| |
| |
| |
|
Page: 1 2
Comments:
<0> hi, how can i uncompress a linux binary kernel?
<1> smoosh: you mean a bzImage?
<1> I think you just bunzip2 ./bzImage
<0> arrummzen: it's not a standard compressed kernel...
<0> arrummzen: it should be compressed with lzma...
<1> well, try decpressing it with something that supports the algorithm then...
<0> arrummzen: i've used p7zip wich is the tool tha support this algorithm. but it doesnt work...
<1> smoosh: how did you compress it in the first place?
<0> arrummzen: the problem, is that i've not compressed the kernel. i've only this kernel and i don't know how decompress it...
<1> where did you get it?
<0> arrummzen: it's a modem linux kernel
<1> smoosh: ah, ok.
<0> arrummzen: bye :)
<2> i think bunzip2 ./bzImage wouldn't work either
<2> but i may be wrong
<2> bzImage is an executable so it contains parts that are not compressed (e.g the decompressor code)
<3> would depend on who made the image
<3> i've seen some embedded kernels compressed with upx
<4> nas
<4> hi
<4> somebody can help me with sata and ide ?
<5> hello guys
<5> is it possible to write like a kernel module that reads the identifier of a ide drive? I mean outside the kernel I can get it from /proc but not sure how to do it inside...
<6> noit's not possible to write like a kernelmodule :P
<6> if it's in proc, all you need too do is find the block of code that wrote it too proc & copy it too your module that you desire
<5> gb, I'm just new to kernel development, not sure about the scope of it... for example from inside the kernel can I reboot the system? like a kernel module that calls reboot or similar? is that possible?
<6> writing it too where you want of course instead of proc
<6> anything running inside init as uid 0 can do anything
<6> why do you keep saying like a?
<5> gb, thanks, I'll try it
<6> it's very confusing too try & ascertain what you are asking
<5> gb, cause I'm not sure
<6> think about what it is you want too type before you type please
<6> then you might have your answer before you ask
<5> gb, don't think so, I have never do anything inside the kernel, just getting info
<5> if I write a kernel module, can I add it to the kernel (when building it) as a 'build it' or I need to somehow make the module aware of that possibility?
<6> if you keep it as auser space thing, you can load it whenever you want via /etc/init.d/
<6> dont quite understand your need too have it as part of the kernel
<5> gb, no, it's not a need just a question, I'm trying to see all the options
<6> well if you keep things in user space it gives you quite a bit more freedom as you can then use it with any kernel
<5> gb, yes, I supposed, but is it simple to just compile in a module into the kernel?? or does it need to be a special (capable of compiling as part of the kernel) module ? just want to know if making a module only or a module with build in option is somehow any different
<6> again you're talking about compiling into the kernel, Why?
<5> gb, because, I want to know...
<6> if you dont know enough about the inner workings of the kernel then you should probably not be trying too compile your code into it
<5> gb, so there are two things totally different
<6> all too often the kernel is made bloated by crap that can function fine outside the kernel but people have put it into the kernel out of ignorance
<6> the only time you put stuff into the kernel is when it causes problems with speed/memory/usability of the system
<6> otherwise it should remain an easy too insert/remove piece of user space code
<6> A sample of code that would be good inside the kernel is something that duplicates a lot of the functionality of the kernel which would otherwise cause double the use of cpu/memory when runniong which could be halved if it simply made some internal kernel calls
<5> gb, ok
<6> Another example would be a piece of code used too mount /
<6> it'd be pretty difficult for such code too be effective outside the kernel
<6> nfs is a good example of where code has been written in both user space & kernel space & caused many problems in compatability
<6> ie at one stage if you were using the kernel space nfsd you need a kernel space nfs client too access it & vice versa
<6> if you dont need too make kernel calls then play it safe & leave it outside the kernel block
<6> if you cant make your code work without putting it inside the kernel then it'sprobably time too look seriously at putting it within the kernel in an ordered manner
<6> ie dont put a piece of code inside the kernel too palmit off too a user space app, thats awful n howalsa used too work if it still doesn't
<6> either stay completely inside or completely outside, the latter being preferable in terms of multiple kernel compatability
<5> gb, thanks I'll take that in consideration ;)
<6> oops
<6> http://pastebin.com & http://paste.husk.org to paste code | Just ask,
<6> don't ask to ask! | #kernel for all your kernel needs | LDD 3rd.
