**** BEGIN LOGGING AT Mon Aug 09 02:59:57 2010 Aug 09 03:29:40 build #68 of au1000 is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/au1000/builds/68 Aug 09 04:22:36 agb * r22552 /trunk/tools/ (Makefile qemu/ qemu/Makefile): [tools] qemu: build qemu-img for packaging vmdk and vdi images Aug 09 04:22:41 agb * r22553 /trunk/target/linux/x86/image/ (Config.in Makefile): Aug 09 04:22:41 [x86] image: use internal qemu-img for vmdk and vdi images Aug 09 04:22:41 drop host dependencies on qemu-utils and VirtualBox Aug 09 04:44:36 agb * r22554 /branches/backfire/ (16 files in 16 dirs): [backfire] kernel: bump 2.6.32 targets to 2.6.32.16 Aug 09 04:44:40 agb * r22555 /branches/backfire/tools/ (Makefile qemu/ qemu/Makefile): Aug 09 04:44:40 [backfire] merge r22552 Aug 09 04:44:40 [tools] qemu: build qemu-img for packaging vmdk and vdi images Aug 09 04:44:44 agb * r22556 /branches/backfire/target/linux/x86/image/ (Config.in Makefile): Aug 09 04:44:44 [backfire] merge r22553 Aug 09 04:44:44 [x86] image: use internal qemu-img for vmdk and vdi images Aug 09 04:44:44 drop host dependencies on qemu-utils and VirtualBox Aug 09 05:45:27 build #66 of ramips is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/ramips/builds/66 Aug 09 07:03:32 build #66 of ps3 is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/ps3/builds/66 Aug 09 08:06:51 _trine: http://openwrt.pastebin.com/raw.php?i=Q0BeqiPV is a start Aug 09 08:09:22 <_trine> hey great Aug 09 08:10:52 <_trine> cherokee is really fast I am using it on my dockstar Aug 09 08:11:12 <_trine> and has minimal configuration needs Aug 09 09:12:06 hi all Aug 09 09:12:28 I need to add the USB subsystem to the ramips platform, how can that be enabled? Aug 09 09:26:43 build #66 of s3c24xx is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/s3c24xx/builds/66 Aug 09 09:34:21 lfcorreia: find a gpl tarball that has an usb driver for ramips, port it to openwrt ;) Aug 09 09:41:25 lfcorreia: also note that accidentially included, non gpl'd/proprietary licensed drivers are not usable Aug 09 09:41:54 KanjiMonster: all I need now is enabling the USB thing for ramips Aug 09 09:42:27 i've already managed to compile that driver for openwrt, but it doesn't work due to missing usb subsystem Aug 09 09:42:55 ah I see Aug 09 09:43:46 and the driver's license notice is somewhat cryptic, I for one, wouldn't say it is GPL... Aug 09 09:43:56 yes Aug 09 09:44:01 this driver is present in FON's svn tree Aug 09 09:44:09 oh, okay. Aug 09 09:44:36 but since FON uses an older kernel, I had to tweak some stuff to make it work Aug 09 09:44:55 and again, for now I want to have USB subsys to test driver functionality Aug 09 09:45:08 then I'll worry about making it a true platform driver Aug 09 09:45:19 so any hel is welcomed :) Aug 09 09:45:48 i meant help Aug 09 09:46:22 FYI, this is the driver: http://svn.fonosfera.org/fon-ng/trunk/fon/ra_usb/src/ Aug 09 09:47:10 lfcorreia: well, the usb subsystem is enabled for the ramips targets, so in theory it should work. Aug 09 09:47:20 In what way does it fail? Aug 09 09:47:38 missing symbols, all of them related to sub subsys Aug 09 09:48:14 if you select ramips, in menuconfig, device drivers-usb doesn't show anything Aug 09 09:49:52 lfcorreia: USB_SUPPORT Aug 09 09:49:55 grep for that Aug 09 09:50:16 i think Aug 09 09:50:17 1 se Aug 