**** BEGIN LOGGING AT Tue Feb 14 02:59:56 2012
Feb 14 04:04:18 <CIA-56> nbd * r30515 /packages/net/samba36/Makefile: samba36: remove the tdbsam backend
Feb 14 04:04:20 <CIA-56> nbd * r30516 /packages/net/samba36/patches/310-remove_error_strings.patch: samba36: do not patch generated files
Feb 14 04:04:22 <CIA-56> nbd * r30517 /packages/net/samba36/patches/111-owrt_smbpasswd.patch: samba36: integrate a cleanup/rewrite of the smbpasswd utility from samba 3.0
Feb 14 07:04:43 <russell--> nbd: not the gain calibration error that i was seeing before
Feb 14 07:05:00 <russell--> so, i consider that a quite positive sign
Feb 14 07:06:11 <russell--> i have logread -f running on all of them
Feb 14 07:06:38 <russell--> nothing but expected crond stuff is showing up
Feb 14 08:01:43 <builder_> build #97 of adm5120 is complete: Failure [failed compile_10]  Build details are at http://buildbot.openwrt.org:8010/builders/adm5120/builds/97
Feb 14 09:06:55 <russell--> nbd: uh-oh, bad news.  One of the four test devices with your patch has failed after < 34 hours. "ath5k phy0: gain calibration timeout (2412MHz)"
Feb 14 09:07:04 <russell--> the other three are still running
Feb 14 09:09:20 <russell--> they've been up 33hrs, 29hrs and 28hrs, respectively
Feb 14 09:25:36 <russell--> i have some "echo all > /sys/kernel/debug/ieee80211/phy0/ath5k/debug" syslog captures, if you think they might be useful
Feb 14 09:31:04 <Olipro> how does the github mirror get synced with OpenWRT's svn repo?
Feb 14 09:32:00 <Olipro> is it automated and updated after every and any svn checkin, or automated periodically at a fixed interval, or just manual
Feb 14 09:34:59 * raorn pokes
Feb 14 09:35:15 <raorn> can someone please review my patches for tayga?
Feb 14 09:35:49 <Olipro> have you posted them to the mailing list?
Feb 14 09:35:58 <raorn> yesterday
Feb 14 09:36:10 <raorn> https://lists.openwrt.org/pipermail/openwrt-devel/2012-February/013978.html https://lists.openwrt.org/pipermail/openwrt-devel/2012-February/013977.html
Feb 14 09:36:21 <Olipro> when you say "review" - do you mean you want someone with commit access to do something with it?
Feb 14 09:36:25 <Olipro> if not, I'll take a look
Feb 14 09:36:45 <raorn> er... yes. i meant "review and apply"
Feb 14 09:37:25 <raorn> but if you don't have commit access, can you please take a look at patch2?
Feb 14 09:37:58 <raorn> i've moved static 4-to-6 mappings to firewall configuration
Feb 14 09:38:19 <Olipro> so your script loads the firewall config in
Feb 14 09:38:33 <Olipro> I'd say that's a nice idea
Feb 14 09:41:00 <russell--> nbd: here's a syslog capture with ath5k debugging on: https://personaltelco.net/~russell/debug-capture-during-gain-calibration-timeout-problem.log
Feb 14 09:41:06 <russell--> if that's any help at all
Feb 14 09:41:09 <Olipro> raorn: so, the functions you're calling for config nat64
Feb 14 09:41:16 <Olipro> do they only work on an individual address
Feb 14 09:41:34 <Olipro> I would say it'd be useful if you can add a nat64 section for both addresses *and* prefixes
Feb 14 09:42:33 <raorn> these are static mappings. you map one ipv6 address to one ipv4 address taken from dynamic pool
Feb 14 09:43:03 * russell-- hits hay
Feb 14 09:43:20 <Olipro> right, but what if I want to have a static mapping of an entire range of addresses
Feb 14 09:43:42 <Olipro> say, 10.0.0.0/24 into 2001:db8::/120
Feb 14 09:44:04 <raorn> tayga doesn't allow that
Feb 14 10:49:46 <builder_> build #92 of octeon is complete: Failure [failed compile_10]  Build details are at http://buildbot.openwrt.org:8010/builders/octeon/builds/92
Feb 14 11:14:31 <Kaloz> SmugLeaf: hm?
