**** BEGIN LOGGING AT Sat Apr 29 09:59:57 2006 Apr 29 10:01:44 * koen is of to buy some groceries Apr 29 10:19:50 XorA|gone: NOTE: package sylpheed-2.2.4-r5: task do_build: completed Apr 29 10:20:40 Ok, I've synced a few fixes with bitbake. I guess I should look at some OE bugs now :) Apr 29 10:27:46 woohoo! ssh to lude via WPA2-protected wireless :-) Apr 29 10:28:06 (using madwifi-ng and wpa-supplicant from OE - commits to follow) Apr 29 10:28:53 rwhitby: nice :) Apr 29 10:29:12 Hmm - not yet stable - ssh session hung after a while :-( Apr 29 10:29:31 :-( Apr 29 10:29:32 03eFfeM 07org.oe.dev * rec447375... 10/packages/pvrusb2-mci/pvrusb2-mci_20060423.bb: pvrusb2-mci_20060423.bb: added HOMEPAGE in bb file Apr 29 10:29:37 03eFfeM 07org.oe.dev * rb4533330... 10/packages/pwc/ (pwc-10.0.12-rc1/Makefile.patch pwc_10.0.12-rc1.bb): pwc: added pwc_10.0.12-rc1 release, fixed build Apr 29 10:39:12 RP: I think I'll put it as slugos-only for the time being, until it is stable at least, just in case it affects other drivers. Apr 29 10:40:31 rwhitby: that will sort of garantee nobody else will look at it :( Apr 29 10:41:56 ok, here's the thing: it's the non-default version of wpa-supplicant (0.5.1, marked default_pref=-1), and it depends on a new madwifi-ng, which no-one else is using yet anyway) Apr 29 10:43:19 rwhitby: Don't make it slugos only if its the experimental version :) Apr 29 10:43:44 rwhitby: We can't fork everything there some small risk - if there are problems we can worry about it then Apr 29 10:43:45 hey eFfeM Apr 29 10:44:55 RP: the defconfig for wpa-supplicant is shared with previous versions, so I will need to fork that into a PV-specific files dir Apr 29 10:45:12 hi koen, thought I'd stop by in oe again Apr 29 10:45:16 but I agree that I shouldn't fork just for slugos Apr 29 10:45:22 rwhitby: That's fair enough Apr 29 10:45:24 just made pwc work again Apr 29 10:46:17 eFfeM: did you see http://www.nl.bol.com/is-bin/INTERSHOP.enfinity/eCS/Store/nl/-/EUR/BOL_DisplayProductInformation-Start?BOL_OWNER_ID=1003004004232493&Section=SOFTWARE ? Apr 29 10:47:35 nope and for some reason my router is doing odd things with me again Apr 29 10:47:38 will check later Apr 29 10:49:09 RP: ok, I worked out how to do it without forking the defconfig - I just append to it from the configure stanza in the experimental bb Apr 29 10:49:57 rwhitby: neat :). I wouldn't worry about files against specific versions too much though - there is a lot of precident for that already Apr 29 10:50:07 rwhitby: if defconfig is changed internally between versions (0.4.8<>0.5.1) I would use separate file Apr 29 10:50:18 so the only slugos-specific file is the defaults file for the ath0 vs eth1 and madwifi driver vs wext Apr 29 10:51:09 hrw: looks like 0.4.7 uses defconfig (openssl) and 0.5.1 uses defconfig-gnutls Apr 29 10:51:38 yep Apr 29 10:51:47 0.5.1 is able to use gnutls 1.3.x Apr 29 10:51:56 oooh, multiple tarball stashes Apr 29 10:52:31 hrw: cheers for that Apr 29 10:52:46 ok. time to eat breakfast and go food shopping Apr 29 10:52:54 XorA|gone: will put into upgrades later Apr 29 10:53:04 XorA|gone: thx for package Apr 29 10:57:19 bleh Apr 29 10:57:25 udevd doesn't start in angstrom Apr 29 10:58:19 hrw: I pointed Pavel at the rc1 images - seem to work for him on spitz Apr 29 10:59:26 kool Apr 29 11:00:27 koen: saw the cams, that is quite inexpensive. have a pair of pwc680k myself so no real need for this though Apr 29 11:02:02 RP: later during day (or tomorrow) I will update defconfig for c7x0 with my space save changes Apr 29 11:04:00 hrw: ok. We should also think about them on the other models Apr 29 11:04:30 bitbake nano does fail to build. This is extremely worrying :-/ Apr 29 11:04:40 RP: diff probably can be applied to most of them Apr 29 11:05:32 hrw: We're also not as tight for space on the cxx00 models as we were - I split akita from spitz making jffs2 modular on spitz and ext3 modular on akita Apr 29 11:06:09 +CONFIG_CC_OPTIMIZE_FOR_SIZE=y Apr 29 11:06:22 is most important change Apr 29 11:06:39 +CONFIG_INET_DIAG=m Apr 29 11:06:40 +CONFIG_INET_TCP_DIAG=m Apr 29 11:06:45 another bonus (was in kernel) Apr 29 11:07:04 -CONFIG_FONT_8x8=y Apr 29 11:07:13 we did not use 8x8 but 8x16 Apr 29 11:07:27 -CONFIG_MSDOS_FS=y Apr 29 11:07:48 and that can be all - cpufreq modules = 2K only Apr 29 11:07:57 iirc Apr 29 11:08:30 The INET_DIAG is a good find :) Apr 29 11:08:33 * koen wants the cute logo Apr 29 11:08:53 RP: 'touch' is broken for eabi :/ Apr 29 11:08:54 koen: OZ one can be? Apr 29 11:09:25 koen: logo is just 1.1M patch ;) Apr 29 11:09:42 koen: 'touch' ? Apr 29 11:09:50 RP: yes Apr 29 11:09:52 hey zecke Apr 29 11:09:56 zecke: hi Apr 29 11:10:00 koen: you mean the touchscreen? Apr 29 11:10:03 hey Apr 29 11:10:07 RP: no, the command Apr 29 11:10:07 hail zecke Apr 29 11:10:15 RP: did you see the meta bitbake backport bug? Apr 29 11:10:17 as in 'touch /var/log/utmp' Apr 29 11:10:25 -> 'function not implemented' Apr 29 11:10:27 zecke: Have a look at it ;-) Apr 29 11:11:03 koen: Ah. Not good :-/ Apr 29 11:11:14 hmmm Apr 29 11:11:22 * koen tries a different busybox config Apr 29 11:11:46 koen: update to busybox 1.1.x Apr 29 11:12:02 10 April 2006 -- BusyBox 1.1.2 (stable) Apr 29 11:12:25 I looked at that, but the missing patch scared me Apr 29 11:12:47 zecke: I've pushed a few bitbake fixes/improvements to svn and added a few backport patches to that meta bug Apr 29 11:13:18 zecke: I also reviewed you patches and commented in the metabug accordingly Apr 29 11:13:38 RP: ah good this was my question Apr 29 11:13:50 RP: self.cachefile = False Apr 29 11:14:11 zecke: That would be fine with me Apr 29 11:14:13 RP: this way you can end up with two different types for cachefile, which even when possible is considered not too good Apr 29 11:14:31 *breakfast* Apr 29 11:15:44 ibot, botmail for reenoo: would you mind copying the latest autotools.bbclass from .dev into the oz354fam083 branch? It looks like this version of autotools_stage_all() is better at avoiding the rpath thing. Apr 29 11:16:14 zecke: I can see your point. I'm not sure I see a problem with allowing a Flase value though... Apr 29 11:17:39 03koen 07org.oe.dev * rc032f363... 10/packages/busybox/ (busybox-1.01/angstrom/defconfig busybox_1.01.bb): busybox: add defconfig for angstrom Apr 29 11:19:18 * rwhitby pushes madwifi-ng r1527 and matching wpa-supplicant 0.5.1 driver Apr 29 11:19:33 zecke: If you try a fresh OE build with "bitbake nano", bitbake is doing something wrong with the dependencies :-( Apr 29 11:19:46 busybox-svn have diff applet.. nice Apr 29 11:19:53 bitbake ncurses shows the problem as well... Apr 29 11:20:48 hrw: http://bugs.openembedded.org/show_bug.cgi?id=745, what does the mount-all type patch do? Apr 29 11:24:18 koen: ask person which added it into OE... Apr 29 11:24:19 ibot, botmail for florian_kc: I created a new branch "eabi-branch-20060429" in the kernel tree for this eabi work. Apr 29 11:32:52 hrw: http://oe-devel.bkbits.net:8080/openembedded/related/packages/busybox/busybox-1.00/mount-all-type.patch?nav=index.html|src/.|src/packages|src/packages/busybox|src/packages/busybox/busybox-1.00 Apr 29 11:40:42 re Apr 29 11:40:46 keyboard died in X11 ;( Apr 29 11:41:22 03rwhitby 07org.oe.dev * r60996b89... 10/packages/wpa-supplicant/wpa-supplicant_0.5.1.bb: wpa-supplicant: Added the missing space at the end of a _prepend in version 0.5.1 Apr 29 11:41:26 03rwhitby 07org.oe.dev * r7ea94f81... 10/packages/meta/slugos-packages.bb: slugos-packages: Removed hostap* - use wpa-supplicant instead. Apr 29 11:41:30 03rwhitby 07org.oe.dev * ra6f5ab39... 10/packages/madwifi/ (madwifi-ng-modules_svn.bb madwifi-ng_r1527-20060425.bb): madwifi: Renamed to madwifi-ng (which emits -modules and -tools) and now stages include files for wpa-supplicant to use. Apr 29 11:41:34 03rwhitby 07org.oe.dev * r1f24df3f... 10/packages/wpa-supplicant/wpa-supplicant_0.5.1.bb: wpa-supplicant: Added madwifi driver to experimental 0.5.1 verision. Apr 29 11:41:39 03rwhitby 07org.oe.dev * r1deebb03... 10/packages/wpa-supplicant/files/slugos/defaults-sane: wpa-supplicant: Added defaults file for slugos. Apr 29 11:41:43 03rwhitby 07org.oe.dev * r73a458a7... 