**** BEGIN LOGGING AT Sat Mar 10 02:59:58 2012 Mar 10 05:01:37 swalker * r30860 /packages/multimedia/fswebcam/Makefile: [packages] fswebcam: update to 20110717 Mar 10 07:30:30 build #118 of kirkwood is complete: Failure [failed shell] Build details are at http://buildbot.openwrt.org:8010/builders/kirkwood/builds/118 Mar 10 13:29:41 nbd * r30861 /trunk/tools/Makefile: tools: add dependency on the seq symlink Mar 10 13:29:44 nbd * r30862 /trunk/package/mac80211/Makefile: mac80211: add the real compat-wireless version + openwrt revision to the compat_version Mar 10 13:29:48 nbd * r30863 /trunk/package/mac80211/patches/ (13 files): mac80211: reorganize patches, fold in patches that were merged upstream Mar 10 13:29:51 nbd * r30864 /trunk/package/mac80211/patches/ (4 files): ath9k: merge a fix for hardware full sleep Mar 10 13:30:00 nbd * r30865 /trunk/package/mac80211/patches/ (18 files): mac80211: more patch reorganization, merge an upstream performance optimization patch Mar 10 13:30:03 nbd * r30866 /trunk/package/mac80211/patches/ (5 files): mac80211/ath9k: some more performance improvements Mar 10 13:30:06 nbd * r30867 /trunk/package/mac80211/patches/300-pending_work.patch: ath9k: merge an ANI improvement patch, improves performance in some environments Mar 10 13:30:10 nbd * r30868 /trunk/package/mac80211/patches/512-ath9k_reduce_rxbuf.patch: ath9k: increase rx buffers for improved performance with 3x3 chipsets Mar 10 13:30:19 nbd: oi, wtf is avgrate? Mar 10 13:30:28 also nudge https://forum.openwrt.org/viewtopic.php?pid=159963 Mar 10 13:31:12 avgrate? context? Mar 10 13:31:25 your qos-scripts Mar 10 13:31:39 at least i think it's still yours Mar 10 13:31:53 it's the average rate Mar 10 13:32:00 as the name implies Mar 10 13:32:08 about opendpi: i think as-is it's crap Mar 10 13:32:22 because the netfilter binding sucks Mar 10 13:32:28 i'm still working on my variant of it, but it's not complete yet Mar 10 13:32:34 i know it's not great but l7 was worse Mar 10 13:32:43 yeah, l7 is worse Mar 10 13:32:46 but that doesn't say much ;) Mar 10 13:32:48 we as long as you have something in the pipe Mar 10 13:32:52 well* Mar 10 13:33:07 is avgrate an upper bound then? Mar 10 13:33:27 yes Mar 10 13:33:34 for my opendpi variant i also wrote an improved bittorrent pattern matching algorithm Mar 10 13:33:58 in my tests, it matched >97% of my bittorrent traffic Mar 10 13:34:12 without false positives Mar 10 13:34:23 so much fucking want Mar 10 13:36:22 actually, i need to figure out what some guys in this building are running over udp. I've had to just brute force filter all udp since it kills everything Mar 10 13:37:13 ok i'm brain farting here. if avgrate is an upper bound wouldn't that mean "Normal" doesn't get more the 10% of avalible bandwidth? Mar 10 13:37:22 i think i need to go over the tc docs Mar 10 13:37:59 sorry, it's not an upper bound Mar 10 13:38:01 limit is the upper bound Mar 10 13:38:11 avgrate is an upper bound only when all queues are fully loaded Mar 10 13:39:01 ah, /doc/uci/qos also uses maxrate when i was told limitrate Mar 10 13:41:11 i don't remember the details, it's been about 4 years since i did any significant work on this thing Mar 10 13:42:48 i keep hoping that guy on the ML will finish his per client sharing extension Mar 10 13:43:03 at some point i will probably rewrite that thing Mar 10 13:43:19 i wouldn't have to car about the UDP flood if one client couldn't hog all BW Mar 10 13:43:22 when i have some time to dig into the improvements that the bufferbloat guys made Mar 10 13:43:25 care* Mar 10 13:44:16 openwrt 12.12.12; rewrite all the things Mar 10 13:45:02 also if you don;t get a release on 12.12.12 @ 12:12 that will be sad Mar 10 13:45:18 * nbd doesn't care much about fancy numbers Mar 10 13:46:10 boo Mar 10 13:46:56 long gone are the days of mbm counter trolling the main channel Mar 10 13:47:55 now everything is serious Mar 10 14:01:51 nbd * r30869 /trunk/package/mac80211/patches/ (2 files): ath9k: fix passing MAC time to mac80211 Mar 10 14:01:54 nbd * r30870 /trunk/package/mac80211/patches/ (2 files): mac80211: limit TID buffering to prevent out-of-memory issues on low-memory systems Mar 10 14:01:57 nbd * r30871 /trunk/package/mac80211/patches/ (3 files): rt2x00: merge a fix for random tx stalls Mar 10 15:21:11 build #119 of ar71xx is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/ar71xx/builds/119 Mar 10 15:25:37 build #135 of ps3 is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/ps3/builds/135 Mar 10 15:52:29 build #112 of iop32x is complete: Failure [failed compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/iop32x/builds/112 Mar 10 16:36:22 build #135 of pxcab is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/pxcab/builds/135 Mar 10 19:23:36 nbd: I'm glad you found that issue with module_init(), though it's too bad the one person affected by it seems to have fallen off the face of the earth. Mar 10 19:52:21 by the way, did the fix get upstreamed? Mar 10 19:54:37 nbd: toolchain/uClibc/patches-0.9.33/160-mips_signalfd_sfd_nonblock.patch is no longer needed; it's in upstream Mar 10 20:12:32 swalker * r30872 /packages/net/weechat/Makefile: [packages] weechat: update to 0.3.7, disable more scripting languages **** ENDING LOGGING AT Sun Mar 11 02:59:58 2012