**** BEGIN LOGGING AT Wed Aug 10 23:59:56 2005 Aug 11 00:23:21 is it really necessary to compile EVERY locale of glic Aug 11 00:23:25 glibc Aug 11 00:24:59 no it's not, but it's not a technical issue Aug 11 00:25:13 it's more of a policy issue on oe's part Aug 11 00:26:34 it's a policy issue that seems to take about 9 hours ;) Aug 11 00:27:16 only 9hrs? you must have a faster machine than me! Aug 11 00:29:05 well, it hasn't finished, but it has been running overnight Aug 11 00:30:03 hmm, is this a native compile? Aug 11 00:38:35 must be Aug 11 00:52:51 ka6sox - no, a cross-compile Aug 11 00:54:38 blaster8, what's your hardware? glibc shouldn't take more than about an hour Aug 11 00:55:10 maybe something's wrong Aug 11 03:41:53 jp30-work:ping Aug 11 03:42:27 jp30: windoer if you could answer a question. I installed php-apache a couple of weeks ago and obviously it installed php as part of that package. Aug 11 03:42:46 the libphp5.so file is isntalled by this in /opt/libexec Aug 11 03:43:32 I am trying to build a package against the libphp5.so and when I stage php-stage it builds libphp5.so in staging/opt/lib ratehr than staging/opt/libexec Aug 11 03:44:05 which is the correct location? and how shoudl I distribute the package when php has already been installed by php-apache? Aug 11 03:44:40 what's the situation here? Aug 11 04:27:49 can anyone with admin access to the CVS repository give me a hand for 5 mins. Aug 11 04:58:08 copperbeech: what seems to be the problem Aug 11 04:58:51 well I have some files I want to add to the repository... having read cvs manuals for some hours I tried to add some new files, but it said I didn't have write access. Aug 11 04:59:03 rwhitby said he put me on the write access list yesterday. Aug 11 04:59:22 how was it checked ot? Aug 11 04:59:24 out? Aug 11 04:59:49 also the cvs status is a little opaque (may be because I don't know what to expect and how to read it) and I need someone to check my work and make sure I don't overwite stuff I souldnt etc. Aug 11 05:00:20 cd nslu Aug 11 05:00:22 cvs -z3 -d:pserver:anonymous@cvs.sourceforge.net:/cvsroot/nslu co unslung Aug 11 05:00:55 I have some files in a new folder called unslung/sources/appweb/2.0.3 which I was trying to add and commit. Aug 11 05:02:16 you need to check them out with ssh. Aug 11 05:02:23 mcneillm@copperian:~/nslu2/unslung/sources/appweb/2.0.3$ cvs -z3 -d:pserver:mat Aug 11 05:02:24 the pserver type does not allow writes Aug 11 05:02:25 _mcneill@cvs.sourceforge.net:/cvsroot/nslu add * Aug 11 05:02:27 cvs add: cannot add special file `CVS'; skipping Aug 11 05:02:28 cvs [server aborted]: "add" requires write access to the repository Aug 11 05:02:50 ok, ok - going to need to do key stuff etc? Aug 11 05:03:08 I do have a putty key for sourceforge already Aug 11 05:03:24 are you using the master makefile? Aug 11 05:04:19 well I probably need to do an update to that, but I was giong to ask someone else to assist with it. Firstly I just want to update an existing makefile. Aug 11 05:04:25 (for a package) Aug 11 05:04:51 Later I want to add a new package (which I guess will affect the master make file) Aug 11 05:04:54 um... Okay since youre using putty, I think I'll assume the answer is no Aug 11 05:05:19 well all using putty means is that I have to transfer the key to my linux box Aug 11 05:05:34 are you developing on your linux box or your windows box? Aug 11 05:06:30 I am developing on a debian box Aug 11 05:06:48 Okay, lets work with that for now then, Aug 11 05:06:49 ... but I already have some sourceforge projects for windows under my ID Aug 11 05:07:00 the master makefile is for building all nsul2-linux related projects. Aug 11 05:07:01 which is why my key is in putty Aug 11 05:07:07 yep Aug 11 05:07:09 if you dont know what it is, youre not using it Aug 11 