merge main into amd-staging

xfails new test: Driver/XRay/xray-shared.cpp

Change-Id: I7f4cb89a3b12ed39ab2221191b6611d581b416fa

Author: ronlieb

clang/docs/ReleaseNotes.rst

@@ -449,6 +449,8 @@ Improvements to Clang's diagnostics
   name was a reserved name, which we improperly allowed to suppress the
   diagnostic.
 
+- Clang now diagnoses ``[[deprecated]]`` attribute usage on local variables (#GH90073).
+
 Improvements to Clang's time-trace
 ----------------------------------
 

clang/include/clang/Basic/CodeGenOptions.def

@@ -136,6 +136,8 @@ CODEGENOPT(XRayIgnoreLoops , 1, 0)
 ///< Emit the XRay function index section.
 CODEGENOPT(XRayFunctionIndex , 1, 1)
 
+///< Set when -fxray-shared is enabled
+CODEGENOPT(XRayShared , 1, 0)
 
 ///< Set the minimum number of instructions in a function to determine selective
 ///< XRay instrumentation.

clang/include/clang/Driver/Options.td

@@ -2953,6 +2953,11 @@ def fxray_selected_function_group :
   HelpText<"When using -fxray-function-groups, select which group of functions to instrument. Valid range is 0 to fxray-function-groups - 1">,
   MarshallingInfoInt<CodeGenOpts<"XRaySelectedFunctionGroup">, "0">;
 
+defm xray_shared : BoolFOption<"xray-shared",
+  CodeGenOpts<"XRayShared">, DefaultFalse,
+  PosFlag<SetTrue, [], [ClangOption, CC1Option],
+          "Enable shared library instrumentation with XRay">,
+  NegFlag<SetFalse>>;
 
 defm fine_grained_bitfield_accesses : BoolOption<"f", "fine-grained-bitfield-accesses",
   CodeGenOpts<"FineGrainedBitfieldAccesses">, DefaultFalse,
@@ -5512,9 +5517,9 @@ def mfrecipe : Flag<["-"], "mfrecipe">, Group<m_loongarch_Features_Group>,
 def mno_frecipe : Flag<["-"], "mno-frecipe">, Group<m_loongarch_Features_Group>,
   HelpText<"Disable frecipe.{s/d} and frsqrte.{s/d}">;
 def mlam_bh : Flag<["-"], "mlam-bh">, Group<m_loongarch_Features_Group>,
-  HelpText<"Enable amswap_[db].{b/h} and amadd_[db].{b/h}">;
+  HelpText<"Enable amswap[_db].{b/h} and amadd[_db].{b/h}">;
 def mno_lam_bh : Flag<["-"], "mno-lam-bh">, Group<m_loongarch_Features_Group>,
-  HelpText<"Disable amswap_[db].{b/h} and amadd_[db].{b/h}">;
+  HelpText<"Disable amswap[_db].{b/h} and amadd[_db].{b/h}">;
 def mannotate_tablejump : Flag<["-"], "mannotate-tablejump">, Group<m_loongarch_Features_Group>,
   HelpText<"Enable annotate table jump instruction to correlate it with the jump table.">;
 def mno_annotate_tablejump : Flag<["-"], "mno-annotate-tablejump">, Group<m_loongarch_Features_Group>,
@@ -6917,10 +6922,6 @@ def flang_deprecated_no_hlfir : Flag<["-"], "flang-deprecated-no-hlfir">,
   Flags<[HelpHidden]>, Visibility<[FlangOption, FC1Option]>,
   HelpText<"Do not use HLFIR lowering (deprecated)">;
 
-def flang_experimental_integer_overflow : Flag<["-"], "flang-experimental-integer-overflow">,
-  Flags<[HelpHidden]>, Visibility<[FlangOption, FC1Option]>,
-  HelpText<"Add nsw flag to internal operations such as do-variable increment (experimental)">;
-
 //===----------------------------------------------------------------------===//
 // FLangOption + CoreOption + NoXarchOption
 //===----------------------------------------------------------------------===//

clang/include/clang/Driver/XRayArgs.h

@@ -27,6 +27,7 @@ class XRayArgs {
   XRayInstrSet InstrumentationBundle;
   llvm::opt::Arg *XRayInstrument = nullptr;
   bool XRayRT = true;
+  bool XRayShared = false;
 
 public:
   /// Parses the XRay arguments from an argument list.
@@ -35,6 +36,7 @@ class XRayArgs {
                llvm::opt::ArgStringList &CmdArgs, types::ID InputType) const;
 
   bool needsXRayRt() const { return XRayInstrument && XRayRT; }
+  bool needsXRayDSORt() const { return XRayInstrument && XRayRT && XRayShared; }
   llvm::ArrayRef<std::string> modeList() const { return Modes; }
   XRayInstrSet instrumentationBundle() const { return InstrumentationBundle; }
 };

clang/lib/CodeGen/CGCall.cpp

@@ -5126,7 +5126,11 @@ RValue CodeGenFunction::EmitCall(const CGFunctionInfo &CallInfo,
   RawAddress SRetAlloca = RawAddress::invalid();
   llvm::Value *UnusedReturnSizePtr = nullptr;
   if (RetAI.isIndirect() || RetAI.isInAlloca() || RetAI.isCoerceAndExpand()) {
-    if (IsVirtualFunctionPointerThunk && RetAI.isIndirect()) {
+    // For virtual function pointer thunks and musttail calls, we must always
+    // forward an incoming SRet pointer to the callee, because a local alloca
+    // would be de-allocated before the call. These cases both guarantee that
+    // there will be an incoming SRet argument of the correct type.
+    if ((IsVirtualFunctionPointerThunk || IsMustTail) && RetAI.isIndirect()) {
       SRetPtr = makeNaturalAddressForPointer(CurFn->arg_begin() +
                                                  IRFunctionArgs.getSRetArgNo(),
                                              RetTy, CharUnits::fromQuantity(1));

clang/lib/CodeGen/CGOpenMPRuntime.cpp

@@ -1214,6 +1214,7 @@ struct PushAndPopStackRAII {
       CodeGenFunction::JumpDest Dest =
           CGF.getOMPCancelDestination(OMPD_parallel);
       CGF.EmitBranchThroughCleanup(Dest);
+      return llvm::Error::success();
     };
 
