__lsx_vexth_w_h
LSX intrinsic · group: Misc
Synopsis
__m128i __lsx_vexth_w_h (__m128i a)
#include <lsxintrin.h>
Instruction: vexth.w.h vr, vr
CPU Flags: LSX
Description
Sign-extend 16-bit elements in the higher half of every 128-bit vector of a to 32-bit.
Operation
for (int i = 0; i < 4; i++) {
dst.word[i] = (s32)(s16)a.half[i + 4];
}
// Expands to:
if (0) {
dst.word[0] = (s32)((s16)a.half[4]);
dst.word[1] = (s32)((s16)a.half[5]);
dst.word[2] = (s32)((s16)a.half[6]);
dst.word[3] = (s32)((s16)a.half[7]);
}
Tested on real machine.
Latency and Throughput
| CPU | µarch | Latency | Throughput (IPC) |
|---|---|---|---|
| 3C5000 | LA464 | 1 | 2 |
| 3A6000 | LA664 | 1 | 4 |
| 3C6000 | LA664 | 1 | 4 |
| 2K1000LA | LA264 | 1 | 1 |
| 2K3000 | LA364E | 1 | 1 |