=== Starting logfile irclogs/ubuntu-ports.log
=== ubuntulog [i=ubuntulo@ubuntu/bot/ubuntulog]  has joined #ubuntu-ports
=== Topic for #ubuntu-ports: hppa, ia64 or new ports discussion goes here | hppa status: Building gutsy in a private repo | http://buildd.mmjgroup.com/buildLogs/byDate/today.html | deb http://people.ubuntu.com/~lamont/hppa/gutsy-stage0 main (subject to .debs changing underneath you, etc)
=== Topic (#ubuntu-ports): set by lamont at Wed Jun 27 18:59:28 2007
=== tmarble [n=tmarble@user-38q4et6.cable.mindspring.com]  has joined #ubuntu-ports
=== tmarble [n=tmarble@user-38q4et6.cable.mindspring.com]  has joined #ubuntu-ports
=== shinmen_ [n=alvaro@200.122.166.131]  has joined #ubuntu-ports
=== shinmen_ is now known as shinmen
=== jbailey [n=jbailey@modemcable178.77-70-69.static.videotron.ca]  has joined #ubuntu-ports
[04:41] <jbailey> lamont: ping
[04:41] <jbailey> jbailey@bld-5:~$ uname -a
[04:41] <jbailey> Linux bld-5.mmjgroup.com 2.6.22.1 #3 Wed Jul 18 08:33:55 MDT 2007 parisc64 GNU/Linux
[04:41] <jbailey> So your git tree works with my config.
[04:41] <jbailey> I'm guessing there's something broken in the Ubuntu config for the A500 wrt to the console.
[04:41] <jbailey> Shall I email you my .config file for comparison?
[07:08] <lamont> 182 upgraded, 44 newly installed, 2 to remove and 3 not upgraded.               
[07:08] <lamont> please diff the configs
=== lamont dist-upgrades the j7k to gutsy
[07:11] <lamont> stage0,  of course
[07:44] <lamont> Linux test130 2.6.22-8-hppa32 #1 SMP Wed Jul 18 04:33:02 UTC 2007 parisc GNU/Linux
[07:44] <lamont> FTW!!!
[07:59] <lamont> jbailey: did you send me the .config?
[08:05] <jbailey> lamont: No, I left for lunch just before 1. =)
[08:05] <jbailey> Nice! 
=== lamont boots the 64 bit kernel on his happy-with-32-bit J7k
[08:07] <jbailey> lamont: Do you mind just fetching it from bld-5?
[08:07] <jbailey> /home/jbailey/Programming/kernel/hppa-ia64/.config
[08:07] <lamont> no problemo\
[08:08] <jbailey> Tx.
=== lamont detects a saddeningly consistent trend in the 64-bit kernel.
[08:10] <lamont> amusingly, baldios is in the INSTALL stage of 2.6.22-8.18hppa2
[08:11] <lamont> fwiw, 'function not implemented' in df output is because of dapper kernel and gutsy userspace
[08:11] <jbailey> lamont: Your git tree boots on my a500 with my config.
[08:11] <lamont> WARNING: Couldn't open directory /lib/modules/2.6.22-8-hppa32: No such file or d
[08:11] <lamont> irectory                                                                        
[08:11] <lamont> FATAL: Could not open /lib/modules/2.6.22-8-hppa32/modules.dep.temp for writing:
[08:11] <lamont>  No such file or directory                                                      
[08:16] <lamont> hrm... maybe  if I gave it the correct initrd
[08:17] <jbailey> =)
[08:24] <jbailey> lamont: I'm about to run off to visit the notary.  Need anything from me before I go?
[08:24] <lamont> blood
[08:25] <lamont> and offsets.h
[08:25] <lamont> but this evening will work for those
[08:25] <jbailey> Luvly.  Grats again on the kernel.
[08:31] <fabbione> evening guys
[08:31] <fabbione> hey tmarble !
[08:32] <tmarble> hi fabbione!
[08:41] <lamont> -CONFIG_SMP=y                                                                   
[08:41] <lamont> -CONFIG_HOTPLUG_CPU=y                                                           
[08:41] <lamont> +# CONFIG_SMP is not set                                                        
[08:41] <lamont> I _HOPE_ that's not why jbailey's config works
[08:41] <fabbione> ROFL
[08:42] <fabbione> you are fighting with UP/SMP
[08:42] <fabbione> we (sparc) are fighting with +CONFIG_HOTPLUG_CPU=y
[08:42] <fabbione> because our SMP works ;)
[08:42] <fabbione> and of that.. only multiple removal of N CPU's is slightly racy
[08:42] <fabbione> ;)
[08:52] <lamont> well, I'm hoping that's not it...
[08:52] <lamont> because SMP should _WORK_ dammit;
[08:52] <lamont> the situation is that jbailey's .config works on a500, gutsy's .config does not.
