Processing module for integrated modular avionics system
The hardware-assisted virtualization technology establishes multiple virtual machines in the IMA system, which solves the problem that the existing technology cannot support multi-process memory isolation and simultaneous operation of complex operating systems, and realizes the concurrent operation and security and stability of multiple operating systems.
Patent Information
- Application Number
- CN202310619959.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-05-29
- Publication Date
- 2024-11-29
AI Technical Summary
The existing IMA processing module architecture cannot support memory isolation between running multiple processes within a single partition, and cannot run ARINC 653 and complex operating systems such as Linux at the same time.
Through hardware-assisted virtualization technology, multiple virtual machines are established, each virtual machine runs its unique operating system and application software. The module operating system serves as a virtual machine monitor to manage hardware resources, realize the concurrent operation of multiple operating systems and provide memory isolation.
Improves system security and performance, allowing multiple operating systems to be run simultaneously on the same hardware, including ARINC 653 and Linux, ensuring security and stability between operating systems.
Smart Images

Figure CN119045943A_ABST
Abstract
Description
Technical Field
[0001] The present application relates generally to the field of avionics systems, and more particularly to a processing module for an integrated modular avionics (IMA) system. Background Art
[0002] The IMA architecture is widely used in aircraft avionics systems. The IMA architecture is a layered architecture in which the program part including code and data is partitioned to ensure the reliability of the entire system operation.
[0003] ARINC 653 (Standard Software Interface for Avionics Applications) is a software specification for space and time partitioning in safety-critical avionics real-time operating systems (RTOS). It allows multiple applications of different software levels to be hosted on the same hardware in the context of an integrated modular avionics architecture. To decouple the RTOS platform from the application software, ARINC 653 defines an API called "APplication EXecutive (APEX)". Each application is called a partition and has its own memory space. It also has dedicated time slots allocated by the APEX API. Within each partition, multitasking is allowed. The APEX API provides services to manage partitions, processes and timing, as well as partition / process communication and error handling.
[0004] However, in the existing IMA universal processing module architecture, the ARINC 653 module operating system (Module OS) can only run one or more ARINC 653 applications or tasks in each partition, and cannot support running complex operating systems such as Linux. In addition, although ARINC 653 MOS can support multi-tasking, since MOS is implemented under kernel privileges and ARINC 653 applications or tasks are implemented under user privileges, it is impossible to provide memory isolation between multiple processes in a single partition.
[0005] Therefore, a new IMA processing module architecture is required. Summary of the invention
[0006] The purpose of the present invention is to provide such a processing module for an integrated modular avionics system, which utilizes the hardware virtualization provided by the CPU to establish multiple virtual machines, each of which has its own operating system and application software, thereby improving security and performance.
[0007] Specifically, an exemplary embodiment of the present invention provides a processing module for an integrated modular avionics system, comprising: hardware resources, including a processor, a memory, and one or more peripherals, wherein the processor supports hardware-assisted virtualization; a hypervisor layer component, including a module operating system, the module operating system establishes multiple virtual machines with the help of the hardware-assisted virtualization and allocates the hardware resources to each virtual machine, wherein the multiple virtual machines include multiple partitioned virtual machines; an operating system layer component, including multiple operating systems running in the multiple virtual machines respectively, wherein the multiple operating systems include multiple partitioned operating systems based on the ARINC 653 standard, and the multiple partitioned operating systems run on the multiple partitioned virtual machines respectively; and an application layer component, the application layer component includes application programs running on the multiple operating systems.
[0008] According to another exemplary embodiment, a method for establishing an integrated modular avionics system is provided, comprising: providing hardware resources, the hardware resources comprising a processor, a memory, and one or more peripherals, wherein the processor supports hardware-assisted virtualization; establishing a plurality of virtual machines with the aid of the hardware-assisted virtualization and allocating the hardware resources to each virtual machine, wherein the plurality of virtual machines comprises a plurality of partitioned virtual machines; running a plurality of operating systems in the plurality of virtual machines, respectively, wherein the plurality of operating systems comprises a plurality of partitioned operating systems based on the ARINC 653 standard, and the plurality of partitioned operating systems respectively run on the plurality of partitioned virtual machines; and running corresponding applications on the plurality of operating systems.
