MCPcopy Create free account
hub / github.com/NVIDIA/TensorRT / add

Function add

plugin/instanceNormalizationPlugin/instanceNormCommon.h:417–424  ·  view source on GitHub ↗

Source from the content-addressed store, hash-verified

415
416template <int32_t N>
417DEVICE_FUNCTION void add(float (&x)[N], float const (&y)[N])
418{
419#pragma unroll
420 for (int32_t i = 0; i < N; ++i)
421 {
422 x[i] += y[i];
423 }
424}
425
426template <int32_t N>
427DEVICE_FUNCTION void normalize(float (&x)[N], float const (&bias)[N], float const (&scale)[N], float const (&m1)[N])

Callers 3

parallelSums_16x2Function · 0.70
parallelSums_8x4Function · 0.70
parallelSumsFunction · 0.70

Calls

no outgoing calls

Tested by

no test coverage detected