| 00:08.55 | _law_ | Bernardo: thanks iŽll try it |
| 00:09.53 | Bernardo | _law_: I did that to build a opie-image 60 minutes ago |
| 00:14.53 | treke|home | <PROTECTED> |
| 00:14.56 | zecke | mickey|tv: funny the debian pysqlite modul is ancient |
| 00:26.05 | Bernardo | mickey|tv: wireless-tools doesn't get built, so no 802.11... |
| 00:28.50 | Bernardo | apparently it isn't a dependency of opie-image or meta-opie |
| 00:41.00 | *** join/#oe tigrux (~tigrux@201.144.60.186) |
| 00:42.40 | Bernardo | 'night (this time for real) |
| 00:44.46 | *** join/#oe imk (~dj@69-175-55-137.ontrca.adelphia.net) |
| 00:52.22 | *** join/#oe keturn (~kevint@c-24-21-141-200.hsd1.or.comcast.net) |
| 01:03.59 | reenoo_ | 'night all |
| 03:51.32 | *** join/#oe |dkey| (~dkey@L0004P23.dipool.highway.telekom.at) |
| 03:53.52 | *** join/#oe hufnus (~slonsiki@m8c3c36d0.tmodns.net) |
| 04:20.52 | ljp | ahhhh.. back to warmth!! |
| 05:23.13 | mickeyl | morning |
| 05:38.52 | *** join/#oe Timelord (~TL@64-147-65-96.access-3.intertex.net) |
| 05:49.41 | CIA-4 | 03mickeyl 07 * r1.3287 10openembedded/packages/qpf-fonts/qpf-arabic_1.0.bb: add arabic qpf fonts courtesy Abdulhaq Lynch |
| 06:17.46 | *** join/#oe uv1 (~Werner@p50892231.dip0.t-ipconnect.de) |
| 06:26.27 | uv1 | Morning! |
| 06:26.33 | uv1 | Ping 255 |
| 06:34.22 | *** join/#oe Timelord (~TL@64-147-65-167.access-4.intertex.net) |
| 07:01.24 | _law_ | Bernardo|Away: yea, thats the clue, now my gpe-image builds!!! |
| 07:05.45 | CIA-4 | 03mickeyl 07 * r1.3289 10openembedded/packages/linux/ (2 files in 2 dirs): apply oz-releasestring to openzaurus-pxa27x kernel to remove the WARNING |
| 07:05.49 | CIA-4 | 03mickeyl 07 * r1.3288 10openembedded/packages/meta/ (opie-image.bb pivotboot-image.bb opie-collections.inc): fix feed locations in image.bb's |
| 07:14.55 | *** join/#oe zap (~zap@217.170.93.196) |
| 07:25.11 | CIA-4 | 03mickeyl 07 * r1.3290 10openembedded/conf/machine/ (zaurus-clamshell-2.4.conf spitz.conf): ship usb host and bluetooth modules in CKX base images |
| 07:30.58 | CoreDump|home | morning |
| 07:31.15 | mickeyl | morning core |
| 07:31.17 | mickeyl | feeling better? |
| 07:31.28 | CoreDump|home | a bit, yeah |
| 07:53.30 | CIA-4 | 03mickeyl 07 * r1.3291 10openembedded/ (2 files in 2 dirs): last bits to fix the warning on CXK kernels |
| 08:04.22 | mickeyl | Noodles: ping me when you're online |
| 08:06.11 | *** join/#oe keturn (~kevint@c-24-21-141-200.hsd1.or.comcast.net) |
| 08:15.14 | Bernardo | 'morning |
| 08:19.22 | CoreDump|home | morning Bernardo |
| 08:20.32 | _law_ | morning Bernardo |
| 08:22.16 | Bernardo | see you guys later - have to go again... :) |
| 08:25.29 | *** join/#oe Timelord (~TL@64-147-65-167.access-4.intertex.net) |
| 08:32.00 | CIA-4 | 03CoreDump 07 * r1.3292 10openembedded/packages/linux/collie-kernel-58-6-debug_2.4.18-rmk7-pxa3-embedix20030509.bb: Add a debug kernel for collie with KERNEL_CONSOLE=/dev/tty0 |
| 08:33.41 | CoreDump|home | interestingly, a current built is missing libgcc_s.so which init requires |
| 08:33.57 | *** join/#oe jh_ (~jason@69.144.85.206) |
| 08:34.30 | CoreDump|home | i guess it's time to trash tmp yet again... |
| 08:52.15 | Bernardo | CoreDump|home: no need for that |
| 08:53.53 | Bernardo | remove tmp/deploy/ipk/*3.4.0-r10*, tmp/stamps/*gcc*package, and rebuild gcc-cross-initial, gcc-cross, gcc |
| 08:55.11 | Bernardo | that "missing libgcc_s.so" problem has already bitten me and _law_ |
| 09:02.29 | CoreDump|home | Bernardo: ah thx :) too late for that now anyways hehe |
| 09:08.06 | RP | morning all |
| 09:08.56 | RP | Sadly this weekend isn't going to plan for me. I've been roped into building a fence so no kernel work or atilib or anything else I had in mind :-(* |
| 09:14.28 | ljp | you cant do that at the same time? |
| 09:15.14 | Noodles | mickey|brfast: I'm about now, but I guess you're eating. :) |
| 09:22.30 | Bernardo | RP: do like I do, write one line of code every few hours... ;) |
| 09:22.34 | Bernardo | hi Noodles |
| 09:24.59 | *** join/#oe koen (~koen@cl-148.ams-05.nl.sixxs.net) |
| 09:28.43 | *** join/#oe bluelightning (~bluelight@203-173-244-25.akl.ihugultra.co.nz) |
| 09:35.52 | *** join/#oe qfh (~qfh@ppp-82-135-2-35.mnet-online.de) |
| 09:35.52 | koen | A |
| 09:35.57 | koen | good morning all |
| 09:36.08 | RP | morning koen |
| 09:36.16 | koen | hey RP |
| 09:41.43 | bluelightning | hi all |
| 09:41.44 | bluelightning | hi RP |
| 09:42.30 | RP | hi bluelightning |
| 09:43.49 | koen | hey bluelightning |
| 09:44.13 | bluelightning | hi koen |
| 09:50.30 | kergoth | morning. |
| 09:51.24 | bluelightning | ki kergoth |
| 10:17.03 | *** join/#oe Laibsch (~leggewie@G14fc.g.pppool.de) |
| 10:23.13 | *** join/#oe mithro (~tim@202.174.36.83) |
| 10:31.52 | kergoth | bleh |
| 10:40.25 | koen | bleh indeed |
| 10:40.38 | koen | (12:40 pm) |
| 10:50.46 | kergoth | heh, go to bed late? |
| 10:53.11 | koen | we had a big party at the uni (9k people) and I opted to sleep from 1 to 7 am and coordinate the cleanup |
| 10:53.25 | koen | so I had my fist pint of guinness at 10 am |
| 10:54.33 | koen | And now I'm stuck doing some work in matlab I have to finish before tomorrow morning |
| 10:54.59 | kergoth | fun fun |
| 10:55.22 | koen | ~botmail for pb_: do you get a response to your question about the eabi unwinder yet? |
| 10:55.48 | koen | "Out of Memory" |
| 11:18.32 | mickeyl | Noodles: just a couple of questions |
| 11:19.54 | mickeyl | Noodles: good work on the pivot_root stuff so far. do you also get "umount: cannot open /proc/mounts and umount: /media/realroot: Device or resource busy" ? |
| 11:20.50 | mickeyl | Noodles: we have to do something about the cardctl order. It's annyoing to have the user remove their CF card when rebooting |
| 11:20.50 | Noodles | Yes, if we want to unmount /media/realroot it'll have to be done after we've started the init on the HD I think, because otherwise the device is in use for the script we're runnnig. |
| 11:21.18 | Noodles | I noticed you'd done a number of fixes yesterday; I'm building an image with them pulled in at present. |
| 11:21.34 | mickeyl | yeah, i fscked up fstab temporarily |
| 11:21.49 | Noodles | Can we configure pcmcia on the pivotroot to just only look at socket 1? |
| 11:22.17 | Noodles | I only have CF IO cards rather than memory cards, so I haven't hit the problem myself. |
| 11:22.30 | mickeyl | Noodles: either that or patch the kernel to expose the hd slot always as 0 |
| 11:23.22 | mickeyl | Noodles: i removed SharpROM without looking at it. How do they do that? |
| 11:23.30 | koen | keep in mind that other platform than c3000 will want to use pivotroot |
| 11:23.40 | Noodles | Dunno. I didn't really prod the SharpROM that much. |
| 11:23.48 | Noodles | Oh, they had the root fs as flash. |
| 11:23.54 | mickeyl | ah i see |
| 11:23.55 | Noodles | With lots of symlinks into the HD. |
| 11:24.00 | mickeyl | ugly ugly |
| 11:24.19 | mickeyl | btw., any idea what we should do with the hdd2 and hdd3? |
| 11:24.32 | mickeyl | how about using hdd2 for home and hdd3 for ... dunno ? |
| 11:24.45 | Noodles | I though one for home made sense and wasn't sure about the other. |
| 11:25.03 | mickeyl | do you have the partitioning offhand? |
| 11:25.08 | mickeyl | s/have/know/ |
