Swap partition reservation method, device, storage medium and program product

Abstract

The application provides a swap exchange partition reservation method, device, storage medium and program product, belongs to the technical field of computers, and comprises the following steps: a new swap partition is used to save a hibernation image, and the swap partition is set in a resume field of system kernel start parameters; a major and minor device number of device description is converted; a total swap space size is acquired; whether the swap partition is hibernation special is identified; if the swap partition is hibernation special, the total swap exchange memory size and the available remaining swap exchange memory size are not counted; swap partition information special for hibernation is viewed through swapon; memory pages of the exchange space are acquired through calling get_swap_pages; and the pages of the hibernation special swap exchange space are directly applied through dev_t. The application solves the problem of insufficient swap partition when the system hibernates.

Description

Technical Field

[0001] This invention relates to a swap partition, and more particularly to a method, device, storage medium, and program product for reserving a swap partition for system hibernation, belonging to the field of computer technology. Background Technology

[0002] With the increasing energy efficiency requirements of embedded devices and PC hosts, Linux hibernation technology has become a core solution for optimizing system low power consumption. S4 hibernation achieves complete power loss by saving the memory state to disk, and the Swap partition provides efficient persistence as a data storage medium.

[0003] The S4 hibernation mechanism utilizes the Swap partition as the storage medium for the hibernation image. Combined with virtual memory management and system state preservation technologies, it achieves efficient and reliable system hibernation and wake-up. During Linux S4 hibernation, the system compresses memory data, kernel state, and device configurations and writes them to the Swap partition, achieving complete power-off hibernation. Therefore, as a critical data storage medium for S4 hibernation, sufficient Swap space must be ensured for S4 hibernation to proceed correctly.

[0004] Insufficient swap space will cause hibernation to fail. By default, the swap size is usually 1-2 times the physical memory, but the actual demand may vary depending on the system load. During actual system operation, as various applications continuously request and consume memory, the swap space will also be consumed. Especially after the system has been running for a long time, the consumption of swap space by various processes in the system will further aggravate the swap pressure, making it more likely that the system will fail during the hibernation phase because it cannot store a complete image due to insufficient swap space.

[0005] Therefore, ensuring that Swap has enough space during the hibernation and wake-up process to guarantee that the system can hibernate normally has become an urgent problem to be solved.

[0006] During system operation, many operations consume swap space, such as: running memory-intensive applications (e.g., databases); certain processes (e.g., zombie processes) continuously occupying swap space; memory-leaking programs failing to release resources before hibernation, causing the system to run for extended periods; and so on. All of these consume swap space, exacerbating swap pressure. Furthermore, in current technology, the swap space consumed by the system and the swap space used to save the image during hibernation are often shared, without differentiation or isolation. The only solution is to allocate a larger swap space. However, excessively large swap space can lead to wasted disk space, especially when the system rarely uses swap; moreover, excessive swap space may come at the cost of performance degradation and hardware wear and tear. Currently, there are no other methods specifically designed to prevent insufficient swap space in Linux systems, thus avoiding hibernation failures.

[0007] Chinese invention patent "A Method for Using Memory (CN200910135339.1)" provides a method for using memory, applicable to a memory system. When main memory is insufficient, a preset swap space is provided to store compressed data. When the swap space is about to become insufficient, a new swap space is added to store compressed data. This patent is mainly applicable to memory systems, where new swap partitions can be continuously added to store compressed data when the system uses memory. However, it does not involve Linux systems and is unrelated to the situation of insufficient swap partitions during hibernation and wake-up.

[0008] Chinese invention patent "A Fast Startup Method for Linux Operating System Hibernation Based on NVDIMM (CN201710795980.2)" discloses a fast startup method for Linux operating system hibernation based on NVDIMM. This method uses NVDIMM media to save the hibernation image, which would normally be saved to the SWAP disk partition, onto the NVDIMM media. This patent primarily implements a method for changing the SWAP partition storage medium for fast system hibernation startup. However, it does not address the situation where the swap partition is insufficient during system hibernation.

