**** BEGIN LOGGING AT Fri Jun 02 02:59:56 2006 Jun 02 03:31:07 ppl, I'm running kernel 2.6.15 on my h2200. However I got this from Matt Reimer. Jun 02 04:00:53 thejapa: did you get an image ? Jun 02 04:01:10 they say the SD failure is a 2.4.x bug Jun 02 04:01:16 so i wanna try 2.6.x Jun 02 04:05:40 it's a zimage for h2200, Jun 02 04:06:14 i see Jun 02 04:06:34 it's on the 0.8.4-rc2.3.. er... this doesnt help much, does it? Jun 02 04:07:06 i have 0.8.4 Jun 02 04:07:24 i mean, will it work on other ipaqs? Jun 02 04:07:29 no idea Jun 02 04:09:19 u wanna try it? the link on the wiki is good, wait... Jun 02 04:10:44 no it wont Jun 02 04:10:52 the hardware has to be the same Jun 02 04:11:01 or atleast close enough Jun 02 04:11:11 oh Jun 02 04:11:35 power managment and chipsets are different Jun 02 04:11:40 i can't find the link anyway Jun 02 04:11:59 emte: do you know if rc2 comes with 2.6.x ? Jun 02 04:12:06 rc2 of what? Jun 02 04:12:13 familiar 0.8.4 Jun 02 04:12:17 which device Jun 02 04:12:24 h55xx Jun 02 04:12:40 yes i belive the h55 uses 2,6 Jun 02 04:12:42 i can't read my sd card anymore Jun 02 04:12:45 sec i'll look Jun 02 04:12:56 i got rc1 and it's still with 2.4.x Jun 02 04:12:58 buy a cf card? :P Jun 02 04:13:22 and that power discharge strap, whatever Jun 02 04:13:38 a power who what? Jun 02 04:13:54 CSMan got static problems. Jun 02 04:14:07 his touch reboots devices Jun 02 04:14:08 what type? Jun 02 04:14:27 then i suspect he has a short on his board Jun 02 04:14:34 CSMan, did you test the SD somewhere else? Jun 02 04:14:41 thejapa: yeah, it's working fine Jun 02 04:14:44 =( Jun 02 04:15:04 so, did you test another SD on your ipaq? Jun 02 04:15:15 that i haven't tried yet Jun 02 04:15:20 hehe Jun 02 04:15:37 hh is slow tonight Jun 02 04:15:52 ignoring mmc irq with null request Jun 02 04:15:52 first_minor 0x00000000 Jun 02 04:15:52 Partition check: Jun 02 04:15:52 /dev/mmc:ignoring mmc irq with null request Jun 02 04:15:52 ignoring mmc irq with null request Jun 02 04:15:55 oops Jun 02 04:16:03 ok Jun 02 04:16:12 u need cf Jun 02 04:16:30 i need 2.6.x =P Jun 02 04:16:44 can't you go back to wince to test it (desperate measures) Jun 02 04:17:00 ouch Jun 02 04:17:20 what's the extension of the wince image ? Jun 02 04:17:26 gee, no idea Jun 02 04:17:33 i flashed mine without a backup Jun 02 04:17:52 its whatever you named it Jun 02 04:17:59 i have it yeah Jun 02 04:18:04 ok, i'll try that Jun 02 04:18:21 hope it was good advice Jun 02 04:18:44 how many flashes does the ipaq resist ? Jun 02 04:19:38 no idea Jun 02 04:20:25 * emte vaguly remembers a posibility of using the h39 image on the h5x Jun 02 04:22:50 anyway Jun 02 04:22:53 http://www.handhelds.org/moin/moin.cgi/HpIpaqH5400 Jun 02 04:22:59 is the h5x series page Jun 02 04:24:57 Receive buffer too small (01FC0000) to accept file size 02F80000 Jun 02 04:28:38 wow Jun 02 04:28:45 no idea Jun 02 04:28:51 not good advice Jun 02 04:30:57 what's the setting to bring console output during boot ? Jun 02 04:31:54 you have serial cradle? Jun 02 04:32:09 oh, sorry, understood wrong Jun 02 04:33:17 console=tty0 ? Jun 02 04:34:44 yeah Jun 02 04:36:31 it works? i was trying to remember it other day... Jun 02 04:44:22 need sleep, good night Jun 02 07:37:48 morning Jun 02 08:55:05 hey lrg Jun 02 08:56:24 hey XorA, I think I've fixed the audio now. Will send a patch later. Jun 02 08:56:58 lrg: cool, I can test Jun 02 08:57:20 http://news.zdnet.co.uk/0,39020330,39272166,00.htm I want half a dozeon of these running linux Jun 02 08:58:44 ask santa claus Jun 02 08:59:24 morning all Jun 02 08:59:49 XorA: looks like it's alchemy, so it will run Linux :) Jun 02 09:00:08 morning lardman|work Jun 02 09:01:33 XorA: heh, the "2 years to perfect" doesn't look good for WinCE in time to market ! Jun 02 09:01:59 hi liam lardman Jun 02 09:02:06 hey Marcin Jun 02 09:02:11 hi lrg, hrw, XorA Jun 02 09:04:20 morning all Jun 02 09:04:46 hey Richard Jun 02 09:07:36 hi RP Jun 02 09:11:24 hey RP Jun 02 09:15:54 with an LCD with vga port that thing would work nice behind my TV as a mythtv frontend Jun 02 09:19:01 hi RP Jun 02 09:20:14 Its a Newcastle based company. It'd be nice if OE supported it. 32MB or 64MB DoC flash... Jun 02 09:24:20 RP: is that your end of the world, maybe you should try and get a couple of demo units :-) Jun 02 09:31:26 hey Jun 02 09:31:57 hi zecke Jun 02 09:32:02 zecke: ello Jun 02 09:32:09 kergoth: which version of BitBake? Jun 02 09:33:00 ~botmail for kergoth Which version of BitBake, if you run with -vvv what does it say? Jun 02 09:33:56 yo zecke Jun 02 09:39:53 damn, I bought crap at the bakery :( Jun 02 09:44:16 XorA: Maybe I should try. I don't really have the time for yet another project atm though :-/ Jun 02 09:44:38 zecke: Did you see kergoth's comment about BBFILES? Jun 02 09:45:10 RP: no Jun 02 09:45:23 RP: I got a BotMail from him claiming we owe him a dependency Jun 02 09:45:49 zecke: owe him a depencency? Jun 02 09:46:08 zecke: He found BBFILES will only take globs now and not