1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2025-02-01 05:01:59 +01:00
llvm-mirror/include/llvm/CodeGen/FunctionLoweringInfo.h

292 lines
10 KiB
C
Raw Normal View History

//===- FunctionLoweringInfo.h - Lower functions from LLVM IR ---*- C++ -*--===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
//
// This implements routines for translating functions from LLVM IR into
// Machine IR.
//
//===----------------------------------------------------------------------===//
#ifndef LLVM_CODEGEN_FUNCTIONLOWERINGINFO_H
#define LLVM_CODEGEN_FUNCTIONLOWERINGINFO_H
#include "llvm/ADT/APInt.h"
[SelectionDAGBuilder] Add restrictions to EmitFuncArgumentDbgValue Summary: This patch fixes PR40587. When a dbg.value instrinsic is emitted to the DAG by using EmitFuncArgumentDbgValue the resulting DBG_VALUE is hoisted to the beginning of the entry block. I think the idea is to be able to locate a formal argument already from the start of the function. However, EmitFuncArgumentDbgValue only checked that the value that was used to describe a variable was originating from a function parameter, not that the variable itself actually was an argument to the function. So when for example assigning a local variable "local" the value from an argument "a", the assocated DBG_VALUE instruction would be hoisted to the beginning of the function, even if the scope for "local" started somewhere else (or if "local" was mapped to other values earlier in the function). This patch adds some logic to EmitFuncArgumentDbgValue to check that the variable being described actually is an argument to the function. And that the dbg.value being lowered already is in the entry block. Otherwise we bail out, and the dbg.value will be handled as an ordinary dbg.value (not as a "FuncArgumentDbgValue"). A tricky situation is when both the variable and the value is related to function arguments, but not neccessarily the same argument. We make sure that we do not describe the same argument more than once as a "FuncArgumentDbgValue". This solution works as long as opt has injected a "first" dbg.value that corresponds to the formal argument at the function entry. Reviewers: jmorse, aprantl Subscribers: jyknight, hiraditya, fedor.sergeev, dstenb, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D57702 llvm-svn: 353735
2019-02-11 19:23:30 +00:00
#include "llvm/ADT/BitVector.h"
#include "llvm/ADT/DenseMap.h"
#include "llvm/ADT/IndexedMap.h"
#include "llvm/ADT/Optional.h"
#include "llvm/ADT/SmallPtrSet.h"
2010-04-29 01:39:13 +00:00
#include "llvm/ADT/SmallVector.h"
#include "llvm/CodeGen/ISDOpcodes.h"
#include "llvm/CodeGen/MachineBasicBlock.h"
#include "llvm/CodeGen/TargetRegisterInfo.h"
#include "llvm/IR/Instructions.h"
#include "llvm/IR/Type.h"
#include "llvm/IR/Value.h"
#include "llvm/Support/KnownBits.h"
#include <cassert>
#include <utility>
#include <vector>
namespace llvm {
class Argument;
class BasicBlock;
class BranchProbabilityInfo;
class LegacyDivergenceAnalysis;
class Function;
class Instruction;
class MachineFunction;
class MachineInstr;
class MachineRegisterInfo;
class MVT;
class SelectionDAG;
class TargetLowering;
//===--------------------------------------------------------------------===//
/// FunctionLoweringInfo - This contains information that is global to a
/// function that is used when lowering a region of the function.
///
class FunctionLoweringInfo {
public:
const Function *Fn;
MachineFunction *MF;
const TargetLowering *TLI;
MachineRegisterInfo *RegInfo;
BranchProbabilityInfo *BPI;
const LegacyDivergenceAnalysis *DA;
/// CanLowerReturn - true iff the function's return value can be lowered to
/// registers.
bool CanLowerReturn;
/// True if part of the CSRs will be handled via explicit copies.
bool SplitCSR;
/// DemoteRegister - if CanLowerReturn is false, DemoteRegister is a vreg
/// allocated to hold a pointer to the hidden sret parameter.
unsigned DemoteRegister;
/// MBBMap - A mapping from LLVM basic blocks to their machine code entry.
