**** BEGIN LOGGING AT Sat Mar 07 03:03:04 2020
Mar 07 09:57:07 <jow> ynezz: thanks for the forum announcement
Mar 07 10:15:38 <ynezz> I'm probably too dumb to use dokuwiki, how could I find that page about release steps?
Mar 07 10:29:51 <aparcar[m]> any wiki mods around at irc?
Mar 07 10:31:46 <aparcar[m]> ynezz: I renewed the x86 image creation code. Should I now fix imx6 ;)?
Mar 07 10:36:55 <Tapper> Hi people hows the bump to k5.4 going for mvebu?
Mar 07 10:37:21 <Tapper> I am having a boring morning and can do some testing.
Mar 07 10:37:31 <Hauke> Tapper: scott is working on something here: https://github.com/TELUS-BBA/openwrt-1
Mar 07 10:37:47 <Tapper> Hauke thanks will look
Mar 07 10:39:07 <Tapper> O there is a open pr https://github.com/openwrt/openwrt/pull/2804
Mar 07 10:39:54 <Tapper> Not ready for me to play with yet.
Mar 07 11:11:54 <kab-el> Tapper: what board are you testing on mvebu?
Mar 07 11:12:30 <Tapper> Hi kab-el I have a wrt1900ac-v2 and a wrt3200acm.
Mar 07 11:12:54 <Tapper> K5.4 is not ready to test yet.
Mar 07 11:13:07 <hsp> ar71xx image-builder not work
Mar 07 11:13:38 <Tapper> Rite now the kernel is getting to big to fit in the partitions set aside for it.
Mar 07 11:13:54 <kab-el> hm
Mar 07 11:14:09 <hsp> Profile "archer-c7-v2" does not exist!
Mar 07 11:14:10 <kab-el> i have turris omnia and turris mox, can maybe test on those, but would need to do additional patching
Mar 07 11:14:13 <Tapper> I don't know enuff to mess around with it in it's current state.
Mar 07 11:16:27 <hsp> ^^ 19.07.2
Mar 07 11:18:05 <kab-el> Tapper: the device has sata port, do you have a drive connected?
Mar 07 11:18:26 <kab-el> Tapper: doesn't u-boot support booting from hdd, or usb?
Mar 07 11:18:29 <Tapper> No mate no E SATA or USB
Mar 07 11:18:52 <kab-el> no usb port?
Mar 07 11:19:01 <Tapper> kab-el I don't know. I have never booted it that way.
Mar 07 11:19:02 <kab-el> USB ports:1x 2.0, 1x 3.0
Mar 07 11:19:28 <Tapper> kab-el yes I have USB ports but no drives plugged in.
Mar 07 11:19:49 <kab-el> hmm, upstream u-boot does not contain device tree for that device
Mar 07 11:20:08 <kab-el> well, you can try porting u-boot ;)
Mar 07 11:45:40 <RobertP> hello - I am an OpenWRT beginner, but I am trying to add support for a new device I own (ASUS RT-AC5300). I've made some progress thanks to many suggestions, but now I am stuck at flashing a custom image stage.
Mar 07 11:45:56 <RobertP> image boots but I got something like this:
Mar 07 11:46:07 <RobertP> [    0.322055] nand: 128 MiB, SLC, erase size: 128 KiB, page size: 2048, OOB size: 64
Mar 07 11:46:11 <RobertP> [    0.329735] iproc_nand 18028000.nand: detected 128MiB total, 128KiB blocks, 2KiB pages, 16B OOB, 8-bit, BCH-8 (1KiB sector)
Mar 07 11:46:36 <RobertP> [    0.340907] Scanning device for bad blocks
Mar 07 11:46:50 <RobertP> [    1.393128] mtd_read error while parsing (offset: 0x0): -74
Mar 07 11:47:42 <RobertP> could anyone give me a hand and school me a bit how to fix it? - any help would be aprecueated :)
Mar 07 12:17:44 <PaulFertser> RobertP: hey :) So now you get error right from the very first block, 0x0 ? It was different before.
Mar 07 12:18:02 <PaulFertser> RobertP: also, are you able to boot an initramfs image? It doesn't need flash so you can check everything else.
