Provided by: opencl-1.2-man-doc_1.0~svn22836-1.1_all bug


       fast_normalize - Normal vector length 1.

       floatn fast_normalize(floatn p);


       Returns a vector in the same direction as p but with a length of 1.  fast_normalize is
       computed as:

           p * sqrt(3clc)(p.x2 + p.y2 +...)

       The result shall be within 8192 ulps error from the infinitely precise result of:

           if ( any(3clc)(p == 0.0f))
           result = p;
           result = p /  sqrt(3clc)(p.x2 + p.y2 +...);

       with the following exceptions:

        1. If the sum of squares is greater than FLT_MAX then the value of the floating-point
           values in the result vector are undefined.

        2. If the sum of squares is less than FLT_MIN then the implementation may return back p.

        3. If the device is in 'denorms are flushed to zero' mode, individual operand elements
           with magnitude less than sqrt(3clc)(FLT_MIN) may be flushed to zero before proceeding
           with the calculation.


       Built-in geometric functions operate component-wise. The description is per-component.
       floatn is float, float2, float3, or float4 and doublen is double, double2, double3, or
       double4. The built-in geometric functions are implemented using the round to nearest even
       rounding mode.

       The geometric functions can be implemented using contractions such as mad(3clc) or


       OpenCL Specification[1]




       The Khronos Group


       Copyright © 2007-2011 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.


        1. OpenCL Specification
           page 262, section 6.12.5 - Geometric Functions