**** BEGIN LOGGING AT Sun Feb 01 02:59:59 2015
Feb 01 08:23:09 <OpenWrt-Commits> nbd r44234 trunk/target/linux/ramips/files/drivers/net/ethernet/ralink/ralink_soc_eth.c * ramips: fix ethernet driver tx completion polling
Feb 01 09:38:22 <OpenWrt-Commits> wigyori r44235 trunk/package/kernel/linux/modules/usb.mk * modules: fix usb-gadget build on 3.18
Feb 01 10:48:00 <_trine> can anyone explain why I get this error http://paste.debian.net/143329/
Feb 01 10:48:41 <plntyk> hauke having a talk about the backports project (wireless drivers used in openwrt) soon  at http://live.fosdem.org/watch.php?room=UB2.252A%20%28Lameere%29
Feb 01 10:51:21 <plntyk> _trine: well you have "--with-ncurses " as CONFIGURE_ARG for util-linux but the configure script from util-linux does not find ncurses
Feb 01 10:51:25 <plntyk> "checking for NCURSESW... no checking for NCURSES... no"
Feb 01 10:51:44 <_trine> yes but I thought I had them installed
Feb 01 10:52:17 <plntyk> yes - libncursesw might be installed on your side if I see your log correctly
Feb 01 10:52:40 <plntyk> there is some error with util-linux - i did investigate it a short while on my end
Feb 01 10:52:47 <_trine> I have installed libncursesw5.dev and libcurses5.dev
Feb 01 10:53:19 <_trine> so does it need fixing in openwrt
Feb 01 10:53:32 <plntyk> yes - the configure script from util-linux looks first for pkgconfig files - atm openwrt does not build these
Feb 01 10:53:46 <_trine> ah thanks for explaining
Feb 01 10:53:48 <plntyk> this is the fail when "NCURSES/NCURSESW ...no"
Feb 01 10:54:00 <plntyk> then the configure script looks at "normal" header location
Feb 01 10:54:08 <plntyk> for the .h files but this fails too
Feb 01 10:55:21 <_trine> has it been like this for long?  could I revert my source a bit to get away from this error
Feb 01 10:58:37 <plntyk> sorry dont knoiw - it might be a build system issue (I tried to build util-linux on arch linux and this failed) - notice that the buildbot has no failed packages (afaik these are debian systems) - but I didnt have time yet to check with my debian VM
Feb 01 10:59:13 <_trine> I use ubuntu
Feb 01 10:59:47 <_trine> and this is a new CO,, maybe its only new CO which are affected
Feb 01 11:00:41 <_trine> do the buildbaot always use new CO's
Feb 01 11:00:48 <_trine> do the buildbots always use new CO's
Feb 01 11:02:50 <jow_laptop> yes
Feb 01 11:03:08 <plntyk> https://dev.openwrt.org/log/trunk/package/utils/util-linux -> might try 43422 (before the 2.25.2 update)
Feb 01 11:04:08 <plntyk> ah now hauke speaking :)
Feb 01 11:06:35 <jow_laptop> the config.log would be interesting
Feb 01 11:06:57 <jow_laptop> _trine: ^
Feb 01 11:07:23 <_trine> ah, yes sorry I went to make a coffee back now
Feb 01 11:09:36 <_trine> the config.log for util-linux ??
Feb 01 11:09:48 <plntyk> yes _trine
Feb 01 11:14:21 <_trine> I need to wait a few minutes before I can put it on my ftp server
Feb 01 11:14:23 <builder> build #850 of ppc40x is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org:8010/builders/ppc40x/builds/850
Feb 01 11:14:32 <_trine> I am copying some files
Feb 01 11:15:03 <_trine> unless it will go in another pastebin
Feb 01 11:16:15 <_trine> http://pastebin.com/U9YSBzW7
Feb 01 11:16:28 <_trine> the first 2 pastebins said it was too big
Feb 01 11:18:09 <jow_laptop> _trine: find /home/johng/store/nano2/trunk/staging_dir/target-mips_34kc_uClibc-0.9.33.2/usr/include -name curses.h
Feb 01 11:20:23 <jow_laptop> _trine: did you build with -j ?
Feb 01 11:20:31 <_trine> yes
Feb 01 11:20:33 <_trine> but
Feb 01 11:20:45 <_trine> after the fail I used make V=s
Feb 01 11:20:51 <_trine> and it still failed
Feb 01 11:21:33 <_trine> I dont see /home/johng/store/nano2/trunk/staging_dir/target-mips_34kc_uClibc-0.9.33.2/usr/include -name curses.h
Feb 01 11:24:05 <_trine> I do see /home/johng/store/nano2/trunk/staging_dir/target-mips_34kc_uClibc-0.9.33.2/usr/include/ncursessw curses.h
Feb 01 11:24:06 <jow_laptop> I think the PKG_BUILD_DEPENDS in util-linux is wrong
Feb 01 11:24:53 <jow_laptop> do you have a build_dir/target-mips_34kc_uClibc-0.9.33.2/libncurses ?
