**** BEGIN LOGGING AT Thu Nov 16 03:00:07 2006 Nov 16 03:03:16 03lenehan 07org.oe.dev * ra86f3c5c... 10/ (4 files in 2 dirs): Nov 16 03:03:16 openssl: Use the new siteinfo stuff to determine the endianess and set the Nov 16 03:03:16 appropriate flag rather then manually processing the site file contents to Nov 16 03:03:16 do this. Nov 16 03:41:12 what is glibc-intermediate used for? Nov 16 03:50:57 VoodooZ: It's some sort of reduced glibc used to build gcc, and then the gcc it builds (using that libc) is then used to build the full glibc... or something like that. Nov 16 03:53:16 where does glibc-initial fit in then? Nov 16 03:53:54 3 steps process or 2? Nov 16 03:56:32 anyways, bed time. Nothing works anyways... Nov 16 03:57:58 VoodooZ: Not really sure.. Nov 16 03:58:04 nite VoodooZ **** BEGIN LOGGING AT Thu Nov 16 04:16:22 2006 Nov 16 07:15:26 morning all Nov 16 08:19:07 morining koen :) Nov 16 08:38:07 good morning all Nov 16 08:38:10 03lenehan 07org.oe.dev * rc04bf3e4... 10/ (1 site/endian-big site/endian-little): Nov 16 08:38:10 site: Add newlines to the last entry in endian-big and endian-little. Nov 16 08:38:10 This is just so that they show up on the console when grep for those Nov 16 08:38:10 entries or when you cat those files. Nov 16 08:38:23 03lenehan 07org.oe.dev * rbd26f8df... 10/ (1 packages/hping/hping2_1.9.9+2.0.0rc3.bb): Nov 16 08:38:23 hping2 2.0.0rc3: Use the new siteinfo stuff to determine the endianess and Nov 16 08:38:23 set the appropriate flag rather then manually processing the site file Nov 16 08:38:23 contents to do this. Nov 16 08:38:33 03lenehan 07org.oe.dev * ra1f1dc8d... 10/ (1 site/sh3-linux): Nov 16 08:38:33 site/sh3-linux: Replace the samaba entries with those from sh4. The current Nov 16 08:38:33 ones will build samaba but all the files will be shown with blank file Nov 16 08:38:33 names. This change has been previously made to all the other site files. Nov 16 08:38:33 This is part of a step towards merging the sh3 and sh4 site files. Nov 16 08:38:42 03lenehan 07org.oe.dev * r929667ff... 10/ (1 packages/net-snmp/net-snmp_5.1.2.bb): Nov 16 08:38:42 net-snmp 5.1.2: Use the new siteinfo stuff to determine the endianess and Nov 16 08:38:42 set the appropriate flag rather then manually processing the site file Nov 16 08:38:42 contents to do this. Nov 16 08:38:52 03lenehan 07org.oe.dev * rab6a3911... 10/ (1 site/sh3-linux site/sh4-linux): Nov 16 08:38:52 site/sh{3,4}: Update the sh3-linux and sh4-linux site files to match be Nov 16 08:38:52 adding missing entries from each site file to the other. Nov 16 08:39:05 03lenehan 07org.oe.dev * r84fe2f0f... 10/ (5 files in 2 dirs): Nov 16 08:39:05 site/sh{3,4}: Merge the sh3 and sh4 site files into a single common site file Nov 16 08:39:05 for both of them. Nov 16 08:39:05 03hrw 07org.oe.dev * r60dba0f5... 10/ (3 files in 3 dirs): perl: add support for x86_64 - close #1591 Nov 16 08:39:11 03hrw 07org.oe.dev * r31613cbe... 10/ (1 packages/tinymail/tinymail_svn.bb): Nov 16 08:39:11 tinymail: switched SRC_URI to HTTP method (was HTTPS) Nov 16 08:39:11 - both methods works, http does not need confirmation of SSL certificate Nov 16 08:39:13 so fetching via cronjob works Nov 16 08:39:18 03hrw 07org.oe.dev * rb7cca68e... 10/ (1 packages/ortp/ortp_0.7.1.bb): ortp: fixed SRC_URI Nov 16 08:39:24 03hrw 07org.oe.dev * r2cf62c3a... 10/ (1 packages/python/python-itools_0.13.0.bb): python-itools: fixed SRC_URI Nov 16 08:39:28 03hrw 07org.oe.dev * rbe5139ab... 10/ (4 files in 2 dirs): offlineimap: updated to 4.0.14 Nov 16 09:08:36 good morning Nov 16 09:12:52 do someone use a git version of xserver-kdrive ? i need a good src_date that is not broken Nov 16 09:13:24 morning Nov 16 09:13:47 yo hrw|work Nov 16 09:14:55 RP: I feel that bitbake trunk fetchall tries to restart failed fetch tasks Nov 16 09:28:03 hi all Nov 16 09:29:22 Genesis: afaik xcalibrate is broken when used with kdrive-git Nov 16 09:35:54 koen, : i think , i've pb with tslib structure & define Nov 16 09:36:00 RP: http://ewi546.ewi.utwente.nl/tmp/hrw/logs/1116-apt.log Nov 16 09:36:22 koen, : i try to fix it ugly for the moment Nov 16 09:37:30 in fact , i need a xserver that support dual head Nov 16 09:38:05 with the august patch of mallum , there is a switch -fb to choose the fb at lauchtime Nov 16 09:56:07 hrw|work: hmm :-/ Nov 16 09:59:46 hrw|work: I tried bitbake apt -c fetchall -k here and it worked as expected. I don't know why it doesn't do that for you :-/ Nov 16 10:00:10 hrw|work: Is your bitbake patched at all? Nov 16 10:08:32 RP: its clean from trunk Nov 16 10:08:46 RP: you tried against .dev or poky? poky has newer apt Nov 16 10:09:27 hrw|work: .dev - fetch failed, it marked the task as failed, tried again and errored out with no tasks executable Nov 16 10:09:44 will do such test here Nov 16 10:09:46 hrw|work: Does "bitbake apt -c fetchall -k" loop for you? Nov 16 10:10:02 started it now Nov 16 10:12:36 NOTE: Resolving missing task queue dependencies Nov 16 10:12:36 NOTE: Preparing Runqueue Nov 16 10:12:36 NOTE: All possible tasks have been run but build incomplete (--continue mode). See errors above for incomplete tasks. Nov 16 10:12:39 NOTE: Executing runqueue Nov 16 10:12:42 NOTE: build 200611160232: completed Nov 16 10:12:48 thats all Nov 16 10:12:59 hrw|work: That's the expected behaviour Nov 16 10:13:33 hrw|work: it try the fetch first though, right? Nov 16 10:14:01 RP: it tried in past Nov 16 10:14:33 hrw|work: Can you pastebin the full output from that run? Nov 16 10:14:55 sure Nov 16 10:15:30 hrw|work: It should look like http://pastebin.com/825593 Nov 16 10:16:46 http://pastebin.com/825595 Nov 16 10:17:19 no stamps at all for apt Nov 16 10:17:33 ah... w8. I cant test it now as I dropped them locally.. Nov 16 10:17:45 ~lart me wice Nov 16 10:17:46 * ibot gives me wice a good seeing to Nov 16 10:17:47 twice Nov 16 10:18:11 reverted, restarted Nov 16 10:20:17 http://ewi546.ewi.utwente.nl/tmp/hrw/logs/apt.log - result of bitbake apt -cfetchall -kv|tee log2 Nov 16 10:20:51 will recreate it without tee Nov 16 10:25:06 http://ewi546.ewi.utwente.nl/tmp/hrw/logs/apt2.log Nov 16 10:30:25 RP: apt 0.6.42.2 in poky is broken. apt-native works Nov 16 10:30:37 RP: autofoo.patch does not apply here Nov 16 10:32:48 hrw|work: That looks like its behaving correctly... Nov 16 10:33:00 I'll keep the apt issue in mind in poky, thanks Nov 16 10:33:24 RP: "bitbake -cfetchall world -vk" looks different Nov 16 10:33:45 hrw|work: I didn't have the -v Nov 16 10:35:00 http://ewi546.ewi.utwente.nl/tmp/hrw/logs/1116-1200-fetchall.txt.bz2 is from my failed build Nov 16 10:40:49 hrw|work: No mention of apt? Nov 16 10:41:27 moment- phone Nov 16 10:42:23 ops. wrong log Nov 16 10:45:21 I love to explain that when XYZ says "ERROR" take it seriously... Nov 16 10:45:25 ~lart users Nov 16 10:45:25 * ibot whacks users with the cluebat Nov 16 10:48:22 http://ewi546.ewi.utwente.nl/tmp/hrw/logs/1116-1200-fetchall2.txt.bz2 Nov 16 10:48:29 this is correct one Nov 16 10:57:04 zecke: no, 'ERROR' means actually 'free candy', so people ignore it Nov 16 10:57:19 koen: ERROR This will fail Nov 16 10:57:27 It failed and I have no clue what is wrong :) Nov 16 10:58:24 zecke: yeah Nov 16 10:58:48 RP: do you have any screenshots of the curses gui yet? Nov 16 10:59:13 zecke: did you notice a lot of the parse errors got fixed? Nov 16 10:59:25 koen: The demo I got from mickey is quite cool... Nov 16 11:00:18 koen: Its going to take a while before it gets implemented :-/ Nov 16 11:02:28 hrw|work: I can see what's wrong... Nov 16 11:02:37 RP: could you put it online somewhere? Nov 16 11:03:00 koen: Just wait ;-) Nov 16 11:08:23 koen: yes, i will have to rerun the test script Nov 16 11:08:26 koen: I will take a day off tomorrow to learn for my theoretical computer science test Nov 16 11:08:36 koen: I think hacking parsers qualify as studying :) Nov 16 11:10:11 RP: what it is? Nov 16 11:14:12 03hrw 07org.oe.dev * r097b9926... 