__lasx_xvsrani_b_h
LASX intrinsic · group: Shift
Synopsis
__m256i __lasx_xvsrani_b_h (__m256i a, __m256i b, imm0_15 imm)
#include <lasxintrin.h>
Instruction: xvsrani.b.h xr, xr, imm
CPU Flags: LASX
Description
Arithmetic right shift the signed 16-bit elements in a and b by imm, write the lower 8 bits of the shifted result into the lower half of each 128-bit lane of dst, and zero the upper half.
Operation
for (int i = 0; i < 16; i++) {
dst.byte[i] =
(i < 8) ? (s8)((s16)b.half[i] >> imm) : (s8)((s16)a.half[i - 8] >> imm);
}
for (int i = 16; i < 32; i++) {
dst.byte[i] = (i < 24) ? (s8)((s16)b.half[i - 8] >> imm)
: (s8)((s16)a.half[i - 16] >> imm);
}
Tested on real machine.
Latency and Throughput
| CPU | µarch | Latency | Throughput (IPC) |
|---|---|---|---|
| 3C5000 | LA464 | 4 | 1 |
| 3A6000 | LA664 | 4 | 2 |
| 3C6000 | LA664 | 4 | 2 |