[TableGen] Add backend to generate command guide for tools using libOption.

[JDevlieghere]

Summary

For lldb and dsymutil, the command guide is essentially a copy of its help output generated by libOption. Making sure the two stay in sync is tedious and error prone. Given that we already generate the help from a tablegen file, we might as well generate the RST as well.

This adds a tablegen backend for generating Sphinx/RST command guides from the tablegen file.

[merge_guards_bot]

Build result: pass - 60250 tests passed, 0 failed and 732 were skipped.

Log files: console-log.txt, CMakeCache.txt

[Bigcheese]

llvm/utils/TableGen/OptEmitter.h

14

Is this used outside the definition of CompareOptionRecords?

[JDevlieghere]

llvm/utils/TableGen/OptEmitter.h

14

Doesn't look like it. I'll move it into the implementation.

[Bigcheese]

lgtm

llvm/utils/TableGen/CMakeLists.txt

add_tablegen(llvm-tblgen LLVM
   FastISelEmitter.cpp
   FixedLenDecoderEmitter.cpp
   GICombinerEmitter.cpp
   GlobalISelEmitter.cpp
   InfoByHwMode.cpp
   InstrInfoEmitter.cpp
   InstrDocsEmitter.cpp
   IntrinsicEmitter.cpp
+  OptEmitter.cpp
   OptParserEmitter.cpp
+  OptRSTEmitter.cpp
   PredicateExpander.cpp
   PseudoLoweringEmitter.cpp
   RISCVCompressInstEmitter.cpp
   RegisterBankEmitter.cpp
   RegisterInfoEmitter.cpp
   SDNodeProperties.cpp
   SearchableTableEmitter.cpp
   SubtargetEmitter.cpp

llvm/utils/TableGen/OptEmitter.h

+//===- OptEmitter.h - Helper for emitting options. --------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_UTILS_TABLEGEN_OPTEMITTER_H
+#define LLVM_UTILS_TABLEGEN_OPTEMITTER_H
+
+namespace llvm {
+class Record;
+int StrCmpOptionName(const char *A, const char *B);
+int CompareOptionRecords(Record *const *Av, Record *const *Bv);
+} // namespace llvm
+#endif

llvm/utils/TableGen/OptEmitter.cpp

+//===- OptEmitter.cpp - Helper for emitting options.----------- -----------===//
+//
+// 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
+//
+//===----------------------------------------------------------------------===//
+
+#include "OptEmitter.h"
+#include "llvm/ADT/Twine.h"
+#include "llvm/TableGen/Error.h"
+#include "llvm/TableGen/Record.h"
+#include <cctype>
+#include <cstring>
+
+namespace llvm {
+
+// Ordering on Info. The logic should match with the consumer-side function in
+// llvm/Option/OptTable.h.
+// FIXME: Make this take StringRefs instead of null terminated strings to
+// simplify callers.
+int StrCmpOptionName(const char *A, const char *B) {
+  const char *X = A, *Y = B;
+  char a = tolower(*A), b = tolower(*B);
+  while (a == b) {
+    if (a == '\0')
+      return strcmp(A, B);
+
+    a = tolower(*++X);
+    b = tolower(*++Y);
+  }
+
+  if (a == '\0') // A is a prefix of B.
+    return 1;
+  if (b == '\0') // B is a prefix of A.
+    return -1;
+
+  // Otherwise lexicographic.
+  return (a < b) ? -1 : 1;
+}
+
+int CompareOptionRecords(Record *const *Av, Record *const *Bv) {
+  const Record *A = *Av;
+  const Record *B = *Bv;
+
+  // Sentinel options precede all others and are only ordered by precedence.
+  bool ASent = A->getValueAsDef("Kind")->getValueAsBit("Sentinel");
+  bool BSent = B->getValueAsDef("Kind")->getValueAsBit("Sentinel");
+  if (ASent != BSent)
+    return ASent ? -1 : 1;
+
+  // Compare options by name, unless they are sentinels.
+  if (!ASent)
+    if (int Cmp = StrCmpOptionName(A->getValueAsString("Name").str().c_str(),
+                                   B->getValueAsString("Name").str().c_str()))
+      return Cmp;
+
+  if (!ASent) {
+    std::vector<StringRef> APrefixes = A->getValueAsListOfStrings("Prefixes");
+    std::vector<StringRef> BPrefixes = B->getValueAsListOfStrings("Prefixes");
+
+    for (std::vector<StringRef>::const_iterator APre = APrefixes.begin(),
+                                                AEPre = APrefixes.end(),
+                                                BPre = BPrefixes.begin(),
+                                                BEPre = BPrefixes.end();
+         APre != AEPre && BPre != BEPre; ++APre, ++BPre) {
+      if (int Cmp = StrCmpOptionName(APre->str().c_str(), BPre->str().c_str()))
+        return Cmp;
+    }
+  }
+
+  // Then by the kind precedence;
+  int APrec = A->getValueAsDef("Kind")->getValueAsInt("Precedence");
+  int BPrec = B->getValueAsDef("Kind")->getValueAsInt("Precedence");
+  if (APrec == BPrec && A->getValueAsListOfStrings("Prefixes") ==
+                            B->getValueAsListOfStrings("Prefixes")) {
+    PrintError(A->getLoc(), Twine("Option is equivalent to"));
+    PrintError(B->getLoc(), Twine("Other defined here"));
+    PrintFatalError("Equivalent Options found.");
+  }
+  return APrec < BPrec ? -1 : 1;
+}
+
+} // namespace llvm

