**** BEGIN LOGGING AT Wed Dec 31 02:59:58 2008 Dec 31 12:30:36 hello everebody Dec 31 12:30:42 hello Dec 31 12:30:52 i would like to ask you something Dec 31 12:31:02 should we send patches to devel mailing list or not? Dec 31 12:31:42 most of the patches i have sent are ignored Dec 31 12:31:58 no comments, no nothing Dec 31 13:55:17 lars * r13781 /packages/Xorg/xorg/app/smproxy/ (. Makefile): Add smproxy package. Dec 31 14:07:56 <[florian]> Acinonyx: sed them to trac Dec 31 14:08:18 <[florian]> Acinonyx: most people do not read the mailing-list not use git-am to pickup the patches Dec 31 14:52:14 nbd * r13782 /trunk/scripts/feeds: Dec 31 14:52:14 scripts/feeds update: Dec 31 14:52:14 * Fix feed update via git. Dec 31 14:52:14 * Add support for bzr and hg. Dec 31 14:52:14 * Proper escaping of command line arguments handed to system. Dec 31 14:52:15 * Reduced code duplication. Dec 31 14:52:17 Patch by Justus Winter Dec 31 14:52:26 nbd * r13783 /trunk/README: change readme to point to scripts/flashing/flash.sh instead of scripts/flash.sh Dec 31 14:52:42 nbd * r13784 /branches/8.09/scripts/feeds: merge r13782 to 8.09 Dec 31 14:59:34 lars * r13785 /packages/libs/libusb/Makefile: Fix libusb install. Dec 31 15:02:27 lars * r13786 /packages/lang/telepathy-python/Makefile: Update telepathy-python to 0.15.3 Dec 31 15:07:23 lars * r13787 /packages/libs/libpam/Makefile: Fix libpam install. Dec 31 18:30:11 how does that "feeds" script work? Dec 31 18:30:18 mb@homer:~/develop/svn/openwrt_moko/trunk$ ./scripts/feeds install enlightenment Dec 31 18:30:20 Ignoring feed 'packages' - index missing Dec 31 18:30:21 Ignoring feed 'xwrt' - index missing Dec 31 18:30:23 Ignoring feed 'luci' - index missing Dec 31 18:30:24 WARNING: No feed for package 'enlightenment' found, maybe it's already part of the standard packages? Dec 31 18:30:36 mb_: run ./scripts/feeds update before Dec 31 18:30:43 like with apt-get or opkg Dec 31 18:31:27 ah ok :) Dec 31 19:02:06 nbd * r13788 /trunk/package/firewall/files/ (firewall.config uci_firewall.sh): disable the MSS fixup hack by default (most ISPs don't require this as a workaround for MTU problems, only some do). this should give a nice speedup for routing on standard-compliant ISPs Dec 31 19:02:59 nbd * r13789 /branches/8.09/package/firewall/files/ (firewall.config uci_firewall.sh): merge r13788 to 8.09 Dec 31 19:05:33 larsc: When using the feeds script it works fine. So I guess I'll just use that :) Dec 31 19:07:39 mb_: ok :) Dec 31 19:09:29 agb * r13790 /trunk/ (3 files in 3 dirs): [package] hostapd,madwifi: rename agmode option to hwmode. 11a and 11g aren't the only allowed values Dec 31 19:35:44 mirko: Thanks a lot for that e17 script. That helped a lot to reduce the pain ;) Dec 31 19:36:46 But I think we should really figure out a way to compile that stuff as part of the toolchain automatically. Dec 31 19:37:17 anybody working on that, by chance? :) Dec 31 19:43:21 mb_: its not that easy since edje_cc needs almost all of the e17 libraries. so we either have to compile all of them for the host manually or implement support for native packages. Dec 31 19:45:43 larsc: yeah I know. But I see no general problem with that. Simply treat edje_cc like gcc and compile it, including all its deps, in the toolchain step. Dec 31 19:47:25 This could also be autodetected. So if edje_cc is available, skip that expensive step. Dec 31 19:48:07 But on a modern CPU it's not that expensive. It took about 5to10 minutes Dec 31 19:49:08 But yeah, that probably needs another type of packag definitions, that are compiled for the builshost Dec 31 19:49:12 buildhost Dec 31 19:49:34 mb_: yep, thats the problem Dec 31 19:50:00 how's gcc handled? is that somehow hardwired? Dec 31 19:53:46 the tools which are build as part of the toolchain are hardwired in toolchain/Makefile Dec 31 19:57:31 mirko: with the patch I just sent to the list e17 starts on the moko :) Dec 31 19:57:57 The cursor has major breakage, but it basically works :) Dec 31 20:02:22 it has x and y swapped :D Dec 31 20:02:48 hey, that's fun :D Dec 31 20:03:51 huh? with openwrt kernel? Dec 31 20:04:08 yeah Dec 31 20:04:25 the moko touchpad is swapped in x and y Dec 31 20:04:34 so if you move in y, the cursor moves in x Dec 31 20:04:55 I guess you can swap that in xorg.conf Dec 31 20:06:36 well with the openmoko kernel its working correctly, so maybe it should be fixed in the driver so it will work with both kernels. Dec 31 20:07:13 yeah well. I _guess_ they swap it in the X config Dec 31 20:07:28 do you have an X config from moko? Dec 31 20:09:07 no, but i have openwrt userland running on my neo with a openmoko kernel and its working without any special config Dec 31 20:09:24 larsc: i had the same problem withg the cursor but using the moko kernel Dec 31 20:09:40 larsc: which WM are you using? Dec 31 20:09:41 i am rebuilding everythin right now to see if that helps Dec 31 20:10:27 mb_: e17 but i;m using kdrive maybe thats the reason Dec 31 20:10:41 what's kdrive? Dec 31 20:11:44 oh framebuffer stuff Dec 31 20:11:51 xserver Dec 31 20:12:45 its using the same codebase as Xorg but doesn't support module loading so everything is build in Dec 31 20:13:08 I'm running xglamo Dec 31 20:13:22 thats a kdrive Dec 31 20:13:51 yeah Dec 31 20:15:09 are you using kdrive-xfbdev? Dec 31 20:15:18 xglamo Dec 31 20:15:42 ok, well. Doesn't work for me :) Dec 31 20:16:01 Are you also running 2.6.26? Dec 31 20:16:30 not sure Dec 31 20:17:57 2.6.24 Dec 31 20:18:21 hm, so maybe the touchscreen driver is simply broken Dec 31 20:20:55 I'll check for new kernel patches in the moko tree. Dec 31 20:21:04 Maybe not now :) **** ENDING LOGGING AT Thu Jan 01 02:59:57 2009