#archlinux32 | Logs for 2024-02-06
Back
[00:08:09] -!- abouvier has quit [Read error: Connection reset by peer]
[00:12:27] -!- abouvier has joined #archlinux32
[02:05:55] -!- morriset has joined #archlinux32
[02:50:49] -!- morriset has quit [Quit: Leaving]
[03:33:25] -!- epony has quit [Remote host closed the connection]
[03:34:20] -!- epony has joined #archlinux32
[04:27:49] -!- epony has quit [Remote host closed the connection]
[06:19:23] -!- abaumann has joined #archlinux32
[06:19:23] <buildmaster> Hi abaumann!
[06:19:24] <buildmaster> !rq abaumann
[06:19:24] <phrik> buildmaster: <abaumann> meson adds another layer of f*ups
[06:20:12] <abaumann> KitsuWhooa: glibc, if you remove the --disable-cet, do you end up with endbr32 in your code?
[09:16:46] -!- epony has joined #archlinux32
[09:37:23] <KitsuWhooa> abaumann: it just doesn't build
[09:37:31] <KitsuWhooa> Configure fails
[09:37:52] <KitsuWhooa> That's what the commit message says
[09:39:54] -!- mvchtz has joined #archlinux32
[09:43:26] -!- abaumann has quit [Quit: Client closed]
[10:08:49] -!- abaumann has joined #archlinux32
[10:08:49] <buildmaster> Hi abaumann!
[10:08:49] <buildmaster> !rq abaumann
[10:08:50] <phrik> buildmaster: <abaumann> stackoverflow is more or less a collection of esotheric modern woodoo instead of a source for technical information :-)
[10:09:34] <abaumann> That's why you can crawl it and feed it to a LLM and have a chatbot continuing maintainance of Linux. ;-)
[10:10:09] <KitsuWhooa> hah
[10:24:03] <KitsuWhooa> I was referring to my commit message fixing it fwiw
[10:24:32] <KitsuWhooa> https://git.archlinux32.org
[10:24:34] <phrik> Title: packages - Archlinux32 package modifications (at git.archlinux32.org)
[10:51:38] <KitsuWhooa> > warning: cannot resolve "dbus-units", a dependency of "systemd"
[10:51:41] <KitsuWhooa> what fresh hell is this
[11:10:02] <KitsuWhooa> gpg: error running '/usr/bin/gpg-agent': exit status 2
[11:10:02] <KitsuWhooa> gpg: failed to start gpg-agent '/usr/bin/gpg-agent': General error
[11:10:07] <KitsuWhooa> eurobuild is spamming that
[11:10:10] <KitsuWhooa> doesn't seem to be fatal
[11:10:16] <KitsuWhooa> > ==> Generating pacman master key. This may take some time.
[11:10:22] <KitsuWhooa> is it supposed to regenerate it every time?
[11:10:55] <KitsuWhooa> but also oh god, with new libc everything needs a rebuild again, ugh
[11:11:10] <KitsuWhooa> we're never getting to a stable state again are we
[11:12:36] <abaumann> aeh, the dbus implementation changed for systemd
[11:13:04] <abaumann> now. hopefully glibc maintains ABI compatibility.
[11:13:25] <abaumann> So, a rebuild is not really necessary (of glibc and the toolchain of course, but nothing else).
[11:13:35] <abaumann> otherwise we are in deep trouble anyway. :-)
[11:14:16] <KitsuWhooa> I was looking at pipewire failing to build
[11:14:25] <KitsuWhooa> 12:54:32 <buildmaster> i686/pipewire is broken (says eurobuild6-3): https://archlinux32.org
[11:14:42] <KitsuWhooa> missing dependencies: systemd
[11:15:25] <KitsuWhooa> unless it's because of those gpg errors
[11:15:45] <KitsuWhooa> > slave3@eurobuild6
[11:20:06] <KitsuWhooa> I've been working on a script that tails build logs for a builder
[11:20:07] <KitsuWhooa> I need to finish it
[11:20:15] <abaumann> usually the order: toolchain, core, then others is the best
[11:20:27] <abaumann> sadly that systemd will most likely fail also on python or so.
[11:20:36] <abaumann> oh.
[11:20:42] <abaumann> a logger, nice.
