FMOV (vector, immediate)

Floating-point move immediate (vector). This instruction copies an immediate floating-point constant into every element of the SIMD&FP destination register.

Depending on the settings in the CPACR_EL1, CPTR_EL2, and CPTR_EL3 registers, and the current Security state and Exception level, an attempt to execute the instruction might be trapped.

It has encodings from 2 classes: Half-precision and Single-precision and double-precision

Half-precision
(FEAT_FP16)

313029282726252423222120191817161514131211109876543210
0Q00111100000abc111111defghRd

FMOV <Vd>.<T>, #<imm>

if !HaveFP16Ext() then UNDEFINED; integer rd = UInt(Rd); integer datasize = if Q == '1' then 128 else 64; bits(datasize) imm; imm8 = a:b:c:d:e:f:g:h; imm16 = imm8<7>:NOT(imm8<6>):Replicate(imm8<6>,2):imm8<5:0>:Zeros(6); imm = Replicate(imm16, datasize DIV 16);

Single-precision and double-precision

313029282726252423222120191817161514131211109876543210
0Qop0111100000abc111101defghRd
cmode

Single-precision (op == 0)

FMOV <Vd>.<T>, #<imm>

Double-precision (Q == 1 && op == 1)

FMOV <Vd>.2D, #<imm>

integer rd = UInt(Rd); integer datasize = if Q == '1' then 128 else 64; bits(datasize) imm; bits(64) imm64; ImmediateOp operation; case cmode:op of when '0xx00' operation = ImmediateOp_MOVI; when '0xx01' operation = ImmediateOp_MVNI; when '0xx10' operation = ImmediateOp_ORR; when '0xx11' operation = ImmediateOp_BIC; when '10x00' operation = ImmediateOp_MOVI; when '10x01' operation = ImmediateOp_MVNI; when '10x10' operation = ImmediateOp_ORR; when '10x11' operation = ImmediateOp_BIC; when '110x0' operation = ImmediateOp_MOVI; when '110x1' operation = ImmediateOp_MVNI; when '1110x' operation = ImmediateOp_MOVI; when '11110' operation = ImmediateOp_MOVI; when '11111' // FMOV Dn,#imm is in main FP instruction set if Q == '0' then UNDEFINED; operation = ImmediateOp_MOVI; imm64 = AdvSIMDExpandImm(op, cmode, a:b:c:d:e:f:g:h); imm = Replicate(imm64, datasize DIV 64);

Assembler Symbols

<Vd>

Is the name of the SIMD&FP destination register, encoded in the "Rd" field.

<T>

For the half-precision variant: is an arrangement specifier, encoded in Q:

Q <T>
0 4H
1 8H

For the single-precision variant: is an arrangement specifier, encoded in Q:

Q <T>
0 2S
1 4S
<imm>

Is a signed floating-point constant with 3-bit exponent and normalized 4 bits of precision, encoded in "a:b:c:d:e:f:g:h". For details of the range of constants available and the encoding of <imm>, see Modified immediate constants in A64 floating-point instructions.


Internal version only: isa v33.11seprel, AdvSIMD v29.05, pseudocode v2021-09_rel, sve v2021-09_rc3d ; Build timestamp: 2021-10-06T11:41

Copyright © 2010-2021 Arm Limited or its affiliates. All rights reserved. This document is Non-Confidential.