**** BEGIN LOGGING AT Tue Jun 21 23:59:56 2005
Jun 22 00:02:59 <jeanfabrice> morning
Jun 22 00:14:19 <CIA-3> 03jp30 * 10unslung/ (make/emacs.mk Makefile): split emacs into multiple ipks, enable cross-compilation
Jun 22 00:41:47 <mr_claus> while compiling openslug i get an error with patching ixp425-eth, it seems the patch does not match, can anybody confirm that?
Jun 22 00:45:20 <jacques> mr_claus, I can confirm that I have not gotten that error
Jun 22 00:47:38 <mr_claus> jacques: when did you the last compile?
Jun 22 00:48:21 <jacques> last night
Jun 22 00:48:40 <mr_claus> NOTE: package ixp425-eth-1.1-r1: task do_patch: failed
Jun 22 00:48:44 <jacques> about 24 hours ago
Jun 22 00:48:54 <jacques> ok, so what's the log say?
Jun 22 00:49:21 <mr_claus> DEBUG: ethhdr ethhdr.patch
Jun 22 00:49:21 <mr_claus> NOTE: import patch '/mnt/daten/projects/nslu2/openslug/openembedded/packages/ixp
Jun 22 00:49:21 <mr_claus> 425-eth/files/ethhdr.patch' as 'ethhdr.patch'
Jun 22 00:49:21 <mr_claus> ERROR: patch did not apply, try again with force (-f -i ...)
Jun 22 00:49:56 <mr_claus> my repo is 2 days old
Jun 22 00:50:31 <jacques> hard to say what could cause that
Jun 22 00:53:47 <mr_claus> could you run a md5sum of openembedded/packages/ixp425-eth/files/ethhdr.patch
Jun 22 00:56:34 <jacques> 9d577de2ea472fb3363188fb9925754b  openembedded/packages/ixp425-eth/files/ethhdr.patch
Jun 22 00:57:54 <mr_claus> tahts the same like my one
Jun 22 00:58:05 <mr_claus> but the source file does not match with the patch
Jun 22 00:59:01 <mr_claus> which line number is that line in your ixp425-eth-1.1-r0/ixp425_eth.c
Jun 22 00:59:03 <mr_claus> struct ethhdr *eth = skb->mac.ethernet
Jun 22 01:03:32 <jacques> uh
Jun 22 01:04:12 <jacques> there are two lines like that
Jun 22 01:04:46 <jacques> 1364 and 1406
Jun 22 01:05:54 <mr_claus> +++ ixp425-eth-1.1-r0/ixp425_eth.c
Jun 22 01:05:54 <mr_claus> @@ -1307,7 +1307,7 @@
Jun 22 01:05:54 <mr_claus>          * and its constants are taken from the eth_type_trans()
Jun 22 01:05:54 <mr_claus>          * function.
Jun 22 01:05:54 <mr_claus>          */
Jun 22 01:05:55 <mr_claus> -       struct ethhdr *eth = skb->mac.ethernet;
Jun 22 01:05:57 <mr_claus> +       struct ethhdr *eth = eth_hdr(skb);
Jun 22 01:05:59 <mr_claus>         unsigned short hproto = ntohs(eth->h_proto);
Jun 22 01:06:01 <mr_claus> then i wonder how that patch is working
Jun 22 01:06:21 <mr_claus> that patch does not fit to the source file
Jun 22 01:06:29 <mr_claus> how it is running at your machine?
Jun 22 01:06:31 <jacques> why not ?
Jun 22 01:06:44 <mr_claus> line number 1307
Jun 22 01:06:50 <jacques> patch can fuzz the lines
Jun 22 01:08:34 <jacques> I just started a build from latest pull
Jun 22 01:08:44 <jacques> we'll see if I get same error
Jun 22 01:08:46 <mr_claus> Hunk #1 succeeded at 1353 (offset 46 lines).
Jun 22 01:08:46 <mr_claus> Hunk #2 FAILED at 1395.
Jun 22 01:08:46 <mr_claus> 1 out of 2 hunks FAILED -- saving rejects to file ixp425_eth.c.rej
Jun 22 01:08:55 <mr_claus> its the second part
Jun 22 01:14:09 <jacques> what does ixp425_eth.c.rej look like?
