**** BEGIN LOGGING AT Sat May 06 09:59:56 2006 May 06 10:03:20 koen: is it broken in trunk as well? May 06 10:03:43 let me check May 06 10:05:50 I just returned from the most inpsiring place May 06 10:05:56 koen: it is not a BitBake bug May 06 10:06:01 base.bbclass should run update_data May 06 10:06:48 Ah, I am guilty for that probelm :-/ May 06 10:09:02 zecke: You think that update_data we moved in bitbake was doing more than its location suggested? May 06 10:09:23 zecke: trunk has the same problem May 06 10:11:17 koen: could you make gcc print which FPU setting it is using? May 06 10:11:24 koen: I'm quite sure it will be soft float May 06 10:11:38 RP: well, it was just an accident base.bbclass had the data available ;) May 06 10:12:04 zecke: I don't have a toolchain at the moment May 06 10:15:10 zecke: I suspect bitbake should call update data before passing off control. We should also clone that data structure first though... May 06 10:16:43 RP: yes before firing events we should create a copy May 06 10:17:12 RP: but we should leave the update_data to bbclass? May 06 10:18:23 zecke: Probably not, the data should be updated and should only need a further update if the class does something to OVERRIDES etc May 06 10:35:10 lol May 06 10:35:22 glibc has been gentoofied May 06 10:35:31 crumbs May 06 10:35:42 pb_: it can not be compiled without optimisation May 06 10:35:52 pb_: at least O2 is accepted as optimisation May 06 10:35:58 ah yes May 06 10:36:23 * koen uses -Os May 06 10:36:42 it does have this --enable-omitfp option which enables the true gentoo cflags May 06 10:53:31 Hello May 06 10:54:23 I recently compiled a bootstrap image for my asus wl500g May 06 10:54:46 Now it's created in tmp/deploy/images, May 06 10:54:54 morning May 06 10:55:18 - it's a file called bootstrap-image-unstable-20060506-wl500g.rootfs.jffs2 - May 06 10:55:35 But how do I now get this file in the flash of my router? May 06 10:55:49 I know how to upload trx files with tftp May 06 10:56:11 but if I upload the image direct via tftp the router doesn't work May 06 10:56:37 so it seems that I have to convert it in a certain way to something that can be uploaded... May 06 10:57:48 sorry, I don't know May 06 11:02:10 Trying to translate this problem to english ... I've got a 4GB SD-Card of A-Data with 150x speed. I'd like to use it with a Zaurus SL-C1000. It doesn't work with the original BS nor with CACKO 1.23 and OZ with OPIE tells me this: mmcblk0: mmc 0:0002 SD 3912704 KiB -how to make a retur without sending the line? -----mmcblk0: p1 < > ---------return---------attempt to access beyond end of device------ret-------mmcblk0p1: rw=0, want=4, limit=2------ May 06 11:03:12 It works good with a Windows PDS :-( May 06 11:18:43 koen: works here (TM) May 06 11:18:57 oh wait May 06 11:32:03 hey pH5 May 06 11:32:38 hey koen May 06 11:34:16 koen: hmm works here May 06 11:34:29 hi zecke May 06 11:35:39 hi ph5 May 06 11:35:54 hi pb_ May 06 11:41:34 Trying to translate this problem to english ... I've got a 4GB SD-Card of A-Data with 150x speed. I'd like to use it with a Zaurus SL-C1000. It doesn't work with the original BS nor with CACKO 1.23 and OZ with OPIE tells me this: mmcblk0: mmc 0:0002 SD 3912704 KiB -how to make a retur without sending the line? -----mmcblk0: p1 < > ---------return---------attempt to access beyond end of device------ret-------mmcblk0p1: rw=0, want=4, limit=2---- May 06 11:46:22 odo_: you didn't get an answer the first time May 06 11:47:23 odo_: and mmc is not a property of Opie May 06 11:47:27 odo_: but of the kernel ;) May 06 11:49:36 May be, I'm not so deep into oz yet - CoreDump told me to ask here ... May 06 11:52:20 ~lart CoreDump May 06 11:52:20 * ibot duct-tapes CoreDump to the floor and drools on him May 06 11:52:26 I suppose #openzaurus is what you need. May 06 11:53:11 more to the point, RP is what he needs ;) May 06 11:54:12 heh May 06 11:55:19 odo_: not meant to be hostile May 06 11:55:26 odo_: A-Data with 150x speed what is that? May 06 11:55:40 odo_: what size is reported by linux, what size is printed on the card May 06 11:55:49 odo_: who reports to access beyond end