**** BEGIN LOGGING AT Tue Jun 05 03:00:15 2018
Jun 05 03:15:01 <owrt-snap-builde> build #520 of ramips/mt76x8 is complete: Success [build successful]  Build details are at http://phase1.builds.lede-project.org/builders/ramips%2Fmt76x8/builds/520
Jun 05 04:07:10 <wilson3542> also if you ever want to chase down something exotic this old demo started running really slowly somewhere along the way   https://files.scene.org/view/demos/groups/orion/orion_purple_final_hq.zip
Jun 05 04:07:27 <wilson3542> at least on radeon
Jun 05 04:12:30 <wilson3542> whoops wrong channelo
Jun 05 04:12:35 <wilson3542> sorry about that
Jun 05 04:32:16 <aparcar> I'm getting some errors compiling latest snapshot. seems to miss selinux/selinux.h
Jun 05 04:32:17 <aparcar> Is that a local problem or have there been some changes (which I can't find)?
Jun 05 07:11:03 <dwmw2_gone> stintel: the actual changes already got committed to master but not 18.06
Jun 05 07:11:08 <dwmw2_gone> (for 11r)
Jun 05 07:39:15 * ldir wanders in for a bit
Jun 05 08:00:52 <Tapper> Hi ldir
Jun 05 08:04:54 <ldir> hiya Tapper
Jun 05 08:05:36 <Tapper> Not much going on on the github.
Jun 05 08:05:53 <Tapper> I was looking to see what's changed but not mutch
Jun 05 08:06:23 <Tapper> It must be the time of year.
Jun 05 08:06:29 <Tapper> people out in the sun
Jun 05 08:08:06 <ultito> morning
Jun 05 08:08:37 <ldir> Tapper: could be!
Jun 05 08:08:52 <Tapper> hi ultito
Jun 05 08:10:02 <ultito> another hot day will be today...
Jun 05 08:55:59 <aparcar> have there beeen recently changed to busybox? https://paste2.org/JngMvnpp it fails due to various missing libs
Jun 05 09:00:10 <jow> aparcar: no, maybe your host system is dorty or too new
Jun 05 09:00:14 <jow> *dirty
Jun 05 09:01:02 <aparcar> to new?
Jun 05 09:01:26 <aparcar> so I'd run make clean or more to do?
Jun 05 09:01:33 <jow> no idea
Jun 05 09:01:41 <jow> I've never seen that selinux error so far
Jun 05 09:02:00 <jow> apparently busybox is picking up selinux capabilities from somewhere
Jun 05 09:02:03 <stintel> did you maybe uninstall libselinux ?
Jun 05 09:02:06 <stintel> on your host system
Jun 05 09:02:47 <stintel> I vaguely remember running into some issue after selinux was removed from my system (due to no packages depending on it anymore)
Jun 05 09:03:04 <aparcar> stintel: it's installed
Jun 05 09:03:15 <stintel> recently?
Jun 05 09:03:31 <aparcar> however I think the ci machine is some strange container, can that interfere?
Jun 05 09:03:46 <aparcar> stintel: recently what?
Jun 05 09:04:50 <aparcar> I think the error only occours on non x86 targets.
Jun 05 09:05:40 <jow> busybox will only do selinux stuff when "CONFIG_SELINUX" is enabled
Jun 05 09:05:47 <jow> did you customize your busybox settings?
Jun 05 09:06:10 <jow> whats the output of scripts/diffconfig.sh ?
Jun 05 09:07:01 <aparcar> Oh I guess I ran something lime feeds install -a -p repo -d m
Jun 05 09:07:23 <aparcar> So the option -a should be before -p I guess
Jun 05 09:07:29 <stintel> was it (selinux) recently installed on your host
Jun 05 09:07:33 <aparcar> So maybe all options are now activates causing the trouble...
