Computer system, starting method, sleeping method and awakening method
By configuring different hardware redundancy and safety firmware on a single SoC chip, the isolation of different functional safety levels subsystems in the autonomous driving system is solved, and the cost problem caused by multiple chips is achieved, and different functions are effectively isolated on a single chip The subsystem of the security level domain reduces costs and meets high isolation requirements.
Patent Information
- Application Number
- CN202310574053.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-05-19
- Publication Date
- 2024-11-22
AI Technical Summary
In the prior art, in order to achieve isolation of subsystems of different functional safety levels in autonomous driving systems, multiple system-level chips are usually required, which leads to high cost and difficulty in achieving effective isolation on a single SoC chip.
By configuring the isolation mechanism on the same SoC chip using different hardware redundancy and secure firmware configuration, subsystem deployment of different functional safety levels, such as the ASIL-D domain and the ASIL-B domain, respectively, configuring the lock step core and the unlock step respectively. and limit access permissions through trusted firmware, achieving full isolation between hardware and software stacks.
Implementing effective isolation of different functional safety level domains on a single SoC chip reduces the cost of computer systems, while meeting the isolation requirements of high functional safety level, and avoiding the impact of low functional safety level subsystems on high functional safety level subsystems .
Smart Images

Figure CN119004559A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of computer technology, and in particular to a computer system, a startup method, a sleep method, and a wake-up method. Background Art
[0002] As computer systems become more and more complex, a computer system will include multiple subsystems. How to ensure the normal operation of each subsystem in the computer system and avoid interference with other normally operating subsystems when some subsystems have abnormalities is a relatively important issue.
[0003] Especially for systems on equipment with high safety requirements, there are higher requirements for the safety of subsystems (also known as reliability) and the isolation between subsystems. For example, the vehicle's autonomous driving system includes multiple subsystems, and multiple subsystems need to meet different functional safety level requirements based on the assessment of safety risks. In order to evaluate the safety level of each subsystem in the vehicle's autonomous driving system, ISO26262 "Road Vehicle Functional Safety" introduces the concept of automotive safety integrity level (ASIL) and defines four different ASILs: ASIL-A, ASIL-B, ASIL-C, and ASIL-D. Among them, ASIL-D is the highest functional safety level and ASIL-A is the lowest functional safety level. In addition, if the risk is identified as QM, there is no functional safety level requirement. According to different functional safety levels, the subsystems in the autonomous driving system can be divided into different functional safety level domains (also called ASIL domains). The subsystems in different functional safety level domains need to meet the isolation requirements. Even if a subsystem in one ASIL domain fails or an accident occurs, it will not affect the normal operation of the subsystem in another ASIL domain. This requires strict isolation between different ASIL domains. If the isolation is not thorough, it will cause catastrophic consequences.
[0004] In the prior art, in order to ensure the isolation requirements between subsystems in different ASIL domains, the usual implementation solution is to use multiple system-on-chips (SoCs) to form autonomous driving systems with different functional safety levels. For example, one ASIL-D MCU and one ASIL-B SoC. However, using multiple SoC chips will lead to relatively high costs. Summary of the invention
[0005] An embodiment of the present application provides a computer system that, through different hardware redundancy and security firmware configuration isolation mechanisms on the same SoC chip, enables subsystems with different functional safety levels to be deployed on the same SoC, and meets the safety requirements and isolation requirements of subsystems with different functional safety levels, effectively reducing costs.
[0006] In a first aspect, the present application provides a computer system comprising a first subsystem and a second subsystem, wherein the hardware layer of the first subsystem comprises a first CPU core, which is configured as a lock-step core, and the software layer comprises a first trusted firmware; the hardware layer of the second subsystem comprises a second CPU core, which is configured as a non-lock-step core, and the software layer comprises a second trusted firmware; wherein the functional safety level of the first subsystem is higher than the functional safety level of the second subsystem, and the first subsystem and the second subsystem are integrated on the same SoC; the first trusted firmware configures a functional safety isolation mechanism of the first subsystem so that the second subsystem has read-only access to resources of the first subsystem.
[0007] The computer system provided by the present application deploys subsystems of different functional safety levels on the same SoC chip, and different hardware redundancies are set on the areas of the SoC where subsystems of different functional safety levels are deployed. For example, for the part where the ASIL-D domain subsystem is deployed, the CPU core is configured as a lock-step core to provide the reliability required for this level, and for the part where the ASIL-B domain subsystem is deployed, the CPU core is configured as a non-lock-step core to provide reliability matching this level, thereby achieving isolation of domains of different functional safety levels at the hardware layer; then, an isolation mechanism is configured through the trusted firmware on the ASIL-D domain to limit the access of the ASIL-B domain subsystem to the resources of the ASIL-D domain subsystem, for example, the ASIL-B domain subsystem is only allowed to read the resources of the ASIL-D domain subsystem, and write access is prohibited, thereby achieving basic isolation of different functional safety levels of the software stack; and then achieving full-stack hardware and software isolation, achieving effective isolation of subsystems of different functional safety level domains on a single SoC, and effectively reducing the cost of the computer system.
[0008] In one possible implementation, the first subsystem is configured to extract functional safety level information from the received access request information, where the functional safety level information indicates the functional safety level of the access initiator; determine that the functional safety level of the access initiator is lower than the functional safety level of the first subsystem, and the access right of the access initiator is read-only access.
[0009] Exemplarily, the second subsystem sends an access request to the first subsystem, and the access request carries functional safety level information of the second subsystem. After receiving the access request sent by the second subsystem, the first subsystem extracts the functional safety level information carried in the access request, obtains the functional safety level of the second subsystem, compares the functional safety level of the second subsystem with the functional safety level of the first subsystem, and determines that the functional safety level of the second subsystem is lower than the functional safety level of the first subsystem. In this case, the access request only has read access permission.
[0010] In one possible implementation, the functional safety level information includes level token information, where the level token information indicates the functional safety level of the access initiator.
[0011] In another possible implementation, the first subsystem further includes a first memory, which is configured as a mirroring attribute; the second subsystem further includes a second memory, which is configured as a non-mirroring attribute.
[0012] That is to say, the first subsystem also has memory redundancy, which improves the storage reliability of the first subsystem and further improves the security of the first subsystem.
[0013] In another possible implementation, the functional safety isolation mechanism instructs the second subsystem to read-only access the first memory, and the second subsystem is also isolated in the memory, further improving the effectiveness of the isolation.
[0014] In another possible implementation, the second CPU core includes a second CPU master core and a second CPU slave core; when powered on, the first trusted firmware runs, and the first CPU core performs the following operations: initializes the first subsystem; resets the second CPU master core; runs the second trusted firmware, and the second CPU master core performs the following operations: resets the second CPU slave core.
