**** BEGIN LOGGING AT Wed Sep 04 02:59:58 2013
Sep 04 05:39:16 <builder> build #277 of mcs814x is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org:8010/builders/mcs814x/builds/277
Sep 04 09:39:56 <Pteridium> KanjiMonster: i'm helping to port a Netgear EVG2000 router but there is a problem with the spi; could it be due to a spi multiplexor in the bottom side of the board? The board has voip.
Sep 04 09:40:04 <Pteridium> http://wiki.openwrt.org/toh/netgear/evg2000
Sep 04 09:41:38 <Pteridium> KanjiMonster: i'm helping to port a Netgear EVG2000 router but there is a problem with the spi; could it be due to a spi multiplexor in the bottom side of the board? The board has voip.
Sep 04 09:42:17 <Pteridium> KanjiMonster: i'm helping to port a Netgear EVG2000 router
Sep 04 09:44:45 <Pteridium> this is the board: http://wiki.openwrt.org/toh/netgear/evg2000
Sep 04 09:47:34 <Pteridium> and this is the diff i've made: http://pastebin.com/ir4vWSVR
Sep 04 10:11:47 <Pteridium> kanjiMonster: forgot to say that the problem is that the switch is not configured: http://pastebin.com/tVrT9y5f ,line 257
Sep 04 10:41:32 <KanjiMonster> Pteridium: are you sure there is an external switch? AFAICT it uses the internal switch with one (or both) rgmii ports connected
Sep 04 11:05:46 <_trine> hi KanjiMonster are we any nearer to updating the kernel on kirkwood
Sep 04 11:10:11 <luka> KanjiMonster: please put the patches somewhere so i could start working on it too
Sep 04 11:30:23 <karlp> I've got this in my bootup log: http://paste.drizzle.org/show/2356/ is it an actual problem?  Should I change my mach/board files to have what it "expects" ?
Sep 04 11:30:54 <jow_laptop> its normal
Sep 04 11:36:42 <karlp> what about "ath79-spi ath79-spi: master is unqueued, this is deprecated" ?
Sep 04 11:37:26 <blogic> karlp: in current trunk ?
Sep 04 11:37:39 <blogic> safe to ignore
Sep 04 11:37:48 <blogic> the kernel api has changed and the driver has not been updated
Sep 04 11:40:47 <karlp> current trunk, yeah, or a couple of days ago.
Sep 04 12:28:03 <stintel> nbd: still having a lot of issues with ath9k on a Buffalo WZR-HP-G300NH with BB r37838: http://pastebin.com/Mf4rmsHc
Sep 04 12:45:45 <Pteridium> KanjiMonster: sorry, i had a bad night and i was slept
Sep 04 12:46:32 <Pteridium> you are right, it uses one of both RGMII interfaces and has a BCM53115 switch
Sep 04 12:47:15 <Pteridium> in addition has an hex inverter: 74hc04
Sep 04 12:49:21 <Pteridium> i am using this repo because it has defined a, more or less, similar board defined: https://github.com/Noltari/openwrt/tree/bcm63xx-r37180
Sep 04 12:50:26 <Pteridium> this is the board: https://github.com/Noltari/openwrt/commit/aac2c38e0034fbfe9e289ff18ed0fcf48bd7175b
Sep 04 12:51:12 <KanjiMonster> Pteridium: do you have a bootlog of the original firmware?
Sep 04 12:53:13 <Pteridium> only a part of it: http://pastebin.com/FAiaBu6y
Sep 04 12:54:25 <KanjiMonster> the interesting parts are missing
Sep 04 12:55:06 <Pteridium> but you can see that spi0 is used by the voip
Sep 04 12:55:50 <merbanan> Pteridium: there are 2 silabs chips there
Sep 04 12:56:02 <Pteridium> yes, two
Sep 04 12:56:09 <KanjiMonster> also there are variants of bcm53115 that are configured through mdio, and this isn't supported yet
Sep 04 12:56:20 <Pteridium> ahhhhhhhhhh
Sep 04 12:57:15 <Pteridium> but mdio is not defined in the b-53 driver in the kernel?
