This is a collection of Intel®’ IA32® Software Developer's Manuals (URL of the day) and AMD' AMD64 Architecture Programmer's Manual together with the related specifications, application notes, white papers, and change logs. The collection aims to keep all available revisions. It was originally created by Michal Necasek, see OS/2 Museum.

If you have a public document, related to the IA32® specifications and missing from the collection, please mail it to me. The content of this URL and all sub-ULRs is available for convenient bulk download by rsync x86docs password "" (empty).

VSELEQ, VSELGE, VSELGT, VSELVS -- AArch32

VSELEQ, VSELGE, VSELGT, VSELVS

Floating-point conditional select allows the destination register to take the value in either one or the other source register according to the condition codes in the APSR.

It has encodings from the following instruction sets: A32 ( A1 ) and T32 ( T1 ) .

A1

313029282726252423222120191817161514131211109876543210
111111100DccVnVd10sizeN0M0Vm

Encoding for the Equal, half-precision scalar variant
(FEAT_FP16)

Applies when (cc == 00 && size == 01)

VSELEQ.F16 <Sd>, <Sn>, <Sm> // (Cannot be conditional)

Encoding for the Equal, single-precision scalar variant

Applies when (cc == 00 && size == 10)

VSELEQ.F32 <Sd>, <Sn>, <Sm> // (Cannot be conditional)

Encoding for the Equal, double-precision scalar variant

Applies when (cc == 00 && size == 11)

VSELEQ.F64 <Dd>, <Dn>, <Dm> // (Cannot be conditional)

Encoding for the Greater than or Equal, half-precision scalar variant
(FEAT_FP16)

Applies when (cc == 10 && size == 01)

VSELGE.F16 <Sd>, <Sn>, <Sm> // (Cannot be conditional)

Encoding for the Greater than or Equal, single-precision scalar variant

Applies when (cc == 10 && size == 10)

VSELGE.F32 <Sd>, <Sn>, <Sm> // (Cannot be conditional)

Encoding for the Greater than or Equal, double-precision scalar variant

Applies when (cc == 10 && size == 11)

VSELGE.F64 <Dd>, <Dn>, <Dm> // (Cannot be conditional)

Encoding for the Greater than, half-precision scalar variant
(FEAT_FP16)

Applies when (cc == 11 && size == 01)

VSELGT.F16 <Sd>, <Sn>, <Sm> // (Cannot be conditional)

Encoding for the Greater than, single-precision scalar variant

Applies when (cc == 11 && size == 10)

VSELGT.F32 <Sd>, <Sn>, <Sm> // (Cannot be conditional)

Encoding for the Greater than, double-precision scalar variant

Applies when (cc == 11 && size == 11)

VSELGT.F64 <Dd>, <Dn>, <Dm> // (Cannot be conditional)

Encoding for the Unordered, half-precision scalar variant
(FEAT_FP16)

Applies when (cc == 01 && size == 01)

VSELVS.F16 <Sd>, <Sn>, <Sm> // (Cannot be conditional)

Encoding for the Unordered, single-precision scalar variant

Applies when (cc == 01 && size == 10)

VSELVS.F32 <Sd>, <Sn>, <Sm> // (Cannot be conditional)

Encoding for the Unordered, double-precision scalar variant

Applies when (cc == 01 && size == 11)

VSELVS.F64 <Dd>, <Dn>, <Dm> // (Cannot be conditional)

Decode for all variants of this encoding

if size == '00' || (size == '01' && !IsFeatureImplemented(FEAT_FP16)) then Undefined(); end; let esize : integer{} = 8 << UInt(size); let d : integer = if size == '11' then UInt(D::Vd) else UInt(Vd::D); let n : integer = if size == '11' then UInt(N::Vn) else UInt(Vn::N); let m : integer = if size == '11' then UInt(M::Vm) else UInt(Vm::M); let condition : bits(4) = cc::(cc[1] XOR cc[0])::'0';

T1

15141312111098765432101514131211109876543210
111111100DccVnVd10sizeN0M0Vm

Encoding for the Equal, half-precision scalar variant
(FEAT_FP16)