10/ (conf/distro/slugos.conf packages/meta/slugos-packages.bb): slugos: Added madwifi-ng and wpa-supplicant (version 0.5.1 required). Apr 29 11:42:30 rwhitby: wpa-supplicant depend on madwifi-ng now? Apr 29 11:50:46 hey pH5 Apr 29 11:50:51 good afternoon Apr 29 11:50:53 hey koen Apr 29 11:51:36 hi ph5 Apr 29 11:52:19 hi pb_ Apr 29 11:56:53 hrw: yep, latest experimental 0.5.1 wpa-supplicant depends on madwifi-ng Apr 29 11:57:24 koen and RP convinced me not to make it slugos-only here one hour ago Apr 29 11:58:06 rwhitby: so it will depend on hostap-modules and madwifi-ng soon... in .oz354x I have wpa-supplicant 0.4.8 depend on hostap-modules Apr 29 11:58:07 hrw: I'm going to build 1.1.0 to so if it solved the busybox <> glibc problems Apr 29 11:58:31 koen: why not 1.1.2? too much patches to review? Apr 29 11:58:39 hrw: is that a problem? I'm not familiar with hostap Apr 29 11:59:10 rwhitby: not for me - does not know does hostap-modules 0.4.7 builds for slugos Apr 29 12:00:18 hrw: yep, too much patches too review Apr 29 12:00:51 * koen adds glibc2.4-icmp6.patch to 1.1.0 Apr 29 12:01:01 koen: some does no apply? Apr 29 12:01:18 hrw: it failed add add-getkeyapplet and I gave up from there Apr 29 12:01:24 ok Apr 29 12:01:28 hrw: just confirmed that hostap-modules 0.4.7 builds no problem for slugos Apr 29 12:03:56 crap Apr 29 12:04:11 by the looks of it we need at least busybox 1.1.1 to have good eabi support Apr 29 12:04:17 aha, hostap is to Prism as madwifi is to Atheros. Apr 29 12:04:39 rwhitby: right Apr 29 12:04:52 hrw: at some point, I will want wpa-supplicant to depend on zd1211 too :-) Apr 29 12:05:09 (for usb wifi modules for the slug) Apr 29 12:05:32 rwhitby: it should probably be recommending those things, not depending on them. Apr 29 12:06:38 you don't really want everyone to be obliged to install the drivers for every possible wifi card. for example, zd1211 is pretty useless on the ipaq since there is (generally) no usb host facility. Apr 29 12:06:41 pb_: for madwifi, it actually needed include files from the source. don't know yet whether zd1211 will need to be depends to rrecommends. Apr 29 12:06:59 s/depends to/depends or/ Apr 29 12:07:07 ah, right, I see what you mean Apr 29 12:07:18 if it's just a build time dependency then that's fine, of course Apr 29 12:07:23 03pH5 07org.oe.dev * rf0d5a9ca... 10/packages/linux-libc-headers/ (2 files in 2 dirs): linux-libc-headers-2.6.15.99: remove extern function declaration from linux/netdevice.h Apr 29 12:07:34 pb_: that's all it is Apr 29 12:07:40 righto Apr 29 12:07:53 hm, stupid apmd still isn't working right Apr 29 12:07:55 * pb_ stabs linux Apr 29 12:07:55 hrw: http://oe.pastebin.com/688497 Apr 29 12:08:40 koen: those 3 patches need work? Apr 29 12:08:54 hrw: I think the last one has been applied Apr 29 12:09:28 hdparm_M.patch can be dropped iirc - no defconfig with hdparm in OE Apr 29 12:09:50 hrw: LOL: http://www.oesf.org/forums/index.php?showtopic=19134&pid=125075&st=0&#entry125075 Apr 29 12:09:53 and as usual: someone need to push those patches upstream/// Apr 29 12:10:25 koen: 1st part is nice, 2nd is his usual kindergarten Apr 29 12:12:15 zecke: I need to talk to you about what looks like a fairly serious bug in bitbake :-( Apr 29 12:12:39 RP: sure :) Apr 29 12:12:53 RP: either now or after my assigments Apr 29 12:13:13 hrw: lol, sashz is not a good teamplayer it seams Apr 29 12:13:23 w8 for my answer Apr 29 12:13:24 zecke: I can quickly explain it and see if you can see the problem? Apr 29 12:14:12 how would you call 'rake' for children? (tools which childs use in sandbox) Apr 29 12:14:12 zecke: I have this applied: http://www.rpsys.net/openzaurus/temp/bitbake_debug.patch Apr 29 12:14:30 zecke: I see this:NOTE: dependencies for ncurses are: autoconf-native automake-native libtool-native gnu-config-native ipkg-utils-native file-native quilt-native virtual/arm-linux-gcc virtual/libc ncurses-native Apr 29 12:14:30 NOTE: dependencies for ncurses are: autoconf-native automake-native libtool-native gnu-config-native ncurses-native Apr 29 12:14:54 zecke: So the results from loadData and loadDataFull are different Apr 29 12:15:11 ERROR: /home/slug/debianslug/openembedded/packages/autoconf/autoconf-native_2.59.bb depends on its Apr 29 12:15:12 elf (eventually) Apr 29 12:15:12 ERROR: upwards chain is: slugos-packages (debianslug-packages) -> file-native (file-native) -> aut Apr 29 12:15:12 oconf-native (autoconf-native) -> m4-native (m4-native) -> gnu-config-native (gnu-config-native) - Apr 29 12:15:12 > quilt-native (quilt-native) -> gettext-native (gettext-native) -> libtool-native (libtool-native Apr 29 12:15:12 ) Apr 29 12:15:37 zecke: Specifically base_dep_prepend in base.bbclass isn't getting run Apr 29 12:15:42 that doesn't look good - anyone know what's changed? Apr 29 12:15:59 rwhitby: an old and broken quilt-native? Apr 29 12:16:27 RP: right this looks like the __anonfunc of base.bbclass is not run Apr 29 12:16:58 zecke: is four hours old too old? Apr 29 12:17:36 zecke: Maybe - base_dep_prepends() is called by DEPENDS_prepend="${@base_dep_prepend(d)} " Apr 29 12:17:51 oops - 1 day, 4 hours. Apr 29 12:18:10 RP: ah right Apr 29 12:18:29 RP: hmm looking at the time, I need to postpone discussion a couple of hours :} Apr 29 12:18:45 zecke: ok Apr 29 12:28:03 speaking of coreutils-native, there is no coreutils-native_5.3.0.bb and coreutils-native_5.1.3 is when I try and build gpe-image. Apr 29 12:28:28 *dying Apr 29 12:30:10 hrw: it seems you need to kill me Apr 29 12:30:46 because? Apr 29 12:31:22 check the forum (and please don't post any reponses) Apr 29 12:32:06 ;) Apr 29 12:32:28 I posted before that post Apr 29 12:32:33 http://www.oesf.org/forums/index.php?showtopic=19134&st=0&p=125079&#entry125079 Apr 29 12:33:11 koen: you just lack diplomacy too^W often Apr 29 12:34:04 I would say "at all" Apr 29 12:36:50 * koen installs winxp Apr 29 12:36:51 I tried to be nto so cruel ;D Apr 29 12:36:58 koen: on corgi? Apr 29 12:37:02 heh Apr 29 12:37:10 no, on my workstation Apr 29 12:38:27 I bought "C&C - the first decade" and that only runs on xp Apr 29 12:38:44 I would have used wine if ubuntu hadn't decided to mess up dri again Apr 29 12:39:17 ;) Apr 29 12:40:26 ugh Apr 29 12:40:41 * koen downloads 50+ mb worth of ati drivers Apr 29 12:41:37 * hrw need to fetch via rhine drivers for w98 Apr 29 12:45:01 time to shopping Apr 29 12:47:02 * rwhitby is worried that Zaurus hacking seems to be going the way of WRT54G hacking ... many disjoint camps throwing spears at each other ... just an outsider's view. Apr 29 12:48:55 going? it is on that way Apr 29 12:49:06 cu Apr 29 12:50:28 OESF seems very OpenZaurus oriented. Is that intentional, or just historical? Apr 29 12:50:56 (i.e. would an NSLU2/OpenSlug forum be out of place on that site?) Apr 29 12:50:56 hrw|gone: cya Apr 29 12:50:56 rwhitby: spear throwing is new Apr 29 12:51:16 rwhitby: OESF has nothing to do with OE Apr 29 12:51:35 rwhitby: I think it is just the majoriy of people are zaurus users Apr 29 12:51:59 zecke: has nothing to do with OE, apart from the first two words of it's name? Apr 29 12:52:45 how many installations of OE-derived firmware are out there? Apr 29 12:53:10 rwhitby: i think this is too hard to estimate? Apr 29 12:53:23 (there are at least 15000 installations of nslu2-linux firmware built on OE) Apr 29 12:53:26 rwhitby: I think it is very possible and wanted to have a NSLU2/OpenSlug forum at OESF Apr 29 12:54:15 rwhitby: OESF stole our name without asking Apr 29 12:54:16 zecke: well, tonight is the first time I have seen OESF, and I'm not sure I'd want to have a forum there ;-) Apr 29 12:54:35 koen: time to get the trademark :} Apr 29 12:54:59 lo Apr 29 12:55:05 gerwinin: hey Apr 29 12:55:16 hey gerwinin Apr 29 12:55:18 zecke: how is everything going ? Apr 29 12:55:59 Koen: not celebrating queensday ? Apr 29 12:56:59 gerwinin: tomorrow evening I'll head to the bar :) Apr 29 12:57:25 Koen: I am a republican so but I go to the bar anyway tonight :) Apr 29 12:57:40 Koen: but not american republican :) Apr 29 12:59:36 I dislike the whole monarchy thing Apr 29 12:59:45 and I really don't like the oranjebitter Apr 29 13:00:09 rwhitby: The differences in this case are long standing sadly - just at the surface atm :-/ Apr 29 13:00:22 Koen me too :) Apr 29 13:00:42 Koen the Netherlands should get a generals regime , I am for a complete regime change Apr 29 13:00:48 heh Apr 29 13:00:53 * koen starts a junta Apr 29 13:01:22 rwhitby: You can probably count the number of times I've posted there on your fingers... Apr 29 13:01:41 koen: I like your title of professional clown :) Apr 29 13:02:01 what is the oranjebitter? Apr 29 13:02:08 People keep suggesting nslu2-linux have web forums, and I keep rejecting the idea. Apr 29 13:02:21 rwhitby: hehe Apr 29 13:02:25 Crofton: a traditional liquor drank at queensday Apr 29 13:02:31 ah Apr 29 13:02:33 Crofton: very bitter and orange Apr 29 13:02:46 kinda like orangemarmalade with alcohol Apr 29 13:02:49 * Crofton makes a note to avoid this Apr 29 13:03:01 of course I like sambuka Apr 29 13:03:36 hehhee schlidorvitsch rulez :) Apr 29 13:03:40 And absinth Apr 29 13:04:52 give me a Coopers Pale Ale anyday Apr 29 13:05:16 NOTE: package glibc-intermediate-2.4-r4: task do_populate_staging: started Apr 29 13:05:19 * koen taps foot Apr 29 13:05:46 http://www.coopers.com.au/beer.php?id=128&pid=1 Apr 29 13:06:18 heheheheh a beer I drink frequently is guinness :) Apr 29 13:06:35 I think I spent already a month salary on it in the local irisch pub :) Apr 29 13:12:52 Hmm: NOTE: package gnutls-1.3.5-r2: task do_populate_staging: failed Apr 29 13:13:13 zecke: I'm halfway rosterifying yesterdays db, so I can update the 0.26 server Apr 29 13:14:44 The error is: /home/slug/debianslug/tmp/work/gnutls-1.3.5-r2/gnutls-1.3.5/doc//signatures.texi:14: Unknown command `euro'. Apr 29 13:17:13 I'm thinking about something like http://www.rpsys.net/openzaurus/temp/glibc_workionprogress.patch (with locales in my local.conf, not the glibc.bb) to speed up glibc. I'm not sure how much of a speed increase this gives or whether it can be further improved. Any thoughts? Apr 29 13:19:22 (now updated to include the .inc) Apr 29 13:27:49 hi, need some advice: I'm trying to craft a .bb file for a new module, but when the package is downloaded it extracts in the wrong dir (in .../tmp/work/ark3116-0.4.1-r0 instead of .../tmp/work/ark3116-0.4.1-r0/ark3116-0.4.1 