__lasx_xvsrlni_h_w
LASX intrinsic · group: Shift
Synopsis
__m256i __lasx_xvsrlni_h_w (__m256i a, __m256i b, imm0_31 imm)
#include <lasxintrin.h>
Instruction: xvsrlni.h.w xr, xr, imm
CPU Flags: LASX
Description
Logical right shift the unsigned 32-bit elements in a and b by imm, write the lower 16 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 < 8; i++) {
dst.half[i] =
(i < 4) ? (u16)((u32)b.word[i] >> imm) : (u16)((u32)a.word[i - 4] >> imm);
}
for (int i = 8; i < 16; i++) {
dst.half[i] = (i < 12) ? (u16)((u32)b.word[i - 4] >> imm)
: (u16)((u32)a.word[i - 8] >> 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 |