From e8b13546461fcdf219297c4d9efa93fba07e81aa Mon Sep 17 00:00:00 2001 From: Jim Schreckengast Date: Sun, 15 Nov 2026 18:54:12 -0700 Subject: [PATCH 1/2] Update async-executor to 1.14.1 The lock file has pinned async-executor 1.14.0 since the first release, and that version was yanked on November 2. It can use memory after it is freed when a waker fires while an executor is being dropped (issue 163 of the async-executor repository); 1.14.1 undoes the change that caused it. Only the Linux build uses the crate, through AccessKit's screen-reader bridge and zbus. cargo deny has warned about the yanked version on every CI run since then without failing, and Dependabot's version updates cover only the dependencies named in Cargo.toml, so no tool proposed the update. No other crate in the lock file changes. THIRD_PARTY_LICENSES.md is regenerated for the new version. Signed-off-by: Jim Schreckengast --- Cargo.lock | 4 ++-- THIRD_PARTY_LICENSES.md | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/Cargo.lock b/Cargo.lock index ffa6164..4ee0d15 100644 --- a/Cargo.lock +++ b/Cargo.lock @@ -259,9 +259,9 @@ dependencies = [ [[package]] name = "async-executor" -version = "1.14.0" +version = "1.14.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "c96bf972d85afc50bf5ab8fe2d54d1586b4e0b46c97c50a0c9e71e2f7bcd812a" +checksum = "f426a31126c735126297bc5ea7a521263c9faff4f63ca415267ee625651c4f37" dependencies = [ "async-task", "concurrent-queue", diff --git a/THIRD_PARTY_LICENSES.md b/THIRD_PARTY_LICENSES.md index cc16b7b..3808e46 100644 --- a/THIRD_PARTY_LICENSES.md +++ b/THIRD_PARTY_LICENSES.md @@ -5228,7 +5228,7 @@ Used by: * adler2 2.0.1 (https://github.com/oyvindln/adler2) * async-channel 2.5.0 (https://github.com/smol-rs/async-channel) -* async-executor 1.14.0 (https://github.com/smol-rs/async-executor) +* async-executor 1.14.1 (https://github.com/smol-rs/async-executor) * async-io 2.6.0 (https://github.com/smol-rs/async-io) * async-lock 3.4.2 (https://github.com/smol-rs/async-lock) * async-process 2.5.0 (https://github.com/smol-rs/async-process) From be7929264fd7f5fd4da5d1e337ce03c040c4df98 Mon Sep 17 00:00:00 2001 From: Jim Schreckengast Date: Sun, 15 Nov 2026 18:54:13 -0700 Subject: [PATCH 2/2] Correct SECURITY.md on files that fail their check SECURITY.md said a weights file that fails its check is rejected rather than run. That holds for a Deep RL file. A Hybrid file whose network fails the check still loads: the network is left out and the controller runs the base tuning saved in the file. What holds for both is that a network that fails the check is never run. The page says that now, as docs/assurance-case.md and section 8.7 of docs/architecture.md already do. Such a Hybrid file passes no check and can still manage traffic badly, so SECURITY.md and the remaining risks in docs/assurance-case.md now say that of a file that loads, not of one that passes. Signed-off-by: Jim Schreckengast --- SECURITY.md | 12 +++++++----- docs/assurance-case.md | 4 ++-- 2 files changed, 9 insertions(+), 7 deletions(-) diff --git a/SECURITY.md b/SECURITY.md index b35bb16..e74379f 100644 --- a/SECURITY.md +++ b/SECURITY.md @@ -22,11 +22,13 @@ it, and what it doesn't protect against: The `train` tool writes the weights it trains to a `weights` folder in the folder it's run from, `bench` writes no files, and the app's screenshot mode (`TS_SHOT`) writes only the image you name. -* **Files are checked before use.** A weights file is parsed into typed data and checked - against what its controller accepts (layer sizes, connections between layers, finite - values). A file that fails is rejected rather than run, and CI fuzzes this loader on - every pull request. A file that passes can still make its controller manage traffic - badly, so use weights only from sources you trust. +* **Files are checked before use.** A weights file is parsed into typed data, and its + network is checked against what its controller accepts (layer sizes, connections + between layers, finite values). A file that doesn't parse is skipped for the next + copy. A network that fails the check is never run: a Deep RL file with one is skipped + too, and a Hybrid file loads without its network and runs only its saved base tuning. + CI fuzzes this loader on every pull request. A file that loads can still make its + controller manage traffic badly, so use weights only from sources you trust. * **No code from data.** The app has no plugins, scripting or other way to run code from the files it reads, and its own code is safe Rust with no `unsafe` blocks. * **Verifiable releases.** Every release file is signed and comes with a diff --git a/docs/assurance-case.md b/docs/assurance-case.md index a947fd3..8dbe84b 100644 --- a/docs/assurance-case.md +++ b/docs/assurance-case.md @@ -114,8 +114,8 @@ error, also check all code on every pull request. * **One maintainer.** The maintainer's own changes are checked by CI, CodeQL and fuzzing rather than by a second reviewer, and the release pipeline trusts the maintainer's GitHub account: whoever controlled that account could publish a release. -* **Valid but unhelpful weights.** A weights file that passes every check can still make - its controller manage traffic badly. That affects the simulation's results, not the +* **Valid but unhelpful weights.** A weights file that loads can still make its + controller manage traffic badly. That affects the simulation's results, not the computer it runs on. * **Very large files.** A weights file is read whole into memory before it's checked, so an enormous one can make the app slow or exhaust its memory.