**** BEGIN LOGGING AT Mon Aug 17 02:59:58 2015 Aug 17 05:58:07 blogic r46613 trunk/target/linux/ (158 files) * ramips: fix indentation and other mistakes in .dts{, i} files Aug 17 05:58:28 blogic r46614 trunk/target/linux/ (8 files in 6 dirs) * ramips: fix Zbtlink ZBT-WR8305RT board name Aug 17 05:58:43 blogic r46615 trunk/target/linux/ (5 files in 2 dirs) * ramips: fix Asus WL-330N{, 3G} boards names Aug 17 05:58:55 blogic r46616 trunk/target/linux/ramips/image/Makefile * ramips: fix Asus RP-N53 and Lenovo Y1{, S} image names Aug 17 05:59:02 blogic r46617 trunk/target/linux/ (10 files in 7 dirs) * ramips: fix Xiaomi MiWiFi Mini board, image and profile names Aug 17 05:59:10 blogic r46618 trunk/target/linux/ ramips/dts/UR-326N4G.dts ramips/dts/UR-336UN.dts ramips/base-files/lib/ramips.sh * ramips: fix UPVEL model names Aug 17 05:59:23 blogic r46619 trunk/target/linux/ ramips/dts/RTN56U.dts ramips/image/Makefile ramips/dts/RT-N56U.dts * ramips: fix Asus RT-N56U dts filename Aug 17 05:59:32 blogic r46620 trunk/target/linux/ramips/image/Makefile * ramips: fix Belkin F7C027 image name Aug 17 05:59:45 blogic r46621 trunk/target/linux/ (6 files in 5 dirs) * ramips: fix Argus ATP-52B board and image names Aug 17 06:00:01 blogic r46622 trunk/target/linux/ (5 files in 4 dirs) * ramips: fix Sitecom WL-341v3 board and image names Aug 17 06:00:17 blogic r46623 trunk/target/linux/ (11 files in 5 dirs) * ramips: fix Olimex models, images and profiles names Aug 17 06:00:24 blogic r46624 trunk/target/linux/ ramips/dts/AIBR100.dts ramips/image/Makefile ramips/dts/AI-BR100.dts ramips/mt7620/profiles/aigale.mk * ramips: fix Aigale Ai-BR100 dts and profile names Aug 17 06:00:37 blogic r46625 trunk/target/linux/ramips/dts/W150M.dts * ramips: fix Tenda W150M model name Aug 17 06:00:45 blogic r46626 trunk/target/linux/ ramips/base-files/lib/ramips.sh ramips/base-files/etc/board.d/01_leds ramips/base-files/etc/diag.sh * ramips: fix Edimax BR-6425 board name Aug 17 06:00:58 blogic r46627 trunk/target/linux/ (13 files in 7 dirs) * ramips: fix Omnima models, images and profiles names Aug 17 06:01:38 blogic r46628 trunk/target/linux/ (7 files in 6 dirs) * ramips: fix Buffalo WMR-300 dts, board, image and profile names Aug 17 06:01:54 blogic r46629 trunk/target/linux/ (12 files in 7 dirs) * ramips: fix 7Links PX-4885 dts{, i}, board, image and profile names Aug 17 06:02:10 blogic r46630 trunk/target/linux/ ramips/base-files/etc/board.d/02_network ramips/rt288x/profiles/edimax.mk ramips/base-files/etc/board.d/01_leds ramips/base-files/etc/diag.sh * ramips: remove old code for Edimax BR-6524N Aug 17 06:02:25 blogic r46631 trunk/target/linux/ (45 files in 5 dirs) * ramips: fix various mistakes in subtargets *.mk profiles Aug 17 06:02:31 blogic r46632 trunk/target/linux/ramips/base-files/lib/ramips.sh * ramips: remove leading spaces, sort boards and optimize base-files/lib/ramips.sh Aug 17 06:02:46 blogic r46633 trunk/target/linux/ramips/base-files/lib/upgrade/platform.sh * ramips: be consistent with case statement in base-files/lib/upgrade/platform.sh Aug 17 06:03:46 blogic r46634 trunk/target/linux/ramips/base-files/etc/board.d/02_network Aug 17 06:03:46 ramips: fix indentation, case statements structure and optimize base-files/etc/board.d/02_network Aug 17 06:04:31 blogic r46635 trunk/target/linux/ramips/base-files/etc/diag.sh * ramips: remove unnecessary LED declaration for WT1520 in diag.sh Aug 17 06:04:44 blogic r46636 trunk/target/linux/ (16 files in 3 dirs) * oxnas: prepare for kernel 4.1 Aug 17 06:04:49 blogic r46637 trunk/target/linux/oxnas/Makefile * oxnas: switch to kernel 4.1 Aug 17 06:04:59 blogic r46638 trunk/package/kernel/brcm2708-gpu-fw/Makefile * brcm2708-gpu-fw: update to latest version Aug 17 06:05:05 blogic r46639 