**** BEGIN LOGGING AT Sun May 16 02:59:57 2010 May 16 06:58:50 juhosg * r21464 /packages/lang/perl/patches/100-static_uudmap.patch: packages/perl: refresh patch May 16 06:58:54 juhosg * r21465 /packages/lang/perl/ (2 files in 2 dirs): packages/perl: fix target system endianness detection (closes #6900) May 16 11:00:38 florian * r21466 /trunk/package/acx-mac80211/ (3 files in 2 dirs): [package] update acx-mac80211 to 2010514, bringing experimental AP support May 16 11:00:52 florian * r21467 /trunk/target/linux/rdc/Makefile: [rdc] switch to 2.6.32.12 May 16 11:05:31 hauke * r21468 /trunk/package/mac80211/Makefile: mac80211: update b43-fwcutter May 16 11:07:23 hauke * r21469 /trunk/package/mac80211/patches/ (2 files): mac80211: remove patches already in mainline compat-wireless May 16 11:16:44 hauke * r21470 /trunk/package/mac80211/Makefile: May 16 11:16:44 mac80211: use stable firmware for b43 driver as the experimental causes problems on some devices with L-PHY May 16 11:16:44 This should close #6907 May 16 11:54:46 Hauke: ping May 16 11:55:04 rtz2: pong May 16 11:55:38 Hauke: about my second patch, the one with the enable/disable stuff May 16 11:55:50 Hauke: I noticed a problem with it May 16 11:56:19 rtz2: what problem? May 16 11:56:25 Hauke: the changes of the function pointers are global for all busses of this type May 16 11:56:49 Hauke: this is ok for the normal ssb bus, because there is only one May 16 11:57:44 Hauke: but, if broadcom starts to sell mini-pci cards with an ai type bus and the user has an old pci card and a new pci card in the same machine, it will blow up May 16 11:58:27 Hauke: not sure, if it's worth to handle this, or if it's better to simply wait until broadcom produces such cards (they may never do so) May 16 11:58:29 rtz2: ah yes you are right that will cause strange problems May 16 12:02:02 Hauke: so should I cook up something, or is it better to wait until broadcom actually sells such cards? May 16 12:03:02 rtz2: changing the structs like you do now is not a realy nice way in general May 16 12:04:01 Hauke: I know, but I didn't have a really clever idea how to do it May 16 12:04:28 rtz2: the two options I see are bouth not realy good May 16 12:04:32 Hauke: create another ops struct for the enable/disable ops? May 16 12:04:44 Hauke: what are your options? May 16 12:04:53 double the ssb_{pci,pcmcia,...}_ops structs or add a new struct May 16 12:05:09 a new ops May 16 12:07:02 Hauke: I would rather go for option 2 then May 16 12:07:53 Hauke: and I also have a splitted up patch for the scan stuff May 16 12:08:01 Hauke: hopefully now working May 16 12:08:09 rtz2: ok thats nice May 16 12:08:27 mb__: ping May 16 12:08:41 Hauke: pong May 16 12:08:57 Hauke: I will see what I can do about the scan functions, but as always, it may take a few days May 16 12:09:15 rtz2: ok May 16 12:10:28 mb__: did you read the patches rtz2 provided for the ssb vus to support a new ssb type? May 16 12:10:42 no May 16 12:10:54 non-broadcom SSB or what? May 16 12:11:18 mb__: it's broadcom, they did change it with new models May 16 12:11:26 mb__: brcm4716/7/8 May 16 12:11:37 ah ok May 16 12:11:49 No I did not receive anything May 16 12:12:27 Hauke: and I also have some changes planned for the mips74k support patch, so you may not want to spend too much time on it yet May 16 12:12:44 mb__: they are on the openwrt-devel list May 16 12:13:16 Oh. Ok. I did not read that for several weeks May 16 12:13:37 rtz2: What is the subject? May 16 12:14:12 mb__: BRCM4716 support part 4 May 16 12:14:58 I should have added [patch] to the