CN100347676C - PowerPC processor based vehicular operating system interrupt management method - Google Patents

PowerPC processor based vehicular operating system interrupt management method Download PDF

Info

Publication number
CN100347676C
CN100347676C CNB2005100113797A CN200510011379A CN100347676C CN 100347676 C CN100347676 C CN 100347676C CN B2005100113797 A CNB2005100113797 A CN B2005100113797A CN 200510011379 A CN200510011379 A CN 200510011379A CN 100347676 C CN100347676 C CN 100347676C
Authority
CN
China
Prior art keywords
real
time
kernel
interrupt
task
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CNB2005100113797A
Other languages
Chinese (zh)
Other versions
CN1655124A (en
Inventor
孙晓民
周书平
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tsinghua University
Original Assignee
Tsinghua University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Tsinghua University filed Critical Tsinghua University
Priority to CNB2005100113797A priority Critical patent/CN100347676C/en
Publication of CN1655124A publication Critical patent/CN1655124A/en
Application granted granted Critical
Publication of CN100347676C publication Critical patent/CN100347676C/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

The present invention relates to an interrupt management method of a vehicular operation system based on a Power PC processor, which belongs to the technical field of an embedded real-time operation system. The present invention is characterized in that aiming at the embedded real-time operation system adopting a double-kernel structure realized by the Power PC processor, the present invention adopts different processors to store fields according to the operation state of the system when interrupt occurs; In the running state of the real-time kernel, the present invention adopts one-step storage and one-step recovery and thus reduces unnecessary stack operation. As one part of an open system of automobile electrons in Qinghua University, the method can be implemented on an MPC555 platform. A test indicates that the method improves the interrupt response speed and the system running efficiency of the real-time kernel in the double-kernel system.