Sep 04 12:57:19 <merbanan> Pteridium: what is the number on the voip chips
Sep 04 12:58:19 <KanjiMonster> Pteridium: the enetswitch driver does not support mdio probing/phy_devices yet (the old enet driver does)
Sep 04 12:58:37 <Pteridium> according to the wiki there are 2: http://wiki.openwrt.org/toh/netgear/evg2000
Sep 04 13:00:52 <Pteridium> KanjiMonster: then must i try with the old driver?
Sep 04 13:02:15 <KanjiMonster> Pteridium: no, the old driver is for the old chips (6345~6358), the enetsw driver is for the new chips (6368~63268)
Sep 04 13:04:28 <Pteridium> to summarize: if the switch is attached to the mdio bus there is no way to make it work wit openwrt at this moment. :-/
Sep 04 13:04:52 <Pteridium> well, work properly
Sep 04 13:06:34 <Pteridium> KanjiMonster: you are right; is attached to the mdio:       ethsw: found bcm53115!       ethsw_mdio_wreg: timeout!
Sep 04 13:09:04 <Pteridium> KanjiMonster: i'll tell the guy that has the board that the best is to send it to you
Sep 04 13:10:39 <KanjiMonster> Pteridium: no need to, I have a ugh ... similar enough board that needs this too (my problem in that case isn't hardware, it's time)
Sep 04 13:11:46 <Pteridium> ok, no problem. All of you are very busy AFAIK.
Sep 04 13:12:10 <KanjiMonster> it didn't help being away for almost two weeks ;)
Sep 04 13:12:45 <Devastator> it helps, if it was arelaxing
Sep 04 13:12:52 <Devastator> it helps, if it was relaxing weeks
Sep 04 13:12:55 <Devastator> :)
Sep 04 13:13:14 <KanjiMonster> definitely not relaxing, but fun
Sep 04 13:13:16 <Pteridium> no problem, the first are you ;)
Sep 04 13:13:57 <Pteridium> a lot of thanks for the advices
Sep 04 13:15:34 <Pteridium> KanjiMonster: forgot to say that take care with the stress; i've suffered it one time and is horrible
Sep 04 13:16:14 <Pteridium> merbanan: thanks for your advices too
Sep 04 13:23:28 <_trine> KanjiMonster, did you notice the line to you from me and then later luka ?
Sep 04 13:33:03 <KanjiMonster> _trine: luka: yes, sorry, I didn't get around doing it beofer I was gone for 1 1/2 weeks, and I likely won't get around to it today
Sep 04 13:39:07 <_trine> KanjiMonster, thanks for the response
Sep 04 14:09:36 <luka> KanjiMonster: no problem, i dont care if your patches are ugly... i'd just like to work on that this weekend
Sep 04 14:24:18 <builder> build #359 of cobalt is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org:8010/builders/cobalt/builds/359
Sep 04 14:27:27 <builder> build #339 of ppc44x is complete: Failure [failed shell_12]  Build details are at http://buildbot.openwrt.org:8010/builders/ppc44x/builds/339
Sep 04 14:29:33 <builder> build #339 of rb532 is complete: Failure [failed shell_12]  Build details are at http://buildbot.openwrt.org:8010/builders/rb532/builds/339
Sep 04 14:40:44 <middey> Good evening
Sep 04 14:41:07 <middey> Are there any news from the easybox 904 lte ?
Sep 04 14:41:41 <Devastator> luka hi, I just got a feedback from mtd guys about my patch, I lost the 3.12 merge window, but it will be merged in 3.13, can I make an openwrt patch with the exact message I commited to mtd list, with that changelog in the wrong place or should I correct the changelog "position"?
Sep 04 14:42:38 <DonkeyHotei> middey: is that the box with 128 ram?
Sep 04 14:42:55 <middey> DonkeyHotei yes i think so
Sep 04 14:43:28 <DonkeyHotei> i'd love to run asterisk or freeswitch on one of those with its fxs port(s)
Sep 04 14:44:42 <DonkeyHotei> the lte part would be useless in the usa, as it doesn't support any lte radio bands used here
Sep 04 14:45:35 <middey> i'd like to flash a second u-boot as 2nd stage bootloader(because the normal u-boot needs a password)  but i don't know which modification i have to do to make it run
Sep 04 14:46:03 <middey> DonkeyHotei there is also a dsl version of the easybox 904
Sep 04 14:46:21 <DonkeyHotei> wasn't the password pasted in this channel a few weeks ago?
