MCPcopy Create free account
hub / github.com/deepspeedai/DeepSpeed / param_update_kernel

Function param_update_kernel

csrc/xpu/common/custom_cuda_kernel.dp.cpp:44–50  ·  view source on GitHub ↗

Source from the content-addressed store, hash-verified

42}
43
44void param_update_kernel(const float* input, sycl::half* output, int size)
45{
46 auto item_ct1 = sycl::ext::oneapi::experimental::this_nd_item<3>();
47 int id = item_ct1.get_group(2) * item_ct1.get_local_range(2) + item_ct1.get_local_id(2);
48
49 if (id < size) { output[id] = (sycl::half)input[id]; }
50}
51
52void launch_param_update(const float* input, sycl::half* output, int size, sycl::queue* stream)
53{

Callers 1

launch_param_updateFunction · 0.85

Calls

no outgoing calls

Tested by

no test coverage detected