CN103870326A - Method and application for moving bottom half portion of interrupt handler to application layer - Google Patents
Method and application for moving bottom half portion of interrupt handler to application layer Download PDFInfo
- Publication number
- CN103870326A CN103870326A CN201210532215.9A CN201210532215A CN103870326A CN 103870326 A CN103870326 A CN 103870326A CN 201210532215 A CN201210532215 A CN 201210532215A CN 103870326 A CN103870326 A CN 103870326A
- Authority
- CN
- China
- Prior art keywords
- application layer
- interrupt
- half portion
- function
- handling routine
- 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.)
- Granted
Links
Images
Abstract
The invention relates to a method for moving the bottom half portion of an interrupt handler to an application layer and a method for processing the bottom half portion of the interrupt handler in the application layer based on the method. According to the method and the application for moving the bottom half portion of the interrupt handler to the application layer, the modification of the interrupt handler is simple, an application only needs to be modified in the application layer, the modification of a kernel layer is not involved, and the requirements for the research and development personnel are low; the modification and debugging work of the interrupt handler is flexible, the cost is low, only application developers need to participate in, and driving developers and the application developers do not need to work together; an interrupt event callback function which is set through the application is called after interruption, interrupt processing is performed through the callback function, the callback function is identical to an interrupt processing function of the kernel layer in form, and the interrupt processing is the same as that the interrupt handler is directly developed in the application layer from the perspective of the application.
Description
Technical field
The present invention relates to the interrupt processing mechanism under linux, more particularly, relate to a kind of method of half portion at the bottom of interrupt handling routine being moved to application layer, and a kind of method that half portion processes at the bottom of application layer is to interrupt handling routine based on described method.
Background technology
Interrupt handling routine is decomposed into two half-unit by Linux: top half (top half) and the end half portion (bottom half).Top half completes the least possible more urgent function, and it just carries out the work of " registration is interrupted " after often just reading simply the interruption status in register and remove interrupt identification.Like this, the speed that top half is carried out will be very fast, can serve more interrupt request.Next more time-consuming task just can be given the end half portion and completed, and can say, the end half portion of interrupting processing completes most tasks of interrupt event, has almost done all things of interrupt handling routine.The realization mechanism that interrupts the processing end half portion under Linux mainly contains tasklet, work queue and soft interruption.
No matter what mechanism the end half portion adopt realize, he realizes in driver, that is to say to realize under kernel mode.If need to interrupt handling routine be modified, for example because the modification of the method for the difference of vehicle environmental to mileage pulse statistics, because driver need to be revised and debug to the modifications of the difference of vehicle environmental to sensor signal jitter removing method etc. under kernel mode.The problem of bringing is like this, the one, revise difficulty large, need to revise at inner nuclear layer, to research staff require high; The 2nd, underaction, expense are large, because the middle layer of driver in operating system and application program, each modification all needs to relate to Driver Development personnel and application developer's cooperation uniting and adjustment, and the human resources of cost are many.
Chinese invention patent application 200810117341.1 provides a kind of method and embedded system that realizes interrupt bottom semi-section mechanism, and wherein, method comprises: embedded OS is received after interrupt request, carries out interrupt operation, forwards to and interrupts processing top half module; Described interruption is processed top half module and is carried out after interrupt handling routine, forwards the end half interrupt management task module to; Described management role module is carried out the end half management role, and what management role of the described end half was embedded OS has the system task higher than the priority of normal work to do, and management role of the described end half is called and interrupted processing the end half module; Described interruption is processed the end half module and is carried out the interruption processing end half program.The present invention completes the linking of interrupting top half and the end half portion by the system task with super objective priority of embedded OS, simple, response fast, size of code is little, is more suitable in Embedded Application field.
The technical scheme of foregoing invention patented claim has been mainly used in solving the problem of interrupting top half and the end half linking weak effect, but still use existing interrupt processing mechanism, same Shortcomings: one, revise difficulty large, need to revise at inner nuclear layer, to research staff require high; Two, underaction, expense are large, because the middle layer of driver in operating system and application program, each modification all needs to relate to Driver Development personnel and application developer's cooperation uniting and adjustment, and the human resources of cost are many.
