Help on floating point support in windows kernel

NTDEV
1

hi all
i wanted to know if the windows 2000 kernel and windows xp kernel
for Intel Pentium IV processors supports the following or not and if yes
how:-
—floating point calculations in kernel
–MMX instructions in kernel
–SSE/SSE2 instructions in kernel
–Assembly language instructions in kernel

any help would be highly appreciated
Also i would like to thank all the developers in helping me understand the
kernel
and write a real time kernel driver in windows 2000

thanks
Mayank

2

> —floating point calculations in kernel

Yes with limitations, KeSave/RestoreFloatingPointState is a must, see the
documentation on these routines for the limitations.
There is also an MSDN article “Using Floating Point or MMX in a WDM Driver”.

–MMX instructions in kernel

Same as with floating point.

–SSE/SSE2 instructions in kernel

Again same as floating point.

–Assembly language instructions in kernel

Why not? Just use __asm{} in the C/C++ source.
Note that this will make the code unportable to 64 bits, and so, this must be
used with care, if there are no other ways of solving the task.

Maxim Shatskih, Windows DDK MVP
StorageCraft Corporation
xxxxx@storagecraft.com
http://www.storagecraft.com