| 00:13.19 | dwery | x1205 is in git! |
| 00:15.10 | rwhitby-treo | woohoo! |
| 00:15.37 | rwhitby-treo | congrats dwery - is that your first patch to make it there? |
| 00:16.18 | dwery | umm.. not.. i think the gpio one liner was the first one :) (exluding the work i did years ago on other drivers...) |
| 00:24.00 | dwery | 'good night all |
| 01:16.00 | *** join/#nslu2-linux echange-m (n=change-m@h20.111.140.67.ip.alltel.net) |
| 01:18.37 | *** join/#nslu2-linux mechange- (n=change-m@h20.111.140.67.ip.alltel.net) |
| 01:41.35 | CIA-8 | 03jbowler 07org.openembedded.dev * rd59b05a4... 10/packages/linux/nslu2-kernel/2.6.14/10-ixp4xx-copy-from.patch: nslu2-kernel: comment update, no change to 2.6.14 |
| 01:41.38 | CIA-8 | 03jbowler 07org.openembedded.dev * r5de60fa1... 10/packages/linux/ (2 files in 2 dirs): nslu2-kernel: patch format, nslu2-power now uses ctrl-alt-del in 2.6.14 |
| 03:37.56 | *** join/#nslu2-linux senneth_ (i=senneth@irssi/staff/senneth) |
| 04:34.18 | *** join/#nslu2-linux ByronT (n=byron-po@nslu2-linux/ByronT) |
| 04:53.52 | *** join/#nslu2-linux mithro (n=tim@c213-100-42-188.swipnet.se) |
| 07:32.57 | *** join/#nslu2-linux odoc (i=odoc@is.borntobooze.de) |
| 08:31.33 | CIA-8 | 03noodles 07org.openembedded.dev * r8d8fd891... 10/packages/ppp/ (ppp-2.4.3/enable-ipv6.patch ppp_2.4.3.bb): ppp-2.4.3: Enable IPv6 support. |
| 08:36.21 | *** join/#nslu2-linux conradl (n=clancast@CPE-60-226-19-234.qld.bigpond.net.au) |
| 08:52.14 | *** join/#nslu2-linux caplink811_log (n=caplink8@dslb-084-059-068-075.pools.arcor-ip.net) |
| 09:02.09 | *** join/#nslu2-linux macsat (n=macsat@0x535b9779.arcnxx9.adsl-dhcp.tele.dk) |
| 09:11.58 | CIA-8 | 03noodles 07org.openembedded.dev * r9117a5cd... 10/packages/net-snmp/ (files/uclibc-fix.patch net-snmp_5.1.2.bb): net-snmp: Fix up compilation with uclibc. |
| 09:30.19 | *** join/#nslu2-linux AdamBaker (n=aab@userch028.dsl.pipex.com) |
| 10:08.54 | *** join/#nslu2-linux jelle (n=jelle@ip54518e8c.direct-adsl.nl) |
| 11:20.58 | CIA-8 | 03eFfeM 07org.openembedded.dev * r7a4b0ead... 10/packages/linux/ (2 files in 2 dirs): |
| 11:20.58 | CIA-8 | nslu2-kernel: added patch in pegasus.h to support Philips CPWUE01/00 network adapter. |
| 11:20.58 | CIA-8 | The maintainer of pegasus.h will put this change in the next official release. |
| 11:21.00 | CIA-8 | 03eFfeM 07org.openembedded.dev * rb706375f... 10/packages/linux/ (2 files in 2 dirs): disapproval of revision 7a4b0eade96c0806e41bfffb4807487e8005ae2a |
| 11:21.03 | CIA-8 | 03eFfeM 07org.openembedded.dev * rf827de78... 10/packages/linux/ (nslu2-kernel/2.6.14/90-pegasus.patch nslu2-kernel_2.6.14.bb): |
| 11:21.03 | CIA-8 | packages/linux/nslu2-kernel/2.6.14/90-pegasus.patch packages/linux/nslu2-kernel_2.6.14.bb: added patch to support Philips CPWUE01/00 usb network adapter |
| 11:21.03 | CIA-8 | This patch is already accepted by the pegasus.h maintainer for inclusion in the next mainstream release. |
| 12:02.54 | rwhitby | eFfeM: nice work! |
| 12:27.42 | eFfeM | rwhitby, tnx. it took quite some effort to go to the head again due to changes in /etc and due to 1.6.14 being a lower version nr than 1.6.14-rc5 |
| 12:28.48 | eFfeM | i'll probably need to update some wiki pages, but I decided to spent the previous hr in the garden, we're having kind of an indian summer here |
| 12:30.01 | eFfeM | still not sure what to do with pwc (for my philips webcam). It is supported by the latest version (10.0.9) but not by the version in 2.6.14 (which is 9.something) |
