1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-19 02:52:53 +02:00
llvm-mirror/lib/TableGen/Record.cpp

2478 lines
73 KiB
C++
Raw Normal View History

//===- Record.cpp - Record implementation ---------------------------------===//
//
// 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
//
//===----------------------------------------------------------------------===//
//
// Implement the tablegen record classes.
//
//===----------------------------------------------------------------------===//
#include "llvm/ADT/ArrayRef.h"
#include "llvm/ADT/DenseMap.h"
#include "llvm/ADT/FoldingSet.h"
#include "llvm/ADT/SmallString.h"
#include "llvm/ADT/SmallVector.h"
[tblgen] Track CodeInit origins when possible Summary: Add an SMLoc to CodeInit that records the source line it originated from. This allows tablegen to point precisely at portions of code when reporting errors within the CodeInit. For example, in the upcoming GlobalISel combiner, it can report undefined expansions and point at the instance of the expansion. This is achieved using something like: SMLoc::getFromPointer(SMLoc::getPointer() + (StringRef - CodeInit::getValue())) The location is lost when producing a CodeInit by string concatenation so a fallback SMLoc is required (e.g. the Record::getLoc()) but that's pretty rare for CodeInits. There's a reasonable case for extending tracking of a couple other Init objects, for example StringInit's are often parsed and it would be good to point inside the string when reporting errors about that. However, location tracking also harms de-duplication. This is fine for CodeInit where there's only a few hundred of them (~160 for X86) and it may be worth it for StringInit (~86k up to ~1.9M for roughly 15MB increase for X86). However the origin tracking would be a _terrible_ idea for IntInit, BitInit, and UnsetInit. I haven't measured either of those three but BitInit would most likely be on the order of increasing the current 2 BitInit values up to billions. Reviewers: volkan, aditya_nandakumar, bogner, paquette, aemerson Reviewed By: paquette Subscribers: javed.absar, kristof.beyls, dexonsmith, llvm-commits, kristina Tags: #llvm Differential Revision: https://reviews.llvm.org/D58141 llvm-svn: 355245
2019-03-02 01:12:57 +01:00
#include "llvm/ADT/Statistic.h"
#include "llvm/ADT/StringExtras.h"
#include "llvm/ADT/StringMap.h"
#include "llvm/ADT/StringRef.h"
[tblgen] Track CodeInit origins when possible Summary: Add an SMLoc to CodeInit that records the source line it originated from. This allows tablegen to point precisely at portions of code when reporting errors within the CodeInit. For example, in the upcoming GlobalISel combiner, it can report undefined expansions and point at the instance of the expansion. This is achieved using something like: SMLoc::getFromPointer(SMLoc::getPointer() + (StringRef - CodeInit::getValue())) The location is lost when producing a CodeInit by string concatenation so a fallback SMLoc is required (e.g. the Record::getLoc()) but that's pretty rare for CodeInits. There's a reasonable case for extending tracking of a couple other Init objects, for example StringInit's are often parsed and it would be good to point inside the string when reporting errors about that. However, location tracking also harms de-duplication. This is fine for CodeInit where there's only a few hundred of them (~160 for X86) and it may be worth it for StringInit (~86k up to ~1.9M for roughly 15MB increase for X86). However the origin tracking would be a _terrible_ idea for IntInit, BitInit, and UnsetInit. I haven't measured either of those three but BitInit would most likely be on the order of increasing the current 2 BitInit values up to billions. Reviewers: volkan, aditya_nandakumar, bogner, paquette, aemerson Reviewed By: paquette Subscribers: javed.absar, kristof.beyls, dexonsmith, llvm-commits, kristina Tags: #llvm Differential Revision: https://reviews.llvm.org/D58141 llvm-svn: 355245
2019-03-02 01:12:57 +01:00
#include "llvm/ADT/StringSet.h"
#include "llvm/Config/llvm-config.h"
#include "llvm/Support/Allocator.h"
#include "llvm/Support/Casting.h"
#include "llvm/Support/Compiler.h"
#include "llvm/Support/ErrorHandling.h"
#include "llvm/Support/SMLoc.h"
#include "llvm/Support/raw_ostream.h"
#include "llvm/TableGen/Error.h"
#include "llvm/TableGen/Record.h"
#include <cassert>
#include <cstdint>
#include <memory>
#include <map>
#include <string>
#include <utility>
#include <vector>
using namespace llvm;
[tblgen] Track CodeInit origins when possible Summary: Add an SMLoc to CodeInit that records the source line it originated from. This allows tablegen to point precisely at portions of code when reporting errors within the CodeInit. For example, in the upcoming GlobalISel combiner, it can report undefined expansions and point at the instance of the expansion. This is achieved using something like: SMLoc::getFromPointer(SMLoc::getPointer() + (StringRef - CodeInit::getValue())) The location is lost when producing a CodeInit by string concatenation so a fallback SMLoc is required (e.g. the Record::getLoc()) but that's pretty rare for CodeInits. There's a reasonable case for extending tracking of a couple other Init objects, for example StringInit's are often parsed and it would be good to point inside the string when reporting errors about that. However, location tracking also harms de-duplication. This is fine for CodeInit where there's only a few hundred of them (~160 for X86) and it may be worth it for StringInit (~86k up to ~1.9M for roughly 15MB increase for X86). However the origin tracking would be a _terrible_ idea for IntInit, BitInit, and UnsetInit. I haven't measured either of those three but BitInit would most likely be on the order of increasing the current 2 BitInit values up to billions. Reviewers: volkan, aditya_nandakumar, bogner, paquette, aemerson Reviewed By: paquette Subscribers: javed.absar, kristof.beyls, dexonsmith, llvm-commits, kristina Tags: #llvm Differential Revision: https://reviews.llvm.org/D58141 llvm-svn: 355245
2019-03-02 01:12:57 +01:00
#define DEBUG_TYPE "tblgen-records"
static BumpPtrAllocator Allocator;
[tblgen] Track CodeInit origins when possible Summary: Add an SMLoc to CodeInit that records the source line it originated from. This allows tablegen to point precisely at portions of code when reporting errors within the CodeInit. For example, in the upcoming GlobalISel combiner, it can report undefined expansions and point at the instance of the expansion. This is achieved using something like: SMLoc::getFromPointer(SMLoc::getPointer() + (StringRef - CodeInit::getValue())) The location is lost when producing a CodeInit by string concatenation so a fallback SMLoc is required (e.g. the Record::getLoc()) but that's pretty rare for CodeInits. There's a reasonable case for extending tracking of a couple other Init objects, for example StringInit's are often parsed and it would be good to point inside the string when reporting errors about that. However, location tracking also harms de-duplication. This is fine for CodeInit where there's only a few hundred of them (~160 for X86) and it may be worth it for StringInit (~86k up to ~1.9M for roughly 15MB increase for X86). However the origin tracking would be a _terrible_ idea for IntInit, BitInit, and UnsetInit. I haven't measured either of those three but BitInit would most likely be on the order of increasing the current 2 BitInit values up to billions. Reviewers: volkan, aditya_nandakumar, bogner, paquette, aemerson Reviewed By: paquette Subscribers: javed.absar, kristof.beyls, dexonsmith, llvm-commits, kristina Tags: #llvm Differential Revision: https://reviews.llvm.org/D58141 llvm-svn: 355245
2019-03-02 01:12:57 +01:00
STATISTIC(CodeInitsConstructed,
"The total number of unique CodeInits constructed");
//===----------------------------------------------------------------------===//
// Type implementations
//===----------------------------------------------------------------------===//
BitRecTy BitRecTy::Shared;
CodeRecTy CodeRecTy::Shared;
IntRecTy IntRecTy::Shared;
StringRecTy StringRecTy::Shared;
DagRecTy DagRecTy::Shared;
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
LLVM_DUMP_METHOD void RecTy::dump() const { print(errs()); }
#endif
ListRecTy *RecTy::getListTy() {
if (!ListTy)
ListTy = new(Allocator) ListRecTy(this);
return ListTy;
}
bool RecTy::typeIsConvertibleTo(const RecTy *RHS) const {
assert(RHS && "NULL pointer");
return Kind == RHS->getRecTyKind();
}
TableGen: Allow !cast of records, cleanup conversion machinery Summary: Distinguish two relationships between types: is-a and convertible-to. For example, a bit is not an int or vice versa, but they can be converted into each other (with range checks that you can think of as "dynamic": unlike other type checks, those range checks do not happen during parsing, but only once the final values have been established). Actually converting initializers between types is subtle: even when values of type A can be converted to type B (e.g. int into string), it may not be possible to do so with a concrete initializer (e.g., a VarInit that refers to a variable of type int cannot be immediately converted to a string). For this reason, distinguish between getCastTo and convertInitializerTo: the latter implements the actual conversion when appropriate, while the former will first try to do the actual conversion and fall back to introducing a !cast operation so that the conversion will be delayed until variable references have been resolved. To make the approach of adding !cast operations to work, !cast needs to fallback to convertInitializerTo when the special string <-> record logic does not apply. This enables casting records to a subclass, although that new functionality is only truly useful together with !isa, which will be added in a later change. The test is removed because it uses !srl on a bit sequence, which cannot really be supported consistently, but luckily isn't used anywhere either. Change-Id: I98168bf52649176654ed2ec61a29bdb29970cfe7 Reviewers: arsenm, craig.topper, tra, MartinO Subscribers: wdng, llvm-commits Differential Revision: https://reviews.llvm.org/D43753 llvm-svn: 326785
2018-03-06 14:48:39 +01:00
bool RecTy::typeIsA(const RecTy *RHS) const { return this == RHS; }
bool BitRecTy::typeIsConvertibleTo(const RecTy *RHS) const{
if (RecTy::typeIsConvertibleTo(RHS) || RHS->getRecTyKind() == IntRecTyKind)
return true;
if (const BitsRecTy *BitsTy = dyn_cast<BitsRecTy>(RHS))
return BitsTy->getNumBits() == 1;
return false;
}
BitsRecTy *BitsRecTy::get(unsigned Sz) {
static std::vector<BitsRecTy*> Shared;
if (Sz >= Shared.size())
Shared.resize(Sz + 1);
BitsRecTy *&Ty = Shared[Sz];
if (!Ty)
Ty = new(Allocator) BitsRecTy(Sz);
return Ty;
}
std::string BitsRecTy::getAsString() const {
return "bits<" + utostr(Size) + ">";
}
bool BitsRecTy::typeIsConvertibleTo(const RecTy *RHS) const {
if (RecTy::typeIsConvertibleTo(RHS)) //argument and the sender are same type
return cast<BitsRecTy>(RHS)->Size == Size;
RecTyKind kind = RHS->getRecTyKind();
return (kind == BitRecTyKind && Size == 1) || (kind == IntRecTyKind);
}
TableGen: Allow !cast of records, cleanup conversion machinery Summary: Distinguish two relationships between types: is-a and convertible-to. For example, a bit is not an int or vice versa, but they can be converted into each other (with range checks that you can think of as "dynamic": unlike other type checks, those range checks do not happen during parsing, but only once the final values have been established). Actually converting initializers between types is subtle: even when values of type A can be converted to type B (e.g. int into string), it may not be possible to do so with a concrete initializer (e.g., a VarInit that refers to a variable of type int cannot be immediately converted to a string). For this reason, distinguish between getCastTo and convertInitializerTo: the latter implements the actual conversion when appropriate, while the former will first try to do the actual conversion and fall back to introducing a !cast operation so that the conversion will be delayed until variable references have been resolved. To make the approach of adding !cast operations to work, !cast needs to fallback to convertInitializerTo when the special string <-> record logic does not apply. This enables casting records to a subclass, although that new functionality is only truly useful together with !isa, which will be added in a later change. The test is removed because it uses !srl on a bit sequence, which cannot really be supported consistently, but luckily isn't used anywhere either. Change-Id: I98168bf52649176654ed2ec61a29bdb29970cfe7 Reviewers: arsenm, craig.topper, tra, MartinO Subscribers: wdng, llvm-commits Differential Revision: https://reviews.llvm.org/D43753 llvm-svn: 326785
2018-03-06 14:48:39 +01:00
bool BitsRecTy::typeIsA(const RecTy *RHS) const {
if (const BitsRecTy *RHSb = dyn_cast<BitsRecTy>(RHS))
return RHSb->Size == Size;
return false;
}
bool IntRecTy::typeIsConvertibleTo(const RecTy *RHS) const {
RecTyKind kind = RHS->getRecTyKind();
return kind==BitRecTyKind || kind==BitsRecTyKind || kind==IntRecTyKind;
}
bool CodeRecTy::typeIsConvertibleTo(const RecTy *RHS) const {
RecTyKind Kind = RHS->getRecTyKind();
return Kind == CodeRecTyKind || Kind == StringRecTyKind;
}
std::string StringRecTy::getAsString() const {
return "string";
}
bool StringRecTy::typeIsConvertibleTo(const RecTy *RHS) const {
RecTyKind Kind = RHS->getRecTyKind();
return Kind == StringRecTyKind || Kind == CodeRecTyKind;
}
std::string ListRecTy::getAsString() const {
return "list<" + Ty->getAsString() + ">";
}
bool ListRecTy::typeIsConvertibleTo(const RecTy *RHS) const {
if (const auto *ListTy = dyn_cast<ListRecTy>(RHS))
return Ty->typeIsConvertibleTo(ListTy->getElementType());
return false;
}
TableGen: Allow !cast of records, cleanup conversion machinery Summary: Distinguish two relationships between types: is-a and convertible-to. For example, a bit is not an int or vice versa, but they can be converted into each other (with range checks that you can think of as "dynamic": unlike other type checks, those range checks do not happen during parsing, but only once the final values have been established). Actually converting initializers between types is subtle: even when values of type A can be converted to type B (e.g. int into string), it may not be possible to do so with a concrete initializer (e.g., a VarInit that refers to a variable of type int cannot be immediately converted to a string). For this reason, distinguish between getCastTo and convertInitializerTo: the latter implements the actual conversion when appropriate, while the former will first try to do the actual conversion and fall back to introducing a !cast operation so that the conversion will be delayed until variable references have been resolved. To make the approach of adding !cast operations to work, !cast needs to fallback to convertInitializerTo when the special string <-> record logic does not apply. This enables casting records to a subclass, although that new functionality is only truly useful together with !isa, which will be added in a later change. The test is removed because it uses !srl on a bit sequence, which cannot really be supported consistently, but luckily isn't used anywhere either. Change-Id: I98168bf52649176654ed2ec61a29bdb29970cfe7 Reviewers: arsenm, craig.topper, tra, MartinO Subscribers: wdng, llvm-commits Differential Revision: https://reviews.llvm.org/D43753 llvm-svn: 326785
2018-03-06 14:48:39 +01:00
bool ListRecTy::typeIsA(const RecTy *RHS) const {
if (const ListRecTy *RHSl = dyn_cast<ListRecTy>(RHS))
return getElementType()->typeIsA(RHSl->getElementType());
return false;
}
std::string DagRecTy::getAsString() const {
return "dag";
}
static void ProfileRecordRecTy(FoldingSetNodeID &ID,
ArrayRef<Record *> Classes) {
ID.AddInteger(Classes.size());
for (Record *R : Classes)
ID.AddPointer(R);
}
RecordRecTy *RecordRecTy::get(ArrayRef<Record *> UnsortedClasses) {
if (UnsortedClasses.empty()) {
static RecordRecTy AnyRecord(0);
return &AnyRecord;
}
FoldingSet<RecordRecTy> &ThePool =
UnsortedClasses[0]->getRecords().RecordTypePool;
SmallVector<Record *, 4> Classes(UnsortedClasses.begin(),
UnsortedClasses.end());
llvm::sort(Classes, [](Record *LHS, Record *RHS) {
return LHS->getNameInitAsString() < RHS->getNameInitAsString();
});
FoldingSetNodeID ID;
ProfileRecordRecTy(ID, Classes);
void *IP = nullptr;
if (RecordRecTy *Ty = ThePool.FindNodeOrInsertPos(ID, IP))
return Ty;
#ifndef NDEBUG
// Check for redundancy.
