**** BEGIN LOGGING AT Sun Feb 07 02:59:58 2016 Feb 07 03:09:37 build #212 of cobalt is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/cobalt/builds/212 Feb 07 03:12:41 build #212 of cns21xx is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/cns21xx/builds/212 Feb 07 03:16:42 build #212 of orion is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/212 Feb 07 05:15:20 build #209 of pxa is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/pxa/builds/209 Feb 07 06:55:36 build #156 of mvebu is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/mvebu/builds/156 Feb 07 07:34:02 build #161 of au1000 is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/au1000/builds/161 Feb 07 07:37:06 build #201 of ppc40x is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/ppc40x/builds/201 Feb 07 08:02:14 build #167 of ar71xx is complete: Exception [exception upload] Build details are at http://buildbot.openwrt.org:8010/builders/ar71xx/builds/167 Feb 07 09:42:23 build #208 of x86.64 is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/x86.64/builds/208 Feb 07 10:46:52 build #156 of adm8668 is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/adm8668/builds/156 Feb 07 11:16:16 build #215 of bcm53xx is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/bcm53xx/builds/215 Feb 07 12:01:20 build #191 of brcm2708 is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/brcm2708/builds/191 Feb 07 12:33:05 build #179 of arm64 is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/arm64/builds/179 Feb 07 12:44:21 build #193 of x86.xen_domu is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/x86.xen_domu/builds/193 Feb 07 12:48:21 build #166 of ramips.mt7620 is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/ramips.mt7620/builds/166 Feb 07 12:50:23 build #186 of ar71xx.nand is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/ar71xx.nand/builds/186 Feb 07 12:57:27 build #172 of avr32 is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/172 Feb 07 13:29:13 nbd r48641 trunk/package/kernel/mac80211/ patches/522-mac80211_configure_antenna_gain.patch patches/530-ath9k_extra_leds.patch patches/326-ath9k-make-NF-load-complete-quickly-and-reliably.patch * ath9k: merge fix for noise floor calibration Feb 07 13:29:21 nbd r48642 trunk/config/Config-devel.in * build: don't add -fno-plt for ARC Feb 07 13:29:31 nbd r48643 trunk/package/network/utils/iptables/Makefile * iptables: using external kernel tree should not alter patch behaviour. Feb 07 13:29:38 nbd r48644 trunk/package/system/udev * udev: delete package, it is old and not used by any core packages/targets Feb 07 13:52:05 build #178 of x86.kvm_guest is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/x86.kvm_guest/builds/178 Feb 07 14:18:40 nbd r48645 trunk/target/linux/mpc85xx/ files/arch/powerpc/boot/dts/tl-wdr4900-v1.dts base-files/etc/board.d/02_network * mpc85xx: fix TL-WDR4900 mac address assignment to match original firmware Feb 07 14:18:46 nbd r48646 trunk/package/base-files/files/bin/config_generate * base-files: set mac address in default network config via device sections Feb 07 14:29:55 nbd r48647 trunk/package/kernel/mac80211/ patches/327-mac80211-Remove-MPP-table-entries-with-MPath.patch patches/329-mac80211-Unify-mesh-and-mpp-path-removal-function.patch patches/328-mac80211-let-unused-MPP-table-entries-timeout.patch * mac80211: merge a few mesh fixes Feb 07 14:30:05 nbd r48648 trunk/package/kernel/mac80211/patches/330-mac80211-minstrel-Change-expected-throughput-unit-ba.patch * mac80211: merge a fix for minstrel expected throughput reporting Feb 07 15:33:29 build #179 of x86 is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/x86/builds/179 Feb 07 17:17:32 build #176 of xburst is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/xburst/builds/176 