| 12:31.21 | eFfeM | for now I have hacked it in locally (by just overwriting tmp/work/something/linux2.6.14/drivers/usb/media/pwc and only compile but that is of course not the nice way |
| 12:31.39 | eFfeM | do we have a policy on how to deal with newer kernel modules? |
| 12:34.29 | rwhitby | isn't there a pwc package in OE ? |
| 12:35.34 | eFfeM | not sure let me check |
| 12:35.53 | eFfeM | ah, yes there is, it is at 10.0.7a |
| 12:37.01 | eFfeM | that's even better, I'll see if I can bring it up to the latest version later today (I'm sure dyoung-away as maintainer would not mind |
| 12:37.57 | eFfeM | I feel I'm beginning comfortable with small edits and clones for new versions within a package, but doing a package from scratch is still a road too far for me |
| 12:38.00 | dyoung-away | eFfeM: You have my blessing to modify any driver that I'm the maintainer of. |
| 12:38.25 | eFfeM | dyoung-away, tnx, I knew you wouldn't mind (you already said so with pvrusb2) |
| 12:38.56 | eFfeM | actually I'm trying to scribble a small recording daemon for pvrusb2 |
| 12:39.04 | dyoung-away | woo ! |
| 12:39.20 | eFfeM | mind you: 3small1 |
| 12:39.27 | dyoung-away | I tried hacking webvcrplus into shape, but I didnt have much luck there. |
| 12:40.22 | eFfeM | basically it is a proggie that reads a file called timers which only contains triplets preset,starttime, stoptime and a file channels that maps preset numbers to frequencies |
| 12:40.31 | dyoung-away | eFfeM: odoc was instrumental in helping test the pwc driver stuff. I just put it in, I dont actually own a pwc yet. :-) |
| 12:40.44 | dyoung-away | I meant to; but never got around to it |
| 12:41.35 | rwhitby | eFfeM: is there a trick to getting it to use 2.6.14 instead of 2.6.14-rc5 ? |
| 12:41.41 | eFfeM | the daemon then copies from /dev/video0 to the file (just fread/fwrite of 1k blocks, could not find out if mmap from /dev/video would actually work if it is an mpeg stream and decided also that this is a little bit too far |
| 12:41.54 | dwery | 2.6.14-git1 has been release |
| 12:41.57 | dwery | d |
| 12:42.06 | dwery | x1205 include. going to modify the cvs. |
| 12:42.39 | rwhitby | 2.6.14 or 2.6.15-git1 ? |
| 12:42.45 | eFfeM | rwhitby, are you referring to ipkg? I just did a -force-downgrade when installing the modules |
| 12:43.02 | rwhitby | eFfeM: oh, I was referring to building |
| 12:43.08 | eFfeM | dyoung-away, I got two 680's from a collegue who had them lying around |
| 12:43.23 | dwery | 14 |
| 12:43.33 | eFfeM | rwhitby, I pulled the head to my system then did bb openslug-image |
| 12:43.45 | rwhitby | so 2.6.14-git1 is *after* the 2.6.14 release ? |
| 12:44.02 | eFfeM | triggered by some advice of jbowler-zzz in #openslug |
| 12:45.29 | eFfeM | then flashed the img, disconnected the hd rebooted, mounted the hd, saved /dev, /etc etc so I could examine the old copies if needed and did a turnup -i -f followed by installing all modules I needed and bashing /etc/init.d again into shape |
| 12:45.29 | dyoung-away | dwery: did I just understand that x1205 is in the 2.6.14-git1 ? |
| 12:45.34 | eFfeM | all from my own builds |
| 12:45.34 | dyoung-away | thats awesome. |
| 12:45.45 | dwery | dyoung-away: yes. |
| 12:46.15 | dyoung-away | I hope my bug isnt propogated that far.... |
