Re: [boost] [AFIO] Callback API

Subject: Re: [boost] [AFIO] Callback API
From: Niall Douglas (s_sourceforge_at_[hidden])
Date: 2014-01-01 11:27:05

On 1 Jan 2014 at 12:31, Bjorn Reese wrote:

> > By "calling the next write operation", I'm thinking that you're
> > thinking that a kernel API ought to be issued right there and then
> > yes just like ASIO does?
> 
> Yes.
> 
> > understand exactly what you're looking for, and then I can implement
> > it.
> 
> I have started working on a prototype to see how feasible it is.

Be aware I didn't reject the naïve ASIO-style API design due to the 
API design itself. No, rather I rejected it because file i/o is 
always blocking [1] and issuing kernel API calls immediately resulted 
in pathological performance. In short, socket i/o API calls are nice 
and deterministic with bounded worst case execution times, whereas 
file i/o is nasty and unpredictable and NP worst case execution 
times.

[1]: This is a very tl;dr discussion worthy of a 90 minute 
presentation, but in short yes this is true on all mainstream 
operating systems. Linux *can* turn on non-blocking for file handles, 
but the feature has "dragons live here" all over it. On Windows an 
overlapped file i/o operation can and does go synchronous on many an 
occasion - I've seen an overlapped WriteFile() take 1.6 seconds to 
return "operation in progress".

Niall

-- 
Currently unemployed and looking for work.
Work Portfolio: http://careers.stackoverflow.com/nialldouglas/