**** BEGIN LOGGING AT Sat Sep 14 02:59:59 2013
Sep 14 04:26:40 <builder> build #325 of ixp4xx is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/ixp4xx/builds/325
Sep 14 05:00:08 <builder> build #37 of imx23 is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/imx23/builds/37
Sep 14 05:09:46 <builder> build #330 of xburst is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/xburst/builds/330
Sep 14 06:36:21 <builder> build #253 of octeon is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/octeon/builds/253
Sep 14 06:43:36 <builder> build #278 of gemini is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/gemini/builds/278
Sep 14 07:10:18 <builder> build #100 of mpc85xx is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/mpc85xx/builds/100
Sep 14 07:50:10 <The_Lizard> blogic: My computer restarted, so I have to re-do the last run
Sep 14 07:50:45 <blogic> hrr hrr
Sep 14 07:50:51 <blogic> i will do more tonight
Sep 14 07:51:11 <blogic> i build a uboot but need to test it today
Sep 14 07:53:53 <builder> build #284 of mcs814x is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/mcs814x/builds/284
Sep 14 07:58:50 <The_Lizard> im re-running it now, doing the lowercase test afterwards
Sep 14 07:59:04 <blogic> ok
Sep 14 09:05:59 <The_Lizard> nothing on the capitals
Sep 14 09:57:14 <The_Lizard> nothing as well...
Sep 14 10:01:52 <The_Lizard> somebody mentioned a while back, it could possibly be something like arcadyanarcadyan, anyone knows how I can use oclhashcat to check it out, but then by testing for each lower and uppercase possibilities?
Sep 14 10:06:30 <nbd> what kind of hash are you trying to crack?
Sep 14 10:07:46 <KanjiMonster> Devastator: yes, and I just confirmed my suspicion. Basically mdiobus->read() will always disregard the phyaddr parameter and use tg->phyaddr.
Sep 14 10:09:10 <The_Lizard> sha-1, 16 char length, from the arcadyan vgv7519 board
Sep 14 10:10:35 <nbd> have you tried google or some rainbow-table based sites?
Sep 14 10:10:41 <KanjiMonster> yes
Sep 14 10:11:45 <The_Lizard> yes, I have searched around, found nothing
Sep 14 10:12:20 <The_Lizard> the problem I guess, is that it is 16 characters long... it is considered way beyond 'unbreakable'
Sep 14 10:12:57 <KanjiMonster> Devastator: try http://inuyasha.ath.cx/~jonas/tg3_fix_mdio_access.patch in addition to hauke's patch(es)
Sep 14 10:13:14 <nbd> how do you know it's 16 characters long?
Sep 14 10:17:46 <The_Lizard> it is on the serial console
Sep 14 10:17:55 <The_Lizard> it waits for 16 characters to be entered
Sep 14 10:19:59 <OpenWrt-Commits> juhosg r37980 trunk/target/linux/generic/patches-3.10/400-mtd-add-rootfs-split-support.patch * generic/3.10: reduce number of ifdefs in the rootfs split code
Sep 14 10:20:00 <OpenWrt-Commits> juhosg r37981 trunk/target/linux/generic/patches-3.10/400-mtd-add-rootfs-split-support.patch * generic/3.10: remove unused defines from the rootfs split patch
Sep 14 10:38:21 <The_Lizard> wrote a program to generate the wordlist for me, didn't help..
