02:35 <Grommish> stintel: Ping

02:50 <kopijahe> Hi, I have xiaomi R3P with busted NAND, so I installed an ordinary SPI on it, I've been compiling for it by myself ( https://forum.openwrt.org/t/openwrt-22-03-23-05-for-xiaomi-mi-router-3-pro-r3p-with-spi/136379 ). If I make a PR for it, is there any chance that it will be accepted?

02:51 <kopijahe> Maybe just split the dts for nand and spi variant, because the rest hw is still the same

04:46 <owrt-images-builds> Build [#38](https://buildbot.openwrt.org/images/#/builders/5/builds/38) of `openwrt-23.05_bmips/bcm63268` failed.

05:29 <Mangix> kopijahe: probably not

05:30 <kopijahe> ok

07:28 <rmilecki> nbd: after 2,5 years I gave my MT7603E another try

07:29 <rmilecki> it was because I noticed your work on mt7603 PSE hang from 2023

07:29 <rmilecki> sadly it didn't help my case :/ MT7603E still suffers from hiccups: https://pastebin.com/hwLcmcKR

07:31 <rmilecki> nbd: is there anything else I could try, anything you may have in mind?

07:43 <rmilecki> nbd: FWIW all counters in /sys/kernel/debug/ieee80211/phy0/mt76/reset are zero

09:13 <f00b4r0> nbd: crash occurred much sooner this time, with only 4 STAs on 2 aps on the same phy. Oddly, the cpu remained hogged for 1h30 (visible on collectd data). HTH

09:14 <f00b4r0> afaict in dmesg this corresponds to a series of Message timeout, until a backtrace is dumped and the timeouts stop

09:50 <robimarko> BTW, any opinion on: https://github.com/openwrt/openwrt/pull/14713

09:50 <robimarko> To me it looks like a good idea, and avoids perl

10:20 <rmilecki> jow: do you think I can push [PATCH] base-files: sysupgrade: handle errors when generating backup

10:26 <jow> rmilecki: sure, looks simple enough

10:26 <jow> robimarko: same for the PR

10:27 <jow> robimarko: line 137 looks like a typo though

10:28 <jow> "is needed recommended"

10:28 <jow> also "is needed to outsmart git" sound a little bit too colloquial for my taste

10:30 <jow> maybe should read "This is an automatically generated commit which aids following Kernel patch history.\n\nSee: https://github.com/openwrt/openwrt/pull/14713" or similar

10:31 <jow> then add a suitable link either pointing to a readme, the mailing list archive, the PR or something

10:31 <jow> 15 years down the road, nobody probably has any idea what "outsmart git" was supposed to mean exactly when browsing the history

11:42 <stintel> Grommish: pong

13:06 <stintel> would maintainer-tools.git not be a better place for this kernel bump script?

13:06 <robimarko> Indeed it would

14:22 <PeterFromSwe> Pull Request testbuilds seem to all fail. I have no clue what's wrong with mine https://github.com/openwrt/packages/pull/23585

14:45 <schmars[m]> PeterFromSwe: if they are built with the SDK there's a little bug on snapshot that was fixed last night and should be fixed in prebuilt SDK some time today

16:04 <schmars[m]> PeterFromSwe yep it's the SDK thing, you can follow https://downloads.openwrt.org/snapshots/targets/ to see the targets being rebuilt. everything from today 12:00 AM UTC should be fine

18:45 <hurricos> mrkiko: I'm braindead, sorry. I meant 6E. HE160 is exclusive to 6GHz I think, but not sure

18:54 <hurricos> blocktrron: If you were still here, I'd ask about u-boot-msm.

19:00 <robimarko> You can do 160 on 5GHz as well, 320 though is 6GHz exclusive

23:36 <Grommish> stintel: You're still playing with rust, yes? If you get bored, check out the rustup package I'm working on? Doesn't require it be pre-installed on host and takes 5 minutes vs 3 hours to build (and doesn't 'rebuild' each target change). Any interest?

23:36 <Grommish> stintel: And, if you got that package you were talking about finishing up somewhere, I can use it for testing