**** BEGIN LOGGING AT Sun Jul 03 02:59:57 2011 Jul 03 03:02:30 cshore * r27368 /trunk/target/linux/x86/geos/target.mk: Jul 03 03:02:30 The Geode CS5535 and CS5536 companion chips are backward compatible with the SCx200 I2C drivers. Jul 03 03:02:30 Include this on Geos. Jul 03 03:02:30 Signed-off-by: Philip Prindeville Jul 03 03:05:22 btw: have you tried the wifi on the box that was sent to you? Jul 03 03:05:34 doh... wrong window. Jul 03 03:18:17 cshore * r27369 /trunk/ (3 files in 3 dirs): Jul 03 03:18:17 Like the earlier mfgpt patch, the new GPIO drivers also require the MFD driver to be present for the BAR's to be mapped. Jul 03 03:18:17 Signed-off-by: Philip Prindeville Jul 03 03:18:20 cshore * r27370 /trunk/ (3 files in 3 dirs): Jul 03 03:18:20 [x86] Revert "Like the earlier mfgpt patch, the new GPIO drivers also require the MFD driver to be present for the BAR's to be mapped." Jul 03 03:18:20 This reverts commit c485487b63d1c1de72a375f1278d8a98546ba3af. Jul 03 03:18:22 cshore * r27371 /trunk/target/linux/generic/ (2 files in 2 dirs): Jul 03 03:18:23 [x86] target/patches: Like the earlier mfgpt patch, the new GPIO drivers also require the MFD driver to be present for the BAR's to be mapped. Jul 03 03:18:23 Signed-off-by: Philip Prindeville Jul 03 03:18:27 cshore * r27372 /trunk/package/kernel/modules/other.mk: Jul 03 03:18:27 [x86] package/kernel/modules: Alexandros pointed me at this patch, and I noted that not only the MFGPT module but also the new GPIO driver both rely on the mfd_core functionality to operate correctly. Jul 03 03:18:27 Signed-off-by: Alexandros C. Couloumbis Jul 03 03:18:27 Signed-off-by: Philip Prindeville Jul 03 03:23:35 cshore * r27373 /trunk/target/linux/generic/ (2 files in 2 dirs): Jul 03 03:23:35 [x86] target/patches: This has been sent upstream, but might not appear for a couple of weeks. Jul 03 03:23:35 Signed-off-by: Philip Prindeville Jul 03 04:23:29 nbd * r27374 /trunk/package/swconfig/src/swlib.c: swconfig: avoid setting "NOVAL" attributes if "0" was provided as a value Jul 03 04:23:47 nbd * r27375 /trunk/package/swconfig/src/uci.c: swconfig: make reset and enable_vlan default to 1 (based on patch by Nikolay Martynov) Jul 03 09:10:00 build #41 of etrax is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/etrax/builds/41 Jul 03 14:16:15 re Jul 03 14:16:33 strange bug wr841 V7 - lan ethernet ports do not work Jul 03 14:16:40 latest svn trunl Jul 03 14:58:42 hi Jul 03 14:58:46 _trine, :) Jul 03 15:03:58 jow * r27376 /trunk/scripts/download.pl: [scripts] download.pl: cope with potential two-digit linux versions Jul 03 15:05:12 juhosg * r27377 /trunk/target/linux/ramips/files/arch/mips/ralink/rt305x/irq.c: ramips: set cp0_perfcount_irq for RT305X Jul 03 15:05:14 juhosg * r27378 /trunk/target/linux/ramips/files/arch/mips/ralink/rt305x/mach-hw550-3g.c: Jul 03 15:05:14 ramips: update HW550-3G mtd partitions Jul 03 15:05:14 This patch updates the mtd partitions for the Aztech HW550-3G Jul 03 15:05:14 Signed-off-by: Layne Edwards Jul 03 15:05:15 juhosg * r27379 /trunk/target/linux/ramips/image/Makefile: ramips: swap target check in image generation Jul 03 15:05:16 juhosg * r27380 /trunk/target/linux/ramips/base-files/lib/upgrade/platform.sh: ramips: disable watchdog on sysupgrade Jul 03 15:05:17 juhosg * r27381 /trunk/target/linux/ramips/files/arch/mips/ralink/common/gpio.c: ramips: remove unused include from the common GPIO code Jul 03 15:05:20 juhosg * r27382 /trunk/target/linux/ramips/files/drivers/net/ramips_esw.c: ramips: fix a section mismatch warning Jul 03 15:05:21 juhosg * r27383 /trunk/target/linux/ramips/files/arch/mips/ralink/ (6 files in 2 dirs): ramips: consistently use 'firmware' name for sysupgrade partitions Jul 03 15:05:23 juhosg * r27384 /trunk/target/linux/ramips/files/arch/mips/ralink/rt305x/mach-dir-300-revb.c: Jul 03 15:05:23 ramips: update partition map for DIR-300-B1 Jul 03 15:05:23 Patch by Layne Edwards from #9252. Jul 03 15:05:24 juhosg * r27385 /trunk/target/linux/ramips/patches-2.6.37/011-mtd-cfi_cmdset_0002-force-word-write.patch: ramips: force word write on flashes with AMD cmdset (#9252) Jul 03 15:05:26 juhosg * r27386 /trunk/target/linux/ramips/image/Makefile: ramips: use different partition names in the DIR-300-B1 image Jul 03 15:05:27 juhosg * r27387 /trunk/target/linux/ramips/ (files/arch/mips/ralink/rt305x/mach-wcr150gn.c image/Makefile): ramips: use lowercase partiion names on the WCR150GN board Jul 03 15:05:28 juhosg * r27388 /trunk/target/linux/ramips/image/Makefile: ramips: check the size of the uImage in image generation Jul 03 15:05:29 juhosg * r27389 /trunk/target/linux/ramips/base-files/lib/ramips.sh: ramips: add all supported boards to lib/ramips.sh Jul 03 15:05:32 juhosg * r27391 /trunk/target/linux/ramips/files/arch/mips/ralink/rt288x/devices.c: ramips: specify eeprom file name on RT288X Jul 03 15:05:33 juhosg * r27392 /trunk/target/linux/ramips/files/arch/mips/ralink/rt305x/devices.c: ramips: specify eeprom file name on RT305X Jul 03 15:05:34 juhosg * r27393 /trunk/target/linux/ramips/base-files/etc/hotplug.d/ (. firmware/ firmware/10-rt2x00-eeprom): ramips: add hotplug script to extract eeprom for the rt2x00 driver Jul 03 15:05:35 juhosg * r27394 /trunk/package/mac80211/patches/ (601-rt2x00-set_pci_mwi.patch 610-set_pci_mwi.patch): mac80211: rename a rt2x00 related patch Jul 03 15:05:37 juhosg * r27390 /trunk/target/linux/generic/files/include/linux/rt2x00_platform.h: generic: add rt2x00_platform.h Jul 03 15:05:38 juhosg * r27395 /trunk/package/mac80211/ (4 files in 2 dirs): max80211: rt2x00: load the eeprom data from a file on Ralink SoCs Jul 03 15:05:39 juhosg * r27396 /trunk/target/linux/ramips/ (12 files in 3 dirs): Jul 03 15:05:40 ramips: add 2.6.39 support Jul 03 15:05:40 Based on the patches from #9625 Jul 03 15:05:41 juhosg * r27397 /trunk/target/linux/ramips/Makefile: ramips: switch to 2.6.39.2 Jul 03 15:05:42 juhosg * r27398 /trunk/target/linux/ramips/ (patches-2.6.37/ rt288x/config-2.6.37 rt305x/config-2.6.37): ramips: remove 2.6.37 stuff Jul 03 15:05:43 juhosg * r27399 /trunk/target/linux/ramips/ (2 files in 2 dirs): ramips: merge 2.6.39 patches Jul 03 15:15:31 jow * r27400 /trunk/package/base-files/files/etc/init.d/boot: [package] base-files: pass conloglevel to syslogd (#9463, #9647) Jul 03 15:16:17 is uhttpd solely for openwrt? trying to compare it to busybox httpd but could not find a tarball for it Jul 03 15:16:33 yes Jul 03 15:17:23 need a light weight httpd to do lua/haserl/sqlite, thought it would be good if i can just use uhttpd Jul 03 15:18:03 boa is not even in openwrt, only mini_httpd, thttpd/lighttpd are too large Jul 03 15:18:32 xMff: why not just use busybox's http? Jul 03 15:19:09 because it cannot be removed from the image without altering busybox Jul 03 15:19:20 and its code is awful Jul 03 15:26:11 will try mini_httpd vs uhttpd then, wish uhttpd has a separate tarball so i can play around with it easily