The embedded software industry is in the midst of a major revolution.
Tremendous amount of new development lies ahead. This new software
needs an actual architecture that is safer, more extensible, and easier to
understand than the usual “shared-state concurrency and blocking” based
on a traditional Real-Time Operating System (RTOS).
Quantum Leaps' QP™ real-time embedded frameworks (RTEFs) and the
QM™ model-based design (MBD) tool provide such a modern, eventdriven architecture based on active objects (actors), hierarchical state
machines (UML statecharts), software tracing, unit testing, model-based
engineering and automatic code generation.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
You can't just look at the QP™ real-time embedded frameworks and the
QM™ model-based design tool as a collection of features, because some
of the features will make no sense in isolation. You can only use these
powerful tools effectively if you are thinking about the overall architecture
and design of your system, not simply coding. And to understand the
tools and the underlying concepts that way, you must understand the
problems with programming real-time and embedded (RTE) systems in
general.
Therefore this presentation starts with discussing problems inherent in
RTE systems, why they are problems, and how active object frameworks
and hierarchical state machines can help.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
Some of the most difficult problems with real-time and embedded (RTE)
programing are related to concurrent code execution
→ these problems are usually intermittent, subtle, hard-to-reproduce,
hard-to-isolate, hard-to-debug, and hard-to-remove
→ they pose the highest risk to the project schedule
#1 Problems due to sharing of resources:
●
Endemic to all shared-state systems (main+ISRs and RTOS)
●
The ripple-effects of preemption in shared-state systems:
→ Race conditions → failure (if unaddressed)
→ mutual exclusion → blocking → missed deadlines
#2 Problems caused by threads synchronization by blocking:
●
Endemic to most conventional RTOS
→ lack of responsiveness → more threads → more mutual exclusion
→ more blocking … → architectural decay
●
No really good options!
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
Experts in the field have learned to avoid shared-state concurrency and to
avoid blocking to synchronize their threads. Instead, experts apply the
following best practices of concurrent programming:
1. Keep data and resources encapsulated inside threads (“sharenothing” principle) and use events to share information
2. Communicate among threads asynchronously via event objects
→ Threads run truly independently, without blocking on each other
In other words, experts combine multi-threading with event-driven
programming:
→ Threads are organized as “message pumps” (event queue + event
loop)
→ Threads process one event at a time (Run-to-Completion, RTC)
→ Threads block only on empty queue and don't block anywhere else
Such event-driven, asynchronous, non-blocking, encapsulated threads
are called Active Objects (a.k.a. Actors)
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
The Active Object (Actor) pattern inherently supports and automatically
enforces the best practices of concurrent programming.
The Active Object pattern is valuable, because it dramatically improves
your ability to reason about your thread's code and operation by giving
you higher-level abstractions and idioms that raise the semantic level of
your program and let you express your intent more directly and safely,
thus improving your productivity.
The concept of autonomous software objects communicating by message
passing dates back to the 1970s (Carl Hewitt came up with Actors). In the
1990s, methodologies like ROOM adapted actors for real-time computing.
More recently, UML has introduced the concept of Active Objects that are
essentially synonymous with the ROOM actors. Today, the actor model is
all the rage in the enterprise computing. A number of actor programming
languages (e.g., Erlang, Scala, D) as well as actor libraries and
frameworks (e.g., Akka, Kilim, Jetlang) are in extensive use. In the realtime embedded space, active objects frameworks provide the backbone
of various modeling and code generation tools. Examples include: IBM
Rational Rhapsody (with OXF/SXF frameworks), National Instruments
LabVIEW (with LabVIEW Actor Framework), and QP™ frameworks from
Quantum Leaps.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
Most conventional RTOSes are not “event-driven”, because RTOSes are
based on blocking (while event-driven programming is all about notblocking). Also RTOSes don't provide the event abstraction (event objects
or messages carrying event signals and parameters) .
→ semaphores or event-flags RTOS primitives are not event instances.
But still, you can manually implement the Active Object pattern on top of a
conventional RTOS by self-imposing the following rules and conventions:
●
You define your own basic event data type, which carries the event
signal and can be extended to carry event parameters
●
Each thread owns an event queue capable of storing your event
objects (could be message queue in RTOS)
●
The treads communicate only by posting events to their queues
→ asynchronous communication without blocking
●
Each thread is organized as a “message pump” (queue + event loop)
→ thread blocks only when its queue is empty, and does not block
when processing an event
●
All data and resources (e.g., peripherals) are bound to threads and can
be accessed only from the owner thread (encapsulation for
concurrency)
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
A framework is a universal, reusable software architecture for
development of specific class of software (e.g., real-time embedded
control systems).
