fp@369: %------------------------------------------------------------------------------
fp@369: %
fp@369: %  IgH EtherCAT Master Documentation
fp@369: %
fp@369: %  $Id$
fp@369: %
fp@1202: %  vi: spell spelllang=en tw=78
fp@2421: %
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@369: \documentclass[a4paper,12pt,BCOR6mm,bibtotoc,idxtotoc]{scrbook}
fp@369: 
fp@369: \usepackage[latin1]{inputenc}
fp@369: \usepackage[automark,headsepline]{scrpage2}
fp@369: \usepackage{graphicx}
fp@369: \usepackage{makeidx}
fp@369: \usepackage[refpage]{nomencl}
fp@369: \usepackage{listings}
fp@1588: \usepackage[nofancy]{rcsinfo}
fp@1085: \usepackage{SIunits}
fp@1517: \usepackage{amsmath} % for \text{}
fp@2297: \usepackage{longtable}
fp@1275: \usepackage{hyperref}
fp@1275: 
fp@1275: \hypersetup{pdfpagelabels,plainpages=false}
fp@1275: \hypersetup{linkcolor=blue,colorlinks=true,urlcolor=blue}
fp@369: 
fp@369: \setlength{\parskip}{0.8ex plus 0.8ex minus 0.5ex}
fp@369: \setlength{\parindent}{0mm}
fp@369: 
fp@369: \setcounter{secnumdepth}{\subsubsectionlevel}
fp@369: 
fp@369: \DeclareFontShape{OT1}{cmtt}{bx}{n}
fp@369: {
fp@369:   <5><6><7><8><9><10><10.95><12><14.4><17.28><20.74><24.88>cmttb10
fp@369: }{}
fp@369: 
fp@369: \lstset{basicstyle=\ttfamily\small,numberstyle=\tiny,aboveskip=4mm,
fp@2515:   belowskip=2mm,escapechar=`,breaklines=true}
fp@369: \renewcommand\lstlistlistingname{List of Listings}
fp@369: 
fp@917: % Workaround for lstlistoflistings bug
fp@917: \makeatletter% --> De-TeX-FAQ
fp@917: \renewcommand*{\lstlistoflistings}{%
fp@917:   \begingroup
fp@917:     \if@twocolumn
fp@917:       \@restonecoltrue\onecolumn
fp@917:     \else
fp@917:       \@restonecolfalse
fp@917:     \fi
fp@917:     \lol@heading
fp@917:     \setlength{\parskip}{\z@}%
fp@917:     \setlength{\parindent}{\z@}%
fp@917:     \setlength{\parfillskip}{\z@ \@plus 1fil}%
fp@917:     \@starttoc{lol}%
fp@917:     \if@restonecol\twocolumn\fi
fp@917:   \endgroup
fp@917: }
fp@917: \makeatother% --> \makeatletter
fp@917: 
fp@369: \renewcommand\nomname{Glossary}
fp@369: 
fp@369: \newcommand{\IgH}{\raisebox{-0.7667ex}
fp@369:   {\includegraphics[height=2.2ex]{images/ighsign}}}
fp@369: 
fp@1589: \rcsInfo $RCSId$
fp@369: 
fp@2515: \newcommand{\masterversion}{1.5.2}
fp@1085: \newcommand{\linenum}[1]{\normalfont\textcircled{\tiny #1}}
fp@487: 
fp@369: \makeindex
fp@917: \makenomenclature
fp@369: 
fp@1589: % Revision and date on inner footer
fp@1589: \ifoot[\scriptsize\rcsInfoRevision, \rcsInfoDate]
fp@1589:     {\scriptsize\rcsInfoRevision, \rcsInfoDate}
fp@1589: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@369: \begin{document}
fp@369: 
fp@369: \pagenumbering{roman}
fp@369: \pagestyle{empty}
fp@369: 
fp@369: \begin{titlepage}
fp@369:   \begin{center}
fp@369:     \rule{\textwidth}{1.5mm}
fp@369: 
fp@1917:     {\Huge\sf\textbf{IgH \includegraphics[height=2.4ex]{images/ethercat}
fp@1917:       Master \masterversion}\\[1ex]
fp@1917:       \textbf{Documentation}}
fp@369: 
fp@369:     \vspace{1ex}
fp@369:     \rule{\textwidth}{1.5mm}
fp@369: 
fp@1204:     \vspace{\fill} {\Large Dipl.-Ing. (FH) Florian Pose,
fp@2691:     \url{fp@igh.de}\\[1ex] Ingenieurgemeinschaft \IgH}
fp@369: 
fp@369:     \vspace{\fill}
fp@1588:     {\Large Essen, \rcsInfoLongDate\\[1ex]
fp@1588:       Revision \rcsInfoRevision}
fp@369:   \end{center}
fp@369: \end{titlepage}
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@1589: \pagestyle{scrplain}
fp@1589: 
fp@369: \tableofcontents
fp@369: \listoftables
fp@369: \listoffigures
fp@1204: %\lstlistoflistings
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@369: \newpage
fp@369: \pagestyle{scrheadings}
fp@369: 
fp@369: \section*{Conventions}
fp@369: \addcontentsline{toc}{section}{Conventions}
fp@369: \markleft{Conventions}
fp@369: 
fp@369: The following typographic conventions are used:
fp@369: 
fp@369: \begin{itemize}
fp@1085: 
fp@1085: \item \textit{Italic face} is used for newly introduced terms and file names.
fp@1085: 
fp@1085: \item \texttt{Typewriter face} is used for code examples and command line
fp@1085: output.
fp@1085: 
fp@1085: \item \texttt{\textbf{Bold typewriter face}} is used for user input in command
fp@1085: lines.
fp@1085: 
fp@369: \end{itemize}
fp@369: 
fp@1085: Data values and addresses are usually specified as hexadecimal values. These
fp@1085: are marked in the \textit{C} programming language style with the prefix
fp@1085: \lstinline+0x+ (example: \lstinline+0x88A4+). Unless otherwise noted, address
fp@1085: values are specified as byte addresses.
fp@369: 
fp@369: Function names are always printed with parentheses, but without
fp@1085: parameters. So, if a function \lstinline+ecrt_request_master()+ has
fp@1085: empty parentheses, this shall not imply that it has no parameters.
fp@1085: 
fp@1085: If shell commands have to be entered, this is marked by a dollar prompt:
fp@1085: 
fp@1085: \begin{lstlisting}
fp@1085: $
fp@369: \end{lstlisting}
fp@369: 
fp@369: Further, if a shell command has to be entered as the superuser, the
fp@1085: prompt is a mesh:
fp@1085: 
fp@1085: \begin{lstlisting}
fp@1085: #
fp@369: \end{lstlisting}
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@369: \chapter{The IgH EtherCAT Master}
fp@369: \label{chapter:master}
fp@369: \pagenumbering{arabic}
fp@369: 
fp@1085: This chapter covers some general information about the EtherCAT master.
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@369: \section{Feature Summary}
fp@369: \label{sec:summary}
fp@369: \index{Master!Features}
fp@369: 
fp@1085: The list below gives a short summary of the master features.
fp@369: 
fp@369: \begin{itemize}
fp@1085: 
fp@2515: \item Designed as a kernel module for Linux 2.6 / 3.x.
fp@1085: 
fp@1086: \item Implemented according to IEC 61158-12 \cite{dlspec} \cite{alspec}.
fp@1086: 
fp@1588: \item Comes with EtherCAT-capable native drivers for several common Ethernet
fp@1588: chips, as well as a generic driver for all chips supported by the Linux
fp@1588: kernel.
fp@1085: 
fp@369:   \begin{itemize}
fp@1085: 
fp@1588:   \item The native drivers operate the hardware without interrupts.
fp@1588: 
fp@1588:   \item Native drivers for additional Ethernet hardware can easily be
fp@2515:   implemented using the common device interface (see~\autoref{sec:ecdev})
fp@1588:   provided by the master module.
fp@1588: 
fp@1588:   \item For any other hardware, the generic driver can be used. It uses the
fp@1588:   lower layers of the Linux network stack.
fp@1085: 
fp@369:   \end{itemize}
fp@1085: 
fp@1085: \item The master module supports multiple EtherCAT masters running in
fp@1085: parallel.
fp@1085: 
fp@1085: \item The master code supports any Linux realtime extension through its
fp@1085: independent architecture.
fp@1085: 
fp@369:   \begin{itemize}
fp@1085: 
fp@2515:   \item RTAI\nomenclature{RTAI}{Realtime Application Interface} \cite{rtai}
fp@2515:   (including LXRT via RTDM), ADEOS\nomenclature{ADEOS}{Adaptive Domain
fp@2515:   Environment for Operating Systems}, RT-Preempt \cite{rt-preempt}, Xenomai
fp@2515:   (including RTDM), etc.
fp@1085: 
fp@369:   \item It runs well even without realtime extensions.
fp@1085: 
fp@369:   \end{itemize}
fp@1085: 
fp@1204: \item Common ``Application Interface'' for applications, that want to use
fp@2515: EtherCAT functionality (see \autoref{chap:api}).
fp@1085: 
fp@1085: \item \textit{Domains} are introduced, to allow grouping of process
fp@1085:   data transfers with different slave groups and task periods.
fp@1085: 
fp@369:   \begin{itemize}
fp@1085: 
fp@1085:   \item Handling of multiple domains with different task periods.
fp@1085: 
fp@1085:   \item Automatic calculation of process data mapping, FMMU and sync manager
fp@1085:   configuration within each domain.
fp@1085: 
fp@369:   \end{itemize}
fp@1085: 
fp@1085: \item Communication through several finite state machines.
fp@1085: 
fp@369:   \begin{itemize}
fp@1085: 
fp@1085:   \item Automatic bus scanning after topology changes.
fp@1085: 
fp@1085:   \item Bus monitoring during operation.
fp@1085: 
fp@1085:   \item Automatic reconfiguration of slaves (for example after power failure)
fp@1085:   during operation.
fp@1085: 
fp@369:   \end{itemize}
fp@1085: 
fp@2515: \item Distributed Clocks support (see \autoref{sec:dc}).
fp@1517: 
fp@1517:   \begin{itemize}
fp@1517: 
fp@1517:   \item Configuration of the slave's DC parameters through the application
fp@1517:   interface.
fp@1517: 
fp@1517:   \item Synchronization (offset and drift compensation) of the distributed
fp@1517:   slave clocks to the reference clock.
fp@1517: 
fp@2515:   \item Optional synchronization of the reference clock to the master clock or
fp@2515:   the other way round.
fp@1517: 
fp@1517:   \end{itemize}
fp@1517: 
fp@1327: \item CANopen over EtherCAT (CoE)
fp@1085: 
fp@369:   \begin{itemize}
fp@1085: 
fp@1327:   \item SDO upload, download and information service.
fp@1327: 
fp@1327:   \item Slave configuration via SDOs.
fp@1327: 
fp@1327:   \item SDO access from userspace and from the application.
fp@1085: 
fp@369:   \end{itemize}
fp@1085: 
fp@1327: \item Ethernet over EtherCAT (EoE)
fp@1085: 
fp@369:   \begin{itemize}
fp@1085: 
fp@1085:   \item Transparent use of EoE slaves via virtual network interfaces.
fp@1085: 
fp@1085:   \item Natively supports either a switched or a routed EoE network
fp@1085:   architecture.
fp@1085: 
fp@369:   \end{itemize}
fp@1085: 
fp@1364: \item Vendor-specific over EtherCAT (VoE)
fp@1364: 
fp@1364:   \begin{itemize}
fp@1364: 
fp@1364:   \item Communication with vendor-specific mailbox protocols via the API.
fp@1364: 
fp@1364:   \end{itemize}
fp@1364: 
fp@1364: \item File Access over EtherCAT (FoE)
fp@1364: 
fp@1364:   \begin{itemize}
fp@1364: 
fp@1364:   \item Loading and storing files via the command-line tool.
fp@1364: 
fp@1364:   \item Updating a slave's firmware can be done easily.
fp@1364: 
fp@1364:   \end{itemize}
fp@1364: 
fp@1917: \item Servo Profile over EtherCAT (SoE)
fp@1917: 
fp@1917:   \begin{itemize}
fp@1917: 
fp@1917:   \item Implemented according to IEC 61800-7 \cite{soespec}.
fp@1917: 
fp@1917:   \item Storing IDN configurations, that are written to the slave during
fp@1917:   startup.
fp@1917: 
fp@1917:   \item Accessing IDNs via the command-line tool.
fp@1917: 
fp@1917:   \item Accessing IDNs at runtime via the the user-space library.
fp@1917: 
fp@1917:   \end{itemize}
fp@1917: 
fp@2515: \item Userspace command-line-tool ``ethercat'' (see \autoref{sec:tool})
fp@1085: 
fp@369:   \begin{itemize}
fp@1085: 
fp@1364:   \item Detailed information about master, slaves, domains and bus
fp@1364:   configuration.
fp@1085:   \item Setting the master's debug level.
fp@1364:   \item Reading/Writing alias addresses.
fp@1364:   \item Listing slave configurations.
fp@1364:   \item Viewing process data.
fp@1364:   \item SDO download/upload; listing SDO dictionaries.
fp@1364:   \item Loading and storing files via FoE.
fp@1917:   \item SoE IDN access.
fp@1364:   \item Access to slave registers.
fp@1364:   \item Slave SII (EEPROM) access.
fp@1364:   \item Controlling application-layer states.
fp@1517:   \item Generation of slave description XML and C-code from existing slaves.
fp@1085: 
fp@369:   \end{itemize}
fp@1085: 
fp@369: \item Seamless system integration though LSB\nomenclature{LSB}{Linux
fp@369:     Standard Base} compliance.
fp@1085: 
fp@369:   \begin{itemize}
fp@1085: 
fp@1095:   \item Master and network device configuration via sysconfig files.
fp@1085: 
fp@1085:   \item Init script for master control.
fp@1085: 
fp@2515:   \item Service file for systemd.
fp@2515: 
fp@369:   \end{itemize}
fp@1085: 
fp@369: \item Virtual read-only network interface for monitoring and debugging
fp@369:   purposes.
fp@1085: 
fp@369: \end{itemize}
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@369: \section{License}
fp@369: \label{sec:license}
fp@369: 
fp@1275: The master code is released under the terms and conditions of the GNU General
fp@1275: Public License (GPL \cite{gpl})\index{GPL}, version 2. Other developers, that
fp@1275: want to use EtherCAT with Linux systems, are invited to use the master code or
fp@1275: even participate on development.
fp@1275: 
fp@1275: To allow static linking of userspace application against the master's
fp@2515: application interface (see \autoref{chap:api}), the userspace library (see
fp@2515: \autoref{sec:userlib}) is licensed under the terms and conditions of the GNU
fp@1275: Lesser General Public License (LGPL \cite{lgpl})\index{LGPL}, version 2.1.
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@1085: \chapter{Architecture}
fp@1299: \label{chap:arch}
fp@369: \index{Master!Architecture}
fp@369: 
fp@2515: The EtherCAT master is integrated into the Linux kernel. This was an early
fp@2515: design decision, which has been made for several reasons:
fp@369: 
fp@369: \begin{itemize}
fp@1085: 
fp@1306: \item Kernel code has significantly better realtime characteristics, i.\,e.\
fp@1306: less latency than userspace code. It was foreseeable, that a fieldbus master
fp@1306: has a lot of cyclic work to do. Cyclic work is usually triggered by timer
fp@1306: interrupts inside the kernel. The execution delay of a function that processes
fp@1306: timer interrupts is less, when it resides in kernelspace, because there is no
fp@1306: need of time-consuming context switches to a userspace process.
fp@1085: 
fp@369: \item It was also foreseeable, that the master code has to directly
fp@1085: communicate with the Ethernet hardware. This has to be done in the kernel
fp@1085: anyway (through network device drivers), which is one more reason for the
fp@1275: master code being in kernelspace.
fp@1085: 
fp@369: \end{itemize}
fp@369: 
fp@2515: \autoref{fig:arch} gives a general overview of the master architecture.
fp@369: 
fp@369: \begin{figure}[htbp]
fp@369:   \centering
fp@1588:   \includegraphics[width=\textwidth]{images/architecture}
fp@1291:   \caption{Master Architecture}
fp@1085:   \label{fig:arch}
fp@369: \end{figure}
fp@369: 
fp@1291: The components of the master environment are described below:
fp@1291: 
fp@1291: \begin{description}
fp@1291: 
fp@1291: \item[Master Module]\index{Master Module} Kernel module containing one or more
fp@2515: EtherCAT master instances (see \autoref{sec:mastermod}), the ``Device
fp@2515: Interface'' (see \autoref{sec:ecdev}) and the ``Application Interface'' (see
fp@2515: \autoref{chap:api}).
fp@1291: 
fp@1291: \item[Device Modules]\index{Device modules} EtherCAT-capable Ethernet device
fp@1291: driver modules\index{Device modules}, that offer their devices to the EtherCAT
fp@2515: master via the device interface (see \autoref{sec:ecdev}). These modified
fp@1291: network drivers can handle network devices used for EtherCAT operation and
fp@1291: ``normal'' Ethernet devices in parallel. A master can accept a certain device
fp@1291: and then is able to send and receive EtherCAT frames. Ethernet devices
fp@1291: declined by the master module are connected to the kernel's network stack as
fp@1291: usual.
fp@1291: 
fp@1291: \item[Application]\index{Application} A program that uses the EtherCAT master
fp@1291: (usually for cyclic exchange of process data with EtherCAT slaves). These
fp@1291: programs are not part of the EtherCAT master code\footnote{Although there are
fp@1291: some examples provided in the \textit{examples/} directory.}, but have to be
fp@2515: generated or written by the user. An application can request a master through
fp@2515: the application interface (see \autoref{chap:api}). If this succeeds, it has
fp@2515: the control over the master: It can provide a bus configuration and exchange
fp@2515: process data.  Applications can be kernel modules (that use the kernel
fp@2515: application interface directly) or userspace programs, that use the
fp@2515: application interface via the EtherCAT library (see \autoref{sec:userlib}), or
fp@2515: the RTDM library (see~\autoref{sec:rtdm}).
fp@1291: 
fp@1291: \end{description}
fp@1085: 
fp@1085: %------------------------------------------------------------------------------
fp@1085: 
fp@1306: \section{Master Module}
fp@1306: \label{sec:mastermod}
fp@1306: \index{Master module}
fp@1306: 
fp@1306: The EtherCAT master kernel module \textit{ec\_master} can contain multiple
fp@2515: master instances. Each master waits for certain Ethernet device(s) identified
fp@2515: by its MAC address(es)\index{MAC address}. These addresses have to be
fp@2515: specified on module loading via the \textit{main\_devices} (and optional:
fp@2515: \textit{backup\_devices}) module parameter.  The number of master instances to
fp@2515: initialize is taken from the number of MAC addresses given.
fp@1306: 
fp@1306: The below command loads the master module with a single master instance that
fp@2515: waits for one Ethernet device with the MAC address
fp@1306: \lstinline+00:0E:0C:DA:A2:20+. The master will be accessible via index $0$.
fp@1306: 
fp@1306: \begin{lstlisting}
fp@1306: # `\textbf{modprobe ec\_master main\_devices=00:0E:0C:DA:A2:20}`
fp@1306: \end{lstlisting}
fp@1306: 
fp@1306: MAC addresses for multiple masters have to be separated by commas:
fp@1306: 
fp@1306: \begin{lstlisting}
fp@1306: # `\textbf{modprobe ec\_master main\_devices=00:0E:0C:DA:A2:20,00:e0:81:71:d5:1c}`
fp@1306: \end{lstlisting}
fp@1306: 
fp@1306: The two masters can be addressed by their indices 0 and 1 respectively (see
fp@2515: \autoref{fig:masters}). The master index is needed for the
fp@1306: \lstinline+ecrt_master_request()+ function of the application interface (see
fp@2515: \autoref{chap:api}) and the \lstinline+--master+ option of the
fp@2515: \textit{ethercat} command-line tool (see \autoref{sec:tool}), which defaults
fp@1306: to $0$.
fp@1306: 
fp@1306: \begin{figure}[htbp]
fp@1306:   \centering
fp@1306:   \includegraphics[width=.5\textwidth]{images/masters}
fp@1306:   \caption{Multiple masters in one module}
fp@1306:   \label{fig:masters}
fp@1306: \end{figure}
fp@1306: 
fp@1399: \paragraph{Debug Level} The master module also has a parameter
fp@1399: \textit{debug\_level} to set the initial debug level for all masters (see
fp@2515: also~\autoref{sec:ethercat-debug}).
fp@1399: 
fp@1306: \paragraph{Init Script}
fp@1306: \index{Init script}
fp@1306: 
fp@1306: In most cases it is not necessary to load the master module and the Ethernet
fp@1306: driver modules manually. There is an init script available, so the master can
fp@2515: be started as a service (see \autoref{sec:system}). For systems that are
fp@2515: managed by systemd \cite{systemd}, there is also a service file available.
