vstorea_halfn(3clc)

vstorea_halfn3clc

Convert float vector to half vector and write to a pointer to memory.

void vstorea_halfn(floatn data, size_t offset, __global half *p);

void vstorea_halfn_rte(floatn data, size_t offset, __global half *p);

void vstorea_halfn_rtz(floatn data, size_t offset, __global half *p);

void vstorea_halfn_rtp(floatn data, size_t offset, __global half *p);

void vstorea_halfn_rtn(floatn data, size_t offset, __global half *p);

void vstorea_halfn(floatn data, size_t offset, __local half *p);

void vstorea_halfn_rte(floatn data, size_t offset, __local half *p);

void vstorea_halfn_rtz(floatn data, size_t offset, __local half *p);

void vstorea_halfn_rtp(floatn data, size_t offset, __local half *p);

void vstorea_halfn_rtn(floatn data, size_t offset, __local half *p);

void vstorea_halfn(floatn data, size_t offset, __private half *p);

void vstorea_halfn_rte(floatn data, size_t offset, __private half *p);

void vstorea_halfn_rtz(floatn data, size_t offset, __private half *p);

void vstorea_halfn_rtp(floatn data, size_t offset, __private half *p);

void vstorea_halfn_rtn(floatn data, size_t offset, __private half *p);

If the optional double is supported, you may use doublen instead of floatn.

Overview

The floatn (or doublen if optional double is supported) value igiven by data is converted to a halfn value using the appropriate rounding mode.

For n = 1, 2, 4, 8 and 16, the halfn value is written to the address computed as (p + (offset * n)). The address computed as (p + (offset * n)) must be aligned to sizeof (halfn) bytes.

For n = 3, the half3 value is written to the address computed as (p + (offset * 4)). The address computed as (p + (offset * 4)) must be aligned to sizeof (half) * 4 bytes.

vstorea_halfn uses the default rounding mode. The default rounding mode is round to the nearest even.

An application that wants to use half and halfn types will need to include the #pragma OPENCL EXTENSION cl_khr_fp16(3clc) : enable directive.

Notes

Vector Data Load and Store Functions allow you to read and write vector types from a pointer to memory.

The generic type gentype is used to indicate the built-in data types char, uchar, short, ushort, int, uint, long, ulong, float, or double.

The generic type name gentypen represents n-element vectors of gentype elements. The suffix n is also used in the function names (i.e. vloadn, vstoren, etc.), where n = 2, 3, 4, 8, or 16.

The results of vector data load and store functions are undefined if the address being read from or written to is not correctly aligned. The pointer argument p can be a pointer to global(3clc), local(3clc), or private(3clc) memory for store functions. The pointer argument p can be a pointer to __global, __local, constant(3clc) or __private memory for load functions.

vstorea_half3 writes x, y, z components from a 3-component vector to address (p + (offset * 4)).

Specification

OpenCL Specification[1]

Notes

1.

OpenCL Specification

page 267, section 6.12.7 - Vector Data Load and Store Functions

Releases

5 releases since Debian 8 (Jessie); the text changed 3 times.

  1. 8–9
  2. 10
  3. 11
  4. 13

Authors

The Khronos Group

From opencl-1.2-man-doc 1.0~svn33624-5.1 in Debian 13 (Trixie). Parsed by carta from the page's man source.