**** BEGIN LOGGING AT Wed Nov 26 02:59:57 2008
Nov 26 08:40:53 <CIA-14> juhosg * r13356 /trunk/target/linux/ar71xx/files/arch/mips/ar71xx/prom.c: [ar71xx] fix AP83 boot failure
Nov 26 10:16:35 <russell_> missing backslash in packages/utils/owfs/Makefile
Nov 26 10:16:51 <xMff> russel_: tell me where I'll fix it asap
Nov 26 10:17:10 <russell_> define Build/Configure
Nov 26 10:17:15 <xMff> ok
Nov 26 10:17:32 <russell_> in the line before ./configure
Nov 26 10:17:52 <russell_> otherwise, all those nice environment variables get applied to nothing, i think.
Nov 26 10:17:57 <xMff> yeah and one too much at the end
Nov 26 10:20:07 <russell_> not tested, just found that part choked during a compile, that looks like likely problem
Nov 26 10:21:20 <russell_> the backslash after $(CONFIGURE_ARGS) is probably okay
Nov 26 10:21:47 <xMff> ty, fix committed
Nov 26 10:24:14 <xMff> you're right that one on the end was right, overlooked the closing ) at the end
Nov 26 10:26:28 <russell_> $ ./scripts/feeds update packages
Nov 26 10:26:28 <russell_> Updating feed 'packages' from 'git://nbd.name/packages.git' ...
Nov 26 10:26:28 <russell_> Usage: /usr/bin/git-clone [--template=<template_directory>] [--use-separate-remote] [--reference <reference-repo>] [--bare] [-l [-s]] [-q] [-u <upload-pack>] [--origin <name>] [-n] <repo> [<dir>]
Nov 26 10:26:45 <russell_> that doesn't look right
Nov 26 10:28:24 <russell_> $ cd feeds/
Nov 26 10:28:24 <russell_> russell@dalles:/src/openwrt/feeds$ git clone git://nbd.name/packages.git packages
Nov 26 10:28:25 <xMff> it does  git clone --depth 1 $src ./feeds/$name
Nov 26 10:28:30 <russell_> that works
Nov 26 10:29:01 * russell_ trying that manually
Nov 26 10:29:32 <xMff> maybe it does not like --depth ?
Nov 26 10:29:49 <russell_> yeah
Nov 26 10:30:03 <russell_> $ git --version
Nov 26 10:30:03 <russell_> git version 1.4.4.4
Nov 26 10:30:44 <russell_> $ cat /etc/debian_version
Nov 26 10:30:45 <russell_> 4.0
Nov 26 10:31:04 <xMff> git 1.5.4.3 here
Nov 26 10:32:05 <russell_> mine, i guess is a depthless git
Nov 26 10:32:15 <xMff> heh
Nov 26 10:32:58 <xMff> you could just fix feeds for yourself, not sure why nbd added the depth param
Nov 26 10:33:31 <russell_> easy to work around, but i wonder if more robustness might be warranted
Nov 26 10:34:29 <Bartman007> xMff: creates a shallow clone (limited, or in this case, no history)
Nov 26 10:34:44 <xMff> yeah, just looked it up in the manpage
Nov 26 10:36:43 <Bartman007> a recent version of git is in etch-backports
Nov 26 10:38:54 <russell_> what is the canonical script/feeds command sequence on a fresh checkout?
Nov 26 10:39:09 <russell_> i've seen a few variants
Nov 26 10:41:12 <Bartman007> take a look at the "package/symlinks" target in $(TOPDIR)/Makefile
Nov 26 10:41:51 <russell_> ah yes, thank you.
Nov 26 10:45:08 * russell_ lauches off a fresh 'Select all' build try for the wgt
Nov 26 10:45:15 <russell_> launches
Nov 26 10:45:56 <loswillios> russell_: have you tried .27 on your WGT?
Nov 26 10:46:30 <russell_> not yet (or recently) i forget if i was trying .27 on x86 or the wgt a week or so back.
Nov 26 10:46:49 <loswillios> ok
Nov 26 10:47:10 <loswillios> did you run into https://dev.openwrt.org/ticket/4178 by any chance? (not .27 related)
Nov 26 10:47:47 <russell_> i have not
Nov 26 10:48:34 <russell_> but, i'm just getting back to the wgt after some neglect in favor of the soekris problems
Nov 26 10:50:32 <russell_> there is a 2.6.27 related patch sitting in the openwrt-devel mailing list though, that has not been applied yet.
Nov 26 10:51:36 <russell_> Subject: [OpenWrt-Devel] esfq patch to compile on 2.6.27
Nov 26 10:52:58 * russell_ goes to sleep to let 'make' do it's thing
Nov 26 10:53:09 <russell_> its
Nov 26 15:17:19 <x-alina> xMf1: Fuege mkstemp() zu Lua POSIX hinzu um sicher temporaere Dateien erzeugen zu koennen. (Symlink-Attacken, usw...)