llvm/utils/TableGen/OptParserEmitter.cpp

 //===- OptParserEmitter.cpp - Table Driven Command Line Parsing -----------===//
 //
 // 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
 //
 //===----------------------------------------------------------------------===//
 
-#include "llvm/TableGen/Error.h"
+#include "OptEmitter.h"
 #include "llvm/ADT/STLExtras.h"
 #include "llvm/ADT/SmallString.h"
 #include "llvm/ADT/Twine.h"
 #include "llvm/TableGen/Record.h"
 #include "llvm/TableGen/TableGenBackend.h"
 #include <cctype>
 #include <cstring>
 #include <map>
 
 using namespace llvm;
 
-// Ordering on Info. The logic should match with the consumer-side function in
-// llvm/Option/OptTable.h.
-// FIXME: Mmake this take StringRefs instead of null terminated strings to
-// simplify callers.
-static int StrCmpOptionName(const char *A, const char *B) {
-  const char *X = A, *Y = B;
-  char a = tolower(*A), b = tolower(*B);
-  while (a == b) {
-    if (a == '\0')
-      return strcmp(A, B);
-
-    a = tolower(*++X);
-    b = tolower(*++Y);
-  }
-
-  if (a == '\0') // A is a prefix of B.
-    return 1;
-  if (b == '\0') // B is a prefix of A.
-    return -1;
-
-  // Otherwise lexicographic.
-  return (a < b) ? -1 : 1;
-}
-
-static int CompareOptionRecords(Record *const *Av, Record *const *Bv) {
-  const Record *A = *Av;
-  const Record *B = *Bv;
-
-  // Sentinel options precede all others and are only ordered by precedence.
-  bool ASent = A->getValueAsDef("Kind")->getValueAsBit("Sentinel");
-  bool BSent = B->getValueAsDef("Kind")->getValueAsBit("Sentinel");
-  if (ASent != BSent)
-    return ASent ? -1 : 1;
-
-  // Compare options by name, unless they are sentinels.
-  if (!ASent)
-    if (int Cmp = StrCmpOptionName(A->getValueAsString("Name").str().c_str(),
-                                   B->getValueAsString("Name").str().c_str()))
-      return Cmp;
-
-  if (!ASent) {
-    std::vector<StringRef> APrefixes = A->getValueAsListOfStrings("Prefixes");
-    std::vector<StringRef> BPrefixes = B->getValueAsListOfStrings("Prefixes");
-
-    for (std::vector<StringRef>::const_iterator APre = APrefixes.begin(),
-                                                AEPre = APrefixes.end(),
-                                                BPre = BPrefixes.begin(),
-                                                BEPre = BPrefixes.end();
-                                                APre != AEPre &&
-                                                BPre != BEPre;
-                                                ++APre, ++BPre) {
-      if (int Cmp = StrCmpOptionName(APre->str().c_str(), BPre->str().c_str()))
-        return Cmp;
-    }
-  }
-
-  // Then by the kind precedence;
-  int APrec = A->getValueAsDef("Kind")->getValueAsInt("Precedence");
-  int BPrec = B->getValueAsDef("Kind")->getValueAsInt("Precedence");
-  if (APrec == BPrec &&
-      A->getValueAsListOfStrings("Prefixes") ==
-      B->getValueAsListOfStrings("Prefixes")) {
-    PrintError(A->getLoc(), Twine("Option is equivalent to"));
-    PrintError(B->getLoc(), Twine("Other defined here"));
-    PrintFatalError("Equivalent Options found.");
-  }
-  return APrec < BPrec ? -1 : 1;
-}
-
 static const std::string getOptionName(const Record &R) {
   // Use the record name unless EnumName is defined.
   if (isa<UnsetInit>(R.getValueInit("EnumName")))
     return R.getName();
 