Jul 03 15:57:41 roh: yes, 11n aps are the wzr-hp-g300n, wndr3700, tl-1043nd, dir-825 Jul 03 16:07:46 xMff: which i guess arent buyable at all in germany? Jul 03 16:07:59 hm? Jul 03 16:08:07 I am in germany and I have all of them Jul 03 16:08:32 which chipset are they using? Jul 03 16:08:37 ar71xx Jul 03 16:08:45 i see. Jul 03 16:09:30 is any reason for not supporting the brcm N devices? or is it just 'not done yet' ? Jul 03 16:09:55 yeah there is Jul 03 16:09:59 there is no specs Jul 03 16:10:08 there are no open drivers Jul 03 16:10:28 xMff: my point is: i dont want to buy new hardware. i dont even have a single N client. but the fonera and other atheros hardware just performs _shitty_ atm with all clients (macs and thinkpads, i guess mostly iwl and brcm clients) Jul 03 16:11:05 the fonera is dead slow Jul 03 16:11:23 and "other atheros hardware" probably runs shitty firmware Jul 03 16:11:44 its only a bridge. but all my 'atheros aps' run openwrt and disassociate clients randomly Jul 03 16:11:44 anyway Jul 03 16:12:05 if you want broadcom go dd-wrt and ask there Jul 03 16:12:14 imho atheros 11n is the most reliable free software solution available at the moment Jul 03 16:12:41 the 5ghz one seems to have problems which go away with rebooting..at some point hostapd seems to not accept any clients anymore. (from the client it seems to not like the crypto) Jul 03 16:12:52 anything else currently either only works with binary-only crap or is not stable enough Jul 03 16:12:58 nbd: i see. tnx. Jul 03 16:13:35 nbd: any idea about my reliability issues with wpa-psk and the atheros aps (fonera and ar5k on a wl500gp) Jul 03 16:13:51 use madwifi Jul 03 16:14:09 ath5k is not really mature Jul 03 16:14:42 xMff: with madwifi half the clients coulnt associate at all. Jul 03 16:15:44 sigh. anyway.. thanks for the infos. seems i should see if updating the fonera helps reducing the reliability problems Jul 03 16:16:23 the fonera is using madwifi still afaik. Jul 03 16:17:19 seems to drop packets here and there which made me look into it at all, replace it with the belkin, learn that one drops all multicast (wtf?) and debug further ;) Jul 03 16:17:50 nbd: will you be at the camp? Jul 03 16:17:59 well I operate an ath9k<->ath9k wds link since two months here Jul 03 16:18:06 and I get no issues Jul 03 16:18:29 my clients are broadcom, iwl5100 an iphone and a windows netbook Jul 03 16:18:52 (those are both dual radio routers) Jul 03 16:19:07 xMff: sure. i believe that. my guess is that the client-zoo i have here combined with the old versions of madwifi and ar5k are the issue. will upgrade and see if that makes it better. Jul 03 16:19:11 I used a fonera as ap for a while Jul 03 16:19:21 but imo its too slow for anything but doing bridged ap Jul 03 16:20:11 bridge is fine. the router is a asus wl500gP, but it needs to be replaced as soon as we get cable anyhow (and do vlan switch and 5ghz ar5k ap only from then on) Jul 03 16:20:51 there's also a big difference between ath5k and ath9k, stability and devlopment wise Jul 03 16:22:10 xMff: ah. didnt know that. do they work with the same hw? Jul 03 16:22:16 no Jul 03 16:22:24 ath9k is atheros 11n stuff Jul 03 16:22:34 * roh is really not up to date with the wifi stuff anymore. too much different hw scrolling by Jul 03 16:22:38 will ath5k is for the old a/b/g radios Jul 03 16:22:43 *while Jul 03 16:23:19 ath5k is basically the fully open source, mac80211 based successor of madwifi Jul 03 