[Mlir-commits] [mlir] [MLIR][GPUToLLVMSPV] Update ConvertGpuOpsToLLVMSPVOps's option (PR #118818)

[llvmlistbot at llvm.org]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-mlir

Author: Jefferson Le Quellec (jle-quel)

<details>
<summary>Changes</summary>

## Description

This PR updates the ConvertGpuOpsToLLVMSPVOps's option by replacing the `index-bitwidth` with a boolean option `use-64bit-index` (similar to the `ConvertGPUToSPIRV` option).

The reason for this modification is because the `ConvertGpuOpsToLLVMSPVOps`:
> Generate LLVM operations to be ingested by a SPIR-V backend for gpu operations

In the context of SPIR-V specifications only two physical addressing models are allowed: `Physical32` and `Physical64`.

This change guarantees output sanity by preventing invalid or unsupported index bitwidths from being specified.

---
Full diff: https://github.com/llvm/llvm-project/pull/118818.diff


3 Files Affected:

- (modified) mlir/include/mlir/Conversion/Passes.td (+3-3) 
- (modified) mlir/lib/Conversion/GPUToLLVMSPV/GPUToLLVMSPV.cpp (+1-3) 
- (modified) mlir/test/Conversion/GPUToLLVMSPV/gpu-to-llvm-spv.mlir (+4-2) 


``````````diff
diff --git a/mlir/include/mlir/Conversion/Passes.td b/mlir/include/mlir/Conversion/Passes.td
index d722bd1f3e296a..8835e0a9099fdd 100644
--- a/mlir/include/mlir/Conversion/Passes.td
+++ b/mlir/include/mlir/Conversion/Passes.td
@@ -553,9 +553,9 @@ def ConvertGpuOpsToLLVMSPVOps : Pass<"convert-gpu-to-llvm-spv", "gpu::GPUModuleO
     "Generate LLVM operations to be ingested by a SPIR-V backend for gpu operations";
   let dependentDialects = ["LLVM::LLVMDialect"];
   let options = [
-    Option<"indexBitwidth", "index-bitwidth", "unsigned",
-           /*default=kDeriveIndexBitwidthFromDataLayout*/"0",
-           "Bitwidth of the index type, 0 to use size of machine word">,
+    Option<"use64bitIndex", "use-64bit-index",
+           "bool", /*default=*/"false",
+           "Use 64-bit integers to convert index types">,
   ];
 }
 
diff --git a/mlir/lib/Conversion/GPUToLLVMSPV/GPUToLLVMSPV.cpp b/mlir/lib/Conversion/GPUToLLVMSPV/GPUToLLVMSPV.cpp
index 03745f4537e99e..a68c0153df4432 100644
--- a/mlir/lib/Conversion/GPUToLLVMSPV/GPUToLLVMSPV.cpp
+++ b/mlir/lib/Conversion/GPUToLLVMSPV/GPUToLLVMSPV.cpp
@@ -409,9 +409,7 @@ struct GPUToLLVMSPVConversionPass final
     RewritePatternSet patterns(context);
 
     LowerToLLVMOptions options(context);
-    if (indexBitwidth != kDeriveIndexBitwidthFromDataLayout)
-      options.overrideIndexBitwidth(indexBitwidth);
-
+    options.overrideIndexBitwidth(this->use64bitIndex ? 64 : 32);
     LLVMTypeConverter converter(context, options);
     LLVMConversionTarget target(*context);
 
diff --git a/mlir/test/Conversion/GPUToLLVMSPV/gpu-to-llvm-spv.mlir b/mlir/test/Conversion/GPUToLLVMSPV/gpu-to-llvm-spv.mlir
index 16b692b9689398..e75225d6d54f55 100644
--- a/mlir/test/Conversion/GPUToLLVMSPV/gpu-to-llvm-spv.mlir
+++ b/mlir/test/Conversion/GPUToLLVMSPV/gpu-to-llvm-spv.mlir
@@ -1,6 +1,8 @@
-// RUN: mlir-opt -pass-pipeline="builtin.module(gpu.module(convert-gpu-to-llvm-spv))" -split-input-file -verify-diagnostics %s \
+// RUN: mlir-opt -pass-pipeline="builtin.module(gpu.module(convert-gpu-to-llvm-spv{use-64bit-index=true}))" -split-input-file -verify-diagnostics %s \
 // RUN: | FileCheck --check-prefixes=CHECK-64,CHECK %s
-// RUN: mlir-opt -pass-pipeline="builtin.module(gpu.module(convert-gpu-to-llvm-spv{index-bitwidth=32}))" -split-input-file -verify-diagnostics %s \
+// RUN: mlir-opt -pass-pipeline="builtin.module(gpu.module(convert-gpu-to-llvm-spv))" -split-input-file -verify-diagnostics %s \
+// RUN: | FileCheck --check-prefixes=CHECK-32,CHECK %s
+// RUN: mlir-opt -pass-pipeline="builtin.module(gpu.module(convert-gpu-to-llvm-spv{use-64bit-index=false}))" -split-input-file -verify-diagnostics %s \
 // RUN: | FileCheck --check-prefixes=CHECK-32,CHECK %s
 
 gpu.module @builtins {

``````````

</details>


https://github.com/llvm/llvm-project/pull/118818