merge main into amd-staging (llvm#3195)

Author: z1-cciauto

clang-tools-extra/docs/clang-tidy/Contributing.rst

@@ -657,6 +657,29 @@ directory.  The path to this directory is available in a lit test with the varia
 .. _FileCheck: https://llvm.org/docs/CommandGuide/FileCheck.html
 .. _test/clang-tidy/checkers/google/readability-casting.cpp: https://github.com/llvm/llvm-project/blob/main/clang-tools-extra/test/clang-tidy/checkers/google/readability-casting.cpp
 
+
+Submitting a Pull Request
+-------------------------
+
+Before submitting a pull request, contributors are encouraged to run
+:program:`clang-tidy` and :program:`clang-format` on their changes to ensure
+code quality and catch potential issues. While :program:`clang-tidy` is not
+currently enforced in CI, following this practice helps maintain code
+consistency and prevent common errors.
+
+Here's useful command to check your staged changes:
+
+.. code-block:: console
+
+  $ git diff --staged -U0 | ./clang-tools-extra/clang-tidy/tool/clang-tidy-diff.py \
+      -j $(nproc) -path build/ -p1 -only-check-in-db
+  $ git clang-format
+
+Note that some warnings may be false positives or require careful consideration
+before fixing. Use your judgment and feel free to discuss in the pull request
+if you're unsure about a particular warning.
+
+
 Out-of-tree check plugins
 -------------------------
 

clang/include/clang/Basic/BuiltinsAMDGPU.def

@@ -675,6 +675,7 @@ TARGET_BUILTIN(__builtin_amdgcn_s_wait_asynccnt, "vIUs", "n", "gfx1250-insts")
 TARGET_BUILTIN(__builtin_amdgcn_s_wait_tensorcnt, "vIUs", "n", "gfx1250-insts")
 
 TARGET_BUILTIN(__builtin_amdgcn_tanhf, "ff", "nc", "tanh-insts")
+TARGET_BUILTIN(__builtin_amdgcn_tanhh, "hh", "nc", "tanh-insts")
 TARGET_BUILTIN(__builtin_amdgcn_tanh_bf16, "yy", "nc", "bf16-trans-insts")
 TARGET_BUILTIN(__builtin_amdgcn_rcp_bf16, "yy", "nc", "bf16-trans-insts")
 TARGET_BUILTIN(__builtin_amdgcn_rsq_bf16, "yy", "nc", "bf16-trans-insts")

clang/lib/CodeGen/TargetBuiltins/AMDGPU.cpp

@@ -516,6 +516,7 @@ Value *CodeGenFunction::EmitAMDGPUBuiltinExpr(unsigned BuiltinID,
     return Builder.CreateCall(F, { Src });
   }
   case AMDGPU::BI__builtin_amdgcn_tanhf:
+  case AMDGPU::BI__builtin_amdgcn_tanhh:
   case AMDGPU::BI__builtin_amdgcn_tanh_bf16:
     return emitBuiltinWithOneOverloadedType<1>(*this, E,
                                                Intrinsic::amdgcn_tanh);

clang/lib/Format/TokenAnnotator.cpp

@@ -2996,14 +2996,18 @@ class AnnotatingParser {
     const FormatToken *PrevToken = Tok.getPreviousNonComment();
     if (!PrevToken)
       return TT_UnaryOperator;
-    if (PrevToken->is(TT_TypeName))
+    if (PrevToken->isTypeName(LangOpts))
       return TT_PointerOrReference;
     if (PrevToken->isPlacementOperator() && Tok.is(tok::ampamp))
       return TT_BinaryOperator;
 
-    const FormatToken *NextToken = Tok.getNextNonComment();
+    auto *NextToken = Tok.getNextNonComment();
     if (!NextToken)
       return TT_PointerOrReference;
+    if (NextToken->is(tok::greater)) {
+      NextToken->setFinalizedType(TT_TemplateCloser);
+      return TT_PointerOrReference;
+    }
 
     if (InTemplateArgument && NextToken->is(tok::kw_noexcept))
       return TT_BinaryOperator;
@@ -3112,7 +3116,7 @@ class AnnotatingParser {
 