Summary of the invention
The object of the invention is to overcome the deficiencies in the prior art, provide a kind of modification to interrupt handling routine to become simple, modification debugging work to interrupt handling routine becomes flexible, under the little a kind of Linux of expense, half portion at the bottom of interrupt handling routine is moved to the method for application layer, and use it in application layer process and interrupt in the application of the end half portion.
Technical scheme of the present invention is as follows:
A method of under Linux, half portion at the bottom of interrupt handling routine being moved to application layer, step is as follows:
1) inner nuclear layer driver completes and interrupts, after top half processing, interrupt event being carried out to broadcast announcement;
2) application layer dynamic base program receives the interrupt event broadcast announcement that inner nuclear layer driver sends;
3) application layer dynamic base program creation overall situation chained list, call back function registration interface, overall chained list is for storing to the call back function of call back function registration interface registration;
4) end half processing function is interrupted in the definition of application layer applications program, and registers to call back function registration interface, is stored in overall chained list, forms the interruption end half portion and processes call back function;
5) at the bottom of application layer dynamic base program is taken out the interruption in overall chained list, half portion processes call back function, completes and interrupts the end half processing.
As preferably, inner nuclear layer driver is positioned at the inner nuclear layer of system, and application layer dynamic base program, application layer applications program are positioned at the application layer of system.
As preferably, step 1) is specially:
1.1) inner nuclear layer driver interrupts top half processing function to system kernel registration, for receiving hardware interrupt and carrying out top half processing;
1.2) inner nuclear layer driver creates a netlink socket, for to system broadcasts interrupt event.
As preferably, step 2) be specially: application layer dynamic base program creation the 2nd netlink socket, broadcast announcement receiving thread, the interrupt event broadcast announcement sending for receiving inner nuclear layer driver.
As preferably, what in overall chained list, store is head pointer, and in chained list, each node storage is the call address of a function.
As preferably, the interface that call back function registration interface provides to application layer for system dynamic base, will join in overall chained list to the function of its registration by call back function registration interface.
As preferably, step 1.1) be specially: self-defining function is passed to system kernel as the parameter of request_irq function, and self-defining function can interrupt top half processes function, and for receiving hardware interrupt and interrupting top half processing.
As preferably, step 1.2) be specially: utilize netlink_kernel_create function, send broadcast announcement by a netlink socket object to system kernel, application layer.
As preferably, the realization of broadcast announcement receiving thread is specially: create self-defined thread, and circulation is read the 2nd netlink socket object and realized the reception to interrupting event broadcast notice in this self-defined thread.
The method that half portion processes at the bottom of application layer is to interrupt handling routine, step is as follows:
A) interrupt producing;
B) half portion at the bottom of interrupt handling routine is moved to application layer;
C) process interrupting the end half portion in application layer;
Step B) utilize the method that half portion at the bottom of interrupt handling routine is moved to application layer to operate.
Beneficial effect of the present invention is as follows:
(1) modification of interrupt handling routine is become simply, only need, in application-level modifications application program, not relate to the modification of inner nuclear layer, lower to research staff's requirement;
(2) the modification debugging work of interrupt handling routine is become flexibly, expense is little.Only need application developer to participate in, not demand motive developer and application developer's collaborative work;
(3) after interrupting producing, the interrupt event call back function that application program arranges is called, and this call back function interrupts processing, in full accord with the interruption processing function of inner nuclear layer in form, from the angle of application program, just as directly develop interrupt handling routine in application layer.
Accompanying drawing explanation
Fig. 1 is the system framework schematic diagram of method application of the present invention.
Embodiment
Below in conjunction with drawings and Examples, the present invention is described in further detail.
