**** BEGIN LOGGING AT Mon Sep 27 02:59:57 2010 Sep 27 03:08:14 build #134 of ppc44x is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/ppc44x/builds/134 Sep 27 04:11:27 build #100 of ifxmips is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/ifxmips/builds/100 Sep 27 06:03:29 nbd: any suggestions how to start tracking down the problem? Sep 27 06:05:45 i tried flipping the console in the CFE variable kernel_args from console=ttyS1,115200 to console=ttyS0,115200, but nothing more showed up on the console. Sep 27 06:06:06 in case the serial port was just switching the wrong way Sep 27 06:59:06 build #95 of ixp4xx is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/ixp4xx/builds/95 Sep 27 09:00:22 nbd * r23125 /trunk/target/linux/ar71xx/ (14 files in 7 dirs): Sep 27 09:00:22 Add support for Atheros AP96 reference board Sep 27 09:00:22 This is still preliminary and not really tested much. Anyway, this Sep 27 09:00:22 version seems to be enough to get both wired and wireless Sep 27 09:00:22 interfaces working. Buttons and leds are also working. Sep 27 09:00:22 mach-ap96.c is based on mach-wndr3700.c. Sep 27 09:00:23 Signed-off-by: Jouni Malinen Sep 27 09:00:31 russell_: no idea at the moment Sep 27 09:19:04 nbd: according to the commit, eth0 is wan, but uses phy0's state for aneg/speed, but eth1, which is wan is hardwired to 1000/full - shouldn't this be the other way round, and probably the phy mask 0x10 instead of 0x1? (IIRC ap96 uses the ar8316 too) Sep 27 09:19:12 err... eth0 is lan Sep 27 09:21:48 hm Sep 27 09:21:49 makes sense Sep 27 09:21:58 i'll look into this some more when i have the hardware at home Sep 27 09:23:21 probably the ar8316 code needs some adjustment, too. I think the 0x8 register needed a slightly different value for ap96 Sep 27 09:25:34 the "is initialized" check could be probably reduced to the lower 16 bits; these are the same on the rspro and ap96 Sep 27 09:29:00 at least according to the athrs16_phy.c ;) Sep 27 09:46:34 build #83 of uml is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/uml/builds/83 Sep 27 10:30:31 nbd: if i knew where, maybe strategic printk's or something? Sep 27 12:26:21 build #86 of ar7 is complete: Failure [failed shell_6] Build details are at http://tksite.gotdns.org:8010/builders/ar7/builds/86 Sep 27 12:29:43 hello Sep 27 13:12:01 hi, when compiling a package for openwrt i get the following error: undefined reference to `clock_gettime', any idea about the reason? Sep 27 13:16:35 you need -lrt flag Sep 27 13:17:55 during the configuration i see this message: checking whether clock_gettime is declared... yes Sep 27 13:18:17 so I think that the configure script is looking at the host machine and not at the target Sep 27 13:19:41 before calling .configure I have put $(TARGET_CONFIGURE_OPTS) in the Makefile Sep 27 13:22:28 build #90 of rdc is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/rdc/builds/90 Sep 27 14:42:21 juhosg * r23126 /trunk/target/linux/generic/files/drivers/net/phy/rtl8366rb.c: Sep 27 14:42:21 generic: add rtl8366rb switch port rate, port and qos enable/disable support Sep 27 14:42:21 Patch-by: Roman Yeryomin Sep 27 14:42:33 juhosg * r23127 /trunk/target/linux/generic/files/drivers/net/phy/ (rtl8366rb.c rtl8366s.c): Sep 27 14:42:33 generic: RTL8366S/RB: Fix autonegotiation for the WAN port. Sep 27 14:42:33 The RTL8366S/RB support connecting a second mac to it and using the 5th Sep 27 14:42:33 port as an independent phy. Therefore we need to handle autonegotiation Sep 27 14:42:33 changes for phy 4 like on a single phy device. Sep 27 14:42:33 This fixes #7980, verified on a WNDR3700. Sep 27 14:42:34 Signed-off-by: Jonas Gorski Sep 27 14:42:39 juhosg * r23128 /trunk/target/linux/ramips/rt305x/ (config-2.6.32 config-2.6.34): Sep 27 14:42:39 ramips: allow USB Sep 27 14:42:39 The current defaults for ramips rt305x does not allow the USB Sep 27 14:42:39 subsystem to be selected/built. Sep 27 14:42:39 This patch is just a small step in that direction. Sep 27 14:42:39 Signed-off-by: Luis Correia Sep 27 14:44:01 juhosg * r23129 /trunk/target/linux/ramips/ (7 files in 4 dirs): Sep 27 14:44:01 ramips: Ralink RT305x, support for Prolink