Feb 01 11:25:28 <_trine> no
Feb 01 11:26:00 <_trine> I have build_dir/target-mips_34kc_uClibc-0.9.33.2/libncursesw
Feb 01 11:26:44 <jow_laptop> can you pstebin the output of ./scripts/diffconfig.sh ?
Feb 01 11:27:36 <_trine> http://paste.debian.net/143331/
Feb 01 11:30:11 <jow_laptop> ok
Feb 01 11:30:32 <jow_laptop> if ncursesw is selected, the headers are installed to .../usr/include/ncursesw where util-linux will not find them
Feb 01 11:30:34 <plntyk> jow_laptop: I previously tried to enable generating .pc files for ncurses/ncursesw but somehow this did not help either
Feb 01 11:30:50 <jow_laptop> if ncrusesw is not selected (the default), all is fine
Feb 01 11:31:24 <jow_laptop> if you check package/libs/ncurses/Makefile you'll see that it has different InstallDev sections, depending on whether cursesw is selected or not
Feb 01 11:32:18 <_trine> are you saying I should deselect it and if so how do I deselect it
Feb 01 11:33:54 <plntyk> _trine: libncursesw got selected by python3 .... so you have to disable or change the python3 dependency if its possible
Feb 01 11:34:24 <_trine> ok thats ok because python3 does not seem to work anyway
Feb 01 11:36:14 <jow_laptop> _trine: http://pastebin.com/4kL4BenW
Feb 01 11:36:15 <jow_laptop> try this
Feb 01 11:36:52 <jow_laptop> then, make package/util-linux/compile V=s
Feb 01 11:37:10 <jow_laptop> it might fail later after configure, if it does, pastebin that log as well
Feb 01 11:41:22 <_trine> is this make file change correct
Feb 01 11:41:24 <_trine> http://paste.debian.net/143333/
Feb 01 11:41:56 <jow_laptop> without the + in front
Feb 01 11:42:13 <_trine> yes sorry I just noticed
Feb 01 11:43:41 <_trine> no it failed to compile
Feb 01 11:43:51 <_trine> same error
Feb 01 11:44:16 <jow_laptop> pastebin config.log again
Feb 01 11:46:11 <_trine> http://pastebin.com/H5aAmz0x
Feb 01 11:53:40 <jow_laptop> _trine: another try: http://pastebin.com/PV0vFeM1 - undo the other change
Feb 01 11:55:36 <_trine> trying now
Feb 01 11:56:58 <_trine> http://paste.debian.net/143336/
Feb 01 12:01:25 <jow_laptop> I need config.log again
Feb 01 12:01:31 <_trine> ok
Feb 01 12:02:59 <_trine> http://pastebin.com/pP3jca9j
Feb 01 12:13:33 <plntyk> WARNING: ncurses.h: accepted by the compiler, rejected by the preprocessor!
Feb 01 12:13:38 <plntyk> thats strange
Feb 01 12:13:47 <_trine> yes that was a change
Feb 01 12:14:16 <_trine> after the change to the makefile
Feb 01 12:15:09 <plntyk> ofc it finds "ncurses.h" but it still thinks thats the ncurses one -> so it links with "lncurses" and not what you want: "-lncursesw"
Feb 01 12:15:46 <jow_laptop> _trine: next try: http://pastebin.com/2hB0z4yM
Feb 01 12:15:53 <_trine> ok
Feb 01 12:16:05 <jow_laptop> again, undo the previous change
Feb 01 12:16:37 <jow_laptop> plntyk: its a special situation - util-linux declars a build dependency on ncurses to satisfy configure
Feb 01 12:16:58 <jow_laptop> three applets actually require libncurses(non-w), cal, cfdisk and setterm
Feb 01 12:17:24 <jow_laptop> if any of those applets is actually enabled, the package dep will take care of enabling ncurses(non-w) and all will work
Feb 01 12:18:13 <jow_laptop> if none of those three applets is enabled, the libncurses(non-w) is not enabled either in menuconfig and thus the build_dep on ncurses will merely trigger the build of ncursesw if that is enabled by any other package
Feb 01 12:18:44 <jow_laptop> so the current attempt to solve it is to simply disable ncurses support entirely in configure if no curses-dependant applet is enabled
Feb 01 12:19:18 <jow_laptop> if such an applet is enabled it will also pull in the proper ncurses dep, so no need for a build dep
Feb 01 12:19:59 <_trine> this is the makefile changes I used and this is the result http://paste.debian.net/143338/
Feb 01 12:20:58 <jow_laptop> _trine: pastebin the svn diff / git diff output
Feb 01 12:21:35 <jow_laptop> I think you made a mistake when editing the makefile
Feb 01 12:21:39 <jow_laptop> a syntax error or something
Feb 01 12:22:17 <_trine> I am not sure what you want me to do > trine: pastebin the svn diff / git diff output
Feb 01 12:22:47 <jow_laptop> did you use svn checkout or git clone?
