<dwfreed> stintel: since /var is default a symlink to /tmp and /tmp is default a tmpfs, it would probably be best to handle making the dir in initscript; if things need to persist across reboots, then yes they should be in /etc
Daanct12 has joined #openwrt-devel
parazyd has quit [Ping timeout: 480 seconds]
parazyd has joined #openwrt-devel
tSYS has quit [Quit: *squeak*]
tSYS has joined #openwrt-devel
rua has quit [Quit: Leaving.]
<owrt-images-builds> Build [#33](https://buildbot.openwrt.org/images/#/builders/152/builds/33) of `openwrt-23.05_ramips/rt288x` completed successfully.
minimal has quit [Quit: Leaving]
goliath has joined #openwrt-devel
rua has joined #openwrt-devel
hanetzer1 has joined #openwrt-devel
hanetzer has quit [Ping timeout: 480 seconds]
rua has quit [Quit: Leaving.]
rua has joined #openwrt-devel
rua has quit [Quit: Leaving.]
robimarko has joined #openwrt-devel
cation has joined #openwrt-devel
<stintel> dwfreed: there is no init script, but I suggested to do mkdir in the proto handler. and no. /etc is not the right place for a file that is changed by the daemon
rua has joined #openwrt-devel
<dwfreed> stintel: I mean, if the file needs to persist across reboots without uci config to recreate it, then /var is the wrong place to put it
torv has quit [Remote host closed the connection]
torv has joined #openwrt-devel
<stintel> I'm well aware of that, but I put it on /var/lib/thread for multiple reasons: it's the default, it won't end up on flash by default so no chance of killing flash and like this the user has to decide themselves if they want to do that, you can just use CONFIG_TARGET_ROOTFS_PERSIST_VAR=y
<dwfreed> does it matter if the file doesn't persist
<stintel> what do you mean?
<dwfreed> meaning, in the default configuration, if the device reboots, is the whole setup still usable without manual intervention
<dwfreed> with non-persistent /var
<stintel> no
<dwfreed> then it should not be in /var, full stop
<dwfreed> because changing that config var requires a full source rebuild, basically (or you try to rebuild just base-files and replace it, but that gets hairy)
<stintel> I'm not putting it in /etc, that is just wrong
<dwfreed> that is the best place for it under the circumstances
<stintel> a user can decide to do that, I will not be making that default in a package I maintain
<dwfreed> then why ship a package that is broken by default?
Mangix has quit [Read error: Connection reset by peer]
<froztbyte> hmm, before I start going down this rabbithole myself I figure it's best to maybe ask: if I wanted to cache parts of the build process in my CI setup, which bits are safe/which bits should I look at for cache invalidation? I've glanced at a couple of the buildfarm jobs to see how they run, but it's a bit of a pain to drill down into them and afaict there's no outwardly visible file that defines
<froztbyte> the buildjobs and their setups there
<froztbyte> right now I'm doing some jobs both targeting variations on 23.05.2 release, as well as some others tracking off snapshot at various commit refs. the 23.05.2 builds I figure I could cache the whole toolchain? but the others I figure there might be moving pieces
<dwfreed> buildbot configuration is defined there
<froztbyte> ah, ty
<dwfreed> also accessible here if you like github better: https://github.com/openwrt/buildbot
<froztbyte> inverse, actually, github annots the crap out of me
<froztbyte> already checked out the repo, glancing at it now
<froztbyte> dwfreed: hmm this is only the defintion for the buildbot itself? I'm looking for the build job definitions
<dwfreed> the build jobs are in there
<dwfreed> buildbot is all python
<froztbyte> ah, they're in the master.cfg files
<froztbyte> it's been forever since I had to touch buildbot, might've evicted all the knowledge from my brain since :>
<froztbyte> ty, will read over this
guidosarducci has joined #openwrt-devel
torv has quit [Quit: torv]
torv has joined #openwrt-devel
theJoker8814 has joined #openwrt-devel
rua has quit [Quit: Leaving.]
rua has joined #openwrt-devel
castiel652 has joined #openwrt-devel
<rmilecki> jow: i just wanted to ask if you saw my question on sysupgrade + tar
vincejv has quit [Quit: Bye bye! Leaving for now...]
castiel652 has quit [Quit: Leaving]
Daanct12 has quit [Quit: WeeChat 4.2.1]
castiel652 has joined #openwrt-devel
vincejv has joined #openwrt-devel
castiel652 has quit []
gch981213 has joined #openwrt-devel
fakuivan has quit [Ping timeout: 480 seconds]
Misanthropos has quit [Ping timeout: 480 seconds]
SlimeyX has quit [Ping timeout: 480 seconds]
<owrt-images-builds> Build [#86](https://buildbot.openwrt.org/images/#/builders/204/builds/86) of `master_sunxi/cortexa53` failed.
rua is now known as Guest1187
rua has joined #openwrt-devel
theJoker8814 has quit [Quit: theJoker8814]
Guest1187 has quit [Ping timeout: 480 seconds]
Misanthropos has joined #openwrt-devel
<owrt-images-builds> Build [#88](https://buildbot.openwrt.org/images/#/builders/217/builds/88) of `master_bcm47xx/generic` completed successfully.
valku has joined #openwrt-devel
goliath has quit [Quit: SIGSEGV]
minimal has joined #openwrt-devel
djfe has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
djfe has joined #openwrt-devel
<rmilecki> nbd: what do you think about / do you plan mt76 update for 23.05 branch?
