I am working on doing some significant optimization of some machine vision code on an embedded AMD Geode LX. I am going as far as to rewrite the computationally intense portions in Assembly, making heavy use of the x86 MMX instructions.
The basic MMX instruction set has many references, but I am looking for specific guides for optimizing for the Geode LX.
The AMD website has a few AMD64 and Athlon optimization guides, but I have had no luck finding any information on the Geode specifically.
I am looking for information on such Geode specific things as pipelining, speed of different instructions, and other speed critical information. Information just on how the Geode works, as well as tips for improving code, would be very helpful.
I am working with an AMD Geode LX 800.
[edit] This website is exactly what I’m looking, except it only focuses on AMD Athlon and Opteron processors:
http://www.agner.org/optimize/
Does this help?
http://support.amd.com/us/Embedded_TechDocs/gx1_ds.pdf
[edit] or this:
http://support.amd.com/us/Embedded_TechDocs/33234H_LX_databook.pdf