irclog2html for #nslu2-linux on 20050221

00:05.27rwhitby-awaydyoung: can you take the switchbox reigns from here for the next 10 hours?
00:06.59dyoungnot for a bit, I need to take care of the yard for a while.
00:10.36rwhitby-awayok
00:11.08rwhitby-away. /home/unslung/packages/toolchain/armv5b-softfloat-linux/gcc-3.3.5-glibc-2.2.5/lib/gcc-lib/armv5b-softfloat-linux/3.3.5/../../../../armv5b-softfloat-linux/bin/ld: warning: libexpat.so.0, needed by ../src/.libs/libfontconfig.so, not found (try using -rpath or -rpath-link)
00:11.08rwhitby-away../src/.libs/libfontconfig.so: undefined reference to `FT_Get_PS_Font_Info'
00:12.29rwhitby-awayjp30: problem in fontconfig
00:13.02rwhitby-away. /home/unslung/packages/toolchain/armv5b-softfloat-linux/gcc-3.3.5-glibc-2.2.5/bin/armv5b-softfloat-linux-gcc -g -O2 -Wl,-rpath -Wl,/opt/lib -o .libs/fc-cache fc-cache.o  -L/home/unslung/packages/staging/opt/lib ../src/.libs/libfontconfig.so -Wl,--rpath -Wl,/opt/lib
00:13.03rwhitby-away. /home/unslung/packages/toolchain/armv5b-softfloat-linux/gcc-3.3.5-glibc-2.2.5/lib/gcc-lib/armv5b-softfloat-linux/3.3.5/../../../../armv5b-softfloat-linux/bin/ld: warning: libfreetype.so.6, needed by ../src/.libs/libfontconfig.so, not found (try using -rpath or -rpath-link)
00:13.03rwhitby-away. /home/unslung/packages/toolchain/armv5b-softfloat-linux/gcc-3.3.5-glibc-2.2.5/lib/gcc-lib/armv5b-softfloat-linux/3.3.5/../../../../armv5b-softfloat-linux/bin/ld: warning: libz.so, needed by ../src/.libs/libfontconfig.so, not found (try using -rpath or -rpath-link)
00:13.08rwhitby-awaythat's the other errors before that
00:17.00jp30rwhitby-away, thanks: looks like i need to sprinkle in some -rpath-links
00:19.34*** join/#nslu2-linux rwhitby-treo (~upirc@rwhitby.nslu2-linux)
00:21.34CIA-103jp30 * 10unslung/make/fontconfig.mk: fontconfig: link with -rpath-link=STAGING_LIB_DIR
00:22.07jp30rwhitby-away, does that work better?
00:27.52jp30i wonder why fontconfig compiled ok for me. perhaps my build system has a different version of libtool from builds.nlsu2-linux. all the x11 stuff is freshly libtoolized by the build system.
00:28.11kergothbleh
00:28.18kergothipkg hardcodes the /usr/lib/ipkg path
00:28.23kergothdoesnt obey the configure script's libdir
00:28.25kergoth~lart ipkg
00:29.45jacques1ii  libtool                   1.5.6-4                   Generic library support script
00:29.52jacquesfairly recent
00:29.53CIA-103g2 07 * r1.2833 10openembedded/ (3 files in 3 dirs): Switch to hotplug-ng and tweak sysconfsetup for /etc/networks/interfaces and ehci-hcd, boot directly to /dev/mtdblock4 until we add /initrd and fix switchbox
00:30.03CIA-103g2 07 * r1.2832 10openembedded/ (conf/distro/openslug.conf BitKeeper/etc/logging_ok):
00:30.03CIA-1Merge nslu2-linux@nslu2-linux.bkbits.net:openembedded
00:30.03CIA-1into giantshoulder.com:/home/tom/dev/openslug/openembedded
00:30.11CIA-103g2 07 * r1.2787.129.44 10openembedded/conf/distro/openslug.conf:
00:30.11CIA-1Merge http://oe-devel.bkbits.net:8080/openembedded
00:30.11CIA-1into giantshoulder.com:/home/tom/dev/openslug/openembedded
00:30.44CIA-103g2 07 * r1.2787.120.2 10openembedded/BitKeeper/etc/logging_ok:
00:30.44CIA-1Merge http://oe-devel.bkbits.net:8080/openembedded
00:30.44CIA-1into giantshoulder.com:/home/tom/dev/openslug/openembedded
00:48.52CIA-103jp30 * 10unslung/make/pango.mk: pango depends on libICE
01:21.27CIA-103jp30 * 10unslung/make/ (pango.mk gtk+.mk): make sure gtk+ and pango aren't distracted by host X11 and freetype
01:43.52CIA-103jp30 * 10unslung/make/gtk+.mk: fix gtk-stage target
02:08.15*** join/#nslu2-linux FireEgl (Ariel@Atlantica.10-1-1-1.Info)
02:24.17*** join/#nslu2-linux JWynen (~chatzilla@CPE0000864bf1c5-CM014480103458.cpe.net.cable.rogers.com)
02:24.21JWynenHi everyone. Anyone have time to help troubleshoot an ipkg problem?
02:28.59[g2]JWynen, just state the problem and check back in a while
02:31.50JWynenthanks g2.  my nslu is 3.17beta, behind a wrt54g.  i have the nslu dns and gateway numbers set thru the webmin interface.  i can ping ipkg.nslu2-linux.org no problem, but ipkg -verbose_wget update fails with error
02:32.22JWynenresolving %22192.168.1.1%22... failed : host not found
02:32.35JWynenthanks for any help you can give
02:33.58[g2]JWynen, you are pinging from the nslu2 right ?
02:34.25JWyneni telnet into the nslu, then run ipkg as root
02:34.58[g2]and you ping from the telnet on the nslu2 fine right ?
02:35.25JWynenyep.  resolves to an address at sourceforge if i remember correct
02:35.52[g2]what does the /etc/ipkg.conf say for source ?
02:35.57[g2]or src
02:36.23JWynenno such file
02:37.38[g2]is that /opt/etc/ipkg.conf ?
02:38.09JWynen<PROTECTED>
02:38.21jp30ipkg config should be a directory /etc/ipkg/*
02:38.32jp30...on unslung 3.17
02:38.49JWynengot arch.conf and unslung-feed.conf
02:39.34jp30so what's in unslung-feed.conf?
02:39.58JWynen# cat unslung-feed.conf
02:40.14JWynensrc/gz unslung http://ipkg.nslu2-linux.org/feeds/unslung/unstable
02:41.23jp30that's OK. can you cat /etc/resolv.conf for us?
02:42.38JWynennameserver 192.168.1.1
02:42.57JWynenthats the ip of the wrt54 i
02:43.20[g2]wget http://ipkg.nslu2-linux.org/feeds/unslung/unstable/Packages
02:44.14[g2]and wget http://66.35.250.210/feeds/unslung/unstable/Packages
02:44.22JWynenResolving %22192.168.1.1%22... failed: Host not found.
02:44.55ka6sox-awayJWynen, is the WRT providing DNS services/
02:44.57ka6sox-away?
02:45.03JWynenand wget http://66.35.250.210/feeds/unslung/unstable/Packages
02:46.13JWynenbelieve so
02:46.53JWyneni have tried to enter the service providers dns ip numbers in webmin, but the nslu webmin gives an error about invalid ip when i do that
02:47.12ka6sox-awayJWynen, are you running OpenWRT on your WRT?
02:47.33JWynenno, stock linksys, 1.42 i think
02:47.49jacqueswhat are the service providers dns ip numbers?
02:47.57[g2]JWynen, the dns isn't resolving
02:48.21[g2]or is resolving to 192.168.1.1
02:49.15JWyneni will reenter the service providers dns numbers in webmin and try again.  Question: do i need to reboot the nslu for the changes to take effect?
02:49.33[g2]JWynen, echo "66.35.250.210  ipkg.nslu2-linux.org" >> /etc/hosts
02:50.22JWynenok, done
02:50.30jp30[g2], that won't work, there's no /etc/nsswitch.conf on unslung, so /etc/hosts is ignored.
02:51.14[g2]well try the wget with the name for grins
02:51.39[g2]it's a couple up arrows and a return right ?
02:51.41[g2]:)
02:52.33JWynenzing, flies over Jwynens head
02:53.10jp30JWynen, perhaps if you echo "hosts: files dns" >>/etc/nsswitch.conf as well as what [g2] was suggesting
02:53.14[g2]JWynen, the up arrow key brings up the last command
02:53.19jacquesI want to know why webmin gives invalid ip error - sounds like he;s entering the IPs wrong
02:53.41jacqueswhy are we going down this /etc/hosts/nsswitch.conf path?
02:53.53jacqueswhat's not going to help him get DNS resolution working
02:54.03JWynenok  have reset the dns numbers in webmin to the service providers two numbers.  do i need to reboot the nslu for it to take effect?
02:54.30jacquespractically? yes
02:54.38jacquestechnically, probably not
02:54.40JWyneng2, I wondered why !! wasnt working :-)
02:55.07[g2]cat /etc/resolv.conf are they changed ?
02:55.32jacquesthe wrt should still distribute itself as the dns server
02:55.47JWynenok just hit restart on webmin.  will have to wait a moment for /etc/resolv.conf
02:56.35jacquesJWynen, did you set the provider DNS IPs in the wrt or the nslu2?
02:56.53JWynenjacques, in the nslu2
02:57.11jacquesand you did it exactly the same but this time it didn't say "invalid IP" ?
02:57.15[g2]is webmin stock in 3.17beta ?
02:57.48jacquessomething fishy here
02:58.47JWynenjacques, i did the exactly the same, but this time no error.  maybe my memory is slipping.  nslu has restarted, still no joy
02:59.45JWynennameserver 24.153.22.195
03:00.18JWynenand nameserver 24.153.23.66
03:00.40JWynenboth are the DNS numbers listed by the wrt54
03:01.08jp30do you still get the same message from wget?
03:02.45JWynenjp30, Resolving %22192.168.1.1%22...failed
03:03.30jp30and that IP number, 192.168.1.1 no longer appears in your /etc/resolv.conf?
03:04.01JWynenjp30, nope, just the two 24.153s
03:04.42jp30is wget maybe trying to use a proxy?
03:05.05JWynenping still resolves to vvhost.sourceforge.net
03:07.33jp30is your slug using dhcp, or a fixed IP address?
03:08.56JWynenmy slug is fixed at 192.168.1.77  Question.  the wrt54 lists the default gateway as 24.114.82.1  should the slugs gateway be set to this or the ip of the wrt54
03:10.19jp30sounds like your wrt's gateway is set correctly
03:11.28jp30can you try a wget command with --proxy=off ?
03:12.09JWynenjp30, that did it,  100% loaded
03:13.08[g2]~praise jp30
03:13.11jbotAll hail jp30!
03:13.11jp30hah! try "set" and look to see if there's anything like proxy="192.168.1.1" in there
03:14.10JWynennothing in there
03:14.28jp30grep 192.168.1.1 /etc/*
03:14.57jp30(or fgrep if you're a purist)
03:16.56JWynenreturns /etc/CGI_ds.conf:dhcp_start_ip=192.168.1.100
03:17.09JWynenand /etc/CGI_ds.conf:gateway=192.168.1.1
03:18.24JWynenjacques, i remembered that webmin error wrong.  i got it when it tried to change the slugs gateway from 192.168.1.1 to something else
03:19.35jp30JWynen, neither of those are suspicious... any lines *not* commented out in /etc/wgetrc?
03:20.54JWynenjust waitretry = 10
03:21.19JWynen#use_proxy = on
03:21.46CIA-103unslung * 10unslung/Makefile: Promoted atk fixesext fontconfig freetype pango renderext ttf-bitstream-vera xcursor xfixes xft xrender to cross. gtk+ must be renamed (target names don't match .mk file name).
03:22.02Tierstenthe proxy support will be turned on if you have any proxy vars in the env
03:22.09Tierstenset | grep proxy
03:22.11Tierstendo that
03:22.51JWynenreturns  _='proxy'
03:23.19Tierstenthat's fine
03:23.38Tierstenif it's not in wgetrc and it's not in the env then are you sure you're not making it pass it on the command line args?
03:24.42JWynenTiersten, I dont think so, I am just entering "ipkg update"
03:25.39JWynencan i specifically tell ipkg to proxy=off
03:25.43Tierstennobody else has experienced this problem so it must be something specific you've done
03:26.11Tierstenlook in /etc/ipkg.conf
03:26.18Tierstenare any of the proxy lines enabled?
03:27.24Tierstenalso check .wgetrc in your home dir
03:27.29JWynenno file /etc/ipkg.conf  there is an /etc/ipkg directory with arch.conf and unslung-feed.conf
03:27.59JWynen~praise Tiersten
03:28.01jbotAll hail Tiersten!
03:28.09Tiersten?
03:28.24JWynenok .wget rc has http_proxy set to 192.168.1.1
03:28.29Tierstenah
03:28.36JWynenI didnt do that, honest :-)
03:28.39Tierstenheh
03:28.47JWynennuke .wgetrc?