A method of under Linux, half portion at the bottom of interrupt handling routine being moved to application layer, step is as follows:
1) inner nuclear layer driver completes and interrupts, after top half processing, interrupt event being carried out to broadcast announcement; Step 1) is specially:
1.1) inner nuclear layer driver interrupts top half processing function to system kernel registration, for receiving hardware interrupt and carrying out top half processing; Be specially: self-defining function is passed to system kernel as the parameter of request_irq function, and self-defining function can interrupt top half processes function, and for receiving hardware interrupt and interrupting top half processing;
1.2) inner nuclear layer driver creates a netlink socket, for to system broadcasts interrupt event; Be specially: utilize netlink_kernel_create function, send broadcast announcement by a netlink socket object to system kernel, application layer;
2) application layer dynamic base program receives the interrupt event broadcast announcement that inner nuclear layer driver sends; Be specially: application layer dynamic base program creation the 2nd netlink socket, broadcast announcement receiving thread, the interrupt event broadcast announcement sending for receiving inner nuclear layer driver; Wherein, the realization of broadcast announcement receiving thread is specially: create self-defined thread, and circulation is read the 2nd netlink socket object realization to interrupting the reception of event broadcast notice in this self-defined thread;
3) application layer dynamic base program creation overall situation chained list, call back function registration interface, overall chained list is for storing to the call back function of call back function registration interface registration; What in overall situation chained list, store is head pointer, and in chained list, each node storage is the call address of a function; The interface that call back function registration interface provides to application layer for system dynamic base, will join in overall chained list to the function of its registration by call back function registration interface;
4) end half processing function is interrupted in the definition of application layer applications program, and registers to call back function registration interface, is stored in overall chained list, forms the interruption end half portion and processes call back function;
5) at the bottom of application layer dynamic base program is taken out the interruption in overall chained list, half portion processes call back function, completes and interrupts the end half processing.
As shown in Figure 1, inner nuclear layer driver is positioned at the inner nuclear layer of system, and application layer dynamic base program, application layer applications program are positioned at the application layer of system, and interrupt event is from the bottom hardware of hardware layer.
Inner nuclear layer driver moves at inner nuclear layer, the broadcast announcement of being mainly responsible for interrupting the processing of top half and having sent an interrupt event to whole system.In the time that a hardware interrupts arrives, it is called that the interruption top half of inner nuclear layer driver is processed function, the processing of this part and common driver are as broad as long, just interrupting the last of top half processing function, have more a processing of interrupt event being carried out to broadcast announcement, this broadcast announcement adopts netlink mechanism to realize, and can receive this broadcast announcement in the program of application layer and the program of inner nuclear layer.
Application layer dynamic base program is moved in application layer, and at the bottom of the interrupt event broadcast announcement that main responsible reception inner nuclear layer driver sends and the interruption that application layer application program is registered, half processing function is adjusted back.
Application layer applications program is moved in application layer, the interface function that mainly utilizes application layer dynamic base program to provide, the business processing that the end half portion processes call back function and interrupts the end half function is interrupted in registration, for example carry out mileage statistics according to the statistical method of mileage pulse, according to the feature of sensor signal, signal is carried out the validity judgement of debounce processing and signal etc.
The method that half portion processes at the bottom of application layer is to interrupt handling routine, step is as follows:
A) interrupt producing;
B) half portion at the bottom of interrupt handling routine is moved to application layer;
C) process interrupting the end half portion in application layer;
Step B) utilize the method that half portion at the bottom of interrupt handling routine is moved to application layer to operate.
Embodiment
Under Linux, half portion at the bottom of interrupt handling routine is moved to the method for application layer, concrete implementation step is as follows:
1) inner nuclear layer driver interrupts top half processing function to one of kernel registration, to receive hardware interrupt and carry out top half processing; Interrupting top half processing function to kernel registration is the interface that linux kernel provides, realize with request_irq function, a common function is passed to kernel as the parameter of request_irq function, this generic function has just become interruption top half and has processed function, can receive hardware interrupt and interrupt top half processing;
2) inner nuclear layer driver creates a netlink socket, to system broadcasts interrupt event; Creating netlink socket is the interface that linux kernel provides, and realizes with netlink_kernel_create function, can send broadcast announcement to kernel spacing and user's space by the netlink socket object creating;
3) netlink socket of application layer dynamic base program creation and a broadcast announcement receiving thread, to receive the interrupt event broadcast announcement that inner nuclear layer driver sends; Broadcast announcement receiving thread be by create a common thread and in this thread circulation read netlink socket object and realize.If there is broadcast announcement, the function that reads netlink socket object just can return to and obtain the particular content of broadcast announcement;
4) one of application layer dynamic base program creation interrupts the overall chained list of the end half processing call back function, registers the interruption end half processing call back function to store all call back function registration interface of utilizing application layer dynamic base program; This overall chained list defines by list_head data structure, and it is the head pointer of a chained list in essence, and in chained list, each node storage is the call address of a function, and we are referred to as to interrupt the end half processing call back function these functions.Originally, interrupting the end half processing function is the delay disposal part for a complete interruption processing procedure, in linux kernel code, could realize, here moved application layer by the delay disposal part of a complete interruption processing procedure, therefore also claims that the function in this overall chained list node is processed function for interrupting the end half portion.