09 09:50:52 unless I've got a huge f*ckup in my build system, it doesn't do what I expect Aug 09 09:50:59 blogic: that's in all configs in current trunk (=y) Aug 09 09:51:03 i don't have my build sys here at work Aug 09 09:57:12 lfcorreia: the driver looks to be written against 2.6.21, perhaps it needs adjustments to compile against 2.6.32/34 Aug 09 09:57:48 I can send you what I've got already, interested? Aug 09 10:00:00 i've changed some functions and it does compile with minor warnings Aug 09 10:00:58 i think I should be able to see some usb related modules in the running board, and I don't Aug 09 10:44:09 also about this, I still fail to understand the basic stuff related to platform drivers, but I'll have to learn into that Aug 09 10:48:40 xMff, ping. Aug 09 10:53:34 lfcorreia: platform drivers need a corresponding platform device, else they never get called (usually the system setup code is responsible for registering the platform devices) Aug 09 11:03:20 build #66 of ar7 is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/ar7/builds/66 Aug 09 11:14:23 good morning Aug 09 11:20:52 build #62 of x86 is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/x86/builds/62 Aug 09 11:27:16 kaloz * r22557 /trunk/target/linux/cns3xxx/patches-2.6.31/210-cns3xxx_smp_serial_fix.patch: [cns3xxx]: workaround serial hang with SMP enabled Aug 09 11:36:11 RealOpty: pong Aug 09 11:36:40 jow_laptop, im working with quilt. think i figured out what i was gonna ask u Aug 09 11:36:49 ok Aug 09 11:37:18 jow_laptop, but Aug 09 11:37:23 make package/example/update V=99 Aug 09 11:37:37 ^ is that not necessary ? Aug 09 11:37:59 it always tells me that there is no rule for update on package/broadcom-diag Aug 09 11:38:19 "no rule for update" means that there is nothing to update Aug 09 11:38:37 e.g. no prepared source tree with quilt series in build_dir Aug 09 11:38:54 after doing quilt changes, the update target will only work once Aug 09 11:39:08 jow_laptop, make[2]: *** No rule to make target `update'. Stop. Aug 09 11:39:13 thats the exact error. Aug 09 11:40:30 doing $ make package/broadcom-diag/{clean,compile} V=99 Aug 09 11:40:42 seems to make the edited patch apply to build dir. Aug 09 11:40:50 so i guess no worries. Aug 09 11:42:57 jow_laptop, just to make sure of something.... Aug 09 11:43:20 backfire/package/broadcom-diag is totally diffrent than trunk/package/broadcom-diag Aug 09 11:43:24 jow_laptop, correct? Aug 09 11:43:34 likely Aug 09 11:46:39 jow_laptop, ok so ive uploaded the patch to https://dev.openwrt.org/ticket/7702 Aug 09 11:46:58 should i reopen the ticket and maybe hauke will notice ive done the grunt work for him? Aug 09 11:48:29 I suppose you can reopen it Aug 09 11:51:52 jow_laptop, night. Aug 09 12:15:58 cant sleep. Aug 09 12:16:21 jow_laptop, how would i go about making a patch for target/linux/brcm47xx/files/etc/etc/ Aug 09 12:17:15 jow_laptop, make target/linux/brcm47xx/{clean,prepare} V=99 QUILT=1 fails Aug 09 12:17:50 and thats step one :( Aug 09 12:18:44 oh Aug 09 12:18:47 maybe i need to do Aug 09 12:18:48 make target/linux/{clean,prepare} V=99 QUILT=1 Aug 09 12:56:27 hi again Aug 09 12:56:37 getting back to the ramips usb issue Aug 09 12:57:29 after selecting the proper settings for the ramips (rt3052) board, and visiting Device Drivers-USB in 'make menuconfig', nothing