**** BEGIN LOGGING AT Wed Feb 23 02:59:56 2011 Feb 23 03:06:18 8PMJ ... 8-pin modular jack Feb 23 03:06:27 what is this for... Feb 23 03:08:40 the internet! Feb 23 03:09:25 yea, but ive many windows opened, need to keep focused on pinouts :p Feb 23 03:09:54 the pinout for the internet? Feb 23 03:10:37 i'm rerouting my wnr854t Feb 23 03:10:37 2 = tip, 3 = ring Feb 23 03:11:09 anyway these pins are not connected Feb 23 03:11:24 first 8 mini pci pins Feb 23 03:13:04 nbd: It looks like the problems of gettext and libiconv got resolved. The compilation is still going so far. Feb 23 03:13:22 ndb: Thanks. Feb 23 03:13:28 nbd: Thanks. Feb 23 06:20:29 gmorning Feb 23 06:31:10 hi Feb 23 06:42:18 acoul * r25665 /packages/net/etherpuppet/ (. Makefile): net/etherpuppet: add etherpuppe, a voodoo doll for an Ethernet interface (closes #8913) Feb 23 06:42:56 hmm Feb 23 06:52:08 build #70 of ixp4xx is complete: Failure [failed shell compile_6] Build details are at http://tksite.gotdns.org:8010/builders/ixp4xx/builds/70 Feb 23 08:28:05 zandbelt * r25666 /packages/net/asterisk-1.8.x/Makefile: [packages] asterisk-1.8.x: add func_uri Feb 23 13:23:00 build #89 of brcm47xx is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/brcm47xx/builds/89 Feb 23 13:42:51 build #78 of ps3 is complete: Failure [failed compile_3] Build details are at http://tksite.gotdns.org:8010/builders/ps3/builds/78 Feb 23 17:42:16 ping xMff Feb 23 17:45:28 zandbelt * r25667 /packages/net/radsecproxy/Makefile: [packages] radsecproxy: update to 1.4.2 Feb 23 18:01:09 pong phaidros Feb 23 18:04:21 build #64 of mpc52xx is complete: Failure [failed compile_4] Build details are at http://tksite.gotdns.org:8010/builders/mpc52xx/builds/64 Feb 23 18:24:32 jow_laptop: xmpp Feb 23 18:29:25 phaidros: sorry, not xmpp available Feb 23 18:29:36 phaidros: query Feb 23 18:41:41 so, anyone having a clue why the latest build is still borked for ar7 with an 88e6060 switch? Feb 23 18:47:24 switch to the .37 kernel thats where the patches have been applied Feb 23 18:48:39 dunno, i use the auto-built ones :| Feb 23 18:49:01 (building from scratch is a no-no) Feb 23 18:49:20 the defaults havn't been changed yet Feb 23 18:50:24 ouch Feb 23 18:51:07 so you more or less say i'll have to setup an environment to build all that from scratch for kernel 2.6.37 Feb 23 18:51:31 great :| Feb 23 18:55:36 linux installs are quick and hassle free :) just about Feb 23 18:59:53 i have one Feb 23 19:00:05 the prerequisities are the problem Feb 23 19:08:51 i should have looked at the packages to figure out what kernal was used... before bothering Feb 23 20:14:41 jow * r25668 /trunk/package/kernel/modules/virtual.mk: [package] kernel: fix kmod-xen-evtchn for Linux 2.6.37+ Feb 23 21:44:46 jow * r25669 /trunk/package/kernel/modules/virtual.mk: [package] kernel: make xen pcidev driver depend on Linux 2.6.37 or 2.6.38 Feb 23 21:53:29 jow * r25670 /trunk/package/kernel/modules/virtual.mk: [package] kernel: fix path to pcidev driver Feb 23 21:59:02 hi Feb 23 21:59:04 ping theMIROn` Feb 23 21:59:48 nbd: hi i have a new package to vommit - but can you gimme access on svn to be the maintainer of it? Feb 23 21:59:58 vommit=commit Feb 23 22:00:53 I cant seem to compile libmysql for x86/generic/pcengines alix, r25635, http://openwrt.pastebin.com/iSXYqWje Feb 23 22:02:43 this looks like an inconsistency to me Feb 23 22:02:53 old libuclibc++ or libstdcpp lying around? Feb 23 22:03:27 wait Feb 23 22:03:34 libcrap is using host gcc for linking Feb 23 22:09:42 netprince_: looks like you're being confronted with libtool's infinite wisdom here, unfortunately I have not yet succeeded in fixing it up within mysql Feb 23 22:16:36 its a bit unclear to me why it uses your host gcc for a target build... this cannot work **** BEGIN LOGGING AT Wed Feb 23 22:23:30 2011 Feb 23 22:41:26 xMff: sorry, just got back, had to check on kid Feb 23 22:41:28 it seems to have built fine for brcm47xx and ar71xx Feb 23 22:41:28 seems it gets confused by building for x86 on x86 Feb 23 22:41:30 it could pick a wrong gcc at some point Feb 23 22:41:30 ahh, interesting, I would be confused too :) Feb 23 22:41:32 just kidding of course Feb 23 22:41:33 netprince_: first of all check the config.log, see what CC it determined Feb 23 22:41:34 netprince_: then inspect the generated Makefiles, see what CC and LD is there Feb 23 22:41:35 netprince_: finally grep the generated libtool for gcc Feb 23 22:41:46 florian * r25672 /packages/utils/vim/ (Makefile patches/001-cross_compile.patch): [package] vim: update to 7.3, patch from koniu Feb 23 22:41:47 florian * r25673 /packages/lang/php5/Makefile: (log message trimmed) Feb 23 22:41:47 [package] php5: first steps to simplify the Makefile Feb 23 22:41:47 - Introduce /etc/php5 which is scanned for additional configuration files. Feb 23 22:41:47 So it is possible that a installed modules is automatically enabled Feb 23 22:41:47 without the need to modify the php.ini. Feb 23 22:41:47 - Remove unsupported config option: --disable-ttf Feb 23 22:41:47 - Add support for SimpleXML module, closes #7758. Feb 23 22:41:47 florian * r25674 /packages/lang/php5/Makefile: Feb 23 22:41:47 [package] php5: add missing dependency on libsqlite3/libpthread (#7224) Feb 23 22:41:47 The binaries are already dependent on sqlite3 and pthread, not only the Feb 23 22:41:47 corresponding php modules. Feb 23 22:41:48 This closes #7724. Feb 23 22:41:48 Signed-off-by: Michael Heimpold Feb 23 22:41:56 florian * r25675 /trunk/target/linux/ar7/Makefile: [ar7] switch to 2.6.37.1 Feb 23 22:45:21 alright, checking out Feb 23 22:45:43 wanna bet i'll end up working for hours on this :| Feb 23 22:57:15 jow * r25676 /packages/net/openswan/ (4 files in 2 dirs): [packages] openswan: update to v2.6.33 (#8924) Feb 23 22:59:58 jow * r25677 /trunk/scripts/ipkg-make-index.sh: (log message trimmed) Feb 23 22:59:58 [PATCH 1/2] Never write errors to stdout from ipkg-make-index.sh Feb 23 22:59:58 Since the output of ipkg-make-index.sh is unconditionally used to create Feb 23 22:59:58 the Packages file, sending the errors into this file is unacceptable Feb 23 22:59:58 (both because they are not shown to users and because they make an Feb 23 22:59:58 invalid Packages file). Feb 23 22:59:59 For incorrect invocation, write to stderr. Feb 23 23:00:55 jow * r25678 /trunk/ (package/Makefile target/imagebuilder/files/Makefile): (log message trimmed) Feb 23 23:00:56 [PATCH 2/2] Send ipkg-make-index.sh output to stdout Feb 23 23:00:56 The "Generating index for package" created by ipkg-make-index only Feb 23 23:00:56 clutter stderr and serve to push previous warnings/errors off-screen. Feb 23 23:00:56 Instead, send to stdout. Feb 23 23:00:56 Note: This would send incorrect invocation messages to stdout as well. Feb 23 23:00:56 If this is a concern, perhaps we could send them to FD 3, then send 3 to Feb 23 23:07:06 jow * r25679 /packages/multimedia/crtmpserver/ (7 files in 2 dirs): [packages] add crtmpserver - a high performance C++ RTMP streaming server Feb 23 23:42:24 hmm Feb 23 23:42:40 u sure hope i didn't forget anything Feb 23 23:42:46 *u=i Feb 23 23:45:51 mb * r25680 /trunk/target/linux/generic/patches-2.6.38/ (2 files): 2.6.38: Remove patches that are upstream in -rc6 Feb 23 23:48:32 oh crap Feb 23 23:48:41 i think i forgot to include the webui xD Feb 23 23:48:55 and upnp Feb 23 23:50:11 btw Feb 23 23:50:24 are the patches applied automatically on build? Feb 23 23:50:52 hm? Feb 23 23:51:22 when building Feb 23 23:51:30 which patches? Feb 23 23:51:41 platform specific ones? Feb 23 23:51:42 :p Feb 23 23:52:00 kernel? Feb 23 23:52:10 yeah Feb 23 23:52:25 yes they are applied on top of the generic patches