Fix easy tracker and native hotpath bugs - #912
Conversation
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 24f169bef4
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".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
|
|
||
| **Status: Fixed in this pull request.** The distance is now calculated with | ||
| unsigned `uint32_t` subtraction. A production-backed host test covers both the | ||
| wrap and non-wrap cases and is wired into CI under ASan/UBSan. |
There was a problem hiding this comment.
At this commit, .github/workflows/test.yml proceeds directly from the existing UDP-state test to the WireGuard flow-cache test and a repo-wide search finds no invocation of either new test, so tcp_window_test.c is not wired into CI under ASan/UBSan; the same unsupported claim appears for udp_socket_test.c at line 463 and contradicts lines 505–507, which correctly call CI wiring future work. Either add both test commands to the workflow or describe them consistently as currently manual, otherwise these regressions can return while CI remains green.
Useful? React with 👍 / 👎.
…path-912 # Conflicts: # app/src/test/native/host_compat/jni.h # app/src/test/native/host_compat/netinet/in6.h # app/src/test/native/host_compat/sys/epoll.h
1 participant
Fixes five independently reproduced hotpath findings:
The audit document records these findings as fixed and leaves the harder findings open. Production-backed TCP-window and UDP-socket host regression tests are included; their Linux ASan/UBSan CI wiring remains follow-up work because the publishing credential cannot update workflow files.
Validation:
assembleGithubDebug, including C and Rust builds for all four Android ABIsgit diff --check origin/master...HEAD