NAME
    floor - returns largest integer not greater than a scalar or each vector
    component.

SYNPOSIS
      float  floor(float a);
      float1 floor(float1 a);
      float2 floor(float2 a);
      float3 floor(float3 a);
      float4 floor(float4 a);
 
      half   floor(half a);
      half1  floor(half1 a);
      half2  floor(half2 a);
      half3  floor(half3 a);
      half4  floor(half4 a);
 
      fixed  floor(fixed a);
      fixed1 floor(fixed1 a);
      fixed2 floor(fixed2 a);
      fixed3 floor(fixed3 a);
      fixed4 floor(fixed4 a);

PARAMETERS
    a       Vector or scalar of which to determine the floor.

DESCRIPTION
    Returns the floor or largest integer not greater than a scalar or each
    vector component.

REFERENCE IMPLEMENTATION
    floor for a float3 vector could be implemented like this.

      float3 floor(float3 v)
      {
        float3 rv;
        int i;

        for (i=0; i<3; i++) {
          rv[i] = v[i] - frac(v[i]);
        }
        return rv;
      }

PROFILE SUPPORT
    floor is supported in all profiles except fp20.

SEE ALSO
    the ceil manpage, the round manpage

