**** BEGIN LOGGING AT Sun Jul 24 02:59:57 2011 Jul 24 04:19:50 How to build bootloaders in backfire ? Jul 24 07:27:01 build #58 of lantiq is complete: Failure [failed compile_6] Build details are at http://buildbot.openwrt.org:8010/builders/lantiq/builds/58 Jul 24 08:00:28 build #58 of atheros is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/atheros/builds/58 Jul 24 08:08:27 nbd * r27739 /trunk/target/linux/ar71xx/files/arch/mips/ar71xx/mach-wzr-hp-ag300h.c: ar71xx: enable the usb power regulator on the wzr-hp-ag300h through gpio 2 (#9684) Jul 24 08:28:59 nbd * r27740 /packages/net/opennhrp/Makefile: opennhrp: enable CONFIG_ARPD if selected (fixes #9751) Jul 24 08:29:01 nbd * r27741 /packages/libs/clearsilver/Config.in: clearsilver: selecting zlib from one place is enough Jul 24 08:40:15 nbd * r27742 /trunk/include/cmake.mk: cmake: fix ccache command when using host ccache (patch from #9611) Jul 24 08:48:05 nbd * r27743 /trunk/target/linux/x86/geos/target.mk: Jul 24 08:48:05 geos: include pppstats and pppdump in default packaging Jul 24 08:48:05 As the Geos box includes an ADSL modem that's targeted towards PPPoE, PPPoA, and bridged utilization, including these useful tools allows for easier configuration and troubleshooting. Jul 24 08:48:05 Signed-off-by: Philip Prindeville Jul 24 09:08:56 nbd * r27744 /trunk/tools/pkg-config/files/pkg-config: (log message trimmed) Jul 24 09:08:56 pkg-config: fix prefix Jul 24 09:08:56 Now: Jul 24 09:08:56 % PATH=./staging_dir/host/bin:$PATH STAGING_PREFIX=/test PKG_CONFIG_PATH=$(echo staging_dir/target-*/usr/lib/pkgconfig) ./staging_dir/host/bin/pkg-config --cflags lua -I/test//include Jul 24 09:08:56 After applying the patch: Jul 24 09:08:57 % PATH=./staging_dir/host/bin:$PATH STAGING_PREFIX=/test PKG_CONFIG_PATH=$(echo staging_dir/target-*/usr/lib/pkgconfig) ./staging_dir/host/bin/pkg-config --cflags lua -I/test/include Jul 24 09:08:58 I guess that there sould not be '//' but only one '/'. Jul 24 09:16:06 nbd * r27745 /packages/net/usbip/Makefile: Jul 24 09:16:06 USB/IP: Fix building for linux 3.0 Jul 24 09:16:06 The config symbol and module names were changed for 3.0. Jul 24 09:16:06 Signed-off-by: Jonas Gorski Jul 24 09:16:08 nbd * r27746 /packages/net/usbip/Makefile: Jul 24 09:16:08 USB/IP: Add missing dependency to kmod-usb-core Jul 24 09:16:08 Signed-off-by: Jonas Gorski Jul 24 09:17:17 nbd * r27747 /trunk/target/linux/atheros/ (2 files in 2 dirs): Jul 24 09:17:17 atheros: fix mii phy read/write ioctl Jul 24 09:17:17 Signed-off-by: Marek Lindner Jul 24 09:23:02 nbd * r27748 /trunk/include/image.mk: default to using gzip with mksquashfs if lzma and xz are unavailable Jul 24 09:26:30 build #48 of ar71xx is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/ar71xx/builds/48 Jul 24 09:47:14 nbd * r27749 /trunk/target/linux/x86/config-default: Jul 24 09:47:14 x86: 2.6.39 requires additional symbol Jul 24 09:47:14 Sysfs support for DMI BIOS was added in 2.6.39. Jul 24 09:47:14 Signed-off-by: Philip Prindeville Jul 24 10:29:33 build #56 of ps3 is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/ps3/builds/56 Jul 24 11:37:51 build #57 of cobalt is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/cobalt/builds/57 Jul 24 12:03:39 updated openwrt/upstream, https://home.comcast.net/~sdwalker/uscan/index.html Jul 24 12:12:23 build #59 of lantiq is complete: Exception [exception interrupted] Build details are at http://buildbot.openwrt.org:8010/builders/lantiq/builds/59 Jul 24 12:35:36 hauke * r27750 /trunk/package/mac80211/patches/006-disable_bcma_build.patch: mac80211: do not build bcma from compat-wireless Jul 24 12:41:59 build #56 of pxcab