diff --git a/mlir/include/mlir/Dialect/EmitC/IR/EmitC.td b/mlir/include/mlir/Dialect/EmitC/IR/EmitC.td
index fb7a108b39fc0..721f9f6b320ad 100644
--- a/mlir/include/mlir/Dialect/EmitC/IR/EmitC.td
+++ b/mlir/include/mlir/Dialect/EmitC/IR/EmitC.td
@@ -402,7 +402,8 @@ def EmitC_CmpOp : EmitC_BinaryOp<"cmp", []> {
   let assemblyFormat = "$predicate `,` operands attr-dict `:` functional-type(operands, results)";
 }
 
-def EmitC_ConstantOp : EmitC_Op<"constant", [ConstantLike]> {
+def EmitC_ConstantOp
+    : EmitC_Op<"constant", [ConstantLike, CExpressionInterface]> {
   let summary = "Constant operation";
   let description = [{
     The `emitc.constant` operation produces an SSA value equal to some constant
@@ -429,6 +430,13 @@ def EmitC_ConstantOp : EmitC_Op<"constant", [ConstantLike]> {
 
   let hasFolder = 1;
   let hasVerifier = 1;
+
+  let extraClassDeclaration = [{
+    bool hasSideEffects() {
+      // If operand is fundamental type, the operation is pure.
+      return !isFundamentalType(getResult().getType());
+    }
+  }];
 }
 
 def EmitC_DivOp : EmitC_BinaryOp<"div", []> {
@@ -466,8 +474,9 @@ def EmitC_ExpressionOp
     its single-basic-block region. The operation doesn't take any arguments.
 
     As the operation is to be emitted as a C expression, the operations within
-    its body must form a single Def-Use tree of emitc ops whose result is
-    yielded by a terminating `emitc.yield`.
+    its body must form a single Def-Use tree, or a DAG trivially expandable to
+    one, i.e. a DAG where each operation with side effects is only reachable
+    once from the expression root.
 
     Example:
 
diff --git a/mlir/lib/Dialect/EmitC/IR/EmitC.cpp b/mlir/lib/Dialect/EmitC/IR/EmitC.cpp
index 00ce3b59bf870..a73470cdf76c5 100644
--- a/mlir/lib/Dialect/EmitC/IR/EmitC.cpp
+++ b/mlir/lib/Dialect/EmitC/IR/EmitC.cpp
@@ -14,7 +14,9 @@
 #include "mlir/IR/DialectImplementation.h"
 #include "mlir/IR/Types.h"
 #include "mlir/Interfaces/FunctionImplementation.h"
+#include "mlir/Support/LLVM.h"
 #include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/SmallVector.h"
 #include "llvm/ADT/TypeSwitch.h"
 #include "llvm/Support/Casting.h"
 
@@ -462,12 +464,34 @@ LogicalResult ExpressionOp::verify() {
     return emitOpError("requires yielded type to match return type");
 
   for (Operation &op : region.front().without_terminator()) {
-    if (!isa<emitc::CExpressionInterface>(op))
+    auto expressionInterface = dyn_cast<emitc::CExpressionInterface>(op);
+    if (!expressionInterface)
       return emitOpError("contains an unsupported operation");
     if (op.getNumResults() != 1)
       return emitOpError("requires exactly one result for each operation");
-    if (!op.getResult(0).hasOneUse())
-      return emitOpError("requires exactly one use for each operation");
+    Value result = op.getResult(0);
+    if (result.use_empty())
+      return emitOpError("contains an unused operation");
+  }
+
+  // Make sure any operation with side effect is only reachable once from
+  // the root op, otherwise emission will be replicating side effects.
+  SmallPtrSet<Operation *, 16> visited;
+  SmallVector<Operation *> worklist;
+  worklist.push_back(rootOp);
+  while (!worklist.empty()) {
+    Operation *op = worklist.back();
+    worklist.pop_back();
+    if (visited.contains(op)) {
+      if (cast<CExpressionInterface>(op).hasSideEffects())
+        return emitOpError(
+            "requires exactly one use for operations with side effects");
+    }
+    visited.insert(op);
+    for (Value operand : op->getOperands())
+      if (Operation *def = operand.getDefiningOp()) {
+        worklist.push_back(def);
+      }
   }
 
