llvm-6502/lib/Target/Sparc/SparcFrameLowering.cpp
Quentin Colombet 2f7322b348 [ShrinkWrap] Add (a simplified version) of shrink-wrapping.
This patch introduces a new pass that computes the safe point to insert the
prologue and epilogue of the function.
The interest is to find safe points that are cheaper than the entry and exits
blocks.

As an example and to avoid regressions to be introduce, this patch also
implements the required bits to enable the shrink-wrapping pass for AArch64.


** Context **

Currently we insert the prologue and epilogue of the method/function in the
entry and exits blocks. Although this is correct, we can do a better job when
those are not immediately required and insert them at less frequently executed
places.
The job of the shrink-wrapping pass is to identify such places.


** Motivating example **

Let us consider the following function that perform a call only in one branch of
a if:
define i32 @f(i32 %a, i32 %b)  {
 %tmp = alloca i32, align 4
 %tmp2 = icmp slt i32 %a, %b
 br i1 %tmp2, label %true, label %false

true:
 store i32 %a, i32* %tmp, align 4
 %tmp4 = call i32 @doSomething(i32 0, i32* %tmp)
 br label %false

false:
 %tmp.0 = phi i32 [ %tmp4, %true ], [ %a, %0 ]
 ret i32 %tmp.0
}

On AArch64 this code generates (removing the cfi directives to ease
readabilities):
_f:                                     ; @f
; BB#0:
  stp x29, x30, [sp, #-16]!
  mov  x29, sp
  sub sp, sp, #16             ; =16
  cmp  w0, w1
  b.ge  LBB0_2
; BB#1:                                 ; %true
  stur  w0, [x29, #-4]
  sub x1, x29, #4             ; =4
  mov  w0, wzr
  bl  _doSomething
LBB0_2:                                 ; %false
  mov  sp, x29
  ldp x29, x30, [sp], #16
  ret

With shrink-wrapping we could generate:
_f:                                     ; @f
; BB#0:
  cmp  w0, w1
  b.ge  LBB0_2
; BB#1:                                 ; %true
  stp x29, x30, [sp, #-16]!
  mov  x29, sp
  sub sp, sp, #16             ; =16
  stur  w0, [x29, #-4]
  sub x1, x29, #4             ; =4
  mov  w0, wzr
  bl  _doSomething
  add sp, x29, #16            ; =16
  ldp x29, x30, [sp], #16
LBB0_2:                                 ; %false
  ret

Therefore, we would pay the overhead of setting up/destroying the frame only if
we actually do the call.


** Proposed Solution **

This patch introduces a new machine pass that perform the shrink-wrapping
analysis (See the comments at the beginning of ShrinkWrap.cpp for more details).
It then stores the safe save and restore point into the MachineFrameInfo
attached to the MachineFunction.
This information is then used by the PrologEpilogInserter (PEI) to place the
related code at the right place. This pass runs right before the PEI.

Unlike the original paper of Chow from PLDI’88, this implementation of
shrink-wrapping does not use expensive data-flow analysis and does not need hack
to properly avoid frequently executed point. Instead, it relies on dominance and
loop properties.

The pass is off by default and each target can opt-in by setting the
EnableShrinkWrap boolean to true in their derived class of TargetPassConfig.
This setting can also be overwritten on the command line by using
-enable-shrink-wrap.

Before you try out the pass for your target, make sure you properly fix your
emitProlog/emitEpilog/adjustForXXX method to cope with basic blocks that are not
necessarily the entry block.


** Design Decisions **

1. ShrinkWrap is its own pass right now. It could frankly be merged into PEI but
for debugging and clarity I thought it was best to have its own file.
2. Right now, we only support one save point and one restore point. At some
point we can expand this to several save point and restore point, the impacted
component would then be:
- The pass itself: New algorithm needed.
- MachineFrameInfo: Hold a list or set of Save/Restore point instead of one
  pointer.
- PEI: Should loop over the save point and restore point.
Anyhow, at least for this first iteration, I do not believe this is interesting
to support the complex cases. We should revisit that when we motivating
examples.

Differential Revision: http://reviews.llvm.org/D9210

<rdar://problem/3201744>


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@236507 91177308-0d34-0410-b5e6-96231b3b80d8
2015-05-05 17:38:16 +00:00

