**** BEGIN LOGGING AT Thu Apr 16 02:59:57 2020
Apr 16 03:00:21 <mirko> looks like the only way is to patch that hotplug functionality into `upgraded` which is run as PID 1 at this moment
Apr 16 06:57:38 <owrt-snap-builds> build #326 of armvirt/32 is complete: Failure [failed dirupload]  Build details are at http://buildbot.openwrt.org/master/images/builders/armvirt%2F32/builds/326  blamelist: Hannu Nyman <hannu.nyman@iki.fi>, Adrian Schmutzler <freifunk@adrianschmutzler.de>
Apr 16 07:41:36 <xback> Hauke: fyi, I've ran into the GCC 8.4.0 issue yesterday.  ath79 Mikrotik Mips target was caught in a bootloop.  switching to gcc 7.x solved it
Apr 16 07:47:18 <f00b4r0> aparcar[m]: please don't. There was an order of magnitude performance boost by using ccache the way it is currently used on phase1. The "comparison" has already been made.
Apr 16 07:47:34 <jow> I'd like to backport some ustream and uhttpd fixes to 19.07 without affecting the abi, so I need to back out the TLS cipher selection changes
Apr 16 07:47:50 <jow> shall I go with branches in ustream-ssl.git and uhttpd.git or shall I backport as patches?
Apr 16 07:48:23 <aparcar[m]> f00b4r0: It looks like it is still enabled, at least the MakeEnv(tryccache=True) is mostly activated
Apr 16 07:48:34 <f00b4r0> it is for native code
Apr 16 07:48:39 <f00b4r0> as I explained yesterday
Apr 16 07:48:40 <aparcar[m]> f00b4r0: I have no access to any relevant things, don't worry nothing will be changed :)
Apr 16 07:49:22 <aparcar[m]> dangole mentioned yesterday it would work fine with cross compiling... that's the only reason I wanted to give it a try
Apr 16 07:49:52 <f00b4r0> it does. It's useless and actually hurt performance.
Apr 16 07:50:34 <f00b4r0> bbiab
Apr 16 07:54:18 <aparcar[m]> jow: I created a trivial patch that only builds targets in buildbot when relevant files changed. Do you want a mail or is that feature deactivated on purpose?
Apr 16 07:56:58 <jow> aparcar[m]: you reenabled the existing feature?
Apr 16 07:57:25 <jow> please send a mail
Apr 16 07:57:37 <jow> blocked with video conferences for the next few hours
Apr 16 08:19:32 <owrt-snap-builds> build #332 of kirkwood/generic is complete: Failure [failed updatefeeds]  Build details are at http://buildbot.openwrt.org/master/images/builders/kirkwood%2Fgeneric/builds/332  blamelist: Hannu Nyman <hannu.nyman@iki.fi>, Adrian Schmutzler <freifunk@adrianschmutzler.de>
Apr 16 08:36:47 <ynezz> jow: I find it easier for the maintenance to have separate openwrt-19.07 branches
Apr 16 08:37:10 <ynezz> procd has already lede-17.01 and openwrt-19.07
Apr 16 08:37:18 <jow> allright
Apr 16 09:02:28 <xback> it looks like something is wrong with wpa_sup in latest 19.07.   I notice huge memleaks in the field, and a few sigsevs
Apr 16 09:25:01 <owrt-1907-builds> build #149 of malta/be is complete: Failure [failed updatefeeds]  Build details are at http://buildbot.openwrt.org/openwrt-19.07/images/builders/malta%2Fbe/builds/149  blamelist: Adrian Schmutzler <freifunk@adrianschmutzler.de>, Josef Schlehofer <pepe.schlehofer@gmail.com>
Apr 16 09:28:11 <owrt-1907-builds> build #147 of mpc85xx/generic is complete: Failure [failed updatefeeds]  Build details are at http://buildbot.openwrt.org/openwrt-19.07/images/builders/mpc85xx%2Fgeneric/builds/147  blamelist: Adrian Schmutzler <freifunk@adrianschmutzler.de>, Josef Schlehofer <pepe.schlehofer@gmail.com>
Apr 16 09:31:05 <owrt-1907-builds> build #144 of layerscape/armv8_64b is complete: Failure [failed updatefeeds]  Build details are at http://buildbot.openwrt.org/openwrt-19.07/images/builders/layerscape%2Farmv8_64b/builds/144  blamelist: Adrian Schmutzler <freifunk@adrianschmutzler.de>, Josef Schlehofer <pepe.schlehofer@gmail.com>
Apr 16 09:32:25 <Tapper> I just spotted this on a forum.
