**** BEGIN LOGGING AT Sat Jul 29 02:59:56 2006 Jul 29 07:50:32 has anybody tried to get the fbvncserver kbdsim module to compile for 2.6.17? Jul 29 07:51:21 I am trying to understand the .bb file, but I think I might need a current example of how to get a 2.6 kernel module to build correctly. Jul 29 07:56:43 see the hostap-modules recipe Jul 29 08:19:25 03pH5 07org.oe.dev * r190f097c... 10/ (4 files in 2 dirs): fontconfig: split out debug info, fix for #1218 Jul 29 08:27:03 hey woglinde Jul 29 08:28:07 he koen Jul 29 08:30:22 03koen 07org.oe.dev * rb0d20942... 10/ (1 conf/distro/angstrom-2006.9.conf): angstrom: catch up with virtual/libx11 Jul 29 08:32:17 hehe Jul 29 08:35:31 BTW, thank you everyone for migrating to monotone .27+.. I just did a pull and had 20+ revs come in in ess than 5 minutes :) Jul 29 08:35:54 That used to take ~1 hour - slow machine :) Jul 29 08:58:13 hi. Jul 29 09:00:07 hi gile Jul 29 09:02:26 I tried to bitbake e-uae and recognized some problems. How can I find out the maintainer of that package? Jul 29 09:04:27 hi polyonymous Jul 29 09:04:56 hm Jul 29 09:05:56 You frightened him Jul 29 09:06:47 nono Jul 29 09:11:03 gile: I just reported a bug this morning about e-uae Jul 29 09:11:27 he zecke Jul 29 09:11:43 with a suggested e-uae_0.8.28.bb Jul 29 09:11:52 gile: http://bugs.openembedded.org/show_bug.cgi?id=1247 Jul 29 09:13:11 hi zecke Jul 29 09:13:22 koen: please do a clean rebuild :} Jul 29 09:13:22 cyrilRomain: Glad you are here. I tried your new .bb but it still fails on my machine. Jul 29 09:14:58 gile: huh :( Can you please pastbin it ? maybe can I try another help ;) Jul 29 09:15:01 zecke: a clean rebuild or what? Jul 29 09:15:12 cyrilRomain: Before I tried your new .bb I fixed that bug in 0.8.27 (some #include missing in some keymap header files) but then I came to a problem that I still have with the 0.8.28version: For some reasons the Makefile for nogui.c contains the include paths /usr/include etc for gtk. Seems to be a problem with pkg-config Jul 29 09:16:00 koen: of dbus :) Jul 29 09:16:28 gile: hm.. no idea then, sorry :-/ Jul 29 09:16:44 cyrilRomain: I dumped a "which pkg-config" and got the correct path (...staging/i686-linux/bin/pkg-config). So I really wonder how pkg-config can return the wrong pathes. Jul 29 09:17:55 zecke: including wiping tmp? Jul 29 09:18:42 hmm Jul 29 09:18:46 cyrilRomain: do yo have gtk installed on your system? Jul 29 09:18:47 too tired atm Jul 29 09:18:49 later Jul 29 09:19:26 gile: on my host system ? no Jul 29 09:20:11 cyrilRomain: so it is possible that there is the same problem on your system but pkg-config does not find gtk+-2.0.pc so no include pathes are defined. Jul 29 09:20:30 gile: possible indeed Jul 29 09:20:46 cyrilRomain: So I have to look a little bit deeper into configure... :( Jul 29 09:32:21 03koen 07org.oe.dev * r4f32b2eb... 10/ (1 packages/uae/e-uae_0.8.27.bb): e-uae: de-opiefy Jul 29 09:43:20 03koen 07org.oe.dev * r7f0b9e12... 10/ (1 conf/distro/angstrom-2006.9.conf): angstrom: stop misspelling virtual Jul 29 10:53:31 god I hate compiling stuff for debian Jul 29 10:57:27 morning Jul 29 10:57:39 hey mickeyl Jul 29 10:58:19 ~praise mtn 0.27 Jul 29 10:58:23 All hail mtn 0.27! Jul 29 11:07:40 morning mickeyl Jul 29 11:10:24 * mickeyl reads http://www.oesf.org/forums/index.php?showtopic=16669&st=30 Jul 29 11:10:58 good to see the majority on the GPL side Jul 29 11:11:19 some consumer-only folks are still complaining, but i guess that's natural Jul 29 12:42:43 mickeyl: good morning Jul 29 12:47:42 03koen 07org.oe.dev * r5c52d2dc... 10/ (1 