FCVTNT (unpredicated)

Single-precision convert, narrow and interleave to 8-bit floating-point (top)

Convert each single-precision element of the group of two source vectors to 8-bit floating-point while scaling the value by 2SInt(FPMR.NSCALE), and place the two-way interleaved results in the corresponding odd-numbered 8-bit elements of the destination vector, leaving the even-numbered elements unchanged. The 8-bit floating-point encoding format is selected by FPMR.F8D.

This instruction is unpredicated.

SVE2
(FEAT_FP8)

313029282726252423222120191817161514131211109876543210
0110010100001010001111Zn0Zd
T

FCVTNT <Zd>.B, { <Zn1>.S-<Zn2>.S }

if ((!IsFeatureImplemented(FEAT_SVE2) && !IsFeatureImplemented(FEAT_SME2)) || !IsFeatureImplemented(FEAT_FP8)) then UNDEFINED; constant integer n = UInt(Zn:'0'); constant integer d = UInt(Zd);

Assembler Symbols

<Zd>

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

<Zn1>

Is the name of the first scalable vector register of the source multi-vector group, encoded as "Zn" times 2.

<Zn2>

Is the name of the second scalable vector register of the source multi-vector group, encoded as "Zn" times 2 plus 1.

Operation

CheckFPMREnabled(); CheckSVEEnabled(); constant integer VL = CurrentVL; constant integer elements = VL DIV 32; bits(VL) result = Z[d, VL]; constant bits(VL) operand1 = Z[n+0, VL]; constant bits(VL) operand2 = Z[n+1, VL]; for e = 0 to elements-1 constant bits(32) element1 = Elem[operand1, e, 32]; constant bits(32) element2 = Elem[operand2, e, 32]; Elem[result, 4*e + 1, 8] = FPConvertFP8(element1, FPCR, FPMR, 8); Elem[result, 4*e + 3, 8] = FPConvertFP8(element2, FPCR, FPMR, 8); Z[d, VL] = result;


Internal version only: aarchmrs v2024-03_relA, pseudocode v2024-03_rel, sve v2024-03_rel ; Build timestamp: 2024-03-26T09:45

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