     // It's more likely that & represents operator& than an uninitialized
     // reference.
-    if (Tok.is(tok::amp) && PrevToken && PrevToken->Tok.isAnyIdentifier() &&
+    if (Tok.is(tok::amp) && PrevToken->Tok.isAnyIdentifier() &&
         IsChainedOperatorAmpOrMember(PrevToken->getPreviousNonComment()) &&
         NextToken && NextToken->Tok.isAnyIdentifier()) {
       if (auto NextNext = NextToken->getNextNonComment();

clang/test/CodeGenOpenCL/builtins-amdgcn-gfx1250.cl

@@ -61,6 +61,26 @@ void test_tanh_f32(global float* out, float a)
   *out = __builtin_amdgcn_tanhf(a);
 }
 
+// CHECK-LABEL: @test_tanh_f16(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[OUT_ADDR:%.*]] = alloca ptr addrspace(1), align 8, addrspace(5)
+// CHECK-NEXT:    [[A_ADDR:%.*]] = alloca ptr addrspace(1), align 8, addrspace(5)
+// CHECK-NEXT:    [[OUT_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[OUT_ADDR]] to ptr
+// CHECK-NEXT:    [[A_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[A_ADDR]] to ptr
+// CHECK-NEXT:    store ptr addrspace(1) [[OUT:%.*]], ptr [[OUT_ADDR_ASCAST]], align 8
+// CHECK-NEXT:    store ptr addrspace(1) [[A:%.*]], ptr [[A_ADDR_ASCAST]], align 8
+// CHECK-NEXT:    [[TMP0:%.*]] = load ptr addrspace(1), ptr [[A_ADDR_ASCAST]], align 8
+// CHECK-NEXT:    [[TMP1:%.*]] = load half, ptr addrspace(1) [[TMP0]], align 2
+// CHECK-NEXT:    [[TMP2:%.*]] = call half @llvm.amdgcn.tanh.f16(half [[TMP1]])
+// CHECK-NEXT:    [[TMP3:%.*]] = load ptr addrspace(1), ptr [[OUT_ADDR_ASCAST]], align 8
+// CHECK-NEXT:    store half [[TMP2]], ptr addrspace(1) [[TMP3]], align 2
+// CHECK-NEXT:    ret void
+//
+void test_tanh_f16(global half* out, global half* a)
+{
+  *out = __builtin_amdgcn_tanhh(*a);
+}
+
 // CHECK-LABEL: @test_tanh_bf16(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[OUT_ADDR:%.*]] = alloca ptr addrspace(1), align 8, addrspace(5)

clang/unittests/Format/TokenAnnotatorTest.cpp

@@ -390,6 +390,10 @@ TEST_F(TokenAnnotatorTest, UnderstandsUsesOfStarAndAmp) {
   EXPECT_TOKEN(Tokens[20], tok::l_brace, TT_CompoundRequirementLBrace);
   EXPECT_TOKEN(Tokens[22], tok::star, TT_BinaryOperator);
 
+  Tokens = annotate("bool foo = requires { static_cast<Foo &&>(1); };");
+  ASSERT_EQ(Tokens.size(), 17u) << Tokens;
+  EXPECT_TOKEN(Tokens[8], tok::ampamp, TT_PointerOrReference);
+
   Tokens = annotate("return s.operator int *();");
   ASSERT_EQ(Tokens.size(), 10u) << Tokens;
   // Not TT_FunctionDeclarationName.

