EP1894066A1

EP1894066A1 - Stored-program control

Info

Publication number: EP1894066A1
Application number: EP06754159A
Authority: EP
Inventors: Thomas Lehner
Original assignee: Contec Steuerungstechnik und Automation GmbH
Current assignee: Contec Steuerungstechnik und Automation GmbH
Priority date: 2005-06-07
Filing date: 2006-06-06
Publication date: 2008-03-05
Also published as: AT501880B1; WO2006131317A1; AT501880A1

Abstract

Disclosed is a stored-program control comprising a CPU operating at a first clock frequency, a configurable logic circuit operating at a second clock frequency, and a bus to which the CPU and the logic circuit are connected. A control program encompassing a first task and a second task is stored in the stored-program control. The first task is stored in the stored-program control in such a way that the same can be executed by the CPU while the second task in configured in the logic circuit. The logic circuit is configured such that the second task (11) can be executed by the logic circuit in exactly one clock pulse of the logic circuit.

Description

Programmable logic controller

The present invention relates to a programmable logic controller (PLC) comprising a CPU operating at a first clock frequency, a configurable logic circuit operating at a second clock frequency and a bus to which the CPU and the logic circuit are connected, a PLC program being stored in the PLC, which comprises a first task and a second task, and wherein the first task is stored in the PLC such that it is executable by the CPU and wherein the second task is configured in the logic circuit.

A task is understood to mean a time sequence unit of a program according to the European standard EN 61131. A task is defined by a name, a priority and a type that determines which condition triggers the start of the task. This condition can either be defined in time (cycle interval) or by an internal or external event at the occurrence of which the task is to be executed, for example the rising edge of a global project variable or an interrupt event of the controller. Each task can be assigned a sequence of programs that are to be executed when the task is executed.

Cycle time of a task is the time between two consecutive times at which the task is executed. The time required for the execution of the task is usually lower and at most equal to the cycle time.

In the context of this description, a first task to be executed by the CPU can also be referred to as a hardware task as a software task and a second task configured in the logic circuit.

Since the programmable logic circuit operates in parallel, the cycle time of the second task can be reduced by several orders of magnitude (10 ⁴ to 10 ⁵ ). However, the scope of the functions implementable in the logic circuit is limited, since this has only a certain size.

The advantages of CPU and programmable logic circuitry can be combined by decomposing a control program into a less time critical first task and a more time consuming second task, with the less time critical first task being executed by the CPU and the more time consuming second task being configured in the logic circuit. The combination of CPU and programmable logic circuitry per se is prior art. However, such solutions are associated with high development costs.

On the one hand, the programming of the CPU and the configuration of the logic circuit takes place in different ways. The CPU is programmed by means of a list of sequential instructions formulated in a programming language. The logic circuit is defined by means of schematic input or via hardware description languages.

On the other hand, means must be provided both in the software of the CPU and in the logic circuit, which allow the data exchange. After the hardware and software programming is typically done by different engineers, the communication between the CPU and the logic circuitry is also the communication between the engineers.

It is therefore a great advantage for the user if he can write the algorithms of the CPU and the logic circuit in the same programming language and the data exchange between the CPU and the logic circuit is automatic. This data exchange should ideally be "transparent", ie not visible to the user. This allows him to simply move program parts between the CPU and the logic circuit without having to change the communication.

Furthermore, it is an advantage if the user does not have to learn a new programming language for programming the overall system, but can use a language that is widely used in the control area, such as IEC61131-3.

A PLC with a CPU and a configurable logic circuit is shown in US 6,219,628 B1.

However, the system has the disadvantage that it does not comply with the standard IEC61131.

First, in addition to two graphical languages, the standard also defines three text-based languages that are mainly used by advanced users. Implementing a text-based language in a logic circuit is more difficult because it describes a sequential flow, whereas a graphical program already has a parallel structure.

Conversely, converting the graphical languages to text-based is state of the art. Typically, PLC systems first translate the graphical languages into a text-based intermediate language (e.g., AWL) prior to translation into machine code. Thus, with the invention described herein, both text-based and graphical languages may be executed.

