merge main into amd-staging

Change-Id: I5a73a052c536d71b74786c373a67f71e4d489413

Author: ronlieb

.ci/metrics/metrics.py

@@ -80,6 +80,18 @@ def get_metrics(github_repo: github.Repository, workflows_to_track: dict[str, in
         completed_at = workflow_jobs[0].completed_at
 
         job_result = int(workflow_jobs[0].conclusion == "success")
+        if job_result:
+            # We still might want to mark the job as a failure if one of the steps
+            # failed. This is required due to use setting continue-on-error in
+            # the premerge pipeline to prevent sending emails while we are
+            # testing the infrastructure.
+            # TODO(boomanaiden154): Remove this once the premerge pipeline is no
+            # longer in a testing state and we can directly assert the workflow
+            # result.
+            for step in workflow_jobs[0].steps:
+                if step.conclusion != "success":
+                    job_result = 0
+                    break
 
         queue_time = started_at - created_at
         run_time = completed_at - started_at

clang-tools-extra/docs/ReleaseNotes.rst

@@ -108,22 +108,19 @@ Improvements to clang-query
 Improvements to clang-tidy
 --------------------------
 
-- Improved :program:`clang-tidy`'s `--verify-config` flag by adding support for
-  the configuration options of the `Clang Static Analyzer Checks
-  <https://clang.llvm.org/docs/analyzer/checkers.html>`_.
-
-- Improved :program:`clang-tidy` by accepting parameters file in command line.
-
 - Improved :program:`run-clang-tidy.py` script. Fixed minor shutdown noise
   happening on certain platforms when interrupting the script.
 
-- Improved :program:`clang-tidy` by fixing incorrect configuration file path
-  resolving when file paths contain ``..``.
+- Improved :program:`clang-tidy`:
 
-- Removed :program:`clang-tidy`'s global options for most of checks. All options
-  are changed to local options except `IncludeStyle`, `StrictMode` and
-  `IgnoreMacros`. Global scoped `StrictMode` and `IgnoreMacros` are deprecated
-  and will be removed in further releases.
+  - add support for `--verify-config` flag to check the configuration options of
+    the `Clang Static Analyzer Checks <https://clang.llvm.org/docs/analyzer/checkers.html>`_.
+  - accept parameters file in command line.
+  - fix incorrect configuration file path resolving when file paths contain ``..``.
+  - remove global options for most of checks. All options are changed to local
+    options except `IncludeStyle`, `StrictMode` and `IgnoreMacros`. Global scoped
+    `StrictMode` and `IgnoreMacros` are deprecated and will be removed in further
+    releases.
 
 .. csv-table::
   :header: "Check", "Options removed from global option"

clang/lib/Sema/SemaDecl.cpp

@@ -15855,8 +15855,8 @@ static void diagnoseImplicitlyRetainedSelf(Sema &S) {
   llvm::DenseMap<const BlockDecl *, bool> EscapeInfo;
 
   auto IsOrNestedInEscapingBlock = [&](const BlockDecl *BD) {
-    if (EscapeInfo.count(BD))
-      return EscapeInfo[BD];
+    if (auto It = EscapeInfo.find(BD); It != EscapeInfo.end())
+      return It->second;
 
     bool R = false;
     const BlockDecl *CurBD = BD;

clang/lib/Sema/SemaOpenMP.cpp

@@ -5400,11 +5400,13 @@ static bool checkNestingOfRegions(Sema &SemaRef, const DSAStackTy *Stack,
     Recommend = ShouldBeInTargetRegion;
   } else if (CurrentRegion == OMPD_scan) {
     if (SemaRef.LangOpts.OpenMP >= 50) {
-      // Make sure that the '-fopenmp-target-xteam-scan' flag is passed to
-      // enable the Xteam-Scan Codegen, if the 'scan' directive is found to be
-      // nested inside the 'target teams distribute parallel for' directive
+      // Make sure that one of the flags - '-fopenmp-target-xteam-scan' or
+      // '-fopenmp-target-xteam-no-loop-scan' flag is passed to enable the
+      // Xteam-Scan Codegen, if the 'scan' directive is found to be nested
+      // inside the 'target teams distribute parallel for' directive
       if (ParentRegion == OMPD_target_teams_distribute_parallel_for &&
-          !SemaRef.getLangOpts().OpenMPTargetXteamScan)
+          !(SemaRef.getLangOpts().OpenMPTargetXteamScan ||
+            SemaRef.getLangOpts().OpenMPTargetXteamNoLoopScan))
         SemaRef.Diag(StartLoc, diag::err_omp_xteam_scan_prohibited)
             << getOpenMPDirectiveName(CurrentRegion) << Recommend;
       // OpenMP spec 5.0 and 5.1 require scan to be directly enclosed by for,

clang/test/OpenMP/xteam_scan_codegen.cpp

@@ -2,8 +2,8 @@
 
 /*---------------------------------------- Test Xteam NoLoop Scan -----------------------------------------*/
 
-// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple x86_64-unknown-unknown -fopenmp-targets=amdgcn-amd-amdhsa -fopenmp-target-ignore-env-vars -fopenmp-assume-no-nested-parallelism -fopenmp-assume-no-thread-state -fopenmp-target-xteam-no-loop-scan -emit-llvm-bc %s -o %t-ppc-host1.bc
-// RUN: %clang_cc1 -target-cpu gfx90a -fopenmp -x c++ -std=c++11 -triple amdgcn-amd-amdhsa -fopenmp-targets=amdgcn-amd-amdhsa -fopenmp-target-ignore-env-vars -fopenmp-assume-no-nested-parallelism -fopenmp-assume-no-thread-state -fopenmp-target-xteam-no-loop-scan -emit-llvm %s -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host1.bc -o - | FileCheck %s --check-prefix=CHECK-64WAVE
+// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple x86_64-unknown-unknown -fopenmp-targets=amdgcn-amd-amdhsa -fopenmp-target-ignore-env-vars -fopenmp-assume-no-nested-parallelism -fopenmp-assume-no-thread-state -fopenmp-target-xteam-no-loop-scan -emit-llvm-bc %s -o %t-ppc-host1.bc 
+// RUN: %clang_cc1 -target-cpu gfx90a -fopenmp -x c++ -std=c++11 -triple amdgcn-amd-amdhsa -fopenmp-targets=amdgcn-amd-amdhsa -fopenmp-target-ignore-env-vars -fopenmp-assume-no-nested-parallelism -fopenmp-assume-no-thread-state -fopenmp-target-xteam-no-loop-scan -emit-llvm %s -fopenmp-is-device  -fopenmp-host-ir-file-path %t-ppc-host1.bc -o - | FileCheck %s --check-prefix=CHECK-64WAVE
 
 // RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple x86_64-unknown-unknown -fopenmp-targets=amdgcn-amd-amdhsa -fopenmp-target-ignore-env-vars -fopenmp-assume-no-nested-parallelism -fopenmp-assume-no-thread-state -fopenmp-target-xteam-no-loop-scan -emit-llvm-bc %s -DNUM_THREADS=512 -o %t-ppc-host2.bc
 // RUN: %clang_cc1 -target-cpu gfx90a -fopenmp -x c++ -std=c++11 -triple amdgcn-amd-amdhsa -fopenmp-targets=amdgcn-amd-amdhsa -fopenmp-target-ignore-env-vars -fopenmp-assume-no-nested-parallelism -fopenmp-assume-no-thread-state -fopenmp-target-xteam-no-loop-scan -emit-llvm %s -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host2.bc -DNUM_THREADS=512 -o - | FileCheck %s --check-prefix=CHECK-64WAVE-512WGSize

libc/cmake/modules/LLVMLibCHeaderRules.cmake

@@ -66,6 +66,7 @@ function(add_header target_name)
   set_target_properties(
     ${fq_target_name}
     PROPERTIES
+      HEADER_NAME ${dest_leaf_filename}
       HEADER_FILE_PATH ${dest_file}
       DEPS "${fq_deps_list}"
   )
@@ -164,6 +165,7 @@ function(add_gen_header target_name)
   set_target_properties(
     ${fq_target_name}
     PROPERTIES
+      HEADER_NAME ${ADD_GEN_HDR_GEN_HDR}
       HEADER_FILE_PATH ${out_file}
       DECLS_FILE_PATH "${decl_out_file}"
       DEPS "${fq_deps_list}"

libc/test/include/CMakeLists.txt

@@ -422,7 +422,7 @@ add_libc_test(
     -Werror
   DEPENDS
     libc.include.llvm-libc-macros.math_function_macros
-)  
+)
 
