2017-09-28 07:26:01 +08:00
|
|
|
//===- MILexer.cpp - Machine instructions lexer implementation ------------===//
|
2015-06-23 04:37:46 +08:00
|
|
|
//
|
|
|
|
// The LLVM Compiler Infrastructure
|
|
|
|
//
|
|
|
|
// This file is distributed under the University of Illinois Open Source
|
|
|
|
// License. See LICENSE.TXT for details.
|
|
|
|
//
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
//
|
|
|
|
// This file implements the lexing of machine instructions.
|
|
|
|
//
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
|
|
|
|
#include "MILexer.h"
|
2017-09-28 07:26:01 +08:00
|
|
|
#include "llvm/ADT/APSInt.h"
|
2016-04-16 15:51:28 +08:00
|
|
|
#include "llvm/ADT/None.h"
|
2017-09-28 07:26:01 +08:00
|
|
|
#include "llvm/ADT/STLExtras.h"
|
2015-07-21 04:31:01 +08:00
|
|
|
#include "llvm/ADT/StringExtras.h"
|
2015-07-07 07:07:26 +08:00
|
|
|
#include "llvm/ADT/StringSwitch.h"
|
2017-09-28 07:26:01 +08:00
|
|
|
#include "llvm/ADT/StringRef.h"
|
2015-06-23 04:37:46 +08:00
|
|
|
#include "llvm/ADT/Twine.h"
|
2017-09-28 07:26:01 +08:00
|
|
|
#include <algorithm>
|
|
|
|
#include <cassert>
|
2015-06-23 04:37:46 +08:00
|
|
|
#include <cctype>
|
2017-09-28 07:26:01 +08:00
|
|
|
#include <string>
|
2015-06-23 04:37:46 +08:00
|
|
|
|
|
|
|
using namespace llvm;
|
|
|
|
|
|
|
|
namespace {
|
|
|
|
|
2017-09-28 07:26:01 +08:00
|
|
|
using ErrorCallbackType =
|
|
|
|
function_ref<void(StringRef::iterator Loc, const Twine &)>;
|
2016-03-19 04:41:11 +08:00
|
|
|
|
2015-06-23 04:37:46 +08:00
|
|
|
/// This class provides a way to iterate and get characters from the source
|
|
|
|
/// string.
|
|
|
|
class Cursor {
|
2017-09-28 07:26:01 +08:00
|
|
|
const char *Ptr = nullptr;
|
|
|
|
const char *End = nullptr;
|
2015-06-23 04:37:46 +08:00
|
|
|
|
|
|
|
public:
|
2017-09-28 07:26:01 +08:00
|
|
|
Cursor(NoneType) {}
|
2015-07-01 00:51:29 +08:00
|
|
|
|
2015-06-23 04:37:46 +08:00
|
|
|
explicit Cursor(StringRef Str) {
|
|
|
|
Ptr = Str.data();
|
|
|
|
End = Ptr + Str.size();
|
|
|
|
}
|
|
|
|
|
|
|
|
bool isEOF() const { return Ptr == End; }
|
|
|
|
|
2015-06-24 14:40:09 +08:00
|
|
|
char peek(int I = 0) const { return End - Ptr <= I ? 0 : Ptr[I]; }
|
2015-06-23 04:37:46 +08:00
|
|
|
|
2015-06-27 00:46:11 +08:00
|
|
|
void advance(unsigned I = 1) { Ptr += I; }
|
2015-06-23 04:37:46 +08:00
|
|
|
|
|
|
|
StringRef remaining() const { return StringRef(Ptr, End - Ptr); }
|
|
|
|
|
|
|
|
StringRef upto(Cursor C) const {
|
|
|
|
assert(C.Ptr >= Ptr && C.Ptr <= End);
|
|
|
|
return StringRef(Ptr, C.Ptr - Ptr);
|
|
|
|
}
|
|
|
|
|
|
|
|
StringRef::iterator location() const { return Ptr; }
|
2015-07-01 00:51:29 +08:00
|
|
|
|
|
|
|
operator bool() const { return Ptr != nullptr; }
|
2015-06-23 04:37:46 +08:00
|
|
|
};
|
|
|
|
|
|
|
|
} // end anonymous namespace
|
|
|
|
|
2015-08-07 07:17:42 +08:00
|
|
|
MIToken &MIToken::reset(TokenKind Kind, StringRef Range) {
|
|
|
|
this->Kind = Kind;
|
|
|
|
this->Range = Range;
|
|
|
|
return *this;
|
|
|
|
}
|
|
|
|
|
|
|
|
MIToken &MIToken::setStringValue(StringRef StrVal) {
|
|
|
|
StringValue = StrVal;
|
|
|
|
return *this;
|
|
|
|
}
|
|
|
|
|
|
|
|
MIToken &MIToken::setOwnedStringValue(std::string StrVal) {
|
|
|
|
StringValueStorage = std::move(StrVal);
|
|
|
|
StringValue = StringValueStorage;
|
|
|
|
return *this;
|
|
|
|
}
|
|
|
|
|
|
|
|
MIToken &MIToken::setIntegerValue(APSInt IntVal) {
|
|
|
|
this->IntVal = std::move(IntVal);
|
|
|
|
return *this;
|
|
|
|
}
|
|
|
|
|
2015-06-23 04:37:46 +08:00
|
|
|
/// Skip the leading whitespace characters and return the updated cursor.
|
|
|
|
static Cursor skipWhitespace(Cursor C) {
|
2015-08-14 07:10:16 +08:00
|
|
|
while (isblank(C.peek()))
|
|
|
|
C.advance();
|
|
|
|
return C;
|
|
|
|
}
|
|
|
|
|
|
|
|
static bool isNewlineChar(char C) { return C == '\n' || C == '\r'; }
|
|
|
|
|
|
|
|
/// Skip a line comment and return the updated cursor.