Apr 16 09:32:27 <Tapper> If you are an open source software developer, you can reply this thread send me PM or to apply for a free sample. If you want to evaluate ROCK Pi E for your project, you can
Apr 16 09:32:34 <Tapper> https://forum.radxa.com/t/rock-pi-e-engineering-sample-is-available-now/3130
Apr 16 09:42:40 <Borromini> i think someone just made a PR for rockchip support on openwrt
Apr 16 09:43:46 <Borromini> https://github.com/openwrt/openwrt/pull/2917
Apr 16 10:08:08 <mangix> free sample is nice
Apr 16 10:08:43 <mangix> wow. usb c for power
Apr 16 10:12:20 <aparcar[m]> mangix: free sample?
Apr 16 10:15:08 <shalzz> Hey, is sqm disable for targets with linux 5.4 or is it completely removed?
Apr 16 10:17:20 <mangix> aparcar[m]: follow Tapper's link
Apr 16 10:19:49 <mangix> shalzz: news to me. Unless you mean cake.
Apr 16 10:20:43 <shalzz> mangix I'm talking about the sqm-scripts and the luci-app-sqm
Apr 16 10:24:52 <aparcar[m]> mangix: thanks
Apr 16 10:25:55 <aparcar[m]> jow: patch sent
Apr 16 10:36:13 <mangix> shalzz: it is not removed
Apr 16 10:49:27 <shalzz> mangix: it doesn't show up in the menuconfif because of depends=!LINUX_4.4
Apr 16 10:53:49 <mangix> Sounds like a bug.
Apr 16 10:53:51 <shalzz> when builds for a mt7620 target
Apr 16 10:53:58 <shalzz> building*
Apr 16 10:54:10 <shalzz> > Sounds like a bug.
Apr 16 10:54:36 <shalzz> Yes so my question was is it intentional or accidental
Apr 16 10:55:40 * shalzz sent a long message:  < https://matrix.org/_matrix/media/r0/download/matrix.org/sOwuQiFfnpqrynqksbCrRyVh >
Apr 16 10:55:45 <mangix> accidental
Apr 16 10:56:17 <mangix> this was probably places when 4.14 was the latest kernel
Apr 16 10:56:33 <mangix> *placed
Apr 16 11:02:38 <owrt-snap-builds> build #68 of bcm47xx/generic is complete: Failure [failed updatefeeds]  Build details are at http://buildbot.openwrt.org/master/images/builders/bcm47xx%2Fgeneric/builds/68  blamelist: Hannu Nyman <hannu.nyman@iki.fi>
Apr 16 11:09:18 <owrt-snap-builds> build #67 of bcm47xx/legacy is complete: Failure [failed updatefeeds]  Build details are at http://buildbot.openwrt.org/master/images/builders/bcm47xx%2Flegacy/builds/67  blamelist: Hannu Nyman <hannu.nyman@iki.fi>
Apr 16 13:25:39 <xback> greearb: fyi, I've just re-tested ath10k-ct (using 5.4 now) in IBSS-RSN.   same issue still exists
Apr 16 13:26:03 <xback> a simple ping works, but as soon as aggregated packets are made .. the link stalls
Apr 16 13:26:26 <xback> it doesn't happen with unencrypted data
Apr 16 13:33:39 <greearb> xback, this is wave-1 radio?
Apr 16 13:33:56 <xback> greearb: yes
Apr 16 13:34:25 <xback> qca988x hw2.0
Apr 16 13:34:51 <xback> i've just reflashed clean images with ath debugging in
Apr 16 13:35:07 <greearb> yeah, I never spent the time to dig into that problem in depth.
Apr 16 13:36:21 <xback> greearb: does your gut feeling tell you if it's a driver or a fw issue?