Apr 29 13:27:59 how can I resolve this ? Apr 29 13:28:22 i started off with a .bb file that does things as expected, so I am kind of flabbergasted that this happens Apr 29 13:34:46 eFfeM, your package (tar.gz/tar.bz2) probably doesn't have the leading ark3116-0.4.1 Apr 29 13:35:12 myopiate: it doesn't thought this was made by bb Apr 29 13:36:10 i now understand the problem Apr 29 13:36:12 eFfeM, i believe it isn't, you usually have to set S = to the directory where autotools and make needs to berun Apr 29 13:36:13 tnx Apr 29 13:36:30 np Apr 29 13:44:21 NOTE: package glibc-2.4-r4: task do_populate_staging: started Apr 29 13:44:26 * koen taps foots Apr 29 13:44:30 ehm Apr 29 13:44:31 feet Apr 29 13:44:39 vg koen Apr 29 13:45:16 re Apr 29 13:45:23 heh Apr 29 13:45:25 rwhitby: avoid OESF forums Apr 29 13:45:26 hre: wb Apr 29 13:46:59 rwhitby: would you have some eligible nslu2 people for google SoC? Apr 29 13:47:33 koen: how do you mean? people to sponsor projects, or people to do projects? Apr 29 13:49:48 to do projects Apr 29 13:50:00 hre: thx for the advice, I think I will avoid them. Apr 29 13:50:09 koen: OE projects? Apr 29 13:50:22 rwhitby: its too much zaurus specific Apr 29 13:51:04 rwhitby: ofcourse some people there would shout 'it need nslu2 section because friend of my friend has a friend which have nslu2' Apr 29 13:51:19 rwhitby: yes, we should be able to get some OE slots in the hh.org SoC Apr 29 13:53:00 * hre apply diff-to-diff to diff Apr 29 13:53:16 hre: /nick hrw Apr 29 13:53:26 hre: hehe, and a different file... Apr 29 13:53:32 there, now I won't get confused :) Apr 29 13:53:43 hrw: I have a svn repository for the keymap driver on ewi Apr 29 13:53:47 hrw: you can co it ;) Apr 29 13:53:48 zecke: I use gvim for it Apr 29 13:54:02 hrw: it is to a different file as well (applying the patch) Apr 29 13:54:55 zecke: so give me svnurl Apr 29 13:55:25 zecke: btw - capslock/numlock works on collie Apr 29 13:56:03 hrw: ?? Apr 29 13:56:21 hrw: svn co file:///home/zecke/qte-hacking/qte-repo Apr 29 13:56:37 03rwhitby 07org.oe.dev * rfd10e7e7... 10/packages/gnutls/ (gnutls-1.3.5/gnutls-texinfo-euro.patch gnutls_1.3.5.bb): gnutls: Minor patch to signatures.texi so that it will parse correctly with older versions of texinfo (like the version in Debian Sarge). Apr 29 13:57:41 thx Apr 29 13:58:25 zecke: booted collie today, upgraded to newest qte patchset and tested capslock and fn^h^hnumlock under opie - works ok Apr 29 14:00:09 zecke: that repo has only kernel-keymap applied? Apr 29 14:02:44 hrw: yes, only kernel keymap :} Apr 29 14:03:15 hrw: wee need to upgrade meta-opie to not depend on opie-calibrate Apr 29 14:05:44 zecke: can you? Apr 29 14:06:03 NOTE: package busybox-1.1.2-r0: task do_compile: started Apr 29 14:07:47 nice Apr 29 14:07:58 hmm.. quilt does not like xfs? Apr 29 14:08:07 hrw: okay Apr 29 14:08:13 or just diff-to-diff? Apr 29 14:12:21 hrw: hmm, is it clear to you that this is a diff-diff agsinst the ROAD repo and can't be applied directly to our keybaord driver? Apr 29 14:13:04 zecke: sure - going though diff your repo our-qte now Apr 29 14:13:13 hrw: do_install does nothing with busybox 1.1.2 Apr 29 14:16:01 koen: all what I can say about busybox: 1.1.0 was iirc working Apr 29 14:16:03 koen + everybody: Do you think bug 26 can be closed? There have been no objections to the patches for the last year. Apr 29 14:16:58 hrw: http://ewi546.ewi.utwente.nl/tmp/busybix-breakage Apr 29 14:17:23 Laibsch: it is a bero patch (loooks good) Apr 29 14:19:42 * chouimat is back. Apr 29 14:19:50 morning Apr 29 14:21:40 I am trying to verify bug 33. bitbake fails to build glibc so I guess something is seriously wrong on my end. Somebody care to take a look: http://rafb.net/paste/results/M185TE45.html Apr 29 14:21:49 ? Apr 29 14:23:02 zecke: got it: Apr 29 14:23:02 kernelkeyboard.cpp | 684 +++++++++++++++++++++++++++++++++++++++++++++++++ Apr 29 14:23:02 keyboard_linux_to_qt.h | 263 ++++++++++++++++++ Apr 29 14:23:02 qkeyboard_qws.cpp | 166 +++++------ Apr 29 14:23:02 3 files changed, 1022 insertions(+), 91 deletions(-) Apr 29 14:24:00 hrw: I think I have a fix for busybox Apr 29 14:26:06 Packaged contents of busybox into /data/build/koen/OE/build/tmp/angstrom/deploy/ipk/busybox_1.1.2-r0_armv5te.ipk Apr 29 14:26:50 zecke: cp -a is bad, right? Apr 29 14:27:14 koen: yes Apr 29 14:27:38 what's the correct one? Apr 29 14:27:50 koen: cp -??? ;) Apr 29 14:28:43 koen: cp -pPR Apr 29 14:29:08 03freyther 07org.oe.dev * r7df389f1... 10/packages/meta/task-opie.bb: Apr 29 14:29:08 packages/meta/task-opie.bb: Remove opie-calibrate DEP Apr 29 14:29:08 The opie-calibrate package has been removed from Apr 29 14:29:08 OE (it is a builtin executable in the qpe launcher). Apr 29 14:29:08 Remove the wrong dependency from it now Apr 29 14:29:14 zecke: thanks Apr 29 14:29:35 koen: scp (schilling cp) ;) Apr 29 14:29:41 heh Apr 29 14:33:05 zecke: NOTE: package qte-2.3.10-r33: task do_patch: completed Apr 29 14:33:31 hrw: I updated http://bugs.openembedded.org/show_bug.cgi?id=745 Apr 29 14:34:12 hrw: nice :) Apr 29 14:36:27 Packaged contents of qte into /home/hrw/devel/build/0420-354x/tmp/deploy/ipk/libqte2_2.3.10-r33_tosa.ik Apr 29 14:39:22 someone switched to libata for pata discs? Apr 29 14:39:32 * cyrilRomain still trying to figure out why coreutils-5.1.3 is built instead of coreutils-5.3.0. Related bug: http://bugs.openembedded.org/show_bug.cgi?id=814 Apr 29 14:39:45 hrw: my ep93xx should use libata, but I haven't tried it yet Apr 29 14:39:58 cyrilRomain: PREFERRED_VERSION somewhere? Apr 29 14:41:55 koen: "egrep -r 'PREFERRED_VERSION_core' *" from org.openembedded.dev/ returns nothing :-/ Apr 29 14:45:27 koen: i'm so blind ! this is coreutils-native, not coreutils ! Apr 29 14:46:23 koen: As Bernhard Rosenkraenzer propose in the bug report, maybe should we cp coreutils-native_5.1.3.bb coreutils-native_5.3.0.bb Apr 29 14:46:34 * cyrilRomain try it right now Apr 29 14:51:41 ~lart tt for shit called qt buildsystem Apr 29 14:51:41 * ibot forces tt to use Outlook Express for shit called qt buildsystem Apr 29 14:52:00 do_compile: completed (with not linked binaries) Apr 29 15:02:47 koen: 'cp coreutils-native_5.1.3.bb coreutils-native_5.3.0.bb' to build coreutils-native_5.3.0 worked like a charm. And seeing its content, I think we can safely add this file in the database. Apr 29 15:04:12 I have a general OE policy question: do you keep maintain old version of packages when new one correctly build for any machine ? Apr 29 15:14:21 03hrw 07org.oe.oz354x * rf77a5313... 10/packages/meta/meta-opie.bb: meta-opie: removed opie-calibrate (removed from OE) Apr 29 15:17:51 cyrilRomain: I updated coreutils Apr 29 15:17:53 -native Apr 29 15:18:53 koen: great :) Apr 29 15:19:09 so the bug can be closed Apr 29 15:19:14 koen: this does not solve the issue of the coreutils-native_5.1.3 though Apr 29 15:19:30 cyrilRomain: 5.1.3 can be dropped Apr 29 15:19:41 hrw: well, ok Apr 29 15:19:48 cyrilRomain: If nobody reports problems with 5.3.0 we'll drop the previous verison Apr 29 15:20:12 RP: ok, this answer to my general OE policy question, thank you guys :) Apr 29 15:20:32 koen: I close the bug right now Apr 29 15:20:38 thanks Apr 29 15:20:40 03koen 07org.oe.dev * r14a4cb7c... 10/packages/coreutils/coreutils-native_5.3.0.bb: coreutils-native: update to 5.3.0 Apr 29 15:21:25 Can I close bug 26? Apr 29 15:22:01 No, I can't since I cannot push the patches into OE Apr 29 15:22:13 But maybe somebody else wants to do it? Apr 29 15:23:27 someone know how to convert bin/cue image to normal iso? Apr 29 15:25:28 there is something ... try to look on freshmeat.net searcing 'cue' Apr 29 15:25:42 nslu2-linux finally got a clean autobuild of all four distros and all package feeds :-) Apr 29 15:25:45 bchunk found Apr 29 15:26:09 Come on guys, I thought today was bug squashing time? Apr 29 15:26:16 rwhitby: congrats Apr 29 15:26:34 rwhitby: Excellent. Do you run autobuilds regularly so we can know if we break it? Apr 29 15:26:46 RP: yes, and they are reported in tinderbox Apr 29 15:26:56 we run every 15 minutes Apr 29 15:27:09 BTW, bitbake segfaults on one of my machines now. What should I do? Apr 29 15:27:15 rwhitby: incremental? hmm... Apr 29 15:27:17 of course, I have forgotten where tinderbox results are ... Apr 29 15:27:23 yes, incremental Apr 29 15:27:43 Laibsch: bug #26 looks nasty as patches for only half the issue appear to be present Apr 29 15:27:56 but we do a full build from scratch every six months or so. this latest one took three weeks to get back to a clean autobuild :-( Apr 29 15:28:35 rwhitby: For a while I've been planning to have a machine do autobuild of a few common settings. We should probably look for volunteers to do that if we can detail the configs we want tested... Apr 29 15:28:47 (from scratch say once a week) Apr 29 15:28:53 RP: OK. Then I guess, I need to look for other bugs. Like those where compilation failed. Apr 29 15:29:11 RP: I could do that autobuild if needed Apr 29 15:30:07 Or I would really like to builg qpobox. But right now, OE aborts half-way through while compiling some intermediate packages on one machine and segfault 5 seconds after being called on the other :-( Apr 29 15:30:41 Laibsch: You have some bugs you can reproduce and fix then :) Apr 29 15:30:52 I have