New PR with Devin's complete changes #507
There are no files selected for viewing
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
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
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
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
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,20 @@ | ||
| //! BSON Document Length Field Fuzzer | ||
| //! | ||
| //! This fuzz test focuses on finding security vulnerabilities related to BSON document length fields. | ||
| //! It specifically targets: | ||
| //! - Integer overflow/underflow in length calculations | ||
| //! - Malformed length fields that could cause buffer overruns | ||
| //! - Mismatches between declared and actual document sizes | ||
| //! - Memory allocation issues with large or invalid lengths | ||
|
|
||
| #![no_main] | ||
| #[macro_use] extern crate libfuzzer_sys; | ||
| extern crate bson; | ||
| use bson::RawDocument; | ||
|
|
||
| fuzz_target!(|buf: &[u8]| { | ||
| if buf.len() >= 4 { | ||
|
There was a problem hiding this comment. Why the length guard? There was a problem hiding this comment. Devin got this from the docs and/or source, I agree it would be better to remove this guard. |
||
| // Focus on document length field manipulation | ||
| let _ = RawDocument::from_bytes(buf); | ||
| } | ||
| }); | ||
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
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,209 @@ | ||
| #![no_main] | ||
| use arbitrary::Arbitrary; | ||
| use bson::{ | ||
| raw::{RawBson, RawBsonRef, RawDocument}, | ||
| spec::BinarySubtype, | ||
| Decimal128, | ||
| }; | ||
| use libfuzzer_sys::fuzz_target; | ||
| use std::str::FromStr; | ||
|
|
||
| fn convert_bson_ref(bson_ref: RawBsonRef) -> Option<RawBson> { | ||
|
There was a problem hiding this comment. I'm not sure I understand the intent of this function (or of the fuzz test in this file, for that matter). This function converts from a borrowed to an owned value (e.g. what The handling of There was a problem hiding this comment. Yes, this is what it did when told to compare the input to the output |
||
| match bson_ref { | ||
| RawBsonRef::Double(d) => { | ||
| if d.is_nan() { | ||
| Some(RawBsonRef::Double(f64::NAN).to_raw_bson()) | ||
| } else if d.is_infinite() { | ||
|
There was a problem hiding this comment. Nit: this case can be collapsed into the final There was a problem hiding this comment. Presumably Devin would fix this kind of issue if asked. You're definitely right. In this case I think I'll fix it myself. |
||
| Some(RawBsonRef::Double(d).to_raw_bson()) | ||
| } else { | ||
| Some(RawBsonRef::Double(d).to_raw_bson()) | ||
| } | ||
| } | ||
| RawBsonRef::String(s) => { | ||
| if !s.is_empty() && !s.contains('\0') && s.len() <= (i32::MAX as usize) { | ||
| Some(RawBsonRef::String(s).to_raw_bson()) | ||
| } else { | ||
| None | ||
| } | ||
| } | ||
| RawBsonRef::Document(d) => { | ||
| let mut valid = true; | ||
| for result in d.iter() { | ||
| match result { | ||
| Ok((key, _)) if key.is_empty() || key.contains('\0') => { | ||
| valid = false; | ||
| break; | ||
| } | ||
| Err(_) => { | ||
| valid = false; | ||
| break; | ||
| } | ||
