Skip to content

deps(cargo): bump the rust-minor group across 1 directory with 19 updates - #243

Closed
dependabot[bot] wants to merge 1 commit into
mainfrom
dependabot/cargo/rust-minor-aa28ceb857
Closed

dependabot[bot] wants to merge 1 commit into
mainfrom
dependabot/cargo/rust-minor-aa28ceb857

Conversation

@dependabot

@dependabot dependabot Bot commented on behalf of github Nov 9, 2026 •

Copy link
Copy Markdown
Contributor

Bumps the rust-minor group with 19 updates in the / directory:

Package From To
serde_json 1.0.151 1.0.154
uuid 1.26.1 1.29.0
tokio 1.53.1 1.54.1
memmap2 0.9.11 0.9.12
log 0.4.34 0.4.35
libc 0.2.189 0.2.191
lazy_static 1.5.0 1.5.1
zstd 0.14.0 0.14.3
trybuild 1.0.121 1.0.122
proptest 1.11.0 1.12.0
syn 3.0.6 3.0.9
cxx 1.0.202 1.0.205
cxx-build 1.0.202 1.0.205
sha2 0.11.0 0.11.1
open 5.4.4 5.4.5
tree-sitter 0.27.0 0.27.1
notify 8.2.0 8.2.1
schemars 1.2.2 1.2.3
insta 1.48.0 1.49.0

Updates serde_json from 1.0.151 to 1.0.154

Release notes

Sourced from serde_json's releases.

v1.0.154

v1.0.153

v1.0.152

Commits
  • 20572a6 Release 1.0.154
  • a91bfaf Fix RandomState links in map documentation
  • 3c932c0 Merge pull request #1333 from razmser/no-std-preserve-order
  • 4b2d726 Release 1.0.153
  • 83335ee Touch up PR 1345
  • 4b4810a Resolve range_plus_one pedantic clippy lint
  • 8af71cc Merge pull request #1345 from 00200200/fix-invalid-utf8-error-column
  • 452afa2 Correct multiline raw UTF-8 error positions and simplify indexing
  • c9f1e3b Release 1.0.152
  • 854fc70 Touch up PR 1347
  • Additional commits viewable in compare view

Updates uuid from 1.26.1 to 1.29.0

Release notes

Sourced from uuid's releases.

v1.29.0

What's Changed

Full Changelog: uuid-rs/uuid@v1.28.0...v1.29.0

v1.28.0

What's Changed

New Contributors

Full Changelog: uuid-rs/uuid@v1.27.0...v1.28.0

v1.27.0

What's Changed

New Contributors

Full Changelog: uuid-rs/uuid@v1.26.1...v1.27.0

Commits
  • 4643e60 Merge pull request #920 from uuid-rs/cargo/v1.29.0
  • a3631fd prepare for 1.29.0 release
  • 304913e Merge pull request #919 from RanaPriyansh/fix/non-nil-versioned-constructors-901
  • d49f963 Reuse existing version tests for NonNilUuid constructors
  • 2f02404 Add versioned constructors to NonNilUuid
  • 276c236 Merge pull request #918 from uuid-rs/cargo/v1.28.0
  • 4b3b9f5 prepare for 1.28.0 release
  • cbf9c81 Merge pull request #916 from RanaPriyansh/fix/optional-serde-adapters-840
  • 784a022 Inline expected tokens in optional compact serde test
  • a5678c3 Simplify optional serde adapter tests to option tokens
  • Additional commits viewable in compare view

Updates tokio from 1.53.1 to 1.54.1

Release notes

Sourced from tokio's releases.

Tokio v1.54.1

1.54.1 (October 21st, 2026)

Fixed

  • time: revert "remove the runtime handle from timers" (#8692)

#8692: tokio-rs/tokio#8692

Tokio v1.54.0

1.54.0 (October 19th, 2026)

The 1.54.0 release includes all changes up to 1.53.3.

The MSRV is increased to 1.85. (#8514)

Added

  • fs: add File::set_times, File::set_modified, and fs::FileTimes (#8504)
  • net: implement Default for named_pipe::{ClientOptions, ServerOptions} (#8378)
  • runtime: add Builder::max_io_events_per_busy_tick (#8410)
  • runtime: stabilize tokio::runtime::worker_index() (#8431)
  • sync: add SemaphorePermit::semaphore() (#8482)
  • sync: add blocking_acquire methods to Semaphore (#8269)
  • sync: add watch::Receiver::is_closed (#8481)
  • wasm: support JSPI for Emscripten target (#8285, #8627)
  • wasm: support the wasm32-unknown-emscripten target (#8281, #8422, #8528, #8628)

Changed

  • fs: reject zero in File::set_max_buf_size (#8449)
  • io: make copy_buf cooperative (#8459)
  • io: panic when BufReader or BufStream is created with zero capacity (#8494)
  • io: panic when in-memory pipe is created with zero capacity (#8397)
  • io: reject zero buffer size in copy_bidirectional_with_sizes (#8454)
  • io: support vectored writes in Sink (#8663)
  • net: start UnixStream::pair and UnixDatagram::pair sockets as writable (#8444)
  • net: start accepted sockets as readable and writable (#8435)
  • net: support UCred on Fuchsia (#8257)
  • process: use mio::unix::pipe for Unix child stdio (#8603)
  • runtime: cache-pad the shared injection queue and idle state (#8622)
  • runtime: decide auto-boxing of spawned futures at monomorphization time (#8414)
  • runtime: make enable_io available whenever rt is enabled (#8387)
  • runtime: move block_in_place setup into a non-generic function (#8315)
  • runtime: move the multi-thread inject queue to its own mutex (#8382)
  • runtime: panic when max_io_events_per_tick is zero (#8476)
  • runtime: suggest .await in nested block_on panic message (#8501)
  • sync: avoid atomic reader count in broadcast slots (#8357)
  • sync: use acquire/release orderings in Notify (#8325)
  • time: remove the runtime handle from timers (#8458)

... (truncated)

Commits

Updates memmap2 from 0.9.11 to 0.9.12

Changelog

Sourced from memmap2's changelog.

[0.9.12] - 2026-10-21

Added

  • Add MmapOptions::map_if_safe() to allow creation read-only mappings of a file from safe code if the kernel can guarantee the file is immutable and large enough. @​alexlarsson
  • Add MmapOptions::no_probe_handle() to disable probing of file handles on Windows to improve performance, at the cost of no longer being able to remap later with more capabilities. @​mat-gas
  • Add support for OpenHarmony targets (*-unknown-linux-ohos). @​hiyms

Changed

  • Cache allocation granularity on Windows to prevent repeated syscalls. @​mat-gas
Commits
  • 4692b05 Merge pull request #180 from RazrFalcon/release/v0.9.12
  • 9e5c0ee Bump version to 0.9.12.
  • b98035d Merge pull request #177 from JaskRendix/extract-test-suite
  • 0986677 Merge branch 'master' into extract-test-suite
  • d26cc84 Fix map_if_safe() stub.
  • 145ee36 Silence clippy warning about trivial identical match arms.
  • b402ef2 Fix check_safe_to_map() on Apple targets.
  • b544238 Fix formatting.
  • f50e480 Move tests to test/ folder.
  • 0994bf7 Merge pull request #176 from hiyms/feat/support-ohos
  • Additional commits viewable in compare view

Updates log from 0.4.34 to 0.4.35

Release notes

Sourced from log's releases.

v0.4.35

This release doesn't include source changes. It verifies log's new CI/CD processes.

Other

  • rework CI/CD to align with Rust project standards
Changelog

Sourced from log's changelog.

0.4.35 - 2026-10-30

This release doesn't include source changes. It verifies log's new CI/CD processes.

Other

  • rework CI/CD to align with Rust project standards
Commits
  • a46450b Update CHANGELOG.md
  • a715f17 chore: release v0.4.35
  • b4e2b12 Merge pull request #747 from rust-lang/chore/release-plz-mgnt
  • 6b490b6 run fmt
  • 2c7fd52 simplify some crate release tasks for easier automation
  • 5bc5545 Merge pull request #745 from marcoienibot/release-plz-disable-semver-check
  • 4234b8c Disable release-plz semver check
  • 95918ff Merge pull request #744 from Kobzol/trusted-publishing
  • e7ed747 Fix Zizmor lints
  • e87a3c6 Setup trusted publishing
  • Additional commits viewable in compare view

Updates libc from 0.2.189 to 0.2.191

Release notes

Sourced from libc's releases.

0.2.191

Support

  • Emscripten: Support v4.0.12 to v6.0.11 (#5610)

Added

  • Apple: Add F_SETNOSIGPIPE and F_GETNOSIGPIPE (#5609)
  • Apple: add ifaliasreq and in6_aliasreq (#5557)
  • FreeBSD: Add PT_* constants (#5625)
  • Haiku: Add AI_* and NI_* constants from netdb.h (#5585)
  • Hurd: Add POSIX_SPAWN_* flags (#5578)
  • NuttX: Add AT_SYMLINK_NOFOLLOW (#5614)
  • OpenBSD: Add sensordev (#5534)
  • QuRT: Add isatty (#5618)
  • WASI: Add pthread_condattr_{get,set}clock (#5626)

Fixed

  • FreeBSD-aarch64: Fix gpregs and fpregs (#5603)

Changed

  • AIX: Restore 'c_ulong' as the type of 'statvfs.f_fsid'. (#5616)
  • Android: Make prop_info an extern type (#5561)
  • Apple: Mach API is now available on non-MacOS targets again. This should resolve some issues building ecosystem crates. (#5612), (#5636)
  • FreeBSD: Add back NOTE_PCTRLMASK with a deprecation warning (#5600)

0.2.190

There is now a single config for enabling 64-bit time_t: libc_unstable_time64. This can be set unconditionally; it opts in to 64-bit time_t on the following platforms that use 32-bit by default:

  • 32-bit Linux-GNU
  • 32-bit Linux-uClibc
  • 32-bit Linux-musl. Note that setting this flag also enables some other changes that happend in musl v1.2.
  • 32-bit Windows-GNU
  • ESP-IDF (all targets with this environment are 32-bit)

Most other 32-bit platforms are either already using 64-bit time_t, or are considered legacy and will not be gaining support from their upstream maintainers.

