**** BEGIN LOGGING AT Fri Aug 05 23:59:56 2005 Aug 06 02:28:46 good morning all Aug 06 02:30:42 koen: morning Aug 06 02:41:36 good morning Aug 06 03:51:32 * LordVan is away: off Aug 06 03:53:19 morning Aug 06 04:13:47 good morning boys and girls Aug 06 04:14:01 is there any libsigc++1.2 expert in da house? Aug 06 04:17:33 hi mickeyl, I have a problem compiling the kernel 2.6.13-rc4-mm1-r0 for openzaurus/c7x0 Aug 06 04:17:59 tosa-2.6.13-r2.patch fails to apply, because linux-2.6.12/drivers/input/touchscreen/wm97xx.c doesn't exist Aug 06 04:18:53 tosa.c patches ok with an offset, but the other file simply isn't there in a 2.6.12 kernel, maybe the mm1 patch is incomplete? Aug 06 04:20:06 i think tosa-2.6.13-r2 is simply broken Aug 06 04:20:17 try to use -r1 for the time being Aug 06 04:20:48 k, thx Aug 06 04:21:15 otherwise, just wait for do13 to get online again and pester him :) Aug 06 04:21:30 k Aug 06 04:38:56 incinerator: wait 5 minutes for a patch from me before bothering to compile that - I have an important fix coming Aug 06 04:39:08 okies Aug 06 04:39:25 has compiled already though ggg Aug 06 04:41:20 Anyone with tinderbox knowledge around? http://ewi546.ewi.utwente.nl/OE_qa/OpenEmbeddedBuild/status.html <- long green columns in nslu2-linux stuff, rather than the little 15 minute delta builds I was expecting to see ... Aug 06 04:42:45 we do a rebuild (without any cleaning or deleting) every 15 minutes. If nothing has changed, then each of Unlsung, OpenSlug and UcSlugC will take less than 15 minutes to build. So there should be white gaps every fifteen minutes on each of those columns. Aug 06 04:43:37 hi Aug 06 04:44:34 rwhitby: iirc, that is expected tinderbox behaviour. Aug 06 04:45:58 So much for 2.6.13-rc4-mm1 being at fault this time. Looks like the patch I've submitted to akpm has a touchscreen bug that makes it run at 1/64th speed :-/ Aug 06 04:46:15 pb_: if you look around 22:00 UTC here it is different: http://ewi546.ewi.utwente.nl/OE_qa/cgi-bin//tinder.cgi?tree=OpenEmbeddedBuild&start-time=1123306333&display-hours=48 Aug 06 04:49:06 incinerator: That should make that kernel work nicely ;-) Aug 06 04:49:49 03rpurdie 07org.oe.dev * r0e3cf43c... 10/packages/linux/linux-openzaurus_2.6.13-rc4-mm1.bb: linux-oz-2.6: Fix a bug causing the touchscreen to run at 1/64th speed. Make the tosa patches apply until Dirk can fix properly. Aug 06 04:49:52 k, pulling right now.... Aug 06 04:50:59 rwhitby: mm, though the gaps are bigger there. There is a certain minimum threshold that you have to reach for tinderbox to treat it as a real gap. Aug 06 04:51:28 any idea what that gap is? Aug 06 04:51:53 if you don't know it off the top of your head, I'll go searching ... Aug 06 04:52:10 I think it's configurable in the tinderbox config files. I don't remember offhand, might be 15 or 30 minutes. Aug 06 04:52:27 ok, I'll ask koen when I see him next Aug 06 04:52:40 righto Aug 06 04:52:53 pb_: thx Aug 06 04:53:18 that said, zecke and koen are using tinderbox2, which is a bit different to the one that I've used before. So I guess it's possible that all this stuff is different. Aug 06 04:53:33 RP the r2 tosa patch still fails, r1 applies Aug 06 04:53:48 but I'll delete my tmp dir to make sure... Aug 06 05:01:47 03nail 07org.oe.nslu2-linux * r3a06ef82... 