Feb 14 11:57:01 <WillieNL> nbd: dunno if the android-x86 port is a usefull project for you ? i have it running here in vbox since a while.
Feb 14 11:59:41 <WillieNL> nbd: also android is kinda like a 2 layer system.. universal java crap + linux kernel.
Feb 14 12:00:02 <WillieNL> as long that kernel is ok with the hw the rest will work ;)
Feb 14 12:00:44 <dape> damn, java over java over java, my tablet is bloated
Feb 14 12:01:06 <WillieNL> heared yesterday someone about wanting a java emulator for android to run java games on :S
Feb 14 12:01:09 <WillieNL> i was like wtf?!
Feb 14 12:01:13 <dape> i guess its easier for development and porting stuff from ios and others to android..
Feb 14 12:01:43 <WillieNL> also it is nice for fun toys :)
Feb 14 15:29:31 <builder_> build #132 of atheros is complete: Failure [failed compile_10]  Build details are at http://buildbot.openwrt.org:8010/builders/atheros/builds/132
Feb 14 15:34:20 <WillieNL> i still have some cobalt raq3 and raq4 amd k6 machines, are these builds still used by people ? (also have a raq550)
Feb 14 15:34:58 <WillieNL> else i'll sell them away here in holland with some linux on them..
Feb 14 15:35:40 <[florian]> I used to have a raq4 at some point
Feb 14 15:35:58 <CIA-56> blogic * r30518 /trunk/ (4 files in 4 dirs): [ltq-dsl] add annex selection support to ltq_dsl package
Feb 14 15:36:00 <CIA-56> blogic * r30519 /trunk/package/owsip/ (6 files in 2 dirs): [lantiq] add a simple sip client
Feb 14 15:36:02 <CIA-56> blogic * r30520 /trunk/package/pjsip/ (8 files in 3 dirs): [pjsip] bump to 1.12, merge improved lantiq aud_dev driver
Feb 14 15:36:04 <CIA-56> blogic * r30521 /trunk/target/linux/lantiq/patches/205-owrt-gpio-export.patch: [lantiq] fixes gpio_export patch
Feb 14 15:36:06 <CIA-56> blogic * r30522 /trunk/target/linux/lantiq/patches/850-etop_irq.patch: [lantiq] fixes irq locking in ethernet driver
Feb 14 15:36:06 <[florian]> then I was planning on cleaning up code and getting it mainline but never got the time to actually do it
Feb 14 15:36:22 <blogic> meep
Feb 14 15:36:33 <blogic> fxs/sip support \o/
Feb 14 15:36:44 <[florian]> WillieNL: if we want these to be running on openwrt, we need to patch the x86 kernel for these
Feb 14 15:36:57 <[florian]> WillieNL: by default you can get a kernel to boot, but that's all, there is no cobalt peripheral support
Feb 14 15:39:08 <WillieNL> [florian]: heh why patch the x86 kernel ? vanilla x86 kernels do run fine on those systems
Feb 14 15:39:22 <[florian]> because there is no peripheral support for the lcd, buttons etc ...
Feb 14 15:39:31 <[florian]> leds etc ...
Feb 14 15:40:04 <WillieNL> ahh ok that point but the cobalt prepal is not more then some sensors and a lpt/parallel port lcd+button circuit with a 74hcxxx multiplexed to 2 lpt ports
Feb 14 15:40:39 <WillieNL> also leds is onboard except the web led also jeffw have done a lot patches for that and i have used it without probs on gentoo since 2005 till a few years ago..
