gpuGather

Fully qualified name: isaacsim::physics::newton::tensors::gpuGather

template<typename LaunchFn>
bool isaacsim::physics::newton::tensors::gpuGather(

LaunchFn &&launch,

const TensorDesc *dstTensor,

int n,

size_t elemFloats,

float *stagingBuffer,

)

Launches a CUDA gather kernel and stages the result to dstTensor.

Parameters:

• launch – Callable(float* gpuDst, int count) → bool, launches the CUDA kernel.

• dstTensor – Caller’s output tensor (may be CPU or GPU).

• n – Number of elements to gather.

• elemFloats – Number of float components per element (for byte size computation).

• stagingBuffer – Pre-allocated device buffer, used when dstTensor is on CPU.