mtop
Build times (auto-updated)
- i686: 28s (2026-11-01 09:36 UTC, MAKEFLAGS=-j6)
- x86_64: 23s (2026-11-01 09:36 UTC, MAKEFLAGS=-j6)
- Category:
local-install/ (not maintained/adapted by us – plain AUR
package, maintainer Christian Hesse eworm). Already installed on this
host for local use (via yay); rebuilt/published to archlinuxaba 2026-10-26
so the cluster has it too, per the standard
[[feedback_local_install_rebuilds]] workflow (fresh AUR clone over the old
/data/INSTALL/mtop checkout, repo_release.sh, backup deleted only after
publish succeeded).
0.6.6-4, no local edits – builds clean out of the box, arch=('i686'
'x86_64') already literal in the upstream PKGBUILD (no $CARCH trap).
Small Perl ExtUtils::MakeMaker build (perl-dbd-mysql, perl-curses
deps), own check() test suite passes trivially (t/use.t).- namcap notes from this build (not acted on, upstream’s own PKGBUILD, not
ours to fix): SPDX license should be
GPL-1.0-or-later not bare GPL;
perl-dbd-mysql/perl-curses flagged as “may not be needed” (false
positive – script runtime deps namcap’s static analysis can’t see).
- 2026-10-26: first build+publish for both i686 and x86_64, clean on the
first try, no PKGBUILD changes needed.
- 2026-10-29: routine rebuild as part of a perl-5.44 ABI rebuild batch
(user-identified via
pacman -Qqo '/usr/lib/perl5/5.42' listing stale
local-install packages). Fresh re-clone + repo_release.sh, same
0.6.6-4, no content change, clean on first try.
- 2026-11-01: same perl-ABI rebuild situation recurred (new
/usr/lib/
perl5/5.42 stale-link batch, same detection method) – this time
explicitly bumped pkgrel 4 -> 4.1 (X.Y, per
[[feedback_local_install_rebuilds]]/[[feedback_pkgrel_vs_pkgver]]:
local-install content-free rebuilds always get X.Y, unlike the
2026-10-29 rebuild above which kept the plain 4). Fresh re-clone,
built+published cleanly both archs at 0.6.6-4.1, no content change.