Jun 05 09:07:51 <aparcar> stintel: no I don't think so
Jun 05 09:08:22 <stintel> well verify it
Jun 05 09:10:28 <ldir> a haaa!  package commit to ldns 55bffaace8b9823330ba1753634597e5b60d1bbd breaks building on macos
Jun 05 09:12:41 * ldir needs an adult to interpret https://pastebin.com/n1muNJ2Y
Jun 05 09:13:29 <jow> ldir: I'd say a case of an overly clever, custom package build system not recognizing that it is cross compiling for linux
Jun 05 09:14:23 <jow> probably does something along the lines of  case "$(uname)" in Darwin) CFLAGS="awesome osx mumbo jumbo" ;; esac
Jun 05 09:14:58 <ldir> lol - goes to look - doesn't grep "mumbo jumbo" ;-)
Jun 05 09:15:40 <jow> uh oh, I see custom m4 macros already
Jun 05 09:15:59 <jow> the only thing worse than autoconf is badly written custom autoconf macros
Jun 05 09:16:11 * ldir doesn't understand autoconf at all.
Jun 05 09:17:09 * ldir locally reverts the ldns bump so that he can get a build completed
Jun 05 09:18:05 <jow> alright, my assumption was almost right
Jun 05 09:18:17 <jow> https://pastebin.com/N48Jatms
Jun 05 09:18:25 <jow> custom crapcode in configure.ac
Jun 05 09:18:40 <jow> luckily autoconf is like cancer, you can usualy cure it by removing as much as possible
Jun 05 09:20:01 <jow> their OS X specific hack should be using `echo $target_os` not `echo $build_os`
Jun 05 09:20:45 <jow> it still wouldn't work for cross-building *to* OS X as it extracts info from the host build system
Jun 05 09:21:00 <jow> but it would not disturb cross builds not targeting OS X
Jun 05 09:21:20 <abenz> hahah
Jun 05 09:21:24 <ldir> ok, now that I've found the line in configure.ac - I'll adjust with a patch and try it.... or are you doing the patch?
Jun 05 09:21:37 <jow> can't right now
Jun 05 09:21:50 <ldir> makes sense if I do it (barely understands anything) 'cos at least I can test it on macos :-)
Jun 05 09:22:14 <ldir> it *is* configure.ac I need to patch right?
Jun 05 09:22:28 <jow> yes
Jun 05 09:22:48 <jow> and if it is not the case already, you need to add PKG_FIXUP:=autoreconf in the OpenWrt Makefile to rebuild configure from configure.ac
Jun 05 09:22:57 <ldir> that's already there
Jun 05 09:23:43 <jow> however to keep the patch as minimal as possible (and possibly usptream-applicable) try just swapping `$build_os` with `$target_os`
Jun 05 09:23:55 <jow> instead of removing the whole if-block
Jun 05 09:24:03 <ldir> that's what I was going to do
Jun 05 09:24:10 <jow> allright, great
Jun 05 09:24:13 <ldir> minimal change possible
Jun 05 09:28:40 * ldir breaths again - cheers - thanks jow!  Will send in a PR
Jun 05 09:31:32 * ldir tries again as is an idiot and patched the wrong version.
Jun 05 09:35:15 * ldir cheers properly this time.
Jun 05 09:42:02 <ultito> ETA to new release...?
Jun 05 09:42:53 <ldir> and ^^^^ there it is!!!
Jun 05 09:43:12 <ldir> I knew the daily question would appear at some point :-)
Jun 05 09:43:21 <blogic> *yawn*
Jun 05 09:43:57 <ldir> blogic: mornin'
Jun 05 09:44:23 <ultito> it is really actual topic to all of us who have too many routers to update
Jun 05 09:53:43 <jow> ultito: if there is definitve news, it will be communicated
Jun 05 09:53:56 <jow> there is a certain likelyhood that it'll happen in june
Jun 05 09:54:09 <jow> depending on the workload it can take longer or not
Jun 05 10:08:44 * ldir sends the patch upstream as well.  Collects brownie points, feels good about self.
Jun 05 10:11:05 <ldir> and thanks jow profusely for without his guidance.... etc. etc.