10/conf/distro/openslug-packages.conf: Add udev Aug 06 05:03:21 *sigh* Aug 06 05:03:21 | /local/pkg/oe/spitz/tmp/staging/arm-linux/include/sigc++-1.2/sigc++/marshal.h:58: Fehler: Instanziierung von »SigC::Marshal::value_« als Typ »void« Aug 06 05:03:21 | /local/pkg/oe/spitz/tmp/staging/arm-linux/include/sigc++-1.2/sigc++/marshal.h:58: Fehler: »SigC::Marshal::value_« hat unvollständigen Typen Aug 06 05:03:37 03mickeyl 07org.oe.dev * ra85c8bb5... 10/packages/libsigc++-1.2/ (3 files in 2 dirs): add libsigc++-1.2.7 (1.2) Aug 06 05:04:49 ~lart sigc++ Aug 06 05:04:49 * ibot hooks into a hydrant and hoses sigc++ down Aug 06 05:07:32 * mickeyl gives up Aug 06 05:07:40 my c++ skillz are not 31337 enough Aug 06 05:22:09 03coredump 07org.oe.dev * r505f6e00... 10/packages/altboot/altboot.bb: altboot: Add pkg_postinst_spitz, courtesy of JustinP @ #openzaurus. Aug 06 05:41:44 * mickeyl writes to sigc++ mailinglist to cry for help Aug 06 05:54:22 RP: how did they b0rke the ts this time? Aug 06 06:21:45 pb_: ping Aug 06 06:22:02 mickeyl: at your service Aug 06 06:22:45 pb_: hi. when i activate the PARALLEL_INSTALL_MODULES (or so), how does RDEPENDS work then? do the modules PROVIDE the short name or do I have to adjust the RDEPENDS? Aug 06 06:23:01 mickeyl: they RPROVIDE the original name Aug 06 06:23:05 excellent. thanks Aug 06 06:23:25 oh one more question Aug 06 06:23:44 at FOSDEM on your husky you showed me an alternative launcher. is that in GPE/OE yet? Aug 06 06:24:09 I guess that was probably gpe-appmgr. I think it is in OE. Aug 06 06:24:17 ah cool. i'll try that. thanks Aug 06 06:24:27 iirc, it might even be the default launcher for corgi and simpad these days. Aug 06 06:24:43 not sure about that though Aug 06 06:24:44 oh cool. it should be the default for spitz and akita as well then Aug 06 06:24:53 i'll have a look Aug 06 06:29:50 03mickeyl 07org.oe.dev * rebc68fd9... 10/conf/distro/openzaurus.conf: openzaurus.conf: enable extended kernel module naming (PARALLEL_MODULES_INSTALL) Aug 06 06:39:21 woot! someone is working on sa1100 udc gadget driver Aug 06 06:39:28 can't believe it Aug 06 06:39:28 :D Aug 06 06:44:59 heh Aug 06 06:45:00 awesome Aug 06 06:45:05 who is the crazy individual? Aug 06 06:45:29 Nick Bane from cecomputing.co.uk Aug 06 06:53:30 how do I add postinst scripts to packages containing a dot in their name? Aug 06 06:53:39 pkg_postinst_uim-gtk2.0 () results in a parser error Aug 06 06:55:16 heh, that's unfortunate Aug 06 06:55:22 I guess that is a parser bug. Aug 06 07:11:14 the magic __config_regexp__ in bitbake's ConfHandler.py doesn't match the line 'pkg_postinst_uim-gtk2.0 () {' Aug 06 07:17:49 This looks interesting (re: monotone merging): http://lists.gnu.org/archive/html/monotone-devel/2005-08/msg00037.html Aug 06 08:32:38 lo Aug 06 10:16:47 * chouimat is back. Aug 06 10:16:58 hey chouimat Aug 06 10:42:53 03mickeyl 07org.oe.dev * ra09349c6... 10/ (4 files in 3 dirs): update gpe-irc to 0.08, makefile patch no longer necessary Aug 06 11:25:10 03ccsmart 07org.oe.nslu2-linux * r27caeec0... 10/packages/postfix/ (files/main.cf_2.0 postfix_2.0.20.bb): Polishing maildir based install defaults. Aug 06 11:25:14 03ccsmart 07org.oe.nslu2-linux * r68aa7712... 