for (unsigned i = 0; i < Classes.size(); ++i) {
for (unsigned j = 0; j < Classes.size(); ++j) {
assert(i == j || !Classes[i]->isSubClassOf(Classes[j]));
}
assert(&Classes[0]->getRecords() == &Classes[i]->getRecords());
}
#endif
void *Mem = Allocator.Allocate(totalSizeToAlloc<Record *>(Classes.size()),
alignof(RecordRecTy));
RecordRecTy *Ty = new(Mem) RecordRecTy(Classes.size());
std::uninitialized_copy(Classes.begin(), Classes.end(),
Ty->getTrailingObjects<Record *>());
ThePool.InsertNode(Ty, IP);
return Ty;
}
void RecordRecTy::Profile(FoldingSetNodeID &ID) const {
ProfileRecordRecTy(ID, getClasses());
}
std::string RecordRecTy::getAsString() const {
if (NumClasses == 1)
return getClasses()[0]->getNameInitAsString();
std::string Str = "{";
bool First = true;
for (Record *R : getClasses()) {
if (!First)
Str += ", ";
First = false;
Str += R->getNameInitAsString();
}
Str += "}";
return Str;
}
bool RecordRecTy::isSubClassOf(Record *Class) const {
return llvm::any_of(getClasses(), [Class](Record *MySuperClass) {
return MySuperClass == Class ||
MySuperClass->isSubClassOf(Class);
});
}
bool RecordRecTy::typeIsConvertibleTo(const RecTy *RHS) const {
if (this == RHS)
return true;
const RecordRecTy *RTy = dyn_cast<RecordRecTy>(RHS);
if (!RTy)
return false;
return llvm::all_of(RTy->getClasses(), [this](Record *TargetClass) {
return isSubClassOf(TargetClass);
});
}
TableGen: Allow !cast of records, cleanup conversion machinery Summary: Distinguish two relationships between types: is-a and convertible-to. For example, a bit is not an int or vice versa, but they can be converted into each other (with range checks that you can think of as "dynamic": unlike other type checks, those range checks do not happen during parsing, but only once the final values have been established). Actually converting initializers between types is subtle: even when values of type A can be converted to type B (e.g. int into string), it may not be possible to do so with a concrete initializer (e.g., a VarInit that refers to a variable of type int cannot be immediately converted to a string). For this reason, distinguish between getCastTo and convertInitializerTo: the latter implements the actual conversion when appropriate, while the former will first try to do the actual conversion and fall back to introducing a !cast operation so that the conversion will be delayed until variable references have been resolved. To make the approach of adding !cast operations to work, !cast needs to fallback to convertInitializerTo when the special string <-> record logic does not apply. This enables casting records to a subclass, although that new functionality is only truly useful together with !isa, which will be added in a later change. The test is removed because it uses !srl on a bit sequence, which cannot really be supported consistently, but luckily isn't used anywhere either. Change-Id: I98168bf52649176654ed2ec61a29bdb29970cfe7 Reviewers: arsenm, craig.topper, tra, MartinO Subscribers: wdng, llvm-commits Differential Revision: https://reviews.llvm.org/D43753 llvm-svn: 326785
2018-03-06 14:48:39 +01:00
bool RecordRecTy::typeIsA(const RecTy *RHS) const {
return typeIsConvertibleTo(RHS);
}
static RecordRecTy *resolveRecordTypes(RecordRecTy *T1, RecordRecTy *T2) {
SmallVector<Record *, 4> CommonSuperClasses;
SmallVector<Record *, 4> Stack;
Stack.insert(Stack.end(), T1->classes_begin(), T1->classes_end());
while (!Stack.empty()) {
Record *R = Stack.back();
Stack.pop_back();
if (T2->isSubClassOf(R)) {
CommonSuperClasses.push_back(R);
} else {
R->getDirectSuperClasses(Stack);
}
}
return RecordRecTy::get(CommonSuperClasses);
}
RecTy *llvm::resolveTypes(RecTy *T1, RecTy *T2) {
if (T1 == T2)
return T1;
if (RecordRecTy *RecTy1 = dyn_cast<RecordRecTy>(T1)) {
if (RecordRecTy *RecTy2 = dyn_cast<RecordRecTy>(T2))
return resolveRecordTypes(RecTy1, RecTy2);
}
if (T1->typeIsConvertibleTo(T2))
return T2;
if (T2->typeIsConvertibleTo(T1))
return T1;
if (ListRecTy *ListTy1 = dyn_cast<ListRecTy>(T1)) {
if (ListRecTy *ListTy2 = dyn_cast<ListRecTy>(T2)) {
RecTy* NewType = resolveTypes(ListTy1->getElementType(),
ListTy2->getElementType());
if (NewType)
return NewType->getListTy();
}
}
return nullptr;
}
//===----------------------------------------------------------------------===//
// Initializer implementations
//===----------------------------------------------------------------------===//
void Init::anchor() {}
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
LLVM_DUMP_METHOD void Init::dump() const { return print(errs()); }
#endif
UnsetInit *UnsetInit::get() {
static UnsetInit TheInit;
return &TheInit;
}
TableGen: Allow !cast of records, cleanup conversion machinery Summary: Distinguish two relationships between types: is-a and convertible-to. For example, a bit is not an int or vice versa, but they can be converted into each other (with range checks that you can think of as "dynamic": unlike other type checks, those range checks do not happen during parsing, but only once the final values have been established). Actually converting initializers between types is subtle: even when values of type A can be converted to type B (e.g. int into string), it may not be possible to do so with a concrete initializer (e.g., a VarInit that refers to a variable of type int cannot be immediately converted to a string). For this reason, distinguish between getCastTo and convertInitializerTo: the latter implements the actual conversion when appropriate, while the former will first try to do the actual conversion and fall back to introducing a !cast operation so that the conversion will be delayed until variable references have been resolved. To make the approach of adding !cast operations to work, !cast needs to fallback to convertInitializerTo when the special string <-> record logic does not apply. This enables casting records to a subclass, although that new functionality is only truly useful together with !isa, which will be added in a later change. The test is removed because it uses !srl on a bit sequence, which cannot really be supported consistently, but luckily isn't used anywhere either. Change-Id: I98168bf52649176654ed2ec61a29bdb29970cfe7 Reviewers: arsenm, craig.topper, tra, MartinO Subscribers: wdng, llvm-commits Differential Revision: https://reviews.llvm.org/D43753 llvm-svn: 326785
2018-03-06 14:48:39 +01:00
Init *UnsetInit::getCastTo(RecTy *Ty) const {
return const_cast<UnsetInit *>(this);
}
TableGen: Allow !cast of records, cleanup conversion machinery Summary: Distinguish two relationships between types: is-a and convertible-to. For example, a bit is not an int or vice versa, but they can be converted into each other (with range checks that you can think of as "dynamic": unlike other type checks, those range checks do not happen during parsing, but only once the final values have been established). Actually converting initializers between types is subtle: even when values of type A can be converted to type B (e.g. int into string), it may not be possible to do so with a concrete initializer (e.g., a VarInit that refers to a variable of type int cannot be immediately converted to a string). For this reason, distinguish between getCastTo and convertInitializerTo: the latter implements the actual conversion when appropriate, while the former will first try to do the actual conversion and fall back to introducing a !cast operation so that the conversion will be delayed until variable references have been resolved. To make the approach of adding !cast operations to work, !cast needs to fallback to convertInitializerTo when the special string <-> record logic does not apply. This enables casting records to a subclass, although that new functionality is only truly useful together with !isa, which will be added in a later change. The test is removed because it uses !srl on a bit sequence, which cannot really be supported consistently, but luckily isn't used anywhere either. Change-Id: I98168bf52649176654ed2ec61a29bdb29970cfe7 Reviewers: arsenm, craig.topper, tra, MartinO Subscribers: wdng, llvm-commits Differential Revision: https://reviews.llvm.org/D43753 llvm-svn: 326785
2018-03-06 14:48:39 +01:00
Init *UnsetInit::convertInitializerTo(RecTy *Ty) const {
return const_cast<UnsetInit *>(this);
}
BitInit *BitInit::get(bool V) {
static BitInit True(true);
static BitInit False(false);
return V ? &True : &False;
}
Init *BitInit::convertInitializerTo(RecTy *Ty) const {
if (isa<BitRecTy>(Ty))
return const_cast<BitInit *>(this);
if (isa<IntRecTy>(Ty))
return IntInit::get(getValue());
if (auto *BRT = dyn_cast<BitsRecTy>(Ty)) {
// Can only convert single bit.
if (BRT->getNumBits() == 1)
return BitsInit::get(const_cast<BitInit *>(this));
}
return nullptr;
}
static void
ProfileBitsInit(FoldingSetNodeID &ID, ArrayRef<Init *> Range) {
ID.AddInteger(Range.size());
for (Init *I : Range)
ID.AddPointer(I);
}
BitsInit *BitsInit::get(ArrayRef<Init *> Range) {
static FoldingSet<BitsInit> ThePool;
FoldingSetNodeID ID;
ProfileBitsInit(ID, Range);
void *IP = nullptr;
if (BitsInit *I = ThePool.FindNodeOrInsertPos(ID, IP))
return I;
void *Mem = Allocator.Allocate(totalSizeToAlloc<Init *>(Range.size()),
alignof(BitsInit));
BitsInit *I = new(Mem) BitsInit(Range.size());
std::uninitialized_copy(Range.begin(), Range.end(),
I->getTrailingObjects<Init *>());
ThePool.InsertNode(I, IP);
return I;
}
void BitsInit::Profile(FoldingSetNodeID &ID) const {
ProfileBitsInit(ID, makeArrayRef(getTrailingObjects<Init *>(), NumBits));
}
Init *BitsInit::convertInitializerTo(RecTy *Ty) const {
if (isa<BitRecTy>(Ty)) {
if (getNumBits() != 1) return nullptr; // Only accept if just one bit!
return getBit(0);
}
if (auto *BRT = dyn_cast<BitsRecTy>(Ty)) {
// If the number of bits is right, return it. Otherwise we need to expand
// or truncate.
if (getNumBits() != BRT->getNumBits()) return nullptr;
return const_cast<BitsInit *>(this);
}
if (isa<IntRecTy>(Ty)) {
int64_t Result = 0;
for (unsigned i = 0, e = getNumBits(); i != e; ++i)
if (auto *Bit = dyn_cast<BitInit>(getBit(i)))
Result |= static_cast<int64_t>(Bit->getValue()) << i;
else
return nullptr;
return IntInit::get(Result);
}
return nullptr;
}
Init *
BitsInit::convertInitializerBitRange(ArrayRef<unsigned> Bits) const {
SmallVector<Init *, 16> NewBits(Bits.size());
for (unsigned i = 0, e = Bits.size(); i != e; ++i) {
if (Bits[i] >= getNumBits())
return nullptr;
NewBits[i] = getBit(Bits[i]);
}
return BitsInit::get(NewBits);
}
bool BitsInit::isConcrete() const {
for (unsigned i = 0, e = getNumBits(); i != e; ++i) {
if (!getBit(i)->isConcrete())
return false;
}
return true;
}
std::string BitsInit::getAsString() const {
std::string Result = "{ ";
for (unsigned i = 0, e = getNumBits(); i != e; ++i) {
if (i) Result += ", ";
if (Init *Bit = getBit(e-i-1))
Result += Bit->getAsString();
else
Result += "*";
}
return Result + " }";
}
// resolveReferences - If there are any field references that refer to fields
// that have been filled in, we can propagate the values now.
Init *BitsInit::resolveReferences(Resolver &R) const {
bool Changed = false;
SmallVector<Init *, 16> NewBits(getNumBits());
Init *CachedBitVarRef = nullptr;
Init *CachedBitVarResolved = nullptr;
for (unsigned i = 0, e = getNumBits(); i != e; ++i) {
Init *CurBit = getBit(i);
Init *NewBit = CurBit;
if (VarBitInit *CurBitVar = dyn_cast<VarBitInit>(CurBit)) {
if (CurBitVar->getBitVar() != CachedBitVarRef) {
CachedBitVarRef = CurBitVar->getBitVar();
CachedBitVarResolved = CachedBitVarRef->resolveReferences(R);
}
assert(CachedBitVarResolved && "Unresolved bitvar reference");
NewBit = CachedBitVarResolved->getBit(CurBitVar->getBitNum());
} else {
// getBit(0) implicitly converts int and bits<1> values to bit.