Jun 05 10:11:51 * ldir decides *this* is the year he *IS GOING* to the openwrt dev conference
Jun 05 10:18:54 <Xiatian> whenever the new release hits I really should go through my installed packages listing and remove weird stuff I do not need
Jun 05 10:26:08 <paulius> chown'ing a directory under /etc/ to a different user doesn't seem to work, it stays root:root. is it because of squashfs?
Jun 05 10:26:36 <paulius> works fine for stuff in /tmp/
Jun 05 10:38:16 <Tapper> ** ldir +10 xp for upstreem patch! :-)
Jun 05 10:39:00 <Tapper> were will the openwrt dev conference be this time?
Jun 05 10:39:11 <Tapper> has it bin desided yet?
Jun 05 10:39:35 <Tapper> decided*
Jun 05 10:39:58 <jow> paulius: this is possible, yes
Jun 05 10:40:27 <paulius> jow: know of a workaround?
Jun 05 10:40:33 <jow> paulius: unfortunately not
Jun 05 10:40:59 <jow> Tapper: possibly Lisbon if it happens next to the Summit
Jun 05 10:41:14 <Tapper> K thanks
Jun 05 10:41:42 <Tapper> I have never bin to Lisbon  I would like to
Jun 05 10:44:04 <owrt-snap-builde> build #482 of at91/sama5d3 is complete: Failure [failed images]  Build details are at http://phase1.builds.lede-project.org/builders/at91%2Fsama5d3/builds/482  blamelist: Felix Fietkau <nbd@nbd.name>, Hans Dedecker <dedeckeh@gmail.com>
Jun 05 11:06:28 <ldir> stintel: ping
Jun 05 11:06:58 <stintel> pong
Jun 05 11:08:18 <ldir> stintel: so we were both having 'fun' over 4.14 bumps and x86.... yes I have the same problem... so whatever we've done we've both done the same thing :-)
Jun 05 11:08:47 <ldir> and of course 14.48 is here.
Jun 05 11:10:23 <stintel> heeh
Jun 05 11:11:52 <ashkan_> can I put some non related changes of a Makefile in a single PR ?
Jun 05 11:12:12 <ldir> stintel: and as you're responsible adult ;-)
Jun 05 11:13:03 <stintel> me responsible? oh dear :)
Jun 05 11:13:37 <blogic> ashkan_: inside a PR yes as long as they are split into single patches
Jun 05 11:13:48 <stintel> ashkan_: if it's the Makefile of the same package, then yes + what blogic said ^
Jun 05 11:14:42 <ldir> the thing that bugs me with that...is the blasted package version number bumping
Jun 05 11:15:35 <ldir> that always puts the pressure back on to 'combining changes'.
Jun 05 11:16:04 <ldir> so I can see ashkan_ 's dilemma - I have it too.  Good to get an answer.
Jun 05 11:17:01 <stintel> you can bump the number just in the final commit of your PR
Jun 05 11:31:42 <stintel> ldir: I found the problem with >4.14.44
Jun 05 11:32:04 <ldir> ah, go on.
Jun 05 11:33:09 * ldir is currently bumping 4.9 and so far there are no patch refresh changes WHATSOEVER
Jun 05 11:34:23 <stintel> I was looking at the wrong file (pci-quirks.h), the problem is in pci-quirks.c
Jun 05 11:35:35 <stintel> the function was added after #endif /* CONFIG_PCI_DISABLE_COMMON_QUIRKS */
Jun 05 11:36:08 <ldir> ahh, that sort of defeats the point - oh dear :-)
Jun 05 11:36:14 <stintel> so it was outside of the #ifndef CONFIG_PCI_DISABLE_COMMON_QUIRKS ... #endif /* CONFIG_PCI_DISABLE_COMMON_QUIRKS */
Jun 05 11:36:40 <stintel> and on x86_64 # CONFIG_PCI_DISABLE_COMMON_QUIRKS is not set
Jun 05 11:37:54 <stintel> building now, if it works I'm going to bump to .48 and do some more compile/runtime testing
Jun 05 12:03:10 <stintel> damnit, forgot quilt refresh
Jun 05 12:05:19 <ldir> ha - refresh and the subsequent make update to copy the updated patch into patches dir... yep forgotten many times myself :-)
Jun 05 12:05:59 <stintel> made an error in my first fix of the .c file
Jun 05 12:06:05 <stintel> looked at it again, now it should be good
Jun 05 12:06:29 <stintel> fix is in my stating tree
Jun 05 12:06:36 <stintel> staging*
Jun 05 12:08:33 <stintel> lol, gotomeeting in chromium is io-bound, wtf
Jun 05 12:08:53 <stintel> (extract the kernel tarbal and gotomeeting audio starts dying
Jun 05 12:18:35 <stintel> ldir: and it builds \o/
Jun 05 12:19:05 <ldir> yay!  I've refreshed 4.9 and currently building for ar71xx
Jun 05 12:19:10 <stintel> bumping to 4.14.48 now
Jun 05 12:20:41 <ldir> 4.9.106 barely touched any patches.... 1 only, that's it.
