__lasx_xvfcvth_s_h
LASX intrinsic · group: Floating Point Conversion
Synopsis
__m256 __lasx_xvfcvth_s_h (__m256i a)
#include <lasxintrin.h>
Instruction: xvfcvth.s.h xr, xr
CPU Flags: LASX
Description
Convert half precision floating point elements in higher half of every 128-bit vector of a to single precision.
Operation
int i;
for (i = 0; i < 4; i++) {
dst.fp32[i] = a.fp16[i + 4];
}
for (; i < 8; i++) {
dst.fp32[i] = a.fp16[i + 8];
}
// Expands to:
if (0) {
dst.fp32[0] = a.fp16[4];
dst.fp32[1] = a.fp16[5];
dst.fp32[2] = a.fp16[6];
dst.fp32[3] = a.fp16[7];
dst.fp32[4] = a.fp16[12];
dst.fp32[5] = a.fp16[13];
dst.fp32[6] = a.fp16[14];
dst.fp32[7] = a.fp16[15];
}
Latency and Throughput
| CPU | µarch | Latency | Throughput (IPC) |
|---|---|---|---|
| 3C5000 | LA464 | 3 | 1 |
| 3A6000 | LA664 | 3 | 2 |
| 3C6000 | LA664 | 3 | 2 |