NewBit = CurBit->resolveReferences(R)->getBit(0);
}
if (isa<UnsetInit>(NewBit) && R.keepUnsetBits())
NewBit = CurBit;
NewBits[i] = NewBit;
Changed |= CurBit != NewBit;
}
if (Changed)
return BitsInit::get(NewBits);
return const_cast<BitsInit *>(this);
}
IntInit *IntInit::get(int64_t V) {
static std::map<int64_t, IntInit*> ThePool;
IntInit *&I = ThePool[V];
if (!I) I = new(Allocator) IntInit(V);
return I;
}
std::string IntInit::getAsString() const {
return itostr(Value);
}
static bool canFitInBitfield(int64_t Value, unsigned NumBits) {
// For example, with NumBits == 4, we permit Values from [-7 .. 15].
return (NumBits >= sizeof(Value) * 8) ||
(Value >> NumBits == 0) || (Value >> (NumBits-1) == -1);
}
Init *IntInit::convertInitializerTo(RecTy *Ty) const {
if (isa<IntRecTy>(Ty))
return const_cast<IntInit *>(this);
if (isa<BitRecTy>(Ty)) {
int64_t Val = getValue();
if (Val != 0 && Val != 1) return nullptr; // Only accept 0 or 1 for a bit!
return BitInit::get(Val != 0);
}
if (auto *BRT = dyn_cast<BitsRecTy>(Ty)) {
int64_t Value = getValue();
// Make sure this bitfield is large enough to hold the integer value.
if (!canFitInBitfield(Value, BRT->getNumBits()))
return nullptr;
SmallVector<Init *, 16> NewBits(BRT->getNumBits());
for (unsigned i = 0; i != BRT->getNumBits(); ++i)
NewBits[i] = BitInit::get(Value & ((i < 64) ? (1LL << i) : 0));
return BitsInit::get(NewBits);
}
return nullptr;
}
Init *
IntInit::convertInitializerBitRange(ArrayRef<unsigned> Bits) const {
SmallVector<Init *, 16> NewBits(Bits.size());
for (unsigned i = 0, e = Bits.size(); i != e; ++i) {
if (Bits[i] >= 64)
return nullptr;
NewBits[i] = BitInit::get(Value & (INT64_C(1) << Bits[i]));
}
return BitsInit::get(NewBits);
}
[tblgen] Track CodeInit origins when possible Summary: Add an SMLoc to CodeInit that records the source line it originated from. This allows tablegen to point precisely at portions of code when reporting errors within the CodeInit. For example, in the upcoming GlobalISel combiner, it can report undefined expansions and point at the instance of the expansion. This is achieved using something like: SMLoc::getFromPointer(SMLoc::getPointer() + (StringRef - CodeInit::getValue())) The location is lost when producing a CodeInit by string concatenation so a fallback SMLoc is required (e.g. the Record::getLoc()) but that's pretty rare for CodeInits. There's a reasonable case for extending tracking of a couple other Init objects, for example StringInit's are often parsed and it would be good to point inside the string when reporting errors about that. However, location tracking also harms de-duplication. This is fine for CodeInit where there's only a few hundred of them (~160 for X86) and it may be worth it for StringInit (~86k up to ~1.9M for roughly 15MB increase for X86). However the origin tracking would be a _terrible_ idea for IntInit, BitInit, and UnsetInit. I haven't measured either of those three but BitInit would most likely be on the order of increasing the current 2 BitInit values up to billions. Reviewers: volkan, aditya_nandakumar, bogner, paquette, aemerson Reviewed By: paquette Subscribers: javed.absar, kristof.beyls, dexonsmith, llvm-commits, kristina Tags: #llvm Differential Revision: https://reviews.llvm.org/D58141 llvm-svn: 355245
2019-03-02 01:12:57 +01:00
CodeInit *CodeInit::get(StringRef V, const SMLoc &Loc) {
static StringSet<BumpPtrAllocator &> ThePool(Allocator);
[tblgen] Track CodeInit origins when possible Summary: Add an SMLoc to CodeInit that records the source line it originated from. This allows tablegen to point precisely at portions of code when reporting errors within the CodeInit. For example, in the upcoming GlobalISel combiner, it can report undefined expansions and point at the instance of the expansion. This is achieved using something like: SMLoc::getFromPointer(SMLoc::getPointer() + (StringRef - CodeInit::getValue())) The location is lost when producing a CodeInit by string concatenation so a fallback SMLoc is required (e.g. the Record::getLoc()) but that's pretty rare for CodeInits. There's a reasonable case for extending tracking of a couple other Init objects, for example StringInit's are often parsed and it would be good to point inside the string when reporting errors about that. However, location tracking also harms de-duplication. This is fine for CodeInit where there's only a few hundred of them (~160 for X86) and it may be worth it for StringInit (~86k up to ~1.9M for roughly 15MB increase for X86). However the origin tracking would be a _terrible_ idea for IntInit, BitInit, and UnsetInit. I haven't measured either of those three but BitInit would most likely be on the order of increasing the current 2 BitInit values up to billions. Reviewers: volkan, aditya_nandakumar, bogner, paquette, aemerson Reviewed By: paquette Subscribers: javed.absar, kristof.beyls, dexonsmith, llvm-commits, kristina Tags: #llvm Differential Revision: https://reviews.llvm.org/D58141 llvm-svn: 355245
2019-03-02 01:12:57 +01:00
CodeInitsConstructed++;
// Unlike StringMap, StringSet doesn't accept empty keys.
if (V.empty())
return new (Allocator) CodeInit("", Loc);
// Location tracking prevents us from de-duping CodeInits as we're never
// called with the same string and same location twice. However, we can at
// least de-dupe the strings for a modest saving.
auto &Entry = *ThePool.insert(V).first;
return new(Allocator) CodeInit(Entry.getKey(), Loc);
}
StringInit *StringInit::get(StringRef V) {
static StringMap<StringInit*, BumpPtrAllocator &> ThePool(Allocator);
auto &Entry = *ThePool.insert(std::make_pair(V, nullptr)).first;
if (!Entry.second)
Entry.second = new(Allocator) StringInit(Entry.getKey());
return Entry.second;
}
Init *StringInit::convertInitializerTo(RecTy *Ty) const {
if (isa<StringRecTy>(Ty))
return const_cast<StringInit *>(this);
if (isa<CodeRecTy>(Ty))
[tblgen] Track CodeInit origins when possible Summary: Add an SMLoc to CodeInit that records the source line it originated from. This allows tablegen to point precisely at portions of code when reporting errors within the CodeInit. For example, in the upcoming GlobalISel combiner, it can report undefined expansions and point at the instance of the expansion. This is achieved using something like: SMLoc::getFromPointer(SMLoc::getPointer() + (StringRef - CodeInit::getValue())) The location is lost when producing a CodeInit by string concatenation so a fallback SMLoc is required (e.g. the Record::getLoc()) but that's pretty rare for CodeInits. There's a reasonable case for extending tracking of a couple other Init objects, for example StringInit's are often parsed and it would be good to point inside the string when reporting errors about that. However, location tracking also harms de-duplication. This is fine for CodeInit where there's only a few hundred of them (~160 for X86) and it may be worth it for StringInit (~86k up to ~1.9M for roughly 15MB increase for X86). However the origin tracking would be a _terrible_ idea for IntInit, BitInit, and UnsetInit. I haven't measured either of those three but BitInit would most likely be on the order of increasing the current 2 BitInit values up to billions. Reviewers: volkan, aditya_nandakumar, bogner, paquette, aemerson Reviewed By: paquette Subscribers: javed.absar, kristof.beyls, dexonsmith, llvm-commits, kristina Tags: #llvm Differential Revision: https://reviews.llvm.org/D58141 llvm-svn: 355245
2019-03-02 01:12:57 +01:00
return CodeInit::get(getValue(), SMLoc());
return nullptr;
}
Init *CodeInit::convertInitializerTo(RecTy *Ty) const {
if (isa<CodeRecTy>(Ty))
return const_cast<CodeInit *>(this);
if (isa<StringRecTy>(Ty))
return StringInit::get(getValue());
return nullptr;
}
static void ProfileListInit(FoldingSetNodeID &ID,
ArrayRef<Init *> Range,
RecTy *EltTy) {
ID.AddInteger(Range.size());
ID.AddPointer(EltTy);
for (Init *I : Range)
ID.AddPointer(I);
}
ListInit *ListInit::get(ArrayRef<Init *> Range, RecTy *EltTy) {
static FoldingSet<ListInit> ThePool;
FoldingSetNodeID ID;
ProfileListInit(ID, Range, EltTy);
void *IP = nullptr;
if (ListInit *I = ThePool.FindNodeOrInsertPos(ID, IP))
return I;
assert(Range.empty() || !isa<TypedInit>(Range[0]) ||
cast<TypedInit>(Range[0])->getType()->typeIsConvertibleTo(EltTy));
void *Mem = Allocator.Allocate(totalSizeToAlloc<Init *>(Range.size()),
alignof(ListInit));
ListInit *I = new(Mem) ListInit(Range.size(), EltTy);
std::uninitialized_copy(Range.begin(), Range.end(),
I->getTrailingObjects<Init *>());
ThePool.InsertNode(I, IP);
return I;
}
void ListInit::Profile(FoldingSetNodeID &ID) const {
RecTy *EltTy = cast<ListRecTy>(getType())->getElementType();
ProfileListInit(ID, getValues(), EltTy);
}
Init *ListInit::convertInitializerTo(RecTy *Ty) const {
if (getType() == Ty)
return const_cast<ListInit*>(this);
if (auto *LRT = dyn_cast<ListRecTy>(Ty)) {
SmallVector<Init*, 8> Elements;
Elements.reserve(getValues().size());
// Verify that all of the elements of the list are subclasses of the
// appropriate class!
bool Changed = false;
RecTy *ElementType = LRT->getElementType();
for (Init *I : getValues())
if (Init *CI = I->convertInitializerTo(ElementType)) {
Elements.push_back(CI);
if (CI != I)
Changed = true;
2017-08-28 08:47:47 +02:00
} else
return nullptr;
if (!Changed)
return const_cast<ListInit*>(this);
return ListInit::get(Elements, ElementType);
}
return nullptr;
}
Init *ListInit::convertInitListSlice(ArrayRef<unsigned> Elements) const {
SmallVector<Init*, 8> Vals;
Vals.reserve(Elements.size());
for (unsigned Element : Elements) {
if (Element >= size())
return nullptr;
Vals.push_back(getElement(Element));
}
return ListInit::get(Vals, getElementType());
}
2007-02-27 23:08:27 +01:00
Record *ListInit::getElementAsRecord(unsigned i) const {
assert(i < NumValues && "List element index out of range!");
DefInit *DI = dyn_cast<DefInit>(getElement(i));
if (!DI)
PrintFatalError("Expected record in list!");
2007-02-27 23:08:27 +01:00
return DI->getDef();
}
Init *ListInit::resolveReferences(Resolver &R) const {
SmallVector<Init*, 8> Resolved;
Resolved.reserve(size());
bool Changed = false;
for (Init *CurElt : getValues()) {
Init *E = CurElt->resolveReferences(R);
Changed |= E != CurElt;
Resolved.push_back(E);
}
if (Changed)
return ListInit::get(Resolved, getElementType());
return const_cast<ListInit *>(this);
}
bool ListInit::isConcrete() const {
for (Init *Element : *this) {
if (!Element->isConcrete())
return false;
}
return true;
}
std::string ListInit::getAsString() const {
std::string Result = "[";
const char *sep = "";
for (Init *Element : *this) {
Result += sep;
sep = ", ";
Result += Element->getAsString();
}
return Result + "]";
}
Init *OpInit::getBit(unsigned Bit) const {
if (getType() == BitRecTy::get())
return const_cast<OpInit*>(this);
return VarBitInit::get(const_cast<OpInit*>(this), Bit);
}
static void
ProfileUnOpInit(FoldingSetNodeID &ID, unsigned Opcode, Init *Op, RecTy *Type) {
ID.AddInteger(Opcode);
ID.AddPointer(Op);
ID.AddPointer(Type);
}
UnOpInit *UnOpInit::get(UnaryOp Opc, Init *LHS, RecTy *Type) {
static FoldingSet<UnOpInit> ThePool;
FoldingSetNodeID ID;
ProfileUnOpInit(ID, Opc, LHS, Type);
void *IP = nullptr;
if (UnOpInit *I = ThePool.FindNodeOrInsertPos(ID, IP))
return I;
UnOpInit *I = new(Allocator) UnOpInit(Opc, LHS, Type);
ThePool.InsertNode(I, IP);
return I;
}
void UnOpInit::Profile(FoldingSetNodeID &ID) const {
ProfileUnOpInit(ID, getOpcode(), getOperand(), getType());
}
Init *UnOpInit::Fold(Record *CurRec, bool IsFinal) const {
2009-05-14 23:22:49 +02:00
switch (getOpcode()) {
case CAST:
if (isa<StringRecTy>(getType())) {
if (StringInit *LHSs = dyn_cast<StringInit>(LHS))
return LHSs;
2009-05-14 23:22:49 +02:00
if (DefInit *LHSd = dyn_cast<DefInit>(LHS))
return StringInit::get(LHSd->getAsString());
if (IntInit *LHSi = dyn_cast<IntInit>(LHS))
return StringInit::get(LHSi->getAsString());
TableGen: Allow !cast of records, cleanup conversion machinery Summary: Distinguish two relationships between types: is-a and convertible-to. For example, a bit is not an int or vice versa, but they can be converted into each other (with range checks that you can think of as "dynamic": unlike other type checks, those range checks do not happen during parsing, but only once the final values have been established). Actually converting initializers between types is subtle: even when values of type A can be converted to type B (e.g. int into string), it may not be possible to do so with a concrete initializer (e.g., a VarInit that refers to a variable of type int cannot be immediately converted to a string). For this reason, distinguish between getCastTo and convertInitializerTo: the latter implements the actual conversion when appropriate, while the former will first try to do the actual conversion and fall back to introducing a !cast operation so that the conversion will be delayed until variable references have been resolved. To make the approach of adding !cast operations to work, !cast needs to fallback to convertInitializerTo when the special string <-> record logic does not apply. This enables casting records to a subclass, although that new functionality is only truly useful together with !isa, which will be added in a later change. The test is removed because it uses !srl on a bit sequence, which cannot really be supported consistently, but luckily isn't used anywhere either. Change-Id: I98168bf52649176654ed2ec61a29bdb29970cfe7 Reviewers: arsenm, craig.topper, tra, MartinO Subscribers: wdng, llvm-commits Differential Revision: https://reviews.llvm.org/D43753 llvm-svn: 326785
2018-03-06 14:48:39 +01:00
} else if (isa<RecordRecTy>(getType())) {
if (StringInit *Name = dyn_cast<StringInit>(LHS)) {
if (!CurRec && !IsFinal)
break;
assert(CurRec && "NULL pointer");
Record *D;
// Self-references are allowed, but their resolution is delayed until
// the final resolve to ensure that we get the correct type for them.
if (Name == CurRec->getNameInit()) {
if (!IsFinal)
break;
D = CurRec;
} else {
D = CurRec->getRecords().getDef(Name->getValue());
if (!D) {
if (IsFinal)
PrintFatalError(CurRec->getLoc(),
Twine("Undefined reference to record: '") +
Name->getValue() + "'\n");
break;
}
}
2009-05-14 23:22:49 +02:00
DefInit *DI = DefInit::get(D);
if (!DI->getType()->typeIsA(getType())) {
PrintFatalError(CurRec->getLoc(),
Twine("Expected type '") +
getType()->getAsString() + "', got '" +
DI->getType()->getAsString() + "' in: " +
getAsString() + "\n");
}
return DI;
}
2009-05-14 23:22:49 +02:00
}
TableGen: Allow !cast of records, cleanup conversion machinery Summary: Distinguish two relationships between types: is-a and convertible-to. For example, a bit is not an int or vice versa, but they can be converted into each other (with range checks that you can think of as "dynamic": unlike other type checks, those range checks do not happen during parsing, but only once the final values have been established). Actually converting initializers between types is subtle: even when values of type A can be converted to type B (e.g. int into string), it may not be possible to do so with a concrete initializer (e.g., a VarInit that refers to a variable of type int cannot be immediately converted to a string). For this reason, distinguish between getCastTo and convertInitializerTo: the latter implements the actual conversion when appropriate, while the former will first try to do the actual conversion and fall back to introducing a !cast operation so that the conversion will be delayed until variable references have been resolved. To make the approach of adding !cast operations to work, !cast needs to fallback to convertInitializerTo when the special string <-> record logic does not apply. This enables casting records to a subclass, although that new functionality is only truly useful together with !isa, which will be added in a later change. The test is removed because it uses !srl on a bit sequence, which cannot really be supported consistently, but luckily isn't used anywhere either. Change-Id: I98168bf52649176654ed2ec61a29bdb29970cfe7 Reviewers: arsenm, craig.topper, tra, MartinO Subscribers: wdng, llvm-commits Differential Revision: https://reviews.llvm.org/D43753 llvm-svn: 326785
2018-03-06 14:48:39 +01:00
if (Init *NewInit = LHS->convertInitializerTo(getType()))
return NewInit;
2009-05-14 23:22:49 +02:00
break;
case HEAD:
if (ListInit *LHSl = dyn_cast<ListInit>(LHS)) {
assert(!LHSl->empty() && "Empty list in head");
return LHSl->getElement(0);
}
break;
case TAIL:
if (ListInit *LHSl = dyn_cast<ListInit>(LHS)) {
assert(!LHSl->empty() && "Empty list in tail");
// Note the +1. We can't just pass the result of getValues()
// directly.