DenseMap<const BasicBlock*, MachineBasicBlock *> MBBMap;
/// ValueMap - Since we emit code for the function a basic block at a time,
/// we must remember which virtual registers hold the values for
/// cross-basic-block values.
DenseMap<const Value *, unsigned> ValueMap;
/// VirtReg2Value map is needed by the Divergence Analysis driven
/// instruction selection. It is reverted ValueMap. It is computed
/// in lazy style - on demand. It is used to get the Value corresponding
/// to the live in virtual register and is called from the
/// TargetLowerinInfo::isSDNodeSourceOfDivergence.
DenseMap<unsigned, const Value*> VirtReg2Value;
/// This method is called from TargetLowerinInfo::isSDNodeSourceOfDivergence
/// to get the Value corresponding to the live-in virtual register.
const Value * getValueFromVirtualReg(unsigned Vreg);
/// Track virtual registers created for exception pointers.
DenseMap<const Value *, unsigned> CatchPadExceptionPointers;
/// Keep track of frame indices allocated for statepoints as they could be
/// used across basic block boundaries. This struct is more complex than a
/// simple map because the stateopint lowering code de-duplicates gc pointers
/// based on their SDValue (so %p and (bitcast %p to T) will get the same
/// slot), and we track that here.
struct StatepointSpillMap {
using SlotMapTy = DenseMap<const Value *, Optional<int>>;
/// Maps uniqued llvm IR values to the slots they were spilled in. If a
/// value is mapped to None it means we visited the value but didn't spill
/// it (because it was a constant, for instance).
SlotMapTy SlotMap;
/// Maps llvm IR values to the values they were de-duplicated to.
DenseMap<const Value *, const Value *> DuplicateMap;
SlotMapTy::const_iterator find(const Value *V) const {
auto DuplIt = DuplicateMap.find(V);
if (DuplIt != DuplicateMap.end())
V = DuplIt->second;
return SlotMap.find(V);
}
SlotMapTy::const_iterator end() const { return SlotMap.end(); }
};
/// Maps gc.statepoint instructions to their corresponding StatepointSpillMap
/// instances.
DenseMap<const Instruction *, StatepointSpillMap> StatepointSpillMaps;
/// StaticAllocaMap - Keep track of frame indices for fixed sized allocas in
/// the entry block. This allows the allocas to be efficiently referenced
/// anywhere in the function.
DenseMap<const AllocaInst*, int> StaticAllocaMap;
/// ByValArgFrameIndexMap - Keep track of frame indices for byval arguments.
DenseMap<const Argument*, int> ByValArgFrameIndexMap;
2010-04-29 06:58:53 +00:00
/// ArgDbgValues - A list of DBG_VALUE instructions created during isel for
/// function arguments that are inserted after scheduling is completed.
SmallVector<MachineInstr*, 8> ArgDbgValues;
[SelectionDAGBuilder] Add restrictions to EmitFuncArgumentDbgValue Summary: This patch fixes PR40587. When a dbg.value instrinsic is emitted to the DAG by using EmitFuncArgumentDbgValue the resulting DBG_VALUE is hoisted to the beginning of the entry block. I think the idea is to be able to locate a formal argument already from the start of the function. However, EmitFuncArgumentDbgValue only checked that the value that was used to describe a variable was originating from a function parameter, not that the variable itself actually was an argument to the function. So when for example assigning a local variable "local" the value from an argument "a", the assocated DBG_VALUE instruction would be hoisted to the beginning of the function, even if the scope for "local" started somewhere else (or if "local" was mapped to other values earlier in the function). This patch adds some logic to EmitFuncArgumentDbgValue to check that the variable being described actually is an argument to the function. And that the dbg.value being lowered already is in the entry block. Otherwise we bail out, and the dbg.value will be handled as an ordinary dbg.value (not as a "FuncArgumentDbgValue"). A tricky situation is when both the variable and the value is related to function arguments, but not neccessarily the same argument. We make sure that we do not describe the same argument more than once as a "FuncArgumentDbgValue". This solution works as long as opt has injected a "first" dbg.value that corresponds to the formal argument at the function entry. Reviewers: jmorse, aprantl Subscribers: jyknight, hiraditya, fedor.sergeev, dstenb, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D57702 llvm-svn: 353735
2019-02-11 19:23:30 +00:00
/// Bitvector with a bit set if corresponding argument is described in
/// ArgDbgValues. Using arg numbers according to Argument numbering.
