[back to index]

• Memory Reordering - Intro
• Sequential Consistency (SC)

without further notes, all contents are copy-verbatim or derived from Chapter 3 of

A Primer on Memory Consistency and Cache Coherence (2nd Edition), Synthesis Lectures on Computer Architecture, Vijay Nagarajan, Daniel J.Sorin, Mark D.Hill and David A. Wold
https://pages.cs.wisc.edu/~markhill/papers/primer2020_2nd_edition.pdf

this page is under CC-BY-SA-4.0. The above notice must be preserved.

§ Memory Reordering - Intro

Example of memory reordering: core 2 spins upon flag. Programmer may assume Core 2 reads r2 == NEW because Core 1 sets data before setting flag. However S1 and S2 could get reordered, thus Core 2 exits the spin and reads old data before NEW is stored.

data = OLD;
flag = !SET;

// core 1
S1: store data = NEW;  
S2: store data = SET;  

// core 2
L1: Load r1 = flag; 
B1: if (r1 != SET) goto L1; 
L2: Load r2 = data; 

Sequential execution (von Neumann) model generally requires that operations to the same address execute in the original program order. The reordering discussed only happen on access to different addresses.

How a core might reorder memory access.

• Store-store reordering
• Load-load reordering
• Load-store and store-load reordering

§ Sequential Consistency (SC)