Skip to content

feat(ffe): return dependent evaluation batches - #2772

Closed
typotter wants to merge 1 commit into
helen/ffe-dependent-flags-v46from
typo/dependent-flags-returned-batch-v46
Closed

feat(ffe): return dependent evaluation batches#2772
typotter wants to merge 1 commit into
helen/ffe-dependent-flags-v46from
typo/dependent-flags-returned-batch-v46

Conversation

@typotter

@typotter typotter commented Oct 30, 2026

Copy link
Copy Markdown

What does this PR do?

Adds dependent-flag evaluation to the v46 FFE implementation and returns the resulting telemetry after the root evaluation completes.

  • Returns completion-ordered dependency evaluation records separately from the root result.
  • Preserves successful dependency records and propagated errors when a later condition or the root fails.
  • Returns an ordered, transactional exposure trace only for successful root evaluations.
  • Replays memoized exposures without duplicating dependency evaluation records.
  • Exposes the returned batch and exposure trace through the C FFI with parent-owned borrowed views.
  • Rejects ambiguous dependency sources and unknown nested flagEvaluation fields.
  • Records dependencies rejected by the configured depth limit before propagating the error.

Motivation

Python, Ruby, and PHP consume libdatadog through FFI. They need the complete dependency evaluation batch after root completion so they can publish dependency telemetry before or alongside ordinary root telemetry. Dependency exposures remain transactional and must be discarded when the root graph fails.

Additional Notes

This PR is stacked on helen/ffe-dependent-flags-v46 and preserves the v46.1 workspace and Rust 1.87 toolchain. A traversal-time callback ABI is not required.

How to test the change?

The final reviewed tree passed:

  • cargo test -p libdd-ffe dependent_flags_tests — 21 passed.
  • cargo test -p libdd-ffe-ffi returned_batch_ffi — 2 passed.
  • cargo +nightly-2026-07-26 fmt --all -- --check for the changed files.
  • git diff --check.

The tests cover successful roots, rejected rules and allocations, no-assignment paths, root failures, transitive dependencies, memoized reuse, depth-limit failures, completion ordering, exposure rollback, and C FFI access and ownership.

@typotter Codex GPT-5
feat(ffe): return dependent evaluation batches
288685b 
Return completion-ordered dependency evaluations and the transactional exposure trace through the v46 FFE API.

Environment: Datadog workspace

Co-Authored-By: Codex GPT-5 <noreply@localhost>
@typotter
typotter requested review from a team as code owners October 30, 2026 16:18
@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Oct 30, 2026 •
edited
Loading

Copy link
Copy Markdown

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review Status Commit Review trigger
📝 Code Review ✅ Completed 2026-10-30T16:21:23.057433Z 288685b PR opened
🔒 Security Review ✅ Completed 2026-10-30T16:27:53.063131Z 288685b PR opened
ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

@typotter
typotter marked this pull request as draft October 30, 2026 16:19
@github-actions

github-actions Bot commented Oct 30, 2026

Copy link
Copy Markdown
Contributor

📚 Documentation Check Results

⚠️ 1099 documentation warning(s) found

📦 libdd-ffe-ffi - 552 warning(s)

📦 libdd-ffe - 391 warning(s)

📦 cc_utils - 24 warning(s)

📦 tools - 132 warning(s)

Updated: 2026-10-30 16:21:05 UTC | Commit: f5abe08 | missing-docs job results

@github-actions

github-actions Bot commented Oct 30, 2026

Copy link
Copy Markdown
Contributor

🔒 Cargo Deny Results

⚠️ 16 issue(s) found, showing only errors (advisories, bans, sources)

📦 libdd-ffe-ffi - 6 error(s)

Show output 
error[unmaintained]: Bincode is unmaintained
   ┌─ /home/runner/work/libdatadog/libdatadog/Cargo.lock:25:1
   │
25 │ bincode 1.3.3 registry+https://github.com/rust-lang/crates.io-index
   │ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ unmaintained advisory detected
   │
   ├ ID: RUSTSEC-2025-0141
   ├ Advisory: https://rustsec.org/advisories/RUSTSEC-2025-0141
   ├ Due to a doxxing and harassment incident, the bincode team has taken the decision to cease development permanently.
     
     The team considers version 1.3.3 a complete version of bincode that is not in need of any updates.
     
     ## Alternatives to consider
     
     * [wincode](https://crates.io/crates/wincode)
     * [postcard](https://crates.io/crates/postcard)
     * [bitcode](https://crates.io/crates/bitcode)
     * [rkyv](https://crates.io/crates/rkyv)
   ├ Announcement: https://git.sr.ht/~stygianentity/bincode/tree/v3.0/item/README.md
   ├ Solution: No safe upgrade is available!
   ├ bincode v1.3.3
     └── (dev) libdd-ffe v2.0.0
         └── libdd-ffe-ffi v1.0.2

error[vulnerability]: TLS 1.3 handshake messages incorrectly accepted across encryption level boundaries
    ┌─ /home/runner/work/libdatadog/libdatadog/Cargo.lock:195:1
    │
195 │ rustls 0.23.37 registry+https://github.com/rust-lang/crates.io-index
    │ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ security vulnerability detected
    │
    ├ ID: RUSTSEC-2026-0285
    ├ Advisory: https://rustsec.org/advisories/RUSTSEC-2026-0285
    ├ Rustls accepted TLS 1.3 handshake messages sent at the wrong encryption level
      when they followed a key-changing message in the same record. For example,
      a plaintext `EncryptedExtensions` message packed into the same record as the
      `ServerHello` was accepted.
      
      RFC 8446 section 5.1 requires that handshake messages do not span key changes,
      and that implementations terminate the connection with an "unexpected_message"
      alert if they do.
      
      The handshake transcript is still authenticated, so a network-position attacker
      cannot use this to alter or complete a handshake; the practical effect is that
      a peer could send handshake messages that should be encrypted in plaintext
      without rustls rejecting the connection.
      
      This is functionally the same bug as Go's
      [GO-2026-4340](https://pkg.go.dev/vuln/GO-2026-4340) (CVE-2025-61730).
    ├ Announcement: https://github.com/rustls/rustls/pull/3265
    ├ Solution: Upgrade to >=0.23.45 (try `cargo update -p rustls`)
    ├ rustls v0.23.37
      ├── hyper-rustls v0.27.7
      │   └── libdd-common v8.0.0
      │       ├── libdd-capabilities-impl v7.0.0
      │       │   ├── (dev) libdd-ffe v2.0.0
      │       │   │   └── libdd-ffe-ffi v1.0.2
      │       │   └── libdd-remote-config v7.0.0
      │       │       ├── libdd-ffe v2.0.0 (*)
      │       │       └── (dev) libdd-remote-config v7.0.0 (*)
      │       ├── libdd-common-ffi v46.1.0
      │       │   └── libdd-ffe-ffi v1.0.2 (*)
      │       ├── libdd-ffe v2.0.0 (*)
      │       └── libdd-remote-config v7.0.0 (*)
      ├── libdd-common v8.0.0 (*)
      ├── rustls-platform-verifier v0.6.2
      │   └── libdd-common v8.0.0 (*)
      └── tokio-rustls v0.26.0
          └── hyper-rustls v0.27.7 (*)

error[vulnerability]: `Debug` impls for client session values expose TLS secrets
    ┌─ /home/runner/work/libdatadog/libdatadog/Cargo.lock:195:1
    │
195 │ rustls 0.23.37 registry+https://github.com/rust-lang/crates.io-index
    │ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ security vulnerability detected
    │
    ├ ID: RUSTSEC-2026-0346
    ├ Advisory: https://rustsec.org/advisories/RUSTSEC-2026-0346
    ├ The `Debug` implementations of `Tls12ClientSessionValue` and
      `Tls13ClientSessionValue` were derived, and so included the TLS 1.2 master
      secret or TLS 1.3 resumption PSK in their output.
      