Apr 16 13:37:00 <greearb> very likely firmware
Apr 16 13:37:40 <xback> greearb: if you could provide a dbg falvour of the latest fw, I'm more than happy to provide a dump
Apr 16 13:37:54 <xback> start dbg,  do a ping,  then start an iperf
Apr 16 13:38:30 <greearb> for this sort of thing, I have to reproduce locally and debug in depth with sniffers and so forth.  I just need a day or two clear combined with motivation to work on a feature I never use :)
Apr 16 13:38:43 <greearb> that combination is rare
Apr 16 13:39:27 <xback> wave1 IBSS-RSN ?
Apr 16 13:39:39 <greearb> yes
Apr 16 13:39:47 <xback> i always wondered how eg. UBNT or other vendors make their PTP stuff
Apr 16 13:39:56 <xback> it seems they just do AP-STA
Apr 16 13:40:29 <greearb> yep, I think someone should make meshing AP to STA easy, using multiple STA vdevs per radio and one AP per radio
Apr 16 13:41:10 <xback> greearb: would it help if I mail you my currently used config?
Apr 16 13:41:21 <xback> just copy it to both DUT's and start wpa_sup
Apr 16 13:41:55 <greearb> no, I don't even run OpenWrt except rarely, and I think I understand how to reproduce
Apr 16 13:42:19 <xback> I mean a native wpa_sup config.   not a OpenWrt one
Apr 16 13:42:51 <greearb> I can easily set it up myself, our LANforge network testing system has all of the support needed to do so
Apr 16 13:44:01 <xback> cool :-)
Apr 16 14:01:17 <owrt-snap-builds> build #363 of lantiq/ase is complete: Failure [failed updatefeeds]  Build details are at http://buildbot.openwrt.org/master/images/builders/lantiq%2Fase/builds/363  blamelist: Hannu Nyman <hannu.nyman@iki.fi>, Koen Vandeputte <koen.vandeputte@ncentric.com>, Roger Pueyo Centelles <roger.pueyo@guifi.net>
Apr 16 14:03:33 <owrt-snap-builds> build #364 of at91/sam9x is complete: Failure [failed]  Build details are at http://buildbot.openwrt.org/master/images/builders/at91%2Fsam9x/builds/364  blamelist: Hannu Nyman <hannu.nyman@iki.fi>, Koen Vandeputte <koen.vandeputte@ncentric.com>, Roger Pueyo Centelles <roger.pueyo@guifi.net>
Apr 16 14:08:24 <owrt-snap-builds> build #281 of octeon/generic is complete: Failure [failed updatefeeds]  Build details are at http://buildbot.openwrt.org/master/images/builders/octeon%2Fgeneric/builds/281  blamelist: Hannu Nyman <hannu.nyman@iki.fi>, Koen Vandeputte <koen.vandeputte@ncentric.com>, Roger Pueyo Centelles <roger.pueyo@guifi.net>
Apr 16 14:42:37 <owrt-1907-builds> build #142 of mediatek/mt7622 is complete: Failure [failed updatefeeds]  Build details are at http://buildbot.openwrt.org/openwrt-19.07/images/builders/mediatek%2Fmt7622/builds/142  blamelist: Adrian Schmutzler <freifunk@adrianschmutzler.de>, Josef Schlehofer <pepe.schlehofer@gmail.com>, Koen Vandeputte <koen.vandeputte@ncentric.com>
Apr 16 14:46:28 <owrt-1907-builds> build #144 of mediatek/mt7623 is complete: Failure [failed updatefeeds]  Build details are at http://buildbot.openwrt.org/openwrt-19.07/images/builders/mediatek%2Fmt7623/builds/144  blamelist: Adrian Schmutzler <freifunk@adrianschmutzler.de>, Josef Schlehofer <pepe.schlehofer@gmail.com>, Koen Vandeputte <koen.vandeputte@ncentric.com>
