**** BEGIN LOGGING AT Mon Aug 05 02:59:58 2013 Aug 05 04:50:50 build #361 of brcm63xx is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/brcm63xx/builds/361 Aug 05 08:33:23 any dev here? Could anyone close this ticket? https://dev.openwrt.org/ticket/13960#comment:3 Aug 05 08:35:34 done Aug 05 08:35:49 blogic: thank you :-) Aug 05 08:37:12 should I make a ticket for my patch? Or should I leave it on patchwork/the mailinglist a bit longer? Aug 05 08:37:59 MartijnVdS: what ticket? what patch? (introduce me :-) ) Aug 05 08:38:15 strasidlo: http://patchwork.openwrt.org/patch/3892/ Aug 05 08:38:19 strasidlo: no ticket yet Aug 05 08:39:49 MartijnVdS: please don't create tickets for patches; this would just annoy devs Aug 05 08:48:56 whats the proper way to send a patch? Aug 05 08:49:13 dape: https://dev.openwrt.org/wiki/SubmittingPatches Aug 05 08:49:52 ah, so still as i knew it Aug 05 09:09:48 juhosg r37695 trunk/target/linux/ar71xx/patches-3.10/608-MIPS-ath79-ubnt-xm-add-more-boards.patch * ar71xx: ubnt unifi outdoor register gpio button Aug 05 09:09:49 juhosg r37696 trunk/target/linux/ar71xx/files/arch/mips/ath79/mach-tl-wa901nd.c * ar71xx: fix tl-wa901nd buttons code Aug 05 09:09:51 juhosg r37697 trunk/package/kernel/mac80211/Makefile * mac80211: enable RT{33,35,53,55}XX chip support in rt2800usb Aug 05 09:09:52 juhosg r37698 trunk/package/kernel/mac80211/patches/619-rt2x00-change-led-polarity-from-OF.patch * mac80211: rt2x00: fix LED polarity patch Aug 05 09:24:30 juhosg r37699 trunk/package/kernel/mac80211/Makefile * mac80211: fix rt2x00 dependency Aug 05 10:07:43 nbd r37700 trunk/package/system/ubus/Makefile * ubus: change section/category to base, libs Aug 05 10:08:15 nbd r37701 trunk/package/system/ubus/Makefile * ubus: fix typo Aug 05 10:13:28 strasidlo: you should have patchwork access now, please check Aug 05 10:14:49 jow_laptop: Yes, I've got access. Thank you so much. Aug 05 10:16:10 build #330 of cobalt is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/cobalt/builds/330 Aug 05 10:20:56 build #309 of rb532 is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/rb532/builds/309 Aug 05 10:24:40 build #309 of ppc44x is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/ppc44x/builds/309 Aug 05 10:32:23 nbd r37702 trunk/package/kernel/gpio-button-hotplug/src/gpio-button-hotplug.c Aug 05 10:32:23 gpio-button-hotplug: debounce the initial button state, the first reads at boot time might be wrong Aug 05 10:39:24 build #342 of orion is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/342 Aug 05 10:41:59 blogic r37703 trunk/target/linux/ (10 files in 9 dirs) * ar71xx: add support for dir 505 a1 Aug 05 10:42:06 KanjiMonster: I haven't created a ticket :) I was just wondering how I get it looked at/included/discussed :) Aug 05 10:42:11 blogic r37704 trunk/target/linux/ lantiq/xway/profiles/arv.mk lantiq/base-files/etc/uci-defaults/02_network lantiq/dts/ARV752DPW22.dts lantiq/base-files/etc/hotplug.d/firmware/10-rt2x00-eeprom * lantiq: several a803 related fixes Aug 05 10:42:19 blogic r37705 trunk/target/linux/octeon/config-default * octeon: add a few missing symbols to the kernel config file Aug 05 11:05:47 sadly, Gabor doesn't use IRC heheh Aug 05 11:07:01 Devastator: what do you need from him ? Aug 05 11:08:40 blogic r37706 trunk/target/linux/ar71xx/generic/profiles/d-link.mk * ar71xx: add usb modules to dir 505 profile Aug 05 11:12:22 cool, this dlink dir 505 looks pretty nice, Aug 05 11:12:41 do i read correctly it has 8 mb of flash? Aug 05 11:15:59 yes Aug 05 11:16:02 and 64mb ram Aug 05 11:16:14 its a annoiance to get serial soldered Aug 05 11:16:34 i ripped of the 2 pad by accident and hat to find a via leading to the uart ad and solder it there Aug 05 11:16:51 apart from that its pretty cool Aug 05 11:17:56 build #323 of uml is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/uml/builds/323 Aug 05 11:49:42 blogic to discuss Kanji's suggestion for that code he