[llvm] [offload] add support for aligned allocations (PR #203353)

[via llvm-commits]

================
@@ -397,6 +397,20 @@ struct CUDADeviceTy : public GenericDeviceTy {
       return Err;
     MaxBlockSharedMemSize = MaxSharedMem;
 
+    CUmemAllocationProp Prop = {};
+    Prop.type = CU_MEM_ALLOCATION_TYPE_PINNED;
+    Prop.location.type = CU_MEM_LOCATION_TYPE_DEVICE;
+    Prop.location.id = DeviceId;
+
+    Res = cuMemGetAllocationGranularity(&Granularity, &Prop,
----------------
EuphoricThinking wrote:

The Memory Manager will eventually handle aligned allocations. This PR is the groundwork for upcoming changes.

https://github.com/llvm/llvm-project/pull/203353