Questions tagged [atomic]
An atomic operation is one which can be considered, from the perspective of the computing context in which it occurs, to be executed in a single step. Atomic operations either succeed entirely or fail, with no intermediate states.
65
questions
4
votes
3
answers
425
views
C11 zero copy lock-free triple buffer
The code is for a single producer, single consumer scenario, where the consumer only cares about the latest information shared by the producer.
This is just a simple proof of concept created for linux....
- 49
0
votes
1
answer
125
views
C++ slim condition notifier
When writing multi-threading code, one often need the thread to wait for some condition being met. A naive approach would look like this:
...
- 137
1
vote
1
answer
167
views
Lock Guard Atomic alternative
I've recently written a Vulkan library for creating 2D applications with ease. The catch was I need std::lock_guard for my window resize event to resize resources ...
- 579
1
vote
1
answer
99
views
Implementation of a lock free queue using CompareAndSwap in Go
Here is my implementation of a lock free queue using CompareAndSwap operation.
...
3
votes
2
answers
239
views
atomic spinlock mutex class
This here is the follow-up to this question. I was recommended to implement a Lockable type (similar to std::mutex) that can work with ...
- 635
8
votes
3
answers
2k
views
Basic RAII spinlock class
I have written the following class that acts as a simple lock for mutual exclusion:
...
- 635
0
votes
1
answer
137
views
Lock-free implementation of getAndUpdate() using atomic CAS (Compare-And-Swap) operation
We have the following class written in Kotlin Native with the new Memory Manager (which doesn't require to freeze objects):
...
- 101
4
votes
2
answers
1k
views
Mutually exclusive execution using std::atomic?
I am currently learning more about lock free programming and wondered how I could implement mutual exclusion using std::atomics. I implemented the following code to ...
- 145
0
votes
1
answer
220
views
Flippable atomic boolean
I was trying to implement a boolean that can be atomically flipped.
The suggestion on Stack Overflow is to use an integer.
This would be my implementation.
...
- 133
1
vote
0
answers
173
views
Test and set a bit atomically using inline aarch64 assembly
The code below has been a vehicle for learning the syntax of inline assembly for gcc/clang and that is the focus of my question. In addition to anything else that I've missed I would also like ...
- 111
3
votes
2
answers
386
views
C++11 revised `std::latch` implementation
This question follows up on this question.
After turning the while-loop into a conditional wait using std::condition_variable, I ...
- 43
1
vote
1
answer
283
views
Implementation of a latch
As an exercise to learn more about multi-threading and atomic operations work in C++, I decided to implement a latch class in C++11 loosely based off of std::latch ...
- 43
22
votes
3
answers
6k
views
C++ lock-free, MPMC Ring buffer in C++20
I have some performance critical inter-thread messaging code in C++. Multiple producers, one or more consumers.
Profiling dozens of iterations of this messaging code over several years of development, ...
- 364
6
votes
1
answer
1k
views
Implementation of std::atomic<std::shared_ptr<T>> for C++20
As you may know, C++20 has added std::atomic<std::shared_ptr<T>> specialization to the standard, but sadly, most compilers have not implemented it yet. ...
- 295
1
vote
1
answer
491
views
3
votes
2
answers
351
views
A toy example of the concurrent stack via atomic variables and CAS pattern
I've written a toy example of the concurrent stack which has only three functions push(), peek(), and ...
- 33
5
votes
1
answer
2k
views
Count of request per second, minute in java
I want to count number of requests per second, minute. I want do it fast and thread-safe. I have 2 variables to count:
...
- 151
2
votes
1
answer
324
views
4
votes
1
answer
885
views
C++ latch implementation
Since std::latch is not in many standard C++ libraries, I tried implementing my own, is it OK from memory ordering perspective or ...
- 141
4
votes
2
answers
1k
views
Atomic fixed-size parallel std::vector wrapper
I'm in need of a simple wrapper for a fixed-size std::vector that can ensure atomic insertion.
Since my only real problem is with atomic insertion and I know ...
1
vote
1
answer
308
views
Custom bounded memory pool allocator
Recently I put together a custom fixed size (bounded) memory pool for my job system implementation to support its internal memory management. Since it will be accessed by different threads I want to ...
- 155
4
votes
0
answers
85
views
C++14 AtomicRoundRobinPool for sharing a pool of keys
The idea here is that we have some resource that is relatively very expensive to generate, but once we have a bunch of them in a pool, we can keep reusing them instead of generating new ones. "HTTP ...
- 19.3k
1
vote
1
answer
155
views
A minimalistic kind of read-copy-update class
This is a very simple read-copy-update (RCU)-inspired synchronization class:
...
- 119
9
votes
3
answers
5k
views
C++ lock free, single producer, single consumer queue
Assumptions for use:
push() will only ever be called by a single thread.
pop() will only ever be called by a single thread.
Requirements:
push() needs to be as fast as possible. Requirements are ...
- 93
1
vote
1
answer
296
views
Passing objects atomically across threads without locks or data races for audio synchronization
I am learning about one of the hardest parts of Audio development: the synchronization between the audio thread and the GUI thread. Per the discussion here https://forum.juce.com/t/timur-doumler-...
- 13
5
votes
0
answers
900
views
A multithreaded, growable vector with immutable elements, which has wait-free reads
Disclaimer for future readers: We experimented around with a similar approach at HASH. The linked Pull Request is working.
I wrote a small vector with a few lines of unsafe Rust.
