**** BEGIN LOGGING AT Wed Jan 25 02:59:56 2012
Jan 25 03:52:22 <Olipro> EqUaTe: if you don't mind not getting from a physical shop, try Rapid Electronics or Farnell
Jan 25 04:42:58 <builder___> build #90 of mpc52xx is complete: Failure [failed compile_10]  Build details are at http://buildbot.openwrt.org:8010/builders/mpc52xx/builds/90
Jan 25 06:45:42 <builder___> build #123 of cobalt is complete: Failure [failed compile_10]  Build details are at http://buildbot.openwrt.org:8010/builders/cobalt/builds/123
Jan 25 07:44:47 <CIA-84> juhosg * r29904 /trunk/target/linux/ar71xx/ (2 files in 2 dirs):
Jan 25 07:44:47 <CIA-84> ar71xx: TL-MR3020: fix GPIO polarity for button and switch
Jan 25 07:44:47 <CIA-84> This patch fixes the GPIO polarity for the button and the sliding
Jan 25 07:44:47 <CIA-84> switch. The buttons are not active low. "Pressed" and "Released"
Jan 25 07:44:47 <CIA-84> events are wrong without the patch.
Jan 25 07:44:47 <CIA-84> Signed-off-by: Christian Cier-Zniewski <c.cier@gmx.de>
Jan 25 07:44:49 <CIA-84> juhosg * r29905 /trunk/target/linux/ar71xx/ (5 files in 5 dirs): (log message trimmed)
Jan 25 07:44:49 <CIA-84> ar71xx: TL-MR3020: fix board detection, fix missing mandatory package and modify LED behaviour
Jan 25 07:44:49 <CIA-84> This patch fixes the board detection of the TL-MR3020.
Jan 25 07:44:49 <CIA-84> The LED behaviour has also been modified:
Jan 25 07:44:50 <CIA-84> * The WPS LED is the diag LED now.
Jan 25 07:44:50 <CIA-84> * A netdev trigger for the LAN LED has been added.
Jan 25 07:44:51 <CIA-84> The profile of the TL-MR3020 has been updated because the needed package
Jan 25 07:44:51 <CIA-84> juhosg * r29906 /trunk/package/mac80211/Makefile: (log message trimmed)
Jan 25 07:44:52 <CIA-84> mac80211: enable support for RaLink Rt53xx USB devices in rt2800usb
Jan 25 10:32:10 <EqUaTe> what jtag cables/software do people recommend? i believe i'm largely lacking in parallel ports..
Jan 25 10:52:28 <mystica555_> EqUaTe: there are always usb parallel ports ;)
Jan 25 10:52:50 <EqUaTe> :P there also appear to be a number of usb jtag devices.
Jan 25 10:53:05 <EqUaTe> might not bother at the moment.. lacking in funds, and serial should be enough
Jan 25 10:54:50 <mystica555_> ah well, there are quite a number of options open when you decide to
Jan 25 10:55:00 <EqUaTe> if i need to
Jan 25 13:09:42 <CIA-84> acinonyx * r29907 /packages/lang/php5/Makefile: [packages] php5: Don't fail on non-existing file removal
Jan 25 13:23:30 <CIA-84> acinonyx * r29908 /packages/libs/mysql/Makefile: [packages] mysql: Add missing dependencies
Jan 25 13:36:34 <luka12345> jow_laptop: ping
Jan 25 15:09:35 <jow_laptop> luka12345: pong
Jan 25 15:23:15 <builder___> build #120 of ps3 is complete: Failure [failed compile_6]  Build details are at http://buildbot.openwrt.org:8010/builders/ps3/builds/120
Jan 25 15:56:21 <luka12345> jow_laptop: ok, i'll try to make it work tonight
Jan 25 15:57:40 <Chocks> kids
Jan 25 17:14:44 <mirko> hmm.. the MTD_PARTITIONS and MTD_CONCAT options got removed but documentation about alternatives is rare.. could anybody point to me sth. recent and useful?