Apr 16 14:49:47 <owrt-1907-builds> build #141 of cns3xxx/generic is complete: Failure [failed updatefeeds]  Build details are at http://buildbot.openwrt.org/openwrt-19.07/images/builders/cns3xxx%2Fgeneric/builds/141  blamelist: Adrian Schmutzler <freifunk@adrianschmutzler.de>, Josef Schlehofer <pepe.schlehofer@gmail.com>, Koen Vandeputte <koen.vandeputte@ncentric.com>
Apr 16 14:54:54 <owrt-1907-builds> build #141 of mvebu/cortexa9 is complete: Failure [failed]  Build details are at http://buildbot.openwrt.org/openwrt-19.07/images/builders/mvebu%2Fcortexa9/builds/141  blamelist: Adrian Schmutzler <freifunk@adrianschmutzler.de>, Josef Schlehofer <pepe.schlehofer@gmail.com>, Koen Vandeputte <koen.vandeputte@ncentric.com>
Apr 16 14:58:05 <owrt-1907-builds> build #141 of mvebu/cortexa53 is complete: Failure [failed updatefeeds]  Build details are at http://buildbot.openwrt.org/openwrt-19.07/images/builders/mvebu%2Fcortexa53/builds/141  blamelist: Adrian Schmutzler <freifunk@adrianschmutzler.de>, Josef Schlehofer <pepe.schlehofer@gmail.com>, Koen Vandeputte <koen.vandeputte@ncentric.com>
Apr 16 15:01:06 <owrt-1907-builds> build #142 of archs38/generic is complete: Failure [failed updatefeeds]  Build details are at http://buildbot.openwrt.org/openwrt-19.07/images/builders/archs38%2Fgeneric/builds/142  blamelist: Adrian Schmutzler <freifunk@adrianschmutzler.de>, Josef Schlehofer <pepe.schlehofer@gmail.com>, Koen Vandeputte <koen.vandeputte@ncentric.com>
Apr 16 15:36:32 <owrt-snap-builds> build #263 of ramips/mt7621 is complete: Failure [failed fetchrefs]  Build details are at http://buildbot.openwrt.org/master/images/builders/ramips%2Fmt7621/builds/263  blamelist: Hannu Nyman <hannu.nyman@iki.fi>, Koen Vandeputte <koen.vandeputte@ncentric.com>, Roger Pueyo Centelles <roger.pueyo@guifi.net>
Apr 16 15:45:36 <rmilecki> jow: https://forum.openwrt.org/t/loving-the-luci-improvements-on-mobile-tablets-with-19-07/60758 :)
Apr 16 15:55:51 <ynezz> mangix: Hauke: FYI elfutils-0.179: aarch64_initreg.c:85:37: error: invalid operands to binary & (have 'long double' and 'unsigned int')
Apr 16 15:56:36 <ynezz> build testing PR#2917
Apr 16 16:01:38 <tmn505> Hauke: yes please backport
Apr 16 16:04:12 <tmn505> Hauke: Re ebin and kernel 5.4, the USB 3 port and SATA do not work. Either we would need to: 1. provide new U-Boot, 2. point to Armbian U-Boot or 3. patch dts.
Apr 16 16:10:17 <owrt-1907-builds> build #142 of x86/geode is complete: Failure [failed dirupload]  Build details are at http://buildbot.openwrt.org/openwrt-19.07/images/builders/x86%2Fgeode/builds/142  blamelist: Adrian Schmutzler <freifunk@adrianschmutzler.de>, Josef Schlehofer <pepe.schlehofer@gmail.com>, Koen Vandeputte <koen.vandeputte@ncentric.com>
Apr 16 16:35:31 <stintel> the inside of that AP7060DN looks interesting too
Apr 16 16:46:38 <stintel> cool, can connect clamp to flash chip
Apr 16 16:59:18 <stintel> but not compatible with flashrom
Apr 16 17:01:06 <owrt-snap-builds> build #276 of x86/64 is complete: Failure [failed updatefeeds]  Build details are at http://buildbot.openwrt.org/master/images/builders/x86%2F64/builds/276  blamelist: Hannu Nyman <hannu.nyman@iki.fi>, Koen Vandeputte <koen.vandeputte@ncentric.com>, Roger Pueyo Centelles <roger.pueyo@guifi.net>
Apr 16 17:01:20 <PaulFertser> Just a matter of adding an ID?