<rmilecki> right now I'm looking at MT7996 support
<rmilecki> but i'm asking in general too
<owrt-images-builds> Build [#89](https://buildbot.openwrt.org/images/#/builders/85/builds/89) of `master_bcm53xx/generic` completed successfully.
<owrt-images-builds> Build [#87](https://buildbot.openwrt.org/images/#/builders/153/builds/87) of `master_bmips/bcm6328` completed successfully.
<owrt-images-builds> Build [#86](https://buildbot.openwrt.org/images/#/builders/141/builds/86) of `master_kirkwood/generic` completed successfully.
<owrt-images-builds> Build [#85](https://buildbot.openwrt.org/images/#/builders/149/builds/85) of `master_mediatek/filogic` completed successfully.
zkrx is now known as Guest1197
zkrx has joined #openwrt-devel
Guest1197 has quit [Ping timeout: 480 seconds]
zkrx has quit []
goliath has joined #openwrt-devel
<nbd> rmilecki: i am planning on updating mt76 in 23.05, however i'm not planning on adding mt7996 support to that branch
<nbd> i think it would require a mac80211 update
goliath has quit [Quit: SIGSEGV]
zkrx has joined #openwrt-devel
bbezak has quit [Ping timeout: 480 seconds]
bbezak has joined #openwrt-devel
bbezak has quit [Quit: The Lounge - https://thelounge.chat]
bbezak has joined #openwrt-devel
bbezak has quit [Quit: The Lounge - https://thelounge.chat]
bbezak has joined #openwrt-devel
bbezak has quit []
bbezak has joined #openwrt-devel
bbezak has quit []
bbezak has joined #openwrt-devel
<rmilecki> nbd: ok, thanks for info
gladiac is now known as Guest1222
gladiac has joined #openwrt-devel
valku has quit [Quit: valku]
Guest1222 has quit [Ping timeout: 480 seconds]
<jow> rmilecki: I saw your message but had no time at all today to look into it, will hopefully check it tomorrow
robimarko has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
GNUmoon has quit [Quit: Leaving]
<rmilecki> jow: ok, thanks!
GNUmoon has joined #openwrt-devel
Borromini has joined #openwrt-devel
<rmilecki> nbd: is that only about 13b24340d7f2 ("mac80211: backport upstream patch required for mt76 update")
<rmilecki> or are more mac80211 changes needed for mt7996?
<stintel> what's going on with GH CI again?
rua has quit [Quit: Leaving.]
<jow> stintel: afair this is known broken in the pacakges feed
<rmilecki> jow: thanks for replying yoday
<rmilecki> jow: any opinion on Pauls' suggestion to use local pad=$(printf '\1%.0s' $(seq 100))
bbezak has quit [Read error: Connection reset by peer]
<stintel> jow: too bad, I tested locally so I just merged the PR
bbezak has joined #openwrt-devel
gch981213 has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
Borromini has quit [Quit: Lost terminal]
<jow> rmilecki: maybe I am dense but that suggestion did not work for me at all
<jow> rmilecki: not sure what it is meant to accomplish
<rmilecki> jow: sorry, I didn't verify it
<rmilecki> I'll take a look at it tomorrow
<rmilecki> thank you for your help
<jow> rmilecki: it also replaces a static string assignment with two subshell invocations
<rmilecki> that was my concern
<jow> so even if it'd work, I'd say it is not worth it, just to eliminate one long line
<jow> rmilecki: also, not sure if we even should add the complexity to support non-root/0 uids/gids
<rmilecki> ok, I can just make them 0 / 0
<jow> rmilecki: as well as support for file members (which are more efficiently handled by simply invoking tar
<rmilecki> yeah, that bash implementation is slow
<rmilecki> i wrote it before I realized how you wanted that tar code used
<jow> so maybe do a v2 with all but the dynamic member support removed, with uid/gid hardcoded to 0 and with the dd stuff inlined
<rmilecki> ack
<rmilecki> thanks
<jow> then it should become pretty small
<jow> and still remains easy to extend later
<rmilecki> ok
<rmilecki> thanks for hep
<rmilecki> i really need to go to sleep, my eyes are closing
<rmilecki> goodnight!
<jow> good night
Borromini has joined #openwrt-devel
Borromini has quit []