Mar 07 12:19:47 <RobertP> PaulFertser: I think I missed this information previously from programming and flashing ... previous logs were from next boot after flashing ... (sorry - still learning :D)
Mar 07 12:21:43 <RobertP> PaulFertser: am I right that OOB mismatch is kind of important? :)
Mar 07 12:22:38 <PaulFertser> RobertP: OOB data is important indeed
Mar 07 12:22:57 <KanjiMonster> RobertP: OOB is where each block's ECC checksum is stored (or the bad block marker)
Mar 07 12:24:54 <RobertP> so if i understand logs it correctly - my device has OOB: 64, while generated images have OOB 16, right?
Mar 07 12:25:13 <PaulFertser> Sometimes bad block lists are stored in BadBlockTable.
Mar 07 12:25:49 <PaulFertser> RobertP: depends on the flashing method. In general you're not supposed to embed any OOB at all unless you flash raw images to the specific device.
Mar 07 12:26:01 <PaulFertser> RobertP: because you can't tell where bad blocks are in advance
Mar 07 12:26:18 <PaulFertser> RobertP: and the target flashing code should take care of that.
Mar 07 12:36:41 <RobertP> tanks Paul - looks like I still have to understand it better :)
Mar 07 12:57:02 <PaulFertser> RobertP: NAND devices 1. Come with some bad blocks marked from the factory; 2. During their lifetime some additional bad blocks appear and get marked. Usually bad block data is part of OOB.
Mar 07 12:58:07 <PaulFertser> RobertP: all of the software involved need to be on the same page with regard to OOB layout and usage.
Mar 07 12:59:21 <RobertP> that's what I thought - thanks ... now I need to figure it out why I got this mismatch in OOB (16 vs 64)
Mar 07 14:08:42 <mamarley> Seriously?  I go to all the trouble to open up my UAP-LR and attach serial to figure out why 5.4 bricks it.  I install an image with 5.4, but it doesn't brick!  WTF‽
Mar 07 14:09:42 <mamarley> Ah, but the Ethernet interface doesn't come up.  That's not good…
Mar 07 14:10:19 <mamarley> I get "ag71xx: probe of 19000000.eth failed with error -2".
Mar 07 14:14:07 <mamarley> Hauke: ^Would you be the right person to tell about this?
Mar 07 14:35:17 <mamarley> I would have to assume that it has something to do with switching to the upstream ag71xx driver in 5.4.
Mar 07 16:09:03 <lynxis> Kunsi: about edgerouter pro, which image do you used?
Mar 07 16:09:43 <Kunsi> openwrt-19.07.1-octeon-er-initramfs-kernel.bin  openwrt-19.07.1-octeon-er-squashfs-sysupgrade.tar
Mar 07 16:10:07 <Kunsi> boot kernel via usb, then use sysupgrade to install to flash
Mar 07 16:11:22 <Kunsi> Basically followed the instructions of a friend: https://blog.mazdermind.de/2019/03/26/openwrt-on-ubiquity-edgerouter-pro/index.html
Mar 07 16:17:04 <lynxis> Kunsi: i'ven't used SFP module under OpenWrt. I don't know what's missing there. Can you put a dmesg somewhere?
Mar 07 16:18:32 <Kunsi> sure
Mar 07 16:19:49 <lynxis> the edgerouter is also not listed in the wiki/toh. only the er-lite is in the wiki.
Mar 07 16:20:09 <Kunsi> i know :>
Mar 07 16:20:22 <lynxis> bbl.
Mar 07 16:20:43 <Kunsi> https://gist.github.com/Kunsi/87a4098e7ff5adfc5b14c5ec2dfbf17e
Mar 07 17:53:29 <zorun> PaulFertser: ah, right, forgot about http://downloads.openwrt.org/ :( I don't have access anyway...
