**** BEGIN LOGGING AT Tue Nov 29 02:59:56 2005 Nov 29 03:03:17 re Nov 29 03:04:12 hi,hrw Nov 29 03:04:31 I need to build a x86, what shall i do ? Nov 29 03:05:03 hrw:I think to set several keyt items in local.conf . would you give directions to me? Nov 29 03:05:16 hrw: =>MACHINE = "x86" | TARGET_ARCH = "i386" | DISTRO = "openzaurus-3.5.4" | KERNEL_VERSION = "2.6" Nov 29 03:08:59 freethinker: kernel_Version is not needed - x86 use linux-x86 which is 2.6.xx Nov 29 03:09:46 hrw: whether need to reserve these following items , rewrite or remove them? Nov 29 03:09:53 hrw:ASSUME_PROVIDED = "virtual/arm-linux-gcc-2.95" Nov 29 03:09:53 PREFERRED_PROVIDERS = "virtual/qte:qte virtual/libqpe:libqpe-opie" Nov 29 03:09:53 PREFERRED_PROVIDERS += " virtual/libsdl:libsdl-qpe" Nov 29 03:09:53 PREFERRED_PROVIDERS += " virtual/${TARGET_PREFIX}gcc-initial:gcc-cross-initial" Nov 29 03:09:53 PREFERRED_PROVIDERS += " virtual/${TARGET_PREFIX}gcc:gcc-cross" Nov 29 03:09:54 PREFERRED_PROVIDERS += " virtual/${TARGET_PREFIX}g++:gcc-cross" Nov 29 03:10:22 freethinker: ever heard about pastebin or privmsg? dont flood channel Nov 29 03:12:19 hrw:oh,sorry I dont know that. Nov 29 03:14:20 hrw:would you give any hint to me about PREFERRED_PROVIDERS ? Nov 29 03:15:04 leave as they are, remove sdl/qpe ones (you rather dont need them) Nov 29 03:15:48 ~lart kghostview Nov 29 03:15:49 * ibot blasts kghostview with a huge firehose then strangles kghostview with it Nov 29 03:16:59 ~curse debian transitions Nov 29 03:17:00 May you be reincarnated as a Windows XP administrator, debian transitions ! Nov 29 03:18:55 argh.. printing 9 page PDF can be pain... Nov 29 03:23:24 hrw:during building glibc_2.3.5+cvs20050627 , it spawn a error " undefined reference to `__GI___finitel" Nov 29 03:24:20 freethinker: I never built for x86 - my arm builds works with that glibc Nov 29 03:25:14 hrw:thanks for your bearing. Nov 29 03:27:04 hi,koen Nov 29 03:27:16 freethinker: I'm not familiar with toolchain fixing so when I have them I usually try other version of failed package. Checking which one works on my hardware is nearly impossibl Nov 29 03:28:51 hi Cwiiis[away] Nov 29 03:29:21 Hey hrw|work Nov 29 03:29:38 hrw: who are familiar about toolchain in our channel? Nov 29 03:31:47 hmm. good question Nov 29 03:32:10 brb Nov 29 03:41:01 re Nov 29 03:42:11 <_law_> has anyone tried to build gnokii 0.6.10? Nov 29 03:42:20 <_law_> doesnt work for me Nov 29 03:45:43 * XorA successfully ppps via rfcomm on both his personal and work phones Nov 29 03:49:39 damnation, its too early for me, wrong channel sorry Nov 29 03:51:30 _law_: 0.6.10? in OE we have 0.6.4 Nov 29 03:52:03 <_law_> hrw|work, i know but i need 0.6.10 for gnome-phone-manager :-( Nov 29 03:52:21 <_law_> gnome-bluetooth already works :-) Nov 29 03:53:52 _law_: I see 0.6.9 on their homepage Nov 29 03:54:18 0.6.10 is in their download area Nov 29 03:54:52 <_law_> hrw|work, but 0.6.5 also doesnt work because nk6510.c is added :-) Nov 29 03:57:20 * _law_ wants to send sms messages via gnome-phone-manager Nov 29 04:00:07 bye Nov 29 05:11:33 * Geo_KM is away: Away at the moment Nov 29 05:23:21 slighlty offtopic, but does anybody know of a jtag programmer design and software that runs on linux? I am looking to develop a system wiht a arm7 TDMI mcu and th eonly prgrammers i can find would require booing winbloze to prgram the device which would be a whole pain in de #!? Nov 29 05:26:29 dan2003: try http://mnementh.co.uk/eseries/debricking.html Nov 29 05:32:48 03florian 07org.oe.dev * r54a8367a... 10/: Change maintainer to myself. Nov 29 05:32:57 03florian 07org.oe.dev * r270dcdae... 10/: gpe-contacts: Add maintainer. Nov 29 05:47:33 * mjc runs Nov 29 05:49:23 re Nov 29 05:57:09 <_law_> could it be that enum has an limit with the arm compiler? Nov 29 05:57:38 _law_: what limit? Nov 29 05:57:43 _law_: in terms of values? Nov 29 05:58:42 <_law_> zecke, yes Nov 29 06:00:44 whats the password for cvs on handhelds? im trying to get debrick Nov 29 06:01:32 _law_: no ;) Nov 29 06:01:43 _law_: what error do you see? Nov 29 06:03:24 <_law_> zecke, hope you understand it :-) http://pastebin.com/442141 Nov 29 06:04:37 someone know where I can find CSR firmwares for BT dongles? Nov 29 06:06:11 dan2003: anoncvs:anoncvs Nov 29 06:06:53 hrw|work: well are you confident you can flash it Nov 29 06:07:57 _law_: nk6510.c:46:26: gnokii/data1.h: No such file or directory Nov 29 06:08:40 zecke: I did