Sep 04 14:46:48 <DonkeyHotei> adsl, or vdsl?
Sep 04 14:47:21 <DonkeyHotei> and annex a, or b?
Sep 04 14:47:28 <middey> DonkeyHotei i know that we have pasted the sha1 hash of it but i don't know if someone has bruted it
Sep 04 14:47:53 <KanjiMonster> didn't someone patch his u-boot to skip the check?
Sep 04 14:48:02 <DonkeyHotei> i think the unhashed passwd was also pasted iirc
Sep 04 14:49:39 <middey> KanjiMonster i have the check patched but when i flash it after the firmware my box goes into bootloops
Sep 04 14:50:32 <middey> DonkeyHotei i llok for a datasheet now they called  the version xdsl but i think this is not a realy standart
Sep 04 14:51:28 <DonkeyHotei> [Mon 2013-08-05 10:51:27 AM PDT]  <blogic> arcadyan filed the patent that discloses the algorithm used for the exploit in 2004
Sep 04 14:51:28 <DonkeyHotei> [Mon 2013-08-05 10:51:45 AM PDT]  <blogic> int he sam patent they also descirbed the algorithm that is used to sign the fw images
Sep 04 14:51:28 <DonkeyHotei> [Mon 2013-08-05 10:51:48 AM PDT]  <blogic> AND
Sep 04 14:51:28 <DonkeyHotei> [Mon 2013-08-05 10:52:04 AM PDT]  <blogic> they also disclosed the ultra mea secret hightech master password which is 0x12345678
Sep 04 14:51:30 <KanjiMonster> xdsl usually means adsl and vdsl supported
Sep 04 14:52:27 <DonkeyHotei> does that generate the correct hash?
Sep 04 14:52:58 <middey> DonkeyHotei i fire up my serial connection now and try it :)
Sep 04 14:53:16 <DonkeyHotei> no, check it against the hash first
Sep 04 14:55:11 <middey> i don't have the hash stored on this machine
Sep 04 14:55:53 <DonkeyHotei> didn't you say it's in the downloadable source tarball?
Sep 04 14:56:05 <middey> yes it is
Sep 04 14:56:24 <middey> 56 kbits/s download here :(
Sep 04 14:56:30 <DonkeyHotei> oh
Sep 04 14:57:04 <DonkeyHotei> wait, i think that was pasted too; lemme check
Sep 04 14:57:41 <DonkeyHotei> [Mon 2013-08-05 10:22:51 AM PDT]  <KanjiMonster> I think it was 23a40c2a4549eae588dca47887bb5d7c559044f2
Sep 04 14:57:41 <DonkeyHotei> [Mon 2013-08-05 10:23:30 AM PDT]  <trine> KanjiMonster, you need a couple of GPU's and hashcat
Sep 04 14:57:42 <DonkeyHotei> [Mon 2013-08-05 10:23:37 AM PDT]  <<?>middey> char dgst[20]={0x23,0xa4,0x0c,0x2a,0x45,0x49,0xea,0xe5,0x88,0xdc,0xa4,0x78,0x87,0xbb,0x5d,0x7c,0x55,0x90,0x44,0xf2};
Sep 04 14:57:56 <russell--> KanjiMonster: /me gently reminds you about b53 problems on wgt634u ...assuming they still exist ;-)
Sep 04 14:58:23 <KanjiMonster> russell--: still no clue why it happens
Sep 04 14:58:49 * russell-- was about to take another look
Sep 04 14:59:57 <middey> DonkeyHotei thats not the same hash
Sep 04 15:00:17 <DonkeyHotei> not the same as in the source? you pasted it
Sep 04 15:01:04 <middey> it's the same as in the source but the posted password from blogic does not match with these hash
Sep 04 15:01:50 <middey> DonkeyHotei router-faq.de says that the easybox 904 xdsl runs with ADSL, ADSL2+, VDSL2
Sep 04 15:03:33 <DonkeyHotei> none of the password cracking programs suggested on Aug 5 seem to support sha1
Sep 04 15:05:37 <middey> i tried two online hash cracking services which wan't money if they crack this hahs but after 5 days they send me a mail that the cracking was not succesfull and they are sorry about it
Sep 04 15:07:24 <middey> DonkeyHotei i think it is more usefull if we can install a second stage bootloader because cracking this hash is to expensive (time and money)
Sep 04 15:09:01 <builder> build #371 of orion is complete: Failure [failed compile_8]  Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/371
Sep 04 15:13:25 <middey> DonkeyHotei can you help me with the uboot as second bootloader ? I have it sourcecode from the GPL tarball so the things to change should be minimal. But i don't know anything about the u-boot code
Sep 04 15:19:34 <builder> build #352 of uml is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/uml/builds/352
Sep 04 15:21:50 <builder> build #320 of sibyte is complete: Failure [failed shell_12]  Build details are at http://buildbot.openwrt.org:8010/builders/sibyte/builds/320
Sep 04 15:27:02 <_trine> if you want an hash cracked I could ask someone i know
Sep 04 15:27:14 <Hauke> russell--: what problems do you have with b53 on wgt634u?
