**** BEGIN LOGGING AT Wed Feb 11 02:59:58 2015 Feb 11 03:12:53 jow_laptop: whatever my wierd issue was, make clean fixed it Feb 11 04:10:08 is this the file I flash? openwrt-r44324-ar71xx-generic-uImage-lzma.bin Feb 11 05:57:58 blogic r44382 branches/barrier_breaker/ target/linux/ramips/base-files/etc/uci-defaults/01_leds target/linux/ramips/base-files/etc/diag.sh target/linux/ramips/dts/MLW221.dts target/linux/ramips/base-files/etc/uci-defaults/02_network * BB:ramips:Kingston MLW221 cleanup Feb 11 06:14:15 build #291 of omap is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/omap/builds/291 Feb 11 08:45:18 rmilecki r44383 trunk/target/linux/bcm53xx/patches-3.18/302-ARM-BCM5301X-select-GPIOLIB-as-optional.patch * bcm53xx: drop unused (and broken) GPIOLIB patch Feb 11 09:17:26 rmilecki r44384 trunk/target/linux/generic/patches-3.18/025-bcma_backport.patch * kernel: 3.18: complete backport of some bcma patch Feb 11 09:22:09 plntyk, what about writing a document where all the compilations steps are explained? Feb 11 09:23:16 similar to this http://wiki.openwrt.org/doc/howto/build but lower level Feb 11 09:24:30 its difficult to write sth when some internal stuff is not that understood :) Feb 11 09:24:35 see https://lists.openwrt.org/pipermail/openwrt-devel/2015-February/031044.html Feb 11 09:25:36 plntyk, I was refering to all the layout of tools/ toolchain/ packages/, etc. how they are linked between them etc. Feb 11 09:25:44 while autotools is nice to have there are several problems that can appear "out of thin air" there Feb 11 09:26:20 ah ok thats what you mean Feb 11 09:27:47 but it can be told how the different stages of the compilation are handled Feb 11 09:27:59 or how to add cmake projects, autotools project etc. Feb 11 09:28:27 (it's usually just an include or a flag, but is not really explicitly writen) Feb 11 09:39:07 maybe we could create an index of a nice to have explanations, how to order them, and then start writing some parts Feb 11 09:44:19 build #67 of ramips.mt7628 is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/ramips.mt7628/builds/67 Feb 11 10:06:46 Hi, I'm experiencing bad performance when having relayd and the driver in cli and sta mode, the hardware only has one radio (less than 50% of bandwidth in sta mode) Feb 11 10:07:00 is this expected? Feb 11 10:07:29 rmilecki r44385 trunk/target/ linux/generic/patches-3.18/026-bcma-from-3.20.patch linux/bcm53xx/patches-3.18/800-bcma-use-two-different-initcalls-if-built-in.patch linux/generic/patches-3.14/026-bcma-from-3.20.patch linux/bcm53xx/patches-3.14/800-bcma-use-two-different-initcalls-if-built-in.patch * kernel: backport bcma patches queued for 3.20 Feb 11 10:12:07 blogic r44386 trunk/target/ (183 files in 3 dirs) * ipq806x: update target to v3.18 Feb 11 10:12:19 blogic r44387 trunk/target/linux/generic/config-3.19 * generic: add a few missing symbols for generic config-3.19. Feb 11 10:12:35 blogic r44388 trunk/target/linux/malta/config-3.19 * malta: add config file for kernel 3.19. Feb 11 10:12:36 nbd r44389 trunk/package/network/services/relayd/Makefile * relayd: update to the latest version, adds fixes by Alejandro Enrique Feb 11 10:13:39 rmilecki r44390 trunk/target/ linux/bcm53xx/patches-3.18/081-watchdog-bcm47xx_wdt.c-allow-enabling-on-BCM5301X-ar.patch linux/bcm53xx/patches-3.18/080-watchdog-bcm47xx_wdt.c-add-restart-handler-support.patch linux/bcm53xx/config-3.18 * bcm53xx: backport bcm47xx_wdt changes to 3.18 to use this driver in future Feb 11 10:20:17 blogic r44391 trunk/target/ linux/lantiq/dts/VG3503J.dtsi linux/lantiq/image/Makefile * lantiq: fix VG3503J partition table for newer kernels Feb 11 10:20:37 blogic r44392 trunk/ (119 files in 4 dirs) * brcm2708: update to v3.18 Feb 11 10:23:40 blogic r44393 trunk/target/linux/cns3xxx/Makefile * cns3xxx: set v3.18 as default Feb 11 10:42:57 build #66 of ramips.mt7621 is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/ramips.mt7621/builds/66 Feb 11 10:50:17 Hello, i cannot understand why i get this - http://codepad.org/iV1yCY0s Feb 11 10:50:35 I even