CN102915292A - Communication method based on multi-core processor and detection method and control method thereof - Google Patents
Communication method based on multi-core processor and detection method and control method thereof Download PDFInfo
- Publication number
- CN102915292A CN102915292A CN2011102203079A CN201110220307A CN102915292A CN 102915292 A CN102915292 A CN 102915292A CN 2011102203079 A CN2011102203079 A CN 2011102203079A CN 201110220307 A CN201110220307 A CN 201110220307A CN 102915292 A CN102915292 A CN 102915292A
- Authority
- CN
- China
- Prior art keywords
- kernel
- polycaryon processor
- take over
- over party
- transmit leg
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/54—Interprogram communication
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F15/00—Digital computers in general; Data processing equipment in general
- G06F15/16—Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs
- G06F15/163—Interprocessor communication
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Hardware Redundancy (AREA)
- Computer And Data Communications (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The application discloses an internal communication method of a computer comprising a multi-core processor and a detection method and a control method thereof, wherein the multi-core processor supports affinity setting and core identification. The communication method comprises the following steps: a sender selects a receiver to run on which core to send information; and the receiver identifies which core to run on to receive information. The invention provides the novel communication method between progresses and virtual machines, and the corresponding detection method and the control method. The efficiency and the flexibility of multitask cooperative processing under a multi-core processor environment can be improved; and the communication method can be regarded as a supplementary communication method when the conventional IPC (inter-process communication) and the communication between the virtual machines are disabled.
Description
Technical field
The present invention relates to comprise communication means and detection method and the control method of the computer-internal of polycaryon processor, particularly, relate between interprocess communication (IPC) and virtual machine and communicating by letter and security mechanism.
Background technology
An important directions of technical development of computer is to use polycaryon processor.Polycaryon processor is integrated two or more complete computing engines (kernel) in one piece of processor (chip).A plurality of kernel coordination work are to reach the effect of performance multiplication.Polycaryon processor can move different course allocation on different kernels simultaneously, thereby can improve the efficient of multitasking.
When distributing multitask, the scheduling mechanism of polycaryon processor arranges respectively the processor compatibility (CPU affinity) of each process, and certain process of being about to is assigned to the tendentiousness of trying one's best for a long time operation on certain processor and not being migrated to other processors.
Each kernel in the polycaryon processor has the Advanced Programmable Interrupt Controllers APICs sign (Advanced Programmable Interrupt Controller Identification Number is abbreviated as APIC ID) of oneself.Polycaryon processor utilizes APIC ID to identify specific kernel and sends instruction to it.
By means of independently communication network realization intercommunication, for example use bus shared structure, cross bar switch interconnection structure and network-on-chip (network on chip, NOC) between the different kernels of polycaryon processor.Above-mentioned communication network provides intercommunicating infrastructure for the kernel of polycaryon processor.Can utilize inside story mechanism to realize intercommunication in the different processes that the different kernels of polycaryon processor move.
Yet, still be desirably in the new method that interprocess communication is provided in the polycaryon processor, to satisfy the requirement of (under the MAC access control) interprocess communication and safety monitoring under the specific condition, management.
Summary of the invention
The purpose of this invention is to provide a kind of communication means based on polycaryon processor and detection method and control method, the method is based on the scheduling mechanism of existing intercommunication infrastructure and existing operating system.
According to an aspect of the present invention, a kind of communication means based on polycaryon processor is provided, described polycaryon processor is supported compatibility setting and the identification of kernel sign, described method comprises: transmit leg comes transmission information by selective reception side in which kernel operation, and the take over party comes reception information by identifying oneself in which kernel operation.
According to the preferred embodiment of communication means of the present invention, wherein transmit leg is the different process of moving at polycaryon processor with the take over party.
According to the preferred embodiment of communication means of the present invention, wherein transmit leg is the different virtual machine that moves at polycaryon processor with the take over party.According to a further aspect in the invention, provide a kind of detection method of the communication based on polycaryon processor, described polycaryon processor is supported compatibility setting and the identification of kernel sign, and described method comprises: the transmit leg that compatibility arranges is carried out in monitoring; The take over party of kernel sign identification is carried out in monitoring; And if not only have transmit leg but also have the take over party within a certain period of time, then judge to exist utilize the intercommunication of selecting nuclear mechanism.
