2004-07-26 20:45:48 +02:00
|
|
|
//===-- X86FloatingPoint.cpp - Floating point Reg -> Stack converter ------===//
|
2003-10-20 21:43:21 +02:00
|
|
|
//
|
|
|
|
// The LLVM Compiler Infrastructure
|
|
|
|
//
|
|
|
|
// This file was developed by the LLVM research group and is distributed under
|
|
|
|
// the University of Illinois Open Source License. See LICENSE.TXT for details.
|
|
|
|
//
|
|
|
|
//===----------------------------------------------------------------------===//
|
2003-01-13 02:01:59 +01:00
|
|
|
//
|
|
|
|
// This file defines the pass which converts floating point instructions from
|
2004-01-30 23:25:18 +01:00
|
|
|
// virtual registers into register stack instructions. This pass uses live
|
|
|
|
// variable information to indicate where the FPn registers are used and their
|
|
|
|
// lifetimes.
|
|
|
|
//
|
|
|
|
// This pass is hampered by the lack of decent CFG manipulation routines for
|
|
|
|
// machine code. In particular, this wants to be able to split critical edges
|
|
|
|
// as necessary, traverse the machine basic block CFG in depth-first order, and
|
|
|
|
// allow there to be multiple machine basic blocks for each LLVM basicblock
|
|
|
|
// (needed for critical edge splitting).
|
|
|
|
//
|
|
|
|
// In particular, this pass currently barfs on critical edges. Because of this,
|
|
|
|
// it requires the instruction selector to insert FP_REG_KILL instructions on
|
|
|
|
// the exits of any basic block that has critical edges going from it, or which
|
|
|
|
// branch to a critical basic block.
|
|
|
|
//
|
|
|
|
// FIXME: this is not implemented yet. The stackifier pass only works on local
|
|
|
|
// basic blocks.
|
2003-01-13 02:01:59 +01:00
|
|
|
//
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
|
2003-08-03 23:14:38 +02:00
|
|
|
#define DEBUG_TYPE "fp"
|
2003-01-13 02:01:59 +01:00
|
|
|
#include "X86.h"
|
|
|
|
#include "X86InstrInfo.h"
|
|
|
|
#include "llvm/CodeGen/MachineFunctionPass.h"
|
|
|
|
#include "llvm/CodeGen/MachineInstrBuilder.h"
|
|
|
|
#include "llvm/CodeGen/LiveVariables.h"
|
2003-12-13 06:36:22 +01:00
|
|
|
#include "llvm/CodeGen/Passes.h"
|
2003-01-14 23:00:31 +01:00
|
|
|
#include "llvm/Target/TargetInstrInfo.h"
|
2003-01-13 02:01:59 +01:00
|
|
|
#include "llvm/Target/TargetMachine.h"
|
2003-08-02 00:21:34 +02:00
|
|
|
#include "Support/Debug.h"
|
2004-01-30 23:25:18 +01:00
|
|
|
#include "Support/DepthFirstIterator.h"
|
2003-01-13 02:01:59 +01:00
|
|
|
#include "Support/Statistic.h"
|
2004-02-14 02:18:34 +01:00
|
|
|
#include "Support/STLExtras.h"
|
2003-01-13 02:01:59 +01:00
|
|
|
#include <algorithm>
|
2004-01-30 23:25:18 +01:00
|
|
|
#include <set>
|
2003-12-20 10:58:55 +01:00
|
|
|
using namespace llvm;
|
2003-11-11 23:41:34 +01:00
|
|
|
|
2003-01-13 02:01:59 +01:00
|
|
|
namespace {
|
|
|
|
Statistic<> NumFXCH("x86-codegen", "Number of fxch instructions inserted");
|
|
|
|
Statistic<> NumFP ("x86-codegen", "Number of floating point instructions");
|
|
|
|
|
|
|
|
struct FPS : public MachineFunctionPass {
|
|
|
|
virtual bool runOnMachineFunction(MachineFunction &MF);
|
|
|
|
|
|
|
|
virtual const char *getPassName() const { return "X86 FP Stackifier"; }
|
|
|
|
|
|
|
|
virtual void getAnalysisUsage(AnalysisUsage &AU) const {
|
|
|
|
AU.addRequired<LiveVariables>();
|
|
|
|
MachineFunctionPass::getAnalysisUsage(AU);
|
|
|
|
}
|
|
|
|
private:
|
|
|
|
LiveVariables *LV; // Live variable info for current function...
|
|
|
|
MachineBasicBlock *MBB; // Current basic block
|
|
|
|
unsigned Stack[8]; // FP<n> Registers in each stack slot...
|
|
|
|
unsigned RegMap[8]; // Track which stack slot contains each register
|
|
|
|
unsigned StackTop; // The current top of the FP stack.
|
|
|
|
|
|
|
|
void dumpStack() const {
|
|
|
|
std::cerr << "Stack contents:";
|
|
|
|
for (unsigned i = 0; i != StackTop; ++i) {
|
|
|
|
std::cerr << " FP" << Stack[i];
|
|
|
|
assert(RegMap[Stack[i]] == i && "Stack[] doesn't match RegMap[]!");
|
|
|
|
}
|
|
|
|
std::cerr << "\n";
|
|
|
|
}
|
|
|
|
private:
|
|
|
|
// getSlot - Return the stack slot number a particular register number is
|
|
|
|
// in...
|
|
|
|
unsigned getSlot(unsigned RegNo) const {
|
|
|
|
assert(RegNo < 8 && "Regno out of range!");
|
|
|
|
return RegMap[RegNo];
|
|
|
|
}
|
|
|
|
|
|
|
|
// getStackEntry - Return the X86::FP<n> register in register ST(i)
|
|
|
|
unsigned getStackEntry(unsigned STi) const {
|
|
|
|
assert(STi < StackTop && "Access past stack top!");
|
|
|
|
return Stack[StackTop-1-STi];
|
|
|
|
}
|
|
|
|
|
|
|
|
// getSTReg - Return the X86::ST(i) register which contains the specified
|
|
|
|
// FP<RegNo> register
|
|
|
|
unsigned getSTReg(unsigned RegNo) const {
|
2003-11-11 23:41:34 +01:00
|
|
|
return StackTop - 1 - getSlot(RegNo) + llvm::X86::ST0;
|
2003-01-13 02:01:59 +01:00
|
|
|
}
|
|
|
|
|
2004-02-02 20:23:15 +01:00
|
|
|
// pushReg - Push the specified FP<n> register onto the stack
|
2003-01-13 02:01:59 +01:00
|
|
|
void pushReg(unsigned Reg) {
|
|
|
|
assert(Reg < 8 && "Register number out of range!");
|
|
|
|
assert(StackTop < 8 && "Stack overflow!");
|
|
|
|
Stack[StackTop] = Reg;
|
|
|
|
RegMap[Reg] = StackTop++;
|
|
|
|
}
|
|
|
|
|
|
|
|
bool isAtTop(unsigned RegNo) const { return getSlot(RegNo) == StackTop-1; }
|
|
|
|
void moveToTop(unsigned RegNo, MachineBasicBlock::iterator &I) {
|
|
|
|
if (!isAtTop(RegNo)) {
|
|
|
|
unsigned Slot = getSlot(RegNo);
|
|
|
|
unsigned STReg = getSTReg(RegNo);
|
|
|
|
unsigned RegOnTop = getStackEntry(0);
|
|
|
|
|
|
|
|
// Swap the slots the regs are in
|
|
|
|
std::swap(RegMap[RegNo], RegMap[RegOnTop]);
|
|
|
|
|
|
|
|
// Swap stack slot contents
|
|
|
|
assert(RegMap[RegOnTop] < StackTop);
|
|
|
|
std::swap(Stack[RegMap[RegOnTop]], Stack[StackTop-1]);
|
|
|
|
|
|
|
|
// Emit an fxch to update the runtime processors version of the state
|
Simplify code by using the more powerful BuildMI forms.
