**** BEGIN LOGGING AT Sat Aug 06 02:59:57 2011
Aug 06 06:22:13 <builder_> build #64 of lantiq is complete: Failure [failed compile_6]  Build details are at http://buildbot.openwrt.org:8010/builders/lantiq/builds/64
Aug 06 09:53:05 <CIA-89> hauke * r27918 /trunk/ (12 files in 9 dirs): kernel: update to kernel version 3.0.1
Aug 06 09:55:07 <CIA-89> hauke * r27919 /trunk/toolchain/binutils/ (Config.in Makefile patches/2.21.1/): binutils: add binutils 2.21.1
Aug 06 10:42:26 <CIA-89> hauke * r27920 /trunk/target/linux/generic/ (8 files):
Aug 06 10:42:26 <CIA-89> kernel: add some missing config options
Aug 06 10:42:26 <CIA-89> These options where found by buildbot
Aug 06 10:52:15 <builder_> build #54 of avr32 is complete: Failure [failed compile_4]  Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/54
Aug 06 11:24:55 <CIA-89> hauke * r27921 /trunk/target/linux/uml/config/x86_64: uml: add some missing options
Aug 06 12:19:07 <CIA-89> hauke * r27922 /packages/net/openswan/ (4 files in 2 dirs):
Aug 06 12:19:07 <CIA-89> openswan: Update Openswan to upstream 2.6.34
Aug 06 12:19:07 <CIA-89> OpenWRT's bulid process currently uses Openswan v2.6.33, which does not build against the 2.6.39 kernel.
Aug 06 12:19:07 <CIA-89> This patch updates the OpenWRT build process to build Openswan v2.6.34, released 2011-06-08.
Aug 06 12:19:07 <CIA-89> hauke:
Aug 06 12:19:07 <CIA-89> * use Openswan v2.6.35
Aug 06 12:19:08 <CIA-89> Signed-off-by: Stephen Oberholtzer <oliverklozoff@gmail.com>
Aug 06 12:40:41 <CIA-89> nbd * r27923 /trunk/target/linux/generic/ (4 files in 2 dirs): kernel: add missing checks in the netfilter optimization patch which broke some rules containing only source/destination address checks
Aug 06 16:45:23 <CIA-89> hauke * r27924 /trunk/package/kernel/modules/block.mk:
Aug 06 16:45:23 <CIA-89> kernel: typo in cs5535 and cs5536 pata driver kernel definitions
Aug 06 16:45:23 <CIA-89> closes #9773
Aug 06 16:47:15 <CIA-89> hauke * r27925 /trunk/target/linux/generic/config-3.0: kernel: add missing option
Aug 06 17:08:54 <CIA-89> hauke * r27926 /trunk/package/mac80211/ (Makefile files/b43-fwcutter-fw-dirname.patch):
Aug 06 17:08:54 <CIA-89> mac80211: always store broadcom firmware in b43 and b43legacy dir
Aug 06 17:08:54 <CIA-89> This is always used to build a firmware for linux systems also if we are on freebsd.
Aug 06 17:08:54 <CIA-89> This is one patch from #9897
Aug 06 19:28:18 <__lore__> ping nbd
Aug 06 19:28:31 <nbd> pong
Aug 06 19:28:39 <__lore__> hi
Aug 06 19:28:46 <__lore__> how are you?
Aug 06 19:29:21 <nbd> hi
Aug 06 19:29:22 <nbd> good, and you?
Aug 06 19:29:56 <__lore__> fine tnx
Aug 06 19:30:00 <__lore__> :)
Aug 06 19:30:08 <__lore__> just a brief question
Aug 06 19:30:28 <__lore__> have you ever seen this kind of issue?
Aug 06 19:30:44 <__lore__> <3>ath: Failed to stop TX DMA!
Aug 06 19:30:44 <__lore__> <3>ath: DMA failed to stop in 10 ms AR_CR=0x00000024
Aug 06 19:30:44 <__lore__> AR_DIAG_SW=0x42000020 DMADBG_7=0x000067c0
Aug 06 19:30:44 <__lore__> <3>ath: Could not stop RX, we could be confusing the DMA engine when we
Aug 06 19:30:44 <__lore__> start RX up
Aug 06 19:30:44 <__lore__> <3>ath: Failed to stop TX DMA!
