__lasx_xvavg_wu
LASX intrinsic · group: Integer Computation
Synopsis
__m256i __lasx_xvavg_wu (__m256i a, __m256i b)
#include <lasxintrin.h>
Instruction: xvavg.wu xr, xr, xr
CPU Flags: LASX
Description
Compute the average (rounded towards negative infinity) of corresponding unsigned 32-bit elements in a and b, save the result in dst.
Operation
for (int i = 0; i < 8; i++) {
dst.word[i] = ((u32)a.word[i] >> 1) + ((u32)b.word[i] >> 1) +
((a.word[i] & b.word[i]) & 1);
}
Tested on real machine.
Latency and Throughput
| CPU | µarch | Latency | Throughput (IPC) |
|---|---|---|---|
| 3C5000 | LA464 | 1 | 2 |
| 3A6000 | LA664 | 1 | 4 |
| 3C6000 | LA664 | 1 | 4 |