it Nov 29 06:08:42 <_law_> koen, this was only a test from me :-) Nov 29 06:08:54 zecke: http://www.hrw.one.pl/2005/11/29/another-device-flashed/ Nov 29 06:09:39 <_law_> so why gnokii doesnt break on GN_OP_Init ? Nov 29 06:10:14 _law_: did you fix the error with the header? Nov 29 06:10:20 _law_: what is the order of the enum and switch Nov 29 06:14:26 koen: is set-default necessary for pull or only for push or also for pull (reading "Updating" paragraph in http://oe.handhelds.org/cgi-bin/moin.cgi/MonotonePhraseBook Nov 29 06:14:58 <_law_> zecke, yes header problem is fixed Nov 29 06:15:49 _law_: post your error again please ;) Nov 29 06:16:03 obergix[work]: iirc it sets them for both, but I'm not sure Nov 29 06:16:16 mickey|zzZZzz: " ".join( map(lambda x: "qtopia-pda-lib"+x, ["abc","def"]) Nov 29 06:16:25 mickey|zzZZzz: dunno if that works but python rockzz Nov 29 06:16:34 <_law_> zecke, http://pastebin.com/442141 Nov 29 06:17:57 <_law_> zecke, here is the new full http://pastebin.com/442151 Nov 29 06:18:56 obergix[work]: I just fish out one big cmdline out of my .bash_history everytime :) Nov 29 06:19:24 koen: hehe Nov 29 06:19:28 _law_: NOTE: package gnokii-0.6.10-r0: task do_build: completed Nov 29 06:20:02 <_law_> hrw|work, could you please post the .bb file? Nov 29 06:20:16 _law_: cp gnokii_0.6.4.bb gnokii_0.6.10.bb Nov 29 06:20:29 hrw|work: Does the tdk dongle use an CSR chipset? Nov 29 06:20:36 <_law_> hrw|work, doesnt work for me :-( Nov 29 06:20:47 _law_: but one fix to makefile.global was needed - otherwise it does not find bluetooth lib Nov 29 06:20:47 <_law_> hrw|work, i have done the same Nov 29 06:21:02 do13: yes - its bluecore1 chip Nov 29 06:21:29 <_law_> hrw|work, perhabs you could post this :-) Nov 29 06:21:46 _law_: did you preprocess the file yet? Nov 29 06:21:47 <_law_> hrw|work, gcc3 or gcc4 Nov 29 06:21:54 _law_: 50 LIBS = -L/home/hrw/devel/build/0915/tmp/staging/arm-linux/lib -lbluetooth Nov 29 06:21:58 _law_: gcc3.4.4 Nov 29 06:22:02 s/-c/-E Nov 29 06:22:13 _law_: gcc 4.0.x (Debian sid) also build it Nov 29 06:22:17 _law_: and then search for the string it complains about? Nov 29 06:23:26 koen: looks like --set-default is not about the branch... Nov 29 06:23:35 hmm, ok Nov 29 06:23:57 do13: coworker have bluecore3 dongle - thats small.. (when compared to my tdk) Nov 29 06:24:28 koen: so I'm afraid I don't understand why I don't get it ... monotone status gives old_revision [b5879d9b74f746b0e59ad029313b62d0c71db8fd] Nov 29 06:24:53 hrw|work: nice. Which tool did you used for the update? Nov 29 06:25:04 maybe I'm not on the same server and your's and mine not synchronised (the purpose of p2p in monotone is to avoid that ?) Nov 29 06:25:17 do13: windows one which was added to older tdk drivers Nov 29 06:25:29 do13: newest drivers does not have that util and firmware files Nov 29 06:30:11 koen: ok, I think I see : database: default-include-pattern org.openembedded.dev ... should be the branch 083 ... ok... Nov 29 06:33:13 koen: will try with monotone set database default-include-pattern org.openembedded.oz354fam083 and see if it helps... Nov 29 06:38:44 koen: that worked ! ... hurray ;) ... I'll add a note to the wiki, I think... Nov 29 06:40:55 * _law_ is rebuilding from scratch Nov 29 06:47:41 koen: updated wiki : http://oe.handhelds.org/cgi-bin/moin.cgi/MonotonePhraseBook?action=diff ... any comments ? Nov 29 06:51:20 obergix[work]: looks pretty good Nov 29 06:59:52 * france is away: Away Nov 29 07:28:56 ~lart TT Nov 29 07:28:57 * ibot puts TT into a headlock and administers a mighty noogie, rubbing half of TT's hair of Nov 29 07:31:05 lol QTopia VoiIP == KPhone Nov 29 07:37:33 zecke: you are reading heise newsticker? Nov 29 07:38:06 do13: well yes, but I knew that before Nov 29 07:38:12 do13: it smells like a GPL violation Nov 29 07:40:38 zecke: not if they got copy on other license from authors Nov 29 07:51:52 hrw|work: I'm not sure if it is one ;) Nov 29 07:52:00 hrw|work: it looks strange though ;) Nov 29 07:52:09 anyway a 55mb tar ball... Nov 29 07:54:58 zecke: what do you use for pasting matlab/maple code into latex? Nov 29 08:29:19 hi, i'm trying to build OE from source for an x86 target from debian. Nov 29 08:29:39 when building glibc from te target task-bootstrap, i get these erors Nov 29 08:29:58 | ../ports/sysdeps/unix/sysv/i386/signal.S: Assembler messages: Nov 29 08:30:08 ../ports/sysdeps/unix/sysv/i386/signal.S:31: Error: open CFI at the end of file; missing .cfi_endproc directive Nov 29 08:35:52 03koen 07org.oe.dev * r43990b9d... 10/: opensync: remove old opensync package and add libopensync 0.18 Nov 29 08:35:57 03koen 07org.oe.dev * r46543b49... 