**** BEGIN LOGGING AT Wed Jul 21 02:59:58 2010 Jul 21 04:44:43 xMff, ping Jul 21 04:44:53 look at http://nuwiki.openwrt.org/doc/howto/ead Jul 21 04:45:06 do u see how the info box is overlapping the toc? Jul 21 06:05:50 hello Jul 21 06:31:04 ola Jul 21 06:59:21 acoul, do you see this a possibility for openwrt? Jul 21 06:59:21 {{page>meta:infobox:dangerous&noheader&nofooter&noeditbtn}} Jul 21 06:59:25 oops damn it! Jul 21 06:59:42 acoul, https://dev.openwrt.org/ticket/7630 Jul 21 07:04:18 interesting Jul 21 07:04:52 indeed :/ Jul 21 07:06:17 * realopty is willing to see if a patch will work for the device Jul 21 07:06:24 it has boot_wait :) Jul 21 07:06:34 hmm Jul 21 09:13:33 xMff, ping Jul 21 10:48:11 build #60 of ps3 is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/ps3/builds/60 Jul 21 11:20:59 kaloz * r22323 /trunk/ (22 files in 9 dirs): add support for the Gateworks Laguna family (Cavium Networks Econa CNS3xxx) Jul 21 14:13:31 build #62 of rdc is complete: Failure [failed compile_7] Build details are at http://tksite.gotdns.org:8010/builders/rdc/builds/62 Jul 21 14:34:19 nbd * r22324 /trunk/target/linux/generic/ (7 files): kernel: add some missing debugging related config symbols Jul 21 14:34:26 nbd * r22325 /trunk/Config.in: add a config option for enabling kernel debugging symbols Jul 21 14:38:31 nbd: I guess the "waveloeten" in the c-base is the regular meeting you mentioned? Jul 21 14:40:29 yes Jul 21 14:43:52 how's the smoking situation there? Jul 21 14:44:15 outside Jul 21 14:44:33 great Jul 21 14:44:34 at least for the main room Jul 21 14:45:06 most of the time Jul 21 14:45:36 but even when they smoke inside, it's not that much Jul 21 14:52:47 build #57 of uml is complete: Failure [failed compile_10] Build details are at http://tksite.gotdns.org:8010/builders/uml/builds/57 Jul 21 14:58:25 <[florian]> nbd: can you fix the missing include for ioremap in crashlog? Jul 21 15:00:15 ok Jul 21 15:00:19 which target did it fail on? Jul 21 15:00:34 i built one target per kernel version and did not see a compile error Jul 21 15:01:01 <[florian]> nbd: uml, maybe others Jul 21 15:01:26 i'd be surprised if uml even has ioremap Jul 21 15:01:37 i think we need to disable crashlog there anyway Jul 21 15:01:38 <[florian]> I guess it has an empty prototype Jul 21 15:01:38 it won't work Jul 21 15:01:45 <[florian]> certainly not Jul 21 15:02:55 nbd * r22326 /trunk/target/linux/uml/config/ (i386 x86_64): disable crash logging for uml, it cannot possibly work there anyway Jul 21 15:03:22 <[florian]> thanks! Jul 21 15:04:23 btw. i do have the right include in place Jul 21 15:04:28 ioremap is typically defined in asm/io.h Jul 21 15:04:48 <[florian]> should not you use linux/io.h instead? Jul 21 15:05:31 not sure Jul 21 15:05:56 <[florian]> anyway, uml should certainly provide an empty ioremap stub Jul 21 15:18:25 nbd * r22327 /trunk/ (Config.in include/kernel-build.mk): add a build system option for collecting all kernel debug information (including modules) in a tarball Jul 21 15:46:44 obsy * r22328 /packages/net/transmission/Makefile: [packages] transmission: update to 2.03 Jul 21 16:06:21 nbd * r22329 /branches/backfire/target/linux/generic-2.6/ (config-2.6.30 config-2.6.32): add missing kernel config symbols Jul 21 16:06:22 nbd * r22330 /branches/backfire/include/kernel-defaults.mk: merge the kernel config cleanup that was added to trunk as part of r20837 Jul 21 16:06:24 nbd * r22331 /branches/backfire/Config.in: merge the kernel debug option from r22325 Jul 21 16:06:28 nbd * r22332 /branches/backfire/ (Config.in include/kernel-build.mk): merge the kernel debug collect option from r22327 Jul 21 16:06:31 nbd * r22333 /branches/backfire/target/linux/ (8 files in 4 dirs): merge the crashlog feature from r22305, r22326 Jul 21 16:50:58 i