Feb 01 12:22:54 <_trine> svn
Feb 01 12:23:06 <jow_laptop> then pastebin the output of "svn diff package/" please
Feb 01 12:23:52 <_trine> http://paste.debian.net/143339/
Feb 01 12:24:47 <builder> build #798 of avr32 is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/798
Feb 01 12:25:05 <jow_laptop> _trine: you should have "--without-udev" twice in the Makefile
Feb 01 12:25:12 <jow_laptop> _trine: remove the one without "\" in the end
Feb 01 12:25:21 <jow_laptop> _trine: also remove "--with-ncurses  \" further up
Feb 01 12:26:17 <_trine> like this > http://paste.debian.net/143340/
Feb 01 12:26:19 <jow_laptop> the end result should look like: http://pastebin.com/cWzaVrW1
Feb 01 12:26:41 <jow_laptop> yeah, and make an empty line between line 9 and line 10
Feb 01 12:30:47 <_trine> it seems to be compiling now,, I will let you know when it finishes
Feb 01 12:31:08 <_trine> this is the makefil I am using http://paste.debian.net/143341/
Feb 01 12:32:07 <jow_laptop> looks well
Feb 01 12:32:18 <jow_laptop> if the build is through, can you repeat the test with:
Feb 01 12:33:02 <_trine> it compile ok
Feb 01 12:33:04 <jow_laptop> echo CONFIG_PACKAGE_cfdisk=m >> .config; make defconfig; make package/util-linux/compile V=s
Feb 01 12:33:43 <_trine> doing that now
Feb 01 12:36:14 <_trine> fail
Feb 01 12:37:05 <_trine> http://paste.debian.net/143343/
Feb 01 12:47:23 <jow_laptop> ah right
Feb 01 12:47:36 <jow_laptop> make package/util-linux/{clean,compile} V=s
Feb 01 12:47:42 <_trine> ok
Feb 01 12:48:09 <jow_laptop> (missing PKG_CONFIG_DEPENDS)
Feb 01 12:48:49 <_trine> I am compiling it now
Feb 01 12:53:30 <_trine> http://paste.debian.net/143346/  it compiled ok
Feb 01 12:56:17 <_trine> good job thanks
Feb 01 13:13:48 <OpenWrt-Commits> jow r44236 trunk/package/utils/util-linux/Makefile * util-linux: only enable curses support if required
Feb 01 13:56:26 <builder> build #56 of ramips.mt7621 is complete: Failure [failed compile_4]  Build details are at http://buildbot.openwrt.org:8010/builders/ramips.mt7621/builds/56
Feb 01 14:12:58 <builder> build #871 of orion is complete: Failure [failed compile_1]  Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/871
Feb 01 14:40:29 <builder> build #324 of pxa is complete: Failure [failed shell_10]  Build details are at http://buildbot.openwrt.org:8010/builders/pxa/builds/324
Feb 01 15:32:06 <puchu> hi
Feb 01 15:32:20 <puchu> which package feed does  Barrier Breaker use?
Feb 01 15:34:02 <plntyk> see https://downloads.openwrt.org/barrier_breaker/14.07/ar71xx/generic/packages/
Feb 01 15:34:07 <plntyk> each feed is a different dir
Feb 01 15:34:21 <plntyk> for ar71xx - other arches in other subdirs
Feb 01 15:35:19 <plntyk> https://dev.openwrt.org/browser/trunk/feeds.conf.default list the URL of some feeds
Feb 01 15:56:04 <puchu> thanks
Feb 01 16:40:21 <_trine> I have another compile error can anyone help http://paste.debian.net/143375/
Feb 01 16:47:37 <_trine> hhmm I just did make package/openssl/{clean,compile} V=s and it compiled
Feb 01 20:45:54 <swalker> Kaloz: given https://secure.ucc.asn.au/hg/dropbear/file/c635aeeaf25c/options.h#l120 should the change be removed from https://dev.openwrt.org/browser/trunk/package/network/services/dropbear/patches/120-openwrt_options.patch?
Feb 01 22:12:23 <builder> build #799 of avr32 is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/799
**** ENDING LOGGING AT Mon Feb 02 02:59:58 2015