10/: multisync: add 0.90.18 Nov 29 08:46:09 'lo all Nov 29 08:47:00 morning ;) Nov 29 08:47:14 ~seen RP Nov 29 08:47:17 rp is currently on #gpe #oe #openzaurus #handhelds.org. Has said a total of 3 messages. Is idling for 6h 1m 18s Nov 29 08:47:28 RP: ? Nov 29 08:48:10 hi greentux Nov 29 08:48:15 hi TyggerBob Nov 29 08:48:18 hrw|work: hi ;) Nov 29 08:48:29 hrw|work, Nov 29 08:52:23 hrw|work: your sd card is on the way to me... Nov 29 08:52:57 greentux: cool Nov 29 08:53:52 so sooner or later I will more / to sd Nov 29 08:54:14 03koen 07org.oe.dev * rd0ba6dfe... 10/: libopensync-plugin-file: add 0.18 Nov 29 08:57:22 hi greentux Nov 29 08:57:31 hir RP... good news Nov 29 08:58:03 rp: the contract is signed ;) you will recieve it in a week or so Nov 29 08:58:21 greentux: excellent :) Nov 29 08:58:59 day of good news ;) any news from the kernel front? Nov 29 08:59:30 greentux: We have a new led class and led trogger code so the leds work on spitz now Nov 29 08:59:53 and on borzoi ... :) Nov 29 09:00:21 yes, and akita and c7x0 and on anything else someone writes a driver for (which is really easy) Nov 29 09:01:13 fine. what do you think about powersafe? is it very complicated? Nov 29 09:03:44 greentux: I'm not too familar with the userspace implmentations. As far as the kernel goes, we need to write a cpufreq driver for the pxa27x (or find one if someone's written one) Nov 29 09:04:22 I know the PXA's registers and have looked over the code for the pxa25x so it shouldn't be too difficult to write Nov 29 09:04:28 or dyntick Nov 29 09:04:29 in userspace you need nothing. a kernel cpufreqd would be very nice. Nov 29 09:06:02 I need to look into dyntick and how well it would work on the pxa Nov 29 09:06:19 RP, greentux: FYI, audio bus interface is finished. I'm now trying to get madplay working with the --with-alsa configure switch, although I might soon revert to OSS emulation. Nov 29 09:06:56 lrg|home: borzoi is on the way now Nov 29 09:07:12 dyntick is userspace? Nov 29 09:07:12 greentux: cool :) Nov 29 09:07:58 lrg|home: ok, that sounds good. Once I get the next update, I'll try and finish off the machine based sound code. I suspect your changes to moving the i2c code into the machine file will help me Nov 29 09:08:00 lrg|home: recording also working? mic? but the MIC switch is not easy, isnt it? Nov 29 09:08:31 greentux: We need some platform specific code there Nov 29 09:08:44 RP: yes, I'll even create a template for corgi, etc Nov 29 09:09:05 greentux: not long now Nov 29 09:10:16 http://hrw.one.pl/tmp/cpu1.png http://hrw.one.pl/tmp/cpu2.png - my cpufreq switcher Nov 29 09:10:34 lrg|home: the borzoi is new (org. flash), you can flash waht you want. Nov 29 09:10:39 http://hrw.one.pl/tmp/usb1.png -> usb3.png - usb gadget switcher Nov 29 09:10:43 RP: I'll aim for 0.8 after lunch tomorrow. I'd like to fix madplay first. I've also started suspend/resume. Nov 29 09:11:12 lrg|home: Is it looking like madplay or alsa is at fault? Nov 29 09:11:15 greentux: I'll flash the latest oz from monotone. I have a nightly build set up. Nov 29 09:11:24 lrg|home: ok, that sounds good, thanks :) Nov 29 09:11:35 hrw|work: cpufreq in user space works? Nov 29 09:11:51 greentux: on pxa25x I use kernel support for cpufreq Nov 29 09:12:09 greentux: from userspace I only switch governors and/or set cpu speed Nov 29 09:12:09 RP: I'm not sure yet. Aplay is faultless, however madplay doesn't work without changing some default oiptions on my FC4 pc. Nov 29 09:12:16 pxa25x is in borzoi like plattforms? Nov 29 09:12:35 greentux: borzoi is pxa27x, not pxa25x Nov 29 09:12:47 rp: ok, i see Nov 29 09:12:54 greentux: They're very similar but it needs a different driver Nov 29 09:13:47 dyntick is another kernel idea which works by slowing down an idle system - I'm not sure how it will work with the pxa but will research that Nov 29 09:14:05 There is also the possibility of changing the core cpu voltage on the pxa27x Nov 29 09:14:17 If I understand the Sharp source correctly anyway Nov 29 09:14:21 RP: will you try to use the 33mhz mode of the PXA? Nov 29 09:14:52 zecke: Until I look into it, I'm not sure Nov 29 09:16:37 RP: the issue I see is that i2c and i2s would need to be reinitialized (so no true dyntick is possible) Nov 29 09:18:48 zecke: i2s isn't a problem in that you could just disable the clock changing when audio was active... Nov 29 09:19:19 zecke, RP: or drive the i2s from the codec Nov 29 09:21:49 lrg|home: Have we got the hardware for that on the zaurus devices? Nov 29 09:22:40 RP: not sure, I would need to check schematics. Nov 29 09:23:53 RP: how often would the tick occur as we could just recall hw_params in the driver every time the clock is changed. We would hear artifacts though :( Nov 29 09:25:12 lrg|home: I'll have a look at what the pxa hardware can do... Nov 29 09:25:42 RP: np, its an interesting problem. Nov 29 09:29:02 hrw|work: have you looked at cpufreqd, I've not got round to trying it yet Nov 29 09:31:34 lardman|work: what for I would use cpufreqd if kernel part works nice for me? Nov 29 09:32:13 hrw|work: cpufreqd interfaces with the userspace governor to actively control the CPU speed; not sure how much better than ondemand it is, but I think it can have more complex rules, etc. Nov 29 09:32:24 echo "1" > /sys/devices/system/timer/timer0/dyn_tick looks interesting (on a recent Zaurus kernel) Nov 29 09:32:53 hrw|work: it can take account of external power, etc., which ondemand doesn't (nor should it) Nov 29 09:32:59 RP: How recent is that? Nov 29 09:33:19 lardman|work: Since I made the defconfig changes suggested by hrw... Nov 29 09:33:25 RP: ah, cool Nov 29 09:33:41 rp: what means "looks interesting" :) i have no Z at the moment Nov 29 09:34:21 greentux: In that it suggests it enables dyn_tick. It should save us some power and if I understand what it does, shouldn't cost us anything Nov 29 09:34:27 lardman|work: ah.. so 'if AC then fullpower' thing? Nov 29 09:35:22 hrw|work: exactly Nov 29 09:35:36 lardman|work: sounds nice Nov 29 09:35:44 hrw|work: Yep Nov 29 09:36:27 c7x0 is hard to use at 100MHz Nov 29 09:36:37 yeah, terrible wait Nov 29 09:36:56 dyn_tick looks to just work on the zaurus. It just suspends the processor core for a while and the peripheral clocks keep running so we don't have aynone of the frequency issues Nov 29 09:37:09 that cpufreq has Nov 29 09:37:22 s/aynone/any/ Nov 29 09:37:45 2.6.14 should have integrated pxa255 dyntick Nov 29 09:38:05 so I guess it's pretty usuable Nov 29 09:38:10 koen: It has Nov 29 09:42:21 I'll enable dyntick by default in future kernels via the cmdline Nov 29 09:42:58 nice Nov 29 09:43:09 no we needs some benchmarks with battery life Nov 29 09:43:41 I want to install some SW in say /home/scaradio Nov 29 09:43:46 s/no/now/ Nov 29 09:43:52 Benchmarks would be interesting... Nov 29 09:43:56 and am thinking of messing with default prefix on configure.ac Nov 29 09:44:13 but OE will break this approach? Nov 29 09:44:16 Although I really want to see 2.4 vs 2.6 :) Nov 29 09:44:48 ondemand + dyntick vs 2.4.18-crappix Nov 29 09:45:20 Do we want to be running something occasionally to force ondemand to do something though? Nov 29 09:46:37 What we need is a device on an ammeter put under different workloads and comparing the readings... Nov 29 09:46:52 including the readings under a suspend Nov 29 09:48:00 I've got the hardware, just need to build the right cables and do it... Nov 29 09:48:42 you have a multimeter with rs232/other interface? Nov 29 09:49:04 RP: I'll join in the fun when I get my borzoi :) Nov 29 09:49:12 koen: I have a multimeter - automated readings aren't that useful for this... Nov 29 09:49:30 is it possible to change to version of glibc that oe builds? Nov 29 09:49:53 RP: depends how long your battery's going to last hey? Nov 29 09:50:04 brendan_: PREFERRED_VERSION Nov 29 09:50:04 PREFERRED_VERSION_glibc="the one you want" Nov 29 09:50:20 in local.conf? Nov 29 09:50:27 yes Nov 29 09:50:33 lardman|work: Not really, you just need a stable current reading in each state - you'd remove the battery... Nov 29 09:50:35 thanks Nov 29 09:50:39 does that work for any package? Nov 29 09:50:45 RP: oh right, I see Nov 29 09:50:52 brendan_: it should Nov 29 09:51:06 see conf/distro/preferred*.conf Nov 29 10:02:00 RP: lowest freq + dyntick + lcd off vs suspened would be a nice measurement too Nov 29 10:07:47 koen: yes Nov 29 10:08:17 according to the nokia people it should be (almost) the same Nov 29 10:10:18 nice Nov 29 10:10:48 and will give us possibility to drop apm at all (so my dtl1_cs card will work) Nov 29 10:10:59 03koen 07org.oe.dev * rf941aea6... 