diff --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst
index c036f66d60bf7a..e810b92c492712 100644
--- a/clang/docs/ReleaseNotes.rst
+++ b/clang/docs/ReleaseNotes.rst
@@ -188,6 +188,8 @@ X86 Support in Clang
 - The x86 intrinsics ``__rorb``, ``__rorw``, ``__rord``, ``__rorq`, ``_rotr``,
   ``_rotwr`` and ``_lrotr`` may now be used within constant expressions.
 
+- Support for -march=sapphirerapids was added.
+
 Internal API Changes
 --------------------
 
diff --git a/clang/lib/CodeGen/CGDebugInfo.cpp b/clang/lib/CodeGen/CGDebugInfo.cpp
index e3442ecd4bd59f..36bab9c22d4378 100644
--- a/clang/lib/CodeGen/CGDebugInfo.cpp
+++ b/clang/lib/CodeGen/CGDebugInfo.cpp
@@ -2260,6 +2260,25 @@ static bool hasExplicitMemberDefinition(CXXRecordDecl::method_iterator I,
   return false;
 }
 
+static bool canUseCtorHoming(const CXXRecordDecl *RD) {
+  // Constructor homing can be used for classes that have at least one
+  // constructor and have no trivial or constexpr constructors.
+  // Skip this optimization if the class or any of its methods are marked
+  // dllimport.
+  if (RD->isLambda() || RD->hasConstexprNonCopyMoveConstructor() ||
+      isClassOrMethodDLLImport(RD))
+    return false;
+
+  if (RD->ctors().empty())
+    return false;
+
+  for (const auto *Ctor : RD->ctors())
+    if (Ctor->isTrivial() && !Ctor->isCopyOrMoveConstructor())
+      return false;
+
+  return true;
+}
+
 static bool shouldOmitDefinition(codegenoptions::DebugInfoKind DebugKind,
                                  bool DebugTypeExtRefs, const RecordDecl *RD,
                                  const LangOptions &LangOpts) {
@@ -2294,23 +2313,6 @@ static bool shouldOmitDefinition(codegenoptions::DebugInfoKind DebugKind,
       !isClassOrMethodDLLImport(CXXDecl))
     return true;
 
-  // In constructor debug mode, only emit debug info for a class when its
-  // constructor is emitted. Skip this optimization if the class or any of
-  // its methods are marked dllimport.
-  //
-  // This applies to classes that don't have any trivial constructors and have
-  // at least one constructor.
-  if (DebugKind == codegenoptions::DebugInfoConstructor &&
-      !CXXDecl->isLambda() && !CXXDecl->hasConstexprNonCopyMoveConstructor() &&
-      !isClassOrMethodDLLImport(CXXDecl)) {
-    if (CXXDecl->ctors().empty())
-      return false;
-    for (const auto *Ctor : CXXDecl->ctors())
-      if (Ctor->isTrivial() && !Ctor->isCopyOrMoveConstructor())
-        return false;
-    return true;
-  }
-
   TemplateSpecializationKind Spec = TSK_Undeclared;
   if (const auto *SD = dyn_cast<ClassTemplateSpecializationDecl>(RD))
     Spec = SD->getSpecializationKind();
@@ -2320,6 +2322,12 @@ static bool shouldOmitDefinition(codegenoptions::DebugInfoKind DebugKind,
                                   CXXDecl->method_end()))
     return true;
 
+  // In constructor homing mode, only emit complete debug info for a class
+  // when its constructor is emitted.
+  if ((DebugKind == codegenoptions::DebugInfoConstructor) &&
+      canUseCtorHoming(CXXDecl))
+    return true;
+
   return false;
 }
 
diff --git a/clang/test/CodeGenCXX/debug-info-template-explicit-specialization.cpp b/clang/test/CodeGenCXX/debug-info-template-explicit-specialization.cpp
index 4e41c4092bf4e3..b756674f54c405 100644
--- a/clang/test/CodeGenCXX/debug-info-template-explicit-specialization.cpp
+++ b/clang/test/CodeGenCXX/debug-info-template-explicit-specialization.cpp
@@ -1,5 +1,8 @@
 // RUN: %clang_cc1 -emit-llvm -triple %itanium_abi_triple -debug-info-kind=limited %s -o - | FileCheck %s
 
+// Make sure this still works with constructor homing.
+// RUN: %clang_cc1 -emit-llvm -triple %itanium_abi_triple -debug-info-kind=constructor %s -o - | FileCheck %s
+
 // Run again with -gline-tables-only or -gline-directives-only and verify we don't crash.  We won't output
 // type info at all.
 // RUN: %clang_cc1 -emit-llvm -triple %itanium_abi_triple -debug-info-kind=line-tables-only %s -o - | FileCheck %s -check-prefix LINES-ONLY
diff --git a/compiler-rt/lib/sanitizer_common/sanitizer_ptrauth.h b/compiler-rt/lib/sanitizer_common/sanitizer_ptrauth.h
index 4d0d96a64f622f..a288068bf94383 100644
--- a/compiler-rt/lib/sanitizer_common/sanitizer_ptrauth.h
+++ b/compiler-rt/lib/sanitizer_common/sanitizer_ptrauth.h
@@ -18,4 +18,6 @@
 #define ptrauth_string_discriminator(__string) ((int)0)
 #endif
 
+#define STRIP_PC(pc) ((uptr)ptrauth_strip(pc, 0))
+
 #endif // SANITIZER_PTRAUTH_H
diff --git a/compiler-rt/lib/tsan/rtl/tsan_external.cpp b/compiler-rt/lib/tsan/rtl/tsan_external.cpp
index 0faa1ee93a139e..466b2bf0f66ce5 100644
--- a/compiler-rt/lib/tsan/rtl/tsan_external.cpp
+++ b/compiler-rt/lib/tsan/rtl/tsan_external.cpp
@@ -11,6 +11,7 @@
 //===----------------------------------------------------------------------===//
 #include "tsan_rtl.h"
 #include "tsan_interceptors.h"
+#include "sanitizer_common/sanitizer_ptrauth.h"
 
 namespace __tsan {
 
@@ -57,13 +58,13 @@ uptr TagFromShadowStackFrame(uptr pc) {
 #if !SANITIZER_GO
 
 typedef void(*AccessFunc)(ThreadState *, uptr, uptr, int);
-void ExternalAccess(void *addr, void *caller_pc, void *tag, AccessFunc access) {
+void ExternalAccess(void *addr, uptr caller_pc, void *tag, AccessFunc access) {
   CHECK_LT(tag, atomic_load(&used_tags, memory_order_relaxed));
   ThreadState *thr = cur_thread();
-  if (caller_pc) FuncEntry(thr, (uptr)caller_pc);
+  if (caller_pc) FuncEntry(thr, caller_pc);
   InsertShadowStackFrameForTag(thr, (uptr)tag);
   bool in_ignored_lib;
-  if (!caller_pc || !libignore()->IsIgnored((uptr)caller_pc, &in_ignored_lib)) {
+  if (!caller_pc || !libignore()->IsIgnored(caller_pc, &in_ignored_lib)) {
     access(thr, CALLERPC, (uptr)addr, kSizeLog1);
   }
   FuncExit(thr);
@@ -110,12 +111,12 @@ void __tsan_external_assign_tag(void *addr, void *tag) {
 
 SANITIZER_INTERFACE_ATTRIBUTE
 void __tsan_external_read(void *addr, void *caller_pc, void *tag) {
-  ExternalAccess(addr, caller_pc, tag, MemoryRead);
+  ExternalAccess(addr, STRIP_PC(caller_pc), tag, MemoryRead);
 }
 
 SANITIZER_INTERFACE_ATTRIBUTE
 void __tsan_external_write(void *addr, void *caller_pc, void *tag) {
-  ExternalAccess(addr, caller_pc, tag, MemoryWrite);
+  ExternalAccess(addr, STRIP_PC(caller_pc), tag, MemoryWrite);
 }
 }  // extern "C"
 
diff --git a/compiler-rt/lib/tsan/rtl/tsan_interface.cpp b/compiler-rt/lib/tsan/rtl/tsan_interface.cpp
index 2b3a0889b70a4a..5c2a617a24c3d6 100644
--- a/compiler-rt/lib/tsan/rtl/tsan_interface.cpp
+++ b/compiler-rt/lib/tsan/rtl/tsan_interface.cpp
@@ -14,6 +14,7 @@
 #include "tsan_interface_ann.h"
 #include "tsan_rtl.h"
 #include "sanitizer_common/sanitizer_internal_defs.h"
+#include "sanitizer_common/sanitizer_ptrauth.h"
 
 #define CALLERPC ((uptr)__builtin_return_address(0))
 
@@ -43,13 +44,13 @@ void __tsan_write16(void *addr) {
 }
 
 void __tsan_read16_pc(void *addr, void *pc) {
-  MemoryRead(cur_thread(), (uptr)pc, (uptr)addr, kSizeLog8);
-  MemoryRead(cur_thread(), (uptr)pc, (uptr)addr + 8, kSizeLog8);
+  MemoryRead(cur_thread(), STRIP_PC(pc), (uptr)addr, kSizeLog8);
+  MemoryRead(cur_thread(), STRIP_PC(pc), (uptr)addr + 8, kSizeLog8);
 }
 
 void __tsan_write16_pc(void *addr, void *pc) {
-  MemoryWrite(cur_thread(), (uptr)pc, (uptr)addr, kSizeLog8);
-  MemoryWrite(cur_thread(), (uptr)pc, (uptr)addr + 8, kSizeLog8);
+  MemoryWrite(cur_thread(), STRIP_PC(pc), (uptr)addr, kSizeLog8);
+  MemoryWrite(cur_thread(), STRIP_PC(pc), (uptr)addr + 8, kSizeLog8);
 }
 
 // __tsan_unaligned_read/write calls are emitted by compiler.
diff --git a/compiler-rt/lib/tsan/rtl/tsan_interface_inl.h b/compiler-rt/lib/tsan/rtl/tsan_interface_inl.h
index f955ddf99247c5..f5d743c10772ec 100644
--- a/compiler-rt/lib/tsan/rtl/tsan_interface_inl.h
+++ b/compiler-rt/lib/tsan/rtl/tsan_interface_inl.h
@@ -12,6 +12,7 @@
 
 #include "tsan_interface.h"
 #include "tsan_rtl.h"
+#include "sanitizer_common/sanitizer_ptrauth.h"
 
 #define CALLERPC ((uptr)__builtin_return_address(0))
 
@@ -50,35 +51,35 @@ void __tsan_write8(void *addr) {
 }
 
 void __tsan_read1_pc(void *addr, void *pc) {
-  MemoryRead(cur_thread(), (uptr)pc, (uptr)addr, kSizeLog1);
+  MemoryRead(cur_thread(), STRIP_PC(pc), (uptr)addr, kSizeLog1);
 }
 
 void __tsan_read2_pc(void *addr, void *pc) {
-  MemoryRead(cur_thread(), (uptr)pc, (uptr)addr, kSizeLog2);
+  MemoryRead(cur_thread(), STRIP_PC(pc), (uptr)addr, kSizeLog2);
 }
 
 void __tsan_read4_pc(void *addr, void *pc) {
-  MemoryRead(cur_thread(), (uptr)pc, (uptr)addr, kSizeLog4);
+  MemoryRead(cur_thread(), STRIP_PC(pc), (uptr)addr, kSizeLog4);
 }
 
 void __tsan_read8_pc(void *addr, void *pc) {
-  MemoryRead(cur_thread(), (uptr)pc, (uptr)addr, kSizeLog8);
+  MemoryRead(cur_thread(), STRIP_PC(pc), (uptr)addr, kSizeLog8);
 }
 
 void __tsan_write1_pc(void *addr, void *pc) {
-  MemoryWrite(cur_thread(), (uptr)pc, (uptr)addr, kSizeLog1);
+  MemoryWrite(cur_thread(), STRIP_PC(pc), (uptr)addr, kSizeLog1);
 }
 
 void __tsan_write2_pc(void *addr, void *pc) {
-  MemoryWrite(cur_thread(), (uptr)pc, (uptr)addr, kSizeLog2);
+  MemoryWrite(cur_thread(), STRIP_PC(pc), (uptr)addr, kSizeLog2);
 }
 
 void __tsan_write4_pc(void *addr, void *pc) {
-  MemoryWrite(cur_thread(), (uptr)pc, (uptr)addr, kSizeLog4);
+  MemoryWrite(cur_thread(), STRIP_PC(pc), (uptr)addr, kSizeLog4);
 }
 
 void __tsan_write8_pc(void *addr, void *pc) {
-  MemoryWrite(cur_thread(), (uptr)pc, (uptr)addr, kSizeLog8);
+  MemoryWrite(cur_thread(), STRIP_PC(pc), (uptr)addr, kSizeLog8);
 }
 
 void __tsan_vptr_update(void **vptr_p, void *new_val) {
@@ -100,7 +101,7 @@ void __tsan_vptr_read(void **vptr_p) {
 }
 
 void __tsan_func_entry(void *pc) {
-  FuncEntry(cur_thread(), (uptr)pc);
+  FuncEntry(cur_thread(), STRIP_PC(pc));
 }
 
 void __tsan_func_exit() {
@@ -124,9 +125,9 @@ void __tsan_write_range(void *addr, uptr size) {
 }
 
 void __tsan_read_range_pc(void *addr, uptr size, void *pc) {
-  MemoryAccessRange(cur_thread(), (uptr)pc, (uptr)addr, size, false);
+  MemoryAccessRange(cur_thread(), STRIP_PC(pc), (uptr)addr, size, false);
 }
 
 void __tsan_write_range_pc(void *addr, uptr size, void *pc) {
-  MemoryAccessRange(cur_thread(), (uptr)pc, (uptr)addr, size, true);
+  MemoryAccessRange(cur_thread(), STRIP_PC(pc), (uptr)addr, size, true);
 }
diff --git a/flang/include/flang/Parser/characters.h b/flang/include/flang/Parser/characters.h
index eefb524e5d879b..120560625da29c 100644
--- a/flang/include/flang/Parser/characters.h
+++ b/flang/include/flang/Parser/characters.h
@@ -151,6 +151,33 @@ inline constexpr std::optional<char> BackslashEscapeChar(char ch) {
   }
 }
 