some bugs. Apr 29 15:31:03 But I am not sure I can fix them :-( Apr 29 15:31:12 Laibsch: bitbake segfaulting sounds bad btw, especially given its written in python - is your python install ok? Apr 29 15:31:21 RP: tested sound on c7x0 recently? Apr 29 15:31:35 hrw: Possibly not. Is it broken? Apr 29 15:31:37 RP: Straight from Debian. Worked OK just yesterday. Apr 29 15:32:01 RP: 'sounds' like - will test a bit Apr 29 15:32:22 can someone check what permissions busybox has? Apr 29 15:32:36 755? Apr 29 15:32:49 755 - cannot be setuid Apr 29 15:32:54 RP: I can help building from scratch with autobuild and report bugs if needed Apr 29 15:32:59 -rwxr-xr-x 1 root root 514456 Jan 1 1970 /bin/busybox Apr 29 15:33:53 rwhitby: Do you want to email oe@ with the setups you'd like to see autobuilt? I'll add the zaurus ones and then perhaps we can organise some people to test them... Apr 29 15:34:15 hrw: thanks Apr 29 15:34:21 np Apr 29 15:34:52 koen: your favorite devel wrote on oesf Apr 29 15:35:07 RP: mine are easy: machine=ixp4xx, distros=openslug, debianslug, ucslugc. machine=nslu2, distro=unslung. Apr 29 15:35:21 cyrilRomain, NA|Gone_ I think we'd best discuss this on the oe mailing list - you're both subscribed? Apr 29 15:35:39 RP: and target is -packages in all cases. Apr 29 15:35:46 RP: Yep Apr 29 15:35:52 rwhitby: Do you use multimachine? Apr 29 15:35:57 RP: nope Apr 29 15:36:08 RP: yes Apr 29 15:36:11 (don't really know what it is) Apr 29 15:36:51 rwhitby: is machine = nslu2 with openslug, debianslug or ucslug a valid combination? Apr 29 15:37:58 rwhitby: Its useful when you have machines like c7x0, spitz and akita, all of which are mostly common code with a few machine specific bits. multimachine would build one copy of glibc for amv5te but a copy of initscripts for each machine Apr 29 15:37:59 hmmm, it's stuck at init it seems Apr 29 15:38:04 stupid eabi Apr 29 15:38:41 RP: multimachine only works with machines, not distros Apr 29 15:38:48 RP: those distros work for multiple machines without change. Apr 29 15:39:12 * hrw have to check alsa mixer ;( Apr 29 15:40:07 koen: I know that and given hat rwhitby is doing it might not help much as he's changing distros, not machines Apr 29 15:40:15 RP: actually machine=nslu2 is fine for all those distros. so machine=nslu2, distros=openslug,debianslug,ucslugc,unslung. Apr 29 15:40:30 RP: 'Jack Function' has to be set to 'Headphones Apr 29 15:40:36 RP: 'Jack Function' has to be set to 'Headphones' to get speaker sound? Apr 29 15:40:57 hrw: That doesn't sound right... Apr 29 15:41:14 hrw: speaker has to be set to on.. or if headphones removed, auto also works Apr 29 15:41:33 NA|Gone_: speaker is other switch - have to 'On' now Apr 29 15:41:42 RP: the "nslu2" machine actually supports the nslu2, ds101, loft, nas100d and dsmg600 targets. Apr 29 15:42:09 RP: and all samples sounds same ;( Apr 29 15:42:42 So I guess we have 5 x 3 + 1 = 16 OE-based distros in nslu2-linux. Apr 29 15:43:09 rwhitby: Right. I guess we did this slightly differently. There are probably more differences in the zaurus machine cases... Apr 29 15:43:22 RP: Re the autobuild stuff, might be worth looking at mozilla's tinderbox to see if that would assist in reporting build outcomes for OE Apr 29 15:43:45 NAbyss: there already is an OE tinderbox. I just forget where it is. Apr 29 15:43:57 ah, okies Apr 29 15:44:02 NAbyss: We already have some tinderbox infrastructure - its just not very well used Apr 29 15:44:04 RP: we NEED volume limit Apr 29 15:44:31 hrw: volume limit for what? headphones? speakers? Apr 29 15:44:40 NAbyss: http://ewi546.ewi.utwente.nl/tinderbox/showbuilds.pl?tree=OpenEmbeddedBuilds Apr 29 15:45:03 RP: speaker - playback>86 is fscked Apr 29 15:45:19 RP: wheezing Apr 29 15:45:27 NAbyss: we just need to improve the code for it Apr 29 15:45:30 hrw: That involves hacking the kernel codec :-/ Apr 29 15:45:43 zecke: ahh Apr 29 15:45:58 NAbyss: e.g. I want to have. When was package XYZ with version Mumba build on a XScale CPU ;) Apr 29 15:46:15 RP: with playback == 83 'Jack function' can be anything and sound exist Apr 29 15:46:39 zecke: I just worry about how many builds have to occur for a full test :) Apr 29 15:46:49 RP: with playback==100 sound does not sound but wheeze badly and 'jack function' does matter Apr 29 15:46:56 zecke: all that loverly nslu2-linux green :-) Apr 29 15:47:10 hrw: oh, look at out2 right inver Apr 29 15:47:11 t Apr 29 15:47:15 rwhitby: hehe, OE Land Security = green Apr 29 15:47:23 NAbyss: you have cxx00? Apr 29 15:47:28 hrw: yeah, 3100 Apr 29 15:47:38 NAbyss: it has more switches then c7x0 Apr 29 15:47:52 hrw: We'll need to talk to Liam about it Apr 29 15:47:53 I have the same issue unless out2 invert is selected.. destructive interference. Apr 29 15:48:02 rwhitby: Do I need to flash another bootldr to run the NSLU2 in little endian mode? Apr 29 15:48:07 NAbyss: Does that get set automatically now? Apr 29 15:48:12 (it should do) Apr 29 15:48:24 koen: did you see the busybox talk from CELF? Apr 29 15:48:33 zecke: nope - we add a small assembly prepend to the kernel which switches the arm endianness Apr 29 15:48:37 RP: Not sure.. my build's from 20060422.. will look next reboot Apr 29 15:48:49 koen: the current maintainer is talking about how it is managed Apr 29 15:49:05 koen: e.g. you cann apply each bugfix seperately from the patches directory Apr 29 15:49:10 zecke: I didn't see it Apr 29 15:49:17 RP: doesn't appear to be set by default Apr 29 15:49:19 zecke: I'm trying to boot it right now Apr 29 15:49:20 zecke: one principle of nslu2-linux is that we never ask Joe User to change bootloader. It's too dangerous for a lot of unskilled people. Apr 29 15:49:23 but it doesn't work Apr 29 15:49:26 rwhitby: ah well. I'm considering getting a 2nd NSLU2 to run it in LE mode Apr 29 15:50:05 zecke: I run all mine in LE now. Apr 29 15:50:19 (oh, except the one which is currently testing Unslung 6.8) Apr 29 15:50:48 rwhitby: what is the reasoning? Apr 29 15:51:00 zecke: to run stock debian arm port usually. Apr 29 15:52:00 anyway, now I have a clean autobuild it's time for bed. Night all. Apr 29 15:52:18 cya Apr 29 15:57:24 heh Apr 29 15:57:27 03eFfeM 07org.oe.dev * rd73fbf99... 10/packages/ark3116/ (3 files in 2 dirs): Apr 29 15:57:27 ark3116: usb serial adapter driver module Apr 29 15:57:27 includes patch to compile under 2.6.16; compiles, untested Apr 29 15:57:33 hrw: more kindergarten Apr 29 15:58:12 hrw: did we kill sash's dog or something? Apr 29 15:59:38 NAbyss: Hmm. So it either got corrupted sometime after or the defaults didn't work properly :-/ Apr 29 16:00:18 RP: Yeah.. how long's the patch for that been in place, roughly? Apr 29 16:01:35 NAbyss: Quite a while Apr 29 16:02:10 koen: URL? Apr 29 16:02:25 Ok, mail sent to oe@ about autobuilds - feel free to reply with your favourite combinations Apr 29 16:02:30 zecke: http://www.oesf.org/forums/index.php?showtopic=19134&view=getlastpost Apr 29 16:02:42 koen: just share some vla with him! Apr 29 16:02:56 RP: Hmm, I haven't noticed it in any builds I've done; let me know if there's any testing needed to be done Apr 29 16:03:03 hrw: bchunk is one way to do it. it would be compatible to cdrdao's cdr img formats if you reworked the description file using a text editor, or say a perl script. bchunk should work most of the time Apr 29 16:03:21 is sashz a bot? Apr 29 16:04:42 Debian is a dirty hack because they need patches to build debian packages Apr 29 16:04:45 good thinking Apr 29 16:05:08 zecke: pdaX just got removed from the list of distros angstrom wants to cooperate with Apr 29 16:05:08 NAbyss: Its a question of checking the default mixer configuration in zaurusd and either correcting it or working out why it wasn't applied Apr 29 16:05:40 koen: ah well, don't over react Apr 29 16:05:52 koen: I think we still should try to cooperate Apr 29 16:06:17 koen: sash sounds childish though, but it could be due the lack of english skills Apr 29 16:06:39 koen: which is very understandable as he comes from a former soviet country where english wasn't too appreciated Apr 29 16:07:40 koen: regarding pdaXrom he seems to be narrow minded, probably due bad NAND somewhere :} (SCNR) Apr 29 16:11:09 zecke: he strikes me as coming from "russian" mentality Apr 29 16:11:23 zecke: his akita has bad nand Apr 29 16:11:59 jnc: I was living for >half life ~50km from ussr Apr 29 16:12:24 ah, yeah that's not like i'm saying all people from the area think a certian way Apr 29 16:13:03 i mean it just seems like a familiar thing, especially from the HPCA groups of the area Apr 29 16:13:40 the way he tries to bash on you in response to your pointing out his own flaw, that's classic. Apr 29 16:14:00 jnc: I know - just added note Apr 29 16:14:25 hpca? Apr 29 16:14:47 hacking, phreaking, cracking, anarchy Apr 29 16:15:09 ah Apr 29 16:15:58 anyway u-boot on Z would be nice Apr 29 16:17:08 I got told all the zaurus PM code was a hack by sash earlier :) Apr 29 16:17:10 zecke: I'm not reacting at all :) Apr 29 16:17:45 zecke: did you see my patch for qte-mt on angstrom-dev@? Apr 29 16:18:33 RP: and you have proffesional patch for pxa27x overlay Apr 29 16:18:47 hrw: Not yet - it needs work Apr 29 16:18:54 03rpurdie 07org.oe.dev * r8f5cc6a2... 