Implement a small optimization. In test/Regression/CodeGen/X86/select.ll,
we now generate this for foldSel3:
foldSel3:
mov %AL, BYTE PTR [%ESP + 4]
fld DWORD PTR [%ESP + 8]
fld DWORD PTR [%ESP + 12]
mov %EAX, DWORD PTR [%ESP + 16]
mov %ECX, DWORD PTR [%ESP + 20]
cmp %EAX, %ECX
fxch %ST(1)
fcmovae %ST(0), %ST(1)
*** fstp %ST(1)
ret
Instead of:
foldSel3:
mov %AL, BYTE PTR [%ESP + 4]
fld DWORD PTR [%ESP + 8]
fld DWORD PTR [%ESP + 12]
mov %EAX, DWORD PTR [%ESP + 16]
mov %ECX, DWORD PTR [%ESP + 20]
cmp %EAX, %ECX
fxch %ST(1)
fcmovae %ST(0), %ST(1)
*** fxch %ST(1)
*** fstp %ST(0)
ret
In practice, this only effects code size: performance should be basically
unaffected.
llvm-svn: 12588
2004-04-01 06:06:09 +02:00
|
|
|
BuildMI(*MBB, I, X86::FXCH, 1).addReg(STReg);
|
2003-01-13 02:01:59 +01:00
|
|
|
NumFXCH++;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
Simplify code by using the more powerful BuildMI forms.
Implement a small optimization. In test/Regression/CodeGen/X86/select.ll,
we now generate this for foldSel3:
foldSel3:
mov %AL, BYTE PTR [%ESP + 4]
fld DWORD PTR [%ESP + 8]
fld DWORD PTR [%ESP + 12]
mov %EAX, DWORD PTR [%ESP + 16]
mov %ECX, DWORD PTR [%ESP + 20]
cmp %EAX, %ECX
fxch %ST(1)
fcmovae %ST(0), %ST(1)
*** fstp %ST(1)
ret
Instead of:
foldSel3:
mov %AL, BYTE PTR [%ESP + 4]
fld DWORD PTR [%ESP + 8]
fld DWORD PTR [%ESP + 12]
mov %EAX, DWORD PTR [%ESP + 16]
mov %ECX, DWORD PTR [%ESP + 20]
cmp %EAX, %ECX
fxch %ST(1)
fcmovae %ST(0), %ST(1)
*** fxch %ST(1)
*** fstp %ST(0)
ret
In practice, this only effects code size: performance should be basically
unaffected.
llvm-svn: 12588
2004-04-01 06:06:09 +02:00
|
|
|
void duplicateToTop(unsigned RegNo, unsigned AsReg, MachineInstr *I) {
|
2003-01-13 02:01:59 +01:00
|
|
|
unsigned STReg = getSTReg(RegNo);
|
|
|
|
pushReg(AsReg); // New register on top of stack
|
|
|
|
|
Simplify code by using the more powerful BuildMI forms.
Implement a small optimization. In test/Regression/CodeGen/X86/select.ll,
we now generate this for foldSel3:
foldSel3:
mov %AL, BYTE PTR [%ESP + 4]
fld DWORD PTR [%ESP + 8]
fld DWORD PTR [%ESP + 12]
mov %EAX, DWORD PTR [%ESP + 16]
mov %ECX, DWORD PTR [%ESP + 20]
cmp %EAX, %ECX
fxch %ST(1)
fcmovae %ST(0), %ST(1)
*** fstp %ST(1)
ret
Instead of:
foldSel3:
mov %AL, BYTE PTR [%ESP + 4]
fld DWORD PTR [%ESP + 8]
fld DWORD PTR [%ESP + 12]
mov %EAX, DWORD PTR [%ESP + 16]
mov %ECX, DWORD PTR [%ESP + 20]
cmp %EAX, %ECX
fxch %ST(1)
fcmovae %ST(0), %ST(1)
*** fxch %ST(1)
*** fstp %ST(0)
ret
In practice, this only effects code size: performance should be basically
unaffected.
llvm-svn: 12588
2004-04-01 06:06:09 +02:00
|
|
|
BuildMI(*MBB, I, X86::FLDrr, 1).addReg(STReg);
|
2003-01-13 02:01:59 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
// popStackAfter - Pop the current value off of the top of the FP stack
|
|
|
|
// after the specified instruction.
|
|
|
|
void popStackAfter(MachineBasicBlock::iterator &I);
|
|
|
|
|
Simplify code by using the more powerful BuildMI forms.
Implement a small optimization. In test/Regression/CodeGen/X86/select.ll,
we now generate this for foldSel3:
foldSel3:
mov %AL, BYTE PTR [%ESP + 4]
fld DWORD PTR [%ESP + 8]
fld DWORD PTR [%ESP + 12]
mov %EAX, DWORD PTR [%ESP + 16]
mov %ECX, DWORD PTR [%ESP + 20]
cmp %EAX, %ECX
fxch %ST(1)
fcmovae %ST(0), %ST(1)
*** fstp %ST(1)
ret
Instead of:
foldSel3:
mov %AL, BYTE PTR [%ESP + 4]
fld DWORD PTR [%ESP + 8]
fld DWORD PTR [%ESP + 12]
mov %EAX, DWORD PTR [%ESP + 16]
mov %ECX, DWORD PTR [%ESP + 20]
cmp %EAX, %ECX
fxch %ST(1)
fcmovae %ST(0), %ST(1)
*** fxch %ST(1)
*** fstp %ST(0)
ret
In practice, this only effects code size: performance should be basically
unaffected.
llvm-svn: 12588
2004-04-01 06:06:09 +02:00
|
|
|
// freeStackSlotAfter - Free the specified register from the register stack,
|
|
|
|
// so that it is no longer in a register. If the register is currently at
|
|
|
|
// the top of the stack, we just pop the current instruction, otherwise we
|
|
|
|
// store the current top-of-stack into the specified slot, then pop the top
|
|
|
|
// of stack.
|
|
|
|
void freeStackSlotAfter(MachineBasicBlock::iterator &I, unsigned Reg);
|
|
|
|
|
2003-01-13 02:01:59 +01:00
|
|
|
bool processBasicBlock(MachineFunction &MF, MachineBasicBlock &MBB);
|
|
|
|
|
|
|
|
void handleZeroArgFP(MachineBasicBlock::iterator &I);
|
|
|
|
void handleOneArgFP(MachineBasicBlock::iterator &I);
|
2004-02-02 20:23:15 +01:00
|
|
|
void handleOneArgFPRW(MachineBasicBlock::iterator &I);
|
2003-01-13 02:01:59 +01:00
|
|
|
void handleTwoArgFP(MachineBasicBlock::iterator &I);
|
2004-06-11 06:25:06 +02:00
|
|
|
void handleCompareFP(MachineBasicBlock::iterator &I);
|
2004-04-01 00:02:36 +02:00
|
|
|
void handleCondMovFP(MachineBasicBlock::iterator &I);
|
2003-01-13 02:01:59 +01:00
|
|
|
void handleSpecialFP(MachineBasicBlock::iterator &I);
|
|
|
|
};
|
|
|
|
}
|
|
|
|
|
2003-12-20 10:58:55 +01:00
|
|
|
FunctionPass *llvm::createX86FloatingPointStackifierPass() { return new FPS(); }
|
2003-01-13 02:01:59 +01:00
|
|
|
|
|
|
|
/// runOnMachineFunction - Loop over all of the basic blocks, transforming FP
|
|
|
|
/// register references into FP stack references.