10/ (4 files in 2 dirs): man-pages: update to 2.41 Nov 16 11:14:37 hrw|work: Its not reporting the failed tasks correctly by the looks of it :-/ Nov 16 11:16:23 hrw|work: We call raise bb.runqueue.TaskFailure(self.runq_fnid[build_pids[result[0]]], then del build_pids[result[0]] in another exception handler. I don't know what python will do with this... Nov 16 11:17:02 zecke: Any ideas what python will do in this case (at the bottom of runqueue.py) Nov 16 11:17:39 test it :) Nov 16 11:18:24 it will likely remove it, just skip the del Nov 16 11:18:39 it will be deleted as soon as the deps drop to zero anyway Nov 16 11:20:12 ah, result is of course different so it shouldn't be an issue Nov 16 11:27:58 heh.. looks like I will need to use linux32 anyway... Nov 16 11:28:05 ~curse 2.4-crapix Nov 16 11:28:15 May you be reincarnated as a Windows XP administrator, 2.4-crapix ! Nov 16 11:28:37 hrw|work: 'linux32 screen' works wonders Nov 16 11:36:52 RP: Is "lose contacts between electric components" proper english for saying that the contacts don't always make contact? Nov 16 11:37:44 loose contacts between electonic components Nov 16 11:37:50 thanks Nov 16 11:38:37 s/contacts/contact/ and maybe s/loose/intermittent/ Nov 16 11:39:48 hi CoreDump|home Nov 16 11:39:56 morning Nov 16 11:39:59 hi CoreDump|home Nov 16 11:41:01 hey CoreDump|home Nov 16 11:45:06 hrw|work: Can you mtn up and apply http://www.rpsys.net/openzaurus/temp/bitbake_rq-temp.patch to bitbake, then try to reproduce the problem? Nov 16 11:45:13 can Nov 16 11:46:32 RP: fetchall world -vk? Nov 16 11:46:47 hrw|work: yes please Nov 16 11:47:14 started Nov 16 11:47:17 [hrw@misc1 celinux-test]$ time bitbake -cfetchall world -vk|tee log.rp-fix1 Nov 16 11:47:19 hrw|work: What I didn't understand last time was the number of "NOTE: Removing failed runtime build target thcrut" Nov 16 11:47:42 RP: bitbake r675+patch Nov 16 11:48:06 hrw|work: You also pulled the fix to base.bbclass from .dev? Nov 16 11:48:58 yes Nov 16 11:49:05 mtn: modifying classes/base.bbclass Nov 16 11:49:05 mtn: updated to base revision 72e01ba791a9f5bc41e2f61a13635ad9cb1dcc87 Nov 16 11:49:06 ok, good :) Nov 16 11:49:24 take a minute to reparse bbfiles Nov 16 11:49:43 I'll be afk for a while... Nov 16 11:49:48 ok Nov 16 11:50:06 RP: the same 'a while' it takes for the ncurses frontend? ;) Nov 16 11:50:42 koen: In the sense that both timeframes are indetermined and will take as long as they take ;-) Nov 16 11:50:49 :) Nov 16 11:51:22 drat, I just found a 'FIXME- verbal description of LMV and NSEM' while doing a spellcheck Nov 16 11:53:03 * koen start writing a chapter about linear minimum variance estimation and null-space error minimization. Nov 16 12:00:18 | /bin/sh: line 1: ./perl: cannot execute binary file Nov 16 12:00:22 * koen stabs perl Nov 16 12:04:42 hi, all! Nov 16 12:05:12 ERROR: /home/slapin/openedhand/org.openembedded.dev/packages/ipkg-utils/ipkg-utils-native_1.6+cvs20050404.bb depends on itself (eventually) Nov 16 12:05:29 How to fix this error? Nov 16 12:05:42 slapin_nb: if your OE up to date and are you using bitbake 1.6.2? Nov 16 12:06:01 s/if/is/ Nov 16 12:06:31 BB_VERSION = "1.6.3" Nov 16 12:06:35 is it bad? Nov 16 12:06:47 no, that's even better as 1.6.2 :) Nov 16 12:06:51 OE_REVISION = "72e01ba791a9f5bc41e2f61a13635ad9cb1dcc87" Nov 16 12:07:16 just updated... Nov 16 12:11:02 any ideas? Nov 16 12:12:38 distro/machine and list of inherited classes Nov 16 12:13:26 MACHINE = "nokia770" Nov 16 12:13:36 DISTRO = "familiar" Nov 16 12:13:45 DISTRO_VERSION = "unstable-20061116" Nov 16 12:14:14 upwards chain is: nano (nano) -> ipkg-utils-native (ipkg-utils-native) -> file-native (file-native) Nov 16 12:14:51 both things are not the best tested things :( Nov 16 12:16:16 zecke, which are tested? Nov 16 12:16:44 I just trying to build rootfs and don't care for kernel for now. Nov 16 12:16:58 slapin_nb: 'tested' I would say openzaurus-unstable and specially angstrom-somenumber Nov 16 12:17:42 or even 'generic' Nov 16 12:18:00 yeah :) Nov 16 12:18:01 koen, DISTRO=generic? Nov 16 12:18:10 something like that Nov 16 12:18:53 * slapin_nb started build again... Nov 16 12:24:24 RP: fetchall is going nice now Nov 16 12:24:34 No, result is the same :( Nov 16 12:24:50 did you wipe TMPDIR in between? Nov 16 12:25:47 no, but it was running slowly... Nov 16 12:26:24 someone want lsof 4.77? Nov 16 12:26:33 you can't switch DISTRO without wiping tmp Nov 16 12:27:12 koen, ther's no tmp dir created... Nov 16 12:27:36 * slapin_nb started build again... Nov 16 12:27:51 (changing build dir) Nov 16 12:33:08 huhu Nov 16 12:33:11 Result is the same :( Nov 16 12:35:50 an attempt to build test package results on early problem Nov 16 12:36:07 ERROR: /home/slapin/openedhand/org.openembedded.dev/packages/ipkg-utils/ipkg-utils-native_1.6+cvs20050404.bb depends on itself (eventually) Nov 16 12:37:47 So, no solution? Nov 16 12:39:14 DISTRO=angstrom-2007.1 works for me Nov 16 12:40:07 koen, what is angstrom? Nov 16 12:40:39 a distribution that is actually maintained by people Nov 16 12:40:49 will it be rootfs-compatible (at least on core level) with n770? Nov 16 12:41:05 ? Nov 16 12:41:06 I mean just glibc+basic set of packages Nov 16 12:41:21 arm-linux omap1 Nov 16 12:41:51 it should produce a working rootfs, yes Nov 16 12:42:08 my 770 still runs IT2006, so I haven't tested it Nov 16 12:43:33 slapin_nb: what are you trying to achieve? Nov 16 12:45:10 Just have some development for OE using n770 as test device Nov 16 12:53:29 hrw|work: great :) Nov 16 12:59:21 NOTE: 4: /home/hrw/devel/oe/org.openembedded.dev/packages/unzip/unzip_552.bb, do_fetch (5777) Nov 16 12:59:28 RP: number in () is task number? Nov 16 12:59:55 DISTRO = "angstrom" Nov 16 13:00:17 same problem :( Nov 16 13:00:26 slapin_nb: angstrom-2007.1 Nov 16 13:04:16 RP: if yes, then I still se repeats Nov 16 13:05:43 RP: http://pastebin.com/825665 Nov 16 13:10:55 hrw|work, yes, it still the same Nov 16 13:12:14 hrw|work, http://pastebin.com/825669 Nov 16 13:13:00 There is all output of build process... Nov 16 13:13:23 slapin_nb: bitbake 1.6? does not looks like it... Nov 16 13:13:31 slapin_nb: did you ever 'installed' bitbake? Nov 16 13:13:56 No, never Nov 16 13:14:05 I used PATH Nov 16 13:14:35 and it prints its version btw Nov 16 13:14:50 16 14:24 < hrw|work> slapin_nb: angstrom-2007.1 Nov 16 13:17:43 http://pastebin.com/825675 Nov 16 13:17:53 here I provide my local.conf Nov 16 13:18:04 after [local.conf] Nov 16 13:19:17 hrw|work, i've split the debian packages again Nov 16 13:19:36 hrw|work, my openembedded-essential now depends on bitbake Nov 16 13:19:59 Shoragan: ok, will look later Nov 16 13:26:19 hey mickeyl Nov 16 13:26:43 hi mickeyl Crofton Nov 16 13:27:05 gm Nov 16 13:27:24 Yo all Nov 16 13:27:54 Is the only non-clamshell Zaurus that's >5500 the 6000? Nov 16 13:28:20 there's the 5600 as well Nov 16 13:28:21 hrw|work: I think the NOTE X: at the start means its aborting at that point and the task didn't complete Nov 16 13:28:55 5600 has the buggy PXA250 :/ Nov 16 13:29:04 I have a 5000D and I want to upgrade to something newer that runs Linux Nov 16 13:29:12 Speedy2: depends. some 5600 have pxa255 Nov 16 13:29:17 I'm considering Zaurus and iPaq >=3600 && <= 3800 Nov 16 13:29:36 Speedy2: ipaq h2210 Nov 16 13:29:41 Why that model? Nov 16 13:29:53 hrw|work: The number in brackets is the pid Nov 16 13:29:54 I was a fan of the StrongARM SA1110 used in the 36-38 and 5500 Nov 16 13:29:59 RP: aha Nov 16 13:30:10 Speedy2: pxa255 == less problems then sa1100 Nov 16 13:30:23 SA1110 didn't have too many problems. I think USB was the biggest Nov 16 13:30:30 Speedy2: and h2200 is small, has BT, cf+sd Nov 16 13:30:35 What's in the ipaq h2210? Nov 16 13:30:38 Speedy2: and is 2.6 supported Nov 16 13:30:40 A Samsung ARM? Nov 16 13:30:43 pxa255/400 Nov 16 13:30:47 hrm Nov 16 13:31:18 Do you think the h2210 is a better choice than a 5600 or 6000 with PXA255? Nov 16 13:32:08 h2210 is better then Zaurus 5xxx Nov 16 13:32:29 6000 is too huge to use as PDA. Nov 16 13:32:38 hrw|work: Basically the code could be more optimal buts not an actual bug Nov 16 13:32:56 hi Nov 16 13:33:22 hi mickeyl Nov 16 13:33:39 RP: ok Nov 16 13:34:18 hrw|work: 2210 is supported by Familair, with a modern 2.6 kernel you say? Nov 16 13:34:53 yes Nov 16 13:35:08 Just curious, do you use one? Nov 16 13:35:12 no Nov 16 13:35:30 I have 5500, 6000, c760, c3000 and do not use them Nov 16 13:35:41 * CoreDump|home _so_ wished collie would finally get a kernel 2.6 SD driver Nov 16 13:36:19 CoreDump|home: bribe kergoth Nov 16 13:36:29 koen: heh Nov 16 13:36:38 CoreDump|home: or do13 Nov 16 13:36:40 5x00 is still with 2.4, right? Nov 16 13:36:53 Speedy2: indeed Nov 16 13:36:54 CoreDump|home: do13 has card detect working :) Nov 16 13:36:58 Speedy2: 5600 isn't Nov 16 13:37:09 card detect is the easy bit :-( Nov 16 13:37:55 Ah, so all the SA based devices are stuck with 2.4 :( Nov 16 13:37:56 * CoreDump|home wants to move collies root natively on SD like he did w/ Poodle Nov 16 13:38:29 Speedy2: until we solve the SD issue Nov 16 13:38:47 What's the deal with SD? Sharp never gave up any English docs on it? Nov 16 13:39:00 I know it's driven through their own ASIC Nov 16 13:39:13 we have no docs on the controller Nov 16 13:40:09 bah. Nov 16 13:40:14 Is Sharp still hoarding docs? Nov 16 13:41:36 [hrw@misc1 celinux-test]$ grep 'NOTE: Task failed: Fetch failed: ftp://ftp.info-zip.org/pub/infozip/src/unzip552.tar.gz' ~/screenlog.1 |wc 27 167 2560 Nov 16 13:41:39 argh Nov 16 13:42:55 hrw|work: Do you know anyone with the h2210? I'd like to ask questions of someone using it. Nov 16 13:43:00 err h2210 and Linux Nov 16 13:43:33 hrw|work: That can't be right. Can you share the log? Nov 16 13:43:36 Speedy2: there are few people. one of them even here - but if prefer to be quiet then I will not bother him Nov 16 13:43:39 RP: sure Nov 16 13:44:09 hrw|work: Ok. Well thanks a LOT for the tip on the h2210. Nov 16 13:44:40 RP: http://ewi546.ewi.utwente.nl/tmp/hrw/logs/unzip.log.bz2 Nov 16 13:45:02 Speedy2: np Nov 16 13:49:18 hrw|work: It is working, its just inefficient Nov 16 13:49:44 hrw|work: If A fails and there are 7 more tasks running, it doesn't mark failures in the remaining tasks Nov 16 13:49:55 ah Nov 16 13:50:31 hrw|work: I can fix that in due course but it will work for now Nov 16 13:50:45 morning all Nov 16 13:51:01 hi Dirk Nov 16 13:51:09 hey Richard Nov 16 13:51:18 hey do13_ Nov 16 13:51:27 hi koen Nov 16 13:51:59 hi Dirk Nov 16 13:52:21 hi Marcin Nov 16 13:56:22 ~seen twuin Nov 16 13:56:32 ~seen twiun Nov 16 13:56:39 i haven't seen 'twuin', koen Nov 16 13:56:41 twiun was last seen on IRC in channel #oe, 298d 20h 56m 42s ago, saying: 'I thought gentoo was _exactly_ what you wanted to use on a slow machine? Compilation optimisations and all that...'