI(MBB, I, DL, get(NVPTX::FMOV64rr), DestReg)
+    .addReg(SrcReg, getKillRegState(KillSrc));
+  else if (NVPTX::V4F32RegsRegClass.contains(DestReg) &&
+      NVPTX::V4F32RegsRegClass.contains(SrcReg))
+    BuildMI(MBB, I, DL, get(NVPTX::V4f32Mov), DestReg)
+    .addReg(SrcReg, getKillRegState(KillSrc));
+  else if (NVPTX::V4I32RegsRegClass.contains(DestReg) &&
+      NVPTX::V4I32RegsRegClass.contains(SrcReg))
+    BuildMI(MBB, I, DL, get(NVPTX::V4i32Mov), DestReg)
+    .addReg(SrcReg, getKillRegState(KillSrc));
+  else if (NVPTX::V2F32RegsRegClass.contains(DestReg) &&
+      NVPTX::V2F32RegsRegClass.contains(SrcReg))
+    BuildMI(MBB, I, DL, get(NVPTX::V2f32Mov), DestReg)
+    .addReg(SrcReg, getKillRegState(KillSrc));
+  else if (NVPTX::V2I32RegsRegClass.contains(DestReg) &&
+      NVPTX::V2I32RegsRegClass.contains(SrcReg))
+    BuildMI(MBB, I, DL, get(NVPTX::V2i32Mov), DestReg)
+    .addReg(SrcReg, getKillRegState(KillSrc));
+  else if (NVPTX::V4I8RegsRegClass.contains(DestReg) &&
+      NVPTX::V4I8RegsRegClass.contains(SrcReg))
+    BuildMI(MBB, I, DL, get(NVPTX::V4i8Mov), DestReg)
+    .addReg(SrcReg, getKillRegState(KillSrc));
+  else if (NVPTX::V2I8RegsRegClass.contains(DestReg) &&
+      NVPTX::V2I8RegsRegClass.contains(SrcReg))
+    BuildMI(MBB, I, DL, get(NVPTX::V2i8Mov), DestReg)
+    .addReg(SrcReg, getKillRegState(KillSrc));
+  else if (NVPTX::V4I16RegsRegClass.contains(DestReg) &&
+      NVPTX::V4I16RegsRegClass.contains(SrcReg))
+    BuildMI(MBB, I, DL, get(NVPTX::V4i16Mov), DestReg)
+    .addReg(SrcReg, getKillRegState(KillSrc));
+  else if (NVPTX::V2I16RegsRegClass.contains(DestReg) &&
+      NVPTX::V2I16RegsRegClass.contains(SrcReg))
+    BuildMI(MBB, I, DL, get(NVPTX::V2i16Mov), DestReg)
+    .addReg(SrcReg, getKillRegState(KillSrc));
+  else if (NVPTX::V2I64RegsRegClass.contains(DestReg) &&
+      NVPTX::V2I64RegsRegClass.contains(SrcReg))
+    BuildMI(MBB, I, DL, get(NVPTX::V2i64Mov), DestReg)
+    .addReg(SrcReg, getKillRegState(KillSrc));
+  else if (NVPTX::V2F64RegsRegClass.contains(DestReg) &&
+      NVPTX::V2F64RegsRegClass.contains(SrcReg))
+    BuildMI(MBB, I, DL, get(NVPTX::V2f64Mov), DestReg)
+    .addReg(SrcReg, getKillRegState(KillSrc));
+  else {
+    assert(0 && "Don't know how to copy a register");
+  }
+}
+
+bool NVPTXInstrInfo::isMoveInstr(const MachineInstr &MI,
+                                 unsigned &SrcReg,
+                                 unsigned &DestReg) const {
+  // Look for the appropriate part of TSFlags
+  bool isMove = false;
+
+  unsigned TSFlags = (MI.getDesc().TSFlags & NVPTX::SimpleMoveMask) >>
+      NVPTX::SimpleMoveShift;
+  isMove = (TSFlags == 1);
+
+  if (isMove) {
+    MachineOperand dest = MI.getOperand(0);
+    MachineOperand src = MI.getOperand(1);
+    assert(dest.isReg() && "dest of a movrr is not a reg");
+    assert(src.isReg() && "src of a movrr is not a reg");
+
+    SrcReg = src.getReg();
+    DestReg = dest.getReg();
+    return true;
+  }
+
+  return false;
+}
+
+bool  NVPTXInstrInfo::isReadSpecialReg(MachineInstr &MI) const
+{
+  switch (MI.getOpcode()) {
+  default: return false;
+  case NVPTX::INT_PTX_SREG_NTID_X:
+  case NVPTX::INT_PTX_SREG_NTID_Y:
+  case NVPTX::INT_PTX_SREG_NTID_Z:
+  case NVPTX::INT_PTX_SREG_TID_X:
+  case NVPTX::INT_PTX_SREG_TID_Y:
+  case NVPTX::INT_PTX_SREG_TID_Z:
+  case NVPTX::INT_PTX_SREG_CTAID_X:
+  case NVPTX::INT_PTX_SREG_CTAID_Y:
+  case NVPTX::INT_PTX_SREG_CTAID_Z:
+  case NVPTX::INT_PTX_SREG_NCTAID_X:
+  case NVPTX::INT_PTX_SREG_NCTAID_Y:
+  case NVPTX::INT_PTX_SREG_NCTAID_Z:
+  case NVPTX::INT_PTX_SREG_WARPSIZE:
+    return true;
+  }
+}
+
+
+bool NVPTXInstrInfo::isLoadInstr(const MachineInstr &MI,
+                                 unsigned &AddrSpace) const {
+  bool isLoad = false;
+  unsigned TSFlags = (MI.getDesc().TSFlags & NVPTX::isLoadMask) >>
+      NVPTX::isLoadShift;
+  isLoad = (TSFlags == 1);
+  if (isLoad)
+    AddrSpace = getLdStCodeAddrSpace(MI);
+  return isLoad;
+}
+
+bool NVPTXInstrInfo::isStoreInstr(const MachineInstr &MI,
+                                  unsigned &AddrSpace) const {
+  bool isStore = false;
+  unsigned TSFlags = (MI.getDesc().TSFlags & NVPTX::isStoreMask) >>
+      NVPTX::isStoreShift;
+  isStore = (TSFlags == 1);
+  if (isStore)
+    AddrSpace = getLdStCodeAddrSpace(MI);
+  return isStore;
+}
+
+
+bool NVPTXInstrInfo::CanTailMerge(const MachineInstr *MI) const {
+  unsigned addrspace = 0;
+  if (MI->getOpcode() == NVPTX::INT_CUDA_SYNCTHREADS)
+    return false;
+  if (isLoadInstr(*MI, addrspace))
+    if (addrspace == NVPTX::PTXLdStInstCode::SHARED)
+      return false;
+  if (isStoreInstr(*MI, addrspace))
+    if (addrspace == NVPTX::PTXLdStInstCode::SHARED)
+      return false;
+  return true;
+}
+
+
+/// AnalyzeBranch - Analyze the branching code at the end of MBB, returning
+/// true if it cannot be understood (e.g. it's a switch dispatch or isn't
+/// implemented for a target).  Upon success, this returns false and returns
+/// with the following information in various cases:
+///
+/// 1. If this block ends with no branches (it just falls through to its succ)
+///    just return false, leaving TBB/FBB null.
+/// 2. If this block ends with only an unconditional branch, it sets TBB to be
+///    the destination block.
+/// 3. If this block ends with an conditional branch and it falls through to
+///    an successor block, it sets TBB to be the branch destination block and a
+///    list of operands that evaluate the condition. These
+///    operands can be passed to other TargetInstrInfo methods to create new
+///    branches.
+/// 4. If this block ends with an conditional branch and an unconditional
+///    block, it returns the 'true' destination in TBB, the 'false' destination
+///    in FBB, and a list of operands that evaluate the condition. These
+///    operands can be passed to other TargetInstrInfo methods to create new
+///    branches.
+///
+/// Note that RemoveBranch and InsertBranch must be implemented to support
+/// cases where this method returns success.
+///
+bool NVPTXInstrInfo::AnalyzeBranch(MachineBasicBlock &MBB,
+                                   MachineBasicBlock *&TBB,
+                                   MachineBasicBlock *&FBB,
+                                   SmallVectorImpl<MachineOperand> &Cond,
+                                   bool AllowModify) const {
+  // If the block has no terminators, it just falls into the block after it.