03:29.29Tierstenanything else in there?
03:29.41Tierstenrename it and try again anyway
03:29.50Tierstenyou can always put it back and edit it to remove the proxy lines
03:30.03JWynenok, my bad.  at 2am, i did enter that .wgetrc after seeing it in an ipaq .wgetrc.  my bad
03:30.23Tierstenheh
03:31.21JWynenthat worked.  thanks everyone, sorry you laboured over my stupid mistake.  I will try to do something constructive for the group to atone.  I am a hardware guy, so if Tiersten needs a favour, I owe
03:32.30Tierstenno prob
03:32.48TierstenIn this project however, I'm also one of the hardware/lowlevel guys :)
03:34.48JWynenwhat kind of stuff?  i have done some arm7 boards
03:36.16TierstenI was doing work on a replacement redboot but I've been side tracked with real work.  A few other people are working on a total replacement using APEX however
03:38.04TierstenThe pad for JTAG are on the back of the PCB if you want to mess about with it
03:38.09Tierstenpads
03:43.01JWynenThanks Tierstan.  I saw the locations of the JTAG points on a post the other day.  maybe I should look into making a jtag jig for the board available.  I have to do that for my dev boards, but its not the most exciting part of the job
03:44.03TierstenAt least the pads are fairly easy to solder to
03:44.28TierstenThe erase blocks on the strataflash used in the NSLU2 are 128KB so you'll want a decent JTAG interface
03:44.57TierstenThe cheapo parallel port ones you can make/buy go incredibly slowly due to the speed of the parallel port and the 498 bit BSR in the IXP420
03:46.33JWynenI have access to a macgregor raven, so its a little faster (but not much)  Intel also pads out the sequence with 1000s of TCKs if I remember right
03:47.27TierstenA Raven is better than nothing.  It's about 22 minutes per 128KB at the moment
03:47.53TierstenSo if you totally erase the flash then the minimum you'll want to do is 256KB to get RedBoot back up and then load the rest via the network
03:47.57[g2]thank god APEX is only 16KB :)
03:48.40Tierstenthat doesn't include the massive NPE access lib yet but it should still come out to be under 128KB :)
03:48.43[g2]and can also run from ram
03:48.53TierstenRedBoot can as well
03:49.40JWyneni am familiar with redboot, but who is developing APEX?
03:49.47[g2]I think when the FatSlug arrives, I'll be flashing APEX and will probably never need the JTAG
03:49.56[g2]except for maybe debugging
03:50.18[g2]JWynen, #openJTAG is the place to be :)
03:50.44TierstenJWynen: http://projects.buici.com/apex/
03:50.48[g2]beewoolie wrote APEX for a different ARM and did the port
03:50.57[g2]for the ixp4xx
03:50.59Tiersten[g2]: How you going to load stuff?
03:51.18[g2]APEX writes to the flash
03:51.28[g2]it boots OpenSlug
03:51.40JWynenthanks g2  i will check it out.  ok, i will stop with the idle chatter and get dropbear on my slug.  Thanks again g2, jacques and Tiersten
03:51.50[g2]unless I screw up APEX it's just like redboot
03:52.05Tierstenyeah but at the moment we're loading images via the network
03:52.15TierstenAPEX can't use the NPEs at the moment
03:52.18[g2]except I'd have to load the kernel via Xmodem over the serial
03:52.22Tierstenah okay
03:52.37TierstenNo flow control either :)
03:52.41TierstenXON/XOFF sucks
03:53.20[g2]but APEX already toggles the LEDs and there's talk about adding boot selection via the reset button and pushing the power button
03:53.32[g2]the LEDS will indicate selection
03:53.42TierstenJWynen: oh yeah.  one more thing.  fatslug is a NSLU2 with double the memory.  stacking another 32MB on the existing chips
03:53.45[g2]So two kernels, two rootfs
03:53.48Tierstenyeah I know
03:54.24Tierstenhad a conversation with beewoolie on the wiki on the sluggo page :)
03:54.25JWynenTiersten, where do you pick up the extra CS# from?
03:54.30TierstenPR12
03:54.37[g2]ahh.
03:54.44Tierstenone of the pins isn't connected on the other side of that.  it's the other bank select line
03:55.00Tierstenonly got two in the IXP420 and one is already wired to both RAM chips
03:55.19Tierstenhttp://www.nslu2-linux.org/wiki/Info/MemoryExpansion
03:56.35JWynenQuestion for you fellows, part of my motivation for picking up a slug was to have something to polish up my embedded linux skills with.  Any usefull materials you could recommend to get started?  I was thinking of getting the Embedded linux book (cant rember the real title).  
03:57.12TierstenNo clue on that one.  Sorry
03:57.19TierstenOff to bed.  Night all
04:02.50CIA-103jp30 * 10unslung/ (4 files in 3 dirs): rename gtk+ -> gtk
04:28.44[g2]nite
04:29.59[g2]JWynen, just jump in on what interests you.
04:30.22[g2]You've basically got a $80 dev board
04:31.19JWyneng2, thanks, that was my thinking too.  i might get a second to play with too.
04:32.26*** join/#nslu2-linux VoodooZ_home (~stephane6@CPE00c0f021712c-CM014110003785.cpe.net.cable.rogers.com)
04:32.43[g2]speaking of hw hackers :)
04:33.07VoodooZ_home:)
04:33.37VoodooZ_homeI noticed there was a lot of changes earlier. What revolution is  brewing?
04:33.37[g2]hey, if you're feeling spunky or brave I checked in the changed for hotplug
04:33.50VoodooZ_homeChange? The missing scripts?
04:34.05[g2]no
04:34.11VoodooZ_homeYou made it the default?
04:34.15[g2]nod
04:34.22VoodooZ_homeI'm game baby! :)
04:34.37VoodooZ_homeActually, I never found my rootfs problem so I can't work.
04:35.02VoodooZ_homeI guess I can only start from scratch and redo all my privates changes.
04:35.19[g2]I bring up the ixp and all the usb stuff comes up with the ehci-hcd driver that gets installed
04:35.22VoodooZ_homeKergoth told me make patchsets using quilt but I haven't had the gut to figure it out.
04:35.39VoodooZ_homeSo everything works as before then?
04:35.51[g2]I think so
04:35.57VoodooZ_homeGreat work!
04:36.02VoodooZ_home~praise [g2]
04:36.04jbotAll hail [g2]!
04:36.04[g2]but it's a bunch faster
04:36.15VoodooZ_homeWhat's your latest benchmark?
04:36.25[g2]praise me *after* it works the same for you :)
04:36.41VoodooZ_homeWell, you know me, i always manager to screw things up!
04:36.51[g2]there is one change for the moment
04:36.59[g2]it boots directly to the jffs2 partition
04:37.03VoodooZ_homeAlso, what other changes were you guys talking about earlier? Switchbox stuffs.
04:37.19VoodooZ_homeperfect. I want that.
04:37.29[g2]rwhitby's working on merging with the latest switchbox, that's almost done
04:37.32VoodooZ_homeI don't need the switchbox/initrd stuff.
04:37.52VoodooZ_homeAre you using unionfs or?
04:37.53[g2]the command line checked in should work for you
04:38.00VoodooZ_homegreat!
04:38.01[g2]no unionfs
04:38.16VoodooZ_homeok. I just saw a changeset for it so I was wondering.
04:38.54VoodooZ_homeSo right now is it booting switchbox or not? (when you say it boots jffs2 directly I tend to think it bypasses switchbox.)
04:38.56[g2]You'll have to tweak the /etc/network/interfaces scripts for BT
04:39.14VoodooZ_homeyep. One of many private changes I have to redo unforturnately
04:39.19[g2]right now no switchbox, straight to mtdblock4
04:39.27VoodooZ_homegood.
04:39.33VoodooZ_homeBut it is just temporary?
04:40.08[g2]Once switchbox is working, that'll support booting to the hd
04:40.44[g2]I think a lot of ppl will want that and it's only a tweak to the kernel command line to boot straigth to jffs2
04:41.09[g2]so I don't think it's that hard to put back
04:41.17VoodooZ_homeyep. It makes more sense as there's more people needing it. I'm a very special case so I don't mind changing it.
04:41.37[g2]and it's a very simple change if the repo isn't hosed :)
04:41.50VoodooZ_homehehehehe :(
04:42.24jacquesdagnabit
04:42.32VoodooZ_homeI wish I could port all my changes to my new fresh repo but there's a chance it might bring in the bad one too.
04:42.35jacquesI cant get my slug to download an image from my camera
04:42.41VoodooZ_homeCool.
04:42.50VoodooZ_home<PROTECTED>
04:43.02VoodooZ_homesimple colour segementation.
04:43.17[g2]Hey you'll have to hook me up with a battery supplier and ideas for a robot kit. I think I'm gonna play with some stuff this summer
04:43.19VoodooZ_homeI could have ported OpenCV to OE but I'd rather code it myself and learn in the process.
04:43.23VoodooZ_homeMuch faster too.
04:43.34VoodooZ_homeNo problem man!
04:43.45VoodooZ_homesend me an email anytime you need stuff.
04:43.52[g2]I was running the Slug from batteries the other day
04:44.06[g2]4 AA NiMh
04:44.09VoodooZ_homeAlthough I'm not experienced with kits as I've never used one. I'm stubborn. Have to build it myself.
04:44.15VoodooZ_homejust 4?
04:44.21[g2]yup
04:44.39VoodooZ_homeI'm doing a battery burn-in test right now using 8 2100mAh AA.
04:44.45[g2]it was *so* cool
04:45.00VoodooZ_homeIt's been running for a bit more than an hour but that's not including any sensors.
04:45.04[g2]4 in parallel with the other 4
04:45.17VoodooZ_home4 is a bit low though
04:45.35VoodooZ_home4 x 1.2V = 4.8V
04:45.54[g2]nod.
04:46.01[g2]better than 6v
04:46.04VoodooZ_homeYeah I guess it would work but in my case I need a 10V+ pack for all my onboard regulators.
04:46.20VoodooZ_homeI was assuming you were using a regulator.
04:46.32[g2]no straight in
04:46.38VoodooZ_homeAnd I have a 12 cell one for my motors.
04:46.42VoodooZ_homeI see.
04:47.14VoodooZ_homeA small DC-DC switcher PS IC like the lm2825 is great. no external component required.
04:47.40[g2]expensive (relatively) thought no ?
04:47.42VoodooZ_homeI heard bad things about running packs in parallel though.
04:48.28VoodooZ_homeyep. Those are high quality OEM tabbed (soldering) ones I bought on special from www.onlybatteries.com
04:48.41VoodooZ_homeAlmost 80+$ CDN total
04:48.52VoodooZ_homeOh, you meant the regulator?
04:49.03VoodooZ_homeI ordered free samples a few years ago.
04:49.04[g2]yeah the regulator
04:49.08VoodooZ_homeI might need more soon.
04:49.15VoodooZ_homeCheap National's page.
04:49.32VoodooZ_homehttp://www.national.com/pf/LM/LM2825.html
04:50.27VoodooZ_homeThey'll send you samples within 24 hours. they are great.
04:50.28[g2]yeah the who 5V battery pack was only $12
04:51.57VoodooZ_homegood.
04:52.27VoodooZ_homeAre you mechanically challenged like I am? Or do you plan on building your own mobile base?
04:52.30[g2]I guess 4 C's would do it better
04:53.32[g2]I'll get a kit first. It's really for my young one to learn about computers hands-on
04:53.37VoodooZ_homeYep. Or even Ds or those cute little sub-c's but my robot space is at a premium.
04:53.43VoodooZ_homeI see.
04:54.11jacqueswhen I was looking it seemed like AA was king - none of the other standard sizes came close on power density
04:54.34VoodooZ_homeThere's so many of them nowadays too. Everybody is getting into robotics. When I started a few years ago there was barely anything for hobbyists
04:54.34jacquesit was pathetic -  a D which weighs like 6x as much had only about 3x mAH
04:54.58VoodooZ_homeActually, the sub-c and c can do 4500mAh I think so it would be nice.
04:55.15jacquesAAs can do 2500mAH now
04:56.05VoodooZ_homeLike this 3500mAh sub-c on special right now at http://www.onlybatteries.com/webimages/images/12027.jpg
04:56.12VoodooZ_homeoops. wrong link.
04:56.21VoodooZ_homehttp://www.onlybatteries.com/showitem.asp?ItemID=12027
04:56.33VoodooZ_homeThey can now yeah
04:57.01VoodooZ_homeit depends how much space/current you have/need
04:57.50[g2]I was thinking BT and flashstick
04:57.54VoodooZ_homeOr those juicy 9500mAh Ds : http://www.onlybatteries.com/showitem.asp?ItemID=12424.61
04:58.08VoodooZ_homefor what? robot?
04:58.42[g2]Stand-alone BT/WIFI AP
04:58.48[g2]portable :)
05:00.18VoodooZ_homeyeah. then it's a matter of space.
05:02.09[g2]Is NiMh the best kind for price/performance ?