conf/distro/angstrom-2006.9.conf): angstrom: prefer a newer dbus Jul 29 13:40:43 I seem to be getting an error when running bitbake-myimage today (I updated my checkout this morning) Jul 29 13:40:47 genext2fs: couldn't allocate a block (no free space) Jul 29 13:40:53 ERROR: function image_cmd failed Jul 29 13:41:11 thats right after it does: bbimage -n stv-image-mini_itx-20060729140054 -t ext2 -e /home/tom/stv/oe-stv/stv/packages/stv/stv-image.bb Jul 29 14:02:07 is there a miximum size limit for bbimage or something? Jul 29 14:02:27 I have two packages - both are quite large Jul 29 14:02:51 I can create an image with one, or the other, but not both at the same time Jul 29 14:03:41 iirc the ext2 image stuff defaults to 64MB Jul 29 14:07:42 koen: is there a way top make it larger? Jul 29 14:08:45 03koen 07org.oe.packaged-staging * r8ad77752... 10/ (3 files in 2 dirs): Jul 29 14:08:45 classes/package_ipk.bbclass: create directories before trying to write files into them Jul 29 14:08:45 classes/packaged-staging.bbclass : various Jul 29 14:08:45 * turn off staging for native and cross packages Jul 29 14:08:45 * create directories before trying to write files into them Jul 29 14:08:46 * blacklist packages trying to poke at staging *and* cross Jul 29 14:08:48 * symlink staging/usr to staging/ Jul 29 14:10:01 tomkp: yes, but don't ask me Jul 29 14:29:29 cyrilRomain: I found the bug in e-uae! Jul 29 14:38:21 gile: excellent :) Jul 29 14:38:55 cyrilRomain: it is a problem in sdl.m4; so it is not a e-uae only problem. Jul 29 14:39:40 cyrilRomain: sdl.m4 modifies the global PATH vairable. It adds $prefix/bin and $prefix/usr/bin in front of the PATH. Jul 29 14:40:18 cyrilRomain: in the special case of e-uae the wrong pkg-config binary is later used to identify gtk. But it may lead to any kind of problem in a cross compiling environment for every sdl package... Jul 29 14:40:42 gile: good to know Jul 29 14:41:36 gile: now someone should check that this correction does not breaks other build ... Jul 29 14:41:37 cyrilRomain: ye, but difficult to solve. sdl.m4 is unfortunatly used in quite a lot of packages and most packages use a local copy of it. Jul 29 14:41:54 gile: yes :-/ Jul 29 14:42:33 cyrilRomain: what is the typical way to solve this in oe? A patch for each impacted package? Jul 29 14:44:08 I don't know. Personnaly, I would first check (building from start) that other packages build correctly. If not, you should ask here Jul 29 14:44:18 But I'm not the right person to answer such question Jul 29 14:44:30 cyrilRomain: who is? Jul 29 14:44:32 Or you may also file the bug Jul 29 14:44:41 cyrilRomain: of course I will. Jul 29 14:44:44 koen, mickeyl ? Jul 29 14:45:21 yes file the bug so that many developper would be aware of that Jul 29 14:45:28 s/would/will Jul 29 14:46:05 koen, mickeyl: Have you read my talk with cyrilRomain? How are problems solved with .m4 files which are often used as local copies in many packages (like sdl.m4)? Jul 29 14:48:22 gile: patch the hell out of those apps Jul 29 14:50:21 koen: And the "master" sdl.m4 which is published with libsdl-qpe? Maybe some package use that one and rely on the bug. Jul 29 14:51:07 that's certainly not the master m4 file Jul 29 14:51:16 libsdl-x11 has the master m4 file Jul 29 14:51:41 not that 'we want framebuffer' whining qt/e monstrocity Jul 29 14:56:17 03koen 07org.oe.dev * r88968185... 