#archlinux-ports | Logs for 2018-03-20
Back
[00:01:16] -!- sunshavi has quit [Remote host closed the connection]
[00:04:12] -!- sunshavi has joined #archlinux-ports
[00:42:27] -!- deep42thought has quit [Quit: Leaving.]
[00:53:25] guys is now known as evasion
[00:53:33] evasion is now known as guys
[01:07:47] -!- guys has quit [Ping timeout: 276 seconds]
[01:09:58] -!- MrBIOS has joined #archlinux-ports
[01:28:40] -!- guys has joined #archlinux-ports
[01:40:09] -!- MrBIOS has quit [Quit: MrBIOS]
[01:42:54] -!- gnugnugnu has quit [Quit: WeeChat 2.0.1]
[01:50:21] -!- gnugnugnu has joined #archlinux-ports
[05:13:29] -!- hringriin_ has joined #archlinux-ports
[05:16:29] -!- hringriin has quit [Ping timeout: 256 seconds]
[05:16:29] hringriin_ is now known as hringriin
[07:22:10] -!- oaken-source has joined #archlinux-ports
[07:57:21] -!- oaken-source has quit [Ping timeout: 264 seconds]
[08:28:07] -!- insecta_ has quit [Quit: WeeChat 2.0.1]
[08:32:34] -!- insecta has joined #archlinux-ports
[08:41:00] -!- oaken-source has joined #archlinux-ports
[12:40:11] -!- guys has quit [Ping timeout: 256 seconds]
[13:53:21] -!- guys has joined #archlinux-ports
[14:02:55] <girls> brtln: how strict is the "toolchain build order" comment in gcc's PKGBUILD?
[14:03:22] <girls> e.g. _when_ do you (and we) need to rebuild the toolchain in the said manner?
[14:04:04] <brtln> #notmycomment
[14:04:26] <brtln> I rebuild everything for new linux-api-headers
[14:04:34] <girls> but you're the maintainer, who needs to follow it - right?
[14:04:37] <girls> ok
[14:05:48] <brtln> the rest usually isn't needed, but it's a good sanity test for all components with new major release to build everything anyway
[14:06:15] <brtln> I don't bother for bugfix releases (or gcc bumps I push every now and then)
[14:07:00] <girls> but you recompile the chain for each linux-api-header release - also for minors?
[14:07:18] <brtln> no, I wouldn't do it for minors
[14:07:25] <girls> ok, thanks
[14:07:44] <brtln> from what I understand, theoretically you don't need a rebuild at all, I just think it's a good exercise
[14:07:56] <girls> so "for major bump of x: rebuild everything after x in the chain"
[14:08:27] <girls> I just don't want to miss anything upstream rebuilds (but probably does not release)
[14:08:32] <brtln> more like: wait for major bumps to accumulate and rebuild everything
[14:08:59] <brtln> yes, usually the final build hits the repo
[14:09:13] <girls> and that's all, we see
[14:09:17] <brtln> so that gcc in the middle of the chain never leaves my laptop and has ${pkgrel}.1 or pkgrel=0
[14:09:32] <brtln> can't wait for reproducible team to get pissed of at me for this…
[14:09:40] <girls> :-D
[14:10:11] <brtln> if you figure out something better, I'm happy to hear it though
[14:10:34] <girls> I'm afraid I don't understand enough of the internals there
[14:11:04] <girls> for archlinux32, I'd just recompile the chain from left to right for major releases and then publish them with a different sub_pkgrel
[14:11:50] <brtln> I have only vague idea what I'm really doing and nothing broke yet
[14:12:05] <girls> sounds familiar
[14:13:57] <brtln> I asked Allan once about relationship between --enable-kernel in glibc and linux-api-headers
[14:13:59] <brtln> "The linux-api-header version and running kernel do not need to match. The rebiuld for linux-api-headers is just to make sure there is no incompatibilites introduced (it is rare bbut happens). I only bumped a major version for that when other rebuilds were happening."
[14:14:13] <brtln> if that's of any help
[14:15:56] <girls> actually: no - but thanks anyway, you already helped me with what you said before :-)
[14:29:05] <guys> brtln: I guess you could make the repro people happy by publishing -stage1 packages, I've seen it for bootstrapping cross-compiler toolchains in the AUR
[14:29:14] <guys> probably not worth it...
[14:29:44] <brtln> yeah I'm fine with making them sad for now
[14:29:55] <girls> !grab brtln
[14:29:55] <phrik> girls: Tada!
[14:31:03] <brtln> in the future I will push everything to staging and watch the world burn
[14:34:57] -!- oaken-source has quit [Ping timeout: 240 seconds]
[16:14:53] -!- MrBIOS has joined #archlinux-ports
[16:18:32] -!- oaken-source has joined #archlinux-ports
[18:35:29] -!- abaumann has joined #archlinux-ports
[19:43:31] -!- abaumann has quit [Quit: leaving]
[19:54:56] -!- Honkeymagoo has joined #archlinux-ports
[22:05:26] -!- phrik has quit [Remote host closed the connection]
[22:10:48] -!- phrik has joined #archlinux-ports
[22:11:27] -!- alyptik has quit [Ping timeout: 240 seconds]
[22:25:14] -!- typikal has joined #archlinux-ports
[22:26:04] -!- typikal has quit [Excess Flood]
[22:26:53] -!- typikal has joined #archlinux-ports
[22:31:33] -!- oaken-source has quit [Ping timeout: 264 seconds]
[22:32:24] -!- MrBIOS has quit [Quit: MrBIOS]
[22:39:00] -!- typikal has quit [Ping timeout: 246 seconds]
[23:04:28] -!- typikal has joined #archlinux-ports
[23:21:21] -!- Alina-malina has quit [Ping timeout: 265 seconds]
[23:45:15] -!- Honkeymagoo has quit [Quit: Bye]