fp@1306: 
fp@1306: \paragraph{Syslog}
fp@1306: 
fp@1306: The master module outputs information about its state and events to the kernel
fp@2515: ring buffer. These also end up in the system logs. The above module loading
fp@1306: command should result in the messages below:
fp@1306: 
fp@1306: \begin{lstlisting}
fp@1306: # `\textbf{dmesg | tail -2}`
fp@1306: EtherCAT: Master driver `\masterversion`
fp@1306: EtherCAT: 2 masters waiting for devices.
fp@1306: 
fp@1306: # `\textbf{tail -2 /var/log/messages}`
fp@1306: Jul  4 10:22:45 ethercat kernel: EtherCAT: Master driver `\masterversion`
fp@1306: Jul  4 10:22:45 ethercat kernel: EtherCAT: 2 masters waiting
fp@1306:                                  for devices.
fp@1306: \end{lstlisting}
fp@1306: 
fp@2515: Master output is prefixed with \lstinline+EtherCAT+ which makes searching the
fp@2515: logs easier.
fp@1306: 
fp@1306: %------------------------------------------------------------------------------
fp@1306: 
fp@1306: \section{Master Phases}
fp@1085: \index{Master phases}
fp@1085: 
fp@1306: Every EtherCAT master provided by the master module (see
fp@2515: \autoref{sec:mastermod}) runs through several phases (see
fp@2515: \autoref{fig:phases}):
fp@1085: 
fp@1085: \begin{figure}[htbp]
fp@1085:   \centering
fp@1085:   \includegraphics[width=.9\textwidth]{images/phases}
fp@1085:   \caption{Master phases and transitions}
fp@1085:   \label{fig:phases}
fp@1085: \end{figure}
fp@1291: 
fp@1085: \begin{description}
fp@1085: 
fp@1085: \item[Orphaned phase]\index{Orphaned phase} This mode takes effect, when the
fp@2515: master still waits for its Ethernet device(s) to connect. No bus communication
fp@2515: is possible until then.
fp@1086: 
fp@1086: \item[Idle phase]\index{Idle phase} takes effect when the master has accepted
fp@2515: all required Ethernet devices, but is not requested by any application yet.
fp@2515: The master runs its state machine (see \autoref{sec:fsm-master}), that
fp@2515: automatically scans the bus for slaves and executes pending operations from
fp@2515: the userspace interface (for example SDO access). The command-line tool can be
fp@2515: used to access the bus, but there is no process data exchange because of the
fp@2515: missing bus configuration.
fp@1085: 
fp@1085: \item[Operation phase]\index{Operation phase} The master is requested by an
fp@1085: application that can provide a bus configuration and exchange process data.
fp@1085: 
fp@1085: \end{description}
fp@1085: 
fp@1085: %------------------------------------------------------------------------------
fp@1085: 
fp@1293: \section{Process Data}
fp@369: \label{sec:processdata}
fp@369: 
fp@1293: This section shall introduce a few terms and ideas how the master handles
fp@1293: process data.
fp@1085: 
fp@369: \paragraph{Process Data Image}
fp@369: \index{Process data}
fp@369: 
fp@1293: Slaves offer their inputs and outputs by presenting the master so-called
fp@1327: ``Process Data Objects'' (PDOs\index{PDO}). The available PDOs can be either
fp@2515: determined by reading out the slave's TxPDO and RxPDO SII categories from the
fp@1327: E$^2$PROM (in case of fixed PDOs) or by reading out the appropriate CoE
fp@2515: objects (see \autoref{sec:coe}), if available.  The application can register
fp@1327: the PDOs' entries for exchange during cyclic operation. The sum of all
fp@1327: registered PDO entries defines the ``process data image'', which is exchanged
fp@1293: via datagrams with ``logical'' memory access (like LWR, LRD or LRW) introduced
fp@1293: in~\cite[sec.~5.4]{dlspec}.
fp@369: 
fp@369: \paragraph{Process Data Domains}
fp@369: \index{Domain}
fp@369: 
fp@1085: The process data image can be easily managed by creating so-called
fp@1327: ``domains'', which allow grouped PDO exchange. They also take care of managing
fp@1327: the datagram structures needed to exchange the PDOs. Domains are mandatory for
fp@1293: process data exchange, so there has to be at least one. They were introduced
fp@1293: for the following reasons:
fp@369: 
fp@369: \begin{itemize}
fp@1293: 
fp@1293: \item The maximum size of a datagram is limited due to the limited size of an
fp@1293: Ethernet frame: The maximum data size is the Ethernet data field size minus
fp@1293: the EtherCAT frame header, EtherCAT datagram header and EtherCAT datagram
fp@1293: footer: $1500 - 2 - 12 - 2 = 1484$ octets. If the size of the process data
fp@1293: image exceeds this limit, multiple frames have to be sent, and the image has
fp@1293: to be partitioned for the use of multiple datagrams. A domain manages this
fp@1293: automatically.
fp@1293: 
fp@1327: \item Not every PDO has to be exchanged with the same frequency: The values of
fp@1327: PDOs can vary slowly over time (for example temperature values), so exchanging
fp@1293: them with a high frequency would just waste bus bandwidth. For this reason,
fp@1327: multiple domains can be created, to group different PDOs and so allow separate
fp@1293: exchange.
fp@1293: 
fp@369: \end{itemize}
fp@369: 
fp@1293: There is no upper limit for the number of domains, but each domain occupies
fp@1293: one FMMU in each slave involved, so the maximum number of domains is de facto
fp@1293: limited by the slaves.
fp@369: 
fp@369: \paragraph{FMMU Configuration}
fp@369: \index{FMMU!Configuration}
fp@369: 
fp@1327: An application can register PDO entries for exchange. Every PDO entry and its
fp@1327: parent PDO is part of a memory area in the slave's physical memory, that is
fp@1293: protected by a sync manager \cite[sec.~6.7]{dlspec} for synchronized access.
fp@1293: In order to make a sync manager react on a datagram accessing its memory, it
fp@1293: is necessary to access the last byte covered by the sync manager. Otherwise
fp@1293: the sync manager will not react on the datagram and no data will be exchanged.
fp@1293: That is why the whole synchronized memory area has to be included into the
fp@1327: process data image: For example, if a certain PDO entry of a slave is
fp@1293: registered for exchange with a certain domain, one FMMU will be configured to
fp@1327: map the complete sync-manager-protected memory, the PDO entry resides in. If a
fp@1327: second PDO entry of the same slave is registered for process data exchange
fp@1293: within the same domain, and it resides in the same sync-manager-protected
fp@1293: memory as the first one, the FMMU configuration is not altered, because the
fp@1293: desired memory is already part of the domain's process data image. If the
fp@1327: second PDO entry would belong to another sync-manager-protected area, this
fp@1293: complete area would also be included into the domains process data image.
fp@1293: 
fp@2515: \autoref{fig:fmmus} gives an overview, how FMMUs are configured to map
fp@1293: physical memory to logical process data images.
fp@369: 
fp@369: \begin{figure}[htbp]
fp@369:   \centering
fp@369:   \includegraphics[width=\textwidth]{images/fmmus}
fp@1293:   \caption{FMMU Configuration}
fp@369:   \label{fig:fmmus}
fp@369: \end{figure}
fp@369: 
fp@1085: %------------------------------------------------------------------------------
fp@1085: 
fp@1085: \chapter{Application Interface}
fp@1289: \label{chap:api}
fp@1085: \index{Application interface}
fp@1085: 
fp@1297: % TODO
fp@1297: %
fp@1297: % Interface version
fp@1297: % Master Requesting and Releasing
fp@1297: % Master Locking
fp@1327: % Configuring PDO assignment and mapping
fp@1297: % Domains (memory)
fp@1327: % PDO entry registration
fp@1327: % SDO configuration
fp@1327: % SDO access
fp@1917: % IDN configurations
fp@1917: % IDN access
fp@1202: 
fp@1094: The application interface provides functions and data structures for
fp@1292: applications to access an EtherCAT master. The complete documentation of the
fp@1292: interface is included as Doxygen~\cite{doxygen} comments in the header file
fp@1292: \textit{include/ecrt.h}. It can either be read directly from the file
fp@1293: comments, or as a more comfortable HTML documentation. The HTML generation is
fp@2515: described in \autoref{sec:gendoc}.
fp@1094: 
fp@1094: The following sections cover a general description of the application
fp@1094: interface.
fp@1094: 
fp@1094: Every application should use the master in two steps:
fp@1094: 
fp@1094: \begin{description}
fp@1094: 
fp@1094: \item[Configuration] The master is requested and the configuration is applied.
fp@1327: For example, domains are created, slaves are configured and PDO entries are
fp@2515: registered (see \autoref{sec:masterconfig}).
fp@1293: 
fp@1293: \item[Operation] Cyclic code is run and process data are exchanged (see
fp@2515: \autoref{sec:cyclic}).
fp@1094: 
fp@1094: \end{description}
fp@1094: 
fp@1293: \paragraph{Example Applications}\index{Example Applications} There are a few
fp@1204: example applications in the \textit{examples/} subdirectory of the master
fp@1204: code. They are documented in the source code.
fp@1204: 
fp@1094: %------------------------------------------------------------------------------
fp@1094: 
fp@1094: \section{Master Configuration}
fp@1094: \label{sec:masterconfig}
fp@1094: 
fp@1293: The bus configuration is supplied via the application interface.
fp@2515: \autoref{fig:app-config} gives an overview of the objects, that can be
fp@1293: configured by the application.
fp@1094: 
fp@1094: \begin{figure}[htbp]
fp@1094:   \centering
fp@1094:   \includegraphics[width=.8\textwidth]{images/app-config}
fp@1203:   \caption{Master Configuration}
fp@1094:   \label{fig:app-config}
fp@1094: \end{figure}
fp@1094: 
fp@1293: \subsection{Slave Configuration}
fp@1293: 
fp@1293: The application has to tell the master about the expected bus topology. This
fp@1293: can be done by creating ``slave configurations''. A slave configuration can be
fp@1293: seen as an expected slave. When a slave configuration is created, the
fp@1293: application provides the bus position (see below), vendor id and product code.
fp@1293: 
fp@1293: When the bus configuration is applied, the master checks, if there is a slave
fp@1293: with the given vendor id and product code at the given position. If this is
fp@1293: the case, the slave configuration is ``attached'' to the real slave on the bus
fp@1293: and the slave is configured according to the settings provided by the
fp@1293: application. The state of a slave configuration can either be queried via the
fp@1293: application interface or via the command-line tool (see
fp@2515: \autoref{sec:ethercat-config}).
fp@1293: 
fp@1293: \paragraph{Slave Position} The slave position has to be specified as a tuple
fp@1306: of ``alias'' and ``position''. This allows addressing slaves either via an
fp@1293: absolute bus position, or a stored identifier called ``alias'', or a mixture
fp@1293: of both. The alias is a 16-bit value stored in the slave's E$^2$PROM. It can
fp@2515: be modified via the command-line tool (see \autoref{sec:ethercat-alias}).
fp@2515: \autoref{tab:slaveposition} shows, how the values are interpreted.
fp@1293: 
fp@1293: \begin{table}[htbp]
fp@1293:   \centering
fp@1293:   \caption{Specifying a Slave Position}
fp@1293:   \label{tab:slaveposition}
fp@1293:   \vspace{2mm}
fp@1293:   \begin{tabular}{c|c|p{70mm}}
fp@1293:     Alias & Position & Interpretation\\
fp@1293:     \hline
fp@1293: 
fp@1293:     \lstinline+0+ & \lstinline+0+ -- \lstinline+65535+ &
fp@1293: 
fp@1293:     Position addressing. The position parameter is interpreted as the absolute
fp@1293:     ring position in the bus.\\ \hline
fp@1293: 
fp@1293:     \lstinline+1+ -- \lstinline+65535+ & \lstinline+0+ -- \lstinline+65535+ &
fp@1293: 
fp@1293:     Alias addressing. The position parameter is interpreted as relative
fp@1293:     position after the first slave with the given alias address. \\ \hline
fp@1293: 
fp@1293:   \end{tabular}
fp@1293: \end{table}
fp@1293: 
fp@2515: \autoref{fig:attach} shows an example of how slave configurations are
fp@1293: attached. Some of the configurations were attached, while others remain
fp@1293: detached. The below lists gives the reasons beginning with the top slave
fp@1293: configuration.
fp@1293: 
fp@1293: \begin{figure}[htbp]
fp@1293:   \centering
fp@1293:   \includegraphics[width=.7\textwidth]{images/attach}
fp@1293:   \caption{Slave Configuration Attachment}
fp@1293:   \label{fig:attach}
fp@1293: \end{figure}
fp@1293: 
fp@1293: \begin{enumerate}
fp@1293: 
fp@1293: \item A zero alias means to use simple position addressing. Slave 1 exists and
fp@1293: vendor id and product code match the expected values.
fp@1293: 
fp@1293: \item Although the slave with position 0 is found, the product code does not
fp@1293: match, so the configuration is not attached.
fp@1293: 
fp@1293: \item The alias is non-zero, so alias addressing is used. Slave 2 is the first
fp@1293: slave with alias \lstinline+0x2000+. Because the position value is zero, the
fp@1293: same slave is used.
fp@1293: 
fp@1293: \item There is no slave with the given alias, so the configuration can not be
fp@1293: attached.
fp@1293: 
fp@1293: \item Slave 2 is again the first slave with the alias \lstinline+0x2000+, but
fp@1293: position is now 1, so slave 3 is attached.
fp@1293: 
fp@1293: \end{enumerate}
fp@1293: 
fp@2691: f the master sources are configured with \lstinline+--enable-wildcards+, then
fp@2515: \lstinline+0xffffffff+ matches every vendor ID and/or product code.
fp@2515: 
fp@1094: %------------------------------------------------------------------------------
fp@1094: 
fp@1094: \section{Cyclic Operation}
fp@1094: \label{sec:cyclic}
fp@1094: 
fp@1293: 
fp@1293: To enter cyclic operation mode, the master has to be ``activated'' to
fp@1293: calculate the process data image and apply the bus configuration for the first
fp@1293: time. After activation, the application is in charge to send and receive
fp@2515: frames. The configuration can not be changed after activation.
fp@1293: 
fp@1297: % TODO
fp@1297: %
fp@1297: % PDO endianess
fp@1297: % Datagram injection
fp@1094: 
fp@1094: %------------------------------------------------------------------------------
fp@1094: 
fp@1298: \section{VoE Handlers}
fp@1298: \label{sec:api-voe}
fp@1298: 
fp@1298: During the configuration phase, the application can create handlers for the
fp@2515: VoE mailbox protocol described in \autoref{sec:voe}. One VoE handler always
fp@1298: belongs to a certain slave configuration, so the creation function is a method
fp@1298: of the slave configuration.
fp@1298: 
fp@1298: A VoE handler manages the VoE data and the datagram used to transmit and
fp@1298: receive VoE messages. Is contains the state machine necessary to transfer VoE
fp@1298: messages.
fp@1298: 
fp@1298: The VoE state machine can only process one operation at a time. As a result,
fp@1298: either a read or write operation may be issued at a time\footnote{If
fp@1298: simultaneous sending and receiving is desired, two VoE handlers can be created
fp@1298: for the slave configuration.}. After the operation is initiated, the handler
fp@1298: must be executed cyclically until it is finished. After that, the results of
fp@1298: the operation can be retrieved.
fp@1298: 
fp@1298: A VoE handler has an own datagram structure, that is marked for exchange after
fp@1298: each execution step. So the application can decide, how many handlers to
fp@1298: execute before sending the corresponding EtherCAT frame(s).
fp@1298: 
fp@1298: For more information about the use of VoE handlers see the documentation of
fp@1298: the application interface functions and the example applications provided in
fp@1298: the \textit{examples/} directory.
fp@1298: 
fp@1298: %------------------------------------------------------------------------------
fp@1298: 
fp@1294: \section{Concurrent Master Access}
fp@1085: \label{sec:concurr}
fp@1085: \index{Concurrency}
fp@1085: 
fp@1085: In some cases, one master is used by several instances, for example when an
fp@1289: application does cyclic process data exchange, and there are EoE-capable
fp@1289: slaves that require to exchange Ethernet data with the kernel (see
fp@2515: \autoref{sec:eoe}). For this reason, the master is a shared resource, and
fp@1289: access to it has to be sequentialized. This is usually done by locking with
fp@1085: semaphores, or other methods to protect critical sections.
fp@1085: 
fp@1085: The master itself can not provide locking mechanisms, because it has no chance
fp@1291: to know the appropriate kind of lock. For example if the application is in
fp@1291: kernelspace and uses RTAI functionality, ordinary kernel semaphores would not
fp@1291: be sufficient. For that, an important design decision was made: The
fp@1291: application that reserved a master must have the total control, therefore it
fp@1291: has to take responsibility for providing the appropriate locking mechanisms.
fp@1517: If another instance wants to access the master, it has to request the bus
fp@1517: access via callbacks, that have to be provided by the application. Moreover
fp@1517: the application can deny access to the master if it considers it to be awkward
fp@1517: at the moment.
fp@369: 
fp@369: \begin{figure}[htbp]
fp@369:   \centering
fp@1085:   \includegraphics[width=.6\textwidth]{images/master-locks}
fp@1295:   \caption{Concurrent Master Access}
fp@1085:   \label{fig:locks}
fp@369: \end{figure}
fp@369: 
fp@2515: \autoref{fig:locks} exemplary shows, how two processes share one master:
fp@1204: The application's cyclic task uses the master for process data exchange, while
fp@1204: the master-internal EoE process uses it to communicate with EoE-capable
fp@1517: slaves. Both have to access the bus from time to time, but the EoE process
fp@1517: does this by ``asking'' the application to do the bus access for it. In this
fp@1517: way, the application can use the appropriate locking mechanism to avoid
fp@1517: accessing the bus at the same time. See the application interface
fp@2515: documentation (\autoref{chap:api}) for how to use these callbacks.
fp@1517: 
fp@1517: %------------------------------------------------------------------------------
fp@1517: 
fp@1517: \section{Distributed Clocks}
fp@1517: \label{sec:dc}
fp@1517: \index{Distributed Clocks}
fp@1517: 
fp@1517: From version 1.5, the master supports EtherCAT's ``Distributed Clocks''
fp@1517: feature. It is possible to synchronize the slave clocks on the bus to the
fp@1517: ``reference clock'' (which is the local clock of the first slave with DC
fp@1517: support) and to synchronize the reference clock to the ``master clock'' (which
fp@1517: is the local clock of the master). All other clocks on the bus (after the
fp@2515: reference clock) are considered as ``slave clocks'' (see \autoref{fig:dc}).
fp@1517: 
fp@1517: \begin{figure}[htbp]
fp@1517:   \centering
fp@1517:   \includegraphics[width=.8\textwidth]{images/dc}
fp@1517:   \caption{Distributed Clocks}
fp@1517:   \label{fig:dc}
fp@1517: \end{figure}
fp@1517: 
fp@1517: \paragraph{Local Clocks} Any EtherCAT slave that supports DC has a local clock
fp@1517: register with nanosecond resolution. If the slave is powered, the clock starts
fp@1517: from zero, meaning that when slaves are powered on at different times, their
fp@1517: clocks will have different values. These ``offsets'' have to be compensated by
fp@1517: the distributed clocks mechanism. On the other hand, the clocks do not run
fp@1517: exactly with the same speed, since the used quarts units have a natural
fp@1517: frequency deviation. This deviation is usually very small, but over longer
fp@1517: periods, the error would accumulate and the difference between local clocks
fp@1517: would grow. This clock ``drift'' has also to be compensated by the DC
fp@1517: mechanism.
fp@1517: 
fp@1517: \paragraph{Application Time} The common time base for the bus has to be
fp@1517: provided by the application. This application time $t_\text{app}$ is used
fp@1517: 
fp@1517: \begin{enumerate}
fp@1517: \item to configure the slaves' clock offsets (see below),
fp@1517: \item to program the slave's start times for sync pulse generation (see
fp@1517: below).
fp@1517: \item to synchronize the reference clock to the master clock (optional).
fp@1517: \end{enumerate}
fp@1517: 
fp@1517: \paragraph{Offset Compensation} For the offset compensation, each slave
fp@1517: provides a ``System Time Offset'' register $t_\text{off}$, that is added to
fp@1517: the internal clock value $t_\text{int}$ to get the ``System Time''
fp@1517: $t_\text{sys}$:
fp@1517: 
fp@1517: \begin{eqnarray}
fp@1517: t_\text{sys} & = & t_\text{int} + t_\text{off} \\
fp@1517: \Rightarrow t_\text{int} & = & t_\text{sys} - t_\text{off} \nonumber
fp@1517: \end{eqnarray}
fp@1517: 
fp@1517: The master reads the values of both registers to calculate a new system time
fp@1517: offset in a way, that the resulting system time shall match the master's
fp@1517: application time $t_\text{app}$:
fp@1517: 
fp@1517: \begin{eqnarray}
fp@1517: t_\text{sys} & \stackrel{!}{=} & t_\text{app} \\
fp@1517: \Rightarrow t_\text{int} + t_\text{off} & \stackrel{!}{=} & t_\text{app} \nonumber \\
fp@1517: \Rightarrow t_\text{off} & = & t_\text{app} - t_\text{int} \nonumber \\
fp@1517: \Rightarrow t_\text{off} & = & t_\text{app} - (t_\text{sys} - t_\text{off}) \nonumber \\
fp@1517: \Rightarrow t_\text{off} & = & t_\text{app} - t_\text{sys} + t_\text{off}
fp@1517: \end{eqnarray}
fp@1517: 
fp@1517: The small time offset error resulting from the different times of reading and
fp@1517: writing the registers will be compensated by the drift compensation.
