| 00:12.17 | rwhitby | monotone.nslu2-linux.org is updated to monotone 0.20 |
| 00:12.22 | rwhitby | all clients must update to 0.20 |
| 00:13.03 | rwhitby | Do: "monotone -d monotone/nslu2-linux.db pull monotone.nslu2-linux.org org.openembedded.* org.nslu2-linux.*" to update your database |
| 00:13.20 | rwhitby | Master Makefile has also been updated. |
| 00:17.39 | *** join/#nslu2-linux DaKa2 (~david@193.13.58.211) |
| 00:32.10 | rwhitby | DaKa2: can you test the monotone upgrade? |
| 00:32.43 | DaKa2 | monotone upgrade? |
| 00:33.12 | DaKa2 | ahh, 0.20? |
| 00:34.22 | rwhitby | yep |
| 00:34.26 | DaKa2 | fantastic, debian has 0.20 packaged |
| 00:34.46 | rwhitby | Do: "monotone -d monotone/nslu2-linux.db pull monotone.nslu2-linux.org org.openembedded.* org.nslu2-linux.*" to update your database after you have installed 0.20 |
| 00:34.47 | DaKa2 | hm.. new syntax? |
| 00:35.05 | rwhitby | gotta go - pass the word on to others who come here today |
| 00:35.30 | DaKa2 | ok, got to sleep in an hour, but sure |
| 00:55.09 | *** join/#nslu2-linux ByronT (~byron-por@ByronT.nslu2-linux) |
| 01:01.19 | *** join/#nslu2-linux rwhitby-treo (~upirc@rwhitby.nslu2-linux) |
| 01:01.54 | rwhitby-treo | anyone updated to 0.20 monotone yet? |
| 01:02.04 | DaKa2 | well, me |
| 01:02.23 | rwhitby-treo | it worked fine? |
| 01:02.28 | DaKa2 | yes |
| 01:02.33 | DaKa2 | atleast a pull |
| 01:03.07 | DaKa2 | make update works after updating the makefile |
| 01:03.25 | rwhitby-treo | ok, good. thx |
| 01:11.11 | rwhitby-treo | ok, back later ... |
| 01:11.30 | *** part/#nslu2-linux rwhitby-treo (~upirc@rwhitby.nslu2-linux) |
| 01:31.47 | jbowler | If anything 0.20 seems to be slightly slower, but that may just be me - I haven't done a scientific test and the db is larger than it was when I tried 0.19 on a slug. |
| 01:32.04 | jbowler | There is, however, a progress indicator, of a sort. |
| 01:32.40 | DaKa2 | I saw that, atleast better than before |
| 01:33.08 | DaKa2 | I just had 15 revs to download, so I dont know about speed |
| 01:36.36 | jbowler | Impossible to measure using monotone.nslu2-linux.dev - to do a real speed test it would be necessary to copy the db file and serve/sync using both 0.19 and 0.20 |
| 01:36.54 | jbowler | The glob syntax is the bash extended one by the looks of things: 'org.{openembedded,nslu2-linux}.*' works. |
| 01:40.58 | jbowler | It doesn't give progress on the "revs out", so there is a push if there is a lot to push back to the server (not typically the case). |
| 01:41.04 | jbowler | s/push/pause/ |
| 01:44.53 | *** join/#nslu2-linux nslu2-log (dyoung@dyoung.nslu2-linux) |
| 01:49.07 | *** join/#nslu2-linux SpanKY (~UserBah@pool-141-154-233-111.bos.east.verizon.net) |
| 01:49.24 | SpanKY | jbowler: hey, feel like reviewing patches for the armeb stat issue again ? :) |
| 01:49.55 | SpanKY | this version i did some testing with and it works on patched nslu2-2.4.22 and it should work on unpatched 2.4 and 2.6 |
| 02:26.02 | jbowler | Can you put it up on pastebin? (I.e. the patch) |
| 02:27.35 | SpanKY | http://dev.gentoo.org/~vapier/armeb-stat-abi.patch |
| 02:28.17 | SpanKY | mine should fail only if the value in st_dev requires more than 4 bytes to describe the device node |
| 02:28.24 | SpanKY | which, atm, i dont think anyone does ... |
| 02:29.53 | SpanKY | especially since all device nodes atm fit into like 2 bytes :) |
| 02:31.45 | jbowler | Ok, so you are reassembling the 8 byte st_dev in the 2.6 kernel from two shorts and the first 4 bytes of a long long |