10/: libopensync-plugin-gpe: add 0.18 Nov 29 10:15:56 Talking about battery readings, are the apm scripts working? They weren't last time I tried to use them (to do battery readings) Nov 29 10:19:26 lardman|work: There are lots of subtle bugs with apm Nov 29 10:19:30 they dont work Nov 29 10:20:00 koen: I suspect with the zaurus, suspend power consumption will be better as a lot of hardware is physically powered down Nov 29 10:20:26 The zaurus wasn't designed to run with dyn_tick and doesn't have the power management chips the N devices no doubt have Nov 29 10:20:37 hrw|work: good, glad it's not just my ineptitude Nov 29 10:20:48 the 770 just powers down bt and wifi when the connection isn't used Nov 29 10:21:13 RP: There's nothing to stop us from powering down peripherals as a matter of course though is there? Nov 29 10:21:17 koen: says who? Nov 29 10:21:53 RP: wlancond and btcond :) Nov 29 10:22:23 lardman|work: Linux doesn't yet have software support for powering down individual devices... Not that I'm aware of much other than the sound which would benefit (and Liam has this in hand) Nov 29 10:22:36 man.. I wish the h5xxx's did that Nov 29 10:22:47 koen: I'm suggesting it contains hardware which also helps... Nov 29 10:31:22 03koen 07org.oe.dev * r099a4507... 10/: libopensync 0.18: fix DEPENDS Nov 29 10:33:03 Okay, another question, does the bluetooth & wifi work with the gpe image (on Nokia 770)? Nov 29 10:33:12 oops wrong channel Nov 29 10:36:39 <_law_> hrw|work: read (http://www.hrw.one.pl/2005/10/31/my-openembedded-enviroment/#more-75)and setup my env new but where are the local.confs placed? Nov 29 10:37:35 _law_: builddir/conf/ Nov 29 10:37:48 night Nov 29 10:37:58 03koen 07org.oe.dev * r06734878... 10/: libopensync-plugin-irmc: add 0.18 Nov 29 10:38:02 _law_: for example build/oz353/conf/ Nov 29 10:38:18 <_law_> hrw|work: thanks Nov 29 10:41:02 Hum... syncing to monotone.vanille.de: Nov 29 10:41:03 monotone: bytes in | bytes out | revs in | revs out | revs written Nov 29 10:41:03 monotone: 100.6 k | 721.5 k | 1 | 188 | 1 Nov 29 10:41:10 I don't like that 'revs out' Nov 29 10:41:48 Still, I think it is managing to exclude them. Nov 29 10:49:04 heh.. so much work lately that I dont even have time to update Z.. Nov 29 10:51:28 <_law_> hrw|work: and you build stuff into build/oz353/tmp? Nov 29 10:51:38 _law_: yes Nov 29 10:52:19 <_law_> hrw|work: TMPDIR = build/oz353/tmp in local.conf Nov 29 10:53:18 <_law_> hrw|work: and how to you handle oe/hrw-packages/ ? Nov 29 10:53:53 <_law_> is this an copy of org.openembedded.dev? Nov 29 10:54:13 <_law_> with your xtra packages? Nov 29 10:54:14 _law_: I dont set TMPDIR because I'm in build/oz353/ when building. Nov 29 10:54:45 _law_: hrw-pacakges are extras. Nov 29 10:54:47 PKGDIR = "/home/hrw/devel/oe/org.openembedded/packages" Nov 29 10:54:47 HRWPKGDIR = "/home/hrw/devel/oe/hrw-packages" Nov 29 10:54:48 BBFILES = "${PKGDIR}/*/*.bb ${HRWPKGDIR}/*/*.bb" Nov 29 10:56:56 <_law_> hrw|work: works realy nice :-) Nov 29 11:00:26 I know Nov 29 11:01:37 koen: do you see multiple heads in your monontone db (for org.openembedded.dev?) Nov 29 11:02:28 where are the defualt vars like prefix set again??? Nov 29 11:02:45 bitbake.conf? Nov 29 11:03:01 thanks Nov 29 11:03:06 senior moment Nov 29 11:03:11 monotone: misuse: branch 'org.openembedded.dev' is merged Nov 29 11:03:40 thought so - it looks like you don't have the problem revisions Nov 29 11:04:15 For example, you probably don't have ef52fb8de8658c30581cea89db5730a944f8a8bc Nov 29 11:05:08 03koen 07org.oe.dev * rc125c820... 10/: libopensync-plugin-irmc: fix DEPENDS, thanks to OE shlibs code Nov 29 11:05:57 _law_: in hrw-pacakges I have unfinished/broken stuff mainly Nov 29 11:06:27 <_law_> hrw|work: i?ll do the same with law-packages :-) Nov 29 11:07:50 <_law_> hrw|work: is it possible to rename /build/test to /build/husky_gpe ? Nov 29 11:08:38 <_law_> or is there any hardcoded stuff? Nov 29 11:09:18 _law_: it is not possible - you have to rebuild all if renamed Nov 29 11:14:23 Well, we'll see: last revision in the OE.db.bz2 snapshot is 44ef2a80f1407b2958e8f750892482362d9bb352, just before the problem happened. Either the next snapshot will have the following revisions (and probably a single head) or it won't... Nov 29 11:16:02 In the OE view of the world (where org.nslu2-linux.dev isn't in the DB) there is only one head, in the nslu2-linux view of the world there are two... Maybe there is no