__bfloat162int_ru

Applicability

Function Usage

Converts bfloat16 data to a signed integer in line with the CAST_CEIL mode and returns the converted value.

Prototype

1

inline int __bfloat162int_ru(const bfloat16_t x)

Parameters

Returns

The signed integer converted from the input in line with the CAST_CEIL mode. Special scenarios are described as follows:

• If x is nan, the return value is 0.
• If x is inf, the return value is 2147483647.
• If x is -inf, the return value is -2147483648.

Restrictions

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.

1

#include "simt_api/asc_bf16.h"

Examples

1
2
3
4
5

__simt_vf__ __launch_bounds__(1024) inline void kernel__bfloat162int_ru(__gm__ int32_t* dst, __gm__ bfloat16_t* x)
{
    int idx = threadIdx.x + blockIdx.x * blockDim.x;
    dst[idx] = __bfloat162int_ru(x[idx]);
}