+// Does not include spaces or line ending characters.
+inline constexpr bool IsValidFortranTokenCharacter(char ch) {
+  switch (ch) {
+  case '"':
+  case '%':
+  case '\'':
+  case '(':
+  case ')':
+  case '*':
+  case '+':
+  case ',':
+  case '-':
+  case '.':
+  case '/':
+  case ':':
+  case ';':
+  case '<':
+  case '=':
+  case '>':
+  case '[':
+  case ']':
+    return true;
+  default:
+    return IsLegalIdentifierStart(ch) || IsDecimalDigit(ch);
+  }
+}
+
 struct EncodedCharacter {
   static constexpr int maxEncodingBytes{6};
   char buffer[maxEncodingBytes];
diff --git a/flang/include/flang/Parser/parse-tree.h b/flang/include/flang/Parser/parse-tree.h
index cbff637007755e..317d772889eee3 100644
--- a/flang/include/flang/Parser/parse-tree.h
+++ b/flang/include/flang/Parser/parse-tree.h
@@ -547,7 +547,8 @@ struct ProgramUnit {
   std::variant<common::Indirection<MainProgram>,
       common::Indirection<FunctionSubprogram>,
       common::Indirection<SubroutineSubprogram>, common::Indirection<Module>,
-      common::Indirection<Submodule>, common::Indirection<BlockData>>
+      common::Indirection<Submodule>, common::Indirection<BlockData>,
+      common::Indirection<CompilerDirective>>
       u;
 };
 
diff --git a/flang/lib/Parser/prescan.cpp b/flang/lib/Parser/prescan.cpp
index 9e90f7f6228feb..5e6f13797646b8 100644
--- a/flang/lib/Parser/prescan.cpp
+++ b/flang/lib/Parser/prescan.cpp
@@ -184,7 +184,8 @@ void Prescanner::Statement() {
     case LineClassification::Kind::PreprocessorDirective:
       Say(preprocessed->GetProvenanceRange(),
           "Preprocessed line resembles a preprocessor directive"_en_US);
-      preprocessed->ToLowerCase().Emit(cooked_);
+      preprocessed->ToLowerCase().CheckBadFortranCharacters(messages_).Emit(
+          cooked_);
       break;
     case LineClassification::Kind::CompilerDirective:
       if (preprocessed->HasRedundantBlanks()) {
@@ -193,7 +194,9 @@ void Prescanner::Statement() {
       NormalizeCompilerDirectiveCommentMarker(*preprocessed);
       preprocessed->ToLowerCase();
       SourceFormChange(preprocessed->ToString());
-      preprocessed->ClipComment(true /* skip first ! */).Emit(cooked_);
+      preprocessed->ClipComment(true /* skip first ! */)
+          .CheckBadFortranCharacters(messages_)
+          .Emit(cooked_);
       break;
     case LineClassification::Kind::Source:
       if (inFixedForm_) {
@@ -205,7 +208,10 @@ void Prescanner::Statement() {
           preprocessed->RemoveRedundantBlanks();
         }
       }
-      preprocessed->ToLowerCase().ClipComment().Emit(cooked_);
+      preprocessed->ToLowerCase()
+          .ClipComment()
+          .CheckBadFortranCharacters(messages_)
+          .Emit(cooked_);
       break;
     }
   } else {
@@ -213,7 +219,7 @@ void Prescanner::Statement() {
     if (line.kind == LineClassification::Kind::CompilerDirective) {
       SourceFormChange(tokens.ToString());
     }
-    tokens.Emit(cooked_);
+    tokens.CheckBadFortranCharacters(messages_).Emit(cooked_);
   }
   if (omitNewline_) {
     omitNewline_ = false;
@@ -245,8 +251,9 @@ void Prescanner::NextLine() {
   }
 }
 
-void Prescanner::LabelField(TokenSequence &token, int outCol) {
+void Prescanner::LabelField(TokenSequence &token) {
   const char *bad{nullptr};
+  int outCol{1};
   for (; *at_ != '\n' && column_ <= 6; ++at_) {
     if (*at_ == '\t') {
       ++at_;
@@ -256,20 +263,26 @@ void Prescanner::LabelField(TokenSequence &token, int outCol) {
     if (*at_ != ' ' &&
         !(*at_ == '0' && column_ == 6)) { // '0' in column 6 becomes space
       EmitChar(token, *at_);
+      ++outCol;
       if (!bad && !IsDecimalDigit(*at_)) {
         bad = at_;
       }
-      ++outCol;
     }
     ++column_;
   }
-  if (outCol > 1) {
+  if (outCol == 1) { // empty label field
+    // Emit a space so that, if the line is rescanned after preprocessing,
+    // a leading 'C' or 'D' won't be left-justified and then accidentally
+    // misinterpreted as a comment card.
+    EmitChar(token, ' ');
+    ++outCol;
+  } else {
     if (bad && !preprocessor_.IsNameDefined(token.CurrentOpenToken())) {
       Say(GetProvenance(bad),
           "Character in fixed-form label field must be a digit"_en_US);
     }
-    token.CloseToken();
   }
+  token.CloseToken();
   SkipToNextSignificantCharacter();
   if (IsDecimalDigit(*at_)) {
     Say(GetProvenance(at_),
diff --git a/flang/lib/Parser/prescan.h b/flang/lib/Parser/prescan.h
index 595f8f701185b2..0b5b64792004ac 100644
--- a/flang/lib/Parser/prescan.h
+++ b/flang/lib/Parser/prescan.h
@@ -147,7 +147,7 @@ class Prescanner {
                     common::LanguageFeature::ClassicCComments)));
   }
 
-  void LabelField(TokenSequence &, int outCol = 1);
+  void LabelField(TokenSequence &);
   void SkipToEndOfLine();
   bool MustSkipToEndOfLine() const;
   void NextChar();
diff --git a/flang/lib/Parser/program-parsers.cpp b/flang/lib/Parser/program-parsers.cpp
index 278cc6fdb51a5c..dee359e240cfa6 100644
--- a/flang/lib/Parser/program-parsers.cpp
+++ b/flang/lib/Parser/program-parsers.cpp
@@ -20,20 +20,6 @@
 
 namespace Fortran::parser {
 
-// R501 program -> program-unit [program-unit]...
-// This is the top-level production for the Fortran language.
-// F'2018 6.3.1 defines a program unit as a sequence of one or more lines,
-// implying that a line can't be part of two distinct program units.
-// Consequently, a program unit END statement should be the last statement
-// on its line.  We parse those END statements via unterminatedStatement()
-// and then skip over the end of the line here.
-TYPE_PARSER(construct<Program>(
-    extension<LanguageFeature::EmptySourceFile>(skipStuffBeforeStatement >>
-        !nextCh >> pure<std::list<ProgramUnit>>()) ||
-    some(StartNewSubprogram{} >> Parser<ProgramUnit>{} / skipMany(";"_tok) /
-            space / recovery(endOfLine, SkipPast<'\n'>{})) /
-        skipStuffBeforeStatement))
-
 // R502 program-unit ->
 //        main-program | external-subprogram | module | submodule | block-data
 // R503 external-subprogram -> function-subprogram | subroutine-subprogram
@@ -49,12 +35,30 @@ TYPE_PARSER(construct<Program>(
 // variant parsers for several productions; giving the "module" production
 // priority here is a cleaner solution, though regrettably subtle.  Enforcing
 // C1547 is done in semantics.
-TYPE_PARSER(construct<ProgramUnit>(indirect(Parser<Module>{})) ||
+static constexpr auto programUnit{
+    construct<ProgramUnit>(indirect(Parser<Module>{})) ||
     construct<ProgramUnit>(indirect(functionSubprogram)) ||
     construct<ProgramUnit>(indirect(subroutineSubprogram)) ||
     construct<ProgramUnit>(indirect(Parser<Submodule>{})) ||
     construct<ProgramUnit>(indirect(Parser<BlockData>{})) ||
-    construct<ProgramUnit>(indirect(Parser<MainProgram>{})))
+    construct<ProgramUnit>(indirect(Parser<MainProgram>{}))};
+static constexpr auto normalProgramUnit{StartNewSubprogram{} >> programUnit /
+        skipMany(";"_tok) / space / recovery(endOfLine, SkipPast<'\n'>{})};
+static constexpr auto globalCompilerDirective{
+    construct<ProgramUnit>(indirect(compilerDirective))};
+
+// R501 program -> program-unit [program-unit]...
+// This is the top-level production for the Fortran language.
+// F'2018 6.3.1 defines a program unit as a sequence of one or more lines,
+// implying that a line can't be part of two distinct program units.
+// Consequently, a program unit END statement should be the last statement
+// on its line.  We parse those END statements via unterminatedStatement()
+// and then skip over the end of the line here.
+TYPE_PARSER(construct<Program>(
+    extension<LanguageFeature::EmptySourceFile>(skipStuffBeforeStatement >>
+        !nextCh >> pure<std::list<ProgramUnit>>()) ||
+    some(globalCompilerDirective || normalProgramUnit) /
+        skipStuffBeforeStatement))
 
 // R504 specification-part ->
 //         [use-stmt]... [import-stmt]... [implicit-part]
diff --git a/flang/lib/Parser/token-sequence.cpp b/flang/lib/Parser/token-sequence.cpp
index 07c5b12e5f759f..4797cb759a72b2 100644
--- a/flang/lib/Parser/token-sequence.cpp
+++ b/flang/lib/Parser/token-sequence.cpp
@@ -8,6 +8,7 @@
 
 #include "token-sequence.h"
 #include "flang/Parser/characters.h"
+#include "flang/Parser/message.h"
 #include "llvm/Support/raw_ostream.h"
 
 namespace Fortran::parser {
@@ -310,4 +311,25 @@ ProvenanceRange TokenSequence::GetIntervalProvenanceRange(
 ProvenanceRange TokenSequence::GetProvenanceRange() const {
   return GetIntervalProvenanceRange(0, start_.size());
 }
+
+const TokenSequence &TokenSequence::CheckBadFortranCharacters(
+    Messages &messages) const {
+  std::size_t tokens{SizeInTokens()};
+  for (std::size_t j{0}; j < tokens; ++j) {
+    CharBlock token{TokenAt(j)};
+    char ch{token.FirstNonBlank()};
+    if (ch != ' ' && !IsValidFortranTokenCharacter(ch)) {
+      if (ch == '!' && j == 0) {
+        // allow in !dir$
+      } else if (ch < ' ' || ch >= '\x7f') {
+        messages.Say(GetTokenProvenanceRange(j),
+            "bad character (0x%02x) in Fortran token"_err_en_US, ch & 0xff);
+      } else {
+        messages.Say(GetTokenProvenanceRange(j),
+            "bad character ('%c') in Fortran token"_err_en_US, ch);
+      }
+    }
+  }
+  return *this;
+}
 } // namespace Fortran::parser
diff --git a/flang/lib/Parser/token-sequence.h b/flang/lib/Parser/token-sequence.h
index d98c0b955c5e93..6a10ef1977d38a 100644
--- a/flang/lib/Parser/token-sequence.h
+++ b/flang/lib/Parser/token-sequence.h
@@ -27,6 +27,8 @@ class raw_ostream;
 
 namespace Fortran::parser {
 
+class Messages;
+
 // Buffers a contiguous sequence of characters that has been partitioned into
 // a sequence of preprocessing tokens with provenances.
 class TokenSequence {
@@ -115,6 +117,7 @@ class TokenSequence {
   TokenSequence &RemoveBlanks(std::size_t firstChar = 0);
   TokenSequence &RemoveRedundantBlanks(std::size_t firstChar = 0);
   TokenSequence &ClipComment(bool skipFirst = false);
+  const TokenSequence &CheckBadFortranCharacters(Messages &) const;
   void Emit(CookedSource &) const;
   void Dump(llvm::raw_ostream &) const;
 
diff --git a/flang/lib/Semantics/check-io.cpp b/flang/lib/Semantics/check-io.cpp
index 250ad492ebc926..d00f56c38042d0 100644
--- a/flang/lib/Semantics/check-io.cpp
+++ b/flang/lib/Semantics/check-io.cpp
@@ -155,7 +155,8 @@ void IoChecker::Enter(const parser::ConnectSpec::CharExpr &spec) {
   }
 }
 
-void IoChecker::Enter(const parser::ConnectSpec::Newunit &) {
+void IoChecker::Enter(const parser::ConnectSpec::Newunit &var) {
+  CheckForDefinableVariable(var, "NEWUNIT");
   SetSpecifier(IoSpecKind::Newunit);
 }
 
@@ -266,10 +267,11 @@ void IoChecker::Enter(const parser::IdExpr &) { SetSpecifier(IoSpecKind::Id); }
 
 void IoChecker::Enter(const parser::IdVariable &spec) {
   SetSpecifier(IoSpecKind::Id);
-  auto expr{GetExpr(spec)};
+  const auto *expr{GetExpr(spec)};
   if (!expr || !expr->GetType()) {
     return;
   }
+  CheckForDefinableVariable(spec, "ID");
   int kind{expr->GetType()->kind()};
   int defaultKind{context_.GetDefaultKind(TypeCategory::Integer)};
   if (kind < defaultKind) {
@@ -281,21 +283,18 @@ void IoChecker::Enter(const parser::IdVariable &spec) {
 
 void IoChecker::Enter(const parser::InputItem &spec) {
   flags_.set(Flag::DataList);
-  if (const parser::Variable * var{std::get_if<parser::Variable>(&spec.u)}) {
-    const parser::Name &name{GetLastName(*var)};
-    if (name.symbol) {
-      if (auto *details{name.symbol->detailsIf<ObjectEntityDetails>()}) {
-        // TODO: Determine if this check is needed at all, and if so, replace
-        // the false subcondition with a check for a whole array.  Otherwise,
-        // the check incorrectly flags array element and section references.
-        if (details->IsAssumedSize() && false) {
-          // This check may be superseded by C928 or C1002.
-          context_.Say(name.source,
-              "'%s' must not be a whole assumed size array"_err_en_US,
-              name.source); // C1231
-        }
-      }
-    }
+  const parser::Variable *var{std::get_if<parser::Variable>(&spec.u)};
+  if (!var) {
+    return;
+  }
+  CheckForDefinableVariable(*var, "Input");
+  const auto &name{GetLastName(*var)};
+  const auto *expr{GetExpr(*var)};
+  if (name.symbol && IsAssumedSizeArray(*name.symbol) && expr &&
+      !evaluate::IsArrayElement(*GetExpr(*var))) {
+    context_.Say(name.source,
+        "Whole assumed size array '%s' may not be an input item"_err_en_US,
+        name.source); // C1231
   }
 }
 
@@ -386,6 +385,8 @@ void IoChecker::Enter(const parser::InquireSpec::CharVar &spec) {
     specKind = IoSpecKind::Dispose;
     break;
   }
+  CheckForDefinableVariable(std::get<parser::ScalarDefaultCharVariable>(spec.t),
+      parser::ToUpperCaseLetters(common::EnumToString(specKind)));
   SetSpecifier(specKind);
 }
 