10/packages/cyrus-imapd/ (cyrus-imapd_2.2.12.bb files/cyrus.conf_2.2): Fixed paths for lmtp and notify. Aug 06 12:18:41 * koen watches monotone grind through a delete Aug 06 12:20:01 03koen 07org.oe.dev * ra94e401d... 10/packages/gstreamer/gpe-gst-plugins_0.8.4.bb: Remove gpe-gst-plugins Aug 06 12:20:17 ah, very good Aug 06 12:20:32 I always disliked that package. Aug 06 12:20:56 it got in the way of a succesfull 'build gpe*' Aug 06 12:23:12 pb_: heh. me too. glad it's gone now :) Aug 06 12:42:26 aargh Aug 06 12:42:34 minimo fails on a different point now Aug 06 13:32:26 mickey|party: It was me who broke the ts ;-) Aug 06 14:07:14 Is anyone actually working on udev? Aug 06 14:18:03 NAiL: not that I know off Aug 06 14:18:46 hmm Aug 06 14:39:21 03mickeyl 07org.oe.dev * rbc710581... 10/packages/gpe-appmgr/ (4 files in 2 dirs): Aug 06 14:39:22 patch gpe-appmgr to compile against the latest cairo Aug 06 14:39:22 remove a couple of older gpe-appmgr Aug 06 14:48:11 aaargh Aug 06 14:48:27 * NAiL tries to disapprove his last revision Aug 06 14:48:32 but it doesn't work Aug 06 14:49:22 mickey|party: don't forget to check in the patch too ;) Aug 06 14:58:40 * NAiL rolls over Aug 06 14:58:48 * NAiL disapproved one revision Aug 06 14:58:55 * NAiL waits for CIA Aug 06 14:59:01 03nail 07org.oe.nslu2-linux * rb8873177... 10/ (3 files in 3 dirs): Added watchdog package Aug 06 14:59:05 03nail 07org.oe.nslu2-linux * raa7c47af... 10/ (3 files in 3 dirs): disapproval of revision b887317745e76c913d221d462248b9555eaf7c4d Aug 06 14:59:09 03nail 07org.oe.nslu2-linux * rae3124f8... 10/ (3 files in 3 dirs): disapproval of revision aa7c47af4a1d20ac50212af0be57a0ab4b597184 Aug 06 14:59:13 03nail 07org.oe.nslu2-linux * rcaf4fea9... 10/ (3 files in 3 dirs): disapproval of revision ae3124f810e521f4a710f11319d143e188e3fee4 Aug 06 14:59:17 03nail 07org.oe.nslu2-linux * r233263c0... 10/packages/watchdog/watchdog_5.2.bb: Added watchdog package. Without adding anything else this time Aug 06 14:59:21 03nail 07org.oe.nslu2-linux * r30035fbe... 10/ (2 files in 2 dirs): Add watchdog to openslug Aug 06 14:59:45 I wonder how on earth monotone figured it'd disapprove the disapproval of the first disapproval Aug 06 15:47:46 There are graphical apps to interface with ipkgs, right? Is there a way to get a message past to the user with then? Aug 06 15:59:58 NAiL: I'd expect them to relay messages Aug 06 16:00:10 especially the "config file altered" stuff Aug 06 16:00:26 * koen always uses ipkg cmdline Aug 06 16:00:39 There's not exactly a gui on my slug either ;) Aug 06 16:01:17 usb hub + usb vga + usb keyboard + usb mouse Aug 06 16:01:36 yes, then I'd have myself a nice, slow desktop Aug 06 16:01:51 I'd still run cli on it though :P Aug 06 16:02:00 :) Aug 06 16:02:12 I don't think I've actually run X for a couple of years Aug 06 16:02:27 I just realise that my most used osX app is the terminal Aug 06 16:02:51 * NAiL admits that the most used app on his PC is putty Aug 06 16:04:00 putty, firefox, thunderbird and winamp. That accounts for probably 96% of my program usage Aug 06 16:56:53 03mickeyl 07org.oe.dev * rfc974e96... 