[0015] On the basis of effective isolation of the full stack of software and hardware of the first subsystem and the second subsystem, the startup of the first subsystem and the second subsystem is achieved through the cooperation between the first trusted firmware and the second trusted firmware.
[0016] In another possible implementation, when powered on, the first trusted firmware runs, and the first CPU core also performs the following operations: initializes the functional safety isolation mechanism of the second subsystem, the functional safety isolation mechanism instructs the second subsystem to extract functional safety level information from the received access request information, the functional safety level information indicates the functional safety level of the access initiator; determines that the functional safety level of the access initiator is lower than the functional safety level of the second subsystem, and the access initiator's access right is read-only access.
[0017] In one example, the first subsystem communicates with the second subsystem by means of inter-core interrupts; or, the first subsystem communicates with the second subsystem by means of shared memory.
[0018] In another possible implementation, the first trusted firmware and the second trusted firmware are both ATF, for example, the first trusted firmware may be called ATF-1, and the second trusted firmware may be called ATF-2.
[0019] In another possible implementation, the second subsystem further includes a virtual layer, on which multiple operating systems are deployed.
[0020] In a second aspect, the present application provides a computer system startup method, which is applied to the computer system provided in the first aspect. The computer system startup method includes initializing a first subsystem with a first CPU core; enabling a functional safety isolation mechanism of the first subsystem with the first CPU core; resetting a second CPU main core with the first CPU core; and resetting a second CPU slave core with the second CPU main core.
[0021] In another possible implementation, the second subsystem further includes a virtual layer; the computer system startup method provided in the embodiment of the present application further includes: the first CPU core sends the address of the virtual layer to the second CPU core.
[0022] In another possible implementation, the computer system startup method provided in the embodiment of the present application further includes initializing the SoC by the first CPU core.
[0023] In another possible implementation, the first CPU core includes a first CPU master core and a first CPU slave core; the startup method provided in the embodiment of the present application also includes: the first CPU master core resets the first CPU slave core.
[0024] In a third aspect, the present application provides a sleep method for a computer system, which is applied to the computer system provided in the first aspect. The sleep method for the computer system includes: the second CPU core executes a sleep operation; the first CPU core shuts down the functional safety isolation mechanism of the second subsystem; the first CPU core executes a sleep operation; the first CPU core and the second CPU core are powered off.
[0025] In a possible implementation, the sleep method of the computer system provided in the embodiment of the present application further includes: the first CPU core detects whether the second CPU core enters a sleep state.
[0026] In another possible implementation, the sleep method of the computer system provided in the embodiment of the present application also includes: the first CPU core executes a SoC data backup operation.
[0027] In a fourth aspect, the present application provides a computer system wake-up method, which is applied to the computer system provided in the first aspect, and the computer system wake-up method includes: waking up the first CPU core; the first CPU core initializes the first subsystem; the first CPU core initializes the functional safety isolation mechanism of the second subsystem; the first CPU core resets the second CPU main core; the second CPU main core resets the second CPU slave core.
[0028] In one possible implementation, the second subsystem further includes a virtual layer; and the computer system wake-up method provided in the embodiment of the present application further includes the first CPU core sending the address of the virtual layer to the second CPU core.
[0029] In another possible implementation, the first CPU core initializes the SoC.
[0030] In another possible implementation, the first CPU core includes a first CPU master core and a first CPU slave core; the computer system wake-up method provided in the embodiment of the present application also includes: the first CPU master core resets the first CPU slave core.
[0031] In the fifth aspect, the present application provides a SoC chip, which is divided into multiple domains, and the multiple domains include at least a first domain and a second domain; the first domain includes a first CPU core, and the first CPU core is configured as a lock-step core; the second domain includes a second CPU core, and the second CPU core is configured as a non-lock-step core; wherein, the functional safety level of the first domain is higher than the functional safety level of the second domain, the first domain runs a first trusted firmware, and the first trusted firmware configures a functional safety isolation mechanism of the first domain so that the second domain has read-only access to the resources of the first domain.
[0032] In a sixth aspect, the present application provides a computing device, including the computer system provided in the first aspect.
[0033] In a seventh aspect, the present application provides a vehicle, comprising the computer system described in the first aspect.
[0034] In an eighth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed in a computer, the computer is caused to execute the computer system startup method provided in the second aspect, the computer system sleep method provided in the third aspect, and the computer system wake-up method provided in the fourth aspect of the present application.
[0035] In the ninth aspect, the present application provides a computer program or a computer program product, which includes instructions. When the instructions are executed, the computer system startup method provided in the second aspect of the present application, the computer system sleep method provided in the third aspect, and the computer system wake-up method provided in the fourth aspect are implemented. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] Figure 1 A schematic block diagram of an intelligent driving system for a vehicle is shown;
[0037] Figure 2 A system architecture diagram of an intelligent driving system in the related art is shown;
[0038] Figure 3a A schematic diagram of the system architecture of a computer system provided in an embodiment of the present application is shown;
[0039] Figure 3b A schematic diagram of the system architecture of an intelligent driving system provided in an embodiment of the present application is shown;
[0040] Figure 4 A flowchart of a method for starting a computer system provided in an embodiment of the present application;
[0041] Figure 5 A schematic diagram showing a specific implementation of the method for starting a computer system provided in an embodiment of the present application is shown;
[0042] Figure 6 A flowchart of a sleep method for a computer system provided in an embodiment of the present application;
[0043] Figure 7 A schematic diagram showing a specific implementation of a sleep method for a computer system provided in an embodiment of the present application is shown;
[0044] Figure 8 A flowchart of a method for waking up a computer system provided in an embodiment of the present application;
[0045] Fig. 9 A schematic diagram showing a specific implementation of the computer system wake-up method provided in an embodiment of the present application is shown;
[0046] Fig.10 A schematic diagram of the structure of a chip provided in an embodiment of the present application. DETAILED DESCRIPTION
[0047] The technical solution of the present application is further described in detail below through the accompanying drawings and embodiments.
[0048] The computer system provided in the embodiment of the present application can be applied to any device with mixed functional safety level deployment requirements, such as vehicles, aircraft, high-speed railways, etc. In particular, it can be applied to the intelligent driving system of the vehicle. When the computer system provided in the embodiment of the present application is applied to the intelligent driving system of the vehicle, the (partition) isolation of different subsystems of the computer system is essentially the isolation of different subsystems in the autonomous driving system.
[0049] The following describes in detail the computer system provided in the embodiment of the present application by taking an intelligent driving system (also referred to as an automatic driving system) as an example.