fp@1517: 
fp@1517: \paragraph{Drift Compensation} The drift compensation is possible due to a
fp@1517: special mechanism in each DC-capable slave: A write operation to the ``System
fp@1517: time'' register will cause the internal time control loop to compare the
fp@1517: written time (minus the programmed transmission delay, see below) to the
fp@1517: current system time. The calculated time error will be used as an input to the
fp@1517: time controller, that will tune the local clock speed to be a little faster or
fp@1517: slower\footnote{The local slave clock will be incremented either with
fp@1517: \unit{9}{\nano\second}, \unit{10}{\nano\second} or \unit{11}{\nano\second}
fp@1517: every \unit{10}{\nano\second}.}, according to the sign of the error.
fp@1517: 
fp@1517: \paragraph{Transmission Delays} The Ethernet frame needs a small amount of
fp@1517: time to get from slave to slave. The resulting transmission delay times
fp@1517: accumulate on the bus and can reach microsecond magnitude and thus have to be
fp@1517: considered during the drift compensation. EtherCAT slaves supporting DC
fp@1517: provide a mechanism to measure the transmission delays: For each of the four
fp@1517: slave ports there is a receive time register. A write operation to the receive
fp@1517: time register of port 0 starts the measuring and the current system time is
fp@1517: latched and stored in a receive time register once the frame is received on
fp@1517: the corresponding port. The master can read out the relative receive times,
fp@1517: then calculate time delays between the slaves (using its knowledge of the bus
fp@1517: topology), and finally calculate the time delays from the reference clock to
fp@1517: each slave. These values are programmed into the slaves' transmission delay
fp@2421: registers. In this way, the drift compensation can reach nanosecond synchrony.
fp@1517: 
fp@1517: \paragraph{Checking Synchrony} DC-capable slaves provide the 32-bit ``System
fp@1517: time difference'' register at address \lstinline+0x092c+, where the system
fp@1517: time difference of the last drift compensation is stored in nanosecond
fp@1517: resolution and in sign-and-magnitude coding\footnote{This allows
fp@1517: broadcast-reading all system time difference registers on the bus to get an
fp@1517: upper approximation}. To check for bus synchrony, the system time difference
fp@1517: registers can also be cyclically read via the command-line-tool (see
fp@2515: \autoref{sec:regaccess}):
fp@1517: 
fp@1517: \begin{lstlisting}
fp@1917: $ `\textbf{watch -n0 "ethercat reg\_read -p4 -tsm32 0x92c"}`
fp@1517: \end{lstlisting}
fp@1517: 
fp@1517: \paragraph{Sync Signals} Synchronous clocks are only the prerequisite for
fp@1517: synchronous events on the bus. Each slave with DC support provides two ``sync
fp@1517: signals'', that can be programmed to create events, that will for example
fp@1517: cause the slave application to latch its inputs on a certain time. A sync
fp@1517: event can either be generated once or cyclically, depending on what makes
fp@1517: sense for the slave application. Programming the sync signals is a matter of
fp@1517: setting the so-called ``AssignActivate'' word and the sync signals' cycle- and
fp@1517: shift times. The AssignActivate word is slave-specific and has to be taken
fp@1517: from the XML slave description (\lstinline+Device+ $\rightarrow$
fp@1517: \lstinline+Dc+), where also typical sync signal configurations ``OpModes'' can
fp@1517: be found.
fp@1085: 
fp@1085: %------------------------------------------------------------------------------
fp@1085: 
fp@1202: \chapter{Ethernet Devices}
fp@1085: \label{sec:devices}
fp@1085: 
fp@1203: The EtherCAT protocol is based on the Ethernet standard, so a master relies on
fp@1203: standard Ethernet hardware to communicate with the bus.
fp@1085: 
fp@1085: The term \textit{device} is used as a synonym for Ethernet network interface
fp@1588: hardware.
fp@1588: 
fp@1588: \paragraph{Native Ethernet Device Drivers} There are native device driver
fp@2515: modules (see \autoref{sec:native-drivers}) that handle Ethernet hardware,
fp@1588: which a master can use to connect to an EtherCAT bus. They offer their
fp@1588: Ethernet hardware to the master module via the device interface (see
fp@2515: \autoref{sec:ecdev}) and must be capable to prepare Ethernet devices either
fp@1588: for EtherCAT (realtime) operation or for ``normal'' operation using the
fp@1588: kernel's network stack. The advantage of this approach is that the master can
fp@1588: operate nearly directly on the hardware, which allows a high performance. The
fp@1588: disadvantage is, that there has to be an EtherCAT-capable version of the
fp@1588: original Ethernet driver.
fp@1588: 
fp@1588: \paragraph{Generic Ethernet Device Driver} From master version 1.5, there is a
fp@2515: generic Ethernet device driver module (see \autoref{sec:generic-driver}),
fp@1588: that uses the lower layers of the network stack to connect to the hardware.
fp@1588: The advantage is, that arbitrary Ethernet hardware can be used for EtherCAT
fp@1588: operation, independently of the actual hardware driver (so all Linux Ethernet
fp@1588: drivers are supported without modifications). The disadvantage is, that this
fp@1588: approach does not support realtime extensions like RTAI, because the Linux
fp@1588: network stack is addressed. Moreover the performance is a little worse than
fp@1588: the native approach, because the Ethernet frame data have to traverse the
fp@1588: network stack.
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@1085: \section{Network Driver Basics}
fp@369: \label{sec:networkdrivers}
fp@369: \index{Network drivers}
fp@369: 
fp@369: EtherCAT relies on Ethernet hardware and the master needs a physical
fp@369: Ethernet device to communicate with the bus. Therefore it is necessary
fp@369: to understand how Linux handles network devices and their drivers,
fp@369: respectively.
fp@369: 
fp@1588: \paragraph{Tasks of a Network Driver} Network device drivers usually handle
fp@1588: the lower two layers of the OSI model, that is the physical layer and the
fp@1588: data-link layer. A network device itself natively handles the physical layer
fp@1588: issues: It represents the hardware to connect to the medium and to send and
fp@1588: receive data in the way, the physical layer protocol describes. The network
fp@1588: device driver is responsible for getting data from the kernel's networking
fp@1588: stack and forwarding it to the hardware, that does the physical transmission.
fp@1588: If data is received by the hardware respectively, the driver is notified
fp@1588: (usually by means of an interrupt) and has to read the data from the hardware
fp@1588: memory and forward it to the network stack. There are a few more tasks, a
fp@1588: network device driver has to handle, including queue control, statistics and
fp@1588: device dependent features.
fp@1588: 
fp@1588: \paragraph{Driver Startup} Usually, a driver searches for compatible devices
fp@1588: on module loading.  For PCI drivers, this is done by scanning the PCI bus and
fp@1588: checking for known device IDs. If a device is found, data structures are
fp@1588: allocated and the device is taken into operation.
fp@1588: 
fp@1588: \paragraph{Interrupt Operation}\index{Interrupt} A network device usually
fp@1588: provides a hardware interrupt that is used to notify the driver of received
fp@1588: frames and success of transmission, or errors, respectively. The driver has to
fp@1588: register an interrupt service routine
fp@1588: (ISR\index{ISR}\nomenclature{ISR}{Interrupt Service Routine}), that is
fp@1588: executed each time, the hardware signals such an event. If the interrupt was
fp@1588: thrown by the own device (multiple devices can share one hardware interrupt),
fp@1588: the reason for the interrupt has to be determined by reading the device's
fp@1588: interrupt register. For example, if the flag for received frames is set, frame
fp@1588: data has to be copied from hardware to kernel memory and passed to the network
fp@1588: stack.
fp@1588: 
fp@1588: \paragraph{The \lstinline+net_device+ Structure}\index{net\_device} The driver
fp@1588: registers a \lstinline+net_device+ structure for each device to communicate
fp@1588: with the network stack and to create a ``network interface''. In case of an
fp@1588: Ethernet driver, this interface appears as \textit{ethX}, where X is a number
fp@1588: assigned by the kernel on registration. The \lstinline+net_device+ structure
fp@1588: receives events (either from userspace or from the network stack) via several
fp@1588: callbacks, which have to be set before registration. Not every callback is
fp@1588: mandatory, but for reasonable operation the ones below are needed in any case:
fp@1202: 
fp@1202: \newsavebox\boxopen
fp@1202: \sbox\boxopen{\lstinline+open()+}
fp@1202: \newsavebox\boxstop
fp@1202: \sbox\boxstop{\lstinline+stop()+}
fp@1202: \newsavebox\boxxmit
fp@1202: \sbox\boxxmit{\lstinline+hard_start_xmit()+}
fp@1202: \newsavebox\boxstats
fp@1202: \sbox\boxstats{\lstinline+get_stats()+}
fp@369: 
fp@369: \begin{description}
fp@1095: 
fp@1202: \item[\usebox\boxopen] This function is called when network communication has
fp@1275: to be started, for example after a command \lstinline+ip link set ethX up+
fp@1275: from userspace. Frame reception has to be enabled by the driver.
fp@1202: 
fp@1297: \item[\usebox\boxstop] The purpose of this callback is to ``close'' the
fp@1306: device, i.\,e.\ make the hardware stop receiving frames.
fp@1202: 
fp@1202: \item[\usebox\boxxmit] This function is called for each frame that has to be
fp@1202: transmitted. The network stack passes the frame as a pointer to an
fp@1202: \lstinline+sk_buff+ structure (``socket buffer''\index{Socket buffer}, see
fp@1095: below), which has to be freed after sending.
fp@1095: 
fp@1202: \item[\usebox\boxstats] This call has to return a pointer to the device's
fp@1202: \lstinline+net_device_stats+ structure, which permanently has to be filled with
fp@1095: frame statistics. This means, that every time a frame is received, sent, or an
fp@1095: error happened, the appropriate counter in this structure has to be increased.
fp@1095: 
fp@1095: \end{description}
fp@1095: 
fp@1095: The actual registration is done with the \lstinline+register_netdev()+ call,
fp@1095: unregistering is done with \lstinline+unregister_netdev()+.
fp@369: 
fp@1588: \paragraph{The \lstinline+netif+ Interface}\index{netif} All other
fp@1588: communication in the direction interface $\to$ network stack is done via the
fp@1588: \lstinline+netif_*()+ calls. For example, on successful device opening, the
fp@1588: network stack has to be notified, that it can now pass frames to the
fp@1202: interface. This is done by calling \lstinline+netif_start_queue()+. After this
fp@1202: call, the \lstinline+hard_start_xmit()+ callback can be called by the network
fp@1588: stack. Furthermore a network driver usually manages a frame transmission
fp@1588: queue.  If this gets filled up, the network stack has to be told to stop
fp@1588: passing further frames for a while. This happens with a call to
fp@1202: \lstinline+netif_stop_queue()+. If some frames have been sent, and there is
fp@1095: enough space again to queue new frames, this can be notified with
fp@1095: \lstinline+netif_wake_queue()+. Another important call is
fp@1095: \lstinline+netif_receive_skb()+\footnote{This function is part of the NAPI
fp@1095: (``New API''), that replaces the kernel 2.4 technique for interfacing to the
fp@1095: network stack (with \lstinline+netif_rx()+). NAPI is a technique to improve
fp@1095: network performance on Linux. Read more in
fp@1095: \url{http://www.cyberus.ca/~hadi/usenix-paper.tgz}.}: It passes a frame to the
fp@1095: network stack, that was just received by the device. Frame data has to be
fp@1202: included in a so-called ``socket buffer'' for that (see below).
fp@369: 
fp@1588: \paragraph{Socket Buffers}\index{Socket buffer} Socket buffers are the basic
fp@1588: data type for the whole network stack. They serve as containers for network
fp@1588: data and are able to quickly add data headers and footers, or strip them off
fp@1588: again. Therefore a socket buffer consists of an allocated buffer and several
fp@1588: pointers that mark beginning of the buffer (\lstinline+head+), beginning of
fp@1588: data (\lstinline+data+), end of data (\lstinline+tail+) and end of buffer
fp@1588: (\lstinline+end+). In addition, a socket buffer holds network header
fp@1588: information and (in case of received data) a pointer to the
fp@1588: \lstinline+net_device+, it was received on. There exist functions that create
fp@1588: a socket buffer (\lstinline+dev_alloc_skb()+), add data either from front
fp@1588: (\lstinline+skb_push()+) or back (\lstinline+skb_put()+), remove data from
fp@1588: front (\lstinline+skb_pull()+) or back (\lstinline+skb_trim()+), or delete the
fp@1588: buffer (\lstinline+kfree_skb()+).  A socket buffer is passed from layer to
fp@1588: layer, and is freed by the layer that uses it the last time. In case of
fp@1588: sending, freeing has to be done by the network driver.
fp@1588: 
fp@1588: %------------------------------------------------------------------------------
fp@1588: 
fp@1588: \section{Native EtherCAT Device Drivers}
fp@1588: \label{sec:native-drivers}
fp@1588: 
fp@1588: There are a few requirements, that applies to Ethernet hardware when used with
fp@1588: a native Ethernet driver with EtherCAT functionality.
fp@1588: 
fp@1588: \paragraph{Dedicated Hardware} For performance and realtime purposes, the
fp@1588: EtherCAT master needs direct and exclusive access to the Ethernet hardware.
fp@1588: This implies that the network device must not be connected to the kernel's
fp@1588: network stack as usual, because the kernel would try to use it as an ordinary
fp@1588: Ethernet device.
fp@1588: 
fp@1588: \paragraph{Interrupt-less Operation}\index{Interrupt} EtherCAT frames travel
fp@1588: through the logical EtherCAT ring and are then sent back to the master.
fp@1588: Communication is highly deterministic: A frame is sent and will be received
fp@1588: again after a constant time, so there is no need to notify the driver about
fp@1588: frame reception: The master can instead query the hardware for received
fp@1588: frames, if it expects them to be already received.
fp@1095: 
fp@2515: \autoref{fig:interrupt} shows two workflows for cyclic frame transmission
fp@1095: and reception with and without interrupts.
fp@369: 
fp@369: \begin{figure}[htbp]
fp@369:   \centering
fp@1202:   \includegraphics[width=.9\textwidth]{images/interrupt}
fp@369:   \caption{Interrupt Operation versus Interrupt-less Operation}
fp@369:   \label{fig:interrupt}
fp@369: \end{figure}
fp@369: 
fp@1095: In the left workflow ``Interrupt Operation'', the data from the last cycle is
fp@1095: first processed and a new frame is assembled with new datagrams, which is then
fp@1095: sent.  The cyclic work is done for now.  Later, when the frame is received
fp@1095: again by the hardware, an interrupt is triggered and the ISR is executed. The
fp@1095: ISR will fetch the frame data from the hardware and initiate the frame
fp@1095: dissection: The datagrams will be processed, so that the data is ready for
fp@1095: processing in the next cycle.
fp@1095: 
fp@1095: In the right workflow ``Interrupt-less Operation'', there is no hardware
fp@1095: interrupt enabled.  Instead, the hardware will be polled by the master by
fp@1095: executing the ISR. If the frame has been received in the meantime, it will be
fp@1095: dissected. The situation is now the same as at the beginning of the left
fp@1095: workflow: The received data is processed and a new frame is assembled and
fp@1095: sent. There is nothing to do for the rest of the cycle.
fp@1095: 
fp@1202: The interrupt-less operation is desirable, because hardware interrupts are not
fp@1202: conducive in improving the driver's realtime behaviour: Their indeterministic
fp@1202: incidences contribute to increasing the jitter. Besides, if a realtime
fp@1202: extension (like RTAI) is used, some additional effort would have to be made to
fp@1202: prioritize interrupts.
fp@369: 
fp@1588: \paragraph{Ethernet and EtherCAT Devices} Another issue lies in the way Linux
fp@1588: handles devices of the same type.  For example, a
fp@1588: PCI\nomenclature{PCI}{Peripheral Component Interconnect, Computer Bus} driver
fp@1588: scans the PCI bus for devices it can handle. Then it registers itself as the
fp@1588: responsible driver for all of the devices found. The problem is, that an
fp@1588: unmodified driver can not be told to ignore a device because it will be used
fp@1588: for EtherCAT later. There must be a way to handle multiple devices of the same
fp@1588: type, where one is reserved for EtherCAT, while the other is treated as an
fp@1588: ordinary Ethernet device.
fp@369: 
fp@1095: For all this reasons, the author decided that the only acceptable solution is
fp@1095: to modify standard Ethernet drivers in a way that they keep their normal
fp@1095: functionality, but gain the ability to treat one or more of the devices as
fp@1095: EtherCAT-capable.
fp@369: 
fp@369: Below are the advantages of this solution:
fp@369: 
fp@369: \begin{itemize}
fp@369: \item No need to tell the standard drivers to ignore certain devices.
fp@369: \item One networking driver for EtherCAT and non-EtherCAT devices.
fp@369: \item No need to implement a network driver from scratch and running
fp@369:   into issues, the former developers already solved.
fp@369: \end{itemize}
fp@369: 
fp@369: The chosen approach has the following disadvantages:
fp@369: 
fp@369: \begin{itemize}
fp@369: \item The modified driver gets more complicated, as it must handle
fp@369:   EtherCAT and non-EtherCAT devices.
fp@369: \item Many additional case differentiations in the driver code.
fp@1085: \item Changes and bug fixes on the standard drivers have to be ported
fp@369:   to the Ether\-CAT-capable versions from time to time.
fp@369: \end{itemize}
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@1588: \section{Generic EtherCAT Device Driver}
fp@1588: \label{sec:generic-driver}
fp@1588: 
fp@1588: Since there are approaches to enable the complete Linux kernel for realtime
fp@1588: operation \cite{rt-preempt}, it is possible to operate without native
fp@1588: implementations of EtherCAT-capable Ethernet device drivers and use the Linux
fp@2515: network stack instead. \autoref{fig:arch} shows the ``Generic Ethernet Driver
fp@1588: Module'', that connects to local Ethernet devices via the network stack. The
fp@1588: kernel module is named \lstinline+ec_generic+ and can be loaded after the
fp@1588: master module like a native EtherCAT-capable Ethernet driver.
fp@1588: 
fp@1588: The generic device driver scans the network stack for interfaces, that have
fp@1588: been registered by Ethernet device drivers. It offers all possible devices to
fp@1588: the EtherCAT master. If the master accepts a device, the generic driver
fp@1588: creates a packet socket (see \lstinline+man 7 packet+) with
fp@1588: \lstinline+socket_type+ set to \lstinline+SOCK_RAW+, bound to that device. All
fp@2515: functions of the device interface (see \autoref{sec:ecdev}) will then operate
fp@1588: on that socket.
fp@1588: 
fp@1588: Below are the advantages of this solution:
fp@1588: 
fp@1588: \begin{itemize}
fp@1588: \item Any Ethernet hardware, that is covered by a Linux Ethernet driver can be
fp@1588: used for EtherCAT.
fp@1588: \item No modifications have to be made to the actual Ethernet drivers.
fp@1588: \end{itemize}
fp@1588: 
fp@1588: The generic approach has the following disadvantages:
fp@1588: 
fp@1588: \begin{itemize}
fp@1588: \item The performance is a little worse than the native approach, because the
fp@1588: frame data have to traverse the lower layers of the network stack.
fp@1588: \item It is not possible to use in-kernel realtime extensions like RTAI with
fp@1588: the generic driver, because the network stack code uses dynamic memory
fp@1588: allocations and other things, that could cause the system to freeze in
fp@1588: realtime context.
fp@1588: \end{itemize}
fp@1588: 
fp@2535: \paragraph{Device Activation} In order to send and receive frames through a
fp@2535: socket, the Ethernet device linked to that socket has to be activated,
fp@2535: otherwise all frames will be rejected. Activation has to take place before the
fp@2535: master module is loaded and can happen in several ways:
fp@2535: 
fp@2535: \begin{itemize}
fp@2535: 
fp@2535: \item Ad-hoc, using the command \lstinline+ip link set dev ethX up+ (or the
fp@2535: older \lstinline+ifconfig ethX up+),
fp@2535: 
fp@2535: \item Configured, depending on the distribution, for example using
fp@2535: \lstinline+ifcfg+ files (\lstinline+/etc/sysconfig/network/ifcfg-ethX+) in
fp@2535: openSUSE and others. This is the better choice, if the EtherCAT master shall
fp@2535: start at system boot time. Since the Ethernet device shall only be activated,
fp@2535: but no IP address etc.\ shall be assigned, it is enough to use
fp@2535: \lstinline+STARTMODE=auto+ as configuration.
fp@2535: 
fp@2535: \end{itemize}
fp@2535: 
fp@1588: %------------------------------------------------------------------------------
fp@1588: 
fp@1588: \section{Providing Ethernet Devices}
fp@1588: \label{sec:providing-devices}
fp@1588: 
fp@1588: After loading the master module, additional module(s) have to be loaded to
fp@2515: offer devices to the master(s) (see \autoref{sec:ecdev}). The master module
fp@1588: knows the devices to choose from the module parameters (see
fp@2515: \autoref{sec:mastermod}). If the init script is used to start the master, the
fp@1588: drivers and devices to use can be specified in the sysconfig file (see
fp@2515: \autoref{sec:sysconfig}).
