Merge remote-tracking branch 'llvm_be_very_careful/main' into corentin/structured_binding_size

Author: cor3ntin

clang/docs/ClangOffloadBundler.rst

@@ -266,15 +266,14 @@ without differentiation based on offload kind.
     The target triple of the code object. See `Target Triple
     <https://clang.llvm.org/docs/CrossCompilation.html#target-triple>`_.
 
-    The bundler accepts target triples with or without the optional environment
-    field:
+    LLVM target triples can be with or without the optional environment field:
 
     ``<arch><sub>-<vendor>-<sys>``, or
     ``<arch><sub>-<vendor>-<sys>-<env>``
 
-    However, in order to standardize outputs for tools that consume bitcode
-    bundles, bundles written by the bundler internally use only the 4-field
-    target triple:
+    However, in order to standardize outputs for tools that consume bitcode bundles
+    and to parse target ID containing dashes, the bundler only accepts target
+    triples in the 4-field format:
 
     ``<arch><sub>-<vendor>-<sys>-<env>``
 
@@ -543,4 +542,4 @@ The compressed offload bundle begins with a header followed by the compressed bi
 - **Compressed Data**:
     The actual compressed binary data follows the header. Its size can be inferred from the total size of the file minus the header size.
 
-    > **Note**: Version 3 of the format is under development. It uses 64-bit fields for Total File Size and Uncompressed Binary Size to support files larger than 4GB. To experiment with version 3, set the environment variable `COMPRESSED_BUNDLE_FORMAT_VERSION=3`. This support is experimental and not recommended for production use.
+    > **Note**: Version 3 of the format is under development. It uses 64-bit fields for Total File Size and Uncompressed Binary Size to support files larger than 4GB. To experiment with version 3, set the environment variable `COMPRESSED_BUNDLE_FORMAT_VERSION=3`. This support is experimental and not recommended for production use.

clang/docs/LanguageExtensions.rst

@@ -1652,6 +1652,7 @@ Designated initializers (N494)                                                 C
 Array & element qualification (N2607)                                          C23           C89
 Attributes (N2335)                                                             C23           C89
 ``#embed`` (N3017)                                                             C23           C89, C++
+Octal literals prefixed with ``0o`` or ``0O``                                  C2y           C89, C++
 ============================================= ================================ ============= =============
 
 Builtin type aliases

clang/docs/ReleaseNotes.rst

@@ -132,6 +132,13 @@ C2y Feature Support
 - Implemented `WG14 N3411 <https://www.open-std.org/jtc1/sc22/wg14/www/docs/n3411.pdf>`_
   which allows a source file to not end with a newline character. This is still
   reported as a conforming extension in earlier language modes.
+- Implemented `WG14 N3353 <https://www.open-std.org/jtc1/sc22/wg14/www/docs/n3353.htm>_`
+  which adds the new ``0o`` and ``0O`` ocal literal prefixes and deprecates
+  octal literals other than ``0`` which do not start with the new prefix. This
+  feature is exposed in earlier language modes and in C++ as an extension. The
+  paper also introduced octal and hexadecimal delimited escape sequences (e.g.,
+  ``"\x{12}\o{12}"``) which are also supported as an extension in older C
+  language modes.
 
 C23 Feature Support
 ^^^^^^^^^^^^^^^^^^^

clang/include/clang/Basic/DiagnosticGroups.td

@@ -92,6 +92,7 @@ def EnumCompare : DiagGroup<"enum-compare", [EnumCompareSwitch,
 def DeprecatedAnonEnumEnumConversion : DiagGroup<"deprecated-anon-enum-enum-conversion">;
 def DeprecatedEnumEnumConversion : DiagGroup<"deprecated-enum-enum-conversion">;
 def DeprecatedEnumFloatConversion : DiagGroup<"deprecated-enum-float-conversion">;
+def DeprecatedOctalLiterals : DiagGroup<"deprecated-octal-literals">;
 def AnonEnumEnumConversion : DiagGroup<"anon-enum-enum-conversion",
                                    [DeprecatedAnonEnumEnumConversion]>;
 def EnumEnumConversion : DiagGroup<"enum-enum-conversion",
@@ -235,7 +236,8 @@ def Deprecated : DiagGroup<"deprecated", [DeprecatedAnonEnumEnumConversion,
                                           DeprecatedVolatile,
                                           DeprecatedWritableStr,
                                           DeprecatedRedundantConstexprStaticDef,
-                                          DeprecatedMissingCommaVariadicParam
+                                          DeprecatedMissingCommaVariadicParam,
+                                          DeprecatedOctalLiterals
                                           ]>,
                  DiagCategory<"Deprecations">;
 

