Repository navigation
fix(zstd): validate frame content sizes - #9164
Merged
Merged
Conversation
Validate serialized zstd output sizes against frame headers before CPU allocation and host-backed GPU planning. Add regressions for zstd arrays and zstd buffers. Tested with: - cargo nextest run -p vortex-zstd --all-features - cargo clippy -p vortex-zstd --all-targets --all-features -- -D warnings Signed-off-by: Lorenz Hübschle <lorenz@firebolt.io>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Rationale for this change
Fuzzing found that zstd (arrays + buffers) serialised output sizes were not validated against the declared uncompressed sizes in frame headers.
What changes are included in this PR?
Validate serialized zstd output sizes against frame headers before CPU allocation and host-backed GPU planning. Add regressions for zstd arrays and zstd buffers.
Note that as long as the size declared in the frame header matches what's declared in metadata, allocations are not bounded. But this is much harder to fix, since zstd can essentially decompress to
⌈compressed_len/3⌉ × 128 KiBwhich is ~42GiB per MiB of compressed data so not particularly useful as a bound.What APIs are changed? Are there any user-facing changes?
Malformed files where output sizes declared in zstd frames don't match the header are now refused instead of allocating a buffer of file-controlled size.
Investigated and built with Codex.