BitVector DescribedArgs;
/// RegFixups - Registers which need to be replaced after isel is done.
DenseMap<unsigned, unsigned> RegFixups;
[FastISel] Sink local value materializations to first use Summary: Local values are constants, global addresses, and stack addresses that can't be folded into the instruction that uses them. For example, when storing the address of a global variable into memory, we need to materialize that address into a register. FastISel doesn't want to materialize any given local value more than once, so it generates all local value materialization code at EmitStartPt, which always dominates the current insertion point. This allows it to maintain a map of local value registers, and it knows that the local value area will always dominate the current insertion point. The downside is that local value instructions are always emitted without a source location. This is done to prevent jumpy line tables, but it means that the local value area will be considered part of the previous statement. Consider this C code: call1(); // line 1 ++global; // line 2 ++global; // line 3 call2(&global, &local); // line 4 Today we end up with assembly and line tables like this: .loc 1 1 callq call1 leaq global(%rip), %rdi leaq local(%rsp), %rsi .loc 1 2 addq $1, global(%rip) .loc 1 3 addq $1, global(%rip) .loc 1 4 callq call2 The LEA instructions in the local value area have no source location and are treated as being on line 1. Stepping through the code in a debugger and correlating it with the assembly won't make much sense, because these materializations are only required for line 4. This is actually problematic for the VS debugger "set next statement" feature, which effectively assumes that there are no registers live across statement boundaries. By sinking the local value code into the statement and fixing up the source location, we can make that feature work. This was filed as https://bugs.llvm.org/show_bug.cgi?id=35975 and https://crbug.com/793819. This change is obviously not enough to make this feature work reliably in all cases, but I felt that it was worth doing anyway because it usually generates smaller, more comprehensible -O0 code. I measured a 0.12% regression in code generation time with LLC on the sqlite3 amalgamation, so I think this is worth doing. There are some special cases worth calling out in the commit message: 1. local values materialized for phis 2. local values used by no-op casts 3. dead local value code Local values can be materialized for phis, and this does not show up as a vreg use in MachineRegisterInfo. In this case, if there are no other uses, this patch sinks the value to the first terminator, EH label, or the end of the BB if nothing else exists. Local values may also be used by no-op casts, which adds the register to the RegFixups table. Without reversing the RegFixups map direction, we don't have enough information to sink these instructions. Lastly, if the local value register has no other uses, we can delete it. This comes up when fastisel tries two instruction selection approaches and the first materializes the value but fails and the second succeeds without using the local value. Reviewers: aprantl, dblaikie, qcolombet, MatzeB, vsk, echristo Subscribers: dotdash, chandlerc, hans, sdardis, amccarth, javed.absar, zturner, llvm-commits, hiraditya Differential Revision: https://reviews.llvm.org/D43093 llvm-svn: 327581
2018-03-14 21:54:21 +00:00
DenseSet<unsigned> RegsWithFixups;
/// StatepointStackSlots - A list of temporary stack slots (frame indices)
[Statepoints 3/4] Statepoint infrastructure for garbage collection: SelectionDAGBuilder This is the third patch in a small series. It contains the CodeGen support for lowering the gc.statepoint intrinsic sequences (223078) to the STATEPOINT pseudo machine instruction (223085). The change also includes the set of helper routines and classes for working with gc.statepoints, gc.relocates, and gc.results since the lowering code uses them. With this change, gc.statepoints should be functionally complete. The documentation will follow in the fourth change, and there will likely be some cleanup changes, but interested parties can start experimenting now. I'm not particularly happy with the amount of code or complexity involved with the lowering step, but at least it's fairly well isolated. The statepoint lowering code is split into it's own files and anyone not working on the statepoint support itself should be able to ignore it. During the lowering process, we currently spill aggressively to stack. This is not entirely ideal (and we have plans to do better), but it's functional, relatively straight forward, and matches closely the implementations of the patchpoint intrinsics. Most of the complexity comes from trying to keep relocated copies of values in the same stack slots across statepoints. Doing so avoids the insertion of pointless load and store instructions to reshuffle the stack. The current implementation isn't as effective as I'd like, but it is functional and 'good enough' for many common use cases. In the long term, I'd like to figure out how to integrate the statepoint lowering with the register allocator. In principal, we shouldn't need to eagerly spill at all. The register allocator should do any spilling required and the statepoint should simply record that fact. Depending on how challenging that turns out to be, we may invest in a smarter global stack slot assignment mechanism as a stop gap measure. Reviewed by: atrick, ributzka llvm-svn: 223137
2014-12-02 18:50:36 +00:00
/// used to spill values at a statepoint. We store them here to enable
/// reuse of the same stack slots across different statepoints in different
/// basic blocks.