flang/lib/Optimizer/CodeGen/CodeGen.cpp

@@ -1123,6 +1123,16 @@ struct AllocMemOpConversion : public fir::FIROpConversion<fir::AllocMemOp> {
     for (mlir::Value opnd : adaptor.getOperands())
       size = rewriter.create<mlir::LLVM::MulOp>(
           loc, ity, size, integerCast(loc, rewriter, ity, opnd));
+
+    // As the return value of malloc(0) is implementation defined, allocate one
+    // byte to ensure the allocation status being true. This behavior aligns to
+    // what the runtime has.
+    mlir::Value zero = genConstantIndex(loc, ity, rewriter, 0);
+    mlir::Value one = genConstantIndex(loc, ity, rewriter, 1);
+    mlir::Value cmp = rewriter.create<mlir::LLVM::ICmpOp>(
+        loc, mlir::LLVM::ICmpPredicate::sgt, size, zero);
+    size = rewriter.create<mlir::LLVM::SelectOp>(loc, cmp, size, one);
+
     auto mallocTyWidth = lowerTy().getIndexTypeBitwidth();
     auto mallocTy =
         mlir::IntegerType::get(rewriter.getContext(), mallocTyWidth);

flang/test/Fir/alloc-32.fir

@@ -20,7 +20,9 @@ func.func @allocmem_scalar_nonchar() -> !fir.heap<i32> {
 // CHECK-SAME: i32 %[[len:.*]])
 // CHECK: %[[mul1:.*]] = sext i32 %[[len]] to i64
 // CHECK: %[[mul2:.*]] = mul i64 1, %[[mul1]]
-// CHECK: %[[trunc:.*]] = trunc i64 %[[mul2]] to i32
+// CHECK: %[[cmp:.*]] = icmp sgt i64 %[[mul2]], 0
+// CHECK: %[[sz:.*]] = select i1 %[[cmp]], i64 %[[mul2]], i64 1
+// CHECK: %[[trunc:.*]] = trunc i64 %[[sz]] to i32
 // CHECK: call ptr @malloc(i32 %[[trunc]])
 func.func @allocmem_scalar_dynchar(%l : i32) -> !fir.heap<!fir.char<1,?>> {
   %1 = fir.allocmem !fir.char<1,?>(%l : i32)

