__lasx_xvmuh_w
LASX intrinsic · group: Integer Computation
Synopsis
__m256i __lasx_xvmuh_w (__m256i a, __m256i b)
#include <lasxintrin.h>
Instruction: xvmuh.w xr, xr, xr
CPU Flags: LASX
Description
Multiply corresponding signed 32-bit elements in a and b, save the upper half of the double-width product in dst.
Operation
for (int i = 0; i < 8; i++) {
dst.word[i] = (((s64)(s32)a.word[i] * (s64)(s32)b.word[i])) >> 32;
}
Tested on real machine.
Latency and Throughput
| CPU | µarch | Latency | Throughput (IPC) |
|---|---|---|---|
| 3C5000 | LA464 | 4 | 2 |
| 3A6000 | LA664 | 4 | 2 |
| 3C6000 | LA664 | 4 | 2 |