@@ -412,6 +413,8 @@ void IoChecker::Enter(const parser::InquireSpec::IntVar &spec) {
     specKind = IoSpecKind::Size;
     break;
   }
+  CheckForDefinableVariable(std::get<parser::ScalarIntVariable>(spec.t),
+      parser::ToUpperCaseLetters(common::EnumToString(specKind)));
   SetSpecifier(specKind);
 }
 
@@ -500,17 +503,23 @@ void IoChecker::Enter(const parser::IoControlSpec::Rec &) {
   SetSpecifier(IoSpecKind::Rec);
 }
 
-void IoChecker::Enter(const parser::IoControlSpec::Size &) {
+void IoChecker::Enter(const parser::IoControlSpec::Size &var) {
+  CheckForDefinableVariable(var, "SIZE");
   SetSpecifier(IoSpecKind::Size);
 }
 
 void IoChecker::Enter(const parser::IoUnit &spec) {
   if (const parser::Variable * var{std::get_if<parser::Variable>(&spec.u)}) {
-    // TODO: C1201 - internal file variable must not be an array section ...
-    if (auto expr{GetExpr(*var)}) {
-      if (!ExprTypeKindIsDefault(*expr, context_)) {
+    if (stmt_ == IoStmtKind::Write) {
+      CheckForDefinableVariable(*var, "Internal file");
+    }
+    if (const auto *expr{GetExpr(*var)}) {
+      if (HasVectorSubscript(*expr)) {
+        context_.Say(parser::FindSourceLocation(*var), // C1201
+            "Internal file must not have a vector subscript"_err_en_US);
+      } else if (!ExprTypeKindIsDefault(*expr, context_)) {
         // This may be too restrictive; other kinds may be valid.
-        context_.Say( // C1202
+        context_.Say(parser::FindSourceLocation(*var), // C1202
             "Invalid character kind for an internal file variable"_err_en_US);
       }
     }
@@ -522,13 +531,26 @@ void IoChecker::Enter(const parser::IoUnit &spec) {
   }
 }
 
-void IoChecker::Enter(const parser::MsgVariable &) {
+void IoChecker::Enter(const parser::MsgVariable &var) {
+  if (stmt_ == IoStmtKind::None) {
+    // allocate, deallocate, image control
+    CheckForDefinableVariable(var, "ERRMSG");
+    return;
+  }
+  CheckForDefinableVariable(var, "IOMSG");
   SetSpecifier(IoSpecKind::Iomsg);
 }
 
-void IoChecker::Enter(const parser::OutputItem &) {
+void IoChecker::Enter(const parser::OutputItem &item) {
   flags_.set(Flag::DataList);
-  // TODO: C1233 - output item must not be a procedure pointer
+  if (const auto *x{std::get_if<parser::Expr>(&item.u)}) {
+    if (const auto *expr{GetExpr(*x)}) {
+      if (IsProcedurePointer(*expr)) {
+        context_.Say(parser::FindSourceLocation(*x),
+            "Output item must not be a procedure pointer"_err_en_US); // C1233
+      }
+    }
+  }
 }
 
 void IoChecker::Enter(const parser::StatusExpr &spec) {
@@ -555,12 +577,14 @@ void IoChecker::Enter(const parser::StatusExpr &spec) {
   }
 }
 
-void IoChecker::Enter(const parser::StatVariable &) {
+void IoChecker::Enter(const parser::StatVariable &var) {
   if (stmt_ == IoStmtKind::None) {
-    // ALLOCATE & DEALLOCATE
-  } else {
-    SetSpecifier(IoSpecKind::Iostat);
+    // allocate, deallocate, image control
+    CheckForDefinableVariable(var, "STAT");
+    return;
   }
+  CheckForDefinableVariable(var, "IOSTAT");
+  SetSpecifier(IoSpecKind::Iostat);
 }
 
 void IoChecker::Leave(const parser::BackspaceStmt &) {
@@ -808,7 +832,7 @@ void IoChecker::CheckStringValue(IoSpecKind specKind, const std::string &value,
 
 // CheckForRequiredSpecifier and CheckForProhibitedSpecifier functions
 // need conditions to check, and string arguments to insert into a message.
-// A IoSpecKind provides both an absence/presence condition and a string
+// An IoSpecKind provides both an absence/presence condition and a string
 // argument (its name).  A (condition, string) pair provides an arbitrary
 // condition and an arbitrary string.
 
@@ -893,6 +917,17 @@ void IoChecker::CheckForProhibitedSpecifier(
   }
 }
 
+template <typename A>
+void IoChecker::CheckForDefinableVariable(
+    const A &var, const std::string &s) const {
+  const Symbol *sym{
+      GetFirstName(*parser::Unwrap<parser::Variable>(var)).symbol};
+  if (WhyNotModifiable(*sym, context_.FindScope(*context_.location()))) {
+    context_.Say(parser::FindSourceLocation(var),
+        "%s variable '%s' must be definable"_err_en_US, s, sym->name());
+  }
+}
+
 void IoChecker::CheckForPureSubprogram() const { // C1597
   CHECK(context_.location());
   if (FindPureProcedureContaining(context_.FindScope(*context_.location()))) {
diff --git a/flang/lib/Semantics/check-io.h b/flang/lib/Semantics/check-io.h
index b5e8f12b5ee643..01bbcd9ba24ff7 100644
--- a/flang/lib/Semantics/check-io.h
+++ b/flang/lib/Semantics/check-io.h
@@ -122,6 +122,11 @@ class IoChecker : public virtual BaseChecker {
   void CheckForProhibitedSpecifier(IoSpecKind, bool, const std::string &) const;
   void CheckForProhibitedSpecifier(bool, const std::string &, IoSpecKind) const;
 
+  template <typename A>
+  void CheckForDefinableVariable(const A &var, const std::string &s) const;
+
+  void CheckForPureSubprogram() const;
+
   void Init(IoStmtKind s) {
     stmt_ = s;
     specifierSet_.reset();
@@ -130,8 +135,6 @@ class IoChecker : public virtual BaseChecker {
 
   void Done() { stmt_ = IoStmtKind::None; }
 
-  void CheckForPureSubprogram() const;
-
   SemanticsContext &context_;
   IoStmtKind stmt_{IoStmtKind::None};
   common::EnumSet<IoSpecKind, common::IoSpecKind_enumSize> specifierSet_;
diff --git a/flang/lib/Semantics/program-tree.cpp b/flang/lib/Semantics/program-tree.cpp
index e6dfd9b2f51cdd..9466a748567e1a 100644
--- a/flang/lib/Semantics/program-tree.cpp
+++ b/flang/lib/Semantics/program-tree.cpp
@@ -112,6 +112,10 @@ ProgramTree ProgramTree::Build(const parser::BlockData &x) {
   return result.set_stmt(stmt).set_endStmt(end);
 }
 
+ProgramTree ProgramTree::Build(const parser::CompilerDirective &) {
+  DIE("ProgramTree::Build() called for CompilerDirective");
+}
+
 const parser::ParentIdentifier &ProgramTree::GetParentId() const {
   const auto *stmt{
       std::get<const parser::Statement<parser::SubmoduleStmt> *>(stmt_)};
diff --git a/flang/lib/Semantics/program-tree.h b/flang/lib/Semantics/program-tree.h
index 69c133bfbde7bb..6b074522820173 100644
--- a/flang/lib/Semantics/program-tree.h
+++ b/flang/lib/Semantics/program-tree.h
@@ -38,6 +38,7 @@ class ProgramTree {
   static ProgramTree Build(const parser::Module &);
   static ProgramTree Build(const parser::Submodule &);
   static ProgramTree Build(const parser::BlockData &);
+  static ProgramTree Build(const parser::CompilerDirective &);
 
   ENUM_CLASS(Kind, // kind of node
       Program, Function, Subroutine, MpSubprogram, Module, Submodule, BlockData)
diff --git a/flang/lib/Semantics/resolve-names.cpp b/flang/lib/Semantics/resolve-names.cpp
index 60376a1b84690d..36c38c1cd99c4b 100644
--- a/flang/lib/Semantics/resolve-names.cpp
+++ b/flang/lib/Semantics/resolve-names.cpp
@@ -6226,6 +6226,11 @@ void ResolveNamesVisitor::Post(const parser::AssignedGotoStmt &x) {
 }
 
 bool ResolveNamesVisitor::Pre(const parser::ProgramUnit &x) {
+  if (std::holds_alternative<common::Indirection<parser::CompilerDirective>>(
+          x.u)) {
+    // TODO: global directives
+    return true;
+  }
   auto root{ProgramTree::Build(x)};
   SetScope(context().globalScope());
   ResolveSpecificationParts(root);
diff --git a/flang/test/Parser/compiler-directives.f90 b/flang/test/Parser/compiler-directives.f90
index 5545d0486e56f9..c916b16b532742 100644
--- a/flang/test/Parser/compiler-directives.f90
+++ b/flang/test/Parser/compiler-directives.f90
@@ -2,6 +2,7 @@
 
 ! Test that compiler directives can appear in various places.
 
+!dir$ integer
 module m
   !dir$ integer
   use iso_fortran_env
diff --git a/flang/test/Preprocessing/fixed-rescan.F b/flang/test/Preprocessing/fixed-rescan.F
new file mode 100644
index 00000000000000..3d6ba9a8c6f450
--- /dev/null
+++ b/flang/test/Preprocessing/fixed-rescan.F
@@ -0,0 +1,7 @@
+! RUN: %f18 -E %s | FileCheck %s
+! CHECK: callbar
+! Ensure that rescanned lines after macro replacement are not
+! misinterpreted as fixed-form comments when they start with C or D.
+#define foo bar
+      call foo
+      end
diff --git a/flang/test/Preprocessing/pp029.F b/flang/test/Preprocessing/pp029.F
index bb8efe6c1a2e03..9be309f75143e5 100644
--- a/flang/test/Preprocessing/pp029.F
+++ b/flang/test/Preprocessing/pp029.F
@@ -1,5 +1,5 @@
 ! RUN: %f18 -E %s 2>&1 | FileCheck %s
-! CHECK: if(77 7.eq.777)then
+! CHECK: if(777.eq.777)then
 * \ newline allowed in #define
       integer, parameter :: KWM = 666
 #define KWM 77\
diff --git a/flang/test/Preprocessing/pp130.F90 b/flang/test/Preprocessing/pp130.F90
index af4ad126e6fa48..be1148807b8bfd 100644
--- a/flang/test/Preprocessing/pp130.F90
+++ b/flang/test/Preprocessing/pp130.F90
@@ -1,5 +1,5 @@
-! RUN: %f18 -E %s 2>&1 | FileCheck %s
-! CHECK: j = j + &
+! RUN: (%f18 -E %s 2>&1 || true) | FileCheck %s
+! CHECK: error: bad character ('&') in Fortran token
 ! #define KWM &, use for continuation w/o pasting (ifort and nag seem to continue #define)
 #define KWM &
 
diff --git a/flang/test/Semantics/deallocate05.f90 b/flang/test/Semantics/deallocate05.f90
index 7524cc88fe0b8b..4a54469e5ab676 100644
--- a/flang/test/Semantics/deallocate05.f90
+++ b/flang/test/Semantics/deallocate05.f90
@@ -21,6 +21,7 @@ Program deallocatetest
 
 Real :: r
 Integer :: s
+Integer, Parameter :: const_s = 13
 Integer :: e
 Integer :: pi
 Character(256) :: ee
@@ -56,6 +57,8 @@ Program deallocatetest
 
 !ERROR: STAT may not be duplicated in a DEALLOCATE statement
 Deallocate(x, stat=s, stat=s)
+!ERROR: STAT variable 'const_s' must be definable
+Deallocate(x, stat=const_s)
 !ERROR: ERRMSG may not be duplicated in a DEALLOCATE statement
 Deallocate(x, errmsg=ee, errmsg=ee)
 !ERROR: STAT may not be duplicated in a DEALLOCATE statement
diff --git a/flang/test/Semantics/io01.f90 b/flang/test/Semantics/io01.f90
index 4238df89f5d0ee..9828d4afe8921f 100644
--- a/flang/test/Semantics/io01.f90
+++ b/flang/test/Semantics/io01.f90
@@ -21,6 +21,7 @@
   integer :: unit10 = 10
   integer :: unit11 = 11
   integer :: n = 40
+  integer, parameter :: const_new_unit = 66
 
   integer(kind=1) :: stat1
   integer(kind=2) :: stat2
@@ -73,6 +74,9 @@
   !ERROR: If NEWUNIT appears, FILE or STATUS must also appear
   open(newunit=n, newunit=nn, iostat=stat4)
 
+  !ERROR: NEWUNIT variable 'const_new_unit' must be definable
+  open(newunit=const_new_unit, status=cc)
+
   !ERROR: Duplicate UNIT specifier
   open(unit=100, unit=100)
 
diff --git a/flang/test/Semantics/io02.f90 b/flang/test/Semantics/io02.f90
index 5fd5fca4bc0c8d..9f5235d353cbd9 100644
--- a/flang/test/Semantics/io02.f90
+++ b/flang/test/Semantics/io02.f90
@@ -1,6 +1,7 @@
 ! RUN: %S/test_errors.sh %s %t %f18
   integer :: unit10 = 10
   integer :: unit11 = 11
+  integer, parameter :: const_stat = 6666
 
   integer(kind=1) :: stat1
   integer(kind=8) :: stat8
@@ -28,5 +29,8 @@
   !ERROR: Invalid STATUS value 'old'
   close(status='old', unit=17)
 
+  !ERROR: IOSTAT variable 'const_stat' must be definable
+  close(14, iostat=const_stat)
+
 9 continue
 end
diff --git a/flang/test/Semantics/io03.f90 b/flang/test/Semantics/io03.f90
index 0041e6cd0f5c44..5eb3420d1aea17 100644
--- a/flang/test/Semantics/io03.f90
+++ b/flang/test/Semantics/io03.f90
@@ -2,13 +2,18 @@
   character(kind=1,len=50) internal_file
   character(kind=2,len=50) internal_file2
   character(kind=4,len=50) internal_file4
+  character(kind=1,len=50) internal_fileA(20)
   character(kind=1,len=111) msg
   character(20) advance
+  character(20) :: cvar;
+  character, parameter :: const_internal_file = "(I6)"
+  character, parameter :: const_cvar = "Ceci n'est pas une pipe."
   integer*1 stat1
   integer*2 stat2, id2
   integer*8 stat8
   integer :: iunit = 10
-  integer, parameter :: junit = 11
+  integer, parameter :: junit = 11, const_size = 13, const_int = 15
+  integer :: vv(10) = 7
 
   namelist /mmm/ mm1, mm2
   namelist /nnn/ nn1, nn2
@@ -29,11 +34,14 @@
   read(fmt='(I4)', unit=*) jj
   read(iunit, *) jj
   read(junit, *) jj
-  read(10, *) jj
+  read(10, *) jj, cvar, cvar(7:17)
   read(internal_file, *) jj
+  read(internal_fileA(3), *) jj
+  read(internal_fileA(4:9), *) jj
   read(10, nnn)
   read(internal_file, nnn)
   read(internal_file, nml=nnn)
+  read(const_internal_file, *)
   read(fmt=*, unit=internal_file)
   read(nml=nnn, unit=internal_file)
   read(iunit, nnn)
@@ -53,6 +61,21 @@
   !ERROR: Invalid character kind for an internal file variable
   read(internal_file4, *) jj
 
