[Bug 5099] Problem installing Command and Conquer Tiberian Sun

[Wine Bugs]

http://bugs.winehq.org/show_bug.cgi?id=5099

------- Additional Comments From [e-mail]  2006-23-04 13:07 -------
| looks like device 9 is really my cd device

Not exactly, Bernd.  "device=9" refers to the device type (that is to
FILE_DEVICE_FILE_SYSTEM) based on which FSCTL_[UN]LOCK_VOLUME consts are
defined.  And the actual reference is a handle to the "\.\D:" (0xd8) :

| 0009:trace:file:CreateFileW L"\\.\D:" GENERIC_READ FILE_SHARE_READ 
|   FILE_SHARE_WRITE  creation 3 attributes 0x0
  ...
| 0009:trace:file:CreateFileW returning 0xd8
| 0009:trace:ntdll:NtFsControlFile 
|   0xd8,(nil),(nil),(nil),0x7fbbee44,0x00090018,(nil),0x00000000,(nil),0x00...)
| 0009:fixme:ntdll:NtFsControlFile Unsupported fsctl 90018 (device=9 access=0 
|   func=6 method=0)
| 
| /dev/hdc is my cd device

Fine.  Now it would be nice to run some test on real Windows:

handleC = CreateFileW("\.\C:", ...);
  handleD = CreateFileW("\.\D:", ...);
  NtQueryObject(handleC, Object...Information , ...);
  NtQueryObject(handleD, Object[Type|Name]Information, &ObjectInformation, ...);

/* Then print some fields of the returned ObjectInformation struct ... */
  /* so we could see the difference between handles to C: and to D:      */

res = NtFsControlFile(handleD, FSCTL_LOCK_VOLUME, ...);

IMHO, this could be a start for additional ntdll-tests.  Vitaliy, what do you think?