trunk/target/linux/brcm2708/image/Makefile * brcm2708: use BuildDTB in image/Makefile Aug 17 06:05:19 blogic r46640 trunk/target/linux/ (147 files in 3 dirs) * brcm2708: update 4.1 patches Aug 17 06:05:26 blogic r46641 trunk/target/linux/brcm2708/modules.mk * brcm2708: fix kmod-sound-soc-iqaudio-dac and add missing symbol Aug 17 06:05:35 blogic r46642 trunk/target/linux/brcm2708/image/config.txt * brcm2708: override disabled device tree module status from config.txt Aug 17 06:16:21 blogic r46643 trunk/target/linux/ramips/patches-3.18/0012-MIPS-ralink-add-MT7621-support.patch * ramips: Fix uart2/uart3 pinmux order on MT7621 Aug 17 06:16:31 blogic r46644 trunk/target/linux/ramips/patches-3.18/0012-MIPS-ralink-add-MT7621-support.patch * ramips: Fix amount of MT7621 pins controlled by spi group Aug 17 06:16:36 blogic r46645 trunk/target/linux/ramips/patches-3.18/0012-MIPS-ralink-add-MT7621-support.patch * ramips: Fix MT7621 pinmux bits for uart3, uart2, mdio Aug 17 06:16:43 blogic r46646 trunk/target/linux/ mcs814x/patches-3.18/014-debuguart.patch mcs814x/config-3.18 * mcs814x: fix debug UART Aug 17 06:16:55 blogic r46647 trunk/target/linux/mcs814x/files-3.18/arch/arm/ mach-mcs814x/timer.c mach-mcs814x/irq.c * mcs814x: fix interrupt handling Aug 17 06:17:07 blogic r46648 trunk/target/linux/ mcs814x/image/Makefile mcs814x/config-3.18 mcs814x/files-3.18/arch/arm/boot/dts/dlan-usb-extender.dts * mcs814x: use firmware partition splitter on dLAN USB Extender Aug 17 06:17:17 blogic r46649 trunk/target/linux/at91/patches/201-ARM-at91-usb-determine_rate-backport.patch * at91: Backport determine_rate prototype fix Aug 17 06:17:26 blogic r46650 trunk/package/base-files/files/etc/rc.button/reset * base-files: rework reset button script to indicate factory reset Aug 17 06:17:43 blogic r46651 trunk/target/linux/ar71xx/ (10 files in 9 dirs) * ar71xx: Add support for Black Swift Aug 17 06:18:03 blogic r46652 trunk/package/network/services/hostapd/files/netifd.sh * hostapd: Add vlan_file option to netifd.sh Aug 17 06:18:10 blogic r46653 trunk/package/base-files/files/lib/functions.sh * base-files: default_postinst: propagate the real postinst return code Aug 17 06:18:25 blogic r46654 trunk/target/linux/ (19 files in 3 dirs) * kernel: move linux, part-probe dt parsing in generic code Aug 17 06:18:36 blogic r46655 trunk/target/linux/ (19 files in 2 dirs) * ipq806x: add hwspinlock support Aug 17 06:18:45 blogic r46656 trunk/target/linux/ (6 files in 2 dirs) * ipq806x: add SMEM support Aug 17 06:18:51 blogic r46657 trunk/target/linux/ ipq806x/patches-3.18/037-mtd-add-SMEM-parser-for-QCOM-platforms.patch ipq806x/patches-4.1/037-mtd-add-SMEM-parser-for-QCOM-platforms.patch * ipq806x: add QCOM SMEM based MTD parser Aug 17 06:19:05 blogic r46658 trunk/target/linux/ (13 files in 3 dirs) * ipq806x: switch AP148 to using SMEM based MTD parser Aug 17 06:19:17 blogic r46659 trunk/target/linux/ (21 files) * ramips: reorder patches Aug 17 06:19:26 blogic r46660 trunk/target/linux/ramips/base-files/lib/upgrade/platform.sh * ramips: reorder the sysupgrade script Aug 17 06:24:30 blogic r46661 trunk/tools/firmware-utils/src/mktplinkfw.c * tools/firmware-utils: add header version 2 support for mktplinkfw Aug 17 06:24:37 blogic r46662 trunk/target/linux/ar71xx/files/drivers/mtd/tplinkpart.c * ar71xx: add 64kb bootloader mtd parser for tplinkpart Aug 17 06:24:52 blogic r46663 trunk/target/linux/ar71xx/ (11 files in 10 dirs) * ar71xx: add support for TP-Link TL-WDR6500 v2 Aug 17 06:25:21 blogic r46664 trunk/target/linux/ (80 files in 3 dirs) * ramips: use consistent naming scheme for LEDs Aug 17 06:25:34 blogic r46665 trunk/target/linux/ ramips/base-files/etc/board.d/01_leds ramips/base-files/etc/diag.sh * ramips: base-files: optimize etc/diag.sh and etc/board.d/01_leds