Glossary
8 accumulators
For ADDPD YMM (latency 4, throughput 2): 4/2 = **2 accumulators**
Learn More
Assembly Language
—
Chapter 33: Performance Analysis and Optimization
Related Terms
"Branch Misprediction Cost"
"Branchless Equivalents of Simple Functions"
"Data Structures in the Linux Kernel"
"Engineering a Compiler" by Cooper and Torczon
"Exploiting the Hard-Working DWARF"
"Falsehoods Programmers Believe About Money"
"Function Call Conventions and Stack Frame Layout"
"Function Call Overhead"
We use cookies to improve your experience and show relevant ads.
Privacy Policy
Essential Only
Accept All