Mar 07 17:54:10 <zorun> ynezz: this page? https://openwrt.org/docs/guide-developer/releases/release-process
Mar 07 18:00:58 <zorun> hsp: looks like archer-c7-v2 works fine in the imagebuilder here
Mar 07 18:00:59 <zorun> Building images for ar71xx - TP-LINK Archer C7 v2
Mar 07 18:01:05 <zorun> in 19.07.2
Mar 07 18:01:22 <zorun> if you are using ath79, be careful, the name has changed, it's tplink_archer-c7-v2 now
Mar 07 18:29:35 <lynxis> Kunsi: you have the Ubiquiti ERPRO-8?
Mar 07 18:30:35 <Kunsi> yes
Mar 07 18:31:17 <lynxis> can you do PCB pictures?
Mar 07 18:31:38 <lynxis> blogic: do you know about the ubiquity edge router pro? The sfp ports seems unsupported by openwrt.
Mar 07 19:00:55 <Kunsi> lynxis: i can try, but I only have one, which is my production router ...
Mar 07 19:01:17 <Kunsi> maybe we have a spare at the office i could open up
Mar 07 19:56:46 <pkgadd> hsp: just to rule out the obvious, ar71xx hasn't been built for master in over half a year by now, that will never work again (it should in openwrt-19.07 though), for master you'll have to switch to ath79
Mar 07 20:08:19 <aparcar[m]> ynezz: hey thanks again for th mail. I'm struggling a bit to figure out where the image target is actually called. From my Makefile understanding we first build up a hole lot of definitions which are then called at some place - which I did not find yet
Mar 07 20:08:25 <aparcar[m]> can you please give me a pointer here?
Mar 07 20:10:28 <ynezz> aparcar[m]: include/image.mk:  $$(_TARGET): $(BIN_DIR)/$(call IMAGE_NAME,$(1),$(2))$$(GZ_SUFFIX)
Mar 07 20:11:44 <aparcar[m]> that calls it? but it defined above the actual functions. or is _TARGET called itself at a later place?
Mar 07 20:12:45 <ynezz> that's how I understand the rules, target: prereq
Mar 07 20:13:13 <ynezz> so the $(BIN_DIR)/$(call IMAGE_NAME,$(1),$(2))$$(GZ_SUFFIX) is prereq for $$(_TARGET)
Mar 07 20:17:05 <aparcar[m]> Yea it does seem to work instantly
Mar 07 20:17:13 <aparcar[m]> ugh
Mar 07 20:24:28 <aparcar[m]> ynezz: can you give me a hint on the make json_overview_info_files command? How can I keep the name while telling Make that a file called $(BIN_DIR)/profiles.json is created?
Mar 07 20:26:33 <ynezz> that $(BIN_DIR)/profiles.json is target
Mar 07 20:35:22 <aparcar[m]> ok
Mar 07 20:36:28 <aparcar[m]> https://paste.debian.net/1133935/ like this?
Mar 07 20:40:06 <aparcar[m]> it does work but looks strange
Mar 07 20:40:08 * aparcar[m] sent a long message:  < https://matrix.org/_matrix/media/r0/download/matrix.org/iBuYkqVwvsUfCLsGerXuoTdX >
Mar 07 20:43:21 <aparcar[m]> ynezz: I think I'm close to the buildroot magic you requested, https://paste.sr.ht/%7Eaparcar/e7f2b71bd86662ec9eb6a9186a82c8609e8dd8bc
Mar 07 21:38:57 <PaulFertser> zorun: thank you for caring! Somebody has already fixed that, so no worries.
Mar 07 23:58:11 <owrt-snap-builds> build #255 of mpc85xx/p1020 is complete: Failure [failed pkgbuild]  Build details are at http://buildbot.openwrt.org/master/images/builders/mpc85xx%2Fp1020/builds/255  blamelist: Davide Fioravanti <pantanastyle@gmail.com>, Manuel Giganto <mgigantoregistros@gmail.com>, Sven Over <sp@cedenti.st>, Daniel Golle <daniel@makrotopia.org>, Adrian Schmutzler
Mar 07 23:58:12 <owrt-snap-builds> <freifunk@adrianschmutzler.de>, David Bauer <mail@david-bauer.net>, Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>, WeiDong Jia <jwdsccd@gmail.com>
**** ENDING LOGGING AT Sun Mar 08 03:01:09 2020