194 lines · plain
1=====================================================2Documentation for userland software suspend interface3=====================================================4 5 (C) 2006 Rafael J. Wysocki <rjw@sisk.pl>6 7First, the warnings at the beginning of swsusp.txt still apply.8 9Second, you should read the FAQ in swsusp.txt _now_ if you have not10done it already.11 12Now, to use the userland interface for software suspend you need special13utilities that will read/write the system memory snapshot from/to the14kernel. Such utilities are available, for example, from15<http://suspend.sourceforge.net>. You may want to have a look at them if you16are going to develop your own suspend/resume utilities.17 18The interface consists of a character device providing the open(),19release(), read(), and write() operations as well as several ioctl()20commands defined in include/linux/suspend_ioctls.h . The major and minor21numbers of the device are, respectively, 10 and 231, and they can22be read from /sys/class/misc/snapshot/dev.23 24The device can be open either for reading or for writing. If open for25reading, it is considered to be in the suspend mode. Otherwise it is26assumed to be in the resume mode. The device cannot be open for simultaneous27reading and writing. It is also impossible to have the device open more than28once at a time.29 30Even opening the device has side effects. Data structures are31allocated, and PM_HIBERNATION_PREPARE / PM_RESTORE_PREPARE chains are32called.33 34The ioctl() commands recognized by the device are:35 36SNAPSHOT_FREEZE37 freeze user space processes (the current process is38 not frozen); this is required for SNAPSHOT_CREATE_IMAGE39 and SNAPSHOT_ATOMIC_RESTORE to succeed40 41SNAPSHOT_UNFREEZE42 thaw user space processes frozen by SNAPSHOT_FREEZE43 44SNAPSHOT_CREATE_IMAGE45 create a snapshot of the system memory; the46 last argument of ioctl() should be a pointer to an int variable,47 the value of which will indicate whether the call returned after48 creating the snapshot (1) or after restoring the system memory state49 from it (0) (after resume the system finds itself finishing the50 SNAPSHOT_CREATE_IMAGE ioctl() again); after the snapshot51 has been created the read() operation can be used to transfer52 it out of the kernel53 54SNAPSHOT_ATOMIC_RESTORE55 restore the system memory state from the56 uploaded snapshot image; before calling it you should transfer57 the system memory snapshot back to the kernel using the write()58 operation; this call will not succeed if the snapshot59 image is not available to the kernel60 61SNAPSHOT_FREE62 free memory allocated for the snapshot image63 64SNAPSHOT_PREF_IMAGE_SIZE65 set the preferred maximum size of the image66 (the kernel will do its best to ensure the image size will not exceed67 this number, but if it turns out to be impossible, the kernel will68 create the smallest image possible)69 70SNAPSHOT_GET_IMAGE_SIZE71 return the actual size of the hibernation image72 (the last argument should be a pointer to a loff_t variable that73 will contain the result if the call is successful)74 75SNAPSHOT_AVAIL_SWAP_SIZE76 return the amount of available swap in bytes77 (the last argument should be a pointer to a loff_t variable that78 will contain the result if the call is successful)79 80SNAPSHOT_ALLOC_SWAP_PAGE81 allocate a swap page from the resume partition82 (the last argument should be a pointer to a loff_t variable that83 will contain the swap page offset if the call is successful)84 85SNAPSHOT_FREE_SWAP_PAGES86 free all swap pages allocated by87 SNAPSHOT_ALLOC_SWAP_PAGE88 89SNAPSHOT_SET_SWAP_AREA90 set the resume partition and the offset (in <PAGE_SIZE>91 units) from the beginning of the partition at which the swap header is92 located (the last ioctl() argument should point to a struct93 resume_swap_area, as defined in kernel/power/suspend_ioctls.h,94 containing the resume device specification and the offset); for swap95 partitions the offset is always 0, but it is different from zero for96 swap files (see Documentation/power/swsusp-and-swap-files.rst for97 details).98 99SNAPSHOT_PLATFORM_SUPPORT100 enable/disable the hibernation platform support,101 depending on the argument value (enable, if the argument is nonzero)102 103SNAPSHOT_POWER_OFF104 make the kernel transition the system to the hibernation105 state (eg. ACPI S4) using the platform (eg. ACPI) driver106 107SNAPSHOT_S2RAM108 suspend to RAM; using this call causes the kernel to109 immediately enter the suspend-to-RAM state, so this call must always110 be preceded by