of device? May 06 11:55:51 odo_: filesystem? May 06 11:55:55 odo_: is it mounted? May 06 11:56:03 odo_: tried creating a filesystem on this card? May 06 11:56:08 odo_: does this problem persist May 06 11:56:19 odo_: you wrote a long text, with little information :) May 06 11:56:43 well it works for windows so I guess VFAT is on there May 06 11:56:44 Didn't know what info you need ... May 06 11:56:56 odo_: don't worry May 06 11:57:03 A-Data is the manufacturing company of that SD-Card May 06 11:57:16 odo_: that part is uninteresting for now May 06 11:57:29 odo_: use windows, copy all the data from the card May 06 11:57:34 Size, as I wrote: mmc 0:0002 SD 3912704 KiB May 06 11:57:58 odo_: I'm not good at numbers, does this match with what you expect? May 06 11:58:06 theres's no windos around here, exept that wich i can open to exchange the air May 06 11:58:48 the card should have about 4GB and that's what mmc tells here May 06 11:58:59 well create a vfat filesystem on this card May 06 11:59:07 koen: mickeyl|bbl: Parallel starting of initscripts doesn't buy us anything :\. If you want to launch everything in parallel you have to handle dependencies (like networking depends on udev etc). As soon as you add that logic you are slower on a Z as a normal boot. May 06 11:59:10 odo_: or when is 'access' beyond... is reported? May 06 11:59:26 CoreDump|home: how did you test it? May 06 11:59:42 access beyond is the last report, than the log stands still May 06 11:59:45 zecke: I wrote a new rcS script for parallel launches May 06 12:00:06 maybe in C it could be faster but not in a shell script... May 06 12:00:56 shell scripts are certainly very slow on the zaurus. That wouldn't be a good way to implement something that is meant to be a performance boost. May 06 12:01:44 it was worth a try tho. If you ignore dependencies and launch selected scripts serial not parallel, you _are_ a little bit faster May 06 12:02:24 dependencies are more elegant IMO ;) May 06 12:05:19 zecke: "attempt ..." comes after "mmcblk0: p1 < >" May 06 12:16:44 hi, I needed the result of uname in a .bb file and looked how others did this. May 06 12:17:02 in several bb files I find references to uname -r. Eg altboot: May 06 12:17:03 altboot/files/altboot.func: if ( uname -r | grep -q "2.6." ) May 06 12:17:28 does this run on the host? if not, am I not getting the wrong uname? May 06 12:17:36 there is no uname in e.g. cross May 06 12:18:01 eFfeM: nope, that runs on the PDA May 06 12:18:12 ok May 06 12:18:29 why do you need the ouput of "uname" on your host anyway? May 06 12:18:36 wrong example. May 06 12:18:37 libnet/libnet-1.0.2a/configure.patch: case "`uname -r`" in May 06 12:19:09 it's using the hosts uname in this example May 06 12:19:35 CoreDump|home: I'm trying to get dhclient working. it didn't compile for arm, found out the cause fixed it then it still does not work because there are #ifdefs to get the kernel major version May 06 12:20:16 CoreDump|home: that's what i figured out as well. I don't know libnet, but it seems odd that a confiure run would depend on the uname of the host .... May 06 12:21:29 I am looking in dclient and the makefile.conf has: -DLINUX_MAJOR=$(MAJORVERSION) -DLINUX_MINOR=$(MINORVERSION) May 06 12:21:40 on Zaurus we define a runtime variable containing our targeted kernel, you could hack the configure / ifdef's to use it if there is such a thing for your distribution May 06 12:22:10 eventually configure resolves MAJORVERSION etc using uname May 06 12:22:17 yeah May 06 12:23:11 I'm fairly sure that dhclient will not work on Zaurus either; didn't see anything in the bb file or patches dir that indicates anything there May 06 12:23:38 it was just a sugestion who you could do it May 06 12:23:48 if there is anyone who runs a cross-compiled dhclient, please speak up May 06 12:24:12 CoreDump|home: I was thinking if we should not have a uname exe in the cross dir May 06 12:24:26 but making such a thing is beyond my capabliities May 06 12:24:29 if your device only supports kernel 2.4 you could hard-code that in. If it supports 2.6 and 2.4 there is a way your distro lets you select the targeted