**** BEGIN LOGGING AT Sun Aug 18 02:59:57 2019 Aug 18 10:34:56 WIFI is broke on wrt3200acm Aug 18 10:34:59 Sat Aug 17 22:09:13 2019 daemon.notice netifd: radio0 (2141): WARNING: Variable 'data' does not exist or is not an array/object Aug 18 10:34:59 Sat Aug 17 22:09:13 2019 daemon.notice netifd: radio1 (2139): WARNING: Variable 'data' does not exist or is not an array/object Aug 18 10:35:12 latest OpenWrt master Aug 18 10:35:30 Sat Aug 17 22:09:13 2019 daemon.notice netifd: radio1 (2083): Could not find PHY for device 'radio1' Aug 18 10:35:31 Sat Aug 17 22:09:13 2019 daemon.notice netifd: radio0 (2082): Could not find PHY for device 'radio0' Aug 18 10:50:22 I have seen the mails on the ML about mvebu so ignore my last posts Aug 18 12:11:33 Tapper: should be fixed in https://git.openwrt.org/?p=openwrt/openwrt.git;a=commitdiff;h=0c43219a3579a5b86e8b62d603f67872329657b0 Aug 18 12:11:58 Hauke cool. Aug 18 12:12:24 This would not happen if marvel where not dicks! Aug 18 12:13:26 :P Aug 18 12:14:26 I think NXP now own the wifi parts Aug 18 12:14:31 *owns Aug 18 12:14:48 * Tapper thinks it's all a mes! Aug 18 12:15:39 NXP should just opensource the driver! Aug 18 12:15:51 https://github.com/kaloz/mwlwifi/issues/360#issuecomment-491966299 < did you see this tapper? Aug 18 12:15:56 the driver is open source, but not in the mainline kernel Aug 18 12:16:17 Hauke sorry I ment the firmware Aug 18 12:17:50 of course when i say out loud on the forum that peoples should stop recommending marvel hardware, dizzy starts his crusade again Aug 18 12:17:57 Borromini Yeah I sor it. Aug 18 12:18:15 because i'm 'only plugging mt7621' Aug 18 12:18:30 Borromini lol Diizzy is not that bad. Aug 18 12:18:44 he is Tapper. you may not have seen how he behaved in here Aug 18 12:18:49 He just does not know when to shut up Aug 18 12:19:31 which can be glossed over once, or twice Aug 18 12:19:34 but not all the time Aug 18 12:19:45 it's like a kid going on about his pet peeve Aug 18 12:19:51 except this kid is probably physicaly grown Aug 18 12:19:57 whole different ball game Aug 18 12:20:00 technically quantenna would be wifi with open source firmware; they run a full-fledged u-boot + linux on their cards Aug 18 12:20:02 Yeah I know mate. I like to see the best in people. it's just the way I am Aug 18 12:20:59 Tapper: good for you... but that doesn't make it any better :P Aug 18 12:21:00 My wife tells me I am to soft with my 5 kids all the time! lol Aug 18 12:21:40 but they struggle since months with how to upstream binaries to linux-firmware and provide the sources in an acceptable way, without putting the full SDK into the git tree Aug 18 12:21:54 why am i not surprised ;) Aug 18 12:22:29 Borromini Is that for me or the monster? Aug 18 12:22:36 KanjiMonster Aug 18 12:22:43 for you Aug 18 12:23:03 * Tapper grins! Aug 18 12:26:09 as it looks like linksys is the only high volume customer of these marvell wifi devices and these devices are already some years old and they are probably working on the ax generation I do not expect there anything any more Aug 18 12:26:26 and linksys probably does not sell so much Aug 18 12:26:45 lots of refurb nowadays it seems for the wrt series Aug 18 12:28:11 ok, with not so much, I anyway mean less then 1M per year Aug 18 12:28:19 :D Aug 18 12:29:52 but if NXP now uses this and they want to use this for automotive and industry it could be that their customers want to have open source and mainline drivers becasue they ahev to maintain this for 15 years Aug 18 12:29:59 *have Aug 18 12:30:41 The NXP SoCs are pretty good supported upstream Aug 18 12:31:06 but probably also expensive Aug 18 13:24:17 finger crossed i say. Aug 18 13:24:24 *fingers Aug 18 16:27:22 updated openwrt/upstream, https://sdwalker.github.io/uscan/index.html Aug 18 18:29:08 any (u)qmi experts around? Aug 18 19:27:21 jow: sort of Aug 18 20:13:22 jow: could you please fix this https://github.com/openwrt/openwrt/pull/2339 as it fixes two targets for the snapshot image server Aug 18 20:13:38 s/fix/merge/g Aug 18 20:57:44 Hauke: thanks for looking into ath10k, confirmed working (nbg6817/ QCA9984 && r10825-2ceee0e023 && kernel v4.19.67) Aug 18 20:59:36 pkgadd: nice Aug 18 21:00:09 I am intretsed how QCA tested this code Aug 18 21:00:19 *interested Aug 18 21:10:38 if at all, probably on x86_64 Aug 18 21:11:14 but "SWBA overrun on vdev 0, skipped old beacon" (which is going rampant) suggests that they didn't test it at all Aug 18 21:12:25 pkgadd: will you send a mail becasue of the "SWBA overrun on vdev 0, skipped old beacon" to the wireless mailling list? Aug 18 21:15:10 I'll have a look (the new version is just running for ~45 minutes now) and will try to check the older backports versions first, to get a bit closer to the culprit (the v4.9 with the same firmware, 10.4-3.10-00047, didn't show it) Aug 18 21:15:28 err, v4.19 based backports Aug 18 21:15:42 good evening Aug 18 21:16:37 I hope I can reproduce it on my BT Home Hub 5 Type A/ QCA9880 as well, as doing testing on that one would be considerably less disruptive Aug 18 21:27:53 Anyone has the network configuration for sky UK ? Aug 18 22:16:26 pkgadd: thank you for takeing care of this problem **** BEGIN LOGGING AT Sun Aug 18 22:16:43 2019 **** ENDING LOGGING AT Mon Aug 19 02:59:57 2019