mirror of
https://github.com/pound-emu/ballistic.git
synced 2026-01-31 01:15:21 +01:00
273 lines
17 KiB
XML
273 lines
17 KiB
XML
<?xml version="1.0" encoding="utf-8"?>
|
|
<?xml-stylesheet type="text/xsl" encoding="UTF-8" href="iform.xsl" version="1.0"?>
|
|
<!DOCTYPE instructionsection PUBLIC "-//ARM//DTD instructionsection //EN" "iform-p.dtd">
|
|
<!-- Copyright (c) 2010-2022 Arm Limited or its affiliates. All rights reserved. -->
|
|
<!-- This document is Non-Confidential. This document may only be used and distributed in accordance with the terms of the agreement entered into by Arm and the party that Arm delivered this document to. -->
|
|
|
|
<instructionsection id="FCMLA_advsimd_vec" title="FCMLA -- A64" type="instruction">
|
|
<docvars>
|
|
<docvar key="advsimd-type" value="simd" />
|
|
<docvar key="instr-class" value="advsimd" />
|
|
<docvar key="isa" value="A64" />
|
|
<docvar key="mnemonic" value="FCMLA" />
|
|
</docvars>
|
|
<heading>FCMLA</heading>
|
|
<desc>
|
|
<brief>
|
|
<para>Floating-point Complex Multiply Accumulate</para>
|
|
</brief>
|
|
<authored>
|
|
<para>Floating-point Complex Multiply Accumulate.</para>
|
|
<para>This instruction operates on complex numbers that are represented in SIMD&FP registers as pairs of elements, with the more significant element holding the imaginary part of the number and the less significant element holding the real part of the number. Each element holds a floating-point value. It performs the following computation on the corresponding complex number element pairs from the two source registers and the destination register:</para>
|
|
<list type="unordered">
|
|
<listitem><content>Considering the complex number from the second source register on an Argand diagram, the number is rotated counterclockwise by 0, 90, 180, or 270 degrees.</content></listitem>
|
|
<listitem><content>The two elements of the transformed complex number are multiplied by:<list type="unordered"><listitem><content>The real element of the complex number from the first source register, if the transformation was a rotation by 0 or 180 degrees.</content></listitem><listitem><content>The imaginary element of the complex number from the first source register, if the transformation was a rotation by 90 or 270 degrees.</content></listitem></list></content></listitem>
|
|
<listitem><content>The complex number resulting from that multiplication is added to the complex number from the destination register.</content></listitem>
|
|
</list>
|
|
<para>The multiplication and addition operations are performed as a fused multiply-add, without any intermediate rounding.</para>
|
|
<para>This instruction can generate a floating-point exception. Depending on the settings in <xref linkend="AArch64.fpcr">FPCR</xref>, the exception results in either a flag being set in <xref linkend="AArch64.fpsr">FPSR</xref> or a synchronous exception being generated. For more information, see <xref linkend="BEIJDDAG">Floating-point exception traps</xref>.</para>
|
|
<para>Depending on the settings in the <xref linkend="AArch64.cpacr_el1">CPACR_EL1</xref>, <xref linkend="AArch64.cptr_el2">CPTR_EL2</xref>, and <xref linkend="AArch64.cptr_el3">CPTR_EL3</xref> registers, and the current Security state and Exception level, an attempt to execute the instruction might be trapped.</para>