fp@1588: 
fp@1588: Modules offering Ethernet devices can be
fp@1588: 
fp@1588: \begin{itemize}
fp@1588: \item native EtherCAT-capable network driver modules (see
fp@2515: \autoref{sec:native-drivers}) or
fp@1588: \item the generic EtherCAT device driver module (see
fp@2515: \autoref{sec:generic-driver}).
fp@1588: \end{itemize}
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@2515: \section{Redundancy}
fp@2515: \label{sec:redundancy}
fp@2515: \index{Redundancy}
fp@2515: 
fp@2515: Redundant bus operation means, that there is more than one Ethernet connection
fp@2515: from the master to the slaves. Process data exchange datagrams are sent out on
fp@2515: every master link, so that the exchange is still complete, even if the bus is
fp@2515: disconnected somewhere in between.
fp@2515: 
fp@2515: Prerequisite for fully redundant bus operation is, that every slave can be
fp@2515: reached by at least one master link. In this case a single connection failure
fp@2515: (i.\,e.~cable break) will never lead to incomplete process data. Double-faults
fp@2515: can not be handled with two Ethernet devices.
fp@2515: 
fp@2515: Redundancy is configured with the \lstinline+--with-devices+ switch at
fp@2515: configure time (see \autoref{sec:installation}) and using the
fp@2515: \lstinline+backup_devices+ parameter of the \lstinline+ec_master+ kernel
fp@2515: module (see \autoref{sec:mastermod}) or the appropriate variable
fp@2515: \lstinline+MASTERx_BACKUP+ in the (sys-)config file (see
fp@2515: \autoref{sec:sysconfig}).
fp@2515: 
fp@2515: Bus scanning is done after a topology change on any Ethernet link. The
fp@2515: application interface (see \autoref{chap:api}) and the command-line tool (see
fp@2515: \autoref{sec:tool}) both have methods to query the status of the redundant
fp@2515: operation.
fp@2515: 
fp@2515: %------------------------------------------------------------------------------
fp@2515: 
fp@1203: \section{EtherCAT Device Interface}
fp@369: \label{sec:ecdev}
fp@369: \index{Device interface}
fp@369: 
fp@369: An anticipation to the section about the master module
fp@2515: (\autoref{sec:mastermod}) has to be made in order to understand the way, a
fp@1289: network device driver module can connect a device to a specific EtherCAT
fp@1289: master.
fp@369: 
fp@1202: The master module provides a ``device interface'' for network device drivers.
fp@1202: To use this interface, a network device driver module must include the header
fp@1202: \textit{devices/ecdev.h}\nomenclature{ecdev}{EtherCAT Device}, coming with the
fp@1202: EtherCAT master code. This header offers a function interface for EtherCAT
fp@1202: devices. All functions of the device interface are named with the prefix
fp@1202: \lstinline+ecdev+.
fp@1202: 
fp@1289: The documentation of the device interface can be found in the header file or
fp@1289: in the appropriate module of the interface documentation (see
fp@2515: \autoref{sec:gendoc} for generation instructions).
fp@1202: 
fp@1296: % TODO general description of the device interface
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@1588: \section{Patching Native Network Drivers}
fp@369: \label{sec:patching}
fp@369: \index{Network drivers}
fp@369: 
fp@1202: This section will describe, how to make a standard Ethernet driver
fp@1588: EtherCAT-capable, using the native approach (see
fp@2515: \autoref{sec:native-drivers}). Unfortunately, there is no standard procedure
fp@1588: to enable an Ethernet driver for use with the EtherCAT master, but there are a
fp@1588: few common techniques.
fp@369: 
fp@369: \begin{enumerate}
fp@1202: 
fp@1202: \item A first simple rule is, that \lstinline+netif_*()+ calls must be avoided
fp@1202: for all EtherCAT devices. As mentioned before, EtherCAT devices have no
fp@1202: connection to the network stack, and therefore must not call its interface
fp@1202: functions.
fp@1202: 
fp@1202: \item Another important thing is, that EtherCAT devices should be operated
fp@1202: without interrupts. So any calls of registering interrupt handlers and enabling
fp@1202: interrupts at hardware level must be avoided, too.
fp@1202: 
fp@1202: \item The master does not use a new socket buffer for each send operation:
fp@1202: Instead there is a fix one allocated on master initialization. This socket
fp@1202: buffer is filled with an EtherCAT frame with every send operation and passed to
fp@1202: the \lstinline+hard_start_xmit()+ callback. For that it is necessary, that the
fp@1202: socket buffer is not be freed by the network driver as usual.
fp@1202: 
fp@369: \end{enumerate}
fp@369: 
fp@1202: An Ethernet driver usually handles several Ethernet devices, each described by
fp@1202: a \lstinline+net_device+ structure with a \lstinline+priv_data+ field to
fp@1202: attach driver-dependent data to the structure. To distinguish between normal
fp@1202: Ethernet devices and the ones used by EtherCAT masters, the private data
fp@1202: structure used by the driver could be extended by a pointer, that points to an
fp@1202: \lstinline+ec_device_t+ object returned by \lstinline+ecdev_offer()+ (see
fp@2515: \autoref{sec:ecdev}) if the device is used by a master and otherwise is zero.
fp@1202: 
fp@1202: The RealTek RTL-8139 Fast Ethernet driver is a ``simple'' Ethernet driver and
fp@1202: can be taken as an example to patch new drivers. The interesting sections can
fp@1202: be found by searching the string ``ecdev" in the file
fp@1202: \textit{devices/8139too-2.6.24-ethercat.c}.
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@1085: \chapter{State Machines}
fp@369: \label{sec:fsm}
fp@369: \index{FSM}
fp@369: 
fp@1085: Many parts of the EtherCAT master are implemented as \textit{finite state
fp@1085: machines} (FSMs\nomenclature{FSM}{Finite State Machine}). Though this leads
fp@1085: to a higher grade of complexity in some aspects, is opens many new
fp@1085: possibilities.
fp@369: 
fp@369: The below short code example exemplary shows how to read all slave
fp@369: states and moreover illustrates the restrictions of ``sequential''
fp@369: coding:
fp@369: 
fp@1085: \begin{lstlisting}[gobble=2,language=C,numbers=left]
fp@369:   ec_datagram_brd(datagram, 0x0130, 2); // prepare datagram
fp@369:   if (ec_master_simple_io(master, datagram)) return -1;
fp@369:   slave_states = EC_READ_U8(datagram->data); // process datagram
fp@369: \end{lstlisting}
fp@369: 
fp@1085: The \textit{ec\_master\_simple\_io()} function provides a simple interface for
fp@1085: synchronously sending a single datagram and receiving the result\footnote{For
fp@1085: all communication issues have been meanwhile sourced out into state machines,
fp@1085: the function is deprecated and stopped existing. Nevertheless it is adequate
fp@1085: for showing it's own restrictions.}. Internally, it queues the specified
fp@1085: datagram, invokes the \textit{ec\_master\_send\_datagrams()} function to send
fp@1085: a frame with the queued datagram and then waits actively for its reception.
fp@369: 
fp@1289: This sequential approach is very simple, reflecting in only three lines of
fp@1289: code. The disadvantage is, that the master is blocked for the time it waits
fp@1289: for datagram reception. There is no difficulty when only one instance is using
fp@1289: the master, but if more instances want to (synchronously\footnote{At this
fp@1289: time, synchronous master access will be adequate to show the advantages of an
fp@2515: FSM. The asynchronous approach will be discussed in \autoref{sec:eoe}}) use
fp@1289: the master, it is inevitable to think about an alternative to the sequential
fp@1289: model.
fp@369: 
fp@369: Master access has to be sequentialized for more than one instance
fp@369: wanting to send and receive datagrams synchronously. With the present
fp@369: approach, this would result in having one phase of active waiting for
fp@369: each instance, which would be non-acceptable especially in realtime
fp@369: circumstances, because of the huge time overhead.
fp@369: 
fp@369: A possible solution is, that all instances would be executed
fp@369: sequentially to queue their datagrams, then give the control to the
fp@369: next instance instead of waiting for the datagram reception. Finally,
fp@369: bus IO is done by a higher instance, which means that all queued
fp@369: datagrams are sent and received. The next step is to execute all
fp@369: instances again, which then process their received datagrams and issue
fp@369: new ones.
fp@369: 
fp@369: This approach results in all instances having to retain their state,
fp@369: when giving the control back to the higher instance. It is quite
fp@369: obvious to use a \textit{finite state machine} model in this case.
fp@2515: \autoref{sec:fsmtheory} will introduce some of the theory used,
fp@369: while the listings below show the basic approach by coding the example
fp@369: from above as a state machine:
fp@369: 
fp@1085: \begin{lstlisting}[gobble=2,language=C,numbers=left]
fp@369:   // state 1
fp@369:   ec_datagram_brd(datagram, 0x0130, 2); // prepare datagram
fp@369:   ec_master_queue(master, datagram); // queue datagram
fp@369:   next_state = state_2;
fp@369:   // state processing finished
fp@369: \end{lstlisting}
fp@369: 
fp@369: After all instances executed their current state and queued their
fp@369: datagrams, these are sent and received. Then the respective next
fp@369: states are executed:
fp@369: 
fp@1085: \begin{lstlisting}[gobble=2,language=C,numbers=left]
fp@369:   // state 2
fp@369:   if (datagram->state != EC_DGRAM_STATE_RECEIVED) {
fp@369:           next_state = state_error;
fp@369:           return; // state processing finished
fp@369:   }
fp@369:   slave_states = EC_READ_U8(datagram->data); // process datagram
fp@369:   // state processing finished.
fp@369: \end{lstlisting}
fp@369: 
fp@2515: See \autoref{sec:statemodel} for an introduction to the state machine
fp@1289: programming concept used in the master code.
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@1085: \section{State Machine Theory}
fp@369: \label{sec:fsmtheory}
fp@369: \index{FSM!Theory}
fp@369: 
fp@369: A finite state machine \cite{automata} is a model of behavior with
fp@369: inputs and outputs, where the outputs not only depend on the inputs,
fp@369: but the history of inputs. The mathematical definition of a finite
fp@369: state machine (or finite automaton) is a six-tuple $(\Sigma, \Gamma,
fp@369: S, s_0, \delta, \omega)$, with
fp@369: 
fp@369: \begin{itemize}
fp@369: \item the input alphabet $\Sigma$, with $\Sigma \neq
fp@369:   \emptyset$, containing all input symbols,
fp@369: \item the output alphabet $\Gamma$, with $\Gamma \neq
fp@369:   \emptyset$, containing all output symbols,
fp@369: \item the set of states $S$, with $S \neq \emptyset$,
fp@369: \item the set of initial states $s_0$ with $s_0 \subseteq S, s_0 \neq
fp@369:   \emptyset$
fp@369: \item the transition function $\delta: S \times \Sigma \rightarrow S
fp@369:   \times \Gamma$
fp@369: \item the output function $\omega$.
fp@369: \end{itemize}
fp@369: 
fp@369: The state transition function $\delta$ is often specified by a
fp@369: \textit{state transition table}, or by a \textit{state transition
fp@369:   diagram}. The transition table offers a matrix view of the state
fp@2515: machine behavior (see \autoref{tab:statetrans}). The matrix rows
fp@369: correspond to the states ($S = \{s_0, s_1, s_2\}$) and the columns
fp@369: correspond to the input symbols ($\Gamma = \{a, b, \varepsilon\}$).
fp@369: The table contents in a certain row $i$ and column $j$ then represent
fp@369: the next state (and possibly the output) for the case, that a certain
fp@369: input symbol $\sigma_j$ is read in the state $s_i$.
fp@369: 
fp@369: \begin{table}[htbp]
fp@369:   \caption{A typical state transition table}
fp@369:   \label{tab:statetrans}
fp@369:   \vspace{2mm}
fp@369:   \centering
fp@369:   \begin{tabular}{l|ccc}
fp@369:     & $a$ & $b$ & $\varepsilon$\\ \hline
fp@369:     $s_0$ & $s_1$ & $s_1$ & $s_2$\\
fp@369:     $s_1$ & $s_2$ & $s_1$ & $s_0$\\
fp@369:     $s_2$ & $s_0$ & $s_0$ & $s_0$\\ \hline
fp@369:   \end{tabular}
fp@369: \end{table}
fp@369: 
fp@369: The state diagram for the same example looks like the one in
fp@2515: \autoref{fig:statetrans}. The states are represented as circles or
fp@369: ellipses and the transitions are drawn as arrows between them. Close
fp@369: to a transition arrow can be the condition that must be fulfilled to
fp@369: allow the transition. The initial state is marked by a filled black
fp@369: circle with an arrow pointing to the respective state.
fp@369: 
fp@369: \begin{figure}[htbp]
fp@369:   \centering
fp@369:   \includegraphics[width=.5\textwidth]{images/statetrans}
fp@369:   \caption{A typical state transition diagram}
fp@369:   \label{fig:statetrans}
fp@369: \end{figure}
fp@369: 
fp@369: \paragraph{Deterministic and non-deterministic state machines}
fp@369: 
fp@369: A state machine can be deterministic, meaning that for one state and
fp@369: input, there is one (and only one) following state. In this case, the
fp@369: state machine has exactly one starting state. Non-deterministic state
fp@369: machines can have more than one transitions for a single state-input
fp@369: combination. There is a set of starting states in the latter case.
fp@369: 
fp@369: \paragraph{Moore and Mealy machines}
fp@369: 
fp@369: There is a distinction between so-called \textit{Moore machines}, and
fp@369: \textit{Mealy machines}. Mathematically spoken, the distinction lies
fp@369: in the output function $\omega$: If it only depends on the current
fp@369: state ($\omega: S \rightarrow \Gamma$), the machine corresponds to the
fp@369: ``Moore Model''. Otherwise, if $\omega$ is a function of a state and
fp@369: the input alphabet ($\omega: S \times \Sigma \rightarrow \Gamma$) the
fp@369: state machine corresponds to the ``Mealy model''. Mealy machines are
fp@369: the more practical solution in most cases, because their design allows
fp@369: machines with a minimum number of states. In practice, a mixture of
fp@369: both models is often used.
fp@369: 
fp@369: \paragraph{Misunderstandings about state machines}
fp@369: 
fp@1085: There is a phenomenon called ``state explosion'', that is often taken as a
fp@1085: counter-argument against general use of state machines in complex environments.
fp@1085: It has to be mentioned, that this point is misleading~\cite{fsmmis}. State
fp@1085: explosions happen usually as a result of a bad state machine design: Common
fp@1085: mistakes are storing the present values of all inputs in a state, or not
fp@1085: dividing a complex state machine into simpler sub state machines. The EtherCAT
fp@1085: master uses several state machines, that are executed hierarchically and so
fp@1085: serve as sub state machines. These are also described below.
fp@1085: 
fp@1085: %------------------------------------------------------------------------------
fp@1085: 
fp@1085: \section{The Master's State Model}
fp@369: \label{sec:statemodel}
fp@369: 
fp@369: This section will introduce the techniques used in the master to
fp@369: implement state machines.
fp@369: 
fp@369: \paragraph{State Machine Programming}
fp@369: 
fp@369: There are certain ways to implement a state machine in \textit{C}
fp@369: code. An obvious way is to implement the different states and actions
fp@369: by one big case differentiation:
fp@369: 
fp@1085: \begin{lstlisting}[gobble=2,language=C,numbers=left]
fp@369:   enum {STATE_1, STATE_2, STATE_3};
fp@369:   int state = STATE_1;
fp@369: 
fp@369:   void state_machine_run(void *priv_data) {
fp@369:           switch (state) {
fp@369:                   case STATE_1:
fp@369:                           action_1();
fp@369:                           state = STATE_2;
fp@369:                           break;
fp@369:                   case STATE_2:
fp@369:                           action_2()
fp@369:                           if (some_condition) state = STATE_1;
fp@369:                           else state = STATE_3;
fp@369:                           break;
fp@369:                   case STATE_3:
fp@369:                           action_3();
fp@369:                           state = STATE_1;
fp@369:                           break;
fp@369:           }
fp@369:   }
fp@369: \end{lstlisting}
fp@369: 
fp@369: For small state machines, this is an option. The disadvantage is, that
fp@369: with an increasing number of states the code soon gets complex and an
fp@369: additional case differentiation is executed each run. Besides, lots of
fp@369: indentation is wasted.
fp@369: 
fp@369: The method used in the master is to implement every state in an own
fp@369: function and to store the current state function with a function
fp@369: pointer:
fp@369: 
fp@1085: \begin{lstlisting}[gobble=2,language=C,numbers=left]
fp@369:   void (*state)(void *) = state1;
fp@369: 
fp@369:   void state_machine_run(void *priv_data) {
fp@369:           state(priv_data);
fp@369:   }
fp@369: 
fp@369:   void state1(void *priv_data) {
fp@369:           action_1();
fp@369:           state = state2;
fp@369:   }
fp@369: 
fp@369:   void state2(void *priv_data) {
fp@369:           action_2();
fp@369:           if (some_condition) state = state1;
fp@369:           else state = state2;
fp@369:   }
fp@369: 
fp@369:   void state3(void *priv_data) {
fp@369:           action_3();
fp@369:           state = state1;
fp@369:   }
fp@369: \end{lstlisting}
fp@369: 
fp@1085: In the master code, state pointers of all state machines\footnote{All except
fp@1085: for the EoE state machine, because multiple EoE slaves have to be handled in
fp@1085: parallel. For this reason each EoE handler object has its own state pointer.}
fp@1202: are gathered in a single object of the \lstinline+ec_fsm_master_t+ class. This
fp@1202: is advantageous, because there is always one instance of every state machine
fp@1085: available and can be started on demand.
fp@369: 
fp@369: \paragraph{Mealy and Moore}
fp@369: 
fp@1202: If a closer look is taken to the above listing, it can be seen that the
fp@1202: actions executed (the ``outputs'' of the state machine) only depend on the
fp@1202: current state. This accords to the ``Moore'' model introduced in
fp@2515: \autoref{sec:fsmtheory}. As mentioned, the ``Mealy'' model offers a higher
fp@1202: flexibility, which can be seen in the listing below:
fp@369: 
fp@1085: \begin{lstlisting}[gobble=2,language=C,numbers=left]
fp@369:   void state7(void *priv_data) {
fp@369:           if (some_condition) {
fp@369:                   action_7a();
fp@369:                   state = state1;
fp@369:           }
fp@369:           else {
fp@369:                   action_7b();
fp@369:                   state = state8;
fp@369:           }
fp@369:   }
fp@369: \end{lstlisting}
fp@369: 
fp@369: \begin{description}
fp@1202: 
fp@1202: \item[\linenum{3} + \linenum{7}] The state function executes the actions
fp@1202: depending on the state transition, that is about to be done.
fp@1202: 
fp@369: \end{description}
fp@369: 
fp@369: The most flexible alternative is to execute certain actions depending
fp@369: on the state, followed by some actions dependent on the state
fp@369: transition:
fp@369: 
fp@1085: \begin{lstlisting}[gobble=2,language=C,numbers=left]
fp@369:   void state9(void *priv_data) {
fp@369:           action_9();
fp@369:           if (some_condition) {
fp@369:                   action_9a();
fp@369:                   state = state7;
fp@369:           }
fp@369:           else {
fp@369:                   action_9b();
fp@369:                   state = state10;
fp@369:           }
fp@369:   }
fp@369: \end{lstlisting}
fp@369: 
fp@1202: This model is often used in the master. It combines the best aspects of both
fp@1202: approaches.
fp@369: 
fp@369: \paragraph{Using Sub State Machines}
fp@369: 
fp@1202: To avoid having too much states, certain functions of the EtherCAT master
fp@1202: state machine have been sourced out into sub state machines.  This helps to
fp@1085: encapsulate the related workflows and moreover avoids the ``state explosion''
fp@2515: phenomenon described in \autoref{sec:fsmtheory}. If the master would instead
fp@1289: use one big state machine, the number of states would be a multiple of the
fp@1289: actual number. This would increase the level of complexity to a non-manageable
fp@1289: grade.
fp@369: 
fp@369: \paragraph{Executing Sub State Machines}
fp@369: 
fp@369: If a state machine starts to execute a sub state machine, it usually
fp@369: remains in one state until the sub state machine terminates. This is
fp@369: usually done like in the listing below, which is taken out of the
fp@369: slave configuration state machine code:
fp@369: 
fp@1085: \begin{lstlisting}[gobble=2,language=C,numbers=left]
fp@813:   void ec_fsm_slaveconf_safeop(ec_fsm_t *fsm)
fp@369:   {
fp@369:           fsm->change_state(fsm); // execute state change
fp@369:                                   // sub state machine
fp@369: 
fp@369:           if (fsm->change_state == ec_fsm_error) {
fp@369:                   fsm->slave_state = ec_fsm_end;
fp@369:                   return;
fp@369:           }
fp@369: 
fp@369:           if (fsm->change_state != ec_fsm_end) return;
fp@369: 
fp@369:           // continue state processing
fp@369:           ...