10/packages/gpe-appmgr/files/cairo-fixes.patch: gpe-appmgr: add cairo patch Aug 06 16:58:58 'night all Aug 06 17:11:23 03jbowler 07org.oe.nslu2-linux * r51b6958c... 10/conf/distro/ucslugc.conf: Aug 06 17:11:23 UcSlugC: make (significantly) smaller by removing the kernel from the root Aug 06 17:11:23 file system - the size of the rootfs is reduced by 35% of the previous Aug 06 17:11:23 size! Aug 06 18:01:19 hmm Aug 06 18:01:32 guess its finally time to go to #edevelop Aug 06 18:01:46 unless ... Aug 06 18:01:55 hmm Aug 06 22:20:28 raster, me bannana ... you here? Aug 06 22:24:57 damn ... Aug 06 22:25:36 i cant figure out how make an evas canvas through an ecore server connection Aug 06 22:26:12 mostly because i dont know what to put for a size parameter Aug 06 22:26:39 in the ecore_con_server_send() which i am guessing i need to use Aug 06 22:27:24 i am hoping there is a simpler way than me trying to figure out how to write the "handler" Aug 06 22:28:34 maybe i need to look for an ipc in ecore_evas Aug 06 22:56:11 eh? Aug 06 22:56:20 why are u playign with ecore_con and evas? Aug 06 22:56:23 they are not related Aug 06 22:57:43 i am aware Aug 06 22:58:08 because i am trying to establish a canvas through the ecore server using a client Aug 06 23:01:31 ecore_ipc_server_send() seems to be almost the exact same finction Aug 06 23:01:47 both requireing an sozeof Aug 06 23:01:51 sizeof* Aug 06 23:02:28 so you are making a canvas server? Aug 06 23:02:39 ie some process runs and creates a canvas Aug 06 23:02:43 and allows clients to access it Aug 06 23:02:45 via ipc Aug 06 23:03:03 yes taht is the ised Aug 06 23:03:05 idea Aug 06 23:03:54 ok Aug 06 23:04:03 well u need to expand on a ipc protcol Aug 06 23:04:12 the size param is the sioze of the data chunk in bytes Aug 06 23:04:20 if data is NULL nothng will be sent int he data Aug 06 23:04:33 e3core_ipc unlike ecore_con guarantees delivery of WHOLE ipc messages Aug 06 23:04:34 yeah ... hmmm Aug 06 23:04:37 no matter how large Aug 06 23:04:39 in 1 lump Aug 06 23:04:51 so u dont get a message/event saying there is a message until its all delivered Aug 06 23:04:58 okay Aug 06 23:05:09 ecore_con does no such guaranteeing Aug 06 23:05:20 k Aug 06 23:05:22 it will tell you how many bytes of raw data u got and what they were - when u get them Aug 06 23:06:45 now how to do this ... i was hoping ecore_evas might have a present ... oh well Aug 06 23:07:38 no Aug 06 23:07:42 there was a project in cvs Aug 06 23:07:43 evoak Aug 06 23:07:48 yeah i know Aug 06 23:07:50 it dfoes a canavs server and client-side library Aug 06 23:07:56 you told me not to use it remember Aug 06 23:08:03 i gave up as its too much work (for now) Aug 06 23:08:14 u are stumbling into a zone of a lot of work Aug 06 23:08:33 yeah .. Aug 06 23:09:14 doesnt make much sense to do it any other way tho Aug 06 23:09:40 it would just be the same as any other wm/env Aug 06 23:10:44 time to print the ipc functions i guess Aug 06 23:12:41 the ipc fn's are there to help u build an ipc protocol on top Aug 06 23:12:55 u still have to do a lot of encode/decode on top Aug 06 23:13:07 yeah, need to be able to read them tho Aug 06 23:13:19 ecore_ipc will encode/decode soem basic values like min/major number reply to, error, etc. Aug 06 23:14:28 so for some things u can avoid using the data stuff]but the data field si for u to attach arbitary datas to extend Aug 06 23:14:41 but then u have to handle encoding/decoding that Aug 06 23:15:17 yeah Aug 06 23:15:44 to be expected Aug 06 23:16:02 i am glad i make my life easy ... Aug 06 23:22:06 time for a beer i think ... **** ENDING LOGGING AT Sat Aug 06 23:59:57 2005