[08:53] <lamont> jbailey's config doesn't work on j7k (pretty sure it's because he excludes lots of buss support...), gutsy's doesn't either
[08:53] <lamont> and so I'm doing the binary sort on the config diffs.  only not so binary
[08:54] <fabbione> oh boy
[08:54] <fabbione> you mean you got to build a kernel that will never work?
[09:15] <lamont> fabbione: yep
[09:16] <lamont> so... care if I toss an smp 32-bit kernel on baldios?
[09:17] <fabbione> not at all... go ahead
[09:17] <fabbione> do you need me to check if it boots?
[09:17] <fabbione> btw.. saturday i am going in holidays for 3 weeks
[09:17] <lamont> I need you if it doesn't boot. :-)
[09:18] <fabbione> there will be nobody babysitting baldios
[09:18] <fabbione> ok
[09:18] <fabbione> go ahead
[09:18] <fabbione> i am here if you need me
[09:18] <fabbione> just make sure not to kill the old kernel :)
[10:00] <lamont> fabbione: got dragged off for a few... installing now
[10:01] <fabbione> lamont: sure
[10:01] <fabbione> i am still here watching a movie
[10:01] <lamont> cool.  gotta walk into a meeting room, then will do the reboot.
[10:13] <lamont> fabbione: rebooting now...
[10:15] <lamont> and baldios is again responding to pings.. woot
=== lamont smacks fabbione
[10:20] <fabbione> lamont: ?
[10:22] <lamont> palo.conf caused the box to always boot your kernel, instead of /boot/vmlinux
[10:22] <lamont> fixed palo.conf, your kernel is now the recovery kernel (0/vmlinux), and /boot/vmlinux (2/vmlinux) is the default kernel
[10:22] <lamont> and the machine is booting again
[10:23] <fabbione> oh right yeah i made sure to always boot a good kernel
[10:23] <lamont> right.  and i want to always boot the current kernel, with a known-good fallback for when tha tfails.
[10:26] <fabbione> did it boot?
[10:26] <lamont> now running 2.6.22-8-hppa32, and buildd running
[10:26] <fabbione> that's SMP or UP?
[10:26] <lamont> feel free to keep a watch on it for panics,e tc
[10:26] <lamont> SMP
[10:26] <fabbione> ok
[10:27] <fabbione> i don't have console on it right now. the machine with serial is powered off
[10:27] <lamont> that's why it's hppa32... that one works, and gives us SMP for the buildd
[10:27] <fabbione> just let it run for the night and we will see
[10:27] <fabbione> we have time till my friday morning to roll back if it crashes
[10:27] <fabbione> then you are on your own till 8th of Aug
[10:28] <lamont> OK.  once I finish debugging the 64 bit kernel, I'll light up another j7k or two as well.
[10:28] <lamont> although I expect that baldios/bld-4 will be current before too long anyway
[10:29] <fabbione> ehhe
[10:33] <fabbione> lamont: did you notice the OOPS in dmesg at boot time?
[10:33] <lamont>  /dev/md1             679749632 265894532 400047224  40% /home
[10:33] <lamont> not bad for a home workstation/devel machine
[10:33] <lamont> that's a feature
[10:34] <lamont> fabbione: 0 byte slab release
[10:34] <lamont> iirc
[10:36] <fabbione> ok
[10:37] <fabbione> lamont: how much is that? 2TB?
[10:38] <lamont> .7TB
[10:38] <lamont> raid 5
[10:38] <fabbione> yeah not bad
[10:38] <lamont> single-user desktop used for devel work and general existance
[10:38] <fabbione> ehehhe
[10:39] <lamont> scary part is that the home mirror /srv2 is smaller... :-(
[10:39] <fabbione>  /dev/fucking/src      340G  298G   39G  89% /usr/src
=== lamont has always been amused by that name
[10:39] <fabbione> + another 250GB in a couple of disks
[10:39] <fabbione> this is the workstation
[10:39] <fabbione> server has a bit more
[10:40] <lamont>  /dev/md2             484489896 394052552  65826624  86% /srv2
[10:40] <fabbione>   VG Size               1.09 TB
[10:40] <lamont> gotta fix that one of these days...
[10:41] <lamont> hwraid would be nice, too.
[10:41] <fabbione> yeah...
[10:41] <fabbione> but i still prefer software to several extents
[10:42] <lamont> well, good hwraid would be nice. :-)
[10:42] <fabbione> it needs to be really good or it can backfire if you get multiple failures
[10:42] <lamont> yeah
[10:43] <fabbione> like in a raid5 sf if by accident 2 disks are gone with no spare, you have a good chance to recover
[10:43] <fabbione> most hw won't allow you to force a start
[11:53] <lamont>  15:53:22 up  3:35,  3 users,  load average: 152.70, 45.03, 16.32               
[11:53] <lamont> hrm...
[11:53] <lamont> maybe make -j wasn't a completely sane idea.
[12:04] <lamont>  15:55:13 up  3:37,  3 users,  load average: 548.09, 220.90, 82.95