[0009] In the above design, with the help of the hardware-assisted virtualization capability provided by the processor, the module operating system can act as a virtual machine monitor (VMM) to divide the hardware resources into multiple independent virtual machines, and each virtual machine is isolated from each other, so that multiple operating systems (such as partitioned operating systems based on the ARINC 653 standard and complex operating systems such as Linux) can be implemented in multiple virtual machines. Concurrent operation, while ensuring the security and stability between operating systems.
[0010] Preferably, the processor is a multi-core processor, wherein the plurality of partitioned virtual machines run alternately on a core of the multi-core processor, and the module operating system allocates a cycle and duration for exclusive use of the core to each partitioned virtual machine.
[0011] Preferably, at least one peripheral among the one or more peripherals is allocated to the multiple partition virtual machines, and when the multiple partition virtual machines switch on the core, the module operating system notifies the at least one peripheral.
[0012] Preferably, when the at least one peripheral device needs to perform direct memory access (DMA), the remaining runtime of the current partition virtual machine running on the core is sent to the at least one peripheral device to determine whether the remaining runtime is sufficient for direct memory access conversion.
[0013] Preferably, the plurality of virtual machines further include a general virtual machine, and other cores of the multi-core processor except the core are allocated to run the general virtual machine.
[0014] Preferably, the plurality of operating systems further include a general operating system that is not based on the ARINC 653 standard, and the general operating system runs on the general virtual machine.
[0015] Preferably, the module operating system configures the plurality of partition virtual machines to be isolated from each other and from the general virtual machine.
[0016] Preferably, the plurality of partition virtual machines communicate with each other using an inter-partition communication (IPC) mechanism.
[0017] Preferably, the module operating system allocates a specific peripheral among the one or more peripherals so that the specific peripheral is used by only one virtual machine among the multiple virtual machines.
[0018] Preferably, the module operating system allocates the one or more peripherals so that the one or more peripherals are shared by the multiple virtual machines.
[0019] Preferably: the one or more peripherals themselves are capable of deterministically sharing their resources among the multiple virtual machines; the module operating system virtualizes the one or more peripherals to establish multiple virtual peripherals and distributes the multiple virtual peripherals among the multiple virtual machines; or one of the multiple virtual machines acts as an agent to relay messages between the one or more peripherals and other virtual machines among the multiple virtual machines.
[0020] Preferably, the processing module further comprises an input-output memory management unit (IOMMU), and the input-output memory management unit is used to isolate access of different virtual machines to the same peripheral device.
[0021] Preferably, the one or more peripheral devices support Single Root I / O Virtualization (SR-IOV).
[0022] Preferably, the partition operating system provides memory isolation between multiple processes in the partition virtual machine where the partition operating system is located. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] The present invention may be better understood by describing exemplary embodiments of the present invention in conjunction with the accompanying drawings, in which:
[0024] Figure 1 A schematic structural block diagram of a processing module for an integrated modular avionics (IMA) system according to an embodiment of the present invention is shown.
[0025] Figure 2 A schematic block diagram showing hardware resources according to an embodiment of the present invention is shown.
[0026] Figure 3 An example of the composition of a plurality of virtual machines according to an embodiment of the present invention is shown.
[0027] Figure 4 An example of a time schedule for allocating usage of a multi-core processor by a processing module to multiple virtual machines according to an embodiment of the present invention is shown.
[0028] Figure 5 The allocation of memory resources to each virtual machine in the processing module is shown.
