__lsx_vreplve_h
LSX intrinsic · group: Misc
Synopsis
__m128i __lsx_vreplve_h (__m128i a, int idx)
#include <lsxintrin.h>
Instruction: vreplve.h vr, vr, r
CPU Flags: LSX
Description
Repeat the element in lane idx (index modulo 8) of a to fill whole vector.
Operation
for (int i = 0; i < 8; i++) {
dst.half[i] = a.half[idx % 8];
}
Tested on real machine.
Latency and Throughput
| CPU | µarch | Latency | Throughput (IPC) |
|---|---|---|---|
| 3C5000 | LA464 | 1 | 1 |
| 3A6000 | LA664 | 1 | 1 |
| 3C6000 | LA664 | 1 | 1 |
| 2K1000LA | LA264 | 1 | 1 |
| 2K3000 | LA364E | 1 | 1 |