redownloaded the patch file via svn and still the same error.. quilt is installed.. Feb 11 11:05:02 build #881 of orion is complete: Failure [failed compile_1] Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/881 Feb 11 11:33:34 nico r44394 trunk/toolchain/gcc/patches/4.8-linaro/951-arm_bug_58595.patch * toolchain/gcc: fix an ICE on ARM (PR58595) Feb 11 11:33:39 nico r44395 trunk/package/utils/util-linux/Makefile * util-linux: disable rpath Feb 11 11:33:55 nico r44396 trunk/package/network/utils/ ebtables/patches ebtables/Makefile ebtables/patches/001-no_rpath.patch * ebtables: disable rpath Feb 11 11:34:09 nico r44397 trunk/config/Config-kernel.in * config: disable kernel tracing on uml Feb 11 11:34:17 nico r44398 trunk/package/devel/oprofile/Makefile * oprofile: mark broken on uml Feb 11 11:34:56 build #67 of ramips.mt7620 is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/ramips.mt7620/builds/67 Feb 11 11:50:01 dape: did you check your .quiltrc ? Feb 11 11:50:51 no, where can i find it? Feb 11 11:51:56 it should be in your home dir- (http://wiki.openwrt.org/doc/devel/patches) - patch: **** malformed patch at line 192: #endif /* _ATH79_MACHTYPE_H */ Feb 11 11:53:01 Raspberry PI 2 with OpenWrt -> http://pastebin.com/TJ52SQxK :D Feb 11 11:53:04 that patch file has no problems on buildbot which is building ar71xx 44382 atm - see http://buildbot.openwrt.org:8010/builders/ar71xx/builds/775/steps/compile_4/logs/stdio Feb 11 11:54:27 Noltari: too easy ! compile + test with gcc 4.9.x and the ssp hardening enabled :) Feb 11 11:55:27 as a first approach it's more than enough ;) Feb 11 11:57:11 plntyk, created the rc file but im still failing at clean,prepare Feb 11 12:07:00 what exact clean/prepare cmd did you use? Feb 11 12:12:27 build #334 of pxa is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/pxa/builds/334 Feb 11 12:12:59 make package/example/{clean,prepare} V=99 QUILT=1 Feb 11 12:13:11 make target/linux/{clean,prepare} V=99 QUILT=1 Feb 11 12:13:31 apologies, be right back in 45 min Feb 11 12:16:28 nico r44399 trunk/target/linux/ramips/mt7620/config-default * ramips/mt7620: add missing config symbol Feb 11 13:03:22 build #67 of ipq806x is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/ipq806x/builds/67 Feb 11 13:06:45 plntyk, i am actually trying to add support for tp-link mr12u by replicating what is added for the similar tp-link mr13u (with the proper changes specific to mr12u) Feb 11 13:07:46 blogic r44400 trunk/package/ kernel/mt76/patches kernel/mt76/Makefile * mt76: update to latest git HEAD Feb 11 13:07:53 blogic r44401 trunk/target/linux/ramips/files/drivers/net/ethernet/ralink/ralink_soc_eth.c * ralink: fix when tx done is 0 also need to clean interrupt status Feb 11 13:08:02 blogic r44402 trunk/target/ linux/brcm2708/patches-3.14 linux/brcm2708/config-3.14 * brcm2708: remove 3.14 support Feb 11 13:08:11 blogic r44403 trunk/target/ (13 files in 5 dirs) * brcm2708: add Raspberry Pi 2 support Feb 11 13:08:16 blogic r44404 trunk/target/linux/brcm2708/config-3.18 * brcm2708: remove CMA_DEBUG Feb 11 13:09:35 plntyk, http://codepad.org/vQbjmyrk here are the modifications i made, i would like to send them to openwrt devel but i am stuck at kernel 610-MIPS-ath79-openwrt-machines.patch right now Feb 11 13:10:22 right now my tp-link mr12u runs the mr3040 image so i will be able to test only the sysupgrade image but.. kernel compiling fails me Feb 11 13:12:18 blogic r44405 trunk/target/linux/brcm2708/Makefile * brcm2708: add missing SUBTARGETS:= to makefile Feb 11 13:12:38 i should probably but directly the mailing list Feb 11 13:26:10 dape: the patch for "tp-link.mk" did not apply (missing trunk/) at my end Feb 11 13:28:44 plntyk, how about this one? http://codepad.org/ZB2Y3miU Feb 11 13:44:26 build #68 of oxnas is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/oxnas/builds/68 Feb 11 13:45:25 dape: i think I know why your patch failed - trying to compile atm Feb 11 13:48:19 plntyk, are you watching the mailing list? i just eliminated the copyed mach and used the mr13u mach and it seems