10/ (1 packages/uae/e-uae_0.8.27.bb packages/uae/e-uae_0.8.28.bb): e-uae: replace .27 with .28 Jul 29 15:06:59 who was having trouble building freetype? Jul 29 15:08:16 Why it is forbidden to manipulate the home directory from a package? (That was meantioned in the old e-uae-0.8.27.bb files) Jul 29 15:08:56 how do you know which user is going to run it? Jul 29 15:09:09 and do you want your handcrafted stuff to get overwritten? Jul 29 15:09:33 a package NEVER EVER touches a homedir Jul 29 15:09:41 03koen 07org.oe.dev * r261f8a41... 10/ (1 packages/freetype/freetype_2.2.1.bb): freetype 2.2.1: add HACK to work around libtool breakage Jul 29 15:09:56 koen: It understand the first argument but the second problem also applies to /etc, doesn't it? Jul 29 15:10:19 gile: /etc is guarded by the CONFFILES globs in OE Jul 29 15:10:26 so ipkg will know it's a config file Jul 29 15:10:59 koen: thanx, I understand. Sound like OE is similar to gentoo, right? Jul 29 15:11:05 give me one solid argument why a package should touch a homedir Jul 29 15:11:32 gile: most of the syntax evolved from the .ebuild format Jul 29 15:14:11 koen: for now my last question before I create the sdl.m4 patch for e-uae (and others): how to I find out the maintainer of a package? Is there a dedicated maintainer per package? Jul 29 15:14:44 see the MAINTAINER field inside the .bb Jul 29 15:14:56 is there is no such field there isn't a maintainer Jul 29 15:17:49 koen: thanx. Jul 29 15:37:40 Finally got xserver-kdrive installed :) but it's giving me segfaults whenever I quit any X appllication Jul 29 15:38:07 when I fire up Xfbdev, I get a couple of errors: "Could not init font path element /usr/lib/X11/fonts/misc, removing from list!" Jul 29 15:38:23 and I have a full backtrace here: http://rafb.net/paste/results/VDThht84.html Jul 29 15:39:06 are there some additional components required to use xserver-kdrive? Jul 29 15:41:27 might be worth mentioning that the crash itself seems to be font related Jul 29 15:45:28 03mickeyl 07org.oe.dev * rf013fa49... 10/ (4 files in 2 dirs): libsdl all install sdl.m4 into aclocal include dir Jul 29 15:49:33 tomkp: the font errors should be harmless Jul 29 15:49:38 I get them, but X comes up Jul 29 15:49:50 ('should' -> I'm not sure) Jul 29 15:51:27 03koen 07org.oe.dev * ra3d194ba... 10/ (6 files in 2 dirs): gpe-edit: add .31 and remove old versions Jul 29 15:53:15 tomkp, koen: The font path message says that the given path does not exist. X removes it from the list automatically. Normally this is harmless. You may change xorg.conf to avoid the message in future. Jul 29 15:53:35 gile: kdrive doesn't use xorg.conf Jul 29 15:54:16 tomkp: do you have a font installed? Jul 29 15:54:35 koen: I haven't explecitly installed one Jul 29 15:55:07 our segfault bt certainly looks related to the font issue Jul 29 15:56:01 as I understand it, kdrive only supports bitmap fonts... I'm just looking for some now Jul 29 15:57:17 are any of the packages in xorg-font suitable for use with kdrive? Jul 29 15:57:48 try 'bitbake ttf-dejavu' and install the packages it generated Jul 29 15:58:16 koen: ok... I was just trying bitbake font-adobe-100dpi Jul 29 15:58:32 which semms to fail due to missing dependency (virtual/libx11-native) Jul 29 15:58:50 hmmm Jul 29 15:58:54 that looks like a bug Jul 29 15:59:03 hmm Jul 29 15:59:04 could you add it to the bugtracker? Jul 29 15:59:05 ERROR: dependency virtual/libx11-native (for mkfontscale-native) not satisfied Jul 29 15:59:07 sure Jul 29 16:07:30 well, I installed the ttf-dejavu package, but still the same error Jul 29 16:07:42 hrm Jul 29 16:07:50 although that package installed the fonts into /usr/share/fonts/truetype/ Jul 29 16:08:10 which is not where kdrive appears to be looking Jul 29 16:08:41 I tried creating the three missing directories and copying the fonts in to each of them... Jul 29 16:08:51 still made no difference Jul 29 16:09:13 are there some bitmap font packages? Jul 29 16:09:36 03koen 07org.oe.dev * r0bb59343... 