Replace RL16S with RB16S, the values are big-endian.
This is related to the recently-fixed bug #1463.
/* Byte 7+8/9+10/11+12/13+14: channel T1/T2/T3/T4 temperature. */
for (i = 0; i < NUM_CHANNELS; i++) {
/* Byte 7+8/9+10/11+12/13+14: channel T1/T2/T3/T4 temperature. */
for (i = 0; i < NUM_CHANNELS; i++) {
- temp_i16 = RL16S(&buf[7 + 2 * i]);
+ temp_i16 = RB16S(&buf[7 + 2 * i]);
info->temp[i] = (float)temp_i16;
}
info->temp[i] = (float)temp_i16;
}