Jan 25 17:28:20 <Chocks> mirko: can you explain further?
Jan 25 17:30:54 <mirko> Chocks: the options MTD_PARTITIONS and MTD_CONCACT got purged
Jan 25 17:31:13 <mirko> and the only information i find are commit-msgs within the git
Jan 25 17:32:10 <mirko> and now i'm unsure what implications this merge has: MTD_PARTITIONS is now set implicitly? MTD_PARTITIONS is deprecated and DEVICE-TREEE should be used now?
Jan 25 17:35:50 <KanjiMonster> mirko: MTD_PARTITIONS is now always active (since ugh... 3.0 IIRC)
Jan 25 17:35:55 <Chocks> right yes, the MTD_PARTITIONS #ifdef was stripped from 3.x platform files
Jan 25 17:36:40 <mirko> it's still present inside the defconfig-files in 3.2
Jan 25 17:36:53 <mirko> as well as some symbols still have it within their SELECT statement
Jan 25 17:36:57 <mirko> so i got confused
Jan 25 17:37:05 <mirko> KanjiMonster: Chocks: thanks for calrification
Jan 25 17:37:13 <mirko> *clarification
Jan 25 17:37:18 <Chocks> not sure I did, but np ;-)
Jan 25 20:58:44 <russell--> weird, ath5k didn't get built into the image and doesn't show up in make menuconfig's kernel / wireless-drivers section
Jan 25 20:59:06 <russell--> unpatched r29884
Jan 25 21:00:41 <philipp64|laptop> russell--: if you figure out why it crashes on initialization, let me know. I've given up on using ath5k because all of the ath9k churn has destabilized it (I don't think anyone tests their changes on ath5k hardware as well as ath9k).
Jan 25 21:12:07 <btsimonh> I'm about to create a flash layout for a new router.  is http://wiki.openwrt.org/doc/techref/flash.layout a good starting point?
Jan 25 21:19:38 <russell--> something between r29834 and r29844 made it disappear
Jan 25 21:20:42 <Chocks> bt: yes
Jan 25 21:20:59 <Chocks> just make sure you don't overwrite anything needed by bootloader, or other parts of the system
Jan 25 21:21:02 <Chocks> like the art stuff
Jan 25 21:21:52 <nbd> russell--: so grep ath5k .config shows nothing?
Jan 25 21:21:57 <nbd> russell--: what device?
Jan 25 21:22:02 <nbd> i mean what target platform?
Jan 25 21:22:44 <Chocks> that page is a bit shouty(!)
Jan 25 21:23:48 <jow_laptop> you don't like pink paths?
Jan 25 21:23:57 <russell--> atheros
Jan 25 21:24:19 <Chocks> more the excessive !s, and pointless Caps Of Kernel, etc
Jan 25 21:24:19 <russell--> the test was looking at the packages for kmod-ath5k
Jan 25 21:24:36 <russell--> in build-all-packages builds
Jan 25 21:25:02 <russell--> russell@frieda[/src/openwrt] $ ls -al bin-r29834-atheros/atheros/packages | grep ath5k
Jan 25 21:25:06 <russell--> -rw-r--r-- 1 russell staff    80858 Jan 19 21:42 kmod-ath5k_2.6.37.6+2011-12-01-2_atheros.ipk
Jan 25 21:25:12 <Chocks> and syntax colouring the shell code is a bit excessive
Jan 25 21:25:20 <russell--> russell@frieda[/src/openwrt] $ ls -al bin-r29844-atheros/atheros/packages | grep ath5k
Jan 25 21:25:27 <russell--> *nothing*
Jan 25 21:27:06 <russell--> russell@frieda[/src/openwrt] $ grep ath5k bin-r29844-atheros/.config
Jan 25 21:27:06 <russell--> # CONFIG_TARGET_brcm47xx_Atheros-ath5k is not set
Jan 25 21:27:06 <russell--> # CONFIG_TARGET_ar71xx_generic_ath5k is not set
Jan 25 21:27:06 <russell--> # CONFIG_TARGET_ixp4xx_generic_Atheros-ath5k is not set
Jan 25 21:27:06 <russell--> CONFIG_DEFAULT_kmod-ath5k=y
Jan 25 21:27:16 <russell--> but it doesn't build, apparently
Jan 25 21:27:38 <russell--> (that grep for r29834 looks identical)
Jan 25 21:27:51 <russell--> (oldconfig'd from one to the other)
Jan 25 21:30:09 <nbd> and you're sure you don't have any messed up local changes in your tree?