| _ => {} | ||
| } | ||
| } | ||
| if valid { | ||
| Some(RawBsonRef::Document(d).to_raw_bson()) | ||
| } else { | ||
| None | ||
| } | ||
| } | ||
| RawBsonRef::Array(a) => { | ||
| let mut valid = true; | ||
| for result in a.into_iter() { | ||
| if result.is_err() { | ||
| valid = false; | ||
| break; | ||
| } | ||
| } | ||
| if valid { | ||
| Some(RawBsonRef::Array(a).to_raw_bson()) | ||
| } else { | ||
| None | ||
| } | ||
| } | ||
| RawBsonRef::Binary(b) => { | ||
| if b.bytes.len() <= i32::MAX as usize && | ||
| match b.subtype { | ||
| BinarySubtype::Generic | | ||
| BinarySubtype::Function | | ||
| BinarySubtype::BinaryOld | | ||
| BinarySubtype::UuidOld | | ||
| BinarySubtype::Uuid | | ||
| BinarySubtype::Md5 | | ||
| BinarySubtype::UserDefined(_) => true, | ||
| _ => false | ||
| } { | ||
| Some(RawBsonRef::Binary(b).to_raw_bson()) | ||
| } else { | ||
| None | ||
| } | ||
| } | ||
| RawBsonRef::ObjectId(id) => Some(RawBsonRef::ObjectId(id).to_raw_bson()), | ||
|
There was a problem hiding this comment. This (and all the other cases with no extra logic) can just be collapsed down to a single There was a problem hiding this comment. This is interesting. I often don't like catch-alls because they can cause errors when new types are added, but I don't think the bson spec will be adding types any time soon. |
||
| RawBsonRef::Boolean(b) => Some(RawBsonRef::Boolean(b).to_raw_bson()), | ||
| RawBsonRef::Null => Some(RawBsonRef::Null.to_raw_bson()), | ||
| RawBsonRef::RegularExpression(regex) => { | ||
| let valid_options = "ilmsux"; | ||
| let mut options_sorted = regex.options.chars().collect::<Vec<_>>(); | ||
| options_sorted.sort_unstable(); | ||
| options_sorted.dedup(); | ||
| let sorted_str: String = options_sorted.into_iter().collect(); | ||
|
|
||
| if sorted_str.chars().all(|c| valid_options.contains(c)) && | ||
| !regex.pattern.contains('\0') && | ||
| regex.pattern.len() <= (i32::MAX as usize) { | ||
| Some(RawBsonRef::RegularExpression(regex).to_raw_bson()) | ||
| } else { | ||
| None | ||
| } | ||
| } | ||
| RawBsonRef::JavaScriptCode(code) => { | ||
| if !code.is_empty() && !code.contains('\0') && code.len() <= (i32::MAX as usize) { | ||
| Some(RawBsonRef::JavaScriptCode(code).to_raw_bson()) | ||
| } else { | ||
| None | ||
| } | ||
| } | ||
| RawBsonRef::JavaScriptCodeWithScope(code_w_scope) => { | ||
| if !code_w_scope.code.is_empty() && | ||
| !code_w_scope.code.contains('\0') && | ||
| code_w_scope.code.len() <= (i32::MAX as usize) { | ||
| Some(RawBsonRef::JavaScriptCodeWithScope(code_w_scope).to_raw_bson()) | ||
| } else { | ||
| None | ||
| } | ||
| } | ||
| RawBsonRef::DbPointer(ptr) => { | ||
| let raw_bson = RawBsonRef::DbPointer(ptr).to_raw_bson(); | ||
| Some(raw_bson) | ||
| } | ||
| RawBsonRef::Symbol(s) => { | ||
| if !s.is_empty() && !s.contains('\0') && s.len() <= i32::MAX as usize { | ||
| Some(RawBsonRef::Symbol(s).to_raw_bson()) | ||
| } else { | ||
| None | ||
| } | ||
| } | ||
| RawBsonRef::Int32(i) => Some(RawBsonRef::Int32(i).to_raw_bson()), | ||
| RawBsonRef::Int64(i) => Some(RawBsonRef::Int64(i).to_raw_bson()), | ||
| RawBsonRef::Timestamp(ts) => Some(RawBsonRef::Timestamp(ts).to_raw_bson()), | ||