kernel. This fucntion could be used in your makefile then May 06 12:25:20 Yeah, did that, but was not too happy about it. not a very clean solution May 06 12:25:46 will probably have to do that, cant fix the whole world ... May 06 12:25:51 heh May 06 12:26:30 yo've got to use the distro mechanism to set the kernel as there is simply no way to know which one is used otherwise May 06 12:27:27 probably not May 06 12:27:31 or at least not easy May 06 12:31:57 odo_: try fdisk May 06 12:32:06 odo_: create a new fs... May 06 12:32:12 zecke: will do May 06 12:32:14 odo_: create a new fs on the mmcblock May 06 12:35:07 zecke: you mean something like "fdisk mmcbl0" ? May 06 12:38:10 mkfs.ext2 -m0 /dev/mmcblkp1 or something like that May 06 12:38:28 I'mnot too sure about the devicefile May 06 12:39:39 CoreDump: thanks I'll try that first, but fdisk /dev/mmcblk0 also is telling me some interesting info. May 06 12:40:08 try deleting the existing partition and create a new one May 06 12:40:52 CoreDump|home: Do not use a partotopn May 06 12:41:08 CoreDump|home: it sounds like a kernel bug or wrong partition size May 06 12:41:12 CoreDump|home: or wrong fs size May 06 12:41:32 and recreating the partition in Linux might fix that ;) May 06 12:41:43 hi guys May 06 12:41:47 mickeyl: hey May 06 12:42:18 hi mickeyl May 06 12:45:14 mkfs tells me that it had problems writing the superblocks but afterwards it's saying that the fs will be checked all 34 mounts ...bla,bla. May 06 12:45:39 odo_: this dsounds bad May 06 12:45:48 trying to mount mmcblk0 fails May 06 12:46:03 odo_: well, slower please May 06 12:46:13 odo_: you did fdisk and created an empty partition table and wrote that? May 06 12:46:31 odo_: can you hook up your Z to the internet w/ reachable IP? May 06 12:46:43 zecke: no, I run mkfs first May 06 12:46:52 no internet for the z May 06 12:47:12 on partition or the complete disk? May 06 12:47:42 on mmcblk0, I think that's the whole thing May 06 12:48:11 meh May 06 12:48:28 delete the partitions first before mkfs'ing the whole disk May 06 12:48:34 nm May 06 12:48:48 that mkfs got rif of the partition most probably... May 06 12:48:58 ok, looking for the options of fdisk May 06 12:50:15 option d tells me that there are no partitions yet May 06 12:51:07 l lists many partition types, which one to takr? May 06 12:52:26 that shouldn't matter May 06 12:52:50 then I take a6: Open BSD May 06 12:53:42 hook that Z up to a reachable IP and I'll have a look May 06 12:55:30 It's on the powerbook now May 06 12:56:25 then create a dummy ssh login on the powerbook for me to login, from the powerbook i can SSH to the Z May 06 12:56:41 I' will have to reconfigure firewalls first May 06 12:57:47 which port is ssh? May 06 12:58:21 22? May 06 12:58:52 ssh 22/tcp # SSH Remote Login Protocol May 06 12:58:52 s May 06 13:05:38 zecke (and other OE SoC mentors): proposal submitted May 06 13:06:02 nest question: which of the connection should I activate? usbf, usb0, usbd0? May 06 13:07:21 03pH5 07org.oe.dev * re8a30a4e... 10/packages/zeroconf/ (4 files in 3 dirs): zeroconf: add 0.9 May 06 13:07:26 03pH5 07org.oe.dev * r1f6955f7... 10/packages/avahi/ (avahi_0.6.9.bb avahi_0.6.10.bb): May 06 13:07:26 avahi: bugfix release 0.6.10 May 06 13:07:26 - See http://avahi.org/milestone/Avahi%200.6.10 May 06 13:07:36 IIRC usb0 for kernel 2.6 May 06 13:09:20 koen: we could have versioned build depends :) May 06 13:16:47 zecke: yes and many other nice features May 06 13:17:04 zecke: like packaged toolchains May 06 13:20:19 my #1 feature is fixing up the -dev packages that way May 06 13:20:27 although it will be a lot of work May 06 13:22:11 mickeyl: is there an easy way to 'compile' python May 06 13:22:21 mickeyl: to find syntax errors even in modules? May 06 13:22:38 yeah May 06 13:23:05 dunno which module it is offhand May 06 13:23:31 probably compiler May 06 13:24:15 alternatively, intergrate pychecker May 06 13:24:20 integrate, even May 06 13:28:14 CoreDump: I'll have to leave,now. Let's try it another time, thanks a lot. May 06 13:28:23 np, cya May 06 13:28:44 zecke: thanks, too, bye May 06 13:29:11 mickeyl: are you able to build