| 11:25.54 | Noodles | No. I think hda3 is 3G+, hda1 is around 100M and hda2 is the rest? Roughly OTTOMH. |
| 11:26.07 | Noodles | And hda2/hda3 are fat IIRC. |
| 11:26.21 | mickeyl | ok, i'll switch it on and have a look |
| 11:27.30 | mickeyl | <PROTECTED> |
| 11:27.46 | mickeyl | hda1 is 102784 blocks |
| 11:27.50 | mickeyl | hda2 411264 |
| 11:27.54 | mickeyl | hda3 3485664 |
| 11:28.03 | mickeyl | that would be 100M, 400M, 3.4G |
| 11:28.09 | Noodles | Not bad from memory then. :P |
| 11:28.15 | mickeyl | indeed |
| 11:28.26 | mickeyl | and a shitload of storaged compared to other PDAs |
| 11:28.26 | mickeyl | heh |
| 11:28.33 | Noodles | 's why I bought one. :) |
| 11:28.38 | mickeyl | can't imagine right now what we can do with all this space |
| 11:28.42 | Noodles | Oggs. |
| 11:28.53 | koen | unstripped binaries + gdb |
| 11:29.08 | mickeyl | record audio |
| 11:29.23 | mickeyl | although mono is pretty lousy |
| 11:29.24 | mickeyl | heh |
| 11:29.36 | mickeyl | guess i should look for a usb sound card |
| 11:30.00 | Noodles | I need to find a 2.5mm to 3.5mm adaptor so I can put a handsfree kit on it and see about VOIP. |
| 11:30.05 | mickeyl | power management is pretty odd btw. |
| 11:30.06 | Noodles | That'd be pretty neat. |
| 11:30.24 | mickeyl | i fixed apm to make the resume button work, but every now and then when i suspend, the hdd light stays on |
| 11:30.30 | mickeyl | that pretty much scares me |
| 11:30.38 | Noodles | Have you figured it out? Every time I think I've figured out its logic I find a case where it's different than I expect. |
| 11:31.07 | mickeyl | i fixed our apm wrapper to kill apmd based on detecting a sharp machine - instead of looking for 'embedix' |
| 11:31.10 | mickeyl | that works a bit better |
| 11:31.11 | Noodles | You can suspend with /proc/sys/pm/suspend (or something similar) as well. |
| 11:31.18 | Noodles | Right. |
| 11:31.36 | Noodles | As I say, I'm building an image which has pulled in all your stuff from yesterday and will see how that goes. |
| 11:31.40 | CIA-4 | 03koen 07 * r1.3293 10openembedded/packages/gpe-tetris/gpe-tetris_0.6-4.bb: Add new gpe-tetris .bb that I forgot to commit a few days ago |
| 11:31.56 | Noodles | Oh, the pivotroot init for Spitz should really fsck the drive before mounting it too. |
| 11:32.18 | mickeyl | ya, there are lots of neat things I'm planning for the pivotroot |
| 11:32.25 | mickeyl | a real boot menu and stuff |
| 11:32.39 | Noodles | That'd be nice. I could put Debian on hda3 and choose at boot time. |
| 11:33.01 | mickeyl | indeed. and even more - because when we have pivotroot anyway, then I want to pivot using nfs and CF/SD |
| 11:34.35 | mickeyl | opie-wise we have to fix keyboard stuff |
| 11:35.00 | mickeyl | perhaps we can use the loadkeys scheme from 2.6 images on 2.4 as well |
| 11:35.10 | mickeyl | since it could take a bit until we have 2.6 on CXK |
| 11:36.51 | CoreDump|home | I'm really curious who you manage to use pivot_root at all. Whatever I try, the "pivot_root" command never returns |
| 11:37.22 | mickeyl | the c3000 is my first usage of it |
| 11:37.29 | mickeyl | seems to work pretty well |
| 11:37.33 | Noodles | How does the keyboard stuff work? I had a look at qte, but I couldn't see how it actually would deal with things like Fn+r giving "+". |
| 11:37.34 | CoreDump|home | :\ |
| 11:37.35 | mickeyl | pivot_root <new> <old> |
| 11:38.11 | Noodles | I did discover Fn+Ctrl+Up would change VT from OPIE though. |
| 11:38.12 | CoreDump|home | i know, it just freezes during pivot_root for me. no matter what I try. It is supposed to be really simple |
| 11:38.14 | mickeyl | Noodles: IIRC qte doesn't compose that.- FN+<key> is done in the kernel afaik |
| 11:38.28 | Noodles | QTE is doing something with it. |
| 11:38.36 | Noodles | Try pressing Fn in a terminal under OPIE. |
| 11:38.43 | Noodles | It's not just a shift modifier. |
| 11:39.13 | mickeyl | that's probably because the keymap is totally off anyway |
| 11:39.14 | Noodles | The kernel keymap patches in the qte package seem to suggest raw mode or something? |
| 11:39.25 | mickeyl | ya, they're reading using mediumraw |
| 11:39.38 | mickeyl | that requires a sanely populated kernel keymap though |
| 11:39.44 | mickeyl | which is not what we have on 2.4.x |
| 11:39.53 | mickeyl | so we need to populate one and then use loadkeys |
| 11:42.38 | Noodles | The keymap works fine on the console though. |
| 11:43.02 | Bernardo | Noodles: fn+ctrl+up is a nice trick |
| 11:43.26 | Bernardo | Noodles: you also get up/down switched with left/right under opie? |
| 11:43.54 | Noodles | Yes. |
| 11:44.13 | Bernardo | qclockchange segfaults... :( |
| 11:44.35 | Bernardo | mickeyl: one thing I miss are the bluetooth plugins that exist for sharp rom |
| 11:44.53 | Bernardo | it would make connecting to a pan network very easy |
| 11:45.18 | Bernardo | mickeyl: the hinge is also related to the keymap? |
| 11:48.40 | *** join/#oe mallum (~mallum@host81-153-52-223.range81-153.btcentralplus.com) |
| 11:52.23 | *** join/#oe dkey| (~dkey@L0016P09.dipool.highway.telekom.at) |
| 11:57.36 | *** join/#oe woglinde (woglinde@pD9E7EE50.dip.t-dialin.net) |
| 11:57.36 | mickeyl | yes, hinge is just a key |
| 11:58.00 | mickeyl | i miss the whole networking apps from Qtopia/Sharp |
| 11:58.06 | mickeyl | they are pretty good |
| 11:58.11 | mickeyl | compared to ours :/ |
| 11:58.23 | Bernardo | the "connect" applet is very good |
| 11:58.30 | Bernardo | and the profiles support is nice |
| 11:58.41 | Bernardo | too bad they don't gpl those |
| 11:58.55 | woglinde | mickeyl hm I find the network apss okay |
| 11:59.11 | woglinde | besides of the missig profiles |
| 11:59.34 | Bernardo | the photo app is good too, but my favourite is the audio player, I like the interface and the fact that I can turn off the screen and it will play on |
| 12:00.02 | woglinde | now formula one |
| 12:00.19 | mickeyl | woglinde: opie network app is pretty limited, i.e. no bluetooth, no wlan-ng |
| 12:00.48 | woglinde | mickeyl but better than to work with vi |
| 12:01.09 | mickeyl | i wonder under what rock wimpie hides |
| 12:01.13 | mickeyl | ~seen wimpie |
| 12:01.37 | ibot | wimpie <u19809@cable-62-205-106-215.upc.chello.be> was last seen on IRC in channel #familiar, 101d 20h 29m 30s ago, saying: 'dmesg shows usb_recv: RPError 27'. |
| 12:01.38 | Bernardo | woglinde: who are you cheering for? |
| 12:01.52 | woglinde | he i saw the new version, zecke shoed me |
| 12:01.58 | woglinde | showed |
| 12:02.01 | mickeyl | ns2? |
| 12:02.03 | koen | last I herad from Wim, was that he wanted to port linux to the h6300 |
| 12:02.05 | woglinde | yepp |
| 12:02.08 | koen | heard* |
| 12:02.12 | mickeyl | well, it's far from finished |
| 12:06.46 | RP | Of course now cardmgr isn't needed for 2.6 we can probably remove the need for pivotroot in due course anyway :) |
| 12:06.56 | CIA-4 | 03koen 07 * r1.3294 10openembedded/packages/gpe-plucker/ (gpe-plucker/remove-render.patch gpe-plucker_0.2.bb): |
| 12:06.56 | CIA-4 | gpe-plucker_0.2.bb: |
| 12:06.56 | CIA-4 | <PROTECTED> |
| 12:08.24 | mickeyl | RP: yeah, that's absolutely great, I've been looking for that feature since long |
| 12:08.55 | mickeyl | boot from nfs via CF-ethernet will now become possible |
| 12:09.32 | RP | mickeyl: We can also probably boot from an SD card as the 2.6 kernel stands now |
