**** BEGIN LOGGING AT Sun Jun 05 23:59:56 2005 Jun 06 00:03:21 good morning all Jun 06 00:43:49 koen: morning Jun 06 00:44:49 hey Luke-Jr Jun 06 00:44:52 moin zecke Jun 06 00:45:54 hi Jun 06 00:46:00 hey hrw|work Jun 06 00:52:24 morsche Jun 06 00:52:55 porsche! Jun 06 00:54:11 koen: now if you would work at Ximian you would have a Porsche by now Jun 06 00:54:17 Hello Jun 06 00:54:21 aaargh Jun 06 00:54:24 hey Securez Jun 06 00:54:50 anybody had used osloader.exe to boot linux from winCE or know any place where i can download it? Jun 06 00:54:56 zecke: but I would be required to say 'mono' every 5 sentences Jun 06 00:55:17 koen: and "anyway" Jun 06 00:55:31 anyway, mono rocks Jun 06 00:55:55 Jun 06 00:56:02 Securez: hh.org? Jun 06 00:56:18 Securez: not haret? Jun 06 00:56:42 yes i read in hh.org that is posible with osloader Jun 06 00:56:54 but i can't find how i can download it Jun 06 00:58:11 morning all Jun 06 00:58:28 hey RP Jun 06 00:59:47 morning Jun 06 01:05:22 now i'm read haret docs, can i boot linux from wince Jun 06 01:08:25 ?? Jun 06 01:09:06 Securez: yes you can Jun 06 01:18:18 anyway maemo must feature Mono Jun 06 01:18:30 Nokia should by the Novell Mono Desktop license... Jun 06 01:19:18 ARM Ltd should license the Mono core technology from Novell to build accelerated and secure Mono executors... Jun 06 01:21:23 hmm.. I have to decide which DE I will use tomorrow.. opie or gpe.. Jun 06 01:21:55 hrw|work: maemo Jun 06 01:22:10 ;))) Jun 06 01:22:35 will it sucks less? Jun 06 01:22:56 ~dict spealer Jun 06 01:22:58 could not find definition for spealer Jun 06 01:23:03 hrw|work: does that imply something currently sucks? Jun 06 01:23:05 ~spell spealer Jun 06 01:23:08 possible spellings for spealer: speller speaker sealer spearer stealer spiller spoiler spooler sepal appealer pearler Peale paler spear specialer speer suppler spellers spell squealer Jun 06 01:23:26 zecke: each DE sucks - some sucks more, some less Jun 06 01:24:05 hrw|work: what particular? Jun 06 01:24:06 its like email clients.. 'all email clients suxx, mutt just suxx less. mutt suxx less, but still suxx' (mutt-ng homepage) Jun 06 01:25:31 zecke: in gpe most of apps is QVGA sized, some things are not changeable without tweaking in files (font in applauncher) - but I like how fast it rotate... but why after rotate taskbar is not on bottom? Jun 06 01:26:48 hrw|work: there an option to pass to matchbox-panel to change that Jun 06 01:27:00 hrw|work: it can do opie style rotate as well Jun 06 01:27:01 opie have more apps now, windows can be clsoed using Cancel button but datebook does not give infos in month view Jun 06 01:27:23 XorA: I just does not find it yet then Jun 06 01:27:47 hrw|work: I played with that on pdaXrom Jun 06 01:28:16 eh.. look&feel still does not like me ;( Jun 06 01:29:52 hrw|work: matchbox-panel --no-flip (If I remeber right) Jun 06 01:30:34 XorA: currently I'm trying to use gpe like user Jun 06 01:30:37 XorA: why are etsi standards written in such a crappy manner? Jun 06 01:31:30 zecke: because no-one is brave enough to change them, and people like nokia block vote sensible changes out as it would mean updating their products/tests Jun 06 01:31:47 n Jun 06 01:32:12 oh and fun for security people, nokia are no longer bounds checking their products, have fun with those overflows :-) Jun 06 01:32:50 XorA: we all know that not security in [ 'nokia', 'symbian'] Jun 06 01:33:38 zecke: they went as far as demanding that ETSI remove all extreme ranges from official ETSI test