 add_libc_test(
   isfinite_c_test
@@ -483,3 +483,70 @@ add_libc_test(
   DEPENDS
     libc.include.llvm-libc-macros.math_function_macros
 )
+
+# Test `#include <...>` of each header in each available language mode.
+# This is gated on -DLLVM_LIBC_BUILD_HEADER_TESTS=ON until all the bugs
+# in headers are fixed so the tests all compile.
+set(TEST_STDC_VERSIONS 89;99;11;17;23)
+set(TEST_STDCXX_VERSIONS 03;11;14;17;20;23;26)
+
+function(add_header_test target_name source_file deps std_mode)
+  if(LLVM_LIBC_BUILD_HEADER_TESTS)
+    add_libc_test(
+      ${target_name}
+      C_TEST
+      HERMETIC_TEST_ONLY
+      SUITE
+	libc_include_tests
+      SRCS
+	${source_file}
+      COMPILE_OPTIONS
+	-Werror
+	-Wsystem-headers
+	-Wall
+	-Wextra
+	-std=${std_mode}
+      DEPENDS
+	${deps}
+    )
+  endif()
+endfunction()
+
+foreach(target ${TARGET_PUBLIC_HEADERS})
+  string(REPLACE "libc.include." "" header ${target})
+  get_target_property(HEADER_NAME ${target} HEADER_NAME)
+
+  set(test_stdc_file "${CMAKE_CURRENT_BINARY_DIR}/${header}_test.c")
+  configure_file(header-test-template.c ${test_stdc_file} @ONLY)
+  foreach(stdc_version ${TEST_STDC_VERSIONS})
+    add_header_test(
+      "${header}_c${stdc_version}_test"
+      ${test_stdc_file}
+      ${target}
+      "c${stdc_version}"
+    )
+    add_header_test(
+      "${header}_gnu${stdc_version}_test"
+      ${test_stdc_file}
+      ${target}
+      "gnu${stdc_version}"
+    )
+  endforeach()
+
+  set(test_stdcxx_file "${CMAKE_CURRENT_BINARY_DIR}/${header}_test.cpp")
+  configure_file(header-test-template.c ${test_stdcxx_file} @ONLY)
+  foreach(stdcxx_version ${TEST_STDCXX_VERSIONS})
+    add_header_test(
+      "${header}_cpp${stdcxx_version}_test"
+      ${test_stdcxx_file}
+      ${target}
+      "c++${stdcxx_version}"
+    )
+    add_header_test(
+      "${header}_gnucpp${stdcxx_version}_test"
+      ${test_stdcxx_file}
+      ${target}
+      "gnu++${stdcxx_version}"
+    )
+  endforeach()
+endforeach()

libc/test/include/header-test-template.c

@@ -0,0 +1,14 @@
+/*===-- Test for <@HEADER_NAME@> ----------------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDXList-License-Identifier: Apache-2.0 WITH LLVM-exception
+ */
+
+#include <@HEADER_NAME@>
+
+int main(int argc, char **argv) {
+  (void)argc;
+  (void)argv;
+  return 0;
+}

llvm/lib/Analysis/InstructionSimplify.cpp

@@ -871,6 +871,14 @@ static Value *simplifySubInst(Value *Op0, Value *Op1, bool IsNSW, bool IsNUW,
   if (Value *V = simplifyByDomEq(Instruction::Sub, Op0, Op1, Q, MaxRecurse))
     return V;
 
+  // (sub nuw C_Mask, (xor X, C_Mask)) -> X
+  if (IsNUW) {
+    Value *X;
+    if (match(Op1, m_Xor(m_Value(X), m_Specific(Op0))) &&
+        match(Op0, m_LowBitMask()))
+      return X;
+  }
+
   return nullptr;
 }
 
@@ -2540,6 +2548,14 @@ static Value *simplifyXorInst(Value *Op0, Value *Op1, const SimplifyQuery &Q,
   if (Value *V = simplifyByDomEq(Instruction::Xor, Op0, Op1, Q, MaxRecurse))
     return V;
 
+  // (xor (sub nuw C_Mask, X), C_Mask) -> X
+  {
+    Value *X;
+    if (match(Op0, m_NUWSub(m_Specific(Op1), m_Value(X))) &&
+        match(Op1, m_LowBitMask()))
+      return X;
+  }
+
   return nullptr;
 }
 

llvm/lib/Analysis/ValueTracking.cpp

@@ -9897,13 +9897,20 @@ static void setLimitsForBinOp(const BinaryOperator &BO, APInt &Lower,
   }
 }
 
-static ConstantRange getRangeForIntrinsic(const IntrinsicInst &II) {
+static ConstantRange getRangeForIntrinsic(const IntrinsicInst &II,
+                                          bool UseInstrInfo) {
   unsigned Width = II.getType()->getScalarSizeInBits();
   const APInt *C;
   switch (II.getIntrinsicID()) {
-  case Intrinsic::ctpop:
   case Intrinsic::ctlz:
-  case Intrinsic::cttz:
+  case Intrinsic::cttz: {
+    APInt Upper(Width, Width);
+    if (!UseInstrInfo || !match(II.getArgOperand(1), m_One()))
+      Upper += 1;
+    // Maximum of set/clear bits is the bit width.
+    return ConstantRange::getNonEmpty(APInt::getZero(Width), Upper);
+  }
+  case Intrinsic::ctpop:
     // Maximum of set/clear bits is the bit width.
     return ConstantRange::getNonEmpty(APInt::getZero(Width),
                                       APInt(Width, Width) + 1);
@@ -10094,7 +10101,7 @@ ConstantRange llvm::computeConstantRange(const Value *V, bool ForSigned,
     setLimitsForBinOp(*BO, Lower, Upper, IIQ, ForSigned);
     CR = ConstantRange::getNonEmpty(Lower, Upper);
   } else if (auto *II = dyn_cast<IntrinsicInst>(V))
-    CR = getRangeForIntrinsic(*II);
+    CR = getRangeForIntrinsic(*II, UseInstrInfo);
   else if (auto *SI = dyn_cast<SelectInst>(V)) {
     ConstantRange CRTrue = computeConstantRange(
         SI->getTrueValue(), ForSigned, UseInstrInfo, AC, CtxI, DT, Depth + 1);