Repository navigation
dkms: fix ldtarball loading only one kernel from a multi-kernel tarball - #612
Merged
Merged
Conversation
load_tarball() declared kernel_arch_to_load and dkms_dir_location with `local -r` inside the loop over dkms_main_tree/*/*. Bash cannot re-declare a readonly variable, so from the second iteration on both kept the values of the first kernel/arch directory. Without --force every further kernel was skipped with a misleading "already exists" warning; with --force each was copied over the first kernel's directory, registering modules built for one kernel under another. Drop -r, same as was done for make_tarball() in dkms-project#453. Add a mktarball/ldtarball round-trip test with three kernels (two of them fake copies of the real build), with and without --force, which covers both fixes.
Member
|
Thank you! |
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
load_tarball() declared kernel_arch_to_load and dkms_dir_location with
local -rinside the loop overdkms_main_tree/*/*. Bash cannot re-declare a readonly variable, so from the second iteration on both kept the values of the first kernel/arch directory. Without--forceevery further kernel was skipped with a misleading "already exists" warning; with--forceeach was copied over the first kernel's directory, registering modules built for one kernel under another.Drop -r, same as was done for make_tarball() in #453.
Add a mktarball/ldtarball round-trip test with three kernels (two of them fake copies of the real build), with and without
--force, which covers both fixes.