__lasx_xvreplve_w
LASX intrinsic · group: Misc
Synopsis
__m256i __lasx_xvreplve_w (__m256i a, int idx)
#include <lasxintrin.h>
Instruction: xvreplve.w xr, xr, r
CPU Flags: LASX
Description
For each 128-bit lane, repeat that lane's element idx (index modulo 4) to fill the same lane.
Operation
for (int i = 0; i < 4; i++) {
dst.word[i] = a.word[idx % 4];
}
for (int i = 4; i < 8; i++) {
dst.word[i] = a.word[(idx % 4) + 4];
}
Tested on real machine.
Latency and Throughput
| CPU | µarch | Latency | Throughput (IPC) |
|---|---|---|---|
| 3C5000 | LA464 | 1 | 1 |
| 3A6000 | LA664 | 1 | 1 |
| 3C6000 | LA664 | 1 | 1 |