**** BEGIN LOGGING AT Mon May 31 02:59:56 2010 May 31 03:37:45 xMff: your suggestion was just what the doctor ordered. May 31 03:37:55 thanks May 31 06:02:03 good morning May 31 07:58:18 why some timeserver dont respond and if they respond why isnt sync posible? http://openwrt.pastebin.com/SXx49M28, and actualy timeserver didnt picked up timezone setup in /e/c/system which is: option timezone CET-1CEST,M3.5.0,M10.5.0/3 May 31 07:58:39 ups wrong channel sorry it was for #openwrt May 31 08:39:00 i tried today kernel 2.6.33.4 on brcm47xx all seems work just wirelees not, and its atheros supported by ath5k, more info here lsmod,dmesg, and ath5k opkg install which fail and thus is also what should be(was just checking if there are missing dependencys) http://openwrt.pastebin.com/dNTWwB8A May 31 08:39:40 tried madwifi? May 31 08:40:16 no as long time someone wrote ath*k is future :) May 31 08:40:29 anyway May 31 08:40:35 you need to build the kmods yourself May 31 08:40:43 the one from the snapshot will not work May 31 08:40:57 unless ypu use the snapshot's kernel May 31 08:41:09 xMff i did, its included in kernel, thus i tried opkg install which fail and there is version info May 31 08:41:34 it was just for info., ath5k module seems loaded May 31 08:41:46 ah ok May 31 08:41:54 well looks like you need madwifi here May 31 08:42:18 xMff line 186, and 199, are they OK? May 31 08:42:23 yes May 31 08:42:38 xMff i am using backfire svn with ath5k worked ok (problem was two cliend==wifi went down) May 31 08:42:55 then it did not work okay ;) May 31 08:43:29 well yea not right ok, anyway i have just one wireless client, second was just a try to see if it went down. May 31 08:44:15 xMff its Atheros AR5213A chipset which should be supported by ath5k May 31 08:44:30 i swaped broadcom wifi with cm9 one, (asus wl-500gpv1) May 31 08:46:34 this one http://www.wneweb.com.tw/Networking/more/CM9-GP.htm May 31 09:08:54 [florian]: ping May 31 09:12:18 <[florian]> rtz2: pong May 31 09:19:09 [florian]: I did think about the wdt driver again May 31 09:19:13 [florian]: I really don't think it's worth the effort May 31 09:19:13 [florian]: it's not possible to the a timeout longer then 5 seconds, so you need some kind of kernel thread for it anyway May 31 09:19:13 [florian]: and in the case you might as well safe the effort of configuring a different timeout and keep it as it is know May 31 09:20:07 sorry, got disconnected again ... May 31 09:20:49 you should get some ssh login where you can host irssi in a screen :) May 31 09:21:58 xMff: I need proper dsl instead of this umts crap :( May 31 09:22:14 xMff: but I will take a look at the irssi option, thanks for the tip May 31 09:22:35 <[florian]> rtz2: a software timer is just fine to prevent the watchdog from expiring May 31 09:22:46 <[florian]> rtz2: also the driver can notice the watchdog daemon in case the timeout is too big May 31 09:24:35 [florian]: right, but in this there isn't anything to configure in the wdt register, and so doing a read-modify-write is kinda useless May 31 09:26:41 this case May 31 09:28:21 <[florian]> rtz2: unless the hardware looses it's setting, you should not rewrite the complete register value, but just change the relevants bits May 31 09:29:45 [florian]: hmmmm, ok May 31 09:35:13 <[florian]> rtz2: anyway, the hardware is just bogus, so let's leave it as-si May 31 09:36:24 [florian]: ok May 31 10:18:43 xMff madwifi works with kernel 2.6.33.4, but i needed to add more config to wirelles file, http://openwrt.pastebin.com/yGdTWrtv May 31 10:31:35 actualy it doesnt work, i mean driver are loaded, wifi setuped but client cant connect May 31 10:39:57 ok its my fauls, sorry for alarm, i forgot to write one more charachter in password :) May 31 10:40:18 xMff so madwifi works with 2.6.33.4 and two clients without wifi daying.. May 31 11:57:35 blogic * r21643 /trunk/target/linux/ramips/files/arch/mips/include/asm/mach-ralink/dev-gpio-buttons.h: May 31 11:57:35 Fix ramips build when CONFIG_RALINK_DEV_GPIO_BUTTONS is not set. May 31 11:57:35 Signed-off-by: Helmut Schaa May 31 12:17:44 kerneis * r21644 /packages/ipv6/ahcpd/ (Makefile files/ahcp.sh): May 31 12:17:45 [package] ahcpd: Fix server mode (broken script) May 31 12:17:45 Signed-off-by: Gabriel Kerneis May 31 13:10:56 build #63 of brcm47xx is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/brcm47xx/builds/63 May 31 13:32:41 [florian]: you still here? May 31 13:37:23 <[florian]> rtz2: yes May 31 13:38:35 [florian]: about ar7, there is an eth1 interface which doesn't seem to be connected to anything on most devices May 31 13:39:02 [florian]: is this simply a second mac like on rdc, which is unused, or this there a more complex story to it? May 31 13:40:14 <[florian]> rtz2: usually this is how it goes: May 31 13:40:33 <[florian]> rtz2: the ar7 device is a dsl router, so designers connect eth0 to a switch and voila May 31 13:40:45 <[florian]> rtz2: but it can happen they pick up a different reference design and use eth1 instead May 31 13:40:51 <[florian]> rtz2: but there are two separate mac like on rdc May 31 13:41:50 [florian]: ahh, ok thanks May 31 13:41:58 [florian]: and one other question :) May 31 13:42:13 [florian]: the rdc southbridge patch I did send you May 31 13:42:35 [florian]: do you still plan to use it, or do you want to keep the mfd device the way it is now? May 31 13:43:39 <[florian]> rtz2: I would like to keep the mfd device like it is, sorry if I did change my mind multiple times about it, but I prefer keeping it as-is May 31 13:43:48 <[florian]> rtz2: even though you might find it a bit over engineered ;) May 31 13:46:31 [florian]: hmmm, ok May 31 13:46:34 if you really think so May 31 14:16:16 build #61 of sibyte is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/sibyte/builds/61 May 31 15:21:28 [florian]: dead? May 31 15:29:24 <[florian]> Weedy: nope May 31 15:36:54 <[florian]> Weedy: what can I do for you? May 31 17:26:00 [florian]: you know who has a tew652brp? May 31 17:26:09 i can't seem to flash it with mtd May 31 17:32:28 <[florian]> Weedy: nope, is that an ar71xx device? May 31 17:38:28 <_trine> Weedy, use the recovery console May 31 20:15:49 build #58 of ep93xx is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/ep93xx/builds/58 May 31 21:26:45 _trine: it doesn't like the sig May 31 21:26:56 i had to hexedit the .bin May 31 21:27:07 [florian]: yes ar71xx **** ENDING LOGGING AT Tue Jun 01 02:59:57 2010