      Custom `ClientSessionStore` implementations that format these values with
      `Debug` (for example, when logging) could disclose these secrets to readers of
      the logs. This conflicts with rustls's
      [key-material redaction policy](https://github.com/rustls/rustls/blob/99f2358cae2954837dbb866faf6727de75489ab9/SECURITY.md#boundary-public-api).
      
      Ordinary connection and built-in session cache `Debug` output is redacted and
      is not affected.
    ├ Announcement: https://github.com/rustls/rustls/pull/3299
    ├ Solution: Upgrade to >=0.23.46 (try `cargo update -p rustls`)
    ├ rustls v0.23.37
      ├── hyper-rustls v0.27.7
      │   └── libdd-common v8.0.0
      │       ├── libdd-capabilities-impl v7.0.0
      │       │   ├── (dev) libdd-ffe v2.0.0
      │       │   │   └── libdd-ffe-ffi v1.0.2
      │       │   └── libdd-remote-config v7.0.0
      │       │       ├── libdd-ffe v2.0.0 (*)
      │       │       └── (dev) libdd-remote-config v7.0.0 (*)
      │       ├── libdd-common-ffi v46.1.0
      │       │   └── libdd-ffe-ffi v1.0.2 (*)
      │       ├── libdd-ffe v2.0.0 (*)
      │       └── libdd-remote-config v7.0.0 (*)
      ├── libdd-common v8.0.0 (*)
      ├── rustls-platform-verifier v0.6.2
      │   └── libdd-common v8.0.0 (*)
      └── tokio-rustls v0.26.0
          └── hyper-rustls v0.27.7 (*)

error[vulnerability]: Brotli certificate decompression can allocate up to 1 GiB before authentication
    ┌─ /home/runner/work/libdatadog/libdatadog/Cargo.lock:195:1
    │
195 │ rustls 0.23.37 registry+https://github.com/rust-lang/crates.io-index
    │ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ security vulnerability detected
    │
    ├ ID: RUSTSEC-2026-0348
    ├ Advisory: https://rustsec.org/advisories/RUSTSEC-2026-0348
    ├ The built-in Brotli certificate decompressor (enabled by the non-default
      `brotli` crate feature) bounded the size of the decompressed output, but
      accepted a peer-selected history window. Brotli's large-window extension was
      enabled, so a five-byte header could cause an allocation of 1 GiB. A peer can
      abort a process whose allocator cannot satisfy that request.
      
      No trusted certificate is required by the attacker. Affected configurations are:
      
      - clients that enable the `brotli` feature and connect to a malicious server, and
      - servers that enable the `brotli` feature and request client certificates.
      
      Applications that do not enable the `brotli` feature are not affected.
    ├ Announcement: https://github.com/rustls/rustls/pull/3297
    ├ Solution: Upgrade to >=0.23.46 (try `cargo update -p rustls`)
    ├ rustls v0.23.37
      ├── hyper-rustls v0.27.7
      │   └── libdd-common v8.0.0
      │       ├── libdd-capabilities-impl v7.0.0
      │       │   ├── (dev) libdd-ffe v2.0.0
      │       │   │   └── libdd-ffe-ffi v1.0.2
      │       │   └── libdd-remote-config v7.0.0
      │       │       ├── libdd-ffe v2.0.0 (*)
      │       │       └── (dev) libdd-remote-config v7.0.0 (*)
      │       ├── libdd-common-ffi v46.1.0
      │       │   └── libdd-ffe-ffi v1.0.2 (*)
      │       ├── libdd-ffe v2.0.0 (*)
      │       └── libdd-remote-config v7.0.0 (*)
      ├── libdd-common v8.0.0 (*)
      ├── rustls-platform-verifier v0.6.2
      │   └── libdd-common v8.0.0 (*)
      └── tokio-rustls v0.26.0
          └── hyper-rustls v0.27.7 (*)

error[vulnerability]: Unauthenticated ServerHello can downgrade later connections to classical key exchange
    ┌─ /home/runner/work/libdatadog/libdatadog/Cargo.lock:195:1
    │
195 │ rustls 0.23.37 registry+https://github.com/rust-lang/crates.io-index
    │ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ security vulnerability detected
    │
    ├ ID: RUSTSEC-2026-0349
    ├ Advisory: https://rustsec.org/advisories/RUSTSEC-2026-0349
    ├ The client recorded the server-selected key exchange group for each server
      name when processing the ServerHello, before the server was authenticated.
      This record is used to predict which key share to send in future ClientHellos
      to that server name.
      
      A network attacker can answer one connection attempt with a ServerHello
      selecting X25519, and then close the connection. Later ClientHellos for that
      name carry only an X25519 key share, while `supported_groups` still lists
      X25519MLKEM768 first. If the genuine server accepts that share rather than
      sending a HelloRetryRequest for its preferred hybrid group, the handshake
      completes with classical-only key exchange and no error. This persists for as
      long as the entry stays in the client's session store.
      
      The impact is loss of post-quantum confidentiality for affected connections
      ("harvest now, decrypt later"). There is no key compromise and no
      authentication bypass.
      
      The following are not affected:
      
      - connections to servers which send a HelloRetryRequest for their preferred group,
      - clients with resumption disabled (as nothing is stored), and
      - other server names.
      
      This behaviour is discussed in
      [draft-ietf-tls-key-share-prediction](https://www.ietf.org/archive/id/draft-ietf-tls-key-share-prediction-01.html#name-security-considerations),
      so the practical impact depends largely on server policy.
    ├ Announcement: https://github.com/rustls/rustls/pull/3355
    ├ Solution: Upgrade to >=0.23.46 (try `cargo update -p rustls`)
    ├ rustls v0.23.37
      ├── hyper-rustls v0.27.7
      │   └── libdd-common v8.0.0
      │       ├── libdd-capabilities-impl v7.0.0
      │       │   ├── (dev) libdd-ffe v2.0.0
      │       │   │   └── libdd-ffe-ffi v1.0.2
      │       │   └── libdd-remote-config v7.0.0
      │       │       ├── libdd-ffe v2.0.0 (*)
      │       │       └── (dev) libdd-remote-config v7.0.0 (*)
      │       ├── libdd-common-ffi v46.1.0
      │       │   └── libdd-ffe-ffi v1.0.2 (*)
      │       ├── libdd-ffe v2.0.0 (*)
      │       └── libdd-remote-config v7.0.0 (*)
      ├── libdd-common v8.0.0 (*)
      ├── rustls-platform-verifier v0.6.2
      │   └── libdd-common v8.0.0 (*)
      └── tokio-rustls v0.26.0
          └── hyper-rustls v0.27.7 (*)

error[unsound]: `AsyncFd::new` and related constructors violate I/O safety
    ┌─ /home/runner/work/libdatadog/libdatadog/Cargo.lock:246:1
    │
246 │ tokio 1.49.0 registry+https://github.com/rust-lang/crates.io-index
    │ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ unsound advisory detected
    │
    ├ ID: RUSTSEC-2026-0386
    ├ Advisory: https://rustsec.org/advisories/RUSTSEC-2026-0386
    ├ `tokio::io::unix::AsyncFd` provides safe constructors (`new`, `with_interest`, `try_new`, and `try_with_interest`) that accept any `T: AsRawFd` rather than requiring proof of ownership or validity of the file descriptor. Because `RawFd` implements `AsRawFd`, safe code can construct an `AsyncFd` from an arbitrary or closed file descriptor, or close/replace the underlying file descriptor while the `AsyncFd` is still alive.
      