directories Jun 02 09:46:10 "1 Jun 02 09:46:27 okay Jun 02 09:46:37 but this code wasn't touched by me/you anyone :) Jun 02 09:47:06 zecke: right, I knew I hadn't touched it :) Jun 02 09:47:21 RP: http://oe.pastebin.com/753152 Jun 02 09:48:04 RP: I will add it to my 'look-into'-list, need to write stuff now :} Jun 02 09:48:13 zecke: Hmm :-/ Jun 02 09:48:43 RP: I bet he just used some old warezed version of BitBake Jun 02 09:50:12 zecke: I think he was using 1.4.2 or 1.5 Jun 02 09:50:42 01 23:10 < hrw|husky> BBFILES="/path/to/*/*.bb" here Jun 02 09:50:42 01 23:10 < hrw|husky> but i use 1.4.x not trunk Jun 02 09:50:44 01 23:11 < kergoth> i'm also using the latest release Jun 02 09:50:44 lol http://blog.janik.cz/archives/2006/06/02/T08_31_26/ Jun 02 09:51:00 kergoth: whiner! Jun 02 09:51:15 later guys Jun 02 09:51:38 cya zecke Jun 02 09:51:43 cu zecke Jun 02 10:05:05 zecke|away: kergoth is right, bitbake is doing something very weird. Its not the dependency code at fault either - it tries to build autoconf-native and somehow automake-native appears first? :-/ Jun 02 10:09:13 can it be base.bbclass being at fault Jun 02 10:10:32 zecke|away: I think autoconf-native RDEPENDS on automake-native and automake-native RDEPENDS on autoconf-native :-} Jun 02 10:11:02 So bitbake might well be doing the right thing Jun 02 10:11:03 RP:bitbake -g automake-native autoconf-native ;) Jun 02 10:11:15 RP: we should see a dashed circle :} Jun 02 10:13:09 Why does autoconf-native RDEPEND on automake-native? That sounds like a bug. You can use autoconf without automake, can't you? Jun 02 10:13:59 pb_: Its actually RRECOMMENDing Jun 02 10:14:34 But bitbake lets RDEPEND == RRECOMMEND as far as its concerned... Jun 02 10:14:39 I'm not convinced it should do that either. I would have thought it should be RSUGGESTS at the most. Jun 02 10:14:59 I'm not sure why its doing that either Jun 02 10:18:41 bitbake does violate a DEPENDS here as automake-native DEPENDS on autoconf-native whereas autoconf-native only RRECOMMENDS automake-native Jun 02 10:34:22 hi all Jun 02 10:34:54 hi florian Jun 02 10:47:33 hi florian_kc Jun 02 11:04:11 so i had to reinstall wince to check the sd card Jun 02 11:04:24 and fortunately it is working Jun 02 11:04:59 so , does anybody have an oe image with the 2.6.x kernel for the ipaq h55xx series ? Jun 02 11:37:05 When a patch is applied with OE what is the base directory for the patch? I have a kernel patch, should I assume it is applied from the root of the kernel source or is it applied one level up? Jun 02 11:37:26 normal rules apply: -p1 patches Jun 02 11:38:04 hrw|work, and I do that by PATCH=1 correct? Jun 02 11:38:32 file://your.patch;patch=1 Jun 02 11:38:48 or any other url Jun 02 11:38:54 hrw|work, yes thats how I have it. Jun 02 11:40:56 S Jun 02 11:42:59 ~pastebin Jun 02 11:43:07 methinks pastebin is a place to paste your stuff without flooding the channel - try http://pastebin.com/, or http://pastebin.ca, or http://channels.debian.net/paste, or for #oe use http://oe.pastebin.com, or http://bzflag.pastebin.ca/ Jun 02 11:43:43 hi cbrake Jun 02 11:43:59 hi cbrake Jun 02 11:44:10 hrw|work: Gerrath morning Jun 02 11:50:16 hrw|work, here is the top part of the patch file: http://bzflag.pastebin.ca/60974 Jun 02 11:51:19 Gerrath: how would you apply this patch by hand? Jun 02 11:51:33 Gerrath: patch should be linux/*/arch/arm Jun 02 11:55:23 zecke|away, I just run patch -p0 < patchfilename from the root of the linux source tree. Jun 02 11:56:03 hrw|work, shouldn't I just be able to set patch=0 and it should be the same as patch -p0 manually? Jun 02 11:56:29 Gerrath: then add ;pnum=0 to the line hrw pasted? Jun 02 11:56:55 Gerrath: patch=1 just says. "I'M A PATCH" Jun 02 11:57:04 Gerrath: pnum=X says use -pX Jun 02 11:57:06 zecke|away, ahh, thanks, thats were my confusion was. Jun 02 11:58:04 and don't omit the ';' Jun 02 11:58:24 URI;option1=value1;option2=value2 Jun 02 11:59:30 zecke|away, thanks again. Jun 02 11:59:57 hrw|work, thanks. Jun 02 12:06:54 zecke|away, hrw|work I just gave the patch a try with pnum=0 and it worked, thanks again. Jun 02 12:07:39 you are welcome Jun 02 12:11:46 np Jun 02 12:31:29 ARGH.... Jun 02 12:31:40 my root-on-sd is destroyed again ;( Jun 02 12:32:13 fscking shit Jun 02 12:34:28 hrw|work: what happened Jun 02 12:35:15 some dirs/files vanished Jun 02 12:35:23 root@c7x0:~# ls -l /usr/lib/ipkg Jun 02 12:35:23 ls: /usr/lib/ipkg/status: Input/output error Jun 02 12:35:25 etc Jun 02 12:35:36 hrw|work: ah, just needs fscked then Jun 02 12:36:24 maybe Jun 02 12:37:52 koen: think I have a theory on abiword plugins, but Im not a ld.so export Jun 02 12:38:22 hrw|work: that is ext2 preventing access to prevent further damage Jun 02 12:38:47 hrw|work: I guess you rebooted without unmount Jun 02 12:39:13 XorA: we need checkrootfs initscript which check where does he boot from Jun 02 12:39:58 XorA: rebooted is wrong word... 'had to reboot due to hang' is better Jun 02 12:42:02 XorA: after fsck I lost /usr/lib/ipkg/status totally Jun 02 12:42:31 hrw|work: process entered D state while writing to card? Jun 02 12:46:11 XorA: hard to tell - yesterday I was able to install stuff. opie hang later when ipkg was not running Jun 02 12:47:04 hrw|work: it sounds like you are falling over same problem with / on SD I did :-( Jun 02 12:47:55 XorA: Z suxx.. Jun 02 12:48:01 * hrw|work want 512MB flash Jun 02 12:50:27 "I must finish to install my IRQ handler on my strongARM (I'm a trainee and my Jun 02 12:50:27 training period ends in one week!!!)." Jun 02 12:50:28 lol Jun 02 12:57:03 hi guys Jun 02 12:57:09 RP: you're here? Jun 02 12:57:36 hi mickeyl Jun 02 13:00:01 hi mickeyl Jun 02 13:00:27 RP: hey. lets say I have two LEDs, I register two led devices, but both have to be set at once. Would it be ok to use one set_brightness() function, find out, which LED triggered that function and then grab the last state of (how?) of the other LED and then set both always at once? Jun 02 13:01:22 mickeyl: Yes, you'd probably be best tracking the state if the leds in some variable Jun 02 13:02:01 RP: ok, now the newbie question (/me hides in shame): can i just add static variables in a module or is there some recommended procedure to alloc/dealloc memory necessary? Jun 02 13:03:16 mickeyl: That depends on how much use you expect this code to see. If you had 20 pairs of LEDs (or a variable number of them), dynamic memory allocation would be preferable. If its just two leds in a very machine specific module, I'd say a static variable is ok Jun 02 13:04:18 RP: excellent. thanks, that'll be all I need to know to add an EZX driver based on your LED subsystem :) Jun 02 13:04:49 mickeyl: You can encapusulate LED state information around the struct led_classdev - see leds-s3c24xx.c for an example Jun 02 13:05:11 but again, its probably overkill for what you want Jun 02 13:06:01 mickeyl: I'm the official LED maintainer so you know where to send the driver when its finished ;-) Jun 02 13:06:13 going home Jun 02 13:07:01 RP: cool :) That already answers my next question whether the subsystem has been accepted into mainline yet Jun 02 13:07:34 i hope you'll be a tough maintainer :) Jun 02 13:07:46 cya Jun 02 13:07:50 hey zecke Jun 02 13:07:56 mickeyl: It will be in 2.6.17 due for release anytime soon Jun 02 13:08:10 congrats. good job. Jun 02 13:08:12 hi mickeyl Jun 02 13:08:17 hey pb_ Jun 02 13:08:37 mickeyl: Along with the RTC subsystem which is nice from the Zaurus PoV Jun 02 13:09:06 The NSLU people pushed that along and we swapped drivers :) Jun 02 13:09:36 very nice. What can we do with the RTC? Jun 02 13:10:28 mickeyl: Basically we get an accurate timer. One of these days it might even survive reboots... Jun 02 13:10:36 cool Jun 02 13:11:10 I have an rtc test program somewhere which illustrates things... Jun 02 13:11:56 i'd be interested to see that when you have a chance Jun 02 13:13:11 mickeyl: http://www.rpsys.net/openzaurus/temp/tests/rtctest.c Jun 02 13:13:24 nfsroot over usb is working nice on the MOTO phones, so even if I have problems reading from SD on the e680, I can develop nicely. Jun 02 13:13:28 RP: thanks. /me grabs Jun 02 13:14:46 That should just work on any pxa2xx now as long as you compile the RTC driver in Jun 02 13:16:01 evtest.c in the same place is also useful Jun 02 13:19:03 the RTC patches are also in mainline .17? Jun 02 13:19:35 mickeyl: yes Jun 02 13:19:41 good Jun 02 13:19:59 mickeyl: Have a look at the difference between the 2.6.16 kernel and 2.6.16+git in OE ;-) Jun 02 13:20:00 means a whole lot less patches for linux-openzaurus 2.6.17 Jun 02 13:20:05 ah yeah Jun 02 13:20:28 and that's a fairly old git kernel its set to atm as well... Jun 02 13:33:04 * mickeyl wanders home Jun 02 13:33:39 * emte wanders mickeyl's home Jun 02 13:55:17 * chouimat is back. Jun 02 13:55:30 morning Jun 02 14:21:54 03mickeyl 07org.oe.dev * r8ca7c176... 10/packages/linux/ (3 files in 2 dirs): linux-ezx: add LED subsystem from 2.6.17, start with LED driver for E680 Jun 02 14:26:08 jo mickeyl Jun 02 14:34:43 hey woglinde, hi zecke Jun 02 14:35:02 hey Jun 02 14:35:09 woglinde: I wanted to ask you something... Jun 02 14:36:28 http://www.cgeo.net/carlos/blog/?p=12 Jun 02 14:37:39 zecke: going to migrate? Jun 02 14:38:33 zecke what you want from me? Jun 02 14:38:58 hrw|work: no Jun 02 14:39:09 hm release manager Jun 02 14:39:12 that would be nice Jun 02 14:39:16 and tough Jun 02 14:39:54 Linux Software Engineer, Address Book Jun 02 14:39:56 hihi Jun 02 14:40:00 yeah Jun 02 14:40:00 they get it Jun 02 14:40:12 they should contract o-hand once more Jun 02 14:40:35 :) Jun 02 14:40:44 Instant messaging, Voice over IP (VoIP) Jun 02 14:40:50 the google stuff Jun 02 14:40:54 yes Jun 02 14:41:01 woglinde: did you see decibel.basyskom.de? Jun 02 14:41:17 nowadays even decibel.kde.org Jun 02 14:42:43 woglinde: SIP, RTP, they should use JMF Jun 02 14:42:53 lol Jun 02 14:44:24 http://www.trolltech.com/careers/open-positions/berlin Jun 02 14:44:25 hmm Jun 02 14:44:32 zecke: nice webpage (decibel) Jun 02 14:45:06 woglinde: http://www.trolltech.com/careers/open-positions/munich Jun 02 14:45:16 woglinde: they actually hired a quite good guy for munich Jun 02 14:45:27 woglinde: I don't get why he started working there... Jun 02 14:45:29 zecke do you find a job for me? Jun 02 14:45:55 hm the wave lorn? Jun 02 14:45:57 they Jun 02 14:46:11 woglinde: no Robert 'god' Griebl Jun 02 14:47:10 mickeyl, no email.. ? Jun 02 14:47:12 sadly the job offers come 2 years too early Jun 02 14:48:18 Pendalar: I'm afraid so. I couldn't manage to find another main server for OZ that allows us to dump the GBytes we need. Jun 02 14:48:47 Pendalar: which means we're stuck with ewi and just one mirror until we find another solution Jun 02 14:49:18 mickeyl: did we get kicked at SF.net? Jun 02 14:49:34 zecke: we get kicked from sf.net in last year Jun 02 14:49:35 zecke: no, but they finally enforced their quota Jun 02 14:50:08 they finally noticied :} Jun 02 14:50:23 "Mother told me to be good, but she's been wrong before." Jun 02 14:50:32 how many gigs? Jun 02 14:50:32 woglinde: I find it funny that nokia is hiring more GNOME/Hildon people Jun 02 14:50:45 zecke why? Jun 02 14:50:47 woglinde: than TT Qtopia devs :} Jun 02 14:50:57 maybe there are comig more devices Jun 02 14:50:58 Pendalar: 3.5.4 is 2.5G now Jun 02 14:51:23 Pendalar: 3.5.1->3.5.4 is 4G Jun 02 14:51:37 do you hold other releases concurrently? Jun 02 14:51:51 302M ./3.5.1 Jun 02 14:51:51 466M ./3.5.2 Jun 02 14:51:51 832M ./3.5.3 Jun 02 14:51:51 2.5G ./3.5.4 Jun 02 14:51:51 4.0G . Jun 02 14:52:10 Pendalar: we can probably drop <3.5.3 Jun 02 14:52:10 not too bad Jun 02 14:52:39 3.5.4 contain sources (before we did not backup them and relied on oesources) Jun 02 14:53:40 Pendalar: and each release == big bandwidth usage.. 3.5.4 was 16M/s for few hours iirc Jun 02 14:54:41 I've got oz@pdai down for 50G transfer a month, but the HD quota would now be shy Jun 02 14:54:56 zecke: it was 2005-11-08 when we got 'quota reached' mail Jun 02 14:56:58 koen: I have solved the abiword plugins problem Jun 02 14:57:02 should I ping TT once again for Qtopia4 GPL? Jun 02 14:57:08 pb__: are you there Jun 02 14:58:49 hrw|work, mickeyl: I'll be getting my server in a month or two. I'm fine for disk space, but what's the bandwith like per month ? Jun 02 15:02:33 lrg: for me its hard to tell as I did not checked it Jun 02 15:06:22 http://oe.pastebin.com/753637 == gcc 3.4.4 from .oz354x does not build on ewi Jun 02 15:06:25 any ideas? Jun 02 15:07:35 hrw|work: uname says? Jun 02 15:07:55 hrw|work: you picked up x86_64 headers but are on 32 bit userspace :} Jun 02 15:08:09 hrw@bitbake:~/devel/build/3541$ uname -a Jun 02 15:08:10 Linux bitbake 2.6.8 #1 SMP Tue Jun 21 22:44:31 CEST 2005 i686 GNU/Linux Jun 02 15:08:25 hrw|work: don't use AMD-64 headers then Jun 02 15:08:57 zecke: its build from scratch. looks like something to investigate.. Jun 02 15:10:29 XorA: hello Jun 02 15:11:06 pb__: this is a longshot, but you are the most genius person I know with autoconf and toolchain stuff Jun 02 15:11:47 pb__: abuword-plugin config file on x86/amd64 will add a -Wl,--export-dynamic to LDFLAGS, when cross compiling for arm it doesnt, is there known obvious mistakes in this area Jun 02 15:12:05 configure script I mean Jun 02 15:13:14 yes Jun 02 15:13:20 livestream form rock am ring Jun 02 15:15:30 bye guys Jun 02 15:20:04 cu Jun 02 15:25:34 XorA: can't think of anything obvious that would cause that, no. I guess you'd need to inspect the configure.in and look for the logic that adds this flag. Jun 02 15:28:26 * chouimat is away: lunch Jun 02 15:29:31 pb__: ok cheers, am running though the scripting now see if I see anything Jun 02 15:43:53 do i need to EXPORT functions in kernel module even if I just call them from inside? Jun 02 15:44:06 mickeyl: no Jun 02 15:44:22 mickeyl: not if they are local calls Jun 02 15:45:45 hmm Jun 02 15:45:46 odd Jun 02 15:46:03 aaah Jun 02 15:46:06 in fact, you don't want to Jun 02 15:46:07 * mickeyl slaps forhead Jun 02 15:46:15 give a hoot, don't pollute the namespace Jun 02 15:46:35 s/exz/ezx/ Jun 02 15:46:36 :D Jun 02 15:46:42 now lets find the last missing symbol Jun 02 15:46:45 set_GPIO_mode Jun 02 15:46:48 * mickeyl browses Jun 02 15:46:57 mickeyl: did you set up a lxr already? Jun 02 15:47:07 not yet Jun 02 15:47:30 tries hard to remember Jun 02 15:50:22 hi Jun 02 15:53:28 how does one set a GPIO in 2.6 ? Jun 02 15:53:43 or configure, for that matter Jun 02 15:54:28 pxa_gpio_mode? Jun 02 15:54:33 mickeyl: sounds right Jun 02 15:54:37 thanks Jun 02 15:54:38 * mickeyl tries Jun 02 15:58:09 yeah! Jun 02 15:58:20 * mickeyl can turn on a red LED Jun 02 15:58:29 mickeyl: can you make it turn green? Jun 02 15:58:31 although I don't understand 90% of the code in my module Jun 02 15:58:34 mickeyl: or white? Jun 02 15:58:45 mickeyl: stop drinking! Jun 02 15:58:46 i guess so, it's a multicolor LED, but that code isn't in place yet Jun 02 15:59:21 all hail kernel h4x0r mickeyl! Jun 02 15:59:27 hehe Jun 02 15:59:46 * mickeyl going into a pub to celebrate his first kernel module Jun 02 15:59:48 l8er Jun 02 16:00:02 mickey|bbl: you wil forget the remaining 10% of you drink now! Jun 02 16:00:15 zecke_: I guess that's what he wants Jun 02 16:00:27 hehe Jun 02 16:00:35 mickey|bbl: did MVista write the code? Jun 02 16:04:19 yes, MVista Jun 02 16:04:57 mickey|bbl: bad luck Jun 02 16:07:32 hehe Jun 02 16:07:43 mickey|bbl: you should ask for a refund Jun 02 16:08:11 * zecke_ just mailed Trolltech... Jun 02 16:15:33 hmm people should just stop developing KDE :} Jun 02 16:17:08 zecke_: uh! Jun 02 16:17:32 without a doubt the KDE/Qt platform is better... Jun 02 16:17:48 but that is not of any importance :} Jun 02 16:18:01 and I start to hate the Not Invented Here syndrom Jun 02 16:18:13 anyway Jun 02 16:20:28 zecke_: maybe it is better... but the major disadvantage seems to be that it is frustrating. Jun 02 16:23:07 any perl dude around? Jun 02 16:25:11 zecke_: nice email btw Jun 02 16:25:32 lol Jun 02 16:25:55 chouimat: I don't get TT. Every company M$, Symbian, Perl hosted or used to host developer conferences Jun 02 16:26:11 chouimat: to get a community, and TT is doing the opposite, they ban the community Jun 02 16:26:21 chouimat: this can't be good for the health of Qtopia... Jun 02 16:26:32 zecke_: maybe qtopia is a second class product for them ... I don't understand too Jun 02 16:26:55 chouimat: it is generating revenue, I think they earn a shitload on implementing custom features :} Jun 02 16:27:16 chouimat: so it is natural they do not anyone else to be able to provide solutions as well Jun 02 16:27:29 chouimat: e.g. community creating patches, features, polising and new apps Jun 02 16:27:45 chouimat: each new app/feature can ... Jun 02 16:57:21 * koen returns from the middle of nowhere Jun 02 16:57:31 hi koen Jun 02 16:57:39 hey florian_kc Jun 02 16:57:46 * florian_kc -> food -> hackermeeting, bbl Jun 02 17:01:21 * koen -> food Jun 02 17:16:28 question about the following statement: FEED_URIS_append += "cross##http://ipkg.nslu2-linux.org/feeds/slugos-lag/cross/${DISTRO_FEED}" Jun 02 17:16:46 does _append and += do the same thing -- are they redundant? Jun 02 17:17:09 But what if you want to add to the _append string? Jun 02 17:19:31 cbrake: somehow redundant Jun 02 17:19:53 cbrake: but the operation is executed at a different time (dunno the semantic though) Jun 02 17:20:05 * chouimat is back. Jun 02 17:20:16 += is applied immediately, _append is applied when overrides are applied, iirc Jun 02 17:20:34 kergoth: right, but what happens with both at the same time ;) Jun 02 17:20:38 ahh Jun 02 17:20:58 cbrake: FEED_URIS_append += "" will append to other users of _append ;) Jun 02 17:21:07 instead of setting this as the only append :} Jun 02 17:21:35 kergoth: did you see RPs comment about the RRECOMMENDS of autoconf-native and automake-native? Jun 02 17:21:50 kergoth: but I did not find the time yet to look into your issue(s) Jun 02 17:23:41 didnt see the comment Jun 02 17:23:46 i should submit some bugs to the bts Jun 02 17:23:58 feel free to do so Jun 02 17:24:24 morning kergoth Jun 02 17:24:37 so far i've only found two of what look like bitbake bugs and one oe one. will go from there :) it just happens that i actually use a lot of the features the average user doesnt Jun 02 17:24:44 we /need/ unit testing. Jun 02 17:24:55 hey pb__ Jun 02 17:27:36 hmm, the oqo sure looks cute. awfully pricey though Jun 02 17:28:07 * cbrake experiments with append ... Jun 02 17:28:22 kergoth: we do, make me a millionar and I would stop uni... :} Jun 02 17:28:42 toi: Hi Transfer Object Identifier Jun 02 17:28:43 append and prepend exist in that form because we needed to have conditional appends/prepends, and our only conditionals were overrides Jun 02 17:28:47 zecke_: hehe Jun 02 17:28:48 toi: how is you ALC session like? Jun 02 17:29:09 Hi Zecke. Jun 02 17:29:35 Sorry, wrong abbreviation ;-) Jun 02 17:35:47 greetings ! Jun 02 18:18:59 anybody here using angstrom? Jun 02 18:22:47 only developers Jun 02 18:23:06 thejapa: but the main devs are not here (today) Jun 02 18:28:36 gosh I hate that pile of shit called OpenOffice Jun 02 18:31:48 zecke: thanks. btw, something must be happening, lists are quiet, it's quiet here, everywhere it's quiet. :{ Jun 02 18:32:07 hehe Jun 02 18:32:08 weekend Jun 02 18:32:11 time to get drunk Jun 02 18:32:15 yay Jun 02 18:32:54 or like pb_ would say "Time to get pissed mates, aye" :} Jun 02 18:32:55 zecke_: enjoy Jun 02 18:33:02 indeed Jun 02 18:33:12 need to finish stuff first :} Jun 02 18:36:32 damn dnb track uses a hard drive physical error sound. i got fuc*ing scared now. Jun 02 18:36:43 lol Jun 02 18:40:22 dnb? Jun 02 18:41:26 zecke_: new issue, bitbake -c fetch task-bootstrap doesnt build everything, it only hits the toolchain and task-bootstrap, none of the other task-bootstrap deps Jun 02 18:41:37 so grabbing the sources to do an offline build isnt feasable now Jun 02 18:42:01 kergoth: deps == RDEPS, okay Jun 02 18:42:10 kergoth: this is to avoid -cclean cleaning everything :} Jun 02 18:42:28 that used to be handled seperately. specific tasks marked as not following deps Jun 02 18:42:49 ah well Jun 02 18:42:59 how exactly do i populate my sources/ dir for an offline build? Jun 02 18:43:05 for i in ..? Jun 02 18:45:15 raduga: dnb -> drums'n'bass, electronic music Jun 02 18:46:49 kergoth: do you want to buy a support contract? Jun 02 18:47:34 kergoth: check bin/bitbake tryBuild Jun 02 18:48:22 kergoth: I have never seen this task flag Jun 02 18:53:19 hmm I can't find the bug report Jun 02 18:54:10 okay back to work Jun 02 19:10:19 * kergoth seds the output of bitbake -n, then passes that to a bitbake -cfetch Jun 02 19:10:20 * kergoth mutters Jun 02 19:10:40 kergoth: feel like trying a patch? Jun 02 19:14:02 kergoth: http://oe.pastebin.com/754110 Jun 02 19:14:19 okay, i'll mess with it later. busy with other things now Jun 02 19:15:47 simple enough, will probably just make it check for a task flag instead, less hardcoding of task naming in bitbake. that belongs in oe :) Jun 02 19:15:55 * kergoth needs to get familiar with the codebase again Jun 02 19:18:58 all: seen this? https://crossdev.timesys.com Jun 02 19:19:46 no Jun 02 19:20:33 Just more proof that making cross compiled images is hard Jun 02 19:24:51 hi folks Jun 02 19:25:08 I do not want to upset any with this topic, but ..... Jun 02 19:25:33 since the website has been moved and kergoth suggest that I remove everything in /home/oe on handhelds.org Jun 02 19:25:45 the proj_oe group it not really needed any more Jun 02 19:25:54 true Jun 02 19:25:56 and since on hh.org we give everybody shell accounts... Jun 02 19:26:39 france_: who would lose shell access? I guess none? Jun 02 19:26:47 so if I remove proj_oe some hh.org accounts will be canceled.. Jun 02 19:27:11 most people are members of many groups but I suspec that a few will be effected. Jun 02 19:27:19 suspect Jun 02 19:28:02 and what do you want to do about the oe@handhelds.org mailing list...are you creating your own as well? Jun 02 19:28:06 and /cvs/oe ??? Jun 02 19:28:50 france, send an email to the mailing list and request people contact you individually? Jun 02 19:28:54 france_: the main reason for moving stuff was to use a real CMS for the website Jun 02 19:29:08 france_: and to have monotone.openembedded.org Jun 02 19:29:34 zecke_: there real reason is that koen made true on his treat to move oe from hh.org along with any other project he could. Jun 02 19:29:45 s/treat/threat/ Jun 02 19:29:54 france_: no Jun 02 19:29:56 zecke_: www.hh.o can't handle drupal? Jun 02 19:30:00 it is not that easy Jun 02 19:30:26 zecke_: sure it was, he started the fight on the oe mailing list and said see, they are evil, lets move. Jun 02 19:30:47 france_: it is sadly not that easy Jun 02 19:31:12 ggilbert_: it would need to run monotone and viewmtn as well Jun 02 19:31:26 ggilbert_: + drupal + holding checkouts of OE + rsync + read only cvs Jun 02 19:31:34 ggilbert: we could have added monontone, we even have an extra server, when the power upgrades are done at osu, there will be 6 new servers. Jun 02 19:32:09 ggilbert: but it is ok, it is their project, they decided, now it is clean up time. Jun 02 19:32:26 france_: we have no plans to switch the mailing list Jun 02 19:32:38 france_: it would be nice if it could be hosted on hh.org for some time Jun 02 19:33:00 france, osu is doing the hosting? Jun 02 19:33:04 zecke: no worries... Jun 02 19:33:05 Those guys seem really freaking cool Jun 02 19:33:24 zecke: about the mailing list, just let me know...so I can clean things up on my end. Jun 02 19:33:38 ggilbert: 1/2 of the servers are hosted at MIT, the other 1/2 as OSU Jun 02 19:33:49 Seems like they host a ton of stuff Jun 02 19:33:55 just in general Jun 02 19:34:39 they do, they are very good people Jun 02 19:35:10 proj_oe:x:5094:ar,andyc,anoncvs,ayer,bavery,cterry,duffekn,france,gitars,hrw,jamey,jg,kergoth,mickeyl,mithro,noda,pb,zecke Jun 02 19:35:34 jg? really? Jun 02 19:35:50 france, you can probably prune my account. If it hasn't been done already Jun 02 19:37:25 ggilbert: your account is not effected as you are in proj_opie :-D Jun 02 19:37:29 ah Jun 02 19:37:33 ggilbert: but not in proj_oe Jun 02 19:37:41 Well I don't actually do any opie stuff either any more :) Jun 02 19:40:10 it is up to you, if you want it removed send a note to admin@handhelds.org, if you want it to remain that is OK was well, just in case u want to do something Jun 02 19:41:31 ok. I'm fine with it staying. Just wasn't sure if you wanted to trim down unused stuff Jun 02 19:42:04 ggilbert: nope, just cleaning up one project... Jun 02 19:42:20 maybe at some point my interest in open source devices will come back to life. For now the treo just does exactly what I've wanted done :) Jun 02 19:42:37 :-D Jun 02 19:43:04 well not exactly, but the closest approximate out there Jun 02 19:46:17 has anything happened in the past few years on the topic of really low power modes and not just