Sep 04 15:30:36 <stintel> ]
Sep 04 15:47:32 <russell--> Hauke: it doesn't work ;-)
Sep 04 15:47:49 <russell--> or, initially, it's not detected iirc
Sep 04 15:48:05 <Hauke> ah it does not get detected
Sep 04 15:48:10 * russell-- is about to reflash a recent image to recheck
Sep 04 15:48:25 <Hauke> russell--: what switch is used in this device?
Sep 04 15:48:34 <russell--> KanjiMonster provided a patch that fixed detection, but then it kernel panic'd
Sep 04 15:48:45 <Hauke> do you have a link for me
Sep 04 15:48:58 <russell--> it was a few weeks ago, i'll dig it up
Sep 04 15:51:41 <russell--> Hauke: https://personaltelco.net/~russell/wgt634u-3.10.4-boot.log
Sep 04 15:54:31 <russell--> Hauke: i think this one is with KanjiMonster's detection patch: https://personaltelco.net/~russell/wgt634u-3.10.4-boot2.log
Sep 04 15:58:43 <luka> Devastator: use the exact same commit message
Sep 04 16:00:18 <russell--> Hauke: from circa Aug 02, if you have logs, i'm reproducing the fixed up patch now
Sep 04 16:07:37 <Hauke> russell--: could you try this patch: http://pastebin.com/Cw6Jp3iD
Sep 04 16:07:53 <Hauke> KanjiMonster: is this the right way to fix this problem?
Sep 04 16:08:37 <KanjiMonster> Hauke: lemme check, and good spot
Sep 04 16:10:36 <KanjiMonster> Hauke: should be right, no ops is only needed for bcm5325 and compatible (as these have no sensible MIB counters - they have e.g. 16 bit packet counters ... m( )
Sep 04 16:15:31 <russell--> Hauke, trying ...
Sep 04 16:15:52 <OpenWrt-Commits> hauke r37895 trunk/target/linux/generic/files/drivers/net/phy/b53/b53_common.c * kernel: b53: allow access the MIB counters on BCM5365
Sep 04 16:18:33 <Hauke> KanjiMonster: the cpu port is the same for BCM5325 and BCM5365, the code assumes this for both port 5?
Sep 04 16:19:26 <KanjiMonster> Hauke: yes, that is correct. Well for most things the cpu port is 5, for some it is 8 (like the mib counters offset/page)
Sep 04 16:19:50 <KanjiMonster> dunno what broadcom did there ;p
Sep 04 16:20:00 <Hauke> KanjiMonster: ok
Sep 04 16:20:08 <KanjiMonster> it seems to be a bit arbitrary.