Jun 22 01:14:57 <mr_claus> i did a dry-run, there is no rej-file, but it seems that the columns are not matching
Jun 22 01:15:15 <mr_claus> the orginal line begins at column 18 and the patch is looking at column 15
Jun 22 01:15:36 <mr_claus> perhaps there was included a additional loop before between the versions
Jun 22 01:19:39 <mr_claus> the patch use the wrong column and it uses blanks instead of \t
Jun 22 01:20:29 <jacques> maybe something changed in quilt
Jun 22 01:22:14 <mr_claus> FATAL: oe_runmake failed
Jun 22 01:22:22 <mr_claus> now it's patched but not compiling
Jun 22 01:25:05 <mr_claus> http://paste.uni.cc/7331
Jun 22 01:26:00 <jacques> mr_claus, please tell me you're clearing the icp build dir in tmp/work and the proper stamps in tmp/stamps between tries
Jun 22 01:26:35 <rwhitby-away> rm -rf tmp/{work,stamps}/ixp*
Jun 22 01:26:40 <mr_claus> jacques: no i did not because there was no compile before because the patch was not working only
Jun 22 01:27:07 <rwhitby-away> mr_claus: that's probably the issue then.  run the command I pasted and try again
Jun 22 01:27:44 <mr_claus> ok
Jun 22 01:31:32 <mr_claus> still the same error
Jun 22 01:33:39 <rwhitby> dunno then
Jun 22 01:33:57 <jacques> my build has a ways to go before it gets to that point
Jun 22 01:37:01 <mr_claus> whats up with the patch, who change it in the repo?
Jun 22 01:38:17 <rwhitby> mr_claus: ask bk - all the history is there ...
Jun 22 01:38:52 <rwhitby> I haven't seen a changeset go by referencing it ...
Jun 22 01:39:15 <mr_claus> rwhitby: i did not mean who changed it in the past, i mean who can upload the fix
Jun 22 01:40:00 <rwhitby> no-one - we are in a repo freeze until after the transition
Jun 22 01:47:22 <jacques> I see no changes on any of the ixp425 patches since 05/06/16
Jun 22 01:48:10 <jacques> hmm interesting
Jun 22 01:48:26 <jacques> that was only a few days ago, and it's for the patch which is failing for him
Jun 22 01:49:42 <mr_claus> perhaps a new source file?
Jun 22 01:50:35 <jacques> still several of us built last night
Jun 22 01:50:49 <jacques> so it's possible the patch was broken for a short time
Jun 22 01:51:00 <jacques> you said your pull was from two days ago
Jun 22 01:51:35 <jacques> tho this change was almost 6 days ago
Jun 22 01:51:50 <mr_claus> Jun 20 12:44 MET+1
Jun 22 01:57:35 <mr_claus> should i resync my repo?
Jun 22 01:58:33 <jacques> not sure yet
Jun 22 01:59:07 <mr_claus> i dont do it to keep the versions with the error
Jun 22 02:08:40 <jacques> my build is on kernel now so soon
Jun 22 02:13:40 <jacques> mr_claus, mine built fine
Jun 22 02:16:22 <mr_claus> jacques: then i will get the new repo
Jun 22 02:18:51 <jacques> yeah I think you should try that
Jun 22 02:26:34 <mr_claus> delete build/tmp and build/sources?
Jun 22 02:28:16 <mr_claus> i think i will only delete build/tmp
Jun 22 02:31:36 <jacques> no need to delete sources
Jun 22 02:38:23 <mr_claus> argll, same error
Jun 22 02:38:36 <mr_claus> NOTE: package ixp425-eth-1.1-r4: task do_patch: failed
Jun 22 02:50:05 <mr_claus> 425-eth/files/intdriven.patch' as 'intdriven.patch'
Jun 22 02:50:05 <mr_claus> ERROR: patch did not apply, try again with force (-f -i ...)
Jun 22 02:55:55 <jacques> different patch?
Jun 22 02:57:13 <mr_claus> yes, but i see that i forgot to delete ROOT/tmp, i only deleted ROOT/build/tmp
Jun 22 02:57:16 <mr_claus> i restarted 
Jun 22 02:57:42 <mr_claus> but i think the error will be the same because the ixp425 was another version, r4 instead of r1
Jun 22 02:58:10 <mr_claus> but first i'm waiting for the result
Jun 22 03:06:07 <rwhitby-away> mr_claus: why do you have two tmp dirs?