try to build a package that needs access to kernel include path. is there a variable like $(KERNEL_INCLUDE) or something like that? Jul 21 16:53:41 $(LINUX_DIR)/include Jul 21 16:54:00 your package needs to include kernel.mk though Jul 21 16:56:22 build #60 of ramips is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/ramips/builds/60 Jul 21 17:02:23 hello all Jul 21 17:07:30 i'm trying to modify nvram.c from kernel but i do not really understand which file i have to modify Jul 21 17:08:18 the one in build_dir or in target/linux/brcm47xx/files-2.6.32/ Jul 21 17:08:27 ? Jul 21 17:08:45 thanks nbd i'' give it a try Jul 21 17:11:41 @clem-vangelis: do you want to patch your kernel? Jul 21 17:12:02 yes Jul 21 17:12:24 clem-vangelis: when the kernel is unpacked, the files from target/linux/brcm47xx/files-* are copied to build_dir/... Jul 21 17:12:50 clem-vangelis: so if you only change the build_dir, then your changes are lost after a make clean Jul 21 17:12:57 ok but there is no arch/mips/brcm47xx/nvram.c for kernel 2.6.34 Jul 21 17:13:10 you need http://wiki.openwrt.org/doc/devel/patches this Jul 21 17:13:10 that was what i believed Jul 21 17:13:13 it's probably part of the patch series Jul 21 17:13:25 001-backport Jul 21 17:13:32 so i have to patch the patch ? Jul 21 17:13:40 why not Jul 21 17:13:48 your build_dir/linux-*/linux-*/ tree has a quilt patch series Jul 21 17:13:56 you can edit patches there or add new ones Jul 21 17:14:00 that change nvram.c Jul 21 17:14:34 ok thanks Jul 21 17:30:17 build #86 of ar71xx is complete: Failure [failed compile_7] Build details are at http://tksite.gotdns.org:8010/builders/ar71xx/builds/86 Jul 21 17:31:20 nbd * r22334 /trunk/tools/squashfs4/patches/140-mode_check.patch: squashfs4: backport an upstream change to fix the file mode check to allow setuid/setgid binaries (thx, ermo) - fixes #7653 Jul 21 17:32:06 nbd * r22335 /branches/backfire/tools/squashfs4/patches/140-mode_check.patch: squashfs4: backport the mode fix from r22334 Jul 21 17:35:39 nbd: is #7100 still wanted now that crashlog is available? Jul 21 17:36:15 i don't think we need that then Jul 21 17:37:05 closed the ticket Jul 21 17:37:23 thanks Jul 21 17:41:57 sure, only ~700 to go :) Jul 21 17:49:55 nbd * r22336 /trunk/scripts/env: scripts/env: fix initialization of files/ when creating a new environment (patch from #7603) Jul 21 17:50:21 nbd * r22337 /branches/backfire/scripts/env: backport scripts/env init fix from r22336 Jul 21 17:57:00 <[Fate]> nbd: ping Jul 21 17:57:17 pong Jul 21 17:58:36 <[Fate]> did you happen to have the time to apply #7528 minus the fifo vals? Jul 21 17:58:50 not yet Jul 21 17:58:55 i'll get to that soon Jul 21 17:59:00 <[Fate]> ok thank you Jul 21 17:59:17 i was mostly busy with my crash logging / debug stuff Jul 21 17:59:19 ;) Jul 21 18:00:17 <[Fate]> no pun intended :) Jul 21 18:00:29 <[Fate]> I'd just like to get rid of large diffs to trunk :) Jul 21 18:00:37 sure Jul 21 18:26:09 jow * r22338 /trunk/package/opkg/patches/012-strip-trailing-conffiles-whitespace.patch: [package] opkg: strip trailing whitespaces from entries in conffiles control files, fixes conffile md5 calculation against ipkg-build generated archives Jul 21 18:28:29 build #60 of ar7 is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/ar7/builds/60 Jul 21 18:28:32 jow * r22339 /branches/backfire/package/opkg/patches/012-strip-trailing-conffiles-whitespace.patch: [backfire] merge r22338 Jul 21 21:14:06 ping nbd, xMff Jul 21 21:34:17 pong cshore Jul 21 21:34:57 xMff: sorry, figured it out, just wondering how to load the lua uci binding (require("uci"), doh!) Jul 21 21:41:19 xMff: wait! but where are the lua bindings on the filesystem? Jul 21 21:41:45 /usr/lib/lua/ Jul 21 21:42:01 (d'oh) :P Jul 21 21:42:38 hmmm...don't have that dir....firing