specs though :-D Jun 06 01:33:39 sadly and ironically Windows Mobile 5 will be/is the most secure mainstream Operating System Jun 06 01:33:54 XorA: lol Jun 06 01:34:12 XorA: I've struggled yesterday implementing the Sim App Toolkit (GSM 11.10) Jun 06 01:35:05 zecke: 11.blah I may know about, I wrote large chuncks of the Sim side of toolkit apps, and the 03.19 Java implementation Jun 06 01:35:32 zecke: though Id have to dredge my memory back for 2 years ago Jun 06 01:35:47 XorA: it was extremely hard to find any meaningfull values... Jun 06 01:36:03 hi Dirk Jun 06 01:36:19 hi hrw|work Jun 06 01:37:21 zecke: 11.10? Jun 06 01:37:55 XorA: yes inside the standard... all the information is so scattered Jun 06 01:38:22 do13: I've found a problem with 2.6.12-rc4-git1 onwards with xscale - rmk has broken it Jun 06 01:38:58 RP: which problem? Jun 06 01:39:36 do13: Random segfaults, floating point errors, invalid instructions and general breakage Jun 06 01:40:06 do13: He removed the preempt locks on copypage which xscale dislikes Jun 06 01:40:22 morning. Jun 06 01:40:26 zecke: youd need 11.14 to make sense of that spec Jun 06 01:40:27 hi Seb Jun 06 01:41:21 RP: Hmm. I recognized an similar problem with 2.6.12-rc5-mm2 (lot's of alignment traps). Jun 06 01:41:40 Hi marcin Jun 06 01:41:46 * zecke thinks he confused numbers again Jun 06 01:42:07 do13: Its the same problem and stems from 2.6.12-rc4-git1 Jun 06 01:42:17 yeah I meant 11.14 Jun 06 01:42:24 wasn't 11.10 for multiplexer stuff? Jun 06 01:42:36 zecke: ah, yeah, that ones confusing, 03.19 clears up some stuff if you understand java Jun 06 01:43:07 XorA: do you know how I can issue a Proactive Command to return to the main menu? Jun 06 01:43:07 do13: http://www.rpsys.net/openzaurus/temp/tempdiff1a.patch is the problematic part of the patch Jun 06 01:44:01 zecke: you mean back to the phones list of applets? Jun 06 01:44:03 RP: How did you found this? Jun 06 01:44:21 do13: Trial and error :-( Jun 06 01:44:53 do13: Rolling back versions until I found the patch the problems started from Jun 06 01:45:34 morning chaps Jun 06 01:45:55 bk support is going away at July 1st. Is there any plan to migrate the oe packages tree ? Jun 06 01:47:00 proti: plans yes, but no capable scm is there yet Jun 06 01:47:20 pb_: morning chief Jun 06 01:47:41 XorA: - If the user has indicated the need to end the proactive SIM session, the ME shall send a TERMINAL Jun 06 01:47:45 RESPONSE with "Proactive SIM session terminated by the user" result value. Jun 06 01:47:50 zecke: So, what's the backup plan ? Is the 1st of july a hard limit ? Jun 06 01:47:52 XorA: figure a meaning out of that Jun 06 01:47:58 proti: yup Jun 06 01:48:06 zecke: heh, let's hope the SCM developers redouble their efforts in the next three weeks. Jun 06 01:48:18 proti: anyway we get a cvs tree Jun 06 01:48:36 proti: in the worst case we can run cvs2svn on it and have *eek* subversion... Jun 06 01:48:47 proti: or cvs2monotone Jun 06 01:48:51 and use svk on our workstations Jun 06 01:48:55 zecke: Will the history & comment be preserved ? Jun 06 01:49:20 proti: I hope so Jun 06 01:49:39 proti: I wonder how he emulates moves + removes... Jun 06 01:50:03 To me, something will be lost in the move. Jun 06 01:50:14 RP: Now you have a working kernel? Jun 06 01:50:23 Hmmm. rmk has now washed his hands of the pxa problem - "kernels are now too large for my lubbock so someone with a working pxa will have to debug this" Jun 06 01:50:37 hah Jun 06 01:50:40 pb_: hehe Jun 06 01:50:43 RP: hehe Jun 06 01:50:51 !ubbock = !supported Jun 06 01:50:52 koen: Dunno? Jun 06 01:50:52 do13: If you reverse the above patch, I suspect it will work Jun 06 01:51:09 !lubbock even Jun 06 01:51:10 koen: Dunno? Jun 06 01:51:12 do13: rmk sent me a patch. It doesn't work though Jun 06 01:51:17 RP: most easies try? compile without prempt? Jun 06 01:51:30 RP: Ok I will check this after work. Jun 06 01:51:34 ~lart broken keyboards Jun 06 01:51:34 * ibot eats broken keyboards and falls over dead Jun 06 01:51:35 zecke: That will make it work. Its not the point though ;-) Jun 06 01:52:00 RP: it would just confirm that the patch is indeed guilty? Jun 06 01:52:24 zecke: I've already done that - the above patch is what breaks things Jun 06 01:52:44 I just haven't tested whether reversing it fixes any of the later kernels Jun 06 01:52:49 RP: does mmc/sd work with preempt nowadys? Jun 06 01:52:58 Reversing it fixes 2.6.12-rc4-git1... Jun 06 01:53:13 I've never run a none preempt kernel... Jun 06 01:53:35 (so yes) Jun 06 01:53:36 RP: but you've an #error in mmcblock.c? Jun 06 01:53:58 zecke: I've never seen one... Jun 06 01:54:03 I might be just confused Jun 06 01:55:05 I serisously doubt they'd have let the mmc driver in if the base code wasn't preempt/smp safe... Jun 06 01:57:57 I must be stoned... Jun 06 01:59:18 zecke: bugger knows, our sim doesnt have a handler for that one :-) Jun 06 02:00:38 RP: Why is 0.15rc1 the preferred version of the orinoco modules in a 2.6 image. 0.15rc1 doesn't compile with my tosa build. Jun 06 02:01:39 do13: It should compile and has done for me. If it doesn't it needs fixing Jun 06 02:02:11 pb_, it seems our unslung/openslug builds want to use arm-csl compiler now, and that don't work so good for armeb Jun 06 02:02:33 RP: Only 0.15rc2 compiles fine. Jun 06 02:02:51 do13: 0.15rc2 won't actually work Jun 06 02:03:12 also, the ifupdown-spurious-environ.patch to busybox causes ifup to segfault on openslug now Jun 06 02:03:36 do13:When I find a spare minute I'll be trying to get 0.15rc2 working but I haven't had a chance recently Jun 06 02:03:46 jacques: ah, oh dear. What goes wrong with the csl compiler on armeb? Jun 06 02:04:37 RP: Ok. Actually I do a new build from scratch. In one hour I'll post the error messages. Jun 06 02:04:39 jacques: can you tell what's causing the ifup segfault thing? Jun 06 02:04:48 (as in, what the patch is doing wrong) Jun 06 02:05:02 pb_, ifup is a one line patch, it's removing something which apparently should not be removed in our case Jun 06 02:05:25 beyond that I can't tell, since strace also seems to have been broken recently Jun 06 02:05:34 for openslug Jun 06 02:05:38 (it used to work) Jun 06 02:06:01 -static char **environ = NULL; Jun 06 02:06:06 that's all the ifup patch does Jun 06 02:06:23 yeah. I thought that ought to be pretty safe. Jun 06 02:06:39 but, well, evidently not. Jun 06 02:06:39 heh Jun 06 02:07:01 I'll back it out for the time being. Jun 06 02:07:09 I'm trying to recall exactly what arm-csl did - oh yeah Jun 06 02:07:17 give me a sec - I've got the error message Jun 06 02:07:45 /home/packages/oe/nslu2/bb_openslug_default/tmp/cross/lib/gcc/armeb-linux/3.4.4/../../../../armeb-linux/bin/ld: unrecognised emulation mode: armelf_linux Jun 06 02:07:45 Supported emulations: armelfb_linux armelfb Jun 06 02:08:15 XorA: thx for --no-flip info Jun 06 02:08:39 jacques: hm, right, looks like csl-arm is missing some kind of armeb patch. Jun 06 02:08:41 hrw|work: no probs, UI assume you have to have a custom session file? Jun 06 02:08:53 pb_, AFAICT it's missing *all* the armeb patches Jun 06 02:08:59 jacques: ah. Jun 06 02:09:04 which raises the question of what good it does for armeb Jun 06 02:09:14 wanna try applying the patches and see what happens? Jun 06 02:09:18 XorA: not yet Jun 06 02:09:44 pb_, ok, but it might have to be a bit later Jun 06 02:09:48 okay Jun 06 02:10:00 I'll turn off csl-arm for armeb for the time being Jun 06 02:10:10 I just think arm-csl was never tested with armeb Jun 06 02:10:20 heck, if it was, it wouldn't need armeb patches Jun 06 02:10:57 everyone forgets about us poor armeb cousins Jun 06 02:11:42 yeah, probably. Jun 06 02:11:46 who wrote the armeb patches for gcc? Jun 06 02:12:16 whoeever it was, you should get them to have the patches checked in to the gcc cvs tree. Jun 06 02:12:38 yeah, I keep thinking they will be but they never are (checked in) Jun 06 02:12:59 it's sad that even gcc4 needs the same patches gcc 3.3 did Jun 06 02:13:41 I mean, these patches are everywhere, even in crosstool - why haven't then been checked in - I gotta think there's some reason other than nobody thought of it Jun 06 02:14:51 it might be holgerschurig who wrote them Jun 06 02:16:26 03pb 07 * r1.3520 10openembedded/packages/busybox/busybox_1.00.bb: disable ifupdown-spurious-environ patch due to segfaults on armeb Jun 06 02:16:31 afaik, it's just that nobody who can check them in knows anything about them. Jun 06 02:16:43 you need to find someone to champion those patches to the gcc folk. Jun 06 02:17:17 are the armel patches being checked in as far as you know? Jun 06 02:17:24 yes Jun 06 02:17:35 ok Jun 06 02:19:29 03pb 07 * r1.3521 10openembedded/packages/gcc/gcc_csl-arm-20050603.bb: remove gcc_csl-arm preference on armeb Jun 06 02:20:00 pb_, I noticed the gcc34-arm-ldm-peephole.patch changed between 3.4.3 and 3.4.4 and some of the differences look... strange to my untrained eyes Jun 06 02:20:13 which bit looked strange? Jun 06 02:20:14 it's the only patch that changed size Jun 06 02:20:58 this for one: Jun 06 02:21:01 > - && (val_diff == 4 || val_diff == -4)); Jun 06 02:21:01 > + && (val_diff == 4 || val_diff == -4)) Jun 06 02:21:23 hi schurig Jun 06 02:21:44 oh I guess I need context on that Jun 06 02:21:49 it's just a return statement Jun 06 02:22:04 looks like a couple of return()'s had their trailing ; removed Jun 06 02:22:15 that does seem a bit odd Jun 06 02:22:26 yeah Jun 06 02:22:39 < - return (reg0 == reg1) && ((val1 - val0) == 4 || (val0 - val1) == 4); Jun 06 02:22:45 hrw|work: hi ho! Jun 06 02:22:45 < + return (reg0 == reg1) && ((val1 - val0) == 4 || (val0 - val1) == 4) Jun 06 02:22:49 hrw|work: back from my baby pause Jun 06 02:23:07 schurig: she sleep now? Jun 06 02:23:09 hrw|work: Sarah Schurig, born on 25. May, 2880 gramm Jun 06 02:23:25 hrw|work: sleep is fine, Sarah eats the last time at 1am, and then again at 8am Jun 06 02:23:34 jacques: ah, no. see, it's just adding another && term to that expression. Jun 06 02:23:35 schurig: a... forgot to gratulate... Gratulations schurig Jun 06 02:23:35 - && (val_diff == 4 || val_diff == -4)); + && (val_diff == 4 || val_diff == -4)) Jun 06 02:23:35 + && (!arm_ld_sched || val0 == 0 || val1 == 0); Jun 06 02:23:45 hrw|work: maybe she inherited that habit from her programmer father :-) Jun 06 02:23:54 schurig: ;)) Jun 06 02:24:05 schurig: keep coffee in safe place then Jun 06 02:24:09 pb_, oh OK, I guess it really was validly changed then :-) Jun 06 02:24:40 yeah, it looks ok to me Jun 06 