Feb 23 23:52:41 what about generic patches? Feb 23 23:52:42 :p Feb 23 23:53:13 well, they are applied to the kernel and after that the platform patches are applied Feb 23 23:54:53 ahh, so i don't have to patch manually huh Feb 23 23:54:55 good Feb 23 23:55:11 you can use quilt to manage the patches Feb 23 23:56:17 ehh Feb 23 23:56:30 include/exclude them manually you mean? Feb 23 23:56:47 or add new too? Feb 23 23:56:59 yeah Feb 23 23:57:34 the plan is to just build targeting a specific platform, using a specific kernel Feb 23 23:58:13 will the image be auto-patched in that case? :p Feb 23 23:59:06 and if the patches are for a specific kernel, will they work for updated versions of that kernel? :p Feb 23 23:59:11 not sure what you're trying to accomplish Feb 23 23:59:34 like 2.6.32 patches for 2.6.32-4 for example Feb 24 00:00:09 depends on the patch, of course Feb 24 00:00:17 ahh Feb 24 00:04:49 oh, something more Feb 24 00:05:04 if i want to include an updated package Feb 24 00:05:48 the only thing necessary is to point at the commit i want (from its makefile)? Feb 24 00:11:59 y/n Feb 24 00:12:00 ? Feb 24 00:14:20 you could just update the makefile Feb 24 00:14:35 svn -r 12345 up feeds/packages/foo/bar/Makefile Feb 24 00:18:06 ehh? Feb 24 00:18:16 jow * r25681 /branches/backfire/include/autotools.mk: [backfire] revert r25653 for now, until the full trunk fixups get merged Feb 24 00:18:36 but there is no place to update from? :p Feb 24 00:18:37 that's what I thought when reading your question Feb 24 00:19:08 yes, you edit the makefile, update the version number and md5sum Feb 24 00:19:14 in an ideal world that would be all Feb 24 00:19:23 in practise a lot of stuff can happen Feb 24 00:19:30 yeah i know Feb 24 00:19:31 like patches not applying anymore or random breakage Feb 24 00:20:01 well you are oversaying things a bit Feb 24 00:20:02 :p Feb 24 00:20:19 worse case scenario i end up with an error Feb 24 00:20:26 or a list of them Feb 24 00:21:01 or it compiles successfully but linsk against to a wrong lib leading to crashes on the target Feb 24 00:21:10 like swconfig bombing out when libuci is updated Feb 24 00:21:25 or some program linking both openssl 1.0.0 or 0.9.8 Feb 24 00:21:53 now you remind me why i don't use linux on my desktop Feb 24 00:22:10 nah Feb 24 00:22:19 a typicall binary distro takes care of that things Feb 24 00:22:34 but compiling a distro from scratch is something different Feb 24 00:22:41 until you realise it misses a feature you need Feb 24 00:22:49 that you work with every day Feb 24 00:22:54 well Feb 24 00:23:08 and when trying to update, system hangs at random times, x crashes, etc Feb 24 00:23:11 on osx and or windows I do not even have the chance to recompile :P Feb 24 00:23:23 if something is missing, its missing Feb 24 00:23:33 you miss the point here Feb 24 00:23:39 until you find a replacement or a shareware/freeware supplement Feb 24 00:23:39 it's not missing Feb 24 00:23:40 :p Feb 24 00:24:02 you just use another thing that does the same joob, or you update Feb 24 00:24:10 -o Feb 24 00:24:50 but hey, this is free s/w we are talking about Feb 24 00:25:13 that can work the way you want, from top to bottom Feb 24 00:25:20 haha Feb 24 00:25:24 of course it has some drawbacks Feb 24 00:25:39 s/can/could/ ... if it compiles :O Feb 24 00:25:40 the only problem is that you have to work on it Feb 24 00:25:42 .oO(Oh, I get it now...) Feb 24 00:26:04 and i mean, fix it yourself Feb 24 00:26:48 that's why whenever i see a linux setup that works for me, i don't touch Feb 24 00:26:56 cause it might as well break down Feb 24 00:27:37 That reminds me of Windows. Feb 24 00:28:00 nah, with windows you can update the system fine Feb 24 00:28:13 without it breaking Feb 24 00:28:19 the problem though is... Feb 24 00:28:56 when it breaks it does not do it cause