**** BEGIN LOGGING AT Sat May 07 02:59:58 2011 May 07 03:02:24 build #30 of s3c24xx is complete: Failure [failed compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/s3c24xx/builds/30 May 07 05:21:16 build #31 of at91 is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/at91/builds/31 May 07 05:42:24 build #31 of ubicom32 is complete: Failure [failed compile_3] Build details are at http://buildbot.openwrt.org:8010/builders/ubicom32/builds/31 May 07 09:25:11 juhosg * r26840 /trunk/target/linux/ar71xx/files/arch/mips/ar71xx/dev-ar9xxx-wmac.c: ar71xx: override wmac device name for the AR934x SoCs May 07 09:25:12 juhosg * r26841 /trunk/package/base-files/files/etc/banner: base-files: fix a typo in etc/banner May 07 09:38:47 build #30 of brcm47xx is complete: Exception [exception interrupted] Build details are at http://buildbot.openwrt.org:8010/builders/brcm47xx/builds/30 May 07 09:38:48 build #29 of atheros is complete: Exception [exception interrupted] Build details are at http://buildbot.openwrt.org:8010/builders/atheros/builds/29 May 07 09:38:50 build #29 of cobalt is complete: Exception [exception interrupted] Build details are at http://buildbot.openwrt.org:8010/builders/cobalt/builds/29 May 07 09:38:52 build #28 of lantiq is complete: Exception [exception interrupted] Build details are at http://buildbot.openwrt.org:8010/builders/lantiq/builds/28 May 07 10:01:54 build #28 of pxcab is complete: Failure [failed compile_3] Build details are at http://buildbot.openwrt.org:8010/builders/pxcab/builds/28 May 07 10:02:28 build #28 of ps3 is complete: Failure [failed compile_3] Build details are at http://buildbot.openwrt.org:8010/builders/ps3/builds/28 May 07 12:16:52 nbd: something like this? -> http://inuyasha.ath.cx/~jonas/break-bridge-ipv6-dependency.patch (only compile tested with bridge=y, ipv6=m, not runtime tested - the (de-)assignment is also just guessed, not really thought about) May 07 12:25:38 how about this instead: May 07 12:25:51 make a weak reference to ipv6_dev_get_saddr May 07 12:25:53 if it's set, use it May 07 12:25:57 if not, fill in a default May 07 12:26:09 (which calculates the link local address for br->dev) May 07 12:26:42 that way ipv6 multicast snooping will function properly even when the bridge device does not have ipv6 support May 07 12:27:42 hm May 07 12:28:48 no, that probably wouldn't be enough May 07 12:31:28 '(IPV6 || IPV6=n)' is also used for a few other modules May 07 12:31:28 i think i'd prefer using a weak reference over using rcu May 07 12:31:51 I just copied the netfilter example given in the thread May 07 12:33:17 actually, i think the weak reference based version could be implemented directly in the header file May 07 12:33:22 requiring no changes to callsites May 07 12:33:40 though it would still make sense to add a wrapper in the bridge code to check the return code and fill in a sane default May 07 12:34:04 my current implementation doesn't either - I didn't change the bridge code itself ;) May 07 12:34:25 but yours fails if callsites do not use rcu May 07 12:34:41 or at least it has a race condition then May 07 12:36:02 hm, rcu could be added to the wrapper though May 07 12:38:58 nbd: so just using rcu_dereference isn't enough? May 07 12:39:11 no May 07 12:40:26 anything accessing data behind the rcu pointer needs to be wrapped in rcu_read_lock() and rcu_read_unlock() May 07 12:41:53 synchronize_rcu() basically waits for all open rcu read locks to be closed (not counting any new ones that were opened during the wait period) May 07 