submitted, "Build Seama Header" Aug 05 11:50:45 or either "Seama header for 4M flash": https://lists.openwrt.org/pipermail/openwrt-devel/2013-July/020885.html Aug 05 12:33:01 blogic thank you, thank you, thank you, 'nuff said :) Aug 05 12:35:33 juhosg r37707 trunk/target/linux/ramips/image/Makefile * ramips: allow to build Seama firmware images with arbitrary size Aug 05 12:38:34 build #293 of sibyte is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/sibyte/builds/293 Aug 05 12:41:09 Devastator: ;) Aug 05 12:43:25 damn, now that I want to see the changes, trac gives me "bad gateway" Aug 05 12:45:13 haha Aug 05 12:45:40 have you 504'ed today ? Aug 05 12:46:00 502 and 504 Aug 05 12:49:30 juhosg r37708 trunk/target/linux/ramips/patches-3.10/0002-MIPS-ralink-add-pinmux-driver.patch * ramips: re-implement PCI mux Aug 05 12:57:47 Devastator: http://git.openwrt.org/?p=openwrt.git;a=log Aug 05 12:57:57 the commitdiff link Aug 05 13:30:03 jow_laptop: according to my package/base-files/files/lib/preinit/80_mount_root is called before target /linux/x86/base-files/lib/preinit/89_move_config - is it possible that this is it? Aug 05 13:30:53 strasidlo_: could be Aug 05 13:31:04 I'll try that Aug 05 13:31:10 strasidlo_: could have happened during the preinit refactoring Aug 05 13:35:42 hmmm Aug 05 13:36:37 strasidlo_: let me also have a look Aug 05 13:37:51 strasidlo_: ok Aug 05 13:37:54 i see the problem Aug 05 13:38:13 strasidlo_: you are using etx4 i assume Aug 05 13:38:26 blogic: I am using squashfs Aug 05 13:38:30 oh Aug 05 13:38:46 well it should work without that file then Aug 05 13:38:58 i will need to look at all 3 of the x86 specifc prenint files Aug 05 13:39:02 they all look broken Aug 05 13:39:12 a bit strange for x86, but I love the way, how can I do factory reset. Aug 05 13:40:39 blogic: it would be nice of you, if you take a look at it. Aug 05 13:43:29 heh, wasn't here someone failing to get pflash to work on rt3883? Aug 05 13:43:58 was there ? Aug 05 13:44:05 zinx: that was oyu or not Aug 05 13:44:52 blogic, yesssssssss the package arrived. Aug 05 13:45:05 td-w8970 v1.2 Aug 05 13:46:04 precioussssss Aug 05 13:46:12 ;) Aug 05 13:46:28 KanjiMonster: that was me, got it, was the gpio settings as you said - pinmux.c didn't support the pcimux option. Aug 05 13:46:49 KanjiMonster: i'd assumed it did, but that's what you get when you assume i guess; D Aug 05 13:47:23 ;) Aug 05 13:47:34 * quackquack nods Aug 05 13:48:43 looks like i just need to add support for loading eeprom from a file for PCI rt2x00 and it might work fully :/ Aug 05 13:51:07 blogic: if I can help you somehow with the sysupgrade, just ask me. I would love to participate .. somehow. Aug 05 13:52:40 blogic, no ready made image? Aug 05 13:52:47 or roll your own? Aug 05 13:53:38 quackquack: ;) Aug 05 13:53:50 damn.... :-p Aug 05 13:53:55 well Aug 05 13:54:14 i have not made it load with the modified uimage header yet Aug 05 13:54:20 blogic: how much confusion to you think there will be over a device with a sysupgrade that's exactly the same as the factory upgrade image? :x Aug 05 13:54:24 roll your own? is it...? it is! Aug 05 13:54:25 erh, s/to/do/ Aug 05 13:54:33 boah Aug 05 13:54:40 input overflow Aug 05 13:54:44 hehe Aug 05 13:54:53 so how to i install? Aug 05 13:54:55 serial? Aug 05 13:55:02 strasidlo_: let me look in a sec Aug 05 13:55:24 quackquack: build an image and then run it from serial, i will look into the uimage header thing today Aug 05 13:55:40 zinx: none, most sane devices are that way Aug 05 13:55:47 blogic: kk Aug 05 13:55:51 zinx: and eeprom loading for cpi device fails ? Aug 05 13:55:57 *pci Aug 05 13:55:59 aah. okay thanks. Aug 05 13:56:00 that is weird Aug 05 13:56:04 blogic: the driver doesn't support any platform data Aug 05 13:56:08 oh Aug 05 13:56:10 blogic: and the pci driver loads from the card Aug 05 13:56:15 one more question. how many pins are there on the device. Aug 05 13:56:18 yes, in that case oyu need