the SNAPSHOT_FREEZE call and it is also necessary111 to use the SNAPSHOT_UNFREEZE call after the system wakes up. This call112 is needed to implement the suspend-to-both mechanism in which the113 suspend image is first created, as though the system had been suspended114 to disk, and then the system is suspended to RAM (this makes it possible115 to resume the system from RAM if there's enough battery power or restore116 its state on the basis of the saved suspend image otherwise)117 118The device's read() operation can be used to transfer the snapshot image from119the kernel. It has the following limitations:120 121- you cannot read() more than one virtual memory page at a time122- read()s across page boundaries are impossible (ie. if you read() 1/2 of123 a page in the previous call, you will only be able to read()124 **at most** 1/2 of the page in the next call)125 126The device's write() operation is used for uploading the system memory snapshot127into the kernel. It has the same limitations as the read() operation.128 129The release() operation frees all memory allocated for the snapshot image130and all swap pages allocated with SNAPSHOT_ALLOC_SWAP_PAGE (if any).131Thus it is not necessary to use either SNAPSHOT_FREE or132SNAPSHOT_FREE_SWAP_PAGES before closing the device (in fact it will also133unfreeze user space processes frozen by SNAPSHOT_UNFREEZE if they are134still frozen when the device is being closed).135 136Currently it is assumed that the userland utilities reading/writing the137snapshot image from/to the kernel will use a swap partition, called the resume138partition, or a swap file as storage space (if a swap file is used, the resume139partition is the partition that holds this file). However, this is not really140required, as they can use, for example, a special (blank) suspend partition or141a file on a partition that is unmounted before SNAPSHOT_CREATE_IMAGE and142mounted afterwards.143 144These utilities MUST NOT make any assumptions regarding the ordering of145data within the snapshot image. The contents of the image are entirely owned146by the kernel and its structure may be changed in future kernel releases.147 148The snapshot image MUST be written to the kernel unaltered (ie. all of the image149data, metadata and header MUST be written in _exactly_ the same amount, form150and order in which they have been read). Otherwise, the behavior of the151resumed system may be totally unpredictable.152 153While executing SNAPSHOT_ATOMIC_RESTORE the kernel checks if the154structure of the snapshot image is consistent with the information stored155in the image header. If any inconsistencies are detected,156SNAPSHOT_ATOMIC_RESTORE will not succeed. Still, this is not a fool-proof157mechanism and the userland utilities using the interface SHOULD use additional158means, such as checksums, to ensure the integrity of the snapshot image.159 160The suspending and resuming utilities MUST lock themselves in memory,161preferably using mlockall(), before calling SNAPSHOT_FREEZE.162 163The suspending utility MUST check the value stored by SNAPSHOT_CREATE_IMAGE164in the memory location pointed to by the last argument of ioctl() and proceed165in accordance with it:166 1671. If the value is 1 (ie. the system memory snapshot has just been168 created and the system is ready for saving it):169 170 (a) The suspending utility MUST NOT close the snapshot device171 _unless_ the whole suspend procedure is to be cancelled, in172 which case, if the snapshot image has already been saved, the173 suspending utility SHOULD destroy it, preferably by zapping174 its header. If the suspend is not to be cancelled, the175 system MUST be powered off or rebooted after the snapshot176 image has been saved.177 (b) The suspending utility SHOULD NOT attempt to perform any178 file system operations (including reads) on the file systems179 that were mounted before SNAPSHOT_CREATE_IMAGE has been180 called. However, it MAY mount a file system that was not181 mounted at that time and perform some operations on it (eg.182 use it for saving the image).183 1842. If the value is 0 (ie. the system state has just been restored from185 the snapshot image), the suspending utility MUST close the snapshot186 device. Afterwards it will be treated as a regular userland process,187 so it need not exit.188 189The resuming utility SHOULD NOT attempt to mount any file systems that could190be mounted before suspend and SHOULD NOT attempt to perform any operations191involving such file systems.192 193For details, please refer to the source code.194