10/packages/tslib/ (tslib/usec_fix.patch tslib_cvs.bb): tslib: Add fix for collie_raw and corgi_raw modules from Alberto Mardegan (#511) Apr 29 16:19:04 zecke: capslock does not work on c7x0 Apr 29 16:19:20 hrm Apr 29 16:19:23 RP: it is in /patches/ so you have it (sash way of thinking) Apr 29 16:19:38 does anyone spot some serious flaw inside http://ewi546.ewi.utwente.nl/tmp/Angstrom-bootstrap-image-test-20060429-c7x0.rootfs.tar.bz2 ? Apr 29 16:20:09 it stops after "freeining init memory: 80k" Apr 29 16:20:24 keyboard works, cf insertions get printed to screen Apr 29 16:20:40 koen: I'm not on that list :} Apr 29 16:20:43 koen: EABI? Apr 29 16:20:48 jnc: yes Apr 29 16:20:54 jnc: with busybox 1.1.2 Apr 29 16:20:59 koen: i built an image that had EABI rootfs, but the kernel was not Apr 29 16:21:04 it did the same thing you describe Apr 29 16:21:22 the kernel should be eabi Apr 29 16:21:32 i.e it booted an eabi rootfs earlier Apr 29 16:21:37 ah Apr 29 16:21:53 can you boot a gpe-image? Apr 29 16:22:16 I'm trying with bootstrap-image since X locks it Apr 29 16:22:43 RP: sending reports to a tinderbox is fairly easy Apr 29 16:22:44 koen: It if never prints the init version number, init broke Apr 29 16:22:44 it's a shame we have trouble to pass init=/bin/ash Apr 29 16:22:45 :) Apr 29 16:23:13 RP: I wonder why, since I only changed busybox Apr 29 16:23:21 * koen builds an altboot image Apr 29 16:23:22 zecke: I don't understand tinderboxe's output either ;-) Apr 29 16:23:43 * RP suspects he would if he thought about it for a while... Apr 29 16:23:44 RP: in what regard? green, yellow, red? Apr 29 16:23:57 RP: or the summary page? Apr 29 16:24:12 (hint use a mozilla for the full pleasure of the tinderbox) Apr 29 16:24:39 I get the idea of the colours (although either it works or doesn't - yellow means currently running?) Apr 29 16:24:58 could we get tinderbox.openembedded.org (and monotone.openembedded.org too)? Apr 29 16:25:05 I just never quite worked out what the boxes applied to... Apr 29 16:25:12 I always forgot where on ewi tinderbox it Apr 29 16:25:17 RP: yellow buiild in progress Apr 29 16:25:24 * RP will try mozilla Apr 29 16:25:41 RP: hrw asked me month ago to add a legend to it Apr 29 16:26:03 zecke: That will help :) Apr 29 16:26:09 RP: now that oe and bitbake are fairly stable Apr 29 16:26:22 RP: I can concentrate (well in my spare time) on tinderbox and QA again Apr 29 16:26:27 http://ewi546.ewi.utwente.nl/tinderbox/ and then? Apr 29 16:26:46 please don't add dns pointing to ewi Apr 29 16:27:01 that means network access will be killed within minutes Apr 29 16:27:09 hrw: if koen would run the cron job... Apr 29 16:27:12 koen: ok Apr 29 16:27:14 koen: a redirect? Apr 29 16:27:23 hrw: click on administrate Apr 29 16:27:28 hrw: and then select any tree :} Apr 29 16:27:54 zecke: This is much improved since I last looked :) Apr 29 16:28:10 zecke: I can arrange another machine to host tinderbox builds Apr 29 16:28:33 zecke: and for E800/year I can have it hosted with unlimited bandwidth Apr 29 16:28:44 koen: could you arrange another zecke as well Apr 29 16:28:52 I think I won't get along with him... Apr 29 16:29:14 zecke versus zecke, next on ultimate developer battle Apr 29 16:29:49 RP: for each build you have a summary Apr 29 16:29:57 RP: e.g. you can see what packages were built Apr 29 16:29:58 i've got a vps thing, i'd offer to host some files, it's a little lean though. only 100G / month transfer and 256mb ram :/ Apr 29 16:30:02 RP: and you can get the full log Apr 29 16:30:16 RP: my next task is to extract informations from the log :) Apr 29 16:30:21 information even Apr 29 16:30:23 jnc: they could both do nothing.. 1st zecke will tell zecke to do it and vice-versa :D Apr 29 16:30:34 hah Apr 29 16:30:39 * hrw has 10G/month limit on hosting Apr 29 16:30:56 well one zecke would get annoyed by the other Apr 29 16:31:32 you can escape from the deadlock by making one more stoned than the another... Apr 29 16:31:39 back to the assignment :} Apr 29 16:31:40 zecke: 3 things which tinderbox page lack: 1. color legend, 2. status legend, 3. join-howto Apr 29 16:32:45 hrw: at my bmco research office, I have some perl nerds around I hope I can exploit :} Apr 29 16:33:43 hrw: sashz makes a decent point though, you two are sort of arguing about which build system/method is more appropriate. sashz has every right to develop the way he is doing so, and any personal attacks should not involve build systems. coredump's comment wasn't helping that much :/ Apr 29 16:34:04 i thought that might be helpful to add, since, i kind of see both sides Apr 29 16:34:40 there's too much invested in OE for it to be a reasonable action for you to start hacking on both OE and pdaXrom Apr 29 16:35:13 beyond that, i don't see why a thread on the forums is needed to "address" sashz's commentary Apr 29 16:35:48 jnc: you can't say "OZ sucks, you're idiots" and expect us to help him with building linux-openzaurus for pdaX Apr 29 16:36:01 jnc: "Don't bite the hand that feeds you" Apr 29 16:36:13 that is sure one way of viewing the situation Apr 29 16:36:31 another might be that the OE reputation procedes its developers Apr 29 16:37:03 jnc: on that forum, yes Apr 29 16:38:21 * koen wonders if blacklisting some IPs on ewi would help Apr 29 16:38:33 Hi, I am checking a h6300 fstab for the bug 1595... Does anyone know what following line would do in fstab? usbdevfs /proc/bus/usb usbdevfs rw 0 0 Apr 29 16:38:55 is usbdevfs still needed on 2.6? Apr 29 16:39:01 off the forum, a typical response might be "we don't have time to fix everything. if you don't like it, our development model is open, and you are welcome to read the GettingStarted doc and contribute patches" Apr 29 16:39:06 usbfs on /proc/bus/usb type usbfs (rw) Apr 29 16:39:17 lamikr|laptop: s/usbdevfs/usbfs Apr 29 16:39:27 chrw@home:~$ grep usb /etc/fstab Apr 29 16:39:27 none /proc/bus/usb usbfs defaults 0 0 Apr 29 16:39:30 koen: it changed names, still there though Apr 29 16:39:32 my desktop Apr 29 16:39:35 koen@dominion:~$ mount | grep usb Apr 29 16:39:36 usbfs on /proc/bus/usb type usbfs (rw) Apr 29 16:39:36 none on /proc/bus/usb type usbdevfs (rw) Apr 29 16:40:01 koen: 800E for bandwidth? how much for the machine? Apr 29 16:40:16 zecke: that's machine + bandwidth Apr 29 16:41:19 opteron 146, 1GB ram, 2x160GB disks and unlimited bandwidth Apr 29 16:41:29 and 160GB of backup space via ftp Apr 29 16:41:51 03rpurdie 07org.oe.dev * r890854f1... 10/packages/qemu/ (files/nodocs_cvs.patch qemu-native_cvs.bb qemu_cvs.bb): qemu cvs: Disable document generation and also use any gcc-3.4 or gcc-3.3 in preference to gcc if present for qemu-native Apr 29 16:42:38 only opty 146? Apr 29 16:42:44 hi CoreDump Apr 29 16:42:49 hi Apr 29 16:42:57 at least a dual core 152.... Apr 29 16:43:12 jnc: 8x875 would be better Apr 29 16:43:19 i don't know how much better Apr 29 16:43:29 mount did not show anything for me even I have that usbdevfs currently in my h6300 fstab... So it apparently not needed when using usbgadget for booting over nfs. Apr 29 16:43:33 you might end up paying more for rack maintainence if a cpu goes out Apr 29 16:44:00 lamikr|laptop: usbfs is needed only for usbhost machines Apr 29 16:44:09 i'm not a big fan of dual core CPUs (security reasons), but they're great in hosting situations Apr 29 16:44:18 koen: oh well Apr 29 16:44:25 03rpurdie 07org.oe.dev * ra63ecbda... 10/conf/machine/blueangel.conf: blueangel.conf: Add autoload some kernel modules (#515 from Robert Woerle) Apr 29 16:44:27 koen: I have ~200 on my paypal account Apr 29 16:45:30 wtf? Apr 29 16:45:37 "you need windows installer 3.0" Apr 29 16:46:06 koen: the latest upgrade to stop piracy Apr 29 16:46:19 which is easy to skip ;) Apr 29 16:46:50 10 minutes with google Apr 29 16:47:17 it's trying to tell you something. "uuuuuuse uuuubuntuuuuu" Apr 29 16:48:41 * jnc reads more posts Apr 29 16:49:21 this should not have gotten any further than the death threat :( Apr 29 16:49:43 heh death threat? Apr 29 16:49:52 kill the posts, change the topic to "OE forums does not tolerate death threats. Go home." and lock it Apr 29 16:50:14 * CoreDump|home didn't see such a thing Apr 29 16:50:29 LOL: http://www.oesf.org/forums/index.php?showtopic=19136&pid=125097&st=0&#entry125097 Apr 29 16:50:41 CoreDump|home: sashz said I should get killed Apr 29 16:50:57 heh this guy is nuts Apr 29 16:51:06 probably a 13yr old Apr 29 16:51:07 you're both in love Apr 29 16:51:09 jnc: the OESF guy hates our guts since I said he stole our name Apr 29 16:51:26 stole the name? i don't understand Apr 29 16:51:35 CoreDump|home: I have his CV at other machine Apr 29 16:51:46 jnc: OESF has nothing todo with OE Apr 29 16:51:47 jnc: OESF - OpenEmbedded Software Fundation Apr 29 16:51:59 koen: except 1 first letters Apr 29 16:52:01 ~oesf Apr 29 16:52:02 well, oesf is newer name of ZUG - it provide forum for Zaurus/Ipaq/Simpad/etc users (see zug). It has nothing common with OpenEmbedded (except first two words), or http://www.oesf.org/ Apr 29 16:52:05 s/1/2 Apr 29 16:52:10 http://www.angstrom.com/ Apr 29 16:52:13 d'oh Apr 29 16:52:19 ljp: yep Apr 29 16:52:21 ljp: hey Apr 29 16:52:39 so sashz is the forum 'min? Apr 29 16:52:48 koen: childisch Apr 29 16:52:51 jnc: too many times I saw 'OE forums' Apr 29 16:53:01 koen: ask him to create ACL for the pdaXrom forum Apr 29 16:53:10 zecke: why should I? Apr 