I just did a pkgbase update on aarch64 and looked at the kernel vs. world timing relationship. The context is:
Tue, 20 Feb 2024 git: b241767f8ef3 - main - Eliminate unnecessary UFS1 integrity checks. Kirk McKusick git: 3aefe6759669 - main - nextboot: fix typo that merged two args into one Gleb Smirnoff git: 0c3ade2cf13d - main - nextboot: fix nextboot -k on ZFS Gleb Smirnoff git: 81dc3a4d3bb9 - main - tools/net80211/wlanwatch: Remove obsolete code. Joseph Koshy git: 445d3d227e68 - main - msdosfs: fix potential inode collision on FAT12 and FAT16 Stefan Eßer git: 9dd5023e955d - main - iwm.4: add iwlwifi cross-reference Ed Maste git: 6a3e87e1306d - main - arp: fix arp -s/-S Gleb Smirnoff git: 1968d71647ba - main - tests/arp: test arp -s/-S work Gleb Smirnoff git: e7308a60a5b7 - main - md5: Add test case for GNU input modes. Dag-Erling Smørgrav git: 4db7ca244705 - main - md5: Fix Perl mode long options. Dag-Erling Smørgrav git: c05533d97ac1 - main - md5: Fix GNU check mode. Dag-Erling Smørgrav git: 702f133fa181 - main - md5: Untabify declarations. Dag-Erling Smørgrav git: fa6a02f50e65 - main - tests/fdgrowtable: open more files in the threaded case Gleb Smirnoff git: 3ac001620e28 - main - tests/fdgrowtable: remove always true check Gleb Smirnoff git: e53b83a849e3 - main - tests/sigsys: initialize parameter passed to sysctlbyname() Gleb Smirnoff The install reports itself as: # uname -apKU FreeBSD aarch64-main-pkgs 15.0-CURRENT FreeBSD 15.0-CURRENT main-n268410-445d3d227e68 GENERIC-NODEBUG arm64 aarch64 1500014 1500014 That is: git: 445d3d227e68 - main - msdosfs: fix potential inode collision on FAT12 and FAT16 Stefan Eßer CommitDate: 2024-02-20 12:02:24 +0000 However, Looking at /usr/src/tests/sys/kern/sigsys.c shows the content from: git: e53b83a849e3 - main - tests/sigsys: initialize parameter passed to sysctlbyname() Gleb Smirnoff (CommitDate: 2024-02-20 22:37:45 +0000) So, source somewhat over 10 hrs later than the commit date/time for what the kernel reported. (But it is just an example. I've little clue what shortest to longest might be across examples.) (Note: The next commit's source was not present.) Looking to futures based on such status for this early pkgbase context . . . Identifying the detailed combination of sources one is dealing with (commits) is a mess as things are so far. Reporting defects based on testing pkgbase materials is, then, also messy to identify the context for --or to reproduce the context for. Similarly, figuring out if a pkgbase update has a source update that one wants to test can also be messy. Even just being able to identify the kernel's main-n268410-445d3d227e68 like text for the published materials (without download or installation of the actual packages) would be helpful for such. Also having such for the world build involved would be good. Food for thought. === Mark Millard marklmi at yahoo.com
