__lsx_vshuf_w

LSX intrinsic · group: Shuffling

Synopsis

__m128i __lsx_vshuf_w (__m128i a, __m128i b, __m128i c)
#include <lsxintrin.h>
Instruction: vshuf.w vr, vr, vr
CPU Flags: LSX

Description

For each 128-bit lane, shuffle 32-bit elements from b and c using the low 3 bits of the indices from the corresponding lane of a, and save the result to dst.

Examples

__m128i __lsx_vshuf_w(__m128i{0x0000000200000004, 0x0000000700000005}, __m128i{0x1122334455667788, 0x99aabbccddeeff00}, __m128i{0xabcdef1314156678, 0x1234123443214321})
= 0x4321432155667788 0x99aabbcc11223344

Operation

for (int i = 0; i < 4; i++) {
  if ((a.word[i] % 256) >= 64 && (UARCH_LA264 || UARCH_LA464)) {
    // Caveat: observed in LA264 and LA464
    dst.word[i] = 0;
  } else if ((a.word[i] % 8) < 4) {
    dst.word[i] = c.word[a.word[i] % 4];
  } else {
    dst.word[i] = b.word[a.word[i] % 4];
  }
}

Tested on real machine.

Latency and Throughput

CPU	µarch	Latency	Throughput (IPC)
3C5000	LA464	1	2
3A6000	LA664	1	2
3C6000	LA664	1	2
2K1000LA	LA264	1	1
2K3000	LA364E	1	1

---

All intrinsics in alphabetical order