|
|
|
|
///
|
|
|
|
bool FPS::runOnMachineFunction(MachineFunction &MF) {
|
|
|
|
LV = &getAnalysis<LiveVariables>();
|
|
|
|
StackTop = 0;
|
|
|
|
|
2004-01-30 23:25:18 +01:00
|
|
|
// Process the function in depth first order so that we process at least one
|
|
|
|
// of the predecessors for every reachable block in the function.
|
2004-05-01 23:27:53 +02:00
|
|
|
std::set<MachineBasicBlock*> Processed;
|
|
|
|
MachineBasicBlock *Entry = MF.begin();
|
2004-01-30 23:25:18 +01:00
|
|
|
|
|
|
|
bool Changed = false;
|
2004-05-01 23:27:53 +02:00
|
|
|
for (df_ext_iterator<MachineBasicBlock*, std::set<MachineBasicBlock*> >
|
2004-01-30 23:25:18 +01:00
|
|
|
I = df_ext_begin(Entry, Processed), E = df_ext_end(Entry, Processed);
|
|
|
|
I != E; ++I)
|
2004-05-01 23:27:53 +02:00
|
|
|
Changed |= processBasicBlock(MF, **I);
|
2004-01-30 23:25:18 +01:00
|
|
|
|
2003-01-13 02:01:59 +01:00
|
|
|
return Changed;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// processBasicBlock - Loop over all of the instructions in the basic block,
|
|
|
|
/// transforming FP instructions into their stack form.
|
|
|
|
///
|
|
|
|
bool FPS::processBasicBlock(MachineFunction &MF, MachineBasicBlock &BB) {
|
2004-06-02 07:55:25 +02:00
|
|
|
const TargetInstrInfo &TII = *MF.getTarget().getInstrInfo();
|
2003-01-13 02:01:59 +01:00
|
|
|
bool Changed = false;
|
|
|
|
MBB = &BB;
|
|
|
|
|
|
|
|
for (MachineBasicBlock::iterator I = BB.begin(); I != BB.end(); ++I) {
|
2004-02-12 03:27:10 +01:00
|
|
|
MachineInstr *MI = I;
|
2003-01-13 02:01:59 +01:00
|
|
|
unsigned Flags = TII.get(MI->getOpcode()).TSFlags;
|
2004-01-30 23:25:18 +01:00
|
|
|
if ((Flags & X86II::FPTypeMask) == X86II::NotFP)
|
|
|
|
continue; // Efficiently ignore non-fp insts!
|
2003-01-13 02:01:59 +01:00
|
|
|
|
2004-02-12 03:27:10 +01:00
|
|
|
MachineInstr *PrevMI = 0;
|
2004-02-14 02:18:34 +01:00
|
|
|
if (I != BB.begin())
|
|
|
|
PrevMI = prior(I);
|
2003-01-13 02:01:59 +01:00
|
|
|
|
|
|
|
++NumFP; // Keep track of # of pseudo instrs
|
|
|
|
DEBUG(std::cerr << "\nFPInst:\t";
|
2004-06-25 02:13:11 +02:00
|
|
|
MI->print(std::cerr, &(MF.getTarget())));
|
2003-01-13 02:01:59 +01:00
|
|
|
|
|
|
|
// Get dead variables list now because the MI pointer may be deleted as part
|
|
|
|
// of processing!
|
|
|
|
LiveVariables::killed_iterator IB = LV->dead_begin(MI);
|
|
|
|
LiveVariables::killed_iterator IE = LV->dead_end(MI);
|
|
|
|
|
|
|
|
DEBUG(const MRegisterInfo *MRI = MF.getTarget().getRegisterInfo();
|
|
|
|
LiveVariables::killed_iterator I = LV->killed_begin(MI);
|
|
|
|
LiveVariables::killed_iterator E = LV->killed_end(MI);
|
|
|
|
if (I != E) {
|
|
|
|
std::cerr << "Killed Operands:";
|
|
|
|
for (; I != E; ++I)
|
|
|
|
std::cerr << " %" << MRI->getName(I->second);
|
|
|
|
std::cerr << "\n";
|
|
|
|
});
|
|
|
|
|
|
|
|
switch (Flags & X86II::FPTypeMask) {
|
2004-02-02 20:23:15 +01:00
|
|
|
case X86II::ZeroArgFP: handleZeroArgFP(I); break;
|
2004-04-01 00:02:36 +02:00
|
|
|
case X86II::OneArgFP: handleOneArgFP(I); break; // fstp ST(0)
|
2004-02-02 20:23:15 +01:00
|
|
|
case X86II::OneArgFPRW: handleOneArgFPRW(I); break; // ST(0) = fsqrt(ST(0))
|
2004-06-11 06:41:24 +02:00
|
|
|
case X86II::TwoArgFP: handleTwoArgFP(I); break;
|
|
|
|
case X86II::CompareFP: handleCompareFP(I); break;
|
2004-04-01 00:02:36 +02:00
|
|
|
case X86II::CondMovFP: handleCondMovFP(I); break;
|
2004-02-02 20:23:15 +01:00
|
|
|
case X86II::SpecialFP: handleSpecialFP(I); break;
|
2003-01-13 02:01:59 +01:00
|
|
|
default: assert(0 && "Unknown FP Type!");
|
|
|
|
}
|
|
|
|
|
|
|
|
// Check to see if any of the values defined by this instruction are dead
|
|
|
|
// after definition. If so, pop them.
|
|
|
|
for (; IB != IE; ++IB) {
|
|
|
|
unsigned Reg = IB->second;
|
|
|
|
if (Reg >= X86::FP0 && Reg <= X86::FP6) {
|
|
|
|
DEBUG(std::cerr << "Register FP#" << Reg-X86::FP0 << " is dead!\n");
|
2004-06-11 06:25:06 +02:00
|
|
|
freeStackSlotAfter(I, Reg-X86::FP0);
|
2003-01-13 02:01:59 +01:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
// Print out all of the instructions expanded to if -debug
|
2004-02-15 01:46:41 +01:00
|
|
|
DEBUG(
|
|
|
|
MachineBasicBlock::iterator PrevI(PrevMI);
|
|
|
|
if (I == PrevI) {
|
Simplify code by using the more powerful BuildMI forms.