10/ (1 packages/xorg-app/mkfontscale-native_X11R7.0-1.0.1.bb): xorg-app/mkfontscale-native_X11R7.0-1.0.1.bb: no need for a virtual/libx11-native Jul 29 16:09:51 ahh.. beat me to it ;) Jul 29 16:15:18 hello Jul 29 16:35:43 * gile recognizes that OE also shares the long compiling times with gentoo after updating the package tree... Jul 29 16:41:18 it does Jul 29 16:41:34 but fortunately I'm targetting a CPU with a sane instruction set :) Jul 29 16:55:02 know about this error when building xserver-xorg: checking for /usr/share/X11/sgml/defs.ent... configure: error: cannot check for file existence when cross compiling? Jul 29 16:59:54 03koen 07org.oe.dev * r43f8fac6... 10/ (6 files in 2 dirs): virtual/libx11: also RPROVIDE it Jul 29 17:00:51 RP: any news on pxa270 cpu frequency scaling or the fb speed hacks? Jul 29 17:05:10 bleh Jul 29 17:05:14 * koen can't spell Jul 29 17:06:27 03koen 07org.oe.dev * rad624098... 10/ (1 conf/machine/netbook-pro.conf): netbook-pro.conf: prefer xserver-kdrive-epson Jul 29 17:06:32 03koen 07org.oe.dev * r1262fff8... 10/ (1 conf/distro/angstrom-2006.9.conf): Jul 29 17:06:32 angstrom-2006.9: dbus 0.90 doesn't include glib bindings, so we'll use Jul 29 17:06:32 0.62 for now Jul 29 17:06:37 03koen 07org.oe.dev * r7ab1a5ee... 10/ (3 files in 2 dirs): task-angstrom, angstrom-gpe-image: catch up with the viertual/libx11 thing Jul 29 17:16:43 03koen 07org.oe.dev * rf43647a6... 10/ (1 conf/machine/netbook-pro.conf): netbook-pro.conf: create initrd directory Jul 29 17:24:52 RP: I went over poky svn changes to merge in bits, but I'm to stupid understand some changes, so I'll leave those up to you Jul 29 17:24:58 03koen 07org.oe.dev * rc1b4a880... 10/ (1 conf/machine/include/zaurus-clamshell.conf): zaurus-clamshell.conf: add working jffs2 summary support. (contributed by o-hand.com) Jul 29 17:28:49 hi folks - is it normal for 'bitbake -s' to take 5 minutes to run every time? is there a way to speed it up? Jul 29 17:30:48 i expected that merely returning the preferred versions of each item in the db (as the -s switch is supposed to do) would be much quicker - say on the order of a few seconds at most Jul 29 17:32:21 koen, rp, kerwood, anybody have an idea about this? im using 512 megs of ram so it shouldnt be an issue of resources Jul 29 17:33:31 poushag, your parsing more than 3k files Jul 29 17:33:44 and bitbake is single threaded Jul 29 17:36:37 03coredump 07org.oe.oz354x * rf23370d0... 10/ (3 files in 3 dirs): oz-bootsplash: Do a chvt 1 at the end of runlevel 2 Jul 29 17:36:38 03koen 07org.oe.dev * ra2037f5a... 10/ (1 packages/angstrom/angstrom-gpe-image.bb): angstrom-gpe-image: zap root passwd when DISTRO_TYPE = "release" Jul 29 17:36:43 03coredump 07org.oe.dev * r04f4f8d1... 10/ (4 files in 4 dirs): altboot: Fix poodle-2.6.conf Jul 29 17:36:52 how much ram is the newest bb using these days? Jul 29 17:38:03 about 40MB according to top Jul 29 17:39:01 much better than once upon a time Jul 29 17:39:04 CosmicPenguin: did you manage to get your command line version of imgloader onto yr site? Jul 29 17:54:07 ok - so, /win 4 Jul 29 17:54:09 oops :) Jul 29 17:54:18 so, i've tracked down the crash Jul 29 17:55:06 its due to the fact that we have a null function pointer inside the function FontFileOpenFont from libxfont Jul 29 17:55:34 the funciton pointer is *scalable->renderer->OpenScalable Jul 29 17:55:52 which is obviously related to