+  !ERROR: Internal file must not have a vector subscript
+  read(internal_fileA(vv), *) jj
+
+  !ERROR: Input variable 'const_int' must be definable
+  read(11, *) const_int
+
+  !ERROR: SIZE variable 'const_size' must be definable
+  read(11, pos=ipos, size=const_size, end=9)
+
+  !ERROR: Input variable 'const_cvar' must be definable
+  read(11, *) const_cvar
+
+  !ERROR: Input variable 'const_cvar' must be definable
+  read(11, *) const_cvar(3:13)
+
   !ERROR: Duplicate IOSTAT specifier
   read(11, pos=ipos, iostat=stat1, iostat=stat2)
 
@@ -136,3 +159,25 @@
 
 9 continue
 end
+
+subroutine s(aa, n)
+  integer :: aa(5,*)
+  integer, intent(in) :: n
+  integer :: bb(10), vv(10)
+  type tt
+    real :: x, y, z
+  end type tt
+  type(tt) :: qq(20)
+
+  vv = 1
+
+  read(*, *) aa(n,1)
+  read(*, *) aa(n:n+2,2)
+  read(*, *) qq(2:5)%y
+
+  !ERROR: Input variable 'n' must be definable
+  read(*, *) n
+
+  !ERROR: Whole assumed size array 'aa' may not be an input item
+  read(*, *) aa
+end
diff --git a/flang/test/Semantics/io04.f90 b/flang/test/Semantics/io04.f90
index 0a37d685d3ee50..6be26047fd5b28 100644
--- a/flang/test/Semantics/io04.f90
+++ b/flang/test/Semantics/io04.f90
@@ -2,6 +2,7 @@
   character(kind=1,len=50) internal_file
   character(kind=1,len=100) msg
   character(20) sign
+  character, parameter :: const_internal_file = "(I6)"
   integer*1 stat1, id1
   integer*2 stat2
   integer*4 stat4
@@ -9,6 +10,8 @@
   integer :: iunit = 10
   integer, parameter :: junit = 11
   integer, pointer :: a(:)
+  integer, parameter :: const_id = 66666
+  procedure(), pointer :: procptr
 
   namelist /nnn/ nn1, nn2
 
@@ -66,6 +69,9 @@
   !ERROR: If NML appears, a data list must not appear
   write(10, nnn, rec=40, fmt=1) 'Ok'
 
+  !ERROR: Internal file variable 'const_internal_file' must be definable
+  write(const_internal_file, fmt=*)
+
   !ERROR: If UNIT=* appears, POS must not appear
   write(*, pos=n, nml=nnn)
 
@@ -118,8 +124,14 @@
   !ERROR: ID kind (1) is smaller than default INTEGER kind (4)
   write(id=id1, unit=10, asynchronous='Yes') 'Ok'
 
+  !ERROR: ID variable 'const_id' must be definable
+  write(10, *, asynchronous='yes', id=const_id, iostat=stat2) 'Ok'
+
   write(*, '(X)')
 
+  !ERROR: Output item must not be a procedure pointer
+  print*, n1, procptr, n2
+
 1 format (A)
 9 continue
 end
diff --git a/flang/test/Semantics/io05.f90 b/flang/test/Semantics/io05.f90
index 1501fbf587f5ea..ed6b77f7d4ad97 100644
--- a/flang/test/Semantics/io05.f90
+++ b/flang/test/Semantics/io05.f90
@@ -1,10 +1,12 @@
 ! RUN: %S/test_errors.sh %s %t %f18
   character*20 c(25), cv
   character(kind=1,len=59) msg
+  character, parameter :: const_round = "c'est quoi?"
   logical*2 v(5), lv
   integer*1 stat1
   integer*2 stat4
   integer*8 stat8, iv
+  integer, parameter :: const_id = 1
 
   inquire(10)
   inquire(file='abc')
@@ -22,6 +24,7 @@
       exist=v(1), named=v(2), opened=v(3), pending=v(4))
   inquire(pending=v(5), file='abc')
   inquire(10, id=id, pending=v(5))
+  inquire(10, id=const_id, pending=v(5))
 
   ! using variable 'cv' multiple times seems to be allowed
   inquire(file='abc', &
@@ -56,5 +59,8 @@
   !ERROR: If ID appears, PENDING must also appear
   inquire(file='abc', id=id)
 
+  !ERROR: ROUND variable 'const_round' must be definable
+  inquire(file='abc', round=const_round)
+
 9 continue
 end
diff --git a/flang/test/Semantics/io06.f90 b/flang/test/Semantics/io06.f90
index 157d831dc33315..fe3b97f0e67e4a 100644
--- a/flang/test/Semantics/io06.f90
+++ b/flang/test/Semantics/io06.f90
@@ -1,6 +1,7 @@
 ! RUN: %S/test_errors.sh %s %t %f18
   character(kind=1,len=100) msg1
   character(kind=2,len=200) msg2
+  character, parameter :: const_msg = 'doof'
   integer(1) stat1
   integer(2) stat2
   integer(8) stat8
@@ -28,6 +29,9 @@
   !ERROR: Duplicate IOSTAT specifier
   endfile(iostat=stat2, err=9, unit=10, iostat=stat8, iomsg=msg1)
 
+  !ERROR: IOMSG variable 'const_msg' must be definable
+  flush(iomsg=const_msg, unit=10, iostat=stat8, err=9)
+
   !ERROR: REWIND statement must have a UNIT number specifier
   rewind(iostat=stat2)
 
diff --git a/lldb/test/API/macosx/queues/TestQueues.py b/lldb/test/API/macosx/queues/TestQueues.py
index e177daa54fa347..711c99a7d400d6 100644
--- a/lldb/test/API/macosx/queues/TestQueues.py
+++ b/lldb/test/API/macosx/queues/TestQueues.py
@@ -192,7 +192,6 @@ def queues(self):
         user_initiated_thread = lldb.SBThread()
         user_interactive_thread = lldb.SBThread()
         utility_thread = lldb.SBThread()
-        unspecified_thread = lldb.SBThread()
         background_thread = lldb.SBThread()
         for th in process.threads:
             if th.GetName() == "user initiated QoS":
@@ -201,8 +200,6 @@ def queues(self):
                 user_interactive_thread = th
             if th.GetName() == "utility QoS":
                 utility_thread = th
-            if th.GetName() == "unspecified QoS":
-                unspecified_thread = th
             if th.GetName() == "background QoS":
                 background_thread = th
 
@@ -213,9 +210,6 @@ def queues(self):
             user_interactive_thread.IsValid(),
             "Found user interactive QoS thread")
         self.assertTrue(utility_thread.IsValid(), "Found utility QoS thread")
-        self.assertTrue(
-            unspecified_thread.IsValid(),
-            "Found unspecified QoS thread")
         self.assertTrue(
             background_thread.IsValid(),
             "Found background QoS thread")
@@ -248,16 +242,6 @@ def queues(self):
             stream.GetData(), "Utility",
             "utility QoS thread name is valid")
         stream.Clear()
-        self.assertTrue(
-            unspecified_thread.GetInfoItemByPathAsString(
-                "requested_qos.printable_name",
-                stream),
-            "Get QoS printable string for unspecified QoS thread")
-        qosName = stream.GetData()
-        self.assertTrue(
-            qosName == "User Initiated" or qosName == "Default",
-            "unspecified QoS thread name is valid: " + str(qosName))
-        stream.Clear()
         self.assertTrue(
             background_thread.GetInfoItemByPathAsString(
                 "requested_qos.printable_name",
diff --git a/lldb/test/API/macosx/queues/main.c b/lldb/test/API/macosx/queues/main.c
index 3978b92bff1a62..2bf390b1330a63 100644
--- a/lldb/test/API/macosx/queues/main.c
+++ b/lldb/test/API/macosx/queues/main.c
@@ -136,15 +136,9 @@ int main (int argc, const char **argv)
             while (1)
                 sleep (10);
       });
-    dispatch_async (dispatch_get_global_queue(QOS_CLASS_UNSPECIFIED, 0), ^{
-            pthread_setname_np ("unspecified QoS");
-            atomic_fetch_add(&thread_count, 1);
-            while (1)
-                sleep (10);
-      });
 
     // Unfortunately there is no pthread_barrier on darwin.
-    while ((atomic_load(&thread_count) < 13) || (finished_enqueueing_work == 0))
+    while ((atomic_load(&thread_count) < 12) || (finished_enqueueing_work == 0))
         sleep (1);
 
     stopper ();
diff --git a/lldb/unittests/Symbol/LocateSymbolFileTest.cpp b/lldb/unittests/Symbol/LocateSymbolFileTest.cpp
index 268faeaf1dbbda..a2f9be56635d1f 100644
--- a/lldb/unittests/Symbol/LocateSymbolFileTest.cpp
+++ b/lldb/unittests/Symbol/LocateSymbolFileTest.cpp
@@ -14,6 +14,7 @@
 #include "lldb/Host/HostInfo.h"
 #include "lldb/Symbol/LocateSymbolFile.h"
 #include "lldb/Target/Target.h"
+#include "lldb/Utility/Reproducer.h"
 
 using namespace lldb_private;
 
@@ -27,15 +28,22 @@ class SymbolsTest : public ::testing::Test {
 TEST_F(
     SymbolsTest,
     TerminateLocateExecutableSymbolFileForUnknownExecutableAndUnknownSymbolFile) {
+  EXPECT_THAT_ERROR(
+      repro::Reproducer::Initialize(repro::ReproducerMode::Off, llvm::None),
+      llvm::Succeeded());
   ModuleSpec module_spec;
   FileSpecList search_paths = Target::GetDefaultDebugFileSearchPaths();
   FileSpec symbol_file_spec =
       Symbols::LocateExecutableSymbolFile(module_spec, search_paths);
   EXPECT_TRUE(symbol_file_spec.GetFilename().IsEmpty());
+  repro::Reproducer::Terminate();
 }
 
 TEST_F(SymbolsTest,
        LocateExecutableSymbolFileForUnknownExecutableAndMissingSymbolFile) {
+  EXPECT_THAT_ERROR(
+      repro::Reproducer::Initialize(repro::ReproducerMode::Off, llvm::None),
+      llvm::Succeeded());
   ModuleSpec module_spec;
   // using a GUID here because the symbol file shouldn't actually exist on disk
   module_spec.GetSymbolFileSpec().SetFile(
@@ -44,4 +52,5 @@ TEST_F(SymbolsTest,
   FileSpec symbol_file_spec =
       Symbols::LocateExecutableSymbolFile(module_spec, search_paths);
   EXPECT_TRUE(symbol_file_spec.GetFilename().IsEmpty());
+  repro::Reproducer::Terminate();
 }
diff --git a/llvm/docs/ReleaseNotes.rst b/llvm/docs/ReleaseNotes.rst
index 2a2775f13dd1fa..c6b2a3ac9bb57a 100644
--- a/llvm/docs/ReleaseNotes.rst
+++ b/llvm/docs/ReleaseNotes.rst
@@ -97,6 +97,7 @@ During this release ...
 * The 'mpx' feature was removed from the backend. It had been removed from clang
   frontend in 10.0. Mention of the 'mpx' feature in an IR file will print a
   message to stderr, but IR should still compile.
+* Support for -march=sapphirerapids was added.
 
 Changes to the AMDGPU Target
 -----------------------------
diff --git a/llvm/examples/OrcV2Examples/LLJITWithThinLTOSummaries/CMakeLists.txt b/llvm/examples/OrcV2Examples/LLJITWithThinLTOSummaries/CMakeLists.txt
index 22262e4eb5b972..175e655007050a 100644
--- a/llvm/examples/OrcV2Examples/LLJITWithThinLTOSummaries/CMakeLists.txt
+++ b/llvm/examples/OrcV2Examples/LLJITWithThinLTOSummaries/CMakeLists.txt
@@ -1,7 +1,7 @@
 set(LLVM_LINK_COMPONENTS
+  BitReader
   Core
   ExecutionEngine
-  IRReader
   OrcJIT
   Support
   nativecodegen
diff --git a/llvm/include/llvm/Analysis/ValueTracking.h b/llvm/include/llvm/Analysis/ValueTracking.h
index f1b9cc906049be..0c82311d857c13 100644
--- a/llvm/include/llvm/Analysis/ValueTracking.h
+++ b/llvm/include/llvm/Analysis/ValueTracking.h
@@ -584,10 +584,10 @@ constexpr unsigned MaxAnalysisRecursionDepth = 6;
   /// getGuaranteedNonPoisonOp.
   bool propagatesPoison(const Instruction *I);
 
-  /// Return either nullptr or an operand of I such that I will trigger
-  /// undefined behavior if I is executed and that operand has a poison
-  /// value.
-  const Value *getGuaranteedNonPoisonOp(const Instruction *I);
+  /// Insert operands of I into Ops such that I will trigger undefined behavior
+  /// if I is executed and that operand has a poison value.
+  void getGuaranteedNonPoisonOps(const Instruction *I,
+                                 SmallPtrSetImpl<const Value *> &Ops);
 
   /// Return true if the given instruction must trigger undefined behavior.
   /// when I is executed with any operands which appear in KnownPoison holding
diff --git a/llvm/include/llvm/ExecutionEngine/Orc/TPCIndirectionUtils.h b/llvm/include/llvm/ExecutionEngine/Orc/TPCIndirectionUtils.h
index db9cd1b98cf9cb..90097f1131f317 100644
--- a/llvm/include/llvm/ExecutionEngine/Orc/TPCIndirectionUtils.h
+++ b/llvm/include/llvm/ExecutionEngine/Orc/TPCIndirectionUtils.h
@@ -206,4 +206,4 @@ TPCIndirectionUtils::CreateWithABI(TargetProcessControl &TPC) {
 } // end namespace orc
 } // end namespace llvm
 