02:25:46 thanks for making those changes to get us back building - I'll try moving some armeb patches to arm-csl later today Jun 06 02:26:59 good-o Jun 06 02:27:09 or, alternatively, you could try with gcc 4. Jun 06 02:27:34 I've been unable to get an armeb soft-float toolchain built with gcc4 Jun 06 02:28:14 do you think there are any advantages to gcc4 for arm right now? Jun 06 02:28:16 sorry, I meant you could try moving the patches to gcc 4. Jun 06 02:28:39 gcc 4 does have better MI optimisations than gcc 3. Jun 06 02:28:47 MI ? Jun 06 02:30:51 machine independent Jun 06 02:33:19 ah Jun 06 02:39:55 koen: and, how does scratchbox compare to BitBake/OE ? Jun 06 02:40:35 schurig: it doesn't Jun 06 02:40:49 scratchbox is a debugtool IMO Jun 06 02:41:53 you can drop in a OE rootfs and debug some code Jun 06 02:42:12 and build apps with braindead buildsystems Jun 06 02:51:07 morning all Jun 06 02:57:03 hey ade|desk Jun 06 03:05:57 moring Jun 06 03:06:00 morning, too Jun 06 03:06:11 hi mickeyl Jun 06 03:07:02 With opie-image, there's no linuxrc or startup.txt supplied, correct? Jun 06 03:08:42 right. we use systemv init Jun 06 03:10:31 Hm, so we should find another way to have dropbear start on default with our opie-image.. Jun 06 03:13:26 another way? Jun 06 03:13:39 Well, on default, dropbear doesn't seem to start.. Or something else is wrong :) Jun 06 03:13:52 the latter. Jun 06 03:14:01 dropbear starts fine here Jun 06 03:14:12 at least on an image built one week ago Jun 06 03:15:15 Okay. Our image is a few days old, so probably the network itself that's down. Jun 06 03:19:46 Ok, lets see how this init works, thanks sofar mickeyl :) Jun 06 03:23:38 np Jun 06 03:24:42 hm, default pass ain't rootme any longer? Jun 06 03:24:47 it never was Jun 06 03:25:08 Hm, so what should it be ? =) Jun 06 03:25:22 on a release image you need to set one Jun 06 03:25:33 open opie-console and issue passwd Jun 06 03:25:37 then restart dropbear Jun 06 03:25:42 and you should be able to log in Jun 06 03:25:46 But the touchscreen doesn't work on default, and we can't ssh in :) Jun 06 03:39:45 Wheeeeeeeeeeeeeeeee Jun 06 03:40:31 ~netsplit Jun 06 03:40:33 rumour has it, netsplit is something that happens when two IRC servers lose their link, thus isolating the users on every side from each other. a normal part of ALL irc networks, despite what some people bitching about larger networks may seem to think, or an orchestra of poips and thwoops, or something which occurs frequently on OPN Jun 06 03:41:15 ~users Jun 06 03:41:17 so this is what we get for sharing our unpaid volunteer work with you? complaints and accuses? well done, this clearly supports our motivation to continue working on open source projects. Jun 06 03:41:41 03mickeyl 07 * r1.3518.1.1 10openembedded/conf/distro/openzaurus-3.5.4.conf: openzaurus-3.5.4: bump CVSDATE Jun 06 03:43:53 mickeyl: opie changes? oz-3.5.4 Jun 06 03:44:08 hrw|work: ? Jun 06 03:44:42 mickeyl: just curious does cvsdate bump is related to some opie changes Jun 06 03:45:06 or gnu-config :) Jun 06 03:45:16 hrw|work: aah. complete sentences are good :) no, I just felt the need to upgrade so that we don't get outdated. Jun 06 03:45:33 only gradual things in Opie missing Jun 06 03:45:33 ;) Jun 06 03:45:42 although someone works on porting it to qt4/E Jun 06 03:45:46 i'm not commenting on that though Jun 06 03:45:58 qt4/e? Jun 06 03:46:07 still going with the whole "we need fb access" routine? Jun 06 03:46:08 I had to ask about it too... Jun 06 03:46:38 