[0029] Figure 6 A flow chart of a method 600 for establishing an IMA system according to an embodiment of the present invention is shown. DETAILED DESCRIPTION
[0030] The specific embodiments of the present invention will be described below. It should be noted that in the specific description of these embodiments, in order to provide a concise description, it is impossible for this specification to provide a detailed description of all the features of the actual embodiments. It should be understood that in the actual implementation of any embodiment, just as in the process of any engineering project or design project, in order to achieve the specific goals of the developer and to meet system-related or business-related restrictions, various specific decisions are often made, and this will also change from one embodiment to another. In addition, it can also be understood that although the efforts made in this development process may be complex and lengthy, for ordinary technicians in the field related to the content disclosed by the present invention, some changes such as design, manufacturing or production based on the technical content disclosed in this disclosure are just conventional technical means, and should not be understood as insufficient content of this disclosure.
[0031] Unless otherwise defined, the technical or scientific terms used in the claims and the specification shall have the usual meaning understood by persons with ordinary skills in the technical field to which the invention belongs. The words "first", "second" and similar words used in the patent application specification and the claims of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "One" or "one" and other similar words do not indicate a quantitative limitation, but indicate the existence of at least one. "Include" or "comprises" and other similar words mean that the elements or objects appearing before "include" or "comprises" include the elements or objects listed after "include" or "comprises" and their equivalent elements, and do not exclude other elements or objects. "Connected" or "connected" and other similar words are not limited to physical or mechanical connections, nor are they limited to direct or indirect connections.
[0032] In this disclosure, a structure may be referred to as being "used for" or "capable of" performing certain tasks, even if the structure is not currently being operated. For example, "a module operating system used for / capable of establishing a virtual machine" is intended to encompass a module operating system having a component that performs that function during operation, even if the module operating system is not currently being operated (e.g., not started).
[0033] In this application, if there is no special description, all embodiments and preferred embodiments mentioned herein can be combined with each other to form a new technical solution. In this application, if there is no special description, all technical features and preferred features mentioned herein can be combined with each other to form a new technical solution.
[0034] In the description of the embodiments of the present application, the term "and / or" is only a description of the association relationship of the associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.
[0035] In this application, some English abbreviations are used to make the description concise. Specifically, in this application, IMA refers to Integrated Modular Avionics, MOS refers to Module Operating System, and RTOS refers to Real-Time Operating Systems.
[0036] In this application, the term "component" may refer to hardware, software, firmware, or any combination thereof.
[0037] A processing module and method for an integrated modular avionics (IMA) system provided by an embodiment of the present invention will be described in detail below with reference to the accompanying drawings.
[0038] From the perspective of IMA system users, it is hoped that ARINC 653 partitions will resemble physical computers in a federated system, and hardware-assisted virtualization is the best way to simulate multiple computers on top of a set of hardware. Since the ARINC 653 standard was developed before hardware-assisted virtualization matured, most ARINC 653 systems are implemented on top of certain operating systems (OS) rather than hypervisors. RTOS services are used to perform space and time partitioning. For example, WindRiver PSC ARINC 653 was developed on top of its VxWorks RTOS. Domestic ARINC 653 implementations also use a similar approach. At present, hardware-assisted virtualization is mature enough and is widely used in desktops, data centers, and even handheld devices. Therefore, this application proposes a solution for implementing an ARINC 653 system using hardware-assisted virtualization capabilities.
[0039] According to one aspect of the present invention, a processing module for an integrated modular avionics system is provided.
[0040] Figure 1 A schematic structural block diagram of a processing module for an integrated modular avionics system according to an embodiment of the present invention is shown.
[0041] The processing module 100 may include hardware resources 110, hypervisor layer components 120, operating system layer components 130, and application layer components 140. In an embodiment of the present invention, the processing module 100 may comply with the ARINC 653 standard.
[0042] The hardware resources 110 may constitute the hardware layer of the processing module 100, and may include hardware devices such as a processor 111, a memory 112, and one or more peripherals 113, such as Figure 2 As shown. The processor 111 may support hardware-assisted virtualization. The processor 111 supporting hardware-assisted virtualization may be based on architectures such as ARM v8 / v9, Intel's VT-x, or AMD's AMD-V.