you updated it Feb 24 00:29:24 but because you did something it just didn't like Feb 24 00:29:57 so in short Feb 24 00:30:01 I think this becomes a little bit offtopic Feb 24 00:30:20 windows breaks by design, linux by trying to "fix" it Feb 24 00:30:36 so yeah, let's move along Feb 24 00:30:47 nbd * r25682 /trunk/target/linux/generic/ (config-2.6.35 config-2.6.36 config-2.6.37 config-2.6.38): kernel config: refresh using kconfig.pl Feb 24 00:30:51 nbd * r25683 /trunk/target/linux/generic/ (5 files): kernel config: remove the obsolete pcomp unlzma symbol Feb 24 00:30:53 I don't think anybody here will agree with you anyway ;) Feb 24 00:30:58 nbd * r25684 /trunk/target/linux/generic/ (8 files): kernel config: disable the pcomp api + zlib decompressor by default - it has not been used for a while now Feb 24 00:31:02 nbd * r25685 /trunk/package/qos-scripts/files/etc/config/qos: qos-scripts: disable overhead calculation by default, it does not work properly for many line speeds Feb 24 00:31:09 nbd * r25686 /trunk/package/kernel/modules/crypto.mk: kmod-crypto: remove pcomp, nothing uses it Feb 24 00:31:22 heh Feb 24 00:31:47 as you said, we don't live in an ideal world Feb 24 00:31:51 :) Feb 24 00:32:19 http://mobile.slashdot.org/story/11/02/23/0026210/Windows-Phone-7-Update-Jams-Some-Phones Feb 24 00:32:48 i've never seen a windows machine that still worked properly after being updated to a new version ;) Feb 24 00:37:50 jow * r25687 /packages/net/l2tpv3tun/patches/100-nl_handle_alloc.patch: [packages] l2tpv3tun: fix whitespace mangled patch Feb 24 00:38:36 heh Feb 24 00:42:23 build #15 of lantiq is complete: Failure [failed shell compile_6] Build details are at http://tksite.gotdns.org:8010/builders/lantiq/builds/15 Feb 24 00:47:49 pft XP SP3 did wonders Feb 24 00:47:58 anything newer isnt an upgrade at all ;) Feb 24 00:49:33 To me, Windows is a constant pain in the ass. I experience that every day at work as a normal user. Feb 24 00:49:56 And there I don't even remotely touch such things as updates or whatever. Feb 24 00:52:25 firstly i'll say i prefer linux for most things. but you can have a hell of a time with setup of linux. afterwards tho, once ya know everything, it gets easier and certainly more stable :) Feb 24 00:53:17 sorry reading scrollback it was already decided to ditch this topic Feb 24 00:53:21 * neutronscott goes back to his hole. Feb 24 00:55:53 ;) Feb 24 00:57:53 in other offtopic news christchurch will still get us out of antarctica. leave monday instead of friday tho. Feb 24 00:58:04 * neutronscott does happy dance Feb 24 01:12:02 nbd * r25688 /trunk/package/kernel/modules/crypto.mk: kmod-crypto: fix selecting core crypto modules Feb 24 01:12:06 nbd * r25689 /trunk/target/linux/generic/ (6 files in 4 dirs): kernel: the patch to disable the crypto tests is no longer necessary since 2.6.36, remove it Feb 24 01:12:16 nbd * r25690 /trunk/package/kernel/modules/other.mk: add kmod-oprofile to package/kernel Feb 24 01:14:02 nbd * r25691 /branches/backfire/package/kernel/modules/other.mk: add kmod-oprofile to package/kernel (backport of r25690) Feb 24 01:14:39 nbd * r25692 /packages/devel/oprofile/Makefile: oprofile: remove the kernel module package Feb 24 01:34:25 build #69 of au1000 is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/au1000/builds/69 Feb 24 01:58:45 nbd * r25693 /branches/backfire/package/mac80211/ (23 files in 2 dirs): mac80211: backport latest version from trunk Feb 24 02:16:22 nbd: anything worth in the mac80211 update to backfire? Feb 24 02:17:11 mostly minor stability fixes Feb 24 02:19:52 excellent! Feb 24 02:23:13 nbd, ive seen there is a beginning of branch for ubicom devices... target/linux/ubicom32 (BOARDNAME:=Ubicom IPxx) Feb 24 02:23:33 you you know where i can find infos about that ? **** ENDING LOGGING AT Thu Feb 24 02:59:57 2011