**** BEGIN LOGGING AT Tue Jan 17 02:59:58 2012 Jan 17 03:14:49 build #127 of at91 is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/at91/builds/127 Jan 17 03:28:46 build #125 of ubicom32 is complete: Failure [failed compile_3] Build details are at http://buildbot.openwrt.org:8010/builders/ubicom32/builds/125 Jan 17 11:10:37 build #85 of octeon is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/octeon/builds/85 Jan 17 11:23:54 blogic * r29762 /trunk/target/linux/ramips/files/drivers/net/ramips.c: (log message trimmed) Jan 17 11:23:54 ramips: Rework ramips_eth to not require irqsave locking anymore Jan 17 11:23:54 Previously the tx housekeeping was done in a spin_lock_irqsave critical Jan 17 11:23:54 section which causes irqs to be disabled during that time. Since the Jan 17 11:23:54 housekeeping is already prepared to be scheduled as a tasklet process Jan 17 11:23:55 the housekeeping only in softirq context and revise the locking between Jan 17 11:23:56 the tx path and the housekeeping tasklet by using a normal spin_lock Jan 17 12:11:19 blogic * r29763 /trunk/target/linux/lantiq/ (2 files in 2 dirs): Jan 17 12:11:19 [lantiq] gigaset 76x Jan 17 12:11:19 * fixes leds Jan 17 12:11:19 * adds all leds found on the sx763 board Jan 17 12:11:19 * removes board_config from flash layout because wireless data is stored in kernel (ath5k) Jan 17 12:11:19 * fix button handling Jan 17 12:11:19 Signed-off-by: Luka Perkov Jan 17 13:43:51 tripolar * r29764 /packages/net/nfs-kernel-server/Makefile: [packages]: nfs-kernel-server - don't link against libcap Jan 17 14:44:00 build #128 of s3c24xx is complete: Failure [failed compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/s3c24xx/builds/128 Jan 17 16:24:50 hi Jan 17 16:25:16 why doesn't ocf work with linux 3.2? - at least it is disabled for linux 3.2 Jan 17 16:27:19 moo Jan 17 16:30:49 is openwrt.org going to go dark for the SOPA/PIPA protest? Jan 17 16:31:19 (and i mean "on purpose" ;-) Jan 17 16:31:30 probably not Jan 17 16:32:19 dnsmasq should get an update to resolve everthing to the EFF site ;-) Jan 17 16:33:05 while i agree with the cause, i don't think we should do that kind of stuff in openwrt Jan 17 16:33:48 also, openwrt is much more european than US based Jan 17 16:33:53 ;) Jan 17 16:34:24 unfortunately, my government has a global reach. jackasses. Jan 17 16:35:10 nbd: hello! Jan 17 16:35:16 also, in the long run such protests will probably not hold against the systemic issues that keep putting things like SOPA/PIPA on the table Jan 17 16:35:39 nbd: if I make an initramfs image, I get a uImage, which u-boot just boots, not flashes. what to do/hack? Jan 17 16:36:12 that wikipedia is doing it is sufficient to raise awareness. Jan 17 16:36:28 btw, would nbd be interested in participating in a podcast interview thingie? i friend of mine is randal schwartz does a thing called floss weekly. they did a show on village telco recently during which openwrt was mentioned prominently. Jan 17 16:36:30 tripolar: because it it doesn't compile, so the patches need to be updated, and I, ugh, forgot about it between 3.2-rc1 and 3.2.1. Feel free to submit a patch ;) Jan 17 16:36:41 russell--: topic? Jan 17 16:37:06 openwrt Jan 17 16:37:24 anything openwrt related in particular? Jan 17 16:37:36 just generally about the project Jan 17 16:37:39 ok Jan 17 16:37:44 raise profile and stuff Jan 17 16:37:48 sounds good Jan 17 16:38:06 i'll contact him, they seem to have something of an audience Jan 17 16:38:08 * Chocks wiggles nbd Jan 17 16:38:28 Chocks: what do you mean by "which u-boot just boots, not flashes"? Jan 17 16:38:54 anything that can be booted over the network can also be booted from flash Jan 17 16:38:57 as long as it fits Jan 17 16:39:24 understood. But I presume