up menuconfig to see which package I need Jul 21 21:43:11 libuci-lua most likeley Jul 21 21:43:22 found it Jul 21 21:46:07 I planning on using lua so that I can use uci directly for FS instead of shell scripting something to convert uci to xml static files. It means the fs config doesn't require restarting fs, most of the time, and is cleaner and easier. Jul 21 21:46:27 good idea Jul 21 21:46:44 but hasn't fs an own embedded lua interpreter anyway? Jul 21 21:46:58 xMff: that's what I'll be using Jul 21 21:47:07 xMff: but I need the uci bindings Jul 21 22:02:28 jow * r22340 /trunk/package/opkg/patches/013-fix-darwin-compile.patch: [package] opkg: fix host compilation on Darwin (libopkg_test failed to link) Jul 21 22:03:19 cshore: true Jul 21 22:13:39 xMff: where is documentation for uci functions (as opposed to cli)? I can't seem to find it, though I know I've seen it before. Jul 21 22:13:57 uhm, in the code ^^ Jul 21 22:14:14 oh, ok, heh Jul 21 22:14:59 the api is quite simple Jul 21 22:15:18 top level entry point is uci.cursor() Jul 21 22:15:35 that instantiates a uci context instance Jul 21 22:16:12 e.g. x = uci.cursor() or x = uci.cursor(nil, "/var/state") if you want to involve state vars Jul 21 22:16:23 on that you can call the usual operations Jul 21 22:16:40 x:get("config", "sectionname", "option") -> returns string or nil for not found Jul 21 22:17:00 x:set("config", "sectionname", "option", "value") -> sets simple string value Jul 21 22:17:15 x:set("config", "sectionname", "option", { "foo", "bar" }) -> sets list value Jul 21 22:17:33 x:delete("config", "section", "option") -> deletes option Jul 21 22:17:40 x:delete("config", "section") -> deletes section Jul 21 22:18:26 x:add("config", "type") -> adds new anon section "type" and returns its name Jul 21 22:19:11 x:set("config", "name", "type") -> adds new section "name" with type "type" Jul 21 22:19:43 x:foreach("config", "type", function(s) ... end) -> iterates over all sections of type "type" and invokes callback function for each Jul 21 22:20:18 "s" within the callback function is a table containing all options and two special properties s['.type'] -> section type and s['.name'] -> section name Jul 21 22:21:01 thats basically all you need Jul 21 22:21:19 ok, I'm going to put that on the wiki Jul 21 22:21:36 see also http://luci.subsignal.org/trac/browser/luci/trunk/libs/uci/luasrc/model/uci.lua Jul 21 22:22:07 thats what LuCI uses. It "patches" the uci context class at runtime and adds a few more convenience functions Jul 21 22:22:23 s/patches/extends/ Jul 21 22:23:12 official api is inline documented beginning at line 231 Jul 21 22:23:50 in the uci source ? Jul 21 22:23:58 in the lua file I linked above Jul 21 22:59:51 <_trine> there are big problems with the either-net speed on my wndr3700 KAMIKAZE (bleeding edge, r22338) Jul 21 23:00:06 <_trine> its very very slow Jul 21 23:18:54 build #62 of au1000 is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/au1000/builds/62 Jul 21 23:33:07 nbd, nice to see that scripts/env is fixed Jul 21 23:35:27 nbd, but i still get errors lol Jul 21 23:35:27 mkdir: cannot create directory `env': File exists Jul 21 23:35:36 * realopty sigh Jul 22 00:10:49 swalker * r22341 /packages/libs/apr/Makefile: [packages] apr: update to 1.4.2 Jul 22 00:18:37 swalker * r22342 /packages/libs/c-ares/Makefile: [packages] c-ares: update to 1.7.3 Jul 22 02:38:25 swalker * r22343 /packages/net/vpzone/Makefile: [packages] vpzone: add missing kmod-tun dependency Jul 22 02:43:46 jow * r22344 /trunk/tools/mkimage/patches/100-portability.patch: [tools] mkimage: fix compilation on FreeBSD Jul 22 02:59:16 jow * r22345 /trunk/tools/yaffs2/patches/100-compile.patch: [tools] yaffs2: fix compilation on FreeBSD **** ENDING LOGGING AT Thu Jul 22 02:59:56 2010