return ListInit::get(LHSl->getValues().slice(1), LHSl->getElementType());
}
break;
case SIZE:
if (ListInit *LHSl = dyn_cast<ListInit>(LHS))
return IntInit::get(LHSl->size());
break;
case EMPTY:
if (ListInit *LHSl = dyn_cast<ListInit>(LHS))
return IntInit::get(LHSl->empty());
if (StringInit *LHSs = dyn_cast<StringInit>(LHS))
return IntInit::get(LHSs->getValue().empty());
break;
case GETOP:
if (DagInit *Dag = dyn_cast<DagInit>(LHS)) {
DefInit *DI = DefInit::get(Dag->getOperatorAsDef({}));
if (!DI->getType()->typeIsA(getType())) {
PrintFatalError(CurRec->getLoc(),
Twine("Expected type '") +
getType()->getAsString() + "', got '" +
DI->getType()->getAsString() + "' in: " +
getAsString() + "\n");
} else {
return DI;
}
}
break;
}
return const_cast<UnOpInit *>(this);
2009-05-14 23:22:49 +02:00
}
Init *UnOpInit::resolveReferences(Resolver &R) const {
Init *lhs = LHS->resolveReferences(R);
2009-11-22 05:24:42 +01:00
if (LHS != lhs || (R.isFinal() && getOpcode() == CAST))
return (UnOpInit::get(getOpcode(), lhs, getType()))
->Fold(R.getCurrentRecord(), R.isFinal());
return const_cast<UnOpInit *>(this);
2009-05-14 23:22:49 +02:00
}
std::string UnOpInit::getAsString() const {
std::string Result;
switch (getOpcode()) {
2009-05-14 23:22:49 +02:00
case CAST: Result = "!cast<" + getType()->getAsString() + ">"; break;
case HEAD: Result = "!head"; break;
case TAIL: Result = "!tail"; break;
case SIZE: Result = "!size"; break;
case EMPTY: Result = "!empty"; break;
case GETOP: Result = "!getop"; break;
2009-05-14 23:22:49 +02:00
}
return Result + "(" + LHS->getAsString() + ")";
}
static void
ProfileBinOpInit(FoldingSetNodeID &ID, unsigned Opcode, Init *LHS, Init *RHS,
RecTy *Type) {
ID.AddInteger(Opcode);
ID.AddPointer(LHS);
ID.AddPointer(RHS);
ID.AddPointer(Type);
}
BinOpInit *BinOpInit::get(BinaryOp Opc, Init *LHS,
Init *RHS, RecTy *Type) {
static FoldingSet<BinOpInit> ThePool;
FoldingSetNodeID ID;
ProfileBinOpInit(ID, Opc, LHS, RHS, Type);
void *IP = nullptr;
if (BinOpInit *I = ThePool.FindNodeOrInsertPos(ID, IP))
return I;
BinOpInit *I = new(Allocator) BinOpInit(Opc, LHS, RHS, Type);
ThePool.InsertNode(I, IP);
return I;
}
void BinOpInit::Profile(FoldingSetNodeID &ID) const {
ProfileBinOpInit(ID, getOpcode(), getLHS(), getRHS(), getType());
}
static StringInit *ConcatStringInits(const StringInit *I0,
const StringInit *I1) {
SmallString<80> Concat(I0->getValue());
Concat.append(I1->getValue());
return StringInit::get(Concat);
}
Init *BinOpInit::getStrConcat(Init *I0, Init *I1) {
// Shortcut for the common case of concatenating two strings.
if (const StringInit *I0s = dyn_cast<StringInit>(I0))
if (const StringInit *I1s = dyn_cast<StringInit>(I1))
return ConcatStringInits(I0s, I1s);
return BinOpInit::get(BinOpInit::STRCONCAT, I0, I1, StringRecTy::get());
}
static ListInit *ConcatListInits(const ListInit *LHS,
const ListInit *RHS) {
SmallVector<Init *, 8> Args;
Args.insert(Args.end(), LHS->begin(), LHS->end());
Args.insert(Args.end(), RHS->begin(), RHS->end());
return ListInit::get(Args, LHS->getElementType());
}
Init *BinOpInit::getListConcat(TypedInit *LHS, Init *RHS) {
assert(isa<ListRecTy>(LHS->getType()) && "First arg must be a list");
// Shortcut for the common case of concatenating two lists.
if (const ListInit *LHSList = dyn_cast<ListInit>(LHS))
if (const ListInit *RHSList = dyn_cast<ListInit>(RHS))
return ConcatListInits(LHSList, RHSList);
return BinOpInit::get(BinOpInit::LISTCONCAT, LHS, RHS, LHS->getType());
}
Init *BinOpInit::getListSplat(TypedInit *LHS, Init *RHS) {
return BinOpInit::get(BinOpInit::LISTSPLAT, LHS, RHS, LHS->getType());
}
Init *BinOpInit::Fold(Record *CurRec) const {
switch (getOpcode()) {
case CONCAT: {
DagInit *LHSs = dyn_cast<DagInit>(LHS);
DagInit *RHSs = dyn_cast<DagInit>(RHS);
if (LHSs && RHSs) {
DefInit *LOp = dyn_cast<DefInit>(LHSs->getOperator());
DefInit *ROp = dyn_cast<DefInit>(RHSs->getOperator());
if ((!LOp && !isa<UnsetInit>(LHSs->getOperator())) ||
(!ROp && !isa<UnsetInit>(RHSs->getOperator())))
break;
if (LOp && ROp && LOp->getDef() != ROp->getDef()) {
PrintFatalError(Twine("Concatenated Dag operators do not match: '") +
LHSs->getAsString() + "' vs. '" + RHSs->getAsString() +
"'");
}
Init *Op = LOp ? LOp : ROp;
if (!Op)
Op = UnsetInit::get();
SmallVector<Init*, 8> Args;
SmallVector<StringInit*, 8> ArgNames;
for (unsigned i = 0, e = LHSs->getNumArgs(); i != e; ++i) {
Args.push_back(LHSs->getArg(i));
ArgNames.push_back(LHSs->getArgName(i));
}
for (unsigned i = 0, e = RHSs->getNumArgs(); i != e; ++i) {
Args.push_back(RHSs->getArg(i));
ArgNames.push_back(RHSs->getArgName(i));
}
return DagInit::get(Op, nullptr, Args, ArgNames);
}
break;
}
case LISTCONCAT: {
ListInit *LHSs = dyn_cast<ListInit>(LHS);
ListInit *RHSs = dyn_cast<ListInit>(RHS);
if (LHSs && RHSs) {
SmallVector<Init *, 8> Args;
Args.insert(Args.end(), LHSs->begin(), LHSs->end());
Args.insert(Args.end(), RHSs->begin(), RHSs->end());
return ListInit::get(Args, LHSs->getElementType());
}
break;
}
case LISTSPLAT: {
TypedInit *Value = dyn_cast<TypedInit>(LHS);
IntInit *Size = dyn_cast<IntInit>(RHS);
if (Value && Size) {
SmallVector<Init *, 8> Args(Size->getValue(), Value);
return ListInit::get(Args, Value->getType());
}
break;
}
case STRCONCAT: {
StringInit *LHSs = dyn_cast<StringInit>(LHS);
StringInit *RHSs = dyn_cast<StringInit>(RHS);
if (LHSs && RHSs)
return ConcatStringInits(LHSs, RHSs);
break;
}
case EQ:
case NE:
case LE:
case LT:
case GE:
case GT: {
// try to fold eq comparison for 'bit' and 'int', otherwise fallback
// to string objects.
IntInit *L =
dyn_cast_or_null<IntInit>(LHS->convertInitializerTo(IntRecTy::get()));
IntInit *R =
dyn_cast_or_null<IntInit>(RHS->convertInitializerTo(IntRecTy::get()));
if (L && R) {
bool Result;
switch (getOpcode()) {
case EQ: Result = L->getValue() == R->getValue(); break;
case NE: Result = L->getValue() != R->getValue(); break;
case LE: Result = L->getValue() <= R->getValue(); break;
case LT: Result = L->getValue() < R->getValue(); break;
case GE: Result = L->getValue() >= R->getValue(); break;
case GT: Result = L->getValue() > R->getValue(); break;
default: llvm_unreachable("unhandled comparison");
}
return BitInit::get(Result);
}
if (getOpcode() == EQ || getOpcode() == NE) {
StringInit *LHSs = dyn_cast<StringInit>(LHS);
StringInit *RHSs = dyn_cast<StringInit>(RHS);
// Make sure we've resolved
if (LHSs && RHSs) {
bool Equal = LHSs->getValue() == RHSs->getValue();
return BitInit::get(getOpcode() == EQ ? Equal : !Equal);
}
}
break;
}
case SETOP: {
DagInit *Dag = dyn_cast<DagInit>(LHS);
DefInit *Op = dyn_cast<DefInit>(RHS);
if (Dag && Op) {
SmallVector<Init*, 8> Args;
SmallVector<StringInit*, 8> ArgNames;
for (unsigned i = 0, e = Dag->getNumArgs(); i != e; ++i) {
Args.push_back(Dag->getArg(i));
ArgNames.push_back(Dag->getArgName(i));
}
return DagInit::get(Op, nullptr, Args, ArgNames);
}
break;
}
case ADD:
case MUL:
case AND:
case OR:
case SHL:
case SRA:
case SRL: {
IntInit *LHSi =
dyn_cast_or_null<IntInit>(LHS->convertInitializerTo(IntRecTy::get()));
IntInit *RHSi =
dyn_cast_or_null<IntInit>(RHS->convertInitializerTo(IntRecTy::get()));
if (LHSi && RHSi) {
int64_t LHSv = LHSi->getValue(), RHSv = RHSi->getValue();
int64_t Result;
switch (getOpcode()) {
default: llvm_unreachable("Bad opcode!");
case ADD: Result = LHSv + RHSv; break;
case MUL: Result = LHSv * RHSv; break;
case AND: Result = LHSv & RHSv; break;
case OR: Result = LHSv | RHSv; break;
case SHL: Result = LHSv << RHSv; break;
case SRA: Result = LHSv >> RHSv; break;
case SRL: Result = (uint64_t)LHSv >> (uint64_t)RHSv; break;
}
return IntInit::get(Result);
}
break;
}
}
return const_cast<BinOpInit *>(this);
}
Init *BinOpInit::resolveReferences(Resolver &R) const {
Init *lhs = LHS->resolveReferences(R);
Init *rhs = RHS->resolveReferences(R);
2009-11-22 05:24:42 +01:00
if (LHS != lhs || RHS != rhs)
return (BinOpInit::get(getOpcode(), lhs, rhs, getType()))
->Fold(R.getCurrentRecord());
return const_cast<BinOpInit *>(this);
}
std::string BinOpInit::getAsString() const {
std::string Result;
switch (getOpcode()) {
case CONCAT: Result = "!con"; break;
case ADD: Result = "!add"; break;
case MUL: Result = "!mul"; break;
case AND: Result = "!and"; break;
case OR: Result = "!or"; break;
case SHL: Result = "!shl"; break;
case SRA: Result = "!sra"; break;
case SRL: Result = "!srl"; break;
case EQ: Result = "!eq"; break;
case NE: Result = "!ne"; break;
case LE: Result = "!le"; break;
case LT: Result = "!lt"; break;
case GE: Result = "!ge"; break;
case GT: Result = "!gt"; break;
case LISTCONCAT: Result = "!listconcat"; break;
case LISTSPLAT: Result = "!listsplat"; break;
case STRCONCAT: Result = "!strconcat"; break;
case SETOP: Result = "!setop"; break;
}
return Result + "(" + LHS->getAsString() + ", " + RHS->getAsString() + ")";
}
static void
ProfileTernOpInit(FoldingSetNodeID &ID, unsigned Opcode, Init *LHS, Init *MHS,
Init *RHS, RecTy *Type) {
ID.AddInteger(Opcode);
ID.AddPointer(LHS);
ID.AddPointer(MHS);
ID.AddPointer(RHS);
ID.AddPointer(Type);
}
TernOpInit *TernOpInit::get(TernaryOp Opc, Init *LHS, Init *MHS, Init *RHS,
RecTy *Type) {
static FoldingSet<TernOpInit> ThePool;
FoldingSetNodeID ID;
ProfileTernOpInit(ID, Opc, LHS, MHS, RHS, Type);
void *IP = nullptr;
if (TernOpInit *I = ThePool.FindNodeOrInsertPos(ID, IP))
return I;
TernOpInit *I = new(Allocator) TernOpInit(Opc, LHS, MHS, RHS, Type);
ThePool.InsertNode(I, IP);
return I;
}
void TernOpInit::Profile(FoldingSetNodeID &ID) const {
ProfileTernOpInit(ID, getOpcode(), getLHS(), getMHS(), getRHS(), getType());
}
static Init *ForeachApply(Init *LHS, Init *MHSe, Init *RHS, Record *CurRec) {
MapResolver R(CurRec);
R.set(LHS, MHSe);
return RHS->resolveReferences(R);
}
static Init *ForeachDagApply(Init *LHS, DagInit *MHSd, Init *RHS,
Record *CurRec) {
bool Change = false;
Init *Val = ForeachApply(LHS, MHSd->getOperator(), RHS, CurRec);
if (Val != MHSd->getOperator())
Change = true;
SmallVector<std::pair<Init *, StringInit *>, 8> NewArgs;
for (unsigned int i = 0; i < MHSd->getNumArgs(); ++i) {
Init *Arg = MHSd->getArg(i);
Init *NewArg;
StringInit *ArgName = MHSd->getArgName(i);
if (DagInit *Argd = dyn_cast<DagInit>(Arg))
NewArg = ForeachDagApply(LHS, Argd, RHS, CurRec);
else
NewArg = ForeachApply(LHS, Arg, RHS, CurRec);
NewArgs.push_back(std::make_pair(NewArg, ArgName));
if (Arg != NewArg)
Change = true;
}
if (Change)
return DagInit::get(Val, nullptr, NewArgs);
return MHSd;
}
// Applies RHS to all elements of MHS, using LHS as a temp variable.