List_head data structure is defined as follows:
struct?list_hed{
struct?list_head*next,*prev;
};
5) call back function registration interface of application layer dynamic base program creation, so that other application layer applications programs are registered and are interrupted the end half processing call back function to him; The effect of this call back function registration interface is exactly half portion at the bottom of registered interruption to be processed to function join in overall chained list mentioned above, and he is an interface that dynamic base provides to application layer, and the concrete function definition below that adopts realizes:
int?set_interrupt_event_callback_func(void(*event_callback)(int?event));
6) end half processing function is interrupted in one of application layer applications program definition, complete the delay disposal to interrupt event, for example carry out mileage statistics according to the statistical method of mileage pulse, according to the feature of sensor signal, signal is carried out the validity judgement of debounce processing and signal etc.
7) application layer applications program is utilized one of the call back function registration interface registration of application layer dynamic base program to interrupt the end half portion and is processed function.
The method that half portion processes at the bottom of application layer is to interrupt handling routine, concrete implementation step is as follows:
A) hardware interrupts produces;
B) the interruption top half of inner nuclear layer driver is processed function operation, and sends interrupt event broadcast announcement;
C) the broadcast announcement receiving thread of application layer dynamic base program is received the interrupt event broadcast announcement that inner nuclear layer driver sends;
D) application layer dynamic base program is taken out one by one call back function from the overall chained list of the interruption end half processing call back function, and to call one by one;
E) at the bottom of the interruption in application layer applications program, half portion processes call back function and is employed a layer dynamic base routine call, carries out according to the content of this function, completes the processing of interrupting the end half portion.
Above-described embodiment is only for the present invention is described, and not as limitation of the invention.As long as according to technical spirit of the present invention, to above-described embodiment change, modification etc. all will drop in the scope of claim of the present invention.
Claims (10)
1. a method of under Linux, half portion at the bottom of interrupt handling routine being moved to application layer, is characterized in that, step is as follows:
1) inner nuclear layer driver completes and interrupts, after top half processing, interrupt event being carried out to broadcast announcement;
2) application layer dynamic base program receives the interrupt event broadcast announcement that inner nuclear layer driver sends;
3) application layer dynamic base program creation overall situation chained list, call back function registration interface, overall chained list is for storing to the call back function of call back function registration interface registration;
4) end half processing function is interrupted in the definition of application layer applications program, and registers to call back function registration interface, is stored in overall chained list, forms the interruption end half portion and processes call back function;
5) at the bottom of application layer dynamic base program is taken out the interruption in overall chained list, half portion processes call back function, completes and interrupts the end half processing.
2. the method for under Linux according to claim 1, half portion at the bottom of interrupt handling routine being moved to application layer, is characterized in that, inner nuclear layer driver is positioned at the inner nuclear layer of system, and application layer dynamic base program, application layer applications program are positioned at the application layer of system.
3. the method for under Linux according to claim 1, half portion at the bottom of interrupt handling routine being moved to application layer, is characterized in that, step 1) is specially:
1.1) inner nuclear layer driver interrupts top half processing function to system kernel registration, for receiving hardware interrupt and carrying out top half processing;
1.2) inner nuclear layer driver creates a netlink socket, for to system broadcasts interrupt event.
4. under Linux according to claim 1, half portion at the bottom of interrupt handling routine is moved to the method for application layer, it is characterized in that, step 2) be specially: application layer dynamic base program creation the 2nd netlink socket, broadcast announcement receiving thread, the interrupt event broadcast announcement sending for receiving inner nuclear layer driver.
5. the method for under Linux according to claim 1, half portion at the bottom of interrupt handling routine being moved to application layer, is characterized in that, what in overall chained list, store is head pointer, and in chained list, each node storage is the call address of a function.