clang/include/clang/Basic/DiagnosticLexKinds.td

@@ -148,14 +148,14 @@ def ext_mathematical_notation : ExtWarn<
   InGroup<DiagGroup<"mathematical-notation-identifier-extension">>;
 
 def ext_delimited_escape_sequence : Extension<
-  "%select{delimited|named}0 escape sequences are a "
-  "%select{Clang|C++23}1 extension">,
-  InGroup<DiagGroup<"delimited-escape-sequence-extension">>;
-
+  "%select{delimited|named}0 escape sequences are a %select{C++23|C2y|Clang}1 "
+  "extension">, InGroup<DiagGroup<"delimited-escape-sequence-extension">>;
 def warn_cxx23_delimited_escape_sequence : Warning<
-  "%select{delimited|named}0 escape sequences are "
-  "incompatible with C++ standards before C++23">,
-  InGroup<CXXPre23Compat>, DefaultIgnore;
+  "%select{delimited|named}0 escape sequences are incompatible with C++ "
+  "standards before C++23">, InGroup<CXXPre23Compat>, DefaultIgnore;
+def warn_c2y_delimited_escape_sequence : Warning<
+  "delimited escape sequences are incompatible with C standards before C2y">,
+  InGroup<CPre2yCompat>, DefaultIgnore;
 
 def err_delimited_escape_empty : Error<
   "delimited escape sequence cannot be empty">;
@@ -256,6 +256,17 @@ def warn_cxx17_hex_literal : Warning<
   "hexadecimal floating literals are incompatible with "
   "C++ standards before C++17">,
   InGroup<CXXPre17CompatPedantic>, DefaultIgnore;
+def ext_octal_literal : Extension<
+  "octal integer literals are a C2y extension">, InGroup<C2y>;
+def ext_cpp_octal_literal : Extension<
+  "octal integer literals are a Clang extension">,
+  InGroup<DiagGroup<"octal-prefix-extension">>;
+def warn_c2y_compat_octal_literal : Warning<
+  "octal integer literals are incompatible with standards before C2y">,
+  InGroup<CPre2yCompat>, DefaultIgnore;
+def warn_unprefixed_octal_deprecated : Warning<
+  "octal literals without a '0o' prefix are deprecated">,
+  InGroup<DeprecatedOctalLiterals>;
 def ext_binary_literal : Extension<
   "binary integer literals are a C23 extension">, InGroup<C23>;
 def warn_c23_compat_binary_literal : Warning<

clang/include/clang/Driver/OffloadBundler.h

@@ -158,6 +158,11 @@ class CompressedOffloadBundle {
   static llvm::Expected<std::unique_ptr<llvm::MemoryBuffer>>
   decompress(const llvm::MemoryBuffer &Input, bool Verbose = false);
 };
+
+/// Check whether the bundle id is in the following format:
+/// <kind>-<triple>[-<target id>[:target features]]
+/// <triple> := <arch>-<vendor>-<os>-<env>
+bool checkOffloadBundleID(const llvm::StringRef Str);
 } // namespace clang
 
 #endif // LLVM_CLANG_DRIVER_OFFLOADBUNDLER_H

clang/include/clang/Lex/Lexer.h

@@ -582,6 +582,12 @@ class Lexer : public PreprocessorLexer {
   /// sequence.
   static bool isNewLineEscaped(const char *BufferStart, const char *Str);
 