Aug 06 19:30:46 <__lore__> <3>ath: DMA failed to stop in 10 ms AR_CR=0x00000024
Aug 06 19:30:49 <__lore__> AR_DIAG_SW=0x42000020 DMADBG_7=0x000067c0
Aug 06 19:31:04 <nbd> yes
Aug 06 19:31:08 <__lore__> I am using ar9280 chipset
Aug 06 19:31:40 <__lore__> I have this msg before a system crash
Aug 06 19:31:50 <nbd> does it happen often?
Aug 06 19:31:56 <__lore__> quite often
Aug 06 19:32:03 <nbd> is it a new issue?
Aug 06 19:32:22 <__lore__> especially on 20MHz channel
Aug 06 19:33:02 <__lore__> what is due to ?
Aug 06 19:33:12 <__lore__> any ideas?
Aug 06 19:33:25 <nbd> no idea
Aug 06 19:33:29 <__lore__> :(
Aug 06 19:33:34 <nbd> have you checked when it was introduced?
Aug 06 19:33:54 <__lore__> not till now...but I have this issue for a long time
Aug 06 19:33:58 <nbd> ok
Aug 06 19:34:30 <__lore__> I am using openwrt trunk
Aug 06 19:35:51 <__lore__> I found this patch 560-ath9k_rx_stop.patch
Aug 06 19:35:58 <__lore__> could it help=
Aug 06 19:36:00 <__lore__> ?
Aug 06 19:46:53 <nbd> probably not in this case
Aug 06 19:47:38 <nbd> try checking whether there's a race condition that causes either new rx buffers to be added or tx frames to be put in the queue during a reset
Aug 06 19:47:51 <nbd> maybe there's a locking issue
Aug 06 20:20:56 <__lore__> ok tnx
Aug 06 20:20:59 <__lore__> bbl
Aug 06 22:00:02 <philipp64|laptop> jow_laptop: is the +(USE_EGLIBC||USE_GLIBC):libbsd syntax still broken?
Aug 06 22:01:44 <philipp64|laptop> nbd: or maybe Felix can answer that?
Aug 06 22:21:09 <philipp64|laptop> well, the input-core dependencies are buggered on x86, but I can't figure out the correct way to fix it.
Aug 07 00:13:20 <philipp64|laptop> say, I noticed that ledtrig-netdev supports having both 'rx tx' triggers... any reason we couldn't have 'phy0' be the trigger and 'rx tx' be the mode?
Aug 07 00:16:43 <nbd> for wireless you don't need ledtrig-netdev
Aug 07 00:17:18 <philipp64|laptop> no, what I'm saying is you can select 'phy0tx' or 'phy0rx'... you can't choose both.
Aug 07 00:18:58 <nbd> or just add a third trigger, txrx
Aug 07 00:21:00 <philipp64|laptop> we could do that too.
Aug 07 00:22:24 <nbd> when you said phy0 i was thinking that you were talking about wireless, which has its own triggers
Aug 07 00:22:33 <nbd> the correct trigger for wireless is phy0tpt
Aug 07 00:47:32 <philipp64|laptop> I was talking about wireless.  what's tpt? I understand 'rx', 'tx', 'assoc', etc. but not tpt
Aug 07 00:47:45 <nbd> throughput
Aug 07 00:47:53 <nbd> slow blinking with little activity, faster blinking with more
Aug 07 00:48:03 <philipp64|laptop> ah... and it's full-duplex.
Aug 07 00:48:17 <nbd> yep
Aug 07 00:48:25 <philipp64|laptop> I still think rxtx would be good as well.
Aug 07 00:48:50 <philipp64|laptop> dumb question... why not just add functionality on top of ledtrigger-netdev that's wifi specific?
Aug 07 00:50:03 <philipp64|laptop> of course, someone like me would probably want to add DSL-specific indications like carrier, etc. to it as well... and before long, it would be bloated.
Aug 07 02:14:34 <builder_> build #46 of ep93xx is complete: Failure [failed compile_4]  Build details are at http://buildbot.openwrt.org:8010/builders/ep93xx/builds/46
**** ENDING LOGGING AT Sun Aug 07 02:59:56 2011