05:07:23 I do know what it is and I think I do need to edit it to add a new package. Aug 11 05:07:31 No you dont. Aug 11 05:07:32 but that's stage #2 Aug 11 05:07:46 in fact you dont have write access to the master makefile. Aug 11 05:08:07 you do however have write access to the unslung/optware top level makefile. Aug 11 05:08:13 and that you WILL need to add to. Aug 11 05:08:32 ok... I see the difference now. Aug 11 05:08:50 So am I right in assuming you are not yet using the master makefile? Aug 11 05:09:10 yes - I don't need access beyone the common make file for adding a new package... Aug 11 05:09:23 Okay. now that thats straight... :-) Aug 11 05:10:15 :-) Aug 11 05:10:31 the command you want is cvs -d :ext:copperbeach@cvs.sf.net:/cvsroot/nslu co unslung Aug 11 05:10:52 ok - I need to save out all the files I have changed then.... Aug 11 05:11:06 you may have to set a envirnment variable CVS_RSH to ssh. Aug 11 05:11:09 then do all the rebuilding of the toolchains again etc? Aug 11 05:11:22 if you have time on your hands. Aug 11 05:11:34 if not, the mv unslung unslung.pserver Aug 11 05:12:03 the re-check it out using your dev key. Aug 11 05:12:37 and in theory ( I havnt actually done this yet ), you should be able to cp/mv your unslung.pserver/toolchain to unslung. Aug 11 05:12:44 and avoid rebuilding it. Aug 11 05:12:49 yep Aug 11 05:13:13 ok ... lets try this and see how we go... Aug 11 05:13:18 move build* and packages* over too, and nothing will rebuild :-) Aug 11 05:13:42 we seem to be a pretty lazy bunch. Aug 11 05:13:44 oh, and staging* too Aug 11 05:13:45 :-) Aug 11 05:14:07 the best programmers are lazy :-) Aug 11 05:14:14 or just use the master makefile and run "make setup-optware-developer" Aug 11 05:14:20 once thats all squared away, you should be able to retrieve your edited/chaged/added files to the new place and cvs add them. Aug 11 05:14:58 rwhitby-away: I was about to suggest that too if copperbeach was ready for using the master makefile. Aug 11 05:15:10 what does that do :-? Aug 11 05:16:24 set you up the same as everyone else, so we can answer questions much simpler in the future :-) Aug 11 05:16:52 if you do that, you'll need to set a environment variable CVS_USER to mcneill if your login isnt that. Aug 11 05:17:11 if you read the master makefile its pretty apparent what it does. Aug 11 05:17:56 ok... is that unslung/Makefile ? Aug 11 05:18:39 http://www.nslu2-linux.org/wiki/Development/MasterMakefile Aug 11 05:18:44 copperbeech: check "NSLU2-Linux Development" on the wiki sidebar Aug 11 05:19:46 I see - a meta-build system Aug 11 05:19:55 jbot, nslu2-master-makefile is http://www.nslu2-linux.org/wiki/Development/MasterMakefile Aug 11 05:19:56 okay, dyoung Aug 11 05:20:00 03rwhitby 07org.nslu2-linux.bitbake * rfbb1b9d9... 10/ (4 files in 3 dirs): Updated to SVN revision 308 Aug 11 05:20:13 I think it's actually a meta-meta-meta-build system Aug 11 05:20:17 ~nslu2-master-makefile Aug 11 05:20:17 methinks nslu2-master-makefile is http://www.nslu2-linux.org/wiki/Development/MasterMakefile Aug 11 05:22:45 dyoung: should we use linestack_file instead of linestack_memory in ctrlproxy? Aug 11 05:23:15 I'm remembering that one or more of the linkstack modules wasnt there. Aug 11 05:23:21 but yes, its probably a good idea. Aug 11 05:23:41 I was still in testing mode; seeing what goes funny... Aug 11 05:23:55 so you're now a willing test subject. :-) Aug 11 05:24:14 ok: now rebuilding the packages for testing in the new tree. Aug 11 05:24:50 copperbeech: so you're using the master makefile now? Aug 11 05:25:24 no - I think dyoung suggested the "cvs co unslung" so I'm just building the packages. Aug 11 05:26:05 make Aug 11 05:26:17 :wq Aug 11 05:26:20 oops Aug 11 05:26:38 dyoung: ctrlproxy -D -r /etc/ctrlproxy/ctrlproxy.conf Aug 11 05:26:53 