Jan 25 21:31:39 <russell--> yeah, it's pristine afaik
Jan 25 21:32:22 <russell--> git diff returns nothing, a fresh make menuconfig with no .config or .config.old doesn't show ath5k either
Jan 25 21:33:13 <russell--> that's at r29884
Jan 25 21:40:08 <nbd> try rm -rf tmp
Jan 25 21:40:12 <nbd> then make oldconfig
Jan 25 21:40:15 <nbd> maybe that helps
Jan 25 21:41:44 <russell--> okay
Jan 25 21:42:08 <russell--> i do see differences between my oldconfig'd .config and a fresh one
Jan 25 21:42:32 <russell--> e.g. the new one uses the LINARO gcc
Jan 25 21:42:57 <nbd> the linaro gcc has been default for quite some time now
Jan 25 21:43:15 <russell--> yeah, actually, it's just a new CONFIG line
Jan 25 21:43:25 <russell-->  CONFIG_GCC_VERSION="4.5-linaro"
Jan 25 21:43:25 <russell-->  CONFIG_GCC_VERSION_4_5=y
Jan 25 21:43:25 <russell--> +CONFIG_GCC_VERSION_4_5_LINARO=y
Jan 25 21:43:31 <nbd> ah
Jan 25 21:43:40 <nbd> should be ok then
Jan 25 21:43:53 <Chocks> last time I tried 4.6, swconfig was crashing.  perhaps that's been fixed now
Jan 25 21:45:27 <Chocks> any "make menuconfig" in one of my trees hangs forever.  suggestions on what's wrong
Jan 25 21:45:54 <btsimonh> for a flash layout, i've gone for http://pastebin.com/Pi8MBdhi
Jan 25 21:46:04 <russell--> the rm -rf tmp seems to have fixed it
Jan 25 21:46:08 <nbd> ok
Jan 25 21:46:19 <Chocks> do you really have an art section on the existing setup?
Jan 25 21:47:02 <btsimonh> yes, u-boot & config is in a nor flash, this is only for art & linux.
Jan 25 21:47:09 <Chocks> I'm not certain that you need to create one, but I might be wrong.  that's atheros wifi stuff
Jan 25 21:47:16 <Chocks> ok
Jan 25 21:48:09 <btsimonh> yes, art is at start of original nand.  original layout was: http://pastebin.com/98cMUkrR
Jan 25 21:48:25 <Chocks> very good
Jan 25 21:48:29 <btsimonh> is my kernel big enough?
Jan 25 21:49:30 <Chocks> depends what you end up with, but it should squeeze into 1MB with lzma
Jan 25 21:50:17 <Delboy_> btsimonh: why didn't you create linux partition instead of kernel and rootfs?, it cerates them automaticly if you use linux part
Jan 25 21:50:49 <btsimonh> and to date, I've been ramdisk.  what image do i burn where? (having unchecked ramdisk)
Jan 25 21:51:00 <btsimonh> (linux part? - pls explain)
Jan 25 21:51:10 <Chocks> what you burn depends upon what your boot loader expects
Jan 25 21:51:59 <btsimonh> it's older u-boot.  to date, ramdisk has been fine.  had to add lzma to u-boot as that seems all that is made?