Feb 14 15:40:52 <[florian]> that's the problem, jeff's patch is just crappy
Feb 14 15:40:57 <WillieNL> still have some plain C based app for the lcd+button to replace the old scripts
Feb 14 15:41:11 <WillieNL> hmm jeffw stopped since kernel 2.6.28 i think
Feb 14 15:41:16 <[florian]> I need to get my hands back on my cobalt and code some decent leds/button driver
Feb 14 15:41:29 <WillieNL> since mosth has been taken in by upstream. like lcd and button
Feb 14 15:41:48 <WillieNL> also i don't run them anymore due no use for them :)
Feb 14 15:41:51 <[florian]> we are only missing the leds driver
Feb 14 15:41:51 <WillieNL> still have a few
Feb 14 15:41:56 <WillieNL> web led ?
Feb 14 15:42:02 <[florian]> yep
Feb 14 15:42:13 <WillieNL> moment that is the last he had done xD
Feb 14 15:42:19 <WillieNL> was not even kernel based
Feb 14 15:42:27 <[florian]> oh they also need to learn how to generate a patch without .orig files :)
Feb 14 15:43:24 <WillieNL> http://files.parvi.us/files/ebuilds/
Feb 14 15:43:35 <WillieNL> yes his patches was done on his way indeed
Feb 14 15:43:49 <WillieNL> i was one of the guys in his channel hoping that he was going to continue but sadly not
Feb 14 15:44:09 <[florian]> I would not mind taking maintenance of this, but I would need a gen V to also test
Feb 14 15:44:28 <WillieNL> ahh my first c app.. -> http://files.parvi.us/files/users/willie/
Feb 14 15:44:55 <WillieNL> look if it still is as what i remember
Feb 14 15:46:14 <[florian]> WillieNL: would you hand a gen v by chance?
Feb 14 15:46:49 <WillieNL> ehh hand ?
Feb 14 15:46:54 <[florian]> give
Feb 14 15:46:59 <[florian]> sell, send whatever
Feb 14 15:47:03 <WillieNL> i do nothing with it right now
Feb 14 15:47:20 <WillieNL> it is here only eating dust so if i can make someone happy withit .. no prob
Feb 14 15:47:26 <WillieNL> have a spare mainboard also
Feb 14 15:48:00 <WillieNL> raq550 ecc ram belive 256m and unknown about the 2 sticks in the spare board 2x seagate 40gb in it.
Feb 14 15:48:37 <WillieNL> [florian]: my last cube3 has been shipped from holland to germany for about 16~20 euro via GLS
Feb 14 15:49:04 <[florian]> ok, let's discuss that in private :)
Feb 14 15:49:56 <WillieNL> ahh webled was included in the ledstuff from jeffw's patches and write 0 or 1 to /proc/cobalt/led or some like that
Feb 14 15:57:35 <CIA-56> jow * r30523 /packages/libs/mysql/Makefile: [packages] mysql: fix compilation with gcc 4.6
Feb 14 16:10:14 <builder_> build #127 of ps3 is complete: Failure [failed compile_4]  Build details are at http://buildbot.openwrt.org:8010/builders/ps3/builds/127
Feb 14 16:20:16 <SmugLeaf> Kaloz: you around?
Feb 14 16:21:09 <Chocks> moo
Feb 14 16:22:26 <SmugLeaf> moo
Feb 14 17:04:49 <dape_x60> oink
Feb 14 17:51:06 <builder_> build #130 of cobalt is complete: Failure [failed compile_10]  Build details are at http://buildbot.openwrt.org:8010/builders/cobalt/builds/130
Feb 14 18:02:47 <CIA-56> juhosg * r30525 /trunk/target/linux/ramips/files/arch/mips/include/asm/mach-ralink/ (irq.h rt288x/irq.h rt305x/irq.h rt3883/irq.h): ramips: use SoC specific irq.h
Feb 14 18:02:48 <CIA-56> juhosg * r30526 /trunk/target/linux/ramips/rt3883/config-2.6.39: ramips: rt3883: sync 2.6.39 config
Feb 14 18:02:51 <CIA-56> juhosg * r30527 /trunk/target/linux/ramips/ (8 files in 6 dirs): ramips: rt3883: add PCI support
Feb 14 18:02:51 <CIA-56> juhosg * r30528 /trunk/target/linux/ramips/files/arch/mips/ralink/rt3883/mach-rt-n56u.c: ramips: rt3883: enable PCI on the RT-N56U board
Feb 14 18:02:53 <CIA-56> juhosg * r30529 /trunk/target/linux/ramips/base-files/etc/hotplug.d/firmware/10-rt2x00-eeprom: ramips: extract eeprom for the PCIe wlan module on the RT-N56U
Feb 14 18:04:16 <CIA-56> cshore * r30530 /packages/ (24 files in 24 dirs):
Feb 14 18:04:16 <CIA-56> [packages] Mass Fix for GCC 4.6
Feb 14 18:04:16 <CIA-56> nodefaultlibs now includes libc so we have to add it where we want it
Feb 14 18:18:09 <CIA-56> blogic * r30531 /trunk/package/pjsip/Config.in: [pjsip] drop obselete Config.in file
Feb 14 18:18:12 <CIA-56> blogic * r30532 /trunk/ (6 files in 5 dirs): (log message trimmed)
Feb 14 18:18:12 <CIA-56> [lantiq] Support booting the Speedport W502V using BRN-BOOT.