that the signature that it's given means u-boot just launches it rather than flashing it. this is of course Buffalo's crazy closed u-boot Jan 17 16:39:34 actually, not closed. I have source to it Jan 17 16:39:36 you mean when uploaded via tftp? Jan 17 16:39:40 yes Jan 17 16:40:26 I haven't looked closely at the problem, I was just wondering if you knew where to start looking Jan 17 16:40:38 i'm not sure how you want to handle the whole recovery vs normal boot thing Jan 17 16:41:24 I should just be able to tftp the recovery image, then have that automatically fetch and flash the rest via sysupgrade. I need a tftp factory recovery of course, in case it all goes pear shaped Jan 17 16:41:37 so it makes sense that that's the recovery image Jan 17 16:41:49 ok Jan 17 16:42:02 well, if you have a uImage, you can just put the firmware header in front of it Jan 17 16:42:26 I think there's a buffalo tool for that, right? if not, just some bytes Jan 17 16:42:38 yeah, look at the image makefile Jan 17 16:42:42 ok Jan 17 16:44:12 Chocks: btw. what buffalo are you talking about, the nh2? Jan 17 16:44:23 NH now, NH2 next Jan 17 16:44:27 ah, ok Jan 17 16:44:45 have you bugged buffalo about the sources for u-boot for the ag300h and the g300nh2 yet? Jan 17 16:44:54 those do not seem to be up on the gpl site Jan 17 16:44:56 difference on NH2 is different flash access on u-boot. assuming that kexec still doesn't work there Jan 17 16:45:20 no, I think that's probably a waste of time. I'll try and add to existing ar71xx u-boot we have Jan 17 16:45:29 nh2 is quite different, not just in the flash chip Jan 17 16:45:36 yes Jan 17 16:45:36 different soc, different wifi, basically different everything Jan 17 16:45:47 but I only need serial and flash for u-boot Jan 17 16:45:48 i'm not sure if it's a waste of time Jan 17 16:45:57 if you've tried bugging them and they don't respond, i can try to push for it through other channels Jan 17 16:46:07 higher up in the food chain Jan 17 16:46:42 we do actually have contacts at buffalo, but they are hard to get hold of. Jan 17 16:46:50 so, maybe Jan 17 16:47:00 I want to see what the flash problem is first. Jan 17 16:47:02 try official channel first, if that exists Jan 17 16:47:06 how is flash connected in NH2? Jan 17 16:47:11 spi Jan 17 16:47:18 it's actually two flash chips Jan 17 16:47:28 behind the same bus, with different chip select Jan 17 16:47:41 from what I've seen, u-boot should handle that pretty easily Jan 17 16:47:43 for write access you probably need to talk spi directly Jan 17 16:47:49 i'm not sure that works through memory mapped acces Jan 17 16:47:52 I just need read Jan 17 16:48:08 ok Jan 17 16:48:52 so, a few defines and crap vs waiting ages for buffalo vs my time constraints. I'd like to see the sources too on principle, but I'm not sure I have the time/energy Jan 17 16:49:01 ok Jan 17 16:52:21 btw. even read access probably will only work for one flash chip when in memory-mapped mode Jan 17 16:52:32 i don't think the spi controller will map more than one chip Jan 17 16:54:46 bbl Jan 17 18:43:23 russell--: ping Jan 17 20:01:49 glp: pong Jan 17 20:24:34 blogic: ping Jan 17 20:24:42 Delboy_: yes Jan 17 20:25:24 hi, do you maybe have a patch for lantiq for 3.2.1 kernel Jan 17 20:28:15 yes Jan 17 20:28:18 well Jan 17 20:28:26 how fast do you need it ? Jan 17 20:29:00 i am rebasing 3.2 on this series http://dev.phrozen.org/gitweb/?p=lantiq-queue.git;a=shortlog;h=refs/heads/upstream Jan 17 20:30:13 ok, because i wanted try out 3.2.1, but thought that maybe you already did a patch for it Jan 18 01:41:02 build #121 of cobalt is complete: Failure [failed compile_10] Build details are at http://buildbot.openwrt.org:8010/builders/cobalt/builds/121 **** ENDING LOGGING AT Wed Jan 18 02:59:56 2012