| 02:32.18 | SpanKY | yeah |
| 02:32.36 | jbowler | That's weird - you only take the low 4 bytes of the st_dev the kernel provides: |
| 02:32.36 | SpanKY | in case the arm guys ever get their act together and re-order the padding/st_dev in the future :) |
| 02:32.58 | jbowler | buf->st_dev = kbuf->st_dev >> 32; |
| 02:33.52 | jbowler | Ok, but why not just do it as unsigned long, unsigned long (obviating the need for the >>32)? |
| 02:34.11 | SpanKY | hmm, i could xor it with << shifted pads |
| 02:34.24 | SpanKY | that'd give us back 2 bytes from the pad01 |
| 02:35.04 | jbowler | A union would work. |
| 02:35.14 | SpanKY | + unsigned short __pad00; |
| 02:35.14 | SpanKY | + unsigned short __pad01; |
| 02:35.14 | SpanKY | + unsigned long st_dev; |
| 02:35.14 | SpanKY | + unsigned long __pad02; |
| 02:35.15 | SpanKY | ? |
| 02:35.42 | SpanKY | i thought about union's but couldnt come up with one that didnt suck let alone work nicely |
| 02:36.27 | SpanKY | do you mean like union { unsigned long a; unsigned long long b; } and then just doing b = a ? |
| 02:37.30 | jbowler | union { unsigned short __small; unsigned long long __large } st_rdev - but it doesn't work in the header. |
| 02:37.53 | jbowler | (i.e. someone who actually uses kernel_stat.h will have code which no longer compiles). |
| 02:38.04 | SpanKY | no one should use kernel_stat.h ... |
| 02:38.11 | SpanKY | especially since it's a uClibcism |
| 02:39.08 | jbowler | Hum. Well, then a union of the new struct and the old (effectively) seems to make it clearer what's going on - but so far as I can see what you have now works, and it works with the version above too. |
| 02:40.17 | SpanKY | a union of structs ... hadnt thought of that ... |
| 02:40.38 | jbowler | I had not either unless 30s ago. |
| 02:40.55 | SpanKY | we could use st_blocks to check whether we have new ABI or old |
| 02:41.05 | SpanKY | since its meaning is swapped with the pad between 2.4 and 2.6 |
| 02:41.28 | SpanKY | hrm well that'd fail if st_blocks was actually 0 |
| 02:41.43 | jbowler | I guess the test is that if the new (2.6) kernel stat64 has st_dev or st_rdev > (1<<48) |
| 02:41.57 | jbowler | >= (1U << 48) |
| 02:42.24 | SpanKY | 2.6 has unsigned long long st_dev; unsigned char __pad0[4]; |
| 02:43.12 | SpanKY | which i think is why nslu2 patches its kernel / uclibc the way it does |
| 02:43.29 | SpanKY | it changes the ABI of 2.4 kernels so its compat with 2.6 |
| 02:43.37 | jbowler | And 2.4 unsigned short, unsigned char [10]. I think if the unsigned short == 0 use the unsigned long. |
| 02:45.25 | jbowler | That's the LinkSys patch is it? It has some clear advantages... |
| 02:45.49 | SpanKY | if (unsigned short != 0 && unsigned long == 0) { abi 2.4 } else { abi 2.6 } |
| 02:46.03 | SpanKY | 2.4.22-xfs-nslu2.patch.bz2 |
| 02:46.18 | SpanKY | it re-orders stat64 in asm-arm/stat.h |
| 02:46.41 | jbowler | unsigned long long above, not unsigned long |
| 02:46.43 | SpanKY | that's why i was having so many problems trying to test originally, i didnt know it was doing that to me heh |
| 02:47.37 | jbowler | That may be what fooled me too - I'm fairly sure I checked unslung, at least that I checked the source code. |
| 02:48.31 | SpanKY | hmm, and as you mentioned on the list, there's no way to cleanly detect abi if the thing being stat-ed is mounted on a 0,0 device |
| 02:48.47 | SpanKY | but oh well, not a corner case we have to worry about ... we'll just err on the side of newer ABI |