static Init *ForeachHelper(Init *LHS, Init *MHS, Init *RHS, RecTy *Type,
Record *CurRec) {
if (DagInit *MHSd = dyn_cast<DagInit>(MHS))
return ForeachDagApply(LHS, MHSd, RHS, CurRec);
if (ListInit *MHSl = dyn_cast<ListInit>(MHS)) {
SmallVector<Init *, 8> NewList(MHSl->begin(), MHSl->end());
for (Init *&Item : NewList) {
Init *NewItem = ForeachApply(LHS, Item, RHS, CurRec);
if (NewItem != Item)
Item = NewItem;
}
return ListInit::get(NewList, cast<ListRecTy>(Type)->getElementType());
}
return nullptr;
}
Init *TernOpInit::Fold(Record *CurRec) const {
switch (getOpcode()) {
case SUBST: {
DefInit *LHSd = dyn_cast<DefInit>(LHS);
VarInit *LHSv = dyn_cast<VarInit>(LHS);
StringInit *LHSs = dyn_cast<StringInit>(LHS);
DefInit *MHSd = dyn_cast<DefInit>(MHS);
VarInit *MHSv = dyn_cast<VarInit>(MHS);
StringInit *MHSs = dyn_cast<StringInit>(MHS);
DefInit *RHSd = dyn_cast<DefInit>(RHS);
VarInit *RHSv = dyn_cast<VarInit>(RHS);
StringInit *RHSs = dyn_cast<StringInit>(RHS);
if (LHSd && MHSd && RHSd) {
Record *Val = RHSd->getDef();
if (LHSd->getAsString() == RHSd->getAsString())
Val = MHSd->getDef();
return DefInit::get(Val);
}
if (LHSv && MHSv && RHSv) {
std::string Val = std::string(RHSv->getName());
if (LHSv->getAsString() == RHSv->getAsString())
Val = std::string(MHSv->getName());
return VarInit::get(Val, getType());
}
if (LHSs && MHSs && RHSs) {
std::string Val = std::string(RHSs->getValue());
std::string::size_type found;
std::string::size_type idx = 0;
while (true) {
found = Val.find(std::string(LHSs->getValue()), idx);
if (found == std::string::npos)
break;
Val.replace(found, LHSs->getValue().size(),
std::string(MHSs->getValue()));
idx = found + MHSs->getValue().size();
}
return StringInit::get(Val);
}
break;
2009-11-22 05:24:42 +01:00
}
case FOREACH: {
if (Init *Result = ForeachHelper(LHS, MHS, RHS, getType(), CurRec))
return Result;
break;
}
case IF: {
TableGen: Allow !cast of records, cleanup conversion machinery Summary: Distinguish two relationships between types: is-a and convertible-to. For example, a bit is not an int or vice versa, but they can be converted into each other (with range checks that you can think of as "dynamic": unlike other type checks, those range checks do not happen during parsing, but only once the final values have been established). Actually converting initializers between types is subtle: even when values of type A can be converted to type B (e.g. int into string), it may not be possible to do so with a concrete initializer (e.g., a VarInit that refers to a variable of type int cannot be immediately converted to a string). For this reason, distinguish between getCastTo and convertInitializerTo: the latter implements the actual conversion when appropriate, while the former will first try to do the actual conversion and fall back to introducing a !cast operation so that the conversion will be delayed until variable references have been resolved. To make the approach of adding !cast operations to work, !cast needs to fallback to convertInitializerTo when the special string <-> record logic does not apply. This enables casting records to a subclass, although that new functionality is only truly useful together with !isa, which will be added in a later change. The test is removed because it uses !srl on a bit sequence, which cannot really be supported consistently, but luckily isn't used anywhere either. Change-Id: I98168bf52649176654ed2ec61a29bdb29970cfe7 Reviewers: arsenm, craig.topper, tra, MartinO Subscribers: wdng, llvm-commits Differential Revision: https://reviews.llvm.org/D43753 llvm-svn: 326785
2018-03-06 14:48:39 +01:00
if (IntInit *LHSi = dyn_cast_or_null<IntInit>(
LHS->convertInitializerTo(IntRecTy::get()))) {
if (LHSi->getValue())
return MHS;
return RHS;
}
break;
}
case DAG: {
ListInit *MHSl = dyn_cast<ListInit>(MHS);
ListInit *RHSl = dyn_cast<ListInit>(RHS);
bool MHSok = MHSl || isa<UnsetInit>(MHS);
bool RHSok = RHSl || isa<UnsetInit>(RHS);
if (isa<UnsetInit>(MHS) && isa<UnsetInit>(RHS))
break; // Typically prevented by the parser, but might happen with template args
if (MHSok && RHSok && (!MHSl || !RHSl || MHSl->size() == RHSl->size())) {
SmallVector<std::pair<Init *, StringInit *>, 8> Children;
unsigned Size = MHSl ? MHSl->size() : RHSl->size();
for (unsigned i = 0; i != Size; ++i) {
Init *Node = MHSl ? MHSl->getElement(i) : UnsetInit::get();
Init *Name = RHSl ? RHSl->getElement(i) : UnsetInit::get();
if (!isa<StringInit>(Name) && !isa<UnsetInit>(Name))
return const_cast<TernOpInit *>(this);
Children.emplace_back(Node, dyn_cast<StringInit>(Name));
}
return DagInit::get(LHS, nullptr, Children);
}
break;
}
}
return const_cast<TernOpInit *>(this);
}
Init *TernOpInit::resolveReferences(Resolver &R) const {
Init *lhs = LHS->resolveReferences(R);
if (getOpcode() == IF && lhs != LHS) {
TableGen: Allow !cast of records, cleanup conversion machinery Summary: Distinguish two relationships between types: is-a and convertible-to. For example, a bit is not an int or vice versa, but they can be converted into each other (with range checks that you can think of as "dynamic": unlike other type checks, those range checks do not happen during parsing, but only once the final values have been established). Actually converting initializers between types is subtle: even when values of type A can be converted to type B (e.g. int into string), it may not be possible to do so with a concrete initializer (e.g., a VarInit that refers to a variable of type int cannot be immediately converted to a string). For this reason, distinguish between getCastTo and convertInitializerTo: the latter implements the actual conversion when appropriate, while the former will first try to do the actual conversion and fall back to introducing a !cast operation so that the conversion will be delayed until variable references have been resolved. To make the approach of adding !cast operations to work, !cast needs to fallback to convertInitializerTo when the special string <-> record logic does not apply. This enables casting records to a subclass, although that new functionality is only truly useful together with !isa, which will be added in a later change. The test is removed because it uses !srl on a bit sequence, which cannot really be supported consistently, but luckily isn't used anywhere either. Change-Id: I98168bf52649176654ed2ec61a29bdb29970cfe7 Reviewers: arsenm, craig.topper, tra, MartinO Subscribers: wdng, llvm-commits Differential Revision: https://reviews.llvm.org/D43753 llvm-svn: 326785
2018-03-06 14:48:39 +01:00
if (IntInit *Value = dyn_cast_or_null<IntInit>(
lhs->convertInitializerTo(IntRecTy::get()))) {
// Short-circuit
TableGen: Allow !cast of records, cleanup conversion machinery Summary: Distinguish two relationships between types: is-a and convertible-to. For example, a bit is not an int or vice versa, but they can be converted into each other (with range checks that you can think of as "dynamic": unlike other type checks, those range checks do not happen during parsing, but only once the final values have been established). Actually converting initializers between types is subtle: even when values of type A can be converted to type B (e.g. int into string), it may not be possible to do so with a concrete initializer (e.g., a VarInit that refers to a variable of type int cannot be immediately converted to a string). For this reason, distinguish between getCastTo and convertInitializerTo: the latter implements the actual conversion when appropriate, while the former will first try to do the actual conversion and fall back to introducing a !cast operation so that the conversion will be delayed until variable references have been resolved. To make the approach of adding !cast operations to work, !cast needs to fallback to convertInitializerTo when the special string <-> record logic does not apply. This enables casting records to a subclass, although that new functionality is only truly useful together with !isa, which will be added in a later change. The test is removed because it uses !srl on a bit sequence, which cannot really be supported consistently, but luckily isn't used anywhere either. Change-Id: I98168bf52649176654ed2ec61a29bdb29970cfe7 Reviewers: arsenm, craig.topper, tra, MartinO Subscribers: wdng, llvm-commits Differential Revision: https://reviews.llvm.org/D43753 llvm-svn: 326785
2018-03-06 14:48:39 +01:00
if (Value->getValue())
return MHS->resolveReferences(R);
return RHS->resolveReferences(R);
}
}
2009-11-22 05:24:42 +01:00
Init *mhs = MHS->resolveReferences(R);
Init *rhs;
if (getOpcode() == FOREACH) {
ShadowResolver SR(R);
SR.addShadow(lhs);
rhs = RHS->resolveReferences(SR);
} else {
rhs = RHS->resolveReferences(R);
}
if (LHS != lhs || MHS != mhs || RHS != rhs)
return (TernOpInit::get(getOpcode(), lhs, mhs, rhs, getType()))
->Fold(R.getCurrentRecord());
return const_cast<TernOpInit *>(this);
}
std::string TernOpInit::getAsString() const {
std::string Result;
bool UnquotedLHS = false;
switch (getOpcode()) {
case SUBST: Result = "!subst"; break;
case FOREACH: Result = "!foreach"; UnquotedLHS = true; break;
2009-11-22 05:24:42 +01:00
case IF: Result = "!if"; break;
case DAG: Result = "!dag"; break;
}
return (Result + "(" +
(UnquotedLHS ? LHS->getAsUnquotedString() : LHS->getAsString()) +
", " + MHS->getAsString() + ", " + RHS->getAsString() + ")");
}
static void ProfileFoldOpInit(FoldingSetNodeID &ID, Init *A, Init *B,
Init *Start, Init *List, Init *Expr,
RecTy *Type) {
ID.AddPointer(Start);
ID.AddPointer(List);
ID.AddPointer(A);
ID.AddPointer(B);
ID.AddPointer(Expr);
ID.AddPointer(Type);
}
FoldOpInit *FoldOpInit::get(Init *Start, Init *List, Init *A, Init *B,
Init *Expr, RecTy *Type) {
static FoldingSet<FoldOpInit> ThePool;
FoldingSetNodeID ID;
ProfileFoldOpInit(ID, Start, List, A, B, Expr, Type);
void *IP = nullptr;
if (FoldOpInit *I = ThePool.FindNodeOrInsertPos(ID, IP))
return I;
FoldOpInit *I = new (Allocator) FoldOpInit(Start, List, A, B, Expr, Type);
ThePool.InsertNode(I, IP);
return I;
}
void FoldOpInit::Profile(FoldingSetNodeID &ID) const {
ProfileFoldOpInit(ID, Start, List, A, B, Expr, getType());
}
Init *FoldOpInit::Fold(Record *CurRec) const {
if (ListInit *LI = dyn_cast<ListInit>(List)) {
Init *Accum = Start;
for (Init *Elt : *LI) {
MapResolver R(CurRec);
R.set(A, Accum);
R.set(B, Elt);
Accum = Expr->resolveReferences(R);
}
return Accum;
}
return const_cast<FoldOpInit *>(this);
}
Init *FoldOpInit::resolveReferences(Resolver &R) const {
Init *NewStart = Start->resolveReferences(R);
Init *NewList = List->resolveReferences(R);
ShadowResolver SR(R);
SR.addShadow(A);
SR.addShadow(B);
Init *NewExpr = Expr->resolveReferences(SR);
if (Start == NewStart && List == NewList && Expr == NewExpr)
return const_cast<FoldOpInit *>(this);
return get(NewStart, NewList, A, B, NewExpr, getType())
->Fold(R.getCurrentRecord());
}
Init *FoldOpInit::getBit(unsigned Bit) const {
return VarBitInit::get(const_cast<FoldOpInit *>(this), Bit);
}
std::string FoldOpInit::getAsString() const {
return (Twine("!foldl(") + Start->getAsString() + ", " + List->getAsString() +
", " + A->getAsUnquotedString() + ", " + B->getAsUnquotedString() +
", " + Expr->getAsString() + ")")
.str();
}
static void ProfileIsAOpInit(FoldingSetNodeID &ID, RecTy *CheckType,
Init *Expr) {
ID.AddPointer(CheckType);
ID.AddPointer(Expr);
}
IsAOpInit *IsAOpInit::get(RecTy *CheckType, Init *Expr) {
static FoldingSet<IsAOpInit> ThePool;
FoldingSetNodeID ID;
ProfileIsAOpInit(ID, CheckType, Expr);
void *IP = nullptr;
if (IsAOpInit *I = ThePool.FindNodeOrInsertPos(ID, IP))
return I;
IsAOpInit *I = new (Allocator) IsAOpInit(CheckType, Expr);
ThePool.InsertNode(I, IP);
return I;
}
void IsAOpInit::Profile(FoldingSetNodeID &ID) const {
ProfileIsAOpInit(ID, CheckType, Expr);
}
Init *IsAOpInit::Fold() const {
if (TypedInit *TI = dyn_cast<TypedInit>(Expr)) {
// Is the expression type known to be (a subclass of) the desired type?
if (TI->getType()->typeIsConvertibleTo(CheckType))
return IntInit::get(1);
if (isa<RecordRecTy>(CheckType)) {
// If the target type is not a subclass of the expression type, or if
// the expression has fully resolved to a record, we know that it can't
// be of the required type.
if (!CheckType->typeIsConvertibleTo(TI->getType()) || isa<DefInit>(Expr))
return IntInit::get(0);
} else {
// We treat non-record types as not castable.
return IntInit::get(0);
}
}
return const_cast<IsAOpInit *>(this);
}
Init *IsAOpInit::resolveReferences(Resolver &R) const {
Init *NewExpr = Expr->resolveReferences(R);
if (Expr != NewExpr)
return get(CheckType, NewExpr)->Fold();
return const_cast<IsAOpInit *>(this);
}
Init *IsAOpInit::getBit(unsigned Bit) const {
return VarBitInit::get(const_cast<IsAOpInit *>(this), Bit);
}
std::string IsAOpInit::getAsString() const {
return (Twine("!isa<") + CheckType->getAsString() + ">(" +
Expr->getAsString() + ")")
.str();
}
RecTy *TypedInit::getFieldType(StringInit *FieldName) const {
if (RecordRecTy *RecordType = dyn_cast<RecordRecTy>(getType())) {
for (Record *Rec : RecordType->getClasses()) {
if (RecordVal *Field = Rec->getValue(FieldName))
return Field->getType();
}
}
return nullptr;
}
Init *
TypedInit::convertInitializerTo(RecTy *Ty) const {
TableGen: Allow !cast of records, cleanup conversion machinery Summary: Distinguish two relationships between types: is-a and convertible-to. For example, a bit is not an int or vice versa, but they can be converted into each other (with range checks that you can think of as "dynamic": unlike other type checks, those range checks do not happen during parsing, but only once the final values have been established). Actually converting initializers between types is subtle: even when values of type A can be converted to type B (e.g. int into string), it may not be possible to do so with a concrete initializer (e.g., a VarInit that refers to a variable of type int cannot be immediately converted to a string). For this reason, distinguish between getCastTo and convertInitializerTo: the latter implements the actual conversion when appropriate, while the former will first try to do the actual conversion and fall back to introducing a !cast operation so that the conversion will be delayed until variable references have been resolved. To make the approach of adding !cast operations to work, !cast needs to fallback to convertInitializerTo when the special string <-> record logic does not apply. This enables casting records to a subclass, although that new functionality is only truly useful together with !isa, which will be added in a later change. The test is removed because it uses !srl on a bit sequence, which cannot really be supported consistently, but luckily isn't used anywhere either. Change-Id: I98168bf52649176654ed2ec61a29bdb29970cfe7 Reviewers: arsenm, craig.topper, tra, MartinO Subscribers: wdng, llvm-commits Differential Revision: https://reviews.llvm.org/D43753 llvm-svn: 326785
2018-03-06 14:48:39 +01:00
if (getType() == Ty || getType()->typeIsA(Ty))
return const_cast<TypedInit *>(this);
TableGen: Allow !cast of records, cleanup conversion machinery Summary: Distinguish two relationships between types: is-a and convertible-to. For example, a bit is not an int or vice versa, but they can be converted into each other (with range checks that you can think of as "dynamic": unlike other type checks, those range checks do not happen during parsing, but only once the final values have been established). Actually converting initializers between types is subtle: even when values of type A can be converted to type B (e.g. int into string), it may not be possible to do so with a concrete initializer (e.g., a VarInit that refers to a variable of type int cannot be immediately converted to a string). For this reason, distinguish between getCastTo and convertInitializerTo: the latter implements the actual conversion when appropriate, while the former will first try to do the actual conversion and fall back to introducing a !cast operation so that the conversion will be delayed until variable references have been resolved. To make the approach of adding !cast operations to work, !cast needs to fallback to convertInitializerTo when the special string <-> record logic does not apply. This enables casting records to a subclass, although that new functionality is only truly useful together with !isa, which will be added in a later change. The test is removed because it uses !srl on a bit sequence, which cannot really be supported consistently, but luckily isn't used anywhere either. Change-Id: I98168bf52649176654ed2ec61a29bdb29970cfe7 Reviewers: arsenm, craig.topper, tra, MartinO Subscribers: wdng, llvm-commits Differential Revision: https://reviews.llvm.org/D43753 llvm-svn: 326785
2018-03-06 14:48:39 +01:00
if (isa<BitRecTy>(getType()) && isa<BitsRecTy>(Ty) &&
cast<BitsRecTy>(Ty)->getNumBits() == 1)
return BitsInit::get({const_cast<TypedInit *>(this)});
return nullptr;
}
Init *TypedInit::convertInitializerBitRange(ArrayRef<unsigned> Bits) const {
BitsRecTy *T = dyn_cast<BitsRecTy>(getType());
if (!T) return nullptr; // Cannot subscript a non-bits variable.