According to the preferred embodiment of detection method of the present invention, wherein transmit leg is the different process of moving at polycaryon processor with the take over party.
According to the preferred embodiment of detection method of the present invention, wherein transmit leg is the different virtual machine that moves at polycaryon processor with the take over party.According to a further aspect in the invention, a kind of control method of the communication based on polycaryon processor is provided, described polycaryon processor is supported compatibility setting and the identification of kernel sign, described method comprises: allow/forbid the compatibility setting of at least one kernel, and/or the kernel sign identification of at least one kernel of permission/forbidding.
According to the preferred embodiment of control method of the present invention, forbid compatibility setting and/or the identification of kernel sign of whole kernels, utilize the passage (channel) of selecting nuclear mechanism to close.
According to the preferred embodiment of control method of the present invention, allow compatibility setting and the identification of kernel sign of whole kernels, the bandwidth of selecting the passage of nuclear mechanism with maximum using.
The present invention is applicable to most of polycaryon processors (for example polycaryon processor of Intel Company and AMD) and operating system (for example, the operating system that compatibility can be set comprises Windows, Linux, Xen etc.).Polycaryon processor and operating system/virtual machine support jointly that at hardware layer and operating system/virtual machine layer compatibility arranges function respectively; Polycaryon processor is independently supported place nuclear recognition command at hardware layer.
The invention provides the new communication means between process and between virtual machine, and corresponding detection method and control method.The present invention has expanded between process and the mode of virtual inter-machine communication, and under the restricted situation of conventional Way of Process Communication, and for example under the environment of forcing access control, it is still effective that the communication means of nuclear mechanism is selected in utilization according to the present invention.Thereby the present invention can improve efficient and the dirigibility of the multitask coordinated processing under the polycaryon processor environment, and as a supplement communication means in can communication means lost efficacy between the interprocess communication (IPC) of routine and virtual machine.
Description of drawings
The schematically illustrated ultimate principle that the communication of nuclear mechanism is selected in utilization according to the first embodiment of the present invention of Fig. 1.
Fig. 2 has schematically shown the flow process that the communication of nuclear mechanism is selected in utilization according to the first embodiment of the present invention.
The schematically illustrated ultimate principle of selecting the communication of nuclear mechanism of utilizing according to a second embodiment of the present invention of Fig. 3.
Fig. 4 has schematically shown and has utilized according to a second embodiment of the present invention the flow process of selecting the communication of nuclear mechanism.
Embodiment
Hereinafter with reference to accompanying drawing the present invention is described in more detail.
The＜the first embodiment 〉
The schematically illustrated ultimate principle that the communication of nuclear mechanism is selected in utilization according to the first embodiment of the present invention of Fig. 1.In this example, this polycaryon processor comprises two kernels, i.e. kernel 1 and kernel 2.According to the scheduling mechanism of polycaryon processor, in original state, process 1 and process 2 run on respectively on kernel 2 and the kernel 1, as shown in Figure 1a.Process 1 can arrange the kernel compatibility of self and with self moving on the kernel 1, shown in Fig. 1 b.According to the scheduling mechanism of polycaryon processor, process 2 will be automatically migrated on the idle kernel 2, shown in Fig. 1 c.The kernel sign that process 2 identification moves self, and think the migratory behaviour of self represent data ' 1 ' or ' 0 '.
Fig. 2 has schematically shown the flow process that the communication of nuclear mechanism is selected in utilization according to the first embodiment of the present invention.This interprocess communication starts from step S01, and in original state, process 1 is as transmit leg and be arranged in kernel 2, and process 2 is as the take over party and be arranged in kernel 1, as shown in Figure 1a.
At step S02, process 1 arranges the kernel compatibility of self, reaches himself be tied to kernel 1 or kernel 2 at least certain hour (for example 500 milliseconds) and full speed running.