fp@369: \end{lstlisting}
fp@369: 
fp@369: \begin{description}
fp@1202: 
fp@1202: \item[\linenum{3}] \lstinline+change_state+ is the state pointer of the state
fp@1202: change state machine. The state function, the pointer points on, is
fp@1202: executed\ldots
fp@1202: 
fp@1202: \item[\linenum{6}] \ldots either until the state machine terminates with the
fp@1202: error state \ldots
fp@1202: 
fp@1202: \item[\linenum{11}] \ldots or until the state machine terminates in the end
fp@1202: state. Until then, the ``higher'' state machine remains in the current state
fp@1202: and executes the sub state machine again in the next cycle.
fp@1202: 
fp@369: \end{description}
fp@369: 
fp@369: \paragraph{State Machine Descriptions}
fp@369: 
fp@1202: The below sections describe every state machine used in the EtherCAT master.
fp@1202: The textual descriptions of the state machines contain references to the
fp@1202: transitions in the corresponding state transition diagrams, that are marked
fp@1202: with an arrow followed by the name of the successive state. Transitions caused
fp@1306: by trivial error cases (i.\,e.\ no response from slave) are not described
fp@1202: explicitly. These transitions are drawn as dashed arrows in the diagrams.
fp@1202: 
fp@1202: %------------------------------------------------------------------------------
fp@1202: 
fp@1202: \section{The Master State Machine}
fp@1202: \label{sec:fsm-master}
fp@1202: \index{FSM!Master}
fp@1202: 
fp@1202: The master state machine is executed in the context of the master thread.
fp@2515: \autoref{fig:fsm-master} shows its transition diagram. Its purposes are:
fp@369: 
fp@369: \begin{figure}[htbp]
fp@369:   \centering
fp@1202:   \includegraphics[width=\textwidth]{graphs/fsm_master}
fp@1202:   \caption{Transition diagram of the master state machine}
fp@1202:   \label{fig:fsm-master}
fp@369: \end{figure}
fp@369: 
fp@369: \begin{description}
fp@1202: 
fp@1202: \item[Bus monitoring] The bus topology is monitored. If it changes, the bus is
fp@1202: (re-)scanned.
fp@1202: 
fp@1202: \item[Slave configuration] The application-layer states of the slaves are
fp@1202: monitored. If a slave is not in the state it supposed to be, the slave is
fp@1202: (re-)configured.
fp@1202: 
fp@1202: \item[Request handling] Requests (either originating from the application or
fp@1203: from external sources) are handled. A request is a job that the master shall
fp@1327: process asynchronously, for example an SII access, SDO access, or similar.
fp@369: 
fp@369: \end{description}
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@1085: \section{The Slave Scan State Machine}
fp@369: \label{sec:fsm-scan}
fp@369: \index{FSM!Slave Scan}
fp@369: 
fp@369: The slave scan state machine, which can be seen in
fp@2515: \autoref{fig:fsm-slavescan}, leads through the process of reading desired
fp@1202: slave information.
fp@369: 
fp@369: \begin{figure}[htbp]
fp@369:   \centering
fp@1202:   \includegraphics[height=.8\textheight]{graphs/fsm_slave_scan}
fp@369:   \caption{Transition diagram of the slave scan state machine}
fp@369:   \label{fig:fsm-slavescan}
fp@369: \end{figure}
fp@369: 
fp@1202: The scan process includes the following steps:
fp@1202: 
fp@369: \begin{description}
fp@1202: 
fp@1202: \item[Node Address] The node address is set for the slave, so that it can be
fp@1202: node-addressed for all following operations.
fp@1202: 
fp@1202: \item[AL State] The initial application-layer state is read.
fp@1202: 
fp@1202: \item[Base Information] Base information (like the number of supported FMMUs)
fp@1202: is read from the lower physical memory.
fp@1202: 
fp@1202: \item[Data Link] Information about the physical ports is read.
fp@1202: 
fp@1202: \item[SII Size] The size of the SII contents is determined to allocate SII
fp@1202: image memory.
fp@1202: 
fp@1202: \item[SII Data] The SII contents are read into the master's image.
fp@1202: 
fp@1202: \item[PREOP] If the slave supports CoE, it is set to PREOP state using the
fp@2515: State change FSM (see \autoref{sec:fsm-change}) to enable mailbox
fp@1327: communication and read the PDO configuration via CoE.
fp@1327: 
fp@1327: \item[PDOs] The PDOs are read via CoE (if supported) using the PDO Reading FSM
fp@2515: (see \autoref{sec:fsm-pdo}). If this is successful, the PDO information from
fp@1289: the SII (if any) is overwritten.
fp@369: 
fp@369: \end{description}
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@1085: \section{The Slave Configuration State Machine}
fp@369: \label{sec:fsm-conf}
fp@369: \index{FSM!Slave Configuration}
fp@369: 
fp@369: The slave configuration state machine, which can be seen in
fp@2515: \autoref{fig:fsm-slaveconf}, leads through the process of configuring a
fp@1202: slave and bringing it to a certain application-layer state.
fp@369: 
fp@369: \begin{figure}[htbp]
fp@369:   \centering
fp@1406:   \includegraphics[height=\textheight]{graphs/fsm_slave_conf}
fp@369:   \caption{Transition diagram of the slave configuration state
fp@369:     machine}
fp@369:   \label{fig:fsm-slaveconf}
fp@369: \end{figure}
fp@369: 
fp@369: \begin{description}
fp@1202: 
fp@1202: \item[INIT] The state change FSM is used to bring the slave to the INIT state.
fp@1202: 
fp@1202: \item[FMMU Clearing] To avoid that the slave reacts on any process data, the
fp@1202: FMMU configuration are cleared. If the slave does not support FMMUs, this
fp@1202: state is skipped. If INIT is the requested state, the state machine is
fp@1202: finished.
fp@1202: 
fp@1202: \item[Mailbox Sync Manager Configuration] If the slaves support mailbox
fp@1202: communication, the mailbox sync managers are configured. Otherwise this state
fp@1202: is skipped.
fp@1202: 
fp@1202: \item[PREOP] The state change FSM is used to bring the slave to PREOP state.
fp@1202: If this is the requested state, the state machine is finished.
fp@1202: 
fp@1327: \item[SDO Configuration] If there is a slave configuration attached (see
fp@2515: \autoref{sec:masterconfig}), and there are any SDO configurations are
fp@1202: provided by the application, these are sent to the slave.
fp@1202: 
fp@1327: \item[PDO Configuration] The PDO configuration state machine is executed to
fp@1327: apply all necessary PDO configurations.
fp@1327: 
fp@1327: \item[PDO Sync Manager Configuration] If any PDO sync managers exist, they are
fp@1202: configured.
fp@1202: 
fp@1202: \item[FMMU Configuration] If there are FMMUs configurations supplied by the
fp@1327: application (i.\,e.\ if the application registered PDO entries), they are
fp@2421: applied.
fp@1202: 
fp@1202: \item[SAFEOP] The state change FSM is used to bring the slave to SAFEOP state.
fp@1202: If this is the requested state, the state machine is finished.
fp@1202: 
fp@1202: \item[OP] The state change FSM is used to bring the slave to OP state.
fp@1202: If this is the requested state, the state machine is finished.
fp@369: 
fp@369: \end{description}
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@1085: \section{The State Change State Machine}
fp@369: \label{sec:fsm-change}
fp@369: \index{FSM!State Change}
fp@369: 
fp@369: The state change state machine, which can be seen in
fp@2515: \autoref{fig:fsm-change}, leads through the process of changing a slave's
fp@1202: application-layer state. This implements the states and transitions described
fp@1289: in \cite[sec.~6.4.1]{alspec}.
fp@369: 
fp@369: \begin{figure}[htbp]
fp@369:   \centering
fp@1203:   \includegraphics[width=.6\textwidth]{graphs/fsm_change}
fp@1202:   \caption{Transition Diagram of the State Change State Machine}
fp@369:   \label{fig:fsm-change}
fp@369: \end{figure}
fp@369: 
fp@369: \begin{description}
fp@1203: 
fp@1203: \item[Start] The new application-layer state is requested via the ``AL Control
fp@1306: Request'' register (see~\cite[sec. 5.3.1]{alspec}).
fp@1203: 
fp@1203: \item[Check for Response] Some slave need some time to respond to an AL state
fp@1203: change command, and do not respond for some time. For this case, the command
fp@1203: is issued again, until it is acknowledged.
fp@1203: 
fp@1203: \item[Check AL Status] If the AL State change datagram was acknowledged, the
fp@1289: ``AL Control Response'' register (see~\cite[sec. 5.3.2]{alspec}) must be read
fp@1289: out until the slave changes the AL state.
fp@1203: 
fp@1203: \item[AL Status Code] If the slave refused the state change command, the
fp@1203: reason can be read from the ``AL Status Code'' field in the ``AL State
fp@1289: Changed'' registers (see~\cite[sec. 5.3.3]{alspec}).
fp@1203: 
fp@1203: \item[Acknowledge State] If the state change was not successful, the master
fp@1203: has to acknowledge the old state by writing to the ``AL Control request''
fp@1203: register again.
fp@1203: 
fp@1203: \item[Check Acknowledge] After sending the acknowledge command, it has to read
fp@1203: out the ``AL Control Response'' register again.
fp@369: 
fp@369: \end{description}
fp@369: 
fp@1203: The ``start\_ack'' state is a shortcut in the state machine for the case, that
fp@1203: the master wants to acknowledge a spontaneous AL state change, that was not
fp@1203: requested.
fp@1203: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@1085: \section{The SII State Machine}
fp@369: \label{sec:fsm-sii}
fp@369: \index{FSM!SII}
fp@369: 
fp@2515: The SII\index{SII} state machine (shown in \autoref{fig:fsm-sii})
fp@1289: implements the process of reading or writing SII data via the Slave
fp@1289: Information Interface described in \cite[sec.~6.4]{dlspec}.
fp@369: 
fp@369: \begin{figure}[htbp]
fp@369:   \centering
fp@1203:   \includegraphics[width=.5\textwidth]{graphs/fsm_sii}
fp@1202:   \caption{Transition Diagram of the SII State Machine}
fp@369:   \label{fig:fsm-sii}
fp@369: \end{figure}
fp@369: 
fp@1203: This is how the reading part of the state machine works:
fp@1202: 
fp@369: \begin{description}
fp@1203: 
fp@1203: \item[Start Reading] The read request and the requested word address are
fp@1203: written to the SII attribute.
fp@1203: 
fp@1203: \item[Check Read Command] If the SII read request command has been
fp@1203: acknowledged, a timer is started. A datagram is issued, that reads out the SII
fp@1203: attribute for state and data.
fp@1203: 
fp@1203: \item[Fetch Data] If the read operation is still busy (the SII is usually
fp@1203: implemented as an E$^2$PROM), the state is read again. Otherwise the data are
fp@1203: copied from the datagram.
fp@1203: 
fp@369: \end{description}
fp@369: 
fp@1203: The writing part works nearly similar:
fp@369: 
fp@369: \begin{description}
fp@1203: 
fp@1203: \item[Start Writing] A write request, the target address and the data word are
fp@1203: written to the SII attribute.
fp@1203: 
fp@1203: \item[Check Write Command] If the SII write request command has been
fp@1203: acknowledged, a timer is started. A datagram is issued, that reads out the SII
fp@1203: attribute for the state of the write operation.
fp@1203: 
fp@1203: \item[Wait while Busy] If the write operation is still busy (determined by a
fp@1203: minimum wait time and the state of the busy flag), the state machine remains in
fp@1203: this state to avoid that another write operation is issued too early.
fp@1203: 
fp@369: \end{description}
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@1327: \section{The PDO State Machines}
fp@1203: \label{sec:fsm-pdo}
fp@1327: \index{FSM!PDO}
fp@1327: 
fp@1327: The PDO state machines are a set of state machines that read or write the PDO
fp@1327: assignment and the PDO mapping via the ``CoE Communication Area'' described in
fp@1327: \cite[sec. 5.6.7.4]{alspec}. For the object access, the CANopen over EtherCAT
fp@2515: access primitives are used (see \autoref{sec:coe}), so the slave must support
fp@1289: the CoE mailbox protocol.
fp@1203: 
fp@2515: \paragraph{PDO Reading FSM} This state machine (\autoref{fig:fsm-pdo-read})
fp@1327: has the purpose to read the complete PDO configuration of a slave. It reads
fp@1327: the PDO assignment for each Sync Manager and uses the PDO Entry Reading FSM
fp@2515: (\autoref{fig:fsm-pdo-entry-read}) to read the mapping for each assigned PDO.
fp@1203: 
fp@1203: \begin{figure}[htbp]
fp@1203:   \centering
fp@1203:   \includegraphics[width=.4\textwidth]{graphs/fsm_pdo_read}
fp@1327:   \caption{Transition Diagram of the PDO Reading State Machine}
fp@1203:   \label{fig:fsm-pdo-read}
fp@1203: \end{figure}
fp@1203: 
fp@1327: Basically it reads the every Sync manager's PDO assignment SDO's
fp@1203: (\lstinline+0x1C1x+) number of elements to determine the number of assigned
fp@1327: PDOs for this sync manager and then reads out the subindices of the SDO to get
fp@1327: the assigned PDO's indices. When a PDO index is read, the PDO Entry Reading
fp@1327: FSM is executed to read the PDO's mapped PDO entries.
fp@1327: 
fp@1327: \paragraph{PDO Entry Reading FSM} This state machine
fp@2515: (\autoref{fig:fsm-pdo-entry-read}) reads the PDO mapping (the PDO entries) of
fp@1327: a PDO. It reads the respective mapping SDO (\lstinline+0x1600+ --
fp@1293: \lstinline+0x17ff+, or \lstinline+0x1a00+ -- \lstinline+0x1bff+) for the given
fp@1327: PDO by reading first the subindex zero (number of elements) to determine the
fp@1327: number of mapped PDO entries. After that, each subindex is read to get the
fp@1327: mapped PDO entry index, subindex and bit size.
fp@1203: 
fp@1203: \begin{figure}[htbp]
fp@1203:   \centering
fp@1203:   \includegraphics[width=.4\textwidth]{graphs/fsm_pdo_entry_read}
fp@1327:   \caption{Transition Diagram of the PDO Entry Reading State Machine}
fp@1204:   \label{fig:fsm-pdo-entry-read}
fp@1203: \end{figure}
fp@1203: 
fp@1203: \begin{figure}[htbp]
fp@1203:   \centering
fp@1203:   \includegraphics[width=.9\textwidth]{graphs/fsm_pdo_conf}
fp@1327:   \caption{Transition Diagram of the PDO Configuration State Machine}
fp@1204:   \label{fig:fsm-pdo-conf}
fp@1203: \end{figure}
fp@1203: 
fp@1203: \begin{figure}[htbp]
fp@1203:   \centering
fp@1203:   \includegraphics[width=.4\textwidth]{graphs/fsm_pdo_entry_conf}
fp@1327:   \caption{Transition Diagram of the PDO Entry Configuration State Machine}
fp@1204:   \label{fig:fsm-pdo-entry-conf}
fp@1203: \end{figure}
fp@1203: 
fp@1203: %------------------------------------------------------------------------------
fp@1203: 
fp@1085: \chapter{Mailbox Protocol Implementations}
fp@369: \index{Mailbox}
fp@369: 
fp@1917: The EtherCAT master implements the CANopen over EtherCAT (CoE), Ethernet over
fp@1917: EtherCAT (EoE), File-access over EtherCAT (FoE), Vendor-specific over EtherCAT
fp@1917: (VoE) and Servo Profile over EtherCAT (SoE) mailbox protocols. See the below
fp@1917: sections for details.
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@1327: \section{Ethernet over EtherCAT (EoE)}
fp@1269: \label{sec:eoe}
fp@369: \index{EoE}
fp@369: 
fp@1306: The EtherCAT master implements the
fp@1327: Ethernet over EtherCAT\nomenclature{EoE}{Ethernet over EtherCAT, Mailbox
fp@1306: Protocol} mailbox protocol~\cite[sec.~5.7]{alspec} to enable the tunneling of
fp@1306: Ethernet frames to special slaves, that can either have physical Ethernet
fp@1306: ports to forward the frames to, or have an own IP stack to receive the frames.
fp@369: 
fp@369: \paragraph{Virtual Network Interfaces}
fp@369: 
fp@1202: The master creates a virtual EoE network interface for every EoE-capable
fp@1202: slave. These interfaces are called either
fp@1202: 
fp@1202: \begin{description}
fp@1202: 
fp@1293: \item[eoeXsY] for a slave without an alias address (see
fp@2515: \autoref{sec:ethercat-alias}), where X is the master index and Y is the
fp@1293: slave's ring position, or
fp@1202: 
fp@1202: \item[eoeXaY] for a slave with a non-zero alias address, where X is the master
fp@1202: index and Y is the decimal alias address.
fp@1202: 
fp@1202: \end{description}
fp@1202: 
fp@1202: Frames sent to these interfaces are forwarded to the associated slaves by the
fp@1202: master. Frames, that are received by the slaves, are fetched by the master and
fp@1202: forwarded to the virtual interfaces.
fp@369: 
fp@369: This bears the following advantages:
fp@369: 
fp@369: \begin{itemize}
fp@1202: 
fp@369: \item Flexibility: The user can decide, how the EoE-capable slaves are
fp@1202: interconnected with the rest of the world.
fp@1202: 
fp@1202: \item Standard tools can be used to monitor the EoE activity and to configure
fp@1202: the EoE interfaces.
fp@1202: 
fp@1202: \item The Linux kernel's layer-2-bridging implementation (according to the
fp@1202: IEEE 802.1D MAC Bridging standard) can be used natively to bridge Ethernet
fp@1202: traffic between EoE-capable slaves.
fp@1202: 
fp@1202: \item The Linux kernel's network stack can be used to route packets between
fp@1202: EoE-capable slaves and to track security issues, just like having physical
fp@1202: network interfaces.
fp@1202: 
fp@369: \end{itemize}
fp@369: 
fp@369: \paragraph{EoE Handlers}
fp@369: 
fp@1202: The virtual EoE interfaces and the related functionality is encapsulated in
fp@1202: the \lstinline+ec_eoe_t+ class. An object of this class is called ``EoE
fp@1202: handler''. For example the master does not create the network interfaces
fp@1202: directly: This is done inside the constructor of an EoE handler. An EoE
fp@1202: handler additionally contains a frame queue. Each time, the kernel passes a
fp@1202: new socket buffer for sending via the interface's
fp@1202: \lstinline+hard_start_xmit()+ callback, the socket buffer is queued for
fp@1202: transmission by the EoE state machine (see below). If the queue gets filled
fp@1202: up, the passing of new socket buffers is suspended with a call to
fp@1202: \lstinline+netif_stop_queue()+.
fp@1202: 
fp@1202: \paragraph{Creation of EoE Handlers}
fp@1202: 
fp@2515: During bus scanning (see \autoref{sec:fsm-scan}), the master determines the
fp@1289: supported mailbox protocols foe each slave. This is done by examining the
fp@1202: ``Supported Mailbox Protocols'' mask field at word address 0x001C of the
fp@1202: SII\index{SII}. If bit 1 is set, the slave supports the EoE protocol. In this
fp@1202: case, an EoE handler is created for that slave.
fp@369: 
fp@369: \paragraph{EoE State Machine}
fp@369: \index{FSM!EoE}
fp@369: 
fp@1202: Every EoE handler owns an EoE state machine, that is used to send frames to
fp@1202: the corresponding slave and receive frames from the it via the EoE
fp@369: communication primitives. This state machine is showed in
fp@2515: \autoref{fig:fsm-eoe}.
fp@369: 
fp@369: \begin{figure}[htbp]
fp@369:   \centering
fp@1202:   \includegraphics[width=.7\textwidth]{images/fsm-eoe} % FIXME
fp@1202:   \caption{Transition Diagram of the EoE State Machine}
fp@369:   \label{fig:fsm-eoe}
fp@369: \end{figure}
fp@369: 
fp@1202: % FIXME
fp@1202: 
fp@369: \begin{description}
fp@369: \item[RX\_START] The beginning state of the EoE state machine. A
fp@369:   mailbox check datagram is sent, to query the slave's mailbox for new
fp@379:   frames. $\rightarrow$~RX\_CHECK
fp@369: 
fp@369: \item[RX\_CHECK] The mailbox check datagram is received. If the
fp@369:   slave's mailbox did not contain data, a transmit cycle is started.
fp@379:   $\rightarrow$~TX\_START
fp@369: 
fp@369:   If there are new data in the mailbox, a datagram is sent to fetch
fp@379:   the new data. $\rightarrow$~RX\_FETCH
fp@369: 
fp@369: \item[RX\_FETCH] The fetch datagram is received. If the mailbox data
fp@369:   do not contain a ``EoE Fragment request'' command, the data are
fp@369:   dropped and a transmit sequence is started.
fp@379:   $\rightarrow$~TX\_START
fp@369: 
fp@369:   If the received Ethernet frame fragment is the first fragment, a new
fp@369:   socket buffer is allocated. In either case, the data are copied into
fp@369:   the correct position of the socket buffer.
fp@369: 
fp@369:   If the fragment is the last fragment, the socket buffer is forwarded
fp@369:   to the network stack and a transmit sequence is started.
fp@379:   $\rightarrow$~TX\_START
fp@369: 
fp@369:   Otherwise, a new receive sequence is started to fetch the next
fp@379:   fragment. $\rightarrow$~RX\_\-START
fp@369: 
fp@369: \item[TX\_START] The beginning state of a transmit sequence. It is
fp@1085:   checked, if the transmission queue contains a frame to send. If not,
fp@379:   a receive sequence is started. $\rightarrow$~RX\_START
fp@369: 
fp@369:   If there is a frame to send, it is dequeued. If the queue was
fp@369:   inactive before (because it was full), the queue is woken up with a
fp@369:   call to \textit{netif\_wake\_queue()}. The first fragment of the
fp@379:   frame is sent. $\rightarrow$~TX\_SENT
fp@369: 
fp@369: \item[TX\_SENT] It is checked, if the first fragment was sent
fp@369:   successfully. If the current frame consists of further fragments,
fp@379:   the next one is sent. $\rightarrow$~TX\_SENT
fp@369: 
fp@369:   If the last fragment was sent, a new receive sequence is started.
fp@379:   $\rightarrow$~RX\_START
fp@369: \end{description}
fp@369: 
fp@369: \paragraph{EoE Processing}
fp@369: 
fp@1202: To execute the EoE state machine of every active EoE handler, there must be a
fp@1202: cyclic process. The easiest solution would be to execute the EoE state
fp@1202: machines synchronously with the master state machine (see
fp@2515: \autoref{sec:fsm-master}). This approach has the following disadvantage:
fp@1202: 
fp@1202: Only one EoE fragment could be sent or received every few cycles. This
fp@1085: causes the data rate to be very low, because the EoE state machines are not
fp@1085: executed in the time between the application cycles. Moreover, the data rate
fp@1085: would be dependent on the period of the application task.
fp@1085: 
fp@1202: To overcome this problem, an own cyclic process is needed to asynchronously
fp@1202: execute the EoE state machines. For that, the master owns a kernel timer, that
fp@1202: is executed each timer interrupt. This guarantees a constant bandwidth, but
fp@1202: poses the new problem of concurrent access to the master. The locking
fp@2515: mechanisms needed for this are introduced in \autoref{sec:concurr}.
fp@369: 
fp@369: \paragraph{Automatic Configuration}
fp@369: 
fp@1202: By default, slaves are left in PREOP state, if no configuration is applied. If
fp@1202: an EoE interface link is set to ``up'', the requested slave's
fp@1202: application-layer state is automatically set to OP.
