v_frexp_exp_i32_f32 GPU Native ISA AMD Vector

Vector Arithmetic

v_frexp_exp_i32_f32

Extract the exponent of a single-precision float input and store the result as a signed 32-bit integer into a vector register.

Encoding

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

The ENC_VOP1 layout from AMD's machine-readable ISA specification (AMD CDNA 5). Opcode 63 in OP rebuilds 0x7E007E00, 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 51, field layout), AMD CDNA 3 (opcode 51, field layout), AMD CDNA 2 (opcode 51, field layout), AMD CDNA 1 (opcode 51, 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 63
Identifier 0x7E007E00

Operands

  • VDST
    Written. Operand must be a vector GPR. Uses an 8-bit operand field. Data format: Signed 32-bit integral data. (OPR_VGPR, FMT_NUM_I32)
  • SRC0
    Read. All operands. Covers all operands that are allowed as scalar or vector sources. Data format: 32-bit single-precision floating point value. (OPR_SRC, FMT_NUM_F32)

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

GFX Target Compatibility

TargetSupport
gfx1100✅ Supported

This operation satisfies the invariant S0.f32 = significand * (2 ** exponent). See also V_FREXP_MANT_F32, which returns the significand. See the C library function frexp() for more information.

Related

More in Vector Arithmetic

Reference

AMDGPU / GFX ISA

Semantics

if ((64'F(S0.f32) == +INF) || (64'F(S0.f32) == -INF) || isNAN(64'F(S0.f32))) then D0.i32 = 0 else D0.i32 = exponent(S0.f32) - 127 + 1 endif

Example

v_frexp_exp_i32_f32 v5, v1

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