**** BEGIN LOGGING AT Mon May 18 02:59:57 2020
May 18 03:56:02 <rr123> GOOS=linux GOARCH=mips go build; tryint to run a helloworld go binary on TP-Link Archer C7, it gavve me "Illegal insutrction", anyone tried to run go binary on MIPS 74Kc?
May 18 04:11:27 <rr123> to answer my own question, I need add floatpoint as the default is using hard float point
May 18 04:11:34 <rr123> now go runs well
May 18 06:36:35 <ynezz> jow: pong
May 18 07:21:58 <dengqf6> blogic: upstream rtl DSA driver does not have tag support
May 18 07:22:24 <dengqf6> blogic: there is a programming guide that describes the rtl tag format
May 18 07:24:22 <dengqf6> linusw: ping
May 18 07:52:27 <jow> ynezz: was about reassembling the buildbot cluster, but I did that yesterday evening. Seems somehow the connection to the build works was interrupted
May 18 07:53:31 <jow> ynezz: not sure if that affected 19.07.3 builds or if all completed successfully
May 18 07:55:57 <russell--> can anyone tell me how rfkill works on boot (i.e. coldplug). the wdr3600 has a physical slide switch that triggers /etc/rc.button/rfkill, but i'm not seeing where the switch position is detected during boot.
May 18 08:09:51 <russell--> ah, interesting. looks like the rc.button script just commits to /etc/config/wireless and so the state is read at boot. if the switch is changed with power off, nothing notices.
May 18 08:09:56 <ynezz> jow: hm, on all build slaves, or just those at OSUOSL ?
May 18 09:51:27 <jow> ynezz: just osuosl
May 18 09:56:07 <ynezz> jow: reassembling the buildbot cluster = `ansible-playbook -i inventory buildslave.yml --tags reconfigure` ?
May 18 10:16:23 <jow> ynezz: that and manually logging into all four osuosl machines and restarting the docker instances
May 18 10:16:36 <jow> ynezz: the somehow were in limbo and didn't reconnect
May 18 10:19:39 <ynezz> docker :)
May 18 10:20:32 <ynezz> FYI I've noticed it as well, two containers restarted, two left in previous state, added that into my todo list
May 18 10:22:05 <jow> I also reshuffled some slaves to get 18.06 package builds going again
May 18 10:22:27 <jow> this time I committed everything to avoid surprises when ansible is rerun the next time
May 18 10:25:22 <ynezz> jow: ok,  but you need to push it as well :)
May 18 10:25:48 <jow> thought I ded, let me check
May 18 10:25:52 <jow> *did
May 18 10:26:15 <jow> oops. now I did
May 18 10:30:42 <ynezz> ok, thanks
May 18 11:29:45 <lynxis> we have ansible scripts?
May 18 11:31:18 <stintel> playbooks* ;)
May 18 11:32:35 <lynxis> ... and roles?
May 18 11:33:31 <lynxis> stintel: do you know where are they? do we have a git repo for it?
May 18 11:34:07 <stintel> lynxis: I believe they are in a private repo
May 18 11:34:19 <stintel> lynxis: jow tell you the details
May 18 11:35:53 <stintel> + can
May 18 12:28:19 <Neighbor11111115> Hi all, I'm a little confused. When I look at /proc/meminfo I've noticed that SwapTotal and SwapFree are not 0. How can that be? I'd assume routers have no swap
May 18 12:29:12 <hell__> Neighbor11111115: on which board does that happen?
May 18 12:29:55 <Neighbor11111115> It's an reference board so could be a mistake. It's based on mt7621
May 18 12:30:07 <Neighbor11111115> Am I right in assuming SwapTotal should be 0?
May 18 12:30:16 <hell__> this TL-WR841N with 32 MiB of RAM and 4 MiB of flash has no swap at all
May 18 12:30:33 <Neighbor11111115> OK, may be we've messed up somewhere
May 18 12:31:17 <hell__> if your board has more RAM/flash, maybe it has some swap. but I know very little
May 18 12:31:56 <Neighbor11111115> Given the wearing properties of flash I'd assume using it as swap is a very bad idea
May 18 12:32:57 <urjaman> i suppose it could have zswap (compressed ram), but i dunno if that's a default openwrt setup anywhere
May 18 12:36:25 <PaulFertser> Neighbor11111115: cat /proc/swaps
May 18 12:36:39 <Neighbor11111115> thanks guys, one sec
May 18 12:38:46 <karlp> swapon -s if you can't remember the proc files.
May 18 12:39:11 <owrt-snap-builds> build #325 of x86/legacy is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org/master/images/builders/x86%2Flegacy/builds/325
May 18 12:42:02 <Neighbor11111115> ok found out were using zram, got this:
May 18 12:42:04 <Neighbor11111115> Filename                                Type            Size    Used    Priority
May 18 12:42:09 <Neighbor11111115> Filename                                Type            Size    Used    Priority
May 18 12:42:09 <Neighbor11111115>  /dev/zram0                              partition      61436   6032    -2
May 18 14:18:03 <Hauke> dengqf6: where do I find the RTL programming guide for tags?