Description

Vehicular operating system interrupt management method based on the PowerPC processor
Technical field
Vehicular operating system interrupt management method based on the PowerPC processor belongs to the embedded real-time operating system technical field, relates in particular to the embedded real-time operating system technical field based on the PowerPC processor structure.
Background technology
Interrupt management is one of the important step of real-time technique that adopts the (SuSE) Linux OS of dual core structure.The interrupt management of at present popular real-time Linux adopts the break simulation technology more, improves the real-time performance of general Linux.Promptly when Linux forbid/during opens interrupters, no longer directly to the hardware interrupts controller function, just by in addition record of global flag position corresponding on the break simulation device is set, and in fact, in non-real-time process and linux kernel operational process, interrupt always opening.When hardware interrupts took place, the real-time interrupt controller was intercepted and captured this interruption, judged then whether this interruption has the real-time requirement, was to handle or hang up this interruption immediately with decision.If in the real-time kernel ready attitude real-time process is arranged when this interruption takes place, and interrupt there is not real-time requirement, then immediately the CPU control is returned to operating system, and skipped over the process of searching the Linux interrupt vector table and carrying out the Linux interrupt service routine.
In the Interrupt Process process, stack management is related to stability, operational efficiency and the operating factor of memory space of operating system.Preserving on-the-spot, Interrupt Process when broken hair is given birth in the middle of its main task finishes the scene of preserving when the back recovers to interrupt taking place, moves definite rational stack space for interrupt service routine, if and had the real-time task of high priority in interrupt handling routine, to enter ready attitude, then need when interrupt service routine withdraws from, carry out task and switch.
At present, the real-time Linux of employing dual core structure is most typical is RTLinux and RTAI.In order to guarantee versatility, their storehouses in the Interrupt Process process switch the better simply implementation framework of employing, to reduce the size of code of system transplantation.But, present various processor is very big at the order set characteristic difference of aspects such as stack manipulation, addressing operation, if at specific processor structure design objective handover mechanism, will more help making full use of the function of par-ticular processor, improve the operational efficiency on this processor platform.
In the interrupt management of RTLinux, no matter which kind of state is system operate under when interrupting taking place, as long as the operation of Real-time Interrupt Processing program causes having the real-time task of high priority to enter ready attitude, system all will preserve single treatment device scene when Interrupt Process begins, and the generation task need be preserved single treatment device scene when switching again.After high-priority task had been moved, the scene that system preserves when at first recovering process is switched went out stacked rule according to function call then and pops, the scene of preserving when recovering to interrupt taking place at last.Yet under the situation that when real-time task is moved, takes place to interrupt, processor scene when the scene of preserving is real-time interrupt service routine end of run for the second time, and this scene is nonsensical, preserves it and has not only reduced the interrupt response speed of real-time kernel but also wasted the stack space of real-time task.
Summary of the invention
Combine with the processor characteristic at stack management in the interrupt management process of the existing real time operating system that adopts the dual core structure that undertighten, scene are preserved unnecessary operation, storehouse uses sub-economic problem, the present invention is directed to the PowerPC structure and proposed a flexible and efficient stack management method, mainly comprise following content:
1. in same Interrupt Process flow process, centering is disconnected to be occurred in different situations and carries out differentiated treatment; At the situation of interrupting occurring in the real-time task operational process, the Interrupt Process flow process is once preserved the strategy that once recovers to real-time task operation field by using.
2. adopt the interrupt stack management at the real-time kernel run mode, reduce the storehouse user demand of real-time task.
The invention is characterized in: it is the real time operating system that adopts dual core mechanism, described dual core mechanism is the method for standard non real-time Linux being carried out hard real-timeization, promptly by between standard non real-time linux kernel and bottom hardware, inserting a little real-time kernel, the management of real-time task is provided by real-time kernel, described linux kernel has only the operation of just obtaining as the minimum real-time task of the priority of real-time kernel when not having real-time task to be in ready attitude; At the situation of interrupting occurring in the real-time task operational process, the Interrupt Process flow process is once preserved the strategy that once recovers to real-time task operation field by using; Correspondingly this vehicular operating system interrupt management method contains following steps successively:
Step 1. is set:
Described real time operating system logically is divided into following four ranks:
The non-real-time process rank: the non real-time program that run user is write, the operation process adopts the dynamic priority strategy, and by the management and running of Linux scheduler program, its processing rank is minimum;
The real-time process rank: real-time application that the user writes and linux kernel run on the real-time process rank, and described linux kernel is minimum as its priority of special processing by the real-time kernel management, is " 0 "; It supports dynamic object to create, and Dram distributes; It carries out the simulation interrupt management;
The real-time kernel rank: real-time kernel is driven by non-real-time process and linux kernel, runs on the real-time kernel rank;
The real-time interrupt rank: its operation real-time interrupt service routine, and drive break simulation management Linux interrupt handling routine;
Described operating system is provided with following three kinds of stack type:
User stack, it is prepared for the program run of user's attitude;
Be the real-time process storehouse that real-time process is distributed, Linux real-time process storehouse wherein is corresponding to the core stack of current process in the linux kernel;
Real-time interrupt storehouse: operation Real-time Interrupt Processing program;
Step 2. makes up on the PowerPC processor with lower module:
Process creation and storehouse distribution module, it comprises:
The establishment of un-real time job and storehouse distribution sub module: it is being each each space below un-real time job distribution on the described linux kernel:
Process control block (PCB), expression state of a process and the resource that has;
Kernel stack, it is the memory headroom for operational system service and break in service distribution;
The task run space is provided with user stack and is used for the operation task code;
Real-time task is created and the storehouse distribution sub module: it runs on the real-time kernel, and described system distributes a real-time task storehouse for each real-time task and is fixed in the memory headroom of reserving when system kernel generates;
System's dual core operational module comprises the activate a task that calls for system start-up code, user task code, interrupt handling routine and choose a user task from the ready task tabulation jumping to a kind of like this scheduler operation that this task is carried out again; System's dual core operation code is supported preemption scheduling, promptly when the user task of user task code or the actuator-activated high priority of interrupt handling routine calling system real-time kernel, the activate a task operation is just called scheduler and is selected new activating of task, pressure interrupts the lower user task of priority that is moving, and the task of forwarding activation to goes to carry out;
The interrupt handling routine operational module, it contains:
The interrupt handling routine inlet: wherein, the external interrupt handling procedure is placed on the 0x500 place, entry address of MPC555 microcontroller, and the counter interrupt handling routine is placed on 0x900 place, entry address;
The real-time interrupt service routine: the priority of setting the Real-time Interrupt Processing program is higher than real-time task priority;
The non real-time interrupt handling routine: the task in the setting non real-time kernel is as the minimum real-time task of the priority of real-time kernel, in the non real-time kernel, non real-time interrupt handling routine and kernel system service run on kernel state, and its priority is higher than the task of user's attitude;
Interrupt handling routine finishes: by the new task that user interrupt service routine has activated, carry out by scheduler after can only waiting the interrupt handling routine to finish again;
The described interrupt handling routine of step 3. contains following each step successively:
Step 3.1 is preserved on-the-spot:
In generation, have no progeny,, just stack pointer is switched to interrupt stack from user stack, enter kernel state, preserve on-the-spot again as long as described system is in real-time or non real-time user's attitude; Otherwise system has been in kernel state, directly keeps on-the-spot;
Step 3.2 is judged the state when taking place to interrupt:
If user's attitude, the operation of carry out preserving process address successively, preserving the process stack pointer and obtaining interrupting channel number;
If kernel state: whether kernel is busy when at first judge to interrupt taking place, if kernel is not in a hurry, then preserves process address, preserves the process stack pointer again and obtains interrupting channel number; If kernel is busy, judge earlier then whether real-time kernel interrupts nested, as if nested, then preserves the process stack pointer, obtain interrupting channel number again; If not nested, then earlier switch to the real-time interrupt storehouse from the real-time process storehouse, remove the variable that real-time kernel need be dispatched, then go to obtain interrupting channel number again;