dunno, i didn't look at the branch yet, but from the talk on the list it sounded like qt4/e Jun 06 03:46:40 gpe has much faster rotate then opie has.. Jun 06 03:46:56 bbl, lunch time Jun 06 03:47:07 good idea Jun 06 03:47:20 bah, is an hour till my lunch :-( Jun 06 03:48:02 I can go even now but I prefer >14:00 Jun 06 03:48:48 sometime BST time sucks Jun 06 04:03:09 afternoon Jun 06 04:09:54 reenoo: moin moin :-) Jun 06 04:13:23 xserver-kdrive-omap_6.6.3.bb: tweak xserver-omap, courtesy Florian Boor Jun 06 04:14:13 moin schurig Jun 06 04:20:58 pb_: modutils-cross_2.4.27.bb doesn't compile anymore. Could you do bk mv packages/modutils/modutils-2.4.27/gcc4.patch packages/modutils/files/ Jun 06 04:21:13 fsck... something failed on my home system and I cannot get shell there ;( Jun 06 04:22:14 hrw|work: I hate when that happens Jun 06 04:22:42 XorA: I hate more things in linux Jun 06 04:24:21 hrw|work: though I hated blowing an ADSL filter more yesterday Jun 06 04:24:41 do13: I will do this Jun 06 04:24:53 thanks Jun 06 04:26:06 do13: bk push'ed Jun 06 04:26:28 * schurig hates this consistency checks Jun 06 04:27:30 hi all Jun 06 04:27:49 hey bluelightning Jun 06 04:27:59 hi reenoo Jun 06 04:28:09 hey bluelightning Jun 06 04:28:16 hi bluelightning Jun 06 04:28:30 hi koen Jun 06 04:28:32 hi RP Jun 06 05:15:34 hi offroadgee Jun 06 05:15:57 hi hrw Jun 06 05:18:44 hi offroadgeek_ Jun 06 05:30:01 is bk clone bk://openembedded.bkbits.net/openembedded supposed to work? Jun 06 05:32:39 That's what I do over here as well gints|wrk Jun 06 05:32:56 so no :( Jun 06 05:33:35 What's not working ? Jun 06 05:33:42 Tried it for the last time last week Jun 06 05:33:47 clone: cannot determine remote pathname Jun 06 05:34:27 bk clone bk://openembedded.bkbits.net/openembedded Jun 06 05:34:33 yup, same command Jun 06 05:37:58 lunch time Jun 06 05:38:47 bon appetit hrw Jun 06 05:40:04 When I bitbake opie-image, does it default to tslib input layer? Jun 06 05:40:48 that depends Jun 06 05:40:53 read the qte bb Jun 06 05:40:59 ah, qte, ok Jun 06 05:41:32 DEPENDS = "zlib libpng jpeg tslib uicmoc-native" Jun 06 05:41:39 My guess is yes, then ? Jun 06 05:42:16 Lots of -DQT_QWS_TSLIB too Jun 06 05:42:38 The problem is, cat /dev/input/event0 gives output when I touch the TS, but in opie, it doesn't respond Jun 06 05:42:41 hi RP (a little delayed) Jun 06 05:42:44 I exported it using export TSLIB_TSDEVICE Jun 06 05:42:49 But that doesn't help Jun 06 05:43:36 any error messages on stdout? Jun 06 05:43:48 Nope. Nothing :| Jun 06 05:44:07 On another rootfs, it works, but I can't find the differences Jun 06 05:44:13 ts.cofn is the same, tslib.sh too Jun 06 05:44:21 'set' gives the same exports Jun 06 05:44:24 I'm puzzled. Jun 06 05:44:53 So the only thing I can think of is opie not being 'compiled' for tslib, is that possible? Jun 06 05:45:26 possible. you can check the compile log to make sure Jun 06 05:45:41 hey mickeyl Jun 06 05:46:49 hi gints Jun 06 05:52:34 ascent, I think there is another setting you need to make ... Jun 06 05:52:48 crofton_|laptop: please, enlighten me :) Jun 06 05:52:48 what machine are you on? Jun 06 05:52:50 2210 Jun 06 05:52:59 hang on, let me check something Jun 06 05:53:01 Thanks :) Jun 06 05:53:08 because I am puzzled. Jun 06 05:53:59 mickeyl: I teached libopie2 to use the tosa 2.6 backlight. Could you checkin this patch http://www.do13.in-berlin.de/openzaurus/temp/libopie2-tosa-lcd.patch Jun 06 05:54:03 [balister@omap-devel omap5912osk]$ more tslib.sh Jun 06 05:54:03 #!