      This violates I/O safety in two ways:
      
      1. `AsyncFd<T>` implements `AsFd` for all `T: AsRawFd` by calling `BorrowedFd::borrow_raw(self.as_raw_fd())`, allowing safe code to obtain an arbitrary `BorrowedFd<'_>` without owning the file descriptor.
      2. Tokio's I/O driver stores a pointer to an internal allocation in the `epoll`/`kqueue` user data for registered file descriptors. If the underlying file descriptor is duplicated and then closed while an `AsyncFd` is alive, and the OS reuses that file descriptor number for a another IO resource on the same runtime, dropping the first `AsyncFd` deregisters the reused file descriptor number and frees the first while the original file description remains registered in `epoll`, leading to a use-after-free when a readiness event fires.
      
      Starting in Tokio 1.51.6, 1.53.3, and 1.54.0, `AsyncFd::new`, `AsyncFd::with_interest`, `AsyncFd::try_new`, and `AsyncFd::try_with_interest` are deprecated in favor of the `unsafe` constructors `AsyncFd::register` and `AsyncFd::register_with_interest`. Callers must ensure that the file descriptor remains valid, open, and unchanged for the lifetime of the `AsyncFd`.
    ├ Announcement: https://github.com/tokio-rs/tokio/pull/8587
    ├ Solution: Upgrade to >=1.51.6, <1.52.0 OR >=1.53.3 (try `cargo update -p tokio`)
    ├ tokio v1.49.0
      ├── h2 v0.4.19
      │   └── hyper v1.11.1
      │       ├── httpmock v0.8.0-alpha.1
      │       │   └── (dev) libdd-ffe v2.0.0
      │       │       └── libdd-ffe-ffi v1.0.2
      │       ├── hyper-rustls v0.27.7
      │       │   └── libdd-common v8.0.0
      │       │       ├── libdd-capabilities-impl v7.0.0
      │       │       │   ├── (dev) libdd-ffe v2.0.0 (*)
      │       │       │   └── libdd-remote-config v7.0.0
      │       │       │       ├── libdd-ffe v2.0.0 (*)
      │       │       │       └── (dev) libdd-remote-config v7.0.0 (*)
      │       │       ├── libdd-common-ffi v46.1.0
      │       │       │   └── libdd-ffe-ffi v1.0.2 (*)
      │       │       ├── libdd-ffe v2.0.0 (*)
      │       │       └── libdd-remote-config v7.0.0 (*)
      │       ├── hyper-util v0.1.17
      │       │   ├── httpmock v0.8.0-alpha.1 (*)
      │       │   ├── hyper-rustls v0.27.7 (*)
      │       │   ├── libdd-common v8.0.0 (*)
      │       │   └── libdd-remote-config v7.0.0 (*)
      │       ├── libdd-common v8.0.0 (*)
      │       ├── libdd-common-ffi v46.1.0 (*)
      │       └── libdd-remote-config v7.0.0 (*)
      ├── httpmock v0.8.0-alpha.1 (*)
      ├── hyper v1.11.1 (*)
      ├── hyper-rustls v0.27.7 (*)
      ├── hyper-util v0.1.17 (*)
      ├── (dev) libdd-capabilities-impl v7.0.0 (*)
      ├── libdd-common v8.0.0 (*)
      ├── libdd-remote-config v7.0.0 (*)
      ├── (dev) libdd-trace-protobuf v5.1.0
      │   └── libdd-remote-config v7.0.0 (*)
      ├── tokio-rustls v0.26.0
      │   └── hyper-rustls v0.27.7 (*)
      └── tokio-util v0.7.12
          ├── h2 v0.4.19 (*)
          └── libdd-remote-config v7.0.0 (*)

advisories FAILED, bans ok, sources ok

📦 libdd-ffe - 7 error(s)

Show output 
error[unmaintained]: Bincode is unmaintained
   ┌─ /home/runner/work/libdatadog/libdatadog/Cargo.lock:22:1
   │
22 │ bincode 1.3.3 registry+https://github.com/rust-lang/crates.io-index
   │ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ unmaintained advisory detected
   │
   ├ ID: RUSTSEC-2025-0141
   ├ Advisory: https://rustsec.org/advisories/RUSTSEC-2025-0141
   ├ Due to a doxxing and harassment incident, the bincode team has taken the decision to cease development permanently.
     
     The team considers version 1.3.3 a complete version of bincode that is not in need of any updates.
     
     ## Alternatives to consider
     
     * [wincode](https://crates.io/crates/wincode)
     * [postcard](https://crates.io/crates/postcard)
     * [bitcode](https://crates.io/crates/bitcode)
     * [rkyv](https://crates.io/crates/rkyv)
   ├ Announcement: https://git.sr.ht/~stygianentity/bincode/tree/v3.0/item/README.md
   ├ Solution: No safe upgrade is available!
   ├ bincode v1.3.3
     └── (dev) libdd-ffe v2.0.0

error[unsound]: Potential use-after-free due to lack of panic safety in `LruCache::pop()`
    ┌─ /home/runner/work/libdatadog/libdatadog/Cargo.lock:128:1
    │
128 │ lru 0.16.4 registry+https://github.com/rust-lang/crates.io-index
    │ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ unsound advisory detected
    │
    ├ ID: RUSTSEC-2026-0253
    ├ Advisory: https://rustsec.org/advisories/RUSTSEC-2026-0253
    ├ `LruCache::pop()` in `lru` was not panic-safe. If the `Drop` implementation of a stored key panics during `pop()`, `self.detach()` is never called, leaving dangling pointers in the internal doubly-linked list.
      
      A subsequent cache operation that triggers eviction can then dereference these dangling pointers:
      - The node is freed from the map, but remains linked in the LRU list due to the skipped `detach()` call
      - When a new insertion causes eviction, the LRU traversal encounters the dangling pointer
      - This results in a write to already-freed memory during the eviction process
      
      ## Impact
      
      - **CWE-416 (Use-After-Free):** memory corruption when subsequent cache operations access freed node pointers in the linked list
      - **CWE-415 (Double Free):** potential heap corruption when the same memory is freed multiple times
      
      Both types of undefined behavior can be invoked in safe Rust, but only if unwinding panics are enabled and `std::panic::catch_unwind` is used with key types that have potentially-panicking `Drop` implementations.
      
      ## Fix
      
      Fixed in `lru` 0.18.2 by detaching the node from the linked list before freeing it and dropping the key ([lru-rs#238](https://github.com/jeromefroe/lru-rs/pull/238)).
    ├ Announcement: https://github.com/jeromefroe/lru-rs/pull/238
    ├ Solution: Upgrade to >=0.18.2 (try `cargo update -p lru`)
    ├ lru v0.16.4
      └── libdd-ffe v2.0.0

error[vulnerability]: TLS 1.3 handshake messages incorrectly accepted across encryption level boundaries
    ┌─ /home/runner/work/libdatadog/libdatadog/Cargo.lock:170:1
    │
170 │ rustls 0.23.37 registry+https://github.com/rust-lang/crates.io-index
    │ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ security vulnerability detected
    │
    ├ ID: RUSTSEC-2026-0285
    ├ Advisory: https://rustsec.org/advisories/RUSTSEC-2026-0285
    ├ Rustls accepted TLS 1.3 handshake messages sent at the wrong encryption level
      when they followed a key-changing message in the same record. For example,
      a plaintext `EncryptedExtensions` message packed into the same record as the
      `ServerHello` was accepted.
      