Feb 07 18:33:44 I am curious as to which mt76/mt72 board nbd has been hacking on of late? The only one I knew of seems out of stock. Feb 07 18:34:01 hey dtaht_sf Feb 07 18:34:14 nice patch yesterday. Cheered me up enormously. Feb 07 18:34:55 ;) Feb 07 18:35:16 it seems that having a-msdu support hooked into the software tx queueing code is creating more incentive for other drivers to start using it Feb 07 18:35:20 seems that iwlwifi will pick it up Feb 07 18:35:24 ath10k is going to use it Feb 07 18:35:35 somebody's working on ath9k as well Feb 07 18:35:39 good for de-bloating ;) Feb 07 18:35:40 yep. congrats! Feb 07 18:36:17 so, the devices i'm hacking on are the witi board and a buffalo wsr-1166dhp Feb 07 18:36:24 that buffalo one is not available yet Feb 07 18:36:29 The witi board is out of stock. Feb 07 18:36:31 hopefully that will change at some point Feb 07 18:36:32 hm Feb 07 18:36:34 grump Feb 07 18:36:36 they're preparing a v2 of it Feb 07 18:36:52 with the same chips Feb 07 18:36:56 Did you see me lose my temper on the archer C5 "v2"? Feb 07 18:37:05 no Feb 07 18:37:09 not yet Feb 07 18:37:22 complete swapout from atheros to broadcom. No visible change on the box. Feb 07 18:38:03 btw. are you going to netdev or battlemesh? Feb 07 18:38:16 That has GOTTA be illegal... somewhere. Grump. Feb 07 18:38:34 where are they, and when ndb ? Feb 07 18:38:46 my wife has been pushing me to go to one of those events eventually Feb 07 18:39:16 http://www.netdevconf.org/1.1/ - netdev starts in a few days on feb 10 Feb 07 18:39:37 I wanted to go to netdev, but have visa issues. I wanted to go to battlemesh but think staying home and working on code (for a change) would be a better idea. Feb 07 18:39:45 battlemesh is may 1st in porto: http://battlemesh.org/BattleMeshV9 Feb 07 18:40:14 dtaht_sf: i'm still thinking if i should do battlemesh or not. because it starts right around the time i return from japan Feb 07 18:40:21 hmm, in portugal, probably not in english then Feb 07 18:40:21 Avery's preso has some great data in it in netdev. Feb 07 18:40:22 so i'm considering going there directly from asia Feb 07 18:41:16 groz: english is standard at battlemesh Feb 07 18:41:18 groz: the event will be in english Feb 07 18:41:23 ok Feb 07 18:42:08 pretty cool. I was thinking I'd head to china in late july. But losing university sponsorship for my travels and having a tight budget overall... agh. Feb 07 18:43:45 nbd: Is there a pcie wifi card/chipset worth hacking on besides ath9k? (turris omnia targetted). I was very down on the ath10k (As were you), I have some hope now... Feb 07 18:47:50 i hope that mt7615 will be good, it's the mediatek 4x4 chipset Feb 07 18:48:09 i don't know anything widely available that's worth hacking on right now Feb 07 18:52:39 Even narrowly available would be good. I am glad you are going to netdev. Wish toke could have made it. I could try to attend remotely for some value of remotely. Feb 07 18:55:48 here's an alternative to the witi board: http://www.aliexpress.com/item/Wireless-Router-1200Mbps-dual-band-AC-Router-Support-a-SIM-card-for-3-g-4-g/32382460744.html?ws_ab_test=searchweb201556_1,searchweb201644_2_505_506_503_504_9985_502_10001_10002_10016_10017_10010_10005_10011_10006_10003_10004_10009_10008,searchweb201560_8,searchweb1451318400_-1,searchweb1451318411_6448&btsid=8d90cef7-f08c-4432-acdd-b Feb 07 18:55:55 4a3c099222d Feb 07 18:56:20 i have one of those at home as well Feb 07 18:56:24 and it works with openwrt trunk Feb 07 18:56:30 nbd: do you have any numbers in your head for the kinds of tradeoffs needed for mu-mimo vs txop size (I'd asked the question on linux-wireless) Feb 07 18:56:43 no, i haven't played with mu-mimo yet Feb 07 18:56:52 haven't even looked at the code for it