4 commits, squashed; the originals stay reachable from tag archive/merge-main-develop.
- d13300b0a jyviko: cargo fmt: the format-40 change and the rebase's export list
- 873124c49 jyviko: CI: develop runs like main; SMLib is fetched after the Rust cache is restored
- 6517d1ee6 jyviko: Arabic and Hebrew catalogs: #4414's thread direction labels out, as in every other catalog
- fddb57c0d jyviko: Thread dialog: its labels in the catalogs; Arabic and Hebrew pending messages compare in display order
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
4 commits, squashed; the originals stay reachable from tag archive/merge-main-develop.
- 4e26dc640 jyviko: The command line and the web app link SMLib by default too
- 826f36a79 jyviko: CI and releases fetch SMLib before building
- 8aea8e5a8 jyviko: Format 40: a thread's hand on tapped holes too; v0.24.0's handedness opens as hand
- 652d3a2cb jyviko: ui: main's browser picks (#4774) read the placed model through session_world
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
4 commits, squashed; the originals stay reachable from tag archive/merge-main-develop.
- 7024ceec8 jyviko: CI: no runs for pull requests into kernel-smlib for now
- 92d433ca9 jyviko: AGENTS.md: the kernel-smlib branch's rules (ADR 0002)
- ee5deffaa jyviko: brep: Curve2 pcurve type and the truck pcurve entry point
- 26ab8981d jyviko: kernel: exact pcurves of lines and splines on planes
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
AGENTS.md drops from 16.8k to 3.7k characters: the hard rules, each naming the
check that enforces it, and a table of what to read for each kind of work.
The rest moves into eleven new skills; the five existing ones stay.
cargo xtask skills fails when AGENTS.md or a SKILL.md cites a path, an xtask
command or a skill that does not exist, when the router table and skills/
disagree, or when the router or a new skill grows past its size cap. It runs
in cargo xtask ci and the repository checks job.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The first test in each GPU path shard compiles the path pipeline cold
on lavapipe (335-502 s on 16 vCPU, 665 s on 8, against the 600 s setup
budget); the later ones read Mesa's disk cache in about 30 s. Put that
cache on a sticky disk that main commits, so the first test reads it
too until the shaders change.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
On 8 vCPU the first merge-queue run after #7066 failed: one gpu path
test went past the 600 s pipeline-setup budget (665 s; 204 s for the
same test on the PR run) and shard 2 took 35 minutes. lavapipe compiles
on every core, so the shards move to 16 vCPU, and four count partitions
(27 tests each) replace three hash ones (41/40/27).
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The 108 gpu path:: tests run one at a time (gpu-path test group) and took
about 36 of Native Linux tests' 45 minutes in a merge-queue run. They now
run in a GPU path tests job, three hash partitions on 8 vCPU, beside the
other tests; Native Linux tests leaves exactly them out. A `gpu` output of
Select checks drives both the job's `if` and the `ci` check.
The Windows jobs now run in a merge group only when its diff reaches what
they exercise (the pull request patterns and the Cargo.lock graphics-stack
check), not on pushes to main, and still nightly and by hand.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* CI: duplicate-PR check and a required 'Changes a default' line (Refs #6439)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* PR checks: enforce the defaults line only on PRs opened from 2026-10-25 (Refs #6439)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Vertical history strip: left of the browser, compact rows, hollow-circle marker, collapsed rail hover
* Vertical history strip: seam handle to expand and collapse names, drag to resize with snap
* Vertical history strip: no empty gap below the transport
* Show the history seam arrow only on hover (#2028)
* claim: #2021
* Redesign the Assistant chat panel with shared UI components
* Show Assistant in the shared floating panel shell
* Extend floating Assistant to viewport height and shift navigation
* Dock Assistant before history and match docking preview bounds
* Align Assistant top margin with floating shell cards
* Extract a shared panel drag grip
* Rebuild responsive history sizing and three-way docking
* Share browser and History collapse hover styling
* Show the History title as soon as its text fits
* Animate stacked browser and History expansion and collapse
* Make panel folding responsive and independent of frame delays
* Let folding panel contents shrink continuously to header height
* Use four explicit History transport actions in every layout
* Resize the shared Browser and History column from either card
* claim: shared compact-mode containers (#2029)
* Centralize layout-aware UI containers and enforce their use
* Keep docked Assistant at the outer edge beside tool sheets
* Use shared surface policy for Assistant tool space
* Let tool sheets drag over shell panels while avoiding Assistant
* Update Parameters drag bounds assertion for shell overlap
* Keep dragged tool sheets below ribbon and workspace selector
* Enforce universal UI containers in mandatory repository checks
* Remove legacy report windows and require shared window composition
* Remove retired UI styling and require shared controls
* claim: compact layout preference (#2029)
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
While a new geometry kernel is evaluated, a workflow labels new issues, pull
requests and discussions about kernel work kernel-paused, so a maintainer finds
them and answers in person. A pull request counts when it changes
crates/kernel/ or vendor/truck*, anything else when its title names a kernel
operation. Crash and data-loss reports and anyone who can push are left alone.