      RFC 8446 section 5.1 requires that handshake messages do not span key changes,
      and that implementations terminate the connection with an "unexpected_message"
      alert if they do.
      
      The handshake transcript is still authenticated, so a network-position attacker
      cannot use this to alter or complete a handshake; the practical effect is that
      a peer could send handshake messages that should be encrypted in plaintext
      without rustls rejecting the connection.
      
      This is functionally the same bug as Go's
      [GO-2026-4340](https://pkg.go.dev/vuln/GO-2026-4340) (CVE-2025-61730).
    ├ Announcement: https://github.com/rustls/rustls/pull/3265
    ├ Solution: Upgrade to >=0.23.45 (try `cargo update -p rustls`)
    ├ rustls v0.23.37
      ├── hyper-rustls v0.27.7
      │   └── libdd-common v8.0.0
      │       ├── libdd-capabilities-impl v7.0.0
      │       │   ├── (dev) libdd-ffe v2.0.0
      │       │   └── libdd-remote-config v7.0.0
      │       │       ├── libdd-ffe v2.0.0 (*)
      │       │       └── (dev) libdd-remote-config v7.0.0 (*)
      │       ├── libdd-ffe v2.0.0 (*)
      │       └── libdd-remote-config v7.0.0 (*)
      ├── libdd-common v8.0.0 (*)
      ├── rustls-platform-verifier v0.6.2
      │   └── libdd-common v8.0.0 (*)
      └── tokio-rustls v0.26.0
          └── hyper-rustls v0.27.7 (*)

error[vulnerability]: `Debug` impls for client session values expose TLS secrets
    ┌─ /home/runner/work/libdatadog/libdatadog/Cargo.lock:170:1
    │
170 │ rustls 0.23.37 registry+https://github.com/rust-lang/crates.io-index
    │ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ security vulnerability detected
    │
    ├ ID: RUSTSEC-2026-0346
    ├ Advisory: https://rustsec.org/advisories/RUSTSEC-2026-0346
    ├ The `Debug` implementations of `Tls12ClientSessionValue` and
      `Tls13ClientSessionValue` were derived, and so included the TLS 1.2 master
      secret or TLS 1.3 resumption PSK in their output.
      
      Custom `ClientSessionStore` implementations that format these values with
      `Debug` (for example, when logging) could disclose these secrets to readers of
      the logs. This conflicts with rustls's
      [key-material redaction policy](https://github.com/rustls/rustls/blob/99f2358cae2954837dbb866faf6727de75489ab9/SECURITY.md#boundary-public-api).
      
      Ordinary connection and built-in session cache `Debug` output is redacted and
      is not affected.
    ├ Announcement: https://github.com/rustls/rustls/pull/3299
    ├ Solution: Upgrade to >=0.23.46 (try `cargo update -p rustls`)
    ├ rustls v0.23.37
      ├── hyper-rustls v0.27.7
      │   └── libdd-common v8.0.0
      │       ├── libdd-capabilities-impl v7.0.0
      │       │   ├── (dev) libdd-ffe v2.0.0
      │       │   └── libdd-remote-config v7.0.0
      │       │       ├── libdd-ffe v2.0.0 (*)
      │       │       └── (dev) libdd-remote-config v7.0.0 (*)
      │       ├── libdd-ffe v2.0.0 (*)
      │       └── libdd-remote-config v7.0.0 (*)
      ├── libdd-common v8.0.0 (*)
      ├── rustls-platform-verifier v0.6.2
      │   └── libdd-common v8.0.0 (*)
      └── tokio-rustls v0.26.0
          └── hyper-rustls v0.27.7 (*)

error[vulnerability]: Brotli certificate decompression can allocate up to 1 GiB before authentication
    ┌─ /home/runner/work/libdatadog/libdatadog/Cargo.lock:170:1
    │
170 │ rustls 0.23.37 registry+https://github.com/rust-lang/crates.io-index
    │ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ security vulnerability detected
    │
    ├ ID: RUSTSEC-2026-0348
    ├ Advisory: https://rustsec.org/advisories/RUSTSEC-2026-0348
    ├ The built-in Brotli certificate decompressor (enabled by the non-default
      `brotli` crate feature) bounded the size of the decompressed output, but
      accepted a peer-selected history window. Brotli's large-window extension was
      enabled, so a five-byte header could cause an allocation of 1 GiB. A peer can
      abort a process whose allocator cannot satisfy that request.
      
      No trusted certificate is required by the attacker. Affected configurations are:
      
      - clients that enable the `brotli` feature and connect to a malicious server, and
      - servers that enable the `brotli` feature and request client certificates.
      
      Applications that do not enable the `brotli` feature are not affected.
    ├ Announcement: https://github.com/rustls/rustls/pull/3297
    ├ Solution: Upgrade to >=0.23.46 (try `cargo update -p rustls`)
    ├ rustls v0.23.37
      ├── hyper-rustls v0.27.7
      │   └── libdd-common v8.0.0
      │       ├── libdd-capabilities-impl v7.0.0
      │       │   ├── (dev) libdd-ffe v2.0.0
      │       │   └── libdd-remote-config v7.0.0
      │       │       ├── libdd-ffe v2.0.0 (*)
      │       │       └── (dev) libdd-remote-config v7.0.0 (*)
      │       ├── libdd-ffe v2.0.0 (*)
      │       └── libdd-remote-config v7.0.0 (*)
      ├── libdd-common v8.0.0 (*)
      ├── rustls-platform-verifier v0.6.2
      │   └── libdd-common v8.0.0 (*)
      └── tokio-rustls v0.26.0
          └── hyper-rustls v0.27.7 (*)

error[vulnerability]: Unauthenticated ServerHello can downgrade later connections to classical key exchange
    ┌─ /home/runner/work/libdatadog/libdatadog/Cargo.lock:170:1
    │
170 │ rustls 0.23.37 registry+https://github.com/rust-lang/crates.io-index
    │ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ security vulnerability detected
    │
    ├ ID: RUSTSEC-2026-0349
    ├ Advisory: https://rustsec.org/advisories/RUSTSEC-2026-0349
    ├ The client recorded the server-selected key exchange group for each server
      name when processing the ServerHello, before the server was authenticated.
      This record is used to predict which key share to send in future ClientHellos
      to that server name.
      
      A network attacker can answer one connection attempt with a ServerHello
      selecting X25519, and then close the connection. Later ClientHellos for that
      name carry only an X25519 key share, while `supported_groups` still lists
      X25519MLKEM768 first. If the genuine server accepts that share rather than
      sending a HelloRetryRequest for its preferred hybrid group, the handshake
      completes with classical-only key exchange and no error. This persists for as
      long as the entry stays in the client's session store.
      
      The impact is loss of post-quantum confidentiality for affected connections
      ("harvest now, decrypt later"). There is no key compromise and no
      authentication bypass.
      
      The following are not affected:
      
      - connections to servers which send a HelloRetryRequest for their preferred group,
      - clients with resumption disabled (as nothing is stored), and
      - other server names.
      