Sep 14 11:13:26 <builder> build #268 of mpc52xx is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/mpc52xx/builds/268
Sep 14 11:25:54 <KanjiMonster> Hauke: hey. Found the reason for probably most issues: http://inuyasha.ath.cx/~jonas/tg3_fix_mdio_access.patch
Sep 14 11:26:37 <Hauke> KanjiMonster: ah yes, thanks
Sep 14 11:28:08 <KanjiMonster> it was probably chance that the register it wrote to had a R/O bit that stayed 0 ad made b53 read/writes seem to work, but actually didn't
Sep 14 11:29:31 <Hauke> KanjiMonster: yes, after the switch driver found a BCM5365, which is the result of reading 0x0 every time, I thought about this
Sep 14 11:30:34 <KanjiMonster> ah well, let's wait for results ;)
Sep 14 11:40:34 <Hauke> KanjiMonster: for the problem zajec is debugging, are there any mib values for the bcm5325 switch? I want to know how far the txed patches are going
Sep 14 11:40:44 <Hauke> I do not see then on the wire
Sep 14 11:58:55 <KanjiMonster> Hauke: there are some 16 bit counters, but I have never found real documentation about them
Sep 14 11:59:26 <Hauke> KanjiMonster: where did you actually found the documentation of the rest of the switch drivers?
Sep 14 12:03:19 <KanjiMonster> Hauke: mostly taken from bcmrobo.c and the old switch driver
Sep 14 12:03:45 <KanjiMonster> also some bits taken from the broadcom bcm63xx ethernet driver
Sep 14 12:03:55 <KanjiMonster> (which includes its own switch support code)
Sep 14 12:05:28 <Hauke> KanjiMonster: ok
Sep 14 12:33:25 <builder> build #286 of adm5120 is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/adm5120/builds/286
Sep 14 12:41:18 <zajec> Hauke: did you reproduce my problem using your BCM5365?
Sep 14 12:41:38 <Hauke> zajec: I haven't looked into the received data
Sep 14 12:41:43 <Hauke> but it looks the same
Sep 14 12:41:56 <zajec> ok
Sep 14 13:10:38 <builder> build #397 of at91 is complete: Failure [failed compile_5 shell_19]  Build details are at http://buildbot.openwrt.org:8010/builders/at91/builds/397
Sep 14 13:30:53 <nbd> stintel: ping
Sep 14 13:52:03 <builder> build #370 of ramips is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/ramips/builds/370
Sep 14 13:56:14 <CRCinAU> Anyone alive in here that isn't in #openwrt? :P
Sep 14 13:56:33 <CRCinAU> re trying to boot the ar71 kernel on a new device we're trying to figure out
Sep 14 13:56:43 <CRCinAU> more for uboot help
Sep 14 14:06:46 <Hauke> zajec: the gmac core you have is not core rev 4?
Sep 14 14:09:11 <Hauke> zajec: and it has a pmu
Sep 14 14:14:39 <zajec> i've to install & boot
Sep 14 14:14:41 <zajec> sec
Sep 14 14:14:59 <zajec> do you think we're lacking sth in PMU?
Sep 14 14:16:06 <Hauke> I am just going throgh the changes between an old and a new etcgmac.c version
Sep 14 14:16:32 <zajec> bgmac is based on some recent one I believe/
Sep 14 14:16:52 <zajec> do you think Broadcom dropped support for that old devices at some point?
Sep 14 14:17:18 <zajec> i'm trying to compile Linksys E900 firmware with et backporter from Asus RT16
Sep 14 14:17:51 <zajec> [    0.084000] bcma: bus0: Core 2 found: GBit MAC (manuf 0x4BF, id 0x82D, rev 0x03, class 0x0)
Sep 14 14:17:55 <zajec> is this what you needed?
Sep 14 14:18:11 <Hauke> yes so no problem, rev 4 needs some more changes
Sep 14 14:18:24 <Hauke> I think rev 4 is used in the arm devices
Sep 14 14:21:59 <zajec> how to *un*define sth with CFLAGS?
Sep 14 14:22:27 <zajec> gcc is called with -D HNDCTF
Sep 14 14:22:45 <zajec> can I put sth in the EXTRA_CFLAGS to undefine that?
Sep 14 14:23:24 <zajec>        -U name
Sep 14 14:23:26 <zajec>            Cancel any previous definition of name, either built in or provided with a -D option.
Sep 14 14:23:50 <Hauke> I didn't know about that
Sep 14 14:28:44 <zajec> i've compiled Linksys's firmware with et backported from Asus!
