[OpenCL] Add image type handling for builtins

svenvh · svenvh · commit 988f1e3e32a9 · 2019-09-05T10:01:24.000Z
Image types were previously available, but not working. This patch adds image type handling. Rename the image type definitions in the .td file to make them consistent with other type names. Use abstract types to represent the unqualified types. Instantiate access-qualified image types at the point of use using, e.g. `ImageType<Image2d, "RO">`. Add/update TableGen definitions for the read_image/write_image builtin functions. Patch by Pierre Gondois and Sven van Haastregt. Differential Revision: https://reviews.llvm.org/D63480 llvm-svn: 371046
diff --git a/clang/lib/Sema/OpenCLBuiltins.td b/clang/lib/Sema/OpenCLBuiltins.td
@@ -87,11 +87,11 @@ class Type<string _Name, QualType _QTName> {
 // OpenCL vector types (e.g. int2, int3, int16, float8, ...).
 class VectorType<Type _Ty, int _VecWidth> : Type<_Ty.Name, _Ty.QTName> {
   let VecWidth = _VecWidth;
+  let AccessQualifier = "";
   // Inherited fields
   let IsPointer = _Ty.IsPointer;
   let IsConst = _Ty.IsConst;
   let IsVolatile = _Ty.IsVolatile;
-  let AccessQualifier = _Ty.AccessQualifier;
   let AddrSpace = _Ty.AddrSpace;
 }
 
@@ -129,10 +129,16 @@ class VolatileType<Type _Ty> : Type<_Ty.Name, _Ty.QTName> {
   let AddrSpace = _Ty.AddrSpace;
 }
 
-// OpenCL image types (e.g. image2d_t, ...)
-class ImageType<Type _Ty, QualType _QTName, string _AccessQualifier> :
-                                              Type<_Ty.Name, _QTName> {
+// OpenCL image types (e.g. image2d).
+class ImageType<Type _Ty, string _AccessQualifier> :
+          Type<_Ty.Name, QualType<_Ty.QTName.Name#_AccessQualifier#"Ty", 0>> {
+  let VecWidth = 0;
   let AccessQualifier = _AccessQualifier;
+  // Inherited fields
+  let IsPointer = _Ty.IsPointer;
+  let IsConst = _Ty.IsConst;
+  let IsVolatile = _Ty.IsVolatile;
+  let AddrSpace = _Ty.AddrSpace;
 }
 
 // List of Types.
@@ -221,37 +227,21 @@ def Void      : Type<"void_t",    QualType<"VoidTy">>;
 // OpenCL v1.0/1.2/2.0 s6.1.2: Built-in Vector Data Types.
 // Built-in vector data types are created by TableGen's OpenCLBuiltinEmitter.
 