. Nov 16 13:56:42 * koen gets some coffee to improve his typing Nov 16 13:57:31 hrm, on HH.org the Todo for h2210 lists "Suspend/resume and power management for HAMCOP GPIO", has that been resolved? Nov 16 13:58:04 suspend/resume seems to work fine Nov 16 13:58:47 298d... that's quite a bit Nov 16 13:59:26 almost 10 months Nov 16 13:59:56 Hey mickeyl Nov 16 14:01:31 um coffee Nov 16 14:02:11 ~lart debian badly Nov 16 14:02:12 * ibot plops debian badly into a giant vat of herring Nov 16 14:02:39 yo Speedy2 Nov 16 14:02:49 this must be the day all the veterans are showing up... Nov 16 14:02:53 :) Nov 16 14:02:57 Speedy2: how has life been to you? Nov 16 14:03:02 OK. Nov 16 14:03:28 I have been resolving software bugs in on a product I've been working on for a few months. I did a demo and all seemed well. I have been tearing my hair out over some of these bugs! Nov 16 14:03:58 Maybe I should just stick with hardware :) Nov 16 14:04:09 mickeyl: How have you been? What's new on your side? Nov 16 14:04:22 hehe Nov 16 14:04:32 I had a terrific 2006 Nov 16 14:04:40 Tell me about it! Nov 16 14:04:53 finished my ph.d thesis (pending approval), starting my freelancer career, being hired by FIC to work on OpenMoko Nov 16 14:05:06 everything really exciting Nov 16 14:05:10 (albeit hard work) Nov 16 14:05:34 Wow! Nov 16 14:05:43 What was your Thesis on Dr. Lauer? Nov 16 14:05:50 When/where did you defend? Nov 16 14:06:10 unfortunately not yet Dr. The defense is scheduled for december or january Nov 16 14:06:23 thesis was on "component-based adaptive middleware for mobile distributed systems" Nov 16 14:06:34 Ah Nov 16 14:06:43 Where are you doing your work? Nov 16 14:07:19 atm. at home in Frankfurt, Germany. Some visits to Shangai and Taiwan next year Nov 16 14:07:27 Would you move to .tw to work with FIC? Nov 16 14:07:27 That's a big change! Nov 16 14:08:17 i would never move permanently. my wife has a career here in Frankfurt and I'm not going to get involved into a remote marriage Nov 16 14:08:19 so either we both move or no one Nov 16 14:08:30 wow. Nov 16 14:08:49 morning Nov 16 14:08:51 Yo chouimat Nov 16 14:08:55 hey chouimat Nov 16 14:10:07 guys!! Nov 16 14:10:10 good news Nov 16 14:10:10 How serious is FIC about their OpenMoko thing? Hopefully they're more serious than Sharp. Nov 16 14:10:33 Speedy2: FIC is dead serious on making OpenMoko a success. They think in terms of platforms, not releases. Nov 16 14:11:02 Speedy2: it has been a really good relationsship. This is much more an Open Source project than anything else. Starting in January, the complete stack will be developed in the open Nov 16 14:11:15 * mickeyl officially announces OpenEmbedded has a stand @ FOSDEM2007 Nov 16 14:11:15 That excellent news. Nov 16 14:12:09 mickeyl: yay! Nov 16 14:12:39 * Genesis noticed :)) Nov 16 14:13:09 mickeyl: Hopefully FIC will make a PDA/Phone with a real keypad Nov 16 14:13:45 if they alter the shape a bit you could use it as a boomerang ;) Nov 16 14:13:45 mickeyl: so I officially have to check how to get to fosdem Nov 16 14:14:01 hrw|work: definitly now. i will send a mail to get the planning mill started. Nov 16 14:14:04 (tomorrow...) Nov 16 14:15:17 Speedy2: you see this first Neo1973 device is a test. if this first device sells reasonably (and I'm almost sure it will), then we will see a whole lot of devices using this platform. and this is not 2008 or later, no... even next year. Nov 16 14:15:29 Wow. Nov 16 14:15:35 mickeyl: plan=tomorrow, or plan:=tomorrow? Nov 16 14:15:53 koen: := ;) Nov 16 14:16:01 bbl, need to go to the grocery Nov 16 14:22:01 hi Nov 16 14:22:53 mickey|bbl: OpenSource or free Software? Nov 16 14:35:38 http://www.intel.com/Please-Read-The-BB-File/IPL_ixp400AccessLibrary-2_1.zip ;D I'm curious does someone in intel wonder why such URL is requested Nov 16 14:37:23 * chouimat is listening to Elected by Bruce Dickinson on Mr. Bean Soundtrack [Amarok] Nov 16 14:42:19 someone know does opie-camera is usable without sharp camera? Nov 16 14:42:36 that would be me Nov 16 14:42:38 hrw|work: last time I checked, it wasn't Nov 16 14:42:40 hrw|work: it's not Nov 16 14:42:44 i didn't write it that way Nov 16 14:43:03 mickeyl: so its now COMPATIBLE_MACHINE = "collie" :D Nov 16 14:43:12 zecke: to be honest, I still don't get the subtle differences, but it's definitly the better one of the two options ;) Nov 16 14:43:23 hrw|work: oh yeah, anything 2.4 at least Nov 16 14:44:49 mickeyl: so collie only Nov 16 14:46:59 ya Nov 16 14:47:01 i'm afraid so Nov 16 14:48:13 2.4 with binary crap module Nov 16 14:49:48 mickeyl: NOTE: Task failed: Fetch failed: http://www.vanille.de/mirror/sip-4.4.5.tar.gz Nov 16 14:52:56 mickeyl: subtle? Nov 16 14:53:11 Gerrath: are you around? :> I'm still fighging DFB.. tried going via SDL... but no real luck so far Nov 16 14:55:12 OE/packages/ixp4xx/ixp-osal_*.bb are nslu2 only or nslu2|ixp4xx? Nov 16 14:55:25 zecke: hugeß Nov 16 14:55:26 ? Nov 16 14:56:01 zecke: Have you read TODO in bitbake recently? :} Nov 16 14:56:23 * RP hadn't realised it existed... Nov 16 14:56:31 RP: do we have one? Nov 16 14:56:48 hrw|work: ixp4xx Nov 16 14:57:03 zecke: it appears we do Nov 16 14:57:06 koen: nslu2 does not use them? Nov 16 14:57:07 hrw|work: it's firmware for the microengines in the CPU Nov 16 14:57:29 I marked them as nslu2|ixp4xx now Nov 16 14:57:59 hrw|work: iirc they are being obsoleted by an open driver, but I'm not entirely sure Nov 16 14:58:09 hrw|work: but yes, the nslu2 uses them Nov 16 14:58:11 koen: opendriver still use firmware Nov 16 14:58:16 I'm about to move poky to use bitbake trunk... Nov 16 15:02:57 I vote for removing mono from OE. it was added over 2 years ago and now it is obsoleted by upstream. Nov 16 15:05:42 hrw|work: start a removal.txt Nov 16 15:05:53 hrw|work: stating reasons and a due-date Nov 16 15:06:31 mickeyl: how up to date is the scons stuff in OE? Nov 16 15:06:51 good idea Nov 16 15:08:54 i've a patch for -fb swith Nov 16 15:10:43 it's fun since i don't know how to deal with 2 touscreen , i've only one that work , so the mouse is one the first and pointer move on the second ;) Nov 16 15:22:36 CIA-3: kick Nov 16 15:23:26 let's assume I have received some distro with prebuild stuff+rootfs + kernel sources from a vendor; I want to purge everything except for the kernel and setup OE to build the kernel from the sources that I have + everything else; basically create an own rootfs with only the stuff I need. how would I do that? I read the documentatoin and also the wiki but could not figure out how to really "start from scratch" Nov 16 15:23:36 hrw|work: the cia server has been down since this morning Nov 16 15:24:09 Jin^eLD: docs: create own machine, then add kernel recipe Nov 16 15:24:21 Jin^eLD: create a machine config for it, create a recipe for the kernel, and pick a distro (generic-unstable would be a good start) Nov 16 15:24:48 I did the machine thing already (right now I have a setup for the prebuilt toolchain) Nov 16 15:24:50 ok thanks Nov 16 15:25:33 packages/linux - those are the kernel recipes, right? Nov 16 15:26:01 let's see :> Nov 16 15:26:42 yes Jin^eLD Nov 16 15:27:05 Jin^eLD, : you should have a bootloader too Nov 16 15:27:16 Genesis: I do have one from the vendor Nov 16 15:27:36 you could want to recompile it with your new toolchain Nov 16 15:27:44 oki Nov 16 15:27:47 do I need to? Nov 16 15:27:56 actually they use redboot Nov 16 15:27:58 no , sometime it's preffered Nov 16 15:28:07 i use redboot too Nov 16 15:28:21 it's a shit. Nov 16 15:28:26 sometimes - like when? sorry if the question is stupid but I am quite new to the embedded linux stuff Nov 16 15:28:53 sorry don't remember well Nov 16 15:28:56 What's wrong with redboot? Nov 16 15:29:30 Speedy2, : no kernel cmdline parameter , no gpl , no maintener , no mmc boot , no usb boot Nov 16 15:29:46 usb boot... Nov 16 15:29:51 no nothing, only kernel-from-flash booting? Nov 16 15:29:51 it's 4x biggest that uboot Nov 16 15:29:52 I was interested in that btw Nov 16 15:30:33 hrw|work: tftp booting with nfs rootfs works afaik Nov 16 15:31:08 so which one you suggest? I'd like to have the usb boot option... Nov 16 15:31:43 depend your board Nov 16 15:32:04 well... some ARM9 with a little ram and a little flash Nov 16 15:32:10 +usb Nov 16 15:32:15 look at uboot Nov 16 15:32:15 and ethernet Nov 16 15:33:03 ok, I will take a look.. thanks Nov 16 15:35:18 I am in a talk by some company taling about how they use tinderbox Nov 16 15:35:44 apparently they have an irc bot that watches