to fix it Aug 05 13:56:23 i mean there is one under the sticker? Aug 05 13:56:24 blogic: yes :) Aug 05 13:56:33 blogic: shouldn't be hard to add dts/of/whatever support Aug 05 13:56:33 *void the warranty* ? Aug 05 13:56:36 zinx: which router is this ? Aug 05 13:56:41 quackquack: pins ? Aug 05 13:56:45 blogic: Edimax BR-6475nD Aug 05 13:56:51 pins = screws. Aug 05 13:57:02 screw the warranty? Aug 05 13:57:07 quackquack: 4 Aug 05 13:57:12 1 in each corner Aug 05 13:57:14 blogic: the driver supports pdata for SoC/SPI modes, but not PCI ;D Aug 05 13:57:20 aah screw the warranty. Aug 05 13:57:28 we void warranties !! Aug 05 13:57:38 (well, the driver supports pdata for PCI, but not through dts) Aug 05 13:57:41 * quackquack goes away screwing my warranty ;-) Aug 05 13:59:00 zinx: should be not too hard to add Aug 05 13:59:15 yeah Aug 05 13:59:18 just gotta find the time Aug 05 13:59:25 ;) Aug 05 13:59:38 on that note, off i go :/ Aug 05 13:59:49 * _rmk_ knows all about that. No one's yet given me a round tuit though. Aug 05 14:00:37 <_rmk_> on the plus side, draytek have sorted out the problems I've reported to them so far... eventually (it takes them about two months a go) Aug 05 14:00:52 ;) Aug 05 14:01:01 your unit still scares me Aug 05 14:01:12 <_rmk_> heh. Aug 05 14:01:20 _rmk_: there are so many nice dsl units out there Aug 05 14:01:26 and oyu picked the ugliest one ;) Aug 05 14:01:28 <_rmk_> I just need to convince them that doing a transparent dns proxy is a bad idea Aug 05 14:01:33 hahah Aug 05 14:01:52 <_rmk_> ... and then I'm all fixed Aug 05 14:02:31 <_rmk_> one thing though - once I get from them a firmware image which works, I'll be sticking with that image. Aug 05 14:02:54 <_rmk_> and if at any time you feel less scared and I have a round tuit... Aug 05 14:05:30 _rmk_: then i will use the time to make the next gen SoC work instread of fixing the problems of yesterday ;) Aug 05 14:05:56 which incidentially is inside the bt home hub 4 Aug 05 14:06:03 i found that out ont he weekend Aug 05 14:06:05 i think i will try screwing during daytime. Aug 05 14:06:11 bthh4 is based on arx368 Aug 05 14:06:49 <_rmk_> quackquack: whatever works for you :) Aug 05 14:07:38 i hope oyur screwing skills are idependent of the day time Aug 05 14:07:46 hehe, pink screen of death on esxi while running machine with OpenWrt on it. Aug 05 14:08:26 _rmk_, its like first use the 8 pin soic and backup the current firmware using a buspirate. Aug 05 14:08:54 and then upgrade the flash chip with a shining 16 MiB chip. Aug 05 14:09:12 and you were worrying about voiding th ewarranty Aug 05 14:09:22 and then if i fail. i can scream here. "help i failed!" Aug 05 14:09:33 tomorrow 10000% sure warranty void. Aug 05 14:09:41 replacing the flashchip. Aug 05 14:13:12 * _rmk_ wishes you luck Aug 05 14:13:27 <_rmk_> and hopes the soic8 pads don't come off the board Aug 05 14:13:57 using soic16, 16 MiB ;-) Aug 05 15:04:59 blogic: if you give me some instructions or hints, how to fix sysupgrade (and the whole preinit for x86), I would love to fix it myself. :-) Aug 05 15:09:35 strasidlo_: as stupid as it sounds the instruction is "find th eproblem and fix it" ;) Aug 05 15:09:53 sure Aug 05 15:10:28 I wonder what differentiates a cold reboot from a reboot Aug 05 15:11:42 I'd say that POST on the beginning of the boot process. Aug 05 15:11:42 the mount_root order Aug 05 15:12:04 i think that x86 has a custo way to handle the tar.gz thing Aug 05 15:12:07 blogic: how's that? some state kept in ram? kernel not restarted? Aug 05 15:12:21 jow_laptop: no idea yet Aug 05 15:12:31 i am fixing the tplink uimage header format just now Aug 05 15:12:43 the tplink adsl router has v2 headers which we dont support yet Aug 05 15:12:56 once that works i will look at the x86 sysupgrade Aug 05 15:44:32 build #303 of avr32 is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/303 Aug 05 15:50:49 finally.. Aug 05 15:51:05 isn't a great afternoon