| RawBsonRef::DateTime(dt) => Some(RawBsonRef::DateTime(dt).to_raw_bson()), | ||
| RawBsonRef::Decimal128(d) => { | ||
| let d_str = d.to_string(); | ||
| if d_str.contains("NaN") { | ||
| if let Ok(nan) = Decimal128::from_str("NaN") { | ||
| Some(RawBsonRef::Decimal128(nan).to_raw_bson()) | ||
| } else { | ||
| None | ||
| } | ||
| } else if d_str == "Infinity" || d_str == "-Infinity" { | ||
| if let Ok(val) = Decimal128::from_str(&d_str) { | ||
| Some(RawBsonRef::Decimal128(val).to_raw_bson()) | ||
| } else { | ||
| None | ||
| } | ||
| } else { | ||
| Some(RawBsonRef::Decimal128(d).to_raw_bson()) | ||
| } | ||
| } | ||
| RawBsonRef::MinKey => Some(RawBsonRef::MinKey.to_raw_bson()), | ||
| RawBsonRef::MaxKey => Some(RawBsonRef::MaxKey.to_raw_bson()), | ||
| RawBsonRef::Undefined => Some(RawBsonRef::Undefined.to_raw_bson()), | ||
| } | ||
| } | ||
|
|
||
| #[derive(Debug, Arbitrary)] | ||
|
There was a problem hiding this comment. Why is this an There was a problem hiding this comment. I rewrote this test entirely. Going to have to say Devin failed on this one. |
||
| struct Input { | ||
| bytes: Vec<u8>, | ||
| } | ||
|
|
||
| fuzz_target!(|input: Input| { | ||
| if let Ok(doc) = RawDocument::from_bytes(&input.bytes) { | ||
| for result in doc.iter() { | ||
| if let Ok((key, value)) = result { | ||
| if let Some(converted) = convert_bson_ref(value) { | ||
| let original_bytes = value.to_raw_bson(); | ||
| match value { | ||
| RawBsonRef::Double(d) if d.is_nan() => { | ||
| if let Some(converted_ref) = converted.as_raw_bson_ref().as_f64() { | ||
| assert!(converted_ref.is_nan(), | ||
| "NaN comparison failed for key: {}", key); | ||
| } | ||
| } | ||
| RawBsonRef::Double(d) if d.is_infinite() => { | ||
| if let Some(converted_ref) = converted.as_raw_bson_ref().as_f64() { | ||
| assert_eq!(d.is_sign_positive(), converted_ref.is_sign_positive(), | ||
| "Infinity sign mismatch for key: {}", key); | ||
| assert!(converted_ref.is_infinite(), | ||
| "Infinity comparison failed for key: {}", key); | ||
| } | ||
| } | ||
| RawBsonRef::Decimal128(d) if d.to_string().contains("NaN") => { | ||
| match converted.as_raw_bson_ref() { | ||
| RawBsonRef::Decimal128(cd) => { | ||
| assert!(cd.to_string().contains("NaN"), | ||
| "Decimal128 NaN comparison failed for key: {}", key); | ||
| } | ||
| _ => panic!("Type mismatch: expected Decimal128, got different type for key: {}", key), | ||
| } | ||
| } | ||
| RawBsonRef::Decimal128(d) if d.to_string().contains("Infinity") => { | ||
| match converted.as_raw_bson_ref() { | ||
| RawBsonRef::Decimal128(cd) => { | ||
| let d_str = d.to_string(); | ||
| let cd_str = cd.to_string(); | ||
| assert_eq!(d_str, cd_str, | ||
| "Decimal128 Infinity comparison failed for key: {}", key); | ||
| } | ||
| _ => panic!("Type mismatch: expected Decimal128, got different type for key: {}", key), | ||
| } | ||
| } | ||
| _ => { | ||
| assert_eq!(converted, original_bytes, | ||
| "Serialization mismatch for key: {}", key); | ||
| } | ||
| } | ||
| } | ||
| } | ||
| } | ||
| } | ||
| }); | ||
Oops, something went wrong.
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.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, devin actually updated this comment