Pad vbmeta images during cvd fetch - #3192
Open
3405691582 wants to merge 1 commit into
Open
3405691582 wants to merge 1 commit into
3405691582 wants to merge 1 commit into
Conversation
Databean
approved these changes
Sep 17, 2026
3405691582
added a commit
to 3405691582/android-cuttlefish
that referenced
this pull request
Sep 21, 2026
In google#3192 we ensure that fetched files are modified during fetch, in order to front-load modifications of fetched resources. Substitution also requires modification of the fetched tree. cvd fetch and cvd start can both trigger substitution -- we require both since we need to operate on fetched and local builds. However, we try to modify the tree during cvd start as well, even when the symlinks are already correct. If we have made the host artefacts read-only after fetching them, the subsequent substitution fails. There's no need to regenerate the symlink; do it once and only once.
3405691582
force-pushed
the
vbmeta_fetch
branch
from
September 21, 2026 15:00
fd4e072 to
85ac242
Compare
3405691582
enabled auto-merge
September 21, 2026 17:02
The guest reads the vbmeta images with libavb, which expects to be able to read the maximum vbmeta size, so we pad the images in place to match this or the read will fail. This implies that the image location must be writable, which is problematic for particular use-cases. It is worthwhile noting that images are also modified during cvd fetch to de-sparse images before the images are quiescent on disk. While this change does not alleviate the need for in-place image modification, it does ensure that the images need not be modifiable after the cvd fetch occurs. assemble_cvd behavior is unchanged -- indeed it must not change, since it will need to operate correctly on images not sourced from cvd fetch.
3405691582
force-pushed
the
vbmeta_fetch
branch
from
September 21, 2026 21:54
85ac242 to
5f797b4
Compare
GoogleCuttlefishTesterBot
removed
the
kokoro:force-run
pull Bot
pushed a commit
to emaheuxPEREN/android-cuttlefish
that referenced
this pull request
Sep 21, 2026
In google#3192 we ensure that fetched files are modified during fetch, in order to front-load modifications of fetched resources. Substitution also requires modification of the fetched tree. cvd fetch and cvd start can both trigger substitution -- we require both since we need to operate on fetched and local builds. However, we try to modify the tree during cvd start as well, even when the symlinks are already correct. If we have made the host artefacts read-only after fetching them, the subsequent substitution fails. There's no need to regenerate the symlink; do it once and only once.
github-merge-queue
Bot
removed this pull request from the merge queue due to failed status checks
Sep 22, 2026
This branch has not been deployed
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.
The guest reads the vbmeta images with libavb, which expects to be able to read the maximum vbmeta size, so we pad the images in place to match this or the read will fail.
This implies that the image location must be writable, which is problematic for particular use-cases. It is worthwhile noting that images are also modified during cvd fetch to de-sparse images before the images are quiescent on disk. While this change does not alleviate the need for in-place image modification, it does ensure that the images need not be modifiable after the cvd fetch occurs.
assemble_cvd behavior is unchanged -- indeed it must not change, since it will need to operate correctly on images not sourced from cvd fetch.