6. under Linux according to claim 1, half portion at the bottom of interrupt handling routine is moved to the method for application layer, it is characterized in that, the interface that call back function registration interface provides to application layer for system dynamic base, will join in overall chained list to the function of its registration by call back function registration interface.
7. under Linux according to claim 3, half portion at the bottom of interrupt handling routine is moved to the method for application layer, it is characterized in that, step 1.1) be specially: self-defining function is passed to system kernel as the parameter of request_irq function, self-defining function can interrupt top half processes function, and for receiving hardware interrupt and interrupting top half processing.
8. under Linux according to claim 3, half portion at the bottom of interrupt handling routine is moved to the method for application layer, it is characterized in that, step 1.2) be specially: utilize netlink_kernel_create function, send broadcast announcement by a netlink socket object to system kernel, application layer.
9. under Linux according to claim 4, half portion at the bottom of interrupt handling routine is moved to the method for application layer, it is characterized in that, the realization of broadcast announcement receiving thread is specially: create self-defined thread, and circulation is read the 2nd netlink socket object realization to interrupting the reception of event broadcast notice in this self-defined thread.
10. the method that half portion processes at the bottom of application layer is to interrupt handling routine, is characterized in that, step is as follows:
A) interrupt producing;
B) half portion at the bottom of interrupt handling routine is moved to application layer;
C) process interrupting the end half portion in application layer;
Step B) utilize the method for half portion at the bottom of interrupt handling routine being moved to application layer described in claim 1 to 9 any one to operate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210532215.9A CN103870326B (en) | 2012-12-11 | 2012-12-11 | A kind of method and application that interrupt handling routine bottom half is moved to application layer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210532215.9A CN103870326B (en) | 2012-12-11 | 2012-12-11 | A kind of method and application that interrupt handling routine bottom half is moved to application layer |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103870326A true CN103870326A (en) | 2014-06-18 |
| CN103870326B CN103870326B (en) | 2018-07-13 |
Family
ID=50908891
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210532215.9A Active CN103870326B (en) | 2012-12-11 | 2012-12-11 | A kind of method and application that interrupt handling routine bottom half is moved to application layer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103870326B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105579963A (en) * | 2014-09-03 | 2016-05-11 | 华为技术有限公司 | Task handling apparatus and method, and electronic device |
| CN110308934A (en) * | 2018-03-22 | 2019-10-08 | 龙芯中科技术有限公司 | HPET driving method and device, readable storage medium storing program for executing and electronic equipment |
| CN111124506A (en) * | 2019-11-12 | 2020-05-08 | 中国电子科技集团公司第三十研究所 | Operation card driving implementation method based on application layer |
| CN111338777A (en) * | 2020-02-26 | 2020-06-26 | 山东超越数控电子股份有限公司 | Low-delay high-stability autonomous platform interrupt response method and equipment |
| CN111400078A (en) * | 2020-03-09 | 2020-07-10 | 广州路派电子科技有限公司 | Method and device for processing task in embedded system |
| CN111984334A (en) * | 2020-08-17 | 2020-11-24 | 上海睿赛德电子科技有限公司 | Lightweight operating system kernel and drive separation method |
| CN112470125A (en) * | 2018-07-24 | 2021-03-09 | 三菱电机株式会社 | Interrupt processing method, computer system and program product |
| CN114416317A (en) * | 2022-01-20 | 2022-04-29 | 科东(广州)软件科技有限公司 | Inter-core interrupt execution method, inter-core interrupt processing device, inter-core interrupt equipment and storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3794119B2 (en) * | 1997-08-29 | 2006-07-05 | ソニー株式会社 | Data processing method, recording medium, and data processing apparatus |
| CN101349975B (en) * | 2008-07-29 | 2014-04-23 | 北京中星微电子有限公司 | Method for implementing interrupt bottom semi-section mechanism in embedded operation system |
-
2012
- 2012-12-11 CN CN201210532215.9A patent/CN103870326B/en active Active
