Gary said:
Not what I was referring to. All operations stop, even those
not dependent on the CD/DVD.
Gar
I'm not sure of my conjecture since I've never looked at the XP code: My
conjecture is that when the drive informs the computer of an insert
event, a dialogue between the OS and the drive occurs where various
pieces of information about the media are gathered. I further conjecture
that the OS does a "busy wait" for the drive to gather each piece of
information and pass it over. Since this involves "spinning up" the
media, the wait is noticeable to a human. My conjecture is that the
hardware interrupt mechanism is not used at this point (other than to
post the insert notification) and isn't invoked until later when the
media is read or written. If this conjecture is correct, this
inefficiency is historical and nobody has bother to rewrite the
low-speed portion of some very old code.
Anyone out there looked at the code to verify or reject the above?