Index
> Windows > mix use16 with use32 |
| Author |
|
|
luish 06 Feb 2018, 07:59
How to mix 16-bit code in a and 32-bit program withe use16 and use32?
|
|||
06 Feb 2018, 07:59 |
|
|
luish 06 Feb 2018, 09:27
revolution wrote: In Windows you can but it is tricky. Windows doesn't officially support it so earlier/later versions might do different things, but basically you need to manipulate the segment registers. It is not for the faint-of-heart, it requires some degree of finesse and a lot of debugging to get it right. i only want know how to create 16/32 bits segments with use16 and use32 |
|||
|
06 Feb 2018, 09:27 |
|
|
revolution 06 Feb 2018, 09:30
In Windows you can't create segments, but there are existing selectors that are predefined that you can use. NTVDM uses these to run 16-bit code. So you can set CS to select the 16-bit selector.
|
|||
|
06 Feb 2018, 09:30 |
|
|
luish 06 Feb 2018, 09:32
for example:
Code: section ".code" code executable readable writeable use32 main: CALL PROCEDURE ret section ".code16" code executable readable writeable use16 proc keylogger in al,60h ... |
|||
|
06 Feb 2018, 09:32 |
|
|
revolution 06 Feb 2018, 09:35
You would need a far call to set CS to a new value. And the IN/OUT instructions are trapped in the OS so you won't get any real values from them. Plus you will need 32-bit code to handle the trap exceptions that Windows passes on to you.
But you can't run a keylogger in 16-bit code. Windows runs in protected mode so interrupts and hardware access won't ever go to user mode code. You will need a signed driver to load code into the kernel space. Or you can start your system in unsigned-driver mode and load an unsigned kernel mode driver. |
|||
|
06 Feb 2018, 09:35 |
|
< Last Thread | Next Thread > |
Forum Rules:
|