**** BEGIN LOGGING AT Fri Apr 25 02:59:57 2008 Apr 25 13:02:22 thepeople * r10935 /trunk/target/linux/rdc/config-2.6.24: add missing symbol Apr 25 13:59:28 thepeople * r10936 /packages/libs/libreadline/patches/ (. 100-fix_avr32_compile): fix compiling on avr32 Apr 25 14:46:00 thepeople * r10937 /trunk/target/linux/avr32/config-default: add missing symbols Apr 25 16:44:21 * Nix waves to nbd Apr 25 17:08:01 nbd: you still around? I am trying to convince my u-boot to build under openwrt Apr 25 17:54:06 thepeople * r10938 /packages/libs/lzo/patches/ (. 100-fix_avr32_compile): fix compile on avr32 Apr 25 18:34:50 does anyone know why this exists https://dev.openwrt.org/browser/trunk/package/ifxmips_adsl?rev=9900 but this doesnt work svn co -r 9900 https://svn.openwrt.org/openwrt/trunk/package/ifxmips_adsl Apr 25 18:48:45 o.0 weird... I can definitely confirm. Apr 25 19:01:20 Nix: try svn co -r 9901 ? Apr 25 19:01:28 but yeah thats weird Apr 25 19:02:13 Weedy: tried that.. doesnt work Apr 25 19:02:37 * Nix pokes blogic Apr 25 19:07:02 drwxr-xr-x 3 weedy users 4096 2008-04-25 15:06 ifenslave Apr 25 19:07:02 drwxr-xr-x 4 weedy users 4096 2008-04-25 15:06 ifxmips_adsl Apr 25 19:07:02 drwxr-xr-x 4 weedy users 4096 2008-04-25 15:06 iproute2 Apr 25 19:07:54 svn co -r 9900 https://svn.openwrt.org/openwrt/trunk/package/ Apr 25 19:08:14 A package/ifxmips_adsl Apr 25 19:08:16 ... Apr 25 19:08:20 A package/ifxmips_adsl/Makefile Apr 25 19:08:25 i can grab it Apr 25 19:09:18 Weedy: yeah, that works fine, but checking out ifxmips_adsl directly does not Apr 25 19:09:27 but grabbing ifenslave directly does. Apr 25 19:09:40 Bartman007: yeah it doesnt Apr 25 19:09:45 which is weird Apr 25 19:46:30 Bartman007: ok. thanks.. I checked out the whole packages dir and got it Apr 25 19:51:17 ok. I need help here Apr 25 19:51:37 I have some conditional config menus Apr 25 19:51:40 like: Apr 25 19:51:41 config IFX_CONFIG_FLASH_SIZE Apr 25 19:51:41 depends UBOOT_BUILD_ifx_mips Apr 25 19:51:42 int "IFX Flash Size" Apr 25 19:51:44 default 4 Apr 25 19:51:50 which works great Apr 25 19:52:33 but if UBOOT_BUILD_ifx_mips=y I also want to just hardcode things like: IFX_CONFIG_CPU="DANUBE" Apr 25 19:52:46 does anyone know the correct Makefile syntax for that? Apr 25 20:49:43 noz * r10939 /trunk/target/linux/brcm47xx/ (files-2.6.25/ patches-2.6.25/): First step to upgrade of brcm47xx to kernel version 2.6.25 Apr 25 20:52:11 \o/ Apr 25 20:54:47 First step - really only the very first step! Apr 25 20:55:30 Just, I had to commit because I was moving things in subversion, and it wouldn't let me do anything else to the files I'd moved until I committed Apr 25 20:56:08 heh ok Apr 25 20:56:14 all the more reason to use git. Apr 25 20:57:01 oh yes Apr 25 20:57:03 :) Apr 25 20:59:43 Yes - I really must learn git. Apr 25 21:01:15 it will allow you to get 2.6.25 fully working then push the entire series of commits to the svn repo. Apr 25 21:01:35 and watch Kaloz cringe as bots rape his server for the commit info. Apr 25 21:02:02 OK. How difficult is it set up initially? (Sucker) Apr 25 21:02:20 not hard, just install git, and clone nbd's repo Apr 25 21:02:43 now you've got the entire svn history in less space than a single svn checkout Apr 25 21:02:47 Best to do alongside my svn working dir to start with, don't you think? Apr 25 21:02:58 yeah Apr 25 21:04:23 OK. Will you stick around for hand-holding if I need it, until it's going? Apr 25 21:05:08 until you need it the most, then I shall flee Apr 25 21:05:30 yeah, I'm still sick, so I've got absolutely nothing better to do Apr 25 21:05:48 :) Wow - 90k objects Apr 25 21:06:51 * thepeople prods Bartman007 to test firewall stuff Apr 25 21:06:59 oh, right. Apr 25 21:07:02 