Implement a small optimization. In test/Regression/CodeGen/X86/select.ll,
we now generate this for foldSel3:
foldSel3:
mov %AL, BYTE PTR [%ESP + 4]
fld DWORD PTR [%ESP + 8]
fld DWORD PTR [%ESP + 12]
mov %EAX, DWORD PTR [%ESP + 16]
mov %ECX, DWORD PTR [%ESP + 20]
cmp %EAX, %ECX
fxch %ST(1)
fcmovae %ST(0), %ST(1)
*** fstp %ST(1)
ret
Instead of:
foldSel3:
mov %AL, BYTE PTR [%ESP + 4]
fld DWORD PTR [%ESP + 8]
fld DWORD PTR [%ESP + 12]
mov %EAX, DWORD PTR [%ESP + 16]
mov %ECX, DWORD PTR [%ESP + 20]
cmp %EAX, %ECX
fxch %ST(1)
fcmovae %ST(0), %ST(1)
*** fxch %ST(1)
*** fstp %ST(0)
ret
In practice, this only effects code size: performance should be basically
unaffected.
llvm-svn: 12588
2004-04-01 06:06:09 +02:00
|
|
|
std::cerr << "Just deleted pseudo instruction\n";
|
2004-02-15 01:46:41 +01:00
|
|
|
} else {
|
|
|
|
MachineBasicBlock::iterator Start = I;
|
|
|
|
// Rewind to first instruction newly inserted.
|
|
|
|
while (Start != BB.begin() && prior(Start) != PrevI) --Start;
|
|
|
|
std::cerr << "Inserted instructions:\n\t";
|
2004-06-25 02:13:11 +02:00
|
|
|
Start->print(std::cerr, &MF.getTarget());
|
2004-02-15 01:46:41 +01:00
|
|
|
while (++Start != next(I));
|
|
|
|
}
|
|
|
|
dumpStack();
|
|
|
|
);
|
2003-01-13 02:01:59 +01:00
|
|
|
|
|
|
|
Changed = true;
|
|
|
|
}
|
|
|
|
|
|
|
|
assert(StackTop == 0 && "Stack not empty at end of basic block?");
|
|
|
|
return Changed;
|
|
|
|
}
|
|
|
|
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
// Efficient Lookup Table Support
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
|
2003-12-20 10:58:55 +01:00
|
|
|
namespace {
|
|
|
|
struct TableEntry {
|
|
|
|
unsigned from;
|
|
|
|
unsigned to;
|
|
|
|
bool operator<(const TableEntry &TE) const { return from < TE.from; }
|
|
|
|
bool operator<(unsigned V) const { return from < V; }
|
|
|
|
};
|
|
|
|
}
|
2003-01-13 02:01:59 +01:00
|
|
|
|
|
|
|
static bool TableIsSorted(const TableEntry *Table, unsigned NumEntries) {
|
|
|
|
for (unsigned i = 0; i != NumEntries-1; ++i)
|
|
|
|
if (!(Table[i] < Table[i+1])) return false;
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int Lookup(const TableEntry *Table, unsigned N, unsigned Opcode) {
|
|
|
|
const TableEntry *I = std::lower_bound(Table, Table+N, Opcode);
|
|
|
|
if (I != Table+N && I->from == Opcode)
|
|
|
|
return I->to;
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
#define ARRAY_SIZE(TABLE) \
|
|
|
|
(sizeof(TABLE)/sizeof(TABLE[0]))
|
|
|
|
|
|
|
|
#ifdef NDEBUG
|
|
|
|
#define ASSERT_SORTED(TABLE)
|
|
|
|
#else
|
|
|
|
#define ASSERT_SORTED(TABLE) \
|
|
|
|
{ static bool TABLE##Checked = false; \
|
|
|
|
if (!TABLE##Checked) \
|
|
|
|
assert(TableIsSorted(TABLE, ARRAY_SIZE(TABLE)) && \
|
|
|
|
"All lookup tables must be sorted for efficient access!"); \
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
|
|
|
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
// Helper Methods
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
|
|
|
|
// PopTable - Sorted map of instructions to their popping version. The first
|
|
|
|
// element is an instruction, the second is the version which pops.
|
|
|
|
//
|
|
|
|
static const TableEntry PopTable[] = {
|
2003-08-03 23:56:36 +02:00
|
|
|
{ X86::FADDrST0 , X86::FADDPrST0 },
|
|
|
|
|
|
|
|
{ X86::FDIVRrST0, X86::FDIVRPrST0 },
|
|
|
|
{ X86::FDIVrST0 , X86::FDIVPrST0 },
|
|
|
|
|
A big X86 instruction rename. The instructions are renamed to make
their names more decriptive. A name consists of the base name, a
default operand size followed by a character per operand with an
optional special size. For example:
ADD8rr -> add, 8-bit register, 8-bit register
IMUL16rmi -> imul, 16-bit register, 16-bit memory, 16-bit immediate
IMUL16rmi8 -> imul, 16-bit register, 16-bit memory, 8-bit immediate
MOVSX32rm16 -> movsx, 32-bit register, 16-bit memory
llvm-svn: 11995
2004-02-29 09:50:03 +01:00
|
|
|
{ X86::FIST16m , X86::FISTP16m },
|
|
|
|
{ X86::FIST32m , X86::FISTP32m },
|
2003-08-03 23:56:36 +02:00
|
|
|
|
|
|
|
{ X86::FMULrST0 , X86::FMULPrST0 },
|
|
|
|
|
A big X86 instruction rename. The instructions are renamed to make
their names more decriptive. A name consists of the base name, a
default operand size followed by a character per operand with an
optional special size. For example:
ADD8rr -> add, 8-bit register, 8-bit register
IMUL16rmi -> imul, 16-bit register, 16-bit memory, 16-bit immediate
IMUL16rmi8 -> imul, 16-bit register, 16-bit memory, 8-bit immediate
MOVSX32rm16 -> movsx, 32-bit register, 16-bit memory
llvm-svn: 11995
2004-02-29 09:50:03 +01:00
|
|
|
{ X86::FST32m , X86::FSTP32m },
|
|
|
|
{ X86::FST64m , X86::FSTP64m },
|
2003-01-13 02:01:59 +01:00
|
|
|
{ X86::FSTrr , X86::FSTPrr },
|
|
|
|
|
|
|
|
{ X86::FSUBRrST0, X86::FSUBRPrST0 },
|
2003-08-03 23:56:36 +02:00
|
|
|
{ X86::FSUBrST0 , X86::FSUBPrST0 },
|
2003-01-13 02:01:59 +01:00
|
|
|
|
2004-04-12 03:39:15 +02:00
|
|
|
{ X86::FUCOMIr , X86::FUCOMIPr },
|
|
|
|
|
2003-01-13 02:01:59 +01:00
|
|
|
{ X86::FUCOMPr , X86::FUCOMPPr },
|
2003-08-03 23:56:36 +02:00
|
|
|
{ X86::FUCOMr , X86::FUCOMPr },
|
2003-01-13 02:01:59 +01:00
|
|
|
};
|
|
|
|
|
|
|
|
/// popStackAfter - Pop the current value off of the top of the FP stack after
|
|
|
|
/// the specified instruction. This attempts to be sneaky and combine the pop
|
|
|
|
/// into the instruction itself if possible. The iterator is left pointing to
|
|
|
|
/// the last instruction, be it a new pop instruction inserted, or the old
|
|
|
|
/// instruction if it was modified in place.
|
|
|
|
///
|
|
|
|
void FPS::popStackAfter(MachineBasicBlock::iterator &I) {
|
|
|
|
ASSERT_SORTED(PopTable);
|
|
|
|
assert(StackTop > 0 && "Cannot pop empty stack!");
|
|
|
|
RegMap[Stack[--StackTop]] = ~0; // Update state
|
|
|
|
|
|
|
|
// Check to see if there is a popping version of this instruction...
|
2004-02-12 03:27:10 +01:00
|
|
|
int Opcode = Lookup(PopTable, ARRAY_SIZE(PopTable), I->getOpcode());
|
2003-01-13 02:01:59 +01:00
|
|
|
if (Opcode != -1) {
|
2004-02-12 03:27:10 +01:00
|
|
|
I->setOpcode(Opcode);
|
2003-01-13 02:01:59 +01:00
|
|
|
if (Opcode == X86::FUCOMPPr)
|
2004-02-12 03:27:10 +01:00
|
|
|
I->RemoveOperand(0);
|
2003-01-13 02:01:59 +01:00
|
|
|
|
|
|
|
} else { // Insert an explicit pop
|
Simplify code by using the more powerful BuildMI forms.
