I use the following routines quite often in my code to determine the type of cpu that is available in the system and for timing routines to determine application and function performance. When writing optimized code for assembler that will run on many machines, it is important to know what capabilities are available on the processor and often times several functions will be written based on the same algorithm (such as one using the general purpose registers, one using MMX, one using SSE, etc.) and the best algo to run on the target system will be determined at runtime using code from the dll. Many of the functions in the dll are only accessible using assembly because they call the cpu directly. I wrote a very simple VB application that shows how to access the dll which you can download below. Please notify me with any corrections or additions to the code.
Download VB Test Application with proc.dll
The proc.dll code is written in MASM and is commented heavily. Reading through it may help if your just starting out with ASM or optionally you can compile your own version. I just list the individual functions and did not post the boiler plate code that is required for creating a win32 dll. If you would like proc.dll customized and compiled specifically for your application please contact me and I would be happy to for a small fee.