flang/test/Fir/alloc.fir

@@ -87,7 +87,9 @@ func.func @alloca_scalar_dynchar_kind(%l : i32) -> !fir.ref<!fir.char<2,?>> {
 // CHECK-SAME: i32 %[[len:.*]])
 // CHECK: %[[mul1:.*]] = sext i32 %[[len]] to i64
 // CHECK: %[[mul2:.*]] = mul i64 1, %[[mul1]]
-// CHECK: call ptr @malloc(i64 %[[mul2]])
+// CHECK: %[[cmp:.*]] = icmp sgt i64 %[[mul2]], 0
+// CHECK: %[[size:.*]] = select i1 %[[cmp]], i64 %[[mul2]], i64 1
+// CHECK: call ptr @malloc(i64 %[[size]])
 func.func @allocmem_scalar_dynchar(%l : i32) -> !fir.heap<!fir.char<1,?>> {
   %1 = fir.allocmem !fir.char<1,?>(%l : i32)
   return %1 : !fir.heap<!fir.char<1,?>>
@@ -97,7 +99,9 @@ func.func @allocmem_scalar_dynchar(%l : i32) -> !fir.heap<!fir.char<1,?>> {
 // CHECK-SAME: i32 %[[len:.*]])
 // CHECK: %[[mul1:.*]] = sext i32 %[[len]] to i64
 // CHECK: %[[mul2:.*]] = mul i64 2, %[[mul1]]
-// CHECK: call ptr @malloc(i64 %[[mul2]])
+// CHECK: %[[cmp:.*]] = icmp sgt i64 %[[mul2]], 0
+// CHECK: %[[size:.*]] = select i1 %[[cmp]], i64 %[[mul2]], i64 1
+// CHECK: call ptr @malloc(i64 %[[size]])
 func.func @allocmem_scalar_dynchar_kind(%l : i32) -> !fir.heap<!fir.char<2,?>>{
   %1 = fir.allocmem !fir.char<2,?>(%l : i32)
   return %1 : !fir.heap<!fir.char<2,?>>
@@ -152,7 +156,9 @@ func.func @allocmem_array_of_char() -> !fir.heap<!fir.array<3x3x!fir.char<1,10>>
 // CHECK-SAME: i32 %[[len:.*]])
 // CHECK: %[[mul1:.*]] = sext i32 %[[len]] to i64
 // CHECK: %[[mul2:.*]] = mul i64 9, %[[mul1]]
-// CHECK: call ptr @malloc(i64 %[[mul2]])
+// CHECK: %[[cmp:.*]] = icmp sgt i64 %[[mul2]], 0
+// CHECK: %[[size:.*]] = select i1 %[[cmp]], i64 %[[mul2]], i64 1
+// CHECK: call ptr @malloc(i64 %[[size]])
 func.func @allocmem_array_of_dynchar(%l: i32) -> !fir.heap<!fir.array<3x3x!fir.char<1,?>>> {
   %1 = fir.allocmem !fir.array<3x3x!fir.char<1,?>>(%l : i32)
   return %1 : !fir.heap<!fir.array<3x3x!fir.char<1,?>>>
@@ -180,7 +186,9 @@ func.func @alloca_dynarray_of_nonchar2(%e: index) -> !fir.ref<!fir.array<?x?xi32
 // CHECK-LABEL: define ptr @allocmem_dynarray_of_nonchar(
 // CHECK-SAME: i64 %[[extent:.*]])
 // CHECK: %[[prod1:.*]] = mul i64 12, %[[extent]]
-// CHECK: call ptr @malloc(i64 %[[prod1]])
+// CHECK: %[[cmp:.*]] = icmp sgt i64 %[[prod1]], 0
+// CHECK: %[[size:.*]] = select i1 %[[cmp]], i64 %[[prod1]], i64 1
+// CHECK: call ptr @malloc(i64 %[[size]])
 func.func @allocmem_dynarray_of_nonchar(%e: index) -> !fir.heap<!fir.array<3x?xi32>> {
   %1 = fir.allocmem !fir.array<3x?xi32>, %e
   return %1 : !fir.heap<!fir.array<3x?xi32>>
@@ -190,7 +198,9 @@ func.func @allocmem_dynarray_of_nonchar(%e: index) -> !fir.heap<!fir.array<3x?xi
 // CHECK-SAME: i64 %[[extent:.*]])
 // CHECK: %[[prod1:.*]] = mul i64 4, %[[extent]]
 // CHECK: %[[prod2:.*]] = mul i64 %[[prod1]], %[[extent]]
-// CHECK: call ptr @malloc(i64 %[[prod2]])
+// CHECK: %[[cmp:.*]] = icmp sgt i64 %[[prod2]], 0
+// CHECK: %[[size:.*]] = select i1 %[[cmp]], i64 %[[prod2]], i64 1
+// CHECK: call ptr @malloc(i64 %[[size]])
 func.func @allocmem_dynarray_of_nonchar2(%e: index) -> !fir.heap<!fir.array<?x?xi32>> {
   %1 = fir.allocmem !fir.array<?x?xi32>, %e, %e
   return %1 : !fir.heap<!fir.array<?x?xi32>>
@@ -218,7 +228,9 @@ func.func @alloca_dynarray_of_char2(%e : index) -> !fir.ref<!fir.array<?x?x!fir.
 // CHECK-LABEL: define ptr @allocmem_dynarray_of_char(
 // CHECK-SAME: i64 %[[extent:.*]])
 // CHECK: %[[prod1:.*]] = mul i64 60, %[[extent]]
-// CHECK: call ptr @malloc(i64 %[[prod1]])
+// CHECK: %[[cmp:.*]] = icmp sgt i64 %[[prod1]], 0
+// CHECK: %[[size:.*]] = select i1 %[[cmp]], i64 %[[prod1]], i64 1
+// CHECK: call ptr @malloc(i64 %[[size]])
 func.func @allocmem_dynarray_of_char(%e : index) -> !fir.heap<!fir.array<3x?x!fir.char<2,10>>> {
   %1 = fir.allocmem !fir.array<3x?x!fir.char<2,10>>, %e
   return %1 : !fir.heap<!fir.array<3x?x!fir.char<2,10>>>
@@ -228,7 +240,9 @@ func.func @allocmem_dynarray_of_char(%e : index) -> !fir.heap<!fir.array<3x?x!fi
 // CHECK-SAME: i64 %[[extent:.*]])
