[03:47] <ph8> hi guys, i'm currently trying to get a Realtek RTL8111B network card working on a P5B
[03:47] <ph8> (Asus P5B)
[03:47] <ph8> there's a driver from Realtek about but it needs to be built into the kernel, is that the only way i'm going to get my card working? Is there some generic thing i can do which is a lot simpler?
[03:47] <ph8> Building my own kernel would be a pain in the arse!
[04:00] <accumulator> is there a quick reference somewhere on how to re-enable boot splash when building a custom kernel?
[04:00] <accumulator> .. with make-kpkg
[04:00] <accumulator> ?
[04:14] <`ph8> hey guys, my internet connection's appalling today. I'm trying to get my Realtek 8168 network card to be supported in edgy-latest - the only option appears to be me compiling my own kernel with the drivers in - is that indeed the simplest solution?
[04:20] <accumulator> the only, i think
[05:01] <BenC> actually, no, the best way would be to compile the module against the linux-headers package
[05:01] <BenC> but oh well :)
[05:20] <``ph8> gnomefreak told me you replied to my previous post BenC - cheers! I didn't see it because my internet connection's too rubbish
[05:20] <``ph8> and ben's not here.. :p
[05:20] <``ph8> did anyone know what he meant about compiling the r1000 module against headers?
[05:20] <zul__> his connection is rubbish as well
[05:20] <gnomefreak> lol
[05:21] <``ph8> great pair :p
[05:32] <tfheen> BenC: I wonder if one of the latest dapper kernel uploads have broken something for me.  One of my hosts (amd64) dies when it gets some I/O load.  The latest one gave me about 110k "scheduling while atomic: swapper/0xfffffffe/0" in the kernel log.
[05:37] <``ph8> gnomefreak told me you replied to my previous post BenC - cheers! I didn't see it because my internet connection's too rubbish
[05:37] <``ph8>  the error with the patch is '/lib/modules/2.6.17-10-generic/build' -> does not exist, during the first make clean modules command - which is strange
[05:38] <``ph8> i get that on both of my systems when i try to do the first make clean (according to the README)
[05:55] <gnomefreak> BenC: we are getting alot of people not beable to boot its sending them right to busy box. im looking for a bug that had a workaround for it but is this a kernel issue or a usplash issue i cant remember
[06:10] <ivoks> that can only be kernel issue
[06:19] <gnomefreak> im begining to think its the UUID
[06:19] <gnomefreak> i have 2 people in #ubuntu+1 with same issue i dont think either have hdxx
[07:28] <BenC> gnomefreak: They need to check /dev/disk/by-uuid/ and see if the device exists
[07:28] <BenC> check "cat /proc/cmdline" to make sure it's right
[07:28] <BenC> gnomefreak: Are these fresh installs, or upgrades?
[07:28] <gnomefreak> upgrades
[07:30] <BenC> it could be failures in the UUID conversion
[07:30] <BenC> we need to check the /dev/disk/by-uuid/ listings against menu.lst, /etc/fstab and /proc/cmdline
[07:31] <gnomefreak> neither of them are answering me. i know one of them was trying to get rid of UUID and add hda3 to it but i had to leave
[07:31] <BenC> also important is if these are actual partitions or evms/lvm/md
[07:33] <gnomefreak> also that guy that was trying to build that reltek module filed a bug on it for it to be included in edgy. is that even possible?
[07:34] <BenC> nope
[07:34] <gnomefreak> i didnt think so
[08:26] <ph8> BenC: Thanks for answering, sorry to waste your time
[08:26] <ph8> i'll see if my 4gb of mem is recognised after this kernel recompile i'm doing
[08:26] <BenC> ph8: Nothing you can compile is going to change the memory
[08:26] <BenC> ph8: BTW, sudo apt-get install linux-headers-`uname -r` will help you compile your module
[08:26] <BenC> lot better than recompiling the whole kernel
[08:30] <ph8> yes ty i got it working
[08:30] <BenC> [   60.556067]  Memory: 4038036k/5242880k available (2206k kernel code, 155236k reserved, 1486k data, 292k init)
[08:30] <ph8> ok, i installed from amd64 daily today and it can only see 3gb or 4
[08:30] <ph8> * of 4
[08:30] <BenC> ph8: For reference, that is the memory on my 4G box
[08:30] <ph8> you think some of my memory is faulty?