-#endif // LLVM_EXECUTIONENGINE_ORC_T_H
+#endif // LLVM_EXECUTIONENGINE_ORC_TPCINDIRECTIONUTILS_H
diff --git a/llvm/lib/Analysis/MemoryDependenceAnalysis.cpp b/llvm/lib/Analysis/MemoryDependenceAnalysis.cpp
index 7f3de0fcf140af..2428d57d2809fa 100644
--- a/llvm/lib/Analysis/MemoryDependenceAnalysis.cpp
+++ b/llvm/lib/Analysis/MemoryDependenceAnalysis.cpp
@@ -362,6 +362,8 @@ MemoryDependenceResults::getInvariantGroupPointerDependency(LoadInst *LI,
 MemDepResult MemoryDependenceResults::getSimplePointerDependencyFrom(
     const MemoryLocation &MemLoc, bool isLoad, BasicBlock::iterator ScanIt,
     BasicBlock *BB, Instruction *QueryInst, unsigned *Limit) {
+  // We can batch AA queries, because IR does not change during a MemDep query.
+  BatchAAResults BatchAA(AA);
   bool isInvariantLoad = false;
 
   unsigned DefaultLimit = getDefaultBlockScanLimit();
@@ -445,7 +447,7 @@ MemDepResult MemoryDependenceResults::getSimplePointerDependencyFrom(
         // pointer, not on query pointers that are indexed off of them.  It'd
         // be nice to handle that at some point (the right approach is to use
         // GetPointerBaseWithConstantOffset).
-        if (AA.isMustAlias(MemoryLocation(II->getArgOperand(1)), MemLoc))
+        if (BatchAA.isMustAlias(MemoryLocation(II->getArgOperand(1)), MemLoc))
           return MemDepResult::getDef(II);
         continue;
       }
@@ -485,7 +487,7 @@ MemDepResult MemoryDependenceResults::getSimplePointerDependencyFrom(
       MemoryLocation LoadLoc = MemoryLocation::get(LI);
 
       // If we found a pointer, check if it could be the same as our pointer.
-      AliasResult R = AA.alias(LoadLoc, MemLoc);
+      AliasResult R = BatchAA.alias(LoadLoc, MemLoc);
 
       if (isLoad) {
         if (R == NoAlias)
@@ -516,7 +518,7 @@ MemDepResult MemoryDependenceResults::getSimplePointerDependencyFrom(
         continue;
 
       // Stores don't alias loads from read-only memory.
-      if (AA.pointsToConstantMemory(LoadLoc))
+      if (BatchAA.pointsToConstantMemory(LoadLoc))
         continue;
 
       // Stores depend on may/must aliased loads.
@@ -547,7 +549,7 @@ MemDepResult MemoryDependenceResults::getSimplePointerDependencyFrom(
       // If alias analysis can tell that this store is guaranteed to not modify
       // the query pointer, ignore it.  Use getModRefInfo to handle cases where
       // the query pointer points to constant memory etc.
-      if (!isModOrRefSet(AA.getModRefInfo(SI, MemLoc)))
+      if (!isModOrRefSet(BatchAA.getModRefInfo(SI, MemLoc)))
         continue;
 
       // Ok, this store might clobber the query pointer.  Check to see if it is
@@ -556,7 +558,7 @@ MemDepResult MemoryDependenceResults::getSimplePointerDependencyFrom(
       MemoryLocation StoreLoc = MemoryLocation::get(SI);
 
       // If we found a pointer, check if it could be the same as our pointer.
-      AliasResult R = AA.alias(StoreLoc, MemLoc);
+      AliasResult R = BatchAA.alias(StoreLoc, MemLoc);
 
       if (R == NoAlias)
         continue;
@@ -575,7 +577,7 @@ MemDepResult MemoryDependenceResults::getSimplePointerDependencyFrom(
     // handled by BasicAA.
     if (isa<AllocaInst>(Inst) || isNoAliasFn(Inst, &TLI)) {
       const Value *AccessPtr = getUnderlyingObject(MemLoc.Ptr);
-      if (AccessPtr == Inst || AA.isMustAlias(Inst, AccessPtr))
+      if (AccessPtr == Inst || BatchAA.isMustAlias(Inst, AccessPtr))
         return MemDepResult::getDef(Inst);
     }
 
@@ -592,9 +594,10 @@ MemDepResult MemoryDependenceResults::getSimplePointerDependencyFrom(
         continue;
 
     // See if this instruction (e.g. a call or vaarg) mod/ref's the pointer.
-    ModRefInfo MR = AA.getModRefInfo(Inst, MemLoc);
+    ModRefInfo MR = BatchAA.getModRefInfo(Inst, MemLoc);
     // If necessary, perform additional analysis.
     if (isModAndRefSet(MR))
+      // TODO: Support callCapturesBefore() on BatchAAResults.
       MR = AA.callCapturesBefore(Inst, MemLoc, &DT);
     switch (clearMust(MR)) {
     case ModRefInfo::NoModRef:
diff --git a/llvm/lib/Analysis/ValueTracking.cpp b/llvm/lib/Analysis/ValueTracking.cpp
index 07c4e57228f114..2835d2f06edd3f 100644
--- a/llvm/lib/Analysis/ValueTracking.cpp
+++ b/llvm/lib/Analysis/ValueTracking.cpp
@@ -5065,46 +5065,69 @@ bool llvm::propagatesPoison(const Instruction *I) {
   }
 }
 
-const Value *llvm::getGuaranteedNonPoisonOp(const Instruction *I) {
+void llvm::getGuaranteedNonPoisonOps(const Instruction *I,
+                                     SmallPtrSetImpl<const Value *> &Operands) {
   switch (I->getOpcode()) {
     case Instruction::Store:
-      return cast<StoreInst>(I)->getPointerOperand();
+      Operands.insert(cast<StoreInst>(I)->getPointerOperand());
+      break;
 
     case Instruction::Load:
-      return cast<LoadInst>(I)->getPointerOperand();
+      Operands.insert(cast<LoadInst>(I)->getPointerOperand());
+      break;
 
     case Instruction::AtomicCmpXchg:
-      return cast<AtomicCmpXchgInst>(I)->getPointerOperand();
+      Operands.insert(cast<AtomicCmpXchgInst>(I)->getPointerOperand());
+      break;
 
     case Instruction::AtomicRMW:
-      return cast<AtomicRMWInst>(I)->getPointerOperand();
+      Operands.insert(cast<AtomicRMWInst>(I)->getPointerOperand());
+      break;
 
     case Instruction::UDiv:
     case Instruction::SDiv:
     case Instruction::URem:
     case Instruction::SRem:
-      return I->getOperand(1);
+      Operands.insert(I->getOperand(1));
+      break;
 
     case Instruction::Call:
+    case Instruction::Invoke: {
       if (auto *II = dyn_cast<IntrinsicInst>(I)) {
         switch (II->getIntrinsicID()) {
         case Intrinsic::assume:
-          return II->getArgOperand(0);
+          Operands.insert(II->getArgOperand(0));
+          break;
         default:
-          return nullptr;
+          break;
         }
       }
-      return nullptr;
+
+      const CallBase *CB = cast<CallBase>(I);
+      if (CB->isIndirectCall())
+        Operands.insert(CB->getCalledOperand());
+      for (unsigned i = 0; i < CB->arg_size(); ++i) {
+        if (CB->paramHasAttr(i, Attribute::NoUndef))
+          Operands.insert(CB->getArgOperand(i));
+      }
+      break;
+    }
 
     default:
-      return nullptr;
+      break;
   }
 }
 
 bool llvm::mustTriggerUB(const Instruction *I,
                          const SmallSet<const Value *, 16>& KnownPoison) {
-  auto *NotPoison = getGuaranteedNonPoisonOp(I);
-  return (NotPoison && KnownPoison.count(NotPoison));
+  SmallPtrSet<const Value *, 4> NonPoisonOps;
+  getGuaranteedNonPoisonOps(I, NonPoisonOps);
+
+  for (const auto *V : NonPoisonOps)
+    if (KnownPoison.count(V))
+      return true;
+
+  return false;
 }
 
 
diff --git a/llvm/lib/LTO/LTO.cpp b/llvm/lib/LTO/LTO.cpp
index 6e1e3998e490e3..6230216aa44666 100644
--- a/llvm/lib/LTO/LTO.cpp
+++ b/llvm/lib/LTO/LTO.cpp
@@ -798,7 +798,7 @@ Error LTO::linkRegularLTO(RegularLTOState::AddedModule Mod,
   for (GlobalValue *GV : Mod.Keep) {
     if (LivenessFromIndex && !ThinLTO.CombinedIndex.isGUIDLive(GV->getGUID())) {
       if (Function *F = dyn_cast<Function>(GV)) {
-        OptimizationRemarkEmitter ORE(F);
+        OptimizationRemarkEmitter ORE(F, nullptr);
         ORE.emit(OptimizationRemark(DEBUG_TYPE, "deadfunction", F)
                  << ore::NV("Function", F)
                  << " not added to the combined module ");
diff --git a/llvm/lib/Transforms/Instrumentation/PoisonChecking.cpp b/llvm/lib/Transforms/Instrumentation/PoisonChecking.cpp
index fa97a194ea2b52..6f785687b50452 100644
--- a/llvm/lib/Transforms/Instrumentation/PoisonChecking.cpp
+++ b/llvm/lib/Transforms/Instrumentation/PoisonChecking.cpp
@@ -282,8 +282,10 @@ static bool rewrite(Function &F) {
 
       // Note: There are many more sources of documented UB, but this pass only
       // attempts to find UB triggered by propagation of poison.
-      if (Value *Op = const_cast<Value*>(getGuaranteedNonPoisonOp(&I)))
-        CreateAssertNot(B, getPoisonFor(ValToPoison, Op));
+      SmallPtrSet<const Value *, 4> NonPoisonOps;
+      getGuaranteedNonPoisonOps(&I, NonPoisonOps);
+      for (const Value *Op : NonPoisonOps)
+        CreateAssertNot(B, getPoisonFor(ValToPoison, const_cast<Value *>(Op)));
 
       if (LocalCheck)
         if (auto *RI = dyn_cast<ReturnInst>(&I))
diff --git a/llvm/test/CodeGen/AArch64/stack-guard-reassign.ll b/llvm/test/CodeGen/AArch64/stack-guard-reassign.ll
index 6b00eda6233a28..71d3d27a1b6490 100644
--- a/llvm/test/CodeGen/AArch64/stack-guard-reassign.ll
+++ b/llvm/test/CodeGen/AArch64/stack-guard-reassign.ll
@@ -3,6 +3,6 @@
 ; Verify that the offset assigned to the stack protector is at the top of the
 ; frame, covering the locals.
 ; CHECK-LABEL: fn:
-; CHECK:      adrp x8, __stack_chk_guard
-; CHECK-NEXT: ldr x8, [x8, :lo12:__stack_chk_guard]
-; CHECK-NEXT: stur x8, [x29, #-8]
+; CHECK:      adrp [[REG:x[0-9]+]], __stack_chk_guard
+; CHECK-NEXT: ldr [[REG]], {{\[}}[[REG]], :lo12:__stack_chk_guard]
+; CHECK-NEXT: stur [[REG]], [x29, #-8]
diff --git a/llvm/test/LTO/Resolution/X86/dead-strip-fulllto.ll b/llvm/test/LTO/Resolution/X86/dead-strip-fulllto.ll
index 43659313f2dc5c..fb4fde7661b7d4 100644
--- a/llvm/test/LTO/Resolution/X86/dead-strip-fulllto.ll
+++ b/llvm/test/LTO/Resolution/X86/dead-strip-fulllto.ll
@@ -1,7 +1,8 @@
 ; RUN: opt -module-summary -o %t %s
 ; RUN: opt -module-summary -o %t2 %S/Inputs/dead-strip-fulllto.ll
 
-; RUN: llvm-lto2 run --pass-remarks-output=%t4.yaml --pass-remarks-filter=. \
+; Adding '--pass-remarks-with-hotness' should not cause crash.
+; RUN: llvm-lto2 run --pass-remarks-output=%t4.yaml --pass-remarks-filter=. --pass-remarks-with-hotness \
 ; RUN:               %t -r %t,main,px -r %t,live1, -r %t,live2,p -r %t,dead2,p \
 ; RUN:               %t2 -r %t2,live1,p -r %t2,live2, -r %t2,dead1,p -r %t2,dead2, -r %t2,odr, \
 ; RUN:               -save-temps -o %t3
diff --git a/llvm/test/Transforms/InstSimplify/freeze-noundef.ll b/llvm/test/Transforms/InstSimplify/freeze-noundef.ll
index f2e897b507e3d2..6a52bd5b257f0e 100644
--- a/llvm/test/Transforms/InstSimplify/freeze-noundef.ll
+++ b/llvm/test/Transforms/InstSimplify/freeze-noundef.ll
@@ -69,7 +69,6 @@ define {i8, i32} @aggr({i8, i32} noundef %x) {
   ret {i8, i32} %y
 }
 