|
|
|
|
static Cursor skipComment(Cursor C) {
|
|
|
|
if (C.peek() != ';')
|
|
|
|
return C;
|
|
|
|
while (!isNewlineChar(C.peek()) && !C.isEOF())
|
2015-06-23 04:37:46 +08:00
|
|
|
C.advance();
|
|
|
|
return C;
|
|
|
|
}
|
|
|
|
|
2015-07-18 06:48:04 +08:00
|
|
|
/// Return true if the given character satisfies the following regular
|
|
|
|
/// expression: [-a-zA-Z$._0-9]
|
2015-06-23 04:37:46 +08:00
|
|
|
static bool isIdentifierChar(char C) {
|
2015-07-18 06:48:04 +08:00
|
|
|
return isalpha(C) || isdigit(C) || C == '_' || C == '-' || C == '.' ||
|
|
|
|
C == '$';
|
2015-06-23 04:37:46 +08:00
|
|
|
}
|
|
|
|
|
2015-08-06 01:35:55 +08:00
|
|
|
/// Unescapes the given string value.
|
|
|
|
///
|
|
|
|
/// Expects the string value to be quoted.
|
|
|
|
static std::string unescapeQuotedString(StringRef Value) {
|
2015-07-21 04:31:01 +08:00
|
|
|
assert(Value.front() == '"' && Value.back() == '"');
|
|
|
|
Cursor C = Cursor(Value.substr(1, Value.size() - 2));
|
|
|
|
|
2015-08-06 01:35:55 +08:00
|
|
|
std::string Str;
|
2015-07-21 04:31:01 +08:00
|
|
|
Str.reserve(C.remaining().size());
|
|
|
|
while (!C.isEOF()) {
|
|
|
|
char Char = C.peek();
|
|
|
|
if (Char == '\\') {
|
|
|
|
if (C.peek(1) == '\\') {
|
|
|
|
// Two '\' become one
|
|
|
|
Str += '\\';
|
|
|
|
C.advance(2);
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
if (isxdigit(C.peek(1)) && isxdigit(C.peek(2))) {
|
|
|
|
Str += hexDigitValue(C.peek(1)) * 16 + hexDigitValue(C.peek(2));
|
|
|
|
C.advance(3);
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
Str += Char;
|
|
|
|
C.advance();
|
|
|
|
}
|
2015-08-06 01:35:55 +08:00
|
|
|
return Str;
|
2015-07-21 04:31:01 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/// Lex a string constant using the following regular expression: \"[^\"]*\"
|
2016-03-19 04:41:11 +08:00
|
|
|
static Cursor lexStringConstant(Cursor C, ErrorCallbackType ErrorCallback) {
|
2015-07-21 04:31:01 +08:00
|
|
|
assert(C.peek() == '"');
|
|
|
|
for (C.advance(); C.peek() != '"'; C.advance()) {
|
2015-08-14 07:10:16 +08:00
|
|
|
if (C.isEOF() || isNewlineChar(C.peek())) {
|
2015-07-21 04:31:01 +08:00
|
|
|
ErrorCallback(
|
|
|
|
C.location(),
|
|
|
|
"end of machine instruction reached before the closing '\"'");
|
|
|
|
return None;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
C.advance();
|
|
|
|
return C;
|
|
|
|
}
|
|
|
|
|
2016-03-19 04:41:11 +08:00
|
|
|
static Cursor lexName(Cursor C, MIToken &Token, MIToken::TokenKind Type,
|
|
|
|
unsigned PrefixLength, ErrorCallbackType ErrorCallback) {
|
2015-07-29 01:03:40 +08:00
|
|
|
auto Range = C;
|
|
|
|
C.advance(PrefixLength);
|
|
|
|
if (C.peek() == '"') {
|
|
|
|
if (Cursor R = lexStringConstant(C, ErrorCallback)) {
|
2015-08-06 01:35:55 +08:00
|
|
|
StringRef String = Range.upto(R);
|
2015-08-07 07:17:42 +08:00
|
|
|
Token.reset(Type, String)
|
|
|
|
.setOwnedStringValue(
|
|
|
|
unescapeQuotedString(String.drop_front(PrefixLength)));
|
2015-07-29 01:03:40 +08:00
|
|
|
return R;
|
|
|
|
}
|
2015-08-07 07:17:42 +08:00
|
|
|
Token.reset(MIToken::Error, Range.remaining());
|
2015-07-29 01:03:40 +08:00
|
|
|
return Range;
|
|
|
|
}
|
|
|
|
while (isIdentifierChar(C.peek()))
|
|
|
|
C.advance();
|
2015-08-07 07:17:42 +08:00
|
|
|
Token.reset(Type, Range.upto(C))
|
|
|
|
.setStringValue(Range.upto(C).drop_front(PrefixLength));
|
2015-07-29 01:03:40 +08:00
|
|
|
return C;
|
|
|
|
}
|
|
|
|
|
2015-07-07 07:07:26 +08:00
|
|
|
static MIToken::TokenKind getIdentifierKind(StringRef Identifier) {
|
|
|
|
return StringSwitch<MIToken::TokenKind>(Identifier)
|
|
|
|
.Case("_", MIToken::underscore)
|
|
|
|
.Case("implicit", MIToken::kw_implicit)
|
|
|
|
.Case("implicit-def", MIToken::kw_implicit_define)
|
2015-08-20 02:55:47 +08:00
|
|
|
.Case("def", MIToken::kw_def)
|
2015-07-08 04:34:53 +08:00
|
|
|
.Case("dead", MIToken::kw_dead)
|
2015-07-09 05:23:34 +08:00
|
|
|
.Case("killed", MIToken::kw_killed)
|
2015-07-09 07:58:31 +08:00
|
|
|
.Case("undef", MIToken::kw_undef)
|
2015-08-15 03:07:07 +08:00
|
|
|
.Case("internal", MIToken::kw_internal)
|
2015-08-06 01:49:03 +08:00
|
|
|
.Case("early-clobber", MIToken::kw_early_clobber)
|
2015-08-06 01:41:17 +08:00
|
|
|
.Case("debug-use", MIToken::kw_debug_use)
|
2017-12-13 01:53:59 +08:00
|
|
|
.Case("renamable", MIToken::kw_renamable)
|
2015-08-20 03:05:34 +08:00
|
|
|
.Case("tied-def", MIToken::kw_tied_def)
|
2015-07-17 08:24:15 +08:00
|
|
|
.Case("frame-setup", MIToken::kw_frame_setup)
|
2018-01-09 19:33:22 +08:00
|
|
|
.Case("frame-destroy", MIToken::kw_frame_destroy)
|
2018-05-03 08:07:56 +08:00
|
|
|
.Case("nnan", MIToken::kw_nnan)
|
|
|
|
.Case("ninf", MIToken::kw_ninf)
|
|
|
|
.Case("nsz", MIToken::kw_nsz)
|
|
|
|
.Case("arcp", MIToken::kw_arcp)