-// OpenCL v1.2 s6.1.3: Other Built-in Data Types
-// These definitions with a "null" name are "abstract". They should not
-// be used in definitions of Builtin functions.
-def image2d_t         : Type<"image2d_t", QualType<"null", 1>>;
-def image3d_t         : Type<"image3d_t", QualType<"null", 1>>;
-def image2d_array_t   : Type<"image2d_array_t", QualType<"null", 1>>;
-def image1d_t         : Type<"image1d_t", QualType<"null", 1>>;
-def image1d_buffer_t  : Type<"image1d_buffer_t", QualType<"null", 1>>;
-def image1d_array_t   : Type<"image1d_array_t", QualType<"null", 1>>;
-// Unlike the few functions above, the following definitions can be used
-// in definitions of Builtin functions (they have a QualType with a name).
-foreach v = ["RO", "WO", "RW"] in {
-  def image2d_#v#_t       : ImageType<image2d_t,
-                                      QualType<"OCLImage2d"#v#"Ty">,
-                                      v>;
-  def image3d_#v#_t       : ImageType<image3d_t,
-                                      QualType<"OCLImage3d"#v#"Ty">,
-                                      v>;
-  def image2d_array#v#_t  : ImageType<image2d_array_t,
-                                      QualType<"OCLImage2dArray"#v#"Ty">,
-                                      v>;
-  def image1d_#v#_t       : ImageType<image1d_t,
-                                      QualType<"OCLImage1d"#v#"Ty">,
-                                      v>;
-  def image1d_buffer#v#_t : ImageType<image1d_buffer_t,
-                                      QualType<"OCLImage1dBuffer"#v#"Ty">,
-                                      v>;
-  def image1d_array#v#_t  : ImageType<image1d_array_t,
-                                      QualType<"OCLImage1dArray"#v#"Ty">,
-                                      v>;
-}
+// OpenCL v1.0/1.2/2.0 s6.1.3: Other Built-in Data Types.
+// The image definitions are "abstract".  They should not be used without
+// specifying an access qualifier (RO/WO/RW).
+def Image1d               : Type<"Image1d", QualType<"OCLImage1d", 1>>;
+def Image2d               : Type<"Image2d", QualType<"OCLImage2d", 1>>;
+def Image3d               : Type<"Image3d", QualType<"OCLImage3d", 1>>;
+def Image1dArray          : Type<"Image1dArray", QualType<"OCLImage1dArray", 1>>;
+def Image1dBuffer         : Type<"Image1dBuffer", QualType<"OCLImage1dBuffer", 1>>;
+def Image2dArray          : Type<"Image2dArray", QualType<"OCLImage2dArray", 1>>;
+def Image2dDepth          : Type<"Image2dDepth", QualType<"OCLImage2dDepth", 1>>;
+def Image2dArrayDepth     : Type<"Image2dArrayDepth", QualType<"OCLImage2dArrayDepth", 1>>;
+def Image2dMsaa           : Type<"Image2dMsaa", QualType<"OCLImage2dMSAA", 1>>;
+def Image2dArrayMsaa      : Type<"Image2dArrayMsaa", QualType<"OCLImage2dArrayMSAA", 1>>;
+def Image2dMsaaDepth      : Type<"Image2dMsaaDepth", QualType<"OCLImage2dMSAADepth", 1>>;
+def Image2dArrayMsaaDepth : Type<"Image2dArrayMsaaDepth", QualType<"OCLImage2dArrayMSAADepth", 1>>;
 
 def Sampler           : Type<"Sampler", QualType<"OCLSamplerTy">>;
 def Event             : Type<"Event", QualType<"OCLEventTy">>;
@@ -398,14 +388,132 @@ foreach name = ["max", "min"] in {
   def : Builtin<name, [AIGenTypeNNoScalar, AIGenTypeNNoScalar, AIGenType1]>;
 }
 