to compile now Feb 11 13:51:34 yes the solution on the mailing list is cleaner Feb 11 13:51:59 see it now - i do not check it every minute :) Feb 11 13:53:19 blogic r44406 trunk/package/kernel/brcm2708-gpu-fw/Makefile Feb 11 13:53:20 brcm2708-gpu-fw: revert to github releases, adapt to bcm2708 target changes and add back InstallDev Feb 11 13:53:35 blogic r44407 trunk/target/linux/oxnas/Makefile * oxnas: disable usbgadget support Feb 11 13:54:03 dape: the patch of your 610-... file was missing the modification of internal offsets - Feb 11 13:54:13 see for comparison the working: http://paste.debian.net/145602/ Feb 11 13:55:12 dape: when you encountered the error - simply change into the build_dir/target/linux../linux dir and do quilt series -v Feb 11 13:55:15 thanks but shouldnt the already existing mr13u mach be used since the devices are identical except the battery? Feb 11 13:55:43 i think i over complicated things with the added kernel patch Feb 11 13:55:44 yes - but your workflow was somewhat flawed when you created the 610 patch Feb 11 13:56:10 i should have used the mr13u mach.. Feb 11 13:56:37 patching a ".patch" file only leads to trouble - you did it the "wrong" way Feb 11 13:57:08 using quilt series -v , quilt edit and add the information in the kernel would have worked flawlessly Feb 11 13:57:15 i am kindof new to quilt but seems interesting, will read about it today Feb 11 13:57:48 oops, right now i am failing at.. xtables-addon compiling Feb 11 14:02:33 # warning You have CONFIG_NF_CONNTRACK enabled, but CONFIG_NF_CONNTRACK_MARK is not (please enable). Feb 11 14:02:40 i think i saw this error a while ago Feb 11 14:03:28 dape: forgot the second minor thing with your patch Feb 11 14:03:47 in tools/firmware-utils/src/mktplinkfw.c you use whitespace instead of a TAB Feb 11 14:03:47 Feb 11 14:03:52 oops Feb 11 14:04:38 for your final send in you can use checkpatch.pl in .scripts/ Feb 11 14:04:39 thanks plntyk i fixed it Feb 11 14:05:47 cool tool, seem three out of four are okay Feb 11 14:08:26 plntyk, if i compile everything okay i don't think i can test the factory image anymore since i already have a mr3040 image running on the mr12u, should sysupgrade image testing be enough? Feb 11 14:09:06 don't have the original chinese factory tp-link image, i got this router from DX.com.. Feb 11 14:15:45 sysupgrade testing should probably be enough - just write about what you did test/didnt/couldnt test in the patch set - since the routers are fairly similar it should work anyway ; add the testing/firmware stuff to the wiki page if it gets accepted so ppl can give feedback via editing or so Feb 11 14:30:08 okay, compile succeeded, i am ready to test the sysupgrade image. One more question for you: do i still need to patch trunk/target/linux/ar71xx/config-3.14 with +CONFIG_ATH79_MACH_TL_MR12U=y if i am using the mach-tl-mr13u.c ? Feb 11 14:33:39 no you shouldnt need that (that config symbol is responsible for arch/mips/ath79/Makefile line that you do not have anymore) Feb 11 14:34:11 maybe you could rename the MR13U symbol to MR12_13U to indicate that it supports multiple units Feb 11 14:34:33 dont know Feb 11 14:40:21 well, i will send just three out of four patches, sysupgrade worked fine, i updated the wiki page at http://wiki.openwrt.org/toh/tp-link/tl-mr12u#openwrt_trunk_r44407_local_patched_bootlog Feb 11 14:40:39 thanks a lot for your hints plntyk, i am gonna send the working patches Feb 11 14:43:46 rmilecki r44408 trunk/target/ linux/bcm53xx/config-3.18 linux/bcm53xx/patches-3.18/131-ARM-BCM5301X-add-restart-support.patch * bcm53xx: use bcm47xx_wdt for restarting device Feb 11 15:15:11 jow r44409 trunk/package/base-files/Makefile * base-files: fix target/subtarget value in /etc/openwrt_release Feb 11 17:37:48 rmilecki r44410 trunk/target/ (8 files in 2 dirs) * kernel: backport mainlined bgmac patches from 3.19 Feb 11 17:44:32 I am curious as to the "plan" for chaos calmer. was under the impression that the hope was to "finish" it by the end of 2014. (I was dubious. :)) Feb 11 17:48:41 some