Feb 14 18:18:12 <CIA-56> While the disadvantage is less available flash space, it's easy and
Feb 14 18:18:12 <CIA-56> safe to flash without opening the device.
Feb 14 18:18:12 <CIA-56> Going back to the original firmware is also possible.
Feb 14 18:18:12 <CIA-56> This patch add two firmware utilities, mkbrncmdline and mkbrnboot.
Feb 14 18:18:13 <CIA-56> mkbrncmdline patches the uncompressed kernel so the registeres a0 to
Feb 14 18:33:15 <builder_> build #127 of pxcab is complete: Failure [failed compile_4]  Build details are at http://buildbot.openwrt.org:8010/builders/pxcab/builds/127
Feb 14 18:39:30 <CIA-56> jow * r30533 /packages/utils/collectd/ (Makefile patches/003-remove-werror.patch): [packages] collectd: remove further instances of -Werror to fix compilation with gcc 4.6 (#10962)
Feb 14 18:40:34 <russell--> nbd: the second of the 4 devices testing your patch started spitting "gain calibration timeout" messages after 41.2 hours.
Feb 14 18:41:13 <russell--> that's two of the four that have failed
Feb 14 18:47:43 <russell--> meanwhile, the revert-01, revert-02, revert-03, revert-04 devices are all still chugging along after 3+ days of uptime, giving further confidence to the "blame 0004" theory.
Feb 14 19:37:12 <CIA-56> blogic * r30534 /trunk/target/linux/lantiq/patches/206-owrt-brnboot.patch: [lantiq] fix compile error in previous commit
Feb 14 19:37:14 <CIA-56> blogic * r30535 /trunk/package/e2fsprogs/ (4 files in 2 dirs):
Feb 14 19:37:14 <CIA-56> [OpenWrt-Devel] update e2fsprogs
Feb 14 19:37:14 <CIA-56> Update e2fsprogs to the last upstream version 1.42.
Feb 14 19:37:14 <CIA-56> blkid works fine without 100_add_missing_libpthread_for_blkid.patch.
Feb 14 19:37:15 <CIA-56> Signed-off-by: Luka Perkov <openwrt@lukaperkov.net>
Feb 14 19:37:17 <CIA-56> blogic * r30536 /trunk/package/bridge-utils/Makefile:
Feb 14 19:37:17 <CIA-56> Update bridge-utils to the last upstream version 1.5.
Feb 14 19:37:17 <CIA-56> Signed-off-by: Luka Perkov <openwrt@lukaperkov.net>
Feb 14 19:37:19 <CIA-56> blogic * r30537 /trunk/target/linux/lantiq/base-files/etc/uci-defaults/uboot_environment:
Feb 14 19:37:19 <CIA-56> [lantiq] generate uboot environment data in target
Feb 14 19:37:19 <CIA-56> With this patch it is possible to generate uboot environment config file
Feb 14 19:37:19 <CIA-56> inside lantiq target. Currently it is only supporting GIGASX76X board.
