**** BEGIN LOGGING AT Thu Feb 18 02:59:59 2016 Feb 18 08:22:09 blogic r48738 trunk/package/network/services/lldpd/Makefile * lldp: Upgrade to 0.9.0 Feb 18 08:22:19 blogic r48739 trunk/toolchain/ uClibc/config/archs gcc/common.mk uClibc/common.mk * toolchain: add support of ARCv2 architecture Feb 18 08:22:34 blogic r48740 trunk/ (28 files in 12 dirs) * linux: add support of Synopsys ARCHS38-based boards Feb 18 08:35:23 blogic r48741 trunk/target/linux/archs38/base-files.mk * archs38: remove network config override Feb 18 12:09:37 jow r48742 trunk/target/linux/ar71xx/config-4.4 * ar71xx: add missing symbol to Linux 4.4 config (#21878) Feb 18 12:22:37 o_O ar71xx 4.4, must have missed that Feb 18 12:26:22 is there any trick to getting dynamic_debug to work? it works for ip_tables, but I can't seem to get anything form usb_storage, or scsi modules? Feb 18 12:28:53 oh, hrm, greping the dynamic control file, there just isn't any debug to turn on there. Feb 18 16:15:49 How does CONFIG_DEBUG work? I made an image with it and still got binaries without debugging symbols Feb 18 16:16:54 it adds -g to the build flags but still strip binaries Feb 18 16:17:06 usually there is no need to run unstripped debug executables on the target Feb 18 16:17:26 it is better to use gdbserver for that Feb 18 16:17:40 right. OK, thanks. Feb 18 16:18:05 if you really want all unstripped you need to dsiable CONFIG_STRIP as well Feb 18 16:18:37 I see Feb 18 16:24:11 hi Feb 18 16:25:01 will the new EU regulation outlaw the usage of openwrt on wifi capable devices? Feb 18 16:47:36 this a german article about it: http://www.heise.de/newsticker/meldung/Funkregulierung-TP-Link-muss-WLAN-Firmware-sperren-3109847.html Feb 18 16:48:20 https://www.reddit.com/r/linux/comments/46dy4q/tplink_has_started_locking_down_firmware_and/ Feb 18 16:48:48 “Verification by radio equipment of the compliance of its combination with software should not be abused in order to prevent its use with software provided by independent parties.” (point 19). Feb 18 16:49:08 https://blog.tohojo.dk/2015/10/the-new-wifi-regulations-in-europe.html Feb 18 16:52:49 Verification by radio equipment of the compliance of its combination with software should not be abused - this is already an abuse on it's own as one has to pay to get closed source crap/or not modifyable compiled open source code verified. measuring the quality of current router operating systems delivered by companies, nearly all routers are exploitable 1 year after the release and most manufactures Feb 18 16:52:55 provide updates never or very late in time Feb 18 16:52:56 @msgctl Feb 18 16:54:37 they will use tivoization, hopefully they use GPL3 code which outlaws this "tradition" Feb 18 17:06:38 puchu: Oh well, EU still has stronger free software lobby than US. Pirate Party MEPs are already working on this. Feb 18 17:23:58 hmm, tried 4.4 on ar71xx / rspro -> boot stops right after "MIPS: machine is Ubiquiti RouterStation Pro" Feb 18 17:36:17 is it a new rspro stintel ? Feb 18 17:36:27 ie, freshly flashed from factory firmware ? Feb 18 17:36:51 what I've discovered with picstations here, that come from factory with 5.6 firmware, uboot now leaves the flash chip locked Feb 18 17:36:57 and older builds dont unlock it Feb 18 17:39:01 groz: it's a 6y old rspro Feb 18 17:39:32 ok Feb 18 17:40:02 it was running DD r48648 until I flashed it with r48742 half an hour ago Feb 18 17:40:19 I've got a couple here that I flashed early last week, fresh from new, picostations Feb 18 17:40:31 didn't check them carefully, because I was in a hurry, seemed to be running fine Feb 18 17:40:38 and the bucket truck was in place, so, up they went Feb 18 17:40:52 Now they are inaccessible without a truck, and, I've realized the flash is read only Feb 18 17:41:02 ah bummer :( Feb 18 17:41:08 so just trying to build a kernel module that'll unlock the flash Feb 18 17:41:17 and I forgot how to flash the RSPro from the recovery menu Feb 18 17:41:17 * groz learning all about the spi subsystem in this process Feb 18 17:41:26 or I didn't, but it no longer works - atftp put timeout Feb 18 17:41:28 ctrl c into uboot Feb 18 17:41:46 if it's like the others, urescue at the uboot prompt will do it Feb 18 17:41:48 yeah did that, pings on 192.168.1.20 