PWH2004 (Abocom5205) Sep 27 14:44:01 The attached patch adds a preliminary support for Prolink PWH2004 board Sep 27 14:44:01 (Ralink RT305x based). LAN and flash appear to work. However, the USB Sep 27 14:44:01 port and WLAN I have not managed to activate properly yet, but that Sep 27 14:44:01 problem appears to be present with other rt305x targets as well(?). Sep 27 14:44:02 Signed-off-by: Esa Hyytia Sep 27 14:54:09 juhosg * r23130 /trunk/target/linux/ar71xx/files/arch/mips/ar71xx/dev-ap91-pci.c: Sep 27 14:54:09 ar71xx: use different address in ap91_pci_fixup for the AR724[012] SoCs Sep 27 14:54:09 With this change ath9k can handle the wireless chip on the TL-WR841N v7 Sep 27 14:54:09 board which is based on the AR7241 SoC. Sep 27 14:54:11 juhosg * r23131 /trunk/target/linux/ar71xx/ (10 files in 2 dirs): ar71xx: move ath9k specific PCI fixup into a separate file Sep 27 14:54:14 juhosg * r23132 /trunk/target/linux/ar71xx/files/arch/mips/ar71xx/devices.c: ar71xx: use the MDIO interface of the 2nd MAC on the ar7241 Sep 27 14:54:23 juhosg * r23133 /trunk/tools/firmware-utils/src/mktplinkfw.c: firmware-utils/mktplinkfw: add support for the TL-WR841N v7 Sep 27 14:54:23 juhosg * r23134 /trunk/target/linux/ar71xx/ (generic/profiles/tp-link.mk image/Makefile): ar71xx: build firmware image for the TL-WR841N v7 Sep 27 14:54:23 juhosg * r23135 /trunk/target/linux/ar71xx/files/arch/mips/ (ar71xx/dev-usb.c include/asm/mach-ar71xx/ar71xx.h): ar71xx: AR724[12] requires different USB setup Sep 27 14:54:25 juhosg * r23136 /trunk/target/linux/ar71xx/ (11 files in 5 dirs): ar71xx: add support for the TL-MR3420 v1 board Sep 27 14:54:27 juhosg * r23137 /trunk/tools/firmware-utils/src/mktplinkfw.c: firmware-utils/mktplinkfw: add support for the TL-MR3420 v1 Sep 27 14:54:29 juhosg * r23138 /trunk/target/linux/ar71xx/ (generic/profiles/tp-link.mk image/Makefile): ar71xx: build firmware image for the TL-MR3420 v1 board Sep 27 15:38:18 build #96 of iop32x is complete: Failure [failed shell_6] Build details are at http://tksite.gotdns.org:8010/builders/iop32x/builds/96 Sep 27 15:52:06 build #97 of adm5120 is complete: Failure [failed shell_6] Build details are at http://tksite.gotdns.org:8010/builders/adm5120/builds/97 Sep 27 16:46:01 build #122 of avr32 is complete: Failure [failed shell_6] Build details are at http://tksite.gotdns.org:8010/builders/avr32/builds/122 Sep 27 18:30:08 hey Hauke Sep 27 18:30:50 hey Hauke Sep 27 18:33:14 Hauke1, can u tell me the headers id need to include in the flash map file(k2.4) to duplicate what we did in 2.6 regarding detecting "board_data"? Sep 27 18:33:33 hi RealOpty Sep 27 18:34:16 RealOpty: what do your mean by header id? Sep 27 18:34:35 the header file for nvram Sep 27 18:35:39 RealOpty: you want to backport these changes to kernel 2.4? Sep 27 18:37:26 basically yes Sep 27 18:38:55 Hauke1, but dont worry about it, theres a possibility that my idea for making wifi working in backfire might not work out. Sep 27 18:42:26 I was gonna attempt to fix backfire so it wont brick the wrt, then i was hoping to try out the broadcom drivers from 8.09 inside backfire, and if it works patch backfire, and explain how to make wireless work by inserting the packages from 8.09 Sep 27 18:43:56 RealOpty: I do not know how the header file for nvram access is names in kernel 2.4 Sep 27 18:44:41 kk, ill use dirty hacks for testing until i can confirm my wild idea will work. Sep 27 20:48:32 build #93 of orion is complete: Failure [failed shell_6] Build details are at http://tksite.gotdns.org:8010/builders/orion/builds/93 Sep 27 21:40:51 mb * r23139 /feeds/lxde/ (17 files in 14 dirs): Sep 27 21:40:52 Update LXDE. Sep 27 21:40:52 lxterminal needs an updated version of VTE. This is currently being worked around by disabling a cursor shaping feature. Sep 27 22:11:35 mb * r23140 /feeds/lxde/apps/lxpanel/ (6 files in 4 dirs): Sep 27 22:11:35 lxpanel: Override the default config to be more embedded-friendly. Sep 27 22:11:35 Removes the wasteful (w.r.t. CPU resources) CPU-monitor and background theme. Sep 27 22:11:35 Users can easily add it back in the config dialog, if they want it. **** ENDING LOGGING AT Tue Sep 28 02:59:57 2010