Sep 14 14:28:47 <zajec> it booted!
Sep 14 14:28:50 <zajec> and network works :)
Sep 14 14:28:59 <zajec> now I can try to put some debugging in the et
Sep 14 14:30:26 <Hauke> nice
Sep 14 14:36:49 <CRCinAU> when we're trying to boot an ar71 kernel on a Atheros CPU, are there any special things to do?
Sep 14 14:36:55 <CRCinAU> ie tftp it across
Sep 14 14:36:57 <CRCinAU> mboot
Sep 14 14:37:19 <CRCinAU> we're finding it either dies at "Starting kernel" or gives us "Bad Magic Number"
Sep 14 14:37:24 <CRCinAU> depending on what we use
Sep 14 14:47:29 <OpenWrt-Commits> nbd r37982 trunk/package/kernel/mac80211/patches/300-pending_work.patch * ath9k: merge a fix for tx processing issues with aggregation
Sep 14 14:47:40 <OpenWrt-Commits> nbd r37983 trunk/package/network/utils/iwinfo/src/iwinfo_nl80211.c
Sep 14 14:47:41 <OpenWrt-Commits> iwinfo: when using nl80211, parse interface combinations instead of creating/deleting interfaces
Sep 14 14:48:16 <OpenWrt-Commits> nbd r37984 branches/attitude_adjustment/package/iwinfo/src/iwinfo_nl80211.c
Sep 14 14:48:16 <OpenWrt-Commits> iwinfo: when using nl80211, parse interface combinations instead of creating/deleting interfaces
Sep 14 14:54:53 <zajec> et_module_init: passivemode set to 0x0
Sep 14 14:54:54 <zajec> PCI: Setting latency timer of device 0000:00:02.0 to 64
Sep 14 14:54:56 <zajec> R_REG(&ch->regs->cmdcfg) -> 0x010000DB
Sep 14 14:54:57 <zajec> R_REG(&regs->clk_ctl_st) -> 0x010B0000
Sep 14 14:55:02 <zajec> I've managed to dump R_REG :)
Sep 14 14:55:48 <zajec> unfortunately it doesn't work for other files...
Sep 14 14:55:53 <zajec> like hnddma.c
Sep 14 15:02:48 <Hauke> KanjiMonster: is there a way to tell if this is an internal switch or a dedicated switch better than assuming bcm5325 is a internal switch?
Sep 14 15:07:44 <zajec> i believe bcm5325 is an internal one
Sep 14 15:14:19 <Devastator> compiling a new build, let's see :D
Sep 14 15:14:22 * Devastator excited!
Sep 14 15:18:38 <zajec> Hauke: my gmac core on E900 seems to be mapped at 0xb8002000
Sep 14 15:18:49 <zajec> Hauke: do you think this value is hardware specific/
Sep 14 15:18:53 <zajec> constant between reboots/
Sep 14 15:19:07 <Hauke> yes it is constant
Sep 14 15:22:29 <Devastator> yesterday I've read b53 code completely.. but my assumptions were wrong, thankfully
Sep 14 15:22:59 <zajec> Devastator: do you also have BCM5356/BCM5357*?
Sep 14 15:23:17 <Devastator> no, mine is 5395
Sep 14 15:23:20 <Devastator> oooold hehe
Sep 14 15:27:32 <Devastator> does b53 replaces switch-robo from Felix?
Sep 14 15:27:44 <Hauke> yes
Sep 14 15:29:10 <nbd> i'm really happy about getting rid of the old switch-* crap
Sep 14 15:29:47 <Devastator> nbd Seb loves your code :)
Sep 14 15:29:59 <nbd> Seb?