You can enable this using RUSTFLAGS:

RUSTFLAGS='--cfg=libc_unstable_time64' cargo ...

Note that there may still be some changes to features gated by this config option, hence "unstable"

... (truncated)

Changelog

Sourced from libc's changelog.

0.2.191 - 2026-10-16

Support

  • Emscripten: Support v4.0.12 to v6.0.11 (#5610)

Added

  • Apple: Add F_SETNOSIGPIPE and F_GETNOSIGPIPE (#5609)
  • Apple: add ifaliasreq and in6_aliasreq (#5557)
  • FreeBSD: Add PT_* constants (#5625)
  • Haiku: Add AI_* and NI_* constants from netdb.h (#5585)
  • Hurd: Add POSIX_SPAWN_* flags (#5578)
  • NuttX: Add AT_SYMLINK_NOFOLLOW (#5614)
  • OpenBSD: Add sensordev (#5534)
  • QuRT: Add isatty (#5618)
  • WASI: Add pthread_condattr_{get,set}clock (#5626)

Fixed

  • FreeBSD-aarch64: Fix gpregs and fpregs (#5603)

Changed

  • AIX: Restore 'c_ulong' as the type of 'statvfs.f_fsid'. (#5616)
  • Android: Make prop_info an extern type (#5561)
  • Apple: Mach API is now available on non-MacOS targets again. This should resolve some issues building ecosystem crates. (#5612), (#5636)
  • FreeBSD: Add back NOTE_PCTRLMASK with a deprecation warning (#5600)

0.2.190 - 2026-10-02

There is now a single config for enabling 64-bit time_t: libc_unstable_time64. This can be set unconditionally; it opts in to 64-bit time_t on the following platforms that use 32-bit by default:

  • 32-bit Linux-GNU
  • 32-bit Linux-uClibc
  • 32-bit Linux-musl. Note that setting this flag also enables some other changes that happend in musl v1.2.
  • 32-bit Windows-GNU
  • ESP-IDF (all targets with this environment are 32-bit)

Most other 32-bit platforms are either already using 64-bit time_t, or are considered legacy and will not be gaining support from their upstream maintainers.

You can enable this using RUSTFLAGS:

RUSTFLAGS='--cfg=libc_unstable_time64' cargo ...
</tr></table> 

... (truncated)

Commits
  • ccc2ff4 libc: Release 0.2.191
  • bce6efb apple: Add mach_task_self and mach_task_self_ to semver checks
  • ecf4fd4 apple: Re-restrict mach_vm_map to MacOS-only
  • b0e0423 Revert "apple: Disable mach.h API on Apple mobile OSs"
  • 9c7f568 apple: add ifaliasreq and in6_aliasreq
  • 018c7e9 wasi: Add pthread_condattr_{get,set}clock
  • 2c77185 openbsd: add sensordev structure
  • e10813c hurd: Add missing POSIX_SPAWN_* flags
  • 2ea4bc1 Fix FreeBSD/aarch64
  • 6d666fd dragonfly: Parse version from headers rather than CLI
  • Additional commits viewable in compare view

Updates lazy_static from 1.5.0 to 1.5.1

Commits

Updates zstd from 0.14.0 to 0.14.3

Commits
  • 75bc740 Release zstd 0.14.3, zstd-safe 8.2.0, and zstd-sys 2.1.2
  • 1c580b6 zstd-safe: keep the one-shot reset check inline
  • 502ee2f zstd-safe: only reset after one-shot decompression if a stream may be open
  • 5ded6e0 zstd-safe: apply rustfmt
  • e224018 zstd-safe: refuse to unload a dictionary under unfinished zstdmt jobs
  • 6a1065b Stop a single_frame() decoder when finish_frame() ends the frame
  • c7156df Don't let finish_frame() start a frame nothing was read from
  • b0fb8a7 Add write::Decoder::finish() to report incomplete frames
  • ab34d01 Don't retry a panicked writer when dropping AutoFlushDecoder/AutoFinishEncoder
  • 74ace2d Reset the borrowed context's session in raw::{Encoder,Decoder}::with_context
  • Additional commits viewable in compare view

Updates trybuild from 1.0.121 to 1.0.122

Release notes

Sourced from trybuild's releases.

1.0.122

  • Avoid triggering cargo::unused_workspace_dependencies warning in synthesized workspace (#344, thanks @​tamaro-skaljic)
Commits
  • 3f6a324 Release 1.0.122
  • ce13e9b Merge pull request 344 from tamaro-skaljic/fix/issue-343-unused-workspace-dep...
  • d027c0b Copy only inherited workspace dependencies into generated manifest
  • See full diff in compare view

Updates proptest from 1.11.0 to 1.12.0

Commits
  • 70cf37d Merge pull request #675 from proptest-rs/prep-1.12
  • 729e0ff prepare next releases and fix panic handling
  • 28c1fad Merge pull request #664 from Expyron/main
  • 5ddd9a4 Merge pull request #668 from toddATavail/fix-fork-persisted-failure
  • a6a9ac7 Merge pull request #673 from glaziermag/fix-book-deploy
  • d57d88b Replace rand_chacha with chacha20
  • 9411213 Deploy the book on pushes to main
  • 1719745 fix: terminate the fork replay when a persisted failure fails
  • 0f55b7a test: a persisted failure in a forked test must be reported
  • 4b0594e Merge pull request #669 from proptest-rs/no-fail-fast
  • Additional commits viewable in compare view

Updates syn from 3.0.6 to 3.0.9

Release notes

Sourced from syn's releases.

3.0.9

  • Add Punctuated::retain (#2088)

3.0.8

  • Fix parsing of <- and -> tokens in expressions (#2086, #2087)

3.0.7

Commits

Updates cxx from 1.0.202 to 1.0.205

Release notes

Sourced from cxx's releases.

1.0.205

  • Fix handling of - in feature names in cfg(feature = "…-…") attributes (#1766)

1.0.204

  • Fix cross-language aliasing UB flagged by BorrowSanitizer in rust::String::c_str (#1764, #1765)

1.0.203

  • Find target dir from OUT_DIR when CARGO_TARGET_DIR is relative (#1763, thanks @​akashchamp)
Commits
  • 598ccd9 Release 1.0.205
  • 0a3abc9 Merge pull request #1767 from dtolnay/debugassertions
  • 7d13af3 Mark debug_assertions special case for future removal
  • d45597d Update bazel apple_support module version
  • 061094a Lockfile update
  • de8799b Merge pull request #1766 from dtolnay/cfgfeature
  • 0047be1 Remove handling of CARGO_FEATURE_ in favor of 1.85's CARGO_CFG_FEATURE
  • c2413dc Release 1.0.204
  • 8f422a5 Lockfile update
  • 940cdca Merge pull request #1765 from dtolnay/data
  • Additional commits viewable in compare view

Updates cxx-build from 1.0.202 to 1.0.205

Release notes

Sourced from cxx-build's releases.

1.0.205

  • Fix handling of - in feature names in cfg(feature = "…-…") attributes (#1766)

1.0.204

  • Fix cross-language aliasing UB flagged by BorrowSanitizer in rust::String::c_str (#1764, #1765)

1.0.203

  • Find target dir from OUT_DIR when CARGO_TARGET_DIR is relative (#1763, thanks @​akashchamp)
Commits
  • 598ccd9 Release 1.0.205
  • 0a3abc9 Merge pull request #1767 from dtolnay/debugassertions
  • 7d13af3 Mark debug_assertions special case for future removal
  • d45597d Update bazel apple_support module version
  • 061094a Lockfile update
  • de8799b Merge pull request #1766 from dtolnay/cfgfeature
  • 0047be1 Remove handling of CARGO_FEATURE_ in favor of 1.85's CARGO_CFG_FEATURE
  • c2413dc Release 1.0.204
  • 8f422a5 Lockfile update
  • 940cdca Merge pull request #1765 from dtolnay/data
  • Additional commits viewable in compare view

Updates sha2 from 0.11.0 to 0.11.1

Commits

Updates open from 5.4.4 to 5.4.5

Release notes

Sourced from open's releases.

v5.4.5

Bug Fixes

  • check that_detached launchers for immediate failures that_detached() returned as soon as a launcher spawned, so a launcher that failed right after starting was reported as success and no fallback ran.

    Observe each launcher for 250 ms by delegating to that_detached_with_timeout(). A launcher that exits unsuccessfully within that window falls through to the next candidate; one still running is treated as handed off. The macOS and shellexecute-on-windows paths are unchanged.

    Extend the fixture tests to call that_detached() directly.

  • restore launcher fallback and support Plasma 6

    When xdg-open starts but exits unsuccessfully, that() and the CLI previously stopped before trying another installed opener. On Plasma 6, the fallback list also omitted the versioned KDE launchers.

    Launcher behavior

    • Retry reported failures from UNIX desktop launchers and retain the first unsuccessful exit when later candidates are missing.
    • Add kde-open6 and kde-open5 before the existing kde-open fallback.
    • Keep launcher ordering, argument boundaries, and exit-status policy together so compatibility decisions are explicit.
    • Preserve Windows' stop-on-failure policy and the existing protections for option-shaped and non-UTF-8 paths.

    Bounded detached checks

    Add that_detached_with_timeout(path, timeout) to opt into checking the actual launcher status for a configurable interval per candidate. Retry confirmed UNIX launcher failures; consider successful exits or launchers still running at the deadline handed off. A zero timeout polls once.

    On UNIX, use setsid() while retaining the launcher as a child. Reserve a background waiter before spawning and transfer ownership after a timeout so the launcher can outlive its caller and is eventually reaped. Stop fallback if observing an already-started launcher fails, avoiding a second launch based only on a wait error. Existing detached calls retain their timing behavior. The new API uses command-based launchers on Windows, including when shellexecute-on-windows is enabled.

    Historical constraints

    55dc10d relaxed exit checks for argumentless availability probes, where usage errors are expected. It did not establish that failed real opening operations should stop fallback. 245c95e introduced that behavior; the earlier implementation retried unsuccessful operations.

... (truncated)

Changelog

Sourced from open's changelog.

5.4.5 (2026-10-22)

Bug Fixes

  • check that_detached launchers for immediate failures that_detached() returned as soon as a launcher spawned, so a launcher that failed right after starting was reported as success and no fallback ran.