SmallVector<unsigned, 50> StatepointStackSlots;
/// MBB - The current block.
MachineBasicBlock *MBB;
/// MBB - The current insert position inside the current block.
MachineBasicBlock::iterator InsertPt;
struct LiveOutInfo {
unsigned NumSignBits : 31;
unsigned IsValid : 1;
KnownBits Known = 1;
LiveOutInfo() : NumSignBits(0), IsValid(true) {}
};
/// Record the preferred extend type (ISD::SIGN_EXTEND or ISD::ZERO_EXTEND)
/// for a value.
DenseMap<const Value *, ISD::NodeType> PreferredExtendType;
/// VisitedBBs - The set of basic blocks visited thus far by instruction
/// selection.
SmallPtrSet<const BasicBlock*, 4> VisitedBBs;
/// PHINodesToUpdate - A list of phi instructions whose operand list will
/// be updated after processing the current basic block.
/// TODO: This isn't per-function state, it's per-basic-block state. But
/// there's no other convenient place for it to live right now.
std::vector<std::pair<MachineInstr*, unsigned> > PHINodesToUpdate;
unsigned OrigNumPHINodesToUpdate;
/// If the current MBB is a landing pad, the exception pointer and exception
/// selector registers are copied into these virtual registers by
/// SelectionDAGISel::PrepareEHLandingPad().
unsigned ExceptionPointerVirtReg, ExceptionSelectorVirtReg;
/// set - Initialize this FunctionLoweringInfo with the given Function
/// and its associated MachineFunction.
///
void set(const Function &Fn, MachineFunction &MF, SelectionDAG *DAG);
/// clear - Clear out all the function-specific state. This returns this
/// FunctionLoweringInfo to an empty state, ready to be used for a
/// different function.
void clear();
/// isExportedInst - Return true if the specified value is an instruction
/// exported from its block.
bool isExportedInst(const Value *V) {
return ValueMap.count(V);
}
unsigned CreateReg(MVT VT, bool isDivergent = false);
unsigned CreateRegs(const Value *V);
2015-10-05 04:43:48 +00:00
unsigned CreateRegs(Type *Ty, bool isDivergent = false);
2015-10-05 04:43:48 +00:00
unsigned InitializeRegForValue(const Value *V) {
// Tokens never live in vregs.