Sep 14 15:30:10 <Devastator> Sebastian, or BrainSlayer
Sep 14 15:30:35 <nbd> he'll take any code as long as it works
Sep 14 15:30:48 <Devastator> hehe
Sep 14 15:31:03 <OpenWrt-Commits> nbd r37985 trunk/package/kernel/mac80211/patches/300-pending_work.patch * ath9k: add another minor tx path fix
Sep 14 15:31:49 <Devastator> I've never seen a pure switch-robo, just dd-wrt switch-robo with lots of hacks
Sep 14 15:32:16 <zajec> nbd: i love it even more :)
Sep 14 15:32:22 <zajec> i hated that old api :)
Sep 14 15:33:35 <Devastator> the thing I like here is, almost everything goes upstream
Sep 14 15:34:28 <nbd> zajec: yeah, most of the old crap was from 2005, i didn't have a lot of kernel experience back then ;)
Sep 14 15:35:16 <zajec> nbd: it's not any complaining about your work... i believe conditions were totally different back then :)
Sep 14 15:35:57 <zajec> i guess noone expected OpenWrt to grow so big ;)
Sep 14 15:36:26 <Devastator> their founders expected I guess hehe
Sep 14 15:36:39 <nbd> well, our ambitions kept growing along with the project ;)
Sep 14 15:36:49 <nbd> and still do
Sep 14 15:36:56 <zajec> :)
Sep 14 15:48:42 * Devastator crossing his fingers
Sep 14 15:55:02 <Devastator> yes!!
Sep 14 15:55:06 <Devastator> yes!
Sep 14 15:55:15 <Devastator> o// \\o o// \\o
Sep 14 15:55:56 <Hauke> Devastator: working?
Sep 14 15:56:11 <Devastator> Hauke seems like it, dmesg in a sec
Sep 14 15:57:20 <Devastator> Hauke http://codepad.org/yfZik431
Sep 14 15:59:12 <Hauke> Devastator: does ethernet work?
Sep 14 15:59:19 <Hauke> are you able to ping something?
Sep 14 15:59:24 <Devastator> Hauke just pinged, yes
Sep 14 16:03:49 <Devastator> I don't know if it matters but it says Link is up at 1000Mbps, when the pci card it is connected is actually 100Mbps
Sep 14 16:04:36 <Hauke> Devastator: what does "swconfig dev switch0 show" say?
Sep 14 16:06:08 <Devastator> Hauke http://codepad.org/xy3F7jNg
Sep 14 16:07:23 <Hauke> the connection between the ethernet mac and the switch uses 1000MBit/s and the connection between your pc and the switch is just 100MBit/s
Sep 14 16:08:38 <Devastator> oh, fine then :)
Sep 14 16:10:50 <jow_laptop> 0/win 10
Sep 14 16:10:53 <jow_laptop> sry
Sep 14 16:11:27 <OpenWrt-Commits> hauke r37986 trunk/target/linux/ brcm47xx/patches-3.10/790-tg3-phylib.patch brcm47xx/patches-3.10/780-tg3-phylib.patch
Sep 14 16:11:27 <OpenWrt-Commits> brcm47xx: tg3: use the phy address from et0phyaddr and not the default phy address.
Sep 14 16:14:52 <OpenWrt-Commits> hauke r37987 trunk/target/linux/ generic/files/drivers/net/phy/b53/b53_priv.h generic/files/drivers/net/phy/b53/b53_common.c * kernel: b53: add soft reset for BCM539x switches
Sep 14 16:18:26 <OpenWrt-Commits> hauke r37988 trunk/target/linux/generic/files/drivers/net/phy/b53/b53_priv.h * kernel: b53: use hard coded switch reset gpio on some boards.
Sep 14 16:20:11 <OpenWrt-Commits> hauke r37989 trunk/target/linux/brcm47xx/patches-3.10/260-MIPS-BCM47XX-add-board-detection.patch * brcm47xx: detect Asus RT-N10
Sep 14 16:21:59 <stintel> nbd: pong
Sep 14 16:26:06 <Devastator> Hauke is there any other thing we could work on or should I just git pull and build a new image to test what was commited?
Sep 14 16:26:51 <Hauke> Devastator: yes it would be nice if you test the things I just committed
Sep 14 16:27:05 <Devastator> no problem sir :)
Sep 14 16:30:05 <nbd> stintel: could you please test the updated ath9k patches from trunk to see if the issues you were having are gone?
