Repository navigation
Conversation
Import the iwd and libell source recipes from Fedora Rawhide and lock their kernel.org release tarballs by SHA-512, with Fedora's lookaside as fallback. iwd links libell at runtime and Hummingbird does not ship it, so both build here; libell's ell.pc orders it first in the build graph. Move iwd to the backlog's resolved records. Hosted build, publication and runtime evidence remain outstanding in #486. Hive-Run: #486 Hive-Plan: dev-tools source-family import Hive-Spec: factory-build-backlog#closing-a-backlog-gap Assisted-by: Claude Opus 5.5 via Claude Code Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Danathar <Danathar@users.noreply.github.com>
Hive-Run: #486 Hive-Plan: dev-tools source-family import Hive-Spec: factory-build-backlog#closing-a-backlog-gap Assisted-by: Claude Opus 5.5 via Claude Code Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Danathar <Danathar@users.noreply.github.com>
|
@Danathar this PR is back in your court ( Review asks:
Reply here or push a new head to this branch (same sign-off); CI and the review swarm re-run automatically and the PR returns to the normal path. Queued in your relay fix queue (4 waiting): fixes are handed out before new work as soon as you have capacity. |
1 similar comment
|
@Danathar this PR is back in your court ( Review asks:
Reply here or push a new head to this branch (same sign-off); CI and the review swarm re-run automatically and the PR returns to the normal path. Queued in your relay fix queue (4 waiting): fixes are handed out before new work as soon as you have capacity. |
There was a problem hiding this comment.
Read the full diff against main and the backlog tooling in tools/factory_build_backlog.py. Two things worth fixing on the rebase the hive already asked for:
-
libellis imported but left pending in the backlog.config/factory-build-backlog.toml:261(main) still lists"libell"under[areas.long-tail]while this PR addspackages/libell/. The regeneratedreports/factory-build-backlog.jsonshows the effect:libellflips frompendingtoalready_recipe(statesalready_recipe13 → 14) instead of leaving the backlog.--checkpasses because the snapshot was regenerated, so nothing flags it. Movelibellto[resolved]withpr = 793, the same wayiwdis moved atconfig/factory-build-backlog.toml:127. #802 does exactly this move for the identicallibellrecipe — see my note there. -
The dist-git commits in the PR body table are swapped.
packages/libell/.hummingbird-upstream.jsonpins4eef73d6…andpackages/iwd/.hummingbird-upstream.jsonpinsa4b3fa90…; the body lists them the other way round. I checkedsrc.fedoraproject.org/rpms/{libell,iwd}/c/<sha>: each sha resolves only for the package the lock file names. The code is right; the body is not.
Overlap: #802 (libell only) is byte-identical in packages/libell/* to this PR except imported_at, and conflicts with it on the shared config/docs files. One of the two should land; this one carries the consumer and the lookaside fallback URLs.
Otherwise the recipe import and upstream-sources.json entries look correct to me: kernel.org primary with Fedora lookaside fallback, SHA-512 matching packages/*/sources.
Confidence: 3/5 (needs attention) — 1 medium finding; 2 perspectives requested changes
— hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88
Conflicts resolved by keeping both sides (source locks, skill docs); docs/architecture.md and reports/factory-build-backlog.json take main's numbers moved by this PR's own delta. Recipe, .packit.yaml and architecture counts checked equal. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01SnzFs53Qcer1xDySyxv1Vj Signed-off-by: Danathar <berths-purger-0g@icloud.com>
|
@Danathar this PR is back in your court ( Review asks:
Reply here or push a new head to this branch (same sign-off); CI and the review swarm re-run automatically and the PR returns to the normal path. Queued in your relay fix queue (5 waiting): fixes are handed out before new work as soon as you have capacity. |
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01SnzFs53Qcer1xDySyxv1Vj Signed-off-by: Danathar <berths-purger-0g@icloud.com>
|
@Danathar this PR is back in your court ( Review asks:
Reply here or push a new head to this branch (same sign-off); CI and the review swarm re-run automatically and the PR returns to the normal path. Queued in your relay fix queue (5 waiting): fixes are handed out before new work as soon as you have capacity. |
1 similar comment
|
@Danathar this PR is back in your court ( Review asks:
Reply here or push a new head to this branch (same sign-off); CI and the review swarm re-run automatically and the PR returns to the normal path. Queued in your relay fix queue (5 waiting): fixes are handed out before new work as soon as you have capacity. |
There was a problem hiding this comment.
Read both recipes against the tree; this looks correct to me.
What I verified at 271ad7d:
packages/iwd/iwd.specandpackages/libell/libell.specare byte-identical to Fedora dist-gitrpms/iwd@a4b3fa9andrpms/libell@4eef73d(fetched and diffed), matching the commits recorded in each.hummingbird-upstream.json.- SHA-512 in both
sourcesfiles andconfig/upstream-sources.jsonmatch the dist-gitsourcesentries and fresh downloads ofkernel.org/pub/linux/libs/ell/ell-0.83.tar.xzandkernel.org/pub/linux/network/wireless/iwd-3.12.tar.xz. iwd.spec:14BuildRequires: pkgconfig(ell) >= 0.43is satisfied bylibell.spec:55(%{_libdir}/pkgconfig/ell.pcinlibell-devel, version 0.83), so the graph can orderlibellbeforeiwdas the skill note says.asciidoc,python3-docutils,readline-devel,libtoolhave no recipe here, which is expected: the build root is Fedora 44 + Hummingbird (docs/skills/repeated-mistakes.md:709-712) and 23 existing specs already BRasciidoc.Recommends: wireless-regdb(iwd.spec:21) has a recipe underpackages/wireless-regdb.