29 16:53:29 zecke: as far as I'm concerned that forum should die Apr 29 16:53:34 zecke: or be renamed Apr 29 16:54:27 hrw: the "in OE/OZ we have Abiword + Abiword-plugin-NAME" post really was unnecessary Apr 29 16:54:52 if sashz wants to keep his confusing forum name, then it's his responsibility to direct people to OpenEmbedded where appropriate Apr 29 16:55:02 jnc: oesf is not sash Apr 29 16:55:08 well then um Apr 29 16:55:10 jnc: it was for naming Apr 29 16:55:17 who is sashz? Apr 29 16:55:39 even better, who runs oesf? Apr 29 16:55:50 offroadgeek Apr 29 16:56:04 jnc: sashz is developer of pdaX (the only one iirc) Apr 29 16:56:20 oh Apr 29 16:57:02 is it likely for OpenZaurus distro to disappear Apr 29 16:57:05 in favor of Angstrom Apr 29 16:57:19 longterm: yes Apr 29 16:57:28 jnc: OZ will stay for 2.4 powered Apr 29 16:57:29 D'oh Apr 29 16:57:36 dvds don't work in a cdrom drive Apr 29 16:57:45 * koen hunts for a pata dvd drive Apr 29 16:58:06 koen: should we ask for donations? Apr 29 16:58:41 * CoreDump|home playes some ET Apr 29 16:58:44 you could perhaps ask to remove OZ forum topics from OESF forums, and then ask that Angstrom be added Apr 29 16:58:47 plays Apr 29 16:59:31 this is assuming, that OESF forums are not the official place for OZ support Apr 29 16:59:53 jnc: we don't want angstrom in those forums Apr 29 17:00:32 zecke: when OE has need for donations (in case of hosting/machines) we should ask for donations Apr 29 17:01:19 zecke: I know of some companies that would sponsor a tinderbox machine Apr 29 17:01:36 koen: we need to host it somewhere? Apr 29 17:01:40 ah, found a dvd drive Apr 29 17:01:54 I have some 'verbal' offers as well, but these people movel slooow Apr 29 17:01:54 hmm. then i have to agree with the rest of you saying that OESF and OE are in a naming conflict Apr 29 17:02:27 does either project posess a formal corporate entity? Apr 29 17:05:56 How goes the bug-hunting weekend guys? Apr 29 17:06:00 jnc: OE has non in any case Apr 29 17:07:30 trying to think of a suggestion to replace OPEN and EMBEDDED in the abbreviation OE Apr 29 17:07:36 not getting very far, heh Apr 29 17:08:09 Cwiiis: koen and hrw are killing most bugs ;) Apr 29 17:08:42 wow my first physics assignment is done Apr 29 17:09:03 zecke: congrats :) Apr 29 17:09:46 RP: I don't know how I could spend ~3 to 4 hours with that... Apr 29 17:09:57 RP: actually it is stuff from high school atm Apr 29 17:10:55 zecke: Its all this chatting on irc - very distracting ;-) Apr 29 17:12:00 np: Katie Melua - Nine Million Bicycles (0:49 / 0:00) Apr 29 17:12:02 wtf is oesf? Apr 29 17:12:04 RP: we will have 12 assignments, I hope I need to do only 6 of them (to reach 50% of the points) Apr 29 17:12:18 Crofton_|laptop: a forum full of clueless zaurus users Apr 29 17:12:28 heh Apr 29 17:12:34 the name is confusing Apr 29 17:12:57 Crofton_|laptop: it was ZUG before (ZaurusUserGroup) Apr 29 17:13:03 RP: how can I backport BitBake changes without a bug report? Apr 29 17:13:09 that would make more sense Apr 29 17:13:46 zecke: Do I have to file one before you can backport the fixes? :-/ Apr 29 17:13:54 zecke: Its obviously broken ;-) Apr 29 17:14:11 RP: it is so obvious, that no one - besides you - noticed it ;) Apr 29 17:15:50 doing the bitbake backports now Apr 29 17:16:28 zecke: If you want bug reports I'll add them. It just seems a little pointless Apr 29 17:16:53 RP: right, but that is the point ;) Apr 29 17:16:57 (don't file reports now) Apr 29 17:18:24 RP: did you notice the chdir bug in the cache? Apr 29 17:19:02 zecke: Yes, I found a way to make it crash bitbake - it should just throw an error Apr 29 17:19:50 zecke: It should probably print a warning if it can't chdir Apr 29 17:20:15 zecke: It was with invalid settings elsewhere but it still shouldn't crash Apr 29 17:21:08 RP: do you know how to mark patches as applied? Apr 29 17:21:32 zecke: in the bugzilla? no... Apr 29 17:21:52 zecke: mark them as obsolete and add coment Apr 29 17:22:07 hrw: Have you been reguarly using #431? Apr 29 17:22:16 zecke: -> details -> [v] obsolete Apr 29 17:24:17 thanks Apr 29 17:25:15 431? have to check what was it Apr 29 17:25:37 #431 is something for uv1 Apr 29 17:25:51 heh Apr 29 17:26:14 RP: 431 is not mine Apr 29 17:26:24 s/for/from even Apr 29 17:26:43 hrw: Sorry, 524 :-/. 431 was the attachment number Apr 29 17:28:00 RP: the pickle clean up showed an error as well? Apr 29 17:28:46 RP: no - did not use it Apr 29 17:29:12 zecke: I guess we don't need that one for the stable branch - its just an obvious cleanup Apr 29 17:29:31 RP: had problem once, wrote 524 then ignored it when workingcopy was fscked Apr 29 17:29:37 hrw: ok - I'm thinking of applying a revised version Apr 29 17:30:34 ok Apr 29 17:33:14 RP: would you be mad if I let this pickle thing drop under the table? Apr 29 17:33:23 zecke: no Apr 29 17:34:18 'Bruce Almighty' in 0.5h... Apr 29 17:36:17 RP: I rejected your patch, please don't lart me at any forum Apr 29 17:36:46 zecke: I'll sulk for days now ;-) Apr 29 17:37:11 RP: and don't switch poky to another build system Apr 29 17:39:24 ;)) Apr 29 17:39:49 RP: okay comitted all changes to bitbake-1.4 Apr 29 17:39:51 RP: feel free to hack my patch into professional one Apr 29 17:40:37 okay food Apr 29 17:40:40 cya later Apr 29 17:40:40 hrw: There's just a couple of more cases it needs to catch :) Apr 29 17:40:55 zecke: thanks - cya Apr 29 17:41:01 food.... Apr 29 17:41:19 zecke: can I go food with you? Apr 29 17:41:21 :D Apr 29 17:45:15 Need to go, goodbye. Apr 29 17:45:41 heh Apr 29 17:46:14 some people really need to go back to primary school and study the so called 'reading' part again Apr 29 17:46:32 where 'some people' read 'sashz' Apr 29 17:46:39 lol Apr 29 17:46:54 I blog something about angstrom, and he reads it as OZ Apr 29 17:47:18 * koen tries to find out how 'angstrom' and 'oz' look like similar words Apr 29 17:47:33 TheMasterMind1: a lot of your patches went in Apr 29 17:48:20 Does anyone have an oz354fam083 checkout handy to check what the SRC_URI of xstroke points at? Apr 29 17:48:42 viewmtn? Apr 29 17:49:03 I bet it points to http://people.freedesktop.org/~pb/xstroke-0.6.tar.gz Apr 29 17:50:04 . o O (nice how all bugfixes go from .dev to branches) Apr 29 17:50:06 *cough* Apr 29 17:50:41 koen: i saw that Apr 29 17:50:59 It points to oesources, same as .dev Apr 29 17:51:44 koen: you are OZ troll - dont remember? Apr 29 17:52:08 hrw|food: I am? Apr 29 17:52:19 hrw|food: I don't remember getting hired by TT Apr 29 17:52:39 ':) Apr 29 17:54:11 I can write bug ridden C++, so it could be true ;) Apr 29 17:54:25 hah Apr 29 18:07:43 koen: who claimed that Apr 29 18:07:44 re Apr 29 18:08:12 zecke: claimed what? Apr 29 18:08:57 koen: being hired by TT? Apr 29 18:09:37 20:02 < hrw|food> koen: you are OZ troll - dont remember? Apr 29 18:09:39 20:02 < koen> hrw|food: I don't remember getting hired by TT Apr 29 18:11:24 hum.. JavaScriptCore_cvs.sourceforge.net__20060212.tar.gz is not available from oesources, and there is a proxy error on the official cvs repository :-/ Apr 29 18:11:48 sourceforge sucks Apr 29 18:12:01 shit... I used too much curry ;( Apr 29 18:12:39 cyrilRomain: http://ewi546.ewi.utwente.nl/mirror/www.openzaurus.org/official/unstable/3.5.4/sources/ - use it as CVS_TARBALL_STASH Apr 29 18:13:50 hrw|tv: thanks Apr 29 18:14:16 hrw|tv: bitbake svn can have multiple tarball stashes Apr 29 18:14:22 hrw|tv what wasu cooking ? Apr 29 18:15:25 koen: cool Apr 29 18:15:44 but we should add it to the oesource, or update the local.conf.sample file with this repository if multiple tarball stashes is possible, don't you think ? Apr 29 18:15:46 gremlin[it]: chicken breast + 'chinese' vegetables - all on flying pan Apr 29 18:16:10 hrw|tv: you are using bitbake to make chinese food? Apr 29 18:16:45 mhh seem good hrw|tv :) .... Apr 29 18:16:54 Zero_Chaos: no - bitbake can order pizza only Apr 29 18:16:55 * cyrilRomain worry about new OE users... maybe because I am :) Apr 29 18:17:10 bitbake curry_rice ... Apr 29 18:17:14 Zero_Chaos: kergoth wrote bitbake to order pizza in one command Apr 29 18:17:18 gremlin[it]: no rice here Apr 29 18:17:35 ... set preferred_provide_curry='just a bit' Apr 29 18:17:36 :) Apr 29 18:19:49 * RP -> food Apr 29 18:20:20 koen: do you know scmbug? Apr 29 18:21:35 zecke: http://www.mkgnu.net/?q=scmbug Apr 29 18:21:48 ? Apr 29 18:22:10 koen: yes Apr 29 18:22:17 looks nice Apr 29 18:22:17 koen: http://www.bugzilla.org/docs/2.22/html/integration.html Apr 29 18:23:13 aaargh Apr 29 18:23:13 perl Apr 29 18:23:36 yes perl :} Apr 29 18:25:18 lol http://bugzilla.mkgnu.net/show_bug.cgi?id=417 Apr 29 18:28:12 hmm I don't get what scmbug is about :} Apr 29 18:31:12 koen: how do you specify multiple tarball stash ? Apr 29 18:31:37 * koen looks at RP Apr 29 18:32:19 treke|home: ping Apr 29 18:38:37 zecke: monotone 0.26 server update to yesterday Apr 29 18:38:57 koen: I need to repull right? Apr 29 18:39:20 mtn pull should do Apr 29 18:39:31 hrm Apr 29 18:39:34 * koen wonders Apr 29 18:40:24 OT: Anyone feels like creating New Data Sets for OE? Apr 29 18:41:14 mtn: warning: error: Mismatched epoch on branch org.openembedded.dev. Server has 'c71a44c1c5f666857dd7946a4a04044ded1f639f', client has '2352aea7f9238e4bad2693fa60f5c7a0a148aadf'. Apr 29 18:42:59 sorry about that Apr 29 18:43:03 * koen needs more coffee Apr 29 18:43:59 koen: hehe Apr 29 18:44:26 zecke: yes? Apr 29 18:45:16 treke|home: how much time do you have to spare? Apr 29 18:45:24 With regards to? Apr 29 18:45:25 treke|home: http://websvn.kde.org/trunk/bugs/bugz/ Apr 29 18:45:36 treke|home: http://bugs.kde.org/weekly-bug-summary.cgi Apr 29 18:46:04 " *grasp* SECURITY!!!!! Who writes such scripts??! Get a new job!" Apr 29 18:46:13 That doesnt really clear it up for me :) Apr 29 18:46:46 treke|home: oh right. I'm asking if you could look into integrating some KDE BugZilla bits into our setup Apr 29 18:46:56 treke|home: e.g. to have a graphing/change overview Apr 29 18:47:26 Not likely to happen Apr 29 18:47:34 fair enough :) Apr 29 18:48:35 zecke: weekly-bug-summary.cgi looks reasonable standalone Apr 29 18:48:38 No idea what version of Bugzilla these are even against :p Apr 29 18:49:04 koen: I failed with creating DataSets for our bugzilla Apr 29 18:49:22 I'm not adverse to installing any of them, I'm just not interested in maintaining a major fork of bugzilla :p Apr 29 18:50:25 treke|home: right, understandable Apr 29 18:50:35 03koen 07org.oe.dev * r87363a60... 