The most important characteristics of a framework is that code provided
by the application developers is being called by the framework, which
results in inversion of control compared to using a toolkit such as a
conventional RTOS.
For example, when you use an RTOS, you write the main body of each
thread and you call the code from the RTOS (such as a semaphore, time
delay, etc.) In contrast, when you use a framework, you reuse the whole
architecture and write the code that it calls (inversion of control).
The inversion of control is very characteristic to virtually all event-driven
systems. It is the main reason for the architectural-reuse and enforcement
of the best practices, as opposed to re-inventing them for each project at
hand. It also leads to a much higher conceptual integrity of the final
product and dramatic improvement of developer's productivity.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
Even though a conventional RTOS can be used to implement eventdriven Active Objects, you must be very careful not to use most of the
RTOS mechanisms, because they block (e.g., semaphores, delays, etc.)
At the same time, a conventional RTOS does not provide much of support
for event-driven programming, which you need to create yourself.
This is all because conventional RTOSes are designed for the sequential
programming model, where you block and wait in-line for the
occurrence of an event. For example, consider the venerable “Blinky”
implementation with delay() functions called to wait in-line.
Event-driven programing represents a paradigm shift, where each event
(such as timeout event) is processed to completion and the handler
returns to the framework, without blocking.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
Another big class of problems in programming real-time and embedded
(RTE) systems arises from the difficulties in responding to events, which
often leads to convoluted program logic (a.k.a. “spaghetti code”):
●
●
●
●
●
●
The response depends on both: the event type and the internal state
of the system
The internal state (history) of the system is represented ad hoc as
multitude flags and variables
Convoluted IF-THEN-ELSE-SWICH logic to test the flags and variables
→ spaghetti code (a.k.a. BBM = Big Ball of Mud)
Multitude flags and variables → inconsistencies
Multitude of paths through the code → hard to understand code
→ hard to test with high cyclomatic complexity
Fragile code → fear of “breaking the logic” → more flags and variables
→ architectural decay
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
_
_
_
_
_
_
_
_
;
_
_
_
_
_
;
Finite State Machines—the best known “spaghetti reducers”
●
“State” captures only the relevant aspects of the system's history and
ignores all irrelevant aspects.
For example, a computer keyboard can be in “default” or “caps_locked”
state, where it generates lower-case or upper-case characters. Only
pressing CAPS_LOCK toggles between these states. Pressing other keys
is irrelevant.
State machines are a natural fit for event-driven programming,
●
State machine is exactly designed to process an event quickly and
return to the caller
●
The context of the system between calls is represented by the single
state-variable ,
→ much more efficient than in sequential programming, where the
context is represented by the whole call stack.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
(a)
(b)
s1
E
1 ;
s2
E
2 ;
E
3 ;
s3
State diagrams (statecharts) should not be confused with flowcharts
The main difference is that state machines need events to perform any
actions and possibly change state (execute transitions).
Flowcharts don't need events. They progress from one stage of processing
to another upon completion of processing.
Graphically, flowcharts reverse the sense of nodes and arcs in the
diagram. In state machines, processing is associated with arcs. In
flowchart with nodes.
The main difference boils down to the different programming paradigms
represented:
●
Statecharts correspond to event-driven programming paradigm
●
Flowcharts correspond to the sequential programming paradigm
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
A lot of state machine examples in various magazines, books, and online
pertain to input-driven (a.k.a. polled) state machines, as opposed to
truly event-driven state machines introduced earlier.
Input-driven state machines are NOT driven by events. Instead, an inputdriven state machine code is called “as fast as possible”, or “periodically”
from while(1) loops to poll for the events. In the code, you can easily
recognize such input-driven state machines by the if() statements that check
the inputs in each state and only after discovering the right combination of
inputs, they execute actions. In the diagrams, you can easily recognize
input-driven state machines by the fact that state transitions are NOT
labeled by events, but rather by guard conditions. The brackets around
those guard conditions, which are required by the UML state machine
notation, are often missing, but you typically can recognize that the labels
are conditions, especially when you see logic operators, like and/or.