   return R.getValueAsString("EnumName");
 }
 

llvm/utils/TableGen/OptRSTEmitter.cpp

+//===- OptParserEmitter.cpp - Table Driven Command Line Parsing -----------===//
+//
+// 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
+//
+//===----------------------------------------------------------------------===//
+
+#include "OptEmitter.h"
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/SmallString.h"
+#include "llvm/ADT/Twine.h"
+#include "llvm/TableGen/Error.h"
+#include "llvm/TableGen/Record.h"
+#include "llvm/TableGen/TableGenBackend.h"
+#include <cctype>
+#include <cstring>
+#include <map>
+
+using namespace llvm;
+
+/// OptParserEmitter - This tablegen backend takes an input .td file
+/// describing a list of options and emits a RST man page.
+namespace llvm {
+void EmitOptRST(RecordKeeper &Records, raw_ostream &OS) {
+  llvm::StringMap<std::vector<Record *>> OptionsByGroup;
+  std::vector<Record *> OptionsWithoutGroup;
+
+  // Get the options.
+  std::vector<Record *> Opts = Records.getAllDerivedDefinitions("Option");
+  array_pod_sort(Opts.begin(), Opts.end(), CompareOptionRecords);
+
+  // Get the option groups.
+  const std::vector<Record *> &Groups =
+      Records.getAllDerivedDefinitions("OptionGroup");
+  for (unsigned i = 0, e = Groups.size(); i != e; ++i) {
+    const Record &R = *Groups[i];
+    OptionsByGroup.try_emplace(R.getValueAsString("Name"));
+  }
+
+  // Map options to their group.
+  for (unsigned i = 0, e = Opts.size(); i != e; ++i) {
+    const Record &R = *Opts[i];
+    const ListInit *GroupFlags = nullptr;
+    if (const DefInit *DI = dyn_cast<DefInit>(R.getValueInit("Group"))) {
+      GroupFlags = DI->getDef()->getValueAsListInit("Flags");
+      OptionsByGroup[DI->getDef()->getValueAsString("Name")].push_back(Opts[i]);
+    } else {
+      OptionsByGroup["options"].push_back(Opts[i]);
+    }
+  }
+
+  // Print options under their group.
+  for (const auto &KV : OptionsByGroup) {
+    std::string GroupName = KV.getKey().upper();
+    OS << GroupName << '\n';
+    OS << std::string(GroupName.size(), '-') << '\n';
+    OS << '\n';
+
+    for (Record *R : KV.getValue()) {
+      OS << ".. option:: ";
+
+      // Print the prefix.
+      std::vector<StringRef> Prefixes = R->getValueAsListOfStrings("Prefixes");
+      if (!Prefixes.empty())
+        OS << Prefixes[0];
+
+      // Print the option name.
+      OS << R->getValueAsString("Name");
+
+      // Print the meta-variable.
+      if (!isa<UnsetInit>(R->getValueInit("MetaVarName"))) {
+        OS << '=';
+        OS.write_escaped(R->getValueAsString("MetaVarName"));
+      }
+
+      OS << "\n\n";
+
+      // The option help text.
+      if (!isa<UnsetInit>(R->getValueInit("HelpText"))) {
+        OS << ' ';
+        OS.write_escaped(R->getValueAsString("HelpText"));
+        OS << "\n\n";
+      }
+    }
+  }
+}
+} // end namespace llvm