| 12:09.34 | koen | booting nfs over usbnet is easier |
| 12:09.35 | mickeyl | hrw|gone: ping |
| 12:09.43 | woglinde | koen *g* |
| 12:09.45 | mickeyl | RP: yeah, very good |
| 12:09.51 | woglinde | hm why is cardmgr gone? |
| 12:09.57 | mickeyl | oh yes... booting nfs over usbnet was one of the reasons why i sold my SIMpad |
| 12:10.04 | mickeyl | *FRUSTRATION* |
| 12:10.14 | koen | it works very good on my h2210 |
| 12:10.15 | RP | mickeyl: The 2.6.12-rc2-mm3 has the pcmcia patches in it. Unfortunately I get lots of segfaults from the cardctl and friends :-/ |
| 12:10.32 | woglinde | mickeyl? nfs uboot over usb worked for me |
| 12:10.39 | mickeyl | woglinde: kernel 2.6 ? |
| 12:10.51 | woglinde | mickeyl pfff |
| 12:10.56 | mickeyl | woglinde: yeah. |
| 12:11.23 | woglinde | it will work again |
| 12:11.32 | woglinde | * I hope* |
| 12:11.45 | woglinde | .o(where is zecke) |
| 12:12.03 | mickeyl | i can't see anyone motivated enough to write an SA USB gadget driver |
| 12:13.13 | RP | There are much more interesting things to be doing :-/ |
| 12:13.21 | mickeyl | ya |
| 12:13.27 | woglinde | pffff |
| 12:13.32 | mickeyl | and SA is really getting obsolete |
| 12:13.50 | mickeyl | proof me wrong and I'll be happy |
| 12:13.54 | mickeyl | we could use it for collie |
| 12:14.42 | CoreDump|home | interesting, it appears that pivot_root does run successfully, but the script by which it was called is no longer executed. So it just hangs there,looking frozen |
| 12:17.02 | Bernardo | do we have any option to start the apps in "zoomed" mode, like on the sharproom? froot and opie-solitaire look way strange in 640x480 |
| 12:17.12 | mickeyl | no way |
| 12:17.23 | mickeyl | on w100 we know how to change resolution |
| 12:17.34 | mickeyl | however tslib doesn't know about that |
| 12:17.42 | mickeyl | which means we can't use a stylus |
| 12:17.54 | Bernardo | froot doesn't need a stylus... :) |
| 12:18.04 | mickeyl | i don't know if we found out how to get into 320x240 on the newer models |
| 12:18.16 | mickeyl | bbl tonite |
| 12:18.21 | Bernardo | have fun |
| 12:18.52 | woglinde | bye mickeyl |
| 12:21.17 | *** join/#oe zedkatuf (~dfutak@82-32-58-45.cable.ubr08.azte.blueyonder.co.uk) |
| 12:23.22 | Bernardo | Noodles: have you looked at the kernel source, drivers/char/spitz*? The keymap defs for the console are there, maybe we can use that to start working on them for opie |
| 12:26.43 | Noodles | I did look at it, but I didn't find it that easy to read. |
| 12:31.20 | Bernardo | I am trying to see how it is done on the c7xx |
| 12:32.39 | Noodles | Me either. |
| 12:34.03 | Bernardo | at least the keymap is there, so we have a starting point |
| 12:34.41 | Noodles | Have you looked in packages/qte/qte-2.3.10 ? |
| 12:36.55 | Bernardo | going to do that now... |
| 12:38.00 | Bernardo | interesting... we really need to look at qkeyboard_qws.cpp |
| 12:39.19 | Bernardo | maybe we can adjust the corgi patch |
| 12:40.41 | Noodles | It gets applied anyway. |
| 12:41.22 | Noodles | It's the tosa patch you want to look at I think. |
| 12:42.07 | *** join/#oe reenoo_ (~r@p5489AF1C.dip.t-dialin.net) |
| 12:42.15 | reenoo_ | afternoon |
| 12:42.33 | Bernardo | I'd thought the c1000/c3000 keymap would be closer to the corgi than to the tosa |
| 12:42.35 | Bernardo | hi reenoo_ |
| 12:43.12 | reenoo_ | hey Bernardo |
| 12:43.34 | Noodles | All the patches get applied though and IIRC the Spitz compilation options define QT_QWS_SL6000, so it gets pulled in. |
| 12:43.46 | Noodles | Whether that's right or wrong is a different matter. |
| 12:48.16 | Bernardo | I've been checking, there are similarities to the corgi keymap, but ours have a lot more keys |
| 12:50.07 | Bernardo | and in the qte bb file, the patch used is the corgi one |
| 12:50.37 | Noodles | Hmmm. |
| 12:50.47 | Noodles | Though there's obviously an event going on with the Fn key in some sense. |
| 12:51.58 | Bernardo | yes, as it seems to work as a cancel key under opie |
| 13:04.51 | Bernardo | well, I think I know what needs to be done, following the keymap, and replacing the differences in the keymap starting in line 293 in qkeyboard_qws.cpp |
| 13:04.57 | Bernardo | now doing it... :) |
| 13:05.48 | Bernardo | the first change is Qt::Key_F22, which should be mapped to ALT |
| 13:05.55 | Bernardo | brb - she called |
| 13:12.29 | koen | I suspect I found a python package mickeyl hasn't added yet: pyglade |
| 13:12.46 | koen | or glade-python, whatever they are calling it |
| 13:14.35 | CIA-4 | 03koen 07 * r1.3295 10openembedded/packages/mono/files/ (libtool-lossage.patch install-lossage.patch): clean up mono/ dir |
| 13:44.26 | *** join/#oe Laibsch (~leggewie@G14fc.g.pppool.de) |
| 14:04.28 | chouimat | morning |
| 14:06.50 | *** join/#oe marcansoft (1337@80-29-50-242.adsl.nuria.telefonica-data.net) |
| 14:15.36 | Noodles | mickeyl++ # Spitz OPIE is much nicer now. |
| 14:23.38 | Noodles | mickey|away: Why tmpfs for /var btw? |
| 14:25.23 | koen | a leftover from flashbased systems |
| 14:52.32 | *** join/#oe eumel (~chatzilla@pD9EE564A.dip0.t-ipconnect.de) |
| 14:53.00 | *** join/#oe aboeglin (~aboeglin@213.154.79.4) |
| 15:18.19 | *** join/#oe Bernardo (~jbbrs@Bernardo.sourcemage) |
| 15:18.42 | Bernardo | re |
| 15:19.39 | Bernardo | Noodles: I've been looking at the code a little more, it looks like we need the same order for the arrow keys as on the tosa, but a new patch, since there are many differences |
| 15:20.06 | Bernardo | I also need to understand how to map fn keys and special keys... |
| 15:26.19 | uv1|away | ping mickey|away |
| 15:52.26 | CoreDump|home | lart bkbits.net |
| 15:52.29 | CoreDump|home | ~lart bkbits.net |
| 16:05.14 | *** join/#oe rob_w (~roboo@p213.54.138.42.tisdip.tiscali.de) |
| 16:05.19 | rob_w | hi |
| 16:05.33 | rob_w | what can i do again this : Entire repository is locked by: RESYNC directory. |
| 16:05.47 | CoreDump|home | try bk resolve |
| 16:05.47 | rob_w | that is see when i do bk pull ? |
| 16:06.05 | CoreDump|home | or trash packages and re-clone |
| 16:07.05 | rob_w | ok thnx |
| 16:08.06 | CoreDump|home | np |
| 16:15.58 | *** join/#oe [g2] (~g2@g2.nslu2-linux) |
| 16:30.43 | RP | Found the corgi sound problem and qss doesn't work the way ljp says it does :) |
| 16:34.18 | XorA|gone | RP: cool |
| 16:34.42 | XorA|gone | RP: I have a a patched xserver that means gpe for 2.6 is a go |
| 16:34.58 | koen | XorA|gone: cool |
| 16:35.08 | XorA|gone | koen: but the patch is a kludge |
| 16:36.13 | RP | XorA|gone: I'm hoping these changes to the sound don't break X. They shouldn't though :) |
| 16:37.01 | XorA|gone | RP: sound break X? |
| 16:37.01 | RP | XorA|gone: I favour getting X working via a hack for now until we can talk the gpe people (or others) into working out a proper solution |
| 16:37.02 | *** join/#oe woglinde (woglinde@pD9E7EE50.dip.t-dialin.net) |
| 16:37.02 | RP | XorA|gone: In fixing the esd problem, I broke opie ;-) |
| 16:37.02 | XorA|gone | RP: Ill only push my patch if someone more senior give me go ahead |
| 16:37.24 | RP | XorA|gone: Push it but make it machine dependent |
| 16:39.10 | XorA|gone | RP: it is c7x0 only |
| 16:39.23 | RP | XorA|gone: Do you have the patch available anywhere? |
| 16:39.32 | XorA|gone | RP: it will break 2.4 |