-; TODO: should look into extract operations
 define i32 @extract({i8, i32} noundef %x) {
 ; CHECK-LABEL: @extract(
 ; CHECK-NEXT:    [[Y:%.*]] = extractvalue { i8, i32 } [[X:%.*]], 1
@@ -91,3 +90,17 @@ define i32 @extract2({i8, {i8, i32}} noundef %x) {
   %w = freeze i32 %z
   ret i32 %w
 }
+
+declare void @use_i1(i1 noundef)
+
+define i1 @used_by_fncall(i1 %x) {
+; CHECK-LABEL: @used_by_fncall(
+; CHECK-NEXT:    [[Y:%.*]] = add nsw i1 [[X:%.*]], true
+; CHECK-NEXT:    call void @use_i1(i1 [[Y]])
+; CHECK-NEXT:    ret i1 [[Y]]
+;
+  %y = add nsw i1 %x, 1
+  call void @use_i1(i1 %y)
+  %f = freeze i1 %y
+  ret i1 %f
+}
diff --git a/llvm/test/Transforms/LoopVectorize/PowerPC/massv-altivec.ll b/llvm/test/Transforms/LoopVectorize/PowerPC/massv-altivec.ll
index c63db5b0b66363..df2aa99f2d4b4d 100644
--- a/llvm/test/Transforms/LoopVectorize/PowerPC/massv-altivec.ll
+++ b/llvm/test/Transforms/LoopVectorize/PowerPC/massv-altivec.ll
@@ -1,4 +1,4 @@
-; RUN: opt -vector-library=MASSV -loop-vectorize -force-vector-interleave=1 -mattr=-altivec -S < %s | FileCheck %s
+; RUN: opt -vector-library=MASSV -inject-tli-mappings -loop-vectorize -force-vector-interleave=1 -mattr=-altivec -S < %s | FileCheck %s
 
 target datalayout = "e-m:e-i64:64-n32:64" 
 target triple = "powerpc64le-unknown-linux-gnu"
diff --git a/llvm/test/Transforms/LoopVectorize/PowerPC/massv-calls.ll b/llvm/test/Transforms/LoopVectorize/PowerPC/massv-calls.ll
index a08c23b5b2dd8b..eafca05c5aa778 100644
--- a/llvm/test/Transforms/LoopVectorize/PowerPC/massv-calls.ll
+++ b/llvm/test/Transforms/LoopVectorize/PowerPC/massv-calls.ll
@@ -1,4 +1,4 @@
-; RUN: opt -vector-library=MASSV -loop-vectorize -force-vector-interleave=1 -S < %s | FileCheck %s
+; RUN: opt -vector-library=MASSV -inject-tli-mappings -loop-vectorize -force-vector-interleave=1 -S < %s | FileCheck %s
 
 target datalayout = "e-m:e-i64:64-n32:64" 
 target triple = "powerpc64le-unknown-linux-gnu"
diff --git a/llvm/test/Transforms/LoopVectorize/PowerPC/massv-nobuiltin.ll b/llvm/test/Transforms/LoopVectorize/PowerPC/massv-nobuiltin.ll
index e7503b615f44a8..872326934ab405 100644
--- a/llvm/test/Transforms/LoopVectorize/PowerPC/massv-nobuiltin.ll
+++ b/llvm/test/Transforms/LoopVectorize/PowerPC/massv-nobuiltin.ll
@@ -1,4 +1,4 @@
-; RUN: opt -vector-library=MASSV -loop-vectorize -force-vector-interleave=1 -S < %s | FileCheck %s
+; RUN: opt -vector-library=MASSV -inject-tli-mappings -loop-vectorize -force-vector-interleave=1 -S < %s | FileCheck %s
 
 target datalayout = "e-m:e-i64:64-n32:64" 
 target triple = "powerpc64le-unknown-linux-gnu"
diff --git a/llvm/test/Transforms/LoopVectorize/PowerPC/massv-unsupported.ll b/llvm/test/Transforms/LoopVectorize/PowerPC/massv-unsupported.ll
index f9b4b72a027e83..575f9b2700ad5b 100644
--- a/llvm/test/Transforms/LoopVectorize/PowerPC/massv-unsupported.ll
+++ b/llvm/test/Transforms/LoopVectorize/PowerPC/massv-unsupported.ll
@@ -1,4 +1,4 @@
-; RUN: opt -vector-library=MASSV -loop-vectorize -force-vector-interleave=1 -S < %s | FileCheck %s
+; RUN: opt -vector-library=MASSV -inject-tli-mappings -loop-vectorize -force-vector-interleave=1 -S < %s | FileCheck %s
 
 target datalayout = "e-m:e-i64:64-n32:64" 
 target triple = "powerpc64le-unknown-linux-gnu"
diff --git a/llvm/test/Transforms/LoopVectorize/PowerPC/widened-massv-call.ll b/llvm/test/Transforms/LoopVectorize/PowerPC/widened-massv-call.ll
index 18098520908046..f5626598f29c34 100644
--- a/llvm/test/Transforms/LoopVectorize/PowerPC/widened-massv-call.ll
+++ b/llvm/test/Transforms/LoopVectorize/PowerPC/widened-massv-call.ll
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 ; RUN: opt < %s -vector-library=MASSV -force-vector-interleave=1 \
-; RUN:   -vectorizer-maximize-bandwidth -O2 -loop-vectorize \
+; RUN:   -vectorizer-maximize-bandwidth -O2 -inject-tli-mappings -loop-vectorize \
 ; RUN:   -mtriple=powerpc64le-unknown-linux -S -mcpu=pwr9 2>&1 | FileCheck %s
 
 define dso_local double @test(float* %Arr) {
diff --git a/llvm/test/Transforms/LoopVectorize/X86/svml-calls-finite.ll b/llvm/test/Transforms/LoopVectorize/X86/svml-calls-finite.ll
index 1e55e7de91bb4c..d6e3469c7bdb79 100644
--- a/llvm/test/Transforms/LoopVectorize/X86/svml-calls-finite.ll
+++ b/llvm/test/Transforms/LoopVectorize/X86/svml-calls-finite.ll
@@ -1,4 +1,4 @@
-; RUN: opt -vector-library=SVML -loop-vectorize -S < %s | FileCheck %s
+; RUN: opt -vector-library=SVML -inject-tli-mappings -loop-vectorize -S < %s | FileCheck %s
 
 ; Test to verify that when math headers are built with
 ; __FINITE_MATH_ONLY__ enabled, causing use of __<func>_finite
diff --git a/llvm/test/Transforms/LoopVectorize/X86/svml-calls.ll b/llvm/test/Transforms/LoopVectorize/X86/svml-calls.ll
index 1f2b71f6ba21a5..aa8a25c3b87f57 100644
--- a/llvm/test/Transforms/LoopVectorize/X86/svml-calls.ll
+++ b/llvm/test/Transforms/LoopVectorize/X86/svml-calls.ll
@@ -1,4 +1,4 @@
-; RUN: opt -vector-library=SVML -loop-vectorize -force-vector-width=4 -force-vector-interleave=1 -mattr=avx -S < %s | FileCheck %s
+; RUN: opt -vector-library=SVML -inject-tli-mappings -loop-vectorize -force-vector-width=4 -force-vector-interleave=1 -mattr=avx -S < %s | FileCheck %s
 
 target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
 target triple = "x86_64-unknown-linux-gnu"
diff --git a/llvm/test/Transforms/LoopVectorize/X86/veclib-calls.ll b/llvm/test/Transforms/LoopVectorize/X86/veclib-calls.ll
index 6f8f5223ce4588..8cbcf9351d83c2 100644
--- a/llvm/test/Transforms/LoopVectorize/X86/veclib-calls.ll
+++ b/llvm/test/Transforms/LoopVectorize/X86/veclib-calls.ll
@@ -1,4 +1,4 @@
-; RUN: opt < %s -vector-library=Accelerate -loop-vectorize -S | FileCheck %s
+; RUN: opt < %s -vector-library=Accelerate -inject-tli-mappings -loop-vectorize -S | FileCheck %s
 
 target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
 target triple = "x86_64-unknown-linux-gnu"
diff --git a/llvm/tools/llvm-mca/Views/BottleneckAnalysis.cpp b/llvm/tools/llvm-mca/Views/BottleneckAnalysis.cpp
index 99deed6eae976d..519b928fda5dc0 100644
--- a/llvm/tools/llvm-mca/Views/BottleneckAnalysis.cpp
+++ b/llvm/tools/llvm-mca/Views/BottleneckAnalysis.cpp
@@ -16,7 +16,6 @@
 #include "llvm/MC/MCInst.h"
 #include "llvm/MCA/Support.h"
 #include "llvm/Support/Format.h"
-#include "llvm/Support/FormattedStream.h"
 
 namespace llvm {
 namespace mca {
@@ -284,21 +283,14 @@ void DependencyGraph::getCriticalSequence(
   }
 }
 
-static void printInstruction(formatted_raw_ostream &FOS,
-                             const MCSubtargetInfo &STI, MCInstPrinter &MCIP,
-                             const MCInst &MCI,
-                             bool UseDifferentColor = false) {
-  std::string Instruction;
-  raw_string_ostream InstrStream(Instruction);
-
+void BottleneckAnalysis::printInstruction(formatted_raw_ostream &FOS,
+                                          const MCInst &MCI,
+                                          bool UseDifferentColor) const {
   FOS.PadToColumn(14);
 
-  MCIP.printInst(&MCI, 0, "", STI, InstrStream);
-  InstrStream.flush();
-
   if (UseDifferentColor)
     FOS.changeColor(raw_ostream::CYAN, true, false);
-  FOS << StringRef(Instruction).ltrim();
+  FOS << printInstructionString(MCI);
   if (UseDifferentColor)
     FOS.resetColor();
 }
@@ -316,6 +308,7 @@ void BottleneckAnalysis::printCriticalSequence(raw_ostream &OS) const {
   OS << "\nCritical sequence based on the simulation:\n\n";
 
   const DependencyEdge &FirstEdge = *Seq[0];
+  ArrayRef<llvm::MCInst> Source = getSource();
   unsigned FromIID = FirstEdge.FromIID % Source.size();
   unsigned ToIID = FirstEdge.ToIID % Source.size();
   bool IsLoopCarried = FromIID >= ToIID;
@@ -331,17 +324,17 @@ void BottleneckAnalysis::printCriticalSequence(raw_ostream &OS) const {
   unsigned CurrentIID = 0;
   if (IsLoopCarried) {
     FOS << "\n +----< " << FromIID << ".";
-    printInstruction(FOS, STI, MCIP, Source[FromIID], HasColors);
+    printInstruction(FOS, Source[FromIID], HasColors);
     FOS << "\n |\n |    < loop carried > \n |";
   } else {
     while (CurrentIID < FromIID) {
       FOS << "\n        " << CurrentIID << ".";
-      printInstruction(FOS, STI, MCIP, Source[CurrentIID]);
+      printInstruction(FOS, Source[CurrentIID]);
       CurrentIID++;
     }
 
     FOS << "\n +----< " << CurrentIID << ".";
-    printInstruction(FOS, STI, MCIP, Source[CurrentIID], HasColors);
+    printInstruction(FOS, Source[CurrentIID], HasColors);
     CurrentIID++;
   }
 
@@ -351,17 +344,17 @@ void BottleneckAnalysis::printCriticalSequence(raw_ostream &OS) const {
 
     while (CurrentIID < LastIID) {
       FOS << "\n |      " << CurrentIID << ".";
-      printInstruction(FOS, STI, MCIP, Source[CurrentIID]);
+      printInstruction(FOS, Source[CurrentIID]);
       CurrentIID++;
     }
 
     if (CurrentIID == ToIID) {
       FOS << "\n +----> " << ToIID << ".";
-      printInstruction(FOS, STI, MCIP, Source[CurrentIID], HasColors);
+      printInstruction(FOS, Source[CurrentIID], HasColors);
     } else {
       FOS << "\n |\n |    < loop carried > \n |"
           << "\n +----> " << ToIID << ".";
-      printInstruction(FOS, STI, MCIP, Source[ToIID], HasColors);
+      printInstruction(FOS, Source[ToIID], HasColors);
     }
     FOS.PadToColumn(58);
 
@@ -373,7 +366,7 @@ void BottleneckAnalysis::printCriticalSequence(raw_ostream &OS) const {
       FOS << "## REGISTER dependency:  ";
       if (HasColors)
         FOS.changeColor(raw_ostream::MAGENTA, true, false);
-      MCIP.printRegName(FOS, Dep.ResourceOrRegID);
+      getInstPrinter().printRegName(FOS, Dep.ResourceOrRegID);
     } else if (Dep.Type == DependencyEdge::DT_MEMORY) {
       FOS << "## MEMORY dependency.";
     } else {
@@ -397,7 +390,7 @@ void BottleneckAnalysis::printCriticalSequence(raw_ostream &OS) const {
 
   while (CurrentIID < Source.size()) {
     FOS << "\n        " << CurrentIID << ".";
-    printInstruction(FOS, STI, MCIP, Source[CurrentIID]);
+    printInstruction(FOS, Source[CurrentIID]);
     CurrentIID++;
   }
 
@@ -451,8 +444,8 @@ void DependencyGraph::addDependency(unsigned From, unsigned To,
 BottleneckAnalysis::BottleneckAnalysis(const MCSubtargetInfo &sti,
                                        MCInstPrinter &Printer,
                                        ArrayRef<MCInst> S, unsigned NumIter)
-    : STI(sti), MCIP(Printer), Tracker(STI.getSchedModel()), DG(S.size() * 3),
-      Source(S), Iterations(NumIter), TotalCycles(0),
+    : InstructionView(sti, Printer, S), Tracker(sti.getSchedModel()),
+      DG(S.size() * 3), Iterations(NumIter), TotalCycles(0),
       PressureIncreasedBecauseOfResources(false),
       PressureIncreasedBecauseOfRegisterDependencies(false),
       PressureIncreasedBecauseOfMemoryDependencies(false),
@@ -461,7 +454,7 @@ BottleneckAnalysis::BottleneckAnalysis(const MCSubtargetInfo &sti,
 void BottleneckAnalysis::addRegisterDep(unsigned From, unsigned To,
                                         unsigned RegID, unsigned Cost) {
   bool IsLoopCarried = From >= To;
-  unsigned SourceSize = Source.size();
+  unsigned SourceSize = getSource().size();
   if (IsLoopCarried) {
     DG.addRegisterDep(From, To + SourceSize, RegID, Cost);
     DG.addRegisterDep(From + SourceSize, To + (SourceSize * 2), RegID, Cost);
@@ -473,7 +466,7 @@ void BottleneckAnalysis::addRegisterDep(unsigned From, unsigned To,
 void BottleneckAnalysis::addMemoryDep(unsigned From, unsigned To,
                                       unsigned Cost) {
   bool IsLoopCarried = From >= To;
-  unsigned SourceSize = Source.size();
+  unsigned SourceSize = getSource().size();
   if (IsLoopCarried) {
     DG.addMemoryDep(From, To + SourceSize, Cost);
     DG.addMemoryDep(From + SourceSize, To + (SourceSize * 2), Cost);
@@ -485,7 +478,7 @@ void BottleneckAnalysis::addMemoryDep(unsigned From, unsigned To,
 void BottleneckAnalysis::addResourceDep(unsigned From, unsigned To,
                                         uint64_t Mask, unsigned Cost) {
   bool IsLoopCarried = From >= To;
-  unsigned SourceSize = Source.size();
+  unsigned SourceSize = getSource().size();
   if (IsLoopCarried) {
     DG.addResourceDep(From, To + SourceSize, Mask, Cost);
     DG.addResourceDep(From + SourceSize, To + (SourceSize * 2), Mask, Cost);
@@ -508,6 +501,7 @@ void BottleneckAnalysis::onEvent(const HWInstructionEvent &Event) {
   if (Event.Type != HWInstructionEvent::Issued)
     return;
 
+  ArrayRef<llvm::MCInst> Source = getSource();
   const Instruction &IS = *Event.IR.getInstruction();
   unsigned To = IID % Source.size();
 