|
|
|
|
.Case("contract", MIToken::kw_contract)
|
|
|
|
.Case("afn", MIToken::kw_afn)
|
|
|
|
.Case("reassoc", MIToken::kw_reassoc)
|
2015-07-23 05:15:11 +08:00
|
|
|
.Case("debug-location", MIToken::kw_debug_location)
|
2016-07-27 02:20:00 +08:00
|
|
|
.Case("same_value", MIToken::kw_cfi_same_value)
|
|
|
|
.Case("offset", MIToken::kw_cfi_offset)
|
2017-12-15 23:17:18 +08:00
|
|
|
.Case("rel_offset", MIToken::kw_cfi_rel_offset)
|
2016-07-27 02:20:00 +08:00
|
|
|
.Case("def_cfa_register", MIToken::kw_cfi_def_cfa_register)
|
|
|
|
.Case("def_cfa_offset", MIToken::kw_cfi_def_cfa_offset)
|
2017-12-15 23:17:18 +08:00
|
|
|
.Case("adjust_cfa_offset", MIToken::kw_cfi_adjust_cfa_offset)
|
|
|
|
.Case("escape", MIToken::kw_cfi_escape)
|
2016-07-27 02:20:00 +08:00
|
|
|
.Case("def_cfa", MIToken::kw_cfi_def_cfa)
|
2017-12-15 23:17:18 +08:00
|
|
|
.Case("remember_state", MIToken::kw_cfi_remember_state)
|
2017-11-02 20:00:58 +08:00
|
|
|
.Case("restore", MIToken::kw_cfi_restore)
|
2017-12-15 23:17:18 +08:00
|
|
|
.Case("restore_state", MIToken::kw_cfi_restore_state)
|
|
|
|
.Case("undefined", MIToken::kw_cfi_undefined)
|
|
|
|
.Case("register", MIToken::kw_cfi_register)
|
|
|
|
.Case("window_save", MIToken::kw_cfi_window_save)
|
2015-07-29 01:28:03 +08:00
|
|
|
.Case("blockaddress", MIToken::kw_blockaddress)
|
2016-07-30 04:32:59 +08:00
|
|
|
.Case("intrinsic", MIToken::kw_intrinsic)
|
2015-07-29 07:02:45 +08:00
|
|
|
.Case("target-index", MIToken::kw_target_index)
|
2015-08-01 04:49:21 +08:00
|
|
|
.Case("half", MIToken::kw_half)
|
|
|
|
.Case("float", MIToken::kw_float)
|
|
|
|
.Case("double", MIToken::kw_double)
|
|
|
|
.Case("x86_fp80", MIToken::kw_x86_fp80)
|
|
|
|
.Case("fp128", MIToken::kw_fp128)
|
|
|
|
.Case("ppc_fp128", MIToken::kw_ppc_fp128)
|
2015-08-06 08:44:07 +08:00
|
|
|
.Case("target-flags", MIToken::kw_target_flags)
|
2015-08-04 08:24:45 +08:00
|
|
|
.Case("volatile", MIToken::kw_volatile)
|
2015-08-07 00:49:30 +08:00
|
|
|
.Case("non-temporal", MIToken::kw_non_temporal)
|
[CodeGen] Split out the notions of MI invariance and MI dereferenceability.
Summary:
An IR load can be invariant, dereferenceable, neither, or both. But
currently, MI's notion of invariance is IR-invariant &&
IR-dereferenceable.
This patch splits up the notions of invariance and dereferenceability at
the MI level. It's NFC, so adds some probably-unnecessary
"is-dereferenceable" checks, which we can remove later if desired.
Reviewers: chandlerc, tstellarAMD
Subscribers: jholewinski, arsenm, nemanjai, llvm-commits
Differential Revision: https://reviews.llvm.org/D23371
llvm-svn: 281151
2016-09-11 09:38:58 +08:00
|
|
|
.Case("dereferenceable", MIToken::kw_dereferenceable)
|
2015-08-07 00:55:53 +08:00
|
|
|
.Case("invariant", MIToken::kw_invariant)
|
2015-08-08 04:48:30 +08:00
|
|
|
.Case("align", MIToken::kw_align)
|
2018-01-26 19:47:28 +08:00
|
|
|
.Case("addrspace", MIToken::kw_addrspace)
|
2015-08-13 04:44:16 +08:00
|
|
|
.Case("stack", MIToken::kw_stack)
|
2015-08-13 05:00:22 +08:00
|
|
|
.Case("got", MIToken::kw_got)
|
2015-08-13 05:11:08 +08:00
|
|
|
.Case("jump-table", MIToken::kw_jump_table)
|
2015-08-13 04:33:26 +08:00
|
|
|
.Case("constant-pool", MIToken::kw_constant_pool)
|
2015-08-20 08:12:57 +08:00
|
|
|
.Case("call-entry", MIToken::kw_call_entry)
|
2015-08-11 07:24:42 +08:00
|
|
|
.Case("liveout", MIToken::kw_liveout)
|
2015-08-14 07:10:16 +08:00
|
|
|
.Case("address-taken", MIToken::kw_address_taken)
|
|
|
|
.Case("landing-pad", MIToken::kw_landing_pad)
|
|
|
|
.Case("liveins", MIToken::kw_liveins)
|
|
|
|
.Case("successors", MIToken::kw_successors)
|
2016-08-18 04:25:25 +08:00
|
|
|
.Case("floatpred", MIToken::kw_floatpred)
|
|
|
|
.Case("intpred", MIToken::kw_intpred)
|
2015-07-07 07:07:26 +08:00
|
|
|
.Default(MIToken::Identifier);
|
|
|
|
}
|
|
|
|
|
2015-07-01 00:51:29 +08:00
|
|
|
static Cursor maybeLexIdentifier(Cursor C, MIToken &Token) {
|
2016-07-27 05:49:34 +08:00
|
|
|
if (!isalpha(C.peek()) && C.peek() != '_')
|
2015-07-01 00:51:29 +08:00
|
|
|
return None;
|
2015-06-23 04:37:46 +08:00
|
|
|
auto Range = C;
|
|
|
|
while (isIdentifierChar(C.peek()))
|
|
|
|
C.advance();
|
2015-06-25 01:34:58 +08:00
|
|
|
auto Identifier = Range.upto(C);
|
2015-08-07 07:17:42 +08:00
|
|
|
Token.reset(getIdentifierKind(Identifier), Identifier)
|
|
|
|
.setStringValue(Identifier);
|
2015-06-23 04:37:46 +08:00
|
|
|
return C;
|
|
|
|
}
|
|
|
|
|
2016-03-19 04:41:11 +08:00
|
|
|
static Cursor maybeLexMachineBasicBlock(Cursor C, MIToken &Token,
|
|
|
|
ErrorCallbackType ErrorCallback) {
|
2015-08-14 07:10:16 +08:00
|
|
|
bool IsReference = C.remaining().startswith("%bb.");
|
|
|
|
if (!IsReference && !C.remaining().startswith("bb."))
