Does OpenCL allow concurrent writes to same memory address?

Posted by Wonko on Stack Overflow See other posts from Stack Overflow or by Wonko
Published on 2010-05-18T21:50:59Z Indexed on 2010/05/21 10:40 UTC
Read the original article Hit count: 172

Filed under:

concurrency

|

shared-memory

|

opencl

Is two (or more) different threads allowed to write to the same memory location in global space in OpenCL? The write is always changing a uchar from 0 to 1 so the outcome should be predictable, but I'm getting erratic results in my program, so I'm wondering if the reason can be that some of the writes fail.

Could it help to declare the buffer write-only and copy it to a read-only buffer afterwards?

© Stack Overflow or respective owner

Related posts about concurrency

SQL SERVER – Concurrency Basics – Guest Post by Vinod Kumar

as seen on SQL Authority - Search for 'SQL Authority'
This guest post is by Vinod Kumar. Vinod Kumar has worked with SQL Server extensively since joining the industry over a decade ago. Working on various versions from SQL Server 7.0, Oracle 7.3 and other database technologies – he now works with the Microsoft Technology Center (MTC) as a Technology… >>> More
Http requests / concurrency?

as seen on Stack Overflow - Search for 'Stack Overflow'
Say a website on my localhost takes about 3 seconds to do each request. This is fine, and as expected (as it is doing some fancy networking behind the scenes). However, if i open the same url in tabs (in firefox), then reload them all at the same time, it appears to load each page sequentially rather… >>> More
Mnesia: A Distributed DBMS Rooted in Concurrency

as seen on Internet.com - Search for 'Internet.com'
Find out what makes Mnesia, the Erlang-based database management system, perfect for distribution across a network of computers. >>> More
C# concurrency vs Java concurrency, which is neatly designed? which is better?

as seen on Stack Overflow - Search for 'Stack Overflow'
I am hard core fan for Doug Lee and his concurrency framework. I have also (from twitter) came across a comment from Joshua Bloch that java concurrency is one of the best framework (guess that he compared to .Net). But some of you worked in both Java Concurrency and .NET concurrency. What is your… >>> More
Queued Loadtest to remove Concurrency issues using Shared Data Service in OpenScript

as seen on Oracle Blogs - Search for 'Oracle Blogs'
Queued Processing to remove Concurrency issues in Loadtest ScriptsSome scripts act on information returned by the server, e.g. act on first item in the returned list of pending tasks/actions. This may lead to concurrency issues if the virtual users simulated in a load test scenario are not synchronized… >>> More

Related posts about shared-memory

Can't connect to SQL Server 2008 - looks like Shared Memory problem

as seen on Stack Overflow - Search for 'Stack Overflow'
I am unable to connect to my local instance of SQL Server 2008 Express using SQL Server Management Studio. I believe the problem is related to a change I made to the connection protocols. Before the error occurred, I had Shared Memory enabled and Named Pipes and TCP/IP disabled. I then enabled both… >>> More
Instantiating class with custom allocator in shared memory

as seen on Stack Overflow - Search for 'Stack Overflow'
I'm pulling my hair due to the following problem: I am following the example given in boost.interprocess documentation to instantiate a fixed-size ring buffer buffer class that I wrote in shared memory. The skeleton constructor for my class is: template<typename ItemType, class Allocator > SharedMemoryBuffer<ItemType… >>> More
Listing all shared memory segments used by a process on AIX5.3+

as seen on Stack Overflow - Search for 'Stack Overflow'
I would like to find all shared memory segments used by a given process. I am especially interested in figuring out the shmid so i can use it in calls to shmctl(). On Solaris i would just read /proc/$PID/map to figure out that information (field pr_shmid). The contents of that file are defined by… >>> More
A quick design question about C++ container classes in shared memory

as seen on Stack Overflow - Search for 'Stack Overflow'
I am writing a simple wrapper around boost::interprocess's vector container to implement a ring buffer in shared memory (shm) for IPC. Assume that buf is an instance of RingBuffer created in shm. Now, in its ctor, buf itself allocates a private boost::interprocess::vector data member to store values… >>> More
CUDA: When to use shared memory and when to rely on L1 caching?

as seen on Stack Overflow - Search for 'Stack Overflow'
After Compute Capability 2.0 (Fermi) was released, I've wondered if there are any use cases left for shared memory. That is, when is it better to use shared memory than just let L1 perform its magic in the background? Is shared memory simply there to let algorithms designed for CC < 2.0 run efficiently… >>> More