wpa-gui? May 06 13:29:27 let me check, could take a while May 06 13:29:30 * mickeyl starts a build May 06 13:29:48 for me it doesn't build uicmoc or qt3 May 06 13:30:00 ok, let me try from scratch May 06 13:30:10 shouldn't the qt3x11 class take care of that? May 06 13:30:23 do you inherit both qmake and qt3x11 ? May 06 13:30:25 you need to May 06 13:30:53 it does May 06 13:30:56 k May 06 13:31:00 inherit qmake qt3x11 May 06 13:31:02 let's see May 06 13:31:13 koen: TARGET_FPU is 'right' for me May 06 13:31:29 zecke: it isn't over here :/ May 06 13:31:43 http://bugs.openembedded.org/show_bug.cgi?id=938 May 06 13:32:45 hi mickeyl May 06 13:33:36 hey pb_ May 06 13:34:07 lol wtf May 06 13:34:18 ~lart pan for beeing a memory hog May 06 13:34:18 * ibot duct-tapes pan to the floor and drools on him for beeing a memory hog May 06 13:34:34 640Mb shesh May 06 13:34:57 zecke: my mentoring application is pending :-} May 06 13:35:38 pb_: good :) May 06 13:35:55 zecke: I guess that I now await Erik's pleasure May 06 13:36:24 pb_: right, I can't approve you May 06 13:36:38 right, no problem May 06 13:41:09 koen: try trunk of bitbake May 06 13:44:08 trying May 06 13:49:35 morningmorning May 06 13:50:18 zecke: trunk works May 06 13:50:31 DISTRO = "angstrom" May 06 13:50:31 TARGET_FPU = "soft" May 06 13:57:01 ah right :} May 06 13:57:11 now I remember, I updated after my first tests May 06 14:12:43 good morning folks May 06 14:16:27 hey chris144 May 06 14:22:15 hi koen May 06 14:22:57 I did the survey yesterday, do you know who initialized it and who are the folks looking at the results? May 06 14:23:20 chris144: zecke, and zecke + other people :) May 06 14:23:29 koen: ok May 06 14:24:31 koen: you too? May 06 14:25:19 no idea, but iirc the result will be published in some form May 06 14:25:22 * koen looks at zecke May 06 14:26:12 bleh May 06 14:26:21 the uni mx is going berzerk again May 06 14:26:54 chris144: Gerwinin created the first set of questions May 06 14:27:08 chris144: I entered them in the survey tool, people helped shaping the questions May 06 14:27:16 chris144: I hit the start button some days ago ;) May 06 14:27:42 we will let it run another two weeks probably May 06 14:28:03 then I will ask RP, pb_, gerwinin, mickeyl, koen to help me getting a meaning from the results May 06 14:28:05 zecke: ok, was quite interesting to do the survey May 06 14:28:37 zecke: especially the questions about what annoys one most.....:-) May 06 14:28:38 I hope the questions weren't too leading May 06 14:29:19 hehe May 06 14:29:30 zecke: I thought quite a while, but figured that I was never so annoyed at oe to freak out... :-) May 06 14:29:43 zecke: well, I am looking forward to the results... May 06 14:29:54 chris144: hehe May 06 14:29:59 :-) May 06 14:30:23 chris144: my hopes are: We see what annoys the most May 06 14:30:35 chris144: maybe we get help to address it May 06 14:31:00 chris144: and if someone asks us what one could/should do for OE May 06 14:31:06 chris144: we can point him to the results ;) May 06 14:31:14 zecke: yeah of course May 06 14:31:33 koen: I'm doing a backport of this change May 06 14:49:14 hey mithro May 06 14:49:40 hey koen May 06 14:49:43 how goes life? May 06 14:49:48 pretty good May 06 14:49:53 it's nice and sunny this weekend May 06 14:51:17 that is good May 06 14:51:33 guess i should head to bed May 06 15:16:30 * RP returns May 06 15:17:08 RP: wb May 06 15:38:34 anything similar to the NSLU2 but with an internal HD?" May 06 15:38:49 luke-jr_: nas100d May 06 15:38:55 supported by OE? May 06 15:39:00 yes May 06 15:39:17 nice May 06 15:39:44 should work for a good low-power Asterisk box, I'm hoping May 06 15:39:45 luke-jr_: check http://article.gmane.org/gmane.comp.misc.nslu2.devel/701 May 06 15:40:23 damn harddisk is full :} May 06 15:40:32 zecke: OE does that ;) May 06 15:41:43 slug*, angstrom, familiar is just too many :} May 06 15:46:45 * RP updates the arm machine registry to standardise all the Zaurus entries... May 06 15:47:20 03koen 07org.oe.dev * ra13924ba... 10/packages/libgpewidget/libgpewidget_0.111.bb: libgpewidget: add 0.111 May 06 15:50:47 ah, rmk gave agreed