Implement a small optimization. In test/Regression/CodeGen/X86/select.ll,
we now generate this for foldSel3:
foldSel3:
mov %AL, BYTE PTR [%ESP + 4]
fld DWORD PTR [%ESP + 8]
fld DWORD PTR [%ESP + 12]
mov %EAX, DWORD PTR [%ESP + 16]
mov %ECX, DWORD PTR [%ESP + 20]
cmp %EAX, %ECX
fxch %ST(1)
fcmovae %ST(0), %ST(1)
*** fstp %ST(1)
ret
Instead of:
foldSel3:
mov %AL, BYTE PTR [%ESP + 4]
fld DWORD PTR [%ESP + 8]
fld DWORD PTR [%ESP + 12]
mov %EAX, DWORD PTR [%ESP + 16]
mov %ECX, DWORD PTR [%ESP + 20]
cmp %EAX, %ECX
fxch %ST(1)
fcmovae %ST(0), %ST(1)
*** fxch %ST(1)
*** fstp %ST(0)
ret
In practice, this only effects code size: performance should be basically
unaffected.
llvm-svn: 12588
2004-04-01 06:06:09 +02:00
|
|
|
I = BuildMI(*MBB, ++I, X86::FSTPrr, 1).addReg(X86::ST0);
|
2003-01-13 02:01:59 +01:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
Simplify code by using the more powerful BuildMI forms.
Implement a small optimization. In test/Regression/CodeGen/X86/select.ll,
we now generate this for foldSel3:
foldSel3:
mov %AL, BYTE PTR [%ESP + 4]
fld DWORD PTR [%ESP + 8]
fld DWORD PTR [%ESP + 12]
mov %EAX, DWORD PTR [%ESP + 16]
mov %ECX, DWORD PTR [%ESP + 20]
cmp %EAX, %ECX
fxch %ST(1)
fcmovae %ST(0), %ST(1)
*** fstp %ST(1)
ret
Instead of:
foldSel3:
mov %AL, BYTE PTR [%ESP + 4]
fld DWORD PTR [%ESP + 8]
fld DWORD PTR [%ESP + 12]
mov %EAX, DWORD PTR [%ESP + 16]
mov %ECX, DWORD PTR [%ESP + 20]
cmp %EAX, %ECX
fxch %ST(1)
fcmovae %ST(0), %ST(1)
*** fxch %ST(1)
*** fstp %ST(0)
ret
In practice, this only effects code size: performance should be basically
unaffected.
llvm-svn: 12588
2004-04-01 06:06:09 +02:00
|
|
|
/// freeStackSlotAfter - Free the specified register from the register stack, so
|
|
|
|
/// that it is no longer in a register. If the register is currently at the top
|
|
|
|
/// of the stack, we just pop the current instruction, otherwise we store the
|
|
|
|
/// current top-of-stack into the specified slot, then pop the top of stack.
|
|
|
|
void FPS::freeStackSlotAfter(MachineBasicBlock::iterator &I, unsigned FPRegNo) {
|
|
|
|
if (getStackEntry(0) == FPRegNo) { // already at the top of stack? easy.
|
|
|
|
popStackAfter(I);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
// Otherwise, store the top of stack into the dead slot, killing the operand
|
|
|
|
// without having to add in an explicit xchg then pop.
|
|
|
|
//
|
|
|
|
unsigned STReg = getSTReg(FPRegNo);
|
|
|
|
unsigned OldSlot = getSlot(FPRegNo);
|
|
|
|
unsigned TopReg = Stack[StackTop-1];
|
|
|
|
Stack[OldSlot] = TopReg;
|
|
|
|
RegMap[TopReg] = OldSlot;
|
|
|
|
RegMap[FPRegNo] = ~0;
|
|
|
|
Stack[--StackTop] = ~0;
|
|
|
|
I = BuildMI(*MBB, ++I, X86::FSTPrr, 1).addReg(STReg);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2003-01-13 02:01:59 +01:00
|
|
|
static unsigned getFPReg(const MachineOperand &MO) {
|
2004-02-10 21:31:28 +01:00
|
|
|
assert(MO.isRegister() && "Expected an FP register!");
|
2003-01-13 02:01:59 +01:00
|
|
|
unsigned Reg = MO.getReg();
|
|
|
|
assert(Reg >= X86::FP0 && Reg <= X86::FP6 && "Expected FP register!");
|
|
|
|
return Reg - X86::FP0;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
// Instruction transformation implementation
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
|
|
|
|
/// handleZeroArgFP - ST(0) = fld0 ST(0) = flds <mem>
|
2004-02-02 20:23:15 +01:00
|
|
|
///
|
2003-01-13 02:01:59 +01:00
|
|
|
void FPS::handleZeroArgFP(MachineBasicBlock::iterator &I) {
|
2004-02-12 03:27:10 +01:00
|
|
|
MachineInstr *MI = I;
|
2003-01-13 02:01:59 +01:00
|
|
|
unsigned DestReg = getFPReg(MI->getOperand(0));
|
|
|
|
MI->RemoveOperand(0); // Remove the explicit ST(0) operand
|
|
|
|
|
|
|
|
// Result gets pushed on the stack...
|
|
|
|
pushReg(DestReg);
|
|
|
|
}
|
|
|
|
|
2004-02-02 20:23:15 +01:00
|
|
|
/// handleOneArgFP - fst <mem>, ST(0)
|
|
|
|
///
|
2003-01-13 02:01:59 +01:00
|
|
|
void FPS::handleOneArgFP(MachineBasicBlock::iterator &I) {
|
2004-02-12 03:27:10 +01:00
|
|
|
MachineInstr *MI = I;
|
2004-02-03 08:27:34 +01:00
|
|
|
assert((MI->getNumOperands() == 5 || MI->getNumOperands() == 1) &&
|
|
|
|
"Can only handle fst* & ftst instructions!");
|
2003-01-13 02:01:59 +01:00
|
|
|
|
2004-02-02 20:23:15 +01:00
|
|
|
// Is this the last use of the source register?
|
2004-02-03 08:27:34 +01:00
|
|
|
unsigned Reg = getFPReg(MI->getOperand(MI->getNumOperands()-1));
|
2003-01-13 02:01:59 +01:00
|
|
|
bool KillsSrc = false;
|
|
|
|
for (LiveVariables::killed_iterator KI = LV->killed_begin(MI),
|
|
|
|
E = LV->killed_end(MI); KI != E; ++KI)
|
|
|
|
KillsSrc |= KI->second == X86::FP0+Reg;
|
|
|
|
|
A big X86 instruction rename. The instructions are renamed to make
their names more decriptive. A name consists of the base name, a
default operand size followed by a character per operand with an
optional special size. For example:
ADD8rr -> add, 8-bit register, 8-bit register
IMUL16rmi -> imul, 16-bit register, 16-bit memory, 16-bit immediate
IMUL16rmi8 -> imul, 16-bit register, 16-bit memory, 8-bit immediate
MOVSX32rm16 -> movsx, 32-bit register, 16-bit memory
llvm-svn: 11995
2004-02-29 09:50:03 +01:00
|
|
|
// FSTP80r and FISTP64r are strange because there are no non-popping versions.
|
2003-01-13 02:01:59 +01:00
|
|
|
// If we have one _and_ we don't want to pop the operand, duplicate the value
|
|
|
|
// on the stack instead of moving it. This ensure that popping the value is
|
|
|
|
// always ok.
