**** BEGIN LOGGING AT Thu Jan 21 02:59:57 2021
Jan 21 03:02:54 <aparcar[m]> well lmk I'll then push it to my buildbots
Jan 21 03:03:31 <mangix> the toolchain compiled on CentOS 7 so that's promising
Jan 21 03:04:07 <mangix> I personally want to get rid of uClibc++
Jan 21 04:38:46 <mangix> aparcar[m]: all finished successfully. do you want me to send a patch or you'll handle it?
Jan 21 04:47:53 <aparcar[m]> Send patches please mangix
Jan 21 04:52:13 <mangix> sent
Jan 21 05:11:21 <rr123> under menuconfig I can easily turn off procd, what will happen if i turn it off? thought it is one of those must-have packages
Jan 21 06:05:03 <owrt-snap-builds> build #57 of realtek/generic is complete: Failure [failed pkgbuild]  Build details are at http://buildbot.openwrt.org/master/images/builders/realtek%2Fgeneric/builds/57  blamelist: Rafa? Mi?ecki <rafal@milecki.pl>
Jan 21 07:49:57 <blogic> Hauke: gpon
Jan 21 07:50:10 <blogic> Hauke: I have a modem running v5.4 kernel here
Jan 21 07:50:21 <blogic> Hauke: already boots, need to work on eth part next
Jan 21 07:54:42 <ynezz> which modem is that?
Jan 21 07:54:51 <blogic> econet
Jan 21 08:01:39 <ynezz> I've here compal which seems to be puma6
Jan 21 08:03:36 <ynezz> ugh, now I found it, you're talking about gpon
Jan 21 08:07:39 <mangix> subsidiary of mediatek. interesting
Jan 21 08:34:45 <mrkiko> blogic: wow, very nice & interesting!
Jan 21 08:36:31 <rsalvaterra> 'morning!
Jan 21 08:38:20 <rsalvaterra> Is anyone around here running OpenWrt on MT7615 hardware, at the moment?
Jan 21 09:00:14 <blocktrron> lmore377: I'll have a look
Jan 21 09:01:05 <blocktrron> I'll do a review about everything, because GitHub is PITA to view past reviews as soon as they have been marked as resolved
Jan 21 09:47:21 <mixi_> hello, are there paid jobs for developing openwrt? I'd love to contribute but somehow have to pay bills
Jan 21 09:50:25 <Borromini> mixi_: i don't think there are any, but there's companies looking for people working on/with openwrt sometimes, like on the forum or the mailing list, e.g. https://forum.openwrt.org/t/hiring-an-openwrt-developer/85924
Jan 21 09:52:49 <mixi_> thank you very much Borromini , thats a helpful hint
Jan 21 09:56:28 <ynezz> ugh, 17.87/hr avg hourly rate paid
Jan 21 09:57:08 <stintel> some people would be very happy with that ;)
Jan 21 09:57:37 <ynezz> yeah, depends on the location
Jan 21 10:18:23 <danitool> hi, does still 18.06 accept device fixes backports?
Jan 21 10:19:04 <ynezz> 18.06 is EOL
Jan 21 10:20:01 <ynezz> there is ongoing discussion whether to do next 18.06 release due to the recent dnsmasq security issue
Jan 21 10:20:05 <danitool> I know it, no new releases
Jan 21 10:20:44 <ynezz> ah, I misread that
Jan 21 10:20:51 <stintel> does it make sense to keep accepting fixes for people who build their own images?
Jan 21 10:20:52 <ynezz> well, you can always try that
Jan 21 10:20:59 <danitool> but without new releases, it could still merge some fixes
Jan 21 10:21:01 <ynezz> send a patch and lets see
Jan 21 10:21:18 <ynezz> as always, fix for one might be regression for other one :)
Jan 21 10:22:47 <danitool> well, for a particular device there is no risk, futhermore if the fix is made on the specific device dts file
Jan 21 10:24:33 <ynezz> the simple fixes and tiny changes sometimes cause big problems
Jan 21 10:25:04 <Borromini> stintel: i'd say it does
Jan 21 10:25:05 <danitool> sure, a butterfly can destroy the world
Jan 21 10:25:12 <ynezz> is the bug reported on bugs.openwrt.org?
Jan 21 10:25:28 <ynezz> is the fix in the tree already, aka it's just backport?
Jan 21 10:25:32 <danitool> no, it isn't
Jan 21 10:25:57 <danitool> no, it isn't <-- the first question
Jan 21 10:26:29 <danitool> yes, it's just a backport <--- 2nd
Jan 21 10:27:31 <danitool> btw I'll search in the bugtracker to be sure there is no report
Jan 21 10:27:40 <ynezz> or forum
Jan 21 10:28:21 <ynezz> it would certainly help in the decision
Jan 21 10:28:24 <danitool> at the forum there was a report
Jan 21 10:29:25 <ynezz> so add `References: https:/forum.openwrt.org/foo/url` to the patch
Jan 21 10:29:42 <danitool> ok, thank you
Jan 21 10:29:43 <ynezz> above the Signed-off:
Jan 21 10:30:22 <danitool> BTW I own the device
Jan 21 10:32:01 <danitool> it's just a wifi problem caused by missing pinctrl
Jan 21 10:33:22 <danitool> I wanted to have a functional 18.06 at least for testing purposes
Jan 21 10:33:37 <ynezz> BTW it's prefered to have backports done with `git cherry-pick -x` as it keeps reference to the upstream commit
Jan 21 10:43:24 <mangix> cannot find -llibubox
Jan 21 10:43:29 <mangix> what the hell?