[0050] The field of intelligent driving needs to meet higher ASIL functional safety indicators. At the same time, different subsystems and modules in the system have different functional safety level requirements. Therefore, subsystems with different functional safety levels will exist in the intelligent driving system. For example, there will be multiple subsystems with different functional safety levels such as ASIL-D and ASIL-B in the intelligent driving system; as well as QM (no functional safety level requirements) subsystems.
[0051] In order to meet the isolation requirements, the intelligent driving system needs to meet the following requirements: subsystems with higher functional safety levels in the intelligent driving system can read and write resources of subsystems with lower functional safety levels, but subsystems with lower functional safety levels can only read but not write resources of subsystems with higher functional safety levels.
[0052] It needs to be explained that functional safety is different from security. Functional safety is mainly designed from the dimension of personal safety, such as protection against electromagnetic interference and single point failure; security is mainly designed from the dimension of anti-attack, such as anti-tampering and prevention of sensitive data leakage.
[0053] Figure 1 FIG. 1 shows a schematic block diagram of an intelligent driving system for a vehicle. Figure 1 As shown, in order to realize the intelligent driving function, the intelligent driving system of the vehicle includes multiple functional domains, such as the vehicle control domain, the sensor (such as radar, camera, etc.) data processing domain and the external communication domain. The vehicle control domain is used to issue vehicle control instructions to control the vehicle. For example, the vehicle control domain issues braking, acceleration, deceleration and turning instructions to control the vehicle to brake, accelerate, decelerate and turn. The sensor data processing domain is used for the collection and processing of sensor data, such as the collection and processing of sensor data such as cameras and radars to perceive the environment around the vehicle. The external communication domain is used for map updates and version updates, such as the external communication domain communicates with the cloud to realize map updates and version updates.
[0054] Multiple domains in the intelligent driving system need to meet different functional safety level requirements based on safety risk assessment. Through ISO26262, it can be determined that the functional safety level of the vehicle control domain in the intelligent driving system is ASIL-D, the functional safety level of the data processing domain such as cameras is ASIL-B, and the external communication domain has no functional safety level requirement and is quality management (QM).
[0055] In the intelligent driving system, the subsystems with different functional safety levels need to meet the isolation requirements. In order to meet the isolation requirements, in the related technology, multiple chips are usually set up, and the subsystems with different functional safety levels are set on different chips to achieve the isolation requirements between different functional safety level domains.
[0056] Figure 2 FIG. 1 shows a system architecture diagram of an intelligent driving system in the related art. Figure 2 As shown, the ASIL-D domain is set on the MCU chip, and the ASIL-B domain is set on the SoC chip to achieve isolation between the ASIL-D domain and the ASIL-B domain.
[0057] In related technologies, intelligent driving systems require a combination of multiple chips to achieve isolation of subsystems with different functional safety levels, which undoubtedly increases the cost of the intelligent driving system.
[0058] It should be explained that the ASIL-D domain includes several subsystems with ASIL-D functional safety levels, and the ASIL-B domain includes several subsystems with ASIL-B functional safety levels.
[0059] In order to solve the above problems, an embodiment of the present application provides a computer system that can achieve isolation of domains with different functional safety levels in a single SoC, and deploy multiple subsystems with different functional safety levels in a single SoC, effectively reducing the number of chips required and reducing costs.
[0060] Figure 3a FIG. 1 shows an architecture diagram of a computer system provided by an embodiment of the present application. Figure 3a As shown, the computer system includes a first subsystem and a second subsystem, wherein the first subsystem includes a software layer and a hardware layer, the software layer of the first subsystem is set with a first trusted firmware, the software layer of the second subsystem is set with a second trusted firmware, and the hardware layers of the first subsystem and the second subsystem share the same chip. In other words, the first subsystem and the second subsystem are integrated and deployed on the same chip (e.g., a SoC chip).
[0061] The functional safety level of the first subsystem is higher than that of the second subsystem. The chip includes multiple CPU cores. The CPU core corresponding to the first subsystem is the first CPU core, and the CPU core corresponding to the second subsystem is the second CPU core. The first CPU core is configured as a lock-step core to meet the hardware redundancy requirements of the functional safety level of the first subsystem, and the second CPU core is configured as a non-lock-step core to match the reliability requirements of the functional safety level of the second subsystem.
[0062] The first trusted firmware configures the functional safety isolation mechanism of the first subsystem so that the second subsystem has read-only access to resources of the first subsystem.
[0063] The computer system provided by the present application deploys subsystems of different functional safety levels on the same SoC chip, and sets different hardware redundancies in the areas where subsystems of different functional safety levels are deployed. For example, for the part where the ASIL-D domain subsystem is deployed, the CPU core is configured as a lock-step core to provide the reliability required for this level, and for the part where the ASIL-B domain subsystem is deployed, the CPU core is configured as a non-lock-step core to provide reliability matching this level, thereby achieving isolation of domains of different functional safety levels at the hardware layer; then, an isolation mechanism is configured through the trusted firmware on the ASIL-D domain to limit the access of the ASIL-B domain subsystem to the resources of the ASIL-D domain subsystem, for example, the ASIL-B domain subsystem is only allowed to read the resources of the ASIL-D domain subsystem, and write access is prohibited, thereby achieving basic isolation of different functional safety levels of the software stack; and then achieving full-stack hardware and software isolation, achieving effective isolation of subsystems of different functional safety level domains on a single SoC, and effectively reducing the cost of the computer system.
[0064] When the computer system provided in the embodiment of the present application is applied to a vehicle, it can also be referred to as an intelligent driving system, which can realize the deployment of subsystems with different functional safety levels of the vehicle's intelligent driving system on a single SoC and meet the isolation requirements of subsystems with different functional safety levels. It is understandable that when the computer system provided in the embodiment of the present application is applied to a vehicle, it can also be referred to as an intelligent driving system. The following takes the intelligent driving system as an example to introduce the computer system provided in the embodiment of the present application in detail.
[0065] The intelligent driving system includes at least ASIL-D domain subsystems and ASIL-B domain subsystems. In order to achieve the isolation of subsystems in different functional safety level domains on a single SoC, some CPU cores on the SoC are configured as ASIL-D level and some CPU cores are configured as ASIL-B level at the hardware layer. For example, some cores on the SoC are configured as lock-step cores to meet the reliability requirements of the ASIL-D level, and some cores on the SoC are configured as non-lock-step cores to meet the reliability requirements of the ASIL-B level, so as to achieve the isolation of different functional safety level domains at the hardware layer. At the software layer, trusted firmware is also set for the ASIL-D domain and the ASIL-B domain respectively. For example, the first trusted firmware is set for the ASIL-D domain to unify the underlying interface standards of the ASIL-D domain subsystem, such as the power status control interface (PSCI), the trusted board boot requirement (TBBR), and the secure monitor mode call (smc) operation for switching between the safe world state (SWS) and the normal world state (NWS); the second trusted firmware is set for the ASIL-B domain to unify the underlying interface standards of the ASIL-B domain subsystem, such as the power status control interface (PSCI), the trusted board boot requirement (TBBR), and the secure monitor mode call (smc) operation for switching between the safe world state (SWS) and the normal world state (NWS). By setting the first trusted firmware and the second trusted firmware respectively at the software layer, basic isolation of different functional safety level domains of the software stack is achieved.