Step 3.3 judges whether to be real-time interrupt: if real-time interrupt is then carried out the Real-time Interrupt Processing program, the break simulation device is set; Otherwise, the break simulation device just directly is set;
Step 3.4 judges whether described system meets the following conditions simultaneously: real-time kernel is not in a hurry, the break simulation device be in open interruption status, the interrupt response passage is unimpeded, if satisfy above three conditions simultaneously, then directly enter the break simulation device and call interrupt service routine, make End of Interrupt then and handle; If real-time kernel is not in a hurry, and other conditions do not satisfy, and then directly make End of Interrupt and handle; If above three conditions do not satisfy, judge then whether the real-time kernel storehouse is nested;
Step 3.5 is then directly made End of Interrupt and is handled if the real-time kernel storehouse is nested; If it is not nested that real-time kernel interrupts, just switch to the real-time process storehouse from the real-time interrupt storehouse, need then to judge whether Real-Time Scheduling, if need Real-Time Scheduling, just carry out the Real-Time Scheduling operation, otherwise, make End of Interrupt and handle.
Draw through testing on the MPC555 platform, interrupting entering user's interrupt handling routine required time with method of the present invention from generation is 14 microseconds, and 25 microseconds during than the employing universal method are lacked, and have improved interrupt response speed.
Description of drawings
Fig. 1 .MPC555 register classification synoptic diagram;
The memory-mapped synoptic diagram at Fig. 2 .MPC555 microcontroller CPU scene;
Fig. 3 .OSEKLinux stack type synoptic diagram;
Fig. 4. un-real time job storage organization synoptic diagram;
Fig. 5. real-time task storage organization synoptic diagram;
Fig. 6. the interrupt management method flow diagram.
Embodiment
The method that the present invention proposes is the part of embedded real-time operating system kernel, with the form realization of operating system code.Below with the OSEKLinux operating system of Tsing-Hua University's independent development, be example based on the hardware platform environment of MPC555 microcontroller, the embodiment of interrupt management method is described.
Dual core mechanism is a kind of method of standard non real-time Linux being carried out hard real-timeization, by between standard Linux kernel and hardware, inserting a little real-time kernel, real-time kernel provides the management of real-time task, linux kernel has only the operation of just obtaining as the minimum real-time task of the priority of real-time kernel when not having real-time task to be in ready attitude.
OSEKLinux adopts the system model design proposal of the real-time framework of dual core: general-purpose operating system linux kernel and a real-time real-time kernel that meets the OSEK standard coexist as in the system, the OSEK real-time kernel is embedded between linux kernel and the bottom hardware, real-time kernel provides running environment for real-time task, and linux kernel is used for moving un-real time job.
OSEKLinux operating system is handled rank from being divided into four in logic:
The non-real-time process rank
The non real-time program run that the user writes is in the non-real-time process rank, and these processes adopt the dynamic priority strategy, by the management and running of Linux scheduler program.Handle in the rank at four, the non-real-time process rank is minimum processing rank, has only when other rank does not have processing requirements, and it just can be handled.
The real-time process rank
Real-time application that the user writes and linux kernel run on the real-time process rank, and linux kernel is as a special processing by the real-time kernel management, and its priority is minimum, and it monopolizes real-time process priority " 0 ".Linux kernel is except being replaced original switch interrupts management by the simulation interrupt management, no longer directly control hardware comes outside the switch interrupts, and other mechanism and DLL (dynamic link library) are basic identical; Linux kernel is supported the establishment of dynamic object in the dual core of OSEKLinux, and Dram distributes, and non-real-time process and linux kernel can not call all system services that real-time kernel provides except starting real-time kernel.Other real-time processes are write by the user, because these processes run under the kernel state of traditional Linux meaning, so the user need guarantee its correctness, otherwise can have influence on the safety of total system.Real-time process is dispatched according to the static priority that the user distributes.
The real-time kernel rank
Real-time kernel runs on the real-time kernel rank, it provides system service for the real-time application that the user writes, the system object of real-time kernel management is static allocation, and by user's static configuration, real-time kernel is not supported dynamic allocation of memory mechanism when generating in system.
The real-time interrupt rank
Interrupt levels moves real-time interrupt service routine (ISR), comprises two class Interrupt Process service routines of real-time kernel, drives break simulation management and Linux interrupt handling routine.
