v_rsq_f16 GPU Native ISA AMD Vector

Vector Arithmetic

v_rsq_f16

Calculate the reciprocal of the square root of the half-precision float input using IEEE rules and store the result into a vector register.

Encoding

Binary Layout (AMD CDNA 5)
0111111
31:25
VDST
24:17
unassigned
16
1010110
15:9
SRC0
8:0
 

The ENC_VOP1 layout from AMD's machine-readable ISA specification (AMD CDNA 5). Opcode 86 in OP rebuilds 0x7E00AC00, an identifier AMD lists for this encoding. Bits marked unassigned have no field in the specification. Not the same in every generation: AMD CDNA 4 (opcode 63, field layout), AMD CDNA 3 (opcode 63, field layout), AMD CDNA 2 (opcode 63, field layout), AMD CDNA 1 (opcode 63, field layout), AMD RDNA 3.5 (field layout), AMD RDNA 3 (field layout), AMD RDNA 2 (field layout), AMD RDNA 1 (field layout). The same in AMD RDNA 4.

Format VOP1
Width 32 bits
Opcode 86
Identifier 0x7E00AC00

Operands

  • VDST
    Written. Operand must be a vector GPR. Uses an 8-bit operand field. Data format: 16-bit half-precision floating point value. (OPR_VGPR, FMT_NUM_F16)
  • SRC0
    Read. All operands. Covers all operands that are allowed as scalar or vector sources. Data format: 16-bit half-precision floating point value. (OPR_SRC, FMT_NUM_F16)

In AMD's order, as its machine-readable ISA specification lists them for the ENC_VOP1 encoding (AMD CDNA 5).

GFX Target Compatibility

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

1ULP accuracy, denormals are supported.

Related

More in Vector Arithmetic

Reference

AMDGPU / GFX ISA

Semantics

D0.f16 = 16'1.0 / sqrt(S0.f16)

Example

V_RSQ_F16(0xfc00) => 0xfe00     // rsq(-INF) = NAN
V_RSQ_F16(0x8000) => 0xfc00     // rsq(-0.0) = -INF
V_RSQ_F16(0x0000) => 0x7c00     // rsq(+0.0) = +INF
V_RSQ_F16(0x4400) => 0x3800     // rsq(+4.0) = +0.5

Reproduced from AMD's official "AMD Instinct MI300" ISA Reference Guide for technical accuracy.

Sources