[08:30] <BenC> that 155236k reserved, is for various system elements
[08:30] <ph8> i'm using top, is that from your dmesg?
[08:31] <ph8> didn't know it was in there
[08:31] <BenC> yes, that's dmesg
[08:31] <ph8> so, on an out of the box amd64 daily install, if it says: [   29.173492]  Memory: 3084280k/3145344k available (2129k kernel code, 60676k reserved, 1424k data, 188k init) - that means one of my memory chips is messed up
[08:31] <ph8> ?
[08:31] <ph8> i can remove them one at a time to make sure
[08:32] <BenC> see, one of your mem sticks isn't being seen I bet
[08:32] <BenC> check BIOS and see what it sees
[08:32] <BenC> run memtest86 too
[08:32] <ph8> memtest sees 3
[08:32] <ph8> not sure about bios
[08:32] <ph8> cheers i'll give it a go after i install this kernel
[08:32] <BenC> if memtest only sees 3, then that's about as bare bones as you can get
[08:32] <BenC> ok
[08:32] <ph8> will the -generic i get with the amd64 daily be the 64bit kernel?
[08:32] <ph8> is there a way to test?
[08:33] <ph8> i don't really need to compile this custom kernel if it is
[08:33] <ph8> Linux rigel 2.6.17-10-generic #2 SMP Fri Oct 13 15:34:39 UTC 2006 x86_64 GNU/Linux
[08:33] <ph8> that looks like 64 to me
[08:34] <ph8> but am i wrong?
[08:46] <ph8> BenC: it's not a faulty stick, just tested them all - saw a post online saying that its a limitation of 'most modern motherboards' that they cannot fully utilise 4gb of RAM,  could this just be Asus' current limit?
[08:47] <mjg59> It's not uncommon for lower-end motherboards to map the PCI space over the top of the RAM
[08:53] <ph8> oo
[08:53] <ph8> there's an option in the bios that says 'allow pci overmapping' or something like that
[08:53] <ph8> do you think that should be off?
[08:54] <ph8> or is it actually a good feature?
[08:55] <mjg59> Try changing that, yeah
[08:58] <ph8> will let you know :)
[09:04] <ph8> well
[09:04] <ph8> it was actually disabled
[09:05] <ph8> enabling it gets my mem to 3584 - which i'm happy with, cheers.
[09:05] <ph8> I'm now experiencing 'NMI Watchdog experienced a LOCKUP on CPU 1' though - is that a common one?
[09:05] <ph8> Many reporters on google but no causes/solutions
[09:22] <ph8> hmm, the memory overlapping option did it - strangely
[10:00] <ph8> just a general newbie question guys - what does the ubuntu kernel have that the stock one doesn't?
[10:22] <johanbr> Lots of things, look at /usr/share/doc/linux-image-*/changelog.Debian.gz
[11:10] <ph8> should IA32 be enabled by defalt in the amd64 kernel?
[11:10] <ph8> i get some weird errors when running a binary that works on other machines
[11:11] <mjg59> ph8: It, uh, is, surely?
[11:11] <ph8> hopefully
[11:11] <ph8> i can't run the linux client from folding.stanford.edu
[11:11] <ph8> on my machine
[11:11] <ph8> weird error though
[11:11] <mjg59> What's the error?
[11:12] <ph8> root@rigel:~/bench# ./FAH502-Linux.exe 
[11:12] <ph8> -su: ./FAH502-Linux.exe: No such file or directory
[11:12] <ph8> it's definitely there and global +x
[11:12] <mjg59> Run strace against it and see what opens fail
[11:12] <mjg59> That usually means that it's missing the run-time dynamic linker
[11:12] <mjg59> Which may been that you're missing the compatibility libraries
[11:13] <mjg59> CONFIG_IA32_EMULATION=y
[11:14] <ph8> ah! i don't have ia32-libs
[11:14] <ph8> lets see if that makes a diff
[11:18] <ph8> mjg59: ia32-libs sorted it
[11:18] <ph8> cheers!
[11:18] <ph8> my install borked a bit at the end you see, i didn't get anything apart from the base packages