Secondly, as already described, IEC 61131 defines execution time units called the tasks. The system of US 6,219,628 B1, however, has a data-driven system. For example, as shown in Figures 16 and 19 of this document, the fast portion is configured in the logic circuit to be in the form of modules interconnected with one another, with the input data being incrementally passed from module to module in response to an enable signal. A cycle time therefore essentially corresponds to the time required for the data flow from the modules arranged at the input to the modules arranged at the output.

Furthermore, this results in the problem that the execution of the fast part always requires several clock cycles. The resulting PLC is faster than if the entire program were executed on the CPU. The speed of the logic circuit is not fully utilized.

The object of the invention is to develop a generic PLC such that it is programmable according to the standard IEC 61131. Furthermore, the PLC should have a higher speed.

The measure according to the invention can be realized, for example, in such a way that at least part of the second task is configured in the logic circuit in the form of a combinatorial logic circuit. In such a combinatorial logic circuit, each input value of a signal is assigned the associated output value in exactly one cycle of the logic circuit. Regarding all variables of the hardware task, registers of corresponding width are provided in the logic circuit and the contents of these registers are determined by means of combinatorial logic from the register contents of the preceding cycle determined. For hardware tasks of a simple nature, the combinatorial logic circuit can be represented in the form of a table, wherein the rows of the table are formed by a final indication of all possible input values and the columns of the table are formed by the respectively associated output values.

The programmer specifies the desired logic in a PLC programming language, such as structured text. A software tool generates the logic in a hardware description language, such as Verilog. This generation involves emulating global variables, configuring a synchronization unit, and generating the hardware task. In simple cases, the hardware task can be generated by a syntax translation. From the hardware description language, a bitstream is subsequently synthesized. The bitstream generated in this way then configures the logic in the logic circuit.

Of course, it can be provided that the control program stored in the PLC has, in addition to the first and the second task, at least one further task, wherein the at least one further task is configured in the logic circuit. In general, a given control program can be split into a whole series of first and second tasks, the first tasks being respectively stored in the PLC such that they are executable by the CPU and the second tasks are respectively configured in the logic circuit. In other words, a given control program can be decomposed into a plurality of software and / or hardware tasks.

In particular - but not exclusively - in the presence of multiple hardware tasks, it is advantageous if it is provided that the logic circuit has a control unit which is designed to control the execution of the configured in the logic circuit hardware tasks.

This allows greater flexibility in terms of the executability of the second tasks configured in the logic circuit. Depending on the application, it may be provided, for example, that the control unit is designed such that it executes the second task at each clock of the logic circuit. Alternatively, however, it can also be provided that the control unit is designed such that it executes the second task at a predetermined multiple of the clock of the logic circuit. In particular, in the presence of at least two hardware tasks can be provided that the control unit is designed such that it the second task (first hardware task) and the at least one another task (second hardware task) executes at different clocks of the logic circuit.

It is particularly preferred that the configurable logic circuit comprises at least one FPGA (Field Programmable Gate Array). The configurable logic circuit can even be completely formed by an FPGA.

A further variant of the invention relates to a programmable logic controller (PLC), in particular according to one of the aforementioned embodiments, comprising a CPU operating at a first clock frequency, a configurable logic circuit operating at a second clock frequency and a bus to which the CPU and the logic circuit are connected wherein the PLC stores a control program comprising a first task and a second task, and wherein the first task is stored in the PLC to be executable by the CPU and wherein the second task is configured in the logic circuit and wherein the first and the second task are designed such that they can exchange signals with each other.

The application of a PLC in the control and measurement technology requires real-time capability of the control program. This is the ability to react to an event within a given period of time. The second tasks configured in the logic circuit must be able to run permanently and can not be stopped for signal or data exchange. For example, if the second task produces an output, then the output would be disrupted at that time. If the second task is configured to count input pulses, then input pulses would be missed during that time. Despite this limitation, a consistent signal or data exchange between the first and the second task (s) must be possible.

This is achieved by a PLC having the features of claim 9. By providing a synchronization unit via which the signal exchange takes place, wherein the synchronization unit is designed such that the signals transmitted in one clock cycle of the receiving task originate from exactly one clock cycle of the transmitting task, a consistent signal or data exchange is made possible since all Get data or signals of a data or signal set at the same time (the same cycle). This will be explained below with reference to an illustrative, nonlimiting example:

Assuming that the first task has a cycle time of 1 ms, and that the second task has a cycle time of 100 ns, and taking into account that both cycle times are approximately expressible as multiples of the cycle time of the logic circuit (hereinafter called system clock) It is that there is a data exchange every 10000th system clock. (In practice, this would not always happen exactly at the 10000th system clock, as the expert knows, a certain bandwidth of, for example, ± 500 system clocks is to be expected.)

