**** BEGIN LOGGING AT Wed Oct 24 03:00:01 2012 Oct 24 09:16:22 hi. i'm missing libpthread in my build. it shows in menuconfig and is selected as M, but isn't built. also i can't find it in packages/libs or any other location. where is it hidden in the buildroot? Oct 24 09:16:56 package/toolchain Oct 24 09:17:04 thx Oct 24 09:25:57 jow_laptop: http://pastebin.com/Z8sZSeNx Oct 24 09:26:12 what can i do here= make package/toolchain/clean doesn't help Oct 24 09:32:09 doesn't load Oct 24 09:32:59 the error is: cp: target `/mnt/data/buildroots/environment/trunk/staging_dir/target-mips_uClibc-0.9.33.2/root-brcm63xx/tmp-libc/lib/libgcc_s_pic.a' is not a directory Oct 24 09:33:15 is the tree clean? Oct 24 09:33:22 probably not Oct 24 10:45:36 RealOpty: Most debian projects are replacing CIA with KGB. http://kgb.alioth.debian.org/ Oct 24 10:46:28 oh the names Oct 24 11:29:04 <[florian]> stintel: one could think it's been named on purpose :) Oct 24 11:30:16 [florian]: that only makes it worse ;) Oct 24 18:09:40 [florian]: generic/patches-3.6/613-netfilter_optional_tcp_window_check.patch needs adjusting after d2ba1fde42af44fbce361202e9af13daff9e4381 Oct 24 20:54:54 hi, can someone tell me if current trunk is broken for rb532 or just here ? Oct 24 21:01:42 <[florian]> wotwot: what's broken exactly? Oct 24 21:02:14 [florian], i'm trying to build netboot images, they only get up to "jumping to kernel image" Oct 24 21:02:33 <[florian]> ok, what about flash images? Oct 24 21:02:52 that would mean climbing quite a bit Oct 24 21:03:07 <[florian]> true, well, you could enable earlyprintk and see if that changes anything Oct 24 21:03:43 i'll do that. was that in menuconfig or kernel_menuconfig ? Oct 24 21:04:10 <[florian]> in make menuconfig under Build options there is an option to enable Oct 24 21:04:18 ty [florian] Oct 24 21:04:21 <[florian]> and you will have to change the kernel command-line too Oct 24 21:04:30 <[florian]> I do not guarantee that this will change anything though Oct 24 21:04:39 <[florian]> the entry point of the kernel might just be completely Oct 24 21:04:41 ah right, and thats in kernel_m->kernel hacking->default cmdline ? Oct 24 21:04:48 <[florian]> exactly Oct 24 21:04:54 thanks Oct 24 21:05:11 <[florian]> you are welcome Oct 24 21:08:54 [florian], is the early printk CONFIG_KERNEL_PRINTK_TIME=y ? that is already set Oct 24 21:09:31 <[florian]> nopeit should CONFIG_SYS_HAS_EARLY_PRINTK=y Oct 24 21:09:32 <[florian]> CONFIG_EARLY_PRINTK=y Oct 24 21:10:28 k, trying to find it.. Oct 24 21:10:47 <[florian]> might need some kernel hacking to actually enable it Oct 24 21:10:49 <[florian]> let me chec Oct 24 21:10:51 <[florian]> +k Oct 24 21:10:53 ah Oct 24 21:11:31 <[florian]> just add SYS_HAS_EARLY_PRINTK in arch/mips/Kconfig for RB532 Oct 24 21:11:38 ah cool Oct 24 21:12:30 Hi, could you please tell me if there is a way to prevent openwrt boot from turning off early0 boot console? Oct 24 21:13:02 <[florian]> tonikasch: that's a kernel feature, once a proper serial driver is found, it will use it Oct 24 21:13:27 <[florian]> I don't know if you can actually keep it on Oct 24 21:13:39 [florian], bootlog shows this: ltq_asc.1: ttyLTQ1 at MMIO 0x1e100c00 (irq = 112) is a ltq_asc Oct 24 21:13:47 but it doesn't enable it :( Oct 24 21:14:20 well, thx anyway :) Oct 24 21:16:14 <[florian]> ask blogic about that, he maintains the lantiq target Oct 24 21:18:59 ok Oct 24 21:21:33 i'm getting "arch/mips/kernel/early_printk.c:22: undefined reference to `prom_putchar'" - do i have to write something else ? Oct 24 21:22:26 <[florian]> you'd probably need to hack a prom_putchar to output something on the uart Oct 24 21:22:43 damn :D Oct 24 21:23:12 <[florian]> cobalt uses the same UART afair, might be a good example to start with Oct 24 21:23:22 k, i'll dig into that Oct 24 23:08:30 i took one rb532, fed it with a CF card, compiled a jffs image, stuffed that onto CF and now it boots. sry, but its a bit late for me to do kernel hacking Oct 24 23:51:04 . Oct 25 00:33:19 I have a Linksys ea3500 and I'm quite unhappy with the stock firmware. If I was to take the kernel and the modules from the original firmware and the rootfs from openwrt I should be able to run openwrt on the router, right? Oct 25 00:50:00 openwrt kernels have a lot of patches Oct 25 00:54:52 yeah, that could be a problem. Oct 25 00:56:05 is it a broadcom-based device? Oct 25 00:56:30 DonkeyHotei: no, it is a marvell one Oct 25 01:02:08 which marvell core? Oct 25 01:04:09 DonkeyHotei: a "Feroceon 88FR131 rev 1 (v5l)" according to this blog http://blog.danwalters.net/blog/2012/06/19/hacking-linksys-ea3500-firmware-for-ssh-access/ Oct 25 01:05:04 that would be the "orion" target in openwrt Oct 25 01:11:24 yeah, I will probably try it out tomorrow. Oct 25 01:13:10 I will need to connect to the serial console first though... **** ENDING LOGGING AT Thu Oct 25 03:00:00 2012