|
|
</authored>
|
|
</desc>
|
|
<alias_list howmany="0"></alias_list>
|
|
<classes>
|
|
<iclass name="Vector" oneof="1" id="iclass_simd" no_encodings="1" isa="A64">
|
|
<docvars>
|
|
<docvar key="advsimd-type" value="simd" />
|
|
<docvar key="instr-class" value="advsimd" />
|
|
<docvar key="isa" value="A64" />
|
|
<docvar key="mnemonic" value="FCMLA" />
|
|
</docvars>
|
|
<iclassintro count="1"></iclassintro>
|
|
<arch_variants>
|
|
<arch_variant name="ARMv8.3" feature="FEAT_FCMA" />
|
|
</arch_variants>
|
|
<regdiagram form="32" psname="aarch64/instrs/vector/arithmetic/binary/uniform/mul/fp/complex">
|
|
<box hibit="31" settings="1">
|
|
<c>0</c>
|
|
</box>
|
|
<box hibit="30" name="Q" usename="1">
|
|
<c></c>
|
|
</box>
|
|
<box hibit="29" name="U" settings="1">
|
|
<c>1</c>
|
|
</box>
|
|
<box hibit="28" width="5" settings="5">
|
|
<c>0</c>
|
|
<c>1</c>
|
|
<c>1</c>
|
|
<c>1</c>
|
|
<c>0</c>
|
|
</box>
|
|
<box hibit="23" width="2" name="size" usename="1">
|
|
<c colspan="2"></c>
|
|
</box>
|
|
<box hibit="21" settings="1">
|
|
<c>0</c>
|
|
</box>
|
|
<box hibit="20" width="5" name="Rm" usename="1">
|
|
<c colspan="5"></c>
|
|
</box>
|
|
<box hibit="15" settings="1">
|
|
<c>1</c>
|
|
</box>
|
|
<box hibit="14" width="2" settings="2">
|
|
<c>1</c>
|
|
<c>0</c>
|
|
</box>
|
|
<box hibit="12" width="2" name="rot" usename="1">
|
|
<c colspan="2"></c>
|
|
</box>
|
|
<box hibit="10" settings="1">
|
|
<c>1</c>
|
|
</box>
|
|
<box hibit="9" width="5" name="Rn" usename="1">
|
|
<c colspan="5"></c>
|
|
</box>
|
|
<box hibit="4" width="5" name="Rd" usename="1">
|
|
<c colspan="5"></c>
|
|
</box>
|
|
</regdiagram>
|
|
<encoding name="FCMLA_asimdsame2_C" oneofinclass="1" oneof="1" label="">
|
|
<docvars>
|
|
<docvar key="advsimd-type" value="simd" />
|
|
<docvar key="instr-class" value="advsimd" />
|
|
<docvar key="isa" value="A64" />
|
|
<docvar key="mnemonic" value="FCMLA" />
|
|
</docvars>
|
|
<asmtemplate><text>FCMLA </text><a link="sa_vd" hover="SIMD&FP destination register (field "Rd")"><Vd></a><text>.</text><a link="sa_t" hover="Arrangement specifier (field "size:Q") [2D,2S,4H,4S,8H]"><T></a><text>, </text><a link="sa_vn" hover="First SIMD&FP source register (field "Rn")"><Vn></a><text>.</text><a link="sa_t" hover="Arrangement specifier (field "size:Q") [2D,2S,4H,4S,8H]"><T></a><text>, </text><a link="sa_vm" hover="Second SIMD&FP source register (field "Rm")"><Vm></a><text>.</text><a link="sa_t" hover="Arrangement specifier (field "size:Q") [2D,2S,4H,4S,8H]"><T></a><text>, #</text><a link="sa_rotate" hover="Rotation (field "rot") [0,90,180,270]"><rotate></a></asmtemplate>
|
|
</encoding>
|
|
<ps_section howmany="1">
|
|
<ps name="aarch64/instrs/vector/arithmetic/binary/uniform/mul/fp/complex" mylink="aarch64.instrs.vector.arithmetic.binary.uniform.mul.fp.complex" enclabels="" sections="1" secttype="noheading">
|
|
<pstext mayhavelinks="1" section="Decode" rep_section="decode">if !<a link="impl-shared.HaveFCADDExt.0" file="shared_pseudocode.xml" hover="function: boolean HaveFCADDExt()">HaveFCADDExt</a>() then UNDEFINED;
|
|
integer d = <a link="impl-shared.UInt.1" file="shared_pseudocode.xml" hover="function: integer UInt(bits(N) x)">UInt</a>(Rd);
|
|
integer n = <a link="impl-shared.UInt.1" file="shared_pseudocode.xml" hover="function: integer UInt(bits(N) x)">UInt</a>(Rn);