is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/pxcab/builds/56 Jul 24 14:16:48 jow * r27751 /trunk/package/broadcom-wl/ (11 files in 4 dirs): Jul 24 14:16:48 [package] broadcom-wl: make it work on the RT-N16 and other device using the BCMA bus Jul 24 14:16:48 - refresh and reorder patches Jul 24 14:16:48 - introduce a unified BCMA/SSB glue driver to accomodate for both bus types on brcm47xx Jul 24 14:16:48 - extend wlunbind init script to detach device from bcma-pci-bridge as well Jul 24 14:20:56 jow * r27752 /trunk/package/broadcom-wl/patches/ (007-add-bcma-support.patch 007-use-glue-driver.patch): [package] broadcom-wl: give the 007 patch a more meaningful name Jul 24 14:24:08 hauke * r27753 /trunk/ (67 files in 24 dirs): (log message trimmed) Jul 24 14:24:08 ocf-linux: version bump to 20110720 Jul 24 14:24:08 Fixes problem with TFM allocation in cryptosoft.c Jul 24 14:24:08 Signed-off-by: Philip Prindeville Jul 24 14:24:08 Hauke: Jul 24 14:24:09 * remove ubsec_ssb package and take it from ocf-linux Jul 24 14:24:09 * use patches from ocf-linux package Jul 24 14:46:31 nbd * r27754 /trunk/target/linux/ (12 files in 5 dirs): rtl8366: unify rtl8366s and rtl8366rb platform data Jul 24 14:46:39 nbd * r27755 /trunk/target/linux/generic/files/ (drivers/net/phy/rtl8366_smi.c include/linux/rtl8366.h): rtl8366_smi: implement a function for detecting whether the attached switch is RTL8366S or RTL8366RB Jul 24 14:46:43 nbd * r27756 /trunk/target/linux/ar71xx/files/arch/mips/ar71xx/ (Kconfig mach-wzr-hp-g300nh.c): ar71xx: autodetect rtl8366s/rtl8366rb on wzr-hp-g300nh Jul 24 15:14:25 hauke * r27757 /trunk/package/kernel/modules/crypto.mk: kernel: crypto: fix spelling Jul 24 15:24:38 hauke * r27758 /trunk/target/linux/brcm47xx/patches-3.0/230-ssb_pci_sprom.patch: Jul 24 15:24:38 brcm47xx: run ssb_pcicore_fix_sprom_core_index just for ssb bus on pci Jul 24 15:24:38 ssb_pcicore_fix_sprom_core_index accesses the sprom on the pci bus but Jul 24 15:24:38 this causes a data bus error (oops) on a SoC. Jul 24 15:42:41 jow * r27759 /trunk/target/linux/generic/patches-3.0/501-yaffs_cvs_2009_04_24.patch: [generic] refresh 501-yaffs_cvs_2009_04_24.patch for Linux 3.0 final Jul 24 15:43:09 jow * r27760 /trunk/target/linux/generic/patches-3.0/475-mtd_cfi_cmdset_0002-add-buffer-write-cmd-timeout.patch: Jul 24 15:43:09 From: George Kashperko Jul 24 15:43:09 Issue map read after Write Buffer Load command to ensure chip is ready Jul 24 15:43:09 to receive data. Jul 24 15:43:09 Signed-off-by: George Kashperko Jul 24 17:35:05 build #56 of orion is complete: Failure [failed compile_6] Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/56 Jul 24 18:40:34 build #52 of ppc44x is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/ppc44x/builds/52 Jul 24 19:39:22 build #52 of uml is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/uml/builds/52 Jul 24 19:41:42 ^- uml and ppc hang with "NEW" kernel options Jul 24 20:05:57 hi Jul 24 20:07:15 brcm47xx is broken in trunk - i created a complete new image (deleted staging and build dir) but the kernel doesnt come up ... r27723 worked nicely with linux 3.0 Jul 24 20:07:55 i dont have a serial cable attacxhed to the device so i cant tell whats the exact error Jul 24 20:15:12 KanjiMonster: broken how? I was just going to build and load the OCF stuff... Jul 24 20:15:43 i got disconnected... can someone verify that brcm47xx is broken? Jul 24 20:17:26 tripolar: ware ypu using kernel 2.6.39 or 3.0? Jul 24 20:17:29 you Jul 24 20:17:32 3.0 Jul 24 20:17:47 r27723 worked with 3.0 Jul 24 20:18:06 maybe ocf is the cause - its the biggest change since 3.0 Jul 24 20:18:17 since r27723 Jul 24 20:19:07 i thought maybe its 