making you the maintainer of all machines? May 06 15:50:58 ehm May 06 15:51:00 s/gave// May 06 15:54:51 koen: All Zaurii except Tosa which Dirk now has May 06 15:55:01 nice May 06 15:56:43 * koen reads May 06 15:56:47 * koen reads http://rpsys.net/openzaurus/ May 06 15:57:01 so poodle support is complete, except for sound May 06 15:57:25 koen: Yes. And sound shouldn't be that much of an issue thanks to ASoC May 06 15:59:08 do you happen to know what the status of uda1380 asoc support is? May 06 15:59:39 koen: Not tested as much as Liam/myself would like May 06 15:59:53 aha May 06 16:00:03 I guess I should revive my h2200 May 06 16:00:40 koen: I think it needs an h2200 kernel person to tie up the loose ends - neither myself to Liam have access to a uda1380 May 06 16:00:56 that would mean mreimer or giorgio May 06 16:01:30 We talked to giorgio about ASoC and he didn't seem to see the benefits :-/ May 06 16:48:17 03mickeyl 07org.oe.dev * rce8b36a5... 10/classes/ (qt3x11.bbclass qt4x11.bbclass): May 06 16:48:17 qt3x11.bbclass: add DEPENDS on qt3 May 06 16:48:17 qt4x11.bbclass: add DEPENDS on qt4 May 06 16:48:21 03mickeyl 07org.oe.dev * r887360a0... 10/packages/ (5 files in 4 dirs): libqanava, lyx, texmaker, wpa-gui: no need to explicitly depend on qt, inheriting will do May 06 16:48:46 * chouimat is away: lunch!!!! May 06 16:50:48 cya later May 06 16:52:21 koen: this should fix the wpa-gui issue May 06 16:52:46 mickey|sun: hehe May 06 16:52:53 SUN Solaris? or SUN SUN ;) May 06 16:52:58 hehe May 06 16:53:01 solarsystem sun May 06 16:53:21 mickey|sun: Do you know a way how I could redefine the semantic of sys.exit? May 06 16:57:15 zecke: the semantics? May 06 16:57:28 atexit might be interesting for you May 06 16:57:35 (guessing) May 06 16:57:59 mickey|shower: sys.exit not exiting ;) May 06 17:14:06 mickey|shower: thanks May 06 17:16:40 koen: you might have seen it already, try bitbake-1.4 again May 06 17:20:42 mickey|shower: what is my pw? May 06 17:21:22 wuahauZ3 May 06 17:21:39 hi zecke May 06 17:21:43 hi mickeyl May 06 17:27:49 drat, raining again May 06 17:28:07 pb here was the sun the whole day May 06 17:28:19 woglinde: I deduce that you are not in the uk May 06 17:28:41 pb ;) I know May 06 17:28:55 very tiresome, I need to cycle into town now May 06 17:29:10 oh well, I guess I will just have to get wet May 06 17:30:09 pb_: no roof for your transportation vehicle yet? May 06 17:31:11 heh May 06 17:31:22 cya later May 06 17:31:37 I was just typing a mail to bero to say that kernel-discuss might give more attention to his patches May 06 17:31:41 zecke: not my bike :-} May 06 17:31:44 oh, he left May 06 17:32:05 I guess I should leave too May 06 17:32:06 pb dont you have such a rain cape for cycles? May 06 17:32:35 sadly no May 06 17:32:39 maybe I should invest in one May 06 17:32:53 later all May 06 17:32:57 * pb_ -> May 06 17:33:01 bye pb May 06 17:47:46 03rpurdie 07org.oe.oz354x * r003e30f7... 10/ (5 files in 3 dirs): zaurusd: c7x0 mixer changes applied upstream. Use our own zaurus-hinge.in script instead of patching. Update the zaurusd cvsdate to obtain the recent changes. May 06 17:47:51 03rpurdie 07org.oe.dev * r224e3cbc... 10/packages/zaurusd/ (3 files in 2 dirs): zaurusd: Use our own copy of zaurus-hinge.in instead of patching. Clean up the .bb a bit. May 06 17:58:27 03rpurdie 07org.oe.oz354x * r10ae2f81... 10/packages/zaurusd/files/zaurus-hinge.in: zaurusd: Add missing script. May 06 18:24:13 (unsigned char, int) May 06 19:42:40 building qt-x11-free is taking forever May 06 19:59:42 JustinP: qt-x11-free provides qvfb May 06 20:00:37 * koen was grepping in the wrong deploydir May 06 20:13:34 hey May 06 20:14:25 hey zecke May 06 20:39:54 njs`: ping May 06 20:45:42 hi, can anyone point me to a .bb file which gives a good example how to add creating a doc package? May 06 20:45:44 now I get: May 06 20:45:44 NOTE: the following files were installed but not shipped in any package: May 06 20:45:47 NOTE: /usr/man/man5/dhcp-options.5 May 06 20:45:47 NOTE: /usr/man/man5/dhcp-eval.5 May 06 20:45:49 ... May 06 20:48:09 changed my mind, but still need advice.... May 06 20:49:14 