the ouput and messgaes when there is an issue Nov 16 15:37:05 now there is a slide of an irc conversation Nov 16 15:47:51 ~logs Nov 16 15:47:52 apt/ibot/jbot/purl all log to http://ibot.rikers.org// where channelname is html encoded ie: %23debian | lines that start with a space are not shown | some channels have stats at http://ibot.rikers.org/stats/.html.gz, or updated "nightly" Nov 16 15:52:52 hrw|work: I disabled cia xmlrpc messaged so the commit mails can go through Nov 16 15:53:09 right now the script is hanging on contacting cia.navi.cx Nov 16 15:55:30 ok Nov 16 15:55:58 might make that a background thread in ciabot Nov 16 15:56:07 should be simple Nov 16 16:00:33 | /a/home/hrw/devel/build/celinux-test/tmp/work/keymaps-1.0-r11/temp/run.do_fetchall.7512: line 544: do_fetchall: command not found Nov 16 16:00:56 RP: fetchall does not work with SRC_URI = file:// recipes Nov 16 16:10:17 mickeyl: my idea would be to have 2 instances of ciabot running Nov 16 16:11:15 one for mailing, one for rpcing? Nov 16 16:11:23 mickeyl: I'll try getting the usbvga thingy working on the loft next week Nov 16 16:11:43 koen: that'd rock Nov 16 16:13:27 hrw|work: ok, we probably need a null fetchall method in base.bbclass Nov 16 16:13:56 Why only file:// urls though? :-/ Nov 16 16:14:08 RP: keymaps.bb Nov 16 16:19:25 hrw|work: fix pushed Nov 16 16:22:20 Poznan, Poland suxx when it comes to flights. Nov 16 16:22:42 I have to decide: Poznan -> Berlin -> fosdem -> Berlin -> Poznan or s/Berlin/Warsaw Nov 16 16:23:48 zecke: how you travel to fosdem? Nov 16 16:24:05 hrw|work: by a unspecified vehicle Nov 16 16:24:11 * IvoC lives 15km from fosdem :) Nov 16 16:24:31 hrw|work: probably I will fly to brussel but I won't take the same cab again Nov 16 16:25:10 zecke: take train to Bxl center (every 10minutes) and take subway or cab there Nov 16 16:25:11 zecke: I meant how 'berlin -> fosdem' ;D Nov 16 16:25:46 IvoC: well, not if you have a queue of 30 asian travelers in front of you Nov 16 16:25:56 IvoC: and the last train departs in 5 minutes :) Nov 16 16:28:46 ok, then you're screwed :-P Nov 16 16:31:00 * koen will use the train Nov 16 16:31:10 * mickeyl will use the plane Nov 16 16:31:19 50 minutes from FRA Nov 16 16:34:27 hmm early booking rocks Nov 16 16:34:46 early booking and last-minute booking Nov 16 16:34:52 mickeyl: using lufthansa? Nov 16 16:34:54 in between you are screwed Nov 16 16:34:58 yeah Nov 16 16:38:21 14:35 - 22:05 in one way, 21:00 -> who know when in return way Nov 16 16:38:36 ~lart Poznan for lack of flights to normal places Nov 16 16:38:36 * ibot duct-tapes Poznan to the floor and drools on him for lack of flights to normal places Nov 16 16:38:46 are we clustering in the same hotel as last year, or are going for the o-hand hotel? Nov 16 16:40:55 i like the Astrid Nov 16 16:41:00 where did the ohand guys hang out? Nov 16 16:41:25 * koen tries to remember Nov 16 16:42:02 novotel? Nov 16 16:42:07 i remember liam was there Nov 16 16:42:14 Yes, Novatel Nov 16 16:42:30 can we convince you guys to book the Astrid next year? :) Nov 16 16:42:37 wtf? Nov 16 16:42:39 http://openembedded.objectis.net/ Nov 16 16:42:42 i'll talk w/ mallum Nov 16 16:42:50 mickeyl: mallum could be open to persuasion :) Nov 16 16:42:56 * raduga curious Nov 16 16:43:01 is it very cold there, right now? Nov 16 16:43:05 koen: eeks? Nov 16 16:43:28 raduga: brussels is always frickin' cold. especially in february Nov 16 16:43:39 * raduga checks the calendar Nov 16 16:43:52 * mickeyl had a dumb looking woolcap on, but was warm Nov 16 16:43:56 * RP makes note to remind mallum about company fleeces, not tshirts Nov 16 16:44:04 hehe Nov 16 16:44:05 yeah! Nov 16 16:44:11 fleeces rock Nov 16 16:44:11 mickeyl: How much - is cold for guys? :) Nov 16 16:44:23 mickeyl: warm and dumb is better than smart and cold Nov 16 16:44:28 den-ros: I'm getting uncomfortable below 6 celcious Nov 16 16:44:35 raduga: :) Nov 16 16:44:36 mickeyl: FWIW, brussels is warmer than here ;-) Nov 16 16:45:32 mickeyl: More hard when high water concetration in air even in not very low temperature. Nov 16 16:45:55 mickeyl: define cold Nov 16 16:46:00 -20°C? Nov 16 16:46:35 I was one time in far Siberia :) ... more then -30 -35... but i live yet :) Nov 16 16:46:46 It also depends a lot on wind and mosture. I was fine in Finland in -20 yet freeze here at -5... Nov 16 16:47:07 RP: Agree. Nov 16 16:51:09 RP: today is mallum's birthday, right? Nov 16 16:51:28 yeah Nov 16 16:53:48 iirc he turns 33 Nov 16 16:53:56 right Nov 16 16:54:01 * mickeyl hits the "send mail" button Nov 16 16:54:14 I'm seriously considering spending a third of the year somewhere else Nov 16 16:54:22 I don't like winter Nov 16 16:54:39 bora bora is nice Nov 16 16:55:17 mickeyl: i won't say SD is nice, but at least its not cold Nov 16 16:55:19 * den-ros googling bora bora Nov 16 16:55:32 hmm, my b-day today, too. wish I were still 33 ;-) Nov 16 16:55:44 tnb: congrats Nov 16 16:55:49 congrats tnb Nov 16 16:55:54 tnb: Grngs! (*) Nov 16 16:56:01 happy 34th :) Nov 16 16:56:09 thanks guys Nov 16 16:56:16 congrats... Nov 16 16:56:16 raduga: funny guy Nov 16 16:56:21 41 Nov 16 16:56:27 close... Nov 16 16:56:29 ehhh... Nov 16 16:56:57 * den-ros whant to turn time back Nov 16 16:57:39 cu Nov 16 16:58:21 den-ros: how far back? Nov 16 16:58:28 you miss being a communist? :) Nov 16 16:58:35 raduga: at 80's Nov 16 16:58:55 raduga: afraid it no no no ... :( Nov 16 16:58:57 :) Nov 16 16:59:35 * raduga threatens den-ros, and his entire evil empire with nuclear annihilation Nov 16 16:59:36 80's is stylish times... Nov 16 16:59:52 raduga: :) Nov 16 16:59:55 the 80s were... interesting. Nov 16 16:59:59 a bit strange :) Nov 16 17:00:36 * mickeyl adores 80s music Nov 16 17:00:57 girls, disco and car shapes of 80's never will be repeated :) Nov 16 17:01:00 modernism was beginning to break down, but hadn't completely crumbled Nov 16 17:01:21 the cracks in the edifice, were fun to poke at. Nov 16 17:01:59 * JustinP thinks of The Terminator Nov 16 17:02:01 kicking loose debris in the rubble that remains from where modernism fell, is rather less fun Nov 16 17:02:13 den-ros: so we hope. Nov 16 17:02:31 * raduga didn't like girls so much, in the 80s. Nov 16 17:02:36 mostly they just annoyed me :( Nov 16 17:02:38 * chouimat thinkings the '80s shouldn't have happened in music ... ok maybe except for some bands like Iron Maiden etc ... but not for the pop crap ... wait pop crap is crap anyday anyway :) Nov 16 17:02:49 heh Nov 16 17:02:54 chouimat: hehe Nov 16 17:03:22 mickeyl: :p Nov 16 17:03:22 :) Nov 16 17:03:29 one oddity though. Nov 16 17:03:53 not only has Madonna survived the 80s, but has thrived :/ Nov 16 17:04:05 damn, I didn't even get to see the complete eighties Nov 16 17:04:14 and is still finding new ways to generate hideous offspring Nov 16 17:04:17 raduga: it's because she slept with a lot of producers :) Nov 16 17:05:08 koen: :-) Nov 16 17:05:11 we must reflash ourbrains into some AI prototipe, before death come... so we have only 30 - 40 years only :) mickeyl can You start new open project for this please? :) Nov 16 17:05:22 i wonder if she's among the progenitors of the Atreides line Nov 16 17:05:35 LOL Nov 16 17:05:43 OpenBrains Nov 16 17:05:53 :)) Nov 16 17:06:31 i'm going to flash my brain into ibot Nov 16 17:06:36 ~brain Nov 16 17:06:42 somebody said brain was a wonderful organ; it starts working the moment you get up Nov 16 17:06:45 brb I just a fairly clear message (aka claw in my foot) that it's time to feed the cat Nov 16 17:07:01 ibot: flash me Nov 16 17:07:04 * ibot starts a firmware update on raduga, then pulls the plug halfway through Nov 16 17:07:25 mickeyl: http://en.wikipedia.org/wiki/Trepanation Nov 16 17:07:25 s/organ/orgazm Nov 16 17:07:50 wow Nov 16 17:07:50 chouimat: that's subtle Nov 16 17:08:19 I used to think the brain is the most wonderfull organ in my body, till I realised what was telling me that Nov 16 17:08:19 my cat tends to jump on top the monitor, and dangle his legs in front of the screen Nov 16 17:09:48 raduga: mine too but that one was more like "Hey!!! I'm starving you big Ape!!! Feed me !!! " Nov 16 17:11:05 chouimat: when my cat wants to be subtle, he'll dig his claws into the carpet, hard enough that a claw-sheath falls off Nov 16 17:11:16 and I'll be walking by, minding my own business, then suddenly Nov 16 17:11:19 *clawed in the foot* Nov 16 17:11:27 and my cat is sleeping, undisturbed in the next room Nov 16 17:11:54 hehe Nov 16 17:13:22 chouimat: this site needs more zaurii Nov 16 17:13:46 http://www.infinitecat.com/infinite/cat-html/1-100/18.html Nov 16 17:14:47 but, aside from the stylus, he pays *no* attention whatsoever, to the Z in any event Nov 16 17:16:09 raduga: hehe Nov 16 17:27:43 * mrz80 is away: Metaphysically I'm still here; physically, I'm elsewhere Nov 16 17:53:15 'If it wasn't for C, we would be using BASI, PASAL and OBOL' Nov 16 17:53:33 127.0.0.1, sweet 127.0.0.1 Nov 16 17:53:46 help Nov 16 17:54:11 * cbrake should have typed /help :-\ Nov 16 17:55:15 back to taglines. Nov 16 17:55:22 hrw: i'd still be using