Support for late resume #139
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
controlled by the 'late_resume' parameter, will move the trigger point for resume to after devices have been decrypted or discovered allows resume from encrypted swap or devices requiring extra handling be visible to the kernel
truffle0
force-pushed
the
dev
branch
2 times, most recently
from
872fe17 to
6067efc
Compare
currently it's more of a rearrangment of the original code, with some added safety/integrity checks `resume` is now a separate function, the plan is to have the code for actually resuming within this function and code for locating the device outside. in the more general case, if `late_resume` is enabled, the initram will check at both the `init_early` and again at the `init_premount` stage for the presence of the resume device. Once host-only is implemented then it'll be able to determine at build time which state it'll need to check at. For safety checks, currently it checks whether the device is a valid block device, whether resume has already been attempted previous (based on the content of `/sys/power/resume`) & whether there are are any other mounted block devices.
desultory
force-pushed
the
dev
branch
2 times, most recently
from
b86a171 to
80f9d3c
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Adds a new
late_resumeparameter, which moves the trigger point frominit_earlytoinit_premount.This allows cryptsetup and LVM to initialize before the system attempts to resume, making encrypted or otherwise non-trival-to-access swap partitions accessible.
In addition it will use allow using
resume=UUID=*to specify the resume device, as any partitions that aren't directly on disk don't have aPARTUUID.Which this option is safe in theory and I have been using it without issue for a few months, it is important to note that configurations that cause writes to the device before triggering resume run the risk of system instability and data loss.
I have purposefully used
init_premountsection to trigger before filesystems are mounted as the kernel shouldn't make any metadata changes (save for possibly LVM access times) before mounting filesystems.I have also made sure to include warnings in the configuration docs and the output of ugrd when this option is enabled to make sure users understand the implications (including a link to the kernel docs that detail the risks).
I'm not sure whether the warning is overkill, as I believe this option is safe to use under normal encrypted swap and lvm setups, but to cater to potentially non-standard setups I think it's important to have.