**** BEGIN LOGGING AT Fri Jan 22 02:59:57 2010 Jan 22 05:21:38 build #22 of atheros-trunk is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/atheros-trunk/builds/22 Jan 22 08:12:23 nico * r19270 /trunk/ (Config.in rules.mk): add a config item to specify arguments to strip Jan 22 08:18:06 nico * r19271 /trunk/package/kernel/modules/ (8 files): [package] kernel: specify kmod packages dependencies in a uniform way Jan 22 10:55:05 florian * r19272 /trunk/toolchain/uClibc/patches-0.9.30.2/700-mips_nptl_dl_find_hash_fix.patch: [toolchain] fix uclibc for mips by adding missing upstream hunks Jan 22 12:41:44 damn, multicast is an annoying search term. it shows up in every ifconfig :( Jan 22 12:42:06 few days ago there was someone who couldn't receive multicast on his 3port switch of the rspro. Jan 22 12:42:26 I'm seeing the same issue, changed the mac of the bridge to the mac of the wired (it took the one of the wifi NIC), but that didn't help Jan 22 12:42:38 is this a known problem to you guys? Jan 22 12:44:48 hm, the rspro switch seems to have issues in general Jan 22 12:45:04 there's also some guy in the forum who can't get traffic forwarded if wan is bridged to the two lan ports Jan 22 12:46:21 ic. well, I am about 99% sure it has worked before - I am running miniupnpd for my Xbox, and I have seen ports being added in iptables Jan 22 12:46:37 I'll try with 2.6.31 and 2.6.32, see if that helps anything Jan 22 14:57:44 jow_laptop: can you close #5988? Jan 22 14:58:00 build #21 of au1000-trunk is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/au1000-trunk/builds/21 Jan 22 15:01:45 jow_laptop: #5692 as well with r19167 Jan 22 15:05:59 nbd: can you close #5652? see r17522 Jan 22 15:10:10 oO Jan 22 15:10:10 done Jan 22 15:10:16 "invalid action resolve" Jan 22 15:10:21 i just closed it ;) Jan 22 15:10:27 I see Jan 22 15:10:38 didn't know if you were there ;) Jan 22 15:11:22 * jow_laptop neither Jan 22 15:11:37 doing sql the whole day, it is starting to drive me crazy Jan 22 15:11:51 heh Jan 22 15:11:53 yeah, sql can be annoying Jan 22 15:12:31 esp. if the usual queries are 400-500 lines long with joins across a dozen tables :-/ Jan 22 15:35:19 build #21 of brcm47xx-trunk is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/brcm47xx-trunk/builds/21 Jan 22 15:53:15 nbd: close #3531? see r17478 Jan 22 15:55:09 closed Jan 22 16:08:54 someone want to take a look at the patch on #5308? Jan 22 16:11:03 I'll look into it Jan 22 16:16:07 jow * r19273 /trunk/package/broadcom-wl/files/lib/wifi/broadcom.sh: [package] broadcom-wl: detect all radios when generating config, do wireless setup per-interface (#5308) Jan 22 16:17:22 #4753 looks to be a dupe of #2421 Jan 22 16:19:24 so does #6521 Jan 22 16:20:00 yes Jan 22 16:20:19 this probably needs to be addressed per-subtarget/profile Jan 22 16:32:02 what whould be the best way to stop an already running wpa_supplicant in wpa_supplicant.sh ? Jan 22 16:32:25 because once wpa_supplicant is running, I'm getting an error when running wifi Jan 22 16:38:45 something like http://openwrt.pastebin.ca/1761707 ? or put it inside an if, together with the executing of the new wpa_supplicant instance ? Jan 22 16:57:07 ah, there's something to kill wpa_supplicant in disable_mac80211, but it checks for phyX in /proc/$pid/cmdline while it should look for wlanX Jan 22 17:05:04 #6442 looks to have been a dupe of #4128 Jan 22 17:18:28 nbd: is the patch on #5528 wanted currently? Jan 22 17:23:44 the patch as is probably won't work Jan 22 17:23:48 since we're using a newer hostapd Jan 22 17:24:04 but if somebody tests this and gets it working, why not Jan 22 17:24:28 i have now started to work on my hostapd/wpa_supplicant replacement Jan 22 17:24:58 stintel: https://dev.openwrt.org/ticket/5991? Jan 22 17:26:41 swalker: oh, that seems like it could fix my problem Jan 22 17:28:07 but, it will no longer kill hostapd I think, since hostapd cmdline does contain phyX rather than wlanX Jan 22 17:37:17 swalker: I would suggest http://openwrt.pastebin.ca/1761779 Jan 22 17:37:42 juhosg * r19274 /trunk/target/linux/ar71xx/files/drivers/net/phy/rtl8366_smi.c: rtl8366_smi: store device pointer instead of plaform_device Jan 22 17:37:49 juhosg * r19275 /trunk/target/linux/ar71xx/files/drivers/net/phy/rtl8366_smi.c: rtl8366_smi: store GPIO numbers directly in the rtl8366_smi struct Jan 22 17:37:51 juhosg * r19276 /trunk/target/linux/ar71xx/files/drivers/net/phy/rtl8366_smi.c: rtl8366_smi: separate gpio initialization Jan 22 17:37:56 juhosg * r19277 /trunk/target/linux/ar71xx/files/drivers/net/phy/rtl8366_smi.c: rtl8366_smi: move chip specific stuff into its own struct Jan 22 17:38:00 juhosg * r19278 /trunk/target/linux/ar71xx/files/drivers/net/phy/rtl8366_smi.c: rtl8366_smi: rename rtl8366s specific functions and definitions Jan 22 17:38:08 juhosg * r19279 /trunk/target/linux/ar71xx/ (17 files in 10 dirs): rtl8366_smi: move RTL8366S driver into a separate file Jan 22 17:39:37 but it will still kill hostapd I think Jan 22 17:40:06 maybe better to wait on what nbd is working on :-) Jan 22 17:40:13 it'll take a while Jan 22 17:40:32 hostapd usually gets killed when i run wifi Jan 22 17:40:38 haven't noticed any problems there Jan 22 17:45:26 [florian]: ping Jan 22 18:01:06 lars * r19280 /trunk/target/linux/xburst/config-2.6.32: [xburst] Enable fireware loading support Jan 22 18:13:04 lars * r19281 /trunk/target/linux/xburst/files-2.6.32/ (6 files in 3 dirs): [xburst] Cleanup clock module a bit and replace last users of __cpm_* Jan 22 18:30:23 nbd: by the way, did you find the time to take a look at the phy stuff? Jan 22 18:31:22 not yet Jan 22 18:31:42 ok Jan 22 18:31:57 no idea when i'm going to have time for it Jan 22 18:32:01 my todo list is way too long at the moment Jan 22 18:33:07 nbd: no problem, but I may ask again sometime next week :) Jan 22 20:51:25 could someone help me? i need to change the acktimeout but i can't find a way to do it Jan 22 20:52:35 i've tried with wl0_distance but it seems not to be working Jan 22 20:53:13 i also found that i could do that with dctrl command Jan 22 20:53:24 but i just can't find a place to download it Jan 22 20:54:57 i really need to do that change, please help me Jan 22 21:22:59 i think this is the place where i can receive help, isn't it? Jan 22 21:24:05 which openwrt version are you running? Jan 22 21:25:08 right now i'm using white russian Jan 22 21:26:04 dunno about whiterussian (so long obsolete that i forgot everything about it), but in 8.09.2 it's easy to configure Jan 22 21:26:40 in 8.09.2 it's 'option distance ' in the wifi-device section of /etc/config/wireless Jan 22 21:26:53 but there's a problem with wrt54gl in that version Jan 22 21:26:56 ? Jan 22 21:28:05 b43 for wrt54gl doesn't work with 8.09.2 Jan 22 21:28:22 but i think the same instruction in 8.09 works Jan 22 21:28:50 all 8.09 based releases offer both linux 2.4 and 2.6 based firmware for broadcom devices Jan 22 21:28:59 8.09.2 as well Jan 22 21:29:09 so if you flash the linux 2.4 based firmware, you get working wifi Jan 22 21:29:45 and one last question, i would like to try a non ack link Jan 22 21:29:56 what should i do? Jan 22 21:30:06 why do you want that? Jan 22 21:30:27 getting rid of ack handling breaks rate control and retransmissions Jan 22 21:31:36 it's just for experimental purposes Jan 22 21:31:51 i'd like to see the impact of no ack Jan 22 21:32:07 we played with it at some point Jan 22 21:32:09 but dropped support for it Jan 22 21:32:13 because there's not much point Jan 22 21:33:22 so i can't try that option Jan 22 21:33:26 right Jan 22 21:35:00 and have you tried to implement a block ack? Jan 22 21:36:27 that's something else i'd like to try Jan 22 21:37:52 nbd thank you very much for your help Jan 22 21:38:16 and for all your effort in openwrt Jan 22 21:38:26 you're welcome Jan 22 21:38:41 it's something we really appreciate Jan 22 21:38:44 take care Jan 22 21:38:58 bye Jan 22 21:39:03 bye Jan 22 22:06:06 build #21 of infineon-trunk is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/infineon-trunk/builds/21 Jan 22 22:43:13 <[Fate]> nbd: ping :) Jan 22 22:45:47 [Fate]: pong Jan 22 22:46:00 <[Fate]> nbd: https://dev.openwrt.org/ticket/6551 trivially ok? Jan 22 22:46:42 ping rtz2 & xMff Jan 22 22:46:54 [Fate]: while the file is being changed, imho find should be replaced by $(FIND) Jan 22 22:46:59 then it should work properly Jan 22 22:47:25 <[Fate]> but that wouldn't fix the "spaces in filenames" problem? Jan 22 22:47:40 no, i mean in addition to your patch Jan 22 22:47:48 i'll do both Jan 22 22:47:50 your patch looks fine Jan 22 22:48:08 <[Fate]> it looked so trivial I thought I'm just not seeing something... Jan 22 22:48:26 just mentioning it, because -print0 might not be portable to all systems and $(FIND) forces the build system to use the GNU version Jan 22 22:49:00 <[Fate]> ah ok Jan 22 22:49:16 <[Fate]> I just noticed it because my openwrt-based VDR uses some channel logos with spaces in their filenames Jan 22 22:50:03 ah Jan 22 22:50:08 will commit the patch in a minute Jan 22 22:50:53 <[Fate]> okthx Jan 22 22:52:33 damn...broke my patch Jan 22 23:10:58 nbd * r19282 /trunk/include/image.mk: fix rootfs preparation with rootfs containing files with whitespaces in the filename (based on patch from #6551) Jan 23 02:17:59 * cshore imagines pitching his computer in the snowbank and cackles in glee Jan 23 02:22:53 No init found. What...it worked yesterday .... grrr.... **** ENDING LOGGING AT Sat Jan 23 02:59:57 2010