|
2015-07-01 00:51:29 +08:00
|
|
|
return None;
|
2015-06-27 00:46:11 +08:00
|
|
|
auto Range = C;
|
2015-08-14 07:10:16 +08:00
|
|
|
unsigned PrefixLength = IsReference ? 4 : 3;
|
|
|
|
C.advance(PrefixLength); // Skip '%bb.' or 'bb.'
|
2015-06-27 00:46:11 +08:00
|
|
|
if (!isdigit(C.peek())) {
|
2015-08-07 07:17:42 +08:00
|
|
|
Token.reset(MIToken::Error, C.remaining());
|
2015-06-27 00:46:11 +08:00
|
|
|
ErrorCallback(C.location(), "expected a number after '%bb.'");
|
|
|
|
return C;
|
|
|
|
}
|
|
|
|
auto NumberRange = C;
|
|
|
|
while (isdigit(C.peek()))
|
|
|
|
C.advance();
|
|
|
|
StringRef Number = NumberRange.upto(C);
|
2015-08-14 07:10:16 +08:00
|
|
|
unsigned StringOffset = PrefixLength + Number.size(); // Drop '%bb.<id>'
|
2017-12-05 01:18:51 +08:00
|
|
|
// TODO: The format bb.<id>.<irname> is supported only when it's not a
|
|
|
|
// reference. Once we deprecate the format where the irname shows up, we
|
|
|
|
// should only lex forward if it is a reference.
|
2015-06-27 00:46:11 +08:00
|
|
|
if (C.peek() == '.') {
|
|
|
|
C.advance(); // Skip '.'
|
|
|
|
++StringOffset;
|
|
|
|
while (isIdentifierChar(C.peek()))
|
|
|
|
C.advance();
|
|
|
|
}
|
2015-08-14 07:10:16 +08:00
|
|
|
Token.reset(IsReference ? MIToken::MachineBasicBlock
|
|
|
|
: MIToken::MachineBasicBlockLabel,
|
|
|
|
Range.upto(C))
|
2015-08-07 07:17:42 +08:00
|
|
|
.setIntegerValue(APSInt(Number))
|
|
|
|
.setStringValue(Range.upto(C).drop_front(StringOffset));
|
2015-06-27 00:46:11 +08:00
|
|
|
return C;
|
|
|
|
}
|
|
|
|
|
2015-07-16 07:38:35 +08:00
|
|
|
static Cursor maybeLexIndex(Cursor C, MIToken &Token, StringRef Rule,
|
|
|
|
MIToken::TokenKind Kind) {
|
|
|
|
if (!C.remaining().startswith(Rule) || !isdigit(C.peek(Rule.size())))
|
|
|
|
return None;
|
|
|
|
auto Range = C;
|
|
|
|
C.advance(Rule.size());
|
|
|
|
auto NumberRange = C;
|
|
|
|
while (isdigit(C.peek()))
|
|
|
|
C.advance();
|
2015-08-07 07:17:42 +08:00
|
|
|
Token.reset(Kind, Range.upto(C)).setIntegerValue(APSInt(NumberRange.upto(C)));
|
2015-07-16 07:38:35 +08:00
|
|
|
return C;
|
|
|
|
}
|
|
|
|
|
2015-07-17 07:37:45 +08:00
|
|
|
static Cursor maybeLexIndexAndName(Cursor C, MIToken &Token, StringRef Rule,
|
|
|
|
MIToken::TokenKind Kind) {
|
|
|
|
if (!C.remaining().startswith(Rule) || !isdigit(C.peek(Rule.size())))
|
|
|
|
return None;
|
|
|
|
auto Range = C;
|
|
|
|
C.advance(Rule.size());
|
|
|
|
auto NumberRange = C;
|
|
|
|
while (isdigit(C.peek()))
|
|
|
|
C.advance();
|
|
|
|
StringRef Number = NumberRange.upto(C);
|
|
|
|
unsigned StringOffset = Rule.size() + Number.size();
|
|
|
|
if (C.peek() == '.') {
|
|
|
|
C.advance();
|
|
|
|
++StringOffset;
|
|
|
|
while (isIdentifierChar(C.peek()))
|
|
|
|
C.advance();
|
|
|
|
}
|
2015-08-07 07:17:42 +08:00
|
|
|
Token.reset(Kind, Range.upto(C))
|
|
|
|
.setIntegerValue(APSInt(Number))
|
|
|
|
.setStringValue(Range.upto(C).drop_front(StringOffset));
|
2015-07-17 07:37:45 +08:00
|
|
|
return C;
|
|
|
|
}
|
|
|
|
|
2015-07-16 07:38:35 +08:00
|
|
|
