[MLIR] emitc: Add fmtArgs to verbatim (llvm#123294)

Allows to print code snippets that refer to arguments or local
variables. E.g. `emitc.verbatim "#pragma abc var={}" args %arg0 :
!emitc.ptr<i32>` is printed as `#pragma abc var=v1` when the translator
had decided to print `%arg0` as `v1`.

As a follow-up PR, we will use the same infra to extend opaque type,
which provides a way to generate template types depending on the
spelling of other types.

Author: mgehre-amd

mlir/include/mlir/Dialect/EmitC/IR/EmitC.h

@@ -27,6 +27,8 @@
 #include "mlir/Dialect/EmitC/IR/EmitCDialect.h.inc"
 #include "mlir/Dialect/EmitC/IR/EmitCEnums.h.inc"
 
+#include <variant>
+
 namespace mlir {
 namespace emitc {
 void buildTerminatedBody(OpBuilder &builder, Location loc);
@@ -47,6 +49,10 @@ bool isSupportedFloatType(mlir::Type type);
 /// Determines whether \p type is a emitc.size_t/ssize_t type.
 bool isPointerWideType(mlir::Type type);
 
+// Either a literal string, or an placeholder for the fmtArgs.
+struct Placeholder {};
+using ReplacementItem = std::variant<StringRef, Placeholder>;
+
 } // namespace emitc
 } // namespace mlir
 

mlir/include/mlir/Dialect/EmitC/IR/EmitC.td

@@ -1269,10 +1269,29 @@ def EmitC_VerbatimOp : EmitC_Op<"verbatim"> {
     }
     #endif
     ```
+
+    If the `emitc.verbatim` op has operands, then the `value` is interpreted as
+    format string, where `{}` is a placeholder for an operand in their order.
+    For example, `emitc.verbatim "#pragma my src={} dst={}" %src, %dest : i32, i32`
+    would be emitted as `#pragma my src=a dst=b` if `%src` became `a` and
+    `%dest` became `b` in the C code.
+    `{{` in the format string is interpreted as a single `{` and doesn't introduce
+    a placeholder.
   }];
 
-  let arguments = (ins StrAttr:$value);
-  let assemblyFormat = "$value attr-dict";
+  let extraClassDeclaration = [{
+    FailureOr<SmallVector<::mlir::emitc::ReplacementItem>> parseFormatString();
+  }];
+
+  let arguments = (ins StrAttr:$value, Variadic<EmitCType>:$fmtArgs);
+
+  let builders = [OpBuilder<(ins "::mlir::StringAttr":$value),
+                            [{ build($_builder, $_state, value, {}); }]>];
+  let builders = [OpBuilder<(ins "::llvm::StringRef":$value),
+                            [{ build($_builder, $_state, value, {}); }]>];
+  let hasVerifier = 1;
+  let assemblyFormat =
+      "$value (`args` $fmtArgs^ `:` type($fmtArgs))? attr-dict";
 }
 
 def EmitC_AssignOp : EmitC_Op<"assign", []> {

mlir/lib/Dialect/EmitC/IR/EmitC.cpp

@@ -19,6 +19,7 @@
 #include "llvm/ADT/StringExtras.h"
 #include "llvm/ADT/TypeSwitch.h"
 #include "llvm/Support/Casting.h"
+#include "llvm/Support/FormatVariadic.h"
 
 using namespace mlir;
 using namespace mlir::emitc;
@@ -167,6 +168,63 @@ static LogicalResult verifyInitializationAttribute(Operation *op,
   return success();
 }
 
