**** BEGIN LOGGING AT Tue Mar 08 02:59:57 2011 Mar 08 03:03:42 build #75 of xburst is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/xburst/builds/75 Mar 08 03:09:38 build #80 of sibyte is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/sibyte/builds/80 Mar 08 03:21:41 nbd: problem persists in distclean'd/rebuilt r25944 Mar 08 03:25:47 that is, still broken Mar 08 03:47:24 build #95 of ubicom32 is complete: Failure [failed compile_3] Build details are at http://tksite.gotdns.org:8010/builders/ubicom32/builds/95 Mar 08 05:07:51 build #79 of x86 is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/x86/builds/79 Mar 08 05:17:49 build #78 of ar71xx is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/ar71xx/builds/78 Mar 08 05:25:37 wtf happened with glib Mar 08 05:26:06 irssi-ssl seems to break the buildbots completely, on all targets Mar 08 06:00:54 jow * r25945 /packages/libs/glib2/Makefile: [packages] glib2: cache two more symbols that cannot be determined during cross compilation Mar 08 06:02:23 jow * r25946 /trunk/include/site/linux: [include] site: cache ac_cv_func_malloc_0_nonnull and ac_cv_func_realloc_0_nonnull Mar 08 06:02:55 jow * r25947 /branches/backfire/include/site/linux: [backfire] merge r25946 Mar 08 07:06:42 build #69 of mpc52xx is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/mpc52xx/builds/69 Mar 08 07:17:57 cshore: looks good Mar 08 07:52:41 kaloz * r25948 /trunk/toolchain/gdb/Makefile: [toolchain/gdb]: upgrade to the 2011.03-0 release from Linaro Mar 08 08:00:12 gmorning Mar 08 09:14:58 jow * r25949 /feeds/desktop/libs/gtkhtml/patches/ (000-disable-editor.patch 010-disable-gconf.patch): [desktop] gtkhtml: extend 000-disable-editor.patch to cover configure.ac, refresh patches Mar 08 10:43:51 build #74 of au1000 is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/au1000/builds/74 Mar 08 11:15:11 build #85 of ps3 is complete: Failure [failed compile_3] Build details are at http://tksite.gotdns.org:8010/builders/ps3/builds/85 Mar 08 12:21:39 build #103 of brcm63xx is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/brcm63xx/builds/103 Mar 08 12:40:40 russell--: which error exactly are you seeing with latest? Mar 08 12:48:54 same one, the dlopen one Mar 08 12:49:28 russell--: ok, i just found the bug Mar 08 12:49:39 iftop: can't resolve symbol 'dlopen' Mar 08 12:49:46 awesome! Mar 08 12:50:07 nbd * r25950 /trunk/package/base-files/Makefile: base-files: the relinked libpthread needs to be linked against libdl, otherwise calls to dlopen() fail (fixes #8978) Mar 08 12:50:36 * russell-- thought -ldl was involved somewhere, didn't know exactly where Mar 08 12:55:45 acinonyx * r25951 /trunk/package/madwifi/files/lib/wifi/madwifi.sh: [package] madwifi: Add support for multiple interfaces Mar 08 13:11:42 nbd * r25952 /trunk/ (2 files in 2 dirs): Mar 08 13:11:42 gcc: move the optimized assembler helpers back into the static libgcc and skip Mar 08 13:11:42 relinking for this arch. due to relocation constraints, the assembler functions Mar 08 13:11:42 cannot be in the shared libgcc and must always be linked in statically Mar 08 14:01:42 build #23 of lantiq is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/lantiq/builds/23 Mar 08 14:11:43 build #76 of kirkwood is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/kirkwood/builds/76 Mar 08 14:17:00 weirdness, iftop exits now but kills the terminal session, going to try distclean/rebuilding Mar 08 14:32:39 I am