| 12:46.19 | dwery | dyoung-away: it is also the most recent version, with atrim and dtrim reading |
| 12:48.23 | eFfeM | dyoung-away, i'm a little bit stuck with version numbers. the latest pwc in the packages is 10.0.7a, the new one is 10.0.9-rc1 |
| 12:48.49 | eFfeM | will that work fine with making the rc1 as latest version? |
| 12:50.11 | dyoung-away | that should in theory work. |
| 12:50.19 | eFfeM | ok, i'll try |
| 12:50.31 | dyoung-away | as longn as its not a '_' |
| 12:51.24 | eFfeM | no it is a - |
| 12:59.45 | dwery | rwhitby: it seems so.. i thought it would have been called 2.6.15-git1 |
| 13:18.48 | *** join/#nslu2-linux Matta (n=matta@YYYDXL.dsl.saunalahti.fi) |
| 13:25.38 | oO[NOVA]Oo | perhaps someone can help me with an annoying nslu2 problem |
| 13:26.02 | oO[NOVA]Oo | my nslu2 keeps resetting the telnet access code to the linksys default |
| 13:26.35 | oO[NOVA]Oo | yes i started the nslu2 without harddrives attached, logged in with uNSLUng and got in fine. i then changed the password and copied the new passwd file to /share/hdd/conf/ |
| 13:26.51 | oO[NOVA]Oo | and it works then, for a while |
| 13:27.05 | oO[NOVA]Oo | until a reboot. then it all goes back to default. it's driving me crazy |
| 13:27.23 | rwhitby | install openssh or dropbear |
| 13:27.31 | oO[NOVA]Oo | i tried that |
| 13:27.38 | oO[NOVA]Oo | then i get an error that i should get a better entropy |
| 13:27.48 | oO[NOVA]Oo | and that the encryption appears to be 'blocked' |
| 14:03.52 | eFfeM | dyoung-away, are you still there |
| 14:04.23 | eFfeM | I tried to make pwc-10.0.9.rc1 and I failed miserably |
| 14:05.50 | eFfeM | I copied the directory and bb file from 7a, it downloads the new version, unpacks it and then barfs that it cannot find pwc/Makefile |
| 14:06.06 | eFfeM | The very same constructs are used in the 7a file and that works like a charm |
| 14:06.11 | eFfeM | what is wrong here? |
| 14:07.39 | eFfeM | anyone else a clue? |
| 14:09.58 | eFfeM | nevermind, I found it, made a typo in the filename ( _ i.s.o - and dyoung-away even warned me about it, silly me) |
| 14:17.58 | rwhitby | widrone: your debootstrap howto depends on running OpenSlug - it doesn't work on UcSlugC. |
| 14:23.37 | rwhitby | Hmm - I wonder how much smaller slugos-btg is than slugos-bag ... |
| 14:26.39 | CIA-8 | 03eFfeM 07org.openembedded.dev * r8bfbc133... 10/packages/pwc/ (pwc-10.0.9-rc1/Makefile pwc_10.0.9-rc1.bb): packages/pwc: added pwc_10.0.9-rc1 release |
| 14:39.39 | CIA-8 | 03azummo * 10kernel/2.6.14/30-i2c-x1205.patch: Removed x1205 i2c driver as it has been included in 2.6.14-git1. |
| 14:40.19 | dwery-away | current tag is NSLU2_2_6_14_git1_1 |
| 14:41.15 | rwhitby | dwery-away: do we really want to make that change in the 2.6.14 dir, or do we want to start a new 2.6.15 dir ? |
| 14:41.40 | dwery-away | given that the kernel is 2.6.41-git1 i would use the same dir |
| 14:42.23 | *** join/#nslu2-linux bullet (n=bullet@139.48.203.62.cust.bluewin.ch) |
| 14:42.25 | rwhitby | yes, but neither OE or debian will be using the git1 kernel, so I would start a new 2.6.15 dir for the patchset that will be used for 2.6.15 when it is released. |
| 14:42.45 | rwhitby | i.e. we want to stabilize on 2.6.14 released for a while. |
| 14:43.20 | dwery-away | dunno.. .15 directory with .14 tags? sounds confusing |
| 14:43.45 | rwhitby | .14 directory which doesn't apply against .14 released? even more confusing. |