Sep 14 16:33:00 <stintel> nbd: I'll try to check it on monday
Sep 14 16:33:11 <stintel> nbd: the device is at work
Sep 14 16:41:13 <nbd> stintel: ok, thx
Sep 14 16:43:02 * Devastator thinks about buying a bricked atheros device..
Sep 14 16:47:58 <builder> build #312 of ar71xx is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/ar71xx/builds/312
Sep 14 17:03:46 <Devastator> Hauke yes, it works after git pull and new image
Sep 14 17:04:48 <Devastator> Hauke a few questions to you: where I define gpios? where I configure lan/wan mac addresses?
Sep 14 17:05:52 <KanjiMonster> zajec: bcm5325 also exists as an external switch chip (in four flavours, bcm5325, bcm5325e, bcm5325m, and bcm5325f)
Sep 14 17:07:11 <zajec> interesting
Sep 14 17:07:33 <zajec> log from Linksys's firmware + Asus driver:
Sep 14 17:07:35 <zajec> et0: internal switch attached
Sep 14 17:07:37 <zajec> et0: reseting switch
Sep 14 17:25:40 <builder> build #313 of au1000 is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/au1000/builds/313
Sep 14 17:31:08 <Hauke> zajec: on bcm47xx /bcm53xx SoC the switch bcm5325 is mostly the internal switch
Sep 14 17:32:03 <Hauke> Devastator: nice to hear that it is finally working
Sep 14 17:32:35 <Hauke> Devastator: the gpios are configured in package/kernel/broadcom-diag/src/diag.c
Sep 14 17:33:31 <Hauke> Devastator: the default mac address is read from nvram, you can change the mac address the device should use in the OpenWrt configuration, wee this: https://forum.openwrt.org/viewtopic.php?id=27131
Sep 14 17:46:19 <zajec> Fucking yeah
Sep 14 17:46:24 <zajec> I've BCM5357 working
Sep 14 17:46:50 <Hauke> what was ist?
Sep 14 17:46:53 <Hauke> zajec:
Sep 14 17:57:14 <KanjiMonster> zajec_: I'm curious too, what was missing?
Sep 14 18:30:59 <Devastator> KanjiMonster do you know how to add devices to diag.c? don't know if it's used by brcm63xx
Sep 14 18:33:02 <KanjiMonster> Devastator: it's bcm47xx only, and no idea
Sep 14 18:33:17 <Devastator> thanks
Sep 14 18:34:17 <Devastator> KanjiMonster another quick question, shouldn't wanmac be lanmac+1?
Sep 14 19:12:22 <Devastator> zajec just teased our curiosity and left :(
Sep 14 19:53:54 <builder> build #350 of rb532 is complete: Failure [failed shell_12]  Build details are at http://buildbot.openwrt.org:8010/builders/rb532/builds/350
Sep 14 19:55:46 <builder> build #350 of ppc44x is complete: Failure [failed shell_12]  Build details are at http://buildbot.openwrt.org:8010/builders/ppc44x/builds/350
Sep 14 20:08:25 <OpenWrt-Commits> nbd r37990 trunk/package/network/services/hostapd/Makefile * hostapd: adjust the md5sum for the uploaded source tarball (fixes #14155)
Sep 14 21:10:41 <builder> build #363 of uml is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/uml/builds/363
Sep 14 21:54:31 <zajec> Hauke, were are you :(
Sep 14 22:46:12 <Devastator> zajec he and Kanji were curious why BCM5357 was not working
Sep 14 22:47:19 <Devastator> me included hehe
Sep 14 22:55:20 <zajec> yeah, i guess ;)
Sep 14 22:55:26 <zajec> i disappeared to test it further
Sep 14 22:55:29 <zajec> cause I still got some problems
Sep 14 22:55:33 <zajec> and i still have
Sep 14 22:55:39 <zajec> bgmac is working fine now
Sep 14 22:55:46 <zajec> but there is some problem with b53