+  MachineBasicBlock::iterator I = MBB.end();
+  if (I == MBB.begin() || !isUnpredicatedTerminator(--I))
+    return false;
+
+  // Get the last instruction in the block.
+  MachineInstr *LastInst = I;
+
+  // If there is only one terminator instruction, process it.
+  if (I == MBB.begin() || !isUnpredicatedTerminator(--I)) {
+    if (LastInst->getOpcode() == NVPTX::GOTO) {
+      TBB = LastInst->getOperand(0).getMBB();
+      return false;
+    } else if (LastInst->getOpcode() == NVPTX::CBranch) {
+      // Block ends with fall-through condbranch.
+      TBB = LastInst->getOperand(1).getMBB();
+      Cond.push_back(LastInst->getOperand(0));
+      return false;
+    }
+    // Otherwise, don't know what this is.
+    return true;
+  }
+
+  // Get the instruction before it if it's a terminator.
+  MachineInstr *SecondLastInst = I;
+
+  // If there are three terminators, we don't know what sort of block this is.
+  if (SecondLastInst && I != MBB.begin() &&
+      isUnpredicatedTerminator(--I))
+    return true;
+
+  // If the block ends with NVPTX::GOTO and NVPTX:CBranch, handle it.
+  if (SecondLastInst->getOpcode() == NVPTX::CBranch &&
+      LastInst->getOpcode() == NVPTX::GOTO) {
+    TBB =  SecondLastInst->getOperand(1).getMBB();
+    Cond.push_back(SecondLastInst->getOperand(0));
+    FBB = LastInst->getOperand(0).getMBB();
+    return false;
+  }
+
+  // If the block ends with two NVPTX:GOTOs, handle it.  The second one is not
+  // executed, so remove it.
+  if (SecondLastInst->getOpcode() == NVPTX::GOTO &&
+      LastInst->getOpcode() == NVPTX::GOTO) {
+    TBB = SecondLastInst->getOperand(0).getMBB();
+    I = LastInst;
+    if (AllowModify)
+      I->eraseFromParent();
+    return false;
+  }
+
+  // Otherwise, can't handle this.
+  return true;
+}
+
+unsigned NVPTXInstrInfo::RemoveBranch(MachineBasicBlock &MBB) const {
+  MachineBasicBlock::iterator I = MBB.end();
+  if (I == MBB.begin()) return 0;
+  --I;
+  if (I->getOpcode() != NVPTX::GOTO && I->getOpcode() != NVPTX::CBranch)
+    return 0;
+
+  // Remove the branch.
+  I->eraseFromParent();
+
+  I = MBB.end();
+
+  if (I == MBB.begin()) return 1;
+  --I;
+  if (I->getOpcode() != NVPTX::CBranch)
+    return 1;
+
+  // Remove the branch.
+  I->eraseFromParent();
+  return 2;
+}
+
+unsigned
+NVPTXInstrInfo::InsertBranch(MachineBasicBlock &MBB, MachineBasicBlock *TBB,
+                             MachineBasicBlock *FBB,
+                             const SmallVectorImpl<MachineOperand> &Cond,
+                             DebugLoc DL) const {
+  // Shouldn't be a fall through.
+  assert(TBB && "InsertBranch must not be told to insert a fallthrough");
+  assert((Cond.size() == 1 || Cond.size() == 0) &&
+         "NVPTX branch conditions have two components!");
+
+  // One-way branch.
+  if (FBB == 0) {
+    if (Cond.empty())   // Unconditional branch
+      BuildMI(&MBB, DL, get(NVPTX::GOTO)).addMBB(TBB);
+    else                // Conditional branch
+      BuildMI(&MBB, DL, get(NVPTX::CBranch))
+      .addReg(Cond[0].getReg()).addMBB(TBB);
+    return 1;
+  }
+
+  // Two-way Conditional Branch.
+  BuildMI(&MBB, DL, get(NVPTX::CBranch))
+  .addReg(Cond[0].getReg()).addMBB(TBB);
+  BuildMI(&MBB, DL, get(NVPTX::GOTO)).addMBB(FBB);
+  return 2;
+}