VLDR (immediate)

Load SIMD&FP register (immediate) loads a single register from the Advanced SIMD and floating-point register file, using an address from a general-purpose register, with an optional offset.

Depending on settings in the CPACR, NSACR, HCPTR, and FPEXC registers, and the Security state and PE mode in which the instruction is executed, an attempt to execute the instruction might be undefined, or trapped to Hyp mode. For more information, see Enabling Advanced SIMD and floating-point support.

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

A1

313029282726252423222120191817161514131211109876543210
!= 11111101UD01!= 1111Vd10sizeimm8
condRn

Encoding for the Half-precision scalar variant
(FEAT_FP16)

Applies when (size == 01)

VLDR{<c>}{<q>}.16 <Sd>, [<Rn> {, #{+/-}<imm>}]

Encoding for the Single-precision scalar variant

Applies when (size == 10)

VLDR{<c>}{<q>}{.32} <Sd>, [<Rn> {, #{+/-}<imm>}]

Encoding for the Double-precision scalar variant

Applies when (size == 11)

VLDR{<c>}{<q>}{.64} <Dd>, [<Rn> {, #{+/-}<imm>}]

Decode for all variants of this encoding

if size == '00' || (size == '01' && !IsFeatureImplemented(FEAT_FP16)) then UNDEFINED; if size == '01' && cond != '1110' then UNPREDICTABLE; constant esize = 8 << UInt(size); constant add = (U == '1'); constant imm32 = if esize == 16 then ZeroExtend(imm8:'0', 32) else ZeroExtend(imm8:'00', 32); constant d = if size == '11' then UInt(D:Vd) else UInt(Vd:D); constant n = UInt(Rn);

CONSTRAINED UNPREDICTABLE behavior

If size == '01' && cond != '1110', then one of the following behaviors must occur:

T1

15141312111098765432101514131211109876543210
11101101UD01!= 1111Vd10sizeimm8
Rn

Encoding for the Half-precision scalar variant
(FEAT_FP16)

Applies when (size == 01)

VLDR{<c>}{<q>}.16 <Sd>, [<Rn> {, #{+/-}<imm>}]

Encoding for the Single-precision scalar variant

Applies when (size == 10)

VLDR{<c>}{<q>}{.32} <Sd>, [<Rn> {, #{+/-}<imm>}]

Encoding for the Double-precision scalar variant

Applies when (size == 11)

VLDR{<c>}{<q>}{.64} <Dd>, [<Rn> {, #{+/-}<imm>}]

Decode for all variants of this encoding

if size == '00' || (size == '01' && !IsFeatureImplemented(FEAT_FP16)) then UNDEFINED; if size == '01' && InITBlock() then UNPREDICTABLE; constant esize = 8 << UInt(size); constant add = (U == '1'); constant imm32 = if esize == 16 then ZeroExtend(imm8:'0', 32) else ZeroExtend(imm8:'00', 32); constant d = if size == '11' then UInt(D:Vd) else UInt(Vd:D); constant n = UInt(Rn);

CONSTRAINED UNPREDICTABLE behavior

If size == '01' && InITBlock(), then one of the following behaviors must occur:

Assembler Symbols

<c>

See Standard assembler syntax fields.

<q>

See Standard assembler syntax fields.

.64

Is an optional data size specifier for 64-bit memory accesses that can be used in the assembler source code, but is otherwise ignored.

<Dd>

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

.32

Is an optional data size specifier for 32-bit memory accesses that can be used in the assembler source code, but is otherwise ignored.

<Sd>

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

<Rn>

Is the general-purpose base register, encoded in the "Rn" field.

+/-

Specifies the offset is added to or subtracted from the base register, defaulting to + if omitted and encoded in U:

U +/-
0 -
1 +
<imm>

For the single-precision scalar or double-precision scalar variants: is the optional unsigned immediate byte offset, a multiple of 4, in the range 0 to 1020, defaulting to 0, and encoded in the "imm8" field as <imm>/4.

For the half-precision scalar variant: is the optional unsigned immediate byte offset, a multiple of 2, in the range 0 to 510, defaulting to 0, and encoded in the "imm8" field as <imm>/2.

Operation

if ConditionPassed() then EncodingSpecificOperations(); CheckVFPEnabled(TRUE); constant base = if n == 15 then Align(PC32,4) else R[n]; constant address = if add then (base + imm32) else (base - imm32); case esize of when 16 H[d] = MemA[address,2]; when 32 S[d] = MemA[address,4]; when 64 constant word1 = MemA[address,4]; constant word2 = MemA[address+4,4]; // Combine the word-aligned words in the correct order for current endianness. D[d] = if BigEndian(AccessType_ASIMD) then word1:word2 else word2:word1;

Operational information

If CPSR.DIT is 1, the timing of this instruction is insensitive to the value of the data being loaded or stored.


Internal version only: isa v01_32, pseudocode v2024-12_rel ; Build timestamp: 2024-12-16T10:54

Copyright © 2010-2024 Arm Limited or its affiliates. All rights reserved. This document is Non-Confidential.