[0009] A foreign invention patent, "METHOD OF ACCESSING A MEMORY (US12482428)," provides a method for accessing memory in a memory system with main memory and auxiliary memory. The main memory includes a compressed data area with a predefined SWAP file region. When the main memory is insufficient for data buffering, the auxiliary storage can be used for data exchange. This patent primarily implements the use of both main and auxiliary memory for swapping, addressing the issue of auxiliary storage as a backup when main memory is insufficient. It does not involve the hibernation / wake-up mechanism of Linux systems. Summary of the Invention

[0010] To address the aforementioned problems, this invention provides a method, device, storage medium, and program product for reserving swap partitions, in order to solve the problem of insufficient swap partitions during hibernation.

[0011] To achieve the above objectives, the technical solution of the present invention is: a method for reserving swap partitions for system hibernation, comprising the following steps:

[0012] When installing the system and partitioning the disk in S001, a new swap partition is added to store the hibernation image. Then, the swap partition is set to be used in the system kernel boot parameter resume field. The kernel boot parameter resume is set to the form resume= / dev / sdx or resume=UUID=xxxxxx.

[0013] During the S002 system kernel startup phase, the kernel identifies the kernel startup parameter resume field. The kernel uses this field to convert resume= / dev / sdx or resume=UUID=xxxxxx into the major and minor device number dev_t of the device description, so that the corresponding swap partition information can be obtained later through dev_t.

[0014] S003 traverses the swap partition. If the swap partition is dedicated to hibernation, it is not included in the total available swap memory size and the available remaining swap memory size. If the swap partition is not dedicated to hibernation, it is included in the total available swap memory size and the available remaining swap memory size.

[0015] S004 When various applications of the system request swap memory, they obtain the memory pages of the swap space by calling get_swap_pages. When calling get_swap_pages, all available swap spaces are traversed. If it is dedicated to hibernation, it is filtered directly; if it is not dedicated to hibernation and the swap space is in an available state, memory pages are requested from this swap space.

[0016] S005 When the system performs a hibernation operation, pages of the dedicated swap space for hibernation are directly requested through dev_t in step S002.

[0017] Furthermore, in steps S003 and S004, the determination of whether it is a dedicated swap space for hibernation is made by comparing the bdev address and start_block address of the dev_t partition dedicated to hibernation obtained in step S002.

[0018] Furthermore, determining whether it is a dedicated swap space for hibernation includes the following steps:

[0019] When S1 iterates through all swap partitions, it will pass the swap_info of the swap partition. First, it will determine whether the partition is valid for use. If not, it will return false; if it is valid, it will proceed to the next step.

[0020] S2 retrieves the device descriptor dev_t for the hibernation-dedicated swap partition. If the retrieval fails, it returns false directly; if the retrieval is successful, it proceeds to the next step.

[0021] S3 retrieves the red-black tree node rb_node associated with the found swap partition, and obtains the extended information swap_extent of the swap partition through rb_node;

[0022] S4 obtains the block device descriptor bdev of the swap partition through the red-black tree node rb_node; and obtains the starting block address start_block of the partition through the partition extension information swap_extent.

[0023] S5 obtains the block device descriptor bdev and the starting block address start_block of the hibernation partition through the device descriptor dev_t of the hibernation partition.

[0024] S6. Compare the block device descriptor bdev of the swap partition and the block device descriptor bdev of the hibernation partition. If they are different, the swap partition is not a hibernation partition. If they are the same, proceed to the next step.

[0025] S7 continues to compare the starting block address start_block of the swap partition and the starting block address start_block of the hibernation partition. If they are not equal, then the partition is not a hibernation partition either.

[0026] S8 If the block device descriptor bdev and the starting block address start_block are the same, it means that the traversed partition is the hibernation partition, and return true.