can be selected in that menu Aug 09 12:57:33 please advise Aug 09 13:09:43 lfcorreia: where are you, the kernel menuconfig or buildroots menuconfig? Aug 09 13:16:46 buildroots menuconfig Aug 09 13:16:59 my coleague Helmut has provided a hint Aug 09 13:18:24 /linux-2.6.34.../drivers/usb/Kconfig, add default y if RALINK_RT305X Aug 09 13:18:46 as I said, the default kernel config file for RAMIPS does not enable USB Aug 09 13:18:51 ah, I found why Aug 09 13:20:25 lfcorreia: try removing the three "# CONFIG_USB_ARCH_HAS_* is not set" lines from the target/ramips/rt305x/config- Aug 09 13:20:41 at the end of the config file Aug 09 13:20:47 KanjiMonster: remove them all? Aug 09 13:21:00 or change with "=y" Aug 09 13:21:12 remove them, they overwrite the default config which is =y Aug 09 13:25:06 acoul * r22558 /packages/net/tinyproxy/ (Makefile patches/010-no-docs-and-tests.patch): net/tinyproxy: update tinyproxy to 1.8.2 Aug 09 13:26:02 mb * r22559 /trunk/package/kexec-tools/ (Makefile patches/0005-mips64_support.patch): Aug 09 13:26:02 kexec-tools: Update to 2.0.2. Aug 09 13:26:02 Mips64 support is builtin in this version. So remove the patch. Aug 09 13:28:30 KanjiMonster: thanks, will try and then post here the results Aug 09 13:28:59 this driver would eventually be made part of the platform drivers for the ramips platform Aug 09 13:29:09 assuming it is working at all Aug 09 13:29:50 and while I personally won't bother much with GPLness, it would be great if we could figure out the proper licensing Aug 09 13:30:23 lfcorreia: as far as I can see, this is a BSD-style license, perhaps even the gpl-compatible one Aug 09 13:31:25 thar license isn't clearly nor BSD of GPL Aug 09 13:31:57 all they ask if for the copyright notice to be present Aug 09 13:32:03 if=is Aug 09 13:35:51 I said bsd-style, not /is/ bsd ;). Going from what they allow you to do with the source. Aug 09 13:40:16 IANAL, but requiring the copyright notice in the source code shouldn't be a further restriction to the gpl and therefore the license would be gpl compatible - otoh, this is probably not an OSI-approved license. Aug 09 14:06:59 IANAL too :) Aug 09 14:07:44 _if_ the driver works, and _if_ it is really working right, then I'll analise whatever i've changed Aug 09 14:22:58 jow_laptop, u around? Aug 09 14:23:06 yea Aug 09 14:24:54 lol every time i figure it out some how.... Aug 09 14:26:28 quilt stuff is confusing when updating edited patches Aug 09 14:47:42 jow_laptop, do you know what 'bcm_strtoul' is? Aug 09 15:01:31 C string to unsingled long Aug 09 15:01:36 *unsigned Aug 09 15:43:44 jow_laptop, now im getting this error... Patch platform/999-fix_board_data does not apply (enforce with -f) Aug 09 15:43:51 any suggestions? Aug 09 15:45:06 quilt push -f Aug 09 15:45:14 then resolv the conflicts manually Aug 09 15:50:22 jow * r22560 /trunk/target/linux/generic/patches-2.6.35/001-fix_localversion.patch: [generic] fix localversion detection for linux 2.6.35 Aug 09 15:53:38 jow_laptop, plz take a min to run me though the commands for ;editing a patch; Aug 09 15:53:45 the files im editing are in backfire/target/linux/brcm-2.4 Aug 09 15:54:28 im reading the wiki. doin all it says but it still not accepting new changes? Aug 09 15:55:13 what did you do so far? Aug 09 15:55:54 well i made a patch. every thing worked out