05:02.38jacquesheh
05:04.40[g2]VoodooZ_home, oh and hotplug isn't working for net
05:05.00VoodooZ_homeI would think yeah.
05:05.01[g2]I think pci, usb, scsi work
05:05.25VoodooZ_homeI don't care much for CSR as I'll be removing it but It might be an issue for my BT.
05:05.40[g2]So if you plug in the BT or usbnet you'll need to ifup
05:06.17VoodooZ_homeactually, It's usually plugged in at bootup so wouldn't that work?
05:06.35[g2]but since the networking scripts runs after the drivers are loaded in S39sysconfsetup they come up fine on boot
05:07.10VoodooZ_homegood. that's all I need really.
05:07.43VoodooZ_homeThe robot is usually turned on before every competition runs and I only use it for debugging.
05:13.39VoodooZ_homeanyways, It's past midnight here so I'll see you guys tommorow morning.
05:13.41VoodooZ_homeTake care,
05:20.40*** join/#nslu2-linux rwhitby-web (www-data@netblock-66-159-209-60.dslextreme.com)
05:20.46rwhitby-webVoodooZ_log: ping
05:21.04rwhitby-webdang.  missed him
05:22.54rwhitby-webdyoung: ping
05:26.19*** join/#nslu2-linux garpinc (~garpinc@pool-151-203-198-44.bos.east.verizon.net)
05:27.59*** part/#nslu2-linux rwhitby-web (www-data@netblock-66-159-209-60.dslextreme.com)
05:41.21*** join/#nslu2-linux rwhitby-away (~rwhitby@rwhitby.nslu2-linux)
06:11.19*** join/#nslu2-linux rwhitby-away (~rwhitby@rwhitby.nslu2-linux)
06:13.55*** join/#nslu2-linux samms71 (~samms71@c-67-168-203-102.client.comcast.net)
06:21.56*** part/#nslu2-linux samms71 (~samms71@c-67-168-203-102.client.comcast.net)
06:22.07jacquesweird - after latest pull my unslung-image build time went down by 9 minutes
06:22.37jacquesthis is versus a previous primed-ccache build
06:37.25*** join/#nslu2-linux mrblack51 (mrblack51@wa-pullman-cuda1k-244.losaca.adelphia.net)
06:49.06*** join/#nslu2-linux NSLU2-IRC (~NSLU2-IRC@S01060010dc1d88de.vn.shawcable.net)
06:57.48*** join/#nslu2-linux mrb51 (mrblack51@wa-pullman-cuda1k-244.losaca.adelphia.net)
08:06.20*** join/#nslu2-linux christopher_logg (~christoph@128-193-244-84.resnet.oregonstate.edu)
08:15.00*** join/#nslu2-linux mrblack51 (mrblack51@wa-pullman-cuda1k-244.losaca.adelphia.net)
08:22.46*** join/#nslu2-linux christopher-logg (~christoph@128-193-244-84.resnet.oregonstate.edu)
09:14.15*** join/#nslu2-linux CGI794 (~ac10370b@cpe-70-93-27-157.hawaii.rr.com)
09:18.13*** join/#nslu2-linux perlguru-work (www-data@netblock-66-159-209-60.dslextreme.com)
09:35.25*** join/#nslu2-linux CIA-1 (~CIA@flapjack.navi.cx)
09:44.00*** join/#nslu2-linux caplink811 (~Joerg@dsl-213-023-142-249.arcor-ip.net)
10:27.05*** join/#nslu2-linux NSLU2-IRC (~NSLU2-IRC@pcp0011259705pcs.newhvn01.in.comcast.net)
10:29.09*** part/#nslu2-linux NSLU2-IRC (~NSLU2-IRC@pcp0011259705pcs.newhvn01.in.comcast.net)
10:41.37CIA-103dereky 07 * r1.2834 10openembedded/packages/openslug-init/openslug-init_0.10.bb: added creation of /initrd for switchbox compatibility
10:47.59CIA-103dyoung 07 * r1.2835 10openembedded/packages/openslug-init/openslug-init_0.10.bb: Oops forgot to bump the PR
11:27.28*** join/#nslu2-linux rwhitby-treo (~upirc@rwhitby.nslu2-linux)
12:13.10*** join/#nslu2-linux perlguru-work (www-data@netblock-66-159-209-60.dslextreme.com)
12:14.50rwhitby-awaykergoth`zzz: ping
12:15.03rwhitby-awayso OpenSlug's /sbin/init has the following code in it:
12:15.12rwhitby-away#ifndef DEBUG_INIT
12:15.12rwhitby-away<PROTECTED>
12:15.12rwhitby-away#ifdef CONFIG_FEATURE_INITRD
12:15.12rwhitby-away<PROTECTED>
12:15.14rwhitby-away#endif
12:15.14rwhitby-away<PROTECTED>
12:15.16rwhitby-away<PROTECTED>
12:15.18rwhitby-away<PROTECTED>
12:15.42rwhitby-awaywhich means that it either has to be PID 1, or invoked as /linuxrc.
12:16.02rwhitby-awaywhich means that it can't be used as a target for the exec chroot that happens after a pivot_root
12:18.14rwhitby-awayso a symlink from /sbin/init to /linuxrc is called for in openslug-init ...
12:18.30rwhitby-awayln -s /sbin/init /linuxrc
12:20.43rwhitby-awayvery quiet here tonight ....
12:20.56gielrwhitby-away: i know and you're wrong
12:21.15gielrwhitby-away: i mean, you're right, but you just have to make sure it is pid 1
12:21.34gielrwhitby-away: and, /sbin/init -i works too (undocumented i guess)
12:22.06gielrwhitby-away: but you really have to make sure it is pid 1, or else you get a swapper process which eats up your cpu time
12:22.15rwhitby-awayhow do you make sure it is pid 1 when you're pivot_rooting to it?
12:22.26gielroot=/dev/ram0 init=/linuxrc
12:22.33gielthen your linuxrc execs as pid 1
12:22.47gielor, with devfs automount: root=/dev/rd/0
12:23.01rwhitby-awayyeah, yeah - that's how you do it without pivot_root.
12:23.04gielno no
12:23.08gielthat is the pivot_root solution
12:23.09rwhitby-awayI asked how you do it with pivot_root
12:23.41rwhitby-awaydidn't [g2] fix openslug to use that exact command line?
12:24.04rwhitby-away(sorry for arguing - you know more about this than me I think)
12:24.48rwhitby-awayok, my bad
12:25.14rwhitby-awayI forgot to change the cmdline_root back to /dev/ram0 ....
12:25.30rwhitby-awaythx giel - you saved me going down a rathole
12:25.49gieluhu
12:26.07gieli've discussed this for a while with the author of initrd
12:26.24gielnp
12:27.16rwhitby-awayok, let me try again with the right kernel cmdline
12:28.47*** join/#nslu2-linux siddy (~sid@p5092716F.dip.t-dialin.net)
12:42.49rwhitby-awayhmm :
12:42.54rwhitby-awayRAMDISK: Compressed image found at block 0
12:42.54rwhitby-awayVFS: Mounted root (ext2 filesystem) readonly.
12:42.54rwhitby-awayFreeing init memory: 72K
12:42.54rwhitby-awayBusyBox v1.00 (2005.02.21-11:08+0000) Built-in shell (ash)
12:42.55rwhitby-awayEnter 'help' for a list of built-in commands.
12:42.55rwhitby-away#
12:43.03rwhitby-awaygiel: any idea what's happening there?
12:43.27rwhitby-away# /linuxrc
12:43.27rwhitby-awayPreferred root device is jffs2
12:43.27rwhitby-awayRoot filesystem will be mounted from /dev/mtdblock4 ...
12:43.27rwhitby-awayUsage: init 0123456SsQqAaBbCcUu
12:43.27rwhitby-away#
12:44.13rwhitby-awayrunning /sbin/init -i after that boots OpenSlug properly ...
12:45.13rwhitby-awayroot@nslu2:~# ps -ef
12:45.13rwhitby-away<PROTECTED>
12:45.13rwhitby-away<PROTECTED>
12:45.13rwhitby-away<PROTECTED>
12:45.13rwhitby-away<PROTECTED>
12:45.14rwhitby-away<PROTECTED>
12:45.20rwhitby-away<PROTECTED>
12:45.20rwhitby-away<PROTECTED>
12:45.20rwhitby-away<PROTECTED>
12:45.21rwhitby-away<PROTECTED>
12:45.23rwhitby-away<PROTECTED>
12:45.25rwhitby-away<PROTECTED>
12:45.27rwhitby-away<PROTECTED>
12:45.29rwhitby-away<PROTECTED>
12:45.31rwhitby-away<PROTECTED>
12:45.33rwhitby-away<PROTECTED>
12:45.35rwhitby-away<PROTECTED>
12:45.37rwhitby-away<PROTECTED>
12:45.39rwhitby-away<PROTECTED>
12:45.41rwhitby-away<PROTECTED>
12:45.43rwhitby-away<PROTECTED>
12:45.45rwhitby-away<PROTECTED>
12:45.47rwhitby-away<PROTECTED>
12:45.49rwhitby-away<PROTECTED>
12:45.51rwhitby-awayroot@nslu2:~#
12:45.55rwhitby-awaythat's the process tree at the end of that manual running of /linuxrc and then manual running on /sbin/init -i
12:46.08rwhitby-awayI guess I should have run exec /sbin/init -i ....
12:46.54rwhitby-awayah - cmdline is wrong:
12:47.08rwhitby-awayroot=/dev/ram0 rootfstype=ext2,jffs2 initrd=0x01000000,10M /init=/linuxrc mem=32M@0x00000000 console=ttyS0,115200n8
12:47.16rwhitby-awayshould just be init=/linuxrc ...
12:47.25rwhitby-awaynot /init= ...
12:49.42gielyeah
12:55.21rwhitbyok, unslung still works after the change to switchbox and slingbox to start using /usr/sbin and /usr/bin again (so that the location of chroot matches openslug)
12:55.45rwhitby(I'll push everything once openslug works too)
12:59.37rwhitbyYES!
13:00.01rwhitbyOpenSlug works with switchbox 4
13:00.11rwhitbythx again giel
13:00.16rwhitby~worship giel
13:00.18jbotACTION bows down before giel.  "I'm not worthy!"
13:00.38perlguru-workWOW!
13:04.05CIA-103rwhitby 07 * r1.2836 10openembedded/ (10 files in 8 dirs):
13:04.05CIA-1Unslung and OpenSlug both working with switchbox. Updated CVSDATE stuff in
13:04.05CIA-1nslu2.conf. Changed CMDLINE in openslug kernel to use switchbox. Changed
13:04.05CIA-1switchbox and slingbox to use /usr/sbin and /usr/bin (just like in OpenSlug
13:04.05CIA-1busybox) so that the location of executables match across SwitchBox, Unslung and
13:04.05CIA-1Openslug. Added /initrd directory to OpenSlug rootfs so that it can be used for
13:04.07CIA-1the pivot_root in switchbox.
13:04.24rwhitbyPhew!
13:04.57rwhitbyI hope everyone feels all unified again now :-)
13:05.42rwhitbyCan I have volunteers to build OpenSlug and Unslung from scratch and make sure they work for you?
13:12.31rwhitbyDon't all step forwards at the same time ;-)
13:13.18perlguru-workno time, today. maybe tommorrow (if I can get my dev. env. to work again)
13:13.36rwhitbycool, thx perlguru-work
13:13.53perlguru-workstill got mu ipkg-native problem
13:13.58perlguru-workmu->my
13:14.20rwhitbywhat problem is that?
13:15.19perlguru-workit won't build. libbb problem. full_read already defined
13:16.10rwhitbybuild under which environment?
13:17.17perlguru-workbitbake unslung-image
13:17.44rwhitbydo you have the latest bitbake?
13:18.05perlguru-worklook at log of  14/2 around 19:00
13:18.13*** join/#nslu2-linux [g2] (~g2@rdu57-8-035.nc.rr.com)
13:18.24perlguru-work( that was also the last time i tried it)
13:18.25[g2]someone's been busy :)
13:18.27rwhitbymorning sunshine - I've got good news for you
13:18.33perlguru-workhey, [g2]
13:18.34[g2]I read the change logs
13:18.44[g2]FILES +=
13:18.45[g2]:)
13:19.02[g2]well changeset logs
13:19.26[g2]that's great news not good news
13:19.51[g2]hi perlguru-work
13:20.08rwhitby[g2]: it needs some independent testing, but it seems to work here ...
13:20.43rwhitbyand I haven't tested anything except switching to jffs2
13:22.28[g2]I've pulled and I'm baking now
13:23.22rwhitbyBTW, [g2], it's init=, not /init= :-)
13:23.57gielrwhitby: you might need to add 'rw' to the command line too
13:23.57[g2]hey :)
13:24.09gielrwhitby: if you're pulling drivers out of the payload or jffs2
13:24.19gielrwhitby: 'cause initrd gets mounted ro n ow
13:24.22gielnow*
13:24.46gielrwhitby: or add a "mount -o remount /" somewhere
13:24.58rwhitbygiel: yeah, looking at that now ...
