diff --git a/.github/workflows/release-binaries.yml b/.github/workflows/release-binaries.yml
diff --git a/.github/workflows/release-sources.yml b/.github/workflows/release-sources.yml
@@ -47,7 +47,7 @@ jobs:
     steps:
       - id: inputs
         run: |
-          ref=${{ inputs.release-version || github.sha }}
+          ref=${{ (inputs.release-version && format('llvmorg-{0}', inputs.release-version)) || github.sha }}
           if [ -n "${{ inputs.release-version }}" ]; then
             export_args="-release ${{ inputs.release-version }} -final"
           else

diff --git a/.github/workflows/release-tasks.yml b/.github/workflows/release-tasks.yml
@@ -81,10 +81,20 @@ jobs:
     needs:
       - validate-tag
       - release-create
+    strategy:
+      fail-fast: false
+      matrix:
+        runs-on:
+          - ubuntu-22.04
+          - windows-2022
+          - macos-13
+          - macos-14
+
     uses: ./.github/workflows/release-binaries.yml
     with:
       release-version: ${{ needs.validate-tag.outputs.release-version }}
       upload: true
+      runs-on: ${{ matrix.runs-on }}
 
   release-sources:
     name: Package Release Sources

diff --git a/bolt/include/bolt/Core/DebugData.h b/bolt/include/bolt/Core/DebugData.h
@@ -475,7 +475,8 @@ class DebugStrOffsetsWriter {
   }
 
   /// Update Str offset in .debug_str in .debug_str_offsets.
-  void updateAddressMap(uint32_t Index, uint32_t Address);
+  void updateAddressMap(uint32_t Index, uint32_t Address,
+                        const DWARFUnit &Unit);
 
   /// Get offset for given index in original .debug_str_offsets section.
   uint64_t getOffset(uint32_t Index) const { return StrOffsets[Index]; }
@@ -507,6 +508,8 @@ class DebugStrOffsetsWriter {
   std::unique_ptr<DebugStrOffsetsBufferVector> StrOffsetsBuffer;
   std::unique_ptr<raw_svector_ostream> StrOffsetsStream;
   std::map<uint32_t, uint32_t> IndexToAddressMap;
+  [[maybe_unused]]
+  DenseSet<uint64_t> DebugStrOffsetFinalized;
   SmallVector<uint32_t, 5> StrOffsets;
   std::unordered_map<uint64_t, uint64_t> ProcessedBaseOffsets;
   bool StrOffsetSectionWasModified = false;

diff --git a/bolt/lib/Core/DIEBuilder.cpp b/bolt/lib/Core/DIEBuilder.cpp
@@ -78,7 +78,7 @@ static void addStringHelper(DebugStrOffsetsWriter &StrOffstsWriter,
   uint32_t NewOffset = StrWriter.addString(Str);
   if (Unit.getVersion() >= 5) {
     StrOffstsWriter.updateAddressMap(DIEAttrInfo.getDIEInteger().getValue(),
-                                     NewOffset);
+                                     NewOffset, Unit);
     return;
   }
   DIEBldr.replaceValue(&Die, DIEAttrInfo.getAttribute(), DIEAttrInfo.getForm(),

diff --git a/bolt/lib/Core/DebugData.cpp b/bolt/lib/Core/DebugData.cpp
@@ -851,7 +851,11 @@ void DebugStrOffsetsWriter::initialize(DWARFUnit &Unit) {
         StrOffsetsSection.Data.data() + Contr->Base + Offset));
 }
 
-void DebugStrOffsetsWriter::updateAddressMap(uint32_t Index, uint32_t Address) {
+void DebugStrOffsetsWriter::updateAddressMap(uint32_t Index, uint32_t Address,
+                                             const DWARFUnit &Unit) {
+  assert(DebugStrOffsetFinalized.count(Unit.getOffset()) == 0 &&
+         "Cannot update address map since debug_str_offsets was already "
+         "finalized for this CU.");
   IndexToAddressMap[Index] = Address;
   StrOffsetSectionWasModified = true;
 }
@@ -906,6 +910,8 @@ void DebugStrOffsetsWriter::finalizeSection(DWARFUnit &Unit,
   }
 
   StrOffsetSectionWasModified = false;
+  assert(DebugStrOffsetFinalized.insert(Unit.getOffset()).second &&
+         "debug_str_offsets was already finalized for this CU.");
   clear();
 }
 

diff --git a/bolt/lib/Rewrite/DWARFRewriter.cpp b/bolt/lib/Rewrite/DWARFRewriter.cpp
@@ -32,6 +32,7 @@
 #include "llvm/MC/MCAssembler.h"
 #include "llvm/MC/MCObjectWriter.h"
 #include "llvm/MC/MCStreamer.h"
+#include "llvm/MC/MCTargetOptionsCommandFlags.h"
 #include "llvm/Object/ObjectFile.h"
 #include "llvm/Support/Casting.h"
 #include "llvm/Support/CommandLine.h"
@@ -56,6 +57,8 @@
 #undef DEBUG_TYPE
 #define DEBUG_TYPE "bolt"
 
+static mc::RegisterMCTargetOptionsFlags MOF;
+
 static void printDie(const DWARFDie &DIE) {
   DIDumpOptions DumpOpts;
   DumpOpts.ShowForm = true;
@@ -711,7 +714,8 @@ void DWARFRewriter::updateDebugInfo() {
       RangesBase = RangesSectionWriter.getSectionOffset() +
                    getDWARF5RngListLocListHeaderSize();
       RangesSectionWriter.initSection(Unit);
-      StrOffstsWriter->finalizeSection(Unit, DIEBlder);
+      if (!SplitCU)
+        StrOffstsWriter->finalizeSection(Unit, DIEBlder);
     } else if (SplitCU) {
       RangesBase = LegacyRangesSectionWriter.get()->getSectionOffset();
     }
@@ -757,6 +761,8 @@ void DWARFRewriter::updateDebugInfo() {
               : std::optional<std::string>(opts::DwarfOutputPath.c_str());
       std::string DWOName = DIEBlder.updateDWONameCompDir(
           *StrOffstsWriter, *StrWriter, *CU, DwarfOutputPath, std::nullopt);
+      if (CU->getVersion() >= 5)
+        StrOffstsWriter->finalizeSection(*CU, DIEBlder);
       processSplitCU(*CU, **SplitCU, DIEBlder, *TempRangesSectionWriter,
                      AddressWriter, DWOName, DwarfOutputPath);
     }

diff --git a/bolt/test/AArch64/dummy-return.s b/bolt/test/AArch64/dummy-return.s
@@ -1,4 +1,6 @@
-# REQUIRES: system-linux,target=aarch64{{.*}}
+# This test checks instrumentation of static binary on AArch64.
+
+# REQUIRES: system-linux,bolt-runtime,target=aarch64{{.*}}
 
 # RUN: llvm-mc -filetype=obj -triple aarch64-unknown-unknown %s -o %t.o
 # RUN: %clang %cflags %t.o -o %t.exe -Wl,-q -static

diff --git a/bolt/test/X86/dwarf5-df-larger-batch-size.test b/bolt/test/X86/dwarf5-df-larger-batch-size.test
@@ -0,0 +1,28 @@
+; RUN: rm -rf %t
+; RUN: mkdir %t
+; RUN: cd %t
+; RUN: llvm-mc -dwarf-version=5 -filetype=obj -triple x86_64-unknown-linux %p/Inputs/dwarf5-df-input-lowpc-ranges-main.s \
+; RUN: -split-dwarf-file=main.dwo -o main.o
+; RUN: llvm-mc -dwarf-version=5 -filetype=obj -triple x86_64-unknown-linux %p/Inputs/dwarf5-df-input-lowpc-ranges-other.s \
+; RUN: -split-dwarf-file=mainOther.dwo -o other.o
+; RUN: %clang %cflags main.o other.o -o main.exe
+; RUN: llvm-bolt main.exe -o main.exe.bolt --update-debug-sections --cu-processing-batch-size=1
+; RUN: llvm-bolt main.exe -o main-batch.exe.bolt --update-debug-sections --cu-processing-batch-size=2
+; RUN: llvm-dwarfdump --show-form --verbose --debug-info main.exe.bolt >> %t/foo.txt
+; RUN: cat %t/foo.txt | FileCheck -check-prefix=BOLT %s
+; RUN: llvm-dwarfdump --show-form --verbose --debug-info main.exe.bolt >> %t/foo-batch.txt
+; RUN: cat %t/foo-batch.txt | FileCheck -check-prefix=BOLT-BATCH %s
+
+;; Tests that BOLT correctly handles DWO name strings with larger batch sizes.
+
+; BOLT: DW_TAG_skeleton_unit
+; BOLT: DW_AT_dwo_name [DW_FORM_strx1]  (indexed (00000001) string = "main.dwo.dwo")
+
+; BOLT: DW_TAG_skeleton_unit
+; BOLT: DW_AT_dwo_name [DW_FORM_strx1]  (indexed (00000001) string = "mainOther.dwo.dwo")
+
+; BOLT-BATCH: DW_TAG_skeleton_unit
+; BOLT-BATCH: DW_AT_dwo_name [DW_FORM_strx1]  (indexed (00000001) string = "main.dwo.dwo")
+
+; BOLT-BATCH: DW_TAG_skeleton_unit
+; BOLT-BATCH: DW_AT_dwo_name [DW_FORM_strx1]  (indexed (00000001) string = "mainOther.dwo.dwo")
diff --git a/clang-tools-extra/clang-tidy/add_new_check.py b/clang-tools-extra/clang-tidy/add_new_check.py
@@ -13,9 +13,12 @@
 
 import argparse
 import io
+import itertools
 import os
 import re
 import sys
+import textwrap
+
 
 # Adapts the module's CMakelist file. Returns 'True' if it could add a new
 # entry and 'False' if the entry already existed.
@@ -53,7 +56,29 @@ def adapt_cmake(module_path, check_name_camel):
 
 
 # Adds a header for the new check.
-def write_header(module_path, module, namespace, check_name, check_name_camel):
+def write_header(
+    module_path,
+    module,
+    namespace,
+    check_name,
+    check_name_camel,
+    description,
+    lang_restrict,
+):
+    wrapped_desc = "\n".join(
+        textwrap.wrap(
+            description, width=80, initial_indent="/// ", subsequent_indent="/// "
+        )
+    )
+    if lang_restrict:
+        override_supported = """
+  bool isLanguageVersionSupported(const LangOptions &LangOpts) const override {
+    return %s;
+  }""" % (
+            lang_restrict % {"lang": "LangOpts"}
+        )
+    else:
+        override_supported = ""
     filename = os.path.join(module_path, check_name_camel) + ".h"
     print("Creating %s..." % filename)
     with io.open(filename, "w", encoding="utf8", newline="\n") as f:
@@ -85,7 +110,7 @@ def write_header(module_path, module, namespace, check_name, check_name_camel):
 
 namespace clang::tidy::%(namespace)s {
 
-/// FIXME: Write a short description.
+%(description)s
 ///
 /// For the user-facing documentation see:
 /// http://clang.llvm.org/extra/clang-tidy/checks/%(module)s/%(check_name)s.html
@@ -94,7 +119,7 @@ class %(check_name_camel)s : public ClangTidyCheck {
   %(check_name_camel)s(StringRef Name, ClangTidyContext *Context)
       : ClangTidyCheck(Name, Context) {}
   void registerMatchers(ast_matchers::MatchFinder *Finder) override;
-  void check(const ast_matchers::MatchFinder::MatchResult &Result) override;
+  void check(const ast_matchers::MatchFinder::MatchResult &Result) override;%(override_supported)s
 };
 
 } // namespace clang::tidy::%(namespace)s
@@ -107,6 +132,8 @@ class %(check_name_camel)s : public ClangTidyCheck {
                 "check_name": check_name,
                 "module": module,
                 "namespace": namespace,
+                "description": wrapped_desc,
+                "override_supported": override_supported,
             }
         )
 
@@ -235,7 +262,12 @@ def adapt_module(module_path, module, check_name, check_name_camel):
 
 
 # Adds a release notes entry.
-def add_release_notes(module_path, module, check_name):
+def add_release_notes(module_path, module, check_name, description):
+    wrapped_desc = "\n".join(
+        textwrap.wrap(
+            description, width=80, initial_indent="  ", subsequent_indent="  "
+        )
+    )
     check_name_dashes = module + "-" + check_name
     filename = os.path.normpath(
         os.path.join(module_path, "../../docs/ReleaseNotes.rst")
@@ -281,18 +313,20 @@ def add_release_notes(module_path, module, check_name):
                             """- New :doc:`%s
   <clang-tidy/checks/%s/%s>` check.
 
-  FIXME: add release notes.
+%s
 
 """
-                            % (check_name_dashes, module, check_name)
+                            % (check_name_dashes, module, check_name, wrapped_desc)
                         )
                         note_added = True
 
             f.write(line)
 
 
 # Adds a test for the check.
-def write_test(module_path, module, check_name, test_extension):
+def write_test(module_path, module, check_name, test_extension, test_standard):
+    if test_standard:
+        test_standard = f"-std={test_standard}-or-later "
     check_name_dashes = module + "-" + check_name
     filename = os.path.normpath(
         os.path.join(
@@ -309,7 +343,7 @@ def write_test(module_path, module, check_name, test_extension):
     print("Creating %s..." % filename)
     with io.open(filename, "w", encoding="utf8", newline="\n") as f:
         f.write(
-            """// RUN: %%check_clang_tidy %%s %(check_name_dashes)s %%t
+            """// RUN: %%check_clang_tidy %(standard)s%%s %(check_name_dashes)s %%t
 
 // FIXME: Add something that triggers the check here.
 void f();
@@ -324,7 +358,7 @@ def write_test(module_path, module, check_name, test_extension):
 // FIXME: Add something that doesn't trigger the check here.
 void awesome_f2();
 """
-            % {"check_name_dashes": check_name_dashes}
+            % {"check_name_dashes": check_name_dashes, "standard": test_standard}
         )
 
 
@@ -497,7 +531,10 @@ def format_link_alias(doc_file):
         if (match or (check_name.startswith("clang-analyzer-"))) and check_name:
             module = doc_file[0]
             check_file = doc_file[1].replace(".rst", "")
-            if not match or match.group(1) == "https://clang.llvm.org/docs/analyzer/checkers":
+            if (
+                not match
+                or match.group(1) == "https://clang.llvm.org/docs/analyzer/checkers"
+            ):
                 title = "Clang Static Analyzer " + check_file
                 # Preserve the anchor in checkers.html from group 2.
                 target = "" if not match else match.group(1) + ".html" + match.group(2)
@@ -515,29 +552,31 @@ def format_link_alias(doc_file):
             if target:
                 # The checker is just a redirect.
                 return (
-                        "   :doc:`%(check_name)s <%(module)s/%(check_file)s>`, %(ref_begin)s`%(title)s <%(target)s>`%(ref_end)s,%(autofix)s\n"
+                    "   :doc:`%(check_name)s <%(module)s/%(check_file)s>`, %(ref_begin)s`%(title)s <%(target)s>`%(ref_end)s,%(autofix)s\n"
                     % {
                         "check_name": check_name,
                         "module": module,
                         "check_file": check_file,
                         "target": target,
                         "title": title,
                         "autofix": autofix,
-                        "ref_begin" : ref_begin,
-                        "ref_end" : ref_end
-                    })
+                        "ref_begin": ref_begin,
+                        "ref_end": ref_end,
+                    }
+                )
             else:
                 # The checker is just a alias without redirect.
                 return (
-                        "   :doc:`%(check_name)s <%(module)s/%(check_file)s>`, %(title)s,%(autofix)s\n"
+                    "   :doc:`%(check_name)s <%(module)s/%(check_file)s>`, %(title)s,%(autofix)s\n"
                     % {
                         "check_name": check_name,
                         "module": module,
                         "check_file": check_file,
                         "target": target,
                         "title": title,
                         "autofix": autofix,
-                    })
+                    }
+                )
         return ""
 
     checks = map(format_link, doc_files)
@@ -599,6 +638,22 @@ def main():
         "objc": "m",
         "objc++": "mm",
     }
+    cpp_language_to_requirements = {
+        "c++98": "CPlusPlus",
+        "c++11": "CPlusPlus11",
+        "c++14": "CPlusPlus14",
+        "c++17": "CPlusPlus17",
+        "c++20": "CPlusPlus20",
+        "c++23": "CPlusPlus23",
+        "c++26": "CPlusPlus26",
+    }
+    c_language_to_requirements = {
+        "c99": None,
+        "c11": "C11",
+        "c17": "C17",
+        "c23": "C23",
+        "c27": "C2Y",
+    }
     parser = argparse.ArgumentParser()
     parser.add_argument(
         "--update-docs",
@@ -609,9 +664,26 @@ def main():
         "--language",
         help="language to use for new check (defaults to c++)",
         choices=language_to_extension.keys(),
-        default="c++",
+        default=None,
         metavar="LANG",
     )
+    parser.add_argument(
+        "--description",
+        "-d",
+        help="short description of what the check does",
+        default="FIXME: Write a short description",
+        type=str,
+    )
+    parser.add_argument(
+        "--standard",
+        help="Specify a specific version of the language",
+        choices=list(
+            itertools.chain(
+                cpp_language_to_requirements.keys(), c_language_to_requirements.keys()
+            )
+        ),
+        default=None,
+    )
     parser.add_argument(
         "module",
         nargs="?",
@@ -652,12 +724,53 @@ def main():
     else:
         namespace = module
 