| 16:39.39 | XorA|gone | RP: I can stick it on website |
| 16:40.03 | RP | XorA|gone: Breaking 2.4 might not be acceptable :-/ |
| 16:40.15 | RP | It is to me, but... ;-) |
| 16:40.41 | koen | RP: the patch will only show up in the *new* 2.6 image |
| 16:41.03 | koen | apply, bake, revert |
| 16:41.04 | RP | koen: It depends whether we'll ever have a 2.4 gpe image again... |
| 16:41.44 | RP | Can we make the patch depend on the KERNEL_VERSION variable? |
| 16:41.55 | RP | (as a short term fix) |
| 16:42.41 | RP | <CIA> linux-oz-2.6: Fix sound bugs |
| 16:46.55 | XorA|gone | RP: http://hyboria.mine.nu/~dp/xserver-kdrive_20050207.bb |
| 16:47.08 | XorA|gone | RP: http://hyboria.mine.nu/~dp/disable-apm.patch |
| 16:47.36 | XorA|gone | RP: basically I disable apm for kdrive as it doesnt need it on 2.6 |
| 16:48.28 | RP | XorA|gone: Is there a config option for xserver we can do that with? |
| 16:49.07 | XorA|gone | RP: not as far as I can see, kdrive doesnt appear to be nicely modular/configurable |
| 16:49.18 | XorA|gone | RP: apm code isnt within #if |
| 16:50.02 | RP | XorA|gone: Ok :-/. I'd send a note to gpe@hh.org with a link to the patch explaining the problem |
| 16:50.10 | koen | XorA|gone: kdrive doesnt appear to be nicely<period> ;) |
| 16:50.33 | XorA|gone | koen: yeah, I am really wondering if xorg proper isnt the way to go |
| 16:51.22 | koen | iirc Xorg is substantially bigger |
| 16:51.34 | XorA|gone | koen: yeah,but Ive got a C7x0 :-) |
| 17:12.41 | *** join/#oe Timelord (~TL@64-147-65-167.access-4.intertex.net) |
| 17:24.10 | CIA-4 | 03CoreDump 07 * r1.3292.1.2 10openembedded/packages/initscripts/initscripts_1.0.bb: Bump PR |
| 17:24.58 | XorA | koen: ever run gpdf? |
| 17:27.04 | koen | XorA: I crashes pdfs that contain text |
| 17:27.12 | koen | picture only ones are reported to work |
| 17:28.29 | XorA | koen: damnation, thats the one app I really need |
| 17:28.39 | koen | I miss it too |
| 17:28.49 | koen | want to attack it with gdb? |
| 17:28.57 | aoe | can anyone comment on this: perl: symbol lookup error: /usr/lib/libperl.so.5: undefined symbol: __eqdf2 |
| 17:28.58 | CIA-4 | 03CoreDump 07 * r1.3292.1.1 10openembedded/packages/initscripts/initscripts-1.0/bootmisc.sh: Use absolute path when symlinking /tmp |
| 17:29.15 | aoe | or, would I risk my health if I reported this as a bug? |
| 17:29.57 | XorA | koen: Ill do it tomorrow if work is quiet |
| 17:30.02 | koen | cool |
| 17:30.14 | XorA | koen: first I will check debian arm patches :-) |
| 17:30.20 | koen | stracing it yielded no clues |
| 17:31.00 | aoe | I installed a bitbaked (tosa) libc6_2.3.2+cvs20040726-r15_armv5te.ipk too, but something seems to be amiss: my libraries, in contrast to the ones in build/tmp/cross-arm/lib, don't contain any traces of the (soft-float?) symbol __eqdf2 |
| 17:31.17 | koen | what does goofle say about it? |
| 17:31.47 | aoe | in conjunction with zaurus, nothing |
| 17:32.23 | aoe | eqdf2 turns up a few soft-float references ... |
| 17:32.25 | koen | why would it be zaurus specific? |
| 17:32.56 | aoe | because perl compiles fine elsewhere? |
| 17:33.17 | koen | no idea, haven't tried it |
| 17:33.34 | aoe | no one seems to have ;) |
| 17:33.54 | aoe | rather s/compiles/runs/ |
| 17:34.22 | rob_w | CoreDump|home: the last message from bk resolve was "clean" but still bk pull says "locked" |
| 17:34.54 | koen | rm -rf RESYNC should 'solve' it |
| 17:35.28 | rob_w | koen: thnx ..looks good |
| 17:36.23 | rob_w | koen: now i also did tar my sources dir and will now try to copy it to my office pc .. |
| 17:37.29 | *** join/#oe zecke_ (~ich@83-169-170-59-dynip.superkabel.de) |
| 17:47.07 | koen | about 6 megs with gconf |
| 17:47.14 | koen | gconf-dbus pull sin less |
| 17:51.42 | rob_w | can someone pastebin his bitbake.conf .. seems i lost it ( maybe under some garbage here in my room ) |
| 17:53.41 | koen | http://ewi546.ewi.utwente.nl/tmp/bitbake.conf |
| 17:53.55 | koen | (openembedded/conf/bitbake.conf) |
| 17:54.59 | rob_w | thny soo mucg |
| 17:55.02 | rob_w | thny soo much |
| 17:57.06 | XorA | Dr Who time :-) |
| 17:57.41 | *** join/#oe [cc]smart (~smart@gw.ptr-62-65-149-158.customer.ch.netstream.com) |
| 18:28.11 | *** join/#oe Philippe (~philippe@ip-213-49-70-74.dsl.scarlet.be) |
| 18:40.43 | *** join/#oe offroadgeek (~zaurus@dsl093-017-213.msp1.dsl.speakeasy.net) |
| 18:41.39 | *** join/#oe Laibsch (~leggewie@G19f1.g.pppool.de) |
| 18:47.20 | RP | mickey|headache: I found and fixed the opie sound problem and have made a new kernel available |
| 18:47.51 | mickey|headache | RP: nice. what was it? |
| 18:48.25 | RP | mickey|headache: In the last patch I tidied up some code and inadvertenly reverse some logic. A few !'s fixed it ;-) |
| 18:49.13 | RP | mickey|headache: I also checked out the fast key repeat. 2.6's key repeat values are fast. I'm not sure I really want to change them though |
| 18:49.47 | RP | 2.6 default is 250ms before first repeat and the repeat every 33ms |
| 18:51.14 | mickey|headache | perhaps 300ms would be better |
| 18:51.27 | mickey|headache | repetition frequency is good |
| 18:52.02 | mickey|headache | btw., isn't there a ioctl interface for that in the console subsystem? or is that an x86 thing? |
| 18:52.12 | RP | mickey|headache: If we change one, I'd probably change the other to 50ms or so to match |
| 18:52.33 | RP | mickey|headache: Its hardcoded into input.c as far as I can see. If there is an ioctl, that would be ideal though |
| 18:53.18 | RP | mickey|headache: drivers/input.c, input_register_device() if you have a copy of the kernel handy... |
| 18:53.27 | mickey|headache | ya |
| 18:54.23 | *** part/#oe Laibsch (~leggewie@G19f1.g.pppool.de) |
| 18:54.49 | mickey|headache | hmm |
| 18:54.51 | mickey|headache | indeed |
| 18:54.52 | mickey|headache | hardcoded |
| 18:54.54 | mickey|headache | pff |
| 18:55.02 | mickey|headache | that should get a sysfs interface |
| 18:55.50 | RP | mickey|headache: or an ioctl added to evdev.c... |
| 18:56.20 | RP | Anyhow, I'm happy enough its not the kernel driver at fault |
| 18:56.20 | *** join/#oe Philippe (~philippe@ip-213-49-117-9.dsl.scarlet.be) |
| 18:56.35 | RP | The timer does seem to run short but I can't see how... |
| 18:57.00 | RP | I've detailed this on the kernel status page |
| 18:58.00 | mickey|headache | ok |
| 18:58.28 | mickey|headache | so do things work with qss or should we still look for a better sound server ? |
| 18:59.11 | RP | mickey|headache: ljp is wrong in that qss does release /dev/dsp when its not using it. I'd therefore just leave it as it stands |
| 18:59.36 | RP | opie works and if you load another app, that just takes it over |
| 18:59.52 | mickey|headache | ok, good. since we have more important things open, we can leave it like that |
| 19:00.20 | RP | The bug I was chasing was keeping /dev/dsp open hence the confusion |
| 19:00.32 | mickey|headache | ah right |
| 19:01.44 | RP | The good thing is I know understand a lot more about the sound code which will help the alsa driver :) |
| 19:01.58 | mickey|headache | indeed a good thing :) |
| 19:02.13 | *** join/#oe molivier (~mac@f01m-31-99.d4.club-internet.fr) |
| 19:03.00 | mickey|headache | i keep pondering whether to patch 2.4.20 to make it more sane or better wait until someone tackles 2.6 for the CXK |