13:25.10gielrwhitby: adding rw to the command line pbly does the trick
13:25.24rwhitbywe need to update switchbox anyway to get drivers from jffs2 instead of payload for nfsroot
13:26.00rwhitby[g2]: does the openslug kernel now have everything in there static for usb disk access?
13:26.42rwhitby[g2]: can we put a .openslug file in openslug-init to capture the version number?
13:28.31[g2]rwhitby, no it's modules
13:29.52[g2].openslug, sure  openslug-1.0-rc1 ?
13:30.01[g2]OpenSlug-1.0-rc1 ?
13:30.05rwhitbyok, that presents a problem for switchbox switching to external usb disk rootfs for openslug
13:30.30rwhitbyOpenSlug-1.0-alpha ?
13:30.52rwhitby[you know how much I love going straight past 1.0 and never looking back ...]
13:30.53[g2]-alpha-1
13:31.07rwhitbyno, next would be 1.1-alpha
13:31.37[g2]actually it'll probably be 1.1-beta soon :)
13:31.52rwhitbyone version number (incremented often) and the status after that
13:32.05rwhitbyMy money is on 1.5-beta
13:32.24[g2]ok -0.9-alpha ?
13:32.31rwhitby(based on Unslung history, it takes 5 versions from feature freeze to beta release)
13:32.54rwhitbyyou're really hooked on that magic 1.0 number, aren't you :-)
13:33.03[g2]or 0.5-alpha :)
13:33.31rwhitbywhat is it about 1.0 that people never want to get past it ?
13:33.52[g2]I'm fine with 1.0 for beta
13:34.12rwhitbywould you be more likely to try a new program if it is marked 1.0 or marked 1.6 ?
13:34.50rwhitbyto me, 1.0 screams "this is flakey and you should wait till the real release with the bug fixes" :-)
13:35.08[g2]google uses beta for years
13:35.40[g2]call it 1.1-alphs then :)
13:36.33[g2]1.x.1
13:36.38[g2]1.y
13:37.04rwhitby1.1-alpha it is then
13:38.24gielsiddy-away: msg me when you get back and have a couple of mins to spare
13:40.23rwhitby[g2]: it's good to all be working on the same thing again.  takes me back to the old days of nslu2-linux ....
13:41.03[g2]rwhitby, it's very fun
13:43.26[g2]rwhitby, I don't think the kernel pr number got bumped
13:44.41rwhitbyyeah, about to push the rw change, and I bumped it then
13:45.01rwhitbyjust flashing it now to make sure the version got in there
13:47.35rwhitbycan't push cause the repo is locked ...
13:49.51[g2]someone doing a pull ?
13:50.18rwhitbysomeone else doing a push it seems ...
13:51.10rwhitbyhmm - that's a long time for a push.  I'm going to remove the lock
13:51.11[g2]or a pull for oe-devel maybe ?
13:51.32[g2]can you see who's doing it ?
13:51.39CIA-103rwhitby 07 * r1.2836.1.1 10openembedded/packages/ (2 files in 2 dirs): OpenSlug-1.1-alpha, versioned, with rw initrd.
13:51.48rwhitbythere ya go
13:51.58*** join/#nslu2-linux VoodooZ_log (~VoodooZ@CPE00c0f021712c-CM014110003785.cpe.net.cable.rogers.com)
13:52.17VoodooZ_logstupid ssh connection died on me.
13:52.26VoodooZ_logrwhitby: a very late PONG.
13:52.56rwhitbyVoodooZ_log: I should be getting a bluetooth dongle (Linksys USBBT100) tomorrow in the post.  You know you are going to get some questions, right :-)
13:53.09VoodooZ_loghehehe NO problem mate.
13:53.33[g2]Applying   1 revisions to packages/kdepimpi/kdepimpi_2.0.9.bb
13:53.37VoodooZ_logCheck my badly written wiki first though and please make corrections if you find I forgot some important stuff.
13:53.50[g2]I'd guess that was the auto oe-devel pull
13:54.18VoodooZ_logrwhitby: Do you know how to use quilt to maintain my large private pool of changes to my repo?
13:54.42rwhitbynow [g2] knows the reason for getting switchbox working for openslug ...
13:54.46rwhitbyVoodooZ_log: nope
13:54.51VoodooZ_logok
13:55.08VoodooZ_logHow would you handle a bunch of private changes though?
13:55.30rwhitbyI don't have to - I make the decision to include them all ;-)
13:56.00VoodooZ_logyeah. That's what I though so. I can't include mine as I usually remove stuff or add weird stuff.
13:56.29rwhitbyIf I had to, I would create a new distro, and conditionalise stuff based on that.
13:56.45VoodooZ_logouch.
13:56.46rwhitbyconf/distro/voodooslug.conf
13:56.56VoodooZ_loghehehe
13:57.02rwhitbyor is that voodoozlug
13:57.04VoodooZ_logIt those sound good!
13:57.09VoodooZ_logsrx2_robot
13:57.50VoodooZ_logIs that simple though? Just creating a distro and configuring my local.conf accordingly?
13:57.57rwhitbyA new distro isn't that much of a deal - we did it for switchbox
13:58.09VoodooZ_lognothing else to do?
13:58.29VoodooZ_logBut that doesn't help
13:58.52VoodooZ_logas I still coulnd't push my private changes
13:59.03VoodooZ_logthey are useless to the community
13:59.18rwhitbyif they are conditionalise with your distro, then you can push them.
13:59.26VoodooZ_logok. i c
13:59.48[g2]VoodooZ_log, most of your changes are localized to the kernel right ?
14:00.00VoodooZ_logI guess eventually it could become useful to other robot hobbyist.
14:00.14rwhitbyyou just have to prove that they don't affect the other distros.  Just like we prove to kergoth that our nslu2-linux stuff doesn't stuff up the other distros in openembedded.
14:00.19VoodooZ_logwell, no. I had to change base-files, udev, bluez, etc...
14:00.50VoodooZ_logWell, most changes were actually fixes to other package's scripts to run under a busybox shell.
14:01.10VoodooZ_logbtw, did somebody push my grep-o-less.patch from the wiki?
14:01.27VoodooZ_logI guess it's a moot point now that we got rid of hotplug.
14:01.32[g2]don't think so,
14:01.36[g2]nod. on the hotplug-linux
14:02.22VoodooZ_logI guess I'll try to figure out what kergoth suggested using quilt to make patch-sets
14:02.36VoodooZ_logsimpler for now and it prevents contention with others.
14:03.46VoodooZ_logUnless I can just do it with a normal patch command?
14:03.46[g2]although I think the changes to use busybox may be valid changes to go upstream
14:03.47rwhitbyVoodooZ_log: I really think a new robot-friendly distro would be easier in the long run
14:04.04VoodooZ_logdefinitely.
14:04.41VoodooZ_logI just feel lazy and overwhelmed right now unfortunately as I have to redo all those changes. Some I forgot about.
14:05.14rwhitbyperfect time to get them into a new distro .... you know you won't do it later ....
14:05.48VoodooZ_loghehehe damn right
14:06.13VoodooZ_logok, but don't be surprised if I ask you questions! :)
14:06.35rwhitby2.06 hours on average between nslu2-linux commits since day 1, 5.5 months ago
14:06.58VoodooZ_loghow would I deal with distro specific changes like creating a /flash dir though? I previously modified base-files to do it.
14:07.00rwhitbyhttp://cia.navi.cx/stats/project/nslu2-linux
14:07.04VoodooZ_logwow
14:08.27VoodooZ_logOr even the interfaces file? do I just have my distro overwrite that file? How do you conditionalize?
14:08.56rwhitbyVoodooZ_log: look for nslu2, unslung and openslug directories in the repo ...
14:09.00rwhitbye.g. in base-files
14:09.23rwhitbyFILES_PATH is the key
14:10.13VoodooZ_logcool. I actually noticed that and put my fstab changes in there.
14:10.19[g2]/etc/rcS.d/S35mountall.sh: 23: test: not found
14:10.19[g2]/etc/rcS.d/S35mountall.sh: 42: test: not found
14:10.19[g2]/etc/init.d/rcS: 92: [: not found
14:10.22[g2]/etc/init.d/rcS: 92: [: not found
14:10.26[g2]/etc/init.d/rcS: 92: [: not found
14:10.30[g2]etc ....
14:10.50rwhitby[g2]: oh - did you have an old version of switchbox-4.8 ?
14:11.18[g2]it got pulled this moring or maybe last night
14:11.37[g2]I can delete it and rebuild
14:11.39rwhitbyremove switchbox-4.8.tar.gz* from your downloads dir
14:11.44[g2]nod.
14:11.47rwhitbyso it gets re-downloaded
14:12.03rwhitbyit was never mentioned in a .bb, so we didn't bother bumping the version
14:12.53rwhitbyheh, in 3 weeks, jp30 has an average of 8.5 hours between commits
14:13.15rwhitby<PROTECTED>
14:13.25VoodooZ_logslacker! ;)
14:13.35rwhitbyyeah, RLB
14:14.05rwhitbyhttp://cia.navi.cx/stats/author/jp30
14:14.43VoodooZ_WorkI wish I could use a longer nick so I could be called VoodooZ_Pretend2Work
14:15.39rwhitby[g2]: let me know when it works, cause then I'll go to bed
14:16.12rwhitbyshould be quick for just a switchbox change
14:17.10[g2]actually it didn't dl switchbox I think I need to remove the stamps also
14:17.16rwhitbyHmm - Gastro and Nudi are still offline ....
14:17.25rwhitbyyeah remove nslu2-switchbox-firmware
14:24.06rwhitbyhttp://slugsite.us/
14:24.33[g2]rwhitby, same thing with 242a308a53818f9d7887b5c4cb86ac36  ../sources/switchbox-4.8.tar.gz
14:25.16rwhitbyhmm
14:25.30rwhitbythat's the right md5sum
14:25.41rwhitbydid the image get rebuilt?
14:25.59[g2]sure
14:26.30rwhitbychanges were to openslug-init, the kernel, and switchbox.
14:26.50[g2]I pulled those
14:27.10[g2]I've obviously got the kernel changes or I'd be booting to mtd4
14:27.12rwhitbyI have no outstanding changes or local mods here ...
14:27.33rwhitbydid chroot run?
14:27.50rwhitby(before all those errors)
14:27.50[g2]dunno switchbox is quiet
14:28.09rwhitbyquiet?
14:28.27[g2]NET: Registered protocol family 1
14:28.27[g2]RAMDISK: Compressed image found at block 0
14:28.27[g2]VFS: Mounted root (ext2 filesystem).
14:28.27[g2]Freeing init memory: 72K
14:28.27[g2]Preferred root device is jffs2
14:28.27[g2]Root filesystem will be mounted from /dev/mtdblock4 ...
14:28.32[g2]maybe not so quiet
14:28.36[g2]:)
14:28.48rwhitbywhat's the next three lines after that?
14:29.05[g2]modprobe: FATAL: Could not load /lib/modules/2.6.9/modules.dep: No such file or directory
14:29.05[g2]INIT: version 2.86 booting
14:29.05[g2]modprobe: FATAL: Could not load /lib/modules/2.6.9/modules.dep: No such file or directory
14:29.05[g2]/etc/rcS.d/S35mountall.sh: 23: test: not found
14:29.07[g2]/etc/rcS.d/S35mountall.sh: 42: test: not found
14:29.09[g2]/etc/init.d/rcS: 92: [: not found
14:29.11[g2]/etc/init.d/rcS: 92: [: not found
14:29.15[g2]the modprobe's are normal
14:29.50[g2]then we are off to the other issue
14:29.54rwhitbywell that's weird.
14:30.06rwhitbyI've kicked off full rebuilds now ..
14:30.15[g2]ok
14:31.24[g2]Filesystem           1k-blocks      Used Available Use% Mounted on
14:31.24[g2]/dev/mtdblock4             955       395       560  41% /initrd
14:31.24[g2]/dev/mtdblock4            6400      4680      1720  73% /
14:31.24[g2]/dev/ram0                 6400      4680      1720  73% /
14:31.24[g2]tmpfs                    15324         8     15316   0% /var
14:31.24[g2]tmpfs                    15324         0     15324   0% /media/ram
14:31.29rwhitbyah - base-files
14:32.00rwhitbyI must have forgotten to bump the PR on base-files - I added jbowler's /etc/fstab fix
14:32.08rwhitbyand that's what causes those errors.
14:32.28[g2]I'll check the /etc/fstab
14:32.48[g2]/dev/ram0       /       ext2    defaults        1  1
14:32.50[g2]:)
14:33.32rwhitbyok, blow away base-files and you should be sweet
14:35.01CIA-103rwhitby 07 * r1.2839 10openembedded/packages/base-files/base-files_3.0.14.bb: Bumped the base-files PR for nslu2
14:35.03[g2]that's the issue,  patching the files in jffs2 works fine, I'll clean and rebuild
14:35.14rwhitbyjust pull and rebuild now :-)
14:35.15[g2]actually, I'll just pull and rebuild
14:38.00[g2]rwhitby, with initrd mounted we could make a little tool that enables switchbox changes from the OpenSlug cmd line
14:38.18[g2]and it can be umounted and freed normally
14:39.17rwhitbyinitrd is an ext2 fs from ram0 - you can't change it and make it stick
14:39.49[g2]you can make a tarball from it
14:40.05[g2]gzip it, and flash that
14:40.06[g2]right ?