[11:20:54] -!- abaumann has quit [Quit: Client closed]
[11:21:07] -!- abaumann has joined #archlinux32
[11:23:43] <KitsuWhooa> it'd help if I could find it
[11:26:45] <abaumann> The log
[11:26:52] <abaumann> should be in builder/work locally
[11:27:08] <abaumann> and somewhere in the buildmaster main page there is a link to all the logfiles
[11:27:10] <KitsuWhooa> I meant the script
[11:27:12] <KitsuWhooa> but yes
[11:27:13] <abaumann> a
[11:27:18] <KitsuWhooa> the problem is that new files come and go all the time
[11:27:21] <abaumann> yeah, that one I don't know. :-)
[11:27:34] <abaumann> that's the rolling aspect of it ;-)
[12:12:48] -!- zxrom has joined #archlinux32
[13:12:10] -!- abaumann has quit [Quit: Client closed]
[13:24:39] <KitsuWhooa> archlinux32.org seems dead
[14:00:16] <zxrom> KitsuWhooa, Works for me.
[14:00:33] <KitsuWhooa> yeah it's back up now
[14:28:59] -!- texus has joined #archlinux32
[14:38:52] -!- ssserpent has joined #archlinux32
[14:44:46] -!- ssserpent has quit [Quit: WeeChat 4.2.1]
[14:46:00] -!- ssserpent has joined #archlinux32
[14:47:25] -!- ssserpent has quit [Client Quit]
[14:52:43] -!- ssserpent has joined #archlinux32
[15:04:47] -!- ssserpent has quit [Quit: WeeChat 4.2.1]
[15:05:08] -!- ssserpent has joined #archlinux32
[15:25:26] -!- ssserpent has quit [Quit: WeeChat 4.2.1]
[15:30:12] -!- ssserpent has joined #archlinux32
[15:35:49] -!- ssserpent has quit [Quit: WeeChat 4.2.1]
[15:36:31] -!- ssserpent has joined #archlinux32
[15:42:51] -!- ssserpent has quit [Quit: WeeChat 4.2.1]
[15:43:09] -!- ssserpent has joined #archlinux32
[15:57:28] -!- titus_livius has joined #archlinux32
[17:00:30] -!- epony has quit [Remote host closed the connection]
[17:09:08] -!- epony has joined #archlinux32
[17:30:11] -!- ssserpent has quit [Quit: WeeChat 4.2.1]
[17:31:17] -!- ssserpent has joined #archlinux32
[17:37:06] -!- abaumann has joined #archlinux32
[17:37:06] <buildmaster> Hi abaumann!
[17:37:06] <buildmaster> !rq abaumann
[17:37:07] <phrik> buildmaster: <abaumann> I think, adding a error to namcap whenever git/hg repos are used, is an idea. :->
[17:37:32] <KitsuWhooa> isn't glibc built from git?
[17:37:35] <KitsuWhooa> :p
[17:38:06] <abaumann> yes, why?
[17:38:22] <abaumann> did you just move the --disable-cet patch to the end for all CARCHs?
[17:38:30] <abaumann> so I misunderstood it. :-)
[17:39:16] -!- texus has quit [Quit: WeeChat 4.2.1]
[17:39:29] <KitsuWhooa> 19:37:07 <phrik> buildmaster: <abaumann> I think, adding a error to namcap whenever git/hg repos are used, is an idea. :->
[17:39:35] <KitsuWhooa> it was in response to that
[17:39:46] <KitsuWhooa> 19:38:23 <abaumann> did you just move the --disable-cet patch to the end for all CARCHs? <-- yes, because it was complaining about CET on i686 and pentium4 and not building
[17:40:15] <abaumann> so, this means CET mitigation doesn't exist anymore at all for 32-bit
[17:40:24] <abaumann> it was never available on i486
[17:40:29] <KitsuWhooa> guess so
[17:40:33] <abaumann> ok.
[17:40:41] <KitsuWhooa> I just wanted a working glibc to get things going so I didn't put much thought into it
[17:41:11] <abaumann> I'm just puzzled because I don't see a CARCH in the diff.