| 02:48.57 | jbowler | I think I checked what that was and it isn't something which can be mounted. |
| 02:49.44 | SpanKY | yeah, i think we're just being overly anal at that point :) |
| 02:50.50 | jbowler | Anyway, it doesn't matter because if st_dev is 0 then both structures have 0 in st_dev... |
| 02:54.27 | jbowler | Ok, so the patch you have works, so far as I can see, and a test of the form using a union of the two kernel stat64 structures reliably detects 2.6 so long as st_rdev < (1U<<48) and works on patched 2.4 too - so that is a good test too. |
| 02:56.30 | SpanKY | refresh the patch |
| 02:56.51 | SpanKY | it uses a union and it works for me on patched 2.4.22 kernel (aka 2.6/new abi) |
| 02:57.09 | SpanKY | also looks much nicer :) |
| 02:58.20 | SpanKY | the corner case it fails on now is 0,0 device on a 2.4 (aka old abi) kernel |
| 02:58.21 | jbowler | Hum, Konqueror gives me KWrite and no way to refresh, I'll use wget. |
| 02:58.46 | SpanKY | but that's ok with me, i'm a 2.6 wh0re |
| 03:01.22 | jbowler | Yes, that's much clearer (i.e. in that patch the st_dev handling is much clearer than the st_rdev handling). |
| 03:02.23 | jbowler | You could test for !st_dev.old_abi && !st_rdev.old_abi |
| 03:02.48 | SpanKY | well that would only affect special files |
| 03:02.54 | SpanKY | regular files would fail the rdev test |
| 03:03.14 | SpanKY | and i'd say a good 99% of the time you deal with regular files ... |
| 03:03.40 | jbowler | Do devices have a non-zero st_blocks ever? |
| 03:04.25 | SpanKY | going by the output of `stat -c %b /dev/* | sort -u` i'd say no |
| 03:04.51 | SpanKY | but directories give back 0 blocks ... |
| 03:04.54 | jbowler | The test case is /dev/full |
| 03:05.47 | SpanKY | i say we live with just the 2.4 / 0,0 error |
| 03:06.35 | jbowler | Yes. |
| 03:06.52 | SpanKY | ok, i'm gonna tag it and bag it then |
| 03:07.02 | SpanKY | thanks a lot :) |
| 03:07.03 | jbowler | I have /dev/full as 0,0, but I get 'no such device or address' when I try to open it ;-) |
| 03:08.03 | jbowler | Thanks for spending so much time looking into it (particularly given that the 2.4 patch is so specific to NSLU2!) |
| 04:09.00 | SpanKY | jbowler-away: heh, and after all that friggin effort, it looks like it's because a patch was lost in Jan 2002 to fix linux-2.4 |
| 04:09.10 | SpanKY | http://www.arm.linux.org.uk/developer/patches/viewpatch.php?id=863/1 |
| 04:09.33 | SpanKY | i'm thinking about reverting all of our diffs and just go with the 'correct' ABI |
| 06:00.13 | *** join/#nslu2-linux ka6sox (www-data@ka6sox.nslu2-linux) |
| 08:15.28 | *** part/#nslu2-linux dyoung (~dyoung@dyoung.nslu2-linux) |
| 08:58.23 | *** join/#nslu2-linux dyoung (~dyoung@dyoung.nslu2-linux) |
| 09:25.42 | *** join/#nslu2-linux eumel (~chatzilla@p54B6AB72.dip0.t-ipconnect.de) |
| 10:41.39 | *** join/#nslu2-linux llagendijk (~louis@lagendijk.xs4all.nl) |
| 11:26.23 | *** join/#nslu2-linux [g2] (~g2@g2.nslu2-linux) |
| 12:13.11 | rwhitby | Anyone seen one of these: http://www.myfmstation.com/html/linex_usb.html ? |
| 12:18.55 | [cc]smart | argh, what this now: monotone: warning: caught bad_decode exception decoding input from peer monotone.nslu2-linux.org: 'protocol version mismatch: wanted '4' got '5'' |
| 12:19.58 | [cc]smart | rwhitby: i don't see sense in this |
| 12:20.10 | [cc]smart | you connect it to you PC via usb |
| 12:20.18 | [cc]smart | probably means your PC should be switched on too. |