[0027] Furthermore, step S003 specifically includes the following steps:

[0028] Sⅰ defines S4_totalswap and S4_freeswap to record the total size and available size of the hibernation-dedicated swap partition;

[0029] Sⅱ begins traversing all swap partitions;

[0030] Sⅲ Determines whether the swap partition is dedicated to hibernation; if it is dedicated to hibernation, it records S4_totalswap and S4_freeswap; if it is not dedicated to hibernation, it proceeds to the next step to obtain the total freeswap and total swap.

[0031] Sⅳ completes the traversal of all swap partitions;

[0032] Sⅴ obtains the total size of the system's available swap partition by subtracting S4_totalswap from the total totalswap; and obtains the remaining size of the system's available swap partition by subtracting S4_freeswap from the total freeswap.

[0033] Furthermore, step S004 specifically includes the following steps:

[0034] After the SⅠ application starts calling get_swap_pages to request swap pages, it first determines whether it is a sleep operation;

[0035] If SⅡ is not a hibernation operation, then iterate through all available swap partitions;

[0036] SⅢ checks if the swap partition is for hibernation only and filters it out, then continues to traverse the next usable swap partition;

[0037] If SⅣ determines that the swap partition is not for hibernation, then it can request to use the swap partition pages.

[0038] Furthermore, step S005 specifically includes the following steps:

[0039] The Sa system application invokes the hibernation operation and then requests a swap page;

[0040] After Sb determines that the operation is hibernation, it directly and normally allocates dedicated swap memory pages for hibernation.

[0041] After Sc allocates dedicated swap memory pages for hibernation, it saves the hibernation image to this swap partition, thus completing the hibernation operation.

[0042] A computer device includes a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the method described above.

[0043] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0044] A computer program product includes a computer program that, when executed by a processor, implements the steps of the method described above.

[0045] The beneficial effects of the swap partition reservation method for system hibernation according to the present invention are as follows:

[0046] In terms of user experience: This invention provides, for the first time, a method for reserving a swap partition specifically for system hibernation in Linux systems, addressing the problem that Linux systems often fail to hibernate due to insufficient swap space. The method of this invention effectively improves the efficiency of swap partition usage by dividing swap into different partitions according to their purpose. These swap partitions for different purposes are independent and do not interfere with each other, thus ensuring a more stable and reliable user experience.

[0047] From the developer's perspective: The method of this invention isolates the swap memory used by the system during normal operation from the swap memory used during hibernation, so they do not interfere with each other. Therefore, if various swap memory-related problems occur in the system, they can be troubleshooted independently, which facilitates debugging and helps to find the cause of the problem. Attached Figure Description

[0048] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0049] Figure 1 This is a detailed flowchart of the present invention;

[0050] Figure 2 This is a flowchart of step S003 of the present invention;

[0051] Figure 3 This is a flowchart of steps S004-S005 of the present invention;

[0052] Figure 4 The flowchart for determining whether the swap partition is a dedicated swap partition for hibernation is shown in this invention. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0054] Definitions of abbreviations and key terms:

[0055] SWAP (swap space) is a region on a disk, which can be a separate partition or a file. The operating system expands available memory resources by temporarily storing inactive data from physical memory to a swap partition or swap file on the hard drive.

[0056] S4 (Suspend to Disk), also known as Hibernate mode, involves saving the computer's current running state (including all data in memory and system state) to the hard drive before completely shutting down the power. When the computer restarts, it reads this data from the hard drive and restores the running state it had before entering S4.

[0057] To address the shortcomings of existing technologies, this invention provides, for the first time, a method for reserving a dedicated swap partition for system hibernation in Linux systems. This method solves the problem of Linux systems being unable to hibernate due to insufficient swap space. By allocating a dedicated swap partition or file and isolating it from swap partitions or files used by system processes for memory allocation, this method ensures that the dedicated hibernation swap partition or file is not used by other processes consuming memory, thereby guaranteeing sufficient space for this dedicated S4 hibernation swap partition or file and maximizing the ability for hibernation to proceed normally. The method of this invention will be described in detail below with reference to the accompanying drawings.