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105579963B (en) * | 2014-09-03 | 2019-10-01 | 华为技术有限公司 | Task Processing Unit, electronic equipment and method |
| CN105579963A (en) * | 2014-09-03 | 2016-05-11 | 华为技术有限公司 | Task handling apparatus and method, and electronic device |
| CN110308934A (en) * | 2018-03-22 | 2019-10-08 | 龙芯中科技术有限公司 | HPET driving method and device, readable storage medium storing program for executing and electronic equipment |
| CN110308934B (en) * | 2018-03-22 | 2022-03-29 | 龙芯中科(成都)技术有限公司 | HPET driving method and device, readable storage medium and electronic equipment |
| CN112470125A (en) * | 2018-07-24 | 2021-03-09 | 三菱电机株式会社 | Interrupt processing method, computer system and program product |
| CN112470125B (en) * | 2018-07-24 | 2024-02-20 | 三菱电机株式会社 | Interrupt processing method, computer system, and storage medium |
| CN111124506A (en) * | 2019-11-12 | 2020-05-08 | 中国电子科技集团公司第三十研究所 | Operation card driving implementation method based on application layer |
| CN111338777B (en) * | 2020-02-26 | 2023-07-07 | 超越科技股份有限公司 | Low-delay high-stability autonomous platform interrupt response method and equipment |
| CN111338777A (en) * | 2020-02-26 | 2020-06-26 | 山东超越数控电子股份有限公司 | Low-delay high-stability autonomous platform interrupt response method and equipment |
| CN111400078A (en) * | 2020-03-09 | 2020-07-10 | 广州路派电子科技有限公司 | Method and device for processing task in embedded system |
| CN111400078B (en) * | 2020-03-09 | 2023-08-25 | 广州路派电子科技有限公司 | Method and device for processing tasks in embedded system |
| CN111984334A (en) * | 2020-08-17 | 2020-11-24 | 上海睿赛德电子科技有限公司 | Lightweight operating system kernel and drive separation method |
| CN114416317B (en) * | 2022-01-20 | 2022-12-02 | 科东(广州)软件科技有限公司 | Inter-core interrupt execution method, inter-core interrupt processing device, inter-core interrupt equipment and storage medium |
| CN114416317A (en) * | 2022-01-20 | 2022-04-29 | 科东(广州)软件科技有限公司 | Inter-core interrupt execution method, inter-core interrupt processing device, inter-core interrupt equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103870326B (en) | 2018-07-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103870326A (en) | Method and application for moving bottom half portion of interrupt handler to application layer | |
| CN101719262B (en) | Graphics processing unit and meta-command processing system | |
| EP3889774A1 (en) | Heterogeneous computing-based task processing method and software-hardware framework system | |
| US8578216B2 (en) | Execution history tracing method | |
| US8381212B2 (en) | Dynamic allocation and partitioning of compute nodes in hierarchical job scheduling | |
| CN106663021B (en) | Intelligent GPU scheduling in virtualized environments | |
| US8832666B2 (en) | Dynamic instrumentation | |
| US20130227194A1 (en) | Active non-volatile memory post-processing | |
| TW201638791A (en) | System, method, and apparatus for synchronization among heterogeneous data sources | |
| CN1842769A (en) | Instruction for initiation of concurrent instruction streams in a multithreading microprocessor | |
| CN103218329A (en) | Digital signal processing data transfer | |
| CN107463357B (en) | Task scheduling system, scheduling method, braking simulation system and simulation method | |
| CN106780149A (en) | A kind of equipment real-time monitoring system based on timed task scheduling | |
| CN102955737A (en) | Program debugging method and system of heterogeneous processor system | |
| US9239732B2 (en) | Unrolling aggregation operations in asynchronous programming code having multiple levels in hierarchy | |
| CN110795150A (en) | Implementation of load fetch/store release instruction by load/store operation according to DMB operation | |
| CN103077140A (en) | Communication method and communication device for motion control card of robot and host | |
| CN102331961B (en) | Method, system and dispatcher for simulating multiple processors in parallel | |
| TW202109286A (en) | System and architecture of pure functional neural network accelerator | |
| Prashanth et al. | Real-time issues in embedded system design | |
| CN1851651A (en) | Method for realizing process priority scheduling for embedded SRAM operating system | |
| US9311225B2 (en) | DMA channels | |
| US10713188B2 (en) | Inter-process signaling system and method | |
| Zagan et al. | Designing a custom CPU architecture based on hardware RTOS and dynamic preemptive scheduler | |
| US11163605B1 (en) | Heterogeneous execution pipeline across different processor architectures and FPGA fabric |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |