Experiments in program optimisation ququMy blogwewe · Scour

From macros to templates

pzemtsov.github.io·49w

How to make C run as fast as Java

pzemtsov.github.io·49w

Size does matter

pzemtsov.github.io·49w

De-multiplexing of E1 streams in Java

pzemtsov.github.io·49w

De-multiplexing of E1 stream: converting to C

pzemtsov.github.io·49w

One handy C idiom: multi-statement macro

pzemtsov.github.io·49w

The mystery of an unstable performance

pzemtsov.github.io·49w

Fast E1 de-multiplexing in C using SSE/AVX

pzemtsov.github.io·49w

A bug story: testing your test code

pzemtsov.github.io·49w

Game of Life, hash tables and hash codes

pzemtsov.github.io·49w

Bug story 2: Unrolling the 16×16 matrix transposition, or be careful with macros

pzemtsov.github.io·49w

How cache affects demultiplexing of E1

pzemtsov.github.io·49w

Standard input, buffering and affinity mask

pzemtsov.github.io·49w

On the benefits of stream buffering in Java

pzemtsov.github.io·49w

How to transpose a 16×16 byte matrix using SSE

pzemtsov.github.io·49w

Macros, call by name, lambda and Jensen's device

pzemtsov.github.io·49w

Lambdas and other callbacks in Java

pzemtsov.github.io·49w

More experiments with cache

pzemtsov.github.io·49w

Hash functions: disassembling and measuring performance

pzemtsov.github.io·49w

More optimisations of division-based hash: Karatsuba's method

pzemtsov.github.io·49w

Log in to enable infinite scrolling