[0066] The first trusted firmware configures the functional safety isolation mechanism of the ASIL-D domain, prohibiting subsystems with low functional safety levels from writing to the resources of the ASIL-D domain (such as registers, memory space, etc.); for example, prohibiting the ASIL-B domain from writing to the resources of the ASIL-D domain (such as registers, memory space, etc.), to achieve full-stack functional safety isolation. For example, configure the external inspection mechanism of the ASIL-D domain. When the subsystem of the ASIL-D domain receives an access request, the subsystem (such as AA, Dispatch, etc. at the subsystem entrance) extracts the functional safety level information carried in the access request and checks whether the functional safety level meets the requirements. If it does not meet the requirements, write access cannot be performed, thus establishing the functional safety isolation of the ASIL-D domain.
[0067] The meaning of satisfying the requirements here refers to whether the functional safety level is greater than or equal to the functional safety level of the accessed subsystem. If so, the requirements are satisfied; otherwise, the requirements are not satisfied.
[0068] When accessing other subsystems, subsystems in the intelligent driving system will carry their own functional safety level information. For example, when a subsystem in the ASIL-B domain accesses a subsystem in the ASIL-D domain, the access request sent to the ASIL-D domain will carry a level token, which indicates that the functional safety level of the subsystem in the ASIL-B domain is ASIL-B. After receiving the access request, the target subsystem in the ASIL-D domain extracts the level token and compares the functional safety level indicated by the level token. If it is less than ASIL-D, write access is prohibited and only read access is allowed.
[0069] That is to say, a subsystem with a high functional safety level can write to a subsystem with the same or lower functional safety level, while a subsystem with a lower functional safety level cannot write to a subsystem with a higher functional safety level, thereby achieving isolation between subsystems with low functional safety levels and subsystems with high functional safety levels.
[0070] Since a subsystem with a low functional safety level cannot write to a subsystem with a high functional safety level, the impact of an abnormality in a subsystem with a low functional safety level on a subsystem with a high functional safety level can be avoided.
[0071] Optionally, the first trusted firmware and the second trusted firmware may be ARM trusted firmware (ATF), the first trusted firmware may be ATF-1, and the second trusted firmware may be ATF-2.
[0072] Of course, the first trusted firmware and the second trusted firmware are not limited to ATF, and may also be other trusted firmware, such as self-developed trusted firmware.
[0073] Furthermore, at the hardware layer, memory space is configured for the ASIL-D domain and the ASIL-B domain respectively. For example, the memory space of the SoC, such as the double data rate synchronous dynamic random (DDR) memory space, configures DDR space for the ASIL-D domain and the ASIL-B domain respectively.
[0074] Optionally, on the SoC, the DDR space of the ASIL-D domain is configured as a mirror attribute, and the storage reliability requirements of the ASIL-D domain are further met through mirror redundancy. The DDR space of the ASIL-B domain is configured as a non-mirror attribute.
[0075] The functional safety isolation mechanism of the ASIL-D domain configured by the first trusted firmware also includes prohibiting subsystems with low functional safety levels from writing to the DDR space of the ASIL-D domain, for example, prohibiting the ASIL-B domain from writing to the DDR space of the ASIL-D domain. Exemplarily, the external inspection mechanism of the ASIL-D domain is configured. When the DDR space of the ASIL-D domain receives an access request, the DHA (part of the DDR access control channel) extracts the functional safety level information carried in the access request, and checks whether the functional safety level meets the requirements. If it does not meet the requirements, the write operation access cannot be performed, thereby establishing the functional safety isolation of the memory.
[0076] Subsystems between different functional safety level domains can communicate through a variety of communication methods, including but not limited to inter-core interrupts and shared memory.
[0077] For example, to share the memory space of an ASIL-B domain, a section of DDR space is divided in the DDR space of the ASIL-B domain as a shared memory space. Both the ASIL-D domain and the ASIL-B domain have the right to read and write the shared memory space, and can communicate by reading and writing in the shared memory. In this way, while subsystems with different functional safety levels are isolated, normal communication between subsystems with different functional safety levels is guaranteed.
[0078] Figure 3b FIG. 1 is a schematic diagram showing the system architecture of an intelligent driving system provided by an embodiment of the present application. Figure 3b As shown in the figure, the intelligent driving system includes domains with different functional safety levels, such as ASIL-D domain and ASIL-B domain. And both ASIL-D domain and ASIL-B domain are deployed on the same SoC chip. Some CPU cores of the SoC are configured as dual-core lock-step (DCLS), and some memory space is configured as Mirror attribute, as the hardware layer of the ASIL-D domain; some CPU cores of the SoC are configured as non-lock-step cores, and some memory space is configured as non-Mirror attribute, as the hardware layer of the ASIL-B domain, to achieve functional safety isolation of the ASIL-D domain and the ASIL-B domain at the hardware layer.
[0079] It is easy to understand that DCLS means assigning a task to two CPU cores for processing, and then comparing the processing results of the two CPU cores. If the comparison results are consistent, the result is passed. If the comparison results are inconsistent, the result is not passed, and the safety abnormal interrupt is reported to the fault processing center. In other words, redundant settings are added to the hardware to improve the reliability of the system.
[0080] The meaning of memory mirroring is to read or write two copies of a piece of data into the memory and compare the results. If they are consistent, it succeeds; otherwise, an error occurs and a safety exception alarm is reported.
[0081] At the software layer, ATF-1 is set for the ASIL-D domain and ATF-2 is set for the ASIL-B domain to achieve functional safety isolation between the ASIL-D domain and the ASIL-B domain in the software stack.
[0082] It can be understood that trusted firmware is a type of firmware, which is also called "hardened software". Firmware is the program code written inside the chip and is computer software stored in a non-lost way. The firmware serves as the most basic and lowest-level software of the computer system.
[0083] The upper layer of ATF-1 in the ASIL-D domain is the virtualization layer (Hypervisor) or the operating system layer (operation system, OS), and above the operating system layer is the application software layer (application, APP).