Feb 18 17:41:51 ah Feb 18 17:41:57 but it's not uboot it's redboot Feb 18 17:42:00 then tftp binary Feb 18 17:42:02 ohhhh Feb 18 17:42:05 hmmz Feb 18 17:42:12 * groz hasn't seen redboot in years, forget how that works Feb 18 17:42:20 currently connected over one of my apu2's :) Feb 18 17:42:42 I would leave it like this, but I need 3 LAN ports here and I don't have a switch :( Feb 18 17:42:48 the RSPro is perfect for that Feb 18 17:42:54 also I don't have wifi in the apu2 :( Feb 18 17:42:55 meh Feb 18 17:42:59 dont have a switch ? Feb 18 17:43:17 you would do a double take then looking at how I did things here at our new house after we bought it Feb 18 17:43:47 house is on a 4' crawlspace, so, I've got a 24 port switch hanging from the joists down there, and ran ethernet to every room in the house Feb 18 17:44:15 don't have a switch here, no Feb 18 17:44:17 I've been using wifi for the link out to the back lot for a couple years, but last fall when the creek was dry Feb 18 17:44:29 but my situation is ... complicated :P Feb 18 17:44:36 we brought in a ditchwitch, and put in a trench for a water pipe Feb 18 17:44:46 while it was open, I put in conduits for power and data too Feb 18 17:44:54 I moved to Bulgaria, my ERLite's are there, routing my 50Mbps fiber uplink in a HA setup Feb 18 17:44:57 but it's a 450 foot run, so the data conduit getsa fiber Feb 18 17:45:20 fiber not in the conduit yet, but, scheduled 'real soon now' Feb 18 17:45:21 originally from Belgium, have an office at my friends' house in Belgium Feb 18 17:45:36 with my old PC and repurposed RSPro Feb 18 17:45:38 but no switch here etc Feb 18 17:46:46 in my case, with the picos, I rushed because the neighbor had a bucket truck in for another job Feb 18 17:46:50 uplinks here are a bit faster, so I'm going to eventually replace the rspro with the apu2's Feb 18 17:46:56 and I grabbed the chance to get the pico way up top of the pole Feb 18 17:47:05 but I need a switch here + solution for wifi :) Feb 18 17:50:29 oh, it's also urescue Feb 18 17:50:35 thanks Feb 18 17:50:45 pffft that flash is so damn slow Feb 18 17:53:13 did a factory flash now and still freeze Feb 18 17:55:44 what I've found with the pico Feb 18 17:56:03 flashing factory, most recent version, thru urescue, creates a device that wont boot Feb 18 17:56:15 it says 'loading ', then about 5 seconds later, resets Feb 18 17:56:48 ubiquiti firmware is starting to implement locks and checks Feb 18 17:56:55 use an older version, and it'll probably work Feb 18 17:57:45 I think the problem is with the 4.4 kernel Feb 18 17:57:54 the target is still on 4.1 in trunk Feb 18 17:58:14 I just noticed patches for 4.4 were added a few weeks ago and thought I'd give the 4.4 kernel a try Feb 18 18:08:16 when I realized my pico was in trouble, did some checking, still have one left here on the desk Feb 18 18:08:27 with he 4.1 build, problem is addressed Feb 18 18:08:44 but the one up on the pole has an older 3.18 build I had kickigng around ready to go when I needed it quickly Feb 18 18:09:10 with that one, flash subsystem is not unlocking the flash chip, which uboot now leaves locked on the gadget Feb 18 18:09:29 but the older uboot with the 5.5.1 firmware, leaves flash unlocked when the device starts Feb 18 18:09:48 so, my trick now is, can I get a kernel module in place that talks directly to the unlock registers on the flash chip Feb 18 18:09:57 then leaves it in a state I can run a sysupgrade Feb 18 18:55:38 nbd r48743 trunk/package/kernel/mac80211/ patches/322-mac80211-add-A-MSDU-tx-support.patch patches/335-mac80211-minstrel_ht-set-A-MSDU-tx-limits-based-on-s.patch * mac80211: update A-MSDU tx support to the latest version Feb 18 18:55:48 nbd r48744 trunk/package/kernel/mac80211/patches/336-mac80211-minstrel_ht-set-default-tx-aggregation-time.patch * mac80211: set default aggregation timeout to 0 to improve stability Feb 19 00:57:27 jow_lapt1p: I've updated my jsonpath librafication some more, to remove the libubox dep from the library portion of it, but I'd still appreciate your feedback on how the tokens and errors are handled. https://github.com/karlp/jsonpath/pull/1 particularly whether things in parser.h need to be exposed to the user. **** ENDING LOGGING AT Fri Feb 19 02:59:58 2016