[Mlir-commits] [mlir] [MLIR] Pass hostShared flag in gpu.alloc op to runtime wrappers (PR #66401)

[Guray Ozen]

================
@@ -811,8 +811,13 @@ LogicalResult ConvertAllocOpToGpuRuntimeCallPattern::matchAndRewrite(
   // descriptor.
   Type elementPtrType = this->getElementPtrType(memRefType);
   auto stream = adaptor.getAsyncDependencies().front();
+
+  auto isHostShared = rewriter.create<mlir::LLVM::ConstantOp>(
+      loc, llvmInt64Type, rewriter.getI64IntegerAttr(isShared));
+
   Value allocatedPtr =
-      allocCallBuilder.create(loc, rewriter, {sizeBytes, stream}).getResult();
+      allocCallBuilder.create(loc, rewriter, {sizeBytes, stream, isHostShared})
----------------
grypp wrote:

I thought we had a consensus about avoiding the use of `host_shared` and `async` because of USM's limitation in handling asynchronous data allocation. Are we still on the same page with this? 

https://github.com/llvm/llvm-project/pull/65539#discussion_r1320546363

https://github.com/llvm/llvm-project/pull/66401