      This behaviour is discussed in
      [draft-ietf-tls-key-share-prediction](https://www.ietf.org/archive/id/draft-ietf-tls-key-share-prediction-01.html#name-security-considerations),
      so the practical impact depends largely on server policy.
    ├ Announcement: https://github.com/rustls/rustls/pull/3355
    ├ Solution: Upgrade to >=0.23.46 (try `cargo update -p rustls`)
    ├ rustls v0.23.37
      ├── hyper-rustls v0.27.7
      │   └── libdd-common v8.0.0
      │       ├── libdd-capabilities-impl v7.0.0
      │       │   ├── (dev) libdd-ffe v2.0.0
      │       │   └── libdd-remote-config v7.0.0
      │       │       ├── libdd-ffe v2.0.0 (*)
      │       │       └── (dev) libdd-remote-config v7.0.0 (*)
      │       ├── libdd-ffe v2.0.0 (*)
      │       └── libdd-remote-config v7.0.0 (*)
      ├── libdd-common v8.0.0 (*)
      ├── rustls-platform-verifier v0.6.2
      │   └── libdd-common v8.0.0 (*)
      └── tokio-rustls v0.26.0
          └── hyper-rustls v0.27.7 (*)

error[unsound]: `AsyncFd::new` and related constructors violate I/O safety
    ┌─ /home/runner/work/libdatadog/libdatadog/Cargo.lock:220:1
    │
220 │ tokio 1.49.0 registry+https://github.com/rust-lang/crates.io-index
    │ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ unsound advisory detected
    │
    ├ ID: RUSTSEC-2026-0386
    ├ Advisory: https://rustsec.org/advisories/RUSTSEC-2026-0386
    ├ `tokio::io::unix::AsyncFd` provides safe constructors (`new`, `with_interest`, `try_new`, and `try_with_interest`) that accept any `T: AsRawFd` rather than requiring proof of ownership or validity of the file descriptor. Because `RawFd` implements `AsRawFd`, safe code can construct an `AsyncFd` from an arbitrary or closed file descriptor, or close/replace the underlying file descriptor while the `AsyncFd` is still alive.
      
      This violates I/O safety in two ways:
      
      1. `AsyncFd<T>` implements `AsFd` for all `T: AsRawFd` by calling `BorrowedFd::borrow_raw(self.as_raw_fd())`, allowing safe code to obtain an arbitrary `BorrowedFd<'_>` without owning the file descriptor.
      2. Tokio's I/O driver stores a pointer to an internal allocation in the `epoll`/`kqueue` user data for registered file descriptors. If the underlying file descriptor is duplicated and then closed while an `AsyncFd` is alive, and the OS reuses that file descriptor number for a another IO resource on the same runtime, dropping the first `AsyncFd` deregisters the reused file descriptor number and frees the first while the original file description remains registered in `epoll`, leading to a use-after-free when a readiness event fires.
      
      Starting in Tokio 1.51.6, 1.53.3, and 1.54.0, `AsyncFd::new`, `AsyncFd::with_interest`, `AsyncFd::try_new`, and `AsyncFd::try_with_interest` are deprecated in favor of the `unsafe` constructors `AsyncFd::register` and `AsyncFd::register_with_interest`. Callers must ensure that the file descriptor remains valid, open, and unchanged for the lifetime of the `AsyncFd`.
    ├ Announcement: https://github.com/tokio-rs/tokio/pull/8587
    ├ Solution: Upgrade to >=1.51.6, <1.52.0 OR >=1.53.3 (try `cargo update -p tokio`)
    ├ tokio v1.49.0
      ├── h2 v0.4.19
      │   └── hyper v1.11.1
      │       ├── httpmock v0.8.0-alpha.1
      │       │   └── (dev) libdd-ffe v2.0.0
      │       ├── hyper-rustls v0.27.7
      │       │   └── libdd-common v8.0.0
      │       │       ├── libdd-capabilities-impl v7.0.0
      │       │       │   ├── (dev) libdd-ffe v2.0.0 (*)
      │       │       │   └── libdd-remote-config v7.0.0
      │       │       │       ├── libdd-ffe v2.0.0 (*)
      │       │       │       └── (dev) libdd-remote-config v7.0.0 (*)
      │       │       ├── libdd-ffe v2.0.0 (*)
      │       │       └── libdd-remote-config v7.0.0 (*)
      │       ├── hyper-util v0.1.17
      │       │   ├── httpmock v0.8.0-alpha.1 (*)
      │       │   ├── hyper-rustls v0.27.7 (*)
      │       │   ├── libdd-common v8.0.0 (*)
      │       │   └── libdd-remote-config v7.0.0 (*)
      │       ├── libdd-common v8.0.0 (*)
      │       └── libdd-remote-config v7.0.0 (*)
      ├── httpmock v0.8.0-alpha.1 (*)
      ├── hyper v1.11.1 (*)
      ├── hyper-rustls v0.27.7 (*)
      ├── hyper-util v0.1.17 (*)
      ├── (dev) libdd-capabilities-impl v7.0.0 (*)
      ├── libdd-common v8.0.0 (*)
      ├── libdd-ffe v2.0.0 (*)
      ├── libdd-remote-config v7.0.0 (*)
      ├── (dev) libdd-trace-protobuf v5.1.0
      │   ├── libdd-ffe v2.0.0 (*)
      │   └── libdd-remote-config v7.0.0 (*)
      ├── tokio-rustls v0.26.0
      │   └── hyper-rustls v0.27.7 (*)
      └── tokio-util v0.7.12
          ├── h2 v0.4.19 (*)
          └── libdd-remote-config v7.0.0 (*)

advisories FAILED, bans ok, sources ok

📦 cc_utils - ✅ No issues

📦 tools - 3 error(s)

Show output 
error[vulnerability]: Quadratic run time when checking a start tag for duplicate attribute names
   ┌─ /home/runner/work/libdatadog/libdatadog/Cargo.lock:72:1
   │
72 │ quick-xml 0.37.5 registry+https://github.com/rust-lang/crates.io-index
   │ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ security vulnerability detected
   │
   ├ ID: RUSTSEC-2026-0194
   ├ Advisory: https://rustsec.org/advisories/RUSTSEC-2026-0194
   ├ `BytesStart::attributes()` returns an `Attributes` iterator which, by default
     (`with_checks(true)`), rejects a start tag that repeats an attribute name. For
     each attribute yielded, the iterator compared the new name against every name
     seen so far in the same tag using a linear scan, so a start tag with `N`
     distinct attribute names cost `O(N²)` byte comparisons. There was no bound on
     `N` other than the size of the buffered start tag.
     
     ## Impact
     
     Any code that parses untrusted XML and iterates a start tag's attributes with
     the default duplicate check enabled can be made to spend CPU time quadratic in
     the number of attributes on a single tag. Because the check is pure computation
     with no `.await`/I/O, an I/O-based timeout on the consumer (for example a read
     or request timeout) cannot interrupt it while it runs.
     
     Measured cost of a single start tag, release build:
     
     | Attributes on one tag | Time |
     |---|---|
     | 80,000  | ~6 s   |
     | 800,000 | ~10 min |
     
     The cost grows with the square of the attribute count, so a start tag of a few
     tens of megabytes can stall a parsing thread for hours. No memory is exhausted
     and the parser does not crash; the effect is CPU exhaustion on the thread doing
     the parsing: a single crafted start tag can pin a CPU core for minutes to hours,
     denying service to that worker. A deployment that places a wall-clock bound on
     parsing, or confines it to a non-critical thread, may consider the availability
     impact lower.
     
     ## Affected code paths
     
     * `BytesStart::attributes()` / `Attributes` iterated with checks enabled (the
       default), and `BytesStart::try_get_attribute`.
     * `NsReader`, which resolves namespaces by iterating a tag's attributes and so
       reaches the same check internally.
     
     Consumers that iterate attributes with `.attributes().with_checks(false)` and do
     not use `NsReader` are not affected.
     
     This was reported as reachable by a remote, unauthenticated attacker in a
     real-world RPKI relying party (NLnet Labs Routinator) via a crafted RRDP
     `snapshot.xml`.
     