| 14:43.56 | dwery-away | we should have a tag for .14 release maybe |
| 14:44.16 | rwhitby | ok, how about a 2.6.14-git directory then ? |
| 14:44.17 | dwery-away | it would be also fine to have archives |
| 14:44.26 | dwery-away | i.e. tar.gz |
| 14:44.41 | dwery-away | with the patches that needs to be applied to a specific kernel version |
| 14:44.55 | dwery-away | so that joe user does not need to |
| 14:44.59 | dwery-away | learn cvs tags usage |
| 14:45.15 | rwhitby | this repo isn't targetted at joe user |
| 14:45.28 | dwery-away | exactly.. we could have archives for that |
| 14:45.38 | rwhitby | basically, I'd like to have the directory name related to the .tar.gz from kernel.org |
| 14:46.12 | rwhitby | what will be the next .tar.gz from kernel.org after 2.6.14.tar.gz ? |
| 14:46.25 | dwery-away | I really hope 2.6.15 :D |
| 14:47.25 | rwhitby | that's why I'd like the new dir to be called 2.6.14-git1 or 2.6.15, but not to change the 2.6.14 files because 2.6.14 has been released. |
| 14:48.27 | rwhitby | (because we may decide to change patches that are applied against 2.6.14 released, independently of patches that are targetted for 2.6.15. |
| 14:48.27 | dwery-away | ok.. i'll try to revert the change |
| 14:48.52 | dwery-away | do you have a copy of the repo just before my change? |
| 14:49.06 | dwery-away | if yes, could you tag it with NSLU2_2_6_14 ? |
| 14:49.11 | rwhitby | you should be able to just cvs add the file back again and then cvs commit |
| 14:49.28 | dwery-away | right.. |
| 14:50.04 | rwhitby | ok, I tagged it |
| 14:54.23 | dwery-away | a cvs add on 30-x1205 will do? |
| 14:54.31 | rwhitby | should do |
| 14:54.37 | rwhitby | it should get it back from the Attic |
| 14:55.20 | dwery-away | cvs add: 30-i2c-x1205.patch already exists, with version number 1.2 |
| 14:55.43 | dwery-away | there must be a way to inform cvs that 1.2 is HEAD |
| 14:56.59 | CIA-8 | 03rwhitby * 10kernel/2.6.14/30-i2c-x1205.patch: Resurrected |
| 14:57.23 | rwhitby | that should do it |
| 14:57.30 | dwery-away | did you used cvs commit -r HEAD 30-i2c-x1205.patch ? |
| 14:57.56 | rwhitby | I just did cvs up ; cvs add 2.6.14/30-i2c-x1205.patch ; cvs commit |
| 14:58.06 | dwery-away | ok |
| 14:58.35 | dwery-away | should I go with 2.6.14-git or 2.6.15 ? |
| 14:59.04 | rwhitby | hmm |
| 14:59.24 | rwhitby | 2.6.14-git - then it's crystal clear |
| 15:00.00 | dwery-away | ok |
| 15:02.40 | eFfeM | hi can any of you perhaps point me to some info on how entries in /dev are created in openslug with 2.6.14 kernel? |
| 15:02.40 | eFfeM | When I plugin my webcam it automatically creates /dev/video0, but no audio device, although the cam also has a mic |
| 15:02.51 | eFfeM | and /proc/bus/usb/devices says I: If#= 2 Alt= 4 #EPs= 1 Cls=01(audio) Sub=02 Prot=00 Driver=snd-usb-audio |
| 15:03.03 | eFfeM | and snd-usb-audio is of course loaded |
| 15:03.12 | rwhitby | eFfeM: I think we're using udev now |
| 15:03.54 | eFfeM | yes and udev is making /dev/video0 for me. /dev/video0 is not there when the cam is disconnected and is created when I connect it, but there is no audio dev created |
| 15:04.06 | eFfeM | so that kind-a confused me |
| 15:09.36 | CIA-8 | 03azummo * 10kernel/2.6.14-git/ (16 files): |
| 15:09.36 | CIA-8 | 2.6.14-git1, initial import. |
| 15:09.36 | CIA-8 | 30-i2c-x1205.pathc is now mainstream. |
| 15:09.54 | rwhitby | dwery-away: thx |