scripts Aug 17 08:27:18 how can i overlay, for example, /etc/rc.button with a tmpfs on a temporary basis (i.e. i don't care if it survives a reboot, in fact i don't want it to, which is the whole point) ? Aug 17 08:54:27 it seems overlaying an overlayfs might not be possible? Aug 17 08:54:43 russell--: it is supposed to work Aug 17 09:00:32 any pointers would be welcome, i'm staring at /lib/functions/preinit.sh and it's pivot stuff Aug 17 09:01:31 s/it's/its/ Aug 17 09:07:35 is there something special about the handling of an external kernel tree ? I get build errors in busybox regarding typical linux/in6.h // netinet/in.h redefinitions but these should be fixed since ~2013-ish according to git - and even the with similar kernels (3.18.20 external) I get this error Aug 17 09:07:46 mkdir -p /tmp/root Aug 17 09:07:46 mkdir -p /mnt/tmpfs Aug 17 09:07:46 /bin/mount -t tmpfs -o noatime,mode=0755 root /tmp/root Aug 17 09:07:46 /bin/mount -o noatime,lowerdir=/,upperdir=/tmp/root -t overlayfs "overlayfs:/tmp/root" /mnt/tmpfs Aug 17 09:07:54 gives me an error: Aug 17 09:08:09 mkdir -p /tmp/rootmount: mounting overlayfs:/tmp/root on /mnt/tmpfs failed: No such device Aug 17 09:08:42 or rather: "mount: mounting overlayfs:/tmp/root on /mnt/tmpfs failed: No such device" Aug 17 09:13:43 perhaps i'm doing something wrong Aug 17 09:16:21 russell--: is http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/fs/overlayfs?id=ef94b1864d1ed5be54376404bb23d22ed0481feb relevant if you are on newer kernel ? Aug 17 09:17:38 http://git.infradead.org/users/dhowells/unionmount-testsuite.git has overlay(fs) support Aug 17 09:18:14 # uname -a Aug 17 09:18:15 Linux OpenWrt 3.18.19 #79 Sun Aug 2 22:46:12 PDT 2015 mips GNU/Linux Aug 17 09:19:22 anyway, those commands presumably work for mounting jffs2 as an overlay Aug 17 09:22:40 dmesg might contain some info Aug 17 09:26:13 hmm. nope. Aug 17 10:13:08 russell--: you are missing a workdir entry - the docs state one example with only multiple "lowerdir" and then with lowerdir,upperdir,workdir Aug 17 10:13:53 blogic r46666 trunk/package/utils/ugps/Makefile * ugps: update to latest git HEAD Aug 17 10:14:34 blogic r46667 trunk/target/linux/ ramips/patches-3.18/0104-fix_bootargs_handling.patch ramips/patches-3.18/0073-fix_bootargs_handling.patch * ramips: move patches around Aug 17 10:14:52 hey the patch I submitted which was merged in 46649 looks like it got mangled when I pasted to my email client, tabs replaced with spaces, and now it doesn't apply when preparing :( sorry about that Aug 17 10:15:46 how can I help fix, shall I resubmit? Aug 17 10:16:30 russell--: look at the "mount" output on router there is upperdir and workdir by the default overlay mount done - and "The "workdir" needs to be an empty directory on the same filesystem Aug 17 10:16:30 as upperdir." -> upperdir=/overlay/upper,workdir=/overlay/work Aug 17 10:18:00 Wipster: look if copy paste is wrong/not working (pasting from terminal window afaik does not keep spaces) or if your email client is at fault Aug 17 10:18:17 *keep tabs* i mean Aug 17 10:35:16 plntyk: seems like I fell fowl of using gmail online, new message in plaintext mode (which I should have guessed) doesn't like tabs. Aug 17 10:35:44 ah yeah those evil web mailers :) Aug 17 10:36:34 mozilla thunderbird can submit patches if it is configured the right way Aug 17 10:36:51 simplest solution might be git send-email Aug 17 10:39:40 plntyk: good plan, I will set that up Aug 17 11:26:34 blogic r46668 trunk/target/linux/ oxnas/config-3.18 oxnas/patches-4.0 oxnas/patches-3.18 oxnas/config-4.0 * oxnas: remove support for kernel 3.18 and 4.0 Aug 17 11:26:39 blogic r46669 trunk/package/network/utils/iwinfo/Makefile * iwinfo: update to latest git HEAD Aug 17 11:26:44 blogic r46670 