[?@yahoo.com: Found problem of floppy drive acces]

Gary R. Van Sickle g.r.vansickle@worldnet.att.net
Sat Dec 18 02:43:00 GMT 2004

> > > Hello again,
> > > 
> > > I found where the floppy drive was being access and it's 
> definately 
> > > a problem of Cygwin.
> > 
> > 
> >   No it is not.  It is WINDOWS that is accessing the drive 
> and there 
> > is absolutely nothing cygwin can do to either cause or prevent this.
>   Actually, that may not be strictly true.  Let me elaborate:
>   The reason the floppy is being accessed is because some 
> win32 api function or other wants to check its status.  It is 
> quite likely some function related to drive or file 
> properties or information, and it probably scans all the 
> known drive letters despite that being superfluous for most of them.
>   It's not just cygwin, it's absolutely definitely part of 
> windows.  For quite some time my home PC used to always scan 
> the floppy whenever I opened the "My computer" window.

???  Doesn't it still?  I thought it always has and always will because MS
refuses to keep track of media properly....  Hmm, seems like on this fancy
Dell of mine it doesn't... Did MS finally fix a bug that affects

>  And 
> if there's an A:\ path in the File...Open MRU list in wordpad 
> or similar, it will scan the floppy when it starts up.  So 
> this is natural windoze behaviour in lots of circumstances.

That's Wordpad telling Windows to hit the floppy.  Just because a menu has a
bunch of file paths in it doesn't result in any disk being hit.

>   So it is conceivable that this is a consequence of some 
> system call that cygwin makes to get drive info and it is 
> also conceivable that there might be some other way cygwin 
> could get the same info using some other system call that 
> doesn't decide to scan the floppy.
>   But it's also conceivable that there's no way to avoid this 
> side-effect,

No, it isn't, since Cygwin did avoid this somehow until recently.  For
suitable definitions of "somehow" and "recently".

> and it's certainly not a cygwin *bug* that it 
> uses a bog-standard win32 api function in the way it is 
> supposed to be used and for the purpose it is supposed to be 
> used for; it is a *windows* bug that it superfluously scans 
> the floppy.


1.  You've entered the realm of wild speculation.  For all anybody in this
discussion knows, Cygwin could have been changed recently to specifically
and superfluously hit the floppy.
2.  Since we're speculating, what probably happened is that a call got
added/changed somewhere in the bowels of cygwin1.dll to GetAllDriveLetters()
(or whatever that one's called) or one of its ilk, and it hits the floppy
for reasons known only to Microsoft instead of properly supporting
change-line.  Pretty much what you speculate above actually.
3.  The Windows bug is not that it superfluously hits the floppy.  It's
that, for an operating system which grew out of a purported Disk Operating
System,  the one thing it handles worst above all others is mass storage.
Apparently MS puts the new grads and the burnouts on the filesystem team.
How else can you explain a filesystem refusing to mount a USB disk drive
because it can't write to it, even though the drive properly reports
"Read-Only" in the two appropriate structures, properly fails the
WriteSector commands it shouldn't be receiving in the first place, and
properly clears the resulting proper errors when told to do so by Windows?
Yeah, the story you just read is true.  Count your blessings if the worst MS
does to you is hit your floppy drive once in a while.

PS: I thought you were Korn-shell-guy. ;-)

Gary R. Van Sickle

Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

More information about the Cygwin mailing list