Feb 14 19:37:19 <CIA-56> uboot environment config is used by fw_printenv fw_setenv tools.
Feb 14 19:37:20 <CIA-56> Signed-off-by: Luka Perkov <openwrt@lukaperkov.net>
Feb 14 19:37:21 <CIA-56> blogic * r30538 /trunk/package/kernel/modules/netsupport.mk:
Feb 14 19:37:21 <CIA-56> Add kernel module package for SLIP
Feb 14 19:37:22 <CIA-56> This patch adds support for a kernel module package for SLIP. We needed it forusing tunslip6 from Contiki OS for 6LoWPAN communications.
Feb 14 19:38:39 <blogic> philipp64|laptop: ping
Feb 14 19:45:05 <philipp64|laptop> blogic: pong
Feb 14 19:49:53 <blogic> you have 3 open patches
Feb 14 19:49:56 <blogic> right ?
Feb 14 19:50:01 <blogic> or patchwork tells me so
Feb 14 19:56:28 <Delboy_> blogic: do you remember if ath5k ever worked on arv4518?
Feb 14 19:57:06 <blogic> ermmmm
Feb 14 19:57:07 <blogic> no
Feb 14 19:57:12 <blogic> never used ath5k
Feb 14 19:57:14 <blogic> i used madwifi
Feb 14 19:57:22 <blogic> but i have 1 at home
Feb 14 19:57:34 <blogic> so i can potentially test
Feb 14 19:59:13 <Delboy_> i have arv4519, i think they have same wlan ar2417, now i dunno if eeprom has some problem because i get ath5k phy0: failed to warm reset the MAC Chip
Feb 14 19:59:39 <blogic> ah ok
Feb 14 19:59:41 <philipp64|laptop> looking...
Feb 14 19:59:41 <blogic> hmmm
Feb 14 20:00:29 <philipp64|laptop> the 2 ATM patches are still good-to-go. the LED patch is seriously out of date.
Feb 14 20:00:47 <blogic> ok
Feb 14 20:41:20 <danitool> Is this caused by the lack of Ram memory? http://pastebin.com/gXhVpeTS
Feb 14 20:41:30 <CIA-56> blogic * r30540 /trunk/target/linux/generic/ (3 files in 3 dirs):
Feb 14 20:41:30 <CIA-56> linux-atm: headroom/padding fixes
Feb 14 20:41:30 <CIA-56> Submitted upstream to netdev by Chas Williams on linux-atm.
Feb 14 20:41:30 <CIA-56> Signed-off-by: Philip Prindeville <philipp_subx@redfish-solutions.com>
Feb 14 20:41:32 <CIA-56> blogic * r30541 /trunk/target/linux/generic/ (3 files in 3 dirs): (log message trimmed)
Feb 14 20:41:32 <CIA-56> linux-atm: update solos-pci driver to upstream
Feb 14 20:41:32 <CIA-56> Fixes for the solos-pci driver.
Feb 14 20:41:32 <CIA-56> - Corrects calculation of headroom for padding.
Feb 14 20:41:33 <CIA-56> - Removes pointless debug messages.
Feb 14 20:41:33 <CIA-56> - Expose Bis A and Bis M annex capabilities.
Feb 14 20:41:34 <CIA-56> - Expose hardware revision.
Feb 14 20:41:36 <CIA-56> blogic * r30542 /trunk/package/kernel/modules/crypto.mk:
Feb 14 20:41:36 <CIA-56> [kernel] add missing autoload shortcut to crypto modules for btrfs
Feb 14 20:41:36 <CIA-56> Currently the btrfs module (and other fs modules) is loaded before other modules
Feb 14 20:41:36 <CIA-56> to provide early extroot compatibility. This will fail for btrfs however,
Feb 14 20:41:36 <CIA-56> because the dependencies crypto-core and crypto-hash are missing. This patch
Feb 14 20:41:39 <danitool> 16 MiB
Feb 14 20:41:41 <jow_laptop> danitool: yes
Feb 14 20:43:35 <CIA-56> florian * r30543 /trunk/package/gdb/Makefile: [package] gdb: use an updated 6.8a tarball
Feb 14 20:44:25 <jow_laptop> [florian]: maybe we should attempt to update it to 7.x while being at it
Feb 14 20:44:54 <[florian]> yes, I was just fixing the download issue
Feb 14 21:50:53 <SmugLeaf> Kaloz: i'll just email
Feb 14 21:56:59 <dape_x60> guys, do we need BUSYBOX_CONFIG_FEATURE_ROTATE_LOGFILE ?