rwhitby: that last commit of yours... that is bitbake? Aug 11 05:26:53 * rwhitby waits for rwhitby-proxy to arrive ... Aug 11 05:26:55 what -D -r do ? Aug 11 05:27:03 NAiL: yep - updating it to latest Aug 11 05:27:17 ok :) Aug 11 05:27:28 make import-bitbake ; cd bitbake ; mt commit ; cd .. ; make push-bitbake Aug 11 05:27:41 dyoung: -D is daemon, -r is conf file. Aug 11 05:27:52 Let's use /etc/ctrlproxy, and put the keys in there too Aug 11 05:27:52 dyoung: packages rebuild and packaged fine... Aug 11 05:28:09 what I need to do is the folloeing 3 things: Aug 11 05:28:24 1) update unslung/make/appweb.mk Aug 11 05:28:40 2) add unslung/make/appweb-php.mk Aug 11 05:28:58 Woo! Aug 11 05:28:59 3) add a folder with some files in to unslung/sources/appweb Aug 11 05:29:10 I didnt know about daemon mode. I been running it piped through a doo-dad. Aug 11 05:29:18 I guess need to do #3 first Aug 11 05:31:53 dyoung: linestack_file.so don't exist. any idea why? Aug 11 05:36:19 ok am about to execute the following to do #3 - does it look sane? Aug 11 05:36:21 cd ~/nslu2/unslung/sources/appweb Aug 11 05:36:22 cvs -d:ext:matt_mcneill@cvs.sf.net:/cvsroot/nslu add -r -m 'new files for appweb 2.0.x' 2.0.3/ Aug 11 05:36:24 cd 2.0.3/ Aug 11 05:36:26 cvs -d:ext:matt_mcneill@cvs.sf.net:/cvsroot/nslu commit * Aug 11 05:36:52 copperbeech: why create a new subdir? Aug 11 05:37:22 and you don't need the -d option in any directory which has ./CVS in it. Aug 11 05:37:37 I dunno why yet. Aug 11 05:37:43 I was still looking into it. Aug 11 05:38:03 I may have simply un-configured it out. Aug 11 05:38:06 * dyoung dunnos Aug 11 05:38:26 copperbeech: in general we haven't had that extra level of subdir under sources. Do you absolutely need it? Aug 11 05:38:45 Can those files just go under sources/appweb directly? Aug 11 05:38:57 rwhitby: no, I was just being paranoid... I can remove it Aug 11 05:42:15 rebuilding and retesting packages... Aug 11 05:45:27 ok, this is through ctrlproxy now .. Aug 11 05:48:08 03matt_mcneill * 10unslung/sources/appweb/ (7 files): updated files for appweb 2.0.3 Aug 11 05:48:55 can I just do cvs commit appweb.mk to do a single file ? Aug 11 05:49:03 yep Aug 11 05:50:38 03matt_mcneill * 10unslung/make/appweb.mk: updated appweb to 2.0.3 Aug 11 05:51:29 treo through ctrlproxy Aug 11 05:51:43 * NAiL looks at the commits... who is matt_mcneill? Aug 11 05:52:01 it's me Aug 11 05:52:06 03matt_mcneill * 10unslung/make/appweb-php.mk: added new file for appweb 2.0.3 with php Aug 11 05:52:14 oh :) Aug 11 05:52:35 ok - now what's the protocol for adding appweb-php to the unslung packages list? Aug 11 05:52:49 in the unslung makefile Aug 11 05:53:16 NAiL: Should there be an option in the master makefile for an unslung stable target Aug 11 05:53:17 put it in the testing section Aug 11 05:53:47 blaster8: No idea. I don't do unslung Aug 11 05:53:52 anyone who does? Aug 11 05:54:25 since the kernel version and glibc version never changes, there's not a lot of point in not building the latest. Aug 11 05:54:28 blaster8: No. Its all fixed to a specific version of gcc/glibc and based on fixed linksys sources. so the latest version IS the stable version Aug 11 05:54:29 should I put the new version of appweb in TESTING too? Aug 11 05:55:27 what do you think dyoung? Aug 11 05:55:36 thanks - someone was having issues yesterday and I couldn't sensibly answer that question Aug 11 05:56:37 rwhitby: btw who is responsible for promoting packages from TESTING to PACKAGES in the unslung Makefile? Aug 11 05:56:59 the optware/nslu2 package manager, jp30. Aug 11 05:57:25 the protocol is to add new stuff to testing. Aug 11 05:57:45 ok - what about significant updates to existing packages ? Aug 11 05:58:07 I would consider that "new" . Aug 11 05:58:13 rwhitby: how is ctrlproxy different from