08:43 <csharper2005> rmilecki: ping

11:27 <owrt-2203-builds> Build [#27](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/39/builds/27) of `layerscape/armv7` failed.

11:36 <\x> nbd: nice mt7921 now has ap mode support ;)

11:37 <nbd> don't expect too much of it, it's limited to one vif and will likely only support few stations

11:38 <\x> does it work with the amd branded ones rz608 and rz616?

11:38 <\x> oh those are mt7921k

11:38 <\x> not supported?

11:53 <nbd> i'd expect them to be supported, but i don't know for sure

11:53 <nbd> i've only used mtk reference cards so far

11:53 <\x> that card also runs mt7921e driver i think but yeah ill wait a little while I guess

12:57 <aiyion_> bkoblitz, raylynnknight are not in these channels? I got in possesion of a DGS-1210-10P F1, and wanted to know whether I had to take special caution during flashing or which commit on master? worked for them.

13:06 <grift> these /etc/rc.button scripts installed by base-files have bad-permissions (+x), the are interpretted and cannot be executed

13:08 <owrt-2203-builds> Build [#26](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/60/builds/26) of `layerscape/armv8_64b` failed.

13:17 <grift> so i guess that stuff just belongs in /usr/lib and without -x

13:33 <jow> this stuff is executed by procd

13:33 <jow> using exec()

13:35 <jow> https://git.openwrt.org/?p=openwrt/openwrt.git;a=blob;f=package/system/procd/files/hotplug-preinit.json;h=658c685474fe0e31dd2667662d07ba037661c147;hb=089eb02abcd7512c6d182953560eb2453ef144ca#l16

13:44 <grift> so the variables those scripts depend on are in procd's environment?

13:44 <grift> why is it in /etc though?

13:44 <dwfreed> because they are config

13:44 <dwfreed> in a sense

13:45 <dwfreed> it's the same as rc.local

13:45 <grift> thanks i will then treat them as such

13:45 <dwfreed> and not having to tell sysupgrade about yet another path that needs backup and restore across upgrades is nice

13:48 <grift> not saying its nice or not, but i saw them depending on variable and they werent sourcing any functions so i was like if theyre executed directly then where do those variables come from

13:48 <grift> but if theyre in procd's environment then i guess that answers it

13:50 <Borromini> aiyion_: seems the F1 revision was merged early on and has no separate commit

13:50 <Borromini> maybe https://svanheule.net/switches/dgs-1210-10p can be of use to you

13:50 <Borromini> PaulFertser has one as well but I believe it's an R1 one not f1.

14:36 <PaulFertser> Borromini: thanks. aiyion_ yes, I have access to R1 revision but most probably F1 is similar other than RAM size.

17:13 <aiyion> PaulFertser, Borromini: I think the device was initially added in c78e123d5a5e7c3baa74f326613a14f7e99a29a3 - in october 19th last year.

17:15 <aiyion> So it should be in v21.02 right?

17:16 <aiyion> https://downloads.openwrt.org/releases/21.02.3/targets/realtek/generic/

17:16 <aiyion> nice.

17:27 <swalker> updated openwrt/upstream, https://sdwalker.github.io/uscan/index.html

17:34 <PaulFertser> aiyion: my patches added support for buttons and LEDs.

17:50 <csrf> what's the process for doing a custom build after upgrading the flash to a bigger size? i.e., what do I need to change in the build config in order to handle the increase in flash space?

17:51 <jow> grift: there's various non-standandalone exectuables in /etc/ and elsewhere that should rather go into /usr/libexec/

17:51 <PaulFertser> csrf: usually it requires a change in the DTS file.

17:51 <jow> the argument that /etc/rc.whatever is configuration is true, but on the other hand its unrestricted shell script code, so not really a declarative configuration either

17:57 <grift> jow: yes the implementation isnt the prettiest.

17:57 <csrf> PaulFertser, where can I learn more about that file?

18:05 <PaulFertser> csrf: https://elinux.org/Device_Tree_Linux , usually they're all in target/linux/.../dts/ . You might also want to edit image/Makefile (or one of the files it includes) to allow creating bigger images.

18:06 <csrf> PaulFertser, ok will go investigate, thanks

20:08 <aiyion> PaulFertser: but are part of 21.02 nevertheless, are thy not?

20:59 <Grommish> When making a Host package (PKG_HOST_ONLY), do I leave out the $(INCLUDE_DIR)/package.mk and $(eval $(call BuildPackage,xxxx)) along with the Package/xxxx calls? I'm trying to get it so that the calls don't need to have xxx/host (or am I misunderstanding the PKG_HOST_ONLY call?)

21:39 <mrkiko> rmilecki: ping