Jan 21 10:44:42 <ynezz> it's cmake
Jan 21 10:46:01 <Borromini> danitool: what hardware?
Jan 21 10:47:30 <rmilecki> mixi_: sure they are, but it's a relatively small market
Jan 21 10:47:47 <rmilecki> mixi_: if you find company that needs something generic developer for OpenWrt, it's usually simpler
Jan 21 10:48:00 <rmilecki> mixi_: some companies need dev working on specific hardware
Jan 21 10:48:20 <rmilecki> so it's sometimes hard for company and devs to "find" each other
Jan 21 10:48:32 <mangix> ynezz: that's completely wrong though.
Jan 21 10:49:01 <rmilecki> e.g. demand for Broadcom is quite low but I got company that uses it & OpenWrt
Jan 21 10:49:32 <mixi_> rmilecki, so how to proceed if you want a career in openwrt?
Jan 21 10:50:02 <stintel> replace the firewalls at your current job with OpenWrt :P
Jan 21 10:50:25 <Borromini> :P
Jan 21 10:50:28 <rmilecki> mixi_: if you are a developer, i suggest posting on ML and forum, let know what you can work on (some target? specific hw? upstream Linux? software?)
Jan 21 10:50:29 <Borromini> stintel is a selfmade man :P
Jan 21 10:50:47 <rmilecki> mixi_: link your contributions (git commits)
Jan 21 10:51:13 <rmilecki> mixi_: also search forum / ML for job offers
Jan 21 10:51:37 <rmilecki> it's a small market, i'd even try replying to some old posts
Jan 21 10:52:27 <mangix> oh i see
Jan 21 10:52:30 <mangix> local changer
Jan 21 10:52:32 <mangix> *change
Jan 21 11:06:38 <danitool> Borromini: bcm63xx, BCM6368 SoC  https://openwrt.org/toh/observatelecom/vh4032n
Jan 21 11:54:30 <owrt-snap-builds> build #738 of apm821xx/sata is complete: Failure [failed updatefeeds]  Build details are at http://buildbot.openwrt.org/master/images/builders/apm821xx%2Fsata/builds/738  blamelist: Rafa? Mi?ecki <rafal@milecki.pl>
Jan 21 12:14:36 <ynezz> rmilecki: hi, is anything preventing merging of https://patchwork.ozlabs.org/project/openwrt/list/?series=225499 ? ("v4 kernel: mtdsplit_uimage: use device tree properties for non-standard uimage parsing")
Jan 21 12:15:04 <ynezz> AFAIK it blocks some development so I would like to get it in
Jan 21 12:20:58 <rmilecki> ynezz: looks good to me!
Jan 21 12:24:01 <ynezz> thanks
Jan 21 12:24:11 <rmilecki> np :)
Jan 21 12:28:06 <svanheule[m]> ynezz: Borromini was testing these patches again, but bricked one of his devices
Jan 21 12:31:54 <Borromini> yes i have a DIR-878 A1 (which uses the sge parser bmork modified)
Jan 21 12:32:13 <ynezz> is bmork aware? :)
Jan 21 12:32:38 <Borromini> and i'll be trying an rt-ac57u which i need to revive first from an unrelated issue
Jan 21 12:33:04 <Borromini> ynezz: the DIR-878 is production, sysupgrade bricked it, since the d-link recovery doesn't like linux browsers i can't say for sure if a clean flash works.
Jan 21 12:33:08 <Borromini> so i'm holding off on that for now
Jan 21 12:33:30 <Borromini> sysupgrade was without -n for sure, so might not have been the best approach either (but 'production') :P
Jan 21 12:34:00 <Borromini> that being said, i flashed a build with the v4 patchset on a realtek switch (gs1900-8hp) and that doesn't seem to show any issues
Jan 21 12:34:09 <Borromini> i recall realtek being part of the targets that needed testing?
Jan 21 12:35:40 <ynezz> any testing is always welcome
Jan 21 12:36:08 <ynezz> but yes, I've built the realtek target as well, since that target was touched as well
Jan 21 12:38:14 <ynezz> Borromini: could you please send a reply about your experience to the v4 thread?
Jan 21 12:38:23 <ynezz> FYI I just kicked another build with v4 of the patches https://gitlab.com/ynezz/openwrt/-/pipelines/244909588/builds when/if it finishes, then it should show up on https://foo.true.cz/minio/openwrt/staging-builds/testing-bjorn-mtdsplit/
Jan 21 12:39:06 <Borromini> ynezz: i will do, but only after i made sure that a clean flash breaks the dir-878 A1 as well
Jan 21 12:39:15 <Borromini> which will hopefully be tonight.