then i needed to make changes so i did. then that one applied like it should Aug 09 15:56:37 now i do make target/linux/{clean,prepare} V=99 QUILT=1 Aug 09 15:57:52 then go into the folder i want to edit Aug 09 15:58:05 quilt push 999 Aug 09 15:58:38 but it says File series fully applied, ends at patch 999-fix_board_data Aug 09 15:58:54 quilt top Aug 09 15:59:08 is it the patch you want to edit? Aug 09 15:59:22 ~/OpenWRT/netgear/backfire/target/linux/brcm-2.4$ quilt top Aug 09 15:59:22 999-fix_board_data Aug 09 15:59:22 yes Aug 09 15:59:42 then just edit... quilt edit foo/bar.c Aug 09 15:59:54 then i did that Aug 09 16:00:02 then i looked at the diff Aug 09 16:00:05 then i refreshed Aug 09 16:00:28 then i do make target/linux/{clean,prepare} V=99 QUILT= Aug 09 16:00:31 correct? Aug 09 16:00:34 no Aug 09 16:00:35 +1 Aug 09 16:00:38 oh Aug 09 16:00:51 make target/linux/{**update**,clean,prepare} Aug 09 16:01:04 otherwise you're just wiping everything you just did Aug 09 16:01:24 update or **update** Aug 09 16:01:38 update Aug 09 16:02:12 okie Aug 09 16:03:07 jow_laptop, http://paste.ubuntu.com/475504/ Aug 09 16:03:12 same errors :\ Aug 09 16:03:51 did you edit the file initially before adding it to the patch? Aug 09 16:04:06 no it was fresh svn Aug 09 16:04:18 im trying to make patches the correct way Aug 09 16:05:36 it looks like your patches are processed in the wrong order Aug 09 16:08:39 * RealOpty starts over. Aug 09 16:08:48 and give it a .patch extension Aug 09 16:09:09 also make sure to create the patch with quilt new platform/999-foo.patch Aug 09 16:09:26 not just the name Aug 09 16:09:40 kernel uses an additional subdir Aug 09 16:12:01 jow_laptop, are patches applied to the file your editing? Aug 09 16:12:11 ? Aug 09 16:12:13 yes Aug 09 16:12:18 no Aug 09 16:12:23 ok cause Aug 09 16:12:28 the question makes no sense Aug 09 16:12:39 lol sorry im tired Aug 09 16:12:54 ok so im editing backfire/target/linux/brcm-2.4/files/drivers/mtd/maps/filename.c with quilt Aug 09 16:13:05 not exactly like that... but u know Aug 09 16:13:26 anyways the patch i made in quilt is applied to backfire/target/linux/brcm-2.4/files/drivers/mtd/maps/file.c Aug 09 16:13:34 is that what should happen? Aug 09 16:14:02 * RealOpty dont think so Aug 09 16:14:10 uhm Aug 09 16:14:16 I don't get it Aug 09 16:14:30 i believe ive messed up some where with the commands. Aug 09 16:14:35 ah you mean the path is wrong? Aug 09 16:15:13 ok let me try to explain again :) Aug 09 16:15:36 I believe the file you're trying to edit is placed in target/linux/brcm-2.4/files/... Aug 09 16:15:38 when i make a patch its saved to my+patch_name.patch Aug 09 16:15:40 you do not need a patch for this Aug 09 16:15:48 you can edit it directly Aug 09 16:16:03 doh!? Aug 09 16:16:09 patches are for stuff that comes from the linux tarball Aug 09 16:16:12 here i was trying to do it the correct way lol! Aug 09 16:16:43 im just gonna role with that cause it would be a lot easier atm lol Aug 09 16:16:51 jow_laptop, how do i undo the patch Aug 09 16:16:55 quilt delete ? Aug 09 16:17:08 yes Aug 09 16:20:10 hmmm Aug 09 16:20:18 somehow this patch got into backfire/target/linux/brcm-2.4/patches Aug 09 16:20:26 i think thats what caused it all... Aug 09 16:23:34 jow_laptop, thanks Aug 09 16:27:13 i was wondering why making a patch for a package