**** BEGIN LOGGING AT Tue Sep 03 02:59:58 2013 Sep 03 05:53:47 build #358 of cobalt is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/cobalt/builds/358 Sep 03 06:30:26 cyrus r37891 trunk/include/netfilter.mk * Fix IPv6 NAT breaking older kernels Sep 03 06:47:15 build #370 of orion is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/370 Sep 03 07:23:06 juhosg r37892 trunk/target/linux/ar71xx/image/Makefile * ar71xx: fix NetGear initramfs image names Sep 03 09:03:40 is the only reason fpu emulation is turned off by default for space reasons? (as indicated by florian in this patch? http://openwrt-for-embedded.googlecode.com/svn-history/r203/trunk/target/linux/generic-2.6/patches-2.6.36/025-mips_disable_fpu.patch Sep 03 09:03:57 was just having a go at building luajit on ar71xx, but it wants mips with fpu, Sep 03 09:12:54 karlp: tried that in the past already - unusable Sep 03 09:13:32 not sure if something significantly changed recently but the last try on ar71xx just resulted in 100% cpu consuming luajit process that didn't do anything Sep 03 09:16:22 fair enough then, I just moved up from old atheros to ar71xx, thought I'd give it another spin. Sep 03 09:36:05 karlp: soft fp is much faster than fpu emulation; and with soft fp there is no need for fpu emulation to be enabled anymore (so it was disabled), saves both kernel space as well as should improve task switching performance a little bit (as the virtual fpu registers don't need to be (re-)stored) Sep 03 09:37:53 yeah, but if luajit is 10 times faster and fpu emulation is 30% slower, it's still winning. Sep 03 09:38:20 soft-fp is fine for anything else, sure. Sep 03 09:38:42 not going to care much, the new ar71xx is so much faster than the old ar2317 anyway :) Sep 03 09:51:42 [florian]: What means this: #define BCM_6348_PERF_BASE Sep 03 09:52:17 found in bcm63xx_cpu.h Sep 03 10:04:21 [florian]: you have defined, for the bcm3368, BCM_3368_MPI_BASE (0xfff80000), but in a source code i've found yesterday that address belongs to EBU, while MPI is defined in 0xfff80100 Sep 03 10:04:41 is defined at* Sep 03 11:24:37 build #338 of rb532 is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/rb532/builds/338 Sep 03 11:27:04 build #338 of ppc44x is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/ppc44x/builds/338 Sep 03 11:27:54 guys... Sep 03 11:28:10 why does luci-app-upnp disables miniupnpd on boot? Sep 03 11:28:44 ah, is that the reason ? Sep 03 11:28:51 security reasons Sep 03 11:29:00 the idea was that the user has to explicitely enable it in the gui Sep 03 11:29:17 and the reason why it was implemented in that way is because there is no disable option in uci Sep 03 11:29:45 if the upnp uci config gains an option enabled 0/1 I can remove the init script fiddling Sep 03 11:31:27 so even if: /etc/init.d/miniupnpd enabled && echo yes Sep 03 11:31:28 yes Sep 03 11:31:41 that fiddling causes miniupnpd to not start at boot, right ? Sep 03 11:32:37 how can i delete a file when building custom image? Sep 03 11:33:12 i want to wipe the whole /etc/uci-defaults thing Sep 03 11:33:27 there is no facility to do that Sep 03 11:33:59 you can make a 00_disable script in uci-defaults which simply deletes the other files Sep 03 11:34:37 i'll add empty scripts then Sep 03 11:51:13 my wndr3700 died yesterday, had to buy wndr3800. i don't even need to rebuild firmware! Sep 03 11:58:54 please remind me, what had to be done after "scripts/feeds update -a" ? Sep 03 12:17:11 build #351 of uml is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/uml/builds/351 Sep 03 12:19:57 build #319 of sibyte is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/sibyte/builds/319 Sep 03 12:39:27 mazilo: ping Sep 03 12:39:50 strasidlo: pong. Sep 03 12:39:57 strasidlo: let's go to PM. Sep 03 14:21:06 build #329 of avr32 is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/329 Sep 03 15:47:18 jow r37893 trunk/package/ utils/usbreset/Makefile utils/usbreset/src/usbreset.c * usbreset: rewrite to not rely on /proc/bus/usb anymore Sep 03 17:26:32 jow_laptop: Hi Jo-Philipp, can you please have a look at Patchwork 3963 when you have a moment? Thanks. Sep 03 17:33:56 can anyone tell me if there is a way to auto mount a USB drive *without* using option