Jan 21 12:40:00 <ynezz> even that sysupgrade issue is unfortunate I would say
Jan 21 12:41:01 <ynezz> I wouldn't expect this kind of breakage for such changes either
Jan 21 12:41:49 <ynezz> let's see what you find out
Jan 21 12:42:11 <Borromini> i'll keep you posted
Jan 21 14:22:21 <ldir> Hello unix gurus - I'm looking for a neat way of doing the following.  I have a command 'foo' that returns two lines of text, the 2nd line contains an access key - in essence "Authentication success\nkey is keytext"   neat way of getting 'keytext' into a variable.
Jan 21 14:23:31 <stintel> | tail -n1 | awk '{print $3}' ?
Jan 21 14:24:24 <ldir> that's close to what I have already  tail -n1 | cut -d' ' -f3
Jan 21 14:34:57 <jow> command | sed -ne '2s#key is ##p'
Jan 21 14:35:05 <jow> command | sed -ne '2s#^key is ##p'
Jan 21 14:35:52 <jow> or real dirty, if keytext never contains whitespace and is never followed by something else:
Jan 21 14:36:10 <jow> for key in $(command); do :; done  # $key contains keytext after the loop
Jan 21 14:37:09 <ldir> oooh yuck :-)
Jan 21 14:38:35 <karlp> oh, that's special :)
Jan 21 14:38:36 <karlp> I like that :)
Jan 21 14:40:01 <Borromini> $command|awk '/key/ {print $3}' should work as well
Jan 21 14:40:04 * Borromini thinks
Jan 21 14:46:31 * ldir went with jow's sed solution
Jan 21 15:40:20 <lmore377> blocktrron: thank you!
Jan 21 15:42:30 <blocktrron> lmore377: ping me once you've updated the PR
Jan 21 15:44:38 <ldir> I ended up with this as it turned out I needed to get rid of some trailing whitespace - sed -ne '2s#^key is ##;2s# *$##p'
Jan 21 15:47:18 <lmore377> blocktrron: one quick note about the led situation, on this device the led is red/blue and it's connected to a microcontroller and based on how you switch those 3 pins on and off it'll do different patterns of either red or blue. if you look at some of the past comments there's one that explains it better (i can copy paste it to yours in a bit)
Jan 21 15:49:05 <adrianschmutzler> lmore377: but still, labels should follow the common scheme
Jan 21 15:52:55 <ynezz> we should really automate that, people won't argue with robots :p
Jan 21 16:09:36 <blocktrron> lmore377: wasn't aware of that, in this case this is fine
Jan 21 16:10:32 <blocktrron> > if you look at some of the past comments there's one that explains it better - This is what i was referring to with "Reviewing long running GItHub PRs as an outsider is pain"
Jan 21 16:13:05 <adrianschmutzler> still, the labels should at least converted to something like just "ledctrl1", "ledctrl2", etc.
Jan 21 17:07:15 <danitool> hi, I can't download the upgrade image into the router from the Openwrt website
Jan 21 17:07:42 <danitool> I get "Failed to redirect to /releases/19.07.6/target....."
Jan 21 17:07:50 <danitool> is this the expected behavior?
Jan 21 17:08:06 <stintel> danitool: you'll have to be more specific than "the upgrade image"
Jan 21 17:08:17 <danitool> the sysupgrade image
Jan 21 17:08:34 <danitool> well, probably any image
Jan 21 17:08:58 <danitool> I mean downloading the sysupgrade image using openwrt itself
Jan 21 17:09:33 <danitool> with the browser at my PC the download is ok
Jan 21 17:12:31 <danitool> mm, probably it is redirecting to the https url, but the "wget" lacks https capacity
Jan 21 17:34:26 <inhaller> hello everyone
Jan 21 17:37:51 <inhaller> i'm trying to port openwrt to an Aerohive AP230, but i'm having trouble getting the squashfs image in the right format
Jan 21 17:51:26 <owrt-snap-builds> build #58 of realtek/generic is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org/master/images/builders/realtek%2Fgeneric/builds/58
Jan 21 19:05:00 <rsalvaterra> mangix: still targeting musl 1.2.1? I just noticed 1.2.2 is out… :P
Jan 21 19:43:24 <adrianschmutzler> Well, I think he started with 1.2.0 and we are still waiting for 21.xx ...
Jan 21 20:16:35 <aliceussr> Hello! Help me, please! Who have driver for MediaTek RT5390 (int. SoC), for full speed WiFi?
Jan 21 21:30:26 <hurricos> How does one tickle the mail bot?
Jan 21 21:36:04 <ldir> very carefully
Jan 22 00:22:04 <owrt-snap-builds> build #739 of apm821xx/sata is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org/master/images/builders/apm821xx%2Fsata/builds/739
Jan 22 00:25:39 <lmore377> blocktrron: so do i just delete the buttons from the device tree? they work fine
Jan 22 00:30:49 <mangix> i didn't know the er-x sfp needed a 24 volt power supply
Jan 22 00:30:50 <mangix> sigh
**** ENDING LOGGING AT Fri Jan 22 02:59:58 2021