by import of Wine commit
313c63e194
merged from current Wine head.
Thanks to patches author Fabian Maurer
and also Doug Lyons for tests and adding initial the merge-patch.
This is done by marking everything that is already loaded but not 'processed'.
After the shim engine is done initializing, the original state is restored,
with the exception of the shim engine itself.
CORE-15846
There was a missing break statement in ftGdiGetGlyphOutline().
Found by JIRA contributor 'I_Kill_Bugs' :-)
Add also brackets around 'case's to make code navigation easier.
Nowadays more and more people try to install ReactOS from removable
drives (e.g. USB sticks) onto fixed HDDs, or try to install it into
USB sticks too.
Both fixed and removable drives, as well as partitions on these, are
represented in NT using the same device name format:
\Device\HarddiskM\PartitionN ,
with an increasing disk number M. Using this number for building the
corresponding firmware-specific ARC multi(x)disk(y)rdisk(z) path used
by the NT/ROS loader (FreeLdr, ...) is then prone to error since there
may have been removable drives inserted and accounted for in the
calculation of the disk number. These drives must be correctly
subtracted in order to generate the correct ARC path, valid once all
the removable drives have been ejected (which should also be the
situation seen from the BIOS when booting up, except of course if you
boot on a USB stick).
This problem is now solved. Note that it matters only for the disks
that have also been enumerated by the firmware (BIOS; Int 13h). We
don't have to care about the other drives, since the ARC path will be
of a different format and will not use the disk number (instead, the
SCSI coordinates are used).
We also try to enumerate all the disks found in all the possible disk
adapters and controllers enumerated in the Hardware registry tree
(and that are visible by FreeLdr) in order to cover all.
Finally, we detect whether a disk reports as a "super-floppy", i.e.
an unpartitioned disk with a valid VBR. This is indeed how a standard
floppy disk looks like, or how USB sticks are partitioned on Windows.
Such disk is reported has having only one single partition starting at
the beginning of the disk, with partition number == 0, its type being
FAT16 non-bootable.
This allows us to forbid creating any new partitions on such disks.
Note that accessing either \Device\HarddiskN\Partition0 or Partition1
on such a disk returns the same data.
Note also that on the contrary, regular MBR-partitioned disks would
report at least four partitions entries, instead of just one.
The other improvements are:
- Do *NOT* write any MBR on a disk partitioned as "super-floppy".
CORE-13703
- Fix the computed disk identifier, of format: %08x-%08x-%c .
The numbers are respectively the checksum of the first sector, and
the disk signature. The terminating letter is A or X, depending
whether the first sector ends with 0x55AA/0xAA55 or not (see also
commit 5053f1f5).
- Warn if the user attempts to install ReactOS on a disk that is not
visible by the firmware of his computer, because it may not be
bootable.
Fixes regression CORE-15785 (Zim Desktop Wiki 0.67 crashed) and
Fixes regression CORE-15755 (NLite 1.4.9.3 used wrong font)
without reintroducing regression CORE-15558 (AbiWord 2.6.8 font enumeration)
This is achieved by partial revert of
0.4.12-dev-320-g
6e4e5a004c
and got ack of Katayama Hirofumi MZ.
Thanks to patches author Doug Lyons.
Test-results: https://reactos.org/testman/compare.php?ids=66264,66267
The last parameter of USBPORT_ParseConfigurationDescriptor is a pointer
to a boolean indicating whether any alternate interface settings were
found for the specified interface.
Interpreting it as an alternate setting value, as we did before,
would always override the alternate setting to '1' (if any alternate
descriptors were present), therefore selecting the wrong interface
setting, and possibly causing a buffer overflow on the InterfaceList's
Pipes array.
Found by Special Pool.
There can be other descriptors between the config descriptor and the
first interface descriptor, so we specifically need to check for
the interface descriptor type and skip anything before that.
We also need to guard against bLength == 0, which would cause an
infinite loop, instead of doing a second bDescriptorType check.