perl Nov 16 17:55:35 raduga: Perl - The only language that looks the same before and after RSA encryption. Nov 16 17:55:50 hrw: hee Nov 16 17:57:32 hrw: *gg* Nov 16 17:59:39 Programming consists of 50% planning, 50% coding and 30% mathematics Nov 16 17:59:59 and 200% debugging Nov 16 18:00:20 that one is nice: If GOD had intended computers to have 1,024 bytes per kilobyte, HE/SHE would have given us SIXTEEN FINGERS!! Nov 16 18:00:28 lol Nov 16 18:01:00 actually, 2^10 = 1024, so maybe he did :-) Nov 16 18:02:30 I tried to count people using the binary finger method once Nov 16 18:03:00 after 7 people I had cramped fingers and a cramped brain Nov 16 18:03:01 god gave us 8 fingers, so we should be counting in octal Nov 16 18:03:20 but silly ppl keep sticking their thumbs in Nov 16 18:03:30 488M /home/hrw/devel/sources/svn_dir Nov 16 18:03:30 1.4G /home/hrw/devel/sources/cvs_dir Nov 16 18:03:30 2.3G /home/hrw/devel/sources/dl_dir Nov 16 18:03:30 69M /home/hrw/devel/sources/git_dir Nov 16 18:03:30 4.1G /home/hrw/devel/sources/ Nov 16 18:03:34 progressing Nov 16 18:03:59 * raduga thinks "thumbs are the work of the Devil" Nov 16 18:04:09 who will write bittorent fetcher? :D Nov 16 18:10:16 home time, cya Nov 16 18:25:32 hrw, i've always been a fan of fred brooks "mythical man month" dictum: take programming time. multiply by 4 to get to actual product. Nov 16 18:47:10 RFC: ${KERNELORG_MIRROR} Nov 16 18:47:51 RFC: ${KERNELORG_MIRROR} instead of http://kernel.org http://www.kernel.org http://ftp.kernel.org ftp://ftp.kernel.org ftp://kernel.org ftp://www.kernel.org Nov 16 18:49:59 hrw: sounds like a good idea Nov 16 18:53:50 thx Nov 16 19:01:11 * koen adds 'scons' and 'blender' to the buildsystem shitlist Nov 16 19:01:26 hm scons suckz? Nov 16 19:02:10 anything that requires python syntax and indenting in config files sucks Nov 16 19:02:35 hm Nov 16 19:02:39 thats make too Nov 16 19:02:53 woglinde: have a look at http://ewi546.ewi.utwente.nl/tmp/blender_2.42a.bb Nov 16 19:03:41 wourgh... Nov 16 19:03:42 hm but whats blender good for on a zaurus? Nov 16 19:03:52 it needs even more cp statements, so I gave up on it Nov 16 19:03:56 jesus Nov 16 19:03:58 woglinde: is the world a zaurus? Nov 16 19:04:03 what the hell Nov 16 19:04:30 Bye. Nov 16 19:04:42 I also run OE on my smp workstation :) Nov 16 19:05:26 okay okay Nov 16 19:06:20 hey likewise Nov 16 19:06:26 koen: hey there Nov 16 19:06:28 morning all Nov 16 19:06:36 hi likewise Nov 16 19:06:37 likewise: afternoo Nov 16 19:06:38 n Nov 16 19:06:54 so, let's look for a FOSDEM hotel :-) Nov 16 19:07:11 likewise: the astrid or the novotel Nov 16 19:07:37 likewise: it looks like the astrid will get the preference for the OE dudes Nov 16 19:07:42 koen: thanks, I will try the novotel then :-) Nov 16 19:15:51 doh.. creating stuff from scratch is really difficult Nov 16 19:17:44 Jin^eLD: yeah, I just made dinner from scratch. Nov 16 19:18:20 likewise: well.. I'm trying to create a distro + rootfs Nov 16 19:18:44 we had a talk yesterday, about dumping everything that comes from the vendor.. Nov 16 19:18:48 Jin^eLD: try using an existing distro first :) Nov 16 19:18:49 yup Nov 16 19:19:38 koen: I am starting up with generic.conf Nov 16 19:20:16 but I do not think that I will manage to setup the kernel compilatoin Nov 16 19:20:32 so I will use the prebuilt kernel for the timebeing Nov 16 19:23:49 but setting up the distro seems like a nightmare :) Nov 16 19:24:03 it was much easier with the prebuilt toolchain - there I just had to follow the howto ;) Nov 16 19:24:44 Jin^eLD: now howto is: MACHINE="yourmachine" DISTRO="generic" bitbake bootstrap-image Nov 16 19:24:47 ;D Nov 16 19:25:39 hrw: but that would build the kernel? Nov 16 19:26:56 if you added it into OE.. Nov 16 19:27:20 I did not.. but the generic.conf has some setions Nov 16 19:27:21 KERNEL = "kernel26" Nov 16 19:27:21 MACHINE_KERNEL_VERSION = "2.6" Nov 16 19:27:23 like that Nov 16 19:27:27 or is it just information about the targt? Nov 16 19:29:00 about target Nov 16 19:29:27 some targets in OE can run 2.4 or 2.6 - MACHINE_KERNEL_VERSION switch it Nov 16 19:29:32 hmm ok then, I am following your howto :> Nov 16 19:29:43 KERNEL = "kernel26" is old, obsolete way Nov 16 19:29:49 but wait, I need to tell what packages I need, right? Nov 16 19:30:05 no, you need to tell what features you need Nov 16 19:30:32 is that also in the distro.conf or somewhere else? Nov 16 19:30:39 * mrz80 is back (gone 02:02:57) Nov 16 19:31:33 distro and machine Nov 16 19:33:24 what should go to which of them? Nov 16 19:33:31 my machine stuff is very simple Nov 16 19:33:41 http://pastebin.ca/250250 Nov 16 19:34:25 Jin^eLD: remove all except TARGET_ARCH IPKG_EXTRA_ARCHS Nov 16 19:34:38 Jin^eLD: use proper conf/machine/include/tune-* includes Nov 16 19:35:11 there was none that did fit, that's why I set the TARGET_CC_ARCH there... Nov 16 19:35:29 Jin^eLD: ah.. gcc3<>gcc4 problem... Nov 16 19:35:38 hrw: yes, also - I used a prebuilt toolchain before Nov 16 19:35:45 and they had gcc3 Nov 16 19:35:49 something like http://pastebin.ca/250251 ? Nov 16 19:36:39 http://pastebin.ca/250252 rather Nov 16 19:37:29 I thought I was supposed to remove everything except ARGET_ARCH IPKG_EXTRA_ARCHS ? Nov 16 19:37:45 Jin^eLD: MACHINE_FEATURES is needed Nov 16 19:38:13 hrw: well ok I will go with what you posted http://pastebin.ca/250252 Nov 16 19:38:14 Jin^eLD: as it say what machine feature (2.4/2.6 kernel, usb host/gadget, pcmcia/cf etc) Nov 16 19:38:27 ah.. so Nov 16 19:38:38 I have to add some stuff then Nov 16 19:38:50 so those "features" are - hardware that the machine has? Nov 16 19:38:54 yes Nov 16 19:39:04 is there a predefined list from which I can choose? Nov 16 19:39:14 look at other confs and task-base.bb Nov 16 19:40:04 ok I see.. let me add a couple of things.. Nov 16 19:41:19 hrw oh nice wlan-cf-card Nov 16 19:43:04 ok.. so the rest is just taking the generic.conf as it is? Nov 16 19:43:23 or do I have to tweak the distro ? Nov 16 19:43:29 you can Nov 16 19:44:54 what's this INHERIT += "debian multimachine" Nov 16 19:45:23 Jin^eLD: look at celinux-test distro Nov 16 19:46:02 nice, commented! Nov 16 19:46:11 I wrote it ;D Nov 16 19:46:24 aah :> Nov 16 19:48:46 well maybe I shoult start off with your distro config file and not with the generic one, yours seems kind of cleaner Nov 16 19:49:15 I think that generic need to get some comments from celinux-test rather Nov 16 19:49:52 what's this task-bootstrap thing in generic? Nov 16 19:49:54 or from angstrom :) Nov 16 19:50:09 Jin^eLD: ignore task-bootstrap, that's obsolete now Nov 16 19:50:13 * likewise watches the Battle of Distros Nov 16 19:50:24 ok.. Nov 16 19:50:53 ok let me explain what I intend to do :) maybe you can give me a better hint then... Nov 16 19:51:10 Jin^eLD: what machine you are adding? Nov 16 19:51:32 I want to create a minimalistic distro with only a few packages; it will be a NAS with a UPnP server on it + samba, that's it Nov 16 19:51:36 hrw: ARM9 based Nov 16 19:51:50 hrw: you just helped me writing the machine.conf for it :> Nov 16 19:52:05 Jin^eLD: is it on market already? Nov 16 19:52:09 hrw: no Nov 16 19:52:25 hrw: and yes :) well this particular one is not Nov 16 19:52:32 but it is based on a reference design which is out already Nov 16 19:52:41 but there is no linux for that yet Nov 16 19:52:50 arm926 is a funny choice for a NAS Nov 16 19:52:51 * chouimat is enjoying the "Mp3 Retagging Festival" Nov 16 19:53:05 koen cool be fun to hack anyway ;) Nov 16 19:53:15 but arm926 is also used in the neuros :) Nov 16 19:53:22 I received the first linux distro from the vendor, so I took a board from a product that we were developing and want to hack around in my spare time Nov 16 19:54:04 koen: well, it was never planned to become a NAS ;) the original reference design can do more, like mp3, pictures, LCD support, etc. Nov 16 19:54:24 Jin^eLD: I see ... Nov 16 19:54:37 I only recently learned about the RDC chips used in the thing is - I am writing a UPnP server, opensource, we got it running on the NSLU2 - some users helped us; so I got interested in making something for myself Nov 16 19:55:37 our company develops a product wich is not related to NAS at all, but the board is very small and has ethernet and usb Nov 16 19:55:49 and that's exactly what I need :) and - the vendor delivered linux recently Nov 16 19:56:07 Jin^eLD: 64M ram or more? less? Nov 16 19:56:10 8mb :) Nov 16 19:56:27 even my WAP/router has more... Nov 16 19:56:34 Jin^eLD: something like http://rafb.net/paste/results/mIayMj94.html ? Nov 16 19:56:38 the thing is - I actually did cross compile and launch my server on it already, and I was surprised it did run with such little ram Nov 16 19:57:02 because we originally wrote the server for a PC enviroenment, all in C++ with reference counting and greedy on memory Nov 16 19:57:52 ehm, like http://rafb.net/paste/results/hYzozk25.html Nov 16 19:58:16 doubled inherits Nov 16 19:58:30 X11 not really needed Nov 16 19:58:32 what is QPE? Nov 16 19:58:40 I can safely remove that I guess, right? Nov 16 19:58:46 yeah