flat_store_short_d16_hi GPU Native ISA AMD Vector

Flat Memory

flat_store_short_d16_hi

Store 16 bits of data from the high 16 bits of a 32-bit vector register into the flat aperture.

Also written as flat_store_d16_hi_b16. 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 (FLAT) is well-documented and stable; exact per-target opcode/field bit positions have not yet been imported from a verified source.
Format FLAT

Operands

  • 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)

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

GFX Target Compatibility

TargetSupport
gfx1100✅ Supported

Related

More in Flat Memory

Reference

AMDGPU / GFX ISA

Semantics

addr = CalcFlatAddr(ADDR.b32, OFFSET.b32); MEM[addr].b16 = VDATA[31 : 16]

Example

flat_store_short_d16_hi v[1:2], v2

A real instruction accepted by the LLVM assembler, taken verbatim from LLVM's own AMDGPU MC test suite (gfx9_asm_flat.s). Not from an AMD document, and not authored here.

Sources