According to the difference of real-time, OSEKLinux is divided into two big classes to process, promptly in common process of moving on the linux kernel and the real-time process of moving on real-time kernel.There is not the task of requirement in real time and more complicated to move in the mode of Linux process; Real-time task is then moved with the form of real-time process.
In the automobile-used embedded real-time operating system of OSEKLinux, when processor or interruption of equipment generation, operating system can respond to this interruption, and the handling procedure of operation response interruption.Because this process takes place very frequent in the real-time system operational process, the quality of this handling procedure directly affects the operational efficiency of total system, the process stacks management that the present invention is directed in the Interrupt Process process has proposed a kind of Governance framework, to improve the treatment effeciency of OSEKLinux Real-time Interrupt Processing process.
Stack type
Three kinds of storehouses are arranged in the automobile-used embedded real-time operating system of OSEKLinux.As shown in Figure 3, a kind of is the user stack of preparing for the program run of user's attitude; A kind of is the real-time process storehouse that distributes for real-time process, and wherein the storehouse of Linux real-time process is an abstract stack management, and what reality was corresponding is the core stack of current process in the linux kernel; Also having a kind of is the real-time interrupt storehouse, moves the Real-time Interrupt Processing program specially.
The Interrupt Process process
As shown in Figure 4, the treatment scheme of this interrupt management method is as follows:
When 1. central broken hair is given birth to, whether be 0 at first by the value of judging the SPRG2 register, this register is used for preserving in the core stack for preserving the plot that stack space is reserved in the interrupt spot by linux kernel, if this value is 0, illustrate that stack pointer is in core stack when interrupting taking place, the system that also we can say can move un-real time job scarcely, and can directly be pressed into storehouse with the scene this moment; If the value of SPRG2 register is not 0, then the scene being saved in this value is in the stack of plot, and current stack pointer is adjusted into this value, will realize the switching of user stack to the core stack.
2. after on-the-spot reservation finished, the break simulation processor of real-time kernel was taken over disposal right, at first judged the running status of system when taking place to interrupt, if be derived from real-time kernel, then system need handle the preparation work of OSEK real-time kernel Interrupt Process.Interrupt source has two kinds of situations from real-time kernel, a kind of is that interrupt nesting does not take place, in this case stack pointer is transformed in the interrupt stack from process stacks, and represents that with one whether interruption needs the global variable zero clearing of dispatching when withdrawing from, enter next step processing then; Another kind of situation is when interrupting generation, real-time kernel is being handled the Real-time Interrupt Processing program, interrupt nesting promptly takes place, need to increase nested progression this moment, but because this moment, system operated in the interrupt stack, and the variable whether expression needs to dispatch is cleared, so can directly enter next step processing.
3. system obtains interrupting channel number by calling special function, and by corresponding position on the rtirq territory on the osek_global structure, judge whether this passage is real-time interrupt, if, then jump to osek_global_handlers[] the respective handling function, carry out the real-time interrupt service routine, use in order to linux kernel if not then directly this passage being recorded on the structural pending of the osek_global territory.
4. next step will determine whether will carry out the linux kernel Interrupt Process, carrying out the linux kernel Interrupt Process will satisfy three conditions simultaneously, and promptly real-time kernel is not in a hurry (by judging the l_busy position on the flags territory of osek_local structure), linux kernel is in and opens interruption status and the not conductively-closed of this interrupting channel.If this is the case, system's standard treatment scheme of entering Linux then.If latter two condition does not satisfy, then the CPU scene is directly recovered by system, turns back to the program when interrupting taking place, and it continues operation; If real-time kernel is busy, then need to handle the aftermath of real-time kernel Interrupt Process, if in interrupt nesting, then recover the scene of upper level interrupt handling routine, proceed Real-time Interrupt Processing, to the last one deck Interrupt Process finishes, before withdrawing from the Real-time Interrupt Processing flow process, need the real-time process storehouse of stack pointer before the real-time interrupt storehouse switches to interruption, judge whether system needs scheduling, if the Real-time Interrupt Processing routine call some system services cause entering ready attitude than the real-time process of current process higher priority, then be provided with and represent whether interruption needs the global variable of dispatching when withdrawing from, if this variable is non-vanishing, dispatch, interrupt preceding real-time process continuation operation otherwise turn back to.

