Background processes setting O_NONBLOCK on ttys

[Stephen McKay]

On Thursday, 13th January 2005, Jilles Tjoelker wrote:

>POSIX states that the O_NONBLOCK flag is per open file description. File
>descriptors created with dup(2), fork(2) or similar methods share the
>same open file description, while new descriptors created with open(2)
>do not.

Can you quote chapter and verse on this?  Or is it merely implied?  A pointer
to an online copy would be best.

I found that "The Open Group Base Specifications Issue 6" at
http://www.opengroup.org/onlinepubs/009695399/toc.htm is not clear
on the matter.  Also "The Single UNIX Specification, Version 2" at
http://www.opengroup.org/onlinepubs/007908799/ is similarly unclear.

>Consequently, a POSIX-compliant implementation of O_NONBLOCK does not
>solve the problem, as all descriptors to the terminal refer to the same
>open file description. O_NONBLOCK is indeed unusable for descriptions
>shared with other unknowing processes :(

If we must be POSIX compliant and if you accept that POSIX demands that
dup'd file descriptors share a single O_NONBLOCK flag, then the logical
end position is that the threads library must not routinely set stdin,
stdout and stderr to non-blocking.  Right?

Personally, I have no qualms being non-POSIX compliant in a good cause.
Wouldn't being non-POSIX on this issue be easier?

Hmm.  Should this be on -arch now?

Stephen.