__float2bfloat16_rn_sat

Applicability

Product	Supported
Atlas 350 Accelerator Card	√
Atlas A3 training product/Atlas A3 inference product	x
Atlas A2 training product/Atlas A2 inference product	x
Atlas 200I/500 A2 inference product	x
Atlas inference product AI Core	x
Atlas inference product Vector Core	x
Atlas training product	x

Function Usage

In saturation mode, converts a floating-point number to bfloat16 precision in line with the CAST_RINT mode and returns the converted value.

Prototype

inline bfloat16_t __float2bfloat16_rn_sat(const float x)

Parameters

**Table 1** Parameters
Parameter	Input/Output	Description
x	Input	Source operand.

Returns

The bfloat16 data converted from the input in saturation mode in line with the CAST_RINT mode.

Restrictions

Before using this API, set bit[60] of the CTRL register to 0. Otherwise, the saturation mode does not take effect. For details, see Methods of Controlling Saturation Behaviors.

For SIMT programming, this API is not supported.

Header File to Be Included

To use this API, the simt_api/asc_bf16.h header file must be included.

#include "simt_api/asc_bf16.h"

Examples

For SIMD and SIMT programming:

__simt_vf__ __launch_bounds__(1024) inline void kernel__float2bfloat16_rn_sat(__gm__ bfloat16_t* dst, __gm__ float* x)
{
    int idx = threadIdx.x + blockIdx.x * blockDim.x;
    dst[idx] = __float2bfloat16_rn_sat(x[idx]);
}

Parent topic: Type Conversion