static Cursor maybeLexJumpTableIndex(Cursor C, MIToken &Token) {
|
|
|
|
return maybeLexIndex(C, Token, "%jump-table.", MIToken::JumpTableIndex);
|
|
|
|
}
|
|
|
|
|
2015-07-17 07:37:45 +08:00
|
|
|
static Cursor maybeLexStackObject(Cursor C, MIToken &Token) {
|
|
|
|
return maybeLexIndexAndName(C, Token, "%stack.", MIToken::StackObject);
|
|
|
|
}
|
|
|
|
|
|
|
|
static Cursor maybeLexFixedStackObject(Cursor C, MIToken &Token) {
|
|
|
|
return maybeLexIndex(C, Token, "%fixed-stack.", MIToken::FixedStackObject);
|
|
|
|
}
|
|
|
|
|
2015-07-21 04:51:18 +08:00
|
|
|
static Cursor maybeLexConstantPoolItem(Cursor C, MIToken &Token) {
|
|
|
|
return maybeLexIndex(C, Token, "%const.", MIToken::ConstantPoolItem);
|
|
|
|
}
|
|
|
|
|
2016-03-29 02:18:46 +08:00
|
|
|
static Cursor maybeLexSubRegisterIndex(Cursor C, MIToken &Token,
|
|
|
|
ErrorCallbackType ErrorCallback) {
|
|
|
|
const StringRef Rule = "%subreg.";
|
|
|
|
if (!C.remaining().startswith(Rule))
|
|
|
|
return None;
|
|
|
|
return lexName(C, Token, MIToken::SubRegisterIndex, Rule.size(),
|
|
|
|
ErrorCallback);
|
|
|
|
}
|
|
|
|
|
2016-03-19 04:41:11 +08:00
|
|
|
static Cursor maybeLexIRBlock(Cursor C, MIToken &Token,
|
|
|
|
ErrorCallbackType ErrorCallback) {
|
2015-07-29 01:28:03 +08:00
|
|
|
const StringRef Rule = "%ir-block.";
|
|
|
|
if (!C.remaining().startswith(Rule))
|
|
|
|
return None;
|
|
|
|
if (isdigit(C.peek(Rule.size())))
|
|
|
|
return maybeLexIndex(C, Token, Rule, MIToken::IRBlock);
|
2015-08-06 01:35:55 +08:00
|
|
|
return lexName(C, Token, MIToken::NamedIRBlock, Rule.size(), ErrorCallback);
|
2015-07-28 06:42:41 +08:00
|
|
|
}
|
|
|
|
|
2016-03-19 04:41:11 +08:00
|
|
|
static Cursor maybeLexIRValue(Cursor C, MIToken &Token,
|
|
|
|
ErrorCallbackType ErrorCallback) {
|
2015-08-04 07:08:19 +08:00
|
|
|
const StringRef Rule = "%ir.";
|
|
|
|
if (!C.remaining().startswith(Rule))
|
|
|
|
return None;
|
2015-08-20 07:31:05 +08:00
|
|
|
if (isdigit(C.peek(Rule.size())))
|
|
|
|
return maybeLexIndex(C, Token, Rule, MIToken::IRValue);
|
2015-08-06 01:35:55 +08:00
|
|
|
return lexName(C, Token, MIToken::NamedIRValue, Rule.size(), ErrorCallback);
|
2015-08-04 07:08:19 +08:00
|
|
|
}
|
|
|
|
|
2017-07-12 06:23:00 +08:00
|
|
|
static Cursor maybeLexStringConstant(Cursor C, MIToken &Token,
|
|
|
|
ErrorCallbackType ErrorCallback) {
|
|
|
|
if (C.peek() != '"')
|
|
|
|
return None;
|
|
|
|
return lexName(C, Token, MIToken::StringConstant, /*PrefixLength=*/0,
|
|
|
|
ErrorCallback);
|
|
|
|
}
|
|
|
|
|
2015-07-11 06:51:20 +08:00
|
|
|
static Cursor lexVirtualRegister(Cursor C, MIToken &Token) {
|
|
|
|
auto Range = C;
|
|
|
|
C.advance(); // Skip '%'
|
|
|
|
auto NumberRange = C;
|
|
|
|
while (isdigit(C.peek()))
|
|
|
|
C.advance();
|
2015-08-07 07:17:42 +08:00
|
|
|
Token.reset(MIToken::VirtualRegister, Range.upto(C))
|
|
|
|
.setIntegerValue(APSInt(NumberRange.upto(C)));
|
2015-07-11 06:51:20 +08:00
|
|
|
return C;
|
|
|
|
}
|
|
|
|
|
2016-07-27 05:49:34 +08:00
|
|
|
/// Returns true for a character allowed in a register name.