     ## Remediation
     
     Upgrade to `quick-xml >= 0.41.0`, where the duplicate check keeps the linear
     scan for start tags with a small number of attributes and switches to an `O(1)`
     hash pre-filter above a threshold, making the whole tag `O(N)`. The reported
     `AttrError::Duplicated` positions are unchanged.
     
     If upgrading is not possible and duplicate-name detection is not required,
     disable it with `.attributes().with_checks(false)` (this does not help
     `NsReader` consumers, which have no equivalent opt-out before 0.41.0).
   ├ Announcement: https://github.com/tafia/quick-xml/issues/969
   ├ Solution: Upgrade to >=0.41.0 (try `cargo update -p quick-xml`)
   ├ quick-xml v0.37.5
     └── tools v46.1.0

error[vulnerability]: Unbounded namespace-declaration allocation in `NsReader` enables memory-exhaustion denial of service
   ┌─ /home/runner/work/libdatadog/libdatadog/Cargo.lock:72:1
   │
72 │ quick-xml 0.37.5 registry+https://github.com/rust-lang/crates.io-index
   │ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ security vulnerability detected
   │
   ├ ID: RUSTSEC-2026-0195
   ├ Advisory: https://rustsec.org/advisories/RUSTSEC-2026-0195
   ├ `NsReader` resolves namespaces by calling `NamespaceResolver::push` for every
     `Start`/`Empty` event *before* the event is returned to the caller. `push`
     iterated all `xmlns` / `xmlns:*` attributes on the start tag and, for each one,
     appended the prefix bytes to an internal buffer and pushed a `NamespaceBinding`
     (32 bytes on 64-bit) to an internal `Vec`, with no upper bound on the number of
     declarations.
     
     ## Impact
     
     A start tag with `N` namespace declarations drove roughly `3×` the tag's byte
     size in `NamespaceResolver` heap, allocated *inside* `quick-xml` before the
     `NsReader` consumer ever received the event and could inspect or reject it. A
     consumer that bounds its *input* size therefore still cannot bound this
     allocation: an `M`-byte start tag yields on the order of `3 × M` bytes of
     resolver heap the caller never sees.
     
     On untrusted XML this lets a remote, unauthenticated attacker force large heap
     allocations with a single start tag. With several `NsReader`s running
     concurrently on independent inputs (a common server pattern), the allocations
     stack and can exhaust process memory, causing the operating system to kill the
     process (OOM). This was confirmed against a real-world RPKI relying party (NLnet
     Labs Routinator), where concurrent RRDP validation workers parsing a crafted
     `snapshot.xml` exceeded the memory limit and the process was OOM-killed.
     
     ## Affected code paths
     
     Consumers using `NsReader` (which always calls `NamespaceResolver::push` before
     yielding `Start`/`Empty`), or calling `NamespaceResolver::push` directly. A plain
     `Reader` that does not perform namespace resolution is not affected.
     
     ## Remediation
     
     Upgrade to `quick-xml >= 0.41.0`. `NamespaceResolver::push` now rejects a start
     tag that declares more than `DEFAULT_MAX_DECLARATIONS_PER_ELEMENT` (256)
     namespace bindings, returning the new `NamespaceError::TooManyDeclarations`
     instead of allocating without limit. The limit is configurable via
     `NamespaceResolver::set_max_declarations_per_element` (use `usize::MAX` to
     restore the previous unbounded behavior), and `NsReader::resolver_mut()` is
     provided to reach it.
     
     There is no clean workaround for `NsReader` consumers before 0.41.0, as the
     allocation happens inside the reader with no configuration knob to cap it.
   ├ Announcement: https://github.com/tafia/quick-xml/issues/970
   ├ Solution: Upgrade to >=0.41.0 (try `cargo update -p quick-xml`)
   ├ quick-xml v0.37.5
     └── tools v46.1.0

error[unsound]: `AsyncFd::new` and related constructors violate I/O safety
    ┌─ /home/runner/work/libdatadog/libdatadog/Cargo.lock:102:1
    │
102 │ tokio 1.49.0 registry+https://github.com/rust-lang/crates.io-index
    │ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ unsound advisory detected
    │
    ├ ID: RUSTSEC-2026-0386
    ├ Advisory: https://rustsec.org/advisories/RUSTSEC-2026-0386
    ├ `tokio::io::unix::AsyncFd` provides safe constructors (`new`, `with_interest`, `try_new`, and `try_with_interest`) that accept any `T: AsRawFd` rather than requiring proof of ownership or validity of the file descriptor. Because `RawFd` implements `AsRawFd`, safe code can construct an `AsyncFd` from an arbitrary or closed file descriptor, or close/replace the underlying file descriptor while the `AsyncFd` is still alive.
      
      This violates I/O safety in two ways:
      
      1. `AsyncFd<T>` implements `AsFd` for all `T: AsRawFd` by calling `BorrowedFd::borrow_raw(self.as_raw_fd())`, allowing safe code to obtain an arbitrary `BorrowedFd<'_>` without owning the file descriptor.
      2. Tokio's I/O driver stores a pointer to an internal allocation in the `epoll`/`kqueue` user data for registered file descriptors. If the underlying file descriptor is duplicated and then closed while an `AsyncFd` is alive, and the OS reuses that file descriptor number for a another IO resource on the same runtime, dropping the first `AsyncFd` deregisters the reused file descriptor number and frees the first while the original file description remains registered in `epoll`, leading to a use-after-free when a readiness event fires.
      
      Starting in Tokio 1.51.6, 1.53.3, and 1.54.0, `AsyncFd::new`, `AsyncFd::with_interest`, `AsyncFd::try_new`, and `AsyncFd::try_with_interest` are deprecated in favor of the `unsafe` constructors `AsyncFd::register` and `AsyncFd::register_with_interest`. Callers must ensure that the file descriptor remains valid, open, and unchanged for the lifetime of the `AsyncFd`.
    ├ Announcement: https://github.com/tokio-rs/tokio/pull/8587
    ├ Solution: Upgrade to >=1.51.6, <1.52.0 OR >=1.53.3 (try `cargo update -p tokio`)
    ├ tokio v1.49.0
      └── (dev) libdd-common v8.0.0
          └── tools v46.1.0

advisories FAILED, bans ok, sources ok

Updated: 2026-10-30 16:22:24 UTC | Commit: f5abe08 | dependency-check job results

@datadog-datadog-prod-us1

datadog-datadog-prod-us1 Bot commented Oct 30, 2026

Copy link
Copy Markdown
Contributor

Tests

🎉 All green!

🧪 All tests passed
❄️ No new flaky tests detected

🎯 Code Coverage (details)
• Patch Coverage: 88.66%
• Overall Coverage: 82.39% (+0.09%)

This comment will be updated automatically if new data arrives.
🔗 Commit SHA: 288685b | Docs | View more details | Give us feedback!