+  /// Diagnose use of a delimited or named escape sequence.
+  static void DiagnoseDelimitedOrNamedEscapeSequence(SourceLocation Loc,
+                                                     bool Named,
+                                                     const LangOptions &Opts,
+                                                     DiagnosticsEngine &Diags);
+
   /// Represents a char and the number of bytes parsed to produce it.
   struct SizedChar {
     char Char;

clang/lib/Driver/OffloadBundler.cpp

@@ -83,32 +83,27 @@ OffloadTargetInfo::OffloadTargetInfo(const StringRef Target,
                                      const OffloadBundlerConfig &BC)
     : BundlerConfig(BC) {
 
-  // TODO: Add error checking from ClangOffloadBundler.cpp
-  auto TargetFeatures = Target.split(':');
-  auto TripleOrGPU = TargetFeatures.first.rsplit('-');
-
-  if (clang::StringToOffloadArch(TripleOrGPU.second) !=
-      clang::OffloadArch::UNKNOWN) {
-    auto KindTriple = TripleOrGPU.first.split('-');
-    this->OffloadKind = KindTriple.first;
-
-    // Enforce optional env field to standardize bundles
-    llvm::Triple t = llvm::Triple(KindTriple.second);
-    this->Triple = llvm::Triple(t.getArchName(), t.getVendorName(),
-                                t.getOSName(), t.getEnvironmentName());
-
-    this->TargetID = Target.substr(Target.find(TripleOrGPU.second));
-  } else {
-    auto KindTriple = TargetFeatures.first.split('-');
-    this->OffloadKind = KindTriple.first;
-
-    // Enforce optional env field to standardize bundles
-    llvm::Triple t = llvm::Triple(KindTriple.second);
-    this->Triple = llvm::Triple(t.getArchName(), t.getVendorName(),
-                                t.getOSName(), t.getEnvironmentName());
-
+  // <kind>-<triple>[-<target id>[:target features]]
+  // <triple> := <arch>-<vendor>-<os>-<env>
+  SmallVector<StringRef, 6> Components;
+  Target.split(Components, '-', /*MaxSplit=*/5);
+  assert((Components.size() == 5 || Components.size() == 6) &&
+         "malformed target string");
+
+  StringRef TargetIdWithFeature =
+      Components.size() == 6 ? Components.back() : "";
+  StringRef TargetId = TargetIdWithFeature.split(':').first;
+  if (!TargetId.empty() &&
+      clang::StringToOffloadArch(TargetId) != clang::OffloadArch::UNKNOWN)
+    this->TargetID = TargetIdWithFeature;
+  else
     this->TargetID = "";
-  }
+
+  this->OffloadKind = Components.front();
+  ArrayRef<StringRef> TripleSlice{&Components[1], /*length=*/4};
+  llvm::Triple T = llvm::Triple(llvm::join(TripleSlice, "-"));
+  this->Triple = llvm::Triple(T.getArchName(), T.getVendorName(), T.getOSName(),
+                              T.getEnvironmentName());
 }
 
 bool OffloadTargetInfo::hasHostKind() const {
@@ -148,7 +143,18 @@ bool OffloadTargetInfo::operator==(const OffloadTargetInfo &Target) const {
 }
 
 std::string OffloadTargetInfo::str() const {
-  return Twine(OffloadKind + "-" + Triple.str() + "-" + TargetID).str();
+  std::string NormalizedTriple;
+  // Unfortunately we need some special sauce for AMDGPU because all the runtime
+  // assumes the triple to be "amdgcn-amd-amdhsa-" (empty environment) instead
+  // of "amdgcn-amd-amdhsa-unknown". It's gonna be very tricky to patch
+  // different layers of runtime.
+  if (Triple.isAMDGPU()) {
+    NormalizedTriple = Triple.normalize(Triple::CanonicalForm::THREE_IDENT);
+    NormalizedTriple.push_back('-');
+  } else {
+    NormalizedTriple = Triple.normalize(Triple::CanonicalForm::FOUR_IDENT);
+  }
+  return Twine(OffloadKind + "-" + NormalizedTriple + "-" + TargetID).str();
 }
 
 static StringRef getDeviceFileExtension(StringRef Device,
@@ -1507,6 +1513,9 @@ Error OffloadBundler::UnbundleFiles() {
   StringMap<StringRef> Worklist;
   auto Output = BundlerConfig.OutputFileNames.begin();
   for (auto &Triple : BundlerConfig.TargetNames) {
+    if (!checkOffloadBundleID(Triple))
+      return createStringError(errc::invalid_argument,
+                               "invalid bundle id from bundle config");
     Worklist[Triple] = *Output;
     ++Output;
   }
@@ -1526,6 +1535,9 @@ Error OffloadBundler::UnbundleFiles() {
 