Jun 05 13:03:49 <owrt-snap-builde> build #12 of mediatek/mt7623 is complete: Success [build successful]  Build details are at http://phase1.builds.lede-project.org/builders/mediatek%2Fmt7623/builds/12
Jun 05 13:42:13 <claviger-pc>  someone gave me a european TL-WR710N yesterday, version 1.1 (8mb/32mb), I flashed pepe2k's uboot and updated, making it basically unbrickable, and then installed the latest lede/openwrt build.  But I have no need for it.  Anyone interested in Europe?
Jun 05 14:21:42 <ultito> claviger-pc: just for information, how much for that?
Jun 05 14:23:56 <claviger-pc> well, they are probably $25-30 new, stock, so it is probably worth something in that range with the bit of work I put into it
Jun 05 14:28:19 <claviger-pc> ultito: you interested and in Europe?
Jun 05 14:37:07 <ultito> claviger-pc: maybe I am interested an I am in europe
Jun 05 14:38:03 <claviger-pc> are you a developer or end user?
Jun 05 14:53:56 <mangix> ldir: any chance you can test my ath10k patch? https://github.com/neheb/source/commit/c474361f5d3159f76ba56b23caac67a0428251a9
Jun 05 15:02:41 <claviger-pc> ultito: I would say that unless you have a very particular need for this form factor or model, you'd be better off with a gl-inet router.  But I am happy to sell it to you
Jun 05 16:23:27 <dangole> somehow reboot on mt7688an broke at some point between lede-17.01 and now
Jun 05 16:49:03 <ldir> mangix: will give it a try
Jun 05 16:57:19 <Borromini> hi
Jun 05 17:24:54 <mangix> dangole: could be related to the kernel. there was a patch that fixed halt for ramips.
Jun 05 17:42:35 <max_g> seems protobuf-c is broken in current master: https://pastebin.com/mH9Gan6J
Jun 05 18:03:45 <ldir> mangix: it no worky - wireless doesn't come up
Jun 05 18:04:15 <mangix> ldir: o_0
Jun 05 18:05:16 <mangix> what does dmesg say?
Jun 05 18:07:21 <ldir> https://pastebin.com/j91VRbZM
Jun 05 18:08:59 <mangix> the failure looks unrelated.
Jun 05 18:09:27 <mangix> firmware and board files were loaded
Jun 05 18:11:46 <ldir> well, let me reboot it and check again.  Then I'll go back to version without the patch and test again.
Jun 05 18:14:44 <ldir> reboot still no worky. - waiting for build to finish.
Jun 05 18:23:11 <mangix> this may have broken if you have test mode enabled. will need to respin.
Jun 05 18:23:41 <ldir> If I have test mode enabled I have no idea how / why
Jun 05 18:24:11 <schlimig> Hi. I'm using LuCI lede-17.01 on a Netgear X6. I have installed all (I think) the usb and samba packages and have an NTFS HDD plugged into the USB3 socket. I have successfully mounted the drive in a folder but it is empty and no files are showing. I have run an error scan on a PC and it's in good health. Can anyone assist please?