Feb 14 21:57:39 <jow_laptop> dape_x60: no
Feb 14 21:58:11 <dape_x60> hehe, right, thanks
Feb 14 22:08:17 <dape_x60> i suppose for a very beginner user i dont need CONFIG_PACKAGE_MAC80211_DEBUGFS neither
Feb 14 22:11:12 <jow_laptop> well
Feb 14 22:50:20 <danmackay> ugh, I need to setup distcc on my 1 core vps
Feb 14 22:51:15 * danmackay hopes everything I added to my custom image fits into my image
Feb 14 22:51:30 <dape_x60> meh, cant compile compat-wireless without debugfs and mesh support, make will break
Feb 14 22:53:59 <danmackay> took me a few minutes to find out where kmod-ppp was so I could uninclude it
Feb 14 22:54:15 <danmackay> who needs pppoe stuff when you have cable internet anyways
Feb 14 22:54:48 <dape_x60> right
Feb 14 22:56:10 <danmackay> hmm, wonder if this builds sysupgrade will work on my router
Feb 14 22:56:25 <danmackay> the trunk build didnt -_-
Feb 14 22:57:18 <dape_x60t> factory didnt and you want sysupgrade to work?
Feb 14 22:57:59 <danmackay> I was trying to sysupgrade my trunk build with a newer trunk sysupgrade
Feb 14 23:45:43 <nbd> russell--: ping
Feb 14 23:46:39 <russell--> nbd: pong
Feb 14 23:47:09 <nbd> so about those devices that started reporting gain calibration timeouts...
Feb 14 23:47:14 <nbd> do they still work?
Feb 14 23:47:24 <nbd> of the ones with my patch
Feb 14 23:47:47 <nbd> or do they stop passing traffic?
Feb 14 23:47:52 <russell--> no, when they start spitting that, the beacons disappear
Feb 14 23:48:17 <nbd> ok
Feb 14 23:52:18 <nbd> i guess i'll have to dig up my old docs and code and do some more reading
Feb 14 23:52:33 <nbd> i'll see if i can come up with another patch for you to test tonight
Feb 14 23:52:36 <russell--> let me know if you need anything from me
Feb 14 23:52:38 <russell--> cool!
Feb 14 23:52:57 * russell-- has 60 of these things ... ample reason to help debug
Feb 14 23:53:33 <nbd> and soon i will probably need ath5k to work properly for a different project as well
Feb 14 23:53:39 <russell--> lol
Feb 14 23:53:42 <nbd> more motivation to just get this stuff fixed ;)
Feb 14 23:53:58 <russell--> i haven't seen the problem on x86 with ath5k radios
Feb 14 23:54:02 <nbd> also, i'd love to be able to finally nuke madwifi from openwrt trunk
Feb 14 23:54:14 <russell--> or, i think on wgt634u, mipsel
Feb 14 23:54:41 <nbd> what was the type of your device and radio again?
Feb 14 23:55:05 <russell--> it's an accton mr3201a, but i've seen the same thing on meraki mini's
Feb 14 23:55:18 <nbd> ar2315 or ar2317?
Feb 14 23:55:18 <russell--> it's an embedded atheros soc
Feb 14 23:55:26 <russell--> 2315 i think
Feb 14 23:55:28 <nbd> ok
Feb 14 23:55:43 <nbd> maybe that one needs some special handling
Feb 14 23:55:54 <russell--> if you want/need, i can get you an ipv6 shell on one
Feb 14 23:56:21 <nbd> i don't think i need one at the moment, but i'll keep it in mind
Feb 14 23:56:28 <russell--> okay
Feb 14 23:58:47 <russell--> i have r29672 (which is post compat-wireless-2012-12-02) running on a wgt634u with the stock radio for weeks without the same problem
Feb 14 23:59:24 <philipp64|laptop> jow_laptop: so about the dnsmasq set/tag stuff...