missing leds under /sys/class/leds.... what module am I missing? hardware specific or generic? Mar 08 14:33:29 using wzr-hp-g300nh Mar 08 14:34:56 hmm, bet its leds_gpio Mar 08 14:42:53 ping xMff Mar 08 14:45:56 pong cshore Mar 08 14:46:05 Shall I commit? Mar 08 14:47:03 cshore: of course Mar 08 14:47:30 xMff: just wanted to make sure you didn't want to wait until LuCI was in place Mar 08 14:47:47 btw, did you had a chance to implement the overlay-enable init action? Mar 08 14:48:20 build #97 of brcm47xx is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/brcm47xx/builds/97 Mar 08 14:48:21 Not yet, I will do that next....I did miniupnpd first because it was broken Mar 08 14:51:13 cshore * r25953 /packages/net/miniupnpd/ (4 files in 2 dirs): [net] miniupnpd: Added uci setting of allow/deny rules, amongst other things. This involved switching to the use of /var/etc/miniupnpd.conf that is generated on the fly, instead of command-line parameters. Mar 08 14:52:47 actually first I will update the wiki Mar 08 14:53:58 cshore: one detail Mar 08 14:54:02 oh, here's something I wanted to ask: should we make the default upload/download large? Mar 08 14:54:04 yes? Mar 08 14:54:16 cshore: in the permission mask, s/ip_mask/int_addr/ Mar 08 14:54:24 erm *pemrission rule Mar 08 14:54:32 doh. Mar 08 14:56:41 oh, that was the localization were it was missing Mar 08 14:57:34 The config_get was right Mar 08 14:59:13 cshore * r25954 /packages/net/miniupnpd/files/miniupnpd.init: [net] miniupnpd: Corrected local ip_mask to int_address. Thanks jow for pointing this out. Mar 08 15:00:24 cshore: there is still a wrong var in the default rule in upnpd.config Mar 08 15:00:31 ah, ok Mar 08 15:00:39 line 13 Mar 08 15:02:58 cshore * r25955 /packages/net/miniupnpd/files/upnpd.config: [net] miniupnpd: Corrected ip_mask to int_addr in upnpd config file too. Thanks jow. Mar 08 15:03:14 now its okay Mar 08 15:03:16 The ironic thing is I got it right in the example Mar 08 15:03:28 indeed Mar 08 15:03:56 build #99 of atheros is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/atheros/builds/99 Mar 08 15:04:08 I wonder wtf is up with the glib crap Mar 08 15:04:20 it did not change at all recently, yet all builders suddenly fail on it Mar 08 15:04:48 But what do you think of increasing the upload/download rates by default (so that we're not limiting by default)? Mar 08 15:04:58 did they change something upstream? Mar 08 15:05:04 nothing Mar 08 15:05:13 neither irssi nor glib changed afaik Mar 08 15:05:39 hmmm...did the builders get new compilers or libc or something? Mar 08 15:05:47 don't know Mar 08 15:05:52 maybe they're just out of memory Mar 08 15:07:08 heh, maybe glib had one of those timebomb things somehow Mar 08 15:07:40 well irssi does not find glib Mar 08 15:07:50 prints some stupid message Mar 08 15:08:02 and does an infinite wait with a "please press enter" message Mar 08 15:08:14 which ultimately kills the build Mar 08 15:08:28 I guess I patch that out now to see whats really going on Mar 08 15:08:55 btw what do you think of changing the upload/download rates by default for miniupnpd Mar 08 15:09:19 or are they informational only? Mar 08 15:09:32 oh sorry Mar 08 15:09:38 they're informational only Mar 08 15:09:50 I suppose for torrent clients and the like to tune their settings Mar 08 15:10:49 will $(( break if users set the rate too high ? Mar 08 15:11:29 yeah Mar 08 15:11:47 but who cares Mar 08 15:12:01 the ui can sanitize and on the cli there are endless ways to break stuff Mar 08 15:12:18 ok Mar 08 15:13:22 so what's the limit for $(( ? 