| 19:03.20 | mickey|headache | 2.4.20-embedix, that is |
| 19:03.21 | mickey|headache | of course |
| 19:04.52 | RP | It depends how soon you think someone will take on 2.6. I keep wondering about starting but without a device I'll probably not get anywhere fast :-/ |
| 19:05.19 | *** join/#oe Cwiiis (~cwiiis@user-214-207-151-83.e7even.com) |
| 19:05.32 | RP | I don't think it will be too bad thought - the PXA270 isn't too bad to work on... |
| 19:05.58 | RP | Although my ipaq randomly segfaults :-/ |
| 19:08.14 | RP | For reference, I have a borrowed c700 at the moment. That should let me do more development/testing as I can keep a stable image on my husky and test on this.... |
| 19:08.39 | mickey|headache | oh good to hear |
| 19:09.50 | RP | Its scary as I can see differences in the firmware. You know you've been working with hardware too closely when :) |
| 19:10.02 | mickey|headache | hehe |
| 19:10.34 | mickey|headache | did you ever got around to looking into the maintenance area? |
| 19:10.37 | hrw | mickey|headache: pong |
| 19:10.40 | mickey|headache | i'd be interested to know what toolkit they used |
| 19:11.12 | mickey|headache | it looked like something nice and small to rip off for boot menu and the like |
| 19:11.15 | hrw | RP: do you know where I can find info how to add resume/suspend actions to driver? |
| 19:11.35 | RP | mickey|headache: I've never really looked at it. To do so I'd need to install 2.4 again... |
| 19:11.41 | mickey|headache | hrw: hi. wanted to ask you to copy a file in wordpress. found out |
| 19:11.49 | hrw | ok |
| 19:12.03 | RP | hrw: You could look at some of my c7x0 2.6 drivers? |
| 19:12.07 | Bernardo|Away | mickey|headache: is qkeyboard_qws.cpp in qte the right file to get the keymap to work? |
| 19:12.30 | mickey|headache | Bernardo: that depends if you want to fix it the right way or if you want to hack |
| 19:12.34 | hrw | RP: probably - which one you sugest? |
| 19:12.35 | mickey|headache | it would be the right file for the hack |
| 19:12.41 | RP | hrw: try arch/arm/common/scoop.c |
| 19:12.47 | mickey|headache | for the right way I recommend looking into the kernel keymap |
| 19:12.53 | hrw | RP: I would like to look for adding suspend/resume into dtl1_cs driver |
| 19:13.04 | RP | I'm assuming you're using the device model. If not, I'd suggest switching the driver to it |
| 19:13.07 | Bernardo | mickey|headache: I started with the corgi patch, and followed the kernel keymap |
| 19:13.10 | RP | It makes suspend/resume hooks easy |
| 19:13.14 | Bernardo | which is very complete, it seems |
| 19:13.27 | hrw | RP: its not my driver |
| 19:13.48 | RP | hrw: Is it in the mainline kernel? |
| 19:13.56 | hrw | RP: yes |
| 19:13.57 | Bernardo | but no mater how many times I rebuild qte with my patch, and reinstall libqte2, it won't work.. The keys act the same way |
| 19:14.04 | hrw | RP: drivers/bluetooth/dtl1_cs.c |
| 19:14.13 | Bernardo | have to go, dinner |
| 19:14.14 | RP | hrw: I'll have a quick look |
| 19:14.42 | mickey|headache | Bernardo: i suggest populating the kernel keymap and using loadkeys to insert it |
| 19:14.50 | mickey|headache | then the qte kernel keymap driver should just work (tm) |
| 19:14.56 | hrw | RP: thx |
| 19:15.42 | RP | hrw: It looks like it recieves suspend/resume events from pcmcia. Look for the CS_EVENT_PM_SUSPEND and CS_EVENT_PM_RESUME events |
| 19:16.19 | hrw | RP: in driver? |
| 19:16.31 | RP | hrw: Yes |
| 19:16.43 | RP | hrw: Do a search on the above |
| 19:16.52 | mickey|headache | hmm i hate when people attempt to fix such problems in userland |
| 19:16.52 | hrw | RP: found - it do nothing then |
| 19:16.53 | RP | hrw: dtl1_event() |
| 19:16.54 | mickey|headache | look here |
| 19:17.01 | mickey|headache | case "$1" in |
| 19:17.01 | mickey|headache | suspend) |
| 19:17.01 | mickey|headache | <PROTECTED> |
| 19:17.01 | mickey|headache | <PROTECTED> |
| 19:17.01 | mickey|headache | <PROTECTED> |
| 19:17.02 | mickey|headache | <PROTECTED> |
| 19:17.03 | mickey|headache | <PROTECTED> |
| 19:17.06 | mickey|headache | <PROTECTED> |
| 19:17.07 | mickey|headache | <PROTECTED> |
| 19:17.10 | mickey|headache | and then |
| 19:17.12 | mickey|headache | ... |
| 19:17.13 | mickey|headache | <PROTECTED> |
| 19:17.15 | mickey|headache | <PROTECTED> |
| 19:17.18 | mickey|headache | <PROTECTED> |
| 19:17.19 | mickey|headache | ... |
| 19:17.25 | mickey|headache | is that really necessary ? |
| 19:17.34 | mickey|headache | IMO the drivers should cope with suspend/resume |
| 19:17.38 | hrw | mickey|headache: wanna see my /etc/network/if.pre-up.d/bnep0? |
| 19:17.39 | RP | hrw: Unless anything looks at link->state, it would appear not |
| 19:17.48 | mickey|headache | hrw: ugly? |
| 19:18.03 | hrw | mickey|headache: cardctl eject/insert, bluetooth restart etc |
| 19:18.09 | mickey|headache | eek, nasty |
| 19:18.13 | hrw | mickey|headache: removed it few minutes after writing |
| 19:18.19 | mickey|headache | heh |
| 19:18.35 | RP | mickey|headache: That does look nasty |
| 19:19.06 | RP | mickey|headache: There are a load of apm scripts in OZ that are equally bad ;-). I'd say 50% aren't needed under 2.6 |
| 19:19.33 | RP | They actually break the usb connection if I understood the reports right |
| 19:20.24 | mickey|headache | looks like we should stop insisting on writing rootfs'es that work for both kernels |
| 19:20.53 | RP | mickey|headache: Or at least have the scripts that aren't needed exit under a 2.6 kernel |
| 19:21.11 | RP | mickey|headache: We're rapidly moving to a state where I'd say bin 2.4 for the c7x0 though |
| 19:21.11 | mickey|headache | yeah, that'd be a start |
| 19:21.44 | mickey|headache | RP: I'm all for it. I'm happy to remove it the second we get both gpe and opie equally working |
| 19:21.59 | RP | I don't think any of us are interested in 2.4 problems and I don't know of anything much 2.6 can't do now |
| 19:22.16 | *** join/#oe dkey (~dkey@193.170.48.234) |
| 19:22.28 | RP | mickey|headache: With the application of XorA's xserver patch, I think a 2.6 kernel +gpe will be working |
| 19:22.47 | RP | mickey|headache: He's just not sure about pushing it as it will break xserver for 2.4 ;-) |
| 19:23.16 | RP | (The patch is a hack to disable xserver's apm access as its no longer needed under 2.6) |
| 19:24.12 | mickey|headache | sounds good to me |
| 19:24.22 | mickey|headache | if it works, let's go to do that for _c7x0 |
| 19:24.55 | RP | XorA|DrWho: You might want to read the above :) |
| 19:27.26 | Luke-Jr | Anyone know of a WM supported by OE that allows resizing windows? Or if it's possible to OE-compile Matchbox w/ such support? |
| 19:28.09 | Luke-Jr | Preferably with OSX-like resizing with a corner designed to help resizing |
| 19:28.41 | koen | xfce? |
| 19:28.50 | koen | otherwise ask in #gpe |
| 19:30.30 | RP | mickey|headache: Are there any other 2.6 bugs I don't know about? |
| 19:31.23 | mickey|headache | don't know of any - other than the freezing that I didn't have time to work out a reproducable sequence :/ |
| 19:31.50 | mickey|headache | I'm working on it though |
| 19:32.04 | RP | mickey|headache: Was this upon a suspend/resume? |
| 19:32.15 | mickey|headache | no |
| 19:32.22 | mickey|headache | while normal operation |
