08:43 <Igel> trying to get merge package into 23.05 pls https://github.com/openwrt/packages/pull/22972

08:52 <Igel> its been over a month. its fair to say there are negative assumptions being made over this.

12:11 <rmilecki> Igel: your're right, as always, we don't have enough man power for all requests

12:12 <rmilecki> Igel: i reviewed it, I'll merge it soon unless someone comments against

12:26 <owrt-images-builds> Build [#16](https://buildbot.openwrt.org/images/#/builders/2/builds/16) of `openwrt-23.05_mediatek/mt7623` completed successfully.

14:20 <Ansuel> robimarko well the only missing values are the max_volt_scale and max_quot_scale

14:20 <Ansuel> but we can make them optional if we really want...

14:21 <Ansuel> they are just sanity check ofr the interpolation... we don't have such thing in the qsdk driver

14:42 <robimarko> Yeah, I set those to the max OPP as a sanity check

14:42 <robimarko> BTW, thanks for merge

15:03 <f00b4r0> hmm, mt7915 woes. kernel: [515274.518857] mt7915e 0000:02:00.0: Message 000025ed (seq 2) timeout followed by an OOPS

15:09 <f00b4r0> that's in 23.05.2

15:09 <f00b4r0> nbd: I have a dmesg output full of scary error messages collected before the device was rebooted to restore connectivity, are you interested? Should I open a GH issue against m76?

15:10 <nbd> please send me a paste and i'll take a look

15:11 <f00b4r0> sure

15:13 <f00b4r0> https://clbin.com/fI4bk full logread output. Trouble starts toward the last 10% but at least you have a fairly long history of what happened

15:19 <Ansuel> wow something really bad happen o.o

15:19 <f00b4r0> it wasn't the first occurence but it's the only time where the operator managed to get me a log ;P

15:20 <f00b4r0> 9 days uptime I should probably mention

15:20 <f00b4r0> lost connectivity after 7 apparently

15:32 <nbd> f00b4r0: firmware crashed, but the log doesn't have firmware messages enabled (those need to be enabled via debugfs)

15:32 <nbd> so unfortunately the relevant information is not in there

15:33 <nbd> if you can trigger it, it would be good to enable firmware messages

15:33 <nbd> and preferably use a newer build

16:01 <owrt-images-builds> Build [#39](https://buildbot.openwrt.org/images/#/builders/151/builds/39) of `master_layerscape/armv7` failed.

16:07 <rmilecki> Ansuel: thank you for fixing phy_interface_num_ports()

16:30 <f00b4r0> nbd: noted. Sadly I'm not running these systems and 23.05 is as high as they'll go; I'll check for enabling the messages

16:56 <schmars[m]> is the matrix bridge working? just seeing joins/parts/kicks and no other messages since tuesday

17:00 <nbd> f00b4r0: in the phy debugfs directory, write 1 to fw_debug_wm (iirc)

17:01 <f00b4r0> nbd: ok thanks

17:05 <Ansuel> robimarko well the quic guy just decided that he is not ok... ffs -.- i'm very pissed

17:05 <Piraty> i'm rebuilding a few packages from the 18.06 branch , finding a lot of them files are reproducible with the openwrt toolchain, but the package building seems to differ in that openwrt's binary packages seem to have used `tar --numeric-owner` or similar, while the packages i just built have `(0) root` in tar table

17:16 <Piraty> (meaning gnu-tar's --numeric-owner wasn't used)

17:18 <Piraty> i see 7a73221332 ("build: use numeric-owner in ipkg-build") but it is pretty recent

17:19 <Piraty> i vaguely remember LEDE being a thing back then, maybe the package on mirror are those built by lede and the code was not merged ? any other ideas?

17:20 <Piraty> openwrt.org gives me 404 oO

17:26 <robimarko> Ansuel: I dont get the guy

17:41 <Piraty> here's a diff : https://paste.rs/pQLxn

17:49 <zorun> Piraty: thanks for the report, there was a disk issue, it's now fixed

17:49 <Piraty> welcome. it was reported an 1h before in #openwrt though

20:42 <Piraty> what is the dotconfig used to build packages found in https://archive.openwrt.org/releases/packages-* ?

20:43 <Piraty> i feel like it's not defconfig

21:21 <f00b4r0> Piraty: and yet it is.

22:24 <Piraty> but defconfig doesn't select all packages..?

22:30 <Piraty> simply comparing libopenssl reveals that a different config was used

23:02 <Dunya> hi everyone

23:37 <Piraty> according to https://openwrt.org/docs/guide-developer/toolchain/use-buildsystem#using_official_build_config old releases should have config.seed but https://archive.openwrt.org/releases/18.06.9/targets/ar71xx/generic/ doesn't have it

23:45 <Ansuel> i think it wasn't a thing in 18.06

23:46 <Piraty> the ramips target has one, so i wonder

23:47 <Piraty> (just try the example in linked wiki page)