| 12:20.20 | [cc]smart | :) |
| 12:20.28 | [cc]smart | so why not have you pc stream it ? |
| 12:20.34 | dyoung | huh? |
| 12:20.49 | rwhitby | [cc]smart: did you see the messages here and on the mailing list about the monotone server upgrade? |
| 12:20.51 | dyoung | stream stuff from slug over FM radio? |
| 12:21.03 | [cc]smart | ah |
| 12:21.04 | rwhitby | [cc]smart: connect it to the slug instead of a PC |
| 12:21.05 | [cc]smart | eek |
| 12:21.06 | [cc]smart | stop |
| 12:21.18 | [cc]smart | misunderstood what its about |
| 12:21.21 | [cc]smart | :D |
| 12:21.38 | [cc]smart | slug could stream it just as well |
| 12:21.54 | [cc]smart | but it's the FM part rwhitby is interested in |
| 12:22.07 | rwhitby | yes, and this is how the slug becomes a radio transmitter |
| 12:22.28 | [cc]smart | i'm ot sure though if i thought this is really what i'd want |
| 12:22.39 | [cc]smart | if you want the sound for your fixed stereo |
| 12:22.53 | [cc]smart | you'd prolly better use on of those USB audio cards |
| 12:23.19 | [cc]smart | if you look at the mbile side, add wireless headphones :) |
| 12:24.19 | [cc]smart | what i was about to try is adding a USB soundcard to the slug myself |
| 12:24.28 | [cc]smart | and connect that to the stereo |
| 12:24.32 | [cc]smart | what is missing is the interface |
| 12:24.43 | [cc]smart | i'v got dect telephone handsets here |
| 12:24.47 | [cc]smart | the support wap |
| 12:25.10 | [cc]smart | so i thought, nifty as is, i'd use the asterisk instalation (this is where the dect phones go to) |
| 12:25.19 | rwhitby | dyoung: did you bump the PR? |
| 12:25.22 | [cc]smart | i'd set up my own mini isp with asterisk ppp support |
| 12:25.37 | dyoung | nope |
| 12:25.47 | [cc]smart | so i could use the dect/wap telephone handsets to control music playing of the slug |
| 12:25.58 | [cc]smart | but i couldn't make asterisk and handesets work together so far |
| 12:26.17 | [cc]smart | so i left the usb soundcard journey out |
| 12:26.30 | dyoung | PR bump on openslug-kernel is it? |
| 12:27.28 | rwhitby | probably on nslu2-kernel |
| 12:28.45 | dyoung | bumped. |
| 12:29.03 | rwhitby | heh, me too - let's see what monotone does with it ... |
| 12:30.24 | dyoung | dual heads eh |
| 12:30.42 | dyoung | but theres only one rev... |
| 12:31.56 | rwhitby | dyoung: the master makefile is set up to handle multiple heads |
| 12:33.12 | dyoung | okay. So are we okay? the file is in there? |
| 12:34.10 | rwhitby | should be in there, and the PR should be bumped too |
| 12:35.00 | dyoung | Okay cool, because my brain turned off at least an hour ago. |
| 12:35.57 | rwhitby | just fixing CIA now |
| 12:36.24 | rwhitby | ok, who created org.openembedded.nslu2-linux.testing ????? |
| 12:36.58 | rwhitby | hmm - maybe it's just in the CIA database ... |
| 12:42.40 | CIA-8 | 03rwhitby 07org.nslu2-linux.dev * r0bf7a652... 10/Makefile: Updated for monotone 0.20 |
| 12:42.55 | CIA-8 | 03dyoung 07org.openembedded.nslu2-linux * rd20d02a1... 10/packages/linux/nslu2-kernel/2.6.12/nslu2-setup.c: Change it to the new serial driver model, enabling serial console |
| 12:43.00 | CIA-8 | 03rwhitby 07org.openembedded.nslu2-linux * r64742a62... 10/packages/linux/nslu2-kernel_2.6.12.2.bb: |
| 12:43.00 | CIA-8 | Bumped the PR |
| 12:43.00 | CIA-8 | bump the pr |
| 12:43.53 | rwhitby | ok, CIA should be working again now |
| 12:44.00 | rwhitby | just had to update it for 0.20 syntax |
| 12:47.13 | NAiL | ah |
| 12:47.21 | NAiL | serial is now working in 2.6.12.2? |