    Observe each launcher for 250 ms by delegating to that_detached_with_timeout(). A launcher that exits unsuccessfully within that window falls through to the next candidate; one still running is treated as handed off. The macOS and shellexecute-on-windows paths are unchanged.

    Extend the fixture tests to call that_detached() directly.

  • restore launcher fallback and support Plasma 6

    When xdg-open starts but exits unsuccessfully, that() and the CLI previously stopped before trying another installed opener. On Plasma 6, the fallback list also omitted the versioned KDE launchers.

    Launcher behavior

    • Retry reported failures from UNIX desktop launchers and retain the first unsuccessful exit when later candidates are missing.
    • Add kde-open6 and kde-open5 before the existing kde-open fallback.
    • Keep launcher ordering, argument boundaries, and exit-status policy together so compatibility decisions are explicit.
    • Preserve Windows' stop-on-failure policy and the existing protections for option-shaped and non-UTF-8 paths.

    Bounded detached checks

    Add that_detached_with_timeout(path, timeout) to opt into checking the actual launcher status for a configurable interval per candidate. Retry confirmed UNIX launcher failures; consider successful exits or launchers still running at the deadline handed off. A zero timeout polls once.

    On UNIX, use setsid() while retaining the launcher as a child. Reserve a background waiter before spawning and transfer ownership after a timeout so the launcher can outlive its caller and is eventually reaped. Stop fallback if observing an already-started launcher fails, avoiding a second launch based only on a wait error. Existing detached calls retain their timing behavior. The new API uses command-based launchers on Windows, including when shellexecute-on-windows is enabled.

    Historical constraints

    55dc10d relaxed exit checks for argumentless availability probes, where usage errors are expected. It did not establish that failed real opening operations should stop fallback. 245c95e introduced that behavior;

... (truncated)

Commits
  • a60a9d5 Release open v5.4.5
  • 4e64f6b Merge pull request #135 from Byron/fix-linux-launchers
  • c88759b auto-review
  • 42a34b0 fix(unix): check that_detached launchers for immediate failures
  • 279c1b2 fix(unix): restore launcher fallback and support Plasma 6
  • See full diff in compare view

Updates tree-sitter from 0.27.0 to 0.27.1

Release notes

Sourced from tree-sitter's releases.

v0.27.1

What's Changed

Full Changelog: tree-sitter/tree-sitter@v0.27.0...v0.27.1

Commits
  • 10eb5ac release v0.27.1
  • 1bf93c4 fix(bindings): embed queries in Bun standalone executables
  • 2173ca8 fix(bindings): await the load check in the Node binding test
  • a89d5b8 fix(bindings): load Node bindings under Bun again
  • 17d6c94 fix(bindings): load node-types.json on Windows
  • 825b3e4 fix(rust): ban std::fs::canonicalize
  • 4eb3973 fix(web): pass the progress callback to the cursor in Query.captures
  • 60cdb55 fix(test): give the query timeout tests room to time out
  • bdf4e7e fix(lib): charge query cursor work for the states it visits
  • dc21d38 fix(lib): halt the query cursor when the progress callback asks to stop
  • Additional commits viewable in compare view

Updates notify from 8.2.0 to 8.2.1

Release notes

Sourced from notify's releases.

notify v8.2.1

notify 8.2.1 (2026-10-25)

  • CHANGE: raise MSRV from 1.77 to 1.85 #719 #723 breaking
  • CHANGE: replace fsevent-sys with objc2-core-foundation and objc2-core-services on macOS, preserving the fsevent-sys feature for compatibility #726
  • CHANGE: update windows-sys to 0.61 #1045
  • FIX: prevent FsEventWatcher from crashing when unwatching an empty path #718
  • FIX: clean up deleted paths in INotifyWatcherDescription has been truncated

…ates

Bumps the rust-minor group with 19 updates in the / directory:

| Package | From | To |
| --- | --- | --- |
| [serde_json](https://github.com/serde-rs/json) | `1.0.151` | `1.0.154` |
| [uuid](https://github.com/uuid-rs/uuid) | `1.26.1` | `1.29.0` |
| [tokio](https://github.com/tokio-rs/tokio) | `1.53.1` | `1.54.1` |
| [memmap2](https://github.com/RazrFalcon/memmap2-rs) | `0.9.11` | `0.9.12` |
| [log](https://github.com/rust-lang/log) | `0.4.34` | `0.4.35` |
| [libc](https://github.com/rust-lang/libc) | `0.2.189` | `0.2.191` |
| [lazy_static](https://github.com/rust-lang-nursery/lazy-static.rs) | `1.5.0` | `1.5.1` |
| [zstd](https://github.com/gyscos/zstd-rs) | `0.14.0` | `0.14.3` |
| [trybuild](https://github.com/dtolnay/trybuild) | `1.0.121` | `1.0.122` |
| [proptest](https://github.com/proptest-rs/proptest) | `1.11.0` | `1.12.0` |
| [syn](https://github.com/dtolnay/syn) | `3.0.6` | `3.0.9` |
| [cxx](https://github.com/dtolnay/cxx) | `1.0.202` | `1.0.205` |
| [cxx-build](https://github.com/dtolnay/cxx) | `1.0.202` | `1.0.205` |
| [sha2](https://github.com/RustCrypto/hashes) | `0.11.0` | `0.11.1` |
| [open](https://github.com/Byron/open-rs) | `5.4.4` | `5.4.5` |
| [tree-sitter](https://github.com/tree-sitter/tree-sitter) | `0.27.0` | `0.27.1` |
| [notify](https://github.com/notify-rs/notify) | `8.2.0` | `8.2.1` |
| [schemars](https://github.com/GREsau/schemars) | `1.2.2` | `1.2.3` |
| [insta](https://github.com/mitsuhiko/insta) | `1.48.0` | `1.49.0` |



Updates `serde_json` from 1.0.151 to 1.0.154
- [Release notes](https://github.com/serde-rs/json/releases)
- [Commits](serde-rs/json@v1.0.151...v1.0.154)

Updates `uuid` from 1.26.1 to 1.29.0
- [Release notes](https://github.com/uuid-rs/uuid/releases)
- [Commits](uuid-rs/uuid@v1.26.1...v1.29.0)

Updates `tokio` from 1.53.1 to 1.54.1
- [Release notes](https://github.com/tokio-rs/tokio/releases)
- [Commits](tokio-rs/tokio@tokio-1.53.1...tokio-1.54.1)

Updates `memmap2` from 0.9.11 to 0.9.12
- [Changelog](https://github.com/RazrFalcon/memmap2-rs/blob/master/CHANGELOG.md)
- [Commits](RazrFalcon/memmap2-rs@v0.9.11...v0.9.12)

Updates `log` from 0.4.34 to 0.4.35
- [Release notes](https://github.com/rust-lang/log/releases)
- [Changelog](https://github.com/rust-lang/log/blob/main/CHANGELOG.md)
- [Commits](rust-lang/log@0.4.34...v0.4.35)

Updates `libc` from 0.2.189 to 0.2.191
- [Release notes](https://github.com/rust-lang/libc/releases)
- [Changelog](https://github.com/rust-lang/libc/blob/0.2.191/CHANGELOG.md)
- [Commits](rust-lang/libc@0.2.189...0.2.191)

Updates `lazy_static` from 1.5.0 to 1.5.1
- [Release notes](https://github.com/rust-lang-nursery/lazy-static.rs/releases)
- [Commits](https://github.com/rust-lang-nursery/lazy-static.rs/commits)

Updates `zstd` from 0.14.0 to 0.14.3
- [Release notes](https://github.com/gyscos/zstd-rs/releases)
- [Commits](gyscos/zstd-rs@v0.14.0...v0.14.3)

Updates `trybuild` from 1.0.121 to 1.0.122
- [Release notes](https://github.com/dtolnay/trybuild/releases)
- [Commits](dtolnay/trybuild@1.0.121...1.0.122)

Updates `proptest` from 1.11.0 to 1.12.0
- [Release notes](https://github.com/proptest-rs/proptest/releases)
- [Changelog](https://github.com/proptest-rs/proptest/blob/main/CHANGELOG.md)
- [Commits](proptest-rs/proptest@v1.11.0...v1.12.0)

Updates `syn` from 3.0.6 to 3.0.9
- [Release notes](https://github.com/dtolnay/syn/releases)
- [Commits](dtolnay/syn@3.0.6...3.0.9)

Updates `cxx` from 1.0.202 to 1.0.205
- [Release notes](https://github.com/dtolnay/cxx/releases)
- [Commits](dtolnay/cxx@1.0.202...1.0.205)

Updates `cxx-build` from 1.0.202 to 1.0.205
- [Release notes](https://github.com/dtolnay/cxx/releases)
- [Commits](dtolnay/cxx@1.0.202...1.0.205)

Updates `sha2` from 0.11.0 to 0.11.1
- [Commits](RustCrypto/hashes@sha2-v0.11.0...sha2-v0.11.1)

Updates `open` from 5.4.4 to 5.4.5
- [Release notes](https://github.com/Byron/open-rs/releases)
- [Changelog](https://github.com/Byron/open-rs/blob/main/changelog.md)
- [Commits](Byron/open-rs@v5.4.4...v5.4.5)

Updates `tree-sitter` from 0.27.0 to 0.27.1
- [Release notes](https://github.com/tree-sitter/tree-sitter/releases)
- [Commits](tree-sitter/tree-sitter@v0.27.0...v0.27.1)

Updates `notify` from 8.2.0 to 8.2.1
- [Release notes](https://github.com/notify-rs/notify/releases)
- [Changelog](https://github.com/notify-rs/notify/blob/notify-8.2.1/CHANGELOG.md)
- [Commits](notify-rs/notify@notify-8.2.0...notify-8.2.1)

Updates `schemars` from 1.2.2 to 1.2.3
- [Release notes](https://github.com/GREsau/schemars/releases)
- [Changelog](https://github.com/GREsau/schemars/blob/master/CHANGELOG.md)
- [Commits](GREsau/schemars@v1.2.2...v1.2.3)

Updates `insta` from 1.48.0 to 1.49.0
- [Release notes](https://github.com/mitsuhiko/insta/releases)
- [Changelog](https://github.com/mitsuhiko/insta/blob/master/CHANGELOG.md)
- [Commits](mitsuhiko/insta@1.48.0...1.49.0)