Jun 22 03:06:34 <mr_claus> rwhitby-away: i dont know, perhaps one is an old one
Jun 22 03:07:03 <mr_claus> rwhitby-away: it seems that build/tmp was the old one (which was unused) and /tmp is the new one
Jun 22 03:10:09 <rwhitby-away> anyone know if I can use WDS to link a wl500g with LAN on 192.168.128.x, with a wl530g with LAN on 192.168.1.x ?
Jun 22 03:12:12 <rwhitby-away> Hmm - I should probably just set my netmask on my lan to 255.255.0.0 and be done with it ....
Jun 22 03:18:39 <mr_claus> anybody know where i can by a board with an arm cpu and three network ports?
Jun 22 03:24:51 <[g2]> rwhitby-away, that's what I was going to say about the netmask
Jun 22 03:38:04 <rwhitby-away> [g2]: the idea is my production stuff is on .128.x, and the test stuff is on .1.x, so that when things go into recovery mode they are automatically on the test part of the subnet
Jun 22 03:38:23 <rwhitby-away> (especially when you have two routers, both of which will go into recovery mode to the same IP address)
Jun 22 03:39:59 <rwhitby-away> So I think I will just put the wl530g into ethernet bridge mode, and have everything hanging off it on .1.x, and everything else on .128.x, but all actually on the same LAN netmask of 255.255.0.0 ...
Jun 22 03:53:53 <mr_claus> arglll, same error
Jun 22 03:53:59 <mr_claus> why its not compiling here
Jun 22 03:56:01 <mr_claus> hunk 2 and 4 failed of patch intdriven
Jun 22 03:58:37 <mr_claus> thats the code
Jun 22 03:58:38 <mr_claus>     if (request_irq(IXP425_INT_LVL_QM1,
Jun 22 03:58:38 <mr_claus>                     dev_qmgr_os_isr,
Jun 22 03:58:38 <mr_claus>                     SA_SHIRQ,
Jun 22 03:58:38 <mr_claus>                     MODULE_NAME,
Jun 22 03:58:47 <mr_claus> and the patch
Jun 22 03:58:49 <mr_claus>      if (request_irq(IXP425_INT_LVL_QM1,
Jun 22 03:58:49 <mr_claus>                      dev_qmgr_os_isr,
Jun 22 03:58:49 <mr_claus> -                    SA_SHIRQ,
Jun 22 03:58:49 <mr_claus> +                    SA_SHIRQ | SA_SAMPLE_RANDOM,
Jun 22 03:58:54 <mr_claus>                      DRV_NAME,
Jun 22 03:59:19 <mr_claus> jacques: how that worked? DRV_NAME should be MODULE_NAME
Jun 22 04:00:04 <jacques> mr_claus, I dunno what to tell you
Jun 22 04:00:26 <mr_claus> please could you take a look into these two files
Jun 22 04:00:47 <jacques> i have to go to sleep now. it's 4am here
Jun 22 04:00:55 <mr_claus> ui
Jun 22 04:01:03 <mr_claus> then gn8 :)
Jun 22 04:01:17 <jacques> later
Jun 22 04:01:29 <mr_claus> i will change the patch try the next steps
Jun 22 04:37:57 <mr_claus> i found something about the problem
Jun 22 04:38:05 <mr_claus> the patches should be ok
Jun 22 04:38:17 <mr_claus> but there is one big patch which will be done with the prepatch task
Jun 22 04:38:26 <mr_claus> do_pre_patch () {
Jun 22 04:38:26 <mr_claus>         patcher -p 0 -i ixp425_eth_1_1_update_nf_bridge.patch
Jun 22 04:38:26 <mr_claus>         patcher -f -p 0 -i ixp425_eth.c.patch
Jun 22 04:38:26 <mr_claus> }
Jun 22 04:38:50 <mr_claus> the ixp425_eth.c.patch contains all the changes the other patches based on
Jun 22 04:39:26 <mr_claus> now the interesting part, in my workdir i got several files
Jun 22 04:39:28 <mr_claus> ixp425_eth.c
Jun 22 04:39:37 <mr_claus> ixp425_eth.c.orig
Jun 22 04:39:43 <mr_claus> ixp425_eth.c.patch
Jun 22 04:40:10 <mr_claus> i got following messages while compiling
Jun 22 04:40:13 <mr_claus> NOTE: package ixp425-eth-1.1-r4: task do_pre_patch: started
Jun 22 04:40:13 <mr_claus> NOTE: package ixp425-eth-1.1-r4: task do_pre_patch: completed
Jun 22 04:40:27 <mr_claus> but ... the ixp425_eth.c is not patched
Jun 22 04:41:00 <mr_claus> what could be the reason of that issue, log tell me it is patched, the .orig file is there, but the .c file is not patched?