[0084] For example, when an ASIL-D domain has multiple operating systems, the ATF-1 upper layer needs to deploy a Hypervisor layer to virtualize and abstract the hardware resources of the ASIL-D domain (such as CPU cores, memory, network, and storage, etc.), presenting an environment that can be divided and combined into one or more computer configurations, so that the upper layer can use resources on demand, freeing up the operating system and deploying multiple operating systems. When an ASIL-D domain has only a single operating system, there is no need to set up a virtualization layer, and the OS layer can be directly deployed on the ATF-1 layer.
[0085] The upper layer of ATF-2 in the ASIL-B domain is the Hypervisor layer, above the Hypervisor layer are multiple OS layers, such as multiple Guest OS layers, and above the OS layer is the APP layer.
[0086] Understandably, Figure 3b The intelligent driving system shown is only an example and does not constitute a limitation on the driving system provided in the embodiment of the present application. The intelligent driving system may include more or fewer domains of functional safety levels. For example, the intelligent driving system may include an ASIL-D domain, an ASIL-C domain, an ASIL-B domain, an ASIL-A domain, and a QM domain.
[0087] When the intelligent driving system also includes a domain with a lower functional safety level than the ASIL-B domain, for example, the intelligent driving system also includes the QM domain, then ATF-1 needs to configure the functional safety isolation mechanism of the ASIL-B domain to prohibit subsystems with low functional safety levels from writing to the resources of the ASIL-B domain; for example, prohibit the QM domain from writing to the resources of the ASIL-B domain to achieve full-stack functional safety isolation. For example, configure the external inspection mechanism of the ASIL-B domain. When the subsystem of the ASIL-B domain receives an access request, the subsystem (such as AA and Dispatch at the subsystem entrance) extracts the functional safety level information carried in the access request and checks whether the functional safety level meets the requirements. If it does not meet the requirements, write access cannot be performed, thus establishing the functional safety isolation of the ASIL-B domain.
[0088] It should be noted that the functional safety level information can be expressed in a variety of ways. For example, the functional safety level information can be a level token. When a subsystem sends an access request to another subsystem, it will carry its own level token, which indicates the functional safety level of the subsystem itself. For another example, the functional safety level information can also be an information bit segment. When a subsystem sends an access request to another subsystem, it will carry an information bit segment indicating its own functional safety level.
[0089] For the intelligent driving system provided in the embodiments of the present application, the embodiments of the present application also provide a startup method, a sleep method and a wake-up method for the intelligent driving system.
[0090] Through the cooperation between ATF-1 and ATF-2, the intelligent driving system can be started, sleep and wake up.
[0091] Figure 4 A flowchart of a method for starting a computer system provided in an embodiment of the present application. The method can be applied to Figure 3a The computer system shown in the figure starts the computer system. Figure 4 As shown, the startup method at least includes steps S401 to S404.
[0092] In step S401 , ATF-1 initializes the subsystem of the ASIL-D domain.
[0093] ATF-1 runs, and the CPU core of the ASIL-D domain corresponding to ATF-1 performs initialization of the small system in the ASIL-D domain. For example, the functional safety mechanism and isolation enable of the ASIL-D domain are initialized; specifically, the CPU core of the ASIL-D domain is initialized and configured as DCLS, and the DDR memory space is initialized and configured as the Mirror attribute to meet the reliability requirements of the ASIL-D domain; and isolation is enabled to prohibit subsystems with low functional safety levels from writing to the resources of the ASIL-D domain. For example, the external inspection mechanism of the ASIL-D domain is initialized and configured. When the subsystem of the ASIL-D domain receives an access request, the entry of the subsystem will extract the functional safety level information carried in the access request and check whether the functional safety level meets the requirements. If it does not meet the requirements, write access cannot be performed, thus establishing the functional safety isolation of the ASIL-D domain.
[0094] It is easy to understand that the SoC chip will expose some registers, which can also be called interfaces, for software to configure. The software can configure the registers corresponding to the required function to enable this function. For example, ATF-1 configures the registers corresponding to DCLS and the registers corresponding to Mirror memory to enable DCLS and Mirror memory functions to achieve the hardware reliability requirements of the ASIL-D domain. ATF-1 configures the registers corresponding to the inspection function at the entrance of the system to achieve isolation enablement, such as extracting the functional safety level information carried in the access request and checking whether the functional safety level meets the requirements. If it does not meet the requirements, write access cannot be performed, thus establishing the functional safety isolation of the ASIL-D domain.
[0095] In step S402 , the functional safety isolation mechanism of the ASIL-B domain is initialized.
[0096] ATF-1 has higher permissions and can also be responsible for initializing the functional safety isolation mechanism of the ASIL-B domain. For example, it initializes the external inspection mechanism of the ASIL-B domain. When the subsystem of the ASIL-B domain receives an access request, the subsystem extracts the functional safety level information carried in the access request and checks whether the functional safety level meets the requirements. If it does not meet the requirements, write access cannot be performed, thus establishing the functional safety isolation of the ASIL-B domain.
[0097] In step S403 , the master core of the ASIL-B domain is de-reset.
[0098] ATF-1 de-resets the main core corresponding to ATF-2 in the ASIL-B domain and pulls up the main core of ATF-2. For example, ATF-1 configures the de-reset address of the main core of ATF-2 to perform a de-reset operation on the main core of ATF-2 and pull up the main core of ATF-2.
[0099] In step S404, ATF-2 deactivates the slave core of the ASIL-B domain.
[0100] After the master core of ATF-2 is pulled up, ATF-2 takes over the control, and ATF-2 resets the slave core of the ASIL-B domain to pull up the slave core of the ASIL-B domain.
[0101] The meaning of ATF-2 taking over control is that ATF-2 runs and the CPU core corresponding to ATF-2 executes the operation.
[0102] In one example, the startup method provided in the embodiment of the present application further includes more steps, such as:
[0103] ATF-1 is also responsible for the initialization of SoC. The meaning of SoC initialization can be understood as: initialization of the entire SoC, that is, initialization of some common modules or subsystems in the ASIL-D domain and ASIL-B domain in the SoC. These modules or subsystems do not belong to the ASIL-D domain or the ASIL-B domain, but the ASIL-D domain and ASIL-B domain are related to them in order to maintain normal operation, such as GICD, ITS, clock, power supply, storage space, etc.
[0104] ATF-1 de-resets the slave core of the ASIL-D domain, that is, ATF-1 de-resets the slave core of the ASIL-D domain and initializes the slave core of the ASIL-D domain.
[0105] ATF-1 sends the address of the virtual layer of the ASIL-B domain of the ATF-2 main core to inform the ATF-2 main core to return the address of the virtual layer. For example, ATF-1 sends the address of the virtual layer to the ATF-2 main core through shared memory or inter-core interrupts.