@@ -617,7 +611,7 @@ void BottleneckAnalysis::printBottleneckHints(raw_ostream &OS) const {
 
   if (BPI.PressureIncreaseCycles) {
     ArrayRef<unsigned> Distribution = Tracker.getResourcePressureDistribution();
-    const MCSchedModel &SM = STI.getSchedModel();
+    const MCSchedModel &SM = getSubTargetInfo().getSchedModel();
     for (unsigned I = 0, E = Distribution.size(); I < E; ++I) {
       unsigned ResourceCycles = Distribution[I];
       if (ResourceCycles) {
diff --git a/llvm/tools/llvm-mca/Views/BottleneckAnalysis.h b/llvm/tools/llvm-mca/Views/BottleneckAnalysis.h
index 9e3bd5978f0931..a0aad9faff4098 100644
--- a/llvm/tools/llvm-mca/Views/BottleneckAnalysis.h
+++ b/llvm/tools/llvm-mca/Views/BottleneckAnalysis.h
@@ -87,6 +87,7 @@
 #include "llvm/MC/MCSchedule.h"
 #include "llvm/MC/MCSubtargetInfo.h"
 #include "llvm/Support/raw_ostream.h"
+#include "llvm/Support/FormattedStream.h"
 
 namespace llvm {
 namespace mca {
@@ -282,13 +283,10 @@ class DependencyGraph {
 };
 
 /// A view that collects and prints a few performance numbers.
-class BottleneckAnalysis : public View {
-  const MCSubtargetInfo &STI;
-  MCInstPrinter &MCIP;
+class BottleneckAnalysis : public InstructionView {
   PressureTracker Tracker;
   DependencyGraph DG;
 
-  ArrayRef<MCInst> Source;
   unsigned Iterations;
   unsigned TotalCycles;
 
@@ -317,6 +315,9 @@ class BottleneckAnalysis : public View {
   void addMemoryDep(unsigned From, unsigned To, unsigned Cy);
   void addResourceDep(unsigned From, unsigned To, uint64_t Mask, unsigned Cy);
 
+  void printInstruction(formatted_raw_ostream &FOS, const MCInst &MCI,
+                        bool UseDifferentColor = false) const;
+
   // Prints a bottleneck message to OS.
   void printBottleneckHints(raw_ostream &OS) const;
   void printCriticalSequence(raw_ostream &OS) const;
diff --git a/llvm/tools/llvm-mca/Views/InstructionInfoView.cpp b/llvm/tools/llvm-mca/Views/InstructionInfoView.cpp
index 548f0b36d0557a..da53f16a991389 100644
--- a/llvm/tools/llvm-mca/Views/InstructionInfoView.cpp
+++ b/llvm/tools/llvm-mca/Views/InstructionInfoView.cpp
@@ -20,9 +20,8 @@ namespace mca {
 void InstructionInfoView::printView(raw_ostream &OS) const {
   std::string Buffer;
   raw_string_ostream TempStream(Buffer);
-  std::string Instruction;
-  raw_string_ostream InstrStream(Instruction);
 
+  ArrayRef<llvm::MCInst> Source = getSource();
   if (!Source.size())
     return;
 
@@ -82,14 +81,7 @@ void InstructionInfoView::printView(raw_ostream &OS) const {
     }
 
     const MCInst &Inst = std::get<1>(I.value());
-    MCIP.printInst(&Inst, 0, "", STI, InstrStream);
-    InstrStream.flush();
-
-    // Consume any tabs or spaces at the beginning of the string.
-    StringRef Str(Instruction);
-    Str = Str.ltrim();
-    TempStream << Str << '\n';
-    Instruction = "";
+    TempStream << printInstructionString(Inst) << '\n';
   }
 
   TempStream.flush();
@@ -98,8 +90,9 @@ void InstructionInfoView::printView(raw_ostream &OS) const {
 
 void InstructionInfoView::collectData(
     MutableArrayRef<InstructionInfoViewData> IIVD) const {
+  const llvm::MCSubtargetInfo &STI = getSubTargetInfo();
   const MCSchedModel &SM = STI.getSchedModel();
-  for (auto I : zip(Source, IIVD)) {
+  for (auto I : zip(getSource(), IIVD)) {
     const MCInst &Inst = std::get<0>(I);
     InstructionInfoViewData &IIVDEntry = std::get<1>(I);
     const MCInstrDesc &MCDesc = MCII.get(Inst.getOpcode());
diff --git a/llvm/tools/llvm-mca/Views/InstructionInfoView.h b/llvm/tools/llvm-mca/Views/InstructionInfoView.h
index aca7e5894a8921..c2093b2d042910 100644
--- a/llvm/tools/llvm-mca/Views/InstructionInfoView.h
+++ b/llvm/tools/llvm-mca/Views/InstructionInfoView.h
@@ -50,13 +50,10 @@ namespace llvm {
 namespace mca {
 
 /// A view that prints out generic instruction information.
-class InstructionInfoView : public View {
-  const llvm::MCSubtargetInfo &STI;
+class InstructionInfoView : public InstructionView {
   const llvm::MCInstrInfo &MCII;
   CodeEmitter &CE;
   bool PrintEncodings;
-  llvm::ArrayRef<llvm::MCInst> Source;
-  llvm::MCInstPrinter &MCIP;
 
   struct InstructionInfoViewData {
     unsigned NumMicroOpcodes = 0;
@@ -76,8 +73,8 @@ class InstructionInfoView : public View {
                       const llvm::MCInstrInfo &II, CodeEmitter &C,
                       bool ShouldPrintEncodings, llvm::ArrayRef<llvm::MCInst> S,
                       llvm::MCInstPrinter &IP)
-      : STI(ST), MCII(II), CE(C), PrintEncodings(ShouldPrintEncodings),
-        Source(S), MCIP(IP) {}
+      : InstructionView(ST, IP, S), MCII(II), CE(C),
+        PrintEncodings(ShouldPrintEncodings) {}
 
   void printView(llvm::raw_ostream &OS) const override;
 };
diff --git a/llvm/tools/llvm-mca/Views/ResourcePressureView.cpp b/llvm/tools/llvm-mca/Views/ResourcePressureView.cpp
index bdb9dc21247b08..a5a74210c67263 100644
--- a/llvm/tools/llvm-mca/Views/ResourcePressureView.cpp
+++ b/llvm/tools/llvm-mca/Views/ResourcePressureView.cpp
@@ -21,10 +21,10 @@ namespace mca {
 ResourcePressureView::ResourcePressureView(const llvm::MCSubtargetInfo &sti,
                                            MCInstPrinter &Printer,
                                            ArrayRef<MCInst> S)
-    : STI(sti), MCIP(Printer), Source(S), LastInstructionIdx(0) {
+    : InstructionView(sti, Printer, S), LastInstructionIdx(0) {
   // Populate the map of resource descriptors.
   unsigned R2VIndex = 0;
-  const MCSchedModel &SM = STI.getSchedModel();
+  const MCSchedModel &SM = getSubTargetInfo().getSchedModel();
   for (unsigned I = 0, E = SM.getNumProcResourceKinds(); I < E; ++I) {
     const MCProcResourceDesc &ProcResource = *SM.getProcResource(I);
     unsigned NumUnits = ProcResource.NumUnits;
@@ -37,7 +37,7 @@ ResourcePressureView::ResourcePressureView(const llvm::MCSubtargetInfo &sti,
   }
 
   NumResourceUnits = R2VIndex;
-  ResourceUsage.resize(NumResourceUnits * (Source.size() + 1));
+  ResourceUsage.resize(NumResourceUnits * (getSource().size() + 1));
   std::fill(ResourceUsage.begin(), ResourceUsage.end(), 0.0);
 }
 
@@ -52,6 +52,7 @@ void ResourcePressureView::onEvent(const HWInstructionEvent &Event) {
     return;
 
   const auto &IssueEvent = static_cast<const HWInstructionIssuedEvent &>(Event);
+  ArrayRef<llvm::MCInst> Source = getSource();
   const unsigned SourceIdx = Event.IR.getSourceIndex() % Source.size();
   for (const std::pair<ResourceRef, ResourceCycles> &Use :
        IssueEvent.UsedResources) {
@@ -105,7 +106,7 @@ void ResourcePressureView::printResourcePressurePerIter(raw_ostream &OS) const {
   formatted_raw_ostream FOS(TempStream);
 
   FOS << "\n\nResources:\n";
-  const MCSchedModel &SM = STI.getSchedModel();
+  const MCSchedModel &SM = getSubTargetInfo().getSchedModel();
   for (unsigned I = 1, ResourceIndex = 0, E = SM.getNumProcResourceKinds();
        I < E; ++I) {
     const MCProcResourceDesc &ProcResource = *SM.getProcResource(I);
@@ -132,6 +133,7 @@ void ResourcePressureView::printResourcePressurePerIter(raw_ostream &OS) const {
   FOS << '\n';
   FOS.flush();
 
+  ArrayRef<llvm::MCInst> Source = getSource();
   const unsigned Executions = LastInstructionIdx / Source.size() + 1;
   for (unsigned I = 0, E = NumResourceUnits; I < E; ++I) {
     double Usage = ResourceUsage[I + Source.size() * E];
@@ -148,13 +150,11 @@ void ResourcePressureView::printResourcePressurePerInst(raw_ostream &OS) const {
   formatted_raw_ostream FOS(TempStream);
 
   FOS << "\n\nResource pressure by instruction:\n";
-  printColumnNames(FOS, STI.getSchedModel());
+  printColumnNames(FOS, getSubTargetInfo().getSchedModel());
   FOS << "Instructions:\n";
 
-  std::string Instruction;
-  raw_string_ostream InstrStream(Instruction);
-
   unsigned InstrIndex = 0;
+  ArrayRef<llvm::MCInst> Source = getSource();
   const unsigned Executions = LastInstructionIdx / Source.size() + 1;
   for (const MCInst &MCI : Source) {
     unsigned BaseEltIdx = InstrIndex * NumResourceUnits;
@@ -163,16 +163,7 @@ void ResourcePressureView::printResourcePressurePerInst(raw_ostream &OS) const {
       printResourcePressure(FOS, Usage / Executions, (J + 1) * 7);
     }
 
-    MCIP.printInst(&MCI, 0, "", STI, InstrStream);
-    InstrStream.flush();
-    StringRef Str(Instruction);
-
-    // Remove any tabs or spaces at the beginning of the instruction.
-    Str = Str.ltrim();
-
-    FOS << Str << '\n';
-    Instruction = "";
-
+    FOS << printInstructionString(MCI) << '\n';
     FOS.flush();
     OS << Buffer;
     Buffer = "";
diff --git a/llvm/tools/llvm-mca/Views/ResourcePressureView.h b/llvm/tools/llvm-mca/Views/ResourcePressureView.h
index 0fa0b9a36aa381..39914f9e2f27d7 100644
--- a/llvm/tools/llvm-mca/Views/ResourcePressureView.h
+++ b/llvm/tools/llvm-mca/Views/ResourcePressureView.h
@@ -69,10 +69,7 @@ namespace mca {
 
 /// This class collects resource pressure statistics and it is able to print
 /// out all the collected information as a table to an output stream.
-class ResourcePressureView : public View {
-  const llvm::MCSubtargetInfo &STI;
-  llvm::MCInstPrinter &MCIP;
-  llvm::ArrayRef<llvm::MCInst> Source;
+class ResourcePressureView : public InstructionView {
   unsigned LastInstructionIdx;
 
   // Map to quickly obtain the ResourceUsage column index from a processor
diff --git a/llvm/tools/llvm-mca/Views/TimelineView.cpp b/llvm/tools/llvm-mca/Views/TimelineView.cpp
index cf5b48e811b88d..f520c5c31d3f84 100644
--- a/llvm/tools/llvm-mca/Views/TimelineView.cpp
+++ b/llvm/tools/llvm-mca/Views/TimelineView.cpp
@@ -20,10 +20,10 @@ namespace mca {
 TimelineView::TimelineView(const MCSubtargetInfo &sti, MCInstPrinter &Printer,
                            llvm::ArrayRef<llvm::MCInst> S, unsigned Iterations,
                            unsigned Cycles)
-    : STI(sti), MCIP(Printer), Source(S), CurrentCycle(0),
+    : InstructionView(sti, Printer, S), CurrentCycle(0),
       MaxCycle(Cycles == 0 ? 80 : Cycles), LastCycle(0), WaitTime(S.size()),
       UsedBuffer(S.size()) {
-  unsigned NumInstructions = Source.size();
+  unsigned NumInstructions = getSource().size();
   assert(Iterations && "Invalid number of iterations specified!");
   NumInstructions *= Iterations;
   Timeline.resize(NumInstructions);
@@ -40,10 +40,10 @@ TimelineView::TimelineView(const MCSubtargetInfo &sti, MCInstPrinter &Printer,
 
 void TimelineView::onReservedBuffers(const InstRef &IR,
                                      ArrayRef<unsigned> Buffers) {
-  if (IR.getSourceIndex() >= Source.size())
+  if (IR.getSourceIndex() >= getSource().size())
     return;
 
-  const MCSchedModel &SM = STI.getSchedModel();
+  const MCSchedModel &SM = getSubTargetInfo().getSchedModel();
   std::pair<unsigned, int> BufferInfo = {0, -1};
   for (const unsigned Buffer : Buffers) {
     const MCProcResourceDesc &MCDesc = *SM.getProcResource(Buffer);
@@ -70,7 +70,7 @@ void TimelineView::onEvent(const HWInstructionEvent &Event) {
     // Update the WaitTime entry which corresponds to this Index.
     assert(TVEntry.CycleDispatched >= 0 && "Invalid TVEntry found!");
     unsigned CycleDispatched = static_cast<unsigned>(TVEntry.CycleDispatched);
-    WaitTimeEntry &WTEntry = WaitTime[Index % Source.size()];
+    WaitTimeEntry &WTEntry = WaitTime[Index % getSource().size()];
     WTEntry.CyclesSpentInSchedulerQueue +=
         TVEntry.CycleIssued - CycleDispatched;
     assert(CycleDispatched <= TVEntry.CycleReady &&
@@ -133,7 +133,7 @@ void TimelineView::printWaitTimeEntry(formatted_raw_ostream &OS,
                                       const WaitTimeEntry &Entry,
                                       unsigned SourceIndex,
                                       unsigned Executions) const {
-  bool PrintingTotals = SourceIndex == Source.size();
+  bool PrintingTotals = SourceIndex == getSource().size();
   unsigned CumulativeExecutions = PrintingTotals ? Timeline.size() : Executions;
 
   if (!PrintingTotals)
@@ -164,7 +164,8 @@ void TimelineView::printWaitTimeEntry(formatted_raw_ostream &OS,
   OS.PadToColumn(27);
   if (!PrintingTotals)
     tryChangeColor(OS, Entry.CyclesSpentAfterWBAndBeforeRetire,
-                   CumulativeExecutions, STI.getSchedModel().MicroOpBufferSize);
+                   CumulativeExecutions,
+                   getSubTargetInfo().getSchedModel().MicroOpBufferSize);
   OS << format("%.1f", floor((AverageTime3 * 10) + 0.5) / 10);
 
   if (OS.has_colors())
@@ -181,33 +182,19 @@ void TimelineView::printAverageWaitTimes(raw_ostream &OS) const {
       "[3]: Average time elapsed from WB until retire stage\n\n"
       "      [0]    [1]    [2]    [3]\n";
   OS << Header;
-
-  // Use a different string stream for printing instructions.
-  std::string Instruction;
-  raw_string_ostream InstrStream(Instruction);
-
   formatted_raw_ostream FOS(OS);
-  unsigned Executions = Timeline.size() / Source.size();
+  unsigned Executions = Timeline.size() / getSource().size();
   unsigned IID = 0;
-  for (const MCInst &Inst : Source) {
+  for (const MCInst &Inst : getSource()) {
     printWaitTimeEntry(FOS, WaitTime[IID], IID, Executions);
-    // Append the instruction info at the end of the line.
-    MCIP.printInst(&Inst, 0, "", STI, InstrStream);
-    InstrStream.flush();
-
-    // Consume any tabs or spaces at the beginning of the string.
-    StringRef Str(Instruction);
-    Str = Str.ltrim();
-    FOS << "   " << Str << '\n';
+    FOS << "   " << printInstructionString(Inst) << '\n';
     FOS.flush();
-    Instruction = "";
-
     ++IID;
   }
 