subject, but I kinda forgot about it ... May 16 12:15:39 Whoa fuck a patch-patch. I'm not gonna read that. May 16 12:22:25 I can also send them as plain patches in the future, if this is better May 16 12:22:58 anyway, I have to go now May 16 12:24:06 The problem is that diffed-diffs are basically unreadable. At least if they are that big :) May 16 12:36:28 florian * r21471 /trunk/target/linux/rdc/patches-2.6.32/013-r6040_fix_mii_check_media.patch: [rdc] fix link checking with switches, only port1 of switches would report link up May 16 12:51:44 <_trine> cshore: are you here May 16 13:13:21 juhosg * r21472 /packages/net/jtg/ (. Makefile): May 16 13:13:22 new package: jtg May 16 13:13:22 add the jtg traffic generator May 16 13:13:22 Signed-off-by: Roberto Riggio May 16 13:13:27 juhosg * r21473 /packages/lang/sigscheme/ (. Makefile): (log message trimmed) May 16 13:13:27 sigscheme: new package in lang section May 16 13:13:27 this patch adds a new package in the lang section: May 16 13:13:27 http://code.google.com/p/sigscheme/ May 16 13:13:27 SigScheme is a R5RS Scheme interpreter for embedded use. It features May 16 13:13:27 small footprint (64KB in library form on the 'small' configuration) May 16 13:13:28 like SIOD and TinyScheme, low memory consumption (2-words per cons May 16 13:40:23 florian * r21474 /trunk/target/linux/rdc/patches-2.6.32/ (100-platform_support.patch 100-rdc_boards.patch): [rdc] squash platform_support.patch into rdc_boards.patch May 16 13:40:28 florian * r21475 /trunk/target/linux/rdc/ (2 files in 2 dirs): [rdc] set vendor to RDC even if CPU id is not known May 16 13:45:40 nico * r21476 /branches/backfire/ (5 files in 4 dirs): [backfire] merge [21362:21365] & [21369] May 16 13:46:58 nico * r21477 /trunk/package/ (6 files in 6 dirs): bump a bunch of packages updated after 10.03 May 16 13:50:41 nico * r21478 /branches/backfire/package/ (6 files in 6 dirs): [backfire] merge r21477 May 16 13:55:37 nico * r21479 /branches/backfire/package/ppp/Makefile: [backfire] package/ppp: bump release number May 16 14:08:38 Is there a way to specify that for a plraform (or, even better, a profile), the kernel should be fetched from a specific git? May 16 14:09:59 I know about CONFIG_KERNEL_GIT_CLONE_URI, but I would like to have git as the default for this platform. May 16 14:22:18 nico * r21480 /packages/ (13 files in 13 dirs): [packages] bump release number for packages modified after 0.03 May 16 15:15:38 hi May 16 15:16:45 @all & jow_laptop: How can I output the connectiontrack mark value with conntrack.htm or conntrack.lua ? May 16 15:17:36 I'm a bit puzzled by the SimpleForm Architecture :-), any pointers or a good way to inspect the luci.sys.net.conntrack() call? May 16 15:18:20 I hope the mark value is already parsed from conntrack input, but I'm unclear make my changes to conntrack.htm work May 16 15:22:42 uhm, that looks like a possible buggy regular expression in sys.lua :-| May 16 15:23:06 https://luci.subsignal.org/trac/browser/luci/trunk/libs/sys/luasrc/sys.lua?rev=5054#L800 May 16 15:23:32 That match can't deal with mypassword="\"lala" ... May 16 17:47:34 build #52 of etrax is complete: Failure [failed compile_4] Build details are at http://tksite.gotdns.org:8010/builders/etrax/builds/52 May 16 18:15:19 <_trine> cshore: ping May 16 18:18:09 Does somebody run the ATNGW100 board (AVR32) with OpenWRT? http://shop.embedit.de/product__590.php May 16 18:21:34 <[florian]> well yes that's how Kaloz did the original port :) May 16 18:26:00 cool May 16 18:28:08 hm, is the USB port on that board host or client (or both?) May 16 18:28:41 It has a B connector. That is rather strange. May 16 18:30:01 ah, seems to be a device controller .