2**32 ? Mar 08 15:13:57 right Mar 08 15:15:27 ok, the max kb is 2**19 ? (10 from 1024, and 3 from 8) since the kb rate is multiplied by those? Mar 08 15:18:24 ok, still more than gigabit (which is what I was concerned about). Mar 08 15:21:15 jow * r25956 /packages/net/irssi-glib1/patches/100-cross_compiling_fix.patch: [packages] irssi-glib1: simply fail if glib1 is not available instead of waiting for a user response Mar 08 15:42:10 build #98 of s3c24xx is complete: Failure [failed compile_10] Build details are at http://tksite.gotdns.org:8010/builders/s3c24xx/builds/98 Mar 08 16:11:34 cshore * r25957 /packages/net/miniupnpd/files/upnpd.config: [net] miniupnpd: In config file corrected int_address to int_addr Mar 08 16:18:50 ping xMff Mar 08 16:18:56 pong Mar 08 16:19:37 I noticed that secure_mode option is currently used as secure_mode, but we've changed it to enable_secure. Should I change it back? Mar 08 16:22:48 no, its ok Mar 08 16:23:12 jow_laptop: ok, then I have to change the defaullt config file Mar 08 16:26:13 jow_laptop: You sure you don't want it for backwards compatibility? Mar 08 16:27:13 xMff: I have to do a commit anyway one way or the other. Mar 08 16:35:07 anyone have a fix for this build error of backfire: Mar 08 16:35:10 make[4]: Entering directory `/vol/backfire_10.03/build_dir/target-mipsel_uClibc-0.9.30.1/busybox-1.15.3' Mar 08 16:35:24 Makefile:431: *** mixed implicit and normal rules. Stop. Mar 08 16:44:57 trying svn now instead of tar ball.... Mar 08 16:56:03 Why http://pastebin.openwrt.org/ points to http://openwrt.org/ ? Mar 08 17:22:05 xMff: http://wiki.openwrt.org/doc/uci/miniupnpd Mar 08 17:25:30 I just deleted the feeds/packages/net/freeswitch/Makefile and did an 'svn up' to pull up a new copy of the Makefile. Then, I did a 'make package/freeswitch/{clean,compile BUILT_LOG=1' and the compilation crashes as shown here (http://pastebin.com/NaW2UmKi). This is still a FreeSWITCH v1.0.6. Has anyone encountered this problem? Mar 08 17:27:21 I don't think you're looking at enough of the compile output Mar 08 17:28:18 cshore: The output is about 110KB in size and PasteBin can't take it. So, I just pasted the last part. Mar 08 17:34:15 build #96 of ubicom32 is complete: Failure [failed compile_3] Build details are at http://tksite.gotdns.org:8010/builders/ubicom32/builds/96 Mar 08 17:35:37 it seems button-hotplug is missing the AUTOLOAD var Mar 08 17:42:33 odd. libgmp compiled but the .h files are not available as dependancy for other packages Mar 08 17:45:50 mazilo: well it's the warnings or errors that matter I'd guess...or other messages other than compiled ok. Mar 08 17:47:00 ohh the path seems wrong... Mar 08 17:49:05 cshore: The compilation stops after the error and did not produce any freeswitch-1.0.6 packages. Mar 08 17:49:38 mazilo: I mean before that error message Mar 08 17:50:20 cshore: Before the error message, there wasn't any error in compilation. Mar 08 17:50:29 mazilo: what about warnings? Mar 08 17:51:20 mazilo: the message itself doesn't really tell me anything, so I'm looking for more Mar 08 17:51:53 mazilo: is the file present in the build_dir? Mar 08 17:52:00 -I$(PKG_BUILD_DIR)/include does not map to the right dir for me Mar 08 17:52:13 i fixed it for the include by hardcoding it. now have to do the same for the lib dir. Mar 08 17:52:23 