miau ? Aug 11 05:58:17 dyoung: thanks Aug 11 06:01:13 "cvs commit: failed to create lock directory for `/cvsroot/nslu/unslung'" what does that mean? Aug 11 06:01:34 trying to "cvs commit Makefile" Aug 11 06:03:27 dunno Aug 11 06:03:51 were you able to commit other files? Aug 11 06:03:59 it worked this time... Aug 11 06:04:11 rwhitby: how is ctrlproxy different from miau ? Aug 11 06:04:23 dunno yet :-) Aug 11 06:04:44 the big thing is the backlog command, but I'm still working out whether that will work for me the way I want. Aug 11 06:05:41 (I want it for the treo, so when I get a phone call, I can get back on IRC and get the last 5 minutes of backlog) Aug 11 06:06:14 I do that with the miau quicklog se to a couple of hours... Aug 11 06:06:35 when I connect back to maiu it just spools it all up as it connects to each of the channels I have set Aug 11 06:07:52 yeah, but on a 12 line screen with less than 50 lines of scrollback, and GPRS charges at $5 per Mb, I need something more controllable Aug 11 06:08:27 you can set it to five mins if you want - but what you're saying is that you need to be able to command it from the prompt? Aug 11 06:08:36 yrp Aug 11 06:08:57 I have it set for an hour for home, but on the phone I need a controllable period Aug 11 06:09:10 yep Aug 11 06:14:35 dyoung: even though ctrlproxy connects to different servers, you have to connect to *it* on different ports to get to them. I was hoping it would somehow collate them onto a single port ... Aug 11 06:19:05 copperbeech: have a look at http://www.nslu2-linux.org/autobuild-nudi-last.txt - appweb has some issues Aug 11 06:24:12 03rwhitby * 10unslung/Makefile: Demoted madplay again - still breaks the autobuild Aug 11 06:28:09 copperbeech: any reason why it's "appweb-php" and not "php-appweb" (like php-apache and php-thttpd) ? Aug 11 06:32:45 copperbeech: all those ../../../... in appweb-php.mk need to be changed to ${STAGING_INCLUDE_DIR} Aug 11 06:34:55 and when you rename it to php-appweb, you will need to rename APPWEB-PHP to PHP_APPWEB too inside the file (as well as appweb-php renamed to php-appweb inside the file too). Aug 11 06:46:57 yay, nslu2 board still seems to work after giving it a couple of standoffs to sit inside its new case Aug 11 06:47:05 always a good thing Aug 11 07:01:53 out of interest, does anyone know any reason why 2.5" HDs shouldn't be run upside down Aug 11 07:32:30 blaster8: the 1s and 0s fall off the platter if you run them upside down .... down here in Australia we actually have to run all our disks upside down cause we're in the southern hemisphere. Aug 11 07:57:20 llagendijk, ping? Aug 11 08:03:26 rwhitby: http://www.hitachigst.com/hdd/research/recording_head/pr/PerpendicularAnimation.html Aug 11 08:10:10 03nail 07org.openembedded.nslu2-linux * r0b40ae52... 10/packages/nail/ (files/nail.spec.diff nail_11.21.bb): Update nail to use /usr/sbin/sendmail instead of /usr/lib/sendmail Aug 11 08:25:31 Cool. These guys are using the Slug for their UAV!! Aug 11 08:25:34 http://www.auvsi.org/competitions/2005/JournalPapers/Uottawa.pdf Aug 11 08:26:09 I remember giving them a demo of the slug at our local robotics club meeting. I guess they really liked it! Aug 11 08:28:00 VoodooZ_Work: wonder if they've removed R83 ? Aug 11 08:28:44 jp30-work pong Aug 11 08:31:03 VoodooZ_Work: I added that link to http://www.nslu2-linux.org/wiki/Main/Articles - perhaps you want to create a "Robots" section on that page ... ? Aug 11 08:32:14 actually, I was wondering the same thing! Aug 11 08:32:21 I'll email them to warn them. Aug 11 08:32:36 I guess it's a bit late though as the competition was yesterday and they finished 15th. Aug 11 08:33:23 Yeah, if I can get my new USB I/O (to replace the slow s/w I2C) finished I'll post something. Aug 11 08:37:41 jp30-work pong? Aug 11 08:40:07 03nail 07org.openembedded.nslu2-linux * rbf63a8b3... 