WO2015113211A1

WO2015113211A1 - Reentrant resource scheduling method, device and system

Info

Publication number: WO2015113211A1
Application number: PCT/CN2014/071681
Authority: WO
Inventors: 祁凯; 王伟; 蔡毅
Original assignee: 华为技术有限公司
Priority date: 2014-01-28
Filing date: 2014-01-28
Publication date: 2015-08-06
Also published as: CN104956328A; CN104956328B

Abstract

Disclosed in an embodiment of the present application is a reentrant resource scheduling method, device and system. The method is applied to a reentrant resource scheduling system; the reentrant resource scheduling system comprises an operating system, a scheduling buffer and a field programmable gate array (FPGA), and the operating system is used for performing the following steps: monitoring the operating condition information of a reentrant resource; determining that the reentrant resource satisfies a scheduling condition according to the operating condition information of the reentrant resource; storing the reentrant resource satisfying the scheduling condition into the scheduling buffer, the format of the reentrant resource being a format having been converted and being identifiable by the FPGA; scheduling the reentrant resource stored in the scheduling buffer into the FPGA, such that the reentrant resource becomes an FPGA service program. The present application can automatically achieve scheduling into the FPGA of the reentrant resource satisfying the scheduling condition according to the operating condition information of the reentrant resource obtained through monitoring, thus improving working efficiency.

Description

Reentrant resource scheduling method, device and system

The present invention relates to the field of computer application technologies, and in particular, to a re-entry resource scheduling method, device, and system. Background technique

FPGA (Field Programmable Gate Array), as a semi-custom circuit in the field of ASICs, users can connect the logic blocks inside the FPGA through an editable connection, just like a circuit test board. Was placed in a chip. The FPGA embedded system consists of a microprocessor unit, an FPGA part, and an externally EEPROM (Electrically Erasable Programmable Read Only Memory) device. The microprocessor unit is the main control part of the system and is mainly responsible for managing the FPGA. The system runs normally, the control command is issued, and the operation result of the receiving FPGA system is returned. The peripheral devices in the FPGA are connected to the micro processing unit and the external input/output interface. The static area in the FPGA is the firmware part of the FPGA and the related configuration part. The reentrant area is the FPGA business program in the EEPROM after powering up the FPGA system, and the FPGA business program inside the FPGA can also be dispatched into the EEPROM; the EEPROM is an executable code cache device running on the FPGA, mainly stored on the FPGA. Before the power is required, the FPGA service program inside the FPGA and the FPGA business program in the EEPROM are scheduled from the FPGA after the FPGA is powered on.

In the prior art, after the FPGA system is powered on, the FPGA service program in the EEPROM is dispatched to the FPGA. In this case, if the program is loaded into the FPGA or the program is uninstalled from the FPGA, the control command needs to be manually input to enable the micro processing unit to issue the data. The control command is sent to the FPGA, and then the FPGA loads or unloads the corresponding program, wherein the loading or unloading program needs manual editing, the operation is cumbersome, and the work efficiency is low. Summary of the invention

The technical problem to be solved by the embodiments of the present invention is to provide a re-entry resource scheduling method, device and system, which can automatically schedule reentrant resources into an FPGA to improve work efficiency. A first aspect of the present invention provides a reentrant resource scheduling method, the method being used in a reentrant resource scheduling system, where the reentrant resource scheduling system includes an operating system, a scheduling buffer, and an FPGA, where the operation The system is used to perform the following steps:

Monitor operational status information of reentrant resources;

Determining, according to the running status information of the reentrant resource, that the reentrant resource meets a scheduling condition;

Storing the reentrant resource satisfying the scheduling condition into the scheduling buffer, wherein the format of the reentrant resource stored in the scheduling buffer is a converted format that can be recognized by the FPGA ;

Dislocating the reentrant resource stored in the scheduling buffer to the FPGA to make the reentrant resource an FPGA service program.

In a first possible implementation manner, the reentrant resource includes a user program;

Determining, according to the operating state information of the reentrant resource, that the reentrant resource meets a scheduling condition includes:

Determining that the user program is in an executable state according to the running status information of the user program; and determining that the reentrant resource satisfies a scheduling condition when the user program is in an operational state.

With reference to the first possible implementation manner of the first aspect, in a second possible implementation manner, when the user program is in an executable state, determining that the reentrant resource meets a scheduling condition includes: ;

When the user program is in an executable state, determining that the memory usage occupied by the user program in the operable state reaches a preset threshold; determining that the reentrant resource meets the scheduling condition.

In conjunction with the possible implementation of the first aspect, in a third possible implementation, the re-entrant resource includes a kernel driver;

Determining that the kernel driver is being enabled according to the running state information of the kernel driver State of use

When the kernel driver is in a used state, it is determined that the reentrant resource satisfies a scheduling condition.

In conjunction with the possible implementation of the first aspect, in a fourth possible implementation, monitoring the running state information of the FPGA service program;

Determining, according to the running state information of the FPGA service program, that the FPGA service program is in a sleep state or an unused state;

The FPGA service program that determines the sleep state or the unused state is deleted.

With reference to the first aspect, or any one of the first aspect of the first aspect to the fourth aspect of the first aspect, in a fifth possible implementation manner, the monitoring re-entrant resource running state Information includes;

Sending status request information to the reentrant resource;

Receiving, by the status information, the operational status information returned by the reentrant resource according to the status request information, the threshold value, or whether the kernel driver is in a used state.

With reference to the fourth possible implementation of the first aspect, in a sixth possible implementation, the monitoring, by using the operating status information of the FPGA service program, includes:

Sending status request information to the FPGA service program;

Receiving operation state information returned by the FPGA service program according to the status request information, the operation status information including whether the FPGA service program is in a sleep state or an unused state.

With reference to the fifth possible implementation manner of the first aspect, in a seventh possible implementation manner, the status request information includes the following information: the version identifier, the information type, the information length, the transmission identifier, and the target resource type;

The operating status information includes:

Version ID, information type, message length, transport ID, and target resource type.

With reference to the second possible implementation manner of the first aspect, in an eighth possible implementation, the storing, by the storing, the re-entrant resource that meets the scheduling condition into the scheduling buffer includes: The statistical identifier is used to indicate that the user program in the operable state is occupied. Memory usage reaches a preset threshold;

And storing, according to the statistical identifier, the reentrant resource that satisfies the scheduling condition into the scheduling buffer.

In conjunction with the third possible implementation of the first aspect, in a ninth possible implementation, the storing the re-entrant resource that meets the scheduling condition into the scheduling buffer includes: The memory usage occupied by the kernel driver reaches a preset threshold;

With reference to the eighth or the ninth possible implementation manner of the first aspect, in the tenth possible implementation manner, the acquiring the statistical identifier includes:

Obtaining statistical information according to the running status information of the reentrant resource, where the statistical information includes a version identifier, an information type, an information length, a transmission identifier, a target resource type, and a statistical identifier.

With reference to the fourth possible implementation of the first aspect, in an eleventh possible implementation, the deleting the FPGA service program that is determined to be in a sleep state or an unused state includes: when the FPGA service program Obtaining a statistical identifier when the user is in a sleep state or an unused state, where the statistical identifier is used to indicate that the FPGA service program is in a sleep state or an unused state; according to the statistical identifier, deleting the determination to be in a sleep state or not The state of the FPGA business process.

With reference to the eleventh possible implementation manner of the foregoing aspect, in a twelfth possible implementation, the obtaining the statistical identifier when the FPGA service program is in a sleep state or an unused state includes:

The statistical information is obtained according to the running state information of the FPGA service program, where the statistical information includes a version identifier, an information type, an information length, a transmission identifier, a target resource type, and a statistical identifier.

In conjunction with the possible implementation of the first aspect, in a thirteenth possible implementation, the re-entry resource stored in the scheduling buffer is scheduled to be included in the FPGA;

Obtaining a scheduling identifier when the reentrant resource stored in the scheduling buffer is scheduled into the FPGA, the scheduling identifier is used to indicate that the reentrant resource stored in the scheduling buffer has succeeded Dispatched into the FPGA. With reference to the thirteenth possible implementation manner of the first aspect, in a fourteenth possible implementation manner, when the reentrant resource stored in the scheduling buffer is scheduled to be sent to the FPGA, The scheduling identifier includes:

Obtaining scheduling status information when scheduling re-entrant resources stored in the scheduling buffer into the FPGA, the scheduling status information including version identification, information type, information length, transmission identification, target resource type, and scheduling Logo.

A second aspect of the present invention provides a computer storage medium, wherein the computer storage medium stores a program, and the program includes all or part of the steps of the reentrant resource scheduling method provided by the first aspect of the embodiment of the present invention.

