armsub.w

LBT intrinsic · group: ARM Integer

Synopsis

Instruction: armsub.w rj, rk, cond
CPU Flags: LBT

Description

ARM-style subtract (32-bit): conditionally subtract value in rk from rj. Update ARMFLAGS (C, V, N, Z). Only ARMFLAGS (LBT4) is updated; the GPR is not modified.

Operation

if (arm_cond_holds(ARMFLAGS, cond)) {
  uint32_t lhs = (uint32_t)a;
  uint32_t rhs = (uint32_t)b;
  uint64_t subtrahend = (uint64_t)rhs;
  uint32_t r = (uint32_t)(lhs - subtrahend);
  ARMFLAGS.C = lhs < subtrahend;
  ARMFLAGS.V = ((lhs ^ rhs) & (lhs ^ r) & 0x80000000) != 0;
  ARMFLAGS.N = (r >> 31) & 1;
  ARMFLAGS.Z = r == 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