| 12:47.30 | NAiL | (.2) |
| 12:47.57 | rwhitby | NAiL: just about to verify that. dyoung vouches for it, which is good enough for ,e |
| 12:47.59 | rwhitby | me |
| 12:48.03 | NAiL | yeah |
| 12:48.10 | NAiL | I'll take a look at that tomorrow |
| 12:48.17 | NAiL | I'm gonna flash both my slugs |
| 12:49.02 | dyoung | it worked for me, but the brain is so dead now, I need verification. |
| 12:49.17 | dyoung | (verify I'm not hallucinating....) |
| 12:49.29 | NAiL | dyoung: you are sleepy. your head feels heavy... |
| 12:49.41 | NAiL | you are unable to keep your eyelids open |
| 12:50.01 | dyoung | aoieucrpow oavwog |
| 12:50.03 | dyoung | ow. |
| 12:50.12 | NAiL | Your head hitting the keyboard? ;) |
| 12:50.37 | senneth | postfix with dovecot would be nice in openslug :] |
| 12:50.43 | NAiL | dovecot? |
| 12:50.54 | senneth | teh best imap server ever :) |
| 12:50.58 | senneth | *hides* |
| 12:50.58 | NAiL | aha |
| 12:51.12 | NAiL | written by you? :P |
| 12:51.18 | senneth | not really :) |
| 12:51.21 | NAiL | heh |
| 12:51.32 | NAiL | postfix is in openslug unstable anyway |
| 12:51.41 | senneth | ahh cool |
| 12:51.42 | NAiL | So it's just to add dovecot then |
| 12:51.55 | NAiL | In unstable, mind you |
| 12:52.11 | NAiL | I haven't had the time to test postfix enough to trust it in stable yet. |
| 12:52.23 | senneth | i've managed to compile dovecot on unslung so there should be no problems |
| 12:52.36 | NAiL | cool |
| 12:53.11 | NAiL | Haven't tried dovecot, but it might be nice. I'm using imap for my mail |
| 12:53.16 | senneth | and it's able to use epoll as the io-loop when the kernel is a 2.6 |
| 12:53.30 | NAiL | what about fam? |
| 12:53.49 | senneth | hum? |
| 12:53.54 | NAiL | file alteration monitor |
| 12:54.11 | NAiL | Think that's what cyrus uses |
| 12:54.25 | rwhitby | dyoung: running reflash now ... |
| 12:54.52 | senneth | dunno, have to look that up |
| 12:55.07 | NAiL | oh well.. Gotta go anyway |
| 12:57.17 | senneth | the main advantage is the authentication i think, you can use /etc/passwd or a virtual passwd file or vpopmail and stuff |
| 12:57.21 | DaKa2 | dovecot is really great |
| 12:57.57 | senneth | It's optimized to use very little memory aswell |
| 12:58.06 | DaKa2 | exactly, perfect for the slug |
| 12:58.24 | DaKa2 | Im using it on all my mailservers |
| 12:58.29 | rwhitby | ooh - testing turnup ramf for the first time ... |
| 13:04.44 | rwhitby | rebooting ... |
| 13:04.45 | dyoung | so I guess it worked? |
| 13:05.04 | rwhitby | just about to find out ... |
| 13:05.30 | rwhitby | oh YEAH baby! |
| 13:05.59 | rwhitby | verified. sweet dreams dyoung! |
| 13:06.01 | NAiL | ~emulate rwhitby |
| 13:06.01 | jbot | Well, my work here is done. |
| 13:06.07 | NAiL | hmm.. wrong one |
| 13:06.33 | dyoung | cool. |
| 13:06.38 | dyoung | ~emulate rwhitby |
| 13:06.38 | jbot | Well, my work here is done. |
| 13:06.43 | rwhitby | heh |
| 13:06.53 | NAiL | No more satisfied customers |
| 13:06.58 | dyoung-zzzz | ~emulate rwhitby |
| 13:06.58 | jbot | Well, my work here is done. |
| 13:07.10 | dyoung-zzzz | ~literal emulate rwhitby |
| 13:07.10 | jbot | "emulate rwhitby" is "<reply> Another Satisfied Customer! || <reply> Well, my work here is done." |
| 13:07.10 | NAiL | ~emulate rwhitby |
| 13:07.10 | jbot | Another Satisfied Customer! |
| 13:07.22 | dyoung-zzzz | slap jbot |
| 13:07.28 | NAiL | There :) |
| 13:07.52 | DaKa2 | NOTE: package perl-native-5.8.4-r0: task do_compile: started |