   return success();
diff --git a/mlir/lib/Target/Cpp/TranslateToCpp.cpp b/mlir/lib/Target/Cpp/TranslateToCpp.cpp
index 570f38c60020b..a5bd80e9d6b8b 100644
--- a/mlir/lib/Target/Cpp/TranslateToCpp.cpp
+++ b/mlir/lib/Target/Cpp/TranslateToCpp.cpp
@@ -97,6 +97,7 @@ static FailureOr<int> getOperatorPrecedence(Operation *operation) {
         return op->emitError("unsupported cmp predicate");
       })
       .Case<emitc::ConditionalOp>([&](auto op) { return 2; })
+      .Case<emitc::ConstantOp>([&](auto op) { return 17; })
       .Case<emitc::DivOp>([&](auto op) { return 13; })
       .Case<emitc::LoadOp>([&](auto op) { return 16; })
       .Case<emitc::LogicalAndOp>([&](auto op) { return 4; })
@@ -267,8 +268,14 @@ struct CppEmitter {
     Operation *def = value.getDefiningOp();
     if (!def)
       return false;
+    return isPartOfCurrentExpression(def);
+  }
+
+  /// Determine whether given operation is part of the expression potentially
+  /// being emitted.
+  bool isPartOfCurrentExpression(Operation *def) {
     auto operandExpression = dyn_cast<ExpressionOp>(def->getParentOp());
-    return operandExpression == emittedExpression;
+    return operandExpression && operandExpression == emittedExpression;
   };
 
   // Resets the value counter to 0.
@@ -408,6 +415,9 @@ static LogicalResult printOperation(CppEmitter &emitter,
   Operation *operation = constantOp.getOperation();
   Attribute value = constantOp.getValue();
 
+  if (emitter.isPartOfCurrentExpression(operation))
+    return emitter.emitAttribute(operation->getLoc(), value);
+
   return printConstantOp(emitter, operation, value);
 }
 
diff --git a/mlir/test/Dialect/EmitC/form-expressions.mlir b/mlir/test/Dialect/EmitC/form-expressions.mlir
index 67cd6fddba638..7b6723989e260 100644
--- a/mlir/test/Dialect/EmitC/form-expressions.mlir
+++ b/mlir/test/Dialect/EmitC/form-expressions.mlir
@@ -2,8 +2,8 @@
 
 // CHECK-LABEL: func.func @single_expression(
 // CHECK-SAME:                               %[[VAL_0:.*]]: i32, %[[VAL_1:.*]]: i32, %[[VAL_2:.*]]: i32, %[[VAL_3:.*]]: i32) -> i1 {
-// CHECK:           %[[VAL_4:.*]] = "emitc.constant"() <{value = 42 : i32}> : () -> i32
-// CHECK:           %[[VAL_5:.*]] = emitc.expression %[[VAL_3]], %[[VAL_2]], %[[VAL_0]], %[[VAL_4]] : (i32, i32, i32, i32) -> i1 {
+// CHECK:           %[[VAL_5:.*]] = emitc.expression %[[VAL_3]], %[[VAL_2]], %[[VAL_0]] : (i32, i32, i32) -> i1 {
+// CHECK:             %[[VAL_4:.*]] = "emitc.constant"() <{value = 42 : i32}> : () -> i32
 // CHECK:             %[[VAL_6:.*]] = mul %[[VAL_0]], %[[VAL_4]] : (i32, i32) -> i32
 // CHECK:             %[[VAL_7:.*]] = sub %[[VAL_6]], %[[VAL_2]] : (i32, i32) -> i32
 // CHECK:             %[[VAL_8:.*]] = cmp lt, %[[VAL_7]], %[[VAL_3]] : (i32, i32) -> i1
@@ -131,7 +131,10 @@ func.func @single_result_requirement() -> (i32, i32) {
 // CHECK-LABEL:   func.func @expression_with_load(
 // CHECK-SAME:                                    %[[VAL_0:.*]]: i32,
 // CHECK-SAME:                                    %[[VAL_1:.*]]: !emitc.ptr<i32>) -> i1 {
-// CHECK:           %[[VAL_2:.*]] = "emitc.constant"() <{value = 0 : i64}> : () -> i64
+// CHECK:           %[[VAL_2:.*]] = emitc.expression : () -> i64 {
+// CHECK:             %[[VAL_C:.*]] = "emitc.constant"() <{value = 0 : i64}> : () -> i64
+// CHECK:             yield %[[VAL_C]] : i64
+// CHECK:           }
 // CHECK:           %[[VAL_3:.*]] = "emitc.variable"() <{value = #emitc.opaque<"42">}> : () -> !emitc.lvalue<i32>
 // CHECK:           %[[VAL_4:.*]] = emitc.expression %[[VAL_3]] : (!emitc.lvalue<i32>) -> i32 {
 // CHECK:             %[[VAL_5:.*]] = load %[[VAL_3]] : <i32>
@@ -162,24 +165,46 @@ func.func @expression_with_load(%arg0: i32, %arg1: !emitc.ptr<i32>) -> i1 {
 }
 
