__lasx_xvhsubw_du_wu

LASX intrinsic · group: Integer Computation

Synopsis

__m256i __lasx_xvhsubw_du_wu (__m256i a, __m256i b)
#include <lasxintrin.h>
Instruction: xvhsubw.du.wu xr, xr, xr
CPU Flags: LASX

Description

Subtract even-positioned unsigned 32-bit elements in b from odd-positioned unsigned 32-bit elements in a, zero-extending both to 64-bit, to get the 64-bit result.

Operation

for (int i = 0; i < 4; i++) {
  dst.dword[i] = (u64)(u32)a.word[2 * i + 1] - (u64)(u32)b.word[2 * i];
}

// Expands to:

if (0) {
  dst.dword[0] = ((u64)((u32)a.word[1])) - ((u64)((u32)b.word[0]));
  dst.dword[1] = ((u64)((u32)a.word[3])) - ((u64)((u32)b.word[2]));
  dst.dword[2] = ((u64)((u32)a.word[5])) - ((u64)((u32)b.word[4]));
  dst.dword[3] = ((u64)((u32)a.word[7])) - ((u64)((u32)b.word[6]));
}

Tested on real machine.

Latency and Throughput

CPU	µarch	Latency	Throughput (IPC)
3C5000	LA464	2	2
3A6000	LA664	2	2
3C6000	LA664	2	2

---

All intrinsics in alphabetical order