optdb5.3

Build times (auto-updated)

• x86_64: 6m10s (2026-09-10 05:46 UTC, serial)
• armv7h: 14m11s (2026-09-10 06:00 UTC, remote)

---

• Path: arch/private/optdb5.3
• 2026-07-15: build failure fixed and published, pkgrel 5 -> 6. Previously flagged as a “genuine compiler/source error” (incompatible function-pointer types – modern GCC strictness vs old C code): gcc 16.1.1 now errors (not just warns) on -Wincompatible-pointer-types by default, and this 2015-era Berkeley DB source (src/env/env_config.c) declares its dispatch tables with old K&R-style unprototyped function pointers (int (*func)();), which the compiler now treats as int (*)(void) and rejects when initialized with the real, typed function pointers (__env_add_data_dir etc).
• Fix: added -Wno-error=incompatible-pointer-types alongside the existing -Wno-error=implicit-function-declaration in build()’s CFLAGS. Packaging-only fix, no source/version change.
• Test-built clean with extra-x86_64-build, then built+signed+ published to the private repo (x86_64). See UPDATES.md’s 2026-07-15 entry for full detail.

2026-09-10: content-free rebuild for armv7h coverage, pkgrel 6 -> 6.1

arch=(x86_64 armv7h) was already correct, but only x86_64 had ever actually been published (the 2026-07-15 fix above) – armv7h had no local archbuild chroot available at the time. Now covered by the eurobuild5 remote-build board (see remote_arm_build_machines.md), so requested as a pure rebuild-for-coverage, no PKGBUILD content change. pkgrel 6 -> 6.1 (not plain 7) per the “content-free rebuilds get X.Y too” rule in memory/PKGREL_VS_PKGVER.md – this package carries local patches on top of Oracle’s real upstream tarball, same shape as adapted/ even though it’s filed under private/. Built cleanly on both x86_64 (6m10s) and armv7h (14m11s remote); no arch-specific fixes needed. Briefly raced adapted/cyrus-imapd2’s own armv7h remote build for the same eurobuild5 board (both requested in the same session) – see that package’s memory note for how the collision was resolved.