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@1327: \section{CANopen over EtherCAT (CoE)}
fp@1269: \label{sec:coe}
fp@369: \index{CoE}
fp@369: 
fp@1327: The CANopen over EtherCAT\nomenclature{CoE}{CANopen over EtherCAT, Mailbox
fp@1306: Protocol} protocol~\cite[sec.~5.6]{alspec} is used to configure slaves and
fp@1306: exchange data objects on application level.
fp@1202: 
fp@1297: % TODO
fp@1203: %
fp@1203: % Download / Upload
fp@1203: % Expedited / Normal
fp@1204: % Segmenting
fp@1327: % SDO Info Services
fp@1203: %
fp@1203: 
fp@1327: \paragraph{SDO Download State Machine}
fp@1327: 
fp@1327: The best time to apply SDO configurations is during the slave's PREOP state,
fp@1289: because mailbox communication is already possible and slave's application will
fp@1289: start with updating input data in the succeeding SAFEOP state. Therefore the
fp@1327: SDO configuration has to be part of the slave configuration state machine (see
fp@2515: \autoref{sec:fsm-conf}): It is implemented via an SDO download state machine,
fp@1289: that is executed just before entering the slave's SAFEOP state. In this way,
fp@1327: it is guaranteed that the SDO configurations are applied each time, the slave
fp@1289: is reconfigured.
fp@369: 
fp@1327: The transition diagram of the SDO Download state machine can be seen
fp@2515: in \autoref{fig:fsm-coedown}.
fp@369: 
fp@369: \begin{figure}[htbp]
fp@369:   \centering
fp@1202:   \includegraphics[width=.9\textwidth]{images/fsm-coedown} % FIXME
fp@369:   \caption{Transition diagram of the CoE download state machine}
fp@369:   \label{fig:fsm-coedown}
fp@369: \end{figure}
fp@369: 
fp@1202: % FIXME
fp@1202: 
fp@369: \begin{description}
fp@369: \item[START] The beginning state of the CoE download state
fp@1327:   machine. The ``SDO Download Normal Request'' mailbox command is
fp@379:   sent. $\rightarrow$~REQUEST
fp@369: 
fp@369: \item[REQUEST] It is checked, if the CoE download request has been
fp@369:   received by the slave. After that, a mailbox check command is issued
fp@379:   and a timer is started. $\rightarrow$~CHECK
fp@369: 
fp@369: \item[CHECK] If no mailbox data is available, the timer is checked.
fp@369:   \begin{itemize}
fp@1327:   \item If it timed out, the SDO download is aborted.
fp@379:     $\rightarrow$~ERROR
fp@369:   \item Otherwise, the mailbox is queried again.
fp@379:     $\rightarrow$~CHECK
fp@369:   \end{itemize}
fp@369: 
fp@369:   If the mailbox contains new data, the response is fetched.
fp@379:   $\rightarrow$~RESPONSE
fp@369: 
fp@369: \item[RESPONSE] If the mailbox response could not be fetched, the data
fp@1327:   is invalid, the wrong protocol was received, or a ``Abort SDO
fp@1327:   Transfer Request'' was received, the SDO download is aborted.
fp@379:   $\rightarrow$~ERROR
fp@369: 
fp@1327:   If a ``SDO Download Normal Response'' acknowledgement was received,
fp@1327:   the SDO download was successful. $\rightarrow$~END
fp@1327: 
fp@1327: \item[END] The SDO download was successful.
fp@1327: 
fp@1327: \item[ERROR] The SDO download was aborted due to an error.
fp@369: 
fp@369: \end{description}
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@1327: \section{Vendor specific over EtherCAT (VoE)}
fp@1270: \label{sec:voe}
fp@1270: \index{VoE}
fp@1270: 
fp@1270: The VoE protocol opens the possibility to implement a vendor-specific mailbox
fp@1270: communication protocol. VoE mailbox messages are prepended by a VoE header
fp@1270: containing a 32-bit vendor ID and a 16-bit vendor-type. There are no more
fp@1270: constraints regarding this protocol.
fp@1270: 
fp@1270: The EtherCAT master allows to create multiple VoE handlers per slave
fp@2515: configuration via the application interface (see \autoref{chap:api}). These
fp@1270: handlers contain the state machine necessary for the communication via VoE.
fp@1298: 
fp@2515: For more information about using VoE handlers, see \autoref{sec:api-voe} or
fp@1298: the example applications provided in the \textit{examples/} subdirectory.
fp@1270: 
fp@1270: %------------------------------------------------------------------------------
fp@1270: 
fp@1917: \section{Servo Profile over EtherCAT (SoE)}
fp@1917: \label{sec:soe}
fp@1917: \index{SoE}
fp@1917: 
fp@1917: The SoE protocol implements the Service Channel layer, specified in IEC
fp@1917: 61800-7 \cite{soespec} via EtherCAT mailboxes.
fp@1917: 
fp@1917: The SoE protocol is quite similar to the CoE protocol (see
fp@2515: \autoref{sec:coe}). Instead of SDO indices and subindices, so-called
fp@1917: identification numbers (IDNs) identify parameters.
fp@1917: 
fp@1917: The implementation covers the ``SCC Read'' and ``SCC Write'' primitives, each
fp@1917: with the ability to fragment data.
fp@1917: 
fp@1917: There are several ways to use the SoE implementation:
fp@1917: 
fp@1917: \begin{itemize}
fp@1917: 
fp@1917: \item Reading and writing IDNs via the command-line tool (see
fp@2515: \autoref{sec:soeaccess}).
fp@1917: 
fp@1917: \item Storing configurations for arbitrary IDNs via the application interface
fp@2515: (see \autoref{chap:api}, i.\,e.~\lstinline+ecrt_slave_config_idn()+). These
fp@1917: configurations are written to the slave during configuration in PREOP state,
fp@1917: before going to SAFEOP.
fp@1917: 
fp@2515: \item The user-space library (see \autoref{sec:userlib}), offers functions to
fp@1917: read/write IDNs in blocking mode (\lstinline+ecrt_master_read_idn()+,
fp@1917: \lstinline+ecrt_master_write_idn()+).
fp@1917: 
fp@1917: \end{itemize}
fp@1917: 
fp@1917: %------------------------------------------------------------------------------
fp@1917: 
fp@1275: \chapter{Userspace Interfaces}
fp@369: \label{sec:user}
fp@1275: \index{Userspace}
fp@1202: 
fp@1203: For the master runs as a kernel module, accessing it is natively limited to
fp@1275: analyzing Syslog messages and controlling using \textit{modutils}.
fp@1275: 
fp@1275: It was necessary to implement further interfaces, that make it easier to access
fp@1275: the master from userspace and allow a finer influence. It should be possible
fp@1203: to view and to change special parameters at runtime.
fp@1203: 
fp@1275: Bus visualization is another point: For development and debugging purposes it
fp@1275: is necessary to show the connected slaves with a single command, for instance
fp@2515: (see \autoref{sec:tool}).
fp@1275: 
fp@1275: The application interface has to be available in userspace, to allow userspace
fp@1275: programs to use EtherCAT master functionality. This was implemented via a
fp@2515: character device and a userspace library (see \autoref{sec:userlib}).
fp@1275: 
fp@1275: Another aspect is automatic startup and configuration. The master must be able
fp@1275: to automatically start up with a persistent configuration (see
fp@2515: \autoref{sec:system}).
fp@1203: 
fp@1203: A last thing is monitoring EtherCAT communication. For debugging purposes,
fp@1203: there had to be a way to analyze EtherCAT datagrams. The best way would be
fp@2515: with a popular network analyzer, like Wireshark \cite{wireshark} or others
fp@2515: (see \autoref{sec:debug}).
fp@1275: 
fp@1275: This chapter covers all these points and introduces the interfaces and tools
fp@1203: to make all that possible.
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@1087: \section{Command-line Tool}
fp@1289: \label{sec:tool}
fp@1087: 
fp@1297: % TODO --master
fp@1087: 
fp@1202: \subsection{Character Devices}
fp@1087: \label{sec:cdev}
fp@1087: 
fp@1214: Each master instance will get a character device as a userspace interface.
fp@1214: The devices are named \textit{/dev/EtherCATx}, where $x \in \{0 \ldots n\}$ is
fp@1214: the index of the master.
fp@1214: 
fp@1214: \paragraph{Device Node Creation} The character device nodes are automatically
fp@1289: created, if the \lstinline+udev+ Package is installed. See
fp@2515: \autoref{sec:autonode} for how to install and configure it.
fp@1087: 
fp@1087: %------------------------------------------------------------------------------
fp@1087: 
fp@1202: \subsection{Setting Alias Addresses}
fp@1293: \label{sec:ethercat-alias}
fp@1140: 
fp@1140: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_alias}
fp@1140: 
fp@1140: %------------------------------------------------------------------------------
fp@1140: 
fp@1202: \subsection{Displaying the Bus Configuration}
fp@1293: \label{sec:ethercat-config}
fp@1140: 
fp@1140: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_config}
fp@1140: 
fp@1140: %------------------------------------------------------------------------------
fp@1140: 
fp@1514: \subsection{Output PDO information in C Language}
fp@1514: \label{sec:ethercat-cstruct}
fp@1514: 
fp@1514: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_cstruct}
fp@1514: 
fp@1514: %------------------------------------------------------------------------------
fp@1514: 
fp@1202: \subsection{Displaying Process Data}
fp@1140: 
fp@1140: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_data}
fp@1140: 
fp@1140: %------------------------------------------------------------------------------
fp@1140: 
fp@1202: \subsection{Setting a Master's Debug Level}
fp@1399: \label{sec:ethercat-debug}
fp@1140: 
fp@1140: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_debug}
fp@1140: 
fp@1140: %------------------------------------------------------------------------------
fp@1140: 
fp@1202: \subsection{Configured Domains}
fp@1140: 
fp@1140: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_domains}
fp@1140: 
fp@1140: %------------------------------------------------------------------------------
fp@1140: 
fp@1423: \subsection{SDO Access}
fp@1423: 
fp@1423: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_download}
fp@1423: 
fp@1423: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_upload}
fp@1423: 
fp@1423: %------------------------------------------------------------------------------
fp@1423: 
fp@1485: \subsection{EoE Statistics}
fp@1485: 
fp@1485: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_eoe}
fp@1485: 
fp@1485: %------------------------------------------------------------------------------
fp@1485: 
fp@1423: \subsection{File-Access over EtherCAT}
fp@1423: 
fp@1423: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_foe_read}
fp@1423: 
fp@1423: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_foe_write}
fp@1423: 
fp@1423: %------------------------------------------------------------------------------
fp@1423: 
fp@1423: \subsection{Creating Topology Graphs}
fp@1423: 
fp@1423: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_graph}
fp@1423: 
fp@1423: %------------------------------------------------------------------------------
fp@1423: 
fp@1202: \subsection{Master and Ethernet Devices}
fp@1140: 
fp@1140: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_master}
fp@1140: 
fp@1140: %------------------------------------------------------------------------------
fp@1140: 
fp@1327: \subsection{Sync Managers, PDOs and PDO Entries}
fp@1140: 
fp@1140: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_pdos}
fp@1140: 
fp@1140: %------------------------------------------------------------------------------
fp@1140: 
fp@1423: \subsection{Register Access}
fp@1517: \label{sec:regaccess}
fp@1423: 
fp@1423: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_reg_read}
fp@1423: 
fp@1423: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_reg_write}
fp@1423: 
fp@1423: %------------------------------------------------------------------------------
fp@1423: 
fp@1327: \subsection{SDO Dictionary}
fp@1140: 
fp@1140: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_sdos}
fp@1140: 
fp@1140: %------------------------------------------------------------------------------
fp@1140: 
fp@1423: \subsection{SII Access}
fp@1423: \label{sec:siiaccess}
fp@1423: \index{SII!Access}
fp@1423: 
fp@1423: It is possible to directly read or write the complete SII contents of the
fp@1423: slaves. This was introduced for the reasons below:
fp@1423: 
fp@1423: \begin{itemize}
fp@1423: 
fp@1423: \item The format of the SII data is still in development and categories can be
fp@1423: added in the future. With read and write access, the complete memory contents
fp@1423: can be easily backed up and restored.
fp@1423: 
fp@1423: \item Some SII data fields have to be altered (like the alias address). A quick
fp@1423: writing must be possible for that.
fp@1423: 
fp@1423: \item Through reading access, analyzing category data is possible from
fp@1423: userspace.
fp@1423: 
fp@1423: \end{itemize}
fp@1423: 
fp@1423: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_sii_read}
fp@1423: 
fp@1423: Reading out SII data is as easy as other commands. Though the data are in
fp@1423: binary format, analysis is easier with a tool like \textit{hexdump}:
fp@1423: 
fp@1423: \begin{lstlisting}
fp@1423: $ `\textbf{ethercat sii\_read --position 3 | hexdump}`
fp@1423: 0000000 0103 0000 0000 0000 0000 0000 0000 008c
fp@1423: 0000010 0002 0000 3052 07f0 0000 0000 0000 0000
fp@1423: 0000020 0000 0000 0000 0000 0000 0000 0000 0000
fp@1423: ...
fp@1423: \end{lstlisting}
fp@1423: 
fp@1423: Backing up SII contents can easily done with a redirection:
fp@1423: 
fp@1423: \begin{lstlisting}
fp@1423: $ `\textbf{ethercat sii\_read --position 3 > sii-of-slave3.bin}`
fp@1423: \end{lstlisting}
fp@1423: 
fp@1423: To download SII contents to a slave, writing access to the master's character
fp@2515: device is necessary (see \autoref{sec:cdev}).
fp@1423: 
fp@1423: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_sii_write}
fp@1423: 
fp@1423: \begin{lstlisting}
fp@1423: # `\textbf{ethercat sii\_write --position 3 sii-of-slave3.bin}`
fp@1423: \end{lstlisting}
fp@1423: 
fp@1423: The SII contents will be checked for validity and then sent to the slave. The
fp@1423: write operation may take a few seconds.
fp@1140: 
fp@1140: %------------------------------------------------------------------------------
fp@1140: 
fp@1202: \subsection{Slaves on the Bus}
fp@1087: 
fp@1087: Slave information can be gathered with the subcommand \lstinline+slaves+:
fp@1087: 
fp@1140: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_slaves}
fp@1140: 
fp@1140: Below is a typical output:
fp@1140: 
fp@1087: \begin{lstlisting}
fp@1087: $ `\textbf{ethercat slaves}`
fp@1087: 0     0:0  PREOP  +  EK1100 Ethernet Kopplerklemme (2A E-Bus)
fp@1087: 1  5555:0  PREOP  +  EL3162 2K. Ana. Eingang 0-10V
fp@1087: 2  5555:1  PREOP  +  EL4102 2K. Ana. Ausgang 0-10V
fp@1087: 3  5555:2  PREOP  +  EL2004 4K. Dig. Ausgang 24V, 0,5A
fp@1087: \end{lstlisting}
fp@1087: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@1917: \subsection{SoE IDN Access}
fp@1917: \label{sec:soeaccess}
fp@1917: 
fp@1917: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_soe_read}
fp@1917: 
fp@1917: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_soe_write}
fp@1917: 
fp@1917: %------------------------------------------------------------------------------
fp@1917: 
fp@1202: \subsection{Requesting Application-Layer States}
fp@1140: 
fp@1140: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_states}
fp@1140: 
fp@1140: %------------------------------------------------------------------------------
fp@1140: 
fp@1423: \subsection{Displaying the Master Version}
fp@1423: 
fp@1423: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_version}
fp@1423: 
fp@1423: %------------------------------------------------------------------------------
fp@1423: 
fp@1202: \subsection{Generating Slave Description XML}
fp@1140: 
fp@1140: \lstinputlisting[basicstyle=\ttfamily\footnotesize]{external/ethercat_xml}
fp@1140: 
fp@1140: %------------------------------------------------------------------------------
fp@1140: 
fp@1275: \section{Userspace Library}
fp@1275: \label{sec:userlib}
fp@1275: 
fp@2515: The native application interface (see \autoref{chap:api}) resides in
fp@1282: kernelspace and hence is only accessible from inside the kernel. To make the
fp@1282: application interface available from userspace programs, a userspace library
fp@1282: has been created, that can be linked to programs under the terms and
fp@1282: conditions of the LGPL, version 2 \cite{lgpl}.
fp@1282: 
fp@1282: The library is named \textit{libethercat}. Its sources reside in the
fp@1282: \textit{lib/} subdirectory and are build by default when using
fp@1282: \lstinline+make+. It is installed in the \textit{lib/} path below the
fp@1282: installation prefix as \textit{libethercat.a} (for static linking),
fp@1282: \textit{libethercat.la} (for the use with \textit{libtool}) and
fp@1282: \textit{libethercat.so} (for dynamic linking).
fp@1282: 
fp@1299: \subsection{Using the Library}
fp@1282: 
fp@1282: The application interface header \textit{ecrt.h} can be used both in kernel
fp@1282: and in user context.
fp@1282: 
fp@1282: The following minimal example shows how to build a program with EtherCAT
fp@1282: functionality. An entire example can be found in the \textit{examples/user/}
fp@1282: path of the master sources.
fp@1282: 
fp@1282: \begin{lstlisting}[language=C]
fp@1282: #include <ecrt.h>
fp@1282: 
fp@1282: int main(void)
fp@1282: {
fp@1282:     ec_master_t *master = ecrt_request_master(0);
fp@1282: 
fp@1282:     if (!master)
fp@1282:         return 1; // error
fp@1282: 
fp@1282:     pause(); // wait for signal
fp@1282:     return 0;
fp@1282: }
fp@1282: \end{lstlisting}
fp@1282: 
fp@1282: The program can be compiled and dynamically linked to the library with the
fp@1282: below command:
fp@1282: 
fp@2515: \begin{lstlisting}[caption=Linker command for using the userspace library,
fp@2515: label=lst:linker-user]
fp@1282: gcc ethercat.c -o ectest -I/opt/etherlab/include \
fp@1282:     -L/opt/etherlab/lib -lethercat \
fp@1282:     -Wl,--rpath -Wl,/opt/etherlab/lib
fp@1282: \end{lstlisting}
fp@1282: 
fp@1282: The library can also be linked statically to the program:
fp@1282: 
fp@1282: \begin{lstlisting}
fp@1282: gcc -static ectest.c -o ectest -I/opt/etherlab/include \
fp@1282:     /opt/etherlab/lib/libethercat.a
fp@1282: \end{lstlisting}
fp@1282: 
fp@1282: \subsection{Implementation}
fp@1282: \label{sec:userimp}
fp@1282: 
fp@1282: Basically the kernel API was transferred into userspace via the master
fp@2515: character device (see \autoref{chap:arch}, \autoref{fig:arch} and
fp@2515: \autoref{sec:cdev}).