Feb 14 23:59:41 <russell--> [   26.580000] ath5k phy0: Atheros AR5213A chip found (MAC: 0x59, PHY: 0x43)
Feb 14 23:59:44 <russell--> [   26.588000] ath5k phy0: RF2112B 2GHz radio found (0x46)
Feb 14 23:59:47 <russell--> is what i see on the wgt634u
Feb 15 00:00:30 <russell--> this is what i see on the failing devices:
Feb 15 00:00:33 <russell--> [   24.360000] ieee80211 phy0: Selected rate control algorithm 'minstrel_ht'
Feb 15 00:00:36 <russell--> [   24.370000] ath5k phy0: Atheros AR2315 chip found (MAC: 0x86, PHY: 0x48)
Feb 15 00:01:04 <danmackay> oh great, build finished
Feb 15 00:02:08 <russell--> on the wgt, i see this:
Feb 15 00:02:11 <russell--> [   22.220000] Compat-wireless backport release: compat-wireless-2011-11-29
Feb 15 00:02:14 <russell--> [   22.228000] Backport based on wireless-testing.git master-2011-12-01
Feb 15 00:19:01 <builder_> build #120 of ppc44x is complete: Failure [failed compile_4]  Build details are at http://buildbot.openwrt.org:8010/builders/ppc44x/builds/120
Feb 15 00:37:47 <nbd> russell--: just noticed another thing, the calibration that fails is one that in my opinion doesn't even need to be re-run periodically
Feb 15 00:37:57 <nbd> and it's not actually the gain calibration that fails
Feb 15 00:38:09 <nbd> it's another calibration, but the error message is a bit confusing
Feb 15 00:38:22 <nbd> so there's a good chance that my next patch will fix this
Feb 15 00:38:33 <nbd> (to be applied on top of the previous one i already gave you)
Feb 15 00:38:46 <nbd> just need to do a bit more reading
Feb 15 00:54:45 <philipp64|laptop> russell--: you thinking about that x86-sysupgrade bounty?
Feb 15 01:14:41 <CIA-56> swalker * r30544 /packages/net/tor-alpha/ (3 files in 2 dirs): [packages] tor-alpha: update to 0.2.3.12-alpha, refresh patches
Feb 15 01:15:00 <CIA-56> swalker * r30545 /packages/net/bind/ (Makefile patches/): [packages] bind: update to 9.8.1-P1, thanks Oliver Smith, drop obsolete patches
Feb 15 01:15:17 <CIA-56> swalker * r30546 /packages/net/unbound/Makefile: [packages] unbound: update to 1.4.16
Feb 15 01:30:33 <nbd> russell--: http://nbd.name/441-ath5k_no_agc_recalibration.patch
Feb 15 01:57:19 <russell--> nbd: cool, i'll build and test again.
Feb 15 02:31:26 <russell--> nbd: any ideas on how to try to trigger the problem?  i reflashed one of the test devices that had not failed in 46+ hours.
Feb 15 02:31:53 <russell--> it would be cool if i could stress things a bit to get it to fail faster if it's going to at all
Feb 15 02:37:48 <nbd> i still don't know what makes the calibration fail
Feb 15 02:37:54 <russell--> lol
Feb 15 02:38:24 <nbd> i only know that other drivers only run it directly after reset
Feb 15 02:38:28 <nbd> and not periodically
Feb 15 02:39:13 <russell--> hmm
Feb 15 02:39:32 <russell--> anyways, i have 5 devices fired up with the new patch
Feb 15 02:40:33 <russell--> most of the failures have occurred by about 30 hours, although some have gone longer
Feb 15 02:44:41 <nbd> ok
**** ENDING LOGGING AT Wed Feb 15 02:59:57 2012