---
updated-dependencies:
- dependency-name: serde_json
  dependency-version: 1.0.154
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: rust-minor
- dependency-name: uuid
  dependency-version: 1.29.0
  dependency-type: direct:production
  update-type: version-update:semver-minor
  dependency-group: rust-minor
- dependency-name: tokio
  dependency-version: 1.54.1
  dependency-type: direct:production
  update-type: version-update:semver-minor
  dependency-group: rust-minor
- dependency-name: memmap2
  dependency-version: 0.9.12
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: rust-minor
- dependency-name: log
  dependency-version: 0.4.35
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: rust-minor
- dependency-name: libc
  dependency-version: 0.2.191
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: rust-minor
- dependency-name: lazy_static
  dependency-version: 1.5.1
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: rust-minor
- dependency-name: zstd
  dependency-version: 0.14.3
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: rust-minor
- dependency-name: trybuild
  dependency-version: 1.0.122
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: rust-minor
- dependency-name: proptest
  dependency-version: 1.12.0
  dependency-type: direct:production
  update-type: version-update:semver-minor
  dependency-group: rust-minor
- dependency-name: syn
  dependency-version: 3.0.9
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: rust-minor
- dependency-name: cxx
  dependency-version: 1.0.205
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: rust-minor
- dependency-name: cxx-build
  dependency-version: 1.0.205
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: rust-minor
- dependency-name: sha2
  dependency-version: 0.11.1
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: rust-minor
- dependency-name: open
  dependency-version: 5.4.5
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: rust-minor
- dependency-name: tree-sitter
  dependency-version: 0.27.1
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: rust-minor
- dependency-name: notify
  dependency-version: 8.2.1
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: rust-minor
- dependency-name: schemars
  dependency-version: 1.2.3
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: rust-minor
- dependency-name: insta
  dependency-version: 1.49.0
  dependency-type: direct:production
  update-type: version-update:semver-minor
  dependency-group: rust-minor
...

Signed-off-by: dependabot[bot] <support@github.com>
@dependabot dependabot Bot added dependencies Pull requests that update a dependency file rust Pull requests that update rust code labels Nov 9, 2026
@dependabot @github

dependabot Bot commented on behalf of github Nov 9, 2026

Copy link
Copy Markdown
Contributor Author

Assignees

The following users could not be added as assignees: lord-patpak. Either the username does not exist or it does not have the correct permissions to be added as an assignee.

Please fix the above issues or remove invalid values from dependabot.yml.

@github-actions
github-actions Bot enabled auto-merge (squash) November 9, 2026 06:28
@github-actions

github-actions Bot commented Nov 9, 2026

Copy link
Copy Markdown
Contributor

Benchmark gate

Baseline: rolling window of 2 run(s) on AMD EPYC 9V45 96-Core Processor. Current arm: median of 6 repetition(s).

Result: PASS (with tail advisories)

Adaptive_CmdVel@16B — 50000 samples
metric baseline current change band (<=) verdict
median 30.0 ns 40.0 ns +33.3% 60.0 ns INCONCLUSIVE
p95 60.0 ns 60.0 ns +0.0% 120.0 ns INCONCLUSIVE
p99 60.0 ns 70.0 ns +16.7% 120.0 ns INCONCLUSIVE
p99.9 90.0 ns 90.0 ns +0.0% 180.0 ns INCONCLUSIVE
p99.99 — 90.0 ns — — N/A (samples)
max 7.79 µs 8.95 µs +15.0% 15.57 µs INCONCLUSIVE
max_jitter 7.78 µs 8.94 µs +15.0% 15.55 µs INCONCLUSIVE
p99/median 2.00x 1.75x -12.5% 4.00x INCONCLUSIVE
p99.9/median 3.00x 2.25x -25.0% 6.00x INCONCLUSIVE
ns/msg 125.7 ns 134.4 ns +6.9% 251.4 ns INCONCLUSIVE
Adaptive_Imu@304B — 50000 samples
metric baseline current change band (<=) verdict
median 40.0 ns 40.0 ns +0.0% 80.0 ns INCONCLUSIVE
p95 60.0 ns 60.0 ns +0.0% 120.0 ns INCONCLUSIVE
p99 80.0 ns 80.0 ns +0.0% 160.0 ns INCONCLUSIVE
p99.9 100.0 ns 100.0 ns +0.0% 200.0 ns INCONCLUSIVE
p99.99 — 110.0 ns — — N/A (samples)
max 7.11 µs 9.27 µs +30.4% 14.22 µs INCONCLUSIVE
max_jitter 7.09 µs 9.25 µs +30.5% 14.18 µs INCONCLUSIVE
p99/median 2.00x 2.00x +0.0% 4.00x INCONCLUSIVE
p99.9/median 2.50x 2.50x +0.0% 5.00x INCONCLUSIVE
ns/msg 201.6 ns 216.2 ns +7.2% 403.3 ns INCONCLUSIVE
Adaptive_JointCommand@928B — 50000 samples
metric baseline current change band (<=) verdict
median 60.0 ns 60.0 ns +0.0% 120.0 ns INCONCLUSIVE
p95 90.0 ns 80.0 ns -11.1% 180.0 ns INCONCLUSIVE
p99 110.0 ns 110.0 ns +0.0% 220.0 ns INCONCLUSIVE
p99.9 125.0 ns 130.0 ns +4.0% 250.0 ns INCONCLUSIVE
p99.99 — 160.0 ns — — N/A (samples)
max 8.52 µs 8.71 µs +2.3% 17.04 µs INCONCLUSIVE
max_jitter 8.49 µs 8.68 µs +2.3% 16.98 µs INCONCLUSIVE
p99/median 1.83x 1.71x -6.5% 3.67x INCONCLUSIVE
p99.9/median 2.08x 2.00x -4.0% 4.17x INCONCLUSIVE
ns/msg 236.4 ns 256.2 ns +8.3% 472.8 ns INCONCLUSIVE
Adaptive_LaserScan@1480B — 10000 samples
metric baseline current change band (<=) verdict
median 60.0 ns 60.0 ns +0.0% 120.0 ns INCONCLUSIVE
p95 80.0 ns 80.0 ns +0.0% 160.0 ns INCONCLUSIVE
p99 100.0 ns 100.0 ns +0.0% 200.0 ns INCONCLUSIVE
p99.9 130.0 ns 120.0 ns -7.7% 260.0 ns INCONCLUSIVE
p99.99 — 150.0 ns — — N/A (samples)
max 3.59 µs 160.0 ns -95.5% 7.18 µs INCONCLUSIVE
max_jitter 3.56 µs 120.0 ns -96.6% 7.12 µs INCONCLUSIVE
p99/median 1.67x 1.67x +0.0% 3.33x INCONCLUSIVE
p99.9/median 2.17x 2.00x -7.7% 4.33x INCONCLUSIVE
ns/msg 387.9 ns 403.4 ns +4.0% 775.8 ns INCONCLUSIVE
Topic_cross_process_oneway_paced_stream@1024B — 200000 samples
metric baseline current change band (<=) verdict
median 162.5 ns 160.0 ns -1.5% 325.0 ns INCONCLUSIVE
p95 197.5 ns 210.0 ns +6.3% 395.0 ns INCONCLUSIVE
p99 232.5 ns 300.0 ns +29.0% 465.0 ns INCONCLUSIVE
p99.9 858.5 ns 1.43 µs +66.8% 1.72 µs INCONCLUSIVE
p99.99 6.92 µs 10.80 µs +56.0% 13.85 µs INCONCLUSIVE
max 37.58 µs 51.84 µs +37.9% 75.16 µs INCONCLUSIVE
max_jitter 37.48 µs 51.73 µs +38.0% 74.95 µs INCONCLUSIVE
p99/median 1.43x 2.00x +39.8% 2.86x INCONCLUSIVE
p99.9/median 5.28x 9.55x +80.7% 10.57x INCONCLUSIVE
ns/msg 10.00 µs 10.00 µs -0.0% 20.00 µs INCONCLUSIVE
Topic_cross_process_oneway_paced_stream@64B — 200000 samples
metric baseline current change band (<=) verdict
median 112.5 ns 130.0 ns +15.6% 225.0 ns INCONCLUSIVE
p95 147.5 ns 160.0 ns +8.5% 295.0 ns INCONCLUSIVE
p99 174.5 ns 250.0 ns +43.3% 349.0 ns INCONCLUSIVE
p99.9 793.5 ns 921.0 ns +16.1% 1.59 µs INCONCLUSIVE
p99.99 5.29 µs 8.29 µs +56.7% 10.58 µs INCONCLUSIVE
max 17.75 µs 50.62 µs +185.2% 35.51 µs REGRESSED
max_jitter 17.68 µs 50.55 µs +185.9% 35.36 µs REGRESSED
p99/median 1.55x 1.92x +23.7% 3.11x INCONCLUSIVE
p99.9/median 7.07x 7.67x +8.5% 14.15x INCONCLUSIVE
ns/msg 10.00 µs 10.00 µs -0.0% 20.00 µs INCONCLUSIVE
Topic_cross_process_oneway_pingpong@1024B — 1000000 samples
metric baseline current change band (<=) verdict
median 175.0 ns 175.0 ns +0.0% 350.0 ns INCONCLUSIVE
p95 212.5 ns 205.0 ns -3.5% 425.0 ns INCONCLUSIVE
p99 235.0 ns 225.0 ns -4.3% 470.0 ns INCONCLUSIVE
p99.9 531.0 ns 536.0 ns +0.9% 1.06 µs INCONCLUSIVE
p99.99 4.00 µs 5.10 µs +27.5% 8.00 µs INCONCLUSIVE
max 14.91 µs 20.13 µs +35.0% 29.82 µs INCONCLUSIVE
max_jitter 14.80 µs 20.02 µs +35.3% 29.59 µs INCONCLUSIVE
p99/median 1.34x 1.29x -4.3% 2.69x INCONCLUSIVE
p99.9/median 3.03x 3.06x +0.9% 6.07x INCONCLUSIVE
ns/msg 212.8 ns 224.6 ns +5.6% 425.5 ns INCONCLUSIVE
Topic_cross_process_oneway_pingpong@64B — 1000000 samples
metric baseline current change band (<=) verdict
median 105.0 ns 120.0 ns +14.3% 210.0 ns INCONCLUSIVE
p95 127.5 ns 140.0 ns +9.8% 255.0 ns INCONCLUSIVE
p99 137.5 ns 145.0 ns +5.5% 275.0 ns INCONCLUSIVE
p99.9 463.5 ns 481.0 ns +3.8% 927.0 ns INCONCLUSIVE
p99.99 3.24 µs 3.86 µs +18.8% 6.49 µs INCONCLUSIVE
max 14.59 µs 16.04 µs +9.9% 29.18 µs INCONCLUSIVE
max_jitter 14.51 µs 15.96 µs +10.0% 29.02 µs INCONCLUSIVE
p99/median 1.31x 1.21x -7.7% 2.62x INCONCLUSIVE
p99.9/median 4.41x 4.01x -9.2% 8.83x INCONCLUSIVE
ns/msg 139.5 ns 157.3 ns +12.7% 279.1 ns INCONCLUSIVE
Topic_cross_process_unpaced_queue_delay@64B — 200000 samples
metric baseline current change band (<=) verdict
median 87.5 ns 110.0 ns +25.7% 175.0 ns INCONCLUSIVE
p95 132.5 ns 150.0 ns +13.2% 265.0 ns INCONCLUSIVE
p99 157.5 ns 170.0 ns +7.9% 315.0 ns INCONCLUSIVE
p99.9 502.5 ns 4.23 µs +741.0% 1.00 µs GROSS REGRESSION
p99.99 5.22 µs 6.33 µs +21.1% 10.45 µs INCONCLUSIVE
max 19.53 µs 39.36 µs +101.6% 39.05 µs REGRESSED
max_jitter 19.46 µs 39.29 µs +101.9% 38.92 µs REGRESSED
p99/median 1.83x 1.55x -15.7% 3.67x INCONCLUSIVE
p99.9/median 6.00x 38.42x +540.3% 12.00x GROSS REGRESSION
ns/msg 70.7 ns 75.3 ns +6.5% 141.5 ns INCONCLUSIVE