|
|
|
|
//
|
A big X86 instruction rename. The instructions are renamed to make
their names more decriptive. A name consists of the base name, a
default operand size followed by a character per operand with an
optional special size. For example:
ADD8rr -> add, 8-bit register, 8-bit register
IMUL16rmi -> imul, 16-bit register, 16-bit memory, 16-bit immediate
IMUL16rmi8 -> imul, 16-bit register, 16-bit memory, 8-bit immediate
MOVSX32rm16 -> movsx, 32-bit register, 16-bit memory
llvm-svn: 11995
2004-02-29 09:50:03 +01:00
|
|
|
if ((MI->getOpcode() == X86::FSTP80m ||
|
|
|
|
MI->getOpcode() == X86::FISTP64m) && !KillsSrc) {
|
2003-01-13 02:01:59 +01:00
|
|
|
duplicateToTop(Reg, 7 /*temp register*/, I);
|
|
|
|
} else {
|
|
|
|
moveToTop(Reg, I); // Move to the top of the stack...
|
|
|
|
}
|
2004-02-03 08:27:34 +01:00
|
|
|
MI->RemoveOperand(MI->getNumOperands()-1); // Remove explicit ST(0) operand
|
2003-01-13 02:01:59 +01:00
|
|
|
|
A big X86 instruction rename. The instructions are renamed to make
their names more decriptive. A name consists of the base name, a
default operand size followed by a character per operand with an
optional special size. For example:
ADD8rr -> add, 8-bit register, 8-bit register
IMUL16rmi -> imul, 16-bit register, 16-bit memory, 16-bit immediate
IMUL16rmi8 -> imul, 16-bit register, 16-bit memory, 8-bit immediate
MOVSX32rm16 -> movsx, 32-bit register, 16-bit memory
llvm-svn: 11995
2004-02-29 09:50:03 +01:00
|
|
|
if (MI->getOpcode() == X86::FSTP80m || MI->getOpcode() == X86::FISTP64m) {
|
2003-01-13 02:01:59 +01:00
|
|
|
assert(StackTop > 0 && "Stack empty??");
|
|
|
|
--StackTop;
|
|
|
|
} else if (KillsSrc) { // Last use of operand?
|
|
|
|
popStackAfter(I);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2004-02-02 20:23:15 +01:00
|
|
|
|
2004-04-11 22:21:06 +02:00
|
|
|
/// handleOneArgFPRW: Handle instructions that read from the top of stack and
|
|
|
|
/// replace the value with a newly computed value. These instructions may have
|
|
|
|
/// non-fp operands after their FP operands.
|
|
|
|
///
|
|
|
|
/// Examples:
|
|
|
|
/// R1 = fchs R2
|
|
|
|
/// R1 = fadd R2, [mem]
|
2004-02-02 20:23:15 +01:00
|
|
|
///
|
|
|
|
void FPS::handleOneArgFPRW(MachineBasicBlock::iterator &I) {
|
2004-02-12 03:27:10 +01:00
|
|
|
MachineInstr *MI = I;
|
2004-04-11 22:21:06 +02:00
|
|
|
assert(MI->getNumOperands() >= 2 && "FPRW instructions must have 2 ops!!");
|
2004-02-02 20:23:15 +01:00
|
|
|
|
|
|
|
// Is this the last use of the source register?
|
|
|
|
unsigned Reg = getFPReg(MI->getOperand(1));
|
|
|
|
bool KillsSrc = false;
|
|
|
|
for (LiveVariables::killed_iterator KI = LV->killed_begin(MI),
|
|
|
|
E = LV->killed_end(MI); KI != E; ++KI)
|
|
|
|
KillsSrc |= KI->second == X86::FP0+Reg;
|
|
|
|
|
|
|
|
if (KillsSrc) {
|
|
|
|
// If this is the last use of the source register, just make sure it's on
|
|
|
|
// the top of the stack.
|
|
|
|
moveToTop(Reg, I);
|
|
|
|
assert(StackTop > 0 && "Stack cannot be empty!");
|
|
|
|
--StackTop;
|
|
|
|
pushReg(getFPReg(MI->getOperand(0)));
|
|
|
|
} else {
|
|
|
|
// If this is not the last use of the source register, _copy_ it to the top
|
|
|
|
// of the stack.
|
|
|
|
duplicateToTop(Reg, getFPReg(MI->getOperand(0)), I);
|
|
|
|
}
|
|
|
|
|
|
|
|
MI->RemoveOperand(1); // Drop the source operand.
|
|
|
|
MI->RemoveOperand(0); // Drop the destination operand.
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2003-01-13 02:01:59 +01:00
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
// Define tables of various ways to map pseudo instructions
|
|
|
|
//
|
|
|
|
|
|
|
|
// ForwardST0Table - Map: A = B op C into: ST(0) = ST(0) op ST(i)
|
|
|
|
static const TableEntry ForwardST0Table[] = {
|
2004-04-12 03:39:15 +02:00
|
|
|
{ X86::FpADD , X86::FADDST0r },
|
|
|
|
{ X86::FpDIV , X86::FDIVST0r },
|
|
|
|
{ X86::FpMUL , X86::FMULST0r },
|
|
|
|
{ X86::FpSUB , X86::FSUBST0r },
|
2003-01-13 02:01:59 +01:00
|
|
|
};
|
|
|
|
|
|
|
|
// ReverseST0Table - Map: A = B op C into: ST(0) = ST(i) op ST(0)
|
|
|
|
static const TableEntry ReverseST0Table[] = {
|
2004-04-12 03:39:15 +02:00
|
|
|
{ X86::FpADD , X86::FADDST0r }, // commutative
|
|
|
|
{ X86::FpDIV , X86::FDIVRST0r },
|
|
|
|
{ X86::FpMUL , X86::FMULST0r }, // commutative
|
|
|
|
{ X86::FpSUB , X86::FSUBRST0r },
|
2003-01-13 02:01:59 +01:00
|
|
|
};
|
|
|
|
|
|
|
|
// ForwardSTiTable - Map: A = B op C into: ST(i) = ST(0) op ST(i)
|
|
|
|
static const TableEntry ForwardSTiTable[] = {
|
2004-04-12 03:39:15 +02:00
|
|
|
{ X86::FpADD , X86::FADDrST0 }, // commutative
|
|
|
|
{ X86::FpDIV , X86::FDIVRrST0 },
|
|
|
|
{ X86::FpMUL , X86::FMULrST0 }, // commutative
|
|
|
|
{ X86::FpSUB , X86::FSUBRrST0 },
|
2003-01-13 02:01:59 +01:00
|
|
|
};
|
|
|
|
|
|
|
|
// ReverseSTiTable - Map: A = B op C into: ST(i) = ST(i) op ST(0)
|
|
|
|
static const TableEntry ReverseSTiTable[] = {
|
2004-04-12 03:39:15 +02:00
|
|
|
{ X86::FpADD , X86::FADDrST0 },
|
|
|
|
{ X86::FpDIV , X86::FDIVrST0 },
|
|
|
|
{ X86::FpMUL , X86::FMULrST0 },
|
|
|
|
{ X86::FpSUB , X86::FSUBrST0 },
|
2003-01-13 02:01:59 +01:00
|
|
|
};
|
|
|
|
|
|
|
|
|
|
|
|
/// handleTwoArgFP - Handle instructions like FADD and friends which are virtual
|
|
|
|
/// instructions which need to be simplified and possibly transformed.
