**** BEGIN LOGGING AT Thu Apr 29 02:59:56 2010 Apr 29 03:03:50 nico * r21246 /trunk/target/linux/x86/image/Makefile: Apr 29 03:03:51 target/x86: various image changes Apr 29 03:03:51 * add a -combined suffix for grub+kernel+rootfs images and -rootfs for root filesystem only images Apr 29 03:03:51 * change ext from .image to .img Apr 29 03:03:51 * fix perms on VirtualBox .vdi image Apr 29 03:03:51 * gzip ext2 rootfs image as well Apr 29 03:57:56 nico * r21247 /trunk/ (2 files in 2 dirs): move kmod-lp package defition specific to brcm-2.4 to package/kernel Apr 29 04:04:21 nico * r21248 /trunk/package/kernel/modules/ (9 files): package/kernel: move AddDepends calls at the end of package definitions, move back some dependencies in DEPENDS for clarity Apr 29 04:35:49 craigc * r21249 /packages/net/multiwan/ (Makefile files/etc/config/multiwan files/usr/bin/multiwan): [packages] multiwan: removed resolv_conf settings and added dns server override & updated default config file. Apr 29 07:07:52 gmorning Apr 29 07:40:43 hi, is x86 build broken or is it just me? Apr 29 07:41:19 grub doesn't seem to think it's cross compiling Apr 29 07:41:35 since I am running on x86 but Glibc Apr 29 07:42:39 http://pastebin.com/5RtCcWid Apr 29 07:53:16 the toolchain path should be on your env Apr 29 07:53:33 therof cross compiled grub should be called Apr 29 07:54:31 issue a: make distclean Apr 29 07:54:38 and see if things are better Apr 29 07:56:30 also try to use a latest trunk Apr 29 07:58:00 actually, I just checked out from trunk and just make menuconfig + make, but i'll try svn up and make distclean Apr 29 08:04:21 somebody submitted a messed up version of my patch to the mailinglist ... Apr 29 08:04:43 <[florian]> rtz2: which patch? Apr 29 08:04:55 brcm4716 support Apr 29 08:05:19 <[florian]> ah, yes ifdef'd with some weird config symbol? Apr 29 08:05:23 right Apr 29 08:05:33 and there is a long list of other problems Apr 29 08:06:11 now I only need a polite way to say no but thank you and keep him working on stuff like usb and network support Apr 29 08:06:19 <[florian]> ah ah Apr 29 08:10:32 hmmmm Apr 29 08:11:19 anybody here has a really old wrt56g with a 4710 soc or any other version without a chipcommon core? Apr 29 08:11:30 wrt54g Apr 29 10:23:39 nico * r21250 /trunk/package/kernel/modules/001-depends.mk: package/kernel: adjust kmod-crc16 dependencies, it does not exist on 2.4 Apr 29 10:25:59 nico * r21251 /trunk/package/kernel/modules/ (7 files): package/kernel: mark a bunch of kmod packages that don't exist on 2.4 as 2.6 only Apr 29 12:15:47 <[florian]> rtz2: seems like you will need patience to explain things :p Apr 29 12:24:39 [florian]: as long as people are willing to learn, i'm usually pretty patient :) Apr 29 12:29:01 [florian]: i'm more worried about driving him away Apr 29 12:50:11 rhk * r21252 /trunk/target/linux/brcm63xx/ (4 files in 2 dirs): [brcm63xx] fix gpio_button device register Apr 29 12:51:04 rhk * r21253 /trunk/target/linux/brcm63xx/ (6 files in 2 dirs): [brcm63xx] fix gpio button key code, match button-hotplug code Apr 29 12:51:43 rhk * r21254 /trunk/target/linux/brcm63xx/ (6 files in 2 dirs): [brcm63xx] add WPS button support for nb4 boards Apr 29 12:54:25 rhk * r21255 /trunk/target/linux/brcm63xx/ (10 files in 2 dirs): [brcm63xx] rename reset_buttons to buttons Apr 29 13:12:46 acoul * r21256 /trunk/target/linux/mpc83xx/ (5 files in 3 dirs): mpc83xx: upgrade to kernel 2.6.33.3, enable rb333 support, sync config options, refresh kernel patches Apr 29 13:25:15 acoul * r21257 /trunk/target/linux/mpc83xx/config-2.6.33: mpc83xx: trim down unnecessary config-2.6.33 options Apr 29 13:40:27 acoul * r21258 /trunk/target/linux/mpc83xx/config-2.6.33: mpc83xx: add some missing options on config-2.6.33 Apr 29 13:48:50 acoul * r21259 /trunk/target/linux/mpc83xx/image/Makefile: mpc83xx: sync image/Makefile to r20834 Apr 29 15:57:04 mirko * r21260 /trunk/target/linux/xburst/config-2.6.32: enable inotify support target xburst Apr 29 16:01:30 nico * r21261 /trunk/target/linux/generic-2.4/config-default: target/brcm-2.4: add missing CONFIG_INPUT_* symbols Apr 29 16:04:22 nico * r21262 /trunk/package/kernel/modules/other.mk: package/kernel: renable kmod-input on 2.4 (closes: #7063) Apr 29 17:02:22 ali1234: ping Apr 29 17:03:25 hi Apr 29 17:03:34 i have about 1 minute to talk then i have to go eat dinner :) Apr 29 17:04:17 ah, ok, just a quick question ;) Apr 29 17:04:49 Since you were poking around in the spi registers for bcm63xx, you do have the register definitions, right? Apr 29 17:05:29 yeah Apr 29 17:05:40 (because I just noticed the spi/adsl