Page table reading method, device and equipment and computer storage medium

文档序号：1242433 发布日期：2020-08-18 浏览：12次中文

阅读说明：本技术 页表的读取方法、装置、设备及计算机存储介质 (Page table reading method, device and equipment and computer storage medium ) 是由袁典涛何建彬邱雪松于 2020-05-14 设计创作，主要内容包括：本申请提供了一种页表的读取方法、装置、设备及计算机存储介质,通过在缓存器中的页表中成功查询得到目标虚拟地址对应的物理地址时,获取所述缓存器中的当前未使用的页表的数量；判断所述缓存器中的当前未使用的页表的数量,是否足以支持一次获取页表行为所需的时间内的虚拟地址到物理地址转换的查询操作；若判断出所述缓存器中的当前未使用的页表的数量,足以支持一次获取页表行为所需的时间,则从存储器中读取预设数量的页表。以达到隐藏从存储器中读取页表的延时,从而提高芯片的处理性能的目的。(The application provides a page table reading method, a page table reading device, page table reading equipment and a computer storage medium, wherein when a physical address corresponding to a target virtual address is obtained through successful query in a page table in a buffer, the number of the currently unused page tables in the buffer is obtained; determining whether the number of currently unused page tables in the cache is sufficient to support a lookup operation for virtual to physical address translation within a time required to obtain page table behavior; if the number of the currently unused page tables in the register is judged to be enough to support the time required for one-time page table acquiring, reading a preset number of page tables from a memory. The purpose of hiding the time delay of reading the page table from the memory and improving the processing performance of the chip is achieved.)

1. A method for reading a page table, comprising:

if the physical address corresponding to the target virtual address is obtained through successful query in a page table of the cache, the number of the current unused page tables in the cache is obtained;

determining whether the number of currently unused page tables in the cache is sufficient to support a lookup operation for virtual to physical address translation within a time required to obtain page table behavior;

if the number of the currently unused page tables in the register is judged to be enough to support the time required for one-time page table acquiring, reading a preset number of page tables from a memory.

2. The method of claim 1, wherein determining whether the number of currently unused page tables in the cache is sufficient to support a lookup operation for a virtual to physical address translation within a time required to obtain page table behavior comprises:

judging whether the number of the current unused page tables in the buffer is equal to the number of the page tables in the time required for acquiring the preset number of page table behaviors at one time or not;

wherein: if the number of the currently unused page tables in the buffer is determined to be equal to the number of the page tables used in the time required for acquiring the preset number of page table behaviors at one time, the number of the currently unused page tables in the buffer is determined to be sufficient to support the time required for acquiring the page table behaviors at one time.

3. The reading method according to claim 1, further comprising:

obtaining the number of the cache page tables of the cache; the number of the cache page tables of the cache is the number of the cache page tables which can be supported by the cache space of the cache;

judging whether the number of the cache page tables of the cache is more than or equal to 2 times of the number of the page tables acquired at one time;

and if the number of the cache page tables of the cache is judged to be more than or equal to 2 times of the number of the page tables acquired at one time, the number of the current unused page tables in the cache is acquired.

4. The reading method according to claim 1, further comprising, after reading the preset number of page tables from the memory:

replacing the page table used in the buffer with the read page table; and the number of the replaced page tables is the same as that of the page tables obtained by reading.

5. An apparatus for reading a page table, comprising:

the device comprises an obtaining unit, a processing unit and a processing unit, wherein the obtaining unit is used for obtaining the number of current unused page tables in a buffer if a physical address corresponding to a target virtual address is obtained by successfully querying in a page table of the buffer;

the judging unit is used for judging whether the number of the current unused page tables in the buffer is enough to support the inquiry operation of converting the virtual address into the physical address within the time required by the behavior of the page table acquisition;

the reading unit is configured to read a preset number of page tables from a memory if the determining unit determines that the number of currently unused page tables in the buffer is sufficient to support the time required for obtaining the page table behavior once.

6. The reading apparatus according to claim 5, wherein the judging unit includes:

a determining subunit, configured to determine whether the number of currently unused page tables in the buffer is equal to the number of page tables used in a time period required for acquiring a preset number of page table behaviors at a time;

wherein: if the determining subunit determines that the number of the currently unused page tables in the buffer is equal to the number of the page tables used in the time required for acquiring the preset number of page table behaviors at one time, the determining unit determines that the number of the currently unused page tables in the buffer is sufficient to support the time required for acquiring the page table behaviors at one time.

7. The reading apparatus according to claim 5, further comprising:

an obtaining unit, configured to obtain the number of the cache page tables of the cache; the number of the cache page tables of the cache is the number of the cache page tables which can be supported by the cache space of the cache;

the comparison unit is used for judging whether the number of the cache page tables of the cache is more than or equal to 2 times of the number of the page tables acquired at one time;

if the comparing unit determines that the number of the page tables cached by the cache is greater than or equal to 2 times of the number of the page tables acquired at one time, the acquiring unit executes the action of acquiring the number of the currently unused page tables in the cache.

8. The reading apparatus according to claim 5, further comprising:

a replacing unit, configured to replace a page table already used in the buffer with a read page table; and the number of the replaced page tables is the same as that of the page tables obtained by reading.

9. A page table reading apparatus, comprising:

one or more processors;

a storage device having one or more programs stored thereon;

the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of any of claims 1-4.

10. A computer storage medium, having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method of any of claims 1 to 4.

Technical Field

The present application relates to the field of computer technologies, and in particular, to a method, an apparatus, a device, and a computer storage medium for reading a page table.

Background

In the design of large computing chips, there is a need to perform address virtualization management on a memory, and when a virtual address is used to perform read-write operation on the memory, the virtual address needs to be converted into a physical address.

At present, in the process of converting a virtual address into a physical address, an existing page table is usually queried in a buffer, and a physical address corresponding to the virtual address is obtained through querying according to mapping data from the virtual address to the physical address recorded in the page table. However, when the existing page table of the register is used, the page table lookup of the register fails, and a new read request is triggered to retrieve the page table to be used. The delay of sending a read request to the memory to read the page table is generally long, which causes a temporary interruption of the whole hardware pipeline, thereby affecting the processing performance of the whole chip.

Therefore, a method for hiding the delay of reading the page table from the memory is needed, so as to improve the processing performance of the chip.

Disclosure of Invention

In view of this, the present application provides a method, an apparatus, a device and a computer storage medium for reading a page table, which are used to hide a latency of reading the page table from a memory, thereby improving the processing performance of a chip.

A first aspect of the present application provides a page table reading method, including:

Optionally, the determining whether the number of currently unused page tables in the buffer is sufficient to support a lookup operation of virtual address to physical address translation within a time required for obtaining a page table behavior includes:

Optionally, the method for reading the page table further includes:

judging whether the number of the cache page tables of the cache is more than or equal to 2 times of the number of the page tables acquired at one time;

Optionally, after reading the preset number of page tables from the memory, the method further includes:

replacing the page table used in the buffer with the read page table; and the number of the replaced page tables is the same as that of the page tables obtained by reading.

A second aspect of the present application provides a page table reading apparatus, including:

Optionally, the determining unit includes:

Optionally, the reading apparatus of the page table further includes:

the comparison unit is used for judging whether the number of the cache page tables of the cache is more than or equal to 2 times of the number of the page tables acquired at one time;

Optionally, the reading apparatus of the page table further includes:

A third aspect of the present application provides a page table reading apparatus, including:

one or more processors;

a storage device having one or more programs stored thereon;

the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of any of the first aspects of the present application.

A fourth aspect of the present application provides a computer storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method according to any one of the first aspect of the present application.

According to the above scheme, in the page table reading method, the device, the equipment and the computer storage medium provided by the application, when the physical address corresponding to the target virtual address is obtained through successful query in the page table of the buffer, the number of the currently unused page tables in the buffer is obtained; determining whether the number of currently unused page tables in the cache is sufficient to support a lookup operation for virtual to physical address translation within a time required to obtain page table behavior; if the number of the current unused page tables in the buffer is judged to be enough to support the time required for acquiring the page table behavior once, a preset number of page tables are read from the memory, so that the delay of reading the page tables from the memory is hidden, and the purpose of improving the processing performance of the chip is achieved.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.

Fig. 1 is a detailed flowchart of a page table reading method according to an embodiment of the present disclosure;

FIG. 2 is a flowchart illustrating a page table reading method according to another embodiment of the present application;

FIG. 3 is a diagram illustrating a reading apparatus of a page table according to another embodiment of the present application;

FIG. 4 is a diagram illustrating a reading apparatus of a page table according to another embodiment of the present application;

fig. 5 is a schematic diagram of an apparatus for performing a page table reading method according to another embodiment of the present application.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

It should be noted that the terms "first", "second", and the like, referred to in this application, are only used for distinguishing different devices, modules or units, and are not used for limiting the order or interdependence of functions performed by these devices, modules or units, but the terms "include", or any other variation thereof are intended to cover a non-exclusive inclusion, so that a process, method, article, or apparatus that includes a series of elements includes not only those elements but also other elements that are not explicitly listed, or includes elements inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

At present, in the design process of a large-scale computing chip, address virtualization management needs to be performed on a memory, so that different use units and users of the memory can have independent address spaces. The read and write operations of the memory by using the virtual address must be converted from the virtual address to the actual physical address, which is a crucial step. This conversion process is typically handled by a specialized module, which we may simply refer to as a VM. Mapping data from virtual addresses to physical addresses is stored in memory and may be referred to as page tables, abbreviated as PTEs. The VM implementation usually includes a page table buffer for storing a page table read back from the memory, and the physical address can be obtained by looking up the page table in the page table buffer according to the virtual address.

When the VM operates, the existing page table in the cache is firstly inquired according to the currently processed virtual address, and if the inquiry is hit, the physical address can be immediately obtained; if the query fails, a read request is sent to the memory to retrieve the required page table, and in the prior art, a plurality of page tables are retrieved at one time in the read request, so that the virtual address to be processed subsequently directly hits the page table in the buffer. The virtual address to physical address translation may then be performed according to the page table in the buffer.

It can be seen that, in the existing VM implementation method, when a read request is sent to a memory read page table, no matter how many page tables are read at a time, when the last page table is used, a failure of querying the page table of the buffer occurs, and a new read request is triggered to retrieve the page table to be used. The delay of sending a read request to the memory to read the page table is generally long, which causes a temporary interruption of the whole hardware pipeline, thereby affecting the processing performance of the whole chip.

Therefore, an embodiment of the present application provides a method for reading a page table, as shown in fig. 1, which specifically includes the following steps:

s101, receiving a query instruction; the query instruction is used for searching a physical address corresponding to the target virtual address.

Specifically, if a user performs a read/write operation, the physical address corresponding to the target virtual address needs to be obtained by querying in the page table currently stored in the buffer, and therefore a query instruction is sent to the VM.

S102, responding to the query instruction, and querying a physical address corresponding to the target virtual address in a page table currently stored in the buffer.

Wherein, the buffer is a page table buffer. The page table buffer will send a page table read instruction to the memory and store the page table read back locally. Therefore, after receiving the query instruction, it is first queried in the locally stored page table whether the physical address corresponding to the target virtual address is recorded.

Specifically, if the physical address corresponding to the target virtual address is obtained by querying in the page table currently stored in the buffer, step S103 is executed; if the physical address corresponding to the target virtual address is not obtained by querying in the page table currently stored in the register, the query is failed. When the physical address corresponding to the target virtual address is not searched in the page table currently stored in the buffer, a read request is sent to the memory to retrieve the required page table.

S103, acquiring the number of the current unused page tables in the buffer.

The currently unused page table is a page table in which the lookup instruction is received in the buffer, but the page table does not match the virtual address of the lookup instruction.

Specifically, the currently unused page tables are obtained by searching all the page tables in the buffer, and the number of the currently unused page tables is counted.

S104, judging whether the number of the current unused page tables in the buffer is enough to support the inquiry operation of the virtual address-to-physical address conversion within the time required by the behavior of the page table acquisition.

It should be noted that, when the number of the currently unused page tables in the buffer is larger, it indicates that there is more time for performing the page table fetching operation.

Specifically, if the time that the unused page table in the current buffer can be used is greater than or equal to the time required for one-time page table obtaining, the number of the current unused page tables in the buffer is determined to be sufficient to support the time required for one-time page table obtaining, and step S105 is executed.

Optionally, in another embodiment of the present application, an implementation manner of step S104 includes:

whether the number of the current unused page tables in the buffer is equal to the number of the page tables used in the time required for acquiring the preset number of page table behaviors at one time is judged.

It should be noted that, in the case that the time used by each page table is the same, it may be determined whether the number of currently unused page tables in the buffer is equal to the number of page tables used in the time required to obtain the predetermined number of page table behaviors at a time.

Specifically, if it is determined that the number of currently unused page tables in the buffer is equal to the number of page tables used in the time required to acquire the predetermined number of page table behaviors at a time, the number of currently unused page tables in the buffer is indicated to be sufficient to support the time required to acquire the page table behaviors at a time.

And S105, reading a preset number of page tables from the memory.

When a preset number of page tables are read, the usable time of reading the preset number of page tables from the memory at a time must be longer than the usable time of reading the preset number of page tables from the memory at a time.

It should be noted that the size of the memory is twice the size of the preset number of page tables.

Specifically, for example, one round of the registers can store 8 page tables. When the first page table is used, the number of currently unused page tables in the buffer must be sufficient to support a lookup operation for virtual to physical address translation within the time required for the page table behavior to be obtained, so that theoretically 8 page tables of the next round can be pre-fetched from the memory during the use of the first page table. However, if the page tables are not used in the order of the page tables themselves in the actual use process, for example, the order of obtaining the page tables is page table 1, page table 2, page table 3, page table 4, etc., and the order of page table 2, page table 8, page table 4, page table 1, etc. is used in the use process. Then the probability that the prefetched page table is used, and during use, is invalid, is increased. Moreover, it can be seen that the more page tables that are currently used, the fewer unused page tables remain in the buffer, and the higher the accuracy of the page tables obtained by prefetching when the page tables are prefetched. Therefore, the timing of prefetching the page table may be, but is not limited to, a calculation using a preset formula. Wherein, the preset formula is that S is max (N/2, M); in the formula, S is the number of page tables remaining in the buffer, N is the number of page tables to be read, and M is the minimum number of page tables remaining in the current buffer, that is, the number of page tables used in the time required for acquiring the page table behavior at one time. When the number of remaining page tables reaches S, prefetching of the page tables is started.

Optionally, in another embodiment of the present application, before step S104, as shown in fig. 2, the method further includes:

s201, obtaining the number of cache page tables of the buffer.

The number of the cache page tables of the cache is the number of the cache page tables which can be supported by the cache space of the cache.

S202, judging whether the number of the cache page tables of the buffer is more than or equal to 2 times of the number of the page tables acquired at one time.

It should be noted that only when the number of the page tables cached by the buffer is greater than or equal to twice the number of the page tables obtained each time, it can be ensured that the new page table obtained back does not replace the page table that is not used yet in the buffer.

Specifically, if it is determined that the number of the cache page tables in the register is greater than or equal to 2 times the number of the page tables obtained at one time, step S203 is executed.

S203, acquiring the number of the current unused page tables in the buffer.

Specifically, the specific implementation process of obtaining the number of the currently unused page tables in the buffer may refer to the implementation process in step S103, and is not described here again.

Optionally, in another embodiment of the present application, after step S105, the method further includes:

the obtained page table is read to replace the used page table in the buffer.

The number of the replaced page tables is the same as that of the read page tables.

Specifically, the page table obtained in advance replaces the page table already used in the buffer, so that the unused page table is always stored in the buffer, and the process of reading the page table from the memory in the prior art is hidden, so that the temporary interruption in the whole hardware flow process is avoided, the processing performance of the whole chip is influenced, and the purpose of improving the processing performance of the whole chip is finally achieved.

According to the scheme, in the page table reading method provided by the application, when the physical address corresponding to the target virtual address is obtained through successful query in the page table of the buffer, the number of the current unused page tables in the buffer is obtained; then, judging whether the number of the current unused page tables in the buffer is enough to support the inquiry operation of converting the virtual address into the physical address within the time required for acquiring the page table behavior; if the number of the currently unused page tables in the register is determined to be sufficient to support the time required for one page table acquiring action, a preset number of page tables are read from the memory. The purpose of hiding the time delay of reading the page table from the memory and improving the processing performance of the chip is achieved.

Another embodiment of the present application provides a reading apparatus of a page table, as shown in fig. 3, specifically including:

an obtaining unit 301, configured to obtain the number of currently unused page tables in the buffer if the physical address corresponding to the target virtual address is obtained through successful lookup in the page table of the buffer.

A determining unit 302 is configured to determine whether the number of currently unused page tables in the buffer is sufficient to support a lookup operation for virtual address to physical address translation within a time required for obtaining page table behavior.

Optionally, in another embodiment of the present application, an implementation manner of the determining unit 302 includes:

and the judging subunit is used for judging whether the number of the current unused page tables in the buffer is equal to the number of the page tables used in the time required for acquiring the preset number of page table behaviors at one time.

Specifically, the method comprises the following steps: if the determining subunit determines that the number of currently unused page tables in the buffer is equal to the number of page tables used in the time required to obtain the predetermined number of page table behaviors at a time, the determining unit 302 determines that the number of currently unused page tables in the buffer is sufficient to support the time required to obtain the page table behaviors at a time.

For specific working processes of the units disclosed in the above embodiments of the present application, reference may be made to the contents of the corresponding method embodiments, which are not described herein again.

A reading unit 303, configured to read a preset number of page tables from the memory if the determining unit 302 determines that the number of currently unused page tables in the register is sufficient to support the time required for one page table fetching behavior.

For a specific working process of the unit disclosed in the above embodiment of the present application, reference may be made to the content of the corresponding method embodiment, as shown in fig. 1, which is not described herein again.

Optionally, in another embodiment of the present application, as shown in fig. 4, the reading apparatus of the page table further includes:

an obtaining unit 401 is configured to obtain the number of cache page tables.

The number of the cache page tables of the cache is the number of the cache page tables which can be supported by the cache space of the cache.

A comparing unit 402, configured to determine whether the number of the cache page tables is greater than or equal to 2 times the number of the one-time fetching page tables.

Specifically, if the comparing unit 402 determines that the number of the page tables cached in the cache is greater than or equal to 2 times the number of the page tables obtained at one time, the obtaining unit 401 performs an operation of obtaining the number of the currently unused page tables in the cache.

Optionally, in another embodiment of the present application, the reading apparatus of the page table further includes:

and the replacing unit is used for replacing the used page table in the buffer with the read page table.

The number of the replaced page tables is the same as that of the read page tables.

As can be seen from the above solutions, in the page table reading apparatus provided in the present application, when the physical address corresponding to the target virtual address is obtained through successful query in the page table of the buffer by the obtaining unit 301, the number of currently unused page tables in the buffer is obtained; then, the determining unit 302 is used to determine whether the number of currently unused page tables in the buffer is sufficient to support a lookup operation for converting a virtual address into a physical address within a time required for acquiring a page table behavior; if the determining unit 302 determines that the number of currently unused page tables in the register is sufficient to support the time required for one page table fetching, the reading unit 303 is utilized to read a preset number of page tables from the memory. The purpose of hiding the time delay of reading the page table from the memory and improving the processing performance of the chip is achieved.

Another embodiment of the present application provides a computer storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method of any one of the above embodiments.

Another embodiment of the present application provides an apparatus, as shown in fig. 5, including:

one or more processors 501.

A storage device 502 on which one or more programs are stored.

The one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method as in any one of the above embodiments.

In the above embodiments disclosed in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The apparatus and method embodiments described above are illustrative only, as the flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

In addition, functional modules in the embodiments of the present disclosure may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part. The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present disclosure may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a live broadcast device, or a network device) to execute all or part of the steps of the method according to the embodiments of the present disclosure. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

Those skilled in the art can make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

13页详细技术资料下载

上一篇：一种医用注射器针头装配设备

下一篇：对页表遍历的冗余进行检测的处理器

Page table reading method, device and equipment and computer storage medium

相关技术

网友询问留言