[mlir] Fix nvvm integration tests build error (#70113)

#69934 broke integration tests that rely on the
kernel-bare-ptr-calling-convention and host-bare-ptr-calling-convention
flags. This PR brings these flags.

Also the kernel-index-bitwidth flag is removed, as kernel pointer size
depends on the host. Separating host (64-bit) and kernel (32-bit) is not
viable.

Author: grypp

mlir/test/Integration/GPU/CUDA/TensorCore/sm80/transform-mma-sync-matmul-f16-f16-accum.mlir

@@ -1,7 +1,7 @@
 // RUN: mlir-opt %s \
 // RUN:  -transform-interpreter \
 // RUN:  -test-transform-dialect-erase-schedule \
-// RUN:  -test-lower-to-nvvm="kernel-index-bitwidth=32 cubin-chip=sm_80 cubin-features=+ptx76 cubin-format=%gpu_compilation_format" \
+// RUN:  -test-lower-to-nvvm="cubin-chip=sm_80 cubin-features=+ptx76 cubin-format=%gpu_compilation_format" \
 // RUN: | mlir-cpu-runner \
 // RUN:   --shared-libs=%mlir_cuda_runtime \
 // RUN:   --shared-libs=%mlir_runner_utils \

mlir/test/Integration/GPU/CUDA/TensorCore/sm80/transform-mma-sync-matmul-f32.mlir

@@ -11,7 +11,7 @@
 // RUN: mlir-opt %s \
 // RUN:   -transform-interpreter \
 // RUN:   -test-transform-dialect-erase-schedule \
-// RUN:   -test-lower-to-nvvm="kernel-index-bitwidth=32 cubin-chip=sm_80 cubin-features=+ptx76 cubin-format=%gpu_compilation_format" \
+// RUN:   -test-lower-to-nvvm="cubin-chip=sm_80 cubin-features=+ptx76 cubin-format=%gpu_compilation_format" \
 // RUN: | mlir-cpu-runner \
 // RUN:   --shared-libs=%mlir_cuda_runtime \
 // RUN:   --shared-libs=%mlir_runner_utils \

mlir/test/lib/Dialect/GPU/TestLowerToNVVM.cpp

@@ -65,6 +65,18 @@ struct TestLowerToNVVMOptions
       *this, "opt-level",
       llvm::cl::desc("Optimization level for NVVM compilation"),
       llvm::cl::init(2)};
+  PassOptions::Option<bool> kernelUseBarePtrCallConv{
+      *this, "kernel-bare-ptr-calling-convention",
+      llvm::cl::desc(
+          "Whether to use the bareptr calling convention on the kernel "
+          "(warning this should be false until the GPU layering is fixed)"),
+      llvm::cl::init(false)};
+  PassOptions::Option<bool> hostUseBarePtrCallConv{
+      *this, "host-bare-ptr-calling-convention",
+      llvm::cl::desc(
+          "Whether to use the bareptr calling convention on the host (warning "
+          "this should be false until the GPU layering is fixed)"),
+      llvm::cl::init(false)};
 };
 
 //===----------------------------------------------------------------------===//
@@ -105,7 +117,10 @@ void buildCommonPassPipeline(OpPassManager &pm,
 void buildGpuPassPipeline(OpPassManager &pm,
                           const TestLowerToNVVMOptions &options) {
   pm.addNestedPass<gpu::GPUModuleOp>(createStripDebugInfoPass());
-  pm.addNestedPass<gpu::GPUModuleOp>(createConvertGpuOpsToNVVMOps());
+  ConvertGpuOpsToNVVMOpsOptions opt;
+  opt.useBarePtrCallConv = options.kernelUseBarePtrCallConv;
+  opt.indexBitwidth = options.indexBitWidth;
+  pm.addNestedPass<gpu::GPUModuleOp>(createConvertGpuOpsToNVVMOps(opt));
   pm.addNestedPass<gpu::GPUModuleOp>(createCanonicalizerPass());
   pm.addNestedPass<gpu::GPUModuleOp>(createCSEPass());
   pm.addNestedPass<gpu::GPUModuleOp>(createReconcileUnrealizedCastsPass());
@@ -116,7 +131,10 @@ void buildGpuPassPipeline(OpPassManager &pm,
 //===----------------------------------------------------------------------===//
 void buildHostPostPipeline(OpPassManager &pm,
                            const TestLowerToNVVMOptions &options) {
-  pm.addPass(createGpuToLLVMConversionPass());
+  GpuToLLVMConversionPassOptions opt;
+  opt.hostBarePtrCallConv = options.hostUseBarePtrCallConv;
+  opt.kernelBarePtrCallConv = options.kernelUseBarePtrCallConv;
+  pm.addPass(createGpuToLLVMConversionPass(opt));
 
   GpuModuleToBinaryPassOptions gpuModuleToBinaryPassOptions;
   gpuModuleToBinaryPassOptions.compilationTarget = options.cubinFormat;