diff --git a/mlir/include/mlir/Dialect/GPU/GPUOps.td b/mlir/include/mlir/Dialect/GPU/GPUOps.td
--- a/mlir/include/mlir/Dialect/GPU/GPUOps.td
+++ b/mlir/include/mlir/Dialect/GPU/GPUOps.td
@@ -454,6 +454,10 @@
     /// the operand will be dropped.  The block argument must not have any uses.
     void eraseKernelArgument(unsigned index);
 
+    /// Add the given value as a kernel argument. Returns the corresponding newly
+    /// added BlockArgument.
+    BlockArgument addKernelArgument(Value argument);
+
     static StringRef getBlocksKeyword() { return "blocks"; }
     static StringRef getThreadsKeyword() { return "threads"; }
     static StringRef getArgsKeyword() { return "args"; }
diff --git a/mlir/lib/Dialect/GPU/IR/GPUDialect.cpp b/mlir/lib/Dialect/GPU/IR/GPUDialect.cpp
--- a/mlir/lib/Dialect/GPU/IR/GPUDialect.cpp
+++ b/mlir/lib/Dialect/GPU/IR/GPUDialect.cpp
@@ -201,6 +201,8 @@
   result.addOperands(
       {gridSizeX, gridSizeY, gridSizeZ, blockSizeX, blockSizeY, blockSizeZ});
   result.addOperands(operands);
+  // We want to be able to add operands later, for instance due to code motion.
+  result.setOperandListToResizable();
 
   // Create a kernel body region with kNumConfigRegionAttributes + N arguments,
   // where the first kNumConfigRegionAttributes arguments have `index` type and
@@ -449,6 +451,15 @@
   getOperation()->eraseOperand(kNumConfigOperands + index);
 }
 
+BlockArgument LaunchOp::addKernelArgument(Value value) {
+  Block &entryBlock = body().front();
+  Operation *op = getOperation();
+  llvm::SmallVector<Value, 8> operands(op->getOperands());
+  operands.push_back(value);
+  op->setOperands(operands);
+  return entryBlock.addArgument(value.getType());
+}
+
 namespace {
 // Clone any known constants passed as operands to the kernel into its body.
 class PropagateConstantBounds : public OpRewritePattern<LaunchOp> {