     StringRef CurTriple = **CurTripleOrErr;
     assert(!CurTriple.empty());
+    if (!checkOffloadBundleID(CurTriple))
+      return createStringError(errc::invalid_argument,
+                               "invalid bundle id read from the bundle");
 
     auto Output = Worklist.begin();
     for (auto E = Worklist.end(); Output != E; Output++) {
@@ -1584,6 +1596,8 @@ Error OffloadBundler::UnbundleFiles() {
         return createFileError(E.second, EC);
 
       // If this entry has a host kind, copy the input file to the output file.
+      // We don't need to check E.getKey() here through checkOffloadBundleID
+      // because the entire WorkList has been checked above.
       auto OffloadInfo = OffloadTargetInfo(E.getKey(), BundlerConfig);
       if (OffloadInfo.hasHostKind())
         OutputFile.write(Input.getBufferStart(), Input.getBufferSize());
@@ -1813,6 +1827,10 @@ Error OffloadBundler::UnbundleArchive() {
     // archive.
     while (!CodeObject.empty()) {
       SmallVector<StringRef> CompatibleTargets;
+      if (!checkOffloadBundleID(CodeObject)) {
+        return createStringError(errc::invalid_argument,
+                                 "Invalid bundle id read from code object");
+      }
       auto CodeObjectInfo = OffloadTargetInfo(CodeObject, BundlerConfig);
       if (getCompatibleOffloadTargets(CodeObjectInfo, CompatibleTargets,
                                       BundlerConfig)) {
@@ -1894,3 +1912,11 @@ Error OffloadBundler::UnbundleArchive() {
 
   return Error::success();
 }
+
+bool clang::checkOffloadBundleID(const llvm::StringRef Str) {
+  // <kind>-<triple>[-<target id>[:target features]]
+  // <triple> := <arch>-<vendor>-<os>-<env>
+  SmallVector<StringRef, 6> Components;
+  Str.split(Components, '-', /*MaxSplit=*/5);
+  return Components.size() == 5 || Components.size() == 6;
+}

clang/lib/Driver/ToolChains/Clang.cpp

@@ -8978,7 +8978,8 @@ void OffloadBundler::ConstructJob(Compilation &C, const JobAction &JA,
     }
     Triples += Action::GetOffloadKindName(CurKind);
     Triples += '-';
-    Triples += CurTC->getTriple().normalize();
+    Triples +=
+        CurTC->getTriple().normalize(llvm::Triple::CanonicalForm::FOUR_IDENT);
     if ((CurKind == Action::OFK_HIP || CurKind == Action::OFK_Cuda) &&
         !StringRef(CurDep->getOffloadingArch()).empty()) {
       Triples += '-';
@@ -9072,7 +9073,8 @@ void OffloadBundler::ConstructJobMultipleOutputs(
     auto &Dep = DepInfo[I];
     Triples += Action::GetOffloadKindName(Dep.DependentOffloadKind);
     Triples += '-';
-    Triples += Dep.DependentToolChain->getTriple().normalize();
+    Triples += Dep.DependentToolChain->getTriple().normalize(
+        llvm::Triple::CanonicalForm::FOUR_IDENT);
     if ((Dep.DependentOffloadKind == Action::OFK_HIP ||
          Dep.DependentOffloadKind == Action::OFK_Cuda) &&
         !Dep.DependentBoundArch.empty()) {

