Wolfgang Rupprecht wrote:

> Simon Burge writes:
> 
> > Would using dup2(2) be a way to hack around this?
> 
> I don't think so, if I understand the issue correctly.  The man page
> seems to be hinting that the /dev/cd0d's reference count must drop
> down to zero before the driver lets the new disc's data appear to the
> user.

I was thinking of something along the lines of:

	close(cdromfd);
	tmpfd = open(device, ...);
	dup2(tmpfd, cdromfd);
	close(tmpfd);

That shouldn't have any ref-count problems (unless of course cdromfd is
dup()'d anywhere else).

Simon.
--
Simon Burge                            <simonb@wasabisystems.com>
NetBSD CDs, Support and Service:    http://www.wasabisystems.com/