Applies when (cc == 00 && size == 01)

VSELEQ.F16 <Sd>, <Sn>, <Sm> // (Not permitted in IT block)

Encoding for the Equal, single-precision scalar variant

Applies when (cc == 00 && size == 10)

VSELEQ.F32 <Sd>, <Sn>, <Sm> // (Not permitted in IT block)

Encoding for the Equal, double-precision scalar variant

Applies when (cc == 00 && size == 11)

VSELEQ.F64 <Dd>, <Dn>, <Dm> // (Not permitted in IT block)

Encoding for the Greater than or Equal, half-precision scalar variant
(FEAT_FP16)

Applies when (cc == 10 && size == 01)

VSELGE.F16 <Sd>, <Sn>, <Sm> // (Not permitted in IT block)

Encoding for the Greater than or Equal, single-precision scalar variant

Applies when (cc == 10 && size == 10)

VSELGE.F32 <Sd>, <Sn>, <Sm> // (Not permitted in IT block)

Encoding for the Greater than or Equal, double-precision scalar variant

Applies when (cc == 10 && size == 11)

VSELGE.F64 <Dd>, <Dn>, <Dm> // (Not permitted in IT block)

Encoding for the Greater than, half-precision scalar variant
(FEAT_FP16)

Applies when (cc == 11 && size == 01)

VSELGT.F16 <Sd>, <Sn>, <Sm> // (Not permitted in IT block)

Encoding for the Greater than, single-precision scalar variant

Applies when (cc == 11 && size == 10)

VSELGT.F32 <Sd>, <Sn>, <Sm> // (Not permitted in IT block)

Encoding for the Greater than, double-precision scalar variant

Applies when (cc == 11 && size == 11)

VSELGT.F64 <Dd>, <Dn>, <Dm> // (Not permitted in IT block)

Encoding for the Unordered, half-precision scalar variant
(FEAT_FP16)

Applies when (cc == 01 && size == 01)

VSELVS.F16 <Sd>, <Sn>, <Sm> // (Not permitted in IT block)

Encoding for the Unordered, single-precision scalar variant

Applies when (cc == 01 && size == 10)

VSELVS.F32 <Sd>, <Sn>, <Sm> // (Not permitted in IT block)

Encoding for the Unordered, double-precision scalar variant

Applies when (cc == 01 && size == 11)

VSELVS.F64 <Dd>, <Dn>, <Dm> // (Not permitted in IT block)

Decode for all variants of this encoding

if InITBlock() then UnpredictableProcedure(); end; if size == '00' || (size == '01' && !IsFeatureImplemented(FEAT_FP16)) then Undefined(); end; let esize : integer{} = 8 << UInt(size); let d : integer = if size == '11' then UInt(D::Vd) else UInt(Vd::D); let n : integer = if size == '11' then UInt(N::Vn) else UInt(Vn::N); let m : integer = if size == '11' then UInt(M::Vm) else UInt(Vm::M); let condition : bits(4) = cc::(cc[1] XOR cc[0])::'0';

CONSTRAINED UNPREDICTABLE behavior

If InITBlock(), then one of the following behaviors must occur:

Assembler Symbols

<Sd>

Is the 32-bit name of the SIMD&FP destination register, encoded in the "Vd:D" field.

<Sn>

Is the 32-bit name of the first SIMD&FP source register, encoded in the "Vn:N" field.

<Sm>

Is the 32-bit name of the second SIMD&FP source register, encoded in the "Vm:M" field.

<Dd>

Is the 64-bit name of the SIMD&FP destination register, encoded in the "D:Vd" field.

<Dn>

Is the 64-bit name of the first SIMD&FP source register, encoded in the "N:Vn" field.

<Dm>

Is the 64-bit name of the second SIMD&FP source register, encoded in the "M:Vm" field.


2025-09_rel_asl1 2026-03-12 12:57:38

Copyright © 2010-2025 Arm Limited or its affiliates. All rights reserved. This document is Non-Confidential.