[0043] In some embodiments, the hardware resources 110 may also include hardware support components ( Figure 2 The hardware support component can be used to provide support for the scheduling and management of the underlying hardware devices. As an example, the hardware support component can provide CPU support, bus / device drivers, system boot, board support packages, etc.
[0044] The hypervisor layer component 120 constitutes the supervisory layer of the processing module 100. The hypervisor layer component 120 may include a module operating system (MOS) and may also include some extended functional components. MOS may run directly on the hardware resources 110 as the underlying OS.
[0045] MOS can establish multiple virtual machines (virtual machine 1, virtual machine 2, ..., virtual machine N) with the help of hardware-assisted virtualization provided by processor 111 and allocate hardware resources 110 to each virtual machine. Hardware-assisted virtualization can hand over some virtualization operations to the hardware in the processor. Specifically, hardware-assisted virtualization usually provides some special instructions that allow the operation of the virtual machine to run directly on the CPU hardware without the need for translation by MOS. Hardware-assisted virtualization can make the process of virtual machines accessing hardware resources 110 such as processors, memory, and peripherals more efficient, and enhance the security of virtual machines. For example, assuming that a virtual machine needs to access memory, the CPU hardware can directly convert the instructions issued by the virtual machine into instructions of the physical machine, and then send them to the memory controller of the physical machine. In this way, the efficiency of virtual machines accessing memory can be greatly improved, and the overhead of virtualization can be reduced.
[0046] The processing module 100 allocates computing resources to each virtual machine by scheduling the use of the processor 111, which will be further described below. Figure 4 The management of the memory 112 by the processing module 200 is achieved by allocating a separate memory space to each virtual machine, which will be further discussed below. Figure 5 The peripherals 113 may include I / O devices, network adapters, etc. for communication. A person skilled in the art will appreciate that, depending on the specific application, the hardware resources 110 may include, for example, Figure 2 More or fewer hardware devices are shown.
[0047] The operating system layer component 130 may include multiple operating systems (OS 1130a, OS 1130b, ..., OS 1130n) running in multiple virtual machines respectively, constituting an OS layer. In some embodiments, the multiple operating systems may include multiple partition operating systems (POS) based on the ARINC 653 standard.
[0048] The application layer component 140 may include application programs (application 1 140a, application 2 140b, . . . , application n 140n) running on the plurality of operating systems 130a - 130n, which may constitute the application layer of the processing module 100.
[0049] In the above four-layer architecture of the processing module 100 for the IMA system, with the hardware-assisted virtualization capability provided by the processor 111, the underlying OS (i.e., MOS) can act as a virtual machine monitor (VMM) to divide hardware resources into multiple independent virtual machines, and each virtual machine is isolated from each other, so that multiple operating systems (such as partitioned operating systems based on the ARINC 653 standard and complex operating systems such as Linux) can be run concurrently, while ensuring the security and stability between the operating systems. On the other hand, within a single virtual machine, the user operating system is implemented in kernel mode, thereby allowing memory isolation (through a memory protection mechanism) to be provided for multiple processes running on it.
[0050] In an embodiment of the present invention, a plurality of virtual machines (virtual machine 1, virtual machine 2, ..., virtual machine N) may include a plurality of partitioned virtual machines 31 and at least one general virtual machine 32, such as Figure 3 As shown. In this document, "partitioned virtual machine" is intended to refer to a virtual machine with partition attributes defined based on the ARINC 653 specification, while "general virtual machine" is intended to refer to a virtual machine that does not participate in ARINC 653 tasks. The difference between the partitioned virtual machine 31 and the general virtual machine 32 is that the partitioned virtual machine 31 has more attributes, such as the cycle and duration of the exclusive processor defined by the ARINC 653 specification.