Jun 05 18:24:18 <mangix> it's an option under kmod-ath
Jun 05 18:24:41 <ldir> will look
Jun 05 18:27:53 <ldir> nope - DFS support & force respect user regdomain settings
Jun 05 18:35:10 <ldir> original firmware without patch worky fine
Jun 05 18:35:29 <mangix> hrm interesting. more testing is needed looks like
Jun 05 18:36:08 <ldir> does it work on your unit then?
Jun 05 18:36:13 <mangix> yes
Jun 05 18:36:25 <mangix> kernel 4.14 though
Jun 05 18:36:56 <ldir> you've an archer c7 working on 4.14 ?
Jun 05 18:37:24 <mangix> no turris omnia. i don't have an archer on 4.14
Jun 05 18:38:30 <mangix> i'm assuming this means my patch works with 4.14 and breaks with 4.9. meaning it's not workable.
Jun 05 18:39:22 <mangix> oh well
Jun 05 18:39:48 <ldir> looks like it.
Jun 05 19:17:33 <Borromini> ldir: where did you get your mi wifi?
Jun 05 19:18:06 <Borromini> i see lots of vendors selling a version with a fast ethernet switch, i suppose you got the gigabit one though
Jun 05 19:19:38 <ldir> the mir3g is sat in the office not even switched on and the archers are still doing archery things.
Jun 05 19:23:03 <Borromini> what's the regular name xiaomi sells that with?
Jun 05 19:23:09 <Borromini> i find their naming scheme confusing :( :(
Jun 05 19:25:43 <stintel> o/
Jun 05 19:27:21 <Borromini> hey stintel
Jun 05 19:27:35 <stintel> going to finish 4.14.48 bump
Jun 05 19:27:37 <Borromini> ldir: when i check amazon i find mt7620 stuff, mt7628...
Jun 05 19:27:45 <Borromini> stintel: can i send one in for 18.06?
Jun 05 19:28:01 <stintel> back from restaurant and gf at her place so I should be able to finish it :)
Jun 05 19:28:13 <Borromini> are you doing master or 18.06?
Jun 05 19:28:16 <stintel> master
Jun 05 19:28:24 <stintel> I am currently not interested in 18.06 due to lack of time
Jun 05 19:28:52 <Borromini> that's ok
Jun 05 19:28:59 <Borromini> i'll send bumps in if ar71xx_tiny checks out here.
Jun 05 19:38:44 <rot13_> I almost can't believe it, but it seems like I've managed to get both a DSL connection and an analog phone using asterisk running on an o2 box 6431! <3 <3 <3
Jun 05 19:38:49 <rot13_> thx to the devs!
Jun 05 19:44:15 <stintel> lbh'er jrypbzr
Jun 05 19:44:59 <Borromini> strange language, bulgarian...
Jun 05 19:46:18 <stintel> кво?
Jun 05 19:46:30 <Borromini> なに
Jun 05 19:46:51 <Borromini> at least i can read cyrillic now, kinda... that's about it :P
Jun 05 19:47:09 <stintel> :P
Jun 05 19:48:17 <stintel> sysupgrading my apu2
Jun 05 19:48:25 <stintel> with master+4.14.48
Jun 05 19:48:27 <Borromini> :)
Jun 05 19:48:38 <stintel> if I am silent after 5 min something went wrong ;)
Jun 05 19:48:44 <Borromini> had some sysupgrade weirdness a few days ago
Jun 05 19:48:51 <stintel> oh the erl already took over
Jun 05 19:48:57 <stintel> love my HA setup ;)
Jun 05 19:49:05 <Borromini> pushed an 18.06 build with a kernel bump, my mt7621 device choked on it
Jun 05 19:49:12 <Borromini> rebooted, was still old build
Jun 05 19:49:21 <stintel> Borromini: read something about that on the ML
Jun 05 19:49:23 <Borromini> only at the fourth try it took the newer one >_>
Jun 05 19:49:26 <Borromini> stintel: oh?
Jun 05 19:49:30 <stintel> something is killing sysupgrade
Jun 05 19:49:34 <Borromini> hmmm.