<6> Edition http://www.oreilly.com/catalog/linuxdrive3/book/index.csp
<6> oops
<7> gpm?
<6> punk-***: ey?
<7> gpm is always a pita to paste multi-line stuff
<6> hehehe yup
<6> nomore user error
<6> shoot user, replace & press a key too continue :)
<7> small price to pay for the simplicity of text based apps
<6> yeah we should shoot all the users
<6> they are the biggest hole in the kernel :)
<7> all i/o should be done away with
<8> should i report this crash as a bug? http://bum.pastebin.com/718129 , http://bum.pastebin.com/718133
<9> my /dev/ttyS* entries are al broken links
<9> *all
<10> any one know if the issue with libata has been fixed, i have a 2.6.8 kernel, i cannot get my sata drives to show
<6> Ugly_Duck: so try a more recent kernel then :P
<6> you might as well be using 2.4.32 then too whinge about faults in a kernel pre 2.6.10
<10> gb: could you please help me with a link? i am currently using debian kernel-sources, not sure what file to grab
<6> Ugly_Duck: ask in #debian then, I use kernel source which hasn't been corrupted by dist patching
<10> kk..
<10> would be nice to be able to find a module thats been patched, instead of compiling my own kernel
<10> i know 2.6.7 worked
<10> 2.6.8 does not appear too
<11> does anyone here know how the on disk caches work with IDE drives?
<11> or if there is a way to disable them somehow (read direction)?
<12> hi _RADIOhead
<12> haow are your networking problems? have you tested grml?
<13> aXp: dude
<13> lo lo
<13> i was out for food
<13> :)
<14> pfft
<6> :P
<14> least I keep it up top date
<14> err.. top = to
<6> err err too you tewe :P
<15> yo _RADIOhead
<13> neil_m: dude :) how are you
<15> am good man and you?
<13> fine :) working
<15> doh :(
<15> heh
<13> damn winbblows
<15> yes :(
<15> install linux
<15> :D
<15> and get the WinXP theme for KDE
<15> they will never know bwahah
<13> hahahhahaha ;)
<15> _RADIOhead get slackware installed now!
<13> neil_m: u?
<13> #uname -a
<13> Linux D1-Box 2.6.16.2-Dloop #2 SMP PREEMPT Wed Apr 12 12:30:34 CEST 2006 i686 prescott i386 GNU/Linux
<16> hi, please have a look at http://pastebin.com/718754
<16> i have a kernel error: >Unable to handle kernel paging request at virtual address
<16> i have being trying many things to solve it, but i couldn't
<16> I ran the memtest86+ memory test but the memory was OK
<16> and I ran a fsck.ext3 from a Knoppix and it is ok as well
<16> I fell due to this kernel problem
<16> can anyone help me?
<6> acidborg: Can?Probably;Want too?Probably not :)
<16> what's pmt?
<6> Percusive Maintenance Therapy/Treatment :)
<6> it's good for the user & the equipment in question :)
<16> ?
<6> removes all that pentup frustration
<6> you take an 8lb sledge hammer & apply it too the faulting hardware until it gives you no more issues :)
<16> . . .
<16> anyone else can help me,please?
<6> well I scrolled back & cant even see where you mention a particular kernel version
<16> 2.6.15
<6> or whether you've tried multiple versions
<16> yeah,I have tried different 2.6.15 version
<16> 2 ubuntu versions
<16> and one compiled myself
<6> no as in 2.6.16 which is later & 2.6.14 which is much earlier
<6> dist versions are of no real help as they have wierd pastches we can do nothing about
<16> now, i can't compile any kernel because of this errors
<16> they freeze my computer randomly
<6> can you run an old kernel prior too this error?
<6> if no then your problem is not kernel related
<16> let me see if I can ran a 2.6.16.11
<6> I doubt it is random, it just appears that way because you do not know the cause of your problem
Return to
#kernel
or
Go to some related
logs:
#osdev
vmware fedora network unreachable
fglrx.ko needs unknown symbol pm_register
spca error opening v4l
iptables -t nat -m tcp -p tcp -A PREROUTING --dport
psyke83
#suse
#suse
gtparted
#nvidia
|
|