**** BEGIN LOGGING AT Wed Nov 11 02:59:58 2015 Nov 11 03:05:07 build #123 of avr32 is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/123 Nov 11 03:27:11 build #114 of mpc83xx is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/mpc83xx/builds/114 Nov 11 03:51:07 build #114 of ep93xx is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/ep93xx/builds/114 Nov 11 06:28:54 build #141 of brcm47xx is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/brcm47xx/builds/141 Nov 11 06:29:32 build #141 of bcm53xx is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/bcm53xx/builds/141 Nov 11 06:34:10 what controls which musl files are copied to staging? Nov 11 06:35:32 build #129 of netlogic is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/netlogic/builds/129 Nov 11 06:39:06 build #137 of x86.64 is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/x86.64/builds/137 Nov 11 06:56:46 build #129 of ar71xx.nand is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/ar71xx.nand/builds/129 Nov 11 07:50:55 build #140 of cobalt is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/cobalt/builds/140 Nov 11 07:54:16 build #140 of cns21xx is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/cns21xx/builds/140 Nov 11 07:57:42 build #140 of orion is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/140 Nov 11 08:34:24 nbd r47453 trunk/package/libs/elfutils/ (8 files in 2 dirs) * elfutils: bump to 0.164 Nov 11 08:34:33 nbd r47454 trunk/package/firmware/ath10k-firmware/Makefile * ath10k-firmware: update firmware version for QCA988X Nov 11 08:36:54 nbd r47455 trunk/package/network/ services/hostapd/Makefile services/hostapd/files/netifd.sh services/hostapd/files/hostapd.sh * hostapd: Use network_get_device instead of uci_get_state Nov 11 09:05:49 build #138 of pxa is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/pxa/builds/138 Nov 11 10:31:19 cyrusff: is it possible when using ip6assign to tell the router "you need to configure adress ::2 in the range" ? Nov 11 10:31:22 instead of it using ::1 Nov 11 10:34:24 yeah there is some sort of option for that, let me see Nov 11 10:35:59 stintel: try "option ip6ifaceid ::2" Nov 11 10:37:27 cyrusff: awesome, thanks! Nov 11 10:38:14 I'll have to set that per interface I guess, or is there a way to have it defined globally ? Nov 11 10:38:27 is there some documentation on building an exact version of a release? Nov 11 10:38:31 e.g. CC Nov 11 10:38:40 especially regarding the feeds Nov 11 10:43:06 stintel: this is per interface, it is not possible to change the default Nov 11 10:43:20 cyrusff: ok Nov 11 11:41:17 nbd r47456 trunk/package/libs/libnl-tiny/src/ include/linux/netlink.h include/netlink/handlers.h * libnl-tiny: include Nov 11 11:54:26 why does uci ship list.c/h instead of using libuboxs'? Nov 11 11:54:49 they share the filename, but thats about it Nov 11 11:55:10 its not a mere copy of the linked implementation Nov 11 11:55:15 *linked list Nov 11 11:56:15 jow_laptop: list.h is not the same... is it? Nov 11 11:56:24 is the same I meant Nov 11 11:57:10 you asked for list.c as well Nov 11 11:57:29 the uci codebase simply existed before libubox did Nov 11 11:57:33 oh ok Nov 11 11:57:47 would it be interesting to delete that duplicate? Nov 11 11:58:45 not sure, it would introduce work for no visible gain Nov 11 11:59:31 yeah, that's why I ask =) Nov 11 12:02:41 seems like libuci already depends on libubox, so you can probably just delete list.h Nov 11 12:03:00 and replace #include "list.h" with #include Nov 11 12:18:23 apparently list.h isn't used in uci anymore Nov 11 12:18:39 the last user was ucimap-example Nov 11 12:18:53 i pushed a commit that deletes list.h Nov 11 12:28:38 nice Nov 11 12:31:09 Hmm.... Nov 11 12:33:51 I am repackaging all the base libraries in a feed with other paths to be able to use upstream stuff in old branches such as aa Nov 11 12:40:05 if interested ping me Nov 11 12:50:24 luka r47457 trunk/package/boot/uboot-envtools/files/kirkwood * uboot-envtools: kirkwood: fix EA35/4500 board names Nov 11 12:50:25 luka r47458 trunk/target/linux/ kirkwood/image/Makefile kirkwood/profiles/100-generic.mk * kirkwood: add Linksys routers to default profile Nov 11 13:12:33 jow_laptop: I sent you a patch for rpcd correct a literal use instead of a define Nov 11 13:12:41 I hope it is correctly formatted Nov 11 13:12:47 I always screw up somewhere Nov 11 14:14:30 also, jow_laptop seems like rpcd does need to have iwinfo but it is not specified in the Makefile, without looking in internals, should it be always a dependency or just if the iwinfo plugin is enabled? Nov 11 14:14:46 nbd r47459 trunk/target/linux/ generic/patches-4.1/103-packet_allow_empty_payload.patch generic/patches-4.3/103-packet_allow_empty_payload.patch * kernel: fix an af_packet regression that was breaking PPPoE on 4.1+ (#20707) Nov 11 14:33:34 seems like it's only distribution... I will add the depends to PKG_BUILD_DEPENDS Nov 11 14:42:46 also, how much is it worth to convert iwinfo to cmake ? Nov 11 16:11:18 another discovery, if you don't have libcrypt on the