60a
Subj : Re: Threading and Timeouts
To   : comp.programming.threads
From : David Schwartz
Date : Wed Jul 20 2005 05:51 pm


"Giancarlo Niccolai" <searchforme@ingoogle.com> wrote in message 
news:dbml2b$ec$1@newsread.albacom.net...

> Again, we are talking about blocking in wait for the system to be ready to
> accept data. I.E. In case of  sync i/o write() is forced to wait untill 
> all
> data are phisically gone on the media. What select() GUARANTEES is that 
> the
> fd is in a status that, if it were O_NONBLOCK, then it may not return
> EAGAIN, that is to say, that the system will immediately process *at 
> least*
> part of the data that can be sent on the fd.

    Except that it doesn't guarantee this. This is a common, but totally 
mistaken, belief. The 'select' function calls gives you information that is 
correct for some time interval between when you called 'select' and when it 
returns about what would have happened to a socket operation in that time 
period. It provides no future guarantees, and the standards are very careful 
*NOT* to say it does.

    This is the same for every other status function. For example, does a 
call to 'access' guarantee that a future access will succeed? Of course not. 
Anything can change in-between when 'access' returns and when you attempt 
the access, and nothing in POSIX requires anything to remain true later.

    I defy you to find any statement that says or implies a *future* 
guarantee about what *will* or *must* happen later.

    DS

.

0