The main problems with input-driven (polled) state machines are that they
might miss changes in the inputs (if sampling is too slow) or recognize the
changes in different order, depending on the timing (race conditions).
They are also wasteful, as they need to run “all the time”. Finally, it is hard
to apply hierarchical event processing, because there are no explicit
events.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
C
(a)
(b)
1
C
1
_0_9, C
C
_0_9
C
2
2
_0_9
_0_9
But even the event-driven FSMs have a major shortcoming known as the
“state and transition explosion”. For example, if you try to represent the
behavior of a simple pocket calculator with a traditional FSM, you'll notice
that many events (e.g., the Clear or Off button presses) are handled
identically in many states. A conventional FSM, has no means of capturing
such a commonality and requires repeating the same actions and
transitions in many states.
Hierarchical State Machines solve this problem by introducing state
nesting with the following semantics: If a system is in the nested state, for
example "result" (called the substate), it also (implicitly) is in the
surrounding state "on" (called the superstate). This state machine will
attempt to handle any event in the context of the substate, which
conceptually is at the lower level of the hierarchy. However, if the substate
"result" does not prescribe how to handle the event, the event is not quietly
discarded as in a traditional "flat" state machine; rather, it is automatically
handled at the higher level context of the superstate "on".
State nesting enables substates to reuse the transitions and actions
defined already in superstates. The substates need only define the
differences from the superstates (programming-by-difference).
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
This section introduces the QP active object frameworks specifically
designed for real-time embedded (RTE) systems, such as single-chip
microcontrollers.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
QP™ is a family of lightweight real-time embedded frameworks (RTEFs)
specifically designed for deeply embedded real-time systems, such as
single chip MCUs (8-, 16-, and 32-bit).
The QP family consists of QP/C, QP/C++, and QP-nano frameworks,
which are all strictly quality controlled, thoroughly documented, and
available in full source code.
The behavior of active objects is specified in QP by means of hierarchical
state machines (UML statecharts). The frameworks support manual coding
of UML state machines in C or C++ as well as automatic code generation
by means of the free QM™ model-based design tool (discussed later).
QP™ frameworks are especially applicable to systems with functionalsafety requirements, such as medical devices, defense, aerospace,
industrial control, robotics, IoT, transportation, automotive, etc.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
QP™ RTEFs are surrounded by a comprehensive suite of host-based
tools that provide the following functionality:
●
●
●
●
●
●
Model-Based Design (MBD) and automatic code generation (QM)
Software Tracing System (QP/Spy)
Unit Testing harness (QUTest) specifically designed for event-driven
embedded systems
System monitoring and visualization (QspyView) for developing custom
remote interfaces to embedded systems
GUI Prototyping on Windows (QWin) for designing realistic embedded
front-panels consisting of LCDs, LEDs and buttons with direct binding to
C and C++.
Dual-targeting of embedded development on Windows, Linux, or
MacOS workstations.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
QP™ frameworks are developed under the increasingly popular, strictly
quality-controlled, professional open source business model that
combines the best of the open source and proprietary software worlds to
make open source a safe choice for the embedded systems vendors. This
includes the accountability for the licensed intellectual property,
professional documentation and technical support expected of a traditional
software vendor as well as transparent development, availability of source
code and active community inherent in open source projects.
QP™ RTEFs address high-reliability applications across a wide variety of
markets. In each of these application areas, the modern QP™ software
architecture has distinct advantages.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
NOTE: All QP™ frameworks are fundamentally object-oriented, which
means that the frameworks themselves and your applications derived from
the frameworks are fundamentally composed of classes and only classes
can have state machines associated with them.
The QP/C and QP/C++ frameworks have very similar features, although
QP/C++ supports directly the C++ object model, while QP/C emulates it
with design patterns and coding conventions.
The QP-nano framework has significantly reduced feature set, specifically
designed for low-end 8-bit CPUs with very limited RAM.
The general guidelines for choosing the QP framework are as follows:
●
8-bit CPU and/or total RAM < 1KB → QP-nano
●
16- or 32-bit CPU and total RAM > 1KB → QP/C or QP/C++
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
In the resource-constrained embedded systems, the biggest concern has
always been about the size and efficiency of Active Object (Actor)
frameworks, especially that the frameworks accompanying various
modeling tools have traditionally been built on top of a conventional RTOS,
which adds memory footprint and CPU overhead to the final solution.