At step S03, the scheduling mechanism of polycaryon processor arranges generation kernel migration instruction according to the kernel compatibility of process 1, runs on which kernel with decision process 1.If process 1 will himself be tied to kernel 1, the scheduling mechanism of polycaryon processor produces the kernel migration instruction of process 1.If process 1 will himself be tied to kernel 2, the scheduling mechanism of polycaryon processor will can not produce the kernel migration instruction of process 1.
At step S04, whether the scheduling mechanism of polycaryon processor basis produces kernel migration instruction and carries out different operations.If produced kernel migration instruction, then proceed to step S05, wherein process 1 migrates to kernel 1 from kernel 2 and reaches at least certain hour (for example 500 milliseconds), thereby moves simultaneously process 1 and process 2 in this instantaneous state on kernel 1, shown in Fig. 1 b.If do not produce kernel migration instruction, then proceed to step S06, wherein process 1 maintains operation on the kernel 2.
Migrate to from kernel 2 under the situation of kernel 1 in process 1, the scheduling mechanism of polycaryon processor will find that subsequently kernel 2 is in idle condition, thereby further produce the kernel migration instruction of process 2, and the process 2 that the kernel compatibility is not set is moved to kernel 2.Process 1 and process 2 run on respectively on kernel 1 and the kernel 2, shown in Fig. 1 c.
At step S07, process 2 identifies for example APIC ID such as identify the own kernel that moves by modes such as CPUID assembly instructions, thereby judges whether self moves.The CPU after 1993 of Intel Company and AMD can carry out the CPUID assembly instruction basically.
At step S08, process 2 is carried out different operations according to self whether moving.If move (shown in Fig. 1 c), then proceed to step S09, wherein process 2 is thought its migratory behaviour ' 1 ' that represents to receive data from process 1.Otherwise, if do not move (not shown), then proceed to step S10, wherein process 2 thinks that migratory behaviour ' 0 ' that represents to receive data from process 1 does not occur for it.
If need to transmit a large amount of data, then repeat above-mentioned steps S02 to S10, until all transmit required data, then finish interprocess communication at step S11.
Preferably, between step S01 and S02, also comprise additional noise level detecting step.This noise may come other processes of moving in the comfortable polycaryon processor, and for example another process 3 (interference source) except process 1 (transmit leg) and process 2 (take over party) also may cause the kernel migration of process 2 (take over party).
This detecting step comprises: when interprocess communication begins, process 1 detects the load level of each kernel, if the load of each kernel less than predetermined threshold level m, for example 50%, then think interference noise in insignificant level, the later step of communicating by letter between can executive process.
Preferably, between step S9, S10 and step S11, also comprise additional checking procedure.This checking procedure comprises: when interprocess communication finishes, utilize check code to judge in the data that transmit whether have mistake.When judging that existence is wrong, the data that can retransfer are to correct a mistake.
According to preferred embodiment, the noise level detecting step when interprocess communication begins and/or the checking procedure when interprocess communication finishes can improve the reliability that data transmit.
As the example of the first embodiment, the execution in step of transmit leg (being above-mentioned process 1) master routine sender is as follows:
/*sender*/
The first step: create a processor consumption molded line journey threadfunc;
Second step: call the pthread_setaffinity_np function, the nucleophilicity of threadfunc thread is made as CPU1
The 3rd step: open the binary file file1 that will send
The 4th step: read the value of a bit among the file f ile1 that fixed length is L, and be assigned to variable ch
The 5th step: carry out loop body
If the value of ch were ' 1 ', the pthread_setaffinity_np function would then be called, the nucleophilicity of change thread threadfunc.If that is: the kernel of the present binding of this thread is CPU1, then change binding CPU2 into; If that this kernel binding is CPU2 at present, then change binding CPU1 into).
If the value of ch is ' 0 ', then what is not done.
Otherwise interrupt withdrawing from loop body.
Call 500 milliseconds of nanosleep sleeps, with take over party's (be above-mentioned process 2) synchronously.
The value of next bit among the file reading file1, and be assigned to variable ch,
Loop variable k is from increasing 1 (such as k++)
If k=L then withdraw from,
Otherwise repeated for the 5th step.
