the difference between 32 bits driver and 64bits driver

Question

0.00/5 (No votes)

See more:

Anyone who know this？ I search the microsoft， It only explain briefly， I see that the windows 7 64bits driver needs a signature。what is that means？

Posted 5-May-11 17:29pm

markfilan

Updated 5-May-11 18:01pm

Hans Dietrich

v2

Add a Solution

3 solutions

Solution 3

Just looking at your title, 32bits, 64bits refers the way that computer process data. In other words it define the architecture that memory address length, handling by the processor. If you have an old CPU you may have to use 32bit support OS, but for the newer once it should be 64bits versions. More importantly, having 64bit OS doesn't run all the application faster because they may be tested and optimized on 32bits OS. So you must choose the best accordingly.

I'm not sure what you have really means by signature. If you can please post the message so I can comment more on it. Anyway driver signatures are kind of certifications that vendors attache with it. Most of the time signatures are comes with commercial applications, which are NOT free of charge. There are some third party tools to disable those, but I don't like to discuss about them. It's not ethical at all.

Posted 5-May-11 17:53pm

CodingLover

Add a Solution

Add your solution here

Treat my content as plain text, not as HTML

Preview 0

…

Existing Members

Sign in to your account

...or Join us

Download, Vote, Comment, Publish.

Your Email
Password
Forgot your password?

Your Email
This email is in use. Do you need your password?
Optional Password

I have read and agree to the Terms of Service and Privacy Policy
Please subscribe me to the CodeProject newsletters

When answering a question please:

Read the question carefully.
Understand that English isn't everyone's first language so be lenient of bad spelling and grammar.
If a question is poorly phrased then either ask for clarification, ignore it, or edit the question and fix the problem. Insults are not welcome.
Don't tell someone to read the manual. Chances are they have and don't get it. Provide an answer or move on to the next question.

Let's work to help developers, not make them feel stupid.

This content, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)

barneyman · Accepted Answer · 2011-05-05T17:45:00

The fundamental difference is the compiler used, and the size of an int - 32bits vs 64bits

64bit operating systems (win7, xp64) need 64bit drivers to access devices, similarly a 32bit OS needs 32bit drivers - they generally operate in kernel space, which is why they need to be native to the chip architecture

In the past, anyone could write a driver and bolt it into your system; drivers, though not hard to write, have very low-level access to the OS and hardware, and a poorly written one could cause chaos - so there are now a couple of safeguards

Code Signing - it's more difficult to install a driver that hasn't been digitally signed

Kernel Mode Signing - you must have a kernel mode driver signed by microsoft, after passing compliance

details here

http://msdn.microsoft.com/en-us/windows/hardware/gg487317[^]

Sergey Alexandrovich Kryukov · Accepted Answer · 2011-05-05T17:50:00

Solution 2

There is no big difference between 32-bit and 64-bit drivers. They are just for different CPU architectures, and the codes for different architectures are incompatible. Win32 applications can work on both 64-bit architectures (I64 (Itanium) and x86-64), but not drivers.

The drives are signed so it's possible to check up if they are authentic. This can help to understand the idea:
http://en.wikipedia.org/wiki/Digital_signature[^],
see also http://en.wikipedia.org/wiki/Asymmetric_key_algorithm[^].

For driver signing, see http://msdn.microsoft.com/en-us/library/ff549499(v=vs.85).aspx[^].

—SA

Posted 5-May-11 17:50pm

Sergey Alexandrovich Kryukov

Comments

markfilan 6-May-11 3:00am

thanks a lot.

Sergey Alexandrovich Kryukov 6-May-11 3:11am

You're welcome.
Will you formally accept this answer (green button)?
--SA

Kurt Degiorgio 9-May-11 10:59am

Reason for 2: when it comes to developing drivers 64bit and 32 bit makes a huge difference Your 32 bit designed driver generally wont work if you just compile it with a 64 compiler (by wont work I mean it will crash the OS a lot) to name a few areas that give problems ptr sizes are different , structures are processed differently also the most problematic area is if the user space application you are communicating with is 32bit. (you can geuss what type of problems arise here)

Sergey Alexandrovich Kryukov 9-May-11 21:14pm

You're absolutely right. Whatever you named is correct; and all of these items are not big difference. Not that we disagree in anything at all. You just voted with no point. You're unhappy with just a words "no big difference". This is just words. If you understand what is really important, you can see that changing the figure 32 to 64 change nothing but... this figure. Pointer sizes double, registers are replaced with 64-bit equivalents. That's all! If driver is written in C, the difference in code will be nearly invisible.

So, I have to repeat in case it is not clear: no big difference.
--SA

Kurt Degiorgio 10-May-11 3:30am

Yes on paper you are right but my point is in reality you can not design a 32 bit driver and expect it too work on a 64bit machine by just "re-compiling it" it simply won't wont work. As such in a professional environment when developing driver if you are targeting 64bit platforms as well you will need to design your driver with this in mind. I can give you countless horror stories of 32bit designed drivers crashing on 64 bit machines. remember kernel mode programming is different from user mode programming since you have too deal with different IRQL levels and paged memory something which the OS abstracts at user mode.
(just my 2 cents)

Sergey Alexandrovich Kryukov 10-May-11 13:27pm

OK, maybe you know better; and I simply do not understand these big differences. I know some kernel-mode programming, but I did it only in Assembly where you really cannot "recompile" but need to re-write almost every line. But I really don't understand how the size of registers and pointers can affect, say, IRQ levels. Let me put it this way: I'm mostly familiar with transition between 16-bit and 32-bit architectures. That was really a change that had very little to do with address size; but it was absolutely different architecture (depending on OS architecture though). So, when I said "not a big difference" in my answer, I referred to that situation. So our discussion is more of the cultural clash... :-)
--SA