27758 but deleting the patch doesnt solve the problem Jul 24 20:22:57 Hauke: what do you think? what could be the cause? Jul 24 20:28:28 tripolar: it is this change: https://dev.openwrt.org/changeset/27736 Jul 24 20:28:46 the device is unable to load modules Jul 24 20:31:08 Hauke: can you revert it? Jul 24 20:31:41 tripolar: or fix it up properly? Jul 24 20:32:42 but why does removing this 999-drop_sublevel.patch file doesnt allow the kernel to load a module? Jul 24 20:32:51 @ Hauke Jul 24 20:33:42 because insmod is looking for the modules in /lib/modules/3.0 instead of 3.0.0 Jul 24 20:34:08 Hauke: how do I refresh the patches to fix the fuzz? Jul 24 20:34:18 err... @nbd Jul 24 20:34:36 philipp64|laptop: make foo/bar/clean && make foo/bar/refresh Jul 24 20:35:09 ok. Jul 24 20:36:07 building 3.0 on the x86 stops here... http://fpaste.org/AbeH/ not sure why. Jul 24 20:36:11 Kaloz but the changeset removed the file so it was looking befor (when the patch was still used) in 3.0 - wasnt it? Jul 24 20:36:37 build #53 of x86 is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/x86/builds/53 Jul 24 20:37:03 tripolar: with 3.0-rc that file made the modules go to /lib/modules/3.0 Jul 24 20:37:33 philipp64|laptop: MELAN (NEW) Jul 24 20:38:23 philipp64|laptop: so CONFIG_MELAN is mising from the config file Jul 24 20:38:25 but rootimage still uses lib/modules/3.0/ Jul 24 20:38:32 Kaloz Jul 24 20:39:16 can someone revert changeset 27736 Jul 24 20:39:18 ? Jul 24 20:39:42 I think KanjiMonster ported everything to use 3.0 (without sublevel) Jul 24 20:40:32 Kaloz: thanks Jul 24 20:42:18 tripolar: revert it locally, if you can't wait for the proper fix Jul 24 20:42:34 loswillios where is it defined to not use the sublevel? Jul 24 20:42:58 for the /lib/modules dir Jul 24 20:43:50 tripolar: best to have a look at his patch series (it was applied a few days/weeks ago) Jul 24 20:46:05 Kaloz: should SCSI_ISCI go into target/linux/generic/config-3.0 or target/linux/x86/config-default ? Jul 24 20:47:54 philipp64|laptop: it seems to be (at least for now) limited to intel's c6xx chipsets, so probably the later Jul 24 20:47:59 anyone here know which piece(s) of software are responsible for creating wds client devices (wlanx.stay) and assigning the mac address they use? the latter is broken Jul 24 20:49:10 there's a software-only iscsi initiator in the kernel. what am I missing there with chipset support? Jul 24 20:49:43 performance Jul 24 20:50:09 if you meant what you miss there versus a hw iscsi initiator Jul 24 20:50:13 that config option applies to any initiator though, right? Jul 24 20:50:26 isci != iscsi Jul 24 20:50:27 bah, i'd have to look at it Jul 24 20:50:33 oh Jul 24 20:50:41 anyway, wds, needs fixing :) Jul 24 21:02:05 tripolar: I did this in advance because of this -> https://lkml.org/lkml/2011/5/29/204 - "( ...), so the kernel is fighting back, and is calling itself 3.0.0-rc1. We'll have the usual 6-7 weeks to wrestle it into submission, and get scripts etc cleaned up, and the final release should be just "3.0"." Jul 24 21:02:20 eh Jul 24 21:02:21 meh Jul 24 21:03:56 ok, patched 3.0 missing symbols for x86. Jul 24 21:04:17 so if I boot this linux-3.0 x86 image that I just built... what's the probability of it working? Jul 24 21:06:21 philipp64|laptop: depends on if you *need* modules to boot/read from the root partition or if you just need modules for "optional" things like network ;) Jul 24 21:06:41 uhh.... yes. Jul 24 21:07:07 Kaloz: does the /refresh thing work with target/linux/refresh as well? Jul 24 21:07:17 yes Jul 24 21:07:36 refreshes the generic patches as well as the current selected target's patches Jul 24 