CONTRIBUTING.md says so.
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* CI: build and run the tests in one job, sccache on a Blacksmith sticky disk
The tests job compiles and runs the tests itself on the 32 vCPU runner, so a pull request or merge
group uploads and downloads no 3 GB archive (the download took 45 to 385 s per shard, and once
14 minutes) and starts beside the repository checks instead of after them. The archive is built
only for the nightly and manual jobs that need it. sccache moves from GitHub's 10 GB cache, where
about 1100 per-ref entries were evicting each other, to a sticky disk that only main commits.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* TEMP: let this branch save caches so the warm run can be measured
* Revert "TEMP: let this branch save caches so the warm run can be measured"
This reverts commit 1bbbfcf5ff.
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* CI: give the slowest tests a shard each
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* CI: spread the slow tests on pull requests too, and keep both junit files
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* CI: pull requests skip the goldens and the action matrix sample unless the kernel changes
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Docs: which pull requests run the goldens
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* CI: run the tests in one job on a big runner, slowest first
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* CI: allow the 32 vCPU runner label in actionlint, read junit.xml in the golden hint
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The merge queue reads the required ci check from the newest run of a commit. Adding any label but
approved-for-ci starts a run whose jobs all skip, and its ci job was skipped too (as
'ci (ignored label)'), so a pull request that had passed could not enter the queue (#1850, #3182,
#3252). On such a run the ci job now waits for the commit's other runs to finish and passes only
when the newest one that ran its checks passed.
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Blacksmith's Windows runners start the app but never present a frame (no interactive desktop),
so the startup smoke failed on main after #3343. The shader job stays on Blacksmith.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* CI: merge-queue groups run on Blacksmith, test shards on 16 vCPU
The merge queue builds up to three groups at once, but every group's
checks, Clippy, doc and reference tests and all four shards ran on the one
self-hosted machine, so the groups waited for each other. Merge groups now
run on Blacksmith like pull requests (test archive built once, shards on
16 vCPU). Pushes to main keep the self-hosted runners and their warm build
directories, and a push the queue already tested still runs only the
repository checks.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
* CI: every job on Blacksmith: pushes to main and the Windows jobs too
Pushes to main ran on the one self-hosted machine and the Windows jobs on
GitHub's runners. Both move to Blacksmith (Windows Server 2025, 8 vCPU),
so nothing in ci.yml depends on a single box any more. With no
self-hosted job left, the local routing goes: no warm build directories,
every run builds the test archive once and the shards run it, and main
saves the build caches every other run restores.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
---------
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The self-hosted-untrusted pool is being retired for CI, so fork PRs whose
number ends in 7, 8 or 9 no longer route there. Build tests, the test shards
and the reference tests reduce to our own runners for main and the merge
queue, Blacksmith for everything else; the build's 90-minute gVisor timeout
goes. Blacksmith VMs are ephemeral and fork runs get no secrets.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Blacksmith runners report runner.environment == 'self-hosted', and a list of
our machines' name prefixes breaks when a box is added or renamed. Select
checks already decides where each job runs; its local and untrusted outputs
(and the trusted-run condition in the repository checks) now drive the warm
build directory, cache and lavapipe steps. Same results as before on every
event.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Re-applies #2237 (reverted in #2277 while the Blacksmith account awaited verification) on today's
ci.yml: every ubuntu-24.04 job or fallback runs on blacksmith-<N>vcpu-ubuntu-2404, sized as #2237
did (quick-compile, new since, gets 8), and steps that asked runner.environment now tell our own
machines apart by runner name, since Blacksmith runners report 'self-hosted' too. The trusted
self-hosted routing and the fork routing are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
From the review: git patch-id --stable ignores whitespace, so a conflict resolution that only
respaced the change (inside a string, or Python indentation) counted as no change and was
carried; --verbatim compares every byte. And a passing run is now taken only from this pull
request (by number, or for a fork, where GitHub lists none on the run, by head repository and
branch), not from another pull request with the same head commit.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
From the review of this PR:
- What counts as code lives only in scripts/ci/affected.py now: nothing selected and not full
means no code changed, on pull requests and pushes to main alike. The untested bash copy in
Select checks disagreed with it. Root files and folders a crate embeds now select that crate
(the licence texts and NOTICE: the engine; skills/: the MCP server's tests), where before a
change to them ran no test at all. Other workflows and the bots' scripts select nothing.
