T2 IRC Log: 2009-08-12

This is the log as captured by an IRC bot in the channel. The statements are those of the individual people and might not neccessarily reflect the policy and legal rules as set forth by the T2 SDE Project.

« prev | next »

--- Log opened Wed Aug 12 00:00:38 2009
00:49 -!- mqueiros_ [n=mqueiros@217.70.75.151] has quit [Client Quit]
00:57 -!- mqueiros_ [n=mqueiros@217.70.75.151] has joined #t2
02:47 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: mjungwirth
02:50 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: mtr, TobiX
02:51 -!- TobiX [i=tobias@zoidberg.org] has joined #t2
02:51 -!- Netsplit over, joins: mjungwirth
02:51 -!- TobiX [i=tobias@zoidberg.org] has quit [Killed by sagan.freenode.net (Nick collision)]
02:51 -!- TobiX [i=tobias@88.198.6.61] has joined #t2
02:51 -!- Netsplit over, joins: mtr
02:51 -!- TobiX [i=tobias@88.198.6.61] has quit [Read error: 104 (Connection reset by peer)]
02:52 -!- TobiX [i=tobias@zoidberg.org] has joined #t2
02:53 -!- TobiX [i=tobias@zoidberg.org] has quit [Killed by wolfe.freenode.net (wolfe.freenode.net (TobiX[i=tobias@zoidberg.org] Ghosted sagan.freenode.net))]
02:57 -!- TobiX [i=tobias@zoidberg.org] has joined #t2
03:02 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: mtr
03:04 -!- Netsplit over, joins: mtr
03:58 -!- koan_ [n=koan@unaffiliated/koan] has joined #t2
03:59 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: koan
07:49 -!- mj-netbits [n=sarpedon@213.174.252.112] has joined #t2
08:25 -!- T-One [n=T-One@62.218.122.66] has joined #t2
09:36 [Users #t2]
09:36 [@ChanServ ] [ CIA-69] [ mj-netbits] [ mtr ] [ Stealth]
09:36 [ _|` ] [ Cyda_ ] [ mjungwirth] [ oliv`] [ T-One ]
09:36 [ bitdancer] [ koan_ ] [ mqueiros_ ] [ rxr ] [ TobiX ]
09:36 -!- Irssi: #t2: Total of 15 nicks [1 ops, 0 halfops, 0 voices, 14 normal]
09:45 -!- T-One [n=T-One@62.218.122.66] has quit ["Verlassend"]
09:53 -!- koan_ is now known as koan
10:25 < CIA-69> rene * r33851 /trunk/package/filesystem/dvdisaster/dvdisaster.desc: * updated dvdisaster (0.72 -> 0.72.1)
10:25 < CIA-69> rene * r33850 /trunk/package/audio/snd/snd.desc: * updated snd (10.7 -> 10.8)
10:25 < CIA-69> rene * r33852 /trunk/package/xorg/xrandr/xrandr.desc: * updated xrandr (1.3.0 -> 1.3.1)
10:25 < CIA-69> rene * r33853 /trunk/package/develop/roundup/roundup.desc: * updated roundup (1.4.8 -> 1.4.9)
10:25 < CIA-69> rene * r33854 /trunk/package/mail/dovecot/dovecot.desc: * updated dovecot (1.2.2 -> 1.2.3)
10:26 < CIA-69> rene * r33855 /trunk/package/develop/diffstat/diffstat.desc: * updated diffstat (1.47 -> 1.48)
10:37 -!- PassingStranger [n=dude@p54BF5691.dip.t-dialin.net] has joined #t2
10:37 [Users #t2]
10:37 [@ChanServ ] [ CIA-69] [ mj-netbits] [ mtr ] [ rxr ]
10:37 [ _|` ] [ Cyda_ ] [ mjungwirth] [ oliv` ] [ Stealth]
10:37 [ bitdancer] [ koan ] [ mqueiros_ ] [ PassingStranger] [ TobiX ]
10:37 -!- Irssi: #t2: Total of 15 nicks [1 ops, 0 halfops, 0 voices, 14 normal]
10:47 -!- T-One [n=T-One@62.218.122.66] has joined #t2
11:18 -!- mj-netbits [n=sarpedon@213.174.252.112] has quit ["Nettalk6 - www.ntalk.de"]
11:21 -!- Baldzius [n=aldas@87-198-133-94.ptr.magnet.ie] has joined #t2
11:21 < Baldzius> moin
12:34 < rxr> moin Baldzius
12:34 * rxr gotta lunch - cu then