14:40.20[g2]add header the header too,
14:40.33rwhitbyyou need genext2fs, don't you?
14:41.34[g2]actually we'd need to create a ram partition and copy it stuff to there
14:41.41[g2]then dd that
14:41.53rwhitbyright
14:42.00rwhitbythat'd work
14:42.31[g2]I think all the stuff is there or should be there
14:42.33rwhitbywe need to get openslug's init scripts (maybe linuxrc) to umount /initrd to free up the ram
14:49.07[g2]~emulate rwhitby
14:49.08jbotAnother Satisfied Customer!
14:49.23[g2]that's the ticket
14:52.33rwhitbycool
14:52.45[g2]I think ppl will be happy :)
14:53.01[g2]umount /initrd works
14:53.10rwhitbydon't forget initrd is still mounted and using up ram - can you deal with that in openslug-init ?
14:53.40[g2]sure, for now ppl maybe interestd in looking around in the ramdisk
14:54.01rwhitbyyep
14:54.53[g2]soon, after booting directly to the hd works (we just need to load drivers) we can free it up
14:55.04[g2]for the beta release
14:55.43rwhitbyyep.  Are you going to compile the drivers into the kernel for hd ?
14:56.30[g2]I'm 50/50 on the issue, 50 compile them in, 50 load from the switchbox linuxrc like the ethernet driver
14:57.21[g2]I'll probably first try with them compiled in, and see how close we are to the 1MB limit
14:57.31rwhitbycool
14:57.34rwhitbynight all
14:57.37rwhitby~emulate rwhitby
14:57.38jbotAnother Satisfied Customer!
14:57.44rwhitby~emulate rwhitb
14:57.49rwhitby~emulate rwhitby
14:57.50jbotAnother Satisfied Customer!
14:57.58[g2]we're using up space in the jffs2  partition cause the zImage is there
14:58.00rwhitby~insult jbot
14:58.11[g2]thx for the good work
14:58.42rwhitbynp - now OpenSlug is ready for my bluetooth dongle when it arrives tomorrow ...
14:58.47VoodooZ_WorkGreat work guys
14:58.55[g2]rwhitby, which one did you get ?
14:59.10rwhitbyLinksys USBBT100 - based on good reports with Treo650
14:59.17VoodooZ_WorkThey are all pretty much supported so it should smooth.
14:59.29[g2]is that a Class 1 device ?
14:59.32rwhitbyI checked bluez compatibility before buying it
14:59.35rwhitbyyep, class 1
14:59.41VoodooZ_WorkJust build and install bluez-nodbus and remove the pcmcia stuff.
14:59.53VoodooZ_Workclass1? cool. mine is class2
14:59.58rwhitbysome guy said he got 1000 feet with his Treo650 and the USBBT100
15:00.00VoodooZ_Workplenty for a small robot.
15:00.12VoodooZ_Worksucks more power though.
15:00.14rwhitbyand that was through walls
15:00.29[g2]yeah my IOGear goes quite far
15:00.40VoodooZ_Work[g2]: is it a class1 too?
15:00.45[g2]nod. class 1
15:00.58[g2]rwhitby-asleep, sweet dreams
15:01.02VoodooZ_WorkMine is rated 35 meters but only takes like 30mAh
15:01.12VoodooZ_Work30mA even
15:34.52VoodooZ_Workgood morning sir kergoth
15:38.19kergoths/good //
15:43.07VoodooZ_WorkI'm starting to look in using quilt for my private stuff as you suggested as I'm f'n tired of having to redo them everytime my repo craps out.
15:52.31VoodooZ_Workwow. I must be dumb as I as feel I need a quilt howto. :(
15:53.54kergothit has a lot of commands, definately go through a tutorial
15:53.58kergothquilt new patchname
15:54.00kergothquilt add filename
15:54.03kergoth<make changes to filename>
15:54.05kergothquilt refresh
15:54.19kergothquilt pop # to unapply
15:54.22kergothquilt push # to apply
15:54.34kergothquilt edit filename # to do a quilt add on the file, then spawn $EDITOR
15:54.41VoodooZ_Workyeah, but I'm having a problem understanding the concepts first. LIke the stack.
15:54.56kergothyou've never applied a patchset before?
15:55.02VoodooZ_WorkThe actual syntax is not hard after you understand it.
15:55.13kergothits a set of patches that must be applied in a specific order. thats all.
15:55.19VoodooZ_Workwell, I guess I need an actual example to get my terminology right
15:55.25VoodooZ_Workyeah ok
15:55.51VoodooZ_WorkBut in the context of OE: if I modify base-files_x.bb for example.
15:56.10kergoththe first ;patch=1 entry gets 'quilt import'd, then quilt pushed.
15:56.12kergoththen the second does
15:56.14kergoththen the third
15:56.17kergothapplied in order
15:56.29kergothquilt import pulls an external patch into the local patchset. quilt push applies it
15:57.16VoodooZ_WorkI'm trying to fit all this with my private changes so I'll have to try it first.
15:57.35kergothcd packages
15:57.39kergothquilt new mybase-files-changes
15:57.45kergothbk edit base-files/base-files_x.bb
15:57.50kergothquilt edit base-files/base-files_x.bb
15:57.56kergoth<make changes in your editor, save, exit>
15:57.58kergothquilt refresh
15:58.04kergothta da, the patch file is in patches/
15:58.08VoodooZ_Workok.
15:58.09kergothquilt pop to unapply it
15:58.12kergothquilt push to apply it again
15:58.33VoodooZ_WorkANd how easy is it to port over my patchset if my repo ever dies again?
15:58.38kergothport?
15:58.45VoodooZ_Workwell, more like move.
15:58.47kergothcp -a patches /somewhere/safe
15:58.52kergothcd /path/to/new/packages
15:58.57kergothcp -a /somewhere/safe/patches patches
15:58.58kergothquilt push -a
15:59.03kergoth<tap foot, wait for patches to apply>
15:59.08VoodooZ_Workok. sounds straightforward enough.
15:59.17kergoththe 'patches' directory contains a 'series' file
15:59.24kergoththe series file describes the patch application order
15:59.28kergothquilt will obey that
15:59.28VoodooZ_WorkPerfect. That's a great howto and in only 5 minutes.
15:59.54kergotha hell of a lot easier than having to cp the file to a .orig or something so you can manually do the diff, thats for sure
15:59.56VoodooZ_WorkI'll make sure to copy this whole conversation to my notes though. so I won't have to ask again.
16:00.02VoodooZ_Workdamn right!
16:00.30kergothone annoying quilt "feature" is that it does a chmod u+w on files when you add them to a patch
16:00.45kergothso if you havent done a bk edit on the file first, it'll screw up bk's perms on the file
16:00.49kergothand make it most unhappy
16:01.34VoodooZ_Workhehehe. I did that a few times to my repo. no wonder it's broken.
16:08.14*** join/#nslu2-linux perlguru (~maurice@cc444495-a.hnglo1.ov.home.nl)
16:09.30*** join/#nslu2-linux rwhitby-away (~rwhitby@rwhitby.nslu2-linux)
16:10.32perlgururwhitby-away, around?
16:11.18[g2]perlguru, sleeping
16:11.26perlguruok
16:11.30[g2]connection reset
16:11.50perlgurustill got my problem with bitbaking ipkg-native
16:12.04[g2]what the repo access ?
16:12.26perlguru?
16:12.38perlguruusing the normal repo
16:12.40[g2]what's the ipkg-native issue ?
16:12.49perlguru| all_read.c:45: error: conflicting types for 'full_read'
16:12.49perlguru| libbb.h:119: error: previous declaration of 'full_read' was here
16:12.49perlguru| all_read.c:45: error: conflicting types for 'full_read'
16:12.49perlguru| libbb.h:119: error: previous declaration of 'full_read' was here
16:12.49perlguru|
16:13.10[g2]hmmm
16:13.11perlguruipkg-native-0.99.140-r0:
16:13.22[g2]that's the new ipkg ?
16:13.30perlguruit does this, even after a clean build
16:13.44perlguruwhat do you mean?
16:13.56[g2]clean build as in rm -f tmp
16:14.06perlguru|  ccache gcc -DHAVE_CONFIG_H -I. -I/home/maurice/nslu2/unslung2/build/tmp/work/ipkg-native-0.99.140-r0/ipkg/C/libbb -I.. -I/home/maurice/nslu2/unslung2/build/tmp/staging/x86_64-linux/include -g -O -Wall -DHOST_CPU_STR=\"x86_64\" -DBUILD_CPU=x86_64 -I/home/maurice/nslu2/unslung2/build/tmp/staging/x86_64-linux/include -O2 -MT libbb_la-all_read.lo -MD -MP -MF .deps/libbb_la-all_read.Tpo -c all_read.c  -fPIC -DPIC -o .libs/libbb_la-all_r
16:14.06perlguruead.o
16:14.24perlguruyes, rm -rf build/tmp
16:14.36perlgurualso did a new clone
16:14.54[g2]did you flush ccache ?
16:15.12perlguruflush?
16:15.25[g2]clear the ccache
16:15.35perlguruhow?
16:16.13[g2]I think there may be options or you can just delete the directory if you don't mind losing the cached stuff
16:17.12perlguruif i delete tmp, the cache dir is also deleted. and i did this several times now.
16:18.00[g2]hmmm, I don't know if OE override the local ccache
16:18.12[g2]I know I've got a ~/.ccache dir
16:18.44perlgururunning ccache -c now
16:19.03[g2]that was the option I saw :)
16:19.14perlguruno luck
16:19.19perlgurustill get the error
16:20.25[g2]then it's probably x86_64 specific
16:20.49[g2]can you ship me your system for a little testing ? :)
16:21.05perlgurusure, come and get it ;)
16:22.02perlguruccache -C now...
16:23.25[g2]that and tmp should be *everything*
16:23.51perlguruok, removing tmp
16:24.30[g2]you could also rm the ipkg code from sources so it gets pulled again, but I doubt that's it
16:24.43perlguru(this could take some time, 'cause i'm emerging also (glibc))
16:24.59[g2]nod.
16:25.20[g2]Am I rushing you along ?
16:25.34[g2]:)
16:27.00siddy-awayhi [g2], concerning point 2) of your mail: installing "modutils-initscripts" solved it for me.
16:27.35[g2]siddy-away, thx I'll check what that point was
16:27.52siddy-awaythe update-modules.postinst message
16:28.10[g2]ah... yeah
16:28.41[g2]I'm not sure whether modutils-initscripts is 2.4 specific
16:29.00siddy-awaydunno, works quite well for me.
16:29.08VoodooZ_WorkI think it is.
16:29.13kergoththe init script modprobes all the modules listed in /etc/modules.  dont recall if it does anything else
16:29.20kergothdont see any reason why it wouldnt work for both
16:29.27[g2]most excellent
16:29.54[g2]kergoth, THX again for adding hotplug-ng, I don't know if you noticed but I pushed it for OpenSlug baseline
16:30.02VoodooZ_Workyep. thanks.
16:30.12[g2]it makes a *huge* difference
16:30.42siddy-awaythat could also help with point 5) (loading ixp425_eth and ehci-hcd). It could be done the "normal way" (via /etc/modultils/whatever)
16:31.23[g2]I'll test it right now
16:34.57[g2]siddy-away, which repo are you pointing at nslu2-linux or oe-devel ?
16:37.05siddy-away[g2], nslu2-linux
16:37.33[g2]you are aware of all the changes that went it recently ?
16:37.59siddy-awayyou mean the switchbox-stuff?
16:38.11[g2]yeah
16:38.55siddy-awayI've read that, but haven't had the time to take a look at it.
16:39.00[g2]it should be in a good state now, but I think it was a little hosed earlier
16:39.32siddy-awayAnd I won't have the time in the near future.... :-\
16:40.55[g2]just FYI
16:41.10[g2]and thx for all the help
16:41.12siddy-awaywill booting to usb-disk work with that switchbox out of the box?
16:41.33[g2]that's what will be testing very soon
16:41.49[g2]it does for unslung, but there kernel has all the modules built in
16:42.13[g2]we'll either change the default kernel config or explicitly load the modules from switchbox via the jffs2 partition
16:42.32siddy-awayoh, not the problem with my current kernel. I already have all thet usb-stuff in there.
16:42.42[g2]and we should be able to boot to .sda[1|2] and .sdb[1|2]
16:43.54siddy-awayhm... of course I need sdd1, so once again I need switchbox-tuning! ;)
16:44.16[g2]currently the initrd is on /initrd
16:44.53[g2]I think we might be able to setup the necessary stuff (if it's not already there) to be able to modify it from the OpenSlug cmdline for the brave and bold
16:50.41[g2]siddy-away, cool that works for both
16:50.54[g2]now I've just got one more question
16:51.01siddy-away?