Nov 16 19:58:47 Jin^eLD: qpe is some obsolete pda soft Nov 16 19:59:14 * chouimat is waiting for qtopia 4.2 gpl ... Nov 16 19:59:15 extra RDEPENDS: that's where I could add my upnp server and all libraries that I am using? like libexif, sqlite3, etc.? Nov 16 19:59:19 ever heard about Obsolete Integrated Palmtop Environment? Nov 16 19:59:28 nope, never had a palm Nov 16 19:59:41 hrw: hehe Nov 16 19:59:41 Jin^eLD: does you server link against those libs? Nov 16 19:59:51 yes Nov 16 20:00:02 Jin^eLD: then install only server ;D Nov 16 20:00:08 Jin^eLD: ok, so you only have to add your server, OE will figure out the libs Nov 16 20:00:28 koen: but my .bb file has to tell what libs it needs, no? Nov 16 20:00:39 Jin^eLD: has to - in DEPENDS Nov 16 20:00:40 Jin^eLD: yes Nov 16 20:01:12 DISTRO_EXTRA_RDEPENDS = "samba leet-upnp-server" Nov 16 20:01:34 but when I do DEPENDS += "sqlite3", but I actually do depend on libsqlite3 - then ipkg complains about a missing sqlite3 dependancy Nov 16 20:01:51 but I can not do DEPENDS += "libsqlite3" because there is no such package (yet - the bbfile is sqlite3) Nov 16 20:01:56 how to cope with that? Nov 16 20:02:26 bbl - need to work some on mails Nov 16 20:03:18 Jin^eLD: OE takes care of the 'sqlite3' -> 'libsqlite3' renaming Nov 16 20:03:26 well, it should :) Nov 16 20:03:38 hmm :) well maybe I messed something up.. heh Nov 16 20:03:39 time to watch some tv Nov 16 20:03:47 thank you all for your help :) Nov 16 20:08:53 libraries are automatically included (they are build-time and run-time dependencies). However, for your own upnp server, you would need to make a bitbake recipe first. Nov 16 20:09:00 Jin^eLD: libraries are automatically included (they are build-time and run-time dependencies). However, for your own upnp server, you would need to make a bitbake recipe first. Nov 16 20:09:39 likewise: I already created the recipe for the server Nov 16 20:10:11 if all works well I will try submitting it after the release of the upcoming version Nov 16 20:10:12 Jin^eLD: bitbake will then inspect which libraries it needs by itself. Nov 16 20:10:56 Jin^eLD: I *guess* it does this by using objdump and grepping for NEEDED entries. Nov 16 20:11:07 I see.. Nov 16 20:11:48 Jin^eLD: when you have a working machine and distro, you might want to see this work with existing packages. It takes a while to learn how the system solves thing for you. Nov 16 20:11:57 s/thing/things/ Nov 16 20:12:13 I am still learning myself in that area. Nov 16 20:12:23 likewise: well, the existing packages seemed to work well so far; until now I used the vendor supplied rootfs,kernel and prebuilt tools Nov 16 20:12:34 and I already have the server with all needed libraries running on the target Nov 16 20:12:45 but I do not want their rootfs and their tools Nov 16 20:12:46 :) Nov 16 20:13:11 especially when I have a special purpose in mind.. so that's why the move to an own distro now Nov 16 20:13:37 it use 'objdump -p' and NEEDED/SONAME Nov 16 20:14:01 package_do_shlibs() in package.bbclass Nov 16 20:14:41 hrw|busy: tnx Nov 16 20:18:13 maybe a stupid qustion.. but how to I build my distro? :) Nov 16 20:18:19 I mean - the whole distro so to speak Nov 16 20:18:33 bitbake world? Nov 16 20:18:38 yes Nov 16 20:19:01 ok then.. let's see how it goes :) Nov 16 20:21:11 build bootstrap-image first ;) Nov 16 20:22:13 Im gone for a couple of days, and somehow everything changes Nov 16 20:22:19 I must start reading those OE mails Nov 16 20:22:40 Kristoffer: so do I Nov 16 20:23:11 Im not even sure anything builds any longer. :D Nov 16 20:23:46 koen angstrom can build on Sa1110 ... or it have some trouble for eabi ? Nov 16 20:24:07 s/on Sa1100/for SA1110/ Nov 16 20:24:27 eabi doesn't support armv4l yet Nov 16 20:24:41 see http://wiki.debian.org/ArmEabiPort Nov 16 20:25:09 thanks ... are people already (activelly) work on that ? Nov 16 20:26:15 the solutions is known, it just needs someone to implement it Nov 16 20:27:53 I think debian is to mangle package names for shared libraries, in the debian style (not sure about all details here). Nov 16 20:28:18 likewise: yes, it enables you to install different versions of libs in parallel Nov 16 20:29:14 gremlin[it], got a jornada 7xx? Nov 16 20:29:27 koen ... touch screen now work on ipaq h3600 :) ... so i'm trying to build a gpe-image :) ... i'll like take care of eabi ... just have to look what have highest priority ... Nov 16 20:29:51 Kristoffer ho i have an ipaq h3600 and a h3800 Nov 16 20:30:01 I cant make my build stop trying to build glibc 2.5....arghhh Nov 16 20:30:25 PREFERRED_PROVIDER_virtual/arm-linux-libc-for-gcc = "glibc-intermediate" Nov 16 20:30:28 should do the trick Nov 16 20:30:30 but it doesnt Nov 16 20:30:54 Kristoffer: that only tells it to use glibc-intermediate if you use nptl Nov 16 20:30:59 nothing about versions Nov 16 20:32:24 koen, explain that again Nov 16 20:33:07 really all I should need is PREFERRED_VERSION_glibc-intermediate = "2.4" Nov 16 20:34:21 Koen: Can you close this bug http://bugs.openembedded.org/show_bug.cgi?id=1371 Nov 16 20:34:40 Koen: it has to do with people using an old version of dbus Nov 16 20:34:53 gerwinin: feel free to close it yourself :) Nov 16 20:34:58 gerwinin what are the solution ? Nov 16 20:36:19 koen, how can I tell it to use glibc-intermediate = 2.4 for all possible events? Nov 16 20:36:46 Kristoffer: glibc 2.4 has a COMPATIBLE_ARCH extry Nov 16 20:38:37 Koen : done Nov 16 20:40:12 Koen: I made a bb file for the via unichrome drivers so in bugzilla attach bb file ? Nov 16 20:40:30 yes Nov 16 20:40:45 gerwinin: as hrw said :) Nov 16 20:41:11 gerwinin : the solution to #1371 is use a newest version of dbus ? (0.92 for example) ... i have #1371 with dbus-glib-0.72 ... Nov 16 20:41:12 ~time OR, USA Nov 16 20:41:28 koen, oki thx Nov 16 20:41:48 someone could teach ibot about giving time for countries/states Nov 16 20:42:08 gremlin[it]: for x86 it is Nov 16 20:42:20 mh no i'm compiling for arm ... Nov 16 20:42:37 koen, doh, why is arm set as DEFAULT_PREFERENCE_arm = "-1" ?.. Nov 16 20:42:55 gremlin[it]: you can also adapt the configure.ac but I think for most people that is a bit too difficult Nov 16 20:43:01 Kristoffer: because it's broken for arm Nov 16 20:43:22 koen, severly broken?.. Cause i've used it on my jornada for quite awhile now Nov 16 20:43:45 gerwinin : but why after 0.72 there is only dbus and no more dbus-glib are merged ? Nov 16 20:44:32 gremlin[it]: because dbus-glib is perfect? :D Nov 16 20:45:38 gremlin[it]: I don't know Nov 16 20:46:13 ah .. so dbus and dbus-glib don't need to be of teh same version :) ... i understand :) Nov 16 20:47:02 to have the same version ... damn my english is horrible !!! :( Nov 16 20:49:50 what you people want to get documented? Nov 16 20:50:49 hrw|busy: who are you talking to? Nov 16 20:50:58 #oe Nov 16 20:51:23 hrw|busy: I have been thinking about documentation: I think what really helps is diagrams instead of textual information. Nov 16 20:51:37 hrw|busy: the stubs in the usermanual :) Nov 16 20:51:54 hrw|busy: well also the stuff I was asking today.. settings in the distro, stubs in the usermanual - I was searching for some explanations there and hit the stubs :) Nov 16 20:51:59 oh koen just mentioned it :) Nov 16 20:52:03 :D Nov 16 20:52:08 koen: currently I think about documenting adding packages and machine integration Nov 16 20:52:31 yeah, machine integration would also be cool :) Nov 16 20:52:47 likewise: thats the problem - as a programmer I do not like to do diagrams on other media then sheets of paper Nov 16 20:52:50 there already is a chapter on how to write a recipe, so you could expand that Nov 16 20:53:02 hrw|busy: I know :-) Nov 16 20:53:12 Adding a new Package Nov 16 20:53:12 This section is a stub, help us by expanding it Nov 16 20:53:17 koen: this one? Nov 16 20:53:31 hrw|busy: I do like creating diagrams, but only in Adobe Illustrator or such Nov 16 20:53:49 hrw, koen: uhm.. did we miss something in my distro config? it is trying to download and compile the kernel Nov 16 20:53:51 hrw|busy: in januari a whole batch of people that are new to embedded stuff will start using OE, you could add docs for them Nov 16 20:54:04 koen: whom? OpenMoko? Nov 16 20:54:13 likewise: yes Nov 16 20:54:40 hrw|busy: I would like to help create some diagrams. We have to agree on the format. I would suggest SVG source formats, PNG output? Nov 16 20:55:16 likewise: that should work Nov 16 20:55:26 likewise: svg's can be diffed :) Nov 16 20:55:28 hrw|busy: problem is, you probably want to create diagrams programmatically at one point. But that can be done later. Nov 16 20:55:48 I do not like the output of 'dot' for example... Nov 16 20:56:13 svg2png and svg2pdf should exist somewhere Nov 16 20:56:25 yeah they do Nov 16 20:57:13 iirc you can use inkscape from the cmdline as well Nov 16 20:58:25 and the rsvg-* tools Nov 16 20:58:31 from librsvg Nov 16 20:59:00 ah, right Nov 16 20:59:09 * v8jlene is back Nov 16 20:59:10 rsvg-convert defaults to PNG (I just saw from its help) Nov 16 20:59:21 wb Nov 16 20:59:31 ibot: EABI is