Nov 26 16:41:41 <x-alina> xMff: Chicken and Egg: luci.sauth needs luci.sys and luci.sauth needs luci.sys. ("sauth.lua:22: loop or previous error loading module 'luci.sys'") Laesst ich das in Lua irgendwie loesen?
Nov 26 16:52:33 <xMff> x-alina: wieso brauch sys jetzt sauth?
Nov 26 16:53:40 <xMff> die dürfen nicht zusammenhängen
Nov 26 16:53:53 <xMff> wär toll wenn man sys auch ohne webfoo nehmen könnte
Nov 26 16:59:43 <x-alina> xMff: das Problem ist das checkpassword in sys ist und dieses braucht jetzt die challage methode aus sauth um die reponse zu ueberpruefen
Nov 26 17:01:47 <xMff> x-alina: das ist schlecht
Nov 26 17:02:09 <xMff> x-alina: luci.sys ist per design standalone, entferne bitte die dependencies nach sauth
Nov 26 17:03:03 <x-alina> xMff: checkpassord nach sauth?
Nov 26 17:03:15 <xMff> nein
Nov 26 17:03:58 <x-alina> xMff: Was dann?
Nov 26 17:04:53 <xMff> k.A.
Nov 26 17:05:05 <xMff> du hast mehr Überblick über deinen Code da
Nov 26 17:08:59 <x-alina> xMff: Alles wuerde darauf hinnaus laufen code doppelt zu haben.
Nov 26 17:10:03 <xMff> dann machst du was fundamental falsch
Nov 26 17:17:29 <x-alina> xMff: checkpasswd muesste aus sys raus und alles waere gut. lagere jetzt das ermitteln der challange in den aufzurufenden code aus, mehr code aber immer noch das eleganteste, was ohne einen kompletten umbau moeglich ist.
Nov 26 17:43:56 <CIA-14> florian * r13365 /trunk/package/kernel/modules/other.mk: ssb module config symbols fixup
Nov 26 18:16:25 <CIA-14> florian * r13366 /trunk/target/linux/generic-2.6/patches-2.6.27/200-sched_esfq.patch:
Nov 26 18:16:25 <CIA-14> the 'dev' member of the structure was removed in 2.6.27 (see:
Nov 26 18:16:25 <CIA-14> http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=5ce2d488fe039ddd86a638496cf704df86c74eeb),
Nov 26 18:16:25 <CIA-14> therefore use the qdisc_dev() function instead.
Nov 26 18:16:26 <CIA-14> Signed-off-by: Russell Senior <seniorr@aracnet.com>
Nov 26 18:39:30 <CIA-14> florian * r13367 /trunk/target/linux/generic-2.6/ (config-2.6.24 config-2.6.25 config-2.6.26 config-2.6.27): Add missing CONFIG_SSB_SLIENT config symbol
Nov 26 18:41:07 <CIA-14> florian * r13368 /trunk/target/linux/brcm63xx/ (80 files in 16 dirs): Flatten brcm63xx patches, should make our life easier to patch files now ;)
Nov 26 19:43:03 <russell_> build failure in collectd
Nov 26 19:44:15 <xMff> russell_: pastebin error, I'll look
Nov 26 19:44:25 <russell_> k, coming ....
Nov 26 19:47:26 <russell_> http://pastebin.ca/1267843
Nov 26 19:48:57 <xMff> hmm okay, iptables... strange it builds here
Nov 26 19:49:16 <russell_> that's for wgt
Nov 26 19:49:50 <russell_> circa r13359
Nov 26 19:49:55 <xMff> brcm-2.4 here and atheros (fonera) yesterday
Nov 26 19:57:40 <xMff> russell_: solution for now is to disable the iptables plugin
Nov 26 19:57:56 <xMff> russell_: will look in 1, 2 hours
Nov 26 19:58:19 <russell_> danke
Nov 26 20:00:55 <CIA-14> juhosg * r13369 /trunk/target/linux/ar71xx/files/ (7 files in 3 dirs): [ar71xx] ag71xx: introduce SoC specific fuctions for DDR flush and PLL setup
Nov 26 20:43:22 <CIA-14> ryd * r13370 /packages/Xorg/lib/imlib2/Makefile: Add missing dependency to imlib2
Nov 26 21:13:47 <russell_> oh, yeah.  another build failure in coreutils
Nov 26 21:13:51 <russell_> i've seen this before
Nov 26 21:15:24 <russell_> http://pastebin.ca/1267931
Nov 26 22:49:31 <russell_> still bumping into this: https://dev.openwrt.org/ticket/4081
Nov 26 23:02:14 <Bartman007> russell_: oops, I guess I never committed. can you confirm the commmit I have on-deck fixes it? http://openwrt.pastebin.com/d565ec0b6
Nov 26 23:13:00 * russell_ looks
Nov 26 23:15:31 <Bartman007> ah, yes. that's why I didn't commit. I wanted to see what the other devs thought about the kernel config tweak, and we decided against it.