10/packages/postfix/postfix_2.0.20.bb: Update postfix to use u-a for /usr/sbin/sendmail (Conflict with ssmtp) Aug 11 08:47:37 hi, llagendijk Aug 11 08:48:14 llagendijk, am i right in thinking that you are the maintainer of bind in optware? Aug 11 08:52:05 rwhitby, I made some changes to that link you created. I pointed it directly to their page instead of the UAV association. Please approve site. Aug 11 08:52:37 jp30: ping Aug 11 08:52:41 llagendijk, if so, can you register yourself with slugbug? someone reported a bug in bind, and i would like to assign it to you Aug 11 08:52:46 hi, copperbeech Aug 11 08:53:29 jp30: I have uploaded some .mk files for appweb to bring it up to the latest version, but I'd like to go over the appweb-php.mk file to get your view on it. Aug 11 08:53:44 It was to do with installing the libphp5.so library etc. Aug 11 08:55:08 ok, looking at it now Aug 11 08:56:05 I got a bit confused with the php dependencies. When I built php-stage I get libphp5.so being built in staging/opt/lib Aug 11 08:56:35 but when I installed php-apache recently it installed libphp5.so in /opt/libexec Aug 11 08:57:05 so php-apache also depends on php... so why are the libraries in different places? Aug 11 08:57:49 yes. the libphp5.so built by php-apache is an apache module; the libphp5.so built by php is the php embed SAPI, a free-standing library. Aug 11 08:58:49 ok... so why didn't the ipkg install of appweb install php (including the libphp5.so) that is the stand alone library? Aug 11 08:59:14 at the moment the appweb build script is having to copy the libphp5.so into /opt/lib/ Aug 11 08:59:53 Thumbs up on the web site side bar reorg. Encourages people to help themselves with the faq and search. Aug 11 09:00:45 Speaking of the FAQ and in particular item 30. Are there any firm plans to resolve the issue with upgrading openssh via an ssh session? Aug 11 09:00:56 the php embed SAPI is in the "php-embed" ipk. appweb should depend on that Aug 11 09:01:06 ahhh -ok great Aug 11 09:01:38 out of interest - why are you replacing configure wholesale rather than patching it in appweb-php.mk? Aug 11 09:01:43 also - I was wondering what to use to test the php build... do you have any nasty scripts which test each of the libraries? Aug 11 09:02:01 never thought of that - what would you suggest? Aug 11 09:02:51 how do you normally do patching of make files? It sounds a bit tricky to maintain... ? Aug 11 09:03:43 jp30: I also wondered if you could cast your eye over the php configuration options (particularly the libraries) Aug 11 09:03:45 --with-php5-libs="php5 crypt db z" Aug 11 09:03:55 to test php? how about doing a from-scratch install of mediawiki and phpmyadmin and playing with them. that'll give you a good sense of whether it's fast enough too Aug 11 09:04:40 I'ts not the speed - I think it will be ok, it's trying to test the compatability and correct linkage with php ... but I'll try a phpmyadmin and see what happens. Aug 11 09:05:19 jp-30:btw - I'm registered as copperbeech on slugbug... Aug 11 09:05:20 about patching: just use diff -u to generate a diff between the configure that appWeb came with, and the one you are replacing it with. the only thing you need to avoid is patching generated files (i.e. never patch an autoconf-generated configure) Aug 11 09:06:08 ok - but that would mean I would have to maintain two .mk files anyway just to generate the diff - what's the advantage in that? Aug 11 09:06:31 i think we must be speaking past each other... Aug 11 09:06:57 probably over my head :-) Aug 11 09:07:01 there's a file, configure in sources/appweb, and your .mk file copies it into the appweb build dir at configuration time... Aug 11 09:07:53 i'm suggesting you might generate a patch containing