JdBP> HOWEVER, that does *not* mean that the device driver can USE those
JdBP> inherited file handles when it is invoked in the context of any
JdBP> arbitrary child process.  For one thing, only certain portions of a
JdBP> device driver, such as the INIT routine, can use application-level
JdBP> kernel APIs, since most parts of a device driver run at ring 0, not ring 
JdBP> 3.

Ok, I'll take your word for it.

JdBP> For another, even if the device driver *could* make application-level
JdBP> API calls to perform file I/O, there's no guarantee that the process, or
JdBP> one of its parents, will not have closed the file handle concerned and
JdBP> possibly opened another one in its stead.  There are no
"system" or
JdBP> "reserved" file handles.  Each process is sovereign as
far as its open
JdBP> file handles are concerned.

But I don't agree with this.  You don't have the right to close arbitrary
file handles.  You're only allowed to close files that you yourself have
opened.  That's a bit like saying that the device driver shouldn't rely on
being able to access memory it just malloc'ed, because "the user might
have run a program to go and deliberately clobber that memory".

JdBP> So, having covered the ground more explicitly this time around, I'll
JdBP> repeat the point that I made above.  Any device driver that assumes that
JdBP> an arbitrary process will have a specific open file handle is badly
JdBP> broken.

OK, maybe IBM fixed the TCP/IP stuff in Warp 4 then, and the problem can be
solved from that end.  Actually, I know how to report bugs for free now, so
we might be able to do something about that.  Although the solution we have
now is working pretty fine.  The reason I say pretty fine is because there
is a report that you need to set maxfh to 80 to make it work completely, 40
won't cut it.  That test hasn't been done to my satisfaction yet. 
(Basically I don't believe 40 isn't good enough). It is possible though,
that something else is using up file handles in the compile process, hence
the probelm.

BTW, if you see a message from me at 5AM, it's NOT because I'm an
earlybird.  :-)  And actually, I haven't adjusted my computer's time, since
we're on 1 hour daylight saving at the moment (unlike you bloody Northern
Hemispherians who don't even know the joy of a hot Christmas :-)).

BFN.  Paul.
@EOT:

---