| 13:07.59 | DaKa2 | this is so not going to work.. |
| 13:08.06 | dyoung-zzzz | Enjoy. :-) |
| 13:08.38 | rwhitby | dyoung-zzzz: as soon as the reboot problem is solved, then that's a binary release candidate in my book .... |
| 13:10.48 | NAiL | But there's one thing I'd like to get clarified before a new release |
| 13:11.53 | NAiL | The address to the feeds... Is there any reason why there shouldn't be openslug/<version>/stable and openslug/<version>/unstable? |
| 13:12.11 | NAiL | Except making the package managers job harder... |
| 13:13.21 | NAiL | Because when a new release comes out, we would want it to pull from the stable feed. But there's been glibc/kernel changes since the original stable feed. |
| 13:14.05 | rwhitby | NAiL: that's your call. we'll push and pull from whatever feed structure you require :-0 |
| 13:14.20 | DaKa2 | why not openslug/<version> as the stable, and openslug/unstable for the unstable? |
| 13:14.48 | NAiL | DaKa2: not a bad idea. I think I had one hitch with that setup earlier though |
| 13:15.00 | NAiL | Can't remember what that was |
| 13:15.09 | NAiL | I'll think about it over dinner |
| 13:15.26 | DaKa2 | I really cannot see the need for versioned unstables, but I may be missing something |
| 13:15.39 | NAiL | no, you're right actually |
| 13:15.41 | NAiL | ok |
| 13:15.47 | rwhitby | unstable should always match the latest bleeding edge monotone |
| 13:15.51 | NAiL | exactly |
| 13:15.59 | DaKa2 | exactly exactly |
| 13:16.02 | NAiL | so unstable is always the same |
| 13:16.11 | [cc]smart | no, unstable is never the same :) |
| 13:16.11 | NAiL | only stable needs a version number |
| 13:16.25 | NAiL | heh, unstable is in flux, but the *path* is always the same, you nitpicker :p |
| 13:16.31 | DaKa2 | exactly :-) |
| 13:16.33 | [cc]smart | hrhr |
| 13:16.33 | rwhitby | so feeds/openslug/2.2-beta and feeds/openslug/unstable |
| 13:16.38 | NAiL | exactly |
| 13:16.47 | NAiL | Will the release be 2.2-beta? |
| 13:17.07 | rwhitby | no, cause we currently have 2.2-beta, so we need to increment on release |
| 13:17.14 | NAiL | yeah |
| 13:17.25 | rwhitby | numbers are cheap. increment early and increment often |
| 13:17.30 | NAiL | :) |
| 13:18.55 | NAiL | I'd like to know the number beforehand, so I can update the .conf-file so it pulls from the right stable feed from the beginning |
| 13:19.06 | DaKa2 | Im actually starting to think my native package build of perl is going to work.. |
| 13:19.10 | DaKa2 | scarry |
| 13:19.41 | rwhitby | NAiL: we will change the openslug.conf file to use the variable |
| 13:20.00 | rwhitby | so it will automatically be correct |
| 13:20.20 | NAiL | rwhitby: ah, good |
| 13:20.56 | rwhitby | just use DISTRO_VERSION in FEED_URIS.... |
| 13:21.24 | *** join/#nslu2-linux caplink811-log (~caplink81@dsl-213-023-144-145.arcor-ip.net) |
| 13:21.28 | rwhitby | NAiL: are you going to have "cross" and "native" in the feed uri too ? |
| 13:22.11 | NAiL | rwhitby: I can't see why that should be necessary |
| 13:22.19 | rwhitby | feeds/openslug/cross/2.2-beta, feeds/openslug/native/2.2-beta, feeds/openslug/cross/unstable, feeds/openslug/native/unstable |
| 13:22.27 | NAiL | or... |
| 13:22.29 | NAiL | yes |
| 13:22.42 | NAiL | it makes the pushing a heck of a lot easier |
| 13:22.50 | NAiL | but from the user side there shouldn't be any diff |
| 13:22.55 | rwhitby | it's what we do for unslung ... |
| 13:23.03 | NAiL | yes, thanks |
| 13:23.49 | NAiL | Ok, thanks for the input all. Taxi outside, gotta go |
