01:51 <KGB-2> https://tests.reproducible-builds.org/openwrt/openwrt_sunxi.html has been updated. (0% images and 100.0% packages reproducible in our current test framework.)

06:23 <\x> Snuupy hah https://ieee802.org/19/arc/stds-802-19list/msg03710.html

06:43 <Snuupy> \x: lol finally

06:43 <Snuupy> China still doesn't have it

06:43 <Snuupy> I wonder why they're dragging their feet

07:29 <\x> so yeah, may 23 + 15 days and i can use 6ghz legally

07:30 <\x> i wonder how will wireless-regdb will handle this rule though since it is a dual rule

07:30 <\x> (5925 - 6425 @ 320??), (23), NO-OUTDOOR

07:30 <\x> (5925 - 6425 @ 320??), (13)

07:41 <rmilecki> nbd: there seems to be a problem with mt7603 compilation

07:41 <rmilecki> caused by https://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless-next.git/commit/?id=b473c0e47f04d3b4ee9d05d2e79234134aad14d5 ("wifi: mt76: mt7603: fix tx queue of loopback packets")

07:41 <rmilecki> mt7603/dma.c: In function 'mt7603_rx_loopback_skb':

07:41 <rmilecki> mt7603/dma.c:60:17: error: ISO C90 forbids mixed declarations and code [-Werror=declaration-after-statement]

07:42 <rmilecki> 60 | u8 qid = tid_to_ac[tid];

07:42 <rmilecki> | ^~

07:46 <nbd> rmilecki: will look into it later

09:25 <\x> Snuupy: china not having 6GHz is kind of yeahh... I guess no cheap wifi hardware for now

12:36 <Snuupy> \x: there's always that cr1000A

13:10 <enyc> Hrrrrrm; Engineering friend definitely confirmed PPPoE fault not working in 23.05.3 but does work in 22.03.6 , and apparently on OpenWRT forums 'Joined the forum, it looks like I'm not the only one to be having trouble with PPPoE with 23 series."

13:10 <enyc> Question for Devs is: ** What to do to debug this or add to or provide suitable fault/bug-report ?

13:19 <enyc> We would, of course, like to see this fixed for 23.05.4 ... PPPoE *is* working oven PTM over VDSL lantiq on AAISP, but PPPoE on ethernet interface to Openreach ONT to Zen internet, does not work ... where to go from there?

13:22 <enyc> I should be ablo to test 23.05.3 with PPPoE to Openreach VDSL separate modem to AAISP, come to think of it

13:22 <T-Bone> hauke: hi, just curious, what's up with 22.03.7?

13:33 <hauke> f00b4r0: it is delayed

13:34 <hauke> probably in 1.5 weeks

13:47 <f00b4r0> ok thanks

13:53 <jow> 11:06 < jow> enyc: well first clue would be to see whether you get a "timeout waiting for PADO" or some different failure mode

13:53 <jow> 11:06 < jow> if it is the former, then the issue might be related to the switch/ethnet

13:53 <jow> 11:06 < jow> e.g. non working vlan config in case the isp requires pppoe via tagged ethernet or some other, lower level frame reception issue

13:53 <jow> 11:07 < jow> I vaguelly recall some bug reports about some broken dsa switches not receiving pppoe frames

15:30 <lu_zero> anybody with a bpi-r3-mini?

15:31 <lu_zero> I keep having the configuration not restored upon sysupgrade

15:34 <lu_zero> I just reinstalled everything and the problem persists :E

15:55 <enyc> jow: in this case, its' not tagged case, and on EA8300 is dedicated wan ethernet port anyhow, no VLAN and (I think) not through switch

16:20 <dwfreed> enyc: nice typo

16:21 <enyc> lol Error: Unknown error (AUTH_TO_PEER_FAILED)

16:26 <DSchndr_> Hi, i am trying to port openwrt to the mt76x8 gardena_smart_gateway-19005 but stumble upon an issue, the vendor uboot throws "Image too large: increase CONFIG_SYS_BOOTM_LEN" when trying to boot the FIT image. I threw out everything out of the kernel but the kernel image and tried to constrain it with "KERNEL_SIZE" but it still keeps the same size :( any ideas?