|
|
|
|
static bool isRegisterChar(char C) {
|
|
|
|
return isIdentifierChar(C) && C != '.';
|
|
|
|
}
|
|
|
|
|
2018-03-31 02:15:54 +08:00
|
|
|
static Cursor lexNamedVirtualRegister(Cursor C, MIToken &Token) {
|
|
|
|
Cursor Range = C;
|
|
|
|
C.advance(); // Skip '%'
|
|
|
|
while (isRegisterChar(C.peek()))
|
|
|
|
C.advance();
|
|
|
|
Token.reset(MIToken::NamedVirtualRegister, Range.upto(C))
|
|
|
|
.setStringValue(Range.upto(C).drop_front(1)); // Drop the '%'
|
|
|
|
return C;
|
|
|
|
}
|
|
|
|
|
2018-02-01 06:04:26 +08:00
|
|
|
static Cursor maybeLexRegister(Cursor C, MIToken &Token,
|
|
|
|
ErrorCallbackType ErrorCallback) {
|
|
|
|
if (C.peek() != '%' && C.peek() != '$')
|
2015-07-01 00:51:29 +08:00
|
|
|
return None;
|
2018-02-01 06:04:26 +08:00
|
|
|
|
|
|
|
if (C.peek() == '%') {
|
|
|
|
if (isdigit(C.peek(1)))
|
|
|
|
return lexVirtualRegister(C, Token);
|
|
|
|
|
2018-03-31 02:15:54 +08:00
|
|
|
if (isRegisterChar(C.peek(1)))
|
|
|
|
return lexNamedVirtualRegister(C, Token);
|
|
|
|
|
2018-02-01 06:04:26 +08:00
|
|
|
return None;
|
|
|
|
}
|
|
|
|
|
|
|
|
assert(C.peek() == '$');
|
2015-06-24 00:35:26 +08:00
|
|
|
auto Range = C;
|
2018-02-01 06:04:26 +08:00
|
|
|
C.advance(); // Skip '$'
|
2016-07-27 05:49:34 +08:00
|
|
|
while (isRegisterChar(C.peek()))
|
2015-06-24 00:35:26 +08:00
|
|
|
C.advance();
|
2015-08-07 07:17:42 +08:00
|
|
|
Token.reset(MIToken::NamedRegister, Range.upto(C))
|
2018-02-01 06:04:26 +08:00
|
|
|
.setStringValue(Range.upto(C).drop_front(1)); // Drop the '$'
|
2015-06-24 00:35:26 +08:00
|
|
|
return C;
|
|
|
|
}
|
|
|
|
|
2016-03-19 04:41:11 +08:00
|
|
|
static Cursor maybeLexGlobalValue(Cursor C, MIToken &Token,
|
|
|
|
ErrorCallbackType ErrorCallback) {
|
2015-07-22 05:23:08 +08:00
|
|
|
if (C.peek() != '@')
|
|
|
|
return None;
|
|
|
|
if (!isdigit(C.peek(1)))
|
2015-08-06 01:35:55 +08:00
|
|
|
return lexName(C, Token, MIToken::NamedGlobalValue, /*PrefixLength=*/1,
|
2015-07-22 05:23:08 +08:00
|
|
|
ErrorCallback);
|
|
|
|
auto Range = C;
|
|
|
|
C.advance(1); // Skip the '@'
|
|
|
|
auto NumberRange = C;
|
|
|
|
while (isdigit(C.peek()))
|
|
|
|
C.advance();
|
2015-08-07 07:17:42 +08:00
|
|
|
Token.reset(MIToken::GlobalValue, Range.upto(C))
|
|
|
|
.setIntegerValue(APSInt(NumberRange.upto(C)));
|
2015-07-22 05:23:08 +08:00
|
|
|
return C;
|
|
|
|
}
|
|
|
|
|
2016-03-19 04:41:11 +08:00
|
|
|
static Cursor maybeLexExternalSymbol(Cursor C, MIToken &Token,
|
|
|
|
ErrorCallbackType ErrorCallback) {
|
2018-01-10 08:56:48 +08:00
|
|
|
if (C.peek() != '&')
|
2015-07-22 00:59:53 +08:00
|
|
|
return None;
|
2015-08-06 01:35:55 +08:00
|
|
|
return lexName(C, Token, MIToken::ExternalSymbol, /*PrefixLength=*/1,
|
|
|
|
ErrorCallback);
|
2015-07-22 00:59:53 +08:00
|
|
|
}
|
|
|
|
|
2015-08-01 04:49:21 +08:00
|
|
|
static bool isValidHexFloatingPointPrefix(char C) {
|
|
|
|
return C == 'H' || C == 'K' || C == 'L' || C == 'M';
|
|
|
|
}
|
|
|
|
|
|
|
|
static Cursor lexFloatingPointLiteral(Cursor Range, Cursor C, MIToken &Token) {
|
|
|
|
C.advance();
|
|
|
|
// Skip over [0-9]*([eE][-+]?[0-9]+)?
|
|
|
|
while (isdigit(C.peek()))
|
|
|
|
C.advance();
|
|
|
|
if ((C.peek() == 'e' || C.peek() == 'E') &&
|
|
|
|
(isdigit(C.peek(1)) ||
|
|
|
|
((C.peek(1) == '-' || C.peek(1) == '+') && isdigit(C.peek(2))))) {
|
|
|
|
C.advance(2);
|
|
|
|
while (isdigit(C.peek()))
|
|
|
|
C.advance();
|
|
|
|
}
|
2015-08-07 07:17:42 +08:00
|
|
|
Token.reset(MIToken::FloatingPointLiteral, Range.upto(C));
|
2015-08-01 04:49:21 +08:00
|
|
|
return C;
|
|
|
|
}
|
|
|
|
|
2016-10-13 05:06:45 +08:00
|
|
|
static Cursor maybeLexHexadecimalLiteral(Cursor C, MIToken &Token) {
|
|
|
|
if (C.peek() != '0' || (C.peek(1) != 'x' && C.peek(1) != 'X'))
|
|
|
|
return None;
|
|
|
|
Cursor Range = C;
|
|
|
|
C.advance(2);
|
|
|
|
unsigned PrefLen = 2;
|
|
|
|
if (isValidHexFloatingPointPrefix(C.peek())) {
|
|
|
|
C.advance();
|
|
|
|
PrefLen++;
|
|
|
|
}
|
|
|
|
while (isxdigit(C.peek()))
|
|
|
|
C.advance();
|
|
|
|
StringRef StrVal = Range.upto(C);
|
|
|
|
if (StrVal.size() <= PrefLen)
|
|
|
|
return None;
|
|
|
|
if (PrefLen == 2)
|
|
|
|
Token.reset(MIToken::HexLiteral, Range.upto(C));
|
|
|
|
else // It must be 3, which means that there was a floating-point prefix.
|
|
|
|
Token.reset(MIToken::FloatingPointLiteral, Range.upto(C));
|
|
|
|
return C;
|
|
|
|
}
|
|
|
|
|
2015-08-01 04:49:21 +08:00
|
|
|
static Cursor maybeLexNumericalLiteral(Cursor C, MIToken &Token) {
|
2015-07-01 00:51:29 +08:00
|
|
|
if (!isdigit(C.peek()) && (C.peek() != '-' || !isdigit(C.peek(1))))
|
|
|
|
return None;
|
2015-06-24 07:42:28 +08:00
|
|
|
auto Range = C;
|
|
|
|
C.advance();
|
|
|
|
while (isdigit(C.peek()))
|
|
|
|
C.advance();
|
2015-08-01 04:49:21 +08:00
|
|
|
if (C.peek() == '.')
