Hello! Maybe I did not quite understand the logic of the work IJobParallelFor or something I missed, but I have the following question. Can we somehow transfer an NativeArray to job without batching. For example:
struct SomeParallelJob : IJobParallelFor
{
public NativeArray<CustomStruct> dontBatchMe; //<-- I do not want to batching this array
public NativeArray<int> batchMeBaby;
public void Execute(int index)
{
int tmp = batchMeBaby[index];
for (int i = 0; i < dontBatchMe.Length; i++)
{
tmp += dontBatchMe[i].someFieldFromStruct;
}
batchMeBaby[index] = tmp;
}
}
And you can still 1 question, so as not to create a new thread. Can we get one output from Job? Approximately how here (Now I’m using a native array of long 1 for this):
struct SomeJob : IJob
{
[ReadOnly] public NativeArray<int> data;
public int counter;
public void Execute()
{
for (int i = 0; i < data.Length; i++)
{
counter++;
}
}
}
protected override void OnUpdate()
{
NativeArray<int> data = new NativeArray<int>(100000, Allocator.TempJob);
var someJob = new SomeJob
{
data = data,
};
JobHandle handle = someJob.Schedule(data.Length, 1250);
handle.Complete();
Debug.Log("Count: " + someJob.counter); // <--- this value always default int (0)
data.Dispose();
}
That’s an acceptable approach. (Which is just a pointer to some allocated, presumably TempJob, memory.) - Any solution that’s specific to one int would just be identical to that anyway. (Although you could make an NativeInt container if you wanted, it’s probably doesn’t buy you anything.)