/bin/sh Jun 06 05:54:03 export TSLIB_TSDEVICE=/dev/input/event1 Jun 06 05:54:03 export QWS_MOUSE_PROTO=TPanel Jun 06 05:54:23 [balister@omap-devel omap5912osk]$ more ts.conf Jun 06 05:54:23 module_raw input Jun 06 05:54:23 module pthres pmin=1 Jun 06 05:54:23 module variance delta=30 Jun 06 05:54:23 module dejitter delta=100 Jun 06 05:54:24 module linear Jun 06 05:54:36 ts.conf is taken care of. The QWS_MOUSE_PROTO looks new to me, lets test Jun 06 05:54:42 TSDEVICE is already set (to event0 in our case) Jun 06 05:54:51 Yeah, that was the one that I had trouble finding out about :) Jun 06 05:54:57 ascent: one is usually keyboard, the other touchscreen Jun 06 05:55:06 ascent: but I don't know your hardware :-) Jun 06 05:55:19 ascent: you can find out with hexdump well, catting event0 and touching the touchscreen gives output :) Jun 06 05:55:22 so.. :) Jun 06 05:55:32 yep, smae method I sued :) Jun 06 05:55:35 schurig: depends on the sequence of the modules Jun 06 05:55:40 hah Jun 06 05:55:41 er used Jun 06 05:55:50 detect-stylus --device :) Jun 06 05:56:17 koen: remember you sugested me somthing to do when I tested my 2.6 kernel... Jun 06 05:56:23 koen: in what ipkg is that program located? Jun 06 05:56:34 also was case that touchscreen does not worked... Jun 06 05:56:58 koen: yeah, the best way would be to interatore over all /dev/input/eventX devices, and then use the exported input API frm Linux to find out if the device submits EV_ABS or EV_KEY events Jun 06 05:56:59 ascent: http://ipkgfind.handhelds.org/details.phtml?package=detect-stylus&official=&format= Jun 06 05:57:01 Ok, let's test that out. Jun 06 05:57:13 ah, gpe :) Jun 06 05:57:21 schurig: that's what detect-stylus does Jun 06 05:57:23 but, using cat /dev/input/event0 is just as reliable, right? Jun 06 05:57:28 ascent: yes, Jun 06 05:57:30 good. Jun 06 05:57:43 ascent: just use "hexdump" (from busybox), doesn't look so garbled ... Jun 06 05:57:54 remove the last 20 lines from detect-stylus.c to kill the X dep Jun 06 05:57:55 heh :) Jun 06 05:58:22 ascent: also, there are some test scripts in tslib-tests.ipk Jun 06 05:58:39 s/scripts/programs/ Jun 06 05:58:39 using ts_calibrate, calibrating works fine Jun 06 05:59:06 The mouse proto var should do the trick for you Jun 06 05:59:20 this sounds familiar to the path I took on the OSK Jun 06 05:59:44 * koen wants an omap too Jun 06 06:00:07 * Crofton_|laptop wants an omap 2 Jun 06 06:00:14 :) Jun 06 06:00:14 nokia 770 Jun 06 06:00:19 road sk101 Jun 06 06:00:28 * schurig wants a oma Jun 06 06:00:37 (german word for Grandmother) :-) Jun 06 06:00:40 heh Jun 06 06:00:44 dutch too :) Jun 06 06:01:07 * koen looks at the omap2420 specs Jun 06 06:01:18 * ascent hugs crofton_|laptop, it works! Jun 06 06:01:20 QWS_MOUSE_PROTO=TPanel rules :) Jun 06 06:01:28 But, how on earth should we have found that one :) Jun 06 06:01:51 I forget who told me, but now three of us know the answer :) Jun 06 06:02:04 Whee, topic it :) Jun 06 06:02:07 Maybe I found it rooting through tslib machine files? Jun 06 06:02:10 by reading the source Jun 06 06:02:56 lots of people know the TSLIB_DEVICE var, but fewer know the QWS one Jun 06 06:03:59 Crofton_|laptop: an omap2 even has 3d accell Jun 06 06:04:16 I wonder how well opengl ES works on that Jun 06 06:04:20 heh, I need high speed USB and DSP combo Jun 06 06:04:42 the dsp is still the same as far I can see Jun 06 06:04:54 Ok, sound now :) Jun 06 06:05:01 I've heard rumours you may be able to get one with a "hotter" DSP Jun 06 06:05:33