Apr 16 17:12:19 <stintel> hopefully it's that simple
Apr 16 17:13:06 <stintel> let's first look into bumping flashrom again
Apr 16 17:13:25 <stintel> it's lagging behind at 1.0, 1.2 is out
Apr 16 17:13:48 <stintel> my attempt to bump to 1.1 failed because one of the variants didn't compile
Apr 16 17:14:01 <stintel> let's hope that was magically fixed in 1.2
Apr 16 18:14:49 <owrt-1907-builds> build #137 of ath25/generic is complete: Failure [failed updatefeeds]  Build details are at http://buildbot.openwrt.org/openwrt-19.07/images/builders/ath25%2Fgeneric/builds/137  blamelist: Adrian Schmutzler <freifunk@adrianschmutzler.de>, Josef Schlehofer <pepe.schlehofer@gmail.com>, Koen Vandeputte <koen.vandeputte@ncentric.com>
Apr 16 18:15:40 <stintel> sol, problem is still there
Apr 16 18:21:55 <Hauke> tmn505: I wanted to build an u-boot for the ebin in OpenWrt
Apr 16 18:22:03 <Hauke> but haven't started with that
Apr 16 18:22:32 <Hauke> what would we need to patch in the dts?
Apr 16 18:22:47 <Hauke> would it work to patch the dts and make it work with old an new u-boot?
Apr 16 18:27:37 <tmn505> Hauke: deleting these additions 8e18c8e58da644018a6217dac5eaa0af8d0f390c bd3d25b0734284dcf603173d54232c538eb6e385 should be enough, but didn't test that yet.
Apr 16 18:29:22 <tmn505> I don't know if patching for both would work. But AFAIK there is no newer factory U-Boot on the boards than what we both have.
Apr 16 18:41:57 <stintel> there seem to be a bunch of undefined kernel symbols for bcm27xx
Apr 16 18:42:06 <stintel> no time for yah shaving right now
Apr 16 19:15:55 <mangix>  ynezz: upstream has been notified.
Apr 16 19:30:20 <stintel> how do you even see what variant failed to build :/.
Apr 16 19:34:20 <stintel> finally
Apr 16 19:36:46 <stintel> that's what you get with undocumented Makefile voodoo and black magic
Apr 16 19:39:00 <stintel> and of course the next variant needs fixing too
Apr 16 19:45:23 <stintel> and then github died
Apr 16 19:46:08 <tmn505> Hauke: this patch works: http://ix.io/2ik7
Apr 16 19:46:58 <tmn505> for both SATA and USB
Apr 16 20:03:19 <blocktrron> Hauke: i've pulled the mac80211 TXQ patch you've mentioned yesterday to my staging tree - so far it didn't seem to break anything
Apr 16 20:06:03 <Hauke> blocktrron: nice
Apr 16 20:07:44 <owrt-snap-builds> build #69 of bcm27xx/bcm2711 is complete: Failure [failed updatefeeds]  Build details are at http://buildbot.openwrt.org/master/images/builders/bcm27xx%2Fbcm2711/builds/69  blamelist: Koen Vandeputte <koen.vandeputte@ncentric.com>, Roger Pueyo Centelles <roger.pueyo@guifi.net>
Apr 16 20:29:02 <stintel> pffft what a pos package
Apr 16 21:09:49 <owrt-snap-builds> build #68 of bcm63xx/smp is complete: Failure [failed updatefeeds]  Build details are at http://buildbot.openwrt.org/master/images/builders/bcm63xx%2Fsmp/builds/68  blamelist: Koen Vandeputte <koen.vandeputte@ncentric.com>, Roger Pueyo Centelles <roger.pueyo@guifi.net>
Apr 16 21:13:59 <owrt-snap-builds> build #360 of ipq806x/generic is complete: Failure [failed updatefeeds]  Build details are at http://buildbot.openwrt.org/master/images/builders/ipq806x%2Fgeneric/builds/360  blamelist: Koen Vandeputte <koen.vandeputte@ncentric.com>, Roger Pueyo Centelles <roger.pueyo@guifi.net>
Apr 16 21:32:47 <owrt-snap-builds> build #68 of bcm47xx/legacy is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org/master/images/builders/bcm47xx%2Flegacy/builds/68
Apr 16 21:38:01 <stintel> yay Found Macronix flash chip "MX25U25635F" (32768 kB, SPI) on linux_spi.