Blocking metrics: median, p99/median. Report-only: p95, p99, p99.9, p99.99, max, max_jitter, p99.9/median, ns/msg. Tail metrics past p99.9 are advisory on a shared runner — one host preemption owns max.

Raw benchmark output
ults=0 ctx_switches vol=62 invol=5
  core freq cpu3   (measured window only): start=n/a end=n/a min=0MHz max=0MHz mean=0MHz spread=0.00% (n=0)
  core freq cpu1   (measured window only): start=n/a end=n/a min=0MHz max=0MHz mean=0MHz spread=0.00% (n=0)
  cross-core TSC offset theta_hat = +0.0 ns (clock B - clock A). The `one-way, offset-free` row does not depend on it.

== cross-process ping-pong -- ONE-WAY LATENCY (2 processes, 2 topics, 1 role per handle) ==
   json metric key: `Topic_cross_process_oneway_pingpong` (identity used by the CI gate: `Topic_cross_process_oneway_pingpong@1024B`)
   payload Probe1k = 1024 B (POD=true) | ring capacity 256 slots | topology 1P/1C | cpu3 <-> cpu1 | pid 7372 <-> pid 7382
------------------------------------------------------------------------------------------------------------------------------------------
  metric (ns)                            n       min    median       p99     p99.9     p99.99         max |  n>p99.9  n>p99.99
------------------------------------------------------------------------------------------------------------------------------------------
  one-way, offset-free             1000000     115.2     175.3     225.3     535.8     4396.5     21266.4 |      965       100
  round trip (RTT)                 1000000     260.4     380.6     470.7    1101.6     9534.1     42562.9 |      997       100
  RTT halved (one-way)             1000000     130.2     190.3     235.3     550.8     4767.0     21281.5 |      997       100
  leg A->B (apparent)              1000000     100.1     170.3     230.3     841.2     5898.7     42292.5 |      991       100
  leg B->A (apparent)              1000000     110.2     180.3     240.4     280.4     5518.2     19118.3 |      905        99
  responder turnaround             1000000      20.0      30.0      30.0      30.0      110.2     33229.1 |      185        84
------------------------------------------------------------------------------------------------------------------------------------------
  samples=1000000  warmup=200000  transferred=2000000  lost=0  invalid(TSC-skew)=0  poll cost=24.3 ns  window=0.449 s
  measured window: minor_faults=0 major_faults=0 ctx_switches vol=88 invol=4
  core freq cpu3   (measured window only): start=n/a end=n/a min=0MHz max=0MHz mean=0MHz spread=0.00% (n=0)
  core freq cpu1   (measured window only): start=n/a end=n/a min=0MHz max=0MHz mean=0MHz spread=0.00% (n=0)
  cross-core TSC offset theta_hat = -5.0 ns (clock B - clock A). The `one-way, offset-free` row does not depend on it.

== cross-process paced stream -- ONE-WAY LATENCY (10000 ns publish period, publisher -> subscriber, direct) ==
   json metric key: `Topic_cross_process_oneway_paced_stream` (identity used by the CI gate: `Topic_cross_process_oneway_paced_stream@64B`)
   payload Probe64 = 64 B (POD=true) | ring capacity 256 slots | topology 1P/1C | cpu3 <-> cpu1 | pid 7372 <-> pid 7384
------------------------------------------------------------------------------------------------------------------------------------------
  metric (ns)                            n       min    median       p99     p99.9     p99.99         max |  n>p99.9  n>p99.99
------------------------------------------------------------------------------------------------------------------------------------------
  one-way (assumes TSC coherence)    200000      80.1     120.2     260.4     921.4     8292.3     35742.9 |      194        20
------------------------------------------------------------------------------------------------------------------------------------------
  samples=200000  warmup=20000  transferred=200000  lost=0  invalid(TSC-skew)=0  poll cost=5.0 ns  window=2.000 s
  measured backlog: 0.01 messages deep on the median sample (ring holds 256); delivered 100.0% of transferred
  measured window: minor_faults=0 major_faults=0 ctx_switches vol=395 invol=17
  core freq cpu3   (measured window only): start=n/a end=n/a min=0MHz max=0MHz mean=0MHz spread=0.00% (n=0)
  core freq cpu1   (measured window only): start=n/a end=n/a min=0MHz max=0MHz mean=0MHz spread=0.00% (n=0)
  this row assumes cross-core TSC coherence; theta_hat measured in the ping-pong at this size = +0.0 ns (add to correct).

== cross-process paced stream -- ONE-WAY LATENCY (10000 ns publish period, publisher -> subscriber, direct) ==
   json metric key: `Topic_cross_process_oneway_paced_stream` (identity used by the CI gate: `Topic_cross_process_oneway_paced_stream@1024B`)
   payload Probe1k = 1024 B (POD=true) | ring capacity 256 slots | topology 1P/1C | cpu3 <-> cpu1 | pid 7372 <-> pid 7386
------------------------------------------------------------------------------------------------------------------------------------------
  metric (ns)                            n       min    median       p99     p99.9     p99.99         max |  n>p99.9  n>p99.99
------------------------------------------------------------------------------------------------------------------------------------------
  one-way (assumes TSC coherence)    200000     110.2     170.3     340.5    1181.7    11777.4     51836.7 |      198        20
------------------------------------------------------------------------------------------------------------------------------------------
  samples=200000  warmup=20000  transferred=200000  lost=0  invalid(TSC-skew)=0  poll cost=10.4 ns  window=2.000 s
  measured backlog: 0.02 messages deep on the median sample (ring holds 256); delivered 100.0% of transferred
  measured window: minor_faults=0 major_faults=0 ctx_switches vol=395 invol=15
  core freq cpu3   (measured window only): start=n/a end=n/a min=0MHz max=0MHz mean=0MHz spread=0.00% (n=0)
  core freq cpu1   (measured window only): start=n/a end=n/a min=0MHz max=0MHz mean=0MHz spread=0.00% (n=0)
  this row assumes cross-core TSC coherence; theta_hat measured in the ping-pong at this size = -5.0 ns (add to correct).

== cross-process unpaced stream -- QUEUE DELAY / BACKPRESSURE, NOT LATENCY (publisher unthrottled; each sample includes its wait in the ring) ==
   !! Every number in this block is a PUBLISH-TO-RECEIVE DELAY that INCLUDES
   !! time spent waiting in the ring. It is a backpressure/throughput figure.
   !! Do not quote it as an IPC latency; that is the exact mistake the retired
   !! `Topic_cross_process` metric shipped for months. Read `measured backlog`
   !! below: ~1 message means the subscriber kept up and no queue formed.
   json metric key: `Topic_cross_process_unpaced_queue_delay` (identity used by the CI gate: `Topic_cross_process_unpaced_queue_delay@64B`)
   payload Probe64 = 64 B (POD=true) | ring capacity 256 slots | topology 1P/1C | cpu3 <-> cpu1 | pid 7372 <-> pid 7389
------------------------------------------------------------------------------------------------------------------------------------------
  metric (ns)                            n       min    median       p99     p99.9     p99.99         max |  n>p99.9  n>p99.99
------------------------------------------------------------------------------------------------------------------------------------------
  publish->receive incl. queue      200000      70.1     110.2     170.3    2563.8     5568.2     41271.0 |      200        19
------------------------------------------------------------------------------------------------------------------------------------------
  samples=200000  warmup=20000  transferred=200000  lost=0  invalid(TSC-skew)=0  poll cost=6.5 ns  window=0.015 s
  measured backlog: 1.46 messages deep on the median sample (ring holds 256); delivered 100.0% of transferred
  measured window: minor_faults=0 major_faults=0 ctx_switches vol=3 invol=1
  core freq cpu3   (measured window only): start=n/a end=n/a min=0MHz max=0MHz mean=0MHz spread=0.00% (n=0)
  core freq cpu1   (measured window only): start=n/a end=n/a min=0MHz max=0MHz mean=0MHz spread=0.00% (n=0)
  this row assumes cross-core TSC coherence; theta_hat measured in the ping-pong at this size = +0.0 ns (add to correct).