16:51.34[g2]I've manually added the two modules to the /etc/modules file but the comments in there say "don't edit" this file it's generated by update-modules
16:51.48[g2]what's the proper way of doing this ?
16:51.55VoodooZ_Workbase-files?
16:52.01siddy-awayyou'll have to create a new file in /etc/modutils and add the two modules there.
16:52.21siddy-awaythe /etc/modules is created by update-modules iirc
16:52.52siddy-away(by copying all files in /etc/modutils/)
16:53.41[g2]that's simple enough
16:54.33[g2]and we may or may not have to run the update-modules earlier in the init sequence
16:55.25*** join/#nslu2-linux ByronT (ByronT@adsl-220-231-152.bhm.bellsouth.net)
16:55.39siddy-awayyes, we'll especially have to take care of not braking the first boot, so the ordering may be important.
16:56.38VoodooZ_Workperlguru: I just tried a fresh installtion of OE on a brand new Sun quad opteron and it crashes at ipkg-native too.
16:57.01VoodooZ_Workthere goes my chance to "borrow" this nice new machine.
16:57.17[g2]root@nslu2:/etc# cat /etc/modutils
16:57.17[g2]# Adding modules to be automatically loaded via update-modules
16:57.17[g2]ipx425_eth
16:57.17[g2]ehci-hcd
16:57.35[g2]siddy-away, I ran update-modules after that but it didn't update /etc/modules.conf
16:57.43[g2]well /etc/modules
16:57.45VoodooZ_WorkWhat about ohci-hcd?
16:58.13[g2]VoodooZ_Work, I think every thing gets loaded via hotplug
16:58.20perlguruVoodooZ_Work, good to know I am not the only one...
16:58.27siddy-awayerr... you need a directory /etc/modutils/ and place some file in there.
16:58.28VoodooZ_Workcool.
16:58.39VoodooZ_Workperlguru: still. it sucks.
16:58.45perlguruand both using AMD64
16:58.52perlguruyes, indeed
16:58.55VoodooZ_Workperlguru: What else did you find out about this problem.
16:58.57[g2]VoodooZ_Work, ok ohci doesn't get loaded
16:59.06[g2]but everything else
16:59.12VoodooZ_Work[g2]: could be useful for USB 1.1 support.
16:59.31[g2]hotplug works for usb
16:59.32VoodooZ_Workperlguru: did you google for this?
16:59.39VoodooZ_Work[g2]: good.
16:59.40[g2]except for net
16:59.42perlguruno
17:00.20VoodooZ_WorkI was looking forward to using the quad opteron to cut my build times in half so I could try more stuff in the same time.
17:01.09perlguruVoodooZ_Work, the funny part is: it used to work
17:01.44VoodooZ_Workperlguru: what linux distro are you using ?
17:01.51perlguruGentoo
17:01.51VoodooZ_Workreally?
17:01.55[g2]siddy-away, THAT works better :)
17:01.57VoodooZ_WorkI'm using fedora core 3.
17:02.20VoodooZ_WorkWhen did it work? did you find out what the change was?
17:02.37VoodooZ_Workif we can pinpoint it then we can ask the OE guys perhaps.
17:02.46perlguruno, not yet. i have problem since XMas
17:03.04perlgurulast succesfull build was just before xmas
17:03.09VoodooZ_Workok. I'll check it out soon.
17:03.18VoodooZ_WorkI have to run now. Lunch time. bbl
17:04.06perlguruVoodooZ_lunch, last image was of 20050128
17:04.42ByronTFrankenslug was purchased on Saturday...
17:08.25ByronTFrakenslug = my test slug
17:28.18[g2]siddy I've got a chicken and egg issue with the update
17:28.31[g2]Starting portmap daemon: portmap.
17:28.31[g2]Configuring kernel-module-ehci-hcd
17:28.31[g2]/etc/modutils/modutils.txt: 2: ixp425_eth: not found
17:28.31[g2]/etc/modutils/modutils.txt: 3: ehci-hcd: not found
17:28.31[g2]/etc/modutils/modutils.txt: 4: ohci-hcd: not found
17:28.32[g2]Error while executing /etc/modutils/modutils.txt, aborting
17:28.34[g2]Configuring kernel-module-ext3
17:28.36[g2]/etc/modutils/modutils.txt: 2: ixp425_eth: not found
17:28.38[g2]/etc/modutils/modutils.txt: 3: ehci-hcd: not found
17:28.40[g2]/etc/modutils/modutils.txt: 4: ohci-hcd: not found
17:28.42[g2]Error while executing /etc/modutils/modutils.txt, aborting
17:32.04perlguru'night all
17:33.36[g2]nite
17:35.57ByronTbtw, when I purchased Frankenslug, I was informed by the Beast Buy guy that there were custom firmwares for the NSLU2
17:39.48*** join/#nslu2-linux CIA-11 (~CIA@flapjack.navi.cx)
17:44.04siddy-away[g2], "executing /etc/modutils/modutils.txt"?! Something looks strange here, that file should definitely not be executed.
17:49.04siddy-away[g2], looking at /usr/sbin/update-modules I guess you made /etc/modutils/modutils.txt executable, so update-modules will execute it. chmod 644 should help.
17:49.38[g2]thx will try
17:59.30[g2]siddy-away, IS THE MAN!  :)   THX we are really close now
17:59.45siddy-awaygreat!
18:00.29[g2]it's all works except the ixp425 driver starts after the networking has already run so the networking isn't brought up automatically
18:01.29[g2]I think the short-term fix is to leave the explicit load of the ixp425 driver in the sysconfsetup and automatically load ehci and ohci
18:02.04[g2]THX for helping speed the process along
18:02.53siddy-awaynp!
18:10.10kergothheh, couldnt you cheat, and do it the way network modules are loaded on a lot of desktops?
18:10.31kergoth(that is, alias the interface name to the driver module, and let it get autoloaded)
18:10.45kergothalias eth0 ixp425_eth .. or whatever
18:10.55kergothin modprobe.conf
18:11.12[g2]kergoth, I could try that too
18:11.41[g2]but I'm not sure if that's more of a drawback than explicitly loading the module
18:12.25kergothcertainly seems cleaner to me.  no more ordering problems, it gets loaded the first time the interface comes up, regardless of when that happens
18:12.49VoodooZ_lunchkergoth: have you heard of a problem building ipkg-native under linux on x86-64?
18:13.12kergothask cosmicpenguin
18:13.17VoodooZ_lunchthanks
18:13.20kergothhe's done x86_64 builds more recently than i
18:13.27VoodooZ_lunchk
18:14.15[g2]siddy-away, that works!
18:14.45[g2]kergoth, I'll have to looks at the alias option too
18:15.01siddy-awaythe way kergoth proposed? yeah, that sounds good.
18:15.23[g2]siddy-away, no, the explicit module right now
18:15.49[g2]I was reflashing when kergoth suggested that
18:16.04[g2]I'd like to understand how that works and try it later
18:16.23[g2]and also understand the tradeoffs
18:16.50*** join/#nslu2-linux [cc]smart (~smart@gw.ptr-62-65-149-158.customer.ch.netstream.com)
18:17.14[g2]by default we are bringing up ixp as eth0, there are a couple ppl that are tailoring it for BT / usbnet, etc.... already
18:17.45[g2]right now its just comment out the 1 modprobe in the sysconfsetup init file
18:17.54[g2]and get the driver loaded
18:18.26[g2]the net hotpluging is an issue with hotplug-ng which will need to be worked out
18:18.59siddy-awaythat would be changed to modifying one alias, not much difference.
18:19.19siddy-awayhotplug... hm... always wondering what we need that for in openslug.
18:19.45[g2]plugging in drives, adapters etc...
18:19.53kergothnet is just a matter of splitting a couple of scripts from linux-hotplug into a hotplug-common package that both linux-hotplug and hotplug-ng RRECOMMENDS.
18:20.34[g2]kergoth, nod. I'm reading the code and doc's and trying to come up to speed on that
18:21.00[g2]and I need to get a grip on how the alias works
18:21.02siddy-awayguys, I'm now really away, meal-time.
18:21.20[g2]me too. lunch is on the table
18:21.35kergothan alias is just that.
18:21.37kergothan alias
18:21.54kergothif i alias ixp425_eth to eth0, then when i run 'modprobe eth0', it loads ixp425_eth
18:22.01kergothjust an aliased name for a module
18:22.17kergothand it happens that when you go to bring up an interface, it attempts a modprobe.
18:23.07[g2]ok so by adding the module alias the S40networking auto will modprobe eth0 a.k.a ixp425_eth and bingo it'll work
18:23.19kergothright.
18:23.35[g2]that's a crystal clear explaination, THX
18:23.43kergothnp
18:24.10[g2]I'll look to add the alias after lunch and test that out. it is a cleaner change
18:24.16[g2]~praise kergoth
18:24.18jbotAll hail kergoth!
18:24.39[g2]~praise siddy-away
18:24.41jbotAll hail siddy-away!
18:24.59[g2]bbiab -- lunch
18:32.47*** join/#nslu2-linux jp30 (~josh@dcn236-147.dcn.davis.ca.us)
18:50.05*** join/#nslu2-linux garpinc (~garpinc@pool-151-203-198-44.bos.east.verizon.net)
19:10.20CIA-1103jp30 * 10unslung/ (make/openldap.mk sources/openldap/hostcc.patch): openldap cross compilation stuff
19:10.40*** join/#nslu2-linux cbrake (~cbrake@a22c174.neo.rr.com)
19:16.32CIA-1103g2 07 * r1.2839.1.1 10openembedded/ (5 files in 3 dirs): Fix modutils punchlist issues
19:17.23VoodooZ_lunchWow! Thanks to cosmicpenguin on OE I was able to fix the AMD64 issue and my build with no psyco JIT is 32minutes on the quad opteron!
19:17.37[g2]-awayVoodooZ_lunch, COOL!
19:17.41VoodooZ_lunchANd that's without tweaking the -j 6 yet.
19:17.49VoodooZ_lunchI'll try again with 7 and 8 for fun.
19:17.53[g2]-awaykergoth, the alias works fine and that's what's checked in
19:18.37[g2]-awayVoodooZ_Work, was that the ipk-native issue that perlguru has too ?
19:18.43VoodooZ_Workyep.
19:18.47[g2]-awaycool
19:18.54[g2]-awayI'm sure he'll be happy
19:19.00VoodooZ_WorkIf you see him before I do let him know that it's a bug in the latest version
19:19.10[g2]-awayoff for a bit
19:19.35[g2]-awayyou might want to try pulling from nslu2-linux and building OpenSlug, the punch list issues are resolved
19:19.39VoodooZ_Workhe can workaround it by add this to his local.conf: PREFERRED_VERSION_ipkg-native = "0.99.138"
19:19.39VoodooZ_WorkPREFERRED_VERSION_ipkg = "0.99.138"
19:19.52[g2]-awayjust tweak your alias in modprobe.conf for BT
19:19.52VoodooZ_WorkI just did a pull.
19:19.58VoodooZ_Workcool.
19:20.04[g2]-awaycool, I'll check the log when I get back
19:20.10VoodooZ_Worklater.
19:20.14VoodooZ_WorkANd good work
19:21.25jp30i have a question about psuedo-terminals on unslung: unslung uses bsd-style ptys rather than the /dev/pty filesystem, right? how does a unprivileged program like xterm get control of a ptyp/ttyp pair on a system like that? normally on glibc you can call getpt(), which hides the details of the difference between bsd and unix98; on unslung getpt() returns -1 unless the caller is root. looks like the problem is the permissions of /dev/ptyp0 and f
19:21.25jp30riends
19:30.51*** join/#nslu2-linux pTweety (~pTweety@dialin-145-254-197-135.arcor-ip.net)
19:31.31CIA-1103ptweety * 10unslung/ (11 files in 3 dirs): corrected permission problem.
19:31.38CIA-1103ptweety * 10unslung/sources/perl-html-parser/control: corrected permission problem.
19:31.44CIA-1103ptweety * 10unslung/sources/perl-html-tagset/control: corrected permission problem.
19:32.10CIA-1103ptweety * 10unslung/sources/ (perl-mime-base64/control perl-net-dns/control): corrected permission problem.
19:32.38CIA-1103ptweety * 10unslung/sources/ (3 files in 3 dirs): corrected permission problem.
19:33.34pTweetysorry for this multiple messages, but sf.net seems to be verry slow today
19:45.55jacquesjp30, I've run into that problem too
19:46.34jacquesI *think* it went away when I mounted /dev/pts like so:
19:46.37jacquesnone                    /dev/pts                devpts  gid=5,mode=620  0 0
19:47.57jp30jacques, thanks, might also need a /dev/ptmx. what say I make a package that does this?