-// OpenCL v1.2 s6.12.14: Built-in Image Read Functions
-def read_imagef : Builtin<"read_imagef",
-                          [VectorType<Float, 4>, image2d_RO_t, VectorType<Int, 2>]>;
-def write_imagef : Builtin<"write_imagef",
-                           [Void,
-                            image2d_WO_t,
-                            VectorType<Int, 2>,
-                            VectorType<Float, 4>]>;
+//--------------------------------------------------------------------
+// OpenCL v1.1 s6.11.3, v1.2 s6.12.14, v2.0 s6.13.14: Image Read and Write Functions
+// OpenCL Extension v2.0 s5.1.8 and s6.1.8: Image Read and Write Functions
+// --- Table 22: Image Read Functions with Samplers ---
+foreach imgTy = [Image1d] in {
+  foreach coordTy = [Int, Float] in {
+    def : Builtin<"read_imagef", [VectorType<Float, 4>, ImageType<imgTy, "RO">, Sampler, coordTy]>;
+    def : Builtin<"read_imagei", [VectorType<Int, 4>, ImageType<imgTy, "RO">, Sampler, coordTy]>;
+    def : Builtin<"read_imageui", [VectorType<UInt, 4>, ImageType<imgTy, "RO">, Sampler, coordTy]>;
+  }
+}
+foreach imgTy = [Image2d, Image1dArray] in {
+  foreach coordTy = [Int, Float] in {
+    def : Builtin<"read_imagef", [VectorType<Float, 4>, ImageType<imgTy, "RO">, Sampler, VectorType<coordTy, 2>]>;
+    def : Builtin<"read_imagei", [VectorType<Int, 4>, ImageType<imgTy, "RO">, Sampler, VectorType<coordTy, 2>]>;
+    def : Builtin<"read_imageui", [VectorType<UInt, 4>, ImageType<imgTy, "RO">, Sampler, VectorType<coordTy, 2>]>;
+  }
+}
+foreach imgTy = [Image3d, Image2dArray] in {
+  foreach coordTy = [Int, Float] in {
+    def : Builtin<"read_imagef", [VectorType<Float, 4>, ImageType<imgTy, "RO">, Sampler, VectorType<coordTy, 4>]>;
+    def : Builtin<"read_imagei", [VectorType<Int, 4>, ImageType<imgTy, "RO">, Sampler, VectorType<coordTy, 4>]>;
+    def : Builtin<"read_imageui", [VectorType<UInt, 4>, ImageType<imgTy, "RO">, Sampler, VectorType<coordTy, 4>]>;
+  }
+}
+foreach coordTy = [Int, Float] in {
+  def : Builtin<"read_imagef", [Float, ImageType<Image2dDepth, "RO">, Sampler, VectorType<coordTy, 2>]>;
+  def : Builtin<"read_imagef", [Float, ImageType<Image2dArrayDepth, "RO">, Sampler, VectorType<coordTy, 4>]>;
+}
+
+// --- Table 23: Sampler-less Read Functions ---
+foreach aQual = ["RO", "RW"] in {
+  foreach imgTy = [Image2d, Image1dArray] in {
+    def : Builtin<"read_imagef", [VectorType<Float, 4>, ImageType<imgTy, aQual>, VectorType<Int, 2>]>;
+    def : Builtin<"read_imagei", [VectorType<Int, 4>, ImageType<imgTy, aQual>, VectorType<Int, 2>]>;
+    def : Builtin<"read_imageui", [VectorType<UInt, 4>, ImageType<imgTy, aQual>, VectorType<Int, 2>]>;
+  }
+  foreach imgTy = [Image3d, Image2dArray] in {
+    def : Builtin<"read_imagef", [VectorType<Float, 4>, ImageType<imgTy, aQual>, VectorType<Int, 4>]>;
+    def : Builtin<"read_imagei", [VectorType<Int, 4>, ImageType<imgTy, aQual>, VectorType<Int, 4>]>;
+    def : Builtin<"read_imageui", [VectorType<UInt, 4>, ImageType<imgTy, aQual>, VectorType<Int, 4>]>;
+  }
+  foreach imgTy = [Image1d, Image1dBuffer] in {
+    def : Builtin<"read_imagef", [VectorType<Float, 4>, ImageType<imgTy, aQual>, Int]>;
+    def : Builtin<"read_imagei", [VectorType<Int, 4>, ImageType<imgTy, aQual>, Int]>;
+    def : Builtin<"read_imageui", [VectorType<UInt, 4>, ImageType<imgTy, aQual>, Int]>;
+  }
+  def : Builtin<"read_imagef", [Float, ImageType<Image2dDepth, aQual>, VectorType<Int, 2>]>;
+  def : Builtin<"read_imagef", [Float, ImageType<Image2dArrayDepth, aQual>, VectorType<Int, 4>]>;
+}
+
+// --- Table 24: Image Write Functions ---
+foreach aQual = ["WO", "RW"] in {
+  foreach imgTy = [Image2d] in {
+    def : Builtin<"write_imagef", [Void, ImageType<imgTy, aQual>, VectorType<Int, 2>, VectorType<Float, 4>]>;
+    def : Builtin<"write_imagei", [Void, ImageType<imgTy, aQual>, VectorType<Int, 2>, VectorType<Int, 4>]>;
