Commissariat &#xc0; L'&#xc9;nergie Atomique Et Aux &#xc9;nergies Alternatives

Krono-safe

Commissariat a L'energie Atomique

Areva Np Gmbh

Scaleo Chip

Silicon Mobility

Vincent David

Mathieu Jan

Stéphane Louise

Adrien Barbot

Christophe Aussagues

Michel Harrand

Thierry Collette

Nicolas Ventroux

Raphaël David

Olivier Debicki

Frédéric Blanc

Emmanuel Ohayon

Jean Delcoigne

Eric Miniere

Damien Chabrol

Matthieu Lemerre

Bruno Salle

A method for constructing off-line a sequencing plan for a plurality of tasks executed on a real-time system, including decomposing each task into a sequence of frames, where each frame is associated with an elementary process having an execution need, and defines a start date from which the process can start and an end date, or deadline, at which the process must complete; associating respective order constraints with pairs of processes; projecting the dates on a time scale associated with an execution unit of the real-time system, thus forming a sequence of time slots; sequentially parsing the time slots of the sequence; identifying eligible processes in a current time slot, a process being eligible if it meets the following two criteria: i) it is executable in the current time slot, and ii) if it has an order constraint.

Method for executing sequencing plans ensuring low-latency communication between real-time tasks

A method for executing two tasks in timesharing, includes: decomposing offline each task in a repetitive sequence of consecutive frames, and defining a start date and deadline by which an associated atomic operation must respectively start and end; verifying for each frame of a first of the repetitive sequences the corresponding operation can be performed between any two successive operations of a group of frames of the second repetitive sequence, overlapping the first repetitive sequence frame; and if the verification is satisfied, allowing the execution of the two tasks. Scheduling the operations of the two tasks, if two operations can start, executing the operation having the shorter deadline; and if a single operation can start, executing it if its execution need is less than the time remaining until the next frame start date of the other sequence, plus the time margin associated with the next frame of the other sequence.

Method for composing and executing a real-time task sequence plan

A device and a method for transmitting data within a network switch is provided. The method implemented by computer makes it possible to transmit to a receiver network domain only the applicational data necessary for this domain. The method identifies in a frame received solely the data corresponding to each addressee network domain, and then the frame is disassembled so as to write each suite of data into an assigned memory area of the network switch. The method moreover makes it possible to dispatch a frame with only the data of an addressee network domain by constructing a new frame with solely the data necessary for the addressee network which have been stored in a memory area of the switch.

Device and method for retransmitting data in a network switch

Method for executing a task composed of a set of sequential and alternative processes. The method includes the steps of: a) assigning to each process a hardware resource need and time constraint; b) allocating to each process a time-slot having a duration corresponding to the time constraint of the process; c) identifying a branch point at which is decided the execution of one or other of two alternative processes; d) allocating to the two alternative processes a common time-slot; e) assigning to the common time-slot a resource need equal to the larger of the resource needs of the two alternative processes; f) iterating from step c) for each branch point; g) organizing the resulting time-slots in an execution template associated with the task; and h) configuring real-time multitasking system to constrain the execution of the task according to the resource needs assigned to the time slots of the execution template.

Method of execution of tasks in a critical real-time system

A method for executing an application in a multitasking system is provided. The application is composed of at least one task for which the temporal triggering is specified in a first temporal reference frame that is asynchronous relative to the physical time, called first external clock domain, defined by a synchronous basic clock with changes of state of a peripheral device of the system. The method comprises a set of steps executed by the system upon reception of an occurrence of an interrupt in order to render the execution of the task deterministic or quasi-deterministic.

Method for executing, within a multitasking onboard system, an application timed according to a plurality of different time domains including interruption management

A method for executing tasks of a real-time application on a multitasking computer, steps including: defining time-windows, each associated with the execution of processing operation of task of the application, allocating to each processing operation having time-window, time-quota and time-margin, time allocated to processing operation by time-quota and time-margin being shorter than duration of time-window of processing operation, during the execution of application, activating each processing operation at the start of time-window with which it is associated, on expiry of time-quota of one of processing operations, activating an error mode if the execution of processing operation has not been completed and, if error mode is active for one of processing operations, executing an error handling operation for processing operation, during remaining time allocated to processing operation by time-quota and time-margin.

Task time allocation method allowing deterministic error recovery in real time

A system including a plurality of processing units for executing tasks in parallel and a communication network. The processing units are organized into clusters of units, each cluster comprising a local memory. The system includes means for statically allocating tasks to each cluster of units, so that a task of an application is processed by the same cluster of units from one execution to another. Each cluster includes cluster management means for allocating tasks to each of its processing units and space in the local memory for executing them, so that a given task of an application may not be processed by the same processing unit from one execution to another. The cluster management means includes means for managing the tasks, means for managing the processing units, means for managing the local memory and means for managing the communications involving its processing units. The management means operate simultaneously and cooperatively.

Statically allocating set of tasks to particular cluster and dynamically allocating each task to different processing units within cluster

An information processing system includes two processing cores. The execution of an application by the system includes the execution of application tasks and the execution of system tasks, and the system includes a micro-kernel executing the system tasks, which are directly linked to hardware resources. The processing system includes a computation part of the micro-kernel executing system tasks relating to the switching of the tasks on a first core, and a control part of the micro-kernel executing, on a second core, system tasks relating to the control of the task allocation order on the first core.

Method for the deterministic execution and synchronization of an information processing system comprising a plurality of processing cores executing system tasks

An allocation and control unit for allocating execution threads for a task to a plurality of auxiliary processing units and for controlling the parallel execution of said execution threads by said auxiliary processing units, the task being executed in a sequential manner by a main processing unit. The allocation and control unit includes means for managing auxiliary logical processing units, means for managing auxiliary physical processing units each corresponding to an auxiliary processing unit, and means for managing the auxiliary processing units. The means for managing the auxiliary processing units include means for allocating an auxiliary logical processing unit to an execution thread to be executed, and means for managing the correspondence between the auxiliary logical processing units and the auxiliary physical processing units. The auxiliary processing units execute in parallel the execution threads for the task by way of the auxiliary logical processing units, which are allocated as late as possible and freed as early as possible.

Allocation and control unit for controlling parallel execution of threads on auxiliary processing units

The method is implemented with a management system of the time-triggered architecture type in association with a processor of a central processor unit that possesses a privileged execution mode to which access is protected by an instruction of the 'call to system layer' type. The only system layer call that is authorized from an application task under consideration to the system layer consists in reporting a change of node in the control graph of the task in question. When the system layer has verified that the call is legal relative to the execution paths of the control graph as described in the constant tables of the application, all of the operations to be performed on the node in question by the system layer are predetermined by the constant tables of the application associated with the system layer. Prior to real-time execution, pre-ordering of the lists of tasks of an application under consideration is performed in a micro-kernel, which subsequently ensures, when called by the system layer during real-time execution, that task lists are updated in ordered manner depending on the new time characteristics of the tasks, specifically their earliest start times d(i) and their latest finish times f(i) as calculated by the system layer.

Growing community of inventors

Marcoussis, France

Vincent David