Claims (1)

1. based on the vehicular operating system interrupt management method of PowerPC processor structure, it is characterized in that: it is the real time operating system that adopts dual core mechanism, described dual core mechanism is the method for standard non real-time Linux being carried out hard real-timeization, promptly by between standard non real-time linux kernel and bottom hardware, inserting a little real-time kernel, the management of real-time task is provided by real-time kernel, described linux kernel has only the operation of just obtaining as the minimum real-time task of the priority of real-time kernel when not having real-time task to be in ready attitude; At the situation of interrupting occurring in the real-time task operational process, the Interrupt Process flow process is once preserved the strategy that once recovers to real-time task operation field by using; Correspondingly this vehicular operating system interrupt management method contains following steps successively:
Step 1. is set:
Described real time operating system logically is divided into following four ranks:
The non-real-time process rank: the non real-time program that run user is write, the operation process adopts the dynamic priority strategy, and by the management and running of Linux scheduler program, its processing rank is minimum;
The real-time process rank: real-time application that the user writes and linux kernel run on the real-time process rank, and described linux kernel is minimum as its priority of special processing by the real-time kernel management, is " 0 "; It supports dynamic object to create, and Dram distributes; It carries out the simulation interrupt management;
The real-time kernel rank: real-time kernel is driven by non-real-time process and linux kernel, runs on the real-time kernel rank;
The real-time interrupt rank: its operation real-time interrupt service routine, and drive break simulation management Linux interrupt handling routine;
Described operating system is provided with following three kinds of stack type:
User stack, it is prepared for the program run of user's attitude;
Be the real-time process storehouse that real-time process is distributed, Linux real-time process storehouse wherein is corresponding to the core stack of current process in the linux kernel;
Real-time interrupt storehouse: operation Real-time Interrupt Processing program;
Step 2. makes up on the PowerPC processor with lower module:
Process creation and storehouse distribution module, it comprises:
The establishment of un-real time job and storehouse distribution sub module: it is being each each space below un-real time job distribution on the described linux kernel:
Process control block (PCB), expression state of a process and the resource that has;
Kernel stack, it is the memory headroom for operational system service and break in service distribution;
The task run space is provided with user stack and is used for the operation task code;
Real-time task is created and the storehouse distribution sub module: it runs on the real-time kernel, and described system distributes a real-time task storehouse for each real-time task and is fixed in the memory headroom of reserving when system kernel generates;
System's dual core operational module comprises the activate a task that calls for system start-up code, user task code, interrupt handling routine and choose a user task from the ready task tabulation jumping to a kind of like this scheduler operation that this task is carried out again; System's dual core operation code is supported preemption scheduling, promptly when the user task of user task code or the actuator-activated high priority of interrupt handling routine calling system real-time kernel, the activate a task operation is just called scheduler and is selected new activating of task, pressure interrupts the lower user task of priority that is moving, and the task of forwarding activation to goes to carry out;
The interrupt handling routine operational module, it contains:
The interrupt handling routine inlet: wherein, the external interrupt handling procedure is placed on the 0x500 place, entry address of MPC555 microcontroller, and the counter interrupt handling routine is placed on 0x900 place, entry address;
The real-time interrupt service routine: the priority of setting the Real-time Interrupt Processing program is higher than real-time task priority;
The non real-time interrupt handling routine: the task in the setting non real-time kernel is as the minimum real-time task of the priority of real-time kernel, in the non real-time kernel, non real-time interrupt handling routine and kernel system service run on kernel state, and its priority is higher than the task of user's attitude;
Interrupt handling routine finishes: by the new task that user interrupt service routine has activated, carry out by scheduler after can only waiting the interrupt handling routine to finish again;
The described interrupt handling routine of step 3. contains following each step successively:
Step 3.1 is preserved on-the-spot:
In generation, have no progeny,, just stack pointer is switched to interrupt stack from user stack, enter kernel state, preserve on-the-spot again as long as described system is in real-time or non real-time user's attitude; Otherwise system has been in kernel state, directly keeps on-the-spot;
Step 3.2 is judged the state when taking place to interrupt:
If user's attitude, the operation of carry out preserving process address successively, preserving the process stack pointer and obtaining interrupting channel number;
If kernel state: whether kernel is busy when at first judge to interrupt taking place, if kernel is not in a hurry, then preserves process address, preserves the process stack pointer again and obtains interrupting channel number; If kernel is busy, judge earlier then whether real-time kernel interrupts nested, as if nested, then preserves the process stack pointer, obtain interrupting channel number again; If not nested, then earlier switch to the real-time interrupt storehouse from the real-time process storehouse, remove the variable that real-time kernel need be dispatched, then go to obtain interrupting channel number again;
Step 3.3 judges whether to be real-time interrupt: if real-time interrupt is then carried out the Real-time Interrupt Processing program, the break simulation device is set; Otherwise, the break simulation device just directly is set;
Step 3.4 judges whether described system meets the following conditions simultaneously: real-time kernel is not in a hurry, the break simulation device be in open interruption status, the interrupt response passage is unimpeded, if satisfy above three conditions simultaneously, then directly enter the break simulation device and call interrupt service routine, make End of Interrupt then and handle; If real-time kernel is not in a hurry, and other conditions do not satisfy, and then directly make End of Interrupt and handle; If above three conditions do not satisfy, judge then whether the real-time kernel storehouse is nested;
Step 3.5 is then directly made End of Interrupt and is handled if the real-time kernel storehouse is nested; If it is not nested that real-time kernel interrupts, just switch to the real-time process storehouse from the real-time interrupt storehouse, need then to judge whether Real-Time Scheduling, if need Real-Time Scheduling, just carry out the Real-Time Scheduling operation, otherwise, make End of Interrupt and handle.
CNB2005100113797A 2005-03-04 2005-03-04 PowerPC processor based vehicular operating system interrupt management method Expired - Fee Related CN100347676C (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CNB2005100113797A CN100347676C (en) 2005-03-04 2005-03-04 PowerPC processor based vehicular operating system interrupt management method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CNB2005100113797A CN100347676C (en) 2005-03-04 2005-03-04 PowerPC processor based vehicular operating system interrupt management method