if (V->getType()->isTokenTy())
return 0;
unsigned &R = ValueMap[V];
assert(R == 0 && "Already initialized this value register!");
AMDGPU: Fix various issues around the VirtReg2Value mapping Summary: The VirtReg2Value mapping is crucial for getting consistently reliable divergence information into the SelectionDAG. This patch fixes a bunch of issues that lead to incorrect divergence info and introduces tight assertions to ensure we don't regress: 1. VirtReg2Value is generated lazily; there were some cases where a lookup was performed before all relevant virtual registers were created, leading to an out-of-sync mapping. Those cases were: - Complex code to lower formal arguments that generated CopyFromReg nodes from live-in registers (fixed by never querying the mapping for live-in registers). - Code that generates CopyToReg for formal arguments that are used outside the entry basic block (fixed by never querying the mapping for Register nodes, which don't need the divergence info anyway). 2. For complex values that are lowered to a sequence of registers, all registers must be reflected in the VirtReg2Value mapping. I am not adding any new tests, since I'm not actually aware of any bugs that these problems are causing with trunk as-is. However, I recently added a test case (in r346423) which fails when D53283 is applied without this change. Also, the new assertions should provide most of the effective test coverage. There is one test change in sdwa-peephole.ll. The underlying issue is that since the divergence info is now correct, the DAGISel will select V_OR_B32 directly instead of S_OR_B32. This leads to an extra COPY which affects the behavior of MachineLICM in a way that ends up with the S_MOV_B32 with the constant in a different basic block than the V_OR_B32, which is presumably what defeats the peephole. Reviewers: alex-t, arsenm, rampitec Subscribers: kzhuravl, jvesely, wdng, yaxunl, dstuttard, tpr, t-tye, llvm-commits Differential Revision: https://reviews.llvm.org/D54340 llvm-svn: 348049
2018-11-30 22:55:29 +00:00
assert(VirtReg2Value.empty());
return R = CreateRegs(V);
}
/// GetLiveOutRegInfo - Gets LiveOutInfo for a register, returning NULL if the
/// register is a PHI destination and the PHI's LiveOutInfo is not valid.
const LiveOutInfo *GetLiveOutRegInfo(unsigned Reg) {
if (!LiveOutRegInfo.inBounds(Reg))
return nullptr;
const LiveOutInfo *LOI = &LiveOutRegInfo[Reg];
if (!LOI->IsValid)
return nullptr;
return LOI;
}
/// GetLiveOutRegInfo - Gets LiveOutInfo for a register, returning NULL if the
/// register is a PHI destination and the PHI's LiveOutInfo is not valid. If
/// the register's LiveOutInfo is for a smaller bit width, it is extended to
/// the larger bit width by zero extension. The bit width must be no smaller
/// than the LiveOutInfo's existing bit width.
const LiveOutInfo *GetLiveOutRegInfo(unsigned Reg, unsigned BitWidth);
/// AddLiveOutRegInfo - Adds LiveOutInfo for a register.
void AddLiveOutRegInfo(unsigned Reg, unsigned NumSignBits,
const KnownBits &Known) {
// Only install this information if it tells us something.
if (NumSignBits == 1 && Known.isUnknown())
return;
LiveOutRegInfo.grow(Reg);
LiveOutInfo &LOI = LiveOutRegInfo[Reg];
LOI.NumSignBits = NumSignBits;
LOI.Known.One = Known.One;
LOI.Known.Zero = Known.Zero;
}
/// ComputePHILiveOutRegInfo - Compute LiveOutInfo for a PHI's destination
/// register based on the LiveOutInfo of its operands.
void ComputePHILiveOutRegInfo(const PHINode*);
/// InvalidatePHILiveOutRegInfo - Invalidates a PHI's LiveOutInfo, to be
/// called when a block is visited before all of its predecessors.
void InvalidatePHILiveOutRegInfo(const PHINode *PN) {
// PHIs with no uses have no ValueMap entry.
DenseMap<const Value*, unsigned>::const_iterator It = ValueMap.find(PN);
if (It == ValueMap.end())
return;
unsigned Reg = It->second;
if (Reg == 0)
return;
LiveOutRegInfo.grow(Reg);
LiveOutRegInfo[Reg].IsValid = false;
}
/// setArgumentFrameIndex - Record frame index for the byval
/// argument.
void setArgumentFrameIndex(const Argument *A, int FI);
/// getArgumentFrameIndex - Get frame index for the byval argument.
int getArgumentFrameIndex(const Argument *A);
unsigned getCatchPadExceptionPointerVReg(const Value *CPI,
const TargetRegisterClass *RC);
private:
void addSEHHandlersForLPads(ArrayRef<const LandingPadInst *> LPads);
/// LiveOutRegInfo - Information about live out vregs.
IndexedMap<LiveOutInfo, VirtReg2IndexFunctor> LiveOutRegInfo;
};
} // end namespace llvm
#endif // LLVM_CODEGEN_FUNCTIONLOWERINGINFO_H