xvrdpi

VSX Vector Round to Double-Precision Integer using round to Nearest Away

xvrdpi XT,XB

Rounds each element of a double-precision floating-point vector to the nearest integer away from zero.

Encoding

Binary Layout

6:10

11:15

201

16:20

21:25

26:31

Format XX2-form

Opcode 0xF0000324

Extension VSX

Operands

XT
Target Vector Register
XB
Source Vector Register

Example

xvrdpi vs1, vs3

Registers Altered

FPSCR, VXSNAN

Reference

View in PowerISA v3.1C Specification ↗ OPF Power ISA v3.1C

Description

For xvrdpi, each element in the source vector VSR[XB] is rounded to an integer using the rounding mode Round to Nearest Away. The result is placed into the target vector VSR[XT]. If a Signalling NaN is encountered, it is converted to a Quiet NaN and VXSNAN is set to 1.

Operation

if MSR.VSX=0 then VSX_Unavailable()

ex_flag ←0b0
do i = 0 to 1
    reset_xflags()
    src ←bfp_CONVERT_FROM_BFP64(VSR[32×BX+B].dword[i])
    rnd ←bfp_ROUND_TO_INTEGER(0b100, src)
    vresult.dword[i] ←bfp64_CONVERT_FROM_BFP(rnd)
    if vxsnan_flag=1 then SetFX(FPSCR.VXSNAN)
    ex_flag ←ex_flag | (FPSCR.VE & vxsnan_flag)
end
if ex_flag=0 then VSR[32×TX+T] ←vresult

Programming Note

This instruction is commonly used for converting floating-point numbers to integers with rounding towards the nearest integer away from zero. Ensure that the VSX (Vector Scalar Extensions) are enabled in the MSR register; otherwise, an exception will be raised. Be cautious of Signalling NaNs, as they are converted to Quiet NaNs and VXSNAN is set, which might affect subsequent operations if not handled properly.