scalable fonts which are disabled in kdrive afaik Jul 29 17:56:40 so the question is, should this code be sanity checking the pointer before calling it or is there a reason why its getting to this point in the first place when we dont have scalable fonts? Jul 29 17:57:59 so, reading the code further Jul 29 17:58:00 jkp_: I'd say it should not crash but print something like 'no scalable fonts' Jul 29 17:58:06 koen: right Jul 29 17:58:21 jkp_: but I'll leave it to the X people to say if X should exit or continue Jul 29 17:58:31 what its doing is trying to locate the fixed size font, then falling back on scalable fonts if it cant find it Jul 29 17:58:44 which begs the question: should the font be on the system somewhere? Jul 29 17:58:59 the font its looking for is "-misc-fixed-medium-r-semicondensed--0-0-75-75-c-0-iso8859-1�9�" Jul 29 18:00:55 what happens if you add a check for the null pointer and allow it to continue? Jul 29 18:01:19 thats what im thinking for a patch to get it to work Jul 29 18:01:21 I know that xfree86 used to exit if it didn't find the fixed font Jul 29 18:01:34 but that was years ago Jul 29 18:02:01 but isnt the fixed font in this case going to be integral to the working of X? Jul 29 18:02:08 but yeah, i will try this now Jul 29 18:02:58 it X *really* needs that font, it should be in RDEPENDS Jul 29 18:03:54 s/it/if/ Jul 29 18:03:54 agreed Jul 29 18:04:03 is there a package for that font? Jul 29 18:04:17 * koen looks Jul 29 18:04:20 we couldnt see anything...all ttf only Jul 29 18:06:11 there is some stuff in xorg-font Jul 29 18:06:43 try font-bitstream-100dpi Jul 29 18:06:51 wait Jul 29 18:06:58 font-bitstream-75dpi Jul 29 18:07:30 if the '75-75' means 75 dpi, that is Jul 29 18:09:57 koen: that package DEPENDS on xserver-xorg... which is failing to build with a cross compiling error: Jul 29 18:10:02 checking for /usr/share/X11/sgml/defs.ent... configure: error: cannot check for file existence when cross compiling Jul 29 18:10:19 try with -b Jul 29 18:11:12 any ideas what package builds libXfont? i see no .bb for it - i was about to write a patch but it seems that there is already something patching the file Jul 29 18:11:25 i want to see what it is - maybe that pacth isnt right Jul 29 18:11:57 jkp, libxfont1 ? Jul 29 18:12:09 sorry man Jul 29 18:12:16 wrong captilization Jul 29 18:12:25 my bad Jul 29 18:12:28 packages/xorg-lib/libxfont_X11R7.1-1.2.0.bb Jul 29 18:12:31 polyonymous_: thanks dude Jul 29 18:12:44 no prob, I've just `ipkg search`-ed ;) Jul 29 18:15:39 koen: -b worked, thanks... Jul 29 18:16:09 does that font really need to DEDPEND on xserver-xorg? Jul 29 18:16:21 tomkp: if it works with -b, it seems that DEPENDS is a bit misplaced Jul 29 18:17:12 well it works... as in, it built the package, although I can't install it on the target due to loads of missing dependencies Jul 29 18:17:25 but I guess that's to be expected Jul 29 18:57:03 03koen 07org.oe.packaged-staging * r5ec3350c... 10/ (1 classes/base.bbclass): classes/base.bbclass: do not call do_stage twice Jul 29 18:57:07 03koen 07org.oe.packaged-staging * r2019b73c... 10/ (1 classes/base.bbclass): classes/base.bbclass: do_install, not install Jul 29 18:57:11 03koen 07org.oe.packaged-staging * rdb433176... 