19:48.54jacquesthis probably needs to be addressed on the image creation level rather than package
19:49.16jp30yes, I suppose that's right
19:50.30jacquesperhaps a slugbug is in order
19:50.44jacques(if there isn't already one - I seem to remember other people finding this)
19:52.03jp30jacques, will do. strictly speaking this is extra functionality rather than a bug, though. the bug is presumably the permissions on the bsd-style ptyp nodes
19:52.38*** join/#nslu2-linux NSLU2-IRC (~NSLU2-IRC@cpc1-brmb1-3-0-cust185.bagu.cable.ntl.com)
19:52.44NSLU2-IRChey peeps
19:53.06Gorxive gotta prob for you all
19:54.31ByronTis it a developer problem or a user problem?
19:54.36ByronT~nslu2-general
19:54.37jbot#nslu2-general is the IRC channel specifically set up for end-user questions about the NSLU2 device. Questions that are not related to the nslu2-linux development projects should not be asked in #nslu2-linux, but should be asked in #nslu2-general instead.
19:55.59*** join/#nslu2-linux NSLU2-IRC (~NSLU2-IRC@kotnet-147.kulnet.kuleuven.be)
19:57.50*** join/#nslu2-linux cbrake (~cbrake@a22c174.neo.lrun.com)
19:59.48bob_tm-awayByronT: Taking care of it
20:01.07*** join/#nslu2-linux FireEgl (Ariel@Atlantica.IPv6Tunnel.Info)
20:01.23sk8jkrhi. I need some advice
20:01.44sk8jkrI have been working on getting usb audio support on the nslu2
20:01.53sk8jkrwhich is now completed
20:02.18sk8jkri have an nslu2 running usb audio and acting as a music server now
20:02.44sk8jkri based the server on mpd
20:03.07sk8jkra lot of packages were built and I can add them to the cvs rep
20:03.20sk8jkrno problem there
20:03.30sk8jkrproblem lies in the kernel changes
20:03.39sk8jkri did those in openembedded
20:03.57sk8jkrbut am not sure how I need to package them up for nslu2 distribution
20:04.15sk8jkrand how to properly insert these driver bitbakes into OE
20:04.36sk8jkrplease advise
20:10.10dyoung-zzzzdid you make it into a changeset>
20:10.54sk8jkrno, not yet. would like to know where to add it first
20:11.16sk8jkrnow added alsa-driver 1.0.8 under packages/alsa
20:11.26sk8jkrand did patches to that to cross compile it
20:11.47sk8jkrbut it might have to sit somewhere else
20:12.00sk8jkri also need a change in defconfig of unslung-able
20:12.16sk8jkrand an extra kernel module included in the base build
20:13.50CIA-1103jp30 * 10unslung/make/gtk.mk: gtk: configure input modules on install; don't get confused by target gdk-pixbuf-csource in staging directory
20:14.14CIA-1103jp30 * 10unslung/sources/gtk/postinst: gtk: configure input modules on install; don't get confused by target gdk-pixbuf-csource in staging directory
20:16.37dyoung-zzzzum..
20:16.56dyoung-zzzzhow did you make these kernel changes?
20:17.54sk8jkri changed the following:
20:18.40sk8jkrCONFIG_SOUND=m in packages/linux/unslung-kernel/able/defconfig
20:19.25sk8jkrthat gives me audio hooks in the kernel
20:19.44sk8jkrthen I patched the autotools files for alsa-driver to have it cross compile
20:19.48sk8jkrno C code changes
20:20.21sk8jkroh yeah, and added alsa-driver_1.0.8 to OE, as it is not there
20:20.42sk8jkrthey seem to rely on 2.6 kernels that include it for other projects
20:20.57CIA-1103ptweety * 10unslung/Makefile: perl-db-file and perl-dbi ready for native_only_testing
20:21.16CIA-1103ptweety * 10unslung/ (4 files in 2 dirs): perl-db-file and perl-dbi ready for native_only_testing
20:21.51CIA-1103ptweety * 10unslung/sources/perl-dbi/control: perl-db-file and perl-dbi ready for native_only_testing
20:22.56jacquessk8jkr, sounds like good work
20:23.09jacques:-\
20:23.10*** join/#nslu2-linux NSLU2-IRC (~NSLU2-IRC@kotnet-147.kulnet.kuleuven.be)
20:23.23Tierstensk8jkr: It's a Java app in the browser.  Don't leave the page :)
20:23.49NSLU2-IRCjust found that out the hard way
20:25.38pTweetyWondering what's wrong with cyrus-imapd.mk and postfix.mk.
20:25.57pTweetyI have no problem with these packages.
20:26.16pTweetyCan someone please send me the log, so I can fix the problems?
20:26.43jacquessk8jkr, you should make a changeset and sent it to nslu-changesets@lists.sourceforge.net - then we can work on getting it pushed into oe
20:29.59dyoung-zzzzsigh
20:30.11dyoung-zzzzoh hes back now
20:30.29sk8jkrok, will do
20:30.55dyoung-zzzzsk8jkr, if youve used bk edit to edit these files; you can do a bk ci of these files and then a bk commit to get a changeset number
20:31.12dyoung-zzzzthen you can bk send -r (some changeset number) nslu-changesets@lists.sf.net
20:31.29sk8jkryes, read that on the Wiki
20:32.08sk8jkrthis was the plan all along, just wanted to know if i am ok with file locations
20:32.21sk8jkrbasically will they allow a new entry in packages/alsa
20:33.18sk8jkrin order to get the right ipk for installation on a slug I have to put in nslu2-specific stuff
20:33.33sk8jkrlike for instance just packaging usb-audio and not all other sound cards
20:33.58sk8jkri thought maybe I should have a nslu2-alsa-driver or something that builds the ipk we need on nslu2
20:34.03dyoung-zzzzKergoth will be better equipped to answer that.
20:34.22jp30pTweety, i have a slug sitting idle. would it be useful for you if i try building cyrus-imapd and see if it fails?
20:34.25dyoung-zzzzI'm not sure what the policy is there.
20:34.40pTweetyyes, verry useful
20:34.59jp30OK, will do, then...
20:35.09jacquesprobably to just have a variable which configures which alsa sound card drivers to built
20:35.09pTweetythx
20:35.36jacqueslike gentoo has: ALSA_CARDS="emu10k1"
20:35.38sk8jkrit's a parameter to ./configure
20:35.55jacquesyeah I know
20:36.45Gorxany1 got a usb printer to work on this little device/
20:36.52sk8jkrI can do it that way. but it will still package alsa header files to /usr/include/sound
20:38.03jp30pTweety, is the only reason cyrus-imapd can't build cross that it likes to build & run binaries at compile time?
20:38.48pTweetyAFAIK, yes. But I checked it not to deep
20:39.48jp30you didn't try configuring it to use HOSTCC in appropriate places?
20:40.54pTweetyI tried with HOSTCC but found that I have to patch the source an several places (to much for me)
20:41.49jp30ulch,... yes that's understandable. well, it's native building OK for me so far
20:41.57*** join/#nslu2-linux Kirk_22 (~Administr@pool-141-152-138-141.norf.east.verizon.net)
20:44.34pTweetyrwhitby, are you arround?
20:46.27ka6sox-away7:16am
20:50.28jp30pTweety, make cyrus-imapd just failed with "armv5b-softfloat-linux-gcc: Internal error: Killed (program cc1)". yurgh!
20:51.50jacquesmight have just run out of memory
20:52.00jp30this was while building chartable.c.
20:52.05pTweetycan you send me the complete log by mail?
20:52.22jacquesI've seen gcc ICE a lot when running our of memory
20:52.25jp30jacques, sounds plausible, it was thrashing the disk before it died... no -pipe in this build. filled up /tmp?
20:52.49pTweetyhm, but my slug is not verry special
20:53.01jacquesdepends how much other stuff you have running
20:53.31jacquessomeone with lots of crap running will see that more than someone who kills everything not necessary for compiling
20:53.45pTweetyhm, nothing more than sshd
20:53.49jp30apache and slapd are running on my slug, that's probably the relevant difference
20:54.04jacqueslol
20:54.21jacquesapache is great, but it does like to use ram
20:54.38pTweetyjp30, can you please retry without apache?
20:54.53jacquesdo you also have standard 56MB swap? that's barely enough
20:54.59jacquesI have 128MB swap on mine
20:55.16pTweetyjepp, I ha everthing default
20:55.22jp30pTweety, will do. might help to try forcing -pipe into CFLAGS, too
20:56.11pTweetyunslung 3.17 with an unmodded slug
20:56.36jacques-pipe just increases ram usage
20:56.49jacqueswhich is why I lobbied for its removal
20:58.28jp30in this case i think the problem might be running out of /tmp. assuming that gcc using /tmp, not /opt/tmp on the slug. i will try without -pipe and without apache running.
20:59.21jp30(the file it's dying on appears to be a very large generated source, making for a large temporary assembly file with -pipe)
20:59.42jp30s/with/without/
21:00.16jacquessigh
21:00.27jacquesI would like it better if you changed one variable at a time
21:00.42jp30jacques, that's what i am doing.
21:00.51jp30i've just switch off apache
21:01.06jacquesok, sorry I misread
21:01.29jacquesI thought you said "with -pipe and without apache running"
21:02.35jp30jacques, s'ok... :)
21:03.02jp30ok, it is thrashing swap again, let's see if it will die this time
21:03.12jacquesjp30, I don't support you have top installed? I use that to monitor how big gcc is getting
21:03.25jacquesI know it takes ram itself, but it's useful in these cases
21:03.39jacquesyou can watch ram ans swap usage
21:03.49jp30sure do, gcc is using ~41M vm, 12M of which is resident
21:03.57jacquesouch
21:03.57jp30looks like I was wrong about /tmp
21:04.14jacquesthis is gcc and now g++ ? g++ takes tons of ram
21:04.17jacquesnow
21:04.19jacquesnot
21:05.53jp30recent g++ seems to have been getting hungrier. not unusual to see it grabbing 100M in a single allocation :(
21:06.15jacquesjp30, hmm why can gcc only get 12MB resident? should be able to get up to 24MB or so
21:06.56jacquesthe reason I lobbied against -pipe for native builds is that gcc and as would run at same time, contending for cpu and vm
21:06.59jp30not sure about that, it's down to about 9MB or so now. puzzling
21:07.10jacquesand as can also get big - I've seen it get > 10MB
21:07.49jacquesuilds were actually slower with -pipe (and some would not even build due to running out of vm)
21:08.10jacquesthis is all native of course, cross still uses -pipe
21:08.19jp30jacques, re -pipe, that sounds reasonable... so long as the temporary files are going in /opt/tmp where there's plenty of space for them
21:10.00jacquesI get what you're saying about temporary files, but I just haven't hit that issue on the nslu2
21:10.08jacquesI don't know where they are going :-\
21:12.35jp30two things to report: the cyrus-imapd build has gotten past the place where it died last time; and native gcc uses /tmp for its temporaries
21:14.49jacquesperhaps a judicious use of a certain ENV var will fix that?
21:14.49pTweetythen we have to wait for another 15 mins (if I recall corectly my cyrus tests)
21:15.48[g2]-awayVoodooZ_Work, any luck with the pull/OpenSlug build ?
21:16.43VoodooZ_Workyep.
21:16.53VoodooZ_WorkWorks great and builds fast now!
21:17.04VoodooZ_WorkActually, I didn't try the default.
21:17.17rwhitbypTweety: gastro is down at the moment, so I can't pull up the failure log
21:17.21[g2]rwhitby, morning sunshine !
21:17.24VoodooZ_WorkI commented out the default commandline and uncommented the jffs2 direct one.
21:17.28rwhitbywas there anything I wrote in the Makefile about it?
21:17.31VoodooZ_Workboots in ~40secs
21:17.32rwhitby[g2]: mrning
21:17.35pTweetynop
21:17.50rwhitby[g2]: everything cool with switchbox overnight?
21:18.13pTweetyrwhitby, you simply demoted cyrus-imapd and postfix
21:18.19[g2]rwhitby, think so the rest of the punchlist was fixed today
21:18.32VoodooZ_WorkI still have to put in my private changes in though so I'll try to optimize other things to get to 30seconds flat.
21:19.10[g2]siddy-away, kergoth and others had some suggestions and insights and it's all checked in and running
21:19.16rwhitbypTweety: I wrote some comments in the Makefile - search for postfix
21:19.44pTweety-feedingbgrr, "ssh: connect to host cvs.sourceforge.net port 22: Connection refused"
21:19.53rwhitby[g2]: are all the PRs bumped? Can I just bitbake without cleaning?
21:20.00[g2]rwhitby, I didn't get to play the /linuxrc much or booting directly to the hd, but I think that's the only issue
21:20.06[g2]rwhitby, nod on the PRs
21:20.11rwhitby# cyrus-imapd fails when compiling the perl stuff
21:20.11rwhitby# postfix fails when packaging
21:20.17[g2]rwhitby, pull first :)
21:21.23rwhitby[g2]: building now, but no time to test this morning
21:23.35rwhitbybase-files, openslug-init and modutils-initscripts - any others?