yet Feb 07 18:57:44 well, gang scheduling looked to be a pita, but my overall thrust was definately to start reducing max txop size with larger numbers of stations present. Feb 07 18:58:12 oh, goodie, that board looks pretty ideal... except for the 15-39 day ship time. Feb 07 18:58:48 512M RAM is also pretty nice Feb 07 19:12:33 nbd: anyway, I guess moving something fq_codel like into a library accessible by your hooks is still on my plate? Feb 07 19:13:29 * dtaht_sf wants to drop or mark packets more often Feb 07 19:14:43 * dtaht_sf but mostly wants to FQ the hell out of stuff as close to the "air" as possible. Feb 07 19:15:31 build #210 of realview is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/realview/builds/210 Feb 07 19:16:31 dtaht_sf: i will talk to people on netdev about this too Feb 07 19:20:05 thx! (darn, do wish I could be there) ok, well, I'm signed up for that lib - but needed hooks and hardware (DO, finally, have some research grant money coming in next month, hopefully). FQ - easy. Codel with aggregates, not so much. I also wanted to tear into retries really hard... multicast... crypto and other management frames... Feb 07 19:21:15 so thanks for the hooks! and the hardware steer! Feb 07 19:22:38 dtaht_sf: you could start with a simple version that does not care about aggregation and only tries to manage the per-sta-per-tid queue well, as it would on a normal qdisc Feb 07 19:22:51 right now there are no real aggregation hooks exposed anyway Feb 07 19:23:03 the driver simply calls the mac80211 dequeue function Feb 07 19:26:01 yep. however, lacking hardware with a simpler driver to understand was something of a barrier. The ath9k codebase + mac80211 intimidates the hell out of me. So, I swear, I'm getting on it. Soon. Feb 07 19:29:55 my take on it was that we'd create a fq_codel-like queue as a per-sta-per-tid arrived (somehow), and then destroy it when the station disassociates. we *could* cut the overall memory overhead from 1024 queues with birthday problems to the "cake" set associative one (64?). Or leverage more of a sch_fq rbtree approach. Feb 07 19:43:37 http://mrncciew.com/2014/11/01/cwap-802-11-block-ack/ - pretty good summary, even has timings from the packet captures. 44us Feb 07 20:15:28 nbd r48649 trunk/package/utils/busybox/patches/020-networking-libiproute-fix-displaying-route-table-for.patch * busybox: fix route tables in ip rule show output Feb 07 20:59:55 nbd r48650 trunk/target/linux/ar71xx/ patches-4.4/622-MIPS-ath79-add-support-for-QCA956x-ethernet.patch files/arch/mips/ath79/dev-eth.c patches-4.1/622-MIPS-ath79-add-support-for-QCA956x-ethernet.patch Feb 07 20:59:55 ar71xx: fold patch 622-MIPS-ath79-add-support-for-QCA956x-ethernet.patch into files/ Feb 07 21:00:06 nbd r48651 trunk/target/linux/ar71xx/files/arch/mips/ath79/dev-eth.c * ar71xx: fix MDIO bus probe on QCA956x Feb 07 21:00:16 nbd r48652 trunk/target/linux/ar71xx/files/arch/mips/ath79/mach-tl-wdr6500-v2.c Feb 07 21:00:16 ar71xx: register only the built-in switch MDIO bus, hook up eth0 to it, should hopefully fix regression reported in #21520 Feb 07 21:44:10 build #155 of mpc85xx is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/mpc85xx/builds/155 Feb 07 21:47:48 build #150 of mpc52xx is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/mpc52xx/builds/150 Feb 07 23:12:43 rmilecki r48653 trunk/package/kernel/mac80211/patches/331-brcmfmac-Increase-nr-of-supported-flowrings.patch * mac80211: fix flowrings infinite loops in brcmfmac on BCM4366 Feb 08 01:31:12 build #187 of ar71xx.nand is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/ar71xx.nand/builds/187 Feb 08 02:00:52 build #159 of uml is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/uml/builds/159 **** ENDING LOGGING AT Mon Feb 08 02:59:58 2016