[0051] like Figure 3 As shown, the multiple operating systems (OS layers) may include multiple partition operating systems (POS) 131 based on the ARINC 653 standard and a general operating system (GOS) 132 not based on the ARINC 653 standard. The general operating system may run on a general virtual machine 32. As an example, the general operating system may include a relatively complex operating system such as Linux, Android, Windows, etc. In this way, multiple ARINC 653 applications 141 may run on multiple POS 131 respectively, and user applications (non-ARINC 653 applications) 142 may run on the general operating system. Note that although in Figure 3 Only one general-purpose virtual machine 32 is shown in the figure, but the present application is not limited thereto and any number of general-purpose virtual machines may be provided as required.
[0052] In an embodiment of the present invention, the hypervisor layer component 120 (specifically, MOS) can configure multiple partition virtual machines 31 to be isolated from each other and from the general virtual machine 32. Each partition virtual machine 31 is isolated from each other, which means that each partition virtual machine 31 has an independent memory space, can use the processor of the general module 100 independently, and has an independent running context, that is, the data and operations of each partition virtual machine 31 are independent of each other and will not be affected by other partition virtual machines 31 or general virtual machines 32. This isolation mechanism of the partition virtual machine 31 of the processing module 100 ensures the stability and reliability of the operation of the POS based on ARINC 653 and the entire avionics system.
[0053] In some embodiments of the present invention, multiple partition virtual machines 31 may communicate with each other using an inter-partition communication (IPC) mechanism. The inter-partition communication may comply with the APEX service interface standard, for example.
[0054] In some embodiments of the present invention, the processor 111 may be a multi-core processor. Multiple partition virtual machines 31 may alternately run on one of the cores of the multi-core processor, and the hypervisor layer component 120 (specifically, MOS) allocates cycles and durations of exclusive use of the core to each partition virtual machine 31. In these embodiments, if a general-purpose virtual machine exists, the other cores of the multi-core processor other than the core are allocated to run the general-purpose virtual machine, and when there is no general-purpose virtual machine, the other cores may be shut down.
[0055] See also Figure 4 , which shows an example of a time schedule for the processing module 100 to allocate the use of a multi-core processor to multiple virtual machines. In this example, the processing module 100 may include three partition virtual machines 31 and one general virtual machine 32. Three POS (POS 1, POS 2, and POS 3) run in the three partition virtual machines 31, respectively, and a general operating system runs on the general virtual machine 32. The processor 111 may use a quad-core (core 0, core 1, core 2, and core 3) processor, such as an AArch64 CPU defined by ARM v8 / v9.
[0056] like Figure 4As shown, MOS sets multiple main time slots for core 0. Each main time slot includes three sub-time slots. Sub-time slot 1 can be allocated to the first partition virtual machine 31, which means that in sub-time slot 1, POS 1 can monopolize core 0 of the processor of the processing module. Similarly, sub-time slot 2 can be allocated to the second partition virtual machine 31, which means that in sub-time slot 2, POS 2 can monopolize core 0 of the processor of the processing module. Sub-time slot 3 can be allocated to the third virtual machine 31, which means that in sub-time slot 3, POS 3 can monopolize core 0 of the processor of the processing module. In operation, each partition virtual machine can synchronize its corresponding clock to the clock of the processing module, so that each partition virtual machine only occupies processing resources in the sub-time slot to which it is allocated, thereby realizing the isolation of partition virtual machines in terms of processing resources. The allocation of the above time slots is also completed by MOS when multiple virtual machines are established, for example by establishing a time schedule. Note that in addition to the sub-time slots allocated to each partition virtual machine 31, the main time slot can also include idle time slots (that is, processing resources are not occupied). As shown Figure 4 As shown, MOS can allocate cores 1, 2, and 3 of a multi-core processor to a general virtual machine 32 for running a general operating system, such as Linux. In other words, in cores 1, 2, and 3 of the multi-core processor, no processes related to ARINC 653 are executed. Similarly, in the case of multiple general virtual machines 32, multiple general operating systems running on multiple general virtual machines 32 can use cores 1, 2, and 3 of the multi-core processor at the same time, and are not strictly restricted by time scheduling like partitioned virtual machines 31, because the requirements for resource isolation and protection between multiple general operating systems and between applications running on the general operating systems are usually low, that is, at a lower development assurance level (DAL).