-    write_header(module_path, module, namespace, check_name, check_name_camel)
+    description = args.description
+    if not description.endswith("."):
+        description += "."
+
+    language = args.language
+
+    if args.standard:
+        if args.standard in cpp_language_to_requirements:
+            if language and language != "c++":
+                raise ValueError("C++ standard chosen when language is not C++")
+            language = "c++"
+        elif args.standard in c_language_to_requirements:
+            if language and language != "c":
+                raise ValueError("C standard chosen when language is not C")
+            language = "c"
+
+    if not language:
+        language = "c++"
+
+    language_restrict = None
+
+    if language == "c":
+        language_restrict = "!%(lang)s.CPlusPlus"
+        extra = c_language_to_requirements.get(args.standard, None)
+        if extra:
+            language_restrict += f" && %(lang)s.{extra}"
+    elif language == "c++":
+        language_restrict = (
+            f"%(lang)s.{cpp_language_to_requirements.get(args.standard, 'CPlusPlus')}"
+        )
+    elif language in ["objc", "objc++"]:
+        language_restrict = "%(lang)s.ObjC"
+
+    write_header(
+        module_path,
+        module,
+        namespace,
+        check_name,
+        check_name_camel,
+        description,
+        language_restrict,
+    )
     write_implementation(module_path, module, namespace, check_name_camel)
     adapt_module(module_path, module, check_name, check_name_camel)
-    add_release_notes(module_path, module, check_name)
-    test_extension = language_to_extension.get(args.language)
-    write_test(module_path, module, check_name, test_extension)
+    add_release_notes(module_path, module, check_name, description)
+    test_extension = language_to_extension.get(language)
+    write_test(module_path, module, check_name, test_extension, args.standard)
     write_docs(module_path, module, check_name)
     update_checks_list(clang_tidy_path)
     print("Done. Now it's your turn!")

diff --git a/clang-tools-extra/test/clang-doc/basic-project.test b/clang-tools-extra/test/clang-doc/basic-project.test
@@ -1,3 +1,6 @@
+// See https://github.com/llvm/llvm-project/issues/97507.
+// UNSUPPORTED: target={{.*}}
+
 // RUN: rm -rf %t && mkdir -p %t/docs %t/build
 // RUN: sed 's|$test_dir|%/S|g' %S/Inputs/basic-project/database_template.json > %t/build/compile_commands.json
 // RUN: clang-doc --format=html --output=%t/docs --executor=all-TUs %t/build/compile_commands.json

diff --git a/clang-tools-extra/test/clang-tidy/check_clang_tidy.py b/clang-tools-extra/test/clang-tidy/check_clang_tidy.py
@@ -205,9 +205,11 @@ def run_clang_tidy(self):
                 self.temp_file_name,
             ]
             + [
-                "-fix"
-                if self.export_fixes is None
-                else "--export-fixes=" + self.export_fixes
+                (
+                    "-fix"
+                    if self.export_fixes is None
+                    else "--export-fixes=" + self.export_fixes
+                )
             ]
             + [
                 "--checks=-*," + self.check_name,
@@ -299,19 +301,37 @@ def run(self):
             self.check_notes(clang_tidy_output)
 
 
+CPP_STANDARDS = [
+    "c++98",
+    "c++11",
+    ("c++14", "c++1y"),
+    ("c++17", "c++1z"),
+    ("c++20", "c++2a"),
+    ("c++23", "c++2b"),
+    ("c++26", "c++2c"),
+]
+C_STANDARDS = ["c99", ("c11", "c1x"), "c17", ("c23", "c2x"), "c2y"]
+
+
 def expand_std(std):
-    if std == "c++98-or-later":
-        return ["c++98", "c++11", "c++14", "c++17", "c++20", "c++23", "c++2c"]
-    if std == "c++11-or-later":
-        return ["c++11", "c++14", "c++17", "c++20", "c++23", "c++2c"]
-    if std == "c++14-or-later":
-        return ["c++14", "c++17", "c++20", "c++23", "c++2c"]
-    if std == "c++17-or-later":
-        return ["c++17", "c++20", "c++23", "c++2c"]
-    if std == "c++20-or-later":
-        return ["c++20", "c++23", "c++2c"]
-    if std == "c++23-or-later":
-        return ["c++23", "c++2c"]
+    split_std, or_later, _ = std.partition("-or-later")
+
+    if not or_later:
+        return [split_std]
+
+    for standard_list in (CPP_STANDARDS, C_STANDARDS):
+        item = next(
+            (
+                i
+                for i, v in enumerate(standard_list)
+                if (split_std in v if isinstance(v, (list, tuple)) else split_std == v)
+            ),
+            None,
+        )
+        if item is not None:
+            return [split_std] + [
+                x if isinstance(x, str) else x[0] for x in standard_list[item + 1 :]
+            ]
     return [std]
 
 

diff --git a/clang/cmake/caches/Release.cmake b/clang/cmake/caches/Release.cmake
@@ -29,9 +29,13 @@ endfunction()
 # cache file to CMake via -C. e.g.
 #
 # cmake -D LLVM_RELEASE_ENABLE_PGO=ON -C Release.cmake
+set (DEFAULT_RUNTIMES "compiler-rt;libcxx")
+if (NOT WIN32)
+  list(APPEND DEFAULT_RUNTIMES "libcxxabi" "libunwind")
+endif()
 set(LLVM_RELEASE_ENABLE_LTO THIN CACHE STRING "")
 set(LLVM_RELEASE_ENABLE_PGO ON CACHE BOOL "")
-set(LLVM_RELEASE_ENABLE_RUNTIMES "compiler-rt;libcxx;libcxxabi;libunwind" CACHE STRING "")
+set(LLVM_RELEASE_ENABLE_RUNTIMES ${DEFAULT_RUNTIMES} CACHE STRING "")
 set(LLVM_RELEASE_ENABLE_PROJECTS "clang;lld;lldb;clang-tools-extra;bolt;polly;mlir;flang" CACHE STRING "")
 # Note we don't need to add install here, since it is one of the pre-defined
 # steps.

diff --git a/clang/docs/LanguageExtensions.rst b/clang/docs/LanguageExtensions.rst
@@ -1546,6 +1546,7 @@ The following type trait primitives are supported by Clang. Those traits marked
 * ``__array_extent(type, dim)`` (Embarcadero):
   The ``dim``'th array bound in the type ``type``, or ``0`` if
   ``dim >= __array_rank(type)``.
+* ``__builtin_is_virtual_base_of`` (C++, GNU, Microsoft)
 * ``__can_pass_in_regs`` (C++)
   Returns whether a class can be passed in registers under the current
   ABI. This type can only be applied to unqualified class types.

diff --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst
@@ -81,6 +81,9 @@ C++23 Feature Support
 C++2c Feature Support
 ^^^^^^^^^^^^^^^^^^^^^
 
+- Add ``__builtin_is_virtual_base_of`` intrinsic, which supports
+  `P2985R0 A type trait for detecting virtual base classes <https://wg21.link/p2985r0>`_
+
 Resolutions to C++ Defect Reports
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
@@ -113,6 +116,9 @@ Attribute Changes in Clang
 
 - Introduced a new format attribute ``__attribute__((format(syslog, 1, 2)))`` from OpenBSD.
 
+- The ``hybrid_patchable`` attribute is now supported on ARM64EC targets. It can be used to specify
+  that a function requires an additional x86-64 thunk, which may be patched at runtime.
+
 Improvements to Clang's diagnostics
 -----------------------------------
 
@@ -130,6 +136,8 @@ Improvements to Clang's diagnostics
 
 - Clang now diagnoses undefined behavior in constant expressions more consistently. This includes invalid shifts, and signed overflow in arithmetic.
 
+- -Wdangling-assignment-gsl is enabled by default.
+
 Improvements to Clang's time-trace
 ----------------------------------
 
@@ -163,6 +171,9 @@ Miscellaneous Bug Fixes
 Miscellaneous Clang Crashes Fixed
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
+- Fixed a crash in C due to incorrect lookup that members in nested anonymous struct/union
+  can be found as ordinary identifiers in struct/union definition. (#GH31295)
+
 OpenACC Specific Changes
 ------------------------
 

diff --git a/clang/include/clang/AST/Type.h b/clang/include/clang/AST/Type.h
@@ -4698,26 +4698,25 @@ class FunctionEffect {
   };
 
 private:
-  LLVM_PREFERRED_TYPE(Kind)
-  unsigned FKind : 3;
+  Kind FKind;
 
   // Expansion: for hypothetical TCB+types, there could be one Kind for TCB,
   // then ~16(?) bits "SubKind" to map to a specific named TCB. SubKind would
   // be considered for uniqueness.
 
 public:
-  FunctionEffect() : FKind(unsigned(Kind::None)) {}
+  FunctionEffect() : FKind(Kind::None) {}
 
-  explicit FunctionEffect(Kind K) : FKind(unsigned(K)) {}
+  explicit FunctionEffect(Kind K) : FKind(K) {}
 
   /// The kind of the effect.
-  Kind kind() const { return Kind(FKind); }
+  Kind kind() const { return FKind; }
 
   /// Return the opposite kind, for effects which have opposites.
   Kind oppositeKind() const;
 
   /// For serialization.
-  uint32_t toOpaqueInt32() const { return FKind; }
+  uint32_t toOpaqueInt32() const { return uint32_t(FKind); }
   static FunctionEffect fromOpaqueInt32(uint32_t Value) {
     return FunctionEffect(Kind(Value));
   }

diff --git a/clang/include/clang/Basic/Attr.td b/clang/include/clang/Basic/Attr.td
@@ -477,6 +477,9 @@ def TargetELF : TargetSpec {
 def TargetELFOrMachO : TargetSpec {
   let ObjectFormats = ["ELF", "MachO"];
 }
+def TargetWindowsArm64EC : TargetSpec {
+  let CustomCode = [{ Target.getTriple().isWindowsArm64EC() }];
+}
 
 def TargetSupportsInitPriority : TargetSpec {
   let CustomCode = [{ !Target.getTriple().isOSzOS() }];
@@ -4027,6 +4030,12 @@ def SelectAny : InheritableAttr {
   let SimpleHandler = 1;
 }
 
+def HybridPatchable : InheritableAttr, TargetSpecificAttr<TargetWindowsArm64EC> {
+  let Spellings = [Declspec<"hybrid_patchable">, Clang<"hybrid_patchable">];
+  let Subjects = SubjectList<[Function]>;
+  let Documentation = [HybridPatchableDocs];
+}
+
 def Thread : Attr {
   let Spellings = [Declspec<"thread">];
   let LangOpts = [MicrosoftExt];

diff --git a/clang/include/clang/Basic/AttrDocs.td b/clang/include/clang/Basic/AttrDocs.td
@@ -5985,6 +5985,16 @@ For more information see
 or `msvc documentation <https://docs.microsoft.com/pl-pl/cpp/cpp/selectany>`_.
 }]; }
 
+def HybridPatchableDocs : Documentation {
+  let Category = DocCatFunction;
+  let Content = [{
+The ``hybrid_patchable`` attribute declares an ARM64EC function with an additional
+x86-64 thunk, which may be patched at runtime.
+
+For more information see
+`ARM64EC ABI documentation <https://learn.microsoft.com/en-us/windows/arm/arm64ec-abi>`_.
+}]; }
+
 def WebAssemblyExportNameDocs : Documentation {
   let Category = DocCatFunction;
   let Content = [{

diff --git a/clang/include/clang/Basic/DiagnosticSemaKinds.td b/clang/include/clang/Basic/DiagnosticSemaKinds.td
@@ -3681,6 +3681,9 @@ def err_attribute_weak_static : Error<
   "weak declaration cannot have internal linkage">;
 def err_attribute_selectany_non_extern_data : Error<
   "'selectany' can only be applied to data items with external linkage">;
+def warn_attribute_hybrid_patchable_non_extern : Warning<
+  "'hybrid_patchable' is ignored on functions without external linkage">,
+  InGroup<IgnoredAttributes>;
 def err_declspec_thread_on_thread_variable : Error<
   "'__declspec(thread)' applied to variable that already has a "
   "thread-local storage specifier">;
@@ -3812,8 +3815,6 @@ def warn_sme_locally_streaming_has_vl_args_returns : Warning<
   InGroup<AArch64SMEAttributes>, DefaultIgnore;
 def err_conflicting_attributes_arm_state : Error<
   "conflicting attributes for state '%0'">;
-def err_sme_streaming_cannot_be_multiversioned : Error<
-  "streaming function cannot be multi-versioned">;
 def err_unknown_arm_state : Error<
   "unknown state '%0'">;
 def err_missing_arm_state : Error<
@@ -10133,7 +10134,7 @@ def warn_dangling_lifetime_pointer : Warning<
   InGroup<DanglingGsl>;
 def warn_dangling_lifetime_pointer_assignment : Warning<"object backing the "
   "pointer %0 will be destroyed at the end of the full-expression">,
-  InGroup<DanglingAssignmentGsl>, DefaultIgnore;
+  InGroup<DanglingAssignmentGsl>;
 def warn_new_dangling_initializer_list : Warning<
   "array backing "
   "%select{initializer list subobject of the allocated object|"

diff --git a/clang/include/clang/Basic/LangOptions.def b/clang/include/clang/Basic/LangOptions.def
@@ -224,7 +224,6 @@ COMPATIBLE_LANGOPT(GNUInline         , 1, 0, "GNU inline semantics")
 COMPATIBLE_LANGOPT(NoInlineDefine    , 1, 0, "__NO_INLINE__ predefined macro")
 COMPATIBLE_LANGOPT(Deprecated        , 1, 0, "__DEPRECATED predefined macro")
 COMPATIBLE_LANGOPT(FastMath          , 1, 0, "fast FP math optimizations, and __FAST_MATH__ predefined macro")
-COMPATIBLE_LANGOPT(FiniteMathOnly    , 1, 0, "__FINITE_MATH_ONLY__ predefined macro")
 COMPATIBLE_LANGOPT(UnsafeFPMath      , 1, 0, "Unsafe Floating Point Math")
 COMPATIBLE_LANGOPT(ProtectParens     , 1, 0, "optimizer honors parentheses "
                    "when floating-point expressions are evaluated")
@@ -340,7 +339,6 @@ LANGOPT(SinglePrecisionConstants , 1, 0, "treating double-precision floating poi
 LANGOPT(FastRelaxedMath , 1, 0, "OpenCL fast relaxed math")
 BENIGN_LANGOPT(CLNoSignedZero , 1, 0, "Permit Floating Point optimization without regard to signed zeros")
 COMPATIBLE_LANGOPT(CLUnsafeMath , 1, 0, "Unsafe Floating Point Math")
-COMPATIBLE_LANGOPT(CLFiniteMathOnly , 1, 0, "__FINITE_MATH_ONLY__ predefined macro")
 /// FP_CONTRACT mode (on/off/fast).
 BENIGN_ENUM_LANGOPT(DefaultFPContractMode, FPModeKind, 2, FPM_Off, "FP contraction type")
 COMPATIBLE_LANGOPT(ExpStrictFP, 1, false, "Enable experimental strict floating point")

diff --git a/clang/include/clang/Basic/TokenKinds.def b/clang/include/clang/Basic/TokenKinds.def
@@ -501,6 +501,7 @@ TYPE_TRAIT_1(__has_trivial_move_assign, HasTrivialMoveAssign, KEYCXX)
 TYPE_TRAIT_1(__has_trivial_move_constructor, HasTrivialMoveConstructor, KEYCXX)
 