Sep 14 22:56:07 <zajec> this line: b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL5_25, vc5);
Sep 14 22:56:13 <zajec> stops packets for me
Sep 14 22:56:36 <zajec> after commenting it out, packets are transmitted, however switch doesn't work
Sep 14 22:56:41 <zajec> it ignores VLAN tag
Sep 14 22:56:53 <zajec> and just passes it to the destination devices
Sep 14 22:57:05 <zajec> my notebook gets answers from the fouter with VLAN tag
Sep 14 22:57:10 <zajec> so it ignores them
Sep 14 22:57:41 <zajec> and the bgmac fix: http://marc.info/?l=linux-netdev&m=137919739607883&w=2
Sep 14 22:58:13 <zajec> as it usually happens... week of more spent just to find a one-liner fix ;/
Sep 14 22:58:33 <zajec> i'm going to work on b53 now
Sep 14 23:18:44 <KanjiMonster> zajec: hrm, but the broadcom code does the same. Well, actually it does not, it just writes VC5_DROP_VTABLE_MISS
Sep 14 23:20:31 <zajec> KanjiMonster: yes, this is correct
Sep 14 23:20:38 <zajec> we must be doing sth wrong at different place
Sep 14 23:22:06 <KanjiMonster> since parts of my code is taken from bcm63xx support code for bcm5325, maybe there's some stuff that is fine for a real bcm5325, but breaks the embedded bcm5325's of bcm536x
Sep 14 23:45:04 <Devastator> zajec have you ported any bcm47xx? I'm curious if openwrt uses wan mac as lan mac + 1
Sep 14 23:48:38 <zajec> Devastator: not idea what you exactly mean by porting bcm47xx
Sep 14 23:48:42 <zajec> i was working on that
Sep 14 23:48:58 <zajec> OpenWrt uses same MAC I believe
Sep 14 23:49:08 <zajec> this is because bcm47xx uses the same switch for WAN port
Sep 14 23:49:15 <zajec> it just assigns different VLAN id
Sep 14 23:49:36 <Devastator> zajec sorry, my english is far from being good :P
Sep 14 23:50:33 <Devastator> strange.. this router has a 100Mbit WAN and gigabit lan from the same switch?!
Sep 14 23:51:35 <zajec> yup
Sep 14 23:51:51 <zajec> err
Sep 14 23:51:53 <zajec> sec
Sep 14 23:51:54 <zajec> not ;)
Sep 14 23:51:58 <zajec> where did you read that?
Sep 14 23:52:09 <Devastator> zajec read what?
Sep 14 23:52:17 <zajec> info about that speeds
Sep 14 23:52:27 <zajec> 100Mbps vs. 1000Mbps
Sep 14 23:52:42 <Devastator> http://infodepot.wikia.com/wiki/Linksys_WRT310N_v1.0
Sep 14 23:53:11 <Devastator> anyway, still learning how openwrt works, dd-wrt creates wan and wireless mac with lan+1 and lan+2 respectively
Sep 14 23:53:58 <zajec> Devastator: I think it's unlikely... that 100 on WAN and 1000 on LANs
Sep 14 23:54:11 <zajec> KanjiMonster: any idea?
Sep 15 00:05:08 <Devastator> my cfe doesn't help either
Sep 15 00:49:11 <builder> build #382 of orion is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/382
Sep 15 01:20:28 <builder> build #388 of brcm47xx is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/brcm47xx/builds/388
Sep 15 01:23:38 <builder> build #331 of sibyte is complete: Failure [failed shell_12]  Build details are at http://buildbot.openwrt.org:8010/builders/sibyte/builds/331
Sep 15 01:39:58 <builder> build #380 of atheros is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/atheros/builds/380
Sep 15 02:23:16 <builder> build #402 of brcm63xx is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/brcm63xx/builds/402
**** ENDING LOGGING AT Sun Sep 15 02:59:59 2013