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

[Ivan Butygin]

================
@@ -167,7 +167,8 @@ class ConvertOpToGpuRuntimeCallPattern : public ConvertOpToLLVMPattern<OpTy> {
       "mgpuMemAlloc",
       llvmPointerType /* void * */,
       {llvmIntPtrType /* intptr_t sizeBytes */,
-       llvmPointerType /* void *stream */}};
+       llvmPointerType /* void *stream */,
+       llvmInt64Type /* bool isHostShared */}};
----------------
Hardcode84 wrote:

I think it should be `i8`  to match C++ bool, or, better just use i32 (int) on both sides.

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