Jun 22 04:52:02 <mr_claus> only with that driver patcher will be called, no other package do that
Jun 22 04:52:06 <mr_claus> patcher -p 0 -i ixp425_eth_1_1_update_nf_bridge.patch
Jun 22 05:37:40 <mr_claus> changed patcher to patch in the bb-file and it works now
Jun 22 06:21:37 <mr_claus> is there an information how to merge the 3 files into on firmware-image?
Jun 22 06:33:59 <mr_claus> contains the flashdisk image the other stuff?
Jun 22 06:34:49 <rwhitby-away> tmp/deploy/images/foo.flashdisk.img is the firmware image (8MB)
Jun 22 08:00:30 <mr_claus> coooool, my first openslug with homemade firmware is running :)
Jun 22 08:24:08 <mr_claus> why i cannot mount my disk with openslug?
Jun 22 08:24:28 <ka6sox-zzzz> turnup
Jun 22 08:24:57 <mr_claus> ??
Jun 22 08:45:14 <[g2]> mr_claus, congrats!
Jun 22 13:06:47 <CIA-3> 03llagendijk * 10unslung/ (5 files in 3 dirs): 
Jun 22 13:06:47 <CIA-3> - updated bind to 9.3.1
Jun 22 13:06:47 <CIA-3> - added openssl dependency in makefile
Jun 22 13:06:47 <CIA-3> - cleaned up postinst to remove check for openssl (moved to makefile/ipkg dependency)
Jun 22 13:06:47 <CIA-3> - cleaned up startup script to remove dependency on function script
Jun 22 13:06:48 <CIA-3> - removed renaming hack for ipkg workaround
Jun 22 13:06:48 <CIA-3> Louis Lagendijk <louis,lagendijk@gmail.com>
Jun 22 13:08:14 <CIA-3> 03llagendijk * 10unslung/make/quagga.mk: set maintainer field
Jun 22 13:09:55 <CIA-3> 03llagendijk * 10unslung/make/quagga.mk: removed double desription line
Jun 22 13:13:08 <macsat> hi
Jun 22 13:21:37 <macsat> hi jeanfabrice :)
Jun 22 13:22:48 <jeanfabrice> macsat: hey
Jun 22 14:30:41 <CIA-3> 03jeanfabrice * 10unslung/Makefile: Promote bind, tsocks for wiley
Jun 22 14:44:03 <jeanfabrice> llagendijk: ping
Jun 22 15:20:18 <CIA-3> 03jeanfabrice * 10unslung/Makefile: Promote oww, libghttp for wiley
Jun 22 15:50:22 <CIA-3> 03aabaker * 10unslung/ (5 files in 2 dirs): 
Jun 22 15:50:22 <CIA-3> Create a package for Linux net-tools. This package currently conflicts with
Jun 22 15:50:22 <CIA-3> the GNU coreutils package as both provide a version of hostname
Jun 22 15:51:29 <AdamBaker> As CIA has just spotted I've packaged up Linux net-tools for unslung
Jun 22 15:53:30 <AdamBaker> Most (all?) Linux distributions seem to use the net-tools version of hostname which provides -s and -f options to request short or full versions of the name rather than the coreutils version which apparently renames your host to -s if you try to give the -s option
Jun 22 15:53:44 <NAiL> haha
Jun 22 15:54:40 <AdamBaker> Should hostname just be removed from coreutils (it is provided by the Linksys busybox anyway) or do we want to set up update-alternatives for it?