| 13:23.56 | rwhitby | bye |
| 13:23.59 | NAiL | I'll look at it later tonight |
| 13:27.01 | [cc]smart | taxi ? he's got plans.... :D |
| 13:27.33 | [cc]smart | wow, ferakin shit. these thunders get hefty |
| 13:30.27 | *** join/#nslu2-linux D|dge557 (~didge@2001:960:624:1:0:0:0:1) |
| 13:30.33 | [cc]smart | and again i say, these are not the weathers i know from 10 years and back... |
| 13:43.05 | *** join/#nslu2-linux senneth (senneth@senneth.staff.irssi) |
| 13:52.49 | *** join/#nslu2-linux rwhitby-away (~rwhitby@rwhitby.nslu2-linux) |
| 13:56.35 | DaKa2 | I did that, and noticed there was no subversion |
| 13:56.56 | DaKa2 | then I just copied everything over instead |
| 13:58.41 | rwhitby-away | I'll let this run overnight and see if it finishes the monotone setup ... |
| 13:59.28 | [g2]-away | rwhitby I like that /root idea too |
| 14:00.44 | rwhitby-asleep | [g2]-away: yeah, unless there are objections I will implement that tomorrow. |
| 14:01.10 | [g2]-away | rwhitby-asleep, not from me :) |
| 14:01.51 | rwhitby-asleep | night all |
| 14:27.35 | NAiL | I can't see any reasonable objections to moving ~root either |
| 14:52.35 | NAiL | hmm.. how do I put two feed lines in the config? |
| 14:52.48 | NAiL | FEED_URIS_append_linux += "oe##http://ipkg.nslu2-linux.org/feeds/openslug/native/${DISTRO_VERSION}" |
| 14:52.57 | NAiL | Just have two lines like that? |
| 14:52.57 | *** join/#nslu2-linux eumel (~chatzilla@p54831284.dip0.t-ipconnect.de) |
| 14:54.45 | NAiL | Not sure how it is parsed |
| 14:58.21 | DaKa2 | different names |
| 14:58.27 | DaKa2 | both cannot be called oe |
| 15:02.13 | NAiL | aha |
| 15:02.14 | NAiL | yes |
| 15:02.36 | NAiL | so I just need to change the name## part, and add another line |
| 15:03.01 | DaKa2 | should work |
| 15:18.59 | [cc]smart | this changes fast than i can comply with |
| 15:19.08 | [cc]smart | ok 'ive updated monotone |
| 15:19.14 | [cc]smart | and am back on this |
| 15:19.39 | [cc]smart | now i made my bitbake openslug-image and i get a seemengly completely different result |
| 15:19.55 | [cc]smart | i don't get a flashdisk.img |
| 15:20.31 | [cc]smart | ah one sec |
| 15:20.36 | [cc]smart | me blind |
| 15:20.39 | [cc]smart | error occurde |
| 15:20.40 | DaKa2 | :) |
| 15:21.24 | NAiL | hmm |
| 15:21.45 | DaKa2 | /usr/lib/ipkg/info/bzip2.postinst: line 3: ldconfig: command not found |
| 15:21.55 | DaKa2 | uhm.. that looks a bit strange.. |
| 15:22.05 | DaKa2 | ahh |
| 15:22.12 | DaKa2 | no sbin in path |
| 16:04.39 | *** join/#nslu2-linux marceln (~marceln@pion.xs4all.nl) |
| 16:05.38 | CIA-8 | 03marceln * 10unslung/Makefile: Added an target for make/%.mk which generates the makefile from the template. |
| 16:51.16 | jbowler-away | NAiL: look in conf/distro/openslug.conf at the commented out lines, the name is, so far as I can see, arbitrary. E.g. ${DISTRO_NAME}-${DISTRO_VERSION} should be fine. |
| 16:55.43 | *** join/#nslu2-linux llagendijk (~louis@lagendijk.xs4all.nl) |
| 17:01.50 | *** join/#nslu2-linux grimreap (grimreap@c-67-165-167-67.hsd1.il.comcast.net) |
| 17:44.15 | [cc]smart | ah finally |
| 17:49.41 | CIA-8 | 03ccsmart 07org.openembedded.nslu2-linux * r75cf64dd... 10/ (3 files in 2 dirs): Added new populate script for directory creation during startup. |
| 17:49.44 | CIA-8 | 03ccsmart 07org.openembedded.nslu2-linux * r685decee... 10/ (5 files in 3 dirs): Moved populate-volatile to suggested place. |
