__lsx_vdiv_hu

LSX intrinsic · group: Integer Computation

Synopsis

__m128i __lsx_vdiv_hu (__m128i a, __m128i b)
#include <lsxintrin.h>
Instruction: vdiv.hu vr, vr, vr
CPU Flags: LSX

Description

Divide each unsigned 16-bit element in a by the corresponding element in b.

Operation

for (int i = 0; i < 8; i++) {
  dst.half[i] = (b.half[i] == 0) ? 0 : ((u16)a.half[i] / (u16)b.half[i]);
}

Tested on real machine.

Latency and Throughput

CPU µarch Latency Throughput (IPC)
3C5000 LA464 17, 21.5 0.07(1/14)
3A6000 LA664 17, 22 0.11(1/9)
3C6000 LA664 17, 21.5 0.13(1/7.5)
2K1000LA LA264 18, 27 0.03(1/31)
2K3000 LA364E 18, 27 0.03(1/31)

All intrinsics in alphabetical order