Introduction to Intel Advanced Vector Extensions

Introduction to Intel Advanced Vector Extensions

I just read this article:

http://software.intel.com/en-us/articles/introduction-to-intel-advanced-vector-extensions/

it's a fairly good introduction to AVX though in the "Instruction Set Overview"the author states that "As a result, the VEX instructions can only be used when running in 64-bit mode." I know from first hand experience that this is wrong since my own AVX coderunfine in32-bit mode on a Sandy Bridge CPU. I'll suggest to fix this error in the article since newcomers to AVX may be misled by this.

2 posts / 0 new
Last post
For more complete information about compiler optimizations, see our Optimization Notice.

nice catch, wrong statement for sure, we will fix it,
thanks! ... author was close in understanding of LDS/LES opcodes re-use for VEX
prefix, but what happens in 32-bit is that VEX prefix, although still using
LDS/LES opcodes, always ends up in unsupported forms of these instructions (those
with memory operands), thanks to bit-complement encoding of some fields, hence everything
is backwards compatible in 32-bit mode too

-Max

Leave a Comment

Please sign in to add a comment. Not a member? Join today