unsigned NumBits = T->getNumBits();
SmallVector<Init *, 16> NewBits;
NewBits.reserve(Bits.size());
for (unsigned Bit : Bits) {
if (Bit >= NumBits)
return nullptr;
NewBits.push_back(VarBitInit::get(const_cast<TypedInit *>(this), Bit));
}
return BitsInit::get(NewBits);
}
TableGen: Allow !cast of records, cleanup conversion machinery Summary: Distinguish two relationships between types: is-a and convertible-to. For example, a bit is not an int or vice versa, but they can be converted into each other (with range checks that you can think of as "dynamic": unlike other type checks, those range checks do not happen during parsing, but only once the final values have been established). Actually converting initializers between types is subtle: even when values of type A can be converted to type B (e.g. int into string), it may not be possible to do so with a concrete initializer (e.g., a VarInit that refers to a variable of type int cannot be immediately converted to a string). For this reason, distinguish between getCastTo and convertInitializerTo: the latter implements the actual conversion when appropriate, while the former will first try to do the actual conversion and fall back to introducing a !cast operation so that the conversion will be delayed until variable references have been resolved. To make the approach of adding !cast operations to work, !cast needs to fallback to convertInitializerTo when the special string <-> record logic does not apply. This enables casting records to a subclass, although that new functionality is only truly useful together with !isa, which will be added in a later change. The test is removed because it uses !srl on a bit sequence, which cannot really be supported consistently, but luckily isn't used anywhere either. Change-Id: I98168bf52649176654ed2ec61a29bdb29970cfe7 Reviewers: arsenm, craig.topper, tra, MartinO Subscribers: wdng, llvm-commits Differential Revision: https://reviews.llvm.org/D43753 llvm-svn: 326785
2018-03-06 14:48:39 +01:00
Init *TypedInit::getCastTo(RecTy *Ty) const {
// Handle the common case quickly
if (getType() == Ty || getType()->typeIsA(Ty))
return const_cast<TypedInit *>(this);
if (Init *Converted = convertInitializerTo(Ty)) {
assert(!isa<TypedInit>(Converted) ||
cast<TypedInit>(Converted)->getType()->typeIsA(Ty));
return Converted;
}
if (!getType()->typeIsConvertibleTo(Ty))
return nullptr;
return UnOpInit::get(UnOpInit::CAST, const_cast<TypedInit *>(this), Ty)
->Fold(nullptr);
TableGen: Allow !cast of records, cleanup conversion machinery Summary: Distinguish two relationships between types: is-a and convertible-to. For example, a bit is not an int or vice versa, but they can be converted into each other (with range checks that you can think of as "dynamic": unlike other type checks, those range checks do not happen during parsing, but only once the final values have been established). Actually converting initializers between types is subtle: even when values of type A can be converted to type B (e.g. int into string), it may not be possible to do so with a concrete initializer (e.g., a VarInit that refers to a variable of type int cannot be immediately converted to a string). For this reason, distinguish between getCastTo and convertInitializerTo: the latter implements the actual conversion when appropriate, while the former will first try to do the actual conversion and fall back to introducing a !cast operation so that the conversion will be delayed until variable references have been resolved. To make the approach of adding !cast operations to work, !cast needs to fallback to convertInitializerTo when the special string <-> record logic does not apply. This enables casting records to a subclass, although that new functionality is only truly useful together with !isa, which will be added in a later change. The test is removed because it uses !srl on a bit sequence, which cannot really be supported consistently, but luckily isn't used anywhere either. Change-Id: I98168bf52649176654ed2ec61a29bdb29970cfe7 Reviewers: arsenm, craig.topper, tra, MartinO Subscribers: wdng, llvm-commits Differential Revision: https://reviews.llvm.org/D43753 llvm-svn: 326785
2018-03-06 14:48:39 +01:00
}
Init *TypedInit::convertInitListSlice(ArrayRef<unsigned> Elements) const {
ListRecTy *T = dyn_cast<ListRecTy>(getType());
if (!T) return nullptr; // Cannot subscript a non-list variable.
if (Elements.size() == 1)
return VarListElementInit::get(const_cast<TypedInit *>(this), Elements[0]);
SmallVector<Init*, 8> ListInits;
ListInits.reserve(Elements.size());
for (unsigned Element : Elements)
ListInits.push_back(VarListElementInit::get(const_cast<TypedInit *>(this),
Element));
return ListInit::get(ListInits, T->getElementType());
}
VarInit *VarInit::get(StringRef VN, RecTy *T) {
Init *Value = StringInit::get(VN);
return VarInit::get(Value, T);
}
VarInit *VarInit::get(Init *VN, RecTy *T) {
using Key = std::pair<RecTy *, Init *>;
static DenseMap<Key, VarInit*> ThePool;
Key TheKey(std::make_pair(T, VN));
VarInit *&I = ThePool[TheKey];
if (!I)
I = new(Allocator) VarInit(VN, T);
return I;
}
StringRef VarInit::getName() const {
StringInit *NameString = cast<StringInit>(getNameInit());
return NameString->getValue();
}
Init *VarInit::getBit(unsigned Bit) const {
if (getType() == BitRecTy::get())
return const_cast<VarInit*>(this);
return VarBitInit::get(const_cast<VarInit*>(this), Bit);
}
Init *VarInit::resolveReferences(Resolver &R) const {
if (Init *Val = R.resolve(VarName))
return Val;
return const_cast<VarInit *>(this);
}
VarBitInit *VarBitInit::get(TypedInit *T, unsigned B) {
using Key = std::pair<TypedInit *, unsigned>;
static DenseMap<Key, VarBitInit*> ThePool;
Key TheKey(std::make_pair(T, B));
VarBitInit *&I = ThePool[TheKey];
if (!I)
I = new(Allocator) VarBitInit(T, B);
return I;
}
std::string VarBitInit::getAsString() const {
return TI->getAsString() + "{" + utostr(Bit) + "}";
}
Init *VarBitInit::resolveReferences(Resolver &R) const {
Init *I = TI->resolveReferences(R);
if (TI != I)
return I->getBit(getBitNum());
return const_cast<VarBitInit*>(this);
}
VarListElementInit *VarListElementInit::get(TypedInit *T,
unsigned E) {
using Key = std::pair<TypedInit *, unsigned>;
static DenseMap<Key, VarListElementInit*> ThePool;
Key TheKey(std::make_pair(T, E));
VarListElementInit *&I = ThePool[TheKey];
if (!I) I = new(Allocator) VarListElementInit(T, E);
return I;
}
std::string VarListElementInit::getAsString() const {
return TI->getAsString() + "[" + utostr(Element) + "]";
}
Init *VarListElementInit::resolveReferences(Resolver &R) const {
Init *NewTI = TI->resolveReferences(R);
if (ListInit *List = dyn_cast<ListInit>(NewTI)) {
// Leave out-of-bounds array references as-is. This can happen without
// being an error, e.g. in the untaken "branch" of an !if expression.
if (getElementNum() < List->size())
return List->getElement(getElementNum());
}
if (NewTI != TI && isa<TypedInit>(NewTI))
return VarListElementInit::get(cast<TypedInit>(NewTI), getElementNum());
return const_cast<VarListElementInit *>(this);
}
Init *VarListElementInit::getBit(unsigned Bit) const {
if (getType() == BitRecTy::get())
return const_cast<VarListElementInit*>(this);
return VarBitInit::get(const_cast<VarListElementInit*>(this), Bit);
}
DefInit::DefInit(Record *D)
TableGen: Streamline how defs are instantiated Summary: Instantiating def's and defm's needs to perform the following steps: - for defm's, clone multiclass def prototypes and subsitute template args - for def's and defm's, add subclass definitions, substituting template args - clone the record based on foreach loops and substitute loop iteration variables - override record variables based on the global 'let' stack - resolve the record name (this should be simple, but unfortunately it's not due to existing .td files relying on rather silly implementation details) - for def(m)s in multiclasses, add the unresolved record as a multiclass prototype - for top-level def(m)s, resolve all internal variable references and add them to the record keeper and any active defsets This change streamlines how we go through these steps, by having both def's and defm's feed into a single addDef() method that handles foreach, final resolve, and routing the record to the right place. This happens to make foreach inside of multiclasses work, as the new test case demonstrates. Previously, foreach inside multiclasses was not forbidden by the parser, but it was de facto broken. Another side effect is that the order of "instantiated from" notes in error messages is reversed, as the modified test case shows. This is arguably clearer, since the initial error message ends up pointing directly to whatever triggered the error, and subsequent notes will point to increasingly outer layers of multiclasses. This is consistent with how C++ compilers report nested #includes and nested template instantiations. Change-Id: Ica146d0db2bc133dd7ed88054371becf24320447 Reviewers: arsenm, craig.topper, tra, MartinO Subscribers: wdng, llvm-commits Differential Revision: https://reviews.llvm.org/D44478 llvm-svn: 328117
2018-03-21 18:12:53 +01:00
: TypedInit(IK_DefInit, D->getType()), Def(D) {}
DefInit *DefInit::get(Record *R) {
return R->getDefInit();
}
Init *DefInit::convertInitializerTo(RecTy *Ty) const {
if (auto *RRT = dyn_cast<RecordRecTy>(Ty))
if (getType()->typeIsConvertibleTo(RRT))
return const_cast<DefInit *>(this);
return nullptr;
}
RecTy *DefInit::getFieldType(StringInit *FieldName) const {
if (const RecordVal *RV = Def->getValue(FieldName))
return RV->getType();
return nullptr;
}
std::string DefInit::getAsString() const { return std::string(Def->getName()); }
static void ProfileVarDefInit(FoldingSetNodeID &ID,
Record *Class,
ArrayRef<Init *> Args) {
ID.AddInteger(Args.size());
ID.AddPointer(Class);
for (Init *I : Args)
ID.AddPointer(I);
}
VarDefInit *VarDefInit::get(Record *Class, ArrayRef<Init *> Args) {
static FoldingSet<VarDefInit> ThePool;
FoldingSetNodeID ID;
ProfileVarDefInit(ID, Class, Args);
void *IP = nullptr;
if (VarDefInit *I = ThePool.FindNodeOrInsertPos(ID, IP))
return I;
void *Mem = Allocator.Allocate(totalSizeToAlloc<Init *>(Args.size()),
alignof(VarDefInit));
VarDefInit *I = new(Mem) VarDefInit(Class, Args.size());
std::uninitialized_copy(Args.begin(), Args.end(),
I->getTrailingObjects<Init *>());
ThePool.InsertNode(I, IP);
return I;
}
void VarDefInit::Profile(FoldingSetNodeID &ID) const {
ProfileVarDefInit(ID, Class, args());
}
DefInit *VarDefInit::instantiate() {
if (!Def) {
RecordKeeper &Records = Class->getRecords();
auto NewRecOwner = std::make_unique<Record>(Records.getNewAnonymousName(),
Class->getLoc(), Records,
/*IsAnonymous=*/true);
Record *NewRec = NewRecOwner.get();
// Copy values from class to instance
TableGen: Streamline the semantics of NAME Summary: The new rules are straightforward. The main rules to keep in mind are: 1. NAME is an implicit template argument of class and multiclass, and will be substituted by the name of the instantiating def/defm. 2. The name of a def/defm in a multiclass must contain a reference to NAME. If such a reference is not present, it is automatically prepended. And for some additional subtleties, consider these: 3. defm with no name generates a unique name but has no special behavior otherwise. 4. def with no name generates an anonymous record, whose name is unique but undefined. In particular, the name won't contain a reference to NAME. Keeping rules 1&2 in mind should allow a predictable behavior of name resolution that is simple to follow. The old "rules" were rather surprising: sometimes (but not always), NAME would correspond to the name of the toplevel defm. They were also plain bonkers when you pushed them to their limits, as the old version of the TableGen test case shows. Having NAME correspond to the name of the toplevel defm introduces "spooky action at a distance" and breaks composability: refactoring the upper layers of a hierarchy of nested multiclass instantiations can cause unexpected breakage by changing the value of NAME at a lower level of the hierarchy. The new rules don't suffer from this problem. Some existing .td files have to be adjusted because they ended up depending on the details of the old implementation. Change-Id: I694095231565b30f563e6fd0417b41ee01a12589 Reviewers: tra, simon_tatham, craig.topper, MartinO, arsenm, javed.absar Subscribers: wdng, llvm-commits Differential Revision: https://reviews.llvm.org/D47430 llvm-svn: 333900
2018-06-04 16:26:05 +02:00
for (const RecordVal &Val : Class->getValues())
NewRec->addValue(Val);
// Substitute and resolve template arguments
ArrayRef<Init *> TArgs = Class->getTemplateArgs();
MapResolver R(NewRec);
for (unsigned i = 0, e = TArgs.size(); i != e; ++i) {
if (i < args_size())
R.set(TArgs[i], getArg(i));
else
R.set(TArgs[i], NewRec->getValue(TArgs[i])->getValue());
NewRec->removeValue(TArgs[i]);
}
NewRec->resolveReferences(R);
// Add superclasses.
