What is a "kernel 32 virus" ????

[Spacebaboon]

Someone is trying to sell me a comp... But he says it has a kernel 32 virus does anyone know what kind of virus it is? Are those removable by a format? Or how do you remove them?

[Structural Integrity]

Originally posted by Spacebaboon
Someone is trying to sell me a comp... But he says it has a kernel 32 virus does anyone know what kind of virus it is? Are those removable by a format? Or how do you remove them?

You know what a "Kernel" is?
Right... Figure that out and you can answer your questions.

[Spacebaboon]

I know what kernel32.dll is But the guy is telling me that it is hard to remove i just want to know can such a virus survive a format?

[Structural Integrity]

Originally posted by Spacebaboon
I know what kernel32.dll is But the guy is telling me that it is hard to remove i just want to know can such a virus survive a format?

No...
The kernel IS the operating system. It's the system that issues tasks, manages the memory, drives the hardware, operates the networking stack, etc etc.

I guess such a virus is hard to remove from an OS without actually removing the OS, but I really doubt it can survive a format.

[ParraCida]

Format will fix it.

[Spacebaboon]

thx i think this guy is trying to rip me off...
First he wants to selle me a Athlon xp 1400+ 17'', etc. for 400€ (desent price) Then he says the comp has a kernel 32 virus and wants to selle me a pentium 800... with bigger hd and more RAM...

[menth0l]

o rly

It's obviously to do with corn

[Nusselt]

whats the 32 for

Originally posted by Nusselt
whats the 32 for

How much corn

[Nusselt]

is 32 a lot of corn?

[Structural Integrity]

Originally posted by Nusselt
whats the 32 for

nr of bits the platform supports.

Originally posted by Nusselt
is 32 a lot of corn?

When its in tonnes

[Nusselt]

Originally posted by Yossarian
When its in tonnes

kurashima could eat 32 tonnes of corn.

[Fifth_teletubbie]

Originally posted by Structural Integrity
nr of bits the platform supports.

Actually it's the maximum range of storage the operating system can address.
32-bit (in this context) just means the OS uses 32-bit pointers to address its storage.

[queball]

Originally posted by Structural Integrity
No...
The kernel IS the operating system. It's the system that issues tasks, manages the memory, drives the hardware, operates the networking stack, etc etc.

I guess such a virus is hard to remove from an OS without actually removing the OS, but I really doubt it can survive a format.

"Reinstall"
kernel32.dll doesn't directly deal with most hardware or the network stack, don't know about the rest.

Originally posted by Fifth_teletubbie
Actually it's the maximum range of storage the operating system can address.
32-bit (in this context) just means the OS uses 32-bit pointers to address its storage.

"Maximum range of storage the operating system can address"? So it can never "address" more than 4GB?

[Fifth_teletubbie]

Originally posted by queball


"Maximum range of storage the operating system can address"? So it can never "address" more than 4GB?

Yup. Though depending on the OS it might use some of those bits in alternative ways. Many use 31 bits for addressing, leaving a 2GB max (which is also a limit on quite a few motherboards for much the same reason).

And no reason to put quotes around 'address', it is the proper term to use.

[queball]

Originally posted by Fifth_teletubbie
Yup. Though depending on the OS it might use some of those bits in alternative ways. Many use 31 bits for addressing, leaving a 2GB max (which is also a limit on quite a few motherboards for much the same reason).

And no reason to put quotes around 'address', it is the proper term to use.

So it's really a 31-bit system? Or could you explain what you mean by address?

Since my HD is far more than 4GB, and I can access all of it. And Win2K AS, with its kernel32.dll, claims to support 8GB of RAM.

[Costa]

Originally posted by queball
So it's really a 31-bit system? Or could you explain what you mean by address?

Since my HD is far more than 4GB, and I can access all of it. And Win2K AS, with its kernel32.dll, claims to support 8GB of RAM.

its the amount of ram,.

and Windows 2000 AS is built apon dual processor technology meaning it can address 32 + 32 bits addresses (in theory)

[MAdnRisKy]

I had one, it was called brazil, it was nasty.

I formated.

now it know's who's the boss, and it went off to the virus graveyard in cyberspace.