| 19:32.28 | RP | mickey|headache: I have seen my c760 freeze on resume but I suspect the sound issue |
| 19:33.00 | RP | I'd advise upgrading the kernel just to rule out the sound problems... |
| 19:33.27 | *** join/#oe Pigi (~NoOne@host171-27.pool80117.interbusiness.it) |
| 19:33.37 | mickey|headache | ok, will do. I'll download your latest version |
| 19:33.38 | Pigi | Ciao all |
| 19:33.38 | hrw | ~lart sharp again for lack of reset |
| 19:34.12 | hrw | hi Pigi |
| 19:34.19 | RP | hrw: SysRq+B |
| 19:34.26 | RP | hi Pigi |
| 19:34.36 | hrw | RP: sysrq is? |
| 19:35.07 | RP | Left Hand Jap key + Home, release the jap key (holding Home), press B |
| 19:35.15 | hrw | o |
| 19:35.17 | hrw | k |
| 19:35.28 | RP | That's reboot. Sync is press S instead which will stop some of the jffs2 errors |
| 19:35.35 | mickey|headache | ah one thing |
| 19:35.39 | mickey|headache | not sure if it's kernel keymap or qte |
| 19:35.42 | mickey|headache | Ok no longer closes dialogs |
| 19:35.52 | mickey|headache | it works as "enter" but it used to also close dialogs |
| 19:35.57 | mickey|headache | i.e. activate the "ok" button |
| 19:36.13 | mickey|headache | probably a qt thing |
| 19:36.20 | RP | OK is mapped to enter by the kernel keymap |
| 19:36.33 | RP | I'm not sure how to split that up again once in qte |
| 19:37.54 | RP | hrw: Sorry, Its the right hand jap yet |
| 19:38.02 | RP | jap *key* |
| 19:38.10 | hrw | thx |
| 19:38.25 | hrw | have to remember it |
| 19:38.30 | RP | (I've been hammering fence panels all day) |
| 19:42.20 | mickey|headache | physical work is good |
| 19:42.28 | mickey|headache | need to do something different once in a while |
| 19:43.35 | RP | mickey|headache: I seem to have a knack of getting caught up in such things :) |
| 19:43.54 | RP | I can't think of many parts of this house I haven't had or seen in bits... |
| 19:46.05 | RP | At school, they all thought I was mad when I went to welding classes :) |
| 19:46.29 | hrw | RP: current OE 2.6.11 is working good on c7x0? |
| 19:46.34 | mickey|headache | RP: hehe |
| 19:46.47 | Noodles | mickey|headache: Did you find your C3K was mounting / as ro btw? |
| 19:47.01 | mickey|headache | RP: ok, well done. sound works good. well... not really good, but up to par with 2.4 :) |
| 19:47.10 | mickey|headache | Noodles: you mean the real / ? |
| 19:47.14 | mickey|headache | Noodles: before pivot_init? |
| 19:47.17 | Noodles | hda1. |
| 19:47.25 | mickey|headache | oh. no, it always mounted it rw |
| 19:47.40 | Noodles | I saw it before, but had stopped my image doing it and now I find with the images I've built today it's ending up as ro again. |
| 19:47.43 | Noodles | So OPIE doesn't start. |
| 19:47.46 | RP | hrw: It seems to be. I've just made an updated kernel available to hopefully sort the sound issues once and for all... |
| 19:47.53 | Noodles | And I don't seem to have a login on the console. |
| 19:48.13 | mickey|headache | Noodles: hmm odd. you have an updated bk ? |
| 19:48.20 | mickey|headache | Noodles: i goofed up fstab for a couple of hours |
| 19:48.23 | Noodles | Yeah. Pulled this morning. |
| 19:48.26 | mickey|headache | maybe the wrong slipped into your image |
| 19:48.27 | RP | mickey|headache: How would you improve the sound out of interest given we can't change the hardware? :) |
| 19:49.00 | mickey|headache | RP: a) improve latency for opening/closing the device b) improve sounds |
| 19:49.35 | mickey|headache | Noodles: I'll be doing a full rebuild from scratch tomorrow morning and flash that |
| 19:49.55 | Noodles | I don't think we want /var on a tmpfs. |
| 19:50.03 | RP | mickey|headache: I had noticed the latency. That can be improved... |
| 19:50.08 | Pigi | koen, ping |
| 19:50.53 | mickey|headache | RP: i can work on the sounds. I'm a musician, heh. At least I was one before I was dragged into OSS :D |
| 19:51.29 | woglinde | mickeyl *g* |
| 19:51.43 | Pigi | is there anyone with high level access on hh ? There seems to be some problem with cvs lock for kernel area. |
| 19:51.43 | mickey|headache | bbl |
| 19:51.49 | hrw | btw - can I have usb0 on desktop connected in one time to collie/2.4 and c7x0/2.6? |
| 19:52.02 | RP | mickey|headache: Sounds like we should manage some improvements between us then :) |
| 19:52.15 | koen | Pigi: pong |
| 19:52.29 | RP | hrw: Looking at that bt driver, note the cases statements are fallthroughs |
| 19:52.42 | Pigi | Do you have some way to fix right acces on /var/cvs on hh.org ? |
| 19:52.53 | RP | hrw: So it should be doing something upon a suspend/resume |
| 19:52.56 | hrw | RP: yep - added just for being there and doing nothing |
| 19:52.56 | koen | Pigi: nope |
| 19:53.08 | Pigi | ok. Will wait for france |
| 19:53.09 | koen | Pigi: france or jamey are able |
| 19:53.11 | RP | hrw: no, they do something |
| 19:53.27 | hrw | RP: but not enough to reinitialize card |
| 19:53.36 | RP | hrw: CS_EVENT_PM_SUSPEND is a CS_EVENT_RESET_PHYSICAL |
| 19:53.45 | hrw | ok |
| 19:53.53 | RP | CS_EVENT_PM_RESUME is a CS_EVENT_CARD_RESET |
| 19:54.07 | RP | Whether that's enough or not remains to be seen ;-) |
| 19:54.45 | hrw | its not enough |
| 19:59.12 | *** join/#oe molivier (~mac@f01v-24-82.d4.club-internet.fr) |
| 20:00.40 | hrw | argh.. it looks like my 256M sd card is not good for my c7x0/2.6 |
| 20:00.58 | hrw | mmcblk0: error 1 sending read/write command |
| 20:00.58 | hrw | end_request: I/O error, dev mmcblk0, sector 180741 |
| 20:01.03 | hrw | for whole card |
| 20:01.20 | hrw | ( |
| 20:04.49 | hrw | RP: any ideas? |
| 20:08.32 | hrw | will have to check that card in collie... |
| 20:10.03 | RP | hrw: I really don't know why it would do that. It could be a timing problem in which case you'll need to turn on mmc debudding and see there the error appears |
| 20:10.15 | RP | see *where* |
| 20:10.59 | *** join/#oe cyn (~cyn@cpe-66-61-101-26.midsouth.res.rr.com) |
| 20:11.19 | cyn | 'press into hole' means spend hours destroying your hands |
| 20:11.56 | cyn | the lcd mount was too low so I just stuck my lcd in normally, no biggs |
| 20:12.04 | hrw | need to recompile? |
| 20:12.20 | hrw | RP: I put card in usb card reader - works ok |
| 20:13.41 | hrw | written to full capacity.. syncing |
| 20:14.17 | hrw | zecke_: inherit patcher |
| 20:15.09 | XorA | re-evening |
| 20:15.20 | hrw | RP: for mmc debugging I need to recompile kernel or can enable it via echo>/sys/something? |
| 20:15.24 | hrw | hi XorA |
| 20:15.45 | RP | hrw: you need to recompile - its an option in configure |
| 20:15.54 | hrw | ah |
| 20:16.29 | RP | reevening XorA |
| 20:16.43 | RP | XorA: I presume you saw mickey|tv's comments? |
| 20:17.00 | XorA | RP: yeah I did, so Im good to go? |
| 20:17.10 | RP | XorA: I'd say so |
| 20:17.45 | hrw | RP: tomorrow will try (when find some time) |
| 20:17.57 | zecke_ | hrw: patcher is worse |
| 20:18.02 | zecke_ | but quilt new foo |
| 20:18.04 | zecke_ | Now in foo |
| 20:18.12 | zecke_ | but it didn't fucking refresh the old patch... |
| 20:18.13 | RP | hrw: Ok. If you send me the log, I'll try and see if anything jumps out at me |
| 20:18.47 | zecke_ | I'm about to stop developing the sqlite backend after having lost my changes due quilt... |
| 20:19.32 | RP | zecke_: I was looking forward to seeing them :-/ |
| 20:20.08 | hrw | bye all |
| 20:20.17 | hrw | RP: ok |
