I see, so the same doesn't apply to read only workflow though right?
private static int dot(int[] heap, int n, int a, int b) {
var sum = 0;
for (var ii = 0; ii < n; ++ii) {
sum += heap[a + ii] * heap[b + ii];
}
return sum;
}
And one last question.
Can VarHandle byte array usage vectorize yet?
It seems to me it might be fastest to do something like:
private static void mul(byte[] heap, int n, int accum, int src) {
int[] newSrc = copyToTempIntBuf(heap, src, n);
int[] newAccum = copyToTempIntBuf(heap, accum, n);
// c2 needs same indices to vectorise
for (var ii = 0; ii < n; ++ii) {
newAccum[ii] *= newSrc[ii];
}
freeTempBuf(newSrc);
copyBack(heap, newAccum, n);
freeTempBuf(newAccum);
}
But that is extremely circumlocutious and I doubt you can efficiently copy
between byte arrays and int arrays without Unsafe anyway (even using byte
array varhandles.)
--
You received this message because you are subscribed to the Google Groups
"mechanical-sympathy" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