[0058] Example 1

[0059] Combination Figure 1 As shown, the present invention provides a method for reserving swap partitions for system hibernation, comprising the following steps:

[0060] When installing the S001 system, the disk will be partitioned. At this time, a new swap partition needs to be added. This partition is specifically used to store the hibernation image, and the system kernel boot parameter resume field is set to use this partition. The kernel boot parameter resume can be in the form of resume= / dev / sdx or resume=UUID=xxxxxx.

[0061] During the S002 system kernel boot phase, the kernel accurately identifies the kernel boot parameter resume field. This field is specifically used for the swap partition during hibernation. The kernel uses this field to convert resume= / dev / sdx or resume=UUID=xxxxxx into the major and minor device number dev_t of the device description, so that the corresponding swap partition information can be obtained later through dev_t.

[0062] S003 iterates through the swap partition. If the swap partition is dedicated to hibernation, it is excluded from the total available swap memory size and the available remaining swap memory size; this excludes swap memory dedicated to hibernation from the system's swap space. If the swap partition is not dedicated to hibernation, it is included in the total available swap memory size and the available remaining swap memory size. For example... Figure 2 As shown, the specific steps include the following:

[0063] (1) Define S4_totalswap and S4_freeswap to record the total size and available size of the hibernation-specific swap partition;

[0064] (2) Begin traversing all swap partitions;

[0065] (3) Determine whether the swap partition is dedicated to hibernation using the following method; if it is dedicated to hibernation, record it in S4_totalswap and S4_freeswap; if it is not dedicated to hibernation, proceed to the next step to obtain the total freeswap and totalswap.

[0066] (4) Complete the traversal of all swap partitions;

[0067] (5) The total size of the system's available swap partition can be obtained by subtracting S4_totalswap from the total totalswap; the remaining size of the system's available swap partition can be obtained by subtracting S4_freeswap from the total freeswap.

[0068] (6) At this point, the swap space that can be viewed through system commands will no longer include the size of the dedicated swap space for hibernation.

[0069] It should be noted that step S003 is primarily used to check the remaining size of the system's available swap partition, providing feedback to the system user on how much swap space is available. This step is a parallel operation to the subsequent steps of normal application swap space allocation and hibernation swap space allocation; they are not sequential. After entering the system, you can enter a command to check the remaining swap size, or you can choose not to check it; this will not affect the subsequent hibernation and application swap usage.

[0070] In addition, the system provides users with an interface to obtain the current available swap size. Specifically, after entering the system, when using the `free` command to view memory information, the swap space dedicated to hibernation will not be visible; only the `swapon` command will show the information about the swap partition dedicated to hibernation. It is important to emphasize that the `free` command can view normally used swap information but will not show hibernation-specific swap information; the `swapon` command can view hibernation-specific swap information.

[0071] S004 When various applications of the system request swap memory, they will obtain the memory pages of the swap space by calling get_swap_pages. get_swap_pages will iterate through all available swap spaces. If it is dedicated to hibernation, it will be filtered directly; if it is not dedicated to hibernation and the swap space is available, then memory pages will be requested from that swap space. For example... Figure 3 As shown, the specific steps include the following:

[0072] (1) After the application starts calling get_swap_pages to request swap pages, it first determines whether it is a sleep operation;

[0073] (2) If it is not a hibernation operation, then traverse all available swap partitions;

[0074] (3) Through the following Figure 4 The method shown determines whether the swap partition is for hibernation purposes and filters it out, then continues to traverse the next usable swap partition.

[0075] (4) If the swap partition is determined to be not for hibernation purposes by the method in the previous step, then the swap partition pages can be used.

[0076] The above steps also require traversing all swap spaces when calling get_swap_pages, and during the traversal process, the hibernation-specific partition can be filtered out.