today? for me at least :) Aug 05 16:17:16 hauke r37709 trunk/ (7 files in 6 dirs) * kernel: fix DMA error when BCM4331 is connected to BCM4706 Aug 05 16:45:16 good evening. I compiled a Image for the EASY80920NAND and flashed it to the easybox 904 lte. I get a kernel panic because the partition for the filesystem is not set. ( I think that the kernel didn't find the right mtd devices but i'm not sure) but the good thing is that my laptop recognized the lan connection. But here is a bootlog of it http://nopaste.info/db32c55aac.html. Can somebody give me an hint what i should do now ? Aug 05 16:46:33 you need a proper DTS file Aug 05 16:47:41 anyway, it's easier to test with initramfs images than actual flashes Aug 05 16:48:30 you should flash the original fw back on now Aug 05 16:52:02 build #261 of iop32x is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/iop32x/builds/261 Aug 05 16:54:37 middey: you should start by using initramfs images Aug 05 16:54:46 however those are hard to make workj with brnboot Aug 05 16:55:13 the device don't use brnboot it use uboot but it is password protected Aug 05 16:55:24 *u-boot Aug 05 16:57:45 you might need a new u-boot Aug 05 16:58:40 ok :( how risky is it to flash a u-boot Aug 05 16:59:16 the unit has jtag Aug 05 16:59:19 so not very risky Aug 05 17:03:38 yes but i have no jtag tool at home if it bricks. But normally it should work when i use the gpl u-boot provided by Vodafone and change the if clause to accept every password but not the right. Or is there any way to flash a u-boot behind the u-boot for the first test if the u-boot works ? Aug 05 17:04:38 you have the uboot source? Aug 05 17:04:44 well then the passwd will be inside Aug 05 17:04:49 well then the passwd will be inside the source Aug 05 17:06:06 yes but it is a sha1 hash and i don't have the hardware/time to crack it Aug 05 17:06:31 have you tried google on the hash? Aug 05 17:06:48 what is the hash in the code ? Aug 05 17:06:52 show me please Aug 05 17:07:11 yes i must just search it Aug 05 17:20:19 nbd: I already tried, and two online sha1 (or 256, don't remember whihc one) rainbow table searches; no matches Aug 05 17:20:55 k Aug 05 17:21:16 I only tried free ones that don't require registration ;) Aug 05 17:21:19 I'm cheap Aug 05 17:21:26 john the ripper to the rescue Aug 05 17:21:49 middey: so the hash is ? Aug 05 17:22:51 I think it was 23a40c2a4549eae588dca47887bb5d7c559044f2 Aug 05 17:23:30 KanjiMonster, you need a couple of GPU's and hashcat Aug 05 17:23:37 char dgst[20]={0x23,0xa4,0x0c,0x2a,0x45,0x49,0xea,0xe5,0x88,0xdc,0xa4,0x78,0x87,0xbb,0x5d,0x7c,0x55,0x90,0x44,0xf2}; Aug 05 17:23:37 Aug 05 17:25:03 trine: for that I would also need a better psu; my system is tailored for low-power cpu + gpu ;) Aug 05 17:25:39 is there a ec2 way to do this ? Aug 05 17:25:48 I have such a system but the computer is being updated with ubuntu 13.04 at the moment Aug 05 17:26:09 blogic: I'm pretty sure there is Aug 05 17:26:19 @echo john the ripper Aug 05 17:26:27 http://sourceforge.net/projects/cryptohaze/files/Cryptohaze-Combined/ Aug 05 17:26:51 apparently that thing can crack sha1 for ~3$ on ec2 Aug 05 17:27:29 brb, upgrading hashes Aug 05 17:27:50 cryptohaze is not as good as hashcat and you will find the people behind cryptohaze on #hashcat Aug 05 17:31:36 i think i found something https://www.evernote.com/shard/s9/note/57d5c23d-3257-4973-90a7-74c4846d74f3/wishi/crazylazy i read this now Aug 05 17:42:31 cryptohaze seems to support sha1 but not 256 Aug 05 17:45:25 I'm waiting for sha3 support (the newly minted standard) Aug 05 17:46:15 (in most tools) Aug 05 17:49:48 lol ... http://www.spiegel.de/netzwelt/web/easybox-802-und-803-schwere-sicherheitsluecke-in-vodafone-routern-a-914857.html Aug 05 17:49:56 both units work with owrt btw ;) Aug 05 17:50:09 oopsie Aug 05 17:50:35 i mean ... if you let arcadyan build your firmware ... what do you expect Aug 05 17:50:36 ;) Aug 05 17:51:13 the funny