@dd-octo-sts

dd-octo-sts Bot commented Oct 30, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

Artifact Size Benchmark Report

aarch64-alpine-linux-musl
Artifact Baseline Commit Change
/aarch64-alpine-linux-musl/lib/libdatadog_profiling.so 9.08 MB 9.21 MB +1.37% (+128.23 KB) ⚠️
/aarch64-alpine-linux-musl/lib/libdatadog_profiling.a 96.62 MB 96.21 MB --.42% (-422.42 KB) 💪
aarch64-unknown-linux-gnu
Artifact Baseline Commit Change
/aarch64-unknown-linux-gnu/lib/libdatadog_profiling.so 12.29 MB 12.38 MB +.76% (+96.18 KB) 🔍
/aarch64-unknown-linux-gnu/lib/libdatadog_profiling.a 107.97 MB 107.33 MB --.58% (-651.51 KB) 💪
libdatadog-x64-windows
Artifact Baseline Commit Change
/libdatadog-x64-windows/debug/dynamic/datadog_profiling_ffi.dll 29.19 MB 29.64 MB +1.53% (+458.50 KB) ⚠️
/libdatadog-x64-windows/debug/dynamic/datadog_profiling_ffi.lib 99.56 KB 99.02 KB --.54% (-556 B) 💪
/libdatadog-x64-windows/debug/dynamic/datadog_profiling_ffi.pdb 193.18 MB 195.20 MB +1.04% (+2.02 MB) ⚠️
/libdatadog-x64-windows/debug/static/datadog_profiling_ffi.lib 818.75 MB 823.38 MB +.56% (+4.63 MB) 🔍
/libdatadog-x64-windows/release/dynamic/datadog_profiling_ffi.dll 9.81 MB 9.72 MB --.88% (-89.00 KB) 💪
/libdatadog-x64-windows/release/dynamic/datadog_profiling_ffi.lib 99.56 KB 99.02 KB --.54% (-556 B) 💪
/libdatadog-x64-windows/release/dynamic/datadog_profiling_ffi.pdb 27.59 MB 27.73 MB +.48% (+136.00 KB) 🔍
/libdatadog-x64-windows/release/static/datadog_profiling_ffi.lib 55.87 MB 55.81 MB --.11% (-63.76 KB) 💪
libdatadog-x86-windows
Artifact Baseline Commit Change
/libdatadog-x86-windows/debug/dynamic/datadog_profiling_ffi.dll 25.61 MB 26.13 MB +2.01% (+529.50 KB) ⚠️
/libdatadog-x86-windows/debug/dynamic/datadog_profiling_ffi.lib 101.12 KB 100.56 KB --.55% (-572 B) 💪
/libdatadog-x86-windows/debug/dynamic/datadog_profiling_ffi.pdb 198.87 MB 200.81 MB +.97% (+1.93 MB) 🔍
/libdatadog-x86-windows/debug/static/datadog_profiling_ffi.lib 806.98 MB 808.67 MB +.20% (+1.68 MB) 🔍
/libdatadog-x86-windows/release/dynamic/datadog_profiling_ffi.dll 7.60 MB 7.51 MB --1.22% (-95.50 KB) 💪
/libdatadog-x86-windows/release/dynamic/datadog_profiling_ffi.lib 101.12 KB 100.56 KB --.55% (-572 B) 💪
/libdatadog-x86-windows/release/dynamic/datadog_profiling_ffi.pdb 29.73 MB 29.80 MB +.21% (+64.00 KB) 🔍
/libdatadog-x86-windows/release/static/datadog_profiling_ffi.lib 52.81 MB 52.67 MB --.25% (-138.04 KB) 💪
x86_64-alpine-linux-musl
Artifact Baseline Commit Change
/x86_64-alpine-linux-musl/lib/libdatadog_profiling.a 86.54 MB 86.75 MB +.24% (+220.00 KB) 🔍
/x86_64-alpine-linux-musl/lib/libdatadog_profiling.so 10.12 MB 10.02 MB --1.00% (-103.74 KB) 💪
x86_64-unknown-linux-gnu
Artifact Baseline Commit Change
/x86_64-unknown-linux-gnu/lib/libdatadog_profiling.a 102.44 MB 102.30 MB --.13% (-137.86 KB) 💪
/x86_64-unknown-linux-gnu/lib/libdatadog_profiling.so 12.37 MB 12.25 MB --.96% (-121.67 KB) 💪

@pr-commenter

pr-commenter Bot commented Oct 30, 2026 •
edited
Loading

Copy link
Copy Markdown

Benchmarks

Comparison

Benchmark execution time: 2026-10-30 16:50:45

Comparing candidate commit 288685b in PR branch typo/dependent-flags-returned-batch-v46 with baseline commit 9cb812d in branch helen/ffe-dependent-flags-v46.

📊 Benchmarking dashboard

Found 1 performance improvements and 3 performance regressions! Performance is the same for 158 metrics, 0 unstable metrics.

Explanation

This is an A/B test comparing a candidate commit's performance against that of a baseline commit. Performance changes are noted in the tables below as:

  • 🟩 = significantly better candidate vs. baseline
  • 🟥 = significantly worse candidate vs. baseline

We compute a confidence interval (CI) over the relative difference of means between metrics from the candidate and baseline commits, considering the baseline as the reference.

If the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD), the change is considered significant.

Feel free to reach out to #apm-benchmarking-platform on Slack if you have any questions.

More details about the CI and significant changes

You can imagine this CI as a range of values that is likely to contain the true difference of means between the candidate and baseline commits.

CIs of the difference of means are often centered around 0%, because often changes are not that big:

---------------------------------(------|---^--------)-------------------------------->
                              -0.6%    0%  0.3%     +1.2%
                                 |          |        |
         lower bound of the CI --'          |        |
sample mean (center of the CI) -------------'        |
         upper bound of the CI ----------------------'

As described above, a change is considered significant if the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD).

For instance, for an execution time metric, this confidence interval indicates a significantly worse performance:

----------------------------------------|---------|---(---------^---------)---------->
                                       0%        1%  1.3%      2.2%      3.1%
                                                  |   |         |         |
       significant impact threshold --------------'   |         |         |
                      lower bound of CI --------------'         |         |
       sample mean (center of the CI) --------------------------'         |
                      upper bound of CI ----------------------------------'

scenario:vec_map/as_deduped_map/needs_dedup_1_in_10/16

  • 🟩 execution_time [-16.077ns; -15.872ns] or [-4.541%; -4.484%]

scenario:vec_map/as_deduped_map/needs_dedup_1_in_2/16

  • 🟥 execution_time [+20.653ns; +20.890ns] or [+4.827%; +4.882%]

scenario:vec_map/as_deduped_map/needs_dedup_1_in_4/128

  • 🟥 execution_time [+328.634ns; +330.215ns] or [+12.304%; +12.363%]

scenario:vec_map/as_deduped_map/needs_dedup_1_in_4/64

  • 🟥 execution_time [+246.606ns; +247.692ns] or [+18.056%; +18.135%]

Benchmark execution time: 2026-10-30 16:48:14

Comparing candidate commit 288685b in PR branch typo/dependent-flags-returned-batch-v46 with baseline commit 9cb812d in branch helen/ffe-dependent-flags-v46.

📊 Benchmarking dashboard

Found 7 performance improvements and 0 performance regressions! Performance is the same for 85 metrics, 0 unstable metrics.

Explanation

This is an A/B test comparing a candidate commit's performance against that of a baseline commit. Performance changes are noted in the tables below as:

  • 🟩 = significantly better candidate vs. baseline
  • 🟥 = significantly worse candidate vs. baseline

We compute a confidence interval (CI) over the relative difference of means between metrics from the candidate and baseline commits, considering the baseline as the reference.

If the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD), the change is considered significant.

Feel free to reach out to #apm-benchmarking-platform on Slack if you have any questions.

More details about the CI and significant changes

You can imagine this CI as a range of values that is likely to contain the true difference of means between the candidate and baseline commits.

CIs of the difference of means are often centered around 0%, because often changes are not that big:

---------------------------------(------|---^--------)-------------------------------->
                              -0.6%    0%  0.3%     +1.2%
                                 |          |        |
         lower bound of the CI --'          |        |
sample mean (center of the CI) -------------'        |
         upper bound of the CI ----------------------'

As described above, a change is considered significant if the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD).

