[atomic] Bikeshedding over address-free waiting/notifying operation names
Hi, As some of you may know, C++20 added new waiting/notifying operations to std::atomic: wait, notify_one and notify_all. These operations are not address-free, i.e. their implementation is allowed to use process-local addresses of the atomics as keys. In particular, it means you can't use them for inter-process communication. In Boost.Atomic I intend to provide address-free variants as well, where possible, and I'm thinking about the naming scheme for them. Currently, I'm using the "_address_free" suffix, so we have wait_address_free, notify_one_address_free and notify_all_address_free. The "address-free" term comes from the C++ standard, but the resulting names seem a bit long. Maybe there are better suggestions? Thanks.
On Sun, 31 May 2020 at 16:05, Andrey Semashev via Boost < boost@lists.boost.org> wrote:
What about "global_"-prefix or "_system"-post-fix ("_sys")? degski -- @systemdeg https://www.instapaper.com http://instapaper.com "We value your privacy, click here!" Sod off! - degski
On Mon, 1 Jun 2020 at 06:18, Andrey Semashev via Boost < boost@lists.boost.org> wrote:
On the assumption that the std-post-fixes have already extensively been bike-shedded, any alternative will likely be a compromise between length and descriptiveness, so your 'hmmm ...' is natural, I'm thinking about words around 'virtual memory' (because that's where the non-local addresses come from), while "_virtual" is out of the question. degski -- @systemdeg https://www.instapaper.com http://instapaper.com "We value your privacy, click here!" Sod off! - degski
On 2020-06-01 15:07, degski wrote:
Many platforms use the term "physical address" to describe the underlying key source. However, wait_physical_address is not really better than wait_address_free. I've also considered "_pshared" suffix, which comes from POSIX, but the address-free operations can also be useful within the same process, when the atomic object is mapped at multiple addresses.
On 2020-06-01 15:05, Phil Endecott via Boost wrote:
That name cannot be used in Boost.Atomic as it is reserved for Boost.Interprocess. Atomics that are lock-free can already be used for inter-process communication, so a hypothetical boost::interprocess::atomic would largely duplicate boost::atomic. However, I'm open to the idea, if it is significantly different from boost::atomic.
The obvious solution then is to use boost::atomic::interprocess ;-) I do see value in having 2 different atomic types in different namespaces as the use case is different: The "regular" atomic usually works. Once you need interprocess communication you use another type, not another member function.
... I was going to suggest the same :)
I do see value in having 2 different atomic types in different namespaces as the use case is different: The "regular" atomic usually works. Once you need interprocess communication you use another type, not another member function.
Agreed, having two different types has benefits: - you can prevent people to pass a boost::atomic where a boost::atomic::interprocess is needed (if that makes any sense..., do atomics ever appear in interfaces?) - you can make the intent clear in the implementation of a class or function
On 2020-06-02 17:28, Hans Dembinski via Boost wrote:
The problem with this is that the resulting name will be long and have the potential of name clash with Boost.Interprocess. Boost.Atomic types are currently defined in namespace boost::atomics and then imported into namespace boost (so that people can use boost::atomic<int> instead of boost::atomics::atomic<int>). Adding namespace boost::atomics::interprocess would make it impossible to import it into namespace boost as it would conflict with Boost.Interprocess. So people will have to spell boost::atomics::interprocess::atomic<int>, which is unfortunate. There is an alternative to name the type differently, e.g. boost::atomics::interprocess_atomic<int> or boost::atomics::pshared_atomic<int>, though I'm not sure I like it either. Or use a different name for the nested namespace instead of interprocess (ipc? other suggestions?).
Yes, I can see the potential benefit of having a separate type for inter-process. That would allow writing generic algorithms over process-local and inter-process atomics. It's those naming details I'm not quite happy about. Also, and more importantly, given that atomics may potentially use runtime detection of availability of lock-free operations, it isn't clear what behavior should inter-process atomics have if the required operations turn out to be unavailable in run time. This is not a problem in the current implementation, as it always uses compile-time detection, but I wouldn't want to prohibit runtime detection in the future.
Makes sense. I guess we all agree that redundancy in names/namespaces is bad. If the hypothetical namespace boost::atomics::interprocess is going to contains only a single type anyway, then boost::atomics::interprocess_atomic<int> seems like a good compromise to me, or maybe boost::atomics::ipc_atomic<int> if you want it shorter. Best regards, Hans
On 2020-06-03 14:44, Hans Dembinski via Boost wrote:
Makes sense. I guess we all agree that redundancy in names/namespaces is bad. If the hypothetical namespace boost::atomics::interprocess is going to contains only a single type anyway, then boost::atomics::interprocess_atomic<int> seems like a good compromise to me, or maybe boost::atomics::ipc_atomic<int> if you want it shorter.
Or inter_atomic<T> if we want something that is easily pronounced.
participants (6)
-
Alexander Grund -
Andrey Semashev -
Bjorn Reese -
degski -
Hans Dembinski -
Phil Endecott