00:11 <Mangix> Ansuel: it already is with host packages

00:48 <Mangix> make -r package/kea/host/compile: build failed. joy

05:52 <damo22> I searched the forum and docs but could not find much about TP-Link Archer MR200 v6, is this ported or likely to be able to be ported?

05:53 <damo22> i could not locate the gpl sources on tplink's website for v6

05:56 <damo22> i didnt do my homework and didnt realise there were slight variations with different version numbers of the hardware, and so i have purchased a v6

07:20 <PaulFertser> damo22: hi

07:21 <PaulFertser> damo22: what's inside that v6?

07:25 <damo22> good question, i havent even unwrapped the plastic yet

07:28 <damo22> i guess i need to take it apart and take hi res photos of the mainboard?

07:36 <damo22> how would i get a shell on it to print the kernel log?

07:38 <KanjiMonster> damo22: usually open the case, locate the serial, connect to it*, let it boot, observe output (* may require soldering)

07:38 <damo22> okay

07:40 <PaulFertser> damo22: and in any case do not really expect someone will do the port for you

07:40 <PaulFertser> damo22: I suggest you instead find the FCCID and check the photos in the FCC filing first. Probably it's a completely different device.

07:42 <jow> at the very least it appears to be a qlacom soc

07:42 <jow> *qualcom

07:42 <jow> so no sudden shift to broadcom or similar

07:42 <damo22> i ported a x86 intel chipset to coreboot, i can probably port a router to openwrt

07:45 <damo22> so FCC has images of routers based on FCCID

07:48 <robimarko> And as long as they are not ultra new they are public

07:53 <PaulFertser> damo22: sure, if the SoC is already supported!

07:53 <damo22> i cant seem to get the plastic cover off... lol

07:53 <damo22> theres no screw holes

07:53 <PaulFertser> Sometimes under stickers or rubber stands

08:28 <damo22> Mediatek MT7613BEN

08:29 <damo22> does that help?

08:31 <PaulFertser> damo22: that's supported PCIe wireless, good.

08:31 <PaulFertser> damo22: what about the SoC?

08:31 <damo22> its hiding under alloy covers

08:32 <PaulFertser> damo22: grab serial log from vendor firmware then, it's the necessary step either way.

08:34 <damo22> where would i find the fcc id?

08:34 <PaulFertser> damo22: on the sticker should be present for sure unless it's for internal chinese market.

08:39 <damo22> no FCC just (EU)

08:39 <damo22> RoHS

08:39 <damo22> and made in china

08:48 <PaulFertser> damo22: I meant on the plastic case, not on the PCB.

08:51 <damo22> yeah thats what i meant

08:52 <damo22> https://www.zamaudio.com/mbox2/mr200v6.jpg

09:03 <PaulFertser> damo22: so far my impression is that porting OpenWrt to this should be relatively straightforward.

09:04 <damo22> im guessing the serial is a block of 4 pins?

09:06 <damo22> is it usually 3v3 or 1v8?

09:06 <PaulFertser> damo22: likely. See the OpenWrt wiki for ideas how to probe for pinout etc.

09:06 <PaulFertser> damo22: Tx idles high, so just measure to be sure

09:06 <damo22> ok

09:06 <PaulFertser> damo22: for mediatek 3.3 V most likely

09:33 <PaulFertser> damo22: ground and power are clearly visible on that 4-pin connector footprint, so one of the other two is target's Tx.

09:34 <PaulFertser> damo22: and even if you connect target's Rx to your adapter's Rx there's no danger, just no reliable transmission either.

09:34 <f00b4r0> if you're only going to read the ouptut you only need GND and device TX, and then voltage doesn't really matter. Either your transceiver will work or it won't, but you won't damage anything

09:35 <f00b4r0> heh, PaulFertser beat me to it :)

09:38 <PaulFertser> 3.3 V UART can read from 1.8 V usually. Doesn't matter in this case, I'd bet money it's 3.3 V target.

09:39 <f00b4r0> same

09:41 <PaulFertser> damo22: the next step would be to try to stop in bootloader to see if you can load and run an OpenWrt initramfs image, you can probably just try one for board with similar features (cell modem doesn't matter, it's essentially an external USB device) and same (!) SoC which you'll know from the log.

15:00 <schmars[m]> blocktrron: did you have the mt76 "message timeout" bug yourself too? Flashed your patch yesterday and it doesnt look like it helped

15:06 <owrt-images-builds> Build [#71](https://buildbot.openwrt.org/images/#/builders/66/builds/71) of `openwrt-23.05_bmips/bcm6358` completed successfully.

20:58 <Mangix> these mbedtls failures are a pain to fix

21:48 <rsalvaterra> Heh. Just noticed we're using the old version of the rtl8821ae firmware. A newer version was added… 7 years ago.

21:56 <al> was just reading usteer code to get an idea of what the different config options actually do and came across this:

21:56 <al> wat.

21:56 <al> reasons |= (1 << UEV_SELECT_REASON_LOAD);

21:56 <al> if (has_better_load(current_node, new_node) && !has_better_load(current_node, new_node))

21:57 <Habbie> ok so which one is it!

22:01 <al> I'm not hallucinating, right? that's always false, right?

22:02 <Habbie> unless that function can somehow return true and then false in the span of nanoseconds, yes

22:02 <al> yeah ok :)

22:02 <al> I'm really not familiar with this code, but I sent https://github.com/openwrt/usteer/pull/7 for now

22:03 <al> I think another bug is that below_load_threshold actually implements above_load_threshold

22:03 <Habbie> heh

22:06 <al> kinda understand all the posts on various platforms questioning that usteer actually works now

22:14 <rsalvaterra> hauke: I found the culprit for the iproute2 build failure. In 6.8.0, support for iptables action was removed: https://git.kernel.org/pub/scm/network/iproute2/iproute2.git/commit/?id=45dc10463dd5a51eef944b110a4f046feb1b46db

22:15 <al> based on starightforward PRs like https://github.com/openwrt/usteer/pull/3 not getting merged, is usteer dead?

22:16 <Habbie> always hard to say for community/volunteer projects

22:17 <Habbie> also https://github.com/openwrt/usteer?tab=readme-ov-file#submitting-patches

22:20 <al> Oh, thanks

23:13 <damo22> thanks for the tips, i have parked my port for now, until i get a voltmeter