Fix issues with RAID1 metadata 1.0 partitions on IEEE1275#212
Open
Rolv-Apneseth wants to merge 2 commits into
Open
Fix issues with RAID1 metadata 1.0 partitions on IEEE1275#212Rolv-Apneseth wants to merge 2 commits into
Rolv-Apneseth wants to merge 2 commits into
Conversation
Contributor
|
There must be an empty newline between commit subject and description, so please fix it. Otherwise LGTM @nfrayer can you also take a look please? |
The powerpc guard in grub_mdraid_detect skips metadata 1.0 to avoid a firmware block issue, but also ends up applying to userspace utilities where reads will be going through the kernel. Extend the conditional with !defined(GRUB_UTIL) so utilities like grub-install can detect metadata 1.0 arrays. Signed-off-by: Rolv Apneseth <rolv.apneseth@gmail.com>
On IEEE1275, grub-install hardcodes an mduuid prefix for RAID abstractions, requiring array assembly at boot. This fails on metadata 1.0 partitions because detection is skipped due to a firmware issue. Since RAID1 mirrors are individually readable, use search.fs_uuid on IEEE1275 to find the boot filesystem on any member partition and avoid requiring array assembly. Signed-off-by: Rolv Apneseth <rolv.apneseth@gmail.com>
Rolv-Apneseth
force-pushed
the
fix_ppc_raid1
branch
from
7fb383b to
01d7dfc
Compare
Contributor
|
lgtm ! |
Author
|
Thanks for the reviews! I believe I've addressed the newline between commit subject and description but let me know if there's anything else |
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
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
I'm here from coreos/fedora-coreos-tracker#2134, where the relevant context for this PR is in coreos/fedora-coreos-tracker#2134 (comment).
This PR contains 2 commits, which when combined fix the issue we're encountering with metadata 1.0 RAID1 arrays on ppc64le on Fedora CoreOS. Additional context can be found in the comment linked above, and in the individual commits, but for a brief overview (both specific to PowerPC):
grub-install, usesearch.fs_uuidfor RAID1 instead of hard-coding the mduuid prefix, which requires array assembly at boot (and hence can hit the metadata 1.0 skip)Theoretically these changes should only affect PowerPC RAID1, but I did test that this didn't break boot on x86 or aarch64 (ran our various boot tests with builds including the patched rpm). And with these patches, all our tests pass on PowerPC.
Please let me know if there are any issues, or there are additional checks/tests you would like me to run.