**** BEGIN LOGGING AT Thu Jun 11 02:59:57 2009
Jun 11 07:18:06 <CIA-29> juhosg * r16415 /trunk/target/linux/ar71xx/files/arch/mips/ar71xx/irq.c: [ar71xx] flush more register writings
Jun 11 07:57:55 <CIA-29> florian * r16416 /trunk/target/linux/brcm47xx/patches-2.6.28/816-ssb_fix_irq_setup.patch: [brcm47xx] fix ssb irq setup, allows ubsec_ssb devices on Asus WL500GP and Netgear WGT634U to work, no side effects noticed on other devices
Jun 11 08:07:16 <CIA-29> juhosg * r16417 /trunk/target/linux/ar71xx/files/ (4 files in 3 dirs): [ar71xx] move definition of ar91xx_flash_platform_data into a separate header file (closes #5321)
Jun 11 08:17:23 <CIA-29> hcg * r16418 /trunk/include/kernel-version.mk: Add 2.6.30 final
Jun 11 08:26:54 <CIA-29> hcg * r16419 /trunk/target/linux/ (8 files in 8 dirs): Bump 2.6.30 kernel version to final
Jun 11 08:49:55 <CIA-29> florian * r16420 /trunk/target/linux/generic-2.6/patches-2.6.30/202-mips-freestanding.patch: [kernel] refresh generic patches
Jun 11 12:40:09 <CIA-29> nbd * r16421 /trunk/package/madwifi/files/lib/wifi/madwifi.sh: madwifi: disable compression by default, as it can cause instabilities on various platforms (thx, acoul)
Jun 11 12:41:05 <CIA-29> nbd * r16422 /branches/8.09/package/madwifi/files/lib/wifi/madwifi.sh: merge r16421 to 8.09
Jun 11 12:45:29 <CIA-29> hcg * r16423 /trunk/target/linux/octeon/patches/001-handle_removal_h_constraint.patch: Removed delay.h update from 001-handle_removal_h_constraint.patch
Jun 11 12:58:24 <CIA-29> lars * r16424 /packages/mail/dovecot/Makefile: [packages] dovecot: Disable linking against libpam. Fixes #5322
Jun 11 13:02:12 <larsc> uhm that should have been "Fixes #5324"
Jun 11 15:33:13 <xxiao> juhosg: are you aware of the ar71xx pci-core bug that nbd mentioned?
Jun 11 15:33:42 <xxiao> juhosg: my ath9k reset the routerstation board sometimes while idle
Jun 11 15:34:04 <xxiao> ath5k crashes, saying "data bus error" with some pci-bus information
Jun 11 15:34:26 <xxiao> similar things happened on a rb433ah board, which is also ar71xx based
Jun 11 15:52:07 <juhosg> xxiao: you mean something similar to this: http://openwrt.pastebin.ca/1456715 ?
Jun 11 16:06:54 <xxiao> juhosg: yes
Jun 11 16:07:46 <xxiao> juhosg: the ath5k /madwifi one also said something about "data bus error"
Jun 11 16:09:22 <nunojpg> I've submitted a patch 2 months ago, so I hope you all forgive me to ask a user question in this channel. I have to solve my problem in 3 hours.
Jun 11 16:09:23 <nunojpg> I can't access ports 580 and 522 from WAN side regardless of the iptables rule
Jun 11 16:09:33 <nunojpg> http://pastebin.com/m3e1a28ac
Jun 11 16:09:49 <nunojpg> http://pastebin.com/m2898a2a1
Jun 11 16:10:54 <juhosg> xxiao: in ath9k this is related to the dynamic power save feature
Jun 11 16:11:11 <juhosg> xxiao: the ath9k driver are trying to read/write some register while the chip is in power-save mode, and sometimes this causes errors on the PCI bus
Jun 11 16:11:29 <juhosg> xxiao: we can't create a workaround in the PCI driver of the ar71xx, imho
Jun 11 16:11:32 <xxiao> juhosg: no wonder the board resets after it's been idle for a while
Jun 11 16:12:10 <xxiao> juhosg: do we have a way to avoid that? ath9k/power-save issue
Jun 11 16:13:08 <xxiao> juhosg: is the root cause on ar71xx or ath9k, guess ath9k does not do that on other boards(e.g. ixp4xx)
Jun 11 16:18:37 <juhosg> xxiao: this might be ar71xx specific thing, although i see weird things on a gateworks board as well
Jun 11 16:23:59 <xxiao> juhosg: sigh...i ordered a few ar71xx boards(routerstation) for some field testing, now this silicon bug gets me a bit uneasy
Jun 11 16:24:48 <xxiao> juhosg: not sure if there are workarounds for madwifi/ath5k/ath9k on ar71xx boards.
Jun 11 16:25:18 <xxiao> juhosg: otherwise it's just unusable
Jun 11 16:30:59 <juhosg> xxiao: you are right, but if a driver try to access a register when that is not allowed to do, then it is a bug in the driver
Jun 11 16:31:04 <juhosg> xxiao: even if the ar71xx PCI controller has a silicon bug
Jun 11 16:33:15 <juhosg> xxiao: the ar913x has no such PCI controller, and the recent ath9k drivers are broken on this as well
Jun 11 16:33:26 <nbd> the pci core bug that i was talking about has nothing to do with powersave
Jun 11 16:33:36 <nbd> i was talking about dma issues
Jun 11 16:35:43 <xxiao> juhosg: hmm, guess Luis is aware of that power-save bug
Jun 11 16:36:15 <xxiao> nbd: will this dma only impact madwifi, or all drivers (athXk) that uses dma(which they do)
Jun 11 16:42:58 <xxiao> i think i just want to dig down and find out what really these bugs are...any pointers nbd and juhosg?
Jun 11 16:58:24 <juhosg> xxiao: it is hard to tell exactly what is causing the issues with the ps mode, i'm still fighting with it
Jun 11 17:12:53 <xxiao> juhosg: i c. but these bugs are really show stoppers, i am pulling in wireless-git to openwrt and trying to debug it too.
Jun 11 19:06:28 <fofware> hello every body, someone know how to install pepperspot?
Jun 11 19:29:25 <xxiao> you need include the package from feeds then either opkg install, or compile it in
Jun 11 19:30:09 <xxiao> ./scripts/feeds install it, make menuconfig select it, opkg install it?
Jun 11 19:52:21 <xxiao> can someone update compat-wireless to the stable 2.6.30 release...
Jun 11 19:52:58 <xxiao> " Stable compat-wireless release for 2.6.30"
Jun 11 23:32:21 <fofware> Hello, what is the diff of snapshot and 8.09.1 and why the image of 8.09.1 is more small
**** ENDING LOGGING AT Fri Jun 12 02:59:56 2009