x86srai.h

LBT intrinsic · group: x86 Shift

Synopsis

Instruction: x86srai.h rj, imm
CPU Flags: LBT

Description

x86-style shift right arithmetic: shift 16-bit value in rj right arithmetically by immediate imm. If shift count is non-zero, update CF, PF, ZF, SF (OF is set to 0 if count is 1). If shift count is zero, flags are unchanged. Only EFLAGS (LBT4) is updated; the GPR is not modified.

Operation

uint16_t v = (uint16_t)a;
unsigned c = (unsigned)imm;
if (c != 0) {
  uint8_t carry_out = c > 16 ? ((int16_t)v < 0) : ((v >> (c - 1)) & 1);
  uint16_t r = c >= 16 ? (uint16_t)((int16_t)v >> (16 - 1))
                       : (uint16_t)((int16_t)v >> c);
  EFLAGS.CF = carry_out;
  EFLAGS.PF = parity_even((uint8_t)r);
  EFLAGS.ZF = r == 0;
  EFLAGS.SF = (int16_t)r < 0;
  if (c == 1) {
    EFLAGS.OF = 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)

All intrinsics in alphabetical order