ArrayRef<std::pair<Record *, SMRange>> SCs = Class->getSuperClasses();
for (const auto &SCPair : SCs)
NewRec->addSuperClass(SCPair.first, SCPair.second);
NewRec->addSuperClass(Class,
SMRange(Class->getLoc().back(),
Class->getLoc().back()));
// Resolve internal references and store in record keeper
NewRec->resolveReferences();
Records.addDef(std::move(NewRecOwner));
Def = DefInit::get(NewRec);
}
return Def;
}
Init *VarDefInit::resolveReferences(Resolver &R) const {
TrackUnresolvedResolver UR(&R);
bool Changed = false;
SmallVector<Init *, 8> NewArgs;
NewArgs.reserve(args_size());
for (Init *Arg : args()) {
Init *NewArg = Arg->resolveReferences(UR);
NewArgs.push_back(NewArg);
Changed |= NewArg != Arg;
}
if (Changed) {
auto New = VarDefInit::get(Class, NewArgs);
if (!UR.foundUnresolved())
return New->instantiate();
return New;
}
return const_cast<VarDefInit *>(this);
}
Init *VarDefInit::Fold() const {
if (Def)
return Def;
TrackUnresolvedResolver R;
for (Init *Arg : args())
Arg->resolveReferences(R);
if (!R.foundUnresolved())
return const_cast<VarDefInit *>(this)->instantiate();
return const_cast<VarDefInit *>(this);
}
std::string VarDefInit::getAsString() const {
std::string Result = Class->getNameInitAsString() + "<";
const char *sep = "";
for (Init *Arg : args()) {
Result += sep;
sep = ", ";
Result += Arg->getAsString();
}
return Result + ">";
}
FieldInit *FieldInit::get(Init *R, StringInit *FN) {
using Key = std::pair<Init *, StringInit *>;
static DenseMap<Key, FieldInit*> ThePool;
Key TheKey(std::make_pair(R, FN));
FieldInit *&I = ThePool[TheKey];
if (!I) I = new(Allocator) FieldInit(R, FN);
return I;
}
Init *FieldInit::getBit(unsigned Bit) const {
if (getType() == BitRecTy::get())
return const_cast<FieldInit*>(this);
return VarBitInit::get(const_cast<FieldInit*>(this), Bit);
}
Init *FieldInit::resolveReferences(Resolver &R) const {
Init *NewRec = Rec->resolveReferences(R);
if (NewRec != Rec)
return FieldInit::get(NewRec, FieldName)->Fold(R.getCurrentRecord());
return const_cast<FieldInit *>(this);
}
Init *FieldInit::Fold(Record *CurRec) const {
if (DefInit *DI = dyn_cast<DefInit>(Rec)) {
Record *Def = DI->getDef();
if (Def == CurRec)
PrintFatalError(CurRec->getLoc(),
Twine("Attempting to access field '") +
FieldName->getAsUnquotedString() + "' of '" +
Rec->getAsString() + "' is a forbidden self-reference");
Init *FieldVal = Def->getValue(FieldName)->getValue();
if (FieldVal->isComplete())
return FieldVal;
}
return const_cast<FieldInit *>(this);
}
bool FieldInit::isConcrete() const {
if (DefInit *DI = dyn_cast<DefInit>(Rec)) {
Init *FieldVal = DI->getDef()->getValue(FieldName)->getValue();
return FieldVal->isConcrete();
}
return false;
}
static void ProfileCondOpInit(FoldingSetNodeID &ID,
ArrayRef<Init *> CondRange,
ArrayRef<Init *> ValRange,
const RecTy *ValType) {
assert(CondRange.size() == ValRange.size() &&
"Number of conditions and values must match!");
ID.AddPointer(ValType);
ArrayRef<Init *>::iterator Case = CondRange.begin();
ArrayRef<Init *>::iterator Val = ValRange.begin();
while (Case != CondRange.end()) {
ID.AddPointer(*Case++);
ID.AddPointer(*Val++);
}
}
void CondOpInit::Profile(FoldingSetNodeID &ID) const {
ProfileCondOpInit(ID,
makeArrayRef(getTrailingObjects<Init *>(), NumConds),
makeArrayRef(getTrailingObjects<Init *>() + NumConds, NumConds),
ValType);
}
CondOpInit *
CondOpInit::get(ArrayRef<Init *> CondRange,
ArrayRef<Init *> ValRange, RecTy *Ty) {
assert(CondRange.size() == ValRange.size() &&
"Number of conditions and values must match!");
static FoldingSet<CondOpInit> ThePool;
FoldingSetNodeID ID;
ProfileCondOpInit(ID, CondRange, ValRange, Ty);
void *IP = nullptr;
if (CondOpInit *I = ThePool.FindNodeOrInsertPos(ID, IP))
return I;
void *Mem = Allocator.Allocate(totalSizeToAlloc<Init *>(2*CondRange.size()),
alignof(BitsInit));
CondOpInit *I = new(Mem) CondOpInit(CondRange.size(), Ty);
std::uninitialized_copy(CondRange.begin(), CondRange.end(),
I->getTrailingObjects<Init *>());
std::uninitialized_copy(ValRange.begin(), ValRange.end(),
I->getTrailingObjects<Init *>()+CondRange.size());
ThePool.InsertNode(I, IP);
return I;
}
Init *CondOpInit::resolveReferences(Resolver &R) const {
SmallVector<Init*, 4> NewConds;
bool Changed = false;
for (const Init *Case : getConds()) {
Init *NewCase = Case->resolveReferences(R);
NewConds.push_back(NewCase);
Changed |= NewCase != Case;
}
SmallVector<Init*, 4> NewVals;
for (const Init *Val : getVals()) {
Init *NewVal = Val->resolveReferences(R);
NewVals.push_back(NewVal);
Changed |= NewVal != Val;
}
if (Changed)
return (CondOpInit::get(NewConds, NewVals,
getValType()))->Fold(R.getCurrentRecord());
return const_cast<CondOpInit *>(this);
}
Init *CondOpInit::Fold(Record *CurRec) const {
for ( unsigned i = 0; i < NumConds; ++i) {
Init *Cond = getCond(i);
Init *Val = getVal(i);
if (IntInit *CondI = dyn_cast_or_null<IntInit>(
Cond->convertInitializerTo(IntRecTy::get()))) {
if (CondI->getValue())
return Val->convertInitializerTo(getValType());
} else
return const_cast<CondOpInit *>(this);
}
PrintFatalError(CurRec->getLoc(),
CurRec->getName() +
" does not have any true condition in:" +
this->getAsString());
return nullptr;
}
bool CondOpInit::isConcrete() const {
for (const Init *Case : getConds())
if (!Case->isConcrete())
return false;
for (const Init *Val : getVals())
if (!Val->isConcrete())
return false;
return true;
}
bool CondOpInit::isComplete() const {
for (const Init *Case : getConds())
if (!Case->isComplete())
return false;
for (const Init *Val : getVals())
if (!Val->isConcrete())
return false;
return true;
}
std::string CondOpInit::getAsString() const {
std::string Result = "!cond(";
for (unsigned i = 0; i < getNumConds(); i++) {
Result += getCond(i)->getAsString() + ": ";
Result += getVal(i)->getAsString();
if (i != getNumConds()-1)
Result += ", ";
}
return Result + ")";
}
Init *CondOpInit::getBit(unsigned Bit) const {
return VarBitInit::get(const_cast<CondOpInit *>(this), Bit);
}
static void ProfileDagInit(FoldingSetNodeID &ID, Init *V, StringInit *VN,
ArrayRef<Init *> ArgRange,
ArrayRef<StringInit *> NameRange) {
ID.AddPointer(V);
ID.AddPointer(VN);
ArrayRef<Init *>::iterator Arg = ArgRange.begin();
ArrayRef<StringInit *>::iterator Name = NameRange.begin();
while (Arg != ArgRange.end()) {
assert(Name != NameRange.end() && "Arg name underflow!");
ID.AddPointer(*Arg++);
ID.AddPointer(*Name++);
}
assert(Name == NameRange.end() && "Arg name overflow!");
}
DagInit *
DagInit::get(Init *V, StringInit *VN, ArrayRef<Init *> ArgRange,
ArrayRef<StringInit *> NameRange) {
static FoldingSet<DagInit> ThePool;
FoldingSetNodeID ID;
ProfileDagInit(ID, V, VN, ArgRange, NameRange);
void *IP = nullptr;
if (DagInit *I = ThePool.FindNodeOrInsertPos(ID, IP))
return I;
void *Mem = Allocator.Allocate(totalSizeToAlloc<Init *, StringInit *>(ArgRange.size(), NameRange.size()), alignof(BitsInit));
DagInit *I = new(Mem) DagInit(V, VN, ArgRange.size(), NameRange.size());
std::uninitialized_copy(ArgRange.begin(), ArgRange.end(),
I->getTrailingObjects<Init *>());
std::uninitialized_copy(NameRange.begin(), NameRange.end(),
I->getTrailingObjects<StringInit *>());
ThePool.InsertNode(I, IP);
return I;
}
DagInit *
DagInit::get(Init *V, StringInit *VN,
ArrayRef<std::pair<Init*, StringInit*>> args) {
SmallVector<Init *, 8> Args;
SmallVector<StringInit *, 8> Names;
for (const auto &Arg : args) {
Args.push_back(Arg.first);
Names.push_back(Arg.second);
}
return DagInit::get(V, VN, Args, Names);
}
void DagInit::Profile(FoldingSetNodeID &ID) const {
ProfileDagInit(ID, Val, ValName, makeArrayRef(getTrailingObjects<Init *>(), NumArgs), makeArrayRef(getTrailingObjects<StringInit *>(), NumArgNames));
}
Record *DagInit::getOperatorAsDef(ArrayRef<SMLoc> Loc) const {
if (DefInit *DefI = dyn_cast<DefInit>(Val))
return DefI->getDef();
PrintFatalError(Loc, "Expected record as operator");
return nullptr;
}
Init *DagInit::resolveReferences(Resolver &R) const {
SmallVector<Init*, 8> NewArgs;
NewArgs.reserve(arg_size());
bool ArgsChanged = false;
for (const Init *Arg : getArgs()) {
Init *NewArg = Arg->resolveReferences(R);
NewArgs.push_back(NewArg);
ArgsChanged |= NewArg != Arg;
}
2009-11-22 05:24:42 +01:00
Init *Op = Val->resolveReferences(R);
if (Op != Val || ArgsChanged)
return DagInit::get(Op, ValName, NewArgs, getArgNames());
2009-11-22 05:24:42 +01:00
return const_cast<DagInit *>(this);
}
bool DagInit::isConcrete() const {
if (!Val->isConcrete())
return false;
for (const Init *Elt : getArgs()) {
if (!Elt->isConcrete())
return false;
}
return true;
}
std::string DagInit::getAsString() const {
std::string Result = "(" + Val->getAsString();
if (ValName)
Result += ":" + ValName->getAsUnquotedString();
if (!arg_empty()) {
Result += " " + getArg(0)->getAsString();
if (getArgName(0)) Result += ":$" + getArgName(0)->getAsUnquotedString();
for (unsigned i = 1, e = getNumArgs(); i != e; ++i) {
Result += ", " + getArg(i)->getAsString();
if (getArgName(i)) Result += ":$" + getArgName(i)->getAsUnquotedString();
}
}
return Result + ")";
}
//===----------------------------------------------------------------------===//
// Other implementations
//===----------------------------------------------------------------------===//
RecordVal::RecordVal(Init *N, RecTy *T, bool P)
: Name(N), TyAndPrefix(T, P) {
TableGen: Allow !cast of records, cleanup conversion machinery Summary: Distinguish two relationships between types: is-a and convertible-to. For example, a bit is not an int or vice versa, but they can be converted into each other (with range checks that you can think of as "dynamic": unlike other type checks, those range checks do not happen during parsing, but only once the final values have been established). Actually converting initializers between types is subtle: even when values of type A can be converted to type B (e.g. int into string), it may not be possible to do so with a concrete initializer (e.g., a VarInit that refers to a variable of type int cannot be immediately converted to a string). For this reason, distinguish between getCastTo and convertInitializerTo: the latter implements the actual conversion when appropriate, while the former will first try to do the actual conversion and fall back to introducing a !cast operation so that the conversion will be delayed until variable references have been resolved. To make the approach of adding !cast operations to work, !cast needs to fallback to convertInitializerTo when the special string <-> record logic does not apply. This enables casting records to a subclass, although that new functionality is only truly useful together with !isa, which will be added in a later change. The test is removed because it uses !srl on a bit sequence, which cannot really be supported consistently, but luckily isn't used anywhere either. Change-Id: I98168bf52649176654ed2ec61a29bdb29970cfe7 Reviewers: arsenm, craig.topper, tra, MartinO Subscribers: wdng, llvm-commits Differential Revision: https://reviews.llvm.org/D43753 llvm-svn: 326785
2018-03-06 14:48:39 +01:00
setValue(UnsetInit::get());
assert(Value && "Cannot create unset value for current type!");
}
StringRef RecordVal::getName() const {
return cast<StringInit>(getNameInit())->getValue();
}
TableGen: Allow !cast of records, cleanup conversion machinery Summary: Distinguish two relationships between types: is-a and convertible-to. For example, a bit is not an int or vice versa, but they can be converted into each other (with range checks that you can think of as "dynamic": unlike other type checks, those range checks do not happen during parsing, but only once the final values have been established). Actually converting initializers between types is subtle: even when values of type A can be converted to type B (e.g. int into string), it may not be possible to do so with a concrete initializer (e.g., a VarInit that refers to a variable of type int cannot be immediately converted to a string). For this reason, distinguish between getCastTo and convertInitializerTo: the latter implements the actual conversion when appropriate, while the former will first try to do the actual conversion and fall back to introducing a !cast operation so that the conversion will be delayed until variable references have been resolved. To make the approach of adding !cast operations to work, !cast needs to fallback to convertInitializerTo when the special string <-> record logic does not apply. This enables casting records to a subclass, although that new functionality is only truly useful together with !isa, which will be added in a later change. The test is removed because it uses !srl on a bit sequence, which cannot really be supported consistently, but luckily isn't used anywhere either. Change-Id: I98168bf52649176654ed2ec61a29bdb29970cfe7 Reviewers: arsenm, craig.topper, tra, MartinO Subscribers: wdng, llvm-commits Differential Revision: https://reviews.llvm.org/D43753 llvm-svn: 326785
2018-03-06 14:48:39 +01:00
bool RecordVal::setValue(Init *V) {
if (V) {
Value = V->getCastTo(getType());
if (Value) {
assert(!isa<TypedInit>(Value) ||
cast<TypedInit>(Value)->getType()->typeIsA(getType()));
if (BitsRecTy *BTy = dyn_cast<BitsRecTy>(getType())) {
if (!isa<BitsInit>(Value)) {
SmallVector<Init *, 64> Bits;
Bits.reserve(BTy->getNumBits());
for (unsigned i = 0, e = BTy->getNumBits(); i < e; ++i)
Bits.push_back(Value->getBit(i));
Value = BitsInit::get(Bits);
}
}
}
return Value == nullptr;
}
Value = nullptr;
return false;
}
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
LLVM_DUMP_METHOD void RecordVal::dump() const { errs() << *this; }
#endif
void RecordVal::print(raw_ostream &OS, bool PrintSem) const {
if (getPrefix()) OS << "field ";
OS << *getType() << " " << getNameInitAsString();
if (getValue())
OS << " = " << *getValue();
if (PrintSem) OS << ";\n";
}
unsigned Record::LastID = 0;
void Record::checkName() {
// Ensure the record name has string type.
const TypedInit *TypedName = cast<const TypedInit>(Name);
if (!isa<StringRecTy>(TypedName->getType()))
TableGen: Streamline how defs are instantiated Summary: Instantiating def's and defm's needs to perform the following steps: - for defm's, clone multiclass def prototypes and subsitute template args - for def's and defm's, add subclass definitions, substituting template args - clone the record based on foreach loops and substitute loop iteration variables - override record variables based on the global 'let' stack - resolve the record name (this should be simple, but unfortunately it's not due to existing .td files relying on rather silly implementation details) - for def(m)s in multiclasses, add the unresolved record as a multiclass prototype - for top-level def(m)s, resolve all internal variable references and add them to the record keeper and any active defsets This change streamlines how we go through these steps, by having both def's and defm's feed into a single addDef() method that handles foreach, final resolve, and routing the record to the right place. This happens to make foreach inside of multiclasses work, as the new test case demonstrates. Previously, foreach inside multiclasses was not forbidden by the parser, but it was de facto broken. Another side effect is that the order of "instantiated from" notes in error messages is reversed, as the modified test case shows. This is arguably clearer, since the initial error message ends up pointing directly to whatever triggered the error, and subsequent notes will point to increasingly outer layers of multiclasses. This is consistent with how C++ compilers report nested #includes and nested template instantiations. Change-Id: Ica146d0db2bc133dd7ed88054371becf24320447 Reviewers: arsenm, craig.topper, tra, MartinO Subscribers: wdng, llvm-commits Differential Revision: https://reviews.llvm.org/D44478 llvm-svn: 328117
2018-03-21 18:12:53 +01:00
PrintFatalError(getLoc(), Twine("Record name '") + Name->getAsString() +
"' is not a string!");
}
RecordRecTy *Record::getType() {
SmallVector<Record *, 4> DirectSCs;
getDirectSuperClasses(DirectSCs);
return RecordRecTy::get(DirectSCs);
}
DefInit *Record::getDefInit() {
if (!TheInit)
TheInit = new(Allocator) DefInit(this);
return TheInit;
}
void Record::setName(Init *NewName) {
Name = NewName;
checkName();
// DO NOT resolve record values to the name at this point because
// there might be default values for arguments of this def. Those
// arguments might not have been resolved yet so we don't want to
// prematurely assume values for those arguments were not passed to
// this def.
