**** BEGIN LOGGING AT Fri Apr 10 02:59:57 2020
Apr 10 03:05:06 <mangix> it is. but you're converting to bool and back to int with +.
Apr 10 03:05:22 <mangix> as an example, 3 ~= 1
Apr 10 03:05:33 <mangix> !=
Apr 10 03:05:58 <mangix> I think you meant to use ||
Apr 10 03:06:31 <mangix>  in which case, !! is not needed
Apr 10 03:22:21 <mangix> jow: it seems that I cannot boot malta with 5.4. I have a feeling it
Apr 10 03:23:33 <mangix> 's due to https://patchwork.ozlabs.org/patch/1261133/ . It fails  right after xt_time: kernel timezone is -0000 .
Apr 10 03:27:29 <mangix> maybe this needs backporting...: https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/net/netfilter/xt_time.c?h=next-20200409&id=fcbad8293d52864d87d0b9f6035fd87a049d59d8
Apr 10 03:32:17 <gch981213> mangix: never tried wireguard
Apr 10 03:33:51 <gch981213> mangix: spi clock is calculated by dividing bus clock with a divider register. Bus clock is CPU_clock/4 which is 220MHz so the nearest one to 80MHz is 220/3=73.33
Apr 10 03:37:08 * mangix gets calculator
Apr 10 03:37:55 <mangix> 225/3 = 75
Apr 10 03:37:57 <mangix> hmmmmm
Apr 10 03:38:48 <mangix> last i tested this, I tested 80 and 66 but not in between
Apr 10 03:41:09 <aparcar[m]> mangix: I might be mistaken but it's not bool but a filename, so one ! makes it a bool and the second negate it. Maybe I'm wrong, really low C experience and I just took what was used in usign
Apr 10 03:42:02 <gch981213> mangix: Oh. I remember you've sent mails about this and wondering why clocks higher than 80MHz doesn't change anything. The clock higher than 220/3=73.33MHz is 220/2=110MHz which is too fast for spi flash.
Apr 10 03:46:02 <mangix> gch981213: are you saying underclocking such that the SPI clock becomes 104MHz would work?
Apr 10 03:47:10 <gch981213> mangix: I think it doesn't really worth to underclock for this purpose
Apr 10 03:47:23 <mangix> no, not at all.
Apr 10 03:49:14 <mangix> AGH. malta is still hangingh on 5.4. time to try glibc. Although I have a feeling that won't do anything...
Apr 10 03:49:45 <mangix> oh perfect, CPU stall
Apr 10 03:54:07 <mangix> regulatory_set_wiphy_regd_sync_rtnl suggests a wifi problem...
Apr 10 04:40:46 <mangix> glibc works
Apr 10 04:40:47 <mangix> sigh
Apr 10 06:04:01 <rmilecki> when I was still studying i spent few hours playing with ipsec, later i was reading about openvpn a bit
Apr 10 06:04:12 <rmilecki> i decided i'll anything not to use vpn ever :P
Apr 10 06:04:33 <rmilecki> yesterday I setup wireguard - a big wow :)
Apr 10 06:05:21 <rmilecki> OpenWrt integration seems still somehow broken - I was using LuCI and changes were not applied immediately - i decided to reboot
Apr 10 06:05:27 <rmilecki> otherwise it's great
Apr 10 06:31:30 <mangix>  rmilecki: using it with DSA?
Apr 10 06:55:54 <rmilecki> mangix: no, swconfig
Apr 10 06:56:11 <rmilecki> if that's what you meant by DSA
Apr 10 07:20:14 <ynezz> stintel: pita? I'm mostly running the stuff from Docker containers which boils down to `wget Makefile && make register run` I did the same on Hetzner and GCP just a few days ago
Apr 10 07:20:37 <ynezz> the runner setup is like 2 minutes including the Docker install
Apr 10 07:21:28 <ynezz> https://gist.githubusercontent.com/ynezz/21de20aa906ba7ab7e6113d4b151146d/raw/9772537f79db5c894e3b4cdf0b53f04ad135c3c9/Makefile
Apr 10 07:22:43 <ynezz> I admit, that I've never used gitlab-runner running as a service directly on the host, not from the container
Apr 10 07:22:51 <stintel> ynezz: I can't use the Docker as I need to be able to run qemu
Apr 10 07:23:18 <stintel> anyway I'll just keep hitting it until it works :)
Apr 10 07:23:21 <ynezz> qemu doesnt run in docker?