- A push to main skips the native jobs only when nothing a build reads changed since the last
push to main whose run passed. Comparing with the push just before was unsafe: a newer push
cancels a waiting run, so that commit may never have been tested.
- Fail fast applies to pull requests only. On main, nightly and Run workflow the test build and
the shards run regardless of the repository checks, so the action matrix and property tests
that hang off Build tests keep their signal. Doc tests (main and nightly only now) no longer
wait for the checks.
- tests/robustness/ turns on the goldens, so a change to goldens.json runs the seed tests.
- tests/oracle/ no longer selects the kernel, which never reads it.
- The workflow header and CONTRIBUTING describe what main and pull requests now run.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Merging main into a pull request (Merge branch 'main', or GitHub's Update branch) restarted the
whole run. Over three days 1,241 such merges started 966 runs and about 21,000 runner-minutes, and
872 of the merges brought in nothing that touched the pull request's files. Select checks now
recognises the case (scripts/ci/main_merge.py): HEAD is a merge whose other parent is on main, the
change's own patch is unchanged (git patch-id), and a ci.yml run of the change before the merge
passed with its test shards. Then only the repository checks run, the ci job names the passing run
in its summary, and main tests the merged result after it lands. Anything else, or any doubt,
runs the full set as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Doc tests move to main and nightly, beside the reference tests and the Windows jobs: on pull
requests they were the only failing job 3 times in about 7,900 runner-minutes over three days.
Build tests, the test shards and the doc tests now start only after the format and repository
checks pass; 212 attempts in those three days ran their heavy jobs beside a failed cheap check.
Also corrects the header, which still said pull requests run the Windows jobs and the reference
tests.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
- Pushes to main use the same code/no-code rule as pull requests (`touches_code` in Select
checks): a push changing only documentation, packaging, other workflows or the bots' scripts
runs the repository checks only. With the merge queue off, every push to main ran the full suite
and the Windows jobs, docs-only ones included.
- affected.py: test data under tests/ selects the packages that read it (robustness, hole matrix,
projection, taper: solidcraft-engine; oracle: the CLI, the kernel, xtask; action matrix: the
CLI, the engine, xtask), and a vendored crate the packages that use it (truck: the kernel and up;
egui-wgpu: the UI and the apps), instead of the whole workspace. tests/tools and unknown vendored
crates still select everything.
- Docker (the web app image) skips pushes touching only docs, test data or bot scripts; Pages skips
test data and bot scripts; Generated files runs only when the manifests, lockfile, command
registry, CLI or xtask change.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
* CI: heavy checks run after review (approved-for-ci label)
Pull requests without the approved-for-ci label run Select checks, the
format and repository checks and a quick cargo check. The label runs the
full suite, and every push after that does too. Other labels start no work
and never cancel a run. Main, merge groups, nightly and Run workflow are
unchanged.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* CI: approved fork PRs run heavy jobs on self-hosted-fast in private directories
An approved-for-ci label on a fork pull request sends Clippy, Build tests,
the test shards and Doc tests to the trusted runners, with private copies
of the build directory, Cargo home and rustup home that are deleted after
the job. Only the labeling event qualifies; later pushes use the old fork
routing. The skipped run of an unrelated label names its ci job differently.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Revert "CI: approved fork PRs run heavy jobs on self-hosted-fast in private directories"
This reverts commit 439f2f7dab.
* CI: name the ci job differently on an ignored label's skipped run
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
the format and repository checks job still took a self-hosted-fast runner for every same-repo
pull request. It now follows the same rule, and the notes at the top say so.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Pull requests that can change a robustness golden check every seed but the 17 slowest, in the
same nextest run as the other tests instead of a second step after it (16-30 minutes on its own).
The slow seeds (about four fifths of the goldens' time) still run in the merge queue, on main and
nightly. A failed golden still prints how to re-record it, and junit.xml now holds the whole run.
- The slowest tests start first (nextest priority), so a shard no longer ends with one of them
running alone.
- tspline::tests::hostile_tsplines_never_panic joins the heavy tests pull requests skip (up to
12 minutes on a GitHub runner; the same kind of check as hostile_params_never_panic).