the diffs between appweb's original configure file and the one you are copying in Aug 11 09:08:11 oh yes... that is because the appweb source has had the configure file updated and I need to replace it... I could have patched I suppose, but originally I was going to try and wget it... but my proxy settings dont work through work's firewall. Aug 11 09:08:37 it wasn't a patch so much as an update from the appweb guys. Aug 11 09:08:53 ah, i see. Aug 11 09:09:03 03repvik * r173 10/releases/OpenSlug-2.5-beta/openembedded/packages/ (9 files in 7 dirs): Update postfix, ssmtp, nail, util-linux and coreutils Aug 11 09:09:09 a little bit messy - I know. Aug 11 09:09:55 jp30: do you think the php build flags (into configure) look ok? Aug 11 09:09:56 i think you have probably done the right thing there. Aug 11 09:11:24 you know better than me what the semantics of appweb's configure options are, i think :) it depends what exactly that --with-php5-libs option does Aug 11 09:12:03 do you need to list all the dependencies of libphp5.so? Aug 11 09:12:13 :-) *grins* it was a bit of a guess by me - I guess it is the libraries php needs to build... Aug 11 09:12:26 jp30-work, ok, will do. I am in the process of upgrading my hw, but will look into asap Aug 11 09:12:42 llagendijk, thanks. no big hurry Aug 11 09:12:44 in the --with-php5-iflags directive? Aug 11 09:13:25 copperbeech, if you need to list all the dependencies of libphp5.so, you can find them all out by using objdump -p staging/opt/lib/libphp5.so Aug 11 09:13:39 ok... I was just going from http://www.appwebserver.org/products/appWeb/doc/common/source/packages.html#php Aug 11 09:14:05 ...(and there are more than you listed). however, i'd try using --with-php5-libs="php5" and seeing if it works Aug 11 09:14:32 the ELF library dependency mechanism should take care of the rest Aug 11 09:17:22 ok - am now updating my build... Aug 11 09:26:04 hmm even though I put the depends php-embed it didn't install it when I installed the ipkg (from file) Aug 11 09:26:47 jp30-work: I just have to put depends as "APPWEB-PHP_DEPENDS=openssl php-embed" ? Aug 11 09:26:49 DEPENDS is comma seperated Aug 11 09:26:54 ahh Aug 11 09:27:12 try that... Aug 11 09:27:19 building... Aug 11 09:30:03 ok - working now :-) Aug 11 09:30:18 ... well it's downloaded php-embed anyway :-) Aug 11 09:32:27 03jp30 * 10unslung/make/ncurses.mk: ncurses: use ln -sf to avoid error when restaging Aug 11 09:33:48 03matt_mcneill * 10unslung/make/appweb.mk: small fix for make file Aug 11 09:34:34 03matt_mcneill * 10unslung/make/appweb-php.mk: small fixes for php libs Aug 11 09:40:29 03matt_mcneill * 10unslung/make/appweb-php.mk: small fixes for php libs Aug 11 09:41:01 03matt_mcneill * 10unslung/make/appweb.mk: removing blank line tabs for emacs Aug 11 09:47:48 03matt_mcneill * 10unslung/make/appweb-php.mk: commented out source code rule duplicated in appweb.mk Aug 11 10:05:05 03jbowler 07org.openembedded.nslu2-linux * rb41e1e0c... 10/packages/meta/ucslugc-packages.bb: Aug 11 10:05:06 ucslugc, UCSLUGC_PACKAGES: make various prerequesite packages explicit in Aug 11 10:05:06 the build list, therefore if the dependent package is removed the Aug 11 10:05:06 prerequisite packages are still built. Aug 11 10:15:47 03matt_mcneill * 10unslung/make/appweb-php.mk: stripped the libraries Aug 11 10:16:24 03matt_mcneill * 10unslung/make/appweb.mk: stripped the libraries Aug 11 12:10:36 jbot, test Aug 11 12:10:37 Test Failed! Aug 11 12:32:08 hmm Aug 11 12:32:09 jbot, test Aug 11 12:32:10 Test Passed! Aug 11 12:32:14 phew Aug 11 12:36:37 jbot, test Aug 11 12:36:38 Test Failed! Aug 11 12:37:19 d`oh Aug 11 12:38:21 jbot, test Aug 11 12:38:21 Test Passed! Aug 11 12:38:24 ha Aug 11 12:38:29 hours of fun ;) Aug 11 12:38:31 ~botsnack Aug 11 12:38:31 caplink811_log: aw, gee Aug 11 12:39:00 