Jan 25 21:52:21 <Chocks> you can do gz too
Jan 25 21:52:25 <Chocks> bigger tho
Jan 25 21:52:54 <Delboy_> when you create linux, there is a code that automatically detects end of kernel and it creates start of rootfs where kernel ends
Jan 25 21:53:31 <Delboy_> to use the flash space maximally
Jan 25 21:53:40 <btsimonh> so in menuconfig, i have ramdisk unchecked, cpio.gz checked, and squashfs checked, no image options available.
Jan 25 21:54:18 <btsimonh> I get a uImage, a rootfs.sq.., various vmlinux.  no combined image?
Jan 25 21:54:18 <Chocks> you might to have look at the image/Makefile for your platform to see what you are making
Jan 25 21:54:54 <CIA-84> swalker * r29909 /packages/net/ddns-scripts/ (Makefile files/usr/lib/ddns/services): [packages] ddns-scripts: add ovh.com (#7340, thanks buggerone), dnsomatic.com (#8804), 3322.org & namecheap.com (#9286, thanks zjhzzyf)
Jan 25 21:55:08 <Delboy_> squashfs is kernel+rootfs
Jan 25 21:55:24 <Delboy_> i use that for ma gigaset
Jan 25 21:56:30 <Delboy_> http://pastebin.com/BsiUFGMc this is how it looks when you use linux partiiton
Jan 25 21:56:37 <btsimonh> Del: should be the same for me then... since I stole the gigiaset image building :).
Jan 25 21:57:00 <btsimonh> So just dump rootfs.squashfs into 'firmware'?
Jan 25 21:57:59 <btsimonh> (but then the linux/root boundary will not be right...  i'll check the gigiaset mach file for flash layout... back in a moment).
Jan 25 21:58:14 <Delboy_> yes but it doesn't matter
Jan 25 21:59:06 <btsimonh> if i use 'linux' as the partition :).  i see now...
Jan 25 21:59:18 <Delboy_> rootfs_data is where config and stuff is sitored
Jan 25 22:00:42 <btsimonh> so as a flash layout, this is simpler: http://pastebin.com/2GSPpnfQ
Jan 25 22:01:07 <luka12345> btsimonh: you shour remove whole flash
Jan 25 22:01:09 <Delboy_> also make sure you create a backup of that atheros eeprom, because if you erase it accidentally than you don't have call data for your board
Jan 25 22:01:12 <Delboy_> yes that is ok
Jan 25 22:01:38 <luka12345> Delboy_: why do you think wholeflash is needed?
Jan 25 22:02:34 <Delboy_> well, it is not needed :)
Jan 25 22:03:01 <luka12345> this is part of the bootlog showing on older trunk: http://pastebin.com/VDeHLB88
Jan 25 22:03:20 <luka12345> you can see here how it gets recognized...
Jan 25 22:05:06 <btsimonh_> got it.  will try now :)
Jan 25 22:08:01 <Delboy_> also you don't need that boardid and mach-bthomehubv2b.h stuff, that is for sx76x because it has no eeprom in flash
Jan 25 23:31:39 <Delboy_> btsimonh: now you need to get wlan working :)
Jan 25 23:32:37 <btsimonh> yep.  if i could get wired back.  just going to rebuild with linux renamed to linux1 to see if it comes back when it's not trying to load a rootfs
Jan 25 23:33:46 <btsimonh> for yours, art area is in code.  for mine, i guess there should be somehting to read it from 'art' mtd?
Jan 25 23:34:09 <Delboy_> yeah check out arv45xx mach
Jan 25 23:35:01 <btsimonh> ta. god i hate it when i have to re-power the router  means i have to get up :)
Jan 25 23:36:17 <Delboy_> :)
Jan 25 23:38:39 <Delboy_> but why on your board there is no pci id printed for wlan chip?
Jan 25 23:39:02 <btsimonh> ??
Jan 25 23:39:05 <Delboy_> i mean inside kernel log you posted
Jan 25 23:39:56 <btsimonh> :)  Not sure I follow. I do have the register madfi commentedat the moment?
