global_atomic_cmpswap_x2 GPU Native ISA AMD Vector

Vector/Global Memory

global_atomic_cmpswap_x2

Compare two unsigned 64-bit integer values stored in the data comparison register and a location in the global aperture.

Also written as global_atomic_cmpswap_b64. AMD's machine-readable ISA specification lists this name for the same instruction.

Encoding

Verified bit-level encoding data is not yet available for this instruction. The instruction-format classification below (GLOBAL) is well-documented and stable; exact per-target opcode/field bit positions have not yet been imported from a verified source.
Format GLOBAL

Operands

  • VDST
    Written. N/A Data format: N/A (OPR_VGPR_OR_ACCVGPR, FMT_NUM_U64)
  • ADDR
    Read. N/A Data format: Value can be anything. (OPR_VGPR, FMT_ANY)
  • DATA
    Read. N/A Data format: Value can be anything. (OPR_VGPR_OR_ACCVGPR, FMT_ANY)
  • SADDR
    Read. N/A Data format: Value can be anything. (OPR_SREG, FMT_ANY)

In AMD's order, as its machine-readable ISA specification lists them for the ENC_FLAT_GLBL encoding (AMD CDNA 4).

GFX Target Compatibility

Per-target GFX compatibility has not yet been verified for this instruction.

Related

More in Vector/Global Memory

Reference

AMDGPU / GFX ISA

Description

Compare two unsigned 64-bit integer values stored in the data comparison register and a location in the global aperture. Modify the memory location with a value in the data source register iff the comparison is equal. Store the original value from global aperture into a vector register iff the SC0 bit is set.

Semantics

addr = CalcGlobalAddr(ADDR.b32, SADDR.b32, OFFSET.b32); tmp = MEM[addr].u64; src = DATA[63 : 0].u64; cmp = DATA[127 : 64].u64; MEM[addr].u64 = tmp == cmp ? src : tmp; RETURN_DATA.u64 = tmp

Example

global_atomic_cmpswap_x2 v1, v[2:5], vcc

A real instruction accepted by the LLVM assembler, taken verbatim from LLVM's own AMDGPU MC test suite. Not from an AMD document, and not authored here.

Sources