16:23:32 while ath9k is a completely new driver implementation sponsored by atheros Jul 03 16:23:44 i see. are there ar9k minipci radios? / recommendable hw to do 'accesspoint with vlan capabilities'? Jul 03 16:24:07 wndr3700v2 or dir-852 if you want both 2.4ghz and 5ghz Jul 03 16:24:23 or zr-hp-g300nh if you only need 2.4 Jul 03 16:24:29 *wzr-hp-g300n Jul 03 16:24:36 it also has loads of flash and ram Jul 03 16:24:53 can be 2 devices. i need both frequencies anyhow (and moving them to different places in the room seems to help on the crowded rf we have here) Jul 03 16:25:00 they all use rtl8366 switches which support jumbo frames and up to 4096 vlans Jul 03 16:25:52 there are also minipci cards Jul 03 16:27:01 for example http://ubnt.com/sr71a Jul 03 16:27:23 heh. i wonder if it would be a good 'marketing opportunity (collecting money for openwrt) to do a 'list of currently recommended and tested with openwrt devices' ;) Jul 03 16:29:14 with current broadcom models the situation is worse, the old stuff works, but the "new generation" devices are mostly unsupported now Jul 03 16:29:22 or rather yet Jul 03 16:30:00 and even if basic platform support is sorted out, the wifi driver situation is still bad Jul 03 16:30:22 i see. Jul 03 16:30:28 b43 only recently gained rudimentary, experimental 11n support Jul 03 16:30:39 it will be certainly less stable than what you have atm Jul 03 16:31:17 the binary blob is available for kernel 2.6 now, but it does not do 11n afair (at least I've never see it doing that) and it does not work on all models Jul 03 16:31:37 though it supports the wrt54gl etc. Jul 03 16:31:55 to be fair, if it was for me i would be glad to have simple, perfectly working 802.11A and G and not care for N, but people keep nagging me. i mostly just use a cable Jul 03 16:32:05 roh: yes, i'll be at the camp Jul 03 16:32:11 also because the iwlwifi client isnt that great ;) Jul 03 16:32:25 about ath5k: i just fixed some stability issues with it yesterday and today Jul 03 16:33:15 nbd: ah. so it could be that i can get a 'reliable' 802.11g bridge with the fonera from a trunk build :) Jul 03 16:36:28 has anybody tried building something like a CI environment, involving the real hw on like a test-stand for openwrt? Jul 03 16:36:40 not yet Jul 03 16:37:46 that would be really cool. i'd like to help set something like that up, from a hardware hackers perspective ;) Jul 03 16:44:10 what is a CI environment ? Jul 03 16:44:40 continous integration ? Jul 03 16:44:43 continous integration. means building every rev, running it through tests and reporting Jul 03 16:44:46 yes Jul 03 16:45:01 ok Jul 03 16:45:23 and best case (in complex projects like this) it involves the real deal (flashing and running real hw in some kind of rig) Jul 03 16:45:35 yes Jul 03 16:45:41 its been planned for about 3 years Jul 03 16:45:55 i just never heard the BUZZword before Jul 03 16:46:15 :) Jul 03 16:47:13 i have been looking at extension cables that can be controlled via udp Jul 03 16:47:16 there is 'bitten' which integrates into trac Jul 03 16:47:41 blogic: to switch power (230V ac) ? Jul 03 16:48:26 yes Jul 03 16:48:32 so you can force reboots Jul 03 16:48:48 the problem is, that wen you flash version X and it hangs that you need an operator Jul 03 16:49:01 thats the part i can help to automate Jul 03 16:49:06 so you need to give it X grace shoots and then if all fail consider the revison borked Jul 03 16:49:13 brb my viggie burger