hehe. Apr 25 21:07:03 being you are sick and all :-P Apr 25 21:09:51 thepeople: you crashed X11 Apr 25 21:10:43 * thepeople feels special :-P Apr 25 21:10:45 OK, cloned the repo. I guess I need to start my own branch now? git branch my-devel Apr 25 21:11:25 noz: you can do `git co -b my-devel' to create a branch and check it out at the same time Apr 25 21:11:53 and I find it more convenient to create a branch for each task, so I'd name this brcm-2.6.25 or something like that Apr 25 21:12:26 then you can pop between tasks without them conflicing Apr 25 21:12:47 Indeed. I had a much longer name, but yours is better Apr 25 21:13:00 IIRC nbd avoids that method and still uses seperate checkouts for different work Apr 25 21:13:37 'git: 'co' is not a git-command' Apr 25 21:13:38 mostly because using multiple different configs in one tree is not yet convenient enough Apr 25 21:13:48 noz: oops, I've aliased it. Apr 25 21:15:12 here's my alias section from ~/.gitconfig I really need to expand it, but it's a start: http://openwrt.pastebin.ca/997805 Apr 25 21:15:34 OK. I'm working in a git branch. When I want to rename things, do i just do so, or do I need the equivalent of svn mv? Apr 25 21:16:10 if you do it manually, you have to use git-add and git-rm Apr 25 21:16:12 there's also a git-mv Apr 25 21:16:29 but if you rename things manually, git is smart enough to use rename tracking in svn anyways Apr 25 21:16:40 OK. I'll look into those. Thanks Apr 25 21:17:27 and fyi, every command can be accessed via git-foo or git foo Apr 25 21:18:12 and you can use either `man git-foo' `git-foo --help' or `git foo --help' to access the man page. Apr 25 22:05:28 thepeople have you fixed the avr config script on readline? Apr 25 22:06:18 matteo: yes, I didn't releaze that it wasn't in the level dir becuase there was a config.sub there that was correct Apr 25 22:08:14 matteo: I ran into a error with uclibc++ that I haven't looked at yet Apr 25 22:18:13 Is there an analog of svn cp when using git? Does it make sense to preserve the history of a file in that way? Apr 25 22:19:14 IIRC git figured that out automatically Apr 25 22:19:28 the two files will have the same sha1 sum, therefore they are identical Apr 25 22:19:42 figure* Apr 25 22:20:08 But then if I change one? Do I have to git commit (or git add) first? Apr 25 22:21:49 I also think it compares text files, I seem to recall seeing it detect renames and state that the files are 54% similar (or maybe it was 54% confidence) when I refreshed a patch for a new version Apr 25 22:22:20 not 100% sure though Apr 25 22:45:25 Ugh! I'm beginning to wonder how much I've bitten off with the 2.6.25 update of brcm. A lot of it has made it upstream, but in a different form. Lots of diffs => high possibility of missing something... Apr 25 22:50:43 noz: just try to get it working somehow. i'll review it later Apr 25 22:51:08 That sounds like a blank cheque! Apr 25 22:52:22 hehe Apr 25 22:52:33 review != fix carnage Apr 25 22:52:51 I can do carnage - just wait and see Apr 25 22:53:12 yeah review might just be preparation for flame ;) Apr 25 22:53:48 nah, don't worry. i want to see this working properly with newer kernels as well Apr 25 22:53:57 preferably in a form that can be submitted upstream Apr 25 22:54:11 so of course i'm going to help Apr 25 22:54:17 Sure. Apr 25 22:54:27 I wan't doubting that you would! Apr 25 22:54:35 *wasn't Apr 25 22:54:37 upstream to OE so we can just be an overlay of patches? ;-) Apr 25 22:56:51 yeah, let's make diffs of different versions of meta-diffs Apr 25 23:41:18 <[mbm]> . Apr 26 02:26:53 thepeople * r10940 /packages/libs/cyrus-sasl/patches/200-fix_avr32_compile: fix avr32 compiling **** ENDING LOGGING AT Sat Apr 26 02:59:57 2008