suspend? Jun 02 19:47:04 * chouimat is away: tv and pub and a lot of beers Jun 02 19:47:13 how low? Jun 02 19:47:38 well low enough that you could leave the device running, but in it all day Jun 02 19:47:41 with cpu scaling on 2.6/akita i can get > 12hr uptime Jun 02 19:47:46 instead of flat out suspending Jun 02 19:48:21 ggilbert: underclock your cpu; any kernel that supports it, should help Jun 02 19:48:38 Similar to what a palm does. Just as an example, it can happily suspend and maintain a working imap connection for most of the day Jun 02 19:48:47 dynamic scaling is new with 2.6; its experimental, and iirc hasn't made it into OZ yet Jun 02 19:49:22 oh. that's a little different. Jun 02 19:49:40 I have no idea how they actually implement it :p Jun 02 19:49:43 actually, I'm not sure what the Palm does Jun 02 19:50:03 Just wondering if anyone has put thought into something similar from a user point of view Jun 02 19:50:06 but with scaling, you can let your machine run at ~ 100mhz or less, until something interesting happens Jun 02 19:50:18 and then when interesting things happen, it scales back up to full speed. Jun 02 19:50:31 cool deal Jun 02 19:50:34 or even beyond that to overclock speed, if things are *really* interesting Jun 02 19:51:16 power usage always worked well enough for me before Jun 02 19:51:26 power usage? Jun 02 19:51:28 it just wasn't very smart when I put it to sleep Jun 02 19:51:34 hrm. Jun 02 19:51:42 what device are you using? Jun 02 19:51:49 battery life, however you want to phrase it Jun 02 19:52:03 raduga: Hmmm. Boxer I think. Jun 02 19:52:16 I haven't actually turned it on in a year or two Jun 02 19:53:01 i have no idea what state cpu scaling is, on boxer Jun 02 19:53:15 might work right-out-of-the-box, might not be possible Jun 02 19:53:24 hehe Jun 02 19:53:38 of course with boxer I wouldn't care since it isn't going to have a network connection anyways Jun 02 19:53:42 look at: http://www.rpsys.net/openzaurus/patches/ Jun 02 19:53:47 see if anything catches your fancy Jun 02 19:55:07 ggilbert: see what patches OE puts in boxer by default. it looks like scaling is fully supported on pxa255 Jun 02 20:21:49 ok, cterry, mithro and noa whould vanish... will send e-mail Jun 02 20:25:03 france_: mithro once up on a time changed the website Jun 02 20:30:23 zecke: maybe someone will request that his account be added to another project? Jun 02 20:30:44 whois cterry and noa? Jun 02 20:31:38 no(d)a was some other website guy, I think Jun 02 20:31:49 might have been the openzaurus webmaster Jun 02 20:31:57 Yeah he was Jun 02 20:31:57 ah oky Jun 02 20:32:14 couldn't we get the real Yoda? Jun 02 20:32:17 that is a shame Jun 02 20:32:22 cterry was the guy who did the website briefly after noda, wasn't he? Jun 02 20:32:28 ah yeah, that's right Jun 02 20:32:34 very briefly though, iirc Jun 02 20:32:38 for the brief diversion into plone Jun 02 20:34:30 he burned quickly and was gone Jun 02 20:53:02 re Jun 02 20:54:46 hi ppl, is there some way in bitbake to make it believe a dep has already been built? Jun 02 20:55:26 thejapa: ASSUME_PROVIDED Jun 02 20:55:30 ASSUME_PROVIDES_ Jun 02 20:55:30 thejapa: see the BitBake manual Jun 02 20:55:33 ahh, and what about mithro? Jun 02 20:55:43 yeah, had just opened the page. Jun 02 20:55:51 france_: is he in any other group? Jun 02 20:56:11 thanks for the quick answer, and the rtfm, I get lost sometimes. Jun 02 20:56:19 france_: does he have anything in his home directory? Jun 02 20:56:24 zecke: cterry, mithro and noda are only in the proj_oe group..maybe someone will request they be added to another project, so they can keep there accounts. Jun 02 20:57:13 zecke: mithro no...node yes...cterry yes... Jun 02 20:57:56 s/there/their/ the typing is going bad, time to take a break Jun 02 20:58:26 heh, i dnot remember teh last time i saw noda or cterry Jun 02 20:58:30 ~seen noda Jun 02 20:58:33 noda was last seen on IRC in channel #debian, 402d 23h 26m 19s ago, saying: 'ok thanks'. Jun 02 20:58:41 ~seen cterry Jun 02 20:58:43 cterry <~cterry@64-126-91-238-dhcp-kc.everestkc.net> was last seen on IRC in channel #oe, 495d 3h 54m 31s ago, saying: 'to bad...'. Jun 02 20:58:46 hehe Jun 02 20:58:49 wow :-D Jun 02 20:59:08 mithro is lurking Jun 02 20:59:12 can be removed :} Jun 02 20:59:20 ~seen cterry Jun 02 20:59:22 cterry <~cterry@64-126-91-238-dhcp-kc.everestkc.net> was last seen on IRC in channel #oe, 495d 3h 55m 10s ago, saying: 'to bad...'. Jun 02 20:59:46 thejapa: yeah, but he has no data in his home dir Jun 02 21:01:11 ~seen mithro Jun 02 21:01:14 mithro is currently on #oe (8h 23m 13s) #nslu2-linux (8h 23m 13s) #openzaurus (8h 23m 13s) #handhelds.org (8h 23m 13s), last said: 'it's slow however?'