fp@1282: 
fp@1282: The function calls of the kernel API are mapped to the userspace via an
fp@1517: \lstinline+ioctl()+ interface. The userspace API functions share a set of
fp@1517: generic \lstinline+ioctl()+ calls. The kernel part of the interface calls the
fp@1517: according API functions directly, what results in a minimum additional delay
fp@2515: (see \autoref{sec:usertiming}).
fp@1282: 
fp@1299: For performance reasons, the actual domain process data (see
fp@2515: \autoref{sec:processdata}) are not copied between kernel and user memory on
fp@1299: every access: Instead, the data are memory-mapped to the userspace
fp@1299: application. Once the master is configured and activated, the master module
fp@1299: creates one process data memory area spanning all domains and maps it to
fp@1299: userspace, so that the application can directly access the process data. As a
fp@1299: result, there is no additional delay when accessing process data from
fp@1299: userspace.
fp@1299: 
fp@1299: \paragraph{Kernel/User API Differences} Because of the memory-mapping of the
fp@1299: process data, the memory is managed internally by the library functions. As a
fp@1299: result, it is not possible to provide external memory for domains, like in the
fp@1299: kernel API. The corresponding functions are only available in kernelspace.
fp@1299: This is the only difference when using the application interface in userspace.
fp@1275: 
fp@1281: \subsection{Timing}
fp@1281: \label{sec:usertiming}
fp@1281: 
fp@1281: An interesting aspect is the timing of the userspace library calls compared to
fp@2515: those of the kernel API. \autoref{tab:usertiming} shows the call times and
fp@1281: standard deviancies of typical (and time-critical) API functions measured on
fp@1281: an Intel Pentium 4 M CPU with \unit{2.2}{\giga\hertz} and a standard 2.6.26
fp@1281: kernel.
fp@1281: 
fp@1281: \begin{table}[htbp]
fp@1281:   \centering
fp@1281:   \caption{Application Interface Timing Comparison}
fp@1281:   \label{tab:usertiming}
fp@1281:   \vspace{2mm}
fp@1281:   \begin{tabular}{l|c|c|c|c}
fp@1281: 
fp@1281:     &
fp@1281:     \multicolumn{2}{|c}{\textbf{Kernelspace}} &
fp@1281:     \multicolumn{2}{|c}{\textbf{Userspace}}  \\
fp@1281: 
fp@1281:     \textbf{Function} &
fp@1281:     $\mu(t)$ &
fp@1281:     $\sigma(t)$ &
fp@1281:     $\mu(t)$ &
fp@1281:     $\sigma(t)$ \\
fp@1281:     \hline
fp@1281: 
fp@1281:     \lstinline+ecrt_master_receive()+ &
fp@1281:     \unit{1.1}{\micro\second} &
fp@1281:     \unit{0.3}{\micro\second} &
fp@1281:     \unit{2.2}{\micro\second} &
fp@1281:     \unit{0.5}{\micro\second} \\
fp@1281: 
fp@1281:     \lstinline+ecrt_domain_process()+ &
fp@1281:     \unit{<0.1}{\micro\second} &
fp@1281:     \unit{<0.1}{\micro\second} &
fp@1281:     \unit{1.0}{\micro\second} &
fp@1281:     \unit{0.2}{\micro\second} \\
fp@1281: 
fp@1281:     \lstinline+ecrt_domain_queue()+ &
fp@1281:     \unit{<0.1}{\micro\second} &
fp@1281:     \unit{<0.1}{\micro\second} &
fp@1281:     \unit{1.0}{\micro\second} &
fp@1281:     \unit{0.1}{\micro\second} \\
fp@1281: 
fp@1281:     \lstinline+ecrt_master_send()+ &
fp@1281:     \unit{1.8}{\micro\second} &
fp@1281:     \unit{0.2}{\micro\second} &
fp@1281:     \unit{2.5}{\micro\second} &
fp@1281:     \unit{0.5}{\micro\second} \\
fp@1281: 
fp@1281:   \end{tabular}
fp@1281: \end{table}
fp@1281: 
fp@1282: The test results show, that for this configuration, the userspace API causes
fp@1282: about \unit{1}{\micro\second} additional delay for each function, compared to
fp@1282: the kernel API.
fp@1281: 
fp@1275: %------------------------------------------------------------------------------
fp@1275: 
fp@2515: \section{RTDM Interface}
fp@2515: \label{sec:rtdm}
fp@2515: 
fp@2515: When using the userspace interfaces of realtime extensions like Xenomai or
fp@2515: RTAI, the use of \textit{ioctl()} is not recommended, because it may disturb
fp@2515: realtime operation. To accomplish this, the Real-Time Device Model (RTDM)
fp@2515: \cite{rtdm} has been developed.  The master module provides an RTDM interface
fp@2515: (see \autoref{fig:arch}) in addition to the normal character device, if the
fp@2515: master sources were configured with \lstinline+--enable-rtdm+ (see
fp@2515: \autoref{sec:installation}).
fp@2515: 
fp@2515: To force an application to use the RTDM interface instead of the normal
fp@2515: character device, it has to be linked with the \textit{libethercat\_rtdm}
fp@2515: library instead of \textit{libethercat}. The use of the
fp@2515: \textit{libethercat\_rtdm} is transparent, so the EtherCAT header
fp@2515: \textit{ecrt.h} with the complete API can be used as usual.
fp@2515: 
fp@2515: To make the example in \autoref{lst:linker-user} use the RTDM library, the
fp@2515: linker command has to be altered as follows:
fp@2515: 
fp@2515: \begin{lstlisting}
fp@2515: gcc ethercat-with-rtdm.c -o ectest -I/opt/etherlab/include \
fp@2515:     -L/opt/etherlab/lib -lethercat_rtdm \
fp@2515:     -Wl,--rpath -Wl,/opt/etherlab/lib
fp@2515: \end{lstlisting}
fp@2515: 
fp@2515: %------------------------------------------------------------------------------
fp@2515: 
fp@1085: \section{System Integration}
fp@369: \label{sec:system}
fp@369: 
fp@1086: To integrate the EtherCAT master as a service into a running system, it comes
fp@2515: with an init script and a sysconfig file, that are described below. Modern
fp@2515: systems may be managed by systemd \cite{systemd}. Integration of the master
fp@2515: with systemd is described in \autoref{sec:systemd}.
fp@1086: 
fp@1086: \subsection{Init Script}
fp@369: \label{sec:init}
fp@369: \index{Init script}
fp@369: 
fp@1086: The EtherCAT master init script conforms to the requirements of the ``Linux
fp@1086: Standard Base'' (LSB\index{LSB}, \cite{lsb}). The script is installed to
fp@1202: \textit{etc/init.d/ethercat} below the installation prefix and has to be
fp@1202: copied (or better: linked) to the appropriate location (see
fp@2515: \autoref{sec:installation}), before the master can be inserted as a service.
fp@1289: Please note, that the init script depends on the sysconfig file described
fp@1289: below.
fp@1086: 
fp@1306: To provide service dependencies (i.\,e.\ which services have to be started
fp@1306: before others) inside the init script code, LSB defines a special comment
fp@1306: block. System tools can extract this information to insert the EtherCAT init
fp@1306: script at the correct place in the startup sequence:
fp@1086: 
fp@1086: \lstinputlisting[firstline=38,lastline=48]
fp@1086:     {../script/init.d/ethercat}
fp@1086: 
fp@1202: \subsection{Sysconfig File}
fp@1086: \label{sec:sysconfig}
fp@1086: \index{Sysconfig file}
fp@1086: 
fp@1086: For persistent configuration, the init script uses a sysconfig file installed
fp@1086: to \textit{etc/sysconfig/ethercat} (below the installation prefix), that is
fp@1086: mandatory for the init script. The sysconfig file contains all configuration
fp@1086: variables needed to operate one or more masters. The documentation is inside
fp@1086: the file and included below:
fp@1086: 
fp@1086: \lstinputlisting[numbers=left,firstline=9,basicstyle=\ttfamily\scriptsize]
fp@1086:     {../script/sysconfig/ethercat}
fp@1086: 
fp@2515: For systems managed by systemd (see \autoref{sec:systemd}), the sysconfig file
fp@2515: has moved to \lstinline+/etc/ethercat.conf+. Both versions are part of the
fp@2515: master sources and are meant to used alternatively.
fp@2515: 
fp@1202: \subsection{Starting the Master as a Service}
fp@1086: \label{sec:service}
fp@1086: \index{Service}
fp@1086: 
fp@1086: After the init script and the sysconfig file are placed into the right
fp@1086: location, the EtherCAT master can be inserted as a service. The different Linux
fp@1086: distributions offer different ways to mark a service for starting and stopping
fp@1086: in certain runlevels. For example, SUSE Linux provides the \textit{insserv}
fp@1086: command:
fp@1086: 
fp@1086: \begin{lstlisting}
fp@1086: # `\textbf{insserv ethercat}`
fp@369: \end{lstlisting}
fp@369: 
fp@369: The init script can also be used for manually starting and stopping
fp@369: the EtherCAT master. It has to be executed with one of the parameters
fp@379: \texttt{start}, \texttt{stop}, \texttt{restart} or \texttt{status}.
fp@369: 
fp@1085: \begin{lstlisting}[gobble=2]
fp@379:   # `\textbf{/etc/init.d/ethercat restart}`
fp@369:   Shutting down EtherCAT master                done
fp@369:   Starting EtherCAT master                     done
fp@369: \end{lstlisting}
fp@369: 
fp@2515: \subsection{Integration with systemd}
fp@2515: \label{sec:systemd}
fp@2515: \index{systemd}
fp@2515: 
fp@2515: Distributions using \textit{systemd} instead of the SysV init system are using service files to describe how a service is to be maintained. \autoref{lst:service} lists the master's service file:
fp@2515: 
fp@2515: \lstinputlisting[basicstyle=\ttfamily\footnotesize,caption=Service file,
fp@2515: label=lst:service]{../script/ethercat.service}
fp@2515: 
fp@2515: The \textit{ethercatctl} command is used to load and unload the master and
fp@2515: network driver modules in a similar way to the former init script
fp@2515: (\autoref{sec:init}). Because it is installed into the \textit{sbin/}
fp@2515: directory, it can also be used separately:
fp@2515: 
fp@2515: \begin{lstlisting}[gobble=2]
fp@2515:   # `\textbf{ethercatctl start}`
fp@2515: \end{lstlisting}
fp@2515: 
fp@2515: When using systemd and/or the \textit{ethercatctl} command, the master
fp@2515: configuration must be in \texttt{/etc/ethercat.conf} instead of
fp@2515: \texttt{/etc/sysconfig/ethercat}! The latter is ignored. The configuration
fp@2515: options are exactly the same.
fp@2515: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@1307: \section{Debug Interfaces}
fp@369: \label{sec:debug}
fp@1307: \index{Debug Interfaces}
fp@369: 
fp@1306: EtherCAT buses can always be monitored by inserting a switch between master
fp@1306: and slaves. This allows to connect another PC with a network monitor like
fp@1588: Wireshark~\cite{wireshark}, for example. It is also possible to listen to
fp@1588: local network interfaces on the machine running the EtherCAT master directly.
fp@2515: If the generic Ethernet driver (see \autoref{sec:generic-driver}) is used,
fp@1588: the network monitor can directly listen on the network interface connected to
fp@1588: the EtherCAT bus.
fp@1588: 
fp@2515: When using native Ethernet drivers (see \autoref{sec:native-drivers}), there
fp@1588: are no local network interfaces to listen to, because the Ethernet devices
fp@1588: used for EtherCAT are not registered at the network stack. For that case,
fp@1588: so-called ``debug interfaces'' are supported, which are virtual network
fp@1588: interfaces allowing to capture EtherCAT traffic with a network monitor (like
fp@1588: Wireshark or tcpdump) running on the master machine without using external
fp@1588: hardware. To use this functionality, the master sources have to be configured
fp@1588: with the \lstinline+--enable-debug-if+ switch (see
fp@2515: \autoref{sec:installation}).
fp@1307: 
fp@1307: Every EtherCAT master registers a read-only network interface per attached
fp@1308: physical Ethernet device. The network interfaces are named \textit{ecdbgmX}
fp@2515: for the main device, and \textit{ecdbgbX} for the backup device, where X is
fp@2515: the master index. The below listing shows a debug interface among some
fp@2515: standard network interfaces:
fp@1306: 
fp@1306: \begin{lstlisting}
fp@1306: # `\textbf{ip link}`
fp@1306: 1: lo: <LOOPBACK,UP> mtu 16436 qdisc noqueue
fp@1306:     link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
fp@1306: 4: eth0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop qlen 1000
fp@1307:     link/ether 00:13:46:3b:ad:d7 brd ff:ff:ff:ff:ff:ff
fp@1308: 8: ecdbgm0: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast
fp@1308:                                                  qlen 1000
fp@1306:     link/ether 00:04:61:03:d1:01 brd ff:ff:ff:ff:ff:ff
fp@1307: \end{lstlisting}
fp@1307: 
fp@1307: While a debug interface is enabled, all frames sent or received to or from the
fp@1307: physical device are additionally forwarded to the debug interface by the
fp@1308: corresponding master. Network interfaces can be enabled with the below
fp@1308: command:
fp@1306: 
fp@1306: \begin{lstlisting}
fp@1306: # `\textbf{ip link set dev ecdbgm0 up}`
fp@1306: \end{lstlisting}
fp@1306: 
fp@1306: Please note, that the frame rate can be very high. With an application
fp@1306: connected, the debug interface can produce thousands of frames per second.
fp@1306: 
fp@1309: \paragraph{Attention} The socket buffers needed for the operation of debug
fp@1588: interfaces have to be allocated dynamically. Some Linux realtime extensions
fp@1588: (like RTAI) do not allow this in realtime context!
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@1085: \chapter{Timing Aspects}
fp@369: \label{sec:timing}
fp@369: 
fp@1085: Although EtherCAT's timing is highly deterministic and therefore timing issues
fp@1085: are rare, there are a few aspects that can (and should be) dealt with.
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@1202: \subsection{Application Interface Profiling}
fp@369: \label{sec:timing-profile}
fp@1204: \index{Profiling}
fp@1204: % FIXME
fp@369: 
fp@1085: One of the most important timing aspects are the execution times of the
fp@1204: application interface functions, that are called in cyclic context. These
fp@1085: functions make up an important part of the overall timing of the application.
fp@1085: To measure the timing of the functions, the following code was used:
fp@369: 
fp@369: \begin{lstlisting}[gobble=2,language=C]
fp@369:   c0 = get_cycles();
fp@369:   ecrt_master_receive(master);
fp@369:   c1 = get_cycles();
fp@369:   ecrt_domain_process(domain1);
fp@369:   c2 = get_cycles();
fp@369:   ecrt_master_run(master);
fp@369:   c3 = get_cycles();
fp@369:   ecrt_master_send(master);
fp@369:   c4 = get_cycles();
fp@369: \end{lstlisting}
fp@369: 
fp@1085: Between each call of an interface function, the CPU timestamp counter is read.
fp@1085: The counter differences are converted to \micro\second\ with help of the
fp@1085: \lstinline+cpu_khz+ variable, that contains the number of increments per
fp@1085: \milli\second.
fp@1085: 
fp@1085: For the actual measuring, a system with a \unit{2.0}{\giga\hertz} CPU was used,
fp@1085: that ran the above code in an RTAI thread with a period of
fp@1085: \unit{100}{\micro\second}. The measuring was repeated $n = 100$ times and the
fp@2515: results were averaged. These can be seen in \autoref{tab:profile}.
fp@369: 
fp@369: \begin{table}[htpb]
fp@369:   \centering
fp@1204:   \caption{Profiling of an Application Cycle on a \unit{2.0}{\giga\hertz}
fp@1085:   Processor}
fp@369:   \label{tab:profile}
fp@369:   \vspace{2mm}
fp@369:   \begin{tabular}{l|r|r}
fp@1085:     Element & Mean Duration [\second] & Standard Deviancy [\micro\second] \\
fp@369:     \hline
fp@369:     \textit{ecrt\_master\_receive()} & 8.04 & 0.48\\
fp@369:     \textit{ecrt\_domain\_process()} & 0.14 & 0.03\\
fp@369:     \textit{ecrt\_master\_run()} & 0.29 & 0.12\\
fp@369:     \textit{ecrt\_master\_send()} & 2.18 & 0.17\\ \hline
fp@369:     Complete Cycle & 10.65 & 0.69\\ \hline
fp@369:   \end{tabular}
fp@369: \end{table}
fp@369: 
fp@1085: It is obvious, that the functions accessing hardware make up the
fp@369: lion's share. The \textit{ec\_master\_receive()} executes the ISR of
fp@369: the Ethernet device, analyzes datagrams and copies their contents into
fp@369: the memory of the datagram objects. The \textit{ec\_master\_send()}
fp@369: assembles a frame out of different datagrams and copies it to the
fp@369: hardware buffers. Interestingly, this makes up only a quarter of the
fp@369: receiving time.
fp@369: 
fp@1085: The functions that only operate on the masters internal data structures are
fp@1085: very fast ($\Delta t < \unit{1}{\micro\second}$). Interestingly the runtime of
fp@1085: \textit{ec\_domain\_process()} has a small standard deviancy relative to the
fp@1085: mean value, while this ratio is about twice as big for
fp@1085: \textit{ec\_master\_run()}: This probably results from the latter function
fp@1085: having to execute code depending on the current state and the different state
fp@1085: functions are more or less complex.
fp@1085: 
fp@1085: For a realtime cycle makes up about \unit{10}{\micro\second}, the theoretical
fp@1085: frequency can be up to \unit{100}{\kilo\hertz}. For two reasons, this frequency
fp@369: keeps being theoretical:
fp@369: 
fp@369: \begin{enumerate}
fp@1085: 
fp@1085: \item The processor must still be able to run the operating system between the
fp@1085: realtime cycles.
fp@1085: 
fp@1085: \item The EtherCAT frame must be sent and received, before the next realtime
fp@1085: cycle begins. The determination of the bus cycle time is difficult and covered
fp@2515: in \autoref{sec:timing-bus}.
fp@1085: 
fp@369: \end{enumerate}
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@369: \subsection{Bus Cycle Measuring}
fp@369: \label{sec:timing-bus}
fp@369: \index{Bus cycle}
fp@369: 
fp@1297: For measuring the time, a frame is ``on the wire'', two timestamps must be
fp@1297: taken:
fp@369: 
fp@369: \begin{enumerate}
fp@1297: 
fp@1297: \item The time, the Ethernet hardware begins with physically sending the
fp@1297: frame.
fp@1297: 
fp@1297: \item The time, the frame is completely received by the Ethernet hardware.
fp@1297: 
fp@369: \end{enumerate}
fp@369: 
fp@369: Both times are difficult to determine. The first reason is, that the
fp@1297: interrupts are disabled and the master is not notified, when a frame is sent
fp@1297: or received (polling would distort the results). The second reason is, that
fp@1297: even with interrupts enabled, the time from the event to the notification is
fp@1297: unknown. Therefore the only way to confidently determine the bus cycle time is
fp@1297: an electrical measuring.
fp@369: 
fp@1204: Anyway, the bus cycle time is an important factor when designing realtime
fp@1204: code, because it limits the maximum frequency for the cyclic task of the
fp@1204: application.  In practice, these timing parameters are highly dependent on the
fp@1204: hardware and often a trial and error method must be used to determine the
fp@1204: limits of the system.
fp@1085: 
fp@1085: The central question is: What happens, if the cycle frequency is too high? The
fp@1297: answer is, that the EtherCAT frames that have been sent at the end of the
fp@1297: cycle are not yet received, when the next cycle starts.  First this is noticed
fp@1297: by \textit{ecrt\_domain\_process()}, because the working counter of the
fp@1297: process data datagrams were not increased. The function will notify the user
fp@1297: via Syslog\footnote{To limit Syslog output, a mechanism has been implemented,
fp@1297: that outputs a summarized notification at maximum once a second.}. In this
fp@1297: case, the process data keeps being the same as in the last cycle, because it
fp@1297: is not erased by the domain. When the domain datagrams are queued again, the
fp@1297: master notices, that they are already queued (and marked as sent). The master
fp@1297: will mark them as unsent again and output a warning, that datagrams were
fp@1085: ``skipped''.
fp@1085: 
fp@1085: On the mentioned \unit{2.0}{\giga\hertz} system, the possible cycle frequency
fp@1085: can be up to \unit{25}{\kilo\hertz} without skipped frames. This value can
fp@1297: surely be increased by choosing faster hardware. Especially the RealTek
fp@1297: network hardware could be replaced by a faster one. Besides, implementing a
fp@1297: dedicated ISR for EtherCAT devices would also contribute to increasing the
fp@1297: latency. These are two points on the author's to-do list.