+    def : Builtin<"write_imageui", [Void, ImageType<imgTy, aQual>, VectorType<Int, 2>, VectorType<UInt, 4>]>;
+  }
+  foreach imgTy = [Image2dArray] in {
+    def : Builtin<"write_imagef", [Void, ImageType<imgTy, aQual>, VectorType<Int, 4>, VectorType<Float, 4>]>;
+    def : Builtin<"write_imagei", [Void, ImageType<imgTy, aQual>, VectorType<Int, 4>, VectorType<Int, 4>]>;
+    def : Builtin<"write_imageui", [Void, ImageType<imgTy, aQual>, VectorType<Int, 4>, VectorType<UInt, 4>]>;
+  }
+  foreach imgTy = [Image1d, Image1dBuffer] in {
+    def : Builtin<"write_imagef", [Void, ImageType<imgTy, aQual>, Int, VectorType<Float, 4>]>;
+    def : Builtin<"write_imagei", [Void, ImageType<imgTy, aQual>, Int, VectorType<Int, 4>]>;
+    def : Builtin<"write_imageui", [Void, ImageType<imgTy, aQual>, Int, VectorType<UInt, 4>]>;
+  }
+  foreach imgTy = [Image1dArray] in {
+    def : Builtin<"write_imagef", [Void, ImageType<imgTy, aQual>, VectorType<Int, 2>, VectorType<Float, 4>]>;
+    def : Builtin<"write_imagei", [Void, ImageType<imgTy, aQual>, VectorType<Int, 2>, VectorType<Int, 4>]>;
+    def : Builtin<"write_imageui", [Void, ImageType<imgTy, aQual>, VectorType<Int, 2>, VectorType<UInt, 4>]>;
+  }
+  foreach imgTy = [Image3d] in {
+    def : Builtin<"write_imagef", [Void, ImageType<imgTy, aQual>, VectorType<Int, 4>, VectorType<Float, 4>]>;
+    def : Builtin<"write_imagei", [Void, ImageType<imgTy, aQual>, VectorType<Int, 4>, VectorType<Int, 4>]>;
+    def : Builtin<"write_imageui", [Void, ImageType<imgTy, aQual>, VectorType<Int, 4>, VectorType<UInt, 4>]>;
+  }
+  def : Builtin<"write_imagef", [Void, ImageType<Image2dDepth, aQual>, VectorType<Int, 2>, Float]>;
+  def : Builtin<"write_imagef", [Void, ImageType<Image2dArrayDepth, aQual>, VectorType<Int, 4>, Float]>;
+}
+
+// OpenCL extension v2.0 s5.1.9: Built-in Image Read Functions
+// --- Table 8 ---
+foreach aQual = ["RO"] in {
+  foreach name = ["read_imageh"] in {
+    foreach coordTy = [Int, Float] in {
+      foreach imgTy = [Image2d, Image1dArray] in {
+        def : Builtin<name, [VectorType<Half, 4>, ImageType<imgTy, aQual>, Sampler, VectorType<coordTy, 2>]>;
+      }
+      foreach imgTy = [Image3d, Image2dArray] in {
+        def : Builtin<name, [VectorType<Half, 4>, ImageType<imgTy, aQual>, Sampler, VectorType<coordTy, 4>]>;
+      }
+      foreach imgTy = [Image1d] in {
+        def : Builtin<name, [VectorType<Half, 4>, ImageType<imgTy, aQual>, Sampler, coordTy]>;
+      }
+    }
+  }
+}
+// OpenCL extension v2.0 s5.1.10: Built-in Image Sampler-less Read Functions
+// --- Table 9 ---
+foreach aQual = ["RO", "RW"] in {
+  foreach name = ["read_imageh"] in {
+    foreach imgTy = [Image2d, Image1dArray] in {
+      def : Builtin<name, [VectorType<Half, 4>, ImageType<imgTy, aQual>, VectorType<Int, 2>]>;
+    }
+    foreach imgTy = [Image3d, Image2dArray] in {
+      def : Builtin<name, [VectorType<Half, 4>, ImageType<imgTy, aQual>, VectorType<Int, 4>]>;
+    }
+    foreach imgTy = [Image1d, Image1dBuffer] in {
+      def : Builtin<name, [VectorType<Half, 4>, ImageType<imgTy, aQual>, Int]>;
+    }
+  }
+}
+// OpenCL extension v2.0 s5.1.11: Built-in Image Write Functions
+// --- Table 10 ---
+foreach aQual = ["WO", "RW"] in {
+  foreach name = ["write_imageh"] in {
+    def : Builtin<name, [Void, ImageType<Image2d, aQual>, VectorType<Int, 2>, VectorType<Half, 4>]>;
+    def : Builtin<name, [Void, ImageType<Image2dArray, aQual>, VectorType<Int, 4>, VectorType<Half, 4>]>;
+    def : Builtin<name, [Void, ImageType<Image1d, aQual>, Int, VectorType<Half, 4>]>;
+    def : Builtin<name, [Void, ImageType<Image1dBuffer, aQual>, Int, VectorType<Half, 4>]>;
+    def : Builtin<name, [Void, ImageType<Image1dArray, aQual>, VectorType<Int, 2>, VectorType<Half, 4>]>;
+    def : Builtin<name, [Void, ImageType<Image3d, aQual>, VectorType<Int, 4>, VectorType<Half, 4>]>;
+  }
+}
 