Apr 10 07:25:23 <stintel> well the test framework I'm trying to automate first creates a VM, this tries to mount an image file which isn't allowed, etc
Apr 10 07:26:23 <stintel> documentation isn't great, weird bugs, runner hanging without any indicator why
Apr 10 07:26:29 <stintel> just frustrating to work with
Apr 10 07:40:21 <aparcar[m]> ynezz: I created your signify patch just in case you miss the github notification
Apr 10 10:20:53 <Rene__> dengqf6: Are you working on the ramips DSA and VLAN issue?
Apr 10 10:36:26 <owrt-snap-builds> build #260 of layerscape/armv7 is complete: Failure [failed images]  Build details are at http://buildbot.openwrt.org/master/images/builders/layerscape%2Farmv7/builds/260  blamelist: ?? <uxgood.org@gmail.com>, Bj?rn Mork <bjorn@mork.no>, Florian Eckert <fe@dev.tdt.de>, Petr ?tetiar <ynezz@true.cz>, Linus Walleij <linus.walleij@linaro.org>, Daniel Golle
Apr 10 10:36:26 <owrt-snap-builds> <daniel@makrotopia.org>, Ansuel Smith <ansuelsmth@gmail.com>, Norbert van Bolhuis <nvbolhuis@aimvalley.nl>, Alexey Dobrovolsky <dobrovolskiy.alexey@gmail.com>, Stefan Lippers-Hollmann <s.l-h@gmx.de>, Felix Fietkau <nbd@nbd.name>, Daniel Engberg <daniel.engberg.lists@pyret.net>, Eneas U de Queiroz <cotequeiroz@gmail.com>, Roman Hampel <rhamp@arcor.de>, Pawel
Apr 10 10:36:26 <owrt-snap-builds> Dembicki <paweldembicki@gmail.com>
Apr 10 10:41:00 <mwarning> current openwrt master gives me "undefined reference to `sym_is_changable'". I build the x86 target. Is that a known issue?
Apr 10 10:54:12 <ldir> mwarning: try make -C scripts/config clean
Apr 10 10:58:50 <mwarning> ldir: thanks, that did it.
Apr 10 11:00:26 <ldir> a make dirclean would have solved it too, but the above is much more accurately aimed :-)
Apr 10 11:03:12 <mwarning> hm, I thought I had a fresh checkout..
Apr 10 11:03:21 <mwarning> *clone
Apr 10 11:04:46 <ldir> that's odd then.
Apr 10 11:07:36 <Borromini> ldir: is there a tell for when one should do scripts/config clean?
Apr 10 11:10:08 <mwarning> ldir: it is no impossible that that copy had some fragments
Apr 10 11:12:32 <ldir> Borromini: not really.  config aka make menuconfig/defconfig & friends is the thing that generates the menu based on available packages.  The things that get changed in the config are 'symbols' - 'sym_is_changable' sounded like a symbol related thing to me, and I knew that config had been bumped recently
Apr 10 11:12:59 <Borromini> ldir: ok, thanks
Apr 10 11:13:11 <ldir> Borromini: so it was an intuitive guess that maybe there were old bits of kconfig lying around and it needed to be re-built.
Apr 10 11:14:19 <ldir> scripts/config tends to 'hang around' and not get re-built as/when it should - that's why recently I put in a step to make dirclean to also clean up scripts/config
Apr 10 11:14:42 <ldir> so hopefully those sorts of oddities would be reduced.
