Is this page helpful?

cast_transpose_noop.h#

Transpose functions with no-op flag.

Functions

void nvte_transpose_with_noop(
const NVTETensor input,
const NVTETensor noop,
NVTETensor output,
cudaStream_t stream,
)#

Transposes the input.

Parameters:

  • input – [in] Input tensor to be cast.

  • noop – [in] If this single element tensor has non-zero value, kernel will exit immediately.

  • output – [inout] Output tensor.

  • stream – [in] CUDA stream used for the operation.

void nvte_cast_transpose_with_noop(
const NVTETensor input,
const NVTETensor noop,
NVTETensor output,
cudaStream_t stream,
)#

Casts and transposes the input.

Parameters:

  • input – [in] Input tensor to be cast.

  • noop – [in] If this single element tensor has non-zero value, kernel will exit immediately.

  • output – [inout] Output quantized tensor.

  • stream – [in] CUDA stream used for the operation.