Jan 25 23:40:41 <btsimonh> I think i will need to clean and rebuild.  swapping between ramdisk and non-ramdisk and back does not seem to work well... :(
Jan 25 23:41:06 <btsimonh> (trying a full make)
Jan 25 23:41:13 <Delboy_> this is what i mean (from my kernel log) http://pastebin.com/iam4RkVG
Jan 25 23:41:53 <Delboy_> whell does it work now with squashfs?
Jan 25 23:43:55 <btsimonh> seems you have more PCI devices?  or are you saying my wifi is missing?
Jan 25 23:44:54 <btsimonh> (just reobooting to ramdisk now)
Jan 25 23:44:55 <Delboy_> well, yes but other one is some serial controller for atheros chip, dunno what it is used for
Jan 25 23:46:26 <btsimonh> back to working ramdisk with br-lan up.  now back to squash with more of a rebuild than last time :)
Jan 25 23:47:47 <btsimonh> (i've been brave and uncommented all the wifi stuff).
**** BEGIN LOGGING AT Wed Jan 25 23:56:08 2012
Jan 25 23:59:45 <Delboy_> does it find mac address in init?
Jan 25 23:59:45 <Delboy_> that message before boardid is...
Jan 26 00:00:04 <btsimonh> eth0 gets my default mac (hardcoded)
Jan 26 00:02:30 <btsimonh> i don't get 'device eth0 entered promiscuous mode'
Jan 26 00:03:05 <Delboy_> did you get br-lan with squashfs before?
Jan 26 00:03:45 <luka12345> btsimonh: can i see your network config and device mach?
Jan 26 00:04:37 <btsimonh> no, only with ramfs.  but...  also did not get it with target/install when swapping back to ram based... (luka, getting now.. which netwoirk config file?)
Jan 26 00:05:05 <luka12345> /etc/config/network
Jan 26 00:05:26 <btsimonh> complete mach.c file: http://pastebin.com/XDhqdRka
Jan 26 00:06:56 <btsimonh> '/etc/config/network' - i don't have one!!!! why does it work for ram based?
Jan 26 00:07:46 <luka12345> you can boot the board right?
Jan 26 00:08:37 <luka12345> and you dont have nothing this file '/etc/config/network' when the board is booted
Jan 26 00:08:40 <luka12345> ?
Jan 26 00:09:13 <btsimonh> yep.  to serial.  and netowkr is fine until i go squash.  I have a feeling that somewhere in lantiq target there is a 'rm network' - just searching...
Jan 26 00:10:17 <btsimonh> :) target\linux\lantiq\base-files.mk = 'rm -f $(1)/etc/config/network'
Jan 26 00:10:22 <luka12345> yes
Jan 26 00:10:24 <luka12345> i know
Jan 26 00:10:49 <btsimonh> :) ????  WHY... :)  is it meant to go get one somewhere else?
Jan 26 00:10:54 <luka12345> network is set up in: target/linux/lantiq/base-files/etc/uci-defaults/network
Jan 26 00:11:37 <luka12345> i'm doing some nasty overwriting
Jan 26 00:12:05 <luka12345> http://pastebin.com/78b47Nai
Jan 26 00:12:22 <luka12345> you could try to do something simmilar
Jan 26 00:12:51 <luka12345> or patch that file so your board is recognized
Jan 26 00:13:07 <luka12345> but first you should try to create /etc/config/network manualy
Jan 26 00:13:20 <luka12345> and then do /etc/init.d/network restart
Jan 26 00:13:23 <luka12345> and see what happens
Jan 26 00:13:49 <philipp64|laptop> what's Nauls' handle?
Jan 26 00:14:31 <Chocks> Naulls
Jan 26 00:14:35 <Chocks> tis me
Jan 26 00:15:58 <btsimonh> do i read it right; if board was GIGASX76X, it would do something?
Jan 26 00:16:20 <philipp64|laptop> ah, thought it might be. what eglibc issues do you have?
