[mlir] Create a gpu.module operation for the GPU Dialect.

Summary:
This is based on the use of code constantly checking for an attribute on
a model and instead represents the distinct operaion with a different
op. Instead, this op can be used to provide better filtering.

Reviewers: herhut, mravishankar, antiagainst, rriddle

Reviewed By: herhut, antiagainst, rriddle

Subscribers: liufengdb, aartbik, jholewinski, mgorny, mehdi_amini, rriddle, jpienaar, burmako, shauheen, antiagainst, nicolasvasilache, csigg, arpith-jacob, mgester, lucyrfox, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D72336

2 files changed, 15 insertions, 19 deletions

diff --git a/mlir/lib/Conversion/GPUToCUDA/ConvertKernelFuncToCubin.cpp b/mlir/lib/Conversion/GPUToCUDA/ConvertKernelFuncToCubin.cpp
index 66a2e66f99a..b111c96313c 100644
--- a/mlir/lib/Conversion/GPUToCUDA/ConvertKernelFuncToCubin.cpp
+++ b/mlir/lib/Conversion/GPUToCUDA/ConvertKernelFuncToCubin.cpp
@@ -46,18 +46,15 @@ static constexpr const char *kCubinAnnotation = "nvvm.cubin";
 /// IR and further to PTX. A user provided CubinGenerator compiles the PTX to
 /// GPU binary code, which is then attached as an attribute to the function. The
 /// function body is erased.
-class GpuKernelToCubinPass : public ModulePass<GpuKernelToCubinPass> {
+class GpuKernelToCubinPass
+    : public OperationPass<GpuKernelToCubinPass, gpu::GPUModuleOp> {
 public:
   GpuKernelToCubinPass(
       CubinGenerator cubinGenerator = compilePtxToCubinForTesting)
       : cubinGenerator(cubinGenerator) {}
 
-  void runOnModule() override {
-    ModuleOp module = getModule();
-    if (!module.getAttrOfType<UnitAttr>(
-            gpu::GPUDialect::getKernelModuleAttrName()) ||
-        !module.getName())
-      return;
+  void runOnOperation() override {
+    gpu::GPUModuleOp module = getOperation();
 
     // Make sure the NVPTX target is initialized.
     LLVMInitializeNVPTXTarget();
@@ -71,8 +68,8 @@ public:
 
     // Translate the module to CUBIN and attach the result as attribute to the
     // module.
-    if (auto cubinAttr = translateGpuModuleToCubinAnnotation(
-            *llvmModule, module.getLoc(), *module.getName()))
+    if (auto cubinAttr = translateGPUModuleToCubinAnnotation(
+            *llvmModule, module.getLoc(), module.getName()))
       module.setAttr(kCubinAnnotation, cubinAttr);
     else
       signalPassFailure();
@@ -92,7 +89,7 @@ private:
                                   StringRef name);
 
   /// Translates llvmModule to cubin and returns the result as attribute.
-  StringAttr translateGpuModuleToCubinAnnotation(llvm::Module &llvmModule,
+  StringAttr translateGPUModuleToCubinAnnotation(llvm::Module &llvmModule,
                                                  Location loc, StringRef name);
 
   CubinGenerator cubinGenerator;
@@ -149,7 +146,7 @@ OwnedCubin GpuKernelToCubinPass::convertModuleToCubin(llvm::Module &llvmModule,
   return cubinGenerator(ptx, loc, name);
 }
 
-StringAttr GpuKernelToCubinPass::translateGpuModuleToCubinAnnotation(
+StringAttr GpuKernelToCubinPass::translateGPUModuleToCubinAnnotation(
     llvm::Module &llvmModule, Location loc, StringRef name) {
   auto cubin = convertModuleToCubin(llvmModule, loc, name);
   if (!cubin)
@@ -157,7 +154,7 @@ StringAttr GpuKernelToCubinPass::translateGpuModuleToCubinAnnotation(
   return StringAttr::get({cubin->data(), cubin->size()}, loc->getContext());
 }
 
-std::unique_ptr<OpPassBase<ModuleOp>>
+std::unique_ptr<OpPassBase<gpu::GPUModuleOp>>
 mlir::createConvertGPUKernelToCubinPass(CubinGenerator cubinGenerator) {
   return std::make_unique<GpuKernelToCubinPass>(cubinGenerator);
 }
diff --git a/mlir/lib/Conversion/GPUToCUDA/ConvertLaunchFuncToCudaCalls.cpp b/mlir/lib/Conversion/GPUToCUDA/ConvertLaunchFuncToCudaCalls.cpp
index 41f69d6e21d..31024d2881b 100644
--- a/mlir/lib/Conversion/GPUToCUDA/ConvertLaunchFuncToCudaCalls.cpp
+++ b/mlir/lib/Conversion/GPUToCUDA/ConvertLaunchFuncToCudaCalls.cpp
@@ -132,9 +132,9 @@ public:
 
     // GPU kernel modules are no longer necessary since we have a global
     // constant with the CUBIN data.
-    for (auto m : llvm::make_early_inc_range(getModule().getOps<ModuleOp>()))
-      if (m.getAttrOfType<UnitAttr>(gpu::GPUDialect::getKernelModuleAttrName()))
-        m.erase();
+    for (auto m :
+         llvm::make_early_inc_range(getModule().getOps<gpu::GPUModuleOp>()))
+      m.erase();
   }
 
 private:
@@ -343,8 +343,8 @@ void GpuLaunchFuncToCudaCallsPass::translateGpuLaunchCalls(
                                                builder.getI32IntegerAttr(0));
   // Create an LLVM global with CUBIN extracted from the kernel annotation and
   // obtain a pointer to the first byte in it.
-  auto kernelModule =
-      getModule().lookupSymbol<ModuleOp>(launchOp.getKernelModuleName());
+  auto kernelModule = getModule().lookupSymbol<gpu::GPUModuleOp>(
+      launchOp.getKernelModuleName());
   assert(kernelModule && "expected a kernel module");
 
   auto cubinAttr = kernelModule.getAttrOfType<StringAttr>(kCubinAnnotation);
@@ -354,8 +354,7 @@ void GpuLaunchFuncToCudaCallsPass::translateGpuLaunchCalls(
     return signalPassFailure();
   }
 
-  assert(kernelModule.getName() && "expected a named module");
-  SmallString<128> nameBuffer(*kernelModule.getName());
+  SmallString<128> nameBuffer(kernelModule.getName());
   nameBuffer.append(kCubinStorageSuffix);
   Value data = LLVM::createGlobalString(
       loc, builder, nameBuffer.str(), cubinAttr.getValue(),