Same merge note as on #786: #786, #789, #793 and #808 all regenerate the same lines in docs/architecture.md, config/factory-build-backlog.toml, reports/factory-build-backlog.json and .packit.yaml; whichever lands later needs the count/report tools re-run on rebase. The recipes themselves do not overlap.
Confidence: 5/5 (safe)
— hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88
|
@Danathar the hive has adopted this PR ( |
|
@Danathar this PR has a new head without confirmed hive push evidence, so the hive has handed this PR back to you ( |
|
@Danathar this PR is back in your court ( Review asks:
Reply here or push a new head to this branch (same sign-off); CI and the review swarm re-run automatically and the PR returns to the normal path. Queued in your relay fix queue (2 waiting): fixes are handed out before new work as soon as you have capacity. |
|
@Danathar the hive has adopted this PR ( |
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01SnzFs53Qcer1xDySyxv1Vj Signed-off-by: Danathar <berths-purger-0g@icloud.com>
|
@Danathar this PR has a new head without confirmed hive push evidence, so the hive has handed this PR back to you ( |
|
@Danathar this PR is back in your court ( Review asks:
Reply here or push a new head to this branch (same sign-off); CI and the review swarm re-run automatically and the PR returns to the normal path. Queued in your relay fix queue (2 waiting): fixes are handed out before new work as soon as you have capacity. |
|
stage 095-pre-merge-soak → 072-returned-to-contributor: conflicting |
|
@Danathar the hive has adopted this PR ( |
|
Rebased onto main and continued in 827 (this PR's head stopped syncing with its branch). |
Summary: This adds build recipes for iwd, a Wi-Fi connection service that Bluefin ships and Utah (the Bluefin-like image this repository builds packages for) does not have yet. It also adds the small system library iwd needs to run, which the base system Utah builds on does not provide. This only prepares the recipes. Whether the package builds, gets published and works on a real Wi-Fi card still has to be checked, so the tracking issue stays open.
Refs #486 — partial: this resolves only
iwd. Remaining in the batch:fedora-release-common,fedora-release-identity-silverblue,fedora-release-ostree-desktop,fedora-release-silverblue,generic-logos(branding, which needs a linked maintainer decision),fuse-encfs(retired in Fedora Rawhide) andhfsplus-tools(upstream source host returns 404). Consumer tracking: projectbluefin/utah#392 (Utah triage listsiwdasplannedunder projectbluefin/utah#262).What changed
libell0.83a4b3fa900dfacba2961ef7e65f816789c31e816ehttps://www.kernel.org/pub/linux/libs/ell/ell-0.83.tar.xziwd)iwd3.124eef73d6390c72617d779cceb4d20063cbbe6dcchttps://www.kernel.org/pub/linux/network/wireless/iwd-3.12.tar.xziwdtools/import_rawhide.py. Fedora'siwd/.gitignoreignoresiwd.specandsources, so those two were staged withgit add -f(same tracked-but-ignored pattern ashplipandibus).config/upstream-sources.json: kernel.org primaryurl, Fedora lookaside only infallback_urls. I downloaded both upstream tarballs and checked that their SHA-512 matches the pins inpackages/*/sources. Both fallback URLs return HTTP 200..packit.yamlanddocs/architecture.mdwere regenerated withtools/render_packit_config.py --writeandtools/render_architecture_counts.py --write.config/factory-build-backlog.toml:iwdmoved from[areas.dev-tools]to[resolved], andreports/factory-build-backlog.jsonwas regenerated (227 backlog entries, 212 pending).docs/skills/factory-build-backlog.md: new section covering iwd and libell, the unrelated Fedoraellpackage, the.gitignoreissue, and whyfuse-encfsandhfsplus-toolscan't be imported straight from Rawhide.Dependencies
iwdwith--enable-external-ell, so it needslibell.soat runtime. The current Hummingbirdprimary.xmlhas nolibell/ell, so it builds here.libell-devellists%{_libdir}/pkgconfig/ell.pcin its%files, sotools/build_graph.pyorders it beforeiwd'spkgconfig(ell) >= 0.43without anystage.elldist-git is an unrelated header-only C++ library. I did not import it.dbus,systemd,readline) are in Hummingbird:dbus-1:1.16.2,systemd-261,readline-8.3(hum1).wireless-regdbis onlyRecommends.asciidocandpython3-docutilsare build-time only and come from the build root, as they do forbluezandgit.Evidence still missing (not claimed)
libellandiwdiwctl --version, startingiwd.serviceon Wi-Fi hardware)Validation
python3 tools/factory_build_backlog.py --check: consistent (227 entries, 212 pending)just check: passed, includingpytest(1390 passed, 2 skipped)pre-commit run --all-files: not run becausepre-commitisn't installed on this host. CI runs it.🤖 Generated with Claude Code
— hive: backend=claude model=claude-opus-5-5
🐝 Hive Agent:
contributor| SHA:4292aea4