**** BEGIN LOGGING AT Fri Mar 01 02:59:58 2013 Mar 01 03:55:48 <3JTAANC26> build #203 of orion is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/203 Mar 01 10:08:33 hi Mar 01 10:11:31 what would be a committable approach to get better bsd/macosx support as a host system. Mar 01 10:11:48 wrapper scripts? Mar 01 10:13:38 either that or a better tool discovery Mar 01 10:13:53 like symlinking gcp into staging_dir/host/bin/cp etc. Mar 01 10:14:10 or handling stuff like gstat vs. stat Mar 01 10:14:30 where gstat on bsd refers to gnu stat but on many linux distros it is an admin tool for some database Mar 01 10:20:18 jow_laptop: is symlinking of gcp already happening? Mar 01 10:20:30 it is supposed to afair Mar 01 10:23:58 yes, it is a symlink on my mac. where can I find the mechanism which creates it Mar 01 10:29:00 tools/Makefile, the PrepareCommand macro Mar 01 10:33:13 ah, thx Mar 01 10:36:24 hi wbx Mar 01 10:36:47 my main dev machine is running mac os x, so while there still are some packages that are broken, most of it works just fine Mar 01 10:39:37 python f.e. is broken. Mar 01 10:39:45 but i like to prepare a patch Mar 01 10:40:43 nbd: what do you think about my new openjdk7 submission? Mar 01 10:42:13 last time i looked at python, it wasn't very easy to fix because many places assume that host system == target system Mar 01 10:42:38 where is the openjdk7 submission? haven't seen it on the list Mar 01 10:44:52 nbd: on the mailing list (user i think) Mar 01 10:45:24 nbd: the python patch is not as big as you think. Mar 01 10:47:43 could you please resubmit to -devel, as it's tracked by patchwork Mar 01 10:54:06 nbd: okay Mar 01 10:55:27 nbd: do you remember the openjdk6 patches? now it is more easy to cross-compile Mar 01 10:55:48 nbd: and i added all prereq to tools/ Mar 01 10:56:29 nbd: the uclibc $ORIGIN patch is still required, but it is used in oe since years Mar 01 11:02:16 wbx: hm, what i don't like about that patch is putting a lot of stuff in tools/ that is only used by openjdk Mar 01 11:02:34 it would be better to have this as host-build packages Mar 01 11:02:38 that openjdk can build-depend on Mar 01 11:03:19 nbd: what are host-build packages? the stuff in tools depend on openjdk Mar 01 11:03:56 grep for HostBuild in packages Mar 01 11:07:40 nbd: hmm. i don't understand it. Mar 01 11:09:38 tools/ should be used mostly for core stuff that is used by either many packages or by target stuff Mar 01 11:10:35 all those host tools that are specific to openjdk should go into /packages as well Mar 01 11:11:51 and if you grep for HostBuild, you can find other examples of packages there that build stuff for the host Mar 01 11:12:18 lol, that aiccu thread was awesome, thank you! Mar 01 11:18:55 here, use my service! \\BOOM// oops, don't step there. Mar 01 11:22:49 nbd r35839 branches/attitude_adjustment/target/linux/ cns3xxx/patches-3.3/470-gpio_irq.patch cns3xxx/patches-3.3/300-laguna_support.patch * cns3xxx: backport laguna UART setup crash fix from r33719, r33721 Mar 01 11:34:39 russell--: ... Mar 01 12:35:10 <3JTAANC26> build #167 of ar71xx is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/ar71xx/builds/167 Mar 01 13:46:22 luka r35840 trunk/ (21 files in 12 dirs) * ar71xx: add support for Buffalo WZR 600DHP Mar 01 15:11:17 <3JTAANC26> build #159 of adm5120 is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/adm5120/builds/159 Mar 01 18:02:54 whee Mar 01 18:03:20 found out why my image won't boot - jffs2 won't init because there's no jffs2 there already Mar 01 18:09:09 hmmmmmmm Mar 01 18:10:00 flash setup gotta be wrong somehow o_O Mar 01 19:37:57 that wzr600dhp broke my wzr600dhp ;-) Mar 01 19:38:04 patch* Mar 01 19:38:31 r35840 -- Mar 01 19:38:54 had been working fine with the wzr-hp-ag300h target Mar 01 19:39:08 * russell-- got a kernel panic during boot Mar 01 19:39:32 [ 0.230000] VFS: Cannot open root device "(null)" or unknown-block(0,0): error -6 Mar 01 19:39:35 [ 0.240000] Please append a correct "root=" boot option; here are the available partitions: Mar 01 19:39:38 [ 0.250000] Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(0,0) Mar 01 20:00:06 luka: your r35840 broke my wzr600dhp Mar 01 20:05:12 [ 0.000000] Kernel command line: board=WZR-600DHP console=ttyS0,115200 rootfstype=squashfs,jffs2 noinitrd Mar 01 20:05:15 [ 0.000000] MIPS: no machine found for id 'WZR-600DHP', supported machines: Mar 01 20:23:16 russell--: let's chat a bit later, we'll fix it Mar 01 20:23:41 luka: okay, when-ish? Mar 01 20:43:52 luka: i think it might just be the classic "did you rebuild your kernel" problem ... trying Mar 01 20:44:08 patches not triggering rebuild, etc Mar 01 20:46:29 * russell-- realizing he hasn't need to flash from the bootloader on this device before Mar 01 21:50:00 russell--: ping Mar 01 21:50:51 luka: yeah, i think the kernel clean fixed it Mar 01 21:51:10 okay Mar 01 21:51:29 i still need to push one fix though Mar 01 21:53:10 * russell-- will test ;-) Mar 01 21:55:03 thx Mar 01 22:04:58 hi Mar 01 22:05:13 on latest openwrt rc, i get it when i use trafic control : http://pastebin.com/jXapeiCY Mar 01 22:39:50 mazilo r35841 packages/ net/freeswitch/patches/107-libs-libzrtp-include-zrtp_config_h.patch net/freeswitch/Makefile * include the forgotten patch for endian detection. Mar 01 22:57:54 jow_laptop: do you think it is possible that before r35484, firewall in trunk was completely broken ? Mar 01 22:58:31 I mean, my network has several vlans, communication between those vlans worked, I then flashed a new image, suddenly almost all inter-vlan communication is blocked Mar 01 23:06:49 <3JTAANC26> build #146 of mpc52xx is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/mpc52xx/builds/146 Mar 01 23:29:48 <3JTAANC26> build #208 of brcm63xx is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/brcm63xx/builds/208 Mar 01 23:36:04 * russell-- wonders if anyone is ever going to fix the svn-tracking bot's nick Mar 01 23:39:57 nope Mar 01 23:55:41 ok i don't know what's going on >_< Mar 01 23:56:13 split_squashfs isn't finding the filesystem, but the kernel boots fine and the partitions are all in the same spot in the kernel messages Mar 02 00:12:47 ? Mar 02 00:13:16 ./include/uapi/linux/magic.h:#define SQUASHFS_MAGIC 0x73717368 Mar 02 00:23:31 hmm it's reading 0 from flash o_O Mar 02 00:29:06 * russell-- just tried out r35840 with v3.8.1, seems happy so far Mar 02 00:33:14 nbd r35842 trunk/target/linux/cns3xxx/patches-3.3/066-pcie_bus_topology.patch * Revert "[cns3xxx]: clean up PCI bus topology" (r35079) Mar 02 00:38:04 mazilo r35843 packages/net/freeswitch/Makefile * updated to freeswitch v1.3.14b Mar 02 01:07:14 O_O ok that explains it Mar 02 01:07:28 <3JTAANC26> build #209 of at91 is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/at91/builds/209 Mar 02 01:09:06 kernel section in the flash is apparently 1048556 bytes. friggen edimax. Mar 02 01:48:41 yay it boots >_< Mar 02 02:43:53 mazilo r35844 packages/net/freeswitch/Makefile * fixed some of install scripts. **** ENDING LOGGING AT Sat Mar 02 02:59:59 2013