| 17:49.46 | CIA-8 | 03ccsmart 07org.openembedded.nslu2-linux * r42c8721f... 10/packages/ (2 files in 2 dirs): Changed to use populate-volatile.sh. |
| 17:53.00 | [cc]smart | this makes an update to turnup necessary which will follow soonish |
| 17:54.32 | [g2]-away | [cc]smart, I've got two tweaks for turnup |
| 17:54.49 | [g2]-away | I can send them your way |
| 18:04.14 | [cc]smart | ok. |
| 18:04.19 | [cc]smart | you have the adress ? |
| 18:04.28 | [cc]smart | or pastebin |
| 18:06.49 | [g2]-away | [cc]smart, here |
| 18:06.56 | [g2]-away | <PROTECTED> |
| 18:06.56 | [g2]-away | <PROTECTED> |
| 18:06.56 | [g2]-away | <PROTECTED> |
| 18:06.56 | [g2]-away | <PROTECTED> |
| 18:06.56 | [g2]-away | <PROTECTED> |
| 18:06.57 | [g2]-away | <PROTECTED> |
| 18:06.59 | [g2]-away | <PROTECTED> |
| 18:07.01 | [g2]-away | <PROTECTED> |
| 18:07.25 | [g2]-away | those 4 lines gett added around line 753 |
| 18:08.14 | [g2]-away | basically in either the static case or dhcp case it updates the /etc/hosts file with the name of the host |
| 18:10.12 | [g2]-away | thx |
| 18:27.08 | *** join/#nslu2-linux AdamBaker (~aab@userch028.dsl.pipex.com) |
| 18:34.07 | [cc]smart | ok |
| 18:34.18 | [cc]smart | i will add this ina second rev |
| 18:37.36 | *** join/#nslu2-linux grimreap (~grimreap@c-67-165-167-67.hsd1.il.comcast.net) |
| 19:11.42 | *** join/#nslu2-linux cbrake (~cbrake@cpe-204-210-216-174.neo.res.rr.com) |
| 19:32.52 | [cc]smart | [g2]-away: ping |
| 19:48.58 | CIA-8 | 03ccsmart 07org.openembedded.nslu2-linux * rbc6e966f... 10/packages/openslug-init/ (openslug-init-0.10/turnup openslug-init_0.10.bb): Modified to work with populate-var and allow per target (nfs,disk...) definition of ramfs objects. |
| 19:49.01 | CIA-8 | 03ccsmart 07org.openembedded.nslu2-linux * rb810c896... 10/packages/openslug-init/ (openslug-init-0.10/turnup openslug-init_0.10.bb): Adding own hostname to hosts. Offered by [g2]. |
| 20:09.15 | *** join/#nslu2-linux Jacmet (~jacmet@50C59F0E.flatrate.dk) |
| 20:37.32 | *** join/#nslu2-linux ka6sox-web (www-data@netblock-66-159-209-60.dslextreme.com) |
| 20:48.34 | *** join/#nslu2-linux grimreap (~grimreap@c-67-165-167-67.hsd1.il.comcast.net) |
| 21:24.30 | *** part/#nslu2-linux marceln (~marceln@pion.xs4all.nl) |
| 21:35.01 | jbowler-away | NOTE: multiple providers are available (glibc, glibc-intermediate); |
| 21:35.01 | jbowler-away | NOTE: consider defining PREFERRED_PROVIDER_virtual/armeb-linux-libc-for-gcc |
| 21:35.19 | jbowler-away | Anyone who understands this enough to say what the correct setting is? |
| 21:47.12 | *** join/#nslu2-linux Ksilebo (russ@d14-69-132-59.try.wideopenwest.com) |
| 22:39.02 | CIA-8 | 03jbowler 07org.openembedded.nslu2-linux * rdbffb700... 10/ (25 files in 16 dirs): |
| 22:39.03 | CIA-8 | propagate from branch 'org.openembedded.dev' (head 50712da157ce819a41034d91e842f659f060e42e) |
| 22:39.03 | CIA-8 | <PROTECTED> |
| 22:53.44 | rwhitby | monotone.nslu2-linux.org is now only serving: org.nslu2-linux.dev. org.openembedded.dev and org.openembedded.nslu2-linux |
| 22:55.29 | rwhitby | that should be "org.nslu2-linux.dev org.openembedded.dev org.openembedded.nslu2-linux" |
| 22:55.38 | rwhitby | I wonder if we need to serve the dreambox branch too ? |
| 22:58.07 | jbowler | The following works for push/pull/serve: 'org.{nslu2-linux.*,openembedded.{dev,dreambox,nslu2-linux}}' |
| 23:00.50 | rwhitby | done |
| 23:14.26 | CIA-8 | 03jbowler 07org.openembedded.nslu2-linux * r4945900d... 10/packages/ (6 files in 2 dirs): Fix the missing quotes round string values. |