[0106] Figure 5 A schematic diagram of a specific implementation of the method for starting a computer system provided in an embodiment of the present application is shown. Figure 5As shown, through the cooperation of ATF-1 and ATF-2, system startup is realized under the condition of isolation of domains with different functional safety levels. For example, the master core of ATF-1 (running in the DDR of the ASIL-D domain) is responsible for the initialization of the overall part of the SOC (such as GICD and ITS shared by ASIL-D and ASIL-B), and then performs the initialization of the small system in the ASIL-D domain, the initialization and isolation enablement of the functional safety mechanism of the ASIL-B domain, and the reset and pull up the slave core of the ASIL-D domain. After pulling up its own slave core, it is responsible for pulling up the master core of ATF-2 (running on the DDR of the ASIL-B domain) (configure the reset address of the ATF-2 master core and perform core reset), and inform the ATF-2 master core by writing the address to the DDR of the shared low safety level domain (such as the DDR of ASIL-B), and return the address of the Hypervisor from ATF-2; then the control is handed over to ATF-2, ATF-2 pulls up its own slave core and initializes the slave core.
[0107] Figure 6 A flowchart of a sleep method for a computer system provided in an embodiment of the present application is shown. The sleep method can be applied to Figure 3a The computer system shown in the figure realizes the sleep of the computer system based on the isolation of different functional safety level domains. Figure 6 As shown, the sleeping method at least includes steps S601 to S604.
[0108] In step S601 , ATF-2 performs a sleep operation on the ASIL-B domain.
[0109] When the intelligent driving system is about to go to sleep, the ASIL-B domain goes to sleep first. For example, when the intelligent driving system is about to go to sleep, if the CPU core of the ASIL-B domain includes slave cores, ATF-2 first processes the slave cores in the ASIL-B domain in sequence to put them in a reset state, and then processes the master core of the ASIL-B domain to put the master core of the ASIL-B domain in a reset state.
[0110] For example, ATF-2 backs up data from the slave core of the ASIL-B domain, and shuts down the slave core of the ASIL-B domain after the data backup is completed; then the master core of the ASIL-B domain backs up data, and shuts down the master core of the ASIL-B domain after the data backup is completed, thus completing the sleep of the ASIL-B domain.
[0111] In step S602, ATF-1 turns off the functional safety isolation mechanism of the ASIL-B domain.
[0112] After ATF-2 completes the sleep of the ASIL-B domain, ATF-1 takes over the control and ATF-1 turns off the functional safety isolation mechanism of the ASIL-B domain.
[0113] In step S603 , ATF-1 performs a sleep operation on the ASIL-D domain.
[0114] ATF-1 performs sleep operations on the ASIL-D domain. For example, if the CPU core of the ASIL-D domain includes slave cores, ATF-1 first processes the slave cores in the ASIL-D domain in sequence to put them in a reset state, and then processes the master core of the ASIL-D domain to put the master core of the ASIL-D domain in a reset state.
[0115] For example, ATF-1 backs up data from the slave core of the ASIL-D domain, and shuts down the slave core of the ASIL-D domain after the data backup is completed; then the master core of the ASIL-D domain backs up data, and shuts down the master core of the ASIL-D domain after the data backup is completed, thus completing the sleep of the ASIL-D domain.
[0116] In step S604, the CPU core is powered off.
[0117] After the CPU cores in the ASIL-B domain and the ASIL-D domain are in the reset state, the CPU cores are powered off uniformly.
[0118] Of course, other power-off methods can also be used. For example, after the CPU cores of the ASIL-B domain and the ASIL-D domain of the intelligent driving system are in a reset state, some of the CPU cores are powered off first, then some of the CPU cores, and so on, until all the CPU cores are powered off.
[0119] In some practical examples, the sleep method of the computer system provided by the embodiments of the present application further includes one or more of the following steps:
[0120] ATF-1 backs up SoC data, that is, it backs up the data of some common modules or subsystems in the ASIL-D domain and ASIL-B domain in the SoC. These modules or subsystems do not belong to the ASIL-D domain or the ASIL-B domain, but the ASIL-D domain and ASIL-B domain are related to them in order to maintain normal operation, such as GICD, ITS, clock, power supply, storage space, etc.
[0121] Before the CPU is powered off, ATF-1 is also responsible for checking whether all CPU cores in the ASIL-B domain are in the reset state. For example, ATF-1 checks the status register corresponding to each CPU core in the ASIL-B domain to find out whether each CPU core in the ASIL-B domain is in the reset state. If a CPU core is found not in the reset state, it will be reset to ensure the reset of the CPU core in the ASIL-B domain.
[0122] Figure 7A schematic diagram of a specific implementation of a sleep method for a computer system provided in an embodiment of the present application is shown. Figure 7 As shown in the figure, when the system needs to sleep, ATF-2 performs ASIL-B domain slave core data backup and shutdown, ASIL-B domain slave core unplug, ASIL-B domain master core data backup and shutdown steps in sequence. Then ATF-1 takes over the control and performs ASIL-D domain slave core data backup and shutdown in sequence, ASIL-D domain slave core unplug, ASIL-D domain master core data backup, SoC data backup, ASIL-B domain functional safety mechanism shutdown, ASIL-D master core WFI, and finally the CPU is powered off and starts to sleep.
[0123] That is to say, when the system is asleep, the ASIL-B domain sleeps first (after the slave core data is backed up and shut down, the master core data is backed up and shut down), and then the ASIL-D domain sleeps (slave core data is backed up and shut down, then the master core data is backed up and waits for shutdown). The difference from the traditional processing flow is that ATF-1 is also responsible for checking the shutdown status of ATF-2 and shutting down the functional safety mechanism of the ASIL-B domain.
[0124] Figure 8 A flowchart of a method for waking up a computer system provided in an embodiment of the present application is shown. The waking up method can be applied to Figure 3a The computer system shown in the figure realizes the wake-up of the computer system based on the isolation of different functional safety level domains. Figure 8 As shown, the wake-up method at least includes steps S801 to S805.
[0125] In step S801 , ATF-1 wakes up the CPU core of the ASIL-D domain.
[0126] When the system is in sleep mode and needs to be awakened again, ATF-1 first wakes up the CPU core of the ASIL-D domain. For example, when there are slave cores in the ASIL-D domain, ATF-1 first wakes up the master core of the ASIL-D domain and then wakes up the slave core of the ASIL-D domain.
[0127] In step S802 , ATF-1 initializes the subsystem of the ASIL-D domain.