Sep 04 16:20:41 <Hauke> this is probably a partly stripped down 8 port switch ;-)
Sep 04 16:22:32 <KanjiMonster> yeah, that's what I think, too
Sep 04 16:24:15 <KanjiMonster> for added fun, bcm53115 (and probably bcm53125, too) has two cpu ports at 5 and 8, and sometimes board designers chose to use the one at 5 (where then the mibcounters are actually at the offset for 5)
Sep 04 16:24:45 <KanjiMonster> so you can't even force 5 to 8 for these parts
Sep 04 16:25:08 <KanjiMonster> (and in theory they might even use both and bond them)
Sep 04 16:29:57 <russell--> flashing r37895
Sep 04 16:32:55 <russell--> ah, we still have the detection problem, need to apply KanjiMonster's patch again
Sep 04 16:51:53 <russell--> ah, yes, things looking good now
Sep 04 16:53:07 <russell--> # dmesg | grep -i b53
Sep 04 16:53:07 <russell--> [    5.516000] b53_common: found switch: BCM5365, rev 0
Sep 04 16:53:08 <russell--> [    5.524000] b44 ssb0:0 (unregistered net_device): attached PHY driver [Broadcom B53 (3)] (mii_bus:phy_addr=1:1e)
Sep 04 16:53:19 <russell--> no panic :-)
Sep 04 16:55:35 <russell--> Hauke: here's KanjiMonster's patch as fixed up: http://sprunge.us/jWFe
Sep 04 16:57:33 <russell--> r37895 + http://sprunge.us/jWFe == "all better now on wgt634u"
Sep 04 17:19:22 <builder> build #330 of avr32 is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/330
Sep 04 17:33:27 <Hauke> KanjiMonster: would you like to commit the BCM5365 detection patch?
Sep 04 17:34:57 <froek> if i wanted to change code in extroot - ie, boot order etc, where would I look?
Sep 04 17:35:17 <Hauke> russell--: nice
Sep 04 17:39:02 <DonkeyHotei> middey_afk: i'm attempting a bruteforce of your hash atm. no results for lengths up to 7, and 21 hours left for trying a length of 8
Sep 04 17:51:51 <middey> DonkeyHotei thank you for trying :)
Sep 04 18:51:39 <Devastator> luka when you have the time: http://codepad.org/cwzm4vCa
Sep 04 19:17:49 <luka> Devastator: file target/linux/generic/patches-3.10/998-mtd_m25p80_add_support_for_esmt_f25l32pa.patch should not be 998 it should be 00X where X is first available number
Sep 04 19:18:34 <luka> look at target/linux/generic/PATCHES to know why
Sep 04 19:19:08 <luka> also, on line 48 of your patch please add links [0] and [1]
Sep 04 19:22:13 <Devastator> luka will do, let me see what's the first available number
Sep 04 20:00:46 <Devastator> luka http://codepad.org/ptkhVOqL
Sep 04 20:00:54 * Devastator crosses his fingers..
Sep 04 20:57:41 <_trine> middey, are you here
Sep 04 20:57:52 <middey> yes _trine
Sep 04 20:57:57 <_trine> pm
Sep 04 20:58:53 <DonkeyHotei> it doesn't need to be in pm
Sep 04 20:59:15 <DonkeyHotei> it's related to owrt development, after all
Sep 04 20:59:39 <middey> DonkeyHotei he/she just asked if the hash is sha1
Sep 04 21:00:05 <luka> Devastator: looks great - one more tiny thing - please run make target/linux/refresh and send that "refreshed" version of the patch
Sep 04 21:01:58 <luka> Devastator: most likey other patches will get refreshed too, but only send a patch with your patch in it
Sep 04 21:15:20 <middey> god night @ all
Sep 04 21:28:13 <builder> build #280 of iop32x is complete: Failure [failed shell_12]  Build details are at http://buildbot.openwrt.org:8010/builders/iop32x/builds/280
Sep 05 00:50:32 <builder> build #247 of octeon is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/octeon/builds/247
Sep 05 01:11:52 <Devastator> luka where should the refreshed version be?
Sep 05 01:18:28 <Devastator> according to git status, my patch was deleted
Sep 05 01:28:24 <Devastator> I was able to recover but I wonder why it is deleting..
Sep 05 02:05:01 <roh> Devastator: add and commit locally (afaik
Sep 05 02:05:25 <roh> or stash
Sep 05 02:06:50 <Devastator> roh that's what I did but when I run make target/linux/refresh it deletes my patch..
Sep 05 02:10:42 <roh> uuh.. dunno. check the manual. i think it was in there somewhere
Sep 05 02:11:31 <roh> http://wiki.openwrt.org/doc/devel/patches
Sep 05 02:12:43 <Devastator> I've checked the wiki, but it never mention why it deletes patches, also I'm not using quilt, should I?
Sep 05 02:14:10 <roh> i guess so
Sep 05 02:14:17 <roh> dunno now.. tired...  n8
Sep 05 02:15:02 <Devastator> night roh and thanks!
**** ENDING LOGGING AT Thu Sep 05 02:59:59 2013