|
 |
|
|
AMD Code Samples for Microsoft Visual Studio Microsoft Visual C++
| Microsoft Visual C++ Support for AMD Processors |
| |
Both Microsoft Visual C++.NET and Visual C++ 6.0 provide
full support for Advanced Micro Devices processors, including support
for 3DNow! Technology, Enhanced 3DNow! and MMX instruction sets. Features such as 3DNow! Compiler Intrinsics & In-Line Assembly, Macro Assembler
(MASM), Run-Time Aligned Memory Management, and Debugging and Disassembly,
are native to the Visual C++.NET compiler and are available via a Processor
Pack for Visual C++6.0. These features will enable easier, straight forward
and more productive code optimization for the AMD Athlon and Duron family
of processors as well as for the K6-family (including embedded versions).
Please
click here to download Microsoft Visual C++ 6.0 Processor
Pack
Microsoft, Visual C++, Visual Studio are either registered
trademarks or trademarks of Microsoft Corporation in the
United States and/or other countries.
|
| |
| AMD Code Samples |
| |
The following "OpTimer" examples demonstrate the use of
3DNow! Compiler Intrinsics and In-Line Assembly in a test program that
can be used as a tool to determine the optimization quality of a function.
For each of the examples, expand the .Zip archives, and open the OpTimer.dsw
workspace to view an unrolled C, an optimized Inline Assembly,
and an Intrinsic Implementation of a 4 x 4 matrix transformation
(Example 1) or multiplication (Example 2). A 3 x 3 Matrix Addition and
detailed documentation on the "OpTimer" test program are available in
the AMD Athlon™ SDK.
*By downloading any of the software from this page, you
agree to be bound by the terms of the software license agreement.
|
| |
| Download
3DNow Intrinsic code samples (.Zip, 12 kb) |
| |
The archived examples demonstrate the
use of 3DNow! Compiler Intrinsics. The intrinsics _m_to_float
and _m_from_float require Microsoft Visual Studio.NET.
|
| |
| Download
memcpy() code sample (.Zip, 39 kb) |
| |
This memcpy() code is optimized for AMD
Athlon and Duron processors. It uses MMX instructions
and prefetch, and employs "non temporal" memory
writes on large blocks, which bypass the cache for better
efficiency.
This code typically provides significantly
improved performance, as indicated in the example graph.
Performance gains are dependent on particular system specs,
including CPU speed, CPU type, chip set, main memory type,
and main memory speed. The data block size and alignment
are also factors. Developers should test their applications
to determine their exact performance benefit.
|
| |
| Download
OpTimer Code Sample 1 (.Zip, 16 kb) - Readme.txt |
| |
OpTimer with a 4x4 Matrix Transformation -
Source & Microsoft Developer Studio build project. |
| |
| Download
OpTimer Code Sample 2 (.Zip, 17 kb) - Readme.txt |
| |
OpTimer with a 4x4 Matrix Multiplication
- Source & Microsoft Developer Studio build project.
|
| |
|
|
|
|
 |
|