[0128] ATF-1 runs, and the CPU core of the ASIL-D domain corresponding to ATF-1 performs initialization of the small system in the ASIL-D domain. For example, the functional safety mechanism and isolation enable of the ASIL-D domain are initialized; specifically, the CPU core of the ASIL-D domain is initialized and configured as DCLS, and the DDR memory space is initialized and configured as the Mirror attribute to meet the reliability requirements of the ASIL-D domain; and isolation is enabled to prohibit subsystems with low functional safety levels from writing to the resources of the ASIL-D domain. For example, the external inspection mechanism of the ASIL-D domain is initialized and configured. When the subsystem of the ASIL-D domain receives an access request, the entry of the subsystem will extract the functional safety level information carried in the access request and check whether the functional safety level meets the requirements. If it does not meet the requirements, write access cannot be performed, thus establishing the functional safety isolation of the ASIL-D domain.
[0129] It is easy to understand that the SoC chip will expose some registers, which can also be called interfaces, for software to configure. The software can configure the registers corresponding to the required function to enable this function. For example, ATF-1 configures the registers corresponding to DCLS and the registers corresponding to Mirror memory to enable DCLS and Mirror memory functions to achieve the hardware reliability requirements of the ASIL-D domain. ATF-1 configures the registers corresponding to the inspection function at the entrance of the system to achieve isolation enablement, such as extracting the functional safety level information carried in the access request and checking whether the functional safety level meets the requirements. If it does not meet the requirements, write access cannot be performed, thus establishing the functional safety isolation of the ASIL-D domain.
[0130] In step S803, ATF-1 initializes the functional safety isolation mechanism of the ASIL-B domain.
[0131] ATF-1 has higher permissions and can also be responsible for initializing the functional safety isolation mechanism of the ASIL-B domain. For example, it initializes the external inspection mechanism of the ASIL-B domain. When the subsystem of the ASIL-B domain receives an access request, the subsystem extracts the functional safety level information carried in the access request and checks whether the functional safety level meets the requirements. If it does not meet the requirements, write access cannot be performed, thus establishing the functional safety isolation of the ASIL-B domain.
[0132] In step S804 , ATF-1 deactivates the master core of the ASIL-B domain.
[0133] ATF-1 de-resets the main core corresponding to ATF-2 in the ASIL-B domain and pulls up the main core of ATF-2. For example, ATF-1 configures the de-reset address of the main core of ATF-2 to perform a de-reset operation on the main core of ATF-2 and pull up the main core of ATF-2.
[0134] In step S805 , ATF-2 deactivates the slave core of the ASIL-B domain.
[0135] After the master core of ATF-2 is pulled up, ATF-2 takes over the control, and ATF-2 resets the slave core of the ASIL-B domain to pull up the slave core of the ASIL-B domain.
[0136] The meaning of ATF-2 taking over control is that ATF-2 runs and the CPU core corresponding to ATF-2 executes the operation.
[0137] In one example, the method for starting a computer system provided in an embodiment of the present application further includes further steps, such as:
[0138] ATF-1 is also responsible for the initialization of SoC. The meaning of SoC initialization can be understood as: initialization of the entire SoC, that is, initialization of some common modules or subsystems in the ASIL-D domain and ASIL-B domain in the SoC. These modules or subsystems do not belong to the ASIL-D domain or the ASIL-B domain, but the ASIL-D domain and ASIL-B domain are related to them in order to maintain normal operation, such as GICD, ITS, clock, power supply, storage space, etc.
[0139] ATF-1 de-resets the slave core of the ASIL-D domain, that is, ATF-1 de-resets the slave core of the ASIL-D domain and initializes the slave core of the ASIL-D domain.
[0140] ATF-1 sends the address of the virtual layer of the ASIL-B domain of the ATF-2 main core to inform the ATF-2 main core to return the address of the virtual layer. For example, ATF-1 sends the address of the virtual layer to the ATF-2 main core through shared memory or inter-core interrupts.
[0141] That is to say, when the system wakes up, ATF-1 first starts the master core of the ASIL-D domain, and then the slave core; then the master core of ATF-1 wakes up the master core of ATF-2, and ATF-2 then starts the slave core of the ASIL-B domain.
[0142] Fig. 9 A schematic diagram of a specific implementation of the computer system wake-up method provided in the embodiment of the present application is shown. Fig. 9 As shown, when the system wakes up, ATF-1 first runs and executes the following steps: initialization of the SoC small system as a whole, initialization of the small system in the ASIL-D domain, initialization of the functional safety mechanism and isolation enablement of the ASIL-B domain, pulling up and initializing the ASIL-D domain slave core, and releasing the reset to pull up the ATF-2 main core; then the small system in the ASIL-B domain is initialized, and the ASIL-B domain slave core is pulled up and initialized.
[0143] The embodiment of the present application also provides a chip.
[0144] Fig.10 A schematic diagram of the structure of a chip provided in an embodiment of the present application.
[0145] like Fig.10 As shown, the chip 1000 includes a processor 1001, a memory 1002 and a communication interface 1003. The processor 1001, the memory 1002 and the communication interface 1003 are connected in communication, and communication can also be achieved through other means such as wireless transmission. The communication interface 1003 is used to communicate with ECUs on other vehicles or the outside of the vehicle, such as receiving version update packages or map update packages from the cloud; the memory 1002 stores executable program code, and the processor 1001 can call the program code stored in the memory 1002 to execute the method in the aforementioned method embodiment.
[0146] The processor 1001 has multiple CPU cores, some of which are configured as ASIL-D level, and some of which are configured as ASIL-B level. For example, some of the cores on the SoC are configured as lock-step cores to meet the reliability requirements of the ASIL-D level, and some of the cores on the SoC are configured as non-lock-step cores to meet the reliability requirements of the ASIL-B level, thereby achieving isolation of different functional safety level domains at the hardware layer.
[0147] The memory 1002 includes DDR memory, some of which are configured as ASIL-D domains and some of which are configured as ASIL-B domains. For example, the DDR space of the ASIL-D domain is configured as a mirror attribute, and through mirror redundancy, the storage reliability requirements of the ASIL-D domain are further met. The DDR space of the ASIL-B domain is configured as a non-mirror attribute.
[0148] The first trusted firmware of the ASIL-D domain and the second trusted firmware of the ASIL-B domain are burned into the memory 1002. The first trusted firmware and the second trusted firmware can cooperate to implement the startup, sleep and wake-up method provided in the embodiment of the present application, so as to realize the startup, sleep and wake-up of the system in the system isolated for different functional safety level domains.
[0149] It should be understood that in the embodiment of the present application, the processor 1001 may be a central processing unit CPU, and the processor 1001 may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.
[0150] The memory 1002 may include a read-only memory and a random access memory, and provide instructions and data to the processor 1001. The memory 1002 may also include a non-volatile random access memory. For example, the memory 1002 may also store a training data set.
