x86rotli.h
LBT intrinsic · group: x86 Shift
Synopsis
Instruction: x86rotli.h rj, imm
CPU Flags: LBT
Description
x86-style rotate left: rotate 16-bit value in rj left by immediate imm. Update CF, OF (for shift count of 1). Only EFLAGS (LBT4) is updated; the GPR is not modified.
Operation
uint16_t v = (uint16_t)a;
unsigned c = (unsigned)imm;
unsigned n = c % 16;
uint16_t r = (uint16_t)((v << n) | (v >> (16 - n)));
EFLAGS.CF = ((v >> ((16 - n) % 16)) & 1);
if (c == 1) {
EFLAGS.OF = ((v ^ r) & 0x8000) != 0;
}
Tested on real machine.
Latency and Throughput
| CPU | µarch | Latency | Throughput (IPC) |
|---|---|---|---|
| 3C5000 | LA464 | 2 | 0.5(1/2) |
| 3C6000 | LA664 | 2 | 0.5(1/2) |