|
|
|
|
return lexFloatingPointLiteral(Range, C, Token);
|
2015-06-24 07:42:28 +08:00
|
|
|
StringRef StrVal = Range.upto(C);
|
2015-08-07 07:17:42 +08:00
|
|
|
Token.reset(MIToken::IntegerLiteral, StrVal).setIntegerValue(APSInt(StrVal));
|
2015-06-24 07:42:28 +08:00
|
|
|
return C;
|
|
|
|
}
|
|
|
|
|
2015-08-18 06:05:15 +08:00
|
|
|
static MIToken::TokenKind getMetadataKeywordKind(StringRef Identifier) {
|
|
|
|
return StringSwitch<MIToken::TokenKind>(Identifier)
|
|
|
|
.Case("!tbaa", MIToken::md_tbaa)
|
2015-08-18 06:06:40 +08:00
|
|
|
.Case("!alias.scope", MIToken::md_alias_scope)
|
2015-08-18 06:08:02 +08:00
|
|
|
.Case("!noalias", MIToken::md_noalias)
|
2015-08-18 06:09:52 +08:00
|
|
|
.Case("!range", MIToken::md_range)
|
2017-08-24 04:31:27 +08:00
|
|
|
.Case("!DIExpression", MIToken::md_diexpr)
|
2015-08-18 06:05:15 +08:00
|
|
|
.Default(MIToken::Error);
|
|
|
|
}
|
|
|
|
|
2016-03-19 04:41:11 +08:00
|
|
|
static Cursor maybeLexExlaim(Cursor C, MIToken &Token,
|
|
|
|
ErrorCallbackType ErrorCallback) {
|
2015-08-18 06:05:15 +08:00
|
|
|
if (C.peek() != '!')
|
|
|
|
return None;
|
|
|
|
auto Range = C;
|
|
|
|
C.advance(1);
|
|
|
|
if (isdigit(C.peek()) || !isIdentifierChar(C.peek())) {
|
|
|
|
Token.reset(MIToken::exclaim, Range.upto(C));
|
|
|
|
return C;
|
|
|
|
}
|
|
|
|
while (isIdentifierChar(C.peek()))
|
|
|
|
C.advance();
|
|
|
|
StringRef StrVal = Range.upto(C);
|
|
|
|
Token.reset(getMetadataKeywordKind(StrVal), StrVal);
|
|
|
|
if (Token.isError())
|
|
|
|
ErrorCallback(Token.location(),
|
|
|
|
"use of unknown metadata keyword '" + StrVal + "'");
|
|
|
|
return C;
|
|
|
|
}
|
|
|
|
|
2015-06-24 00:35:26 +08:00
|
|
|
static MIToken::TokenKind symbolToken(char C) {
|
|
|
|
switch (C) {
|
|
|
|
case ',':
|
|
|
|
return MIToken::comma;
|
2016-07-27 05:49:34 +08:00
|
|
|
case '.':
|
|
|
|
return MIToken::dot;
|
2015-06-24 00:35:26 +08:00
|
|
|
case '=':
|
|
|
|
return MIToken::equal;
|
2015-07-14 07:24:34 +08:00
|
|
|
case ':':
|
|
|
|
return MIToken::colon;
|
2015-07-29 01:28:03 +08:00
|
|
|
case '(':
|
|
|
|
return MIToken::lparen;
|
|
|
|
case ')':
|
|
|
|
return MIToken::rparen;
|
2015-08-15 02:57:24 +08:00
|
|
|
case '{':
|
|
|
|
return MIToken::lbrace;
|
|
|
|
case '}':
|
|
|
|
return MIToken::rbrace;
|
2015-08-06 06:26:15 +08:00
|
|
|
case '+':
|
|
|
|
return MIToken::plus;
|
|
|
|
case '-':
|
|
|
|
return MIToken::minus;
|
2016-03-08 08:57:31 +08:00
|
|
|
case '<':
|
2016-03-08 10:00:43 +08:00
|
|
|
return MIToken::less;
|
2016-03-08 08:57:31 +08:00
|
|
|
case '>':
|
2016-03-08 10:00:43 +08:00
|
|
|
return MIToken::greater;
|
2015-06-24 00:35:26 +08:00
|
|
|
default:
|
|
|
|
return MIToken::Error;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2015-07-01 00:51:29 +08:00
|
|
|
static Cursor maybeLexSymbol(Cursor C, MIToken &Token) {
|
2015-08-04 07:08:19 +08:00
|
|
|
MIToken::TokenKind Kind;
|
|
|
|
unsigned Length = 1;
|
|
|
|
if (C.peek() == ':' && C.peek(1) == ':') {
|
|
|
|
Kind = MIToken::coloncolon;
|
|
|
|
Length = 2;
|
|
|
|
} else
|
|
|
|
Kind = symbolToken(C.peek());
|
2015-07-01 00:51:29 +08:00
|
|
|
if (Kind == MIToken::Error)
|
|
|
|
return None;
|
2015-06-24 00:35:26 +08:00
|
|
|
auto Range = C;
|
2015-08-04 07:08:19 +08:00
|
|
|
C.advance(Length);
|
2015-08-07 07:17:42 +08:00
|
|
|
Token.reset(Kind, Range.upto(C));
|
2015-06-24 00:35:26 +08:00
|
|
|
return C;
|
|
|
|
}
|
|
|
|
|
2015-08-14 07:10:16 +08:00
|
|
|
static Cursor maybeLexNewline(Cursor C, MIToken &Token) {
|
|
|
|
if (!isNewlineChar(C.peek()))
|
|
|
|
return None;
|
|
|
|
auto Range = C;
|
|
|
|
C.advance();
|
|
|
|
Token.reset(MIToken::Newline, Range.upto(C));
|
|
|
|
return C;
|
|
|
|
}
|
|
|
|
|
2016-03-19 04:41:11 +08:00
|
|
|
static Cursor maybeLexEscapedIRValue(Cursor C, MIToken &Token,
|
|
|
|
ErrorCallbackType ErrorCallback) {