Jun 05 19:49:44 <stintel> hostapd hanging or so? don't recall exactly
Jun 05 19:49:57 <Borromini> yeah i have no serial so no way to tell what happens after sysupgrade kills all network connections
Jun 05 19:50:00 <Borromini> ok
Jun 05 19:50:41 <Borromini> my brother's paypal account got hacked... i hope it's just bad luck and not some db leak or anything
Jun 05 19:51:04 <ldir> Borromini: is this why you were asking about my mir3g?
Jun 05 19:51:27 <Borromini> ldir: no, just out of curiosity. i have a dir-860l
Jun 05 19:51:39 <Borromini> mir3g looks like a neat cheap AC solution
Jun 05 19:51:40 <salcedo> Borromini: my mom's paypal got hacked. i told her to turn on two factor auth. was a bad idea.
Jun 05 19:51:50 <Borromini> salcedo: turning on 2FA?
Jun 05 19:51:53 <salcedo> never tell normies to turn on 2FA.
Jun 05 19:51:53 <Borromini> i told my brother as much...
Jun 05 19:52:06 <Borromini> my brother can handle it, although i'll need to walk him through it
Jun 05 19:52:07 <stintel> looking good, apu2 back online with 4.14.48 \o/
Jun 05 19:52:24 <salcedo> she got locked out and it was a bigger hastle than when it was hacked
Jun 05 19:52:50 <stintel> 2FA or no whining :)
Jun 05 19:52:55 <stintel> even for normies
Jun 05 19:53:34 <Borromini> i use freeotp myself
Jun 05 19:53:42 <ausjke> can I static-routing-ip(no NAT) from wired-interface to wireless-interface? wired-to-wired works fine, anything to do with wireless driver?
Jun 05 19:53:58 <Borromini> i recommend authy to regular users, authy allows encrypted backups
Jun 05 19:54:07 <Borromini> freeotp doesn't do backups (adb does though :D à
Jun 05 19:54:10 * ldir agrees with stintel
Jun 05 19:54:41 <ausjke> laptop -<wired>-- linuxRouter -- <wireless> -- openwrt_gateway -- internet
Jun 05 19:55:37 <ausjke> laptop can never reach openwrt_gateway no matter what I do on linuxRouter, with ip_forward to 1, no iptables/nat is used, as I want to use static route
Jun 05 19:56:07 <ausjke> if linuxrouter has two wired interface it works I believe, tested before, but not with wireless
Jun 05 19:56:48 <ausjke> NAT, proxyarp?, may get it work, but i want to play with static-route
Jun 05 19:58:27 <schm0> hmmm.... i now have some problem with my ingress markings for cake...
Jun 05 19:58:50 <schm0> i see packets hitting the voice tin that arent there?
Jun 05 20:06:20 <schm0> when i only use the tc to set cs0.... i see only packets hitting the best effort tin, as it should, but also packets hitting the voice tin?
Jun 05 20:12:20 <schm0> the tc catch all mirror to ifb rule shows 1748116 pkts, cake cs0 tin shows 1749385 pkts and the voice tin shows 33896 pkts
Jun 05 20:12:36 <schm0> what are those ~33 packets?
Jun 05 20:12:43 <schm0> 33k
Jun 05 20:17:01 <schm0> @tohojo ping
Jun 05 20:21:06 <luaraneda> blogic: ping (zynq PR reminder)
Jun 05 20:31:15 <schm0> hmm those must non ip packets... i think those are arp packets... but 4mb of arp traffic in like 20min
Jun 05 20:31:19 <schm0> i dont know
Jun 05 20:36:25 <stintel> ldir: pushed 4.14.48
Jun 05 20:36:51 <ldir> responsible adult!
Jun 05 20:41:38 <Borromini> hmmm
Jun 05 20:42:27 <schm0> hmm those arp packets are normal for docsis :-/ https://github.com/dtaht/sch_cake/commit/644b7efb2a552ba76871e65033c58275c15207f9#diff-f53c48c3fbc2331ed9b893e8157f66c6
Jun 05 20:43:01 <stintel> hah yeah docsis has some annoyances
Jun 05 20:43:12 <stintel> but I'll take it over *any* DSL any day
Jun 05 20:43:21 <schm0> but i dont want them to mess up my diffserv. so filter them out with tc or remove the cake arp/cs7 commit?
