Comparison Operators
The equality and inequality comparison operands can have mixed signedness, but they must be of the same size. The comparison operators for less-than and greater-than must be of the same sign and size. For example:
- The nearest common ancestor is returned for compare for equal/not-equal operations:Iu8vec8 A; Is8vec8 B; I8vec8 C; C = cmpneq(A,B);
- Type cast needed for different-sized elements for equal/not-equal comparisons:Iu8vec8 A, C; Is16vec4 B; C = cmpeq(A,(Iu8vec8)B);
- Type cast needed for sign or size differences for less-than and greater-than comparisons:Iu16vec4 A; Is16vec4 B, C; C = cmpge((Is16vec4)A,B); C = cmpgt(B,C);
Inequality Comparison Symbols and Corresponding Intrinsics
Comparison
| Operators
| Syntax
| Intrinsic
|
|---|---|---|---|
Equality
| cmpeq | R = cmpeq(A, B) | _mm_cmpeq_pi32 _mm_cmpeq_pi16 _mm_cmpeq_pi8 |
Inequality
| cmpneq | R = cmpneq(A, B) | _mm_cmpeq_pi32 _mm_cmpeq_pi16 _mm_cmpeq_pi8 _mm_andnot_si64 |
Greater Than
| cmpgt | R = cmpgt(A, B) | _mm_cmpgt_pi32 _mm_cmpgt_pi16 _mm_cmpgt_pi8 |
Greater Thanor Equal To
| cmpge | R = cmpge(A, B) | _mm_cmpgt_pi32 _mm_cmpgt_pi16 _mm_cmpgt_pi8 _mm_andnot_si64 |
Less Than
| cmplt | R = cmplt(A, B) | _mm_cmpgt_pi32 _mm_cmpgt_pi16 _mm_cmpgt_pi8 |
Less Thanor Equal To
| cmple | R = cmple(A, B) | _mm_cmpgt_pi32 _mm_cmpgt_pi16 _mm_cmpgt_pi8 _mm_andnot_si64 |
Compare Operator Overloading
Comparison operators have the restriction that the operands must be the size and sign as listed in the following table.
R
| Comparison
| Operand A
| Operand B
|
|---|---|---|---|
I32vec2 R | cmpeq cmpne | I[s|u]32vec2 B | I[s|u]32vec2 B |
I16vec4 R | I[s|u]16vec4 B | I[s|u]16vec4 B | |
I8vec8 R | I[s|u]8vec8 B | I[s|u]8vec8 B | |
I32vec2 R | cmpgt cmpge cmplt cmple | Is32vec2 B | Is32vec2 B |
I16vec4 R | Is16vec4 B | Is16vec4 B | |
I8vec8 R | Is8vec8 B | Is8vec8 B |