|
|
integer m = <a link="impl-shared.UInt.1" file="shared_pseudocode.xml" hover="function: integer UInt(bits(N) x)">UInt</a>(Rm);
|
|
if size == '00' then UNDEFINED;
|
|
if Q == '0' && size == '11' then UNDEFINED;
|
|
integer esize = 8 << <a link="impl-shared.UInt.1" file="shared_pseudocode.xml" hover="function: integer UInt(bits(N) x)">UInt</a>(size);
|
|
if !<a link="impl-shared.HaveFP16Ext.0" file="shared_pseudocode.xml" hover="function: boolean HaveFP16Ext()">HaveFP16Ext</a>() && esize == 16 then UNDEFINED;
|
|
integer datasize = if Q == '1' then 128 else 64;
|
|
integer elements = datasize DIV esize;</pstext>
|
|
</ps>
|
|
</ps_section>
|
|
</iclass>
|
|
</classes>
|
|
<explanations scope="all">
|
|
<explanation enclist="FCMLA_asimdsame2_C" symboldefcount="1">
|
|
<symbol link="sa_vd"><Vd></symbol>
|
|
<account encodedin="Rd">
|
|
<intro>
|
|
<para>Is the name of the SIMD&FP destination register, encoded in the "Rd" field.</para>
|
|
</intro>
|
|
</account>
|
|
</explanation>
|
|
<explanation enclist="FCMLA_asimdsame2_C" symboldefcount="1">
|
|
<symbol link="sa_t"><T></symbol>
|
|
<definition encodedin="size:Q">
|
|
<intro>Is an arrangement specifier, </intro>
|
|
<table class="valuetable">
|
|
<tgroup cols="3">
|
|
<thead>
|
|
<row>
|
|
<entry class="bitfield">size</entry>
|
|
<entry class="bitfield">Q</entry>
|
|
<entry class="symbol"><T></entry>
|
|
</row>
|
|
</thead>
|
|
<tbody>
|
|
<row>
|
|
<entry class="bitfield">00</entry>
|
|
<entry class="bitfield">x</entry>
|
|
<entry class="symbol">RESERVED</entry>
|
|
</row>
|
|
<row>
|
|
<entry class="bitfield">01</entry>
|
|
<entry class="bitfield">0</entry>
|
|
<entry class="symbol">4H</entry>
|
|
</row>
|
|
<row>
|
|
<entry class="bitfield">01</entry>
|
|
<entry class="bitfield">1</entry>
|
|
<entry class="symbol">8H</entry>
|
|
</row>
|
|
<row>
|
|
<entry class="bitfield">10</entry>
|
|
<entry class="bitfield">0</entry>
|
|
<entry class="symbol">2S</entry>
|
|
</row>
|
|
<row>
|
|
<entry class="bitfield">10</entry>
|
|
<entry class="bitfield">1</entry>
|
|
<entry class="symbol">4S</entry>
|
|
</row>
|
|
<row>
|
|
<entry class="bitfield">11</entry>
|
|
<entry class="bitfield">0</entry>
|
|
<entry class="symbol">RESERVED</entry>
|
|
</row>
|
|
<row>
|
|
<entry class="bitfield">11</entry>
|
|
<entry class="bitfield">1</entry>
|
|
<entry class="symbol">2D</entry>
|
|
</row>
|
|
</tbody>
|
|
</tgroup>
|
|
</table>
|
|
</definition>
|
|
</explanation>
|
|
<explanation enclist="FCMLA_asimdsame2_C" symboldefcount="1">
|
|
<symbol link="sa_vn"><Vn></symbol>
|
|
<account encodedin="Rn">
|
|
<intro>
|
|
<para>Is the name of the first SIMD&FP source register, encoded in the "Rn" field.</para>
|
|
</intro>
|
|
</account>
|
|
</explanation>
|
|
<explanation enclist="FCMLA_asimdsame2_C" symboldefcount="1">
|
|
<symbol link="sa_vm"><Vm></symbol>
|
|
<account encodedin="Rm">
|
|
<intro>
|
|
<para>Is the name of the second SIMD&FP source register, encoded in the "Rm" field.</para>
|
|
</intro>
|
|
</account>
|
|
</explanation>
|
|
<explanation enclist="FCMLA_asimdsame2_C" symboldefcount="1">
|
|
<symbol link="sa_rotate"><rotate></symbol>
|
|
<definition encodedin="rot">
|
|
<intro>Is the rotation, </intro>
|
|
<table class="valuetable">
|
|
<tgroup cols="2">
|
|
<thead>
|
|