 // GNU and MS Type Traits
+TYPE_TRAIT_2(__builtin_is_virtual_base_of, IsVirtualBaseOf, KEYCXX)
 TYPE_TRAIT_1(__has_nothrow_assign, HasNothrowAssign, KEYCXX)
 TYPE_TRAIT_1(__has_nothrow_copy, HasNothrowCopy, KEYCXX)
 TYPE_TRAIT_1(__has_nothrow_constructor, HasNothrowConstructor, KEYCXX)

diff --git a/clang/include/clang/Driver/Options.td b/clang/include/clang/Driver/Options.td
@@ -1117,8 +1117,7 @@ def cl_single_precision_constant : Flag<["-"], "cl-single-precision-constant">,
   MarshallingInfoFlag<LangOpts<"SinglePrecisionConstants">>;
 def cl_finite_math_only : Flag<["-"], "cl-finite-math-only">, Group<opencl_Group>,
   Visibility<[ClangOption, CC1Option]>,
-  HelpText<"OpenCL only. Allow floating-point optimizations that assume arguments and results are not NaNs or +-Inf.">,
-  MarshallingInfoFlag<LangOpts<"CLFiniteMathOnly">>;
+  HelpText<"OpenCL only. Allow floating-point optimizations that assume arguments and results are not NaNs or +-Inf.">;
 def cl_kernel_arg_info : Flag<["-"], "cl-kernel-arg-info">, Group<opencl_Group>,
   Visibility<[ClangOption, CC1Option]>,
   HelpText<"OpenCL only. Generate kernel argument metadata.">,
@@ -2609,13 +2608,12 @@ defm approx_func : BoolFOption<"approx-func", LangOpts<"ApproxFunc">, DefaultFal
            "with an approximately equivalent calculation",
            [funsafe_math_optimizations.KeyPath]>,
    NegFlag<SetFalse, [], [ClangOption, CC1Option, FC1Option, FlangOption]>>;
-defm finite_math_only : BoolFOption<"finite-math-only",
-  LangOpts<"FiniteMathOnly">, DefaultFalse,
+defm finite_math_only : BoolOptionWithoutMarshalling<"f", "finite-math-only",
   PosFlag<SetTrue, [], [ClangOption, CC1Option],
           "Allow floating-point optimizations that "
           "assume arguments and results are not NaNs or +-inf. This defines "
           "the \\_\\_FINITE\\_MATH\\_ONLY\\_\\_ preprocessor macro.",
-          [cl_finite_math_only.KeyPath, ffast_math.KeyPath]>,
+          [ffast_math.KeyPath]>,
   NegFlag<SetFalse>>;
 defm signed_zeros : BoolFOption<"signed-zeros",
   LangOpts<"NoSignedZero">, DefaultFalse,
@@ -7815,10 +7813,10 @@ def mreassociate : Flag<["-"], "mreassociate">,
   MarshallingInfoFlag<LangOpts<"AllowFPReassoc">>, ImpliedByAnyOf<[funsafe_math_optimizations.KeyPath]>;
 def menable_no_nans : Flag<["-"], "menable-no-nans">,
   HelpText<"Allow optimization to assume there are no NaNs.">,
-  MarshallingInfoFlag<LangOpts<"NoHonorNaNs">>, ImpliedByAnyOf<[ffinite_math_only.KeyPath]>;
-def menable_no_infinities : Flag<["-"], "menable-no-infs">,
+  MarshallingInfoFlag<LangOpts<"NoHonorNaNs">>, ImpliedByAnyOf<[ffast_math.KeyPath]>;
+def menable_no_infs : Flag<["-"], "menable-no-infs">,
   HelpText<"Allow optimization to assume there are no infinities.">,
-  MarshallingInfoFlag<LangOpts<"NoHonorInfs">>, ImpliedByAnyOf<[ffinite_math_only.KeyPath]>;
+  MarshallingInfoFlag<LangOpts<"NoHonorInfs">>, ImpliedByAnyOf<[ffast_math.KeyPath]>;
 
 def pic_level : Separate<["-"], "pic-level">,
   HelpText<"Value for __PIC__">,

diff --git a/clang/include/clang/StaticAnalyzer/Checkers/Checkers.td b/clang/include/clang/StaticAnalyzer/Checkers/Checkers.td
@@ -1045,18 +1045,6 @@ def MallocOverflowSecurityChecker : Checker<"MallocOverflow">,
 
 def MmapWriteExecChecker : Checker<"MmapWriteExec">,
   HelpText<"Warn on mmap() calls that are both writable and executable">,
-  CheckerOptions<[
-    CmdLineOption<Integer,
-                  "MmapProtExec",
-                  "Specifies the value of PROT_EXEC",
-                  "0x04",
-                  Released>,
-    CmdLineOption<Integer,
-                  "MmapProtRead",
-                  "Specifies the value of PROT_READ",
-                  "0x01",
-                  Released>
-  ]>,
   Documentation<HasDocumentation>;
 
 def ReturnPointerRangeChecker : Checker<"ReturnPtrRange">,

diff --git a/clang/lib/AST/DeclBase.cpp b/clang/lib/AST/DeclBase.cpp
@@ -879,8 +879,6 @@ unsigned Decl::getIdentifierNamespaceForKind(Kind DeclKind) {
       return IDNS_Ordinary;
     case Label:
       return IDNS_Label;
-    case IndirectField:
-      return IDNS_Ordinary | IDNS_Member;
 
     case Binding:
     case NonTypeTemplateParm:
@@ -918,6 +916,7 @@ unsigned Decl::getIdentifierNamespaceForKind(Kind DeclKind) {
       return IDNS_ObjCProtocol;
 
     case Field:
+    case IndirectField:
     case ObjCAtDefsField:
     case ObjCIvar:
       return IDNS_Member;

diff --git a/clang/lib/AST/Interp/Disasm.cpp b/clang/lib/AST/Interp/Disasm.cpp
@@ -278,10 +278,15 @@ LLVM_DUMP_METHOD void InterpFrame::dump(llvm::raw_ostream &OS,
   OS << "\n";
   OS.indent(Spaces) << "This: " << getThis() << "\n";
   OS.indent(Spaces) << "RVO: " << getRVOPtr() << "\n";
-
-  while (const InterpFrame *F = this->Caller) {
+  OS.indent(Spaces) << "Depth: " << Depth << "\n";
+  OS.indent(Spaces) << "ArgSize: " << ArgSize << "\n";
+  OS.indent(Spaces) << "Args: " << (void *)Args << "\n";
+  OS.indent(Spaces) << "FrameOffset: " << FrameOffset << "\n";
+  OS.indent(Spaces) << "FrameSize: " << (Func ? Func->getFrameSize() : 0)
+                    << "\n";
+
+  for (const InterpFrame *F = this->Caller; F; F = F->Caller) {
     F->dump(OS, Indent + 1);
-    F = F->Caller;
   }
 }
 

diff --git a/clang/lib/AST/Interp/InterpBuiltin.cpp b/clang/lib/AST/Interp/InterpBuiltin.cpp
@@ -1170,27 +1170,30 @@ static bool interp__builtin_constant_p(InterpState &S, CodePtr OpPC,
       Stk.clear();
     }
 
-    const APValue &LV = Res.toAPValue();
-    if (!Res.isInvalid() && LV.isLValue()) {
-      APValue::LValueBase Base = LV.getLValueBase();
-      if (Base.isNull()) {
-        // A null base is acceptable.
-        return returnInt(true);
-      } else if (const auto *E = Base.dyn_cast<const Expr *>()) {
-        if (!isa<StringLiteral>(E))
+    if (!Res.isInvalid() && !Res.empty()) {
+      const APValue &LV = Res.toAPValue();
+      if (LV.isLValue()) {
+        APValue::LValueBase Base = LV.getLValueBase();
+        if (Base.isNull()) {
+          // A null base is acceptable.
+          return returnInt(true);
+        } else if (const auto *E = Base.dyn_cast<const Expr *>()) {
+          if (!isa<StringLiteral>(E))
+            return returnInt(false);
+          return returnInt(LV.getLValueOffset().isZero());
+        } else if (Base.is<TypeInfoLValue>()) {
+          // Surprisingly, GCC considers __builtin_constant_p(&typeid(int)) to
+          // evaluate to true.
+          return returnInt(true);
+        } else {
+          // Any other base is not constant enough for GCC.
           return returnInt(false);
-        return returnInt(LV.getLValueOffset().isZero());
-      } else if (Base.is<TypeInfoLValue>()) {
-        // Surprisingly, GCC considers __builtin_constant_p(&typeid(int)) to
-        // evaluate to true.
-        return returnInt(true);
-      } else {
-        // Any other base is not constant enough for GCC.
-        return returnInt(false);
+        }
       }
     }
 
-    return returnInt(!Res.isInvalid() && !Res.empty());
+    // Otherwise, any constant value is good enough.
+    return returnInt(true);
   }
 
   return returnInt(false);

diff --git a/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp b/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp
@@ -416,7 +416,7 @@ class ResultObjectVisitor : public AnalysisASTVisitor<ResultObjectVisitor> {
     // below them can initialize the same object (or part of it).
     if (isa<CXXConstructExpr>(E) || isa<CallExpr>(E) || isa<LambdaExpr>(E) ||
         isa<CXXDefaultArgExpr>(E) || isa<CXXStdInitializerListExpr>(E) ||
-        isa<AtomicExpr>(E) ||
+        isa<AtomicExpr>(E) || isa<CXXInheritedCtorInitExpr>(E) ||
         // We treat `BuiltinBitCastExpr` as an "original initializer" too as
         // it may not even be casting from a record type -- and even if it is,
         // the two objects are in general of unrelated type.

diff --git a/clang/lib/Basic/LangOptions.cpp b/clang/lib/Basic/LangOptions.cpp
@@ -34,8 +34,8 @@ void LangOptions::resetNonModularOptions() {
   // invocations that cannot be round-tripped to arguments.
   // FIXME: we should derive this automatically from ImpliedBy in tablegen.
   AllowFPReassoc = UnsafeFPMath;
-  NoHonorNaNs = FiniteMathOnly;
-  NoHonorInfs = FiniteMathOnly;
+  NoHonorInfs = FastMath;
+  NoHonorNaNs = FastMath;
 
   // These options do not affect AST generation.
   NoSanitizeFiles.clear();

diff --git a/clang/lib/Basic/Targets/OSTargets.cpp b/clang/lib/Basic/Targets/OSTargets.cpp
@@ -173,10 +173,10 @@ static void addVisualCDefines(const LangOptions &Opts, MacroBuilder &Builder) {
   // "Under /fp:precise and /fp:strict, the compiler doesn't do any mathematical
   // transformation unless the transformation is guaranteed to produce a bitwise
   // identical result."
-  const bool any_imprecise_flags =
-      Opts.FastMath || Opts.FiniteMathOnly || Opts.UnsafeFPMath ||
-      Opts.AllowFPReassoc || Opts.NoHonorNaNs || Opts.NoHonorInfs ||
-      Opts.NoSignedZero || Opts.AllowRecip || Opts.ApproxFunc;
+  const bool any_imprecise_flags = Opts.FastMath || Opts.UnsafeFPMath ||
+                                   Opts.AllowFPReassoc || Opts.NoHonorNaNs ||
+                                   Opts.NoHonorInfs || Opts.NoSignedZero ||
+                                   Opts.AllowRecip || Opts.ApproxFunc;
 
   // "Under both /fp:precise and /fp:fast, the compiler generates code intended
   // to run in the default floating-point environment."

diff --git a/clang/lib/CodeGen/CGAtomic.cpp b/clang/lib/CodeGen/CGAtomic.cpp
@@ -727,7 +727,7 @@ static void EmitAtomicOp(CodeGenFunction &CGF, AtomicExpr *E, Address Dest,
 
   llvm::Value *LoadVal1 = CGF.Builder.CreateLoad(Val1);
   llvm::AtomicRMWInst *RMWI =
-      CGF.Builder.CreateAtomicRMW(Op, Ptr, LoadVal1, Order, Scope);
+      CGF.emitAtomicRMWInst(Op, Ptr, LoadVal1, Order, Scope);
   RMWI->setVolatile(E->isVolatile());
 
   // For __atomic_*_fetch operations, perform the operation again to
@@ -2034,6 +2034,17 @@ std::pair<RValue, llvm::Value *> CodeGenFunction::EmitAtomicCompareExchange(
                                            IsWeak);
 }
 
+llvm::AtomicRMWInst *
+CodeGenFunction::emitAtomicRMWInst(llvm::AtomicRMWInst::BinOp Op, Address Addr,
+                                   llvm::Value *Val, llvm::AtomicOrdering Order,
+                                   llvm::SyncScope::ID SSID) {
+
+  llvm::AtomicRMWInst *RMW =
+      Builder.CreateAtomicRMW(Op, Addr, Val, Order, SSID);
+  getTargetHooks().setTargetAtomicMetadata(*this, *RMW);
+  return RMW;
+}
+
 void CodeGenFunction::EmitAtomicUpdate(
     LValue LVal, llvm::AtomicOrdering AO,
     const llvm::function_ref<RValue(RValue)> &UpdateOp, bool IsVolatile) {

diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -5986,8 +5986,6 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl GD, unsigned BuiltinID,
         getTarget().getTriple().isAMDGCN() ||
         (getTarget().getTriple().isSPIRV() &&
          getTarget().getTriple().getVendor() == Triple::VendorType::AMD)) {
-      if (getLangOpts().OpenMPIsTargetDevice)
-        return EmitOpenMPDevicePrintfCallExpr(E);
       if (getTarget().getTriple().isNVPTX())
         return EmitNVPTXDevicePrintfCallExpr(E);
       if ((getTarget().getTriple().isAMDGCN() ||

diff --git a/clang/lib/CodeGen/CGExprScalar.cpp b/clang/lib/CodeGen/CGExprScalar.cpp
@@ -2841,9 +2841,10 @@ ScalarExprEmitter::EmitScalarPrePostIncDec(const UnaryOperator *E, LValue LV,
           isInc ? llvm::Instruction::FAdd : llvm::Instruction::FSub;
       llvm::Value *amt = llvm::ConstantFP::get(
           VMContext, llvm::APFloat(static_cast<float>(1.0)));
-      llvm::Value *old =
-          Builder.CreateAtomicRMW(aop, LV.getAddress(), amt,
-                                  llvm::AtomicOrdering::SequentiallyConsistent);
+      llvm::AtomicRMWInst *old =
+          CGF.emitAtomicRMWInst(aop, LV.getAddress(), amt,
+                                llvm::AtomicOrdering::SequentiallyConsistent);
+
       return isPre ? Builder.CreateBinOp(op, old, amt) : old;
     }
     value = EmitLoadOfLValue(LV, E->getExprLoc());
@@ -3583,9 +3584,9 @@ LValue ScalarExprEmitter::EmitCompoundAssignLValue(
             EmitScalarConversion(OpInfo.RHS, E->getRHS()->getType(), LHSTy,
                                  E->getExprLoc()),
             LHSTy);
-        Value *OldVal = Builder.CreateAtomicRMW(
-            AtomicOp, LHSLV.getAddress(), Amt,
-            llvm::AtomicOrdering::SequentiallyConsistent);
+
+        llvm::AtomicRMWInst *OldVal =
+            CGF.emitAtomicRMWInst(AtomicOp, LHSLV.getAddress(), Amt);
 
         // Since operation is atomic, the result type is guaranteed to be the
         // same as the input in LLVM terms.

diff --git a/clang/lib/CodeGen/CGGPUBuiltin.cpp b/clang/lib/CodeGen/CGGPUBuiltin.cpp
@@ -42,28 +42,6 @@ llvm::Function *GetVprintfDeclaration(llvm::Module &M) {
       VprintfFuncType, llvm::GlobalVariable::ExternalLinkage, "vprintf", &M);
 }
 
-llvm::Function *GetOpenMPVprintfDeclaration(CodeGenModule &CGM) {
-  const char *Name = "__llvm_omp_vprintf";
-  llvm::Module &M = CGM.getModule();
-  llvm::Type *ArgTypes[] = {llvm::PointerType::getUnqual(M.getContext()),
-                            llvm::PointerType::getUnqual(M.getContext()),
-                            llvm::Type::getInt32Ty(M.getContext())};
-  llvm::FunctionType *VprintfFuncType = llvm::FunctionType::get(
-      llvm::Type::getInt32Ty(M.getContext()), ArgTypes, false);
-
-  if (auto *F = M.getFunction(Name)) {
-    if (F->getFunctionType() != VprintfFuncType) {
-      CGM.Error(SourceLocation(),
-                "Invalid type declaration for __llvm_omp_vprintf");
-      return nullptr;
-    }
-    return F;
-  }
-
-  return llvm::Function::Create(
-      VprintfFuncType, llvm::GlobalVariable::ExternalLinkage, Name, &M);
-}
-
 // Transforms a call to printf into a call to the NVPTX vprintf syscall (which
 // isn't particularly special; it's invoked just like a regular function).
 // vprintf takes two args: A format string, and a pointer to a buffer containing
@@ -213,10 +191,3 @@ RValue CodeGenFunction::EmitAMDGPUDevicePrintfCallExpr(const CallExpr *E) {
   Builder.SetInsertPoint(IRB.GetInsertBlock(), IRB.GetInsertPoint());
   return RValue::get(Printf);
 }
-
-RValue CodeGenFunction::EmitOpenMPDevicePrintfCallExpr(const CallExpr *E) {
-  assert(getTarget().getTriple().isNVPTX() ||
-         getTarget().getTriple().isAMDGCN());
-  return EmitDevicePrintfCallExpr(E, this, GetOpenMPVprintfDeclaration(CGM),
-                                  true);
-}
diff --git a/clang/lib/CodeGen/CGStmtOpenMP.cpp b/clang/lib/CodeGen/CGStmtOpenMP.cpp
@@ -6326,8 +6326,8 @@ static std::pair<bool, RValue> emitOMPAtomicRMW(CodeGenFunction &CGF, LValue X,
       UpdateVal = CGF.Builder.CreateCast(llvm::Instruction::CastOps::UIToFP, IC,
                                          X.getAddress().getElementType());
   }
-  llvm::Value *Res =
-      CGF.Builder.CreateAtomicRMW(RMWOp, X.getAddress(), UpdateVal, AO);
+  llvm::AtomicRMWInst *Res =
+      CGF.emitAtomicRMWInst(RMWOp, X.getAddress(), UpdateVal, AO);
   return std::make_pair(true, RValue::get(Res));
 }
 

diff --git a/clang/lib/CodeGen/CodeGenFunction.cpp b/clang/lib/CodeGen/CodeGenFunction.cpp
@@ -991,6 +991,9 @@ void CodeGenFunction::StartFunction(GlobalDecl GD, QualType RetTy,
   if (D && D->hasAttr<NoProfileFunctionAttr>())
     Fn->addFnAttr(llvm::Attribute::NoProfile);
 