Let N be an index indicating the number of each cycle of the second task, and let M be the corresponding index for the particular cycle of the first task, and suppose that the first task reads three measurements (A, B ₁ C) from the second task, then the following results:

With reference to the Mth cycle of the first task, all three measured values must come from the cycle N = 10 000 x M or N = 10 001 x M of the second task. In order to ensure consistent data transfer from the second task to the first task, however, it must not happen that, for example, the measured value A is still from the cycle N = 10 000 x M and the measured values B and C are already from the cycle N = 10 001 x M come from.

The same applies if data is to be transferred from the first task to the second task. All values passed from the first task to the second task must change from their old to the new value within one cycle of the second task.

This can be realized, for example, by providing that the synchronization unit has at least two registers connected in series for temporarily storing the signals to be transmitted.

This enables a method of controlling the transfer of data from the first tasks executed by the CPU of a PLC to the second tasks configured in the logic circuit, in which two values are exchanged for each value exchanged between the first and the second tasks Register are provided and in which the earlier data remain stored in the second register until the new transferring data has been transferred completely over the bus into the first register and then in a clock cycle of the receiving second task, the data to be transmitted from the first register to the second register are transmitted.

Likewise, a method of controlling the data transfer from the second tasks configured in the logic circuit to the first tasks executed by the CPU is provided in which two registers connected in series are provided for each value exchanged between the first and the second tasks the data to be transmitted is stored in the second register with each clock cycle of the second task, the data to be transferred is stored in the first register at a particular time, then the data in the first register is kept constant until it is sequentially sent over the bus the first task was transferred.

In a particularly preferred embodiment of this variant of the invention, it is provided that the logic circuit has the synchronization unit.

A further variant of the invention relates to a PLC, in particular according to one of the aforementioned embodiments comprising a CPU, a configurable logic circuit and a bus to which the CPU and the logic circuit are connected, wherein in the PLC a control program is stored, which has a first task and a second task and wherein the first task is stored in the PLC such that it is executable by the CPU and wherein the second task is configured in the logic circuit and wherein the control program comprises at least one global variable.

Such global variables are available for all tasks in such a way that, in principle, each task can access the global variable both in writing and in reading. As such, it is known to a programmer that when it comes to the situation that multiple tasks write to a global variable, the global variable gets the value of the later write access.

The above behavior results from the sequential execution of software. However, since the configurable logic circuit operates in parallel rather than sequentially, the behavior of the global variable must be emulated.

By a PLC with the features of claim 12 this is achieved. For example, it can be provided that a data register is created in the emulation unit for each global variable. For each global variable, an output with a bit width adapted to the size of the respective global variable is created in the second task and connected to the respective data register of the emulation unit. Likewise, for each global variable, another output (Enable, Fig. 5) is applied and connected to the emulation unit.

The invention further relates to a method for programming a PLC with a CPU and a configurable logic circuit, in particular according to one of claims 1 to 13, with a control program to be executed by the PLC, the control program being in a text-based programming language.

The European standard EN 61131-3 defines five different programming languages, two of which are graphical programming languages and three text-based programming languages. While the method defined in US 6,219,628 B1 is only applicable to graphical programming languages, the method defined in the embodiments described above is applicable to text-based programming languages.

This can be done by a software tool decomposing the control program into a first part and a second part, the first part being translated into a machine code readable by the PLC's CPU, and a hardware description being generated from the second part, and being from the hardware description a bit stream is synthesized, and then the translated first part is stored in the PLC so that it is executable by the CPU and the synthesized bit stream is configured in the logic circuit.

This is important insofar as the users of text-based or graphical programming languages are completely different groups with regard to their qualification as programmers. Users of text-based programming languages generally have profound programming skills and also have higher requirements for the functionality to be handled by the PLC.

The invention further relates to a software tool for a text-based software for programming a PLC according to one of claims 1 to 13 with a control program to be executed by the PLC, wherein the software tool is designed such that the control program can be divided into a first part and a second part, and wherein the first part can be translated into a machine code readable by the CPU of the PLC and a hardware description can be generated from the second part and from the hardware description a bitstream configurable in the logic circuit of the PLC can be synthesized is and a disk on which a software tool of the aforementioned type is stored.