| 15:12.17 | CIA-8 | 03azummo * 10kernel/2.6.14-git/defconfig: |
| 15:12.17 | CIA-8 | Enabled the new x1205 driver. |
| 15:12.17 | CIA-8 | Changed CMDLINE to cope with it. |
| 15:18.01 | rwhitby | widrone: when following the Bootstrap HowTo, I get the following after turnup disk to the debootstrapped partition: |
| 15:18.15 | rwhitby | Starting periodic command scheduler: cron. |
| 15:18.16 | rwhitby | INIT: Id "1" respawning too fast: disabled for 5 minutes |
| 15:18.16 | rwhitby | INIT: Id "2" respawning too fast: disabled for 5 minutes |
| 15:18.16 | rwhitby | INIT: Id "3" respawning too fast: disabled for 5 minutes |
| 15:18.16 | rwhitby | INIT: Id "4" respawning too fast: disabled for 5 minutes |
| 15:18.17 | rwhitby | INIT: Id "5" respawning too fast: disabled for 5 minutes |
| 15:18.18 | rwhitby | INIT: Id "6" respawning too fast: disabled for 5 minutes |
| 15:18.20 | rwhitby | INIT: no more processes left in this runlevel |
| 15:18.39 | rwhitby | will investigate further tomorrow |
| 15:31.03 | eFfeM | silly q, for some reason I can only ssh to my slug as root after upgrading and not as a regular user (although the user does exist in /etc/passwd |
| 15:33.53 | eFfeM | this must be a dumb user error but I cannot find an answer in the wiki about this and it used to work in the past .... |
| 15:46.51 | dwery-away | ARGH! 2.6.14-git1 is crashing badly: http://rafb.net/paste/results/n9f99212.html |
| 15:58.50 | dwery-away | i think i'll need help to debug it.. |
| 16:00.00 | *** join/#nslu2-linux Tiersten_ (i=tman@nslu2-linux/Tiersten) |
| 16:06.44 | *** join/#nslu2-linux mcfly587 (n=mcfly587@17.227-200-80.adsl.skynet.be) |
| 16:57.31 | jbowler-zzz | rwhitby-asleep, dwery-away: 2.6.14-git is, I think, 2.6.15, but somewhere else is the fork for 2.6.14.1 |
| 16:58.02 | jbowler-zzz | And 2.6.14-gitX is going to be majorly unstable because it is open for massive changes (if I understand this correctly) |
| 16:59.04 | jbowler-zzz | The major question in my mind is whether we will have to move to 2.6.14.1 because of some serious generic or ARM issue, I hope not... |
| 17:14.34 | *** join/#nslu2-linux NSLU2-IRC (n=NSLU2-IR@209.242.5.110) |
| 17:40.47 | eFfeM | <PROTECTED> |
| 17:51.04 | *** part/#nslu2-linux odoc (i=odoc@is.borntobooze.de) |
| 18:00.33 | *** join/#nslu2-linux bullet (n=bullet@139.48.203.62.cust.bluewin.ch) |
| 18:06.50 | dwery-away | jbowler-zzz: We should maybe test it under x86, but I can't do that on my machines right now. |
| 18:39.17 | *** join/#nslu2-linux Hurray (n=NovaFlas@sev.demon.nl) |
| 18:41.05 | *** join/#nslu2-linux bullet (n=bullet@139.48.203.62.cust.bluewin.ch) |
| 18:52.14 | *** join/#nslu2-linux caplink811-log (n=caplink8@dslb-084-059-096-060.pools.arcor-ip.net) |
| 19:55.42 | *** join/#nslu2-linux mcfly587_ (n=mcfly587@82-120.240.81.adsl.skynet.be) |
| 20:39.49 | *** join/#nslu2-linux mcfly587_ (n=mcfly587@99-115.240.81.adsl.skynet.be) |
| 20:48.25 | rwhitby-asleep | jbowler-zzz: from here on, I'd like to use the Debian kernel source packages somehow for the kernels - the main reason being that Debian keeps the modules path at /lib/modules/2.6.14/, even for kernel source that is based on 2.6.14.1 |
| 20:57.37 | *** join/#nslu2-linux caplink811_log (n=caplink8@dslb-084-059-090-195.pools.arcor-ip.net) |
| 21:26.01 | *** join/#nslu2-linux jacques (n=username@nslu2-linux/jacques) |