     // TODO: Remove this once we emit parallel regions through the
@@ -2355,8 +2356,11 @@ void CGOpenMPRuntime::emitBarrierCall(CodeGenFunction &CGF, SourceLocation Loc,
   auto *OMPRegionInfo =
       dyn_cast_or_null<CGOpenMPRegionInfo>(CGF.CapturedStmtInfo);
   if (CGF.CGM.getLangOpts().OpenMPIRBuilder) {
-    CGF.Builder.restoreIP(OMPBuilder.createBarrier(
-        CGF.Builder, Kind, ForceSimpleCall, EmitChecks));
+    llvm::OpenMPIRBuilder::InsertPointOrErrorTy AfterIP =
+        OMPBuilder.createBarrier(CGF.Builder, Kind, ForceSimpleCall,
+                                 EmitChecks);
+    assert(AfterIP && "unexpected error creating barrier");
+    CGF.Builder.restoreIP(*AfterIP);
     return;
   }
 
@@ -5977,8 +5981,10 @@ void CGOpenMPRuntime::emitTargetOutlinedFunctionHelper(
             /*CanHaveMultiDeviceArgs*/ true, /*IsTopKernel*/ true);
       };
 
-  OMPBuilder.emitTargetRegionFunction(EntryInfo, GenerateOutlinedFunction,
-                                      IsOffloadEntry, OutlinedFn, OutlinedFnID);
+  llvm::Error Err = OMPBuilder.emitTargetRegionFunction(
+      EntryInfo, GenerateOutlinedFunction, IsOffloadEntry, OutlinedFn,
+      OutlinedFnID);
+  assert(!Err && "unexpected error creating target region");
 
   if (!OutlinedFn)
     return;
@@ -9970,9 +9976,12 @@ static void emitTargetCallKernelLaunch(
         NumTargetItems, RTArgs, NumIterations, NumTeams, NumThreads,
         DynCGGroupMem, HasNoWait);
 
-    CGF.Builder.restoreIP(OMPRuntime->getOMPBuilder().emitKernelLaunch(
-        CGF.Builder, OutlinedFnID, EmitTargetCallFallbackCB, Args, DeviceID,
-        RTLoc, AllocaIP));
+    llvm::OpenMPIRBuilder::InsertPointOrErrorTy AfterIP =
+        OMPRuntime->getOMPBuilder().emitKernelLaunch(
+            CGF.Builder, OutlinedFnID, EmitTargetCallFallbackCB, Args, DeviceID,
+            RTLoc, AllocaIP);
+    assert(AfterIP && "unexpected error creating kernel launch");
+    CGF.Builder.restoreIP(*AfterIP);
   };
 
   if (RequiresOuterTask) {
@@ -10668,9 +10677,12 @@ void CGOpenMPRuntime::emitTargetDataCalls(
   InsertPointTy CodeGenIP(CGF.Builder.GetInsertBlock(),
                           CGF.Builder.GetInsertPoint());
   llvm::OpenMPIRBuilder::LocationDescription OmpLoc(CodeGenIP);
-  CGF.Builder.restoreIP(OMPBuilder.createTargetData(
-      OmpLoc, AllocaIP, CodeGenIP, DeviceID, IfCondVal, Info, GenMapInfoCB,
-      /*MapperFunc=*/nullptr, BodyCB, DeviceAddrCB, CustomMapperCB, RTLoc));
+  llvm::OpenMPIRBuilder::InsertPointOrErrorTy AfterIP =
+      OMPBuilder.createTargetData(
+          OmpLoc, AllocaIP, CodeGenIP, DeviceID, IfCondVal, Info, GenMapInfoCB,
+          /*MapperFunc=*/nullptr, BodyCB, DeviceAddrCB, CustomMapperCB, RTLoc);
+  assert(AfterIP && "unexpected error creating target data");
+  CGF.Builder.restoreIP(*AfterIP);
 }
 
 void CGOpenMPRuntime::emitTargetDataStandAloneCall(

clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp

@@ -1984,11 +1984,14 @@ void CGOpenMPRuntimeGPU::emitReduction(
     Idx++;
   }
 
-  CGF.Builder.restoreIP(OMPBuilder.createReductionsGPU(
-      OmpLoc, AllocaIP, CodeGenIP, ReductionInfos, false, TeamsReduction,
-      DistributeReduction, llvm::OpenMPIRBuilder::ReductionGenCBKind::Clang,
-      CGF.getTarget().getGridValue(), C.getLangOpts().OpenMPCUDAReductionBufNum,
-      RTLoc));
+  llvm::OpenMPIRBuilder::InsertPointOrErrorTy AfterIP =
+      OMPBuilder.createReductionsGPU(
+          OmpLoc, AllocaIP, CodeGenIP, ReductionInfos, false, TeamsReduction,
+          DistributeReduction, llvm::OpenMPIRBuilder::ReductionGenCBKind::Clang,
+          CGF.getTarget().getGridValue(),
+          C.getLangOpts().OpenMPCUDAReductionBufNum, RTLoc);
+  assert(AfterIP && "unexpected error creating GPU reductions");
+  CGF.Builder.restoreIP(*AfterIP);
   return;
 }