+  if (D && D->hasAttr<HybridPatchableAttr>())
+    Fn->addFnAttr(llvm::Attribute::HybridPatchable);
+
   if (D) {
     // Function attributes take precedence over command line flags.
     if (auto *A = D->getAttr<FunctionReturnThunksAttr>()) {

diff --git a/clang/lib/CodeGen/CodeGenFunction.h b/clang/lib/CodeGen/CodeGenFunction.h
@@ -4160,6 +4160,13 @@ class CodeGenFunction : public CodeGenTypeCache {
           llvm::AtomicOrdering::SequentiallyConsistent,
       bool IsWeak = false, AggValueSlot Slot = AggValueSlot::ignored());
 
+  /// Emit an atomicrmw instruction, and applying relevant metadata when
+  /// applicable.
+  llvm::AtomicRMWInst *emitAtomicRMWInst(
+      llvm::AtomicRMWInst::BinOp Op, Address Addr, llvm::Value *Val,
+      llvm::AtomicOrdering Order = llvm::AtomicOrdering::SequentiallyConsistent,
+      llvm::SyncScope::ID SSID = llvm::SyncScope::System);
+
   void EmitAtomicUpdate(LValue LVal, llvm::AtomicOrdering AO,
                         const llvm::function_ref<RValue(RValue)> &UpdateOp,
                         bool IsVolatile);
@@ -4529,7 +4536,6 @@ class CodeGenFunction : public CodeGenTypeCache {
 
   RValue EmitNVPTXDevicePrintfCallExpr(const CallExpr *E);
   RValue EmitAMDGPUDevicePrintfCallExpr(const CallExpr *E);
-  RValue EmitOpenMPDevicePrintfCallExpr(const CallExpr *E);
 
   RValue EmitBuiltinExpr(const GlobalDecl GD, unsigned BuiltinID,
                          const CallExpr *E, ReturnValueSlot ReturnValue);

diff --git a/clang/lib/CodeGen/TargetInfo.h b/clang/lib/CodeGen/TargetInfo.h
@@ -334,6 +334,10 @@ class TargetCodeGenInfo {
                                                  llvm::AtomicOrdering Ordering,
                                                  llvm::LLVMContext &Ctx) const;
 
+  /// Allow the target to apply other metadata to an atomic instruction
+  virtual void setTargetAtomicMetadata(CodeGenFunction &CGF,
+                                       llvm::AtomicRMWInst &RMW) const {}
+
   /// Interface class for filling custom fields of a block literal for OpenCL.
   class TargetOpenCLBlockHelper {
   public:

diff --git a/clang/lib/CodeGen/Targets/AMDGPU.cpp b/clang/lib/CodeGen/Targets/AMDGPU.cpp
@@ -311,6 +311,8 @@ class AMDGPUTargetCodeGenInfo : public TargetCodeGenInfo {
                                          SyncScope Scope,
                                          llvm::AtomicOrdering Ordering,
                                          llvm::LLVMContext &Ctx) const override;
+  void setTargetAtomicMetadata(CodeGenFunction &CGF,
+                               llvm::AtomicRMWInst &RMW) const override;
   llvm::Value *createEnqueuedBlockKernel(CodeGenFunction &CGF,
                                          llvm::Function *BlockInvokeFunc,
                                          llvm::Type *BlockTy) const override;
@@ -546,6 +548,23 @@ AMDGPUTargetCodeGenInfo::getLLVMSyncScopeID(const LangOptions &LangOpts,
   return Ctx.getOrInsertSyncScopeID(Name);
 }
 
+void AMDGPUTargetCodeGenInfo::setTargetAtomicMetadata(
+    CodeGenFunction &CGF, llvm::AtomicRMWInst &RMW) const {
+  if (!CGF.getTarget().allowAMDGPUUnsafeFPAtomics())
+    return;
+
+  // TODO: Introduce new, more controlled options that also work for integers,
+  // and deprecate allowAMDGPUUnsafeFPAtomics.
+  llvm::AtomicRMWInst::BinOp RMWOp = RMW.getOperation();
+  if (llvm::AtomicRMWInst::isFPOperation(RMWOp)) {
+    llvm::MDNode *Empty = llvm::MDNode::get(CGF.getLLVMContext(), {});
+    RMW.setMetadata("amdgpu.no.fine.grained.memory", Empty);
+
+    if (RMWOp == llvm::AtomicRMWInst::FAdd && RMW.getType()->isFloatTy())
+      RMW.setMetadata("amdgpu.ignore.denormal.mode", Empty);
+  }
+}
+
 bool AMDGPUTargetCodeGenInfo::shouldEmitStaticExternCAliases() const {
   return false;
 }

diff --git a/clang/lib/Driver/ToolChains/AMDGPU.cpp b/clang/lib/Driver/ToolChains/AMDGPU.cpp
@@ -620,8 +620,10 @@ void amdgpu::Linker::ConstructJob(Compilation &C, const JobAction &JA,
                                   const char *LinkingOutput) const {
   std::string Linker = getToolChain().GetLinkerPath();
   ArgStringList CmdArgs;
-  CmdArgs.push_back("--no-undefined");
-  CmdArgs.push_back("-shared");
+  if (!Args.hasArg(options::OPT_r)) {
+    CmdArgs.push_back("--no-undefined");
+    CmdArgs.push_back("-shared");
+  }
 
   addLinkerCompressDebugSectionsOption(getToolChain(), Args, CmdArgs);
   Args.AddAllArgs(CmdArgs, options::OPT_L);

diff --git a/clang/lib/Driver/ToolChains/Clang.cpp b/clang/lib/Driver/ToolChains/Clang.cpp
@@ -3384,9 +3384,28 @@ static void RenderFloatingPointOptions(const ToolChain &TC, const Driver &D,
     CmdArgs.push_back("-ffast-math");
 
   // Handle __FINITE_MATH_ONLY__ similarly.
-  if (!HonorINFs && !HonorNaNs)
+  // The -ffinite-math-only is added to CmdArgs when !HonorINFs && !HonorNaNs.
+  // Otherwise process the Xclang arguments to determine if -menable-no-infs and
+  // -menable-no-nans are set by the user.
+  bool shouldAddFiniteMathOnly = false;
+  if (!HonorINFs && !HonorNaNs) {
+    shouldAddFiniteMathOnly = true;
+  } else {
+    bool InfValues = true;
+    bool NanValues = true;
+    for (const auto *Arg : Args.filtered(options::OPT_Xclang)) {
+      StringRef ArgValue = Arg->getValue();
+      if (ArgValue == "-menable-no-nans")
+        NanValues = false;
+      else if (ArgValue == "-menable-no-infs")
+        InfValues = false;
+    }
+    if (!NanValues && !InfValues)
+      shouldAddFiniteMathOnly = true;
+  }
+  if (shouldAddFiniteMathOnly) {
     CmdArgs.push_back("-ffinite-math-only");
-
+  }
   if (const Arg *A = Args.getLastArg(options::OPT_mfpmath_EQ)) {
     CmdArgs.push_back("-mfpmath");
     CmdArgs.push_back(A->getValue());
@@ -3755,6 +3774,11 @@ static void RenderOpenCLOptions(const ArgList &Args, ArgStringList &CmdArgs,
     CmdArgs.push_back(Args.MakeArgString(CLExtStr));
   }
 
+  if (Args.hasArg(options::OPT_cl_finite_math_only)) {
+    CmdArgs.push_back("-menable-no-infs");
+    CmdArgs.push_back("-menable-no-nans");
+  }
+
   for (const auto &Arg : ForwardedArguments)
     if (const auto *A = Args.getLastArg(Arg))
       CmdArgs.push_back(Args.MakeArgString(A->getOption().getPrefixedName()));

diff --git a/clang/lib/Driver/ToolChains/PS4CPU.cpp b/clang/lib/Driver/ToolChains/PS4CPU.cpp
@@ -266,6 +266,10 @@ void tools::PS5cpu::Linker::ConstructJob(Compilation &C, const JobAction &JA,
   if (UseJMC)
     AddLTOFlag("-enable-jmc-instrument");
 
+  if (Args.hasFlag(options::OPT_fstack_size_section,
+                   options::OPT_fno_stack_size_section, false))
+    AddLTOFlag("-stack-size-section");
+
   if (Arg *A = Args.getLastArg(options::OPT_fcrash_diagnostics_dir))
     AddLTOFlag(Twine("-crash-diagnostics-dir=") + A->getValue());
 

diff --git a/clang/lib/Frontend/InitPreprocessor.cpp b/clang/lib/Frontend/InitPreprocessor.cpp
@@ -1318,7 +1318,7 @@ static void InitializePredefinedMacros(const TargetInfo &TI,
   if (!LangOpts.MathErrno)
     Builder.defineMacro("__NO_MATH_ERRNO__");
 
-  if (LangOpts.FastMath || LangOpts.FiniteMathOnly)
+  if (LangOpts.FastMath || (LangOpts.NoHonorInfs && LangOpts.NoHonorNaNs))
     Builder.defineMacro("__FINITE_MATH_ONLY__", "1");
   else
     Builder.defineMacro("__FINITE_MATH_ONLY__", "0");

diff --git a/clang/lib/Frontend/PrintPreprocessedOutput.cpp b/clang/lib/Frontend/PrintPreprocessedOutput.cpp
@@ -980,7 +980,6 @@ static void PrintPreprocessedTokens(Preprocessor &PP, Token &Tok,
         *Callbacks->OS << static_cast<unsigned>(*Iter);
         PrintComma = true;
       }
-      IsStartOfLine = true;
     } else if (Tok.isAnnotation()) {
       // Ignore annotation tokens created by pragmas - the pragmas themselves
       // will be reproduced in the preprocessed output.

diff --git a/clang/lib/Lex/PPMacroExpansion.cpp b/clang/lib/Lex/PPMacroExpansion.cpp
@@ -1604,6 +1604,16 @@ static bool isTargetVariantEnvironment(const TargetInfo &TI,
   return false;
 }
 
+#if defined(__sun__) && defined(__svr4__)
+// GCC mangles std::tm as tm for binary compatibility on Solaris (Issue
+// #33114).  We need to match this to allow the std::put_time calls to link
+// (PR #99075).
+asm("_ZNKSt8time_putIcSt19ostreambuf_iteratorIcSt11char_traitsIcEEE3putES3_"
+    "RSt8ios_basecPKSt2tmPKcSB_ = "
+    "_ZNKSt8time_putIcSt19ostreambuf_iteratorIcSt11char_traitsIcEEE3putES3_"
+    "RSt8ios_basecPK2tmPKcSB_");
+#endif
+
 /// ExpandBuiltinMacro - If an identifier token is read that is to be expanded
 /// as a builtin macro, handle it and return the next token as 'Tok'.
 void Preprocessor::ExpandBuiltinMacro(Token &Tok) {

diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp
@@ -1477,6 +1477,18 @@ static bool BuiltinSEHScopeCheck(Sema &SemaRef, CallExpr *TheCall,
   return false;
 }
 
+// In OpenCL, __builtin_alloca_* should return a pointer to address space
+// that corresponds to the stack address space i.e private address space.
+static void builtinAllocaAddrSpace(Sema &S, CallExpr *TheCall) {
+  QualType RT = TheCall->getType();
+  assert((RT->isPointerType() && !(RT->getPointeeType().hasAddressSpace())) &&
+         "__builtin_alloca has invalid address space");
+
+  RT = RT->getPointeeType();
+  RT = S.Context.getAddrSpaceQualType(RT, LangAS::opencl_private);
+  TheCall->setType(S.Context.getPointerType(RT));
+}
+
 namespace {
 enum PointerAuthOpKind {
   PAO_Strip,
@@ -2214,6 +2226,9 @@ Sema::CheckBuiltinFunctionCall(FunctionDecl *FDecl, unsigned BuiltinID,
   case Builtin::BI__builtin_alloca_uninitialized:
     Diag(TheCall->getBeginLoc(), diag::warn_alloca)
         << TheCall->getDirectCallee();
+    if (getLangOpts().OpenCL) {
+      builtinAllocaAddrSpace(*this, TheCall);
+    }
     break;
   case Builtin::BI__arithmetic_fence:
     if (BuiltinArithmeticFence(TheCall))

diff --git a/clang/lib/Sema/SemaDecl.cpp b/clang/lib/Sema/SemaDecl.cpp
@@ -6890,6 +6890,11 @@ static void checkAttributesAfterMerging(Sema &S, NamedDecl &ND) {
     }
   }
 
+  if (HybridPatchableAttr *Attr = ND.getAttr<HybridPatchableAttr>()) {
+    if (!ND.isExternallyVisible())
+      S.Diag(Attr->getLocation(),
+             diag::warn_attribute_hybrid_patchable_non_extern);
+  }
   if (const InheritableAttr *Attr = getDLLAttr(&ND)) {
     auto *VD = dyn_cast<VarDecl>(&ND);
     bool IsAnonymousNS = false;
@@ -11009,6 +11014,9 @@ static bool AttrCompatibleWithMultiVersion(attr::Kind Kind,
   switch (Kind) {
   default:
     return false;
+  case attr::ArmLocallyStreaming:
+    return MVKind == MultiVersionKind::TargetVersion ||
+           MVKind == MultiVersionKind::TargetClones;
   case attr::Used:
     return MVKind == MultiVersionKind::Target;
   case attr::NonNull:
@@ -11145,7 +11153,21 @@ bool Sema::areMultiversionVariantFunctionsCompatible(
     FunctionType::ExtInfo OldTypeInfo = OldType->getExtInfo();
     FunctionType::ExtInfo NewTypeInfo = NewType->getExtInfo();
 
-    if (OldTypeInfo.getCC() != NewTypeInfo.getCC())
+    const auto *OldFPT = OldFD->getType()->getAs<FunctionProtoType>();
+    const auto *NewFPT = NewFD->getType()->getAs<FunctionProtoType>();
+
+    bool ArmStreamingCCMismatched = false;
+    if (OldFPT && NewFPT) {
+      unsigned Diff =
+          OldFPT->getAArch64SMEAttributes() ^ NewFPT->getAArch64SMEAttributes();
+      // Arm-streaming, arm-streaming-compatible and non-streaming versions
+      // cannot be mixed.
+      if (Diff & (FunctionType::SME_PStateSMEnabledMask |
+                  FunctionType::SME_PStateSMCompatibleMask))
+        ArmStreamingCCMismatched = true;
+    }
+
+    if (OldTypeInfo.getCC() != NewTypeInfo.getCC() || ArmStreamingCCMismatched)
       return Diag(DiffDiagIDAt.first, DiffDiagIDAt.second) << CallingConv;
 
     QualType OldReturnType = OldType->getReturnType();
@@ -11165,9 +11187,8 @@ bool Sema::areMultiversionVariantFunctionsCompatible(
     if (!CLinkageMayDiffer && OldFD->isExternC() != NewFD->isExternC())
       return Diag(DiffDiagIDAt.first, DiffDiagIDAt.second) << LanguageLinkage;
 
-    if (CheckEquivalentExceptionSpec(
-            OldFD->getType()->getAs<FunctionProtoType>(), OldFD->getLocation(),
-            NewFD->getType()->getAs<FunctionProtoType>(), NewFD->getLocation()))
+    if (CheckEquivalentExceptionSpec(OldFPT, OldFD->getLocation(), NewFPT,
+                                     NewFD->getLocation()))
       return true;
   }
   return false;

diff --git a/clang/lib/Sema/SemaDeclAttr.cpp b/clang/lib/Sema/SemaDeclAttr.cpp
@@ -3034,9 +3034,6 @@ bool Sema::checkTargetVersionAttr(SourceLocation LiteralLoc, Decl *D,
       return Diag(LiteralLoc, diag::warn_unsupported_target_attribute)
              << Unsupported << None << CurFeature << TargetVersion;
   }
-  if (IsArmStreamingFunction(cast<FunctionDecl>(D),
-                             /*IncludeLocallyStreaming=*/false))
-    return Diag(LiteralLoc, diag::err_sme_streaming_cannot_be_multiversioned);
   return false;
 }
 