[0057] Figure 5 The allocation of memory resources to each virtual machine in the processing module is shown. Specifically, Figure 5 The memory 500 is shown, which may be the memory 500 described above with reference to Figure 2 The memory 500 may include a memory space 1 510, a memory space 2 520, ..., a memory space N 530. As an example, the memory space 1 510 may be Figure 1 In the virtual machine 1 allocation, the memory space 2 520 can be Figure 1 The virtual machine 2 in the allocation, and the memory space N can be Figure 1 The above allocation is also completed by MOS when multiple virtual machines are established, for example, by establishing an address mapping table. As an example, the AArch64Stage 2 translation table can be used to perform memory partitioning.
[0058] In some embodiments of the present invention, the processing module 100 may also include an input-output memory management unit (IOMMU). The I / O memory management unit (IOMMU) can provide isolation and protection to prevent I / O devices from performing direct memory access (DMA) to memory. With the advancement of I / O device virtualization such as single-root I / O virtualization (SR-IOV), the IOMMU can also facilitate the direct assignment of peripherals to an operating system (OS) running on a virtual machine. In this way, the IOMMU can be used to isolate access to the same peripheral by different virtual machines. Compared with the prior art, the use of IOMMU and SR-IOV for peripheral partitioning can significantly improve I / O performance, because in the existing IMA processing module, the partition accesses the peripherals with the help of a device driver (e.g., end system) located in the MOS, which is usually low-performance.
[0059] During the configuration process of the virtual machine, MOS can allocate one or more peripherals 113 to multiple virtual machines for calling by multiple virtual machines. Ideally, a single peripheral 113 is only allocated to a single virtual machine to enhance the stability and reliability of the system. For example, MOS can allocate a specific peripheral in one or more peripherals so that the specific peripheral is only used by one virtual machine in multiple virtual machines. However, this requires too much quantity of peripherals 113 in some cases, which may be difficult to meet. Thus, in some embodiments, MOS can allocate one or more peripherals 113 so that one or more peripherals 113 are shared by multiple virtual machines. When one or more peripherals 113 are shared by multiple virtual machines, the following three methods can generally be used to implement the allocation of peripherals 113: (1) One or more peripherals 113 themselves can deterministically share their resources among multiple virtual machines. For example, a physical network interface card that supports single root I / O virtualization (SR-IOV) can switch network access permissions between different partitioned virtual machines and provide hardware-level isolation and protection; (2) MOS can virtualize one or more peripherals 113 to establish multiple virtual peripherals and allocate the multiple virtual peripherals among multiple virtual machines; or (3) one of the multiple virtual machines can be used as an agent to relay messages between one or more peripherals 113 and other virtual machines in the multiple virtual machines.
[0060] As previously described, multiple partition virtual machines 31 can alternately run on a single core (e.g., core 0) of a multi-core processor. In this case, if a peripheral 113 is shared by multiple partition virtual machines 31, then when multiple partition virtual machines 31 switch on the core, MOS will notify the peripheral 113 of the switch. In this way, one or more peripherals 113 can also perform corresponding switching to provide services to the current partition virtual machine running on the core and ensure data isolation. In some cases, when the peripheral 113 needs to perform direct memory access (DMA), the remaining running time of the current partition virtual machine running on the core is sent to the peripheral 113 to determine whether the remaining running time is sufficient for DMA conversion.
[0061] According to another aspect of the present invention, a method for establishing an IMA system is provided.
[0062] Figure 6 A flowchart of a method 600 for establishing an IMA system according to an embodiment of the present invention is shown. The method may be, for example, Figure 1-5 The processing module 100 described here performs.
