Contents
Contents
1
Introduction
2
Overview of Alpha Family
3
Linux Performance Analysis Tools
3.1
Accurately Measuring Time
3.2
Performance Counters
3.3
GNU
gprof
4
Optimization Techniques: Making Your Applications Fly
4.1
Avoiding Integer Division
4.2
Keeping the Translation Lookaside Buffer in Mind
4.3
Avoiding Memory Accesses
4.4
Lending
gcc
a Hand: Separating Loads/Stores from Computation
4.5
Improving the Memory Access Pattern
4.6
Data-parallel Processing: MPEG Core Loop
5
Conclusions