21:24.22[g2]openslug-init
21:24.40[g2]I guess that was the second one :)
21:25.03[g2]distro/openslug.conf
21:25.38[g2]rwhitby, you should be fine, VoodooZ_Work pulled and built with it
21:25.56rwhitbyflashing now
21:26.23VoodooZ_Workyep.
21:26.57rwhitbyso is this 1.2-alpha ;-)
21:27.26dyoung-zzzzthat was fast
21:27.30*** join/#nslu2-linux NSLU2-IRC (~NSLU2-IRC@62.21.190.133)
21:27.35[g2]tiny changes
21:27.41dyoung-zzzzwhat happened to 1.0?
21:27.52dyoung-zzzzin fact, wasnt there a 0.7 ?
21:27.56[g2]dyoung-zzzz, rwhitby is 1.0 phobic
21:27.57rwhitbyyep, that's what the number after the decimal point is for :-)
21:28.07dyoung-zzzzEvery time I blink, the version changes.
21:28.16dyoung-zzzzYou guys are too fast.
21:28.17dyoung-zzzzsigh
21:29.01rwhitbymodprobe: FATAL: Could not load /lib/modules/2.6.9/modules.dep: No such file or directory
21:29.05[g2]it's a plan rwhitby and I have to make you feel like a slacker :)
21:29.16VoodooZ_Workrwhitby: that normal on first boot
21:29.24[g2]rwhitby, on first boot you get a couple of those
21:30.45rwhitbyok, second boot is squeaky clean
21:30.50rwhitbynice job!
21:31.00[g2]and to you mate!
21:31.01VoodooZ_Workcongrats!
21:31.03pTweetyjp30, any news about the cyrus compile?
21:31.12[g2]~praise nslu2-linux
21:31.15jbotAll hail nslu2-linux!
21:31.17rwhitbynow we just need a package feed, right?
21:31.37[g2]a lot of packages work
21:31.51dyoung-zzzzis openslug supposered to be booting using switchbox?
21:32.12dyoung-zzzzor do I need to manually change that?
21:32.15jp30pTweety, it is still running... I suggest you add TMPDIR=/opt/tmp to the environment though, that should make it less memory hungry
21:32.30rwhitby[g2]: nod.  that's why we need a feed - so I don't have to work out which :-)
21:32.34dyoung-zzzz(ie: did I miss some push in the last few hours)
21:32.36[g2]dyoung-zzzz, openslug boots using switchbox since last night
21:32.45dyoung-zzzzaw crap.
21:32.47[g2]rwhitby, put the changes ing
21:32.48dyoung-zzzzok
21:33.12[g2]rwhitby, we need to get OpenSlug booting to hd
21:33.33rwhitbyI recommend compiling in the required modules into the kernel
21:33.45rwhitbythen it will work for recovery from flash key in recovery mode
21:34.14[g2]ok, do you want to play with the defconfig and make that happen ?
21:34.18rwhitbyyep
21:34.44[g2]ok go for it, just make sure the kernel stays < 1M
21:34.45rwhitbymy rule of thumb is that if the hardware is permanently attached, then compile in instead of make module.  then modules for everything else
21:35.02rwhitbyoh - I thought you said "do you want me to play .."
21:35.05pTweetyjp30, where do I add this? To configure or make?
21:35.10rwhitbyI don't have time today, can you do itA?
21:35.28[g2]actually I think there's an exception to that rule.
21:35.32rwhitbyif not, I'll look at it in 13 hours
21:35.38rwhitbyyeah, exception is ixp
21:35.41[g2]I think ehci has a bug in it
21:35.49rwhitbyoh
21:36.03[g2]which will need patching, I've got a stack dump from the ehci isr
21:36.09jp30pTweety, it would need to in the environment of the make invocation that actually does the build. e.g. "TMPDIR=/opt/tmp $(MAKE) -C ..."
21:36.15rwhitbyok, gotta go
21:36.33[g2]rwhitby-away, cheers
21:36.37[g2]and THX
21:37.02rwhitby-away[g2]: I think we're beta-ready now.
21:37.09VoodooZ_loglater...
21:37.16dyoung-zzzzJust need some alpha testesr
21:37.37jp30pTweety, my build has completed without trouble. "Works For Me"
21:37.38[g2]rwhitby-away, I think so, just a few more users, a manifest/snapshot
21:37.49rwhitby-awayand a README ?
21:38.42[g2]I was thinking maybe the NTP fix and IXP as interrupt in the driver
21:38.51[g2]and maybe NET console
21:39.22dyoungso the question is... Did they diddle the tick *because* of the ixp polling?
21:40.05[g2]that's a good question, if we go to interrupt ixp driven "do we even care" ?
21:41.16[g2]the interrupt driver ixp needs testing, but I really like the idea of it because on balance I'm guessing the ethernets these boxen are running on are on average less loaded
21:41.27pTweetyIs it ok to change TMPDIR to /opt/tmp? On my slug that dir does not exist by default.
21:41.57[g2]also, we could change to /dev/random instead of /dev/urandom for the dropbear keygen and probably ssh
21:42.13[g2]we would get some entrophy
21:42.24dyoungThats a scary message.
21:42.38dyoungwhen it tells you there isnt enough entropy available.
21:42.54[g2]on openslug entrophy is 0
21:43.59jp30pTweety, hmm, that's a worry, I was under the impression it did by default. well, better not refer to a dir that doesn't necessarily exist. does no harm to not have that TMPDIR set, just makes the build use more VM
21:45.22pTweetyso it compiled without the TMPDIR set for you? You just freed some mem with shutting down apache?
21:45.44jp30pTweety, yes, that's right, i didn't mess with TMPDIR myself
21:46.03jacquesI thought there was a variable just for gcc
21:47.11jacqueshmm tho I cannot find it
21:47.11jp30gcc only pays attention to TMPDIR as far as I can tell (from man gcc). the trouble is that we can't count on /opt/tmp to exist
21:48.12pTweetyrwhitby-away, what is running on the slug building the packages? Can you free some more mem on it?
21:48.20jp30could always mkdir $(BUILD_DIR)/tmp, and set TMPDIR to that, but it's a lot of hackery just to speed up a build
21:48.50jp30pTweety, are you sure rwhitby had memory problems? I think his build died in a different place from mine
21:48.55jacquesnot that bad - could just be in the makefile
21:48.57jacquestoplevel
21:49.42pTweetyhm, he wrote something about perl-stuff and your broke at char*.c, right?
21:49.52jp30that's right
21:51.35pTweetyyou have perl installed?
21:51.46rwhitby-awaypTweety: the build errors on gastro were not mem problems
21:52.25pTweetygastro is the official build slug?
21:53.05rwhitby-awayyep
21:53.06jp30yes, i have perl installed and staged. i noticed that my build used the installed perl.
21:53.51jp30... and used it by its full path. (i.e. /share.../perl)
21:54.38pTweetyrwhitby-away, is perl 5.8.6  installed on gastro?
21:55.50rwhitby-awaypTweety: the packages on gastro are listed on the native compilation wiki page
21:58.11jp30rwhitby-away, i notice that pkgconfig and libtool aren't listed on the wiki page - should they be added?
21:59.05rwhitby-awayif they are installed, yes.  but gastro is down so I can't check
21:59.08rwhitby-awayback later
21:59.20pTweetyok, found perl. But now I'm lost. I need more info about the error occurred on gastro
22:00.02pTweetyjp30, can you try make postfix on sour slug as well?
22:00.07rwhitby-awayka6sox or dyoung will be able to tell you the error from the make.log on gastro unslung packages when it comes up later
22:00.20rwhitby-awaypostfix problem was in ipk target, not build (it built fine)
22:00.28rwhitby-awayciao
22:00.32pTweetybye
22:00.42jp30pTweety, I'll try postfix
22:01.54pTweetythanks
22:02.53jp30this might take longer... i offered to do cyrus-imapd because i already had its dependencies staged.
22:04.00*** join/#nslu2-linux mrblack51 (mrblack51@wa-pullman-cuda1k-244.losaca.adelphia.net)
22:05.33jp30as I suspected, now my poor slug is embarking on pcre. i'll leave it running, but may be a while before i have news for you
22:05.44pTweetyjp30, I've to go to bed now but will read the irc-log. Thanks for your testing (since I know how anoying such a native build can be ;-)
22:06.00jp30no probs
22:06.01pTweetynight
22:06.04jp30bfn
22:06.16jacquesthe time involved is the most annoying aspect for me
22:06.36jacquestakes so long to test things to see if they're fixed etc
22:07.29jp30good incentive to make things cross-compile if at all possible...
22:07.44jacquesyep :-)
22:44.50ka6sox-awaydyoung: ping
22:51.59garpincrwhitby-away u around?
23:01.53garpincHe's never around these days when I am... Must be avoiding me :-)
23:02.24*** join/#nslu2-linux NSLU2-IRC (~NSLU2-IRC@82-40-186-72.cable.ubr04.newm.blueyonder.co.uk)
23:07.26*** join/#nslu2-linux rwhitby-web (www-data@netblock-66-159-209-60.dslextreme.com)
23:08.55rwhitby-webgarpinc: you rang?
23:09.26[g2]I hear he's never around :)
23:09.34garpincyes I did....
23:10.35garpincrwhitby-web I was wondering if there are any issues with the latest stuff I checked in... i.e: transcode and the latest version of libAres...e..
23:10.54rwhitby-webis it in READY_FOR_TESTING?
23:12.19garpincWell libAres was a change to the make file to check out from CVS instead.
23:12.54garpinctranscode and the others haven't been put in READY_FOR_TESTING. I'll do that when I get home..
23:13.34*** join/#nslu2-linux FireEgl (Ariel@Atlantica.US.TO)
23:22.56*** join/#nslu2-linux kergoth_ (~kergoth@proxyle02.ext.ti.com)
23:22.58kergoth_re
23:23.06ka6sox-awaywb
23:23.22kergoth_forgot to join.. my irc proxy at home isnt accessible
23:23.32kergoth_(damn wrt handed my desktop a new ip, so all my forwards are dead)
23:23.59ByronTstatic ip is your friend
23:24.03kergoth_heh, indeed
23:24.10kergoth_hadnt gotten around to doing that
23:24.15kergoth_this is definate motivation to do so
23:24.38ByronTit's about the first thing I do when I work on a new client's network
23:24.57ByronTonly dhcp addressee's will be wireless ones
23:25.45[g2]kergoth_, thx for the help earlier -- the alias worked and that's what's checked in
23:26.02kergoth_great
23:26.09kergoth_up to 1,333 unique packages in oe
23:26.16[g2]awesone
23:26.19kergoth_i just checked in target path obeying fixes for over 215 packages
23:26.28kergoth_heh
23:26.31[g2]awesome
23:26.42kergoth_still quite a few to do though. uclibc for example obeys prefix and exec_prefix
23:26.45kergoth_but thats all
23:26.51kergoth_assumes it knows where everything goes from there
23:28.10[g2]kergoth_, In the next week or two I'd like to setup an OE package status for OpenSlug on a web page
23:28.24kergoth_there are already wiki pages for that, arent there?
23:29.10[g2]There's a packages page, but I'm more focus on OE builds status
23:29.22kergoth_what compiles and what doesnt?
23:29.25*** join/#nslu2-linux garpinc (~garpinc@pool-151-203-198-44.bos.east.verizon.net)
23:29.29kergoth_thats _really_ _really_ hard to maintain
23:29.31kergoth_trust me
23:29.35kergoth_people have tried before
23:29.45[g2]compiles is the first part, ipk install and runs
23:29.58kergoth_thatd be useful
23:30.03kergoth_i add lots of things to oe and never know if they work
23:30.08[g2]well that's the whole point of having the community updating it
23:30.25[g2]right and we need ot close the loop on that stuff
23:34.02caplink811-awaygnite all
23:34.09[g2]nite
23:34.22ByronTlater all
23:46.07*** join/#nslu2-linux rwhitby-nslu2 (~rwhitby@rwhitby.nslu2-linux)
23:46.10[g2]kergoth_, what do you think of adding some stats production to the "non-broken" packages in OE
23:46.29kergoth_what do you mean?
23:46.37[g2]we could keep a tiny sqlite3 db in the BB dir
23:47.05[g2]and inject records straight from bb as it complete the compiles/patches etc...
23:47.10kergoth_[g2]: just write a new eventhandler.
23:47.32kergoth_bb already supports writing custom event handlers to do things when tasks and apckages and builds complete or fail
23:47.42[g2]right and take the arch/distro/etc.. and roll up the compile build status
23:48.15kergoth_write a new bbclass or .inc to do it
23:49.03[g2]Ok I may do that for OpenSlug and it could probably be used for any distro
23:55.33*** join/#nslu2-linux copperbeech-log (~copperbee@82-40-186-72.cable.ubr04.newm.blueyonder.co.uk)

Generated by irclog2html.pl by Jeff Waugh - find it at freshmeat.net! Modified by Tim Riker to work with blootbot logs, split per channel, etc.