Wherein, the threadfunc function is as the thread of computational resource depletion type, and this thread belongs to process 1, and process 1 is controlled the nucleophilicity of self by the nucleophilicity that the thread under it is set.Pthread_setaffinity_np is that the kernel compatibility arranges function, is used for arranging the kernel compatibility of thread or process.In the program of this example, the kernel compatibility arranges function and is sending data ' 1 ' time and arrange the kernel compatibility of thread threadfunc.
The execution in step of take over party's (being above-mentioned process 2) master routine receiver is as follows:
/*receiver*/
The first step: open/create a file f ile2, be used for writing the information content that the other side sends.
Second step: the cycle calculations (such as circulation a++) of in 500 milliseconds, doing a large amount of consumption cpu resources
The 3rd step: obtain the ID of this process operation place kernel with GetAPIC ID function, and this ID value is assigned to variable p
The 4th step: carry out loop body
Call 500 milliseconds of nanosleep sleeps, with transmit leg (be above-mentioned process 1) synchronously.
Call the GetAPIC_ID function, obtain the kernel ID at this process operation place, be assigned to variable apicid.
The relatively value of apicid and the value of variable p, if equate, the value that then receives a bit is ' 0 ', otherwise value ' 1 ' of reception, and it is write the file2 file.
With the place kernel ID value that just now obtained, namely the value of variable apicid is assigned to variable p,
Loop variable k is from increasing 1 (such as k++)
If k=L (being the fixed length of file1 in the process 1) then finishes.
Otherwise the 4th step of repetitive cycling.
Wherein, GetAPIC_ID is kernel sign function reading.When interprocess communication, process 2 circulation is read the kernel sign of self and is judged whether thus to move, judge accordingly from process 1 receive data ' 1 ' still ' 0 '.
Although described interprocess communication in above-mentioned communication means, the method for communicating by letter between the modification realization virtual machine according to this embodiment is feasible.For example, under the half virtual condition of XEN, with after the different VCPU of transmit leg virtual machine and the different physical cpu kernels bindings, said method can be implemented to communicate by letter between different virtual machine.
Although in above-mentioned communication means, described interprocess communication under the dual core processor environment, realize that according to the distortion of this embodiment the communication means under the polycaryon processor environment more is obvious.
According to the communication means of the first embodiment of the present invention can communication means loses efficacy between the interprocess communication (IPC) of routine and virtual machine in as a supplement communication means.
The＜the second embodiment 〉
The schematically illustrated ultimate principle of selecting the communication of nuclear mechanism of utilizing according to a second embodiment of the present invention of Fig. 3.In this example, this polycaryon processor comprises two kernels, i.e. kernel 1 and kernel 2.According to the scheduling mechanism of polycaryon processor, in original state, process 1 and process 2 run on respectively on kernel 2 and the kernel 1, shown in Fig. 3 a.Process 1 can arrange the kernel compatibility of process 2 and process 2 is moved on the kernel 2, shown in Fig. 3 b.According to the scheduling mechanism of polycaryon processor, process 1 will be automatically migrated on the idle kernel 1, shown in Fig. 3 c.The kernel sign that process 2 identification moves self, and think the migratory behaviour of self represent data ' 1 ' or ' 0 '.
Fig. 4 has schematically shown and has utilized according to a second embodiment of the present invention the flow process of selecting the communication of nuclear mechanism, only points out the difference of the second embodiment and the first embodiment at this.
After the step S01 according to the first embodiment of the present invention begins interprocess communication, replace the step S02-S06 according to the first embodiment of the present invention, carry out following steps:
At step S02 ', process 1 directly arranges the kernel compatibility of process 2, process 2 is tied to kernel 1 or kernel 2 reaches certain hour (for example 500 milliseconds) at least.
At step S03 ', the scheduling mechanism of polycaryon processor determines whether to produce kernel migration instruction according to the kernel compatibility setting of process 2, runs on which kernel with decision process 2.If process 1 is tied to kernel 2 with process 2, the scheduling mechanism of polycaryon processor produces the kernel migration instruction of process 2.If process 1 is tied to kernel 1 with process 2, the scheduling mechanism of polycaryon processor will can not produce the kernel migration instruction of process 2.