did some variable change name in backfire? Mar 08 17:52:26 cshore: Is there a place I can upload the whole compilation output that you can take a look at it? Mar 08 17:53:37 Since the error is on switch_swigable_cpp.h file, perhaps this "hecking for Swift... configure: WARNING: not found - looked for /opt/swift" in configure will make sense? Mar 08 17:54:33 mazilo: is swigable something to do with a plugin that requires swift? Mar 08 17:54:46 mazilo: i.e. are you adding broken packages? Mar 08 17:54:51 LetoTo: well shouldn't it rather be $(STAGING_DIR)/usr/include & $(STAGING_DIR)/usr/lib ? Mar 08 17:55:10 cshore: Honestly, I have no idea. As with broken package, I believe it is disabled. Mar 08 17:55:51 mazilo: ok, so you're just doing a standard take things from menuconfig without broken packages set to be included? Mar 08 17:55:51 xMff: yeah i think so. it found the libgmp that was compiled for the host, not the target Mar 08 17:56:07 but I see the target one is not yet billed. despite the dependancy on libgmp in the package Mar 08 17:56:13 heh Mar 08 17:56:18 cshore: I am looking at the 'make menuconfig'. Mar 08 17:57:11 mazilo: have you done something like show broken packages in menuconfig? Mar 08 17:57:23 mazilo: or is this with default options? Mar 08 17:57:26 cshore the "Show broken platforms / packages " is not selected. Mar 08 17:57:38 mazilo: hmm..ok Mar 08 17:57:54 tmp/.config-package.in:1069:warning: 'select' used by config symbol 'PACKAGE_openswan' refer to undefined symbol 'PACKAGE_libgmp' Mar 08 17:57:55 ahh Mar 08 17:57:56 mazilo: that swift is not the error then Mar 08 17:57:57 cshore: I am using GCC Sourcery enhancements. Mar 08 17:58:06 cshore: OK. Mar 08 17:58:51 cshore: Is there a way for the developers to contact Nico to upgrade freeswitch to 1.0.7? Mar 08 17:59:30 mazilo: I don't know what the situation is Mar 08 17:59:38 cshore: OK. Mar 08 17:59:56 mazilo: you could make a patch yourself ? Mar 08 18:00:06 and post to -devel Mar 08 18:00:56 cshore: Not really. I haven't got a chance to look into 1.0.7, yet. When it was just released, I gave it a try with compilation errors. Mar 08 18:02:06 cshore: I hope to be able to find some free time so that I can spend some time on making 1.0.7 compilable. When, up load the patches. Mar 08 18:33:41 cshore * r25958 /packages/net/miniupnpd/files/miniupnpd.init: [net] miniupnpd: Changed secure mode uci option back to secure_mode from accidentally used enable_secure Mar 08 18:33:42 cshore * r25959 /packages/net/miniupnpd/files/ (upnpd.config upnpd.config.example): [net]: miniupnpd: Added enable_upnp option and port option to default and example uci config files Mar 08 18:33:47 cshore * r25960 /packages/net/miniupnpd/Makefile: [net] miniupnpd: Remove keeping miniupnpd.conf example on the device - it's rarely needed documentation and can be found in build_dir/.../miniupnpd... Mar 08 18:45:46 odd. the dvn has no package/gmp Mar 08 18:45:56 I have to grab it from the packages git repo Mar 08 18:46:05 and update the openswan to depend on gmp, not libgmp Mar 08 18:48:45 jow * r25961 /packages/libs/glib/ (8 files in 2 dirs): Mar 08 18:48:45 [packages] glib: the nineties called, they want their macros back... Mar 08 18:48:45 - correct and cleanup configure.ac to make it work with autoreconf Mar 08 18:48:45 - fix other outstanding autofail issues Mar 08 18:48:45 - cleanup and refresh patches Mar 08 18:48:46 - drop uneeded debian patch Mar 08 19:08:50 