x86rcr.w
LBT intrinsic · group: x86 Shift
Synopsis
Instruction: x86rcr.w rj, rk
CPU Flags: LBT
Description
x86-style rotate right through carry: rotate 32-bit value in rj and CF together as a 33-bit ring right by the amount specified in rk. Update CF, OF (for shift count of 1). Only EFLAGS (LBT4) is updated; the GPR is not modified.
Operation
uint32_t v = (uint32_t)a;
unsigned c = (unsigned)(b & 0x1f);
unsigned n = c;
if (n != 0) {
unsigned carry_in = EFLAGS.CF;
unsigned carry_out = ((v >> ((n - 1) % (32 + 1))) & 1);
EFLAGS.CF = carry_out;
if (c == 1) {
uint32_t r = (uint32_t)((v >> 1) | (carry_in << 31));
EFLAGS.OF = (((r >> 31) ^ (r >> 30)) & 1) != 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) |