manpower was missing so no CC release Feb 11 17:49:20 and if i read the recent submit history right its 3.18 for CC Feb 11 17:49:26 yea, well, everybody limps along as best we can. Was it a matter of finding a funded maintainer? Feb 11 17:50:12 somebody that attended those secret developer meetings probably knows more Feb 11 17:50:18 heh Feb 11 17:51:09 and afaik nbd is doing vacation / something in new zealand Feb 11 17:51:14 yep Feb 11 17:51:28 I was going to stick around NZ so I could meet up. Just got back from there yesterday. Feb 11 17:51:38 There was a lot of interest there for a "NZwrt" Feb 11 17:53:53 was there some interesting wireless event ? Maybe some slides / presentations one can read ? Hauke presented Backports at fosdem Feb 11 17:55:47 lemme dig up the relevant preso Feb 11 17:58:53 linux conf.au was in .nz this year wasn' tit? Feb 11 17:59:01 yes, I missed that tho Feb 11 17:59:02 plntyk: what's "Backports" ? Feb 11 17:59:36 you're dtaht the cerowrt guy right? Feb 11 18:00:07 heh. I don't really define myself as such, more as "dave taht, the burnt out on bufferbloat guy", but yea. Feb 11 18:01:07 I think you've done great work on the bufferbloat case, even if it's felt like pissing in the wind for years :) Feb 11 18:01:50 Anyway, the guy fed up with the CPE available for NZ's ultrawide fiber broadband rollout and seriously proposing openwrt as a replacement Feb 11 18:02:05 gave a talk here, thursday 29th, afternoon, starting at about 2:50 Feb 11 18:02:06 http://new.livestream.com/i-filmservices/NZNOG2015/videos/75358960 Feb 11 18:02:45 heh, the box vodafone uses here for dsl2+ often falls over when you switch to our fibre network too, most people end up having to rent the "better" router they sell if they want to use the fibre properly. Feb 11 18:02:46 In particular the Q&A session was remarkably positive at 3:30 or so Feb 11 18:03:14 oh man, that's a big video Feb 11 18:03:40 my own talk, starting at about 2:15 friday morning is also on the page. Feb 11 18:04:00 ah those Network Operator Group conferences :) Feb 11 18:04:00 yea, well, they DID get them up on the website 15 minutes after the show closed, but a little editing would be nice. :) Feb 11 18:04:16 hang on, what's the talk at 2:50? seems to be a cloudflare talk? Feb 11 18:04:25 oh, nvm, misradidng Feb 11 18:04:42 The NZnog folk were wonderful. There was one session where people were asked to peer, and person after person got up and shared their AS number. Feb 11 18:05:10 reminded me of the old days where we didn't have to go through three layers of management, an accountant and a lawyer to drag a cable between cages. Feb 11 18:05:43 karlp: I gave the best estimate as to times and the date above. I could be off by 10 minutes in either direction. Feb 11 18:05:55 karlp: "Backports" is the infrastructure where the mac80211 package in openwrt comes from Feb 11 18:06:10 from which fq_codel was ripped out 9 months ago, darn it. Feb 11 18:06:22 dtaht2: no, my problem, I was looking at the first video on that page, not the day you said to look at :) Feb 11 18:06:54 ah, jed laundry's talk starts at 3:36 thursday Feb 11 18:07:14 his co has 180,000 crappy home gateways to deploy Feb 11 18:07:22 (thursday afternoon) Feb 11 18:09:45 karlp: thx for the appreciation. Pissing in the wind is fun, so long as you are pissing downwind Feb 11 18:10:27 how's the guy giving the talk? is that you? black t-shirt? Feb 11 18:10:36 the RGW talk? Feb 11 18:10:54 thursday afternoon around 2:55 onwards, someone talking about openwrt? Feb 11 18:11:09 that's jed laundry, who I'd never met before, who ended up singing my tune, out of the blue. Totally fed up with the crap they have to deliver. Feb 11 18:11:37 My talk is friday morning, roughly 2:15 onwards. But it was jed's talk I wanted openwrt folk to see.... Feb 11 18:11:51 * dtaht2 doesn't mind more folk digging his own talks either Feb 11 18:11:51 ok, cued up for tomorrow morning now :) Feb 11 18:14:01 rmilecki r44411 trunk/target/ (6 files in 3 dirs) * kernel: fix bgmac check for chip ID Feb 11 18:14:58 