At step S04 ', whether the scheduling mechanism of polycaryon processor basis produces kernel migration instruction and carries out different operations.If produced kernel migration instruction, then proceed to step S05 ', wherein process 2 migrates to kernel 2 from kernel 1 and reaches at least certain hour (for example 500 milliseconds).If do not produce kernel migration instruction, then proceed to step S06 ', wherein process 2 maintains operation on the kernel 1.
Then, continue the step S07 to S11 of the executive basis first embodiment of the present invention, thereby the data of finishing interprocess communication transmit.
Preferably, in according to a second embodiment of the present invention, also can comprise above-described additional noise level detecting step and/or checking procedure.
According to preferred embodiment, the noise level detecting step when interprocess communication begins and the checking procedure when interprocess communication finishes can improve the reliability that data transmit.
As the example of the second embodiment, the execution in step of transmit leg (being above-mentioned process 1) master routine sender is as follows:
/*sender*/
The first step: the cycle calculations (such as circulation a++) of in 500 milliseconds, doing a large amount of consumption cpu resources
Second step: call setaffinity np function, the nucleophilicity of receiver process is made as CPU1
The 3rd step: opening the fixed length that will send is the binary file file1 of L
The 4th step: the value of a bit among the file reading file1, and be assigned to variable ch
The 5th step: carry out loop body
If the value of ch were ' 1 ', the setaffinity_np function would then be called, the nucleophilicity of change receiver process.If that is: the kernel of the present binding of this process is CPU1, then change binding CPU2 into; If that this kernel binding is CPU2 at present, then change binding CPU1 into).
If the value of ch is ' 0 ', then what is not done.
Otherwise interrupt withdrawing from loop body.
Call 500 milliseconds of nanosleep sleeps, with take over party's (be above-mentioned process 2) synchronously.
The value of next bit among the file reading file1, and be assigned to variable ch,
Loop variable k is from increasing 1 (such as k++)
If k=L then withdraw from,
Otherwise repeated for the 5th step.
Wherein, sender is as process 1, and process 1 directly or indirectly has system's highest weight limit, and receiver as mentioned below is as process 2, and setaffinity_up is that the kernel compatibility arranges function, is used for arranging the kernel compatibility of thread or process.In the program of this example, the kernel compatibility arranges function arranges the receiver process when sending data ' 1 ' kernel compatibility.
The execution in step of take over party's's (being above-mentioned process 2) master routine receiver is as follows:
/*receiver*/
The first step: open/create a file f ile2, be used for writing the information content that the other side sends.
Second step: the cycle calculations (such as circulation a++) of in 500 milliseconds, doing a large amount of consumption cpu resources
The 3rd step: obtain the ID of this process operation place kernel with the GetAPIC_ID function, and this ID value is assigned to variable p
The 4th step: carry out loop body
Call 500 milliseconds of nanosleep sleeps, with transmit leg (be above-mentioned process 1) synchronously.
Call GetAPIC ID function, obtain the kernel ID at this process operation place, be assigned to variable apicid.
The relatively value of apicid and the value of variable p, if equate, the value that then receives a bit is ' 0 ', otherwise value ' 1 ' of reception, and it is write the file2 file.
With the place kernel ID value that just now obtained, namely the value of variable apicid is assigned to variable p,
Loop variable k is from increasing 1 (such as k++)
If k=L (being the fixed length of file1 in the process 1) then finishes.
Otherwise the 4th step of repetitive cycling.
Wherein, GetAPIC_ID is kernel sign function reading.When interprocess communication, process 2 liaisons are taken from the kernel sign of body and are judged whether thus to move, and judge accordingly from process 1 still ' 0 ', ' 1 ' that receives data.
Although describing interprocess communication in above-mentioned communication means, the method for communicating by letter between the modification realization virtual machine according to this embodiment is feasible.For example, between the Domain0 of XEN and DomainU, because Domain0 can move the supervisory routine " xm vcpu-pin " that has system's highest weight limit, this program can realize the nucleophilicity setting of virtual machine, therefore said method can be implemented communication by xm vcpu-pin order between Domain0 and a DomainU.
Although in above-mentioned communication means, described interprocess communication under the dual core processor environment, realize that according to the distortion of this embodiment the communication means under the polycaryon processor environment more is obvious.