the new "Embedded" ABI by ARM ltd which allow mixing soft and hardfloat code. Not yet supported for StrongARM. Palmtop distro with EABI: Ã…ngström. More info: http://wiki.debian.org/ArmEabiPort Nov 16 20:59:32 i'm not going to learn illegal characters, hrw|busy Nov 16 20:59:41 koen: nasty, isn;t it? Nov 16 20:59:51 ibot: EABI is the new "Embedded" ABI by ARM ltd which allow mixing soft and hardfloat code. Not yet supported for StrongARM. Palmtop distro with EABI: Angstrom. More info: http://wiki.debian.org/ArmEabiPort Nov 16 20:59:53 hrw|busy: okay Nov 16 21:00:52 ibot: no, EABI is the new "Embedded" ABI by ARM ltd which allow mixing soft and hardfloat code. Not yet supported for armv4l cpu's like StrongARM. Embedded distro with EABI: Angstrom. More info: http://wiki.debian.org/ArmEabiPort and http://www.angstrom-distribution.org Nov 16 21:00:54 koen: okay Nov 16 21:00:54 I have to remember to print OE usermanual tomorrow Nov 16 21:01:02 palmtop, schmalmtop Nov 16 21:01:08 ;D Nov 16 21:01:28 I have most of my bitbake recipes chapter completed for the manual... just got to find some time to finish it up, maybe next week... Nov 16 21:01:48 hrw|busy , koen : as a temporary solution have sense to compile Angstrom with "--with-arch=armv4 --with-no-thumb-interwork" options ? Nov 16 21:01:51 v8jlene: can you send me draft? so we will not duplicate Nov 16 21:02:06 gremlin[it]: koen is Ã…ngström maintainer Nov 16 21:02:24 gremlin[it]: no, gcc will use BX isns anyway Nov 16 21:02:38 btw - due to Ã…ngström I learnt how to enter those funny chars in X11 Nov 16 21:02:41 gremlin[it]: change conf/distro/include/angstrom.inc Nov 16 21:02:47 but still no Euro sign Nov 16 21:03:17 has anyone used an SD card with the PXA270? Nov 16 21:03:29 conf/distro/include/angstrom-glibc.inc, that is Nov 16 21:03:39 Gerrath: few times Nov 16 21:03:48 change TARGET_OS = "linux${@['','-gnueabi'][bb.data.getVar('TARGET_ARCH',d,1)=='arm']}" to TARGET_OS = "linux" Nov 16 21:04:04 Gerrath: yes, but with an external controller (asic3) Nov 16 21:04:18 ok ok koen ... i don't do now :) ... just to figure out :) Nov 16 21:04:26 hrw|busy, is there any patches I need to apply to the 2.6 kernel, or is the PXA MMC controller already supported? Nov 16 21:05:08 hrw|busy, I was looking through the Kconfig and I couldn't find any options specific to the PXA and MMC. Nov 16 21:05:37 already Nov 16 21:07:04 Gerrath: I'll write patch for kernel then.. Nov 16 21:07:19 hrw|busy, uhm, well I just grep'd the kernel for MMC and PXA and it looks like it is there.. just need to dig in the source and find out the correct Kconfig settings :-) Nov 16 21:07:27 hrw|busy, thanks. Nov 16 21:07:41 Gerrath: MMC, MMC_PXA Nov 16 21:07:59 PXA_MCI ? Nov 16 21:08:46 hrw|busy, thanks. Nov 16 21:08:46 pxamci.c even Nov 16 21:08:46 v8jlene, need an editor/test reader? Nov 16 21:08:46 Gerrath: http://pastebin.com/826070 Nov 16 21:09:26 HopsNBarley: Everything I've done in the user manual so far (reference section) could do with an editor/reader ;) Nov 16 21:09:50 hrw|busy, thanks again :-) Nov 16 21:11:23 http://pastebin.com/826075 is for lkml ;D Nov 16 21:12:36 hello; is there anyway to force bitbake to do only and nothing but compilation (no further things like packaging etc) Nov 16 21:14:27 ideally if it could do regardless of stamps:) Nov 16 21:15:08 mmp: bitbake -h Nov 16 21:15:15 looks for 'task' Nov 16 21:15:45 koen: ahaa; I so far used just bitbake -i, and there I have not seen anything like that:) Nov 16 21:15:55 -c Nov 16 21:16:31 Gerrath: another patch sent to lkml Nov 16 21:20:48 "hats mean ubuntu? Nov 16 21:20:48 it's african word for "can't configure debian"" Nov 16 21:20:53 +w somewhere Nov 16 21:22:27 koen hehe ... no it's african "Dude I think I fscked up something" Nov 16 21:22:43 v8jlene: so how it will be with docs? Nov 16 21:22:56 v8jlene: I planned writing some during weekend Nov 16 21:23:03 chouimat|away: http://xkcd.com/c178.html Nov 16 21:23:52 ggilbert, I'm going to Ojai today. Nov 16 21:24:08 ggilbert_, I'm going to Ojai today Nov 16 21:24:33 koen hehe Nov 16 21:24:51 ka6sox-office: Ojai, as in Ojai, CA? Nov 16 21:25:00 yes Nov 16 21:25:03 koen: did you see that the cairo 1.3.2 snapshot was released Nov 16 21:25:06 hrw|busy: I'll look at getting it out to you later today. (Currently it's mostly in the form of electronic post-it-notes attached to my openembedded.info bb receiped page) Nov 16 21:25:08 ka6sox-office: my boss lives there Nov 16 21:25:27 why is it called "populate_staging" instead of "stage". (to) stage is a verb, right? Nov 16 21:25:29 small world. Nov 16 21:25:39 ka6sox-office: oh? Nov 16 21:25:41 ka6sox-office: yes, Ojai is a small word :-) Nov 16 21:25:50 I have the machine here at the office in Ventura :) Nov 16 21:26:00 you don't happen to be taking the 101 to the 33 do you? Nov 16 21:26:15 ggilbert_, 101 to 126 to 150 Nov 16 21:26:19 ah Nov 16 21:26:30 ka6sox-office: Do you work in LA? Nov 16 21:26:35 but I can stop anywhere Nov 16 21:26:46 Speedy2, I'm in Santa Barbara Nov 16 21:26:52 Nice. Nov 16 21:27:08 We're off the 33 at Canada Larga. If you want the machine today. If not I'll be up next week Nov 16 21:27:17 <-- San Diego. Nov 16 21:27:19 v8jlene: ok Nov 16 21:27:32 ggilbert_, I'll stop by if thats okay. Nov 16 21:27:39 RP, you here man? Nov 16 21:27:57 sure thing Nov 16 21:28:15 hrw|busy, you're next in line for spitz questions :D Nov 16 21:28:17 What time? I just need a chance to make sure there was nothing valuable on the drive. :) Nov 16 21:28:28 ggilbert, you tell me Nov 16 21:28:51 I'm going to hook up a Stereo Generator up on the mountian as soon as it arrives. Nov 16 21:29:07 I'll be here until 5. I need probably about an hour to go grab the machine and take a look over it Nov 16 21:29:12 any time after that is fine though Nov 16 21:29:26 ggilbert_, it'll be at least an hour Nov 16 21:29:38 mreimer: cairo 1.3.2? http://cairographics.org/releases/ does not show. Nov 16 21:29:50 perfect Nov 16 21:30:02 likewise: http://cairographics.org/snapshots/cairo-1.3.2.tar.gz Nov 16 21:30:08 likewise: http://cairographics.org/news/cairo-1.3.2 Nov 16 21:30:09 We're off the 33 at Canada Larga. There's a big building labeled Brooks Nov 16 21:30:18 mreimer: thanks. will bb it. Nov 16 21:30:23 likewise: but it needs pkgconfig 0.19 or greater, which isn't in oe yet Nov 16 21:30:26 thanks likewise Nov 16 21:30:38 mreimer: I added a patch. It seems to work with 0.15 here Nov 16 21:30:46 likewise: great Nov 16 21:33:41 mreimer: I didn't see that yet Nov 16 21:33:49 mreimer: does that have the magic number patches? Nov 16 21:33:58 koen: yes, it should Nov 16 21:34:31 koen: Carl said this snapshot renders text on his n770 5x faster than before Nov 16 21:34:38 koen: among other nice speedups Nov 16 21:34:47 * koen reads http://cairographics.org/news/cairo-1.3.2 Nov 16 21:35:31 "Here we see that the subimage_copy improvement was only about half as Nov 16 21:35:31 large as the corresponding improvement on my laptop, (27x faster Nov 16 21:35:32 compared to 54x) and the floating-point conversion fix also was quite Nov 16 21:35:32 as significant, (2x compared to 3x). Oddly the improvement to text Nov 16 21:35:32 rendering performance was more than twice as good (5x compared to Nov 16 21:35:32 2x). I don't know what the reason for that is, but I don't think it's Nov 16 21:35:34 anything anybody should complain about." Nov 16 21:35:36 about the n770 Nov 16 21:36:24 if someone creates a .bb for it, I'll put it in the angstrom feeds for people to test :) Nov 16 21:36:54 koen: building as we speak Nov 16 21:36:54 koen: likewise is cooking one now Nov 16 21:37:13 koen: needs a small patch to persuade to cope with pkgconfig 0.15 in configure.in Nov 16 21:37:33 updating pkgconfig is hell.... Nov 16 21:37:51 hi all Nov 16 21:37:52 koen: i figured... who did the 0.15 one? Nov 16 21:37:57 the diff to its internall glib needs to get redone Nov 16 21:38:03 likewise: no idea Nov 16 21:38:13 likewise: but florian added a newer m4 iirc Nov 16 21:38:31 likewise: so it has features of a newer pkgconfig Nov 16 21:38:46 * likewise is sponsoring $25 for latest pkgconfig Nov 16 21:38:49 :-) Nov 16 21:39:00 likewise: maybe cairoclock can run in realtime now Nov 16 21:39:10 1fps ? Nov 16 21:39:37 koen: depends on size. I get 1 fps on a 800x800 cairoclock on the IXP420@533MHz Nov 16 21:39:54 koen: correction: ~3 fps on a 800x800 cairoclock on the IXP420@533MHz Nov 16 21:40:24 suihkulokki: ^ Nov 16 21:40:48 NOTE: package cairo-1.3.2: completed Nov 16 21:41:56 how does OE or quilt help me getting a patch out? I normally do this by hand but am wondering if bitbake interactive can help me? Nov 16 21:42:21 quilt new my.patch;quilt add file file2 file3 Nov 16 21:42:30 edit file file2 file3; quilt refresh Nov 16 21:42:43 and then copy patches/my.patch somewhere Nov 16 21:43:02 simple Nov 16 21:43:05 likewise: 'mtn add packages/cairo ; mtn diff packages/cairo' Nov 16 21:43:14 hrw|busy: right after do_patch? Nov 16 21:43:32 ah, that kind of patches Nov 16 21:44:18 hrw|busy: ok, so interactive bitbake up to "patch", then