[17:41:24] <KitsuWhooa> it was in the i486 one
[17:41:32] <KitsuWhooa> I think it's too high and cut off by the diff
[17:41:36] <KitsuWhooa> https://buildmaster.archlinux32.org
[17:41:38] <KitsuWhooa> this is hte log fwiw
[17:41:39] <abaumann> ah
[17:41:40] <KitsuWhooa> *the
[17:42:17] <abaumann> this will most likely cause problems with gcc and binutils
[17:42:20] <KitsuWhooa> and this is one commit earlier https://git.archlinux32.org
[17:42:22] <phrik> Title: PKGBUILD « glibc « core - packages - Archlinux32 package modifications (at git.archlinux32.org)
[17:42:39] <abaumann> # i486-specific
[17:42:39] <KitsuWhooa> will it?
[17:42:39] <abaumann> if [ "${CARCH}" = "i486" ]; then
[17:42:39] <abaumann> # disable CET (Control Flow instructions endbr32/enbr64)
[17:42:39] <abaumann> eval "$(
[17:42:39] <abaumann> declare -f build | \
[17:42:41] <abaumann> sed '
[17:42:44] <abaumann> s/--enable-cet/--enable-cet=no/
[17:42:46] <abaumann> '
[17:42:49] <abaumann> )"
[17:42:51] <abaumann> in binutils
[17:42:53] <KitsuWhooa> yup
[17:42:54] <abaumann> dito gcc
[17:43:07] <abaumann> you might end up in a non working too chain in staging
[17:43:13] <abaumann> *tool chain
[17:43:14] <KitsuWhooa> oh dear
[17:43:29] <KitsuWhooa> well, nothing has blown up *yet*
[17:43:30] <abaumann> also fcf-protection
[17:43:37] <KitsuWhooa> so uh
[17:43:40] <KitsuWhooa> please fix it before it blows up? :p
[17:43:41] <KitsuWhooa> I tried
[17:44:07] <KitsuWhooa> there's already new gcc
[17:44:16] <abaumann> mmh
[17:46:35] <abaumann> https://lists.gnu.org
[17:46:36] <phrik> Title: The GNU C Library version 2.39 is now available (at lists.gnu.org)
[17:46:38] <abaumann> * Sync with Linux kernel 6.6 shadow stack interface. The --enable-cet configure option is only supported on x86-64.
[17:46:42] <abaumann> glibc 2.39
[17:46:49] <KitsuWhooa> RIP
[17:47:09] <gehidore> but is it 6.6.6!? best kernel ever
[17:47:15] <abaumann> well, what we loose is mitigation of branching errors on pentium4 and i686
[17:47:17] <KitsuWhooa> hah
[17:47:34] <KitsuWhooa> abaumann: it is what it is...
[17:47:56] <abaumann> yes, we cannot add CET for IA32 into the toolchain
[17:48:00] <abaumann> makes no sense
[17:56:42] <abaumann> aeh.. can you log out of the buildmaster - just fast. some kernel updates I would like to play and reboot..
[17:58:39] <KitsuWhooa> done
[17:58:45] <abaumann> thanks
[17:59:13] -!- buildmaster has quit [Read error: Connection reset by peer]
[18:02:09] -!- buildmaster has joined #archlinux32
[18:03:37] -!- ssserpent has quit [Quit: WeeChat 4.2.1]
[18:04:38] -!- ssserpent has joined #archlinux32
[18:05:20] -!- ssserpent has quit [Client Quit]
[18:05:46] -!- ssserpent has joined #archlinux32
[18:05:59] <abaumann> up again. sanity-check -w run manually, all good.
[18:08:14] <KitsuWhooa> nice
[18:09:11] -!- ssserpent has quit [Client Quit]
[18:09:37] -!- ssserpent has joined #archlinux32
[18:09:44] <abaumann> Running /build/binutils/src/binutils-gdb/binutils/testsuite/binutils-all/ar.exp ...
[18:09:47] <abaumann> FAIL: replacing non-deterministic member: (failed to replace file)
[18:09:49] <abaumann> FAIL: replacing SOURCE_DATE_EPOCH deterministic member: (failed to replace file)
[18:09:52] <abaumann> mmh.
[18:18:57] -!- ssserpent has quit [Quit: WeeChat 4.2.1]
[18:45:59] -!- epony has quit [Remote host closed the connection]
[19:17:43] -!- abaumann has quit [Quit: leaving]
[23:08:44] -!- GNUtoo has quit [Ping timeout: 255 seconds]
[23:14:22] -!- GNUtoo has joined #archlinux32