is ready Jul 03 16:49:19 roh Jul 03 16:49:24 roh: that is 0.1% ot Jul 03 16:49:28 roh: that is 0.1% of the work Jul 03 16:49:35 * blogic cant tyoe today :D Jul 03 16:49:42 ;) Jul 03 16:51:44 sure Jul 03 17:08:22 roh: making a test setup is quite huge as a project Jul 03 17:08:39 i spent a lot of time thinking about this already Jul 03 17:09:04 i hope towards the end of the year i can get luci2 ready and start on the test system beginning of next year Jul 03 17:09:17 the important thing is that we need a good way to define test cases Jul 03 17:09:55 the actual physical setup takes a few hours to hack together Jul 03 17:10:10 doing the actual regression testing is a lot more complicated Jul 03 17:10:20 for example you need drones that you can control Jul 03 17:27:09 blogic * r27401 /trunk/package/kernel/modules/other.mk: add kmod definitions for at24(i2c) and at25(spi) eeproms Jul 03 17:40:51 blogic * r27402 /trunk/package/kernel/modules/netdevices.mk: add kmod for dm9000 Jul 03 17:41:05 blogic * r27403 /trunk/ (Config.in rules.mk): add menuconfig option to enable log files during build process Jul 03 17:41:20 blogic * r27404 /trunk/include/image.mk: extend squashfs padding for 256k flash sectors Jul 03 17:41:31 blogic * r27405 /trunk/target/linux/lantiq/Makefile: fix description for lantiq target Jul 03 17:41:44 blogic * r27406 /trunk/target/linux/lantiq/ (16 files in 5 dirs): update lantiq/falcon patches Jul 03 17:41:56 blogic * r27407 /trunk/ (include/target.mk target/linux/lantiq/Makefile): automatic subtarget enumaration Jul 03 17:42:14 blogic * r27408 /trunk/package/ltq-vmmc/ (Config.in Makefile patches/400-falcon.patch): add falcon support to vmmc Jul 03 17:42:25 blogic * r27409 /trunk/package/ (ltq-vmmc/Makefile pjsip/Makefile): fixes dependencies of pjsip and ltq-vmmc Jul 03 17:42:39 blogic * r27410 /trunk/target/linux/lantiq/patches-2.6.39/280-netif_receive_skb.patch: Jul 03 17:42:39 use netif_receive_skb on lantiq Jul 03 17:42:39 Signed-off-by: Luca Olivetti Jul 03 17:42:57 blogic * r27411 /trunk/target/linux/x86/ (alix2/target.mk geos/target.mk): Jul 03 17:42:58 alix2: add/subtract features as needed Jul 03 17:42:58 Signed-off-by: Philip Prindeville Jul 03 18:18:18 updated openwrt/upstream, https://home.comcast.net/~sdwalker/uscan/index.html Jul 03 18:20:58 build #54 of ps3 is complete: Failure [failed compile_3] Build details are at http://buildbot.openwrt.org:8010/builders/ps3/builds/54 Jul 03 18:55:33 blogic * r27412 /trunk/target/linux/at91/ (18 files in 9 dirs): Jul 03 18:55:34 AT91: Add flexibity board support Jul 03 18:55:34 The patch below adds AT91 Flexibity Connect board support. Jul 03 18:55:34 Signed-off-by: Maxim Osipov Jul 03 18:59:30 blogic * r27413 /trunk/target/linux/at91/flexibity/base-files/etc/config/system: AT91: flexibity: add system uci file Jul 03 19:23:26 blogic * r27414 /packages/utils/loop-aes/Makefile: Jul 03 19:23:26 Upgrade loop-aes to 3.6c Jul 03 19:23:26 This patch upgrades loop-aes to latest version 3.6c. This fixes #9614, Jul 03 19:23:26 however, I didn't test other platforms or other kernel version than 2.6.39.2. Jul 03 19:23:26 Signed-off-by: Michael Heimpold Jul 03 19:23:28 blogic * r27415 /packages/lang/jsonpath/ (. Makefile): (log message trimmed) Jul 03 19:23:29 New package jsonpath Jul 03 19:23:29 This patch adds a new python package called jsonpath. Jul 03 19:23:29 JSONPath allows you to run XPATH-like queries over json data structures. Jul 03 19:23:29 Changes