| 20:20.23 | zecke_ | RP: I need to redo the patches by hand now |
| 20:20.29 | zecke_ | I really hate broken tools |
| 20:20.33 | RP | hrw: bye |
| 20:20.43 | cyn | haha - soz -when i started typing my focus was in a diff channel (autoconnecting/etc.) |
| 20:20.45 | RP | zecke_: I hate it when that happens :-/ |
| 20:21.26 | zecke_ | fucking tool |
| 20:21.43 | XorA | RP: patch is in bk |
| 20:22.25 | zecke_ | fuck I hate two 8 hours of work... I really hate it. Next week I might have time to continue |
| 20:23.01 | XorA | CIA bot is down I see |
| 20:23.33 | *** join/#oe Laibsch (~leggewie@G19f1.g.pppool.de) |
| 20:23.46 | *** part/#oe Laibsch (~leggewie@G19f1.g.pppool.de) |
| 20:25.34 | RP | XorA: I'll bake some new images tomorrow and see how I get on |
| 20:26.08 | XorA | RP: Ive been running that since friday with no problems |
| 20:26.15 | RP | My current builds are way too hacked about for any more testing :) |
| 20:27.01 | XorA | RP: :-) |
| 20:27.24 | RP | I still can't work out what this blummin AtiCore_PolyScanline() is :-/ |
| 20:28.38 | koen | a blt buffer? |
| 20:28.41 | RP | What can you draw that takes a variable number of packets where each packet contains a number, and a variable number of what look like X,Y coords |
| 20:29.07 | RP | I've fed it values and it doesn't appear to do anything... |
| 20:32.00 | *** join/#oe SFred (~mteira@231.Red-83-42-134.pooles.rima-tde.net) |
| 20:34.35 | Bernardo | mickey|tv: the kernel for the akita has a fully populated keymap (at least it seems it is complete). |
| 20:36.03 | Bernardo | what ipk provides loadkeys? |
| 20:36.06 | SirFred | RP: Did you have any progress with polyscanline ? |
| 20:36.56 | RP | SirFred: hello. I was just talking about that. No, I still don't know what it does and its irritating me! :) |
| 20:37.19 | SirFred | RP: I think it should paint a whole horizontal line on the screen. |
| 20:37.26 | SirFred | RP: That's what an scanline is. |
| 20:37.45 | SirFred | RP: Perhaps you need to setup some surface as when blittering before using it. |
| 20:37.54 | RP | SirFred: Why all the numbers it takes then? |
| 20:38.11 | SirFred | RP: What do you mean with all the numbers? |
| 20:38.28 | RP | SirFred: Why does it take so many parameters? |
| 20:38.40 | RP | SirFred: If you can make it do anything, I'll be be impressed :) |
| 20:38.52 | SirFred | RP: Perhaps it needs the number of lines, the pitch,... |
| 20:39.42 | RP | SirFred: It doesn't fix the pattern of the parameters... |
| 20:41.02 | SirFred | RP: It's too hard just looking at that lot of params. |
| 20:41.12 | SirFred | RP: Have you seen this before? |
| 20:41.18 | SirFred | http://www.mnementh.co.uk/eseries/docs/imageon/ |
| 20:41.24 | RP | SirFred: I'd tend to agree |
| 20:41.45 | RP | SirFred: Yes, I have. Ian Molton has been doing a lot of work with me in getting w100fb to run on other platforms |
| 20:42.21 | SirFred | RP: Looking for a way to draw lines using a pattern; I've found this: |
| 20:42.31 | SirFred | fb_writel(0xffffffff, regs + mmDP_WRITE_MSK); |
| 20:42.43 | SirFred | So, that seems to be the pattern to apply to the line. |
| 20:43.14 | SirFred | But, unfortunately, I tried to find that mmDP_WRITE_MSK value used on the aticore assembler files, and found nothing. |
| 20:43.56 | SirFred | I've also found that AtiCore_BrushType is a really complex function. It calls to a lot of another functions. |
| 20:43.56 | RP | SirFred: We can probably add our own function to set that |
| 20:44.28 | RP | SirFred: Yes, I know. I'm avoiding the complex ones in the hope I learn from the simple ones first... |
| 20:44.57 | SirFred | RP: Well, I probed some values for AtiCore_BrushType but without luck. |
| 20:45.24 | RP | SirFred: You can write register values directly using some of the atiCore cuntions |
| 20:45.27 | RP | functions |
| 20:45.43 | SirFred | RP: Another thing I wanted to comment with you is the use of AtiCore_WaitComplete. |
| 20:46.05 | SirFred | RP: The qte driver is now calling that function for any operation on the imageon. |
| 20:46.19 | RP | SirFred: You can use ReadRegDirect and WriteRegDirect to set mmDP_WRITE_MSK in the shortterm |
| 20:46.54 | RP | AtiCore_WaitComplete will wait until the queue of w100 instructions is empty. Think of it as a sync function |
| 20:46.56 | SirFred | RP: I find that really suboptimal because you just need to wait for some fifo entries to the operation and not for the fifo to be empty. |
| 20:47.14 | RP | SirFred: So don't call that... |
| 20:47.44 | SirFred | RP: I suppose that any of the functions of aticore is waiting for the right amount of fifo entries to perform its task. |
| 20:48.00 | RP | SirFred: Correct - see the PM4_WaitForEntry() calls |
| 20:48.10 | SirFred | RP: Thanks for all this info. |
| 20:48.18 | RP | I'm guessing that waits until enough space in the fifo is available |
| 20:48.41 | SirFred | RP: I'm going to try that WriteRegDirect. |
| 20:49.23 | RP | SirFred: Keep in mind that I may not be 100% right about all this. I'm just giving you my best guessed based on what I've seen in the code :) |
| 20:49.38 | SirFred | RP: No problem. |
| 20:49.47 | SirFred | RP: I expect just not to burn the imageon. |
| 20:50.27 | RP | SirFred: That thought does worry me although more when I was messing with the PLL's rather than the acceleration code! :) |
| 20:52.33 | SirFred | RP: I see that WriteRegDirect takes 4 params. |
| 20:53.07 | SirFred | RP: I guess the first one always is 0, the second one is the register offset, the third one is the number of bytes to write and the fourth is .. |
| 20:53.13 | SirFred | the pointer to the source address? |
| 20:55.35 | RP | SirFred: I think you should use AtiCore_WriteReg instead - that will take a register offset and the data directly |
| 20:56.17 | RP | I was getting the functions mixed up - it ends up calling WriteRegDirect() |
| 20:56.43 | RP | It knows how to put commands in the fifo though if acceleration is running as well |
| 20:57.27 | SirFred | RP: Humm, nothing happened using WriteRegDirect |
| 20:57.32 | SirFred | Let's try with that other one. |
| 20:58.15 | RP | SirFred: You could always check it worked by calling AtiCore_ReadReg |
| 20:58.54 | SirFred | RP: Perhaps that value is changed by AtiCore_Polyline. |
| 21:00.51 | RP | SirFred: I'd very much doubt that. Its more likely it has no effect on Polyline |
| 21:01.10 | SirFred | RP: A pretty segfault. |
| 21:01.12 | SirFred | Humm. |
| 21:01.54 | SirFred | Perhaps WriteReg needs a pointer to the value and not the value directly? |
| 21:02.20 | RP | ReadReg will probably want a pointer to store the returned value in as the second argument |
| 21:02.49 | SirFred | RP: I think WriteReg also works that way. At least after giving a fast look at your code. |
| 21:03.35 | RP | SirFred: I think you may be right. The aticode always uses WriteRegDirect so WriteReg is untested and probably wrong |
| 21:03.56 | SirFred | RP: Well, now the register got written. |
| 21:04.00 | RP | It was from the very start of this project when I didn't know as much as I did now. Looking at it now, its looks wrong :) |
| 21:04.10 | SirFred | RP: But the lines have not different pattern. |
| 21:04.28 | SirFred | RP: I've tried with 0xf0f0f0f0 |
| 21:05.27 | RP | SirFred: Maybe you also have to tell it to use a patterned line somewhere |