   // If the timeline contains more than one instruction,
   // let's also print global averages.
-  if (Source.size() != 1) {
+  if (getSource().size() != 1) {
     WaitTimeEntry TotalWaitTime = std::accumulate(
         WaitTime.begin(), WaitTime.end(), WaitTimeEntry{0, 0, 0},
         [](const WaitTimeEntry &A, const WaitTimeEntry &B) {
@@ -220,7 +207,7 @@ void TimelineView::printAverageWaitTimes(raw_ostream &OS) const {
     printWaitTimeEntry(FOS, TotalWaitTime, IID, Executions);
     FOS << "   "
         << "<total>" << '\n';
-    InstrStream.flush();
+    FOS.flush();
   }
 }
 
@@ -292,11 +279,8 @@ void TimelineView::printTimeline(raw_ostream &OS) const {
   printTimelineHeader(FOS, LastCycle);
   FOS.flush();
 
-  // Use a different string stream for the instruction.
-  std::string Instruction;
-  raw_string_ostream InstrStream(Instruction);
-
   unsigned IID = 0;
+  ArrayRef<llvm::MCInst> Source = getSource();
   const unsigned Iterations = Timeline.size() / Source.size();
   for (unsigned Iteration = 0; Iteration < Iterations; ++Iteration) {
     for (const MCInst &Inst : Source) {
@@ -306,16 +290,8 @@ void TimelineView::printTimeline(raw_ostream &OS) const {
 
       unsigned SourceIndex = IID % Source.size();
       printTimelineViewEntry(FOS, Entry, Iteration, SourceIndex);
-      // Append the instruction info at the end of the line.
-      MCIP.printInst(&Inst, 0, "", STI, InstrStream);
-      InstrStream.flush();
-
-      // Consume any tabs or spaces at the beginning of the string.
-      StringRef Str(Instruction);
-      Str = Str.ltrim();
-      FOS << "   " << Str << '\n';
+      FOS << "   " << printInstructionString(Inst) << '\n';
       FOS.flush();
-      Instruction = "";
 
       ++IID;
     }
diff --git a/llvm/tools/llvm-mca/Views/TimelineView.h b/llvm/tools/llvm-mca/Views/TimelineView.h
index 9bec3b87db45d3..528579edf7087d 100644
--- a/llvm/tools/llvm-mca/Views/TimelineView.h
+++ b/llvm/tools/llvm-mca/Views/TimelineView.h
@@ -118,11 +118,7 @@ namespace mca {
 /// a TimelineViewEntry object. TimelineViewEntry objects are then used
 /// to print the timeline information, as well as the "average wait times"
 /// for every instruction in the input assembly sequence.
-class TimelineView : public View {
-  const llvm::MCSubtargetInfo &STI;
-  llvm::MCInstPrinter &MCIP;
-  llvm::ArrayRef<llvm::MCInst> Source;
-
+class TimelineView : public InstructionView {
   unsigned CurrentCycle;
   unsigned MaxCycle;
   unsigned LastCycle;
diff --git a/llvm/tools/llvm-mca/Views/View.cpp b/llvm/tools/llvm-mca/Views/View.cpp
index 8e5c34d2d5c247..4cef7456f36616 100644
--- a/llvm/tools/llvm-mca/Views/View.cpp
+++ b/llvm/tools/llvm-mca/Views/View.cpp
@@ -17,5 +17,13 @@ namespace llvm {
 namespace mca {
 
 void View::anchor() {}
+
+StringRef InstructionView::printInstructionString(const llvm::MCInst &MCI) const {
+    InstructionString = "";
+    MCIP.printInst(&MCI, 0, "", STI, InstrStream);
+    InstrStream.flush();
+    // Remove any tabs or spaces at the beginning of the instruction.
+    return StringRef(InstructionString).ltrim();
+  }
 } // namespace mca
 } // namespace llvm
diff --git a/llvm/tools/llvm-mca/Views/View.h b/llvm/tools/llvm-mca/Views/View.h
index 3b52511b4d2922..1af6e5959f3125 100644
--- a/llvm/tools/llvm-mca/Views/View.h
+++ b/llvm/tools/llvm-mca/Views/View.h
@@ -15,6 +15,7 @@
 #ifndef LLVM_TOOLS_LLVM_MCA_VIEW_H
 #define LLVM_TOOLS_LLVM_MCA_VIEW_H
 
+#include "llvm/MC/MCInstPrinter.h"
 #include "llvm/MCA/HWEventListener.h"
 #include "llvm/Support/raw_ostream.h"
 
@@ -27,6 +28,33 @@ class View : public HWEventListener {
   virtual ~View() = default;
   void anchor() override;
 };
+
+// The base class for views that deal with individual machine instructions.
+class InstructionView : public View {
+  const llvm::MCSubtargetInfo &STI;
+  llvm::MCInstPrinter &MCIP;
+  llvm::ArrayRef<llvm::MCInst> Source;
+
+  mutable std::string InstructionString;
+  mutable raw_string_ostream InstrStream;
+
+protected:
+  InstructionView(const llvm::MCSubtargetInfo &STI,
+                  llvm::MCInstPrinter &Printer,
+                  llvm::ArrayRef<llvm::MCInst> S)
+      : STI(STI), MCIP(Printer), Source(S), InstrStream(InstructionString) {}
+
+  virtual ~InstructionView() = default;
+
+  // Return a reference to a string representing a given machine instruction.
+  // The result should be used or copied before the next call to
+  // printInstructionString() as it will overwrite the previous result.
+  StringRef printInstructionString(const llvm::MCInst &MCI) const;
+  
+  const llvm::MCSubtargetInfo &getSubTargetInfo() const { return STI; }
+  llvm::MCInstPrinter &getInstPrinter() const { return MCIP; }
+  llvm::ArrayRef<llvm::MCInst> getSource() const { return Source; }
+};
 } // namespace mca
 } // namespace llvm
 
diff --git a/mlir/include/mlir/Dialect/LLVMIR/LLVMOps.td b/mlir/include/mlir/Dialect/LLVMIR/LLVMOps.td
index a4a0db171e810a..d2bcd64529b5d0 100644
--- a/mlir/include/mlir/Dialect/LLVMIR/LLVMOps.td
+++ b/mlir/include/mlir/Dialect/LLVMIR/LLVMOps.td
@@ -984,6 +984,14 @@ def LLVM_MatrixTransposeOp
 // LLVM masked operations.
 //
 
+/// Create a llvm.get.active.lane.mask to set a mask up to a given position.
+def LLVM_GetActiveLaneMaskOp
+    : LLVM_OneResultIntrOp<"get.active.lane.mask", [0], [0], [NoSideEffect]>,
+      Arguments<(ins LLVM_Type:$base, LLVM_Type:$n)> {
+  let assemblyFormat = "$base `,` $n attr-dict `:` "
+    "type($base) `,` type($n) `to` type($res)";
+}
+
 /// Create a call to Masked Load intrinsic.
 def LLVM_MaskedLoadOp
     : LLVM_OneResultOp<"intr.masked.load">,
diff --git a/mlir/include/mlir/Dialect/OpenACC/OpenACC.h b/mlir/include/mlir/Dialect/OpenACC/OpenACC.h
index 8d3798103b0ce6..399db74fae2ca8 100644
--- a/mlir/include/mlir/Dialect/OpenACC/OpenACC.h
+++ b/mlir/include/mlir/Dialect/OpenACC/OpenACC.h
@@ -35,7 +35,7 @@ namespace acc {
 ///
 /// Value can be combined bitwise to reflect the mapping applied to the
 /// construct. e.g. `acc.loop gang vector`, the `gang` and `vector` could be
-/// combined and the final mapping value would be 5 (4 & 1).
+/// combined and the final mapping value would be 5 (4 | 1).
 enum OpenACCExecMapping { NONE = 0, VECTOR = 1, WORKER = 2, GANG = 4, SEQ = 8 };
 
 } // end namespace acc
diff --git a/mlir/test/Conversion/StandardToLLVM/calling-convention.mlir b/mlir/test/Conversion/StandardToLLVM/calling-convention.mlir
index 4cdf9d6bace87d..4c318c14a95e60 100644
--- a/mlir/test/Conversion/StandardToLLVM/calling-convention.mlir
+++ b/mlir/test/Conversion/StandardToLLVM/calling-convention.mlir
@@ -173,7 +173,7 @@ func @return_var_memref(%arg0: memref<4x3xf32>) -> memref<*xf32> {
   // CHECK: %[[RANK:.*]] = llvm.extractvalue %[[DESC_2]][0] : !llvm.struct<(i64, ptr<i8>)>
   // CHECK: %[[NEW_DESC_1:.*]] = llvm.insertvalue %[[RANK]], %[[NEW_DESC]][0]
   // CHECK: %[[NEW_DESC_2:.*]] = llvm.insertvalue %[[ALLOCATED]], %[[NEW_DESC_1]][1]
-  // CHECL: llvm.return %[[NEW_DESC_2]]
+  // CHECK: llvm.return %[[NEW_DESC_2]]
   return %0 : memref<*xf32>
 }
 
diff --git a/mlir/test/Target/llvmir-intrinsics.mlir b/mlir/test/Target/llvmir-intrinsics.mlir
index 6bf9b9768dd3ae..8a598e67d17b89 100644
--- a/mlir/test/Target/llvmir-intrinsics.mlir
+++ b/mlir/test/Target/llvmir-intrinsics.mlir
@@ -237,6 +237,13 @@ llvm.func @matrix_intrinsics(%A: !llvm.vec<64 x float>, %B: !llvm.vec<48 x float
   llvm.return
 }
 
+// CHECK-LABEL: @get_active_lane_mask
+llvm.func @get_active_lane_mask(%base: !llvm.i64, %n: !llvm.i64) -> (!llvm.vec<7 x i1>) {
+  // CHECK: call <7 x i1> @llvm.get.active.lane.mask.v7i1.i64(i64 %0, i64 %1)
+  %0 = llvm.intr.get.active.lane.mask %base, %n : !llvm.i64, !llvm.i64 to !llvm.vec<7 x i1>
+  llvm.return %0 : !llvm.vec<7 x i1>
+}
+
 // CHECK-LABEL: @masked_load_store_intrinsics
 llvm.func @masked_load_store_intrinsics(%A: !llvm.ptr<vec<7 x float>>, %mask: !llvm.vec<7 x i1>) {
   // CHECK: call <7 x float> @llvm.masked.load.v7f32.p0v7f32(<7 x float>* %{{.*}}, i32 1, <7 x i1> %{{.*}}, <7 x float> undef)
@@ -314,7 +321,12 @@ llvm.func @memcpy_test(%arg0: !llvm.i32, %arg1: !llvm.i1, %arg2: !llvm.ptr<i8>,
 // CHECK-DAG: declare <48 x float> @llvm.matrix.transpose.v48f32(<48 x float>, i32 immarg, i32 immarg)
 // CHECK-DAG: declare <48 x float> @llvm.matrix.column.major.load.v48f32(float* nocapture, i64, i1 immarg, i32 immarg, i32 immarg)
 // CHECK-DAG: declare void @llvm.matrix.column.major.store.v48f32(<48 x float>, float* nocapture writeonly, i64, i1 immarg, i32 immarg, i32 immarg)
+// CHECK-DAG: declare <7 x i1> @llvm.get.active.lane.mask.v7i1.i64(i64, i64)
 // CHECK-DAG: declare <7 x float> @llvm.masked.load.v7f32.p0v7f32(<7 x float>*, i32 immarg, <7 x i1>, <7 x float>)
 // CHECK-DAG: declare void @llvm.masked.store.v7f32.p0v7f32(<7 x float>, <7 x float>*, i32 immarg, <7 x i1>)
+// CHECK-DAG: declare <7 x float> @llvm.masked.gather.v7f32.v7p0f32(<7 x float*>, i32 immarg, <7 x i1>, <7 x float>)
+// CHECK-DAG: declare void @llvm.masked.scatter.v7f32.v7p0f32(<7 x float>, <7 x float*>, i32 immarg, <7 x i1>)
+// CHECK-DAG: declare <7 x float> @llvm.masked.expandload.v7f32(float*, <7 x i1>, <7 x float>)
+// CHECK-DAG: declare void @llvm.masked.compressstore.v7f32(<7 x float>, float*, <7 x i1>)
 // CHECK-DAG: declare void @llvm.memcpy.p0i8.p0i8.i32(i8* noalias nocapture writeonly, i8* noalias nocapture readonly, i32, i1 immarg)
 // CHECK-DAG: declare void @llvm.memcpy.inline.p0i8.p0i8.i64(i8* noalias nocapture writeonly, i8* noalias nocapture readonly, i64 immarg, i1 immarg)
diff --git a/mlir/test/mlir-tblgen/op-decl.td b/mlir/test/mlir-tblgen/op-decl.td
index a30de0959d5790..0cb793b2cd2ac2 100644
--- a/mlir/test/mlir-tblgen/op-decl.td
+++ b/mlir/test/mlir-tblgen/op-decl.td
@@ -56,7 +56,7 @@ def NS_AOp : NS_Op<"a_op", [IsolatedFromAbove, IsolatedFromAbove]> {
 // CHECK:   ::mlir::Value a();
 // CHECK:   ::mlir::ValueRange b();
 // CHECK:   ::mlir::IntegerAttr attr1();
-// CHECL:   ::mlir::FloatAttr attr2();
+// CHECK:   ::mlir::FloatAttr attr2();
 // CHECK: private:
 // CHECK:   ::mlir::ValueRange odsOperands;
 // CHECK: };