Jun 22 15:58:00 <NAiL> [g2]: ping?
Jun 22 15:59:11 <NAiL> [g2]: Did you run make test when you built perl native? Did it work? :)
Jun 22 16:04:21 <NAiL> [g2]: nm, making test now
Jun 22 16:16:06 <rwhitby-away> AdamBaker: remove from coreutils is my vote.  Have a chat to Inge (bob_tm) about it, or email the list.
Jun 22 16:16:08 <rwhitby-away> back later
Jun 22 16:24:35 <NAiL> hmm... Any ETA on when oe will move to monotone?
Jun 22 16:34:04 <Sutak> Please be kind as this is not strictly a developer question, but I tried the wiki, #nslu2-general and searching the web. I want to start ircd as nobody, but "su" only tells me the output of tinylogin that does not support "su". What I am supposed to do?
Jun 22 16:35:22 <Sutak> I am using unslung 5.5, have unslinged to disk and installed a few packages.
Jun 22 16:40:52 <AdamBaker> Sutak: you could try installing the coreutils package - that includes a working version of su (but remember to check you have got /opt/bin before /bin in your path). Alternatively you could install the sudo package which provides what I think is a better alternative to using su.
Jun 22 16:48:11 <Sutak> Thank you.
Jun 22 17:03:21 <[g2]> NAiL, jacques build perl natively with the glibc 2.3.5 and got 100% on the make test iirc
Jun 22 17:06:01 <NAiL> ah, ok
Jun 22 17:06:28 <[g2]> how's the make test going ?
Jun 22 17:06:41 <[g2]> You seem to be having fun :) I hope you are.
Jun 22 17:06:52 <NAiL> Oh yes, I am ;)
Jun 22 17:07:09 <NAiL> Well.. looking at the list, it says "FAILED at test 35"
Jun 22 17:07:31 <[g2]> just one test ?
Jun 22 17:07:34 <NAiL> "t/op/magic"
Jun 22 17:07:35 <NAiL> Yeah
Jun 22 17:07:47 <NAiL> It has skipped a few, as usual
Jun 22 17:08:40 <[g2]> are you still working up to the MySQL build ? :)
Jun 22 17:08:48 <NAiL> hehehe, yeah
Jun 22 17:09:33 <[g2]> I think there were only 4 Python tests that failed, and one was random which probably passes now
Jun 22 17:11:29 <NAiL> I'm setting up a rather custom environment on my slug now. We've got a temporary plane we're gonna use :D
Jun 22 17:12:20 <[g2]> sounds fun
Jun 22 17:12:30 <NAiL> My old trainer will do for development, until we can build the large scale model.
Jun 22 17:12:59 <NAiL> It's not large enough to accomodate all the stuff, so redunda-slug won't be in the dev version :P
Jun 22 17:13:36 <NAiL> oh... yes... brilliant idea! I need to add bluetooth to the slug as well..
Jun 22 17:14:03 <NAiL> Eliminating having to plug my laptop into the plane to update it 
Jun 22 17:14:20 <[g2]> and the class 1 stuff goes 100M
Jun 22 17:14:30 <[g2]> outdoors :)
Jun 22 17:14:44 <NAiL> Haha... yeah, I can just fly the plane in a circle around my head, and update it mid-air :P
Jun 22 17:17:16 <NAiL> Of course, sparkfun has just what I need :D
Jun 22 17:20:30 <jacques> the "t/op/magic" "FAILED at test 35" is because you have busybox ps
Jun 22 17:20:42 <jacques> if you install procps ps it will succeed
Jun 22 17:21:24 <NAiL> ah, ok
Jun 22 17:29:42 <NAiL> Hmm.. I forget. Someone told me how much current the slug uses at the most (excluding any usb-connected stuff). Anyone remember?
Jun 22 17:39:35 <NAiL> Ah.. and does anyone know the weight of the board? I don't have a scale that is accurate enough :(
Jun 22 19:03:22 <CIA-3> 03jp30 * 10unslung/Makefile: promote bind
Jun 22 19:06:29 <CIA-3> 03jp30 * 10unslung/make/ftpd-topfield.mk: ftpd-topfield depends on libusb
**** ENDING LOGGING AT Wed Jun 22 23:59:57 2005