10/ (1 classes/package_ipk.bbclass): classes/package_ipk.bbclass: fix thinko which effectively disabled packaged-staging Jul 29 18:58:23 * chouimat wonders why all the linux distributions that got released this last 2 years SUCK!!! Jul 29 18:58:44 chouimat, cmon, caos is pretty neat Jul 29 18:58:58 caos? Jul 29 18:59:03 lol Jul 29 18:59:09 i guess you missed that one Jul 29 18:59:25 its a distributed Linux distro Jul 29 18:59:32 pretty cool Jul 29 18:59:48 especially if you use mobile devices Jul 29 19:00:12 sound like I will simply rebuild a nice gentoo or a linuxfromscratch on thhis computer Jul 29 19:00:14 and like to use a server/workstation processing power occasionally Jul 29 19:00:37 well LFS is faster than gentoo Jul 29 19:00:48 from src -> usable Jul 29 19:00:49 emte: I know gentoo so ... Jul 29 19:01:15 sec i'll grab you the caos link Jul 29 19:01:54 emte: I'm really pissed off ... last week all the distro I tried got in my way so ... Jul 29 19:02:08 http://caos.caosity.org/ Jul 29 19:02:13 in what respect? Jul 29 19:02:56 tried kubuntu on my ppc ... can't compile anything gcc suck on it, I'm trying to replace suse 10.1 (nice piece of crap) on my x86 ... Jul 29 19:03:45 ah Jul 29 19:03:58 maybe I will go opensolaris ;) Jul 29 19:04:02 lol Jul 29 19:04:10 give caos a spin Jul 29 19:04:35 emte: I'm looking at it Jul 29 19:04:44 if you like it on one system you can leverage the processing power of any other systems you install it on Jul 29 19:05:01 kind of like adaptive clustring Jul 29 19:10:59 03koen 07org.oe.packaged-staging * r1aa312b4... 10/ (1 classes/packaged-staging.bbclass): classes/packaged-staging.bbclass: fix detection logic Jul 29 19:22:59 so, it seems that kdrive is broken then - its not finding the default font when it tries to look it up Jul 29 19:23:15 my patch stops it segfaulting but now it quits saying it cant find the default font Jul 29 19:23:21 the font its looking for is "-misc-fixed-medium-r-semicondensed--0-0-75-75-c-0-iso8859-1�9�" Jul 29 19:23:30 any ideas where this would come from? Jul 29 19:23:41 or what actual filename that maps to? Jul 29 19:28:32 so we'll need the misc font package then, but its broken :/ Jul 29 19:48:10 hello everyone Jul 29 19:53:48 nothing like the sound of a doritos bag ... to help find where the cat is hidding Jul 29 20:01:22 hey Greg2 Jul 29 20:02:36 hi koen Jul 29 20:02:46 hi Greg2 Jul 29 20:02:56 hi CoreDump|home Jul 29 20:07:27 koen: did you see in the irc logs my message about your Angstrom-gpe-image-test-poodle.rootfs.jffs2.bin size? Jul 29 20:07:42 Greg2: yes Jul 29 20:07:56 ok Jul 29 20:08:00 Greg2: right now we're including a lot of debug utils, so it might max out the flash Jul 29 20:09:20 koen: where would be a proper place for me to reduce the size for my test image? Jul 29 20:11:56 Greg2: it's 21MB max, right? Jul 29 20:12:09 more like 20MB Jul 29 20:14:14 let me see how big an angstrom-gpe-image is Jul 29 20:14:21 ok, so it's font-alias that DEPENDS on xserver-xorg... I dont think it's necessary since it built fine without it Jul 29 20:18:46 CoreDump|home: are you Herr Hentges? -> from oesf :) Jul 29 20:18:49 Hallo Herr Hentges :-) Jul 29 20:19:29 yep Jul 29 20:19:46 03koen 07org.oe.dev * r7ca686a6... 10/ (1 packages/xorg-font/font-alias_X11R7.0-1.0.1.bb): font-alias: depend on virtual/xserver instead of xserver-xorg Jul 29 20:20:09 tomkp: that should help a bit with the font troubles Jul 29 20:21:01 CoreDump|home: I know, my great granddad was from Germany :) Jul 29 20:21:28 =) Jul 29 20:21:38 nice one koen Jul 29 20:21:40 Ausweis bitte Jul 29 20:26:03 koen: my document would be my name... Shifler Jul 29 20:26:25 17M deploy/images/poodle/Angstrom-gpe-image-test-20060729-poodle.rootfs.jffs2.bin Jul 29 20:26:35 including strace and gdb :) Jul 29 20:27:19 tomkp: could you attach the patch you made for kdrive to the bugtracker? Jul 29 20:34:55 03koen 07org.oe.dev * r9b52874c... 10/ (1 packages/linux/linux-efika_2.6.16.bb): linux-efika: set compatible machine Jul 29 20:44:38 koen: http://bugs.openembedded.org/show_bug.cgi?id=1250 Jul 29 20:47:55 koen: I'm a bit slow from the heat here... thanks Jul 29 20:48:56 it's at http://www.angstrom-distribution.org/unstable/images/20060729/poodle/ , but I'm pretty sure X will be pretty unstable Jul 29 20:49:15 ok, thanks Jul 29 20:55:41 03koen 07org.oe.dev * r8512ea49... 