|
|
|
|
///
|
|
|
|
/// Result: ST(0) = fsub ST(0), ST(i)
|
|
|
|
/// ST(i) = fsub ST(0), ST(i)
|
|
|
|
/// ST(0) = fsubr ST(0), ST(i)
|
|
|
|
/// ST(i) = fsubr ST(0), ST(i)
|
|
|
|
///
|
|
|
|
void FPS::handleTwoArgFP(MachineBasicBlock::iterator &I) {
|
|
|
|
ASSERT_SORTED(ForwardST0Table); ASSERT_SORTED(ReverseST0Table);
|
|
|
|
ASSERT_SORTED(ForwardSTiTable); ASSERT_SORTED(ReverseSTiTable);
|
2004-02-12 03:27:10 +01:00
|
|
|
MachineInstr *MI = I;
|
2003-01-13 02:01:59 +01:00
|
|
|
|
|
|
|
unsigned NumOperands = MI->getNumOperands();
|
2004-06-11 06:25:06 +02:00
|
|
|
assert(NumOperands == 3 && "Illegal TwoArgFP instruction!");
|
2003-01-13 02:01:59 +01:00
|
|
|
unsigned Dest = getFPReg(MI->getOperand(0));
|
|
|
|
unsigned Op0 = getFPReg(MI->getOperand(NumOperands-2));
|
|
|
|
unsigned Op1 = getFPReg(MI->getOperand(NumOperands-1));
|
|
|
|
bool KillsOp0 = false, KillsOp1 = false;
|
|
|
|
|
|
|
|
for (LiveVariables::killed_iterator KI = LV->killed_begin(MI),
|
|
|
|
E = LV->killed_end(MI); KI != E; ++KI) {
|
|
|
|
KillsOp0 |= (KI->second == X86::FP0+Op0);
|
|
|
|
KillsOp1 |= (KI->second == X86::FP0+Op1);
|
|
|
|
}
|
|
|
|
|
|
|
|
unsigned TOS = getStackEntry(0);
|
|
|
|
|
|
|
|
// One of our operands must be on the top of the stack. If neither is yet, we
|
|
|
|
// need to move one.
|
|
|
|
if (Op0 != TOS && Op1 != TOS) { // No operand at TOS?
|
|
|
|
// We can choose to move either operand to the top of the stack. If one of
|
|
|
|
// the operands is killed by this instruction, we want that one so that we
|
|
|
|
// can update right on top of the old version.
|
|
|
|
if (KillsOp0) {
|
|
|
|
moveToTop(Op0, I); // Move dead operand to TOS.
|
|
|
|
TOS = Op0;
|
|
|
|
} else if (KillsOp1) {
|
|
|
|
moveToTop(Op1, I);
|
|
|
|
TOS = Op1;
|
|
|
|
} else {
|
|
|
|
// All of the operands are live after this instruction executes, so we
|
|
|
|
// cannot update on top of any operand. Because of this, we must
|
|
|
|
// duplicate one of the stack elements to the top. It doesn't matter
|
|
|
|
// which one we pick.
|
|
|
|
//
|
|
|
|
duplicateToTop(Op0, Dest, I);
|
|
|
|
Op0 = TOS = Dest;
|
|
|
|
KillsOp0 = true;
|
|
|
|
}
|
2004-06-11 06:25:06 +02:00
|
|
|
} else if (!KillsOp0 && !KillsOp1) {
|
2003-01-13 02:01:59 +01:00
|
|
|
// If we DO have one of our operands at the top of the stack, but we don't
|
|
|
|
// have a dead operand, we must duplicate one of the operands to a new slot
|
|
|
|
// on the stack.
|
|
|
|
duplicateToTop(Op0, Dest, I);
|
|
|
|
Op0 = TOS = Dest;
|
|
|
|
KillsOp0 = true;
|
|
|
|
}
|
|
|
|
|
|
|
|
// Now we know that one of our operands is on the top of the stack, and at
|
|
|
|
// least one of our operands is killed by this instruction.
|
2004-06-11 06:25:06 +02:00
|
|
|
assert((TOS == Op0 || TOS == Op1) && (KillsOp0 || KillsOp1) &&
|
2003-01-13 02:01:59 +01:00
|
|
|
"Stack conditions not set up right!");
|
|
|
|
|
|
|
|
// We decide which form to use based on what is on the top of the stack, and
|
|
|
|
// which operand is killed by this instruction.
|
|
|
|
const TableEntry *InstTable;
|
|
|
|
bool isForward = TOS == Op0;
|
|
|
|
bool updateST0 = (TOS == Op0 && !KillsOp1) || (TOS == Op1 && !KillsOp0);
|
|
|
|
if (updateST0) {
|
|
|
|
if (isForward)
|
|
|
|
InstTable = ForwardST0Table;
|
|
|
|
else
|
|
|
|
InstTable = ReverseST0Table;
|
|
|
|
} else {
|
|
|
|
if (isForward)
|
|
|
|
InstTable = ForwardSTiTable;
|
|
|
|
else
|
|
|
|
InstTable = ReverseSTiTable;
|
|
|
|
}
|
|
|
|
|
|
|
|
int Opcode = Lookup(InstTable, ARRAY_SIZE(ForwardST0Table), MI->getOpcode());
|
|
|
|
assert(Opcode != -1 && "Unknown TwoArgFP pseudo instruction!");
|
|
|
|
|
|
|
|
// NotTOS - The register which is not on the top of stack...
|
|
|
|
unsigned NotTOS = (TOS == Op0) ? Op1 : Op0;
|
|
|
|
|
|
|
|
// Replace the old instruction with a new instruction
|
2004-04-01 00:02:36 +02:00
|
|
|
MBB->remove(I++);
|
Simplify code by using the more powerful BuildMI forms.
Implement a small optimization. In test/Regression/CodeGen/X86/select.ll,
we now generate this for foldSel3:
foldSel3:
mov %AL, BYTE PTR [%ESP + 4]
fld DWORD PTR [%ESP + 8]
fld DWORD PTR [%ESP + 12]
mov %EAX, DWORD PTR [%ESP + 16]
mov %ECX, DWORD PTR [%ESP + 20]
cmp %EAX, %ECX
fxch %ST(1)
fcmovae %ST(0), %ST(1)
*** fstp %ST(1)
ret
Instead of:
foldSel3:
mov %AL, BYTE PTR [%ESP + 4]
fld DWORD PTR [%ESP + 8]
fld DWORD PTR [%ESP + 12]
mov %EAX, DWORD PTR [%ESP + 16]
mov %ECX, DWORD PTR [%ESP + 20]
cmp %EAX, %ECX
fxch %ST(1)
fcmovae %ST(0), %ST(1)
*** fxch %ST(1)
*** fstp %ST(0)
ret
In practice, this only effects code size: performance should be basically
unaffected.
llvm-svn: 12588
2004-04-01 06:06:09 +02:00
|
|
|
I = BuildMI(*MBB, I, Opcode, 1).addReg(getSTReg(NotTOS));
|
2003-01-13 02:01:59 +01:00
|
|
|
|
|
|
|
// If both operands are killed, pop one off of the stack in addition to
|
|
|
|
// overwriting the other one.
|
|
|
|
if (KillsOp0 && KillsOp1 && Op0 != Op1) {
|
|
|
|
assert(!updateST0 && "Should have updated other operand!");
|
|
|
|
popStackAfter(I); // Pop the top of stack
|
|
|
|
}
|
|
|
|
|
|
|
|
// Update stack information so that we know the destination register is now on
|
|
|
|
// the stack.
