Zstd bomb fix #937
Open
Zstd bomb fix #937
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
l-monninger
reviewed
Dec 10, 2024
| bcs::from_bytes(decompressed.as_slice()).context("failed to deserialize blob")?; | ||
| // decompress the blob and deserialize the data with bcs | ||
| let decoder = zstd::Decoder::new(blob.data.as_slice())?; | ||
| let blob = bcs::from_reader(decoder).context("failed to deserialize blob")?; |
There was a problem hiding this comment.
Do you know how fill_buf would be called here? I wondering if this in fact going to raise an Error or whether it might just panic if the buffer size is restricted.
There was a problem hiding this comment.
I don't think I fully understand this question, and we didn't restrict the size of the internal buffer.
But you did bring to my attention that the slice also supports BufRead and we can take advantage of it, bypassing the intermediate buffer.
There was a problem hiding this comment.
Yes, bcs would return an Error.
I have implemented a customizable limit on length of any variable-length fields in zefchain/bcs#12.
Use streamed decompression API to avoid potentially huge allocations for blob data decompressed from zstd. The bcs decoder enforces the length limit of 2^31 - 1 on byte arrays.
mzabaluev
force-pushed
the
mikhail/zstd-bomb
branch
from
16ea004 to
b2fd870
Compare
Avoid an intermediate buffer when serializing and compressing for CelestiaBlob, plugging the encoder into the serializer.
The data slice is already BufRead, just use it directly.
Use a crafted zstd payload as submitted in #876 (comment)
Need this to enable length limit enforcement on deserialization of blobs.
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.
Fixes #876 (but see outstanding issues below).
Dependencies
movementlabsxyz/aptos-core#111
Summary
protocol-units.Use streamed decompression API to avoid potentially huge allocations
for blob data decompressed from zstd. The bcs decoder enforces the
length limit of 2^31 - 1 on byte arrays.
Changelog
Testing
Added unit tests to movement-celestia-da-util to exercise compression and decompression,
with supercritical and valid lengths for compressed payload and the data field. The test creating legit data structures with super-long blobs is ignored in default runs because of the memory and processing requirements.
Outstanding issues
As commented in #876 (comment), there are four byte array fields in the blob data structure that can each be bloated up to 2 GiB.