The idea is to have ...
- 231
3
votes
1
answer
69
views
Concurrent Data Manager with minimal synchronization
So I have my own library to organize advanced critical sections at GitHub. It's based on the ResourceDataStorage class which is responsible for creating, storing and clearing resource records. I've ...
- 133
2
votes
0
answers
166
views
Lock-free statically allocated async-interrupt-safe multi-consumer double buffer
I have a single threaded embedded system with nested interrupts.
A writer interrupt will periodically update some global data structure with data.
Reader interrupts / calls from main thread will read ...
- 121
11
votes
1
answer
4k
views
Lock-free atomic shared pointer in C++14
I'm trying to write a lock-free implementation for atomic shared pointer. Basically, there are two class templates shared_ptr and ...
- 818
6
votes
2
answers
169
views
Simple atomic pseudo-random number generation
I like to use a single global PRNG in my c programs (similar to Javascript's Math.random) because it simplifies things and the unpredictable nature of a global is ...
- 233
1
vote
1
answer
3k
views
A simple implementation of the producer/consumer pattern
Consider this class that consumes some items and handles them with the provided handler function:
...
- 211
4
votes
0
answers
89
views
Swift framework that facilitates making atomic changes to several objects at a time
For the purposes of an app I work on I have created a separate framework that abstracts away the mechanism of making atomic changes to the complex object structures.
Each class that has to ...
- 141
2
votes
0
answers
414
views
Concurrent ObjectPool using ConcurrentLinkedQueue
Main ObjectPool Class: It uses atomic operations and a concurrent linked queue to achieve the best performance compared to Apache commons pool. I want to make sure there are no concurrency issues even ...
- 21
3
votes
1
answer
334
views
Double checked locking with shared_ptr<>
I have an unordered_map<int, std::shared_ptr<foo>> into which I want to lazily create the instances of foo based on ...
- 460
6
votes
2
answers
1k
views
Atomic floating-point addition
I need to be able to atomically add a 32-bit floating point value to a location in memory. Here is what I came up with. While the code is Windows-specific, I'll extend it with Linux support using <...
- 333
2
votes
1
answer
46
views
Stored procedure to swap account holders
I'm doing an exercise, and would like to know if what I'm doing is correct and follows the correct structure for rollback in case a transaction fails. I would like to know as well if my ...
- 23
3
votes
1
answer
157
views
Async safe threadpool
I'm trying the write a threadpool that can safely be added inside a signal handler or in code forked by multithreaded code. Are there any corner cases that would cause this code to fail? What could be ...
- 133
12
votes
1
answer
4k
views
Copyable Atomic
I find it pretty annoying that c++11 atomics can't be copied. The reasons for this have been discussed e.g. here and I don't want to argue about them now. However, I find myself repeatedly in ...
- 593
2
votes
0
answers
139
views
Controller method to add a product, stored across three database tables
I have some questions on how I can improve this "add" action method in a controller.
Using a single form, the user can upload a product to be displayed on the site. A Product is composed of three ...
- 265
3
votes
2
answers
144
views
Single reader - multiple writer waitable lock-free unreliably ordered stack
The main body of code is located here. In the code linted_error is a platform specific type for error codes. As well, as an optimization the code can use Linux ...
5
votes
2
answers
157
views
Lock implementation using atomics
I was recently given an interview and was asked to write my own lock implementation using atomics, but not necessary reentrant.
I didn't answer the question, but at home, I wrote this code. Please ...
- 563
7
votes
1
answer
732
views
Fast non-fair semaphore
This is a simple non-fair semaphore implementation using Java atomic objects. It currently performs a little over twice as fast as java.util.concurrent.Semaphore on ...
- 3,533
3
votes
3
answers
565
views
SPSC Wait Free Ring Buffer for incoming messages
This is for a single producer and single consumer wait free ring buffer. The writes need to be wait free for sure. It pre-allocates messages slots and uses a claim strategy to capture a buffer for ...
- 33
2
votes
1
answer
333
views
Unreliable shared memory circular buffer with read-only consumer
This code is slightly platform dependent but should be pretty easy to port. The goal with this code was to create a circular buffer where the consumer could be limited to read-only access of the ...
3
votes
1
answer
873
views
4
votes
1
answer
89
views
Adding hostname to block list after x consecutive failures in multithreading application
I am using Callable in my code which will be called by multiple threads as shown below. Given a user id in DataKey object, I will find out what are the ...
- 2,167
6
votes
1
answer
4k
views
Lockless multi user, multi consumer FIFO queue using C++11
One of the features that is missing from C++11 are lockless queues. Therefore I have decided to write one myself.
A few remarks:
I know the code isn't technically lockless. However the lock is only ...
- 165
6
votes
1
answer
1k
views
C++ concurrency library
I started a C++11 library of concurrency primitives in order to
study and compare their performance;
provide high-quality implementation of those to use in my projects.
Its main target platform is ...
- 611
5
votes
1
answer
5k
views
Lock-free ringbuffer with multiple readers in C++11
Basic Info
I needed a lock-free ringbuffer with multiple readers (but one writer). However, I did not want the writer to check all readers every time in order to prevent an overrun. Thus, I decided ...
- 153
6
votes
1
answer
688
views
General Batched Job Runner
This question is inspired by:
Generic Task Scheduler
where the problem is to run tasks on a scheduled basis, in parallel, and have individual timeouts for each job.
For example, consider this use-...
- 97.3k