261 lines
8.8 KiB
C++

//===-- SparcFrameLowering.cpp - Sparc Frame 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 Sparc implementation of TargetFrameLowering class.
//
//===----------------------------------------------------------------------===//
#include "SparcFrameLowering.h"
#include "SparcInstrInfo.h"
#include "SparcMachineFunctionInfo.h"
#include "SparcSubtarget.h"
#include "llvm/CodeGen/MachineFrameInfo.h"
#include "llvm/CodeGen/MachineFunction.h"
#include "llvm/CodeGen/MachineInstrBuilder.h"
#include "llvm/CodeGen/MachineModuleInfo.h"
#include "llvm/CodeGen/MachineRegisterInfo.h"
#include "llvm/IR/DataLayout.h"
#include "llvm/IR/Function.h"
#include "llvm/Support/CommandLine.h"
#include "llvm/Target/TargetOptions.h"
using namespace llvm;
static cl::opt<bool>
DisableLeafProc("disable-sparc-leaf-proc",
cl::init(false),
cl::desc("Disable Sparc leaf procedure optimization."),
cl::Hidden);
SparcFrameLowering::SparcFrameLowering(const SparcSubtarget &ST)
: TargetFrameLowering(TargetFrameLowering::StackGrowsDown,
ST.is64Bit() ? 16 : 8, 0, ST.is64Bit() ? 16 : 8) {}
void SparcFrameLowering::emitSPAdjustment(MachineFunction &MF,
MachineBasicBlock &MBB,
MachineBasicBlock::iterator MBBI,
int NumBytes,
unsigned ADDrr,
unsigned ADDri) const {
DebugLoc dl = (MBBI != MBB.end()) ? MBBI->getDebugLoc() : DebugLoc();
const SparcInstrInfo &TII =
*static_cast<const SparcInstrInfo *>(MF.getSubtarget().getInstrInfo());
if (NumBytes >= -4096 && NumBytes < 4096) {
BuildMI(MBB, MBBI, dl, TII.get(ADDri), SP::O6)
.addReg(SP::O6).addImm(NumBytes);
return;
}
// Emit this the hard way. This clobbers G1 which we always know is
// available here.
if (NumBytes >= 0) {
// Emit nonnegative numbers with sethi + or.
// sethi %hi(NumBytes), %g1
// or %g1, %lo(NumBytes), %g1
// add %sp, %g1, %sp
BuildMI(MBB, MBBI, dl, TII.get(SP::SETHIi), SP::G1)
.addImm(HI22(NumBytes));
BuildMI(MBB, MBBI, dl, TII.get(SP::ORri), SP::G1)
.addReg(SP::G1).addImm(LO10(NumBytes));
BuildMI(MBB, MBBI, dl, TII.get(ADDrr), SP::O6)
.addReg(SP::O6).addReg(SP::G1);
return ;
}
// Emit negative numbers with sethi + xor.
// sethi %hix(NumBytes), %g1
// xor %g1, %lox(NumBytes), %g1
// add %sp, %g1, %sp
BuildMI(MBB, MBBI, dl, TII.get(SP::SETHIi), SP::G1)
.addImm(HIX22(NumBytes));
BuildMI(MBB, MBBI, dl, TII.get(SP::XORri), SP::G1)
.addReg(SP::G1).addImm(LOX10(NumBytes));
BuildMI(MBB, MBBI, dl, TII.get(ADDrr), SP::O6)
.addReg(SP::O6).addReg(SP::G1);
}
void SparcFrameLowering::emitPrologue(MachineFunction &MF,
MachineBasicBlock &MBB) const {
SparcMachineFunctionInfo *FuncInfo = MF.getInfo<SparcMachineFunctionInfo>();
assert(&MF.front() == &MBB && "Shrink-wrapping not yet supported");
MachineFrameInfo *MFI = MF.getFrameInfo();
const SparcInstrInfo &TII =
*static_cast<const SparcInstrInfo *>(MF.getSubtarget().getInstrInfo());
MachineBasicBlock::iterator MBBI = MBB.begin();
DebugLoc dl = MBBI != MBB.end() ? MBBI->getDebugLoc() : DebugLoc();
// Get the number of bytes to allocate from the FrameInfo
int NumBytes = (int) MFI->getStackSize();
unsigned SAVEri = SP::SAVEri;
unsigned SAVErr = SP::SAVErr;
if (FuncInfo->isLeafProc()) {
if (NumBytes == 0)
return;
SAVEri = SP::ADDri;
SAVErr = SP::ADDrr;
}
NumBytes = -MF.getSubtarget<SparcSubtarget>().getAdjustedFrameSize(NumBytes);
emitSPAdjustment(MF, MBB, MBBI, NumBytes, SAVErr, SAVEri);
MachineModuleInfo &MMI = MF.getMMI();
const MCRegisterInfo *MRI = MMI.getContext().getRegisterInfo();
unsigned regFP = MRI->getDwarfRegNum(SP::I6, true);
// Emit ".cfi_def_cfa_register 30".
unsigned CFIIndex =
MMI.addFrameInst(MCCFIInstruction::createDefCfaRegister(nullptr, regFP));
BuildMI(MBB, MBBI, dl, TII.get(TargetOpcode::CFI_INSTRUCTION))
.addCFIIndex(CFIIndex);
// Emit ".cfi_window_save".
CFIIndex = MMI.addFrameInst(MCCFIInstruction::createWindowSave(nullptr));
BuildMI(MBB, MBBI, dl, TII.get(TargetOpcode::CFI_INSTRUCTION))
.addCFIIndex(CFIIndex);