I wanted dhclient, but the package was not working for my distro. I noticed that the version was very old, so I created a new bb file. However now I found that there is another package dhcp which actually is a much newer version May 06 20:49:55 shouldn't we just retire dhclient or add a warning somewhere (this took me quite some time and I want to avoid others falling in the same pit ....) May 06 20:50:12 eFfeM: FILES_${PN}-doc += "/usr/man/man5/*" May 06 20:50:49 eFfeM: right, we should remove that pitfal May 06 20:50:51 zecke, thanks for the info, I'll try to rememver for next time May 06 20:51:22 eFfeM: in bitbae.conf we have PACKAGES = "${PN} ${PN}-dev ${PN}-doc" May 06 20:51:40 i think I can remove, but it might also be in some conf/distro files; also I do not feel experienced enough to do this May 06 20:51:59 eFfeM: in that case mail oeAThh.org May 06 20:52:27 zecke: the whole bitbake is still a little bit of black magic to me; I know how a bb file should look from studying others but that is about it. May 06 20:52:42 (and of course from bothering koen with all kind of questions) May 06 20:53:06 eFfeM: as mickeyl claimed "a solution is coming" May 06 20:53:11 he plans writing a book ;) May 06 20:53:27 I did as well :} May 06 20:53:52 yeah, probably I should get on the oe mailing list as well (I'm an nslu2 person and for now only following the nslu2 mailing lists) May 06 20:54:07 * eFfeM already has his name on a book (on Minix, lol) May 06 20:55:28 eFfeM: post it :) May 06 20:55:34 eFfeM: we will cc you if you ask to May 06 20:55:47 currently 99% of the mail are bugzilla mails May 06 20:57:16 zecke: not "post the book" I hope. Actually I don't know if Andy Tanenbaum would agree :-) I still have some LaTeX source somewhere .... May 06 20:57:32 not too interested in the bugzilla mails May 06 20:57:40 * eFfeM already gets enough mail :-) May 06 20:58:39 anyway: package dhcp works for me, I'll remove dhclient from the slug distro (I feel confident about that) and mail to oe May 06 21:09:51 mail sent May 06 21:36:31 koen: ping! May 06 21:36:52 03eFfeM 07org.oe.dev * r80048cf1... 10/conf/distro/slugos-packages.conf: slugos-packages.conf: added dhcp (includes dhclient) May 06 21:54:19 *hmm* May 06 21:55:36 hmm damn barbecue May 06 22:02:19 zecke: pong May 06 22:03:46 " In particular, May 06 22:03:46 it starts off "you should already know C and pointers", which might May 06 22:03:46 have scared off most of the unserious proposals." May 06 22:03:48 heh May 06 22:03:56 (dan kegel about wine SoCs) May 06 22:04:37 hehe May 06 22:14:48 rwhitby: your folks are invited with to participate in the survey as well May 06 22:14:54 (if it wasn't obvious) May 06 22:19:34 zecke: can we have NAS added to the list of target applications? May 06 22:19:47 rwhitby: sadly not :( May 06 22:20:02 I'm sorry for that May 06 22:20:15 but these are the things one forgets :} May 06 22:20:29 I should have showed you the draft May 06 22:20:39 no problem - I'll get people to write 'NAS' in other, so you can sort on it. May 06 22:20:55 rwhitby: two people already did that May 06 22:22:35 (heh, one was me) May 06 22:22:54 what's the survey url again? I'll do a post to nslu2-linux now. May 06 22:24:10 survey.openembedded.org/index.php?sid=6 May 06 22:24:40 heh - I can't see it again cause I have completed it. What is the question number where we should put 'NAS' in Other ? May 06 22:25:12 What types of devices are you developing for? May 06 22:25:18 0004 May 06 22:25:56 whooo pretty late now, goodf nite folks May 06 22:26:08 cya May 06 22:26:11 Do you only want developers to fill out the survey, or nslu2-linux users too? May 06 22:26:20 cu May 06 22:27:18 (we have over 8 thousand users on the nslu2-linux lists) May 06 22:27:52 (and 150 on the developers list) May 06 22:28:49 I would propose only sending it to the developers list, as you don't want 8000 nslu2-linux users giving you feedback about the nslu2-linux firmware distribution, as opposed to the OpenEmbedded system on which it is based. May 06 22:29:21 right, that would be nice May 06 22:31:09 sent. May 06 22:31:48 rwhitby: would you - once done collecting the results - be interested in looking at the raw results? May 06 22:32:03 zecke: yes May 06 22:32:34 rwhitby: if it wouldn't be php I would give you an account right away ;) May 06 22:36:38 BTW: we now have a regression suite for BitBake as well May 06 22:40:44 rwhitby: did SlugOS win any awards? May 06 22:41:20 zecke: we haven't entered it in any competitions :-) May 06 22:41:40 pretty much the same applies for OE May 06 22:41:54 I want to create a http://www.kde.org/awards/ like page for OE May 06 23:08:58 03rpurdie 07org.oe.dev * re8d99763... 