[xtothez]

Originally posted by Costa
its the amount of ram,.

and Windows 2000 AS is built apon dual processor technology meaning it can address 32 + 32 bits addresses (in theory)

"heh"

queball just put them out of their misery and stop trolling on semantics : )

[Fifth_teletubbie]

Originally posted by queball


stuff

Satisfying the questions of little boys who are full of themselves isnt my job.

If you truly want to know more, go read a book on the subject, but I doubt it.

[Structural Integrity]

Originally posted by Costa
its the amount of ram,.

and Windows 2000 AS is built apon dual processor technology meaning it can address 32 + 32 bits addresses (in theory)

hmmmm....
Untrue

A computer does not only have to address RAM, it also has to address hardware devices like parallel ports, USB thingies, harddrives and other external devices. FAR more than you could address in 32 bits.

I've always learnt that the 32 does not indicate the addressed space, but the max length of the processor opcodes and register size. In normal language that means that the datachunks processed in one processor cycle can be 32 bits.

Next generation of windows will have the 64 bit kernel (and thus probably the kernel64.dll file) for the new 64 bit processor familly like the Athlon 64 and new Pentium 64 processors.

So, the 32 has NOTHING to do with address range but with register size of the platform.

[queball]

Originally posted by Structural Integrity
So, the 32 has NOTHING to do with address range but with register size of the platform.

That's still confusing since ia32 still has 16-bit and 8-bit registers, and I'm sure kernel32.dll uses its 8-bit registers plenty.

kernel32.dll is part of something called "Win32". From a developer's point of view one big difference from the old API is that HANDLE's are now 32-bit. From a user's point of view there's no fundamental difference, it's simply a different name. Certainly nothing to do with RAM.

[W]

Originally posted by Structural Integrity
hmmmm....
Untrue

A computer does not only have to address RAM, it also has to address hardware devices like parallel ports, USB thingies, harddrives and other external devices. FAR more than you could address in 32 bits.

I've always learnt that the 32 does not indicate the addressed space, but the max length of the processor opcodes and register size. In normal language that means that the datachunks processed in one processor cycle can be 32 bits.

Next generation of windows will have the 64 bit kernel (and thus probably the kernel64.dll file) for the new 64 bit processor familly like the Athlon 64 and new Pentium 64 processors.

So, the 32 has NOTHING to do with address range but with register size of the platform.

The register size does however limit the size of a directly adressable segment/memory page

[Fifth_teletubbie]

Originally posted by Structural Integrity

So, the 32 has NOTHING to do with address range but with register size of the platform.

Those of us old enough to have done 16-to-32 bit software conversions (and 8 to 16 for that matter) would beg to differ

Simply put, every bit of data on any given system (not specifically talking PCs here) has an address which describes where it can be found in storage (memory). This address is a number which indicates the location, and is used in basically every instance where data is referenced, transferred, manipulated etc.
For this we use 'pointers', which everyone who has ever done any C or PL/1 will know.

Moving from 16-bit to 32-bit means that your pointers (which indicate locations within storage) move from using 16 bits to using 32 bits, or rather go from 00-FF to 0000-FFFF.

A lot of 'clever' old programming had to be completely rewritten because of programmers making use of either specific memory locations or using system limits instead of programmed limits etc.

The 32 in 32-bit when talking about _operating systems_ has EVERYTHING to do with addressing ranges.
32-bit is however also often used for hardware, where it (usually) indicates the width of a databus or something, and is an entirely different subject really.

[queball]

Originally posted by Fifth_teletubbie

Hence Medium/Compact/Large memory model hacks. I don't deny that lots of 16-bit things turned into 32-bit things but the difference between Win16 and Win32 is more than bitwidths.

So, saying "Actually it's the maximum range of storage the operating system can address." is just wrong. Any explanation is as good as any other. Correcting someone, then proclaiming how oldschool you are (where do you get this idea that you have more experience than others, anyway?), is just arrogance.

[Fifth_teletubbie]

Originally posted by queball
Hence Medium/Compact/Large memory model hacks. I don't deny that lots of 16-bit things turned into 32-bit things but the difference between Win16 and Win32 is more than bitwidths.

So, saying "Actually it's the maximum range of storage the operating system can address." is just wrong. Any explanation is as good as any other. Correcting someone, then proclaiming how oldschool you are (where do you get this idea that you have more experience than others, anyway?), is just arrogance.

Like I said, if you truly want to know what you are blabbering about, read a book for once kid.

