Hi Tim, Now that I've been using lockfree::spsc_queue a bit, I was wondering why there is no method to retrieve write_available() on the producer end. It seems like it would be more efficient to be able to retrieve this before writing in some cases, like say when you want to overwrite old values from the producer thread in order to keep the data fresh. Instead, I'm having to do two push()'s. cheers, Rich p.s. Please let me know if I should resort to the boost users forum instead of this one for questions like these. On Sun, Jul 8, 2012 at 10:54 PM, Rich E <reakinator@gmail.com> wrote:
Thanks for the clarification, that makes perfect sense. Good to know about being able to configure memory usage. I look forward to trying these out in the coming days..
On Sun, Jul 8, 2012 at 3:27 AM, Tim Blechmann <tim@klingt.org> wrote:
First off, all Boost.Lockfree tests in trunk are passing for me on clang 4.0 on OS X, hurrah! Thank you Tim for working on this for so long...
I'm running through the examples now, only with the outdated documentation as a guide (more than a couple years old). I cannot find the ringbuffer container - was it renamed or removed altogether during the review period?
I notice spsc_queue acts awefully like a ringbuffer, and it's implementation contains many
... ringbuffer-like classes - is this the decided class for creating a lockfree buffer that reuses memory?
the ringbuffer was renamed to spsc_queue: internal data structure is a ringbuffer, but it models a single-producer single-consumer wait-free queue. so the name reflects the properties rather than the implementation. iirc the same data structure is called `bounded queue' in shavit/herlihy, fifo in the linux kernel and ringbuffer in many audio-related projects.
btw, all data structures in boost.lockfree can be configured to reuse memory in order to be usable in systems where you want to avoid memory allocations
tim