[0151] The memory 1002 may be a volatile memory or a nonvolatile memory, or may include both volatile and nonvolatile memories. Among them, the nonvolatile memory may be a read-only memory (ROM), a programmable ROM (PROM), an erasable programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).
[0152] It should be understood that the chip 1000 according to the embodiment of the present application can execute the Figures 4 to 9 For the specific implementation method, see the above Figures 4 to 9 For the sake of brevity, the specific description is omitted here.
[0153] The embodiment of the present application also provides a vehicle, on which a Figure 3b The intelligent driving system shown reduces the number of chips and effectively reduces the generation cost of the intelligent system by deploying subsystems with different functional safety levels on a single SoC.
[0154] An embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer instructions are executed by a processor, the above-mentioned method is implemented.
[0155] An embodiment of the present application provides a computing device, including a memory and a processor, wherein instructions are stored in the memory, and when the instructions are executed by the processor, the above-mentioned method is implemented.
[0156] An embodiment of the present application provides a computer program or a computer program product, wherein the computer program or the computer program product comprises instructions, and when the instructions are executed, the computer is caused to execute the above-mentioned method.
[0157] Those of ordinary skill in the art should further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those of ordinary skill in the art may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0158] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented by hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
[0159] The specific implementation methods described above further illustrate the purpose, technical solutions and beneficial effects of the present application in detail. It should be understood that the above description is only the specific implementation method of the present application and is not intended to limit the scope of protection of the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.
Claims
1. A computer system, characterized in that: include: A first subsystem, wherein the hardware layer includes a first CPU core, the first CPU core is configured as a lockstep core, and the software layer includes a first trusted firmware; A second subsystem, the hardware layer includes a second CPU core, the second CPU core is configured as a non-lockstep core, and the software layer includes a second trusted firmware; Wherein, the functional safety level of the first subsystem is higher than the functional safety level of the second subsystem, and the first subsystem and the second subsystem are integrated on the same SoC; The first trusted firmware configures a functional safety isolation mechanism of the first subsystem so that the second subsystem has read-only access to resources of the first subsystem.
2. The system according to claim 1, characterized in that The first subsystem is configured as follows: Extracting functional safety level information from the received access request information, wherein the functional safety level information indicates the functional safety level of the access initiator; It is determined that the functional safety level of the access initiator is lower than the functional safety level of the first subsystem, and the access right of the access initiator is read-only access.
3. The system according to claim 2, characterized in that The functional safety level information includes level token information indicating a functional safety level of an access initiator.
4. The system according to any one of claims 1 to 3, characterized in that: The first subsystem further includes a first memory, wherein the first memory is configured as a mirroring attribute; The second subsystem further includes a second memory, and the second memory is configured as a non-mirroring attribute.
5. The system according to claim 4, characterized in that The functional safety isolation mechanism instructs the second subsystem to access the first memory only through read-mode.
6. The system according to any one of claims 1 to 5, characterized in that: The second CPU core includes a second CPU master core and a second CPU slave core; When powered on, the first trusted firmware runs, and the first CPU core performs the following operations: Performing an initialization operation on the first subsystem; Resetting the second CPU main core; The second trusted firmware is run, and the second CPU main core performs the following operations: Reset the second CPU slave core.
7. The system according to claim 6, characterized in that When powered on, the first trusted firmware runs, and the first CPU core also performs the following operations: Initializing a functional safety isolation mechanism of the second subsystem, the functional safety isolation mechanism indicating that the second subsystem extracts functional safety level information from the received access request information, the functional safety level information indicating the functional safety level of the access initiator; It is determined that the functional safety level of the access initiator is lower than the functional safety level of the second subsystem, and the access right of the access initiator is read-only access.
8. The system according to any one of claims 1 to 7, characterized in that: The first subsystem communicates with the second subsystem via an inter-core interrupt; Alternatively, the first subsystem and the second subsystem communicate with each other by sharing memory.
9. The system according to any one of claims 1 to 8, characterized in that: The first trusted firmware and the second trusted firmware are both ATF.
10. A method for starting a computer system, characterized in that: Applied to the computer system according to any one of claims 1 to 9, the method comprises: The first CPU core initializes the first subsystem; The first CPU core initializes a functional safety isolation mechanism of the second subsystem; The first CPU core de-resets the second CPU main core; The second CPU master core resets the second CPU slave core.
11. The method according to claim 10, characterized in that The second subsystem also includes a virtual layer; The method further comprises: The first CPU core sends the address of the virtual layer to the second CPU core.
12. The method according to claim 10 or 11, characterized in that: Also includes: The first CPU core initializes the SoC.
13. The method according to any one of claims 10 to 12, characterized in that: The first CPU core includes a first CPU master core and a first CPU slave core; The method further comprises: The first CPU master core resets the first CPU slave core.
14. A sleep method for a computer system, characterized in that: Applied to the computer system according to any one of claims 1 to 9, the method comprises: The second CPU core performs a sleep operation; The first CPU core shuts down a functional safety isolation mechanism of the second subsystem; The first CPU core performs a sleep operation; The first CPU core and the second CPU core are powered off.
15. The method according to claim 14, characterized in that Also includes: The first CPU core detects whether the second CPU core enters a sleep state.
16. The method according to claim 14 or 15, characterized in that Also includes: The first CPU core performs a SoC data backup operation.
17. A method for waking up a computer system, characterized in that: Applied to the computer system according to any one of claims 1 to 9, the method comprises: Waking up the first CPU core; The first CPU core initializes the first subsystem; The first CPU core initializes a functional safety isolation mechanism of the second subsystem; The first CPU core de-resets the second CPU main core; The second CPU master core resets the second CPU slave core.
18. The method according to claim 17, characterized in that The second subsystem also includes a virtual layer; The method further comprises: The first CPU core sends the address of the virtual layer to the second CPU core.
19. The method according to claim 17 or 18, characterized in that The first CPU core initializes the SoC.
20. The method according to any one of claims 17 to 19, characterized in that: The first CPU core includes a first CPU master core and a first CPU slave core; The method further comprises: The first CPU master core resets the first CPU slave core.
21. A SoC chip, characterized in that: is divided into a plurality of domains, the plurality of domains comprising at least a first domain and a second domain; The first domain includes a first CPU core, the first CPU core being configured as a lockstep core; The second domain includes a second CPU core, the second CPU core being configured as a non-lockstep core; Among them, the functional safety level of the first domain is higher than the functional safety level of the second domain, the first domain runs a first trusted firmware, and the first trusted firmware configures the functional safety isolation mechanism of the first domain so that the second domain has read-only access to the resources of the first domain.
22. A computing device, characterized in that Comprising a computer system as described in any one of claims 1-9.
23. A vehicle, characterized in that: Comprising a computer system as described in any one of claims 1-9.
