**** BEGIN LOGGING AT Wed Sep 02 02:59:58 2015 Sep 02 09:24:53 nbd r46764 trunk/package/network/services/mdns/files/mdns.init * mdns: fix having mulitple network entries in uci Sep 02 09:24:59 nbd r46765 trunk/target/linux/ generic/patches-3.18/645-bridge_multicast_to_unicast.patch generic/patches-4.1/645-bridge_multicast_to_unicast.patch * kernel: bridge, multicast-to-unicast: fix echoes on STA Sep 02 09:25:05 nbd r46766 trunk/target/linux/ramips/patches-3.18/0065-fix_dts_cache_issues.patch Sep 02 09:25:05 ramips: fix devicetree corruption with some boot loaders if the caches are not ready at boot Sep 02 09:27:22 nbd r46767 branches/chaos_calmer/target/linux/ramips/patches-3.18/0065-fix_dts_cache_issues.patch Sep 02 09:27:22 ramips: fix devicetree corruption with some boot loaders if the caches are not ready at boot Sep 02 09:44:54 build #85 of ppc40x is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/ppc40x/builds/85 Sep 02 10:19:15 jogo r46768 trunk/ (20 files in 8 dirs) * kernel: update 3.18 to 3.18.21 Sep 02 11:49:57 cyrus r46769 trunk/package/network/ (7 files in 2 dirs) * dropbear: bump to 2015.68 Sep 02 11:50:00 cyrus r46770 trunk/package/network/ services/dnsmasq/files/dnsmasq.init services/dnsmasq/Makefile * dnsmasq: make /tmp/dnsmasq.d and /tmp/hosts preferred over UCI settings Sep 02 11:55:03 jow r46771 trunk/package/base-files/ files/etc/hosts Makefile * base-files: add /etc/hosts entries for ::1, ff02::1 and ff02::2 Sep 02 13:26:48 cyrusff: Noticed Linus' multicast STA patch was pushed to git - all is very nice here now :-) Sep 02 13:29:34 rmilecki: yeah, /sys/bus/usb worked, been trying too many other places in /sys and /proc :) (was on a board that didn't have usbutils installed) Sep 02 13:31:11 anyone know why "time lsusb" has different output (and speed) than lsusb? https://pastee.org/s3mfg Sep 02 13:32:03 time sh -c 'lsusb' prints the full text, and takes 1.7 secs, much like lsusb by itself. Sep 02 14:53:08 swalker: you gave me the makefile for pygattlib the other day. do you happen to have by any chance the dependencies also? boot-thread and boost-python iirc. Sep 02 15:30:03 jow_laptop: got around to the set_list clearing: http://palmtree.beeroclock.net/~karlp/patches/0001-uci-set_list-Delete-option-if-the-list-is-empty.patch Sep 02 15:30:25 also, get_list doc updates, and make it so that it always returns a table, not just most of the time: http://palmtree.beeroclock.net/~karlp/patches/0002-uci-get_list-always-return-a-table-no-matter-what.patch Sep 02 15:45:17 nbd r46772 trunk/toolchain/ (7 files in 2 dirs) * musl: update to version 1.1.11 Sep 02 16:07:57 jow_laptop: hrm, I'll try and put them up on github as pulls with the others, Sep 02 16:08:31 https://github.com/openwrt/luci/pull/440 is doing pretty much exactly what I was doign, (well, in name at least, I didn't look at the implementation) and has the same check on list before deleting Sep 02 16:13:17 karlp: I'd rather want to have a function that keeps the order Sep 02 16:13:34 yeah, I don't do the set recreation that that person does, Sep 02 16:13:43 I'm just adding a comment there, Sep 02 16:13:56 karlp: means remove_list(c, s, o, value) -> traverse list, remove all elements equal value, write back remainder Sep 02 16:14:09 add_list(c, s, o, value) -> append value to list Sep 02 16:14:53 for any more complex handling one should use the low level set() call with the entire list as value, it will be written to uci as-is then Sep 02 16:15:23 yeah, that's pretty much what I did, I just didn't build it into the cursor module, I just had them as generic table/list methods. Sep 02 16:19:04 I should really learn how to make some rpcd plugins one day... Sep 02 16:23:44 nbd r46773 trunk/target/linux/ ar7/patches-3.18/100-fix-highmem-offset.patch ar7/patches-4.1/100-fix-highmem-offset.patch * ar7: fix HIGHMEM_START (#20427) Sep 02 16:27:10 karlp: minor nitpick - the "not value or #value == 0" expression can cause runtime exceptions for any type not being a string or table Sep 02 16:28:54 damnit. Sep 02 16:30:02 you get an explosion right now if you pass an empty set into set_list anyway though :) Sep 02 17:00:46 build #85 of ar71xx.nand is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/ar71xx.nand/builds/85 Sep 02 17:07:06 karlp: :) Sep 02 21:29:00 attila_lendvai: boost is in the packages feed, the pygattlib Makefile would need "DEPENDS:=+boost-python +boost-thread" added within the Package/pygattlib define Sep 02 21:39:11 * attila_lendvai retries after adding the missing deps Sep 02 21:44:42 swalker: the error remains the same, although I tries many things: "fatal error: boost/python/list.hpp: No such file or directory". how to I add dirs to the cc include path? pointers to learn from are also welcome Sep 02 22:06:38 attila_lendvai: TARGET_CFLAGS += -I$(STAGING_DIR)/usr/include Sep 02 22:33:15 swalker: I got to the point when it tries to link the gattlib.so, but I get this "g++: error: unrecognized command line option '-mno-branch-likely'", plus 6 more with different args. is this what got you stuck? any ideas? (I didn't add any gcc args by myself, or at least certainly not this many) Sep 02 22:34:28 attila_lendvai: I didn't get past glib2/libffi iirc Sep 02 22:38:03 build #78 of mpc52xx is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/mpc52xx/builds/78 Sep 03 00:13:59 oh, I see now. somehow the host g++ gets called instead of the target compiler... seen a fix for this in the python-pcapy example. will look at it tomorrow. Sep 03 00:15:23 hrm, actually that fix is something else, but related... :/ **** ENDING LOGGING AT Thu Sep 03 02:59:59 2015