**** BEGIN LOGGING AT Tue May 08 02:59:57 2007
May 08 06:02:43 <[florian]> thepeople_work: let me see
May 08 06:10:44 <[florian]> thepeople_work: have you make clean; make ?
May 08 06:10:54 <[florian]> nls packages are generated here
May 08 14:59:19 <CIA-13> florian * r7134 /packages/libs/glib2/ (. Makefile): Add glib2 (#1650)
May 08 15:02:53 <crazy_imp> [florian]: thanks :)
May 08 15:12:20 <[florian]> crazy_imp: you are welcome
May 08 15:56:00 <thepeople_work> [florian]: http://pastebin.ca/477307
May 08 15:57:25 <[florian]> thepeople_work: thanks
May 08 15:59:10 <thepeople_work> nls built I had to do make clean
May 08 15:59:21 <[florian]> thepeople_work: ok
May 08 16:00:47 <[florian]> thepeople_work: multipath is enabeled in generic-2.6 kernel config, and not explicitely disabled in x86-2.6
May 08 16:03:01 <thepeople_work> [florian]: ok, should it be?
May 08 16:03:33 <[florian]> thepeople_work: ok, I see the problem
May 08 16:04:13 <CIA-13> florian * r7135 /trunk/package/kernel/modules/network.mk: Typo, again
May 08 16:05:47 <CIA-13> florian * r7136 /trunk/target/linux/ (3 files in 3 dirs): Build multipath caching modules
May 08 18:40:02 <CIA-13> nbd * r7137 /trunk/package/spca5xx/Makefile: update spca5xx to latest upstream release - old version no longer available
May 08 19:09:24 <CIA-13> florian * r7138 /packages/libs/ (bzip2/Makefile clinkc/Makefile): Fix warnings and should also fix compilation (#1593)
May 08 19:20:18 <CIA-13> florian * r7139 /trunk/package/shfs/Makefile: Use shfs fixes from #1530, unmark as broken
May 08 19:20:29 <crazy_imp> hm, if the sources for a package are not anymore in dl/ the package gets rebuild (?)
May 08 19:21:02 <nbd> yes
May 08 19:21:09 <nbd> there's a file dependency on the file in dl/
May 08 19:22:16 <crazy_imp> nbd: hm, but it's annoying if you removed them to save some space and the stuff gets rebuild
May 08 19:26:36 <crazy_imp> it would make more sense imho if it checks first for the .built in the build-dirs of the packages
May 08 19:27:26 <nbd> if i remove that dependency it messes up some of the build code
May 08 19:27:33 <nbd> i don't want to do that
May 08 19:28:16 <crazy_imp> ok
May 08 19:29:21 <[mbm]> dep tree should be something like: ipkg > build_*/package > package/package > dl/package.tar.gz
May 08 19:29:32 <[mbm]> haven't looked to see what it actually is with all the recent changes
May 08 19:29:59 <nbd> it's somewhat like that
May 08 19:31:11 <crazy_imp> [mbm]: checking first for the ipkg could cause trouble if theres also a  Build/InstallDev section in the makefile but nothing in the staging-dir
May 08 19:33:39 <nbd> it doesn't check for the ipkg first
May 08 19:33:53 <nbd> it's dynamic enough to handle packages without actual ipkgs and just Build/InstallDev
May 08 19:35:09 <crazy_imp> nbd: so if a package has both (ipkg & Build/InstallDev) it won't fail if there's for some reason the ipgk but nothing got installed in the staging-dir?
May 08 19:36:12 <nbd> the compile target depends on InstallDev if present
May 08 19:36:15 <nbd> and on the ipkgs
May 08 19:36:20 <nbd> if one of them fails, make complains
May 08 19:39:56 <CIA-13> florian * r7140 /trunk/package/ar7-atm/patches/100-compile_fix.patch: Fix objects linking order (#1643)
May 08 20:23:05 <CIA-13> nbd * r7141 /trunk/package/fuse/patches/111-uclibc_changes.patch: remove problematic patch from fuse (was only needed with devfs on 2.6)
May 08 20:29:07 <[mbm]> should bump busybox to 1.4.2
May 08 20:30:37 <[mbm]> I'd say 1.5.0 but now's not the time to do that
May 08 20:33:19 <[florian]> seems too much work
May 08 20:34:00 <[mbm]> ?
May 08 20:34:43 <[florian]> bumping to 1.5.0
May 08 20:35:01 <[mbm]> that's why I said not to use 1.5.0
May 08 20:35:36 <[mbm]> we should do 1.4.2 though since there were a few bugfixes
May 08 21:45:04 <CIA-13> florian * r7142 /trunk/package/shfs/patches/ (13 files): Forgot comitting patches
May 08 21:48:19 <CIA-13> nbd * r7143 /trunk/package/nozomi/Makefile: fix nozomi unpack command
**** ENDING LOGGING AT Wed May 09 02:59:57 2007