| |
| |
| |
|
Page: 1 2
Comments:
<0> seems like you might have tried everything I could suggest already though
<1> gb: then start trying to figure out what went wrong by trimming it down a tiny bit at a time
<0> umm exact same config?
<1> gb: .config
<0> make oldconfig you mean I hope?
<1> gb: yes
<0> ok yeah in theory that should work
<1> gb: copy the fc4 .config to 2.6.15.3 dir and make oldconfig && make bzImage && make modules && make modules_install && make install
<0> unfortunately it doesn't always & you may need too await next kernel patch
<1> gb: if it doesn't then it's some obscure fc4 patch that makes that box work
<1> gb: the fc4 patches are available
<1> obviously
<0> sounds like you got it sussed, just got to go through the actions
<1> gb: anyway, i tried of trying the latest fc5 test on this box. it's atm just a toy, i won't need it for serious work until the 28th or so
<1> s/i tried/i thought/
<0> time for a potasium toasted sanga
<2> can somebody explain me what a "kernel paging request" means ?
<1> crich1999: do you get "unable to handle" ?
<1> crich1999: kernel paging request means the kernel would love to have another page of mem
<2> mord, yes
<2> mord, ok
<1> crich1999: unable to handle /might/ mean bad ram
<1> crich1999: it could mean a fscked up kernel conf too
<2> mord, i'm experienced user land c coder, i'd like to track down this oops
<2> mord, no it's something wrong with this particular module
<2> mord, i just wasn't sure what the exact meaning was to understand what's happening
<1> crich1999: i think theres something in ksyms that might help you track down the issue. can't remember much though.
<1> crich1999: the "page" is a page of memory
<2> mord, ok, are there common mistakes one can do to get this (unable to handle .. ) , like double free or something like that ?
<3> hey
<1> crich1999: i suppose
<3> My kernel is panicing for fun
<1> crich1999: i'm not sure you can much control kernel paging from userland though
<1> (cig)
<0> xbmodder_: thats the job of your bofh, dont remove their pleasure
<3> me = bofh
<3> That was bad though
<3> it kept panicing
<0> xbmodder_: you need a pfy then
<3> pfy?
<3> The hardware is bad
<3> something inclduing MMU, CPU, mobo
<0> 10lb sledge hammer then
<0> percusive maintenance on such hardware will alwaya fix it
<3> I just fryed out a server like 10 hours ago
<3> That was sad
<3> I cry
<0> why I thought you were the bofh?
<3> the core fried.
<2> mord: course i can't, it's a kernel mod i'm debugging right now
<3> when I say fried I mean FRIED!
<1> crich1999: ah. the "experienced userland" thingy confused me
<3> it has like a hole in it
<0> thats for cooling
<1> crich1999: think of it as a kernel level segfault and you're pretty close to what it means if i've understood correctly
<2> mord, sorry i just wanted to say that i know about c coding, and it's not a standard issue i'm facing
<1> crich1999: nah. i just misunderstood
<2> mord, but it's not a Null pointer reference, that's sure .. so it must be some sort of double free or so
<1> crich1999: something like that
<3> :!
<2> mord, hard to track down if it's not your code though
<1> crich1999: the prob is as i see it is that the paging request could mean several things, kalloc and free being the most likely culprits
<2> just playing around with kdb .. but it doesn't tell me more than the oops itself ..
<1> crich1999: lots of debug output? ;)
<2> mord, not that muhc .. but it doesn't contain more info
<2> mord, well my culprit seems: kobject_put
<1> crich1999: but seriously, try and find out what it's doing whenever it's fiddling around with mem allocs
<2> don't know what kobject_put or kobject_del really do
<2> mord, anyways thanks so far need to jump off
<1> crich1999: kernel source root dir has kobject.txt, read it
<2> mord, thanks
<2> will do that
<2> bye
<3> :-/
<1> bah. fc4 stock .config, 2.15.3 built with it, hangs at boot at AGP aperture conflicts with PCI mappings
<1> 2.6.15.3
<4> hi
<4> I'm confused on how dual core processors work, as far as the kernel is concerned. are they just smp machines, or single machines with transparent benefits ?
<1> pheorehs: smp i suppose
<4> mord, ok I guess, thx
<5> hai
<6> oh noes
<5> ohh noes its teh bobbens
<5> feckin PS2 my need for speed most wanted save is corrupt
<5> i was number 9 in the black list
<5> grr
<6> meh i only have my super nintendo
<5> yes super nintendo is the ultimate machine
<5> i miss the days of kick *** games!
<6> emulators :)
<5> yes
<5> snes9x
<5> fo shoa
<6> zsnes :)
<5> yes
<7> hi
<7> i have weird problem whit dhcp server
<7> it give the ip addrs to it self
<7> and put that addres to nextcomputer how ask ip
<7> and change his own ip
<7> in debina
<7> debian
<8> so has anyone compiled 2.6.16 yet?
<8> any bugs?
<6> 2.6.16?
<8> yes..
<6> stable is 2.6.15.4
<8> ok ok
<8> ill get it then
<6> must of come out today
<6> i have 2.6.15.2 from like 2 days ago, but i dont want to restart my computer so soon, makes me feel dirty all over
<8> :)
<8> yes
<9> amazing: http://milw0rm.org/id.php?id=1397
<8> oh right
<8> isnt
<9> Drakas, hahaha....you've just test it ?
Return to
#kernel
or
Go to some related
logs:
#linuxhelp
googleplex numeric
red hat console ownership
#python
dialguard peoplepc
#linux
find the value of k that makes this a probability density function
#css
a920 https parser
#sendmail
|
|