was easier than this... lol Aug 09 16:27:57 build #63 of uml is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/uml/builds/63 Aug 09 16:47:37 Hauke, ping. Aug 09 16:52:13 Hauke, when ever you have time, ive added support in broadcom-diag for backfire https://dev.openwrt.org/ticket/7702 Aug 09 16:53:44 hi there Aug 09 16:54:33 I have submitted some patches to support a new board but it doesn't seems to have been taken in account Aug 09 16:55:00 https://dev.openwrt.org/ticket/7672 Aug 09 16:55:59 nico * r22561 /trunk/package/comgt/ (Makefile files/3g.sh): package/comgt: only set pincode if it's supplied (closes: #7334), bump release number Aug 09 16:59:30 nico * r22562 /branches/backfire/package/comgt/ (Makefile files/3g.sh): Aug 09 16:59:30 [backfire] merge r22561 Aug 09 16:59:30 package/comgt: only set pincode if it's supplied (closes: #7334), bump release number Aug 09 16:59:54 clem-vangelis, wrote any patches for a 2.4 kernel? Aug 09 17:02:22 RealOpty, im not a developer but your patch here https://dev.openwrt.org/attachment/ticket/7672/bcm47xx.patch just might break functionality on every device. Aug 09 17:04:03 humi have written it to support every device Aug 09 17:04:27 oups you are right i submit the wrong patch Aug 09 17:04:43 EDIMAX_PS_HEADER_LEN <- need some kind of if statement for that under lines 373,375,376 Aug 09 17:06:19 clem-vangelis, but seriously any patches for 2.4? Aug 09 17:06:26 I am annoyed with just the guys from memcached. they claim that the openwrt compile does not support standards. could somebody have a look at http://code.google.com/p/memcached/issues/detail?id=123 and help me to get in buisness again? Aug 09 17:06:29 no why ? Aug 09 17:06:37 * RealOpty needs some advice. Aug 09 17:07:23 hmm and a smoke too Aug 09 17:08:37 jow_laptop, how do i know what headers that i can include in an app? Aug 09 17:09:02 not just jow_laptop any one for that matter :) Aug 09 17:30:40 heilterminalcons: yeah, the issues posted in there clearly have nothing to do with the compiler Aug 09 17:31:25 heilterminalcons: the configure error comes from autoconf scripts trying to do runtime tests (as opposed to just compile tests), which in general should be avoided at all costs Aug 09 17:31:31 because they cannot possibly work with cross compiling Aug 09 17:31:50 as for the other build error: no idea how they normally define the endian type for the hash stuff Aug 09 17:31:59 probably build system screwup as well Aug 09 17:54:22 nbd: i can remove the exit 1; from the autoconf but this results in other errors Aug 09 18:07:16 memcached up to 1.2.8 builds without any problems Aug 09 18:43:22 mb * r22563 /trunk/target/linux/omap24xx/patches-2.6.35/811-mmc-multiplexer-fixes.patch: omap24xx: Fix MMC multiplexer Aug 09 18:43:45 mb * r22564 /trunk/target/linux/omap24xx/ (config-2.6.35 patches-2.6.35/500-cbus.patch): omap24xx: Update kernel config. This also fixes module loading. Aug 09 18:50:21 l2tpv3 and kvm guest works like a charm thanks xMff Aug 09 18:51:16 great job Aug 09 19:25:25 florian * r22565 /branches/backfire/target/linux/rdc/ (config-2.6.32 files-2.6.32/ patches-2.6.32/): remove rdc 2.6.32 support from backfire, not used Aug 09 19:31:43 florian * r22566 /branches/backfire/target/linux/rdc/config-2.6.30: backport r22259 to backfire Aug 09 19:50:34 nico * r22567 /packages/net/l2tpv3tun/Makefile: [packages] l2tpv3tun: mark as 2.6.35 only (prevent from trying to build it on backfire) Aug 09 19:58:21 <_trine> looks like rtorrent isnt compiling Aug 09 20:00:34 <{Nico}> _trine: which target ? Aug 09 20:00:56 <_trine> kirkwood Aug 09 20:01:17 captain kirkwood please ;o) Aug 09 20:01:27 <{Nico}> hehe Aug 09 20:01:35 <_trine> :) beam me down Aug 09 20:05:42 <{Nico}> _trine: do you have a build log somewhere ? Aug 09 20:06:20 <_trine> {Nico}, not now I just un selected it and rebuilt Aug 09 20:06:37 <_trine> sorry I should have thought about that Aug 09 20:07:03 florian * r22568 /trunk/target/linux/au1000/ (6 files in 3 dirs): [au1000] add support for 2.6.35 Aug 09 20:07:16 <_trine> I could reselect it and do it again Aug 09 20:07:32 <{Nico}> _trine: are you using a recent packages feed ? Aug 09 20:08:29 <_trine> before I compiled I did svn up and ./scripts/feeds update -a and ./scripts/feeds install -a Aug 09 20:12:26 <{Nico}> _trine: ( cd feeds/packages; svn info | grep 'Last Changed Rev') Aug 09 20:13:05 <_trine> Last Changed Rev: 22567 Aug 09 21:43:11 <{Nico}> _trine: can't reproduce your issue, i definitely need a build log ;) Aug 09 21:52:19 KanjiMonster, hi Aug 09 21:52:40 lfcorreia: hi Aug 09 21:52:53 KanjiMonster, none of the proposed edits made usb selectable in make menuconfig Aug 09 21:53:23 as a comparison, bcm4700 does have it enabled, but I couldn't figure out how ot why Aug 09 21:53:51 and yes, i'm a dumb f*ck regarding to owrt confusing build system, sorry Aug 09 21:54:55 (confusing because each target has different configs) Aug 09 22:00:01 lfcorreia: mh, try deleting the tmp of your buildroot directory, it seems it caches the parsed config there - this should force it to reparse it and make usb selectable for rt305x (at least that worked here) Aug 09 22:00:27 i did a full svn checouy as well Aug 09 22:00:33 svn co Aug 09 22:00:41 but will do that again Aug 09 22:01:07 you did leave the config_usb_support=y in? Aug 09 22:05:05 ok, my bad Aug 09 22:05:33 i guess I ran make menuconfig BEFORE editing the config file and that file was indeed cached Aug 09 22:06:05 great. and don't be so hard on yourself, the openwrt build system can be confusing at times Aug 09 22:06:23 especially all the magic stuff happening in the background Aug 09 22:07:05 KanjiMonster, do you have commit rights on svn? Aug 09 22:07:14 nope Aug 09 22:07:56 ok, i'll propose a patch later (perhaps MUCH later) Aug 09 22:08:09 KanjiMonster, and thanks for putting up with me :) Aug 09 22:08:22 you're welcome :) Aug 09 22:23:39 <_trine> {Nico}, Aug 09 22:23:45 <_trine> are you stil here Aug 09 22:33:17 <{Nico}> _trine: yep Aug 09 22:37:15 <_trine> I have reproduced it Aug 09 22:37:23 <_trine> what do you want me to do Aug 09 23:13:06 <{Nico}> _trine: give me a build log :) Aug 09 23:13:12 <{Nico}> make using V=99 Aug 09 23:16:59 <_trine> {Nico}, I did Aug 09 23:22:05 <_trine> {Nico}, http://pastebin.com/g1pQ6LMX Aug 10 00:32:52 RealOpty: if you're back, please try this simple patch: http://luci.subsignal.org/~jow/998-pci_delay.patch and put it into target/linux/brcm47xx/patches-2.6.34/ Aug 10 00:35:05 RealOpty: then check whether pci is detected Aug 10 02:47:42 xMff, i was about to go to bed but this should be exciting to find out **** ENDING LOGGING AT Tue Aug 10 02:59:58 2010