 // CHECK-LABEL:   func.func @opaque_type_expression(%arg0: i32, %arg1: !emitc.opaque<"T0">, %arg2: i32) -> i1 {
-// CHECK:           %0 = "emitc.constant"() <{value = 42 : i32}> : () -> i32
-// CHECK:           %1 = emitc.expression %arg1, %arg0, %0 : (!emitc.opaque<"T0">, i32, i32) -> i32 {
-// CHECK:             %3 = mul %arg0, %0 : (i32, i32) -> i32
-// CHECK:             %4 = sub %3, %arg1 : (i32, !emitc.opaque<"T0">) -> i32
+// CHECK:           %0 = emitc.expression : () -> !emitc.opaque<"T1"> {
+// CHECK:             %4 = "emitc.constant"() <{value = #emitc.opaque<"V">}> : () -> !emitc.opaque<"T1">
+// CHECK:             yield %4 : !emitc.opaque<"T1">
+// CHECK:           }
+// CHECK:           %1 = emitc.expression %arg0, %0 : (i32, !emitc.opaque<"T1">) -> i32 {
+// CHECK:             %4 = mul %arg0, %0 : (i32, !emitc.opaque<"T1">) -> i32
+// CHECK:             yield %4 : i32
+// CHECK:           }
+// CHECK:           %2 = emitc.expression %1, %arg1 : (i32, !emitc.opaque<"T0">) -> i32 {
+// CHECK:             %4 = sub %1, %arg1 : (i32, !emitc.opaque<"T0">) -> i32
 // CHECK:             yield %4 : i32
 // CHECK:           }
-// CHECK:           %2 = emitc.expression %1, %arg2 : (i32, i32) -> i1 {
-// CHECK:             %3 = cmp lt, %1, %arg2 : (i32, i32) -> i1
-// CHECK:             yield %3 : i1
+// CHECK:           %3 = emitc.expression %2, %arg2 : (i32, i32) -> i1 {
+// CHECK:             %4 = cmp lt, %2, %arg2 : (i32, i32) -> i1
+// CHECK:             yield %4 : i1
 // CHECK:           }
-// CHECK:           return %2 : i1
+// CHECK:           return %3 : i1
 // CHECK:         }
 
 
 func.func @opaque_type_expression(%arg0: i32,  %arg1: !emitc.opaque<"T0">, %arg2: i32) -> i1 {
-  %c42 = "emitc.constant"(){value = 42 : i32} : () -> i32
-  %a = emitc.mul %arg0, %c42 : (i32, i32) -> i32
+  %c42 = "emitc.constant"(){value = #emitc.opaque<"V">} : () -> !emitc.opaque<"T1">
+  %a = emitc.mul %arg0, %c42 : (i32, !emitc.opaque<"T1">) -> i32
   %b = emitc.sub %a, %arg1 : (i32, !emitc.opaque<"T0">) -> i32
   %c = emitc.cmp lt, %b, %arg2 :(i32, i32) -> i1
   return %c : i1
 }
+
+// CHECK-LABEL:   func.func @expression_with_constant(
+// CHECK-SAME:                                        %[[VAL_0:.*]]: i32) -> i32 {
+// CHECK:           %[[VAL_1:.*]] = emitc.expression %[[VAL_0]] : (i32) -> i32 {
+// CHECK:             %[[VAL_2:.*]] = "emitc.constant"() <{value = 42 : i32}> : () -> i32
+// CHECK:             %[[VAL_3:.*]] = mul %[[VAL_0]], %[[VAL_2]] : (i32, i32) -> i32
+// CHECK:             yield %[[VAL_3]] : i32
+// CHECK:           }
+// CHECK:           return %[[VAL_1]] : i32
+// CHECK:         }
+
+func.func @expression_with_constant(%arg0: i32) -> i32 {
+  %c42 = "emitc.constant"(){value = 42 : i32} : () -> i32
+  %a = emitc.mul %arg0, %c42 : (i32, i32) -> i32
+  return %a : i32
+}
diff --git a/mlir/test/Dialect/EmitC/invalid_ops.mlir b/mlir/test/Dialect/EmitC/invalid_ops.mlir
index fdfb0eb46f7c5..a97474401645c 100644
--- a/mlir/test/Dialect/EmitC/invalid_ops.mlir
+++ b/mlir/test/Dialect/EmitC/invalid_ops.mlir
@@ -311,7 +311,7 @@ func.func @test_expression_illegal_op(%arg0 : i1) -> i32 {
 // -----
 