clang/lib/Driver/ToolChains/CommonArgs.cpp

@@ -2577,7 +2577,8 @@ static void GetSDLFromOffloadArchive(
   SmallString<128> DeviceTriple;
   DeviceTriple += Action::GetOffloadKindName(JA.getOffloadingDeviceKind());
   DeviceTriple += '-';
-  std::string NormalizedTriple = T.getToolChain().getTriple().normalize();
+  std::string NormalizedTriple = T.getToolChain().getTriple().normalize(
+      llvm::Triple::CanonicalForm::FOUR_IDENT);
   DeviceTriple += NormalizedTriple;
   if (!Target.empty()) {
     DeviceTriple += '-';

clang/lib/Driver/ToolChains/HIPUtility.cpp

@@ -45,7 +45,7 @@ static std::string normalizeForBundler(const llvm::Triple &T,
   return HasTargetID ? (T.getArchName() + "-" + T.getVendorName() + "-" +
                         T.getOSName() + "-" + T.getEnvironmentName())
                            .str()
-                     : T.normalize();
+                     : T.normalize(llvm::Triple::CanonicalForm::FOUR_IDENT);
 }
 
 // Collect undefined __hip_fatbin* and __hip_gpubin_handle* symbols from all

clang/lib/Lex/Lexer.cpp

@@ -3405,6 +3405,30 @@ bool Lexer::isCodeCompletionPoint(const char *CurPtr) const {
   return false;
 }
 
+void Lexer::DiagnoseDelimitedOrNamedEscapeSequence(SourceLocation Loc,
+                                                   bool Named,
+                                                   const LangOptions &Opts,
+                                                   DiagnosticsEngine &Diags) {
+  unsigned DiagId;
+  if (Opts.CPlusPlus23)
+    DiagId = diag::warn_cxx23_delimited_escape_sequence;
+  else if (Opts.C2y && !Named)
+    DiagId = diag::warn_c2y_delimited_escape_sequence;
+  else
+    DiagId = diag::ext_delimited_escape_sequence;
+
+  // The trailing arguments are only used by the extension warning; either this
+  // is a C2y extension or a C++23 extension, unless it's a named escape
+  // sequence in C, then it's a Clang extension.
+  unsigned Ext;
+  if (!Opts.CPlusPlus)
+    Ext = Named ? 2 /* Clang extension */ : 1 /* C2y extension */;
+  else
+    Ext = 0; // C++23 extension
+
+  Diags.Report(Loc, DiagId) << Named << Ext;
+}
+
 std::optional<uint32_t> Lexer::tryReadNumericUCN(const char *&StartPtr,
                                                  const char *SlashLoc,
                                                  Token *Result) {
@@ -3496,12 +3520,10 @@ std::optional<uint32_t> Lexer::tryReadNumericUCN(const char *&StartPtr,
     return std::nullopt;
   }
 
-  if (Delimited && PP) {
-    Diag(SlashLoc, PP->getLangOpts().CPlusPlus23
-                       ? diag::warn_cxx23_delimited_escape_sequence
-                       : diag::ext_delimited_escape_sequence)
-        << /*delimited*/ 0 << (PP->getLangOpts().CPlusPlus ? 1 : 0);
-  }
+  if (Delimited && PP)
+    DiagnoseDelimitedOrNamedEscapeSequence(getSourceLocation(SlashLoc), false,
+                                           PP->getLangOpts(),
+                                           PP->getDiagnostics());
 
   if (Result) {
     Result->setFlag(Token::HasUCN);
@@ -3585,10 +3607,9 @@ std::optional<uint32_t> Lexer::tryReadNamedUCN(const char *&StartPtr,
   }
 
   if (Diagnose && Match)
-    Diag(SlashLoc, PP->getLangOpts().CPlusPlus23
-                       ? diag::warn_cxx23_delimited_escape_sequence
-                       : diag::ext_delimited_escape_sequence)
-        << /*named*/ 1 << (PP->getLangOpts().CPlusPlus ? 1 : 0);
+    DiagnoseDelimitedOrNamedEscapeSequence(getSourceLocation(SlashLoc), true,
+                                           PP->getLangOpts(),
+                                           PP->getDiagnostics());
 
   // If no diagnostic has been emitted yet, likely because we are doing a
   // tentative lexing, we do not want to recover here to make sure the token