Jun 05 20:46:19 <schm0> docsis sucks
Jun 05 20:47:00 <stintel> well there's a good result: http://www.dslreports.com/speedtest/34267493
Jun 05 20:47:14 <schm0> the sqm scripts come with an redirect all to ifb tc rule.. when i change this rule to ip only? can it break somethin?
Jun 05 20:48:45 <Borromini> stintel: just run-tested ar71xx, will do ramips now and then e-mail my patches :)
Jun 05 20:48:48 <Borromini> night gents.
Jun 05 21:01:57 <stintel> ooh and then my apu2 rebooted
Jun 05 21:02:15 <stintel> I suspect flow offloading again
Jun 05 21:06:03 <stintel>  /nr 54
Jun 05 21:24:03 <stintel> hmmm now keepalived seems to be broken
Jun 05 21:31:34 <paulius> and random disconnects often
Jun 05 21:31:54 <stintel> paulius: with flow offload?
Jun 05 21:31:58 <paulius> yes
Jun 05 21:32:09 <paulius> over wifi
Jun 05 21:32:26 <stintel> interesting, might be related to my keepalived switching to master on the backup node while master is still active
Jun 05 21:32:56 <stintel> if random packets disappear
Jun 05 21:33:03 <paulius> should I use a different branch for now or?
Jun 05 21:33:22 <stintel> well you can just disable flow offloading and see if that actually improves your situation
Jun 05 21:33:42 <stintel> if it does, report a bug
Jun 05 21:34:26 <paulius> ok
Jun 05 21:34:34 <paulius> thanks
Jun 05 21:34:44 <paulius> will report back
Jun 05 21:34:48 <stintel> great, thanks
Jun 05 21:34:52 <paulius> tu be sure yo
Jun 05 21:34:59 <paulius> you mean hw or sw?
Jun 05 21:35:07 <stintel> here it's only software
Jun 05 21:35:33 <stintel> might be a different problem altogether though, I've seen weirdness with my HA setup (keepalived) before
Jun 05 21:35:46 <paulius> i have wf offload enabled
Jun 05 21:35:51 <paulius> hw*
Jun 05 21:36:04 <paulius> oh right i see what you mean
Jun 05 21:36:20 <paulius> hw offload not applicable for wifi
Jun 05 21:41:10 <paulius> apparently 2.4ghz works fine, havent changed anything yet
Jun 06 02:14:30 <mangix> what the
Jun 06 02:14:41 <mangix> ipq8 toolchain's not compiling
Jun 06 02:24:02 <owrt-snap-builde> build #870 of mpc85xx/generic is complete: Failure [failed sourceupload]  Build details are at http://phase1.builds.lede-project.org/builders/mpc85xx%2Fgeneric/builds/870  blamelist: Felix Fietkau <nbd@nbd.name>, Stijn Tintel <stijn@linux-ipv6.be>, Daniel Golle <daniel@makrotopia.org>, Ted Hess <thess@kitschensync.net>
Jun 06 02:36:09 <pkgadd> mangix: I've just successfully built and tested ipq8065, including your ath10k patch
Jun 06 02:37:09 <pkgadd> mangix: http://paste.debian.net/hidden/f5b5ba8b/
Jun 06 02:40:36 <mangix> pkgadd: must be a GCC8 issue then
Jun 06 02:40:48 <mangix> ath10k's working fine on 4.14 looks like
Jun 06 02:42:10 <pkgadd> I don't have any ar71xx devices with ath10k wireless, I could revert to kernel 4.9 on ipq8068 or lantiq for testing, but...
Jun 06 02:42:19 <mangix> no worries
Jun 06 02:42:31 <mangix> it's a cosmetic patch anyways
**** ENDING LOGGING AT Wed Jun 06 03:00:10 2018