BF1CVT, BF2CVT

8-bit floating-point convert to BFloat16

This instruction converts each even-numbered 8-bit floating-point element of the source vector to BFloat16 while downscaling the value, and places the results in the overlapping 16-bit elements of the destination vector. BF1CVT scales the values by 2-UInt(FPMR.LSCALE[5:0]). BF2CVT scales the values by 2-UInt(FPMR.LSCALE2[5:0]).

The 8-bit floating-point encoding format for BF1CVT is selected by FPMR.F8S1. The 8-bit floating-point encoding format for BF2CVT is selected by FPMR.F8S2.

This instruction is unpredicated.

It has encodings from 2 classes: BF1CVT and BF2CVT

BF1CVT
((FEAT_SVE2 || FEAT_SME2) && FEAT_FP8)

313029282726252423222120191817161514131211109876543210
0110010100001000001110ZnZd
Lopc

Encoding

BF1CVT <Zd>.H, <Zn>.B

Decode for this encoding

if ((!IsFeatureImplemented(FEAT_SVE2) && !IsFeatureImplemented(FEAT_SME2)) || !IsFeatureImplemented(FEAT_FP8)) then EndOfDecode(Decode_UNDEF); end; let esize : integer{} = 16; let n : integer = UInt(Zn); let d : integer = UInt(Zd); let s_esize : integer{} = 8; let d_esize : integer{} = 16; let issrc2 : boolean = FALSE;

BF2CVT
((FEAT_SVE2 || FEAT_SME2) && FEAT_FP8)

313029282726252423222120191817161514131211109876543210
0110010100001000001111ZnZd
Lopc

Encoding

BF2CVT <Zd>.H, <Zn>.B

Decode for this encoding

if ((!IsFeatureImplemented(FEAT_SVE2) && !IsFeatureImplemented(FEAT_SME2)) || !IsFeatureImplemented(FEAT_FP8)) then EndOfDecode(Decode_UNDEF); end; let esize : integer{} = 16; let n : integer = UInt(Zn); let d : integer = UInt(Zd); let s_esize : integer{} = 8; let d_esize : integer{} = 16; let issrc2 : boolean = TRUE;

Assembler Symbols

<Zd>

Is the name of the destination scalable vector register, encoded in the "Zd" field.

<Zn>

Is the name of the source scalable vector register, encoded in the "Zn" field.


2025-09_diff_tag 2025-11-28 16:29:05

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