21:11:22 philipp64|laptop: +# CONFIG_OCF_UBSEC_SSB is not set <- was already added in r27753 (with the ocf update) Jul 24 21:13:39 philipp64|laptop: yup Jul 24 21:14:04 philipp64|laptop: fixup the insmod issue and it will boot and work just fine Jul 24 21:14:26 just updating now... will I pick up everything I need? Jul 24 21:16:15 hmmm... did a "cd /lib/modules ; ln -s 3.0 3.0.0 " and rebooted. seems to work. Jul 24 21:16:37 "cryptotest -p -z" works too! Jul 24 21:22:53 :) Jul 24 21:27:09 Hauke: I'm confused... http://fpaste.org/XyPv/ ... why leave this out? it was part of the upstream patches. Jul 24 21:27:46 in any case, if we're forcing CONFIG_CRYPTO_MANAGER for kmod-cryptosoft it's a moot point, right? Jul 24 21:28:14 philipp64|laptop: without it I got this warning while loading the module: http://pastebin.com/CfNC7nPf Jul 24 21:29:33 interesting. I'll report it upstream. Jul 24 21:30:28 I already send a patch upstream Jul 24 21:32:27 patch? not just reverting the fix to leave the dependency? Jul 24 21:32:51 also, why did the new Kirkwood files get left out? Jul 24 21:33:31 thei are not referenced anywhere Jul 24 21:33:35 and what is mv_hal? Jul 24 21:33:40 ok. Jul 24 21:39:44 KanjiMonster: redid the config-3.0 patch without UBSEC_SSB Jul 24 21:48:45 jogo * r27761 /trunk/include/kernel.mk: Jul 24 21:48:45 kernel.mk: Fix kernel module path for 3.0 Jul 24 21:48:45 Linux 3.0 calls itself 3.0.0, so insmod expects the modules in Jul 24 21:48:45 /lib/modules/3.0.0. Jul 24 21:49:06 so what about 1243? Jul 24 21:49:29 that's the Alix2 patch that adds in a default "wan" interface definition. Jul 24 21:50:19 Russell's comment was that the labels on his box are the opposite as mine. I'm fine either way... but I think the default config should have a WAN and LAN interface. Jul 24 21:50:27 or a LAN and WAN interface. :-) Jul 24 21:51:25 Hauke: did your fix to McCullough include 3.0 patches? Jul 24 21:51:39 no Jul 24 21:51:50 ok. I'll send those along. Jul 24 21:52:52 It just contained this dependency missing, I send it to the mailing list but I am not subscripted so the moderator has to approve it. Jul 24 21:54:02 philipp64|laptop: there, now the images should work again ootb ;) Jul 24 21:55:58 I will go to sleep now Jul 24 22:10:19 kaloz * r27762 /trunk/target/linux/ (generic/config-3.0 x86/config-default): add missing symbols for generic and x86 on 3.0, based on Philip Prindeville's patch Jul 24 22:25:44 btw: 2.6.39.3 has been declared latest/stable. Jul 24 22:27:46 mhei * r27763 /packages/lang/php5/patches/950-Fix-dl-cross-compiling-issue.patch: Jul 24 22:27:46 [packages] php5: fix cross compiling issue with -ldl Jul 24 22:27:46 Php5's configure script seems to be a little bit buggy as Jul 24 22:27:46 it disables dynamic library support during cross-compiling. Jul 24 22:27:46 See also http://www.apachefriends.org/f/viewtopic.php?f=6&t=32367 Jul 24 22:27:46 For now, fix this by preventing the configure script overriding Jul 24 22:27:47 the previously detected value. Jul 24 23:31:51 build #56 of ramips is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/ramips/builds/56 Jul 25 00:05:24 build #49 of avr32 is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/49 Jul 25 02:19:22 build #53 of ppc40x is complete: Exception [exception interrupted] Build details are at http://buildbot.openwrt.org:8010/builders/ppc40x/builds/53 Jul 25 02:20:17 build #50 of sibyte is complete: Exception [exception interrupted] Build details are at http://buildbot.openwrt.org:8010/builders/sibyte/builds/50 Jul 25 02:22:48 * philipp64|laptop draws a deep breath and clicks 'Install Lion'... **** ENDING LOGGING AT Mon Jul 25 02:59:57 2011