[StrTable] Switch RISCV to leverage sharded, prefixed builtins w/ TableGen

chandlerc · chandlerc · commit 1cb979f001b2 · 2025-02-04T18:04:57.000Z
This lets the TableGen-ed code be much cleaner, directly building an
efficient string table without duplicates and without the repeated
prefix.
diff --git a/clang/include/clang/Basic/BuiltinsRISCVVector.def b/clang/include/clang/Basic/BuiltinsRISCVVector.def
diff --git a/clang/include/clang/Basic/TargetBuiltins.h b/clang/include/clang/Basic/TargetBuiltins.h
@@ -156,8 +156,12 @@ namespace clang {
   namespace RISCVVector {
   enum {
     LastTIBuiltin = clang::Builtin::FirstTSBuiltin - 1,
-#define BUILTIN(ID, TYPE, ATTRS) BI##ID,
-#include "clang/Basic/BuiltinsRISCVVector.def"
+#define GET_RISCVV_BUILTIN_ENUMERATORS
+#include "clang/Basic/riscv_vector_builtins.inc"
+    FirstSiFiveBuiltin,
+    LastRVVBuiltin = FirstSiFiveBuiltin - 1,
+#include "clang/Basic/riscv_sifive_vector_builtins.inc"
+#undef GET_RISCVV_BUILTIN_ENUMERATORS
     FirstTSBuiltin,
   };
   }
diff --git a/clang/lib/Basic/Targets/RISCV.cpp b/clang/lib/Basic/Targets/RISCV.cpp
@@ -241,32 +241,50 @@ void RISCVTargetInfo::getTargetDefines(const LangOptions &Opts,
 }
 
 static constexpr int NumRVVBuiltins =
-    clang::RISCVVector::FirstTSBuiltin - Builtin::FirstTSBuiltin;
+    RISCVVector::FirstSiFiveBuiltin - Builtin::FirstTSBuiltin;
+static constexpr int NumRVVSiFiveBuiltins =
+    RISCVVector::FirstTSBuiltin - RISCVVector::FirstSiFiveBuiltin;
 static constexpr int NumRISCVBuiltins =
-    clang::RISCV::LastTSBuiltin - RISCVVector::FirstTSBuiltin;
+    RISCV::LastTSBuiltin - RISCVVector::FirstTSBuiltin;
 static constexpr int NumBuiltins =
-    clang::RISCV::LastTSBuiltin - Builtin::FirstTSBuiltin;
-static_assert(NumBuiltins == (NumRVVBuiltins + NumRISCVBuiltins));
-
-static constexpr llvm::StringTable BuiltinRVVStrings =
-    CLANG_BUILTIN_STR_TABLE_START
-#define BUILTIN CLANG_BUILTIN_STR_TABLE
-#define TARGET_BUILTIN CLANG_TARGET_BUILTIN_STR_TABLE
-#include "clang/Basic/BuiltinsRISCVVector.def"
-    ;
-static constexpr llvm::StringTable BuiltinRISCVStrings =
+    RISCV::LastTSBuiltin - Builtin::FirstTSBuiltin;
+static_assert(NumBuiltins ==
+              (NumRVVBuiltins + NumRVVSiFiveBuiltins + NumRISCVBuiltins));
+
+namespace RVV {
+#define GET_RISCVV_BUILTIN_STR_TABLE
+#include "clang/Basic/riscv_vector_builtins.inc"
+#undef GET_RISCVV_BUILTIN_STR_TABLE
+static_assert(BuiltinStrings.size() < 100'000);
+
+static constexpr std::array<Builtin::Info, NumRVVBuiltins> BuiltinInfos = {
+#define GET_RISCVV_BUILTIN_INFOS
+#include "clang/Basic/riscv_vector_builtins.inc"
+#undef GET_RISCVV_BUILTIN_INFOS
+};
+} // namespace RVV
+
+namespace RVVSiFive {
+#define GET_RISCVV_BUILTIN_STR_TABLE
+#include "clang/Basic/riscv_sifive_vector_builtins.inc"
+#undef GET_RISCVV_BUILTIN_STR_TABLE
+
+static constexpr std::array<Builtin::Info, NumRVVSiFiveBuiltins> BuiltinInfos =
+    {
+#define GET_RISCVV_BUILTIN_INFOS
+#include "clang/Basic/riscv_sifive_vector_builtins.inc"
+#undef GET_RISCVV_BUILTIN_INFOS
+};
+} // namespace RVVSiFive
+
+static constexpr llvm::StringTable BuiltinStrings =
     CLANG_BUILTIN_STR_TABLE_START
 #define BUILTIN CLANG_BUILTIN_STR_TABLE
 #define TARGET_BUILTIN CLANG_TARGET_BUILTIN_STR_TABLE
 #include "clang/Basic/BuiltinsRISCV.inc"
     ;
 