fp@1085: 
fp@1085: %------------------------------------------------------------------------------
fp@1085: 
fp@1085: \chapter{Installation}
fp@1085: \label{sec:installation}
fp@1085: \index{Master!Installation}
fp@369: 
fp@1772: \section{Getting the Software}
fp@1772: \label{sec:getting}
fp@1772: 
fp@1772: There are several ways to get the master software:
fp@1772: 
fp@1772: \begin{enumerate}
fp@1772: 
fp@1772: \item An official release (for example \masterversion), can be downloaded from
fp@1772: the master's website\footnote{\url{http://etherlab.org/en/ethercat/index.php}}
fp@1772: at~the EtherLab project~\cite{etherlab} as a tarball.
fp@1772: 
fp@1772: \item The most recent development revision (and moreover any other revision)
fp@1772: can be obtained via the Mercurial~\cite{mercurial} repository on the master's
fp@1772: project page on
fp@1772: SourceForge.net\footnote{\url{http://sourceforge.net/projects/etherlabmaster}}.
fp@1772: The whole repository can be cloned with the command
fp@1772: 
fp@1772: \begin{lstlisting}[breaklines=true]
fp@1772: hg clone http://etherlabmaster.hg.sourceforge.net/hgweb/etherlabmaster/etherlabmaster `\textit{local-dir}`
fp@1772: \end{lstlisting}
fp@1772: 
fp@1772: \item Without a local Mercurial installation, tarballs of arbitrary revisions
fp@1772: can be downloaded via the ``bz2'' links in the browsable repository
fp@1772: pages\footnote{\url{http://etherlabmaster.hg.sourceforge.net/hgweb/etherlabmaster/etherlabmaster}}.
fp@1772: 
fp@1772: \end{enumerate}
fp@1772: 
fp@1202: \section{Building the Software}
fp@1094: 
fp@1772: After downloading a tarball or cloning the repository as described in
fp@2515: \autoref{sec:getting}, the sources have to be prepared and configured for the
fp@1772: build process.
fp@1772: 
fp@1772: When a tarball was downloaded, it has to be extracted with the following
fp@1772: commands:
fp@369: 
fp@1085: \begin{lstlisting}[gobble=2]
fp@1202:   $ `\textbf{tar xjf ethercat-\masterversion.tar.bz2}`
fp@1202:   $ `\textbf{cd ethercat-\masterversion/}`
fp@374: \end{lstlisting}
fp@374: 
fp@1772: The software configuration is managed with Autoconf~\cite{autoconf} so the
fp@1772: released versions contain a \lstinline+configure+ shell script, that has to be
fp@1772: executed for configuration (see below).
fp@1772: 
fp@1772: \paragraph{Bootstrap} When downloading or cloning directly from the
fp@1772: repository, the \lstinline+configure+ script does not yet exist. It can be
fp@1772: created via the \lstinline+bootstrap.sh+ script in the master sources. The
fp@1772: autoconf and automake packages are required for this.
fp@1772: 
fp@1772: \paragraph{Configuration and Build} The configuration and the build process
fp@1772: follow the below commands:
fp@369: 
fp@1085: \begin{lstlisting}[gobble=2]
fp@1202:   $ `\textbf{./configure}`
fp@1202:   $ `\textbf{make}`
fp@1202:   $ `\textbf{make modules}`
fp@374: \end{lstlisting}
fp@374: 
fp@2515: \autoref{tab:config} lists important configuration switches and options.
fp@1085: 
fp@2297: \begin{longtable}{l|p{.4\textwidth}|l}
fp@2297:   \caption{Configuration options}\rule[-5ex]{0mm}{0mm}
fp@2297:   \label{tab:config}\\
fp@2297: 
fp@2297: \textbf{Option/Switch} & \textbf{Description} & \textbf{Default}\\\hline
fp@2297: \endfirsthead
fp@2297: 
fp@2297: \textbf{Option/Switch} & \textbf{Description} & \textbf{Default}\\\hline
fp@2297: \endhead
fp@1085: 
fp@1085: \lstinline+--prefix+ & Installation prefix & \textit{/opt/etherlab}\\
fp@1085: 
fp@1085: \lstinline+--with-linux-dir+ & Linux kernel sources & Use running kernel\\
fp@1085: 
fp@2297: \lstinline+--with-module-dir+ & Subdirectory in the kernel module tree, where
fp@2297: the EtherCAT kernel modules shall be installed. & \textit{ethercat}\\
fp@2297: 
fp@2297: \hline
fp@2297: 
fp@2297: \lstinline+--enable-generic+ & Build the generic Ethernet driver (see
fp@2515: \autoref{sec:generic-driver}). & yes\\
fp@2297: 
fp@2297: \lstinline+--enable-8139too+ & Build the 8139too driver & yes\\
fp@2297: 
fp@2297: \lstinline+--with-8139too-kernel+ & 8139too kernel & $\dagger$\\
fp@2297: 
fp@2297: \lstinline+--enable-e100+ & Build the e100 driver & no\\
fp@2297: 
fp@2297: \lstinline+--with-e100-kernel+ & e100 kernel & $\dagger$\\
fp@2297: 
fp@2297: \lstinline+--enable-e1000+ & Enable e1000 driver & no\\
fp@2297: 
fp@2297: \lstinline+--with-e1000-kernel+ & e1000 kernel & $\dagger$\\
fp@2297: 
fp@2364: \lstinline+--enable-e1000e+ & Enable e1000e driver & no\\
fp@2364: 
fp@2364: \lstinline+--with-e1000e-kernel+ & e1000e kernel & $\dagger$\\
fp@2364: 
fp@2297: \lstinline+--enable-r8169+ & Enable r8169 driver & no\\
fp@2297: 
fp@2297: \lstinline+--with-r8169-kernel+ & r8169 kernel & $\dagger$\\
fp@2297: 
fp@2691: \lstinline+--enable-ccat+ & Enable ccat driver (independent of kernel version)
fp@2691: & no\\
fp@2691: 
fp@2691: \lstinline+--enable-igb+ & Enable igb driver & no\\
fp@2691: 
fp@2691: \lstinline+--with-igb-kernel+ & igb kernel & $\dagger$\\
fp@2691: 
fp@2297: \hline
fp@2297: 
fp@2691: \lstinline+--enable-kernel+ & Build the master kernel modules & yes\\
fp@2691: 
fp@2515: \lstinline+--enable-rtdm+ & Create the RTDM interface (RTAI or Xenomai
fp@2515: directory needed, see below) & no\\
fp@2515: 
fp@2515: \lstinline+--with-rtai-dir+ & RTAI path (for RTAI examples and RTDM interface)
fp@2515: & \\
fp@2515: 
fp@2515: \lstinline+--with-xenomai-dir+ & Xenomai path (for Xenomai examples and RTDM
fp@2515: interface) & \\
fp@2515: 
fp@2515: \lstinline+--with-devices+ & Number of Ethernet devices for redundant
fp@2515: operation ($>1$ switches redundancy on) & 1\\
fp@1085: 
fp@2297: \lstinline+--enable-debug-if+ & Create a debug interface for each master & no\\
fp@2297: 
fp@2297: \lstinline+--enable-debug-ring+ & Create a debug ring to record frames & no\\
fp@2297: 
fp@2297: \lstinline+--enable-eoe+ & Enable EoE support & yes\\
fp@2297: 
fp@2297: \lstinline+--enable-cycles+ & Use CPU timestamp counter. Enable this on Intel
fp@2297: architecture to get finer timing calculation. & no\\
fp@2297: 
fp@2297: \lstinline+--enable-hrtimer+ & Use high-resolution timer to let the master
fp@2297: state machine sleep between sending frames. & no\\
fp@2297: 
fp@2691: \lstinline+--enable-regalias+ & Read alias address from register & no\\
fp@1085: 
fp@1588: \lstinline+--enable-tool+ & Build the command-line tool ``ethercat'' (see
fp@2691: \autoref{sec:tool}) & yes\\
fp@2691: 
fp@2691: \lstinline+--enable-userlib+ & Build the userspace library & yes\\
fp@2691: 
fp@2691: \lstinline+--enable-tty+ & Build the TTY driver & no\\
fp@2297: 
fp@2297: \lstinline+--enable-wildcards+ & Enable \textit{0xffffffff} to be wildcards
fp@2691: for vendor ID and product code & no\\
fp@1085: 
fp@2515: \lstinline+--enable-sii-assign+ & Enable assigning SII access to the PDI layer
fp@2691: during slave configuration & no\\
fp@2691: 
fp@2691: \lstinline+--enable-rt-syslog+ & Enable syslog statements in realtime context &
fp@2691: yes\\
fp@2515: 
fp@1085: \hline
fp@1085: 
fp@2297: \end{longtable}
fp@1085: 
fp@1085: \begin{description}
fp@1085: 
fp@1085: \item[$\dagger$] If this option is not specified, the kernel version to use is
fp@1085: extracted from the Linux kernel sources.
fp@1085: 
fp@1085: \end{description}
fp@1085: 
fp@1202: \section{Building the Interface Documentation}
fp@1094: \label{sec:gendoc}
fp@1094: 
fp@1094: The source code is documented using Doxygen~\cite{doxygen}. To build the HTML
fp@1202: documentation, the Doxygen software has to be installed. The below command
fp@1095: will generate the documents in the subdirectory \textit{doxygen-output}:
fp@1094: 
fp@1094: \begin{lstlisting}
fp@1094: $ `\textbf{make doc}`
fp@1094: \end{lstlisting}
fp@1094: 
fp@1204: The interface documentation can be viewed by pointing a browser to the file
fp@1293: \textit{doxygen-output/html/index.html}. The functions and data structures of
fp@1293: the application interface a covered by an own module ``Application
fp@1293: Interface''.
fp@1202: 
fp@1202: \section{Installing the Software}
fp@1094: 
fp@1106: The below commands have to be entered as \textit{root}: The first one will
fp@1275: install the EtherCAT header, init script, sysconfig file and the userspace
fp@1202: tool to the prefix path. The second one will install the kernel modules to the
fp@1202: kernel's modules directory. The final \lstinline+depmod+ call is necessary to
fp@1202: include the kernel modules into the \textit{modules.dep} file to make it
fp@2421: available to the \lstinline+modprobe+ command, used in the init script.
fp@369: 
fp@1094: \begin{lstlisting}
fp@1106: # `\textbf{make install}`
fp@1094: # `\textbf{make modules\_install}`
fp@369: \end{lstlisting}
fp@369: 
fp@1095: If the target kernel's modules directory is not under \textit{/lib/modules}, a
fp@1095: different destination directory can be specified with the \lstinline+DESTDIR+
fp@1095: make variable. For example:
fp@487: 
fp@1094: \begin{lstlisting}
fp@1094: # `\textbf{make DESTDIR=/vol/nfs/root modules\_install}`
fp@487: \end{lstlisting}
fp@487: 
fp@487: This command will install the compiled kernel modules to
fp@487: \textit{/vol/nfs/root/lib/modules}, prepended by the kernel release.
fp@487: 
fp@1085: If the EtherCAT master shall be run as a service\footnote{Even if the EtherCAT
fp@1085: master shall not be loaded on system startup, the use of the init script is
fp@2515: recommended for manual (un-)loading.} (see \autoref{sec:system}), the init
fp@2515: script and the sysconfig file (or the systemd service file, respectively) have
fp@2515: to be copied (or linked) to the appropriate locations. The below example is
fp@2515: suitable for SUSE Linux. It may vary for other distributions.
fp@1085: 
fp@1107: % FIXME relative ln -s?
fp@1094: \begin{lstlisting}
fp@1094: # `\textbf{cd /opt/etherlab}`
fp@1094: # `\textbf{cp etc/sysconfig/ethercat /etc/sysconfig/}`
fp@1094: # `\textbf{ln -s etc/init.d/ethercat /etc/init.d/}`
fp@1094: # `\textbf{insserv ethercat}`
fp@374: \end{lstlisting}
fp@374: 
fp@376: Now the sysconfig file \texttt{/etc/sysconfig/ethercat} (see
fp@2515: \autoref{sec:sysconfig}), or the configuration file
fp@2515: \textit{/etc/ethercat.conf}, if using systemd, has to be customized. The
fp@2515: minimal customization is to set the \lstinline+MASTER0_DEVICE+ variable to the
fp@2515: MAC address of the Ethernet device to use (or \lstinline+ff:ff:ff:ff:ff:ff+ to
fp@2515: use the first device offered) and selecting the driver(s) to load via the
fp@1085: \lstinline+DEVICE_MODULES+ variable.
fp@369: 
fp@1297: After the basic configuration is done, the master can be started with the
fp@1297: below command:
fp@369: 
fp@1094: \begin{lstlisting}
fp@1094: # `\textbf{/etc/init.d/ethercat start}`
fp@369: \end{lstlisting}
fp@369: 
fp@2515: When using systemd, the following command can be used alternatively:
fp@2515: 
fp@2515: \begin{lstlisting}
fp@2515: # `\textbf{ethercatctl start}`
fp@2515: \end{lstlisting}
fp@2515: 
fp@1214: At this time, the operation of the master can be observed by viewing the
fp@1214: Syslog\index{Syslog} messages, which should look like the ones below. If
fp@1214: EtherCAT slaves are connected to the master's EtherCAT device, the activity
fp@1214: indicators should begin to flash.
fp@369: 
fp@369: \begin{lstlisting}[numbers=left]
fp@1085: EtherCAT: Master driver `\masterversion`
fp@1085: EtherCAT: 1 master waiting for devices.
fp@1085: EtherCAT Intel(R) PRO/1000 Network Driver - version 6.0.60-k2
fp@1085: Copyright (c) 1999-2005 Intel Corporation.
fp@1085: PCI: Found IRQ 12 for device 0000:01:01.0
fp@1085: PCI: Sharing IRQ 12 with 0000:00:1d.2
fp@1085: PCI: Sharing IRQ 12 with 0000:00:1f.1
fp@1085: EtherCAT: Accepting device 00:0E:0C:DA:A2:20 for master 0.
fp@1085: EtherCAT: Starting master thread.
fp@1085: ec_e1000: ec0: e1000_probe: Intel(R) PRO/1000 Network
fp@1085:           Connection
fp@1085: ec_e1000: ec0: e1000_watchdog_task: NIC Link is Up 100 Mbps
fp@1085:           Full Duplex
fp@1085: EtherCAT: Link state changed to UP.
fp@1085: EtherCAT: 7 slave(s) responding.
fp@1085: EtherCAT: Slave states: PREOP.
fp@1085: EtherCAT: Scanning bus.
fp@1085: EtherCAT: Bus scanning completed in 431 ms.
fp@369: \end{lstlisting}
fp@369: 
fp@369: \begin{description}
fp@1085: 
fp@1085: \item[\linenum{1} -- \linenum{2}] The master module is loading, and one master
fp@1085: is initialized.
fp@1085: 
fp@1085: \item[\linenum{3} -- \linenum{8}] The EtherCAT-capable e1000 driver is
fp@1085: loading. The master accepts the device with the address
fp@1085: \lstinline+00:0E:0C:DA:A2:20+.
fp@1085: 
fp@1085: \item[\linenum{9} -- \linenum{16}] The master goes to idle phase, starts its
fp@1085: state machine and begins scanning the bus.
fp@1085: 
fp@369: \end{description}
fp@369: 
fp@1214: \section{Automatic Device Node Creation}
fp@1214: \label{sec:autonode}
fp@1214: 
fp@2515: The \lstinline+ethercat+ command-line tool (see \autoref{sec:tool})
fp@1214: communicates with the master via a character device. The corresponding device
fp@1289: nodes are created automatically, if the udev daemon is running.  Note, that on
fp@1289: some distributions, the \lstinline+udev+ package is not installed by default.
fp@1214: 
fp@1214: The device nodes will be created with mode \lstinline+0660+ and group
fp@1292: \lstinline+root+ by default. If ``normal'' users shall have reading access, a
fp@1292: udev rule file (for example \textit{/etc/udev/rules.d/99-EtherCAT.rules}) has
fp@1292: to be created with the following contents:
fp@1214: 
fp@1214: \begin{lstlisting}
fp@1214: KERNEL=="EtherCAT[0-9]*", MODE="0664"
fp@1214: \end{lstlisting}
fp@1214: 
fp@1214: After the udev rule file is created and the EtherCAT master is restarted with
fp@1214: \lstinline[breaklines=true]+/etc/init.d/ethercat restart+, the device node
fp@1214: will be automatically created with the desired rights:
fp@1214: 
fp@1214: \begin{lstlisting}
fp@1214: # `\textbf{ls -l /dev/EtherCAT0}`
fp@1214: crw-rw-r--  1 root root 252, 0 2008-09-03 16:19 /dev/EtherCAT0
fp@1214: \end{lstlisting}
fp@1214: 
fp@2515: Now, the \lstinline+ethercat+ tool can be used (see \autoref{sec:tool}) even
fp@1289: as a non-root user.
fp@1214: 
fp@1306: If non-root users shall have writing access, the following udev rule can be
fp@1306: used instead:
fp@1306: 
fp@1306: \begin{lstlisting}
fp@1306: KERNEL=="EtherCAT[0-9]*", MODE="0664", GROUP="users"
fp@1306: \end{lstlisting}
fp@1306: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@369: \begin{thebibliography}{99}
fp@1094: 
fp@1094: \bibitem{etherlab} Ingenieurgemeinschaft IgH: EtherLab -- Open Source Toolkit
fp@1094: for rapid realtime code generation under Linux with Simulink/RTW and EtherCAT
fp@1094: technology. \url{http://etherlab.org/en}, 2008.
fp@1094: 
fp@369: \bibitem{dlspec} IEC 61158-4-12: Data-link Protocol Specification.
fp@1095: International Electrotechnical Commission (IEC), 2005.
fp@1094: 
fp@1094: \bibitem{alspec} IEC 61158-6-12: Application Layer Protocol Specification.
fp@1095: International Electrotechnical Commission (IEC), 2005.
fp@1094: 
fp@1094: \bibitem{gpl} GNU General Public License, Version 2.
fp@1275: \url{http://www.gnu.org/licenses/gpl-2.0.html}. October~15, 2008.
fp@1275: 
fp@1275: \bibitem{lgpl} GNU Lesser General Public License, Version 2.1.
fp@1275: \url{http://www.gnu.org/licenses/old-licenses/lgpl-2.1.html}. October~15,
fp@1275: 2008.
fp@1094: 
fp@1094: \bibitem{lsb} Linux Standard Base.
fp@1306: \url{http://www.linuxfoundation.org/en/LSB}. August~9, 2006.
fp@1094: 
fp@2515: \bibitem{systemd} systemd System and Service Manager
fp@2515: \url{http://freedesktop.org/wiki/Software/systemd}. January~18, 2013.
fp@2515: 
fp@1094: \bibitem{wireshark} Wireshark. \url{http://www.wireshark.org}. 2008.
fp@1094: 
fp@1306: \bibitem{automata} {\it Hopcroft, J.\,E.\ / Ullman, J.\,D.}: Introduction to
fp@1094: Automata Theory, Languages and Computation. Adison-Wesley, Reading,
fp@1094: Mass.~1979.
fp@1094: 
fp@1306: \bibitem{fsmmis} {\it Wagner, F.\ / Wolstenholme, P.}: State machine
fp@1094: misunderstandings. In: IEE journal ``Computing and Control Engineering'',
fp@1094: 2004.
fp@1094: 
fp@1094: \bibitem{rtai} RTAI. The RealTime Application Interface for Linux from DIAPM.
fp@1588: \url{https://www.rtai.org}, 2010.
fp@1588: 
fp@1588: \bibitem{rt-preempt} RT PREEMPT HOWTO.
fp@1588: \url{http://rt.wiki.kernel.org/index.php/RT_PREEMPT_HOWTO}, 2010.
fp@1094: 
fp@1094: \bibitem{doxygen} Doxygen. Source code documentation generator tool.
fp@1094: \url{http://www.stack.nl/~dimitri/doxygen}, 2008.
fp@1094: 
fp@1772: \bibitem{mercurial} Mercurial SCM. \url{http://mercurial.selenic.com}, 2010.
fp@1772: 
fp@1772: \bibitem{autoconf} Autoconf -- GNU Project -- Free Software Foundation (FSF).
fp@1772: \url{http://www.gnu.org/software/autoconf}, 2010.
fp@1772: 
fp@1917: \bibitem{soespec} IEC 61800-7-304: Adjustable speed electrical power drive
fp@1917: systems - Part 7-300: Generic interface and use of profiles for power drive
fp@1917: systems - Mapping of profiles to network technologies. International
fp@1917: Electrotechnical Commission (IEC), 2007.
fp@1917: 
fp@2515: \bibitem{rtdm} {\it J. Kiszka}: The Real-Time Driver Model and First
fp@2515: Applications.
fp@2515: \url{http://svn.gna.org/svn/xenomai/tags/v2.4.0/doc/nodist/pdf/RTDM-and-Applications.pdf},
fp@2515: 2013.
fp@2515: 
fp@369: \end{thebibliography}
fp@369: 
fp@917: \printnomenclature
fp@369: \addcontentsline{toc}{chapter}{\nomname}
fp@369: \markleft{\nomname}
fp@369: 
fp@369: \printindex
fp@369: \markleft{Index}
fp@369: 
fp@369: %------------------------------------------------------------------------------
fp@369: 
fp@369: \end{document}
fp@369: 
fp@369: %------------------------------------------------------------------------------