@@ -3133,10 +3130,6 @@ bool Sema::checkTargetClonesAttrString(
           HasNotDefault = true;
         }
       }
-      if (IsArmStreamingFunction(cast<FunctionDecl>(D),
-                                 /*IncludeLocallyStreaming=*/false))
-        return Diag(LiteralLoc,
-                    diag::err_sme_streaming_cannot_be_multiversioned);
     } else {
       // Other targets ( currently X86 )
       if (Cur.starts_with("arch=")) {
@@ -6878,6 +6871,9 @@ ProcessDeclAttribute(Sema &S, Scope *scope, Decl *D, const ParsedAttr &AL,
   case ParsedAttr::AT_MSConstexpr:
     handleMSConstexprAttr(S, D, AL);
     break;
+  case ParsedAttr::AT_HybridPatchable:
+    handleSimpleAttribute<HybridPatchableAttr>(S, D, AL);
+    break;
 
   // HLSL attributes:
   case ParsedAttr::AT_HLSLNumThreads:

diff --git a/clang/lib/Sema/SemaDeclCXX.cpp b/clang/lib/Sema/SemaDeclCXX.cpp
@@ -10385,7 +10385,7 @@ void Sema::checkIncorrectVTablePointerAuthenticationAttribute(
   while (1) {
     assert(PrimaryBase);
     const CXXRecordDecl *Base = nullptr;
-    for (auto BasePtr : PrimaryBase->bases()) {
+    for (const CXXBaseSpecifier &BasePtr : PrimaryBase->bases()) {
       if (!BasePtr.getType()->getAsCXXRecordDecl()->isDynamicClass())
         continue;
       Base = BasePtr.getType()->getAsCXXRecordDecl();

diff --git a/clang/lib/Sema/SemaExprCXX.cpp b/clang/lib/Sema/SemaExprCXX.cpp
@@ -6030,6 +6030,32 @@ static bool EvaluateBinaryTypeTrait(Sema &Self, TypeTrait BTT, const TypeSourceI
     return cast<CXXRecordDecl>(rhsRecord->getDecl())
       ->isDerivedFrom(cast<CXXRecordDecl>(lhsRecord->getDecl()));
   }
+  case BTT_IsVirtualBaseOf: {
+    const RecordType *BaseRecord = LhsT->getAs<RecordType>();
+    const RecordType *DerivedRecord = RhsT->getAs<RecordType>();
+
+    if (!BaseRecord || !DerivedRecord) {
+      DiagnoseVLAInCXXTypeTrait(Self, Lhs,
+                                tok::kw___builtin_is_virtual_base_of);
+      DiagnoseVLAInCXXTypeTrait(Self, Rhs,
+                                tok::kw___builtin_is_virtual_base_of);
+      return false;
+    }
+
+    if (BaseRecord->isUnionType() || DerivedRecord->isUnionType())
+      return false;
+
+    if (!BaseRecord->isStructureOrClassType() ||
+        !DerivedRecord->isStructureOrClassType())
+      return false;
+
+    if (Self.RequireCompleteType(Rhs->getTypeLoc().getBeginLoc(), RhsT,
+                                 diag::err_incomplete_type))
+      return false;
+
+    return cast<CXXRecordDecl>(DerivedRecord->getDecl())
+        ->isVirtuallyDerivedFrom(cast<CXXRecordDecl>(BaseRecord->getDecl()));
+  }
   case BTT_IsSame:
     return Self.Context.hasSameType(LhsT, RhsT);
   case BTT_TypeCompatible: {

diff --git a/clang/lib/Sema/SemaPPC.cpp b/clang/lib/Sema/SemaPPC.cpp
@@ -93,7 +93,6 @@ bool SemaPPC::CheckPPCBuiltinFunctionCall(const TargetInfo &TI,
                                           unsigned BuiltinID,
                                           CallExpr *TheCall) {
   ASTContext &Context = getASTContext();
-  unsigned i = 0, l = 0, u = 0;
   bool IsTarget64Bit = TI.getTypeWidth(TI.getIntPtrType()) == 64;
   llvm::APSInt Result;
 
@@ -248,7 +247,7 @@ bool SemaPPC::CheckPPCBuiltinFunctionCall(const TargetInfo &TI,
     return BuiltinPPCMMACall(TheCall, BuiltinID, Types);
 #include "clang/Basic/BuiltinsPPC.def"
   }
-  return SemaRef.BuiltinConstantArgRange(TheCall, i, l, u);
+  llvm_unreachable("must return from switch");
 }
 
 // Check if the given type is a non-pointer PPC MMA type. This function is used

diff --git a/clang/lib/Sema/SemaTemplateDeduction.cpp b/clang/lib/Sema/SemaTemplateDeduction.cpp
@@ -951,9 +951,11 @@ class PackDeductionScope {
 
     // Skip over the pack elements that were expanded into separate arguments.
     // If we partially expanded, this is the number of partial arguments.
+    // FIXME: `&& FixedNumExpansions` is a workaround for UB described in
+    // https://github.com/llvm/llvm-project/issues/100095
     if (IsPartiallyExpanded)
       PackElements += NumPartialPackArgs;
-    else if (IsExpanded)
+    else if (IsExpanded && FixedNumExpansions)
       PackElements += *FixedNumExpansions;
 
     for (auto &Pack : Packs) {

diff --git a/clang/lib/StaticAnalyzer/Checkers/MmapWriteExecChecker.cpp b/clang/lib/StaticAnalyzer/Checkers/MmapWriteExecChecker.cpp
@@ -21,49 +21,56 @@
 #include "clang/StaticAnalyzer/Core/PathSensitive/CallDescription.h"
 #include "clang/StaticAnalyzer/Core/PathSensitive/CallEvent.h"
 #include "clang/StaticAnalyzer/Core/PathSensitive/CheckerContext.h"
+#include "clang/StaticAnalyzer/Core/PathSensitive/CheckerHelpers.h"
 
 using namespace clang;
 using namespace ento;
 
 namespace {
-class MmapWriteExecChecker : public Checker<check::PreCall> {
+class MmapWriteExecChecker
+    : public Checker<check::ASTDecl<TranslationUnitDecl>, check::PreCall> {
   CallDescription MmapFn{CDM::CLibrary, {"mmap"}, 6};
   CallDescription MprotectFn{CDM::CLibrary, {"mprotect"}, 3};
-  static int ProtWrite;
-  static int ProtExec;
-  static int ProtRead;
   const BugType BT{this, "W^X check fails, Write Exec prot flags set",
                    "Security"};
 
+  // Default values are used if definition of the flags is not found.
+  mutable int ProtRead = 0x01;
+  mutable int ProtWrite = 0x02;
+  mutable int ProtExec = 0x04;
+
 public:
+  void checkASTDecl(const TranslationUnitDecl *TU, AnalysisManager &Mgr,
+                    BugReporter &BR) const;
   void checkPreCall(const CallEvent &Call, CheckerContext &C) const;
-  int ProtExecOv;
-  int ProtReadOv;
 };
 }
 
-int MmapWriteExecChecker::ProtWrite = 0x02;
-int MmapWriteExecChecker::ProtExec  = 0x04;
-int MmapWriteExecChecker::ProtRead  = 0x01;
+void MmapWriteExecChecker::checkASTDecl(const TranslationUnitDecl *TU,
+                                        AnalysisManager &Mgr,
+                                        BugReporter &BR) const {
+  Preprocessor &PP = Mgr.getPreprocessor();
+  const std::optional<int> FoundProtRead = tryExpandAsInteger("PROT_READ", PP);
+  const std::optional<int> FoundProtWrite =
+      tryExpandAsInteger("PROT_WRITE", PP);
+  const std::optional<int> FoundProtExec = tryExpandAsInteger("PROT_EXEC", PP);
+  if (FoundProtRead && FoundProtWrite && FoundProtExec) {
+    ProtRead = *FoundProtRead;
+    ProtWrite = *FoundProtWrite;
+    ProtExec = *FoundProtExec;
+  }
+}
 
 void MmapWriteExecChecker::checkPreCall(const CallEvent &Call,
-                                         CheckerContext &C) const {
+                                        CheckerContext &C) const {
   if (matchesAny(Call, MmapFn, MprotectFn)) {
     SVal ProtVal = Call.getArgSVal(2);
     auto ProtLoc = ProtVal.getAs<nonloc::ConcreteInt>();
     if (!ProtLoc)
       return;
     int64_t Prot = ProtLoc->getValue().getSExtValue();
-    if (ProtExecOv != ProtExec)
-      ProtExec = ProtExecOv;
-    if (ProtReadOv != ProtRead)
-      ProtRead = ProtReadOv;
-
-    // Wrong settings
-    if (ProtRead == ProtExec)
-      return;
 
-    if ((Prot & (ProtWrite | ProtExec)) == (ProtWrite | ProtExec)) {
+    if ((Prot & ProtWrite) && (Prot & ProtExec)) {
       ExplodedNode *N = C.generateNonFatalErrorNode();
       if (!N)
         return;
@@ -80,17 +87,10 @@ void MmapWriteExecChecker::checkPreCall(const CallEvent &Call,
   }
 }
 
-void ento::registerMmapWriteExecChecker(CheckerManager &mgr) {
-  MmapWriteExecChecker *Mwec =
-      mgr.registerChecker<MmapWriteExecChecker>();
-  Mwec->ProtExecOv =
-    mgr.getAnalyzerOptions()
-      .getCheckerIntegerOption(Mwec, "MmapProtExec");
-  Mwec->ProtReadOv =
-    mgr.getAnalyzerOptions()
-      .getCheckerIntegerOption(Mwec, "MmapProtRead");
+void ento::registerMmapWriteExecChecker(CheckerManager &Mgr) {
+  Mgr.registerChecker<MmapWriteExecChecker>();
 }
 
-bool ento::shouldRegisterMmapWriteExecChecker(const CheckerManager &mgr) {
+bool ento::shouldRegisterMmapWriteExecChecker(const CheckerManager &) {
   return true;
 }
diff --git a/clang/test/AST/Interp/builtins.cpp b/clang/test/AST/Interp/builtins.cpp
@@ -31,3 +31,8 @@ constexpr bool assume() {
   return true;
 }
 static_assert(assume(), "");
+
+void test_builtin_os_log(void *buf, int i, const char *data) {
+  constexpr int len = __builtin_os_log_format_buffer_size("%d %{public}s %{private}.16P", i, data, data);
+  static_assert(len > 0, "Expect len > 0");
+}
diff --git a/clang/test/Analysis/analyzer-config.c b/clang/test/Analysis/analyzer-config.c
@@ -9,8 +9,6 @@
 // CHECK-NEXT: alpha.clone.CloneChecker:ReportNormalClones = true
 // CHECK-NEXT: alpha.cplusplus.STLAlgorithmModeling:AggressiveStdFindModeling = false
 // CHECK-NEXT: alpha.osx.cocoa.DirectIvarAssignment:AnnotatedFunctions = false
-// CHECK-NEXT: alpha.security.MmapWriteExec:MmapProtExec = 0x04
-// CHECK-NEXT: alpha.security.MmapWriteExec:MmapProtRead = 0x01
 // CHECK-NEXT: alpha.security.taint.TaintPropagation:Config = ""
 // CHECK-NEXT: apply-fixits = false
 // CHECK-NEXT: assume-controlled-environment = false

diff --git a/clang/test/Analysis/mmap-writeexec.c b/clang/test/Analysis/mmap-writeexec.c
@@ -1,13 +1,14 @@
-// RUN: %clang_analyze_cc1 -triple i686-unknown-linux -analyzer-checker=alpha.security.MmapWriteExec -analyzer-config alpha.security.MmapWriteExec:MmapProtExec=1 -analyzer-config alpha.security.MmapWriteExec:MmapProtRead=4 -DUSE_ALTERNATIVE_PROT_EXEC_DEFINITION -verify %s
+// RUN: %clang_analyze_cc1 -triple i686-unknown-linux -analyzer-checker=alpha.security.MmapWriteExec -DUSE_ALTERNATIVE_PROT_EXEC_DEFINITION -verify %s
 // RUN: %clang_analyze_cc1 -triple x86_64-unknown-apple-darwin10 -analyzer-checker=alpha.security.MmapWriteExec -verify %s
 
-#define PROT_WRITE  0x02
 #ifndef USE_ALTERNATIVE_PROT_EXEC_DEFINITION
-#define PROT_EXEC   0x04
-#define PROT_READ   0x01
-#else
 #define PROT_EXEC   0x01
+#define PROT_WRITE  0x02
 #define PROT_READ   0x04
+#else
+#define PROT_EXEC   0x08
+#define PROT_WRITE  0x04
+#define PROT_READ   0x02
 #endif
 #define MAP_PRIVATE 0x0002
 #define MAP_ANON    0x1000