A third aspect of the present invention provides a reentrant resource scheduling device, including:

a status monitoring unit, configured to monitor running status information of reentrant resources;

a condition determining unit, configured to determine, according to the running status information of the reentrant resource monitored by the status monitoring unit, that the reentrant resource meets a scheduling condition;

a resource storage unit, configured to store, by the condition determining unit, a reentrant resource that meets a scheduling condition, into a scheduling buffer, where the format of the reentrant resource stored in the scheduling buffer is converted a format that can be recognized by the FPGA;

And a resource scheduling unit, configured to allocate the reentrant resource stored in the scheduling buffer to the FPGA, so that the reentrant resource becomes an FPGA service program.

The condition determining unit includes:

a status determining module, configured to determine, according to the running status information of the obtained user program by the status monitoring unit, that the user program is in an operable state;

And a condition determining module, configured to determine, when the state determining module determines that the user program is in an executable state, determine that the reentrant resource meets a scheduling condition.

In conjunction with the first possible implementation of the third aspect, in a second possible implementation, the condition determining module includes:

a memory determining submodule, configured to: when the state determining module determines that the user program is in a runnable value; And a condition determining submodule, configured to determine, when the memory determining submodule determines that the memory usage occupied by the user program in the operable state reaches a preset threshold, determining that the reentrant resource meets the scheduling condition.

In conjunction with the possible implementation of the third aspect, in a third possible implementation, the re-entrant resource includes a kernel driver;

The state determining module is further configured to determine, according to the running state information of the kernel driver that is obtained by the state monitoring unit, that the kernel driver is in a used state;

The condition determining module is further configured to: when the state determining module determines that the kernel driver is in a used state, determine that the reentrant resource satisfies a scheduling condition.

In conjunction with the possible implementation of the third aspect, in a fourth possible implementation, the state monitoring unit is further configured to monitor operation state information of the FPGA service program.

The condition determining unit is further configured to determine, according to the running state information of the FPGA service program that is monitored by the state monitoring unit, that the FPGA service program is in a sleep state or an unused state;

The reentrant resource scheduling device further includes a resource deletion unit, configured to delete the FPGA service program that the condition determining unit determines to be in a sleep state or an unused state.

With reference to the third aspect, or any one of the first to third aspects of the third aspect, in a fifth possible implementation, the state monitoring unit includes:

a status request sending module, configured to send status request information to the reentrant resource, and an operation status receiving module, configured to receive operation status information returned by the re-entrant resource according to the status request information sent by the status request sending module The running status information includes whether the memory usage occupied by the user program in the operable state reaches a preset threshold, or whether the kernel driver is in the used state.

With the fourth possible implementation of the third aspect, in a sixth possible implementation, the status request sending module is further configured to send status request information to the FPGA service program; And for receiving the running status information returned by the FPGA service program according to the status request information sent by the status request sending module, where the running status information includes whether the FPGA service program is in a sleep state or an unused state.

In combination with the fifth possible implementation of the third aspect, in a seventh possible implementation, The status request information includes an indication requesting the following information: a version identifier, an information type, an information length, a transmission identifier, and a target resource type;

The operating status information includes:

With reference to the second possible implementation of the third aspect, in an eighth possible implementation, the resource storage unit includes:

a statistical identifier obtaining module, configured to acquire a statistical identifier when the condition determining module determines that the memory usage of the user program in the operable state reaches a preset threshold, where the statistical identifier is used by the resource storage module, The statistical identifier obtained by the statistic identifier obtaining module is stored in the scheduling buffer by the reentrant resource that satisfies the scheduling condition.

With reference to the third possible implementation manner of the third aspect, in a ninth possible implementation manner, the statistic identifier is obtained, and when the condition determining module determines that the kernel driver is in the Obtaining a statistical identifier when the state is used, where the statistical identifier is used to indicate that the memory usage occupied by the kernel driver reaches a preset threshold;

The resource storage module is further configured to store, according to the statistical identifier obtained by the statistical identifier obtaining module, the reentrant resource that meets the scheduling condition into the scheduling buffer.

With reference to the eighth or the ninth possible implementation manner of the third aspect, in the tenth possible implementation manner, the statistics identifier obtaining module is further configured to obtain statistics according to the running state information of the reentrant resource The information includes the version identifier, the information type, the information length, the transmission identifier, the target resource type, and the statistical identifier.

With the fourth possible implementation of the third aspect, in an eleventh possible implementation manner, the resource deleting unit includes:

And a statistical identifier obtaining module, configured to acquire a statistical identifier when the condition determining unit determines that the FPGA service program is in a sleep state or an unused state, where the statistical identifier is used to indicate that the FPGA service program is in a sleep state or not status of use;

And a resource deletion module, configured to delete, according to the statistical identifier acquired by the statistical identifier obtaining module, the FPGA service program that is determined to be in a sleep state or not in use.

In conjunction with the eleventh possible implementation of the third aspect, in a twelfth possible implementation The statistical identifier obtaining module is further configured to obtain statistical information according to the running state information of the FPGA service program, where the statistical information includes a version identifier, an information type, an information length, a transmission identifier, a target resource type, and a statistical identifier.

In conjunction with the possible implementation of the first aspect, in a thirteenth possible implementation, the resource scheduling unit is further configured to: when the resource storage unit is stored in the scheduling buffer, reentrant resource scheduling A scheduling identifier is obtained when the FPGA is in use, and the scheduling identifier is used to indicate that the reentrant resource stored in the scheduling buffer has been successfully scheduled into the FPGA.

With reference to the thirteenth possible implementation manner of the first aspect, in a fourteenth possible implementation, the resource scheduling unit is further configured to: when the resource storage unit is to be stored in the scheduling buffer The scheduling state information is obtained when the reentrant resource is scheduled into the FPGA, and the scheduling state information includes a version identifier, an information type, an information length, a transmission identifier, a target resource type, and a scheduling identifier.

A fourth aspect of the present invention provides a re-entrant resource scheduling device, where the device includes an FPGA, a memory, and a processor, where the memory includes a scheduling buffer, the memory stores a set of program codes, and the processing It is used to call the program code stored in the memory to perform the following operations:

Monitor operational status information of reentrant resources;

The reentrant resources stored in the scheduling buffer are dispatched to the FPGA to cause the reentrant resources to become FPGA service programs.

In the embodiment of the present invention, by monitoring the obtained running state information of the reentrant resource, determining that the reentrant resource meets the scheduling condition, and storing the reentrant resource that meets the scheduling condition into the scheduling buffer, where The format of the reentrant resource stored in the scheduling buffer is a converted format recognizable by the FPGA, and further scheduling the reentrant resource stored in the scheduling buffer to In the FPGA, the reentrant resource is made into an FPGA service program, and the running state information of the reentrant resource obtained according to the monitoring is automatically implemented, and the re-entry resource satisfying the scheduling condition is used. The source is dispatched to the FPGA to improve work efficiency. DRAWINGS

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings to be used in the embodiments will be briefly described below. Obviously, the drawings in the following description are only the present invention. In some embodiments, other drawings may be obtained from those of ordinary skill in the art in light of the inventive work.

FIG. 1 is a schematic flowchart of a re-entry resource scheduling method according to an embodiment of the present invention; FIG. 2 is a schematic flowchart of a reentrant resource scheduling method according to another embodiment of the present invention; FIG. 4 is a schematic flowchart of a re-entry resource scheduling method according to another embodiment of the present invention; FIG. 5 is a schematic diagram of a kernel driver according to an embodiment of the present invention; FIG. 6 is a schematic flowchart of a re-entry resource scheduling method according to another embodiment of the present invention; FIG. 7 is a schematic diagram of a scheduling state machine of an FPGA service program according to an embodiment of the present invention; 8 is a schematic structural diagram of a re-entrant resource scheduling device according to an embodiment of the present invention; FIG. 9 is a schematic structural diagram of a condition determining unit of FIG. 8 according to an embodiment of the present invention; FIG. 10 is a schematic diagram of FIG. FIG. 11 is a schematic structural diagram of a state monitoring unit of FIG. 8 according to an embodiment of the present invention; FIG. 12 is a schematic diagram of the present invention; FIG. 13 is a schematic structural diagram of a resource deletion unit of FIG. 8 according to an embodiment of the present invention; FIG. 14 is a re-entrant resource scheduling system according to an embodiment of the present invention; Schematic diagram of the structure. detailed description

BRIEF DESCRIPTION OF THE DRAWINGS The technical solutions in the embodiments of the present invention will be described in detail below with reference to the accompanying drawings. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present invention without creative work are within the scope of the present invention.

Referring to FIG. 1, FIG. 1 is a flowchart of a reentrant resource scheduling method according to an embodiment of the present invention. The method is used in a re-entrant resource scheduling system, where the re-entrant resource scheduling system includes an operating system, a scheduling buffer, and a field editable logic gate array FPGA, wherein the operating system is configured to perform the following steps:

S101. Monitor running status information of reentrant resources.

The operating system can monitor operational status information for reentrant resources, which can include user programs and kernel drivers. Optionally, the operating system can also monitor the operational status information of the FPGA business program.

Specifically, the monitoring module in the operating system may send status request information to the reentrant resource and the FPGA service program, so that the reentrant resource and the FPGA service program return the running status information according to the status request information, where the operation The status information may include whether the memory usage occupied by the user program in a runnable state reaches a preset threshold, or whether the kernel driver is in a used state, or whether the FPGA service program is in a sleep state or an unused state, thereby monitoring The module may receive the re-entrant resource or operational status information returned by the FPGA service program, where the status request information may be used to indicate the following information: a version identifier, an information type, an information length, a transmission identifier, and a target resource type, The operational status information may include a version identification, an information type, an information length, a transmission identification, and a target resource type.

Busy Information Request Message

Header

Version: 0x00

Message Type: FPGA Busy Request Information

Length: 24

Transaction ID: 0

Busy Information Request

Type: User APP/ Kernel Module I FPGA img

Body: ***** //Different type fill request section

The information format of the exemplary status request information is as shown above. The version identifier in the status request information may be 0x00, the information type may be FPGA busy request information, the information length may be 24 bit bits, and the transmission identifier may be “0”. The target resource type may be a user program, a kernel driver, or an FPGA service program, and the status request information may further include identification information of the target resource, where the target The identification information of the target resource may be a fill request of the target resource. The operating system broadcasts status request information to the user program, the kernel driver, and the FPGA service program, and the user program, the kernel driver, and the FPGA service program return the running status information after receiving the status request information, and the running status information may be Includes version ID, information type, message length, transport ID, and target resource type.

Busy Information Reply Message

Header

Version: 0x00

Message Type: FPGA Busy Reply Information

Length: 24

Transaction ID: 0

Busy Information Reply

Type: User APP/ Kernel Module I FPGA img

Body: ***** // User App Status : high / low

// kernel module Used by : used / not used

// FPGA img Status : high / low

The information format of the exemplary running state information is as shown above, the version identifier in the running state information may be 0x00, the information type may be FPGA busy response information, the information length may be 24 bits, and the transmission identifier may be “0”, the target The resource type may be a user program, a kernel driver, or an FPGA business program. For example, when the memory usage occupied by the user program in a runnable state reaches a preset threshold, the running state of the user program may be high load; the kernel driver When in the used state, the running state of the kernel driver may be used; when the FPGA service program is in a sleep state or an unused state, the running state of the FPGA service program may be idle.

S102. Determine, according to the running status information of the reentrant resource, that the reentrant resource satisfies the scheduling condition. The operating system may determine, according to the running status information of the reentrant resource, that the reentrant resource satisfies the scheduling condition.

In a specific implementation, when the reentrant resource is a user program, the monitoring module in the operating system may determine, according to the received running state information of the user program, that the user program is in a runnable state. And when the user program is in an operational state, determining that the reentrant resource satisfies a scheduling condition. Optionally, when determining that the user program is in a runnable state, the operating system may further determine that the memory usage occupied by the user program in the operable state reaches a preset threshold, when the user program is in a runnable state. When the occupied memory usage reaches a preset threshold, it is determined that the reentrant resource satisfies the scheduling condition.

When the reentrant resource is a kernel driver, the monitoring module in the operating system may determine, according to the received running state information of the kernel driver, that the kernel driver is in a used state, and then when the kernel driver When the program is in the used state, it is determined that the reentrant resource satisfies the scheduling condition.

Optionally, the monitoring module in the operating system may further determine, according to the running status information of the received FPGA service program, that the FPGA service program is in a sleep state or an unused state.

S103. Store the reentrant resource that meets the scheduling condition into the scheduling buffer.

The operating system may store the reentrant resource satisfying the scheduling condition into a scheduling buffer, wherein the format of the reentrant resource stored in the scheduling buffer is converted to be recognized by the FPGA format. Optionally, the operating system may further delete the FPGA service program that is determined to be in a sleep state or an unused state. Further, the operating system may further use the FPGA service program that is determined to be in a sleep state or an unused state. Dispatched into the dispatch buffer.

Specifically, after the monitoring module in the operating system receives the running status information, the management module in the operating system may generate statistical request information, and send the statistical request information to the monitoring module to enable the threshold, or when the kernel When the driver is in the used state, or when the FPGA service program is in the sleep state or the unused state, the management module acquires the statistical identifier returned by the monitoring module, and the statistical identifier is used to indicate that the user program is in the operable state. The memory usage reaches a preset threshold, or the kernel driver is in a used state, or the FPGA service program is in a sleep state or an unused state. Further, the management module may further acquire statistics returned by the monitoring module. The further operating system may store the reentrant resource in the scheduling buffer or delete the FPGA service program according to the obtained re-entrant resource or statistical information of the FPGA service program. After the operating system stores the reentrant resource in the scheduling buffer, the re-entrant resource may also be prioritized, wherein the statistical request information may be used to indicate the following letter Information: a version identifier, an information type, an information length, a transmission identifier, and a target resource type, where the statistical information may include a version identifier, an information type, an information length, a transmission identifier, a target resource type, and a statistical identifier, such as the reentrant resource. Is a user program, where the running status of the user program is high, and the statistical identifier of the user program acquired by the management module is “Γ”, which is used to indicate that the memory usage occupied by the user program in the executable state reaches a preset threshold; if the running state of the user program is a low load, the statistical identifier of the user program acquired by the management module is "0", which is used to indicate that the user program is not in an operable state, or is in a The memory usage occupied by the running user program does not reach the preset threshold.

FPGA Busy Stastic Information Message

Header

Version: 0x00

Message Type: FPGA Busy Stastic Information

Length: 32

Transaction ID: 1

Busy Stastic Request

Message Type: FPGA Busy Stastic Information Request

Type: User APP/ Kernel Module I FPGA img

Flag: 0/1 〃 Resources are idle and busy

Busy Stastic Reply

Message Type: FPGA Busy Stastic Information Reply

Type: User APP/ Kernel Module I FPGA img

Flag: 0/1 〃 Resources are idle and busy

The information format of the exemplary statistical request information and the statistical information is as shown in the above, and the statistical request information is used to indicate the version identifier, the information type, the information length, the transmission identifier, and the target resource type, and the statistical information received by the management module. The version identifier can be 0x00, the information type can be FPGA busy statistics, the information length can be 32bit, the transmission identifier can be "Γ, the target resource type can be a user program, a kernel driver or an FPGA service program, and the statistical identifier is used to indicate The amount of memory used by a user program in a runnable state reaches a preset threshold, or the kernel driver is in a used state, or the FPGA business program is in a sleep state or is not in use, such as when When the memory usage of the user program in the runnable state reaches the preset threshold, the obtained statistical identifier may be “Γ”, indicating that the user program is in a busy state.

S104. Dispatch the reentrant resource stored in the scheduling buffer into the FPGA, so that the reentrant resource becomes an FPGA service program.

The operating system can schedule the reentrant resources stored in the scheduling buffer into the FPGA, so that the reentrant resource becomes an FPGA service program, wherein the operating system can prioritize the scheduling buffer according to a preset priority. The highest re-entrant resource is scheduled into the FPGA, making the reentrant resource an FPGA service program and running the reentrant resource in the FPGA.

Specifically, the scheduling module in the operating system may generate scheduling request information, and send the scheduling request information to the management module, so that the management module returns a statistical identifier corresponding to the target resource type in the scheduling request information, and the scheduling module further The statistical identifier schedules the reentrant resource stored in the scheduling buffer into the scheduling buffer, or deletes the FPGA service program in the FPGA according to the statistical identifier, when the storage into the scheduling buffer is The scheduling module may obtain a scheduling identifier, where the scheduling identifier is used to indicate that the reentrant resource stored in the scheduling buffer has been successfully scheduled into the FPGA, for example, When the reentrant resource stored in the scheduling buffer has been successfully scheduled into the FPGA, the scheduling identifier acquired by the operating system may be “Γ; when stored in the scheduling buffer When the reentrant resource is not successfully scheduled into the FPGA, the scheduling identifier obtained by the operating system may be "0". In the sequence, when the operating system has successfully deleted the FPGA service program, the scheduling identifier obtained by the operating system may be “Γ; when the FPGA service program is not successfully deleted, the scheduling identifier acquired by the operating system may be "0". . The operating system may also acquire scheduling state information, where the scheduling state information may include a version identifier, an information type, an information length, and a transmission identifier, when the reentrant resource stored in the scheduling buffer is scheduled into the FPGA. , target resource type, and schedule ID.

Further, the operating system may determine, according to the scheduling identifier in the scheduling state information, whether the reentrant resource or the FPGA service program scheduling is successful, and if the scheduling identifier in the scheduling state information is “1”, the If the re-entrant resource scheduling is successful or the FPGA service program has been successfully deleted; if the scheduling identifier in the scheduling status information is "0", it indicates that the reentrant resource scheduling is unsuccessful or the FPGA service program is not successfully deleted, optionally, operation The system may re-schedule the reentrant resource stored in the scheduling buffer to the FPGA when the preset time is reached, or delete the second time when the preset time is reached. FPGA business process.

HW Process Schedule Message

Header

Version: 0x00

Message Type: HW Process Schedule Information

Length: 64

Transaction ID: 2

HW Processs Schdule Request

Message Type: HW Process Schedule Information Request

Type: User APP/ Kernel Module I FPGA img

Flag: 0/1 //The resource schedule failed or succeeded.

HW Processs Schdule Reply

Message Type: HW Process Schedule Information Reply

Type: User APP/ Kernel Module I FPGA img

Flag: 0/1 //The resource schedule failed or succeeded.

The information format of the exemplary scheduling state information is as shown above. The version identifier in the scheduling state information may be 0x00, the information type may be resource scheduling information, the information length may be 64 bits, and the transmission identifier may be "2", the target resource. The type may be a user program, a kernel driver, or an FPGA business program, and the scheduling identifier may be "0" or "Γ, where the scheduling identifier "0" is used to indicate that the reentrant resource scheduling fails or the FPGA service program is not successfully deleted. The scheduling identifier "" is used to indicate that the reentrant resource scheduling is successful or the FPGA service program has been successfully deleted.

In the re-entrant resource scheduling method shown in FIG. 1 , the operating system determines that the reentrant resource meets the scheduling condition by monitoring the obtained running state information of the reentrant resource, and the re-entrant that satisfies the scheduling condition is determined. The resource is stored in the scheduling buffer, and the reentrant resource stored in the scheduling buffer is scheduled into the FPGA, so that the reentrant resource becomes an FPGA service program, and the obtained according to the monitoring can be automatically implemented. Re-enter the running status information of the resource, and schedule the reentrant resources satisfying the scheduling condition to the FPGA to improve work efficiency. Referring to FIG. 2, FIG. 2 is a flowchart of a reentrant resource scheduling method according to another embodiment of the present invention. In the embodiment of the present invention, the re-entrant resource scheduling method is described by using a user program as an example. The method is used in a re-entrant resource scheduling system, where the re-entrant resource scheduling system includes an operating system. A scheduling buffer and a field editable logic gate array FPGA, wherein the operating system is configured to perform the following steps:

S201. Monitor operation status information of the user program.

The operating system can monitor the running status information of the user program. Specifically, the monitoring module in the operating system can send status request information to the user program, so that the user program returns running status information according to the status request information, where the operating status is The information may include whether the memory usage occupied by the user program in the operable state reaches a preset threshold, and the monitoring module receives the running status information returned by the user program. The status request information may be used to indicate the following information: a version identifier, an information type, an information length, a transmission identifier, and a target resource type, where the operation status information may include a version identifier, an information type, an information length, a transmission identifier, and a target resource. Types of.

As shown in FIG. 3, FIG. 3 is a schematic diagram of a scheduling state machine of a user program, and the operating system can monitor running state information of the user program, and the running state of the user program may include preparing a scheduling state, an operable state, a busy state, and an idle state.

S202. Determine, according to the running status information of the user program, that the user program is in an operable state. The operating system may determine that the user program is in an executable state according to running state information of the user program, and the operable state may include a running state and a ready running state. After the operating system determines that the user program is in a runnable state, it may also determine that the memory usage of the user program reaches a preset threshold, for example, the CPU (Central Processing Unit) usage rate of the user program reaches a preset. After the monitoring module in the operating system receives the running status information, the management module in the operating system may generate statistical request information, and send the statistical request information to the monitoring module, so that the monitoring module determines that the operating module is operable. When the memory usage of the user program of the state reaches the preset threshold, the statistical identifier returned by the management module to obtain the monitoring module may be "Γ, statistical identifier" Γ used to indicate that the memory usage occupied by the user program in the operable state reaches the pre-pre- The threshold value indicates that the user program is in a busy state, and the management module may further obtain statistics information returned by the monitoring module, where the statistics information may include a version identifier, an information type, an information length, a transmission identifier, a target resource type, and a statistical identifier. S204: Store the user program that is determined to be in a runnable state and the system resource usage reaches a preset threshold into a scheduling buffer. The statistical identifier of the user program is "Γ", and the management module in the operating system may store the user program into the scheduling buffer according to the statistical identifier.

5205. Dispatched the user program stored in the scheduling buffer into the FPGA.

The operating system may schedule the user program stored in the scheduling buffer to the FPGA according to the statistical identifier in the statistical information of the user program. Specifically, the scheduling module in the operating system may generate scheduling request information, where The scheduling request information is sent to the management module, so that the management module returns a statistical identifier corresponding to the target resource type in the scheduling request information, and the scheduling module then schedules the user program stored in the scheduling buffer to the scheduling buffer according to the statistical identifier. In the area, the operating system may schedule the user program with the highest priority in the scheduling buffer to the FPGA according to a preset priority, so that the user program becomes an FPGA service program, and runs in the FPGA. The user program.

Taking the scheduling state machine diagram of the user program shown in FIG. 3 as an example, the operating system monitors that the user program is in a preparation scheduling state, and when the time slice allocated by the user program reaches a preset time, the user program is manipulated. The system resource usage reaches a preset threshold, and the user program can be switched from a runnable state to a busy state, and the operating system can store the user program in a scheduling buffer, and the Dispatching the user program into the FPGA, making the user program an FPGA service program, and running the user program in the FPGA; if the operating system monitors that the system resource usage of the user program does not reach a preset threshold, Switching the user program from an operational state to an idle state, and if the operating system monitors that the system resource usage of the user program reaches a preset threshold, the user program may be switched from an idle state to a busy state, and then the operating system The user program can be stored in a dispatch buffer and will be in the dispatch buffer User scheduler to the FPGA, the FPGA user program becomes business processes, and the user runs the program in the FPGA.

5206. Determine whether the user program stored in the scheduling buffer is successfully dispatched to the FPGA. The scheduling module in the operating system dispatches the user program stored in the dispatch buffer to the FPGA In the middle, it can be determined whether the user program stored in the scheduling buffer has been successfully dispatched into the FPGA, and if so, step S207 is performed; if not, step S208 is performed.

S207. If yes, the acquired scheduling identifier in the scheduling status information for the user program is T.

If the user program stored in the scheduling buffer is successfully dispatched to the FPGA, the scheduling identifier for the user program acquired by the operating system may be “Γ, the scheduling identifier is used to indicate that the user program is successful. The scheduling may be performed by the operating system, where the operating system may also obtain scheduling status information, where the scheduling status information may include a version identifier, an information type, an information length, a transmission identifier, a target resource type, and a scheduling identifier.

S208. If no, the scheduling identifier in the acquired scheduling state information for the user program is "0".

If the user program stored in the dispatch buffer is not successfully dispatched to the FPGA, the scheduling identifier obtained by the operating system is "0".

Further, the operating system may determine, according to the scheduling identifier in the scheduling state information, whether the scheduling of the user program is successful, and if the scheduling identifier in the scheduling state information is “Γ, indicating that the user program is successfully scheduled; If the scheduling identifier in the information is "0", it indicates that the user program scheduling is unsuccessful. Optionally, the operating system can re-store the user program into the scheduling buffer when the preset time is reached, and store it in the scheduling buffer. The user program in the buffer is dispatched to the FPGA.

In the reentrant resource scheduling method shown in FIG. 2 and FIG. 3, the operating system determines that the user program is in a runnable state and the system resource usage reaches a preset threshold according to the monitored running state information of the user program. Storing the user program in a scheduling buffer, and scheduling the user program in the scheduling buffer to the FPGA, and then setting a scheduling identifier in the scheduling state information of the user program according to the scheduling result, which can be automatically implemented. The user program that satisfies the scheduling condition is dispatched to the FPGA according to the running state information of the monitored user program, thereby improving work efficiency. Referring to FIG. 4, FIG. 4 is a schematic flowchart of a method for re-entry resource scheduling according to another embodiment of the present invention. The embodiment of the present invention uses a kernel driver as an example to describe the reentrant resource scheduling method. The method is used in a reentrant resource scheduling system, where the reentrant resource scheduling system includes an operating system, a scheduling buffer, and a field editable logic gate array FPGA, wherein the operating system The system is used to perform the following steps:

5401. Monitor the running status information of the kernel driver.

The operating system can monitor the running status information of the kernel driver. Specifically, the monitoring module in the operating system can send status request information to the kernel driver, so that the kernel driver returns the running status information according to the status request information. The running status information may include whether the kernel driver program is in a used state, and the monitoring module receives the running status information returned by the kernel driver. The status request information may indicate the following information by the user: a version identifier, an information type, an information length, a transmission identifier, and a target resource type, where the operation status information may include a version identifier, an information type, an information length, a transmission identifier, and a target resource type. .

As shown in FIG. 5, FIG. 5 is a schematic diagram of a scheduling state machine of a kernel driver, and an operating system can monitor running state information of a kernel driver. The running state of the kernel driver may include preparing a scheduling state, an executable state, a busy state, and an idle state. Status, etc.

5402. Determine, according to the running status information of the kernel driver, that the kernel driver is in a used state.

The operating system can determine that the kernel driver is in use according to the running state information of the kernel driver.

Specifically, after the monitoring module in the operating system receives the running status information, the management module in the operating system may generate statistical request information, and send the statistical request information to the monitoring module, so that the monitoring module determines when the kernel driver is When the program is in the used state, the management module acquires a statistical identifier of the kernel driver returned by the monitoring module, where the statistical identifier is used to indicate that the kernel driver is in a used state, for example, the operating system determines that the kernel driver is in a state In the used state, the statistical identifier in the statistical information of the kernel driver can be set to "Γ, the statistical identifier "1" indicates that the kernel driver is in a busy state.

In the rushing area. Specifically, the operating system may store the kernel driver in the used state into the scheduling buffer according to the statistical identifier of the kernel driver.

S404, scheduling the kernel driver stored in the scheduling buffer into the FPGA. The operating system can store the statistic identifier in the statistics of the kernel driver to Dispatching the kernel driver in the scheduling buffer to the FPGA. Specifically, the scheduling module in the operating system may generate scheduling request information, and send the scheduling request information to the management module, so that the management module returns the scheduling request. The statistical identifier corresponding to the target resource type in the information, and the scheduling module then schedules the kernel driver stored in the scheduling buffer to the scheduling buffer according to the statistical identifier, wherein the operating system may schedule according to the preset priority. The kernel driver with the highest priority in the buffer is dispatched into the FPGA, making the kernel driver an FPGA business program, and running the kernel driver in the FPGA.

Taking the schematic diagram of the scheduling state machine of the kernel driver shown in FIG. 5 as an example, the operating system monitors that the kernel driver is in a ready-to-schedule state, and when the kernel driver is loaded into the operating system, the kernel driver is monitored. The program is currently being used to switch the kernel driver from a runnable state to a busy state, and the operating system can store the kernel driver in a dispatch buffer that will dispatch the kernel driver in the buffer. Dispatching into the FPGA, making the kernel driver an FPGA business program, and running the kernel driver in the FPGA; if the operating system monitors that the kernel driver is not used, the kernel driver may be Switching from an operational state to an idle state, when the operating system monitors that the kernel driver is used, the kernel driver can be switched from an operational state to a busy state, and the operating system stores the kernel driver to In the dispatch buffer, the kernel drive in the dispatch buffer will be further The program is programmed into the FPGA to make the kernel driver an FPGA business program and to run the kernel driver in the FPGA.

5405. Determine whether the kernel driver stored in the scheduling buffer is successfully dispatched into the FPGA.

When the operating system dispatches the kernel driver stored in the scheduling buffer to the FPGA, it can determine whether the kernel driver stored in the scheduling buffer has been successfully dispatched into the FPGA, and if yes, execute step S406; if not, Then step S407 is performed.

5406. If yes, the obtained scheduling identifier in the scheduling status information for the kernel driver is T.

If the kernel driver stored in the scheduling buffer has been successfully dispatched into the FPGA, the scheduling identifier obtained by the operating system for the kernel driver may be "Γ, where the scheduling state The state information may include a version identifier, an information type, an information length, a transmission identifier, a target resource type, and a scheduling identifier.

S407. If not, the acquired scheduling identifier in the scheduling status information for the kernel driver is "0".

If the kernel driver stored in the dispatch buffer is not successfully dispatched to the FPGA, the scheduling identifier obtained by the operating system may be "0".

Further, the operating system may determine, according to the scheduling identifier in the scheduling state information, whether the kernel driver scheduling is successful, and if the scheduling identifier in the scheduling state information is “Γ, indicating that the kernel driver is successfully scheduled; The scheduling identifier in the scheduling status information is "0", indicating that the kernel driver scheduling is unsuccessful. Optionally, the operating system may re-store the kernel driver into the scheduling buffer when the preset time is reached, and The kernel driver stored in the dispatch buffer is dispatched into the FPGA.

In the reentrant resource scheduling method shown in FIG. 4 and FIG. 5, the operating system determines that the kernel driver is in a used state according to the monitored running state information of the kernel driver, and stores the kernel driver. Go to the scheduling buffer, and schedule the kernel driver in the scheduling buffer to the FPGA, and then set the scheduling identifier in the scheduling state information for the kernel driver according to the scheduling result, which can automatically implement the monitoring according to the monitoring result. The running status information of the kernel driver, and the kernel driver that satisfies the scheduling condition is dispatched to the FPGA to improve work efficiency. Referring to FIG. 6, FIG. 6 is a schematic flowchart of a method for re-entry resource scheduling according to another embodiment of the present invention. The embodiment of the present invention uses an FPGA service program as an example to describe the reentrant resource scheduling method. The method is used in a re-entrant resource scheduling system, where the re-entrant resource scheduling system includes an operating system, a scheduling buffer, and a field editable logic gate array FPGA, wherein the operating system is configured to perform the following steps:

S601. Monitor operation status information of the FPGA service program.

The operating system can monitor the running status information of the FPGA service program. Specifically, the monitoring module in the operating system can send status request information to the FPGA service program, so that the FPGA service program returns the running status information according to the status request information. The running status information may include whether the FPGA service program is in a sleep state or an unused state, and the monitoring module can receive the The running status information returned by the FPGA service program, where the status request information may be used to indicate the following information: a version identifier, an information type, an information length, a transmission identifier, and a target resource type, where the running status information may include a version identifier and information. Type, length of information, transport ID, and target resource type.

As shown in FIG. 7, FIG. 7 is a schematic diagram of a scheduling state machine of an FPGA service program. The reentrant resource scheduling device can monitor operation state information of an FPGA service program, and the running state of the FPGA service program can include preparing a scheduling state, an operable state, Busy and idle status.

S602: Determine, according to the running status information of the FPGA service program, that the FPGA service program is in a sleep state or an unused state.

The operating system may determine that the FPGA service program is in a sleep state according to the running status information of the FPGA service program, wherein the sleep state is waiting for an event to occur or waiting to acquire system resources. Optionally, the reentrant resource scheduling device may further determine that the FPGA service program is in an unused state according to the running status information of the FPGA service program.

After the monitoring module in the operating system receives the running status information, the management module in the operating system may generate statistical request information, and send the statistical request information to the monitoring module, so that the monitoring module determines that the FPGA service program is in sleep. The management module obtains the statistical identifier returned by the monitoring module, and the statistical identifier is used to indicate that the FPGA service program is in a sleep state or an unused state. Further, the management module may obtain the monitoring module to return. Statistics, statistics information can include version identifier, information type, information length, transmission identifier, target resource type, and statistical identifier. If the reentrant resource scheduling device determines that the FPGA service program is in a sleep state or an unused state, the statistical identifier of the FPGA service program acquired by the management module may be "0", and the statistical identifier "0" indicates the The FPGA business process is idle.

Taking the scheduling state machine of the FPGA service program shown in FIG. 7 as an example, the operating system may determine that the FPGA service program is in an idle state according to the monitored operating state information of the FPGA service program, that is, the FPGA service program is in a sleep state. Or not in use.

S603. Delete the FPGA service program that is determined to be in a sleep state or an unused state. The operating system may delete the FPGA service program according to the statistical identifier in the statistical information of the FPGA service program. Optionally, the operating system may also schedule the FPGA service program in the sleep state or the unused state to the scheduling buffer. In the district. Specifically, if the FPGA service program The statistical identifier is "0", that is, the FPGA service program is in an idle state, and also indicates that the FPGA service program is in a sleep state or an unused state, and the FPGA service that is determined to be in a sleep state or an unused state is deleted. The program, optionally, the operating system can schedule the FPGA business program into a scheduling buffer. In an embodiment of the present invention, by deleting an FPGA service program that is determined to be in a sleep state or determined to be in an unused state, there may be an extra run queue in the FPGA for the operating system to schedule the user program or the kernel driver to the Runs in the queue and runs in the run queue.

In the scheduling state machine diagram of the FPGA service program shown in FIG. 7, when the operating system monitors that the reentrant resource is in the ready scheduling state, the reentrant resource in the scheduling buffer may be scheduled into the FPGA. The reentrant resource becomes an FPGA service program, and the FPGA service program is run in the FPGA, and the FPGA service program is in a state of preparation for scheduling; and the operating system can prepare the FPGA service program from preparation for scheduling by powering on the FPGA. The state is switched to the runnable state; if the operating system monitors that the FPGA service program load is low, the FPGA service program may be switched from the runnable state to the idle state, and the operating system may be based on the statistical identifier of the FPGA service program. "0", deleting the FPGA service program, and scheduling the FPGA service program from the FPGA to the scheduling buffer; if the operating system monitors that the FPGA service program load is high, the FPGA service program may be The operating state is switched to a busy state, and the operation is accelerated in the FPGA; if the operating system monitors the FPGA service Order low load, the FPGA program service can be switched from the busy state to the idle state, and thus the operating system may delete the service program FPGA, and the FPGA from the FPGA traffic scheduler in the scheduling buffer.

S604. Determine whether the FPGA service program is successfully deleted.

After the operating system deletes the FPGA service program, it may be determined whether the FPGA service program has been successfully deleted. If the FPGA service program has been successfully deleted, step S605 is performed; if the FPGA service program is not successfully deleted, the step is performed. S606.

S605. If yes, the obtained scheduling identifier in the scheduling status information for the FPGA service program is T.

If the FPGA service program has been successfully deleted, the scheduling identifier for the FPGA service program obtained by the operating system may be “Γ, where the scheduling status information obtained by the operating system may include a version identifier, an information type, an information length, Transport ID, target resource type, and schedule ID. S606. If no, the scheduling identifier in the acquired scheduling state information for the FPGA service program may be "0".

If the FPGA service program is not successfully deleted, the scheduling identifier in the scheduling status information acquired by the operating system may be "0".

Further, the operating system may determine, according to the scheduling identifier, whether the scheduling of the FPGA service program is successful. If the scheduling identifier in the scheduling status information is “Γ, the FPGA service program has been successfully deleted; if the scheduling status information is If the scheduling identifier is "0", it indicates that the FPGA service program is not successfully deleted. Optionally, the operating system may delete the FPGA service program again when the preset time is reached, and schedule the FPGA service program to the scheduling buffer. In the district.

In the reentrant resource scheduling method shown in FIG. 6 and FIG. 7, the operating system is obtained according to the monitoring.

The operating state information of the FPGA service program in the FPGA determines that the FPGA service program is in a sleep state or an unused state, and further deletes the FPGA service program that is determined to be in a sleep state or an unused state, thereby acquiring a scheduling identifier, The scheduling identifier is used to indicate whether the FPGA service program is successfully deleted, and the operating state information of the FPGA service program obtained according to the monitoring is automatically implemented, and the FPGA service program that meets the scheduling condition is deleted from the FPGA to improve work efficiency. The embodiment of the present invention further provides a computer storage medium, wherein the computer storage medium can store a program, and the program includes the reentrant resource scheduling described in the foregoing method embodiments shown in FIG. 1 to FIG. Part or all of the steps of the method. Referring to FIG. 8, FIG. 8 is a schematic structural diagram of a re-entrant resource scheduling device according to an embodiment of the present invention. The re-entrant resource scheduling device may be used to implement the method embodiment shown in FIG. 1 to FIG. In some or all of the steps of the re-entrant resource scheduling method, the re-entrant resource scheduling device may include at least a status monitoring unit 801, a condition determining unit 802, a resource storage unit 803, and a resource scheduling unit 804, where:

The status monitoring unit 801 is configured to monitor running status information of the reentrant resource.

The condition determining unit 802 is configured to determine, according to the running state information of the reentrant resource monitored by the state monitoring unit 801, that the reentrant resource meets the scheduling condition. The resource storage unit 803 is configured to store, by the condition determining unit 802, a reentrant resource that satisfies a scheduling condition, into a scheduling buffer, where the format of the reentrant resource stored in the scheduling buffer is The format of the conversion that can be recognized by the FPGA.

The resource scheduling unit 804 is configured to schedule the reentrant resource stored in the scheduling buffer by the resource storage unit 803 into the FPGA, so that the reentrant resource becomes an FPGA service program.

As an optional implementation, the re-entrant resource includes a user program, and the condition determining unit 802 may be as shown in FIG.

The status determining module 901 is configured to determine, according to the running status information of the user program that is monitored by the status monitoring unit 801, that the user program is in an operable state.

The condition determining module 902 is configured to determine, when the state determining module 901 determines that the user program is in an executable state, determine that the reentrant resource satisfies a scheduling condition.

Further, the condition determining module 902 may further include: a memory determining submodule 1001, configured to: when the state determining module 901 determines that the user program reaches a preset threshold;

The condition determining sub-module 1002 is configured to determine, when the memory determining sub-module 1001 determines that the memory usage occupied by the user program in the operable state reaches a preset threshold, determining that the re-entrant resource meets the scheduling condition.

As an optional implementation, the re-entrant resource includes a kernel driver, where: the state determining module 901 is further configured to monitor, according to the state monitoring unit 801, the running state information of the obtained kernel driver, It is determined that the kernel driver is in a used state.

The condition determining module 902 is further configured to: when the state determining module 901 determines that the kernel driver is in a used state, determine that the reentrant resource satisfies a scheduling condition.

As an optional implementation manner, the status monitoring unit 801 is further configured to monitor running status information of the FPGA service program.

The condition determining unit 802 is further configured to determine, according to the running state information of the FPGA service program that is monitored by the state monitoring unit 801, that the FPGA service program is in a sleep state or an unused state.

The reentrant resource scheduling device may further include: The resource deleting unit 805 is configured to delete the FPGA service program that the condition determining unit 802 determines to be in a sleep state or an unused state.

As an optional implementation manner, the status monitoring unit 801 may be as shown in FIG. 11, and further includes:

The status request sending module 1101 is configured to send status request information to the reentrant resource. The operating state receiving module 1102 is configured to receive running state information returned by the re-entrant resource according to the state request information sent by the state request sending module 1101, where the running state information includes a memory occupied by a user program in a runnable state. Whether the usage reaches a preset threshold or whether the kernel driver is in use.

Further, the status request sending module 1101 is further configured to send status request information to the FPGA service program.

The operating state receiving module 1102 is further configured to receive the running state information returned by the FPGA service program according to the state request information sent by the state request sending module 1101, where the running state information includes whether the FPGA service program is in sleep. Status or unused status.

Further optionally, the status request information includes an indication requesting the following information: a version identifier, an information type, an information length, a transmission identifier, and a target resource type.

The running status information includes a version identifier, an information type, an information length, a transmission identifier, and a target resource type.

As an optional implementation manner, the resource storage unit 803 may be as shown in FIG. 12, and further includes:

The statistical identifier obtaining module 1201 is configured to obtain a statistical identifier when the condition determining module 902 determines that the memory usage occupied by the user program in the operable state reaches a preset threshold, where the threshold is obtained.

The resource storage module 1202 is configured to store, according to the statistical identifier acquired by the statistical identifier obtaining module 1201, the reentrant resource that meets the scheduling condition into the scheduling buffer.

Optionally, the statistic identifier obtaining module 1201 is further configured to: when the condition determining module 902 determines that the kernel driver is in a used state, obtain a statistic identifier, where the statistic identifier is used to indicate the kernel driver. The occupied memory usage reaches the preset threshold. The resource storage module 1202 is further configured to store, according to the statistical identifier acquired by the statistical identifier obtaining module 1201, the reentrant resource that meets the scheduling condition into the scheduling buffer.

Further, the statistic identifier obtaining module 1201 is further configured to obtain the statistic information according to the running status information of the reentrant resource, where the statistic information includes a version identifier, an information type, an information length, a transmission identifier, and a target resource. Type and statistical ID.

As an optional implementation manner, the resource deleting unit 805 may be as shown in FIG. 13, and further includes:

The statistical identifier obtaining module 1301 is configured to: when the condition determining unit 802 determines that the FPGA service program is in a sleep state or an unused state, the statistical identifier is used to indicate that the FPGA service program is in a sleep state or Unused state.

The resource deletion module 1302 is configured to delete, according to the statistical identifier acquired by the statistical identifier obtaining module 1301, the FPGA service program that is determined to be in a sleep state or not in use.

Further, the statistical identifier obtaining module 1301 is further configured to obtain statistical information according to the running state information of the FPGA service program, where the statistical information includes a version identifier, an information type, an information length, a transmission identifier, and a target resource type. And statistical identification.

As an optional implementation manner, the resource scheduling unit 804 is further configured to acquire a scheduling identifier when the re-entrant resource stored in the scheduling buffer is scheduled to be sent to the FPGA by using the resource storage unit. The scheduling identifier is configured to indicate that the reentrant resource stored in the scheduling buffer has been successfully scheduled into the FPGA.

Further, the resource scheduling unit 804 is further configured to: when the resource storage unit schedules the reentrant resource stored in the scheduling buffer to the FPGA, acquire scheduling state information, where the scheduling is performed. Status information includes version identification, information type, information length, transmission identification, target resource type, and scheduling identification.

In the re-entrant resource scheduling device shown in FIG. 8 to FIG. 13 , the condition determining unit 802 monitors the obtained operating state information of the reentrant resource through the state monitoring unit 801, and determines that the reentrant resource satisfies the scheduling condition, the resource. The storage unit 803 stores the reentrant resource in a scheduling buffer, and the resource scheduling unit 804 schedules the reentrant resource stored in the scheduling buffer into the FPGA, so that the reentrant resource becomes The FPGA service program can automatically implement the running state information of the reentrant resources obtained according to the monitoring, and schedule the reentrant resources satisfying the scheduling condition into the FPGA, and improve Work efficiency. Referring to FIG. 14, FIG. 14 is a schematic structural diagram of a reentrant resource scheduling system according to an embodiment of the present invention, which is used to perform a reentrant resource scheduling method provided by an embodiment of the present invention. As shown in FIG. 14, the reentrant resource scheduling system includes: at least one processor 1401, such as a CPU, at least one FPGA 1405, a memory 1404, a scheduling buffer 1403, and at least one communication bus 1402. Communication bus 1402 is used to implement connection communication between these components. The memory 1404 may include a high speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory. The memory 1404 can optionally include at least one storage device located remotely from the aforementioned processor 1401. The scheduling buffer 1403 may be located in a hardware storage medium such as an EEPROM, or may be located in the memory 1404. The memory 1404 stores a set of program codes, and the processor 1401 calls the program code stored in the memory 1404 to perform the following operations:

Monitor operational status information of reentrant resources;

The reentrant resources stored in the scheduling buffer 1403 are scheduled into the FPGA 1405 to make the reentrant resources an FPGA service program.

In an optional embodiment, the reentrant resource includes a user program, and the processor 1401 calls the program code stored in the memory 1404 to determine, according to the running status information of the reentrant resource, that the reentrant resource meets a scheduling condition. Specifically:

Further, the processor 1401 calls the program code stored in the memory 1404. When the user program is in an executable state, determining that the reentrant resource meets the scheduling condition is specifically:

Determining the user program in a runnable state when the user program is in a runnable state The occupied memory usage reaches the preset threshold. Determining that the reentrant resource satisfies a scheduling condition.

In an optional embodiment, the reentrant resource includes a kernel driver, and the processor 1401 calls the program code stored in the memory 1404 to determine that the reentrant resource satisfies the schedule according to the running state information of the reentrant resource. The conditions are as follows:

Determining that the kernel driver is in a used state according to the running state information of the kernel driver;

In an alternative embodiment, the processor 1401 may also invoke the program code stored in the memory 1404 to perform the following steps:

Monitoring operating state information of the FPGA service program;

In an alternative embodiment, the processor 1401 invokes the program code stored in the memory 1404 to monitor the running status information of the reentrant resource.

Sending status request information to the reentrant resource.

Sending status request information to the FPGA service program;

The status request information includes an indication requesting the following information: a version identifier, an information type, an information length, a transmission identifier, and a target resource type; The running status information includes a version identifier, an information type, an information length, a transmission identifier, and a target resource type.

In an alternative embodiment, the processor 1401 calls the program code stored in the memory 1404 to store the reentrant resource that meets the scheduling condition into the scheduling buffer 1403. Specifically, the statistical identifier is obtained, where the statistics are collected. The identifier is used to indicate that the memory usage occupied by the user program in the operable state reaches a preset threshold;

And according to the statistical identifier, the reentrant resource that satisfies the scheduling condition is stored in the scheduling buffer 1403.

In an alternative embodiment, the processor 1401 calls the program code stored in the memory 1404 to store the reentrant resource that meets the scheduling condition into the scheduling buffer 1403. Specifically, the processor driver is configured to: The occupied memory usage reaches a preset threshold;

The processor 1401 may also acquire the program information stored in the memory 1404 to obtain statistical information according to the running status information of the reentrant resource, where the statistical information includes a version identifier, an information type, an information length, a transmission identifier, and a target resource type. And statistical identification.

In an alternative embodiment, the processor 1401 invokes the program code stored in the memory 1404 to delete the FPGA service program that is determined to be in a sleep state or not in use.

Obtaining a statistical identifier when the FPGA service program is in a sleep state or an unused state, where the statistical identifier is used to indicate that the FPGA service program is in a sleep state or an unused state; according to the statistical identifier, deleting the determining An FPGA business program that is in a sleep state or is not in use.

The processor 1401 may also call the program code stored in the memory 1404 according to the

The running status information of the FPGA service program acquires statistical information, where the statistical information includes a version identifier, an information type, an information length, a transmission identifier, a target resource type, and a statistical identifier.

In an alternative embodiment, the processor 1401 calls the program code stored in the memory 1404 to schedule the reentrant resources stored in the scheduling buffer to the FPGA 1405, specifically: Obtaining a scheduling identifier when the reentrant resource stored in the scheduling buffer is scheduled into the FPGA 1405, the scheduling identifier is used to indicate that the reentrant resource stored in the scheduling buffer has succeeded Dispatched into the FPGA 1405.

The processor 1401 may also call the program code stored in the memory 1404 to acquire scheduling state information when the reentrant resource stored in the scheduling buffer 1403 is scheduled into the FPGA 1405, where the scheduling state information includes a version. ID, information type, message length, transport ID, target resource type, and schedule ID.

A person skilled in the art can understand that all or part of the process of implementing the above embodiment method can be completed by a computer program to instruct related hardware, and the program can be stored in a computer readable storage medium, the program When executed, the flow of an embodiment of the methods as described above may be included. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

It should be noted that, in the foregoing embodiments, the descriptions of the various embodiments are different, and the parts that are not described in detail in an embodiment may refer to related descriptions of other embodiments. In the following, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and units involved are not necessarily required by the present invention.

The above is only the preferred embodiment of the present invention, and the scope of the present invention is not limited thereto, and the equivalent changes made by the claims of the present invention are still within the scope of the present invention.

Claims

Rights request

1. A reentrant resource scheduling method, characterized in that the method is used in a reentrant resource scheduling system, and the reentrant resource scheduling system includes an operating system, a scheduling buffer and a field editable logic gate array FPGA , where the operating system is used to perform the following steps:

Monitor the running status information of reentrant resources;

Determine that the reentrant resource satisfies the scheduling condition according to the running status information of the reentrant resource;

Store the reentrant resources that meet the scheduling conditions in the scheduling buffer, wherein the format of the reentrant resources stored in the scheduling buffer is a converted format that can be recognized by the FPGA ;

The reentrant resources stored in the scheduling buffer are scheduled to the FPGA, so that the reentrant resources become FPGA business programs.

2. The method of claim 1, wherein the reentrant resources include user programs;

Determining that the reentrant resource satisfies the scheduling condition based on the running status information of the reentrant resource includes:

According to the running status information of the user program, it is determined that the user program is in a runnable state; when the user program is in a runnable state, it is determined that the reentrant resource satisfies the scheduling condition.

3. The method of claim 2, wherein when the user program is in a runnable state, determining that the reentrant resource satisfies the scheduling condition includes:

When the user program is in a runnable state, it is determined that the memory usage occupied by the user program in the runnable state reaches a preset threshold; and it is determined that the reentrant resource satisfies the scheduling condition.

4. The method of claim 1, wherein the reentrant resources include kernel drivers;

Determine that the kernel driver is in use according to the running status information of the kernel driver;

When the kernel driver is in a used state, it is determined that the reentrant resource satisfies the scheduling condition.

5. The method of claim 1, wherein the method further includes:

Monitor the running status information of the FPGA business program;

According to the running status information of the FPGA business program, determine that the FPGA business program is in a sleep state or is not in use;

Delete the FPGA business program determined to be in a sleeping state or not in use.

6. The method according to any one of claims 1 to 4, characterized in that the monitoring of the operating status information of reentrant resources includes:

Send status request information to the reentrant resource;

Receive the running status information returned by the reentrant resource according to the status request information, the operation threshold, or include whether the kernel driver is in a used state.

7. The method of claim 5, wherein the monitoring the running status information of the FPGA business program includes:

Send status request information to the FPGA business program;

Receive running status information returned by the FPGA business program according to the status request information, where the running status information includes whether the FPGA business program is in a sleep state or an unused state.

8. The method of claim 6, wherein the status request information includes an instruction to request the following information: version identification, information type, information length, transmission identification, and target resource type; the running status information includes:

Version ID, message type, message length, transport ID, and target resource type.

9. The method of claim 3, wherein storing the reentrant resources that satisfy scheduling conditions into the scheduling buffer includes: obtaining a statistical identifier, wherein the statistical identifier is used to indicate The memory usage occupied by the user program in the runnable state reaches the preset threshold;

According to the statistical identification, the reentrant resources that meet the scheduling conditions are stored in the scheduling buffer.

10. The method of claim 4, wherein storing the reentrant resources that satisfy scheduling conditions into the scheduling buffer includes: indicating memory usage occupied by the kernel driver. The amount reaches the preset threshold;

11. The method according to any one of claims 9 or 10, characterized in that the obtaining statistical identification includes:

Statistical information is obtained according to the running status information of the reentrant resource, and the statistical information includes a version identifier, information type, information length, transmission identifier, target resource type, and statistical identifier.

12. The method of claim 5, wherein deleting the FPGA business program determined to be in a sleep state or an unused state includes:

Obtain statistical identification when the FPGA business program is in a sleeping state or is not in use, and the statistical identification is used to indicate that the FPGA business program is in a sleeping state or in an unused state; According to the statistical identification, the FPGA service program determined to be in a sleeping state or unused state is deleted.

13. The method of claim 12, wherein obtaining the statistical identifier when the FPGA business program is in a sleep state or is not in use includes:

Statistical information is obtained according to the running status information of the FPGA business program. The statistical information includes version identification, information type, information length, transmission identification, target resource type and statistical identification.

14. The method of claim 1, wherein scheduling the reentrant resources stored in the scheduling buffer to the FPGA includes:

The scheduling identifier is obtained when the reentrant resources stored in the scheduling buffer are scheduled to the FPGA, and the scheduling identifier is used to indicate that the reentrant resources stored in the scheduling buffer have been successful. scheduled into the FPGA.

15. The method of claim 14, wherein obtaining the scheduling identifier when scheduling the reentrant resources stored in the scheduling buffer to the FPGA includes:

Scheduling status information is obtained when the reentrant resources stored in the scheduling buffer are scheduled to the FPGA. The scheduling status information includes version identification, information type, information length, transmission identification, target resource type and scheduling. logo.

16. A computer storage medium, characterized in that the computer storage medium stores a program, and when executed, the program includes the steps described in any one of claims 1 to 15.

17. A reentrant resource scheduling device, characterized in that the reentrant resource scheduling device includes:

Status monitoring unit, used to monitor the running status information of reentrant resources;

A condition determination unit, configured to determine that the reentrant resource satisfies the scheduling condition based on the operating status information of the reentrant resource monitored by the status monitoring unit;

Resource storage unit, used to determine the reentrant resources that meet the scheduling conditions by the condition determination unit. Store in the scheduling buffer, wherein the format of the reentrant resources stored in the scheduling buffer is a converted format that can be recognized by the FPGA;

A resource scheduling unit is configured to schedule the reentrant resources stored in the scheduling buffer by the resource storage unit into the FPGA, so that the reentrant resources become FPGA business programs.

18. The reentrant resource scheduling device according to claim 17, wherein the reentrant resource includes a user program;

The condition determination unit includes:

A status determination module, configured to determine that the user program is in a runnable state based on the running status information of the user program monitored by the status monitoring unit;

A condition determination module, configured to determine that the reentrant resource satisfies the scheduling condition when the status determination module determines that the user program is in a runnable state.

19. The reentrant resource scheduling device according to claim 18, characterized in that the condition determination module includes:

Memory determination submodule, used when the status determination module determines that the user program is at a executable value;

A condition determination submodule, configured to determine that the reentrant resource satisfies the scheduling condition when the memory determination submodule determines that the memory usage occupied by the user program in the runnable state reaches a preset threshold.

20. The reentrant resource scheduling device according to claim 17, wherein the reentrant resource includes a kernel driver;

The status determination module is also used to determine that the kernel driver is in a used state based on the running status information of the kernel driver monitored by the status monitoring unit;

The condition determination module is also configured to determine that the reentrant resource satisfies the scheduling condition when the status determination module determines that the kernel driver is in a used state.

21. The reentrant resource scheduling device according to claim 17, characterized in that the status monitoring unit is also used to:

Monitor the running status information of the FPGA business program;

The condition determination unit is also configured to determine that the FPGA business program is in a sleep state or an unused state based on the running status information of the FPGA business program monitored by the status monitoring unit;

The reentrant resource scheduling device also includes:

A resource deletion unit is used to delete the FPGA business program that the condition determination unit determines is in a sleep state or an unused state.

22. The reentrant resource scheduling device according to any one of claims 17 to 20, characterized in that the status monitoring unit includes:

a status request sending module, used to send status request information to the reentrant resource; a running status receiving module, used to receive the running status information returned by the reentrant resource according to the status request information sent by the status request sending module , the running status information includes whether the memory usage occupied by the user program in the runnable state reaches a preset threshold, or includes whether the kernel driver is in use.

23. The reentrant resource scheduling device according to claim 21, characterized in that the status request sending module is also used to:

Send status request information to the FPGA business program;

The running status receiving module is also used to receive the running status information returned by the FPGA business program according to the status request information sent by the status request sending module. The running status information includes whether the FPGA business program is in a sleep state or Not in use.

24. The reentrant resource scheduling device according to claim 22, wherein the status request information includes an instruction to request the following information: version identification, information type, information length, transmission identification and target resource type;

The running status information includes: Version ID, message type, message length, transport ID, and target resource type.

25. The reentrant resource scheduling device according to claim 19, wherein the resource storage unit includes:

A statistical identification acquisition module, configured to obtain a statistical identification when the condition determination module determines that the memory usage occupied by the user program in a runnable state reaches a preset threshold, wherein the statistical identification is used in the resource storage module, and is used to obtain the statistical identification according to the The statistical identification obtained by the statistical identification acquisition module stores the reentrant resources that meet the scheduling conditions in the scheduling buffer.

26. The reentrant resource scheduling device according to claim 20, wherein the statistical identification acquisition module is further used to: identify, wherein the statistical identification is used to indicate the memory usage occupied by the kernel driver. reaches the preset threshold;

The resource storage module is also configured to store the reentrant resources that meet the scheduling conditions into the scheduling buffer according to the statistical identification obtained by the statistical identification obtaining module.

27. The reentrant resource scheduling device according to any one of claims 25 or 26, characterized in that, the statistical identification acquisition module is also used to:

28. The method of claim 21, wherein the resource deletion unit includes: a statistical identification acquisition module, used when the condition determination unit determines that the FPGA business program is in a sleep state or is not in use. Obtain a statistical identifier, which is used to indicate that the FPGA business program is in a sleep state or is not in use;

A resource deletion module, configured to delete the FPGA business program determined to be in a sleep state or unused state according to the statistical identifier obtained by the statistical identifier acquisition module.

29. The method according to claim 28, characterized in that the statistical identification acquisition module is also used to:

30. The reentrant resource scheduling device according to claim 17, characterized in that the resource scheduling unit is also used to:

The scheduling identifier is obtained when the reentrant resources stored in the scheduling buffer by the resource storage unit are scheduled to the FPGA. The scheduling identifier is used to indicate the reentrant resources stored in the scheduling buffer. Reentrant resources have been successfully scheduled into the FPGA.

31. The reentrant resource scheduling device according to claim 30, characterized in that the resource scheduling unit is also used to:

Scheduling status information is obtained when the resource storage unit schedules the reentrant resources stored in the scheduling buffer into the FPGA. The scheduling status information includes version identification, information type, information length, transmission identification, Target resource type and scheduling ID.

32. A reentrant resource scheduling system, characterized in that the reentrant resource scheduling system includes an FPGA, a memory, a scheduling buffer and a processor, wherein a set of program codes is stored in the memory, and the The processor is used to call the program code stored in the memory to perform the following operations: monitor the running status information of the reentrant resources;