 // CHECK: %[[prod1:.*]] = mul i64 20, %[[extent]]
 // CHECK: %[[prod2:.*]] = mul i64 %[[prod1]], %[[extent]]
-// CHECK: call ptr @malloc(i64 %[[prod2]])
+// CHECK: %[[cmp:.*]] = icmp sgt i64 %[[prod2]], 0
+// CHECK: %[[size:.*]] = select i1 %[[cmp]], i64 %[[mul2]], i64 1
+// CHECK: call ptr @malloc(i64 %[[size]])
 func.func @allocmem_dynarray_of_char2(%e : index) -> !fir.heap<!fir.array<?x?x!fir.char<2,10>>> {
   %1 = fir.allocmem !fir.array<?x?x!fir.char<2,10>>, %e, %e
   return %1 : !fir.heap<!fir.array<?x?x!fir.char<2,10>>>
@@ -261,7 +275,9 @@ func.func @alloca_dynarray_of_dynchar2(%l: i32, %e : index) -> !fir.ref<!fir.arr
 // CHECK: %[[prod1:.*]] = sext i32 %[[len]] to i64
 // CHECK: %[[prod2:.*]] = mul i64 6, %[[prod1]]
 // CHECK: %[[prod3:.*]] = mul i64 %[[prod2]], %[[extent]]
-// CHECK: call ptr @malloc(i64 %[[prod3]])
+// CHECK: %[[cmp:.*]] = icmp sgt i64 %[[prod3]], 0
+// CHECK: %[[size:.*]] = select i1 %[[cmp]], i64 %[[prod3]], i64 1
+// CHECK: call ptr @malloc(i64 %[[size]])
 func.func @allocmem_dynarray_of_dynchar(%l: i32, %e : index) -> !fir.heap<!fir.array<3x?x!fir.char<2,?>>> {
   %1 = fir.allocmem !fir.array<3x?x!fir.char<2,?>>(%l : i32), %e
   return %1 : !fir.heap<!fir.array<3x?x!fir.char<2,?>>>
@@ -273,7 +289,9 @@ func.func @allocmem_dynarray_of_dynchar(%l: i32, %e : index) -> !fir.heap<!fir.a
 // CHECK: %[[prod1:.*]] = mul i64 2, %[[a]]
 // CHECK: %[[prod2:.*]] = mul i64 %[[prod1]], %[[extent]]
 // CHECK: %[[prod3:.*]] = mul i64 %[[prod2]], %[[extent]]
-// CHECK: call ptr @malloc(i64 %[[prod3]])
+// CHECK: %[[cmp:.*]] = icmp sgt i64 %[[prod3]], 0
+// CHECK: %[[size:.*]] = select i1 %[[cmp]], i64 %[[prod3]], i64 1
+// CHECK: call ptr @malloc(i64 %[[size]])
 func.func @allocmem_dynarray_of_dynchar2(%l: i32, %e : index) -> !fir.heap<!fir.array<?x?x!fir.char<2,?>>> {
   %1 = fir.allocmem !fir.array<?x?x!fir.char<2,?>>(%l : i32), %e, %e
   return %1 : !fir.heap<!fir.array<?x?x!fir.char<2,?>>>
@@ -312,7 +330,9 @@ func.func @alloca_array_with_holes_dynchar(%arg0: index, %arg1: index) -> !fir.r
 // CHECK-SAME: i64 %[[e1:.*]], i64 %[[e2:.*]])
 // CHECK: %[[a:.*]] = mul i64 240, %[[e1]]
 // CHECK: %[[b:.*]] = mul i64 %3, %[[e2]]
-// CHECK: call ptr @malloc(i64 %[[b]])
+// CHECK: %[[cmp:.*]] = icmp sgt i64 %[[b]], 0
+// CHECK: %[[size:.*]] = select i1 %[[cmp]], i64 %[[b]], i64 1
+// CHECK: call ptr @malloc(i64 %[[size]])
 func.func @allocmem_array_with_holes_nonchar(%0 : index, %1 : index) -> !fir.heap<!fir.array<4x?x3x?x5xi32>> {
   %a = fir.allocmem !fir.array<4x?x3x?x5xi32>, %0, %1
   return %a : !fir.heap<!fir.array<4x?x3x?x5xi32>>
@@ -321,7 +341,9 @@ func.func @allocmem_array_with_holes_nonchar(%0 : index, %1 : index) -> !fir.hea
 // CHECK-LABEL: define ptr @allocmem_array_with_holes_char(
 // CHECK-SAME: i64 %[[e:.*]])
 // CHECK: %[[mul:.*]] = mul i64 240, %[[e]]
-// CHECK: call ptr @malloc(i64 %[[mul]])
+// CHECK: %[[cmp:.*]] = icmp sgt i64 %[[mul]], 0
+// CHECK: %[[size:.*]] = select i1 %[[cmp]], i64 %[[mul]], i64 1
+// CHECK: call ptr @malloc(i64 %[[size]])
 func.func @allocmem_array_with_holes_char(%e: index) -> !fir.heap<!fir.array<3x?x4x!fir.char<2,10>>> {
   %1 = fir.allocmem !fir.array<3x?x4x!fir.char<2,10>>, %e
   return %1 : !fir.heap<!fir.array<3x?x4x!fir.char<2,10>>>
@@ -331,7 +353,9 @@ func.func @allocmem_array_with_holes_char(%e: index) -> !fir.heap<!fir.array<3x?
 // CHECK-SAME: i64 %[[len:.*]], i64 %[[extent:.*]])
 // CHECK: %[[a:.*]] = mul i64 24, %[[len]]
 // CHECK: %[[b:.*]] = mul i64 %[[a]], %[[extent]]
-// CHECK: call ptr @malloc(i64 %[[b]])
+// CHECK: %[[cmp:.*]] = icmp sgt i64 %[[b]], 0
+// CHECK: %[[size:.*]] = select i1 %[[cmp]], i64 %[[b]], i64 1
+// CHECK: call ptr @malloc(i64 %[[size]])
 func.func @allocmem_array_with_holes_dynchar(%arg0: index, %arg1: index) -> !fir.heap<!fir.array<3x?x4x!fir.char<2,?>>> {
   %1 = fir.allocmem !fir.array<3x?x4x!fir.char<2,?>>(%arg0 : index), %arg1
   return %1 : !fir.heap<!fir.array<3x?x4x!fir.char<2,?>>>

