VFNCVT.F.F.W

Vector Narrowing Float Convert

VFNCVT.F.F.W vd, vs2, vm

Converts 2*N-bit floats to N-bit floats (e.g., FP32 -> FP16).

Details

Converts vector elements between floating-point and integer types, or between floating-point precisions. Conversions respect the active rounding mode.

Pseudocode Operation

foreach(i < vl): vd[i] = narrow_float(vs2[i]);

Example

VFNCVT.F.F.W v1, v4, v0.t

Encoding

Binary Layout
010010
31:26
vm
25
vs2
24:20
10100
19:15
001
14:12
vd
11:7
1010111
6:0
 
Format OPFVV
Opcode 0x57
Extension V

Operands

  • vd
    Dest (Narrow Float)
  • vs2
    Src (Wide Float)