 
 // OpenCL v2.0 s9.17.3: Additions to section 6.13.1: Work-Item Functions
diff --git a/clang/test/SemaOpenCL/fdeclare-opencl-builtins.cl b/clang/test/SemaOpenCL/fdeclare-opencl-builtins.cl
@@ -4,11 +4,14 @@
 
 // Test the -fdeclare-opencl-builtins option.
 
+#pragma OPENCL EXTENSION cl_khr_fp16 : enable
+
 // Provide typedefs when invoking clang without -finclude-default-header.
 #ifdef NO_HEADER
 typedef char char2 __attribute__((ext_vector_type(2)));
 typedef char char4 __attribute__((ext_vector_type(4)));
 typedef float float4 __attribute__((ext_vector_type(4)));
+typedef half half4 __attribute__((ext_vector_type(4)));
 typedef int int2 __attribute__((ext_vector_type(2)));
 typedef int int4 __attribute__((ext_vector_type(4)));
 typedef long long2 __attribute__((ext_vector_type(2)));
@@ -46,6 +49,33 @@ char4 test_int(char c, char4 c4) {
   return max(c4, c);
 }
 
+kernel void basic_image_readonly(read_only image2d_t image_read_only_image2d) {
+  int2 i2;
+  sampler_t sampler;
+  half4 res;
+  float4 resf;
+
+  resf = read_imagef(image_read_only_image2d, i2);
+  res = read_imageh(image_read_only_image2d, i2);
+  res = read_imageh(image_read_only_image2d, sampler, i2);
+}
+
+kernel void basic_image_readwrite(read_write image3d_t image_read_write_image3d) {
+  half4 h4;
+  int4 i4;
+
+  write_imageh(image_read_write_image3d, i4, h4);
+}
+
+kernel void basic_image_writeonly(write_only image1d_buffer_t image_write_only_image1d_buffer) {
+  half4 h4;
+  float4 f4;
+  int i;
+
+  write_imagef(image_write_only_image1d_buffer, i, f4);
+  write_imageh(image_write_only_image1d_buffer, i, h4);
+}
+
 kernel void basic_subgroup(global uint *out) {
   out[0] = get_sub_group_size();
 }
diff --git a/clang/utils/TableGen/ClangOpenCLBuiltinEmitter.cpp b/clang/utils/TableGen/ClangOpenCLBuiltinEmitter.cpp
@@ -56,6 +56,7 @@
 #include "llvm/ADT/StringExtras.h"
 #include "llvm/ADT/StringRef.h"
 #include "llvm/ADT/StringSet.h"
+#include "llvm/ADT/StringSwitch.h"
 #include "llvm/Support/ErrorHandling.h"
 #include "llvm/Support/raw_ostream.h"
 #include "llvm/TableGen/Error.h"
@@ -233,6 +234,13 @@ void BuiltinNameEmitter::EmitDeclarations() {
 
   // Structure definitions.
   OS << R"(
+// Image access qualifier.
+enum OpenCLAccessQual : unsigned char {
+  OCLAQ_None,
+  OCLAQ_ReadOnly,
+  OCLAQ_WriteOnly,
+  OCLAQ_ReadWrite
+};
 