Further advantages and details will become apparent from the following figures and the accompanying description of the figures. Showing:

Fig. 1 is a block diagram of an embodiment of an inventive

PLC, Fig. 2 is a block diagram of the configurable logic circuit of the PLC after

Fig. 1, Fig. 3 is a block diagram of the logic circuit shown in Fig. 2, which is configured for a concrete application example, Fig. 4 is a block diagram of an embodiment of an inventive

Synchronization unit and Fig. 5 is a block diagram of an embodiment of an inventive

Emulator unit.

1 shows a block diagram of a PLC 1, which comprises a CPU 2 and an FPGA 3, which are interconnected by a bus 8. Digital inputs 4 as well as analogue inputs 5 and digital outputs 6 as well as analogue outputs 7 are provided. The inputs 4, 5 and the outputs 6, 7 are connected to the FPGA 3, so that a quick response is possible. However, it remains up to the user whether the inputs 4, 5 or the outputs 6, 7 are to be processed in the first task or in the second task 11. Not shown are further interfaces of the CPU 2 to further controls, to a visualization unit or to a development system (download, debugging). Shown is an optional providable expansion module 9 for the FPGA 3, which may also represent a fast input-output module. The module 9 can be connected to the FPGA 3 via a high-speed link 10.

FIG. 2 shows a block diagram of the FPGA 3. The bold edged modules 11, 19 and 20 are configured depending on the control program. The remaining modules 12 to 16 and 18 are configured by default in FPGA 3. Registers 15, 16 and 18 represent the interface to the CPU 2. The registers 15 to 18 are connected to the bus 21 so that the data in the memory area of the CPU 2 are visible. The input-output unit 13 takes over tasks such as the synchronization of the inputs 4,5 or the driving of external analog-digital converters. This unit also resolves write conflicts to the outputs 6,7, so if software task (s) and hardware task (s) want to simultaneously write to one of the outputs 6,7. The emulation unit 20 for emulating global variables and the second task 11 configured in the FPGA 3 are also recognizable. The second task 11 communicates via a synchronization unit 19, which will be described in detail below via the bus 21 or the bus 8 with the CPU 2 The control unit 12 assumes general control functions, such as the control of the execution of the second task 11, the definition of the cycle time of the second task 11 and additional functions such as debugging. The further module 14 can realize functions that would otherwise have to be done by additional chips. This may be, for example, a display controller or a CAN bus.

In a simplified example, explained with reference to FIG. 3 example is to be measured with a PLC 2, the frequency of a digital input signal which is applied to the digital input 4. If the frequency is below an imaginary minimum, this should be signaled via one of the digital outputs 6, if the frequency is above a preset maximum, this should be signaled via another of the digital outputs 6.

The control program is split into a first task and a second task 11. The second task 11 measures the frequency of the input signal by counting the positive edges of the pulses. The slower first task evaluates the counter readings and switches the digital outputs 6.

In the PLC programming language "structured text" defined in EN 61131-3, the second task 11 (hardware task) is formulated as follows: TASK FastMain;

CounterModuleO; END_TASK

PROGRAM CounterModule VAR previousLevel: BOOL; END_VAR VAR_OUTPUT counterValue: INT; END_VAR

IF parln [1] AND NOT previousLevel THEN counterValue: = counterValue + 1; ENDJF previousLevel: = parln [1]; END PROGRAM

The CounterModule program defines a Boolean variable named previousLevel. This stores the previous value of the input. Before the first execution of the program, previousLevel is LOW (thus has the logical value 0).

Then the program defines an output variable called counterValue. After the definition of the variable, the instructions follow, which are executed cyclically.

In the example it is assumed that the digital inputs are provided in the PLC with the name parln. The expression parln [1] then describes the state of input number 1 to which the signal to be counted is applied.

The IF statement checks whether there is a positive edge of the signal. This is always given if the current state is already HIGH (that is, has the logical value 1), but the state stored in the previous level is still LOW.