Jan 26 00:16:36 <Chocks> 2.14 doesn't build despite latest patches
Jan 26 00:16:53 <Chocks> otherwise, nothing that isn't addressed in my patches of 6 months ago, albeit in sometimes hacky fashion
Jan 26 00:17:01 <luka12345> btsimonh: yes
Jan 26 00:17:28 <luka12345> but it's wierd that this is not executed
Jan 26 00:17:28 <luka12345> set_interface_loopback
Jan 26 00:17:28 <luka12345> set_interface_lan 'eth0'
Jan 26 00:17:37 <luka12345> it's before GIGASX76X definition
Jan 26 00:17:53 <luka12345> so try to create the file manually
Jan 26 00:18:12 <btsimonh> i got loopback, and eth0, just normally i have br-lan which has the actual ip...
Jan 26 00:18:37 <luka12345> ok
Jan 26 00:18:50 <luka12345> and you did not have that before?
Jan 26 00:19:27 <luka12345> no sorry, i misread
Jan 26 00:19:57 <luka12345> if you dont have /etc/config/network that is the reason why you are missing br-lan
Jan 26 00:21:23 <btsimonh> i can't understand why i get it with ramboot :).  I think now a tomorrow issue.
Jan 26 00:21:58 <btsimonh> I really need to ensure that the root image contains everything, as I'm hoping to get this as a patch for others,
Jan 26 00:22:09 <btsimonh> and they won't want to solder on serial :)
Jan 26 00:22:44 <luka12345> i gave up that model long ago
Jan 26 00:23:36 <Delboy_> btsimonh: about that pci, maybe you need to set that ltq_pci_data.clock to PCI_CLOCK_EXT
Jan 26 00:24:30 <btsimonh> i've not checked the clocks yet.  i'm pretty sure i will need somehting special -; everything else has been special.
Jan 26 00:25:07 <Delboy_> check inside mach-arv45xx.c, there some boards have clock and irq different
Jan 26 00:25:41 <btsimonh> on the art nand region, i don't think the code you pointed me to helped - nand is not loaded at the point the avr code copies.  there must be some board which load art after nand.
Jan 26 00:27:24 <Delboy_> thany copy it to nor flash :D
Jan 26 00:27:42 <philipp64|laptop> Chocks: for what architecture?
Jan 26 00:27:52 <btsimonh> it's an option :)
Jan 26 00:27:57 <Chocks> ar71xx and ramips
Jan 26 00:28:30 <philipp64|laptop> email me a list and I'll have a look.
Jan 26 00:28:54 <Chocks> of?
Jan 26 00:29:28 <philipp64|laptop> links to the bugs and/or patches?
Jan 26 00:29:45 <Delboy_> but on gigaset, nor also isn't loaded but it still finds mac and boardid, or for nand flash it must be loaded?
Jan 26 00:30:09 <Chocks> https://dev.openwrt.org/ticket/9483
Jan 26 00:31:18 <btsimonh> can't read from NAND directly like you can with NOR :)
Jan 26 00:34:32 <Delboy_> look at https://dev.openwrt.org/browser/trunk/target/linux/lantiq/base-files/etc/hotplug.d/firmware/10-rt2x00-eeprom
Jan 26 00:34:33 <btsimonh> yep tomorrow guys.  thanks for all your help :)
Jan 26 00:37:55 <btsimonh> :) good find.  stored that one away for later. will need to find out how to invoke it right.  The openWRT scripting is hell from a windows programmer's perspective :)
Jan 26 00:44:53 <btsimonh> paaa - it's never going to work.  just did df, and my overlay is ro.  need to test my flash better before I worry about scripts..... i think it thinks my erase blocks are larger than they are...
Jan 26 00:46:39 <luka12345> btsimonh: you should get Delboy_, blogic and me some boards ;)
Jan 26 00:47:29 <btsimonh> Local car boot should start up again after winter shutdown.  If i find some I will buy them :)
Jan 26 00:47:49 <btsimonh> goodnight for today.....
Jan 26 00:48:27 <luka12345> btsimonh: see you
**** ENDING LOGGING AT Thu Jan 26 02:59:57 2012