However, it turns out that a real-time embedded framework (RTEF) can be
actually smaller than a traditional RTOS. This is possible, because active
objects don't need to block internally, so most blocking mechanisms (e.g.,
semaphores) of a conventional RTOS are not needed.
For example, the diagram shows the RAM/ROM sizes of the QP/C, QP/C+
+, and QP-nano RTEFs versus a number of conventional (RT)OSes. The
diagram shows the total system size as opposed to just the RTOS/OS
footprints. As you can see, when compared to conventional RTOSes,
QP™ frameworks require significantly less RAM (the most precious
resource in single-chip MCUs).
All these characteristics make event-driven RTEF is a perfect fit for singlechip microcontrollers (MCUs). Not only you get the productivity boost by
working at a higher level of abstraction than raw RTOS tasks, but you get it
at a lower resource utilization and better power efficiency.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
The QP™ RTEFs have a layered structure:
●
The Target hardware sits at the bottom.
●
The Board Support Package (BSP) above it provides access to the
board-specific features, such as the peripherals.
●
The real-time kernel (QV, QK, QXK, or a conventional 3rd-party RTOS)
provides the foundation for multitasking, such as task scheduling,
context-switching, and inter-task communication.
●
The event-driven framework (QF) supplies the event-driven
infrastructure for executing active objects and ensuring thread-safe
event-driven exchanges among them.
●
The event-processor (QEP) implements the hierarchical state machine
semantics (based on UML statecharts). The top layer is the applicationlevel code consisting of loosely-coupled active objects.
●
QS is software tracing system that enables developers to monitor live
event-driven QP™ applications with minimal target system resources
and without stopping or significantly slowing down the code. QS is an
ideal tool for testing, troubleshooting, and optimizing QP™ applications.
QS can even be used to support acceptance testing in product
manufacturing.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
-
The package and class structure reflects the layered architecture. The QP
framework provides a few base classes to be subclassed and specialized
in the applications. This is a very common approach characteristic of most
frameworks. The framework also uses the underlying kernel or RTOS for
basic multitasking and event-queuing services.
The most important base classes provided by the framework are:
●
QHsm base class for deriving application-specific HSMs.
●
QActive base class for deriving application-specific Active Objects
●
QEvt base class for deriving application-specific events with parameters
or to be used directly for events without parameters.
●
QTimeEvt class to be used “as is” for time events or to be further subclassed into application-specific time events.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
/
/
C
The QEP event processor provides efficient implementation for
hierarchical state machines and enables developers to code hierarchical
state machines in an intuitive, straightforward way, where each state
machine elements maps to code precisely, unambiguously, and exactly
once (traceability between code and design).
QEP supports the following state machine concepts:
Hierarchical state nesting
●
Entry/exit actions in states
●
Regular transitions
●
Internal transitions
●
Nested initial transitions
●
Guard conditions on all transition types
●
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
1
2
_1
_2
The main job of an RTEF is to provide execution environment (thread) to
each active object and to provide thread-safe, asynchronous mechanisms
to exchange events.
The QF real-time framework serves as a “software bus” to connect active
objects. The framework supports direct event posting as well as publishsubscribe event exchange.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
(1)
(3)
1
(2)
2
(
)
The QP/C and QP/C++ frameworks support “zero copy” event delivery for
exchanging events with arbitrary parameters, which works as follows:
1. Application allocates an event from one of the fixed-size event-pools
2. Application posts or publishes the event to active objects
3. AOs process the event, whereas they can re-post or re-publish it
4. QP automatically detects if the event is still in use and recycles it if isn't
5. QP supports immutable “static events” as an optimization. Such static
events don't need to be dynamically allocated and recycled.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
(
-
)
-
n
n-1
.
(e);
;
0
1
.
(e);
;
.
;
(e);
The QP™ framework provides an assortment of real-time kernels that the
developers can choose to execute their active objects.