//
// Nonetheless, it may be that some of this Record's values
// reference the record name. Indeed, the reason for having the
// record name be an Init is to provide this flexibility. The extra
// resolve steps after completely instantiating defs takes care of
// this. See TGParser::ParseDef and TGParser::ParseDefm.
}
void Record::getDirectSuperClasses(SmallVectorImpl<Record *> &Classes) const {
ArrayRef<std::pair<Record *, SMRange>> SCs = getSuperClasses();
while (!SCs.empty()) {
// Superclasses are in reverse preorder, so 'back' is a direct superclass,
// and its transitive superclasses are directly preceding it.
Record *SC = SCs.back().first;
SCs = SCs.drop_back(1 + SC->getSuperClasses().size());
Classes.push_back(SC);
}
}
void Record::resolveReferences(Resolver &R, const RecordVal *SkipVal) {
for (RecordVal &Value : Values) {
if (SkipVal == &Value) // Skip resolve the same field as the given one
continue;
if (Init *V = Value.getValue()) {
Init *VR = V->resolveReferences(R);
if (Value.setValue(VR)) {
std::string Type;
if (TypedInit *VRT = dyn_cast<TypedInit>(VR))
Type =
(Twine("of type '") + VRT->getType()->getAsString() + "' ").str();
PrintFatalError(getLoc(), Twine("Invalid value ") + Type +
"is found when setting '" +
Value.getNameInitAsString() +
"' of type '" +
Value.getType()->getAsString() +
"' after resolving references: " +
VR->getAsUnquotedString() + "\n");
}
}
}
Init *OldName = getNameInit();
Init *NewName = Name->resolveReferences(R);
if (NewName != OldName) {
// Re-register with RecordKeeper.
setName(NewName);
}
}
void Record::resolveReferences() {
RecordResolver R(*this);
R.setFinal(true);
resolveReferences(R);
}
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
LLVM_DUMP_METHOD void Record::dump() const { errs() << *this; }
#endif
raw_ostream &llvm::operator<<(raw_ostream &OS, const Record &R) {
OS << R.getNameInitAsString();
ArrayRef<Init *> TArgs = R.getTemplateArgs();
if (!TArgs.empty()) {
OS << "<";
bool NeedComma = false;
for (const Init *TA : TArgs) {
if (NeedComma) OS << ", ";
NeedComma = true;
const RecordVal *RV = R.getValue(TA);
assert(RV && "Template argument record not found??");
RV->print(OS, false);
}
OS << ">";
}
OS << " {";
ArrayRef<std::pair<Record *, SMRange>> SC = R.getSuperClasses();
if (!SC.empty()) {
OS << "\t//";
for (const auto &SuperPair : SC)
OS << " " << SuperPair.first->getNameInitAsString();
}
OS << "\n";
for (const RecordVal &Val : R.getValues())
if (Val.getPrefix() && !R.isTemplateArg(Val.getNameInit()))
OS << Val;
for (const RecordVal &Val : R.getValues())
if (!Val.getPrefix() && !R.isTemplateArg(Val.getNameInit()))
OS << Val;
return OS << "}\n";
}
Init *Record::getValueInit(StringRef FieldName) const {
const RecordVal *R = getValue(FieldName);
if (!R || !R->getValue())
PrintFatalError(getLoc(), "Record `" + getName() +
"' does not have a field named `" + FieldName + "'!\n");
return R->getValue();
}
StringRef Record::getValueAsString(StringRef FieldName) const {
const RecordVal *R = getValue(FieldName);
if (!R || !R->getValue())
PrintFatalError(getLoc(), "Record `" + getName() +
"' does not have a field named `" + FieldName + "'!\n");
if (StringInit *SI = dyn_cast<StringInit>(R->getValue()))
return SI->getValue();
if (CodeInit *CI = dyn_cast<CodeInit>(R->getValue()))
return CI->getValue();
PrintFatalError(getLoc(), "Record `" + getName() + "', field `" +
FieldName + "' does not have a string initializer!");
}
BitsInit *Record::getValueAsBitsInit(StringRef FieldName) const {
const RecordVal *R = getValue(FieldName);
if (!R || !R->getValue())
PrintFatalError(getLoc(), "Record `" + getName() +
"' does not have a field named `" + FieldName + "'!\n");
if (BitsInit *BI = dyn_cast<BitsInit>(R->getValue()))
return BI;
PrintFatalError(getLoc(), "Record `" + getName() + "', field `" +
FieldName + "' does not have a BitsInit initializer!");
}
ListInit *Record::getValueAsListInit(StringRef FieldName) const {
const RecordVal *R = getValue(FieldName);
if (!R || !R->getValue())
PrintFatalError(getLoc(), "Record `" + getName() +
"' does not have a field named `" + FieldName + "'!\n");
if (ListInit *LI = dyn_cast<ListInit>(R->getValue()))
return LI;
PrintFatalError(getLoc(), "Record `" + getName() + "', field `" +
FieldName + "' does not have a list initializer!");
}
2009-11-22 05:24:42 +01:00
std::vector<Record*>
Record::getValueAsListOfDefs(StringRef FieldName) const {
ListInit *List = getValueAsListInit(FieldName);
std::vector<Record*> Defs;
for (Init *I : List->getValues()) {
if (DefInit *DI = dyn_cast<DefInit>(I))
Defs.push_back(DI->getDef());
else
PrintFatalError(getLoc(), "Record `" + getName() + "', field `" +
FieldName + "' list is not entirely DefInit!");
}
return Defs;
}
int64_t Record::getValueAsInt(StringRef FieldName) const {
const RecordVal *R = getValue(FieldName);
if (!R || !R->getValue())
PrintFatalError(getLoc(), "Record `" + getName() +
"' does not have a field named `" + FieldName + "'!\n");
if (IntInit *II = dyn_cast<IntInit>(R->getValue()))
return II->getValue();
PrintFatalError(getLoc(), Twine("Record `") + getName() + "', field `" +
FieldName +
"' does not have an int initializer: " +
R->getValue()->getAsString());
}
2009-11-22 05:24:42 +01:00
std::vector<int64_t>
Record::getValueAsListOfInts(StringRef FieldName) const {
ListInit *List = getValueAsListInit(FieldName);
std::vector<int64_t> Ints;
for (Init *I : List->getValues()) {
if (IntInit *II = dyn_cast<IntInit>(I))
Ints.push_back(II->getValue());
else
PrintFatalError(getLoc(),
Twine("Record `") + getName() + "', field `" + FieldName +
"' does not have a list of ints initializer: " +
I->getAsString());
}
return Ints;
}
std::vector<StringRef>
Record::getValueAsListOfStrings(StringRef FieldName) const {
ListInit *List = getValueAsListInit(FieldName);
std::vector<StringRef> Strings;
for (Init *I : List->getValues()) {
if (StringInit *SI = dyn_cast<StringInit>(I))
Strings.push_back(SI->getValue());
else if (CodeInit *CI = dyn_cast<CodeInit>(I))
Strings.push_back(CI->getValue());
else
PrintFatalError(getLoc(),
Twine("Record `") + getName() + "', field `" + FieldName +
"' does not have a list of strings initializer: " +
I->getAsString());
}
return Strings;
}
Record *Record::getValueAsDef(StringRef FieldName) const {
const RecordVal *R = getValue(FieldName);
if (!R || !R->getValue())
PrintFatalError(getLoc(), "Record `" + getName() +
"' does not have a field named `" + FieldName + "'!\n");
if (DefInit *DI = dyn_cast<DefInit>(R->getValue()))
return DI->getDef();
PrintFatalError(getLoc(), "Record `" + getName() + "', field `" +
FieldName + "' does not have a def initializer!");
}
Record *Record::getValueAsOptionalDef(StringRef FieldName) const {
const RecordVal *R = getValue(FieldName);
if (!R || !R->getValue())
PrintFatalError(getLoc(), "Record `" + getName() +
"' does not have a field named `" + FieldName + "'!\n");
if (DefInit *DI = dyn_cast<DefInit>(R->getValue()))
return DI->getDef();
if (isa<UnsetInit>(R->getValue()))
return nullptr;
PrintFatalError(getLoc(), "Record `" + getName() + "', field `" +
FieldName + "' does not have either a def initializer or '?'!");
}
bool Record::getValueAsBit(StringRef FieldName) const {
const RecordVal *R = getValue(FieldName);
if (!R || !R->getValue())
PrintFatalError(getLoc(), "Record `" + getName() +
"' does not have a field named `" + FieldName + "'!\n");
if (BitInit *BI = dyn_cast<BitInit>(R->getValue()))
return BI->getValue();
PrintFatalError(getLoc(), "Record `" + getName() + "', field `" +
FieldName + "' does not have a bit initializer!");
}
bool Record::getValueAsBitOrUnset(StringRef FieldName, bool &Unset) const {
const RecordVal *R = getValue(FieldName);
if (!R || !R->getValue())
PrintFatalError(getLoc(), "Record `" + getName() +
"' does not have a field named `" + FieldName.str() + "'!\n");
if (isa<UnsetInit>(R->getValue())) {
Unset = true;
return false;
}
Unset = false;
if (BitInit *BI = dyn_cast<BitInit>(R->getValue()))
return BI->getValue();
PrintFatalError(getLoc(), "Record `" + getName() + "', field `" +
FieldName + "' does not have a bit initializer!");
}
DagInit *Record::getValueAsDag(StringRef FieldName) const {
const RecordVal *R = getValue(FieldName);
if (!R || !R->getValue())
PrintFatalError(getLoc(), "Record `" + getName() +
"' does not have a field named `" + FieldName + "'!\n");
if (DagInit *DI = dyn_cast<DagInit>(R->getValue()))
return DI;
PrintFatalError(getLoc(), "Record `" + getName() + "', field `" +
FieldName + "' does not have a dag initializer!");
}
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
LLVM_DUMP_METHOD void RecordKeeper::dump() const { errs() << *this; }
#endif
raw_ostream &llvm::operator<<(raw_ostream &OS, const RecordKeeper &RK) {
OS << "------------- Classes -----------------\n";
for (const auto &C : RK.getClasses())
OS << "class " << *C.second;
OS << "------------- Defs -----------------\n";
for (const auto &D : RK.getDefs())
OS << "def " << *D.second;
return OS;
}
/// GetNewAnonymousName - Generate a unique anonymous name that can be used as
/// an identifier.
Init *RecordKeeper::getNewAnonymousName() {
return StringInit::get("anonymous_" + utostr(AnonCounter++));
}
std::vector<Record *>
RecordKeeper::getAllDerivedDefinitions(StringRef ClassName) const {
Record *Class = getClass(ClassName);
if (!Class)
PrintFatalError("ERROR: Couldn't find the `" + ClassName + "' class!\n");
std::vector<Record*> Defs;
for (const auto &D : getDefs())
if (D.second->isSubClassOf(Class))
Defs.push_back(D.second.get());
return Defs;
}
Init *MapResolver::resolve(Init *VarName) {
auto It = Map.find(VarName);
if (It == Map.end())
return nullptr;
Init *I = It->second.V;
if (!It->second.Resolved && Map.size() > 1) {
// Resolve mutual references among the mapped variables, but prevent
// infinite recursion.
Map.erase(It);
I = I->resolveReferences(*this);
Map[VarName] = {I, true};
}
return I;
}
Init *RecordResolver::resolve(Init *VarName) {
Init *Val = Cache.lookup(VarName);
if (Val)
return Val;
for (Init *S : Stack) {
if (S == VarName)
return nullptr; // prevent infinite recursion
}
if (RecordVal *RV = getCurrentRecord()->getValue(VarName)) {
if (!isa<UnsetInit>(RV->getValue())) {
Val = RV->getValue();
Stack.push_back(VarName);
Val = Val->resolveReferences(*this);
Stack.pop_back();
}
}
Cache[VarName] = Val;
return Val;
}
Init *TrackUnresolvedResolver::resolve(Init *VarName) {
Init *I = nullptr;
if (R) {
I = R->resolve(VarName);
if (I && !FoundUnresolved) {
// Do not recurse into the resolved initializer, as that would change
// the behavior of the resolver we're delegating, but do check to see
// if there are unresolved variables remaining.
TrackUnresolvedResolver Sub;
I->resolveReferences(Sub);
FoundUnresolved |= Sub.FoundUnresolved;
}
}
if (!I)
FoundUnresolved = true;
return I;
}
TableGen: Streamline the semantics of NAME Summary: The new rules are straightforward. The main rules to keep in mind are: 1. NAME is an implicit template argument of class and multiclass, and will be substituted by the name of the instantiating def/defm. 2. The name of a def/defm in a multiclass must contain a reference to NAME. If such a reference is not present, it is automatically prepended. And for some additional subtleties, consider these: 3. defm with no name generates a unique name but has no special behavior otherwise. 4. def with no name generates an anonymous record, whose name is unique but undefined. In particular, the name won't contain a reference to NAME. Keeping rules 1&2 in mind should allow a predictable behavior of name resolution that is simple to follow. The old "rules" were rather surprising: sometimes (but not always), NAME would correspond to the name of the toplevel defm. They were also plain bonkers when you pushed them to their limits, as the old version of the TableGen test case shows. Having NAME correspond to the name of the toplevel defm introduces "spooky action at a distance" and breaks composability: refactoring the upper layers of a hierarchy of nested multiclass instantiations can cause unexpected breakage by changing the value of NAME at a lower level of the hierarchy. The new rules don't suffer from this problem. Some existing .td files have to be adjusted because they ended up depending on the details of the old implementation. Change-Id: I694095231565b30f563e6fd0417b41ee01a12589 Reviewers: tra, simon_tatham, craig.topper, MartinO, arsenm, javed.absar Subscribers: wdng, llvm-commits Differential Revision: https://reviews.llvm.org/D47430 llvm-svn: 333900
2018-06-04 16:26:05 +02:00
Init *HasReferenceResolver::resolve(Init *VarName)
{
if (VarName == VarNameToTrack)
Found = true;
return nullptr;
}