| 21:28.10 | *** join/#nslu2-linux mcfly587_ (n=mcfly587@170.147-200-80.adsl.skynet.be) |
| 21:50.48 | *** join/#nslu2-linux caplink811-log (n=caplink8@dslb-084-059-074-006.pools.arcor-ip.net) |
| 22:20.16 | jbowler-zzz | rwhitby-asleep: it's not a kernel issue, the kernel doesn't care where a module is in the file system, it's the module utilities. |
| 22:21.35 | CIA-8 | 03rpurdie 07org.openembedded.dev * rb3e2a703... 10/packages/linux/linux-openzaurus_2.6.14-rc4.bb: linux-oz-2.6: Apply some SoC Alsa updates from Liam to dev kernel. Some moves forward, some regressions... |
| 22:21.39 | CIA-8 | 03rpurdie 07org.openembedded.dev * ree91eab4... 10/packages/linux/ (2 files in 2 dirs): |
| 22:21.40 | CIA-8 | linux-oz-2.6: Update dev cxx00 kernel to include wm8750. I'm pleased to announce |
| 22:21.40 | CIA-8 | that sound has been heard on spitz through alsa with this driver with some |
| 22:21.40 | CIA-8 | trickery :). Also fix the SADIV bug in the i2s layer so c7x0 doesn't suffer from |
| 22:21.40 | CIA-8 | whale music. |
| 22:28.26 | rwhitby | jbowler-zzz: but don't those utilities ask the kernel what version it is? and debian kernels always returned 2.6.12 whilst OE kernels returned 2.6.12.2, even though they both had the same patchsets |
| 22:28.38 | rwhitby | (the same -2 kernel patchsets that is) |
| 22:32.13 | kolla_ | isnt it what sysctl says that counts? :) |
| 22:32.18 | kolla_ | sysctl kernel.osrelease |
| 22:33.07 | rwhitby | that's just asking the kernel too, right? |
| 22:34.05 | kolla_ | yes |
| 22:39.16 | *** join/#nslu2-linux conradl (n=clancast@CPE-60-226-19-234.qld.bigpond.net.au) |
| 23:22.10 | jbowler-zzz | rwhitby: UTS_RELEASE in linux/include/linux/version.h, which seems to be a generated file |
| 23:26.17 | jbowler-zzz | rwhitby: it comes from KERNELRELEASE in the top level Makefile and the problem is that EXTRAVERSION is set, as I remember it a Debian build can set this too, along with LOCALVERSION - it's up to the user when the kernel is build from the Debian package. |
| 23:27.14 | jbowler-zzz | (I used to routinely change these, and the expectation is that there will be independent module trees, because part of the point is to be able to move things in and out of the kernel). |
| 23:28.24 | jbowler-zzz | So what we do is to set EXTRAVERSION="" (along with LOCALVERSION) in the arguments to make... |
| 23:50.30 | rwhitby | jbowler-zzz: that must be what the debian make-kpkg program is doing when it is used to build a debian kernel |
| 23:53.42 | rwhitby | since we have total freedom in OE, I'd like to make our OE kernel configuration match the debian kernel configuration so that we can share kernels. |
| 23:55.50 | jbowler-zzz | rwhitby: I was looking at kernel.bbclass, it should be pretty straightforward to kill EXTRAVERSION, just add to EXTRA_OEMAKE. |
| 23:57.28 | jbowler-zzz | rwhitby: in NSLU2 it makes no sense to have multiple /lib/modules/<foo> directories because they simply won't fit, so there's no reason to have anything other than the kernel version (2.6.14) in there, indeed, we could even live without that |
| 23:59.03 | rwhitby | ok, so we'll stick with /lib/modules/2.6.14/ until 2.6.15 comes out (and we decide to move to it) ? |
| 23:59.35 | jbowler-zzz | Because we can only have one set of modules at once (reliably) it does not harm to not have versions - indeed it is better because the old modules might work on the new kernel... |