The simplest and most efficient is the cooperative QV (“Vanilla”) kernel,
which operates as follows:
The kernel runs in a single main loop, which constantly polls the event
queues of all active objects. The kernel always selects the highest-priority,
not-empty event queue. Every event is always processed to completion in
the main loop. If any new events are produced during the RTC step (e.g.,
by ISRs or by actions in the currently running active object) they are just
queued, but the current RTC step is not preempted. The kernel very easily
detects a situation where all event queues are empty, in which case it
invokes the idle callack, where the application can put the CPU into a lowlevel sleep mode (power-efficient kernel)
The task-level response of this kernel is the longest RTC step in the whole
system, but without blocking the RTC steps are naturally very short.
Therefore the QV kernel is adequate to many systems, including safetycritical systems.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
high priority task
(3)
low priority task
/
(4)
(2)
(1)
(5)
5
0
interrupt
(3)
(4)
-
(5) (6)
high priority task
(7)
(8)
_
_
_
(9)
(
low priority task
_
/
)
(2)
,
(1)
(
)
-
0
5
0
1
2
3
4
5
The QP™ frameworks also contain a very efficient, preemptive, prioritybased, run-to-completion kernel called QK. This kernel does not allow
threads to block in the middle of run-to-completion step, but allows them to
preempt each other (such threads are classified as “basic threads” in the
OSEK/VDX terminology). The non-blocking limitation is irrelevant for
event-driven active objects, where blocking is not needed anyway.
The threads in the QK kernel operate a lot like interrupts with a prioritized
interrupt controller, except that the priority management happens in
software (with up to 64 priority levels). The limitation of not-blocking allows
the QK kernel to nest all threads on the single stack, the same way as all
prioritized interrupts nest on the same stack. This use of the natural stack
protocol of the CPU makes the QK kernel very efficient and requires much
less stack space than traditional blocking kernels.
Still, the QK kernel meets all the requirements of the Rate Monotonic
Analysis and can be used to in hard real-time systems.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
Finally, the QP/C and QP/C++ frameworks contain a traditional
preemptive, blocking kernel called QXK. QXK allows threads to block
anywhere in the code, so it works just like most traditional blocking kernel.
The main purpose of QXK is to allow sequential code (middleware or
legacy code) to coexist with event-driven active objects, without a need for
any 3rd-party RTOS kernel.
QXK provides typical blocking mechanisms, such as semaphores,
mutextes, and time delays. Such primitives are typically expected by
various middleware libraries (TCP/IP stacks, File systems, USB libraries,
etc.)
The main advantage of QXK is that is integrates very tightly with QP™ and
reuses most of the common facilites, such as event queues. QXK is
currently available for all ARM Cortex-M cores (M0/M0+/M3/M4/M4F/M7).
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
-
-
A
B
C
Testing, debugging, and fine-tuning of embedded software often takes
more calendar time than design and coding combined. The biggest
problem is the limited visibility in to the deeply embedded system
Software tracing is a method for obtaining diagnostic information in a live
environment without the need to stop or significantly slow-down the code.
Software tracing is especially effective in event-driven systems, where all
important system interactions funnel through the active object framework
and the state machine event processor.
QP/C and QP/C++ frameworks contain QS/QSPY software tracing system
that is an ideal tool for testing, troubleshooting, and optimizing QP™
applications. QS can even be used to support acceptance testing in
product manufacturing
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
QUTest™ (pronounced 'cutest') is a unit testing harness (a.k.a. unit testing
framework), which is specifically designed for deeply embedded systems,
but also supports unit testing of embedded code on host computers ("dual
targeting"). QUTest™ is the fundamental tooling for Test-Driven
Development (TDD) of QP/C/C++ applications, which is a highly
recommended best-practice.
In a nutshell, working with QUTest™ is similar to "debugging with printf",
where you instrument the code with the printf statements (or sprintf or
similar). You then run the code with a controlled set of inputs, and examine
the produced output from the printfs to determine whether the code under
test operates correctly. The main differences from using printfs are: (1) that
the much more efficient QP/Spy is used instead and (2) that both
generating the inputs and the checking of the test outputs are automated.
QUTest™ separates the execution of the CUT (Code Under Test) from
checking of the "test assertions". The embedded target is concerned only
with running a test fixture that exercises the CUT and produces QP/Spy™
trace, but it does not check the "test assertions". Checking the "test
assertions" against the expectations is performed on the Host computer by
means of test scripts (Python and Tcl are supported).
