| 05:33.38 | *** join/#arm-netbook SQlvpapir (~teis@188.177.95.62) |
| 10:46.31 | *** join/#arm-netbook lkcl (~lkcl@nat67.mia.three.co.uk) |
| 11:10.40 | *** join/#arm-netbook SQlvpapir (~teis@188.177.95.62) |
| 13:48.21 | lkcl | zecrazytux: you see the 20k pricing from chitech? $USD 102 |
| 13:50.27 | zecrazytux | lkcl: sorry ? |
| 13:50.40 | lkcl | $102 per unit, in volumes of 20,000 |
| 13:50.46 | zecrazytux | oh, ok |
| 13:51.07 | lkcl | haaaa |
| 13:51.20 | zecrazytux | but hackable-devices won't stock any stuff. there is no rooms where to store materials |
| 13:51.48 | zecrazytux | that's the problem with selling that kind of machines in europe |
| 13:52.03 | zecrazytux | pretty cheap :) |
| 13:52.49 | lkcl | 20,000 units is a hell of a lot. that's $USD 2m of stock! |
| 13:53.24 | lkcl | i'm uploading http://lkcl.net/arm_systems/CT-PC89E/rootfs.emdebian.tgz btw |
| 13:53.57 | lkcl | rsync says it's going to take 54 minutes |
| 13:54.06 | lkcl | 49 |
| 13:54.07 | lkcl | 50 |
| 13:54.22 | lkcl | 56 |
| 13:54.23 | lkcl | 53 |
| 13:55.28 | zecrazytux | yup but even a few hundreds is too much |
| 13:55.35 | zecrazytux | great |
| 13:55.57 | zecrazytux | did you note the tweaks that were needed to boot , |
| 13:56.03 | lkcl | yehh i know the usual thing: bury money into stock |
| 13:56.47 | lkcl | naah, i just followed wookey's advice. on the debian/lenny rootfs i'd done earlier, i simply added emdebian to /etc/apt/sources.list and did apt-get upgrade |
| 13:58.14 | zecrazytux | and it did actually upgrade *all* debian packages ? |
| 14:02.56 | lkcl | about .... 40% of them. |
| 14:04.10 | lkcl | that's the way emdebian "squeeze" works: they're directly equivalent (just using less space), so where some packages don't exist in emdebian, the standard debian/lenny ones will do perfectly well. |
| 14:10.05 | zecrazytux | yup. but I could get an entire grip/testing rootfs working under qemu without any package from mainstream debian |
| 14:11.50 | zecrazytux | I shoud "debug" the boot sequence with some echoes on fb0 as you did (I put some but it does not tel lme anything relevant for now) |
| 14:13.31 | lkcl | at least putting those in there it shows you which one of the /etc/init.d scripts has been reached. |
| 14:14.28 | lkcl | did you modify /etc/fstab to make it like the one in datang-epc.tar.gz ? |
| 14:18.02 | zecrazytux | yes |
| 14:20.58 | lkcl | do you at least get some pixels top-left by adding echo 11 > /dev/fb0 to /etc/init.d/rcS ? |
| 14:28.22 | zecrazytux | yes |
| 15:07.16 | mnemoc | lkcl: from last night's log of #s3cxxxx ... 23:31:55 < bjdooks> lkcl: sorry, was away 4th-5th and then didn't fancy spending another weekend in front of a compuer |
| 15:08.01 | mnemoc | lkcl: you had left the channel just a couple of minutes before |
| 15:24.33 | lkcl | ta mnemoc. i have to switch off the USB 3G connection |
| 15:32.03 | mnemoc | cu |
| 18:12.11 | bjdooks | as an fyi, i'm on/off for most of the next week or two, trying to get kitchen sorted out |
| 18:12.20 | bjdooks | then i'm possibly away for 6 weeks with work |
| 18:27.04 | lkcl | bjdooks: ack. no problem. i've run IDApro (legally-owned licensed copy) on the binary kernel, matched it up with /proc/ksymall function addresses: the key stuff (s3cfb_init_hw) etc is pretty much matching up. |
| 18:27.32 | lkcl | i'm just biding my time whilst waiting for the kernel source from the factory. |
| 18:27.43 | bjdooks | right |
| 18:28.01 | bjdooks | the sooner something can be submitted, the sooner we can clean up and get it ready for next kernel release |
| 18:30.57 | lkcl | ack |
| 18:32.56 | lkcl | frans reviewed the original samsung-generated 2.6.24.7 patch (on which we presume that chitech based their work, back-porting to 2.6.24.2) is, quoting but summarising fjp, "the usual commercial vendor mess ignoring standard linux kernel subdirectory/file conventions and including binary blobs" |
| 18:33.26 | mnemoc | what would be the point on backporting from .24.7 to .24.2?? |
| 18:33.41 | lkcl | but mer-smartq-kernel (chronolytics) attacked the same patch and have done a stellar job |
| 18:35.11 | lkcl | mnemoc: either lack of experience or being supplied with binary-only drivers which only work with a particular kernel, particular compiler etc. or having source code which you aren't being paid enough or given enough time to adjust to the ever-changing linux kernel datastructures and APIs. |
| 18:35.48 | mnemoc | but it's a microversion... |
| 18:36.01 | lkcl | shrugs |
| 18:36.03 | lkcl | :) |
| 18:37.20 | mnemoc | i find backporting so absurd in this case that i actually doubt it's the base they used |
| 18:40.37 | mnemoc | no that i discard that people can do absurd things |
| 19:22.40 | *** join/#arm-netbook SQlvpapir (~teis@188.177.95.62) |
| 20:05.24 | bjdooks | lkcl: sounds like it would be easier to start from scratch |
| 20:16.46 | bjdooks | if you need any help with it, let me know |
| 20:18.38 | mnemoc | i had the same feeling |
| 20:19.19 | mnemoc | bjdooks: upon next-samnsung? :) |
| 20:19.41 | bjdooks | i was hoping 2.6.34-rc1, but linus has just announced it without RMK's last pull |
| 20:20.23 | mnemoc | going for rc2 or .35? |
| 20:20.32 | bjdooks | no idea |
| 20:34.06 | *** join/#arm-netbook lkcl (~lkcl@nat67.mia.three.co.uk) |
| 20:35.42 | bjdooks | re |
| 20:36.07 | lkcl | arse. xchat segfaulted |
| 20:36.13 | bjdooks | bah, BT have just rewound their FTTC date back to now |
| 20:36.31 | bjdooks | I should have ordered that second line + fttc before I Was going off |
| 20:37.12 | lkcl | oops :) |
| 20:44.29 | bjdooks | it looks like i'll be away 19th March - 10th May |
| 20:50.18 | lkcl | bjdooks: ack. |
| 20:50.39 | bjdooks | six and a bit weeks of South Korea |
| 20:51.42 | lkcl | cool. |
| 22:48.33 | *** join/#arm-netbook Baybal (~baybal@S010600179a23239f.vc.shawcable.net) |