SUMMARY — one row per published JSON metric (nanoseconds, wall clock)
------------------------------------------------------------------------------------------------------------------------------------------
  json metric key                                what it measures                  n    median       p99     p99.9     p99.99         max
------------------------------------------------------------------------------------------------------------------------------------------
  Topic_cross_process_oneway_pingpong@64B        ONE-WAY latency             1000000     120.2     145.2     480.7     3339.9     13469.9
  Topic_cross_process_oneway_pingpong@1024B      ONE-WAY latency             1000000     175.3     225.3     535.8     4396.5     21266.4
  Topic_cross_process_oneway_paced_stream@64B    ONE-WAY latency              200000     120.2     260.4     921.4     8292.3     35742.9
  Topic_cross_process_oneway_paced_stream@1024B  ONE-WAY latency              200000     170.3     340.5    1181.7    11777.4     51836.7
  Topic_cross_process_unpaced_queue_delay@64B    QUEUE DELAY (not latency)    200000     110.2     170.3    2563.8     5568.2     41271.0
------------------------------------------------------------------------------------------------------------------------------------------
  Instrumentation is INCLUDED in every figure and never subtracted.
  `Topic_cross_process_oneway_pingpong` is the true one-way figure: it assumes
  nothing about cross-core TSC synchronisation and excludes the responder's turnaround.
  `RTT halved (one-way)` in the per-scenario tables is a ROUND TRIP DIVIDED BY TWO; it is
  the row to compare against any competitor publishing a halved ping-pong.
  `Topic_cross_process_unpaced_queue_delay` is a BACKPRESSURE figure: it includes
  however long the message waited in the ring, so it scales with the achieved backlog
  (printed in its scenario block) and is NOT a latency.
  A figure without its payload size, ring depth and topology is not comparable to anything.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!! THE SAME WARNINGS AGAIN — they apply to every number above
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!  CPU governor is ?/?, not `performance`. RDTSC is a wall clock while the work costs a fixed number of CORE cycles, so every nanosecond below scales with a frequency the governor is free to change. iceoryx2's maintainers measured ~300 ns of difference from the governor alone.
!!  No isolcpus= on the kernel command line: the benchmark cores are shared with everything else on this box. The involuntary context-switch count in each scenario block is the mechanism behind the far tail.
!!  paced_stream (64 B) `one-way (assumes TSC coherence)`: p99.99 rests on 20 observations. Below ~100 exceedances it is a max wearing a percentile's name; +/-5% needs ~400 (--iterations 4000000).
!!  paced_stream (1024 B) `one-way (assumes TSC coherence)`: p99.99 rests on 20 observations. Below ~100 exceedances it is a max wearing a percentile's name; +/-5% needs ~400 (--iterations 4000000).
!!  unpaced_queue (64 B): measured backlog is only 1.46 messages — the subscriber kept up, so the ring never filled and `Topic_cross_process_unpaced_queue_delay` is NOT a backpressure figure on this run. It is a publish-to-receive delay with no queueing in it, and it is still not a latency (see the paced and ping-pong rows for that).
!!  unpaced_queue (64 B) `publish->receive incl. queue`: p99.99 rests on 19 observations. Below ~100 exceedances it is a max wearing a percentile's name; +/-5% needs ~400 (--iterations 4000000).
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

Gate report written to /home/runner/work/horus/horus/benchmarks/gate-reports/cross_process-2.json
  keys: Topic_cross_process_oneway_pingpong@64B, Topic_cross_process_oneway_pingpong@1024B, Topic_cross_process_oneway_paced_stream@64B, Topic_cross_process_oneway_paced_stream@1024B, Topic_cross_process_unpaced_queue_delay@64B
  NOTE: `Topic_cross_process` is retired. A baseline window that still holds it
  will report it as missing and fail the gate until the window is refreshed —
  that is intended, and is why the key was not silently reused.
  Raw samples are omitted (pass --raw-in-json, or --raw-dir for a lossless cycle dump).
  6 warning(s) apply to these numbers and are NOT in this file; use --detail-json.
╔══════════════════════════════════════════════════════════════════╗
║        HORUS Robotics Message Types Benchmark                    ║
╠══════════════════════════════════════════════════════════════════╣
║  Testing real-world robotics message latency (REP 2014)          ║
╚══════════════════════════════════════════════════════════════════╝

Platform: AMD EPYC 9V45 96-Core Processor (4 cores)
Iterations: 50000
Warmup: 5000

╔═════════════════════════════════════════════════════════════════════════════╗
║ Message Type    │ Size (bytes) │ Typical Rate │ Use Case                    ║
╠═════════════════════════════════════════════════════════════════════════════╣
║ CmdVel          │           16 │ 1000+ Hz     │ Velocity control commands   ║
║ Imu             │          296 │ 500+ Hz      │ IMU sensor fusion           ║
║ LaserScan       │         1480 │ 10-40 Hz     │ 2D lidar navigation         ║
║ JointCommand    │         1032 │ 500+ Hz      │ Manipulator control         ║
╚═════════════════════════════════════════════════════════════════════════════╝


[Topic] Running benchmarks — timing tx.send() enqueue cost only
        (cross-thread consumer drains in the background; no sample
         waits for delivery, so these are NOT end-to-end latencies)
─────────────────────────────────────────────────
  CmdVel              16 bytes │ median:      40ns │ p99:      70ns │ CV: 1.3521
  Imu                304 bytes │ median:      30ns │ p99:      80ns │ CV: 1.4030
  LaserScan         1480 bytes │ median:      70ns │ p99:     110ns │ CV: 0.2301
  JointCommand       928 bytes │ median:      60ns │ p99:     100ns │ CV: 1.1873

╔══════════════════════════════════════════════════════════════════════════════════════════╗
║           SUMMARY BY MESSAGE TYPE — send() ENQUEUE COST, not end-to-end latency           ║
╠══════════════════════════════════════════════════════════════════════════════════════════╣
║ Message      │   Size │ Median (ns) │ p99 (ns) │    CV   │ Misses │ Sustained rate       ║
╠══════════════════════════════════════════════════════════════════════════════════════════╣
║ CmdVel       │     16 │          40 │       70 │  1.3521 │      0 │      7.44 M msg/s     ║
║ Imu          │    304 │          30 │       80 │  1.4030 │      0 │      4.74 M msg/s     ║
║ LaserScan    │   1480 │          70 │      110 │  0.2301 │      0 │      2.38 M msg/s     ║
║ JointCommand │    928 │          60 │      100 │  1.1873 │      0 │      4.08 M msg/s     ║
╚══════════════════════════════════════════════════════════════════════════════════════════╝
Sustained rate is wall-clock over the measured loop, so it includes the batch
drain-waits. It is a publish rate for this producer/consumer pair, not a bus limit.

╔══════════════════════════════════════════════════════════════════╗
║              REAL-TIME SUITABILITY ANALYSIS                      ║
╠══════════════════════════════════════════════════════════════════╣
║ These gates grade the p99 of the send() ENQUEUE COST against one ║
║ whole control period. They are not end-to-end control-loop       ║
║ guarantees: a real loop also pays the receive side, the node's   ║
║ compute, and the scheduler's jitter, none of which is measured   ║
║ here. The headroom is printed so a pass by four orders of        ║
║ magnitude cannot be read as a tight result.                      ║
╠══════════════════════════════════════════════════════════════════╣
║ CmdVel    (1kHz  control): ~ PASS by 14286x — gate not discriminating (p99=70ns vs 1000000ns)
║ CmdVel    (10kHz control): ~ PASS by 1429x — gate not discriminating (p99=70ns vs 100000ns)
║ Imu       (500Hz fusion) : ~ PASS by 25000x — gate not discriminating (p99=80ns vs 2000000ns)
║ LaserScan (40Hz  lidar)  : ~ PASS by 227273x — gate not discriminating (p99=110ns vs 25000000ns)
║ JointCmd  (500Hz control): ~ PASS by 20000x — gate not discriminating (p99=100ns vs 2000000ns)
╚══════════════════════════════════════════════════════════════════╝

Results written to: /home/runner/work/horus/horus/benchmarks/gate-reports/robotics-2.json
===== repetition 3/3 =====

HORUS cross-process benchmark — ONE-WAY LATENCY (headline) + unpaced queue delay
Two OS processes on two distinct physical cores over the real Topic SHM transport.
Every handle holds exactly one role, so the `role == Both` same-thread fast path that
produces the repo's headline send-cost figure is never taken. Round-trip rows are
labelled where they are a halved round trip. See the module header for the clock
decomposition and for why the `Topic_cross_process` JSON key was retired.

MACHINE — every nanosecond below is conditional on this block
------------------------------------------------------------------------------------------------------------------------------------------
  cpu model              : AMD EPYC 9V45 96-Core Processor (2 physical / 4 logical cores)
  kernel                 : 7.0.0-1017-azure (x86_64)
  benchmark cores        : cpu3 = process A (pinger / subscriber), cpu1 = process B (responder / publisher)  [DistinctPhysical]
  governor               : cpu3=n/a cpu1=n/a  driver=n/a
  scaling range          : n/a .. n/a  (hw max n/a)  turbo_disabled=unknown
  clocksource            : tsc   tsc flags: [tsc, rdtscp, constant_tsc, tsc_reliable, nonstop_tsc]
  TSC rate (calibrated)  : 2596.152 MHz (0.3852 ns/cycle) — a WALL clock; NOT the core clock
  clock instrumentation  : 26 cycles = 10.0 ns for serialize()+rdtsc()+rdtscp(), INCLUDED in every sample and never subtracted
  transparent_hugepage   : always [madvise] never
  sched_rt_runtime/period: 950000 / 1000000 us    perf_event_paranoid=4
  isolcpus=(none)  nohz_full=(none)  smt_active=1  mlockall(MCL_CURRENT)=true
------------------------------------------------------------------------------------------------------------------------------------------

running ping-pong (one-way latency), 64 B payload ...
running ping-pong (one-way latency), 1024 B payload ...
running paced stream (one-way latency), 64 B payload ...
running paced stream (one-way latency), 1024 B payload ...
running unpaced stream (QUEUE DELAY, not latency), 64 B payload ...