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
/
-
QSpyView™ is a tool for remote monitoring and control of embedded
devices. QSpyView™ leverages the QP/Spy software tracing system to
provide:
●
●
●
●
Provide a customizable (Tcl/Tk) remote User Interface to your
embedded devices
Graphically display information about the life, running Target
Dynamically interact with the running Target
Remotely reset of the Target
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
Design by Contract™ (DbC) is a philosophy that views a software system
as a set of components whose collaboration is based on precisely defined
specifications of mutual obligations — the contracts. The central idea of
this method is to inherently embed the contracts in the code and validate
them automatically at runtime.
In C and C++, the most important aspects of DbC (the contracts) can be
implemented with assertions. Assertions are increasingly popular among
the developers of mission-critical software. For example, NASA requires
certain density of assertions in such software.
In the context of active object frameworks, such as QP™, DbC provides an
excellent methodology for implementing a very robust, redundancy layer
for monitoring error-free operation. Due to inversion of control so typical in
all event-driven systems, an active object framework controls many more
aspects of the application than a traditional (Real-Time) Operating System.
Such a framework is in a much better position to ensure that the
application is performing correctly, rather than the application to check
error codes or catch exceptions originating from the framework.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
This section introduces the QM graphical modeling tool for the QP
frameworks.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
QM™ (QP™ Modeler) is a freeware, graphical modeling tool for designing
and implementing real-time embedded applications based on the QP™
frameworks and hierarchical state machines (UML statecharts). QM™ is
available for Windows 64-bit, Linux 64-bit, and Mac OS X 64-bit.
QM™ and QP™ beautifully complement each other:
●
QM™ provides a diagram editor for building models of the QP™
applications, to take advantage of the very expressive visual
representation of HSMs as state diagrams (UML statecharts)
●
QP™ frameworks provide an excellent target for code generation
The main goals of the QM™ modeling tool are:
●
to help you break down your software into active objects;
●
to help you graphically design the hierarchical state machines
associated with these active objects, and
●
to automatically generate code that is of production-quality and is fully
traceable from your design.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
Compared to most other "high ceremony" modeling tools on the market
today, QM™ is much simpler, code-centric, and relatively low-level.
This characterization is not pejorative. It simply means that QM™ maps
the design unambiguously and directly to C or C++ code, without
intermediate layers of "Platform-Independent Models" (PIMs), "PlatformSpecific Models" (PSMs), complex "Model-Transformations", or "Action
Languages". All actions executed by state machines are entered into the
model directly in C or C++.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
As most modeling tools, QM™ allows you to capture the logical structure
of your application in terms of packages, classes, and state machines. The
tool provides several views of the abstract model, such as the hierarchical
tree-like Model Explorer, the Diagrams, and Property Sheets associated
with the selected model element.
A lot of thought went into drawing hierarchical state diagrams in QM™. In
this respect, the tool is innovative and might work differently than other
graphical state machine tools on the market. For example, QM does not
use "pseudostates", such as the initial pseudostate or choice point. Instead
QM uses higher-level primitives of initial-transition and choice-segment,
respectively. This simplifies state diagramming immensely, because you
don't need to separately position pseudostates and then connect them.
Also, QM introduces a new notation for internal transitions, which allows
actual drawing of internal transitions (in standard UML notation internal
transitions are just text in the state body). This notation enables you to
attach internal transitions and/or regular state transitions to a single choice
point–something that comes up very often in practice and was never
addressed well in the standard UML.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
QM™ is a unique model-based design tool on the market that allows you
to capture the physical design of your code as an integral part of the
model, whereas "physical design" is the partitioning of the code into
directories and files, such as header files (.h) and implementation files (.c
or .cpp files).
This unique approach gives you the ultimate flexibly in the source code
structure and mitigates the needs to make manual changes to the
generated code in order to combine it with hand-written code or existing
3rd-party code.
Also, QM™ provides mechanisms to quickly go back and forth between
the model and the generated code so that any changes can be
conveniently applied directly to the model rather than the code.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
QM™ provides a mechanism to extend the tool with external commands,
which can be executed directly from QM™.
Examples of external tools include make to perform a software build, lint to
statically check your code, your own unit tests that run on the host
computer, or a command-line tool to download the code to the embedded
target, directly from QM™ with just one key press. The output generated
by the external tool will appear in the QM's Log Console.
© 2005-2019, Quantum Leaps
https://www.state-machine.com/
© 2005-2019, Quantum Leaps
https://www.state-machine.com/