tighten language for NaN to integer conversions (KhronosGroup#1546)

Author: bashbaug

OpenCL_C.txt

@@ -1804,13 +1804,13 @@ specified by <<C99-spec,section 6.3 of the C99 Specification>>.
 When converting from a floating-point type to integer type, the behavior is
 implementation-defined.
 
-Conversions to integer type may opt to convert using the optional saturated
+Conversions to integer types may opt to convert using the optional saturated
 mode by appending the `_sat` modifier to the conversion function name.
 When in saturated mode, values that are outside the representable range
-shall clamp to the nearest representable value in the destination format.
-(NaN should be converted to 0).
+are clamped to the nearest representable value in the destination format,
+and NaN is converted to zero.
 
-Conversions to floating-point type shall conform to IEEE-754 rounding rules.
+Conversions to floating-point types conform to IEEE-754 rounding rules.
 The `_sat` modifier may not be used for conversions to floating-point
 formats.
 
@@ -1824,7 +1824,7 @@ Example 1:
 ----------
 short4 s;
 
-// negative values clamped to 0
+// negative values clamped to zero
 ushort4 u = convert_ushort4_sat( s );
 
 // values > CHAR_MAX converted to CHAR_MAX
@@ -1843,7 +1843,7 @@ float4 f;
 int4 i = convert_int4( f );
 
 // values > INT_MAX clamp to INT_MAX, values < INT_MIN clamp
-// to INT_MIN. NaN should produce 0.
+// to INT_MIN, and NaN is converted to zero
 // The _rtz_ rounding mode is used to produce the integer values.
 int4 i2 = convert_int4_sat( f );
 

man/static/convert_T.txt

@@ -139,13 +139,13 @@ specified by <<C99-spec,section 6.3 of the C99 Specification>>.
 When converting from a floating-point type to integer type, the behavior is
 implementation-defined.
 
-Conversions to integer type may opt to convert using the optional saturated
+Conversions to integer types may opt to convert using the optional saturated
 mode by appending the _sat modifier to the conversion function name.
 When in saturated mode, values that are outside the representable range
-shall clamp to the nearest representable value in the destination format.
-(NaN should be converted to 0).
+are clamped to the nearest representable value in the destination format,
+and NaN is converted to zero.
 
-Conversions to floating-point type shall conform to IEEE-754 rounding rules.
+Conversions to floating-point types conform to IEEE-754 rounding rules.
 The `_sat` modifier may not be used for conversions to floating-point
 formats.
 
@@ -159,7 +159,7 @@ Example 1:
 ----------
 short4 s;
 
-// negative values clamped to 0
+// negative values clamped to zero
 ushort4 u = convert_ushort4_sat( s );
 
 // values > CHAR_MAX converted to CHAR_MAX
@@ -178,7 +178,7 @@ float4 f;
 int4 i = convert_int4( f );
 
 // values > INT_MAX clamp to INT_MAX, values < INT_MIN clamp
-// to INT_MIN. NaN should produce 0.
+// to INT_MIN, and NaN is converted to zero
 // The _rtz_ rounding mode is used to produce the integer values.
 int4 i2 = convert_int4_sat( f );