Apr 10 11:15:05 <Borromini> :)
Apr 10 11:15:11 <Borromini> cool, thanks
Apr 10 11:27:31 * ldir shows his age in the latest commit
Apr 10 11:29:45 <Borromini> :D
Apr 10 11:30:10 <Borromini> KRS-One <3
Apr 10 11:30:19 <ldir> yarp :-)
Apr 10 11:32:39 <Borromini> nice. Signal app might leave the US if that new encryption law passes.
Apr 10 11:32:57 <stintel> many sane people might too
Apr 10 11:34:03 <Borromini> stintel: we all know how well that 'if you guys vote in a republican president i'm off to canada' worked out :)
Apr 10 11:34:21 <Borromini> hollywood hasn't relocated to canada, and they already promised twice
Apr 10 11:52:59 <f00b4r0> Borromini: which new encryption law?
Apr 10 11:53:41 <Borromini> titled 'EARN IT' act
Apr 10 11:53:55 <Borromini> https://it.slashdot.org/story/20/04/09/2014207/signal-threatens-to-dump-us-market-if-earn-it-act-passes
Apr 10 11:54:28 <f00b4r0> ah that. I remember reading about it on El Reg
Apr 10 11:55:07 <Borromini> el reg?
Apr 10 11:55:16 <f00b4r0> theregister :)
Apr 10 11:55:45 <f00b4r0> i thought the bill was mostly technically inapplicable though
Apr 10 11:56:17 <f00b4r0> Borromini: in fact they have a piece on exactly what you pasted: https://www.theregister.co.uk/2020/04/09/signal_earn_it/
Apr 10 11:56:22 <f00b4r0> I'm just lagging behind news ;)
Apr 10 12:00:26 <Borromini> :)
Apr 10 15:32:54 <owrt-1907-builds> build #118 of sunxi/cortexa7 is complete: Failure [failed pkgbuild]  Build details are at http://buildbot.openwrt.org/openwrt-19.07/images/builders/sunxi%2Fcortexa7/builds/118  blamelist: Rafa? Mi?ecki <rafal@milecki.pl>, Dan Haab <dan.haab@legrand.com>
Apr 10 17:47:00 <jow> gch981213: ah okay. But that missing vlan tag on outgoing frames is a bug?
Apr 10 17:47:21 <jow> gch981213: or is it expected that you need bridge filtering for dsa to properly handle vlans?
Apr 10 20:30:19 <owrt-snap-builds> build #315 of layerscape/armv8_64b is complete: Failure [failed images]  Build details are at http://buildbot.openwrt.org/master/images/builders/layerscape%2Farmv8_64b/builds/315  blamelist: ?? <uxgood.org@gmail.com>, DENG Qingfang <dengqf6@mail2.sysu.edu.cn>, Daniel Golle <daniel@makrotopia.org>, Christian Lamparter <chunkeey@gmail.com>, Kevin
Apr 10 20:30:19 <owrt-snap-builds> Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>, Felix Fietkau <nbd@nbd.name>, Daniel Engberg <daniel.engberg.lists@pyret.net>, Eneas U de Queiroz <cotequeiroz@gmail.com>, Stefan Lippers-Hollmann <s.l-h@gmx.de>, Pawel Dembicki <paweldembicki@gmail.com>
Apr 10 20:30:21 <aparcar[m]> jow: json patch :S?
Apr 10 23:00:43 <russell--> hey, does anyone have a sense for what people are using these days for openwrt audio players?  that is, render audio through an openwrt-local audio device from a network source?
Apr 10 23:24:53 <pkgadd> russell--: if you find a good solution, tell me ;)  I'd assume that in many cases the answer will be something like a RPi-0W with USB soundcard + small amp - and not necessarily running OpenWrt (my main grasp with any solution I've seen so far was the input side, as in on/off vol +/-)
Apr 10 23:26:29 <pkgadd> russell--: this also has some charme, https://olimex.wordpress.com/2019/07/30/esp32-adf-internet-radio-example-now-uses-littlevgl-with-2-8-tft-lcd-and-touch-screen/
Apr 10 23:34:42 <russell--> pkgadd: i'm looking for the software part of an an openwrt-based network-connected speaker. the controls will be external.
**** ENDING LOGGING AT Sat Apr 11 02:59:57 2020