10/packages/xlibs/ (35 files): xlibs: Update XORG_MIRROR SRC_URIs to match upstream changes May 06 23:16:51 RP: Did you close bug 935? May 06 23:17:04 RP: Did you try to compile libXau for example? May 06 23:17:12 Laibsch: Not quite yet - one more commit needed, then we can close May 06 23:19:00 Cool. May 06 23:19:04 ok, we can close it now :) May 06 23:19:13 Looking forward to the stuff. May 06 23:20:14 03rpurdie 07org.oe.dev * rf139a1f3... 10/packages/ (9 files in 9 dirs): various: Update remaining XORG_MIRROR SRC_URIs to match upstream changes May 06 23:24:00 305 open bugs... May 06 23:24:58 304... :) May 06 23:28:08 zecke: What do you want to do with bug 829? May 06 23:29:48 03coredump 07org.oe.dev * r776e1905... 10/packages/altboot/ (11 files in 8 dirs): altboot: Split altboot into altboot (PACKAGE_ARCH = all) and altboot-conf (ARCH = ${MACHINE}, cleanup altboot_0.0.0.bb and rework some redirects to use tty1, not tty0. May 06 23:30:13 RP: target it for 1.6 May 06 23:30:42 RP: I have a patch for 'now' support in fetchers May 06 23:31:23 zecke: ok. Didn't we have some issues raised about "now" support? May 06 23:31:53 sure ;) May 06 23:32:06 by rm -fr the work directory of an app, will I force a refetch ? May 06 23:32:17 CSMan: no May 06 23:32:25 bitbake -b file -cfetch -f May 06 23:32:31 but it won't be unpacked ;) May 06 23:32:50 lovely bitbake May 06 23:33:18 CSMan: well you could have changes in your workdir ;) May 06 23:33:38 RP: bitbake-tests is slowly getting useful May 06 23:39:44 cya May 06 23:39:57 hmm tomorrow is my non OE/non computer day :} May 06 23:41:21 odd... once again i am confused on OE policy ... May 06 23:41:52 why would you remove older version packages? May 06 23:42:31 the bb and patches woudl still exist for it and if someoen chose the old version for a reason it would be availible May 06 23:42:42 emte: Security holes? bugs? May 06 23:43:02 emte: We can't maintain too many versions? May 06 23:43:24 RP, this is regarding the dhclinet request to remove old version May 06 23:43:36 dhclient* May 06 23:43:44 emte: His reason was it caused him a lot of confusion May 06 23:43:57 that can be said of any package May 06 23:44:12 it doenst mean we should toss away the other 10 versions of glibc May 06 23:44:26 ( as an example ) May 06 23:44:51 No. I don't know enough to comment on the specific case of dhclient May 06 23:45:15 It is true we have stale bits of metadata though and when we identify them, we should really remove then if they're not needed/used May 06 23:45:24 me either which is why i am asking, as it seems a dangerous precident May 06 23:46:01 remove would cause problems if someone expects to be able to build older images May 06 23:46:11 move to a stale branch i would agree May 06 23:46:23 We remove old packages all the time... May 06 23:46:43 i am aware May 06 23:47:14 At some point i'm going to nuke LTT in OE in favour of LTTng. You could argue for keeping LTT but its not really going to be much use to anyone... May 06 23:47:53 anyone that you know of, or anyone who olny wants newer versions May 06 23:48:59 LTT development is dead. It will only run on a certain set of old 2.6 kernels. If you're going to do profiling, you're likely to be using something much less broken... May 06 23:49:21 If you want the files, you can always pull them out the archives. This is why we have an SCM afterall May 06 23:49:43 i guess my thought is that you should be able to build any image within OE history if you wanted May 06 23:50:10 In practice that won't happen - you're better turning back the clock on your SCM