problem here outside nslu2-linux. Nov 29 11:30:50 RP, greentux: madplay alsa playback is broken (even on my PC) with alsa 1.0.9 :( Madplay does now however work with alsa under OSS emulation :) Nov 29 11:35:45 koen, where do the keys get stored when you do a keygen? Nov 29 11:38:42 in the db, 'monotone list keys' Nov 29 11:39:01 'monotone privkey ' should output the privkey Nov 29 11:39:06 pubkey should work analogous Nov 29 11:39:10 ok.. Nov 29 11:39:26 can you export them that way and them import them into a new snapshot of the db? Nov 29 11:46:03 monotone: connecting to monotone.vanille.de Nov 29 11:46:03 monotone: network error: failed to connect: Connection refused Nov 29 11:46:05 poop Nov 29 11:46:57 It's very heavily loaded Nov 29 11:47:51 koen: or just copy from ~/.monotone/keys/ with 0.24+ Nov 29 11:50:14 monotone privkey > file Nov 29 11:50:22 monotone read < file Nov 29 11:51:05 TyggerBob: you can connect to anyone of the backup servers Nov 29 11:57:23 koen, you have a zippy link there.. Nov 29 11:58:14 lrg|home: Ok, as long as I know its madplay, I know how to reply to the bug reports :) Nov 29 11:58:54 TyggerBob: yeah, ewi546 is on the uni, so it has a 100Mbit/s uplink Nov 29 11:59:04 :) Nov 29 12:01:01 hrm.. this doesn't sound good: Nov 29 12:01:03 monotone: error: sqlite error: 1: SQL logic error or missing database Nov 29 12:01:04 monotone: error: make sure database and containing directory are writeable Nov 29 12:03:40 koen: that's using latest snapshot from your OE dir Nov 29 12:04:31 http://ewi546.ewi.utwente.nl/OE/OE.db.bz2 Nov 29 12:04:39 am I using the wrong one? Nov 29 12:05:03 that should be the correct one Nov 29 12:05:08 do the md5's match? Nov 29 12:05:40 http://pastebin.ca/31706 Nov 29 12:05:51 This bb file builds one binary Nov 29 12:05:59 I want it in /home/sca/bin Nov 29 12:06:07 I have to redown load it to check.. Nov 29 12:06:12 but this do_configure doesn't make it happen Nov 29 12:06:13 re-download, that is Nov 29 12:07:10 TyggerBob: just check the MD5. The database is fine (I've been using it for the last few hours). Nov 29 12:08:06 md5 is fine.. Nov 29 12:08:52 re-bunziping and I'll try agian Nov 29 12:10:05 "and then a miracle occured" Nov 29 12:10:17 did the same thing three times.. Nov 29 12:10:19 and now it works.. Nov 29 12:10:31 as it typical for me. Nov 29 12:12:12 lol.. I thought the more current db would have fewer revisions.. Nov 29 12:12:14 not Nov 29 12:27:28 cu Nov 29 12:28:15 hmmm Nov 29 12:30:59 I need to override the $prefix used in oe_runconf Nov 29 12:31:20 oe_runconf gets its vars expanded before run time Nov 29 12:31:48 if I overide prefix outside do_configure, will this fore oe-runconf to use my prefix? Nov 29 12:34:55 yep, it does :) Nov 29 12:35:42 isn't it easier to ipkg -dest /home/foo install bla.ipkg ? Nov 29 12:35:49 03koen 07org.oe.dev * rbf370406... 10/: gpe-todo: add 0.55 and it's hildon counterpart Nov 29 12:36:53 well, if I build an image, can I do that? Nov 29 12:37:09 These aren't packages that will have a wide audience Nov 29 12:38:39 koen: Is there any reason why the busybox wget bug, 1332 has not been marked resolved? The busybox defconfig in OE does have the ipv6 options. Nov 29 12:40:56 let's try this in the right channel. Nov 29 12:41:00 koen: Is there any reason why the busybox wget bug, 1332 has not been marked resolved? The busybox defconfig in OE does have the ipv6 options. Nov 29 12:50:05 uE: bugzilla won't let me Nov 29 12:50:26 koen: http://handhelds.org/~bugzilla/show_bug.cgi?id=1332 Nov 29 12:50:33 I know Nov 29 12:50:44 Oh, you mean it won't let you resolve it. Nov 29 12:50:46 got it. Nov 29 12:50:47 I wanted to close resolved bugs, but hh.org bugzilla won't let me Nov 29 12:51:07 I applied for bugzilla super powers, but noone has acted on that yet Nov 29 12:51:28 hey marceln Nov 29 12:51:37 Hi koen? Nov 29 12:52:01 marceln: if you send your monotone key to koen@handhelds.org and mickeyl@handhelds.org, we will give you access Nov 29 12:52:03 You have spoken with Rod? Nov 29 12:52:11 as long as the key id is @openemembedded.org Nov 29 12:52:14 marceln: yes Nov 29 12:52:15 Ok. Nov 29 12:52:51 Is there a simple introduction to monotome. I have never worked with it? Nov 29 12:53:02 I am used to cvs. Nov 29 12:53:09 hi uE Nov 29 12:53:30 http://oe.handhelds.org/cgi-bin/moin.cgi/MonotonePhraseBook and http://venge.net/monotone/monotone.html Nov 29 12:53:43 Thanks. Nov 29 12:54:50 koen: No, not that I can say. I committed the lifedrive and palm t3 patches to the kernel26 tree. Nov 29 12:55:25 what about hte h3800 patch? Nov 29 12:56:06 koen: The problem with the h3800 patch is that it contains some SoC code and does not patch very cleanly in ALSA. I asked Ahmed to resubmit a patch. Nov 29 12:56:26 ah, ok Nov 29 12:56:45 koen: He responded and said he had some suspend/resume patches to go along with it. I am waiting. But I liked his patch. So I think if he resubmits we should be jake. Nov 29 12:57:39 so pb_ I'm guessing your not going to LCA 2006? Nov 29 12:58:03 Hi pb_, And an official hi to Koen as well. Nov 29 12:59:05 uE: regarding your comment to bug 1332, does familiar actually contain a version of busybox where ipv6 works in wget? We don't usually close those bugs until the package is at least in the unstable feed. Nov 29 12:59:17 mithro: right, probably not Nov 29 12:59:48 pb_: This is tricky part. busybox advertises the applet as workable with ipv6. But I doubt anyone has actually tried it. Nov 29 13:00:19 is it normal for this to sit here for a while: Nov 29 13:00:20 http://ewi546.ewi.utwente.nl/OE/OE.db.bz2 Nov 29 13:00:22 do Nov 29 13:00:25 monotone: bytes in | bytes out | certs in | revs in | revs written Nov 29 13:00:25 monotone: 1.2 M | 197.8 k | 1704 | 212 | 61 Nov 29 13:00:28 doh Nov 29 13:00:48 uE: okay. well, if busybox says it works, I guess that's good enough. Someone can reopen the bug if it still doesn't work. Nov 29 13:03:19 TyggerBob: monotone can sit on the pulling part for quite some time (half an hour can be seen in some situations). Nov 29 13:03:31 Ahhh.. Nov 29 13:03:32 ok. Nov 29 13:03:52 the past couple tries, it seems to sit at 61 for some time Nov 29 13:04:56 uE: well, I closed that bug. let's see what happens. Nov 29 13:08:09 Yay!! 62!! Nov 29 13:08:11 :) Nov 29 13:10:35 dyn_tick makes no difference to power consumption on the c7x0... Nov 29 13:11:15 03jbowler 07org.oe.dev * r10189d45... 10/: Nov 29 13:11:15 ixp4xx-csr: force rebuild, correct error handling in 2.1 Nov 29 13:11:15 - two different revisions were 'r1', this commit makes a single revision Nov 29 13:11:15 'r2' because one 'r1' revision was broken. Nov 29 13:11:15 - modpost error handling should be fixed now - the package should error Nov 29 13:11:15 out if modpost reports unresolved symbols. Nov 29 13:11:23 either dyntick is faulty or the device isn't idle Nov 29 13:12:00 koen: or dyntick isn't a magic cure. Did you see rmk's analysis of it? Nov 29 13:13:37 pb_: Would you know what a "good" or "normal" suspended current draw for a pxa25x based system would be offhand? Nov 29 13:14:18 RP: the one with the bad math? Nov 29 13:15:17 There's no such thing as a "normal" value really; it depends too much on the details of the circuit. But I would expect something on the order of 10mW. Nov 29 13:19:16 fwiw, the pxa255 itself, when correctly configured, draws almost negligible current in sleep mode; it probably dissipates well under a milliwatt. Most of your power budget goes to the RAM. Nov 29 13:25:48 koen: Could have been. His argument seemed logical though... Nov 29 13:26:17 pb_: I'm seeing more like 70mW here :-/ Nov 29 13:27:39 What does your RAM datasheet say for typical power in self-refresh? Nov 29 13:33:50 * chouimat is away: shower Nov 29 13:35:12 pb_: I should be looking at 6mW. The 70 was incorrect, its drawing more like 35mW now Nov 29 13:36:53 Mmm. In that case, I guess you either have a problem with your power supply (i.e. excessive quiescent current through your regulators) or your system is not correctly configured in some way. Nov 29 13:37:34 Just one or two CPU pins set wrong could easily account for your missing 20mW. Nov 29 13:38:39 pb_: Pulling the CF and SD cards in and out when suspended changes the current reading by a small amount which is a bad start... Nov 29 13:38:57 That is indeed a bad start. Nov 29 13:39:20 If you have any volts on any of that stuff during suspend, you almost certainly have a problem. Nov 29 13:40:28 Shame I don't have a circuit diagram for this :-/ Nov 29 13:40:54 Yeah, that is unfortunate. Nov 29 13:49:56 * chouimat is back. Nov 29 14:31:52 * Geo_KM is back Nov 29 14:42:42 cya Nov 29 14:55:58 CSMan: Sorry for the late reply. I was just saying that you can pretty much copy and paste the usb vendor and device ids from the berlios driver. Nov 29 15:25:38 koen, just to wanted to confirm. Nov 29 15:25:47 I've