In this case, the CounterValue counter is incremented by 1. Then, the next cycle is prepared by storing the current value in the previousValue variable. The first task to be performed by the CPU 2 is as follows:

TASK Main;

PLC_PRG (); END_TASK •

PROGRAM PLC_PRG VAR minValue: INT; maxValue: INT; previousValue: INT; difference: INT; END_VAR

difference: = CounterModule.counterValue - previousValue; previousValue: = CounterModule.counterValue;

parOut [1]: = difference <minValue; parOut [2]: = difference> maxValue;

END_ PROGRAM

First, the task is created again. This executes the main program of the control program, which always bears the name PLC_PRG.

The program first defines the variables again. minValue and maxValue are the preset limits of the frequency. In previousValue the old counter reading is buffered. Before the first run, this variable is 0. The variable difference stores the difference between the current count and the count of the previous cycle.

After defining the variables, the instructions start again.

First the difference of the counter readings is calculated. The value of the second task 11 is accessed with the syntax "CounterModule.counterValue". Then the current counter reading is stored in previousValue so that the value is available in the next cycle.

Then the two digital outputs are switched. parθut [1] is always HIGH if the difference is smaller than the preset value minValue. ParOut [2] is always HIGH if the difference is greater than the maximum (maxValue).

The difference is a measure of the input frequency. The input frequency in hertz is calculated from the formula "input frequency = (difference / cycle time of the slow task)".

The overflow of the counter does not need to be treated, as in case of overflow of the counter, an overflow also occurs during the subtraction. The two overflows cancel each other out so that the difference is correct in each case.

The first task can take over other tasks, such as transmitting the measured frequency or saving in a log file.

The second task 11 is generated in the hardware description language Verilog as follows:

module FastTask (clk, parln, counterValue);

input clk; input [7: 0] parln; output reg [15: 0] counterValue;

reg previousLevel;

always @ (posedge clk) begin if (parln [1] & ipreviousLevel) counterValue = counterValue + 1; previousl_evel = parln [1]; end

endmodule A module with the name FastTask is defined. All the input / output signals of the module are in parenthesis. These are then created in detail:

clk is the system clock parln are the inputs, bit widths are given in square brackets.

CounterValue is the generated counter value. The clause "output reg" means that for the

Output signal automatically creates a register.

Then previsousLevel is created as a 1-bit register.

The line "always @ (posedge clk)" starts a task. This is executed at every system clock clk, "if" corresponds to the IF of the structured text. "&" Is AND, "!" Is NOT, the rest is the same as in the structured text

The second task 11 is configured in the FPGA 3 as shown schematically in FIG. 3:

The system clock goes to units 24, 25 and 19 (not shown)

In this example, it is assumed that the second task 11 is executed every cycle of the system clock.

At the input 23, the input signal is present. The input signal corresponds to parln [1] in the program text above. The input-output unit 13 is not shown in this illustration. The input 23 leads to an AND-GATE 32 and to a D-FLIP FLOP 25 (data FLIP FLOP). The D-FLIP FLOP 25 is connected to the AND-GATTER 32 by a line 35. The input signal is fed to the D-FLIP FLOP 25. The output of the D-FLIP FLOP 25 corresponds to the input signal but is delayed by exactly one system clock. The output signal thus corresponds to the value "previousLevel" in the program text The AND gate 32 has an inverted input and corresponds to the IF instruction in the second task 1 1. If the condition of the IF instruction is fulfilled, then the output of the AND signal Gates 32 HIGH This causes the counter 24 to be incremented in this system clock.

The output of the counter 24 is transferred via the synchronization unit 19 and the bus 8 to the CPU 2. The construction of a synchronization unit 19 is shown schematically in FIG. There are write registers 28, 29 and read registers 30, 31 are provided. The write registers 28, 29 and the read registers 30, 31 communicate with the second task 11. The data D to be read are applied to the registers 30 via the lines 33. Via lines 34, the read registers 31, a clock enable signal (CE signal) can be fed. Upon receipt of the CE signal, the data in the read registers 31 are overwritten with the data D currently present in the registers 30. As long as the CE signal is equal to zero, the output at the read registers 31 remains constant, even though the input value applied to the lines 33 should change. Via the bus 21, the synchronization unit 19 communicates with the CPU 2.