-static constexpr auto BuiltinRVVInfos = Builtin::MakeInfos<NumRVVBuiltins>({
-#define BUILTIN CLANG_BUILTIN_ENTRY
-#define TARGET_BUILTIN CLANG_TARGET_BUILTIN_ENTRY
-#include "clang/Basic/BuiltinsRISCVVector.def"
-});
-static constexpr auto BuiltinRISCVInfos = Builtin::MakeInfos<NumRISCVBuiltins>({
+static constexpr auto BuiltinInfos = Builtin::MakeInfos<NumRISCVBuiltins>({
 #define BUILTIN CLANG_BUILTIN_ENTRY
 #define TARGET_BUILTIN CLANG_TARGET_BUILTIN_ENTRY
 #include "clang/Basic/BuiltinsRISCV.inc"
@@ -275,8 +293,9 @@ static constexpr auto BuiltinRISCVInfos = Builtin::MakeInfos<NumRISCVBuiltins>({
 llvm::SmallVector<Builtin::InfosShard>
 RISCVTargetInfo::getTargetBuiltins() const {
   return {
-      {&BuiltinRVVStrings, BuiltinRVVInfos},
-      {&BuiltinRISCVStrings, BuiltinRISCVInfos},
+      {&RVV::BuiltinStrings, RVV::BuiltinInfos, "__builtin_rvv_"},
+      {&RVVSiFive::BuiltinStrings, RVVSiFive::BuiltinInfos, "__builtin_rvv_"},
+      {&BuiltinStrings, BuiltinInfos},
   };
 }
 
diff --git a/clang/utils/TableGen/RISCVVEmitter.cpp b/clang/utils/TableGen/RISCVVEmitter.cpp
@@ -18,10 +18,12 @@
 #include "llvm/ADT/ArrayRef.h"
 #include "llvm/ADT/StringExtras.h"
 #include "llvm/ADT/StringMap.h"
+#include "llvm/ADT/StringRef.h"
 #include "llvm/ADT/StringSwitch.h"
 #include "llvm/ADT/Twine.h"
 #include "llvm/TableGen/Error.h"
 #include "llvm/TableGen/Record.h"
+#include "llvm/TableGen/StringToOffsetTable.h"
 #include <optional>
 
 using namespace llvm;
@@ -498,31 +500,68 @@ void RVVEmitter::createBuiltins(raw_ostream &OS) {
   std::vector<std::unique_ptr<RVVIntrinsic>> Defs;
   createRVVIntrinsics(Defs);
 
-  // Map to keep track of which builtin names have already been emitted.
-  StringMap<RVVIntrinsic *> BuiltinMap;
+  llvm::StringToOffsetTable Table;
+  // Ensure offset zero is the empty string.
+  Table.GetOrAddStringOffset("");
+  // Hard coded strings used in the builtin structures.
+  Table.GetOrAddStringOffset("n");
+  Table.GetOrAddStringOffset("zve32x");
 