diff --git a/clang/test/CodeGen/AMDGPU/amdgpu-atomic-float.c b/clang/test/CodeGen/AMDGPU/amdgpu-atomic-float.c
diff --git a/clang/test/CodeGen/aarch64-fmv-streaming.c b/clang/test/CodeGen/aarch64-fmv-streaming.c
@@ -0,0 +1,107 @@
+// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sme -emit-llvm -o - %s | FileCheck %s
+
+
+// CHECK-LABEL: define {{[^@]+}}@n_callee._Msve
+// CHECK-SAME: () #[[ATTR0:[0-9]+]] {
+//
+// CHECK-LABEL: define {{[^@]+}}@n_callee._Msimd
+// CHECK-SAME: () #[[ATTR1:[0-9]+]] {
+//
+__arm_locally_streaming __attribute__((target_clones("sve", "simd"))) void n_callee(void) {}
+// CHECK-LABEL: define {{[^@]+}}@n_callee._Msme2
+// CHECK-SAME: () #[[ATTR2:[0-9]+]] {
+//
+__attribute__((target_version("sme2"))) void n_callee(void) {}
+// CHECK-LABEL: define {{[^@]+}}@n_callee.default
+// CHECK-SAME: () #[[ATTR3:[0-9]+]] {
+//
+__attribute__((target_version("default"))) void n_callee(void) {}
+
+
+// CHECK-LABEL: define {{[^@]+}}@s_callee._Msve
+// CHECK-SAME: () #[[ATTR4:[0-9]+]] {
+//
+// CHECK-LABEL: define {{[^@]+}}@s_callee._Msimd
+// CHECK-SAME: () #[[ATTR5:[0-9]+]] {
+//
+__attribute__((target_clones("sve", "simd"))) void s_callee(void) __arm_streaming {}
+// CHECK-LABEL: define {{[^@]+}}@s_callee._Msme2
+// CHECK-SAME: () #[[ATTR6:[0-9]+]] {
+//
+__arm_locally_streaming __attribute__((target_version("sme2"))) void s_callee(void) __arm_streaming {}
+// CHECK-LABEL: define {{[^@]+}}@s_callee.default
+// CHECK-SAME: () #[[ATTR7:[0-9]+]] {
+//
+__attribute__((target_version("default"))) void s_callee(void) __arm_streaming {}
+
+
+// CHECK-LABEL: define {{[^@]+}}@sc_callee._Msve
+// CHECK-SAME: () #[[ATTR8:[0-9]+]] {
+//
+// CHECK-LABEL: define {{[^@]+}}@sc_callee._Msimd
+// CHECK-SAME: () #[[ATTR9:[0-9]+]] {
+//
+__attribute__((target_clones("sve", "simd"))) void sc_callee(void) __arm_streaming_compatible {}
+// CHECK-LABEL: define {{[^@]+}}@sc_callee._Msme2
+// CHECK-SAME: () #[[ATTR10:[0-9]+]] {
+//
+__arm_locally_streaming __attribute__((target_version("sme2"))) void sc_callee(void) __arm_streaming_compatible {}
+// CHECK-LABEL: define {{[^@]+}}@sc_callee.default
+// CHECK-SAME: () #[[ATTR11:[0-9]+]] {
+//
+__attribute__((target_version("default"))) void sc_callee(void) __arm_streaming_compatible {}
+
+
+// CHECK-LABEL: define {{[^@]+}}@n_caller
+// CHECK-SAME: () #[[ATTR3:[0-9]+]] {
+// CHECK:    call void @n_callee()
+// CHECK:    call void @s_callee() #[[ATTR12:[0-9]+]]
+// CHECK:    call void @sc_callee() #[[ATTR13:[0-9]+]]
+//
+void n_caller(void) {
+  n_callee();
+  s_callee();
+  sc_callee();
+}
+
+
+// CHECK-LABEL: define {{[^@]+}}@s_caller
+// CHECK-SAME: () #[[ATTR7:[0-9]+]] {
+// CHECK:    call void @n_callee()
+// CHECK:    call void @s_callee() #[[ATTR12]]
+// CHECK:    call void @sc_callee() #[[ATTR13]]
+//
+void s_caller(void) __arm_streaming {
+  n_callee();
+  s_callee();
+  sc_callee();
+}
+
+
+// CHECK-LABEL: define {{[^@]+}}@sc_caller
+// CHECK-SAME: () #[[ATTR11:[0-9]+]] {
+// CHECK:    call void @n_callee()
+// CHECK:    call void @s_callee() #[[ATTR12]]
+// CHECK:    call void @sc_callee() #[[ATTR13]]
+//
+void sc_caller(void) __arm_streaming_compatible {
+  n_callee();
+  s_callee();
+  sc_callee();
+}
+
+
+// CHECK: attributes #[[ATTR0:[0-9]+]] = {{.*}} "aarch64_pstate_sm_body"
+// CHECK: attributes #[[ATTR1:[0-9]+]] = {{.*}} "aarch64_pstate_sm_body"
+// CHECK: attributes #[[ATTR2:[0-9]+]] = {{.*}}
+// CHECK: attributes #[[ATTR3]] = {{.*}}
+// CHECK: attributes #[[ATTR4:[0-9]+]] = {{.*}} "aarch64_pstate_sm_enabled"
+// CHECK: attributes #[[ATTR5:[0-9]+]] = {{.*}} "aarch64_pstate_sm_enabled"
+// CHECK: attributes #[[ATTR6:[0-9]+]] = {{.*}} "aarch64_pstate_sm_body" "aarch64_pstate_sm_enabled"
+// CHECK: attributes #[[ATTR7]] = {{.*}} "aarch64_pstate_sm_enabled"
+// CHECK: attributes #[[ATTR8:[0-9]+]] = {{.*}} "aarch64_pstate_sm_compatible"
+// CHECK: attributes #[[ATTR9:[0-9]+]] = {{.*}} "aarch64_pstate_sm_compatible"
+// CHECK: attributes #[[ATTR10]] = {{.*}} "aarch64_pstate_sm_body" "aarch64_pstate_sm_compatible"
+// CHECK: attributes #[[ATTR11]] = {{.*}} "aarch64_pstate_sm_compatible"
+// CHECK: attributes #[[ATTR12]] = {{.*}} "aarch64_pstate_sm_enabled"
+// CHECK: attributes #[[ATTR13]] = {{.*}} "aarch64_pstate_sm_compatible"
diff --git a/clang/test/CodeGen/arm64ec-hybrid-patchable.c b/clang/test/CodeGen/arm64ec-hybrid-patchable.c
@@ -0,0 +1,34 @@
+// REQUIRES: aarch64-registered-target
+// RUN: %clang_cc1 -triple arm64ec-pc-windows -fms-extensions -emit-llvm -o - %s -verify | FileCheck %s
+
+// CHECK: ;    Function Attrs: hybrid_patchable noinline nounwind optnone
+// CHECK-NEXT: define dso_local i32 @func() #0 {
+int __attribute__((hybrid_patchable)) func(void) {  return 1; }
+
+// CHECK: ;    Function Attrs: hybrid_patchable noinline nounwind optnone
+// CHECK-NEXT: define dso_local i32 @func2() #0 {
+int __declspec(hybrid_patchable) func2(void) {  return 2; }
+
+// CHECK: ;    Function Attrs: hybrid_patchable noinline nounwind optnone
+// CHECK-NEXT: define dso_local i32 @func3() #0 {
+int __declspec(hybrid_patchable) func3(void);
+int func3(void) {  return 3; }
+
+// CHECK: ;    Function Attrs: hybrid_patchable noinline nounwind optnone
+// CHECK-NEXT: define dso_local i32 @func4() #0 {
+[[clang::hybrid_patchable]] int func4(void);
+int func4(void) {  return 3; }
+
+// CHECK: ; Function Attrs: hybrid_patchable noinline nounwind optnone
+// CHECK-NEXT: define internal void @static_func() #0 {
+// expected-warning@+1 {{'hybrid_patchable' is ignored on functions without external linkage}}
+static void __declspec(hybrid_patchable) static_func(void) {}
+
+// CHECK: ;    Function Attrs: hybrid_patchable noinline nounwind optnone
+// CHECK-NEXT: define linkonce_odr dso_local i32 @func5() #0 comdat {
+int inline __declspec(hybrid_patchable) func5(void) {  return 4; }
+
+void caller(void) {
+  static_func();
+  func5();
+}
diff --git a/clang/test/CodeGen/finite-math.c b/clang/test/CodeGen/finite-math.c
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -ffinite-math-only -emit-llvm -o - %s | FileCheck %s -check-prefix=CHECK -check-prefix=FINITE
+// RUN: %clang_cc1 -menable-no-infs -menable-no-nans -emit-llvm -o - %s | FileCheck %s -check-prefix=CHECK -check-prefix=FINITE
 // RUN: %clang_cc1 -fno-signed-zeros -emit-llvm -o - %s | FileCheck %s -check-prefix=CHECK  -check-prefix=NSZ
 // RUN: %clang_cc1 -freciprocal-math -emit-llvm -o - %s | FileCheck %s -check-prefix=CHECK  -check-prefix=RECIP
 // RUN: %clang_cc1 -mreassociate -emit-llvm -o - %s | FileCheck %s -check-prefix=CHECK  -check-prefix=REASSOC

diff --git a/clang/test/CodeGen/fp-floatcontrol-stack.cpp b/clang/test/CodeGen/fp-floatcontrol-stack.cpp
@@ -1,7 +1,7 @@
 // RUN: %clang_cc1 -triple x86_64-linux-gnu -ffp-contract=on -DDEFAULT=1 -emit-llvm -o - %s | FileCheck --check-prefix=CHECK-DDEFAULT %s
 // RUN: %clang_cc1 -triple x86_64-linux-gnu -ffp-contract=on -DEBSTRICT=1 -ffp-exception-behavior=strict -emit-llvm -o - %s | FileCheck --check-prefix=CHECK-DEBSTRICT %s
 // RUN: %clang_cc1 -triple x86_64-linux-gnu -DFAST=1 -ffast-math -ffp-contract=fast -emit-llvm -o - %s | FileCheck --check-prefix=CHECK-FAST %s
-// RUN: %clang_cc1 -triple x86_64-linux-gnu -ffp-contract=on -DNOHONOR=1 -menable-no-infs -menable-no-nans -emit-llvm -o - %s | FileCheck --check-prefix=CHECK-NOHONOR %s
+// RUN: %clang_cc1 -triple x86_64-linux-gnu -ffp-contract=on -DNOHONOR=1 -ffinite-math-only -menable-no-infs -menable-no-nans -emit-llvm -o - %s | FileCheck --check-prefix=CHECK-NOHONOR %s
 
 #define FUN(n) \
   (float z) { return n * z + n; }

diff --git a/clang/test/CodeGen/fp-options-to-fast-math-flags.c b/clang/test/CodeGen/fp-options-to-fast-math-flags.c
@@ -1,7 +1,7 @@
 // RUN: %clang_cc1 -triple x86_64-unknown-unknown -emit-llvm -o - %s | FileCheck -check-prefix CHECK-PRECISE %s
 // RUN: %clang_cc1 -triple x86_64-unknown-unknown -menable-no-nans -emit-llvm -o - %s | FileCheck -check-prefix CHECK-NO-NANS %s
 // RUN: %clang_cc1 -triple x86_64-unknown-unknown -menable-no-infs -emit-llvm -o - %s | FileCheck -check-prefix CHECK-NO-INFS %s
-// RUN: %clang_cc1 -triple x86_64-unknown-unknown -ffinite-math-only -emit-llvm -o - %s | FileCheck -check-prefix CHECK-FINITE %s
+// RUN: %clang_cc1 -triple x86_64-unknown-unknown -menable-no-infs -menable-no-nans -emit-llvm -o - %s | FileCheck -check-prefix CHECK-FINITE %s
 // RUN: %clang_cc1 -triple x86_64-unknown-unknown -fno-signed-zeros -emit-llvm -o - %s | FileCheck -check-prefix CHECK-NO-SIGNED-ZEROS %s
 // RUN: %clang_cc1 -triple x86_64-unknown-unknown -mreassociate -emit-llvm -o - %s | FileCheck -check-prefix CHECK-REASSOC %s
 // RUN: %clang_cc1 -triple x86_64-unknown-unknown -freciprocal-math -emit-llvm -o - %s | FileCheck -check-prefix CHECK-RECIP %s

diff --git a/clang/test/CodeGen/nofpclass.c b/clang/test/CodeGen/nofpclass.c
@@ -1,7 +1,7 @@
 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --check-attributes --version 2
 // REQUIRES: x86-registered-target
-// RUN: %clang_cc1 -triple x86_64-unknown-unknown -target-feature +avx -fenable-matrix -ffinite-math-only -emit-llvm -o - %s | FileCheck -check-prefixes=CFINITEONLY %s
-// RUN: %clang_cc1 -x cl -triple x86_64-unknown-unknown -target-feature +avx -fenable-matrix -cl-finite-math-only -emit-llvm -o - %s | FileCheck -check-prefixes=CLFINITEONLY %s
+// RUN: %clang_cc1 -triple x86_64-unknown-unknown -target-feature +avx -fenable-matrix -menable-no-infs -menable-no-nans -emit-llvm -o - %s | FileCheck -check-prefixes=CFINITEONLY %s
+// RUN: %clang_cc1 -x cl -triple x86_64-unknown-unknown -target-feature +avx -fenable-matrix -menable-no-nans -menable-no-infs -emit-llvm -o - %s | FileCheck -check-prefixes=CLFINITEONLY %s
 
 // RUN: %clang_cc1 -triple x86_64-unknown-unknown -target-feature +avx -fenable-matrix -menable-no-nans -emit-llvm -o - %s | FileCheck -check-prefixes=NONANS %s
 // RUN: %clang_cc1 -triple x86_64-unknown-unknown -target-feature +avx -fenable-matrix -menable-no-infs -emit-llvm -o - %s | FileCheck -check-prefixes=NOINFS %s

diff --git a/clang/test/CodeGenCUDA/amdgpu-atomic-ops.cu b/clang/test/CodeGenCUDA/amdgpu-atomic-ops.cu
@@ -1,6 +1,10 @@
 // RUN: %clang_cc1 -x hip %s -emit-llvm -o - -triple=amdgcn-amd-amdhsa \
 // RUN:   -fcuda-is-device -target-cpu gfx906 -fnative-half-type \
-// RUN:   -fnative-half-arguments-and-returns | FileCheck %s
+// RUN:   -fnative-half-arguments-and-returns | FileCheck -check-prefixes=CHECK,SAFEIR %s
+
+// RUN: %clang_cc1 -x hip %s -emit-llvm -o - -triple=amdgcn-amd-amdhsa \
+// RUN:   -fcuda-is-device -target-cpu gfx906 -fnative-half-type \
+// RUN:   -fnative-half-arguments-and-returns -munsafe-fp-atomics | FileCheck -check-prefixes=CHECK,UNSAFEIR %s
 
 // RUN: %clang_cc1 -x hip %s -O3 -S -o - -triple=amdgcn-amd-amdhsa \
 // RUN:   -fcuda-is-device -target-cpu gfx1100 -fnative-half-type \
@@ -18,24 +22,38 @@
 
 __global__ void ffp1(float *p) {
   // CHECK-LABEL: @_Z4ffp1Pf
-  // CHECK: atomicrmw fadd ptr {{.*}} monotonic
-  // CHECK: atomicrmw fmax ptr {{.*}} monotonic
-  // CHECK: atomicrmw fmin ptr {{.*}} monotonic
-  // CHECK: atomicrmw fmax ptr {{.*}} syncscope("agent-one-as") monotonic
-  // CHECK: atomicrmw fmin ptr {{.*}} syncscope("workgroup-one-as") monotonic
+  // SAFEIR: atomicrmw fadd ptr {{.*}} monotonic, align 4{{$}}
+  // SAFEIR: atomicrmw fsub ptr {{.*}} monotonic, align 4{{$}}
+  // SAFEIR: atomicrmw fmax ptr {{.*}} monotonic, align 4{{$}}
+  // SAFEIR: atomicrmw fmin ptr {{.*}} monotonic, align 4{{$}}
+  // SAFEIR: atomicrmw fmax ptr {{.*}} syncscope("agent-one-as") monotonic, align 4{{$}}
+  // SAFEIR: atomicrmw fmin ptr {{.*}} syncscope("workgroup-one-as") monotonic, align 4{{$}}
+
+  // UNSAFEIR: atomicrmw fadd ptr {{.*}} monotonic, align 4, !amdgpu.no.fine.grained.memory !{{[0-9]+}}, !amdgpu.ignore.denormal.mode !{{[0-9]+$}}
+  // UNSAFEIR: atomicrmw fsub ptr {{.*}} monotonic, align 4, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+  // UNSAFEIR: atomicrmw fmax ptr {{.*}} monotonic, align 4, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+  // UNSAFEIR: atomicrmw fmin ptr {{.*}} monotonic, align 4, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+  // UNSAFEIR: atomicrmw fmax ptr {{.*}} syncscope("agent-one-as") monotonic, align 4, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+  // UNSAFEIR: atomicrmw fmin ptr {{.*}} syncscope("workgroup-one-as") monotonic, align 4, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+
   // SAFE: _Z4ffp1Pf
   // SAFE: global_atomic_cmpswap
   // SAFE: global_atomic_cmpswap
   // SAFE: global_atomic_cmpswap
   // SAFE: global_atomic_cmpswap
   // SAFE: global_atomic_cmpswap
+  // SAFE: global_atomic_cmpswap
+
   // UNSAFE: _Z4ffp1Pf
   // UNSAFE: global_atomic_add_f32
   // UNSAFE: global_atomic_cmpswap
   // UNSAFE: global_atomic_cmpswap
   // UNSAFE: global_atomic_cmpswap
   // UNSAFE: global_atomic_cmpswap
+  // UNSAFE: global_atomic_cmpswap
+
   __atomic_fetch_add(p, 1.0f, memory_order_relaxed);
+  __atomic_fetch_sub(p, 1.0f, memory_order_relaxed);
   __atomic_fetch_max(p, 1.0f, memory_order_relaxed);
   __atomic_fetch_min(p, 1.0f, memory_order_relaxed);
   __hip_atomic_fetch_max(p, 1.0f, memory_order_relaxed, __HIP_MEMORY_SCOPE_AGENT);
@@ -44,23 +62,36 @@ __global__ void ffp1(float *p) {
 