| 21:05.49 | SirFred | RP: Yes, maybe. |
| 21:06.24 | SirFred | RP: Unfortunately, it's time to sleep. |
| 21:06.39 | SirFred | RP: Tomorrow I'll try to test this with more tome. |
| 21:06.52 | SirFred | RP: Tomorrow I'll try to test this with more time. And also take a look at that Scanline |
| 21:06.53 | RP | SirFred: Ok. I should really sleep soon as well :) |
| 21:07.17 | RP | Ok. I'm going to continue to decode functions and see what I can find... |
| 21:07.38 | RP | SirFred: 'night |
| 21:07.59 | SirFred | RP: OK. Good night. |
| 21:09.52 | Bernardo | mickey|tv: the kernel keymap is very complete (there's even support for the hinge in drivers/char/spitz_keyb.c), how do I use the output of dumpkeys in opie? |
| 21:11.19 | *** join/#oe imk (~dj@69-175-55-137.ontrca.adelphia.net) |
| 21:19.34 | zecke_ | ok now onto fighting with SQLite |
| 21:21.12 | RP | zecke_: How does it help things like speed and memory consumption? |
| 21:21.25 | zecke_ | RP: nothing yet |
| 21:21.34 | zecke_ | RP: it is a reorganisation of bb.data |
| 21:21.49 | zecke_ | RP: now one can plug 'any' persistent and non persistent data implementation there |
| 21:22.17 | RP | zecke_: That sounds like a nice feature and a very good first step :) |
| 21:22.44 | zecke_ | RP: yes, that looks like a good step |
| 21:23.09 | zecke_ | RP: speed wise build.py could be patched to not always execute all functions |
| 21:23.27 | zecke_ | RP: for example if I've a do_install() and call a python method |
| 21:23.43 | RP | zecke_: I long for the day I can start hacking threading :) |
| 21:23.47 | zecke_ | RP: this python method is called every time... |
| 21:24.14 | RP | zecke_: That does sound like there's room for improvement |
| 21:26.00 | koen | yeah, we should replace python with x86 and ppc asm |
| 21:26.40 | RP | koen: I'd settle for C ;-) |
| 21:28.05 | zecke_ | koen: all we need is a turing machine |
| 21:28.11 | zecke_ | koen: probably with two or more tapes |
| 21:28.21 | koen | heh |
| 21:28.23 | Pigi | time to sleep. |
| 21:28.27 | Pigi | Nite all |
| 21:28.28 | koen | 'night Pigi |
| 21:28.30 | zecke_ | cya |
| 21:28.43 | zecke_ | oh message too big |
| 21:29.00 | koen | it chokes on > 40kb iirc |
| 21:29.21 | zecke_ | koen: yupp I've still kmail from kde3.3 with kde3.4 I could compress mails on the fly :} |
| 21:29.36 | zecke_ | I get these patches via fish from my development machine... |
| 21:29.59 | koen | try putting the patch on hh.org/~zecke |
| 21:32.10 | zecke_ | http://handhelds.org/~zecke/downloads/oe/ |
| 21:33.11 | koen | that incorperates COW too? |
| 21:33.28 | zecke_ | koen: yes as one Data Implementation |
| 21:33.53 | zecke_ | cow is pretty much useless for persistent storage |
| 21:34.48 | zecke_ | now if python-sqlite wouldn't suck so much |
| 21:35.54 | RP | For all people rave about sqlite, I've never got on with it. Its SQL implementation is too limited/slow in my experience. Perhaps mysql has spoilt me... |
| 21:37.54 | zecke_ | RP: don't look at my table design... |
| 21:38.53 | RP | zecke_: I'm sure I've done worse without looking ;-) |
| 21:39.06 | *** join/#oe aboeglin (~aboeglin@213.154.79.4) |
| 21:40.56 | kergoth | zecke_: might want to take a look at what i did for bitbake-data for the rewrite. http://svn.berlios.de/viewcvs/bitbake/trunk/bitbake-data/doc/test_scoped_lookups.sql?rev=148&view=auto was a proof of concept setup for the sql tables. |
| 21:41.50 | zecke_ | kergoth: how do you store the bb.dirgraph? |
| 21:42.27 | kergoth | thats a persistant datastore that implements our scopes. that's all. |
| 21:43.19 | zecke_ | kergoth: so basicly it is responsible for saving itself? |
| 21:43.21 | kergoth | you'd use that to lookup a given variable or variables for any files that have been parsed. it doesnt store the graph, just the metadata. |
| 21:45.24 | zecke_ | kergoth: yeah that sounds right but might be not possible with bitbake 1.x |
| 21:46.25 | zecke_ | kergoth: I'll something that might suck... |
| 21:47.40 | kergoth | of course it is. you'd just have to add a call to the parser to store the info about what file the data is associated with in the instance created by bb.data.init(). that's the main difference, is that that sql setup requires that you know the file/recipe that's associated with each datastore. the bitbake 1.x datastores don't require that. |
| 21:48.00 | kergoth | brb, food |
| 21:50.10 | zecke_ | note to myself, stop forgetting writing words... |
| 21:50.58 | zecke_ | kergoth: yes but you do setVar('_task_...', a_bb.dirgraph(), data) |
| 21:51.16 | zecke_ | kergoth: and at that point you've a binary structure you need to store |
| 21:51.28 | zecke_ | kergoth: sure you can easily find out the internal data structure and save it... |
| 21:51.32 | kergoth | ahh right, not all the stuff stored is strings. |
| 21:51.51 | zecke_ | but imho using pickle for that is OKaay for bitbake 1.x external patches |
| 21:51.52 | kergoth | well, does that stuff really need to be persistant? those steps could be re-run on the data at start |
| 21:52.01 | kergoth | thats true |
| 21:53.20 | zecke_ | kergoth: now only python-sqlite randomly throws exceptions... |
| 21:53.27 | kergoth | ouch |
| 21:56.41 | ljp | zecke: whats the price of petrol in germany? |
| 21:59.37 | Bernardo | 'night |
| 21:59.46 | koen | 'night Bernardo|Away |
| 22:01.47 | zecke_ | ljp: fucking expensive, I'll look it up |
| 22:02.08 | zecke_ | ljp: writing a Qtopia Petrol Lookup Manager? |
| 22:04.22 | ljp | no. just curious and comparing prices in au and us and de |
| 22:04.57 | ljp | gas in us is about $.60 USD a litre |
| 22:05.01 | koen | the US go to war to keep the petrol cheap |
| 22:05.19 | koen | 2 euro/liter or so over here |
| 22:05.31 | koen | (the netherlands) |
| 22:06.50 | koen | ehm |
| 22:06.56 | koen | 1,4 E/l that is |
| 22:07.29 | zecke_ | ljp: javascript:fenster('http://www.aral.de/tankstelle/_home/index.cfm?tstnr_6=110109%20',800,600) |
| 22:07.53 | zecke_ | ljp: that is for berlin (west) but I currently do not drive... |
| 22:11.44 | XorA | hey mickeyl |
| 22:11.55 | ljp | wow.. german petrol is about 4x what US is |
| 22:11.58 | koen | zecke_: I lnow remember why my parents use to fill up in Germany :) |
| 22:12.22 | zecke_ | ljp: yupp |
| 22:12.29 | XorA | someone worked out the finest champange is cheaper than petrol in uk |
| 22:13.16 | koen | XorA: run your car on distilled cane sugar like the brazilians do |
| 22:13.39 | koen | 'night all |
| 22:14.18 | ljp | americans are whiners, hey :) |
| 22:18.50 | zecke_ | kergoth: did you try git yet? |
| 22:18.59 | kergoth | not yet |
| 22:25.19 | RP | ljp: 0.819GBP/L in the UK |
| 22:33.34 | zecke_ | kergoth: how are you? |
| 22:33.37 | Kristoffer | 1.4$/Liter in sweden |
| 22:47.01 | zecke_ | ~lart python-sqlite |
| 22:47.14 | zecke_ | I can't even create a fuckting table without getting an exception.. |
| 22:52.30 | reenoo_ | 'night all |
| 22:56.08 | zecke_ | cya |
| 22:56.30 | *** join/#oe mallum (~mallum@host81-153-51-218.range81-153.btcentralplus.com) |
| 23:16.46 | mickeyl | g'night |
| 23:17.00 | RP | 'night mickeyl |
| 23:30.43 | *** join/#oe xep (misha@porter-kresge-apts-57-234.resnet.ucsc.edu) |
| 23:32.12 | zecke_ | good nite |
| 23:37.40 | *** join/#oe [g2] (~g2@g2.nslu2-linux) |