trunk/target/linux/lantiq/ patches-3.18/0023-NET-PHY-adds-driver-for-lantiq-PHY11G.patch patches-4.1/0023-NET-PHY-adds-driver-for-lantiq-PHY11G.patch * lantiq: Support newer versions of the PEF7071 ethernet PHYs Aug 17 11:26:50 blogic r46671 trunk/target/linux/ generic/files/crypto/ocf/pasemi/pasemi.c generic/files/crypto/ocf/kirkwood/cesa_ocf_drv.c generic/files/crypto/ocf/ubsec_ssb/ubsec_ssb.c * generic: crypto: remove IRQF_DISABLED Aug 17 11:27:03 blogic r46672 trunk/target/linux/orion/files/arch/arm/mach-orion5x/dt2-setup.c * orion: remove IRQF_DISABLED Aug 17 12:37:58 build #77 of ppc40x is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/ppc40x/builds/77 Aug 17 13:19:38 blogic r46673 branches/chaos_calmer/target/linux/ar71xx/base-files/etc/hotplug.d/firmware/11-ath10k-caldata * ar71xx: Enable ath10k calibration data loading for EnGenius ESR1750 and EPG500. Aug 17 13:19:43 blogic r46674 branches/ (10 files in 9 dirs) * ar71xx: Add support for Black Swift Aug 17 13:19:49 blogic r46675 branches/chaos_calmer/package/utils/ugps/Makefile * ugps: update to latest git HEAD Aug 17 13:20:10 what package provides libubox/ustream-ssl.h? I'm trying to build uhttpd2 locally, and I've got libubox installed, ubus installed, what else am I looking for, and why is something else installing into the libubox/ prefix space? Aug 17 13:25:53 after turning off TLS in uhttpd, now I get a warnings as errors problems, with BSD_SOURCE crap: https://pastee.org/nzpqm Aug 17 13:28:33 well remove -Werror :) Aug 17 13:28:51 yeah, doing that, but gak, this is rather painful to try and build it locally Aug 17 13:32:50 -Werror just seems like a damn landmine whenever anyone uses a different compiler. Aug 17 13:34:52 ok, now... how to get a luci build to even try this out :) Aug 17 13:52:33 <_trine> can I ask if the openvpn bug which I noticed yesterday is an openwrt problem to resolve or an upstream one Aug 17 14:14:49 plntyk: yeah, adding a workdir doesn't help, still get: "mount: mounting overlayfs:/tmp/root on /mnt/tmpfs failed: No such device" Aug 17 14:22:02 maybe running an strace on that command will help ? Aug 17 14:22:04 mount -t overlay doesntmatter -o lowerdir=/mnt/lower,upperdir=/mnt/upper,workdir=/mnt/work /mnt/merged Aug 17 14:22:07 results in Aug 17 14:22:19 doesntmatter on /mnt/merged type overlay (rw,relatime,lowerdir=/mnt/lower,upperdir=/mnt/upper,workdir=/mnt/work) Aug 17 14:22:42 all empty dirs created in /mnt/ Aug 17 14:23:04 on a newly flashed ar71xx tplink wdr3600 with 4.1.x kernel Aug 17 14:24:18 so some part is not recognized as a real directory ? Aug 17 14:24:29 on your end @russell-- Aug 17 14:33:56 i'm still at 3.18.x Aug 17 14:34:31 i think there was some work to support overlay-on-overlay in recent kernels Aug 17 15:26:54 jow_laptop: does http://paste.fedoraproject.org/255904/82508014/ look acceptable to you? (for passing the X-HTTP-Method-Override header to cgis) Aug 17 15:58:53 karlp: sure. is that a somewhat standardized header (or a common convention) or just some arbritary name? Aug 17 16:02:13 relatively standardized. MS uses https://msdn.microsoft.com/en-us/library/dd541471.aspx (X-HTTP-Method) but most other tools seem to use X-HTTP-Method-Override by default, at least. http://fandry.blogspot.com/2012/03/x-http-header-method-override-and-rest.html implies IBM uses X-METHOD-OVERRIDE, but I've never seen that Aug 17 16:04:00 flask patterns uses it, http://flask.pocoo.org/docs/0.10/patterns/methodoverrides/ hanselman uses it, django uses it too: http://www.django-rest-framework.org/topics/browser-enhancements/ Aug 17 16:06:35 google gdata uses it too. https://developers.google.com/gdata/docs/2.0/reference Aug 17 16:10:05 ok, sounds good to me Aug 17 16:11:37 blogic r46676 