 // Represents a return type or argument type.
 struct OpenCLTypeStruct {
@@ -246,6 +254,8 @@ struct OpenCLTypeStruct {
   const bool IsConst;
   // 0 if the type is not volatile.
   const bool IsVolatile;
+  // Access qualifier.
+  const OpenCLAccessQual AccessQualifier;
   // Address space of the pointer (if applicable).
   const LangAS AS;
 };
@@ -347,12 +357,20 @@ void BuiltinNameEmitter::GetOverloads() {
 void BuiltinNameEmitter::EmitTypeTable() {
   OS << "static const OpenCLTypeStruct TypeTable[] = {\n";
   for (const auto &T : TypeMap) {
-    OS << "  // " << T.second << "\n";
-    OS << "  {OCLT_" << T.first->getValueAsString("Name") << ", "
+    const char *AccessQual =
+        StringSwitch<const char *>(T.first->getValueAsString("AccessQualifier"))
+            .Case("RO", "OCLAQ_ReadOnly")
+            .Case("WO", "OCLAQ_WriteOnly")
+            .Case("RW", "OCLAQ_ReadWrite")
+            .Default("OCLAQ_None");
+
+    OS << "  // " << T.second << "\n"
+       << "  {OCLT_" << T.first->getValueAsString("Name") << ", "
        << T.first->getValueAsInt("VecWidth") << ", "
        << T.first->getValueAsBit("IsPointer") << ", "
        << T.first->getValueAsBit("IsConst") << ", "
        << T.first->getValueAsBit("IsVolatile") << ", "
+       << AccessQual << ", "
        << T.first->getValueAsString("AddrSpace") << "},\n";
   }
   OS << "};\n\n";
@@ -455,6 +473,47 @@ static void OCL2Qual(ASTContext &Context, const OpenCLTypeStruct &Ty,
   // Start of switch statement over all types.
   OS << "\n  switch (Ty.ID) {\n";
 
+  // Switch cases for image types (Image2d, Image3d, ...)
+  std::vector<Record *> ImageTypes =
+      Records.getAllDerivedDefinitions("ImageType");
+
+  // Map an image type name to its 3 access-qualified types (RO, WO, RW).
+  std::map<StringRef, SmallVector<Record *, 3>> ImageTypesMap;
+  for (auto *IT : ImageTypes) {
+    auto Entry = ImageTypesMap.find(IT->getValueAsString("Name"));
+    if (Entry == ImageTypesMap.end()) {
+      SmallVector<Record *, 3> ImageList;
+      ImageList.push_back(IT);
+      ImageTypesMap.insert(
+          std::make_pair(IT->getValueAsString("Name"), ImageList));
+    } else {
+      Entry->second.push_back(IT);
+    }
+  }
+
+  // Emit the cases for the image types.  For an image type name, there are 3
+  // corresponding QualTypes ("RO", "WO", "RW").  The "AccessQualifier" field
+  // tells which one is needed.  Emit a switch statement that puts the
+  // corresponding QualType into "QT".
+  for (const auto &ITE : ImageTypesMap) {
+    OS << "    case OCLT_" << ITE.first.str() << ":\n"
+       << "      switch (Ty.AccessQualifier) {\n"
+       << "        case OCLAQ_None:\n"
+       << "          llvm_unreachable(\"Image without access qualifier\");\n";
+    for (const auto &Image : ITE.second) {
+      OS << StringSwitch<const char *>(
+                Image->getValueAsString("AccessQualifier"))
+                .Case("RO", "        case OCLAQ_ReadOnly:\n")
+                .Case("WO", "        case OCLAQ_WriteOnly:\n")
+                .Case("RW", "        case OCLAQ_ReadWrite:\n")
+         << "          QT.push_back(Context."
+         << Image->getValueAsDef("QTName")->getValueAsString("Name") << ");\n"
+         << "          break;\n";
+    }
+    OS << "      }\n"
+       << "      break;\n";
+  }
+
   // Switch cases for generic types.
   for (const auto *GenType : Records.getAllDerivedDefinitions("GenericType")) {
     OS << "    case OCLT_" << GenType->getValueAsString("Name") << ":\n";
@@ -495,6 +554,9 @@ static void OCL2Qual(ASTContext &Context, const OpenCLTypeStruct &Ty,
   StringMap<bool> TypesSeen;
 
   for (const auto *T : Types) {
+    // Check this is not an image type
+    if (ImageTypesMap.find(T->getValueAsString("Name")) != ImageTypesMap.end())
+      continue;
     // Check we have not seen this Type
     if (TypesSeen.find(T->getValueAsString("Name")) != TypesSeen.end())
       continue;
@@ -512,7 +574,9 @@ static void OCL2Qual(ASTContext &Context, const OpenCLTypeStruct &Ty,
   }
 
   // End of switch statement.
-  OS << "  } // end of switch (Ty.ID)\n\n";
+  OS << "    default:\n"
+     << "      llvm_unreachable(\"OpenCL builtin type not handled yet\");\n"
+     << "  } // end of switch (Ty.ID)\n\n";
 
   // Step 2.
   // Add ExtVector types if this was a generic type, as the switch statement