libpath, build fails but not on make Nov 11 16:11:29 (it fails as an undefined symbol) Nov 11 16:15:05 Infinite trouble with glibc stuff, for some reason aa won't let me use any feature_test_macro Nov 11 16:15:16 it's just not letting me see any Nov 11 16:25:53 txomon: rpcd-mod-iwinfo depends on libiwinfo already Nov 11 16:26:22 jow_laptop: the problem is that the compilation doesn't depend on that option Nov 11 16:26:27 ? Nov 11 16:26:33 build I mean Nov 11 16:26:55 it works fine here Nov 11 16:26:59 not sure what you mean Nov 11 16:27:01 it builds everything, iwinfo, ubus etc. and then the option selects what you distribute Nov 11 16:28:19 $ grep 'rpcd/compile += ' tmp/.packagedeps Nov 11 16:28:35 $(curdir)/system/rpcd/compile += ... $(curdir)/network/utils/iwinfo/compile ... Nov 11 16:28:52 no need for redundant pkg build depends Nov 11 16:29:38 any unconditional binary package dependency implies a source package wide build dependency Nov 11 16:30:13 so if a rpcd-mod-iwinfo declared by source rpcd depends on libiwinfo then source rpcd build-depends on libiwinfo Nov 11 16:31:19 regardless of whether rpcd-mod-iwinfo is going to be packaged or not Nov 11 16:32:58 take into account that the way I reproduced this was taking rpcd and packaging it again for aa Nov 11 16:33:28 it won't mark iwinfo as a dependency and it will fail on build Nov 11 16:36:41 maybe AA's dependency tracking is different then, on CC the rpcd build clearly depends on the iwinfo build, as evident in tmp/.packagedeps Nov 11 16:38:27 jow_laptop: may it be because iwinfo is part of base packages? Nov 11 16:39:43 no idea Nov 11 16:40:43 * txomon is trying to unselect iwinfo Nov 11 16:40:56 jow_laptop: and you know something about the crypt() thing in aa? Nov 11 16:41:09 the feature macros I mean Nov 11 16:41:16 or someone that could help me on that Nov 11 16:41:20 no, I never cared about glibc Nov 11 16:41:30 it was always broken and is far too big for my usecase anyway Nov 11 16:45:22 ok Nov 11 17:00:23 build #115 of cns3xxx is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/cns3xxx/builds/115 Nov 11 17:24:57 build #115 of uml is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/uml/builds/115 Nov 11 17:35:14 build #138 of ppc40x is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/ppc40x/builds/138 Nov 11 17:52:22 what controls which musl files are copied to staging? Nov 11 18:51:45 nbd: would it be acceptable to install the octeon crypto mods like this: https://bpaste.net/show/31fe490cdcee ? idea taken from twofish-i586 Nov 11 18:54:03 hmmm, missing autoload like that Nov 11 18:54:12 there is probably a better way Nov 11 18:54:25 and you can drop the ifdef Nov 11 18:55:31 you can overwrite AUTOLOAD for octeon Nov 11 19:00:59 nbd: easy as this? https://bpaste.net/show/63256eb50564 Nov 11 19:01:54 looks good Nov 11 19:02:37 what controls which musl files are copied to staging? Nov 11 19:02:46 ok, thanks. I'll include that in my octeon -> 4.3 series as well then Nov 11 19:02:57 with sha{1,256,512} Nov 11 19:04:36 in the build_dir, musl has things like net/ethernet.h but owrt does not use them. how can this be changed? Nov 11 19:06:50 it's in the toolchain dir Nov 11 19:07:20 so it is exported and available Nov 11 19:07:31 it's not there Nov 11 19:07:45 it's only in the build dir Nov 11 19:07:48 in all of my builds it's there Nov 11 19:09:09 find staging_dir | grep net/ethernet.h Nov 11 19:09:17 staging_dir/toolchain-mips_34kc_gcc-5.2.0_musl-1.1.11/include/net/ethernet.h Nov 11 19:09:46 hmm Nov 11 19:16:48 nbd: for sha512, is it OK like this? https://bpaste.net/show/751de81aab1d it seems crypto-misc has no autoload at all Nov 11 19:16:57 or move sha512 to a separate package Nov 11 20:15:08 went for the 2nd way, just sent out the patch series Nov 11 20:15:20 distemper: see ML :) Nov 11 20:16:03 if you use strongswan (or possible other IPsec software), you might want to wait Nov 11 20:18:00 build #142 of brcm47xx is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/brcm47xx/builds/142 Nov 11 20:32:55 https://dev.openwrt.org/ticket/19175 and https://dev.openwrt.org/ticket/18021 are up for closing Nov 11 23:36:54 the new ToH is really cool Nov 12 00:26:24 nbd r47460 trunk/package/network/config/netifd/Makefile Nov 12 00:26:25 netifd: update to the latest version, contains several fixes, including one for interface ip4/ip6table for device routes Nov 12 01:08:06 nbd r47461 trunk/scripts/diffconfig.sh * scripts/diffconfig.sh: reduce config output with CONFIG_ALL_KMODS Nov 12 01:40:43 build #141 of brcm63xx is complete: Failure [failed shell_14] Build details are at http://buildbot.openwrt.org:8010/builders/brcm63xx/builds/141 Nov 12 02:13:31 build #142 of bcm53xx is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/bcm53xx/builds/142 Nov 12 02:18:36 build #141 of ramips.mt7628 is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/ramips.mt7628/builds/141 Nov 12 02:21:57 build #141 of ramips.mt7621 is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/ramips.mt7621/builds/141 **** ENDING LOGGING AT Thu Nov 12 02:59:58 2015