17:20 <stintel> I think there are some tricks, like lzma-loader? unfortunately I never properly understood how it worked and it's poorly documented

17:20 <stintel> not sure if it's actually lzma-loader but that's what I remember

17:24 <Habbie> plenty of lzma in target/linux/ramips/image/mt7621.mk and some files near it

17:25 <Habbie> ah, and target/linux/mediatek/image/mt7622.mk

17:25 <Habbie> DSchndr_, so maybe work off of those examples and see what happens?

20:22 <DSchndr_> Reee, so close but now it fails to build the target image because somwhere the bs parameter for dd gets lost (dd if=...19005-squashfs-factory.bin of=...19005-squashfs-factory.bin.new bs= conv=sync)

20:24 <DSchndr_> At least uboot didn't complain on the last non working image (but the watchdog kicked in because it just hung), so lzma-loader is the way to go, thanks :)

20:37 <Habbie> DSchndr_, nice

20:58 <Mangix> rmilecki: that's a kernel 5.15 issue I think

20:58 <Mangix> modern kernels use gnu11

21:11 <Mangix> https://www.phoronix.com/news/Sysctl-Finish-Line-Linux-6.10

22:17 <schmars[m]> kmod-mt7915-firmware seems to be missing for 23.05 ramips/mt7621 - https://downloads.openwrt.org/releases/23.05-SNAPSHOT/targets/ramips/mt7621/kmods/5.15.158-1-082b752fd06fa7edca4e1f0a6f15e5de/

22:26 <Mangix> schmars[m]: gnu11 patch probably needs backporting

22:31 <schmars[m]> i don't think the c standard was changed on 23.05 recently? i'll dig up buildbot logs in a moment

22:32 <Mangix> schmars[m]: https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?h=next-20240517&id=e8c07082a810fbb9db303a2b66b66b8d7e588b53

22:33 <Mangix> in place of backporting that (could cause other issues), mt76 could be patched.

22:36 <schmars[m]> did you already look at logs or how did that pop into your mind :)

22:36 <schmars[m]> it was fine a few days ago so my first assumption is the large recent mt76 changeset on openwrt.git 23.05 branch

22:37 <Mangix> right. mt76 is an upstream driver. it works with modern kernels. There's no requirement to be compatible with older ones

22:37 <schmars[m]> oooh i see. https://github.com/openwrt/openwrt/issues/15510

22:37 <schmars[m]> gt it

22:38 <schmars[m]> looks like rmilecki will fix it soon

23:14 <KGB-2> https://tests.reproducible-builds.org/openwrt/openwrt_omap.html has been updated. (11.1% images and 100.0% packages reproducible in our current test framework.)

23:20 <damo22> is there any way by trial and error to safely find the right serial pins without a voltmeter?

23:21 <Habbie> do you know the voltage for your device?

23:21 <damo22> no

23:21 <Habbie> do you have anything else to measure or test with? a single LED?

23:22 <damo22> hmm i'll have a look

23:22 <Habbie> plus a resistor perhaps :D

23:22 <Habbie> or any thing that is low power and that you would normally feed from 2xAA

23:23 <Habbie> i don't know where you live but here in .nl a voltmeter is 5 EUR, btw

23:23 <damo22> i can go buy one, i just lent mine to someone

23:24 <Habbie> right

23:25 <Habbie> i also have a collection of these https://www.aliexpress.com/item/33017235960.html

23:25 <Habbie> but i don't know how they do on 'signal' pins

23:25 <damo22> theyre awesome

23:30 <damo22> damn i ran out of headers

23:30 <Habbie> wires and pegs? :)

23:31 <damo22> i dont have pegs

23:31 <Habbie> wires and five hands then

23:31 <damo22> yeah

23:31 <Habbie> (been there)

23:31 <damo22> what i need is dupont with adapters

23:32 <damo22> male-male

23:33 <damo22> my usb-serial has female dupont on the ends

23:34 <Habbie> right

23:34 <Habbie> buy one big set of male-female and you can make any wire you want

23:35 <damo22> true

23:46 <damo22> heh found some headers