 func.func @test_expression_no_use(%arg0: i32, %arg1: i32) -> i32 {
-  // expected-error @+1 {{'emitc.expression' op requires exactly one use for each operation}}
+  // expected-error @+1 {{'emitc.expression' op contains an unused operation}}
   %r = emitc.expression %arg0, %arg1 : (i32, i32) -> i32 {
     %a = emitc.add %arg0, %arg1 : (i32, i32) -> i32
     %b = emitc.rem %arg0, %arg1 : (i32, i32) -> i32
@@ -323,12 +323,12 @@ func.func @test_expression_no_use(%arg0: i32, %arg1: i32) -> i32 {
 // -----
 
 func.func @test_expression_multiple_uses(%arg0: i32, %arg1: i32) -> i32 {
-  // expected-error @+1 {{'emitc.expression' op requires exactly one use for each operation}}
+  // expected-error @+1 {{'emitc.expression' op requires exactly one use for operations with side effects}}
   %r = emitc.expression %arg0, %arg1 : (i32, i32) -> i32 {
-    %a = emitc.rem %arg0, %arg1 : (i32, i32) -> i32
+    %a = emitc.call_opaque "foo"(%arg0, %arg1) : (i32, i32) -> i32
     %b = emitc.add %a, %arg0 : (i32, i32) -> i32
-    %c = emitc.mul %arg1, %a : (i32, i32) -> i32
-    emitc.yield %a : i32
+    %c = emitc.mul %b, %a : (i32, i32) -> i32
+    emitc.yield %c : i32
   }
   return %r : i32
 }
diff --git a/mlir/test/Dialect/EmitC/ops.mlir b/mlir/test/Dialect/EmitC/ops.mlir
index e890f77173de7..84c9b65d775d2 100644
--- a/mlir/test/Dialect/EmitC/ops.mlir
+++ b/mlir/test/Dialect/EmitC/ops.mlir
@@ -203,6 +203,16 @@ func.func @test_expression(%arg0: i32, %arg1: i32, %arg2: i32, %arg3: f32, %arg4
   return %r : i32
 }
 
+func.func @test_expression_multiple_uses(%arg0: i32, %arg1: i32) -> i32 {
+  %r = emitc.expression %arg0, %arg1 : (i32, i32) -> i32 {
+    %a = emitc.rem %arg0, %arg1 : (i32, i32) -> i32
+    %b = emitc.add %a, %arg0 : (i32, i32) -> i32
+    %c = emitc.mul %b, %a : (i32, i32) -> i32
+    emitc.yield %c : i32
+  }
+  return %r : i32
+}
+
 func.func @test_for(%arg0 : index, %arg1 : index, %arg2 : index) {
   emitc.for %i0 = %arg0 to %arg1 step %arg2 {
     %0 = emitc.call_opaque "func_const"(%i0) : (index) -> i32
diff --git a/mlir/test/Target/Cpp/expressions.mlir b/mlir/test/Target/Cpp/expressions.mlir
index 433a67ccb3f39..4281f41d0b3fb 100644
--- a/mlir/test/Target/Cpp/expressions.mlir
+++ b/mlir/test/Target/Cpp/expressions.mlir
@@ -1,6 +1,25 @@
 // RUN: mlir-translate -mlir-to-cpp %s | FileCheck %s -check-prefix=CPP-DEFAULT
 // RUN: mlir-translate -mlir-to-cpp -declare-variables-at-top %s | FileCheck %s -check-prefix=CPP-DECLTOP
 