10/ (3 files in 3 dirs): dbus 0.62: 'fix' segfault using patch from dbus 0.90 Jul 29 21:03:32 I created a patch for that sdl.m4 problem. Is it enough to attach it to the bug report or should I mail it to someone / mailing list? Jul 29 21:55:02 goodnight Jul 29 23:26:27 03pH5 07org.oe.dev * rc1912994... 10/ (16 files in 6 dirs): dbus, dbus-glib, dbus-python: update to 0.91/0.71 Jul 30 00:31:13 does anyone know if there is a bb package that provides the switchto command for programatically switching vt's? Jul 30 01:42:41 emte, thx for answering me earlier - i was called away just after i posted that and havent been able to get back til now - so you are saying that a 5 min (bitbake -s) response is typical and theres no way to speed it up? Jul 30 01:48:17 anybody else who knows the answer, feel free to chime in here Jul 30 01:51:22 poushag: I think emte is right: 3000 .bb file to parse takes naturally quite many time. You may ask zecke to confirm it can not speed it up though Jul 30 01:52:36 well i think these files need to be hashed or put into a speedier db format or something - i have seen equivalent amounts of data parsed much more quickly Jul 30 01:52:50 poushag: IIRC, they're cached after they're parsed the first time Jul 30 01:52:54 Look under tmp/cache Jul 30 01:53:13 NAbyss: exactly, thank you Jul 30 01:53:35 yeah they are cached but it seems to only speed things up from 15 minutes to 5 Jul 30 01:53:50 5 mins? On what kind of hardware? Jul 30 01:53:55 Mine takes about 20 sec at most. Jul 30 01:54:02 poushag: then something's wrong Jul 30 01:54:13 mine has 512 mb of ram Jul 30 01:54:14 That's on an AMD running at 1466MHz.. Jul 30 01:54:46 and i think a 2.something ghz p4 Jul 30 01:55:05 poushag: Even at peak usage, python takes about 90mb on my box. I don't think RAM's going to be the problem Jul 30 01:55:08 well thats why im asking - i suspect something is wrong Jul 30 01:55:27 and i need help to figure out what it is Jul 30 01:55:28 It floats at about 50mb during parsing Jul 30 01:55:37 poushag: Local filesystem? Jul 30 01:55:58 checking... Jul 30 01:56:31 poushag: you shoudl eliminate the usual causes of slowness, like NFS, mounted sync, vfat fs, etc. Jul 30 01:56:43 poushag: I can understand being slow if it's NFS.. Jul 30 01:56:53 ext3 Jul 30 01:57:02 Also, run top.. see if anything else's stealing CPU. Jul 30 01:57:25 poushag: are you running root with an active chroot? Jul 30 01:57:49 im running bitbake as regular user Jul 30 01:57:55 hmmm Jul 30 01:58:42 the cpu seems pretty much idle Jul 30 01:59:02 poushag: Can you run 'time bitbake foo', and see if it's spent in system/user/io? Jul 30 01:59:18 ok Jul 30 01:59:36 here on AMD MP 1500+, ~15 secondes to parses the files/cache, ~2 minutes to get the output of 'bitbake -s' Jul 30 02:00:02 I'm no bitbake developer.. but I get the feeling the issue is with your setup rather than BB.. it shouldn't take 5 min on a 2GHz box Jul 30 02:00:37 its running - will prob be 5 minutes before i can post results Jul 30 02:06:19 real 3m39.261s; user 2m41.891s; sys 0m16.292s Jul 30 02:14:35 nabyss - i ran just 'bitbake -s' and posted the results **** ENDING LOGGING AT Sun Jul 30 02:59:56 2006