__lasx_xvdiv_h
LASX intrinsic · group: Integer Computation
Synopsis
__m256i __lasx_xvdiv_h (__m256i a, __m256i b)
#include <lasxintrin.h>
Instruction: xvdiv.h xr, xr, xr
CPU Flags: LASX
Description
Divide each signed 16-bit element in a by the corresponding element in b.
Operation
for (int i = 0; i < 16; i++) {
dst.half[i] = (b.half[i] == 0) ? 0 : ((s16)a.half[i] / (s16)b.half[i]);
}
Tested on real machine.
Latency and Throughput
| CPU | µarch | Latency | Throughput (IPC) |
|---|---|---|---|
| 3C5000 | LA464 | 21.5, 22 | 0.08(1/13) |
| 3A6000 | LA664 | 17 | 0.12(1/8.5) |
| 3C6000 | LA664 | 34, 40 | 0.05(1/19) |