- The ci profile builds dependencies without line tables: the test archive pull requests' shards
download goes from 3.4 GB to 2.2 GB.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
On main and in the merge queue every test shard ran on the self-hosted machine but waited for
the GitHub-hosted Build tests job and then spent about 12.5 minutes downloading its 3.2 GB
archive (four shards at once over one connection). There, each shard now builds the tests in
the runner's warm build directory (the existing build-tests one) and runs its slice directly;
Build tests runs only where the shards run on GitHub's or the untrusted runners.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Same-repo pull requests ran their heavy jobs on the three fast runners too, and the merge
queue's test shards waited behind them for hours. Pull requests now use the GitHub-hosted path.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
#2278 dropped the golden sample from pull requests, and two geometry PRs then
broke main together. Pull requests that touch the kernel, engine, doc, sketch
or geom crates now run every seed, split over the shards, with the command to
re-record in the failure output.
Fork PRs whose number ends in 7, 8 or 9 stay on the untrusted pool; the rest
build and test on GitHub's runners, which have free slots.
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
Build tests on the untrusted runners (fork PRs, gVisor on shared boxes) fail intermittently with
rustc killed mid-crate ("exit status: 2", no diagnostic) on the biggest crates: solidcraft-engine,
ui-egui. Nothing limits Cargo's parallelism, so it starts one rustc per core the sandbox reports.
scripts/ci/build-jobs.sh sets CARGO_BUILD_JOBS from the memory available (and the cgroup limit,
where lower) at 3 GB per job, capped at nproc, and logs the numbers it saw.
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Fork Build tests jobs were hitting the 45-minute limit while uploading their archive, after
the build had finished, and every one had to start over.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The trusted runs' test archive is still built on GitHub's ubuntu-24.04
runners; minisforum's newer glibc runs it unchanged, so shards 3-4 can join
1-2 on the self-hosted runners.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Fork pull requests now run on the untrusted self-hosted runners, which start
every job empty. Let them restore main's rust-cache like GitHub's runners do
(save-if stays main only, so a fork job never writes a cache).
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Lists and Packaging lint each took their own runner slot on every PR push. The list checks now run
in the existing checks job; Lists stays on main. Packaging lint skips PRs that only touch icons.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
With Windows off pull requests, GitHub's 20 runners sat idle while every
fork job queued for the self-hosted pool. Select checks, repository
checks, Clippy, doc tests and the summary go back to GitHub's runners;
Build tests and the test shards stay on the isolated self-hosted pool.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Pull requests now run the repository checks, Clippy, the affected
crates' tests in 2 shards, and the doc tests. Windows, the reference
tests and the robustness goldens run on pushes to main and in the
nightly, where a break is fixed forward. This is the split Bevy, Zed
and Ruff use, and it cuts each pull request's CI to about a third.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Blacksmith isn't provisioning runners: its dashboard says the account
"requires additional verification", so every job labelled blacksmith-*
stays queued with no runner. Fork PRs' Linux jobs go back to the
self-hosted-untrusted pool (#2249/#2252) and the rest to GitHub's runners
until Blacksmith verifies the account; then #2237 can be re-applied.
This reverts commit dc40f051b1.
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Fork pull requests, the nightly run, Run workflow, the test archive and
test shards 3-4 move from ubuntu-24.04 to blacksmith-<N>vcpu-ubuntu-2404
(16 vCPU for Clippy and the test archive, 8 for test shards, doc tests,
reference tests and cross-compiles). Trusted runs keep the self-hosted
routing. Blacksmith runners count as self-hosted to GitHub, so steps now
tell our own machines apart by runner name.
TEMPORARY: this branch routes its own run like a fork pull request.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Every fork run waited in GitHub's shared queue just to start. The gate
and summary jobs now run on the isolated runners as well; the local
output stays false for forks so later jobs never pick the trusted pool.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Fork PRs were waiting hours for GitHub's shared runners. Their Linux
jobs (checks, Clippy, Build tests, the test shards, Doc tests, Reference
tests) now run on the ephemeral, network-jailed self-hosted-untrusted
runners, without the trusted runners' warm build directories.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Pull requests from branches of this repository share the work with the self-hosted runners like main does; fork pull requests stay on GitHub's runners. Select checks and ci run on a light self-hosted runner; the repository checks, Clippy, doc tests, reference tests and test shards 1-2 on the fast ones. The test archive is still built on GitHub's runners (glibc).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The merge queue runs the full suite on exactly the commit it then puts on
main, and main's push run then ran the full suite again on the same files.
Select checks now asks the Actions API whether a passing merge_group CI run
exists for the pushed commit; if so the push runs only the repository checks.
A push that skipped the queue, or an API that cannot answer, still runs
everything. The nightly still runs the whole suite.
Pull requests also run less: the reference tests only when solidcraft-cli (or
anything it depends on) is affected, and doc tests only for the affected
packages that have a library. The ci check requires the reference tests only
when they were selected.
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
jyviko