__lasx_xvexth_d_w
LASX intrinsic · group: Misc
Synopsis
__m256i __lasx_xvexth_d_w (__m256i a)
#include <lasxintrin.h>
Instruction: xvexth.d.w xr, xr
CPU Flags: LASX
Description
Sign-extend 32-bit elements in the higher half of every 128-bit vector of a to 64-bit.
Operation
int i;
for (i = 0; i < 2; i++) {
dst.dword[i] = (s64)(s32)a.word[i + 2];
}
for (; i < 4; i++) {
dst.dword[i] = (s64)(s32)a.word[i + 4];
}
// Expands to:
if (0) {
dst.dword[0] = (s64)((s32)a.word[2]);
dst.dword[1] = (s64)((s32)a.word[3]);
dst.dword[2] = (s64)((s32)a.word[6]);
dst.dword[3] = (s64)((s32)a.word[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 |