trunk/target/linux/at91/patches/201-ARM-at91-usb-determine_rate-backport.patch * at91: fix usb rate backport patch tab mangle Aug 17 16:18:24 jow r46677 trunk/package/network/services/uhttpd/Makefile * uhttpd: pass X-HTTP-Method-Override header to cgi scripts Aug 17 16:39:34 jow_laptop: you want me to send that to openwrt-devel then? or do you want to take it from here? Aug 17 16:41:52 it is already comitted Aug 17 16:42:41 jow r46678 branches/chaos_calmer/package/network/services/uhttpd/Makefile * CC: uhttpd: pass X-HTTP-Method-Override header to cgi scripts Aug 17 16:46:21 groovy :) Aug 17 16:47:11 I consider it a stop-gap solution until I get around to refactor the code to pass all x-headers Aug 17 16:47:26 or any header actually Aug 17 16:48:09 yeah, I didn't want to try any major change like that, was't sure where you wanted to go with it. Aug 17 16:48:28 can someone push my patch https://lists.openwrt.org/pipermail/openwrt-devel/2015-August/035137.html [PATCH packages] lighttpd: backport a fix for running out of filedescriptors Aug 17 16:48:47 i'm afraid maintainer is somehwere away Aug 17 16:48:53 and it's needed for 15.05 release Aug 17 17:19:28 jow_laptop: how should the setfilehandler() calls work for non-encoded input? I think the sink = filecb here: https://github.com/openwrt/luci/blob/master/modules/luci-base/luasrc/http/protocol.lua#L561-L569 isn't correct. A normal filehandler is cb(meta, chunk, eof) but sink is snk(chunk, eof) Aug 17 17:20:24 so if i have a filehandler registered, and I upload multiapart/form, I get a meta, and the data in chunk, but if I upload text/plain directly, I get the actual content in the "meta" argument of the filehandler, instead of in chunk. Aug 17 17:38:25 this makes it work more consistently, but I'm not sure what would be best to put in the meta object for that case? http://paste.fedoraproject.org/255941/14398330/ Aug 17 17:45:40 though something happens to not get the eof passed through at the right time there or something. Aug 17 17:50:22 I guess oassing the msg itself as meta field would be okay Aug 17 17:50:29 or msg.env Aug 17 17:52:23 anything you can pass by reference or the nil value Aug 17 17:52:38 constructing a new table with each invocation will waste a little bit of memory Aug 17 17:53:15 as for the eof, lets see Aug 17 17:54:42 karlp: your wrapper around the filecb does not propagate the return value Aug 17 17:55:20 karlp: the source->sink loop is only continued if the callback returns true Aug 17 18:44:52 hm, I was seeing it called twice, but with eof set to true both times or something, was only copying the existing filehandler type behaviour in failsafe/system/mini that don't return anything. will look at it again tomorrow morning, thanks for looking at. Aug 17 18:59:50 hauke r46679 branches/chaos_calmer/package/system/procd/Makefile * procd: fix seccomp build on some architectures Aug 17 19:02:06 hauke r46680 branches/ (12 files in 5 dirs) * CC: kernel: update 3.18 to 3.18.20 Aug 17 19:24:22 hauke r46681 branches/chaos_calmer/tools/xz/Makefile * xz: update to version 5.2.1 Aug 17 19:40:54 This doesn't look right: https://dev.openwrt.org/changeset/45687 was pptpd deleted on purpose? Aug 17 20:02:53 netprince: it says its on github and its still there Aug 17 20:03:12 https://github.com/openwrt/packages/tree/master/net/pptpd Aug 17 20:21:14 build #77 of x86.64 is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/x86.64/builds/77 Aug 17 21:20:23 hauke r46682 trunk/package/kernel/mac80211/Makefile * mac80211: add mt7601u driver Aug 18 01:12:14 build #73 of mpc83xx is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/mpc83xx/builds/73 Aug 18 01:33:03 build #73 of ep93xx is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/ep93xx/builds/73 **** ENDING LOGGING AT Tue Aug 18 02:59:58 2015