May 18 14:18:24 <dengqf6> Hauke: https://github.com/libc0607/Realtek_switch_hacking/tree/files?files=1
May 18 14:18:54 <dengqf6> Hauke: I have informed linusw (Linus Walleij)
May 18 14:19:40 <Hauke> dengqf6: thank you for the link
May 18 14:42:49 <zorun> I had forgotten how hard it is to plug those small SMA connectors (it's called U.FL I think?)
May 18 14:43:40 <zorun> the hardest part is resisting the urge to smash the things together...
May 18 14:49:40 <f00b4r0> zorun: I usually use the back of a pencil for that exact purpose :)
May 18 14:50:08 <f00b4r0> flat surface, even pressure.
May 18 15:04:24 <zorun> f00b4r0: thanks for the tip
May 18 15:04:35 <zorun> I used the tip of a knife but it's indeed too thin
May 18 17:20:44 <malwar3hun73r> the original maintainer of softlow has not kept it up-to-date in years... he pointed me to https://github.com/irino/softflowd
May 18 17:21:03 <malwar3hun73r> this is more up-to-date and has many fixes - would it be possible to add this version to openwrt?
May 18 17:26:42 <PaulFertser> malwar3hun73r: packages feed accepts "github pull requests"
May 18 17:31:52 <malwar3hun73r> i guess i need to read up on that (i'm sure there's a link somewhere)... not sure of the process or if this is even compatible
May 18 19:05:14 <Hauke> dengqf6: we have the CPU tag informations even under GPL: https://git.openwrt.org/?p=openwrt/openwrt.git;a=blob;f=target/linux/mediatek/files-5.4/drivers/net/phy/rtk/rtl8367c/rtl8367c_asicdrv_cputag.c;h=d22bf65eaa10a82dffe71c3e718713a26b33df2d;hb=HEAD
May 18 19:56:39 <adrianschmutzler> xback: ping
May 18 19:58:51 <adrianschmutzler> just wanted to merge Thibaut's license patches, but you changed one line in rb_hardconfig.c, so I have to ask you for an Acked-by :-)
May 18 20:08:04 <f00b4r0> ?_?
May 18 20:11:54 <f00b4r0> adrianschmutzler: I'm sure xback won't oppose but I'd like to point out that 1) I'm the sole copyright owner for these files and 2) I do not relicense, I merely clarify the SPDX license identifier ("GPL-2.0" isn't valid).
May 18 20:12:49 <f00b4r0> i'm not sure what's there to warrant an Acked-By.
May 18 20:29:17 <lnslbrty> Hello (.*), does any1 know a comfortable way to debug build issues which occur only if running make with multiple build jobs?
May 18 20:29:57 <lnslbrty> (make -j1 V=s is not a solution in this case)
May 18 20:31:45 <Hauke> lnslbrty: you can activate logging
May 18 20:32:08 <Hauke> lnslbrty: CONFIG_BUILD_LOG
May 18 20:32:42 <Hauke> lnslbrty: then you can grep like this: "grep Error logs/ -r"
May 18 20:33:55 <lnslbrty> Hauke: didn't know about CONFIG_BUILD_LOG, thanks
May 18 20:35:58 <adrianschmutzler> f00b4r0: I agree with your second point, but still it doesn't hurt to ask for a merely cosmetic change, so there is no hurry.
May 18 20:37:03 <adrianschmutzler> considering your first point, my experience here is that it was requested that everybody touching the file (except mere copying/splitting) has to be asked
May 18 20:45:43 <f00b4r0> adrianschmutzler: IANAL but I understand enough of the copyright law to tell you that only the copyright owner(s) have any claim to licensing issues. The people who commit and do not add their copyright (I've been in that case) wave any such claim.
May 18 20:46:07 <f00b4r0> as for the second point, it's a fix: the files will be rejected by upstream kernel maintainers with a broken SPDX identifier.
May 18 20:46:38 <f00b4r0> I'm trying really hard not to take this the wrong way (I had a bad day), but it's not easy :P
May 18 20:48:22 <f00b4r0> The point being there is no change to speak of. I'm simply fixing the identifier to match the license I chose and explicited in the file boilerplate.
May 18 20:49:52 <f00b4r0> (which is imo a very good reason to always have such a boilerplate: I have yet to see files with only SPDX and none of the license required text be challenged in court.
May 18 20:49:54 <f00b4r0> )
May 18 21:00:18 <adrianschmutzler> Since you take it so seriously, and I'm not keen on spending my evening researching copyright law, I will set the patches back to "New" and leave this to somebody more familiar with the subject ...
May 18 21:06:02 <f00b4r0> well now I take it the wrong way. But thank you.
May 19 01:50:59 <mangix> lxc 4.0.2 somehow went in the packages feed without a PR.
May 19 01:51:47 <mangix> interesting as that's quite a major version jump. I'm sure the Turris people would have had feedback.
May 19 02:20:42 <owrt-snap-builds> build #114 of bcm63xx/smp is complete: Failure [failed images]  Build details are at http://buildbot.openwrt.org/master/images/builders/bcm63xx%2Fsmp/builds/114  blamelist: ?lvaro Fern?ndez Rojas <noltari@gmail.com>, David Bauer <mail@david-bauer.net>, Fabian Bl?se <fabian@blaese.de>, Adrian Schmutzler <freifunk@adrianschmutzler.de>
**** ENDING LOGGING AT Tue May 19 02:59:57 2020