fontenot, your turn ;-) Aug 11 12:39:26 too bad tim removed roulette Aug 11 12:39:49 you'd say jbot roulette Aug 11 12:39:56 and it's say "click" Aug 11 12:40:04 or "bang" and then kick you from the channel Aug 11 12:40:19 haha Aug 11 12:40:20 :) Aug 11 12:45:07 jbot, roulette Aug 11 12:45:07 * jbot watches fontenot pull the trigger: Click! Aug 11 12:45:11 jbot, roulette Aug 11 12:45:51 jbot, roulette Aug 11 12:45:51 * jbot watches blaster8 pull the trigger: Click! Aug 11 12:46:12 feeling lucky Aug 11 12:46:13 jbot, roulette Aug 11 12:46:13 * jbot watches blaster8 pull the trigger: Click! Aug 11 12:47:12 jbot, roulette Aug 11 12:47:12 * jbot watches caplink811 pull the trigger: Click! Aug 11 12:47:22 puh Aug 11 12:48:00 jbot, roulette Aug 11 12:48:00 * jbot watches caplink811 pull the trigger: Click! Aug 11 12:50:28 no risk, no fun.. Aug 11 12:50:30 jbot, roulette Aug 11 12:50:30 * jbot watches caplink811 pull the trigger: Click! Aug 11 12:50:51 ~botsnack Aug 11 12:50:52 caplink811: aw, gee Aug 11 12:52:24 I think that's just a stock answer now - not the original game Aug 11 12:54:18 the original game took over entire channels Aug 11 13:05:08 03jbowler 07org.openembedded.nslu2-linux * r073419be... 10/packages/ (14 files in 5 dirs): (log message trimmed) Aug 11 13:05:08 This set of changes ensures that TARGET_CC_ARCH is passed reliably to all Aug 11 13:05:08 packages in a build. The change fixes problems in the following packages: Aug 11 13:05:08 binutils{,-cross}_2.15.94.0.1 Aug 11 13:05:08 HOST_CC_ARCH was not being passed to HOST_CC. The problem does Aug 11 13:05:09 not seem to be present in binutils* 2.16, however this version Aug 11 13:05:11 is still used for uClibc builds. Aug 11 13:10:49 jp30-work, ping? Aug 11 13:27:19 llagendijk, pong Aug 11 13:27:59 I have created a Slugbug account, feel free to allocate the bind bug to me. Thanks Aug 11 13:28:22 jp30-work, accountname = llagendijk Aug 11 13:29:06 llagendijk, ok. i think you may wish to mark the bug "won't fix" once you see it; it's about building optware on a MacOS build system Aug 11 13:29:52 jp30-work, ok, I will look into it Aug 11 13:30:39 03jbowler 07org.openembedded.nslu2-linux * rde3c1699... 10/packages/libmad/libmad_0.15.0b.bb: Aug 11 13:30:39 libmad: thumb only (so no PR change): remove the erroneous use of 32 bit Aug 11 13:30:39 ARM assembler when compiling thumb Aug 11 13:30:42 03jbowler 07org.openembedded.nslu2-linux * rccb298ba... 10/packages/libvorbis/libvorbis_1.0.1.bb: Aug 11 13:30:42 libvorbis: thumb only (no PR change): prevent all optimisation to hide an Aug 11 13:30:42 apparent compiler bug (register spilling) when compiling thumb code. Aug 11 13:30:44 03jbowler 07org.openembedded.nslu2-linux * r99bf5683... 10/packages/db/db_4.3.27.bb: Aug 11 13:30:45 db-4: thumb only (no PR bump): prevent the erroneous use of 32 bit arm Aug 11 13:30:47 assembly when compiling thumb instructions. Aug 11 13:30:49 03jbowler 07org.openembedded.nslu2-linux * rf53fd480... 10/ (135 files in 63 dirs): Aug 11 13:30:51 propagate from branch 'org.openembedded.dev' (head af7da4406babf078f2747a3588cfa451ae711315) Aug 11 13:30:53 to branch 'org.openembedded.nslu2-linux' (head 99bf568313454a05bf17df79cb135d7e4bb6c643) Aug 11 13:37:08 llagendijk. i can see that you have registered with slugbug, but you don't appear in the pulldown "assign to" menu! how odd. it is bug #233 - can you visit it and assign it to yourself? Aug 11 13:43:23 jp30-work, ok, I let's see if that works Aug 11 13:49:11 jp30-work, Josh, I can not assign it myself either.... Aug 11 13:49:44 jp30-work: He's not a "developer", so he cannot be assigned bugs Aug 11 13:49:55 in slugbug, that is Aug 11 13:55:51 03jbowler 07org.openembedded.nslu2-linux * rc7dbf3bd... 10/packages/musicpd/mpd_0.11.2.bb: Aug 11 13:55:51 mpd was building, but not necessarily using, it's own local cache of the Aug 11 13:55:51 libmad and libid3tag source. Removing the --enable-mpd-* settings causes Aug 11 13:55:51 it not to build the local source (and it uses the versions in staging). Aug 11 13:56:01 03jbowler 07org.openembedded.nslu2-linux * r823a1ea1... 