[Mong]

Any virus will be removed by a complete format and reinstall. Which is what you should do anyway, if you buy a machine off somebody.

In fact, say to him "I'll take it, but I'm gonna have to format and reinstall, so knock the price down".

M.

[Structural Integrity]

Originally posted by Fifth_teletubbie
Like I said, if you truly want to know what you are blabbering about, read a book for once kid.

The main difference between 8, 16 and 32 bit platforms is again the data size which resulted in the ability to process bigger numbers in one cycle. 8 bit processors processed a 4 byte integer in four CPU cycles. A 32 bit system does this in one cycle. That's why people are sceptical about the newest 64 bit systems because there is little need for PC's to process such big numbers. Addressable memory is not such an issue because you already can address more memory than the register width allows you to.

I never really paid attention in class when about memory management, but I guess that you can address more memory when you use more CPU cycles (ie: split the address in two, a page and an address within that page... and there W has a point that the register size limits the page size).
Ofcourse, there a loss of speed, but it's possible (and used).

[Fifth_teletubbie]

Originally posted by Structural Integrity
The main difference between 8, 16 and 32 bit platforms is again the data size which resulted in the ability to process bigger numbers in one cycle. 8 bit processors processed a 4 byte integer in four CPU cycles. A 32 bit system does this in one cycle. That's why people are sceptical about the newest 64 bit systems because there is little need for PC's to process such big numbers. Addressable memory is not such an issue because you already can address more memory than the register width allows you to.

I never really paid attention in class when about memory management, but I guess that you can address more memory when you use more CPU cycles (ie: split the address in two, a page and an address within that page... and there W has a point that the register size limits the page size).
Ofcourse, there a loss of speed, but it's possible (and used).

You are correct, but you are talking about CPUs. I am talking about Operating Systems. It isnt the same.
Also note I am not specifically talking about PCs either, in fact most of my experience comes from IBM s390 systems.

[Structural Integrity]

Originally posted by Fifth_teletubbie
You are correct, but you are talking about CPUs. I am talking about Operating Systems. It isnt the same.
Also note I am not specifically talking about PCs either, in fact most of my experience comes from IBM s390 systems.

Now you got me thinking... is a pointer in a 32bit environment actually 32 bits large? I mean, you have pointers to files, to external ports and other devices. Certainly the range in which these addresses lie is bigger than 32 bits?
I don't think a program can only address one page of memory... that would be too much of a limitation. No?

[Fifth_teletubbie]

Originally posted by Structural Integrity
Now you got me thinking... is a pointer in a 32bit environment actually 32 bits large? I mean, you have pointers to files, to external ports and other devices. Certainly the range in which these addresses lie is bigger than 32 bits?
I don't think a program can only address one page of memory... that would be too much of a limitation. No?

Well with 32 bits you can potentially address 4GB. Some however use a parity bit and so only address 2GB. This is also the limit for memory on some motherboards btw, which probably isnt a coincidence.

I dont really understand what you mean by 'range in which these adresses lie is bigger than 32 bits'. An address is just a number, and can be as small as '1'.

'paging' as far as I know is moving data from RAM to permanent storage and back. So I'd say that a program by definition can only access one 'page' at a time, otherwise there would be no sense in 'paging' since keeping things in RAM is much faster.
Of course a 'page' would only be as big as the space allotted for that particular program, which on shared systems could be anything.

[queball]

Originally posted by Fifth_teletubbie
You are correct, but you are talking about CPUs. I am talking about Operating Systems. It isnt the same.
Also note I am not specifically talking about PCs either, in fact most of my experience comes from IBM s390 systems.

Win16 used 16-bit handles. This has nothing to do with storage space or even memory (though what W said is correct). By using segment magic 16-bit MSWindows 3.11 can address up to 32MB of RAM, and Win2K can address up to 8GB. PCs are somewhat different from s390s it would seem.



Edit: here's an MSDN article that explains this in depth for anyone interested.

[Miccksey]

euh heh

so basically FAT32 can store less data then FAT16 ;-)

[Fifth_teletubbie]

Originally posted by queball
... storage space or even memory ...

When I talk about storage, I mean program storage , which is memory. Data storage (ie tape, DASD, harddisk, floppy etc) is something else entirely.
Unclear perhaps for those used to different terminology.

STD aint it?