jow * r25962 /packages/libs/glib/patches/001-automake-compat.patch: [packages] glib: remove some unneeded deltas in the previous patch commit Mar 08 21:18:35 build #93 of cobalt is complete: Failure [failed shell compile_6] Build details are at http://tksite.gotdns.org:8010/builders/cobalt/builds/93 Mar 08 21:41:26 build #71 of rdc is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/rdc/builds/71 Mar 08 21:45:22 build #69 of octeon is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/octeon/builds/69 Mar 08 21:50:36 make[7]: Entering directory `/vol/svn/backfire/build_dir/linux-brcm47xx/openswan-2.6.33/OBJ.linux.mips/lib/libopenswan' Mar 08 21:50:36 mkdir: cannot create directory `/usr/man/man3': Permission denied Mar 08 21:50:39 ugh Mar 08 21:50:51 someone is not setting DESTDIR :P Mar 08 22:46:25 root@OpenWrt:/tmp# lsmod|grep soft Mar 08 22:46:25 cryptosoft 8640 0 Mar 08 22:46:25 ocf 22624 4 ipsec,ubsec_ssb,cryptosoft,cryptodev Mar 08 22:46:29 weeh. getting there..... Mar 08 22:47:14 now to finish getting userland IKE acceleration with ubsec_ssb Mar 08 22:53:53 build #87 of ramips is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/ramips/builds/87 Mar 08 23:10:59 nbd: iftop is still behaving oddly, emitting "Aborted\n @" (at least, as if it emitted that) and freezing the terminal session. Mar 08 23:11:27 weird, that didn't happen in my tests Mar 08 23:11:43 i did a distclean/rebuild to confirm Mar 08 23:12:10 oh, i have updated the kernel from vanilla trunk Mar 08 23:12:19 to 2.6.37.2 Mar 08 23:13:01 maybe i should try plain vanilla Mar 08 23:17:47 it looks/does the same regardless of whether it's on a serial console or ssh session Mar 08 23:18:05 but can still ssh again (to reboot or whatever) Mar 08 23:18:45 i haven't tested it on broadcom yet Mar 08 23:18:49 my tests were on ar71xx Mar 08 23:19:07 okay, my testing has been solely on brcm47xx Mar 09 00:38:59 nbd * r25963 /trunk/package/mac80211/patches/550-ath9k_no_vif_promisc_handling.patch: ath9k: remove support for in-bss promisc mode, the rx filter flag is too broad and in-bss promisc mode is not particularly useful. should fix most (or all) instances of "ath: Could not stop RX..." error messages Mar 09 00:42:21 is it possible to use the /etc/functions.sh interface to save settings back into the files in /etc/config? I would like to make some changes from a function called with config_foreach Mar 09 02:07:06 netprince_: isn't that what uci set ... ; uci commit does? Mar 09 02:14:53 build #86 of ps3 is complete: Failure [failed compile_3] Build details are at http://tksite.gotdns.org:8010/builders/ps3/builds/86 Mar 09 02:17:06 russell--: yes, I think uci will work, I was wondering about the config_load/get/set methods though, I can use uci Mar 09 02:21:59 build #98 of brcm47xx is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/brcm47xx/builds/98 Mar 09 02:23:23 build #72 of adm5120 is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/adm5120/builds/72 Mar 09 02:28:36 Does anyone remember how to compile 8.09 with the symbol table ? Mar 09 02:30:05 Hi Mar 09 02:30:10 I'm back too :) Mar 09 02:30:16 Sorry wrong channel Mar 09 02:36:13 build #104 of brcm63xx is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/brcm63xx/builds/104 Mar 09 02:37:25 build #87 of pxcab is complete: Failure [failed compile_3] Build details are at http://tksite.gotdns.org:8010/builders/pxcab/builds/87 **** ENDING LOGGING AT Wed Mar 09 02:59:57 2011