|
2004-06-11 06:25:06 +02:00
|
|
|
unsigned UpdatedSlot = getSlot(updateST0 ? TOS : NotTOS);
|
|
|
|
assert(UpdatedSlot < StackTop && Dest < 7);
|
|
|
|
Stack[UpdatedSlot] = Dest;
|
|
|
|
RegMap[Dest] = UpdatedSlot;
|
|
|
|
delete MI; // Remove the old instruction
|
|
|
|
}
|
|
|
|
|
2004-06-11 06:49:02 +02:00
|
|
|
/// handleCompareFP - Handle FUCOM and FUCOMI instructions, which have two FP
|
2004-06-11 06:25:06 +02:00
|
|
|
/// register arguments and no explicit destinations.
|
|
|
|
///
|
|
|
|
void FPS::handleCompareFP(MachineBasicBlock::iterator &I) {
|
|
|
|
ASSERT_SORTED(ForwardST0Table); ASSERT_SORTED(ReverseST0Table);
|
|
|
|
ASSERT_SORTED(ForwardSTiTable); ASSERT_SORTED(ReverseSTiTable);
|
|
|
|
MachineInstr *MI = I;
|
|
|
|
|
|
|
|
unsigned NumOperands = MI->getNumOperands();
|
2004-06-11 06:49:02 +02:00
|
|
|
assert(NumOperands == 2 && "Illegal FUCOM* instruction!");
|
2004-06-11 06:25:06 +02:00
|
|
|
unsigned Op0 = getFPReg(MI->getOperand(NumOperands-2));
|
|
|
|
unsigned Op1 = getFPReg(MI->getOperand(NumOperands-1));
|
|
|
|
bool KillsOp0 = false, KillsOp1 = false;
|
|
|
|
|
|
|
|
for (LiveVariables::killed_iterator KI = LV->killed_begin(MI),
|
|
|
|
E = LV->killed_end(MI); KI != E; ++KI) {
|
|
|
|
KillsOp0 |= (KI->second == X86::FP0+Op0);
|
|
|
|
KillsOp1 |= (KI->second == X86::FP0+Op1);
|
2003-01-13 02:01:59 +01:00
|
|
|
}
|
2004-06-11 06:25:06 +02:00
|
|
|
|
|
|
|
// Make sure the first operand is on the top of stack, the other one can be
|
|
|
|
// anywhere.
|
|
|
|
moveToTop(Op0, I);
|
|
|
|
|
2004-06-11 07:22:44 +02:00
|
|
|
MI->getOperand(0).setReg(getSTReg(Op1));
|
|
|
|
MI->RemoveOperand(1);
|
|
|
|
|
2004-06-11 06:25:06 +02:00
|
|
|
// If any of the operands are killed by this instruction, free them.
|
|
|
|
if (KillsOp0) freeStackSlotAfter(I, Op0);
|
|
|
|
if (KillsOp1 && Op0 != Op1) freeStackSlotAfter(I, Op1);
|
2003-01-13 02:01:59 +01:00
|
|
|
}
|
|
|
|
|
2004-04-01 00:02:36 +02:00
|
|
|
/// handleCondMovFP - Handle two address conditional move instructions. These
|
|
|
|
/// instructions move a st(i) register to st(0) iff a condition is true. These
|
|
|
|
/// instructions require that the first operand is at the top of the stack, but
|
|
|
|
/// otherwise don't modify the stack at all.
|
|
|
|
void FPS::handleCondMovFP(MachineBasicBlock::iterator &I) {
|
|
|
|
MachineInstr *MI = I;
|
|
|
|
|
|
|
|
unsigned Op0 = getFPReg(MI->getOperand(0));
|
|
|
|
unsigned Op1 = getFPReg(MI->getOperand(1));
|
|
|
|
|
|
|
|
// The first operand *must* be on the top of the stack.
|
|
|
|
moveToTop(Op0, I);
|
|
|
|
|
|
|
|
// Change the second operand to the stack register that the operand is in.
|
|
|
|
MI->RemoveOperand(0);
|
|
|
|
MI->getOperand(0).setReg(getSTReg(Op1));
|
|
|
|
|
|
|
|
// If we kill the second operand, make sure to pop it from the stack.
|
2004-04-14 04:42:32 +02:00
|
|
|
if (Op0 != Op1)
|
|
|
|
for (LiveVariables::killed_iterator KI = LV->killed_begin(MI),
|
|
|
|
E = LV->killed_end(MI); KI != E; ++KI)
|
|
|
|
if (KI->second == X86::FP0+Op1) {
|
|
|
|
// Get this value off of the register stack.
|
|
|
|
freeStackSlotAfter(I, Op1);
|
|
|
|
break;
|
|
|
|
}
|
2004-04-01 00:02:36 +02:00
|
|
|
}
|
|
|
|
|
2003-01-13 02:01:59 +01:00
|
|
|
|
|
|
|
/// handleSpecialFP - Handle special instructions which behave unlike other
|
2003-10-10 19:57:28 +02:00
|
|
|
/// floating point instructions. This is primarily intended for use by pseudo
|
2003-01-13 02:01:59 +01:00
|
|
|
/// instructions.
|
|
|
|
///
|
|
|
|
void FPS::handleSpecialFP(MachineBasicBlock::iterator &I) {
|
2004-02-12 03:27:10 +01:00
|
|
|
MachineInstr *MI = I;
|
2003-01-13 02:01:59 +01:00
|
|
|
switch (MI->getOpcode()) {
|
|
|
|
default: assert(0 && "Unknown SpecialFP instruction!");
|
|
|
|
case X86::FpGETRESULT: // Appears immediately after a call returning FP type!
|
|
|
|
assert(StackTop == 0 && "Stack should be empty after a call!");
|
|
|
|
pushReg(getFPReg(MI->getOperand(0)));
|
|
|
|
break;
|
|
|
|
case X86::FpSETRESULT:
|
|
|
|
assert(StackTop == 1 && "Stack should have one element on it to return!");
|
|
|
|
--StackTop; // "Forget" we have something on the top of stack!
|
|
|
|
break;
|
|
|
|
case X86::FpMOV: {
|
|
|
|
unsigned SrcReg = getFPReg(MI->getOperand(1));
|
|
|
|
unsigned DestReg = getFPReg(MI->getOperand(0));
|
|
|
|
bool KillsSrc = false;
|
|
|
|
for (LiveVariables::killed_iterator KI = LV->killed_begin(MI),
|
|
|
|
E = LV->killed_end(MI); KI != E; ++KI)
|
|
|
|
KillsSrc |= KI->second == X86::FP0+SrcReg;
|
|
|
|
|
|
|
|
if (KillsSrc) {
|
|
|
|
// If the input operand is killed, we can just change the owner of the
|
|
|
|
// incoming stack slot into the result.
|
|
|
|
unsigned Slot = getSlot(SrcReg);
|
|
|
|
assert(Slot < 7 && DestReg < 7 && "FpMOV operands invalid!");
|
|
|
|
Stack[Slot] = DestReg;
|
|
|
|
RegMap[DestReg] = Slot;
|
|
|
|
|
|
|
|
} else {
|
|
|
|
// For FMOV we just duplicate the specified value to a new stack slot.
|
|
|
|
// This could be made better, but would require substantial changes.
|
|
|
|
duplicateToTop(SrcReg, DestReg, I);
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2004-02-12 03:27:10 +01:00
|
|
|
I = MBB->erase(I); // Remove the pseudo instruction
|
|
|
|
--I;
|
2003-01-13 02:01:59 +01:00
|
|
|
}
|