Function tvm::tirx::gpu_dp4a_op

Function tvm::tirx::gpu_dp4a_op#

Function Documentation#

const Op &tvm::tirx::gpu_dp4a_op()#

Dot product of two int8x4 vectors and add an optional accumulator.

Arguments, in order:

  • args[0]: vec1, The first input vector.

  • args[1]: vec2, The second input vector.

  • args[2]: acc, The accumulator.