your easy stuff, then build? Nov 16 21:45:52 you could make a create_patch.bbclass that extracts the source to another dir and invokes quilt Nov 16 21:46:16 ~change 1 eur to pln Nov 16 21:46:22 1.00 Euro (EUR) makes 3.79159 Zloty (PLN) (from http://www.xe.com/) Nov 16 21:46:36 koen: I never used quilt much, but yes, get the idea. Nov 16 21:46:50 ~change 1 usd to pln Nov 16 21:46:55 1.00 United States Dollar (USD) makes 2.96051 Zloty (PLN) (from http://www.xe.com/) Nov 16 21:48:38 hmm.. looks like my fetching failed Nov 16 21:49:32 koen: do we use some dir on openembedded.org website for mirroring stuff? Nov 16 21:50:00 hrw: http://www.openembedded.org/sources/ Nov 16 21:50:11 (yes, that 403s) Nov 16 21:50:19 ok Nov 16 21:52:20 lrg? Nov 16 21:52:44 ~seen lrg Nov 16 21:53:05 lrg was last seen on IRC in channel #oe, 1d 4h 40m 33s ago, saying: 'hvontres|poodle: yep and in latest patch'. Nov 16 21:53:24 Spyro: he is mostly during european workday time Nov 16 21:53:42 :) Nov 16 21:54:03 hey mr Molton Nov 16 21:54:12 hi koen... Nov 16 21:54:18 I'm looking for liam :) Nov 16 21:54:34 Drive north to Edinburgh Nov 16 21:54:45 cross the river and start yelling ;) Nov 16 21:54:50 very funny. Nov 16 21:54:58 wonder if my car would make it that far :) Nov 16 21:55:36 did you get your engine patched up? Nov 16 21:57:11 koen since the turbo blew? Nov 16 21:57:13 yeah Nov 16 21:57:27 was ok for a while now I have a water leak somewhere I cant find. Nov 16 21:59:37 I'm off home, my distro is still building the toolchain :> Thanks everyone for your help Nov 16 22:02:45 mreimer: is cairo 1.3.2 a stable snapshot? Nov 16 22:02:55 likewise: no, it's dev snapshot Nov 16 22:03:01 so -1 Nov 16 22:03:38 mreimer: I forgot what goes before the -1. PREFERRER_PRIORITY? Nov 16 22:03:50 DEFAULT_PREFERENCES = "-1" Nov 16 22:03:51 s/PREFERRER/PREFERRED/ Nov 16 22:03:53 ah yes Nov 16 22:04:08 likewise: oops, drop the trailing 'S', I think Nov 16 22:05:32 Must say the new site layout is quite impressive Nov 16 22:06:32 re Nov 16 22:07:08 hi florian Nov 16 22:07:51 hey Spyro Nov 16 22:09:36 mreimer: http://bugs.openembedded.org/show_bug.cgi?id=1594 Nov 16 22:09:42 mreimer: adds cairo 1.3.2 Nov 16 22:09:51 likewise: thanks! Nov 16 22:10:12 mreimer: where are you deploying cairo on? (Quite busy myself with cairo as well!) Nov 16 22:10:30 likewise: just using it on my hx4700 Nov 16 22:11:02 mreimer: ok. my girl has a hx4700 on lend for her work. My fingers are tingling to flash it... :-) Nov 16 22:11:08 :-) Nov 16 22:11:21 likewise: you can boot from haret, so you don't have to flash :-) Nov 16 22:11:27 haret? Nov 16 22:11:37 likewise: http://handhelds.org/moin/moin.cgi/HaRET Nov 16 22:12:47 mreimer: thanks, didn't know this. Another device, too little time. Nov 16 22:17:46 Kristoffer: I am around now... Nov 16 22:18:53 wb RP Nov 16 22:21:10 hi RP Nov 16 22:21:18 hi mreimer! Nov 16 22:21:19 hi Spyro, hrw Nov 16 22:24:05 likewise: cairo 1.3.2 works for me :-) Now to rebuild gtk+ 2.10.6 without the pangoxft2 patch... Nov 16 22:24:43 I bet xft is still loads faster as cairo, but we can only know by testing :) Nov 16 22:24:45 mreimer: I just uploaded a MTN diff instead of the seperate .bb and .patch, like koen told me. Nov 16 22:25:42 likewise: shall I go ahead and commit it? I see no reason not to commit it Nov 16 22:26:02 mreimer: sure. I have no commit access so go ahead if you have those rights. Nov 16 22:26:49 RP, got it fixed. Had a problem configuring alsamixer :D Nov 16 22:27:36 koen, likewise: the new tessellator didn't make it into 1.3.2. :-( Nov 16 22:28:01 mreimer: it is still in a seperate branch, right? Maybe a new GIT checkout bitbake recipe will do? Nov 16 22:28:02 hrm Nov 16 22:28:24 "I ended up not throwing the new tessellator into the 1.3.2 snapshot. I had gotten it to be 1.5x faster on world_map, but I also got the same 1.5x speedup on the old tessellator with the same custom quadrilateral tessellation fix. Nov 16 22:28:27 or be a little bit more patient. The new tesselator is not always better AFAIK Nov 16 22:28:41 likewise: I think you're right, on both counts Nov 16 22:29:20 mreimer: I truely hoped the new tesselator to be a factor higher faster, but it seems not yet. Nov 16 22:29:35 mreimer: Carl is also looking at completely removing the tesselator Nov 16 22:29:45 likewise: interesting Nov 16 22:30:18 mreimer: it does not make much sense to triangulate if there is no triangles hardware support (everything except high-end GPUs) Nov 16 22:30:33 likewise: agreed Nov 16 22:31:00 mreimer: so I suspect the 'span" approach of freetype could be used. But my graphics knowledge is not as in-depth as the cairo hackers. Nov 16 22:31:16 s/the/that of the/ Nov 16 22:33:22 likewise: 3dfx had triangles, S3 Virge3D also had them Nov 16 22:33:39 hrw: sure: every GPU has them, but do our targets have them? Nov 16 22:33:45 I remember old Amiga times when those cards were highend in Amiga machines Nov 16 22:34:21 likewise: my current target rather does not have. its just 4M gfx chipset Nov 16 22:34:39 i2700g and i.mx31 have '3d' Nov 16 22:34:46 and omap3 Nov 16 22:35:30 koen: I think cairo needs opengl interfacing to the cards in order to really benefit from hw accel Nov 16 22:35:53 koen: or one has to write a custom driver / cairo backend for it Nov 16 22:36:11 hrw: Ahh Amiga. Nov 16 22:36:34 hrw: The totally unbloated machine. Nov 16 22:38:13 likewise: like glitz? Nov 16 22:39:38 likewise: amiga with ppc604/233 and Permedia2 was quite nice machine Nov 16 22:40:32 hrw: never got a PowerPC :-/ Nov 16 22:40:50 hrw: but aiming to work with the PowerPC 8343 soon Nov 16 22:41:16 likewise: neither do I - it was cheaper and better to drop amiga and switch to x86 for me Nov 16 22:42:04 hrw: same here, although even years later I still could not do a smooth real-time Scala - like broadcast presentation on an x86 system... Nov 16 22:42:29 guys: how to isntall something on RHEL? no yum, no apt-get Nov 16 22:42:44 rpm -qipl foo.rpm? Nov 16 22:43:01 koen: you forgot --force-lack-of-depends Nov 16 22:43:28 haven't used rpms since mandrake 6 Nov 16 22:44:02 I since RedHat 5.1 unofficial Amiga port 8 years ago Nov 16 22:44:03 gah, my git was way out of date... Nov 16 22:44:27 I ditched RPM 8 years ago too (on arm) Nov 16 22:44:30 Spyro: now you need to re-port just ported e740? Nov 16 22:45:42 I'm curious does someone warned siemiens sx1 guys before pavel machek started hacking on it.. Nov 16 22:46:20 hrw: ? Nov 16 22:46:47 Spyro: joking Nov 16 22:47:03 hrw :) Nov 16 22:47:26 hrw: he hasn't bothered the openezx just either Nov 16 22:47:38 sadly lkcl did show up Nov 16 22:47:49 Spyro: you ported e740 to current, now you say that your tree was old, so... Nov 16 22:48:05 hrw: no my git binaries were old Nov 16 22:48:10 koen: maybe he fear "I'm GPL" guy Nov 16 22:48:27 it was looking like it updated when I pulled it but it didnt so it got out of date. was using like 1.1.3 !!! Nov 16 22:49:38 goodnite guys, cya tomorrow Nov 16 22:52:27 btw - someone tried to use pata_pcmcia instead of ide-cs? Nov 16 22:54:41 nope Nov 16 22:54:49 I should try the libata stuff soon Nov 16 22:54:54 12-9 = 15:20 Nov 16 22:55:00 * koen waits for lennert and dwery to finish the patch Nov 16 22:55:09 koen: I use libata on my home machine (PATA devices) Nov 16 22:55:39 9h time differency between me and customer hurts. Nov 16 22:56:17 from one side it is good - I can use most of bandwidth during their night Nov 16 22:56:37 but if need to discuss something I have to be during late evening online Nov 16 23:02:15 Compiling per-native takes forever : Nov 16 23:02:17 : Nov 16 23:02:19 :( Nov 16 23:03:22 perl-native? Nov 16 23:03:31 hrw: yes Nov 16 23:04:18 was quite fast here Nov 16 23:04:35 ~hail haerwu business for giving me amd64 Nov 16 23:04:36 ;D Nov 16 23:04:42 * ibot bows down to haerwu business for giving me amd64 and chants, "I'M NOT WORTHY!!" Nov 16 23:06:57 You can logoff, but you can never leave. -- Nina Liedtke, "CyberJoly Drim" Nov 16 23:07:28 so its time for me to logoff to bed Nov 16 23:07:28 'night hrw Nov 16 23:08:16 night hrw Nov 16 23:08:16 weekend will be work, work, work Nov 16 23:08:29 but it will be OE related work Nov 16 23:08:52 mreimer: cairo stuff committed in a slightly modified fashion Nov 16 23:09:11 koen: great, was just about to do that. thanks Nov 16 23:09:16 koen: seems to be working here Nov 16 23:19:17 ~hail ibot for being ibot Nov 16 23:19:24 * ibot bows down to ibot for being ibot and chants, "I'M NOT WORTHY!!" Nov 16 23:19:30 mreimer: http://www.angstrom-distribution.org/repo/?action=details&pnm=libcairo2 Nov 16 23:19:31 ibot: yes you are Nov 16 23:19:47 raduga: just say ~botsnack :) Nov 16 23:19:58 ~koensnack Nov 16 23:20:19 "Upgrading libcairo2 on root from 1.2.4-r2 to 1.3.2-r0..." Nov 16 23:20:55 ~emulate raduga Nov 16 23:21:01 cu Nov 16 23:22:07 hrw: to emulate raduga would be to transform yourself into something lame and worthless Nov 16 23:22:31 hrw: but, have a nice time, being gone Nov 17 00:45:35 where's the best channel to discuss haret? **** ENDING LOGGING AT Fri Nov 17 02:59:57 2006