The process is as follows:

The registers 28, 31 is supplied via the lines 34, the CE signal. When these lines are switched to 1, the respective registers 28, 31 take the data D in each cycle from the input to the output. If CE = 0, the data registers 28, 31 retain the current value. First, the CE signal is set to 1. The read registers 30 permanently receive current values from the second task 11. The first task then sets the CE signal to 0 via the control unit 12. This causes the data in the read registers 31 to be frozen. The first task reads in the data sequentially via the bus 8 and the bus 21 and stores them in the main memory of the CPU 2. The first task then writes the output values sequentially via the bus 8 and the bus 21 in the write registers 29. Subsequently the CE signal is set to 1. As a result, the write values within a cycle in the register 28 and then in the second task 11 are adopted. At the same time, the data in registers 31 is updated. After a completed cycle of the first task, this process begins again.

The representation chosen here is simplified. In real use, two additional signals are required: the clock enable signal for controlling the cycle time and a reset signal for initialization.

Furthermore, the block diagram already represents an optimized version The direct implementation according to claim 1 would include a data register for storing the variables instead of the counter 24 and an adder with the constant "+1", which takes over the state change, but current synthesis tools are able to recognize this combination in Verilog and replace it with a counter. These optimizations allow the example described here to be clocked in currently available 500 MHz FPGAs. This corresponds to a cycle time of the hardware task of 2 ns. The PLC described here can therefore count signals with frequencies up to 250 MHz.

5 shows a block diagram of an emulation unit 20.

The data register 37 holds the current value of the global variable. This is given on the one hand to the hardware task (right) as well as the read register 38, the buffer 39 and the address data bus 21 to the first task.

The read register 38 has only the function of keeping the value constant while read from the bus 8; Access via the address data bus 21 generally takes several clock cycles of the system clock.

All changes to the register 37 pass through the unit 36 (arbitration unit). This supplies to the register 37 the data (shown in bold) and an enable signal (thin). If the enable signal has the logic value 1, then the new data in the register 37 are adopted.

The changes can be made either via bus 8 (in FIG. 5 on the right) or via the first task (in FIG. 5 on the left).

Changes via bus 8:

For write accesses to the global variable, the data is stored in register 40. The unit 41 detects a write access and in this case provides an enable signal that is HIGH for a system clock.

Changes via the hardware task:

When the hardware task (second task 11) changes the global variable, it sets the enable signal HIGH (thin drawn) for one clock cycle. At the same time, he creates the new data. (drawn thickly). The unit 36 regulates the accesses to the register 37. It does this, for example, according to the following algorithm:

• If there is no access, then the enable signal of the register is LOW. The data does not matter in this case.

• If there is a write access of the first task, then the enable signal is HIGH and the data is the data of the first task.

If there is a write access of the second task 11 but no access of the first task, then the enable signal is HIGH and the data is the data of the second task 11.

From Fig. 5 it can be seen that the second task 11 needs three signals for each global variable.

An input corresponding to the bit width of the variable (xxx_in)

• an output corresponding to the bit width of the variable (xxx_out)

• an enable output which signals a write access (xxx_changed)

Of course, this must be taken into account when generating the hardware description, as follows:

• All read accesses to a global variable are replaced with accesses to the corresponding input.

• All write accesses are replaced by two commands:

1. Set the output to the appropriate value

2. Set the enable signal HIGH for one clock cycle

Although the description refers to cyclic tasks, that is, tasks performed at a given periodicity, the disclosure and the scope of protection equally encompass non-cyclic tasks. As already mentioned at the outset, this includes tasks that are always executed in the presence of a specific internal or external event.

Claims

claims:

1. PLC comprising:

a CPU operating at a first clock rate

a configurable logic circuit operating at a second clock frequency

a bus to which the CPU and the logic circuit are connected, wherein the PLC stores a control program comprising a first task and a second task, and wherein the first task is stored in the PLC such that it can be executed by the CPU is and wherein the second task is configured in the logic circuit, characterized in that the logic circuit is configured such that the second task (1 1) is executable in exactly one clock of the logic circuit of the logic circuit.

2. PLC according to claim 1, characterized in that at least a part of the second task (11) is configured in the logic circuit in the form of a combinatorial logic circuit.

3. PLC according to claim 1 or 2, characterized in that in the PLC (1) stored control program in addition to the first and the second task (1 1) has at least one further task, wherein the at least one further task is configured in the logic circuit.

