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

Function writeToSmem

plugin/instanceNormalizationPlugin/instanceNormCommon.h:376–379  ·  view source on GitHub ↗

Source from the content-addressed store, hash-verified

374}
375
376DEVICE_FUNCTION void writeToSmem(float* smem, int32_t idx, float const (&x)[2])
377{
378 reinterpret_cast<float2*>(&smem[2 * idx])[0] = make_float2(x[0], x[1]);
379}
380
381DEVICE_FUNCTION void writeToSmem(float* smem, int32_t idx, float const (&x)[4])
382{

Callers 3

parallelSums_16x2Function · 0.85
parallelSums_8x4Function · 0.85
parallelSumsFunction · 0.85

Calls

no outgoing calls

Tested by

no test coverage detected