+/// Parse a format string and return a list of its parts.
+/// A part is either a StringRef that has to be printed as-is, or
+/// a Placeholder which requires printing the next operand of the VerbatimOp.
+/// In the format string, all `{}` are replaced by Placeholders, except if the
+/// `{` is escaped by `{{` - then it doesn't start a placeholder.
+template <class ArgType>
+FailureOr<SmallVector<ReplacementItem>>
+parseFormatString(StringRef toParse, ArgType fmtArgs,
+                  std::optional<llvm::function_ref<mlir::InFlightDiagnostic()>>
+                      emitError = {}) {
+  SmallVector<ReplacementItem> items;
+
+  // If there are not operands, the format string is not interpreted.
+  if (fmtArgs.empty()) {
+    items.push_back(toParse);
+    return items;
+  }
+
+  while (!toParse.empty()) {
+    size_t idx = toParse.find('{');
+    if (idx == StringRef::npos) {
+      // No '{'
+      items.push_back(toParse);
+      break;
+    }
+    if (idx > 0) {
+      // Take all chars excluding the '{'.
+      items.push_back(toParse.take_front(idx));
+      toParse = toParse.drop_front(idx);
+      continue;
+    }
+    if (toParse.size() < 2) {
+      return (*emitError)()
+             << "expected '}' after unescaped '{' at end of string";
+    }
+    // toParse contains at least two characters and starts with `{`.
+    char nextChar = toParse[1];
+    if (nextChar == '{') {
+      // Double '{{' -> '{' (escaping).
+      items.push_back(toParse.take_front(1));
+      toParse = toParse.drop_front(2);
+      continue;
+    }
+    if (nextChar == '}') {
+      items.push_back(Placeholder{});
+      toParse = toParse.drop_front(2);
+      continue;
+    }
+
+    if (emitError.has_value()) {
+      return (*emitError)() << "expected '}' after unescaped '{'";
+    }
+    return failure();
+  }
+  return items;
+}
+
 //===----------------------------------------------------------------------===//
 // AddOp
 //===----------------------------------------------------------------------===//
@@ -909,6 +967,55 @@ LogicalResult emitc::SubscriptOp::verify() {
   return success();
 }
 
+//===----------------------------------------------------------------------===//
+// VerbatimOp
+//===----------------------------------------------------------------------===//
+
+LogicalResult emitc::VerbatimOp::verify() {
+  auto errorCallback = [&]() -> InFlightDiagnostic {
+    return this->emitOpError();
+  };
+  FailureOr<SmallVector<ReplacementItem>> fmt =
+      ::parseFormatString(getValue(), getFmtArgs(), errorCallback);
+  if (failed(fmt))
+    return failure();
+
+  size_t numPlaceholders = llvm::count_if(*fmt, [](ReplacementItem &item) {
+    return std::holds_alternative<Placeholder>(item);
+  });
+
+  if (numPlaceholders != getFmtArgs().size()) {
+    return emitOpError()
+           << "requires operands for each placeholder in the format string";
+  }
+  return success();
+}
+
+static ParseResult parseVariadicTypeFmtArgs(AsmParser &p,
+                                            SmallVector<Type> &params) {
+  Type type;
+  if (p.parseType(type))
+    return failure();
+
+  params.push_back(type);
+  while (succeeded(p.parseOptionalComma())) {
+    if (p.parseType(type))
+      return failure();
+    params.push_back(type);
+  }
+
+  return success();
+}
+
+static void printVariadicTypeFmtArgs(AsmPrinter &p, ArrayRef<Type> params) {
+  llvm::interleaveComma(params, p, [&](Type type) { p.printType(type); });
+}
+
+FailureOr<SmallVector<ReplacementItem>> emitc::VerbatimOp::parseFormatString() {
+  // Error checking is done in verify.
+  return ::parseFormatString(getValue(), getFmtArgs());
+}
+
 //===----------------------------------------------------------------------===//
 // EmitC Enums
 //===----------------------------------------------------------------------===//

mlir/lib/Target/Cpp/TranslateToCpp.cpp

@@ -568,7 +568,21 @@ static LogicalResult printOperation(CppEmitter &emitter,
                                     emitc::VerbatimOp verbatimOp) {
   raw_ostream &os = emitter.ostream();
 
-  os << verbatimOp.getValue();
+  FailureOr<SmallVector<ReplacementItem>> items =
+      verbatimOp.parseFormatString();
+  if (failed(items))
+    return failure();
+
+  auto fmtArg = verbatimOp.getFmtArgs().begin();
+
+  for (ReplacementItem &item : *items) {
+    if (auto *str = std::get_if<StringRef>(&item)) {
+      os << *str;
+    } else {
+      if (failed(emitter.emitOperand(*fmtArg++)))
+        return failure();
+    }
+  }
 
