__lasx_xvaddwod_h_bu

LASX intrinsic · group: Integer Computation

Synopsis

__m256i __lasx_xvaddwod_h_bu (__m256i a, __m256i b)
#include <lasxintrin.h>
Instruction: xvaddwod.h.bu xr, xr, xr
CPU Flags: LASX

Description

Add the odd-positioned unsigned 8-bit elements in a and the corresponding unsigned 8-bit elements in b, extending both to 16-bit, and save the 16-bit result in dst.

Operation

for (int i = 0; i < 16; i++) {
  dst.half[i] = (u16)(u8)a.byte[2 * i + 1] + (u16)(u8)b.byte[2 * i + 1];
}

// Expands to:

if (0) {
  dst.half[0] = ((u16)((u8)a.byte[1])) + ((u16)((u8)b.byte[1]));
  dst.half[1] = ((u16)((u8)a.byte[3])) + ((u16)((u8)b.byte[3]));
  dst.half[2] = ((u16)((u8)a.byte[5])) + ((u16)((u8)b.byte[5]));
  dst.half[3] = ((u16)((u8)a.byte[7])) + ((u16)((u8)b.byte[7]));
  dst.half[4] = ((u16)((u8)a.byte[9])) + ((u16)((u8)b.byte[9]));
  dst.half[5] = ((u16)((u8)a.byte[11])) + ((u16)((u8)b.byte[11]));
  dst.half[6] = ((u16)((u8)a.byte[13])) + ((u16)((u8)b.byte[13]));
  dst.half[7] = ((u16)((u8)a.byte[15])) + ((u16)((u8)b.byte[15]));
  dst.half[8] = ((u16)((u8)a.byte[17])) + ((u16)((u8)b.byte[17]));
  dst.half[9] = ((u16)((u8)a.byte[19])) + ((u16)((u8)b.byte[19]));
  dst.half[10] = ((u16)((u8)a.byte[21])) + ((u16)((u8)b.byte[21]));
  dst.half[11] = ((u16)((u8)a.byte[23])) + ((u16)((u8)b.byte[23]));
  dst.half[12] = ((u16)((u8)a.byte[25])) + ((u16)((u8)b.byte[25]));
  dst.half[13] = ((u16)((u8)a.byte[27])) + ((u16)((u8)b.byte[27]));
  dst.half[14] = ((u16)((u8)a.byte[29])) + ((u16)((u8)b.byte[29]));
  dst.half[15] = ((u16)((u8)a.byte[31])) + ((u16)((u8)b.byte[31]));
}

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