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!! READ THIS BEFORE QUOTING ANY NUMBER BELOW
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!  CPU governor is ?/?, not `performance`. RDTSC is a wall clock while the work costs a fixed number of CORE cycles, so every nanosecond below scales with a frequency the governor is free to change. iceoryx2's maintainers measured ~300 ns of difference from the governor alone.
!!  No isolcpus= on the kernel command line: the benchmark cores are shared with everything else on this box. The involuntary context-switch count in each scenario block is the mechanism behind the far tail.
!!  pingpong (64 B) `one-way, offset-free`: p99.99 rests on 99 observations. Below ~100 exceedances it is a max wearing a percentile's name; +/-5% needs ~400 (--iterations 4000000).
!!  pingpong (1024 B) `one-way, offset-free`: p99.99 rests on 99 observations. Below ~100 exceedances it is a max wearing a percentile's name; +/-5% needs ~400 (--iterations 4000000).
!!  paced_stream (64 B) `one-way (assumes TSC coherence)`: p99.99 rests on 20 observations. Below ~100 exceedances it is a max wearing a percentile's name; +/-5% needs ~400 (--iterations 4000000).
!!  paced_stream (1024 B) `one-way (assumes TSC coherence)`: p99.99 rests on 20 observations. Below ~100 exceedances it is a max wearing a percentile's name; +/-5% needs ~400 (--iterations 4000000).
!!  unpaced_queue (64 B): measured backlog is only 1.45 messages — the subscriber kept up, so the ring never filled and `Topic_cross_process_unpaced_queue_delay` is NOT a backpressure figure on this run. It is a publish-to-receive delay with no queueing in it, and it is still not a latency (see the paced and ping-pong rows for that).
!!  unpaced_queue (64 B) `publish->receive incl. queue`: p99.99 rests on 20 observations. Below ~100 exceedances it is a max wearing a percentile's name; +/-5% needs ~400 (--iterations 4000000).
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

== cross-process ping-pong -- ONE-WAY LATENCY (2 processes, 2 topics, 1 role per handle) ==
   json metric key: `Topic_cross_process_oneway_pingpong` (identity used by the CI gate: `Topic_cross_process_oneway_pingpong@64B`)
   payload Probe64 = 64 B (POD=true) | ring capacity 256 slots | topology 1P/1C | cpu3 <-> cpu1 | pid 7411 <-> pid 7419
------------------------------------------------------------------------------------------------------------------------------------------
  metric (ns)                            n       min    median       p99     p99.9     p99.99         max |  n>p99.9  n>p99.99
------------------------------------------------------------------------------------------------------------------------------------------
  one-way, offset-free             1000000      80.1     120.2     145.2     480.7     3855.7     16038.7 |      965        99
  round trip (RTT)                 1000000     190.3     270.4     310.5     991.5     7901.7     32107.5 |      964       100
  RTT halved (one-way)             1000000      95.1     135.2     155.2     495.7     3950.8     16053.8 |      964       100
  leg A->B (apparent)              1000000      80.1     120.2     160.2     781.2     5498.1     31957.3 |      995        98
  leg B->A (apparent)              1000000      80.1     120.2     160.2     180.3     5428.0     19268.5 |      816        98
  responder turnaround             1000000      20.0      30.0      30.0      30.0       80.1     21682.1 |      138        99
------------------------------------------------------------------------------------------------------------------------------------------
  samples=1000000  warmup=200000  transferred=2000000  lost=0  invalid(TSC-skew)=0  poll cost=7.3 ns  window=0.312 s
  measured window: minor_faults=0 major_faults=0 ctx_switches vol=62 invol=5
  core freq cpu3   (measured window only): start=n/a end=n/a min=0MHz max=0MHz mean=0MHz spread=0.00% (n=0)
  core freq cpu1   (measured window only): start=n/a end=n/a min=0MHz max=0MHz mean=0MHz spread=0.00% (n=0)
  cross-core TSC offset theta_hat = +0.0 ns (clock B - clock A). The `one-way, offset-free` row does not depend on it.

== cross-process ping-pong -- ONE-WAY LATENCY (2 processes, 2 topics, 1 role per handle) ==
   json metric key: `Topic_cross_process_oneway_pingpong` (identity used by the CI gate: `Topic_cross_process_oneway_pingpong@1024B`)
   payload Probe1k = 1024 B (POD=true) | ring capacity 256 slots | topology 1P/1C | cpu3 <-> cpu1 | pid 7411 <-> pid 7421
------------------------------------------------------------------------------------------------------------------------------------------
  metric (ns)                            n       min    median       p99     p99.9     p99.99         max |  n>p99.9  n>p99.99
------------------------------------------------------------------------------------------------------------------------------------------
  one-way, offset-free             1000000     115.2     175.3     220.3     535.8     5102.6     16224.0 |      970        99
  round trip (RTT)                 1000000     260.4     370.5     470.7    1101.6    11487.0     32558.2 |      982       100
  RTT halved (one-way)             1000000     130.2     185.3     235.3     550.8     5743.5     16279.1 |      982       100
  leg A->B (apparent)              1000000     110.2     170.3     230.3     851.3     7921.7     32297.8 |      981       100
  leg B->A (apparent)              1000000     110.2     170.3     240.4     280.4     5658.4     25718.1 |      953        96
  responder turnaround             1000000      20.0      30.0      30.0      30.0      120.2     13970.7 |      239        97
------------------------------------------------------------------------------------------------------------------------------------------
  samples=1000000  warmup=200000  transferred=2000000  lost=0  invalid(TSC-skew)=0  poll cost=24.7 ns  window=0.446 s
  measured window: minor_faults=0 major_faults=0 ctx_switches vol=88 invol=2
  core freq cpu3   (measured window only): start=n/a end=n/a min=0MHz max=0MHz mean=0MHz spread=0.00% (n=0)
  core freq cpu1   (measured window only): start=n/a end=n/a min=0MHz max=0MHz mean=0MHz spread=0.00% (n=0)
  cross-core TSC offset theta_hat = +0.0 ns (clock B - clock A). The `one-way, offset-free` row does not depend on it.

== cross-process paced stream -- ONE-WAY LATENCY (10000 ns publish period, publisher -> subscriber, direct) ==
   json metric key: `Topic_cross_process_oneway_paced_stream` (identity used by the CI gate: `Topic_cross_process_oneway_paced_stream@64B`)
   payload Probe64 = 64 B (POD=true) | ring capacity 256 slots | topology 1P/1C | cpu3 <-> cpu1 | pid 7411 <-> pid 7423
------------------------------------------------------------------------------------------------------------------------------------------
  metric (ns)                            n       min    median       p99     p99.9     p99.99         max |  n>p99.9  n>p99.99
------------------------------------------------------------------------------------------------------------------------------------------
  one-way (assumes TSC coherence)    200000      80.1     130.2     230.3    1642.4   203911.8    204272.3 |      200        20
------------------------------------------------------------------------------------------------------------------------------------------
  samples=200000  warmup=20000  transferred=200000  lost=0  invalid(TSC-skew)=0  poll cost=5.0 ns  window=2.000 s
  measured backlog: 0.01 messages deep on the median sample (ring holds 256); delivered 100.0% of transferred
  measured window: minor_faults=0 major_faults=0 ctx_switches vol=395 invol=19
  core freq cpu3   (measured window only): start=n/a end=n/a min=0MHz max=0MHz mean=0MHz spread=0.00% (n=0)
  core freq cpu1   (measured window only): start=n/a end=n/a min=0MHz max=0MHz mean=0MHz spread=0.00% (n=0)
  this row assumes cross-core TSC coherence; theta_hat measured in the ping-pong at this size = +0.0 ns (add to correct).

== cross-process paced stream -- ONE-WAY LATENCY (10000 ns publish period, publisher -> subscriber, direct) ==
   json metric key: `Topic_cross_process_oneway_paced_stream` (identity used by the CI gate: `Topic_cross_process_oneway_paced_stream@1024B`)
   payload Probe1k = 1024 B (POD=true) | ring capacity 256 slots | topology 1P/1C | cpu3 <-> cpu1 | pid 7411 <-> pid 7425
------------------------------------------------------------------------------------------------------------------------------------------
  metric (ns)                            n       min    median       p99     p99.9     p99.99         max |  n>p99.9  n>p99.99
------------------------------------------------------------------------------------------------------------------------------------------
  one-way (assumes TSC coherence)    200000     100.1     150.2     300.4    1432.1    10796.0    109041.4 |      200        20
------------------------------------------------------------------------------------------------------------------------------------------
  samples=200000  warmup=20000  transferred=200000  lost=0  invalid(TSC-skew)=0  poll cost=9.2 ns  window=2.000 s
  measured backlog: 0.02 messages deep on the median sample (ring holds 256); delivered 100.0% of transferred
  measured window: minor_faults=0 major_faults=0 ctx_switches vol=395 invol=17
  core freq cpu3   (measured window only): start=n/a end=n/a min=0MHz max=0MHz mean=0MHz spread=0.00% (n=0)
  core freq cpu1   (measured window only): start=n/a end=n/a min=0MHz max=0MHz mean=0MHz spread=0.00% (n=0)
  this row assumes cross-core TSC coherence; theta_hat measured in the ping-pong at this size = +0.0 ns (add to correct).

== cross-process unpaced stream -- QUEUE DELAY / BACKPRESSURE, NOT LATENCY (publisher unthrottled; each sample includes its wait in the ring) ==
   !! Every number in this block is a PUBLISH-TO-RECEIVE DELAY that INCLUDES
   !! time spent waiting in the ring. It is a backpressure/throughput figure.
   !! Do not quote it as an IPC latency; that is the exact mistake the retired
   !! `Topic_cross_process` metric shipped for months. Read `measured backlog`
   !! below: ~1 message means the subscriber kept up and no queue formed.
   json metric key: `Topic_cross_process_unpaced_queue_delay` (identity used by the CI gate: `Topic_cross_process_unpaced_queue_delay@64B`)
   payload Probe64 = 64 B (POD=true) | ring capacity 256 slots | topology 1P/1C | cpu3 <-> cpu1 | pid 7411 <-> pid 7447
------------------------------------------------------------------------------------------------------------------------------------------
  metric (ns)                            n       min    median       p99     p99.9     p99.99         max |  n>p99.9  n>p99.99
------------------------------------------------------------------------------------------------------------------------------------------
  publish->receive incl. queue      200000      70.1     110.2     170.3    4226.3     6970.3     33850.1 |      200        20
------------------------------------------------------------------------------------------------------------------------------------------
  samples=200000  warmup=20000  transferred=200000  lost=0  invalid(TSC-skew)=0  poll cost=7.7 ns  window=0.015 s
  measured backlog: 1.45 messages deep on the median sample (ring holds 256); delivered 100.0% of transferred
  measured window: minor_faults=0 major_faults=0 ctx_switches vol=3 invol=0
  core freq cpu3   (measured window only): start=n/a end=n/a min=0MHz max=0MHz mean=0MHz spread=0.00% (n=0)
  core freq cpu1   (measured window only): start=n/a end=n/a min=0MHz max=0MHz mean=0MHz spread=0.00% (n=0)
  this row assumes cross-core TSC coherence; theta_hat measured in the ping-pong at this size = +0.0 ns (add to correct).