[0063] The method 600 may include: providing hardware resources at step 610. The hardware resources may include a processor, a memory, and one or more peripherals, and the processor supports hardware-assisted virtualization.
[0064] The method 600 may further include: at step 630, establishing multiple virtual machines by means of hardware-assisted virtualization and allocating hardware resources to each virtual machine. The multiple virtual machines may include multiple partitioned virtual machines and at least one general virtual machine.
[0065] The method 600 may further include: at step 650, running multiple operating systems in multiple virtual machines respectively. The multiple operating systems may include multiple partition operating systems based on the ARINC 653 standard and a general operating system not based on the ARINC 653 standard. The multiple partition operating systems are respectively run on the multiple partition virtual machines, and the general operating system is run on the general virtual machine.
[0066] The method 600 may further include: at step 670, running corresponding application programs on multiple operating systems.
[0067] The processing module and related methods for integrated modular avionics systems according to exemplary embodiments of the present invention are described in detail above. The present invention proposes to use MOS as a hypervisor to implement the ARINC 653 system using the hardware virtualization extension provided by the CPU. Specifically, MOS can use the hardware support of the CPU to create multiple virtual machines, each of which has its own operating system (POS or GOS) and application software. These virtual machines are isolated from each other, thereby improving security and reliability. As a hypervisor, MOS provides a virtualization interface for POS and GOS to access hardware resources such as memory, processors and peripherals. When establishing a virtual machine, the hardware virtualization extension can help MOS to handle some virtualization operations, thereby simplifying the creation process of the virtual machine and improving the efficiency of creating the virtual machine. During the operation of the virtual machine, the hardware virtualization extension can hand over some virtualization operations to the CPU hardware for processing, thereby reducing the burden on the supervisory layer and improving the operating efficiency of the virtual machine. Preferably, peripheral partitioning can be further implemented by using an input / output memory management unit (IOMMU) and a single root I / O virtualization (SR-IOV). The advantages of using hardware virtualization to implement ARINC 653 are that it allows complex operating systems (such as Linux) to be hosted on the same hardware resources as ARINC 653 partitioned operating systems, because each operating system runs in its own virtual machine and is isolated from the other virtual machines; the partitions are almost equivalent to running directly on bare metal hardware, so the performance of the partitions will be improved; due to hardware virtualization, the software complexity of implementing space and time partitions can also be significantly reduced.
[0068] It should be understood that the above description is illustrative and not restrictive. For example, the above embodiments (and / or their various aspects) can be used in combination with each other. In addition, without departing from the scope of the present invention, many modifications can be made to adapt specific conditions or materials to the teachings of the various embodiments of the present invention. Although the size and type of the material described herein are used to define the parameters of the various embodiments of the present invention, the various embodiments are not meant to be restrictive, but exemplary embodiments. In the case of reading the above description, many other embodiments are obvious to those skilled in the art. Therefore, the scope of the various embodiments of the present invention should be determined with reference to the attached claims, and the full range of equivalent forms claimed by these claims.
Claims
1. A processing module for an integrated modular avionics system, comprising: Hardware resources including a processor, a memory, and one or more peripherals, wherein the processor supports hardware-assisted virtualization; A hypervisor layer component, including a module operating system, the module operating system is used to establish a plurality of virtual machines with the aid of the hardware-assisted virtualization and allocate the hardware resources to each virtual machine, wherein the plurality of virtual machines includes a plurality of partitioned virtual machines; An operating system layer component, comprising a plurality of operating systems respectively running in the plurality of virtual machines, wherein the plurality of operating systems comprises a plurality of partition operating systems based on the ARINC 653 standard, and the plurality of partition operating systems respectively run on the plurality of partition virtual machines; as well as An application layer component includes application programs running on the multiple operating systems.
2. The processing module according to claim 1, characterized in that The processor is a multi-core processor, wherein the multiple partition virtual machines run alternately on one core of the multi-core processor, and the module operating system allocates a cycle and duration for exclusive use of the core to each partition virtual machine.