Communication means according to a second embodiment of the present invention can communication means loses efficacy between the interprocess communication (IPC) of routine and virtual machine in as a supplement communication means.
The＜the three embodiment 〉
A third embodiment in accordance with the invention relates to a kind of detection method of the communication based on polycaryon processor, and described polycaryon processor is supported compatibility setting and the identification of kernel sign.
This detection method comprises: the transmit leg that compatibility arranges work is carried out in monitoring; The take over party of kernel identification is carried out in monitoring; And if in certain hour t (for example t=3 second), not only had the compatibility setting operation of transmit leg but also had take over party's kernel sign identifying operation, then judge to exist and utilize the intercommunication of selecting nuclear mechanism.
The＜the four embodiment 〉
A fourth embodiment in accordance with the invention relates to a kind of control method of the communication based on polycaryon processor, described polycaryon processor is supported compatibility setting and the identification of kernel sign, described method comprises: allow/forbid the compatibility setting of at least one kernel, and/or the kernel sign identification of at least one kernel of permission/forbidding.
Preferably, forbid compatibility setting and/or the identification of kernel sign of whole kernels, utilize the passage of selecting nuclear mechanism to close.
Preferably, allow compatibility setting and the identification of kernel sign of whole kernels, thereby between process, share whole kernels, the bandwidth of selecting the passage of nuclear mechanism with maximum using.
Above description just illustrates for example and describes the present invention, but not is intended to exhaustive and restriction the present invention.Therefore, the present invention is not limited to described embodiment.For obvious modification or the change as can be known of those skilled in the art, all within protection scope of the present invention.
Claims (14)
1. communication means based on polycaryon processor, described polycaryon processor are supported compatibility setting and the identification of kernel sign, and described method comprises:
Transmit leg comes transmission information by selective reception side in which kernel operation; And
The take over party comes reception information by identifying oneself in which kernel operation.
2. method according to claim 1, the step that wherein sends information comprises: transmit leg arranges the kernel compatibility of self and moves instruction to determine the kernel that whether produces transmit leg, and the scheduling mechanism of polycaryon processor determines whether to produce take over party's kernel migration instruction.
3. method according to claim 1, the step that wherein sends information comprises: transmit leg arranges take over party's kernel compatibility to determine whether to produce take over party's kernel migration instruction.
4. each described method according to claim 1-3, the step that wherein receives information comprises: the take over party adopts the CPUID assembly instruction to read the APIC ID of the kernel that self moves.
5. each described method according to claim 1-3, wherein transmit leg is the different process of moving at polycaryon processor with the take over party.
6. each described method according to claim 1-3, wherein transmit leg is the different virtual machine that moves at polycaryon processor with the take over party.
7. each described method according to claim 1-3, the step that wherein sends information comprises: when beginning to send information, transmit leg detects the load level of each kernel, and just continues to carry out the subsequent step of transmission information during less than predetermined threshold in the load level.
8. each described method according to claim 1-3, the step that wherein receives information comprises: when finishing reception information, the information that take over party's verification receives, and the data that when having mistake, retransfer.
9. detection method based on the communication of polycaryon processor, described polycaryon processor are supported compatibility setting and the identification of kernel sign, and described method comprises:
The transmit leg that compatibility arranges is carried out in monitoring;
The take over party of kernel sign identification is carried out in monitoring; And
If not only have transmit leg but also have the take over party within a certain period of time, then judge to have the intercommunication that utilizes the passage of selecting nuclear mechanism.
10. method according to claim 9, wherein transmit leg is the different process of moving at polycaryon processor with the take over party.
11. method according to claim 9, wherein transmit leg is the different virtual machine that moves at polycaryon processor with the take over party.
12. the control method based on the communication of polycaryon processor, described polycaryon processor are supported compatibility setting and the identification of kernel sign, described method comprises:
Allow/forbid the compatibility setting of at least one kernel, and/or
Allow/forbid the kernel sign identification of at least one kernel.
13. method according to claim 12, compatibility setting and/or the identification of kernel sign of wherein forbidding whole kernels utilize the passage of selecting nuclear mechanism to close.