 __global__ void ffp2(double *p) {
   // CHECK-LABEL: @_Z4ffp2Pd
-  // CHECK: atomicrmw fsub ptr {{.*}} monotonic
-  // CHECK: atomicrmw fmax ptr {{.*}} monotonic
-  // CHECK: atomicrmw fmin ptr {{.*}} monotonic
-  // CHECK: atomicrmw fmax ptr {{.*}} syncscope("agent-one-as") monotonic
-  // CHECK: atomicrmw fmin ptr {{.*}} syncscope("workgroup-one-as") monotonic
+  // SAFEIR: atomicrmw fadd ptr {{.*}} monotonic, align 8{{$}}
+  // SAFEIR: atomicrmw fsub ptr {{.*}} monotonic, align 8{{$}}
+  // SAFEIR: atomicrmw fmax ptr {{.*}} monotonic, align 8{{$}}
+  // SAFEIR: atomicrmw fmin ptr {{.*}} monotonic, align 8{{$}}
+  // SAFEIR: atomicrmw fmax ptr {{.*}} syncscope("agent-one-as") monotonic, align 8{{$}}
+  // SAFEIR: atomicrmw fmin ptr {{.*}} syncscope("workgroup-one-as") monotonic, align 8{{$}}
+
+  // UNSAFEIR: atomicrmw fadd ptr {{.*}} monotonic, align 8, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+  // UNSAFEIR: atomicrmw fsub ptr {{.*}} monotonic, align 8, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+  // UNSAFEIR: atomicrmw fmax ptr {{.*}} monotonic, align 8, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+  // UNSAFEIR: atomicrmw fmin ptr {{.*}} monotonic, align 8, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+  // UNSAFEIR: atomicrmw fmax ptr {{.*}} syncscope("agent-one-as") monotonic, align 8, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+  // UNSAFEIR: atomicrmw fmin ptr {{.*}} syncscope("workgroup-one-as") monotonic, align 8, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+
   // SAFE-LABEL: @_Z4ffp2Pd
   // SAFE: global_atomic_cmpswap_b64
   // SAFE: global_atomic_cmpswap_b64
   // SAFE: global_atomic_cmpswap_b64
   // SAFE: global_atomic_cmpswap_b64
   // SAFE: global_atomic_cmpswap_b64
+  // SAFE: global_atomic_cmpswap_b64
+
   // UNSAFE-LABEL: @_Z4ffp2Pd
+  // UNSAFE: global_atomic_add_f64
   // UNSAFE: global_atomic_cmpswap_x2
   // UNSAFE: global_atomic_cmpswap_x2
   // UNSAFE: global_atomic_cmpswap_x2
   // UNSAFE: global_atomic_max_f64
   // UNSAFE: global_atomic_min_f64
+  __atomic_fetch_add(p, 1.0, memory_order_relaxed);
   __atomic_fetch_sub(p, 1.0, memory_order_relaxed);
   __atomic_fetch_max(p, 1.0, memory_order_relaxed);
   __atomic_fetch_min(p, 1.0, memory_order_relaxed);
@@ -71,11 +102,20 @@ __global__ void ffp2(double *p) {
 // long double is the same as double for amdgcn.
 __global__ void ffp3(long double *p) {
   // CHECK-LABEL: @_Z4ffp3Pe
-  // CHECK: atomicrmw fsub ptr {{.*}} monotonic
-  // CHECK: atomicrmw fmax ptr {{.*}} monotonic
-  // CHECK: atomicrmw fmin ptr {{.*}} monotonic
-  // CHECK: atomicrmw fmax ptr {{.*}} syncscope("agent-one-as") monotonic
-  // CHECK: atomicrmw fmin ptr {{.*}} syncscope("workgroup-one-as") monotonic
+  // SAFEIR: atomicrmw fadd ptr {{.*}} monotonic, align 8{{$}}
+  // SAFEIR: atomicrmw fsub ptr {{.*}} monotonic, align 8{{$}}
+  // SAFEIR: atomicrmw fmax ptr {{.*}} monotonic, align 8{{$}}
+  // SAFEIR: atomicrmw fmin ptr {{.*}} monotonic, align 8{{$}}
+  // SAFEIR: atomicrmw fmax ptr {{.*}} syncscope("agent-one-as") monotonic, align 8{{$}}
+  // SAFEIR: atomicrmw fmin ptr {{.*}} syncscope("workgroup-one-as") monotonic, align 8{{$}}
+
+  // UNSAFEIR: atomicrmw fadd ptr {{.*}} monotonic, align 8, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+  // UNSAFEIR: atomicrmw fsub ptr {{.*}} monotonic, align 8, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+  // UNSAFEIR: atomicrmw fmax ptr {{.*}} monotonic, align 8, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+  // UNSAFEIR: atomicrmw fmin ptr {{.*}} monotonic, align 8, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+  // UNSAFEIR: atomicrmw fmax ptr {{.*}} syncscope("agent-one-as") monotonic, align 8, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+  // UNSAFEIR: atomicrmw fmin ptr {{.*}} syncscope("workgroup-one-as") monotonic, align 8, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+
   // SAFE-LABEL: @_Z4ffp3Pe
   // SAFE: global_atomic_cmpswap_b64
   // SAFE: global_atomic_cmpswap_b64
@@ -88,6 +128,7 @@ __global__ void ffp3(long double *p) {
   // UNSAFE: global_atomic_cmpswap_x2
   // UNSAFE: global_atomic_max_f64
   // UNSAFE: global_atomic_min_f64
+  __atomic_fetch_add(p, 1.0L, memory_order_relaxed);
   __atomic_fetch_sub(p, 1.0L, memory_order_relaxed);
   __atomic_fetch_max(p, 1.0L, memory_order_relaxed);
   __atomic_fetch_min(p, 1.0L, memory_order_relaxed);
@@ -98,37 +139,52 @@ __global__ void ffp3(long double *p) {
 __device__ double ffp4(double *p, float f) {
   // CHECK-LABEL: @_Z4ffp4Pdf
   // CHECK: fpext float {{.*}} to double
-  // CHECK: atomicrmw fsub ptr {{.*}} monotonic
+  // SAFEIR: atomicrmw fsub ptr {{.*}} monotonic, align 8{{$}}
+  // UNSAFEIR: atomicrmw fsub ptr {{.*}} monotonic, align 8, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
   return __atomic_fetch_sub(p, f, memory_order_relaxed);
 }
 
 __device__ double ffp5(double *p, int i) {
   // CHECK-LABEL: @_Z4ffp5Pdi
   // CHECK: sitofp i32 {{.*}} to double
-  // CHECK: atomicrmw fsub ptr {{.*}} monotonic
+  // SAFEIR: atomicrmw fsub ptr {{.*}} monotonic, align 8{{$}}
+  // UNSAFEIR: atomicrmw fsub ptr {{.*}} monotonic, align 8, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
   return __atomic_fetch_sub(p, i, memory_order_relaxed);
 }
 
 __global__ void ffp6(_Float16 *p) {
   // CHECK-LABEL: @_Z4ffp6PDF16
-  // CHECK: atomicrmw fadd ptr {{.*}} monotonic
-  // CHECK: atomicrmw fmax ptr {{.*}} monotonic
-  // CHECK: atomicrmw fmin ptr {{.*}} monotonic
-  // CHECK: atomicrmw fmax ptr {{.*}} syncscope("agent-one-as") monotonic
-  // CHECK: atomicrmw fmin ptr {{.*}} syncscope("workgroup-one-as") monotonic
+  // SAFEIR: atomicrmw fadd ptr {{.*}} monotonic, align 2{{$}}
+  // SAFEIR: atomicrmw fsub ptr {{.*}} monotonic, align 2{{$}}
+  // SAFEIR: atomicrmw fmax ptr {{.*}} monotonic, align 2{{$}}
+  // SAFEIR: atomicrmw fmin ptr {{.*}} monotonic, align 2{{$}}
+  // SAFEIR: atomicrmw fmax ptr {{.*}} syncscope("agent-one-as") monotonic, align 2{{$}}
+  // SAFEIR: atomicrmw fmin ptr {{.*}} syncscope("workgroup-one-as") monotonic, align 2{{$}}
+
+  // UNSAFEIR: atomicrmw fadd ptr {{.*}} monotonic, align 2, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+  // UNSAFEIR: atomicrmw fsub ptr {{.*}} monotonic, align 2, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+  // UNSAFEIR: atomicrmw fmax ptr {{.*}} monotonic, align 2, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+  // UNSAFEIR: atomicrmw fmin ptr {{.*}} monotonic, align 2, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+  // UNSAFEIR: atomicrmw fmax ptr {{.*}} syncscope("agent-one-as") monotonic, align 2, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+  // UNSAFEIR: atomicrmw fmin ptr {{.*}} syncscope("workgroup-one-as") monotonic, align 2, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+
   // SAFE: _Z4ffp6PDF16
   // SAFE: global_atomic_cmpswap
   // SAFE: global_atomic_cmpswap
   // SAFE: global_atomic_cmpswap
   // SAFE: global_atomic_cmpswap
   // SAFE: global_atomic_cmpswap
+  // SAFE: global_atomic_cmpswap
+
   // UNSAFE: _Z4ffp6PDF16
   // UNSAFE: global_atomic_cmpswap
   // UNSAFE: global_atomic_cmpswap
   // UNSAFE: global_atomic_cmpswap
   // UNSAFE: global_atomic_cmpswap
   // UNSAFE: global_atomic_cmpswap
+  // UNSAFE: global_atomic_cmpswap
   __atomic_fetch_add(p, 1.0, memory_order_relaxed);
+  __atomic_fetch_sub(p, 1.0, memory_order_relaxed);
   __atomic_fetch_max(p, 1.0, memory_order_relaxed);
   __atomic_fetch_min(p, 1.0, memory_order_relaxed);
   __hip_atomic_fetch_max(p, 1.0f, memory_order_relaxed, __HIP_MEMORY_SCOPE_AGENT);

diff --git a/clang/test/CodeGenOpenCL/builtins-alloca.cl b/clang/test/CodeGenOpenCL/builtins-alloca.cl
@@ -0,0 +1,141 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 5
+// RUN: %clang_cc1 %s -O0 -triple amdgcn-amd-amdhsa -cl-std=CL1.2 \
+// RUN:     -emit-llvm -o - | FileCheck --check-prefixes=OPENCL %s
+// RUN: %clang_cc1 %s -O0 -triple amdgcn-amd-amdhsa -cl-std=CL2.0 \
+// RUN:     -emit-llvm -o - | FileCheck --check-prefixes=OPENCL %s
+// RUN: %clang_cc1 %s -O0 -triple amdgcn-amd-amdhsa -cl-std=CL3.0 \
+// RUN:     -emit-llvm -o - | FileCheck --check-prefixes=OPENCL %s
+// RUN: %clang_cc1 %s -O0 -triple amdgcn-amd-amdhsa -cl-std=CL3.0 -cl-ext=+__opencl_c_generic_address_space \
+// RUN:     -emit-llvm -o - | FileCheck --check-prefixes=OPENCL %s
+
+// OPENCL-LABEL: define dso_local void @test1_builtin_alloca(
+// OPENCL-SAME: i32 noundef [[N:%.*]]) #[[ATTR0:[0-9]+]] {
+// OPENCL-NEXT:  [[ENTRY:.*:]]
+// OPENCL-NEXT:    [[N_ADDR:%.*]] = alloca i32, align 4, addrspace(5)
+// OPENCL-NEXT:    [[ALLOC_PTR:%.*]] = alloca ptr addrspace(5), align 4, addrspace(5)
+// OPENCL-NEXT:    store i32 [[N]], ptr addrspace(5) [[N_ADDR]], align 4
+// OPENCL-NEXT:    [[TMP0:%.*]] = load i32, ptr addrspace(5) [[N_ADDR]], align 4
+// OPENCL-NEXT:    [[CONV:%.*]] = zext i32 [[TMP0]] to i64
+// OPENCL-NEXT:    [[MUL:%.*]] = mul i64 [[CONV]], 4
+// OPENCL-NEXT:    [[TMP1:%.*]] = alloca i8, i64 [[MUL]], align 8, addrspace(5)
+// OPENCL-NEXT:    store ptr addrspace(5) [[TMP1]], ptr addrspace(5) [[ALLOC_PTR]], align 4
+// OPENCL-NEXT:    ret void
+//
+void test1_builtin_alloca(unsigned n) {
+    __private float* alloc_ptr = (__private float*)__builtin_alloca(n*sizeof(int));
+}
+
+// OPENCL-LABEL: define dso_local void @test1_builtin_alloca_uninitialized(
+// OPENCL-SAME: i32 noundef [[N:%.*]]) #[[ATTR0]] {
+// OPENCL-NEXT:  [[ENTRY:.*:]]
+// OPENCL-NEXT:    [[N_ADDR:%.*]] = alloca i32, align 4, addrspace(5)
+// OPENCL-NEXT:    [[ALLOC_PTR_UNINITIALIZED:%.*]] = alloca ptr addrspace(5), align 4, addrspace(5)
+// OPENCL-NEXT:    store i32 [[N]], ptr addrspace(5) [[N_ADDR]], align 4
+// OPENCL-NEXT:    [[TMP0:%.*]] = load i32, ptr addrspace(5) [[N_ADDR]], align 4
+// OPENCL-NEXT:    [[CONV:%.*]] = zext i32 [[TMP0]] to i64
+// OPENCL-NEXT:    [[MUL:%.*]] = mul i64 [[CONV]], 4
+// OPENCL-NEXT:    [[TMP1:%.*]] = alloca i8, i64 [[MUL]], align 8, addrspace(5)
+// OPENCL-NEXT:    store ptr addrspace(5) [[TMP1]], ptr addrspace(5) [[ALLOC_PTR_UNINITIALIZED]], align 4
+// OPENCL-NEXT:    ret void
+//
+void test1_builtin_alloca_uninitialized(unsigned n) {
+    __private float* alloc_ptr_uninitialized = (__private float*)__builtin_alloca_uninitialized(n*sizeof(int));
+}
+
+// OPENCL-LABEL: define dso_local void @test1_builtin_alloca_with_align(
+// OPENCL-SAME: i32 noundef [[N:%.*]]) #[[ATTR0]] {
+// OPENCL-NEXT:  [[ENTRY:.*:]]
+// OPENCL-NEXT:    [[N_ADDR:%.*]] = alloca i32, align 4, addrspace(5)
+// OPENCL-NEXT:    [[ALLOC_PTR_ALIGN:%.*]] = alloca ptr addrspace(5), align 4, addrspace(5)
+// OPENCL-NEXT:    store i32 [[N]], ptr addrspace(5) [[N_ADDR]], align 4
+// OPENCL-NEXT:    [[TMP0:%.*]] = load i32, ptr addrspace(5) [[N_ADDR]], align 4
+// OPENCL-NEXT:    [[CONV:%.*]] = zext i32 [[TMP0]] to i64
+// OPENCL-NEXT:    [[MUL:%.*]] = mul i64 [[CONV]], 4
+// OPENCL-NEXT:    [[TMP1:%.*]] = alloca i8, i64 [[MUL]], align 1, addrspace(5)
+// OPENCL-NEXT:    store ptr addrspace(5) [[TMP1]], ptr addrspace(5) [[ALLOC_PTR_ALIGN]], align 4
+// OPENCL-NEXT:    ret void
+//
+void test1_builtin_alloca_with_align(unsigned n) {
+    __private float* alloc_ptr_align = (__private float*)__builtin_alloca_with_align((n*sizeof(int)), 8);
+}
+
+// OPENCL-LABEL: define dso_local void @test1_builtin_alloca_with_align_uninitialized(
+// OPENCL-SAME: i32 noundef [[N:%.*]]) #[[ATTR0]] {
+// OPENCL-NEXT:  [[ENTRY:.*:]]
+// OPENCL-NEXT:    [[N_ADDR:%.*]] = alloca i32, align 4, addrspace(5)
+// OPENCL-NEXT:    [[ALLOC_PTR_ALIGN_UNINITIALIZED:%.*]] = alloca ptr addrspace(5), align 4, addrspace(5)
+// OPENCL-NEXT:    store i32 [[N]], ptr addrspace(5) [[N_ADDR]], align 4
+// OPENCL-NEXT:    [[TMP0:%.*]] = load i32, ptr addrspace(5) [[N_ADDR]], align 4
+// OPENCL-NEXT:    [[CONV:%.*]] = zext i32 [[TMP0]] to i64
+// OPENCL-NEXT:    [[MUL:%.*]] = mul i64 [[CONV]], 4
+// OPENCL-NEXT:    [[TMP1:%.*]] = alloca i8, i64 [[MUL]], align 1, addrspace(5)
+// OPENCL-NEXT:    store ptr addrspace(5) [[TMP1]], ptr addrspace(5) [[ALLOC_PTR_ALIGN_UNINITIALIZED]], align 4
+// OPENCL-NEXT:    ret void
+//
+void test1_builtin_alloca_with_align_uninitialized(unsigned n) {
+    __private float* alloc_ptr_align_uninitialized = (__private float*)__builtin_alloca_with_align_uninitialized((n*sizeof(int)), 8);
+}
+
+// OPENCL-LABEL: define dso_local void @test2_builtin_alloca(
+// OPENCL-SAME: i32 noundef [[N:%.*]]) #[[ATTR0]] {
+// OPENCL-NEXT:  [[ENTRY:.*:]]
+// OPENCL-NEXT:    [[N_ADDR:%.*]] = alloca i32, align 4, addrspace(5)
+// OPENCL-NEXT:    [[ALLOC_PTR:%.*]] = alloca ptr addrspace(5), align 4, addrspace(5)
+// OPENCL-NEXT:    store i32 [[N]], ptr addrspace(5) [[N_ADDR]], align 4
+// OPENCL-NEXT:    [[TMP0:%.*]] = load i32, ptr addrspace(5) [[N_ADDR]], align 4
+// OPENCL-NEXT:    [[CONV:%.*]] = zext i32 [[TMP0]] to i64
+// OPENCL-NEXT:    [[TMP1:%.*]] = alloca i8, i64 [[CONV]], align 8, addrspace(5)
+// OPENCL-NEXT:    store ptr addrspace(5) [[TMP1]], ptr addrspace(5) [[ALLOC_PTR]], align 4
+// OPENCL-NEXT:    ret void
+//
+void test2_builtin_alloca(unsigned n) {
+    __private void *alloc_ptr = __builtin_alloca(n);
+}
+
+// OPENCL-LABEL: define dso_local void @test2_builtin_alloca_uninitialized(
+// OPENCL-SAME: i32 noundef [[N:%.*]]) #[[ATTR0]] {
+// OPENCL-NEXT:  [[ENTRY:.*:]]
+// OPENCL-NEXT:    [[N_ADDR:%.*]] = alloca i32, align 4, addrspace(5)
+// OPENCL-NEXT:    [[ALLOC_PTR_UNINITIALIZED:%.*]] = alloca ptr addrspace(5), align 4, addrspace(5)
+// OPENCL-NEXT:    store i32 [[N]], ptr addrspace(5) [[N_ADDR]], align 4
+// OPENCL-NEXT:    [[TMP0:%.*]] = load i32, ptr addrspace(5) [[N_ADDR]], align 4
+// OPENCL-NEXT:    [[CONV:%.*]] = zext i32 [[TMP0]] to i64
+// OPENCL-NEXT:    [[TMP1:%.*]] = alloca i8, i64 [[CONV]], align 8, addrspace(5)
+// OPENCL-NEXT:    store ptr addrspace(5) [[TMP1]], ptr addrspace(5) [[ALLOC_PTR_UNINITIALIZED]], align 4
+// OPENCL-NEXT:    ret void
+//
+void test2_builtin_alloca_uninitialized(unsigned n) {
+    __private void *alloc_ptr_uninitialized = __builtin_alloca_uninitialized(n);
+}
+
+// OPENCL-LABEL: define dso_local void @test2_builtin_alloca_with_align(
+// OPENCL-SAME: i32 noundef [[N:%.*]]) #[[ATTR0]] {
+// OPENCL-NEXT:  [[ENTRY:.*:]]
+// OPENCL-NEXT:    [[N_ADDR:%.*]] = alloca i32, align 4, addrspace(5)
+// OPENCL-NEXT:    [[ALLOC_PTR_ALIGN:%.*]] = alloca ptr addrspace(5), align 4, addrspace(5)
+// OPENCL-NEXT:    store i32 [[N]], ptr addrspace(5) [[N_ADDR]], align 4
+// OPENCL-NEXT:    [[TMP0:%.*]] = load i32, ptr addrspace(5) [[N_ADDR]], align 4
+// OPENCL-NEXT:    [[CONV:%.*]] = zext i32 [[TMP0]] to i64
+// OPENCL-NEXT:    [[TMP1:%.*]] = alloca i8, i64 [[CONV]], align 1, addrspace(5)
+// OPENCL-NEXT:    store ptr addrspace(5) [[TMP1]], ptr addrspace(5) [[ALLOC_PTR_ALIGN]], align 4
+// OPENCL-NEXT:    ret void
+//
+void test2_builtin_alloca_with_align(unsigned n) {
+    __private void *alloc_ptr_align = __builtin_alloca_with_align(n, 8);
+}
+
+// OPENCL-LABEL: define dso_local void @test2_builtin_alloca_with_align_uninitialized(
+// OPENCL-SAME: i32 noundef [[N:%.*]]) #[[ATTR0]] {
+// OPENCL-NEXT:  [[ENTRY:.*:]]
+// OPENCL-NEXT:    [[N_ADDR:%.*]] = alloca i32, align 4, addrspace(5)
+// OPENCL-NEXT:    [[ALLOC_PTR_ALIGN_UNINITIALIZED:%.*]] = alloca ptr addrspace(5), align 4, addrspace(5)
+// OPENCL-NEXT:    store i32 [[N]], ptr addrspace(5) [[N_ADDR]], align 4
+// OPENCL-NEXT:    [[TMP0:%.*]] = load i32, ptr addrspace(5) [[N_ADDR]], align 4
+// OPENCL-NEXT:    [[CONV:%.*]] = zext i32 [[TMP0]] to i64
+// OPENCL-NEXT:    [[TMP1:%.*]] = alloca i8, i64 [[CONV]], align 1, addrspace(5)
+// OPENCL-NEXT:    store ptr addrspace(5) [[TMP1]], ptr addrspace(5) [[ALLOC_PTR_ALIGN_UNINITIALIZED]], align 4
+// OPENCL-NEXT:    ret void
+//
+void test2_builtin_alloca_with_align_uninitialized(unsigned n) {
+    __private void *alloc_ptr_align_uninitialized = __builtin_alloca_with_align_uninitialized(n, 8);
+}
diff --git a/clang/test/CodeGenOpenCL/relaxed-fpmath.cl b/clang/test/CodeGenOpenCL/relaxed-fpmath.cl
@@ -1,6 +1,6 @@
 // RUN: %clang_cc1 %s -emit-llvm -o - | FileCheck %s -check-prefix=NORMAL
 // RUN: %clang_cc1 %s -emit-llvm -cl-fast-relaxed-math -o - | FileCheck %s -check-prefix=FAST
-// RUN: %clang_cc1 %s -emit-llvm -cl-finite-math-only -o - | FileCheck %s -check-prefix=FINITE
+// RUN: %clang_cc1 %s -emit-llvm -menable-no-infs -menable-no-nans -cl-finite-math-only -o - | FileCheck %s -check-prefix=FINITE
 // RUN: %clang_cc1 %s -emit-llvm -cl-unsafe-math-optimizations -o - | FileCheck %s -check-prefix=UNSAFE
 // RUN: %clang_cc1 %s -emit-llvm -cl-mad-enable -o - | FileCheck %s -check-prefix=MAD
 // RUN: %clang_cc1 %s -emit-llvm -cl-no-signed-zeros -o - | FileCheck %s -check-prefix=NOSIGNED
@@ -9,7 +9,7 @@
 // RUN: %clang_cc1 %s -DGEN_PCH=1 -finclude-default-header -triple spir-unknown-unknown -emit-pch -o %t.pch
 // RUN: %clang_cc1 %s -include-pch %t.pch -fno-validate-pch -emit-llvm -o - | FileCheck %s -check-prefix=NORMAL
 // RUN: %clang_cc1 %s -include-pch %t.pch -fno-validate-pch -emit-llvm -cl-fast-relaxed-math -o - | FileCheck %s -check-prefix=FAST
-// RUN: %clang_cc1 %s -include-pch %t.pch -fno-validate-pch -emit-llvm -cl-finite-math-only -o - | FileCheck %s -check-prefix=FINITE
+// RUN: %clang_cc1 %s -include-pch %t.pch -fno-validate-pch -emit-llvm -menable-no-infs -menable-no-nans -cl-finite-math-only -o - | FileCheck %s -check-prefix=FINITE
 // RUN: %clang_cc1 %s -include-pch %t.pch -fno-validate-pch -emit-llvm -cl-unsafe-math-optimizations -o - | FileCheck %s -check-prefix=UNSAFE
 // RUN: %clang_cc1 %s -include-pch %t.pch -fno-validate-pch -emit-llvm -cl-mad-enable -o - | FileCheck %s -check-prefix=MAD
 // RUN: %clang_cc1 %s -include-pch %t.pch -fno-validate-pch -emit-llvm -cl-no-signed-zeros -o - | FileCheck %s -check-prefix=NOSIGNED