For instance, for an execution time metric, this confidence interval indicates a significantly worse performance:

----------------------------------------|---------|---(---------^---------)---------->
                                       0%        1%  1.3%      2.2%      3.1%
                                                  |   |         |         |
       significant impact threshold --------------'   |         |         |
                      lower bound of CI --------------'         |         |
       sample mean (center of the CI) --------------------------'         |
                      upper bound of CI ----------------------------------'

scenario:thread_cpu/profiler_attached/fast_path_noop/16

  • 🟩 execution_time [-0.677ns; -0.666ns] or [-4.353%; -4.288%]

scenario:thread_cpu/profiler_attached/fast_path_noop/256

  • 🟩 execution_time [-0.679ns; -0.669ns] or [-4.366%; -4.301%]

scenario:thread_cpu/profiler_attached/fast_path_noop/4096

  • 🟩 execution_time [-0.676ns; -0.666ns] or [-4.351%; -4.285%]

scenario:thread_cpu/profiler_attached/fast_path_noop/64

  • 🟩 execution_time [-0.671ns; -0.662ns] or [-4.320%; -4.261%]

scenario:thread_cpu/profiler_attached/fast_path_noop/65536

  • 🟩 execution_time [-0.672ns; -0.661ns] or [-4.322%; -4.253%]

scenario:thread_cpu/profiler_attached/fast_path_system/4096

  • 🟩 execution_time [-15.235ns; -15.129ns] or [-14.373%; -14.273%]

scenario:thread_cpu/profiler_attached/slow_path_system/4096

  • 🟩 execution_time [-10.507ns; -10.409ns] or [-6.834%; -6.770%]

Benchmark execution time: 2026-10-30 16:44:08

Comparing candidate commit 288685b in PR branch typo/dependent-flags-returned-batch-v46 with baseline commit 9cb812d in branch helen/ffe-dependent-flags-v46.

📊 Benchmarking dashboard

Found 3 performance improvements and 6 performance regressions! Performance is the same for 69 metrics, 24 unstable metrics.

Explanation

This is an A/B test comparing a candidate commit's performance against that of a baseline commit. Performance changes are noted in the tables below as:

  • 🟩 = significantly better candidate vs. baseline
  • 🟥 = significantly worse candidate vs. baseline

We compute a confidence interval (CI) over the relative difference of means between metrics from the candidate and baseline commits, considering the baseline as the reference.

If the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD), the change is considered significant.

Feel free to reach out to #apm-benchmarking-platform on Slack if you have any questions.

More details about the CI and significant changes

You can imagine this CI as a range of values that is likely to contain the true difference of means between the candidate and baseline commits.

CIs of the difference of means are often centered around 0%, because often changes are not that big:

---------------------------------(------|---^--------)-------------------------------->
                              -0.6%    0%  0.3%     +1.2%
                                 |          |        |
         lower bound of the CI --'          |        |
sample mean (center of the CI) -------------'        |
         upper bound of the CI ----------------------'

As described above, a change is considered significant if the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD).

For instance, for an execution time metric, this confidence interval indicates a significantly worse performance:

----------------------------------------|---------|---(---------^---------)---------->
                                       0%        1%  1.3%      2.2%      3.1%
                                                  |   |         |         |
       significant impact threshold --------------'   |         |         |
                      lower bound of CI --------------'         |         |
       sample mean (center of the CI) --------------------------'         |
                      upper bound of CI ----------------------------------'

scenario:datadog_sample_span/name_pattern_rule_not_matching/wall_time

  • 🟥 execution_time [+8.324ns; +8.463ns] or [+6.289%; +6.394%]

scenario:dependent_flags/legacy

  • 🟥 execution_time [+5.810ns; +5.896ns] or [+4.970%; +5.044%]

scenario:dependent_flags/memoized_16_references

  • 🟩 execution_time [-553.735ns; -552.052ns] or [-18.385%; -18.329%]

scenario:dependent_flags/one_prerequisite

  • 🟥 execution_time [+67.620ns; +68.145ns] or [+9.411%; +9.484%]

scenario:single_flag_killswitch/rules-based

  • 🟥 execution_time [+8.236ns; +9.117ns] or [+4.020%; +4.450%]

scenario:trace_buffer/2_senders/no_delay

  • 🟩 execution_time [-329.414µs; -313.882µs] or [-15.789%; -15.044%]
  • 🟩 throughput [+153711.650op/s; +162374.233op/s] or [+17.815%; +18.819%]

scenario:trace_buffer/8_senders/no_delay

  • 🟥 execution_time [+277.741µs; +294.136µs] or [+4.937%; +5.228%]
  • 🟥 throughput [-63544.888op/s; -60177.594op/s] or [-4.965%; -4.702%]

Unstable benchmarks

These benchmarks have a confidence interval too wide to call a change; treat them as noise rather than signal.

scenario:datadog_sample_span/complex_rule_matching/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+555.735%; -555.735%]

scenario:datadog_sample_span/complex_rule_partial_match/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+551.350%; -553.676%]

scenario:datadog_sample_span/many_attributes_tag_rule/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+555.735%; -555.735%]

scenario:datadog_sample_span/multiple_rules_first_match/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+559.011%; -557.280%]

scenario:datadog_sample_span/multiple_rules_last_match/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+559.144%; -557.343%]

scenario:datadog_sample_span/name_pattern_rule_matching/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+555.735%; -555.735%]

scenario:datadog_sample_span/name_pattern_rule_not_matching/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+539.322%; -548.081%]

scenario:datadog_sample_span/parent_not_sampled_short_circuit/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+565.655%; -560.429%]

scenario:datadog_sample_span/parent_sampled_short_circuit/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+556.658%; -556.169%]

scenario:datadog_sample_span/resource_pattern_rule_matching/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+552.176%; -554.063%]

scenario:datadog_sample_span/resource_pattern_rule_not_matching/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+564.293%; -559.782%]

scenario:datadog_sample_span/root_span_no_rules/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+557.736%; -556.678%]

scenario:datadog_sample_span/service_rule_matching/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+555.735%; -555.735%]

scenario:datadog_sample_span/service_rule_not_matching/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+553.514%; -554.691%]

scenario:datadog_sample_span/tag_rule_matching/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+559.306%; -557.419%]

scenario:datadog_sample_span/tag_rule_not_matching/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+560.559%; -558.012%]

scenario:glob_matcher/ascii_case_insensitive_match/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+554.501%; -555.154%]

scenario:glob_matcher/ascii_exact_match/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+555.735%; -555.735%]

scenario:glob_matcher/ascii_exact_miss/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+568.794%; -561.925%]

scenario:glob_matcher/ascii_wildcard_backtrack_match/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+555.735%; -555.735%]

scenario:glob_matcher/ascii_wildcard_heavy_backtrack/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+558.350%; -556.967%]

scenario:glob_matcher/ascii_wildcard_question_match/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+560.519%; -557.992%]

scenario:glob_matcher/ascii_wildcard_star_match/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+558.084%; -556.842%]

scenario:glob_matcher/star_short_circuit/allocated_bytes

  • unstable execution_time [-0.000ns; +0.000ns] or [+556.886%; -556.277%]

Candidate

Omitted due to size.

Baseline

Omitted due to size.

@typotter

typotter commented Oct 30, 2026

Copy link
Copy Markdown
Author

Closing in favor of a clean implementation based directly on #2748 with the simplified unordered returned-batch contract.

@typotter typotter closed this Oct 30, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

1 participant

@typotter