Nov 26 23:18:13 <russell_> Bartman007: the md5sum needs fixing
Nov 26 23:19:02 <russell_> i recall there is an issue of it *working*, which needed a CONFIG_HZ change.  all i really care about is it compiling. ;-)
Nov 26 23:22:57 <Bartman007> updated: http://openwrt.pastebin.com/d59406ca
Nov 26 23:23:29 <russell_> you want me to test?  or is it moot.
Nov 26 23:25:05 <russell_> http://mirror2.openwrt.org/sources/eina-r37637.tar.gz
Nov 26 23:25:05 <russell_> Download failed.
Nov 26 23:25:17 <russell_> No more mirrors to try - giving up.
Nov 26 23:25:17 <russell_> Checking out files from the svn repository...
Nov 26 23:25:17 <russell_> svn: Syntax error in revision argument 'r37637'
Nov 26 23:26:15 <Bartman007> I don't have time to test at the moment. I will test before I commit, but that will be later tonight
Nov 26 23:27:08 <Bartman007> IIRC the kernel module yells if it doesn't find a suitable timer, users will have to tweak the kernel themselves if they don't have a suitable timing source (of which there are several)
Nov 26 23:27:15 <russell_> oh.  i see.  the "decided against it" was the kernel config tweak.  zaptel update is a go.
Nov 26 23:28:01 * russell_ trying http://openwrt.pastebin.com/d59406ca
Nov 26 23:28:06 <Bartman007> russell_: correct.
Nov 26 23:30:56 <russell_> any ideas on my svn package checkout problem (e.g. eina above)?
Nov 26 23:31:58 <russell_> back on zaptel:
Nov 26 23:32:02 <russell_> make[4]: Entering directory `/src/openwrt/build_dir/linux-brcm47xx/zaptel-1.2.27'
Nov 26 23:32:02 <russell_> mipsel-linux-uclibc-gcc  -DSTANDALONE_ZAPATA -DBUILDING_TONEZONE -DHOTPLUG_FIRMWARE -I. -O4 -g -fPIC -Wall -m64 -DZAPTEL_CONFIG=\"/etc/zaptel.conf\" -c ztcfg.c
Nov 26 23:32:02 <russell_> cc1: error: unrecognized command line option "-m64"
Nov 26 23:32:51 <Bartman007> ok, I've had to fix that previously in asterisk, I'll deal with it later.
Nov 26 23:32:58 <CyrusFF> russell_: maybe try svn revision without the leading r
Nov 26 23:33:31 <russell_> CyrusFF: it's something the magic makefile stuff is doing for me
Nov 26 23:33:38 <russell_> i don't even see the command line
Nov 26 23:34:47 <CyrusFF> ah ok
Nov 26 23:35:48 <Bartman007> russell_: does http://openwrt.pastebin.com/d25b3a136 fix the eina failure?
Nov 26 23:38:05 <russell_> trying
Nov 26 23:39:18 <Bartman007> that was a really quick hack, there is a better way to fix it, but I should really be preparing things for class
Nov 26 23:39:33 <russell_> Checking out files from the svn repository...
Nov 26 23:39:33 <russell_> svn: PROPFIND request failed on '/svn/e/trunk/eina'
Nov 26 23:39:33 <russell_> svn: PROPFIND of '/svn/e/trunk/eina': Server certificate verification failed: certificate issued for a different hostname, issuer is not trusted (https://svn.enlightenment.org)
Nov 26 23:41:00 <Bartman007> russell_: run add IGNORE_ERRORS=m to your make command for now, that will ignore any failures on packages compiled as modules, I'll fix these later.
Nov 26 23:41:10 <Bartman007> s/run //
Nov 26 23:41:20 <russell_> k
Nov 26 23:43:02 <russell_> grep -r to the rescue: include/download.mk
Nov 26 23:44:42 <Bartman007> that Server cert issue is because you haven't indicated you trust that cert (I assume it isn't signed by a trusted CA)
Nov 26 23:45:18 <Bartman007> once I fix these later tarballs of the source tree will be on the openwrt mirrors and it won't be an issue
Nov 26 23:45:55 <russell_> yeah
Nov 26 23:54:26 <russell_> the other fix is to s/https/http/ on the PKG_SOURCE_URL
Nov 26 23:54:33 <russell_> similar problem for eet
Nov 26 23:57:25 <Bartman007> you'll probably see it for everything enlightment related
Nov 27 00:28:51 <russell_> eet fails for different reasons
Nov 27 00:30:29 <russell_> for a while there, xorg wasn't pulled in for wgt634u
Nov 27 00:35:25 <Bartman007> I'll be reviewing the dependencies when I work on it later, that struck me as odd also.
**** ENDING LOGGING AT Thu Nov 27 00:45:18 2008