+// CPP-DEFAULT:      bool single_expression(int32_t [[VAL_1:v[0-9]+]], int32_t [[VAL_2:v[0-9]+]], int32_t [[VAL_3:v[0-9]+]]) {
+// CPP-DEFAULT-NEXT: return [[VAL_1]] * 42 - [[VAL_2]] < [[VAL_3]];
+// CPP-DEFAULT-NEXT: }
+
+// CPP-DECLTOP:      bool single_expression(int32_t [[VAL_1:v[0-9]+]], int32_t [[VAL_2:v[0-9]+]], int32_t [[VAL_3:v[0-9]+]]) {
+// CPP-DECLTOP-NEXT: return [[VAL_1]] * 42 - [[VAL_2]] < [[VAL_3]];
+// CPP-DECLTOP-NEXT: }
+
+func.func @single_expression(%arg0: i32, %arg1: i32, %arg2: i32) -> i1 {
+  %e = emitc.expression %arg0, %arg1, %arg2 : (i32, i32, i32) -> i1 {
+    %c42 = "emitc.constant"(){value = 42 : i32} : () -> i32
+    %a = emitc.mul %arg0, %c42 : (i32, i32) -> i32
+    %b = emitc.sub %a, %arg1 : (i32, i32) -> i32
+    %c = emitc.cmp lt, %b, %arg2 :(i32, i32) -> i1
+    emitc.yield %c : i1
+  }
+  return %e : i1
+}
+
 // CPP-DEFAULT:      int32_t single_use(int32_t [[VAL_1:v[0-9]+]], int32_t [[VAL_2:v[0-9]+]], int32_t [[VAL_3:v[0-9]+]], int32_t [[VAL_4:v[0-9]+]]) {
 // CPP-DEFAULT-NEXT:   bool [[VAL_5:v[0-9]+]] = bar([[VAL_1]] * M_PI, [[VAL_3]]) - [[VAL_4]] < [[VAL_2]];
 // CPP-DEFAULT-NEXT:   int32_t [[VAL_6:v[0-9]+]];
@@ -185,7 +204,7 @@ func.func @user_with_expression_trait(%arg0: i32, %arg1: i32, %arg2: i32) -> i32
 }
 
 // CPP-DEFAULT:      int32_t multiple_uses(int32_t [[VAL_1:v[0-9]+]], int32_t [[VAL_2:v[0-9]+]], int32_t [[VAL_3:v[0-9]+]], int32_t [[VAL_4:v[0-9]+]]) {
-// CPP-DEFAULT-NEXT:   bool [[VAL_5:v[0-9]+]] = bar([[VAL_1]] * [[VAL_2]], [[VAL_3]]) - [[VAL_4]] < [[VAL_2]];
+// CPP-DEFAULT-NEXT:   bool [[VAL_5:v[0-9]+]] = bar([[VAL_1]] * [[VAL_2]], [[VAL_3]]) - [[VAL_1]] * [[VAL_2]] < [[VAL_4]];
 // CPP-DEFAULT-NEXT:   int32_t [[VAL_6:v[0-9]+]];
 // CPP-DEFAULT-NEXT:   if ([[VAL_5]]) {
 // CPP-DEFAULT-NEXT:     [[VAL_6]] = [[VAL_1]];
@@ -203,7 +222,7 @@ func.func @user_with_expression_trait(%arg0: i32, %arg1: i32, %arg2: i32) -> i32
 // CPP-DECLTOP-NEXT:   int32_t [[VAL_6:v[0-9]+]];
 // CPP-DECLTOP-NEXT:   bool [[VAL_7:v[0-9]+]];
 // CPP-DECLTOP-NEXT:   int32_t [[VAL_8:v[0-9]+]];
-// CPP-DECLTOP-NEXT:   [[VAL_5]] = bar([[VAL_1]] * [[VAL_2]], [[VAL_3]]) - [[VAL_4]] < [[VAL_2]];
+// CPP-DECLTOP-NEXT:   [[VAL_5]] = bar([[VAL_1]] * [[VAL_2]], [[VAL_3]]) - [[VAL_1]] * [[VAL_2]] < [[VAL_4]];
 // CPP-DECLTOP-NEXT:   ;
 // CPP-DECLTOP-NEXT:   if ([[VAL_5]]) {
 // CPP-DECLTOP-NEXT:     [[VAL_6]] = [[VAL_1]];
@@ -220,8 +239,8 @@ func.func @multiple_uses(%arg0: i32, %arg1: i32, %arg2: i32, %arg3: i32) -> i32
   %e = emitc.expression %arg0, %arg1, %arg2, %arg3 : (i32, i32, i32, i32) -> i1 {
     %a = emitc.mul %arg0, %arg1 : (i32, i32) -> i32
     %b = emitc.call_opaque "bar" (%a, %arg2) : (i32, i32) -> (i32)
-    %c = emitc.sub %b, %arg3 : (i32, i32) -> i32
-    %d = emitc.cmp lt, %c, %arg1 :(i32, i32) -> i1
+    %c = emitc.sub %b, %a : (i32, i32) -> i32
+    %d = emitc.cmp lt, %c, %arg3 :(i32, i32) -> i1
     emitc.yield %d : i1
   }
   %v = "emitc.variable"(){value = #emitc.opaque<"">} : () -> !emitc.lvalue<i32>