uuid in /etc/config/fstab Sep 03 17:34:02 I'm doing extroot Sep 03 18:06:42 build #279 of iop32x is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/iop32x/builds/279 Sep 03 19:29:01 froek: any reason you can't use: option device "/dev/sdXn" Sep 03 19:32:10 philipp64: I tried.. but it never recognizes the drive.. extroot fails to mount that way Sep 03 19:33:51 as soon as I switch to option uuid, it works flawlessly Sep 03 19:34:13 i've really like ot see this changed.. because everytime I use a different usb disk, I have to manually change the fstab to recognize the uuid.. :( Sep 03 21:19:39 build #281 of ep93xx is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/ep93xx/builds/281 Sep 03 23:22:56 luka r37894 trunk/package/ boot/kobs-ng/Makefile boot/kobs-ng boot/kobs-ng/patches/001-fix-mtd-defines.patch boot/kobs-ng/patches * [package] kobs-ng: add new package Sep 03 23:40:35 luka hi, sorry to bother you so late, I reviewed my patch and this is the new description with the tips you gave me, it's ready to be send to openwrt mailing list, if it's ok by you Sep 03 23:40:38 here: http://codepad.org/82nRXn2d Sep 03 23:41:05 I tried my best to document it well Sep 03 23:48:15 Devastator: ", yet is not kernel upstream." that part is a bit confusing Sep 03 23:48:49 Devastator: and the rest of your patch? Sep 03 23:49:01 hum, I meant it is not in kernel upstream yet Sep 03 23:49:16 but as you notice, english is not my strong point Sep 03 23:49:39 no problem, but please rephrase it Sep 03 23:50:14 also i'd like to see your patch too, because links [2] and [3] i would like to see bellow --- line Sep 03 23:50:33 http://codepad.org/lPZFT5rt Sep 03 23:51:23 well, that is not a patch for openwrt, right? Sep 03 23:51:31 you plan to send that to mtd list? Sep 03 23:51:42 it's already there Sep 03 23:51:46 no, sorry, you already gave links Sep 03 23:51:47 yes Sep 03 23:51:51 too late here Sep 03 23:52:02 no worries, we can talk tomorrow Sep 03 23:52:04 well, like i said that is not a patch for openwrt Sep 03 23:52:10 we can talk now Sep 03 23:52:35 do you have suggestion for a patch to get accepted in platform/generic? Sep 03 23:52:41 similar to this on Sep 03 23:52:43 one.. Sep 03 23:55:41 well, commit message is ok Sep 03 23:55:50 when you change what i pointed out Sep 03 23:56:08 but that is not a patch i can apply to openwrt Sep 03 23:57:45 I understand.. I see a few routers got support with a flashchip patch that went in platform/generic, I just don't know where/how to patch for it to get applied Sep 03 23:58:10 that patch should go into generic Sep 03 23:58:33 but you should send a patch for openwrt which contains that patch in it Sep 03 23:58:55 hummm, a patch of a patch, I see Sep 03 23:59:07 yes Sep 03 23:59:22 thanks luka! Sep 03 23:59:36 but in the patch should be original commit message that you sent to the mtd list Sep 03 23:59:53 + put links [2] and [3] bellow --- line Sep 04 00:00:46 ok, I'm looking at one already in openwrt, so I can base mine Sep 04 00:05:19 luka below --- above drivers/...? Sep 04 00:07:18 another try with the same commit message I've sent to the mtd list and [2] [3] below ---, here: http://codepad.org/iMYjm49C Sep 04 00:07:27 build #304 of au1000 is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/au1000/builds/304 Sep 04 00:07:43 and thank you for your patience Sep 04 00:13:07 Devastator: http://codepad.org/XE4fTQQo Sep 04 00:13:14 something like that Sep 04 00:14:58 ah! I get it! Sep 04 00:20:16 Devastator: fyi, here http://patchwork.ozlabs.org/patch/272438/ the "changelog v2" should go bellow --- line Sep 04 00:21:22 damn Sep 04 00:21:26 Devastator: also this link is broken http://lists.infradead.org/pipermail/linux-mtd/2013-Sptember/048511.html Sep 04 00:22:28 damn you, keyboard Sep 04 00:22:37 September! Sep 04 00:23:36 well, I will talk to mtd guys again, to see if these mistakes are "fine" or I need a v3 Sep 04 00:24:00 in the mean time I will work on openwrt patch Sep 04 00:28:14 good luck :) Sep 04 00:28:34 thanks again Sep 04 00:29:44 I've learned a lot in the process, almost there! Sep 04 01:10:44 build #313 of kirkwood is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/kirkwood/builds/313 **** ENDING LOGGING AT Wed Sep 04 02:59:58 2013