-  OS << "#if defined(TARGET_BUILTIN) && !defined(RISCVV_BUILTIN)\n";
-  OS << "#define RISCVV_BUILTIN(ID, TYPE, ATTRS) TARGET_BUILTIN(ID, TYPE, "
-        "ATTRS, \"zve32x\")\n";
-  OS << "#endif\n";
+  // Map to unique the builtin names.
+  StringMap<RVVIntrinsic *> BuiltinMap;
+  std::vector<RVVIntrinsic *> UniqueDefs;
   for (auto &Def : Defs) {
-    auto P =
-        BuiltinMap.insert(std::make_pair(Def->getBuiltinName(), Def.get()));
-    if (!P.second) {
-      // Verf that this would have produced the same builtin definition.
-      if (P.first->second->hasBuiltinAlias() != Def->hasBuiltinAlias())
-        PrintFatalError("Builtin with same name has different hasAutoDef");
-      else if (!Def->hasBuiltinAlias() &&
-               P.first->second->getBuiltinTypeStr() != Def->getBuiltinTypeStr())
-        PrintFatalError("Builtin with same name has different type string");
+    auto P = BuiltinMap.insert({Def->getBuiltinName(), Def.get()});
+    if (P.second) {
+      Table.GetOrAddStringOffset(Def->getBuiltinName());
+      if (!Def->hasBuiltinAlias())
+        Table.GetOrAddStringOffset(Def->getBuiltinTypeStr());
+      UniqueDefs.push_back(Def.get());
       continue;
     }
-    OS << "RISCVV_BUILTIN(__builtin_rvv_" << Def->getBuiltinName() << ",\"";
-    if (!Def->hasBuiltinAlias())
-      OS << Def->getBuiltinTypeStr();
-    OS << "\", \"n\")\n";
+
+    // Verf that this would have produced the same builtin definition.
+    if (P.first->second->hasBuiltinAlias() != Def->hasBuiltinAlias())
+      PrintFatalError("Builtin with same name has different hasAutoDef");
+    else if (!Def->hasBuiltinAlias() &&
+             P.first->second->getBuiltinTypeStr() != Def->getBuiltinTypeStr())
+      PrintFatalError("Builtin with same name has different type string");
+  }
+
+  // Emit the enumerators of RVV builtins. Note that these are emitted without
+  // any outer context to enable concatenating them.
+  OS << "// RISCV Vector builtin enumerators\n";
+  OS << "#ifdef GET_RISCVV_BUILTIN_ENUMERATORS\n";
+  for (RVVIntrinsic *Def : UniqueDefs)
+    OS << "  BI__builtin_rvv_" << Def->getBuiltinName() << ",\n";
+  OS << "#endif // GET_RISCVV_BUILTIN_ENUMERATORS\n\n";
+
+  // Emit the string table for the RVV builtins.
+  OS << "// RISCV Vector builtin enumerators\n";
+  OS << "#ifdef GET_RISCVV_BUILTIN_STR_TABLE\n";
+  Table.EmitStringTableDef(OS, "BuiltinStrings");
+  OS << "#endif // GET_RISCVV_BUILTIN_STR_TABLE\n\n";
+
+  // Emit the info structs of RVV builtins. Note that these are emitted without
+  // any outer context to enable concatenating them.
+  OS << "// RISCV Vector builtin infos\n";
+  OS << "#ifdef GET_RISCVV_BUILTIN_INFOS\n";
+  for (RVVIntrinsic *Def : UniqueDefs) {
+    OS << "    Builtin::Info{Builtin::Info::StrOffsets{"
+       << Table.GetStringOffset(Def->getBuiltinName()) << " /* "
+       << Def->getBuiltinName() << " */, ";
+    if (Def->hasBuiltinAlias()) {
+      OS << "0, ";
+    } else {
+      OS << Table.GetStringOffset(Def->getBuiltinTypeStr()) << " /* "
+         << Def->getBuiltinTypeStr() << " */, ";
+    }
+    OS << Table.GetStringOffset("n") << " /* n */, ";
+    OS << Table.GetStringOffset("zve32x") << " /* zve32x */}, ";
+
+    OS << "HeaderDesc::NO_HEADER, ALL_LANGUAGES},\n";
   }
-  OS << "#undef RISCVV_BUILTIN\n";
+  OS << "#endif // GET_RISCVV_BUILTIN_INFOS\n\n";
 }
 
 void RVVEmitter::createCodeGen(raw_ostream &OS) {
