__lsx_vshuf4i_h
LSX intrinsic · group: Shuffling
Synopsis
__m128i __lsx_vshuf4i_h (__m128i a, imm0_255 imm)
#include <lsxintrin.h>
Instruction: vshuf4i.h vr, vr, imm
CPU Flags: LSX
Description
For each group of four consecutive 16-bit elements in a, permute the group according to the four 2-bit fields of imm, save the result to dst.
Examples
__m128i __lsx_vshuf4i_h(__m128i{0xabcdef1314156678, 0x1234123443214321}, 0x12)
= 0x667814156678ef13 0x4321432143211234
Operation
for (int i = 0; i < 8; i++) {
dst.half[i] = a.half[(i & ~0x3) + ((imm >> (2 * (i & 0x3))) & 0x3)];
}
// Expands to:
if (0) {
dst.half[0] = a.half[0 + ((imm >> 0) & 0x3)];
dst.half[1] = a.half[0 + ((imm >> 2) & 0x3)];
dst.half[2] = a.half[0 + ((imm >> 4) & 0x3)];
dst.half[3] = a.half[0 + ((imm >> 6) & 0x3)];
dst.half[4] = a.half[4 + ((imm >> 0) & 0x3)];
dst.half[5] = a.half[4 + ((imm >> 2) & 0x3)];
dst.half[6] = a.half[4 + ((imm >> 4) & 0x3)];
dst.half[7] = a.half[4 + ((imm >> 6) & 0x3)];
}
Tested on real machine.
Latency and Throughput
| CPU | µarch | Latency | Throughput (IPC) |
|---|---|---|---|
| 3C5000 | LA464 | 1 | 2 |
| 3A6000 | LA664 | 1 | 4 |
| 3C6000 | LA664 | 1 | 4 |
| 2K1000LA | LA264 | 1 | 1 |
| 2K3000 | LA364E | 1 | 1 |