<row>
|
|
<entry class="bitfield">rot</entry>
|
|
<entry class="symbol"><rotate></entry>
|
|
</row>
|
|
</thead>
|
|
<tbody>
|
|
<row>
|
|
<entry class="bitfield">00</entry>
|
|
<entry class="symbol">0</entry>
|
|
</row>
|
|
<row>
|
|
<entry class="bitfield">01</entry>
|
|
<entry class="symbol">90</entry>
|
|
</row>
|
|
<row>
|
|
<entry class="bitfield">10</entry>
|
|
<entry class="symbol">180</entry>
|
|
</row>
|
|
<row>
|
|
<entry class="bitfield">11</entry>
|
|
<entry class="symbol">270</entry>
|
|
</row>
|
|
</tbody>
|
|
</tgroup>
|
|
</table>
|
|
</definition>
|
|
</explanation>
|
|
</explanations>
|
|
<ps_section howmany="1">
|
|
<ps name="aarch64/instrs/vector/arithmetic/binary/uniform/mul/fp/complex" mylink="execute" enclabels="" sections="1" secttype="Operation">
|
|
<pstext mayhavelinks="1" section="Execute" rep_section="execute"><a link="impl-aarch64.CheckFPAdvSIMDEnabled64.0" file="shared_pseudocode.xml" hover="function: CheckFPAdvSIMDEnabled64()">CheckFPAdvSIMDEnabled64</a>();
|
|
bits(datasize) operand1 = <a link="impl-aarch64.V.read.2" file="shared_pseudocode.xml" hover="accessor: bits(width) V[integer n, integer width]">V</a>[n, datasize];
|
|
bits(datasize) operand2 = <a link="impl-aarch64.V.read.2" file="shared_pseudocode.xml" hover="accessor: bits(width) V[integer n, integer width]">V</a>[m, datasize];
|
|
bits(datasize) operand3 = <a link="impl-aarch64.V.read.2" file="shared_pseudocode.xml" hover="accessor: bits(width) V[integer n, integer width]">V</a>[d, datasize];
|
|
bits(datasize) result;
|
|
bits(esize) element1;
|
|
bits(esize) element2;
|
|
bits(esize) element3;
|
|
bits(esize) element4;
|
|
<a link="FPCRType" file="shared_pseudocode.xml" hover="type FPCRType">FPCRType</a> fpcr = FPCR[];
|
|
|
|
for e = 0 to (elements DIV 2) -1
|
|
case rot of
|
|
when '00'
|
|
element1 = <a link="impl-shared.Elem.read.3" file="shared_pseudocode.xml" hover="accessor: bits(size) Elem[bits(N) vector, integer e, integer size]">Elem</a>[operand2, e*2, esize];
|
|
element2 = <a link="impl-shared.Elem.read.3" file="shared_pseudocode.xml" hover="accessor: bits(size) Elem[bits(N) vector, integer e, integer size]">Elem</a>[operand1, e*2, esize];
|
|
element3 = <a link="impl-shared.Elem.read.3" file="shared_pseudocode.xml" hover="accessor: bits(size) Elem[bits(N) vector, integer e, integer size]">Elem</a>[operand2, e*2+1, esize];
|
|
element4 = <a link="impl-shared.Elem.read.3" file="shared_pseudocode.xml" hover="accessor: bits(size) Elem[bits(N) vector, integer e, integer size]">Elem</a>[operand1, e*2, esize];
|
|
when '01'
|
|
element1 = <a link="impl-shared.FPNeg.1" file="shared_pseudocode.xml" hover="function: bits(N) FPNeg(bits(N) op)">FPNeg</a>(<a link="impl-shared.Elem.read.3" file="shared_pseudocode.xml" hover="accessor: bits(size) Elem[bits(N) vector, integer e, integer size]">Elem</a>[operand2, e*2+1, esize]);
|
|
element2 = <a link="impl-shared.Elem.read.3" file="shared_pseudocode.xml" hover="accessor: bits(size) Elem[bits(N) vector, integer e, integer size]">Elem</a>[operand1, e*2+1, esize];
|
|
element3 = <a link="impl-shared.Elem.read.3" file="shared_pseudocode.xml" hover="accessor: bits(size) Elem[bits(N) vector, integer e, integer size]">Elem</a>[operand2, e*2, esize];
|
|