diff --git a/clang/test/Driver/amdgpu-toolchain.c b/clang/test/Driver/amdgpu-toolchain.c
@@ -28,3 +28,7 @@
 // RUN: %clang -### --target=amdgcn-amd-amdhsa -mcpu=gfx906 -nogpulib \
 // RUN:   -fuse-ld=ld %s 2>&1 | FileCheck -check-prefixes=LD %s
 // LD: ld.lld
+
+// RUN: %clang -### --target=amdgcn-amd-amdhsa -mcpu=gfx906 -nogpulib \
+// RUN:   -r %s 2>&1 | FileCheck -check-prefixes=RELO %s
+// RELO-NOT: -shared
diff --git a/clang/test/Driver/opencl.cl b/clang/test/Driver/opencl.cl
@@ -35,7 +35,7 @@
 // CHECK-OPT-DISABLE: "-cc1" {{.*}} "-cl-opt-disable"
 // CHECK-STRICT-ALIASING: "-cc1" {{.*}} "-cl-strict-aliasing"
 // CHECK-SINGLE-PRECISION-CONST: "-cc1" {{.*}} "-cl-single-precision-constant"
-// CHECK-FINITE-MATH-ONLY: "-cc1" {{.*}} "-cl-finite-math-only"
+// CHECK-FINITE-MATH-ONLY: "-cc1" {{.*}} "-menable-no-infs" "-menable-no-nans" "-cl-finite-math-only"
 // CHECK-KERNEL-ARG-INFO: "-cc1" {{.*}} "-cl-kernel-arg-info"
 // CHECK-UNSAFE-MATH-OPT: "-cc1" {{.*}} "-cl-unsafe-math-optimizations"
 // CHECK-FAST-RELAXED-MATH: "-cc1" {{.*}} "-cl-fast-relaxed-math"

diff --git a/clang/test/Driver/stack-size-section.c b/clang/test/Driver/stack-size-section.c
@@ -14,6 +14,7 @@
 
 // RUN: %clang -### --target=x86_64-linux-gnu -flto -fstack-size-section %s 2>&1 | FileCheck %s --check-prefix=LTO
 // RUN: %clang -### --target=x86_64-linux-gnu -flto -fstack-size-section -fno-stack-size-section %s 2>&1 | FileCheck %s --check-prefix=LTO-NO
+// RUN: %clang -### --target=x86_64-sie-ps5 -fstack-size-section %s 2>&1 | FileCheck %s --check-prefix=LTO
 
 // LTO: "-plugin-opt=-stack-size-section"
 // LTO-NO-NOT: "-plugin-opt=-stack-size-section"

diff --git a/clang/test/Headers/__clang_hip_cmath.hip b/clang/test/Headers/__clang_hip_cmath.hip
@@ -13,7 +13,8 @@
 // RUN:   -internal-isystem %S/../../lib/Headers/cuda_wrappers \
 // RUN:   -internal-isystem %S/Inputs/include \
 // RUN:   -triple amdgcn-amd-amdhsa -aux-triple x86_64-unknown-unknown \
-// RUN:   -target-cpu gfx906 -emit-llvm %s -fcuda-is-device -O1 -ffinite-math-only -o - \
+// RUN:   -target-cpu gfx906 -emit-llvm %s -fcuda-is-device -O1 -menable-no-infs \
+// RUN:   -menable-no-nans -o - \
 // RUN:   -D__HIPCC_RTC__ | FileCheck -check-prefix=FINITEONLY %s
 
 // DEFAULT-LABEL: @test_fma_f16(

diff --git a/clang/test/Headers/__clang_hip_math.hip b/clang/test/Headers/__clang_hip_math.hip
@@ -14,7 +14,8 @@
 // RUN:   -internal-isystem %S/../../lib/Headers/cuda_wrappers \
 // RUN:   -internal-isystem %S/Inputs/include \
 // RUN:   -triple amdgcn-amd-amdhsa -aux-triple x86_64-unknown-unknown \
-// RUN:   -target-cpu gfx906 -emit-llvm %s -fcuda-is-device -O1 -ffinite-math-only -o - \
+// RUN:   -target-cpu gfx906 -emit-llvm %s -fcuda-is-device -O1 -menable-no-infs \
+// RUN:   -menable-no-nans -o - \
 // RUN:   -D__HIPCC_RTC__ | FileCheck -check-prefixes=CHECK,FINITEONLY %s
 
 // Check that we end up with -fapprox-func set on intrinsic calls

diff --git a/clang/test/Headers/float.c b/clang/test/Headers/float.c
@@ -2,7 +2,7 @@
 // RUN: %clang_cc1 -fsyntax-only -verify -std=c99 -ffreestanding %s
 // RUN: %clang_cc1 -fsyntax-only -verify -std=c11 -ffreestanding %s
 // RUN: %clang_cc1 -fsyntax-only -verify -std=c23 -ffreestanding %s
-// RUN: %clang_cc1 -fsyntax-only -verify=finite -std=c23 -ffreestanding -ffinite-math-only %s
+// RUN: %clang_cc1 -fsyntax-only -verify=finite -std=c23 -ffreestanding -menable-no-nans -menable-no-infs %s
 // RUN: %clang_cc1 -fsyntax-only -verify -xc++ -std=c++11 -ffreestanding %s
 // RUN: %clang_cc1 -fsyntax-only -verify -xc++ -std=c++14 -ffreestanding %s
 // RUN: %clang_cc1 -fsyntax-only -verify -xc++ -std=c++17 -ffreestanding %s

diff --git a/clang/test/Misc/pragma-attribute-supported-attributes-list.test b/clang/test/Misc/pragma-attribute-supported-attributes-list.test
@@ -83,6 +83,7 @@
 // CHECK-NEXT: HIPManaged (SubjectMatchRule_variable)
 // CHECK-NEXT: HLSLResourceClass (SubjectMatchRule_record_not_is_union)
 // CHECK-NEXT: Hot (SubjectMatchRule_function)
+// CHECK-NEXT: HybridPatchable (SubjectMatchRule_function)
 // CHECK-NEXT: IBAction (SubjectMatchRule_objc_method_is_instance)
 // CHECK-NEXT: IFunc (SubjectMatchRule_function)
 // CHECK-NEXT: InitPriority (SubjectMatchRule_variable)

diff --git a/clang/test/OpenMP/amdgpu-unsafe-fp-atomics.cpp b/clang/test/OpenMP/amdgpu-unsafe-fp-atomics.cpp
@@ -0,0 +1,59 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 5
+// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple amdgcn-amd-amdhsa -fopenmp-targets=amdgcn-amd-amdhsa -emit-llvm %s -fopenmp-is-target-device -o - | FileCheck -check-prefix=DEFAULT %s
+// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple amdgcn-amd-amdhsa -fopenmp-targets=amdgcn-amd-amdhsa -munsafe-fp-atomics -emit-llvm %s -fopenmp-is-target-device -o - | FileCheck -check-prefix=UNSAFE-FP-ATOMICS %s
+
+#pragma omp declare target
+
+float fv, fx;
+double dv, dx;
+
+// DEFAULT-LABEL: define hidden void @_Z15atomic_fadd_f32v(
+// DEFAULT-SAME: ) #[[ATTR0:[0-9]+]] {
+// DEFAULT-NEXT:  [[ENTRY:.*:]]
+// DEFAULT-NEXT:    [[TMP0:%.*]] = load float, ptr addrspacecast (ptr addrspace(1) @fv to ptr), align 4
+// DEFAULT-NEXT:    [[TMP1:%.*]] = atomicrmw fadd ptr addrspacecast (ptr addrspace(1) @fx to ptr), float [[TMP0]] monotonic, align 4
+// DEFAULT-NEXT:    [[ADD:%.*]] = fadd float [[TMP1]], [[TMP0]]
+// DEFAULT-NEXT:    store float [[ADD]], ptr addrspacecast (ptr addrspace(1) @fv to ptr), align 4
+// DEFAULT-NEXT:    ret void
+//
+// UNSAFE-FP-ATOMICS-LABEL: define hidden void @_Z15atomic_fadd_f32v(
+// UNSAFE-FP-ATOMICS-SAME: ) #[[ATTR0:[0-9]+]] {
+// UNSAFE-FP-ATOMICS-NEXT:  [[ENTRY:.*:]]
+// UNSAFE-FP-ATOMICS-NEXT:    [[TMP0:%.*]] = load float, ptr addrspacecast (ptr addrspace(1) @fv to ptr), align 4
+// UNSAFE-FP-ATOMICS-NEXT:    [[TMP1:%.*]] = atomicrmw fadd ptr addrspacecast (ptr addrspace(1) @fx to ptr), float [[TMP0]] monotonic, align 4, !amdgpu.no.fine.grained.memory [[META5:![0-9]+]], !amdgpu.ignore.denormal.mode [[META5]]
+// UNSAFE-FP-ATOMICS-NEXT:    [[ADD:%.*]] = fadd float [[TMP1]], [[TMP0]]
+// UNSAFE-FP-ATOMICS-NEXT:    store float [[ADD]], ptr addrspacecast (ptr addrspace(1) @fv to ptr), align 4
+// UNSAFE-FP-ATOMICS-NEXT:    ret void
+//
+void atomic_fadd_f32() {
+#pragma omp atomic capture
+  fv = fx = fx + fv;
+}
+
+// DEFAULT-LABEL: define hidden void @_Z15atomic_fadd_f64v(
+// DEFAULT-SAME: ) #[[ATTR0]] {
+// DEFAULT-NEXT:  [[ENTRY:.*:]]
+// DEFAULT-NEXT:    [[TMP0:%.*]] = load double, ptr addrspacecast (ptr addrspace(1) @dv to ptr), align 8
+// DEFAULT-NEXT:    [[TMP1:%.*]] = atomicrmw fadd ptr addrspacecast (ptr addrspace(1) @dx to ptr), double [[TMP0]] monotonic, align 8
+// DEFAULT-NEXT:    [[ADD:%.*]] = fadd double [[TMP1]], [[TMP0]]
+// DEFAULT-NEXT:    store double [[ADD]], ptr addrspacecast (ptr addrspace(1) @dv to ptr), align 8
+// DEFAULT-NEXT:    ret void
+//
+// UNSAFE-FP-ATOMICS-LABEL: define hidden void @_Z15atomic_fadd_f64v(
+// UNSAFE-FP-ATOMICS-SAME: ) #[[ATTR0]] {
+// UNSAFE-FP-ATOMICS-NEXT:  [[ENTRY:.*:]]
+// UNSAFE-FP-ATOMICS-NEXT:    [[TMP0:%.*]] = load double, ptr addrspacecast (ptr addrspace(1) @dv to ptr), align 8
+// UNSAFE-FP-ATOMICS-NEXT:    [[TMP1:%.*]] = atomicrmw fadd ptr addrspacecast (ptr addrspace(1) @dx to ptr), double [[TMP0]] monotonic, align 8, !amdgpu.no.fine.grained.memory [[META5]]
+// UNSAFE-FP-ATOMICS-NEXT:    [[ADD:%.*]] = fadd double [[TMP1]], [[TMP0]]
+// UNSAFE-FP-ATOMICS-NEXT:    store double [[ADD]], ptr addrspacecast (ptr addrspace(1) @dv to ptr), align 8
+// UNSAFE-FP-ATOMICS-NEXT:    ret void
+//
+void atomic_fadd_f64() {
+#pragma omp atomic capture
+  dv = dx = dx + dv;
+}
+
+#pragma omp end declare target
+//.
+// UNSAFE-FP-ATOMICS: [[META5]] = !{}
+//.