( May 16 18:30:01 <[florian]> client only May 16 18:30:02 :( May 16 18:31:55 Ok, that reduces usability of the whole device significantly for me May 16 18:33:21 whats state of Sheevaplug and openwrt May 16 18:34:06 <[florian]> it should work just fine May 16 18:34:38 thnx for info, still looking how to order it in EU May 16 18:34:54 <[florian]> I managed to order on on globalscale technology website May 16 18:36:08 oh nice yea i am just looking at it, now as i sow plus version i am in doubt :) May 16 18:36:47 so small device with so much power.. with maybe usb wireless card as AP router.. May 16 18:40:03 <_trine> the guruplug looks nice May 16 18:40:27 <[florian]> rtz2: ping? May 16 18:40:35 [florian]: pong May 16 18:40:48 <[florian]> rtz2: did you implement mdiobus support for r6040? May 16 18:41:24 [florian]: only my hacked up version May 16 18:41:38 <[florian]> rtz2: ok May 16 18:41:43 to bad that tey want 43$ for shiping to EU... May 16 18:41:44 [florian]: I could probably cook something up without too much effort May 16 18:42:04 <[florian]> rtz2: I am doing it right now to see if phylib can attach to the ic+ switch May 16 18:42:22 _trine yea that wan i called plus... :) May 16 18:42:56 <_trine> anyone know if cshore has gone away somewhere May 16 18:44:28 [florian]: it can May 16 18:44:38 [florian]: if you have a driver for it May 16 18:44:55 <[florian]> rtz2: seems like there is one already May 16 18:45:00 [florian]: there is one floating round somewhere May 16 18:45:04 right May 16 18:45:28 <[florian]> rtz2: this could let me remove all the ic+ specific phy init May 16 18:45:46 <[florian]> s/phy/switch/ May 16 18:48:01 [florian]: that's probably what's causing the problems on rtl switch May 16 18:48:31 [florian]: and I really hope to have the time this week to send a mail to the netdev list May 16 18:48:41 <[florian]> rtz2: that would be cool May 16 18:49:00 [florian] you know if guruplug server/plus are also fully supported May 16 18:49:21 <[florian]> crow: no idea about these, but if mainline supports it, not hard to add support for it in openwrt May 16 18:54:36 I thought there was a problem with the *plugs, that they wouldn't support master mode, so far May 16 18:54:39 Is this still current ? May 16 18:54:53 <[florian]> that's specific to the marvell topdog driver no? May 16 19:02:14 <[florian]> rtz2: my modification seems to work so far, now I need to write a better adjust link function May 16 19:12:04 <[florian]> rtz2: I managed to break eth1 however ;) May 16 19:12:14 <[florian]> rtz2: this the "WAN" port of the switch May 16 19:20:06 <[florian]> rtz2: ah ah, my bad, made a mistake between eth0 and eth1, they all work fine May 16 19:23:29 [florian]: I know, you work with the kernel a lot. Have you been successful in tellling backfire to use your kernel git repo? May 16 19:39:12 build #52 of mpc52xx is complete: Failure [failed compile_4] Build details are at http://tksite.gotdns.org:8010/builders/mpc52xx/builds/52 May 16 20:51:22 claudio * r21481 /trunk/package/kernel/modules/hwmon.mk: [kernel] hwmon: fix depends definition May 17 02:26:25 swalker * r21482 /packages/net/jtg/Makefile: [packages] jtg: put Roberto's email address inside angle brackets instead of parens May 17 02:41:42 updated openwrt/upstream, http://home.comcast.net/~sdwalker/uscan/uscan.shtml **** ENDING LOGGING AT Mon May 17 02:46:33 2010 **** BEGIN LOGGING AT Mon May 17 02:48:14 2010 **** ENDING LOGGING AT Mon May 17 02:59:57 2010