[0077] S005 When the system performs a hibernation operation, it also needs to allocate pages to save the hibernation image, just like in S004. However, when allocating pages, it no longer needs to traverse all swap spaces. Instead, it directly allocates pages for the dedicated hibernation swap space through the dev_t function in step S002. Figure 3 As shown, the specific steps include the following:

[0078] (1) When a system application calls the hibernation operation, it also needs to request a swap page during hibernation;

[0079] (2) After determining that the operation is hibernation, directly and normally request the dedicated swap memory pages for hibernation;

[0080] (3) After obtaining the dedicated swap memory pages for hibernation, the hibernation image can be saved to the swap partition to complete the hibernation operation.

[0081] It should be noted that in this step, because there is only one dedicated swap partition for hibernation, there is no need to traverse it; it can be requested and used directly. Generally, there may be multiple non-hibernation dedicated swap partitions, while there is only one dedicated hibernation swap partition. When an application requests swap, even if there is no dedicated hibernation swap partition, it still needs to traverse all swap partitions, specifically obtaining the appropriate swap partition based on the requested size and type. However, since there is only one dedicated hibernation swap partition, this distinction has already been made in step S002, so there is no need to traverse it again.

[0082] After the S006 system has been running for a long time or after executing an application that consumes a lot of swap space, the dedicated swap space for hibernation will not be used, and hibernation can be performed normally at this time.

[0083] The fact that the dedicated swap space for hibernation will not be used means that when an application requests to use get_swap_pages, it needs to traverse all swap partitions. During the traversal, the dedicated swap partition for hibernation can be filtered out, ensuring that the dedicated swap space for hibernation will not be used.

[0084] The above steps determine whether the swap space is a hibernation space using the following method. Specifically, when traversing all swap partitions, it is necessary to identify whether the partition is dedicated to hibernation. By comparing the bdev address and start_block address of the partition with the dev_t partition dedicated to hibernation obtained in S002, it can be determined whether the swap partition is a dedicated hibernation swap partition. (Further details can be found in the original text.) Figure 4 As shown, it includes the following steps:

[0085] (1) When traversing all swap partitions, the swap_info of the swap partition will be passed. First, it is determined whether the partition is valid for use. If not, return false; if valid, proceed to the next step.

[0086] (2) Obtain the device descriptor of the hibernation-dedicated swap partition. If the acquisition fails, return false directly. Normally, as long as the kernel boot parameter resume sets the hibernation-dedicated swap partition, the dev_t device descriptor of the partition can be obtained normally. If the acquisition is successful, proceed to the next step.

[0087] (3) Obtain the red-black tree node rb_node associated with the swap partition being searched. The extended information swap_extent of the swap partition can be obtained through rb_node.

[0088] (4) Obtain the block device descriptor bdev of the swap partition through the red-black tree node rb_node; obtain the starting block address start_block of the partition through the partition extension information swap_extent;

[0089] (5) The block device descriptor bdev and the starting block address start_block of the hibernation partition can be obtained through the device descriptor dev_t of the hibernation partition.

[0090] (6) Compare the block device descriptor bdev of the swap partition and the block device descriptor bdev of the hibernation partition. If they are different, the swap partition is not a hibernation partition. If they are the same, proceed to the next step.

[0091] (7) Continue to compare the starting block address start_block of the swap partition and the starting block address start_block of the hibernation partition. If they are not equal, then the partition is not a hibernation partition.

[0092] (8) If the block device descriptor bdev and the starting block address start_block are the same, it means that the traversed partition is the hibernation partition and return true.

