zero-copy API

[Denis Koroskin]

On Fri, 15 Oct 2010 08:03:21 +0400, Sean Kelly <sean at invisibleduck.org>  
wrote:

> Andrei Alexandrescu <SeeWebsiteForEmail at erdani.org> wrote:
>> On 10/14/2010 07:50 PM, Denis Koroskin wrote:
>>> On Fri, 15 Oct 2010 03:08:23 +0400, Andrei Alexandrescu
>>> <SeeWebsiteForEmail at erdani.org> wrote:
>>>
>>>> http://docs.google.com/viewer?a=v&q=cache:K15RE_6zxSwJ:citeseerx.ist.psu.edu/viewdoc/download%3Fdoi%3D10.1.1.134.4874%26rep%3Drep1%26type%3Dpdf+zero+copy+i/o&hl=en&gl=us&pid=bl&srcid=ADGEESjBkiUxG4hRImVjOFy886GrJxRuhFcePjbadiUw9h1c_iicbhhArOgd55vpk0tP6ST4KjhY1j6rl1_PN-msIExUvxSPJWuXfQTbljj4ZYyutY6wvp3mc3t2LuA2-5kKPbbEp7z6&sig=AHIEtbSmuH-Y2AGdwSQxyJcbBXLRB3mJdg
>>>>
>>>>
>>>>
>>>> Andrei
>>>
>>> That paper is an ancient one and the API they demonstrated isn't a
>> > good
>>> one.
>>
>> What is a better, more recent one? I noticed that libfbufs has little
>> traction (googling for it only reveals 4 results) but the paper is
>> quoted by 63 others, so it's fairly influential.
>
> Perhaps not terribly relevant, but IOCP on Windows does this. To perform
> an asynchronous read you supply the buffer to recv and then an event is
> signaled when the read completes. The same API works for file io and
> there's support for scatter/gather as well. It makes for a more
> complicated program though.

We are using IOCP for a game server with a great success. I'm not even  
sure we could achieve the same level of latency if we were using Linux and  
epoll.

I've implemented a Stream API (FileStream and SocketStream) on top of IOCP  
(both sync and async), and it doesn't complicate a program when used  
right. Automatic thread pooling for event callbacks is very handy, and  
since events for the same socket don't overlap you don't even need to  
synchronize anything but global state (in which case a simple  
ReadWriteMutex is more than enough).

Unfortunately, I'm not sure how to integrate that concept into D2 design.  
Even though events don't overlap, the callback thread is always different  
and as such thread local variables are unaccessible. I was  thinking about  
dynamic thread local variables switching, but Walter says it is either  
highly inefficient or impossible to to implement.