12:35 < Baldzius> hi rxr
13:26 -!- PassingStranger [n=dude@p54BF5691.dip.t-dialin.net] has quit ["Leaving"]
13:53 -!- michael-i [n=michael-@141.41.40.153] has joined #t2
13:53 < michael-i> afternoon rxr
13:54 < michael-i> I've gotten the blackfin stuff working :) But (and it's a huge but) it requires two separate toolchains. It seems this is the norm now with analog devices' configuration. One for the kernel and one for everything else
14:01 < rxr> oh my god
14:01 < rxr> this sounds _insane_
14:02 < rxr> do they have a "reasonable" epxlanation for this "mess" ?
14:10 < michael-i> different binary formats...
14:10 < michael-i> http://docs.blackfin.uclinux.org/doku.php?id=toolchain:targets
14:10 < rxr> ough: http://peterbelanger.com/posts/36-cover-creation
14:12 < rxr> michael-i: still strange, normally you would have one compoiler, linker assembler and pass an argument for the object file variant
14:14 < michael-i> rxr: that's why it took me so long to wrap my head around this. That's really the way their buildscript does it. The company that we're porting this for also confirmed this is the only way they're able to build kernels
14:14 < rxr> bummer
14:16 < michael-i> So, I've manually copied in the binary release of that toolchain into TOOLCHAIN/ and changed archtest.out to the new tuple. It compiles fine now but this needs to be automated somehow.
14:18 < michael-i> I'm thinking of making a package (blackfin/toolchain-bin-kernel) which does the copying into that directory. And I guess I'll override CROSSCC for kernel compilation in the arch/blackfin/package/linux26 .conf file
14:18 < michael-i> It's ugly...but it should get the job done
14:18 < rxr> I guess the most useful solution would be to let the toolchain package rebuilt themselves a second time
14:19 < rxr> I do such tricks for some package especially for sparc64
14:19 < rxr> where I default to 32bit and want some 64bit version of some basic libs notheless
14:20 < rxr> architecture/sparc6/parse-config
15:49 < michael-i> Looks good. I think I just need something simple to change the tuple for linux26 in parse-config
16:14 < michael-i> rxr, can I put a condition in archtest.out which will set the arch_target depending on which package is being built?
16:25 -!- T-One [n=T-One@62.218.122.66] has quit ["Verlassend"]
16:31 < michael-i> here's my attempt at a arch/bfin/archtest.sh (http://pastebin.ca/1526569)
16:32 < michael-i> It doesn't seem to switch on the $pkg var which tells me that's the wrong variable in this context. Is it even possible to switch on package name in archtest.sh
16:37 -!- Baldzius [n=aldas@87-198-133-94.ptr.magnet.ie] has quit ["Leaving"]
17:42 < michael-i> Giving up for now. So far I've tried: switching on $pkg in archtest.sh, overriding $arch_target in linux26.conf, manually patching the kernel Makefile. I'm stuck with commenting/uncommenting lines in archtest.out for now.
17:43 < michael-i> If anyone knows the best way to override arch_target for a specific package, I'd be very grateful
17:51 -!- michael-i [n=michael-@141.41.40.153] has quit []
20:54 -!- koan [n=koan@unaffiliated/koan] has quit ["Lost terminal"]
20:54 -!- koan [n=koan@unaffiliated/koan] has joined #t2
--- Log closed Thu Aug 13 00:00:38 2009