10/conf/distro/angstrom-2006.9.conf: Angstrom 2006.9: switch to newer binutils Apr 29 18:50:44 http://bugs.kde.org/reports.cgi?product=konqueror&output=show_chart&datasets=NEW%3A&datasets=ASSIGNED%3A&datasets=REOPENED%3A&datasets=UNCONFIRMED%3A&datasets=RESOLVED%3A&banner=1 <- I want such reports ;) Apr 29 18:51:12 any idea what release of bugzilla they use? Apr 29 18:51:47 treke|home: no :( Apr 29 18:52:14 kde has own fork iirc Apr 29 18:52:25 treke|home: well, I think we can get graphing working without including any KDE bits Apr 29 18:52:39 mhh after a coplete wipeout of /build/tmp dir i got this errror trying to 'bitbake opie-image' Apr 29 18:52:50 ld: cannot find -lqte Apr 29 18:53:03 NOTE: package libqpe-opie-1.2.1+cvs-20060429-r1: task do_compile: failed Apr 29 18:53:04 check their SVN Apr 29 18:53:08 gremlin[it]: ~pastebin Apr 29 18:53:23 zecke just 2lines :) Apr 29 18:53:37 gremlin[it]: check the makefile Apr 29 18:53:41 gremlin[it]: hey Apr 29 18:53:41 hey gerwinin Apr 29 18:53:46 hey Koen Apr 29 19:03:23 * koen tries an altboot enabled abi image Apr 29 19:05:40 zecke simply seem no libqte.so is builded :( ... Apr 29 19:05:54 only libqte.a is present ... Apr 29 19:06:27 which machine? Apr 29 19:06:57 gremlin[it]: staging/arm-linux/qt2/lib has no libs? Apr 29 19:07:00 ep93xx Apr 29 19:07:34 zecke yes ... dir is emply Apr 29 19:08:58 RP: survey is now updated Apr 29 19:09:00 gremlin[it]: hmm bad Apr 29 19:16:09 hrm Apr 29 19:16:14 busybox 1.1.2 is fscked Apr 29 19:23:13 busybox 'cat' has one option... '-u' - in bash it is ignored... Apr 29 19:48:20 hmm.. one thing I like in nvidia - under x11 I have more then one xv port so can watch movie without closeing tvtime Apr 29 20:03:14 my tv card only ever works with capture (as opposed to overlay) :/ Apr 29 20:04:05 Cwiiis: x11? Apr 29 20:04:15 hrw|tv: Yup Apr 29 20:04:43 Cwiiis: with XFree4.3/Xorg6.8 I used 'v4l' module as extra Apr 29 20:05:35 hrw|tv: I'm not sure how much difference it makes anyway :) Apr 29 20:06:06 Cwiiis: dont know too - xorg 6.9/7.0 does not use it Apr 29 20:25:23 hrw|tv: something is very wrong with the new busybox Apr 29 20:25:33 hrw|tv: trying to find out what... Apr 29 20:26:13 * koen is tempted to install bash + coreutils + textutils Apr 29 20:40:30 03rw 07org.oe.dev * r8c282fac... 10/packages/ (classpath/classpath.inc jikes/jikes_1.22.bb): classpath, jikes: reinstate myself as MAINTAINER. Apr 29 20:42:12 koen: unsermake for konq-e? strange Apr 29 20:42:38 zecke: ? Apr 29 20:42:51 koen: latest angstrom konq-e error Apr 29 20:43:18 it builds for me Apr 29 20:43:29 ah I thought it was your bug report Apr 29 20:44:22 with the patch to qte-mt opie builds like a charm for anstrom Apr 29 20:44:27 angstrom* Apr 29 20:51:54 ok. time to bed - cu Apr 29 21:21:56 * chouimat is away: food and stuff Apr 29 21:23:31 hrw|gone: cya Apr 29 21:26:08 koen: I will make you flithy rich (some day) if you stop trolling with reenoo and france ;) Apr 29 21:30:00 koen: we could put tinderbox on monotone.nslu2-linux.org Apr 29 21:30:15 (it's hosted for free to OSUOSL) Apr 29 21:31:29 rwhitby: would it be possible to get the mandantory access for me? Apr 29 21:32:20 zecke: what access is required? Apr 29 21:32:23 rwhitby: I would prefer to run something like bonsai for OE and BitBake? Apr 29 21:32:35 rwhitby: me hacking a cgi-bin directory Apr 29 21:32:42 rwhitby: Postgres or MySQL Apr 29 21:32:51 rwhitby: and then email Apr 29 21:33:58 mysql, email, adding users with perms to dirs - all no problem Apr 29 21:34:48 (it's just a Debian Sarge system) Apr 29 21:35:29 zecke: it's not much load, right? Apr 29 21:36:12 rwhitby: atm it is not much load Apr 29 21:36:56 rwhitby: once my bonsai is done, I might ask you to host it to help the tinderbox out Apr 29 21:38:12 zecke: do you know ka6sox? Apr 29 21:39:49 rwhitby: no Apr 29 21:41:35 I'm trying to bitbake readline-4.3-r2 for armv4 (ipaq-pxa270), but it fails in the compilation with errors like this: Apr 29 21:42:19 kokos: use pastebin Apr 29 21:42:25 ~pastebin Apr 29 21:42:26 well, pastebin is a place to paste your stuff without flooding the channel - try http://pastebin.com/, or http://pastebin.ca, or http://channels.debian.net/paste, or for #oe use http://oe.pastebin.com Apr 29 21:43:25 just one line (simplified): .../bin/ld: ERROR: /home/..../libgcc.a(_divsi3.oS) uses hardware FP, whereas libreadline.so.4.3 uses software FP Apr 29 21:43:52 koen: did you change anything after building your toolchain? Apr 29 21:44:04 kokos: did you change anything after building your toolchain? Apr 29 21:44:30 * cyrilRomain found a pastebin plugin for xchat http://www.limewire.org/blog/?p=27 and would like to find an irssi equivalent :) Apr 29 21:44:49 zecke: no, I don't think so... do you mean something specifically? Apr 29 21:45:21 also, I can build other packages. It's just this one which seems to have a problem Apr 29 21:46:45 I checked the configure logs for any suspicious messages, but didn't found anything relating to fpu, float, etc. Apr 29 21:52:09 how well is this suspend to disk working in linux? Apr 29 21:58:19 Who can control the main site of OE? Apr 29 21:58:46 zecke: Its on hh.org so you need hh.org access and the approriate permissions Apr 29 22:00:53 RP: Isn't it using a content management system? Apr 29 22:01:24 zecke: not sure Apr 29 22:02:56 kokos: I'd check its using the right compiler Apr 29 22:03:46 koen: Have you tried 2.4 and 2.6 machine builds in the same tmp dir with multimachine? Apr 29 22:04:00 RP: Can I check that in the configure logs? Apr 29 22:04:16 koen: I'm wondering if glibc is compatible. I guess it must be for the old Zaurus rootfs systems to have worked... Apr 29 22:05:04 kokos: probably. Also check the paths its searching for libraries - is it pulling in some from your build system? Apr 29 22:08:01 03rpurdie 07org.oe.dev * r780f2c7e... 10/packages/mpfr/mpfr_2.1.1.bb: mpfr: Set infodir to avoid touching build system /usr (fixes #581) Apr 29 22:08:05 03rpurdie 07org.oe.dev * r2ec7d3e6... 10/ (6 files in 4 dirs): Install depmod-2.4 and depmod-2.6 in cross staging instead of plain depmod so 2.4 and 2.6 kernel builds can coexist with multimachine (bug 581) Apr 29 22:25:52 gosh I'm tired again Apr 29 22:27:31 zecke: What do you mean about patches applied in that bug? Apr 29 22:30:43 'do_patch' Apr 29 22:31:23 RP: I had/have a broken BitBake that will not apply this patch to libtool-native Apr 29 22:31:35 njs: you don't have 35 grants? Apr 29 22:31:47 zecke: I think they're just rerun bitbake Apr 29 22:31:55 but I could be wrong Apr 29 22:32:06 njs: git now has fsck objects as well ;) Apr 29 22:32:10 zecke: huh? Apr 29 22:32:20 njs: http://lists.gnu.org/archive/html/monotone-devel/2006-04/msg00302.html <-reading it now Apr 29 22:32:23 no configure/compile either yet it has something there to stage... Apr 29 22:32:30 ah :-) Apr 29 22:32:47 (what is a "fsck object"?) Apr 29 22:33:08 njs: you can fsck your git objects nowadays Apr 29 22:33:16 njs: Its an integrity check of the objects in the git repo Apr 29 22:33:22 ah, yeah, you've been able to do that forever Apr 29 22:33:31 it's the same as monotone's "db check" Apr 29 22:39:16 03rpurdie 07org.oe.dev * reaa71390... 