llvm/utils/TableGen/TableGen.cpp

enum ActionType {
   GenSubtarget,
   GenIntrinsicEnums,
   GenIntrinsicImpl,
   GenTgtIntrinsicEnums,
   GenTgtIntrinsicImpl,
   PrintEnums,
   PrintSets,
   GenOptParserDefs,
+  GenOptRST,
   GenCTags,
   GenAttributes,
   GenSearchableTables,
   GenGlobalISel,
   GenGICombiner,
   GenX86EVEX2VEXTables,
   GenX86FoldTables,
   GenRegisterBank,
cl::values(
                    "Generate target intrinsic enums"),
         clEnumValN(GenTgtIntrinsicImpl, "gen-tgt-intrinsic-impl",
                    "Generate target intrinsic information"),
         clEnumValN(PrintEnums, "print-enums", "Print enum values for a class"),
         clEnumValN(PrintSets, "print-sets",
                    "Print expanded sets for testing DAG exprs"),
         clEnumValN(GenOptParserDefs, "gen-opt-parser-defs",
                    "Generate option definitions"),
+        clEnumValN(GenOptRST, "gen-opt-rst", "Generate option RST"),
         clEnumValN(GenCTags, "gen-ctags", "Generate ctags-compatible index"),
         clEnumValN(GenAttributes, "gen-attrs", "Generate attributes"),
         clEnumValN(GenSearchableTables, "gen-searchable-tables",
                    "Generate generic binary-searchable table"),
         clEnumValN(GenGlobalISel, "gen-global-isel",
                    "Generate GlobalISel selector"),
         clEnumValN(GenGICombiner, "gen-global-isel-combiner",
                    "Generate GlobalISel combiner"),
         clEnumValN(GenX86EVEX2VEXTables, "gen-x86-EVEX2VEX-tables",
                    "Generate X86 EVEX to VEX compress tables"),
         clEnumValN(GenX86FoldTables, "gen-x86-fold-tables",
                    "Generate X86 fold tables"),
         clEnumValN(GenRegisterBank, "gen-register-bank",
                    "Generate registers bank descriptions"),
         clEnumValN(GenExegesis, "gen-exegesis",
                    "Generate llvm-exegesis tables"),
-        clEnumValN(GenAutomata, "gen-automata",
-                   "Generate generic automata")));
+        clEnumValN(GenAutomata, "gen-automata", "Generate generic automata")));
 
 cl::OptionCategory PrintEnumsCat("Options for -print-enums");
 cl::opt<std::string> Class("class", cl::desc("Print Enum list for this class"),
                            cl::value_desc("class name"),
                            cl::cat(PrintEnumsCat));
 
 cl::opt<bool, true>
     TimeRegionsOpt("time-regions",
case GenTgtIntrinsicEnums:
     EmitIntrinsicEnums(Records, OS, true);
     break;
   case GenTgtIntrinsicImpl:
     EmitIntrinsicImpl(Records, OS, true);
     break;
   case GenOptParserDefs:
     EmitOptParser(Records, OS);
     break;
+  case GenOptRST:
+    EmitOptRST(Records, OS);
+    break;
   case PrintEnums:
   {
     for (Record *Rec : Records.getAllDerivedDefinitions(Class))
       OS << Rec->getName() << ", ";
     OS << "\n";
     break;
   }
   case PrintSets:

llvm/utils/TableGen/TableGenBackends.h

 void EmitInstrInfo(RecordKeeper &RK, raw_ostream &OS);
 void EmitInstrDocs(RecordKeeper &RK, raw_ostream &OS);
 void EmitPseudoLowering(RecordKeeper &RK, raw_ostream &OS);
 void EmitCompressInst(RecordKeeper &RK, raw_ostream &OS);
 void EmitRegisterInfo(RecordKeeper &RK, raw_ostream &OS);
 void EmitSubtarget(RecordKeeper &RK, raw_ostream &OS);
 void EmitMapTable(RecordKeeper &RK, raw_ostream &OS);
 void EmitOptParser(RecordKeeper &RK, raw_ostream &OS);
+void EmitOptRST(RecordKeeper &RK, raw_ostream &OS);
 void EmitCTags(RecordKeeper &RK, raw_ostream &OS);
 void EmitAttributes(RecordKeeper &RK, raw_ostream &OS);
 void EmitSearchableTables(RecordKeeper &RK, raw_ostream &OS);
 void EmitGlobalISel(RecordKeeper &RK, raw_ostream &OS);
 void EmitGICombiner(RecordKeeper &RK, raw_ostream &OS);
 void EmitX86EVEX2VEXTables(RecordKeeper &RK, raw_ostream &OS);
 void EmitX86FoldTables(RecordKeeper &RK, raw_ostream &OS);
 void EmitRegisterBank(RecordKeeper &RK, raw_ostream &OS);
 void EmitExegesis(RecordKeeper &RK, raw_ostream &OS);
 void EmitAutomata(RecordKeeper &RK, raw_ostream &OS);
 
 } // End llvm namespace
 
 #endif