__lsx_vdiv_w
LSX intrinsic · group: Integer Computation
Synopsis
__m128i __lsx_vdiv_w (__m128i a, __m128i b)
#include <lsxintrin.h>
Instruction: vdiv.w vr, vr, vr
CPU Flags: LSX
Description
Divide each signed 32-bit element in a by the corresponding element in b.
Operation
for (int i = 0; i < 4; i++) {
dst.word[i] = (b.word[i] == 0) ? 0 : ((s32)a.word[i] / (s32)b.word[i]);
}
Tested on real machine.
Latency and Throughput
| CPU | µarch | Latency | Throughput (IPC) |
|---|---|---|---|
| 3C5000 | LA464 | 11, 17.5 | 0.09(1/11.5) |
| 3A6000 | LA664 | 11 | 0.18(1/5.5) |
| 3C6000 | LA664 | 11, 17.5 | 0.22(1/4.5) |
| 2K1000LA | LA264 | 12, 25 | 0.06(1/18) |
| 2K3000 | LA364E | 12, 18.5 | 0.06(1/18) |