reinitialized my bitbakery... Nov 29 15:26:10 I followed the instructions from the Monontone phrase book. Nov 29 15:26:20 should I be building using .dev? or oz/ Nov 29 15:26:22 ? Nov 29 15:29:30 should I be checking them both out? Nov 29 15:30:04 then, under my ../oe dir, do I create two build dirs? Nov 29 15:30:11 one for dev and one for oz...? Nov 29 15:30:48 anyway if you could leave me an answer here, or in a PM, that would be great.. Nov 29 15:30:55 5:30.. gotta go relieve the nanny Nov 29 15:56:18 TyggerBob: .dev would be ok, as your patches/commits would be in that branch Nov 29 16:41:46 I already have a toolchain for my Hx4700. How do I configure my local.conf to let oe know that I have a cross compiler so it doesnt download and compile it? Nov 29 16:57:46 jt_dnb: its in your best interest to let OE build the compiler Nov 29 17:02:22 ok Nov 29 17:07:54 jt_dnb: But if you want to provide your own compiler then use the ASSUME_PROVIDED variable. Nov 29 17:15:14 jt_dnb: Its use is described in the example local.conf. Nov 29 19:10:42 hey ljp Nov 29 22:47:11 hello Nov 30 01:43:59 morning Nov 30 01:44:04 ~curse debian transitions Nov 30 01:44:05 May you be reincarnated as a Windows XP administrator, debian transitions ! Nov 30 01:47:39 gcc4, xorg, gcc4 again, freetype - soon 'unstable' will be so fscked that upgrading will need solid cherrypicking.. Nov 30 01:48:37 5 versions of KDE are available for debian - none installable Nov 30 01:52:12 hi hrw, you came so early today. Nov 30 01:54:07 freethinker: deadline for website Nov 30 02:07:19 hi Lrg Nov 30 02:07:43 hrw|work: morning Nov 30 02:08:02 Liam is scottish name? Nov 30 02:08:22 iirc it's Irish Nov 30 02:08:55 ah good to know Nov 30 02:09:20 It's short for William Nov 30 02:11:40 argh... I suspect that someone broke cifs/smbfs in 2.6.15-rc Nov 30 02:12:30 no it was lack of coffee Nov 30 02:13:06 hrw|work: That'll do it all right ;) Nov 30 02:15:19 ~lart kwallet Nov 30 02:15:19 * ibot blames kwallet for all the evil in the world Nov 30 02:15:46 fscking kwallet from time to time decide to forgot all passwords.. Nov 30 02:16:52 lovely situation.. I have a tool to remember passwords but that tool decides to forget them so I have to re-login... Nov 30 02:16:56 argh.. Nov 30 02:17:15 Well, makes your passwords secure from being stolen I guess. "The wallet subsytem provides a convenient and secure way to loose all your passwords"... Nov 30 02:18:58 <_law_> hey hrw|work i have got a little problem with local.conf Nov 30 02:19:50 <_law_> BBFILES = "${PKGDIR}/*/*.bb ${LAWPKGDIR}/*/*.bb" doesnt work :-( Nov 30 02:19:59 and? Nov 30 02:20:53 <_law_> PKGDIR = "/home/law/work/programming/zaurus/oe/oe/org.openembedded.dev/packages"LAWPKGDIR = "/home/law/work/programming/zaurus/oe/oe/law-packages" Nov 30 02:21:28 thats two separate lines? Nov 30 02:21:48 <_law_> hrw|work, found the error thanks Nov 30 02:22:07 and it was? Nov 30 02:23:25 <_law_> i have copy and paste the 2 lines yesterday and put the 2 lines into one Nov 30 02:23:38 <_law_> the terminal shows 2 lines Nov 30 02:23:51 <_law_> because it was too small :-) Nov 30 02:24:58 <_law_> now i can build my gnome-bluetooth .bb files :-) Nov 30 02:24:59 thats where ":set number" in vi comes with help Nov 30 02:25:15 <_law_> hrw|work, you are right Nov 30 02:25:25 brb - have to relogin to get kwallet working again Nov 30 02:25:33 ~die kde madafackers Nov 30 02:25:35 * ibot takes two shots to the head, murders kde madafackers then crumples to the ground, lifeless... Nov 30 02:26:37 re Nov 30 02:27:00 <_law_> hrw|work, kde 3.5? Nov 30 02:27:13 _law_: kde at all Nov 30 02:27:22 I have some bad news. We use twice as much power when suspended under 2.6 on the c7x0 than under 2.4 :-( Nov 30 02:27:59 ack Nov 30 02:28:04 RP: ;( Nov 30 02:28:22 kergoth: how goes dns moving for openembedded.org? Nov 30 02:28:33 ? Nov 30 02:28:42 i did that last week. i'd think it would have propogated by now Nov 30 02:28:50 We also use loads more power when idle, when booting or when doing most other things. We've got some improving to do... Nov 30 02:28:52 works here Nov 30 02:29:05 o.. works. sorry Nov 30 02:29:05 RP: Using the same setting of HZ in both? Nov 30 02:29:32 v8jlene: Arm has never changed its HZ value that I know of... Nov 30 02:30:26 RP: Ok. That's the only power related thing I remember people complaing about on lkml. (I don't use ARM myself). **** ENDING LOGGING AT Wed Nov 30 02:59:56 2005