OpenCL 中是否有等同于 native_recip() 的 CUDA？
Is there a CUDA equivalent of native_recip() in OpenCL?
OpenCL 有一个名为 native_recip:
的内置函数
gentype native_recip(gentype x);
native_recip computes reciprocal over an implementation-defined range. The maximum error is implementation-defined.
The vector versions of the math functions operate component-wise. The description is per-component.
The built-in math functions are not affected by the prevailing rounding mode in the calling environment, and always return the same value as they would if called with the round to nearest even rounding mode.
CUDA 中是否有与该函数等效的函数？
如评论中所述，double 的 __frcp_rn() for float's and __drcp_rn()；以及向量类型（例如 float4）的实现，使得 frcp/drcp 被逐元素应用。
注：“rcp”是“倒数”的缩写，“rn”是舍入模式"round to nearest even"。
OpenCL 有一个名为 native_recip:
gentype native_recip(gentype x);
native_recipcomputes reciprocal over an implementation-defined range. The maximum error is implementation-defined.The vector versions of the math functions operate component-wise. The description is per-component.
The built-in math functions are not affected by the prevailing rounding mode in the calling environment, and always return the same value as they would if called with the round to nearest even rounding mode.
CUDA 中是否有与该函数等效的函数？
如评论中所述，double 的 __frcp_rn() for float's and __drcp_rn()；以及向量类型（例如 float4）的实现，使得 frcp/drcp 被逐元素应用。
注：“rcp”是“倒数”的缩写，“rn”是舍入模式"round to nearest even"。