Using the fact that a negative 64-bit integer is also a negative
32-bit integer when truncated, you could use 2x SSE2 32-bit
comparison.
- one for [v0 , v1] - [vlow, vlow] compared to [0, 0] (result
should be >= 0, signed comparison)
- one for [v0 , v1] - [vhigh, vhigh] compared to [0, 0] (result
should be < 0, signed comparison)
Then apply logical and on the masks.
This doesn't work if vhigh - vlow spans a too large area.
On Monday, 8 December 2014 at 16:32:50 UTC, Martin Nowak wrote:
I want to do bounds checking of 2 (4 on avx) ulongs (64-bit) at
a time.
ulong2 vval = [v0, v1];
ulong2 vlow = [low, low];
ulong2 vhigh = [high, high];
int res = PMOVMSKB(vval >= vlow & vval < vhigh);
I figured out sort of a solution, but it seems way too
complicated, because there is only signed comparison.
Usually (scalar) I'd use this, which makes use of unsigned wrap
to safe one conditional
immutable size = cast(ulong)(vhigh - vlow);
if (cast(ulong)(v0 - vlow) < size) {}
if (cast(ulong)(v1 - vlow) < size) {}
over
if (v0 >= vlow && v0 < vhigh) {}
Maybe this can be used on SIMD too (saturated sub or so)?
-Martin