flang/test/Fir/arrexp.fir

@@ -146,7 +146,9 @@ func.func @f6(%arg0: !fir.box<!fir.array<?xf32>>, %arg1: f32) {
   // CHECK: %[[EXT_GEP:.*]] = getelementptr {{.*}} %[[A]], i32 0, i32 7, i64 0, i32 1
   // CHECK: %[[EXTENT:.*]] = load i64, ptr %[[EXT_GEP]]
   // CHECK: %[[SIZE:.*]] = mul i64 4, %[[EXTENT]]
-  // CHECK: %[[MALLOC:.*]] = call ptr @malloc(i64 %[[SIZE]])
+  // CHECK: %[[CMP:.*]] = icmp sgt i64 %[[SIZE]], 0
+  // CHECK: %[[SZ:.*]] = select i1 %[[CMP]], i64 %[[SIZE]], i64 1
+  // CHECK: %[[MALLOC:.*]] = call ptr @malloc(i64 %[[SZ]])
   %1 = fir.slice %c2, %c10, %c1 : (index, index, index) -> !fir.slice<1>
   %2 = fir.array_load %arg0 [%1] : (!fir.box<!fir.array<?xf32>>, !fir.slice<1>) -> !fir.array<?xf32>
   %3 = fir.slice %c1, %c9, %c1 : (index, index, index) -> !fir.slice<1>