SUMMARY — one row per published JSON metric (nanoseconds, wall clock)
------------------------------------------------------------------------------------------------------------------------------------------
  json metric key                                what it measures                  n    median       p99     p99.9     p99.99         max
------------------------------------------------------------------------------------------------------------------------------------------
  Topic_cross_process_oneway_pingpong@64B        ONE-WAY latency             1000000     120.2     145.2     480.7     3855.7     16038.7
  Topic_cross_process_oneway_pingpong@1024B      ONE-WAY latency             1000000     175.3     220.3     535.8     5102.6     16224.0
  Topic_cross_process_oneway_paced_stream@64B    ONE-WAY latency              200000     130.2     230.3    1642.4   203911.8    204272.3
  Topic_cross_process_oneway_paced_stream@1024B  ONE-WAY latency              200000     150.2     300.4    1432.1    10796.0    109041.4
  Topic_cross_process_unpaced_queue_delay@64B    QUEUE DELAY (not latency)    200000     110.2     170.3    4226.3     6970.3     33850.1
------------------------------------------------------------------------------------------------------------------------------------------
  Instrumentation is INCLUDED in every figure and never subtracted.
  `Topic_cross_process_oneway_pingpong` is the true one-way figure: it assumes
  nothing about cross-core TSC synchronisation and excludes the responder's turnaround.
  `RTT halved (one-way)` in the per-scenario tables is a ROUND TRIP DIVIDED BY TWO; it is
  the row to compare against any competitor publishing a halved ping-pong.
  `Topic_cross_process_unpaced_queue_delay` is a BACKPRESSURE figure: it includes
  however long the message waited in the ring, so it scales with the achieved backlog
  (printed in its scenario block) and is NOT a latency.
  A figure without its payload size, ring depth and topology is not comparable to anything.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!! THE SAME WARNINGS AGAIN — they apply to every number above
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!  CPU governor is ?/?, not `performance`. RDTSC is a wall clock while the work costs a fixed number of CORE cycles, so every nanosecond below scales with a frequency the governor is free to change. iceoryx2's maintainers measured ~300 ns of difference from the governor alone.
!!  No isolcpus= on the kernel command line: the benchmark cores are shared with everything else on this box. The involuntary context-switch count in each scenario block is the mechanism behind the far tail.
!!  pingpong (64 B) `one-way, offset-free`: p99.99 rests on 99 observations. Below ~100 exceedances it is a max wearing a percentile's name; +/-5% needs ~400 (--iterations 4000000).
!!  pingpong (1024 B) `one-way, offset-free`: p99.99 rests on 99 observations. Below ~100 exceedances it is a max wearing a percentile's name; +/-5% needs ~400 (--iterations 4000000).
!!  paced_stream (64 B) `one-way (assumes TSC coherence)`: p99.99 rests on 20 observations. Below ~100 exceedances it is a max wearing a percentile's name; +/-5% needs ~400 (--iterations 4000000).
!!  paced_stream (1024 B) `one-way (assumes TSC coherence)`: p99.99 rests on 20 observations. Below ~100 exceedances it is a max wearing a percentile's name; +/-5% needs ~400 (--iterations 4000000).
!!  unpaced_queue (64 B): measured backlog is only 1.45 messages — the subscriber kept up, so the ring never filled and `Topic_cross_process_unpaced_queue_delay` is NOT a backpressure figure on this run. It is a publish-to-receive delay with no queueing in it, and it is still not a latency (see the paced and ping-pong rows for that).
!!  unpaced_queue (64 B) `publish->receive incl. queue`: p99.99 rests on 20 observations. Below ~100 exceedances it is a max wearing a percentile's name; +/-5% needs ~400 (--iterations 4000000).
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

Gate report written to /home/runner/work/horus/horus/benchmarks/gate-reports/cross_process-3.json
  keys: Topic_cross_process_oneway_pingpong@64B, Topic_cross_process_oneway_pingpong@1024B, Topic_cross_process_oneway_paced_stream@64B, Topic_cross_process_oneway_paced_stream@1024B, Topic_cross_process_unpaced_queue_delay@64B
  NOTE: `Topic_cross_process` is retired. A baseline window that still holds it
  will report it as missing and fail the gate until the window is refreshed —
  that is intended, and is why the key was not silently reused.
  Raw samples are omitted (pass --raw-in-json, or --raw-dir for a lossless cycle dump).
  8 warning(s) apply to these numbers and are NOT in this file; use --detail-json.
╔══════════════════════════════════════════════════════════════════╗
║        HORUS Robotics Message Types Benchmark                    ║
╠══════════════════════════════════════════════════════════════════╣
║  Testing real-world robotics message latency (REP 2014)          ║
╚══════════════════════════════════════════════════════════════════╝

Platform: AMD EPYC 9V45 96-Core Processor (4 cores)
Iterations: 50000
Warmup: 5000

╔═════════════════════════════════════════════════════════════════════════════╗
║ Message Type    │ Size (bytes) │ Typical Rate │ Use Case                    ║
╠═════════════════════════════════════════════════════════════════════════════╣
║ CmdVel          │           16 │ 1000+ Hz     │ Velocity control commands   ║
║ Imu             │          296 │ 500+ Hz      │ IMU sensor fusion           ║
║ LaserScan       │         1480 │ 10-40 Hz     │ 2D lidar navigation         ║
║ JointCommand    │         1032 │ 500+ Hz      │ Manipulator control         ║
╚═════════════════════════════════════════════════════════════════════════════╝


[Topic] Running benchmarks — timing tx.send() enqueue cost only
        (cross-thread consumer drains in the background; no sample
         waits for delivery, so these are NOT end-to-end latencies)
─────────────────────────────────────────────────
  CmdVel              16 bytes │ median:      30ns │ p99:      60ns │ CV: 1.7252
  Imu                304 bytes │ median:      40ns │ p99:      80ns │ CV: 2.0224
  LaserScan         1480 bytes │ median:      50ns │ p99:     100ns │ CV: 1.3973
  JointCommand       928 bytes │ median:      60ns │ p99:     110ns │ CV: 0.7867

╔══════════════════════════════════════════════════════════════════════════════════════════╗
║           SUMMARY BY MESSAGE TYPE — send() ENQUEUE COST, not end-to-end latency           ║
╠══════════════════════════════════════════════════════════════════════════════════════════╣
║ Message      │   Size │ Median (ns) │ p99 (ns) │    CV   │ Misses │ Sustained rate       ║
╠══════════════════════════════════════════════════════════════════════════════════════════╣
║ CmdVel       │     16 │          30 │       60 │  1.7252 │      0 │      7.50 M msg/s     ║
║ Imu          │    304 │          40 │       80 │  2.0224 │      0 │      4.50 M msg/s     ║
║ LaserScan    │   1480 │          50 │      100 │  1.3973 │      0 │      2.52 M msg/s     ║
║ JointCommand │    928 │          60 │      110 │  0.7867 │      0 │      3.90 M msg/s     ║
╚══════════════════════════════════════════════════════════════════════════════════════════╝
Sustained rate is wall-clock over the measured loop, so it includes the batch
drain-waits. It is a publish rate for this producer/consumer pair, not a bus limit.

╔══════════════════════════════════════════════════════════════════╗
║              REAL-TIME SUITABILITY ANALYSIS                      ║
╠══════════════════════════════════════════════════════════════════╣
║ These gates grade the p99 of the send() ENQUEUE COST against one ║
║ whole control period. They are not end-to-end control-loop       ║
║ guarantees: a real loop also pays the receive side, the node's   ║
║ compute, and the scheduler's jitter, none of which is measured   ║
║ here. The headroom is printed so a pass by four orders of        ║
║ magnitude cannot be read as a tight result.                      ║
╠══════════════════════════════════════════════════════════════════╣
║ CmdVel    (1kHz  control): ~ PASS by 16667x — gate not discriminating (p99=60ns vs 1000000ns)
║ CmdVel    (10kHz control): ~ PASS by 1667x — gate not discriminating (p99=60ns vs 100000ns)
║ Imu       (500Hz fusion) : ~ PASS by 25000x — gate not discriminating (p99=80ns vs 2000000ns)
║ LaserScan (40Hz  lidar)  : ~ PASS by 250000x — gate not discriminating (p99=100ns vs 25000000ns)
║ JointCmd  (500Hz control): ~ PASS by 18182x — gate not discriminating (p99=110ns vs 2000000ns)
╚══════════════════════════════════════════════════════════════════╝

Results written to: /home/runner/work/horus/horus/benchmarks/gate-reports/robotics-3.json


Gated against a rolling window of recent main runs. Each metric's band is max(relative, center + k·sigma_observed, absolute floor), so it is derived from the noise the baseline itself measured — where a band is printed wider than the change you care about, this run did not detect a regression and could not have. Thresholds and their reasoning: RegressionPolicy in benchmarks/src/output.rs.
Commit: 7dabaf8

@dependabot @github

dependabot Bot commented on behalf of github Nov 16, 2026

Copy link
Copy Markdown
Contributor Author

Looks like these dependencies are updatable in another way, so this is no longer needed.

@dependabot dependabot Bot closed this Nov 16, 2026
auto-merge was automatically disabled November 16, 2026 06:26

Pull request was closed

@dependabot
dependabot Bot deleted the dependabot/cargo/rust-minor-aa28ceb857 branch November 16, 2026 06:26
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

dependencies Pull requests that update a dependency file rust Pull requests that update rust code

Projects

None yet

Development

Successfully merging this pull request may close these issues.

0 participants