registers weren't described in the _map.h in an older tarball, but in a recent tarball they were Apr 29 17:05:41 ) Apr 29 17:05:42 ok Apr 29 17:05:50 the broadcom one? Apr 29 17:05:55 yeah Apr 29 17:06:00 the registers are also defined in the openwrt kernel Apr 29 17:06:16 the definitions are pretty much ccorrect, if a little bit... undescriptive Apr 29 17:06:28 ah I see Apr 29 17:06:33 what do you need it for anyway? Apr 29 17:06:46 also note they are different on 6338 6348 etc Apr 29 17:06:57 i only have 6338 Apr 29 17:07:14 hmm gotta go Apr 29 17:07:18 I don't need them, I just saw them (while inspecting the tarball for the polish router linked on trac) Apr 29 17:18:29 lars * r21263 /trunk/package/uboot-xburst/files/cpu/mips/nanonote_gpm940b0.h: uboot-xburst: Fix lcd display init sequence. Apr 29 18:13:57 aport: I just saw your post (and answered it ;) Apr 29 18:27:56 KanjiMonster, thanks I'll check it out Apr 29 18:32:02 KanjiMonster, are these values set in bootloader or switch driver? Apr 29 18:32:09 aport: boot loader Apr 29 18:32:37 KanjiMonster, thanks, that would make sense. Time to hit the datasheets again Apr 29 18:33:00 aport: oh, you have datasheets? Apr 29 18:33:23 KanjiMonster, I have sheets for AR7161 and AR8316 Apr 29 18:33:32 ah, nice Apr 29 18:33:45 (I wish I had them) Apr 29 18:34:07 I'll have to review my NDA to see if I can divulge any useful information Apr 29 18:34:16 more than likely not Apr 29 18:34:21 probably Apr 29 18:34:41 if its the default "only for business evaluation"-nda Apr 29 18:35:24 there may be an underlying issue with my setup though Apr 29 18:35:47 Though the LAN ports are functional, no "eth0 link up/down" messages appear when the ethernet is inserted or removed Apr 29 18:35:54 though on the WAN port it works fine Apr 29 18:37:19 that's because eth0 is the connection between the mac and the switch Apr 29 18:40:09 you could e.g. generate messages in the ar8216_read_status() based on the phy registers (0-3 are the lan ports), or the switch's port status registers Apr 29 18:46:46 thanks KanjiMonster, I appreciate it Apr 29 18:47:00 to be honest this type of stuff is pretty new to me Apr 29 18:47:20 kind of bit off a lot and trying to learn how to chew it Apr 29 19:26:53 KanjiMonster, could I perhaps get your patch that used to be at http://page.mi.fu-berlin.de/jgorski/openwrt/0001-Add-support-for-the-ar8316-switch_debug.patch Apr 29 19:57:08 aport: I don't know if I have it anymore, let me take a look Apr 29 20:02:36 aport: You probably have to manually apply the stuff, but here it is: http://page.mi.fu-berlin.de/jgorski/openwrt/temp/0001-Add-support-for-the-ar8316-switch_debug.patch Apr 29 20:02:53 KanjiMonster, it's okay. I added some debug statements to the switch driver to show me the values of those LED registers Apr 29 20:03:22 I'm just confused as to why it works fine in u-boot but not in OpenWrt Apr 29 20:03:44 Is there anything that would prevent backfire from working on a really old kernel, like 2.6.11? I'l like to test something.. Apr 29 20:04:10 aport: perhaps the u-boot does a hw reset of the switch before booting linux? Apr 29 20:04:57 KanjiMonster, it's possible, I'll take a look at that Apr 29 20:06:16 xl0: you will probably run into lots of subtle problems Apr 29 20:06:48 aport: when doing a hw reset the switch flashes all leds simultaneusly on the rs pro, this should be quite obvious Apr 29 20:06:50 xl0: mtd maps, drivers, iptables apis and lot's of other things Apr 29 20:06:56 values of b0, b4, and b8 are cc35, ca35, and c935 respectively. Those match my AR8316 sheets Apr 29 20:07:06 KanjiMonster, yeah it's not doing that then Apr 29 20:22:56 I set the registers to routerstation settings (cc37cc37 all three) and saw that the LEDs turned off once the driver was loaded. Looks like I'm getting warmer. Thanks again KanjiMonster Apr 29 20:25:54 aport: the rs has cc37cc37 in b0 and b4, but b8 is ca37ca37 Apr 29 20:26:30 KanjiMonster, I see Apr 29 20:26:30 eh Apr 29 20:26:39 no, cc37* only in b0 Apr 29 20:26:49 and ca37ca37 in b4 and b8 Apr 29 20:26:56 CA/CC37 isn't documented in my datasheet :( Apr 29 20:27:05 only CC/CA/C935 Apr 29 20:31:22 rtz2: The drivers are in the kernel, and I probably won't need iptables. What bothers me is the hotplug dependency, not sure if it would work the same way in an ancient kernel. Apr 29 20:34:01 no clue, you will have to try **** ENDING LOGGING AT Thu Apr 29 21:57:51 2010 **** BEGIN LOGGING AT Thu Apr 29 21:58:15 2010 **** ENDING LOGGING AT Thu Apr 29 21:58:51 2010 **** BEGIN LOGGING AT Thu Apr 29 22:00:14 2010 **** ENDING LOGGING AT Fri Apr 30 02:59:57 2010