|
2015-08-22 05:54:12 +08:00
|
|
|
if (C.peek() != '`')
|
|
|
|
return None;
|
|
|
|
auto Range = C;
|
|
|
|
C.advance();
|
|
|
|
auto StrRange = C;
|
|
|
|
while (C.peek() != '`') {
|
|
|
|
if (C.isEOF() || isNewlineChar(C.peek())) {
|
|
|
|
ErrorCallback(
|
|
|
|
C.location(),
|
|
|
|
"end of machine instruction reached before the closing '`'");
|
|
|
|
Token.reset(MIToken::Error, Range.remaining());
|
|
|
|
return C;
|
|
|
|
}
|
|
|
|
C.advance();
|
|
|
|
}
|
|
|
|
StringRef Value = StrRange.upto(C);
|
|
|
|
C.advance();
|
|
|
|
Token.reset(MIToken::QuotedIRValue, Range.upto(C)).setStringValue(Value);
|
|
|
|
return C;
|
|
|
|
}
|
|
|
|
|
2016-03-19 04:41:11 +08:00
|
|
|
StringRef llvm::lexMIToken(StringRef Source, MIToken &Token,
|
|
|
|
ErrorCallbackType ErrorCallback) {
|
2015-08-14 07:10:16 +08:00
|
|
|
auto C = skipComment(skipWhitespace(Cursor(Source)));
|
2015-06-23 04:37:46 +08:00
|
|
|
if (C.isEOF()) {
|
2015-08-07 07:17:42 +08:00
|
|
|
Token.reset(MIToken::Eof, C.remaining());
|
2015-06-23 04:37:46 +08:00
|
|
|
return C.remaining();
|
|
|
|
}
|
|
|
|
|
2015-07-01 00:51:29 +08:00
|
|
|
if (Cursor R = maybeLexMachineBasicBlock(C, Token, ErrorCallback))
|
|
|
|
return R.remaining();
|
2015-08-14 07:10:16 +08:00
|
|
|
if (Cursor R = maybeLexIdentifier(C, Token))
|
|
|
|
return R.remaining();
|
2015-07-16 07:38:35 +08:00
|
|
|
if (Cursor R = maybeLexJumpTableIndex(C, Token))
|
|
|
|
return R.remaining();
|
2015-07-17 07:37:45 +08:00
|
|
|
if (Cursor R = maybeLexStackObject(C, Token))
|
|
|
|
return R.remaining();
|
|
|
|
if (Cursor R = maybeLexFixedStackObject(C, Token))
|
|
|
|
return R.remaining();
|
2015-07-21 04:51:18 +08:00
|
|
|
if (Cursor R = maybeLexConstantPoolItem(C, Token))
|
|
|
|
return R.remaining();
|
2016-03-29 02:18:46 +08:00
|
|
|
if (Cursor R = maybeLexSubRegisterIndex(C, Token, ErrorCallback))
|
|
|
|
return R.remaining();
|
2015-07-29 01:28:03 +08:00
|
|
|
if (Cursor R = maybeLexIRBlock(C, Token, ErrorCallback))
|
2015-07-28 06:42:41 +08:00
|
|
|
return R.remaining();
|
2015-08-04 07:08:19 +08:00
|
|
|
if (Cursor R = maybeLexIRValue(C, Token, ErrorCallback))
|
|
|
|
return R.remaining();
|
2018-02-01 06:04:26 +08:00
|
|
|
if (Cursor R = maybeLexRegister(C, Token, ErrorCallback))
|
2015-07-01 00:51:29 +08:00
|
|
|
return R.remaining();
|
2015-07-21 04:31:01 +08:00
|
|
|
if (Cursor R = maybeLexGlobalValue(C, Token, ErrorCallback))
|
2015-07-01 00:51:29 +08:00
|
|
|
return R.remaining();
|
2015-07-22 00:59:53 +08:00
|
|
|
if (Cursor R = maybeLexExternalSymbol(C, Token, ErrorCallback))
|
|
|
|
return R.remaining();
|
2016-10-13 05:06:45 +08:00
|
|
|
if (Cursor R = maybeLexHexadecimalLiteral(C, Token))
|
2015-08-01 04:49:21 +08:00
|
|
|
return R.remaining();
|
|
|
|
if (Cursor R = maybeLexNumericalLiteral(C, Token))
|
2015-07-01 00:51:29 +08:00
|
|
|
return R.remaining();
|
2015-08-18 06:05:15 +08:00
|
|
|
if (Cursor R = maybeLexExlaim(C, Token, ErrorCallback))
|
|
|
|
return R.remaining();
|
2015-07-01 00:51:29 +08:00
|
|
|
if (Cursor R = maybeLexSymbol(C, Token))
|
|
|
|
return R.remaining();
|
2015-08-14 07:10:16 +08:00
|
|
|
if (Cursor R = maybeLexNewline(C, Token))
|
|
|
|
return R.remaining();
|
2015-08-22 05:54:12 +08:00
|
|
|
if (Cursor R = maybeLexEscapedIRValue(C, Token, ErrorCallback))
|
|
|
|
return R.remaining();
|
2017-07-12 06:23:00 +08:00
|
|
|
if (Cursor R = maybeLexStringConstant(C, Token, ErrorCallback))
|
|
|
|
return R.remaining();
|
2015-07-01 00:51:29 +08:00
|
|
|
|
2015-08-07 07:17:42 +08:00
|
|
|
Token.reset(MIToken::Error, C.remaining());
|
2015-06-23 04:37:46 +08:00
|
|
|
ErrorCallback(C.location(),
|
2015-07-01 00:51:29 +08:00
|
|
|
Twine("unexpected character '") + Twine(C.peek()) + "'");
|
2015-06-23 04:37:46 +08:00
|
|
|
return C.remaining();
|
|
|
|
}
|