   return success();
 }

mlir/test/Dialect/EmitC/invalid_ops.mlir

@@ -566,3 +566,59 @@ func.func @emitc_switch() {
   }
   return
 }
+
+// -----
+
+func.func @test_verbatim(%arg0 : !emitc.ptr<i32>, %arg1 : i32) {
+  // expected-error @+1 {{'emitc.verbatim' op requires operands for each placeholder in the format string}}
+  emitc.verbatim "" args %arg0, %arg1 : !emitc.ptr<i32>, i32
+  return
+}
+
+// -----
+
+func.func @test_verbatim(%arg0 : !emitc.ptr<i32>, %arg1 : i32) {
+  // expected-error @+1 {{'emitc.verbatim' op expected '}' after unescaped '{' at end of string}}
+  emitc.verbatim "{} + {} {" args %arg0, %arg1 : !emitc.ptr<i32>, i32
+  return
+}
+
+// -----
+
+func.func @test_verbatim(%arg0 : !emitc.ptr<i32>, %arg1 : i32) {
+  // expected-error @+1 {{'emitc.verbatim' op requires operands for each placeholder in the format string}}
+  emitc.verbatim "abc" args %arg0, %arg1 : !emitc.ptr<i32>, i32
+  return
+}
+
+// -----
+
+func.func @test_verbatim(%arg0 : !emitc.ptr<i32>, %arg1 : i32) {
+  // expected-error @+1 {{'emitc.verbatim' op requires operands for each placeholder in the format string}}
+  emitc.verbatim "{}" args %arg0, %arg1 : !emitc.ptr<i32>, i32
+  return
+}
+
+// -----
+
+func.func @test_verbatim(%arg0 : !emitc.ptr<i32>, %arg1 : i32) {
+  // expected-error @+1 {{'emitc.verbatim' op requires operands for each placeholder in the format string}}
+  emitc.verbatim "{} {} {}" args %arg0, %arg1 : !emitc.ptr<i32>, i32
+  return
+}
+
+// -----
+
+func.func @test_verbatim(%arg0 : !emitc.ptr<i32>, %arg1 : i32) {
+  // expected-error @+1 {{'emitc.verbatim' op expected '}' after unescaped '{'}}
+  emitc.verbatim "{ " args %arg0, %arg1 : !emitc.ptr<i32>, i32
+  return
+}
+
+// -----
+
+func.func @test_verbatim(%arg0 : !emitc.ptr<i32>, %arg1 : i32) {
+  // expected-error @+1 {{'emitc.verbatim' op expected '}' after unescaped '{'}}
+  emitc.verbatim "{a} " args %arg0, %arg1 : !emitc.ptr<i32>, i32
+  return
+}

mlir/test/Dialect/EmitC/ops.mlir

@@ -238,6 +238,18 @@ emitc.verbatim "#endif  // __cplusplus"
 emitc.verbatim "typedef int32_t i32;"
 emitc.verbatim "typedef float f32;"
 
+// The value is not interpreted as format string if there are no operands.
+emitc.verbatim "{} {  }"
+
+func.func @test_verbatim(%arg0 : !emitc.ptr<i32>, %arg1 : i32) {
+  emitc.verbatim "{} + {};" args %arg0, %arg1 : !emitc.ptr<i32>, i32
+
+  // Check there is no ambiguity whether %a is the argument to the emitc.verbatim op.
+  emitc.verbatim "a"
+  %a = "emitc.constant"(){value = 42 : i32} : () -> i32
+
+  return
+}
 
 emitc.global @uninit : i32
 emitc.global @myglobal_int : i32 = 4

mlir/test/Target/Cpp/verbatim.mlir

@@ -1,5 +1,5 @@
-// RUN: mlir-translate -mlir-to-cpp %s | FileCheck %s
-// RUN: mlir-translate -mlir-to-cpp -declare-variables-at-top %s | FileCheck %s
+// RUN: mlir-translate -mlir-to-cpp %s | FileCheck %s --match-full-lines
+// RUN: mlir-translate -mlir-to-cpp -declare-variables-at-top %s | FileCheck %s --match-full-lines
 
 
 emitc.verbatim "#ifdef __cplusplus"
@@ -19,3 +19,23 @@ emitc.verbatim "typedef int32_t i32;"
 // CHECK-NEXT: typedef int32_t i32;
 emitc.verbatim "typedef float f32;"
 // CHECK-NEXT: typedef float f32;
+
+emitc.func @func(%arg: f32) {
+  // CHECK: void func(float [[V0:[^ ]*]]) {
+  %a = "emitc.variable"(){value = #emitc.opaque<"">} : () -> !emitc.array<3x7xi32>
+  // CHECK: int32_t [[A:[^ ]*]][3][7];
+
+  emitc.verbatim "{}" args %arg : f32
+  // CHECK: [[V0]]
+
+  emitc.verbatim "{} {{a" args %arg : f32
+  // CHECK-NEXT: [[V0]] {a
+
+  emitc.verbatim "#pragma my var={} property" args %arg : f32
+  // CHECK-NEXT: #pragma my var=[[V0]] property
+
+  emitc.verbatim "#pragma my2 var={} property" args %a : !emitc.array<3x7xi32>
+  // CHECK-NEXT: #pragma my2 var=[[A]] property
+
+  emitc.return
+}