unsigned regInRA = MRI->getDwarfRegNum(SP::I7, true);
unsigned regOutRA = MRI->getDwarfRegNum(SP::O7, true);
// Emit ".cfi_register 15, 31".
CFIIndex = MMI.addFrameInst(
MCCFIInstruction::createRegister(nullptr, regOutRA, regInRA));
BuildMI(MBB, MBBI, dl, TII.get(TargetOpcode::CFI_INSTRUCTION))
.addCFIIndex(CFIIndex);
}
void SparcFrameLowering::
eliminateCallFramePseudoInstr(MachineFunction &MF, MachineBasicBlock &MBB,
MachineBasicBlock::iterator I) const {
if (!hasReservedCallFrame(MF)) {
MachineInstr &MI = *I;
int Size = MI.getOperand(0).getImm();
if (MI.getOpcode() == SP::ADJCALLSTACKDOWN)
Size = -Size;
if (Size)
emitSPAdjustment(MF, MBB, I, Size, SP::ADDrr, SP::ADDri);
}
MBB.erase(I);
}
void SparcFrameLowering::emitEpilogue(MachineFunction &MF,
MachineBasicBlock &MBB) const {
SparcMachineFunctionInfo *FuncInfo = MF.getInfo<SparcMachineFunctionInfo>();
MachineBasicBlock::iterator MBBI = MBB.getLastNonDebugInstr();
const SparcInstrInfo &TII =
*static_cast<const SparcInstrInfo *>(MF.getSubtarget().getInstrInfo());
DebugLoc dl = MBBI->getDebugLoc();
assert(MBBI->getOpcode() == SP::RETL &&
"Can only put epilog before 'retl' instruction!");
if (!FuncInfo->isLeafProc()) {
BuildMI(MBB, MBBI, dl, TII.get(SP::RESTORErr), SP::G0).addReg(SP::G0)
.addReg(SP::G0);
return;
}
MachineFrameInfo *MFI = MF.getFrameInfo();
int NumBytes = (int) MFI->getStackSize();
if (NumBytes == 0)
return;
NumBytes = MF.getSubtarget<SparcSubtarget>().getAdjustedFrameSize(NumBytes);
emitSPAdjustment(MF, MBB, MBBI, NumBytes, SP::ADDrr, SP::ADDri);
}
bool SparcFrameLowering::hasReservedCallFrame(const MachineFunction &MF) const {
// Reserve call frame if there are no variable sized objects on the stack.
return !MF.getFrameInfo()->hasVarSizedObjects();
}
// hasFP - Return true if the specified function should have a dedicated frame
// pointer register. This is true if the function has variable sized allocas or
// if frame pointer elimination is disabled.
bool SparcFrameLowering::hasFP(const MachineFunction &MF) const {
const MachineFrameInfo *MFI = MF.getFrameInfo();
return MF.getTarget().Options.DisableFramePointerElim(MF) ||
MFI->hasVarSizedObjects() || MFI->isFrameAddressTaken();
}
static bool LLVM_ATTRIBUTE_UNUSED verifyLeafProcRegUse(MachineRegisterInfo *MRI)
{
for (unsigned reg = SP::I0; reg <= SP::I7; ++reg)
if (MRI->isPhysRegUsed(reg))
return false;
for (unsigned reg = SP::L0; reg <= SP::L7; ++reg)
if (MRI->isPhysRegUsed(reg))
return false;
return true;
}
bool SparcFrameLowering::isLeafProc(MachineFunction &MF) const
{
MachineRegisterInfo &MRI = MF.getRegInfo();
MachineFrameInfo *MFI = MF.getFrameInfo();
return !(MFI->hasCalls() // has calls
|| MRI.isPhysRegUsed(SP::L0) // Too many registers needed
|| MRI.isPhysRegUsed(SP::O6) // %SP is used
|| hasFP(MF)); // need %FP
}
void SparcFrameLowering::remapRegsForLeafProc(MachineFunction &MF) const {
MachineRegisterInfo &MRI = MF.getRegInfo();
// Remap %i[0-7] to %o[0-7].
for (unsigned reg = SP::I0; reg <= SP::I7; ++reg) {
if (!MRI.isPhysRegUsed(reg))
continue;
unsigned mapped_reg = (reg - SP::I0 + SP::O0);
assert(!MRI.isPhysRegUsed(mapped_reg));
// Replace I register with O register.
MRI.replaceRegWith(reg, mapped_reg);
// Mark the reg unused.
MRI.setPhysRegUnused(reg);
}
// Rewrite MBB's Live-ins.
for (MachineFunction::iterator MBB = MF.begin(), E = MF.end();
MBB != E; ++MBB) {
for (unsigned reg = SP::I0; reg <= SP::I7; ++reg) {
if (!MBB->isLiveIn(reg))
continue;
MBB->removeLiveIn(reg);
MBB->addLiveIn(reg - SP::I0 + SP::O0);
}
}
assert(verifyLeafProcRegUse(&MRI));
#ifdef XDEBUG
MF.verify(0, "After LeafProc Remapping");
#endif
}
void SparcFrameLowering::processFunctionBeforeCalleeSavedScan
(MachineFunction &MF, RegScavenger *RS) const {
if (!DisableLeafProc && isLeafProc(MF)) {
SparcMachineFunctionInfo *MFI = MF.getInfo<SparcMachineFunctionInfo>();
MFI->setLeafProc(true);
remapRegsForLeafProc(MF);
}
}