10/packages/ (musicpd/mpd_0.11.2.bb ppp/ppp_2.4.3.bb): Aug 11 13:56:01 propagate from branch 'org.openembedded.dev' (head 2733583ee4724b13e2625abab589a558da043628) Aug 11 13:56:01 to branch 'org.openembedded.nslu2-linux' (head c7dbf3bdc213b3c3f611e1fb56a743c593138db5) Aug 11 13:56:08 03jbowler 07org.openembedded.nslu2-linux * r2860daf5... 10/ (4 files in 4 dirs): Aug 11 13:56:08 Change the NSLU2 kernel build configuration method to allow the Aug 11 13:56:08 configuration override to be specified in the machine.conf rather than Aug 11 13:56:08 hardwiring -mno-thumb-interwork into nslu2-kernel.inc Aug 11 13:56:09 03jbowler 07org.openembedded.nslu2-linux * ra02108b1... 10/conf/distro/ucslugc.conf: Aug 11 13:56:09 Set -mthumb-interwork on ucslugc builds to allow building (and maybe Aug 11 13:56:11 someday running) of thumb exectuables. Aug 11 13:56:37 I see someone is trying to overtake me in commits ;-) Aug 11 13:59:29 NAiL, does being in the "Developer" group suffice for being a developer? Aug 11 13:59:38 ...slugbugwise that is Aug 11 14:00:08 jp30-work: I have no idea. Aug 11 14:00:27 I know ingeba-away can fix it. Dunno who else can ... Aug 11 14:00:47 hi - i just installed cups from unslung feed on openslug - works fine - is this worth a seperate wiki-page? Aug 11 14:01:31 jp30-work: Yes Aug 11 14:02:11 Joschi: I'd say so. Aug 11 14:02:19 hm, where should i put it? under howtos? or under the openslug section? Aug 11 14:02:30 preferrable, we'd get cups to compile directly in openslug :) Aug 11 14:03:07 I didn't have this idea before... ;) Aug 11 14:03:42 ? Aug 11 14:03:56 btw - when compiling openslug2.5, i got an "internal compiler error" when gcc compiles itself... Aug 11 14:04:09 Joschi: reproducable? Aug 11 14:04:13 yes Aug 11 14:04:28 Does anybody now why appweb.mk gives errors in the unslung package cvs? Aug 11 14:04:40 maybe i should mention i'm compiling in 64bit machine Aug 11 14:04:48 ok, what distro? Aug 11 14:04:53 fc4 Aug 11 14:05:47 Is the distro 64bit as well, or are you running a 32bit distro on a 64bit machine? Aug 11 14:05:56 64bit-distro Aug 11 14:06:25 otherwise it wouldn't matter i think... Aug 11 14:06:26 I think DaKa2 did a successful compile on some 64bit distro Aug 11 14:06:38 no, it shouldn't matter otherwise Aug 11 14:06:50 And I've got ubuntu64 installed on my box Aug 11 14:06:54 i just threw out gcc - then everythink worked fine Aug 11 14:06:58 So I guess I can boot and try a build Aug 11 14:06:59 thanks ingeba-away. can you look into why llagendijk doesn't appear on the list of assignable developers? Aug 11 14:07:21 another little problem was with the ppp package Aug 11 14:07:23 Joschi: File a slugbug. Compile gcc, and attach the log to the bug Aug 11 14:07:28 Yes -- tomorrow - I'm leaving in 5. Will send myself an email about it. Aug 11 14:07:48 ok, i'll do so Aug 11 14:07:50 ingeba-away, thanks Aug 11 14:08:30 ppp-issue: i had to add an index to lib....radius...a manually Aug 11 14:10:06 jp30-work: Fixed with dev. status for llagendijk Aug 11 14:15:41 can i delete an account from slugbug?? Aug 11 14:15:50 i accidently registered twice... Aug 11 14:15:50 Test passed! Aug 11 14:16:10 oops Aug 11 14:17:21 dyoung! morning! Aug 11 14:19:40 ok, bug added Aug 11 14:36:22 NAiL: how do I restart the autobuild (for ucslugc) from scratch? (I changed uclibc, incremental builds will not work). Aug 11 14:38:02 03marceln * 10unslung/ (sources/dump/control make/dump.mk Makefile): Added bzip2 and zlib staging so that build works Aug 11 14:41:19