12:42:04 juhosg * r26842 /trunk/target/linux/ar71xx/files/drivers/net/ag71xx/ (ag71xx.h ag71xx_debugfs.c): ar71xx: ag71xx: use debugfs_remove_recursive May 07 12:42:07 juhosg * r26843 /trunk/target/linux/ar71xx/files/drivers/net/ag71xx/ag71xx_debugfs.c: ar71xx: ag71xx: use dynamically allocated buffer for napi_stats May 07 12:42:10 juhosg * r26844 /trunk/target/linux/ar71xx/files/drivers/net/ag71xx/ (ag71xx.h ag71xx_main.c): ar71xx: ag71xx: add timestamps to ag71xx_buf May 07 12:42:13 juhosg * r26845 /trunk/target/linux/ar71xx/files/drivers/net/ag71xx/ag71xx_debugfs.c: ar71xx: ag71xx: add debugfs entry for [rt]x_ring May 07 12:42:46 nbd: interesting - the netfilter module just uses rcu_dereference, so that't then racy, too? (-> http://lxr.linux.no/#linux+v2.6.38/net/netfilter/nf_conntrack_sip.c#L1224 ) May 07 12:44:22 i guess the entire netfilter match call might be wrapped in rcu_read_lock/unlock May 07 12:45:27 nbd: ah, hm, yeah, that might be it. Haven't looked deep (or high) enough May 07 12:46:46 ah, rcu_read_lock/unlock wraps the entire netfilter packet hook May 07 12:50:57 hauke * r26846 /trunk/target/linux/brcm47xx/ (14 files in 2 dirs): brcm47xx: update patches May 07 14:05:34 Lately, python-xapian package fails to compile as shown here (https://forum.openwrt.org/viewtopic.php?id=29915). Has anyone seen this? May 07 16:54:00 build #31 of s3c24xx is complete: Failure [failed compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/s3c24xx/builds/31 May 07 17:38:02 nbd: ping May 07 18:06:06 Hauke: howdy. May 07 18:06:36 Odd how some weeks that everyone seems to hack all weekend long, and others no one seems to be around. May 07 18:06:44 it's like swarms. May 07 18:10:26 philipp64|laptop: hi ;-) May 07 18:10:49 quiet weekend, huh? you could hear a pin drop... May 07 18:13:14 philipp64|laptop: for now I just hear a bird peeping quite loudly ;-) May 07 18:30:57 anyone have a comment on http://patchwork.midlink.org/patch/970/ ? May 07 18:31:47 it's been sent upstream to Wietse... May 07 19:05:28 * philipp64|laptop relaunches firefox after it hangs trying to print a window... May 07 21:08:02 build #21 of etrax is complete: Failure [failed compile_3] Build details are at http://buildbot.openwrt.org:8010/builders/etrax/builds/21 May 07 22:48:31 * russell-- wonders if someone will apply my rsyncd patch: http://patchwork.midlink.org/patch/926/ May 07 23:17:01 build #17 of iop32x is complete: Exception [exception interrupted] Build details are at http://buildbot.openwrt.org:8010/builders/iop32x/builds/17 May 07 23:17:01 build #19 of adm5120 is complete: Exception [exception interrupted] Build details are at http://buildbot.openwrt.org:8010/builders/adm5120/builds/19 May 07 23:39:37 build #19 of mpc52xx is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/mpc52xx/builds/19 May 08 00:37:38 build #32 of at91 is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/at91/builds/32 May 08 00:57:14 build #32 of ubicom32 is complete: Failure [failed compile_3] Build details are at http://buildbot.openwrt.org:8010/builders/ubicom32/builds/32 May 08 01:21:13 russell--: if you're lighting candles, light one for me too. :-) May 08 01:35:15 build #22 of iop32x is complete: Exception [exception interrupted] Build details are at http://buildbot.openwrt.org:8010/builders/iop32x/builds/22 May 08 01:35:16 build #30 of atheros is complete: Exception [exception interrupted] Build details are at http://buildbot.openwrt.org:8010/builders/atheros/builds/30 **** ENDING LOGGING AT Sun May 08 02:59:58 2011