element4 = <a link="impl-shared.Elem.read.3" file="shared_pseudocode.xml" hover="accessor: bits(size) Elem[bits(N) vector, integer e, integer size]">Elem</a>[operand1, e*2+1, esize];
|
|
when '10'
|
|
element1 = <a link="impl-shared.FPNeg.1" file="shared_pseudocode.xml" hover="function: bits(N) FPNeg(bits(N) op)">FPNeg</a>(<a link="impl-shared.Elem.read.3" file="shared_pseudocode.xml" hover="accessor: bits(size) Elem[bits(N) vector, integer e, integer size]">Elem</a>[operand2, e*2, esize]);
|
|
element2 = <a link="impl-shared.Elem.read.3" file="shared_pseudocode.xml" hover="accessor: bits(size) Elem[bits(N) vector, integer e, integer size]">Elem</a>[operand1, e*2, esize];
|
|
element3 = <a link="impl-shared.FPNeg.1" file="shared_pseudocode.xml" hover="function: bits(N) FPNeg(bits(N) op)">FPNeg</a>(<a link="impl-shared.Elem.read.3" file="shared_pseudocode.xml" hover="accessor: bits(size) Elem[bits(N) vector, integer e, integer size]">Elem</a>[operand2, e*2+1, esize]);
|
|
element4 = <a link="impl-shared.Elem.read.3" file="shared_pseudocode.xml" hover="accessor: bits(size) Elem[bits(N) vector, integer e, integer size]">Elem</a>[operand1, e*2, esize];
|
|
when '11'
|
|
element1 = <a link="impl-shared.Elem.read.3" file="shared_pseudocode.xml" hover="accessor: bits(size) Elem[bits(N) vector, integer e, integer size]">Elem</a>[operand2, e*2+1, esize];
|
|
element2 = <a link="impl-shared.Elem.read.3" file="shared_pseudocode.xml" hover="accessor: bits(size) Elem[bits(N) vector, integer e, integer size]">Elem</a>[operand1, e*2+1, esize];
|
|
element3 = <a link="impl-shared.FPNeg.1" file="shared_pseudocode.xml" hover="function: bits(N) FPNeg(bits(N) op)">FPNeg</a>(<a link="impl-shared.Elem.read.3" file="shared_pseudocode.xml" hover="accessor: bits(size) Elem[bits(N) vector, integer e, integer size]">Elem</a>[operand2, e*2, esize]);
|
|
element4 = <a link="impl-shared.Elem.read.3" file="shared_pseudocode.xml" hover="accessor: bits(size) Elem[bits(N) vector, integer e, integer size]">Elem</a>[operand1, e*2+1, esize];
|
|
|
|
<a link="impl-shared.Elem.write.3" file="shared_pseudocode.xml" hover="accessor: Elem[bits(N) &vector, integer e, integer size] = bits(size) value">Elem</a>[result, e*2, esize] = <a link="impl-shared.FPMulAdd.4" file="shared_pseudocode.xml" hover="function: bits(N) FPMulAdd(bits(N) addend, bits(N) op1, bits(N) op2, FPCRType fpcr)">FPMulAdd</a>(<a link="impl-shared.Elem.read.3" file="shared_pseudocode.xml" hover="accessor: bits(size) Elem[bits(N) vector, integer e, integer size]">Elem</a>[operand3, e*2, esize], element2, element1, fpcr);
|
|
<a link="impl-shared.Elem.write.3" file="shared_pseudocode.xml" hover="accessor: Elem[bits(N) &vector, integer e, integer size] = bits(size) value">Elem</a>[result, e*2+1, esize] = <a link="impl-shared.FPMulAdd.4" file="shared_pseudocode.xml" hover="function: bits(N) FPMulAdd(bits(N) addend, bits(N) op1, bits(N) op2, FPCRType fpcr)">FPMulAdd</a>(<a link="impl-shared.Elem.read.3" file="shared_pseudocode.xml" hover="accessor: bits(size) Elem[bits(N) vector, integer e, integer size]">Elem</a>[operand3, e*2+1, esize], element4, element3, fpcr);
|
|
|
|
<a link="impl-aarch64.V.write.2" file="shared_pseudocode.xml" hover="accessor: V[integer n, integer width] = bits(width) value">V</a>[d, datasize] = result;</pstext>
|
|
</ps>
|
|
</ps_section>
|
|
</instructionsection>
|