[mlir][gpu] Refactor GpuOpsToROCDLOps pass interface (NFC) (#157402)

This PR deletes the `createLowerGpuOpsToROCDLOpsPass` constructor from
the .td file, making the `createConvertGpuOpsToROCDLOps` pass available
to users. This has the following effects:

1. `createLowerGpuOpsToROCDLOpsPass` is not available anymore. Instead,
`createConvertGpuOpsToROCDLOps` should be used. This makes the interface
consistent with ConvertGpuOpsToNVVMOps.

2. To call `createConvertGpuOpsToROCDLOps`, the options must be passed
via ConvertGpuOpsToROCDLOpsOptions. This has the side effect of
making the `allowed-dialects` option available, which was not accessible
via C++ before.

Author: pabloantoniom

mlir/include/mlir/Conversion/GPUToROCDL/GPUToROCDLPass.h

@@ -13,6 +13,7 @@
 #include <memory>
 
 namespace mlir {
+class Pass;
 class LLVMTypeConverter;
 class ConversionTarget;
 class RewritePatternSet;
@@ -42,16 +43,6 @@ void populateGpuToROCDLConversionPatterns(const LLVMTypeConverter &converter,
 /// Configure target to convert from the GPU dialect to ROCDL.
 void configureGpuToROCDLConversionLegality(ConversionTarget &target);
 
-/// Creates a pass that lowers GPU dialect operations to ROCDL counterparts. The
-/// index bitwidth used for the lowering of the device side index computations
-/// is configurable.
-std::unique_ptr<OperationPass<gpu::GPUModuleOp>>
-createLowerGpuOpsToROCDLOpsPass(
-    const std::string &chipset = "gfx900",
-    unsigned indexBitwidth = kDeriveIndexBitwidthFromDataLayout,
-    bool useBarePtrCallConv = false,
-    gpu::amd::Runtime runtime = gpu::amd::Runtime::Unknown);
-
 } // namespace mlir
 
 #endif // MLIR_CONVERSION_GPUTOROCDL_GPUTOROCDLPASS_H_

mlir/include/mlir/Conversion/Passes.td

@@ -624,7 +624,6 @@ def ConvertGpuOpsToNVVMOps : Pass<"convert-gpu-to-nvvm", "gpu::GPUModuleOp"> {
 
 def ConvertGpuOpsToROCDLOps : Pass<"convert-gpu-to-rocdl", "gpu::GPUModuleOp"> {
   let summary = "Generate ROCDL operations for gpu operations";
-  let constructor = "mlir::createLowerGpuOpsToROCDLOpsPass()";
   let dependentDialects = [
     "ROCDL::ROCDLDialect",
     "amdgpu::AMDGPUDialect",

mlir/lib/Conversion/GPUToROCDL/LowerGpuOpsToROCDLOps.cpp

@@ -36,7 +36,6 @@
 #include "mlir/Dialect/MemRef/IR/MemRef.h"
 #include "mlir/Dialect/Vector/IR/VectorOps.h"
 #include "mlir/IR/BuiltinAttributes.h"
-#include "mlir/Pass/Pass.h"
 #include "mlir/Transforms/DialectConversion.h"
 #include "mlir/Transforms/GreedyPatternRewriteDriver.h"
 
@@ -287,19 +286,7 @@ struct GPUShuffleOpLowering : public ConvertOpToLLVMPattern<gpu::ShuffleOp> {
 // code.
 struct LowerGpuOpsToROCDLOpsPass final
     : public impl::ConvertGpuOpsToROCDLOpsBase<LowerGpuOpsToROCDLOpsPass> {
-  LowerGpuOpsToROCDLOpsPass() = default;
-  LowerGpuOpsToROCDLOpsPass(const std::string &chipset, unsigned indexBitwidth,
-                            bool useBarePtrCallConv,
-                            gpu::amd::Runtime runtime) {
-    if (this->chipset.getNumOccurrences() == 0)
-      this->chipset = chipset;
-    if (this->indexBitwidth.getNumOccurrences() == 0)
-      this->indexBitwidth = indexBitwidth;
-    if (this->useBarePtrCallConv.getNumOccurrences() == 0)
-      this->useBarePtrCallConv = useBarePtrCallConv;
-    if (this->runtime.getNumOccurrences() == 0)
-      this->runtime = runtime;
-  }
+  using Base::Base;
 
   void getDependentDialects(DialectRegistry &registry) const override {
     Base::getDependentDialects(registry);
@@ -499,12 +486,3 @@ void mlir::populateGpuToROCDLConversionPatterns(
 
   populateMathToROCDLConversionPatterns(converter, patterns);
 }
-
-std::unique_ptr<OperationPass<gpu::GPUModuleOp>>
-mlir::createLowerGpuOpsToROCDLOpsPass(const std::string &chipset,
-                                      unsigned indexBitwidth,
-                                      bool useBarePtrCallConv,
-                                      gpu::amd::Runtime runtime) {
-  return std::make_unique<LowerGpuOpsToROCDLOpsPass>(
-      chipset, indexBitwidth, useBarePtrCallConv, runtime);
-}