4. PLC according to one of claims 1 to 3, characterized in that the logic circuit comprises a control unit (12) which is designed to control the execution of the configured in the logic circuit tasks (11).

5. PLC according to claim 4, characterized in that the control unit (12) is designed such that it executes the second task (1 1) at each clock of the logic circuit.

6. PLC according to claim 4, characterized in that the control unit (10) is designed such that it executes the second task (11) at a predetermined multiple of the clock of the logic circuit.

7. PLC according to one of claims 4 to 6, characterized in that the control unit (12) is designed such that it executes the second task (1 1) and the at least one further task at different clocks of the logic circuit.

8. PLC according to one of claims 1 to 7, characterized in that the configurable logic circuit comprises at least one FPGA (3).

9. SPS, in particular according to one of claims 1 to 8, comprising:

a CPU operating at a first clock rate

a configurable logic circuit operating at a second clock frequency

a bus to which the CPU and the logic circuit are connected, wherein the PLC stores a control program comprising a first task and a second task, and wherein the first task is stored in the PLC such that it can be executed by the CPU is and wherein the second task is configured in the logic circuit and wherein the first and the second task are designed such that they can exchange signals with each other, characterized in that the PLC (1) has a synchronization unit (19) via which the signal exchange takes place wherein the synchronization unit (19) is designed such that the signals transmitted in one clock cycle of the receiving task originate from exactly one clock cycle of the sending task.

10. The PLC according to claim 9, characterized in that the synchronization unit (19) has at least two registers (29, 30) connected in series for buffering the signals to be transmitted.

11. PLC according to claim 9 or 10, characterized in that the logic circuit comprises the synchronization unit (19).

12. A PLC, in particular according to one of claims 1 to 11, comprising a CPU, a configurable logic circuit and a bus to which the CPU and the logic circuit are connected, wherein in the PLC a control program is stored, which has a first task and a second task and wherein the first task is stored in the PLC such that it is executable by the CPU and wherein the second task is configured in the logic circuit and wherein the Control program comprises at least one global variable, characterized in that in the logic circuit, an emulation unit (20) is configured, which is designed such that it assigns the value of the later write access in a write access of multiple tasks to the global variable of the global variable.

A PLC according to claim 12, characterized in that the emulation unit (20) is adapted to assign the value of the write access of the first task to the global variable during simultaneous write access of the first task and the second task (11).

14. A method for programming a PLC with a CPU and a configurable logic circuit, in particular according to one of claims 1 to 13, with a PLC program to be executed by the PLC, wherein the control program is in a text-based programming language, characterized in that a software tool, the control program in decomposing a first part and a second part, the first part being translated into a machine code readable by the CPU (2) of the PLC (1), and a hardware description being generated from the second part and a bit stream being synthesized from the hardware description; then the translated first part is stored in the PLC (1) such that it is executable by the CPU (2) and the synthesized bit stream is configured in the logic circuit.

15. A method for controlling the data transfer from the first tasks executed by the CPU of a PLC according to one of claims 9 to 13 to the second tasks configured in the logic circuit, characterized in that for each value between the first and the second tasks (11 ) is replaced two registers (28, 29) connected in series, and that

- The previous data remain stored in the second register (28) until the new data to be transferred have been completely transferred over the bus (21) in the first register (29)

- And then in a clock cycle, the data to be transmitted from the first register (29) in the second register (28) are transmitted.

16. A method of controlling the data transfer from the second tasks configured in the logic circuit of a PLC according to claim 9 to the first tasks executed by the CPU, characterized in that for each value that is between the first and the second tasks (11 ) is exchanged two registers (30, 31) connected in series are provided and that

the data to be transmitted is stored in the second register (30) with each clock cycle of the second task (11),

- The data to be transmitted at a given time in the first register (31) are stored

- And then the data in the first register (31) as long as »held constant until they were transmitted sequentially via the bus (21) to the first task.

17. Software tool for a text-based software for programming a PLC according to one of claims 1 to 13 with a PLC program to be executed by the PLC, characterized in that the software tool is designed such that the control program can be disassembled into a first part and a second part and in that the first part can be translated into a machine code which can be read by the CPU (2) of the PLC (1) and a hardware description can be generated from the second part, and a bitstream which can be configured in the logic circuit of the PLC (1) can be synthesized from the hardware description.

18. disk on which a software tool according to claim 17 is stored.