. Jun 02 21:01:20 france_: he is online Jun 02 21:02:57 afaik, mithro applied for SoC too. Jun 02 21:03:04 but didn't get accepted Jun 02 21:03:12 have no idea what his project was Jun 02 21:03:16 or is Jun 02 21:03:25 BitBake C Parser and Unit Testing Jun 02 21:03:29 guess it was for hh.org Jun 02 21:04:05 zecke: in the worst case, I have daily backups for /home Jun 02 21:04:27 zecke: but it would be better / easier if they got there data sooner than later.. Jun 02 21:04:30 the most disruptive thing i could think of would be his email address Jun 02 21:04:39 i'll go to class. see ya. Jun 02 21:05:05 france_: do they have a .forward Jun 02 21:05:36 zecke: they all have postalias entries Jun 02 21:06:02 I have not seen mithro using a AThh.org address Jun 02 21:06:30 france_: he is using thousandparsec.net most of the time Jun 02 21:07:42 zecke: ;-D that is good to hear, I hope that it does not cause them any issues... Jun 02 21:11:52 france_: I will try hard to get an announcement in a reasonable time :} Jun 02 21:12:55 zecke: no worries...it probably does not matter,. as nobody has noticed.. Jun 02 21:25:57 * cbrake discovers package-index :-) Jun 02 21:32:05 zecke: do you know if there is an oe image with the 2.6.x kernel for the ipaq h55xx series ? Jun 02 21:44:35 CSMan: sorry, I don't know Jun 02 21:46:02 i reinstalled wince to test the SD card Jun 02 21:46:11 and it's working without any problem Jun 02 21:46:49 btw, do you guys know how to transfer files from linux to wince ? Jun 02 21:46:57 using the serial port that is. Jun 02 21:52:50 umm... zmodem? Jun 02 21:53:01 yeah, but howto Jun 02 21:53:33 any perl dude around? Jun 02 21:55:04 zecke: I use perl - but I'm not wizard Jun 02 21:55:09 a wizard even Jun 02 21:55:13 you probably know mor then me Jun 02 21:55:18 CosmicPenguin: How can I check if a key is in the dict? Jun 02 21:55:21 lol no Jun 02 21:55:35 umm... seesh, I should know that Jun 02 21:55:49 I tried Jun 02 21:56:00 exists Jun 02 21:56:04 $moo{$var} or $moo{$var} = insert Jun 02 21:56:05 ah Jun 02 21:56:11 exists Dict, value? Jun 02 21:56:14 if exists $hash{$key} Jun 02 21:56:20 ah Jun 02 21:56:31 Returns true if the key exists, even if the corresponding value is undefined Jun 02 21:56:44 and inserting? Jun 02 21:56:46 using push? Jun 02 21:57:30 or just define it Jun 02 21:58:01 oh okay Jun 02 22:00:34 how do I convert from String to Int Jun 02 22:51:21 good night Jun 02 23:46:16 anyone got a monotone 0.25 for AMD64? Jun 02 23:46:46 .deb package I would like. Jun 03 00:09:34 all: Can someone add "autoconf" to http://www.openembedded.org/wiki/RequiredSoftware Jun 03 00:12:20 what is autoconf required for? It should be build as a native package Jun 03 00:12:25 likewise: yes, you can Jun 03 00:12:49 Oh, though I needed a login request with a dev first Jun 03 00:13:22 CosmicPenguin: it is not. I just started a fresh build on a fresh host, and got the error autoconf was not available. Jun 03 00:13:27 what package? Jun 03 00:14:33 CosmicPenguin: package automake-native-1.9.3-r5 Jun 03 00:14:54 CosmicPenguin: but I am not fully up sync with HEAD. Jun 03 00:14:59 didn't somebody hit that recently? Jun 03 00:15:26 CosmicPenguin: sorry for the noise, should have tested against HEAD first. Jun 03 00:15:33 it sounds familiar Jun 03 00:15:35 yay, libtool-native builds for me again. Jun 03 00:17:27 likewise: from automake-native.inc: Jun 03 00:17:28 DEPENDS = "autoconf-native" Jun 03 00:17:49 so unless you have autoconf-native in ASSUMED_PROVIDED, then there is something wonky with your depends - are you sure you have bitbake 1.4.2? Jun 03 00:19:39 yes Jun 03 00:19:46 CosmicPenguin: theres a bug right now. 'bitbake bootstrap-image' for example doesnt end up building autoconf-native before automake-native, despite the latter needing the former, and having it in the deps. Jun 03 00:19:51 * kergoth got bit by it yesterday Jun 03 00:19:58 Thats where I remember that from Jun 03 00:19:58 had to 'bitbake autoconf-native bootstrap-image' Jun 03 00:20:21 i think its a bug in bitbake itself Jun 03 00:20:34 well, it has to be. violating a DEPENDS is bad, mmk Jun 03 00:21:44 kergoth: yes I think so too, my org.oe.dev checkout is exactly the same and has been working before, only bitbake went from 1.4 to 1.4.2 Jun 03 00:22:13 kergoth: your workaround seems to work. Jun 03 00:22:19 its ugly, but at least you can build Jun 03 00:22:23 heh Jun 03 00:23:10 We've got the right people on the case Jun 03 00:23:44 Not only that, also the right people at IRC :-) Jun 03 00:24:40 Thanks. Let's see where the build goes. Jun 03 00:40:01 Good nite all and thanks for all the fish Jun 03 02:48:31 ~seen hrw Jun 03 02:48:39 /~seen hrw Jun 03 02:48:43 hrw was last seen on IRC in channel #oe, 1d 8h 12m 26s ago, saying: 'morning'. Jun 03 02:51:13 ^help Jun 03 02:51:17 ~help Jun 03 02:52:16 wtf is ASSUME_PROVIDED Jun 03 02:52:29 ~wtf is ASSUME_PROVIDED Jun 03 02:52:31 usage: wtf . Jun 03 02:52:45 ~wtf ASSUME_PROVIDED Jun 03 02:52:47 Gee... I don't know what ASSUME_PROVIDED means... Jun 03 02:53:03 ~karma Jun 03 02:53:03 thejapa has neutral karma Jun 03 02:57:31 * thejapa chuckles to himself Jun 03 02:59:44 ASSUME_PROVIDED tells OE not to build something - it assumes it has been made available via some other method... **** ENDING LOGGING AT Sat Jun 03 02:59:56 2006