| Index | index by Group | index by Distribution | index by Vendor | index by creation date | index by Name | Mirrors | Help | Search |
| Name: atomic-queue-devel | Distribution: Fedora Project |
| Version: 1.1 | Vendor: Fedora Project |
| Release: 1.fc38 | Build date: Sat Feb 25 16:03:31 2023 |
| Group: Unspecified | Build host: buildvm-x86-12.iad2.fedoraproject.org |
| Size: 58288 | Source RPM: atomic-queue-1.1-1.fc38.src.rpm |
| Packager: Fedora Project | |
| Url: https://github.com/max0x7ba/atomic_queue | |
| Summary: Development files for atomic-queue | |
C++14 multiple-producer-multiple-consumer lockless queues based on circular
buffer with std::atomic.
It has been developed, tested and benchmarked on Linux, but should support any
C++14 platforms which implement std::atomic.
The main design principle these queues follow is minimalism: the bare minimum
of atomic operations, fixed size buffer, value semantics.
These qualities are also limitations:
• The maximum queue size must be set at compile time or construction time.
The circular buffer side-steps the memory reclamation problem inherent in
linked-list based queues for the price of fixed buffer size. See Effective
memory reclamation for lock-free data structures in C++ for more details.
Fixed buffer size may not be that much of a limitation, since once the
queue gets larger than the maximum expected size that indicates a problem
that elements aren’t processed fast enough, and if the queue keeps growing
it may eventually consume all available memory which may affect the entire
system, rather than the problematic process only. The only apparent
inconvenience is that one has to do an upfront back-of-the-envelope
calculation on what would be the largest expected/acceptable queue size.
• There are no OS-blocking push/pop functions. This queue is designed for
ultra-low-latency scenarios and using an OS blocking primitive would be
sacrificing push-to-pop latency. For lowest possible latency one cannot
afford blocking in the OS kernel because the wake-up latency of a blocked
thread is about 1-3 microseconds, whereas this queue’s round-trip time can
be as low as 150 nanoseconds.
Ultra-low-latency applications need just that and nothing more. The minimalism
pays off, see the throughput and latency benchmarks.
Available containers are:
• AtomicQueue - a fixed size ring-buffer for atomic elements.
• OptimistAtomicQueue - a faster fixed size ring-buffer for atomic elements
which busy-waits when empty or full.
• AtomicQueue2 - a fixed size ring-buffer for non-atomic elements.
• OptimistAtomicQueue2 - a faster fixed size ring-buffer for non-atomic
elements which busy-waits when empty or full.
These containers have corresponding AtomicQueueB, OptimistAtomicQueueB,
AtomicQueueB2, OptimistAtomicQueueB2 versions where the buffer size is
specified as an argument to the constructor.
Totally ordered mode is supported. In this mode consumers receive messages in
the same FIFO order the messages were posted. This mode is supported for push
and pop functions, but for not the try_ versions. On Intel x86 the totally
ordered mode has 0 cost, as of 2019.
Single-producer-single-consumer mode is supported. In this mode, no
read-modify-write instructions are necessary, only the atomic loads and stores.
That improves queue throughput significantly.
Move-only queue element types are fully supported. For example, a queue of
std::unique_ptr<T> elements would be AtomicQueue2B<std::unique_ptr<T>> or
AtomicQueue2<std::unique_ptr<T>, CAPACITY>.
The atomic-queue-devel package contains libraries and header files for
developing applications that use atomic-queue.
MIT
* Sat Feb 25 2023 Benjamin A. Beasley <code@musicinmybrain.net> - 1.1-1 - Update to 1.1 (close RHBZ#2173288) * Wed Jan 18 2023 Fedora Release Engineering <releng@fedoraproject.org> - 1.0-9 - Rebuilt for https://fedoraproject.org/wiki/Fedora_38_Mass_Rebuild * Mon Dec 19 2022 Benjamin A. Beasley <code@musicinmybrain.net> - 1.0-8 - Indicate dirs. in files list with trailing slashes * Mon Dec 19 2022 Benjamin A. Beasley <code@musicinmybrain.net> - 1.0-7 - Leaf package: remove i686 support * Sun Jul 31 2022 Benjamin A. Beasley <code@musicinmybrain.net> - 1.0-6 - Confirm that License is SPDX MIT (no License field change) * Wed Jul 20 2022 Fedora Release Engineering <releng@fedoraproject.org> - 1.0-5 - Rebuilt for https://fedoraproject.org/wiki/Fedora_37_Mass_Rebuild * Wed Jan 19 2022 Fedora Release Engineering <releng@fedoraproject.org> - 1.0-4 - Rebuilt for https://fedoraproject.org/wiki/Fedora_36_Mass_Rebuild * Thu Dec 16 2021 Benjamin A. Beasley <code@musicinmybrain.net> - 1.0-3 - Let the devel subpackage be noarch * Thu Dec 09 2021 Benjamin A. Beasley <code@musicinmybrain.net> - 1.0-2 - Fix release number—no longer a pre-release * Thu Dec 09 2021 Benjamin A. Beasley <code@musicinmybrain.net> - 1.0-0.1 - Update to 1.0 (close RHBZ#2030645) * Fri Oct 22 2021 Benjamin A. Beasley <code@musicinmybrain.net> - 0-0.6 - Update to 7013a8b (support up to 256 byte cache lines) * Wed Oct 20 2021 Benjamin A. Beasley <code@musicinmybrain.net> - 0-0.5 - Update to ee3d91c: fix RHBZ#1994598, fix RHBZ#1994599 - Enable verbose test output - Run the example as an additional test * Tue Oct 19 2021 Benjamin A. Beasley <code@musicinmybrain.net> - 0-0.4 - Update to commit dfd2cbe * Fri Sep 24 2021 Benjamin A. Beasley <code@musicinmybrain.net> - 0-0.3 - We really don’t need to adjust the C++ standard for the tests * Mon Aug 16 2021 Benjamin A. Beasley <code@musicinmybrain.net> - 0-0.2 - Fix an long line in the -devel package description * Mon Aug 16 2021 Benjamin A. Beasley <code@musicinmybrain.net> - 0-0.1 - Initial package
/usr/include/atomic_queue /usr/include/atomic_queue/atomic_queue.h /usr/include/atomic_queue/atomic_queue_mutex.h /usr/include/atomic_queue/barrier.h /usr/include/atomic_queue/defs.h /usr/include/atomic_queue/spinlock.h /usr/share/doc/atomic-queue-devel /usr/share/doc/atomic-queue-devel/README.md /usr/share/licenses/atomic-queue-devel /usr/share/licenses/atomic-queue-devel/LICENSE
Generated by rpm2html 1.8.1
Fabrice Bellet, Tue Apr 9 23:07:34 2024