Does Linux have a variably-sized hibernation file like hiberfil.sys?

JASlinux · 12-12-2021, 06:58 AM

I would hibernate in Linux if it were possible without a large swap file that is always in use.

Quote:

the speed with which the system resumes depends on how big the hiberfil.sys file is

https://www.techrepublic.com/blog/wi...feature-works/

TB0ne · 12-12-2021, 11:57 AM

Quote:

Originally Posted by JASlinux

I would hibernate in Linux if it were possible without a large swap file that is always in use.

Code:

the speed with which the system resumes depends on how big the hiberfil.sys file is

https://www.techrepublic.com/blog/wi...feature-works/

From a brief search about Linux hibernation; bolded a piece for emphasis only:
https://wiki.archlinux.org/title/Pow...tion/file_size

Quote:

Originally Posted by Docs

You may either decrease the value of /sys/power/image_size to make the suspend image as small as possible (for small swap partitions), or increase it to possibly speed up the hibernation process. For systems with a large amount of RAM, smaller values may drastically increase the speed of resuming a hibernating system. See systemd#systemd-tmpfiles - temporary files to make this change persistent.

Linux isn't Windows, and you should look up the differences between suspend and hibernate, what they are, and how they work.

Not sure what the actual issue/problem is; suspend is always going to be faster, and there are well known issues with both suspend and hibernate in Linux. My laptops boot cold to desktop in less than 30 seconds...not sure why that would be too long for you to wait.

rnturn · 12-12-2021, 02:23 PM

Quote:

Originally Posted by JASlinux

I would hibernate in Linux if it were possible without a large swap file that is always in use.

https://www.techrepublic.com/blog/wi...feature-works/

The swap area, as you probably already know (regardless of whether it's a file or a partition), needs to be large enough to hold everything in RAM and likely more for the situations where you're already using more memory than you have physical RAM for your running processes. My rule of thumb comes from my days administering Oracle servers: 4 x physical RAM (disk space is cheap nowadays). With a large-enough swap partition, I've had no problems hibernating. I've only hibernated on my laptop, though. It ought to work on my desktop system but I've just never used it.

There is a kernel boot command line option, "resume=...", that points to a disk partition. I seriously doubt it could be used to specify a file in a filesystem. (Does the kernel know about what filesystem is built on a particular partition during bootstrap and know how to look for a file within it? I'm doubting it.) There are other boot command line parameters that are used for suspend/hibernation in "/usr/src/linux/Documentation/admin-guide/kernel-parameters.txt". A little research into their use might get you what you want.

Hope this helps...

allend · 12-12-2021, 07:45 PM

Quote:

There is a kernel boot command line option, "resume=...", that points to a disk partition. I seriously doubt it could be used to specify a file in a filesystem.

From /usr/src/linux/Documentation/power/swsusp-and-swap-files.rst

Quote:

In principle the location of a swap file's header may be determined with the
help of appropriate filesystem driver. Unfortunately, however, it requires the
filesystem holding the swap file to be mounted, and if this filesystem is
journaled, it cannot be mounted during resume from disk. For this reason to
identify a swap file swsusp uses the name of the partition that holds the file
and the offset from the beginning of the partition at which the swap file's
header is located. For convenience, this offset is expressed in <PAGE_SIZE>
units.

sundialsvcs · 12-14-2021, 09:06 AM

There are several ways that the Linux kernel can implement "suspend." One way is to use the swap area, which may or may not be a file. The kernel architecture is somewhat flexible here.