3. The processing module according to claim 2, characterized in that At least one peripheral among the one or more peripherals is allocated to the plurality of partition virtual machines, and when the plurality of partition virtual machines are switched on the core, the module operating system notifies the at least one peripheral.
4. The processing module according to claim 3, characterized in that When the at least one peripheral device needs to perform direct memory access (DMA), the remaining running time of the current partition virtual machine running on the core is sent to the at least one peripheral device to determine whether the remaining running time is sufficient for direct memory access conversion.
5. The processing module according to claim 2, characterized in that The plurality of virtual machines further include a general-purpose virtual machine, and other cores of the multi-core processor except the core are allocated to run the general-purpose virtual machine.
6. The processing module according to claim 1, characterized in that The multiple operating systems also include a general operating system that is not based on the ARINC653 standard, and the general operating system runs on the general virtual machine.
7. The processing module according to claim 6, characterized in that The module operating system configures the plurality of partition virtual machines to be isolated from each other and from the general virtual machine.
8. The processing module according to claim 1, characterized in that The plurality of partition virtual machines communicate with each other using an inter-partition communication (IPC) mechanism.
9. The processing module according to claim 1, characterized in that The module operating system allocates a specific peripheral among the one or more peripherals so that the specific peripheral is used by only one virtual machine among the plurality of virtual machines.
10. The processing module according to claim 1, characterized in that The module operating system allocates the one or more peripherals such that the one or more peripherals are shared by the multiple virtual machines.
11. The processing module according to claim 10, characterized in that: The one or more peripheral devices themselves are capable of deterministically sharing their resources among the multiple virtual machines; The module operating system virtualizes the one or more peripherals to establish a plurality of virtual peripherals and distributes the plurality of virtual peripherals among the plurality of virtual machines; or A virtual machine among the plurality of virtual machines acts as a proxy to relay messages between the one or more peripherals and other virtual machines among the plurality of virtual machines.
12. The processing module according to claim 10, characterized in that The processing module further includes an input-output memory management unit (IOMMU), which is used to isolate access of different virtual machines to the same peripheral device.
13. The processing module according to claim 1, characterized in that The one or more peripheral devices support single root I / O virtualization (SR-IOV).
14. The processing module according to claim 1, characterized in that The partition operating system provides memory isolation between multiple processes in the partition virtual machine where the partition operating system is located.
15. A method for building an integrated modular avionics system, comprising: Providing hardware resources, the hardware resources comprising a processor, a memory, and one or more peripherals, wherein the processor supports hardware-assisted virtualization; Establishing a plurality of virtual machines by means of the hardware-assisted virtualization and allocating the hardware resources to each virtual machine, wherein the plurality of virtual machines includes a plurality of partitioned virtual machines; Running a plurality of operating systems in the plurality of virtual machines respectively, wherein the plurality of operating systems include a plurality of partition operating systems based on the ARINC653 standard, and the plurality of partition operating systems are respectively run on the plurality of partition virtual machines; as well as Corresponding application programs are run on the multiple operating systems.
16. The method according to claim 15, characterized in that The processor is a multi-core processor, wherein the method further comprises enabling the plurality of partition virtual machines to run alternately on a core of the multi-core processor, and allocating a cycle and a duration for exclusive use of the core to each partition virtual machine.
17. The method according to claim 16, characterized in that The plurality of virtual machines further include a general-purpose virtual machine, and other cores of the multi-core processor except the core are allocated to run the general-purpose virtual machine.
18. The method according to claim 15, characterized in that The multiple operating systems also include a general operating system that is not based on the ARINC653 standard, and the general operating system runs on the general virtual machine.
19. The method according to claim 15, characterized in that Allocating the hardware resources includes allocating the one or more peripherals such that the one or more peripherals are shared by the plurality of virtual machines.
20. The method of claim 19, wherein: The method also includes using an input-output memory management unit to isolate access to the same peripheral device by different virtual machines.