OpenWrt-libtool: link: mips-openwrt-linux-uclibc-gcc-ar cru .libs/libjson-c.a arraylist.o debug.o json_c_version.o json_object.o json_object_iterator.o json_tokener.o json_util.o linkhash.o printbuf.o Feb 11 18:14:58 sorry - this program has been built without plugin support Feb 11 18:14:58 Feb 11 18:15:00 grump Feb 11 18:15:08 * dtaht2 scratches head Feb 11 18:16:41 make dirclean Feb 11 18:16:46 you toolchain needs a rebuild Feb 11 18:16:50 thx Feb 11 18:17:21 while I am at it, is the intent to keep ar71xx frozen on 3.14? Feb 11 18:17:51 * dtaht2 fires off a new build Feb 11 18:17:53 unlikely Feb 11 18:17:57 release goal is 3.18 Feb 11 18:18:18 * dtaht2 notes some GREAT-looking routing related patches landed in 3.19+ Feb 11 18:18:24 cool Feb 11 18:18:36 there's always new toys in current_release + 1 Feb 11 18:18:43 I am hoping to start up make-wifi-fast in april, and that's going to be net next based Feb 11 18:18:44 heh Feb 11 18:18:44 sure Feb 11 18:19:05 iirc nbd already backported the new "routing cache" in netfilter Feb 11 18:19:18 but 3.14 is long in the tooth at this point. I did not mind the idea of CC freezing in january on 3.14. :) cool Feb 11 18:19:27 I imagine nftables isn't ready yet. Feb 11 18:19:43 it works already and probably supports all features we need Feb 11 18:20:05 :) Feb 11 18:20:09 someone needs to rewrite firewall3 into firewall4 to emit equivalent nftables rules from the same configuration Feb 11 18:20:24 maybe there are still some symbol issues regarding all the different/changed iptables/nftables stuff in 3.18 compared to 3.14 Feb 11 18:20:51 I'll probably do it if noone beats me to it, but can't promise any eta Feb 11 18:21:08 also didn't investigate the possibility to let iptables and nftables coexist Feb 11 18:21:34 nftables will be a culture shock for those used to iptables so there should be some sort of migration path Feb 11 18:21:42 people solely using the uci firewall will probably not notice Feb 11 18:22:23 I am under the impression there isn't a compelling performance gain with nftables. Feb 11 18:23:06 a few mbps Feb 11 18:23:35 also little tweaks like ability to decide whether to do strcmp() style interface matching or matching by ifindex Feb 11 18:24:07 * dtaht2 is still fond of interface pattern matches Feb 11 18:24:28 * dtaht2 figures he'll never get traction on routing vs bridging Feb 11 18:26:49 Jed proposes NZWRT at 3:36:20 or so. Feb 11 18:28:34 grump, actually the timestamp Feb 11 18:28:50 was closer to 3:12 Feb 11 18:28:52 grump Feb 11 18:28:59 * dtaht2 hates how the internet handles video Feb 11 18:42:24 * dtaht2 wonders how far along the netgear X4 port of openwrt is along Feb 11 18:45:40 X4? Feb 11 18:46:26 nighthawk x4 ac2350 Feb 11 18:58:53 seems all 11ac drivers use firmware blobs Feb 11 19:01:31 yep Feb 11 19:02:01 at least there is one non chipvendor firmware - thats interesting Feb 11 19:02:10 oh? Feb 11 19:02:55 look at http://www.candelatech.com/ath10k.php Feb 11 19:10:28 build #312 of brcm2708 is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/brcm2708/builds/312 Feb 11 19:23:04 build #879 of cobalt is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/cobalt/builds/879 Feb 11 20:25:45 how nice - https://forum.openwrt.org/viewtopic.php?id=55725 ask "Please write ALL configs for my sitaution" :) Feb 11 20:41:28 build #312 of cns3xxx is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/cns3xxx/builds/312 Feb 11 21:59:31 Package cryptsetup is missing dependencies for the following libraries: libpopt.so.0 Feb 11 21:59:51 however the Makefile had it +libpopt and the libpopt is installed to target.../usr/lib Feb 11 22:00:14 ot Feb 11 22:00:35 it's under staging_dir/target.../usr/lib/libpopt.so.0 Feb 11 22:03:23 make package/libs/popt/clean; make .../cryptsetup/compile worked Feb 11 22:03:42 so it needs rebuild all its dependencies? odd Feb 11 22:18:59 make clean for the lib then rebuild it fix it Feb 12 01:59:55 build #809 of avr32 is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/809 **** ENDING LOGGING AT Thu Feb 12 02:59:58 2015