FMINNMP (vector)

Floating-point Minimum Number Pairwise (vector). This instruction creates a vector by concatenating the vector elements of the first source SIMD&FP register after the vector elements of the second source SIMD&FP register, reads each pair of adjacent vector elements in the two source SIMD&FP registers, writes the smallest of each pair of floating-point values into a vector, and writes the vector to the destination SIMD&FP register. All the values in this instruction are floating-point values.

Regardless of the value of FPCR.AH, the behavior is as follows for each pairwise operation:

This instruction can generate a floating-point exception. Depending on the settings in FPCR, the exception results in either a flag being set in FPSR or a synchronous exception being generated. For more information, see Floating-point exception traps.

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
0Q101110110Rm000001RnRd
Uaopcode

FMINNMP <Vd>.<T>, <Vn>.<T>, <Vm>.<T>

if !IsFeatureImplemented(FEAT_FP16) then UNDEFINED; integer d = UInt(Rd); integer n = UInt(Rn); integer m = UInt(Rm); constant integer esize = 16; constant integer datasize = 64 << UInt(Q); integer elements = datasize DIV esize; boolean pair = (U == '1'); boolean minimum = (a == '1');

Single-precision and double-precision

313029282726252423222120191817161514131211109876543210
0Q1011101sz1Rm110001RnRd
Uo1opcode

FMINNMP <Vd>.<T>, <Vn>.<T>, <Vm>.<T>

integer d = UInt(Rd); integer n = UInt(Rn); integer m = UInt(Rm); if sz:Q == '10' then UNDEFINED; constant integer esize = 32 << UInt(sz); constant integer datasize = 64 << UInt(Q); integer elements = datasize DIV esize; boolean pair = (U == '1'); boolean minimum = (o1 == '1');

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 and double-precision variant: is an arrangement specifier, encoded in sz:Q:

sz Q <T>
0 0 2S
0 1 4S
1 0 RESERVED
1 1 2D
<Vn>

Is the name of the first SIMD&FP source register, encoded in the "Rn" field.

<Vm>

Is the name of the second SIMD&FP source register, encoded in the "Rm" field.

Operation

CheckFPAdvSIMDEnabled64(); bits(datasize) operand1 = V[n, datasize]; bits(datasize) operand2 = V[m, datasize]; bits(datasize) result; bits(2*datasize) concat = operand2:operand1; bits(esize) element1; bits(esize) element2; for e = 0 to elements-1 if pair then element1 = Elem[concat, 2*e, esize]; element2 = Elem[concat, (2*e)+1, esize]; else element1 = Elem[operand1, e, esize]; element2 = Elem[operand2, e, esize]; if minimum then Elem[result, e, esize] = FPMinNum(element1, element2, FPCR); else Elem[result, e, esize] = FPMaxNum(element1, element2, FPCR); V[d, datasize] = 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.