Consider the following kernel:
__constant float4 oneVec = (float4)(1.0f, 1.0f, 1.0f, 1.0f);
__kernel __attribute__((vec_type_hint(float4)))
void inverter2(__global float4* input, __global float4* output)
{
int tid = get_global_id(0);
output[tid] = oneVec – input[tid];
output[tid].w = input[tid].w;
output[tid] = sqrt(output[tid]);
}
For this example of the explicit vector code, the extraction of the
w component is very costly. The reason is that the next vector
operation forces reloading the same vector from memory. Consider loading
a vector once and performing all changes by use of vector operations even
for a single component.
In this specific case, two changes are required:
oneVec, so that its w component
is zero, causing only a sign flip in the w
component of the input vector.float representation to manually flip the sign
bit of the w component back.As a result, the kernel appears as follows:
__constant float4 oneVec = (float4)(1.0f, 1.0f, 1.0f, 0.0f);
__constant int4 signChanger = (int4)(0, 0, 0, 0x80000000);
__kernel __attribute__((vec_type_hint(float4)))
void inverter3(__global float4* input, __global float4* output)
{
int tid = get_global_id(0);
output[tid] = oneVec – input[tid];
output[tid] = as_float4(as_int4(output[tid]) ^ signChanger);
output[tid] = sqrt(output[tid]);
}
At the cost of another constant vector, this implementation performs
all the required operations addressing only full vectors. All the computations
might also be performed in float8.