May 06 23:50:35 which is still not viable until an SCM is decided May 06 23:50:51 We have an SCM May 06 23:51:04 apperently its not concrete May 06 23:51:21 If we change, tailor can covert the history May 06 23:51:28 yeah ... May 06 23:51:46 btw tailor doesnt work properly with nrewer monotone's again May 06 23:52:03 or mercurial May 06 23:52:50 but yes, i do see what your saying May 06 23:59:54 'night all May 07 00:28:31 somebody please merge OE! May 07 00:28:39 i see... xmms-embedded runs only in opie /o\ May 07 01:14:43 packages/xlibs/libx11_X11R7.0-1.0.0.bb needs to be renamed for fetching to work May 07 01:19:46 Laibsch, are you refering to all the x11 rename bugs that have been filed and commited? May 07 01:39:01 emte: They are committed. but libx11 is not properly done. May 07 01:39:12 Both for fetching and for compiling, BTW. May 07 01:39:36 The S = from xorg-common.inc throws it off, too. May 07 01:39:50 The "S = blah"-line from xorg-common.inc throws it off, too. May 07 01:41:03 xorg-libs.inc May 07 01:41:57 Since the bb file is named libx11_X11R7.0-1.0.0.bb, bot the fetch URL and the S directory will have X11R7.0 twice May 07 01:45:11 both May 07 02:31:38 RP: i sold that c3000. no more whining from me about it not suspend/resume'ing properly :P May 07 03:34:20 fun stuff: monotone: read from fd 6 (peer monotone.vanille.de) failed, disconnecting May 07 03:34:25 mickey|shower: monotone: read from fd 6 (peer monotone.vanille.de) failed, disconnecting May 07 03:34:42 mickey|shower: hmmm, seems to be working this time... May 07 07:12:04 hi all ... May 07 07:12:53 just a question, to build a familiar-0.8.3 image i have to change the DISTRO tag in local.conf or change repository ? May 07 07:25:42 you should be using the oz354fam083 branch and you have to change DISTRO May 07 07:25:53 and you should either clean tmp or make a new build dir May 07 07:28:41 mhh ok ... at now i'm using the devel branch ... but with DISTRO = familiar and DISTRO_VERSION = 0.8.3 ... seem to work May 07 07:33:11 morning all May 07 08:10:54 Good morning! May 07 08:11:09 good morning May 07 08:11:17 good morning all May 07 08:11:41 morning pH5 koen Laibsch May 07 08:11:51 gremlin[it]: you need to be using the oz354fam083 branch for the real thing May 07 08:13:05 mhh i hope that choosing a stable familiat it create an image ... current unstable don't finish neither opie of gpe image :( May 07 08:31:56 I am a bit confused. qpobox inherits opie. But "bitbake opie" cannot be done because nothing provides dependency opie. How would I check that all dependencies/inherits build fine for qpobox? May 07 08:37:26 Laibsch: opie is a class, not a separate package. just 'bitbake qpobox' should try to build all its dependencies. May 07 08:49:33 pH5: Did you ever get qpobox to compile? It is the "next-generation" of Japanese input methods from zaurus-ja.sourceforge.jp May 07 08:50:05 pH5: I wanted to make sure that everything but qpobox works before digging further down. May 07 08:58:46 Laibsch: no, I didn't. I just noticed that there are no DEPENDS yet in the qpobox .bb, so you'd have to look at the source or just try to compile to find the dependencies. I guess at least libqpe is needed... May 07 09:06:50 OK, I thought ashikase had provided a complete patch. May 07 09:07:38 qpobox fails because Roma2Kana is not defined. IIRC, libqpe is being built in the process due to inherit opie. May 07 09:07:55 What do you suggest to do find where Roma2Kana comes from? May 07 09:18:05 Laibsch: I guess it's somewhere in OpenPOBox, so this looks like a problem with the poboxserver build. May 07 09:20:11 OK, I guess too heavy for me now. May 07 09:21:36 But IIRC, ashikase mentioned in oesf.org that he successfully compiled this beast. May 07 09:22:01 I'll start a poboxserver build and try to find out where it fails. May 07 09:31:01 pH5: That's cool. Let me know if there is anything I can do to help. Looking forward to your patches. May 07 09:31:28 Maybe today will finally be the day that I can dump Sharp ROM. I WISH! May 07 09:38:51 bbl May 07 09:39:00 bye May 07 09:50:49 hi all **** ENDING LOGGING AT Sun May 07 09:59:57 2006