since v1: Jul 03 19:23:29 * removed unrelated changes Jul 03 19:23:30 * fix copyright Jul 03 19:36:58 blogic * r27416 /trunk/scripts/download.pl: ignore empty lines in scripts/localmirrors Jul 03 19:40:36 blogic * r27417 /packages/multimedia/minidlna/Makefile: Jul 03 19:40:36 mark minidlna.conf as config file Jul 03 19:40:36 Signed-off-by: Mathias Kresin Jul 03 19:59:24 blogic, hey Jul 03 20:00:17 blogic, with automatic subtarget generation, how do you go about selecting the default one? or does this matter ??? Jul 03 20:02:58 RealOpty: currently only lantiq/ uses this Jul 03 20:03:10 i am playing with it and will push a fix to select a default tomoorw Jul 03 20:03:20 i have a patch for metadata.pl in the pipeline Jul 03 20:05:25 blogic, thats untrue Jul 03 20:05:36 blogic, x86 and various others Jul 03 20:05:45 i was in the process of making a patch :/ Jul 03 20:07:52 blogic, http://paste.ubuntu.com/637623/ Jul 03 20:08:02 theres the ones ive gathered so far Jul 03 20:10:08 im just bored lol :( Jul 03 21:08:05 swalker * r27418 /packages/libs/apr/ (Makefile patches/001-autoconf-compat.patch): [packages] apr: update to 1.4.5 (CVE-2011-0419, CVE-2011-1928), refresh patches Jul 03 21:08:45 swalker * r27419 /packages/libs/apr-util/Makefile: [packages] apr-util: update to 1.3.12, switch to working mirror Jul 03 21:09:16 swalker * r27420 /packages/net/bwping/Makefile: [packages] bwping: update to 1.4 Jul 03 21:10:47 swalker * r27421 /packages/net/darkstat/ (Makefile files/darkstat.init): [packages] darkstat: update to 3.0.714, use service_kill Jul 03 21:11:25 swalker * r27422 /packages/mail/fetchmail/ (Makefile patches/): [packages] fetchmail: update to 6.3.20 (CVE-2011-1947), drop obsolete patch Jul 03 21:15:27 swalker * r27423 /packages/ipv6/ipv6calc/Makefile: [packages] ipv6calc: update to 0.90.0 Jul 03 21:24:43 build #54 of orion is complete: Failure [failed compile_7] Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/54 Jul 03 21:31:18 who owns Perl these days? Jul 03 21:32:01 there are a lot of fingerprints all over it, but nothing major. Jul 03 21:38:09 build #54 of pxcab is complete: Failure [failed compile_3] Build details are at http://buildbot.openwrt.org:8010/builders/pxcab/builds/54 Jul 03 22:13:00 word from McCullough is that ocf 20110520 is required for 2.6.38 and later... http://fpaste.org/nv80/ Jul 03 23:56:16 alive, maybe Jul 04 00:51:15 swalker * r27424 /packages/libs/neon/Makefile: [packages] neon: update to 0.29.6 Jul 04 00:52:02 swalker * r27425 /packages/libs/sqlite3/Makefile: [packages] sqlite3: update to 3070701 Jul 04 00:52:36 swalker * r27426 /packages/net/lftp/Makefile: [packages] lftp: update to 4.3.1 Jul 04 00:53:14 swalker * r27427 /packages/net/nbd/Makefile: [packages] nbd: update to 2.9.22 Jul 04 00:54:12 swalker * r27428 /packages/utils/usbutils/ (Makefile patches/120-fix-zlib-support.patch): [packages] usbutils: update to 003, drop obsolete patch Jul 04 01:58:28 nbd * r27429 /branches/backfire/target/linux/generic-2.6/ (2 files in 2 dirs): kernel: backport the ssb_commit_settings export, required for the new compat-wireless version Jul 04 01:58:33 nbd * r27430 /branches/backfire/package/iw/patches/120-ibss_ht.patch: iw: backport the ibss ht patch Jul 04 01:58:41 nbd * r27431 /branches/backfire/package/mac80211/ (64 files in 3 dirs): mac80211: backport latest version from trunk as of r27395 Jul 04 02:36:46 anyone seen thepeople? **** ENDING LOGGING AT Mon Jul 04 02:59:57 2011