Publications (2)

Publication Number Publication Date
CN1655124A CN1655124A (en) 2005-08-17
CN100347676C true CN100347676C (en) 2007-11-07

Family

ID=34894182

Family Applications (1)

Application Number Title Priority Date Filing Date
CNB2005100113797A Expired - Fee Related CN100347676C (en) 2005-03-04 2005-03-04 PowerPC processor based vehicular operating system interrupt management method

Country Status (1)

Country Link
CN (1) CN100347676C (en)

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR2958341B1 (en) * 2010-03-31 2018-07-13 Valeo Systemes De Controle Moteur METHOD FOR MANAGING A HYBRID COMPRESSOR OF AIR CONDITIONING CIRCUIT
CN102012842A (en) * 2010-11-09 2011-04-13 北京神舟航天软件技术有限公司 Independent interrupt stack-based method for preemptive scheduling during interrupt exiting
CN102043668B (en) * 2010-12-10 2013-05-22 成电汽车电子产业园(昆山)有限公司 Method for activating task in embedded real-time operating system for multiple times
CN103676727A (en) * 2013-11-27 2014-03-26 南京国电南自美卓控制系统有限公司 Preemptible and nestable interrupt control method based on embedded chip
CN104476409B (en) * 2014-11-25 2019-10-18 苏州谷夫道自动化科技有限公司 Glass grinding process equipment control method based on real-time Linux platform
CN106874072B (en) * 2015-12-14 2020-12-29 中国航空工业第六一八研究所 Embedded operating system partition virtual interrupt processing method based on PowerPC processor
CN107451045A (en) * 2016-05-31 2017-12-08 北京信威通信技术股份有限公司 A kind of method and device of abnormal information positioning
CN109947539A (en) * 2017-12-20 2019-06-28 广州中国科学院先进技术研究所 A kind of robot controller
JP6656485B1 (en) * 2018-07-24 2020-03-04 三菱電機株式会社 Interrupt processing method, computer system and program
CN109857530B (en) * 2019-01-21 2021-09-03 湖北亿咖通科技有限公司 Application request scheduling method and device
CN115080122A (en) * 2022-07-22 2022-09-20 飞腾信息技术有限公司 Processor, device, method and chip for saving and restoring context data

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1490722A (en) * 2003-09-19 2004-04-21 清华大学 Graded task switching method based on PowerPC processor structure

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1490722A (en) * 2003-09-19 2004-04-21 清华大学 Graded task switching method based on PowerPC processor structure

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
一种基于Linux的车用实时操作系统 蔡俊良,孙晓民.清华大学学报(自然科学版),第44卷第1期 2004 *
清华OSEK车用嵌入式实时操作系统设计 张宝民,孙晓民.计算机工程与设计,第25卷第5期 2004 *

Also Published As

Publication number Publication date
CN1655124A (en) 2005-08-17

Similar Documents

Publication Publication Date Title
CN100347676C (en) PowerPC processor based vehicular operating system interrupt management method
CN1208721C (en) Graded task switching method based on PowerPC processor structure
CN100342342C (en) Java virtual machine implementation method supporting multi-process
CN101051282A (en) Method for realizing multiple operation system synergistic working
CN101751289B (en) Mixed scheduling method of embedded real-time operating system
KR101541344B1 (en) Memory device and controll method of memory device
WO2017166777A1 (en) Task scheduling method and device
CN101051280A (en) Intelligent card embedded operation system and its control method
CN101075196A (en) Portable electronic device and method for loading resource data for the portable electronic device
KR970067231A (en) How to Control Queuing in a CD-ROM Drive
CN1940858A (en) Multi-task realization of media player
CN1783016A (en) Computer system
CN101290590A (en) Embedded operating system task switching method and unit
CN100351793C (en) Task scheduling method for embedded real-time operation system supporting OSEK standard
CN1687901A (en) Real time task dispatching method based on interrupt
CN1740973A (en) A real-time task management and scheduling method
CN103309734A (en) Embedded task scheduling method based on priority grouping
CN1238787C (en) Binary chop type task dispatching method for embedding real-time operating system
CN1851652A (en) Method for realizing process priority-level round robin scheduling for embedded SRAM operating system
CN1519735A (en) Process scheduling method in embedded type real time operating system
CN1829971A (en) Microcontroller with an interrupt structure having programmable priority levels with each priority level associated with a different register set
CN1851651A (en) Method for realizing process priority scheduling for embedded SRAM operating system
CN1825288A (en) Method for implementing process multi-queue dispatching of embedded SRAM operating system
CN1652083A (en) Method of program delay executing and its device
CN1384431A (en) Context saving

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
C17 Cessation of patent right
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20071107