Are indivisible operations still indivisible on multiprocessor and multicore systems?

Posted by Steve314 on Stack Overflow See other posts from Stack Overflow or by Steve314
Published on 2010-03-04T22:41:51Z Indexed on 2010/03/08 8:51 UTC
Read the original article Hit count: 268

Filed under:

multicore

|

multiprocessing

|

multithreading

|

thread-safety

As per the title, plus what are the limitations and gotchas.

For example, on x86 processors, alignment for most data types is optional - an optimisation rather than a requirement. That means that a pointer may be stored at an unaligned address, which in turn means that pointer might be split over a cache page boundary.

Obviously this could be done if you work hard enough on any processor (picking out particular bytes etc), but not in a way where you'd still expect the write operation to be indivisible.

I seriously doubt that a multicore processor can ensure that other cores can guarantee a consistent all-before or all-after view of a written pointer in this unaligned-write-crossing-a-page-boundary situation.

Am I right? And are there any similar gotchas I haven't thought of?

© Stack Overflow or respective owner

Related posts about multicore

Multicore Chips Require OS Rework, Windows Architect Advises

as seen on Devx - Search for 'Devx'
Microsoft's Dave Probert says the OS kernel needs to act more like a hypervisor. >>> More
Multicore Chips Require OS Rework, Windows Architect Advises

as seen on Internet.com - Search for 'Internet.com'
Microsoft's Dave Probert says the OS kernel needs to act more like a hypervisor. >>> More
Ct Technology: A Flexible Parallel Programming Model for Multicore and Many-Core Architectures

as seen on Dr Dobbs - Search for 'Dr Dobbs'
Ct leverages the strengths of data parallel programming to help address challenges of software development for multicore and many-core Parallel computing - Multi-core processor - Programming - Microsoft - Languages >>> More
A Real-Time HPC Approach for Optimizing Multicore Architectures

as seen on Dr Dobbs - Search for 'Dr Dobbs'
Complex math is at the heart of many of the biggest technical challenges. With multicore processors, the type of calculations that would have required a supercomputer can now be performed in real-time, embedded environments. High-performance computing - Supercomputer - Real-time computing - Operating… >>> More
How are you taking advantage of Multicore?

as seen on Stack Overflow - Search for 'Stack Overflow'
As someone in the world of HPC who came from the world of enterprise web development, I'm always curious to see how developers back in the "real world" are taking advantage of parallel computing. This is much more relevant now that all chips are going multicore, and it'll be even more relevant when… >>> More

Related posts about multiprocessing

Comparison of the multiprocessing module and pyro?

as seen on Stack Overflow - Search for 'Stack Overflow'
I use pyro for basic management of parallel jobs on a compute cluster. I just moved to a cluster where I will be responsible for using all the cores on each compute node. (On previous clusters, each core has been a separate node.) The python multiprocessing module seems like a good fit for this… >>> More
How to synchronize a python dict with multiprocessing

as seen on Stack Overflow - Search for 'Stack Overflow'
I am using Python 2.6 and the multiprocessing module for multi-threading. Now I would like to have a synchronized dict (where the only atomic operation I really need is the += operator on a value). Should I wrap the dict with a multiprocessing.sharedctypes.synchronized() call? Or is another way… >>> More
Python - multithreading / multiprocessing, very strange problem.

as seen on Stack Overflow - Search for 'Stack Overflow'
import uuid import time import multiprocessing def sleep_then_write(content): time.sleep(5) print(content) if __name__ == '__main__': for i in range(15): p = multiprocessing.Process(target=sleep_then_write, args=('Hello World',)) p… >>> More
Python - Help with multiprocessing / threading basics.

as seen on Stack Overflow - Search for 'Stack Overflow'
I haven't ever used multi-threading, and I decided to learn it today. I was reluctant to ever use it before, but when I tried it out it seemed way to easy, which makes me wary. Are there any gotchas in my code, or is it really that simple? import uuid import time import multiprocessing def sleep_then_write(content): … >>> More
Solving embarassingly parallel problems using Python multiprocessing

as seen on Stack Overflow - Search for 'Stack Overflow'
How does one use multiprocessing to tackle embarrassingly parallel problems? Embarassingly parallel problems typically consist of three basic parts: Read input data (from a file, database, tcp connection, etc.). Run calculations on the input data, where each calculation is independent of any other… >>> More