thing is ... they make it sound like a new exploit Aug 05 17:51:27 arcadyan filed the patent that discloses the algorithm used for the exploit in 2004 Aug 05 17:51:45 int he sam patent they also descirbed the algorithm that is used to sign the fw images Aug 05 17:51:48 AND Aug 05 17:52:04 they also disclosed the ultra mea secret hightech master password which is 0x12345678 Aug 05 17:52:10 so what could possibly go wrong ?! Aug 05 17:52:15 ಠ_ಠ Aug 05 17:52:35 middey: ^ Aug 05 17:55:51 blogic works the easybox 803 also with the lte modem ? Aug 05 17:56:15 no idea Aug 05 17:56:20 there is no lte modem in the unit Aug 05 17:56:41 ok and what about the analog telephone ports Aug 05 17:57:06 yes but vodafone shipped a Lg fm300 with it Aug 05 17:57:57 not on all units Aug 05 17:58:02 and yes the analog phone ports work Aug 05 17:59:08 blogic: on the patent stuff i recall this: http://devopsreactions.tumblr.com/post/56593401192/pushed-sensitive-data-to-github Aug 05 17:59:53 wigyori: yes Aug 05 18:00:04 well filing a patent wit sensitive data is not really a fauxpaus Aug 05 18:00:11 its more a totally braindead thing to do Aug 05 18:00:40 maybe they have a patent for the easybox 904 password too :))) Aug 05 18:13:24 blogic the frimware update's for the easybox are better they have signed of the with openssl with a private key and test if the file signatur matches with one generated with the public key. But they commented out the check if the public key that is on the image also is on the device. So you can sign the image with any private key attach the signature and the public key and the firmware will pass the check Aug 05 18:15:12 btw have anybody used amazons e2 here ? Aug 05 18:23:54 build #310 of ar7 is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/ar7/builds/310 Aug 05 18:31:39 what happens if i flash a second uboot on the place of the kernel could this work ? (I see that on the dockstar wiki page http://wiki.openwrt.org/toh/seagate/dockstar) Aug 05 18:40:50 middey: only if you also change the uboot's bootcmd (note how that one's changed, too) Aug 05 18:41:39 that should not be the problem because the uboot env tools are running on the original firmware Aug 05 19:19:47 i'm now trying to flash the uboot as second stage loader i hope it works :) Aug 05 19:24:57 well Aug 05 19:25:08 it wont unless its textbase is changed Aug 05 19:27:55 blogic: im trying to bruteforce the firmware recovery page PIN code... Aug 05 19:33:47 blogic can you explain me what you would say with this ? Aug 05 19:38:29 middey: there is a define inside uboot source which you might need to change Aug 05 19:38:32 +#define CONFIG_SYS_TEXT_BASE 0x80800000 Aug 05 19:38:52 so when you load uboot to some ram location (say 0x80800000) Aug 05 19:39:15 uboot needs to be compiled with CONFIG_SYS_TEXT_BASE having the same value Aug 05 19:39:39 otherwise it wont work Aug 05 20:03:57 build #264 of ep93xx is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/ep93xx/builds/264 Aug 05 20:05:44 thanks luka i will try this tomorow Aug 05 20:14:40 np Aug 05 20:29:35 https://twitter.com/merlyn/status/364482243403591681 Aug 05 20:51:52 ? Aug 05 20:55:29 build #295 of xburst is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/xburst/builds/295 Aug 05 20:58:16 florian r37710 trunk/target/linux/generic/config-3.10 * kernel: add some missing config symbols Aug 05 21:07:52 russell--: interesting Aug 05 21:10:13 fatal: http://feeds.openwrt.nanl.de/openwrt/telephony.git/info/refs not found: did you run git update-server-info on the server? Aug 05 21:58:00 hauke r37711 trunk/package/kernel/mac80211/Makefile * mac80211: rtlwifi depends on PCI or USB support in the Kconfig file Aug 06 01:49:54 so, everything working........ except it doesn't transmit or receive :X Aug 06 01:51:04 makes me want to grab a cheap receive-only SDR dongle :/ Aug 06 01:58:55 grab a hammer and.. Aug 06 01:59:00 you know **** ENDING LOGGING AT Tue Aug 06 02:59:58 2013