10/classes/package.bbclass: package.bbclass: Add sanity check for duplicates in PACKAGES (bug 674) Apr 29 22:40:55 hehe, see linuxdevices.com Apr 29 22:42:23 hmm... Apr 29 22:43:42 Software Index :) Apr 29 22:43:49 zecke: well great news :) Apr 29 22:44:23 gosh compiling uclibc takes 10 seconds Apr 29 22:44:31 gcc initial and gcc takes hours :} Apr 29 22:45:01 sourceforge cvs server not working :( can I get gtk-webcore packages from somebody? Apr 29 22:45:24 myopiate: yes Apr 29 22:45:32 myopiate: familiar.handhelds.org Apr 29 22:46:00 http://familiar.handhelds.org/source/v0.8.4-rc1/sources/ Apr 29 22:46:11 add this as tarball stash, or download the stuff manually Apr 29 22:46:33 zecke: How do you add a tarball stash ? Apr 29 22:47:08 cyrilRomain: conf/local.conf Apr 29 22:47:20 zecke: yes, but what is the syntax ? Apr 29 22:47:34 cyrilRomain: see local.conf.sample Apr 29 22:47:50 zecke: ooops, thx Apr 29 22:47:52 cyrilRomain: SRC_TARBALL_STASH = " http://your.stash.biz/foo" Apr 29 22:49:46 currentNRCore_cvs.sourceforge.net__20050430.tar.gz ist not there :( Apr 29 22:50:19 why does automake_1.7.7.bb say RDEPENDS_${PN} = "autoconf-native perl-native" ?! Apr 29 22:50:55 RP: I think I found the source of my problem: this option looks quite suspicious: -Wl,-rpath,/usr/lib Apr 29 22:51:19 kokos: Yes, thats normally a bad sign Apr 29 22:51:24 RP: hmm in contrast to RDEPENDS = ""? Apr 29 22:51:34 zecke: hum.. SRC_TARBALL_STASH ? this is not documented. However there is CVS_TARBALL_STASH. But i did not succeed to use multiple tarball stash: i tried multiple CVS_TARBALL_STASH and even with += without success :-/ Apr 29 22:51:38 zecke: note this isn't automake-native Apr 29 22:51:54 RP: oh that is bad Apr 29 22:52:03 cyrilRomain: Which version of bitbake? Apr 29 22:52:13 RP: 1.4.1 Apr 29 22:52:26 cyrilRomain: mutliple entries will only work with latest svn (I checked that in this morning) Apr 29 22:52:59 RP: backport candidate even if it is a feature? Apr 29 22:53:04 RP: oh ok, :) Apr 29 22:53:21 zecke: I thought you'd tell me off for adding features ;-) Apr 29 22:53:32 But I'm not sure, where can fix this. In the Makefile? or must it be somewhere in my bb configuration? Apr 29 22:53:55 kokos: You'll have to work out where it comes from... Apr 29 22:57:33 RP: okay, it might take a while until I find it, as I'm not that experienced yet. But at least I seem to be on the right track now. Thanks for your help. Apr 29 22:57:38 cyrilRomain, do you have currentNRCore_cvs.sourceforge.net__20050430.tar.gz, I noticed you were looking for JavaScriptCore_cvs.sourceforge.net__20060212.tar.gz earlier Apr 29 22:58:07 kokos: I'm fairly sure its the source of the problem so you're on the right track :) Apr 29 22:58:37 myopiate: I do not have such package in my DL_DIR Apr 29 23:00:59 myopiate: the two package can be found http://ewi546.ewi.utwente.nl/mirror/www.openzaurus.org/official/unstable/3.5.4/sources/ but not on the default oesource repository. Apr 29 23:01:44 :( Apr 29 23:01:44 gcc4.0.2 and uclibc are not working... Apr 29 23:03:09 03rpurdie 07org.oe.dev * r8d934dcb... 10/packages/automake/ (7 files): automake: Fix automake RDEPENDS and add the perl modules needed (bug 852) Apr 29 23:06:38 cyrilRomain, thanks... looked there before but didn't find it ... found it now. Apr 29 23:06:53 * myopiate puts his glasses back on Apr 29 23:09:14 rwhitby: did you test gcc > 4 for armeb already? Apr 29 23:10:18 my server looks upset. I'll probably disappear shortly :-/ Apr 29 23:16:45 zecke: I don't use armeb - you need to ask lennert Apr 29 23:17:24 I'm tired, I'll probably disappear shortly ;) Apr 29 23:17:37 nite all Apr 29 23:27:26 pb__: ping Apr 29 23:27:33 pb__: how can we install setuid binaries? Apr 29 23:29:31 mickey|weekend: you are the 2nd hit for hotshot ;) Apr 29 23:32:09 mickey|weekend: python hotshot that is Apr 29 23:54:44 RP: hey Apr 29 23:55:02 RP: bitbake profiles shows: most time is spend in handle_bb_data Apr 29 23:55:19 RP: (4.5 secs) and almost two seconds in 'sync' Apr 30 00:08:43 'night all Apr 30 00:08:52 cya Apr 30 00:23:45 ET_DYN PeMRxS --- --- RW- - /home/ich/hacking/embedded/oe/builds/simpad/tmp/staging/arm-linux/lib LAZY tmp/staging/arm-linux//lib/libgcrypt.so.11.2.1 Apr 30 00:23:49 ouch Apr 30 00:26:06 unbelievable the security libs have rpath set Apr 30 01:00:40 pb__: for the records, my FBSD crashed for the first time (in vmware though) Apr 30 01:03:47 nite Apr 30 01:22:09 * chouimat is away: Zzzz Apr 30 06:27:31 my build of gpe-image is dying on the install of bluez-utils-dbus. ipkg-cl is complaining that a file exists and it is expecting input. Apr 30 06:31:59 It's because it's installing both bluez-utils-dbus and bluez-utils-nodbus Apr 30 06:36:52 that would be a bug Apr 30 06:37:13 Good Morning Apr 30 06:38:15 koen: Is it a known bug, that sc/mmc access does not work any more (at least with my cl860) ? ... Apr 30 06:38:36 yesterdays kernel works on my c700 Apr 30 06:38:37 Partiotion is mounted, but gives an IO error once you want to access ... Apr 30 06:38:53 well, it mounted the partition Apr 30 06:38:59 I didn't try to read from it Apr 30 06:39:18 Even a dd causes an IO error ! Apr 30 06:44:42 koen, is there a way to find out what is demanding bluez-utils-nodbus? maybe like ... bitbake -i ... whowants bluez-utils-nodbus Apr 30 06:45:01 myopiate: not that I know off Apr 30 06:45:46 so grep it is Apr 30 06:46:44 koen: Also sound is still not working. Alsamixer thinks my boxer is a corgi. No sound at all makes it unusable for me, because alarm does not work. Apr 30 06:47:14 pulled .dev &; building opie-image Apr 30 06:50:48 ~botmail for zecke: do you think it's time for releasing bitbake 1.4.1? Apr 30 06:52:15 koen, bluez-utils-nodbus was an extradepends in my machine config because I put it there, silly me. Apr 30 07:04:17 good morning Apr 30 07:04:26 Bernardo: morning Apr 30 07:05:51 whooo Apr 30 07:05:57 * koen fixed the segfaulting Apr 30 07:06:01 ~lart busybix Apr 30 07:06:01 * ibot flings poo at busybix Apr 30 07:06:05 ~hail coreutils Apr 30 07:06:06 * ibot bows down to coreutils and chants, "I'M NOT WORTHY!!" Apr 30 07:07:30 03koen 07org.oe.dev * rb0c52710... 10/conf/distro/angstrom.conf: Apr 30 07:07:30 angstrom.conf: ship coreutils by default Apr 30 07:07:30 * this is ~900kB more in jffs2 Apr 30 07:07:30 * it doesn't segfault or do 'function not implemented' like busybox Apr 30 07:21:48 morning Apr 30 07:21:50 hey hrw Apr 30 07:22:08 uv1: 1. all c7x0 are 'corgi' for most of parts in kernel Apr 30 07:22:31 uv1: 2. sound is working on c7x0 - set playback volume to 80 or less Apr 30 07:22:48 uv1: 3. some sd/mmc cards does not work under 2.6 ;( Apr 30 07:23:11 ~curse linuxdevices for 'OpenEmbedded forums' Apr 30 07:23:12 May you be reincarnated as a Windows XP administrator, linuxdevices for 'OpenEmbedded forums' ! Apr 30 07:23:43 hrw: yeah we stole the 'bitbake' name from OESF Apr 30 07:24:52 lardman|gone: I agree (oesf msg) Apr 30 07:27:09 yay! Apr 30 07:27:12 I won the lottery Apr 30 07:27:21 a whopping 16.40 euros Apr 30 07:29:25 ;) Apr 30 07:30:16 gremlin[it]: bon giorno Apr 30 07:31:07 ciao koen :) Apr 30 07:31:20 are u study italian ? :) Apr 30 07:31:42 no, I just remember a few words from when I last visited rome :) Apr 30 07:32:14 hihihi :) Apr 30 07:32:47 this remember me i have to start plan my hollidays ;) ... Apr 30 07:35:07 koen: it's safe to change the MACHINE in local.conf and just relaunch 'bitbake xxx-image' ? how i can work with 2 different machine minimizing disk space ? Apr 30 07:35:39 gremlin[it]: if you use multimachine.inc, yes Apr 30 07:38:32 yay! Apr 30 07:38:45 I can ping hosts from my angstrom eabi image Apr 30 07:38:51 ~lar busybox again Apr 30 07:38:59 ~lart busybox again Apr 30 07:38:59 * ibot beats busybox again over the head with a microkernel Apr 30 07:48:58 koen: i'm not able fo find info about multimachine.inc :( Apr 30 07:49:08 No sound at all after trying hrw's hints. Also the MMC card was working with the 2.6.15 kernel ... Apr 30 07:50:06 Can anyone send me a alsamixer configuration file /etc/asound.state of a sound working c7x0 ? Apr 30 07:54:34 EABI rocks :) Apr 30 08:57:53 Just have seen, that bitbake complains, that lots of preferred opie packages (1.2.2) are not available. Does that cause them not in image ? Apr 30 08:58:46 uv1: no, it will build another version Apr 30 08:58:57 uv1: opie 1.2.2 isn't out yet :) Apr 30 09:21:41 cu Apr 30 09:23:05 hey zecke Apr 30 09:23:07 hey pH5 Apr 30 09:23:09 good morning Apr 30 09:23:56 koen: hey, do we see a preassure for this (bitbake 1.4.1)? Apr 30 09:23:58 pH5: hey Apr 30 09:24:38 hey koen, is there a quick way to overwrite angstrom's SRCDATE? I wonder if angstrom-2006.9.conf should use ?= to set the SRCDATE.. Apr 30 09:24:40 hey zecke Apr 30 09:25:02 pH5: ah right, feel free to change that Apr 30 09:25:08 zecke: all those bugfixes Apr 30 09:25:14 zecke: especially the cvs fetcher one Apr 30 09:26:13 koen: cvs fetcher? that was a change to bitbake.conf Apr 30 09:26:18 koen: and is in .dev and branches Apr 30 09:28:33 03pH5 07org.oe.dev * r206b62d3... 10/conf/distro/angstrom-2006.9.conf: angstrom-2006.9.conf: set SRCDATE via ?= so it can be overwritten in local.conf Apr 30 09:44:09 hi all anyone home? Apr 30 09:44:17 home=here Apr 30 09:44:57 I'm using monotone 0.26 and tried darmou@firebird:/stuff $ monotone --db=OE.db pull monotone.vanille.de org.openembedded.dev Apr 30 09:45:12 message back is monotone: misuse: database /stuff/OE.db contains revisions but no rosters Apr 30 09:45:38 I migrated the database so what gives? Apr 30 09:45:45 darmou: venge.net/monotone Apr 30 09:45:51 darmou: read all about 0.26 Apr 30 09:46:01 darmou: also monotone 0.26 binary is called mtn ;) Apr 30 09:46:31 ok special things I need to do to get 0.26 working ok thanks, yep I know I symlikned mtn to monotone Apr 30 09:46:56 darmou: wait until enough people have monotone 0.26+ installed Apr 30 09:47:08 darmou: until then use monotone < 0.26 and > 0.22 for OE Apr 30 09:47:22 ok thanks zecke Apr 30 09:47:44 darmou: monotone had a major rework of how they store the data (roster) Apr 30 09:48:01 darmou: sadly this makes mtn 0.25 and 0.26 incompatible (you can only upgrade to 0.26) **** ENDING LOGGING AT Sun Apr 30 09:59:57 2006