14. method according to claim 12 wherein allows compatibility setting and the identification of kernel sign of whole kernels, the bandwidth of selecting the passage of nuclear mechanism with maximum using.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110220307.9A CN102915292B (en) | 2011-08-02 | 2011-08-02 | Based on the communication means of polycaryon processor and detection method thereof and control method |
| PCT/CN2011/079608 WO2013016891A1 (en) | 2011-08-02 | 2011-09-14 | Communication method based on multi-core processor and detection method and control method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110220307.9A CN102915292B (en) | 2011-08-02 | 2011-08-02 | Based on the communication means of polycaryon processor and detection method thereof and control method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102915292A true CN102915292A (en) | 2013-02-06 |
| CN102915292B CN102915292B (en) | 2015-12-09 |
Family
ID=47613664
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110220307.9A Expired - Fee Related CN102915292B (en) | 2011-08-02 | 2011-08-02 | Based on the communication means of polycaryon processor and detection method thereof and control method |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN102915292B (en) |
| WO (1) | WO2013016891A1 (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104090826A (en) * | 2014-06-30 | 2014-10-08 | 中国电子科技集团公司第三十二研究所 | Task optimization deployment method based on correlation |
| CN107924343A (en) * | 2015-09-26 | 2018-04-17 | 英特尔公司 | Low latency inter-node communication |
| CN108768768A (en) * | 2018-05-15 | 2018-11-06 | 苏州科达科技股份有限公司 | Occlusion detection method, apparatus and storage medium for operation system |
| CN110262839A (en) * | 2019-06-17 | 2019-09-20 | 腾讯科技（深圳）有限公司 | A kind of dispatching method of processor, device and storage medium |
| CN114374848A (en) * | 2021-12-20 | 2022-04-19 | 杭州当虹科技股份有限公司 | Video coding optimization method and system |
| US11431820B1 (en) | 2021-11-05 | 2022-08-30 | Industrial Technology Research Institute | Method and system for implementing inter-process communications |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101089824A (en) * | 2006-06-13 | 2007-12-19 | 英业达股份有限公司 | Method for solving failure problem of self testing built-in central process unit |
| CN101169731A (en) * | 2007-12-05 | 2008-04-30 | 华为技术有限公司 | Multiple-path multiple-core server and its CPU virtualization processing method |
| US20120124339A1 (en) * | 2010-11-17 | 2012-05-17 | Krishnapura Sheshaprasad G | Processor core selection based at least in part upon at least one inter-dependency |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100517239C (en) * | 2007-06-14 | 2009-07-22 | 中兴通讯股份有限公司 | Multi-nuclear processor and serial port multiplexing method |
| JP5210730B2 (en) * | 2007-11-28 | 2013-06-12 | 株式会社日立製作所 | Virtual machine monitor and multiprocessor system |
| CN101266561B (en) * | 2008-04-29 | 2010-06-09 | 中兴通讯股份有限公司 | Inter-core message communication method for multi-core multithread processor |
| US8856794B2 (en) * | 2009-10-13 | 2014-10-07 | Empire Technology Development Llc | Multicore runtime management using process affinity graphs |
-
2011
- 2011-08-02 CN CN201110220307.9A patent/CN102915292B/en not_active Expired - Fee Related
- 2011-09-14 WO PCT/CN2011/079608 patent/WO2013016891A1/en active Application Filing
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101089824A (en) * | 2006-06-13 | 2007-12-19 | 英业达股份有限公司 | Method for solving failure problem of self testing built-in central process unit |
| CN101169731A (en) * | 2007-12-05 | 2008-04-30 | 华为技术有限公司 | Multiple-path multiple-core server and its CPU virtualization processing method |
| US20120124339A1 (en) * | 2010-11-17 | 2012-05-17 | Krishnapura Sheshaprasad G | Processor core selection based at least in part upon at least one inter-dependency |
Non-Patent Citations (2)
| Title |
|---|
| 李静梅，王军锋，张歧: "一种适应多核处理器核间通信机制的设计", 《智能计算机与应用》 * |
| 杨国芳: "多核处理器核间通信技术研究", 《哈尔滨工程大学硕士论文》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104090826A (en) * | 2014-06-30 | 2014-10-08 | 中国电子科技集团公司第三十二研究所 | Task optimization deployment method based on correlation |
