vstore_half
Convert float to half and write to a pointer to memory.
Description
The single-precision floating-point value given by data
is first converted to a half value
using the appropriate rounding mode. The half value is then written to address computed as (p
+ offset
). The write address computed as (p
+ offset
) must be 16-bit aligned.
vstore_half
uses the current rounding mode. The default current rounding mode is round to nearest even.
Notes
If extended with cl_khr_fp64, the function is extended to allow a double precision scalar or vector value to be written to memory as half values. In this case, the functions use double instead of float.
Copyright © 2007-2009 The Khronos Group Inc.
Permission is hereby granted, free of charge, to any person obtaining a
copy of this software and/or associated documentation files (the
"Materials"), to deal in the Materials without restriction, including
without limitation the rights to use, copy, modify, merge, publish,
distribute, sublicense, and/or sell copies of the Materials, and to
permit persons to whom the Materials are furnished to do so, subject to
the condition that this copyright notice and permission notice shall be included
in all copies or substantial portions of the Materials.