Apr 16 21:38:15 <stintel> new flashrom seems to detect it
Apr 16 22:22:43 <mangix> stintel: speaking of new flashrom :)
Apr 16 22:24:31 <stintel> oh don't remind me of that pos package
Apr 16 22:24:34 <stintel> just merge the PR :P
Apr 16 22:24:42 <mangix> what PR?
Apr 16 22:25:55 <stintel> https://github.com/openwrt/packages/pull/10578
Apr 16 22:26:49 <mangix> oh. didn't realize you had a PR.
Apr 16 22:27:55 <stintel> 16|19:13:06 < stintel> let's first look into bumping flashrom again
Apr 16 22:30:24 <stintel> wow ipq807x needs some love. the amount of missing kernel symbols is quite high
Apr 16 22:30:43 <stintel> Press f or F  to stop Auto-Boot in 3 seconds:  2   1 0
Apr 16 22:30:44 <stintel> Password for uboot cmd line :
Apr 16 22:30:47 <stintel> fffffuuuuuu
Apr 16 22:30:51 <stintel> that would have been too easy :P
Apr 16 22:35:15 <blocktrron> stintel: which board uses this "feature"
Apr 16 22:38:02 <stintel> ap7060dn
Apr 16 22:42:47 <stintel> finding things like get_pbkdf2_password with binwalk
Apr 16 22:43:08 <stintel> that's going to be a bitch to find
Apr 16 22:44:24 <stintel> heh! it's the same as the default password in the CLI after boot
Apr 16 22:44:38 <stintel> > Info: You are advised to change the password to ensure security.
Apr 16 22:44:46 <stintel> found this as well via binwalk
Apr 16 22:45:04 <stintel> which prompted me to try the default password :)
Apr 16 22:48:21 <stintel> it seems to be quite crippled though
Apr 16 22:48:40 <stintel> no tftpboot for example
Apr 16 22:48:44 <Hauke> is the a not so expensive ipq807x board out there?
Apr 16 22:49:20 <stintel> no idea, this wasn't cheap (EUR540 ex VAT)
Apr 16 22:50:52 <Hauke> saw the Xiaomi AX3600
Apr 16 22:56:34 <Hauke> https://de.aliexpress.com/item/4000785767529.html
Apr 16 22:58:36 <stintel> heh that's cheap
Apr 16 22:59:54 <Hauke> this is the normal price for this device
Apr 16 23:00:12 <Hauke> but still
Apr 16 23:00:17 <Hauke> looks nice
Apr 16 23:01:30 <Hauke> stintel: https://forum.openwrt.org/t/xiaomi-s-upcoming-ax3600-wi-fi-6-router-is-based-on-qualcomm-s-enterprise-grade-networking-platform/55049
Apr 16 23:04:19 <stintel> oh, but it can do tftp!
Apr 16 23:04:47 <stintel> but I'm getting a `Check Software fail` which was to be expected :)
Apr 16 23:05:12 <stintel> and it looks like this thing has a NOR flash for u-boot but the OS is on NAND
Apr 17 00:09:12 <aparcar[m]> jow: do you use the same shared-workdir over multiple branches? like master and openwrt-19.07 or do you setup a different infrastructure? Wondering how to set it up for the staging builds
Apr 17 00:24:45 * mangix wonders why git versions of glibc are used
Apr 17 02:28:23 <Namidairo> they lock down their u-boot to only take their signed images by default
Apr 17 02:29:48 <Namidairo> but you should be able to load images on that if you manage to overwrite their boot_delay in uboot env
Apr 17 02:33:19 <Namidairo> oh you're already in, derp reading.
Apr 17 02:51:31 <owrt-snap-builds> build #327 of armvirt/32 is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org/master/images/builders/armvirt%2F32/builds/327
**** ENDING LOGGING AT Fri Apr 17 02:59:57 2020