| CN104090826B (en) * | 2014-06-30 | 2017-07-11 | 中国电子科技集团公司第三十二研究所 | Task optimization deployment method based on correlation |
| CN107924343A (en) * | 2015-09-26 | 2018-04-17 | 英特尔公司 | Low latency inter-node communication |
| CN107924343B (en) * | 2015-09-26 | 2021-10-08 | 英特尔公司 | Low latency inter-node communication |
| CN108768768A (en) * | 2018-05-15 | 2018-11-06 | 苏州科达科技股份有限公司 | Occlusion detection method, apparatus and storage medium for operation system |
| CN108768768B (en) * | 2018-05-15 | 2020-12-01 | 苏州科达科技股份有限公司 | Blocking detection method and device for business system and storage medium |
| CN110262839A (en) * | 2019-06-17 | 2019-09-20 | 腾讯科技（深圳）有限公司 | A kind of dispatching method of processor, device and storage medium |
| CN110262839B (en) * | 2019-06-17 | 2021-08-31 | 腾讯科技（深圳）有限公司 | Scheduling method and device of processor and storage medium |
| US11431820B1 (en) | 2021-11-05 | 2022-08-30 | Industrial Technology Research Institute | Method and system for implementing inter-process communications |
| TWI791316B (en) * | 2021-11-05 | 2023-02-01 | 財團法人工業技術研究院 | Method and system for implementing inter-process communications |
| CN114374848A (en) * | 2021-12-20 | 2022-04-19 | 杭州当虹科技股份有限公司 | Video coding optimization method and system |
| CN114374848B (en) * | 2021-12-20 | 2024-03-19 | 杭州当虹科技股份有限公司 | Video coding optimization method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2013016891A1 (en) | 2013-02-07 |
| CN102915292B (en) | 2015-12-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102915292B (en) | Based on the communication means of polycaryon processor and detection method thereof and control method | |
| CN103678003B (en) | The virtual cpu dispatching method that a kind of real-time strengthens | |
| EP2696284B1 (en) | Virtualized management method for hardware resources of communication device and relevant apparatus | |
| RU2285947C2 (en) | Method for ensuring safety with determined execution in real time of multi-task application of control-adjustment type with localization of errors | |
| EP2128759B1 (en) | Operating system startup control method and information processing apparatus | |
| CN103942178A (en) | Communication method between real-time operating system and non-real-time operating system on multi-core processor | |
| US20150033228A1 (en) | Method and apparatus for recovering lock holder preemption | |
| CN101894047A (en) | Kernel virtual machine scheduling policy-based implementation method | |
| CN104598298A (en) | Virtual machine dispatching algorithm based on task load and current work property of virtual machine | |
| US10782999B2 (en) | Method, device, and single-tasking system for implementing multi-tasking in single-tasking system | |
| CN103049332A (en) | Virtual CPU scheduling method | |
| CN108804109B (en) | Industrial deployment and control method based on multi-path functional equivalent module redundancy arbitration | |
| CN103955398A (en) | Virtual machine coexisting scheduling method based on processor performance monitoring | |
| US11115232B2 (en) | Method and device for operating a control unit | |
| CN108829510A (en) | Thread binds treating method and apparatus | |
| CN104156260A (en) | Concurrent queue access control method and system based on task eavesdropping | |
| CN100383743C (en) | Real-time task scheduling method in Java operating system | |
| CN103077081B (en) | Method and device of adjusting resources | |
| CN111694787A (en) | Chip starting method, network equipment and machine readable storage medium | |
| EP2113841A1 (en) | Allocating resources in a multicore environment | |
| Kim et al. | Reducing memory interference latency of safety-critical applications via memory request throttling and Linux Cgroup | |
| Chiba et al. | A distributed real-time operating system with distributed shared memory for embedded control systems | |
| CN114281529A (en) | Distributed virtualized client operating system scheduling optimization method, system and terminal | |
| CN104765613B (en) | Towards the optimization method of tasks in parallel programming model under a kind of virtualized environment | |
| Wellings | Multiprocessors and the real-time specification for java |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20151209 Termination date: 20210802 |