[0093] Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

Claims

1. A method for reserving swap partitions for system hibernation, characterized in that, Includes the following steps: When installing the system and partitioning the disk in S001, a new swap partition is added to store the hibernation image. Then, the swap partition is set to be used in the system kernel boot parameter resume field. The kernel boot parameter resume is set to the form resume= / dev / sdx or resume=UUID=xxxxxx. During the S002 system kernel startup phase, the kernel identifies the kernel startup parameter resume field. The kernel uses this field to convert resume= / dev / sdx or resume=UUID=xxxxxx into the major and minor device number dev_t of the device description, so that the corresponding swap partition information can be obtained later through dev_t. S003 traverses the swap partition. If the swap partition is dedicated to hibernation, it is not included in the total available swap memory size and the available remaining swap memory size. If the swap partition is not dedicated to hibernation, it is included in the total available swap memory size and the available remaining swap memory size. S004 When various applications of the system request swap memory, they obtain the memory pages of the swap space by calling get_swap_pages. When calling get_swap_pages, all available swap spaces are traversed. If it is dedicated to hibernation, it is filtered directly; if it is not dedicated to hibernation and the swap space is in an available state, memory pages are requested from this swap space. S005 When the system performs a hibernation operation, it directly requests pages of the dedicated swap space for hibernation through dev_t in step S002; In steps S003 and S004, it is determined whether the swap space is dedicated to hibernation by comparing the bdev address and start_block address of the dev_t swap partition dedicated to hibernation obtained in step S002. Determining whether a space is dedicated to hibernation swap space involves the following steps: When S1 iterates through all swap partitions, it will pass the swap_info of the swap partition. First, it will determine whether the partition is valid for use. If not, it will return false; if it is valid, it will proceed to the next step. S2 retrieves the device descriptor dev_t for the hibernation-dedicated swap partition. If the retrieval fails, it returns false directly; if the retrieval is successful, it proceeds to the next step. S3 retrieves the red-black tree node rb_node associated with the found swap partition, and obtains the extended information swap_extent of the swap partition through rb_node; S4 obtains the block device descriptor bdev of the swap partition through the red-black tree node rb_node; The starting block address of the partition is obtained by using the partition extension information swap_extent; S5 obtains the block device descriptor bdev and the starting block address start_block of the hibernation partition through the device descriptor dev_t of the hibernation partition. S6. Compare the block device descriptor bdev of the swap partition with the block device descriptor bdev of the hibernation partition. If they are different, the swap partition is not a hibernation partition. If they are the same, proceed to the next step. S7 continues to compare the starting block address start_block of the swap partition with the starting block address start_block of the hibernation partition. If they are not equal, then the partition is not a hibernation partition. S8 If the block device descriptor bdev and the starting block address start_block are the same, it means that the traversed partition is the hibernation partition, and return true.

2. The method according to claim 1, characterized in that, Step S003 specifically includes the following steps: Sⅰ defines S4_totalswap and S4_freeswap to record the total size and available size of the hibernation-dedicated swap partition; Sⅱ begins traversing all swap partitions; Sⅲ Determines whether the swap partition is dedicated to hibernation; if it is dedicated to hibernation, it records S4_totalswap and S4_freeswap; if it is not dedicated to hibernation, it proceeds to the next step to obtain the total freeswap and total swap. Sⅳ completes the traversal of all swap partitions; Sⅴ obtains the total size of the system's available swap partition by subtracting S4_totalswap from the total totalswap. The remaining size of the system's available swap partition is obtained by subtracting S4_freeswap from the total freeswap.

3. The method according to claim 1, characterized in that, Step S004 specifically includes the following steps: After the SⅠ application starts calling get_swap_pages to request swap pages, it first determines whether it is a sleep operation; If SⅡ is not a hibernation operation, then iterate through all available swap partitions; SⅢ checks if the swap partition is for hibernation only and filters it out, then continues to traverse the next usable swap partition; If SⅣ determines that the swap partition is not for hibernation, then it can request to use the swap partition pages.

4. The method according to claim 1, characterized in that, Step S005 specifically includes the following steps: The Sa system application invokes the hibernation operation and then requests a swap page; After Sb determines that the operation is hibernation, it directly and normally allocates dedicated swap memory pages for hibernation. After Sc allocates dedicated swap memory pages for hibernation, it saves the hibernation image to this swap partition, thus completing the hibernation operation.

5. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1-4.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method described in any one of claims 1-4.

7. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method described in any one of claims 1-4.

Images