sincospif

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

Obtains the sine and cosine of the input data multiplied by π.

$\text{[math]}$

Prototype

inline void sincospif(float x, float *s, float *c)

Parameters

**Table 1** Parameters
Parameter	Input/Output	Description
x	Input	Source operand.
s	Output	Unified Buffer, Global Memory, or stack address, which is used to store the sine of the input data multiplied by π.
c	Output	Unified Buffer, Global Memory, or stack address, which is used to store the cosine of the input data multiplied by π.

Returns

If x*π exceeds the maximum or minimum range of float, the output value is nan.
If the input x is inf, -inf, or nan, the output value is nan.

Restrictions

When this API is used, the maximum number of threads cannot exceed 1024. Otherwise, stack overflow may occur.

For SIMT programming, this API is not supported.

Header File to Be Included

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

#include "simt_api/math_functions.h"

Examples

For SIMD and SIMT programming:

In SIMD and SIMT programming, the address space qualifier must be explicitly used to indicate the address space. __gm__ indicates the Global Memory memory space, and __ubuf__ indicates the Unified Buffer memory space. No address space qualifier is required for the stack space.

__simt_vf__ __launch_bounds__(1024) inline void KernelSinCospi(__gm__ float* s, __gm__ float* c, __gm__ float* x)
{
    int idx = threadIdx.x + blockIdx.x * blockDim.x;
    sincospif(x[idx], s + idx, c + idx); // Obtain the sine and cosine of the idxth element at the source address multiplied by π.
}

Parent topic: Mathematical Functions